From 13fe24f37df20e580a5a364e67ec8cf3219d8f8c Mon Sep 17 00:00:00 2001
From: Robin Getz <robin.getz@analog.com>
Date: Sun, 27 Jan 2008 15:38:56 +0800
Subject: [Blackfin] arch: fix bug - trap_tests fails to recover on some tests.

http://blackfin.uclinux.org/gf/project/uclinux-dist/tracker/?action=TrackerItemEdit&tracker_item_id=3719

When the CPLBs get a miss, we do:
  - find a victim in the HW table
  - remove the victim
  - find the replacement in the software table
  - put it into the HW table.

If we can't find a replacement in the software table, we accidently
leave a duplicate in the HW table. This patch ensures that duplicate
is marked as not valid.

What we should do is find the replacement in the software table, before
we find a victim in the HW table - but its too late in the release cycle
to do that much restructuring of this code.

Rather that duplicate code, connect Hardware Errors (irq5) into trap_c,
so user space processes get killed properly.

The rest of irq_panic() can be moved into traps.c (later)

There is still a small corner case that causes problems when a
pheriperal interrupt goes off a single cycle before a user space
hardware error. This causes a kernel panic, rather than the user
space process being killed.

But, this checkin makes things work in 99.9% of the cases, and is a vast
improvement from what is there today (which fails 100% of the time).

Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 include/asm-blackfin/traps.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-blackfin/traps.h b/include/asm-blackfin/traps.h
index ee1cbf73a9ab..f0e5f940d9ca 100644
--- a/include/asm-blackfin/traps.h
+++ b/include/asm-blackfin/traps.h
@@ -45,6 +45,10 @@
 #define VEC_CPLB_I_M	(44)
 #define VEC_CPLB_I_MHIT	(45)
 #define VEC_ILL_RES	(46)	/* including unvalid supervisor mode insn */
+/* The hardware reserves (63) for future use - we use it to tell our
+ * normal exception handling code we have a hardware error
+ */
+#define VEC_HWERR	(63)
 
 #ifndef __ASSEMBLY__
 
-- 
cgit v1.2.3


From a5bb85dfffd38714d83cf7798886462d7314b90a Mon Sep 17 00:00:00 2001
From: Cliff Cai <cliff.cai@analog.com>
Date: Fri, 21 Dec 2007 21:04:40 +0800
Subject: [Blackfin] arch: fix bug SDIO driver fails to build for BF542/BF548

modified CMD_TIMEOUT and DAT_TIMEOUT to CMD_TIME_OUT and DAT_TIME_OUT

Signed-off-by: Cliff Cai <cliff.cai@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 include/asm-blackfin/mach-bf548/defBF542.h | 4 ++--
 include/asm-blackfin/mach-bf548/defBF548.h | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-blackfin/mach-bf548/defBF542.h b/include/asm-blackfin/mach-bf548/defBF542.h
index 32d07130200c..a7c809f29ede 100644
--- a/include/asm-blackfin/mach-bf548/defBF542.h
+++ b/include/asm-blackfin/mach-bf548/defBF542.h
@@ -432,8 +432,8 @@
 
 #define              CMD_CRC_FAIL  0x1        /* CMD CRC Fail */
 #define              DAT_CRC_FAIL  0x2        /* Data CRC Fail */
-#define               CMD_TIMEOUT  0x4        /* CMD Time Out */
-#define               DAT_TIMEOUT  0x8        /* Data Time Out */
+#define               CMD_TIME_OUT  0x4        /* CMD Time Out */
+#define               DAT_TIME_OUT  0x8        /* Data Time Out */
 #define               TX_UNDERRUN  0x10       /* Transmit Underrun */
 #define                RX_OVERRUN  0x20       /* Receive Overrun */
 #define              CMD_RESP_END  0x40       /* CMD Response End */
diff --git a/include/asm-blackfin/mach-bf548/defBF548.h b/include/asm-blackfin/mach-bf548/defBF548.h
index ecbca952985c..e46f56891e6a 100644
--- a/include/asm-blackfin/mach-bf548/defBF548.h
+++ b/include/asm-blackfin/mach-bf548/defBF548.h
@@ -1095,8 +1095,8 @@
 
 #define              CMD_CRC_FAIL  0x1        /* CMD CRC Fail */
 #define              DAT_CRC_FAIL  0x2        /* Data CRC Fail */
-#define               CMD_TIMEOUT  0x4        /* CMD Time Out */
-#define               DAT_TIMEOUT  0x8        /* Data Time Out */
+#define               CMD_TIME_OUT  0x4        /* CMD Time Out */
+#define               DAT_TIME_OUT  0x8        /* Data Time Out */
 #define               TX_UNDERRUN  0x10       /* Transmit Underrun */
 #define                RX_OVERRUN  0x20       /* Receive Overrun */
 #define              CMD_RESP_END  0x40       /* CMD Response End */
-- 
cgit v1.2.3


From fb5f00492a748facc9f069c95621e05c148edf53 Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sun, 23 Dec 2007 23:02:13 +0800
Subject: [Blackfin] arch: Fix bug to Enable kernel to build for bf548 with PM.

On BF548-EZKIT, build kernel faills with power management, video and audio enabled.
This patch fix this.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/Kconfig                      |  2 ++
 arch/blackfin/mach-bf548/head.S            |  8 +++---
 arch/blackfin/mach-bf548/ints-priority.c   |  2 +-
 arch/blackfin/mach-common/dpmc.S           | 46 ++++++++++++++++++++++++------
 arch/blackfin/mach-common/pm.c             | 16 +++++++++++
 include/asm-blackfin/mach-bf548/irq.h      |  4 +--
 include/asm-blackfin/mach-bf548/mem_init.h |  1 +
 7 files changed, 64 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 25232ba08119..5ebcfd226ed8 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -951,6 +951,8 @@ config PM_WAKEUP_SIC_IWR
 	depends on PM_WAKEUP_GPIO_BY_SIC_IWR
 	default 0x80000000 if (BF537 || BF536 || BF534)
 	default 0x100000 if (BF533 || BF532 || BF531)
+	default 0x800000 if (BF549 || BF548 || BF547 || BF542)
+	default 0x800000 if (BF527 || BF524 || BF522)
 
 config PM_WAKEUP_GPIO_NUMBER
 	int "Wakeup GPIO number"
diff --git a/arch/blackfin/mach-bf548/head.S b/arch/blackfin/mach-bf548/head.S
index 74b34c7f3629..745662e88759 100644
--- a/arch/blackfin/mach-bf548/head.S
+++ b/arch/blackfin/mach-bf548/head.S
@@ -298,8 +298,8 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
-	p0.h = hi(SIC_IWR);
-	p0.l = lo(SIC_IWR);
+	p0.h = hi(SIC_IWR0);
+	p0.l = lo(SIC_IWR0);
 	r0.l = 0x1;
 	r0.h = 0x0;
 	[p0] = r0;
@@ -395,8 +395,8 @@ ENTRY(_start_dma_code)
 	[P2] = R1;
 	SSYNC;
 
-	p0.h = hi(SIC_IWR);
-	p0.l = lo(SIC_IWR);
+	p0.h = hi(SIC_IWR0);
+	p0.l = lo(SIC_IWR0);
 	r0.l = lo(IWR_ENABLE_ALL);
 	r0.h = hi(IWR_ENABLE_ALL);
 	[p0] = r0;
diff --git a/arch/blackfin/mach-bf548/ints-priority.c b/arch/blackfin/mach-bf548/ints-priority.c
index cb0ebac53c79..05e3b3cb4c0a 100644
--- a/arch/blackfin/mach-bf548/ints-priority.c
+++ b/arch/blackfin/mach-bf548/ints-priority.c
@@ -58,7 +58,7 @@ void program_IAR(void)
 			    ((CONFIG_IRQ_PINT1 - 7) << IRQ_PINT1_POS) |
 			    ((CONFIG_IRQ_MDMAS0 - 7) << IRQ_MDMAS0_POS) |
 			    ((CONFIG_IRQ_MDMAS1 - 7) << IRQ_MDMAS1_POS) |
-			    ((CONFIG_IRQ_WATCHDOG - 7) << IRQ_WATCHDOG_POS));
+			    ((CONFIG_IRQ_WATCHDOG - 7) << IRQ_WATCH_POS));
 
 	bfin_write_SIC_IAR3(((CONFIG_IRQ_DMAC1_ERR - 7) << IRQ_DMAC1_ERR_POS) |
 			    ((CONFIG_IRQ_SPORT2_ERR - 7) << IRQ_SPORT2_ERR_POS) |
diff --git a/arch/blackfin/mach-common/dpmc.S b/arch/blackfin/mach-common/dpmc.S
index 39fbc2861107..301ac1b6681f 100644
--- a/arch/blackfin/mach-common/dpmc.S
+++ b/arch/blackfin/mach-common/dpmc.S
@@ -38,6 +38,9 @@ ENTRY(_unmask_wdog_wakeup_evt)
 #if defined(CONFIG_BF561)
 	P0.H = hi(SICA_IWR1);
 	P0.L = lo(SICA_IWR1);
+#elif defined(CONFIG_BF54x) || defined(CONFIG_BF52x)
+	P0.h = HI(SIC_IWR0);
+	P0.l = LO(SIC_IWR0);
 #else
 	P0.h = HI(SIC_IWR);
 	P0.l = LO(SIC_IWR);
@@ -236,7 +239,7 @@ ENTRY(_deep_sleep)
 
 	call _set_sic_iwr;
 
-	call _set_sdram_srfs;
+	call _set_dram_srfs;
 
 	/* Clear all the interrupts,bits sticky */
 	R0 = 0xFFFF (Z);
@@ -253,7 +256,7 @@ ENTRY(_deep_sleep)
 	SSYNC;
 	IDLE;
 
-	call _unset_sdram_srfs;
+	call _unset_dram_srfs;
 
 	call _test_pll_locked;
 
@@ -285,7 +288,7 @@ ENTRY(_sleep_deeper)
 	P3 = R0;
 	R0 = IWR_ENABLE(0);
 	call _set_sic_iwr;
-	call _set_sdram_srfs;
+	call _set_dram_srfs;
 
 	/* Clear all the interrupts,bits sticky */
 	R0 = 0xFFFF (Z);
@@ -360,7 +363,7 @@ ENTRY(_sleep_deeper)
 	IDLE;
 	call _test_pll_locked;
 
-	call _unset_sdram_srfs;
+	call _unset_dram_srfs;
 
 	STI R4;
 
@@ -368,25 +371,47 @@ ENTRY(_sleep_deeper)
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
 
-ENTRY(_set_sdram_srfs)
-	/*  set the sdram to self refresh mode */
+ENTRY(_set_dram_srfs)
+	/*  set the dram to self refresh mode */
+#if defined(CONFIG_BF54x)
+	P0.H = hi(EBIU_RSTCTL);
+	P0.L = lo(EBIU_RSTCTL);
+	R2 = [P0];
+	R3.H = hi(SRREQ);
+	R3.L = lo(SRREQ);
+#else
 	P0.H = hi(EBIU_SDGCTL);
 	P0.L = lo(EBIU_SDGCTL);
 	R2 = [P0];
 	R3.H = hi(SRFS);
 	R3.L = lo(SRFS);
+#endif
 	R2 = R2|R3;
 	[P0] = R2;
 	ssync;
+#if defined(CONFIG_BF54x)
+.LSRR_MODE:
+	R2 = [P0];
+	CC = BITTST(R2, 4);
+	if !CC JUMP .LSRR_MODE;
+#endif
 	RTS;
 
-ENTRY(_unset_sdram_srfs)
-	/*  set the sdram out of self refresh mode */
+ENTRY(_unset_dram_srfs)
+	/*  set the dram out of self refresh mode */
+#if defined(CONFIG_BF54x)
+	P0.H = hi(EBIU_RSTCTL);
+	P0.L = lo(EBIU_RSTCTL);
+	R2 = [P0];
+	R3.H = hi(SRREQ);
+	R3.L = lo(SRREQ);
+#else
 	P0.H = hi(EBIU_SDGCTL);
 	P0.L = lo(EBIU_SDGCTL);
 	R2 = [P0];
 	R3.H = hi(SRFS);
 	R3.L = lo(SRFS);
+#endif
 	R3 = ~R3;
 	R2 = R2&R3;
 	[P0] = R2;
@@ -394,8 +419,13 @@ ENTRY(_unset_sdram_srfs)
 	RTS;
 
 ENTRY(_set_sic_iwr)
+#if defined(CONFIG_BF54x) || defined(CONFIG_BF52x)
+	P0.H = hi(SIC_IWR0);
+	P0.L = lo(SIC_IWR0);
+#else
 	P0.H = hi(SIC_IWR);
 	P0.L = lo(SIC_IWR);
+#endif
 	[P0] = R0;
 	SSYNC;
 	RTS;
diff --git a/arch/blackfin/mach-common/pm.c b/arch/blackfin/mach-common/pm.c
index dac51fb06f22..81930f7d06f1 100644
--- a/arch/blackfin/mach-common/pm.c
+++ b/arch/blackfin/mach-common/pm.c
@@ -77,7 +77,15 @@ void bfin_pm_suspend_standby_enter(void)
 
 		gpio_pm_restore();
 
+#if defined(CONFIG_BF54x) || defined(CONFIG_BF52x)
+		bfin_write_SIC_IWR0(IWR_ENABLE_ALL);
+		bfin_write_SIC_IWR1(IWR_ENABLE_ALL);
+# ifdef CONFIG_BF54x
+		bfin_write_SIC_IWR2(IWR_ENABLE_ALL);
+# endif
+#else
 		bfin_write_SIC_IWR(IWR_ENABLE_ALL);
+#endif
 
 		local_irq_restore(flags);
 	}
@@ -85,7 +93,15 @@ void bfin_pm_suspend_standby_enter(void)
 
 #if defined(CONFIG_PM_WAKEUP_GPIO_BY_SIC_IWR)
 	sleep_deeper(CONFIG_PM_WAKEUP_SIC_IWR);
+# if defined(CONFIG_BF54x) || defined(CONFIG_BF52x)
+	bfin_write_SIC_IWR0(IWR_ENABLE_ALL);
+	bfin_write_SIC_IWR1(IWR_ENABLE_ALL);
+#  ifdef CONFIG_BF54x
+	bfin_write_SIC_IWR2(IWR_ENABLE_ALL);
+#  endif
+# else
 	bfin_write_SIC_IWR(IWR_ENABLE_ALL);
+# endif
 #endif				/* CONFIG_PM_WAKEUP_GPIO_BY_SIC_IWR */
 }
 
diff --git a/include/asm-blackfin/mach-bf548/irq.h b/include/asm-blackfin/mach-bf548/irq.h
index 9fb7bc5399a8..c34507a3f1df 100644
--- a/include/asm-blackfin/mach-bf548/irq.h
+++ b/include/asm-blackfin/mach-bf548/irq.h
@@ -88,7 +88,7 @@ Events         (highest priority)  EMU         0
 #define IRQ_PINT1		BFIN_IRQ(20)	/* PINT1 Interrupt */
 #define IRQ_MDMAS0		BFIN_IRQ(21)	/* MDMA Stream 0 Interrupt */
 #define IRQ_MDMAS1		BFIN_IRQ(22)	/* MDMA Stream 1 Interrupt */
-#define IRQ_WATCHDOG		BFIN_IRQ(23)	/* Watchdog Interrupt */
+#define IRQ_WATCH		BFIN_IRQ(23)	/* Watchdog Interrupt */
 #define IRQ_DMAC1_ERROR		BFIN_IRQ(24)	/* DMAC1 Status (Error) Interrupt */
 #define IRQ_SPORT2_ERROR	BFIN_IRQ(25)	/* SPORT2 Error Interrupt */
 #define IRQ_SPORT3_ERROR	BFIN_IRQ(26)	/* SPORT3 Error Interrupt */
@@ -406,7 +406,7 @@ Events         (highest priority)  EMU         0
 #define IRQ_PINT1_POS		16
 #define IRQ_MDMAS0_POS		20
 #define IRQ_MDMAS1_POS		24
-#define IRQ_WATCHDOG_POS	28
+#define IRQ_WATCH_POS		28
 
 /* IAR3 BIT FIELDS */
 #define IRQ_DMAC1_ERR_POS	0
diff --git a/include/asm-blackfin/mach-bf548/mem_init.h b/include/asm-blackfin/mach-bf548/mem_init.h
index 0cb279e973d7..e792eb7ae6a0 100644
--- a/include/asm-blackfin/mach-bf548/mem_init.h
+++ b/include/asm-blackfin/mach-bf548/mem_init.h
@@ -30,6 +30,7 @@
  */
 
 #if (CONFIG_MEM_MT46V32M16)
+#endif
 
 #if defined CONFIG_CLKIN_HALF
 #define CLKIN_HALF       1
-- 
cgit v1.2.3


From b03b08ba9c7235861adf4dde712dade0bb756fe0 Mon Sep 17 00:00:00 2001
From: Robin Getz <robin.getz@analog.com>
Date: Sun, 23 Dec 2007 22:57:01 +0800
Subject: [Blackfin] arch: Clean up dump_bfin_mem

Clean up dump_bfin_mem so that it will display
content from the kernel, as well as l1 instruction, when deferred
HW errors happen, print out the last frame info if it makes sense.

Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/kernel/early_printk.c |   2 +-
 arch/blackfin/kernel/process.c      |   1 +
 arch/blackfin/kernel/traps.c        | 132 +++++++++++++++++++++++-------------
 include/asm-blackfin/bfin-global.h  |   2 +-
 4 files changed, 87 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/kernel/early_printk.c b/arch/blackfin/kernel/early_printk.c
index 724f4a5a1d46..c892777267d5 100644
--- a/arch/blackfin/kernel/early_printk.c
+++ b/arch/blackfin/kernel/early_printk.c
@@ -205,7 +205,7 @@ asmlinkage void __init early_trap_c(struct pt_regs *fp, void *retaddr)
 	if (likely(early_console == NULL))
 		setup_early_printk(DEFAULT_EARLY_PORT);
 
-	dump_bfin_mem((void *)fp->retx);
+	dump_bfin_mem(fp);
 	show_regs(fp);
 	dump_bfin_trace_buffer();
 
diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 5bf15125f0d6..fff3dd9eadda 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -327,6 +327,7 @@ void finish_atomic_sections (struct pt_regs *regs)
 }
 
 #if defined(CONFIG_ACCESS_CHECK)
+/* Return 1 if access to memory range is OK, 0 otherwise */
 int _access_ok(unsigned long addr, unsigned long size)
 {
 	if (size == 0)
diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c
index 8bbfef31666b..c90f16825f98 100644
--- a/arch/blackfin/kernel/traps.c
+++ b/arch/blackfin/kernel/traps.c
@@ -39,6 +39,7 @@
 #include <linux/irq.h>
 #include <asm/trace.h>
 #include <asm/fixed_code.h>
+#include <asm/dma.h>
 
 #ifdef CONFIG_KGDB
 # include <linux/debugger.h>
@@ -171,7 +172,7 @@ asmlinkage void double_fault_c(struct pt_regs *fp)
 	oops_in_progress = 1;
 	printk(KERN_EMERG "\n" KERN_EMERG "Double Fault\n");
 	dump_bfin_process(fp);
-	dump_bfin_mem((void *)fp->retx);
+	dump_bfin_mem(fp);
 	show_regs(fp);
 	panic("Double Fault - unrecoverable event\n");
 
@@ -196,6 +197,10 @@ asmlinkage void trap_c(struct pt_regs *fp)
 	 * we will kernel panic, so the system reboots.
 	 * If KGDB is enabled, don't set this for kernel breakpoints
 	*/
+
+	/* TODO: check to see if we are in some sort of deferred HWERR
+	 * that we should be able to recover from, not kernel panic
+	 */
 	if ((bfin_read_IPEND() & 0xFFC0)
 #ifdef CONFIG_KGDB
 		&& trapnr != VEC_EXCPT02
@@ -478,11 +483,7 @@ asmlinkage void trap_c(struct pt_regs *fp)
 	if (sig != SIGTRAP) {
 		unsigned long stack;
 		dump_bfin_process(fp);
-		/* Is it an interrupt, or an exception? */
-		if (trapnr == VEC_HWERR)
-			dump_bfin_mem((void *)fp->pc);
-		else
-			dump_bfin_mem((void *)fp->retx);
+		dump_bfin_mem(fp);
 		show_regs(fp);
 
 		/* Print out the trace buffer if it makes sense */
@@ -644,8 +645,10 @@ void dump_bfin_process(struct pt_regs *fp)
 	if (oops_in_progress)
 		printk(KERN_EMERG "Kernel OOPS in progress\n");
 
-	if (context & 0x0020)
-		printk(KERN_NOTICE "Deferred excecption or HW Error context\n");
+	if (context & 0x0020 && (fp->seqstat & SEQSTAT_EXCAUSE) == VEC_HWERR)
+		printk(KERN_NOTICE "HW Error context\n");
+	else if (context & 0x0020)
+		printk(KERN_NOTICE "Defered Exception context\n");
 	else if (context & 0x3FC0)
 		printk(KERN_NOTICE "Interrupt context\n");
 	else if (context & 0x4000)
@@ -673,49 +676,82 @@ void dump_bfin_process(struct pt_regs *fp)
 		     "No Valid process in current context\n");
 }
 
-void dump_bfin_mem(void *retaddr)
+void dump_bfin_mem(struct pt_regs *fp)
 {
+	unsigned short *addr, *erraddr, val = 0, err = 0;
+	char sti = 0, buf[6];
 
-	if (retaddr >= (void *)FIXED_CODE_START  && retaddr < (void *)physical_mem_end
-#if L1_CODE_LENGTH != 0
-	    /* FIXME: Copy the code out of L1 Instruction SRAM through dma
-	       memcpy.  */
-	    && !(retaddr >= (void *)L1_CODE_START
-	         && retaddr < (void *)(L1_CODE_START + L1_CODE_LENGTH))
-#endif
-	) {
-		int i = ((unsigned int)retaddr & 0xFFFFFFF0) - 32;
-		unsigned short x = 0;
-		printk(KERN_NOTICE "return address: [0x%p]; contents of:", retaddr);
-		for (; i < ((unsigned int)retaddr & 0xFFFFFFF0) + 32; i += 2) {
-			if (!(i & 0xF))
-				printk("\n" KERN_NOTICE "0x%08x: ", i);
-
-			if (get_user(x, (unsigned short *)i))
-				break;
+	if (unlikely((fp->seqstat & SEQSTAT_EXCAUSE) == VEC_HWERR))
+		erraddr = (void *)fp->pc;
+	else
+		erraddr = (void *)fp->retx;
+
+	printk(KERN_NOTICE "return address: [0x%p]; contents of:", erraddr);
+
+	for (addr = (unsigned short *)((unsigned long)erraddr & ~0xF) - 0x10;
+	     addr < (unsigned short *)((unsigned long)erraddr & ~0xF) + 0x10;
+	     addr++) {
+		if (!((unsigned long)addr & 0xF))
+			printk("\n" KERN_NOTICE "0x%p: ", addr);
+
+		if (get_user(val, addr)) {
+			if (addr >= (unsigned short *)L1_CODE_START &&
+			    addr < (unsigned short *)(L1_CODE_START + L1_CODE_LENGTH)) {
+				dma_memcpy(&val, addr, sizeof(val));
+				sprintf(buf, "%04x", val);
+			} else if (addr >= (unsigned short *)FIXED_CODE_START &&
+				addr <= (unsigned short *)memory_start) {
+				val = bfin_read16(addr);
+				sprintf(buf, "%04x", val);
+			} else {
+				val = 0;
+				sprintf(buf, "????");
+			}
+		} else
+			sprintf(buf, "%04x", val);
+
+		if (addr == erraddr) {
+			printk("[%s]", buf);
+			err = val;
+		} else
+			printk(" %s ", buf);
+
+		/* Do any previous instructions turn on interrupts? */
+		if (addr <= erraddr &&				/* in the past */
+		    ((val >= 0x0040 && val <= 0x0047) ||	/* STI instruction */
+		      val == 0x017b))				/* [SP++] = RETI */
+			sti = 1;
+	}
+
+	printk("\n");
+
+	/* Hardware error interrupts can be deferred */
+	if (unlikely(sti && (fp->seqstat & SEQSTAT_EXCAUSE) == VEC_HWERR &&
+	    oops_in_progress)){
+		printk(KERN_NOTICE "Looks like this was a deferred error - sorry\n");
 #ifndef CONFIG_DEBUG_HWERR
-			/* If one of the last few instructions was a STI
-			 * it is likely that the error occured awhile ago
-			 * and we just noticed. This only happens in kernel
-			 * context, which should mean an oops is happening
-			 */
-			if (oops_in_progress && x >= 0x0040 && x <= 0x0047 && i <= 0)
-				printk(KERN_EMERG "\n"
-					KERN_EMERG "WARNING : You should reconfigure"
-					" the kernel to turn on\n"
-					KERN_EMERG " 'Hardware error interrupt debugging'\n"
-					KERN_EMERG " The rest of this error is meanless\n");
-#endif
-			if (i == (unsigned int)retaddr)
-				printk("[%04x]", x);
-			else
-				printk(" %04x ", x);
+		printk(KERN_NOTICE "The remaining message may be meaningless\n"
+			KERN_NOTICE "You should enable CONFIG_DEBUG_HWERR to get a"
+			 " better idea where it came from\n");
+#else
+		/* If we are handling only one peripheral interrupt
+		 * and current mm and pid are valid, and the last error
+		 * was in that user space process's text area
+		 * print it out - because that is where the problem exists
+		 */
+		if ((!(((fp)->ipend & ~0x30) & (((fp)->ipend & ~0x30) - 1))) &&
+		     (current->pid && current->mm)) {
+			/* And the last RETI points to the current userspace context */
+			if ((fp + 1)->pc >= current->mm->start_code &&
+			    (fp + 1)->pc <= current->mm->end_code) {
+				printk(KERN_NOTICE "It might be better to look around here : \n");
+				printk(KERN_NOTICE "-------------------------------------------\n");
+				show_regs(fp + 1);
+				printk(KERN_NOTICE "-------------------------------------------\n");
+			}
 		}
-		printk("\n");
-	} else
-		printk("\n" KERN_NOTICE
-			"Cannot look at the [PC] <%p> for it is"
-			" in unreadable memory - sorry\n", retaddr);
+#endif
+	}
 }
 
 void show_regs(struct pt_regs *fp)
@@ -885,7 +921,7 @@ void panic_cplb_error(int cplb_panic, struct pt_regs *fp)
 	printk(KERN_EMERG "DCPLB_FAULT_ADDR=%p\n", (void *)bfin_read_DCPLB_FAULT_ADDR());
 	printk(KERN_EMERG "ICPLB_FAULT_ADDR=%p\n", (void *)bfin_read_ICPLB_FAULT_ADDR());
 	dump_bfin_process(fp);
-	dump_bfin_mem((void *)fp->retx);
+	dump_bfin_mem(fp);
 	show_regs(fp);
 	dump_stack();
 	panic("Unrecoverable event\n");
diff --git a/include/asm-blackfin/bfin-global.h b/include/asm-blackfin/bfin-global.h
index 39bdd86871cf..6ae0619d7696 100644
--- a/include/asm-blackfin/bfin-global.h
+++ b/include/asm-blackfin/bfin-global.h
@@ -51,7 +51,7 @@ extern unsigned long sclk_to_usecs(unsigned long sclk);
 extern unsigned long usecs_to_sclk(unsigned long usecs);
 
 extern void dump_bfin_process(struct pt_regs *regs);
-extern void dump_bfin_mem(void *retaddr);
+extern void dump_bfin_mem(struct pt_regs *regs);
 extern void dump_bfin_trace_buffer(void);
 
 extern int init_arch_irq(void);
-- 
cgit v1.2.3


From 971d5bc4e5c75bfc4466deaff09839cd6f918eca Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sun, 27 Jan 2008 16:32:31 +0800
Subject: [Blackfin] arch: Fix bug to Enable bf548 to Re-program Clocks while
 Kernel boots.

Reprogram DDR EBIU register properly for bf548.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/Kconfig                           | 19 +++++++-
 arch/blackfin/mach-bf548/head.S                 | 47 +++++++++++++++++++
 include/asm-blackfin/mach-bf548/defBF54x_base.h | 31 ++++++++++++-
 include/asm-blackfin/mach-bf548/mem_init.h      | 61 ++++++++++++++++++++++++-
 4 files changed, 153 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 5ebcfd226ed8..1364dcaccc18 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -317,7 +317,7 @@ config VCO_MULT
 	range 1 64
 	default "22" if BFIN533_EZKIT
 	default "45" if BFIN533_STAMP
-	default "20" if (BFIN537_STAMP || BFIN527_EZKIT)
+	default "20" if (BFIN537_STAMP || BFIN527_EZKIT || BFIN548_EZKIT)
 	default "22" if BFIN533_BLUETECHNIX_CM
 	default "20" if BFIN537_BLUETECHNIX_CM
 	default "20" if BFIN561_BLUETECHNIX_CM
@@ -354,7 +354,7 @@ config SCLK_DIV
 	range 1 15
 	default 5 if BFIN533_EZKIT
 	default 5 if BFIN533_STAMP
-	default 4 if (BFIN537_STAMP || BFIN527_EZKIT)
+	default 4 if (BFIN537_STAMP || BFIN527_EZKIT || BFIN548_EZKIT)
 	default 5 if BFIN533_BLUETECHNIX_CM
 	default 4 if BFIN537_BLUETECHNIX_CM
 	default 4 if BFIN561_BLUETECHNIX_CM
@@ -409,6 +409,7 @@ config MEM_SIZE
 	default  32 if BFIN533_EZKIT
 	default  64 if BFIN527_EZKIT
 	default  64 if BFIN537_STAMP
+	default  64 if BFIN548_EZKIT
 	default  64 if BFIN561_EZKIT
 	default 128 if BFIN533_STAMP
 	default  64 if PNAV10
@@ -416,6 +417,7 @@ config MEM_SIZE
 
 config MEM_ADD_WIDTH
 	int "SDRAM Memory Address Width"
+	depends on (!BF54x)
 	default  9 if BFIN533_EZKIT
 	default  9 if BFIN561_EZKIT
 	default  9 if H8606_HVSISTEMAS
@@ -424,6 +426,19 @@ config MEM_ADD_WIDTH
 	default 11 if BFIN533_STAMP
 	default 10 if PNAV10
 
+
+choice
+	prompt "DDR SDRAM Chip Type"
+	depends on BFIN548_EZKIT
+	default MEM_MT46V32M16_5B
+
+config MEM_MT46V32M16_6T
+        bool "MT46V32M16_6T"
+
+config MEM_MT46V32M16_5B
+        bool "MT46V32M16_5B"
+endchoice
+
 config ENET_FLASH_PIN
 	int "PF port/pin used for flash and ethernet sharing"
 	depends on (BFIN533_STAMP)
diff --git a/arch/blackfin/mach-bf548/head.S b/arch/blackfin/mach-bf548/head.S
index 745662e88759..74fe258421a5 100644
--- a/arch/blackfin/mach-bf548/head.S
+++ b/arch/blackfin/mach-bf548/head.S
@@ -324,12 +324,25 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
+#if defined(CONFIG_BF54x)
+	P2.H = hi(EBIU_RSTCTL);
+	P2.L = lo(EBIU_RSTCTL);
+	R0 = [P2];
+	BITSET (R0, 3);
+#else
 	P2.H = hi(EBIU_SDGCTL);
 	P2.L = lo(EBIU_SDGCTL);
 	R0 = [P2];
 	BITSET (R0, 24);
+#endif
 	[P2] = R0;
 	SSYNC;
+#if defined(CONFIG_BF54x)
+.LSRR_MODE:
+	R0 = [P2];
+	CC = BITTST(R0, 4);
+	if !CC JUMP .LSRR_MODE;
+#endif
 
 	r0 = CONFIG_VCO_MULT & 63;       /* Load the VCO multiplier         */
 	r0 = r0 << 9;                    /* Shift it over,                  */
@@ -361,6 +374,39 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
+#if defined(CONFIG_BF54x)
+	P2.H = hi(EBIU_RSTCTL);
+	P2.L = lo(EBIU_RSTCTL);
+	R0 = [P2];
+	CC = BITTST(R0, 0);
+	if CC jump .Lskipddrrst;
+	BITSET (R0, 0);
+.Lskipddrrst:
+	BITCLR (R0, 3);
+	[P2] = R0;
+	SSYNC;
+
+	p0.l = lo(EBIU_DDRCTL0);
+	p0.h = hi(EBIU_DDRCTL0);
+	r0.l = lo(mem_DDRCTL0);
+	r0.h = hi(mem_DDRCTL0);
+	[p0] = r0;
+	ssync;
+
+	p0.l = lo(EBIU_DDRCTL1);
+	p0.h = hi(EBIU_DDRCTL1);
+	r0.l = lo(mem_DDRCTL1);
+	r0.h = hi(mem_DDRCTL1);
+	[p0] = r0;
+	ssync;
+
+	p0.l = lo(EBIU_DDRCTL2);
+	p0.h = hi(EBIU_DDRCTL2);
+	r0.l = lo(mem_DDRCTL2);
+	r0.h = hi(mem_DDRCTL2);
+	[p0] = r0;
+	ssync;
+#else
 	p0.l = lo(EBIU_SDRRC);
 	p0.h = hi(EBIU_SDRRC);
 	r0 = mem_SDRRC;
@@ -394,6 +440,7 @@ ENTRY(_start_dma_code)
 	R1 = R1 | R0;
 	[P2] = R1;
 	SSYNC;
+#endif
 
 	p0.h = hi(SIC_IWR0);
 	p0.l = lo(SIC_IWR0);
diff --git a/include/asm-blackfin/mach-bf548/defBF54x_base.h b/include/asm-blackfin/mach-bf548/defBF54x_base.h
index 319a48590c9c..08f90c21fe8a 100644
--- a/include/asm-blackfin/mach-bf548/defBF54x_base.h
+++ b/include/asm-blackfin/mach-bf548/defBF54x_base.h
@@ -1772,17 +1772,36 @@
 #define                       TRP  0x3c0000   /* Pre charge-to-active command period */
 #define                      TRAS  0x3c00000  /* Min Active-to-pre charge time */
 #define                       TRC  0x3c000000 /* Active-to-active time */
+#define DDR_TRAS(x)		((x<<22)&TRAS)	/* DDR tRAS = (1~15) cycles */
+#define DDR_TRP(x)		((x<<18)&TRP)	/* DDR tRP = (1~15) cycles */
+#define DDR_TRC(x)		((x<<26)&TRC)	/* DDR tRC = (1~15) cycles */
+#define DDR_TRFC(x)		((x<<14)&TRFC)	/* DDR tRFC = (1~15) cycles */
+#define DDR_TREFI(x)		(x&TREFI)	/* DDR tRFC = (1~15) cycles */
 
 /* Bit masks for EBIU_DDRCTL1 */
 
 #define                      TRCD  0xf        /* Active-to-Read/write delay */
-#define                       MRD  0xf0       /* Mode register set to active */
+#define                      TMRD  0xf0       /* Mode register set to active */
 #define                       TWR  0x300      /* Write Recovery time */
 #define               DDRDATWIDTH  0x3000     /* DDR data width */
 #define                  EXTBANKS  0xc000     /* External banks */
 #define               DDRDEVWIDTH  0x30000    /* DDR device width */
 #define                DDRDEVSIZE  0xc0000    /* DDR device size */
-#define                     TWWTR  0xf0000000 /* Write-to-read delay */
+#define                      TWTR  0xf0000000 /* Write-to-read delay */
+#define DDR_TWTR(x)		((x<<28)&TWTR)	/* DDR tWTR = (1~15) cycles */
+#define DDR_TMRD(x)		((x<<4)&TMRD)	/* DDR tMRD = (1~15) cycles */
+#define DDR_TWR(x)		((x<<8)&TWR)	/* DDR tWR = (1~15) cycles */
+#define DDR_TRCD(x)		(x&TRCD)	/* DDR tRCD = (1~15) cycles */
+#define DDR_DATWIDTH		0x2000		/* DDR data width */
+#define EXTBANK_1		0		/* 1 external bank */
+#define EXTBANK_2		0x4000		/* 2 external banks */
+#define DEVSZ_64		0x40000		/* DDR External Bank Size = 64MB */
+#define DEVSZ_128		0x80000		/* DDR External Bank Size = 128MB */
+#define DEVSZ_256		0xc0000		/* DDR External Bank Size = 256MB */
+#define DEVSZ_512		0		/* DDR External Bank Size = 512MB */
+#define DEVWD_4			0		/* DDR Device Width = 4 Bits    */
+#define DEVWD_8			0x10000		/* DDR Device Width = 8 Bits    */
+#define DEVWD_16		0x20000		/* DDR Device Width = 16 Bits    */
 
 /* Bit masks for EBIU_DDRCTL2 */
 
@@ -1790,6 +1809,10 @@
 #define                CASLATENCY  0x70       /* CAS latency */
 #define                  DLLRESET  0x100      /* DLL Reset */
 #define                      REGE  0x1000     /* Register mode enable */
+#define CL_1_5			0x50		/* DDR CAS Latency = 1.5 cycles */
+#define CL_2			0x20		/* DDR CAS Latency = 2 cycles */
+#define CL_2_5			0x60		/* DDR CAS Latency = 2.5 cycles */
+#define CL_3			0x30		/* DDR CAS Latency = 3 cycles */
 
 /* Bit masks for EBIU_DDRCTL3 */
 
@@ -2257,6 +2280,10 @@
 
 #define                      CSEL  0x30       /* Core Select */
 #define                      SSEL  0xf        /* System Select */
+#define			CSEL_DIV1	0x0000	/* CCLK = VCO / 1 */
+#define			CSEL_DIV2	0x0010	/* CCLK = VCO / 2 */
+#define			CSEL_DIV4	0x0020	/* CCLK = VCO / 4 */
+#define			CSEL_DIV8	0x0030	/* CCLK = VCO / 8 */
 
 /* Bit masks for PLL_CTL */
 
diff --git a/include/asm-blackfin/mach-bf548/mem_init.h b/include/asm-blackfin/mach-bf548/mem_init.h
index e792eb7ae6a0..befc2903d5a5 100644
--- a/include/asm-blackfin/mach-bf548/mem_init.h
+++ b/include/asm-blackfin/mach-bf548/mem_init.h
@@ -28,10 +28,69 @@
  * If not, write to the Free Software Foundation,
  * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  */
+#define MIN_DDR_SCLK(x)	(x*(CONFIG_SCLK_HZ/1000/1000)/1000 + 1)
 
-#if (CONFIG_MEM_MT46V32M16)
+#if (CONFIG_MEM_MT46V32M16_6T)
+#define DDR_SIZE	DEVSZ_512
+#define DDR_WIDTH	DEVWD_16
+
+#define DDR_tRC		DDR_TRC(MIN_DDR_SCLK(60))
+#define DDR_tRAS	DDR_TRAS(MIN_DDR_SCLK(42))
+#define DDR_tRP		DDR_TRP(MIN_DDR_SCLK(15))
+#define DDR_tRFC	DDR_TRFC(MIN_DDR_SCLK(72))
+#define DDR_tREFI	DDR_TREFI(MIN_DDR_SCLK(7800))
+
+#define DDR_tRCD	DDR_TRCD(MIN_DDR_SCLK(15))
+#define DDR_tWTR	DDR_TWTR(1)
+#define DDR_tMRD	DDR_TMRD(MIN_DDR_SCLK(12))
+#define DDR_tWR		DDR_TWR(MIN_DDR_SCLK(15))
+#endif
+
+#if (CONFIG_MEM_MT46V32M16_5B)
+#define DDR_SIZE	DEVSZ_512
+#define DDR_WIDTH	DEVWD_16
+
+#define DDR_tRC		DDR_TRC(MIN_DDR_SCLK(55))
+#define DDR_tRAS	DDR_TRAS(MIN_DDR_SCLK(40))
+#define DDR_tRP		DDR_TRP(MIN_DDR_SCLK(15))
+#define DDR_tRFC	DDR_TRFC(MIN_DDR_SCLK(70))
+#define DDR_tREFI	DDR_TREFI(MIN_DDR_SCLK(7800))
+
+#define DDR_tRCD	DDR_TRCD(MIN_DDR_SCLK(15))
+#define DDR_tWTR	DDR_TWTR(2)
+#define DDR_tMRD	DDR_TMRD(MIN_DDR_SCLK(10))
+#define DDR_tWR		DDR_TWR(MIN_DDR_SCLK(15))
+#endif
+
+#if (CONFIG_MEM_GENERIC_BOARD)
+#define DDR_SIZE	DEVSZ_512
+#define DDR_WIDTH	DEVWD_16
+
+#define DDR_tRCD	DDR_TRCD(3)
+#define DDR_tWTR	DDR_TWTR(2)
+#define DDR_tWR		DDR_TWR(2)
+#define DDR_tMRD	DDR_TMRD(2)
+#define DDR_tRP		DDR_TRP(3)
+#define DDR_tRAS	DDR_TRAS(7)
+#define DDR_tRC		DDR_TRC(10)
+#define DDR_tRFC	DDR_TRFC(12)
+#define DDR_tREFI	DDR_TREFI(1288)
 #endif
 
+#if (CONFIG_SCLK_HZ <= 133333333)
+#define	DDR_CL		CL_2
+#elif (CONFIG_SCLK_HZ <= 166666666)
+#define	DDR_CL		CL_2_5
+#else
+#define	DDR_CL		CL_3
+#endif
+
+#define mem_DDRCTL0	(DDR_tRP | DDR_tRAS | DDR_tRC | DDR_tRFC | DDR_tREFI)
+#define mem_DDRCTL1	(DDR_DATWIDTH | EXTBANK_1 | DDR_SIZE | DDR_WIDTH | DDR_tWTR \
+			| DDR_tMRD | DDR_tWR | DDR_tRCD)
+#define mem_DDRCTL2	DDR_CL
+
+
 #if defined CONFIG_CLKIN_HALF
 #define CLKIN_HALF       1
 #else
-- 
cgit v1.2.3


From 3ca32c1dc62dabe67d7e51f6e8568e3965f52236 Mon Sep 17 00:00:00 2001
From: Bernd Schmidt <bernd.schmidt@analog.com>
Date: Mon, 24 Dec 2007 12:40:29 +0800
Subject: [Blackfin] arch: fix up coding style in uaccess.h

Signed-off-by: Bernd Schmidt <bernd.schmidt@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 include/asm-blackfin/uaccess.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-blackfin/uaccess.h b/include/asm-blackfin/uaccess.h
index 2233f8f9314d..22a410b8003b 100644
--- a/include/asm-blackfin/uaccess.h
+++ b/include/asm-blackfin/uaccess.h
@@ -31,7 +31,7 @@ static inline void set_fs(mm_segment_t fs)
 #define VERIFY_READ	0
 #define VERIFY_WRITE	1
 
-#define access_ok(type,addr,size) _access_ok((unsigned long)(addr),(size))
+#define access_ok(type, addr, size) _access_ok((unsigned long)(addr), (size))
 
 static inline int is_in_rom(unsigned long addr)
 {
-- 
cgit v1.2.3


From fac3cf432ef9b6bfd64b35b95afe0b7e0079da74 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Mon, 24 Dec 2007 20:07:03 +0800
Subject: [Blackfin] arch: Fix gpio label handling

early serial init also utilizes the peripheral request api - however
at this point bfin_gpio_init didn't allocate memory for the labels.
So we always have two zombies (allocated pin functions without labels)

This happens before the initcalls - We now allocate memory statically.
Define MAX_RESOURCES individually for each cpu.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/kernel/bfin_gpio.c          | 31 ++++++++++++++++++-------------
 include/asm-blackfin/mach-bf527/portmux.h |  2 ++
 include/asm-blackfin/mach-bf533/portmux.h |  2 ++
 include/asm-blackfin/mach-bf537/portmux.h |  2 ++
 include/asm-blackfin/mach-bf548/portmux.h |  2 ++
 include/asm-blackfin/mach-bf561/portmux.h |  2 ++
 6 files changed, 28 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c
index ea13d2a22d69..ffee36910288 100644
--- a/arch/blackfin/kernel/bfin_gpio.c
+++ b/arch/blackfin/kernel/bfin_gpio.c
@@ -179,15 +179,13 @@ static struct gpio_port_t *gpio_array[gpio_bank(MAX_BLACKFIN_GPIOS)] = {
 #endif
 
 static unsigned short reserved_gpio_map[gpio_bank(MAX_BLACKFIN_GPIOS)];
-static unsigned short reserved_peri_map[gpio_bank(MAX_BLACKFIN_GPIOS + 16)];
+static unsigned short reserved_peri_map[gpio_bank(MAX_RESOURCES)];
 
-#define MAX_RESOURCES 		256
 #define RESOURCE_LABEL_SIZE 	16
 
-struct str_ident {
+static struct str_ident {
 	char name[RESOURCE_LABEL_SIZE];
-} *str_ident;
-
+} str_ident[MAX_RESOURCES];
 
 #ifdef CONFIG_PM
 static unsigned short wakeup_map[gpio_bank(MAX_BLACKFIN_GPIOS)];
@@ -251,6 +249,11 @@ static char *get_label(unsigned short ident)
 
 static int cmp_label(unsigned short ident, const char *label)
 {
+	if (label == NULL) {
+		dump_stack();
+		printk(KERN_ERR "Please provide none-null label\n");
+	}
+
 	if (label && str_ident)
 		return strncmp(str_ident[ident].name,
 				 label, strlen(label));
@@ -419,12 +422,6 @@ static void default_gpio(unsigned short gpio)
 
 static int __init bfin_gpio_init(void)
 {
-	str_ident = kcalloc(MAX_RESOURCES,
-				 sizeof(struct str_ident), GFP_KERNEL);
-	if (str_ident == NULL)
-		return -ENOMEM;
-
-	memset(str_ident, 0, MAX_RESOURCES * sizeof(struct str_ident));
 
 	printk(KERN_INFO "Blackfin GPIO Controller\n");
 
@@ -785,6 +782,14 @@ void gpio_pm_restore(void)
 }
 
 #endif
+#else /* BF548_FAMILY */
+
+unsigned short get_gpio_dir(unsigned short gpio)
+{
+	return (0x01 & (gpio_array[gpio_bank(gpio)]->port_dir_clear >> gpio_sub_n(gpio)));
+}
+EXPORT_SYMBOL(get_gpio_dir);
+
 #endif /* BF548_FAMILY */
 
 /***********************************************************
@@ -1204,10 +1209,10 @@ static int gpio_proc_read(char *buf, char **start, off_t offset,
 
 	for (c = 0; c < MAX_RESOURCES; c++) {
 		if (!check_gpio(c) && (reserved_gpio_map[gpio_bank(c)] & gpio_bit(c)))
-			len = sprintf(buf, "GPIO_%d: %s \tGPIO %s\n", c,
+			len = sprintf(buf, "GPIO_%d: %s \t\tGPIO %s\n", c,
 				 get_label(c), get_gpio_dir(c) ? "OUTPUT" : "INPUT");
 		else if (reserved_peri_map[gpio_bank(c)] & gpio_bit(c))
-			len = sprintf(buf, "GPIO_%d: %s \tPeripheral\n", c, get_label(c));
+			len = sprintf(buf, "GPIO_%d: %s \t\tPeripheral\n", c, get_label(c));
 		else
 			continue;
 		buf += len;
diff --git a/include/asm-blackfin/mach-bf527/portmux.h b/include/asm-blackfin/mach-bf527/portmux.h
index dcf001adc63c..ae4d205bfcf5 100644
--- a/include/asm-blackfin/mach-bf527/portmux.h
+++ b/include/asm-blackfin/mach-bf527/portmux.h
@@ -1,6 +1,8 @@
 #ifndef _MACH_PORTMUX_H_
 #define _MACH_PORTMUX_H_
 
+#define MAX_RESOURCES 	MAX_BLACKFIN_GPIOS
+
 #define P_PPI0_D0	(P_DEFINED | P_IDENT(GPIO_PF0) | P_FUNCT(0))
 #define P_PPI0_D1	(P_DEFINED | P_IDENT(GPIO_PF1) | P_FUNCT(0))
 #define P_PPI0_D2	(P_DEFINED | P_IDENT(GPIO_PF2) | P_FUNCT(0))
diff --git a/include/asm-blackfin/mach-bf533/portmux.h b/include/asm-blackfin/mach-bf533/portmux.h
index 137f4884acfe..685a2651dcda 100644
--- a/include/asm-blackfin/mach-bf533/portmux.h
+++ b/include/asm-blackfin/mach-bf533/portmux.h
@@ -1,6 +1,8 @@
 #ifndef _MACH_PORTMUX_H_
 #define _MACH_PORTMUX_H_
 
+#define MAX_RESOURCES 	MAX_BLACKFIN_GPIOS
+
 #define P_PPI0_CLK	(P_DONTCARE)
 #define P_PPI0_FS1	(P_DONTCARE)
 #define P_PPI0_FS2	(P_DONTCARE)
diff --git a/include/asm-blackfin/mach-bf537/portmux.h b/include/asm-blackfin/mach-bf537/portmux.h
index 5a3f7d3bf73d..78fee6e0f237 100644
--- a/include/asm-blackfin/mach-bf537/portmux.h
+++ b/include/asm-blackfin/mach-bf537/portmux.h
@@ -1,6 +1,8 @@
 #ifndef _MACH_PORTMUX_H_
 #define _MACH_PORTMUX_H_
 
+#define MAX_RESOURCES 	(MAX_BLACKFIN_GPIOS + GPIO_BANKSIZE)	/* We additionally handle PORTJ */
+
 #define P_UART0_TX	(P_DEFINED | P_IDENT(GPIO_PF0) | P_FUNCT(0))
 #define P_UART0_RX	(P_DEFINED | P_IDENT(GPIO_PF1) | P_FUNCT(0))
 #define P_UART1_TX	(P_DEFINED | P_IDENT(GPIO_PF2) | P_FUNCT(0))
diff --git a/include/asm-blackfin/mach-bf548/portmux.h b/include/asm-blackfin/mach-bf548/portmux.h
index 6b485120015f..8177a567dcdb 100644
--- a/include/asm-blackfin/mach-bf548/portmux.h
+++ b/include/asm-blackfin/mach-bf548/portmux.h
@@ -1,6 +1,8 @@
 #ifndef _MACH_PORTMUX_H_
 #define _MACH_PORTMUX_H_
 
+#define MAX_RESOURCES 	MAX_BLACKFIN_GPIOS
+
 #define P_SPORT2_TFS	(P_DEFINED | P_IDENT(GPIO_PA0) | P_FUNCT(0))
 #define P_SPORT2_DTSEC	(P_DEFINED | P_IDENT(GPIO_PA1) | P_FUNCT(0))
 #define P_SPORT2_DTPRI	(P_DEFINED | P_IDENT(GPIO_PA2) | P_FUNCT(0))
diff --git a/include/asm-blackfin/mach-bf561/portmux.h b/include/asm-blackfin/mach-bf561/portmux.h
index 132ad31665e3..a6ee8206efb6 100644
--- a/include/asm-blackfin/mach-bf561/portmux.h
+++ b/include/asm-blackfin/mach-bf561/portmux.h
@@ -1,6 +1,8 @@
 #ifndef _MACH_PORTMUX_H_
 #define _MACH_PORTMUX_H_
 
+#define MAX_RESOURCES 	MAX_BLACKFIN_GPIOS
+
 #define P_PPI0_CLK	(P_DONTCARE)
 #define P_PPI0_FS1	(P_DONTCARE)
 #define P_PPI0_FS2	(P_DONTCARE)
-- 
cgit v1.2.3


From 7cc1c4b2c44d7807f55da6a36f5b2e49977c67b7 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <michael.frysinger@analog.com>
Date: Mon, 24 Dec 2007 20:05:09 +0800
Subject: [Blackfin] arch: update to latest anomaly sheets

Signed-off-by: Mike Frysinger <michael.frysinger@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 include/asm-blackfin/mach-bf533/anomaly.h | 12 ++++---
 include/asm-blackfin/mach-bf537/anomaly.h | 14 +++++---
 include/asm-blackfin/mach-bf548/anomaly.h | 54 +++++++++++++++++++------------
 include/asm-blackfin/mach-bf561/anomaly.h | 18 +++++++++--
 4 files changed, 67 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/asm-blackfin/mach-bf533/anomaly.h b/include/asm-blackfin/mach-bf533/anomaly.h
index f36ff5af1b91..98209d40abba 100644
--- a/include/asm-blackfin/mach-bf533/anomaly.h
+++ b/include/asm-blackfin/mach-bf533/anomaly.h
@@ -7,9 +7,7 @@
  */
 
 /* This file shoule be up to date with:
- *  - Revision X,  March 23, 2007; ADSP-BF533 Blackfin Processor Anomaly List
- *  - Revision AB, March 23, 2007; ADSP-BF532 Blackfin Processor Anomaly List
- *  - Revision W,  March 23, 2007; ADSP-BF531 Blackfin Processor Anomaly List
+ *  - Revision B, 12/10/2007; ADSP-BF531/BF532/BF533 Blackfin Processor Anomaly List
  */
 
 #ifndef _MACH_ANOMALY_H_
@@ -17,7 +15,7 @@
 
 /* We do not support 0.1 or 0.2 silicon - sorry */
 #if __SILICON_REVISION__ < 3
-# error Kernel will not work on BF533 silicon version 0.0, 0.1, or 0.2
+# error will not work on BF533 silicon version 0.0, 0.1, or 0.2
 #endif
 
 #if defined(__ADSPBF531__)
@@ -251,6 +249,12 @@
 #define ANOMALY_05000192 (__SILICON_REVISION__ < 3)
 /* Internal Voltage Regulator may not start up */
 #define ANOMALY_05000206 (__SILICON_REVISION__ < 3)
+/* Serial Port (SPORT) Multichannel Transmit Failure when Channel 0 Is Disabled */
+#define ANOMALY_05000357 (1)
+/* PPI Underflow Error Goes Undetected in ITU-R 656 Mode */
+#define ANOMALY_05000366 (1)
+/* Possible RETS Register Corruption when Subroutine Is under 5 Cycles in Duration */
+#define ANOMALY_05000371 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000266 (0)
diff --git a/include/asm-blackfin/mach-bf537/anomaly.h b/include/asm-blackfin/mach-bf537/anomaly.h
index 2b66ecf489f7..746a794b3119 100644
--- a/include/asm-blackfin/mach-bf537/anomaly.h
+++ b/include/asm-blackfin/mach-bf537/anomaly.h
@@ -7,9 +7,7 @@
  */
 
 /* This file shoule be up to date with:
- *  - Revision M, March 13, 2007; ADSP-BF537 Blackfin Processor Anomaly List
- *  - Revision L, March 13, 2007; ADSP-BF536 Blackfin Processor Anomaly List
- *  - Revision M, March 13, 2007; ADSP-BF534 Blackfin Processor Anomaly List
+ *  - Revision A, 09/04/2007; ADSP-BF534/ADSP-BF536/ADSP-BF537 Blackfin Processor Anomaly List
  */
 
 #ifndef _MACH_ANOMALY_H_
@@ -17,7 +15,7 @@
 
 /* We do not support 0.1 silicon - sorry */
 #if __SILICON_REVISION__ < 2
-# error Kernel will not work on BF537 silicon version 0.0 or 0.1
+# error will not work on BF537 silicon version 0.0 or 0.1
 #endif
 
 #if defined(__ADSPBF534__)
@@ -44,6 +42,8 @@
 #define ANOMALY_05000122 (1)
 /* Killed 32-bit MMR write leads to next system MMR access thinking it should be 32-bit */
 #define ANOMALY_05000157 (__SILICON_REVISION__ < 2)
+/* Turning SPORTs on while External Frame Sync Is Active May Corrupt Data */
+#define ANOMALY_05000167 (1)
 /* PPI_DELAY not functional in PPI modes with 0 frame syncs */
 #define ANOMALY_05000180 (1)
 /* Instruction Cache Is Not Functional */
@@ -130,6 +130,12 @@
 #define ANOMALY_05000321 (__SILICON_REVISION__ < 3)
 /* EMAC RMII mode at 10-Base-T speed: RX frames not received properly */
 #define ANOMALY_05000322 (1)
+/* Ethernet MAC MDIO Reads Do Not Meet IEEE Specification */
+#define ANOMALY_05000341 (__SILICON_REVISION__ >= 3)
+/* Serial Port (SPORT) Multichannel Transmit Failure when Channel 0 Is Disabled */
+#define ANOMALY_05000357 (1)
+/* DMAs that Go Urgent during Tight Core Writes to External Memory Are Blocked */
+#define ANOMALY_05000359 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000125 (0)
diff --git a/include/asm-blackfin/mach-bf548/anomaly.h b/include/asm-blackfin/mach-bf548/anomaly.h
index c5b63759cdee..850dc12eb7f2 100644
--- a/include/asm-blackfin/mach-bf548/anomaly.h
+++ b/include/asm-blackfin/mach-bf548/anomaly.h
@@ -7,7 +7,7 @@
  */
 
 /* This file shoule be up to date with:
- *  - Revision C, July 16, 2007; ADSP-BF549 Silicon Anomaly List
+ *  - Revision E, 11/28/2007; ADSP-BF542/BF544/BF547/BF548/BF549 Blackfin Processor Anomaly List
  */
 
 #ifndef _MACH_ANOMALY_H_
@@ -26,47 +26,59 @@
 /* Certain Data Cache Writethrough Modes Fail for Vddint <= 0.9V */
 #define ANOMALY_05000272 (1)
 /* False Hardware Error Exception when ISR context is not restored */
-#define ANOMALY_05000281 (1)
+#define ANOMALY_05000281 (__SILICON_REVISION__ < 1)
 /* SSYNCs After Writes To CAN/DMA MMR Registers Are Not Always Handled Correctly */
-#define ANOMALY_05000304 (1)
+#define ANOMALY_05000304 (__SILICON_REVISION__ < 1)
 /* False Hardware Errors Caused by Fetches at the Boundary of Reserved Memory */
 #define ANOMALY_05000310 (1)
 /* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
-#define ANOMALY_05000312 (1)
+#define ANOMALY_05000312 (__SILICON_REVISION__ < 1)
 /* TWI Slave Boot Mode Is Not Functional */
-#define ANOMALY_05000324 (1)
+#define ANOMALY_05000324 (__SILICON_REVISION__ < 1)
 /* External FIFO Boot Mode Is Not Functional */
-#define ANOMALY_05000325 (1)
+#define ANOMALY_05000325 (__SILICON_REVISION__ < 1)
 /* Data Lost When Core and DMA Accesses Are Made to the USB FIFO Simultaneously */
-#define ANOMALY_05000327 (1)
+#define ANOMALY_05000327 (__SILICON_REVISION__ < 1)
 /* Incorrect Access of OTP_STATUS During otp_write() Function */
-#define ANOMALY_05000328 (1)
+#define ANOMALY_05000328 (__SILICON_REVISION__ < 1)
 /* Synchronous Burst Flash Boot Mode Is Not Functional */
-#define ANOMALY_05000329 (1)
+#define ANOMALY_05000329 (__SILICON_REVISION__ < 1)
 /* Host DMA Boot Mode Is Not Functional */
-#define ANOMALY_05000330 (1)
+#define ANOMALY_05000330 (__SILICON_REVISION__ < 1)
 /* Inadequate Timing Margins on DDR DQS to DQ and DQM Skew */
-#define ANOMALY_05000334 (1)
+#define ANOMALY_05000334 (__SILICON_REVISION__ < 1)
 /* Inadequate Rotary Debounce Logic Duration */
-#define ANOMALY_05000335 (1)
+#define ANOMALY_05000335 (__SILICON_REVISION__ < 1)
 /* Phantom Interrupt Occurs After First Configuration of Host DMA Port */
-#define ANOMALY_05000336 (1)
+#define ANOMALY_05000336 (__SILICON_REVISION__ < 1)
 /* Disallowed Configuration Prevents Subsequent Allowed Configuration on Host DMA Port */
-#define ANOMALY_05000337 (1)
+#define ANOMALY_05000337 (__SILICON_REVISION__ < 1)
 /* Slave-Mode SPI0 MISO Failure With CPHA = 0 */
-#define ANOMALY_05000338 (1)
+#define ANOMALY_05000338 (__SILICON_REVISION__ < 1)
 /* If Memory Reads Are Enabled on SDH or HOSTDP, Other DMAC1 Peripherals Cannot Read */
-#define ANOMALY_05000340 (1)
+#define ANOMALY_05000340 (__SILICON_REVISION__ < 1)
 /* Boot Host Wait (HWAIT) and Boot Host Wait Alternate (HWAITA) Signals Are Swapped */
-#define ANOMALY_05000344 (1)
+#define ANOMALY_05000344 (__SILICON_REVISION__ < 1)
 /* USB Calibration Value Is Not Intialized */
-#define ANOMALY_05000346 (1)
+#define ANOMALY_05000346 (__SILICON_REVISION__ < 1)
 /* Boot ROM Kernel Incorrectly Alters Reset Value of USB Register */
-#define ANOMALY_05000347 (1)
+#define ANOMALY_05000347 (__SILICON_REVISION__ < 1)
 /* Data Lost when Core Reads SDH Data FIFO */
-#define ANOMALY_05000349 (1)
+#define ANOMALY_05000349 (__SILICON_REVISION__ < 1)
 /* PLL Status Register Is Inaccurate */
-#define ANOMALY_05000351 (1)
+#define ANOMALY_05000351 (__SILICON_REVISION__ < 1)
+/* Serial Port (SPORT) Multichannel Transmit Failure when Channel 0 Is Disabled */
+#define ANOMALY_05000357 (1)
+/* External Memory Read Access Hangs Core With PLL Bypass */
+#define ANOMALY_05000360 (1)
+/* DMAs that Go Urgent during Tight Core Writes to External Memory Are Blocked */
+#define ANOMALY_05000365 (1)
+/* Addressing Conflict between Boot ROM and Asynchronous Memory */
+#define ANOMALY_05000369 (1)
+/* Mobile DDR Operation Not Functional */
+#define ANOMALY_05000377 (1)
+/* Security/Authentication Speedpath Causes Authentication To Fail To Initiate */
+#define ANOMALY_05000378 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000125 (0)
diff --git a/include/asm-blackfin/mach-bf561/anomaly.h b/include/asm-blackfin/mach-bf561/anomaly.h
index bed956456884..0c1d46193939 100644
--- a/include/asm-blackfin/mach-bf561/anomaly.h
+++ b/include/asm-blackfin/mach-bf561/anomaly.h
@@ -7,7 +7,7 @@
  */
 
 /* This file shoule be up to date with:
- *  - Revision N, March 28, 2007; ADSP-BF561 Silicon Anomaly List
+ *  - Revision O, 11/15/2007; ADSP-BF561 Blackfin Processor Anomaly List
  */
 
 #ifndef _MACH_ANOMALY_H_
@@ -15,7 +15,7 @@
 
 /* We do not support 0.1, 0.2, or 0.4 silicon - sorry */
 #if __SILICON_REVISION__ < 3 || __SILICON_REVISION__ == 4
-# error Kernel will not work on BF561 silicon version 0.0, 0.1, 0.2, or 0.4
+# error will not work on BF561 silicon version 0.0, 0.1, 0.2, or 0.4
 #endif
 
 /* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot 2 Not Supported */
@@ -208,6 +208,8 @@
 #define ANOMALY_05000275 (__SILICON_REVISION__ > 2)
 /* Timing Requirements Change for External Frame Sync PPI Modes with Non-Zero PPI_DELAY */
 #define ANOMALY_05000276 (__SILICON_REVISION__ < 5)
+/* Writes to an I/O data register one SCLK cycle after an edge is detected may clear interrupt */
+#define ANOMALY_05000277 (__SILICON_REVISION__ < 3)
 /* Disabling Peripherals with DMA Running May Cause DMA System Instability */
 #define ANOMALY_05000278 (__SILICON_REVISION__ < 5)
 /* False Hardware Error Exception When ISR Context Is Not Restored */
@@ -246,6 +248,18 @@
 #define ANOMALY_05000332 (__SILICON_REVISION__ < 5)
 /* Flag Data Register Writes One SCLK Cycle After Edge Is Detected May Clear Interrupt Status */
 #define ANOMALY_05000333 (__SILICON_REVISION__ < 5)
+/* New Feature: Additional PPI Frame Sync Sampling Options (Not Available on Older Silicon) */
+#define ANOMALY_05000339 (__SILICON_REVISION__ < 5)
+/* Memory DMA FIFO Causes Throughput Degradation on Writes to External Memory */
+#define ANOMALY_05000343 (__SILICON_REVISION__ < 5)
+/* Serial Port (SPORT) Multichannel Transmit Failure when Channel 0 Is Disabled */
+#define ANOMALY_05000357 (1)
+/* Conflicting Column Address Widths Causes SDRAM Errors */
+#define ANOMALY_05000362 (1)
+/* PPI Underflow Error Goes Undetected in ITU-R 656 Mode */
+#define ANOMALY_05000366 (1)
+/* Possible RETS Register Corruption when Subroutine Is under 5 Cycles in Duration */
+#define ANOMALY_05000371 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000158 (0)
-- 
cgit v1.2.3


From 2047e40d724d42928c0b5994a1568c1b738efdb7 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Tue, 22 Jan 2008 15:29:18 +0800
Subject: [Blackfin] arch: set_bfin_dma_config shouldnt set SYNC or RESTART by
 default - add argument or option

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/kernel/bfin_dma_5xx.c | 4 ++--
 drivers/serial/bfin_5xx.c           | 6 ++++--
 drivers/video/bf54x-lq043fb.c       | 3 ++-
 include/asm-blackfin/dma.h          | 6 +++++-
 4 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c
index b54446055a43..fa9debe8d5f4 100644
--- a/arch/blackfin/kernel/bfin_dma_5xx.c
+++ b/arch/blackfin/kernel/bfin_dma_5xx.c
@@ -339,13 +339,13 @@ EXPORT_SYMBOL(set_dma_config);
 
 unsigned short
 set_bfin_dma_config(char direction, char flow_mode,
-		    char intr_mode, char dma_mode, char width)
+		    char intr_mode, char dma_mode, char width, char syncmode)
 {
 	unsigned short config;
 
 	config =
 	    ((direction << 1) | (width << 2) | (dma_mode << 4) |
-	     (intr_mode << 6) | (flow_mode << 12) | RESTART);
+	     (intr_mode << 6) | (flow_mode << 12) | (syncmode << 5));
 	return config;
 }
 EXPORT_SYMBOL(set_bfin_dma_config);
diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c
index 6f475b609864..ac2a3ef28d55 100644
--- a/drivers/serial/bfin_5xx.c
+++ b/drivers/serial/bfin_5xx.c
@@ -442,7 +442,8 @@ static void bfin_serial_dma_tx_chars(struct bfin_serial_port *uart)
 		set_bfin_dma_config(DIR_READ, DMA_FLOW_STOP,
 			INTR_ON_BUF,
 			DIMENSION_LINEAR,
-			DATA_SIZE_8));
+			DATA_SIZE_8,
+			DMA_SYNC_RESTART));
 	set_dma_start_addr(uart->tx_dma_channel, (unsigned long)(xmit->buf+xmit->tail));
 	set_dma_x_count(uart->tx_dma_channel, uart->tx_count);
 	set_dma_x_modify(uart->tx_dma_channel, 1);
@@ -689,7 +690,8 @@ static int bfin_serial_startup(struct uart_port *port)
 	set_dma_config(uart->rx_dma_channel,
 		set_bfin_dma_config(DIR_WRITE, DMA_FLOW_AUTO,
 				INTR_ON_ROW, DIMENSION_2D,
-				DATA_SIZE_8));
+				DATA_SIZE_8,
+				DMA_SYNC_RESTART));
 	set_dma_x_count(uart->rx_dma_channel, DMA_RX_XCOUNT);
 	set_dma_x_modify(uart->rx_dma_channel, 1);
 	set_dma_y_count(uart->rx_dma_channel, DMA_RX_YCOUNT);
diff --git a/drivers/video/bf54x-lq043fb.c b/drivers/video/bf54x-lq043fb.c
index 74d11c318987..1b7e54de0d76 100644
--- a/drivers/video/bf54x-lq043fb.c
+++ b/drivers/video/bf54x-lq043fb.c
@@ -224,7 +224,8 @@ static int config_dma(struct bfin_bf54xfb_info *fbi)
 	set_dma_config(CH_EPPI0,
 		       set_bfin_dma_config(DIR_READ, DMA_FLOW_AUTO,
 					   INTR_DISABLE, DIMENSION_2D,
-					   DATA_SIZE_32));
+					   DATA_SIZE_32,
+					   DMA_NOSYNC_KEEP_DMA_BUF));
 	set_dma_x_count(CH_EPPI0, (LCD_X_RES * LCD_BPP) / DMA_BUS_SIZE);
 	set_dma_x_modify(CH_EPPI0, DMA_BUS_SIZE / 8);
 	set_dma_y_count(CH_EPPI0, LCD_Y_RES);
diff --git a/include/asm-blackfin/dma.h b/include/asm-blackfin/dma.h
index b469505af364..5abaa2cee8db 100644
--- a/include/asm-blackfin/dma.h
+++ b/include/asm-blackfin/dma.h
@@ -76,6 +76,9 @@ enum dma_chan_status {
 #define INTR_ON_BUF    2
 #define INTR_ON_ROW    3
 
+#define DMA_NOSYNC_KEEP_DMA_BUF	0
+#define DMA_SYNC_RESTART	1
+
 struct dmasg {
 	unsigned long next_desc_addr;
 	unsigned long start_addr;
@@ -157,7 +160,8 @@ void set_dma_y_count(unsigned int channel, unsigned short y_count);
 void set_dma_y_modify(unsigned int channel, short y_modify);
 void set_dma_config(unsigned int channel, unsigned short config);
 unsigned short set_bfin_dma_config(char direction, char flow_mode,
-				   char intr_mode, char dma_mode, char width);
+				   char intr_mode, char dma_mode, char width,
+				   char syncmode);
 void set_dma_curr_addr(unsigned int channel, unsigned long addr);
 
 /* get curr status for polling */
-- 
cgit v1.2.3


From b97b8a998397e8c64699559099fa9febffae2b4d Mon Sep 17 00:00:00 2001
From: Bernd Schmidt <bernd.schmidt@analog.com>
Date: Sun, 27 Jan 2008 18:39:16 +0800
Subject: [Blackfin] arch: Initial checkin of the memory protection support.

Enable it with CONFIG_MPU.

Signed-off-by: Bernd Schmidt <bernd.schmidt@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/Kconfig                     |   9 +
 arch/blackfin/Makefile                    |   6 +-
 arch/blackfin/kernel/cplb-mpu/Makefile    |   8 +
 arch/blackfin/kernel/cplb-mpu/cacheinit.c |  62 ++++++
 arch/blackfin/kernel/cplb-mpu/cplbinfo.c  | 144 +++++++++++++
 arch/blackfin/kernel/cplb-mpu/cplbinit.c  |  91 ++++++++
 arch/blackfin/kernel/cplb-mpu/cplbmgr.c   | 338 ++++++++++++++++++++++++++++++
 arch/blackfin/kernel/setup.c              |  10 +
 arch/blackfin/mach-common/entry.S         |  19 ++
 arch/blackfin/mm/init.c                   |   4 +-
 include/asm-blackfin/cplb-mpu.h           |  61 ++++++
 include/asm-blackfin/cplb.h               |   4 +
 include/asm-blackfin/cplbinit.h           |   8 +
 include/asm-blackfin/mmu.h                |   4 +-
 include/asm-blackfin/mmu_context.h        |  62 +++++-
 15 files changed, 822 insertions(+), 8 deletions(-)
 create mode 100644 arch/blackfin/kernel/cplb-mpu/Makefile
 create mode 100644 arch/blackfin/kernel/cplb-mpu/cacheinit.c
 create mode 100644 arch/blackfin/kernel/cplb-mpu/cplbinfo.c
 create mode 100644 arch/blackfin/kernel/cplb-mpu/cplbinit.c
 create mode 100644 arch/blackfin/kernel/cplb-mpu/cplbmgr.c
 create mode 100644 include/asm-blackfin/cplb-mpu.h

(limited to 'include')

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index ce521a921bee..6b96d7d21aea 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -765,6 +765,15 @@ config L1_MAX_PIECE
 	  Set the max memory pieces for the L1 SRAM allocation algorithm.
 	  Min value is 16. Max value is 1024.
 
+
+config MPU
+	bool "Enable the memory protection unit (EXPERIMENTAL)"
+	default n
+	help
+	  Use the processor's MPU to protect applications from accessing
+	  memory they do not own.  This comes at a performance penalty
+	  and is recommended only for debugging.
+
 comment "Asynchonous Memory Configuration"
 
 menu "EBIU_AMGCTL Global Control"
diff --git a/arch/blackfin/Makefile b/arch/blackfin/Makefile
index 2fc899c5e45f..0edc402fef54 100644
--- a/arch/blackfin/Makefile
+++ b/arch/blackfin/Makefile
@@ -82,7 +82,11 @@ core-y   += arch/$(ARCH)/mach-$(MACHINE)/
 core-y   += arch/$(ARCH)/mach-$(MACHINE)/boards/
 endif
 
-core-y	 += arch/$(ARCH)/kernel/cplb-nompu/
+ifeq ($(CONFIG_MPU),y)
+core-y	+= arch/$(ARCH)/kernel/cplb-mpu/
+else
+core-y	+= arch/$(ARCH)/kernel/cplb-nompu/
+endif
 
 libs-y   += arch/$(ARCH)/lib/
 
diff --git a/arch/blackfin/kernel/cplb-mpu/Makefile b/arch/blackfin/kernel/cplb-mpu/Makefile
new file mode 100644
index 000000000000..286b69357f97
--- /dev/null
+++ b/arch/blackfin/kernel/cplb-mpu/Makefile
@@ -0,0 +1,8 @@
+#
+# arch/blackfin/kernel/cplb-nompu/Makefile
+#
+
+obj-y := cplbinit.o cacheinit.o cplbmgr.o
+
+obj-$(CONFIG_CPLB_INFO) += cplbinfo.o
+
diff --git a/arch/blackfin/kernel/cplb-mpu/cacheinit.c b/arch/blackfin/kernel/cplb-mpu/cacheinit.c
new file mode 100644
index 000000000000..9eecfa403187
--- /dev/null
+++ b/arch/blackfin/kernel/cplb-mpu/cacheinit.c
@@ -0,0 +1,62 @@
+/*
+ *               Copyright 2004-2007 Analog Devices Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include <linux/cpu.h>
+
+#include <asm/cacheflush.h>
+#include <asm/blackfin.h>
+#include <asm/cplb.h>
+#include <asm/cplbinit.h>
+
+#if defined(CONFIG_BFIN_ICACHE)
+void bfin_icache_init(void)
+{
+	unsigned long ctrl;
+	int i;
+
+	SSYNC();
+	for (i = 0; i < MAX_CPLBS; i++) {
+		bfin_write32(ICPLB_ADDR0 + i * 4, icplb_tbl[i].addr);
+		bfin_write32(ICPLB_DATA0 + i * 4, icplb_tbl[i].data);
+	}
+	ctrl = bfin_read_IMEM_CONTROL();
+	ctrl |= IMC | ENICPLB;
+	bfin_write_IMEM_CONTROL(ctrl);
+	SSYNC();
+}
+#endif
+
+#if defined(CONFIG_BFIN_DCACHE)
+void bfin_dcache_init(void)
+{
+	unsigned long ctrl;
+	int i;
+
+	SSYNC();
+	for (i = 0; i < MAX_CPLBS; i++) {
+		bfin_write32(DCPLB_ADDR0 + i * 4, dcplb_tbl[i].addr);
+		bfin_write32(DCPLB_DATA0 + i * 4, dcplb_tbl[i].data);
+	}
+
+	ctrl = bfin_read_DMEM_CONTROL();
+	ctrl |= DMEM_CNTR;
+	bfin_write_DMEM_CONTROL(ctrl);
+	SSYNC();
+}
+#endif
diff --git a/arch/blackfin/kernel/cplb-mpu/cplbinfo.c b/arch/blackfin/kernel/cplb-mpu/cplbinfo.c
new file mode 100644
index 000000000000..bd072299f7f2
--- /dev/null
+++ b/arch/blackfin/kernel/cplb-mpu/cplbinfo.c
@@ -0,0 +1,144 @@
+/*
+ * File:         arch/blackfin/mach-common/cplbinfo.c
+ * Based on:
+ * Author:       Sonic Zhang <sonic.zhang@analog.com>
+ *
+ * Created:      Jan. 2005
+ * Description:  Display CPLB status
+ *
+ * Modified:
+ *               Copyright 2004-2006 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/proc_fs.h>
+#include <linux/uaccess.h>
+
+#include <asm/current.h>
+#include <asm/system.h>
+#include <asm/cplb.h>
+#include <asm/cplbinit.h>
+#include <asm/blackfin.h>
+
+#define CPLB_I 1
+#define CPLB_D 2
+
+#define SYNC_SYS    SSYNC()
+#define SYNC_CORE   CSYNC()
+
+#define CPLB_BIT_PAGESIZE 0x30000
+
+static char page_size_string_table[][4] = { "1K", "4K", "1M", "4M" };
+
+static char *cplb_print_entry(char *buf, struct cplb_entry *tbl, int switched)
+{
+	int i;
+	buf += sprintf(buf, "Index\tAddress\t\tData\tSize\tU/RD\tU/WR\tS/WR\tSwitch\n");
+	for (i = 0; i < MAX_CPLBS; i++) {
+		unsigned long data = tbl[i].data;
+		unsigned long addr = tbl[i].addr;
+		if (!(data & CPLB_VALID))
+			continue;
+
+		buf +=
+		    sprintf(buf,
+			    "%d\t0x%08lx\t%06lx\t%s\t%c\t%c\t%c\t%c\n",
+			    i, addr, data,
+			    page_size_string_table[(data & 0x30000) >> 16],
+			    (data & CPLB_USER_RD) ? 'Y' : 'N',
+			    (data & CPLB_USER_WR) ? 'Y' : 'N',
+			    (data & CPLB_SUPV_WR) ? 'Y' : 'N',
+			    i < switched ? 'N' : 'Y');
+	}
+	buf += sprintf(buf, "\n");
+
+	return buf;
+}
+
+int cplbinfo_proc_output(char *buf)
+{
+	char *p;
+
+	p = buf;
+
+	p += sprintf(p, "------------------ CPLB Information ------------------\n\n");
+
+	if (bfin_read_IMEM_CONTROL() & ENICPLB) {
+		p += sprintf(p, "Instruction CPLB entry:\n");
+		p = cplb_print_entry(p, icplb_tbl, first_switched_icplb);
+	} else
+		p += sprintf(p, "Instruction CPLB is disabled.\n\n");
+
+	if (1 || bfin_read_DMEM_CONTROL() & ENDCPLB) {
+		p += sprintf(p, "Data CPLB entry:\n");
+		p = cplb_print_entry(p, dcplb_tbl, first_switched_dcplb);
+	} else
+		p += sprintf(p, "Data CPLB is disabled.\n");
+
+	p += sprintf(p, "ICPLB miss: %d\nICPLB supervisor miss: %d\n",
+		     nr_icplb_miss, nr_icplb_supv_miss);
+	p += sprintf(p, "DCPLB miss: %d\nDCPLB protection fault:%d\n",
+		     nr_dcplb_miss, nr_dcplb_prot);
+	p += sprintf(p, "CPLB flushes: %d\n",
+		     nr_cplb_flush);
+
+	return p - buf;
+}
+
+static int cplbinfo_read_proc(char *page, char **start, off_t off,
+			      int count, int *eof, void *data)
+{
+	int len;
+
+	len = cplbinfo_proc_output(page);
+	if (len <= off + count)
+		*eof = 1;
+	*start = page + off;
+	len -= off;
+	if (len > count)
+		len = count;
+	if (len < 0)
+		len = 0;
+	return len;
+}
+
+static int __init cplbinfo_init(void)
+{
+	struct proc_dir_entry *entry;
+
+	entry = create_proc_entry("cplbinfo", 0, NULL);
+	if (!entry)
+		return -ENOMEM;
+
+	entry->read_proc = cplbinfo_read_proc;
+	entry->data = NULL;
+
+	return 0;
+}
+
+static void __exit cplbinfo_exit(void)
+{
+	remove_proc_entry("cplbinfo", NULL);
+}
+
+module_init(cplbinfo_init);
+module_exit(cplbinfo_exit);
diff --git a/arch/blackfin/kernel/cplb-mpu/cplbinit.c b/arch/blackfin/kernel/cplb-mpu/cplbinit.c
new file mode 100644
index 000000000000..e2e2b5079f5b
--- /dev/null
+++ b/arch/blackfin/kernel/cplb-mpu/cplbinit.c
@@ -0,0 +1,91 @@
+/*
+ * Blackfin CPLB initialization
+ *
+ *               Copyright 2004-2007 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include <linux/module.h>
+
+#include <asm/blackfin.h>
+#include <asm/cplb.h>
+#include <asm/cplbinit.h>
+
+struct cplb_entry icplb_tbl[MAX_CPLBS];
+struct cplb_entry dcplb_tbl[MAX_CPLBS];
+
+int first_switched_icplb, first_switched_dcplb;
+int first_mask_dcplb;
+
+void __init generate_cpl_tables(void)
+{
+	int i_d, i_i;
+	unsigned long addr;
+	unsigned long d_data, i_data;
+	unsigned long d_cache = 0, i_cache = 0;
+
+#ifdef CONFIG_BFIN_ICACHE
+	i_cache = CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
+#endif
+
+#ifdef CONFIG_BFIN_DCACHE
+	d_cache = CPLB_L1_CHBL;
+#ifdef CONFIG_BLKFIN_WT
+	d_cache |= CPLB_L1_AOW | CPLB_WT;
+#endif
+#endif
+	i_d = i_i = 0;
+
+	/* Set up the zero page.  */
+	dcplb_tbl[i_d].addr = 0;
+	dcplb_tbl[i_d++].data = SDRAM_OOPS | PAGE_SIZE_1KB;
+
+#if 0
+	icplb_tbl[i_i].addr = 0;
+	icplb_tbl[i_i++].data = i_cache | CPLB_USER_RD | PAGE_SIZE_4KB;
+#endif
+
+	/* Cover kernel memory with 4M pages.  */
+	addr = 0;
+	d_data = d_cache | CPLB_SUPV_WR | CPLB_VALID | PAGE_SIZE_4MB | CPLB_DIRTY;
+	i_data = i_cache | CPLB_VALID | CPLB_PORTPRIO | PAGE_SIZE_4MB;
+
+	for (; addr < memory_start; addr += 4 * 1024 * 1024) {
+		dcplb_tbl[i_d].addr = addr;
+		dcplb_tbl[i_d++].data = d_data;
+		icplb_tbl[i_i].addr = addr;
+		icplb_tbl[i_i++].data = i_data | (addr == 0 ? CPLB_USER_RD : 0);
+	}
+
+	/* Cover L1 memory.  One 4M area for code and data each is enough.  */
+#if L1_DATA_A_LENGTH > 0 || L1_DATA_B_LENGTH > 0
+	dcplb_tbl[i_d].addr = L1_DATA_A_START;
+	dcplb_tbl[i_d++].data = L1_DMEMORY | PAGE_SIZE_4MB;
+#endif
+	icplb_tbl[i_i].addr = L1_CODE_START;
+	icplb_tbl[i_i++].data = L1_IMEMORY | PAGE_SIZE_4MB;
+
+	first_mask_dcplb = i_d;
+	first_switched_dcplb = i_d + (1 << page_mask_order);
+	first_switched_icplb = i_i;
+
+	while (i_d < MAX_CPLBS)
+		dcplb_tbl[i_d++].data = 0;
+	while (i_i < MAX_CPLBS)
+		icplb_tbl[i_i++].data = 0;
+}
diff --git a/arch/blackfin/kernel/cplb-mpu/cplbmgr.c b/arch/blackfin/kernel/cplb-mpu/cplbmgr.c
new file mode 100644
index 000000000000..c426a22f9907
--- /dev/null
+++ b/arch/blackfin/kernel/cplb-mpu/cplbmgr.c
@@ -0,0 +1,338 @@
+/*
+ *               Blackfin CPLB exception handling.
+ *               Copyright 2004-2007 Analog Devices Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include <linux/module.h>
+#include <linux/mm.h>
+
+#include <asm/blackfin.h>
+#include <asm/cplbinit.h>
+#include <asm/mmu_context.h>
+
+#ifdef CONFIG_BFIN_ICACHE
+
+#define FAULT_RW	(1 << 16)
+#define FAULT_USERSUPV	(1 << 17)
+
+int page_mask_nelts;
+int page_mask_order;
+unsigned long *current_rwx_mask;
+
+int nr_dcplb_miss, nr_icplb_miss, nr_icplb_supv_miss, nr_dcplb_prot;
+int nr_cplb_flush;
+
+static inline void disable_dcplb(void)
+{
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_DMEM_CONTROL();
+	ctrl &= ~ENDCPLB;
+	bfin_write_DMEM_CONTROL(ctrl);
+	SSYNC();
+}
+
+static inline void enable_dcplb(void)
+{
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_DMEM_CONTROL();
+	ctrl |= ENDCPLB;
+	bfin_write_DMEM_CONTROL(ctrl);
+	SSYNC();
+}
+
+static inline void disable_icplb(void)
+{
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_IMEM_CONTROL();
+	ctrl &= ~ENICPLB;
+	bfin_write_IMEM_CONTROL(ctrl);
+	SSYNC();
+}
+
+static inline void enable_icplb(void)
+{
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_IMEM_CONTROL();
+	ctrl |= ENICPLB;
+	bfin_write_IMEM_CONTROL(ctrl);
+	SSYNC();
+}
+
+/*
+ * Given the contents of the status register, return the index of the
+ * CPLB that caused the fault.
+ */
+static inline int faulting_cplb_index(int status)
+{
+	int signbits = __builtin_bfin_norm_fr1x32(status & 0xFFFF);
+	return 30 - signbits;
+}
+
+/*
+ * Given the contents of the status register and the DCPLB_DATA contents,
+ * return true if a write access should be permitted.
+ */
+static inline int write_permitted(int status, unsigned long data)
+{
+	if (status & FAULT_USERSUPV)
+		return !!(data & CPLB_SUPV_WR);
+	else
+		return !!(data & CPLB_USER_WR);
+}
+
+/* Counters to implement round-robin replacement.  */
+static int icplb_rr_index, dcplb_rr_index;
+
+/*
+ * Find an ICPLB entry to be evicted and return its index.
+ */
+static int evict_one_icplb(void)
+{
+	int i;
+	for (i = first_switched_icplb; i < MAX_CPLBS; i++)
+		if ((icplb_tbl[i].data & CPLB_VALID) == 0)
+			return i;
+	i = first_switched_icplb + icplb_rr_index;
+	if (i >= MAX_CPLBS) {
+		i -= MAX_CPLBS - first_switched_icplb;
+		icplb_rr_index -= MAX_CPLBS - first_switched_icplb;
+	}
+	icplb_rr_index++;
+	return i;
+}
+
+static int evict_one_dcplb(void)
+{
+	int i;
+	for (i = first_switched_dcplb; i < MAX_CPLBS; i++)
+		if ((dcplb_tbl[i].data & CPLB_VALID) == 0)
+			return i;
+	i = first_switched_dcplb + dcplb_rr_index;
+	if (i >= MAX_CPLBS) {
+		i -= MAX_CPLBS - first_switched_dcplb;
+		dcplb_rr_index -= MAX_CPLBS - first_switched_dcplb;
+	}
+	dcplb_rr_index++;
+	return i;
+}
+
+static noinline int dcplb_miss(void)
+{
+	unsigned long addr = bfin_read_DCPLB_FAULT_ADDR();
+	int status = bfin_read_DCPLB_STATUS();
+	unsigned long *mask;
+	int idx;
+	unsigned long d_data;
+
+	nr_dcplb_miss++;
+	if (addr >= _ramend)
+		return CPLB_PROT_VIOL;
+
+	d_data = CPLB_SUPV_WR | CPLB_VALID | CPLB_DIRTY | PAGE_SIZE_4KB;
+#ifdef CONFIG_BFIN_DCACHE
+	d_data |= CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
+#ifdef CONFIG_BLKFIN_WT
+	d_data |= CPLB_L1_AOW | CPLB_WT;
+#endif
+#endif
+	mask = current_rwx_mask;
+	if (mask) {
+		int page = addr >> PAGE_SHIFT;
+		int offs = page >> 5;
+		int bit = 1 << (page & 31);
+
+		if (mask[offs] & bit)
+			d_data |= CPLB_USER_RD;
+
+		mask += page_mask_nelts;
+		if (mask[offs] & bit)
+			d_data |= CPLB_USER_WR;
+	}
+
+	idx = evict_one_dcplb();
+
+	addr &= PAGE_MASK;
+	dcplb_tbl[idx].addr = addr;
+	dcplb_tbl[idx].data = d_data;
+
+	disable_dcplb();
+	bfin_write32(DCPLB_DATA0 + idx * 4, d_data);
+	bfin_write32(DCPLB_ADDR0 + idx * 4, addr);
+	enable_dcplb();
+
+	return 0;
+}
+
+static noinline int icplb_miss(void)
+{
+	unsigned long addr = bfin_read_ICPLB_FAULT_ADDR();
+	int status = bfin_read_ICPLB_STATUS();
+	int idx;
+	unsigned long i_data;
+
+	nr_icplb_miss++;
+	if (status & FAULT_USERSUPV)
+		nr_icplb_supv_miss++;
+
+	if (addr >= _ramend)
+		return CPLB_PROT_VIOL;
+
+	/*
+	 * First, try to find a CPLB that matches this address.  If we
+	 * find one, then the fact that we're in the miss handler means
+	 * that the instruction crosses a page boundary.
+	 */
+	for (idx = first_switched_icplb; idx < MAX_CPLBS; idx++) {
+		if (icplb_tbl[idx].data & CPLB_VALID) {
+			unsigned long this_addr = icplb_tbl[idx].addr;
+			if (this_addr <= addr && this_addr + PAGE_SIZE > addr) {
+				addr += PAGE_SIZE;
+				break;
+			}
+		}
+	}
+
+	i_data = CPLB_VALID | CPLB_PORTPRIO | PAGE_SIZE_4KB;
+#ifdef CONFIG_BFIN_ICACHE
+	i_data |= CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
+#endif
+
+	/*
+	 * Two cases to distinguish - a supervisor access must necessarily
+	 * be for a module page; we grant it unconditionally (could do better
+	 * here in the future).  Otherwise, check the x bitmap of the current
+	 * process.
+	 */
+	if (!(status & FAULT_USERSUPV)) {
+		unsigned long *mask = current_rwx_mask;
+
+		if (mask) {
+			int page = addr >> PAGE_SHIFT;
+			int offs = page >> 5;
+			int bit = 1 << (page & 31);
+
+			mask += 2 * page_mask_nelts;
+			if (mask[offs] & bit)
+				i_data |= CPLB_USER_RD;
+		}
+	}
+
+	idx = evict_one_icplb();
+	addr &= PAGE_MASK;
+	icplb_tbl[idx].addr = addr;
+	icplb_tbl[idx].data = i_data;
+
+	disable_icplb();
+	bfin_write32(ICPLB_DATA0 + idx * 4, i_data);
+	bfin_write32(ICPLB_ADDR0 + idx * 4, addr);
+	enable_icplb();
+
+	return 0;
+}
+
+static noinline int dcplb_protection_fault(void)
+{
+	unsigned long addr = bfin_read_DCPLB_FAULT_ADDR();
+	int status = bfin_read_DCPLB_STATUS();
+
+	nr_dcplb_prot++;
+
+	if (status & FAULT_RW) {
+		int idx = faulting_cplb_index(status);
+		unsigned long data = dcplb_tbl[idx].data;
+		if (!(data & CPLB_WT) && !(data & CPLB_DIRTY) &&
+		    write_permitted(status, data)) {
+			data |= CPLB_DIRTY;
+			dcplb_tbl[idx].data = data;
+			bfin_write32(DCPLB_DATA0 + idx * 4, data);
+			return 0;
+		}
+	}
+	return CPLB_PROT_VIOL;
+}
+
+int cplb_hdr(int seqstat, struct pt_regs *regs)
+{
+	int cause = seqstat & 0x3f;
+	switch (cause) {
+	case 0x23:
+		return dcplb_protection_fault();
+	case 0x2C:
+		return icplb_miss();
+	case 0x26:
+		return dcplb_miss();
+	default:
+	    return 1;
+		panic_cplb_error(seqstat, regs);
+	}
+}
+
+void flush_switched_cplbs(void)
+{
+	int i;
+
+	nr_cplb_flush++;
+
+	disable_icplb();
+	for (i = first_switched_icplb; i < MAX_CPLBS; i++) {
+		icplb_tbl[i].data = 0;
+		bfin_write32(ICPLB_DATA0 + i * 4, 0);
+	}
+	enable_icplb();
+
+	disable_dcplb();
+	for (i = first_mask_dcplb; i < MAX_CPLBS; i++) {
+		dcplb_tbl[i].data = 0;
+		bfin_write32(DCPLB_DATA0 + i * 4, 0);
+	}
+	enable_dcplb();
+}
+
+void set_mask_dcplbs(unsigned long *masks)
+{
+	int i;
+	unsigned long addr = (unsigned long)masks;
+	unsigned long d_data;
+	current_rwx_mask = masks;
+
+	if (!masks)
+		return;
+
+	d_data = CPLB_SUPV_WR | CPLB_VALID | CPLB_DIRTY | PAGE_SIZE_4KB;
+#ifdef CONFIG_BFIN_DCACHE
+	d_data |= CPLB_L1_CHBL;
+#ifdef CONFIG_BLKFIN_WT
+	d_data |= CPLB_L1_AOW | CPLB_WT;
+#endif
+#endif
+
+	disable_dcplb();
+	for (i = first_mask_dcplb; i < first_switched_dcplb; i++) {
+		dcplb_tbl[i].addr = addr;
+		dcplb_tbl[i].data = d_data;
+		bfin_write32(DCPLB_DATA0 + i * 4, d_data);
+		bfin_write32(DCPLB_ADDR0 + i * 4, addr);
+		addr += PAGE_SIZE;
+	}
+	enable_dcplb();
+}
+
+#endif
diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index a03c2dfff4a3..1a942a721d51 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -238,7 +238,12 @@ void __init setup_arch(char **cmdline_p)
 	memory_end = _ramend - DMA_UNCACHED_REGION;
 
 	_ramstart = (unsigned long)__bss_stop;
+#ifdef CONFIG_MPU
+	/* Round up to multiple of 4MB.  */
+	memory_start = (_ramstart + 0x3fffff) & ~0x3fffff;
+#else
 	memory_start = PAGE_ALIGN(_ramstart);
+#endif
 
 #if defined(CONFIG_MTD_UCLINUX)
 	/* generic memory mapped MTD driver */
@@ -307,6 +312,11 @@ void __init setup_arch(char **cmdline_p)
 	printk(KERN_NOTICE "Warning: limiting memory to %liMB due to hardware anomaly 05000263\n", memory_end >> 20);
 #endif				/* ANOMALY_05000263 */
 
+#ifdef CONFIG_MPU
+	page_mask_nelts = ((_ramend >> PAGE_SHIFT) + 31) / 32;
+	page_mask_order = get_order(3 * page_mask_nelts * sizeof(long));
+#endif
+
 #if !defined(CONFIG_MTD_UCLINUX)
 	memory_end -= SIZE_4K; /*In case there is no valid CPLB behind memory_end make sure we don't get to close*/
 #endif
diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S
index 58f7ad617992..c2e81a10c47b 100644
--- a/arch/blackfin/mach-common/entry.S
+++ b/arch/blackfin/mach-common/entry.S
@@ -95,24 +95,43 @@ ENTRY(_ex_workaround_261)
 	R6 = 0x26;	/* Data CPLB Miss */
 	cc = R6 == R7;
 	if cc jump _ex_dcplb_miss (BP);
+	R6 = 0x23;	/* Data CPLB Miss */
+	cc = R6 == R7;
+	if cc jump _ex_dcplb_viol (BP);
 	/* Handle 0x23 Data CPLB Protection Violation
 	 * and Data CPLB Multiple Hits - Linux Trap Zero
 	 */
 	jump _ex_trap_c;
 ENDPROC(_ex_workaround_261)
 
+#else
+#ifdef CONFIG_MPU
+#define _ex_dviol _ex_dcplb_viol
 #else
 #define _ex_dviol _ex_trap_c
+#endif
 #define _ex_dmiss _ex_dcplb_miss
 #define _ex_dmult _ex_trap_c
 #endif
 
+
+ENTRY(_ex_dcplb_viol)
 ENTRY(_ex_dcplb_miss)
 ENTRY(_ex_icplb_miss)
 	(R7:6,P5:4) = [sp++];
 	ASTAT = [sp++];
 	SAVE_ALL_SYS
+#ifdef CONFIG_MPU
+	R0 = SEQSTAT;
+	R1 = SP;
+	sp += -12;
+	call _cplb_hdr;
+	sp += 12;
+	CC = R0 == 0;
+	IF !CC JUMP _handle_bad_cplb;
+#else
 	call __cplb_hdr;
+#endif
 	DEBUG_START_HWTRACE(p5, r7)
 	RESTORE_ALL_SYS
 	SP = EX_SCRATCH_REG;
diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index e97ea8fc8dc4..9f007cace6e6 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -184,13 +184,15 @@ static __init void free_init_pages(const char *what, unsigned long begin, unsign
 #ifdef CONFIG_BLK_DEV_INITRD
 void __init free_initrd_mem(unsigned long start, unsigned long end)
 {
+#ifndef CONFIG_MPU
 	free_init_pages("initrd memory", start, end);
+#endif
 }
 #endif
 
 void __init free_initmem(void)
 {
-#ifdef CONFIG_RAMKERNEL
+#if defined CONFIG_RAMKERNEL && !defined CONFIG_MPU
 	free_init_pages("unused kernel memory",
 			(unsigned long)(&__init_begin),
 			(unsigned long)(&__init_end));
diff --git a/include/asm-blackfin/cplb-mpu.h b/include/asm-blackfin/cplb-mpu.h
new file mode 100644
index 000000000000..75c67b99d607
--- /dev/null
+++ b/include/asm-blackfin/cplb-mpu.h
@@ -0,0 +1,61 @@
+/*
+ * File:         include/asm-blackfin/cplbinit.h
+ * Based on:
+ * Author:
+ *
+ * Created:
+ * Description:
+ *
+ * Modified:
+ *               Copyright 2004-2006 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#ifndef __ASM_BFIN_CPLB_MPU_H
+#define __ASM_BFIN_CPLB_MPU_H
+
+struct cplb_entry {
+	unsigned long data, addr;
+};
+
+struct mem_region {
+	unsigned long start, end;
+	unsigned long dcplb_data;
+	unsigned long icplb_data;
+};
+
+extern struct cplb_entry dcplb_tbl[MAX_CPLBS];
+extern struct cplb_entry icplb_tbl[MAX_CPLBS];
+extern int first_switched_icplb;
+extern int first_mask_dcplb;
+extern int first_switched_dcplb;
+
+extern int nr_dcplb_miss, nr_icplb_miss, nr_icplb_supv_miss, nr_dcplb_prot;
+extern int nr_cplb_flush;
+
+extern int page_mask_order;
+extern int page_mask_nelts;
+
+extern unsigned long *current_rwx_mask;
+
+extern void flush_switched_cplbs(void);
+extern void set_mask_dcplbs(unsigned long *);
+
+extern void __noreturn panic_cplb_error(int seqstat, struct pt_regs *);
+
+#endif /* __ASM_BFIN_CPLB_MPU_H */
diff --git a/include/asm-blackfin/cplb.h b/include/asm-blackfin/cplb.h
index 06828d77a58f..654375c2b746 100644
--- a/include/asm-blackfin/cplb.h
+++ b/include/asm-blackfin/cplb.h
@@ -65,7 +65,11 @@
 #define SIZE_1M 0x00100000      /* 1M */
 #define SIZE_4M 0x00400000      /* 4M */
 
+#ifdef CONFIG_MPU
+#define MAX_CPLBS 16
+#else
 #define MAX_CPLBS (16 * 2)
+#endif
 
 #define ASYNC_MEMORY_CPLB_COVERAGE	((ASYNC_BANK0_SIZE + ASYNC_BANK1_SIZE + \
 				 ASYNC_BANK2_SIZE + ASYNC_BANK3_SIZE) / SIZE_4M)
diff --git a/include/asm-blackfin/cplbinit.h b/include/asm-blackfin/cplbinit.h
index c4d0596e8e9f..0eb1c1b685a7 100644
--- a/include/asm-blackfin/cplbinit.h
+++ b/include/asm-blackfin/cplbinit.h
@@ -33,6 +33,12 @@
 #include <asm/blackfin.h>
 #include <asm/cplb.h>
 
+#ifdef CONFIG_MPU
+
+#include <asm/cplb-mpu.h>
+
+#else
+
 #define INITIAL_T 0x1
 #define SWITCH_T  0x2
 #define I_CPLB    0x4
@@ -79,6 +85,8 @@ extern u_long ipdt_swapcount_table[];
 extern u_long dpdt_swapcount_table[];
 #endif
 
+#endif /* CONFIG_MPU */
+
 extern unsigned long reserved_mem_dcache_on;
 extern unsigned long reserved_mem_icache_on;
 
diff --git a/include/asm-blackfin/mmu.h b/include/asm-blackfin/mmu.h
index 11d52f1167d0..757e43906ed4 100644
--- a/include/asm-blackfin/mmu.h
+++ b/include/asm-blackfin/mmu.h
@@ -24,7 +24,9 @@ typedef struct {
 	unsigned long	exec_fdpic_loadmap;
 	unsigned long	interp_fdpic_loadmap;
 #endif
-
+#ifdef CONFIG_MPU
+	unsigned long *page_rwx_mask;
+#endif
 } mm_context_t;
 
 #endif
diff --git a/include/asm-blackfin/mmu_context.h b/include/asm-blackfin/mmu_context.h
index c5c71a6aaf19..b5eb67596ad5 100644
--- a/include/asm-blackfin/mmu_context.h
+++ b/include/asm-blackfin/mmu_context.h
@@ -30,9 +30,12 @@
 #ifndef __BLACKFIN_MMU_CONTEXT_H__
 #define __BLACKFIN_MMU_CONTEXT_H__
 
+#include <linux/gfp.h>
+#include <linux/sched.h>
 #include <asm/setup.h>
 #include <asm/page.h>
 #include <asm/pgalloc.h>
+#include <asm/cplbinit.h>
 
 extern void *current_l1_stack_save;
 extern int nr_l1stack_tasks;
@@ -50,6 +53,12 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 static inline int
 init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 {
+#ifdef CONFIG_MPU
+	unsigned long p = __get_free_pages(GFP_KERNEL, page_mask_order);
+	mm->context.page_rwx_mask = (unsigned long *)p;
+	memset(mm->context.page_rwx_mask, 0,
+	       page_mask_nelts * 3 * sizeof(long));
+#endif
 	return 0;
 }
 
@@ -73,6 +82,11 @@ static inline void destroy_context(struct mm_struct *mm)
 		sram_free(tmp->addr);
 		kfree(tmp);
 	}
+#ifdef CONFIG_MPU
+	if (current_rwx_mask == mm->context.page_rwx_mask)
+		current_rwx_mask = NULL;
+	free_pages((unsigned long)mm->context.page_rwx_mask, page_mask_order);
+#endif
 }
 
 static inline unsigned long
@@ -106,9 +120,21 @@ activate_l1stack(struct mm_struct *mm, unsigned long sp_base)
 
 #define deactivate_mm(tsk,mm)	do { } while (0)
 
-static inline void activate_mm(struct mm_struct *prev_mm,
-			       struct mm_struct *next_mm)
+#define activate_mm(prev, next) switch_mm(prev, next, NULL)
+
+static inline void switch_mm(struct mm_struct *prev_mm, struct mm_struct *next_mm,
+			     struct task_struct *tsk)
 {
+	if (prev_mm == next_mm)
+		return;
+#ifdef CONFIG_MPU
+	if (prev_mm->context.page_rwx_mask == current_rwx_mask) {
+		flush_switched_cplbs();
+		set_mask_dcplbs(next_mm->context.page_rwx_mask);
+	}
+#endif
+
+	/* L1 stack switching.  */
 	if (!next_mm->context.l1_stack_save)
 		return;
 	if (next_mm->context.l1_stack_save == current_l1_stack_save)
@@ -120,10 +146,36 @@ static inline void activate_mm(struct mm_struct *prev_mm,
 	memcpy(l1_stack_base, current_l1_stack_save, l1_stack_len);
 }
 
-static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
-			     struct task_struct *tsk)
+#ifdef CONFIG_MPU
+static inline void protect_page(struct mm_struct *mm, unsigned long addr,
+				unsigned long flags)
+{
+	unsigned long *mask = mm->context.page_rwx_mask;
+	unsigned long page = addr >> 12;
+	unsigned long idx = page >> 5;
+	unsigned long bit = 1 << (page & 31);
+
+	if (flags & VM_MAYREAD)
+		mask[idx] |= bit;
+	else
+		mask[idx] &= ~bit;
+	mask += page_mask_nelts;
+	if (flags & VM_MAYWRITE)
+		mask[idx] |= bit;
+	else
+		mask[idx] &= ~bit;
+	mask += page_mask_nelts;
+	if (flags & VM_MAYEXEC)
+		mask[idx] |= bit;
+	else
+		mask[idx] &= ~bit;
+}
+
+static inline void update_protections(struct mm_struct *mm)
 {
-	activate_mm(prev, next);
+	flush_switched_cplbs();
+	set_mask_dcplbs(mm->context.page_rwx_mask);
 }
+#endif
 
 #endif
-- 
cgit v1.2.3


From acbcd2631975cf6f0be5cd294cbfd12226cd9958 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Tue, 22 Jan 2008 18:36:20 +0800
Subject: [Blackfin] arch: Fix BUG gpio_direction_output API is not
 compatitable with GENERIC_GPIO API interface

signef-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/kernel/bfin_gpio.c                  | 55 +++++++++++++++++------
 arch/blackfin/mach-bf527/boards/ezkit.c           |  7 +--
 arch/blackfin/mach-bf537/boards/generic_board.c   |  6 +--
 arch/blackfin/mach-bf537/boards/pnav10.c          |  6 +--
 arch/blackfin/mach-bf537/boards/stamp.c           |  7 +--
 drivers/video/bf54x-lq043fb.c                     |  3 +-
 include/asm-blackfin/gpio.h                       | 12 ++---
 include/asm-blackfin/mach-bf527/bfin_serial_5xx.h |  2 +-
 include/asm-blackfin/mach-bf533/bfin_serial_5xx.h |  2 +-
 include/asm-blackfin/mach-bf537/bfin_serial_5xx.h |  2 +-
 include/asm-blackfin/mach-bf548/bfin_serial_5xx.h |  2 +-
 include/asm-blackfin/mach-bf561/bfin_serial_5xx.h |  2 +-
 12 files changed, 58 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c
index ffee36910288..7312011a4211 100644
--- a/arch/blackfin/kernel/bfin_gpio.c
+++ b/arch/blackfin/kernel/bfin_gpio.c
@@ -229,6 +229,11 @@ inline int check_gpio(unsigned short gpio)
 }
 #endif
 
+void gpio_error(unsigned gpio)
+{
+	printk(KERN_ERR "bfin-gpio: GPIO %d wasn't requested!\n", gpio);
+}
+
 static void set_label(unsigned short ident, const char *label)
 {
 
@@ -1034,7 +1039,7 @@ EXPORT_SYMBOL(peripheral_free_list);
 * MODIFICATION HISTORY :
 **************************************************************/
 
-int gpio_request(unsigned short gpio, const char *label)
+int gpio_request(unsigned gpio, const char *label)
 {
 	unsigned long flags;
 
@@ -1081,7 +1086,7 @@ int gpio_request(unsigned short gpio, const char *label)
 }
 EXPORT_SYMBOL(gpio_request);
 
-void gpio_free(unsigned short gpio)
+void gpio_free(unsigned gpio)
 {
 	unsigned long flags;
 
@@ -1091,7 +1096,7 @@ void gpio_free(unsigned short gpio)
 	local_irq_save(flags);
 
 	if (unlikely(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)))) {
-		printk(KERN_ERR "bfin-gpio: GPIO %d wasn't reserved!\n", gpio);
+		gpio_error(gpio);
 		dump_stack();
 		local_irq_restore(flags);
 		return;
@@ -1107,34 +1112,47 @@ void gpio_free(unsigned short gpio)
 }
 EXPORT_SYMBOL(gpio_free);
 
+
 #ifdef BF548_FAMILY
-void gpio_direction_input(unsigned short gpio)
+int gpio_direction_input(unsigned gpio)
 {
 	unsigned long flags;
 
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
+	if (!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))) {
+		gpio_error(gpio);
+		return -EINVAL;
+	}
+
 
 	local_irq_save(flags);
 	gpio_array[gpio_bank(gpio)]->port_dir_clear = gpio_bit(gpio);
 	gpio_array[gpio_bank(gpio)]->port_inen |= gpio_bit(gpio);
 	local_irq_restore(flags);
+
+	return 0;
 }
 EXPORT_SYMBOL(gpio_direction_input);
 
-void gpio_direction_output(unsigned short gpio)
+int gpio_direction_output(unsigned gpio, int value)
 {
 	unsigned long flags;
 
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
+	if (!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))) {
+		gpio_error(gpio);
+		return -EINVAL;
+	}
 
 	local_irq_save(flags);
 	gpio_array[gpio_bank(gpio)]->port_inen &= ~gpio_bit(gpio);
+	gpio_set_value(gpio, value);
 	gpio_array[gpio_bank(gpio)]->port_dir_set = gpio_bit(gpio);
 	local_irq_restore(flags);
+
+	return 0;
 }
 EXPORT_SYMBOL(gpio_direction_output);
 
-void gpio_set_value(unsigned short gpio, unsigned short arg)
+void gpio_set_value(unsigned gpio, int arg)
 {
 	if (arg)
 		gpio_array[gpio_bank(gpio)]->port_set = gpio_bit(gpio);
@@ -1144,7 +1162,7 @@ void gpio_set_value(unsigned short gpio, unsigned short arg)
 }
 EXPORT_SYMBOL(gpio_set_value);
 
-unsigned short gpio_get_value(unsigned short gpio)
+int gpio_get_value(unsigned gpio)
 {
 	return (1 & (gpio_array[gpio_bank(gpio)]->port_data >> gpio_sub_n(gpio)));
 }
@@ -1152,31 +1170,42 @@ EXPORT_SYMBOL(gpio_get_value);
 
 #else
 
-void gpio_direction_input(unsigned short gpio)
+int gpio_direction_input(unsigned gpio)
 {
 	unsigned long flags;
 
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
+	if (!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))) {
+		gpio_error(gpio);
+		return -EINVAL;
+	}
 
 	local_irq_save(flags);
 	gpio_bankb[gpio_bank(gpio)]->dir &= ~gpio_bit(gpio);
 	gpio_bankb[gpio_bank(gpio)]->inen |= gpio_bit(gpio);
 	AWA_DUMMY_READ(inen);
 	local_irq_restore(flags);
+
+	return 0;
 }
 EXPORT_SYMBOL(gpio_direction_input);
 
-void gpio_direction_output(unsigned short gpio)
+int gpio_direction_output(unsigned gpio, int value)
 {
 	unsigned long flags;
 
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
+	if (!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))) {
+		gpio_error(gpio);
+		return -EINVAL;
+	}
 
 	local_irq_save(flags);
 	gpio_bankb[gpio_bank(gpio)]->inen &= ~gpio_bit(gpio);
+	gpio_set_value(gpio, value);
 	gpio_bankb[gpio_bank(gpio)]->dir |= gpio_bit(gpio);
 	AWA_DUMMY_READ(dir);
 	local_irq_restore(flags);
+
+	return 0;
 }
 EXPORT_SYMBOL(gpio_direction_output);
 
diff --git a/arch/blackfin/mach-bf527/boards/ezkit.c b/arch/blackfin/mach-bf527/boards/ezkit.c
index bc256811a5e3..6d70aae4cc9b 100644
--- a/arch/blackfin/mach-bf527/boards/ezkit.c
+++ b/arch/blackfin/mach-bf527/boards/ezkit.c
@@ -317,12 +317,7 @@ static struct resource sl811_hcd_resources[] = {
 void sl811_port_power(struct device *dev, int is_on)
 {
 	gpio_request(CONFIG_USB_SL811_BFIN_GPIO_VBUS, "usb:SL811_VBUS");
-	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS);
-
-	if (is_on)
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 1);
-	else
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 0);
+	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS, is_on);
 }
 #endif
 
diff --git a/arch/blackfin/mach-bf537/boards/generic_board.c b/arch/blackfin/mach-bf537/boards/generic_board.c
index 1c97219a2ab5..3225b1588398 100644
--- a/arch/blackfin/mach-bf537/boards/generic_board.c
+++ b/arch/blackfin/mach-bf537/boards/generic_board.c
@@ -205,12 +205,8 @@ static struct resource sl811_hcd_resources[] = {
 void sl811_port_power(struct device *dev, int is_on)
 {
 	gpio_request(CONFIG_USB_SL811_BFIN_GPIO_VBUS, "usb:SL811_VBUS");
-	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS);
+	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS, is_on);
 
-	if (is_on)
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 1);
-	else
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 0);
 }
 #endif
 
diff --git a/arch/blackfin/mach-bf537/boards/pnav10.c b/arch/blackfin/mach-bf537/boards/pnav10.c
index 23e10c7dab5b..4c4870590bd8 100644
--- a/arch/blackfin/mach-bf537/boards/pnav10.c
+++ b/arch/blackfin/mach-bf537/boards/pnav10.c
@@ -134,12 +134,8 @@ static struct resource sl811_hcd_resources[] = {
 void sl811_port_power(struct device *dev, int is_on)
 {
 	gpio_request(CONFIG_USB_SL811_BFIN_GPIO_VBUS, "usb:SL811_VBUS");
-	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS);
+	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS, is_on);
 
-	if (is_on)
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 1);
-	else
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 0);
 }
 #endif
 
diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c
index 3e0ad04d8d75..37759ac7df2a 100644
--- a/arch/blackfin/mach-bf537/boards/stamp.c
+++ b/arch/blackfin/mach-bf537/boards/stamp.c
@@ -250,12 +250,7 @@ static struct resource sl811_hcd_resources[] = {
 void sl811_port_power(struct device *dev, int is_on)
 {
 	gpio_request(CONFIG_USB_SL811_BFIN_GPIO_VBUS, "usb:SL811_VBUS");
-	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS);
-
-	if (is_on)
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 1);
-	else
-		gpio_set_value(CONFIG_USB_SL811_BFIN_GPIO_VBUS, 0);
+	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS, is_on);
 }
 #endif
 
diff --git a/drivers/video/bf54x-lq043fb.c b/drivers/video/bf54x-lq043fb.c
index 1b7e54de0d76..c8e7427a0bc8 100644
--- a/drivers/video/bf54x-lq043fb.c
+++ b/drivers/video/bf54x-lq043fb.c
@@ -264,8 +264,7 @@ static int request_ports(struct bfin_bf54xfb_info *fbi)
 		}
 	}
 
-	gpio_direction_output(disp);
-	gpio_set_value(disp, 1);
+	gpio_direction_output(disp, 1);
 
 	return 0;
 }
diff --git a/include/asm-blackfin/gpio.h b/include/asm-blackfin/gpio.h
index 33ce98ef7e0f..7bba9b17654e 100644
--- a/include/asm-blackfin/gpio.h
+++ b/include/asm-blackfin/gpio.h
@@ -426,19 +426,19 @@ struct gpio_port_s {
 * MODIFICATION HISTORY :
 **************************************************************/
 
-int gpio_request(unsigned short, const char *);
-void gpio_free(unsigned short);
+int gpio_request(unsigned, const char *);
+void gpio_free(unsigned);
 
-void gpio_set_value(unsigned short gpio, unsigned short arg);
-unsigned short gpio_get_value(unsigned short gpio);
+void gpio_set_value(unsigned gpio, int arg);
+int gpio_get_value(unsigned gpio);
 
 #ifndef BF548_FAMILY
 #define gpio_get_value(gpio) 		get_gpio_data(gpio)
 #define gpio_set_value(gpio, value)	set_gpio_data(gpio, value)
 #endif
 
-void gpio_direction_input(unsigned short gpio);
-void gpio_direction_output(unsigned short gpio);
+int gpio_direction_input(unsigned gpio);
+int gpio_direction_output(unsigned gpio, int value);
 
 #include <asm-generic/gpio.h>		/* cansleep wrappers */
 #include <asm/irq.h>
diff --git a/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h
index 0b867e6a76c4..15dbc21eed8b 100644
--- a/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h
+++ b/include/asm-blackfin/mach-bf527/bfin_serial_5xx.h
@@ -146,7 +146,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 
 	if (uart->rts_pin >= 0) {
 		gpio_request(uart->rts_pin, DRIVER_NAME);
-		gpio_direction_output(uart->rts_pin);
+		gpio_direction_output(uart->rts_pin, 0);
 	}
 #endif
 }
diff --git a/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h
index 69b9f8e120e9..7871d4313f49 100644
--- a/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h
+++ b/include/asm-blackfin/mach-bf533/bfin_serial_5xx.h
@@ -111,7 +111,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 	}
 	if (uart->rts_pin >= 0) {
 		gpio_request(uart->rts_pin, DRIVER_NAME);
-		gpio_direction_input(uart->rts_pin);
+		gpio_direction_input(uart->rts_pin, 0);
 	}
 #endif
 }
diff --git a/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h
index 6fb328f5186a..86e45c379838 100644
--- a/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h
+++ b/include/asm-blackfin/mach-bf537/bfin_serial_5xx.h
@@ -146,7 +146,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 
 	if (uart->rts_pin >= 0) {
 		gpio_request(uart->rts_pin, DRIVER_NAME);
-		gpio_direction_output(uart->rts_pin);
+		gpio_direction_output(uart->rts_pin, 0);
 	}
 #endif
 }
diff --git a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h
index f21a1620e6bd..3770aa38ee9f 100644
--- a/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h
+++ b/include/asm-blackfin/mach-bf548/bfin_serial_5xx.h
@@ -186,7 +186,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 
 	if (uart->rts_pin >= 0) {
 		gpio_request(uart->rts_pin, DRIVER_NAME);
-		gpio_direction_output(uart->rts_pin);
+		gpio_direction_output(uart->rts_pin, 0);
 	}
 #endif
 }
diff --git a/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h b/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h
index 69b9f8e120e9..7871d4313f49 100644
--- a/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h
+++ b/include/asm-blackfin/mach-bf561/bfin_serial_5xx.h
@@ -111,7 +111,7 @@ static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 	}
 	if (uart->rts_pin >= 0) {
 		gpio_request(uart->rts_pin, DRIVER_NAME);
-		gpio_direction_input(uart->rts_pin);
+		gpio_direction_input(uart->rts_pin, 0);
 	}
 #endif
 }
-- 
cgit v1.2.3


From a2c8cfef6abb33ee49d80c58391ebfc4f94221ef Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Tue, 22 Jan 2008 17:20:10 +0800
Subject: [Blackfin] arch: GPIO API cleanup and anomaly update

 - Add anomaly workaround for bfin_gpio_reset_spi0_ssel1
 - Fix style
 - Update copyright
 - Remove BUG_ON checks for functions intended to be used only by arch
   support. GPIO users should only access using the generic GPIO API
 - Make all GPIO identifier unsigned int

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/kernel/bfin_gpio.c | 63 ++++++++++++++++-------------------
 include/asm-blackfin/gpio.h      | 72 ++++++++++++++++++++--------------------
 2 files changed, 64 insertions(+), 71 deletions(-)

(limited to 'include')

diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c
index 7312011a4211..6bbe0a2fccb8 100644
--- a/arch/blackfin/kernel/bfin_gpio.c
+++ b/arch/blackfin/kernel/bfin_gpio.c
@@ -7,7 +7,7 @@
  * Description:  GPIO Abstraction Layer
  *
  * Modified:
- *               Copyright 2007 Analog Devices Inc.
+ *               Copyright 2008 Analog Devices Inc.
  *
  * Bugs:         Enter bugs at http://blackfin.uclinux.org/
  *
@@ -137,7 +137,6 @@ static unsigned short *port_fer[gpio_bank(MAX_BLACKFIN_GPIOS)] = {
 	(unsigned short *) PORTG_FER,
 	(unsigned short *) PORTH_FER,
 };
-
 #endif
 
 #ifdef BF527_FAMILY
@@ -211,7 +210,7 @@ static unsigned int sic_iwr_irqs[gpio_bank(MAX_BLACKFIN_GPIOS)] = {IRQ_PROG0_INT
 #endif /* CONFIG_PM */
 
 #if defined(BF548_FAMILY)
-inline int check_gpio(unsigned short gpio)
+inline int check_gpio(unsigned gpio)
 {
 	if (gpio == GPIO_PB15 || gpio == GPIO_PC14 || gpio == GPIO_PC15
 	    || gpio == GPIO_PH14 || gpio == GPIO_PH15
@@ -221,7 +220,7 @@ inline int check_gpio(unsigned short gpio)
 	return 0;
 }
 #else
-inline int check_gpio(unsigned short gpio)
+inline int check_gpio(unsigned gpio)
 {
 	if (gpio >= MAX_BLACKFIN_GPIOS)
 		return -EINVAL;
@@ -236,7 +235,6 @@ void gpio_error(unsigned gpio)
 
 static void set_label(unsigned short ident, const char *label)
 {
-
 	if (label && str_ident) {
 		strncpy(str_ident[ident].name, label,
 			 RESOURCE_LABEL_SIZE);
@@ -267,7 +265,7 @@ static int cmp_label(unsigned short ident, const char *label)
 }
 
 #if defined(BF527_FAMILY) || defined(BF537_FAMILY)
-static void port_setup(unsigned short gpio, unsigned short usage)
+static void port_setup(unsigned gpio, unsigned short usage)
 {
 	if (!check_gpio(gpio)) {
 		if (usage == GPIO_USAGE)
@@ -278,7 +276,7 @@ static void port_setup(unsigned short gpio, unsigned short usage)
 	}
 }
 #elif defined(BF548_FAMILY)
-static void port_setup(unsigned short gpio, unsigned short usage)
+static void port_setup(unsigned gpio, unsigned short usage)
 {
 	if (usage == GPIO_USAGE)
 		gpio_array[gpio_bank(gpio)]->port_fer &= ~gpio_bit(gpio);
@@ -399,7 +397,7 @@ inline void portmux_setup(unsigned short portno, unsigned short function)
 #endif
 
 #ifndef BF548_FAMILY
-static void default_gpio(unsigned short gpio)
+static void default_gpio(unsigned gpio)
 {
 	unsigned short bank, bitmask;
 	unsigned long flags;
@@ -419,7 +417,6 @@ static void default_gpio(unsigned short gpio)
 	gpio_bankb[bank]->edge &= ~bitmask;
 	AWA_DUMMY_READ(edge);
 	local_irq_restore(flags);
-
 }
 #else
 # define default_gpio(...)  do { } while (0)
@@ -457,10 +454,9 @@ arch_initcall(bfin_gpio_init);
 /* Set a specific bit */
 
 #define SET_GPIO(name) \
-void set_gpio_ ## name(unsigned short gpio, unsigned short arg) \
+void set_gpio_ ## name(unsigned gpio, unsigned short arg) \
 { \
 	unsigned long flags; \
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))); \
 	local_irq_save(flags); \
 	if (arg) \
 		gpio_bankb[gpio_bank(gpio)]->name |= gpio_bit(gpio); \
@@ -480,10 +476,9 @@ SET_GPIO(both)
 
 #if ANOMALY_05000311 || ANOMALY_05000323
 #define SET_GPIO_SC(name) \
-void set_gpio_ ## name(unsigned short gpio, unsigned short arg) \
+void set_gpio_ ## name(unsigned gpio, unsigned short arg) \
 { \
 	unsigned long flags; \
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))); \
 	local_irq_save(flags); \
 	if (arg) \
 		gpio_bankb[gpio_bank(gpio)]->name ## _set = gpio_bit(gpio); \
@@ -495,9 +490,8 @@ void set_gpio_ ## name(unsigned short gpio, unsigned short arg) \
 EXPORT_SYMBOL(set_gpio_ ## name);
 #else
 #define SET_GPIO_SC(name) \
-void set_gpio_ ## name(unsigned short gpio, unsigned short arg) \
+void set_gpio_ ## name(unsigned gpio, unsigned short arg) \
 { \
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio))); \
 	if (arg) \
 		gpio_bankb[gpio_bank(gpio)]->name ## _set = gpio_bit(gpio); \
 	else \
@@ -511,19 +505,17 @@ SET_GPIO_SC(maskb)
 SET_GPIO_SC(data)
 
 #if ANOMALY_05000311 || ANOMALY_05000323
-void set_gpio_toggle(unsigned short gpio)
+void set_gpio_toggle(unsigned gpio)
 {
 	unsigned long flags;
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
 	local_irq_save(flags);
 	gpio_bankb[gpio_bank(gpio)]->toggle = gpio_bit(gpio);
 	AWA_DUMMY_READ(toggle);
 	local_irq_restore(flags);
 }
 #else
-void set_gpio_toggle(unsigned short gpio)
+void set_gpio_toggle(unsigned gpio)
 {
-	BUG_ON(!(reserved_gpio_map[gpio_bank(gpio)] & gpio_bit(gpio)));
 	gpio_bankb[gpio_bank(gpio)]->toggle = gpio_bit(gpio);
 }
 #endif
@@ -534,7 +526,7 @@ EXPORT_SYMBOL(set_gpio_toggle);
 
 #if ANOMALY_05000311 || ANOMALY_05000323
 #define SET_GPIO_P(name) \
-void set_gpiop_ ## name(unsigned short gpio, unsigned short arg) \
+void set_gpiop_ ## name(unsigned gpio, unsigned short arg) \
 { \
 	unsigned long flags; \
 	local_irq_save(flags); \
@@ -545,7 +537,7 @@ void set_gpiop_ ## name(unsigned short gpio, unsigned short arg) \
 EXPORT_SYMBOL(set_gpiop_ ## name);
 #else
 #define SET_GPIO_P(name) \
-void set_gpiop_ ## name(unsigned short gpio, unsigned short arg) \
+void set_gpiop_ ## name(unsigned gpio, unsigned short arg) \
 { \
 	gpio_bankb[gpio_bank(gpio)]->name = arg; \
 } \
@@ -561,11 +553,10 @@ SET_GPIO_P(both)
 SET_GPIO_P(maska)
 SET_GPIO_P(maskb)
 
-
 /* Get a specific bit */
 #if ANOMALY_05000311 || ANOMALY_05000323
 #define GET_GPIO(name) \
-unsigned short get_gpio_ ## name(unsigned short gpio) \
+unsigned short get_gpio_ ## name(unsigned gpio) \
 { \
 	unsigned long flags; \
 	unsigned short ret; \
@@ -578,7 +569,7 @@ unsigned short get_gpio_ ## name(unsigned short gpio) \
 EXPORT_SYMBOL(get_gpio_ ## name);
 #else
 #define GET_GPIO(name) \
-unsigned short get_gpio_ ## name(unsigned short gpio) \
+unsigned short get_gpio_ ## name(unsigned gpio) \
 { \
 	return (0x01 & (gpio_bankb[gpio_bank(gpio)]->name >> gpio_sub_n(gpio))); \
 } \
@@ -598,7 +589,7 @@ GET_GPIO(maskb)
 
 #if ANOMALY_05000311 || ANOMALY_05000323
 #define GET_GPIO_P(name) \
-unsigned short get_gpiop_ ## name(unsigned short gpio) \
+unsigned short get_gpiop_ ## name(unsigned gpio) \
 { \
 	unsigned long flags; \
 	unsigned short ret; \
@@ -611,7 +602,7 @@ unsigned short get_gpiop_ ## name(unsigned short gpio) \
 EXPORT_SYMBOL(get_gpiop_ ## name);
 #else
 #define GET_GPIO_P(name) \
-unsigned short get_gpiop_ ## name(unsigned short gpio) \
+unsigned short get_gpiop_ ## name(unsigned gpio) \
 { \
 	return (gpio_bankb[gpio_bank(gpio)]->name);\
 } \
@@ -648,7 +639,7 @@ GET_GPIO_P(maskb)
 *************************************************************
 * MODIFICATION HISTORY :
 **************************************************************/
-int gpio_pm_wakeup_request(unsigned short gpio, unsigned char type)
+int gpio_pm_wakeup_request(unsigned gpio, unsigned char type)
 {
 	unsigned long flags;
 
@@ -656,7 +647,6 @@ int gpio_pm_wakeup_request(unsigned short gpio, unsigned char type)
 		return -EINVAL;
 
 	local_irq_save(flags);
-
 	wakeup_map[gpio_bank(gpio)] |= gpio_bit(gpio);
 	wakeup_flags_map[gpio] = type;
 	local_irq_restore(flags);
@@ -665,7 +655,7 @@ int gpio_pm_wakeup_request(unsigned short gpio, unsigned char type)
 }
 EXPORT_SYMBOL(gpio_pm_wakeup_request);
 
-void gpio_pm_wakeup_free(unsigned short gpio)
+void gpio_pm_wakeup_free(unsigned gpio)
 {
 	unsigned long flags;
 
@@ -680,7 +670,7 @@ void gpio_pm_wakeup_free(unsigned short gpio)
 }
 EXPORT_SYMBOL(gpio_pm_wakeup_free);
 
-static int bfin_gpio_wakeup_type(unsigned short gpio, unsigned char type)
+static int bfin_gpio_wakeup_type(unsigned gpio, unsigned char type)
 {
 	port_setup(gpio, GPIO_USAGE);
 	set_gpio_dir(gpio, 0);
@@ -789,7 +779,7 @@ void gpio_pm_restore(void)
 #endif
 #else /* BF548_FAMILY */
 
-unsigned short get_gpio_dir(unsigned short gpio)
+unsigned short get_gpio_dir(unsigned gpio)
 {
 	return (0x01 & (gpio_array[gpio_bank(gpio)]->port_dir_clear >> gpio_sub_n(gpio)));
 }
@@ -1123,7 +1113,6 @@ int gpio_direction_input(unsigned gpio)
 		return -EINVAL;
 	}
 
-
 	local_irq_save(flags);
 	gpio_array[gpio_bank(gpio)]->port_dir_clear = gpio_bit(gpio);
 	gpio_array[gpio_bank(gpio)]->port_inen |= gpio_bit(gpio);
@@ -1158,7 +1147,6 @@ void gpio_set_value(unsigned gpio, int arg)
 		gpio_array[gpio_bank(gpio)]->port_set = gpio_bit(gpio);
 	else
 		gpio_array[gpio_bank(gpio)]->port_clear = gpio_bit(gpio);
-
 }
 EXPORT_SYMBOL(gpio_set_value);
 
@@ -1200,7 +1188,12 @@ int gpio_direction_output(unsigned gpio, int value)
 
 	local_irq_save(flags);
 	gpio_bankb[gpio_bank(gpio)]->inen &= ~gpio_bit(gpio);
-	gpio_set_value(gpio, value);
+
+	if (value)
+		gpio_bankb[gpio_bank(gpio)]->data_set = gpio_bit(gpio);
+	else
+		gpio_bankb[gpio_bank(gpio)]->data_clear = gpio_bit(gpio);
+
 	gpio_bankb[gpio_bank(gpio)]->dir |= gpio_bit(gpio);
 	AWA_DUMMY_READ(dir);
 	local_irq_restore(flags);
@@ -1225,6 +1218,7 @@ void bfin_gpio_reset_spi0_ssel1(void)
 
 	port_setup(gpio, GPIO_USAGE);
 	gpio_bankb[gpio_bank(gpio)]->data_set = gpio_bit(gpio);
+	AWA_DUMMY_READ(data_set);
 	udelay(1);
 }
 
@@ -1259,6 +1253,5 @@ static __init int gpio_register_proc(void)
 		proc_gpio->read_proc = gpio_proc_read;
 	return proc_gpio != NULL;
 }
-
 __initcall(gpio_register_proc);
 #endif
diff --git a/include/asm-blackfin/gpio.h b/include/asm-blackfin/gpio.h
index 7bba9b17654e..d0426c108262 100644
--- a/include/asm-blackfin/gpio.h
+++ b/include/asm-blackfin/gpio.h
@@ -7,7 +7,7 @@
  * Description:
  *
  * Modified:
- *               Copyright 2004-2006 Analog Devices Inc.
+ *               Copyright 2004-2008 Analog Devices Inc.
  *
  * Bugs:         Enter bugs at http://blackfin.uclinux.org/
  *
@@ -304,39 +304,39 @@
 **************************************************************/
 
 #ifndef BF548_FAMILY
-void set_gpio_dir(unsigned short, unsigned short);
-void set_gpio_inen(unsigned short, unsigned short);
-void set_gpio_polar(unsigned short, unsigned short);
-void set_gpio_edge(unsigned short, unsigned short);
-void set_gpio_both(unsigned short, unsigned short);
-void set_gpio_data(unsigned short, unsigned short);
-void set_gpio_maska(unsigned short, unsigned short);
-void set_gpio_maskb(unsigned short, unsigned short);
-void set_gpio_toggle(unsigned short);
-void set_gpiop_dir(unsigned short, unsigned short);
-void set_gpiop_inen(unsigned short, unsigned short);
-void set_gpiop_polar(unsigned short, unsigned short);
-void set_gpiop_edge(unsigned short, unsigned short);
-void set_gpiop_both(unsigned short, unsigned short);
-void set_gpiop_data(unsigned short, unsigned short);
-void set_gpiop_maska(unsigned short, unsigned short);
-void set_gpiop_maskb(unsigned short, unsigned short);
-unsigned short get_gpio_dir(unsigned short);
-unsigned short get_gpio_inen(unsigned short);
-unsigned short get_gpio_polar(unsigned short);
-unsigned short get_gpio_edge(unsigned short);
-unsigned short get_gpio_both(unsigned short);
-unsigned short get_gpio_maska(unsigned short);
-unsigned short get_gpio_maskb(unsigned short);
-unsigned short get_gpio_data(unsigned short);
-unsigned short get_gpiop_dir(unsigned short);
-unsigned short get_gpiop_inen(unsigned short);
-unsigned short get_gpiop_polar(unsigned short);
-unsigned short get_gpiop_edge(unsigned short);
-unsigned short get_gpiop_both(unsigned short);
-unsigned short get_gpiop_maska(unsigned short);
-unsigned short get_gpiop_maskb(unsigned short);
-unsigned short get_gpiop_data(unsigned short);
+void set_gpio_dir(unsigned, unsigned short);
+void set_gpio_inen(unsigned, unsigned short);
+void set_gpio_polar(unsigned, unsigned short);
+void set_gpio_edge(unsigned, unsigned short);
+void set_gpio_both(unsigned, unsigned short);
+void set_gpio_data(unsigned, unsigned short);
+void set_gpio_maska(unsigned, unsigned short);
+void set_gpio_maskb(unsigned, unsigned short);
+void set_gpio_toggle(unsigned);
+void set_gpiop_dir(unsigned, unsigned short);
+void set_gpiop_inen(unsigned, unsigned short);
+void set_gpiop_polar(unsigned, unsigned short);
+void set_gpiop_edge(unsigned, unsigned short);
+void set_gpiop_both(unsigned, unsigned short);
+void set_gpiop_data(unsigned, unsigned short);
+void set_gpiop_maska(unsigned, unsigned short);
+void set_gpiop_maskb(unsigned, unsigned short);
+unsigned short get_gpio_dir(unsigned);
+unsigned short get_gpio_inen(unsigned);
+unsigned short get_gpio_polar(unsigned);
+unsigned short get_gpio_edge(unsigned);
+unsigned short get_gpio_both(unsigned);
+unsigned short get_gpio_maska(unsigned);
+unsigned short get_gpio_maskb(unsigned);
+unsigned short get_gpio_data(unsigned);
+unsigned short get_gpiop_dir(unsigned);
+unsigned short get_gpiop_inen(unsigned);
+unsigned short get_gpiop_polar(unsigned);
+unsigned short get_gpiop_edge(unsigned);
+unsigned short get_gpiop_both(unsigned);
+unsigned short get_gpiop_maska(unsigned);
+unsigned short get_gpiop_maskb(unsigned);
+unsigned short get_gpiop_data(unsigned);
 
 struct gpio_port_t {
 	unsigned short data;
@@ -382,8 +382,8 @@ struct gpio_port_t {
 #define PM_WAKE_LOW	0x8
 #define PM_WAKE_BOTH_EDGES	(PM_WAKE_RISING | PM_WAKE_FALLING)
 
-int gpio_pm_wakeup_request(unsigned short gpio, unsigned char type);
-void gpio_pm_wakeup_free(unsigned short gpio);
+int gpio_pm_wakeup_request(unsigned gpio, unsigned char type);
+void gpio_pm_wakeup_free(unsigned gpio);
 unsigned int gpio_pm_setup(void);
 void gpio_pm_restore(void);
 
-- 
cgit v1.2.3


From fc97551db9e4e9402ff2b5c94be8267b2e5f32f4 Mon Sep 17 00:00:00 2001
From: Bernd Schmidt <bernd.schmidt@analog.com>
Date: Sun, 27 Jan 2008 19:56:43 +0800
Subject: [Blackfin] arch: Add the semtimedop syscall.  Upstream uClibc doesn't
 compile without it.

Signed-off-by: Bernd Schmidt <bernd.schmidt@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 arch/blackfin/mach-common/entry.S | 1 +
 include/asm-blackfin/unistd.h     | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S
index c2e81a10c47b..56ff51bc8c21 100644
--- a/arch/blackfin/mach-common/entry.S
+++ b/arch/blackfin/mach-common/entry.S
@@ -1381,6 +1381,7 @@ ENTRY(_sys_call_table)
 	.long _sys_set_robust_list
 	.long _sys_get_robust_list	/* 355 */
 	.long _sys_fallocate
+	.long _sys_semtimedop
 	.rept NR_syscalls-(.-_sys_call_table)/4
 	.long _sys_ni_syscall
 	.endr
diff --git a/include/asm-blackfin/unistd.h b/include/asm-blackfin/unistd.h
index 07ffe8b718c5..e98167358d26 100644
--- a/include/asm-blackfin/unistd.h
+++ b/include/asm-blackfin/unistd.h
@@ -369,8 +369,9 @@
 #define __NR_set_robust_list	354
 #define __NR_get_robust_list	355
 #define __NR_fallocate		356
+#define __NR_semtimedop		357
 
-#define __NR_syscall		357
+#define __NR_syscall		358
 #define NR_syscalls		__NR_syscall
 
 /* Old optional stuff no one actually uses */
-- 
cgit v1.2.3


From 408dbc0da29913f99e56001db892eb653b47c13b Mon Sep 17 00:00:00 2001
From: Bryan Wu <bryan.wu@analog.com>
Date: Sun, 27 Jan 2008 18:38:12 +0800
Subject: [Blackfin] arch: remove old I2C BF54x porting.

Signed-off-by: Bryan Wu <bryan.wu@analog.com>
---
 include/asm-blackfin/mach-bf548/cdefBF54x_base.h | 33 ------------------------
 1 file changed, 33 deletions(-)

(limited to 'include')

diff --git a/include/asm-blackfin/mach-bf548/cdefBF54x_base.h b/include/asm-blackfin/mach-bf548/cdefBF54x_base.h
index aefab3f618c1..19ddcd83c71f 100644
--- a/include/asm-blackfin/mach-bf548/cdefBF54x_base.h
+++ b/include/asm-blackfin/mach-bf548/cdefBF54x_base.h
@@ -244,39 +244,6 @@ static __inline__ void bfin_write_VR_CTL(unsigned int val)
 #define bfin_read_TWI0_RCV_DATA16()		bfin_read16(TWI0_RCV_DATA16)
 #define bfin_write_TWI0_RCV_DATA16(val)		bfin_write16(TWI0_RCV_DATA16, val)
 
-#define bfin_read_TWI_CLKDIV()			bfin_read16(TWI0_CLKDIV)
-#define bfin_write_TWI_CLKDIV(val)		bfin_write16(TWI0_CLKDIV, val)
-#define bfin_read_TWI_CONTROL()			bfin_read16(TWI0_CONTROL)
-#define bfin_write_TWI_CONTROL(val)		bfin_write16(TWI0_CONTROL, val)
-#define bfin_read_TWI_SLAVE_CTRL()		bfin_read16(TWI0_SLAVE_CTRL)
-#define bfin_write_TWI_SLAVE_CTRL(val)		bfin_write16(TWI0_SLAVE_CTRL, val)
-#define bfin_read_TWI_SLAVE_STAT()		bfin_read16(TWI0_SLAVE_STAT)
-#define bfin_write_TWI_SLAVE_STAT(val)		bfin_write16(TWI0_SLAVE_STAT, val)
-#define bfin_read_TWI_SLAVE_ADDR()		bfin_read16(TWI0_SLAVE_ADDR)
-#define bfin_write_TWI_SLAVE_ADDR(val)		bfin_write16(TWI0_SLAVE_ADDR, val)
-#define bfin_read_TWI_MASTER_CTL()		bfin_read16(TWI0_MASTER_CTRL)
-#define bfin_write_TWI_MASTER_CTL(val)		bfin_write16(TWI0_MASTER_CTRL, val)
-#define bfin_read_TWI_MASTER_STAT()		bfin_read16(TWI0_MASTER_STAT)
-#define bfin_write_TWI_MASTER_STAT(val)		bfin_write16(TWI0_MASTER_STAT, val)
-#define bfin_read_TWI_MASTER_ADDR()		bfin_read16(TWI0_MASTER_ADDR)
-#define bfin_write_TWI_MASTER_ADDR(val)		bfin_write16(TWI0_MASTER_ADDR, val)
-#define bfin_read_TWI_INT_STAT()		bfin_read16(TWI0_INT_STAT)
-#define bfin_write_TWI_INT_STAT(val)		bfin_write16(TWI0_INT_STAT, val)
-#define bfin_read_TWI_INT_MASK()		bfin_read16(TWI0_INT_MASK)
-#define bfin_write_TWI_INT_MASK(val)		bfin_write16(TWI0_INT_MASK, val)
-#define bfin_read_TWI_FIFO_CTL()		bfin_read16(TWI0_FIFO_CTRL)
-#define bfin_write_TWI_FIFO_CTL(val)		bfin_write16(TWI0_FIFO_CTRL, val)
-#define bfin_read_TWI_FIFO_STAT()		bfin_read16(TWI0_FIFO_STAT)
-#define bfin_write_TWI_FIFO_STAT(val)		bfin_write16(TWI0_FIFO_STAT, val)
-#define bfin_read_TWI_XMT_DATA8()		bfin_read16(TWI0_XMT_DATA8)
-#define bfin_write_TWI_XMT_DATA8(val)		bfin_write16(TWI0_XMT_DATA8, val)
-#define bfin_read_TWI_XMT_DATA16()		bfin_read16(TWI0_XMT_DATA16)
-#define bfin_write_TWI_XMT_DATA16(val)		bfin_write16(TWI0_XMT_DATA16, val)
-#define bfin_read_TWI_RCV_DATA8()		bfin_read16(TWI0_RCV_DATA8)
-#define bfin_write_TWI_RCV_DATA8(val)		bfin_write16(TWI0_RCV_DATA8, val)
-#define bfin_read_TWI_RCV_DATA16()		bfin_read16(TWI0_RCV_DATA16)
-#define bfin_write_TWI_RCV_DATA16(val)		bfin_write16(TWI0_RCV_DATA16, val)
-
 /* SPORT0 is not defined in the shared file because it is not available on the ADSP-BF542 and ADSP-BF544 bfin_read_()rocessors */
 
 /* SPORT1 Registers */
-- 
cgit v1.2.3


From 16d004a2eda7be2c6a2de63eca2ad3c6b57307b3 Mon Sep 17 00:00:00 2001
From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Date: Thu, 11 Oct 2007 19:48:58 +0800
Subject: [CRYPTO] des: Create header file for common macros

This patch creates include/crypto/des.h for common macros shared between
DES implementations.

Signed-off-by: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/des_generic.c |  8 +-------
 include/crypto/des.h | 19 +++++++++++++++++++
 2 files changed, 20 insertions(+), 7 deletions(-)
 create mode 100644 include/crypto/des.h

(limited to 'include')

diff --git a/crypto/des_generic.c b/crypto/des_generic.c
index 59966d14b8e0..f75eafe1a875 100644
--- a/crypto/des_generic.c
+++ b/crypto/des_generic.c
@@ -20,13 +20,7 @@
 #include <linux/crypto.h>
 #include <linux/types.h>
 
-#define DES_KEY_SIZE		8
-#define DES_EXPKEY_WORDS	32
-#define DES_BLOCK_SIZE		8
-
-#define DES3_EDE_KEY_SIZE	(3 * DES_KEY_SIZE)
-#define DES3_EDE_EXPKEY_WORDS	(3 * DES_EXPKEY_WORDS)
-#define DES3_EDE_BLOCK_SIZE	DES_BLOCK_SIZE
+#include <crypto/des.h>
 
 #define ROL(x, r) ((x) = rol32((x), (r)))
 #define ROR(x, r) ((x) = ror32((x), (r)))
diff --git a/include/crypto/des.h b/include/crypto/des.h
new file mode 100644
index 000000000000..2971c6304ade
--- /dev/null
+++ b/include/crypto/des.h
@@ -0,0 +1,19 @@
+/* 
+ * DES & Triple DES EDE Cipher Algorithms.
+ */
+
+#ifndef __CRYPTO_DES_H
+#define __CRYPTO_DES_H
+
+#define DES_KEY_SIZE		8
+#define DES_EXPKEY_WORDS	32
+#define DES_BLOCK_SIZE		8
+
+#define DES3_EDE_KEY_SIZE	(3 * DES_KEY_SIZE)
+#define DES3_EDE_EXPKEY_WORDS	(3 * DES_EXPKEY_WORDS)
+#define DES3_EDE_BLOCK_SIZE	DES_BLOCK_SIZE
+
+
+extern unsigned long des_ekey(u32 *pe, const u8 *k);
+
+#endif /* __CRYPTO_DES_H */
-- 
cgit v1.2.3


From 89e12654312dddbbdbf17b5adc95b22cb672f947 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <sebastian@breakpoint.cc>
Date: Wed, 17 Oct 2007 23:18:57 +0800
Subject: [CRYPTO] aes: Move common defines into a header file

This three defines are used in all AES related hardware.

Signed-off-by: Sebastian Siewior <sebastian@breakpoint.cc>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 arch/s390/crypto/aes_s390.c  |  7 +------
 arch/x86/crypto/aes_32.c     |  4 +---
 arch/x86/crypto/aes_64.c     |  6 +-----
 crypto/aes_generic.c         |  6 +-----
 drivers/crypto/geode-aes.c   |  1 +
 drivers/crypto/padlock-aes.c |  4 +---
 include/crypto/aes.h         | 15 +++++++++++++++
 7 files changed, 21 insertions(+), 22 deletions(-)
 create mode 100644 include/crypto/aes.h

(limited to 'include')

diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c
index 512669691ad0..812511bbb540 100644
--- a/arch/s390/crypto/aes_s390.c
+++ b/arch/s390/crypto/aes_s390.c
@@ -16,17 +16,12 @@
  *
  */
 
+#include <crypto/aes.h>
 #include <crypto/algapi.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include "crypt_s390.h"
 
-#define AES_MIN_KEY_SIZE	16
-#define AES_MAX_KEY_SIZE	32
-
-/* data block size for all key lengths */
-#define AES_BLOCK_SIZE		16
-
 #define AES_KEYLEN_128		1
 #define AES_KEYLEN_192		2
 #define AES_KEYLEN_256		4
diff --git a/arch/x86/crypto/aes_32.c b/arch/x86/crypto/aes_32.c
index 49aad9397f10..9b0ab50394b0 100644
--- a/arch/x86/crypto/aes_32.c
+++ b/arch/x86/crypto/aes_32.c
@@ -38,6 +38,7 @@
  */
 
 #include <asm/byteorder.h>
+#include <crypto/aes.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -48,9 +49,6 @@
 asmlinkage void aes_enc_blk(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 asmlinkage void aes_dec_blk(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 
-#define AES_MIN_KEY_SIZE	16
-#define AES_MAX_KEY_SIZE	32
-#define AES_BLOCK_SIZE		16
 #define AES_KS_LENGTH		4 * AES_BLOCK_SIZE
 #define RC_LENGTH		29
 
diff --git a/arch/x86/crypto/aes_64.c b/arch/x86/crypto/aes_64.c
index 5cdb13ea5cc2..0b38a4cd2ce1 100644
--- a/arch/x86/crypto/aes_64.c
+++ b/arch/x86/crypto/aes_64.c
@@ -54,6 +54,7 @@
 */
 
 #include <asm/byteorder.h>
+#include <crypto/aes.h>
 #include <linux/bitops.h>
 #include <linux/crypto.h>
 #include <linux/errno.h>
@@ -61,11 +62,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 
-#define AES_MIN_KEY_SIZE	16
-#define AES_MAX_KEY_SIZE	32
-
-#define AES_BLOCK_SIZE		16
-
 /*
  * #define byte(x, nr) ((unsigned char)((x) >> (nr*8)))
  */
diff --git a/crypto/aes_generic.c b/crypto/aes_generic.c
index 9401dca85e87..6683260475f9 100644
--- a/crypto/aes_generic.c
+++ b/crypto/aes_generic.c
@@ -52,6 +52,7 @@
     s/RIJNDAEL(d_key)/D_KEY/g
 */
 
+#include <crypto/aes.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
@@ -59,11 +60,6 @@
 #include <linux/crypto.h>
 #include <asm/byteorder.h>
 
-#define AES_MIN_KEY_SIZE	16
-#define AES_MAX_KEY_SIZE	32
-
-#define AES_BLOCK_SIZE		16
-
 /*
  * #define byte(x, nr) ((unsigned char)((x) >> (nr*8))) 
  */
diff --git a/drivers/crypto/geode-aes.c b/drivers/crypto/geode-aes.c
index fa4c9904346f..5008a1cddffb 100644
--- a/drivers/crypto/geode-aes.c
+++ b/drivers/crypto/geode-aes.c
@@ -13,6 +13,7 @@
 #include <linux/crypto.h>
 #include <linux/spinlock.h>
 #include <crypto/algapi.h>
+#include <crypto/aes.h>
 
 #include <asm/io.h>
 #include <asm/delay.h>
diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c
index 5f7e71810489..c33334ac987e 100644
--- a/drivers/crypto/padlock-aes.c
+++ b/drivers/crypto/padlock-aes.c
@@ -44,6 +44,7 @@
  */
 
 #include <crypto/algapi.h>
+#include <crypto/aes.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
@@ -53,9 +54,6 @@
 #include <asm/byteorder.h>
 #include "padlock.h"
 
-#define AES_MIN_KEY_SIZE	16	/* in uint8_t units */
-#define AES_MAX_KEY_SIZE	32	/* ditto */
-#define AES_BLOCK_SIZE		16	/* ditto */
 #define AES_EXTENDED_KEY_SIZE	64	/* in uint32_t units */
 #define AES_EXTENDED_KEY_SIZE_B	(AES_EXTENDED_KEY_SIZE * sizeof(uint32_t))
 
diff --git a/include/crypto/aes.h b/include/crypto/aes.h
new file mode 100644
index 000000000000..9ff842fc6b89
--- /dev/null
+++ b/include/crypto/aes.h
@@ -0,0 +1,15 @@
+/*
+ * Common values for AES algorithms
+ */
+
+#ifndef _CRYPTO_AES_H
+#define _CRYPTO_AES_H
+
+#define AES_MIN_KEY_SIZE	16
+#define AES_MAX_KEY_SIZE	32
+#define AES_KEYSIZE_128		16
+#define AES_KEYSIZE_192		24
+#define AES_KEYSIZE_256		32
+#define AES_BLOCK_SIZE		16
+
+#endif
-- 
cgit v1.2.3


From 96e82e4551d38e0863b366a7b61185bc4a9946cc Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <sebastian@breakpoint.cc>
Date: Thu, 8 Nov 2007 21:20:30 +0800
Subject: [CRYPTO] aes-generic: Make key generation exportable

This patch exports four tables and the set_key() routine. This ressources
can be shared by other AES implementations (aes-x86_64 for instance).
The decryption key has been turned around (deckey[0] is the first piece
of the key instead of deckey[keylen+20]). The encrypt/decrypt functions
are looking now identical (except they are using different tables and
key).

Signed-off-by: Sebastian Siewior <sebastian@breakpoint.cc>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/aes_generic.c | 249 +++++++++++++++++++++++++--------------------------
 include/crypto/aes.h |  16 ++++
 2 files changed, 136 insertions(+), 129 deletions(-)

(limited to 'include')

diff --git a/crypto/aes_generic.c b/crypto/aes_generic.c
index df8df4d346d2..cf30af74480f 100644
--- a/crypto/aes_generic.c
+++ b/crypto/aes_generic.c
@@ -47,11 +47,6 @@
  * ---------------------------------------------------------------------------
  */
 
-/* Some changes from the Gladman version:
-    s/RIJNDAEL(e_key)/E_KEY/g
-    s/RIJNDAEL(d_key)/D_KEY/g
-*/
-
 #include <crypto/aes.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -60,32 +55,26 @@
 #include <linux/crypto.h>
 #include <asm/byteorder.h>
 
-/*
- * #define byte(x, nr) ((unsigned char)((x) >> (nr*8))) 
- */
 static inline u8 byte(const u32 x, const unsigned n)
 {
 	return x >> (n << 3);
 }
 
-struct aes_ctx {
-	int key_length;
-	u32 buf[120];
-};
-
-#define E_KEY (&ctx->buf[0])
-#define D_KEY (&ctx->buf[60])
-
 static u8 pow_tab[256] __initdata;
 static u8 log_tab[256] __initdata;
 static u8 sbx_tab[256] __initdata;
 static u8 isb_tab[256] __initdata;
 static u32 rco_tab[10];
-static u32 ft_tab[4][256];
-static u32 it_tab[4][256];
 
-static u32 fl_tab[4][256];
-static u32 il_tab[4][256];
+u32 crypto_ft_tab[4][256];
+u32 crypto_fl_tab[4][256];
+u32 crypto_it_tab[4][256];
+u32 crypto_il_tab[4][256];
+
+EXPORT_SYMBOL_GPL(crypto_ft_tab);
+EXPORT_SYMBOL_GPL(crypto_fl_tab);
+EXPORT_SYMBOL_GPL(crypto_it_tab);
+EXPORT_SYMBOL_GPL(crypto_il_tab);
 
 static inline u8 __init f_mult(u8 a, u8 b)
 {
@@ -134,37 +123,37 @@ static void __init gen_tabs(void)
 		p = sbx_tab[i];
 
 		t = p;
-		fl_tab[0][i] = t;
-		fl_tab[1][i] = rol32(t, 8);
-		fl_tab[2][i] = rol32(t, 16);
-		fl_tab[3][i] = rol32(t, 24);
+		crypto_fl_tab[0][i] = t;
+		crypto_fl_tab[1][i] = rol32(t, 8);
+		crypto_fl_tab[2][i] = rol32(t, 16);
+		crypto_fl_tab[3][i] = rol32(t, 24);
 
 		t = ((u32) ff_mult(2, p)) |
 		    ((u32) p << 8) |
 		    ((u32) p << 16) | ((u32) ff_mult(3, p) << 24);
 
-		ft_tab[0][i] = t;
-		ft_tab[1][i] = rol32(t, 8);
-		ft_tab[2][i] = rol32(t, 16);
-		ft_tab[3][i] = rol32(t, 24);
+		crypto_ft_tab[0][i] = t;
+		crypto_ft_tab[1][i] = rol32(t, 8);
+		crypto_ft_tab[2][i] = rol32(t, 16);
+		crypto_ft_tab[3][i] = rol32(t, 24);
 
 		p = isb_tab[i];
 
 		t = p;
-		il_tab[0][i] = t;
-		il_tab[1][i] = rol32(t, 8);
-		il_tab[2][i] = rol32(t, 16);
-		il_tab[3][i] = rol32(t, 24);
+		crypto_il_tab[0][i] = t;
+		crypto_il_tab[1][i] = rol32(t, 8);
+		crypto_il_tab[2][i] = rol32(t, 16);
+		crypto_il_tab[3][i] = rol32(t, 24);
 
 		t = ((u32) ff_mult(14, p)) |
 		    ((u32) ff_mult(9, p) << 8) |
 		    ((u32) ff_mult(13, p) << 16) |
 		    ((u32) ff_mult(11, p) << 24);
 
-		it_tab[0][i] = t;
-		it_tab[1][i] = rol32(t, 8);
-		it_tab[2][i] = rol32(t, 16);
-		it_tab[3][i] = rol32(t, 24);
+		crypto_it_tab[0][i] = t;
+		crypto_it_tab[1][i] = rol32(t, 8);
+		crypto_it_tab[2][i] = rol32(t, 16);
+		crypto_it_tab[3][i] = rol32(t, 24);
 	}
 }
 
@@ -184,69 +173,69 @@ static void __init gen_tabs(void)
 } while (0)
 
 #define ls_box(x)		\
-	fl_tab[0][byte(x, 0)] ^	\
-	fl_tab[1][byte(x, 1)] ^	\
-	fl_tab[2][byte(x, 2)] ^	\
-	fl_tab[3][byte(x, 3)]
+	crypto_fl_tab[0][byte(x, 0)] ^	\
+	crypto_fl_tab[1][byte(x, 1)] ^	\
+	crypto_fl_tab[2][byte(x, 2)] ^	\
+	crypto_fl_tab[3][byte(x, 3)]
 
 #define loop4(i)	do {		\
 	t = ror32(t, 8);		\
 	t = ls_box(t) ^ rco_tab[i];	\
-	t ^= E_KEY[4 * i];		\
-	E_KEY[4 * i + 4] = t;		\
-	t ^= E_KEY[4 * i + 1];		\
-	E_KEY[4 * i + 5] = t;		\
-	t ^= E_KEY[4 * i + 2];		\
-	E_KEY[4 * i + 6] = t;		\
-	t ^= E_KEY[4 * i + 3];		\
-	E_KEY[4 * i + 7] = t;		\
+	t ^= ctx->key_enc[4 * i];		\
+	ctx->key_enc[4 * i + 4] = t;		\
+	t ^= ctx->key_enc[4 * i + 1];		\
+	ctx->key_enc[4 * i + 5] = t;		\
+	t ^= ctx->key_enc[4 * i + 2];		\
+	ctx->key_enc[4 * i + 6] = t;		\
+	t ^= ctx->key_enc[4 * i + 3];		\
+	ctx->key_enc[4 * i + 7] = t;		\
 } while (0)
 
 #define loop6(i)	do {		\
 	t = ror32(t, 8);		\
 	t = ls_box(t) ^ rco_tab[i];	\
-	t ^= E_KEY[6 * i];		\
-	E_KEY[6 * i + 6] = t;		\
-	t ^= E_KEY[6 * i + 1];		\
-	E_KEY[6 * i + 7] = t;		\
-	t ^= E_KEY[6 * i + 2];		\
-	E_KEY[6 * i + 8] = t;		\
-	t ^= E_KEY[6 * i + 3];		\
-	E_KEY[6 * i + 9] = t;		\
-	t ^= E_KEY[6 * i + 4];		\
-	E_KEY[6 * i + 10] = t;		\
-	t ^= E_KEY[6 * i + 5];		\
-	E_KEY[6 * i + 11] = t;		\
+	t ^= ctx->key_enc[6 * i];		\
+	ctx->key_enc[6 * i + 6] = t;		\
+	t ^= ctx->key_enc[6 * i + 1];		\
+	ctx->key_enc[6 * i + 7] = t;		\
+	t ^= ctx->key_enc[6 * i + 2];		\
+	ctx->key_enc[6 * i + 8] = t;		\
+	t ^= ctx->key_enc[6 * i + 3];		\
+	ctx->key_enc[6 * i + 9] = t;		\
+	t ^= ctx->key_enc[6 * i + 4];		\
+	ctx->key_enc[6 * i + 10] = t;		\
+	t ^= ctx->key_enc[6 * i + 5];		\
+	ctx->key_enc[6 * i + 11] = t;		\
 } while (0)
 
 #define loop8(i)	do {			\
 	t = ror32(t, 8);			\
 	t = ls_box(t) ^ rco_tab[i];		\
-	t ^= E_KEY[8 * i];			\
-	E_KEY[8 * i + 8] = t;			\
-	t ^= E_KEY[8 * i + 1];			\
-	E_KEY[8 * i + 9] = t;			\
-	t ^= E_KEY[8 * i + 2];			\
-	E_KEY[8 * i + 10] = t;			\
-	t ^= E_KEY[8 * i + 3];			\
-	E_KEY[8 * i + 11] = t;			\
-	t  = E_KEY[8 * i + 4] ^ ls_box(t);	\
-	E_KEY[8 * i + 12] = t;			\
-	t ^= E_KEY[8 * i + 5];			\
-	E_KEY[8 * i + 13] = t;			\
-	t ^= E_KEY[8 * i + 6];			\
-	E_KEY[8 * i + 14] = t;			\
-	t ^= E_KEY[8 * i + 7];			\
-	E_KEY[8 * i + 15] = t;			\
+	t ^= ctx->key_enc[8 * i];			\
+	ctx->key_enc[8 * i + 8] = t;			\
+	t ^= ctx->key_enc[8 * i + 1];			\
+	ctx->key_enc[8 * i + 9] = t;			\
+	t ^= ctx->key_enc[8 * i + 2];			\
+	ctx->key_enc[8 * i + 10] = t;			\
+	t ^= ctx->key_enc[8 * i + 3];			\
+	ctx->key_enc[8 * i + 11] = t;			\
+	t  = ctx->key_enc[8 * i + 4] ^ ls_box(t);	\
+	ctx->key_enc[8 * i + 12] = t;			\
+	t ^= ctx->key_enc[8 * i + 5];			\
+	ctx->key_enc[8 * i + 13] = t;			\
+	t ^= ctx->key_enc[8 * i + 6];			\
+	ctx->key_enc[8 * i + 14] = t;			\
+	t ^= ctx->key_enc[8 * i + 7];			\
+	ctx->key_enc[8 * i + 15] = t;			\
 } while (0)
 
-static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
+int crypto_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 		unsigned int key_len)
 {
-	struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
 	u32 *flags = &tfm->crt_flags;
-	u32 i, t, u, v, w;
+	u32 i, t, u, v, w, j;
 
 	if (key_len % 8) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
@@ -255,54 +244,55 @@ static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 
 	ctx->key_length = key_len;
 
-	E_KEY[0] = le32_to_cpu(key[0]);
-	E_KEY[1] = le32_to_cpu(key[1]);
-	E_KEY[2] = le32_to_cpu(key[2]);
-	E_KEY[3] = le32_to_cpu(key[3]);
+	ctx->key_dec[key_len + 24] = ctx->key_enc[0] = le32_to_cpu(key[0]);
+	ctx->key_dec[key_len + 25] = ctx->key_enc[1] = le32_to_cpu(key[1]);
+	ctx->key_dec[key_len + 26] = ctx->key_enc[2] = le32_to_cpu(key[2]);
+	ctx->key_dec[key_len + 27] = ctx->key_enc[3] = le32_to_cpu(key[3]);
 
 	switch (key_len) {
 	case 16:
-		t = E_KEY[3];
+		t = ctx->key_enc[3];
 		for (i = 0; i < 10; ++i)
 			loop4(i);
 		break;
 
 	case 24:
-		E_KEY[4] = le32_to_cpu(key[4]);
-		t = E_KEY[5] = le32_to_cpu(key[5]);
+		ctx->key_enc[4] = le32_to_cpu(key[4]);
+		t = ctx->key_enc[5] = le32_to_cpu(key[5]);
 		for (i = 0; i < 8; ++i)
 			loop6(i);
 		break;
 
 	case 32:
-		E_KEY[4] = le32_to_cpu(key[4]);
-		E_KEY[5] = le32_to_cpu(key[5]);
-		E_KEY[6] = le32_to_cpu(key[6]);
-		t = E_KEY[7] = le32_to_cpu(key[7]);
+		ctx->key_enc[4] = le32_to_cpu(key[4]);
+		ctx->key_enc[5] = le32_to_cpu(key[5]);
+		ctx->key_enc[6] = le32_to_cpu(key[6]);
+		t = ctx->key_enc[7] = le32_to_cpu(key[7]);
 		for (i = 0; i < 7; ++i)
 			loop8(i);
 		break;
 	}
 
-	D_KEY[0] = E_KEY[0];
-	D_KEY[1] = E_KEY[1];
-	D_KEY[2] = E_KEY[2];
-	D_KEY[3] = E_KEY[3];
+	ctx->key_dec[0] = ctx->key_enc[key_len + 24];
+	ctx->key_dec[1] = ctx->key_enc[key_len + 25];
+	ctx->key_dec[2] = ctx->key_enc[key_len + 26];
+	ctx->key_dec[3] = ctx->key_enc[key_len + 27];
 
 	for (i = 4; i < key_len + 24; ++i) {
-		imix_col(D_KEY[i], E_KEY[i]);
+		j = key_len + 24 - (i & ~3) + (i & 3);
+		imix_col(ctx->key_dec[j], ctx->key_enc[i]);
 	}
-
 	return 0;
 }
+EXPORT_SYMBOL_GPL(crypto_aes_set_key);
 
 /* encrypt a block of text */
 
 #define f_rn(bo, bi, n, k)	do {				\
-	bo[n] = ft_tab[0][byte(bi[n], 0)] ^			\
-		ft_tab[1][byte(bi[(n + 1) & 3], 1)] ^		\
-		ft_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
-		ft_tab[3][byte(bi[(n + 3) & 3], 3)] ^ *(k + n);	\
+	bo[n] = crypto_ft_tab[0][byte(bi[n], 0)] ^			\
+		crypto_ft_tab[1][byte(bi[(n + 1) & 3], 1)] ^		\
+		crypto_ft_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
+		crypto_ft_tab[3][byte(bi[(n + 3) & 3], 3)] ^ *(k + n);	\
 } while (0)
 
 #define f_nround(bo, bi, k)	do {\
@@ -314,10 +304,10 @@ static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 } while (0)
 
 #define f_rl(bo, bi, n, k)	do {				\
-	bo[n] = fl_tab[0][byte(bi[n], 0)] ^			\
-		fl_tab[1][byte(bi[(n + 1) & 3], 1)] ^		\
-		fl_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
-		fl_tab[3][byte(bi[(n + 3) & 3], 3)] ^ *(k + n);	\
+	bo[n] = crypto_fl_tab[0][byte(bi[n], 0)] ^			\
+		crypto_fl_tab[1][byte(bi[(n + 1) & 3], 1)] ^		\
+		crypto_fl_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
+		crypto_fl_tab[3][byte(bi[(n + 3) & 3], 3)] ^ *(k + n);	\
 } while (0)
 
 #define f_lround(bo, bi, k)	do {\
@@ -329,23 +319,24 @@ static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 
 static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
 {
-	const struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	const struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *src = (const __le32 *)in;
 	__le32 *dst = (__le32 *)out;
 	u32 b0[4], b1[4];
-	const u32 *kp = E_KEY + 4;
+	const u32 *kp = ctx->key_enc + 4;
+	const int key_len = ctx->key_length;
 
-	b0[0] = le32_to_cpu(src[0]) ^ E_KEY[0];
-	b0[1] = le32_to_cpu(src[1]) ^ E_KEY[1];
-	b0[2] = le32_to_cpu(src[2]) ^ E_KEY[2];
-	b0[3] = le32_to_cpu(src[3]) ^ E_KEY[3];
+	b0[0] = le32_to_cpu(src[0]) ^ ctx->key_enc[0];
+	b0[1] = le32_to_cpu(src[1]) ^ ctx->key_enc[1];
+	b0[2] = le32_to_cpu(src[2]) ^ ctx->key_enc[2];
+	b0[3] = le32_to_cpu(src[3]) ^ ctx->key_enc[3];
 
-	if (ctx->key_length > 24) {
+	if (key_len > 24) {
 		f_nround(b1, b0, kp);
 		f_nround(b0, b1, kp);
 	}
 
-	if (ctx->key_length > 16) {
+	if (key_len > 16) {
 		f_nround(b1, b0, kp);
 		f_nround(b0, b1, kp);
 	}
@@ -370,10 +361,10 @@ static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
 /* decrypt a block of text */
 
 #define i_rn(bo, bi, n, k)	do {				\
-	bo[n] = it_tab[0][byte(bi[n], 0)] ^			\
-		it_tab[1][byte(bi[(n + 3) & 3], 1)] ^		\
-		it_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
-		it_tab[3][byte(bi[(n + 1) & 3], 3)] ^ *(k + n);	\
+	bo[n] = crypto_it_tab[0][byte(bi[n], 0)] ^			\
+		crypto_it_tab[1][byte(bi[(n + 3) & 3], 1)] ^		\
+		crypto_it_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
+		crypto_it_tab[3][byte(bi[(n + 1) & 3], 3)] ^ *(k + n);	\
 } while (0)
 
 #define i_nround(bo, bi, k)	do {\
@@ -381,14 +372,14 @@ static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
 	i_rn(bo, bi, 1, k);	\
 	i_rn(bo, bi, 2, k);	\
 	i_rn(bo, bi, 3, k);	\
-	k -= 4;			\
+	k += 4;			\
 } while (0)
 
 #define i_rl(bo, bi, n, k)	do {			\
-	bo[n] = il_tab[0][byte(bi[n], 0)] ^		\
-	il_tab[1][byte(bi[(n + 3) & 3], 1)] ^		\
-	il_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
-	il_tab[3][byte(bi[(n + 1) & 3], 3)] ^ *(k + n);	\
+	bo[n] = crypto_il_tab[0][byte(bi[n], 0)] ^		\
+	crypto_il_tab[1][byte(bi[(n + 3) & 3], 1)] ^		\
+	crypto_il_tab[2][byte(bi[(n + 2) & 3], 2)] ^		\
+	crypto_il_tab[3][byte(bi[(n + 1) & 3], 3)] ^ *(k + n);	\
 } while (0)
 
 #define i_lround(bo, bi, k)	do {\
@@ -400,17 +391,17 @@ static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
 
 static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
 {
-	const struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	const struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *src = (const __le32 *)in;
 	__le32 *dst = (__le32 *)out;
 	u32 b0[4], b1[4];
 	const int key_len = ctx->key_length;
-	const u32 *kp = D_KEY + key_len + 20;
+	const u32 *kp = ctx->key_dec + 4;
 
-	b0[0] = le32_to_cpu(src[0]) ^ E_KEY[key_len + 24];
-	b0[1] = le32_to_cpu(src[1]) ^ E_KEY[key_len + 25];
-	b0[2] = le32_to_cpu(src[2]) ^ E_KEY[key_len + 26];
-	b0[3] = le32_to_cpu(src[3]) ^ E_KEY[key_len + 27];
+	b0[0] = le32_to_cpu(src[0]) ^  ctx->key_dec[0];
+	b0[1] = le32_to_cpu(src[1]) ^  ctx->key_dec[1];
+	b0[2] = le32_to_cpu(src[2]) ^  ctx->key_dec[2];
+	b0[3] = le32_to_cpu(src[3]) ^  ctx->key_dec[3];
 
 	if (key_len > 24) {
 		i_nround(b1, b0, kp);
@@ -445,7 +436,7 @@ static struct crypto_alg aes_alg = {
 	.cra_priority		=	100,
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	AES_BLOCK_SIZE,
-	.cra_ctxsize		=	sizeof(struct aes_ctx),
+	.cra_ctxsize		=	sizeof(struct crypto_aes_ctx),
 	.cra_alignmask		=	3,
 	.cra_module		=	THIS_MODULE,
 	.cra_list		=	LIST_HEAD_INIT(aes_alg.cra_list),
@@ -453,7 +444,7 @@ static struct crypto_alg aes_alg = {
 		.cipher = {
 			.cia_min_keysize	=	AES_MIN_KEY_SIZE,
 			.cia_max_keysize	=	AES_MAX_KEY_SIZE,
-			.cia_setkey		=	aes_set_key,
+			.cia_setkey		=	crypto_aes_set_key,
 			.cia_encrypt		=	aes_encrypt,
 			.cia_decrypt		=	aes_decrypt
 		}
diff --git a/include/crypto/aes.h b/include/crypto/aes.h
index 9ff842fc6b89..d480b76715a8 100644
--- a/include/crypto/aes.h
+++ b/include/crypto/aes.h
@@ -5,6 +5,9 @@
 #ifndef _CRYPTO_AES_H
 #define _CRYPTO_AES_H
 
+#include <linux/types.h>
+#include <linux/crypto.h>
+
 #define AES_MIN_KEY_SIZE	16
 #define AES_MAX_KEY_SIZE	32
 #define AES_KEYSIZE_128		16
@@ -12,4 +15,17 @@
 #define AES_KEYSIZE_256		32
 #define AES_BLOCK_SIZE		16
 
+struct crypto_aes_ctx {
+	u32 key_length;
+	u32 key_enc[60];
+	u32 key_dec[60];
+};
+
+extern u32 crypto_ft_tab[4][256];
+extern u32 crypto_fl_tab[4][256];
+extern u32 crypto_it_tab[4][256];
+extern u32 crypto_il_tab[4][256];
+
+int crypto_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
+		unsigned int key_len);
 #endif
-- 
cgit v1.2.3


From cd12fb906d2591e80da9edcbd4794b9b916d7489 Mon Sep 17 00:00:00 2001
From: Jonathan Lynch <jonathan.lynch@intel.com>
Date: Sat, 10 Nov 2007 20:08:25 +0800
Subject: [CRYPTO] sha256-generic: Extend sha256_generic.c to support SHA-224

Resubmitting this patch which extends sha256_generic.c to support SHA-224 as
described in FIPS 180-2 and RFC 3874. HMAC-SHA-224 as described in RFC4231
is then supported through the hmac interface.

Patch includes test vectors for SHA-224 and HMAC-SHA-224.

SHA-224 chould be chosen as a hash algorithm when 112 bits of security
strength is required.

Patch generated against the 2.6.24-rc1 kernel and tested against
2.6.24-rc1-git14 which includes fix for scatter gather implementation for HMAC.

Signed-off-by: Jonathan Lynch <jonathan.lynch@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig          |   5 +-
 crypto/sha256_generic.c |  72 +++++++++++++++++++++---
 crypto/tcrypt.c         |  22 +++++++-
 crypto/tcrypt.h         | 142 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/crypto/sha.h    |  12 ++++
 5 files changed, 241 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index cf115b14079e..7758454b9f17 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -91,7 +91,7 @@ config CRYPTO_SHA1
 	  SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2).
 
 config CRYPTO_SHA256
-	tristate "SHA256 digest algorithm"
+	tristate "SHA224 and SHA256 digest algorithm"
 	select CRYPTO_ALGAPI
 	help
 	  SHA256 secure hash standard (DFIPS 180-2).
@@ -99,6 +99,9 @@ config CRYPTO_SHA256
 	  This version of SHA implements a 256 bit hash with 128 bits of
 	  security against collision attacks.
 
+          This code also includes SHA-224, a 224 bit hash with 112 bits
+          of security against collision attacks.
+
 config CRYPTO_SHA512
 	tristate "SHA384 and SHA512 digest algorithms"
 	select CRYPTO_ALGAPI
diff --git a/crypto/sha256_generic.c b/crypto/sha256_generic.c
index fd3918be58b5..3cc93fd61043 100644
--- a/crypto/sha256_generic.c
+++ b/crypto/sha256_generic.c
@@ -9,6 +9,7 @@
  * Copyright (c) Jean-Luc Cooke <jlcooke@certainkey.com>
  * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk>
  * Copyright (c) 2002 James Morris <jmorris@intercode.com.au>
+ * SHA224 Support Copyright 2007 Intel Corporation <jonathan.lynch@intel.com>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the Free
@@ -218,6 +219,22 @@ static void sha256_transform(u32 *state, const u8 *input)
 	memset(W, 0, 64 * sizeof(u32));
 }
 
+
+static void sha224_init(struct crypto_tfm *tfm)
+{
+	struct sha256_ctx *sctx = crypto_tfm_ctx(tfm);
+	sctx->state[0] = SHA224_H0;
+	sctx->state[1] = SHA224_H1;
+	sctx->state[2] = SHA224_H2;
+	sctx->state[3] = SHA224_H3;
+	sctx->state[4] = SHA224_H4;
+	sctx->state[5] = SHA224_H5;
+	sctx->state[6] = SHA224_H6;
+	sctx->state[7] = SHA224_H7;
+	sctx->count[0] = 0;
+	sctx->count[1] = 0;
+}
+
 static void sha256_init(struct crypto_tfm *tfm)
 {
 	struct sha256_ctx *sctx = crypto_tfm_ctx(tfm);
@@ -294,8 +311,17 @@ static void sha256_final(struct crypto_tfm *tfm, u8 *out)
 	memset(sctx, 0, sizeof(*sctx));
 }
 
+static void sha224_final(struct crypto_tfm *tfm, u8 *hash)
+{
+	u8 D[SHA256_DIGEST_SIZE];
+
+	sha256_final(tfm, D);
+
+	memcpy(hash, D, SHA224_DIGEST_SIZE);
+	memset(D, 0, SHA256_DIGEST_SIZE);
+}
 
-static struct crypto_alg alg = {
+static struct crypto_alg sha256 = {
 	.cra_name	=	"sha256",
 	.cra_driver_name=	"sha256-generic",
 	.cra_flags	=	CRYPTO_ALG_TYPE_DIGEST,
@@ -303,28 +329,58 @@ static struct crypto_alg alg = {
 	.cra_ctxsize	=	sizeof(struct sha256_ctx),
 	.cra_module	=	THIS_MODULE,
 	.cra_alignmask	=	3,
-	.cra_list       =       LIST_HEAD_INIT(alg.cra_list),
+	.cra_list	=	LIST_HEAD_INIT(sha256.cra_list),
 	.cra_u		=	{ .digest = {
 	.dia_digestsize	=	SHA256_DIGEST_SIZE,
-	.dia_init   	= 	sha256_init,
-	.dia_update 	=	sha256_update,
-	.dia_final  	=	sha256_final } }
+	.dia_init	=	sha256_init,
+	.dia_update	=	sha256_update,
+	.dia_final	=	sha256_final } }
+};
+
+static struct crypto_alg sha224 = {
+	.cra_name	= "sha224",
+	.cra_driver_name = "sha224-generic",
+	.cra_flags	= CRYPTO_ALG_TYPE_DIGEST,
+	.cra_blocksize	= SHA224_BLOCK_SIZE,
+	.cra_ctxsize	= sizeof(struct sha256_ctx),
+	.cra_module	= THIS_MODULE,
+	.cra_alignmask	= 3,
+	.cra_list	= LIST_HEAD_INIT(sha224.cra_list),
+	.cra_u		= { .digest = {
+	.dia_digestsize = SHA224_DIGEST_SIZE,
+	.dia_init	= sha224_init,
+	.dia_update	= sha256_update,
+	.dia_final	= sha224_final } }
 };
 
 static int __init init(void)
 {
-	return crypto_register_alg(&alg);
+	int ret = 0;
+
+	ret = crypto_register_alg(&sha224);
+
+	if (ret < 0)
+		return ret;
+
+	ret = crypto_register_alg(&sha256);
+
+	if (ret < 0)
+		crypto_unregister_alg(&sha224);
+
+	return ret;
 }
 
 static void __exit fini(void)
 {
-	crypto_unregister_alg(&alg);
+	crypto_unregister_alg(&sha224);
+	crypto_unregister_alg(&sha256);
 }
 
 module_init(init);
 module_exit(fini);
 
 MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("SHA256 Secure Hash Algorithm");
+MODULE_DESCRIPTION("SHA-224 and SHA-256 Secure Hash Algorithm");
 
+MODULE_ALIAS("sha224");
 MODULE_ALIAS("sha256");
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index aa84bc4f2313..4d364ccacbb2 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -12,6 +12,7 @@
  * Software Foundation; either version 2 of the License, or (at your option)
  * any later version.
  *
+ * 2007-11-06 Added SHA-224 and SHA-224-HMAC tests
  * 2006-12-07 Added SHA384 HMAC and SHA512 HMAC tests
  * 2004-08-09 Added cipher speed tests (Reyk Floeter <reyk@vantronix.net>)
  * 2003-09-14 Rewritten by Kartikey Mahendra Bhatt
@@ -74,8 +75,9 @@ static char *xbuf;
 static char *tvmem;
 
 static char *check[] = {
-	"des", "md5", "des3_ede", "rot13", "sha1", "sha256", "blowfish",
-	"twofish", "serpent", "sha384", "sha512", "md4", "aes", "cast6",
+	"des", "md5", "des3_ede", "rot13", "sha1", "sha224", "sha256",
+	"blowfish", "twofish", "serpent", "sha384", "sha512", "md4", "aes",
+	"cast6", "arc4", "michael_mic", "deflate", "crc32c", "tea", "xtea",
 	"arc4", "michael_mic", "deflate", "crc32c", "tea", "xtea",
 	"khazad", "wp512", "wp384", "wp256", "tnepres", "xeta",  "fcrypt",
 	"camellia", "seed", NULL
@@ -918,6 +920,8 @@ static void do_test(void)
 
 		test_hash("md4", md4_tv_template, MD4_TEST_VECTORS);
 
+		test_hash("sha224", sha224_tv_template, SHA224_TEST_VECTORS);
+
 		test_hash("sha256", sha256_tv_template, SHA256_TEST_VECTORS);
 
 		//BLOWFISH
@@ -1067,6 +1071,8 @@ static void do_test(void)
 			  HMAC_MD5_TEST_VECTORS);
 		test_hash("hmac(sha1)", hmac_sha1_tv_template,
 			  HMAC_SHA1_TEST_VECTORS);
+		test_hash("hmac(sha224)", hmac_sha224_tv_template,
+			  HMAC_SHA224_TEST_VECTORS);
 		test_hash("hmac(sha256)", hmac_sha256_tv_template,
 			  HMAC_SHA256_TEST_VECTORS);
 		test_hash("hmac(sha384)", hmac_sha384_tv_template,
@@ -1299,6 +1305,9 @@ static void do_test(void)
 			    camellia_cbc_dec_tv_template,
 			    CAMELLIA_CBC_DEC_TEST_VECTORS);
 		break;
+	case 33:
+		test_hash("sha224", sha224_tv_template, SHA224_TEST_VECTORS);
+		break;
 
 	case 100:
 		test_hash("hmac(md5)", hmac_md5_tv_template,
@@ -1324,7 +1333,10 @@ static void do_test(void)
 		test_hash("hmac(sha512)", hmac_sha512_tv_template,
 			  HMAC_SHA512_TEST_VECTORS);
 		break;
-
+	case 105:
+		test_hash("hmac(sha224)", hmac_sha224_tv_template,
+			  HMAC_SHA224_TEST_VECTORS);
+		break;
 
 	case 200:
 		test_cipher_speed("ecb(aes)", ENCRYPT, sec, NULL, 0,
@@ -1459,6 +1471,10 @@ static void do_test(void)
 		test_hash_speed("tgr192", sec, generic_hash_speed_template);
 		if (mode > 300 && mode < 400) break;
 
+	case 313:
+		test_hash_speed("sha224", sec, generic_hash_speed_template);
+		if (mode > 300 && mode < 400) break;
+
 	case 399:
 		break;
 
diff --git a/crypto/tcrypt.h b/crypto/tcrypt.h
index f7f9b2379270..b91585ea1362 100644
--- a/crypto/tcrypt.h
+++ b/crypto/tcrypt.h
@@ -173,6 +173,33 @@ static struct hash_testvec sha1_tv_template[] = {
 	}
 };
 
+
+/*
+ * SHA224 test vectors from from FIPS PUB 180-2
+ */
+#define SHA224_TEST_VECTORS     2
+
+static struct hash_testvec sha224_tv_template[] = {
+	{
+		.plaintext = "abc",
+		.psize  = 3,
+		.digest = { 0x23, 0x09, 0x7D, 0x22, 0x34, 0x05, 0xD8, 0x22,
+			0x86, 0x42, 0xA4, 0x77, 0xBD, 0xA2, 0x55, 0xB3,
+			0x2A, 0xAD, 0xBC, 0xE4, 0xBD, 0xA0, 0xB3, 0xF7,
+			0xE3, 0x6C, 0x9D, 0xA7},
+	}, {
+		.plaintext =
+		"abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq",
+		.psize  = 56,
+		.digest = { 0x75, 0x38, 0x8B, 0x16, 0x51, 0x27, 0x76, 0xCC,
+			0x5D, 0xBA, 0x5D, 0xA1, 0xFD, 0x89, 0x01, 0x50,
+			0xB0, 0xC6, 0x45, 0x5C, 0xB4, 0xF5, 0x8B, 0x19,
+			0x52, 0x52, 0x25, 0x25 },
+		.np     = 2,
+		.tap    = { 28, 28 }
+	}
+};
+
 /*
  * SHA256 test vectors from from NIST
  */
@@ -817,6 +844,121 @@ static struct hash_testvec hmac_sha1_tv_template[] = {
 	},
 };
 
+
+/*
+ * SHA224 HMAC test vectors from RFC4231
+ */
+#define HMAC_SHA224_TEST_VECTORS    4
+
+static struct hash_testvec hmac_sha224_tv_template[] = {
+	{
+		.key    = { 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b,
+			0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b,
+			0x0b, 0x0b, 0x0b, 0x0b },
+		.ksize  = 20,
+		/*  ("Hi There") */
+		.plaintext = { 0x48, 0x69, 0x20, 0x54, 0x68, 0x65, 0x72, 0x65 },
+		.psize  = 8,
+		.digest = { 0x89, 0x6f, 0xb1, 0x12, 0x8a, 0xbb, 0xdf, 0x19,
+			0x68, 0x32, 0x10, 0x7c, 0xd4, 0x9d, 0xf3, 0x3f,
+			0x47, 0xb4, 0xb1, 0x16, 0x99, 0x12, 0xba, 0x4f,
+			0x53, 0x68, 0x4b, 0x22},
+	}, {
+		.key    = { 0x4a, 0x65, 0x66, 0x65 }, /* ("Jefe") */
+		.ksize  = 4,
+		/* ("what do ya want for nothing?") */
+		.plaintext = { 0x77, 0x68, 0x61, 0x74, 0x20, 0x64, 0x6f, 0x20,
+			0x79, 0x61, 0x20, 0x77, 0x61, 0x6e, 0x74, 0x20,
+			0x66, 0x6f, 0x72, 0x20, 0x6e, 0x6f, 0x74, 0x68,
+			0x69, 0x6e, 0x67, 0x3f },
+		.psize  = 28,
+		.digest = { 0xa3, 0x0e, 0x01, 0x09, 0x8b, 0xc6, 0xdb, 0xbf,
+			0x45, 0x69, 0x0f, 0x3a, 0x7e, 0x9e, 0x6d, 0x0f,
+			0x8b, 0xbe, 0xa2, 0xa3, 0x9e, 0x61, 0x48, 0x00,
+			0x8f, 0xd0, 0x5e, 0x44 },
+		.np = 4,
+		.tap    = { 7, 7, 7, 7 }
+	}, {
+		.key    = { 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa },
+		.ksize  = 131,
+		/* ("Test Using Larger Than Block-Size Key - Hash Key First") */
+		.plaintext = { 0x54, 0x65, 0x73, 0x74, 0x20, 0x55, 0x73, 0x69,
+			0x6e, 0x67, 0x20, 0x4c, 0x61, 0x72, 0x67, 0x65,
+			0x72, 0x20, 0x54, 0x68, 0x61, 0x6e, 0x20, 0x42,
+			0x6c, 0x6f, 0x63, 0x6b, 0x2d, 0x53, 0x69, 0x7a,
+			0x65, 0x20, 0x4b, 0x65, 0x79, 0x20, 0x2d, 0x20,
+			0x48, 0x61, 0x73, 0x68, 0x20, 0x4b, 0x65, 0x79,
+			0x20, 0x46, 0x69, 0x72, 0x73, 0x74 },
+		.psize  = 54,
+		.digest = { 0x95, 0xe9, 0xa0, 0xdb, 0x96, 0x20, 0x95, 0xad,
+			0xae, 0xbe, 0x9b, 0x2d, 0x6f, 0x0d, 0xbc, 0xe2,
+			0xd4, 0x99, 0xf1, 0x12, 0xf2, 0xd2, 0xb7, 0x27,
+			0x3f, 0xa6, 0x87, 0x0e },
+	}, {
+		.key    = { 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa,
+			0xaa, 0xaa, 0xaa },
+		.ksize  = 131,
+		/* ("This is a test using a larger than block-size key and a")
+		(" larger than block-size data. The key needs to be")
+			(" hashed before being used by the HMAC algorithm.") */
+		.plaintext = { 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20,
+			0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0x20, 0x75,
+			0x73, 0x69, 0x6e, 0x67, 0x20, 0x61, 0x20, 0x6c,
+			0x61, 0x72, 0x67, 0x65, 0x72, 0x20, 0x74, 0x68,
+			0x61, 0x6e, 0x20, 0x62, 0x6c, 0x6f, 0x63, 0x6b,
+			0x2d, 0x73, 0x69, 0x7a, 0x65, 0x20, 0x6b, 0x65,
+			0x79, 0x20, 0x61, 0x6e, 0x64, 0x20, 0x61, 0x20,
+			0x6c, 0x61, 0x72, 0x67, 0x65, 0x72, 0x20, 0x74,
+			0x68, 0x61, 0x6e, 0x20, 0x62, 0x6c, 0x6f, 0x63,
+			0x6b, 0x2d, 0x73, 0x69, 0x7a, 0x65, 0x20, 0x64,
+			0x61, 0x74, 0x61, 0x2e, 0x20, 0x54, 0x68, 0x65,
+			0x20, 0x6b, 0x65, 0x79, 0x20, 0x6e, 0x65, 0x65,
+			0x64, 0x73, 0x20, 0x74, 0x6f, 0x20, 0x62, 0x65,
+			0x20, 0x68, 0x61, 0x73, 0x68, 0x65, 0x64, 0x20,
+			0x62, 0x65, 0x66, 0x6f, 0x72, 0x65, 0x20, 0x62,
+			0x65, 0x69, 0x6e, 0x67, 0x20, 0x75, 0x73, 0x65,
+			0x64, 0x20, 0x62, 0x79, 0x20, 0x74, 0x68, 0x65,
+			0x20, 0x48, 0x4d, 0x41, 0x43, 0x20, 0x61, 0x6c,
+			0x67, 0x6f, 0x72, 0x69, 0x74, 0x68, 0x6d, 0x2e },
+		.psize  = 152,
+		.digest = { 0x3a, 0x85, 0x41, 0x66, 0xac, 0x5d, 0x9f, 0x02,
+			0x3f, 0x54, 0xd5, 0x17, 0xd0, 0xb3, 0x9d, 0xbd,
+			0x94, 0x67, 0x70, 0xdb, 0x9c, 0x2b, 0x95, 0xc9,
+			0xf6, 0xf5, 0x65, 0xd1 },
+	},
+};
+
 /*
  * HMAC-SHA256 test vectors from
  * draft-ietf-ipsec-ciph-sha-256-01.txt
diff --git a/include/crypto/sha.h b/include/crypto/sha.h
index 0686e1f7a24b..c0ccc2b1a2d8 100644
--- a/include/crypto/sha.h
+++ b/include/crypto/sha.h
@@ -8,6 +8,9 @@
 #define SHA1_DIGEST_SIZE        20
 #define SHA1_BLOCK_SIZE         64
 
+#define SHA224_DIGEST_SIZE	28
+#define SHA224_BLOCK_SIZE	64
+
 #define SHA256_DIGEST_SIZE      32
 #define SHA256_BLOCK_SIZE       64
 
@@ -23,6 +26,15 @@
 #define SHA1_H3		0x10325476UL
 #define SHA1_H4		0xc3d2e1f0UL
 
+#define SHA224_H0	0xc1059ed8UL
+#define SHA224_H1	0x367cd507UL
+#define SHA224_H2	0x3070dd17UL
+#define SHA224_H3	0xf70e5939UL
+#define SHA224_H4	0xffc00b31UL
+#define SHA224_H5	0x68581511UL
+#define SHA224_H6	0x64f98fa7UL
+#define SHA224_H7	0xbefa4fa4UL
+
 #define SHA256_H0	0x6a09e667UL
 #define SHA256_H1	0xbb67ae85UL
 #define SHA256_H2	0x3c6ef372UL
-- 
cgit v1.2.3


From 332f8840f7095d294f9bb066b175a100bcde214c Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 15 Nov 2007 22:36:07 +0800
Subject: [CRYPTO] ablkcipher: Add distinct ABLKCIPHER type

Up until now we have ablkcipher algorithms have been identified as
type BLKCIPHER with the ASYNC bit set.  This is suboptimal because
ablkcipher refers to two things.  On the one hand it refers to the
top-level ablkcipher interface with requests.  On the other hand it
refers to and algorithm type underneath.

As it is you cannot request a synchronous block cipher algorithm
with the ablkcipher interface on top.  This is a problem because
we want to be able to eventually phase out the blkcipher top-level
interface.

This patch fixes this by making ABLKCIPHER its own type, just as
we have distinct types for HASH and DIGEST.  The type it associated
with the algorithm implementation only.

Which top-level interface is used for synchronous block ciphers is
then determined by the mask that's used.  If it's a specific mask
then the old blkcipher interface is given, otherwise we go with the
new ablkcipher interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/authenc.c           |  2 +-
 crypto/blkcipher.c         |  9 +++------
 crypto/cryptd.c            |  4 ++--
 drivers/crypto/hifn_795x.c |  2 +-
 include/crypto/algapi.h    |  4 ++--
 include/linux/crypto.h     | 18 ++++++++++--------
 6 files changed, 19 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/crypto/authenc.c b/crypto/authenc.c
index 126a529b496d..bc4e608ca841 100644
--- a/crypto/authenc.c
+++ b/crypto/authenc.c
@@ -292,7 +292,7 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 		goto out_put_auth;
 
 	enc = crypto_attr_alg(tb[3], CRYPTO_ALG_TYPE_BLKCIPHER,
-			      CRYPTO_ALG_TYPE_MASK);
+			      CRYPTO_ALG_TYPE_BLKCIPHER_MASK);
 	inst = ERR_PTR(PTR_ERR(enc));
 	if (IS_ERR(enc))
 		goto out_put_auth;
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index f6c67f9d4e5c..180d91451476 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -433,9 +433,8 @@ static unsigned int crypto_blkcipher_ctxsize(struct crypto_alg *alg, u32 type,
 	struct blkcipher_alg *cipher = &alg->cra_blkcipher;
 	unsigned int len = alg->cra_ctxsize;
 
-	type ^= CRYPTO_ALG_ASYNC;
-	mask &= CRYPTO_ALG_ASYNC;
-	if ((type & mask) && cipher->ivsize) {
+	if ((mask & CRYPTO_ALG_TYPE_MASK) == CRYPTO_ALG_TYPE_MASK &&
+	    cipher->ivsize) {
 		len = ALIGN(len, (unsigned long)alg->cra_alignmask + 1);
 		len += cipher->ivsize;
 	}
@@ -482,9 +481,7 @@ static int crypto_init_blkcipher_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 	if (alg->ivsize > PAGE_SIZE / 8)
 		return -EINVAL;
 
-	type ^= CRYPTO_ALG_ASYNC;
-	mask &= CRYPTO_ALG_ASYNC;
-	if (type & mask)
+	if ((mask & CRYPTO_ALG_TYPE_MASK) == CRYPTO_ALG_TYPE_MASK)
 		return crypto_init_blkcipher_ops_sync(tfm);
 	else
 		return crypto_init_blkcipher_ops_async(tfm);
diff --git a/crypto/cryptd.c b/crypto/cryptd.c
index 8bf2da835f7b..1a5c45b96852 100644
--- a/crypto/cryptd.c
+++ b/crypto/cryptd.c
@@ -228,7 +228,7 @@ static struct crypto_instance *cryptd_alloc_blkcipher(
 	struct crypto_alg *alg;
 
 	alg = crypto_get_attr_alg(tb, CRYPTO_ALG_TYPE_BLKCIPHER,
-				  CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
+				  CRYPTO_ALG_TYPE_MASK);
 	if (IS_ERR(alg))
 		return ERR_PTR(PTR_ERR(alg));
 
@@ -236,7 +236,7 @@ static struct crypto_instance *cryptd_alloc_blkcipher(
 	if (IS_ERR(inst))
 		goto out_put_alg;
 
-	inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER | CRYPTO_ALG_ASYNC;
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC;
 	inst->alg.cra_type = &crypto_ablkcipher_type;
 
 	inst->alg.cra_ablkcipher.ivsize = alg->cra_blkcipher.ivsize;
diff --git a/drivers/crypto/hifn_795x.c b/drivers/crypto/hifn_795x.c
index fec32aa1ec52..bf817d4ecae2 100644
--- a/drivers/crypto/hifn_795x.c
+++ b/drivers/crypto/hifn_795x.c
@@ -2361,7 +2361,7 @@ static int hifn_alg_alloc(struct hifn_device *dev, struct hifn_alg_template *t)
 	snprintf(alg->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME, "%s", t->drv_name);
 
 	alg->alg.cra_priority = 300;
-	alg->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER | CRYPTO_ALG_ASYNC;
+	alg->alg.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC;
 	alg->alg.cra_blocksize = t->bsize;
 	alg->alg.cra_ctxsize = sizeof(struct hifn_context);
 	alg->alg.cra_alignmask = 15;
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index b9b05d399d2b..88619f902c10 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -191,7 +191,7 @@ static inline struct crypto_ablkcipher *crypto_spawn_ablkcipher(
 	struct crypto_spawn *spawn)
 {
 	u32 type = CRYPTO_ALG_TYPE_BLKCIPHER;
-	u32 mask = CRYPTO_ALG_TYPE_MASK;
+	u32 mask = CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 
 	return __crypto_ablkcipher_cast(crypto_spawn_tfm(spawn, type, mask));
 }
@@ -200,7 +200,7 @@ static inline struct crypto_blkcipher *crypto_spawn_blkcipher(
 	struct crypto_spawn *spawn)
 {
 	u32 type = CRYPTO_ALG_TYPE_BLKCIPHER;
-	u32 mask = CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC;
+	u32 mask = CRYPTO_ALG_TYPE_MASK;
 
 	return __crypto_blkcipher_cast(crypto_spawn_tfm(spawn, type, mask));
 }
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index f3110ebe894a..f56ae8721bc9 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -33,10 +33,12 @@
 #define CRYPTO_ALG_TYPE_DIGEST		0x00000002
 #define CRYPTO_ALG_TYPE_HASH		0x00000003
 #define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000004
-#define CRYPTO_ALG_TYPE_COMPRESS	0x00000005
-#define CRYPTO_ALG_TYPE_AEAD		0x00000006
+#define CRYPTO_ALG_TYPE_ABLKCIPHER	0x00000005
+#define CRYPTO_ALG_TYPE_COMPRESS	0x00000008
+#define CRYPTO_ALG_TYPE_AEAD		0x00000009
 
 #define CRYPTO_ALG_TYPE_HASH_MASK	0x0000000e
+#define CRYPTO_ALG_TYPE_BLKCIPHER_MASK	0x0000000c
 
 #define CRYPTO_ALG_LARVAL		0x00000010
 #define CRYPTO_ALG_DEAD			0x00000020
@@ -530,7 +532,7 @@ static inline struct crypto_ablkcipher *crypto_alloc_ablkcipher(
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
-	mask |= CRYPTO_ALG_TYPE_MASK;
+	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 
 	return __crypto_ablkcipher_cast(
 		crypto_alloc_base(alg_name, type, mask));
@@ -552,7 +554,7 @@ static inline int crypto_has_ablkcipher(const char *alg_name, u32 type,
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
-	mask |= CRYPTO_ALG_TYPE_MASK;
+	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 
 	return crypto_has_alg(alg_name, type, mask);
 }
@@ -841,9 +843,9 @@ static inline struct crypto_blkcipher *crypto_blkcipher_cast(
 static inline struct crypto_blkcipher *crypto_alloc_blkcipher(
 	const char *alg_name, u32 type, u32 mask)
 {
-	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
+	type &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
-	mask |= CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC;
+	mask |= CRYPTO_ALG_TYPE_MASK;
 
 	return __crypto_blkcipher_cast(crypto_alloc_base(alg_name, type, mask));
 }
@@ -861,9 +863,9 @@ static inline void crypto_free_blkcipher(struct crypto_blkcipher *tfm)
 
 static inline int crypto_has_blkcipher(const char *alg_name, u32 type, u32 mask)
 {
-	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
+	type &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
-	mask |= CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC;
+	mask |= CRYPTO_ALG_TYPE_MASK;
 
 	return crypto_has_alg(alg_name, type, mask);
 }
-- 
cgit v1.2.3


From 984e976f5382ff09351ddd3b023937611396d739 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 21 Nov 2007 12:24:45 +0800
Subject: [HWRNG]: move status polling loop to data_present callbacks

Handle waiting for new random within the drivers themselves, this allows to
use better suited timeouts for the individual rngs.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 drivers/char/hw_random/amd-rng.c    | 12 ++++++++++--
 drivers/char/hw_random/core.c       | 24 ++++++------------------
 drivers/char/hw_random/geode-rng.c  | 12 ++++++++++--
 drivers/char/hw_random/intel-rng.c  | 15 ++++++++++++---
 drivers/char/hw_random/omap-rng.c   | 13 +++++++++++--
 drivers/char/hw_random/pasemi-rng.c | 14 +++++++++++---
 drivers/char/hw_random/via-rng.c    | 19 ++++++++++++-------
 include/linux/hw_random.h           |  2 +-
 8 files changed, 73 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/drivers/char/hw_random/amd-rng.c b/drivers/char/hw_random/amd-rng.c
index 556fd81fa815..c422e870dc52 100644
--- a/drivers/char/hw_random/amd-rng.c
+++ b/drivers/char/hw_random/amd-rng.c
@@ -28,6 +28,7 @@
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/hw_random.h>
+#include <linux/delay.h>
 #include <asm/io.h>
 
 
@@ -52,11 +53,18 @@ MODULE_DEVICE_TABLE(pci, pci_tbl);
 static struct pci_dev *amd_pdev;
 
 
-static int amd_rng_data_present(struct hwrng *rng)
+static int amd_rng_data_present(struct hwrng *rng, int wait)
 {
 	u32 pmbase = (u32)rng->priv;
+	int data, i;
 
-      	return !!(inl(pmbase + 0xF4) & 1);
+	for (i = 0; i < 20; i++) {
+		data = !!(inl(pmbase + 0xF4) & 1);
+		if (data || !wait)
+			break;
+		udelay(10);
+	}
+	return data;
 }
 
 static int amd_rng_data_read(struct hwrng *rng, u32 *data)
diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c
index 26a860adcb38..0118b9817a95 100644
--- a/drivers/char/hw_random/core.c
+++ b/drivers/char/hw_random/core.c
@@ -66,11 +66,11 @@ static inline void hwrng_cleanup(struct hwrng *rng)
 		rng->cleanup(rng);
 }
 
-static inline int hwrng_data_present(struct hwrng *rng)
+static inline int hwrng_data_present(struct hwrng *rng, int wait)
 {
 	if (!rng->data_present)
 		return 1;
-	return rng->data_present(rng);
+	return rng->data_present(rng, wait);
 }
 
 static inline int hwrng_data_read(struct hwrng *rng, u32 *data)
@@ -94,8 +94,7 @@ static ssize_t rng_dev_read(struct file *filp, char __user *buf,
 {
 	u32 data;
 	ssize_t ret = 0;
-	int i, err = 0;
-	int data_present;
+	int err = 0;
 	int bytes_read;
 
 	while (size) {
@@ -107,21 +106,10 @@ static ssize_t rng_dev_read(struct file *filp, char __user *buf,
 			err = -ENODEV;
 			goto out;
 		}
-		if (filp->f_flags & O_NONBLOCK) {
-			data_present = hwrng_data_present(current_rng);
-		} else {
-			/* Some RNG require some time between data_reads to gather
-			 * new entropy. Poll it.
-			 */
-			for (i = 0; i < 20; i++) {
-				data_present = hwrng_data_present(current_rng);
-				if (data_present)
-					break;
-				udelay(10);
-			}
-		}
+
 		bytes_read = 0;
-		if (data_present)
+		if (hwrng_data_present(current_rng,
+				       !(filp->f_flags & O_NONBLOCK)))
 			bytes_read = hwrng_data_read(current_rng, &data);
 		mutex_unlock(&rng_mutex);
 
diff --git a/drivers/char/hw_random/geode-rng.c b/drivers/char/hw_random/geode-rng.c
index 8e8658dcd2e3..fed4ef5569f5 100644
--- a/drivers/char/hw_random/geode-rng.c
+++ b/drivers/char/hw_random/geode-rng.c
@@ -28,6 +28,7 @@
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/hw_random.h>
+#include <linux/delay.h>
 #include <asm/io.h>
 
 
@@ -61,11 +62,18 @@ static int geode_rng_data_read(struct hwrng *rng, u32 *data)
 	return 4;
 }
 
-static int geode_rng_data_present(struct hwrng *rng)
+static int geode_rng_data_present(struct hwrng *rng, int wait)
 {
 	void __iomem *mem = (void __iomem *)rng->priv;
+	int data, i;
 
-	return !!(readl(mem + GEODE_RNG_STATUS_REG));
+	for (i = 0; i < 20; i++) {
+		data = !!(readl(mem + GEODE_RNG_STATUS_REG));
+		if (data || !wait)
+			break;
+		udelay(10);
+	}
+	return data;
 }
 
 
diff --git a/drivers/char/hw_random/intel-rng.c b/drivers/char/hw_random/intel-rng.c
index 753f46052b87..5cc651ef75eb 100644
--- a/drivers/char/hw_random/intel-rng.c
+++ b/drivers/char/hw_random/intel-rng.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/stop_machine.h>
+#include <linux/delay.h>
 #include <asm/io.h>
 
 
@@ -162,11 +163,19 @@ static inline u8 hwstatus_set(void __iomem *mem,
 	return hwstatus_get(mem);
 }
 
-static int intel_rng_data_present(struct hwrng *rng)
+static int intel_rng_data_present(struct hwrng *rng, int wait)
 {
 	void __iomem *mem = (void __iomem *)rng->priv;
-
-	return !!(readb(mem + INTEL_RNG_STATUS) & INTEL_RNG_DATA_PRESENT);
+	int data, i;
+
+	for (i = 0; i < 20; i++) {
+		data = !!(readb(mem + INTEL_RNG_STATUS) &
+			  INTEL_RNG_DATA_PRESENT);
+		if (data || !wait)
+			break;
+		udelay(10);
+	}
+	return data;
 }
 
 static int intel_rng_data_read(struct hwrng *rng, u32 *data)
diff --git a/drivers/char/hw_random/omap-rng.c b/drivers/char/hw_random/omap-rng.c
index 3f35a1c562b1..7e319951fa41 100644
--- a/drivers/char/hw_random/omap-rng.c
+++ b/drivers/char/hw_random/omap-rng.c
@@ -29,6 +29,7 @@
 #include <linux/err.h>
 #include <linux/platform_device.h>
 #include <linux/hw_random.h>
+#include <linux/delay.h>
 
 #include <asm/io.h>
 
@@ -65,9 +66,17 @@ static void omap_rng_write_reg(int reg, u32 val)
 }
 
 /* REVISIT: Does the status bit really work on 16xx? */
-static int omap_rng_data_present(struct hwrng *rng)
+static int omap_rng_data_present(struct hwrng *rng, int wait)
 {
-	return omap_rng_read_reg(RNG_STAT_REG) ? 0 : 1;
+	int data, i;
+
+	for (i = 0; i < 20; i++) {
+		data = omap_rng_read_reg(RNG_STAT_REG) ? 0 : 1;
+		if (data || !wait)
+			break;
+		udelay(10);
+	}
+	return data;
 }
 
 static int omap_rng_data_read(struct hwrng *rng, u32 *data)
diff --git a/drivers/char/hw_random/pasemi-rng.c b/drivers/char/hw_random/pasemi-rng.c
index fa6040b6c8f2..621adf25e580 100644
--- a/drivers/char/hw_random/pasemi-rng.c
+++ b/drivers/char/hw_random/pasemi-rng.c
@@ -23,6 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/hw_random.h>
+#include <linux/delay.h>
 #include <asm/of_platform.h>
 #include <asm/io.h>
 
@@ -44,9 +45,16 @@
 static int pasemi_rng_data_present(struct hwrng *rng)
 {
 	void __iomem *rng_regs = (void __iomem *)rng->priv;
-
-	return (in_le32(rng_regs + SDCRNG_CTL_REG)
-		& SDCRNG_CTL_FVLD_M) ? 1 : 0;
+	int data, i;
+
+	for (i = 0; i < 20; i++) {
+		data = (in_le32(rng_regs + SDCRNG_CTL_REG)
+			& SDCRNG_CTL_FVLD_M) ? 1 : 0;
+		if (data || !wait)
+			break;
+		udelay(10);
+	}
+	return data;
 }
 
 static int pasemi_rng_data_read(struct hwrng *rng, u32 *data)
diff --git a/drivers/char/hw_random/via-rng.c b/drivers/char/hw_random/via-rng.c
index ec435cb25c4f..868e39fd42e4 100644
--- a/drivers/char/hw_random/via-rng.c
+++ b/drivers/char/hw_random/via-rng.c
@@ -27,6 +27,7 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/hw_random.h>
+#include <linux/delay.h>
 #include <asm/io.h>
 #include <asm/msr.h>
 #include <asm/cpufeature.h>
@@ -77,10 +78,11 @@ static inline u32 xstore(u32 *addr, u32 edx_in)
 	return eax_out;
 }
 
-static int via_rng_data_present(struct hwrng *rng)
+static int via_rng_data_present(struct hwrng *rng, int wait)
 {
 	u32 bytes_out;
 	u32 *via_rng_datum = (u32 *)(&rng->priv);
+	int i;
 
 	/* We choose the recommended 1-byte-per-instruction RNG rate,
 	 * for greater randomness at the expense of speed.  Larger
@@ -95,12 +97,15 @@ static int via_rng_data_present(struct hwrng *rng)
 	 * completes.
 	 */
 
-	*via_rng_datum = 0; /* paranoia, not really necessary */
-	bytes_out = xstore(via_rng_datum, VIA_RNG_CHUNK_1);
-	bytes_out &= VIA_XSTORE_CNT_MASK;
-	if (bytes_out == 0)
-		return 0;
-	return 1;
+	for (i = 0; i < 20; i++) {
+		*via_rng_datum = 0; /* paranoia, not really necessary */
+		bytes_out = xstore(via_rng_datum, VIA_RNG_CHUNK_1);
+		bytes_out &= VIA_XSTORE_CNT_MASK;
+		if (bytes_out || !wait)
+			break;
+		udelay(10);
+	}
+	return bytes_out ? 1 : 0;
 }
 
 static int via_rng_data_read(struct hwrng *rng, u32 *data)
diff --git a/include/linux/hw_random.h b/include/linux/hw_random.h
index 21ea7610e177..85d11916e9ea 100644
--- a/include/linux/hw_random.h
+++ b/include/linux/hw_random.h
@@ -33,7 +33,7 @@ struct hwrng {
 	const char *name;
 	int (*init)(struct hwrng *rng);
 	void (*cleanup)(struct hwrng *rng);
-	int (*data_present)(struct hwrng *rng);
+	int (*data_present)(struct hwrng *rng, int wait);
 	int (*data_read)(struct hwrng *rng, u32 *data);
 	unsigned long priv;
 
-- 
cgit v1.2.3


From 7613636def82092a5c7b6322078a2af832410417 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 20 Nov 2007 17:26:06 +0800
Subject: [CRYPTO] api: Add crypto_inc and crypto_xor

With the addition of more stream ciphers we need to curb the proliferation
of ad-hoc xor functions.  This patch creates a generic pair of functions,
crypto_inc and crypto_xor which does big-endian increment and exclusive or,
respectively.

For optimum performance, they both use u32 operations so alignment must be
as that of u32 even though the arguments are of type u8 *.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/algapi.c         | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 include/crypto/algapi.h |  4 ++++
 2 files changed, 51 insertions(+)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index 8383282de1dd..08eca6da1d53 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -605,6 +605,53 @@ int crypto_tfm_in_queue(struct crypto_queue *queue, struct crypto_tfm *tfm)
 }
 EXPORT_SYMBOL_GPL(crypto_tfm_in_queue);
 
+static inline void crypto_inc_byte(u8 *a, unsigned int size)
+{
+	u8 *b = (a + size);
+	u8 c;
+
+	for (; size; size--) {
+		c = *--b + 1;
+		*b = c;
+		if (c)
+			break;
+	}
+}
+
+void crypto_inc(u8 *a, unsigned int size)
+{
+	__be32 *b = (__be32 *)(a + size);
+	u32 c;
+
+	for (; size >= 4; size -= 4) {
+		c = be32_to_cpu(*--b) + 1;
+		*b = cpu_to_be32(c);
+		if (c)
+			return;
+	}
+
+	crypto_inc_byte(a, size);
+}
+EXPORT_SYMBOL_GPL(crypto_inc);
+
+static inline void crypto_xor_byte(u8 *a, const u8 *b, unsigned int size)
+{
+	for (; size; size--)
+		*a++ ^= *b++;
+}
+
+void crypto_xor(u8 *dst, const u8 *src, unsigned int size)
+{
+	u32 *a = (u32 *)dst;
+	u32 *b = (u32 *)src;
+
+	for (; size >= 4; size -= 4)
+		*a++ ^= *b++;
+
+	crypto_xor_byte((u8 *)a, (u8 *)b, size);
+}
+EXPORT_SYMBOL_GPL(crypto_xor);
+
 static int __init crypto_algapi_init(void)
 {
 	crypto_init_proc();
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 88619f902c10..2cdb227fc39d 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -124,6 +124,10 @@ int crypto_enqueue_request(struct crypto_queue *queue,
 struct crypto_async_request *crypto_dequeue_request(struct crypto_queue *queue);
 int crypto_tfm_in_queue(struct crypto_queue *queue, struct crypto_tfm *tfm);
 
+/* These functions require the input/output to be aligned as u32. */
+void crypto_inc(u8 *a, unsigned int size);
+void crypto_xor(u8 *dst, const u8 *src, unsigned int size);
+
 int blkcipher_walk_done(struct blkcipher_desc *desc,
 			struct blkcipher_walk *walk, int err);
 int blkcipher_walk_virt(struct blkcipher_desc *desc,
-- 
cgit v1.2.3


From 7ba683a6deba70251756aa5a021cdaa5c875a7a2 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 2 Dec 2007 18:49:21 +1100
Subject: [CRYPTO] aead: Make authsize a run-time parameter

As it is authsize is an algorithm paramter which cannot be changed at
run-time.  This is inconvenient because hardware that implements such
algorithms would have to register each authsize that they support
separately.

Since authsize is a property common to all AEAD algorithms, we can add
a function setauthsize that sets it at run-time, just like setkey.

This patch does exactly that and also changes authenc so that authsize
is no longer a parameter of its template.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/aead.c          | 24 +++++++++++++++++++++---
 crypto/authenc.c       | 39 ++++++++++++---------------------------
 crypto/gcm.c           |  2 +-
 include/linux/crypto.h |  5 ++++-
 4 files changed, 38 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/crypto/aead.c b/crypto/aead.c
index 84a3501fb478..f23c2b0ee009 100644
--- a/crypto/aead.c
+++ b/crypto/aead.c
@@ -53,6 +53,24 @@ static int setkey(struct crypto_aead *tfm, const u8 *key, unsigned int keylen)
 	return aead->setkey(tfm, key, keylen);
 }
 
+int crypto_aead_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
+{
+	int err;
+
+	if (authsize > crypto_aead_alg(tfm)->maxauthsize)
+		return -EINVAL;
+
+	if (crypto_aead_alg(tfm)->setauthsize) {
+		err = crypto_aead_alg(tfm)->setauthsize(tfm, authsize);
+		if (err)
+			return err;
+	}
+
+	crypto_aead_crt(tfm)->authsize = authsize;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(crypto_aead_setauthsize);
+
 static unsigned int crypto_aead_ctxsize(struct crypto_alg *alg, u32 type,
 					u32 mask)
 {
@@ -64,14 +82,14 @@ static int crypto_init_aead_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 	struct aead_alg *alg = &tfm->__crt_alg->cra_aead;
 	struct aead_tfm *crt = &tfm->crt_aead;
 
-	if (max(alg->authsize, alg->ivsize) > PAGE_SIZE / 8)
+	if (max(alg->maxauthsize, alg->ivsize) > PAGE_SIZE / 8)
 		return -EINVAL;
 
 	crt->setkey = setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
 	crt->ivsize = alg->ivsize;
-	crt->authsize = alg->authsize;
+	crt->authsize = alg->maxauthsize;
 
 	return 0;
 }
@@ -85,7 +103,7 @@ static void crypto_aead_show(struct seq_file *m, struct crypto_alg *alg)
 	seq_printf(m, "type         : aead\n");
 	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
 	seq_printf(m, "ivsize       : %u\n", aead->ivsize);
-	seq_printf(m, "authsize     : %u\n", aead->authsize);
+	seq_printf(m, "maxauthsize  : %u\n", aead->maxauthsize);
 }
 
 const struct crypto_type crypto_aead_type = {
diff --git a/crypto/authenc.c b/crypto/authenc.c
index 66fb2aa5c325..5df5fb169cbe 100644
--- a/crypto/authenc.c
+++ b/crypto/authenc.c
@@ -24,7 +24,6 @@ struct authenc_instance_ctx {
 	struct crypto_spawn auth;
 	struct crypto_spawn enc;
 
-	unsigned int authsize;
 	unsigned int enckeylen;
 };
 
@@ -76,8 +75,6 @@ out:
 static int crypto_authenc_hash(struct aead_request *req)
 {
 	struct crypto_aead *authenc = crypto_aead_reqtfm(req);
-	struct authenc_instance_ctx *ictx =
-		crypto_instance_ctx(crypto_aead_alg_instance(authenc));
 	struct crypto_authenc_ctx *ctx = crypto_aead_ctx(authenc);
 	struct crypto_hash *auth = ctx->auth;
 	struct hash_desc desc = {
@@ -111,7 +108,8 @@ auth_unlock:
 	if (err)
 		return err;
 
-	scatterwalk_map_and_copy(hash, dst, cryptlen, ictx->authsize, 1);
+	scatterwalk_map_and_copy(hash, dst, cryptlen,
+				 crypto_aead_authsize(authenc), 1);
 	return 0;
 }
 
@@ -147,8 +145,6 @@ static int crypto_authenc_encrypt(struct aead_request *req)
 static int crypto_authenc_verify(struct aead_request *req)
 {
 	struct crypto_aead *authenc = crypto_aead_reqtfm(req);
-	struct authenc_instance_ctx *ictx =
-		crypto_instance_ctx(crypto_aead_alg_instance(authenc));
 	struct crypto_authenc_ctx *ctx = crypto_aead_ctx(authenc);
 	struct crypto_hash *auth = ctx->auth;
 	struct hash_desc desc = {
@@ -186,7 +182,7 @@ auth_unlock:
 	if (err)
 		return err;
 
-	authsize = ictx->authsize;
+	authsize = crypto_aead_authsize(authenc);
 	scatterwalk_map_and_copy(ihash, src, cryptlen, authsize, 0);
 	return memcmp(ihash, ohash, authsize) ? -EINVAL : 0;
 }
@@ -224,18 +220,12 @@ static int crypto_authenc_init_tfm(struct crypto_tfm *tfm)
 	struct crypto_authenc_ctx *ctx = crypto_tfm_ctx(tfm);
 	struct crypto_hash *auth;
 	struct crypto_ablkcipher *enc;
-	unsigned int digestsize;
 	int err;
 
 	auth = crypto_spawn_hash(&ictx->auth);
 	if (IS_ERR(auth))
 		return PTR_ERR(auth);
 
-	err = -EINVAL;
-	digestsize = crypto_hash_digestsize(auth);
-	if (ictx->authsize > digestsize)
-		goto err_free_hash;
-
 	enc = crypto_spawn_ablkcipher(&ictx->enc);
 	err = PTR_ERR(enc);
 	if (IS_ERR(enc))
@@ -246,7 +236,7 @@ static int crypto_authenc_init_tfm(struct crypto_tfm *tfm)
 	tfm->crt_aead.reqsize = max_t(unsigned int,
 				      (crypto_hash_alignmask(auth) &
 				       ~(crypto_tfm_ctx_alignment() - 1)) +
-				      digestsize * 2,
+				      crypto_hash_digestsize(auth) * 2,
 				      sizeof(struct ablkcipher_request) +
 				      crypto_ablkcipher_reqsize(enc));
 
@@ -273,7 +263,6 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 	struct crypto_alg *auth;
 	struct crypto_alg *enc;
 	struct authenc_instance_ctx *ctx;
-	unsigned int authsize;
 	unsigned int enckeylen;
 	int err;
 
@@ -286,18 +275,13 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 	if (IS_ERR(auth))
 		return ERR_PTR(PTR_ERR(auth));
 
-	err = crypto_attr_u32(tb[2], &authsize);
-	inst = ERR_PTR(err);
-	if (err)
-		goto out_put_auth;
-
-	enc = crypto_attr_alg(tb[3], CRYPTO_ALG_TYPE_BLKCIPHER,
+	enc = crypto_attr_alg(tb[2], CRYPTO_ALG_TYPE_BLKCIPHER,
 			      CRYPTO_ALG_TYPE_BLKCIPHER_MASK);
 	inst = ERR_PTR(PTR_ERR(enc));
 	if (IS_ERR(enc))
 		goto out_put_auth;
 
-	err = crypto_attr_u32(tb[4], &enckeylen);
+	err = crypto_attr_u32(tb[3], &enckeylen);
 	if (err)
 		goto out_put_enc;
 
@@ -308,18 +292,17 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 
 	err = -ENAMETOOLONG;
 	if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
-		     "authenc(%s,%u,%s,%u)", auth->cra_name, authsize,
+		     "authenc(%s,%s,%u)", auth->cra_name,
 		     enc->cra_name, enckeylen) >= CRYPTO_MAX_ALG_NAME)
 		goto err_free_inst;
 
 	if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
-		     "authenc(%s,%u,%s,%u)", auth->cra_driver_name,
-		     authsize, enc->cra_driver_name, enckeylen) >=
+		     "authenc(%s,%s,%u)", auth->cra_driver_name,
+		     enc->cra_driver_name, enckeylen) >=
 	    CRYPTO_MAX_ALG_NAME)
 		goto err_free_inst;
 
 	ctx = crypto_instance_ctx(inst);
-	ctx->authsize = authsize;
 	ctx->enckeylen = enckeylen;
 
 	err = crypto_init_spawn(&ctx->auth, auth, inst, CRYPTO_ALG_TYPE_MASK);
@@ -337,7 +320,9 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 	inst->alg.cra_type = &crypto_aead_type;
 
 	inst->alg.cra_aead.ivsize = enc->cra_blkcipher.ivsize;
-	inst->alg.cra_aead.authsize = authsize;
+	inst->alg.cra_aead.maxauthsize = auth->cra_type == &crypto_hash_type ?
+					 auth->cra_hash.digestsize :
+					 auth->cra_digest.dia_digestsize;
 
 	inst->alg.cra_ctxsize = sizeof(struct crypto_authenc_ctx);
 
diff --git a/crypto/gcm.c b/crypto/gcm.c
index ad8b8b9aeef2..5681c7957b88 100644
--- a/crypto/gcm.c
+++ b/crypto/gcm.c
@@ -414,7 +414,7 @@ static struct crypto_instance *crypto_gcm_alloc(struct rtattr **tb)
 	inst->alg.cra_alignmask = __alignof__(u32) - 1;
 	inst->alg.cra_type = &crypto_aead_type;
 	inst->alg.cra_aead.ivsize = 12;
-	inst->alg.cra_aead.authsize = 16;
+	inst->alg.cra_aead.maxauthsize = 16;
 	inst->alg.cra_ctxsize = sizeof(struct crypto_gcm_ctx);
 	inst->alg.cra_init = crypto_gcm_init_tfm;
 	inst->alg.cra_exit = crypto_gcm_exit_tfm;
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index f56ae8721bc9..48aa5959abbb 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -187,11 +187,12 @@ struct ablkcipher_alg {
 struct aead_alg {
 	int (*setkey)(struct crypto_aead *tfm, const u8 *key,
 	              unsigned int keylen);
+	int (*setauthsize)(struct crypto_aead *tfm, unsigned int authsize);
 	int (*encrypt)(struct aead_request *req);
 	int (*decrypt)(struct aead_request *req);
 
 	unsigned int ivsize;
-	unsigned int authsize;
+	unsigned int maxauthsize;
 };
 
 struct blkcipher_alg {
@@ -754,6 +755,8 @@ static inline int crypto_aead_setkey(struct crypto_aead *tfm, const u8 *key,
 	return crypto_aead_crt(tfm)->setkey(tfm, key, keylen);
 }
 
+int crypto_aead_setauthsize(struct crypto_aead *tfm, unsigned int authsize);
+
 static inline struct crypto_aead *crypto_aead_reqtfm(struct aead_request *req)
 {
 	return __crypto_aead_cast(req->base.tfm);
-- 
cgit v1.2.3


From e236d4a89a2ffbc8aa18064161f4f159c4d89b4a Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 22 Nov 2007 23:11:53 +0800
Subject: [CRYPTO] authenc: Move enckeylen into key itself

Having enckeylen as a template parameter makes it a pain for hardware
devices that implement ciphers with many key sizes since each one would
have to be registered separately.

Since the authenc algorithm is mainly used for legacy purposes where its
key is going to be constructed out of two separate keys, we can in fact
embed this value into the key itself.

This patch does this by prepending an rtnetlink header to the key that
contains the encryption key length.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/authenc.c         | 49 ++++++++++++++++++++++++++++--------------------
 include/crypto/authenc.h | 27 ++++++++++++++++++++++++++
 2 files changed, 56 insertions(+), 20 deletions(-)
 create mode 100644 include/crypto/authenc.h

(limited to 'include')

diff --git a/crypto/authenc.c b/crypto/authenc.c
index 5df5fb169cbe..a61dea1c2fe6 100644
--- a/crypto/authenc.c
+++ b/crypto/authenc.c
@@ -11,10 +11,12 @@
  */
 
 #include <crypto/algapi.h>
+#include <crypto/authenc.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/rtnetlink.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
@@ -23,8 +25,6 @@
 struct authenc_instance_ctx {
 	struct crypto_spawn auth;
 	struct crypto_spawn enc;
-
-	unsigned int enckeylen;
 };
 
 struct crypto_authenc_ctx {
@@ -36,19 +36,31 @@ struct crypto_authenc_ctx {
 static int crypto_authenc_setkey(struct crypto_aead *authenc, const u8 *key,
 				 unsigned int keylen)
 {
-	struct authenc_instance_ctx *ictx =
-		crypto_instance_ctx(crypto_aead_alg_instance(authenc));
-	unsigned int enckeylen = ictx->enckeylen;
 	unsigned int authkeylen;
+	unsigned int enckeylen;
 	struct crypto_authenc_ctx *ctx = crypto_aead_ctx(authenc);
 	struct crypto_hash *auth = ctx->auth;
 	struct crypto_ablkcipher *enc = ctx->enc;
+	struct rtattr *rta = (void *)key;
+	struct crypto_authenc_key_param *param;
 	int err = -EINVAL;
 
-	if (keylen < enckeylen) {
-		crypto_aead_set_flags(authenc, CRYPTO_TFM_RES_BAD_KEY_LEN);
-		goto out;
-	}
+	if (keylen < sizeof(*rta))
+		goto badkey;
+	if (rta->rta_type != CRYPTO_AUTHENC_KEYA_PARAM)
+		goto badkey;
+	if (RTA_PAYLOAD(rta) < sizeof(*param))
+		goto badkey;
+
+	param = RTA_DATA(rta);
+	enckeylen = be32_to_cpu(param->enckeylen);
+
+	key += RTA_ALIGN(rta->rta_len);
+	keylen -= RTA_ALIGN(rta->rta_len);
+
+	if (keylen < enckeylen)
+		goto badkey;
+
 	authkeylen = keylen - enckeylen;
 
 	crypto_hash_clear_flags(auth, CRYPTO_TFM_REQ_MASK);
@@ -70,6 +82,10 @@ static int crypto_authenc_setkey(struct crypto_aead *authenc, const u8 *key,
 
 out:
 	return err;
+
+badkey:
+	crypto_aead_set_flags(authenc, CRYPTO_TFM_RES_BAD_KEY_LEN);
+	goto out;
 }
 
 static int crypto_authenc_hash(struct aead_request *req)
@@ -263,7 +279,6 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 	struct crypto_alg *auth;
 	struct crypto_alg *enc;
 	struct authenc_instance_ctx *ctx;
-	unsigned int enckeylen;
 	int err;
 
 	err = crypto_check_attr_type(tb, CRYPTO_ALG_TYPE_AEAD);
@@ -281,10 +296,6 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 	if (IS_ERR(enc))
 		goto out_put_auth;
 
-	err = crypto_attr_u32(tb[3], &enckeylen);
-	if (err)
-		goto out_put_enc;
-
 	inst = kzalloc(sizeof(*inst) + sizeof(*ctx), GFP_KERNEL);
 	err = -ENOMEM;
 	if (!inst)
@@ -292,18 +303,16 @@ static struct crypto_instance *crypto_authenc_alloc(struct rtattr **tb)
 
 	err = -ENAMETOOLONG;
 	if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
-		     "authenc(%s,%s,%u)", auth->cra_name,
-		     enc->cra_name, enckeylen) >= CRYPTO_MAX_ALG_NAME)
+		     "authenc(%s,%s)", auth->cra_name, enc->cra_name) >=
+	    CRYPTO_MAX_ALG_NAME)
 		goto err_free_inst;
 
 	if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
-		     "authenc(%s,%s,%u)", auth->cra_driver_name,
-		     enc->cra_driver_name, enckeylen) >=
-	    CRYPTO_MAX_ALG_NAME)
+		     "authenc(%s,%s)", auth->cra_driver_name,
+		     enc->cra_driver_name) >= CRYPTO_MAX_ALG_NAME)
 		goto err_free_inst;
 
 	ctx = crypto_instance_ctx(inst);
-	ctx->enckeylen = enckeylen;
 
 	err = crypto_init_spawn(&ctx->auth, auth, inst, CRYPTO_ALG_TYPE_MASK);
 	if (err)
diff --git a/include/crypto/authenc.h b/include/crypto/authenc.h
new file mode 100644
index 000000000000..e47b044929a8
--- /dev/null
+++ b/include/crypto/authenc.h
@@ -0,0 +1,27 @@
+/*
+ * Authenc: Simple AEAD wrapper for IPsec
+ *
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+#ifndef _CRYPTO_AUTHENC_H
+#define _CRYPTO_AUTHENC_H
+
+#include <linux/types.h>
+
+enum {
+	CRYPTO_AUTHENC_KEYA_UNSPEC,
+	CRYPTO_AUTHENC_KEYA_PARAM,
+};
+
+struct crypto_authenc_key_param {
+	__be32 enckeylen;
+};
+
+#endif	/* _CRYPTO_AUTHENC_H */
+
-- 
cgit v1.2.3


From 42c271c6c538857cb13c5ead5184d264d745f675 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 7 Dec 2007 18:52:49 +0800
Subject: [CRYPTO] scatterwalk: Move scatterwalk.h to linux/crypto

The scatterwalk infrastructure is used by algorithms so it needs to
move out of crypto for future users that may live in drivers/crypto
or asm/*/crypto.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/authenc.c             |   3 +-
 crypto/blkcipher.c           |   2 +-
 crypto/digest.c              |   3 +-
 crypto/gcm.c                 |   3 +-
 crypto/internal.h            |  29 ------------
 crypto/scatterwalk.c         |   5 +-
 crypto/scatterwalk.h         |  75 ------------------------------
 crypto/xcbc.c                |   2 +-
 include/crypto/scatterwalk.h | 107 +++++++++++++++++++++++++++++++++++++++++++
 9 files changed, 115 insertions(+), 114 deletions(-)
 delete mode 100644 crypto/scatterwalk.h
 create mode 100644 include/crypto/scatterwalk.h

(limited to 'include')

diff --git a/crypto/authenc.c b/crypto/authenc.c
index 6c9104ebf2f4..fbbc2b505a21 100644
--- a/crypto/authenc.c
+++ b/crypto/authenc.c
@@ -12,6 +12,7 @@
 
 #include <crypto/algapi.h>
 #include <crypto/authenc.h>
+#include <crypto/scatterwalk.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -20,8 +21,6 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
-#include "scatterwalk.h"
-
 struct authenc_instance_ctx {
 	struct crypto_spawn auth;
 	struct crypto_spawn enc;
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index 180d91451476..7939504dfd8b 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -14,6 +14,7 @@
  *
  */
 
+#include <crypto/scatterwalk.h>
 #include <linux/crypto.h>
 #include <linux/errno.h>
 #include <linux/hardirq.h>
@@ -25,7 +26,6 @@
 #include <linux/string.h>
 
 #include "internal.h"
-#include "scatterwalk.h"
 
 enum {
 	BLKCIPHER_WALK_PHYS = 1 << 0,
diff --git a/crypto/digest.c b/crypto/digest.c
index d3e827a141f1..52845f53f8e1 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -12,6 +12,7 @@
  *
  */
 
+#include <crypto/scatterwalk.h>
 #include <linux/mm.h>
 #include <linux/errno.h>
 #include <linux/hardirq.h>
@@ -20,8 +21,6 @@
 #include <linux/module.h>
 #include <linux/scatterlist.h>
 
-#include "internal.h"
-
 static int init(struct hash_desc *desc)
 {
 	struct crypto_tfm *tfm = crypto_hash_tfm(desc->tfm);
diff --git a/crypto/gcm.c b/crypto/gcm.c
index d60c340b0b9d..27483f361e80 100644
--- a/crypto/gcm.c
+++ b/crypto/gcm.c
@@ -10,13 +10,14 @@
 
 #include <crypto/algapi.h>
 #include <crypto/gf128mul.h>
+#include <crypto/scatterwalk.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 
-#include "scatterwalk.h"
+#include "internal.h"
 
 struct gcm_instance_ctx {
 	struct crypto_spawn ctr;
diff --git a/crypto/internal.h b/crypto/internal.h
index abb01f71f817..cb13952f82bf 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -25,7 +25,6 @@
 #include <linux/notifier.h>
 #include <linux/rwsem.h>
 #include <linux/slab.h>
-#include <asm/kmap_types.h>
 
 /* Crypto notification events. */
 enum {
@@ -50,34 +49,6 @@ extern struct list_head crypto_alg_list;
 extern struct rw_semaphore crypto_alg_sem;
 extern struct blocking_notifier_head crypto_chain;
 
-static inline enum km_type crypto_kmap_type(int out)
-{
-	enum km_type type;
-
-	if (in_softirq())
-		type = out * (KM_SOFTIRQ1 - KM_SOFTIRQ0) + KM_SOFTIRQ0;
-	else
-		type = out * (KM_USER1 - KM_USER0) + KM_USER0;
-
-	return type;
-}
-
-static inline void *crypto_kmap(struct page *page, int out)
-{
-	return kmap_atomic(page, crypto_kmap_type(out));
-}
-
-static inline void crypto_kunmap(void *vaddr, int out)
-{
-	kunmap_atomic(vaddr, crypto_kmap_type(out));
-}
-
-static inline void crypto_yield(u32 flags)
-{
-	if (flags & CRYPTO_TFM_REQ_MAY_SLEEP)
-		cond_resched();
-}
-
 #ifdef CONFIG_PROC_FS
 void __init crypto_init_proc(void);
 void __exit crypto_exit_proc(void);
diff --git a/crypto/scatterwalk.c b/crypto/scatterwalk.c
index 206c39a97e57..12d19019e178 100644
--- a/crypto/scatterwalk.c
+++ b/crypto/scatterwalk.c
@@ -13,6 +13,8 @@
  * any later version.
  *
  */
+
+#include <crypto/scatterwalk.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
 #include <linux/module.h>
@@ -20,9 +22,6 @@
 #include <linux/highmem.h>
 #include <linux/scatterlist.h>
 
-#include "internal.h"
-#include "scatterwalk.h"
-
 static inline void memcpy_dir(void *buf, void *sgdata, size_t nbytes, int out)
 {
 	void *src = out ? buf : sgdata;
diff --git a/crypto/scatterwalk.h b/crypto/scatterwalk.h
deleted file mode 100644
index fd5517d2a7a9..000000000000
--- a/crypto/scatterwalk.h
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
- * Cryptographic API.
- *
- * Copyright (c) 2002 James Morris <jmorris@intercode.com.au>
- * Copyright (c) 2002 Adam J. Richter <adam@yggdrasil.com>
- * Copyright (c) 2004 Jean-Luc Cooke <jlcooke@certainkey.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- *
- */
-
-#ifndef _CRYPTO_SCATTERWALK_H
-#define _CRYPTO_SCATTERWALK_H
-
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-
-#include "internal.h"
-
-static inline unsigned long scatterwalk_samebuf(struct scatter_walk *walk_in,
-						struct scatter_walk *walk_out)
-{
-	return !(((sg_page(walk_in->sg) - sg_page(walk_out->sg)) << PAGE_SHIFT) +
-		 (int)(walk_in->offset - walk_out->offset));
-}
-
-static inline unsigned int scatterwalk_pagelen(struct scatter_walk *walk)
-{
-	unsigned int len = walk->sg->offset + walk->sg->length - walk->offset;
-	unsigned int len_this_page = offset_in_page(~walk->offset) + 1;
-	return len_this_page > len ? len : len_this_page;
-}
-
-static inline unsigned int scatterwalk_clamp(struct scatter_walk *walk,
-					     unsigned int nbytes)
-{
-	unsigned int len_this_page = scatterwalk_pagelen(walk);
-	return nbytes > len_this_page ? len_this_page : nbytes;
-}
-
-static inline void scatterwalk_advance(struct scatter_walk *walk,
-				       unsigned int nbytes)
-{
-	walk->offset += nbytes;
-}
-
-static inline unsigned int scatterwalk_aligned(struct scatter_walk *walk,
-					       unsigned int alignmask)
-{
-	return !(walk->offset & alignmask);
-}
-
-static inline struct page *scatterwalk_page(struct scatter_walk *walk)
-{
-	return sg_page(walk->sg) + (walk->offset >> PAGE_SHIFT);
-}
-
-static inline void scatterwalk_unmap(void *vaddr, int out)
-{
-	crypto_kunmap(vaddr, out);
-}
-
-void scatterwalk_start(struct scatter_walk *walk, struct scatterlist *sg);
-void scatterwalk_copychunks(void *buf, struct scatter_walk *walk,
-			    size_t nbytes, int out);
-void *scatterwalk_map(struct scatter_walk *walk, int out);
-void scatterwalk_done(struct scatter_walk *walk, int out, int more);
-
-void scatterwalk_map_and_copy(void *buf, struct scatterlist *sg,
-			      unsigned int start, unsigned int nbytes, int out);
-
-#endif  /* _CRYPTO_SCATTERWALK_H */
diff --git a/crypto/xcbc.c b/crypto/xcbc.c
index ac68f3b62fde..789cdeee6b7d 100644
--- a/crypto/xcbc.c
+++ b/crypto/xcbc.c
@@ -19,6 +19,7 @@
  * 	Kazunori Miyazawa <miyazawa@linux-ipv6.org>
  */
 
+#include <crypto/scatterwalk.h>
 #include <linux/crypto.h>
 #include <linux/err.h>
 #include <linux/hardirq.h>
@@ -27,7 +28,6 @@
 #include <linux/rtnetlink.h>
 #include <linux/slab.h>
 #include <linux/scatterlist.h>
-#include "internal.h"
 
 static u_int32_t ks[12] = {0x01010101, 0x01010101, 0x01010101, 0x01010101,
 			   0x02020202, 0x02020202, 0x02020202, 0x02020202,
diff --git a/include/crypto/scatterwalk.h b/include/crypto/scatterwalk.h
new file mode 100644
index 000000000000..07b6f17122d2
--- /dev/null
+++ b/include/crypto/scatterwalk.h
@@ -0,0 +1,107 @@
+/*
+ * Cryptographic scatter and gather helpers.
+ *
+ * Copyright (c) 2002 James Morris <jmorris@intercode.com.au>
+ * Copyright (c) 2002 Adam J. Richter <adam@yggdrasil.com>
+ * Copyright (c) 2004 Jean-Luc Cooke <jlcooke@certainkey.com>
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_SCATTERWALK_H
+#define _CRYPTO_SCATTERWALK_H
+
+#include <asm/kmap_types.h>
+#include <crypto/algapi.h>
+#include <linux/hardirq.h>
+#include <linux/highmem.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/scatterlist.h>
+
+static inline enum km_type crypto_kmap_type(int out)
+{
+	enum km_type type;
+
+	if (in_softirq())
+		type = out * (KM_SOFTIRQ1 - KM_SOFTIRQ0) + KM_SOFTIRQ0;
+	else
+		type = out * (KM_USER1 - KM_USER0) + KM_USER0;
+
+	return type;
+}
+
+static inline void *crypto_kmap(struct page *page, int out)
+{
+	return kmap_atomic(page, crypto_kmap_type(out));
+}
+
+static inline void crypto_kunmap(void *vaddr, int out)
+{
+	kunmap_atomic(vaddr, crypto_kmap_type(out));
+}
+
+static inline void crypto_yield(u32 flags)
+{
+	if (flags & CRYPTO_TFM_REQ_MAY_SLEEP)
+		cond_resched();
+}
+
+static inline unsigned long scatterwalk_samebuf(struct scatter_walk *walk_in,
+						struct scatter_walk *walk_out)
+{
+	return !(((sg_page(walk_in->sg) - sg_page(walk_out->sg)) << PAGE_SHIFT) +
+		 (int)(walk_in->offset - walk_out->offset));
+}
+
+static inline unsigned int scatterwalk_pagelen(struct scatter_walk *walk)
+{
+	unsigned int len = walk->sg->offset + walk->sg->length - walk->offset;
+	unsigned int len_this_page = offset_in_page(~walk->offset) + 1;
+	return len_this_page > len ? len : len_this_page;
+}
+
+static inline unsigned int scatterwalk_clamp(struct scatter_walk *walk,
+					     unsigned int nbytes)
+{
+	unsigned int len_this_page = scatterwalk_pagelen(walk);
+	return nbytes > len_this_page ? len_this_page : nbytes;
+}
+
+static inline void scatterwalk_advance(struct scatter_walk *walk,
+				       unsigned int nbytes)
+{
+	walk->offset += nbytes;
+}
+
+static inline unsigned int scatterwalk_aligned(struct scatter_walk *walk,
+					       unsigned int alignmask)
+{
+	return !(walk->offset & alignmask);
+}
+
+static inline struct page *scatterwalk_page(struct scatter_walk *walk)
+{
+	return sg_page(walk->sg) + (walk->offset >> PAGE_SHIFT);
+}
+
+static inline void scatterwalk_unmap(void *vaddr, int out)
+{
+	crypto_kunmap(vaddr, out);
+}
+
+void scatterwalk_start(struct scatter_walk *walk, struct scatterlist *sg);
+void scatterwalk_copychunks(void *buf, struct scatter_walk *walk,
+			    size_t nbytes, int out);
+void *scatterwalk_map(struct scatter_walk *walk, int out);
+void scatterwalk_done(struct scatter_walk *walk, int out, int more);
+
+void scatterwalk_map_and_copy(void *buf, struct scatterlist *sg,
+			      unsigned int start, unsigned int nbytes, int out);
+
+#endif  /* _CRYPTO_SCATTERWALK_H */
-- 
cgit v1.2.3


From b2ab4a57b018aafbba35bff088218f5cc3d2142e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 5 Dec 2007 20:59:25 +1100
Subject: [CRYPTO] scatterwalk: Restore custom sg chaining for now

Unfortunately the generic chaining hasn't been ported to all architectures
yet, and notably not s390.  So this patch restores the chainging that we've
been using previously which does work everywhere.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/digest.c              |  2 +-
 crypto/gcm.c                 |  2 +-
 crypto/hmac.c                |  3 ++-
 crypto/scatterwalk.c         |  4 ++--
 include/crypto/scatterwalk.h | 11 +++++++++++
 5 files changed, 17 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/crypto/digest.c b/crypto/digest.c
index 52845f53f8e1..6fd43bddd545 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -75,7 +75,7 @@ static int update2(struct hash_desc *desc,
 
 		if (!nbytes)
 			break;
-		sg = sg_next(sg);
+		sg = scatterwalk_sg_next(sg);
 	}
 
 	return 0;
diff --git a/crypto/gcm.c b/crypto/gcm.c
index 27483f361e80..502da929a5fc 100644
--- a/crypto/gcm.c
+++ b/crypto/gcm.c
@@ -100,7 +100,7 @@ static void crypto_gcm_ghash_update_sg(struct crypto_gcm_ghash_ctx *ctx,
 		n = scatterwalk_clamp(&walk, len);
 
 		if (!n) {
-			scatterwalk_start(&walk, sg_next(walk.sg));
+			scatterwalk_start(&walk, scatterwalk_sg_next(walk.sg));
 			n = scatterwalk_clamp(&walk, len);
 		}
 
diff --git a/crypto/hmac.c b/crypto/hmac.c
index 34c3706db85d..a1d016a50e7d 100644
--- a/crypto/hmac.c
+++ b/crypto/hmac.c
@@ -17,6 +17,7 @@
  */
 
 #include <crypto/algapi.h>
+#include <crypto/scatterwalk.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -160,7 +161,7 @@ static int hmac_digest(struct hash_desc *pdesc, struct scatterlist *sg,
 
 	sg_init_table(sg1, 2);
 	sg_set_buf(sg1, ipad, bs);
-	sg_chain(sg1, 2, sg);
+	scatterwalk_sg_chain(sg1, 2, sg);
 
 	sg_init_table(sg2, 1);
 	sg_set_buf(sg2, opad, bs + ds);
diff --git a/crypto/scatterwalk.c b/crypto/scatterwalk.c
index 12d19019e178..297e19d13c4e 100644
--- a/crypto/scatterwalk.c
+++ b/crypto/scatterwalk.c
@@ -61,7 +61,7 @@ static void scatterwalk_pagedone(struct scatter_walk *walk, int out,
 		walk->offset += PAGE_SIZE - 1;
 		walk->offset &= PAGE_MASK;
 		if (walk->offset >= walk->sg->offset + walk->sg->length)
-			scatterwalk_start(walk, sg_next(walk->sg));
+			scatterwalk_start(walk, scatterwalk_sg_next(walk->sg));
 	}
 }
 
@@ -112,7 +112,7 @@ void scatterwalk_map_and_copy(void *buf, struct scatterlist *sg,
 			break;
 
 		offset += sg->length;
-		sg = sg_next(sg);
+		sg = scatterwalk_sg_next(sg);
 	}
 
 	scatterwalk_advance(&walk, start - offset);
diff --git a/include/crypto/scatterwalk.h b/include/crypto/scatterwalk.h
index 07b6f17122d2..bd62431c33bd 100644
--- a/include/crypto/scatterwalk.h
+++ b/include/crypto/scatterwalk.h
@@ -52,6 +52,17 @@ static inline void crypto_yield(u32 flags)
 		cond_resched();
 }
 
+static inline void scatterwalk_sg_chain(struct scatterlist *sg1, int num,
+					struct scatterlist *sg2)
+{
+	sg_set_page(&sg1[num - 1], (void *)sg2, 0, 0);
+}
+
+static inline struct scatterlist *scatterwalk_sg_next(struct scatterlist *sg)
+{
+	return (++sg)->length ? sg : (void *)sg_page(sg);
+}
+
 static inline unsigned long scatterwalk_samebuf(struct scatter_walk *walk_in,
 						struct scatter_walk *walk_out)
 {
-- 
cgit v1.2.3


From 551a09a7a954f720067f207657bbbd26a3fe156a Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 1 Dec 2007 21:47:07 +1100
Subject: [CRYPTO] api: Sanitise mask when allocating ablkcipher/hash

When allocating ablkcipher/hash objects, we use a mask that's wider than
the usual type mask.  This patch sanitises the mask supplied by the user
so we don't end up using a narrower mask which may lead to unintended
results.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/crypto.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 48aa5959abbb..ef7642ed3e42 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -532,6 +532,7 @@ static inline struct crypto_ablkcipher *crypto_alloc_ablkcipher(
 	const char *alg_name, u32 type, u32 mask)
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
 	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 
@@ -554,6 +555,7 @@ static inline int crypto_has_ablkcipher(const char *alg_name, u32 type,
 					u32 mask)
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
 	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 
@@ -1086,6 +1088,7 @@ static inline struct crypto_hash *crypto_alloc_hash(const char *alg_name,
 						    u32 type, u32 mask)
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_HASH;
 	mask |= CRYPTO_ALG_TYPE_HASH_MASK;
 
@@ -1105,6 +1108,7 @@ static inline void crypto_free_hash(struct crypto_hash *tfm)
 static inline int crypto_has_hash(const char *alg_name, u32 type, u32 mask)
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_HASH;
 	mask |= CRYPTO_ALG_TYPE_HASH_MASK;
 
-- 
cgit v1.2.3


From 68b6c7d6919be7c732fc6229c55e35d0166e9258 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 7 Dec 2007 20:18:17 +0800
Subject: [CRYPTO] api: Add crypto_attr_alg_name

This patch adds a new helper crypto_attr_alg_name which is basically the
first half of crypto_attr_alg.  That is, it returns an algorithm name
parameter as a string without looking it up.  The caller can then look it
up immediately or defer it until later.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/algapi.c         | 18 ++++++++++++++++--
 include/crypto/algapi.h |  1 +
 2 files changed, 17 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index 08eca6da1d53..e65cb50cf4af 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -472,7 +472,7 @@ int crypto_check_attr_type(struct rtattr **tb, u32 type)
 }
 EXPORT_SYMBOL_GPL(crypto_check_attr_type);
 
-struct crypto_alg *crypto_attr_alg(struct rtattr *rta, u32 type, u32 mask)
+const char *crypto_attr_alg_name(struct rtattr *rta)
 {
 	struct crypto_attr_alg *alga;
 
@@ -486,7 +486,21 @@ struct crypto_alg *crypto_attr_alg(struct rtattr *rta, u32 type, u32 mask)
 	alga = RTA_DATA(rta);
 	alga->name[CRYPTO_MAX_ALG_NAME - 1] = 0;
 
-	return crypto_alg_mod_lookup(alga->name, type, mask);
+	return alga->name;
+}
+EXPORT_SYMBOL_GPL(crypto_attr_alg_name);
+
+struct crypto_alg *crypto_attr_alg(struct rtattr *rta, u32 type, u32 mask)
+{
+	const char *name;
+	int err;
+
+	name = crypto_attr_alg_name(rta);
+	err = PTR_ERR(name);
+	if (IS_ERR(name))
+		return ERR_PTR(err);
+
+	return crypto_alg_mod_lookup(name, type, mask);
 }
 EXPORT_SYMBOL_GPL(crypto_attr_alg);
 
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 2cdb227fc39d..726a765e5ec6 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -113,6 +113,7 @@ struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn, u32 type,
 
 struct crypto_attr_type *crypto_get_attr_type(struct rtattr **tb);
 int crypto_check_attr_type(struct rtattr **tb, u32 type);
+const char *crypto_attr_alg_name(struct rtattr *rta);
 struct crypto_alg *crypto_attr_alg(struct rtattr *rta, u32 type, u32 mask);
 int crypto_attr_u32(struct rtattr *rta, u32 *num);
 struct crypto_instance *crypto_alloc_instance(const char *name,
-- 
cgit v1.2.3


From 5311f248b7764ba8b59e6d477355f766e5609686 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 17 Dec 2007 21:34:32 +0800
Subject: [CRYPTO] ctr: Refactor into ctr and rfc3686

As discussed previously, this patch moves the basic CTR functionality
into a chainable algorithm called ctr.  The IPsec-specific variant of
it is now placed on top with the name rfc3686.

So ctr(aes) gives a chainable cipher with IV size 16 while the IPsec
variant will be called rfc3686(ctr(aes)).  This patch also adjusts
gcm accordingly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ctr.c         | 334 +++++++++++++++++++++++++++++++--------------------
 crypto/gcm.c         |   7 +-
 crypto/tcrypt.c      |   8 +-
 include/crypto/ctr.h |  20 +++
 4 files changed, 233 insertions(+), 136 deletions(-)
 create mode 100644 include/crypto/ctr.h

(limited to 'include')

diff --git a/crypto/ctr.c b/crypto/ctr.c
index 57da7d0affcb..1052b318aed2 100644
--- a/crypto/ctr.c
+++ b/crypto/ctr.c
@@ -11,6 +11,7 @@
  */
 
 #include <crypto/algapi.h>
+#include <crypto/ctr.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -19,16 +20,13 @@
 #include <linux/scatterlist.h>
 #include <linux/slab.h>
 
-struct ctr_instance_ctx {
-	struct crypto_spawn alg;
-	unsigned int noncesize;
-	unsigned int ivsize;
-	unsigned int countersize;
-};
-
 struct crypto_ctr_ctx {
 	struct crypto_cipher *child;
-	u8 *nonce;
+};
+
+struct crypto_rfc3686_ctx {
+	struct crypto_blkcipher *child;
+	u8 nonce[CTR_RFC3686_NONCE_SIZE];
 };
 
 static int crypto_ctr_setkey(struct crypto_tfm *parent, const u8 *key,
@@ -36,18 +34,7 @@ static int crypto_ctr_setkey(struct crypto_tfm *parent, const u8 *key,
 {
 	struct crypto_ctr_ctx *ctx = crypto_tfm_ctx(parent);
 	struct crypto_cipher *child = ctx->child;
-	struct ctr_instance_ctx *ictx =
-		crypto_instance_ctx(crypto_tfm_alg_instance(parent));
-	unsigned int noncelen = ictx->noncesize;
-	int err = 0;
-
-	/* the nonce is stored in bytes at end of key */
-	if (keylen < noncelen)
-		return  -EINVAL;
-
-	memcpy(ctx->nonce, key + (keylen - noncelen), noncelen);
-
-	keylen -=  noncelen;
+	int err;
 
 	crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK);
 	crypto_cipher_set_flags(child, crypto_tfm_get_flags(parent) &
@@ -60,11 +47,13 @@ static int crypto_ctr_setkey(struct crypto_tfm *parent, const u8 *key,
 }
 
 static void crypto_ctr_crypt_final(struct blkcipher_walk *walk,
-				   struct crypto_cipher *tfm, u8 *ctrblk,
-				   unsigned int countersize)
+				   struct crypto_cipher *tfm)
 {
 	unsigned int bsize = crypto_cipher_blocksize(tfm);
-	u8 *keystream = ctrblk + bsize;
+	unsigned long alignmask = crypto_cipher_alignmask(tfm);
+	u8 *ctrblk = walk->iv;
+	u8 tmp[bsize + alignmask];
+	u8 *keystream = PTR_ALIGN(tmp + 0, alignmask + 1);
 	u8 *src = walk->src.virt.addr;
 	u8 *dst = walk->dst.virt.addr;
 	unsigned int nbytes = walk->nbytes;
@@ -72,15 +61,17 @@ static void crypto_ctr_crypt_final(struct blkcipher_walk *walk,
 	crypto_cipher_encrypt_one(tfm, keystream, ctrblk);
 	crypto_xor(keystream, src, nbytes);
 	memcpy(dst, keystream, nbytes);
+
+	crypto_inc(ctrblk, bsize);
 }
 
 static int crypto_ctr_crypt_segment(struct blkcipher_walk *walk,
-				    struct crypto_cipher *tfm, u8 *ctrblk,
-				    unsigned int countersize)
+				    struct crypto_cipher *tfm)
 {
 	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
 		   crypto_cipher_alg(tfm)->cia_encrypt;
 	unsigned int bsize = crypto_cipher_blocksize(tfm);
+	u8 *ctrblk = walk->iv;
 	u8 *src = walk->src.virt.addr;
 	u8 *dst = walk->dst.virt.addr;
 	unsigned int nbytes = walk->nbytes;
@@ -91,7 +82,7 @@ static int crypto_ctr_crypt_segment(struct blkcipher_walk *walk,
 		crypto_xor(dst, src, bsize);
 
 		/* increment counter in counterblock */
-		crypto_inc(ctrblk + bsize - countersize, countersize);
+		crypto_inc(ctrblk, bsize);
 
 		src += bsize;
 		dst += bsize;
@@ -101,15 +92,17 @@ static int crypto_ctr_crypt_segment(struct blkcipher_walk *walk,
 }
 
 static int crypto_ctr_crypt_inplace(struct blkcipher_walk *walk,
-				    struct crypto_cipher *tfm, u8 *ctrblk,
-				    unsigned int countersize)
+				    struct crypto_cipher *tfm)
 {
 	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
 		   crypto_cipher_alg(tfm)->cia_encrypt;
 	unsigned int bsize = crypto_cipher_blocksize(tfm);
+	unsigned long alignmask = crypto_cipher_alignmask(tfm);
 	unsigned int nbytes = walk->nbytes;
+	u8 *ctrblk = walk->iv;
 	u8 *src = walk->src.virt.addr;
-	u8 *keystream = ctrblk + bsize;
+	u8 tmp[bsize + alignmask];
+	u8 *keystream = PTR_ALIGN(tmp + 0, alignmask + 1);
 
 	do {
 		/* create keystream */
@@ -117,7 +110,7 @@ static int crypto_ctr_crypt_inplace(struct blkcipher_walk *walk,
 		crypto_xor(src, keystream, bsize);
 
 		/* increment counter in counterblock */
-		crypto_inc(ctrblk + bsize - countersize, countersize);
+		crypto_inc(ctrblk, bsize);
 
 		src += bsize;
 	} while ((nbytes -= bsize) >= bsize);
@@ -134,41 +127,22 @@ static int crypto_ctr_crypt(struct blkcipher_desc *desc,
 	struct crypto_ctr_ctx *ctx = crypto_blkcipher_ctx(tfm);
 	struct crypto_cipher *child = ctx->child;
 	unsigned int bsize = crypto_cipher_blocksize(child);
-	struct ctr_instance_ctx *ictx =
-		crypto_instance_ctx(crypto_tfm_alg_instance(&tfm->base));
-	unsigned long alignmask = crypto_cipher_alignmask(child) |
-				  (__alignof__(u32) - 1);
-	u8 cblk[bsize * 2 + alignmask];
-	u8 *counterblk = (u8 *)ALIGN((unsigned long)cblk, alignmask + 1);
 	int err;
 
 	blkcipher_walk_init(&walk, dst, src, nbytes);
 	err = blkcipher_walk_virt_block(desc, &walk, bsize);
 
-	/* set up counter block */
-	memset(counterblk, 0 , bsize);
-	memcpy(counterblk, ctx->nonce, ictx->noncesize);
-	memcpy(counterblk + ictx->noncesize, walk.iv, ictx->ivsize);
-
-	/* initialize counter portion of counter block */
-	crypto_inc(counterblk + bsize - ictx->countersize, ictx->countersize);
-
 	while (walk.nbytes >= bsize) {
 		if (walk.src.virt.addr == walk.dst.virt.addr)
-			nbytes = crypto_ctr_crypt_inplace(&walk, child,
-							  counterblk,
-							  ictx->countersize);
+			nbytes = crypto_ctr_crypt_inplace(&walk, child);
 		else
-			nbytes = crypto_ctr_crypt_segment(&walk, child,
-							  counterblk,
-							  ictx->countersize);
+			nbytes = crypto_ctr_crypt_segment(&walk, child);
 
 		err = blkcipher_walk_done(desc, &walk, nbytes);
 	}
 
 	if (walk.nbytes) {
-		crypto_ctr_crypt_final(&walk, child, counterblk,
-				       ictx->countersize);
+		crypto_ctr_crypt_final(&walk, child);
 		err = blkcipher_walk_done(desc, &walk, 0);
 	}
 
@@ -178,15 +152,11 @@ static int crypto_ctr_crypt(struct blkcipher_desc *desc,
 static int crypto_ctr_init_tfm(struct crypto_tfm *tfm)
 {
 	struct crypto_instance *inst = (void *)tfm->__crt_alg;
-	struct ctr_instance_ctx *ictx = crypto_instance_ctx(inst);
+	struct crypto_spawn *spawn = crypto_instance_ctx(inst);
 	struct crypto_ctr_ctx *ctx = crypto_tfm_ctx(tfm);
 	struct crypto_cipher *cipher;
 
-	ctx->nonce = kzalloc(ictx->noncesize, GFP_KERNEL);
-	if (!ctx->nonce)
-		return -ENOMEM;
-
-	cipher = crypto_spawn_cipher(&ictx->alg);
+	cipher = crypto_spawn_cipher(spawn);
 	if (IS_ERR(cipher))
 		return PTR_ERR(cipher);
 
@@ -199,7 +169,6 @@ static void crypto_ctr_exit_tfm(struct crypto_tfm *tfm)
 {
 	struct crypto_ctr_ctx *ctx = crypto_tfm_ctx(tfm);
 
-	kfree(ctx->nonce);
 	crypto_free_cipher(ctx->child);
 }
 
@@ -207,10 +176,6 @@ static struct crypto_instance *crypto_ctr_alloc(struct rtattr **tb)
 {
 	struct crypto_instance *inst;
 	struct crypto_alg *alg;
-	struct ctr_instance_ctx *ictx;
-	unsigned int noncesize;
-	unsigned int ivsize;
-	unsigned int countersize;
 	int err;
 
 	err = crypto_check_attr_type(tb, CRYPTO_ALG_TYPE_BLKCIPHER);
@@ -222,71 +187,28 @@ static struct crypto_instance *crypto_ctr_alloc(struct rtattr **tb)
 	if (IS_ERR(alg))
 		return ERR_PTR(PTR_ERR(alg));
 
-	err = crypto_attr_u32(tb[2], &noncesize);
-	if (err)
-		goto out_put_alg;
-
-	err = crypto_attr_u32(tb[3], &ivsize);
-	if (err)
-		goto out_put_alg;
-
-	err = crypto_attr_u32(tb[4], &countersize);
-	if (err)
-		goto out_put_alg;
-
-	/* verify size of nonce + iv + counter
-	 * counter must be >= 4 bytes.
-	 */
+	/* Block size must be >= 4 bytes. */
 	err = -EINVAL;
-	if (((noncesize + ivsize + countersize) < alg->cra_blocksize) ||
-	    ((noncesize + ivsize) > alg->cra_blocksize) ||
-	    (countersize > alg->cra_blocksize) || (countersize < 4))
+	if (alg->cra_blocksize < 4)
 		goto out_put_alg;
 
 	/* If this is false we'd fail the alignment of crypto_inc. */
-	if ((alg->cra_blocksize - countersize) % 4)
+	if (alg->cra_blocksize % 4)
 		goto out_put_alg;
 
-	inst = kzalloc(sizeof(*inst) + sizeof(*ictx), GFP_KERNEL);
-	err = -ENOMEM;
-	if (!inst)
-		goto out_put_alg;
-
-	err = -ENAMETOOLONG;
-	if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
-		     "ctr(%s,%u,%u,%u)", alg->cra_name, noncesize,
-		     ivsize, countersize) >= CRYPTO_MAX_ALG_NAME) {
-		goto err_free_inst;
-	}
-
-	if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
-		     "ctr(%s,%u,%u,%u)", alg->cra_driver_name, noncesize,
-		     ivsize, countersize) >= CRYPTO_MAX_ALG_NAME) {
-		goto err_free_inst;
-	}
-
-	ictx = crypto_instance_ctx(inst);
-	ictx->noncesize = noncesize;
-	ictx->ivsize = ivsize;
-	ictx->countersize = countersize;
-
-	err = crypto_init_spawn(&ictx->alg, alg, inst,
-		CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
-	if (err)
-		goto err_free_inst;
+	inst = crypto_alloc_instance("ctr", alg);
+	if (IS_ERR(inst))
+		goto out;
 
-	err = 0;
 	inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER;
 	inst->alg.cra_priority = alg->cra_priority;
 	inst->alg.cra_blocksize = 1;
 	inst->alg.cra_alignmask = alg->cra_alignmask | (__alignof__(u32) - 1);
 	inst->alg.cra_type = &crypto_blkcipher_type;
 
-	inst->alg.cra_blkcipher.ivsize = ivsize;
-	inst->alg.cra_blkcipher.min_keysize = alg->cra_cipher.cia_min_keysize
-					      + noncesize;
-	inst->alg.cra_blkcipher.max_keysize = alg->cra_cipher.cia_max_keysize
-					      + noncesize;
+	inst->alg.cra_blkcipher.ivsize = alg->cra_blocksize;
+	inst->alg.cra_blkcipher.min_keysize = alg->cra_cipher.cia_min_keysize;
+	inst->alg.cra_blkcipher.max_keysize = alg->cra_cipher.cia_max_keysize;
 
 	inst->alg.cra_ctxsize = sizeof(struct crypto_ctr_ctx);
 
@@ -297,24 +219,18 @@ static struct crypto_instance *crypto_ctr_alloc(struct rtattr **tb)
 	inst->alg.cra_blkcipher.encrypt = crypto_ctr_crypt;
 	inst->alg.cra_blkcipher.decrypt = crypto_ctr_crypt;
 
-err_free_inst:
-	if (err)
-		kfree(inst);
-
-out_put_alg:
+out:
 	crypto_mod_put(alg);
-
-	if (err)
-		inst = ERR_PTR(err);
-
 	return inst;
+
+out_put_alg:
+	inst = ERR_PTR(err);
+	goto out;
 }
 
 static void crypto_ctr_free(struct crypto_instance *inst)
 {
-	struct ctr_instance_ctx *ictx = crypto_instance_ctx(inst);
-
-	crypto_drop_spawn(&ictx->alg);
+	crypto_drop_spawn(crypto_instance_ctx(inst));
 	kfree(inst);
 }
 
@@ -325,13 +241,174 @@ static struct crypto_template crypto_ctr_tmpl = {
 	.module = THIS_MODULE,
 };
 
+static int crypto_rfc3686_setkey(struct crypto_tfm *parent, const u8 *key,
+				 unsigned int keylen)
+{
+	struct crypto_rfc3686_ctx *ctx = crypto_tfm_ctx(parent);
+	struct crypto_blkcipher *child = ctx->child;
+	int err;
+
+	/* the nonce is stored in bytes at end of key */
+	if (keylen < CTR_RFC3686_NONCE_SIZE)
+		return -EINVAL;
+
+	memcpy(ctx->nonce, key + (keylen - CTR_RFC3686_NONCE_SIZE),
+	       CTR_RFC3686_NONCE_SIZE);
+
+	keylen -= CTR_RFC3686_NONCE_SIZE;
+
+	crypto_blkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK);
+	crypto_blkcipher_set_flags(child, crypto_tfm_get_flags(parent) &
+					  CRYPTO_TFM_REQ_MASK);
+	err = crypto_blkcipher_setkey(child, key, keylen);
+	crypto_tfm_set_flags(parent, crypto_blkcipher_get_flags(child) &
+				     CRYPTO_TFM_RES_MASK);
+
+	return err;
+}
+
+static int crypto_rfc3686_crypt(struct blkcipher_desc *desc,
+				struct scatterlist *dst,
+				struct scatterlist *src, unsigned int nbytes)
+{
+	struct crypto_blkcipher *tfm = desc->tfm;
+	struct crypto_rfc3686_ctx *ctx = crypto_blkcipher_ctx(tfm);
+	struct crypto_blkcipher *child = ctx->child;
+	unsigned long alignmask = crypto_blkcipher_alignmask(tfm);
+	u8 ivblk[CTR_RFC3686_BLOCK_SIZE + alignmask];
+	u8 *iv = PTR_ALIGN(ivblk + 0, alignmask + 1);
+	u8 *info = desc->info;
+	int err;
+
+	/* set up counter block */
+	memcpy(iv, ctx->nonce, CTR_RFC3686_NONCE_SIZE);
+	memcpy(iv + CTR_RFC3686_NONCE_SIZE, info, CTR_RFC3686_IV_SIZE);
+
+	/* initialize counter portion of counter block */
+	*(__be32 *)(iv + CTR_RFC3686_NONCE_SIZE + CTR_RFC3686_IV_SIZE) =
+		cpu_to_be32(1);
+
+	desc->tfm = child;
+	desc->info = iv;
+	err = crypto_blkcipher_encrypt_iv(desc, dst, src, nbytes);
+	desc->tfm = tfm;
+	desc->info = info;
+
+	return err;
+}
+
+static int crypto_rfc3686_init_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = (void *)tfm->__crt_alg;
+	struct crypto_spawn *spawn = crypto_instance_ctx(inst);
+	struct crypto_rfc3686_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct crypto_blkcipher *cipher;
+
+	cipher = crypto_spawn_blkcipher(spawn);
+	if (IS_ERR(cipher))
+		return PTR_ERR(cipher);
+
+	ctx->child = cipher;
+
+	return 0;
+}
+
+static void crypto_rfc3686_exit_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_rfc3686_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	crypto_free_blkcipher(ctx->child);
+}
+
+static struct crypto_instance *crypto_rfc3686_alloc(struct rtattr **tb)
+{
+	struct crypto_instance *inst;
+	struct crypto_alg *alg;
+	int err;
+
+	err = crypto_check_attr_type(tb, CRYPTO_ALG_TYPE_BLKCIPHER);
+	if (err)
+		return ERR_PTR(err);
+
+	alg = crypto_attr_alg(tb[1], CRYPTO_ALG_TYPE_BLKCIPHER,
+				  CRYPTO_ALG_TYPE_MASK);
+	err = PTR_ERR(alg);
+	if (IS_ERR(alg))
+		return ERR_PTR(err);
+
+	/* We only support 16-byte blocks. */
+	err = -EINVAL;
+	if (alg->cra_blkcipher.ivsize != CTR_RFC3686_BLOCK_SIZE)
+		goto out_put_alg;
+
+	/* Not a stream cipher? */
+	if (alg->cra_blocksize != 1)
+		goto out_put_alg;
+
+	inst = crypto_alloc_instance("rfc3686", alg);
+	if (IS_ERR(inst))
+		goto out;
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER;
+	inst->alg.cra_priority = alg->cra_priority;
+	inst->alg.cra_blocksize = 1;
+	inst->alg.cra_alignmask = alg->cra_alignmask;
+	inst->alg.cra_type = &crypto_blkcipher_type;
+
+	inst->alg.cra_blkcipher.ivsize = CTR_RFC3686_IV_SIZE;
+	inst->alg.cra_blkcipher.min_keysize = alg->cra_blkcipher.min_keysize
+					      + CTR_RFC3686_NONCE_SIZE;
+	inst->alg.cra_blkcipher.max_keysize = alg->cra_blkcipher.max_keysize
+					      + CTR_RFC3686_NONCE_SIZE;
+
+	inst->alg.cra_ctxsize = sizeof(struct crypto_rfc3686_ctx);
+
+	inst->alg.cra_init = crypto_rfc3686_init_tfm;
+	inst->alg.cra_exit = crypto_rfc3686_exit_tfm;
+
+	inst->alg.cra_blkcipher.setkey = crypto_rfc3686_setkey;
+	inst->alg.cra_blkcipher.encrypt = crypto_rfc3686_crypt;
+	inst->alg.cra_blkcipher.decrypt = crypto_rfc3686_crypt;
+
+out:
+	crypto_mod_put(alg);
+	return inst;
+
+out_put_alg:
+	inst = ERR_PTR(err);
+	goto out;
+}
+
+static struct crypto_template crypto_rfc3686_tmpl = {
+	.name = "rfc3686",
+	.alloc = crypto_rfc3686_alloc,
+	.free = crypto_ctr_free,
+	.module = THIS_MODULE,
+};
+
 static int __init crypto_ctr_module_init(void)
 {
-	return crypto_register_template(&crypto_ctr_tmpl);
+	int err;
+
+	err = crypto_register_template(&crypto_ctr_tmpl);
+	if (err)
+		goto out;
+
+	err = crypto_register_template(&crypto_rfc3686_tmpl);
+	if (err)
+		goto out_drop_ctr;
+
+out:
+	return err;
+
+out_drop_ctr:
+	crypto_unregister_template(&crypto_ctr_tmpl);
+	goto out;
 }
 
 static void __exit crypto_ctr_module_exit(void)
 {
+	crypto_unregister_template(&crypto_rfc3686_tmpl);
 	crypto_unregister_template(&crypto_ctr_tmpl);
 }
 
@@ -340,3 +417,4 @@ module_exit(crypto_ctr_module_exit);
 
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("CTR Counter block mode");
+MODULE_ALIAS("rfc3686");
diff --git a/crypto/gcm.c b/crypto/gcm.c
index 08183171913c..c54d478948a0 100644
--- a/crypto/gcm.c
+++ b/crypto/gcm.c
@@ -160,7 +160,7 @@ static void crypto_gcm_ghash_final_xor(struct crypto_gcm_ghash_ctx *ctx,
 
 static inline void crypto_gcm_set_counter(u8 *counterblock, u32 value)
 {
-	*((u32 *)&counterblock[12]) = cpu_to_be32(value);
+	*((u32 *)&counterblock[12]) = cpu_to_be32(value + 1);
 }
 
 static int crypto_gcm_encrypt_counter(struct crypto_aead *aead, u8 *block,
@@ -400,9 +400,8 @@ static struct crypto_instance *crypto_gcm_alloc(struct rtattr **tb)
 		return inst;
 
 	inst = ERR_PTR(ENAMETOOLONG);
-	if (snprintf(
-		    ctr_name, CRYPTO_MAX_ALG_NAME,
-		    "ctr(%s,0,16,4)", cipher->cra_name) >= CRYPTO_MAX_ALG_NAME)
+	if (snprintf(ctr_name, CRYPTO_MAX_ALG_NAME, "ctr(%s)",
+		     cipher->cra_name) >= CRYPTO_MAX_ALG_NAME)
 		return inst;
 
 	ctr = crypto_alg_mod_lookup(ctr_name, CRYPTO_ALG_TYPE_BLKCIPHER,
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index 0cfb8ebb22ba..1142b4998c84 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -1193,9 +1193,9 @@ static void do_test(void)
 			    AES_XTS_ENC_TEST_VECTORS);
 		test_cipher("xts(aes)", DECRYPT, aes_xts_dec_tv_template,
 			    AES_XTS_DEC_TEST_VECTORS);
-		test_cipher("ctr(aes,4,8,4)", ENCRYPT, aes_ctr_enc_tv_template,
+		test_cipher("rfc3686(ctr(aes))", ENCRYPT, aes_ctr_enc_tv_template,
 			    AES_CTR_ENC_TEST_VECTORS);
-		test_cipher("ctr(aes,4,8,4)", DECRYPT, aes_ctr_dec_tv_template,
+		test_cipher("rfc3686(ctr(aes))", DECRYPT, aes_ctr_dec_tv_template,
 			    AES_CTR_DEC_TEST_VECTORS);
 		test_aead("gcm(aes)", ENCRYPT, aes_gcm_enc_tv_template,
 			  AES_GCM_ENC_TEST_VECTORS);
@@ -1394,9 +1394,9 @@ static void do_test(void)
 			    AES_XTS_ENC_TEST_VECTORS);
 		test_cipher("xts(aes)", DECRYPT, aes_xts_dec_tv_template,
 			    AES_XTS_DEC_TEST_VECTORS);
-		test_cipher("ctr(aes,4,8,4)", ENCRYPT, aes_ctr_enc_tv_template,
+		test_cipher("rfc3686(ctr(aes))", ENCRYPT, aes_ctr_enc_tv_template,
 			    AES_CTR_ENC_TEST_VECTORS);
-		test_cipher("ctr(aes,4,8,4)", DECRYPT, aes_ctr_dec_tv_template,
+		test_cipher("rfc3686(ctr(aes))", DECRYPT, aes_ctr_dec_tv_template,
 			    AES_CTR_DEC_TEST_VECTORS);
 		break;
 
diff --git a/include/crypto/ctr.h b/include/crypto/ctr.h
new file mode 100644
index 000000000000..4180fc080e3b
--- /dev/null
+++ b/include/crypto/ctr.h
@@ -0,0 +1,20 @@
+/*
+ * CTR: Counter mode
+ *
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_CTR_H
+#define _CRYPTO_CTR_H
+
+#define CTR_RFC3686_NONCE_SIZE 4
+#define CTR_RFC3686_IV_SIZE 8
+#define CTR_RFC3686_BLOCK_SIZE 16
+
+#endif  /* _CRYPTO_CTR_H */
-- 
cgit v1.2.3


From 378f4f51f9fdd8df80ea875320e2bf1d7c6e6e77 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 17 Dec 2007 20:07:31 +0800
Subject: [CRYPTO] skcipher: Add crypto_grab_skcipher interface

Note: From now on the collective of ablkcipher/blkcipher/givcipher will
be known as skcipher, i.e., symmetric key cipher.  The name blkcipher has
always been much of a misnomer since it supports stream ciphers too.

This patch adds the function crypto_grab_skcipher as a new way of getting
an ablkcipher spawn.  The problem is that previously we did this in two
steps, first getting the algorithm and then calling crypto_init_spawn.

This meant that each spawn user had to be aware of what type and mask to
use for these two steps.  This is difficult and also presents a problem
when the type/mask changes as they're about to be for IV generators.

The new interface does both steps together just like crypto_alloc_ablkcipher.

As a side-effect this also allows us to be stronger on type enforcement
for spawns.  For now this is only done for ablkcipher but it's trivial
to extend for other types.

This patch also moves the type/mask logic for skcipher into the helpers
crypto_skcipher_type and crypto_skcipher_mask.

Finally this patch introduces the function crypto_require_sync to determine
whether the user is specifically requesting a sync algorithm.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ablkcipher.c                | 25 +++++++++++++++++--
 include/crypto/algapi.h            | 22 +++++++++++++---
 include/crypto/internal/skcipher.h | 51 ++++++++++++++++++++++++++++++++++++++
 include/linux/crypto.h             | 26 +++++++++++--------
 4 files changed, 108 insertions(+), 16 deletions(-)
 create mode 100644 include/crypto/internal/skcipher.h

(limited to 'include')

diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index 2731acb86e7d..0083140304d2 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -13,14 +13,16 @@
  *
  */
 
-#include <crypto/algapi.h>
-#include <linux/errno.h>
+#include <crypto/internal/skcipher.h>
+#include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
 
+#include "internal.h"
+
 static int setkey_unaligned(struct crypto_ablkcipher *tfm, const u8 *key,
 			    unsigned int keylen)
 {
@@ -105,5 +107,24 @@ const struct crypto_type crypto_ablkcipher_type = {
 };
 EXPORT_SYMBOL_GPL(crypto_ablkcipher_type);
 
+int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
+			 u32 type, u32 mask)
+{
+	struct crypto_alg *alg;
+	int err;
+
+	type = crypto_skcipher_type(type);
+	mask = crypto_skcipher_mask(mask);
+
+	alg = crypto_alg_mod_lookup(name, type, mask);
+	if (IS_ERR(alg))
+		return PTR_ERR(alg);
+
+	err = crypto_init_spawn(&spawn->base, alg, spawn->base.inst, mask);
+	crypto_mod_put(alg);
+	return err;
+}
+EXPORT_SYMBOL_GPL(crypto_grab_skcipher);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Asynchronous block chaining cipher type");
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 726a765e5ec6..fda1759ffe2d 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -111,6 +111,12 @@ void crypto_drop_spawn(struct crypto_spawn *spawn);
 struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn, u32 type,
 				    u32 mask);
 
+static inline void crypto_set_spawn(struct crypto_spawn *spawn,
+				    struct crypto_instance *inst)
+{
+	spawn->inst = inst;
+}
+
 struct crypto_attr_type *crypto_get_attr_type(struct rtattr **tb);
 int crypto_check_attr_type(struct rtattr **tb, u32 type);
 const char *crypto_attr_alg_name(struct rtattr *rta);
@@ -195,10 +201,9 @@ static inline struct crypto_instance *crypto_aead_alg_instance(
 static inline struct crypto_ablkcipher *crypto_spawn_ablkcipher(
 	struct crypto_spawn *spawn)
 {
-	u32 type = CRYPTO_ALG_TYPE_BLKCIPHER;
-	u32 mask = CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
-
-	return __crypto_ablkcipher_cast(crypto_spawn_tfm(spawn, type, mask));
+	return __crypto_ablkcipher_cast(
+		crypto_spawn_tfm(spawn, crypto_skcipher_type(0),
+				 crypto_skcipher_mask(0)));
 }
 
 static inline struct crypto_blkcipher *crypto_spawn_blkcipher(
@@ -308,5 +313,14 @@ static inline struct crypto_alg *crypto_get_attr_alg(struct rtattr **tb,
 	return crypto_attr_alg(tb[1], type, mask);
 }
 
+/*
+ * Returns CRYPTO_ALG_ASYNC if type/mask requires the use of sync algorithms.
+ * Otherwise returns zero.
+ */
+static inline int crypto_requires_sync(u32 type, u32 mask)
+{
+	return (type ^ CRYPTO_ALG_ASYNC) & mask & CRYPTO_ALG_ASYNC;
+}
+
 #endif	/* _CRYPTO_ALGAPI_H */
 
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
new file mode 100644
index 000000000000..87879e64ff4c
--- /dev/null
+++ b/include/crypto/internal/skcipher.h
@@ -0,0 +1,51 @@
+/*
+ * Symmetric key ciphers.
+ * 
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_INTERNAL_SKCIPHER_H
+#define _CRYPTO_INTERNAL_SKCIPHER_H
+
+#include <crypto/algapi.h>
+
+struct crypto_skcipher_spawn {
+	struct crypto_spawn base;
+};
+
+static inline void crypto_set_skcipher_spawn(
+	struct crypto_skcipher_spawn *spawn, struct crypto_instance *inst)
+{
+	crypto_set_spawn(&spawn->base, inst);
+}
+
+int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
+			 u32 type, u32 mask);
+
+static inline void crypto_drop_skcipher(struct crypto_skcipher_spawn *spawn)
+{
+	crypto_drop_spawn(&spawn->base);
+}
+
+static inline struct crypto_alg *crypto_skcipher_spawn_alg(
+	struct crypto_skcipher_spawn *spawn)
+{
+	return spawn->base.alg;
+}
+
+static inline struct crypto_ablkcipher *crypto_spawn_skcipher(
+	struct crypto_skcipher_spawn *spawn)
+{
+	return __crypto_ablkcipher_cast(
+		crypto_spawn_tfm(&spawn->base, crypto_skcipher_type(0),
+				 crypto_skcipher_mask(0)));
+}
+
+#endif	/* _CRYPTO_INTERNAL_SKCIPHER_H */
+
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index ef7642ed3e42..d6962b409489 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -528,16 +528,26 @@ static inline struct crypto_ablkcipher *__crypto_ablkcipher_cast(
 	return (struct crypto_ablkcipher *)tfm;
 }
 
-static inline struct crypto_ablkcipher *crypto_alloc_ablkcipher(
-	const char *alg_name, u32 type, u32 mask)
+static inline u32 crypto_skcipher_type(u32 type)
 {
 	type &= ~CRYPTO_ALG_TYPE_MASK;
-	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
+	return type;
+}
+
+static inline u32 crypto_skcipher_mask(u32 mask)
+{
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
 	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
+	return mask;
+}
 
+static inline struct crypto_ablkcipher *crypto_alloc_ablkcipher(
+	const char *alg_name, u32 type, u32 mask)
+{
 	return __crypto_ablkcipher_cast(
-		crypto_alloc_base(alg_name, type, mask));
+		crypto_alloc_base(alg_name, crypto_skcipher_type(type),
+				  crypto_skcipher_mask(mask)));
 }
 
 static inline struct crypto_tfm *crypto_ablkcipher_tfm(
@@ -554,12 +564,8 @@ static inline void crypto_free_ablkcipher(struct crypto_ablkcipher *tfm)
 static inline int crypto_has_ablkcipher(const char *alg_name, u32 type,
 					u32 mask)
 {
-	type &= ~CRYPTO_ALG_TYPE_MASK;
-	mask &= ~CRYPTO_ALG_TYPE_MASK;
-	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
-	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
-
-	return crypto_has_alg(alg_name, type, mask);
+	return crypto_has_alg(alg_name, crypto_skcipher_type(type),
+			      crypto_skcipher_mask(mask));
 }
 
 static inline struct ablkcipher_tfm *crypto_ablkcipher_crt(
-- 
cgit v1.2.3


From 61da88e2b800eed2b03834a73c46cc89ad48716d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 17 Dec 2007 21:51:27 +0800
Subject: [CRYPTO] skcipher: Add givcrypt operations and givcipher type

Different block cipher modes have different requirements for intialisation
vectors.  For example, CBC can use a simple randomly generated IV while
modes such as CTR must use an IV generation mechanisms that give a stronger
guarantee on the lack of collisions.  Furthermore, disk encryption modes
have their own IV generation algorithms.

Up until now IV generation has been left to the users of the symmetric
key cipher API.  This is inconvenient as the number of block cipher modes
increase because the user needs to be aware of which mode is supposed to
be paired with which IV generation algorithm.

Therefore it makes sense to integrate the IV generation into the crypto
API.  This patch takes the first step in that direction by creating two
new ablkcipher operations, givencrypt and givdecrypt that generates an
IV before performing the actual encryption or decryption.

The operations are currently not exposed to the user.  That will be done
once the underlying functionality has actually been implemented.

It also creates the underlying givcipher type.  Algorithms that directly
generate IVs would use it instead of ablkcipher.  All other algorithms
(including all existing ones) would generate a givcipher algorithm upon
registration.  This givcipher algorithm will be constructed from the geniv
string that's stored in every algorithm.  That string will locate a template
which is instantiated by the blkcipher/ablkcipher algorithm in question to
give a givcipher algorithm.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ablkcipher.c                | 46 ++++++++++++++++++++++++++++++++++++++
 include/crypto/internal/skcipher.h |  9 ++++++++
 include/crypto/skcipher.h          | 38 +++++++++++++++++++++++++++++++
 include/linux/crypto.h             |  7 ++++++
 4 files changed, 100 insertions(+)
 create mode 100644 include/crypto/skcipher.h

(limited to 'include')

diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index 0083140304d2..e403d8137ecd 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -107,6 +107,52 @@ const struct crypto_type crypto_ablkcipher_type = {
 };
 EXPORT_SYMBOL_GPL(crypto_ablkcipher_type);
 
+static int no_givdecrypt(struct skcipher_givcrypt_request *req)
+{
+	return -ENOSYS;
+}
+
+static int crypto_init_givcipher_ops(struct crypto_tfm *tfm, u32 type,
+				      u32 mask)
+{
+	struct ablkcipher_alg *alg = &tfm->__crt_alg->cra_ablkcipher;
+	struct ablkcipher_tfm *crt = &tfm->crt_ablkcipher;
+
+	if (alg->ivsize > PAGE_SIZE / 8)
+		return -EINVAL;
+
+	crt->setkey = setkey;
+	crt->encrypt = alg->encrypt;
+	crt->decrypt = alg->decrypt;
+	crt->givencrypt = alg->givencrypt;
+	crt->givdecrypt = alg->givdecrypt ?: no_givdecrypt;
+	crt->ivsize = alg->ivsize;
+
+	return 0;
+}
+
+static void crypto_givcipher_show(struct seq_file *m, struct crypto_alg *alg)
+	__attribute__ ((unused));
+static void crypto_givcipher_show(struct seq_file *m, struct crypto_alg *alg)
+{
+	struct ablkcipher_alg *ablkcipher = &alg->cra_ablkcipher;
+
+	seq_printf(m, "type         : givcipher\n");
+	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
+	seq_printf(m, "min keysize  : %u\n", ablkcipher->min_keysize);
+	seq_printf(m, "max keysize  : %u\n", ablkcipher->max_keysize);
+	seq_printf(m, "ivsize       : %u\n", ablkcipher->ivsize);
+}
+
+const struct crypto_type crypto_givcipher_type = {
+	.ctxsize = crypto_ablkcipher_ctxsize,
+	.init = crypto_init_givcipher_ops,
+#ifdef CONFIG_PROC_FS
+	.show = crypto_givcipher_show,
+#endif
+};
+EXPORT_SYMBOL_GPL(crypto_givcipher_type);
+
 int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
 			 u32 type, u32 mask)
 {
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index 87879e64ff4c..c9402dd12d03 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -14,11 +14,14 @@
 #define _CRYPTO_INTERNAL_SKCIPHER_H
 
 #include <crypto/algapi.h>
+#include <crypto/skcipher.h>
 
 struct crypto_skcipher_spawn {
 	struct crypto_spawn base;
 };
 
+extern const struct crypto_type crypto_givcipher_type;
+
 static inline void crypto_set_skcipher_spawn(
 	struct crypto_skcipher_spawn *spawn, struct crypto_instance *inst)
 {
@@ -47,5 +50,11 @@ static inline struct crypto_ablkcipher *crypto_spawn_skcipher(
 				 crypto_skcipher_mask(0)));
 }
 
+static inline void *skcipher_givcrypt_reqctx(
+	struct skcipher_givcrypt_request *req)
+{
+	return ablkcipher_request_ctx(&req->creq);
+}
+
 #endif	/* _CRYPTO_INTERNAL_SKCIPHER_H */
 
diff --git a/include/crypto/skcipher.h b/include/crypto/skcipher.h
new file mode 100644
index 000000000000..c283fab5eddb
--- /dev/null
+++ b/include/crypto/skcipher.h
@@ -0,0 +1,38 @@
+/*
+ * Symmetric key ciphers.
+ * 
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_SKCIPHER_H
+#define _CRYPTO_SKCIPHER_H
+
+#include <linux/crypto.h>
+
+/**
+ *	struct skcipher_givcrypt_request - Crypto request with IV generation
+ *	@seq: Sequence number for IV generation
+ *	@giv: Space for generated IV
+ *	@creq: The crypto request itself
+ */
+struct skcipher_givcrypt_request {
+	u64 seq;
+	u8 *giv;
+
+	struct ablkcipher_request creq;
+};
+
+static inline struct crypto_ablkcipher *skcipher_givcrypt_reqtfm(
+	struct skcipher_givcrypt_request *req)
+{
+	return crypto_ablkcipher_reqtfm(&req->creq);
+}
+
+#endif	/* _CRYPTO_SKCIPHER_H */
+
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index d6962b409489..3656a24ea7f0 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -34,6 +34,7 @@
 #define CRYPTO_ALG_TYPE_HASH		0x00000003
 #define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000004
 #define CRYPTO_ALG_TYPE_ABLKCIPHER	0x00000005
+#define CRYPTO_ALG_TYPE_GIVCIPHER	0x00000006
 #define CRYPTO_ALG_TYPE_COMPRESS	0x00000008
 #define CRYPTO_ALG_TYPE_AEAD		0x00000009
 
@@ -99,6 +100,7 @@ struct crypto_blkcipher;
 struct crypto_hash;
 struct crypto_tfm;
 struct crypto_type;
+struct skcipher_givcrypt_request;
 
 typedef void (*crypto_completion_t)(struct crypto_async_request *req, int err);
 
@@ -178,6 +180,8 @@ struct ablkcipher_alg {
 	              unsigned int keylen);
 	int (*encrypt)(struct ablkcipher_request *req);
 	int (*decrypt)(struct ablkcipher_request *req);
+	int (*givencrypt)(struct skcipher_givcrypt_request *req);
+	int (*givdecrypt)(struct skcipher_givcrypt_request *req);
 
 	unsigned int min_keysize;
 	unsigned int max_keysize;
@@ -320,6 +324,9 @@ struct ablkcipher_tfm {
 	              unsigned int keylen);
 	int (*encrypt)(struct ablkcipher_request *req);
 	int (*decrypt)(struct ablkcipher_request *req);
+	int (*givencrypt)(struct skcipher_givcrypt_request *req);
+	int (*givdecrypt)(struct skcipher_givcrypt_request *req);
+
 	unsigned int ivsize;
 	unsigned int reqsize;
 };
-- 
cgit v1.2.3


From 23508e11ab3bb405dca66bf4d77e488bf2b07b0c Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 27 Nov 2007 21:33:24 +0800
Subject: [CRYPTO] skcipher: Added geniv field

This patch introduces the geniv field which indicates the default IV
generator for each algorithm.  It should point to a string that is not
freed as long as the algorithm is registered.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ablkcipher.c    | 2 ++
 crypto/blkcipher.c     | 2 ++
 include/linux/crypto.h | 4 ++++
 3 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index e403d8137ecd..8f48b4d58fca 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -96,6 +96,7 @@ static void crypto_ablkcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	seq_printf(m, "min keysize  : %u\n", ablkcipher->min_keysize);
 	seq_printf(m, "max keysize  : %u\n", ablkcipher->max_keysize);
 	seq_printf(m, "ivsize       : %u\n", ablkcipher->ivsize);
+	seq_printf(m, "geniv        : %s\n", ablkcipher->geniv ?: "<default>");
 }
 
 const struct crypto_type crypto_ablkcipher_type = {
@@ -142,6 +143,7 @@ static void crypto_givcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	seq_printf(m, "min keysize  : %u\n", ablkcipher->min_keysize);
 	seq_printf(m, "max keysize  : %u\n", ablkcipher->max_keysize);
 	seq_printf(m, "ivsize       : %u\n", ablkcipher->ivsize);
+	seq_printf(m, "geniv        : %s\n", ablkcipher->geniv ?: "<built-in>");
 }
 
 const struct crypto_type crypto_givcipher_type = {
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index 7939504dfd8b..7a3e1a308074 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -496,6 +496,8 @@ static void crypto_blkcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	seq_printf(m, "min keysize  : %u\n", alg->cra_blkcipher.min_keysize);
 	seq_printf(m, "max keysize  : %u\n", alg->cra_blkcipher.max_keysize);
 	seq_printf(m, "ivsize       : %u\n", alg->cra_blkcipher.ivsize);
+	seq_printf(m, "geniv        : %s\n", alg->cra_blkcipher.geniv ?:
+					     "<default>");
 }
 
 const struct crypto_type crypto_blkcipher_type = {
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 3656a24ea7f0..facafa1bd8ca 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -183,6 +183,8 @@ struct ablkcipher_alg {
 	int (*givencrypt)(struct skcipher_givcrypt_request *req);
 	int (*givdecrypt)(struct skcipher_givcrypt_request *req);
 
+	const char *geniv;
+
 	unsigned int min_keysize;
 	unsigned int max_keysize;
 	unsigned int ivsize;
@@ -209,6 +211,8 @@ struct blkcipher_alg {
 		       struct scatterlist *dst, struct scatterlist *src,
 		       unsigned int nbytes);
 
+	const char *geniv;
+
 	unsigned int min_keysize;
 	unsigned int max_keysize;
 	unsigned int ivsize;
-- 
cgit v1.2.3


From ecfc43292f68566c144afca966b46b371c26d56c Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 5 Dec 2007 21:08:36 +1100
Subject: [CRYPTO] skcipher: Add skcipher_geniv_alloc/skcipher_geniv_free

This patch creates the infrastructure to help the construction of givcipher
templates that wrap around existing blkcipher/ablkcipher algorithms by adding
an IV generator to them.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ablkcipher.c                |  10 +-
 crypto/blkcipher.c                 | 185 ++++++++++++++++++++++++++++++++++++-
 include/crypto/internal/skcipher.h |  18 ++++
 include/linux/crypto.h             |  18 +++-
 4 files changed, 225 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index 8f48b4d58fca..092d9659b861 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -80,6 +80,7 @@ static int crypto_init_ablkcipher_ops(struct crypto_tfm *tfm, u32 type,
 	crt->setkey = setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
+	crt->base = __crypto_ablkcipher_cast(tfm);
 	crt->ivsize = alg->ivsize;
 
 	return 0;
@@ -122,11 +123,13 @@ static int crypto_init_givcipher_ops(struct crypto_tfm *tfm, u32 type,
 	if (alg->ivsize > PAGE_SIZE / 8)
 		return -EINVAL;
 
-	crt->setkey = setkey;
+	crt->setkey = tfm->__crt_alg->cra_flags & CRYPTO_ALG_GENIV ?
+		      alg->setkey : setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
 	crt->givencrypt = alg->givencrypt;
 	crt->givdecrypt = alg->givdecrypt ?: no_givdecrypt;
+	crt->base = __crypto_ablkcipher_cast(tfm);
 	crt->ivsize = alg->ivsize;
 
 	return 0;
@@ -155,6 +158,11 @@ const struct crypto_type crypto_givcipher_type = {
 };
 EXPORT_SYMBOL_GPL(crypto_givcipher_type);
 
+const char *crypto_default_geniv(const struct crypto_alg *alg)
+{
+	return alg->cra_flags & CRYPTO_ALG_ASYNC ? "eseqiv" : "chainiv";
+}
+
 int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
 			 u32 type, u32 mask)
 {
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index 7a3e1a308074..ca6ef065cded 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -14,8 +14,8 @@
  *
  */
 
+#include <crypto/internal/skcipher.h>
 #include <crypto/scatterwalk.h>
-#include <linux/crypto.h>
 #include <linux/errno.h>
 #include <linux/hardirq.h>
 #include <linux/kernel.h>
@@ -450,6 +450,7 @@ static int crypto_init_blkcipher_ops_async(struct crypto_tfm *tfm)
 	crt->setkey = async_setkey;
 	crt->encrypt = async_encrypt;
 	crt->decrypt = async_decrypt;
+	crt->base = __crypto_ablkcipher_cast(tfm);
 	crt->ivsize = alg->ivsize;
 
 	return 0;
@@ -509,5 +510,187 @@ const struct crypto_type crypto_blkcipher_type = {
 };
 EXPORT_SYMBOL_GPL(crypto_blkcipher_type);
 
+static int crypto_grab_nivcipher(struct crypto_skcipher_spawn *spawn,
+				const char *name, u32 type, u32 mask)
+{
+	struct crypto_alg *alg;
+	int err;
+
+	type = crypto_skcipher_type(type);
+	mask = crypto_skcipher_mask(mask) | CRYPTO_ALG_GENIV;
+
+	alg = crypto_alg_mod_lookup(name, type, mask);
+	if (IS_ERR(alg))
+		return PTR_ERR(alg);
+
+	err = crypto_init_spawn(&spawn->base, alg, spawn->base.inst, mask);
+	crypto_mod_put(alg);
+	return err;
+}
+
+struct crypto_instance *skcipher_geniv_alloc(struct crypto_template *tmpl,
+					     struct rtattr **tb, u32 type,
+					     u32 mask)
+{
+	struct {
+		int (*setkey)(struct crypto_ablkcipher *tfm, const u8 *key,
+			      unsigned int keylen);
+		int (*encrypt)(struct ablkcipher_request *req);
+		int (*decrypt)(struct ablkcipher_request *req);
+
+		unsigned int min_keysize;
+		unsigned int max_keysize;
+		unsigned int ivsize;
+
+		const char *geniv;
+	} balg;
+	const char *name;
+	struct crypto_skcipher_spawn *spawn;
+	struct crypto_attr_type *algt;
+	struct crypto_instance *inst;
+	struct crypto_alg *alg;
+	int err;
+
+	algt = crypto_get_attr_type(tb);
+	err = PTR_ERR(algt);
+	if (IS_ERR(algt))
+		return ERR_PTR(err);
+
+	if ((algt->type ^ (CRYPTO_ALG_TYPE_GIVCIPHER | CRYPTO_ALG_GENIV)) &
+	    algt->mask)
+		return ERR_PTR(-EINVAL);
+
+	name = crypto_attr_alg_name(tb[1]);
+	err = PTR_ERR(name);
+	if (IS_ERR(name))
+		return ERR_PTR(err);
+
+	inst = kzalloc(sizeof(*inst) + sizeof(*spawn), GFP_KERNEL);
+	if (!inst)
+		return ERR_PTR(-ENOMEM);
+
+	spawn = crypto_instance_ctx(inst);
+
+	/* Ignore async algorithms if necessary. */
+	mask |= crypto_requires_sync(algt->type, algt->mask);
+
+	crypto_set_skcipher_spawn(spawn, inst);
+	err = crypto_grab_nivcipher(spawn, name, type, mask);
+	if (err)
+		goto err_free_inst;
+
+	alg = crypto_skcipher_spawn_alg(spawn);
+
+	if ((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
+	    CRYPTO_ALG_TYPE_BLKCIPHER) {
+		balg.ivsize = alg->cra_blkcipher.ivsize;
+		balg.min_keysize = alg->cra_blkcipher.min_keysize;
+		balg.max_keysize = alg->cra_blkcipher.max_keysize;
+
+		balg.setkey = async_setkey;
+		balg.encrypt = async_encrypt;
+		balg.decrypt = async_decrypt;
+
+		balg.geniv = alg->cra_blkcipher.geniv;
+	} else {
+		balg.ivsize = alg->cra_ablkcipher.ivsize;
+		balg.min_keysize = alg->cra_ablkcipher.min_keysize;
+		balg.max_keysize = alg->cra_ablkcipher.max_keysize;
+
+		balg.setkey = alg->cra_ablkcipher.setkey;
+		balg.encrypt = alg->cra_ablkcipher.encrypt;
+		balg.decrypt = alg->cra_ablkcipher.decrypt;
+
+		balg.geniv = alg->cra_ablkcipher.geniv;
+	}
+
+	err = -EINVAL;
+	if (!balg.ivsize)
+		goto err_drop_alg;
+
+	/*
+	 * This is only true if we're constructing an algorithm with its
+	 * default IV generator.  For the default generator we elide the
+	 * template name and double-check the IV generator.
+	 */
+	if (algt->mask & CRYPTO_ALG_GENIV) {
+		if (!balg.geniv)
+			balg.geniv = crypto_default_geniv(alg);
+		err = -EAGAIN;
+		if (strcmp(tmpl->name, balg.geniv))
+			goto err_drop_alg;
+
+		memcpy(inst->alg.cra_name, alg->cra_name, CRYPTO_MAX_ALG_NAME);
+		memcpy(inst->alg.cra_driver_name, alg->cra_driver_name,
+		       CRYPTO_MAX_ALG_NAME);
+	} else {
+		err = -ENAMETOOLONG;
+		if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
+			     "%s(%s)", tmpl->name, alg->cra_name) >=
+		    CRYPTO_MAX_ALG_NAME)
+			goto err_drop_alg;
+		if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
+			     "%s(%s)", tmpl->name, alg->cra_driver_name) >=
+		    CRYPTO_MAX_ALG_NAME)
+			goto err_drop_alg;
+	}
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_GIVCIPHER | CRYPTO_ALG_GENIV;
+	inst->alg.cra_flags |= alg->cra_flags & CRYPTO_ALG_ASYNC;
+	inst->alg.cra_priority = alg->cra_priority;
+	inst->alg.cra_blocksize = alg->cra_blocksize;
+	inst->alg.cra_alignmask = alg->cra_alignmask;
+	inst->alg.cra_type = &crypto_givcipher_type;
+
+	inst->alg.cra_ablkcipher.ivsize = balg.ivsize;
+	inst->alg.cra_ablkcipher.min_keysize = balg.min_keysize;
+	inst->alg.cra_ablkcipher.max_keysize = balg.max_keysize;
+	inst->alg.cra_ablkcipher.geniv = balg.geniv;
+
+	inst->alg.cra_ablkcipher.setkey = balg.setkey;
+	inst->alg.cra_ablkcipher.encrypt = balg.encrypt;
+	inst->alg.cra_ablkcipher.decrypt = balg.decrypt;
+
+out:
+	return inst;
+
+err_drop_alg:
+	crypto_drop_skcipher(spawn);
+err_free_inst:
+	kfree(inst);
+	inst = ERR_PTR(err);
+	goto out;
+}
+EXPORT_SYMBOL_GPL(skcipher_geniv_alloc);
+
+void skcipher_geniv_free(struct crypto_instance *inst)
+{
+	crypto_drop_skcipher(crypto_instance_ctx(inst));
+	kfree(inst);
+}
+EXPORT_SYMBOL_GPL(skcipher_geniv_free);
+
+int skcipher_geniv_init(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = (void *)tfm->__crt_alg;
+	struct crypto_ablkcipher *cipher;
+
+	cipher = crypto_spawn_skcipher(crypto_instance_ctx(inst));
+	if (IS_ERR(cipher))
+		return PTR_ERR(cipher);
+
+	tfm->crt_ablkcipher.base = cipher;
+	tfm->crt_ablkcipher.reqsize += crypto_ablkcipher_reqsize(cipher);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(skcipher_geniv_init);
+
+void skcipher_geniv_exit(struct crypto_tfm *tfm)
+{
+	crypto_free_ablkcipher(tfm->crt_ablkcipher.base);
+}
+EXPORT_SYMBOL_GPL(skcipher_geniv_exit);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Generic block chaining cipher type");
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index c9402dd12d03..07e7c82324ad 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -15,6 +15,9 @@
 
 #include <crypto/algapi.h>
 #include <crypto/skcipher.h>
+#include <linux/types.h>
+
+struct rtattr;
 
 struct crypto_skcipher_spawn {
 	struct crypto_spawn base;
@@ -50,6 +53,21 @@ static inline struct crypto_ablkcipher *crypto_spawn_skcipher(
 				 crypto_skcipher_mask(0)));
 }
 
+const char *crypto_default_geniv(const struct crypto_alg *alg);
+
+struct crypto_instance *skcipher_geniv_alloc(struct crypto_template *tmpl,
+					     struct rtattr **tb, u32 type,
+					     u32 mask);
+void skcipher_geniv_free(struct crypto_instance *inst);
+int skcipher_geniv_init(struct crypto_tfm *tfm);
+void skcipher_geniv_exit(struct crypto_tfm *tfm);
+
+static inline struct crypto_ablkcipher *skcipher_geniv_cipher(
+	struct crypto_ablkcipher *geniv)
+{
+	return crypto_ablkcipher_crt(geniv)->base;
+}
+
 static inline void *skcipher_givcrypt_reqctx(
 	struct skcipher_givcrypt_request *req)
 {
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index facafa1bd8ca..fa7afa9b9f4f 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -52,6 +52,12 @@
  */
 #define CRYPTO_ALG_NEED_FALLBACK	0x00000100
 
+/*
+ * This bit is set for symmetric key ciphers that have already been wrapped
+ * with a generic IV generator to prevent them from being wrapped again.
+ */
+#define CRYPTO_ALG_GENIV		0x00000200
+
 /*
  * Transform masks and values (for crt_flags).
  */
@@ -331,6 +337,8 @@ struct ablkcipher_tfm {
 	int (*givencrypt)(struct skcipher_givcrypt_request *req);
 	int (*givdecrypt)(struct skcipher_givcrypt_request *req);
 
+	struct crypto_ablkcipher *base;
+
 	unsigned int ivsize;
 	unsigned int reqsize;
 };
@@ -541,14 +549,14 @@ static inline struct crypto_ablkcipher *__crypto_ablkcipher_cast(
 
 static inline u32 crypto_skcipher_type(u32 type)
 {
-	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
 	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
 	return type;
 }
 
 static inline u32 crypto_skcipher_mask(u32 mask)
 {
-	mask &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
 	mask |= CRYPTO_ALG_TYPE_BLKCIPHER_MASK;
 	return mask;
 }
@@ -623,7 +631,9 @@ static inline void crypto_ablkcipher_clear_flags(struct crypto_ablkcipher *tfm,
 static inline int crypto_ablkcipher_setkey(struct crypto_ablkcipher *tfm,
 					   const u8 *key, unsigned int keylen)
 {
-	return crypto_ablkcipher_crt(tfm)->setkey(tfm, key, keylen);
+	struct ablkcipher_tfm *crt = crypto_ablkcipher_crt(tfm);
+
+	return crt->setkey(crt->base, key, keylen);
 }
 
 static inline struct crypto_ablkcipher *crypto_ablkcipher_reqtfm(
@@ -655,7 +665,7 @@ static inline unsigned int crypto_ablkcipher_reqsize(
 static inline void ablkcipher_request_set_tfm(
 	struct ablkcipher_request *req, struct crypto_ablkcipher *tfm)
 {
-	req->base.tfm = crypto_ablkcipher_tfm(tfm);
+	req->base.tfm = crypto_ablkcipher_tfm(crypto_ablkcipher_crt(tfm)->base);
 }
 
 static inline struct ablkcipher_request *ablkcipher_request_cast(
-- 
cgit v1.2.3


From 15c67286685cddce207b646306e8819ec8268ede Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 30 Nov 2007 20:17:28 +1100
Subject: [CRYPTO] skcipher: Added skcipher_givcrypt_complete

This patch adds the helper skcipher_givcrypt_complete which should be
called when an ablkcipher algorithm has completed a givcrypt request.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/crypto/internal/skcipher.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index 07e7c82324ad..80c5bfb14a60 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -74,5 +74,17 @@ static inline void *skcipher_givcrypt_reqctx(
 	return ablkcipher_request_ctx(&req->creq);
 }
 
+static inline void ablkcipher_request_complete(struct ablkcipher_request *req,
+					       int err)
+{
+	req->base.complete(&req->base, err);
+}
+
+static inline void skcipher_givcrypt_complete(
+	struct skcipher_givcrypt_request *req, int err)
+{
+	ablkcipher_request_complete(&req->creq, err);
+}
+
 #endif	/* _CRYPTO_INTERNAL_SKCIPHER_H */
 
-- 
cgit v1.2.3


From b9c55aa475599183d0eab6833ea23e70c52dd24b Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 4 Dec 2007 12:46:48 +1100
Subject: [CRYPTO] skcipher: Create default givcipher instances

This patch makes crypto_alloc_ablkcipher/crypto_grab_skcipher always
return algorithms that are capable of generating their own IVs through
givencrypt and givdecrypt.  Each algorithm may specify its default IV
generator through the geniv field.

For algorithms that do not set the geniv field, the blkcipher layer will
pick a default.  Currently it's chainiv for synchronous algorithms and
eseqiv for asynchronous algorithms.  Note that if these wrappers do not
work on an algorithm then that algorithm must specify its own geniv or
it can't be used at all.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ablkcipher.c                | 158 ++++++++++++++++++++++++++++++++++++-
 crypto/api.c                       |  19 +++--
 crypto/blkcipher.c                 |   4 +
 crypto/internal.h                  |   2 +
 include/crypto/internal/skcipher.h |   2 +
 include/linux/crypto.h             |   9 +--
 6 files changed, 181 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index 092d9659b861..d1df528c8fa3 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -18,6 +18,8 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/rtnetlink.h>
+#include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
 
@@ -68,6 +70,16 @@ static unsigned int crypto_ablkcipher_ctxsize(struct crypto_alg *alg, u32 type,
 	return alg->cra_ctxsize;
 }
 
+int skcipher_null_givencrypt(struct skcipher_givcrypt_request *req)
+{
+	return crypto_ablkcipher_encrypt(&req->creq);
+}
+
+int skcipher_null_givdecrypt(struct skcipher_givcrypt_request *req)
+{
+	return crypto_ablkcipher_decrypt(&req->creq);
+}
+
 static int crypto_init_ablkcipher_ops(struct crypto_tfm *tfm, u32 type,
 				      u32 mask)
 {
@@ -80,6 +92,10 @@ static int crypto_init_ablkcipher_ops(struct crypto_tfm *tfm, u32 type,
 	crt->setkey = setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
+	if (!alg->ivsize) {
+		crt->givencrypt = skcipher_null_givencrypt;
+		crt->givdecrypt = skcipher_null_givdecrypt;
+	}
 	crt->base = __crypto_ablkcipher_cast(tfm);
 	crt->ivsize = alg->ivsize;
 
@@ -163,6 +179,108 @@ const char *crypto_default_geniv(const struct crypto_alg *alg)
 	return alg->cra_flags & CRYPTO_ALG_ASYNC ? "eseqiv" : "chainiv";
 }
 
+static int crypto_givcipher_default(struct crypto_alg *alg, u32 type, u32 mask)
+{
+	struct rtattr *tb[3];
+	struct {
+		struct rtattr attr;
+		struct crypto_attr_type data;
+	} ptype;
+	struct {
+		struct rtattr attr;
+		struct crypto_attr_alg data;
+	} palg;
+	struct crypto_template *tmpl;
+	struct crypto_instance *inst;
+	struct crypto_alg *larval;
+	const char *geniv;
+	int err;
+
+	larval = crypto_larval_lookup(alg->cra_driver_name,
+				      CRYPTO_ALG_TYPE_GIVCIPHER,
+				      CRYPTO_ALG_TYPE_MASK);
+	err = PTR_ERR(larval);
+	if (IS_ERR(larval))
+		goto out;
+
+	err = -EAGAIN;
+	if (!crypto_is_larval(larval))
+		goto drop_larval;
+
+	ptype.attr.rta_len = sizeof(ptype);
+	ptype.attr.rta_type = CRYPTOA_TYPE;
+	ptype.data.type = type | CRYPTO_ALG_GENIV;
+	/* GENIV tells the template that we're making a default geniv. */
+	ptype.data.mask = mask | CRYPTO_ALG_GENIV;
+	tb[0] = &ptype.attr;
+
+	palg.attr.rta_len = sizeof(palg);
+	palg.attr.rta_type = CRYPTOA_ALG;
+	/* Must use the exact name to locate ourselves. */
+	memcpy(palg.data.name, alg->cra_driver_name, CRYPTO_MAX_ALG_NAME);
+	tb[1] = &palg.attr;
+
+	tb[2] = NULL;
+
+	if ((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
+	    CRYPTO_ALG_TYPE_BLKCIPHER)
+		geniv = alg->cra_blkcipher.geniv;
+	else
+		geniv = alg->cra_ablkcipher.geniv;
+
+	if (!geniv)
+		geniv = crypto_default_geniv(alg);
+
+	tmpl = crypto_lookup_template(geniv);
+	err = -ENOENT;
+	if (!tmpl)
+		goto kill_larval;
+
+	inst = tmpl->alloc(tb);
+	err = PTR_ERR(inst);
+	if (IS_ERR(inst))
+		goto put_tmpl;
+
+	if ((err = crypto_register_instance(tmpl, inst))) {
+		tmpl->free(inst);
+		goto put_tmpl;
+	}
+
+	/* Redo the lookup to use the instance we just registered. */
+	err = -EAGAIN;
+
+put_tmpl:
+	crypto_tmpl_put(tmpl);
+kill_larval:
+	crypto_larval_kill(larval);
+drop_larval:
+	crypto_mod_put(larval);
+out:
+	crypto_mod_put(alg);
+	return err;
+}
+
+static struct crypto_alg *crypto_lookup_skcipher(const char *name, u32 type,
+						 u32 mask)
+{
+	struct crypto_alg *alg;
+
+	alg = crypto_alg_mod_lookup(name, type, mask);
+	if (IS_ERR(alg))
+		return alg;
+
+	if ((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
+	    CRYPTO_ALG_TYPE_GIVCIPHER)
+		return alg;
+
+	if (!((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
+	      CRYPTO_ALG_TYPE_BLKCIPHER ? alg->cra_blkcipher.ivsize :
+					  alg->cra_ablkcipher.ivsize))
+		return alg;
+
+	return ERR_PTR(crypto_givcipher_default(alg, type, mask));
+}
+
 int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
 			 u32 type, u32 mask)
 {
@@ -172,7 +290,7 @@ int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
 	type = crypto_skcipher_type(type);
 	mask = crypto_skcipher_mask(mask);
 
-	alg = crypto_alg_mod_lookup(name, type, mask);
+	alg = crypto_lookup_skcipher(name, type, mask);
 	if (IS_ERR(alg))
 		return PTR_ERR(alg);
 
@@ -182,5 +300,43 @@ int crypto_grab_skcipher(struct crypto_skcipher_spawn *spawn, const char *name,
 }
 EXPORT_SYMBOL_GPL(crypto_grab_skcipher);
 
+struct crypto_ablkcipher *crypto_alloc_ablkcipher(const char *alg_name,
+						  u32 type, u32 mask)
+{
+	struct crypto_tfm *tfm;
+	int err;
+
+	type = crypto_skcipher_type(type);
+	mask = crypto_skcipher_mask(mask);
+
+	for (;;) {
+		struct crypto_alg *alg;
+
+		alg = crypto_lookup_skcipher(alg_name, type, mask);
+		if (IS_ERR(alg)) {
+			err = PTR_ERR(alg);
+			goto err;
+		}
+
+		tfm = __crypto_alloc_tfm(alg, type, mask);
+		if (!IS_ERR(tfm))
+			return __crypto_ablkcipher_cast(tfm);
+
+		crypto_mod_put(alg);
+		err = PTR_ERR(tfm);
+
+err:
+		if (err != -EAGAIN)
+			break;
+		if (signal_pending(current)) {
+			err = -EINTR;
+			break;
+		}
+	}
+
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(crypto_alloc_ablkcipher);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Asynchronous block chaining cipher type");
diff --git a/crypto/api.c b/crypto/api.c
index 1f5c72477356..a2496d1bc6d4 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -137,7 +137,7 @@ static struct crypto_alg *crypto_larval_alloc(const char *name, u32 type,
 	return alg;
 }
 
-static void crypto_larval_kill(struct crypto_alg *alg)
+void crypto_larval_kill(struct crypto_alg *alg)
 {
 	struct crypto_larval *larval = (void *)alg;
 
@@ -147,6 +147,7 @@ static void crypto_larval_kill(struct crypto_alg *alg)
 	complete_all(&larval->completion);
 	crypto_alg_put(alg);
 }
+EXPORT_SYMBOL_GPL(crypto_larval_kill);
 
 static struct crypto_alg *crypto_larval_wait(struct crypto_alg *alg)
 {
@@ -176,11 +177,9 @@ static struct crypto_alg *crypto_alg_lookup(const char *name, u32 type,
 	return alg;
 }
 
-struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
+struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask)
 {
 	struct crypto_alg *alg;
-	struct crypto_alg *larval;
-	int ok;
 
 	if (!name)
 		return ERR_PTR(-ENOENT);
@@ -193,7 +192,17 @@ struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
 	if (alg)
 		return crypto_is_larval(alg) ? crypto_larval_wait(alg) : alg;
 
-	larval = crypto_larval_alloc(name, type, mask);
+	return crypto_larval_alloc(name, type, mask);
+}
+EXPORT_SYMBOL_GPL(crypto_larval_lookup);
+
+struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
+{
+	struct crypto_alg *alg;
+	struct crypto_alg *larval;
+	int ok;
+
+	larval = crypto_larval_lookup(name, type, mask);
 	if (IS_ERR(larval) || !crypto_is_larval(larval))
 		return larval;
 
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index ca6ef065cded..4a7e65c4df4d 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -450,6 +450,10 @@ static int crypto_init_blkcipher_ops_async(struct crypto_tfm *tfm)
 	crt->setkey = async_setkey;
 	crt->encrypt = async_encrypt;
 	crt->decrypt = async_decrypt;
+	if (!alg->ivsize) {
+		crt->givencrypt = skcipher_null_givencrypt;
+		crt->givdecrypt = skcipher_null_givdecrypt;
+	}
 	crt->base = __crypto_ablkcipher_cast(tfm);
 	crt->ivsize = alg->ivsize;
 
diff --git a/crypto/internal.h b/crypto/internal.h
index cb13952f82bf..32f4c2145603 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -93,6 +93,8 @@ void crypto_exit_digest_ops(struct crypto_tfm *tfm);
 void crypto_exit_cipher_ops(struct crypto_tfm *tfm);
 void crypto_exit_compress_ops(struct crypto_tfm *tfm);
 
+void crypto_larval_kill(struct crypto_alg *alg);
+struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask);
 void crypto_larval_error(const char *name, u32 type, u32 mask);
 
 void crypto_shoot_alg(struct crypto_alg *alg);
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index 80c5bfb14a60..2071999d4b5e 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -53,6 +53,8 @@ static inline struct crypto_ablkcipher *crypto_spawn_skcipher(
 				 crypto_skcipher_mask(0)));
 }
 
+int skcipher_null_givencrypt(struct skcipher_givcrypt_request *req);
+int skcipher_null_givdecrypt(struct skcipher_givcrypt_request *req);
 const char *crypto_default_geniv(const struct crypto_alg *alg);
 
 struct crypto_instance *skcipher_geniv_alloc(struct crypto_template *tmpl,
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index fa7afa9b9f4f..835dcaf3fe4e 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -561,13 +561,8 @@ static inline u32 crypto_skcipher_mask(u32 mask)
 	return mask;
 }
 
-static inline struct crypto_ablkcipher *crypto_alloc_ablkcipher(
-	const char *alg_name, u32 type, u32 mask)
-{
-	return __crypto_ablkcipher_cast(
-		crypto_alloc_base(alg_name, crypto_skcipher_type(type),
-				  crypto_skcipher_mask(mask)));
-}
+struct crypto_ablkcipher *crypto_alloc_ablkcipher(const char *alg_name,
+						  u32 type, u32 mask);
 
 static inline struct crypto_tfm *crypto_ablkcipher_tfm(
 	struct crypto_ablkcipher *tfm)
-- 
cgit v1.2.3


From 45d44eb56ad197cfccb8f84b5df64abff8b7cb96 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 2 Dec 2007 21:21:02 +1100
Subject: [CRYPTO] skcipher: Remove crypto_spawn_ablkcipher

Now that gcm and authenc have been converted to crypto_spawn_skcipher,
this patch removes the obsolete crypto_spawn_ablkcipher function.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/crypto/algapi.h | 8 --------
 1 file changed, 8 deletions(-)

(limited to 'include')

diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index fda1759ffe2d..60d06e784be3 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -198,14 +198,6 @@ static inline struct crypto_instance *crypto_aead_alg_instance(
 	return crypto_tfm_alg_instance(&aead->base);
 }
 
-static inline struct crypto_ablkcipher *crypto_spawn_ablkcipher(
-	struct crypto_spawn *spawn)
-{
-	return __crypto_ablkcipher_cast(
-		crypto_spawn_tfm(spawn, crypto_skcipher_type(0),
-				 crypto_skcipher_mask(0)));
-}
-
 static inline struct crypto_blkcipher *crypto_spawn_blkcipher(
 	struct crypto_spawn *spawn)
 {
-- 
cgit v1.2.3


From 03bf712fb4defc7831c727d1e32d0269f7f96de0 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 1 Dec 2007 18:35:38 +1100
Subject: [CRYPTO] skcipher: Add top-level givencrypt/givdecrypt calls

This patch finally makes the givencrypt/givdecrypt operations available
to users by adding crypto_skcipher_givencrypt and crypto_skcipher_givdecrypt.
A suite of helpers to allocate and fill in the request is also available.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/crypto/skcipher.h | 72 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 72 insertions(+)

(limited to 'include')

diff --git a/include/crypto/skcipher.h b/include/crypto/skcipher.h
index c283fab5eddb..25fd6126522d 100644
--- a/include/crypto/skcipher.h
+++ b/include/crypto/skcipher.h
@@ -14,6 +14,8 @@
 #define _CRYPTO_SKCIPHER_H
 
 #include <linux/crypto.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
 
 /**
  *	struct skcipher_givcrypt_request - Crypto request with IV generation
@@ -34,5 +36,75 @@ static inline struct crypto_ablkcipher *skcipher_givcrypt_reqtfm(
 	return crypto_ablkcipher_reqtfm(&req->creq);
 }
 
+static inline int crypto_skcipher_givencrypt(
+	struct skcipher_givcrypt_request *req)
+{
+	struct ablkcipher_tfm *crt =
+		crypto_ablkcipher_crt(skcipher_givcrypt_reqtfm(req));
+	return crt->givencrypt(req);
+};
+
+static inline int crypto_skcipher_givdecrypt(
+	struct skcipher_givcrypt_request *req)
+{
+	struct ablkcipher_tfm *crt =
+		crypto_ablkcipher_crt(skcipher_givcrypt_reqtfm(req));
+	return crt->givdecrypt(req);
+};
+
+static inline void skcipher_givcrypt_set_tfm(
+	struct skcipher_givcrypt_request *req, struct crypto_ablkcipher *tfm)
+{
+	req->creq.base.tfm = crypto_ablkcipher_tfm(tfm);
+}
+
+static inline struct skcipher_givcrypt_request *skcipher_givcrypt_cast(
+	struct crypto_async_request *req)
+{
+	return container_of(ablkcipher_request_cast(req),
+			    struct skcipher_givcrypt_request, creq);
+}
+
+static inline struct skcipher_givcrypt_request *skcipher_givcrypt_alloc(
+	struct crypto_ablkcipher *tfm, gfp_t gfp)
+{
+	struct skcipher_givcrypt_request *req;
+
+	req = kmalloc(sizeof(struct skcipher_givcrypt_request) +
+		      crypto_ablkcipher_reqsize(tfm), gfp);
+
+	if (likely(req))
+		skcipher_givcrypt_set_tfm(req, tfm);
+
+	return req;
+}
+
+static inline void skcipher_givcrypt_free(struct skcipher_givcrypt_request *req)
+{
+	kfree(req);
+}
+
+static inline void skcipher_givcrypt_set_callback(
+	struct skcipher_givcrypt_request *req, u32 flags,
+	crypto_completion_t complete, void *data)
+{
+	ablkcipher_request_set_callback(&req->creq, flags, complete, data);
+}
+
+static inline void skcipher_givcrypt_set_crypt(
+	struct skcipher_givcrypt_request *req,
+	struct scatterlist *src, struct scatterlist *dst,
+	unsigned int nbytes, void *iv)
+{
+	ablkcipher_request_set_crypt(&req->creq, src, dst, nbytes, iv);
+}
+
+static inline void skcipher_givcrypt_set_giv(
+	struct skcipher_givcrypt_request *req, u8 *giv, u64 seq)
+{
+	req->giv = giv;
+	req->seq = seq;
+}
+
 #endif	/* _CRYPTO_SKCIPHER_H */
 
-- 
cgit v1.2.3


From 743edf57272fd420348e148bf94f9e48ed6abb70 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 10 Dec 2007 16:18:01 +0800
Subject: [CRYPTO] aead: Add givcrypt operations

This patch adds the underlying givcrypt operations for aead and associated
support elements.  The rationale is identical to that of the skcipher
givcrypt operations, i.e., sometimes only the algorithm knows how the
IV should be generated.

A new request type aead_givcrypt_request is added which contains an
embedded aead_request structure with two new elements to support this
operation.  The new elements are seq and giv.  The seq field should
contain a strictly increasing 64-bit integer which may be used by
certain IV generators as an input value.  The giv field will be used
to store the generated IV.  It does not need to obey the alignment
requirements of the algorithm because it's not used during the operation.

The existing iv field must still be available as it will be used to store
intermediate IVs and the output IV if chaining is desired.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/aead.c          |  7 +++++++
 include/crypto/aead.h  | 38 ++++++++++++++++++++++++++++++++++++++
 include/linux/crypto.h |  5 +++++
 3 files changed, 50 insertions(+)
 create mode 100644 include/crypto/aead.h

(limited to 'include')

diff --git a/crypto/aead.c b/crypto/aead.c
index f23c2b0ee009..0402b606fcfd 100644
--- a/crypto/aead.c
+++ b/crypto/aead.c
@@ -77,6 +77,11 @@ static unsigned int crypto_aead_ctxsize(struct crypto_alg *alg, u32 type,
 	return alg->cra_ctxsize;
 }
 
+static int no_givdecrypt(struct aead_givcrypt_request *req)
+{
+	return -ENOSYS;
+}
+
 static int crypto_init_aead_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 {
 	struct aead_alg *alg = &tfm->__crt_alg->cra_aead;
@@ -88,6 +93,8 @@ static int crypto_init_aead_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 	crt->setkey = setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
+	crt->givencrypt = alg->givencrypt;
+	crt->givdecrypt = alg->givdecrypt ?: no_givdecrypt;
 	crt->ivsize = alg->ivsize;
 	crt->authsize = alg->maxauthsize;
 
diff --git a/include/crypto/aead.h b/include/crypto/aead.h
new file mode 100644
index 000000000000..083920312da0
--- /dev/null
+++ b/include/crypto/aead.h
@@ -0,0 +1,38 @@
+/*
+ * AEAD: Authenticated Encryption with Associated Data
+ * 
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_AEAD_H
+#define _CRYPTO_AEAD_H
+
+#include <linux/crypto.h>
+#include <linux/kernel.h>
+
+/**
+ *	struct aead_givcrypt_request - AEAD request with IV generation
+ *	@seq: Sequence number for IV generation
+ *	@giv: Space for generated IV
+ *	@areq: The AEAD request itself
+ */
+struct aead_givcrypt_request {
+	u64 seq;
+	u8 *giv;
+
+	struct aead_request areq;
+};
+
+static inline struct crypto_aead *aead_givcrypt_reqtfm(
+	struct aead_givcrypt_request *req)
+{
+	return crypto_aead_reqtfm(&req->areq);
+}
+
+#endif	/* _CRYPTO_AEAD_H */
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 835dcaf3fe4e..7524928bff93 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -106,6 +106,7 @@ struct crypto_blkcipher;
 struct crypto_hash;
 struct crypto_tfm;
 struct crypto_type;
+struct aead_givcrypt_request;
 struct skcipher_givcrypt_request;
 
 typedef void (*crypto_completion_t)(struct crypto_async_request *req, int err);
@@ -202,6 +203,8 @@ struct aead_alg {
 	int (*setauthsize)(struct crypto_aead *tfm, unsigned int authsize);
 	int (*encrypt)(struct aead_request *req);
 	int (*decrypt)(struct aead_request *req);
+	int (*givencrypt)(struct aead_givcrypt_request *req);
+	int (*givdecrypt)(struct aead_givcrypt_request *req);
 
 	unsigned int ivsize;
 	unsigned int maxauthsize;
@@ -348,6 +351,8 @@ struct aead_tfm {
 	              unsigned int keylen);
 	int (*encrypt)(struct aead_request *req);
 	int (*decrypt)(struct aead_request *req);
+	int (*givencrypt)(struct aead_givcrypt_request *req);
+	int (*givdecrypt)(struct aead_givcrypt_request *req);
 	unsigned int ivsize;
 	unsigned int authsize;
 	unsigned int reqsize;
-- 
cgit v1.2.3


From 3a282bd2e77966e7361fffbd5d1cea6eb0499b6c Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 8 Dec 2007 20:13:15 +0800
Subject: [CRYPTO] aead: Add top-level givencrypt/givdecrypt calls

This patch finally makes the givencrypt/givdecrypt operations available
to users by adding crypto_aead_givencrypt and crypto_aead_givdecrypt.
A suite of helpers to allocate and fill in the request is also available.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/crypto/aead.h | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 67 insertions(+)

(limited to 'include')

diff --git a/include/crypto/aead.h b/include/crypto/aead.h
index 083920312da0..0edf949f6369 100644
--- a/include/crypto/aead.h
+++ b/include/crypto/aead.h
@@ -15,6 +15,7 @@
 
 #include <linux/crypto.h>
 #include <linux/kernel.h>
+#include <linux/slab.h>
 
 /**
  *	struct aead_givcrypt_request - AEAD request with IV generation
@@ -35,4 +36,70 @@ static inline struct crypto_aead *aead_givcrypt_reqtfm(
 	return crypto_aead_reqtfm(&req->areq);
 }
 
+static inline int crypto_aead_givencrypt(struct aead_givcrypt_request *req)
+{
+	struct aead_tfm *crt = crypto_aead_crt(aead_givcrypt_reqtfm(req));
+	return crt->givencrypt(req);
+};
+
+static inline int crypto_aead_givdecrypt(struct aead_givcrypt_request *req)
+{
+	struct aead_tfm *crt = crypto_aead_crt(aead_givcrypt_reqtfm(req));
+	return crt->givdecrypt(req);
+};
+
+static inline void aead_givcrypt_set_tfm(struct aead_givcrypt_request *req,
+					 struct crypto_aead *tfm)
+{
+	req->areq.base.tfm = crypto_aead_tfm(tfm);
+}
+
+static inline struct aead_givcrypt_request *aead_givcrypt_alloc(
+	struct crypto_aead *tfm, gfp_t gfp)
+{
+	struct aead_givcrypt_request *req;
+
+	req = kmalloc(sizeof(struct aead_givcrypt_request) +
+		      crypto_aead_reqsize(tfm), gfp);
+
+	if (likely(req))
+		aead_givcrypt_set_tfm(req, tfm);
+
+	return req;
+}
+
+static inline void aead_givcrypt_free(struct aead_givcrypt_request *req)
+{
+	kfree(req);
+}
+
+static inline void aead_givcrypt_set_callback(
+	struct aead_givcrypt_request *req, u32 flags,
+	crypto_completion_t complete, void *data)
+{
+	aead_request_set_callback(&req->areq, flags, complete, data);
+}
+
+static inline void aead_givcrypt_set_crypt(struct aead_givcrypt_request *req,
+					   struct scatterlist *src,
+					   struct scatterlist *dst,
+					   unsigned int nbytes, void *iv)
+{
+	aead_request_set_crypt(&req->areq, src, dst, nbytes, iv);
+}
+
+static inline void aead_givcrypt_set_assoc(struct aead_givcrypt_request *req,
+					   struct scatterlist *assoc,
+					   unsigned int assoclen)
+{
+	aead_request_set_assoc(&req->areq, assoc, assoclen);
+}
+
+static inline void aead_givcrypt_set_giv(struct aead_givcrypt_request *req,
+					 u8 *giv, u64 seq)
+{
+	req->giv = giv;
+	req->seq = seq;
+}
+
 #endif	/* _CRYPTO_AEAD_H */
-- 
cgit v1.2.3


From 5b6d2d7fdf806f2b5a9352416f9e670911fc4748 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 12 Dec 2007 19:23:36 +0800
Subject: [CRYPTO] aead: Add aead_geniv_alloc/aead_geniv_free

This patch creates the infrastructure to help the construction of IV
generator templates that wrap around AEAD algorithms by adding an IV
generator to them.  This is useful for AEAD algorithms with no built-in
IV generator or to replace their built-in generator.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/aead.c                  | 216 ++++++++++++++++++++++++++++++++++++++++-
 include/crypto/internal/aead.h |  77 +++++++++++++++
 include/linux/crypto.h         |  11 ++-
 3 files changed, 297 insertions(+), 7 deletions(-)
 create mode 100644 include/crypto/internal/aead.h

(limited to 'include')

diff --git a/crypto/aead.c b/crypto/aead.c
index 15335ed9010a..9f7aca891926 100644
--- a/crypto/aead.c
+++ b/crypto/aead.c
@@ -12,14 +12,16 @@
  *
  */
 
-#include <crypto/algapi.h>
-#include <linux/errno.h>
+#include <crypto/internal/aead.h>
+#include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
 
+#include "internal.h"
+
 static int setkey_unaligned(struct crypto_aead *tfm, const u8 *key,
 			    unsigned int keylen)
 {
@@ -55,18 +57,20 @@ static int setkey(struct crypto_aead *tfm, const u8 *key, unsigned int keylen)
 
 int crypto_aead_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
 {
+	struct aead_tfm *crt = crypto_aead_crt(tfm);
 	int err;
 
 	if (authsize > crypto_aead_alg(tfm)->maxauthsize)
 		return -EINVAL;
 
 	if (crypto_aead_alg(tfm)->setauthsize) {
-		err = crypto_aead_alg(tfm)->setauthsize(tfm, authsize);
+		err = crypto_aead_alg(tfm)->setauthsize(crt->base, authsize);
 		if (err)
 			return err;
 	}
 
-	crypto_aead_crt(tfm)->authsize = authsize;
+	crypto_aead_crt(crt->base)->authsize = authsize;
+	crt->authsize = authsize;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(crypto_aead_setauthsize);
@@ -90,11 +94,13 @@ static int crypto_init_aead_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 	if (max(alg->maxauthsize, alg->ivsize) > PAGE_SIZE / 8)
 		return -EINVAL;
 
-	crt->setkey = setkey;
+	crt->setkey = tfm->__crt_alg->cra_flags & CRYPTO_ALG_GENIV ?
+		      alg->setkey : setkey;
 	crt->encrypt = alg->encrypt;
 	crt->decrypt = alg->decrypt;
 	crt->givencrypt = alg->givencrypt ?: no_givcrypt;
 	crt->givdecrypt = alg->givdecrypt ?: no_givcrypt;
+	crt->base = __crypto_aead_cast(tfm);
 	crt->ivsize = alg->ivsize;
 	crt->authsize = alg->maxauthsize;
 
@@ -111,6 +117,7 @@ static void crypto_aead_show(struct seq_file *m, struct crypto_alg *alg)
 	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
 	seq_printf(m, "ivsize       : %u\n", aead->ivsize);
 	seq_printf(m, "maxauthsize  : %u\n", aead->maxauthsize);
+	seq_printf(m, "geniv        : %s\n", aead->geniv ?: "<built-in>");
 }
 
 const struct crypto_type crypto_aead_type = {
@@ -122,5 +129,204 @@ const struct crypto_type crypto_aead_type = {
 };
 EXPORT_SYMBOL_GPL(crypto_aead_type);
 
+static int aead_null_givencrypt(struct aead_givcrypt_request *req)
+{
+	return crypto_aead_encrypt(&req->areq);
+}
+
+static int aead_null_givdecrypt(struct aead_givcrypt_request *req)
+{
+	return crypto_aead_decrypt(&req->areq);
+}
+
+static int crypto_init_nivaead_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
+{
+	struct aead_alg *alg = &tfm->__crt_alg->cra_aead;
+	struct aead_tfm *crt = &tfm->crt_aead;
+
+	if (max(alg->maxauthsize, alg->ivsize) > PAGE_SIZE / 8)
+		return -EINVAL;
+
+	crt->setkey = setkey;
+	crt->encrypt = alg->encrypt;
+	crt->decrypt = alg->decrypt;
+	if (!alg->ivsize) {
+		crt->givencrypt = aead_null_givencrypt;
+		crt->givdecrypt = aead_null_givdecrypt;
+	}
+	crt->base = __crypto_aead_cast(tfm);
+	crt->ivsize = alg->ivsize;
+	crt->authsize = alg->maxauthsize;
+
+	return 0;
+}
+
+static void crypto_nivaead_show(struct seq_file *m, struct crypto_alg *alg)
+	__attribute__ ((unused));
+static void crypto_nivaead_show(struct seq_file *m, struct crypto_alg *alg)
+{
+	struct aead_alg *aead = &alg->cra_aead;
+
+	seq_printf(m, "type         : nivaead\n");
+	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
+	seq_printf(m, "ivsize       : %u\n", aead->ivsize);
+	seq_printf(m, "maxauthsize  : %u\n", aead->maxauthsize);
+	seq_printf(m, "geniv        : %s\n", aead->geniv);
+}
+
+const struct crypto_type crypto_nivaead_type = {
+	.ctxsize = crypto_aead_ctxsize,
+	.init = crypto_init_nivaead_ops,
+#ifdef CONFIG_PROC_FS
+	.show = crypto_nivaead_show,
+#endif
+};
+EXPORT_SYMBOL_GPL(crypto_nivaead_type);
+
+static int crypto_grab_nivaead(struct crypto_aead_spawn *spawn,
+			       const char *name, u32 type, u32 mask)
+{
+	struct crypto_alg *alg;
+	int err;
+
+	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	type |= CRYPTO_ALG_TYPE_AEAD;
+	mask |= CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV;
+
+	alg = crypto_alg_mod_lookup(name, type, mask);
+	if (IS_ERR(alg))
+		return PTR_ERR(alg);
+
+	err = crypto_init_spawn(&spawn->base, alg, spawn->base.inst, mask);
+	crypto_mod_put(alg);
+	return err;
+}
+
+struct crypto_instance *aead_geniv_alloc(struct crypto_template *tmpl,
+					 struct rtattr **tb, u32 type,
+					 u32 mask)
+{
+	const char *name;
+	struct crypto_aead_spawn *spawn;
+	struct crypto_attr_type *algt;
+	struct crypto_instance *inst;
+	struct crypto_alg *alg;
+	int err;
+
+	algt = crypto_get_attr_type(tb);
+	err = PTR_ERR(algt);
+	if (IS_ERR(algt))
+		return ERR_PTR(err);
+
+	if ((algt->type ^ (CRYPTO_ALG_TYPE_AEAD | CRYPTO_ALG_GENIV)) &
+	    algt->mask)
+		return ERR_PTR(-EINVAL);
+
+	name = crypto_attr_alg_name(tb[1]);
+	err = PTR_ERR(name);
+	if (IS_ERR(name))
+		return ERR_PTR(err);
+
+	inst = kzalloc(sizeof(*inst) + sizeof(*spawn), GFP_KERNEL);
+	if (!inst)
+		return ERR_PTR(-ENOMEM);
+
+	spawn = crypto_instance_ctx(inst);
+
+	/* Ignore async algorithms if necessary. */
+	mask |= crypto_requires_sync(algt->type, algt->mask);
+
+	crypto_set_aead_spawn(spawn, inst);
+	err = crypto_grab_nivaead(spawn, name, type, mask);
+	if (err)
+		goto err_free_inst;
+
+	alg = crypto_aead_spawn_alg(spawn);
+
+	err = -EINVAL;
+	if (!alg->cra_aead.ivsize)
+		goto err_drop_alg;
+
+	/*
+	 * This is only true if we're constructing an algorithm with its
+	 * default IV generator.  For the default generator we elide the
+	 * template name and double-check the IV generator.
+	 */
+	if (algt->mask & CRYPTO_ALG_GENIV) {
+		if (strcmp(tmpl->name, alg->cra_aead.geniv))
+			goto err_drop_alg;
+
+		memcpy(inst->alg.cra_name, alg->cra_name, CRYPTO_MAX_ALG_NAME);
+		memcpy(inst->alg.cra_driver_name, alg->cra_driver_name,
+		       CRYPTO_MAX_ALG_NAME);
+	} else {
+		err = -ENAMETOOLONG;
+		if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
+			     "%s(%s)", tmpl->name, alg->cra_name) >=
+		    CRYPTO_MAX_ALG_NAME)
+			goto err_drop_alg;
+		if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
+			     "%s(%s)", tmpl->name, alg->cra_driver_name) >=
+		    CRYPTO_MAX_ALG_NAME)
+			goto err_drop_alg;
+	}
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_AEAD | CRYPTO_ALG_GENIV;
+	inst->alg.cra_flags |= alg->cra_flags & CRYPTO_ALG_ASYNC;
+	inst->alg.cra_priority = alg->cra_priority;
+	inst->alg.cra_blocksize = alg->cra_blocksize;
+	inst->alg.cra_alignmask = alg->cra_alignmask;
+	inst->alg.cra_type = &crypto_aead_type;
+
+	inst->alg.cra_aead.ivsize = alg->cra_aead.ivsize;
+	inst->alg.cra_aead.maxauthsize = alg->cra_aead.maxauthsize;
+	inst->alg.cra_aead.geniv = alg->cra_aead.geniv;
+
+	inst->alg.cra_aead.setkey = alg->cra_aead.setkey;
+	inst->alg.cra_aead.setauthsize = alg->cra_aead.setauthsize;
+	inst->alg.cra_aead.encrypt = alg->cra_aead.encrypt;
+	inst->alg.cra_aead.decrypt = alg->cra_aead.decrypt;
+
+out:
+	return inst;
+
+err_drop_alg:
+	crypto_drop_aead(spawn);
+err_free_inst:
+	kfree(inst);
+	inst = ERR_PTR(err);
+	goto out;
+}
+EXPORT_SYMBOL_GPL(aead_geniv_alloc);
+
+void aead_geniv_free(struct crypto_instance *inst)
+{
+	crypto_drop_aead(crypto_instance_ctx(inst));
+	kfree(inst);
+}
+EXPORT_SYMBOL_GPL(aead_geniv_free);
+
+int aead_geniv_init(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = (void *)tfm->__crt_alg;
+	struct crypto_aead *aead;
+
+	aead = crypto_spawn_aead(crypto_instance_ctx(inst));
+	if (IS_ERR(aead))
+		return PTR_ERR(aead);
+
+	tfm->crt_aead.base = aead;
+	tfm->crt_aead.reqsize += crypto_aead_reqsize(aead);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(aead_geniv_init);
+
+void aead_geniv_exit(struct crypto_tfm *tfm)
+{
+	crypto_free_aead(tfm->crt_aead.base);
+}
+EXPORT_SYMBOL_GPL(aead_geniv_exit);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Authenticated Encryption with Associated Data (AEAD)");
diff --git a/include/crypto/internal/aead.h b/include/crypto/internal/aead.h
new file mode 100644
index 000000000000..eb4eee7a7804
--- /dev/null
+++ b/include/crypto/internal/aead.h
@@ -0,0 +1,77 @@
+/*
+ * AEAD: Authenticated Encryption with Associated Data
+ * 
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_INTERNAL_AEAD_H
+#define _CRYPTO_INTERNAL_AEAD_H
+
+#include <crypto/aead.h>
+#include <crypto/algapi.h>
+#include <linux/types.h>
+
+struct rtattr;
+
+struct crypto_aead_spawn {
+	struct crypto_spawn base;
+};
+
+extern const struct crypto_type crypto_nivaead_type;
+
+static inline void crypto_set_aead_spawn(
+	struct crypto_aead_spawn *spawn, struct crypto_instance *inst)
+{
+	crypto_set_spawn(&spawn->base, inst);
+}
+
+static inline void crypto_drop_aead(struct crypto_aead_spawn *spawn)
+{
+	crypto_drop_spawn(&spawn->base);
+}
+
+static inline struct crypto_alg *crypto_aead_spawn_alg(
+	struct crypto_aead_spawn *spawn)
+{
+	return spawn->base.alg;
+}
+
+static inline struct crypto_aead *crypto_spawn_aead(
+	struct crypto_aead_spawn *spawn)
+{
+	return __crypto_aead_cast(
+		crypto_spawn_tfm(&spawn->base, CRYPTO_ALG_TYPE_AEAD,
+				 CRYPTO_ALG_TYPE_MASK));
+}
+
+struct crypto_instance *aead_geniv_alloc(struct crypto_template *tmpl,
+					 struct rtattr **tb, u32 type,
+					 u32 mask);
+void aead_geniv_free(struct crypto_instance *inst);
+int aead_geniv_init(struct crypto_tfm *tfm);
+void aead_geniv_exit(struct crypto_tfm *tfm);
+
+static inline struct crypto_aead *aead_geniv_base(struct crypto_aead *geniv)
+{
+	return crypto_aead_crt(geniv)->base;
+}
+
+static inline void *aead_givcrypt_reqctx(struct aead_givcrypt_request *req)
+{
+	return aead_request_ctx(&req->areq);
+}
+
+static inline void aead_givcrypt_complete(struct aead_givcrypt_request *req,
+					  int err)
+{
+	aead_request_complete(&req->areq, err);
+}
+
+#endif	/* _CRYPTO_INTERNAL_AEAD_H */
+
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 7524928bff93..639385a9672d 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -206,6 +206,8 @@ struct aead_alg {
 	int (*givencrypt)(struct aead_givcrypt_request *req);
 	int (*givdecrypt)(struct aead_givcrypt_request *req);
 
+	const char *geniv;
+
 	unsigned int ivsize;
 	unsigned int maxauthsize;
 };
@@ -353,6 +355,9 @@ struct aead_tfm {
 	int (*decrypt)(struct aead_request *req);
 	int (*givencrypt)(struct aead_givcrypt_request *req);
 	int (*givdecrypt)(struct aead_givcrypt_request *req);
+
+	struct crypto_aead *base;
+
 	unsigned int ivsize;
 	unsigned int authsize;
 	unsigned int reqsize;
@@ -781,7 +786,9 @@ static inline void crypto_aead_clear_flags(struct crypto_aead *tfm, u32 flags)
 static inline int crypto_aead_setkey(struct crypto_aead *tfm, const u8 *key,
 				     unsigned int keylen)
 {
-	return crypto_aead_crt(tfm)->setkey(tfm, key, keylen);
+	struct aead_tfm *crt = crypto_aead_crt(tfm);
+
+	return crt->setkey(crt->base, key, keylen);
 }
 
 int crypto_aead_setauthsize(struct crypto_aead *tfm, unsigned int authsize);
@@ -809,7 +816,7 @@ static inline unsigned int crypto_aead_reqsize(struct crypto_aead *tfm)
 static inline void aead_request_set_tfm(struct aead_request *req,
 					struct crypto_aead *tfm)
 {
-	req->base.tfm = crypto_aead_tfm(tfm);
+	req->base.tfm = crypto_aead_tfm(crypto_aead_crt(tfm)->base);
 }
 
 static inline struct aead_request *aead_request_alloc(struct crypto_aead *tfm,
-- 
cgit v1.2.3


From d29ce988aeb459203c74f14747f4f77e1829ef78 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 12 Dec 2007 19:24:27 +0800
Subject: [CRYPTO] aead: Create default givcipher instances

This patch makes crypto_alloc_aead always return algorithms that is
capable of generating their own IVs through givencrypt and givdecrypt.
All existing AEAD algorithms already do.  New ones must either supply
their own or specify a generic IV generator with the geniv field.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/aead.c                  | 153 +++++++++++++++++++++++++++++++++++++++++
 include/crypto/internal/aead.h |   3 +
 include/linux/crypto.h         |  10 +--
 3 files changed, 157 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/crypto/aead.c b/crypto/aead.c
index 9f7aca891926..f5b1add31976 100644
--- a/crypto/aead.c
+++ b/crypto/aead.c
@@ -17,6 +17,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/rtnetlink.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
 
@@ -328,5 +329,157 @@ void aead_geniv_exit(struct crypto_tfm *tfm)
 }
 EXPORT_SYMBOL_GPL(aead_geniv_exit);
 
+static int crypto_nivaead_default(struct crypto_alg *alg, u32 type, u32 mask)
+{
+	struct rtattr *tb[3];
+	struct {
+		struct rtattr attr;
+		struct crypto_attr_type data;
+	} ptype;
+	struct {
+		struct rtattr attr;
+		struct crypto_attr_alg data;
+	} palg;
+	struct crypto_template *tmpl;
+	struct crypto_instance *inst;
+	struct crypto_alg *larval;
+	const char *geniv;
+	int err;
+
+	larval = crypto_larval_lookup(alg->cra_driver_name,
+				      CRYPTO_ALG_TYPE_AEAD | CRYPTO_ALG_GENIV,
+				      CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	err = PTR_ERR(larval);
+	if (IS_ERR(larval))
+		goto out;
+
+	err = -EAGAIN;
+	if (!crypto_is_larval(larval))
+		goto drop_larval;
+
+	ptype.attr.rta_len = sizeof(ptype);
+	ptype.attr.rta_type = CRYPTOA_TYPE;
+	ptype.data.type = type | CRYPTO_ALG_GENIV;
+	/* GENIV tells the template that we're making a default geniv. */
+	ptype.data.mask = mask | CRYPTO_ALG_GENIV;
+	tb[0] = &ptype.attr;
+
+	palg.attr.rta_len = sizeof(palg);
+	palg.attr.rta_type = CRYPTOA_ALG;
+	/* Must use the exact name to locate ourselves. */
+	memcpy(palg.data.name, alg->cra_driver_name, CRYPTO_MAX_ALG_NAME);
+	tb[1] = &palg.attr;
+
+	tb[2] = NULL;
+
+	geniv = alg->cra_aead.geniv;
+
+	tmpl = crypto_lookup_template(geniv);
+	err = -ENOENT;
+	if (!tmpl)
+		goto kill_larval;
+
+	inst = tmpl->alloc(tb);
+	err = PTR_ERR(inst);
+	if (IS_ERR(inst))
+		goto put_tmpl;
+
+	if ((err = crypto_register_instance(tmpl, inst))) {
+		tmpl->free(inst);
+		goto put_tmpl;
+	}
+
+	/* Redo the lookup to use the instance we just registered. */
+	err = -EAGAIN;
+
+put_tmpl:
+	crypto_tmpl_put(tmpl);
+kill_larval:
+	crypto_larval_kill(larval);
+drop_larval:
+	crypto_mod_put(larval);
+out:
+	crypto_mod_put(alg);
+	return err;
+}
+
+static struct crypto_alg *crypto_lookup_aead(const char *name, u32 type,
+					     u32 mask)
+{
+	struct crypto_alg *alg;
+
+	alg = crypto_alg_mod_lookup(name, type, mask);
+	if (IS_ERR(alg))
+		return alg;
+
+	if (alg->cra_type == &crypto_aead_type)
+		return alg;
+
+	if (!alg->cra_aead.ivsize)
+		return alg;
+
+	return ERR_PTR(crypto_nivaead_default(alg, type, mask));
+}
+
+int crypto_grab_aead(struct crypto_aead_spawn *spawn, const char *name,
+		     u32 type, u32 mask)
+{
+	struct crypto_alg *alg;
+	int err;
+
+	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	type |= CRYPTO_ALG_TYPE_AEAD;
+	mask &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	alg = crypto_lookup_aead(name, type, mask);
+	if (IS_ERR(alg))
+		return PTR_ERR(alg);
+
+	err = crypto_init_spawn(&spawn->base, alg, spawn->base.inst, mask);
+	crypto_mod_put(alg);
+	return err;
+}
+EXPORT_SYMBOL_GPL(crypto_grab_aead);
+
+struct crypto_aead *crypto_alloc_aead(const char *alg_name, u32 type, u32 mask)
+{
+	struct crypto_tfm *tfm;
+	int err;
+
+	type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	type |= CRYPTO_ALG_TYPE_AEAD;
+	mask &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	for (;;) {
+		struct crypto_alg *alg;
+
+		alg = crypto_lookup_aead(alg_name, type, mask);
+		if (IS_ERR(alg)) {
+			err = PTR_ERR(alg);
+			goto err;
+		}
+
+		tfm = __crypto_alloc_tfm(alg, type, mask);
+		if (!IS_ERR(tfm))
+			return __crypto_aead_cast(tfm);
+
+		crypto_mod_put(alg);
+		err = PTR_ERR(tfm);
+
+err:
+		if (err != -EAGAIN)
+			break;
+		if (signal_pending(current)) {
+			err = -EINTR;
+			break;
+		}
+	}
+
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(crypto_alloc_aead);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Authenticated Encryption with Associated Data (AEAD)");
diff --git a/include/crypto/internal/aead.h b/include/crypto/internal/aead.h
index eb4eee7a7804..d838c945575a 100644
--- a/include/crypto/internal/aead.h
+++ b/include/crypto/internal/aead.h
@@ -31,6 +31,9 @@ static inline void crypto_set_aead_spawn(
 	crypto_set_spawn(&spawn->base, inst);
 }
 
+int crypto_grab_aead(struct crypto_aead_spawn *spawn, const char *name,
+		     u32 type, u32 mask);
+
 static inline void crypto_drop_aead(struct crypto_aead_spawn *spawn)
 {
 	crypto_drop_spawn(&spawn->base);
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 639385a9672d..0aba10460201 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -723,15 +723,7 @@ static inline struct crypto_aead *__crypto_aead_cast(struct crypto_tfm *tfm)
 	return (struct crypto_aead *)tfm;
 }
 
-static inline struct crypto_aead *crypto_alloc_aead(const char *alg_name,
-						    u32 type, u32 mask)
-{
-	type &= ~CRYPTO_ALG_TYPE_MASK;
-	type |= CRYPTO_ALG_TYPE_AEAD;
-	mask |= CRYPTO_ALG_TYPE_MASK;
-
-	return __crypto_aead_cast(crypto_alloc_base(alg_name, type, mask));
-}
+struct crypto_aead *crypto_alloc_aead(const char *alg_name, u32 type, u32 mask);
 
 static inline struct crypto_tfm *crypto_aead_tfm(struct crypto_aead *tfm)
 {
-- 
cgit v1.2.3


From 3631c650c495d61b1dabf32eb26b46873636e918 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 13 Dec 2007 22:28:59 +0800
Subject: [CRYPTO] null: Add null blkcipher algorithm

This patch adds a null blkcipher algorithm called ecb(cipher_null) for
backwards compatibility.  Previously the null algorithm when used by
IPsec copied the data byte by byte.  This new algorithm optimises that
to a straight memcpy which lets us better measure inherent overheads in
our IPsec code.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/crypto_null.c               | 69 ++++++++++++++++++++++++++++++++------
 include/crypto/internal/skcipher.h |  5 +++
 2 files changed, 64 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/crypto/crypto_null.c b/crypto/crypto_null.c
index 29f77477d701..fc2d2a4d7789 100644
--- a/crypto/crypto_null.c
+++ b/crypto/crypto_null.c
@@ -16,15 +16,17 @@
  * (at your option) any later version.
  *
  */
+
+#include <crypto/internal/skcipher.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mm.h>
-#include <linux/crypto.h>
 #include <linux/string.h>
 
 #define NULL_KEY_SIZE		0
 #define NULL_BLOCK_SIZE		1
 #define NULL_DIGEST_SIZE	0
+#define NULL_IV_SIZE		0
 
 static int null_compress(struct crypto_tfm *tfm, const u8 *src,
 			 unsigned int slen, u8 *dst, unsigned int *dlen)
@@ -55,6 +57,26 @@ static void null_crypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 	memcpy(dst, src, NULL_BLOCK_SIZE);
 }
 
+static int skcipher_null_crypt(struct blkcipher_desc *desc,
+			       struct scatterlist *dst,
+			       struct scatterlist *src, unsigned int nbytes)
+{
+	struct blkcipher_walk walk;
+	int err;
+
+	blkcipher_walk_init(&walk, dst, src, nbytes);
+	err = blkcipher_walk_virt(desc, &walk);
+
+	while (walk.nbytes) {
+		if (walk.src.virt.addr != walk.dst.virt.addr)
+			memcpy(walk.dst.virt.addr, walk.src.virt.addr,
+			       walk.nbytes);
+		err = blkcipher_walk_done(desc, &walk, 0);
+	}
+
+	return err;
+}
+
 static struct crypto_alg compress_null = {
 	.cra_name		=	"compress_null",
 	.cra_flags		=	CRYPTO_ALG_TYPE_COMPRESS,
@@ -96,6 +118,25 @@ static struct crypto_alg cipher_null = {
 	.cia_decrypt		=	null_crypt } }
 };
 
+static struct crypto_alg skcipher_null = {
+	.cra_name		=	"ecb(cipher_null)",
+	.cra_driver_name	=	"ecb-cipher_null",
+	.cra_priority		=	100,
+	.cra_flags		=	CRYPTO_ALG_TYPE_BLKCIPHER,
+	.cra_blocksize		=	NULL_BLOCK_SIZE,
+	.cra_type		=	&crypto_blkcipher_type,
+	.cra_ctxsize		=	0,
+	.cra_module		=	THIS_MODULE,
+	.cra_list		=	LIST_HEAD_INIT(skcipher_null.cra_list),
+	.cra_u			=	{ .blkcipher = {
+	.min_keysize		=	NULL_KEY_SIZE,
+	.max_keysize		=	NULL_KEY_SIZE,
+	.ivsize			=	NULL_IV_SIZE,
+	.setkey			= 	null_setkey,
+	.encrypt		=	skcipher_null_crypt,
+	.decrypt		=	skcipher_null_crypt } }
+};
+
 MODULE_ALIAS("compress_null");
 MODULE_ALIAS("digest_null");
 MODULE_ALIAS("cipher_null");
@@ -108,27 +149,35 @@ static int __init init(void)
 	if (ret < 0)
 		goto out;
 
+	ret = crypto_register_alg(&skcipher_null);
+	if (ret < 0)
+		goto out_unregister_cipher;
+
 	ret = crypto_register_alg(&digest_null);
-	if (ret < 0) {
-		crypto_unregister_alg(&cipher_null);
-		goto out;
-	}
+	if (ret < 0)
+		goto out_unregister_skcipher;
 
 	ret = crypto_register_alg(&compress_null);
-	if (ret < 0) {
-		crypto_unregister_alg(&digest_null);
-		crypto_unregister_alg(&cipher_null);
-		goto out;
-	}
+	if (ret < 0)
+		goto out_unregister_digest;
 
 out:	
 	return ret;
+
+out_unregister_digest:
+	crypto_unregister_alg(&digest_null);
+out_unregister_skcipher:
+	crypto_unregister_alg(&skcipher_null);
+out_unregister_cipher:
+	crypto_unregister_alg(&cipher_null);
+	goto out;
 }
 
 static void __exit fini(void)
 {
 	crypto_unregister_alg(&compress_null);
 	crypto_unregister_alg(&digest_null);
+	crypto_unregister_alg(&skcipher_null);
 	crypto_unregister_alg(&cipher_null);
 }
 
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index 2071999d4b5e..0053f34764ff 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -88,5 +88,10 @@ static inline void skcipher_givcrypt_complete(
 	ablkcipher_request_complete(&req->creq, err);
 }
 
+static inline u32 ablkcipher_request_flags(struct ablkcipher_request *req)
+{
+	return req->base.flags;
+}
+
 #endif	/* _CRYPTO_INTERNAL_SKCIPHER_H */
 
-- 
cgit v1.2.3


From e7cd2514ea506f06bd4f7b13a9b62afd60f9c73b Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 14 Dec 2007 22:28:14 +0800
Subject: [CRYPTO] chainiv: Avoid lock spinning where possible

This patch makes chainiv avoid spinning by postponing requests on lock
contention if the user allows the use of asynchronous algorithms.  If
a synchronous algorithm is requested then we behave as before.

This should improve IPsec performance on SMP when two CPUs attempt to
transmit over the same SA.  Currently one of them will spin doing nothing
waiting for the other CPU to finish its encryption.  This patch makes it
postpone the request and get on with other work.

If only one CPU is transmitting for a given SA, then we will process
the request synchronously as before.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/chainiv.c                   | 208 +++++++++++++++++++++++++++++++++++--
 include/crypto/internal/skcipher.h |  13 +++
 2 files changed, 213 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/crypto/chainiv.c b/crypto/chainiv.c
index 38868d160b47..d17fa0454dc3 100644
--- a/crypto/chainiv.c
+++ b/crypto/chainiv.c
@@ -16,16 +16,34 @@
 #include <crypto/internal/skcipher.h>
 #include <linux/err.h>
 #include <linux/init.h>
+#include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/spinlock.h>
 #include <linux/string.h>
+#include <linux/workqueue.h>
+
+enum {
+	CHAINIV_STATE_INUSE = 0,
+};
 
 struct chainiv_ctx {
 	spinlock_t lock;
 	char iv[];
 };
 
+struct async_chainiv_ctx {
+	unsigned long state;
+
+	spinlock_t lock;
+	int err;
+
+	struct crypto_queue queue;
+	struct work_struct postponed;
+
+	char iv[];
+};
+
 static int chainiv_givencrypt(struct skcipher_givcrypt_request *req)
 {
 	struct crypto_ablkcipher *geniv = skcipher_givcrypt_reqtfm(req);
@@ -80,26 +98,187 @@ unlock:
 	return chainiv_givencrypt(req);
 }
 
+static int chainiv_init_common(struct crypto_tfm *tfm)
+{
+	tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request);
+
+	return skcipher_geniv_init(tfm);
+}
+
 static int chainiv_init(struct crypto_tfm *tfm)
 {
-	struct crypto_ablkcipher *geniv = __crypto_ablkcipher_cast(tfm);
-	struct chainiv_ctx *ctx = crypto_ablkcipher_ctx(geniv);
+	struct chainiv_ctx *ctx = crypto_tfm_ctx(tfm);
 
 	spin_lock_init(&ctx->lock);
 
-	tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request);
+	return chainiv_init_common(tfm);
+}
 
-	return skcipher_geniv_init(tfm);
+static int async_chainiv_schedule_work(struct async_chainiv_ctx *ctx)
+{
+	int queued;
+
+	if (!ctx->queue.qlen) {
+		smp_mb__before_clear_bit();
+		clear_bit(CHAINIV_STATE_INUSE, &ctx->state);
+
+		if (!ctx->queue.qlen ||
+		    test_and_set_bit(CHAINIV_STATE_INUSE, &ctx->state))
+			goto out;
+	}
+
+	queued = schedule_work(&ctx->postponed);
+	BUG_ON(!queued);
+
+out:
+	return ctx->err;
+}
+
+static int async_chainiv_postpone_request(struct skcipher_givcrypt_request *req)
+{
+	struct crypto_ablkcipher *geniv = skcipher_givcrypt_reqtfm(req);
+	struct async_chainiv_ctx *ctx = crypto_ablkcipher_ctx(geniv);
+	int err;
+
+	spin_lock_bh(&ctx->lock);
+	err = skcipher_enqueue_givcrypt(&ctx->queue, req);
+	spin_unlock_bh(&ctx->lock);
+
+	if (test_and_set_bit(CHAINIV_STATE_INUSE, &ctx->state))
+		return err;
+
+	ctx->err = err;
+	return async_chainiv_schedule_work(ctx);
+}
+
+static int async_chainiv_givencrypt_tail(struct skcipher_givcrypt_request *req)
+{
+	struct crypto_ablkcipher *geniv = skcipher_givcrypt_reqtfm(req);
+	struct async_chainiv_ctx *ctx = crypto_ablkcipher_ctx(geniv);
+	struct ablkcipher_request *subreq = skcipher_givcrypt_reqctx(req);
+	unsigned int ivsize = crypto_ablkcipher_ivsize(geniv);
+
+	memcpy(req->giv, ctx->iv, ivsize);
+	memcpy(subreq->info, ctx->iv, ivsize);
+
+	ctx->err = crypto_ablkcipher_encrypt(subreq);
+	if (ctx->err)
+		goto out;
+
+	memcpy(ctx->iv, subreq->info, ivsize);
+
+out:
+	return async_chainiv_schedule_work(ctx);
+}
+
+static int async_chainiv_givencrypt(struct skcipher_givcrypt_request *req)
+{
+	struct crypto_ablkcipher *geniv = skcipher_givcrypt_reqtfm(req);
+	struct async_chainiv_ctx *ctx = crypto_ablkcipher_ctx(geniv);
+	struct ablkcipher_request *subreq = skcipher_givcrypt_reqctx(req);
+
+	ablkcipher_request_set_tfm(subreq, skcipher_geniv_cipher(geniv));
+	ablkcipher_request_set_callback(subreq, req->creq.base.flags,
+					req->creq.base.complete,
+					req->creq.base.data);
+	ablkcipher_request_set_crypt(subreq, req->creq.src, req->creq.dst,
+				     req->creq.nbytes, req->creq.info);
+
+	if (test_and_set_bit(CHAINIV_STATE_INUSE, &ctx->state))
+		goto postpone;
+
+	if (ctx->queue.qlen) {
+		clear_bit(CHAINIV_STATE_INUSE, &ctx->state);
+		goto postpone;
+	}
+
+	return async_chainiv_givencrypt_tail(req);
+
+postpone:
+	return async_chainiv_postpone_request(req);
+}
+
+static int async_chainiv_givencrypt_first(struct skcipher_givcrypt_request *req)
+{
+	struct crypto_ablkcipher *geniv = skcipher_givcrypt_reqtfm(req);
+	struct async_chainiv_ctx *ctx = crypto_ablkcipher_ctx(geniv);
+
+	if (test_and_set_bit(CHAINIV_STATE_INUSE, &ctx->state))
+		goto out;
+
+	if (crypto_ablkcipher_crt(geniv)->givencrypt !=
+	    async_chainiv_givencrypt_first)
+		goto unlock;
+
+	crypto_ablkcipher_crt(geniv)->givencrypt = async_chainiv_givencrypt;
+	get_random_bytes(ctx->iv, crypto_ablkcipher_ivsize(geniv));
+
+unlock:
+	clear_bit(CHAINIV_STATE_INUSE, &ctx->state);
+
+out:
+	return async_chainiv_givencrypt(req);
+}
+
+static void async_chainiv_do_postponed(struct work_struct *work)
+{
+	struct async_chainiv_ctx *ctx = container_of(work,
+						     struct async_chainiv_ctx,
+						     postponed);
+	struct skcipher_givcrypt_request *req;
+	struct ablkcipher_request *subreq;
+
+	/* Only handle one request at a time to avoid hogging keventd. */
+	spin_lock_bh(&ctx->lock);
+	req = skcipher_dequeue_givcrypt(&ctx->queue);
+	spin_unlock_bh(&ctx->lock);
+
+	if (!req) {
+		async_chainiv_schedule_work(ctx);
+		return;
+	}
+
+	subreq = skcipher_givcrypt_reqctx(req);
+	subreq->base.flags |= CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	async_chainiv_givencrypt_tail(req);
+}
+
+static int async_chainiv_init(struct crypto_tfm *tfm)
+{
+	struct async_chainiv_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	spin_lock_init(&ctx->lock);
+
+	crypto_init_queue(&ctx->queue, 100);
+	INIT_WORK(&ctx->postponed, async_chainiv_do_postponed);
+
+	return chainiv_init_common(tfm);
+}
+
+static void async_chainiv_exit(struct crypto_tfm *tfm)
+{
+	struct async_chainiv_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	BUG_ON(test_bit(CHAINIV_STATE_INUSE, &ctx->state) || ctx->queue.qlen);
+
+	skcipher_geniv_exit(tfm);
 }
 
 static struct crypto_template chainiv_tmpl;
 
 static struct crypto_instance *chainiv_alloc(struct rtattr **tb)
 {
+	struct crypto_attr_type *algt;
 	struct crypto_instance *inst;
+	int err;
 
-	inst = skcipher_geniv_alloc(&chainiv_tmpl, tb, 0,
-				    CRYPTO_ALG_ASYNC);
+	algt = crypto_get_attr_type(tb);
+	err = PTR_ERR(algt);
+	if (IS_ERR(algt))
+		return ERR_PTR(err);
+
+	inst = skcipher_geniv_alloc(&chainiv_tmpl, tb, 0, 0);
 	if (IS_ERR(inst))
 		goto out;
 
@@ -108,8 +287,21 @@ static struct crypto_instance *chainiv_alloc(struct rtattr **tb)
 	inst->alg.cra_init = chainiv_init;
 	inst->alg.cra_exit = skcipher_geniv_exit;
 
-	inst->alg.cra_ctxsize = sizeof(struct chainiv_ctx) +
-				inst->alg.cra_ablkcipher.ivsize;
+	inst->alg.cra_ctxsize = sizeof(struct chainiv_ctx);
+
+	if (!crypto_requires_sync(algt->type, algt->mask)) {
+		inst->alg.cra_flags |= CRYPTO_ALG_ASYNC;
+
+		inst->alg.cra_ablkcipher.givencrypt =
+			async_chainiv_givencrypt_first;
+
+		inst->alg.cra_init = async_chainiv_init;
+		inst->alg.cra_exit = async_chainiv_exit;
+
+		inst->alg.cra_ctxsize = sizeof(struct async_chainiv_ctx);
+	}
+
+	inst->alg.cra_ctxsize += inst->alg.cra_ablkcipher.ivsize;
 
 out:
 	return inst;
diff --git a/include/crypto/internal/skcipher.h b/include/crypto/internal/skcipher.h
index 0053f34764ff..2ba42cd7d6aa 100644
--- a/include/crypto/internal/skcipher.h
+++ b/include/crypto/internal/skcipher.h
@@ -70,6 +70,19 @@ static inline struct crypto_ablkcipher *skcipher_geniv_cipher(
 	return crypto_ablkcipher_crt(geniv)->base;
 }
 
+static inline int skcipher_enqueue_givcrypt(
+	struct crypto_queue *queue, struct skcipher_givcrypt_request *request)
+{
+	return ablkcipher_enqueue_request(queue, &request->creq);
+}
+
+static inline struct skcipher_givcrypt_request *skcipher_dequeue_givcrypt(
+	struct crypto_queue *queue)
+{
+	return container_of(ablkcipher_dequeue_request(queue),
+			    struct skcipher_givcrypt_request, creq);
+}
+
 static inline void *skcipher_givcrypt_reqctx(
 	struct skcipher_givcrypt_request *req)
 {
-- 
cgit v1.2.3


From 1c5dfe6a959b79215c0f73d793169a7d5755900e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 24 Dec 2007 08:54:24 +0800
Subject: [CRYPTO] api: Include sched.h for cond_resched in scatterwalk.h

As Andrew Morton correctly points out, we need to explicitly include
sched.h as we use the function cond_resched in crypto/scatterwalk.h.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/crypto/scatterwalk.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/crypto/scatterwalk.h b/include/crypto/scatterwalk.h
index bd62431c33bd..224658b8d806 100644
--- a/include/crypto/scatterwalk.h
+++ b/include/crypto/scatterwalk.h
@@ -23,6 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/mm.h>
 #include <linux/scatterlist.h>
+#include <linux/sched.h>
 
 static inline enum km_type crypto_kmap_type(int out)
 {
-- 
cgit v1.2.3


From 6eb7228421c01ba48a6a88a7a5b3e71cfb70d4a9 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 8 Jan 2008 17:16:44 +1100
Subject: [CRYPTO] api: Set default CRYPTO_MINALIGN to unsigned long long

Thanks to David Miller for pointing out that the SLAB (or SLOB/SLUB)
cache uses the alignment of unsigned long long if the architecture
kmalloc/slab alignment macros are not defined.

This patch changes the CRYPTO_MINALIGN so that it uses the same default
value.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/crypto.h | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 0aba10460201..5e02d1b46370 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -90,13 +90,11 @@
 #define CRYPTO_MINALIGN ARCH_KMALLOC_MINALIGN
 #elif defined(ARCH_SLAB_MINALIGN)
 #define CRYPTO_MINALIGN ARCH_SLAB_MINALIGN
+#else
+#define CRYPTO_MINALIGN __alignof__(unsigned long long)
 #endif
 
-#ifdef CRYPTO_MINALIGN
 #define CRYPTO_MINALIGN_ATTR __attribute__ ((__aligned__(CRYPTO_MINALIGN)))
-#else
-#define CRYPTO_MINALIGN_ATTR
-#endif
 
 struct scatterlist;
 struct crypto_ablkcipher;
-- 
cgit v1.2.3


From 285e9670d91cdeb6b6693729950339cb45410fdc Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Tue, 14 Aug 2007 14:10:39 +0200
Subject: [SCSI] sr,sd: send media state change modification events

This will send for a card reader slot (remove/add media):
  UEVENT[1187091572.155884] change   /devices/pci0000:00/0000:00:1d.7/usb5/5-2/5-2:1.0/host7/target7:0:0/7:0:0:0 (scsi)
  UEVENT[1187091572.162314] remove   /block/sdb/sdb1 (block)
  UEVENT[1187091572.172464] add      /block/sdb/sdb1 (block)
  UEVENT[1187091572.175408] change   /devices/pci0000:00/0000:00:1d.7/usb5/5-2/5-2:1.0/host7/target7:0:0/7:0:0:0 (scsi)

and for a DVD drive (add/eject media):
  UEVENT[1187091590.189159] change   /devices/pci0000:00/0000:00:1f.1/host4/target4:0:0/4:0:0:0 (scsi)
  UEVENT[1187091590.957124] add      /module/isofs (module)
  UEVENT[1187091604.468207] change   /devices/pci0000:00/0000:00:1f.1/host4/target4:0:0/4:0:0:0 (scsi)

Userspace gets events, even for unpartitioned media. This unifies
the event handling for asynchronoous events (AN) and events caused by
perodical polling the device from userspace.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>

[jejb: modified for new event API]

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sd.c | 24 +++++++++++++++---------
 drivers/scsi/sr.c | 14 +++++++++++---
 drivers/scsi/sr.h |  1 +
 include/scsi/sd.h |  1 +
 4 files changed, 28 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index a69b155f39a2..18343a6acd8e 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -749,8 +749,11 @@ static int sd_media_changed(struct gendisk *disk)
 	 * can deal with it then.  It is only because of unrecoverable errors
 	 * that we would ever take a device offline in the first place.
 	 */
-	if (!scsi_device_online(sdp))
-		goto not_present;
+	if (!scsi_device_online(sdp)) {
+		set_media_not_present(sdkp);
+		retval = 1;
+		goto out;
+	}
 
 	/*
 	 * Using TEST_UNIT_READY enables differentiation between drive with
@@ -762,6 +765,7 @@ static int sd_media_changed(struct gendisk *disk)
 	 * sd_revalidate() is called.
 	 */
 	retval = -ENODEV;
+
 	if (scsi_block_when_processing_errors(sdp))
 		retval = scsi_test_unit_ready(sdp, SD_TIMEOUT, SD_MAX_RETRIES);
 
@@ -771,8 +775,11 @@ static int sd_media_changed(struct gendisk *disk)
 	 * and we will figure it out later once the drive is
 	 * available again.
 	 */
-	if (retval)
-		 goto not_present;
+	if (retval) {
+		set_media_not_present(sdkp);
+		retval = 1;
+		goto out;
+	}
 
 	/*
 	 * For removable scsi disk we have to recognise the presence
@@ -783,12 +790,11 @@ static int sd_media_changed(struct gendisk *disk)
 
 	retval = sdp->changed;
 	sdp->changed = 0;
-
+out:
+	if (retval != sdkp->previous_state)
+		sdev_evt_send_simple(sdp, SDEV_EVT_MEDIA_CHANGE, GFP_KERNEL);
+	sdkp->previous_state = retval;
 	return retval;
-
-not_present:
-	set_media_not_present(sdkp);
-	return 1;
 }
 
 static int sd_sync_cache(struct scsi_disk *sdkp)
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index c61999031141..7702681d93f9 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -192,8 +192,9 @@ static int sr_media_change(struct cdrom_device_info *cdi, int slot)
 		 * and we will figure it out later once the drive is
 		 * available again.  */
 		cd->device->changed = 1;
-		return 1;	/* This will force a flush, if called from
-				 * check_disk_change */
+		/* This will force a flush, if called from check_disk_change */
+		retval = 1;
+		goto out;
 	};
 
 	retval = cd->device->changed;
@@ -203,9 +204,16 @@ static int sr_media_change(struct cdrom_device_info *cdi, int slot)
 	if (retval) {
 		/* check multisession offset etc */
 		sr_cd_check(cdi);
-
 		get_sectorsize(cd);
 	}
+
+out:
+	/* Notify userspace, that media has changed. */
+	if (retval != cd->previous_state)
+		sdev_evt_send_simple(cd->device, SDEV_EVT_MEDIA_CHANGE,
+				     GFP_KERNEL);
+	cd->previous_state = retval;
+
 	return retval;
 }
  
diff --git a/drivers/scsi/sr.h b/drivers/scsi/sr.h
index d65de9621b27..0d04e2878c9d 100644
--- a/drivers/scsi/sr.h
+++ b/drivers/scsi/sr.h
@@ -37,6 +37,7 @@ typedef struct scsi_cd {
 	unsigned xa_flag:1;	/* CD has XA sectors ? */
 	unsigned readcd_known:1;	/* drive supports READ_CD (0xbe) */
 	unsigned readcd_cdda:1;	/* reading audio data using READ_CD */
+	unsigned previous_state:1;	/* media has changed */
 	struct cdrom_device_info cdi;
 	/* We hold gendisk and scsi_device references on probe and use
 	 * the refs on this kref to decide when to release them */
diff --git a/include/scsi/sd.h b/include/scsi/sd.h
index f7513313ef0d..8ea9f7358ac1 100644
--- a/include/scsi/sd.h
+++ b/include/scsi/sd.h
@@ -41,6 +41,7 @@ struct scsi_disk {
 	u32		index;
 	u8		media_present;
 	u8		write_prot;
+	unsigned	previous_state : 1;
 	unsigned	WCE : 1;	/* state of disk WCE bit */
 	unsigned	RCD : 1;	/* state of disk RCD bit, unused */
 	unsigned	DPOFUA : 1;	/* state of disk DPOFUA bit */
-- 
cgit v1.2.3


From 5929faf3334f4c69f3bb02be59d7c127e0cefa1f Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <djwong@us.ibm.com>
Date: Mon, 5 Nov 2007 11:51:17 -0800
Subject: [SCSI] libsas: Convert sas_proto users to sas_protocol

sparse complains about the mixing of enums in libsas.  Since the
underlying numeric values of both enums are the same, combine them
to get rid of the warning.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/aic94xx/aic94xx_dev.c  |  6 +++---
 drivers/scsi/aic94xx/aic94xx_dump.c |  4 ++--
 drivers/scsi/aic94xx/aic94xx_hwi.c  |  2 +-
 drivers/scsi/aic94xx/aic94xx_scb.c  |  6 +++---
 drivers/scsi/aic94xx/aic94xx_task.c | 30 +++++++++++++++---------------
 drivers/scsi/aic94xx/aic94xx_tmf.c  | 12 ++++++------
 drivers/scsi/libsas/sas_discover.c  |  2 +-
 drivers/scsi/libsas/sas_expander.c  |  6 +++---
 drivers/scsi/libsas/sas_internal.h  |  2 +-
 include/scsi/libsas.h               | 18 +++++++++---------
 include/scsi/sas.h                  | 13 ++++++-------
 include/scsi/scsi_transport_sas.h   |  8 +-------
 12 files changed, 51 insertions(+), 58 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/aic94xx/aic94xx_dev.c b/drivers/scsi/aic94xx/aic94xx_dev.c
index 3dce618bf414..72042cae7768 100644
--- a/drivers/scsi/aic94xx/aic94xx_dev.c
+++ b/drivers/scsi/aic94xx/aic94xx_dev.c
@@ -165,7 +165,7 @@ static int asd_init_target_ddb(struct domain_device *dev)
 	if (dev->port->oob_mode != SATA_OOB_MODE) {
 		flags |= OPEN_REQUIRED;
 		if ((dev->dev_type == SATA_DEV) ||
-		    (dev->tproto & SAS_PROTO_STP)) {
+		    (dev->tproto & SAS_PROTOCOL_STP)) {
 			struct smp_resp *rps_resp = &dev->sata_dev.rps_resp;
 			if (rps_resp->frame_type == SMP_RESPONSE &&
 			    rps_resp->function == SMP_REPORT_PHY_SATA &&
@@ -193,7 +193,7 @@ static int asd_init_target_ddb(struct domain_device *dev)
 	asd_ddbsite_write_byte(asd_ha, ddb, DDB_TARG_FLAGS, flags);
 
 	flags = 0;
-	if (dev->tproto & SAS_PROTO_STP)
+	if (dev->tproto & SAS_PROTOCOL_STP)
 		flags |= STP_CL_POL_NO_TX;
 	asd_ddbsite_write_byte(asd_ha, ddb, DDB_TARG_FLAGS2, flags);
 
@@ -201,7 +201,7 @@ static int asd_init_target_ddb(struct domain_device *dev)
 	asd_ddbsite_write_word(asd_ha, ddb, SEND_QUEUE_TAIL, 0xFFFF);
 	asd_ddbsite_write_word(asd_ha, ddb, SISTER_DDB, 0xFFFF);
 
-	if (dev->dev_type == SATA_DEV || (dev->tproto & SAS_PROTO_STP)) {
+	if (dev->dev_type == SATA_DEV || (dev->tproto & SAS_PROTOCOL_STP)) {
 		i = asd_init_sata(dev);
 		if (i < 0) {
 			asd_free_ddb(asd_ha, ddb);
diff --git a/drivers/scsi/aic94xx/aic94xx_dump.c b/drivers/scsi/aic94xx/aic94xx_dump.c
index 6bd8e3059d27..3d8c4ff1f2ef 100644
--- a/drivers/scsi/aic94xx/aic94xx_dump.c
+++ b/drivers/scsi/aic94xx/aic94xx_dump.c
@@ -903,11 +903,11 @@ void asd_dump_frame_rcvd(struct asd_phy *phy,
 	int i;
 
 	switch ((dl->status_block[1] & 0x70) >> 3) {
-	case SAS_PROTO_STP:
+	case SAS_PROTOCOL_STP:
 		ASD_DPRINTK("STP proto device-to-host FIS:\n");
 		break;
 	default:
-	case SAS_PROTO_SSP:
+	case SAS_PROTOCOL_SSP:
 		ASD_DPRINTK("SAS proto IDENTIFY:\n");
 		break;
 	}
diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c
index 0cd7eed9196c..098b5f39cd31 100644
--- a/drivers/scsi/aic94xx/aic94xx_hwi.c
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.c
@@ -91,7 +91,7 @@ static int asd_init_phy(struct asd_phy *phy)
 
 	sas_phy->enabled = 1;
 	sas_phy->class = SAS;
-	sas_phy->iproto = SAS_PROTO_ALL;
+	sas_phy->iproto = SAS_PROTOCOL_ALL;
 	sas_phy->tproto = 0;
 	sas_phy->type = PHY_TYPE_PHYSICAL;
 	sas_phy->role = PHY_ROLE_INITIATOR;
diff --git a/drivers/scsi/aic94xx/aic94xx_scb.c b/drivers/scsi/aic94xx/aic94xx_scb.c
index db6ab1a3b81e..0febad4dd75f 100644
--- a/drivers/scsi/aic94xx/aic94xx_scb.c
+++ b/drivers/scsi/aic94xx/aic94xx_scb.c
@@ -788,12 +788,12 @@ void asd_build_control_phy(struct asd_ascb *ascb, int phy_id, u8 subfunc)
 
 		/* initiator port settings are in the hi nibble */
 		if (phy->sas_phy.role == PHY_ROLE_INITIATOR)
-			control_phy->port_type = SAS_PROTO_ALL << 4;
+			control_phy->port_type = SAS_PROTOCOL_ALL << 4;
 		else if (phy->sas_phy.role == PHY_ROLE_TARGET)
-			control_phy->port_type = SAS_PROTO_ALL;
+			control_phy->port_type = SAS_PROTOCOL_ALL;
 		else
 			control_phy->port_type =
-				(SAS_PROTO_ALL << 4) | SAS_PROTO_ALL;
+				(SAS_PROTOCOL_ALL << 4) | SAS_PROTOCOL_ALL;
 
 		/* link reset retries, this should be nominal */
 		control_phy->link_reset_retries = 10;
diff --git a/drivers/scsi/aic94xx/aic94xx_task.c b/drivers/scsi/aic94xx/aic94xx_task.c
index ee0a98bffcd4..663fdef38c52 100644
--- a/drivers/scsi/aic94xx/aic94xx_task.c
+++ b/drivers/scsi/aic94xx/aic94xx_task.c
@@ -187,7 +187,7 @@ static void asd_get_response_tasklet(struct asd_ascb *ascb,
 	ts->buf_valid_size = 0;
 	edb = asd_ha->seq.edb_arr[edb_id + escb->edb_index];
 	r = edb->vaddr;
-	if (task->task_proto == SAS_PROTO_SSP) {
+	if (task->task_proto == SAS_PROTOCOL_SSP) {
 		struct ssp_response_iu *iu =
 			r + 16 + sizeof(struct ssp_frame_hdr);
 
@@ -341,14 +341,14 @@ Again:
 	}
 
 	switch (task->task_proto) {
-	case SATA_PROTO:
-	case SAS_PROTO_STP:
+	case SAS_PROTOCOL_SATA:
+	case SAS_PROTOCOL_STP:
 		asd_unbuild_ata_ascb(ascb);
 		break;
-	case SAS_PROTO_SMP:
+	case SAS_PROTOCOL_SMP:
 		asd_unbuild_smp_ascb(ascb);
 		break;
-	case SAS_PROTO_SSP:
+	case SAS_PROTOCOL_SSP:
 		asd_unbuild_ssp_ascb(ascb);
 	default:
 		break;
@@ -586,17 +586,17 @@ int asd_execute_task(struct sas_task *task, const int num,
 	list_for_each_entry(a, &alist, list) {
 		t = a->uldd_task;
 		a->uldd_timer = 1;
-		if (t->task_proto & SAS_PROTO_STP)
-			t->task_proto = SAS_PROTO_STP;
+		if (t->task_proto & SAS_PROTOCOL_STP)
+			t->task_proto = SAS_PROTOCOL_STP;
 		switch (t->task_proto) {
-		case SATA_PROTO:
-		case SAS_PROTO_STP:
+		case SAS_PROTOCOL_SATA:
+		case SAS_PROTOCOL_STP:
 			res = asd_build_ata_ascb(a, t, gfp_flags);
 			break;
-		case SAS_PROTO_SMP:
+		case SAS_PROTOCOL_SMP:
 			res = asd_build_smp_ascb(a, t, gfp_flags);
 			break;
-		case SAS_PROTO_SSP:
+		case SAS_PROTOCOL_SSP:
 			res = asd_build_ssp_ascb(a, t, gfp_flags);
 			break;
 		default:
@@ -633,14 +633,14 @@ out_err_unmap:
 			t->task_state_flags &= ~SAS_TASK_AT_INITIATOR;
 			spin_unlock_irqrestore(&t->task_state_lock, flags);
 			switch (t->task_proto) {
-			case SATA_PROTO:
-			case SAS_PROTO_STP:
+			case SAS_PROTOCOL_SATA:
+			case SAS_PROTOCOL_STP:
 				asd_unbuild_ata_ascb(a);
 				break;
-			case SAS_PROTO_SMP:
+			case SAS_PROTOCOL_SMP:
 				asd_unbuild_smp_ascb(a);
 				break;
-			case SAS_PROTO_SSP:
+			case SAS_PROTOCOL_SSP:
 				asd_unbuild_ssp_ascb(a);
 			default:
 				break;
diff --git a/drivers/scsi/aic94xx/aic94xx_tmf.c b/drivers/scsi/aic94xx/aic94xx_tmf.c
index c0d0b7d7a8ce..87b2f6e6adfe 100644
--- a/drivers/scsi/aic94xx/aic94xx_tmf.c
+++ b/drivers/scsi/aic94xx/aic94xx_tmf.c
@@ -372,21 +372,21 @@ int asd_abort_task(struct sas_task *task)
 	scb->header.opcode = ABORT_TASK;
 
 	switch (task->task_proto) {
-	case SATA_PROTO:
-	case SAS_PROTO_STP:
+	case SAS_PROTOCOL_SATA:
+	case SAS_PROTOCOL_STP:
 		scb->abort_task.proto_conn_rate = (1 << 5); /* STP */
 		break;
-	case SAS_PROTO_SSP:
+	case SAS_PROTOCOL_SSP:
 		scb->abort_task.proto_conn_rate  = (1 << 4); /* SSP */
 		scb->abort_task.proto_conn_rate |= task->dev->linkrate;
 		break;
-	case SAS_PROTO_SMP:
+	case SAS_PROTOCOL_SMP:
 		break;
 	default:
 		break;
 	}
 
-	if (task->task_proto == SAS_PROTO_SSP) {
+	if (task->task_proto == SAS_PROTOCOL_SSP) {
 		scb->abort_task.ssp_frame.frame_type = SSP_TASK;
 		memcpy(scb->abort_task.ssp_frame.hashed_dest_addr,
 		       task->dev->hashed_sas_addr, HASHED_SAS_ADDR_SIZE);
@@ -512,7 +512,7 @@ static int asd_initiate_ssp_tmf(struct domain_device *dev, u8 *lun,
 	int res = 1;
 	struct scb *scb;
 
-	if (!(dev->tproto & SAS_PROTO_SSP))
+	if (!(dev->tproto & SAS_PROTOCOL_SSP))
 		return TMF_RESP_FUNC_ESUPP;
 
 	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL);
diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c
index 5f3a0d7b18de..31b9af224243 100644
--- a/drivers/scsi/libsas/sas_discover.c
+++ b/drivers/scsi/libsas/sas_discover.c
@@ -98,7 +98,7 @@ static int sas_get_port_device(struct asd_sas_port *port)
 			dev->dev_type = SATA_PM;
 		else
 			dev->dev_type = SATA_DEV;
-		dev->tproto = SATA_PROTO;
+		dev->tproto = SAS_PROTOCOL_SATA;
 	} else {
 		struct sas_identify_frame *id =
 			(struct sas_identify_frame *) dev->frame_rcvd;
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index 8727436b222d..27674fe468f5 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -656,9 +656,9 @@ static struct domain_device *sas_ex_discover_end_dev(
 	sas_ex_get_linkrate(parent, child, phy);
 
 #ifdef CONFIG_SCSI_SAS_ATA
-	if ((phy->attached_tproto & SAS_PROTO_STP) || phy->attached_sata_dev) {
+	if ((phy->attached_tproto & SAS_PROTOCOL_STP) || phy->attached_sata_dev) {
 		child->dev_type = SATA_DEV;
-		if (phy->attached_tproto & SAS_PROTO_STP)
+		if (phy->attached_tproto & SAS_PROTOCOL_STP)
 			child->tproto = phy->attached_tproto;
 		if (phy->attached_sata_dev)
 			child->tproto |= SATA_DEV;
@@ -695,7 +695,7 @@ static struct domain_device *sas_ex_discover_end_dev(
 		}
 	} else
 #endif
-	  if (phy->attached_tproto & SAS_PROTO_SSP) {
+	  if (phy->attached_tproto & SAS_PROTOCOL_SSP) {
 		child->dev_type = SAS_END_DEV;
 		rphy = sas_end_device_alloc(phy->port);
 		/* FIXME: error handling */
diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h
index 2b8213b1832d..baa0666ffa80 100644
--- a/drivers/scsi/libsas/sas_internal.h
+++ b/drivers/scsi/libsas/sas_internal.h
@@ -45,7 +45,7 @@
 void sas_scsi_recover_host(struct Scsi_Host *shost);
 
 int sas_show_class(enum sas_class class, char *buf);
-int sas_show_proto(enum sas_proto proto, char *buf);
+int sas_show_proto(enum sas_protocol proto, char *buf);
 int sas_show_linkrate(enum sas_linkrate linkrate, char *buf);
 int sas_show_oob_mode(enum sas_oob_mode oob_mode, char *buf);
 
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index a466c2cb8955..e42374cebf95 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -122,8 +122,8 @@ struct ex_phy {
 	u8   attached_sata_dev:1;
 	u8   attached_sata_ps:1;
 
-	enum sas_proto attached_tproto;
-	enum sas_proto attached_iproto;
+	enum sas_protocol attached_tproto;
+	enum sas_protocol attached_iproto;
 
 	u8   attached_sas_addr[SAS_ADDR_SIZE];
 	u8   attached_phy_id;
@@ -191,8 +191,8 @@ struct domain_device {
 
         struct list_head dev_list_node;
 
-        enum sas_proto    iproto;
-        enum sas_proto    tproto;
+        enum sas_protocol    iproto;
+        enum sas_protocol    tproto;
 
         struct sas_rphy *rphy;
 
@@ -245,8 +245,8 @@ struct asd_sas_port {
 	enum sas_class   class;
 	u8               sas_addr[SAS_ADDR_SIZE];
 	u8               attached_sas_addr[SAS_ADDR_SIZE];
-	enum sas_proto   iproto;
-	enum sas_proto   tproto;
+	enum sas_protocol   iproto;
+	enum sas_protocol   tproto;
 
 	enum sas_oob_mode oob_mode;
 
@@ -289,8 +289,8 @@ struct asd_sas_phy {
 
 	int            id;	  /* must be set */
 	enum sas_class class;
-	enum sas_proto iproto;
-	enum sas_proto tproto;
+	enum sas_protocol iproto;
+	enum sas_protocol tproto;
 
 	enum sas_phy_type  type;
 	enum sas_phy_role  role;
@@ -537,7 +537,7 @@ struct sas_task {
 	spinlock_t   task_state_lock;
 	unsigned     task_state_flags;
 
-	enum   sas_proto      task_proto;
+	enum   sas_protocol      task_proto;
 
 	/* Used by the discovery code. */
 	struct timer_list     timer;
diff --git a/include/scsi/sas.h b/include/scsi/sas.h
index 2f4b6afa34fc..e9fd02281381 100644
--- a/include/scsi/sas.h
+++ b/include/scsi/sas.h
@@ -102,13 +102,12 @@ enum sas_dev_type {
 	SATA_PM_PORT= 8,
 };
 
-/* Partly from IDENTIFY address frame. */
-enum sas_proto {
-	SATA_PROTO    = 1,
-	SAS_PROTO_SMP = 2,	  /* protocol */
-	SAS_PROTO_STP = 4,	  /* protocol */
-	SAS_PROTO_SSP = 8,	  /* protocol */
-	SAS_PROTO_ALL = 0xE,
+enum sas_protocol {
+	SAS_PROTOCOL_SATA		= 0x01,
+	SAS_PROTOCOL_SMP		= 0x02,
+	SAS_PROTOCOL_STP		= 0x04,
+	SAS_PROTOCOL_SSP		= 0x08,
+	SAS_PROTOCOL_ALL		= 0x0E,
 };
 
 /* From the spec; local phys only */
diff --git a/include/scsi/scsi_transport_sas.h b/include/scsi/scsi_transport_sas.h
index abdfd2e27dd7..007d92904207 100644
--- a/include/scsi/scsi_transport_sas.h
+++ b/include/scsi/scsi_transport_sas.h
@@ -4,6 +4,7 @@
 #include <linux/transport_class.h>
 #include <linux/types.h>
 #include <linux/mutex.h>
+#include <scsi/sas.h>
 
 struct scsi_transport_template;
 struct sas_rphy;
@@ -16,13 +17,6 @@ enum sas_device_type {
 	SAS_FANOUT_EXPANDER_DEVICE,
 };
 
-enum sas_protocol {
-	SAS_PROTOCOL_SATA		= 0x01,
-	SAS_PROTOCOL_SMP		= 0x02,
-	SAS_PROTOCOL_STP		= 0x04,
-	SAS_PROTOCOL_SSP		= 0x08,
-};
-
 static inline int sas_protocol_ata(enum sas_protocol proto)
 {
 	return ((proto & SAS_PROTOCOL_SATA) ||
-- 
cgit v1.2.3


From 8ee24023ef43c77cdaea1add00e1e45cc4fc3261 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <djwong@us.ibm.com>
Date: Mon, 5 Nov 2007 11:52:14 -0800
Subject: [SCSI] libsas: Fix various sparse complaints

Annotate sas_queuecommand with locking details, and clean up a few
more sparse warnings about static/non-static declarations.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libsas/sas_scsi_host.c | 6 +++++-
 include/scsi/libsas.h               | 4 +---
 2 files changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 7663841eb4cf..b784089b18d0 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -200,6 +200,10 @@ int sas_queue_up(struct sas_task *task)
  */
 int sas_queuecommand(struct scsi_cmnd *cmd,
 		     void (*scsi_done)(struct scsi_cmnd *))
+	__releases(host->host_lock)
+	__acquires(dev->sata_dev.ap->lock)
+	__releases(dev->sata_dev.ap->lock)
+	__acquires(host->host_lock)
 {
 	int res = 0;
 	struct domain_device *dev = cmd_to_domain_dev(cmd);
@@ -410,7 +414,7 @@ static int sas_recover_I_T(struct domain_device *dev)
 }
 
 /* Find the sas_phy that's attached to this device */
-struct sas_phy *find_local_sas_phy(struct domain_device *dev)
+static struct sas_phy *find_local_sas_phy(struct domain_device *dev)
 {
 	struct domain_device *pdev = dev->parent;
 	struct ex_phy *exphy = NULL;
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index e42374cebf95..93248cd48373 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -563,7 +563,7 @@ struct sas_task {
 	struct work_struct abort_work;
 };
 
-
+extern struct kmem_cache *sas_task_cache;
 
 #define SAS_TASK_STATE_PENDING      1
 #define SAS_TASK_STATE_DONE         2
@@ -573,7 +573,6 @@ struct sas_task {
 
 static inline struct sas_task *sas_alloc_task(gfp_t flags)
 {
-	extern struct kmem_cache *sas_task_cache;
 	struct sas_task *task = kmem_cache_zalloc(sas_task_cache, flags);
 
 	if (task) {
@@ -590,7 +589,6 @@ static inline struct sas_task *sas_alloc_task(gfp_t flags)
 static inline void sas_free_task(struct sas_task *task)
 {
 	if (task) {
-		extern struct kmem_cache *sas_task_cache;
 		BUG_ON(!list_empty(&task->list));
 		kmem_cache_free(sas_task_cache, task);
 	}
-- 
cgit v1.2.3


From 25d7c363f2663fe399e623c9bd819258c9760bdc Mon Sep 17 00:00:00 2001
From: Tony Battersby <tonyb@cybernetics.com>
Date: Mon, 12 Nov 2007 10:00:44 -0500
Subject: [SCSI] move single_lun flag from scsi_device to scsi_target

Some SCSI tape medium changers that need the BLIST_SINGLELUN flag have
the medium changer at one LUN and the tape drive at a different LUN.
The inquiry string of the tape drive may be different from that of the
medium changer.  In order for single_lun to be effective, every
scsi_device under a given scsi_target must have it set.  This means that
there needs to be a blacklist entry for BOTH the medium changer AND the
tape drive, which is impractical because some medium changers may be
paired with a variety of different tape drive models.  It makes more
sense to put the single_lun flag in scsi_target instead of scsi_device,
which causes every device at a given target ID to inherit the single_lun
flag from one LUN.  This makes it possible to blacklist just the medium
changer and not the tape drive.

Signed-off-by: Tony Battersby <tonyb@cybernetics.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_lib.c    | 4 ++--
 drivers/scsi/scsi_scan.c   | 2 +-
 include/scsi/scsi_device.h | 6 +++---
 3 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 199b02452259..b0c59ae63a2f 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -526,7 +526,7 @@ static void scsi_run_queue(struct request_queue *q)
 	struct Scsi_Host *shost = sdev->host;
 	unsigned long flags;
 
-	if (sdev->single_lun)
+	if (scsi_target(sdev)->single_lun)
 		scsi_single_lun_run(sdev);
 
 	spin_lock_irqsave(shost->host_lock, flags);
@@ -1559,7 +1559,7 @@ static void scsi_request_fn(struct request_queue *q)
 
 		if (!scsi_host_queue_ready(q, shost, sdev))
 			goto not_ready;
-		if (sdev->single_lun) {
+		if (scsi_target(sdev)->single_lun) {
 			if (scsi_target(sdev)->starget_sdev_user &&
 			    scsi_target(sdev)->starget_sdev_user != sdev)
 				goto not_ready;
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 7e8b2045ccfa..ada72af0a6dd 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -865,7 +865,7 @@ static int scsi_add_lun(struct scsi_device *sdev, unsigned char *inq_result,
 		sdev->no_start_on_add = 1;
 
 	if (*bflags & BLIST_SINGLELUN)
-		sdev->single_lun = 1;
+		scsi_target(sdev)->single_lun = 1;
 
 	sdev->use_10_for_rw = 1;
 
diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index 6c2d80b36aa1..45bb12b54175 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -122,9 +122,6 @@ struct scsi_device {
 	unsigned tagged_supported:1;	/* Supports SCSI-II tagged queuing */
 	unsigned simple_tags:1;	/* simple queue tag messages are enabled */
 	unsigned ordered_tags:1;/* ordered queue tag messages are enabled */
-	unsigned single_lun:1;	/* Indicates we should only allow I/O to
-				 * one of the luns for the device at a 
-				 * time. */
 	unsigned was_reset:1;	/* There was a bus reset on the bus for 
 				 * this device */
 	unsigned expecting_cc_ua:1; /* Expecting a CHECK_CONDITION/UNIT_ATTN
@@ -202,6 +199,9 @@ struct scsi_target {
 	unsigned int		id; /* target id ... replace
 				     * scsi_device.id eventually */
 	unsigned int		create:1; /* signal that it needs to be added */
+	unsigned int		single_lun:1;	/* Indicates we should only
+						 * allow I/O to one of the luns
+						 * for the device at a time. */
 	unsigned int		pdt_1f_for_no_lun;	/* PDT = 0x1f */
 						/* means no lun present */
 
-- 
cgit v1.2.3


From 001aac257cf8adbe90cdcba6e07f8d12dfc8fa6b Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Sun, 2 Dec 2007 19:10:40 +0200
Subject: [SCSI] sd,sr: add early detection of medium not present

The current scsi_test_unit_ready() is updated to return sense code
information (in struct scsi_sense_hdr).  The sd and sr drivers are
changed to interpret the sense code return asc 0x3a as no media and
adjust the device status accordingly.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_ioctl.c  |  2 +-
 drivers/scsi/scsi_lib.c    | 46 ++++++++++++++++++++++++++++++++++++++--------
 drivers/scsi/sd.c          | 13 ++++++++++---
 drivers/scsi/sr.c          | 19 +++++++++++++------
 include/scsi/scsi_device.h |  2 +-
 5 files changed, 63 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c
index 83e144716901..28b19ef26309 100644
--- a/drivers/scsi/scsi_ioctl.c
+++ b/drivers/scsi/scsi_ioctl.c
@@ -244,7 +244,7 @@ int scsi_ioctl(struct scsi_device *sdev, int cmd, void __user *arg)
 		return scsi_set_medium_removal(sdev, SCSI_REMOVAL_ALLOW);
 	case SCSI_IOCTL_TEST_UNIT_READY:
 		return scsi_test_unit_ready(sdev, IOCTL_NORMAL_TIMEOUT,
-					    NORMAL_RETRIES);
+					    NORMAL_RETRIES, NULL);
 	case SCSI_IOCTL_START_UNIT:
 		scsi_cmd[0] = START_STOP;
 		scsi_cmd[1] = 0;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index aa17e718666e..db52222885b7 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1973,27 +1973,57 @@ scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage,
 }
 EXPORT_SYMBOL(scsi_mode_sense);
 
+/**
+ *	scsi_test_unit_ready - test if unit is ready
+ *	@sdev:	scsi device to change the state of.
+ *	@timeout: command timeout
+ *	@retries: number of retries before failing
+ *	@sshdr_external: Optional pointer to struct scsi_sense_hdr for
+ *		returning sense. Make sure that this is cleared before passing
+ *		in.
+ *
+ *	Returns zero if unsuccessful or an error if TUR failed.  For
+ *	removable media, a return of NOT_READY or UNIT_ATTENTION is
+ *	translated to success, with the ->changed flag updated.
+ **/
 int
-scsi_test_unit_ready(struct scsi_device *sdev, int timeout, int retries)
+scsi_test_unit_ready(struct scsi_device *sdev, int timeout, int retries,
+		     struct scsi_sense_hdr *sshdr_external)
 {
 	char cmd[] = {
 		TEST_UNIT_READY, 0, 0, 0, 0, 0,
 	};
-	struct scsi_sense_hdr sshdr;
+	struct scsi_sense_hdr *sshdr;
 	int result;
-	
-	result = scsi_execute_req(sdev, cmd, DMA_NONE, NULL, 0, &sshdr,
-				  timeout, retries);
+
+	if (!sshdr_external)
+		sshdr = kzalloc(sizeof(*sshdr), GFP_KERNEL);
+	else
+		sshdr = sshdr_external;
+
+	/* try to eat the UNIT_ATTENTION if there are enough retries */
+	do {
+		result = scsi_execute_req(sdev, cmd, DMA_NONE, NULL, 0, sshdr,
+					  timeout, retries);
+	} while ((driver_byte(result) & DRIVER_SENSE) &&
+		 sshdr && sshdr->sense_key == UNIT_ATTENTION &&
+		 --retries);
+
+	if (!sshdr)
+		/* could not allocate sense buffer, so can't process it */
+		return result;
 
 	if ((driver_byte(result) & DRIVER_SENSE) && sdev->removable) {
 
-		if ((scsi_sense_valid(&sshdr)) &&
-		    ((sshdr.sense_key == UNIT_ATTENTION) ||
-		     (sshdr.sense_key == NOT_READY))) {
+		if ((scsi_sense_valid(sshdr)) &&
+		    ((sshdr->sense_key == UNIT_ATTENTION) ||
+		     (sshdr->sense_key == NOT_READY))) {
 			sdev->changed = 1;
 			result = 0;
 		}
 	}
+	if (!sshdr_external)
+		kfree(sshdr);
 	return result;
 }
 EXPORT_SYMBOL(scsi_test_unit_ready);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 18343a6acd8e..212f6bcfd457 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -736,6 +736,7 @@ static int sd_media_changed(struct gendisk *disk)
 {
 	struct scsi_disk *sdkp = scsi_disk(disk);
 	struct scsi_device *sdp = sdkp->device;
+	struct scsi_sense_hdr *sshdr = NULL;
 	int retval;
 
 	SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp, "sd_media_changed\n"));
@@ -766,8 +767,11 @@ static int sd_media_changed(struct gendisk *disk)
 	 */
 	retval = -ENODEV;
 
-	if (scsi_block_when_processing_errors(sdp))
-		retval = scsi_test_unit_ready(sdp, SD_TIMEOUT, SD_MAX_RETRIES);
+	if (scsi_block_when_processing_errors(sdp)) {
+		sshdr  = kzalloc(sizeof(*sshdr), GFP_KERNEL);
+		retval = scsi_test_unit_ready(sdp, SD_TIMEOUT, SD_MAX_RETRIES,
+					      sshdr);
+	}
 
 	/*
 	 * Unable to test, unit probably not ready.   This usually
@@ -775,7 +779,9 @@ static int sd_media_changed(struct gendisk *disk)
 	 * and we will figure it out later once the drive is
 	 * available again.
 	 */
-	if (retval) {
+	if (retval || (scsi_sense_valid(sshdr) &&
+		       /* 0x3a is medium not present */
+		       sshdr->asc == 0x3a)) {
 		set_media_not_present(sdkp);
 		retval = 1;
 		goto out;
@@ -794,6 +800,7 @@ out:
 	if (retval != sdkp->previous_state)
 		sdev_evt_send_simple(sdp, SDEV_EVT_MEDIA_CHANGE, GFP_KERNEL);
 	sdkp->previous_state = retval;
+	kfree(sshdr);
 	return retval;
 }
 
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index 7702681d93f9..896be4ab285d 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -179,18 +179,24 @@ static int sr_media_change(struct cdrom_device_info *cdi, int slot)
 {
 	struct scsi_cd *cd = cdi->handle;
 	int retval;
+	struct scsi_sense_hdr *sshdr;
 
 	if (CDSL_CURRENT != slot) {
 		/* no changer support */
 		return -EINVAL;
 	}
 
-	retval = scsi_test_unit_ready(cd->device, SR_TIMEOUT, MAX_RETRIES);
-	if (retval) {
-		/* Unable to test, unit probably not ready.  This usually
-		 * means there is no disc in the drive.  Mark as changed,
-		 * and we will figure it out later once the drive is
-		 * available again.  */
+	sshdr =  kzalloc(sizeof(*sshdr), GFP_KERNEL);
+	retval = scsi_test_unit_ready(cd->device, SR_TIMEOUT, MAX_RETRIES,
+				      sshdr);
+	if (retval || (scsi_sense_valid(sshdr) &&
+		       /* 0x3a is medium not present */
+		       sshdr->asc == 0x3a)) {
+		/* Media not present or unable to test, unit probably not
+		 * ready. This usually means there is no disc in the drive.
+		 * Mark as changed, and we will figure it out later once
+		 * the drive is available again.
+		 */
 		cd->device->changed = 1;
 		/* This will force a flush, if called from check_disk_change */
 		retval = 1;
@@ -213,6 +219,7 @@ out:
 		sdev_evt_send_simple(cd->device, SDEV_EVT_MEDIA_CHANGE,
 				     GFP_KERNEL);
 	cd->previous_state = retval;
+	kfree(sshdr);
 
 	return retval;
 }
diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index 45bb12b54175..e0c645ac5014 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -295,7 +295,7 @@ extern int scsi_mode_select(struct scsi_device *sdev, int pf, int sp,
 			    struct scsi_mode_data *data,
 			    struct scsi_sense_hdr *);
 extern int scsi_test_unit_ready(struct scsi_device *sdev, int timeout,
-				int retries);
+				int retries, struct scsi_sense_hdr *sshdr);
 extern int scsi_device_set_state(struct scsi_device *sdev,
 				 enum scsi_device_state state);
 extern struct scsi_event *sdev_evt_alloc(enum scsi_device_event evt_type,
-- 
cgit v1.2.3


From 843c0a8a76078cf961b244b839683d0667313740 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 13 Dec 2007 12:43:20 -0600
Subject: [SCSI] libiscsi, iscsi_tcp: add device support

This patch adds logical unit reset support. This should work for ib_iser,
but I have not finished testing that driver so it is not hooked in yet.

This patch also temporarily reverts the iscsi_tcp r2t write out patch.
That code is completely rewritten in this patchset.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/infiniband/ulp/iser/iscsi_iser.c |   6 -
 drivers/scsi/iscsi_tcp.c                 | 151 +++++-----
 drivers/scsi/iscsi_tcp.h                 |  34 +--
 drivers/scsi/libiscsi.c                  | 494 ++++++++++++++++++-------------
 drivers/scsi/scsi_transport_iscsi.c      |   4 +-
 include/scsi/iscsi_if.h                  |   2 +
 include/scsi/iscsi_proto.h               |   2 +
 include/scsi/libiscsi.h                  |  25 +-
 8 files changed, 396 insertions(+), 322 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index bad8dacafd10..2eadb6d0ad8b 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -220,12 +220,6 @@ iscsi_iser_ctask_xmit(struct iscsi_conn *conn,
 	debug_scsi("ctask deq [cid %d itt 0x%x]\n",
 		   conn->id, ctask->itt);
 
-	/*
-	 * serialize with TMF AbortTask
-	 */
-	if (ctask->mtask)
-		return error;
-
 	/* Send the cmd PDU */
 	if (!iser_ctask->command_sent) {
 		error = iser_send_command(conn, ctask);
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 57ce2251abc8..4b226b88b68a 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -197,7 +197,7 @@ iscsi_tcp_cleanup_ctask(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	if (unlikely(!sc))
 		return;
 
-	tcp_ctask->xmstate = XMSTATE_VALUE_IDLE;
+	tcp_ctask->xmstate = XMSTATE_IDLE;
 	tcp_ctask->r2t = NULL;
 }
 
@@ -369,8 +369,7 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	spin_lock(&session->lock);
 	iscsi_update_cmdsn(session, (struct iscsi_nopin*)rhdr);
 
-	if (!ctask->sc || ctask->mtask ||
-	     session->state != ISCSI_STATE_LOGGED_IN) {
+	if (!ctask->sc || session->state != ISCSI_STATE_LOGGED_IN) {
 		printk(KERN_INFO "iscsi_tcp: dropping R2T itt %d in "
 		       "recovery...\n", ctask->itt);
 		spin_unlock(&session->lock);
@@ -409,11 +408,10 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 
 	tcp_ctask->exp_datasn = r2tsn + 1;
 	__kfifo_put(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*));
-	set_bit(XMSTATE_BIT_SOL_HDR_INIT, &tcp_ctask->xmstate);
-	list_move_tail(&ctask->running, &conn->xmitqueue);
-
-	scsi_queue_work(session->host, &conn->xmitwork);
+	tcp_ctask->xmstate |= XMSTATE_SOL_HDR_INIT;
 	conn->r2t_pdus_cnt++;
+
+	iscsi_requeue_ctask(ctask);
 	spin_unlock(&session->lock);
 
 	return 0;
@@ -1254,7 +1252,7 @@ static void iscsi_set_padding(struct iscsi_tcp_cmd_task *tcp_ctask,
 
 	tcp_ctask->pad_count = ISCSI_PAD_LEN - tcp_ctask->pad_count;
 	debug_scsi("write padding %d bytes\n", tcp_ctask->pad_count);
-	set_bit(XMSTATE_BIT_W_PAD, &tcp_ctask->xmstate);
+	tcp_ctask->xmstate |= XMSTATE_W_PAD;
 }
 
 /**
@@ -1269,7 +1267,7 @@ iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask)
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 
 	BUG_ON(__kfifo_len(tcp_ctask->r2tqueue));
-	tcp_ctask->xmstate = 1 << XMSTATE_BIT_CMD_HDR_INIT;
+	tcp_ctask->xmstate = XMSTATE_CMD_HDR_INIT;
 }
 
 /**
@@ -1283,10 +1281,10 @@ iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask)
  *	xmit.
  *
  *	Management xmit state machine consists of these states:
- *		XMSTATE_BIT_IMM_HDR_INIT - calculate digest of PDU Header
- *		XMSTATE_BIT_IMM_HDR      - PDU Header xmit in progress
- *		XMSTATE_BIT_IMM_DATA     - PDU Data xmit in progress
- *		XMSTATE_VALUE_IDLE       - management PDU is done
+ *		XMSTATE_IMM_HDR_INIT	- calculate digest of PDU Header
+ *		XMSTATE_IMM_HDR 	- PDU Header xmit in progress
+ *		XMSTATE_IMM_DATA 	- PDU Data xmit in progress
+ *		XMSTATE_IDLE		- management PDU is done
  **/
 static int
 iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
@@ -1297,12 +1295,12 @@ iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 	debug_scsi("mtask deq [cid %d state %x itt 0x%x]\n",
 		conn->id, tcp_mtask->xmstate, mtask->itt);
 
-	if (test_bit(XMSTATE_BIT_IMM_HDR_INIT, &tcp_mtask->xmstate)) {
+	if (tcp_mtask->xmstate & XMSTATE_IMM_HDR_INIT) {
 		iscsi_buf_init_iov(&tcp_mtask->headbuf, (char*)mtask->hdr,
 				   sizeof(struct iscsi_hdr));
 
 		if (mtask->data_count) {
-			set_bit(XMSTATE_BIT_IMM_DATA, &tcp_mtask->xmstate);
+			tcp_mtask->xmstate |= XMSTATE_IMM_DATA;
 			iscsi_buf_init_iov(&tcp_mtask->sendbuf,
 					   (char*)mtask->data,
 					   mtask->data_count);
@@ -1315,20 +1313,21 @@ iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 					(u8*)tcp_mtask->hdrext);
 
 		tcp_mtask->sent = 0;
-		clear_bit(XMSTATE_BIT_IMM_HDR_INIT, &tcp_mtask->xmstate);
-		set_bit(XMSTATE_BIT_IMM_HDR, &tcp_mtask->xmstate);
+		tcp_mtask->xmstate &= ~XMSTATE_IMM_HDR_INIT;
+		tcp_mtask->xmstate |= XMSTATE_IMM_HDR;
 	}
 
-	if (test_bit(XMSTATE_BIT_IMM_HDR, &tcp_mtask->xmstate)) {
+	if (tcp_mtask->xmstate & XMSTATE_IMM_HDR) {
 		rc = iscsi_sendhdr(conn, &tcp_mtask->headbuf,
 				   mtask->data_count);
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_IMM_HDR, &tcp_mtask->xmstate);
+		tcp_mtask->xmstate &= ~XMSTATE_IMM_HDR;
 	}
 
-	if (test_and_clear_bit(XMSTATE_BIT_IMM_DATA, &tcp_mtask->xmstate)) {
+	if (tcp_mtask->xmstate & XMSTATE_IMM_DATA) {
 		BUG_ON(!mtask->data_count);
+		tcp_mtask->xmstate &= ~XMSTATE_IMM_DATA;
 		/* FIXME: implement.
 		 * Virtual buffer could be spreaded across multiple pages...
 		 */
@@ -1338,13 +1337,13 @@ iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 			rc = iscsi_sendpage(conn, &tcp_mtask->sendbuf,
 					&mtask->data_count, &tcp_mtask->sent);
 			if (rc) {
-				set_bit(XMSTATE_BIT_IMM_DATA, &tcp_mtask->xmstate);
+				tcp_mtask->xmstate |= XMSTATE_IMM_DATA;
 				return rc;
 			}
 		} while (mtask->data_count);
 	}
 
-	BUG_ON(tcp_mtask->xmstate != XMSTATE_VALUE_IDLE);
+	BUG_ON(tcp_mtask->xmstate != XMSTATE_IDLE);
 	if (mtask->hdr->itt == RESERVED_ITT) {
 		struct iscsi_session *session = conn->session;
 
@@ -1364,7 +1363,7 @@ iscsi_send_cmd_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 	int rc = 0;
 
-	if (test_bit(XMSTATE_BIT_CMD_HDR_INIT, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_CMD_HDR_INIT) {
 		tcp_ctask->sent = 0;
 		tcp_ctask->sg_count = 0;
 		tcp_ctask->exp_datasn = 0;
@@ -1389,21 +1388,21 @@ iscsi_send_cmd_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 		if (conn->hdrdgst_en)
 			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
 					 (u8*)tcp_ctask->hdrext);
-		clear_bit(XMSTATE_BIT_CMD_HDR_INIT, &tcp_ctask->xmstate);
-		set_bit(XMSTATE_BIT_CMD_HDR_XMIT, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_CMD_HDR_INIT;
+		tcp_ctask->xmstate |= XMSTATE_CMD_HDR_XMIT;
 	}
 
-	if (test_bit(XMSTATE_BIT_CMD_HDR_XMIT, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_CMD_HDR_XMIT) {
 		rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->imm_count);
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_CMD_HDR_XMIT, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_CMD_HDR_XMIT;
 
 		if (sc->sc_data_direction != DMA_TO_DEVICE)
 			return 0;
 
 		if (ctask->imm_count) {
-			set_bit(XMSTATE_BIT_IMM_DATA, &tcp_ctask->xmstate);
+			tcp_ctask->xmstate |= XMSTATE_IMM_DATA;
 			iscsi_set_padding(tcp_ctask, ctask->imm_count);
 
 			if (ctask->conn->datadgst_en) {
@@ -1413,10 +1412,9 @@ iscsi_send_cmd_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 			}
 		}
 
-		if (ctask->unsol_count) {
-			set_bit(XMSTATE_BIT_UNS_HDR, &tcp_ctask->xmstate);
-			set_bit(XMSTATE_BIT_UNS_INIT, &tcp_ctask->xmstate);
-		}
+		if (ctask->unsol_count)
+			tcp_ctask->xmstate |=
+					XMSTATE_UNS_HDR | XMSTATE_UNS_INIT;
 	}
 	return rc;
 }
@@ -1428,25 +1426,25 @@ iscsi_send_padding(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 	int sent = 0, rc;
 
-	if (test_bit(XMSTATE_BIT_W_PAD, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_W_PAD) {
 		iscsi_buf_init_iov(&tcp_ctask->sendbuf, (char*)&tcp_ctask->pad,
 				   tcp_ctask->pad_count);
 		if (conn->datadgst_en)
 			crypto_hash_update(&tcp_conn->tx_hash,
 					   &tcp_ctask->sendbuf.sg,
 					   tcp_ctask->sendbuf.sg.length);
-	} else if (!test_bit(XMSTATE_BIT_W_RESEND_PAD, &tcp_ctask->xmstate))
+	} else if (!(tcp_ctask->xmstate & XMSTATE_W_RESEND_PAD))
 		return 0;
 
-	clear_bit(XMSTATE_BIT_W_PAD, &tcp_ctask->xmstate);
-	clear_bit(XMSTATE_BIT_W_RESEND_PAD, &tcp_ctask->xmstate);
+	tcp_ctask->xmstate &= ~XMSTATE_W_PAD;
+	tcp_ctask->xmstate &= ~XMSTATE_W_RESEND_PAD;
 	debug_scsi("sending %d pad bytes for itt 0x%x\n",
 		   tcp_ctask->pad_count, ctask->itt);
 	rc = iscsi_sendpage(conn, &tcp_ctask->sendbuf, &tcp_ctask->pad_count,
 			   &sent);
 	if (rc) {
 		debug_scsi("padding send failed %d\n", rc);
-		set_bit(XMSTATE_BIT_W_RESEND_PAD, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate |= XMSTATE_W_RESEND_PAD;
 	}
 	return rc;
 }
@@ -1465,11 +1463,11 @@ iscsi_send_digest(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 	tcp_ctask = ctask->dd_data;
 	tcp_conn = conn->dd_data;
 
-	if (!test_bit(XMSTATE_BIT_W_RESEND_DATA_DIGEST, &tcp_ctask->xmstate)) {
+	if (!(tcp_ctask->xmstate & XMSTATE_W_RESEND_DATA_DIGEST)) {
 		crypto_hash_final(&tcp_conn->tx_hash, (u8*)digest);
 		iscsi_buf_init_iov(buf, (char*)digest, 4);
 	}
-	clear_bit(XMSTATE_BIT_W_RESEND_DATA_DIGEST, &tcp_ctask->xmstate);
+	tcp_ctask->xmstate &= ~XMSTATE_W_RESEND_DATA_DIGEST;
 
 	rc = iscsi_sendpage(conn, buf, &tcp_ctask->digest_count, &sent);
 	if (!rc)
@@ -1478,7 +1476,7 @@ iscsi_send_digest(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 	else {
 		debug_scsi("sending digest 0x%x failed for itt 0x%x!\n",
 			  *digest, ctask->itt);
-		set_bit(XMSTATE_BIT_W_RESEND_DATA_DIGEST, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate |= XMSTATE_W_RESEND_DATA_DIGEST;
 	}
 	return rc;
 }
@@ -1526,8 +1524,8 @@ iscsi_send_unsol_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	struct iscsi_data_task *dtask;
 	int rc;
 
-	set_bit(XMSTATE_BIT_UNS_DATA, &tcp_ctask->xmstate);
-	if (test_bit(XMSTATE_BIT_UNS_INIT, &tcp_ctask->xmstate)) {
+	tcp_ctask->xmstate |= XMSTATE_UNS_DATA;
+	if (tcp_ctask->xmstate & XMSTATE_UNS_INIT) {
 		dtask = &tcp_ctask->unsol_dtask;
 
 		iscsi_prep_unsolicit_data_pdu(ctask, &dtask->hdr);
@@ -1537,14 +1535,14 @@ iscsi_send_unsol_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
 					(u8*)dtask->hdrext);
 
-		clear_bit(XMSTATE_BIT_UNS_INIT, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_UNS_INIT;
 		iscsi_set_padding(tcp_ctask, ctask->data_count);
 	}
 
 	rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->data_count);
 	if (rc) {
-		clear_bit(XMSTATE_BIT_UNS_DATA, &tcp_ctask->xmstate);
-		set_bit(XMSTATE_BIT_UNS_HDR, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA;
+		tcp_ctask->xmstate |= XMSTATE_UNS_HDR;
 		return rc;
 	}
 
@@ -1565,15 +1563,16 @@ iscsi_send_unsol_pdu(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 	int rc;
 
-	if (test_and_clear_bit(XMSTATE_BIT_UNS_HDR, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_UNS_HDR) {
 		BUG_ON(!ctask->unsol_count);
+		tcp_ctask->xmstate &= ~XMSTATE_UNS_HDR;
 send_hdr:
 		rc = iscsi_send_unsol_hdr(conn, ctask);
 		if (rc)
 			return rc;
 	}
 
-	if (test_bit(XMSTATE_BIT_UNS_DATA, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_UNS_DATA) {
 		struct iscsi_data_task *dtask = &tcp_ctask->unsol_dtask;
 		int start = tcp_ctask->sent;
 
@@ -1583,14 +1582,14 @@ send_hdr:
 		ctask->unsol_count -= tcp_ctask->sent - start;
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_UNS_DATA, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA;
 		/*
 		 * Done with the Data-Out. Next, check if we need
 		 * to send another unsolicited Data-Out.
 		 */
 		if (ctask->unsol_count) {
 			debug_scsi("sending more uns\n");
-			set_bit(XMSTATE_BIT_UNS_INIT, &tcp_ctask->xmstate);
+			tcp_ctask->xmstate |= XMSTATE_UNS_INIT;
 			goto send_hdr;
 		}
 	}
@@ -1606,7 +1605,7 @@ static int iscsi_send_sol_pdu(struct iscsi_conn *conn,
 	struct iscsi_data_task *dtask;
 	int left, rc;
 
-	if (test_bit(XMSTATE_BIT_SOL_HDR_INIT, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_SOL_HDR_INIT) {
 		if (!tcp_ctask->r2t) {
 			spin_lock_bh(&session->lock);
 			__kfifo_get(tcp_ctask->r2tqueue, (void*)&tcp_ctask->r2t,
@@ -1620,19 +1619,19 @@ send_hdr:
 		if (conn->hdrdgst_en)
 			iscsi_hdr_digest(conn, &r2t->headbuf,
 					(u8*)dtask->hdrext);
-		clear_bit(XMSTATE_BIT_SOL_HDR_INIT, &tcp_ctask->xmstate);
-		set_bit(XMSTATE_BIT_SOL_HDR, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR_INIT;
+		tcp_ctask->xmstate |= XMSTATE_SOL_HDR;
 	}
 
-	if (test_bit(XMSTATE_BIT_SOL_HDR, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_SOL_HDR) {
 		r2t = tcp_ctask->r2t;
 		dtask = &r2t->dtask;
 
 		rc = iscsi_sendhdr(conn, &r2t->headbuf, r2t->data_count);
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_SOL_HDR, &tcp_ctask->xmstate);
-		set_bit(XMSTATE_BIT_SOL_DATA, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR;
+		tcp_ctask->xmstate |= XMSTATE_SOL_DATA;
 
 		if (conn->datadgst_en) {
 			iscsi_data_digest_init(conn->dd_data, tcp_ctask);
@@ -1645,7 +1644,7 @@ send_hdr:
 			r2t->sent);
 	}
 
-	if (test_bit(XMSTATE_BIT_SOL_DATA, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_SOL_DATA) {
 		r2t = tcp_ctask->r2t;
 		dtask = &r2t->dtask;
 
@@ -1654,7 +1653,7 @@ send_hdr:
 				     &dtask->digestbuf, &dtask->digest);
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_SOL_DATA, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_SOL_DATA;
 
 		/*
 		 * Done with this Data-Out. Next, check if we have
@@ -1699,32 +1698,32 @@ send_hdr:
  *	xmit stages.
  *
  *iscsi_send_cmd_hdr()
- *	XMSTATE_BIT_CMD_HDR_INIT - prepare Header and Data buffers Calculate
- *	                           Header Digest
- *	XMSTATE_BIT_CMD_HDR_XMIT - Transmit header in progress
+ *	XMSTATE_CMD_HDR_INIT - prepare Header and Data buffers Calculate
+ *	                       Header Digest
+ *	XMSTATE_CMD_HDR_XMIT - Transmit header in progress
  *
  *iscsi_send_padding
- *	XMSTATE_BIT_W_PAD        - Prepare and send pading
- *	XMSTATE_BIT_W_RESEND_PAD - retry send pading
+ *	XMSTATE_W_PAD        - Prepare and send pading
+ *	XMSTATE_W_RESEND_PAD - retry send pading
  *
  *iscsi_send_digest
- *	XMSTATE_BIT_W_RESEND_DATA_DIGEST - Finalize and send Data Digest
- *	XMSTATE_BIT_W_RESEND_DATA_DIGEST - retry sending digest
+ *	XMSTATE_W_RESEND_DATA_DIGEST - Finalize and send Data Digest
+ *	XMSTATE_W_RESEND_DATA_DIGEST - retry sending digest
  *
  *iscsi_send_unsol_hdr
- *	XMSTATE_BIT_UNS_INIT     - prepare un-solicit data header and digest
- *	XMSTATE_BIT_UNS_HDR      - send un-solicit header
+ *	XMSTATE_UNS_INIT     - prepare un-solicit data header and digest
+ *	XMSTATE_UNS_HDR      - send un-solicit header
  *
  *iscsi_send_unsol_pdu
- *	XMSTATE_BIT_UNS_DATA     - send un-solicit data in progress
+ *	XMSTATE_UNS_DATA     - send un-solicit data in progress
  *
  *iscsi_send_sol_pdu
- *	XMSTATE_BIT_SOL_HDR_INIT - solicit data header and digest initialize
- *	XMSTATE_BIT_SOL_HDR      - send solicit header
- *	XMSTATE_BIT_SOL_DATA     - send solicit data
+ *	XMSTATE_SOL_HDR_INIT - solicit data header and digest initialize
+ *	XMSTATE_SOL_HDR      - send solicit header
+ *	XMSTATE_SOL_DATA     - send solicit data
  *
  *iscsi_tcp_ctask_xmit
- *	XMSTATE_BIT_IMM_DATA     - xmit managment data (??)
+ *	XMSTATE_IMM_DATA     - xmit managment data (??)
  **/
 static int
 iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
@@ -1741,13 +1740,13 @@ iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	if (ctask->sc->sc_data_direction != DMA_TO_DEVICE)
 		return 0;
 
-	if (test_bit(XMSTATE_BIT_IMM_DATA, &tcp_ctask->xmstate)) {
+	if (tcp_ctask->xmstate & XMSTATE_IMM_DATA) {
 		rc = iscsi_send_data(ctask, &tcp_ctask->sendbuf, &tcp_ctask->sg,
 				     &tcp_ctask->sent, &ctask->imm_count,
 				     &tcp_ctask->immbuf, &tcp_ctask->immdigest);
 		if (rc)
 			return rc;
-		clear_bit(XMSTATE_BIT_IMM_DATA, &tcp_ctask->xmstate);
+		tcp_ctask->xmstate &= ~XMSTATE_IMM_DATA;
 	}
 
 	rc = iscsi_send_unsol_pdu(conn, ctask);
@@ -1980,7 +1979,7 @@ static void
 iscsi_tcp_mgmt_init(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 {
 	struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data;
-	tcp_mtask->xmstate = 1 << XMSTATE_BIT_IMM_HDR_INIT;
+	tcp_mtask->xmstate = XMSTATE_IMM_HDR_INIT;
 }
 
 static int
@@ -2226,6 +2225,7 @@ static struct scsi_host_template iscsi_sht = {
 	.max_sectors		= 0xFFFF,
 	.cmd_per_lun		= ISCSI_DEF_CMD_PER_LUN,
 	.eh_abort_handler       = iscsi_eh_abort,
+	.eh_device_reset_handler= iscsi_eh_device_reset,
 	.eh_host_reset_handler	= iscsi_eh_host_reset,
 	.use_clustering         = DISABLE_CLUSTERING,
 	.slave_configure        = iscsi_tcp_slave_configure,
@@ -2257,7 +2257,8 @@ static struct iscsi_transport iscsi_tcp_transport = {
 				  ISCSI_PERSISTENT_ADDRESS |
 				  ISCSI_TARGET_NAME | ISCSI_TPGT |
 				  ISCSI_USERNAME | ISCSI_PASSWORD |
-				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN,
+				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN |
+				  ISCSI_FAST_ABORT,
 	.host_param_mask	= ISCSI_HOST_HWADDRESS | ISCSI_HOST_IPADDRESS |
 				  ISCSI_HOST_INITIATOR_NAME |
 				  ISCSI_HOST_NETDEV_NAME,
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index 68c36cc8997e..7eba44df0a7f 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -32,21 +32,21 @@
 #define IN_PROGRESS_PAD_RECV		0x4
 
 /* xmit state machine */
-#define XMSTATE_VALUE_IDLE			0
-#define XMSTATE_BIT_CMD_HDR_INIT		0
-#define XMSTATE_BIT_CMD_HDR_XMIT		1
-#define XMSTATE_BIT_IMM_HDR			2
-#define XMSTATE_BIT_IMM_DATA			3
-#define XMSTATE_BIT_UNS_INIT			4
-#define XMSTATE_BIT_UNS_HDR			5
-#define XMSTATE_BIT_UNS_DATA			6
-#define XMSTATE_BIT_SOL_HDR			7
-#define XMSTATE_BIT_SOL_DATA			8
-#define XMSTATE_BIT_W_PAD			9
-#define XMSTATE_BIT_W_RESEND_PAD		10
-#define XMSTATE_BIT_W_RESEND_DATA_DIGEST	11
-#define XMSTATE_BIT_IMM_HDR_INIT		12
-#define XMSTATE_BIT_SOL_HDR_INIT		13
+#define XMSTATE_IDLE			0x0
+#define XMSTATE_CMD_HDR_INIT		0x1
+#define XMSTATE_CMD_HDR_XMIT		0x2
+#define XMSTATE_IMM_HDR			0x4
+#define XMSTATE_IMM_DATA		0x8
+#define XMSTATE_UNS_INIT		0x10
+#define XMSTATE_UNS_HDR			0x20
+#define XMSTATE_UNS_DATA		0x40
+#define XMSTATE_SOL_HDR			0x80
+#define XMSTATE_SOL_DATA		0x100
+#define XMSTATE_W_PAD			0x200
+#define XMSTATE_W_RESEND_PAD		0x400
+#define XMSTATE_W_RESEND_DATA_DIGEST	0x800
+#define XMSTATE_IMM_HDR_INIT		0x1000
+#define XMSTATE_SOL_HDR_INIT		0x2000
 
 #define ISCSI_PAD_LEN			4
 #define ISCSI_SG_TABLESIZE		SG_ALL
@@ -122,7 +122,7 @@ struct iscsi_data_task {
 struct iscsi_tcp_mgmt_task {
 	struct iscsi_hdr	hdr;
 	char			hdrext[sizeof(__u32)]; /* Header-Digest */
-	unsigned long		xmstate;	/* mgmt xmit progress */
+	int			xmstate;	/* mgmt xmit progress */
 	struct iscsi_buf	headbuf;	/* header buffer */
 	struct iscsi_buf	sendbuf;	/* in progress buffer */
 	int			sent;
@@ -150,7 +150,7 @@ struct iscsi_tcp_cmd_task {
 	int			pad_count;		/* padded bytes */
 	struct iscsi_buf	headbuf;		/* header buf (xmit) */
 	struct iscsi_buf	sendbuf;		/* in progress buffer*/
-	unsigned long		xmstate;		/* xmit xtate machine */
+	int			xmstate;		/* xmit xtate machine */
 	int			sent;
 	struct scatterlist	*sg;			/* per-cmd SG list  */
 	struct scatterlist	*bad_sg;		/* assert statement */
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 8b57af5baaec..176458f35316 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -86,7 +86,7 @@ iscsi_update_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr)
 		 * xmit thread
 		 */
 		if (!list_empty(&session->leadconn->xmitqueue) ||
-		    __kfifo_len(session->leadconn->mgmtqueue))
+		    !list_empty(&session->leadconn->mgmtqueue))
 			scsi_queue_work(session->host,
 					&session->leadconn->xmitwork);
 	}
@@ -318,15 +318,15 @@ static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 	conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1;
 	conn->tmfrsp_pdus_cnt++;
 
-	if (conn->tmabort_state != TMABORT_INITIAL)
+	if (conn->tmf_state != TMF_QUEUED)
 		return;
 
 	if (tmf->response == ISCSI_TMF_RSP_COMPLETE)
-		conn->tmabort_state = TMABORT_SUCCESS;
+		conn->tmf_state = TMF_SUCCESS;
 	else if (tmf->response == ISCSI_TMF_RSP_NO_TASK)
-		conn->tmabort_state = TMABORT_NOT_FOUND;
+		conn->tmf_state = TMF_NOT_FOUND;
 	else
-		conn->tmabort_state = TMABORT_FAILED;
+		conn->tmf_state = TMF_FAILED;
 	wake_up(&conn->ehwait);
 }
 
@@ -429,7 +429,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			 */
 			if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen))
 				rc = ISCSI_ERR_CONN_FAILED;
-			list_del(&mtask->running);
+			list_del_init(&mtask->running);
 			if (conn->login_mtask != mtask)
 				__kfifo_put(session->mgmtpool.queue,
 					    (void*)&mtask, sizeof(void*));
@@ -451,10 +451,9 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 
 			if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen))
 				rc = ISCSI_ERR_CONN_FAILED;
-			list_del(&mtask->running);
-			if (conn->login_mtask != mtask)
-				__kfifo_put(session->mgmtpool.queue,
-					    (void*)&mtask, sizeof(void*));
+			list_del_init(&mtask->running);
+			__kfifo_put(session->mgmtpool.queue,
+				    (void*)&mtask, sizeof(void*));
 			break;
 		default:
 			rc = ISCSI_ERR_BAD_OPCODE;
@@ -609,7 +608,8 @@ static void iscsi_prep_mtask(struct iscsi_conn *conn,
 		session->tt->init_mgmt_task(conn, mtask);
 
 	debug_scsi("mgmtpdu [op 0x%x hdr->itt 0x%x datalen %d]\n",
-		   hdr->opcode, hdr->itt, mtask->data_count);
+		   hdr->opcode & ISCSI_OPCODE_MASK, hdr->itt,
+		   mtask->data_count);
 }
 
 static int iscsi_xmit_mtask(struct iscsi_conn *conn)
@@ -658,27 +658,35 @@ static int iscsi_check_cmdsn_window_closed(struct iscsi_conn *conn)
 static int iscsi_xmit_ctask(struct iscsi_conn *conn)
 {
 	struct iscsi_cmd_task *ctask = conn->ctask;
-	int rc = 0;
-
-	/*
-	 * serialize with TMF AbortTask
-	 */
-	if (ctask->state == ISCSI_TASK_ABORTING)
-		goto done;
+	int rc;
 
 	__iscsi_get_ctask(ctask);
 	spin_unlock_bh(&conn->session->lock);
 	rc = conn->session->tt->xmit_cmd_task(conn, ctask);
 	spin_lock_bh(&conn->session->lock);
 	__iscsi_put_ctask(ctask);
-
-done:
 	if (!rc)
 		/* done with this ctask */
 		conn->ctask = NULL;
 	return rc;
 }
 
+/**
+ * iscsi_requeue_ctask - requeue ctask to run from session workqueue
+ * @ctask: ctask to requeue
+ *
+ * LLDs that need to run a ctask from the session workqueue should call
+ * this. The session lock must be held.
+ */
+void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask)
+{
+	struct iscsi_conn *conn = ctask->conn;
+
+	list_move_tail(&ctask->running, &conn->requeue);
+	scsi_queue_work(conn->session->host, &conn->xmitwork);
+}
+EXPORT_SYMBOL_GPL(iscsi_requeue_ctask);
+
 /**
  * iscsi_data_xmit - xmit any command into the scheduled connection
  * @conn: iscsi connection
@@ -717,36 +725,27 @@ static int iscsi_data_xmit(struct iscsi_conn *conn)
 	 * overflow us with nop-ins
 	 */
 check_mgmt:
-	while (__kfifo_get(conn->mgmtqueue, (void*)&conn->mtask,
-			   sizeof(void*))) {
+	while (!list_empty(&conn->mgmtqueue)) {
+		conn->mtask = list_entry(conn->mgmtqueue.next,
+					 struct iscsi_mgmt_task, running);
 		iscsi_prep_mtask(conn, conn->mtask);
-		list_add_tail(&conn->mtask->running, &conn->mgmt_run_list);
+		list_move_tail(conn->mgmtqueue.next, &conn->mgmt_run_list);
 		rc = iscsi_xmit_mtask(conn);
 		if (rc)
 			goto again;
 	}
 
-	/* process command queue */
+	/* process pending command queue */
 	while (!list_empty(&conn->xmitqueue)) {
-		/*
-		 * iscsi tcp may readd the task to the xmitqueue to send
-		 * write data
-		 */
+		if (conn->tmf_state == TMF_QUEUED)
+			break;
+
 		conn->ctask = list_entry(conn->xmitqueue.next,
 					 struct iscsi_cmd_task, running);
-		switch (conn->ctask->state) {
-		case ISCSI_TASK_ABORTING:
-			break;
-		case ISCSI_TASK_PENDING:
-			iscsi_prep_scsi_cmd_pdu(conn->ctask);
-			conn->session->tt->init_cmd_task(conn->ctask);
-			/* fall through */
-		default:
-			conn->ctask->state = ISCSI_TASK_RUNNING;
-			break;
-		}
+		iscsi_prep_scsi_cmd_pdu(conn->ctask);
+		conn->session->tt->init_cmd_task(conn->ctask);
+		conn->ctask->state = ISCSI_TASK_RUNNING;
 		list_move_tail(conn->xmitqueue.next, &conn->run_list);
-
 		rc = iscsi_xmit_ctask(conn);
 		if (rc)
 			goto again;
@@ -755,7 +754,22 @@ check_mgmt:
 		 * we need to check the mgmt queue for nops that need to
 		 * be sent to aviod starvation
 		 */
-		if (__kfifo_len(conn->mgmtqueue))
+		if (!list_empty(&conn->mgmtqueue))
+			goto check_mgmt;
+	}
+
+	while (!list_empty(&conn->requeue)) {
+		if (conn->session->fast_abort && conn->tmf_state != TMF_INITIAL)
+			break;
+
+		conn->ctask = list_entry(conn->requeue.next,
+					 struct iscsi_cmd_task, running);
+		conn->ctask->state = ISCSI_TASK_RUNNING;
+		list_move_tail(conn->requeue.next, &conn->run_list);
+		rc = iscsi_xmit_ctask(conn);
+		if (rc)
+			goto again;
+		if (!list_empty(&conn->mgmtqueue))
 			goto check_mgmt;
 	}
 	spin_unlock_bh(&conn->session->lock);
@@ -859,7 +873,6 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
 
 	atomic_set(&ctask->refcount, 1);
 	ctask->state = ISCSI_TASK_PENDING;
-	ctask->mtask = NULL;
 	ctask->conn = conn;
 	ctask->sc = sc;
 	INIT_LIST_HEAD(&ctask->running);
@@ -929,9 +942,9 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 	} else
 		mtask->data_count = 0;
 
-	INIT_LIST_HEAD(&mtask->running);
 	memcpy(mtask->hdr, hdr, sizeof(struct iscsi_hdr));
-	__kfifo_put(conn->mgmtqueue, (void*)&mtask, sizeof(void*));
+	INIT_LIST_HEAD(&mtask->running);
+	list_add_tail(&mtask->running, &conn->mgmtqueue);
 	return mtask;
 }
 
@@ -954,13 +967,12 @@ EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu);
 void iscsi_session_recovery_timedout(struct iscsi_cls_session *cls_session)
 {
 	struct iscsi_session *session = class_to_transport_session(cls_session);
-	struct iscsi_conn *conn = session->leadconn;
 
 	spin_lock_bh(&session->lock);
 	if (session->state != ISCSI_STATE_LOGGED_IN) {
 		session->state = ISCSI_STATE_RECOVERY_FAILED;
-		if (conn)
-			wake_up(&conn->ehwait);
+		if (session->leadconn)
+			wake_up(&session->leadconn->ehwait);
 	}
 	spin_unlock_bh(&session->lock);
 }
@@ -971,7 +983,6 @@ int iscsi_eh_host_reset(struct scsi_cmnd *sc)
 	struct Scsi_Host *host = sc->device->host;
 	struct iscsi_session *session = iscsi_hostdata(host->hostdata);
 	struct iscsi_conn *conn = session->leadconn;
-	int fail_session = 0;
 
 	spin_lock_bh(&session->lock);
 	if (session->state == ISCSI_STATE_TERMINATE) {
@@ -982,19 +993,13 @@ failed:
 		return FAILED;
 	}
 
-	if (sc->SCp.phase == session->age) {
-		debug_scsi("failing connection CID %d due to SCSI host reset\n",
-			   conn->id);
-		fail_session = 1;
-	}
 	spin_unlock_bh(&session->lock);
 
 	/*
 	 * we drop the lock here but the leadconn cannot be destoyed while
 	 * we are in the scsi eh
 	 */
-	if (fail_session)
-		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+	iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
 
 	debug_scsi("iscsi_eh_host_reset wait for relogin\n");
 	wait_event_interruptible(conn->ehwait,
@@ -1015,62 +1020,43 @@ failed:
 }
 EXPORT_SYMBOL_GPL(iscsi_eh_host_reset);
 
-static void iscsi_tmabort_timedout(unsigned long data)
+static void iscsi_tmf_timedout(unsigned long data)
 {
-	struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)data;
-	struct iscsi_conn *conn = ctask->conn;
+	struct iscsi_conn *conn = (struct iscsi_conn *)data;
 	struct iscsi_session *session = conn->session;
 
 	spin_lock(&session->lock);
-	if (conn->tmabort_state == TMABORT_INITIAL) {
-		conn->tmabort_state = TMABORT_TIMEDOUT;
-		debug_scsi("tmabort timedout [sc %p itt 0x%x]\n",
-			ctask->sc, ctask->itt);
+	if (conn->tmf_state == TMF_QUEUED) {
+		conn->tmf_state = TMF_TIMEDOUT;
+		debug_scsi("tmf timedout\n");
 		/* unblock eh_abort() */
 		wake_up(&conn->ehwait);
 	}
 	spin_unlock(&session->lock);
 }
 
-static int iscsi_exec_abort_task(struct scsi_cmnd *sc,
-				 struct iscsi_cmd_task *ctask)
+static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
+				   struct iscsi_tm *hdr, int age)
 {
-	struct iscsi_conn *conn = ctask->conn;
 	struct iscsi_session *session = conn->session;
-	struct iscsi_tm *hdr = &conn->tmhdr;
-
-	/*
-	 * ctask timed out but session is OK requests must be serialized.
-	 */
-	memset(hdr, 0, sizeof(struct iscsi_tm));
-	hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE;
-	hdr->flags = ISCSI_TM_FUNC_ABORT_TASK;
-	hdr->flags |= ISCSI_FLAG_CMD_FINAL;
-	memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun));
-	hdr->rtt = ctask->hdr->itt;
-	hdr->refcmdsn = ctask->hdr->cmdsn;
+	struct iscsi_mgmt_task *mtask;
 
-	ctask->mtask = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)hdr,
-					    NULL, 0);
-	if (!ctask->mtask) {
+	mtask = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)hdr,
+				      NULL, 0);
+	if (!mtask) {
 		spin_unlock_bh(&session->lock);
 		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-		spin_lock_bh(&session->lock)
-		debug_scsi("abort sent failure [itt 0x%x]\n", ctask->itt);
+		spin_lock_bh(&session->lock);
+		debug_scsi("tmf exec failure\n");
 		return -EPERM;
 	}
-	ctask->state = ISCSI_TASK_ABORTING;
-
-	debug_scsi("abort sent [itt 0x%x]\n", ctask->itt);
+	conn->tmfcmd_pdus_cnt++;
+	conn->tmf_timer.expires = 30 * HZ + jiffies;
+	conn->tmf_timer.function = iscsi_tmf_timedout;
+	conn->tmf_timer.data = (unsigned long)conn;
+	add_timer(&conn->tmf_timer);
+	debug_scsi("tmf set timeout\n");
 
-	if (conn->tmabort_state == TMABORT_INITIAL) {
-		conn->tmfcmd_pdus_cnt++;
-		conn->tmabort_timer.expires = 20*HZ + jiffies;
-		conn->tmabort_timer.function = iscsi_tmabort_timedout;
-		conn->tmabort_timer.data = (unsigned long)ctask;
-		add_timer(&conn->tmabort_timer);
-		debug_scsi("abort set timeout [itt 0x%x]\n", ctask->itt);
-	}
 	spin_unlock_bh(&session->lock);
 	mutex_unlock(&session->eh_mutex);
 	scsi_queue_work(session->host, &conn->xmitwork);
@@ -1078,61 +1064,30 @@ static int iscsi_exec_abort_task(struct scsi_cmnd *sc,
 	/*
 	 * block eh thread until:
 	 *
-	 * 1) abort response
-	 * 2) abort timeout
+	 * 1) tmf response
+	 * 2) tmf timeout
 	 * 3) session is terminated or restarted or userspace has
 	 * given up on recovery
 	 */
-	wait_event_interruptible(conn->ehwait,
-				 sc->SCp.phase != session->age ||
+	wait_event_interruptible(conn->ehwait, age != session->age ||
 				 session->state != ISCSI_STATE_LOGGED_IN ||
-				 conn->tmabort_state != TMABORT_INITIAL);
+				 conn->tmf_state != TMF_QUEUED);
 	if (signal_pending(current))
 		flush_signals(current);
-	del_timer_sync(&conn->tmabort_timer);
+	del_timer_sync(&conn->tmf_timer);
+
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
-	return 0;
-}
-
-/*
- * session lock must be held
- */
-static struct iscsi_mgmt_task *
-iscsi_remove_mgmt_task(struct kfifo *fifo, uint32_t itt)
-{
-	int i, nr_tasks = __kfifo_len(fifo) / sizeof(void*);
-	struct iscsi_mgmt_task *task;
-
-	debug_scsi("searching %d tasks\n", nr_tasks);
-
-	for (i = 0; i < nr_tasks; i++) {
-		__kfifo_get(fifo, (void*)&task, sizeof(void*));
-		debug_scsi("check task %u\n", task->itt);
-
-		if (task->itt == itt) {
-			debug_scsi("matched task\n");
-			return task;
-		}
+	/* if the session drops it will clean up the mtask */
+	if (age != session->age ||
+	    session->state != ISCSI_STATE_LOGGED_IN)
+		return -ENOTCONN;
 
-		__kfifo_put(fifo, (void*)&task, sizeof(void*));
+	if (!list_empty(&mtask->running)) {
+		list_del_init(&mtask->running);
+		__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
+			    sizeof(void*));
 	}
-	return NULL;
-}
-
-static int iscsi_ctask_mtask_cleanup(struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_conn *conn = ctask->conn;
-	struct iscsi_session *session = conn->session;
-
-	if (!ctask->mtask)
-		return -EINVAL;
-
-	if (!iscsi_remove_mgmt_task(conn->mgmtqueue, ctask->mtask->itt))
-		list_del(&ctask->mtask->running);
-	__kfifo_put(session->mgmtpool.queue, (void*)&ctask->mtask,
-		    sizeof(void*));
-	ctask->mtask = NULL;
 	return 0;
 }
 
@@ -1156,7 +1111,6 @@ static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 		conn->session->queued_cmdsn--;
 	else
 		conn->session->tt->cleanup_cmd_task(conn, ctask);
-	iscsi_ctask_mtask_cleanup(ctask);
 
 	sc->result = err;
 	scsi_set_resid(sc, scsi_bufflen(sc));
@@ -1166,6 +1120,44 @@ static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 	__iscsi_put_ctask(ctask);
 }
 
+/*
+ * Fail commands. session lock held and recv side suspended and xmit
+ * thread flushed
+ */
+static void fail_all_commands(struct iscsi_conn *conn, unsigned lun)
+{
+	struct iscsi_cmd_task *ctask, *tmp;
+
+	if (conn->ctask && (conn->ctask->sc->device->lun == lun || lun == -1))
+		conn->ctask = NULL;
+
+	/* flush pending */
+	list_for_each_entry_safe(ctask, tmp, &conn->xmitqueue, running) {
+		if (lun == ctask->sc->device->lun || lun == -1) {
+			debug_scsi("failing pending sc %p itt 0x%x\n",
+				   ctask->sc, ctask->itt);
+			fail_command(conn, ctask, DID_BUS_BUSY << 16);
+		}
+	}
+
+	list_for_each_entry_safe(ctask, tmp, &conn->requeue, running) {
+		if (lun == ctask->sc->device->lun || lun == -1) {
+			debug_scsi("failing requeued sc %p itt 0x%x\n",
+				   ctask->sc, ctask->itt);
+			fail_command(conn, ctask, DID_BUS_BUSY << 16);
+		}
+	}
+
+	/* fail all other running */
+	list_for_each_entry_safe(ctask, tmp, &conn->run_list, running) {
+		if (lun == ctask->sc->device->lun || lun == -1) {
+			debug_scsi("failing in progress sc %p itt 0x%x\n",
+				   ctask->sc, ctask->itt);
+			fail_command(conn, ctask, DID_BUS_BUSY << 16);
+		}
+	}
+}
+
 static void iscsi_suspend_tx(struct iscsi_conn *conn)
 {
 	set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx);
@@ -1178,13 +1170,26 @@ static void iscsi_start_tx(struct iscsi_conn *conn)
 	scsi_queue_work(conn->session->host, &conn->xmitwork);
 }
 
+static void iscsi_prep_abort_task_pdu(struct iscsi_cmd_task *ctask,
+				      struct iscsi_tm *hdr)
+{
+	memset(hdr, 0, sizeof(*hdr));
+	hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE;
+	hdr->flags = ISCSI_TM_FUNC_ABORT_TASK & ISCSI_FLAG_TM_FUNC_MASK;
+	hdr->flags |= ISCSI_FLAG_CMD_FINAL;
+	memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun));
+	hdr->rtt = ctask->hdr->itt;
+	hdr->refcmdsn = ctask->hdr->cmdsn;
+}
+
 int iscsi_eh_abort(struct scsi_cmnd *sc)
 {
 	struct Scsi_Host *host = sc->device->host;
 	struct iscsi_session *session = iscsi_hostdata(host->hostdata);
-	struct iscsi_cmd_task *ctask;
 	struct iscsi_conn *conn;
-	int rc;
+	struct iscsi_cmd_task *ctask;
+	struct iscsi_tm *hdr;
+	int rc, age;
 
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
@@ -1199,19 +1204,23 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 		return SUCCESS;
 	}
 
-	ctask = (struct iscsi_cmd_task *)sc->SCp.ptr;
-	conn = ctask->conn;
-
-	conn->eh_abort_cnt++;
-	debug_scsi("aborting [sc %p itt 0x%x]\n", sc, ctask->itt);
-
 	/*
 	 * If we are not logged in or we have started a new session
 	 * then let the host reset code handle this
 	 */
-	if (session->state != ISCSI_STATE_LOGGED_IN ||
-	    sc->SCp.phase != session->age)
-		goto failed;
+	if (!session->leadconn || session->state != ISCSI_STATE_LOGGED_IN ||
+	    sc->SCp.phase != session->age) {
+		spin_unlock_bh(&session->lock);
+		mutex_unlock(&session->eh_mutex);
+		return FAILED;
+	}
+
+	conn = session->leadconn;
+	conn->eh_abort_cnt++;
+	age = session->age;
+
+	ctask = (struct iscsi_cmd_task *)sc->SCp.ptr;
+	debug_scsi("aborting [sc %p itt 0x%x]\n", sc, ctask->itt);
 
 	/* ctask completed before time out */
 	if (!ctask->sc) {
@@ -1219,27 +1228,26 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 		goto success;
 	}
 
-	/* what should we do here ? */
-	if (conn->ctask == ctask) {
-		printk(KERN_INFO "iscsi: sc %p itt 0x%x partially sent. "
-		       "Failing abort\n", sc, ctask->itt);
-		goto failed;
-	}
-
 	if (ctask->state == ISCSI_TASK_PENDING) {
 		fail_command(conn, ctask, DID_ABORT << 16);
 		goto success;
 	}
 
-	conn->tmabort_state = TMABORT_INITIAL;
-	rc = iscsi_exec_abort_task(sc, ctask);
-	if (rc || sc->SCp.phase != session->age ||
-	    session->state != ISCSI_STATE_LOGGED_IN)
+	/* only have one tmf outstanding at a time */
+	if (conn->tmf_state != TMF_INITIAL)
 		goto failed;
-	iscsi_ctask_mtask_cleanup(ctask);
+	conn->tmf_state = TMF_QUEUED;
 
-	switch (conn->tmabort_state) {
-	case TMABORT_SUCCESS:
+	hdr = &conn->tmhdr;
+	iscsi_prep_abort_task_pdu(ctask, hdr);
+
+	if (iscsi_exec_task_mgmt_fn(conn, hdr, age)) {
+		rc = FAILED;
+		goto failed;
+	}
+
+	switch (conn->tmf_state) {
+	case TMF_SUCCESS:
 		spin_unlock_bh(&session->lock);
 		iscsi_suspend_tx(conn);
 		/*
@@ -1248,22 +1256,26 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 		write_lock_bh(conn->recv_lock);
 		spin_lock(&session->lock);
 		fail_command(conn, ctask, DID_ABORT << 16);
+		conn->tmf_state = TMF_INITIAL;
 		spin_unlock(&session->lock);
 		write_unlock_bh(conn->recv_lock);
 		iscsi_start_tx(conn);
 		goto success_unlocked;
-	case TMABORT_NOT_FOUND:
-		if (!ctask->sc) {
+	case TMF_TIMEDOUT:
+		spin_unlock_bh(&session->lock);
+		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+		goto failed_unlocked;
+	case TMF_NOT_FOUND:
+		if (!sc->SCp.ptr) {
+			conn->tmf_state = TMF_INITIAL;
 			/* ctask completed before tmf abort response */
 			debug_scsi("sc completed while abort in progress\n");
 			goto success;
 		}
 		/* fall through */
 	default:
-		/* timedout or failed */
-		spin_unlock_bh(&session->lock);
-		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-		goto failed_unlocked;
+		conn->tmf_state = TMF_INITIAL;
+		goto failed;
 	}
 
 success:
@@ -1276,12 +1288,93 @@ success_unlocked:
 failed:
 	spin_unlock_bh(&session->lock);
 failed_unlocked:
-	debug_scsi("abort failed [sc %lx itt 0x%x]\n", (long)sc, ctask->itt);
+	debug_scsi("abort failed [sc %p itt 0x%x]\n", sc,
+		    ctask ? ctask->itt : 0);
 	mutex_unlock(&session->eh_mutex);
 	return FAILED;
 }
 EXPORT_SYMBOL_GPL(iscsi_eh_abort);
 
+static void iscsi_prep_lun_reset_pdu(struct scsi_cmnd *sc, struct iscsi_tm *hdr)
+{
+	memset(hdr, 0, sizeof(*hdr));
+	hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE;
+	hdr->flags = ISCSI_TM_FUNC_LOGICAL_UNIT_RESET & ISCSI_FLAG_TM_FUNC_MASK;
+	hdr->flags |= ISCSI_FLAG_CMD_FINAL;
+	int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun);
+	hdr->rtt = ISCSI_RESERVED_TAG;
+}
+
+int iscsi_eh_device_reset(struct scsi_cmnd *sc)
+{
+	struct Scsi_Host *host = sc->device->host;
+	struct iscsi_session *session = iscsi_hostdata(host->hostdata);
+	struct iscsi_conn *conn;
+	struct iscsi_tm *hdr;
+	int rc = FAILED;
+
+	debug_scsi("LU Reset [sc %p lun %u]\n", sc, sc->device->lun);
+
+	mutex_lock(&session->eh_mutex);
+	spin_lock_bh(&session->lock);
+	/*
+	 * Just check if we are not logged in. We cannot check for
+	 * the phase because the reset could come from a ioctl.
+	 */
+	if (!session->leadconn || session->state != ISCSI_STATE_LOGGED_IN)
+		goto unlock;
+	conn = session->leadconn;
+
+	/* only have one tmf outstanding at a time */
+	if (conn->tmf_state != TMF_INITIAL)
+		goto unlock;
+	conn->tmf_state = TMF_QUEUED;
+
+	hdr = &conn->tmhdr;
+	iscsi_prep_lun_reset_pdu(sc, hdr);
+
+	if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age)) {
+		rc = FAILED;
+		goto unlock;
+	}
+
+	switch (conn->tmf_state) {
+	case TMF_SUCCESS:
+		break;
+	case TMF_TIMEDOUT:
+		spin_unlock_bh(&session->lock);
+		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+		goto done;
+	default:
+		conn->tmf_state = TMF_INITIAL;
+		goto unlock;
+	}
+
+	rc = SUCCESS;
+	spin_unlock_bh(&session->lock);
+
+	iscsi_suspend_tx(conn);
+	/* need to grab the recv lock then session lock */
+	write_lock_bh(conn->recv_lock);
+	spin_lock(&session->lock);
+	fail_all_commands(conn, sc->device->lun);
+	conn->tmf_state = TMF_INITIAL;
+	spin_unlock(&session->lock);
+	write_unlock_bh(conn->recv_lock);
+
+	iscsi_start_tx(conn);
+	goto done;
+
+unlock:
+	spin_unlock_bh(&session->lock);
+done:
+	debug_scsi("iscsi_eh_device_reset %s\n",
+		  rc == SUCCESS ? "SUCCESS" : "FAILED");
+	mutex_unlock(&session->eh_mutex);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(iscsi_eh_device_reset);
+
 int
 iscsi_pool_init(struct iscsi_queue *q, int max, void ***items, int item_size)
 {
@@ -1546,17 +1639,12 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 	conn->c_stage = ISCSI_CONN_INITIAL_STAGE;
 	conn->id = conn_idx;
 	conn->exp_statsn = 0;
-	conn->tmabort_state = TMABORT_INITIAL;
+	conn->tmf_state = TMF_INITIAL;
 	INIT_LIST_HEAD(&conn->run_list);
 	INIT_LIST_HEAD(&conn->mgmt_run_list);
+	INIT_LIST_HEAD(&conn->mgmtqueue);
 	INIT_LIST_HEAD(&conn->xmitqueue);
-
-	/* initialize general immediate & non-immediate PDU commands queue */
-	conn->mgmtqueue = kfifo_alloc(session->mgmtpool_max * sizeof(void*),
-			                GFP_KERNEL, NULL);
-	if (conn->mgmtqueue == ERR_PTR(-ENOMEM))
-		goto mgmtqueue_alloc_fail;
-
+	INIT_LIST_HEAD(&conn->requeue);
 	INIT_WORK(&conn->xmitwork, iscsi_xmitworker);
 
 	/* allocate login_mtask used for the login/text sequences */
@@ -1574,7 +1662,7 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 		goto login_mtask_data_alloc_fail;
 	conn->login_mtask->data = conn->data = data;
 
-	init_timer(&conn->tmabort_timer);
+	init_timer(&conn->tmf_timer);
 	init_waitqueue_head(&conn->ehwait);
 
 	return cls_conn;
@@ -1583,8 +1671,6 @@ login_mtask_data_alloc_fail:
 	__kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask,
 		    sizeof(void*));
 login_mtask_alloc_fail:
-	kfifo_free(conn->mgmtqueue);
-mgmtqueue_alloc_fail:
 	iscsi_destroy_conn(cls_conn);
 	return NULL;
 }
@@ -1604,7 +1690,6 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 	unsigned long flags;
 
 	spin_lock_bh(&session->lock);
-	set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx);
 	conn->c_stage = ISCSI_CONN_CLEANUP_WAIT;
 	if (session->leadconn == conn) {
 		/*
@@ -1637,7 +1722,7 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 	}
 
 	/* flush queued up work because we free the connection below */
-	scsi_flush_work(session->host);
+	iscsi_suspend_tx(conn);
 
 	spin_lock_bh(&session->lock);
 	kfree(conn->data);
@@ -1648,8 +1733,6 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 		session->leadconn = NULL;
 	spin_unlock_bh(&session->lock);
 
-	kfifo_free(conn->mgmtqueue);
-
 	iscsi_destroy_conn(cls_conn);
 }
 EXPORT_SYMBOL_GPL(iscsi_conn_teardown);
@@ -1684,7 +1767,7 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn)
 		 * commands after successful recovery
 		 */
 		conn->stop_stage = 0;
-		conn->tmabort_state = TMABORT_INITIAL;
+		conn->tmf_state = TMF_INITIAL;
 		session->age++;
 		spin_unlock_bh(&session->lock);
 
@@ -1709,10 +1792,11 @@ flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn)
 	struct iscsi_mgmt_task *mtask, *tmp;
 
 	/* handle pending */
-	while (__kfifo_get(conn->mgmtqueue, (void*)&mtask, sizeof(void*))) {
+	list_for_each_entry_safe(mtask, tmp, &conn->mgmtqueue, running) {
+		debug_scsi("flushing pending mgmt task itt 0x%x\n", mtask->itt);
+		list_del_init(&mtask->running);
 		if (mtask == conn->login_mtask)
 			continue;
-		debug_scsi("flushing pending mgmt task itt 0x%x\n", mtask->itt);
 		__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
 			    sizeof(void*));
 	}
@@ -1720,7 +1804,7 @@ flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn)
 	/* handle running */
 	list_for_each_entry_safe(mtask, tmp, &conn->mgmt_run_list, running) {
 		debug_scsi("flushing running mgmt task itt 0x%x\n", mtask->itt);
-		list_del(&mtask->running);
+		list_del_init(&mtask->running);
 
 		if (mtask == conn->login_mtask)
 			continue;
@@ -1731,28 +1815,6 @@ flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn)
 	conn->mtask = NULL;
 }
 
-/* Fail commands. Mutex and session lock held and recv side suspended */
-static void fail_all_commands(struct iscsi_conn *conn)
-{
-	struct iscsi_cmd_task *ctask, *tmp;
-
-	/* flush pending */
-	list_for_each_entry_safe(ctask, tmp, &conn->xmitqueue, running) {
-		debug_scsi("failing pending sc %p itt 0x%x\n", ctask->sc,
-			   ctask->itt);
-		fail_command(conn, ctask, DID_BUS_BUSY << 16);
-	}
-
-	/* fail all other running */
-	list_for_each_entry_safe(ctask, tmp, &conn->run_list, running) {
-		debug_scsi("failing in progress sc %p itt 0x%x\n",
-			   ctask->sc, ctask->itt);
-		fail_command(conn, ctask, DID_BUS_BUSY << 16);
-	}
-
-	conn->ctask = NULL;
-}
-
 static void iscsi_start_session_recovery(struct iscsi_session *session,
 					 struct iscsi_conn *conn, int flag)
 {
@@ -1818,7 +1880,7 @@ static void iscsi_start_session_recovery(struct iscsi_session *session,
 	 * flush queues.
 	 */
 	spin_lock_bh(&session->lock);
-	fail_all_commands(conn);
+	fail_all_commands(conn, -1);
 	flush_control_queues(session, conn);
 	spin_unlock_bh(&session->lock);
 	mutex_unlock(&session->eh_mutex);
@@ -1869,6 +1931,9 @@ int iscsi_set_param(struct iscsi_cls_conn *cls_conn,
 	uint32_t value;
 
 	switch(param) {
+	case ISCSI_PARAM_FAST_ABORT:
+		sscanf(buf, "%d", &session->fast_abort);
+		break;
 	case ISCSI_PARAM_MAX_RECV_DLENGTH:
 		sscanf(buf, "%d", &conn->max_recv_dlength);
 		break;
@@ -1983,6 +2048,9 @@ int iscsi_session_get_param(struct iscsi_cls_session *cls_session,
 	int len;
 
 	switch(param) {
+	case ISCSI_PARAM_FAST_ABORT:
+		len = sprintf(buf, "%d\n", session->fast_abort);
+		break;
 	case ISCSI_PARAM_INITIAL_R2T_EN:
 		len = sprintf(buf, "%d\n", session->initial_r2t_en);
 		break;
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index cb48b80c0865..75d3069ecaa0 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -30,7 +30,7 @@
 #include <scsi/scsi_transport_iscsi.h>
 #include <scsi/iscsi_if.h>
 
-#define ISCSI_SESSION_ATTRS 15
+#define ISCSI_SESSION_ATTRS 16
 #define ISCSI_CONN_ATTRS 11
 #define ISCSI_HOST_ATTRS 4
 #define ISCSI_TRANSPORT_VERSION "2.0-724"
@@ -1218,6 +1218,7 @@ iscsi_session_attr(username, ISCSI_PARAM_USERNAME, 1);
 iscsi_session_attr(username_in, ISCSI_PARAM_USERNAME_IN, 1);
 iscsi_session_attr(password, ISCSI_PARAM_PASSWORD, 1);
 iscsi_session_attr(password_in, ISCSI_PARAM_PASSWORD_IN, 1);
+iscsi_session_attr(fast_abort, ISCSI_PARAM_FAST_ABORT, 1);
 
 #define iscsi_priv_session_attr_show(field, format)			\
 static ssize_t								\
@@ -1439,6 +1440,7 @@ iscsi_register_transport(struct iscsi_transport *tt)
 	SETUP_SESSION_RD_ATTR(password_in, ISCSI_USERNAME_IN);
 	SETUP_SESSION_RD_ATTR(username, ISCSI_PASSWORD);
 	SETUP_SESSION_RD_ATTR(username_in, ISCSI_PASSWORD_IN);
+	SETUP_SESSION_RD_ATTR(fast_abort, ISCSI_FAST_ABORT);
 	SETUP_PRIV_SESSION_RD_ATTR(recovery_tmo);
 
 	BUG_ON(count > ISCSI_SESSION_ATTRS);
diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h
index 50e907f42048..bff0b1f7857b 100644
--- a/include/scsi/iscsi_if.h
+++ b/include/scsi/iscsi_if.h
@@ -236,6 +236,7 @@ enum iscsi_param {
 	ISCSI_PARAM_PASSWORD,
 	ISCSI_PARAM_PASSWORD_IN,
 
+	ISCSI_PARAM_FAST_ABORT,
 	/* must always be last */
 	ISCSI_PARAM_MAX,
 };
@@ -266,6 +267,7 @@ enum iscsi_param {
 #define ISCSI_USERNAME_IN		(1 << ISCSI_PARAM_USERNAME_IN)
 #define ISCSI_PASSWORD			(1 << ISCSI_PARAM_PASSWORD)
 #define ISCSI_PASSWORD_IN		(1 << ISCSI_PARAM_PASSWORD_IN)
+#define ISCSI_FAST_ABORT		(1 << ISCSI_PARAM_FAST_ABORT)
 
 /* iSCSI HBA params */
 enum iscsi_host_param {
diff --git a/include/scsi/iscsi_proto.h b/include/scsi/iscsi_proto.h
index 8d1e4e8026fe..751c81eaa7f3 100644
--- a/include/scsi/iscsi_proto.h
+++ b/include/scsi/iscsi_proto.h
@@ -600,6 +600,8 @@ struct iscsi_reject {
 #define ISCSI_MIN_MAX_BURST_LEN			512
 #define ISCSI_MAX_MAX_BURST_LEN			16777215
 
+#define ISCSI_DEF_TIME2WAIT			2
+
 /************************* RFC 3720 End *****************************/
 
 #endif /* ISCSI_PROTO_H */
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index b4b31132618b..89429f433f85 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -57,11 +57,14 @@ struct iscsi_nopin;
 #define ISCSI_MAX_CMD_PER_LUN		128
 
 /* Task Mgmt states */
-#define TMABORT_INITIAL			0x0
-#define TMABORT_SUCCESS			0x1
-#define TMABORT_FAILED			0x2
-#define TMABORT_TIMEDOUT		0x3
-#define TMABORT_NOT_FOUND		0x4
+enum {
+	TMF_INITIAL,
+	TMF_QUEUED,
+	TMF_SUCCESS,
+	TMF_FAILED,
+	TMF_TIMEDOUT,
+	TMF_NOT_FOUND,
+};
 
 /* Connection suspend "bit" */
 #define ISCSI_SUSPEND_BIT		1
@@ -91,7 +94,6 @@ enum {
 	ISCSI_TASK_COMPLETED,
 	ISCSI_TASK_PENDING,
 	ISCSI_TASK_RUNNING,
-	ISCSI_TASK_ABORTING,
 };
 
 struct iscsi_cmd_task {
@@ -110,7 +112,6 @@ struct iscsi_cmd_task {
 	unsigned		data_count;	/* remaining Data-Out */
 	struct scsi_cmnd	*sc;		/* associated SCSI cmd*/
 	struct iscsi_conn	*conn;		/* used connection    */
-	struct iscsi_mgmt_task	*mtask;		/* tmf mtask in progr */
 
 	/* state set/tested under session->lock */
 	int			state;
@@ -152,10 +153,11 @@ struct iscsi_conn {
 	struct iscsi_cmd_task	*ctask;		/* xmit ctask in progress */
 
 	/* xmit */
-	struct kfifo		*mgmtqueue;	/* mgmt (control) xmit queue */
+	struct list_head	mgmtqueue;	/* mgmt (control) xmit queue */
 	struct list_head	mgmt_run_list;	/* list of control tasks */
 	struct list_head	xmitqueue;	/* data-path cmd queue */
 	struct list_head	run_list;	/* list of cmds in progress */
+	struct list_head	requeue;	/* tasks needing another run */
 	struct work_struct	xmitwork;	/* per-conn. xmit workqueue */
 	unsigned long		suspend_tx;	/* suspend Tx */
 	unsigned long		suspend_rx;	/* suspend Rx */
@@ -163,8 +165,8 @@ struct iscsi_conn {
 	/* abort */
 	wait_queue_head_t	ehwait;		/* used in eh_abort() */
 	struct iscsi_tm		tmhdr;
-	struct timer_list	tmabort_timer;
-	int			tmabort_state;	/* see TMABORT_INITIAL, etc.*/
+	struct timer_list	tmf_timer;
+	int			tmf_state;	/* see TMF_INITIAL, etc.*/
 
 	/* negotiated params */
 	unsigned		max_recv_dlength; /* initiator_max_recv_dsl*/
@@ -231,6 +233,7 @@ struct iscsi_session {
 	int			pdu_inorder_en;
 	int			dataseq_inorder_en;
 	int			erl;
+	int			fast_abort;
 	int			tpgt;
 	char			*username;
 	char			*username_in;
@@ -268,6 +271,7 @@ struct iscsi_session {
 extern int iscsi_change_queue_depth(struct scsi_device *sdev, int depth);
 extern int iscsi_eh_abort(struct scsi_cmnd *sc);
 extern int iscsi_eh_host_reset(struct scsi_cmnd *sc);
+extern int iscsi_eh_device_reset(struct scsi_cmnd *sc);
 extern int iscsi_queuecommand(struct scsi_cmnd *sc,
 			      void (*done)(struct scsi_cmnd *));
 
@@ -326,6 +330,7 @@ extern int __iscsi_complete_pdu(struct iscsi_conn *, struct iscsi_hdr *,
 				char *, int);
 extern int iscsi_verify_itt(struct iscsi_conn *, struct iscsi_hdr *,
 			    uint32_t *);
+extern void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask);
 
 /*
  * generic helpers
-- 
cgit v1.2.3


From da32dd681f7a1a17073c42b375fc23cf73c92155 Mon Sep 17 00:00:00 2001
From: Olaf Kirch <olaf.kirch@oracle.com>
Date: Thu, 13 Dec 2007 12:43:21 -0600
Subject: [SCSI] iscsi_tcp: rewrite recv path

Rewrite recv path. Fixes:
- data digest processing and error handling.
- ahs support.

Some fixups by Mike Christie

Signed-off-by: Olaf Kirch <olaf.kirch@oracle.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/iscsi_tcp.c | 1018 +++++++++++++++++++++++-----------------------
 drivers/scsi/iscsi_tcp.h |   66 +--
 include/scsi/libiscsi.h  |    4 +
 3 files changed, 552 insertions(+), 536 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 4b226b88b68a..1b540e03f5bf 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -48,7 +48,7 @@ MODULE_AUTHOR("Dmitry Yusupov <dmitry_yus@yahoo.com>, "
 	      "Alex Aizman <itn780@yahoo.com>");
 MODULE_DESCRIPTION("iSCSI/TCP data-path");
 MODULE_LICENSE("GPL");
-/* #define DEBUG_TCP */
+#undef DEBUG_TCP
 #define DEBUG_ASSERT
 
 #ifdef DEBUG_TCP
@@ -67,10 +67,15 @@ MODULE_LICENSE("GPL");
 static unsigned int iscsi_max_lun = 512;
 module_param_named(max_lun, iscsi_max_lun, uint, S_IRUGO);
 
+static int iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
+				   struct iscsi_chunk *chunk);
+
 static inline void
 iscsi_buf_init_iov(struct iscsi_buf *ibuf, char *vbuf, int size)
 {
-	sg_init_one(&ibuf->sg, vbuf, size);
+	ibuf->sg.page = virt_to_page(vbuf);
+	ibuf->sg.offset = offset_in_page(vbuf);
+	ibuf->sg.length = size;
 	ibuf->sent = 0;
 	ibuf->use_sendmsg = 1;
 }
@@ -78,12 +83,13 @@ iscsi_buf_init_iov(struct iscsi_buf *ibuf, char *vbuf, int size)
 static inline void
 iscsi_buf_init_sg(struct iscsi_buf *ibuf, struct scatterlist *sg)
 {
-	sg_init_table(&ibuf->sg, 1);
-	sg_set_page(&ibuf->sg, sg_page(sg), sg->length, sg->offset);
+	ibuf->sg.page = sg->page;
+	ibuf->sg.offset = sg->offset;
+	ibuf->sg.length = sg->length;
 	/*
 	 * Fastpath: sg element fits into single page
 	 */
-	if (sg->length + sg->offset <= PAGE_SIZE && !PageSlab(sg_page(sg)))
+	if (sg->length + sg->offset <= PAGE_SIZE && !PageSlab(sg->page))
 		ibuf->use_sendmsg = 0;
 	else
 		ibuf->use_sendmsg = 1;
@@ -110,72 +116,331 @@ iscsi_hdr_digest(struct iscsi_conn *conn, struct iscsi_buf *buf,
 	buf->sg.length += sizeof(u32);
 }
 
+/*
+ * Scatterlist handling: inside the iscsi_chunk, we
+ * remember an index into the scatterlist, and set data/size
+ * to the current scatterlist entry. For highmem pages, we
+ * kmap as needed.
+ *
+ * Note that the page is unmapped when we return from
+ * TCP's data_ready handler, so we may end up mapping and
+ * unmapping the same page repeatedly. The whole reason
+ * for this is that we shouldn't keep the page mapped
+ * outside the softirq.
+ */
+
+/**
+ * iscsi_tcp_chunk_init_sg - init indicated scatterlist entry
+ * @chunk: the buffer object
+ * @idx: index into scatterlist
+ * @offset: byte offset into that sg entry
+ *
+ * This function sets up the chunk so that subsequent
+ * data is copied to the indicated sg entry, at the given
+ * offset.
+ */
+static inline void
+iscsi_tcp_chunk_init_sg(struct iscsi_chunk *chunk,
+			unsigned int idx, unsigned int offset)
+{
+	struct scatterlist *sg;
+
+	BUG_ON(chunk->sg == NULL);
+
+	sg = &chunk->sg[idx];
+	chunk->sg_index = idx;
+	chunk->sg_offset = offset;
+	chunk->size = min(sg->length - offset, chunk->total_size);
+	chunk->data = NULL;
+}
+
+/**
+ * iscsi_tcp_chunk_map - map the current S/G page
+ * @chunk: iscsi chunk
+ *
+ * We only need to possibly kmap data if scatter lists are being used,
+ * because the iscsi passthrough and internal IO paths will never use high
+ * mem pages.
+ */
+static inline void
+iscsi_tcp_chunk_map(struct iscsi_chunk *chunk)
+{
+	struct scatterlist *sg;
+
+	if (chunk->data != NULL || !chunk->sg)
+		return;
+
+	sg = &chunk->sg[chunk->sg_index];
+	BUG_ON(chunk->sg_mapped);
+	BUG_ON(sg->length == 0);
+	chunk->sg_mapped = kmap_atomic(sg->page, KM_SOFTIRQ0);
+	chunk->data = chunk->sg_mapped + sg->offset + chunk->sg_offset;
+}
+
+static inline void
+iscsi_tcp_chunk_unmap(struct iscsi_chunk *chunk)
+{
+	if (chunk->sg_mapped) {
+		kunmap_atomic(chunk->sg_mapped, KM_SOFTIRQ0);
+		chunk->sg_mapped = NULL;
+		chunk->data = NULL;
+	}
+}
+
+/*
+ * Splice the digest buffer into the buffer
+ */
+static inline void
+iscsi_tcp_chunk_splice_digest(struct iscsi_chunk *chunk, void *digest)
+{
+	chunk->data = digest;
+	chunk->digest_len = ISCSI_DIGEST_SIZE;
+	chunk->total_size += ISCSI_DIGEST_SIZE;
+	chunk->size = ISCSI_DIGEST_SIZE;
+	chunk->copied = 0;
+	chunk->sg = NULL;
+	chunk->sg_index = 0;
+	chunk->hash = NULL;
+}
+
+/**
+ * iscsi_tcp_chunk_done - check whether the chunk is complete
+ * @chunk: iscsi chunk to check
+ *
+ * Check if we're done receiving this chunk. If the receive
+ * buffer is full but we expect more data, move on to the
+ * next entry in the scatterlist.
+ *
+ * If the amount of data we received isn't a multiple of 4,
+ * we will transparently receive the pad bytes, too.
+ *
+ * This function must be re-entrant.
+ */
 static inline int
-iscsi_hdr_extract(struct iscsi_tcp_conn *tcp_conn)
+iscsi_tcp_chunk_done(struct iscsi_chunk *chunk)
 {
-	struct sk_buff *skb = tcp_conn->in.skb;
+	static unsigned char padbuf[ISCSI_PAD_LEN];
+
+	if (chunk->copied < chunk->size) {
+		iscsi_tcp_chunk_map(chunk);
+		return 0;
+	}
 
-	tcp_conn->in.zero_copy_hdr = 0;
+	chunk->total_copied += chunk->copied;
+	chunk->copied = 0;
+	chunk->size = 0;
 
-	if (tcp_conn->in.copy >= tcp_conn->hdr_size &&
-	    tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER) {
-		/*
-		 * Zero-copy PDU Header: using connection context
-		 * to store header pointer.
-		 */
-		if (skb_shinfo(skb)->frag_list == NULL &&
-		    !skb_shinfo(skb)->nr_frags) {
-			tcp_conn->in.hdr = (struct iscsi_hdr *)
-				((char*)skb->data + tcp_conn->in.offset);
-			tcp_conn->in.zero_copy_hdr = 1;
-		} else {
-			/* ignoring return code since we checked
-			 * in.copy before */
-			skb_copy_bits(skb, tcp_conn->in.offset,
-				&tcp_conn->hdr, tcp_conn->hdr_size);
-			tcp_conn->in.hdr = &tcp_conn->hdr;
+	/* Unmap the current scatterlist page, if there is one. */
+	iscsi_tcp_chunk_unmap(chunk);
+
+	/* Do we have more scatterlist entries? */
+	if (chunk->total_copied < chunk->total_size) {
+		/* Proceed to the next entry in the scatterlist. */
+		iscsi_tcp_chunk_init_sg(chunk, chunk->sg_index + 1, 0);
+		iscsi_tcp_chunk_map(chunk);
+		BUG_ON(chunk->size == 0);
+		return 0;
+	}
+
+	/* Do we need to handle padding? */
+	if (chunk->total_copied & (ISCSI_PAD_LEN-1)) {
+		unsigned int pad;
+
+		pad = ISCSI_PAD_LEN - (chunk->total_copied & (ISCSI_PAD_LEN-1));
+		debug_tcp("consume %d pad bytes\n", pad);
+		chunk->total_size += pad;
+		chunk->size = pad;
+		chunk->data = padbuf;
+		return 0;
+	}
+
+	/*
+	 * Set us up for receiving the data digest. hdr digest
+	 * is completely handled in hdr done function.
+	 */
+	if (chunk->hash) {
+		if (chunk->digest_len == 0) {
+			crypto_hash_final(chunk->hash, chunk->digest);
+			iscsi_tcp_chunk_splice_digest(chunk,
+						      chunk->recv_digest);
+			return 0;
 		}
-		tcp_conn->in.offset += tcp_conn->hdr_size;
-		tcp_conn->in.copy -= tcp_conn->hdr_size;
-	} else {
-		int hdr_remains;
-		int copylen;
+	}
 
-		/*
-		 * PDU header scattered across SKB's,
-		 * copying it... This'll happen quite rarely.
-		 */
+	return 1;
+}
+
+/**
+ * iscsi_tcp_chunk_recv - copy data to chunk
+ * @tcp_conn: the iSCSI TCP connection
+ * @chunk: the buffer to copy to
+ * @ptr: data pointer
+ * @len: amount of data available
+ *
+ * This function copies up to @len bytes to the
+ * given buffer, and returns the number of bytes
+ * consumed, which can actually be less than @len.
+ *
+ * If hash digest is enabled, the function will update the
+ * hash while copying.
+ * Combining these two operations doesn't buy us a lot (yet),
+ * but in the future we could implement combined copy+crc,
+ * just way we do for network layer checksums.
+ */
+static int
+iscsi_tcp_chunk_recv(struct iscsi_tcp_conn *tcp_conn,
+		     struct iscsi_chunk *chunk, const void *ptr,
+		     unsigned int len)
+{
+	struct scatterlist sg;
+	unsigned int copy, copied = 0;
+
+	while (!iscsi_tcp_chunk_done(chunk)) {
+		if (copied == len)
+			goto out;
+
+		copy = min(len - copied, chunk->size - chunk->copied);
+		memcpy(chunk->data + chunk->copied, ptr + copied, copy);
+
+		if (chunk->hash) {
+			sg_init_one(&sg, ptr + copied, copy);
+			crypto_hash_update(chunk->hash, &sg, copy);
+		}
+		chunk->copied += copy;
+		copied += copy;
+	}
+
+out:
+	return copied;
+}
 
-		if (tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER)
-			tcp_conn->in.hdr_offset = 0;
+static inline void
+iscsi_tcp_dgst_header(struct hash_desc *hash, const void *hdr, size_t hdrlen,
+		      unsigned char digest[ISCSI_DIGEST_SIZE])
+{
+	struct scatterlist sg;
+
+	sg_init_one(&sg, hdr, hdrlen);
+	crypto_hash_digest(hash, &sg, hdrlen, digest);
+}
+
+static inline int
+iscsi_tcp_dgst_verify(struct iscsi_tcp_conn *tcp_conn,
+		      struct iscsi_chunk *chunk)
+{
+	if (!chunk->digest_len)
+		return 1;
+
+	if (memcmp(chunk->recv_digest, chunk->digest, chunk->digest_len)) {
+		debug_scsi("digest mismatch\n");
+		return 0;
+	}
 
-		hdr_remains = tcp_conn->hdr_size - tcp_conn->in.hdr_offset;
-		BUG_ON(hdr_remains <= 0);
+	return 1;
+}
 
-		copylen = min(tcp_conn->in.copy, hdr_remains);
-		skb_copy_bits(skb, tcp_conn->in.offset,
-			(char*)&tcp_conn->hdr + tcp_conn->in.hdr_offset,
-			copylen);
+/*
+ * Helper function to set up chunk buffer
+ */
+static inline void
+__iscsi_chunk_init(struct iscsi_chunk *chunk, size_t size,
+		   iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+{
+	memset(chunk, 0, sizeof(*chunk));
+	chunk->total_size = size;
+	chunk->done = done;
 
-		debug_tcp("PDU gather offset %d bytes %d in.offset %d "
-		       "in.copy %d\n", tcp_conn->in.hdr_offset, copylen,
-		       tcp_conn->in.offset, tcp_conn->in.copy);
+	if (hash) {
+		chunk->hash = hash;
+		crypto_hash_init(hash);
+	}
+}
 
-		tcp_conn->in.offset += copylen;
-		tcp_conn->in.copy -= copylen;
-		if (copylen < hdr_remains)  {
-			tcp_conn->in_progress = IN_PROGRESS_HEADER_GATHER;
-			tcp_conn->in.hdr_offset += copylen;
-		        return -EAGAIN;
+static inline void
+iscsi_chunk_init_linear(struct iscsi_chunk *chunk, void *data, size_t size,
+			iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+{
+	__iscsi_chunk_init(chunk, size, done, hash);
+	chunk->data = data;
+	chunk->size = size;
+}
+
+static inline int
+iscsi_chunk_seek_sg(struct iscsi_chunk *chunk,
+		    struct scatterlist *sg, unsigned int sg_count,
+		    unsigned int offset, size_t size,
+		    iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+{
+	unsigned int i;
+
+	__iscsi_chunk_init(chunk, size, done, hash);
+	for (i = 0; i < sg_count; ++i) {
+		if (offset < sg[i].length) {
+			chunk->sg = sg;
+			chunk->sg_count = sg_count;
+			iscsi_tcp_chunk_init_sg(chunk, i, offset);
+			return 0;
 		}
-		tcp_conn->in.hdr = &tcp_conn->hdr;
-		tcp_conn->discontiguous_hdr_cnt++;
-	        tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
+		offset -= sg[i].length;
 	}
 
+	return ISCSI_ERR_DATA_OFFSET;
+}
+
+/**
+ * iscsi_tcp_hdr_recv_prep - prep chunk for hdr reception
+ * @tcp_conn: iscsi connection to prep for
+ *
+ * This function always passes NULL for the hash argument, because when this
+ * function is called we do not yet know the final size of the header and want
+ * to delay the digest processing until we know that.
+ */
+static void
+iscsi_tcp_hdr_recv_prep(struct iscsi_tcp_conn *tcp_conn)
+{
+	debug_tcp("iscsi_tcp_hdr_recv_prep(%p%s)\n", tcp_conn,
+		  tcp_conn->iscsi_conn->hdrdgst_en ? ", digest enabled" : "");
+	iscsi_chunk_init_linear(&tcp_conn->in.chunk,
+				tcp_conn->in.hdr_buf, sizeof(struct iscsi_hdr),
+				iscsi_tcp_hdr_recv_done, NULL);
+}
+
+/*
+ * Handle incoming reply to any other type of command
+ */
+static int
+iscsi_tcp_data_recv_done(struct iscsi_tcp_conn *tcp_conn,
+			 struct iscsi_chunk *chunk)
+{
+	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
+	int rc = 0;
+
+	if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+		return ISCSI_ERR_DATA_DGST;
+
+	rc = iscsi_complete_pdu(conn, tcp_conn->in.hdr,
+			conn->data, tcp_conn->in.datalen);
+	if (rc)
+		return rc;
+
+	iscsi_tcp_hdr_recv_prep(tcp_conn);
 	return 0;
 }
 
+static void
+iscsi_tcp_data_recv_prep(struct iscsi_tcp_conn *tcp_conn)
+{
+	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
+	struct hash_desc *rx_hash = NULL;
+
+	if (conn->datadgst_en)
+		rx_hash = &tcp_conn->rx_hash;
+
+	iscsi_chunk_init_linear(&tcp_conn->in.chunk,
+				conn->data, tcp_conn->in.datalen,
+				iscsi_tcp_data_recv_done, rx_hash);
+}
+
 /*
  * must be called with session lock
  */
@@ -417,16 +682,49 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	return 0;
 }
 
+/*
+ * Handle incoming reply to DataIn command
+ */
 static int
-iscsi_tcp_hdr_recv(struct iscsi_conn *conn)
+iscsi_tcp_process_data_in(struct iscsi_tcp_conn *tcp_conn,
+			  struct iscsi_chunk *chunk)
+{
+	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
+	struct iscsi_hdr *hdr = tcp_conn->in.hdr;
+	int rc;
+
+	if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+		return ISCSI_ERR_DATA_DGST;
+
+	/* check for non-exceptional status */
+	if (hdr->flags & ISCSI_FLAG_DATA_STATUS) {
+		rc = iscsi_complete_pdu(conn, tcp_conn->in.hdr, NULL, 0);
+		if (rc)
+			return rc;
+	}
+
+	iscsi_tcp_hdr_recv_prep(tcp_conn);
+	return 0;
+}
+
+/**
+ * iscsi_tcp_hdr_dissect - process PDU header
+ * @conn: iSCSI connection
+ * @hdr: PDU header
+ *
+ * This function analyzes the header of the PDU received,
+ * and performs several sanity checks. If the PDU is accompanied
+ * by data, the receive buffer is set up to copy the incoming data
+ * to the correct location.
+ */
+static int
+iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 {
 	int rc = 0, opcode, ahslen;
-	struct iscsi_hdr *hdr;
 	struct iscsi_session *session = conn->session;
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	uint32_t cdgst, rdgst = 0, itt;
-
-	hdr = tcp_conn->in.hdr;
+	struct iscsi_cmd_task *ctask;
+	uint32_t itt;
 
 	/* verify PDU length */
 	tcp_conn->in.datalen = ntoh24(hdr->dlength);
@@ -435,77 +733,72 @@ iscsi_tcp_hdr_recv(struct iscsi_conn *conn)
 		       tcp_conn->in.datalen, conn->max_recv_dlength);
 		return ISCSI_ERR_DATALEN;
 	}
-	tcp_conn->data_copied = 0;
 
-	/* read AHS */
+	/* Additional header segments. So far, we don't
+	 * process additional headers.
+	 */
 	ahslen = hdr->hlength << 2;
-	tcp_conn->in.offset += ahslen;
-	tcp_conn->in.copy -= ahslen;
-	if (tcp_conn->in.copy < 0) {
-		printk(KERN_ERR "iscsi_tcp: can't handle AHS with length "
-		       "%d bytes\n", ahslen);
-		return ISCSI_ERR_AHSLEN;
-	}
-
-	/* calculate read padding */
-	tcp_conn->in.padding = tcp_conn->in.datalen & (ISCSI_PAD_LEN-1);
-	if (tcp_conn->in.padding) {
-		tcp_conn->in.padding = ISCSI_PAD_LEN - tcp_conn->in.padding;
-		debug_scsi("read padding %d bytes\n", tcp_conn->in.padding);
-	}
-
-	if (conn->hdrdgst_en) {
-		struct scatterlist sg;
-
-		sg_init_one(&sg, (u8 *)hdr,
-			    sizeof(struct iscsi_hdr) + ahslen);
-		crypto_hash_digest(&tcp_conn->rx_hash, &sg, sg.length,
-				   (u8 *)&cdgst);
-		rdgst = *(uint32_t*)((char*)hdr + sizeof(struct iscsi_hdr) +
-				     ahslen);
-		if (cdgst != rdgst) {
-			printk(KERN_ERR "iscsi_tcp: hdrdgst error "
-			       "recv 0x%x calc 0x%x\n", rdgst, cdgst);
-			return ISCSI_ERR_HDR_DGST;
-		}
-	}
 
 	opcode = hdr->opcode & ISCSI_OPCODE_MASK;
 	/* verify itt (itt encoding: age+cid+itt) */
 	rc = iscsi_verify_itt(conn, hdr, &itt);
 	if (rc == ISCSI_ERR_NO_SCSI_CMD) {
+		/* XXX: what does this do? */
 		tcp_conn->in.datalen = 0; /* force drop */
 		return 0;
 	} else if (rc)
 		return rc;
 
-	debug_tcp("opcode 0x%x offset %d copy %d ahslen %d datalen %d\n",
-		  opcode, tcp_conn->in.offset, tcp_conn->in.copy,
-		  ahslen, tcp_conn->in.datalen);
+	debug_tcp("opcode 0x%x ahslen %d datalen %d\n",
+		  opcode, ahslen, tcp_conn->in.datalen);
 
 	switch(opcode) {
 	case ISCSI_OP_SCSI_DATA_IN:
-		tcp_conn->in.ctask = session->cmds[itt];
-		rc = iscsi_data_rsp(conn, tcp_conn->in.ctask);
+		ctask = session->cmds[itt];
+		rc = iscsi_data_rsp(conn, ctask);
 		if (rc)
 			return rc;
+		if (tcp_conn->in.datalen) {
+			struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
+			struct hash_desc *rx_hash = NULL;
+
+			/*
+			 * Setup copy of Data-In into the Scsi_Cmnd
+			 * Scatterlist case:
+			 * We set up the iscsi_chunk to point to the next
+			 * scatterlist entry to copy to. As we go along,
+			 * we move on to the next scatterlist entry and
+			 * update the digest per-entry.
+			 */
+			if (conn->datadgst_en)
+				rx_hash = &tcp_conn->rx_hash;
+
+			debug_tcp("iscsi_tcp_begin_data_in(%p, offset=%d, "
+				  "datalen=%d)\n", tcp_conn,
+				  tcp_ctask->data_offset,
+				  tcp_conn->in.datalen);
+			return iscsi_chunk_seek_sg(&tcp_conn->in.chunk,
+						scsi_sglist(ctask->sc),
+						scsi_sg_count(ctask->sc),
+						tcp_ctask->data_offset,
+						tcp_conn->in.datalen,
+						iscsi_tcp_process_data_in,
+						rx_hash);
+		}
 		/* fall through */
 	case ISCSI_OP_SCSI_CMD_RSP:
-		tcp_conn->in.ctask = session->cmds[itt];
-		if (tcp_conn->in.datalen)
-			goto copy_hdr;
-
-		spin_lock(&session->lock);
-		rc = __iscsi_complete_pdu(conn, hdr, NULL, 0);
-		spin_unlock(&session->lock);
+		if (tcp_conn->in.datalen) {
+			iscsi_tcp_data_recv_prep(tcp_conn);
+			return 0;
+		}
+		rc = iscsi_complete_pdu(conn, hdr, NULL, 0);
 		break;
 	case ISCSI_OP_R2T:
-		tcp_conn->in.ctask = session->cmds[itt];
+		ctask = session->cmds[itt];
 		if (ahslen)
 			rc = ISCSI_ERR_AHSLEN;
-		else if (tcp_conn->in.ctask->sc->sc_data_direction ==
-								DMA_TO_DEVICE)
-			rc = iscsi_r2t_rsp(conn, tcp_conn->in.ctask);
+		else if (ctask->sc->sc_data_direction == DMA_TO_DEVICE)
+			rc = iscsi_r2t_rsp(conn, ctask);
 		else
 			rc = ISCSI_ERR_PROTO;
 		break;
@@ -518,8 +811,7 @@ iscsi_tcp_hdr_recv(struct iscsi_conn *conn)
 		 * than 8K, but there are no targets that currently do this.
 		 * For now we fail until we find a vendor that needs it
 		 */
-		if (ISCSI_DEF_MAX_RECV_SEG_LEN <
-		    tcp_conn->in.datalen) {
+		if (ISCSI_DEF_MAX_RECV_SEG_LEN < tcp_conn->in.datalen) {
 			printk(KERN_ERR "iscsi_tcp: received buffer of len %u "
 			      "but conn buffer is only %u (opcode %0x)\n",
 			      tcp_conn->in.datalen,
@@ -528,8 +820,13 @@ iscsi_tcp_hdr_recv(struct iscsi_conn *conn)
 			break;
 		}
 
-		if (tcp_conn->in.datalen)
-			goto copy_hdr;
+		/* If there's data coming in with the response,
+		 * receive it to the connection's buffer.
+		 */
+		if (tcp_conn->in.datalen) {
+			iscsi_tcp_data_recv_prep(tcp_conn);
+			return 0;
+		}
 	/* fall through */
 	case ISCSI_OP_LOGOUT_RSP:
 	case ISCSI_OP_NOOP_IN:
@@ -541,129 +838,15 @@ iscsi_tcp_hdr_recv(struct iscsi_conn *conn)
 		break;
 	}
 
-	return rc;
-
-copy_hdr:
-	/*
-	 * if we did zero copy for the header but we will need multiple
-	 * skbs to complete the command then we have to copy the header
-	 * for later use
-	 */
-	if (tcp_conn->in.zero_copy_hdr && tcp_conn->in.copy <=
-	   (tcp_conn->in.datalen + tcp_conn->in.padding +
-	    (conn->datadgst_en ? 4 : 0))) {
-		debug_tcp("Copying header for later use. in.copy %d in.datalen"
-			  " %d\n", tcp_conn->in.copy, tcp_conn->in.datalen);
-		memcpy(&tcp_conn->hdr, tcp_conn->in.hdr,
-		       sizeof(struct iscsi_hdr));
-		tcp_conn->in.hdr = &tcp_conn->hdr;
-		tcp_conn->in.zero_copy_hdr = 0;
-	}
-	return 0;
-}
-
-/**
- * iscsi_ctask_copy - copy skb bits to the destanation cmd task
- * @conn: iscsi tcp connection
- * @ctask: scsi command task
- * @buf: buffer to copy to
- * @buf_size: size of buffer
- * @offset: offset within the buffer
- *
- * Notes:
- *	The function calls skb_copy_bits() and updates per-connection and
- *	per-cmd byte counters.
- *
- *	Read counters (in bytes):
- *
- *	conn->in.offset		offset within in progress SKB
- *	conn->in.copy		left to copy from in progress SKB
- *				including padding
- *	conn->in.copied		copied already from in progress SKB
- *	conn->data_copied	copied already from in progress buffer
- *	ctask->sent		total bytes sent up to the MidLayer
- *	ctask->data_count	left to copy from in progress Data-In
- *	buf_left		left to copy from in progress buffer
- **/
-static inline int
-iscsi_ctask_copy(struct iscsi_tcp_conn *tcp_conn, struct iscsi_cmd_task *ctask,
-		void *buf, int buf_size, int offset)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	int buf_left = buf_size - (tcp_conn->data_copied + offset);
-	unsigned size = min(tcp_conn->in.copy, buf_left);
-	int rc;
-
-	size = min(size, ctask->data_count);
-
-	debug_tcp("ctask_copy %d bytes at offset %d copied %d\n",
-	       size, tcp_conn->in.offset, tcp_conn->in.copied);
-
-	BUG_ON(size <= 0);
-	BUG_ON(tcp_ctask->sent + size > scsi_bufflen(ctask->sc));
-
-	rc = skb_copy_bits(tcp_conn->in.skb, tcp_conn->in.offset,
-			   (char*)buf + (offset + tcp_conn->data_copied), size);
-	/* must fit into skb->len */
-	BUG_ON(rc);
-
-	tcp_conn->in.offset += size;
-	tcp_conn->in.copy -= size;
-	tcp_conn->in.copied += size;
-	tcp_conn->data_copied += size;
-	tcp_ctask->sent += size;
-	ctask->data_count -= size;
-
-	BUG_ON(tcp_conn->in.copy < 0);
-	BUG_ON(ctask->data_count < 0);
-
-	if (buf_size != (tcp_conn->data_copied + offset)) {
-		if (!ctask->data_count) {
-			BUG_ON(buf_size - tcp_conn->data_copied < 0);
-			/* done with this PDU */
-			return buf_size - tcp_conn->data_copied;
-		}
-		return -EAGAIN;
+	if (rc == 0) {
+		/* Anything that comes with data should have
+		 * been handled above. */
+		if (tcp_conn->in.datalen)
+			return ISCSI_ERR_PROTO;
+		iscsi_tcp_hdr_recv_prep(tcp_conn);
 	}
 
-	/* done with this buffer or with both - PDU and buffer */
-	tcp_conn->data_copied = 0;
-	return 0;
-}
-
-/**
- * iscsi_tcp_copy - copy skb bits to the destanation buffer
- * @conn: iscsi tcp connection
- *
- * Notes:
- *	The function calls skb_copy_bits() and updates per-connection
- *	byte counters.
- **/
-static inline int
-iscsi_tcp_copy(struct iscsi_conn *conn, int buf_size)
-{
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	int buf_left = buf_size - tcp_conn->data_copied;
-	int size = min(tcp_conn->in.copy, buf_left);
-	int rc;
-
-	debug_tcp("tcp_copy %d bytes at offset %d copied %d\n",
-	       size, tcp_conn->in.offset, tcp_conn->data_copied);
-	BUG_ON(size <= 0);
-
-	rc = skb_copy_bits(tcp_conn->in.skb, tcp_conn->in.offset,
-			   (char*)conn->data + tcp_conn->data_copied, size);
-	BUG_ON(rc);
-
-	tcp_conn->in.offset += size;
-	tcp_conn->in.copy -= size;
-	tcp_conn->in.copied += size;
-	tcp_conn->data_copied += size;
-
-	if (buf_size != tcp_conn->data_copied)
-		return -EAGAIN;
-
-	return 0;
+	return rc;
 }
 
 static inline void
@@ -677,325 +860,146 @@ partial_sg_digest_update(struct hash_desc *desc, struct scatterlist *sg,
 	crypto_hash_update(desc, &temp, length);
 }
 
-static void
-iscsi_recv_digest_update(struct iscsi_tcp_conn *tcp_conn, char* buf, int len)
-{
-	struct scatterlist tmp;
-
-	sg_init_one(&tmp, buf, len);
-	crypto_hash_update(&tcp_conn->rx_hash, &tmp, len);
-}
-
-static int iscsi_scsi_data_in(struct iscsi_conn *conn)
+/**
+ * iscsi_tcp_hdr_recv_done - process PDU header
+ *
+ * This is the callback invoked when the PDU header has
+ * been received. If the header is followed by additional
+ * header segments, we go back for more data.
+ */
+static int
+iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
+			struct iscsi_chunk *chunk)
 {
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	struct iscsi_cmd_task *ctask = tcp_conn->in.ctask;
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct scsi_cmnd *sc = ctask->sc;
-	struct scatterlist *sg;
-	int i, offset, rc = 0;
-
-	BUG_ON((void*)ctask != sc->SCp.ptr);
-
-	offset = tcp_ctask->data_offset;
-	sg = scsi_sglist(sc);
-
-	if (tcp_ctask->data_offset)
-		for (i = 0; i < tcp_ctask->sg_count; i++)
-			offset -= sg[i].length;
-	/* we've passed through partial sg*/
-	if (offset < 0)
-		offset = 0;
-
-	for (i = tcp_ctask->sg_count; i < scsi_sg_count(sc); i++) {
-		char *dest;
-
-		dest = kmap_atomic(sg_page(&sg[i]), KM_SOFTIRQ0);
-		rc = iscsi_ctask_copy(tcp_conn, ctask, dest + sg[i].offset,
-				      sg[i].length, offset);
-		kunmap_atomic(dest, KM_SOFTIRQ0);
-		if (rc == -EAGAIN)
-			/* continue with the next SKB/PDU */
-			return rc;
-		if (!rc) {
-			if (conn->datadgst_en) {
-				if (!offset)
-					crypto_hash_update(
-							&tcp_conn->rx_hash,
-							&sg[i], sg[i].length);
-				else
-					partial_sg_digest_update(
-							&tcp_conn->rx_hash,
-							&sg[i],
-							sg[i].offset + offset,
-							sg[i].length - offset);
-			}
-			offset = 0;
-			tcp_ctask->sg_count++;
-		}
-
-		if (!ctask->data_count) {
-			if (rc && conn->datadgst_en)
-				/*
-				 * data-in is complete, but buffer not...
-				 */
-				partial_sg_digest_update(&tcp_conn->rx_hash,
-							 &sg[i],
-							 sg[i].offset,
-							 sg[i].length-rc);
-			rc = 0;
-			break;
-		}
-
-		if (!tcp_conn->in.copy)
-			return -EAGAIN;
-	}
-	BUG_ON(ctask->data_count);
+	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
+	struct iscsi_hdr *hdr;
 
-	/* check for non-exceptional status */
-	if (tcp_conn->in.hdr->flags & ISCSI_FLAG_DATA_STATUS) {
-		debug_scsi("done [sc %lx res %d itt 0x%x flags 0x%x]\n",
-			   (long)sc, sc->result, ctask->itt,
-			   tcp_conn->in.hdr->flags);
-		spin_lock(&conn->session->lock);
-		__iscsi_complete_pdu(conn, tcp_conn->in.hdr, NULL, 0);
-		spin_unlock(&conn->session->lock);
+	/* Check if there are additional header segments
+	 * *prior* to computing the digest, because we
+	 * may need to go back to the caller for more.
+	 */
+	hdr = (struct iscsi_hdr *) tcp_conn->in.hdr_buf;
+	if (chunk->copied == sizeof(struct iscsi_hdr) && hdr->hlength) {
+		/* Bump the header length - the caller will
+		 * just loop around and get the AHS for us, and
+		 * call again. */
+		unsigned int ahslen = hdr->hlength << 2;
+
+		/* Make sure we don't overflow */
+		if (sizeof(*hdr) + ahslen > sizeof(tcp_conn->in.hdr_buf))
+			return ISCSI_ERR_AHSLEN;
+
+		chunk->total_size += ahslen;
+		chunk->size += ahslen;
+		return 0;
 	}
 
-	return rc;
-}
-
-static int
-iscsi_data_recv(struct iscsi_conn *conn)
-{
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	int rc = 0, opcode;
-
-	opcode = tcp_conn->in.hdr->opcode & ISCSI_OPCODE_MASK;
-	switch (opcode) {
-	case ISCSI_OP_SCSI_DATA_IN:
-		rc = iscsi_scsi_data_in(conn);
-		break;
-	case ISCSI_OP_SCSI_CMD_RSP:
-	case ISCSI_OP_TEXT_RSP:
-	case ISCSI_OP_LOGIN_RSP:
-	case ISCSI_OP_ASYNC_EVENT:
-	case ISCSI_OP_REJECT:
-		/*
-		 * Collect data segment to the connection's data
-		 * placeholder
-		 */
-		if (iscsi_tcp_copy(conn, tcp_conn->in.datalen)) {
-			rc = -EAGAIN;
-			goto exit;
+	/* We're done processing the header. See if we're doing
+	 * header digests; if so, set up the recv_digest buffer
+	 * and go back for more. */
+	if (conn->hdrdgst_en) {
+		if (chunk->digest_len == 0) {
+			iscsi_tcp_chunk_splice_digest(chunk,
+						      chunk->recv_digest);
+			return 0;
 		}
+		iscsi_tcp_dgst_header(&tcp_conn->rx_hash, hdr,
+				      chunk->total_copied - ISCSI_DIGEST_SIZE,
+				      chunk->digest);
 
-		rc = iscsi_complete_pdu(conn, tcp_conn->in.hdr, conn->data,
-					tcp_conn->in.datalen);
-		if (!rc && conn->datadgst_en && opcode != ISCSI_OP_LOGIN_RSP)
-			iscsi_recv_digest_update(tcp_conn, conn->data,
-			  			tcp_conn->in.datalen);
-		break;
-	default:
-		BUG_ON(1);
+		if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+			return ISCSI_ERR_HDR_DGST;
 	}
-exit:
-	return rc;
+
+	tcp_conn->in.hdr = hdr;
+	return iscsi_tcp_hdr_dissect(conn, hdr);
 }
 
 /**
- * iscsi_tcp_data_recv - TCP receive in sendfile fashion
+ * iscsi_tcp_recv - TCP receive in sendfile fashion
  * @rd_desc: read descriptor
  * @skb: socket buffer
  * @offset: offset in skb
  * @len: skb->len - offset
  **/
 static int
-iscsi_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
-		unsigned int offset, size_t len)
+iscsi_tcp_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
+	       unsigned int offset, size_t len)
 {
-	int rc;
 	struct iscsi_conn *conn = rd_desc->arg.data;
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	int processed;
-	char pad[ISCSI_PAD_LEN];
-	struct scatterlist sg;
-
-	/*
-	 * Save current SKB and its offset in the corresponding
-	 * connection context.
-	 */
-	tcp_conn->in.copy = skb->len - offset;
-	tcp_conn->in.offset = offset;
-	tcp_conn->in.skb = skb;
-	tcp_conn->in.len = tcp_conn->in.copy;
-	BUG_ON(tcp_conn->in.copy <= 0);
-	debug_tcp("in %d bytes\n", tcp_conn->in.copy);
+	struct iscsi_chunk *chunk = &tcp_conn->in.chunk;
+	struct skb_seq_state seq;
+	unsigned int consumed = 0;
+	int rc = 0;
 
-more:
-	tcp_conn->in.copied = 0;
-	rc = 0;
+	debug_tcp("in %d bytes\n", skb->len - offset);
 
 	if (unlikely(conn->suspend_rx)) {
 		debug_tcp("conn %d Rx suspended!\n", conn->id);
 		return 0;
 	}
 
-	if (tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER ||
-	    tcp_conn->in_progress == IN_PROGRESS_HEADER_GATHER) {
-		rc = iscsi_hdr_extract(tcp_conn);
-		if (rc) {
-		       if (rc == -EAGAIN)
-				goto nomore;
-		       else {
-				iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-				return 0;
-		       }
-		}
-
-		/*
-		 * Verify and process incoming PDU header.
-		 */
-		rc = iscsi_tcp_hdr_recv(conn);
-		if (!rc && tcp_conn->in.datalen) {
-			if (conn->datadgst_en)
-				crypto_hash_init(&tcp_conn->rx_hash);
-			tcp_conn->in_progress = IN_PROGRESS_DATA_RECV;
-		} else if (rc) {
-			iscsi_conn_failure(conn, rc);
-			return 0;
-		}
-	}
-
-	if (tcp_conn->in_progress == IN_PROGRESS_DDIGEST_RECV &&
-	    tcp_conn->in.copy) {
-		uint32_t recv_digest;
+	skb_prepare_seq_read(skb, offset, skb->len, &seq);
+	while (1) {
+		unsigned int avail;
+		const u8 *ptr;
 
-		debug_tcp("extra data_recv offset %d copy %d\n",
-			  tcp_conn->in.offset, tcp_conn->in.copy);
-
-		if (!tcp_conn->data_copied) {
-			if (tcp_conn->in.padding) {
-				debug_tcp("padding -> %d\n",
-					  tcp_conn->in.padding);
-				memset(pad, 0, tcp_conn->in.padding);
-				sg_init_one(&sg, pad, tcp_conn->in.padding);
-				crypto_hash_update(&tcp_conn->rx_hash,
-						   &sg, sg.length);
+		avail = skb_seq_read(consumed, &ptr, &seq);
+		if (avail == 0)
+			break;
+		BUG_ON(chunk->copied >= chunk->size);
+
+		debug_tcp("skb %p ptr=%p avail=%u\n", skb, ptr, avail);
+		rc = iscsi_tcp_chunk_recv(tcp_conn, chunk, ptr, avail);
+		BUG_ON(rc == 0);
+		consumed += rc;
+
+		if (chunk->total_copied >= chunk->total_size) {
+			rc = chunk->done(tcp_conn, chunk);
+			if (rc != 0) {
+				skb_abort_seq_read(&seq);
+				goto error;
 			}
-			crypto_hash_final(&tcp_conn->rx_hash,
-					  (u8 *) &tcp_conn->in.datadgst);
-			debug_tcp("rx digest 0x%x\n", tcp_conn->in.datadgst);
-		}
-
-		rc = iscsi_tcp_copy(conn, sizeof(uint32_t));
-		if (rc) {
-			if (rc == -EAGAIN)
-				goto again;
-			iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-			return 0;
-		}
-
-		memcpy(&recv_digest, conn->data, sizeof(uint32_t));
-		if (recv_digest != tcp_conn->in.datadgst) {
-			debug_tcp("iscsi_tcp: data digest error!"
-				  "0x%x != 0x%x\n", recv_digest,
-				  tcp_conn->in.datadgst);
-			iscsi_conn_failure(conn, ISCSI_ERR_DATA_DGST);
-			return 0;
-		} else {
-			debug_tcp("iscsi_tcp: data digest match!"
-				  "0x%x == 0x%x\n", recv_digest,
-				  tcp_conn->in.datadgst);
-			tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
-		}
-	}
-
-	if (tcp_conn->in_progress == IN_PROGRESS_DATA_RECV &&
-	    tcp_conn->in.copy) {
-		debug_tcp("data_recv offset %d copy %d\n",
-		       tcp_conn->in.offset, tcp_conn->in.copy);
 
-		rc = iscsi_data_recv(conn);
-		if (rc) {
-			if (rc == -EAGAIN)
-				goto again;
-			iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-			return 0;
+			/* The done() functions sets up the
+			 * next chunk. */
 		}
-
-		if (tcp_conn->in.padding)
-			tcp_conn->in_progress = IN_PROGRESS_PAD_RECV;
-		else if (conn->datadgst_en)
-			tcp_conn->in_progress = IN_PROGRESS_DDIGEST_RECV;
-		else
-			tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
-		tcp_conn->data_copied = 0;
-	}
-
-	if (tcp_conn->in_progress == IN_PROGRESS_PAD_RECV &&
-	    tcp_conn->in.copy) {
-		int copylen = min(tcp_conn->in.padding - tcp_conn->data_copied,
-				  tcp_conn->in.copy);
-
-		tcp_conn->in.copy -= copylen;
-		tcp_conn->in.offset += copylen;
-		tcp_conn->data_copied += copylen;
-
-		if (tcp_conn->data_copied != tcp_conn->in.padding)
-			tcp_conn->in_progress = IN_PROGRESS_PAD_RECV;
-		else if (conn->datadgst_en)
-			tcp_conn->in_progress = IN_PROGRESS_DDIGEST_RECV;
-		else
-			tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
-		tcp_conn->data_copied = 0;
 	}
 
-	debug_tcp("f, processed %d from out of %d padding %d\n",
-	       tcp_conn->in.offset - offset, (int)len, tcp_conn->in.padding);
-	BUG_ON(tcp_conn->in.offset - offset > len);
+	conn->rxdata_octets += consumed;
+	return consumed;
 
-	if (tcp_conn->in.offset - offset != len) {
-		debug_tcp("continue to process %d bytes\n",
-		       (int)len - (tcp_conn->in.offset - offset));
-		goto more;
-	}
-
-nomore:
-	processed = tcp_conn->in.offset - offset;
-	BUG_ON(processed == 0);
-	return processed;
-
-again:
-	processed = tcp_conn->in.offset - offset;
-	debug_tcp("c, processed %d from out of %d rd_desc_cnt %d\n",
-	          processed, (int)len, (int)rd_desc->count);
-	BUG_ON(processed == 0);
-	BUG_ON(processed > len);
-
-	conn->rxdata_octets += processed;
-	return processed;
+error:
+	debug_tcp("Error receiving PDU, errno=%d\n", rc);
+	iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+	return 0;
 }
 
 static void
 iscsi_tcp_data_ready(struct sock *sk, int flag)
 {
 	struct iscsi_conn *conn = sk->sk_user_data;
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 	read_descriptor_t rd_desc;
 
 	read_lock(&sk->sk_callback_lock);
 
 	/*
-	 * Use rd_desc to pass 'conn' to iscsi_tcp_data_recv.
+	 * Use rd_desc to pass 'conn' to iscsi_tcp_recv.
 	 * We set count to 1 because we want the network layer to
-	 * hand us all the skbs that are available. iscsi_tcp_data_recv
+	 * hand us all the skbs that are available. iscsi_tcp_recv
 	 * handled pdus that cross buffers or pdus that still need data.
 	 */
 	rd_desc.arg.data = conn;
 	rd_desc.count = 1;
-	tcp_read_sock(sk, &rd_desc, iscsi_tcp_data_recv);
+	tcp_read_sock(sk, &rd_desc, iscsi_tcp_recv);
 
 	read_unlock(&sk->sk_callback_lock);
+
+	/* If we had to (atomically) map a highmem page,
+	 * unmap it now. */
+	iscsi_tcp_chunk_unmap(&tcp_conn->in.chunk);
 }
 
 static void
@@ -1097,9 +1101,9 @@ iscsi_send(struct iscsi_conn *conn, struct iscsi_buf *buf, int size, int flags)
 	 * slab case.
 	 */
 	if (buf->use_sendmsg)
-		res = sock_no_sendpage(sk, sg_page(&buf->sg), offset, size, flags);
+		res = sock_no_sendpage(sk, buf->sg.page, offset, size, flags);
 	else
-		res = tcp_conn->sendpage(sk, sg_page(&buf->sg), offset, size, flags);
+		res = tcp_conn->sendpage(sk, buf->sg.page, offset, size, flags);
 
 	if (res >= 0) {
 		conn->txdata_octets += res;
@@ -1783,9 +1787,6 @@ iscsi_tcp_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 
 	conn->dd_data = tcp_conn;
 	tcp_conn->iscsi_conn = conn;
-	tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
-	/* initial operational parameters */
-	tcp_conn->hdr_size = sizeof(struct iscsi_hdr);
 
 	tcp_conn->tx_hash.tfm = crypto_alloc_hash("crc32c", 0,
 						  CRYPTO_ALG_ASYNC);
@@ -1862,11 +1863,9 @@ static void
 iscsi_tcp_conn_stop(struct iscsi_cls_conn *cls_conn, int flag)
 {
 	struct iscsi_conn *conn = cls_conn->dd_data;
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 
 	iscsi_conn_stop(cls_conn, flag);
 	iscsi_tcp_release_conn(conn);
-	tcp_conn->hdr_size = sizeof(struct iscsi_hdr);
 }
 
 static int iscsi_tcp_get_addr(struct iscsi_conn *conn, struct socket *sock,
@@ -1966,7 +1965,7 @@ iscsi_tcp_conn_bind(struct iscsi_cls_session *cls_session,
 	/*
 	 * set receive state machine into initial state
 	 */
-	tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER;
+	iscsi_tcp_hdr_recv_prep(tcp_conn);
 	return 0;
 
 free_socket:
@@ -2059,9 +2058,6 @@ iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param,
 	switch(param) {
 	case ISCSI_PARAM_HDRDGST_EN:
 		iscsi_set_param(cls_conn, param, buf, buflen);
-		tcp_conn->hdr_size = sizeof(struct iscsi_hdr);
-		if (conn->hdrdgst_en)
-			tcp_conn->hdr_size += sizeof(__u32);
 		break;
 	case ISCSI_PARAM_DATADGST_EN:
 		iscsi_set_param(cls_conn, param, buf, buflen);
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index 7eba44df0a7f..f1c541151100 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -24,13 +24,6 @@
 
 #include <scsi/libiscsi.h>
 
-/* Socket's Receive state machine */
-#define IN_PROGRESS_WAIT_HEADER		0x0
-#define IN_PROGRESS_HEADER_GATHER	0x1
-#define IN_PROGRESS_DATA_RECV		0x2
-#define IN_PROGRESS_DDIGEST_RECV	0x3
-#define IN_PROGRESS_PAD_RECV		0x4
-
 /* xmit state machine */
 #define XMSTATE_IDLE			0x0
 #define XMSTATE_CMD_HDR_INIT		0x1
@@ -54,41 +47,64 @@
 
 struct crypto_hash;
 struct socket;
+struct iscsi_tcp_conn;
+struct iscsi_chunk;
+
+typedef int iscsi_chunk_done_fn_t(struct iscsi_tcp_conn *,
+				  struct iscsi_chunk *);
+
+struct iscsi_chunk {
+	unsigned char		*data;
+	unsigned int		size;
+	unsigned int		copied;
+	unsigned int		total_size;
+	unsigned int		total_copied;
+
+	struct hash_desc	*hash;
+	unsigned char		recv_digest[ISCSI_DIGEST_SIZE];
+	unsigned char		digest[ISCSI_DIGEST_SIZE];
+	unsigned int		digest_len;
+
+	struct scatterlist	*sg;
+	void			*sg_mapped;
+	unsigned int		sg_offset;
+	unsigned int		sg_index;
+	unsigned int		sg_count;
+
+	iscsi_chunk_done_fn_t	*done;
+};
 
 /* Socket connection recieve helper */
 struct iscsi_tcp_recv {
 	struct iscsi_hdr	*hdr;
-	struct sk_buff		*skb;
-	int			offset;
-	int			len;
-	int			hdr_offset;
-	int			copy;
-	int			copied;
-	int			padding;
-	struct iscsi_cmd_task	*ctask;		/* current cmd in progress */
+	struct iscsi_chunk	chunk;
+
+	/* Allocate buffer for BHS + AHS */
+	uint32_t		hdr_buf[64];
 
 	/* copied and flipped values */
 	int			datalen;
-	int			datadgst;
-	char			zero_copy_hdr;
+};
+
+/* Socket connection send helper */
+struct iscsi_tcp_send {
+	struct iscsi_hdr	*hdr;
+	struct iscsi_chunk	chunk;
+	struct iscsi_chunk	data_chunk;
+
+	/* Allocate buffer for BHS + AHS */
+	uint32_t		hdr_buf[64];
 };
 
 struct iscsi_tcp_conn {
 	struct iscsi_conn	*iscsi_conn;
 	struct socket		*sock;
-	struct iscsi_hdr	hdr;		/* header placeholder */
-	char			hdrext[4*sizeof(__u16) +
-				    sizeof(__u32)];
-	int			data_copied;
 	int			stop_stage;	/* conn_stop() flag: *
 						 * stop to recover,  *
 						 * stop to terminate */
-	/* iSCSI connection-wide sequencing */
-	int			hdr_size;	/* PDU header size */
-
 	/* control data */
 	struct iscsi_tcp_recv	in;		/* TCP receive context */
-	int			in_progress;	/* connection state machine */
+	struct iscsi_tcp_send	out;		/* TCP send context */
 
 	/* old values for socket callbacks */
 	void			(*old_data_ready)(struct sock *, int);
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index 89429f433f85..e1fb3d0927b0 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -77,6 +77,10 @@ enum {
 
 #define ISCSI_ADDRESS_BUF_LEN		64
 
+enum {
+	ISCSI_DIGEST_SIZE = sizeof(__u32),
+};
+
 struct iscsi_mgmt_task {
 	/*
 	 * Becuae LLDs allocate their hdr differently, this is a pointer to
-- 
cgit v1.2.3


From 004d6530f83bee43a55b51bb5960db96e7ae0ffa Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Thu, 13 Dec 2007 12:43:23 -0600
Subject: [SCSI] iscsi_tcp, libiscsi: initial AHS Support

  at libiscsi generic code
  - currently code assumes a storage space of pdu header is allocated
    at llds ctask and is pointed to by iscsi_cmd_task->hdr. Here I add
    a hdr_max field pertaining to that storage, and an hdr_len that
    accumulates the current use of the pdu-header.

  - Add an iscsi_next_hdr() inline which returns the next free space
    to write new Header at. Also iscsi_next_hdr() is used to retrieve
    the address at which to write the header-digest.

  - Add iscsi_add_hdr(length). What the user do is calls iscsi_next_hdr()
    for address of the new header, than calls iscsi_add_hdr(length) with
    the size of the new header. iscsi_add_hdr() will check if space is
    available and update to the new size. length must be padded according
    to standard.

  - Add 2 padding inline helpers thanks to Olaf. Current patch does not
    use them but Following patches will.
    Also moved definition of ISCSI_PAD_LEN to iscsi_proto.h which had
    PAD_WORD_LEN that was never used anywhere.

  - Let iscsi_prep_scsi_cmd_pdu() signal an Error return since now  it is
    possible that it will fail.

  - I was tired of yet again writing a "this is a digest" comment next to
    sizeof(__u32) so I defined a new ISCSI_DIGEST_SIZE. Now I don't need
    any comments. Changed all places that used sizeof(__u32) or "4" in
    connection to a digest.

  iscsi_tcp specific code
  - At struct iscsi_tcp_cmd_task allocate maximum space allowed in
    standard for all headers following the iscsi_cmd header. and mark
    it so in iscsi_tcp_session_create()
  - At iscsi_send_cmd_hdr() retrieve the correct headers size and
    write header digest at iscsi_next_hdr().

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Olaf Kirch <olaf.kirch@oracle.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/iscsi_tcp.c   | 16 ++++++++--------
 drivers/scsi/iscsi_tcp.h   | 13 +++++++------
 drivers/scsi/libiscsi.c    | 41 +++++++++++++++++++++++++++++++++++++++--
 include/scsi/iscsi_proto.h | 10 +++++++++-
 include/scsi/libiscsi.h    | 33 +++++++++++++++++++++++++++++++--
 5 files changed, 94 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index fd88777df28b..491845f18710 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -113,7 +113,7 @@ iscsi_hdr_digest(struct iscsi_conn *conn, struct iscsi_buf *buf,
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 
 	crypto_hash_digest(&tcp_conn->tx_hash, &buf->sg, buf->sg.length, crc);
-	buf->sg.length += sizeof(u32);
+	buf->sg.length += ISCSI_DIGEST_SIZE;
 }
 
 /*
@@ -220,6 +220,7 @@ static inline int
 iscsi_tcp_chunk_done(struct iscsi_chunk *chunk)
 {
 	static unsigned char padbuf[ISCSI_PAD_LEN];
+	unsigned int pad;
 
 	if (chunk->copied < chunk->size) {
 		iscsi_tcp_chunk_map(chunk);
@@ -243,10 +244,8 @@ iscsi_tcp_chunk_done(struct iscsi_chunk *chunk)
 	}
 
 	/* Do we need to handle padding? */
-	if (chunk->total_copied & (ISCSI_PAD_LEN-1)) {
-		unsigned int pad;
-
-		pad = ISCSI_PAD_LEN - (chunk->total_copied & (ISCSI_PAD_LEN-1));
+	pad = iscsi_padding(chunk->total_copied);
+	if (pad != 0) {
 		debug_tcp("consume %d pad bytes\n", pad);
 		chunk->total_size += pad;
 		chunk->size = pad;
@@ -1385,11 +1384,11 @@ iscsi_send_cmd_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 		}
 
 		iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)ctask->hdr,
-				  sizeof(struct iscsi_hdr));
+				  ctask->hdr_len);
 
 		if (conn->hdrdgst_en)
 			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
-					 (u8*)tcp_ctask->hdrext);
+					 iscsi_next_hdr(ctask));
 		tcp_ctask->xmstate &= ~XMSTATE_CMD_HDR_INIT;
 		tcp_ctask->xmstate |= XMSTATE_CMD_HDR_XMIT;
 	}
@@ -2176,7 +2175,8 @@ iscsi_tcp_session_create(struct iscsi_transport *iscsit,
 		struct iscsi_cmd_task *ctask = session->cmds[cmd_i];
 		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 
-		ctask->hdr = &tcp_ctask->hdr;
+		ctask->hdr = &tcp_ctask->hdr.cmd_hdr;
+		ctask->hdr_max = sizeof(tcp_ctask->hdr) - ISCSI_DIGEST_SIZE;
 	}
 
 	for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) {
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index f1c541151100..eb3784f949fd 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -41,7 +41,6 @@
 #define XMSTATE_IMM_HDR_INIT		0x1000
 #define XMSTATE_SOL_HDR_INIT		0x2000
 
-#define ISCSI_PAD_LEN			4
 #define ISCSI_SG_TABLESIZE		SG_ALL
 #define ISCSI_TCP_MAX_CMD_LEN		16
 
@@ -130,14 +129,14 @@ struct iscsi_buf {
 
 struct iscsi_data_task {
 	struct iscsi_data	hdr;			/* PDU */
-	char			hdrext[sizeof(__u32)];	/* Header-Digest */
+	char			hdrext[ISCSI_DIGEST_SIZE];/* Header-Digest */
 	struct iscsi_buf	digestbuf;		/* digest buffer */
 	uint32_t		digest;			/* data digest */
 };
 
 struct iscsi_tcp_mgmt_task {
 	struct iscsi_hdr	hdr;
-	char			hdrext[sizeof(__u32)]; /* Header-Digest */
+	char			hdrext[ISCSI_DIGEST_SIZE]; /* Header-Digest */
 	int			xmstate;	/* mgmt xmit progress */
 	struct iscsi_buf	headbuf;	/* header buffer */
 	struct iscsi_buf	sendbuf;	/* in progress buffer */
@@ -159,9 +158,11 @@ struct iscsi_r2t_info {
 };
 
 struct iscsi_tcp_cmd_task {
-	struct iscsi_cmd	hdr;
-	char			hdrext[4*sizeof(__u16)+	/* AHS */
-				    sizeof(__u32)];	/* HeaderDigest */
+	struct iscsi_hdr_buff {
+		struct iscsi_cmd	cmd_hdr;
+		char			hdrextbuf[ISCSI_MAX_AHS_SIZE +
+		                                  ISCSI_DIGEST_SIZE];
+	} hdr;
 	char			pad[ISCSI_PAD_LEN];
 	int			pad_count;		/* padded bytes */
 	struct iscsi_buf	headbuf;		/* header buf (xmit) */
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 0beb4c620962..0d7914fd8ef3 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -37,6 +37,9 @@
 #include <scsi/scsi_transport_iscsi.h>
 #include <scsi/libiscsi.h>
 
+static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
+			 int err);
+
 struct iscsi_session *
 class_to_transport_session(struct iscsi_cls_session *cls_session)
 {
@@ -122,6 +125,20 @@ void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *ctask,
 }
 EXPORT_SYMBOL_GPL(iscsi_prep_unsolicit_data_pdu);
 
+static int iscsi_add_hdr(struct iscsi_cmd_task *ctask, unsigned len)
+{
+	unsigned exp_len = ctask->hdr_len + len;
+
+	if (exp_len > ctask->hdr_max) {
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
+	WARN_ON(len & (ISCSI_PAD_LEN - 1)); /* caller must pad the AHS */
+	ctask->hdr_len = exp_len;
+	return 0;
+}
+
 /**
  * iscsi_prep_scsi_cmd_pdu - prep iscsi scsi cmd pdu
  * @ctask: iscsi cmd task
@@ -129,13 +146,19 @@ EXPORT_SYMBOL_GPL(iscsi_prep_unsolicit_data_pdu);
  * Prep basic iSCSI PDU fields for a scsi cmd pdu. The LLD should set
  * fields like dlength or final based on how much data it sends
  */
-static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
+static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 {
 	struct iscsi_conn *conn = ctask->conn;
 	struct iscsi_session *session = conn->session;
 	struct iscsi_cmd *hdr = ctask->hdr;
 	struct scsi_cmnd *sc = ctask->sc;
+	unsigned hdrlength;
+	int rc;
 
+	ctask->hdr_len = 0;
+	rc = iscsi_add_hdr(ctask, sizeof(*hdr));
+	if (rc)
+		return rc;
         hdr->opcode = ISCSI_OP_SCSI_CMD;
         hdr->flags = ISCSI_ATTR_SIMPLE;
         int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun);
@@ -199,6 +222,15 @@ static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 			hdr->flags |= ISCSI_FLAG_CMD_READ;
 	}
 
+	/* calculate size of additional header segments (AHSs) */
+	hdrlength = ctask->hdr_len - sizeof(*hdr);
+
+	WARN_ON(hdrlength & (ISCSI_PAD_LEN-1));
+	hdrlength /= ISCSI_PAD_LEN;
+
+	WARN_ON(hdrlength >= 256);
+	hdr->hlength = hdrlength & 0xFF;
+
 	conn->scsicmd_pdus_cnt++;
 
         debug_scsi("iscsi prep [%s cid %d sc %p cdb 0x%x itt 0x%x len %d "
@@ -206,6 +238,7 @@ static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
                 sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read",
 		conn->id, sc, sc->cmnd[0], ctask->itt, scsi_bufflen(sc),
                 session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1);
+	return 0;
 }
 
 /**
@@ -744,7 +777,10 @@ check_mgmt:
 
 		conn->ctask = list_entry(conn->xmitqueue.next,
 					 struct iscsi_cmd_task, running);
-		iscsi_prep_scsi_cmd_pdu(conn->ctask);
+		if (iscsi_prep_scsi_cmd_pdu(conn->ctask)) {
+			fail_command(conn, conn->ctask, DID_ABORT << 16);
+			continue;
+		}
 		conn->session->tt->init_cmd_task(conn->ctask);
 		conn->ctask->state = ISCSI_TASK_RUNNING;
 		list_move_tail(conn->xmitqueue.next, &conn->run_list);
@@ -1534,6 +1570,7 @@ iscsi_session_setup(struct iscsi_transport *iscsit,
 		if (cmd_task_size)
 			ctask->dd_data = &ctask[1];
 		ctask->itt = cmd_i;
+		ctask->hdr_max = sizeof(struct iscsi_cmd);
 		INIT_LIST_HEAD(&ctask->running);
 	}
 
diff --git a/include/scsi/iscsi_proto.h b/include/scsi/iscsi_proto.h
index 751c81eaa7f3..6947082eee6d 100644
--- a/include/scsi/iscsi_proto.h
+++ b/include/scsi/iscsi_proto.h
@@ -27,7 +27,7 @@
 #define ISCSI_LISTEN_PORT	3260
 
 /* Padding word length */
-#define PAD_WORD_LEN		4
+#define ISCSI_PAD_LEN		4
 
 /*
  * useful common(control and data pathes) macro
@@ -147,6 +147,14 @@ struct iscsi_rlength_ahdr {
 	__be32 read_length;
 };
 
+/* Extended CDB AHS */
+struct iscsi_ecdb_ahdr {
+	__be16 ahslength;	/* CDB length - 15, including reserved byte */
+	uint8_t ahstype;
+	uint8_t reserved;
+	uint8_t ecdb[260 - 16];	/* 4-byte aligned extended CDB spillover */
+};
+
 /* SCSI Response Header */
 struct iscsi_cmd_rsp {
 	uint8_t opcode;
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index e1fb3d0927b0..a9a9e869188d 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -78,6 +78,9 @@ enum {
 #define ISCSI_ADDRESS_BUF_LEN		64
 
 enum {
+	/* this is the maximum possible storage for AHSs */
+	ISCSI_MAX_AHS_SIZE = sizeof(struct iscsi_ecdb_ahdr) +
+				sizeof(struct iscsi_rlength_ahdr),
 	ISCSI_DIGEST_SIZE = sizeof(__u32),
 };
 
@@ -102,10 +105,13 @@ enum {
 
 struct iscsi_cmd_task {
 	/*
-	 * Becuae LLDs allocate their hdr differently, this is a pointer to
-	 * that storage. It must be setup at session creation time.
+	 * Because LLDs allocate their hdr differently, this is a pointer
+	 * and length to that storage. It must be setup at session
+	 * creation time.
 	 */
 	struct iscsi_cmd	*hdr;
+	unsigned short		hdr_max;
+	unsigned short		hdr_len;	/* accumulated size of hdr used */
 	int			itt;		/* this ITT */
 
 	uint32_t		unsol_datasn;
@@ -124,6 +130,11 @@ struct iscsi_cmd_task {
 	void			*dd_data;	/* driver/transport data */
 };
 
+static inline void* iscsi_next_hdr(struct iscsi_cmd_task *ctask)
+{
+	return (void*)ctask->hdr + ctask->hdr_len;
+}
+
 struct iscsi_conn {
 	struct iscsi_cls_conn	*cls_conn;	/* ptr to class connection */
 	void			*dd_data;	/* iscsi_transport data */
@@ -342,4 +353,22 @@ extern void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask);
 extern void iscsi_pool_free(struct iscsi_queue *, void **);
 extern int iscsi_pool_init(struct iscsi_queue *, int, void ***, int);
 
+/*
+ * inline functions to deal with padding.
+ */
+static inline unsigned int
+iscsi_padded(unsigned int len)
+{
+	return (len + ISCSI_PAD_LEN - 1) & ~(ISCSI_PAD_LEN - 1);
+}
+
+static inline unsigned int
+iscsi_padding(unsigned int len)
+{
+	len &= (ISCSI_PAD_LEN - 1);
+	if (len)
+		len = ISCSI_PAD_LEN - len;
+	return len;
+}
+
 #endif
-- 
cgit v1.2.3


From 6320377fd94316e58f75d0be6f6e7a644950a4ee Mon Sep 17 00:00:00 2001
From: Olaf Kirch <olaf.kirch@oracle.com>
Date: Thu, 13 Dec 2007 12:43:25 -0600
Subject: [SCSI] libiscsi, iscsi_tcp: iscsi pool cleanup

iscsi_pool_init simplified

iscsi_pool_init currently has a lot of duplicate kfree() calls it does
when some allocation fails. This patch simplifies the code a little by
using iscsi_pool_free to tear down the pool in case of an error.

iscsi_pool_init also returns a copy of the item array to the caller.
Not all callers use this array, so we make it optional.

Instead of allocating a second array and return that, allocate just one
array, of twice the size.

Update users of iscsi_pool_{init,free}

This patch drops the (now useless) second argument to
iscsi_pool_free, and updates all callers.

It also removes the ctask->r2ts array, which was never
used anyway. Since the items argument to iscsi_pool_init
is now optional, we can pass NULL instead.

Signed-off-by: Olaf Kirch <olaf.kirch@oracle.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/iscsi_tcp.c | 12 +++-----
 drivers/scsi/iscsi_tcp.h |  3 +-
 drivers/scsi/libiscsi.c  | 75 ++++++++++++++++++++++++++----------------------
 include/scsi/libiscsi.h  | 10 +++----
 4 files changed, 50 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 491845f18710..f79a457099e6 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -1998,8 +1998,7 @@ iscsi_r2tpool_alloc(struct iscsi_session *session)
 		 */
 
 		/* R2T pool */
-		if (iscsi_pool_init(&tcp_ctask->r2tpool, session->max_r2t * 4,
-				    (void***)&tcp_ctask->r2ts,
+		if (iscsi_pool_init(&tcp_ctask->r2tpool, session->max_r2t * 4, NULL,
 				    sizeof(struct iscsi_r2t_info))) {
 			goto r2t_alloc_fail;
 		}
@@ -2008,8 +2007,7 @@ iscsi_r2tpool_alloc(struct iscsi_session *session)
 		tcp_ctask->r2tqueue = kfifo_alloc(
 		      session->max_r2t * 4 * sizeof(void*), GFP_KERNEL, NULL);
 		if (tcp_ctask->r2tqueue == ERR_PTR(-ENOMEM)) {
-			iscsi_pool_free(&tcp_ctask->r2tpool,
-					(void**)tcp_ctask->r2ts);
+			iscsi_pool_free(&tcp_ctask->r2tpool);
 			goto r2t_alloc_fail;
 		}
 	}
@@ -2022,8 +2020,7 @@ r2t_alloc_fail:
 		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 
 		kfifo_free(tcp_ctask->r2tqueue);
-		iscsi_pool_free(&tcp_ctask->r2tpool,
-				(void**)tcp_ctask->r2ts);
+		iscsi_pool_free(&tcp_ctask->r2tpool);
 	}
 	return -ENOMEM;
 }
@@ -2038,8 +2035,7 @@ iscsi_r2tpool_free(struct iscsi_session *session)
 		struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 
 		kfifo_free(tcp_ctask->r2tqueue);
-		iscsi_pool_free(&tcp_ctask->r2tpool,
-				(void**)tcp_ctask->r2ts);
+		iscsi_pool_free(&tcp_ctask->r2tpool);
 	}
 }
 
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index eb3784f949fd..d49d87611e82 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -175,9 +175,8 @@ struct iscsi_tcp_cmd_task {
 	uint32_t		exp_datasn;		/* expected target's R2TSN/DataSN */
 	int			data_offset;
 	struct iscsi_r2t_info	*r2t;			/* in progress R2T    */
-	struct iscsi_queue	r2tpool;
+	struct iscsi_pool	r2tpool;
 	struct kfifo		*r2tqueue;
-	struct iscsi_r2t_info	**r2ts;
 	int			digest_count;
 	uint32_t		immdigest;		/* for imm data */
 	struct iscsi_buf	immbuf;			/* for imm data digest */
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 59365864c989..d43f909a022c 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -1413,59 +1413,64 @@ done:
 }
 EXPORT_SYMBOL_GPL(iscsi_eh_device_reset);
 
+/*
+ * Pre-allocate a pool of @max items of @item_size. By default, the pool
+ * should be accessed via kfifo_{get,put} on q->queue.
+ * Optionally, the caller can obtain the array of object pointers
+ * by passing in a non-NULL @items pointer
+ */
 int
-iscsi_pool_init(struct iscsi_queue *q, int max, void ***items, int item_size)
+iscsi_pool_init(struct iscsi_pool *q, int max, void ***items, int item_size)
 {
-	int i;
+	int i, num_arrays = 1;
 
-	*items = kmalloc(max * sizeof(void*), GFP_KERNEL);
-	if (*items == NULL)
-		return -ENOMEM;
+	memset(q, 0, sizeof(*q));
 
 	q->max = max;
-	q->pool = kmalloc(max * sizeof(void*), GFP_KERNEL);
-	if (q->pool == NULL) {
-		kfree(*items);
-		return -ENOMEM;
-	}
+
+	/* If the user passed an items pointer, he wants a copy of
+	 * the array. */
+	if (items)
+		num_arrays++;
+	q->pool = kzalloc(num_arrays * max * sizeof(void*), GFP_KERNEL);
+	if (q->pool == NULL)
+		goto enomem;
 
 	q->queue = kfifo_init((void*)q->pool, max * sizeof(void*),
 			      GFP_KERNEL, NULL);
-	if (q->queue == ERR_PTR(-ENOMEM)) {
-		kfree(q->pool);
-		kfree(*items);
-		return -ENOMEM;
-	}
+	if (q->queue == ERR_PTR(-ENOMEM))
+		goto enomem;
 
 	for (i = 0; i < max; i++) {
-		q->pool[i] = kmalloc(item_size, GFP_KERNEL);
+		q->pool[i] = kzalloc(item_size, GFP_KERNEL);
 		if (q->pool[i] == NULL) {
-			int j;
-
-			for (j = 0; j < i; j++)
-				kfree(q->pool[j]);
-
-			kfifo_free(q->queue);
-			kfree(q->pool);
-			kfree(*items);
-			return -ENOMEM;
+			q->max = i;
+			goto enomem;
 		}
-		memset(q->pool[i], 0, item_size);
-		(*items)[i] = q->pool[i];
 		__kfifo_put(q->queue, (void*)&q->pool[i], sizeof(void*));
 	}
+
+	if (items) {
+		*items = q->pool + max;
+		memcpy(*items, q->pool, max * sizeof(void *));
+	}
+
 	return 0;
+
+enomem:
+	iscsi_pool_free(q);
+	return -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(iscsi_pool_init);
 
-void iscsi_pool_free(struct iscsi_queue *q, void **items)
+void iscsi_pool_free(struct iscsi_pool *q)
 {
 	int i;
 
 	for (i = 0; i < q->max; i++)
-		kfree(items[i]);
-	kfree(q->pool);
-	kfree(items);
+		kfree(q->pool[i]);
+	if (q->pool)
+		kfree(q->pool);
 }
 EXPORT_SYMBOL_GPL(iscsi_pool_free);
 
@@ -1610,9 +1615,9 @@ module_put:
 cls_session_fail:
 	scsi_remove_host(shost);
 add_host_fail:
-	iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds);
+	iscsi_pool_free(&session->mgmtpool);
 mgmtpool_alloc_fail:
-	iscsi_pool_free(&session->cmdpool, (void**)session->cmds);
+	iscsi_pool_free(&session->cmdpool);
 cmdpool_alloc_fail:
 	scsi_host_put(shost);
 	return NULL;
@@ -1635,8 +1640,8 @@ void iscsi_session_teardown(struct iscsi_cls_session *cls_session)
 	iscsi_unblock_session(cls_session);
 	scsi_remove_host(shost);
 
-	iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds);
-	iscsi_pool_free(&session->cmdpool, (void**)session->cmds);
+	iscsi_pool_free(&session->mgmtpool);
+	iscsi_pool_free(&session->cmdpool);
 
 	kfree(session->password);
 	kfree(session->password_in);
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index a9a9e869188d..4b3e3c15121a 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -215,7 +215,7 @@ struct iscsi_conn {
 	uint32_t		eh_abort_cnt;
 };
 
-struct iscsi_queue {
+struct iscsi_pool {
 	struct kfifo		*queue;		/* FIFO Queue */
 	void			**pool;		/* Pool of elements */
 	int			max;		/* Max number of elements */
@@ -274,10 +274,10 @@ struct iscsi_session {
 
 	int			cmds_max;	/* size of cmds array */
 	struct iscsi_cmd_task	**cmds;		/* Original Cmds arr */
-	struct iscsi_queue	cmdpool;	/* PDU's pool */
+	struct iscsi_pool	cmdpool;	/* PDU's pool */
 	int			mgmtpool_max;	/* size of mgmt array */
 	struct iscsi_mgmt_task	**mgmt_cmds;	/* Original mgmt arr */
-	struct iscsi_queue	mgmtpool;	/* Mgmt PDU's pool */
+	struct iscsi_pool	mgmtpool;	/* Mgmt PDU's pool */
 };
 
 /*
@@ -350,8 +350,8 @@ extern void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask);
 /*
  * generic helpers
  */
-extern void iscsi_pool_free(struct iscsi_queue *, void **);
-extern int iscsi_pool_init(struct iscsi_queue *, int, void ***, int);
+extern void iscsi_pool_free(struct iscsi_pool *);
+extern int iscsi_pool_init(struct iscsi_pool *, int, void ***, int);
 
 /*
  * inline functions to deal with padding.
-- 
cgit v1.2.3


From b3a7ea8d50f6028964b468d13a095dfb2508b2fb Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 13 Dec 2007 12:43:26 -0600
Subject: [SCSI] libiscsi: do not block session during logout

There is not need to block the session during logout. Since
we are going to fail the commands that were blocked just fail them
immediately instead.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/infiniband/ulp/iser/iser_initiator.c |   4 +-
 drivers/scsi/iscsi_tcp.c                     |   4 +-
 drivers/scsi/libiscsi.c                      | 153 +++++++++++++++------------
 include/scsi/libiscsi.h                      |   2 +
 include/scsi/scsi_transport_iscsi.h          |   1 +
 5 files changed, 88 insertions(+), 76 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c
index a6f2303ed14a..47f716ca0026 100644
--- a/drivers/infiniband/ulp/iser/iser_initiator.c
+++ b/drivers/infiniband/ulp/iser/iser_initiator.c
@@ -621,9 +621,7 @@ void iser_snd_completion(struct iser_desc *tx_desc)
 			struct iscsi_session *session = conn->session;
 
 			spin_lock(&conn->session->lock);
-			list_del(&mtask->running);
-			__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
-				    sizeof(void*));
+			iscsi_free_mgmt_task(conn, mtask);
 			spin_unlock(&session->lock);
 		}
 	}
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index f79a457099e6..90eae8e0d978 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -1349,9 +1349,7 @@ iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 		struct iscsi_session *session = conn->session;
 
 		spin_lock_bh(&session->lock);
-		list_del(&conn->mtask->running);
-		__kfifo_put(session->mgmtpool.queue, (void*)&conn->mtask,
-			    sizeof(void*));
+		iscsi_free_mgmt_task(conn, mtask);
 		spin_unlock_bh(&session->lock);
 	}
 	return 0;
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index d43f909a022c..b7a2b9ad3a97 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -37,9 +37,6 @@
 #include <scsi/scsi_transport_iscsi.h>
 #include <scsi/libiscsi.h>
 
-static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
-			 int err);
-
 struct iscsi_session *
 class_to_transport_session(struct iscsi_cls_session *cls_session)
 {
@@ -274,6 +271,53 @@ static void __iscsi_put_ctask(struct iscsi_cmd_task *ctask)
 		iscsi_complete_command(ctask);
 }
 
+/*
+ * session lock must be held
+ */
+static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
+			 int err)
+{
+	struct scsi_cmnd *sc;
+
+	sc = ctask->sc;
+	if (!sc)
+		return;
+
+	if (ctask->state == ISCSI_TASK_PENDING)
+		/*
+		 * cmd never made it to the xmit thread, so we should not count
+		 * the cmd in the sequencing
+		 */
+		conn->session->queued_cmdsn--;
+	else
+		conn->session->tt->cleanup_cmd_task(conn, ctask);
+
+	sc->result = err;
+	scsi_set_resid(sc, scsi_bufflen(sc));
+	if (conn->ctask == ctask)
+		conn->ctask = NULL;
+	/* release ref from queuecommand */
+	__iscsi_put_ctask(ctask);
+}
+
+/**
+ * iscsi_free_mgmt_task - return mgmt task back to pool
+ * @conn: iscsi connection
+ * @mtask: mtask
+ *
+ * Must be called with session lock.
+ */
+void iscsi_free_mgmt_task(struct iscsi_conn *conn,
+			  struct iscsi_mgmt_task *mtask)
+{
+	list_del_init(&mtask->running);
+	if (conn->login_mtask == mtask)
+		return;
+	__kfifo_put(conn->session->mgmtpool.queue,
+		    (void*)&mtask, sizeof(void*));
+}
+EXPORT_SYMBOL_GPL(iscsi_free_mgmt_task);
+
 /**
  * iscsi_cmd_rsp - SCSI Command Response processing
  * @conn: iscsi connection
@@ -464,10 +508,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			 */
 			if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen))
 				rc = ISCSI_ERR_CONN_FAILED;
-			list_del_init(&mtask->running);
-			if (conn->login_mtask != mtask)
-				__kfifo_put(session->mgmtpool.queue,
-					    (void*)&mtask, sizeof(void*));
+			iscsi_free_mgmt_task(conn, mtask);
 			break;
 		case ISCSI_OP_SCSI_TMFUNC_RSP:
 			if (datalen) {
@@ -476,6 +517,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			}
 
 			iscsi_tmf_rsp(conn, hdr);
+			iscsi_free_mgmt_task(conn, mtask);
 			break;
 		case ISCSI_OP_NOOP_IN:
 			if (hdr->ttt != cpu_to_be32(ISCSI_RESERVED_TAG) || datalen) {
@@ -486,9 +528,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 
 			if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen))
 				rc = ISCSI_ERR_CONN_FAILED;
-			list_del_init(&mtask->running);
-			__kfifo_put(session->mgmtpool.queue,
-				    (void*)&mtask, sizeof(void*));
+			iscsi_free_mgmt_task(conn, mtask);
 			break;
 		default:
 			rc = ISCSI_ERR_BAD_OPCODE;
@@ -650,14 +690,12 @@ static void iscsi_prep_mtask(struct iscsi_conn *conn,
 static int iscsi_xmit_mtask(struct iscsi_conn *conn)
 {
 	struct iscsi_hdr *hdr = conn->mtask->hdr;
-	int rc, was_logout = 0;
+	int rc;
 
+	if ((hdr->opcode & ISCSI_OPCODE_MASK) == ISCSI_OP_LOGOUT)
+		conn->session->state = ISCSI_STATE_LOGGING_OUT;
 	spin_unlock_bh(&conn->session->lock);
-	if ((hdr->opcode & ISCSI_OPCODE_MASK) == ISCSI_OP_LOGOUT) {
-		conn->session->state = ISCSI_STATE_IN_RECOVERY;
-		iscsi_block_session(session_to_cls(conn->session));
-		was_logout = 1;
-	}
+
 	rc = conn->session->tt->xmit_mgmt_task(conn, conn->mtask);
 	spin_lock_bh(&conn->session->lock);
 	if (rc)
@@ -665,11 +703,6 @@ static int iscsi_xmit_mtask(struct iscsi_conn *conn)
 
 	/* done with this in-progress mtask */
 	conn->mtask = NULL;
-
-	if (was_logout) {
-		set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx);
-		return -ENODATA;
-	}
 	return 0;
 }
 
@@ -763,6 +796,12 @@ check_mgmt:
 	while (!list_empty(&conn->mgmtqueue)) {
 		conn->mtask = list_entry(conn->mgmtqueue.next,
 					 struct iscsi_mgmt_task, running);
+		if (conn->session->state == ISCSI_STATE_LOGGING_OUT) {
+			iscsi_free_mgmt_task(conn, conn->mtask);
+			conn->mtask = NULL;
+			continue;
+		}
+
 		iscsi_prep_mtask(conn, conn->mtask);
 		list_move_tail(conn->mgmtqueue.next, &conn->mgmt_run_list);
 		rc = iscsi_xmit_mtask(conn);
@@ -777,6 +816,10 @@ check_mgmt:
 
 		conn->ctask = list_entry(conn->xmitqueue.next,
 					 struct iscsi_cmd_task, running);
+		if (conn->session->state == ISCSI_STATE_LOGGING_OUT) {
+			fail_command(conn, conn->ctask, DID_NO_CONNECT << 16);
+			continue;
+		}
 		if (iscsi_prep_scsi_cmd_pdu(conn->ctask)) {
 			fail_command(conn, conn->ctask, DID_ABORT << 16);
 			continue;
@@ -800,6 +843,12 @@ check_mgmt:
 		if (conn->session->fast_abort && conn->tmf_state != TMF_INITIAL)
 			break;
 
+		/*
+		 * we always do fastlogout - conn stop code will clean up.
+		 */
+		if (conn->session->state == ISCSI_STATE_LOGGING_OUT)
+			break;
+
 		conn->ctask = list_entry(conn->requeue.next,
 					 struct iscsi_cmd_task, running);
 		conn->ctask->state = ISCSI_TASK_RUNNING;
@@ -842,6 +891,7 @@ enum {
 	FAILURE_SESSION_TERMINATE,
 	FAILURE_SESSION_IN_RECOVERY,
 	FAILURE_SESSION_RECOVERY_TIMEOUT,
+	FAILURE_SESSION_LOGGING_OUT,
 };
 
 int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
@@ -879,12 +929,19 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
 			goto reject;
 		}
 
-		if (session->state == ISCSI_STATE_RECOVERY_FAILED)
+		switch (session->state) {
+		case ISCSI_STATE_RECOVERY_FAILED:
 			reason = FAILURE_SESSION_RECOVERY_TIMEOUT;
-		else if (session->state == ISCSI_STATE_TERMINATE)
+			break;
+		case ISCSI_STATE_TERMINATE:
 			reason = FAILURE_SESSION_TERMINATE;
-		else
+			break;
+		case ISCSI_STATE_LOGGING_OUT:
+			reason = FAILURE_SESSION_LOGGING_OUT;
+			break;
+		default:
 			reason = FAILURE_SESSION_FREED;
+		}
 		goto fault;
 	}
 
@@ -1120,44 +1177,9 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
 	if (age != session->age ||
 	    session->state != ISCSI_STATE_LOGGED_IN)
 		return -ENOTCONN;
-
-	if (!list_empty(&mtask->running)) {
-		list_del_init(&mtask->running);
-		__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
-			    sizeof(void*));
-	}
 	return 0;
 }
 
-/*
- * session lock must be held
- */
-static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
-			 int err)
-{
-	struct scsi_cmnd *sc;
-
-	sc = ctask->sc;
-	if (!sc)
-		return;
-
-	if (ctask->state == ISCSI_TASK_PENDING)
-		/*
-		 * cmd never made it to the xmit thread, so we should not count
-		 * the cmd in the sequencing
-		 */
-		conn->session->queued_cmdsn--;
-	else
-		conn->session->tt->cleanup_cmd_task(conn, ctask);
-
-	sc->result = err;
-	scsi_set_resid(sc, scsi_bufflen(sc));
-	if (conn->ctask == ctask)
-		conn->ctask = NULL;
-	/* release ref from queuecommand */
-	__iscsi_put_ctask(ctask);
-}
-
 /*
  * Fail commands. session lock held and recv side suspended and xmit
  * thread flushed
@@ -1837,22 +1859,13 @@ flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn)
 	/* handle pending */
 	list_for_each_entry_safe(mtask, tmp, &conn->mgmtqueue, running) {
 		debug_scsi("flushing pending mgmt task itt 0x%x\n", mtask->itt);
-		list_del_init(&mtask->running);
-		if (mtask == conn->login_mtask)
-			continue;
-		__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
-			    sizeof(void*));
+		iscsi_free_mgmt_task(conn, mtask);
 	}
 
 	/* handle running */
 	list_for_each_entry_safe(mtask, tmp, &conn->mgmt_run_list, running) {
 		debug_scsi("flushing running mgmt task itt 0x%x\n", mtask->itt);
-		list_del_init(&mtask->running);
-
-		if (mtask == conn->login_mtask)
-			continue;
-		__kfifo_put(session->mgmtpool.queue, (void*)&mtask,
-			   sizeof(void*));
+		iscsi_free_mgmt_task(conn, mtask);
 	}
 
 	conn->mtask = NULL;
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index 4b3e3c15121a..d68f74523f2e 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -346,6 +346,8 @@ extern int __iscsi_complete_pdu(struct iscsi_conn *, struct iscsi_hdr *,
 extern int iscsi_verify_itt(struct iscsi_conn *, struct iscsi_hdr *,
 			    uint32_t *);
 extern void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask);
+extern void iscsi_free_mgmt_task(struct iscsi_conn *conn,
+				 struct iscsi_mgmt_task *mtask);
 
 /*
  * generic helpers
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index 7ff6199cbd55..b8d97bd20f6e 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -176,6 +176,7 @@ struct iscsi_cls_conn {
 #define ISCSI_STATE_TERMINATE		4
 #define ISCSI_STATE_IN_RECOVERY		5
 #define ISCSI_STATE_RECOVERY_FAILED	6
+#define ISCSI_STATE_LOGGING_OUT		7
 
 struct iscsi_cls_session {
 	struct list_head sess_list;		/* item in session_list */
-- 
cgit v1.2.3


From 2697478903d0ad8bdbf868b1346ae4b891662bb1 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 13 Dec 2007 12:43:29 -0600
Subject: [SCSI] libiscsi: fix shutdown

We were using the device delete sysfs file to remove each device
then logout. Now in 2.6.21 this will not work because
the sysfs delete file returns immediately and does not wait for
the device removal to complete. This causes a hang if a cache sync
is needed during shutdown. Before .21, that approach had other
problems, so this patch fixes the shutdown code so that we remove the target
and unbind the session before logging out and shut down the session

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libiscsi.c             |   4 +-
 drivers/scsi/qla4xxx/ql4_init.c     |   4 +-
 drivers/scsi/qla4xxx/ql4_os.c       |   7 +-
 drivers/scsi/scsi_transport_iscsi.c | 289 +++++++++++++++++++-----------------
 include/scsi/iscsi_if.h             |   7 +
 include/scsi/iscsi_proto.h          |   2 +
 include/scsi/scsi_transport_iscsi.h |   7 +-
 7 files changed, 176 insertions(+), 144 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 441e351b4456..5205ef2c29b2 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -1662,7 +1662,7 @@ void iscsi_session_teardown(struct iscsi_cls_session *cls_session)
 	struct iscsi_session *session = iscsi_hostdata(shost->hostdata);
 	struct module *owner = cls_session->transport->owner;
 
-	iscsi_unblock_session(cls_session);
+	iscsi_remove_session(cls_session);
 	scsi_remove_host(shost);
 
 	iscsi_pool_free(&session->mgmtpool);
@@ -1677,7 +1677,7 @@ void iscsi_session_teardown(struct iscsi_cls_session *cls_session)
 	kfree(session->hwaddress);
 	kfree(session->initiatorname);
 
-	iscsi_destroy_session(cls_session);
+	iscsi_free_session(cls_session);
 	scsi_host_put(shost);
 	module_put(owner);
 }
diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c
index d692c713416a..cbe0a17ced5f 100644
--- a/drivers/scsi/qla4xxx/ql4_init.c
+++ b/drivers/scsi/qla4xxx/ql4_init.c
@@ -5,6 +5,7 @@
  * See LICENSE.qla4xxx for copyright and licensing details.
  */
 
+#include <scsi/iscsi_if.h>
 #include "ql4_def.h"
 #include "ql4_glbl.h"
 #include "ql4_dbg.h"
@@ -1305,7 +1306,8 @@ int qla4xxx_process_ddb_changed(struct scsi_qla_host *ha,
 		atomic_set(&ddb_entry->relogin_timer, 0);
 		clear_bit(DF_RELOGIN, &ddb_entry->flags);
 		clear_bit(DF_NO_RELOGIN, &ddb_entry->flags);
-		iscsi_if_create_session_done(ddb_entry->conn);
+		iscsi_session_event(ddb_entry->sess,
+				    ISCSI_KEVENT_CREATE_SESSION);
 		/*
 		 * Change the lun state to READY in case the lun TIMEOUT before
 		 * the device came back.
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index 89460d27c689..f55b9f7d9396 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -298,8 +298,7 @@ void qla4xxx_destroy_sess(struct ddb_entry *ddb_entry)
 		return;
 
 	if (ddb_entry->conn) {
-		iscsi_if_destroy_session_done(ddb_entry->conn);
-		iscsi_destroy_conn(ddb_entry->conn);
+		atomic_set(&ddb_entry->state, DDB_STATE_DEAD);
 		iscsi_remove_session(ddb_entry->sess);
 	}
 	iscsi_free_session(ddb_entry->sess);
@@ -309,6 +308,7 @@ int qla4xxx_add_sess(struct ddb_entry *ddb_entry)
 {
 	int err;
 
+	ddb_entry->sess->recovery_tmo = ddb_entry->ha->port_down_retry_count;
 	err = iscsi_add_session(ddb_entry->sess, ddb_entry->fw_ddb_index);
 	if (err) {
 		DEBUG2(printk(KERN_ERR "Could not add session.\n"));
@@ -321,9 +321,6 @@ int qla4xxx_add_sess(struct ddb_entry *ddb_entry)
 		DEBUG2(printk(KERN_ERR "Could not add connection.\n"));
 		return -ENOMEM;
 	}
-
-	ddb_entry->sess->recovery_tmo = ddb_entry->ha->port_down_retry_count;
-	iscsi_if_create_session_done(ddb_entry->conn);
 	return 0;
 }
 
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 9cc2cc8e87b3..b82139dc4830 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -116,6 +116,8 @@ static struct attribute_group iscsi_transport_group = {
 	.attrs = iscsi_transport_attrs,
 };
 
+
+
 static int iscsi_setup_host(struct transport_container *tc, struct device *dev,
 			    struct class_device *cdev)
 {
@@ -125,13 +127,30 @@ static int iscsi_setup_host(struct transport_container *tc, struct device *dev,
 	memset(ihost, 0, sizeof(*ihost));
 	INIT_LIST_HEAD(&ihost->sessions);
 	mutex_init(&ihost->mutex);
+
+	snprintf(ihost->unbind_workq_name, KOBJ_NAME_LEN, "iscsi_unbind_%d",
+		shost->host_no);
+	ihost->unbind_workq = create_singlethread_workqueue(
+						ihost->unbind_workq_name);
+	if (!ihost->unbind_workq)
+		return -ENOMEM;
+	return 0;
+}
+
+static int iscsi_remove_host(struct transport_container *tc, struct device *dev,
+			     struct class_device *cdev)
+{
+	struct Scsi_Host *shost = dev_to_shost(dev);
+	struct iscsi_host *ihost = shost->shost_data;
+
+	destroy_workqueue(ihost->unbind_workq);
 	return 0;
 }
 
 static DECLARE_TRANSPORT_CLASS(iscsi_host_class,
 			       "iscsi_host",
 			       iscsi_setup_host,
-			       NULL,
+			       iscsi_remove_host,
 			       NULL);
 
 static DECLARE_TRANSPORT_CLASS(iscsi_session_class,
@@ -266,6 +285,35 @@ void iscsi_block_session(struct iscsi_cls_session *session)
 }
 EXPORT_SYMBOL_GPL(iscsi_block_session);
 
+static void __iscsi_unbind_session(struct work_struct *work)
+{
+	struct iscsi_cls_session *session =
+			container_of(work, struct iscsi_cls_session,
+				     unbind_work);
+	struct Scsi_Host *shost = iscsi_session_to_shost(session);
+	struct iscsi_host *ihost = shost->shost_data;
+
+	/* Prevent new scans and make sure scanning is not in progress */
+	mutex_lock(&ihost->mutex);
+	if (list_empty(&session->host_list)) {
+		mutex_unlock(&ihost->mutex);
+		return;
+	}
+	list_del_init(&session->host_list);
+	mutex_unlock(&ihost->mutex);
+
+	scsi_remove_target(&session->dev);
+	iscsi_session_event(session, ISCSI_KEVENT_UNBIND_SESSION);
+}
+
+static int iscsi_unbind_session(struct iscsi_cls_session *session)
+{
+	struct Scsi_Host *shost = iscsi_session_to_shost(session);
+	struct iscsi_host *ihost = shost->shost_data;
+
+	return queue_work(ihost->unbind_workq, &session->unbind_work);
+}
+
 struct iscsi_cls_session *
 iscsi_alloc_session(struct Scsi_Host *shost,
 		    struct iscsi_transport *transport)
@@ -282,6 +330,7 @@ iscsi_alloc_session(struct Scsi_Host *shost,
 	INIT_DELAYED_WORK(&session->recovery_work, session_recovery_timedout);
 	INIT_LIST_HEAD(&session->host_list);
 	INIT_LIST_HEAD(&session->sess_list);
+	INIT_WORK(&session->unbind_work, __iscsi_unbind_session);
 
 	/* this is released in the dev's release function */
 	scsi_host_get(shost);
@@ -298,6 +347,7 @@ int iscsi_add_session(struct iscsi_cls_session *session, unsigned int target_id)
 {
 	struct Scsi_Host *shost = iscsi_session_to_shost(session);
 	struct iscsi_host *ihost;
+	unsigned long flags;
 	int err;
 
 	ihost = shost->shost_data;
@@ -314,9 +364,15 @@ int iscsi_add_session(struct iscsi_cls_session *session, unsigned int target_id)
 	}
 	transport_register_device(&session->dev);
 
+	spin_lock_irqsave(&sesslock, flags);
+	list_add(&session->sess_list, &sesslist);
+	spin_unlock_irqrestore(&sesslock, flags);
+
 	mutex_lock(&ihost->mutex);
 	list_add(&session->host_list, &ihost->sessions);
 	mutex_unlock(&ihost->mutex);
+
+	iscsi_session_event(session, ISCSI_KEVENT_CREATE_SESSION);
 	return 0;
 
 release_host:
@@ -352,19 +408,58 @@ iscsi_create_session(struct Scsi_Host *shost,
 }
 EXPORT_SYMBOL_GPL(iscsi_create_session);
 
+static void iscsi_conn_release(struct device *dev)
+{
+	struct iscsi_cls_conn *conn = iscsi_dev_to_conn(dev);
+	struct device *parent = conn->dev.parent;
+
+	kfree(conn);
+	put_device(parent);
+}
+
+static int iscsi_is_conn_dev(const struct device *dev)
+{
+	return dev->release == iscsi_conn_release;
+}
+
+static int iscsi_iter_destroy_conn_fn(struct device *dev, void *data)
+{
+	if (!iscsi_is_conn_dev(dev))
+		return 0;
+	return iscsi_destroy_conn(iscsi_dev_to_conn(dev));
+}
+
 void iscsi_remove_session(struct iscsi_cls_session *session)
 {
 	struct Scsi_Host *shost = iscsi_session_to_shost(session);
 	struct iscsi_host *ihost = shost->shost_data;
+	unsigned long flags;
+	int err;
+
+	spin_lock_irqsave(&sesslock, flags);
+	list_del(&session->sess_list);
+	spin_unlock_irqrestore(&sesslock, flags);
 
+	/*
+	 * If we are blocked let commands flow again. The lld or iscsi
+	 * layer should set up the queuecommand to fail commands.
+	 */
+	iscsi_unblock_session(session);
+	iscsi_unbind_session(session);
+	/*
+	 * If the session dropped while removing devices then we need to make
+	 * sure it is not blocked
+	 */
 	if (!cancel_delayed_work(&session->recovery_work))
 		flush_workqueue(iscsi_eh_timer_workq);
+	flush_workqueue(ihost->unbind_workq);
 
-	mutex_lock(&ihost->mutex);
-	list_del(&session->host_list);
-	mutex_unlock(&ihost->mutex);
-
-	scsi_remove_target(&session->dev);
+	/* hw iscsi may not have removed all connections from session */
+	err = device_for_each_child(&session->dev, NULL,
+				    iscsi_iter_destroy_conn_fn);
+	if (err)
+		dev_printk(KERN_ERR, &session->dev, "iscsi: Could not delete "
+			   "all connections for session. Error %d.\n", err);
 
 	transport_unregister_device(&session->dev);
 	device_del(&session->dev);
@@ -373,9 +468,9 @@ EXPORT_SYMBOL_GPL(iscsi_remove_session);
 
 void iscsi_free_session(struct iscsi_cls_session *session)
 {
+	iscsi_session_event(session, ISCSI_KEVENT_DESTROY_SESSION);
 	put_device(&session->dev);
 }
-
 EXPORT_SYMBOL_GPL(iscsi_free_session);
 
 /**
@@ -393,20 +488,6 @@ int iscsi_destroy_session(struct iscsi_cls_session *session)
 }
 EXPORT_SYMBOL_GPL(iscsi_destroy_session);
 
-static void iscsi_conn_release(struct device *dev)
-{
-	struct iscsi_cls_conn *conn = iscsi_dev_to_conn(dev);
-	struct device *parent = conn->dev.parent;
-
-	kfree(conn);
-	put_device(parent);
-}
-
-static int iscsi_is_conn_dev(const struct device *dev)
-{
-	return dev->release == iscsi_conn_release;
-}
-
 /**
  * iscsi_create_conn - create iscsi class connection
  * @session: iscsi cls session
@@ -426,6 +507,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid)
 {
 	struct iscsi_transport *transport = session->transport;
 	struct iscsi_cls_conn *conn;
+	unsigned long flags;
 	int err;
 
 	conn = kzalloc(sizeof(*conn) + transport->conndata_size, GFP_KERNEL);
@@ -454,6 +536,11 @@ iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid)
 		goto release_parent_ref;
 	}
 	transport_register_device(&conn->dev);
+
+	spin_lock_irqsave(&connlock, flags);
+	list_add(&conn->conn_list, &connlist);
+	conn->active = 1;
+	spin_unlock_irqrestore(&connlock, flags);
 	return conn;
 
 release_parent_ref:
@@ -469,15 +556,21 @@ EXPORT_SYMBOL_GPL(iscsi_create_conn);
  * iscsi_destroy_conn - destroy iscsi class connection
  * @conn: iscsi cls session
  *
- * This can be called from an LLD or iscsi_transport.
+ * This can be called from a LLD or iscsi_transport.
  */
 int iscsi_destroy_conn(struct iscsi_cls_conn *conn)
 {
+	unsigned long flags;
+
+	spin_lock_irqsave(&connlock, flags);
+	conn->active = 0;
+	list_del(&conn->conn_list);
+	spin_unlock_irqrestore(&connlock, flags);
+
 	transport_unregister_device(&conn->dev);
 	device_unregister(&conn->dev);
 	return 0;
 }
-
 EXPORT_SYMBOL_GPL(iscsi_destroy_conn);
 
 /*
@@ -687,132 +780,74 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh)
 }
 
 /**
- * iscsi_if_destroy_session_done - send session destr. completion event
- * @conn: last connection for session
- *
- * This is called by HW iscsi LLDs to notify userpsace that its HW has
- * removed a session.
+ * iscsi_session_event - send session destr. completion event
+ * @session: iscsi class session
+ * @event: type of event
  */
-int iscsi_if_destroy_session_done(struct iscsi_cls_conn *conn)
+int iscsi_session_event(struct iscsi_cls_session *session,
+			enum iscsi_uevent_e event)
 {
 	struct iscsi_internal *priv;
-	struct iscsi_cls_session *session;
 	struct Scsi_Host *shost;
 	struct iscsi_uevent *ev;
 	struct sk_buff  *skb;
 	struct nlmsghdr *nlh;
-	unsigned long flags;
 	int rc, len = NLMSG_SPACE(sizeof(*ev));
 
-	priv = iscsi_if_transport_lookup(conn->transport);
+	priv = iscsi_if_transport_lookup(session->transport);
 	if (!priv)
 		return -EINVAL;
-
-	session = iscsi_dev_to_session(conn->dev.parent);
 	shost = iscsi_session_to_shost(session);
 
 	skb = alloc_skb(len, GFP_KERNEL);
 	if (!skb) {
-		dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of "
-			  "session creation event\n");
+		dev_printk(KERN_ERR, &session->dev, "Cannot notify userspace "
+			  "of session event %u\n", event);
 		return -ENOMEM;
 	}
 
 	nlh = __nlmsg_put(skb, priv->daemon_pid, 0, 0, (len - sizeof(*nlh)), 0);
 	ev = NLMSG_DATA(nlh);
-	ev->transport_handle = iscsi_handle(conn->transport);
-	ev->type = ISCSI_KEVENT_DESTROY_SESSION;
-	ev->r.d_session.host_no = shost->host_no;
-	ev->r.d_session.sid = session->sid;
-
-	/*
-	 * this will occur if the daemon is not up, so we just warn
-	 * the user and when the daemon is restarted it will handle it
-	 */
-	rc = iscsi_broadcast_skb(skb, GFP_KERNEL);
-	if (rc < 0)
-		dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of "
-			  "session destruction event. Check iscsi daemon\n");
-
-	spin_lock_irqsave(&sesslock, flags);
-	list_del(&session->sess_list);
-	spin_unlock_irqrestore(&sesslock, flags);
-
-	spin_lock_irqsave(&connlock, flags);
-	conn->active = 0;
-	list_del(&conn->conn_list);
-	spin_unlock_irqrestore(&connlock, flags);
-
-	return rc;
-}
-EXPORT_SYMBOL_GPL(iscsi_if_destroy_session_done);
-
-/**
- * iscsi_if_create_session_done - send session creation completion event
- * @conn: leading connection for session
- *
- * This is called by HW iscsi LLDs to notify userpsace that its HW has
- * created a session or a existing session is back in the logged in state.
- */
-int iscsi_if_create_session_done(struct iscsi_cls_conn *conn)
-{
-	struct iscsi_internal *priv;
-	struct iscsi_cls_session *session;
-	struct Scsi_Host *shost;
-	struct iscsi_uevent *ev;
-	struct sk_buff  *skb;
-	struct nlmsghdr *nlh;
-	unsigned long flags;
-	int rc, len = NLMSG_SPACE(sizeof(*ev));
+	ev->transport_handle = iscsi_handle(session->transport);
 
-	priv = iscsi_if_transport_lookup(conn->transport);
-	if (!priv)
+	ev->type = event;
+	switch (event) {
+	case ISCSI_KEVENT_DESTROY_SESSION:
+		ev->r.d_session.host_no = shost->host_no;
+		ev->r.d_session.sid = session->sid;
+		break;
+	case ISCSI_KEVENT_CREATE_SESSION:
+		ev->r.c_session_ret.host_no = shost->host_no;
+		ev->r.c_session_ret.sid = session->sid;
+		break;
+	case ISCSI_KEVENT_UNBIND_SESSION:
+		ev->r.unbind_session.host_no = shost->host_no;
+		ev->r.unbind_session.sid = session->sid;
+		break;
+	default:
+		dev_printk(KERN_ERR, &session->dev, "Invalid event %u.\n",
+			   event);
+		kfree_skb(skb);
 		return -EINVAL;
-
-	session = iscsi_dev_to_session(conn->dev.parent);
-	shost = iscsi_session_to_shost(session);
-
-	skb = alloc_skb(len, GFP_KERNEL);
-	if (!skb) {
-		dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of "
-			  "session creation event\n");
-		return -ENOMEM;
 	}
 
-	nlh = __nlmsg_put(skb, priv->daemon_pid, 0, 0, (len - sizeof(*nlh)), 0);
-	ev = NLMSG_DATA(nlh);
-	ev->transport_handle = iscsi_handle(conn->transport);
-	ev->type = ISCSI_UEVENT_CREATE_SESSION;
-	ev->r.c_session_ret.host_no = shost->host_no;
-	ev->r.c_session_ret.sid = session->sid;
-
 	/*
 	 * this will occur if the daemon is not up, so we just warn
 	 * the user and when the daemon is restarted it will handle it
 	 */
 	rc = iscsi_broadcast_skb(skb, GFP_KERNEL);
 	if (rc < 0)
-		dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of "
-			  "session creation event. Check iscsi daemon\n");
-
-	spin_lock_irqsave(&sesslock, flags);
-	list_add(&session->sess_list, &sesslist);
-	spin_unlock_irqrestore(&sesslock, flags);
-
-	spin_lock_irqsave(&connlock, flags);
-	list_add(&conn->conn_list, &connlist);
-	conn->active = 1;
-	spin_unlock_irqrestore(&connlock, flags);
+		dev_printk(KERN_ERR, &session->dev, "Cannot notify userspace "
+			  "of session event %u. Check iscsi daemon\n", event);
 	return rc;
 }
-EXPORT_SYMBOL_GPL(iscsi_if_create_session_done);
+EXPORT_SYMBOL_GPL(iscsi_session_event);
 
 static int
 iscsi_if_create_session(struct iscsi_internal *priv, struct iscsi_uevent *ev)
 {
 	struct iscsi_transport *transport = priv->iscsi_transport;
 	struct iscsi_cls_session *session;
-	unsigned long flags;
 	uint32_t hostno;
 
 	session = transport->create_session(transport, &priv->t,
@@ -823,10 +858,6 @@ iscsi_if_create_session(struct iscsi_internal *priv, struct iscsi_uevent *ev)
 	if (!session)
 		return -ENOMEM;
 
-	spin_lock_irqsave(&sesslock, flags);
-	list_add(&session->sess_list, &sesslist);
-	spin_unlock_irqrestore(&sesslock, flags);
-
 	ev->r.c_session_ret.host_no = hostno;
 	ev->r.c_session_ret.sid = session->sid;
 	return 0;
@@ -837,7 +868,6 @@ iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev)
 {
 	struct iscsi_cls_conn *conn;
 	struct iscsi_cls_session *session;
-	unsigned long flags;
 
 	session = iscsi_session_lookup(ev->u.c_conn.sid);
 	if (!session) {
@@ -856,28 +886,17 @@ iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev)
 
 	ev->r.c_conn_ret.sid = session->sid;
 	ev->r.c_conn_ret.cid = conn->cid;
-
-	spin_lock_irqsave(&connlock, flags);
-	list_add(&conn->conn_list, &connlist);
-	conn->active = 1;
-	spin_unlock_irqrestore(&connlock, flags);
-
 	return 0;
 }
 
 static int
 iscsi_if_destroy_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev)
 {
-	unsigned long flags;
 	struct iscsi_cls_conn *conn;
 
 	conn = iscsi_conn_lookup(ev->u.d_conn.sid, ev->u.d_conn.cid);
 	if (!conn)
 		return -EINVAL;
-	spin_lock_irqsave(&connlock, flags);
-	conn->active = 0;
-	list_del(&conn->conn_list);
-	spin_unlock_irqrestore(&connlock, flags);
 
 	if (transport->destroy_conn)
 		transport->destroy_conn(conn);
@@ -1004,7 +1023,6 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	struct iscsi_internal *priv;
 	struct iscsi_cls_session *session;
 	struct iscsi_cls_conn *conn;
-	unsigned long flags;
 
 	priv = iscsi_if_transport_lookup(iscsi_ptr(ev->transport_handle));
 	if (!priv)
@@ -1022,13 +1040,16 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		break;
 	case ISCSI_UEVENT_DESTROY_SESSION:
 		session = iscsi_session_lookup(ev->u.d_session.sid);
-		if (session) {
-			spin_lock_irqsave(&sesslock, flags);
-			list_del(&session->sess_list);
-			spin_unlock_irqrestore(&sesslock, flags);
-
+		if (session)
 			transport->destroy_session(session);
-		} else
+		else
+			err = -EINVAL;
+		break;
+	case ISCSI_UEVENT_UNBIND_SESSION:
+		session = iscsi_session_lookup(ev->u.d_session.sid);
+		if (session)
+			iscsi_unbind_session(session);
+		else
 			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_CREATE_CONN:
diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h
index bff0b1f7857b..8a4426df6c3a 100644
--- a/include/scsi/iscsi_if.h
+++ b/include/scsi/iscsi_if.h
@@ -49,12 +49,15 @@ enum iscsi_uevent_e {
 
 	ISCSI_UEVENT_TGT_DSCVR		= UEVENT_BASE + 15,
 	ISCSI_UEVENT_SET_HOST_PARAM	= UEVENT_BASE + 16,
+	ISCSI_UEVENT_UNBIND_SESSION	= UEVENT_BASE + 17,
 
 	/* up events */
 	ISCSI_KEVENT_RECV_PDU		= KEVENT_BASE + 1,
 	ISCSI_KEVENT_CONN_ERROR		= KEVENT_BASE + 2,
 	ISCSI_KEVENT_IF_ERROR		= KEVENT_BASE + 3,
 	ISCSI_KEVENT_DESTROY_SESSION	= KEVENT_BASE + 4,
+	ISCSI_KEVENT_UNBIND_SESSION	= KEVENT_BASE + 5,
+	ISCSI_KEVENT_CREATE_SESSION	= KEVENT_BASE + 6,
 };
 
 enum iscsi_tgt_dscvr {
@@ -156,6 +159,10 @@ struct iscsi_uevent {
 			uint32_t	sid;
 			uint32_t	cid;
 		} c_conn_ret;
+		struct msg_unbind_session {
+			uint32_t	sid;
+			uint32_t	host_no;
+		} unbind_session;
 		struct msg_recv_req {
 			uint32_t	sid;
 			uint32_t	cid;
diff --git a/include/scsi/iscsi_proto.h b/include/scsi/iscsi_proto.h
index 6947082eee6d..318a909e7ae1 100644
--- a/include/scsi/iscsi_proto.h
+++ b/include/scsi/iscsi_proto.h
@@ -21,6 +21,8 @@
 #ifndef ISCSI_PROTO_H
 #define ISCSI_PROTO_H
 
+#include <linux/types.h>
+
 #define ISCSI_DRAFT20_VERSION	0x00
 
 /* default iSCSI listen port for incoming connections */
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index b8d97bd20f6e..093b4036f8db 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -186,6 +186,7 @@ struct iscsi_cls_session {
 	/* recovery fields */
 	int recovery_tmo;
 	struct delayed_work recovery_work;
+	struct work_struct unbind_work;
 
 	int target_id;
 
@@ -206,6 +207,8 @@ struct iscsi_cls_session {
 struct iscsi_host {
 	struct list_head sessions;
 	struct mutex mutex;
+	struct workqueue_struct *unbind_workq;
+	char unbind_workq_name[KOBJ_NAME_LEN];
 };
 
 /*
@@ -215,8 +218,8 @@ extern struct iscsi_cls_session *iscsi_alloc_session(struct Scsi_Host *shost,
 					struct iscsi_transport *transport);
 extern int iscsi_add_session(struct iscsi_cls_session *session,
 			     unsigned int target_id);
-extern int iscsi_if_create_session_done(struct iscsi_cls_conn *conn);
-extern int iscsi_if_destroy_session_done(struct iscsi_cls_conn *conn);
+extern int iscsi_session_event(struct iscsi_cls_session *session,
+			       enum iscsi_uevent_e event);
 extern struct iscsi_cls_session *iscsi_create_session(struct Scsi_Host *shost,
 						struct iscsi_transport *t,
 						unsigned int target_id);
-- 
cgit v1.2.3


From f6d5180c78780d63b0577edeb3ce41eeb3e93eea Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 13 Dec 2007 12:43:30 -0600
Subject: [SCSI] libiscsi: fix nop handling

During root boot and shutdown the target could send us nops.
At this time iscsid cannot be running, so the target will drop
the session and the boot or shutdown will hang.

To handle this and allow us to better control when to check the network
this patch moves the nop handling to the kernel.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/infiniband/ulp/iser/iscsi_iser.c |   4 +-
 drivers/scsi/iscsi_tcp.c                 |   4 +-
 drivers/scsi/libiscsi.c                  | 331 +++++++++++++++++++++++++------
 drivers/scsi/scsi_transport_iscsi.c      |   4 +
 include/scsi/iscsi_if.h                  |  11 +
 include/scsi/libiscsi.h                  |   8 +
 6 files changed, 294 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index a2622f4bd858..265606422278 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -577,7 +577,9 @@ static struct iscsi_transport iscsi_iser_transport = {
 				  ISCSI_PERSISTENT_ADDRESS |
 				  ISCSI_TARGET_NAME | ISCSI_TPGT |
 				  ISCSI_USERNAME | ISCSI_PASSWORD |
-				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN,
+				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN |
+				  ISCSI_FAST_ABORT | ISCSI_ABORT_TMO |
+				  ISCSI_PING_TMO | ISCSI_RECV_TMO,
 	.host_param_mask	= ISCSI_HOST_HWADDRESS |
 				  ISCSI_HOST_NETDEV_NAME |
 				  ISCSI_HOST_INITIATOR_NAME,
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 90eae8e0d978..9b418522c9bf 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -2246,7 +2246,9 @@ static struct iscsi_transport iscsi_tcp_transport = {
 				  ISCSI_TARGET_NAME | ISCSI_TPGT |
 				  ISCSI_USERNAME | ISCSI_PASSWORD |
 				  ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN |
-				  ISCSI_FAST_ABORT,
+				  ISCSI_FAST_ABORT | ISCSI_ABORT_TMO |
+				  ISCSI_LU_RESET_TMO |
+				  ISCSI_PING_TMO | ISCSI_RECV_TMO,
 	.host_param_mask	= ISCSI_HOST_HWADDRESS | ISCSI_HOST_IPADDRESS |
 				  ISCSI_HOST_INITIATOR_NAME |
 				  ISCSI_HOST_NETDEV_NAME,
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 5205ef2c29b2..96883614ba08 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -313,11 +313,70 @@ void iscsi_free_mgmt_task(struct iscsi_conn *conn,
 	list_del_init(&mtask->running);
 	if (conn->login_mtask == mtask)
 		return;
+
+	if (conn->ping_mtask == mtask)
+		conn->ping_mtask = NULL;
 	__kfifo_put(conn->session->mgmtpool.queue,
 		    (void*)&mtask, sizeof(void*));
 }
 EXPORT_SYMBOL_GPL(iscsi_free_mgmt_task);
 
+static struct iscsi_mgmt_task *
+__iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
+		      char *data, uint32_t data_size)
+{
+	struct iscsi_session *session = conn->session;
+	struct iscsi_mgmt_task *mtask;
+
+	if (session->state == ISCSI_STATE_TERMINATE)
+		return NULL;
+
+	if (hdr->opcode == (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) ||
+	    hdr->opcode == (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE))
+		/*
+		 * Login and Text are sent serially, in
+		 * request-followed-by-response sequence.
+		 * Same mtask can be used. Same ITT must be used.
+		 * Note that login_mtask is preallocated at conn_create().
+		 */
+		mtask = conn->login_mtask;
+	else {
+		BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE);
+		BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED);
+
+		if (!__kfifo_get(session->mgmtpool.queue,
+				 (void*)&mtask, sizeof(void*)))
+			return NULL;
+	}
+
+	if (data_size) {
+		memcpy(mtask->data, data, data_size);
+		mtask->data_count = data_size;
+	} else
+		mtask->data_count = 0;
+
+	memcpy(mtask->hdr, hdr, sizeof(struct iscsi_hdr));
+	INIT_LIST_HEAD(&mtask->running);
+	list_add_tail(&mtask->running, &conn->mgmtqueue);
+	return mtask;
+}
+
+int iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr,
+			char *data, uint32_t data_size)
+{
+	struct iscsi_conn *conn = cls_conn->dd_data;
+	struct iscsi_session *session = conn->session;
+	int err = 0;
+
+	spin_lock_bh(&session->lock);
+	if (!__iscsi_conn_send_pdu(conn, hdr, data, data_size))
+		err = -EPERM;
+	spin_unlock_bh(&session->lock);
+	scsi_queue_work(session->host, &conn->xmitwork);
+	return err;
+}
+EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu);
+
 /**
  * iscsi_cmd_rsp - SCSI Command Response processing
  * @conn: iscsi connection
@@ -409,6 +468,39 @@ static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 	wake_up(&conn->ehwait);
 }
 
+static void iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr)
+{
+        struct iscsi_nopout hdr;
+	struct iscsi_mgmt_task *mtask;
+
+	if (!rhdr && conn->ping_mtask)
+		return;
+
+	memset(&hdr, 0, sizeof(struct iscsi_nopout));
+	hdr.opcode = ISCSI_OP_NOOP_OUT | ISCSI_OP_IMMEDIATE;
+	hdr.flags = ISCSI_FLAG_CMD_FINAL;
+
+	if (rhdr) {
+		memcpy(hdr.lun, rhdr->lun, 8);
+		hdr.ttt = rhdr->ttt;
+		hdr.itt = RESERVED_ITT;
+	} else
+		hdr.ttt = RESERVED_ITT;
+
+	mtask = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)&hdr, NULL, 0);
+	if (!mtask) {
+		printk(KERN_ERR "Could not send nopout\n");
+		return;
+	}
+
+	/* only track our nops */
+	if (!rhdr) {
+		conn->ping_mtask = mtask;
+		conn->last_ping = jiffies;
+	}
+	scsi_queue_work(conn->session->host, &conn->xmitwork);
+}
+
 static int iscsi_handle_reject(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			       char *data, int datalen)
 {
@@ -453,6 +545,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 	struct iscsi_mgmt_task *mtask;
 	uint32_t itt;
 
+	conn->last_recv = jiffies;
 	if (hdr->itt != RESERVED_ITT)
 		itt = get_itt(hdr->itt);
 	else
@@ -520,14 +613,22 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			iscsi_free_mgmt_task(conn, mtask);
 			break;
 		case ISCSI_OP_NOOP_IN:
-			if (hdr->ttt != cpu_to_be32(ISCSI_RESERVED_TAG) || datalen) {
+			if (hdr->ttt != cpu_to_be32(ISCSI_RESERVED_TAG) ||
+			    datalen) {
 				rc = ISCSI_ERR_PROTO;
 				break;
 			}
 			conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1;
 
-			if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen))
-				rc = ISCSI_ERR_CONN_FAILED;
+			if (conn->ping_mtask != mtask) {
+				/*
+				 * If this is not in response to one of our
+				 * nops then it must be from userspace.
+				 */
+				if (iscsi_recv_pdu(conn->cls_conn, hdr, data,
+						   datalen))
+					rc = ISCSI_ERR_CONN_FAILED;
+			}
 			iscsi_free_mgmt_task(conn, mtask);
 			break;
 		default:
@@ -547,8 +648,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			if (hdr->ttt == cpu_to_be32(ISCSI_RESERVED_TAG))
 				break;
 
-			if (iscsi_recv_pdu(conn->cls_conn, hdr, NULL, 0))
-				rc = ISCSI_ERR_CONN_FAILED;
+			iscsi_send_nopout(conn, (struct iscsi_nopin*)hdr);
 			break;
 		case ISCSI_OP_REJECT:
 			rc = iscsi_handle_reject(conn, hdr, data, datalen);
@@ -1003,62 +1103,6 @@ int iscsi_change_queue_depth(struct scsi_device *sdev, int depth)
 }
 EXPORT_SYMBOL_GPL(iscsi_change_queue_depth);
 
-static struct iscsi_mgmt_task *
-__iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
-		      char *data, uint32_t data_size)
-{
-	struct iscsi_session *session = conn->session;
-	struct iscsi_mgmt_task *mtask;
-
-	if (session->state == ISCSI_STATE_TERMINATE)
-		return NULL;
-
-	if (hdr->opcode == (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) ||
-	    hdr->opcode == (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE))
-		/*
-		 * Login and Text are sent serially, in
-		 * request-followed-by-response sequence.
-		 * Same mtask can be used. Same ITT must be used.
-		 * Note that login_mtask is preallocated at conn_create().
-		 */
-		mtask = conn->login_mtask;
-	else {
-		BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE);
-		BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED);
-
-		if (!__kfifo_get(session->mgmtpool.queue,
-				 (void*)&mtask, sizeof(void*)))
-			return NULL;
-	}
-
-	if (data_size) {
-		memcpy(mtask->data, data, data_size);
-		mtask->data_count = data_size;
-	} else
-		mtask->data_count = 0;
-
-	memcpy(mtask->hdr, hdr, sizeof(struct iscsi_hdr));
-	INIT_LIST_HEAD(&mtask->running);
-	list_add_tail(&mtask->running, &conn->mgmtqueue);
-	return mtask;
-}
-
-int iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr,
-			char *data, uint32_t data_size)
-{
-	struct iscsi_conn *conn = cls_conn->dd_data;
-	struct iscsi_session *session = conn->session;
-	int err = 0;
-
-	spin_lock_bh(&session->lock);
-	if (!__iscsi_conn_send_pdu(conn, hdr, data, data_size))
-		err = -EPERM;
-	spin_unlock_bh(&session->lock);
-	scsi_queue_work(session->host, &conn->xmitwork);
-	return err;
-}
-EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu);
-
 void iscsi_session_recovery_timedout(struct iscsi_cls_session *cls_session)
 {
 	struct iscsi_session *session = class_to_transport_session(cls_session);
@@ -1134,7 +1178,8 @@ static void iscsi_tmf_timedout(unsigned long data)
 }
 
 static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
-				   struct iscsi_tm *hdr, int age)
+				   struct iscsi_tm *hdr, int age,
+				   int timeout)
 {
 	struct iscsi_session *session = conn->session;
 	struct iscsi_mgmt_task *mtask;
@@ -1149,7 +1194,7 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
 		return -EPERM;
 	}
 	conn->tmfcmd_pdus_cnt++;
-	conn->tmf_timer.expires = 30 * HZ + jiffies;
+	conn->tmf_timer.expires = timeout * HZ + jiffies;
 	conn->tmf_timer.function = iscsi_tmf_timedout;
 	conn->tmf_timer.data = (unsigned long)conn;
 	add_timer(&conn->tmf_timer);
@@ -1233,6 +1278,106 @@ static void iscsi_start_tx(struct iscsi_conn *conn)
 	scsi_queue_work(conn->session->host, &conn->xmitwork);
 }
 
+static enum scsi_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *scmd)
+{
+	struct iscsi_cls_session *cls_session;
+	struct iscsi_session *session;
+	struct iscsi_conn *conn;
+	enum scsi_eh_timer_return rc = EH_NOT_HANDLED;
+
+	cls_session = starget_to_session(scsi_target(scmd->device));
+	session = class_to_transport_session(cls_session);
+
+	debug_scsi("scsi cmd %p timedout\n", scmd);
+
+	spin_lock(&session->lock);
+	if (session->state != ISCSI_STATE_LOGGED_IN) {
+		/*
+		 * We are probably in the middle of iscsi recovery so let
+		 * that complete and handle the error.
+		 */
+		rc = EH_RESET_TIMER;
+		goto done;
+	}
+
+	conn = session->leadconn;
+	if (!conn) {
+		/* In the middle of shuting down */
+		rc = EH_RESET_TIMER;
+		goto done;
+	}
+
+	if (!conn->recv_timeout && !conn->ping_timeout)
+		goto done;
+	/*
+	 * if the ping timedout then we are in the middle of cleaning up
+	 * and can let the iscsi eh handle it
+	 */
+	if (time_before_eq(conn->last_recv + (conn->recv_timeout * HZ) +
+			    (conn->ping_timeout * HZ), jiffies))
+		rc = EH_RESET_TIMER;
+	/*
+	 * if we are about to check the transport then give the command
+	 * more time
+	 */
+	if (time_before_eq(conn->last_recv + (conn->recv_timeout * HZ),
+			   jiffies))
+		rc = EH_RESET_TIMER;
+	/* if in the middle of checking the transport then give us more time */
+	if (conn->ping_mtask)
+		rc = EH_RESET_TIMER;
+done:
+	spin_unlock(&session->lock);
+	debug_scsi("return %s\n", rc == EH_RESET_TIMER ? "timer reset" : "nh");
+	return rc;
+}
+
+static void iscsi_check_transport_timeouts(unsigned long data)
+{
+	struct iscsi_conn *conn = (struct iscsi_conn *)data;
+	struct iscsi_session *session = conn->session;
+	unsigned long timeout, next_timeout = 0, last_recv;
+
+	spin_lock(&session->lock);
+	if (session->state != ISCSI_STATE_LOGGED_IN)
+		goto done;
+
+	timeout = conn->recv_timeout;
+	if (!timeout)
+		goto done;
+
+	timeout *= HZ;
+	last_recv = conn->last_recv;
+	if (time_before_eq(last_recv + timeout + (conn->ping_timeout * HZ),
+			   jiffies)) {
+		printk(KERN_ERR "ping timeout of %d secs expired, "
+		       "last rx %lu, last ping %lu, now %lu\n",
+		       conn->ping_timeout, last_recv,
+		       conn->last_ping, jiffies);
+		spin_unlock(&session->lock);
+		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+		return;
+	}
+
+	if (time_before_eq(last_recv + timeout, jiffies)) {
+		if (time_before_eq(conn->last_ping, last_recv)) {
+			/* send a ping to try to provoke some traffic */
+			debug_scsi("Sending nopout as ping on conn %p\n", conn);
+			iscsi_send_nopout(conn, NULL);
+		}
+		next_timeout = last_recv + timeout + (conn->ping_timeout * HZ);
+	} else {
+		next_timeout = last_recv + timeout;
+	}
+
+	if (next_timeout) {
+		debug_scsi("Setting next tmo %lu\n", next_timeout);
+		mod_timer(&conn->transport_timer, next_timeout);
+	}
+done:
+	spin_unlock(&session->lock);
+}
+
 static void iscsi_prep_abort_task_pdu(struct iscsi_cmd_task *ctask,
 				      struct iscsi_tm *hdr)
 {
@@ -1304,7 +1449,7 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 	hdr = &conn->tmhdr;
 	iscsi_prep_abort_task_pdu(ctask, hdr);
 
-	if (iscsi_exec_task_mgmt_fn(conn, hdr, age)) {
+	if (iscsi_exec_task_mgmt_fn(conn, hdr, age, session->abort_timeout)) {
 		rc = FAILED;
 		goto failed;
 	}
@@ -1365,7 +1510,7 @@ static void iscsi_prep_lun_reset_pdu(struct scsi_cmnd *sc, struct iscsi_tm *hdr)
 	hdr->flags = ISCSI_TM_FUNC_LOGICAL_UNIT_RESET & ISCSI_FLAG_TM_FUNC_MASK;
 	hdr->flags |= ISCSI_FLAG_CMD_FINAL;
 	int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun);
-	hdr->rtt = ISCSI_RESERVED_TAG;
+	hdr->rtt = RESERVED_ITT;
 }
 
 int iscsi_eh_device_reset(struct scsi_cmnd *sc)
@@ -1396,7 +1541,8 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc)
 	hdr = &conn->tmhdr;
 	iscsi_prep_lun_reset_pdu(sc, hdr);
 
-	if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age)) {
+	if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age,
+				    session->lu_reset_timeout)) {
 		rc = FAILED;
 		goto unlock;
 	}
@@ -1572,12 +1718,14 @@ iscsi_session_setup(struct iscsi_transport *iscsit,
 	shost->max_cmd_len = iscsit->max_cmd_len;
 	shost->transportt = scsit;
 	shost->transportt->create_work_queue = 1;
+	shost->transportt->eh_timed_out = iscsi_eh_cmd_timed_out;
 	*hostno = shost->host_no;
 
 	session = iscsi_hostdata(shost->hostdata);
 	memset(session, 0, sizeof(struct iscsi_session));
 	session->host = shost;
 	session->state = ISCSI_STATE_FREE;
+	session->fast_abort = 1;
 	session->mgmtpool_max = ISCSI_MGMT_CMDS_MAX;
 	session->cmds_max = cmds_max;
 	session->queued_cmdsn = session->cmdsn = initial_cmdsn;
@@ -1708,6 +1856,11 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 	conn->id = conn_idx;
 	conn->exp_statsn = 0;
 	conn->tmf_state = TMF_INITIAL;
+
+	init_timer(&conn->transport_timer);
+	conn->transport_timer.data = (unsigned long)conn;
+	conn->transport_timer.function = iscsi_check_transport_timeouts;
+
 	INIT_LIST_HEAD(&conn->run_list);
 	INIT_LIST_HEAD(&conn->mgmt_run_list);
 	INIT_LIST_HEAD(&conn->mgmtqueue);
@@ -1757,6 +1910,8 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 	struct iscsi_session *session = conn->session;
 	unsigned long flags;
 
+	del_timer_sync(&conn->transport_timer);
+
 	spin_lock_bh(&session->lock);
 	conn->c_stage = ISCSI_CONN_CLEANUP_WAIT;
 	if (session->leadconn == conn) {
@@ -1823,11 +1978,29 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn)
 		return -EINVAL;
 	}
 
+	if (conn->ping_timeout && !conn->recv_timeout) {
+		printk(KERN_ERR "iscsi: invalid recv timeout of zero "
+		      "Using 5 seconds\n.");
+		conn->recv_timeout = 5;
+	}
+
+	if (conn->recv_timeout && !conn->ping_timeout) {
+		printk(KERN_ERR "iscsi: invalid ping timeout of zero "
+		      "Using 5 seconds.\n");
+		conn->ping_timeout = 5;
+	}
+
 	spin_lock_bh(&session->lock);
 	conn->c_stage = ISCSI_CONN_STARTED;
 	session->state = ISCSI_STATE_LOGGED_IN;
 	session->queued_cmdsn = session->cmdsn;
 
+	conn->last_recv = jiffies;
+	conn->last_ping = jiffies;
+	if (conn->recv_timeout && conn->ping_timeout)
+		mod_timer(&conn->transport_timer,
+			  jiffies + (conn->recv_timeout * HZ));
+
 	switch(conn->stop_stage) {
 	case STOP_CONN_RECOVER:
 		/*
@@ -1879,6 +2052,8 @@ static void iscsi_start_session_recovery(struct iscsi_session *session,
 {
 	int old_stop_stage;
 
+	del_timer_sync(&conn->transport_timer);
+
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
 	if (conn->stop_stage == STOP_CONN_TERM) {
@@ -1993,6 +2168,18 @@ int iscsi_set_param(struct iscsi_cls_conn *cls_conn,
 	case ISCSI_PARAM_FAST_ABORT:
 		sscanf(buf, "%d", &session->fast_abort);
 		break;
+	case ISCSI_PARAM_ABORT_TMO:
+		sscanf(buf, "%d", &session->abort_timeout);
+		break;
+	case ISCSI_PARAM_LU_RESET_TMO:
+		sscanf(buf, "%d", &session->lu_reset_timeout);
+		break;
+	case ISCSI_PARAM_PING_TMO:
+		sscanf(buf, "%d", &conn->ping_timeout);
+		break;
+	case ISCSI_PARAM_RECV_TMO:
+		sscanf(buf, "%d", &conn->recv_timeout);
+		break;
 	case ISCSI_PARAM_MAX_RECV_DLENGTH:
 		sscanf(buf, "%d", &conn->max_recv_dlength);
 		break;
@@ -2110,6 +2297,12 @@ int iscsi_session_get_param(struct iscsi_cls_session *cls_session,
 	case ISCSI_PARAM_FAST_ABORT:
 		len = sprintf(buf, "%d\n", session->fast_abort);
 		break;
+	case ISCSI_PARAM_ABORT_TMO:
+		len = sprintf(buf, "%d\n", session->abort_timeout);
+		break;
+	case ISCSI_PARAM_LU_RESET_TMO:
+		len = sprintf(buf, "%d\n", session->lu_reset_timeout);
+		break;
 	case ISCSI_PARAM_INITIAL_R2T_EN:
 		len = sprintf(buf, "%d\n", session->initial_r2t_en);
 		break;
@@ -2167,6 +2360,12 @@ int iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn,
 	int len;
 
 	switch(param) {
+	case ISCSI_PARAM_PING_TMO:
+		len = sprintf(buf, "%u\n", conn->ping_timeout);
+		break;
+	case ISCSI_PARAM_RECV_TMO:
+		len = sprintf(buf, "%u\n", conn->recv_timeout);
+		break;
 	case ISCSI_PARAM_MAX_RECV_DLENGTH:
 		len = sprintf(buf, "%u\n", conn->max_recv_dlength);
 		break;
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index b82139dc4830..36aa50e45ec7 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -1202,6 +1202,8 @@ iscsi_conn_attr(port, ISCSI_PARAM_CONN_PORT);
 iscsi_conn_attr(exp_statsn, ISCSI_PARAM_EXP_STATSN);
 iscsi_conn_attr(persistent_address, ISCSI_PARAM_PERSISTENT_ADDRESS);
 iscsi_conn_attr(address, ISCSI_PARAM_CONN_ADDRESS);
+iscsi_conn_attr(ping_tmo, ISCSI_PARAM_PING_TMO);
+iscsi_conn_attr(recv_tmo, ISCSI_PARAM_RECV_TMO);
 
 #define iscsi_cdev_to_session(_cdev) \
 	iscsi_dev_to_session(_cdev->dev)
@@ -1437,6 +1439,8 @@ iscsi_register_transport(struct iscsi_transport *tt)
 	SETUP_CONN_RD_ATTR(exp_statsn, ISCSI_EXP_STATSN);
 	SETUP_CONN_RD_ATTR(persistent_address, ISCSI_PERSISTENT_ADDRESS);
 	SETUP_CONN_RD_ATTR(persistent_port, ISCSI_PERSISTENT_PORT);
+	SETUP_CONN_RD_ATTR(ping_tmo, ISCSI_PING_TMO);
+	SETUP_CONN_RD_ATTR(recv_tmo, ISCSI_RECV_TMO);
 
 	BUG_ON(count > ISCSI_CONN_ATTRS);
 	priv->conn_attrs[count] = NULL;
diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h
index 8a4426df6c3a..e19e58423166 100644
--- a/include/scsi/iscsi_if.h
+++ b/include/scsi/iscsi_if.h
@@ -244,6 +244,12 @@ enum iscsi_param {
 	ISCSI_PARAM_PASSWORD_IN,
 
 	ISCSI_PARAM_FAST_ABORT,
+	ISCSI_PARAM_ABORT_TMO,
+	ISCSI_PARAM_LU_RESET_TMO,
+	ISCSI_PARAM_HOST_RESET_TMO,
+
+	ISCSI_PARAM_PING_TMO,
+	ISCSI_PARAM_RECV_TMO,
 	/* must always be last */
 	ISCSI_PARAM_MAX,
 };
@@ -275,6 +281,11 @@ enum iscsi_param {
 #define ISCSI_PASSWORD			(1 << ISCSI_PARAM_PASSWORD)
 #define ISCSI_PASSWORD_IN		(1 << ISCSI_PARAM_PASSWORD_IN)
 #define ISCSI_FAST_ABORT		(1 << ISCSI_PARAM_FAST_ABORT)
+#define ISCSI_ABORT_TMO			(1 << ISCSI_PARAM_ABORT_TMO)
+#define ISCSI_LU_RESET_TMO		(1 << ISCSI_PARAM_LU_RESET_TMO)
+#define ISCSI_HOST_RESET_TMO		(1 << ISCSI_PARAM_HOST_RESET_TMO)
+#define ISCSI_PING_TMO			(1 << ISCSI_PARAM_PING_TMO)
+#define ISCSI_RECV_TMO			(1 << ISCSI_PARAM_RECV_TMO)
 
 /* iSCSI HBA params */
 enum iscsi_host_param {
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index d68f74523f2e..889f51fabab9 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -148,6 +148,12 @@ struct iscsi_conn {
 	 * conn_stop() flag: stop to recover, stop to terminate
 	 */
         int			stop_stage;
+	struct timer_list	transport_timer;
+	unsigned long		last_recv;
+	unsigned long		last_ping;
+	int			ping_timeout;
+	int			recv_timeout;
+	struct iscsi_mgmt_task	*ping_mtask;
 
 	/* iSCSI connection-wide sequencing */
 	uint32_t		exp_statsn;
@@ -238,6 +244,8 @@ struct iscsi_session {
 	uint32_t		queued_cmdsn;
 
 	/* configuration */
+	int			abort_timeout;
+	int			lu_reset_timeout;
 	int			initial_r2t_en;
 	unsigned		max_r2t;
 	int			imm_data_en;
-- 
cgit v1.2.3


From a8ac6311cc21d78fa284cd43f56df2063f536bf1 Mon Sep 17 00:00:00 2001
From: Olaf Kirch <olaf.kirch@oracle.com>
Date: Thu, 13 Dec 2007 12:43:35 -0600
Subject: [SCSI] iscsi: convert xmit path to iscsi chunks

Convert xmit to iscsi chunks.

from michaelc@cs.wisc.edu:

Bug fixes, more digest integration, sg chaining conversion and other
sg wrapper changes, coding style sync up, and removal of io fields,
like pdu_sent, that are not needed.

Signed-off-by: Olaf Kirch <olaf.kirch@oracle.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/infiniband/ulp/iser/iscsi_iser.c |    3 +-
 drivers/scsi/iscsi_tcp.c                 | 1333 ++++++++++++------------------
 drivers/scsi/iscsi_tcp.h                 |   75 +-
 drivers/scsi/libiscsi.c                  |   37 +-
 include/scsi/scsi_transport_iscsi.h      |    2 +-
 5 files changed, 563 insertions(+), 887 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 265606422278..fd69fb35e3b3 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -129,7 +129,7 @@ error:
  * iscsi_iser_cmd_init - Initialize iSCSI SCSI_READ or SCSI_WRITE commands
  *
  **/
-static void
+static int
 iscsi_iser_cmd_init(struct iscsi_cmd_task *ctask)
 {
 	struct iscsi_iser_conn     *iser_conn  = ctask->conn->dd_data;
@@ -138,6 +138,7 @@ iscsi_iser_cmd_init(struct iscsi_cmd_task *ctask)
 	iser_ctask->command_sent = 0;
 	iser_ctask->iser_conn    = iser_conn;
 	iser_ctask_rdma_init(iser_ctask);
+	return 0;
 }
 
 /**
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 9b418522c9bf..7212fe95a66d 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -68,56 +68,10 @@ static unsigned int iscsi_max_lun = 512;
 module_param_named(max_lun, iscsi_max_lun, uint, S_IRUGO);
 
 static int iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
-				   struct iscsi_chunk *chunk);
-
-static inline void
-iscsi_buf_init_iov(struct iscsi_buf *ibuf, char *vbuf, int size)
-{
-	ibuf->sg.page = virt_to_page(vbuf);
-	ibuf->sg.offset = offset_in_page(vbuf);
-	ibuf->sg.length = size;
-	ibuf->sent = 0;
-	ibuf->use_sendmsg = 1;
-}
-
-static inline void
-iscsi_buf_init_sg(struct iscsi_buf *ibuf, struct scatterlist *sg)
-{
-	ibuf->sg.page = sg->page;
-	ibuf->sg.offset = sg->offset;
-	ibuf->sg.length = sg->length;
-	/*
-	 * Fastpath: sg element fits into single page
-	 */
-	if (sg->length + sg->offset <= PAGE_SIZE && !PageSlab(sg->page))
-		ibuf->use_sendmsg = 0;
-	else
-		ibuf->use_sendmsg = 1;
-	ibuf->sent = 0;
-}
-
-static inline int
-iscsi_buf_left(struct iscsi_buf *ibuf)
-{
-	int rc;
-
-	rc = ibuf->sg.length - ibuf->sent;
-	BUG_ON(rc < 0);
-	return rc;
-}
-
-static inline void
-iscsi_hdr_digest(struct iscsi_conn *conn, struct iscsi_buf *buf,
-		 u8* crc)
-{
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-
-	crypto_hash_digest(&tcp_conn->tx_hash, &buf->sg, buf->sg.length, crc);
-	buf->sg.length += ISCSI_DIGEST_SIZE;
-}
+				   struct iscsi_segment *segment);
 
 /*
- * Scatterlist handling: inside the iscsi_chunk, we
+ * Scatterlist handling: inside the iscsi_segment, we
  * remember an index into the scatterlist, and set data/size
  * to the current scatterlist entry. For highmem pages, we
  * kmap as needed.
@@ -130,60 +84,72 @@ iscsi_hdr_digest(struct iscsi_conn *conn, struct iscsi_buf *buf,
  */
 
 /**
- * iscsi_tcp_chunk_init_sg - init indicated scatterlist entry
- * @chunk: the buffer object
- * @idx: index into scatterlist
+ * iscsi_tcp_segment_init_sg - init indicated scatterlist entry
+ * @segment: the buffer object
+ * @sg: scatterlist
  * @offset: byte offset into that sg entry
  *
- * This function sets up the chunk so that subsequent
+ * This function sets up the segment so that subsequent
  * data is copied to the indicated sg entry, at the given
  * offset.
  */
 static inline void
-iscsi_tcp_chunk_init_sg(struct iscsi_chunk *chunk,
-			unsigned int idx, unsigned int offset)
+iscsi_tcp_segment_init_sg(struct iscsi_segment *segment,
+			  struct scatterlist *sg, unsigned int offset)
 {
-	struct scatterlist *sg;
-
-	BUG_ON(chunk->sg == NULL);
-
-	sg = &chunk->sg[idx];
-	chunk->sg_index = idx;
-	chunk->sg_offset = offset;
-	chunk->size = min(sg->length - offset, chunk->total_size);
-	chunk->data = NULL;
+	segment->sg = sg;
+	segment->sg_offset = offset;
+	segment->size = min(sg->length - offset,
+			    segment->total_size - segment->total_copied);
+	segment->data = NULL;
 }
 
 /**
- * iscsi_tcp_chunk_map - map the current S/G page
- * @chunk: iscsi chunk
+ * iscsi_tcp_segment_map - map the current S/G page
+ * @segment: iscsi_segment
+ * @recv: 1 if called from recv path
  *
  * We only need to possibly kmap data if scatter lists are being used,
  * because the iscsi passthrough and internal IO paths will never use high
  * mem pages.
  */
 static inline void
-iscsi_tcp_chunk_map(struct iscsi_chunk *chunk)
+iscsi_tcp_segment_map(struct iscsi_segment *segment, int recv)
 {
 	struct scatterlist *sg;
 
-	if (chunk->data != NULL || !chunk->sg)
+	if (segment->data != NULL || !segment->sg)
 		return;
 
-	sg = &chunk->sg[chunk->sg_index];
-	BUG_ON(chunk->sg_mapped);
+	sg = segment->sg;
+	BUG_ON(segment->sg_mapped);
 	BUG_ON(sg->length == 0);
-	chunk->sg_mapped = kmap_atomic(sg->page, KM_SOFTIRQ0);
-	chunk->data = chunk->sg_mapped + sg->offset + chunk->sg_offset;
+
+	/*
+	 * If the page count is greater than one it is ok to send
+	 * to the network layer's zero copy send path. If not we
+	 * have to go the slow sendmsg path. We always map for the
+	 * recv path.
+	 */
+	if (page_count(sg_page(sg)) >= 1 && !recv)
+		return;
+
+	debug_tcp("iscsi_tcp_segment_map %s %p\n", recv ? "recv" : "xmit",
+		  segment);
+	segment->sg_mapped = kmap_atomic(sg_page(sg), KM_SOFTIRQ0);
+	segment->data = segment->sg_mapped + sg->offset + segment->sg_offset;
 }
 
 static inline void
-iscsi_tcp_chunk_unmap(struct iscsi_chunk *chunk)
+iscsi_tcp_segment_unmap(struct iscsi_segment *segment)
 {
-	if (chunk->sg_mapped) {
-		kunmap_atomic(chunk->sg_mapped, KM_SOFTIRQ0);
-		chunk->sg_mapped = NULL;
-		chunk->data = NULL;
+	debug_tcp("iscsi_tcp_segment_unmap %p\n", segment);
+
+	if (segment->sg_mapped) {
+		debug_tcp("iscsi_tcp_segment_unmap valid\n");
+		kunmap_atomic(segment->sg_mapped, KM_SOFTIRQ0);
+		segment->sg_mapped = NULL;
+		segment->data = NULL;
 	}
 }
 
@@ -191,23 +157,24 @@ iscsi_tcp_chunk_unmap(struct iscsi_chunk *chunk)
  * Splice the digest buffer into the buffer
  */
 static inline void
-iscsi_tcp_chunk_splice_digest(struct iscsi_chunk *chunk, void *digest)
+iscsi_tcp_segment_splice_digest(struct iscsi_segment *segment, void *digest)
 {
-	chunk->data = digest;
-	chunk->digest_len = ISCSI_DIGEST_SIZE;
-	chunk->total_size += ISCSI_DIGEST_SIZE;
-	chunk->size = ISCSI_DIGEST_SIZE;
-	chunk->copied = 0;
-	chunk->sg = NULL;
-	chunk->sg_index = 0;
-	chunk->hash = NULL;
+	segment->data = digest;
+	segment->digest_len = ISCSI_DIGEST_SIZE;
+	segment->total_size += ISCSI_DIGEST_SIZE;
+	segment->size = ISCSI_DIGEST_SIZE;
+	segment->copied = 0;
+	segment->sg = NULL;
+	segment->hash = NULL;
 }
 
 /**
- * iscsi_tcp_chunk_done - check whether the chunk is complete
- * @chunk: iscsi chunk to check
+ * iscsi_tcp_segment_done - check whether the segment is complete
+ * @segment: iscsi segment to check
+ * @recv: set to one of this is called from the recv path
+ * @copied: number of bytes copied
  *
- * Check if we're done receiving this chunk. If the receive
+ * Check if we're done receiving this segment. If the receive
  * buffer is full but we expect more data, move on to the
  * next entry in the scatterlist.
  *
@@ -217,62 +184,145 @@ iscsi_tcp_chunk_splice_digest(struct iscsi_chunk *chunk, void *digest)
  * This function must be re-entrant.
  */
 static inline int
-iscsi_tcp_chunk_done(struct iscsi_chunk *chunk)
+iscsi_tcp_segment_done(struct iscsi_segment *segment, int recv, unsigned copied)
 {
 	static unsigned char padbuf[ISCSI_PAD_LEN];
+	struct scatterlist sg;
 	unsigned int pad;
 
-	if (chunk->copied < chunk->size) {
-		iscsi_tcp_chunk_map(chunk);
+	debug_tcp("copied %u %u size %u %s\n", segment->copied, copied,
+		  segment->size, recv ? "recv" : "xmit");
+	if (segment->hash && copied) {
+		/*
+		 * If a segment is kmapd we must unmap it before sending
+		 * to the crypto layer since that will try to kmap it again.
+		 */
+		iscsi_tcp_segment_unmap(segment);
+
+		if (!segment->data) {
+			sg_init_table(&sg, 1);
+			sg_set_page(&sg, sg_page(segment->sg), copied,
+				    segment->copied + segment->sg_offset +
+							segment->sg->offset);
+		} else
+			sg_init_one(&sg, segment->data + segment->copied,
+				    copied);
+		crypto_hash_update(segment->hash, &sg, copied);
+	}
+
+	segment->copied += copied;
+	if (segment->copied < segment->size) {
+		iscsi_tcp_segment_map(segment, recv);
 		return 0;
 	}
 
-	chunk->total_copied += chunk->copied;
-	chunk->copied = 0;
-	chunk->size = 0;
+	segment->total_copied += segment->copied;
+	segment->copied = 0;
+	segment->size = 0;
 
 	/* Unmap the current scatterlist page, if there is one. */
-	iscsi_tcp_chunk_unmap(chunk);
+	iscsi_tcp_segment_unmap(segment);
 
 	/* Do we have more scatterlist entries? */
-	if (chunk->total_copied < chunk->total_size) {
+	debug_tcp("total copied %u total size %u\n", segment->total_copied,
+		   segment->total_size);
+	if (segment->total_copied < segment->total_size) {
 		/* Proceed to the next entry in the scatterlist. */
-		iscsi_tcp_chunk_init_sg(chunk, chunk->sg_index + 1, 0);
-		iscsi_tcp_chunk_map(chunk);
-		BUG_ON(chunk->size == 0);
+		iscsi_tcp_segment_init_sg(segment, sg_next(segment->sg),
+					  0);
+		iscsi_tcp_segment_map(segment, recv);
+		BUG_ON(segment->size == 0);
 		return 0;
 	}
 
 	/* Do we need to handle padding? */
-	pad = iscsi_padding(chunk->total_copied);
+	pad = iscsi_padding(segment->total_copied);
 	if (pad != 0) {
 		debug_tcp("consume %d pad bytes\n", pad);
-		chunk->total_size += pad;
-		chunk->size = pad;
-		chunk->data = padbuf;
+		segment->total_size += pad;
+		segment->size = pad;
+		segment->data = padbuf;
 		return 0;
 	}
 
 	/*
-	 * Set us up for receiving the data digest. hdr digest
+	 * Set us up for transferring the data digest. hdr digest
 	 * is completely handled in hdr done function.
 	 */
-	if (chunk->hash) {
-		if (chunk->digest_len == 0) {
-			crypto_hash_final(chunk->hash, chunk->digest);
-			iscsi_tcp_chunk_splice_digest(chunk,
-						      chunk->recv_digest);
-			return 0;
-		}
+	if (segment->hash) {
+		crypto_hash_final(segment->hash, segment->digest);
+		iscsi_tcp_segment_splice_digest(segment,
+				 recv ? segment->recv_digest : segment->digest);
+		return 0;
 	}
 
 	return 1;
 }
 
 /**
- * iscsi_tcp_chunk_recv - copy data to chunk
+ * iscsi_tcp_xmit_segment - transmit segment
  * @tcp_conn: the iSCSI TCP connection
- * @chunk: the buffer to copy to
+ * @segment: the buffer to transmnit
+ *
+ * This function transmits as much of the buffer as
+ * the network layer will accept, and returns the number of
+ * bytes transmitted.
+ *
+ * If CRC hashing is enabled, the function will compute the
+ * hash as it goes. When the entire segment has been transmitted,
+ * it will retrieve the hash value and send it as well.
+ */
+static int
+iscsi_tcp_xmit_segment(struct iscsi_tcp_conn *tcp_conn,
+		       struct iscsi_segment *segment)
+{
+	struct socket *sk = tcp_conn->sock;
+	unsigned int copied = 0;
+	int r = 0;
+
+	while (!iscsi_tcp_segment_done(segment, 0, r)) {
+		struct scatterlist *sg;
+		unsigned int offset, copy;
+		int flags = 0;
+
+		r = 0;
+		offset = segment->copied;
+		copy = segment->size - offset;
+
+		if (segment->total_copied + segment->size < segment->total_size)
+			flags |= MSG_MORE;
+
+		/* Use sendpage if we can; else fall back to sendmsg */
+		if (!segment->data) {
+			sg = segment->sg;
+			offset += segment->sg_offset + sg->offset;
+			r = tcp_conn->sendpage(sk, sg_page(sg), offset, copy,
+					       flags);
+		} else {
+			struct msghdr msg = { .msg_flags = flags };
+			struct kvec iov = {
+				.iov_base = segment->data + offset,
+				.iov_len = copy
+			};
+
+			r = kernel_sendmsg(sk, &msg, &iov, 1, copy);
+		}
+
+		if (r < 0) {
+			iscsi_tcp_segment_unmap(segment);
+			if (copied || r == -EAGAIN)
+				break;
+			return r;
+		}
+		copied += r;
+	}
+	return copied;
+}
+
+/**
+ * iscsi_tcp_segment_recv - copy data to segment
+ * @tcp_conn: the iSCSI TCP connection
+ * @segment: the buffer to copy to
  * @ptr: data pointer
  * @len: amount of data available
  *
@@ -287,29 +337,24 @@ iscsi_tcp_chunk_done(struct iscsi_chunk *chunk)
  * just way we do for network layer checksums.
  */
 static int
-iscsi_tcp_chunk_recv(struct iscsi_tcp_conn *tcp_conn,
-		     struct iscsi_chunk *chunk, const void *ptr,
-		     unsigned int len)
+iscsi_tcp_segment_recv(struct iscsi_tcp_conn *tcp_conn,
+		       struct iscsi_segment *segment, const void *ptr,
+		       unsigned int len)
 {
-	struct scatterlist sg;
-	unsigned int copy, copied = 0;
-
-	while (!iscsi_tcp_chunk_done(chunk)) {
-		if (copied == len)
-			goto out;
+	unsigned int copy = 0, copied = 0;
 
-		copy = min(len - copied, chunk->size - chunk->copied);
-		memcpy(chunk->data + chunk->copied, ptr + copied, copy);
-
-		if (chunk->hash) {
-			sg_init_one(&sg, ptr + copied, copy);
-			crypto_hash_update(chunk->hash, &sg, copy);
+	while (!iscsi_tcp_segment_done(segment, 1, copy)) {
+		if (copied == len) {
+			debug_tcp("iscsi_tcp_segment_recv copied %d bytes\n",
+				  len);
+			break;
 		}
-		chunk->copied += copy;
+
+		copy = min(len - copied, segment->size - segment->copied);
+		debug_tcp("iscsi_tcp_segment_recv copying %d\n", copy);
+		memcpy(segment->data + segment->copied, ptr + copied, copy);
 		copied += copy;
 	}
-
-out:
 	return copied;
 }
 
@@ -325,12 +370,13 @@ iscsi_tcp_dgst_header(struct hash_desc *hash, const void *hdr, size_t hdrlen,
 
 static inline int
 iscsi_tcp_dgst_verify(struct iscsi_tcp_conn *tcp_conn,
-		      struct iscsi_chunk *chunk)
+		      struct iscsi_segment *segment)
 {
-	if (!chunk->digest_len)
+	if (!segment->digest_len)
 		return 1;
 
-	if (memcmp(chunk->recv_digest, chunk->digest, chunk->digest_len)) {
+	if (memcmp(segment->recv_digest, segment->digest,
+		   segment->digest_len)) {
 		debug_scsi("digest mismatch\n");
 		return 0;
 	}
@@ -339,55 +385,59 @@ iscsi_tcp_dgst_verify(struct iscsi_tcp_conn *tcp_conn,
 }
 
 /*
- * Helper function to set up chunk buffer
+ * Helper function to set up segment buffer
  */
 static inline void
-__iscsi_chunk_init(struct iscsi_chunk *chunk, size_t size,
-		   iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+__iscsi_segment_init(struct iscsi_segment *segment, size_t size,
+		     iscsi_segment_done_fn_t *done, struct hash_desc *hash)
 {
-	memset(chunk, 0, sizeof(*chunk));
-	chunk->total_size = size;
-	chunk->done = done;
+	memset(segment, 0, sizeof(*segment));
+	segment->total_size = size;
+	segment->done = done;
 
 	if (hash) {
-		chunk->hash = hash;
+		segment->hash = hash;
 		crypto_hash_init(hash);
 	}
 }
 
 static inline void
-iscsi_chunk_init_linear(struct iscsi_chunk *chunk, void *data, size_t size,
-			iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+iscsi_segment_init_linear(struct iscsi_segment *segment, void *data,
+			  size_t size, iscsi_segment_done_fn_t *done,
+			  struct hash_desc *hash)
 {
-	__iscsi_chunk_init(chunk, size, done, hash);
-	chunk->data = data;
-	chunk->size = size;
+	__iscsi_segment_init(segment, size, done, hash);
+	segment->data = data;
+	segment->size = size;
 }
 
 static inline int
-iscsi_chunk_seek_sg(struct iscsi_chunk *chunk,
-		    struct scatterlist *sg, unsigned int sg_count,
-		    unsigned int offset, size_t size,
-		    iscsi_chunk_done_fn_t *done, struct hash_desc *hash)
+iscsi_segment_seek_sg(struct iscsi_segment *segment,
+		      struct scatterlist *sg_list, unsigned int sg_count,
+		      unsigned int offset, size_t size,
+		      iscsi_segment_done_fn_t *done, struct hash_desc *hash)
 {
+	struct scatterlist *sg;
 	unsigned int i;
 
-	__iscsi_chunk_init(chunk, size, done, hash);
-	for (i = 0; i < sg_count; ++i) {
-		if (offset < sg[i].length) {
-			chunk->sg = sg;
-			chunk->sg_count = sg_count;
-			iscsi_tcp_chunk_init_sg(chunk, i, offset);
+	debug_scsi("iscsi_segment_seek_sg offset %u size %llu\n",
+		  offset, size);
+	__iscsi_segment_init(segment, size, done, hash);
+	for_each_sg(sg_list, sg, sg_count, i) {
+		debug_scsi("sg %d, len %u offset %u\n", i, sg->length,
+			   sg->offset);
+		if (offset < sg->length) {
+			iscsi_tcp_segment_init_sg(segment, sg, offset);
 			return 0;
 		}
-		offset -= sg[i].length;
+		offset -= sg->length;
 	}
 
 	return ISCSI_ERR_DATA_OFFSET;
 }
 
 /**
- * iscsi_tcp_hdr_recv_prep - prep chunk for hdr reception
+ * iscsi_tcp_hdr_recv_prep - prep segment for hdr reception
  * @tcp_conn: iscsi connection to prep for
  *
  * This function always passes NULL for the hash argument, because when this
@@ -399,7 +449,7 @@ iscsi_tcp_hdr_recv_prep(struct iscsi_tcp_conn *tcp_conn)
 {
 	debug_tcp("iscsi_tcp_hdr_recv_prep(%p%s)\n", tcp_conn,
 		  tcp_conn->iscsi_conn->hdrdgst_en ? ", digest enabled" : "");
-	iscsi_chunk_init_linear(&tcp_conn->in.chunk,
+	iscsi_segment_init_linear(&tcp_conn->in.segment,
 				tcp_conn->in.hdr_buf, sizeof(struct iscsi_hdr),
 				iscsi_tcp_hdr_recv_done, NULL);
 }
@@ -409,12 +459,12 @@ iscsi_tcp_hdr_recv_prep(struct iscsi_tcp_conn *tcp_conn)
  */
 static int
 iscsi_tcp_data_recv_done(struct iscsi_tcp_conn *tcp_conn,
-			 struct iscsi_chunk *chunk)
+			 struct iscsi_segment *segment)
 {
 	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
 	int rc = 0;
 
-	if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+	if (!iscsi_tcp_dgst_verify(tcp_conn, segment))
 		return ISCSI_ERR_DATA_DGST;
 
 	rc = iscsi_complete_pdu(conn, tcp_conn->in.hdr,
@@ -435,7 +485,7 @@ iscsi_tcp_data_recv_prep(struct iscsi_tcp_conn *tcp_conn)
 	if (conn->datadgst_en)
 		rx_hash = &tcp_conn->rx_hash;
 
-	iscsi_chunk_init_linear(&tcp_conn->in.chunk,
+	iscsi_segment_init_linear(&tcp_conn->in.segment,
 				conn->data, tcp_conn->in.datalen,
 				iscsi_tcp_data_recv_done, rx_hash);
 }
@@ -448,7 +498,6 @@ iscsi_tcp_cleanup_ctask(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 {
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
 	struct iscsi_r2t_info *r2t;
-	struct scsi_cmnd *sc;
 
 	/* flush ctask's r2t queues */
 	while (__kfifo_get(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*))) {
@@ -457,12 +506,12 @@ iscsi_tcp_cleanup_ctask(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 		debug_scsi("iscsi_tcp_cleanup_ctask pending r2t dropped\n");
 	}
 
-	sc = ctask->sc;
-	if (unlikely(!sc))
-		return;
-
-	tcp_ctask->xmstate = XMSTATE_IDLE;
-	tcp_ctask->r2t = NULL;
+	r2t = tcp_ctask->r2t;
+	if (r2t != NULL) {
+		__kfifo_put(tcp_ctask->r2tpool.queue, (void*)&r2t,
+			    sizeof(void*));
+		tcp_ctask->r2t = NULL;
+	}
 }
 
 /**
@@ -481,11 +530,6 @@ iscsi_data_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	int datasn = be32_to_cpu(rhdr->datasn);
 
 	iscsi_update_cmdsn(session, (struct iscsi_nopin*)rhdr);
-	/*
-	 * setup Data-In byte counter (gets decremented..)
-	 */
-	ctask->data_count = tcp_conn->in.datalen;
-
 	if (tcp_conn->in.datalen == 0)
 		return 0;
 
@@ -543,9 +587,6 @@ iscsi_solicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 			struct iscsi_r2t_info *r2t)
 {
 	struct iscsi_data *hdr;
-	struct scsi_cmnd *sc = ctask->sc;
-	int i, sg_count = 0;
-	struct scatterlist *sg;
 
 	hdr = &r2t->dtask.hdr;
 	memset(hdr, 0, sizeof(struct iscsi_data));
@@ -569,34 +610,6 @@ iscsi_solicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 	conn->dataout_pdus_cnt++;
 
 	r2t->sent = 0;
-
-	iscsi_buf_init_iov(&r2t->headbuf, (char*)hdr,
-			   sizeof(struct iscsi_hdr));
-
-	sg = scsi_sglist(sc);
-	r2t->sg = NULL;
-	for (i = 0; i < scsi_sg_count(sc); i++, sg += 1) {
-		/* FIXME: prefetch ? */
-		if (sg_count + sg->length > r2t->data_offset) {
-			int page_offset;
-
-			/* sg page found! */
-
-			/* offset within this page */
-			page_offset = r2t->data_offset - sg_count;
-
-			/* fill in this buffer */
-			iscsi_buf_init_sg(&r2t->sendbuf, sg);
-			r2t->sendbuf.sg.offset += page_offset;
-			r2t->sendbuf.sg.length -= page_offset;
-
-			/* xmit logic will continue with next one */
-			r2t->sg = sg + 1;
-			break;
-		}
-		sg_count += sg->length;
-	}
-	BUG_ON(r2t->sg == NULL);
 }
 
 /**
@@ -670,7 +683,6 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 
 	tcp_ctask->exp_datasn = r2tsn + 1;
 	__kfifo_put(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*));
-	tcp_ctask->xmstate |= XMSTATE_SOL_HDR_INIT;
 	conn->r2t_pdus_cnt++;
 
 	iscsi_requeue_ctask(ctask);
@@ -684,13 +696,13 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
  */
 static int
 iscsi_tcp_process_data_in(struct iscsi_tcp_conn *tcp_conn,
-			  struct iscsi_chunk *chunk)
+			  struct iscsi_segment *segment)
 {
 	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
 	struct iscsi_hdr *hdr = tcp_conn->in.hdr;
 	int rc;
 
-	if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+	if (!iscsi_tcp_dgst_verify(tcp_conn, segment))
 		return ISCSI_ERR_DATA_DGST;
 
 	/* check for non-exceptional status */
@@ -762,7 +774,7 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 			/*
 			 * Setup copy of Data-In into the Scsi_Cmnd
 			 * Scatterlist case:
-			 * We set up the iscsi_chunk to point to the next
+			 * We set up the iscsi_segment to point to the next
 			 * scatterlist entry to copy to. As we go along,
 			 * we move on to the next scatterlist entry and
 			 * update the digest per-entry.
@@ -774,13 +786,13 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 				  "datalen=%d)\n", tcp_conn,
 				  tcp_ctask->data_offset,
 				  tcp_conn->in.datalen);
-			return iscsi_chunk_seek_sg(&tcp_conn->in.chunk,
-						scsi_sglist(ctask->sc),
-						scsi_sg_count(ctask->sc),
-						tcp_ctask->data_offset,
-						tcp_conn->in.datalen,
-						iscsi_tcp_process_data_in,
-						rx_hash);
+			return iscsi_segment_seek_sg(&tcp_conn->in.segment,
+						     scsi_sglist(ctask->sc),
+						     scsi_sg_count(ctask->sc),
+						     tcp_ctask->data_offset,
+						     tcp_conn->in.datalen,
+						     iscsi_tcp_process_data_in,
+						     rx_hash);
 		}
 		/* fall through */
 	case ISCSI_OP_SCSI_CMD_RSP:
@@ -846,17 +858,6 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 	return rc;
 }
 
-static inline void
-partial_sg_digest_update(struct hash_desc *desc, struct scatterlist *sg,
-			 int offset, int length)
-{
-	struct scatterlist temp;
-
-	sg_init_table(&temp, 1);
-	sg_set_page(&temp, sg_page(sg), length, offset);
-	crypto_hash_update(desc, &temp, length);
-}
-
 /**
  * iscsi_tcp_hdr_recv_done - process PDU header
  *
@@ -866,7 +867,7 @@ partial_sg_digest_update(struct hash_desc *desc, struct scatterlist *sg,
  */
 static int
 iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
-			struct iscsi_chunk *chunk)
+			struct iscsi_segment *segment)
 {
 	struct iscsi_conn *conn = tcp_conn->iscsi_conn;
 	struct iscsi_hdr *hdr;
@@ -876,7 +877,7 @@ iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
 	 * may need to go back to the caller for more.
 	 */
 	hdr = (struct iscsi_hdr *) tcp_conn->in.hdr_buf;
-	if (chunk->copied == sizeof(struct iscsi_hdr) && hdr->hlength) {
+	if (segment->copied == sizeof(struct iscsi_hdr) && hdr->hlength) {
 		/* Bump the header length - the caller will
 		 * just loop around and get the AHS for us, and
 		 * call again. */
@@ -886,8 +887,8 @@ iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
 		if (sizeof(*hdr) + ahslen > sizeof(tcp_conn->in.hdr_buf))
 			return ISCSI_ERR_AHSLEN;
 
-		chunk->total_size += ahslen;
-		chunk->size += ahslen;
+		segment->total_size += ahslen;
+		segment->size += ahslen;
 		return 0;
 	}
 
@@ -895,16 +896,16 @@ iscsi_tcp_hdr_recv_done(struct iscsi_tcp_conn *tcp_conn,
 	 * header digests; if so, set up the recv_digest buffer
 	 * and go back for more. */
 	if (conn->hdrdgst_en) {
-		if (chunk->digest_len == 0) {
-			iscsi_tcp_chunk_splice_digest(chunk,
-						      chunk->recv_digest);
+		if (segment->digest_len == 0) {
+			iscsi_tcp_segment_splice_digest(segment,
+							segment->recv_digest);
 			return 0;
 		}
 		iscsi_tcp_dgst_header(&tcp_conn->rx_hash, hdr,
-				      chunk->total_copied - ISCSI_DIGEST_SIZE,
-				      chunk->digest);
+				      segment->total_copied - ISCSI_DIGEST_SIZE,
+				      segment->digest);
 
-		if (!iscsi_tcp_dgst_verify(tcp_conn, chunk))
+		if (!iscsi_tcp_dgst_verify(tcp_conn, segment))
 			return ISCSI_ERR_HDR_DGST;
 	}
 
@@ -925,7 +926,7 @@ iscsi_tcp_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
 {
 	struct iscsi_conn *conn = rd_desc->arg.data;
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	struct iscsi_chunk *chunk = &tcp_conn->in.chunk;
+	struct iscsi_segment *segment = &tcp_conn->in.segment;
 	struct skb_seq_state seq;
 	unsigned int consumed = 0;
 	int rc = 0;
@@ -943,27 +944,31 @@ iscsi_tcp_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
 		const u8 *ptr;
 
 		avail = skb_seq_read(consumed, &ptr, &seq);
-		if (avail == 0)
+		if (avail == 0) {
+			debug_tcp("no more data avail. Consumed %d\n",
+				  consumed);
 			break;
-		BUG_ON(chunk->copied >= chunk->size);
+		}
+		BUG_ON(segment->copied >= segment->size);
 
 		debug_tcp("skb %p ptr=%p avail=%u\n", skb, ptr, avail);
-		rc = iscsi_tcp_chunk_recv(tcp_conn, chunk, ptr, avail);
+		rc = iscsi_tcp_segment_recv(tcp_conn, segment, ptr, avail);
 		BUG_ON(rc == 0);
 		consumed += rc;
 
-		if (chunk->total_copied >= chunk->total_size) {
-			rc = chunk->done(tcp_conn, chunk);
+		if (segment->total_copied >= segment->total_size) {
+			debug_tcp("segment done\n");
+			rc = segment->done(tcp_conn, segment);
 			if (rc != 0) {
 				skb_abort_seq_read(&seq);
 				goto error;
 			}
 
 			/* The done() functions sets up the
-			 * next chunk. */
+			 * next segment. */
 		}
 	}
-
+	skb_abort_seq_read(&seq);
 	conn->rxdata_octets += consumed;
 	return consumed;
 
@@ -996,7 +1001,7 @@ iscsi_tcp_data_ready(struct sock *sk, int flag)
 
 	/* If we had to (atomically) map a highmem page,
 	 * unmap it now. */
-	iscsi_tcp_chunk_unmap(&tcp_conn->in.chunk);
+	iscsi_tcp_segment_unmap(&tcp_conn->in.segment);
 }
 
 static void
@@ -1076,121 +1081,173 @@ iscsi_conn_restore_callbacks(struct iscsi_tcp_conn *tcp_conn)
 }
 
 /**
- * iscsi_send - generic send routine
- * @sk: kernel's socket
- * @buf: buffer to write from
- * @size: actual size to write
- * @flags: socket's flags
- */
-static inline int
-iscsi_send(struct iscsi_conn *conn, struct iscsi_buf *buf, int size, int flags)
+ * iscsi_xmit - TCP transmit
+ **/
+static int
+iscsi_xmit(struct iscsi_conn *conn)
 {
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	struct socket *sk = tcp_conn->sock;
-	int offset = buf->sg.offset + buf->sent, res;
+	struct iscsi_segment *segment = &tcp_conn->out.segment;
+	unsigned int consumed = 0;
+	int rc = 0;
 
-	/*
-	 * if we got use_sg=0 or are sending something we kmallocd
-	 * then we did not have to do kmap (kmap returns page_address)
-	 *
-	 * if we got use_sg > 0, but had to drop down, we do not
-	 * set clustering so this should only happen for that
-	 * slab case.
-	 */
-	if (buf->use_sendmsg)
-		res = sock_no_sendpage(sk, buf->sg.page, offset, size, flags);
-	else
-		res = tcp_conn->sendpage(sk, buf->sg.page, offset, size, flags);
-
-	if (res >= 0) {
-		conn->txdata_octets += res;
-		buf->sent += res;
-		return res;
+	while (1) {
+		rc = iscsi_tcp_xmit_segment(tcp_conn, segment);
+		if (rc < 0)
+			goto error;
+		if (rc == 0)
+			break;
+
+		consumed += rc;
+
+		if (segment->total_copied >= segment->total_size) {
+			if (segment->done != NULL) {
+				rc = segment->done(tcp_conn, segment);
+				if (rc < 0)
+					goto error;
+			}
+		}
 	}
 
-	tcp_conn->sendpage_failures_cnt++;
-	if (res == -EAGAIN)
-		res = -ENOBUFS;
-	else
-		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
-	return res;
+	debug_tcp("xmit %d bytes\n", consumed);
+
+	conn->txdata_octets += consumed;
+	return consumed;
+
+error:
+	/* Transmit error. We could initiate error recovery
+	 * here. */
+	debug_tcp("Error sending PDU, errno=%d\n", rc);
+	iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+	return rc;
 }
 
 /**
- * iscsi_sendhdr - send PDU Header via tcp_sendpage()
- * @conn: iscsi connection
- * @buf: buffer to write from
- * @datalen: lenght of data to be sent after the header
- *
- * Notes:
- *	(Tx, Fast Path)
- **/
+ * iscsi_tcp_xmit_qlen - return the number of bytes queued for xmit
+ */
 static inline int
-iscsi_sendhdr(struct iscsi_conn *conn, struct iscsi_buf *buf, int datalen)
+iscsi_tcp_xmit_qlen(struct iscsi_conn *conn)
 {
-	int flags = 0; /* MSG_DONTWAIT; */
-	int res, size;
-
-	size = buf->sg.length - buf->sent;
-	BUG_ON(buf->sent + size > buf->sg.length);
-	if (buf->sent + size != buf->sg.length || datalen)
-		flags |= MSG_MORE;
-
-	res = iscsi_send(conn, buf, size, flags);
-	debug_tcp("sendhdr %d bytes, sent %d res %d\n", size, buf->sent, res);
-	if (res >= 0) {
-		if (size != res)
-			return -EAGAIN;
-		return 0;
-	}
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
+	struct iscsi_segment *segment = &tcp_conn->out.segment;
 
-	return res;
+	return segment->total_copied - segment->total_size;
 }
 
-/**
- * iscsi_sendpage - send one page of iSCSI Data-Out.
- * @conn: iscsi connection
- * @buf: buffer to write from
- * @count: remaining data
- * @sent: number of bytes sent
- *
- * Notes:
- *	(Tx, Fast Path)
- **/
 static inline int
-iscsi_sendpage(struct iscsi_conn *conn, struct iscsi_buf *buf,
-	       int *count, int *sent)
+iscsi_tcp_flush(struct iscsi_conn *conn)
 {
-	int flags = 0; /* MSG_DONTWAIT; */
-	int res, size;
-
-	size = buf->sg.length - buf->sent;
-	BUG_ON(buf->sent + size > buf->sg.length);
-	if (size > *count)
-		size = *count;
-	if (buf->sent + size != buf->sg.length || *count != size)
-		flags |= MSG_MORE;
-
-	res = iscsi_send(conn, buf, size, flags);
-	debug_tcp("sendpage: %d bytes, sent %d left %d sent %d res %d\n",
-		  size, buf->sent, *count, *sent, res);
-	if (res >= 0) {
-		*count -= res;
-		*sent += res;
-		if (size != res)
+	int rc;
+
+	while (iscsi_tcp_xmit_qlen(conn)) {
+		rc = iscsi_xmit(conn);
+		if (rc == 0)
 			return -EAGAIN;
-		return 0;
+		if (rc < 0)
+			return rc;
 	}
 
-	return res;
+	return 0;
 }
 
-static inline void
-iscsi_data_digest_init(struct iscsi_tcp_conn *tcp_conn,
-		      struct iscsi_tcp_cmd_task *tcp_ctask)
+/*
+ * This is called when we're done sending the header.
+ * Simply copy the data_segment to the send segment, and return.
+ */
+static int
+iscsi_tcp_send_hdr_done(struct iscsi_tcp_conn *tcp_conn,
+			struct iscsi_segment *segment)
 {
-	crypto_hash_init(&tcp_conn->tx_hash);
-	tcp_ctask->digest_count = 4;
+	tcp_conn->out.segment = tcp_conn->out.data_segment;
+	debug_tcp("Header done. Next segment size %u total_size %u\n",
+		  tcp_conn->out.segment.size, tcp_conn->out.segment.total_size);
+	return 0;
+}
+
+static void
+iscsi_tcp_send_hdr_prep(struct iscsi_conn *conn, void *hdr, size_t hdrlen)
+{
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
+
+	debug_tcp("%s(%p%s)\n", __FUNCTION__, tcp_conn,
+			conn->hdrdgst_en? ", digest enabled" : "");
+
+	/* Clear the data segment - needs to be filled in by the
+	 * caller using iscsi_tcp_send_data_prep() */
+	memset(&tcp_conn->out.data_segment, 0, sizeof(struct iscsi_segment));
+
+	/* If header digest is enabled, compute the CRC and
+	 * place the digest into the same buffer. We make
+	 * sure that both iscsi_tcp_ctask and mtask have
+	 * sufficient room.
+	 */
+	if (conn->hdrdgst_en) {
+		iscsi_tcp_dgst_header(&tcp_conn->tx_hash, hdr, hdrlen,
+				      hdr + hdrlen);
+		hdrlen += ISCSI_DIGEST_SIZE;
+	}
+
+	/* Remember header pointer for later, when we need
+	 * to decide whether there's a payload to go along
+	 * with the header. */
+	tcp_conn->out.hdr = hdr;
+
+	iscsi_segment_init_linear(&tcp_conn->out.segment, hdr, hdrlen,
+				iscsi_tcp_send_hdr_done, NULL);
+}
+
+/*
+ * Prepare the send buffer for the payload data.
+ * Padding and checksumming will all be taken care
+ * of by the iscsi_segment routines.
+ */
+static int
+iscsi_tcp_send_data_prep(struct iscsi_conn *conn, struct scatterlist *sg,
+			 unsigned int count, unsigned int offset,
+			 unsigned int len)
+{
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
+	struct hash_desc *tx_hash = NULL;
+	unsigned int hdr_spec_len;
+
+	debug_tcp("%s(%p, offset=%d, datalen=%d%s)\n", __FUNCTION__,
+			tcp_conn, offset, len,
+			conn->datadgst_en? ", digest enabled" : "");
+
+	/* Make sure the datalen matches what the caller
+	   said he would send. */
+	hdr_spec_len = ntoh24(tcp_conn->out.hdr->dlength);
+	WARN_ON(iscsi_padded(len) != iscsi_padded(hdr_spec_len));
+
+	if (conn->datadgst_en)
+		tx_hash = &tcp_conn->tx_hash;
+
+	return iscsi_segment_seek_sg(&tcp_conn->out.data_segment,
+				   sg, count, offset, len,
+				   NULL, tx_hash);
+}
+
+static void
+iscsi_tcp_send_linear_data_prepare(struct iscsi_conn *conn, void *data,
+				   size_t len)
+{
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
+	struct hash_desc *tx_hash = NULL;
+	unsigned int hdr_spec_len;
+
+	debug_tcp("%s(%p, datalen=%d%s)\n", __FUNCTION__, tcp_conn, len,
+		  conn->datadgst_en? ", digest enabled" : "");
+
+	/* Make sure the datalen matches what the caller
+	   said he would send. */
+	hdr_spec_len = ntoh24(tcp_conn->out.hdr->dlength);
+	WARN_ON(iscsi_padded(len) != iscsi_padded(hdr_spec_len));
+
+	if (conn->datadgst_en)
+		tx_hash = &tcp_conn->tx_hash;
+
+	iscsi_segment_init_linear(&tcp_conn->out.data_segment,
+				data, len, NULL, tx_hash);
 }
 
 /**
@@ -1206,12 +1263,17 @@ iscsi_data_digest_init(struct iscsi_tcp_conn *tcp_conn,
  *
  *	Called under connection lock.
  **/
-static void
+static int
 iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
-			struct iscsi_r2t_info *r2t, int left)
+			struct iscsi_r2t_info *r2t)
 {
 	struct iscsi_data *hdr;
-	int new_offset;
+	int new_offset, left;
+
+	BUG_ON(r2t->data_length - r2t->sent < 0);
+	left = r2t->data_length - r2t->sent;
+	if (left == 0)
+		return 0;
 
 	hdr = &r2t->dtask.hdr;
 	memset(hdr, 0, sizeof(struct iscsi_data));
@@ -1232,43 +1294,46 @@ iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 		r2t->data_count = left;
 		hdr->flags = ISCSI_FLAG_CMD_FINAL;
 	}
-	conn->dataout_pdus_cnt++;
-
-	iscsi_buf_init_iov(&r2t->headbuf, (char*)hdr,
-			   sizeof(struct iscsi_hdr));
-
-	if (iscsi_buf_left(&r2t->sendbuf))
-		return;
-
-	iscsi_buf_init_sg(&r2t->sendbuf, r2t->sg);
-	r2t->sg += 1;
-}
-
-static void iscsi_set_padding(struct iscsi_tcp_cmd_task *tcp_ctask,
-			      unsigned long len)
-{
-	tcp_ctask->pad_count = len & (ISCSI_PAD_LEN - 1);
-	if (!tcp_ctask->pad_count)
-		return;
 
-	tcp_ctask->pad_count = ISCSI_PAD_LEN - tcp_ctask->pad_count;
-	debug_scsi("write padding %d bytes\n", tcp_ctask->pad_count);
-	tcp_ctask->xmstate |= XMSTATE_W_PAD;
+	conn->dataout_pdus_cnt++;
+	return 1;
 }
 
 /**
- * iscsi_tcp_cmd_init - Initialize iSCSI SCSI_READ or SCSI_WRITE commands
+ * iscsi_tcp_ctask - Initialize iSCSI SCSI_READ or SCSI_WRITE commands
  * @conn: iscsi connection
  * @ctask: scsi command task
  * @sc: scsi command
  **/
-static void
-iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask)
+static int
+iscsi_tcp_ctask_init(struct iscsi_cmd_task *ctask)
 {
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
+	struct iscsi_conn *conn = ctask->conn;
+	struct scsi_cmnd *sc = ctask->sc;
+	int err;
 
 	BUG_ON(__kfifo_len(tcp_ctask->r2tqueue));
-	tcp_ctask->xmstate = XMSTATE_CMD_HDR_INIT;
+	tcp_ctask->sent = 0;
+	tcp_ctask->exp_datasn = 0;
+
+	/* Prepare PDU, optionally w/ immediate data */
+	debug_scsi("ctask deq [cid %d itt 0x%x imm %d unsol %d]\n",
+		    conn->id, ctask->itt, ctask->imm_count,
+		    ctask->unsol_count);
+	iscsi_tcp_send_hdr_prep(conn, ctask->hdr, ctask->hdr_len);
+
+	if (!ctask->imm_count)
+		return 0;
+
+	/* If we have immediate data, attach a payload */
+	err = iscsi_tcp_send_data_prep(conn, scsi_sglist(sc), scsi_sg_count(sc),
+				       0, ctask->imm_count);
+	if (err)
+		return err;
+	tcp_ctask->sent += ctask->imm_count;
+	ctask->imm_count = 0;
+	return 0;
 }
 
 /**
@@ -1280,71 +1345,17 @@ iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask)
  *	The function can return -EAGAIN in which case caller must
  *	call it again later, or recover. '0' return code means successful
  *	xmit.
- *
- *	Management xmit state machine consists of these states:
- *		XMSTATE_IMM_HDR_INIT	- calculate digest of PDU Header
- *		XMSTATE_IMM_HDR 	- PDU Header xmit in progress
- *		XMSTATE_IMM_DATA 	- PDU Data xmit in progress
- *		XMSTATE_IDLE		- management PDU is done
  **/
 static int
 iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 {
-	struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data;
 	int rc;
 
-	debug_scsi("mtask deq [cid %d state %x itt 0x%x]\n",
-		conn->id, tcp_mtask->xmstate, mtask->itt);
-
-	if (tcp_mtask->xmstate & XMSTATE_IMM_HDR_INIT) {
-		iscsi_buf_init_iov(&tcp_mtask->headbuf, (char*)mtask->hdr,
-				   sizeof(struct iscsi_hdr));
-
-		if (mtask->data_count) {
-			tcp_mtask->xmstate |= XMSTATE_IMM_DATA;
-			iscsi_buf_init_iov(&tcp_mtask->sendbuf,
-					   (char*)mtask->data,
-					   mtask->data_count);
-		}
-
-		if (conn->c_stage != ISCSI_CONN_INITIAL_STAGE &&
-		    conn->stop_stage != STOP_CONN_RECOVER &&
-		    conn->hdrdgst_en)
-			iscsi_hdr_digest(conn, &tcp_mtask->headbuf,
-					(u8*)tcp_mtask->hdrext);
-
-		tcp_mtask->sent = 0;
-		tcp_mtask->xmstate &= ~XMSTATE_IMM_HDR_INIT;
-		tcp_mtask->xmstate |= XMSTATE_IMM_HDR;
-	}
-
-	if (tcp_mtask->xmstate & XMSTATE_IMM_HDR) {
-		rc = iscsi_sendhdr(conn, &tcp_mtask->headbuf,
-				   mtask->data_count);
-		if (rc)
-			return rc;
-		tcp_mtask->xmstate &= ~XMSTATE_IMM_HDR;
-	}
-
-	if (tcp_mtask->xmstate & XMSTATE_IMM_DATA) {
-		BUG_ON(!mtask->data_count);
-		tcp_mtask->xmstate &= ~XMSTATE_IMM_DATA;
-		/* FIXME: implement.
-		 * Virtual buffer could be spreaded across multiple pages...
-		 */
-		do {
-			int rc;
-
-			rc = iscsi_sendpage(conn, &tcp_mtask->sendbuf,
-					&mtask->data_count, &tcp_mtask->sent);
-			if (rc) {
-				tcp_mtask->xmstate |= XMSTATE_IMM_DATA;
-				return rc;
-			}
-		} while (mtask->data_count);
-	}
+	/* Flush any pending data first. */
+	rc = iscsi_tcp_flush(conn);
+	if (rc < 0)
+		return rc;
 
-	BUG_ON(tcp_mtask->xmstate != XMSTATE_IDLE);
 	if (mtask->hdr->itt == RESERVED_ITT) {
 		struct iscsi_session *session = conn->session;
 
@@ -1352,411 +1363,112 @@ iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 		iscsi_free_mgmt_task(conn, mtask);
 		spin_unlock_bh(&session->lock);
 	}
+
 	return 0;
 }
 
+/*
+ * iscsi_tcp_ctask_xmit - xmit normal PDU task
+ * @conn: iscsi connection
+ * @ctask: iscsi command task
+ *
+ * We're expected to return 0 when everything was transmitted succesfully,
+ * -EAGAIN if there's still data in the queue, or != 0 for any other kind
+ * of error.
+ */
 static int
-iscsi_send_cmd_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
+iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 {
-	struct scsi_cmnd *sc = ctask->sc;
 	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
+	struct scsi_cmnd *sc = ctask->sc;
 	int rc = 0;
 
-	if (tcp_ctask->xmstate & XMSTATE_CMD_HDR_INIT) {
-		tcp_ctask->sent = 0;
-		tcp_ctask->sg_count = 0;
-		tcp_ctask->exp_datasn = 0;
-
-		if (sc->sc_data_direction == DMA_TO_DEVICE) {
-			struct scatterlist *sg = scsi_sglist(sc);
-
-			iscsi_buf_init_sg(&tcp_ctask->sendbuf, sg);
-			tcp_ctask->sg = sg + 1;
-			tcp_ctask->bad_sg = sg + scsi_sg_count(sc);
-
-			debug_scsi("cmd [itt 0x%x total %d imm_data %d "
-				   "unsol count %d, unsol offset %d]\n",
-				   ctask->itt, scsi_bufflen(sc),
-				   ctask->imm_count, ctask->unsol_count,
-				   ctask->unsol_offset);
-		}
-
-		iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)ctask->hdr,
-				  ctask->hdr_len);
-
-		if (conn->hdrdgst_en)
-			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
-					 iscsi_next_hdr(ctask));
-		tcp_ctask->xmstate &= ~XMSTATE_CMD_HDR_INIT;
-		tcp_ctask->xmstate |= XMSTATE_CMD_HDR_XMIT;
-	}
-
-	if (tcp_ctask->xmstate & XMSTATE_CMD_HDR_XMIT) {
-		rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->imm_count);
-		if (rc)
-			return rc;
-		tcp_ctask->xmstate &= ~XMSTATE_CMD_HDR_XMIT;
-
-		if (sc->sc_data_direction != DMA_TO_DEVICE)
-			return 0;
-
-		if (ctask->imm_count) {
-			tcp_ctask->xmstate |= XMSTATE_IMM_DATA;
-			iscsi_set_padding(tcp_ctask, ctask->imm_count);
-
-			if (ctask->conn->datadgst_en) {
-				iscsi_data_digest_init(ctask->conn->dd_data,
-						       tcp_ctask);
-				tcp_ctask->immdigest = 0;
-			}
-		}
-
-		if (ctask->unsol_count)
-			tcp_ctask->xmstate |=
-					XMSTATE_UNS_HDR | XMSTATE_UNS_INIT;
-	}
-	return rc;
-}
-
-static int
-iscsi_send_padding(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	int sent = 0, rc;
-
-	if (tcp_ctask->xmstate & XMSTATE_W_PAD) {
-		iscsi_buf_init_iov(&tcp_ctask->sendbuf, (char*)&tcp_ctask->pad,
-				   tcp_ctask->pad_count);
-		if (conn->datadgst_en)
-			crypto_hash_update(&tcp_conn->tx_hash,
-					   &tcp_ctask->sendbuf.sg,
-					   tcp_ctask->sendbuf.sg.length);
-	} else if (!(tcp_ctask->xmstate & XMSTATE_W_RESEND_PAD))
-		return 0;
-
-	tcp_ctask->xmstate &= ~XMSTATE_W_PAD;
-	tcp_ctask->xmstate &= ~XMSTATE_W_RESEND_PAD;
-	debug_scsi("sending %d pad bytes for itt 0x%x\n",
-		   tcp_ctask->pad_count, ctask->itt);
-	rc = iscsi_sendpage(conn, &tcp_ctask->sendbuf, &tcp_ctask->pad_count,
-			   &sent);
-	if (rc) {
-		debug_scsi("padding send failed %d\n", rc);
-		tcp_ctask->xmstate |= XMSTATE_W_RESEND_PAD;
-	}
-	return rc;
-}
-
-static int
-iscsi_send_digest(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
-			struct iscsi_buf *buf, uint32_t *digest)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask;
-	struct iscsi_tcp_conn *tcp_conn;
-	int rc, sent = 0;
-
-	if (!conn->datadgst_en)
-		return 0;
-
-	tcp_ctask = ctask->dd_data;
-	tcp_conn = conn->dd_data;
-
-	if (!(tcp_ctask->xmstate & XMSTATE_W_RESEND_DATA_DIGEST)) {
-		crypto_hash_final(&tcp_conn->tx_hash, (u8*)digest);
-		iscsi_buf_init_iov(buf, (char*)digest, 4);
-	}
-	tcp_ctask->xmstate &= ~XMSTATE_W_RESEND_DATA_DIGEST;
-
-	rc = iscsi_sendpage(conn, buf, &tcp_ctask->digest_count, &sent);
-	if (!rc)
-		debug_scsi("sent digest 0x%x for itt 0x%x\n", *digest,
-			  ctask->itt);
-	else {
-		debug_scsi("sending digest 0x%x failed for itt 0x%x!\n",
-			  *digest, ctask->itt);
-		tcp_ctask->xmstate |= XMSTATE_W_RESEND_DATA_DIGEST;
-	}
-	return rc;
-}
-
-static int
-iscsi_send_data(struct iscsi_cmd_task *ctask, struct iscsi_buf *sendbuf,
-		struct scatterlist **sg, int *sent, int *count,
-		struct iscsi_buf *digestbuf, uint32_t *digest)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct iscsi_conn *conn = ctask->conn;
-	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
-	int rc, buf_sent, offset;
-
-	while (*count) {
-		buf_sent = 0;
-		offset = sendbuf->sent;
-
-		rc = iscsi_sendpage(conn, sendbuf, count, &buf_sent);
-		*sent = *sent + buf_sent;
-		if (buf_sent && conn->datadgst_en)
-			partial_sg_digest_update(&tcp_conn->tx_hash,
-				&sendbuf->sg, sendbuf->sg.offset + offset,
-				buf_sent);
-		if (!iscsi_buf_left(sendbuf) && *sg != tcp_ctask->bad_sg) {
-			iscsi_buf_init_sg(sendbuf, *sg);
-			*sg = *sg + 1;
-		}
-
-		if (rc)
-			return rc;
-	}
-
-	rc = iscsi_send_padding(conn, ctask);
-	if (rc)
+flush:
+	/* Flush any pending data first. */
+	rc = iscsi_tcp_flush(conn);
+	if (rc < 0)
 		return rc;
 
-	return iscsi_send_digest(conn, ctask, digestbuf, digest);
-}
-
-static int
-iscsi_send_unsol_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct iscsi_data_task *dtask;
-	int rc;
-
-	tcp_ctask->xmstate |= XMSTATE_UNS_DATA;
-	if (tcp_ctask->xmstate & XMSTATE_UNS_INIT) {
-		dtask = &tcp_ctask->unsol_dtask;
-
-		iscsi_prep_unsolicit_data_pdu(ctask, &dtask->hdr);
-		iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)&dtask->hdr,
-				   sizeof(struct iscsi_hdr));
-		if (conn->hdrdgst_en)
-			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
-					(u8*)dtask->hdrext);
-
-		tcp_ctask->xmstate &= ~XMSTATE_UNS_INIT;
-		iscsi_set_padding(tcp_ctask, ctask->data_count);
-	}
-
-	rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->data_count);
-	if (rc) {
-		tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA;
-		tcp_ctask->xmstate |= XMSTATE_UNS_HDR;
-		return rc;
-	}
+	/* Are we done already? */
+	if (sc->sc_data_direction != DMA_TO_DEVICE)
+		return 0;
 
-	if (conn->datadgst_en) {
-		dtask = &tcp_ctask->unsol_dtask;
-		iscsi_data_digest_init(ctask->conn->dd_data, tcp_ctask);
-		dtask->digest = 0;
-	}
+	if (ctask->unsol_count != 0) {
+		struct iscsi_data *hdr = &tcp_ctask->unsol_dtask.hdr;
 
-	debug_scsi("uns dout [itt 0x%x dlen %d sent %d]\n",
-		   ctask->itt, ctask->unsol_count, tcp_ctask->sent);
-	return 0;
-}
+		/* Prepare a header for the unsolicited PDU.
+		 * The amount of data we want to send will be
+		 * in ctask->data_count.
+		 * FIXME: return the data count instead.
+		 */
+		iscsi_prep_unsolicit_data_pdu(ctask, hdr);
 
-static int
-iscsi_send_unsol_pdu(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	int rc;
+		debug_tcp("unsol dout [itt 0x%x doff %d dlen %d]\n",
+				ctask->itt, tcp_ctask->sent, ctask->data_count);
 
-	if (tcp_ctask->xmstate & XMSTATE_UNS_HDR) {
-		BUG_ON(!ctask->unsol_count);
-		tcp_ctask->xmstate &= ~XMSTATE_UNS_HDR;
-send_hdr:
-		rc = iscsi_send_unsol_hdr(conn, ctask);
+		iscsi_tcp_send_hdr_prep(conn, hdr, sizeof(*hdr));
+		rc = iscsi_tcp_send_data_prep(conn, scsi_sglist(sc),
+					      scsi_sg_count(sc),
+					      tcp_ctask->sent,
+					      ctask->data_count);
 		if (rc)
-			return rc;
-	}
-
-	if (tcp_ctask->xmstate & XMSTATE_UNS_DATA) {
-		struct iscsi_data_task *dtask = &tcp_ctask->unsol_dtask;
-		int start = tcp_ctask->sent;
+			goto fail;
+		tcp_ctask->sent += ctask->data_count;
+		ctask->unsol_count -= ctask->data_count;
+		goto flush;
+	} else {
+		struct iscsi_session *session = conn->session;
+		struct iscsi_r2t_info *r2t;
 
-		rc = iscsi_send_data(ctask, &tcp_ctask->sendbuf, &tcp_ctask->sg,
-				     &tcp_ctask->sent, &ctask->data_count,
-				     &dtask->digestbuf, &dtask->digest);
-		ctask->unsol_count -= tcp_ctask->sent - start;
-		if (rc)
-			return rc;
-		tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA;
-		/*
-		 * Done with the Data-Out. Next, check if we need
-		 * to send another unsolicited Data-Out.
+		/* All unsolicited PDUs sent. Check for solicited PDUs.
 		 */
-		if (ctask->unsol_count) {
-			debug_scsi("sending more uns\n");
-			tcp_ctask->xmstate |= XMSTATE_UNS_INIT;
-			goto send_hdr;
+		spin_lock_bh(&session->lock);
+		r2t = tcp_ctask->r2t;
+		if (r2t != NULL) {
+			/* Continue with this R2T? */
+			if (!iscsi_solicit_data_cont(conn, ctask, r2t)) {
+				debug_scsi("  done with r2t %p\n", r2t);
+
+				__kfifo_put(tcp_ctask->r2tpool.queue,
+					    (void*)&r2t, sizeof(void*));
+				tcp_ctask->r2t = r2t = NULL;
+			}
 		}
-	}
-	return 0;
-}
 
-static int iscsi_send_sol_pdu(struct iscsi_conn *conn,
-			      struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct iscsi_session *session = conn->session;
-	struct iscsi_r2t_info *r2t;
-	struct iscsi_data_task *dtask;
-	int left, rc;
-
-	if (tcp_ctask->xmstate & XMSTATE_SOL_HDR_INIT) {
-		if (!tcp_ctask->r2t) {
-			spin_lock_bh(&session->lock);
+		if (r2t == NULL) {
 			__kfifo_get(tcp_ctask->r2tqueue, (void*)&tcp_ctask->r2t,
 				    sizeof(void*));
-			spin_unlock_bh(&session->lock);
+			r2t = tcp_ctask->r2t;
 		}
-send_hdr:
-		r2t = tcp_ctask->r2t;
-		dtask = &r2t->dtask;
-
-		if (conn->hdrdgst_en)
-			iscsi_hdr_digest(conn, &r2t->headbuf,
-					(u8*)dtask->hdrext);
-		tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR_INIT;
-		tcp_ctask->xmstate |= XMSTATE_SOL_HDR;
-	}
-
-	if (tcp_ctask->xmstate & XMSTATE_SOL_HDR) {
-		r2t = tcp_ctask->r2t;
-		dtask = &r2t->dtask;
-
-		rc = iscsi_sendhdr(conn, &r2t->headbuf, r2t->data_count);
-		if (rc)
-			return rc;
-		tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR;
-		tcp_ctask->xmstate |= XMSTATE_SOL_DATA;
+		spin_unlock_bh(&session->lock);
 
-		if (conn->datadgst_en) {
-			iscsi_data_digest_init(conn->dd_data, tcp_ctask);
-			dtask->digest = 0;
+		/* Waiting for more R2Ts to arrive. */
+		if (r2t == NULL) {
+			debug_tcp("no R2Ts yet\n");
+			return 0;
 		}
 
-		iscsi_set_padding(tcp_ctask, r2t->data_count);
-		debug_scsi("sol dout [dsn %d itt 0x%x dlen %d sent %d]\n",
-			r2t->solicit_datasn - 1, ctask->itt, r2t->data_count,
-			r2t->sent);
-	}
+		debug_scsi("sol dout %p [dsn %d itt 0x%x doff %d dlen %d]\n",
+			r2t, r2t->solicit_datasn - 1, ctask->itt,
+			r2t->data_offset + r2t->sent, r2t->data_count);
 
-	if (tcp_ctask->xmstate & XMSTATE_SOL_DATA) {
-		r2t = tcp_ctask->r2t;
-		dtask = &r2t->dtask;
+		iscsi_tcp_send_hdr_prep(conn, &r2t->dtask.hdr,
+					sizeof(struct iscsi_hdr));
 
-		rc = iscsi_send_data(ctask, &r2t->sendbuf, &r2t->sg,
-				     &r2t->sent, &r2t->data_count,
-				     &dtask->digestbuf, &dtask->digest);
+		rc = iscsi_tcp_send_data_prep(conn, scsi_sglist(sc),
+					      scsi_sg_count(sc),
+					      r2t->data_offset + r2t->sent,
+					      r2t->data_count);
 		if (rc)
-			return rc;
-		tcp_ctask->xmstate &= ~XMSTATE_SOL_DATA;
-
-		/*
-		 * Done with this Data-Out. Next, check if we have
-		 * to send another Data-Out for this R2T.
-		 */
-		BUG_ON(r2t->data_length - r2t->sent < 0);
-		left = r2t->data_length - r2t->sent;
-		if (left) {
-			iscsi_solicit_data_cont(conn, ctask, r2t, left);
-			goto send_hdr;
-		}
-
-		/*
-		 * Done with this R2T. Check if there are more
-		 * outstanding R2Ts ready to be processed.
-		 */
-		spin_lock_bh(&session->lock);
-		tcp_ctask->r2t = NULL;
-		__kfifo_put(tcp_ctask->r2tpool.queue, (void*)&r2t,
-			    sizeof(void*));
-		if (__kfifo_get(tcp_ctask->r2tqueue, (void*)&r2t,
-				sizeof(void*))) {
-			tcp_ctask->r2t = r2t;
-			spin_unlock_bh(&session->lock);
-			goto send_hdr;
-		}
-		spin_unlock_bh(&session->lock);
+			goto fail;
+		tcp_ctask->sent += r2t->data_count;
+		r2t->sent += r2t->data_count;
+		goto flush;
 	}
 	return 0;
-}
-
-/**
- * iscsi_tcp_ctask_xmit - xmit normal PDU task
- * @conn: iscsi connection
- * @ctask: iscsi command task
- *
- * Notes:
- *	The function can return -EAGAIN in which case caller must
- *	call it again later, or recover. '0' return code means successful
- *	xmit.
- *	The function is devided to logical helpers (above) for the different
- *	xmit stages.
- *
- *iscsi_send_cmd_hdr()
- *	XMSTATE_CMD_HDR_INIT - prepare Header and Data buffers Calculate
- *	                       Header Digest
- *	XMSTATE_CMD_HDR_XMIT - Transmit header in progress
- *
- *iscsi_send_padding
- *	XMSTATE_W_PAD        - Prepare and send pading
- *	XMSTATE_W_RESEND_PAD - retry send pading
- *
- *iscsi_send_digest
- *	XMSTATE_W_RESEND_DATA_DIGEST - Finalize and send Data Digest
- *	XMSTATE_W_RESEND_DATA_DIGEST - retry sending digest
- *
- *iscsi_send_unsol_hdr
- *	XMSTATE_UNS_INIT     - prepare un-solicit data header and digest
- *	XMSTATE_UNS_HDR      - send un-solicit header
- *
- *iscsi_send_unsol_pdu
- *	XMSTATE_UNS_DATA     - send un-solicit data in progress
- *
- *iscsi_send_sol_pdu
- *	XMSTATE_SOL_HDR_INIT - solicit data header and digest initialize
- *	XMSTATE_SOL_HDR      - send solicit header
- *	XMSTATE_SOL_DATA     - send solicit data
- *
- *iscsi_tcp_ctask_xmit
- *	XMSTATE_IMM_DATA     - xmit managment data (??)
- **/
-static int
-iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	int rc = 0;
-
-	debug_scsi("ctask deq [cid %d xmstate %x itt 0x%x]\n",
-		conn->id, tcp_ctask->xmstate, ctask->itt);
-
-	rc = iscsi_send_cmd_hdr(conn, ctask);
-	if (rc)
-		return rc;
-	if (ctask->sc->sc_data_direction != DMA_TO_DEVICE)
-		return 0;
-
-	if (tcp_ctask->xmstate & XMSTATE_IMM_DATA) {
-		rc = iscsi_send_data(ctask, &tcp_ctask->sendbuf, &tcp_ctask->sg,
-				     &tcp_ctask->sent, &ctask->imm_count,
-				     &tcp_ctask->immbuf, &tcp_ctask->immdigest);
-		if (rc)
-			return rc;
-		tcp_ctask->xmstate &= ~XMSTATE_IMM_DATA;
-	}
-
-	rc = iscsi_send_unsol_pdu(conn, ctask);
-	if (rc)
-		return rc;
-
-	rc = iscsi_send_sol_pdu(conn, ctask);
-	if (rc)
-		return rc;
-
-	return rc;
+fail:
+	iscsi_conn_failure(conn, rc);
+	return -EIO;
 }
 
 static struct iscsi_cls_conn *
@@ -1970,10 +1682,17 @@ free_socket:
 
 /* called with host lock */
 static void
-iscsi_tcp_mgmt_init(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
+iscsi_tcp_mtask_init(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask)
 {
-	struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data;
-	tcp_mtask->xmstate = XMSTATE_IMM_HDR_INIT;
+	debug_scsi("mtask deq [cid %d itt 0x%x]\n", conn->id, mtask->itt);
+
+	/* Prepare PDU, optionally w/ immediate data */
+	iscsi_tcp_send_hdr_prep(conn, mtask->hdr, sizeof(*mtask->hdr));
+
+	/* If we have immediate data, attach a payload */
+	if (mtask->data_count)
+		iscsi_tcp_send_linear_data_prepare(conn, mtask->data,
+						   mtask->data_count);
 }
 
 static int
@@ -2177,7 +1896,7 @@ iscsi_tcp_session_create(struct iscsi_transport *iscsit,
 		struct iscsi_mgmt_task *mtask = session->mgmt_cmds[cmd_i];
 		struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data;
 
-		mtask->hdr = &tcp_mtask->hdr;
+		mtask->hdr = (struct iscsi_hdr *) &tcp_mtask->hdr;
 	}
 
 	if (iscsi_r2tpool_alloc(class_to_transport_session(cls_session)))
@@ -2274,8 +1993,8 @@ static struct iscsi_transport iscsi_tcp_transport = {
 	/* IO */
 	.send_pdu		= iscsi_conn_send_pdu,
 	.get_stats		= iscsi_conn_get_stats,
-	.init_cmd_task		= iscsi_tcp_cmd_init,
-	.init_mgmt_task		= iscsi_tcp_mgmt_init,
+	.init_cmd_task		= iscsi_tcp_ctask_init,
+	.init_mgmt_task		= iscsi_tcp_mtask_init,
 	.xmit_cmd_task		= iscsi_tcp_ctask_xmit,
 	.xmit_mgmt_task		= iscsi_tcp_mtask_xmit,
 	.cleanup_cmd_task	= iscsi_tcp_cleanup_ctask,
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index d49d87611e82..893cd2e1701e 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -24,35 +24,18 @@
 
 #include <scsi/libiscsi.h>
 
-/* xmit state machine */
-#define XMSTATE_IDLE			0x0
-#define XMSTATE_CMD_HDR_INIT		0x1
-#define XMSTATE_CMD_HDR_XMIT		0x2
-#define XMSTATE_IMM_HDR			0x4
-#define XMSTATE_IMM_DATA		0x8
-#define XMSTATE_UNS_INIT		0x10
-#define XMSTATE_UNS_HDR			0x20
-#define XMSTATE_UNS_DATA		0x40
-#define XMSTATE_SOL_HDR			0x80
-#define XMSTATE_SOL_DATA		0x100
-#define XMSTATE_W_PAD			0x200
-#define XMSTATE_W_RESEND_PAD		0x400
-#define XMSTATE_W_RESEND_DATA_DIGEST	0x800
-#define XMSTATE_IMM_HDR_INIT		0x1000
-#define XMSTATE_SOL_HDR_INIT		0x2000
-
 #define ISCSI_SG_TABLESIZE		SG_ALL
 #define ISCSI_TCP_MAX_CMD_LEN		16
 
 struct crypto_hash;
 struct socket;
 struct iscsi_tcp_conn;
-struct iscsi_chunk;
+struct iscsi_segment;
 
-typedef int iscsi_chunk_done_fn_t(struct iscsi_tcp_conn *,
-				  struct iscsi_chunk *);
+typedef int iscsi_segment_done_fn_t(struct iscsi_tcp_conn *,
+				    struct iscsi_segment *);
 
-struct iscsi_chunk {
+struct iscsi_segment {
 	unsigned char		*data;
 	unsigned int		size;
 	unsigned int		copied;
@@ -67,16 +50,14 @@ struct iscsi_chunk {
 	struct scatterlist	*sg;
 	void			*sg_mapped;
 	unsigned int		sg_offset;
-	unsigned int		sg_index;
-	unsigned int		sg_count;
 
-	iscsi_chunk_done_fn_t	*done;
+	iscsi_segment_done_fn_t	*done;
 };
 
 /* Socket connection recieve helper */
 struct iscsi_tcp_recv {
 	struct iscsi_hdr	*hdr;
-	struct iscsi_chunk	chunk;
+	struct iscsi_segment	segment;
 
 	/* Allocate buffer for BHS + AHS */
 	uint32_t		hdr_buf[64];
@@ -88,11 +69,8 @@ struct iscsi_tcp_recv {
 /* Socket connection send helper */
 struct iscsi_tcp_send {
 	struct iscsi_hdr	*hdr;
-	struct iscsi_chunk	chunk;
-	struct iscsi_chunk	data_chunk;
-
-	/* Allocate buffer for BHS + AHS */
-	uint32_t		hdr_buf[64];
+	struct iscsi_segment	segment;
+	struct iscsi_segment	data_segment;
 };
 
 struct iscsi_tcp_conn {
@@ -118,29 +96,19 @@ struct iscsi_tcp_conn {
 	uint32_t		sendpage_failures_cnt;
 	uint32_t		discontiguous_hdr_cnt;
 
-	ssize_t (*sendpage)(struct socket *, struct page *, int, size_t, int);
-};
+	int			error;
 
-struct iscsi_buf {
-	struct scatterlist	sg;
-	unsigned int		sent;
-	char			use_sendmsg;
+	ssize_t (*sendpage)(struct socket *, struct page *, int, size_t, int);
 };
 
 struct iscsi_data_task {
 	struct iscsi_data	hdr;			/* PDU */
 	char			hdrext[ISCSI_DIGEST_SIZE];/* Header-Digest */
-	struct iscsi_buf	digestbuf;		/* digest buffer */
-	uint32_t		digest;			/* data digest */
 };
 
 struct iscsi_tcp_mgmt_task {
 	struct iscsi_hdr	hdr;
 	char			hdrext[ISCSI_DIGEST_SIZE]; /* Header-Digest */
-	int			xmstate;	/* mgmt xmit progress */
-	struct iscsi_buf	headbuf;	/* header buffer */
-	struct iscsi_buf	sendbuf;	/* in progress buffer */
-	int			sent;
 };
 
 struct iscsi_r2t_info {
@@ -148,13 +116,10 @@ struct iscsi_r2t_info {
 	__be32			exp_statsn;	/* copied from R2T */
 	uint32_t		data_length;	/* copied from R2T */
 	uint32_t		data_offset;	/* copied from R2T */
-	struct iscsi_buf	headbuf;	/* Data-Out Header Buffer */
-	struct iscsi_buf	sendbuf;	/* Data-Out in progress buffer*/
 	int			sent;		/* R2T sequence progress */
 	int			data_count;	/* DATA-Out payload progress */
-	struct scatterlist	*sg;		/* per-R2T SG list */
 	int			solicit_datasn;
-	struct iscsi_data_task   dtask;        /* which data task */
+	struct iscsi_data_task	dtask;		/* Data-Out header buf */
 };
 
 struct iscsi_tcp_cmd_task {
@@ -163,24 +128,14 @@ struct iscsi_tcp_cmd_task {
 		char			hdrextbuf[ISCSI_MAX_AHS_SIZE +
 		                                  ISCSI_DIGEST_SIZE];
 	} hdr;
-	char			pad[ISCSI_PAD_LEN];
-	int			pad_count;		/* padded bytes */
-	struct iscsi_buf	headbuf;		/* header buf (xmit) */
-	struct iscsi_buf	sendbuf;		/* in progress buffer*/
-	int			xmstate;		/* xmit xtate machine */
+
 	int			sent;
-	struct scatterlist	*sg;			/* per-cmd SG list  */
-	struct scatterlist	*bad_sg;		/* assert statement */
-	int			sg_count;		/* SG's to process  */
-	uint32_t		exp_datasn;		/* expected target's R2TSN/DataSN */
+	uint32_t		exp_datasn;	/* expected target's R2TSN/DataSN */
 	int			data_offset;
-	struct iscsi_r2t_info	*r2t;			/* in progress R2T    */
+	struct iscsi_r2t_info	*r2t;		/* in progress R2T    */
 	struct iscsi_pool	r2tpool;
 	struct kfifo		*r2tqueue;
-	int			digest_count;
-	uint32_t		immdigest;		/* for imm data */
-	struct iscsi_buf	immbuf;			/* for imm data digest */
-	struct iscsi_data_task	unsol_dtask;	/* unsol data task */
+	struct iscsi_data_task	unsol_dtask;	/* Data-Out header buf */
 };
 
 #endif /* ISCSI_H */
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index b0bc8c3b0ecb..f15df8d75fd8 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -156,20 +156,19 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 	rc = iscsi_add_hdr(ctask, sizeof(*hdr));
 	if (rc)
 		return rc;
-        hdr->opcode = ISCSI_OP_SCSI_CMD;
-        hdr->flags = ISCSI_ATTR_SIMPLE;
-        int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun);
-        hdr->itt = build_itt(ctask->itt, conn->id, session->age);
-        hdr->data_length = cpu_to_be32(scsi_bufflen(sc));
-        hdr->cmdsn = cpu_to_be32(session->cmdsn);
-        session->cmdsn++;
-        hdr->exp_statsn = cpu_to_be32(conn->exp_statsn);
-        memcpy(hdr->cdb, sc->cmnd, sc->cmd_len);
+	hdr->opcode = ISCSI_OP_SCSI_CMD;
+	hdr->flags = ISCSI_ATTR_SIMPLE;
+	int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun);
+	hdr->itt = build_itt(ctask->itt, conn->id, session->age);
+	hdr->data_length = cpu_to_be32(scsi_bufflen(sc));
+	hdr->cmdsn = cpu_to_be32(session->cmdsn);
+	session->cmdsn++;
+	hdr->exp_statsn = cpu_to_be32(conn->exp_statsn);
+	memcpy(hdr->cdb, sc->cmnd, sc->cmd_len);
 	if (sc->cmd_len < MAX_COMMAND_SIZE)
 		memset(&hdr->cdb[sc->cmd_len], 0,
 			MAX_COMMAND_SIZE - sc->cmd_len);
 
-	ctask->data_count = 0;
 	ctask->imm_count = 0;
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
 		hdr->flags |= ISCSI_FLAG_CMD_WRITE;
@@ -198,9 +197,9 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 			else
 				ctask->imm_count = min(scsi_bufflen(sc),
 							conn->max_xmit_dlength);
-			hton24(ctask->hdr->dlength, ctask->imm_count);
+			hton24(hdr->dlength, ctask->imm_count);
 		} else
-			zero_data(ctask->hdr->dlength);
+			zero_data(hdr->dlength);
 
 		if (!session->initial_r2t_en) {
 			ctask->unsol_count = min((session->first_burst),
@@ -210,7 +209,7 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 
 		if (!ctask->unsol_count)
 			/* No unsolicit Data-Out's */
-			ctask->hdr->flags |= ISCSI_FLAG_CMD_FINAL;
+			hdr->flags |= ISCSI_FLAG_CMD_FINAL;
 	} else {
 		hdr->flags |= ISCSI_FLAG_CMD_FINAL;
 		zero_data(hdr->dlength);
@@ -228,13 +227,15 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 	WARN_ON(hdrlength >= 256);
 	hdr->hlength = hdrlength & 0xFF;
 
-	conn->scsicmd_pdus_cnt++;
+	if (conn->session->tt->init_cmd_task(conn->ctask))
+		return EIO;
 
-        debug_scsi("iscsi prep [%s cid %d sc %p cdb 0x%x itt 0x%x len %d "
+	conn->scsicmd_pdus_cnt++;
+	debug_scsi("iscsi prep [%s cid %d sc %p cdb 0x%x itt 0x%x len %d "
 		"cmdsn %d win %d]\n",
-                sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read",
+		sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read",
 		conn->id, sc, sc->cmnd[0], ctask->itt, scsi_bufflen(sc),
-                session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1);
+		session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1);
 	return 0;
 }
 
@@ -927,7 +928,7 @@ check_mgmt:
 			fail_command(conn, conn->ctask, DID_ABORT << 16);
 			continue;
 		}
-		conn->session->tt->init_cmd_task(conn->ctask);
+
 		conn->ctask->state = ISCSI_TASK_RUNNING;
 		list_move_tail(conn->xmitqueue.next, &conn->run_list);
 		rc = iscsi_xmit_ctask(conn);
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index 093b4036f8db..404f11d331d6 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -118,7 +118,7 @@ struct iscsi_transport {
 			 char *data, uint32_t data_size);
 	void (*get_stats) (struct iscsi_cls_conn *conn,
 			   struct iscsi_stats *stats);
-	void (*init_cmd_task) (struct iscsi_cmd_task *ctask);
+	int (*init_cmd_task) (struct iscsi_cmd_task *ctask);
 	void (*init_mgmt_task) (struct iscsi_conn *conn,
 				struct iscsi_mgmt_task *mtask);
 	int (*xmit_cmd_task) (struct iscsi_conn *conn,
-- 
cgit v1.2.3


From b98e66fa0b687149dc3b26179096dc542c7d5001 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Fri, 28 Dec 2007 16:35:17 -0600
Subject: [SCSI] libsas: add host SMP processing

This adds support for host side SMP processing, via a separate
SMP interpreter file.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libsas/Kconfig        |   9 ++
 drivers/scsi/libsas/Makefile       |   1 +
 drivers/scsi/libsas/sas_expander.c |   8 +-
 drivers/scsi/libsas/sas_host_smp.c | 274 +++++++++++++++++++++++++++++++++++++
 drivers/scsi/libsas/sas_internal.h |  14 ++
 include/scsi/scsi_transport_sas.h  |   8 +-
 6 files changed, 305 insertions(+), 9 deletions(-)
 create mode 100644 drivers/scsi/libsas/sas_host_smp.c

(limited to 'include')

diff --git a/drivers/scsi/libsas/Kconfig b/drivers/scsi/libsas/Kconfig
index c01a40d321d4..18f33cd54411 100644
--- a/drivers/scsi/libsas/Kconfig
+++ b/drivers/scsi/libsas/Kconfig
@@ -38,6 +38,15 @@ config SCSI_SAS_ATA
 		Builds in ATA support into libsas.  Will necessitate
 		the loading of libata along with libsas.
 
+config SCSI_SAS_HOST_SMP
+	bool "Support for SMP interpretation for SAS hosts"
+	default y
+	depends on SCSI_SAS_LIBSAS
+	help
+		Allows sas hosts to receive SMP frames.  Selecting this
+		option builds an SMP interpreter into libsas.  Say
+		N here if you want to save the few kb this consumes.
+
 config SCSI_SAS_LIBSAS_DEBUG
 	bool "Compile the SAS Domain Transport Attributes in debug mode"
 	default y
diff --git a/drivers/scsi/libsas/Makefile b/drivers/scsi/libsas/Makefile
index fd387b91856e..60d6e93dd949 100644
--- a/drivers/scsi/libsas/Makefile
+++ b/drivers/scsi/libsas/Makefile
@@ -35,3 +35,4 @@ libsas-y +=  sas_init.o     \
 		sas_expander.o \
 		sas_scsi_host.o
 libsas-$(CONFIG_SCSI_SAS_ATA) +=	sas_ata.o
+libsas-$(CONFIG_SCSI_SAS_HOST_SMP) +=	sas_host_smp.o
\ No newline at end of file
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index 27674fe468f5..76555b1ddfd5 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -1896,11 +1896,9 @@ int sas_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
 	}
 
 	/* no rphy means no smp target support (ie aic94xx host) */
-	if (!rphy) {
-		printk("%s: can we send a smp request to a host?\n",
-		       __FUNCTION__);
-		return -EINVAL;
-	}
+	if (!rphy)
+		return sas_smp_host_handler(shost, req, rsp);
+
 	type = rphy->identify.device_type;
 
 	if (type != SAS_EDGE_EXPANDER_DEVICE &&
diff --git a/drivers/scsi/libsas/sas_host_smp.c b/drivers/scsi/libsas/sas_host_smp.c
new file mode 100644
index 000000000000..16f93123271e
--- /dev/null
+++ b/drivers/scsi/libsas/sas_host_smp.c
@@ -0,0 +1,274 @@
+/*
+ * Serial Attached SCSI (SAS) Expander discovery and configuration
+ *
+ * Copyright (C) 2007 James E.J. Bottomley
+ *		<James.Bottomley@HansenPartnership.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 only.
+ */
+#include <linux/scatterlist.h>
+#include <linux/blkdev.h>
+
+#include "sas_internal.h"
+
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+static void sas_host_smp_discover(struct sas_ha_struct *sas_ha, u8 *resp_data,
+				  u8 phy_id)
+{
+	struct sas_phy *phy;
+	struct sas_rphy *rphy;
+
+	if (phy_id >= sas_ha->num_phys) {
+		resp_data[2] = SMP_RESP_NO_PHY;
+		return;
+	}
+	resp_data[2] = SMP_RESP_FUNC_ACC;
+
+	phy = sas_ha->sas_phy[phy_id]->phy;
+	resp_data[9] = phy_id;
+	resp_data[13] = phy->negotiated_linkrate;
+	memcpy(resp_data + 16, sas_ha->sas_addr, SAS_ADDR_SIZE);
+	memcpy(resp_data + 24, sas_ha->sas_phy[phy_id]->attached_sas_addr,
+	       SAS_ADDR_SIZE);
+	resp_data[40] = (phy->minimum_linkrate << 4) |
+		phy->minimum_linkrate_hw;
+	resp_data[41] = (phy->maximum_linkrate << 4) |
+		phy->maximum_linkrate_hw;
+
+	if (!sas_ha->sas_phy[phy_id]->port ||
+	    !sas_ha->sas_phy[phy_id]->port->port_dev)
+		return;
+
+	rphy = sas_ha->sas_phy[phy_id]->port->port_dev->rphy;
+	resp_data[12] = rphy->identify.device_type << 4;
+	resp_data[14] = rphy->identify.initiator_port_protocols;
+	resp_data[15] = rphy->identify.target_port_protocols;
+}
+
+static void sas_report_phy_sata(struct sas_ha_struct *sas_ha, u8 *resp_data,
+				u8 phy_id)
+{
+	struct sas_rphy *rphy;
+	struct dev_to_host_fis *fis;
+	int i;
+
+	if (phy_id >= sas_ha->num_phys) {
+		resp_data[2] = SMP_RESP_NO_PHY;
+		return;
+	}
+
+	resp_data[2] = SMP_RESP_PHY_NO_SATA;
+
+	if (!sas_ha->sas_phy[phy_id]->port)
+		return;
+
+	rphy = sas_ha->sas_phy[phy_id]->port->port_dev->rphy;
+	fis = (struct dev_to_host_fis *)
+		sas_ha->sas_phy[phy_id]->port->port_dev->frame_rcvd;
+	if (rphy->identify.target_port_protocols != SAS_PROTOCOL_SATA)
+		return;
+
+	resp_data[2] = SMP_RESP_FUNC_ACC;
+	resp_data[9] = phy_id;
+	memcpy(resp_data + 16, sas_ha->sas_phy[phy_id]->attached_sas_addr,
+	       SAS_ADDR_SIZE);
+
+	/* check to see if we have a valid d2h fis */
+	if (fis->fis_type != 0x34)
+		return;
+
+	/* the d2h fis is required by the standard to be in LE format */
+	for (i = 0; i < 20; i += 4) {
+		u8 *dst = resp_data + 24 + i, *src =
+			&sas_ha->sas_phy[phy_id]->port->port_dev->frame_rcvd[i];
+		dst[0] = src[3];
+		dst[1] = src[2];
+		dst[2] = src[1];
+		dst[3] = src[0];
+	}
+}
+
+static void sas_phy_control(struct sas_ha_struct *sas_ha, u8 phy_id,
+			    u8 phy_op, enum sas_linkrate min,
+			    enum sas_linkrate max, u8 *resp_data)
+{
+	struct sas_internal *i =
+		to_sas_internal(sas_ha->core.shost->transportt);
+	struct sas_phy_linkrates rates;
+
+	if (phy_id >= sas_ha->num_phys) {
+		resp_data[2] = SMP_RESP_NO_PHY;
+		return;
+	}
+	switch (phy_op) {
+	case PHY_FUNC_NOP:
+	case PHY_FUNC_LINK_RESET:
+	case PHY_FUNC_HARD_RESET:
+	case PHY_FUNC_DISABLE:
+	case PHY_FUNC_CLEAR_ERROR_LOG:
+	case PHY_FUNC_CLEAR_AFFIL:
+	case PHY_FUNC_TX_SATA_PS_SIGNAL:
+		break;
+
+	default:
+		resp_data[2] = SMP_RESP_PHY_UNK_OP;
+		return;
+	}
+
+	rates.minimum_linkrate = min;
+	rates.maximum_linkrate = max;
+
+	if (i->dft->lldd_control_phy(sas_ha->sas_phy[phy_id], phy_op, &rates))
+		resp_data[2] = SMP_RESP_FUNC_FAILED;
+	else
+		resp_data[2] = SMP_RESP_FUNC_ACC;
+}
+
+int sas_smp_host_handler(struct Scsi_Host *shost, struct request *req,
+			 struct request *rsp)
+{
+	u8 *req_data = NULL, *resp_data = NULL, *buf;
+	struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost);
+	int error = -EINVAL, resp_data_len = rsp->data_len;
+
+	/* eight is the minimum size for request and response frames */
+	if (req->data_len < 8 || rsp->data_len < 8)
+		goto out;
+
+	if (bio_offset(req->bio) + req->data_len > PAGE_SIZE ||
+	    bio_offset(rsp->bio) + rsp->data_len > PAGE_SIZE) {
+		shost_printk(KERN_ERR, shost,
+			"SMP request/response frame crosses page boundary");
+		goto out;
+	}
+
+	req_data = kzalloc(req->data_len, GFP_KERNEL);
+
+	/* make sure frame can always be built ... we copy
+	 * back only the requested length */
+	resp_data = kzalloc(max(rsp->data_len, 128U), GFP_KERNEL);
+
+	if (!req_data || !resp_data) {
+		error = -ENOMEM;
+		goto out;
+	}
+
+	local_irq_disable();
+	buf = kmap_atomic(bio_page(req->bio), KM_USER0) + bio_offset(req->bio);
+	memcpy(req_data, buf, req->data_len);
+	kunmap_atomic(buf - bio_offset(req->bio), KM_USER0);
+	local_irq_enable();
+
+	if (req_data[0] != SMP_REQUEST)
+		goto out;
+
+	/* always succeeds ... even if we can't process the request
+	 * the result is in the response frame */
+	error = 0;
+
+	/* set up default don't know response */
+	resp_data[0] = SMP_RESPONSE;
+	resp_data[1] = req_data[1];
+	resp_data[2] = SMP_RESP_FUNC_UNK;
+
+	switch (req_data[1]) {
+	case SMP_REPORT_GENERAL:
+		req->data_len -= 8;
+		resp_data_len -= 32;
+		resp_data[2] = SMP_RESP_FUNC_ACC;
+		resp_data[9] = sas_ha->num_phys;
+		break;
+
+	case SMP_REPORT_MANUF_INFO:
+		req->data_len -= 8;
+		resp_data_len -= 64;
+		resp_data[2] = SMP_RESP_FUNC_ACC;
+		memcpy(resp_data + 12, shost->hostt->name,
+		       SAS_EXPANDER_VENDOR_ID_LEN);
+		memcpy(resp_data + 20, "libsas virt phy",
+		       SAS_EXPANDER_PRODUCT_ID_LEN);
+		break;
+
+	case SMP_READ_GPIO_REG:
+		/* FIXME: need GPIO support in the transport class */
+		break;
+
+	case SMP_DISCOVER:
+		req->data_len =- 16;
+		if (req->data_len < 0) {
+			req->data_len = 0;
+			error = -EINVAL;
+			goto out;
+		}
+		resp_data_len -= 56;
+		sas_host_smp_discover(sas_ha, resp_data, req_data[9]);
+		break;
+
+	case SMP_REPORT_PHY_ERR_LOG:
+		/* FIXME: could implement this with additional
+		 * libsas callbacks providing the HW supports it */
+		break;
+
+	case SMP_REPORT_PHY_SATA:
+		req->data_len =- 16;
+		if (req->data_len < 0) {
+			req->data_len = 0;
+			error = -EINVAL;
+			goto out;
+		}
+		resp_data_len -= 60;
+		sas_report_phy_sata(sas_ha, resp_data, req_data[9]);
+		break;
+
+	case SMP_REPORT_ROUTE_INFO:
+		/* Can't implement; hosts have no routes */
+		break;
+
+	case SMP_WRITE_GPIO_REG:
+		/* FIXME: need GPIO support in the transport class */
+		break;
+
+	case SMP_CONF_ROUTE_INFO:
+		/* Can't implement; hosts have no routes */
+		break;
+
+	case SMP_PHY_CONTROL:
+		req->data_len =- 44;
+		if (req->data_len < 0) {
+			req->data_len = 0;
+			error = -EINVAL;
+			goto out;
+		}
+		resp_data_len -= 8;
+		sas_phy_control(sas_ha, req_data[9], req_data[10],
+				req_data[32] >> 4, req_data[33] >> 4,
+				resp_data);
+		break;
+
+	case SMP_PHY_TEST_FUNCTION:
+		/* FIXME: should this be implemented? */
+		break;
+
+	default:
+		/* probably a 2.0 function */
+		break;
+	}
+
+	local_irq_disable();
+	buf = kmap_atomic(bio_page(rsp->bio), KM_USER0) + bio_offset(rsp->bio);
+	memcpy(buf, resp_data, rsp->data_len);
+	flush_kernel_dcache_page(bio_page(rsp->bio));
+	kunmap_atomic(buf - bio_offset(rsp->bio), KM_USER0);
+	local_irq_enable();
+	rsp->data_len = resp_data_len;
+
+ out:
+	kfree(req_data);
+	kfree(resp_data);
+	return error;
+}
diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h
index baa0666ffa80..b4f9368f116a 100644
--- a/drivers/scsi/libsas/sas_internal.h
+++ b/drivers/scsi/libsas/sas_internal.h
@@ -80,6 +80,20 @@ struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy);
 
 void sas_hae_reset(struct work_struct *work);
 
+#ifdef CONFIG_SCSI_SAS_HOST_SMP
+extern int sas_smp_host_handler(struct Scsi_Host *shost, struct request *req,
+				struct request *rsp);
+#else
+static inline int sas_smp_host_handler(struct Scsi_Host *shost,
+				       struct request *req,
+				       struct request *rsp)
+{
+	shost_printk(KERN_ERR, shost,
+		"Cannot send SMP to a sas host (not enabled in CONFIG)\n");
+	return -EINVAL;
+}
+#endif
+
 static inline void sas_queue_event(int event, spinlock_t *lock,
 				   unsigned long *pending,
 				   struct work_struct *work,
diff --git a/include/scsi/scsi_transport_sas.h b/include/scsi/scsi_transport_sas.h
index 007d92904207..09125fa95b93 100644
--- a/include/scsi/scsi_transport_sas.h
+++ b/include/scsi/scsi_transport_sas.h
@@ -11,10 +11,10 @@ struct sas_rphy;
 struct request;
 
 enum sas_device_type {
-	SAS_PHY_UNUSED,
-	SAS_END_DEVICE,
-	SAS_EDGE_EXPANDER_DEVICE,
-	SAS_FANOUT_EXPANDER_DEVICE,
+	SAS_PHY_UNUSED = 0,
+	SAS_END_DEVICE = 1,
+	SAS_EDGE_EXPANDER_DEVICE = 2,
+	SAS_FANOUT_EXPANDER_DEVICE = 3,
 };
 
 static inline int sas_protocol_ata(enum sas_protocol proto)
-- 
cgit v1.2.3


From 32e8ae36b8f80372015b88b63c4358a376c9af0f Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Sun, 30 Dec 2007 12:37:31 -0600
Subject: [SCSI] libsas: don't use made up error codes

This is bad for two reasons:

     1. If they're returned to outside applications, no-one knows what
        they mean.
     2. Eventually they'll clash with the ever expanding standard error
        codes.

The problem error code in question is ETASK.  I've replaced this by
ECOMM (communications error on send) a network error code that seems to
most closely relay what ETASK meant.

Acked-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libsas/sas_ata.c      | 2 +-
 drivers/scsi/libsas/sas_expander.c | 2 +-
 include/scsi/libsas.h              | 2 --
 3 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 0829b55c64d2..adc47d481f97 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -500,7 +500,7 @@ static int sas_execute_task(struct sas_task *task, void *buffer, int size,
 			goto ex_err;
 		}
 		wait_for_completion(&task->completion);
-		res = -ETASK;
+		res = -ECOMM;
 		if (task->task_state_flags & SAS_TASK_STATE_ABORTED) {
 			int res2;
 			SAS_DPRINTK("task aborted, flags:0x%x\n",
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index 8aeaad95242c..aefd865a5788 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -96,7 +96,7 @@ static int smp_execute_task(struct domain_device *dev, void *req, int req_size,
 		}
 
 		wait_for_completion(&task->completion);
-		res = -ETASK;
+		res = -ECOMM;
 		if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) {
 			SAS_DPRINTK("smp task timed out or aborted\n");
 			i->dft->lldd_abort_task(task);
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 93248cd48373..a075f130303c 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -91,8 +91,6 @@ enum discover_event {
 
 /* ---------- Expander Devices ---------- */
 
-#define ETASK 0xFA
-
 #define to_dom_device(_obj) container_of(_obj, struct domain_device, dev_obj)
 #define to_dev_attr(_attr)  container_of(_attr, struct domain_dev_attribute,\
                                          attr)
-- 
cgit v1.2.3


From 11c3e689f1c3a73e3af7b0ea767b1b0626da8033 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Mon, 31 Dec 2007 16:37:00 -0600
Subject: [SCSI] block: Introduce new blk_queue_update_dma_alignment interface

The purpose of this is to allow stacked alignment settings, with the
ultimate queue alignment being set to the largest alignment requirement
in the stack.

The reason for this is so that the SCSI mid-layer can relax the default
alignment requirements (which are basically causing a lot of superfluous
copying to go on in the SG_IO interface) while allowing transports,
devices or HBAs to add stricter limits if they need them.

Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 block/ll_rw_blk.c      | 24 ++++++++++++++++++++++++
 include/linux/blkdev.h |  1 +
 2 files changed, 25 insertions(+)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 8b919940b2ab..14af36c5cdb2 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -759,6 +759,30 @@ void blk_queue_dma_alignment(struct request_queue *q, int mask)
 
 EXPORT_SYMBOL(blk_queue_dma_alignment);
 
+/**
+ * blk_queue_update_dma_alignment - update dma length and memory alignment
+ * @q:     the request queue for the device
+ * @mask:  alignment mask
+ *
+ * description:
+ *    update required memory and length aligment for direct dma transactions.
+ *    If the requested alignment is larger than the current alignment, then
+ *    the current queue alignment is updated to the new value, otherwise it
+ *    is left alone.  The design of this is to allow multiple objects
+ *    (driver, device, transport etc) to set their respective
+ *    alignments without having them interfere.
+ *
+ **/
+void blk_queue_update_dma_alignment(struct request_queue *q, int mask)
+{
+	BUG_ON(mask > PAGE_SIZE);
+
+	if (mask > q->dma_alignment)
+		q->dma_alignment = mask;
+}
+
+EXPORT_SYMBOL(blk_queue_update_dma_alignment);
+
 /**
  * blk_queue_find_tag - find a request by its tag and queue
  * @q:	 The request queue for the device
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d18ee67b40f8..81e99e516302 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -766,6 +766,7 @@ extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
 extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
 extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
 extern void blk_queue_dma_alignment(struct request_queue *, int);
+extern void blk_queue_update_dma_alignment(struct request_queue *, int);
 extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
 extern int blk_queue_ordered(struct request_queue *, unsigned, prepare_flush_fn *);
-- 
cgit v1.2.3


From ae8d4ee7ff429136c8b482c3b38ed994c021d3fc Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Sun, 4 Nov 2007 22:05:49 -0500
Subject: libata: Disable ATA8-ACS proposed Trusted Computing features by
 default

Historically word 48 in the identify data was used to mean 32bit I/O
was supported for VLB IDE etc. ATA8 reassigns this word to the Trusted
Computing Group, where it is used for TCG features. This means that
an ATA8 TCG drive is going to trigger 32bit I/O on some systems which
will be funny.

Anyway we need to sort this out ready for ATA8 so:
- Reorder the ata.h header a bit so the ata_version function occurs early
  in it
- Make dword_io check the ATA version
- Add an ATA8 version checking TCG presence test

While we are at it the current drafts have a flaw where it may not be
possible to disable TCG features at boot (and opt out of the trusted
model) as TCG intends because it relies on presence of a different
optional feature (DCS). Handle this in software by refusing the TCG
commands if libata.allow_tpm is not set. (We must make it possible
as some environments such as proprietary VDR devices will doubtless
want to use it to lock up content)

Finally as with CPRM print a warning so that the user knows they may
not be able to full access and use the device.

Signed-off-by: Alan Cox <alan@redhat.com>
---
 drivers/ata/libata-core.c | 12 +++++++++++-
 drivers/ata/libata-scsi.c | 18 ++++++++++++++++++
 drivers/ata/libata.h      |  1 +
 include/linux/ata.h       | 22 +++++++++++++++++++++-
 4 files changed, 51 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 6380726f7538..9b7f3c477730 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -119,6 +119,10 @@ int libata_noacpi = 0;
 module_param_named(noacpi, libata_noacpi, int, 0444);
 MODULE_PARM_DESC(noacpi, "Disables the use of ACPI in probe/suspend/resume when set");
 
+int libata_allow_tpm = 0;
+module_param_named(allow_tpm, libata_allow_tpm, int, 0444);
+MODULE_PARM_DESC(allow_tpm, "Permit the use of TPM commands");
+
 MODULE_AUTHOR("Jeff Garzik");
 MODULE_DESCRIPTION("Library module for ATA devices");
 MODULE_LICENSE("GPL");
@@ -2161,8 +2165,14 @@ int ata_dev_configure(struct ata_device *dev)
 					       "supports DRM functions and may "
 					       "not be fully accessable.\n");
 			snprintf(revbuf, 7, "CFA");
-		} else
+		} else {
 			snprintf(revbuf, 7, "ATA-%d", ata_id_major_version(id));
+			/* Warn the user if the device has TPM extensions */
+			if (ata_id_has_tpm(id))
+				ata_dev_printk(dev, KERN_WARNING,
+					       "supports DRM functions and may "
+					       "not be fully accessable.\n");
+		}
 
 		dev->n_sectors = ata_id_n_sectors(id);
 
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 14daf4848f09..f802dbce41ae 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2690,6 +2690,24 @@ static unsigned int ata_scsi_pass_thru(struct ata_queued_cmd *qc)
 	if ((tf->protocol = ata_scsi_map_proto(cdb[1])) == ATA_PROT_UNKNOWN)
 		goto invalid_fld;
 
+	/*
+	 * Filter TPM commands by default. These provide an
+	 * essentially uncontrolled encrypted "back door" between
+	 * applications and the disk. Set libata.allow_tpm=1 if you
+	 * have a real reason for wanting to use them. This ensures
+	 * that installed software cannot easily mess stuff up without
+	 * user intent. DVR type users will probably ship with this enabled
+	 * for movie content management.
+	 *
+	 * Note that for ATA8 we can issue a DCS change and DCS freeze lock
+	 * for this and should do in future but that it is not sufficient as
+	 * DCS is an optional feature set. Thus we also do the software filter
+	 * so that we comply with the TC consortium stated goal that the user
+	 * can turn off TC features of their system.
+	 */
+	if (tf->command >= 0x5C && tf->command <= 0x5F && !libata_allow_tpm)
+		goto invalid_fld;
+
 	/* We may not issue DMA commands if no DMA mode is set */
 	if (tf->protocol == ATA_PROT_DMA && dev->dma_mode == 0)
 		goto invalid_fld;
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index bbe59c2fd1e2..048e26cfb339 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -60,6 +60,7 @@ extern int atapi_dmadir;
 extern int atapi_passthru16;
 extern int libata_fua;
 extern int libata_noacpi;
+extern int libata_allow_tpm;
 extern struct ata_queued_cmd *ata_qc_new_init(struct ata_device *dev);
 extern int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev,
 			   u64 block, u32 n_block, unsigned int tf_flags,
diff --git a/include/linux/ata.h b/include/linux/ata.h
index e672e80202a8..3fbe6d7784ab 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -379,7 +379,6 @@ struct ata_taskfile {
 #define ata_id_has_ncq(id)	((id)[76] & (1 << 8))
 #define ata_id_queue_depth(id)	(((id)[75] & 0x1f) + 1)
 #define ata_id_removeable(id)	((id)[0] & (1 << 7))
-#define ata_id_has_dword_io(id)	((id)[48] & (1 << 0))
 #define ata_id_has_atapi_AN(id)	\
 	( (((id)[76] != 0x0000) && ((id)[76] != 0xffff)) && \
 	  ((id)[78] & (1 << 5)) )
@@ -415,6 +414,7 @@ static inline bool ata_id_has_dipm(const u16 *id)
 	return val & (1 << 3);
 }
 
+
 static inline int ata_id_has_fua(const u16 *id)
 {
 	if ((id[84] & 0xC000) != 0x4000)
@@ -519,6 +519,26 @@ static inline int ata_id_is_sata(const u16 *id)
 	return ata_id_major_version(id) >= 5 && id[93] == 0;
 }
 
+static inline int ata_id_has_tpm(const u16 *id)
+{
+	/* The TPM bits are only valid on ATA8 */
+	if (ata_id_major_version(id) < 8)
+		return 0;
+	if ((id[48] & 0xC000) != 0x4000)
+		return 0;
+	return id[48] & (1 << 0);
+}
+
+static inline int ata_id_has_dword_io(const u16 *id)
+{
+	/* ATA 8 reuses this flag for "trusted" computing */
+	if (ata_id_major_version(id) > 7)
+		return 0;
+	if (id[48] & (1 << 0))
+		return 1;
+	return 0;
+}
+
 static inline int ata_id_current_chs_valid(const u16 *id)
 {
 	/* For ATA-1 devices, if the INITIALIZE DEVICE PARAMETERS command
-- 
cgit v1.2.3


From f20ded38aa54b92dd0af32578b8916d0aa2d9e05 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:28:52 +0900
Subject: libata: rearrange ATA_DFLAG_*

Area for DFLAGs which are cleared on INIT is full.  Extend it by 8
bits.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/linux/libata.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index 124033cb5e9b..ca347b018649 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -143,10 +143,10 @@ enum {
 	ATA_DFLAG_NCQ_OFF	= (1 << 13), /* device limited to non-NCQ mode */
 	ATA_DFLAG_SPUNDOWN	= (1 << 14), /* XXX: for spindown_compat */
 	ATA_DFLAG_SLEEPING	= (1 << 15), /* device is sleeping */
-	ATA_DFLAG_INIT_MASK	= (1 << 16) - 1,
+	ATA_DFLAG_INIT_MASK	= (1 << 24) - 1,
 
-	ATA_DFLAG_DETACH	= (1 << 16),
-	ATA_DFLAG_DETACHED	= (1 << 17),
+	ATA_DFLAG_DETACH	= (1 << 24),
+	ATA_DFLAG_DETACHED	= (1 << 25),
 
 	ATA_DEV_UNKNOWN		= 0,	/* unknown device */
 	ATA_DEV_ATA		= 1,	/* ATA device */
-- 
cgit v1.2.3


From 405e66b38797875e80669eaf72d313dbb76533c3 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:28:53 +0900
Subject: libata: implement protocol tests

Implement protocol tests - ata_is_atapi(), ata_is_nodata(),
ata_is_pio(), ata_is_dma(), ata_is_ncq() and ata_is_data() and use
them to replace is_atapi_taskfile() and hard coded protocol tests.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ahci.c            |  2 +-
 drivers/ata/libata-core.c     | 36 +++++-----------------
 drivers/ata/sata_fsl.c        |  2 +-
 drivers/ata/sata_sil.c        | 10 +++---
 drivers/ata/sata_sil24.c      | 24 +++------------
 drivers/scsi/libsas/sas_ata.c |  2 +-
 include/linux/ata.h           | 71 ++++++++++++++++++++++++++++++++++++++-----
 7 files changed, 82 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index ef5e6b6e6e69..5eee91c73c90 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1511,7 +1511,7 @@ static void ahci_qc_prep(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ahci_port_priv *pp = ap->private_data;
-	int is_atapi = is_atapi_taskfile(&qc->tf);
+	int is_atapi = ata_is_atapi(qc->tf.protocol);
 	void *cmd_tbl;
 	u32 opts;
 	const u32 cmd_fis_len = 5; /* five dwords */
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index bb9e025522ba..8c7af2c8e8d4 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5358,7 +5358,7 @@ static inline int ata_hsm_ok_in_wq(struct ata_port *ap, struct ata_queued_cmd *q
 		    (qc->tf.flags & ATA_TFLAG_WRITE))
 		    return 1;
 
-		if (is_atapi_taskfile(&qc->tf) &&
+		if (ata_is_atapi(qc->tf.protocol) &&
 		    !(qc->dev->flags & ATA_DFLAG_CDB_INTR))
 			return 1;
 	}
@@ -5955,30 +5955,6 @@ int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
 	return nr_done;
 }
 
-static inline int ata_should_dma_map(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-
-	switch (qc->tf.protocol) {
-	case ATA_PROT_NCQ:
-	case ATA_PROT_DMA:
-	case ATA_PROT_ATAPI_DMA:
-		return 1;
-
-	case ATA_PROT_ATAPI:
-	case ATA_PROT_PIO:
-		if (ap->flags & ATA_FLAG_PIO_DMA)
-			return 1;
-
-		/* fall through */
-
-	default:
-		return 0;
-	}
-
-	/* never reached */
-}
-
 /**
  *	ata_qc_issue - issue taskfile to device
  *	@qc: command to issue to device
@@ -5995,6 +5971,7 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_link *link = qc->dev->link;
+	u8 prot = qc->tf.protocol;
 
 	/* Make sure only one non-NCQ command is outstanding.  The
 	 * check is skipped for old EH because it reuses active qc to
@@ -6002,7 +5979,7 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	 */
 	WARN_ON(ap->ops->error_handler && ata_tag_valid(link->active_tag));
 
-	if (qc->tf.protocol == ATA_PROT_NCQ) {
+	if (prot == ATA_PROT_NCQ) {
 		WARN_ON(link->sactive & (1 << qc->tag));
 
 		if (!link->sactive)
@@ -6018,7 +5995,8 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	qc->flags |= ATA_QCFLAG_ACTIVE;
 	ap->qc_active |= 1 << qc->tag;
 
-	if (ata_should_dma_map(qc)) {
+	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
+				 (ap->flags & ATA_FLAG_PIO_DMA))) {
 		if (qc->flags & ATA_QCFLAG_SG) {
 			if (ata_sg_setup(qc))
 				goto sg_err;
@@ -6217,8 +6195,8 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 		 */
 
 		/* Check the ATA_DFLAG_CDB_INTR flag is enough here.
-		 * The flag was turned on only for atapi devices.
-		 * No need to check is_atapi_taskfile(&qc->tf) again.
+		 * The flag was turned on only for atapi devices.  No
+		 * need to check ata_is_atapi(qc->tf.protocol) again.
 		 */
 		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
 			goto idle_irq;
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index d015b4adcfe0..a3c33f165427 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -417,7 +417,7 @@ static void sata_fsl_qc_prep(struct ata_queued_cmd *qc)
 	}
 
 	/* setup "ACMD - atapi command" in cmd. desc. if this is ATAPI cmd */
-	if (is_atapi_taskfile(&qc->tf)) {
+	if (ata_is_atapi(qc->tf.protocol)) {
 		desc_info |= ATAPI_CMD;
 		memset((void *)&cd->acmd, 0, 32);
 		memcpy((void *)&cd->acmd, qc->cdb, qc->dev->cdb_len);
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index f5119bf40c24..0b8191b52f97 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -416,15 +416,14 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 		 */
 
 		/* Check the ATA_DFLAG_CDB_INTR flag is enough here.
-		 * The flag was turned on only for atapi devices.
-		 * No need to check is_atapi_taskfile(&qc->tf) again.
+		 * The flag was turned on only for atapi devices.  No
+		 * need to check ata_is_atapi(qc->tf.protocol) again.
 		 */
 		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
 			goto err_hsm;
 		break;
 	case HSM_ST_LAST:
-		if (qc->tf.protocol == ATA_PROT_DMA ||
-		    qc->tf.protocol == ATA_PROT_ATAPI_DMA) {
+		if (ata_is_dma(qc->tf.protocol)) {
 			/* clear DMA-Start bit */
 			ap->ops->bmdma_stop(qc);
 
@@ -451,8 +450,7 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 	/* kick HSM in the ass */
 	ata_hsm_move(ap, qc, status, 0);
 
-	if (unlikely(qc->err_mask) && (qc->tf.protocol == ATA_PROT_DMA ||
-				       qc->tf.protocol == ATA_PROT_ATAPI_DMA))
+	if (unlikely(qc->err_mask) && ata_is_dma(qc->tf.protocol))
 		ata_ehi_push_desc(ehi, "BMDMA2 stat 0x%x", bmdma2);
 
 	return;
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 864c1c1b8511..fdd3ceac329b 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -852,9 +852,7 @@ static int sil24_qc_defer(struct ata_queued_cmd *qc)
 	 *   set.
 	 *
  	 */
-	int is_excl = (prot == ATA_PROT_ATAPI ||
-		       prot == ATA_PROT_ATAPI_NODATA ||
-		       prot == ATA_PROT_ATAPI_DMA ||
+	int is_excl = (ata_is_atapi(prot) ||
 		       (qc->flags & ATA_QCFLAG_RESULT_TF));
 
 	if (unlikely(ap->excl_link)) {
@@ -885,35 +883,21 @@ static void sil24_qc_prep(struct ata_queued_cmd *qc)
 
 	cb = &pp->cmd_block[sil24_tag(qc->tag)];
 
-	switch (qc->tf.protocol) {
-	case ATA_PROT_PIO:
-	case ATA_PROT_DMA:
-	case ATA_PROT_NCQ:
-	case ATA_PROT_NODATA:
+	if (!ata_is_atapi(qc->tf.protocol)) {
 		prb = &cb->ata.prb;
 		sge = cb->ata.sge;
-		break;
-
-	case ATA_PROT_ATAPI:
-	case ATA_PROT_ATAPI_DMA:
-	case ATA_PROT_ATAPI_NODATA:
+	} else {
 		prb = &cb->atapi.prb;
 		sge = cb->atapi.sge;
 		memset(cb->atapi.cdb, 0, 32);
 		memcpy(cb->atapi.cdb, qc->cdb, qc->dev->cdb_len);
 
-		if (qc->tf.protocol != ATA_PROT_ATAPI_NODATA) {
+		if (ata_is_data(qc->tf.protocol)) {
 			if (qc->tf.flags & ATA_TFLAG_WRITE)
 				ctrl = PRB_CTRL_PACKET_WRITE;
 			else
 				ctrl = PRB_CTRL_PACKET_READ;
 		}
-		break;
-
-	default:
-		prb = NULL;	/* shut up, gcc */
-		sge = NULL;
-		BUG();
 	}
 
 	prb->ctrl = cpu_to_le16(ctrl);
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 0829b55c64d2..831294de1d8d 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -176,7 +176,7 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
 
 	ata_tf_to_fis(&qc->tf, 1, 0, (u8*)&task->ata_task.fis);
 	task->uldd_task = qc;
-	if (is_atapi_taskfile(&qc->tf)) {
+	if (ata_is_atapi(qc->tf.protocol)) {
 		memcpy(task->ata_task.atapi_packet, qc->cdb, qc->dev->cdb_len);
 		task->total_xfer_len = qc->nbytes + qc->pad_len;
 		task->num_scatter = qc->pad_len ? qc->n_elem + 1 : qc->n_elem;
diff --git a/include/linux/ata.h b/include/linux/ata.h
index 3fbe6d7784ab..43fecf62773a 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -324,6 +324,13 @@ enum {
 	ATA_TFLAG_LBA		= (1 << 4), /* enable LBA */
 	ATA_TFLAG_FUA		= (1 << 5), /* enable FUA */
 	ATA_TFLAG_POLLING	= (1 << 6), /* set nIEN to 1 and use polling */
+
+	/* protocol flags */
+	ATA_PROT_FLAG_PIO	= (1 << 0), /* is PIO */
+	ATA_PROT_FLAG_DMA	= (1 << 1), /* is DMA */
+	ATA_PROT_FLAG_DATA	= ATA_PROT_FLAG_PIO | ATA_PROT_FLAG_DMA,
+	ATA_PROT_FLAG_NCQ	= (1 << 2), /* is NCQ */
+	ATA_PROT_FLAG_ATAPI	= (1 << 3), /* is ATAPI */
 };
 
 enum ata_tf_protocols {
@@ -373,6 +380,63 @@ struct ata_taskfile {
 	u8			command;	/* IO operation */
 };
 
+/*
+ * protocol tests
+ */
+static inline unsigned int ata_prot_flags(u8 prot)
+{
+	switch (prot) {
+	case ATA_PROT_NODATA:
+		return 0;
+	case ATA_PROT_PIO:
+		return ATA_PROT_FLAG_PIO;
+	case ATA_PROT_DMA:
+		return ATA_PROT_FLAG_DMA;
+	case ATA_PROT_NCQ:
+		return ATA_PROT_FLAG_DMA | ATA_PROT_FLAG_NCQ;
+	case ATA_PROT_ATAPI_NODATA:
+		return ATA_PROT_FLAG_ATAPI;
+	case ATA_PROT_ATAPI:
+		return ATA_PROT_FLAG_ATAPI | ATA_PROT_FLAG_PIO;
+	case ATA_PROT_ATAPI_DMA:
+		return ATA_PROT_FLAG_ATAPI | ATA_PROT_FLAG_DMA;
+	}
+	return 0;
+}
+
+static inline int ata_is_atapi(u8 prot)
+{
+	return ata_prot_flags(prot) & ATA_PROT_FLAG_ATAPI;
+}
+
+static inline int ata_is_nodata(u8 prot)
+{
+	return !(ata_prot_flags(prot) & ATA_PROT_FLAG_DATA);
+}
+
+static inline int ata_is_pio(u8 prot)
+{
+	return ata_prot_flags(prot) & ATA_PROT_FLAG_PIO;
+}
+
+static inline int ata_is_dma(u8 prot)
+{
+	return ata_prot_flags(prot) & ATA_PROT_FLAG_DMA;
+}
+
+static inline int ata_is_ncq(u8 prot)
+{
+	return ata_prot_flags(prot) & ATA_PROT_FLAG_NCQ;
+}
+
+static inline int ata_is_data(u8 prot)
+{
+	return ata_prot_flags(prot) & ATA_PROT_FLAG_DATA;
+}
+
+/*
+ * id tests
+ */
 #define ata_id_is_ata(id)	(((id)[0] & (1 << 15)) == 0)
 #define ata_id_has_lba(id)	((id)[49] & (1 << 9))
 #define ata_id_has_dma(id)	((id)[49] & (1 << 8))
@@ -594,13 +658,6 @@ static inline int atapi_command_packet_set(const u16 *dev_id)
 	return (dev_id[0] >> 8) & 0x1f;
 }
 
-static inline int is_atapi_taskfile(const struct ata_taskfile *tf)
-{
-	return (tf->protocol == ATA_PROT_ATAPI) ||
-	       (tf->protocol == ATA_PROT_ATAPI_NODATA) ||
-	       (tf->protocol == ATA_PROT_ATAPI_DMA);
-}
-
 static inline int is_multi_taskfile(struct ata_taskfile *tf)
 {
 	return (tf->command == ATA_CMD_READ_MULTI) ||
-- 
cgit v1.2.3


From 3884f7b0a8382b89d8ca5da23bd98e3e15fc805b Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:28:56 +0900
Subject: libata: clean up EH speed down implementation

Clean up EH speed down implementation.

* is_io boolean variable is replaced eflags.  is_io is ATA_EFLAG_IS_IO.

* Error categories now have names.

* Better comments.

* Reorder 5min and 10min rules in ata_eh_speed_down_verdict()

* Use local variable @link to cache @dev->link in ata_eh_speed_down()

These changes are to improve readability and ease further changes.
This patch doesn't introduce any behavior change.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-eh.c | 117 ++++++++++++++++++++++++++++++------------------
 include/linux/libata.h  |   2 +-
 2 files changed, 74 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index ebab75958900..b01ade102727 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -46,9 +46,20 @@
 #include "libata.h"
 
 enum {
+	/* speed down verdicts */
 	ATA_EH_SPDN_NCQ_OFF		= (1 << 0),
 	ATA_EH_SPDN_SPEED_DOWN		= (1 << 1),
 	ATA_EH_SPDN_FALLBACK_TO_PIO	= (1 << 2),
+
+	/* error flags */
+	ATA_EFLAG_IS_IO			= (1 << 0),
+
+	/* error categories */
+	ATA_ECAT_NONE			= 0,
+	ATA_ECAT_ATA_BUS		= 1,
+	ATA_ECAT_TOUT_HSM		= 2,
+	ATA_ECAT_UNK_DEV		= 3,
+	ATA_ECAT_NR			= 4,
 };
 
 /* Waiting in ->prereset can never be reliable.  It's sometimes nice
@@ -218,7 +229,7 @@ void ata_port_pbar_desc(struct ata_port *ap, int bar, ssize_t offset,
 
 #endif /* CONFIG_PCI */
 
-static void ata_ering_record(struct ata_ering *ering, int is_io,
+static void ata_ering_record(struct ata_ering *ering, unsigned int eflags,
 			     unsigned int err_mask)
 {
 	struct ata_ering_entry *ent;
@@ -229,7 +240,7 @@ static void ata_ering_record(struct ata_ering *ering, int is_io,
 	ering->cursor %= ATA_ERING_SIZE;
 
 	ent = &ering->ring[ering->cursor];
-	ent->is_io = is_io;
+	ent->eflags = eflags;
 	ent->err_mask = err_mask;
 	ent->timestamp = get_jiffies_64();
 }
@@ -1451,20 +1462,20 @@ static unsigned int ata_eh_analyze_tf(struct ata_queued_cmd *qc,
 	return action;
 }
 
-static int ata_eh_categorize_error(int is_io, unsigned int err_mask)
+static int ata_eh_categorize_error(unsigned int eflags, unsigned int err_mask)
 {
 	if (err_mask & AC_ERR_ATA_BUS)
-		return 1;
+		return ATA_ECAT_ATA_BUS;
 
 	if (err_mask & AC_ERR_TIMEOUT)
-		return 2;
+		return ATA_ECAT_TOUT_HSM;
 
-	if (is_io) {
+	if (eflags & ATA_EFLAG_IS_IO) {
 		if (err_mask & AC_ERR_HSM)
-			return 2;
+			return ATA_ECAT_TOUT_HSM;
 		if ((err_mask &
 		     (AC_ERR_DEV|AC_ERR_MEDIA|AC_ERR_INVALID)) == AC_ERR_DEV)
-			return 3;
+			return ATA_ECAT_UNK_DEV;
 	}
 
 	return 0;
@@ -1472,13 +1483,13 @@ static int ata_eh_categorize_error(int is_io, unsigned int err_mask)
 
 struct speed_down_verdict_arg {
 	u64 since;
-	int nr_errors[4];
+	int nr_errors[ATA_ECAT_NR];
 };
 
 static int speed_down_verdict_cb(struct ata_ering_entry *ent, void *void_arg)
 {
 	struct speed_down_verdict_arg *arg = void_arg;
-	int cat = ata_eh_categorize_error(ent->is_io, ent->err_mask);
+	int cat = ata_eh_categorize_error(ent->eflags, ent->err_mask);
 
 	if (ent->timestamp < arg->since)
 		return -1;
@@ -1495,22 +1506,33 @@ static int speed_down_verdict_cb(struct ata_ering_entry *ent, void *void_arg)
  *	whether NCQ needs to be turned off, transfer speed should be
  *	stepped down, or falling back to PIO is necessary.
  *
- *	Cat-1 is ATA_BUS error for any command.
+ *	ECAT_ATA_BUS	: ATA_BUS error for any command
+ *
+ *	ECAT_TOUT_HSM	: TIMEOUT for any command or HSM violation for
+ *			  IO commands
+ *
+ *	ECAT_UNK_DEV	: Unknown DEV error for IO commands
+ *
+ *	Verdicts are
  *
- *	Cat-2 is TIMEOUT for any command or HSM violation for known
- *	supported commands.
+ *	NCQ_OFF		: Turn off NCQ.
  *
- *	Cat-3 is is unclassified DEV error for known supported
- *	command.
+ *	SPEED_DOWN	: Speed down transfer speed but don't fall back
+ *			  to PIO.
  *
- *	NCQ needs to be turned off if there have been more than 3
- *	Cat-2 + Cat-3 errors during last 10 minutes.
+ *	FALLBACK_TO_PIO	: Fall back to PIO.
  *
- *	Speed down is necessary if there have been more than 3 Cat-1 +
- *	Cat-2 errors or 10 Cat-3 errors during last 10 minutes.
+ *	Even if multiple verdicts are returned, only one action is
+ *	taken per error.  ering is cleared after an action is taken.
  *
- *	Falling back to PIO mode is necessary if there have been more
- *	than 10 Cat-1 + Cat-2 + Cat-3 errors during last 5 minutes.
+ *	1. If more than 10 ATA_BUS, TOUT_HSM or UNK_DEV errors
+ *	   ocurred during last 5 mins, FALLBACK_TO_PIO
+ *
+ *	2. If more than 3 TOUT_HSM or UNK_DEV errors occurred
+ *	   during last 10 mins, NCQ_OFF.
+ *
+ *	3. If more than 3 ATA_BUS or TOUT_HSM errors, or more than 10
+ *	   UNK_DEV errors occurred during last 10 mins, SPEED_DOWN.
  *
  *	LOCKING:
  *	Inherited from caller.
@@ -1525,23 +1547,29 @@ static unsigned int ata_eh_speed_down_verdict(struct ata_device *dev)
 	struct speed_down_verdict_arg arg;
 	unsigned int verdict = 0;
 
-	/* scan past 10 mins of error history */
+	/* scan past 5 mins of error history */
 	memset(&arg, 0, sizeof(arg));
-	arg.since = j64 - min(j64, j10mins);
+	arg.since = j64 - min(j64, j5mins);
 	ata_ering_map(&dev->ering, speed_down_verdict_cb, &arg);
 
-	if (arg.nr_errors[2] + arg.nr_errors[3] > 3)
-		verdict |= ATA_EH_SPDN_NCQ_OFF;
-	if (arg.nr_errors[1] + arg.nr_errors[2] > 3 || arg.nr_errors[3] > 10)
-		verdict |= ATA_EH_SPDN_SPEED_DOWN;
+	if (arg.nr_errors[ATA_ECAT_ATA_BUS] +
+	    arg.nr_errors[ATA_ECAT_TOUT_HSM] +
+	    arg.nr_errors[ATA_ECAT_UNK_DEV] > 10)
+		verdict |= ATA_EH_SPDN_FALLBACK_TO_PIO;
 
-	/* scan past 3 mins of error history */
+	/* scan past 10 mins of error history */
 	memset(&arg, 0, sizeof(arg));
-	arg.since = j64 - min(j64, j5mins);
+	arg.since = j64 - min(j64, j10mins);
 	ata_ering_map(&dev->ering, speed_down_verdict_cb, &arg);
 
-	if (arg.nr_errors[1] + arg.nr_errors[2] + arg.nr_errors[3] > 10)
-		verdict |= ATA_EH_SPDN_FALLBACK_TO_PIO;
+	if (arg.nr_errors[ATA_ECAT_TOUT_HSM] +
+	    arg.nr_errors[ATA_ECAT_UNK_DEV] > 3)
+		verdict |= ATA_EH_SPDN_NCQ_OFF;
+
+	if (arg.nr_errors[ATA_ECAT_ATA_BUS] +
+	    arg.nr_errors[ATA_ECAT_TOUT_HSM] > 3 ||
+	    arg.nr_errors[ATA_ECAT_UNK_DEV] > 10)
+		verdict |= ATA_EH_SPDN_SPEED_DOWN;
 
 	return verdict;
 }
@@ -1549,7 +1577,7 @@ static unsigned int ata_eh_speed_down_verdict(struct ata_device *dev)
 /**
  *	ata_eh_speed_down - record error and speed down if necessary
  *	@dev: Failed device
- *	@is_io: Did the device fail during normal IO?
+ *	@eflags: mask of ATA_EFLAG_* flags
  *	@err_mask: err_mask of the error
  *
  *	Record error and examine error history to determine whether
@@ -1563,18 +1591,19 @@ static unsigned int ata_eh_speed_down_verdict(struct ata_device *dev)
  *	RETURNS:
  *	Determined recovery action.
  */
-static unsigned int ata_eh_speed_down(struct ata_device *dev, int is_io,
-				      unsigned int err_mask)
+static unsigned int ata_eh_speed_down(struct ata_device *dev,
+				unsigned int eflags, unsigned int err_mask)
 {
+	struct ata_link *link = dev->link;
 	unsigned int verdict;
 	unsigned int action = 0;
 
 	/* don't bother if Cat-0 error */
-	if (ata_eh_categorize_error(is_io, err_mask) == 0)
+	if (ata_eh_categorize_error(eflags, err_mask) == 0)
 		return 0;
 
 	/* record error and determine whether speed down is necessary */
-	ata_ering_record(&dev->ering, is_io, err_mask);
+	ata_ering_record(&dev->ering, eflags, err_mask);
 	verdict = ata_eh_speed_down_verdict(dev);
 
 	/* turn off NCQ? */
@@ -1590,7 +1619,7 @@ static unsigned int ata_eh_speed_down(struct ata_device *dev, int is_io,
 	/* speed down? */
 	if (verdict & ATA_EH_SPDN_SPEED_DOWN) {
 		/* speed down SATA link speed if possible */
-		if (sata_down_spd_limit(dev->link) == 0) {
+		if (sata_down_spd_limit(link) == 0) {
 			action |= ATA_EH_HARDRESET;
 			goto done;
 		}
@@ -1621,7 +1650,7 @@ static unsigned int ata_eh_speed_down(struct ata_device *dev, int is_io,
 	 * SATA.  Consider it only for PATA.
 	 */
 	if ((verdict & ATA_EH_SPDN_FALLBACK_TO_PIO) && (dev->spdn_cnt >= 2) &&
-	    (dev->link->ap->cbl != ATA_CBL_SATA) &&
+	    (link->ap->cbl != ATA_CBL_SATA) &&
 	    (dev->xfer_shift != ATA_SHIFT_PIO)) {
 		if (ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO) == 0) {
 			dev->spdn_cnt = 0;
@@ -1653,8 +1682,8 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 	struct ata_port *ap = link->ap;
 	struct ata_eh_context *ehc = &link->eh_context;
 	struct ata_device *dev;
-	unsigned int all_err_mask = 0;
-	int tag, is_io = 0;
+	unsigned int all_err_mask = 0, eflags = 0;
+	int tag;
 	u32 serror;
 	int rc;
 
@@ -1713,15 +1742,15 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 		ehc->i.dev = qc->dev;
 		all_err_mask |= qc->err_mask;
 		if (qc->flags & ATA_QCFLAG_IO)
-			is_io = 1;
+			eflags |= ATA_EFLAG_IS_IO;
 	}
 
 	/* enforce default EH actions */
 	if (ap->pflags & ATA_PFLAG_FROZEN ||
 	    all_err_mask & (AC_ERR_HSM | AC_ERR_TIMEOUT))
 		ehc->i.action |= ATA_EH_SOFTRESET;
-	else if ((is_io && all_err_mask) ||
-		 (!is_io && (all_err_mask & ~AC_ERR_DEV)))
+	else if (((eflags & ATA_EFLAG_IS_IO) && all_err_mask) ||
+		 (!(eflags & ATA_EFLAG_IS_IO) && (all_err_mask & ~AC_ERR_DEV)))
 		ehc->i.action |= ATA_EH_REVALIDATE;
 
 	/* If we have offending qcs and the associated failed device,
@@ -1744,7 +1773,7 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 	    dev = link->device;
 
 	if (dev)
-		ehc->i.action |= ata_eh_speed_down(dev, is_io, all_err_mask);
+		ehc->i.action |= ata_eh_speed_down(dev, eflags, all_err_mask);
 
 	DPRINTK("EXIT\n");
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index ca347b018649..74f1255e2524 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -482,7 +482,7 @@ struct ata_port_stats {
 };
 
 struct ata_ering_entry {
-	int			is_io;
+	unsigned int		eflags;
 	unsigned int		err_mask;
 	u64			timestamp;
 };
-- 
cgit v1.2.3


From 00115e0f5bc3bfdf3f3855ad89c8895f10458f92 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:28:58 +0900
Subject: libata: implement ATA_DFLAG_DUBIOUS_XFER

ATA_DFLAG_DUBIOUS_XFER is set whenever data transfer speed or method
changes and gets cleared when data transfer command succeeds in the
newly configured transfer mode.

This will be used to improve speed down logic.

Signed-off-by: Tejun Heo <htejun@gmail.com<
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 19 +++++++++++++++++++
 drivers/ata/libata-eh.c   | 33 +++++++++++++++++++++++++++++++--
 include/linux/libata.h    |  3 +++
 3 files changed, 53 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 86c10cc751e9..124b6f111cc3 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5797,6 +5797,22 @@ static void fill_result_tf(struct ata_queued_cmd *qc)
 	ap->ops->tf_read(ap, &qc->result_tf);
 }
 
+static void ata_verify_xfer(struct ata_queued_cmd *qc)
+{
+	struct ata_device *dev = qc->dev;
+
+	if (ata_tag_internal(qc->tag))
+		return;
+
+	if (ata_is_nodata(qc->tf.protocol))
+		return;
+
+	if ((dev->mwdma_mask || dev->udma_mask) && ata_is_pio(qc->tf.protocol))
+		return;
+
+	dev->flags &= ~ATA_DFLAG_DUBIOUS_XFER;
+}
+
 /**
  *	ata_qc_complete - Complete an active ATA command
  *	@qc: Command to complete
@@ -5868,6 +5884,9 @@ void ata_qc_complete(struct ata_queued_cmd *qc)
 			break;
 		}
 
+		if (unlikely(dev->flags & ATA_DFLAG_DUBIOUS_XFER))
+			ata_verify_xfer(qc);
+
 		__ata_qc_complete(qc);
 	} else {
 		if (qc->flags & ATA_QCFLAG_EH_SCHEDULED)
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 7d766ada7a5d..359a5ace8473 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -456,9 +456,20 @@ void ata_scsi_error(struct Scsi_Host *host)
 		spin_lock_irqsave(ap->lock, flags);
 
 		__ata_port_for_each_link(link, ap) {
+			struct ata_eh_context *ehc = &link->eh_context;
+			struct ata_device *dev;
+
 			memset(&link->eh_context, 0, sizeof(link->eh_context));
 			link->eh_context.i = link->eh_info;
 			memset(&link->eh_info, 0, sizeof(link->eh_info));
+
+			ata_link_for_each_dev(dev, link) {
+				int devno = dev->devno;
+
+				ehc->saved_xfer_mode[devno] = dev->xfer_mode;
+				if (ata_ncq_enabled(dev))
+					ehc->saved_ncq_enabled |= 1 << devno;
+			}
 		}
 
 		ap->pflags |= ATA_PFLAG_EH_IN_PROGRESS;
@@ -2376,11 +2387,27 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link,
 int ata_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 {
 	struct ata_port *ap = link->ap;
+	struct ata_device *dev;
+	int rc;
 
 	/* has private set_mode? */
 	if (ap->ops->set_mode)
-		return ap->ops->set_mode(link, r_failed_dev);
-	return ata_do_set_mode(link, r_failed_dev);
+		rc = ap->ops->set_mode(link, r_failed_dev);
+	else
+		rc = ata_do_set_mode(link, r_failed_dev);
+
+	/* if transfer mode has changed, set DUBIOUS_XFER on device */
+	ata_link_for_each_dev(dev, link) {
+		struct ata_eh_context *ehc = &link->eh_context;
+		u8 saved_xfer_mode = ehc->saved_xfer_mode[dev->devno];
+		u8 saved_ncq = !!(ehc->saved_ncq_enabled & (1 << dev->devno));
+
+		if (dev->xfer_mode != saved_xfer_mode ||
+		    ata_ncq_enabled(dev) != saved_ncq)
+			dev->flags |= ATA_DFLAG_DUBIOUS_XFER;
+	}
+
+	return rc;
 }
 
 static int ata_link_nr_enabled(struct ata_link *link)
@@ -2441,6 +2468,8 @@ static int ata_eh_schedule_probe(struct ata_device *dev)
 	ata_dev_init(dev);
 	ehc->did_probe_mask |= (1 << dev->devno);
 	ehc->i.action |= ATA_EH_SOFTRESET;
+	ehc->saved_xfer_mode[dev->devno] = 0;
+	ehc->saved_ncq_enabled &= ~(1 << dev->devno);
 
 	return 1;
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 74f1255e2524..131fb6625e14 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -143,6 +143,7 @@ enum {
 	ATA_DFLAG_NCQ_OFF	= (1 << 13), /* device limited to non-NCQ mode */
 	ATA_DFLAG_SPUNDOWN	= (1 << 14), /* XXX: for spindown_compat */
 	ATA_DFLAG_SLEEPING	= (1 << 15), /* device is sleeping */
+	ATA_DFLAG_DUBIOUS_XFER	= (1 << 16), /* data transfer not verified */
 	ATA_DFLAG_INIT_MASK	= (1 << 24) - 1,
 
 	ATA_DFLAG_DETACH	= (1 << 24),
@@ -560,6 +561,8 @@ struct ata_eh_context {
 	int			tries[ATA_MAX_DEVICES];
 	unsigned int		classes[ATA_MAX_DEVICES];
 	unsigned int		did_probe_mask;
+	unsigned int		saved_ncq_enabled;
+	u8			saved_xfer_mode[ATA_MAX_DEVICES];
 };
 
 struct ata_acpi_drive
-- 
cgit v1.2.3


From 6357357cae7794dcb89cace758108dec612e7ed5 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:43:39 +0900
Subject: libata: export xfermode / PATA timing related functions

Export the following xfermode related functions.

* ata_pack_xfermask()
* ata_unpack_xfermask()
* ata_xfer_mask2mode()
* ata_xfer_mode2mask()
* ata_xfer_mode2shift()
* ata_mode_string()
* ata_id_xfermask()
* ata_timing_find_mode()

These functions will be used later by LLD updates.  While at it,
change unsigned short @speed to u8 @xfer_mode in
ata_timing_find_mode() for consistency.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 33 +++++++++++++++++++--------------
 include/linux/libata.h    | 10 ++++++++++
 2 files changed, 29 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 124b6f111cc3..a2f2ff864b37 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -454,9 +454,8 @@ int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev,
  *	RETURNS:
  *	Packed xfer_mask.
  */
-static unsigned int ata_pack_xfermask(unsigned int pio_mask,
-				      unsigned int mwdma_mask,
-				      unsigned int udma_mask)
+unsigned int ata_pack_xfermask(unsigned int pio_mask,
+			       unsigned int mwdma_mask, unsigned int udma_mask)
 {
 	return ((pio_mask << ATA_SHIFT_PIO) & ATA_MASK_PIO) |
 		((mwdma_mask << ATA_SHIFT_MWDMA) & ATA_MASK_MWDMA) |
@@ -473,10 +472,8 @@ static unsigned int ata_pack_xfermask(unsigned int pio_mask,
  *	Unpack @xfer_mask into @pio_mask, @mwdma_mask and @udma_mask.
  *	Any NULL distination masks will be ignored.
  */
-static void ata_unpack_xfermask(unsigned int xfer_mask,
-				unsigned int *pio_mask,
-				unsigned int *mwdma_mask,
-				unsigned int *udma_mask)
+void ata_unpack_xfermask(unsigned int xfer_mask, unsigned int *pio_mask,
+			 unsigned int *mwdma_mask, unsigned int *udma_mask)
 {
 	if (pio_mask)
 		*pio_mask = (xfer_mask & ATA_MASK_PIO) >> ATA_SHIFT_PIO;
@@ -509,7 +506,7 @@ static const struct ata_xfer_ent {
  *	RETURNS:
  *	Matching XFER_* value, 0 if no match found.
  */
-static u8 ata_xfer_mask2mode(unsigned int xfer_mask)
+u8 ata_xfer_mask2mode(unsigned int xfer_mask)
 {
 	int highbit = fls(xfer_mask) - 1;
 	const struct ata_xfer_ent *ent;
@@ -532,7 +529,7 @@ static u8 ata_xfer_mask2mode(unsigned int xfer_mask)
  *	RETURNS:
  *	Matching xfer_mask, 0 if no match found.
  */
-static unsigned int ata_xfer_mode2mask(u8 xfer_mode)
+unsigned int ata_xfer_mode2mask(u8 xfer_mode)
 {
 	const struct ata_xfer_ent *ent;
 
@@ -554,7 +551,7 @@ static unsigned int ata_xfer_mode2mask(u8 xfer_mode)
  *	RETURNS:
  *	Matching xfer_shift, -1 if no match found.
  */
-static int ata_xfer_mode2shift(unsigned int xfer_mode)
+int ata_xfer_mode2shift(unsigned int xfer_mode)
 {
 	const struct ata_xfer_ent *ent;
 
@@ -578,7 +575,7 @@ static int ata_xfer_mode2shift(unsigned int xfer_mode)
  *	Constant C string representing highest speed listed in
  *	@mode_mask, or the constant C string "<n/a>".
  */
-static const char *ata_mode_string(unsigned int xfer_mask)
+const char *ata_mode_string(unsigned int xfer_mask)
 {
 	static const char * const xfer_mode_str[] = {
 		"PIO0",
@@ -1468,7 +1465,7 @@ static inline void ata_dump_id(const u16 *id)
  *	RETURNS:
  *	Computed xfermask
  */
-static unsigned int ata_id_xfermask(const u16 *id)
+unsigned int ata_id_xfermask(const u16 *id)
 {
 	unsigned int pio_mask, mwdma_mask, udma_mask;
 
@@ -2855,11 +2852,11 @@ void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
 	if (what & ATA_TIMING_UDMA   ) m->udma    = max(a->udma,    b->udma);
 }
 
-static const struct ata_timing *ata_timing_find_mode(unsigned short speed)
+const struct ata_timing *ata_timing_find_mode(u8 xfer_mode)
 {
 	const struct ata_timing *t;
 
-	for (t = ata_timing; t->mode != speed; t++)
+	for (t = ata_timing; t->mode != xfer_mode; t++)
 		if (t->mode == 0xFF)
 			return NULL;
 	return t;
@@ -7590,6 +7587,13 @@ EXPORT_SYMBOL_GPL(ata_std_dev_select);
 EXPORT_SYMBOL_GPL(sata_print_link_status);
 EXPORT_SYMBOL_GPL(ata_tf_to_fis);
 EXPORT_SYMBOL_GPL(ata_tf_from_fis);
+EXPORT_SYMBOL_GPL(ata_pack_xfermask);
+EXPORT_SYMBOL_GPL(ata_unpack_xfermask);
+EXPORT_SYMBOL_GPL(ata_xfer_mask2mode);
+EXPORT_SYMBOL_GPL(ata_xfer_mode2mask);
+EXPORT_SYMBOL_GPL(ata_xfer_mode2shift);
+EXPORT_SYMBOL_GPL(ata_mode_string);
+EXPORT_SYMBOL_GPL(ata_id_xfermask);
 EXPORT_SYMBOL_GPL(ata_check_status);
 EXPORT_SYMBOL_GPL(ata_altstatus);
 EXPORT_SYMBOL_GPL(ata_exec_command);
@@ -7655,6 +7659,7 @@ EXPORT_SYMBOL_GPL(ata_id_to_dma_mode);
 EXPORT_SYMBOL_GPL(ata_scsi_simulate);
 
 EXPORT_SYMBOL_GPL(ata_pio_need_iordy);
+EXPORT_SYMBOL_GPL(ata_timing_find_mode);
 EXPORT_SYMBOL_GPL(ata_timing_compute);
 EXPORT_SYMBOL_GPL(ata_timing_merge);
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 131fb6625e14..083dd77b120d 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -851,6 +851,15 @@ extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 extern void ata_tf_to_fis(const struct ata_taskfile *tf,
 			  u8 pmp, int is_cmd, u8 *fis);
 extern void ata_tf_from_fis(const u8 *fis, struct ata_taskfile *tf);
+extern unsigned int ata_pack_xfermask(unsigned int pio_mask,
+			unsigned int mwdma_mask, unsigned int udma_mask);
+extern void ata_unpack_xfermask(unsigned int xfer_mask, unsigned int *pio_mask,
+			unsigned int *mwdma_mask, unsigned int *udma_mask);
+extern u8 ata_xfer_mask2mode(unsigned int xfer_mask);
+extern unsigned int ata_xfer_mode2mask(u8 xfer_mode);
+extern int ata_xfer_mode2shift(unsigned int xfer_mode);
+extern const char *ata_mode_string(unsigned int xfer_mask);
+extern unsigned int ata_id_xfermask(const u16 *id);
 extern void ata_noop_dev_select(struct ata_port *ap, unsigned int device);
 extern void ata_std_dev_select(struct ata_port *ap, unsigned int device);
 extern u8 ata_check_status(struct ata_port *ap);
@@ -920,6 +929,7 @@ extern int ata_cable_unknown(struct ata_port *ap);
  */
 
 extern unsigned int ata_pio_need_iordy(const struct ata_device *);
+extern const struct ata_timing *ata_timing_find_mode(u8 xfer_mode);
 extern int ata_timing_compute(struct ata_device *, unsigned short,
 			      struct ata_timing *, int, int);
 extern void ata_timing_merge(const struct ata_timing *,
-- 
cgit v1.2.3


From 70cd071e4ecc06c985189665af75c108601fd5a3 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:43:40 +0900
Subject: libata: clean up xfermode / PATA timing related stuff

* s/ATA_BITS_(PIO|MWDMA|UDMA)/ATA_NR_\1_MODES/g

* Consistently use 0xff to indicate invalid transfer mode (0x00 is
  valid for PIO_SLOW).

* Make ata_xfer_mode2mask() return proper mode mask instead of just
  the highest bit.

* Sort ata_timing table in increasing xfermode order and update
  ata_timing_find_mode() accordingly.

This patch doesn't introduce any behavior change.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 82 +++++++++++++++++++++++------------------------
 include/linux/libata.h    | 21 ++++++------
 2 files changed, 52 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index a2f2ff864b37..a70ed7b3bc55 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -487,9 +487,9 @@ static const struct ata_xfer_ent {
 	int shift, bits;
 	u8 base;
 } ata_xfer_tbl[] = {
-	{ ATA_SHIFT_PIO, ATA_BITS_PIO, XFER_PIO_0 },
-	{ ATA_SHIFT_MWDMA, ATA_BITS_MWDMA, XFER_MW_DMA_0 },
-	{ ATA_SHIFT_UDMA, ATA_BITS_UDMA, XFER_UDMA_0 },
+	{ ATA_SHIFT_PIO, ATA_NR_PIO_MODES, XFER_PIO_0 },
+	{ ATA_SHIFT_MWDMA, ATA_NR_MWDMA_MODES, XFER_MW_DMA_0 },
+	{ ATA_SHIFT_UDMA, ATA_NR_UDMA_MODES, XFER_UDMA_0 },
 	{ -1, },
 };
 
@@ -504,7 +504,7 @@ static const struct ata_xfer_ent {
  *	None.
  *
  *	RETURNS:
- *	Matching XFER_* value, 0 if no match found.
+ *	Matching XFER_* value, 0xff if no match found.
  */
 u8 ata_xfer_mask2mode(unsigned int xfer_mask)
 {
@@ -514,7 +514,7 @@ u8 ata_xfer_mask2mode(unsigned int xfer_mask)
 	for (ent = ata_xfer_tbl; ent->shift >= 0; ent++)
 		if (highbit >= ent->shift && highbit < ent->shift + ent->bits)
 			return ent->base + highbit - ent->shift;
-	return 0;
+	return 0xff;
 }
 
 /**
@@ -535,7 +535,8 @@ unsigned int ata_xfer_mode2mask(u8 xfer_mode)
 
 	for (ent = ata_xfer_tbl; ent->shift >= 0; ent++)
 		if (xfer_mode >= ent->base && xfer_mode < ent->base + ent->bits)
-			return 1 << (ent->shift + xfer_mode - ent->base);
+			return ((2 << (ent->shift + xfer_mode - ent->base)) - 1)
+				& ~((1 << ent->shift) - 1);
 	return 0;
 }
 
@@ -1314,7 +1315,7 @@ void ata_id_to_dma_mode(struct ata_device *dev, u8 unknown)
 	/* Select the mode in use */
 	mode = ata_xfer_mask2mode(mask);
 
-	if (mode != 0) {
+	if (mode != 0xff) {
 		ata_dev_printk(dev, KERN_INFO, "configured for %s\n",
 		       ata_mode_string(mask));
 	} else {
@@ -2788,38 +2789,33 @@ int sata_set_spd(struct ata_link *link)
  */
 
 static const struct ata_timing ata_timing[] = {
+/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 }, */
+	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
+	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
+	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
+	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
+	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
+	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
+	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
 
-	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
-	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
-	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
-	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
+	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
+	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
+	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
 
-	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
+	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
+	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
+	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
 	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 100,   0 },
-	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
-	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
-	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
+	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
 
 /*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0,   0, 150 }, */
-
-	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
-	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
-	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
-
-	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
-	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
-	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
-
-	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
-	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
-	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
-	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
-
-	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
-	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
-	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
-
-/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 }, */
+	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
+	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
+	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
+	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
+	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
+	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
+	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
 
 	{ 0xFF }
 };
@@ -2854,12 +2850,14 @@ void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
 
 const struct ata_timing *ata_timing_find_mode(u8 xfer_mode)
 {
-	const struct ata_timing *t;
+	const struct ata_timing *t = ata_timing;
+
+	while (xfer_mode > t->mode)
+		t++;
 
-	for (t = ata_timing; t->mode != xfer_mode; t++)
-		if (t->mode == 0xFF)
-			return NULL;
-	return t;
+	if (xfer_mode == t->mode)
+		return t;
+	return NULL;
 }
 
 int ata_timing_compute(struct ata_device *adev, unsigned short speed,
@@ -3122,7 +3120,7 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 		dev->dma_mode = ata_xfer_mask2mode(dma_mask);
 
 		found = 1;
-		if (dev->dma_mode)
+		if (dev->dma_mode != 0xff)
 			used_dma = 1;
 	}
 	if (!found)
@@ -3133,7 +3131,7 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 		if (!ata_dev_enabled(dev))
 			continue;
 
-		if (!dev->pio_mode) {
+		if (dev->pio_mode == 0xff) {
 			ata_dev_printk(dev, KERN_WARNING, "no PIO support\n");
 			rc = -EINVAL;
 			goto out;
@@ -3147,7 +3145,7 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 
 	/* step 3: set host DMA timings */
 	ata_link_for_each_dev(dev, link) {
-		if (!ata_dev_enabled(dev) || !dev->dma_mode)
+		if (!ata_dev_enabled(dev) || dev->dma_mode == 0xff)
 			continue;
 
 		dev->xfer_mode = dev->dma_mode;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 083dd77b120d..e2ed3bac8c5b 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -269,17 +269,20 @@ enum {
 	/* encoding various smaller bitmaps into a single
 	 * unsigned int bitmap
 	 */
-	ATA_BITS_PIO		= 7,
-	ATA_BITS_MWDMA		= 5,
-	ATA_BITS_UDMA		= 8,
+	ATA_NR_PIO_MODES	= 7,
+	ATA_NR_MWDMA_MODES	= 5,
+	ATA_NR_UDMA_MODES	= 8,
 
 	ATA_SHIFT_PIO		= 0,
-	ATA_SHIFT_MWDMA		= ATA_SHIFT_PIO + ATA_BITS_PIO,
-	ATA_SHIFT_UDMA		= ATA_SHIFT_MWDMA + ATA_BITS_MWDMA,
-
-	ATA_MASK_PIO		= ((1 << ATA_BITS_PIO) - 1) << ATA_SHIFT_PIO,
-	ATA_MASK_MWDMA		= ((1 << ATA_BITS_MWDMA) - 1) << ATA_SHIFT_MWDMA,
-	ATA_MASK_UDMA		= ((1 << ATA_BITS_UDMA) - 1) << ATA_SHIFT_UDMA,
+	ATA_SHIFT_MWDMA		= ATA_SHIFT_PIO + ATA_NR_PIO_MODES,
+	ATA_SHIFT_UDMA		= ATA_SHIFT_MWDMA + ATA_NR_MWDMA_MODES,
+
+	ATA_MASK_PIO		= ((1 << ATA_NR_PIO_MODES) - 1)
+					<< ATA_SHIFT_PIO,
+	ATA_MASK_MWDMA		= ((1 << ATA_NR_MWDMA_MODES) - 1)
+					<< ATA_SHIFT_MWDMA,
+	ATA_MASK_UDMA		= ((1 << ATA_NR_UDMA_MODES) - 1)
+					<< ATA_SHIFT_UDMA,
 
 	/* size of buffer to pad xfers ending on unaligned boundaries */
 	ATA_DMA_PAD_SZ		= 4,
-- 
cgit v1.2.3


From 9d3501ab962b1506d93974faf8509251b4a85fbc Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:43:41 +0900
Subject: libata: kill ata_id_to_dma_mode()

ata_id_to_dma_mode() isn't quite generic.  The function is basically
privately implemented ata_id_xfermask() combined with hardcoded mode
printing and configuration which are specific to ata_generic.

Kill the function and open code it in generic_set_mode() using generic
xfermode handling functions.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ata_generic.c | 17 ++++++++++++++++-
 drivers/ata/libata-core.c | 43 -------------------------------------------
 include/linux/libata.h    |  1 -
 3 files changed, 16 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index fb5434c8ae6e..d11a4f1fe86a 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -63,7 +63,22 @@ static int generic_set_mode(struct ata_link *link, struct ata_device **unused)
 		/* We do need the right mode information for DMA or PIO
 		   and this comes from the current configuration flags */
 		if (dma_enabled & (1 << (5 + dev->devno))) {
-			ata_id_to_dma_mode(dev, XFER_MW_DMA_0);
+			unsigned int xfer_mask = ata_id_xfermask(dev->id);
+			const char *name;
+
+			if (xfer_mask & (ATA_MASK_MWDMA | ATA_MASK_UDMA))
+				name = ata_mode_string(xfer_mask);
+			else {
+				/* SWDMA perhaps? */
+				name = "DMA";
+				xfer_mask |= ata_xfer_mode2mask(XFER_MW_DMA_0);
+			}
+
+			ata_dev_printk(dev, KERN_INFO, "configured for %s\n",
+				       name);
+
+			dev->xfer_mode = ata_xfer_mask2mode(xfer_mask);
+			dev->xfer_shift = ata_xfer_mode2shift(dev->xfer_mode);
 			dev->flags &= ~ATA_DFLAG_PIO;
 		} else {
 			ata_dev_printk(dev, KERN_INFO, "configured for PIO\n");
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index a70ed7b3bc55..ab84772235c9 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1287,48 +1287,6 @@ static int ata_hpa_resize(struct ata_device *dev)
 	return 0;
 }
 
-/**
- *	ata_id_to_dma_mode	-	Identify DMA mode from id block
- *	@dev: device to identify
- *	@unknown: mode to assume if we cannot tell
- *
- *	Set up the timing values for the device based upon the identify
- *	reported values for the DMA mode. This function is used by drivers
- *	which rely upon firmware configured modes, but wish to report the
- *	mode correctly when possible.
- *
- *	In addition we emit similarly formatted messages to the default
- *	ata_dev_set_mode handler, in order to provide consistency of
- *	presentation.
- */
-
-void ata_id_to_dma_mode(struct ata_device *dev, u8 unknown)
-{
-	unsigned int mask;
-	u8 mode;
-
-	/* Pack the DMA modes */
-	mask = ((dev->id[63] >> 8) << ATA_SHIFT_MWDMA) & ATA_MASK_MWDMA;
-	if (dev->id[53] & 0x04)
-		mask |= ((dev->id[88] >> 8) << ATA_SHIFT_UDMA) & ATA_MASK_UDMA;
-
-	/* Select the mode in use */
-	mode = ata_xfer_mask2mode(mask);
-
-	if (mode != 0xff) {
-		ata_dev_printk(dev, KERN_INFO, "configured for %s\n",
-		       ata_mode_string(mask));
-	} else {
-		/* SWDMA perhaps ? */
-		mode = unknown;
-		ata_dev_printk(dev, KERN_INFO, "configured for DMA\n");
-	}
-
-	/* Configure the device reporting */
-	dev->xfer_mode = mode;
-	dev->xfer_shift = ata_xfer_mode2shift(mode);
-}
-
 /**
  *	ata_noop_dev_select - Select device 0/1 on ATA bus
  *	@ap: ATA channel to manipulate
@@ -7653,7 +7611,6 @@ EXPORT_SYMBOL_GPL(ata_host_resume);
 #endif /* CONFIG_PM */
 EXPORT_SYMBOL_GPL(ata_id_string);
 EXPORT_SYMBOL_GPL(ata_id_c_string);
-EXPORT_SYMBOL_GPL(ata_id_to_dma_mode);
 EXPORT_SYMBOL_GPL(ata_scsi_simulate);
 
 EXPORT_SYMBOL_GPL(ata_pio_need_iordy);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index e2ed3bac8c5b..d33702fe78f9 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -890,7 +890,6 @@ extern void ata_id_string(const u16 *id, unsigned char *s,
 			  unsigned int ofs, unsigned int len);
 extern void ata_id_c_string(const u16 *id, unsigned char *s,
 			    unsigned int ofs, unsigned int len);
-extern void ata_id_to_dma_mode(struct ata_device *dev, u8 unknown);
 extern void ata_bmdma_setup(struct ata_queued_cmd *qc);
 extern void ata_bmdma_start(struct ata_queued_cmd *qc);
 extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
-- 
cgit v1.2.3


From 7dc951aefdc1dc20228691b04867fb6195864d67 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:43:42 +0900
Subject: libata: xfer_mask is unsigned long not unsigned int

Jeff says xfer_mask is unsigned long not unsigned int.  Convert all
xfermask fields and handling functions to deal with unsigned longs.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 29 +++++++++++++++--------------
 include/linux/libata.h    | 46 +++++++++++++++++++++++++---------------------
 2 files changed, 40 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index ab84772235c9..711595390996 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -454,8 +454,9 @@ int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev,
  *	RETURNS:
  *	Packed xfer_mask.
  */
-unsigned int ata_pack_xfermask(unsigned int pio_mask,
-			       unsigned int mwdma_mask, unsigned int udma_mask)
+unsigned long ata_pack_xfermask(unsigned long pio_mask,
+				unsigned long mwdma_mask,
+				unsigned long udma_mask)
 {
 	return ((pio_mask << ATA_SHIFT_PIO) & ATA_MASK_PIO) |
 		((mwdma_mask << ATA_SHIFT_MWDMA) & ATA_MASK_MWDMA) |
@@ -472,8 +473,8 @@ unsigned int ata_pack_xfermask(unsigned int pio_mask,
  *	Unpack @xfer_mask into @pio_mask, @mwdma_mask and @udma_mask.
  *	Any NULL distination masks will be ignored.
  */
-void ata_unpack_xfermask(unsigned int xfer_mask, unsigned int *pio_mask,
-			 unsigned int *mwdma_mask, unsigned int *udma_mask)
+void ata_unpack_xfermask(unsigned long xfer_mask, unsigned long *pio_mask,
+			 unsigned long *mwdma_mask, unsigned long *udma_mask)
 {
 	if (pio_mask)
 		*pio_mask = (xfer_mask & ATA_MASK_PIO) >> ATA_SHIFT_PIO;
@@ -506,7 +507,7 @@ static const struct ata_xfer_ent {
  *	RETURNS:
  *	Matching XFER_* value, 0xff if no match found.
  */
-u8 ata_xfer_mask2mode(unsigned int xfer_mask)
+u8 ata_xfer_mask2mode(unsigned long xfer_mask)
 {
 	int highbit = fls(xfer_mask) - 1;
 	const struct ata_xfer_ent *ent;
@@ -529,7 +530,7 @@ u8 ata_xfer_mask2mode(unsigned int xfer_mask)
  *	RETURNS:
  *	Matching xfer_mask, 0 if no match found.
  */
-unsigned int ata_xfer_mode2mask(u8 xfer_mode)
+unsigned long ata_xfer_mode2mask(u8 xfer_mode)
 {
 	const struct ata_xfer_ent *ent;
 
@@ -552,7 +553,7 @@ unsigned int ata_xfer_mode2mask(u8 xfer_mode)
  *	RETURNS:
  *	Matching xfer_shift, -1 if no match found.
  */
-int ata_xfer_mode2shift(unsigned int xfer_mode)
+int ata_xfer_mode2shift(unsigned long xfer_mode)
 {
 	const struct ata_xfer_ent *ent;
 
@@ -576,7 +577,7 @@ int ata_xfer_mode2shift(unsigned int xfer_mode)
  *	Constant C string representing highest speed listed in
  *	@mode_mask, or the constant C string "<n/a>".
  */
-const char *ata_mode_string(unsigned int xfer_mask)
+const char *ata_mode_string(unsigned long xfer_mask)
 {
 	static const char * const xfer_mode_str[] = {
 		"PIO0",
@@ -1424,9 +1425,9 @@ static inline void ata_dump_id(const u16 *id)
  *	RETURNS:
  *	Computed xfermask
  */
-unsigned int ata_id_xfermask(const u16 *id)
+unsigned long ata_id_xfermask(const u16 *id)
 {
-	unsigned int pio_mask, mwdma_mask, udma_mask;
+	unsigned long pio_mask, mwdma_mask, udma_mask;
 
 	/* Usual case. Word 53 indicates word 64 is valid */
 	if (id[ATA_ID_FIELD_VALID] & (1 << 1)) {
@@ -2050,7 +2051,7 @@ int ata_dev_configure(struct ata_device *dev)
 	struct ata_eh_context *ehc = &dev->link->eh_context;
 	int print_info = ehc->i.flags & ATA_EHI_PRINTINFO;
 	const u16 *id = dev->id;
-	unsigned int xfer_mask;
+	unsigned long xfer_mask;
 	char revbuf[7];		/* XYZ-99\0 */
 	char fwrevbuf[ATA_ID_FW_REV_LEN+1];
 	char modelbuf[ATA_ID_PROD_LEN+1];
@@ -2907,8 +2908,8 @@ int ata_timing_compute(struct ata_device *adev, unsigned short speed,
 int ata_down_xfermask_limit(struct ata_device *dev, unsigned int sel)
 {
 	char buf[32];
-	unsigned int orig_mask, xfer_mask;
-	unsigned int pio_mask, mwdma_mask, udma_mask;
+	unsigned long orig_mask, xfer_mask;
+	unsigned long pio_mask, mwdma_mask, udma_mask;
 	int quiet, highbit;
 
 	quiet = !!(sel & ATA_DNXFER_QUIET);
@@ -3052,7 +3053,7 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 
 	/* step 1: calculate xfer_mask */
 	ata_link_for_each_dev(dev, link) {
-		unsigned int pio_mask, dma_mask;
+		unsigned long pio_mask, dma_mask;
 		unsigned int mode_mask;
 
 		if (!ata_dev_enabled(dev))
diff --git a/include/linux/libata.h b/include/linux/libata.h
index d33702fe78f9..d9eb20c67bb6 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -267,7 +267,7 @@ enum {
 	PORT_DISABLED		= 2,
 
 	/* encoding various smaller bitmaps into a single
-	 * unsigned int bitmap
+	 * unsigned long bitmap
 	 */
 	ATA_NR_PIO_MODES	= 7,
 	ATA_NR_MWDMA_MODES	= 5,
@@ -277,13 +277,6 @@ enum {
 	ATA_SHIFT_MWDMA		= ATA_SHIFT_PIO + ATA_NR_PIO_MODES,
 	ATA_SHIFT_UDMA		= ATA_SHIFT_MWDMA + ATA_NR_MWDMA_MODES,
 
-	ATA_MASK_PIO		= ((1 << ATA_NR_PIO_MODES) - 1)
-					<< ATA_SHIFT_PIO,
-	ATA_MASK_MWDMA		= ((1 << ATA_NR_MWDMA_MODES) - 1)
-					<< ATA_SHIFT_MWDMA,
-	ATA_MASK_UDMA		= ((1 << ATA_NR_UDMA_MODES) - 1)
-					<< ATA_SHIFT_UDMA,
-
 	/* size of buffer to pad xfers ending on unaligned boundaries */
 	ATA_DMA_PAD_SZ		= 4,
 	ATA_DMA_PAD_BUF_SZ	= ATA_DMA_PAD_SZ * ATA_MAX_QUEUE,
@@ -355,6 +348,15 @@ enum {
 	ATA_DMA_MASK_CFA	= (1 << 2),	/* DMA on CF Card */
 };
 
+enum ata_xfer_mask {
+	ATA_MASK_PIO		= ((1LU << ATA_NR_PIO_MODES) - 1)
+					<< ATA_SHIFT_PIO,
+	ATA_MASK_MWDMA		= ((1LU << ATA_NR_MWDMA_MODES) - 1)
+					<< ATA_SHIFT_MWDMA,
+	ATA_MASK_UDMA		= ((1LU << ATA_NR_UDMA_MODES) - 1)
+					<< ATA_SHIFT_UDMA,
+};
+
 enum hsm_task_states {
 	HSM_ST_IDLE,		/* no command on going */
 	HSM_ST_FIRST,		/* (waiting the device to)
@@ -526,9 +528,9 @@ struct ata_device {
 	unsigned int		cdb_len;
 
 	/* per-dev xfer mask */
-	unsigned int		pio_mask;
-	unsigned int		mwdma_mask;
-	unsigned int		udma_mask;
+	unsigned long		pio_mask;
+	unsigned long		mwdma_mask;
+	unsigned long		udma_mask;
 
 	/* for CHS addressing */
 	u16			cylinders;	/* Number of cylinders */
@@ -854,15 +856,16 @@ extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 extern void ata_tf_to_fis(const struct ata_taskfile *tf,
 			  u8 pmp, int is_cmd, u8 *fis);
 extern void ata_tf_from_fis(const u8 *fis, struct ata_taskfile *tf);
-extern unsigned int ata_pack_xfermask(unsigned int pio_mask,
-			unsigned int mwdma_mask, unsigned int udma_mask);
-extern void ata_unpack_xfermask(unsigned int xfer_mask, unsigned int *pio_mask,
-			unsigned int *mwdma_mask, unsigned int *udma_mask);
-extern u8 ata_xfer_mask2mode(unsigned int xfer_mask);
-extern unsigned int ata_xfer_mode2mask(u8 xfer_mode);
-extern int ata_xfer_mode2shift(unsigned int xfer_mode);
-extern const char *ata_mode_string(unsigned int xfer_mask);
-extern unsigned int ata_id_xfermask(const u16 *id);
+extern unsigned long ata_pack_xfermask(unsigned long pio_mask,
+			unsigned long mwdma_mask, unsigned long udma_mask);
+extern void ata_unpack_xfermask(unsigned long xfer_mask,
+			unsigned long *pio_mask, unsigned long *mwdma_mask,
+			unsigned long *udma_mask);
+extern u8 ata_xfer_mask2mode(unsigned long xfer_mask);
+extern unsigned long ata_xfer_mode2mask(u8 xfer_mode);
+extern int ata_xfer_mode2shift(unsigned long xfer_mode);
+extern const char *ata_mode_string(unsigned long xfer_mask);
+extern unsigned long ata_id_xfermask(const u16 *id);
 extern void ata_noop_dev_select(struct ata_port *ap, unsigned int device);
 extern void ata_std_dev_select(struct ata_port *ap, unsigned int device);
 extern u8 ata_check_status(struct ata_port *ap);
@@ -1001,7 +1004,8 @@ extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
 				    const struct ata_port_info * const * ppi,
 				    struct ata_host **r_host);
 extern int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits);
-extern unsigned long ata_pci_default_filter(struct ata_device *, unsigned long);
+extern unsigned long ata_pci_default_filter(struct ata_device *dev,
+					    unsigned long xfer_mask);
 #endif /* CONFIG_PCI */
 
 /*
-- 
cgit v1.2.3


From c88f90c3779cd5e710f2acdf59ad2bd0380de98d Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 27 Nov 2007 19:43:48 +0900
Subject: libata: add ATA_CBL_PATA_IGN

ATA_CBL_PATA_UNK indicates that the cable type can't be determined
from the host side and might be either 80c or 40c.  libata applies
drive or other generic limit in this case.  However, there are
controllers where both host and drive side detections are
misimplemented and the driver has to rely solely on private method -
peeking BIOS or ACPI configuration or using some other private
mechanism.

This patch adds ATA_CBL_PATA_IGN which tells libata to ignore the
cable type completely and just let the LLD determine the transfer mode
via host transfer mode masks and ->mode_filter().

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 13 +++++++++++++
 include/linux/ata.h       |  7 ++++---
 include/linux/libata.h    |  1 +
 3 files changed, 18 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 711595390996..62c4b328b0b6 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -2353,6 +2353,18 @@ int ata_cable_unknown(struct ata_port *ap)
 	return ATA_CBL_PATA_UNK;
 }
 
+/**
+ *	ata_cable_ignore	-	return ignored PATA cable.
+ *	@ap: port
+ *
+ *	Helper method for drivers which don't use cable type to limit
+ *	transfer mode.
+ */
+int ata_cable_ignore(struct ata_port *ap)
+{
+	return ATA_CBL_PATA_IGN;
+}
+
 /**
  *	ata_cable_sata	-	return SATA cable type
  *	@ap: port
@@ -7665,4 +7677,5 @@ EXPORT_SYMBOL_GPL(ata_dev_try_classify);
 EXPORT_SYMBOL_GPL(ata_cable_40wire);
 EXPORT_SYMBOL_GPL(ata_cable_80wire);
 EXPORT_SYMBOL_GPL(ata_cable_unknown);
+EXPORT_SYMBOL_GPL(ata_cable_ignore);
 EXPORT_SYMBOL_GPL(ata_cable_sata);
diff --git a/include/linux/ata.h b/include/linux/ata.h
index 43fecf62773a..c17e9404c88d 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -286,9 +286,10 @@ enum {
 	ATA_CBL_NONE		= 0,
 	ATA_CBL_PATA40		= 1,
 	ATA_CBL_PATA80		= 2,
-	ATA_CBL_PATA40_SHORT	= 3,		/* 40 wire cable to high UDMA spec */
-	ATA_CBL_PATA_UNK	= 4,
-	ATA_CBL_SATA		= 5,
+	ATA_CBL_PATA40_SHORT	= 3,	/* 40 wire cable to high UDMA spec */
+	ATA_CBL_PATA_UNK	= 4,	/* don't know, maybe 80c? */
+	ATA_CBL_PATA_IGN	= 5,	/* don't know, ignore cable handling */
+	ATA_CBL_SATA		= 6,
 
 	/* SATA Status and Control Registers */
 	SCR_STATUS		= 0,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index d9eb20c67bb6..bdb7c6e13993 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -927,6 +927,7 @@ extern u8 ata_irq_on(struct ata_port *ap);
 extern int ata_cable_40wire(struct ata_port *ap);
 extern int ata_cable_80wire(struct ata_port *ap);
 extern int ata_cable_sata(struct ata_port *ap);
+extern int ata_cable_ignore(struct ata_port *ap);
 extern int ata_cable_unknown(struct ata_port *ap);
 
 /*
-- 
cgit v1.2.3


From 7c77fa4d51b1480bcec2e898c94d6912fe063c16 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 18 Dec 2007 16:33:03 +0900
Subject: libata: separate out ata_acpi_gtm_xfermask() from
 pacpi_discover_modes()

Finding out matching transfer mode from ACPI GTM values is useful for
other purposes too.  Separate out the function and timing tables from
pata_acpi::pacpi_discover_modes().

Other than checking shared-configuration bit after doing
ata_acpi_gtm() in pacpi_discover_modes() which should be safe, this
patch doesn't introduce any behavior change.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-acpi.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ata/pata_acpi.c   | 66 +++++++--------------------------------
 include/linux/libata.h    | 25 +++++++++++++++
 3 files changed, 114 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index ebc4dfcf2f19..d6697baf243d 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -441,6 +441,84 @@ static int ata_dev_get_GTF(struct ata_device *dev, struct ata_acpi_gtf **gtf)
 	return rc;
 }
 
+/* Welcome to ACPI, bring a bucket */
+const unsigned int ata_acpi_pio_cycle[7] = {
+	600, 383, 240, 180, 120, 100, 80
+};
+EXPORT_SYMBOL_GPL(ata_acpi_pio_cycle);
+
+const unsigned int ata_acpi_mwdma_cycle[5] = {
+	480, 150, 120, 100, 80
+};
+EXPORT_SYMBOL_GPL(ata_acpi_mwdma_cycle);
+
+const unsigned int ata_acpi_udma_cycle[7] = {
+	120, 80, 60, 45, 30, 20, 15
+};
+EXPORT_SYMBOL_GPL(ata_acpi_udma_cycle);
+
+/**
+ * ata_acpi_gtm_xfermode - determine xfermode from GTM parameter
+ * @dev: target device
+ * @gtm: GTM parameter to use
+ *
+ * Determine xfermask for @dev from @gtm.
+ *
+ * LOCKING:
+ * None.
+ *
+ * RETURNS:
+ * Determined xfermask.
+ */
+unsigned long ata_acpi_gtm_xfermask(struct ata_device *dev,
+				    const struct ata_acpi_gtm *gtm)
+{
+	int unit, i;
+	u32 t;
+	unsigned long mask = (0x7f << ATA_SHIFT_UDMA) | (0x7 << ATA_SHIFT_MWDMA) | (0x1F << ATA_SHIFT_PIO);
+
+	/* we always use the 0 slot for crap hardware */
+	unit = dev->devno;
+	if (!(gtm->flags & 0x10))
+		unit = 0;
+
+	/* start by scanning for PIO modes */
+	for (i = 0; i < 7; i++) {
+		t = gtm->drive[unit].pio;
+		if (t <= ata_acpi_pio_cycle[i]) {
+			mask |= (2 << (ATA_SHIFT_PIO + i)) - 1;
+			break;
+		}
+	}
+
+	/* See if we have MWDMA or UDMA data. We don't bother with
+	 * MWDMA if UDMA is available as this means the BIOS set UDMA
+	 * and our error changedown if it works is UDMA to PIO anyway.
+	 */
+	if (gtm->flags & (1 << (2 * unit))) {
+		/* MWDMA */
+		for (i = 0; i < 5; i++) {
+			t = gtm->drive[unit].dma;
+			if (t <= ata_acpi_mwdma_cycle[i]) {
+				mask |= (2 << (ATA_SHIFT_MWDMA + i)) - 1;
+				break;
+			}
+		}
+	} else {
+		/* UDMA */
+		for (i = 0; i < 7; i++) {
+			t = gtm->drive[unit].dma;
+			if (t <= ata_acpi_udma_cycle[i]) {
+				mask |= (2 << (ATA_SHIFT_UDMA + i)) - 1;
+				break;
+			}
+		}
+	}
+
+	return mask;
+}
+EXPORT_SYMBOL_GPL(ata_acpi_gtm_xfermask);
+
 /**
  * ata_acpi_cbl_80wire		-	Check for 80 wire cable
  * @ap: Port to check
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index e4542ab9c7f8..a4737a3d31cb 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -81,17 +81,6 @@ static void pacpi_error_handler(struct ata_port *ap)
 				  NULL, ata_std_postreset);
 }
 
-/* Welcome to ACPI, bring a bucket */
-static const unsigned int pio_cycle[7] = {
-	600, 383, 240, 180, 120, 100, 80
-};
-static const unsigned int mwdma_cycle[5] = {
-	480, 150, 120, 100, 80
-};
-static const unsigned int udma_cycle[7] = {
-	120, 80, 60, 45, 30, 20, 15
-};
-
 /**
  *	pacpi_discover_modes	-	filter non ACPI modes
  *	@adev: ATA device
@@ -103,56 +92,20 @@ static const unsigned int udma_cycle[7] = {
 
 static unsigned long pacpi_discover_modes(struct ata_port *ap, struct ata_device *adev)
 {
-	int unit = adev->devno;
 	struct pata_acpi *acpi = ap->private_data;
-	int i;
-	u32 t;
-	unsigned long mask = (0x7f << ATA_SHIFT_UDMA) | (0x7 << ATA_SHIFT_MWDMA) | (0x1F << ATA_SHIFT_PIO);
-
 	struct ata_acpi_gtm probe;
+	unsigned int xfer_mask;
 
 	probe = acpi->gtm;
 
-	/* We always use the 0 slot for crap hardware */
-	if (!(probe.flags & 0x10))
-		unit = 0;
-
 	ata_acpi_gtm(ap, &probe);
 
-	/* Start by scanning for PIO modes */
-	for (i = 0; i < 7; i++) {
-		t = probe.drive[unit].pio;
-		if (t <= pio_cycle[i]) {
-			mask |= (2 << (ATA_SHIFT_PIO + i)) - 1;
-			break;
-		}
-	}
+	xfer_mask = ata_acpi_gtm_xfermask(adev, &probe);
 
-	/* See if we have MWDMA or UDMA data. We don't bother with MWDMA
-	   if UDMA is availabe as this means the BIOS set UDMA and our
-	   error changedown if it works is UDMA to PIO anyway */
-	if (probe.flags & (1 << (2 * unit))) {
-		/* MWDMA */
-		for (i = 0; i < 5; i++) {
-			t = probe.drive[unit].dma;
-			if (t <= mwdma_cycle[i]) {
-				mask |= (2 << (ATA_SHIFT_MWDMA + i)) - 1;
-				break;
-			}
-		}
-	} else {
-		/* UDMA */
-		for (i = 0; i < 7; i++) {
-			t = probe.drive[unit].dma;
-			if (t <= udma_cycle[i]) {
-				mask |= (2 << (ATA_SHIFT_UDMA + i)) - 1;
-				break;
-			}
-		}
-	}
-	if (mask & (0xF8 << ATA_SHIFT_UDMA))
+	if (xfer_mask & (0xF8 << ATA_SHIFT_UDMA))
 		ap->cbl = ATA_CBL_PATA80;
-	return mask;
+
+	return xfer_mask;
 }
 
 /**
@@ -185,7 +138,8 @@ static void pacpi_set_piomode(struct ata_port *ap, struct ata_device *adev)
 		unit = 0;
 
 	/* Now stuff the nS values into the structure */
-	acpi->gtm.drive[unit].pio = pio_cycle[adev->pio_mode - XFER_PIO_0];
+	acpi->gtm.drive[unit].pio =
+		ata_acpi_pio_cycle[adev->pio_mode - XFER_PIO_0];
 	ata_acpi_stm(ap, &acpi->gtm);
 	/* See what mode we actually got */
 	ata_acpi_gtm(ap, &acpi->gtm);
@@ -207,10 +161,12 @@ static void pacpi_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 
 	/* Now stuff the nS values into the structure */
 	if (adev->dma_mode >= XFER_UDMA_0) {
-		acpi->gtm.drive[unit].dma = udma_cycle[adev->dma_mode - XFER_UDMA_0];
+		acpi->gtm.drive[unit].dma =
+			ata_acpi_udma_cycle[adev->dma_mode - XFER_UDMA_0];
 		acpi->gtm.flags |= (1 << (2 * unit));
 	} else {
-		acpi->gtm.drive[unit].dma = mwdma_cycle[adev->dma_mode - XFER_MW_DMA_0];
+		acpi->gtm.drive[unit].dma =
+			ata_acpi_mwdma_cycle[adev->dma_mode - XFER_MW_DMA_0];
 		acpi->gtm.flags &= ~(1 << (2 * unit));
 	}
 	ata_acpi_stm(ap, &acpi->gtm);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index bdb7c6e13993..8022e5b2224d 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -961,6 +961,10 @@ enum {
 
 /* libata-acpi.c */
 #ifdef CONFIG_ATA_ACPI
+extern const unsigned int ata_acpi_pio_cycle[7];
+extern const unsigned int ata_acpi_mwdma_cycle[5];
+extern const unsigned int ata_acpi_udma_cycle[7];
+
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
 	if (ap->pflags & ATA_PFLAG_INIT_GTM_VALID)
@@ -970,12 +974,33 @@ static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 extern int ata_acpi_cbl_80wire(struct ata_port *ap);
 int ata_acpi_stm(struct ata_port *ap, const struct ata_acpi_gtm *stm);
 int ata_acpi_gtm(struct ata_port *ap, struct ata_acpi_gtm *stm);
+unsigned long ata_acpi_gtm_xfermask(struct ata_device *dev,
+				    const struct ata_acpi_gtm *gtm);
+
 #else
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
 	return NULL;
 }
 static inline int ata_acpi_cbl_80wire(struct ata_port *ap) { return 0; }
+
+static inline int ata_acpi_stm(const struct ata_port *ap,
+			       struct ata_acpi_gtm *stm)
+{
+	return -ENOSYS;
+}
+
+static inline int ata_acpi_gtm(const struct ata_port *ap,
+			       struct ata_acpi_gtm *stm)
+{
+	return -ENOSYS;
+}
+
+static inline unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev,
+					const struct ata_acpi_gtm *gtm)
+{
+	return 0;
+}
 #endif
 
 #ifdef CONFIG_PCI
-- 
cgit v1.2.3


From a0f79b929acaba10d4780acd2543eff20bf4b5b0 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 18 Dec 2007 16:33:05 +0900
Subject: libata: implement ata_timing_cycle2mode() and use it in libata-acpi
 and pata_acpi

libata-acpi is using separate timing tables for transfer modes
although libata-core has the complete ata_timing table.  Implement
ata_timing_cycle2mode() to look for matching mode given transfer type
and cycle duration and use it in libata-acpi and pata_acpi to replace
private timing tables.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-acpi.c | 62 ++++++++++++-----------------------------------
 drivers/ata/libata-core.c | 52 +++++++++++++++++++++++++++++++++++++++
 drivers/ata/pata_acpi.c   | 13 +++++-----
 include/linux/libata.h    |  5 +---
 4 files changed, 75 insertions(+), 57 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index 9f0b208afcf5..a6f1a6b56d8c 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -441,22 +441,6 @@ static int ata_dev_get_GTF(struct ata_device *dev, struct ata_acpi_gtf **gtf)
 	return rc;
 }
 
-/* Welcome to ACPI, bring a bucket */
-const unsigned int ata_acpi_pio_cycle[7] = {
-	600, 383, 240, 180, 120, 100, 80
-};
-EXPORT_SYMBOL_GPL(ata_acpi_pio_cycle);
-
-const unsigned int ata_acpi_mwdma_cycle[5] = {
-	480, 150, 120, 100, 80
-};
-EXPORT_SYMBOL_GPL(ata_acpi_mwdma_cycle);
-
-const unsigned int ata_acpi_udma_cycle[7] = {
-	120, 80, 60, 45, 30, 20, 15
-};
-EXPORT_SYMBOL_GPL(ata_acpi_udma_cycle);
-
 /**
  * ata_acpi_gtm_xfermode - determine xfermode from GTM parameter
  * @dev: target device
@@ -473,49 +457,33 @@ EXPORT_SYMBOL_GPL(ata_acpi_udma_cycle);
 unsigned long ata_acpi_gtm_xfermask(struct ata_device *dev,
 				    const struct ata_acpi_gtm *gtm)
 {
-	unsigned long pio_mask = 0, mwdma_mask = 0, udma_mask = 0;
-	int unit, i;
-	u32 t;
+	unsigned long xfer_mask = 0;
+	unsigned int type;
+	int unit;
+	u8 mode;
 
 	/* we always use the 0 slot for crap hardware */
 	unit = dev->devno;
 	if (!(gtm->flags & 0x10))
 		unit = 0;
 
-	/* Values larger than the longest cycle results in 0 mask
-	 * while values equal to smaller than the shortest cycle
-	 * results in mask which includes all supported modes.
-	 * Disabled transfer method has the value of 0xffffffff which
-	 * will always result in 0 mask.
-	 */
-
-	/* start by scanning for PIO modes */
-	t = gtm->drive[unit].pio;
-	for (i = 0; i < ARRAY_SIZE(ata_acpi_pio_cycle); i++)
-		if (t > ata_acpi_pio_cycle[i])
-			break;
-	pio_mask = (1 << i) - 1;
+	/* PIO */
+	mode = ata_timing_cycle2mode(ATA_SHIFT_PIO, gtm->drive[unit].pio);
+	xfer_mask |= ata_xfer_mode2mask(mode);
 
 	/* See if we have MWDMA or UDMA data. We don't bother with
 	 * MWDMA if UDMA is available as this means the BIOS set UDMA
 	 * and our error changedown if it works is UDMA to PIO anyway.
 	 */
-	t = gtm->drive[unit].dma;
-	if (!(gtm->flags & (1 << (2 * unit)))) {
-		/* MWDMA */
-		for (i = 0; i < ARRAY_SIZE(ata_acpi_mwdma_cycle); i++)
-			if (t > ata_acpi_mwdma_cycle[i])
-				break;
-		mwdma_mask = (1 << i) - 1;
-	} else {
-		/* UDMA */
-		for (i = 0; i < ARRAY_SIZE(ata_acpi_udma_cycle); i++)
-			if (t > ata_acpi_udma_cycle[i])
-				break;
-		udma_mask = (1 << i) - 1;
-	}
+	if (!(gtm->flags & (1 << (2 * unit))))
+		type = ATA_SHIFT_MWDMA;
+	else
+		type = ATA_SHIFT_UDMA;
+
+	mode = ata_timing_cycle2mode(type, gtm->drive[unit].dma);
+	xfer_mask |= ata_xfer_mode2mask(mode);
 
-	return ata_pack_xfermask(pio_mask, mwdma_mask, udma_mask);
+	return xfer_mask;
 }
 EXPORT_SYMBOL_GPL(ata_acpi_gtm_xfermask);
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 62c4b328b0b6..f01c1734b1d0 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -2902,6 +2902,57 @@ int ata_timing_compute(struct ata_device *adev, unsigned short speed,
 	return 0;
 }
 
+/**
+ *	ata_timing_cycle2mode - find xfer mode for the specified cycle duration
+ *	@xfer_shift: ATA_SHIFT_* value for transfer type to examine.
+ *	@cycle: cycle duration in ns
+ *
+ *	Return matching xfer mode for @cycle.  The returned mode is of
+ *	the transfer type specified by @xfer_shift.  If @cycle is too
+ *	slow for @xfer_shift, 0xff is returned.  If @cycle is faster
+ *	than the fastest known mode, the fasted mode is returned.
+ *
+ *	LOCKING:
+ *	None.
+ *
+ *	RETURNS:
+ *	Matching xfer_mode, 0xff if no match found.
+ */
+u8 ata_timing_cycle2mode(unsigned int xfer_shift, int cycle)
+{
+	u8 base_mode = 0xff, last_mode = 0xff;
+	const struct ata_xfer_ent *ent;
+	const struct ata_timing *t;
+
+	for (ent = ata_xfer_tbl; ent->shift >= 0; ent++)
+		if (ent->shift == xfer_shift)
+			base_mode = ent->base;
+
+	for (t = ata_timing_find_mode(base_mode);
+	     t && ata_xfer_mode2shift(t->mode) == xfer_shift; t++) {
+		unsigned short this_cycle;
+
+		switch (xfer_shift) {
+		case ATA_SHIFT_PIO:
+		case ATA_SHIFT_MWDMA:
+			this_cycle = t->cycle;
+			break;
+		case ATA_SHIFT_UDMA:
+			this_cycle = t->udma;
+			break;
+		default:
+			return 0xff;
+		}
+
+		if (cycle > this_cycle)
+			break;
+
+		last_mode = t->mode;
+	}
+
+	return last_mode;
+}
+
 /**
  *	ata_down_xfermask_limit - adjust dev xfer masks downward
  *	@dev: Device to adjust xfer masks
@@ -7630,6 +7681,7 @@ EXPORT_SYMBOL_GPL(ata_pio_need_iordy);
 EXPORT_SYMBOL_GPL(ata_timing_find_mode);
 EXPORT_SYMBOL_GPL(ata_timing_compute);
 EXPORT_SYMBOL_GPL(ata_timing_merge);
+EXPORT_SYMBOL_GPL(ata_timing_cycle2mode);
 
 #ifdef CONFIG_PCI
 EXPORT_SYMBOL_GPL(pci_test_config_bits);
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index a4737a3d31cb..244098a80ce4 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -133,13 +133,14 @@ static void pacpi_set_piomode(struct ata_port *ap, struct ata_device *adev)
 {
 	int unit = adev->devno;
 	struct pata_acpi *acpi = ap->private_data;
+	const struct ata_timing *t;
 
 	if (!(acpi->gtm.flags & 0x10))
 		unit = 0;
 
 	/* Now stuff the nS values into the structure */
-	acpi->gtm.drive[unit].pio =
-		ata_acpi_pio_cycle[adev->pio_mode - XFER_PIO_0];
+	t = ata_timing_find_mode(adev->pio_mode);
+	acpi->gtm.drive[unit].pio = t->cycle;
 	ata_acpi_stm(ap, &acpi->gtm);
 	/* See what mode we actually got */
 	ata_acpi_gtm(ap, &acpi->gtm);
@@ -155,18 +156,18 @@ static void pacpi_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 {
 	int unit = adev->devno;
 	struct pata_acpi *acpi = ap->private_data;
+	const struct ata_timing *t;
 
 	if (!(acpi->gtm.flags & 0x10))
 		unit = 0;
 
 	/* Now stuff the nS values into the structure */
+	t = ata_timing_find_mode(adev->dma_mode);
 	if (adev->dma_mode >= XFER_UDMA_0) {
-		acpi->gtm.drive[unit].dma =
-			ata_acpi_udma_cycle[adev->dma_mode - XFER_UDMA_0];
+		acpi->gtm.drive[unit].dma = t->udma;
 		acpi->gtm.flags |= (1 << (2 * unit));
 	} else {
-		acpi->gtm.drive[unit].dma =
-			ata_acpi_mwdma_cycle[adev->dma_mode - XFER_MW_DMA_0];
+		acpi->gtm.drive[unit].dma = t->cycle;
 		acpi->gtm.flags &= ~(1 << (2 * unit));
 	}
 	ata_acpi_stm(ap, &acpi->gtm);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 8022e5b2224d..8ede93b5c7a6 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -941,6 +941,7 @@ extern int ata_timing_compute(struct ata_device *, unsigned short,
 extern void ata_timing_merge(const struct ata_timing *,
 			     const struct ata_timing *, struct ata_timing *,
 			     unsigned int);
+extern u8 ata_timing_cycle2mode(unsigned int xfer_shift, int cycle);
 
 enum {
 	ATA_TIMING_SETUP	= (1 << 0),
@@ -961,10 +962,6 @@ enum {
 
 /* libata-acpi.c */
 #ifdef CONFIG_ATA_ACPI
-extern const unsigned int ata_acpi_pio_cycle[7];
-extern const unsigned int ata_acpi_mwdma_cycle[5];
-extern const unsigned int ata_acpi_udma_cycle[7];
-
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
 	if (ap->pflags & ATA_PFLAG_INIT_GTM_VALID)
-- 
cgit v1.2.3


From 021ee9a6da1cfc57f6a6c769c3c898bdd4753108 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 18 Dec 2007 16:33:06 +0900
Subject: libata: reimplement ata_acpi_cbl_80wire() using
 ata_acpi_gtm_xfermask()

Reimplement ata_acpi_cbl_80wire() using ata_acpi_gtm_xfermask() and
while at it relocate the function below ata_acpi_gtm_xfermask().

New ata_acpi_cbl_80wire() implementation takes @gtm, in both pata_via
and pata_amd, use the initial GTM value.  Both are trying to peek
initial BIOS configuration, so using initial caching value makes
sense.  This fixes ACPI part of cable detection in pata_amd which
previously always returned 0 because configuring PIO0 during reset
clears DMA configuration.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-acpi.c | 41 ++++++++++++++++-------------------------
 drivers/ata/pata_amd.c    |  3 ++-
 drivers/ata/pata_via.c    |  3 ++-
 include/linux/libata.h    | 10 +++++++---
 4 files changed, 27 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index a6f1a6b56d8c..9e8ec19260af 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -490,38 +490,29 @@ EXPORT_SYMBOL_GPL(ata_acpi_gtm_xfermask);
 /**
  * ata_acpi_cbl_80wire		-	Check for 80 wire cable
  * @ap: Port to check
+ * @gtm: GTM data to use
  *
- * Return 1 if the ACPI mode data for this port indicates the BIOS selected
- * an 80wire mode.
+ * Return 1 if the @gtm indicates the BIOS selected an 80wire mode.
  */
-
-int ata_acpi_cbl_80wire(struct ata_port *ap)
+int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm)
 {
-	const struct ata_acpi_gtm *gtm = ata_acpi_init_gtm(ap);
-	int valid = 0;
+	struct ata_device *dev;
 
-	if (!gtm)
-		return 0;
+	ata_link_for_each_dev(dev, &ap->link) {
+		unsigned long xfer_mask, udma_mask;
+
+		if (!ata_dev_enabled(dev))
+			continue;
+
+		xfer_mask = ata_acpi_gtm_xfermask(dev, gtm);
+		ata_unpack_xfermask(xfer_mask, NULL, NULL, &udma_mask);
+
+		if (udma_mask & ~ATA_UDMA_MASK_40C)
+			return 1;
+	}
 
-	/* Split timing, DMA enabled */
-	if ((gtm->flags & 0x11) == 0x11 && gtm->drive[0].dma < 55)
-		valid |= 1;
-	if ((gtm->flags & 0x14) == 0x14 && gtm->drive[1].dma < 55)
-		valid |= 2;
-	/* Shared timing, DMA enabled */
-	if ((gtm->flags & 0x11) == 0x01 && gtm->drive[0].dma < 55)
-		valid |= 1;
-	if ((gtm->flags & 0x14) == 0x04 && gtm->drive[0].dma < 55)
-		valid |= 2;
-
-	/* Drive check */
-	if ((valid & 1) && ata_dev_enabled(&ap->link.device[0]))
-		return 1;
-	if ((valid & 2) && ata_dev_enabled(&ap->link.device[1]))
-		return 1;
 	return 0;
 }
-
 EXPORT_SYMBOL_GPL(ata_acpi_cbl_80wire);
 
 static void ata_acpi_gtf_to_tf(struct ata_device *dev,
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 3cc27b514654..e71125a4bd9b 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -272,7 +272,8 @@ static int nv_cable_detect(struct ata_port *ap)
  	if ((udma & 0xC4) == 0xC4 || (udma & 0xC400) == 0xC400)
 		cbl = ATA_CBL_PATA80;
 	/* And a triple check across suspend/resume with ACPI around */
-	if (ata_acpi_cbl_80wire(ap))
+	if (ata_acpi_init_gtm(ap) &&
+	    ata_acpi_cbl_80wire(ap, ata_acpi_init_gtm(ap)))
 		cbl = ATA_CBL_PATA80;
 	return cbl;
 }
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 453d72bf2598..39627ab684bf 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -185,7 +185,8 @@ static int via_cable_detect(struct ata_port *ap) {
 	if (ata66 & (0x10100000 >> (16 * ap->port_no)))
 		return ATA_CBL_PATA80;
 	/* Check with ACPI so we can spot BIOS reported SATA bridges */
-	if (ata_acpi_cbl_80wire(ap))
+	if (ata_acpi_init_gtm(ap) &&
+	    ata_acpi_cbl_80wire(ap, ata_acpi_init_gtm(ap)))
 		return ATA_CBL_PATA80;
 	return ATA_CBL_PATA40;
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 8ede93b5c7a6..cc4eaef6f889 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -968,18 +968,16 @@ static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 		return &ap->__acpi_init_gtm;
 	return NULL;
 }
-extern int ata_acpi_cbl_80wire(struct ata_port *ap);
 int ata_acpi_stm(struct ata_port *ap, const struct ata_acpi_gtm *stm);
 int ata_acpi_gtm(struct ata_port *ap, struct ata_acpi_gtm *stm);
 unsigned long ata_acpi_gtm_xfermask(struct ata_device *dev,
 				    const struct ata_acpi_gtm *gtm);
-
+int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm);
 #else
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
 	return NULL;
 }
-static inline int ata_acpi_cbl_80wire(struct ata_port *ap) { return 0; }
 
 static inline int ata_acpi_stm(const struct ata_port *ap,
 			       struct ata_acpi_gtm *stm)
@@ -998,6 +996,12 @@ static inline unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev,
 {
 	return 0;
 }
+
+static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
+				      const struct ata_acpi_gtm *gtm)
+{
+	return 0;
+}
 #endif
 
 #ifdef CONFIG_PCI
-- 
cgit v1.2.3


From 537b53c1692960b8b3b0324e886fbe48cb9e5c00 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:04 +0900
Subject: cdrom: add more GPCMD_* constants

Add GPCMD_* constants for READ_BUFFER, WRITE_12 and WRITE_BUFFER for
completeness.  These will be used libata.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/linux/cdrom.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/linux/cdrom.h b/include/linux/cdrom.h
index c6d3e22c0624..fcdc11b9609b 100644
--- a/include/linux/cdrom.h
+++ b/include/linux/cdrom.h
@@ -451,6 +451,7 @@ struct cdrom_generic_command
 #define GPCMD_PREVENT_ALLOW_MEDIUM_REMOVAL  0x1e
 #define GPCMD_READ_10			    0x28
 #define GPCMD_READ_12			    0xa8
+#define GPCMD_READ_BUFFER		    0x3c
 #define GPCMD_READ_BUFFER_CAPACITY	    0x5c
 #define GPCMD_READ_CDVD_CAPACITY	    0x25
 #define GPCMD_READ_CD			    0xbe
@@ -480,7 +481,9 @@ struct cdrom_generic_command
 #define GPCMD_TEST_UNIT_READY		    0x00
 #define GPCMD_VERIFY_10			    0x2f
 #define GPCMD_WRITE_10			    0x2a
+#define GPCMD_WRITE_12			    0xaa
 #define GPCMD_WRITE_AND_VERIFY_10	    0x2e
+#define GPCMD_WRITE_BUFFER		    0x3b
 /* This is listed as optional in ATAPI 2.6, but is (curiously) 
  * missing from Mt. Fuji, Table 57.  It _is_ mentioned in Mt. Fuji
  * Table 377 as an MMC command for SCSi devices though...  Most ATAPI
-- 
cgit v1.2.3


From 0dc36888d4422140f9eaf50f24953ec109f750a3 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 18 Dec 2007 16:34:43 -0500
Subject: libata: rename ATA_PROT_ATAPI_* to ATAPI_PROT_*

ATA_PROT_ATAPI_* are ugly and naming schemes between ATA_PROT_* and
ATA_PROT_ATAPI_* are inconsistent causing confusion.  Rename them to
ATAPI_PROT_* and make them consistent with ATA counterpart.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c       | 28 ++++++++++++++--------------
 drivers/ata/libata-eh.c         |  8 ++++----
 drivers/ata/libata-scsi.c       | 10 +++++-----
 drivers/ata/libata-sff.c        |  2 +-
 drivers/ata/pata_pdc202xx_old.c |  5 ++---
 drivers/ata/pdc_adma.c          |  2 +-
 drivers/ata/sata_inic162x.c     |  2 +-
 drivers/ata/sata_promise.c      | 26 ++++++++++++--------------
 drivers/ata/sata_qstor.c        |  2 +-
 drivers/ata/sata_sx4.c          |  2 +-
 drivers/scsi/ipr.c              |  6 +++---
 drivers/scsi/libsas/sas_ata.c   |  2 +-
 include/linux/ata.h             | 12 ++++++------
 13 files changed, 52 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index c9e6bd4d0686..2c9745a74d50 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4678,8 +4678,8 @@ int ata_check_atapi_dma(struct ata_queued_cmd *qc)
  */
 static int atapi_qc_may_overflow(struct ata_queued_cmd *qc)
 {
-	if (qc->tf.protocol != ATA_PROT_ATAPI &&
-	    qc->tf.protocol != ATA_PROT_ATAPI_DMA)
+	if (qc->tf.protocol != ATAPI_PROT_PIO &&
+	    qc->tf.protocol != ATAPI_PROT_DMA)
 		return 0;
 
 	if (qc->tf.flags & ATA_TFLAG_WRITE)
@@ -5161,13 +5161,13 @@ static void atapi_send_cdb(struct ata_port *ap, struct ata_queued_cmd *qc)
 	ata_altstatus(ap); /* flush */
 
 	switch (qc->tf.protocol) {
-	case ATA_PROT_ATAPI:
+	case ATAPI_PROT_PIO:
 		ap->hsm_task_state = HSM_ST;
 		break;
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_NODATA:
 		ap->hsm_task_state = HSM_ST_LAST;
 		break;
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		ap->hsm_task_state = HSM_ST_LAST;
 		/* initiate bmdma */
 		ap->ops->bmdma_start(qc);
@@ -5518,7 +5518,7 @@ fsm_start:
 
 	case HSM_ST:
 		/* complete command or read/write the data register */
-		if (qc->tf.protocol == ATA_PROT_ATAPI) {
+		if (qc->tf.protocol == ATAPI_PROT_PIO) {
 			/* ATAPI PIO protocol */
 			if ((status & ATA_DRQ) == 0) {
 				/* No more data to transfer or device error.
@@ -6073,11 +6073,11 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 		switch (qc->tf.protocol) {
 		case ATA_PROT_PIO:
 		case ATA_PROT_NODATA:
-		case ATA_PROT_ATAPI:
-		case ATA_PROT_ATAPI_NODATA:
+		case ATAPI_PROT_PIO:
+		case ATAPI_PROT_NODATA:
 			qc->tf.flags |= ATA_TFLAG_POLLING;
 			break;
-		case ATA_PROT_ATAPI_DMA:
+		case ATAPI_PROT_DMA:
 			if (qc->dev->flags & ATA_DFLAG_CDB_INTR)
 				/* see ata_dma_blacklisted() */
 				BUG();
@@ -6141,8 +6141,8 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 
 		break;
 
-	case ATA_PROT_ATAPI:
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_PIO:
+	case ATAPI_PROT_NODATA:
 		if (qc->tf.flags & ATA_TFLAG_POLLING)
 			ata_qc_set_polling(qc);
 
@@ -6156,7 +6156,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 			ata_port_queue_task(ap, ata_pio_task, qc, 0);
 		break;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
 
 		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
@@ -6217,7 +6217,7 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 		break;
 	case HSM_ST_LAST:
 		if (qc->tf.protocol == ATA_PROT_DMA ||
-		    qc->tf.protocol == ATA_PROT_ATAPI_DMA) {
+		    qc->tf.protocol == ATAPI_PROT_DMA) {
 			/* check status of DMA engine */
 			host_stat = ap->ops->bmdma_status(ap);
 			VPRINTK("ata%u: host_stat 0x%X\n",
@@ -6259,7 +6259,7 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 	ata_hsm_move(ap, qc, status, 0);
 
 	if (unlikely(qc->err_mask) && (qc->tf.protocol == ATA_PROT_DMA ||
-				       qc->tf.protocol == ATA_PROT_ATAPI_DMA))
+				       qc->tf.protocol == ATAPI_PROT_DMA))
 		ata_ehi_push_desc(ehi, "BMDMA stat 0x%x", host_stat);
 
 	return 1;	/* irq handled */
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 1bc1acf3bbbb..419552603a16 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1299,10 +1299,10 @@ static unsigned int atapi_eh_request_sense(struct ata_queued_cmd *qc)
 
 	/* is it pointless to prefer PIO for "safety reasons"? */
 	if (ap->flags & ATA_FLAG_PIO_DMA) {
-		tf.protocol = ATA_PROT_ATAPI_DMA;
+		tf.protocol = ATAPI_PROT_DMA;
 		tf.feature |= ATAPI_PKT_DMA;
 	} else {
-		tf.protocol = ATA_PROT_ATAPI;
+		tf.protocol = ATAPI_PROT_PIO;
 		tf.lbam = SCSI_SENSE_BUFFERSIZE;
 		tf.lbah = 0;
 	}
@@ -1979,8 +1979,8 @@ static void ata_eh_link_report(struct ata_link *link)
 				[ATA_PROT_PIO]		= "pio",
 				[ATA_PROT_DMA]		= "dma",
 				[ATA_PROT_NCQ]		= "ncq",
-				[ATA_PROT_ATAPI]	= "pio",
-				[ATA_PROT_ATAPI_DMA]	= "dma",
+				[ATAPI_PROT_PIO]	= "pio",
+				[ATAPI_PROT_DMA]	= "dma",
 			};
 
 			snprintf(data_buf, sizeof(data_buf), " %s %u %s",
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 021cdc4cc260..5fd780e509d4 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2354,10 +2354,10 @@ static void atapi_request_sense(struct ata_queued_cmd *qc)
 	qc->tf.command = ATA_CMD_PACKET;
 
 	if (ata_pio_use_silly(ap)) {
-		qc->tf.protocol = ATA_PROT_ATAPI_DMA;
+		qc->tf.protocol = ATAPI_PROT_DMA;
 		qc->tf.feature |= ATAPI_PKT_DMA;
 	} else {
-		qc->tf.protocol = ATA_PROT_ATAPI;
+		qc->tf.protocol = ATAPI_PROT_PIO;
 		qc->tf.lbam = SCSI_SENSE_BUFFERSIZE;
 		qc->tf.lbah = 0;
 	}
@@ -2528,12 +2528,12 @@ static unsigned int atapi_xlat(struct ata_queued_cmd *qc)
 	if (using_pio || nodata) {
 		/* no data, or PIO data xfer */
 		if (nodata)
-			qc->tf.protocol = ATA_PROT_ATAPI_NODATA;
+			qc->tf.protocol = ATAPI_PROT_NODATA;
 		else
-			qc->tf.protocol = ATA_PROT_ATAPI;
+			qc->tf.protocol = ATAPI_PROT_PIO;
 	} else {
 		/* DMA data xfer */
-		qc->tf.protocol = ATA_PROT_ATAPI_DMA;
+		qc->tf.protocol = ATAPI_PROT_DMA;
 		qc->tf.feature |= ATAPI_PKT_DMA;
 
 		if (atapi_dmadir && (scmd->sc_data_direction != DMA_TO_DEVICE))
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index fd5fe4e7e75a..edeb4bea586c 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -417,7 +417,7 @@ void ata_bmdma_drive_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
 	ap->hsm_task_state = HSM_ST_IDLE;
 
 	if (qc && (qc->tf.protocol == ATA_PROT_DMA ||
-		   qc->tf.protocol == ATA_PROT_ATAPI_DMA)) {
+		   qc->tf.protocol == ATAPI_PROT_DMA)) {
 		u8 host_stat;
 
 		host_stat = ap->ops->bmdma_status(ap);
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 6c9689b59b06..3ed866723e0c 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -168,8 +168,7 @@ static void pdc2026x_bmdma_start(struct ata_queued_cmd *qc)
 	pdc202xx_set_dmamode(ap, qc->dev);
 
 	/* Cases the state machine will not complete correctly without help */
-	if ((tf->flags & ATA_TFLAG_LBA48) ||  tf->protocol == ATA_PROT_ATAPI_DMA)
-	{
+	if ((tf->flags & ATA_TFLAG_LBA48) ||  tf->protocol == ATAPI_PROT_DMA) {
 		len = qc->nbytes / 2;
 
 		if (tf->flags & ATA_TFLAG_WRITE)
@@ -208,7 +207,7 @@ static void pdc2026x_bmdma_stop(struct ata_queued_cmd *qc)
 	void __iomem *atapi_reg = master + 0x20 + (4 * ap->port_no);
 
 	/* Cases the state machine will not complete correctly */
-	if (tf->protocol == ATA_PROT_ATAPI_DMA || ( tf->flags & ATA_TFLAG_LBA48)) {
+	if (tf->protocol == ATAPI_PROT_DMA || (tf->flags & ATA_TFLAG_LBA48)) {
 		iowrite32(0, atapi_reg);
 		iowrite8(ioread8(clock) & ~sel66, clock);
 	}
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index bd4c2a3c88d7..459cb7bb7d7f 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -455,7 +455,7 @@ static unsigned int adma_qc_issue(struct ata_queued_cmd *qc)
 		adma_packet_start(qc);
 		return 0;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		BUG();
 		break;
 
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 323c087e8cc1..96e614a1c169 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -585,7 +585,7 @@ static struct ata_port_operations inic_port_ops = {
 };
 
 static struct ata_port_info inic_port_info = {
-	/* For some reason, ATA_PROT_ATAPI is broken on this
+	/* For some reason, ATAPI_PROT_PIO is broken on this
 	 * controller, and no, PIO_POLLING does't fix it.  It somehow
 	 * manages to report the wrong ireason and ignoring ireason
 	 * results in machine lock up.  Tell libata to always prefer
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 9638faaa8111..01738d736d44 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -456,13 +456,13 @@ static void pdc_atapi_pkt(struct ata_queued_cmd *qc)
 	 * and seq id (byte 2)
 	 */
 	switch (qc->tf.protocol) {
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		if (!(qc->tf.flags & ATA_TFLAG_WRITE))
 			buf32[0] = cpu_to_le32(PDC_PKT_READ);
 		else
 			buf32[0] = 0;
 		break;
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_NODATA:
 		buf32[0] = cpu_to_le32(PDC_PKT_NODATA);
 		break;
 	default:
@@ -489,7 +489,7 @@ static void pdc_atapi_pkt(struct ata_queued_cmd *qc)
 	buf[19] = qc->tf.lbal;
 
 	/* set feature and byte counter registers */
-	if (qc->tf.protocol != ATA_PROT_ATAPI_DMA)
+	if (qc->tf.protocol != ATAPI_PROT_DMA)
 		feature = PDC_FEATURE_ATAPI_PIO;
 	else
 		feature = PDC_FEATURE_ATAPI_DMA;
@@ -619,14 +619,14 @@ static void pdc_qc_prep(struct ata_queued_cmd *qc)
 		pdc_pkt_footer(&qc->tf, pp->pkt, i);
 		break;
 
-	case ATA_PROT_ATAPI:
+	case ATAPI_PROT_PIO:
 		pdc_fill_sg(qc);
 		break;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		pdc_fill_sg(qc);
 		/*FALLTHROUGH*/
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_NODATA:
 		pdc_atapi_pkt(qc);
 		break;
 
@@ -746,8 +746,8 @@ static inline unsigned int pdc_host_intr(struct ata_port *ap,
 	switch (qc->tf.protocol) {
 	case ATA_PROT_DMA:
 	case ATA_PROT_NODATA:
-	case ATA_PROT_ATAPI_DMA:
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_DMA:
+	case ATAPI_PROT_NODATA:
 		qc->err_mask |= ac_err_mask(ata_wait_idle(ap));
 		ata_qc_complete(qc);
 		handled = 1;
@@ -892,7 +892,7 @@ static inline void pdc_packet_start(struct ata_queued_cmd *qc)
 static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc)
 {
 	switch (qc->tf.protocol) {
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_NODATA:
 		if (qc->dev->flags & ATA_DFLAG_CDB_INTR)
 			break;
 		/*FALLTHROUGH*/
@@ -900,7 +900,7 @@ static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc)
 		if (qc->tf.flags & ATA_TFLAG_POLLING)
 			break;
 		/*FALLTHROUGH*/
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 	case ATA_PROT_DMA:
 		pdc_packet_start(qc);
 		return 0;
@@ -914,16 +914,14 @@ static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc)
 
 static void pdc_tf_load_mmio(struct ata_port *ap, const struct ata_taskfile *tf)
 {
-	WARN_ON(tf->protocol == ATA_PROT_DMA ||
-		tf->protocol == ATA_PROT_ATAPI_DMA);
+	WARN_ON(tf->protocol == ATA_PROT_DMA || tf->protocol == ATAPI_PROT_DMA);
 	ata_tf_load(ap, tf);
 }
 
 static void pdc_exec_command_mmio(struct ata_port *ap,
 				  const struct ata_taskfile *tf)
 {
-	WARN_ON(tf->protocol == ATA_PROT_DMA ||
-		tf->protocol == ATA_PROT_ATAPI_DMA);
+	WARN_ON(tf->protocol == ATA_PROT_DMA || tf->protocol == ATAPI_PROT_DMA);
 	ata_exec_command(ap, tf);
 }
 
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index c68b241805fd..4e5f07bdd069 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -376,7 +376,7 @@ static unsigned int qs_qc_issue(struct ata_queued_cmd *qc)
 		qs_packet_start(qc);
 		return 0;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		BUG();
 		break;
 
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 4d857185f33b..3de0c27caf53 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -700,7 +700,7 @@ static unsigned int pdc20621_qc_issue_prot(struct ata_queued_cmd *qc)
 		pdc20621_packet_start(qc);
 		return 0;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		BUG();
 		break;
 
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 0841df01bc19..3e78bc2d9170 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -5222,12 +5222,12 @@ static unsigned int ipr_qc_issue(struct ata_queued_cmd *qc)
 		regs->flags |= IPR_ATA_FLAG_XFER_TYPE_DMA;
 		break;
 
-	case ATA_PROT_ATAPI:
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_PIO:
+	case ATAPI_PROT_NODATA:
 		regs->flags |= IPR_ATA_FLAG_PACKET_CMD;
 		break;
 
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		regs->flags |= IPR_ATA_FLAG_PACKET_CMD;
 		regs->flags |= IPR_ATA_FLAG_XFER_TYPE_DMA;
 		break;
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 831294de1d8d..f78d0605747d 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -200,7 +200,7 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
 	case ATA_PROT_NCQ:
 		task->ata_task.use_ncq = 1;
 		/* fall through */
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 	case ATA_PROT_DMA:
 		task->ata_task.dma_xfer = 1;
 		break;
diff --git a/include/linux/ata.h b/include/linux/ata.h
index c17e9404c88d..bc55471a4b2c 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -341,9 +341,9 @@ enum ata_tf_protocols {
 	ATA_PROT_PIO,		/* PIO data xfer */
 	ATA_PROT_DMA,		/* DMA */
 	ATA_PROT_NCQ,		/* NCQ */
-	ATA_PROT_ATAPI,		/* packet command, PIO data xfer*/
-	ATA_PROT_ATAPI_NODATA,	/* packet command, no data */
-	ATA_PROT_ATAPI_DMA,	/* packet command with special DMA sauce */
+	ATAPI_PROT_NODATA,	/* packet command, no data */
+	ATAPI_PROT_PIO,		/* packet command, PIO data xfer*/
+	ATAPI_PROT_DMA,		/* packet command with special DMA sauce */
 };
 
 enum ata_ioctls {
@@ -395,11 +395,11 @@ static inline unsigned int ata_prot_flags(u8 prot)
 		return ATA_PROT_FLAG_DMA;
 	case ATA_PROT_NCQ:
 		return ATA_PROT_FLAG_DMA | ATA_PROT_FLAG_NCQ;
-	case ATA_PROT_ATAPI_NODATA:
+	case ATAPI_PROT_NODATA:
 		return ATA_PROT_FLAG_ATAPI;
-	case ATA_PROT_ATAPI:
+	case ATAPI_PROT_PIO:
 		return ATA_PROT_FLAG_ATAPI | ATA_PROT_FLAG_PIO;
-	case ATA_PROT_ATAPI_DMA:
+	case ATAPI_PROT_DMA:
 		return ATA_PROT_FLAG_ATAPI | ATA_PROT_FLAG_DMA;
 	}
 	return 0;
-- 
cgit v1.2.3


From ceb0c642624f634c5b4f46b0e22df19be87a2e53 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:06 +0900
Subject: libata: add ATAPI_* cmd types and implement atapi_cmd_type()

Add ATAPI command types - ATAPI_READ, WRITE, RW_BUF, READ_CD and MISC,
and implement atapi_cmd_type() which takes SCSI opcode and returns to
which class the opcode belongs.  This will be used later to improve
ATAPI handling.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/linux/libata.h | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

(limited to 'include')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index cc4eaef6f889..03afcd63202d 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -35,6 +35,7 @@
 #include <linux/workqueue.h>
 #include <scsi/scsi_host.h>
 #include <linux/acpi.h>
+#include <linux/cdrom.h>
 
 /*
  * Define if arch has non-standard setup.  This is a _PCI_ standard
@@ -346,6 +347,12 @@ enum {
 	ATA_DMA_MASK_ATA	= (1 << 0),	/* DMA on ATA Disk */
 	ATA_DMA_MASK_ATAPI	= (1 << 1),	/* DMA on ATAPI */
 	ATA_DMA_MASK_CFA	= (1 << 2),	/* DMA on CF Card */
+
+	/* ATAPI command types */
+	ATAPI_READ		= 0,		/* READs */
+	ATAPI_WRITE		= 1,		/* WRITEs */
+	ATAPI_READ_CD		= 2,		/* READ CD [MSF] */
+	ATAPI_MISC		= 3,		/* the rest */
 };
 
 enum ata_xfer_mask {
@@ -1408,6 +1415,27 @@ static inline int ata_try_flush_cache(const struct ata_device *dev)
 	       ata_id_has_flush_ext(dev->id);
 }
 
+static inline int atapi_cmd_type(u8 opcode)
+{
+	switch (opcode) {
+	case GPCMD_READ_10:
+	case GPCMD_READ_12:
+		return ATAPI_READ;
+
+	case GPCMD_WRITE_10:
+	case GPCMD_WRITE_12:
+	case GPCMD_WRITE_AND_VERIFY_10:
+		return ATAPI_WRITE;
+
+	case GPCMD_READ_CD:
+	case GPCMD_READ_CD_MSF:
+		return ATAPI_READ_CD;
+
+	default:
+		return ATAPI_MISC;
+	}
+}
+
 static inline unsigned int ac_err_mask(u8 status)
 {
 	if (status & (ATA_BUSY | ATA_DRQ))
-- 
cgit v1.2.3


From 55dba3120fbcbea6800f9a18503d25f73212a347 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:07 +0900
Subject: libata: update ->data_xfer hook for ATAPI

Depending on how many bytes are transferred as a unit, PIO data
transfer may consume more bytes than requested.  Knowing how much
data is consumed is necessary to determine how much is left for
draining.  This patch update ->data_xfer such that it returns the
number of consumed bytes.

While at it, it also makes the following changes.

* s/adev/dev/
* use READ/WRITE constants for rw indication
* misc clean ups

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c    | 46 +++++++++++++++++++++++++++++---------------
 drivers/ata/pata_bf54x.c     | 28 ++++++++++++++-------------
 drivers/ata/pata_ixp4xx_cf.c | 26 +++++++++++++------------
 drivers/ata/pata_legacy.c    | 36 ++++++++++++++++++----------------
 drivers/ata/pata_qdi.c       | 30 ++++++++++++++++-------------
 drivers/ata/pata_scc.c       | 30 +++++++++++++++--------------
 drivers/ata/pata_winbond.c   | 28 +++++++++++++++------------
 include/linux/libata.h       | 11 ++++++-----
 8 files changed, 133 insertions(+), 102 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 2c9745a74d50..39cedd949ed4 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4994,7 +4994,7 @@ void swap_buf_le16(u16 *buf, unsigned int buf_words)
 
 /**
  *	ata_data_xfer - Transfer data by PIO
- *	@adev: device to target
+ *	@dev: device to target
  *	@buf: data buffer
  *	@buflen: buffer length
  *	@write_data: read/write
@@ -5003,37 +5003,44 @@ void swap_buf_le16(u16 *buf, unsigned int buf_words)
  *
  *	LOCKING:
  *	Inherited from caller.
+ *
+ *	RETURNS:
+ *	Bytes consumed.
  */
-void ata_data_xfer(struct ata_device *adev, unsigned char *buf,
-		   unsigned int buflen, int write_data)
+unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
+			   unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
+	struct ata_port *ap = dev->link->ap;
+	void __iomem *data_addr = ap->ioaddr.data_addr;
 	unsigned int words = buflen >> 1;
 
 	/* Transfer multiple of 2 bytes */
-	if (write_data)
-		iowrite16_rep(ap->ioaddr.data_addr, buf, words);
+	if (rw == READ)
+		ioread16_rep(data_addr, buf, words);
 	else
-		ioread16_rep(ap->ioaddr.data_addr, buf, words);
+		iowrite16_rep(data_addr, buf, words);
 
 	/* Transfer trailing 1 byte, if any. */
 	if (unlikely(buflen & 0x01)) {
 		u16 align_buf[1] = { 0 };
 		unsigned char *trailing_buf = buf + buflen - 1;
 
-		if (write_data) {
-			memcpy(align_buf, trailing_buf, 1);
-			iowrite16(le16_to_cpu(align_buf[0]), ap->ioaddr.data_addr);
-		} else {
-			align_buf[0] = cpu_to_le16(ioread16(ap->ioaddr.data_addr));
+		if (rw == READ) {
+			align_buf[0] = cpu_to_le16(ioread16(data_addr));
 			memcpy(trailing_buf, align_buf, 1);
+		} else {
+			memcpy(align_buf, trailing_buf, 1);
+			iowrite16(le16_to_cpu(align_buf[0]), data_addr);
 		}
+		words++;
 	}
+
+	return words << 1;
 }
 
 /**
  *	ata_data_xfer_noirq - Transfer data by PIO
- *	@adev: device to target
+ *	@dev: device to target
  *	@buf: data buffer
  *	@buflen: buffer length
  *	@write_data: read/write
@@ -5043,14 +5050,21 @@ void ata_data_xfer(struct ata_device *adev, unsigned char *buf,
  *
  *	LOCKING:
  *	Inherited from caller.
+ *
+ *	RETURNS:
+ *	Bytes consumed.
  */
-void ata_data_xfer_noirq(struct ata_device *adev, unsigned char *buf,
-			 unsigned int buflen, int write_data)
+unsigned int ata_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
+				 unsigned int buflen, int rw)
 {
 	unsigned long flags;
+	unsigned int consumed;
+
 	local_irq_save(flags);
-	ata_data_xfer(adev, buf, buflen, write_data);
+	consumed = ata_data_xfer(dev, buf, buflen, rw);
 	local_irq_restore(flags);
+
+	return consumed;
 }
 
 
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 7842cc487359..41cd921082ba 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1167,34 +1167,36 @@ static unsigned char bfin_bmdma_status(struct ata_port *ap)
  *	Note: Original code is ata_data_xfer().
  */
 
-static void bfin_data_xfer(struct ata_device *adev, unsigned char *buf,
-			   unsigned int buflen, int write_data)
+static unsigned int bfin_data_xfer(struct ata_device *dev, unsigned char *buf,
+				   unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
-	unsigned int words = buflen >> 1;
-	unsigned short *buf16 = (u16 *) buf;
+	struct ata_port *ap = dev->link->ap;
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
+	unsigned int words = buflen >> 1;
+	unsigned short *buf16 = (u16 *)buf;
 
 	/* Transfer multiple of 2 bytes */
-	if (write_data) {
-		write_atapi_data(base, words, buf16);
-	} else {
+	if (rw == READ)
 		read_atapi_data(base, words, buf16);
-	}
+	else
+		write_atapi_data(base, words, buf16);
 
 	/* Transfer trailing 1 byte, if any. */
 	if (unlikely(buflen & 0x01)) {
 		unsigned short align_buf[1] = { 0 };
 		unsigned char *trailing_buf = buf + buflen - 1;
 
-		if (write_data) {
-			memcpy(align_buf, trailing_buf, 1);
-			write_atapi_data(base, 1, align_buf);
-		} else {
+		if (rw == READ) {
 			read_atapi_data(base, 1, align_buf);
 			memcpy(trailing_buf, align_buf, 1);
+		} else {
+			memcpy(align_buf, trailing_buf, 1);
+			write_atapi_data(base, 1, align_buf);
 		}
+		words++;
 	}
+
+	return words << 1;
 }
 
 /**
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index 120b5bfa7ce6..030878fedeb5 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -42,13 +42,13 @@ static int ixp4xx_set_mode(struct ata_link *link, struct ata_device **error)
 	return 0;
 }
 
-static void ixp4xx_mmio_data_xfer(struct ata_device *adev, unsigned char *buf,
-				unsigned int buflen, int write_data)
+static unsigned int ixp4xx_mmio_data_xfer(struct ata_device *dev,
+				unsigned char *buf, unsigned int buflen, int rw)
 {
 	unsigned int i;
 	unsigned int words = buflen >> 1;
 	u16 *buf16 = (u16 *) buf;
-	struct ata_port *ap = adev->link->ap;
+	struct ata_port *ap = dev->link->ap;
 	void __iomem *mmio = ap->ioaddr.data_addr;
 	struct ixp4xx_pata_data *data = ap->host->dev->platform_data;
 
@@ -59,30 +59,32 @@ static void ixp4xx_mmio_data_xfer(struct ata_device *adev, unsigned char *buf,
 	udelay(100);
 
 	/* Transfer multiple of 2 bytes */
-	if (write_data) {
-		for (i = 0; i < words; i++)
-			writew(buf16[i], mmio);
-	} else {
+	if (rw == READ)
 		for (i = 0; i < words; i++)
 			buf16[i] = readw(mmio);
-	}
+	else
+		for (i = 0; i < words; i++)
+			writew(buf16[i], mmio);
 
 	/* Transfer trailing 1 byte, if any. */
 	if (unlikely(buflen & 0x01)) {
 		u16 align_buf[1] = { 0 };
 		unsigned char *trailing_buf = buf + buflen - 1;
 
-		if (write_data) {
-			memcpy(align_buf, trailing_buf, 1);
-			writew(align_buf[0], mmio);
-		} else {
+		if (rw == READ) {
 			align_buf[0] = readw(mmio);
 			memcpy(trailing_buf, align_buf, 1);
+		} else {
+			memcpy(align_buf, trailing_buf, 1);
+			writew(align_buf[0], mmio);
 		}
+		words++;
 	}
 
 	udelay(100);
 	*data->cs0_cfg |= 0x01;
+
+	return words << 1;
 }
 
 static struct scsi_host_template ixp4xx_sht = {
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index 17159b5e1e43..dae85aa12e32 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -249,13 +249,14 @@ static void pdc20230_set_piomode(struct ata_port *ap, struct ata_device *adev)
 
 }
 
-static void pdc_data_xfer_vlb(struct ata_device *adev, unsigned char *buf, unsigned int buflen, int write_data)
+static unsigned int pdc_data_xfer_vlb(struct ata_device *dev,
+				unsigned char *buf, unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
-	int slop = buflen & 3;
-	unsigned long flags;
-
 	if (ata_id_has_dword_io(adev->id)) {
+		struct ata_port *ap = dev->link->ap;
+		int slop = buflen & 3;
+		unsigned long flags;
+
 		local_irq_save(flags);
 
 		/* Perform the 32bit I/O synchronization sequence */
@@ -264,26 +265,27 @@ static void pdc_data_xfer_vlb(struct ata_device *adev, unsigned char *buf, unsig
 		ioread8(ap->ioaddr.nsect_addr);
 
 		/* Now the data */
-
-		if (write_data)
-			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
-		else
+		if (rw == READ)
 			ioread32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
+		else
+			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
 
 		if (unlikely(slop)) {
-			__le32 pad = 0;
-			if (write_data) {
-				memcpy(&pad, buf + buflen - slop, slop);
-				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
-			} else {
+			u32 pad;
+			if (rw == READ) {
 				pad = cpu_to_le32(ioread32(ap->ioaddr.data_addr));
 				memcpy(buf + buflen - slop, &pad, slop);
+			} else {
+				memcpy(&pad, buf + buflen - slop, slop);
+				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
 			}
+			buflen += 4 - slop;
 		}
 		local_irq_restore(flags);
-	}
-	else
-		ata_data_xfer_noirq(adev, buf, buflen, write_data);
+	} else
+		buflen = ata_data_xfer_noirq(dev, buf, buflen, rw);
+
+	return buflen;
 }
 
 static struct ata_port_operations pdc20230_port_ops = {
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index a4c0e502cb42..9f308ed76cc8 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -124,29 +124,33 @@ static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
 	return ata_qc_issue_prot(qc);
 }
 
-static void qdi_data_xfer(struct ata_device *adev, unsigned char *buf, unsigned int buflen, int write_data)
+static unsigned int qdi_data_xfer(struct ata_device *dev, unsigned char *buf,
+				  unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
-	int slop = buflen & 3;
+	if (ata_id_has_dword_io(dev->id)) {
+		struct ata_port *ap = dev->link->ap;
+		int slop = buflen & 3;
 
-	if (ata_id_has_dword_io(adev->id)) {
-		if (write_data)
-			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
-		else
+		if (rw == READ)
 			ioread32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
+		else
+			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
 
 		if (unlikely(slop)) {
-			__le32 pad = 0;
-			if (write_data) {
-				memcpy(&pad, buf + buflen - slop, slop);
-				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
-			} else {
+			u32 pad;
+			if (rw == READ) {
 				pad = cpu_to_le32(ioread32(ap->ioaddr.data_addr));
 				memcpy(buf + buflen - slop, &pad, slop);
+			} else {
+				memcpy(&pad, buf + buflen - slop, slop);
+				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
 			}
+			buflen += 4 - slop;
 		}
 	} else
-		ata_data_xfer(adev, buf, buflen, write_data);
+		buflen = ata_data_xfer(dev, buf, buflen, rw);
+
+	return buflen;
 }
 
 static struct scsi_host_template qdi_sht = {
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index ea2ef9fc15be..55055b27524c 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -768,45 +768,47 @@ static u8 scc_bmdma_status (struct ata_port *ap)
 
 /**
  *	scc_data_xfer - Transfer data by PIO
- *	@adev: device for this I/O
+ *	@dev: device for this I/O
  *	@buf: data buffer
  *	@buflen: buffer length
- *	@write_data: read/write
+ *	@rw: read/write
  *
  *	Note: Original code is ata_data_xfer().
  */
 
-static void scc_data_xfer (struct ata_device *adev, unsigned char *buf,
-			   unsigned int buflen, int write_data)
+static unsigned int scc_data_xfer (struct ata_device *dev, unsigned char *buf,
+				   unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
+	struct ata_port *ap = dev->link->ap;
 	unsigned int words = buflen >> 1;
 	unsigned int i;
 	u16 *buf16 = (u16 *) buf;
 	void __iomem *mmio = ap->ioaddr.data_addr;
 
 	/* Transfer multiple of 2 bytes */
-	if (write_data) {
-		for (i = 0; i < words; i++)
-			out_be32(mmio, cpu_to_le16(buf16[i]));
-	} else {
+	if (rw == READ)
 		for (i = 0; i < words; i++)
 			buf16[i] = le16_to_cpu(in_be32(mmio));
-	}
+	else
+		for (i = 0; i < words; i++)
+			out_be32(mmio, cpu_to_le16(buf16[i]));
 
 	/* Transfer trailing 1 byte, if any. */
 	if (unlikely(buflen & 0x01)) {
 		u16 align_buf[1] = { 0 };
 		unsigned char *trailing_buf = buf + buflen - 1;
 
-		if (write_data) {
-			memcpy(align_buf, trailing_buf, 1);
-			out_be32(mmio, cpu_to_le16(align_buf[0]));
-		} else {
+		if (rw == READ) {
 			align_buf[0] = le16_to_cpu(in_be32(mmio));
 			memcpy(trailing_buf, align_buf, 1);
+		} else {
+			memcpy(align_buf, trailing_buf, 1);
+			out_be32(mmio, cpu_to_le16(align_buf[0]));
 		}
+		words++;
 	}
+
+	return words << 1;
 }
 
 /**
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index 7116a9e7a8b2..7312e9182d69 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -92,29 +92,33 @@ static void winbond_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 
-static void winbond_data_xfer(struct ata_device *adev, unsigned char *buf, unsigned int buflen, int write_data)
+static void winbond_data_xfer(struct ata_device *dev, unsigned char *buf,
+			      unsigned int buflen, int rw)
 {
-	struct ata_port *ap = adev->link->ap;
+	struct ata_port *ap = dev->link->ap;
 	int slop = buflen & 3;
 
-	if (ata_id_has_dword_io(adev->id)) {
-		if (write_data)
-			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
-		else
+	if (ata_id_has_dword_io(dev->id)) {
+		if (rw == READ)
 			ioread32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
+		else
+			iowrite32_rep(ap->ioaddr.data_addr, buf, buflen >> 2);
 
 		if (unlikely(slop)) {
-			__le32 pad = 0;
-			if (write_data) {
-				memcpy(&pad, buf + buflen - slop, slop);
-				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
-			} else {
+			u32 pad;
+			if (rw == READ) {
 				pad = cpu_to_le32(ioread32(ap->ioaddr.data_addr));
 				memcpy(buf + buflen - slop, &pad, slop);
+			} else {
+				memcpy(&pad, buf + buflen - slop, slop);
+				iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
 			}
+			buflen += 4 - slop;
 		}
 	} else
-		ata_data_xfer(adev, buf, buflen, write_data);
+		buflen = ata_data_xfer(dev, buf, buflen, rw);
+
+	return buflen;
 }
 
 static struct scsi_host_template winbond_sht = {
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 03afcd63202d..7fa96cb4f6db 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -701,7 +701,8 @@ struct ata_port_operations {
 	void (*bmdma_setup) (struct ata_queued_cmd *qc);
 	void (*bmdma_start) (struct ata_queued_cmd *qc);
 
-	void (*data_xfer) (struct ata_device *, unsigned char *, unsigned int, int);
+	unsigned int (*data_xfer) (struct ata_device *dev, unsigned char *buf,
+				   unsigned int buflen, int rw);
 
 	int (*qc_defer) (struct ata_queued_cmd *qc);
 	void (*qc_prep) (struct ata_queued_cmd *qc);
@@ -881,10 +882,10 @@ extern void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
 extern int ata_port_start(struct ata_port *ap);
 extern int ata_sff_port_start(struct ata_port *ap);
 extern irqreturn_t ata_interrupt(int irq, void *dev_instance);
-extern void ata_data_xfer(struct ata_device *adev, unsigned char *buf,
-			  unsigned int buflen, int write_data);
-extern void ata_data_xfer_noirq(struct ata_device *adev, unsigned char *buf,
-				unsigned int buflen, int write_data);
+extern unsigned int ata_data_xfer(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
+extern unsigned int ata_data_xfer_noirq(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
 extern int ata_std_qc_defer(struct ata_queued_cmd *qc);
 extern void ata_dumb_qc_prep(struct ata_queued_cmd *qc);
 extern void ata_qc_prep(struct ata_queued_cmd *qc);
-- 
cgit v1.2.3


From 001102d7859be0e7f7b9f2d62b841f2c0f9c2640 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:09 +0900
Subject: libata: kill non-sg DMA interface

With atapi_request_sense() converted to use sg, there's no user of
non-sg interface.  Kill non-sg interface.

* ATA_QCFLAG_SINGLE and ATA_QCFLAG_SG are removed.  ATA_QCFLAG_DMAMAP
  is used instead.  (this way no LLD change is necessary)

* qc->buf_virt is removed.

* ata_sg_init_one() and ata_sg_setup_one() are removed.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 148 +++++-----------------------------------------
 include/linux/libata.h    |   7 +--
 2 files changed, 16 insertions(+), 139 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 39cedd949ed4..76360f0ca20d 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4478,9 +4478,6 @@ void ata_sg_clean(struct ata_queued_cmd *qc)
 	WARN_ON(!(qc->flags & ATA_QCFLAG_DMAMAP));
 	WARN_ON(sg == NULL);
 
-	if (qc->flags & ATA_QCFLAG_SINGLE)
-		WARN_ON(qc->n_elem > 1);
-
 	VPRINTK("unmapping %u sg elements\n", qc->n_elem);
 
 	/* if we padded the buffer out to 32-bit bound, and data
@@ -4490,27 +4487,15 @@ void ata_sg_clean(struct ata_queued_cmd *qc)
 	if (qc->pad_len && !(qc->tf.flags & ATA_TFLAG_WRITE))
 		pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
 
-	if (qc->flags & ATA_QCFLAG_SG) {
-		if (qc->n_elem)
-			dma_unmap_sg(ap->dev, sg, qc->n_elem, dir);
-		/* restore last sg */
-		sg_last(sg, qc->orig_n_elem)->length += qc->pad_len;
-		if (pad_buf) {
-			struct scatterlist *psg = &qc->pad_sgent;
-			void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
-			memcpy(addr + psg->offset, pad_buf, qc->pad_len);
-			kunmap_atomic(addr, KM_IRQ0);
-		}
-	} else {
-		if (qc->n_elem)
-			dma_unmap_single(ap->dev,
-				sg_dma_address(&sg[0]), sg_dma_len(&sg[0]),
-				dir);
-		/* restore sg */
-		sg->length += qc->pad_len;
-		if (pad_buf)
-			memcpy(qc->buf_virt + sg->length - qc->pad_len,
-			       pad_buf, qc->pad_len);
+	if (qc->n_elem)
+		dma_unmap_sg(ap->dev, sg, qc->n_elem, dir);
+	/* restore last sg */
+	sg_last(sg, qc->orig_n_elem)->length += qc->pad_len;
+	if (pad_buf) {
+		struct scatterlist *psg = &qc->pad_sgent;
+		void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
+		memcpy(addr + psg->offset, pad_buf, qc->pad_len);
+		kunmap_atomic(addr, KM_IRQ0);
 	}
 
 	qc->flags &= ~ATA_QCFLAG_DMAMAP;
@@ -4764,33 +4749,6 @@ void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
 
 void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
 
-/**
- *	ata_sg_init_one - Associate command with memory buffer
- *	@qc: Command to be associated
- *	@buf: Memory buffer
- *	@buflen: Length of memory buffer, in bytes.
- *
- *	Initialize the data-related elements of queued_cmd @qc
- *	to point to a single memory buffer, @buf of byte length @buflen.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-
-void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen)
-{
-	qc->flags |= ATA_QCFLAG_SINGLE;
-
-	qc->__sg = &qc->sgent;
-	qc->n_elem = 1;
-	qc->orig_n_elem = 1;
-	qc->buf_virt = buf;
-	qc->nbytes = buflen;
-	qc->cursg = qc->__sg;
-
-	sg_init_one(&qc->sgent, buf, buflen);
-}
-
 /**
  *	ata_sg_init - Associate command with scatter-gather table.
  *	@qc: Command to be associated
@@ -4808,82 +4766,13 @@ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen)
 void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 		 unsigned int n_elem)
 {
-	qc->flags |= ATA_QCFLAG_SG;
+	qc->flags |= ATA_QCFLAG_DMAMAP;
 	qc->__sg = sg;
 	qc->n_elem = n_elem;
 	qc->orig_n_elem = n_elem;
 	qc->cursg = qc->__sg;
 }
 
-/**
- *	ata_sg_setup_one - DMA-map the memory buffer associated with a command.
- *	@qc: Command with memory buffer to be mapped.
- *
- *	DMA-map the memory buffer associated with queued_cmd @qc.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	Zero on success, negative on error.
- */
-
-static int ata_sg_setup_one(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	int dir = qc->dma_dir;
-	struct scatterlist *sg = qc->__sg;
-	dma_addr_t dma_address;
-	int trim_sg = 0;
-
-	/* we must lengthen transfers to end on a 32-bit boundary */
-	qc->pad_len = sg->length & 3;
-	if (qc->pad_len) {
-		void *pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
-		struct scatterlist *psg = &qc->pad_sgent;
-
-		WARN_ON(qc->dev->class != ATA_DEV_ATAPI);
-
-		memset(pad_buf, 0, ATA_DMA_PAD_SZ);
-
-		if (qc->tf.flags & ATA_TFLAG_WRITE)
-			memcpy(pad_buf, qc->buf_virt + sg->length - qc->pad_len,
-			       qc->pad_len);
-
-		sg_dma_address(psg) = ap->pad_dma + (qc->tag * ATA_DMA_PAD_SZ);
-		sg_dma_len(psg) = ATA_DMA_PAD_SZ;
-		/* trim sg */
-		sg->length -= qc->pad_len;
-		if (sg->length == 0)
-			trim_sg = 1;
-
-		DPRINTK("padding done, sg->length=%u pad_len=%u\n",
-			sg->length, qc->pad_len);
-	}
-
-	if (trim_sg) {
-		qc->n_elem--;
-		goto skip_map;
-	}
-
-	dma_address = dma_map_single(ap->dev, qc->buf_virt,
-				     sg->length, dir);
-	if (dma_mapping_error(dma_address)) {
-		/* restore sg */
-		sg->length += qc->pad_len;
-		return -1;
-	}
-
-	sg_dma_address(sg) = dma_address;
-	sg_dma_len(sg) = sg->length;
-
-skip_map:
-	DPRINTK("mapped buffer of %d bytes for %s\n", sg_dma_len(sg),
-		qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
-
-	return 0;
-}
-
 /**
  *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
  *	@qc: Command with scatter-gather table to be mapped.
@@ -4906,7 +4795,7 @@ static int ata_sg_setup(struct ata_queued_cmd *qc)
 	int n_elem, pre_n_elem, dir, trim_sg = 0;
 
 	VPRINTK("ENTER, ata%u\n", ap->print_id);
-	WARN_ON(!(qc->flags & ATA_QCFLAG_SG));
+	WARN_ON(!(qc->flags & ATA_QCFLAG_DMAMAP));
 
 	/* we must lengthen transfers to end on a 32-bit boundary */
 	qc->pad_len = lsg->length & 3;
@@ -6025,16 +5914,10 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 
 	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
 				 (ap->flags & ATA_FLAG_PIO_DMA))) {
-		if (qc->flags & ATA_QCFLAG_SG) {
-			if (ata_sg_setup(qc))
-				goto sg_err;
-		} else if (qc->flags & ATA_QCFLAG_SINGLE) {
-			if (ata_sg_setup_one(qc))
-				goto sg_err;
-		}
-	} else {
-		qc->flags &= ~ATA_QCFLAG_DMAMAP;
-	}
+		if (ata_sg_setup(qc))
+			goto sg_err;
+	} else
+		qc->flags &= ATA_QCFLAG_DMAMAP;
 
 	/* if device is sleeping, schedule softreset and abort the link */
 	if (unlikely(qc->dev->flags & ATA_DFLAG_SLEEPING)) {
@@ -7612,7 +7495,6 @@ EXPORT_SYMBOL_GPL(ata_host_register);
 EXPORT_SYMBOL_GPL(ata_host_activate);
 EXPORT_SYMBOL_GPL(ata_host_detach);
 EXPORT_SYMBOL_GPL(ata_sg_init);
-EXPORT_SYMBOL_GPL(ata_sg_init_one);
 EXPORT_SYMBOL_GPL(ata_hsm_move);
 EXPORT_SYMBOL_GPL(ata_qc_complete);
 EXPORT_SYMBOL_GPL(ata_qc_complete_multiple);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 7fa96cb4f6db..acd90ad78417 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -219,9 +219,7 @@ enum {
 
 	/* struct ata_queued_cmd flags */
 	ATA_QCFLAG_ACTIVE	= (1 << 0), /* cmd not yet ack'd to scsi lyer */
-	ATA_QCFLAG_SG		= (1 << 1), /* have s/g table? */
-	ATA_QCFLAG_SINGLE	= (1 << 2), /* no s/g, just a single buffer */
-	ATA_QCFLAG_DMAMAP	= ATA_QCFLAG_SG | ATA_QCFLAG_SINGLE,
+	ATA_QCFLAG_DMAMAP	= (1 << 1), /* SG table is DMA mapped */
 	ATA_QCFLAG_IO		= (1 << 3), /* standard IO command */
 	ATA_QCFLAG_RESULT_TF	= (1 << 4), /* result TF requested */
 	ATA_QCFLAG_CLEAR_EXCL	= (1 << 5), /* clear excl_link on completion */
@@ -475,7 +473,6 @@ struct ata_queued_cmd {
 
 	struct scatterlist	sgent;
 	struct scatterlist	pad_sgent;
-	void			*buf_virt;
 
 	/* DO NOT iterate over __sg manually, use ata_for_each_sg() */
 	struct scatterlist	*__sg;
@@ -891,8 +888,6 @@ extern void ata_dumb_qc_prep(struct ata_queued_cmd *qc);
 extern void ata_qc_prep(struct ata_queued_cmd *qc);
 extern void ata_noop_qc_prep(struct ata_queued_cmd *qc);
 extern unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc);
-extern void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf,
-		unsigned int buflen);
 extern void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 		 unsigned int n_elem);
 extern unsigned int ata_dev_classify(const struct ata_taskfile *tf);
-- 
cgit v1.2.3


From ff2aeb1eb64c8a4770a6304f9addbae9f9828646 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:11 +0900
Subject: libata: convert to chained sg

libata used private sg iterator to handle padding sg.  Now that sg can
be chained, padding can be handled using standard sg ops.  Convert to
chained sg.

* s/qc->__sg/qc->sg/

* s/qc->pad_sgent/qc->extra_sg[]/.  Because chaining consumes one sg
  entry.  There need to be two extra sg entries.  The renaming is also
  for future addition of other extra sg entries.

* Padding setup is moved into ata_sg_setup_extra() which is organized
  in a way that future addition of other extra sg entries is easy.

* qc->orig_n_elem is unused and removed.

* qc->n_elem now contains the number of sg entries that LLDs should
  map.  qc->mapped_n_elem is added to carry the original number of
  mapped sgs for unmapping.

* The last sg of the original sg list is used to chain to extra sg
  list.  The original last sg is pointed to by qc->last_sg and the
  content is stored in qc->saved_last_sg.  It's restored during
  ata_sg_clean().

* All sg walking code has been updated.  Unnecessary assertions and
  checks for conditions the core layer already guarantees are removed.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ahci.c            |  18 ++--
 drivers/ata/libata-core.c     | 201 +++++++++++++++++++++++++-----------------
 drivers/ata/libata-scsi.c     |   2 +-
 drivers/ata/pata_bf54x.c      |  13 +--
 drivers/ata/pata_icside.c     |   3 +-
 drivers/ata/pdc_adma.c        |   3 +-
 drivers/ata/sata_fsl.c        |   4 +-
 drivers/ata/sata_mv.c         |   3 +-
 drivers/ata/sata_nv.c         |  25 ++----
 drivers/ata/sata_promise.c    |  40 ++++-----
 drivers/ata/sata_qstor.c      |  13 +--
 drivers/ata/sata_sil24.c      |   6 +-
 drivers/ata/sata_sx4.c        |   4 +-
 drivers/scsi/ipr.c            |   3 +-
 drivers/scsi/libsas/sas_ata.c |  10 +--
 include/linux/libata.h        |  42 ++-------
 16 files changed, 193 insertions(+), 197 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 5eee91c73c90..cffad07c65bf 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1483,28 +1483,24 @@ static void ahci_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 static unsigned int ahci_fill_sg(struct ata_queued_cmd *qc, void *cmd_tbl)
 {
 	struct scatterlist *sg;
-	struct ahci_sg *ahci_sg;
-	unsigned int n_sg = 0;
+	struct ahci_sg *ahci_sg = cmd_tbl + AHCI_CMD_TBL_HDR_SZ;
+	unsigned int si;
 
 	VPRINTK("ENTER\n");
 
 	/*
 	 * Next, the S/G list.
 	 */
-	ahci_sg = cmd_tbl + AHCI_CMD_TBL_HDR_SZ;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		dma_addr_t addr = sg_dma_address(sg);
 		u32 sg_len = sg_dma_len(sg);
 
-		ahci_sg->addr = cpu_to_le32(addr & 0xffffffff);
-		ahci_sg->addr_hi = cpu_to_le32((addr >> 16) >> 16);
-		ahci_sg->flags_size = cpu_to_le32(sg_len - 1);
-
-		ahci_sg++;
-		n_sg++;
+		ahci_sg[si].addr = cpu_to_le32(addr & 0xffffffff);
+		ahci_sg[si].addr_hi = cpu_to_le32((addr >> 16) >> 16);
+		ahci_sg[si].flags_size = cpu_to_le32(sg_len - 1);
 	}
 
-	return n_sg;
+	return si;
 }
 
 static void ahci_qc_prep(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 865428a64de3..e998028302da 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4471,13 +4471,13 @@ static unsigned int ata_dev_init_params(struct ata_device *dev,
 void ata_sg_clean(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	struct scatterlist *sg = qc->__sg;
+	struct scatterlist *sg = qc->sg;
 	int dir = qc->dma_dir;
 	void *pad_buf = NULL;
 
 	WARN_ON(sg == NULL);
 
-	VPRINTK("unmapping %u sg elements\n", qc->n_elem);
+	VPRINTK("unmapping %u sg elements\n", qc->mapped_n_elem);
 
 	/* if we padded the buffer out to 32-bit bound, and data
 	 * xfer direction is from-device, we must copy from the
@@ -4486,19 +4486,20 @@ void ata_sg_clean(struct ata_queued_cmd *qc)
 	if (qc->pad_len && !(qc->tf.flags & ATA_TFLAG_WRITE))
 		pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
 
-	if (qc->n_elem)
-		dma_unmap_sg(ap->dev, sg, qc->n_elem, dir);
+	if (qc->mapped_n_elem)
+		dma_unmap_sg(ap->dev, sg, qc->mapped_n_elem, dir);
 	/* restore last sg */
-	sg_last(sg, qc->orig_n_elem)->length += qc->pad_len;
+	if (qc->last_sg)
+		*qc->last_sg = qc->saved_last_sg;
 	if (pad_buf) {
-		struct scatterlist *psg = &qc->pad_sgent;
+		struct scatterlist *psg = &qc->extra_sg[1];
 		void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
 		memcpy(addr + psg->offset, pad_buf, qc->pad_len);
 		kunmap_atomic(addr, KM_IRQ0);
 	}
 
 	qc->flags &= ~ATA_QCFLAG_DMAMAP;
-	qc->__sg = NULL;
+	qc->sg = NULL;
 }
 
 /**
@@ -4516,13 +4517,10 @@ static void ata_fill_sg(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg;
-	unsigned int idx;
+	unsigned int si, pi;
 
-	WARN_ON(qc->__sg == NULL);
-	WARN_ON(qc->n_elem == 0 && qc->pad_len == 0);
-
-	idx = 0;
-	ata_for_each_sg(sg, qc) {
+	pi = 0;
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u32 addr, offset;
 		u32 sg_len, len;
 
@@ -4539,18 +4537,17 @@ static void ata_fill_sg(struct ata_queued_cmd *qc)
 			if ((offset + sg_len) > 0x10000)
 				len = 0x10000 - offset;
 
-			ap->prd[idx].addr = cpu_to_le32(addr);
-			ap->prd[idx].flags_len = cpu_to_le32(len & 0xffff);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx, addr, len);
+			ap->prd[pi].addr = cpu_to_le32(addr);
+			ap->prd[pi].flags_len = cpu_to_le32(len & 0xffff);
+			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
 
-			idx++;
+			pi++;
 			sg_len -= len;
 			addr += len;
 		}
 	}
 
-	if (idx)
-		ap->prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
 }
 
 /**
@@ -4570,13 +4567,10 @@ static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg;
-	unsigned int idx;
-
-	WARN_ON(qc->__sg == NULL);
-	WARN_ON(qc->n_elem == 0 && qc->pad_len == 0);
+	unsigned int si, pi;
 
-	idx = 0;
-	ata_for_each_sg(sg, qc) {
+	pi = 0;
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u32 addr, offset;
 		u32 sg_len, len, blen;
 
@@ -4594,25 +4588,24 @@ static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
 				len = 0x10000 - offset;
 
 			blen = len & 0xffff;
-			ap->prd[idx].addr = cpu_to_le32(addr);
+			ap->prd[pi].addr = cpu_to_le32(addr);
 			if (blen == 0) {
 			   /* Some PATA chipsets like the CS5530 can't
 			      cope with 0x0000 meaning 64K as the spec says */
-				ap->prd[idx].flags_len = cpu_to_le32(0x8000);
+				ap->prd[pi].flags_len = cpu_to_le32(0x8000);
 				blen = 0x8000;
-				ap->prd[++idx].addr = cpu_to_le32(addr + 0x8000);
+				ap->prd[++pi].addr = cpu_to_le32(addr + 0x8000);
 			}
-			ap->prd[idx].flags_len = cpu_to_le32(blen);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx, addr, len);
+			ap->prd[pi].flags_len = cpu_to_le32(blen);
+			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
 
-			idx++;
+			pi++;
 			sg_len -= len;
 			addr += len;
 		}
 	}
 
-	if (idx)
-		ap->prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
 }
 
 /**
@@ -4764,54 +4757,48 @@ void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
 void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 		 unsigned int n_elem)
 {
-	qc->__sg = sg;
+	qc->sg = sg;
 	qc->n_elem = n_elem;
-	qc->orig_n_elem = n_elem;
-	qc->cursg = qc->__sg;
+	qc->cursg = qc->sg;
 }
 
-/**
- *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
- *	@qc: Command with scatter-gather table to be mapped.
- *
- *	DMA-map the scatter-gather table associated with queued_cmd @qc.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	Zero on success, negative on error.
- *
- */
-
-static int ata_sg_setup(struct ata_queued_cmd *qc)
+static unsigned int ata_sg_setup_extra(struct ata_queued_cmd *qc,
+				       unsigned int *n_elem_extra)
 {
 	struct ata_port *ap = qc->ap;
-	struct scatterlist *sg = qc->__sg;
-	struct scatterlist *lsg = sg_last(qc->__sg, qc->n_elem);
-	int n_elem, pre_n_elem, dir, trim_sg = 0;
+	unsigned int n_elem = qc->n_elem;
+	struct scatterlist *lsg, *copy_lsg = NULL, *tsg = NULL, *esg = NULL;
 
-	VPRINTK("ENTER, ata%u\n", ap->print_id);
+	*n_elem_extra = 0;
+
+	/* needs padding? */
+	qc->pad_len = qc->nbytes & 3;
+
+	if (likely(!qc->pad_len))
+		return n_elem;
+
+	/* locate last sg and save it */
+	lsg = sg_last(qc->sg, n_elem);
+	qc->last_sg = lsg;
+	qc->saved_last_sg = *lsg;
+
+	sg_init_table(qc->extra_sg, ARRAY_SIZE(qc->extra_sg));
 
-	/* we must lengthen transfers to end on a 32-bit boundary */
-	qc->pad_len = lsg->length & 3;
 	if (qc->pad_len) {
+		struct scatterlist *psg = &qc->extra_sg[1];
 		void *pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
-		struct scatterlist *psg = &qc->pad_sgent;
 		unsigned int offset;
 
 		WARN_ON(qc->dev->class != ATA_DEV_ATAPI);
 
 		memset(pad_buf, 0, ATA_DMA_PAD_SZ);
 
-		/*
-		 * psg->page/offset are used to copy to-be-written
+		/* psg->page/offset are used to copy to-be-written
 		 * data in this function or read data in ata_sg_clean.
 		 */
 		offset = lsg->offset + lsg->length - qc->pad_len;
-		sg_init_table(psg, 1);
 		sg_set_page(psg, nth_page(sg_page(lsg), offset >> PAGE_SHIFT),
-				qc->pad_len, offset_in_page(offset));
+			    qc->pad_len, offset_in_page(offset));
 
 		if (qc->tf.flags & ATA_TFLAG_WRITE) {
 			void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
@@ -4821,36 +4808,84 @@ static int ata_sg_setup(struct ata_queued_cmd *qc)
 
 		sg_dma_address(psg) = ap->pad_dma + (qc->tag * ATA_DMA_PAD_SZ);
 		sg_dma_len(psg) = ATA_DMA_PAD_SZ;
-		/* trim last sg */
+
+		/* Trim the last sg entry and chain the original and
+		 * padding sg lists.
+		 *
+		 * Because chaining consumes one sg entry, one extra
+		 * sg entry is allocated and the last sg entry is
+		 * copied to it if the length isn't zero after padded
+		 * amount is removed.
+		 *
+		 * If the last sg entry is completely replaced by
+		 * padding sg entry, the first sg entry is skipped
+		 * while chaining.
+		 */
 		lsg->length -= qc->pad_len;
-		if (lsg->length == 0)
-			trim_sg = 1;
+		if (lsg->length) {
+			copy_lsg = &qc->extra_sg[0];
+			tsg = &qc->extra_sg[0];
+		} else {
+			n_elem--;
+			tsg = &qc->extra_sg[1];
+		}
+
+		esg = &qc->extra_sg[1];
 
-		DPRINTK("padding done, sg[%d].length=%u pad_len=%u\n",
-			qc->n_elem - 1, lsg->length, qc->pad_len);
+		(*n_elem_extra)++;
 	}
 
-	pre_n_elem = qc->n_elem;
-	if (trim_sg && pre_n_elem)
-		pre_n_elem--;
+	if (copy_lsg)
+		sg_set_page(copy_lsg, sg_page(lsg), lsg->length, lsg->offset);
 
-	if (!pre_n_elem) {
-		n_elem = 0;
-		goto skip_map;
+	sg_chain(lsg, 1, tsg);
+	sg_mark_end(esg);
+
+	/* sglist can't start with chaining sg entry, fast forward */
+	if (qc->sg == lsg) {
+		qc->sg = tsg;
+		qc->cursg = tsg;
 	}
 
-	dir = qc->dma_dir;
-	n_elem = dma_map_sg(ap->dev, sg, pre_n_elem, dir);
-	if (n_elem < 1) {
-		/* restore last sg */
-		lsg->length += qc->pad_len;
-		return -1;
+	return n_elem;
+}
+
+/**
+ *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
+ *	@qc: Command with scatter-gather table to be mapped.
+ *
+ *	DMA-map the scatter-gather table associated with queued_cmd @qc.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	Zero on success, negative on error.
+ *
+ */
+static int ata_sg_setup(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	unsigned int n_elem, n_elem_extra;
+
+	VPRINTK("ENTER, ata%u\n", ap->print_id);
+
+	n_elem = ata_sg_setup_extra(qc, &n_elem_extra);
+
+	if (n_elem) {
+		n_elem = dma_map_sg(ap->dev, qc->sg, n_elem, qc->dma_dir);
+		if (n_elem < 1) {
+			/* restore last sg */
+			if (qc->last_sg)
+				*qc->last_sg = qc->saved_last_sg;
+			return -1;
+		}
+		DPRINTK("%d sg elements mapped\n", n_elem);
 	}
 
-	DPRINTK("%d sg elements mapped\n", n_elem);
+	qc->n_elem = qc->mapped_n_elem = n_elem;
+	qc->n_elem += n_elem_extra;
 
-skip_map:
-	qc->n_elem = n_elem;
 	qc->flags |= ATA_QCFLAG_DMAMAP;
 
 	return 0;
@@ -5912,7 +5947,7 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	/* We guarantee to LLDs that they will have at least one
 	 * non-zero sg if the command is a data command.
 	 */
-	BUG_ON(ata_is_data(prot) && (!qc->__sg || !qc->n_elem || !qc->nbytes));
+	BUG_ON(ata_is_data(prot) && (!qc->sg || !qc->n_elem || !qc->nbytes));
 
 	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
 				 (ap->flags & ATA_FLAG_PIO_DMA)))
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 5fd780e509d4..42bf61599730 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -517,7 +517,7 @@ static struct ata_queued_cmd *ata_scsi_qc_new(struct ata_device *dev,
 		qc->scsicmd = cmd;
 		qc->scsidone = done;
 
-		qc->__sg = scsi_sglist(cmd);
+		qc->sg = scsi_sglist(cmd);
 		qc->n_elem = scsi_sg_count(cmd);
 	} else {
 		cmd->result = (DID_OK << 16) | (QUEUE_FULL << 1);
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 41cd921082ba..a32e3c44a606 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -832,6 +832,7 @@ static void bfin_bmdma_setup(struct ata_queued_cmd *qc)
 {
 	unsigned short config = WDSIZE_16;
 	struct scatterlist *sg;
+	unsigned int si;
 
 	pr_debug("in atapi dma setup\n");
 	/* Program the ATA_CTRL register with dir */
@@ -839,7 +840,7 @@ static void bfin_bmdma_setup(struct ata_queued_cmd *qc)
 		/* fill the ATAPI DMA controller */
 		set_dma_config(CH_ATAPI_TX, config);
 		set_dma_x_modify(CH_ATAPI_TX, 2);
-		ata_for_each_sg(sg, qc) {
+		for_each_sg(qc->sg, sg, qc->n_elem, si) {
 			set_dma_start_addr(CH_ATAPI_TX, sg_dma_address(sg));
 			set_dma_x_count(CH_ATAPI_TX, sg_dma_len(sg) >> 1);
 		}
@@ -848,7 +849,7 @@ static void bfin_bmdma_setup(struct ata_queued_cmd *qc)
 		/* fill the ATAPI DMA controller */
 		set_dma_config(CH_ATAPI_RX, config);
 		set_dma_x_modify(CH_ATAPI_RX, 2);
-		ata_for_each_sg(sg, qc) {
+		for_each_sg(qc->sg, sg, qc->n_elem, si) {
 			set_dma_start_addr(CH_ATAPI_RX, sg_dma_address(sg));
 			set_dma_x_count(CH_ATAPI_RX, sg_dma_len(sg) >> 1);
 		}
@@ -867,6 +868,7 @@ static void bfin_bmdma_start(struct ata_queued_cmd *qc)
 	struct ata_port *ap = qc->ap;
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
 	struct scatterlist *sg;
+	unsigned int si;
 
 	pr_debug("in atapi dma start\n");
 	if (!(ap->udma_mask || ap->mwdma_mask))
@@ -881,7 +883,7 @@ static void bfin_bmdma_start(struct ata_queued_cmd *qc)
 		 * data cache is enabled. Otherwise, this loop
 		 * is an empty loop and optimized out.
 		 */
-		ata_for_each_sg(sg, qc) {
+		for_each_sg(qc->sg, sg, qc->n_elem, si) {
 			flush_dcache_range(sg_dma_address(sg),
 				sg_dma_address(sg) + sg_dma_len(sg));
 		}
@@ -910,7 +912,7 @@ static void bfin_bmdma_start(struct ata_queued_cmd *qc)
 	ATAPI_SET_CONTROL(base, ATAPI_GET_CONTROL(base) | TFRCNT_RST);
 
 		/* Set transfer length to buffer len */
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		ATAPI_SET_XFER_LEN(base, (sg_dma_len(sg) >> 1));
 	}
 
@@ -932,6 +934,7 @@ static void bfin_bmdma_stop(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg;
+	unsigned int si;
 
 	pr_debug("in atapi dma stop\n");
 	if (!(ap->udma_mask || ap->mwdma_mask))
@@ -950,7 +953,7 @@ static void bfin_bmdma_stop(struct ata_queued_cmd *qc)
 			 * data cache is enabled. Otherwise, this loop
 			 * is an empty loop and optimized out.
 			 */
-			ata_for_each_sg(sg, qc) {
+			for_each_sg(qc->sg, sg, qc->n_elem, si) {
 				invalidate_dcache_range(
 					sg_dma_address(sg),
 					sg_dma_address(sg)
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index 842fe08a3c13..5b8586dac63b 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -224,6 +224,7 @@ static void pata_icside_bmdma_setup(struct ata_queued_cmd *qc)
 	struct pata_icside_state *state = ap->host->private_data;
 	struct scatterlist *sg, *rsg = state->sg;
 	unsigned int write = qc->tf.flags & ATA_TFLAG_WRITE;
+	unsigned int si;
 
 	/*
 	 * We are simplex; BUG if we try to fiddle with DMA
@@ -234,7 +235,7 @@ static void pata_icside_bmdma_setup(struct ata_queued_cmd *qc)
 	/*
 	 * Copy ATAs scattered sg list into a contiguous array of sg
 	 */
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		memcpy(rsg, sg, sizeof(*sg));
 		rsg++;
 	}
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index 459cb7bb7d7f..8e1b7e9c0ae4 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -321,8 +321,9 @@ static int adma_fill_sg(struct ata_queued_cmd *qc)
 	u8  *buf = pp->pkt, *last_buf = NULL;
 	int i = (2 + buf[3]) * 8;
 	u8 pFLAGS = pORD | ((qc->tf.flags & ATA_TFLAG_WRITE) ? pDIRO : 0);
+	unsigned int si;
 
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u32 addr;
 		u32 len;
 
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index a3c33f165427..d041709dee1a 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -323,6 +323,7 @@ static unsigned int sata_fsl_fill_sg(struct ata_queued_cmd *qc, void *cmd_desc,
 	struct scatterlist *sg;
 	unsigned int num_prde = 0;
 	u32 ttl_dwords = 0;
+	unsigned int si;
 
 	/*
 	 * NOTE : direct & indirect prdt's are contigiously allocated
@@ -333,13 +334,14 @@ static unsigned int sata_fsl_fill_sg(struct ata_queued_cmd *qc, void *cmd_desc,
 	struct prde *prd_ptr_to_indirect_ext = NULL;
 	unsigned indirect_ext_segment_sz = 0;
 	dma_addr_t indirect_ext_segment_paddr;
+	unsigned int si;
 
 	VPRINTK("SATA FSL : cd = 0x%x, prd = 0x%x\n", cmd_desc, prd);
 
 	indirect_ext_segment_paddr = cmd_desc_paddr +
 	    SATA_FSL_CMD_DESC_OFFSET_TO_PRDT + SATA_FSL_MAX_PRD_DIRECT * 16;
 
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		dma_addr_t sg_addr = sg_dma_address(sg);
 		u32 sg_len = sg_dma_len(sg);
 
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 37b850ae0845..7e72463a90eb 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -1136,9 +1136,10 @@ static void mv_fill_sg(struct ata_queued_cmd *qc)
 	struct mv_port_priv *pp = qc->ap->private_data;
 	struct scatterlist *sg;
 	struct mv_sg *mv_sg, *last_sg = NULL;
+	unsigned int si;
 
 	mv_sg = pp->sg_tbl;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		dma_addr_t addr = sg_dma_address(sg);
 		u32 sg_len = sg_dma_len(sg);
 
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index ed5dc7cb50cd..a0f98fdab7a0 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -1336,21 +1336,18 @@ static void nv_adma_fill_aprd(struct ata_queued_cmd *qc,
 static void nv_adma_fill_sg(struct ata_queued_cmd *qc, struct nv_adma_cpb *cpb)
 {
 	struct nv_adma_port_priv *pp = qc->ap->private_data;
-	unsigned int idx;
 	struct nv_adma_prd *aprd;
 	struct scatterlist *sg;
+	unsigned int si;
 
 	VPRINTK("ENTER\n");
 
-	idx = 0;
-
-	ata_for_each_sg(sg, qc) {
-		aprd = (idx < 5) ? &cpb->aprd[idx] :
-			       &pp->aprd[NV_ADMA_SGTBL_LEN * qc->tag + (idx-5)];
-		nv_adma_fill_aprd(qc, sg, idx, aprd);
-		idx++;
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
+		aprd = (si < 5) ? &cpb->aprd[si] :
+			       &pp->aprd[NV_ADMA_SGTBL_LEN * qc->tag + (si-5)];
+		nv_adma_fill_aprd(qc, sg, si, aprd);
 	}
-	if (idx > 5)
+	if (si > 5)
 		cpb->next_aprd = cpu_to_le64(((u64)(pp->aprd_dma + NV_ADMA_SGTBL_SZ * qc->tag)));
 	else
 		cpb->next_aprd = cpu_to_le64(0);
@@ -1995,17 +1992,14 @@ static void nv_swncq_fill_sg(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg;
-	unsigned int idx;
 	struct nv_swncq_port_priv *pp = ap->private_data;
 	struct ata_prd *prd;
-
-	WARN_ON(qc->__sg == NULL);
-	WARN_ON(qc->n_elem == 0 && qc->pad_len == 0);
+	unsigned int si, idx;
 
 	prd = pp->prd + ATA_MAX_PRD * qc->tag;
 
 	idx = 0;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u32 addr, offset;
 		u32 sg_len, len;
 
@@ -2027,8 +2021,7 @@ static void nv_swncq_fill_sg(struct ata_queued_cmd *qc)
 		}
 	}
 
-	if (idx)
-		prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+	prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
 }
 
 static unsigned int nv_swncq_issue_atacmd(struct ata_port *ap,
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 01738d736d44..a07d319f6e8c 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -533,17 +533,15 @@ static void pdc_fill_sg(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg;
-	unsigned int idx;
 	const u32 SG_COUNT_ASIC_BUG = 41*4;
+	unsigned int si, idx;
+	u32 len;
 
 	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
 		return;
 
-	WARN_ON(qc->__sg == NULL);
-	WARN_ON(qc->n_elem == 0 && qc->pad_len == 0);
-
 	idx = 0;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u32 addr, offset;
 		u32 sg_len, len;
 
@@ -570,29 +568,27 @@ static void pdc_fill_sg(struct ata_queued_cmd *qc)
 		}
 	}
 
-	if (idx) {
-		u32 len = le32_to_cpu(ap->prd[idx - 1].flags_len);
+	len = le32_to_cpu(ap->prd[idx - 1].flags_len);
 
-		if (len > SG_COUNT_ASIC_BUG) {
-			u32 addr;
+	if (len > SG_COUNT_ASIC_BUG) {
+		u32 addr;
 
-			VPRINTK("Splitting last PRD.\n");
+		VPRINTK("Splitting last PRD.\n");
 
-			addr = le32_to_cpu(ap->prd[idx - 1].addr);
-			ap->prd[idx - 1].flags_len = cpu_to_le32(len - SG_COUNT_ASIC_BUG);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx - 1, addr, SG_COUNT_ASIC_BUG);
+		addr = le32_to_cpu(ap->prd[idx - 1].addr);
+		ap->prd[idx - 1].flags_len = cpu_to_le32(len - SG_COUNT_ASIC_BUG);
+		VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx - 1, addr, SG_COUNT_ASIC_BUG);
 
-			addr = addr + len - SG_COUNT_ASIC_BUG;
-			len = SG_COUNT_ASIC_BUG;
-			ap->prd[idx].addr = cpu_to_le32(addr);
-			ap->prd[idx].flags_len = cpu_to_le32(len);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx, addr, len);
+		addr = addr + len - SG_COUNT_ASIC_BUG;
+		len = SG_COUNT_ASIC_BUG;
+		ap->prd[idx].addr = cpu_to_le32(addr);
+		ap->prd[idx].flags_len = cpu_to_le32(len);
+		VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", idx, addr, len);
 
-			idx++;
-		}
-
-		ap->prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+		idx++;
 	}
+
+	ap->prd[idx - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
 }
 
 static void pdc_qc_prep(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 4e5f07bdd069..91cc12c82040 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -287,14 +287,10 @@ static unsigned int qs_fill_sg(struct ata_queued_cmd *qc)
 	struct scatterlist *sg;
 	struct ata_port *ap = qc->ap;
 	struct qs_port_priv *pp = ap->private_data;
-	unsigned int nelem;
 	u8 *prd = pp->pkt + QS_CPB_BYTES;
+	unsigned int si;
 
-	WARN_ON(qc->__sg == NULL);
-	WARN_ON(qc->n_elem == 0 && qc->pad_len == 0);
-
-	nelem = 0;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		u64 addr;
 		u32 len;
 
@@ -306,12 +302,11 @@ static unsigned int qs_fill_sg(struct ata_queued_cmd *qc)
 		*(__le32 *)prd = cpu_to_le32(len);
 		prd += sizeof(u64);
 
-		VPRINTK("PRD[%u] = (0x%llX, 0x%X)\n", nelem,
+		VPRINTK("PRD[%u] = (0x%llX, 0x%X)\n", si,
 					(unsigned long long)addr, len);
-		nelem++;
 	}
 
-	return nelem;
+	return si;
 }
 
 static void qs_qc_prep(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index fdd3ceac329b..b4b1f91ea693 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -813,8 +813,9 @@ static inline void sil24_fill_sg(struct ata_queued_cmd *qc,
 {
 	struct scatterlist *sg;
 	struct sil24_sge *last_sge = NULL;
+	unsigned int si;
 
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		sge->addr = cpu_to_le64(sg_dma_address(sg));
 		sge->cnt = cpu_to_le32(sg_dma_len(sg));
 		sge->flags = 0;
@@ -823,8 +824,7 @@ static inline void sil24_fill_sg(struct ata_queued_cmd *qc,
 		sge++;
 	}
 
-	if (likely(last_sge))
-		last_sge->flags = cpu_to_le32(SGE_TRM);
+	last_sge->flags = cpu_to_le32(SGE_TRM);
 }
 
 static int sil24_qc_defer(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 3de0c27caf53..211ba8da64f4 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -473,7 +473,7 @@ static void pdc20621_dma_prep(struct ata_queued_cmd *qc)
 	void __iomem *mmio = ap->host->iomap[PDC_MMIO_BAR];
 	void __iomem *dimm_mmio = ap->host->iomap[PDC_DIMM_BAR];
 	unsigned int portno = ap->port_no;
-	unsigned int i, idx, total_len = 0, sgt_len;
+	unsigned int i, si, idx, total_len = 0, sgt_len;
 	u32 *buf = (u32 *) &pp->dimm_buf[PDC_DIMM_HEADER_SZ];
 
 	WARN_ON(!(qc->flags & ATA_QCFLAG_DMAMAP));
@@ -487,7 +487,7 @@ static void pdc20621_dma_prep(struct ata_queued_cmd *qc)
 	 * Build S/G table
 	 */
 	idx = 0;
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		buf[idx++] = cpu_to_le32(sg_dma_address(sg));
 		buf[idx++] = cpu_to_le32(sg_dma_len(sg));
 		total_len += sg_dma_len(sg);
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 3e78bc2d9170..aa0df0a4b22a 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -5142,6 +5142,7 @@ static void ipr_build_ata_ioadl(struct ipr_cmnd *ipr_cmd,
 	struct ipr_ioadl_desc *last_ioadl = NULL;
 	int len = qc->nbytes + qc->pad_len;
 	struct scatterlist *sg;
+	unsigned int si;
 
 	if (len == 0)
 		return;
@@ -5159,7 +5160,7 @@ static void ipr_build_ata_ioadl(struct ipr_cmnd *ipr_cmd,
 			cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
 	}
 
-	ata_for_each_sg(sg, qc) {
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
 		ioadl->flags_and_data_len = cpu_to_be32(ioadl_flags | sg_dma_len(sg));
 		ioadl->address = cpu_to_be32(sg_dma_address(sg));
 
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index f78d0605747d..827cfb132f21 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -158,8 +158,8 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
 	struct Scsi_Host *host = sas_ha->core.shost;
 	struct sas_internal *i = to_sas_internal(host->transportt);
 	struct scatterlist *sg;
-	unsigned int num = 0;
 	unsigned int xfer = 0;
+	unsigned int si;
 
 	task = sas_alloc_task(GFP_ATOMIC);
 	if (!task)
@@ -181,17 +181,15 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
 		task->total_xfer_len = qc->nbytes + qc->pad_len;
 		task->num_scatter = qc->pad_len ? qc->n_elem + 1 : qc->n_elem;
 	} else {
-		ata_for_each_sg(sg, qc) {
-			num++;
+		for_each_sg(qc->sg, sg, qc->n_elem, si)
 			xfer += sg->length;
-		}
 
 		task->total_xfer_len = xfer;
-		task->num_scatter = num;
+		task->num_scatter = si;
 	}
 
 	task->data_dir = qc->dma_dir;
-	task->scatter = qc->__sg;
+	task->scatter = qc->sg;
 	task->ata_task.retry_count = 1;
 	task->task_state_flags = SAS_TASK_STATE_PENDING;
 	qc->lldd_task = task;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index acd90ad78417..162f8b5509ac 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -458,7 +458,7 @@ struct ata_queued_cmd {
 	unsigned int		tag;
 	unsigned int		n_elem;
 	unsigned int		n_iter;
-	unsigned int		orig_n_elem;
+	unsigned int		mapped_n_elem;
 
 	int			dma_dir;
 
@@ -471,11 +471,12 @@ struct ata_queued_cmd {
 	struct scatterlist	*cursg;
 	unsigned int		cursg_ofs;
 
+	struct scatterlist	*last_sg;
+	struct scatterlist	saved_last_sg;
 	struct scatterlist	sgent;
-	struct scatterlist	pad_sgent;
+	struct scatterlist	extra_sg[2];
 
-	/* DO NOT iterate over __sg manually, use ata_for_each_sg() */
-	struct scatterlist	*__sg;
+	struct scatterlist	*sg;
 
 	unsigned int		err_mask;
 	struct ata_taskfile	result_tf;
@@ -1123,35 +1124,6 @@ extern void ata_port_pbar_desc(struct ata_port *ap, int bar, ssize_t offset,
 			       const char *name);
 #endif
 
-/*
- * qc helpers
- */
-static inline struct scatterlist *
-ata_qc_first_sg(struct ata_queued_cmd *qc)
-{
-	qc->n_iter = 0;
-	if (qc->n_elem)
-		return qc->__sg;
-	if (qc->pad_len)
-		return &qc->pad_sgent;
-	return NULL;
-}
-
-static inline struct scatterlist *
-ata_qc_next_sg(struct scatterlist *sg, struct ata_queued_cmd *qc)
-{
-	if (sg == &qc->pad_sgent)
-		return NULL;
-	if (++qc->n_iter < qc->n_elem)
-		return sg_next(sg);
-	if (qc->pad_len)
-		return &qc->pad_sgent;
-	return NULL;
-}
-
-#define ata_for_each_sg(sg, qc) \
-	for (sg = ata_qc_first_sg(qc); sg; sg = ata_qc_next_sg(sg, qc))
-
 static inline unsigned int ata_tag_valid(unsigned int tag)
 {
 	return (tag < ATA_MAX_QUEUE) ? 1 : 0;
@@ -1386,15 +1358,17 @@ static inline void ata_tf_init(struct ata_device *dev, struct ata_taskfile *tf)
 static inline void ata_qc_reinit(struct ata_queued_cmd *qc)
 {
 	qc->dma_dir = DMA_NONE;
-	qc->__sg = NULL;
+	qc->sg = NULL;
 	qc->flags = 0;
 	qc->cursg = NULL;
 	qc->cursg_ofs = 0;
 	qc->nbytes = qc->curbytes = 0;
 	qc->n_elem = 0;
+	qc->mapped_n_elem = 0;
 	qc->n_iter = 0;
 	qc->err_mask = 0;
 	qc->pad_len = 0;
+	qc->last_sg = NULL;
 	qc->sect_size = ATA_SECT_SIZE;
 
 	ata_tf_init(qc->dev, &qc->tf);
-- 
cgit v1.2.3


From 0bcc65ad78ae517de16b2ca07a2891f49d44d156 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 5 Dec 2007 16:43:12 +0900
Subject: libata: make qc->nbytes include extra buffers

qc->nbytes didn't use to include extra buffers setup by libata core
layer and my be odd.  This patch makes qc->nbytes include any extra
buffers setup by libata core layer and guaranteed to be aligned on 4
byte boundary.

This value is to be used to program the host controller.  As this
represents the actual length of buffer available to the controller and
the controller must be able to deal with short transfers for ATAPI
commands which can transfer variable length, this shouldn't break any
controllers while making problems like rounding-down and controllers
choking up on odd transfer bytes much less likely.

The unmodified value is stored in new field qc->raw_nbytes.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 14 ++++++++++----
 include/linux/libata.h    |  3 ++-
 2 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e998028302da..ee72994500a3 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4763,13 +4763,15 @@ void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 }
 
 static unsigned int ata_sg_setup_extra(struct ata_queued_cmd *qc,
-				       unsigned int *n_elem_extra)
+				       unsigned int *n_elem_extra,
+				       unsigned int *nbytes_extra)
 {
 	struct ata_port *ap = qc->ap;
 	unsigned int n_elem = qc->n_elem;
 	struct scatterlist *lsg, *copy_lsg = NULL, *tsg = NULL, *esg = NULL;
 
 	*n_elem_extra = 0;
+	*nbytes_extra = 0;
 
 	/* needs padding? */
 	qc->pad_len = qc->nbytes & 3;
@@ -4833,6 +4835,7 @@ static unsigned int ata_sg_setup_extra(struct ata_queued_cmd *qc,
 		esg = &qc->extra_sg[1];
 
 		(*n_elem_extra)++;
+		(*nbytes_extra) += 4 - qc->pad_len;
 	}
 
 	if (copy_lsg)
@@ -4866,11 +4869,11 @@ static unsigned int ata_sg_setup_extra(struct ata_queued_cmd *qc,
 static int ata_sg_setup(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	unsigned int n_elem, n_elem_extra;
+	unsigned int n_elem, n_elem_extra, nbytes_extra;
 
 	VPRINTK("ENTER, ata%u\n", ap->print_id);
 
-	n_elem = ata_sg_setup_extra(qc, &n_elem_extra);
+	n_elem = ata_sg_setup_extra(qc, &n_elem_extra, &nbytes_extra);
 
 	if (n_elem) {
 		n_elem = dma_map_sg(ap->dev, qc->sg, n_elem, qc->dma_dir);
@@ -4885,7 +4888,7 @@ static int ata_sg_setup(struct ata_queued_cmd *qc)
 
 	qc->n_elem = qc->mapped_n_elem = n_elem;
 	qc->n_elem += n_elem_extra;
-
+	qc->nbytes += nbytes_extra;
 	qc->flags |= ATA_QCFLAG_DMAMAP;
 
 	return 0;
@@ -5949,6 +5952,9 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 	 */
 	BUG_ON(ata_is_data(prot) && (!qc->sg || !qc->n_elem || !qc->nbytes));
 
+	/* ata_sg_setup() may update nbytes */
+	qc->raw_nbytes = qc->nbytes;
+
 	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
 				 (ap->flags & ATA_FLAG_PIO_DMA)))
 		if (ata_sg_setup(qc))
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 162f8b5509ac..7b7c78e42077 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -466,6 +466,7 @@ struct ata_queued_cmd {
 	unsigned int		sect_size;
 
 	unsigned int		nbytes;
+	unsigned int		raw_nbytes;
 	unsigned int		curbytes;
 
 	struct scatterlist	*cursg;
@@ -1362,7 +1363,7 @@ static inline void ata_qc_reinit(struct ata_queued_cmd *qc)
 	qc->flags = 0;
 	qc->cursg = NULL;
 	qc->cursg_ofs = 0;
-	qc->nbytes = qc->curbytes = 0;
+	qc->nbytes = qc->raw_nbytes = qc->curbytes = 0;
 	qc->n_elem = 0;
 	qc->mapped_n_elem = 0;
 	qc->n_iter = 0;
-- 
cgit v1.2.3


From 442eacc362c2576aac8ebfd41b99252e28e0f49c Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Wed, 19 Dec 2007 04:25:10 -0500
Subject: libata: make ata_port_queue_task() an internal function

ata_port_queue_task() served a single user:  ata_pio_task()

Rename to ata_pio_queue_task() and un-export it, as nobody outside of
libata-core.c uses it.

Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 24 +++++++++++-------------
 include/linux/libata.h    |  2 --
 2 files changed, 11 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 627703cba9a6..4020a1d3eb2c 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1480,7 +1480,7 @@ unsigned long ata_id_xfermask(const u16 *id)
 }
 
 /**
- *	ata_port_queue_task - Queue port_task
+ *	ata_pio_queue_task - Queue port_task
  *	@ap: The ata_port to queue port_task for
  *	@fn: workqueue function to be scheduled
  *	@data: data for @fn to use
@@ -1492,16 +1492,15 @@ unsigned long ata_id_xfermask(const u16 *id)
  *	one task is active at any given time.
  *
  *	libata core layer takes care of synchronization between
- *	port_task and EH.  ata_port_queue_task() may be ignored for EH
+ *	port_task and EH.  ata_pio_queue_task() may be ignored for EH
  *	synchronization.
  *
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_port_queue_task(struct ata_port *ap, work_func_t fn, void *data,
-			 unsigned long delay)
+static void ata_pio_queue_task(struct ata_port *ap, void *data,
+			       unsigned long delay)
 {
-	PREPARE_DELAYED_WORK(&ap->port_task, fn);
 	ap->port_task_data = data;
 
 	/* may fail if ata_port_flush_task() in progress */
@@ -5618,7 +5617,7 @@ fsm_start:
 		msleep(2);
 		status = ata_busy_wait(ap, ATA_BUSY, 10);
 		if (status & ATA_BUSY) {
-			ata_port_queue_task(ap, ata_pio_task, qc, ATA_SHORT_PAUSE);
+			ata_pio_queue_task(ap, qc, ATA_SHORT_PAUSE);
 			return;
 		}
 	}
@@ -6041,7 +6040,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 		ap->hsm_task_state = HSM_ST_LAST;
 
 		if (qc->tf.flags & ATA_TFLAG_POLLING)
-			ata_port_queue_task(ap, ata_pio_task, qc, 0);
+			ata_pio_queue_task(ap, qc, 0);
 
 		break;
 
@@ -6063,7 +6062,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 		if (qc->tf.flags & ATA_TFLAG_WRITE) {
 			/* PIO data out protocol */
 			ap->hsm_task_state = HSM_ST_FIRST;
-			ata_port_queue_task(ap, ata_pio_task, qc, 0);
+			ata_pio_queue_task(ap, qc, 0);
 
 			/* always send first data block using
 			 * the ata_pio_task() codepath.
@@ -6073,7 +6072,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 			ap->hsm_task_state = HSM_ST;
 
 			if (qc->tf.flags & ATA_TFLAG_POLLING)
-				ata_port_queue_task(ap, ata_pio_task, qc, 0);
+				ata_pio_queue_task(ap, qc, 0);
 
 			/* if polling, ata_pio_task() handles the rest.
 			 * otherwise, interrupt handler takes over from here.
@@ -6094,7 +6093,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 		/* send cdb by polling if no cdb interrupt */
 		if ((!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) ||
 		    (qc->tf.flags & ATA_TFLAG_POLLING))
-			ata_port_queue_task(ap, ata_pio_task, qc, 0);
+			ata_pio_queue_task(ap, qc, 0);
 		break;
 
 	case ATAPI_PROT_DMA:
@@ -6106,7 +6105,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 
 		/* send cdb by polling if no cdb interrupt */
 		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
-			ata_port_queue_task(ap, ata_pio_task, qc, 0);
+			ata_pio_queue_task(ap, qc, 0);
 		break;
 
 	default:
@@ -6722,7 +6721,7 @@ struct ata_port *ata_port_alloc(struct ata_host *host)
 	ap->msg_enable = ATA_MSG_DRV | ATA_MSG_ERR | ATA_MSG_WARN;
 #endif
 
-	INIT_DELAYED_WORK(&ap->port_task, NULL);
+	INIT_DELAYED_WORK(&ap->port_task, ata_pio_task);
 	INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug);
 	INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan);
 	INIT_LIST_HEAD(&ap->eh_done_q);
@@ -7599,7 +7598,6 @@ EXPORT_SYMBOL_GPL(ata_wait_register);
 EXPORT_SYMBOL_GPL(ata_busy_sleep);
 EXPORT_SYMBOL_GPL(ata_wait_after_reset);
 EXPORT_SYMBOL_GPL(ata_wait_ready);
-EXPORT_SYMBOL_GPL(ata_port_queue_task);
 EXPORT_SYMBOL_GPL(ata_scsi_ioctl);
 EXPORT_SYMBOL_GPL(ata_scsi_queuecmd);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 7b7c78e42077..ccb055636f37 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -847,8 +847,6 @@ extern int ata_busy_sleep(struct ata_port *ap,
 			  unsigned long timeout_pat, unsigned long timeout);
 extern void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline);
 extern int ata_wait_ready(struct ata_port *ap, unsigned long deadline);
-extern void ata_port_queue_task(struct ata_port *ap, work_func_t fn,
-				void *data, unsigned long delay);
 extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 			     unsigned long interval_msec,
 			     unsigned long timeout_msec);
-- 
cgit v1.2.3


From 4ca4e439640cd1d3659cbcf60e7a73c2ae0450b3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ZenIV.linux.org.uk>
Date: Sun, 30 Dec 2007 09:32:22 +0000
Subject: libata annotations and fixes

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ahci.c          | 18 +++++++++---------
 drivers/ata/libata-core.c   |  2 +-
 drivers/ata/libata-scsi.c   |  4 ++--
 drivers/ata/pata_cs5520.c   |  2 +-
 drivers/ata/pata_pdc2027x.c |  2 +-
 drivers/ata/sata_promise.h  |  2 +-
 drivers/ata/sata_sx4.c      |  9 +++++----
 include/linux/ata.h         |  4 ++--
 8 files changed, 22 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index cffad07c65bf..49761bc12cf2 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -198,18 +198,18 @@ enum {
 };
 
 struct ahci_cmd_hdr {
-	u32			opts;
-	u32			status;
-	u32			tbl_addr;
-	u32			tbl_addr_hi;
-	u32			reserved[4];
+	__le32			opts;
+	__le32			status;
+	__le32			tbl_addr;
+	__le32			tbl_addr_hi;
+	__le32			reserved[4];
 };
 
 struct ahci_sg {
-	u32			addr;
-	u32			addr_hi;
-	u32			reserved;
-	u32			flags_size;
+	__le32			addr;
+	__le32			addr_hi;
+	__le32			reserved;
+	__le32			flags_size;
 };
 
 struct ahci_host_priv {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 4020a1d3eb2c..8c82193b077e 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4949,7 +4949,7 @@ unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
 
 	/* Transfer trailing 1 byte, if any. */
 	if (unlikely(buflen & 0x01)) {
-		u16 align_buf[1] = { 0 };
+		__le16 align_buf[1] = { 0 };
 		unsigned char *trailing_buf = buf + buflen - 1;
 
 		if (rw == READ) {
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 42bf61599730..15a4c8a20358 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2210,7 +2210,7 @@ unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf,
 
 		/* sector size */
 		ATA_SCSI_RBUF_SET(6, ATA_SECT_SIZE >> 8);
-		ATA_SCSI_RBUF_SET(7, ATA_SECT_SIZE);
+		ATA_SCSI_RBUF_SET(7, ATA_SECT_SIZE & 0xff);
 	} else {
 		/* sector count, 64-bit */
 		ATA_SCSI_RBUF_SET(0, last_lba >> (8 * 7));
@@ -2224,7 +2224,7 @@ unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf,
 
 		/* sector size */
 		ATA_SCSI_RBUF_SET(10, ATA_SECT_SIZE >> 8);
-		ATA_SCSI_RBUF_SET(11, ATA_SECT_SIZE);
+		ATA_SCSI_RBUF_SET(11, ATA_SECT_SIZE & 0xff);
 	}
 
 	return 0;
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index 33f7f0843f4f..d4590f546c49 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -198,7 +198,7 @@ static int __devinit cs5520_init_one(struct pci_dev *pdev, const struct pci_devi
 	};
 	const struct ata_port_info *ppi[2];
 	u8 pcicfg;
-	void *iomap[5];
+	void __iomem *iomap[5];
 	struct ata_host *host;
 	struct ata_ioports *ioaddr;
 	int i, rc;
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index 2622577521a1..028af5dbeed6 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -348,7 +348,7 @@ static unsigned long pdc2027x_mode_filter(struct ata_device *adev, unsigned long
 	ata_id_c_string(pair->id, model_num, ATA_ID_PROD,
 			  ATA_ID_PROD_LEN + 1);
 	/* If the master is a maxtor in UDMA6 then the slave should not use UDMA 6 */
-	if (strstr(model_num, "Maxtor") == 0 && pair->dma_mode == XFER_UDMA_6)
+	if (strstr(model_num, "Maxtor") == NULL && pair->dma_mode == XFER_UDMA_6)
 		mask &= ~ (1 << (6 + ATA_SHIFT_UDMA));
 
 	return ata_pci_default_filter(adev, mask);
diff --git a/drivers/ata/sata_promise.h b/drivers/ata/sata_promise.h
index 6ee5e190262d..00d6000e546f 100644
--- a/drivers/ata/sata_promise.h
+++ b/drivers/ata/sata_promise.h
@@ -46,7 +46,7 @@ static inline unsigned int pdc_pkt_header(struct ata_taskfile *tf,
 					  unsigned int devno, u8 *buf)
 {
 	u8 dev_reg;
-	u32 *buf32 = (u32 *) buf;
+	__le32 *buf32 = (__le32 *) buf;
 
 	/* set control bits (byte 0), zero delay seq id (byte 3),
 	 * and seq id (byte 2)
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 211ba8da64f4..e3d56bc6726d 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -334,7 +334,7 @@ static inline void pdc20621_ata_sg(struct ata_taskfile *tf, u8 *buf,
 {
 	u32 addr;
 	unsigned int dw = PDC_DIMM_APKT_PRD >> 2;
-	u32 *buf32 = (u32 *) buf;
+	__le32 *buf32 = (__le32 *) buf;
 
 	/* output ATA packet S/G table */
 	addr = PDC_20621_DIMM_BASE + PDC_20621_DIMM_DATA +
@@ -356,7 +356,7 @@ static inline void pdc20621_host_sg(struct ata_taskfile *tf, u8 *buf,
 {
 	u32 addr;
 	unsigned int dw = PDC_DIMM_HPKT_PRD >> 2;
-	u32 *buf32 = (u32 *) buf;
+	__le32 *buf32 = (__le32 *) buf;
 
 	/* output Host DMA packet S/G table */
 	addr = PDC_20621_DIMM_BASE + PDC_20621_DIMM_DATA +
@@ -377,7 +377,7 @@ static inline unsigned int pdc20621_ata_pkt(struct ata_taskfile *tf,
 					    unsigned int portno)
 {
 	unsigned int i, dw;
-	u32 *buf32 = (u32 *) buf;
+	__le32 *buf32 = (__le32 *) buf;
 	u8 dev_reg;
 
 	unsigned int dimm_sg = PDC_20621_DIMM_BASE +
@@ -429,7 +429,8 @@ static inline void pdc20621_host_pkt(struct ata_taskfile *tf, u8 *buf,
 				     unsigned int portno)
 {
 	unsigned int dw;
-	u32 tmp, *buf32 = (u32 *) buf;
+	u32 tmp;
+	__le32 *buf32 = (__le32 *) buf;
 
 	unsigned int host_sg = PDC_20621_DIMM_BASE +
 			       (PDC_DIMM_WINDOW_STEP * portno) +
diff --git a/include/linux/ata.h b/include/linux/ata.h
index bc55471a4b2c..78bbacaed8c4 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -354,8 +354,8 @@ enum ata_ioctls {
 /* core structures */
 
 struct ata_prd {
-	u32			addr;
-	u32			flags_len;
+	__le32			addr;
+	__le32			flags_len;
 };
 
 struct ata_taskfile {
-- 
cgit v1.2.3


From 4e6b79fa61091a0ed9b0af0f573cc257772cd88d Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Fri, 18 Jan 2008 18:36:28 +0900
Subject: libata: factor out ata_pci_activate_sff_host() from ata_pci_one()

Factor out ata_pci_activate_sff_host() from ata_pci_one().  This does
about the same thing as ata_host_activate() but needs to be separate
because SFF controllers use different and multiple IRQs in legacy
mode.

This will be used to make SFF LLD initialization more flexible.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c |   1 +
 drivers/ata/libata-sff.c  | 187 +++++++++++++++++++++++++---------------------
 include/linux/libata.h    |   3 +
 3 files changed, 107 insertions(+), 84 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 8c82193b077e..ce803d18e96a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -7629,6 +7629,7 @@ EXPORT_SYMBOL_GPL(pci_test_config_bits);
 EXPORT_SYMBOL_GPL(ata_pci_init_sff_host);
 EXPORT_SYMBOL_GPL(ata_pci_init_bmdma);
 EXPORT_SYMBOL_GPL(ata_pci_prepare_sff_host);
+EXPORT_SYMBOL_GPL(ata_pci_activate_sff_host);
 EXPORT_SYMBOL_GPL(ata_pci_init_one);
 EXPORT_SYMBOL_GPL(ata_pci_remove_one);
 #ifdef CONFIG_PM
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 97dd8f2796b7..60cd4b179766 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -713,6 +713,99 @@ int ata_pci_prepare_sff_host(struct pci_dev *pdev,
 	return rc;
 }
 
+/**
+ *	ata_pci_activate_sff_host - start SFF host, request IRQ and register it
+ *	@host: target SFF ATA host
+ *	@irq_handler: irq_handler used when requesting IRQ(s)
+ *	@sht: scsi_host_template to use when registering the host
+ *
+ *	This is the counterpart of ata_host_activate() for SFF ATA
+ *	hosts.  This separate helper is necessary because SFF hosts
+ *	use two separate interrupts in legacy mode.
+ *
+ *	LOCKING:
+ *	Inherited from calling layer (may sleep).
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
+ */
+int ata_pci_activate_sff_host(struct ata_host *host,
+			      irq_handler_t irq_handler,
+			      struct scsi_host_template *sht)
+{
+	struct device *dev = host->dev;
+	struct pci_dev *pdev = to_pci_dev(dev);
+	const char *drv_name = dev_driver_string(host->dev);
+	int legacy_mode = 0, rc;
+
+	rc = ata_host_start(host);
+	if (rc)
+		return rc;
+
+	if ((pdev->class >> 8) == PCI_CLASS_STORAGE_IDE) {
+		u8 tmp8, mask;
+
+		/* TODO: What if one channel is in native mode ... */
+		pci_read_config_byte(pdev, PCI_CLASS_PROG, &tmp8);
+		mask = (1 << 2) | (1 << 0);
+		if ((tmp8 & mask) != mask)
+			legacy_mode = 1;
+#if defined(CONFIG_NO_ATA_LEGACY)
+		/* Some platforms with PCI limits cannot address compat
+		   port space. In that case we punt if their firmware has
+		   left a device in compatibility mode */
+		if (legacy_mode) {
+			printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n");
+			return -EOPNOTSUPP;
+		}
+#endif
+	}
+
+	if (!devres_open_group(dev, NULL, GFP_KERNEL))
+		return -ENOMEM;
+
+	if (!legacy_mode && pdev->irq) {
+		rc = devm_request_irq(dev, pdev->irq, irq_handler,
+				      IRQF_SHARED, drv_name, host);
+		if (rc)
+			goto out;
+
+		ata_port_desc(host->ports[0], "irq %d", pdev->irq);
+		ata_port_desc(host->ports[1], "irq %d", pdev->irq);
+	} else if (legacy_mode) {
+		if (!ata_port_is_dummy(host->ports[0])) {
+			rc = devm_request_irq(dev, ATA_PRIMARY_IRQ(pdev),
+					      irq_handler, IRQF_SHARED,
+					      drv_name, host);
+			if (rc)
+				goto out;
+
+			ata_port_desc(host->ports[0], "irq %d",
+				      ATA_PRIMARY_IRQ(pdev));
+		}
+
+		if (!ata_port_is_dummy(host->ports[1])) {
+			rc = devm_request_irq(dev, ATA_SECONDARY_IRQ(pdev),
+					      irq_handler, IRQF_SHARED,
+					      drv_name, host);
+			if (rc)
+				goto out;
+
+			ata_port_desc(host->ports[1], "irq %d",
+				      ATA_SECONDARY_IRQ(pdev));
+		}
+	}
+
+	rc = ata_host_register(host, sht);
+ out:
+	if (rc == 0)
+		devres_remove_group(dev, NULL);
+	else
+		devres_release_group(dev, NULL);
+
+	return rc;
+}
+
 /**
  *	ata_pci_init_one - Initialize/register PCI IDE host controller
  *	@pdev: Controller to be initialized
@@ -742,9 +835,6 @@ int ata_pci_init_one(struct pci_dev *pdev,
 	struct device *dev = &pdev->dev;
 	const struct ata_port_info *pi = NULL;
 	struct ata_host *host = NULL;
-	const char *drv_name = dev_driver_string(&pdev->dev);
-	u8 mask;
-	int legacy_mode = 0;
 	int i, rc;
 
 	DPRINTK("ENTER\n");
@@ -766,95 +856,24 @@ int ata_pci_init_one(struct pci_dev *pdev,
 	if (!devres_open_group(dev, NULL, GFP_KERNEL))
 		return -ENOMEM;
 
-	/* FIXME: Really for ATA it isn't safe because the device may be
-	   multi-purpose and we want to leave it alone if it was already
-	   enabled. Secondly for shared use as Arjan says we want refcounting
-
-	   Checking dev->is_enabled is insufficient as this is not set at
-	   boot for the primary video which is BIOS enabled
-	  */
-
 	rc = pcim_enable_device(pdev);
 	if (rc)
-		goto err_out;
+		goto out;
 
-	if ((pdev->class >> 8) == PCI_CLASS_STORAGE_IDE) {
-		u8 tmp8;
-
-		/* TODO: What if one channel is in native mode ... */
-		pci_read_config_byte(pdev, PCI_CLASS_PROG, &tmp8);
-		mask = (1 << 2) | (1 << 0);
-		if ((tmp8 & mask) != mask)
-			legacy_mode = 1;
-#if defined(CONFIG_NO_ATA_LEGACY)
-		/* Some platforms with PCI limits cannot address compat
-		   port space. In that case we punt if their firmware has
-		   left a device in compatibility mode */
-		if (legacy_mode) {
-			printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n");
-			rc = -EOPNOTSUPP;
-			goto err_out;
-		}
-#endif
-	}
-
-	/* prepare host */
+	/* prepare and activate SFF host */
 	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
 	if (rc)
-		goto err_out;
+		goto out;
 
 	pci_set_master(pdev);
+	rc = ata_pci_activate_sff_host(host, pi->port_ops->irq_handler,
+				       pi->sht);
+ out:
+	if (rc == 0)
+		devres_remove_group(&pdev->dev, NULL);
+	else
+		devres_release_group(&pdev->dev, NULL);
 
-	/* start host and request IRQ */
-	rc = ata_host_start(host);
-	if (rc)
-		goto err_out;
-
-	if (!legacy_mode && pdev->irq) {
-		/* We may have no IRQ assigned in which case we can poll. This
-		   shouldn't happen on a sane system but robustness is cheap
-		   in this case */
-		rc = devm_request_irq(dev, pdev->irq, pi->port_ops->irq_handler,
-				      IRQF_SHARED, drv_name, host);
-		if (rc)
-			goto err_out;
-
-		ata_port_desc(host->ports[0], "irq %d", pdev->irq);
-		ata_port_desc(host->ports[1], "irq %d", pdev->irq);
-	} else if (legacy_mode) {
-		if (!ata_port_is_dummy(host->ports[0])) {
-			rc = devm_request_irq(dev, ATA_PRIMARY_IRQ(pdev),
-					      pi->port_ops->irq_handler,
-					      IRQF_SHARED, drv_name, host);
-			if (rc)
-				goto err_out;
-
-			ata_port_desc(host->ports[0], "irq %d",
-				      ATA_PRIMARY_IRQ(pdev));
-		}
-
-		if (!ata_port_is_dummy(host->ports[1])) {
-			rc = devm_request_irq(dev, ATA_SECONDARY_IRQ(pdev),
-					      pi->port_ops->irq_handler,
-					      IRQF_SHARED, drv_name, host);
-			if (rc)
-				goto err_out;
-
-			ata_port_desc(host->ports[1], "irq %d",
-				      ATA_SECONDARY_IRQ(pdev));
-		}
-	}
-
-	/* register */
-	rc = ata_host_register(host, pi->sht);
-	if (rc)
-		goto err_out;
-
-	devres_remove_group(dev, NULL);
-	return 0;
-
-err_out:
-	devres_release_group(dev, NULL);
 	return rc;
 }
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index ccb055636f37..4374c4277780 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1033,6 +1033,9 @@ extern int ata_pci_init_bmdma(struct ata_host *host);
 extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
 				    const struct ata_port_info * const * ppi,
 				    struct ata_host **r_host);
+extern int ata_pci_activate_sff_host(struct ata_host *host,
+				     irq_handler_t irq_handler,
+				     struct scsi_host_template *sht);
 extern int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits);
 extern unsigned long ata_pci_default_filter(struct ata_device *dev,
 					    unsigned long xfer_mask);
-- 
cgit v1.2.3


From 5e8f757cb2e0f67bf43f71d5180a8bf0ab3484eb Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Sat, 19 Jan 2008 16:07:58 +0000
Subject: ata_generic: Cenatek support

Not much to do here. It's an ata memory as disk.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ata_generic.c | 7 ++++++-
 include/linux/pci_ids.h   | 3 +++
 2 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index d11a4f1fe86a..20534202fc79 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -26,7 +26,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "ata_generic"
-#define DRV_VERSION "0.2.13"
+#define DRV_VERSION "0.2.15"
 
 /*
  *	A generic parallel ATA driver using libata
@@ -48,11 +48,15 @@ static int generic_set_mode(struct ata_link *link, struct ata_device **unused)
 	struct ata_port *ap = link->ap;
 	int dma_enabled = 0;
 	struct ata_device *dev;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
 	/* Bits 5 and 6 indicate if DMA is active on master/slave */
 	if (ap->ioaddr.bmdma_addr)
 		dma_enabled = ioread8(ap->ioaddr.bmdma_addr + ATA_DMA_STATUS);
 
+	if (pdev->vendor == PCI_VENDOR_ID_CENATEK)
+		dma_enabled = 0xFF;
+
 	ata_link_for_each_dev(dev, link) {
 		if (!ata_dev_enabled(dev))
 			continue;
@@ -201,6 +205,7 @@ static struct pci_device_id ata_generic[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_HINT,   PCI_DEVICE_ID_HINT_VXPROII_IDE), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_VIA,    PCI_DEVICE_ID_VIA_82C561), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_OPTI,   PCI_DEVICE_ID_OPTI_82C558), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_CENATEK,PCI_DEVICE_ID_CENATEK_IDE), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2),  },
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 7f2215139e9a..1fbd0256e86b 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2066,6 +2066,9 @@
 #define PCI_VENDOR_ID_NETCELL		0x169c
 #define PCI_DEVICE_ID_REVOLUTION	0x0044
 
+#define PCI_VENDOR_ID_CENATEK		0x16CA
+#define PCI_DEVICE_ID_CENATEK_IDE	0x0001
+
 #define PCI_VENDOR_ID_VITESSE		0x1725
 #define PCI_DEVICE_ID_VITESSE_VSC7174	0x7174
 
-- 
cgit v1.2.3


From fd1109711d7f76126e7cef947999f139b198dc15 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 2 Jan 2008 18:48:47 -0600
Subject: [SCSI] attribute_container: update to use the group interface

This patch is the beginning of moving the attribute_containers to use
attribute groups exclusively.  The attr element is now deprecated and
will eventually be removed (along with all the hand rolled code for
doing exactly what attribute groups do) when all the consumers are
converted to attribute groups.

Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/base/attribute_container.c  | 14 ++++++++++++--
 include/linux/attribute_container.h |  1 +
 2 files changed, 13 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/attribute_container.c b/drivers/base/attribute_container.c
index 7370d7cf5988..5bf25c6f966c 100644
--- a/drivers/base/attribute_container.c
+++ b/drivers/base/attribute_container.c
@@ -320,9 +320,14 @@ attribute_container_add_attrs(struct class_device *classdev)
 	struct class_device_attribute **attrs =	cont->attrs;
 	int i, error;
 
-	if (!attrs)
+	BUG_ON(attrs && cont->grp);
+
+	if (!attrs && !cont->grp)
 		return 0;
 
+	if (cont->grp)
+		return sysfs_create_group(&classdev->kobj, cont->grp);
+
 	for (i = 0; attrs[i]; i++) {
 		error = class_device_create_file(classdev, attrs[i]);
 		if (error)
@@ -378,9 +383,14 @@ attribute_container_remove_attrs(struct class_device *classdev)
 	struct class_device_attribute **attrs =	cont->attrs;
 	int i;
 
-	if (!attrs)
+	if (!attrs && !cont->grp)
 		return;
 
+	if (cont->grp) {
+		sysfs_remove_group(&classdev->kobj, cont->grp);
+		return ;
+	}
+
 	for (i = 0; attrs[i]; i++)
 		class_device_remove_file(classdev, attrs[i]);
 }
diff --git a/include/linux/attribute_container.h b/include/linux/attribute_container.h
index 8ff274933948..f5582332af04 100644
--- a/include/linux/attribute_container.h
+++ b/include/linux/attribute_container.h
@@ -17,6 +17,7 @@ struct attribute_container {
 	struct list_head	node;
 	struct klist		containers;
 	struct class		*class;
+	struct attribute_group	*grp;
 	struct class_device_attribute **attrs;
 	int (*match)(struct attribute_container *, struct device *);
 #define	ATTRIBUTE_CONTAINER_NO_CLASSDEVS	0x01
-- 
cgit v1.2.3


From d4acd722b7bb5f48b9fc3848e8c2a845b100d84f Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 31 Oct 2007 09:38:04 -0500
Subject: [SCSI] sysfs: add filter function to groups

This patch allows the various users of attribute_groups to selectively
allow the appearance of group attributes.  The primary consumer of
this will be the transport classes in which we currently have
elaborate attribute selection algorithms to do this same thing.

Acked-by: Greg KH <greg@kroah.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 fs/sysfs/group.c      | 26 ++++++++++++++++----------
 include/linux/sysfs.h |  2 ++
 kernel/params.c       |  2 +-
 3 files changed, 19 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c
index d1972374655a..0871c3dadce1 100644
--- a/fs/sysfs/group.c
+++ b/fs/sysfs/group.c
@@ -16,25 +16,31 @@
 #include "sysfs.h"
 
 
-static void remove_files(struct sysfs_dirent *dir_sd,
+static void remove_files(struct sysfs_dirent *dir_sd, struct kobject *kobj,
 			 const struct attribute_group *grp)
 {
 	struct attribute *const* attr;
+	int i;
 
-	for (attr = grp->attrs; *attr; attr++)
-		sysfs_hash_and_remove(dir_sd, (*attr)->name);
+	for (i = 0, attr = grp->attrs; *attr; i++, attr++)
+		if (!grp->is_visible ||
+		    grp->is_visible(kobj, *attr, i))
+			sysfs_hash_and_remove(dir_sd, (*attr)->name);
 }
 
-static int create_files(struct sysfs_dirent *dir_sd,
+static int create_files(struct sysfs_dirent *dir_sd, struct kobject *kobj,
 			const struct attribute_group *grp)
 {
 	struct attribute *const* attr;
-	int error = 0;
+	int error = 0, i;
 
-	for (attr = grp->attrs; *attr && !error; attr++)
-		error = sysfs_add_file(dir_sd, *attr, SYSFS_KOBJ_ATTR);
+	for (i = 0, attr = grp->attrs; *attr && !error; i++, attr++)
+		if (!grp->is_visible ||
+		    grp->is_visible(kobj, *attr, i))
+			error |=
+				sysfs_add_file(dir_sd, *attr, SYSFS_KOBJ_ATTR);
 	if (error)
-		remove_files(dir_sd, grp);
+		remove_files(dir_sd, kobj, grp);
 	return error;
 }
 
@@ -54,7 +60,7 @@ int sysfs_create_group(struct kobject * kobj,
 	} else
 		sd = kobj->sd;
 	sysfs_get(sd);
-	error = create_files(sd, grp);
+	error = create_files(sd, kobj, grp);
 	if (error) {
 		if (grp->name)
 			sysfs_remove_subdir(sd);
@@ -75,7 +81,7 @@ void sysfs_remove_group(struct kobject * kobj,
 	} else
 		sd = sysfs_get(dir_sd);
 
-	remove_files(sd, grp);
+	remove_files(sd, kobj, grp);
 	if (grp->name)
 		sysfs_remove_subdir(sd);
 
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 149ab62329e2..802710438a9e 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -32,6 +32,8 @@ struct attribute {
 
 struct attribute_group {
 	const char		*name;
+	int			(*is_visible)(struct kobject *,
+					      struct attribute *, int);
 	struct attribute	**attrs;
 };
 
diff --git a/kernel/params.c b/kernel/params.c
index 7686417ee00e..dfef46474e55 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -472,7 +472,7 @@ param_sysfs_setup(struct module_kobject *mk,
 			sizeof(mp->grp.attrs[0]));
 	size[1] = (valid_attrs + 1) * sizeof(mp->grp.attrs[0]);
 
-	mp = kmalloc(size[0] + size[1], GFP_KERNEL);
+	mp = kzalloc(size[0] + size[1], GFP_KERNEL);
 	if (!mp)
 		return ERR_PTR(-ENOMEM);
 
-- 
cgit v1.2.3


From a0899d4df534d2bcf671b0f647b809842309a9ae Mon Sep 17 00:00:00 2001
From: Hans de Goede <j.w.r.degoede@hhs.nl>
Date: Sun, 20 Jan 2008 11:12:26 +0100
Subject: [SCSI] sd: add fix for devices with last sector access problems

This patch adds a new scsi_device flag (last_sector_bug) for devices
which contain a bug where the device crashes when the last sector is
read in a larger then 1 sector read.

This is for example the case with sdcards in the HP PSC1350 printer
cardreader and in the HP PSC1610 printer cardreader.

Signed-off-by: Hans de Goede <j.w.r.degoede@hhs.nl>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sd.c          | 9 +++++++++
 include/scsi/scsi_device.h | 1 +
 2 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 212f6bcfd457..24eba3118b5a 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -395,6 +395,15 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq)
 		goto out;
 	}
 
+	/*
+	 * Some devices (some sdcards for one) don't like it if the
+	 * last sector gets read in a larger then 1 sector read.
+	 */
+	if (unlikely(sdp->last_sector_bug &&
+	    rq->nr_sectors > sdp->sector_size / 512 &&
+	    block + this_count == get_capacity(disk)))
+		this_count -= sdp->sector_size / 512;
+
 	SCSI_LOG_HLQUEUE(2, scmd_printk(KERN_INFO, SCpnt, "block=%llu\n",
 					(unsigned long long)block));
 
diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index e0c645ac5014..3c8f898b160e 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -139,6 +139,7 @@ struct scsi_device {
 	unsigned fix_capacity:1;	/* READ_CAPACITY is too high by 1 */
 	unsigned guess_capacity:1;	/* READ_CAPACITY might be too high by 1 */
 	unsigned retry_hwerror:1;	/* Retry HARDWARE_ERROR */
+	unsigned last_sector_bug:1;	/* Always read last sector in a 1 sector read */
 
 	DECLARE_BITMAP(supported_events, SDEV_EVT_MAXBITS); /* supported events */
 	struct list_head event_list;	/* asserted events */
-- 
cgit v1.2.3


From b30c2fc1113edfb2371427c10503ff942b0a0370 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Sun, 20 Jan 2008 09:09:40 -0600
Subject: [SCSI] scsi.h: add macro for enclosure bit of inquiry data

The macro tells us whether the device is (or contains) an enclosure device.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 include/scsi/scsi_device.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index 3c8f898b160e..ab7acbe80960 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -387,6 +387,10 @@ static inline int scsi_device_qas(struct scsi_device *sdev)
 		return 0;
 	return sdev->inquiry[56] & 0x02;
 }
+static inline int scsi_device_enclosure(struct scsi_device *sdev)
+{
+	return sdev->inquiry[6] & (1<<6);
+}
 
 #define MODULE_ALIAS_SCSI_DEVICE(type) \
 	MODULE_ALIAS("scsi:t-" __stringify(type) "*")
-- 
cgit v1.2.3


From de25deb18016f66dcdede165d07654559bb332bc Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 16 Jan 2008 13:32:17 +0900
Subject: [SCSI] use dynamically allocated sense buffer

This removes static array sense_buffer in scsi_cmnd and uses
dynamically allocated sense_buffer (with GFP_DMA).

The reason for doing this is that some architectures need cacheline
aligned buffer for DMA:

http://lkml.org/lkml/2007/11/19/2

The problems are that scsi_eh_prep_cmnd puts scsi_cmnd::sense_buffer
to sglist and some LLDs directly DMA to scsi_cmnd::sense_buffer. It's
necessary to DMA to scsi_cmnd::sense_buffer safely. This patch solves
these issues.

__scsi_get_command allocates sense_buffer via kmem_cache_alloc and
attaches it to a scsi_cmnd so everything just work as before.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/hosts.c     |  9 ++++++-
 drivers/scsi/scsi.c      | 61 ++++++++++++++++++++++++++++++++++++++++++++++--
 drivers/scsi/scsi_priv.h |  2 ++
 include/scsi/scsi_cmnd.h |  2 +-
 4 files changed, 70 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 9a10b4335e76..f5d3fbb55717 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -268,6 +268,7 @@ static void scsi_host_dev_release(struct device *dev)
 	}
 
 	scsi_destroy_command_freelist(shost);
+	scsi_destroy_command_sense_buffer(shost);
 	if (shost->bqt)
 		blk_free_tags(shost->bqt);
 
@@ -372,10 +373,14 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 	else
 		shost->dma_boundary = 0xffffffff;
 
-	rval = scsi_setup_command_freelist(shost);
+	rval = scsi_setup_command_sense_buffer(shost);
 	if (rval)
 		goto fail_kfree;
 
+	rval = scsi_setup_command_freelist(shost);
+	if (rval)
+		goto fail_destroy_sense;
+
 	device_initialize(&shost->shost_gendev);
 	snprintf(shost->shost_gendev.bus_id, BUS_ID_SIZE, "host%d",
 		shost->host_no);
@@ -399,6 +404,8 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 
  fail_destroy_freelist:
 	scsi_destroy_command_freelist(shost);
+ fail_destroy_sense:
+	scsi_destroy_command_sense_buffer(shost);
  fail_kfree:
 	kfree(shost);
 	return NULL;
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 54ff611b8677..0a4a5b8b87c6 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -161,6 +161,9 @@ static struct scsi_host_cmd_pool scsi_cmd_dma_pool = {
 
 static DEFINE_MUTEX(host_cmd_pool_mutex);
 
+static struct kmem_cache *sense_buffer_slab;
+static int sense_buffer_slab_users;
+
 /**
  * __scsi_get_command - Allocate a struct scsi_cmnd
  * @shost: host to transmit command
@@ -172,6 +175,7 @@ static DEFINE_MUTEX(host_cmd_pool_mutex);
 struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost, gfp_t gfp_mask)
 {
 	struct scsi_cmnd *cmd;
+	unsigned char *buf;
 
 	cmd = kmem_cache_alloc(shost->cmd_pool->slab,
 			gfp_mask | shost->cmd_pool->gfp_mask);
@@ -186,6 +190,21 @@ struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost, gfp_t gfp_mask)
 			list_del_init(&cmd->list);
 		}
 		spin_unlock_irqrestore(&shost->free_list_lock, flags);
+
+		if (cmd) {
+			buf = cmd->sense_buffer;
+			memset(cmd, 0, sizeof(*cmd));
+			cmd->sense_buffer = buf;
+		}
+	} else {
+		buf = kmem_cache_alloc(sense_buffer_slab, __GFP_DMA|gfp_mask);
+		if (likely(buf)) {
+			memset(cmd, 0, sizeof(*cmd));
+			cmd->sense_buffer = buf;
+		} else {
+			kmem_cache_free(shost->cmd_pool->slab, cmd);
+			cmd = NULL;
+		}
 	}
 
 	return cmd;
@@ -212,7 +231,6 @@ struct scsi_cmnd *scsi_get_command(struct scsi_device *dev, gfp_t gfp_mask)
 	if (likely(cmd != NULL)) {
 		unsigned long flags;
 
-		memset(cmd, 0, sizeof(*cmd));
 		cmd->device = dev;
 		init_timer(&cmd->eh_timeout);
 		INIT_LIST_HEAD(&cmd->list);
@@ -246,8 +264,10 @@ void __scsi_put_command(struct Scsi_Host *shost, struct scsi_cmnd *cmd,
 	}
 	spin_unlock_irqrestore(&shost->free_list_lock, flags);
 
-	if (likely(cmd != NULL))
+	if (likely(cmd != NULL)) {
+		kmem_cache_free(sense_buffer_slab, cmd->sense_buffer);
 		kmem_cache_free(shost->cmd_pool->slab, cmd);
+	}
 
 	put_device(dev);
 }
@@ -290,6 +310,7 @@ int scsi_setup_command_freelist(struct Scsi_Host *shost)
 {
 	struct scsi_host_cmd_pool *pool;
 	struct scsi_cmnd *cmd;
+	unsigned char *sense_buffer;
 
 	spin_lock_init(&shost->free_list_lock);
 	INIT_LIST_HEAD(&shost->free_list);
@@ -319,9 +340,18 @@ int scsi_setup_command_freelist(struct Scsi_Host *shost)
 			GFP_KERNEL | shost->cmd_pool->gfp_mask);
 	if (!cmd)
 		goto fail2;
+
+	sense_buffer = kmem_cache_alloc(sense_buffer_slab,
+					GFP_KERNEL | __GFP_DMA);
+	if (!sense_buffer)
+		goto destroy_backup;
+
+	cmd->sense_buffer = sense_buffer;
 	list_add(&cmd->list, &shost->free_list);
 	return 0;
 
+destroy_backup:
+	kmem_cache_free(shost->cmd_pool->slab, cmd);
  fail2:
 	mutex_lock(&host_cmd_pool_mutex);
 	if (!--pool->users)
@@ -342,6 +372,7 @@ void scsi_destroy_command_freelist(struct Scsi_Host *shost)
 
 		cmd = list_entry(shost->free_list.next, struct scsi_cmnd, list);
 		list_del_init(&cmd->list);
+		kmem_cache_free(sense_buffer_slab, cmd->sense_buffer);
 		kmem_cache_free(shost->cmd_pool->slab, cmd);
 	}
 
@@ -351,6 +382,32 @@ void scsi_destroy_command_freelist(struct Scsi_Host *shost)
 	mutex_unlock(&host_cmd_pool_mutex);
 }
 
+int scsi_setup_command_sense_buffer(struct Scsi_Host *shost)
+{
+	mutex_lock(&host_cmd_pool_mutex);
+	if (!sense_buffer_slab_users) {
+		sense_buffer_slab = kmem_cache_create("scsi_sense_buffer",
+						      SCSI_SENSE_BUFFERSIZE,
+						      0, SLAB_CACHE_DMA, NULL);
+		if (!sense_buffer_slab) {
+			mutex_unlock(&host_cmd_pool_mutex);
+			return -ENOMEM;
+		}
+	}
+	sense_buffer_slab_users++;
+	mutex_unlock(&host_cmd_pool_mutex);
+
+	return 0;
+}
+
+void scsi_destroy_command_sense_buffer(struct Scsi_Host *shost)
+{
+	mutex_lock(&host_cmd_pool_mutex);
+	if (!--sense_buffer_slab_users)
+		kmem_cache_destroy(sense_buffer_slab);
+	mutex_unlock(&host_cmd_pool_mutex);
+}
+
 #ifdef CONFIG_SCSI_LOGGING
 void scsi_log_send(struct scsi_cmnd *cmd)
 {
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index 3f34e9376b0a..55c6f71a9aee 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -27,6 +27,8 @@ extern void scsi_exit_hosts(void);
 extern int scsi_dispatch_cmd(struct scsi_cmnd *cmd);
 extern int scsi_setup_command_freelist(struct Scsi_Host *shost);
 extern void scsi_destroy_command_freelist(struct Scsi_Host *shost);
+extern int scsi_setup_command_sense_buffer(struct Scsi_Host *shost);
+extern void scsi_destroy_command_sense_buffer(struct Scsi_Host *shost);
 extern void __scsi_done(struct scsi_cmnd *cmd);
 #ifdef CONFIG_SCSI_LOGGING
 void scsi_log_send(struct scsi_cmnd *cmd);
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index 3f47e522a1ec..abd7479ff452 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -88,7 +88,7 @@ struct scsi_cmnd {
 				   	   working on */
 
 #define SCSI_SENSE_BUFFERSIZE 	96
-	unsigned char sense_buffer[SCSI_SENSE_BUFFERSIZE];
+	unsigned char *sense_buffer;
 				/* obtained by REQUEST SENSE when
 				 * CHECK CONDITION is received on original
 				 * command (auto-sense) */
-- 
cgit v1.2.3


From 81b4e1f6269cea345f17d3aa349ec9beb31a8cd3 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Wed, 16 Jan 2008 17:20:37 -0500
Subject: DMI: move dmi_available declaration to linux/dmi.h

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/firmware/dmi-id.c | 2 --
 include/linux/dmi.h       | 2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/firmware/dmi-id.c b/drivers/firmware/dmi-id.c
index bc132d8f79cb..313c99cbdc62 100644
--- a/drivers/firmware/dmi-id.c
+++ b/drivers/firmware/dmi-id.c
@@ -173,8 +173,6 @@ static struct device *dmi_dev;
 	if (dmi_get_system_info(_field)) \
 		sys_dmi_attributes[i++] = &sys_dmi_##_name##_attr.dev_attr.attr;
 
-extern int dmi_available;
-
 /* In a separate function to keep gcc 3.2 happy - do NOT merge this in
    dmi_id_init! */
 static void __init dmi_id_init_attr_table(void)
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index 00fc7a9c35ec..b1251b2af568 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -78,6 +78,7 @@ extern const struct dmi_device * dmi_find_device(int type, const char *name,
 extern void dmi_scan_machine(void);
 extern int dmi_get_year(int field);
 extern int dmi_name_in_vendors(const char *str);
+extern int dmi_available;
 
 #else
 
@@ -87,6 +88,7 @@ static inline const struct dmi_device * dmi_find_device(int type, const char *na
 	const struct dmi_device *from) { return NULL; }
 static inline int dmi_get_year(int year) { return 0; }
 static inline int dmi_name_in_vendors(const char *s) { return 0; }
+#define dmi_available 0
 
 #endif
 
-- 
cgit v1.2.3


From f89e3b0620a0dc19f313218f55373b9361142203 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Wed, 23 Jan 2008 16:36:45 -0500
Subject: DMI: create dmi_get_slot()

This simply allows other sub-systems (such as ACPI)
to access and print out slots in static dmi_ident[].

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/firmware/dmi_scan.c | 8 ++++++++
 include/linux/dmi.h         | 2 ++
 2 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index 0cdadea7a40e..5e596a7e3601 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -470,3 +470,11 @@ int dmi_get_year(int field)
 	return year;
 }
 
+/**
+ *	dmi_get_slot - return dmi_ident[slot]
+ *	@slot:	index into dmi_ident[]
+ */
+char *dmi_get_slot(int slot)
+{
+	return(dmi_ident[slot]);
+}
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index b1251b2af568..5b42a659a308 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -79,6 +79,7 @@ extern void dmi_scan_machine(void);
 extern int dmi_get_year(int field);
 extern int dmi_name_in_vendors(const char *str);
 extern int dmi_available;
+extern char *dmi_get_slot(int slot);
 
 #else
 
@@ -89,6 +90,7 @@ static inline const struct dmi_device * dmi_find_device(int type, const char *na
 static inline int dmi_get_year(int year) { return 0; }
 static inline int dmi_name_in_vendors(const char *s) { return 0; }
 #define dmi_available 0
+static inline char *dmi_get_slot(int slot) { return NULL; }
 
 #endif
 
-- 
cgit v1.2.3


From d4b7dc499daae909e62dc260b95cd618f2970ded Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Wed, 23 Jan 2008 20:50:56 -0500
Subject: ACPI: make _OSI(Linux) console messages smarter

If BIOS invokes _OSI(Linux), the kernel response
depends on what the ACPI DMI list knows about the system,
and that is reflectd in dmesg:

1) System unknown to DMI:

ACPI: BIOS _OSI(Linux) query ignored
ACPI: DMI System Vendor: LENOVO
ACPI: DMI Product Name: 7661W1P
ACPI: DMI Product Version: ThinkPad T61
ACPI: DMI Board Name: 7661W1P
ACPI: DMI BIOS Vendor: LENOVO
ACPI: DMI BIOS Date: 10/18/2007
ACPI: Please send DMI info above to linux-acpi@vger.kernel.org
ACPI: If "acpi_osi=Linux" works better, please notify linux-acpi@vger.kernel.org

2) System known to DMI, but effect of OSI(Linux) unknown:

ACPI: DMI detected: Lenovo ThinkPad T61
...
ACPI: BIOS _OSI(Linux) query ignored via DMI
ACPI: If "acpi_osi=Linux" works better, please notify linux-acpi@vger.kernel.org

3) System known to DMI, which disables _OSI(Linux):

ACPI: DMI detected: Lenovo ThinkPad T61
...
ACPI: BIOS _OSI(Linux) query ignored via DMI

4) System known to DMI, which enable _OSI(Linux):

ACPI: DMI detected: Lenovo ThinkPad T61
ACPI: Added _OSI(Linux)
...
ACPI: BIOS _OSI(Linux) query honored via DMI

cmdline overrides take precidence over the built-in
default and the DMI prescribed default.
cmdline "acpi_osi=Linux" results in:

ACPI: BIOS _OSI(Linux) query honored via cmdline

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/blacklist.c |  11 +++++
 drivers/acpi/osl.c       | 122 +++++++++++++++++++++++++++++++++++++++--------
 include/linux/acpi.h     |   7 ++-
 3 files changed, 118 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/blacklist.c b/drivers/acpi/blacklist.c
index 3ec110ce00c8..018fc16c44ce 100644
--- a/drivers/acpi/blacklist.c
+++ b/drivers/acpi/blacklist.c
@@ -3,6 +3,7 @@
  *
  *  Check to see if the given machine has a known bad ACPI BIOS
  *  or if the BIOS is too old.
+ *  Check given machine against acpi_osi_dmi_table[].
  *
  *  Copyright (C) 2004 Len Brown <len.brown@intel.com>
  *  Copyright (C) 2002 Andy Grover <andrew.grover@intel.com>
@@ -50,6 +51,8 @@ struct acpi_blacklist_item {
 	u32 is_critical_error;
 };
 
+static struct dmi_system_id acpi_osi_dmi_table[] __initdata;
+
 /*
  * POLICY: If *anything* doesn't work, put it on the blacklist.
  *	   If they are critical errors, mark it critical, and abort driver load.
@@ -165,5 +168,13 @@ int __init acpi_blacklisted(void)
 
 	blacklisted += blacklist_by_year();
 
+	dmi_check_system(acpi_osi_dmi_table);
+
 	return blacklisted;
 }
+#ifdef CONFIG_DMI
+static struct dmi_system_id acpi_osi_dmi_table[] __initdata = {
+	{}
+};
+
+#endif /* CONFIG_DMI */
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 15f095ea795e..e53fb516f9d4 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -77,7 +77,55 @@ static struct workqueue_struct *kacpi_notify_wq;
 #define	OSI_STRING_LENGTH_MAX 64	/* arbitrary */
 static char osi_additional_string[OSI_STRING_LENGTH_MAX];
 
-static int osi_linux;		/* disable _OSI(Linux) by default */
+/*
+ * "Ode to _OSI(Linux)"
+ *
+ * osi_linux -- Control response to BIOS _OSI(Linux) query.
+ *
+ * As Linux evolves, the features that it supports change.
+ * So an OSI string such as "Linux" is not specific enough
+ * to be useful across multiple versions of Linux.  It
+ * doesn't identify any particular feature, interface,
+ * or even any particular version of Linux...
+ *
+ * Unfortunately, Linux-2.6.22 and earlier responded "yes"
+ * to a BIOS _OSI(Linux) query.  When
+ * a reference mobile BIOS started using it, its use
+ * started to spread to many vendor platforms.
+ * As it is not supportable, we need to halt that spread.
+ *
+ * Today, most BIOS references to _OSI(Linux) are noise --
+ * they have no functional effect and are just dead code
+ * carried over from the reference BIOS.
+ *
+ * The next most common case is that _OSI(Linux) harms Linux,
+ * usually by causing the BIOS to follow paths that are
+ * not tested during Windows validation.
+ *
+ * Finally, there is a short list of platforms
+ * where OSI(Linux) benefits Linux.
+ *
+ * In Linux-2.6.23, OSI(Linux) is first disabled by default.
+ * DMI is used to disable the dmesg warning about OSI(Linux)
+ * on platforms where it is known to have no effect.
+ * But a dmesg warning remains for systems where
+ * we do not know if OSI(Linux) is good or bad for the system.
+ * DMI is also used to enable OSI(Linux) for the machines
+ * that are known to need it.
+ *
+ * BIOS writers should NOT query _OSI(Linux) on future systems.
+ * It will be ignored by default, and to get Linux to
+ * not ignore it will require a kernel source update to
+ * add a DMI entry, or a boot-time "acpi_osi=Linux" invocation.
+ */
+#define OSI_LINUX_ENABLE 0
+
+struct osi_linux {
+	unsigned int	enable:1;
+	unsigned int	dmi:1;
+	unsigned int	cmdline:1;
+	unsigned int	known:1;
+} osi_linux = { OSI_LINUX_ENABLE, 0, 0, 0};
 
 static void __init acpi_request_region (struct acpi_generic_address *addr,
 	unsigned int length, char *desc)
@@ -959,13 +1007,37 @@ static int __init acpi_os_name_setup(char *str)
 
 __setup("acpi_os_name=", acpi_os_name_setup);
 
-static void enable_osi_linux(int enable) {
+static void __init set_osi_linux(unsigned int enable)
+{
+	if (osi_linux.enable != enable) {
+		osi_linux.enable = enable;
+		printk(KERN_NOTICE PREFIX "%sed _OSI(Linux)\n",
+			enable ? "Add": "Delet");
+	}
+	return;
+}
+
+static void __init acpi_cmdline_osi_linux(unsigned int enable)
+{
+	osi_linux.cmdline = 1;	/* cmdline set the default */
+	set_osi_linux(enable);
+
+	return;
+}
+
+void __init acpi_dmi_osi_linux(int enable, const struct dmi_system_id *d)
+{
+	osi_linux.dmi = 1;	/* DMI knows that this box asks OSI(Linux) */
+
+	printk(KERN_NOTICE PREFIX "DMI detected: %s\n", d->ident);
+
+	if (enable == -1)
+		return;
+
+	osi_linux.known = 1;	/* DMI knows which OSI(Linux) default needed */
 
-	if (osi_linux != enable)
-		printk(KERN_INFO PREFIX "%sabled _OSI(Linux)\n",
-			enable ? "En": "Dis");
+	set_osi_linux(enable);
 
-	osi_linux = enable;
 	return;
 }
 
@@ -982,12 +1054,12 @@ static int __init acpi_osi_setup(char *str)
 		printk(KERN_INFO PREFIX "_OSI method disabled\n");
 		acpi_gbl_create_osi_method = FALSE;
 	} else if (!strcmp("!Linux", str)) {
-		enable_osi_linux(0);
+		acpi_cmdline_osi_linux(0);	/* !enable */
 	} else if (*str == '!') {
 		if (acpi_osi_invalidate(++str) == AE_OK)
 			printk(KERN_INFO PREFIX "Deleted _OSI(%s)\n", str);
 	} else if (!strcmp("Linux", str)) {
-		enable_osi_linux(1);
+		acpi_cmdline_osi_linux(1);	/* enable */
 	} else if (*osi_additional_string == '\0') {
 		strncpy(osi_additional_string, str, OSI_STRING_LENGTH_MAX);
 		printk(KERN_INFO PREFIX "Added _OSI(%s)\n", str);
@@ -1183,19 +1255,29 @@ acpi_os_validate_interface (char *interface)
 	if (!strncmp(osi_additional_string, interface, OSI_STRING_LENGTH_MAX))
 		return AE_OK;
 	if (!strcmp("Linux", interface)) {
-		printk(KERN_WARNING PREFIX
-			"System BIOS is requesting _OSI(Linux)\n");
-		if (acpi_dmi_dump())
-			printk(KERN_NOTICE PREFIX
-				"[please extract dmidecode output]\n");
-		printk(KERN_NOTICE PREFIX
-			"Please send DMI info above to "
-			"linux-acpi@vger.kernel.org\n");
+
 		printk(KERN_NOTICE PREFIX
-			"If \"acpi_osi=%sLinux\" works better, "
-			"please notify linux-acpi@vger.kernel.org\n",
-			osi_linux ? "!" : "");
-		if(osi_linux)
+			"BIOS _OSI(Linux) query %s%s\n",
+			osi_linux.enable ? "honored" : "ignored",
+			osi_linux.cmdline ? " via cmdline" :
+			osi_linux.dmi ? " via DMI" : "");
+
+		if (!osi_linux.dmi) {
+			if (acpi_dmi_dump())
+				printk(KERN_NOTICE PREFIX
+					"[please extract dmidecode output]\n");
+			printk(KERN_NOTICE PREFIX
+				"Please send DMI info above to "
+				"linux-acpi@vger.kernel.org\n");
+		}
+		if (!osi_linux.known && !osi_linux.cmdline) {
+			printk(KERN_NOTICE PREFIX
+				"If \"acpi_osi=%sLinux\" works better, "
+				"please notify linux-acpi@vger.kernel.org\n",
+				osi_linux.enable ? "!" : "");
+		}
+
+		if (osi_linux.enable)
 			return AE_OK;
 	}
 	return AE_SUPPORT;
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index e3c16c981e46..63f2e6ed698f 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -40,6 +40,7 @@
 #include <acpi/acpi_drivers.h>
 #include <acpi/acpi_numa.h>
 #include <asm/acpi.h>
+#include <linux/dmi.h>
 
 
 #ifdef CONFIG_ACPI
@@ -192,7 +193,9 @@ extern int ec_transaction(u8 command,
 #endif /*CONFIG_ACPI_EC*/
 
 extern int acpi_blacklisted(void);
-extern void acpi_bios_year(char *s);
+#ifdef CONFIG_DMI
+extern void acpi_dmi_osi_linux(int enable, const struct dmi_system_id *d);
+#endif
 
 #ifdef CONFIG_ACPI_NUMA
 int acpi_get_pxm(acpi_handle handle);
@@ -226,5 +229,5 @@ static inline int acpi_boot_table_init(void)
 	return 0;
 }
 
-#endif	/* CONFIG_ACPI */
+#endif	/* !CONFIG_ACPI */
 #endif	/*_LINUX_ACPI_H*/
-- 
cgit v1.2.3


From ec68373c04495edbe39fb94fad963fb781e062e5 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Wed, 23 Jan 2008 22:41:20 -0500
Subject: Revert "ACPI: Fan: Drop force_power_state acpi_device option"

This reverts commit 93ad7c07ad487b036add8760dabcc35666a550ef.

http://bugzilla.kernel.org/show_bug.cgi?id=9798

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c      |  2 +-
 drivers/acpi/fan.c      | 40 ++++++++++++++++++++++++++++++++++++++++
 include/acpi/acpi_bus.h |  3 ++-
 3 files changed, 43 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index d7a115c362d1..f4487c38d9f2 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -200,7 +200,7 @@ int acpi_bus_set_power(acpi_handle handle, int state)
 	 * Get device's current power state
 	 */
 	acpi_bus_get_power(device->handle, &device->power.state);
-	if (state == device->power.state) {
+	if ((state == device->power.state) && !device->flags.force_power_state) {
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device is already at D%d\n",
 				  state));
 		return 0;
diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
index a5a5532db268..a6e149d692cb 100644
--- a/drivers/acpi/fan.c
+++ b/drivers/acpi/fan.c
@@ -47,6 +47,8 @@ MODULE_LICENSE("GPL");
 
 static int acpi_fan_add(struct acpi_device *device);
 static int acpi_fan_remove(struct acpi_device *device, int type);
+static int acpi_fan_suspend(struct acpi_device *device, pm_message_t state);
+static int acpi_fan_resume(struct acpi_device *device);
 
 static const struct acpi_device_id fan_device_ids[] = {
 	{"PNP0C0B", 0},
@@ -61,6 +63,8 @@ static struct acpi_driver acpi_fan_driver = {
 	.ops = {
 		.add = acpi_fan_add,
 		.remove = acpi_fan_remove,
+		.suspend = acpi_fan_suspend,
+		.resume = acpi_fan_resume,
 		},
 };
 
@@ -191,6 +195,10 @@ static int acpi_fan_add(struct acpi_device *device)
 		goto end;
 	}
 
+	device->flags.force_power_state = 1;
+	acpi_bus_set_power(device->handle, state);
+	device->flags.force_power_state = 0;
+
 	result = acpi_fan_add_fs(device);
 	if (result)
 		goto end;
@@ -216,6 +224,38 @@ static int acpi_fan_remove(struct acpi_device *device, int type)
 	return 0;
 }
 
+static int acpi_fan_suspend(struct acpi_device *device, pm_message_t state)
+{
+	if (!device)
+		return -EINVAL;
+
+	acpi_bus_set_power(device->handle, ACPI_STATE_D0);
+
+	return AE_OK;
+}
+
+static int acpi_fan_resume(struct acpi_device *device)
+{
+	int result = 0;
+	int power_state = 0;
+
+	if (!device)
+		return -EINVAL;
+
+	result = acpi_bus_get_power(device->handle, &power_state);
+	if (result) {
+		ACPI_DEBUG_PRINT((ACPI_DB_ERROR,
+				  "Error reading fan power state\n"));
+		return result;
+	}
+
+	device->flags.force_power_state = 1;
+	acpi_bus_set_power(device->handle, power_state);
+	device->flags.force_power_state = 0;
+
+	return result;
+}
+
 static int __init acpi_fan_init(void)
 {
 	int result = 0;
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 19c3ead2a90b..7b74b60a68a4 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -168,7 +168,8 @@ struct acpi_device_flags {
 	u32 power_manageable:1;
 	u32 performance_manageable:1;
 	u32 wake_capable:1;	/* Wakeup(_PRW) supported? */
-	u32 reserved:20;
+	u32 force_power_state:1;
+	u32 reserved:19;
 };
 
 /* File System */
-- 
cgit v1.2.3


From c9180a57a9ab2d5525faf8815a332364ee9e89b7 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Fri, 30 Nov 2007 13:00:35 -0500
Subject: Security: add get, set, and cloning of superblock security
 information

Adds security_get_sb_mnt_opts, security_set_sb_mnt_opts, and
security_clont_sb_mnt_opts to the LSM and to SELinux.  This will allow
filesystems to directly own and control all of their mount options if they
so choose.  This interface deals only with option identifiers and strings so
it should generic enough for any LSM which may come in the future.

Filesystems which pass text mount data around in the kernel (almost all of
them) need not currently make use of this interface when dealing with
SELinux since it will still parse those strings as it always has.  I assume
future LSM's would do the same.  NFS is the primary FS which does not use
text mount data and thus must make use of this interface.

An LSM would need to implement these functions only if they had mount time
options, such as selinux has context= or fscontext=.  If the LSM has no
mount time options they could simply not implement and let the dummy ops
take care of things.

An LSM other than SELinux would need to define new option numbers in
security.h and any FS which decides to own there own security options would
need to be patched to use this new interface for every possible LSM.  This
is because it was stated to me very clearly that LSM's should not attempt to
understand FS mount data and the burdon to understand security should be in
the FS which owns the options.

Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Stephen D. Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h          |  36 ++
 security/dummy.c                  |  26 ++
 security/security.c               |  20 +
 security/selinux/hooks.c          | 748 +++++++++++++++++++++++++-------------
 security/selinux/include/objsec.h |   1 +
 5 files changed, 577 insertions(+), 254 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index ac050830a873..cbd970a735f2 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -34,6 +34,12 @@
 #include <linux/xfrm.h>
 #include <net/flow.h>
 
+/* only a char in selinux superblock security struct flags */
+#define FSCONTEXT_MNT		0x01
+#define CONTEXT_MNT		0x02
+#define ROOTCONTEXT_MNT		0x04
+#define DEFCONTEXT_MNT		0x08
+
 /*
  * Bounding set
  */
@@ -261,6 +267,22 @@ struct request_sock;
  *	Update module state after a successful pivot.
  *	@old_nd contains the nameidata structure for the old root.
  *      @new_nd contains the nameidata structure for the new root.
+ * @sb_get_mnt_opts:
+ *	Get the security relevant mount options used for a superblock
+ *	@sb the superblock to get security mount options from
+ *	@mount_options array for pointers to mount options
+ *	@mount_flags array of ints specifying what each mount options is
+ *	@num_opts number of options in the arrays
+ * @sb_set_mnt_opts:
+ *	Set the security relevant mount options used for a superblock
+ *	@sb the superblock to set security mount options for
+ *	@mount_options array for pointers to mount options
+ *	@mount_flags array of ints specifying what each mount options is
+ *	@num_opts number of options in the arrays
+ * @sb_clone_mnt_opts:
+ *	Copy all security options from a given superblock to another
+ *	@oldsb old superblock which contain information to clone
+ *	@newsb new superblock which needs filled in
  *
  * Security hooks for inode operations.
  *
@@ -1242,6 +1264,13 @@ struct security_operations {
 			     struct nameidata * new_nd);
 	void (*sb_post_pivotroot) (struct nameidata * old_nd,
 				   struct nameidata * new_nd);
+	int (*sb_get_mnt_opts) (const struct super_block *sb,
+				char ***mount_options, int **flags,
+				int *num_opts);
+	int (*sb_set_mnt_opts) (struct super_block *sb, char **mount_options,
+				int *flags, int num_opts);
+	void (*sb_clone_mnt_opts) (const struct super_block *oldsb,
+				   struct super_block *newsb);
 
 	int (*inode_alloc_security) (struct inode *inode);	
 	void (*inode_free_security) (struct inode *inode);
@@ -1499,6 +1528,13 @@ void security_sb_post_mountroot(void);
 void security_sb_post_addmount(struct vfsmount *mnt, struct nameidata *mountpoint_nd);
 int security_sb_pivotroot(struct nameidata *old_nd, struct nameidata *new_nd);
 void security_sb_post_pivotroot(struct nameidata *old_nd, struct nameidata *new_nd);
+int security_sb_get_mnt_opts(const struct super_block *sb, char ***mount_options,
+			     int **flags, int *num_opts);
+int security_sb_set_mnt_opts(struct super_block *sb, char **mount_options,
+			     int *flags, int num_opts);
+void security_sb_clone_mnt_opts(const struct super_block *oldsb,
+				struct super_block *newsb);
+
 int security_inode_alloc(struct inode *inode);
 void security_inode_free(struct inode *inode);
 int security_inode_init_security(struct inode *inode, struct inode *dir,
diff --git a/security/dummy.c b/security/dummy.c
index 3ccfbbe973b6..a3b29d0d00e5 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -245,6 +245,29 @@ static void dummy_sb_post_pivotroot (struct nameidata *old_nd, struct nameidata
 	return;
 }
 
+static int dummy_sb_get_mnt_opts(const struct super_block *sb, char ***mount_options,
+				 int **flags, int *num_opts)
+{
+	*mount_options = NULL;
+	*flags = NULL;
+	*num_opts = 0;
+	return 0;
+}
+
+static int dummy_sb_set_mnt_opts(struct super_block *sb, char **mount_options,
+				 int *flags, int num_opts)
+{
+	if (unlikely(num_opts))
+		return -EOPNOTSUPP;
+	return 0;
+}
+
+static void dummy_sb_clone_mnt_opts(const struct super_block *oldsb,
+				    struct super_block *newsb)
+{
+	return;
+}
+
 static int dummy_inode_alloc_security (struct inode *inode)
 {
 	return 0;
@@ -998,6 +1021,9 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sb_post_addmount);
 	set_to_dummy_if_null(ops, sb_pivotroot);
 	set_to_dummy_if_null(ops, sb_post_pivotroot);
+	set_to_dummy_if_null(ops, sb_get_mnt_opts);
+	set_to_dummy_if_null(ops, sb_set_mnt_opts);
+	set_to_dummy_if_null(ops, sb_clone_mnt_opts);
 	set_to_dummy_if_null(ops, inode_alloc_security);
 	set_to_dummy_if_null(ops, inode_free_security);
 	set_to_dummy_if_null(ops, inode_init_security);
diff --git a/security/security.c b/security/security.c
index 0e1f1f124368..b13b54f0af85 100644
--- a/security/security.c
+++ b/security/security.c
@@ -308,6 +308,26 @@ void security_sb_post_pivotroot(struct nameidata *old_nd, struct nameidata *new_
 	security_ops->sb_post_pivotroot(old_nd, new_nd);
 }
 
+int security_sb_get_mnt_opts(const struct super_block *sb,
+			      char ***mount_options,
+			      int **flags, int *num_opts)
+{
+	return security_ops->sb_get_mnt_opts(sb, mount_options, flags, num_opts);
+}
+
+int security_sb_set_mnt_opts(struct super_block *sb,
+			      char **mount_options,
+			      int *flags, int num_opts)
+{
+	return security_ops->sb_set_mnt_opts(sb, mount_options, flags, num_opts);
+}
+
+void security_sb_clone_mnt_opts(const struct super_block *oldsb,
+				struct super_block *newsb)
+{
+	security_ops->sb_clone_mnt_opts(oldsb, newsb);
+}
+
 int security_inode_alloc(struct inode *inode)
 {
 	inode->i_security = NULL;
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 9f3124b08867..233c8b97462f 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -82,6 +82,8 @@
 #define XATTR_SELINUX_SUFFIX "selinux"
 #define XATTR_NAME_SELINUX XATTR_SECURITY_PREFIX XATTR_SELINUX_SUFFIX
 
+#define NUM_SEL_MNT_OPTS 4
+
 extern unsigned int policydb_loaded_version;
 extern int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm);
 extern int selinux_compat_net;
@@ -321,8 +323,8 @@ enum {
 	Opt_error = -1,
 	Opt_context = 1,
 	Opt_fscontext = 2,
-	Opt_defcontext = 4,
-	Opt_rootcontext = 8,
+	Opt_defcontext = 3,
+	Opt_rootcontext = 4,
 };
 
 static match_table_t tokens = {
@@ -366,150 +368,317 @@ static int may_context_mount_inode_relabel(u32 sid,
 	return rc;
 }
 
-static int try_context_mount(struct super_block *sb, void *data)
+static int sb_finish_set_opts(struct super_block *sb)
 {
-	char *context = NULL, *defcontext = NULL;
-	char *fscontext = NULL, *rootcontext = NULL;
-	const char *name;
-	u32 sid;
-	int alloc = 0, rc = 0, seen = 0;
-	struct task_security_struct *tsec = current->security;
 	struct superblock_security_struct *sbsec = sb->s_security;
+	struct dentry *root = sb->s_root;
+	struct inode *root_inode = root->d_inode;
+	int rc = 0;
 
-	if (!data)
-		goto out;
+	if (sbsec->behavior == SECURITY_FS_USE_XATTR) {
+		/* Make sure that the xattr handler exists and that no
+		   error other than -ENODATA is returned by getxattr on
+		   the root directory.  -ENODATA is ok, as this may be
+		   the first boot of the SELinux kernel before we have
+		   assigned xattr values to the filesystem. */
+		if (!root_inode->i_op->getxattr) {
+			printk(KERN_WARNING "SELinux: (dev %s, type %s) has no "
+			       "xattr support\n", sb->s_id, sb->s_type->name);
+			rc = -EOPNOTSUPP;
+			goto out;
+		}
+		rc = root_inode->i_op->getxattr(root, XATTR_NAME_SELINUX, NULL, 0);
+		if (rc < 0 && rc != -ENODATA) {
+			if (rc == -EOPNOTSUPP)
+				printk(KERN_WARNING "SELinux: (dev %s, type "
+				       "%s) has no security xattr handler\n",
+				       sb->s_id, sb->s_type->name);
+			else
+				printk(KERN_WARNING "SELinux: (dev %s, type "
+				       "%s) getxattr errno %d\n", sb->s_id,
+				       sb->s_type->name, -rc);
+			goto out;
+		}
+	}
 
-	name = sb->s_type->name;
+	sbsec->initialized = 1;
 
-	if (sb->s_type->fs_flags & FS_BINARY_MOUNTDATA) {
+	if (sbsec->behavior > ARRAY_SIZE(labeling_behaviors))
+		printk(KERN_ERR "SELinux: initialized (dev %s, type %s), unknown behavior\n",
+		       sb->s_id, sb->s_type->name);
+	else
+		printk(KERN_DEBUG "SELinux: initialized (dev %s, type %s), %s\n",
+		       sb->s_id, sb->s_type->name,
+		       labeling_behaviors[sbsec->behavior-1]);
 
-		/* NFS we understand. */
-		if (!strcmp(name, "nfs")) {
-			struct nfs_mount_data *d = data;
+	/* Initialize the root inode. */
+	rc = inode_doinit_with_dentry(root_inode, root);
 
-			if (d->version <  NFS_MOUNT_VERSION)
-				goto out;
+	/* Initialize any other inodes associated with the superblock, e.g.
+	   inodes created prior to initial policy load or inodes created
+	   during get_sb by a pseudo filesystem that directly
+	   populates itself. */
+	spin_lock(&sbsec->isec_lock);
+next_inode:
+	if (!list_empty(&sbsec->isec_head)) {
+		struct inode_security_struct *isec =
+				list_entry(sbsec->isec_head.next,
+					   struct inode_security_struct, list);
+		struct inode *inode = isec->inode;
+		spin_unlock(&sbsec->isec_lock);
+		inode = igrab(inode);
+		if (inode) {
+			if (!IS_PRIVATE(inode))
+				inode_doinit(inode);
+			iput(inode);
+		}
+		spin_lock(&sbsec->isec_lock);
+		list_del_init(&isec->list);
+		goto next_inode;
+	}
+	spin_unlock(&sbsec->isec_lock);
+out:
+	return rc;
+}
 
-			if (d->context[0]) {
-				context = d->context;
-				seen |= Opt_context;
-			}
-		} else
-			goto out;
+/*
+ * This function should allow an FS to ask what it's mount security
+ * options were so it can use those later for submounts, displaying
+ * mount options, or whatever.
+ */
+static int selinux_get_mnt_opts(const struct super_block *sb,
+				char ***mount_options, int **mnt_opts_flags,
+				int *num_opts)
+{
+	int rc = 0, i;
+	struct superblock_security_struct *sbsec = sb->s_security;
+	char *context = NULL;
+	u32 len;
+	char tmp;
 
-	} else {
-		/* Standard string-based options. */
-		char *p, *options = data;
+	*num_opts = 0;
+	*mount_options = NULL;
+	*mnt_opts_flags = NULL;
 
-		while ((p = strsep(&options, "|")) != NULL) {
-			int token;
-			substring_t args[MAX_OPT_ARGS];
+	if (!sbsec->initialized)
+		return -EINVAL;
 
-			if (!*p)
-				continue;
+	if (!ss_initialized)
+		return -EINVAL;
 
-			token = match_token(p, tokens, args);
+	/*
+	 * if we ever use sbsec flags for anything other than tracking mount
+	 * settings this is going to need a mask
+	 */
+	tmp = sbsec->flags;
+	/* count the number of mount options for this sb */
+	for (i = 0; i < 8; i++) {
+		if (tmp & 0x01)
+			(*num_opts)++;
+		tmp >>= 1;
+	}
 
-			switch (token) {
-			case Opt_context:
-				if (seen & (Opt_context|Opt_defcontext)) {
-					rc = -EINVAL;
-					printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
-					goto out_free;
-				}
-				context = match_strdup(&args[0]);
-				if (!context) {
-					rc = -ENOMEM;
-					goto out_free;
-				}
-				if (!alloc)
-					alloc = 1;
-				seen |= Opt_context;
-				break;
+	*mount_options = kcalloc(*num_opts, sizeof(char *), GFP_ATOMIC);
+	if (!*mount_options) {
+		rc = -ENOMEM;
+		goto out_free;
+	}
 
-			case Opt_fscontext:
-				if (seen & Opt_fscontext) {
-					rc = -EINVAL;
-					printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
-					goto out_free;
-				}
-				fscontext = match_strdup(&args[0]);
-				if (!fscontext) {
-					rc = -ENOMEM;
-					goto out_free;
-				}
-				if (!alloc)
-					alloc = 1;
-				seen |= Opt_fscontext;
-				break;
+	*mnt_opts_flags = kcalloc(*num_opts, sizeof(int), GFP_ATOMIC);
+	if (!*mnt_opts_flags) {
+		rc = -ENOMEM;
+		goto out_free;
+	}
 
-			case Opt_rootcontext:
-				if (seen & Opt_rootcontext) {
-					rc = -EINVAL;
-					printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
-					goto out_free;
-				}
-				rootcontext = match_strdup(&args[0]);
-				if (!rootcontext) {
-					rc = -ENOMEM;
-					goto out_free;
-				}
-				if (!alloc)
-					alloc = 1;
-				seen |= Opt_rootcontext;
-				break;
+	i = 0;
+	if (sbsec->flags & FSCONTEXT_MNT) {
+		rc = security_sid_to_context(sbsec->sid, &context, &len);
+		if (rc)
+			goto out_free;
+		(*mount_options)[i] = context;
+		(*mnt_opts_flags)[i++] = FSCONTEXT_MNT;
+	}
+	if (sbsec->flags & CONTEXT_MNT) {
+		rc = security_sid_to_context(sbsec->mntpoint_sid, &context, &len);
+		if (rc)
+			goto out_free;
+		(*mount_options)[i] = context;
+		(*mnt_opts_flags)[i++] = CONTEXT_MNT;
+	}
+	if (sbsec->flags & DEFCONTEXT_MNT) {
+		rc = security_sid_to_context(sbsec->def_sid, &context, &len);
+		if (rc)
+			goto out_free;
+		(*mount_options)[i] = context;
+		(*mnt_opts_flags)[i++] = DEFCONTEXT_MNT;
+	}
+	if (sbsec->flags & ROOTCONTEXT_MNT) {
+		struct inode *root = sbsec->sb->s_root->d_inode;
+		struct inode_security_struct *isec = root->i_security;
 
-			case Opt_defcontext:
-				if (sbsec->behavior != SECURITY_FS_USE_XATTR) {
-					rc = -EINVAL;
-					printk(KERN_WARNING "SELinux:  "
-					       "defcontext option is invalid "
-					       "for this filesystem type\n");
-					goto out_free;
-				}
-				if (seen & (Opt_context|Opt_defcontext)) {
-					rc = -EINVAL;
-					printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
-					goto out_free;
-				}
-				defcontext = match_strdup(&args[0]);
-				if (!defcontext) {
-					rc = -ENOMEM;
-					goto out_free;
-				}
-				if (!alloc)
-					alloc = 1;
-				seen |= Opt_defcontext;
-				break;
+		rc = security_sid_to_context(isec->sid, &context, &len);
+		if (rc)
+			goto out_free;
+		(*mount_options)[i] = context;
+		(*mnt_opts_flags)[i++] = ROOTCONTEXT_MNT;
+	}
 
-			default:
-				rc = -EINVAL;
-				printk(KERN_WARNING "SELinux:  unknown mount "
-				       "option\n");
-				goto out_free;
+	BUG_ON(i != *num_opts);
 
-			}
-		}
-	}
+	return 0;
+
+out_free:
+	/* don't leak context string if security_sid_to_context had an error */
+	if (*mount_options && i)
+		for (; i > 0; i--)
+			kfree((*mount_options)[i-1]);
+	kfree(*mount_options);
+	*mount_options = NULL;
+	kfree(*mnt_opts_flags);
+	*mnt_opts_flags = NULL;
+	*num_opts = 0;
+	return rc;
+}
+
+static int bad_option(struct superblock_security_struct *sbsec, char flag,
+		      u32 old_sid, u32 new_sid)
+{
+	/* check if the old mount command had the same options */
+	if (sbsec->initialized)
+		if (!(sbsec->flags & flag) ||
+		    (old_sid != new_sid))
+			return 1;
+
+	/* check if we were passed the same options twice,
+	 * aka someone passed context=a,context=b
+	 */
+	if (!sbsec->initialized)
+		if (sbsec->flags & flag)
+			return 1;
+	return 0;
+}
+/*
+ * Allow filesystems with binary mount data to explicitly set mount point
+ * labeling information.
+ */
+int selinux_set_mnt_opts(struct super_block *sb, char **mount_options,
+				 int *flags, int num_opts)
+{
+	int rc = 0, i;
+	struct task_security_struct *tsec = current->security;
+	struct superblock_security_struct *sbsec = sb->s_security;
+	const char *name = sb->s_type->name;
+	struct inode *inode = sbsec->sb->s_root->d_inode;
+	struct inode_security_struct *root_isec = inode->i_security;
+	u32 fscontext_sid = 0, context_sid = 0, rootcontext_sid = 0;
+	u32 defcontext_sid = 0;
 
-	if (!seen)
+	mutex_lock(&sbsec->lock);
+
+	if (!ss_initialized) {
+		if (!num_opts) {
+			/* Defer initialization until selinux_complete_init,
+			   after the initial policy is loaded and the security
+			   server is ready to handle calls. */
+			spin_lock(&sb_security_lock);
+			if (list_empty(&sbsec->list))
+				list_add(&sbsec->list, &superblock_security_head);
+			spin_unlock(&sb_security_lock);
+			goto out;
+		}
+		rc = -EINVAL;
+		printk(KERN_WARNING "Unable to set superblock options before "
+		       "the security server is initialized\n");
 		goto out;
+	}
 
-	/* sets the context of the superblock for the fs being mounted. */
-	if (fscontext) {
-		rc = security_context_to_sid(fscontext, strlen(fscontext), &sid);
+	/*
+	 * parse the mount options, check if they are valid sids.
+	 * also check if someone is trying to mount the same sb more
+	 * than once with different security options.
+	 */
+	for (i = 0; i < num_opts; i++) {
+		u32 sid;
+		rc = security_context_to_sid(mount_options[i],
+					     strlen(mount_options[i]), &sid);
 		if (rc) {
 			printk(KERN_WARNING "SELinux: security_context_to_sid"
 			       "(%s) failed for (dev %s, type %s) errno=%d\n",
-			       fscontext, sb->s_id, name, rc);
-			goto out_free;
+			       mount_options[i], sb->s_id, name, rc);
+			goto out;
+		}
+		switch (flags[i]) {
+		case FSCONTEXT_MNT:
+			fscontext_sid = sid;
+
+			if (bad_option(sbsec, FSCONTEXT_MNT, sbsec->sid,
+					fscontext_sid))
+				goto out_double_mount;
+
+			sbsec->flags |= FSCONTEXT_MNT;
+			break;
+		case CONTEXT_MNT:
+			context_sid = sid;
+
+			if (bad_option(sbsec, CONTEXT_MNT, sbsec->mntpoint_sid,
+					context_sid))
+				goto out_double_mount;
+
+			sbsec->flags |= CONTEXT_MNT;
+			break;
+		case ROOTCONTEXT_MNT:
+			rootcontext_sid = sid;
+
+			if (bad_option(sbsec, ROOTCONTEXT_MNT, root_isec->sid,
+					rootcontext_sid))
+				goto out_double_mount;
+
+			sbsec->flags |= ROOTCONTEXT_MNT;
+
+			break;
+		case DEFCONTEXT_MNT:
+			defcontext_sid = sid;
+
+			if (bad_option(sbsec, DEFCONTEXT_MNT, sbsec->def_sid,
+					defcontext_sid))
+				goto out_double_mount;
+
+			sbsec->flags |= DEFCONTEXT_MNT;
+
+			break;
+		default:
+			rc = -EINVAL;
+			goto out;
 		}
+	}
+
+	if (sbsec->initialized) {
+		/* previously mounted with options, but not on this attempt? */
+		if (sbsec->flags && !num_opts)
+			goto out_double_mount;
+		rc = 0;
+		goto out;
+	}
 
-		rc = may_context_mount_sb_relabel(sid, sbsec, tsec);
+	if (strcmp(sb->s_type->name, "proc") == 0)
+		sbsec->proc = 1;
+
+	/* Determine the labeling behavior to use for this filesystem type. */
+	rc = security_fs_use(sb->s_type->name, &sbsec->behavior, &sbsec->sid);
+	if (rc) {
+		printk(KERN_WARNING "%s: security_fs_use(%s) returned %d\n",
+		       __FUNCTION__, sb->s_type->name, rc);
+		goto out;
+	}
+
+	/* sets the context of the superblock for the fs being mounted. */
+	if (fscontext_sid) {
+
+		rc = may_context_mount_sb_relabel(fscontext_sid, sbsec, tsec);
 		if (rc)
-			goto out_free;
+			goto out;
 
-		sbsec->sid = sid;
+		sbsec->sid = fscontext_sid;
 	}
 
 	/*
@@ -517,182 +686,250 @@ static int try_context_mount(struct super_block *sb, void *data)
 	 * sets the label used on all file below the mountpoint, and will set
 	 * the superblock context if not already set.
 	 */
-	if (context) {
-		rc = security_context_to_sid(context, strlen(context), &sid);
-		if (rc) {
-			printk(KERN_WARNING "SELinux: security_context_to_sid"
-			       "(%s) failed for (dev %s, type %s) errno=%d\n",
-			       context, sb->s_id, name, rc);
-			goto out_free;
-		}
-
-		if (!fscontext) {
-			rc = may_context_mount_sb_relabel(sid, sbsec, tsec);
+	if (context_sid) {
+		if (!fscontext_sid) {
+			rc = may_context_mount_sb_relabel(context_sid, sbsec, tsec);
 			if (rc)
-				goto out_free;
-			sbsec->sid = sid;
+				goto out;
+			sbsec->sid = context_sid;
 		} else {
-			rc = may_context_mount_inode_relabel(sid, sbsec, tsec);
+			rc = may_context_mount_inode_relabel(context_sid, sbsec, tsec);
 			if (rc)
-				goto out_free;
+				goto out;
 		}
-		sbsec->mntpoint_sid = sid;
+		if (!rootcontext_sid)
+			rootcontext_sid = context_sid;
 
+		sbsec->mntpoint_sid = context_sid;
 		sbsec->behavior = SECURITY_FS_USE_MNTPOINT;
 	}
 
-	if (rootcontext) {
-		struct inode *inode = sb->s_root->d_inode;
-		struct inode_security_struct *isec = inode->i_security;
-		rc = security_context_to_sid(rootcontext, strlen(rootcontext), &sid);
-		if (rc) {
-			printk(KERN_WARNING "SELinux: security_context_to_sid"
-			       "(%s) failed for (dev %s, type %s) errno=%d\n",
-			       rootcontext, sb->s_id, name, rc);
-			goto out_free;
-		}
-
-		rc = may_context_mount_inode_relabel(sid, sbsec, tsec);
+	if (rootcontext_sid) {
+		rc = may_context_mount_inode_relabel(rootcontext_sid, sbsec, tsec);
 		if (rc)
-			goto out_free;
+			goto out;
 
-		isec->sid = sid;
-		isec->initialized = 1;
+		root_isec->sid = rootcontext_sid;
+		root_isec->initialized = 1;
 	}
 
-	if (defcontext) {
-		rc = security_context_to_sid(defcontext, strlen(defcontext), &sid);
-		if (rc) {
-			printk(KERN_WARNING "SELinux: security_context_to_sid"
-			       "(%s) failed for (dev %s, type %s) errno=%d\n",
-			       defcontext, sb->s_id, name, rc);
-			goto out_free;
+	if (defcontext_sid) {
+		if (sbsec->behavior != SECURITY_FS_USE_XATTR) {
+			rc = -EINVAL;
+			printk(KERN_WARNING "SELinux: defcontext option is "
+			       "invalid for this filesystem type\n");
+			goto out;
 		}
 
-		if (sid == sbsec->def_sid)
-			goto out_free;
-
-		rc = may_context_mount_inode_relabel(sid, sbsec, tsec);
-		if (rc)
-			goto out_free;
+		if (defcontext_sid != sbsec->def_sid) {
+			rc = may_context_mount_inode_relabel(defcontext_sid,
+							     sbsec, tsec);
+			if (rc)
+				goto out;
+		}
 
-		sbsec->def_sid = sid;
+		sbsec->def_sid = defcontext_sid;
 	}
 
-out_free:
-	if (alloc) {
-		kfree(context);
-		kfree(defcontext);
-		kfree(fscontext);
-		kfree(rootcontext);
-	}
+	rc = sb_finish_set_opts(sb);
 out:
+	mutex_unlock(&sbsec->lock);
 	return rc;
+out_double_mount:
+	rc = -EINVAL;
+	printk(KERN_WARNING "SELinux: mount invalid.  Same superblock, different "
+	       "security settings for (dev %s, type %s)\n", sb->s_id, name);
+	goto out;
 }
 
-static int superblock_doinit(struct super_block *sb, void *data)
+static void selinux_sb_clone_mnt_opts(const struct super_block *oldsb,
+					struct super_block *newsb)
 {
-	struct superblock_security_struct *sbsec = sb->s_security;
-	struct dentry *root = sb->s_root;
-	struct inode *inode = root->d_inode;
-	int rc = 0;
+	const struct superblock_security_struct *oldsbsec = oldsb->s_security;
+	struct superblock_security_struct *newsbsec = newsb->s_security;
 
-	mutex_lock(&sbsec->lock);
-	if (sbsec->initialized)
-		goto out;
+	int set_fscontext =	(oldsbsec->flags & FSCONTEXT_MNT);
+	int set_context =	(oldsbsec->flags & CONTEXT_MNT);
+	int set_rootcontext =	(oldsbsec->flags & ROOTCONTEXT_MNT);
 
-	if (!ss_initialized) {
-		/* Defer initialization until selinux_complete_init,
-		   after the initial policy is loaded and the security
-		   server is ready to handle calls. */
-		spin_lock(&sb_security_lock);
-		if (list_empty(&sbsec->list))
-			list_add(&sbsec->list, &superblock_security_head);
-		spin_unlock(&sb_security_lock);
-		goto out;
+	/* we can't error, we can't save the info, this shouldn't get called
+	 * this early in the boot process. */
+	BUG_ON(!ss_initialized);
+
+	/* this might go away sometime down the line if there is a new user
+	 * of clone, but for now, nfs better not get here... */
+	BUG_ON(newsbsec->initialized);
+
+	/* how can we clone if the old one wasn't set up?? */
+	BUG_ON(!oldsbsec->initialized);
+
+	mutex_lock(&newsbsec->lock);
+
+	newsbsec->flags = oldsbsec->flags;
+
+	newsbsec->sid = oldsbsec->sid;
+	newsbsec->def_sid = oldsbsec->def_sid;
+	newsbsec->behavior = oldsbsec->behavior;
+
+	if (set_context) {
+		u32 sid = oldsbsec->mntpoint_sid;
+
+		if (!set_fscontext)
+			newsbsec->sid = sid;
+		if (!set_rootcontext) {
+			struct inode *newinode = newsb->s_root->d_inode;
+			struct inode_security_struct *newisec = newinode->i_security;
+			newisec->sid = sid;
+		}
+		newsbsec->mntpoint_sid = sid;
 	}
+	if (set_rootcontext) {
+		const struct inode *oldinode = oldsb->s_root->d_inode;
+		const struct inode_security_struct *oldisec = oldinode->i_security;
+		struct inode *newinode = newsb->s_root->d_inode;
+		struct inode_security_struct *newisec = newinode->i_security;
 
-	/* Determine the labeling behavior to use for this filesystem type. */
-	rc = security_fs_use(sb->s_type->name, &sbsec->behavior, &sbsec->sid);
-	if (rc) {
-		printk(KERN_WARNING "%s:  security_fs_use(%s) returned %d\n",
-		       __FUNCTION__, sb->s_type->name, rc);
-		goto out;
+		newisec->sid = oldisec->sid;
 	}
 
-	rc = try_context_mount(sb, data);
-	if (rc)
+	sb_finish_set_opts(newsb);
+	mutex_unlock(&newsbsec->lock);
+}
+
+/*
+ * string mount options parsing and call set the sbsec
+ */
+static int superblock_doinit(struct super_block *sb, void *data)
+{
+	char *context = NULL, *defcontext = NULL;
+	char *fscontext = NULL, *rootcontext = NULL;
+	int rc = 0;
+	char *p, *options = data;
+	/* selinux only know about a fixed number of mount options */
+	char *mnt_opts[NUM_SEL_MNT_OPTS];
+	int mnt_opts_flags[NUM_SEL_MNT_OPTS], num_mnt_opts = 0;
+
+	if (!data)
 		goto out;
 
-	if (sbsec->behavior == SECURITY_FS_USE_XATTR) {
-		/* Make sure that the xattr handler exists and that no
-		   error other than -ENODATA is returned by getxattr on
-		   the root directory.  -ENODATA is ok, as this may be
-		   the first boot of the SELinux kernel before we have
-		   assigned xattr values to the filesystem. */
-		if (!inode->i_op->getxattr) {
-			printk(KERN_WARNING "SELinux: (dev %s, type %s) has no "
-			       "xattr support\n", sb->s_id, sb->s_type->name);
-			rc = -EOPNOTSUPP;
-			goto out;
-		}
-		rc = inode->i_op->getxattr(root, XATTR_NAME_SELINUX, NULL, 0);
-		if (rc < 0 && rc != -ENODATA) {
-			if (rc == -EOPNOTSUPP)
-				printk(KERN_WARNING "SELinux: (dev %s, type "
-				       "%s) has no security xattr handler\n",
-				       sb->s_id, sb->s_type->name);
-			else
-				printk(KERN_WARNING "SELinux: (dev %s, type "
-				       "%s) getxattr errno %d\n", sb->s_id,
-				       sb->s_type->name, -rc);
+	/* with the nfs patch this will become a goto out; */
+	if (sb->s_type->fs_flags & FS_BINARY_MOUNTDATA) {
+		const char *name = sb->s_type->name;
+		/* NFS we understand. */
+		if (!strcmp(name, "nfs")) {
+			struct nfs_mount_data *d = data;
+
+			if (d->version !=  NFS_MOUNT_VERSION)
+				goto out;
+
+			if (d->context[0]) {
+				context = kstrdup(d->context, GFP_KERNEL);
+				if (!context) {
+					rc = -ENOMEM;
+					goto out;
+				}
+			}
+			goto build_flags;
+		} else
 			goto out;
-		}
 	}
 
-	if (strcmp(sb->s_type->name, "proc") == 0)
-		sbsec->proc = 1;
+	/* Standard string-based options. */
+	while ((p = strsep(&options, "|")) != NULL) {
+		int token;
+		substring_t args[MAX_OPT_ARGS];
 
-	sbsec->initialized = 1;
+		if (!*p)
+			continue;
 
-	if (sbsec->behavior > ARRAY_SIZE(labeling_behaviors)) {
-		printk(KERN_ERR "SELinux: initialized (dev %s, type %s), unknown behavior\n",
-		       sb->s_id, sb->s_type->name);
-	}
-	else {
-		printk(KERN_DEBUG "SELinux: initialized (dev %s, type %s), %s\n",
-		       sb->s_id, sb->s_type->name,
-		       labeling_behaviors[sbsec->behavior-1]);
-	}
+		token = match_token(p, tokens, args);
 
-	/* Initialize the root inode. */
-	rc = inode_doinit_with_dentry(sb->s_root->d_inode, sb->s_root);
+		switch (token) {
+		case Opt_context:
+			if (context || defcontext) {
+				rc = -EINVAL;
+				printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
+				goto out_err;
+			}
+			context = match_strdup(&args[0]);
+			if (!context) {
+				rc = -ENOMEM;
+				goto out_err;
+			}
+			break;
+
+		case Opt_fscontext:
+			if (fscontext) {
+				rc = -EINVAL;
+				printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
+				goto out_err;
+			}
+			fscontext = match_strdup(&args[0]);
+			if (!fscontext) {
+				rc = -ENOMEM;
+				goto out_err;
+			}
+			break;
+
+		case Opt_rootcontext:
+			if (rootcontext) {
+				rc = -EINVAL;
+				printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
+				goto out_err;
+			}
+			rootcontext = match_strdup(&args[0]);
+			if (!rootcontext) {
+				rc = -ENOMEM;
+				goto out_err;
+			}
+			break;
+
+		case Opt_defcontext:
+			if (context || defcontext) {
+				rc = -EINVAL;
+				printk(KERN_WARNING SEL_MOUNT_FAIL_MSG);
+				goto out_err;
+			}
+			defcontext = match_strdup(&args[0]);
+			if (!defcontext) {
+				rc = -ENOMEM;
+				goto out_err;
+			}
+			break;
+
+		default:
+			rc = -EINVAL;
+			printk(KERN_WARNING "SELinux:  unknown mount option\n");
+			goto out_err;
 
-	/* Initialize any other inodes associated with the superblock, e.g.
-	   inodes created prior to initial policy load or inodes created
-	   during get_sb by a pseudo filesystem that directly
-	   populates itself. */
-	spin_lock(&sbsec->isec_lock);
-next_inode:
-	if (!list_empty(&sbsec->isec_head)) {
-		struct inode_security_struct *isec =
-				list_entry(sbsec->isec_head.next,
-				           struct inode_security_struct, list);
-		struct inode *inode = isec->inode;
-		spin_unlock(&sbsec->isec_lock);
-		inode = igrab(inode);
-		if (inode) {
-			if (!IS_PRIVATE (inode))
-				inode_doinit(inode);
-			iput(inode);
 		}
-		spin_lock(&sbsec->isec_lock);
-		list_del_init(&isec->list);
-		goto next_inode;
 	}
-	spin_unlock(&sbsec->isec_lock);
+
+build_flags:
+	if (fscontext) {
+		mnt_opts[num_mnt_opts] = fscontext;
+		mnt_opts_flags[num_mnt_opts++] = FSCONTEXT_MNT;
+	}
+	if (context) {
+		mnt_opts[num_mnt_opts] = context;
+		mnt_opts_flags[num_mnt_opts++] = CONTEXT_MNT;
+	}
+	if (rootcontext) {
+		mnt_opts[num_mnt_opts] = rootcontext;
+		mnt_opts_flags[num_mnt_opts++] = ROOTCONTEXT_MNT;
+	}
+	if (defcontext) {
+		mnt_opts[num_mnt_opts] = defcontext;
+		mnt_opts_flags[num_mnt_opts++] = DEFCONTEXT_MNT;
+	}
+
 out:
-	mutex_unlock(&sbsec->lock);
+	rc = selinux_set_mnt_opts(sb, mnt_opts, mnt_opts_flags, num_mnt_opts);
+out_err:
+	kfree(context);
+	kfree(defcontext);
+	kfree(fscontext);
+	kfree(rootcontext);
 	return rc;
 }
 
@@ -4800,6 +5037,9 @@ static struct security_operations selinux_ops = {
 	.sb_statfs =			selinux_sb_statfs,
 	.sb_mount =			selinux_mount,
 	.sb_umount =			selinux_umount,
+	.sb_get_mnt_opts =		selinux_get_mnt_opts,
+	.sb_set_mnt_opts =		selinux_set_mnt_opts,
+	.sb_clone_mnt_opts = 		selinux_sb_clone_mnt_opts,
 
 	.inode_alloc_security =		selinux_inode_alloc_security,
 	.inode_free_security =		selinux_inode_free_security,
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 642a9fd319ad..4138a80f8e27 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -65,6 +65,7 @@ struct superblock_security_struct {
 	u32 mntpoint_sid;		/* SECURITY_FS_USE_MNTPOINT context for files */
 	unsigned int behavior;          /* labeling behavior */
 	unsigned char initialized;      /* initialization flag */
+	unsigned char flags;		/* which mount options were specified */
 	unsigned char proc;             /* proc fs */
 	struct mutex lock;
 	struct list_head isec_head;
-- 
cgit v1.2.3


From 42d7896ebc5f7268b1fe6bbd20f2282e20ae7895 Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@sdv.(none)>
Date: Wed, 26 Dec 2007 15:12:37 +1100
Subject: Security: remove security.h include from mm.h

Remove security.h include from mm.h, as it is only needed for a single
extern declaration, and pulls in all kinds of crud.

Fine-by-me: David Chinner <dgc@sgi.com>
Acked-by: Eric Paris <eparis@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/mm.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 1b7b95c67aca..1897ca223eca 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -12,7 +12,6 @@
 #include <linux/prio_tree.h>
 #include <linux/debug_locks.h>
 #include <linux/mm_types.h>
-#include <linux/security.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -34,6 +33,8 @@ extern int sysctl_legacy_va_layout;
 #define sysctl_legacy_va_layout 0
 #endif
 
+extern unsigned long mmap_min_addr;
+
 #include <asm/page.h>
 #include <asm/pgtable.h>
 #include <asm/processor.h>
-- 
cgit v1.2.3


From bced95283e9434611cbad8f2ff903cd396eaea72 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Sat, 29 Dec 2007 16:20:25 -0800
Subject: security: remove security_sb_post_mountroot hook

The security_sb_post_mountroot() hook is long-since obsolete, and is
fundamentally broken: it is never invoked if someone uses initramfs.
This is particularly damaging, because the existence of this hook has
been used as motivation for not using initramfs.

Stephen Smalley confirmed on 2007-07-19 that this hook was originally
used by SELinux but can now be safely removed:

     http://marc.info/?l=linux-kernel&m=118485683612916&w=2

Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: James Morris <jmorris@namei.org>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h | 8 --------
 init/do_mounts.c         | 1 -
 security/dummy.c         | 6 ------
 security/security.c      | 5 -----
 4 files changed, 20 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index cbd970a735f2..2e2c63faead7 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -249,9 +249,6 @@ struct request_sock;
  *	@mnt contains the mounted file system.
  *	@flags contains the new filesystem flags.
  *	@data contains the filesystem-specific data.
- * @sb_post_mountroot:
- *	Update the security module's state when the root filesystem is mounted.
- *	This hook is only called if the mount was successful.
  * @sb_post_addmount:
  *	Update the security module's state when a filesystem is mounted.
  *	This hook is called any time a mount is successfully grafetd to
@@ -1257,7 +1254,6 @@ struct security_operations {
 	void (*sb_umount_busy) (struct vfsmount * mnt);
 	void (*sb_post_remount) (struct vfsmount * mnt,
 				 unsigned long flags, void *data);
-	void (*sb_post_mountroot) (void);
 	void (*sb_post_addmount) (struct vfsmount * mnt,
 				  struct nameidata * mountpoint_nd);
 	int (*sb_pivotroot) (struct nameidata * old_nd,
@@ -1524,7 +1520,6 @@ int security_sb_umount(struct vfsmount *mnt, int flags);
 void security_sb_umount_close(struct vfsmount *mnt);
 void security_sb_umount_busy(struct vfsmount *mnt);
 void security_sb_post_remount(struct vfsmount *mnt, unsigned long flags, void *data);
-void security_sb_post_mountroot(void);
 void security_sb_post_addmount(struct vfsmount *mnt, struct nameidata *mountpoint_nd);
 int security_sb_pivotroot(struct nameidata *old_nd, struct nameidata *new_nd);
 void security_sb_post_pivotroot(struct nameidata *old_nd, struct nameidata *new_nd);
@@ -1813,9 +1808,6 @@ static inline void security_sb_post_remount (struct vfsmount *mnt,
 					     unsigned long flags, void *data)
 { }
 
-static inline void security_sb_post_mountroot (void)
-{ }
-
 static inline void security_sb_post_addmount (struct vfsmount *mnt,
 					      struct nameidata *mountpoint_nd)
 { }
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 4efa1e5385e3..31b2185ce304 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -470,6 +470,5 @@ void __init prepare_namespace(void)
 out:
 	sys_mount(".", "/", NULL, MS_MOVE, NULL);
 	sys_chroot(".");
-	security_sb_post_mountroot();
 }
 
diff --git a/security/dummy.c b/security/dummy.c
index a3b29d0d00e5..8e34e03415f8 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -225,11 +225,6 @@ static void dummy_sb_post_remount (struct vfsmount *mnt, unsigned long flags,
 }
 
 
-static void dummy_sb_post_mountroot (void)
-{
-	return;
-}
-
 static void dummy_sb_post_addmount (struct vfsmount *mnt, struct nameidata *nd)
 {
 	return;
@@ -1017,7 +1012,6 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sb_umount_close);
 	set_to_dummy_if_null(ops, sb_umount_busy);
 	set_to_dummy_if_null(ops, sb_post_remount);
-	set_to_dummy_if_null(ops, sb_post_mountroot);
 	set_to_dummy_if_null(ops, sb_post_addmount);
 	set_to_dummy_if_null(ops, sb_pivotroot);
 	set_to_dummy_if_null(ops, sb_post_pivotroot);
diff --git a/security/security.c b/security/security.c
index b13b54f0af85..5068808343d0 100644
--- a/security/security.c
+++ b/security/security.c
@@ -288,11 +288,6 @@ void security_sb_post_remount(struct vfsmount *mnt, unsigned long flags, void *d
 	security_ops->sb_post_remount(mnt, flags, data);
 }
 
-void security_sb_post_mountroot(void)
-{
-	security_ops->sb_post_mountroot();
-}
-
 void security_sb_post_addmount(struct vfsmount *mnt, struct nameidata *mountpoint_nd)
 {
 	security_ops->sb_post_addmount(mnt, mountpoint_nd);
-- 
cgit v1.2.3


From 63cb34492351078479b2d4bae6a881806a396286 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 15 Jan 2008 23:47:35 +0000
Subject: security: add a secctx_to_secid() hook

Add a secctx_to_secid() LSM hook to go along with the existing
secid_to_secctx() LSM hook.  This patch also includes the SELinux
implementation for this hook.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h | 13 +++++++++++++
 security/dummy.c         |  6 ++++++
 security/security.c      |  6 ++++++
 security/selinux/hooks.c |  6 ++++++
 4 files changed, 31 insertions(+)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 2e2c63faead7..e4a91cb1b187 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1202,6 +1202,10 @@ struct request_sock;
  *	Convert secid to security context.
  *	@secid contains the security ID.
  *	@secdata contains the pointer that stores the converted security context.
+ * @secctx_to_secid:
+ *      Convert security context to secid.
+ *      @secid contains the pointer to the generated security ID.
+ *      @secdata contains the security context.
  *
  * @release_secctx:
  *	Release the security context.
@@ -1396,6 +1400,7 @@ struct security_operations {
  	int (*getprocattr)(struct task_struct *p, char *name, char **value);
  	int (*setprocattr)(struct task_struct *p, char *name, void *value, size_t size);
 	int (*secid_to_secctx)(u32 secid, char **secdata, u32 *seclen);
+	int (*secctx_to_secid)(char *secdata, u32 seclen, u32 *secid);
 	void (*release_secctx)(char *secdata, u32 seclen);
 
 #ifdef CONFIG_SECURITY_NETWORK
@@ -1634,6 +1639,7 @@ int security_setprocattr(struct task_struct *p, char *name, void *value, size_t
 int security_netlink_send(struct sock *sk, struct sk_buff *skb);
 int security_netlink_recv(struct sk_buff *skb, int cap);
 int security_secid_to_secctx(u32 secid, char **secdata, u32 *seclen);
+int security_secctx_to_secid(char *secdata, u32 seclen, u32 *secid);
 void security_release_secctx(char *secdata, u32 seclen);
 
 #else /* CONFIG_SECURITY */
@@ -2308,6 +2314,13 @@ static inline int security_secid_to_secctx(u32 secid, char **secdata, u32 *secle
 	return -EOPNOTSUPP;
 }
 
+static inline int security_secctx_to_secid(char *secdata,
+					   u32 seclen,
+					   u32 *secid)
+{
+	return -EOPNOTSUPP;
+}
+
 static inline void security_release_secctx(char *secdata, u32 seclen)
 {
 }
diff --git a/security/dummy.c b/security/dummy.c
index 8e34e03415f8..48d4b0a52737 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -946,6 +946,11 @@ static int dummy_secid_to_secctx(u32 secid, char **secdata, u32 *seclen)
 	return -EOPNOTSUPP;
 }
 
+static int dummy_secctx_to_secid(char *secdata, u32 seclen, u32 *secid)
+{
+	return -EOPNOTSUPP;
+}
+
 static void dummy_release_secctx(char *secdata, u32 seclen)
 {
 }
@@ -1106,6 +1111,7 @@ void security_fixup_ops (struct security_operations *ops)
  	set_to_dummy_if_null(ops, getprocattr);
  	set_to_dummy_if_null(ops, setprocattr);
  	set_to_dummy_if_null(ops, secid_to_secctx);
+	set_to_dummy_if_null(ops, secctx_to_secid);
  	set_to_dummy_if_null(ops, release_secctx);
 #ifdef CONFIG_SECURITY_NETWORK
 	set_to_dummy_if_null(ops, unix_stream_connect);
diff --git a/security/security.c b/security/security.c
index 5068808343d0..ca475ca206e4 100644
--- a/security/security.c
+++ b/security/security.c
@@ -831,6 +831,12 @@ int security_secid_to_secctx(u32 secid, char **secdata, u32 *seclen)
 }
 EXPORT_SYMBOL(security_secid_to_secctx);
 
+int security_secctx_to_secid(char *secdata, u32 seclen, u32 *secid)
+{
+	return security_ops->secctx_to_secid(secdata, seclen, secid);
+}
+EXPORT_SYMBOL(security_secctx_to_secid);
+
 void security_release_secctx(char *secdata, u32 seclen)
 {
 	return security_ops->release_secctx(secdata, seclen);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 233c8b97462f..0396354fff95 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -4947,6 +4947,11 @@ static int selinux_secid_to_secctx(u32 secid, char **secdata, u32 *seclen)
 	return security_sid_to_context(secid, secdata, seclen);
 }
 
+static int selinux_secctx_to_secid(char *secdata, u32 seclen, u32 *secid)
+{
+	return security_context_to_sid(secdata, seclen, secid);
+}
+
 static void selinux_release_secctx(char *secdata, u32 seclen)
 {
 	kfree(secdata);
@@ -5138,6 +5143,7 @@ static struct security_operations selinux_ops = {
 	.setprocattr =                  selinux_setprocattr,
 
 	.secid_to_secctx =		selinux_secid_to_secctx,
+	.secctx_to_secid =		selinux_secctx_to_secid,
 	.release_secctx =		selinux_release_secctx,
 
         .unix_stream_connect =		selinux_socket_unix_stream_connect,
-- 
cgit v1.2.3


From 1996a10948e50e546dc2b64276723c0b64d3173b Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Wed, 23 Jan 2008 00:02:58 +0100
Subject: security/selinux: constify function pointer tables and fields

Constify function pointer tables and fields.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h     | 2 +-
 security/keys/proc.c         | 4 ++--
 security/selinux/selinuxfs.c | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index e4a91cb1b187..d24974262dc6 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -2300,7 +2300,7 @@ static inline struct dentry *securityfs_create_file(const char *name,
 						mode_t mode,
 						struct dentry *parent,
 						void *data,
-						struct file_operations *fops)
+						const struct file_operations *fops)
 {
 	return ERR_PTR(-ENODEV);
 }
diff --git a/security/keys/proc.c b/security/keys/proc.c
index 3e0d0a6e224f..694126003ed3 100644
--- a/security/keys/proc.c
+++ b/security/keys/proc.c
@@ -26,7 +26,7 @@ static void *proc_keys_next(struct seq_file *p, void *v, loff_t *_pos);
 static void proc_keys_stop(struct seq_file *p, void *v);
 static int proc_keys_show(struct seq_file *m, void *v);
 
-static struct seq_operations proc_keys_ops = {
+static const struct seq_operations proc_keys_ops = {
 	.start	= proc_keys_start,
 	.next	= proc_keys_next,
 	.stop	= proc_keys_stop,
@@ -47,7 +47,7 @@ static void *proc_key_users_next(struct seq_file *p, void *v, loff_t *_pos);
 static void proc_key_users_stop(struct seq_file *p, void *v);
 static int proc_key_users_show(struct seq_file *m, void *v);
 
-static struct seq_operations proc_key_users_ops = {
+static const struct seq_operations proc_key_users_ops = {
 	.start	= proc_key_users_start,
 	.next	= proc_key_users_next,
 	.stop	= proc_key_users_stop,
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index 2fa483f26113..397fd4955fe1 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -1222,7 +1222,7 @@ static int sel_avc_stats_seq_show(struct seq_file *seq, void *v)
 static void sel_avc_stats_seq_stop(struct seq_file *seq, void *v)
 { }
 
-static struct seq_operations sel_avc_cache_stats_seq_ops = {
+static const struct seq_operations sel_avc_cache_stats_seq_ops = {
 	.start		= sel_avc_stats_seq_start,
 	.next		= sel_avc_stats_seq_next,
 	.show		= sel_avc_stats_seq_show,
-- 
cgit v1.2.3


From 775b64d2b6ca37697de925f70799c710aab5849a Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sat, 12 Jan 2008 20:40:46 +0100
Subject: PM: Acquire device locks on suspend

This patch reorganizes the way suspend and resume notifications are
sent to drivers.  The major changes are that now the PM core acquires
every device semaphore before calling the methods, and calls to
device_add() during suspends will fail, while calls to device_del()
during suspends will block.

It also provides a way to safely remove a suspended device with the
help of the PM core, by using the device_pm_schedule_removal() callback
introduced specifically for this purpose, and updates two drivers (msr
and cpuid) that need to use it.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/x86/kernel/cpuid.c    |   6 +-
 arch/x86/kernel/msr.c      |   6 +-
 drivers/base/core.c        |  65 +++++-
 drivers/base/power/main.c  | 502 +++++++++++++++++++++++++++++----------------
 drivers/base/power/power.h |  12 ++
 include/linux/device.h     |   8 +
 6 files changed, 413 insertions(+), 186 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index 05c9936a16cc..d387c770c518 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -157,15 +157,15 @@ static int __cpuinit cpuid_class_cpu_callback(struct notifier_block *nfb,
 
 	switch (action) {
 	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
 		err = cpuid_device_create(cpu);
 		break;
 	case CPU_UP_CANCELED:
-	case CPU_UP_CANCELED_FROZEN:
 	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
 		cpuid_device_destroy(cpu);
 		break;
+	case CPU_UP_CANCELED_FROZEN:
+		destroy_suspended_device(cpuid_class, MKDEV(CPUID_MAJOR, cpu));
+		break;
 	}
 	return err ? NOTIFY_BAD : NOTIFY_OK;
 }
diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c
index ee6eba4ecfea..21f6e3c0be18 100644
--- a/arch/x86/kernel/msr.c
+++ b/arch/x86/kernel/msr.c
@@ -155,15 +155,15 @@ static int __cpuinit msr_class_cpu_callback(struct notifier_block *nfb,
 
 	switch (action) {
 	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
 		err = msr_device_create(cpu);
 		break;
 	case CPU_UP_CANCELED:
-	case CPU_UP_CANCELED_FROZEN:
 	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
 		msr_device_destroy(cpu);
 		break;
+	case CPU_UP_CANCELED_FROZEN:
+		destroy_suspended_device(msr_class, MKDEV(MSR_MAJOR, cpu));
+		break;
 	}
 	return err ? NOTIFY_BAD : NOTIFY_OK;
 }
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 2683eac30c68..ce6b64c489ad 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -726,11 +726,20 @@ int device_add(struct device *dev)
 {
 	struct device *parent = NULL;
 	struct class_interface *class_intf;
-	int error = -EINVAL;
+	int error;
+
+	error = pm_sleep_lock();
+	if (error) {
+		dev_warn(dev, "Suspicious %s during suspend\n", __FUNCTION__);
+		dump_stack();
+		return error;
+	}
 
 	dev = get_device(dev);
-	if (!dev || !strlen(dev->bus_id))
+	if (!dev || !strlen(dev->bus_id)) {
+		error = -EINVAL;
 		goto Error;
+	}
 
 	pr_debug("DEV: registering device: ID = '%s'\n", dev->bus_id);
 
@@ -795,6 +804,7 @@ int device_add(struct device *dev)
 	}
  Done:
 	put_device(dev);
+	pm_sleep_unlock();
 	return error;
  BusError:
 	device_pm_remove(dev);
@@ -905,6 +915,7 @@ void device_del(struct device * dev)
 	struct device * parent = dev->parent;
 	struct class_interface *class_intf;
 
+	device_pm_remove(dev);
 	if (parent)
 		klist_del(&dev->knode_parent);
 	if (MAJOR(dev->devt))
@@ -981,7 +992,6 @@ void device_del(struct device * dev)
 	if (dev->bus)
 		blocking_notifier_call_chain(&dev->bus->bus_notifier,
 					     BUS_NOTIFY_DEL_DEVICE, dev);
-	device_pm_remove(dev);
 	kobject_uevent(&dev->kobj, KOBJ_REMOVE);
 	kobject_del(&dev->kobj);
 	if (parent)
@@ -1156,14 +1166,11 @@ error:
 EXPORT_SYMBOL_GPL(device_create);
 
 /**
- * device_destroy - removes a device that was created with device_create()
+ * find_device - finds a device that was created with device_create()
  * @class: pointer to the struct class that this device was registered with
  * @devt: the dev_t of the device that was previously registered
- *
- * This call unregisters and cleans up a device that was created with a
- * call to device_create().
  */
-void device_destroy(struct class *class, dev_t devt)
+static struct device *find_device(struct class *class, dev_t devt)
 {
 	struct device *dev = NULL;
 	struct device *dev_tmp;
@@ -1176,12 +1183,54 @@ void device_destroy(struct class *class, dev_t devt)
 		}
 	}
 	up(&class->sem);
+	return dev;
+}
 
+/**
+ * device_destroy - removes a device that was created with device_create()
+ * @class: pointer to the struct class that this device was registered with
+ * @devt: the dev_t of the device that was previously registered
+ *
+ * This call unregisters and cleans up a device that was created with a
+ * call to device_create().
+ */
+void device_destroy(struct class *class, dev_t devt)
+{
+	struct device *dev;
+
+	dev = find_device(class, devt);
 	if (dev)
 		device_unregister(dev);
 }
 EXPORT_SYMBOL_GPL(device_destroy);
 
+#ifdef CONFIG_PM_SLEEP
+/**
+ * destroy_suspended_device - asks the PM core to remove a suspended device
+ * @class: pointer to the struct class that this device was registered with
+ * @devt: the dev_t of the device that was previously registered
+ *
+ * This call notifies the PM core of the necessity to unregister a suspended
+ * device created with a call to device_create() (devices cannot be
+ * unregistered directly while suspended, since the PM core holds their
+ * semaphores at that time).
+ *
+ * It can only be called within the scope of a system sleep transition.  In
+ * practice this means it has to be directly or indirectly invoked either by
+ * a suspend or resume method, or by the PM core (e.g. via
+ * disable_nonboot_cpus() or enable_nonboot_cpus()).
+ */
+void destroy_suspended_device(struct class *class, dev_t devt)
+{
+	struct device *dev;
+
+	dev = find_device(class, devt);
+	if (dev)
+		device_pm_schedule_removal(dev);
+}
+EXPORT_SYMBOL_GPL(destroy_suspended_device);
+#endif /* CONFIG_PM_SLEEP */
+
 /**
  * device_rename - renames a device
  * @dev: the pointer to the struct device to be renamed
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 691ffb64cc37..200ed5fafd50 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -24,20 +24,45 @@
 #include <linux/mutex.h>
 #include <linux/pm.h>
 #include <linux/resume-trace.h>
+#include <linux/rwsem.h>
 
 #include "../base.h"
 #include "power.h"
 
+/*
+ * The entries in the dpm_active list are in a depth first order, simply
+ * because children are guaranteed to be discovered after parents, and
+ * are inserted at the back of the list on discovery.
+ *
+ * All the other lists are kept in the same order, for consistency.
+ * However the lists aren't always traversed in the same order.
+ * Semaphores must be acquired from the top (i.e., front) down
+ * and released in the opposite order.  Devices must be suspended
+ * from the bottom (i.e., end) up and resumed in the opposite order.
+ * That way no parent will be suspended while it still has an active
+ * child.
+ *
+ * Since device_pm_add() may be called with a device semaphore held,
+ * we must never try to acquire a device semaphore while holding
+ * dpm_list_mutex.
+ */
+
 LIST_HEAD(dpm_active);
+static LIST_HEAD(dpm_locked);
 static LIST_HEAD(dpm_off);
 static LIST_HEAD(dpm_off_irq);
+static LIST_HEAD(dpm_destroy);
 
-static DEFINE_MUTEX(dpm_mtx);
 static DEFINE_MUTEX(dpm_list_mtx);
 
-int (*platform_enable_wakeup)(struct device *dev, int is_on);
+static DECLARE_RWSEM(pm_sleep_rwsem);
 
+int (*platform_enable_wakeup)(struct device *dev, int is_on);
 
+/**
+ *	device_pm_add - add a device to the list of active devices
+ *	@dev:	Device to be added to the list
+ */
 void device_pm_add(struct device *dev)
 {
 	pr_debug("PM: Adding info for %s:%s\n",
@@ -48,8 +73,36 @@ void device_pm_add(struct device *dev)
 	mutex_unlock(&dpm_list_mtx);
 }
 
+/**
+ *	device_pm_remove - remove a device from the list of active devices
+ *	@dev:	Device to be removed from the list
+ *
+ *	This function also removes the device's PM-related sysfs attributes.
+ */
 void device_pm_remove(struct device *dev)
 {
+	/*
+	 * If this function is called during a suspend, it will be blocked,
+	 * because we're holding the device's semaphore at that time, which may
+	 * lead to a deadlock.  In that case we want to print a warning.
+	 * However, it may also be called by unregister_dropped_devices() with
+	 * the device's semaphore released, in which case the warning should
+	 * not be printed.
+	 */
+	if (down_trylock(&dev->sem)) {
+		if (down_read_trylock(&pm_sleep_rwsem)) {
+			/* No suspend in progress, wait on dev->sem */
+			down(&dev->sem);
+			up_read(&pm_sleep_rwsem);
+		} else {
+			/* Suspend in progress, we may deadlock */
+			dev_warn(dev, "Suspicious %s during suspend\n",
+				__FUNCTION__);
+			dump_stack();
+			/* The user has been warned ... */
+			down(&dev->sem);
+		}
+	}
 	pr_debug("PM: Removing info for %s:%s\n",
 		 dev->bus ? dev->bus->name : "No Bus",
 		 kobject_name(&dev->kobj));
@@ -57,25 +110,124 @@ void device_pm_remove(struct device *dev)
 	dpm_sysfs_remove(dev);
 	list_del_init(&dev->power.entry);
 	mutex_unlock(&dpm_list_mtx);
+	up(&dev->sem);
+}
+
+/**
+ *	device_pm_schedule_removal - schedule the removal of a suspended device
+ *	@dev:	Device to destroy
+ *
+ *	Moves the device to the dpm_destroy list for further processing by
+ *	unregister_dropped_devices().
+ */
+void device_pm_schedule_removal(struct device *dev)
+{
+	pr_debug("PM: Preparing for removal: %s:%s\n",
+		dev->bus ? dev->bus->name : "No Bus",
+		kobject_name(&dev->kobj));
+	mutex_lock(&dpm_list_mtx);
+	list_move_tail(&dev->power.entry, &dpm_destroy);
+	mutex_unlock(&dpm_list_mtx);
+}
+
+/**
+ *	pm_sleep_lock - mutual exclusion for registration and suspend
+ *
+ *	Returns 0 if no suspend is underway and device registration
+ *	may proceed, otherwise -EBUSY.
+ */
+int pm_sleep_lock(void)
+{
+	if (down_read_trylock(&pm_sleep_rwsem))
+		return 0;
+
+	return -EBUSY;
+}
+
+/**
+ *	pm_sleep_unlock - mutual exclusion for registration and suspend
+ *
+ *	This routine undoes the effect of device_pm_add_lock
+ *	when a device's registration is complete.
+ */
+void pm_sleep_unlock(void)
+{
+	up_read(&pm_sleep_rwsem);
 }
 
 
 /*------------------------- Resume routines -------------------------*/
 
 /**
- *	resume_device - Restore state for one device.
+ *	resume_device_early - Power on one device (early resume).
  *	@dev:	Device.
  *
+ *	Must be called with interrupts disabled.
  */
-
-static int resume_device(struct device * dev)
+static int resume_device_early(struct device *dev)
 {
 	int error = 0;
 
 	TRACE_DEVICE(dev);
 	TRACE_RESUME(0);
 
-	down(&dev->sem);
+	if (dev->bus && dev->bus->resume_early) {
+		dev_dbg(dev, "EARLY resume\n");
+		error = dev->bus->resume_early(dev);
+	}
+
+	TRACE_RESUME(error);
+	return error;
+}
+
+/**
+ *	dpm_power_up - Power on all regular (non-sysdev) devices.
+ *
+ *	Walk the dpm_off_irq list and power each device up. This
+ *	is used for devices that required they be powered down with
+ *	interrupts disabled. As devices are powered on, they are moved
+ *	to the dpm_off list.
+ *
+ *	Must be called with interrupts disabled and only one CPU running.
+ */
+static void dpm_power_up(void)
+{
+
+	while (!list_empty(&dpm_off_irq)) {
+		struct list_head *entry = dpm_off_irq.next;
+		struct device *dev = to_device(entry);
+
+		list_move_tail(entry, &dpm_off);
+		resume_device_early(dev);
+	}
+}
+
+/**
+ *	device_power_up - Turn on all devices that need special attention.
+ *
+ *	Power on system devices, then devices that required we shut them down
+ *	with interrupts disabled.
+ *
+ *	Must be called with interrupts disabled.
+ */
+void device_power_up(void)
+{
+	sysdev_resume();
+	dpm_power_up();
+}
+EXPORT_SYMBOL_GPL(device_power_up);
+
+/**
+ *	resume_device - Restore state for one device.
+ *	@dev:	Device.
+ *
+ */
+static int resume_device(struct device *dev)
+{
+	int error = 0;
+
+	TRACE_DEVICE(dev);
+	TRACE_RESUME(0);
 
 	if (dev->bus && dev->bus->resume) {
 		dev_dbg(dev,"resuming\n");
@@ -92,126 +244,94 @@ static int resume_device(struct device * dev)
 		error = dev->class->resume(dev);
 	}
 
-	up(&dev->sem);
-
 	TRACE_RESUME(error);
 	return error;
 }
 
-
-static int resume_device_early(struct device * dev)
-{
-	int error = 0;
-
-	TRACE_DEVICE(dev);
-	TRACE_RESUME(0);
-	if (dev->bus && dev->bus->resume_early) {
-		dev_dbg(dev,"EARLY resume\n");
-		error = dev->bus->resume_early(dev);
-	}
-	TRACE_RESUME(error);
-	return error;
-}
-
-/*
- * Resume the devices that have either not gone through
- * the late suspend, or that did go through it but also
- * went through the early resume
+/**
+ *	dpm_resume - Resume every device.
+ *
+ *	Resume the devices that have either not gone through
+ *	the late suspend, or that did go through it but also
+ *	went through the early resume.
+ *
+ *	Take devices from the dpm_off_list, resume them,
+ *	and put them on the dpm_locked list.
  */
 static void dpm_resume(void)
 {
 	mutex_lock(&dpm_list_mtx);
 	while(!list_empty(&dpm_off)) {
-		struct list_head * entry = dpm_off.next;
-		struct device * dev = to_device(entry);
-
-		get_device(dev);
-		list_move_tail(entry, &dpm_active);
+		struct list_head *entry = dpm_off.next;
+		struct device *dev = to_device(entry);
 
+		list_move_tail(entry, &dpm_locked);
 		mutex_unlock(&dpm_list_mtx);
 		resume_device(dev);
 		mutex_lock(&dpm_list_mtx);
-		put_device(dev);
 	}
 	mutex_unlock(&dpm_list_mtx);
 }
 
-
 /**
- *	device_resume - Restore state of each device in system.
+ *	unlock_all_devices - Release each device's semaphore
  *
- *	Walk the dpm_off list, remove each entry, resume the device,
- *	then add it to the dpm_active list.
+ *	Go through the dpm_off list.  Put each device on the dpm_active
+ *	list and unlock it.
  */
-
-void device_resume(void)
+static void unlock_all_devices(void)
 {
-	might_sleep();
-	mutex_lock(&dpm_mtx);
-	dpm_resume();
-	mutex_unlock(&dpm_mtx);
-}
-
-EXPORT_SYMBOL_GPL(device_resume);
+	mutex_lock(&dpm_list_mtx);
+	while (!list_empty(&dpm_locked)) {
+		struct list_head *entry = dpm_locked.prev;
+		struct device *dev = to_device(entry);
 
+		list_move(entry, &dpm_active);
+		up(&dev->sem);
+	}
+	mutex_unlock(&dpm_list_mtx);
+}
 
 /**
- *	dpm_power_up - Power on some devices.
- *
- *	Walk the dpm_off_irq list and power each device up. This
- *	is used for devices that required they be powered down with
- *	interrupts disabled. As devices are powered on, they are moved
- *	to the dpm_active list.
+ *	unregister_dropped_devices - Unregister devices scheduled for removal
  *
- *	Interrupts must be disabled when calling this.
+ *	Unregister all devices on the dpm_destroy list.
  */
-
-static void dpm_power_up(void)
+static void unregister_dropped_devices(void)
 {
-	while(!list_empty(&dpm_off_irq)) {
-		struct list_head * entry = dpm_off_irq.next;
-		struct device * dev = to_device(entry);
+	mutex_lock(&dpm_list_mtx);
+	while (!list_empty(&dpm_destroy)) {
+		struct list_head *entry = dpm_destroy.next;
+		struct device *dev = to_device(entry);
 
-		list_move_tail(entry, &dpm_off);
-		resume_device_early(dev);
+		up(&dev->sem);
+		mutex_unlock(&dpm_list_mtx);
+		/* This also removes the device from the list */
+		device_unregister(dev);
+		mutex_lock(&dpm_list_mtx);
 	}
+	mutex_unlock(&dpm_list_mtx);
 }
 
-
 /**
- *	device_power_up - Turn on all devices that need special attention.
+ *	device_resume - Restore state of each device in system.
  *
- *	Power on system devices then devices that required we shut them down
- *	with interrupts disabled.
- *	Called with interrupts disabled.
+ *	Resume all the devices, unlock them all, and allow new
+ *	devices to be registered once again.
  */
-
-void device_power_up(void)
+void device_resume(void)
 {
-	sysdev_resume();
-	dpm_power_up();
+	might_sleep();
+	dpm_resume();
+	unlock_all_devices();
+	unregister_dropped_devices();
+	up_write(&pm_sleep_rwsem);
 }
-
-EXPORT_SYMBOL_GPL(device_power_up);
+EXPORT_SYMBOL_GPL(device_resume);
 
 
 /*------------------------- Suspend routines -------------------------*/
 
-/*
- * The entries in the dpm_active list are in a depth first order, simply
- * because children are guaranteed to be discovered after parents, and
- * are inserted at the back of the list on discovery.
- *
- * All list on the suspend path are done in reverse order, so we operate
- * on the leaves of the device tree (or forests, depending on how you want
- * to look at it ;) first. As nodes are removed from the back of the list,
- * they are inserted into the front of their destintation lists.
- *
- * Things are the reverse on the resume path - iterations are done in
- * forward order, and nodes are inserted at the back of their destination
- * lists. This way, the ancestors will be accessed before their descendents.
- */
-
 static inline char *suspend_verb(u32 event)
 {
 	switch (event) {
@@ -222,7 +342,6 @@ static inline char *suspend_verb(u32 event)
 	}
 }
 
-
 static void
 suspend_device_dbg(struct device *dev, pm_message_t state, char *info)
 {
@@ -232,16 +351,73 @@ suspend_device_dbg(struct device *dev, pm_message_t state, char *info)
 }
 
 /**
- *	suspend_device - Save state of one device.
+ *	suspend_device_late - Shut down one device (late suspend).
  *	@dev:	Device.
  *	@state:	Power state device is entering.
+ *
+ *	This is called with interrupts off and only a single CPU running.
  */
+static int suspend_device_late(struct device *dev, pm_message_t state)
+{
+	int error = 0;
 
-static int suspend_device(struct device * dev, pm_message_t state)
+	if (dev->bus && dev->bus->suspend_late) {
+		suspend_device_dbg(dev, state, "LATE ");
+		error = dev->bus->suspend_late(dev, state);
+		suspend_report_result(dev->bus->suspend_late, error);
+	}
+	return error;
+}
+
+/**
+ *	device_power_down - Shut down special devices.
+ *	@state:		Power state to enter.
+ *
+ *	Power down devices that require interrupts to be disabled
+ *	and move them from the dpm_off list to the dpm_off_irq list.
+ *	Then power down system devices.
+ *
+ *	Must be called with interrupts disabled and only one CPU running.
+ */
+int device_power_down(pm_message_t state)
+{
+	int error = 0;
+
+	while (!list_empty(&dpm_off)) {
+		struct list_head *entry = dpm_off.prev;
+		struct device *dev = to_device(entry);
+
+		list_del_init(&dev->power.entry);
+		error = suspend_device_late(dev, state);
+		if (error) {
+			printk(KERN_ERR "Could not power down device %s: "
+					"error %d\n",
+					kobject_name(&dev->kobj), error);
+			if (list_empty(&dev->power.entry))
+				list_add(&dev->power.entry, &dpm_off);
+			break;
+		}
+		if (list_empty(&dev->power.entry))
+			list_add(&dev->power.entry, &dpm_off_irq);
+	}
+
+	if (!error)
+		error = sysdev_suspend(state);
+	if (error)
+		dpm_power_up();
+	return error;
+}
+EXPORT_SYMBOL_GPL(device_power_down);
+
+/**
+ *	suspend_device - Save state of one device.
+ *	@dev:	Device.
+ *	@state:	Power state device is entering.
+ */
+int suspend_device(struct device *dev, pm_message_t state)
 {
 	int error = 0;
 
-	down(&dev->sem);
 	if (dev->power.power_state.event) {
 		dev_dbg(dev, "PM: suspend %d-->%d\n",
 			dev->power.power_state.event, state.event);
@@ -264,123 +440,105 @@ static int suspend_device(struct device * dev, pm_message_t state)
 		error = dev->bus->suspend(dev, state);
 		suspend_report_result(dev->bus->suspend, error);
 	}
-	up(&dev->sem);
-	return error;
-}
-
-
-/*
- * This is called with interrupts off, only a single CPU
- * running. We can't acquire a mutex or semaphore (and we don't
- * need the protection)
- */
-static int suspend_device_late(struct device *dev, pm_message_t state)
-{
-	int error = 0;
-
-	if (dev->bus && dev->bus->suspend_late) {
-		suspend_device_dbg(dev, state, "LATE ");
-		error = dev->bus->suspend_late(dev, state);
-		suspend_report_result(dev->bus->suspend_late, error);
-	}
 	return error;
 }
 
 /**
- *	device_suspend - Save state and stop all devices in system.
- *	@state:		Power state to put each device in.
+ *	dpm_suspend - Suspend every device.
+ *	@state:	Power state to put each device in.
  *
- *	Walk the dpm_active list, call ->suspend() for each device, and move
- *	it to the dpm_off list.
+ *	Walk the dpm_locked list.  Suspend each device and move it
+ *	to the dpm_off list.
  *
  *	(For historical reasons, if it returns -EAGAIN, that used to mean
  *	that the device would be called again with interrupts disabled.
  *	These days, we use the "suspend_late()" callback for that, so we
  *	print a warning and consider it an error).
- *
- *	If we get a different error, try and back out.
- *
- *	If we hit a failure with any of the devices, call device_resume()
- *	above to bring the suspended devices back to life.
- *
  */
-
-int device_suspend(pm_message_t state)
+static int dpm_suspend(pm_message_t state)
 {
 	int error = 0;
 
-	might_sleep();
-	mutex_lock(&dpm_mtx);
 	mutex_lock(&dpm_list_mtx);
-	while (!list_empty(&dpm_active) && error == 0) {
-		struct list_head * entry = dpm_active.prev;
-		struct device * dev = to_device(entry);
+	while (!list_empty(&dpm_locked)) {
+		struct list_head *entry = dpm_locked.prev;
+		struct device *dev = to_device(entry);
 
-		get_device(dev);
+		list_del_init(&dev->power.entry);
 		mutex_unlock(&dpm_list_mtx);
-
 		error = suspend_device(dev, state);
-
-		mutex_lock(&dpm_list_mtx);
-
-		/* Check if the device got removed */
-		if (!list_empty(&dev->power.entry)) {
-			/* Move it to the dpm_off list */
-			if (!error)
-				list_move(&dev->power.entry, &dpm_off);
-		}
-		if (error)
+		if (error) {
 			printk(KERN_ERR "Could not suspend device %s: "
-				"error %d%s\n",
-				kobject_name(&dev->kobj), error,
-				error == -EAGAIN ? " (please convert to suspend_late)" : "");
-		put_device(dev);
+					"error %d%s\n",
+					kobject_name(&dev->kobj),
+					error,
+					(error == -EAGAIN ?
+					" (please convert to suspend_late)" :
+					""));
+			mutex_lock(&dpm_list_mtx);
+			if (list_empty(&dev->power.entry))
+				list_add(&dev->power.entry, &dpm_locked);
+			mutex_unlock(&dpm_list_mtx);
+			break;
+		}
+		mutex_lock(&dpm_list_mtx);
+		if (list_empty(&dev->power.entry))
+			list_add(&dev->power.entry, &dpm_off);
 	}
 	mutex_unlock(&dpm_list_mtx);
-	if (error)
-		dpm_resume();
 
-	mutex_unlock(&dpm_mtx);
 	return error;
 }
 
-EXPORT_SYMBOL_GPL(device_suspend);
-
 /**
- *	device_power_down - Shut down special devices.
- *	@state:		Power state to enter.
+ *	lock_all_devices - Acquire every device's semaphore
  *
- *	Walk the dpm_off_irq list, calling ->power_down() for each device that
- *	couldn't power down the device with interrupts enabled. When we're
- *	done, power down system devices.
+ *	Go through the dpm_active list. Carefully lock each device's
+ *	semaphore and put it in on the dpm_locked list.
  */
-
-int device_power_down(pm_message_t state)
+static void lock_all_devices(void)
 {
-	int error = 0;
-	struct device * dev;
+	mutex_lock(&dpm_list_mtx);
+	while (!list_empty(&dpm_active)) {
+		struct list_head *entry = dpm_active.next;
+		struct device *dev = to_device(entry);
 
-	while (!list_empty(&dpm_off)) {
-		struct list_head * entry = dpm_off.prev;
+		/* Required locking order is dev->sem first,
+		 * then dpm_list_mutex.  Hence this awkward code.
+		 */
+		get_device(dev);
+		mutex_unlock(&dpm_list_mtx);
+		down(&dev->sem);
+		mutex_lock(&dpm_list_mtx);
 
-		dev = to_device(entry);
-		error = suspend_device_late(dev, state);
-		if (error)
-			goto Error;
-		list_move(&dev->power.entry, &dpm_off_irq);
+		if (list_empty(entry))
+			up(&dev->sem);		/* Device was removed */
+		else
+			list_move_tail(entry, &dpm_locked);
+		put_device(dev);
 	}
+	mutex_unlock(&dpm_list_mtx);
+}
+
+/**
+ *	device_suspend - Save state and stop all devices in system.
+ *
+ *	Prevent new devices from being registered, then lock all devices
+ *	and suspend them.
+ */
+int device_suspend(pm_message_t state)
+{
+	int error;
 
-	error = sysdev_suspend(state);
- Done:
+	might_sleep();
+	down_write(&pm_sleep_rwsem);
+	lock_all_devices();
+	error = dpm_suspend(state);
+	if (error)
+		device_resume();
 	return error;
- Error:
-	printk(KERN_ERR "Could not power down device %s: "
-		"error %d\n", kobject_name(&dev->kobj), error);
-	dpm_power_up();
-	goto Done;
 }
-
-EXPORT_SYMBOL_GPL(device_power_down);
+EXPORT_SYMBOL_GPL(device_suspend);
 
 void __suspend_report_result(const char *function, void *fn, int ret)
 {
diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h
index 379da4e958e0..10c20840395e 100644
--- a/drivers/base/power/power.h
+++ b/drivers/base/power/power.h
@@ -20,6 +20,9 @@ static inline struct device *to_device(struct list_head *entry)
 
 extern void device_pm_add(struct device *);
 extern void device_pm_remove(struct device *);
+extern void device_pm_schedule_removal(struct device *);
+extern int pm_sleep_lock(void);
+extern void pm_sleep_unlock(void);
 
 #else /* CONFIG_PM_SLEEP */
 
@@ -32,6 +35,15 @@ static inline void device_pm_remove(struct device *dev)
 {
 }
 
+static inline int pm_sleep_lock(void)
+{
+	return 0;
+}
+
+static inline void pm_sleep_unlock(void)
+{
+}
+
 #endif
 
 #ifdef CONFIG_PM
diff --git a/include/linux/device.h b/include/linux/device.h
index 2e15822fe409..cf4ae5c5d193 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -521,6 +521,14 @@ extern struct device *device_create(struct class *cls, struct device *parent,
 				    dev_t devt, const char *fmt, ...)
 				    __attribute__((format(printf,4,5)));
 extern void device_destroy(struct class *cls, dev_t devt);
+#ifdef CONFIG_PM_SLEEP
+extern void destroy_suspended_device(struct class *cls, dev_t devt);
+#else /* !CONFIG_PM_SLEEP */
+static inline void destroy_suspended_device(struct class *cls, dev_t devt)
+{
+	device_destroy(cls, devt);
+}
+#endif /* !CONFIG_PM_SLEEP */
 
 /*
  * Platform "fixup" functions - allow the platform to have their say
-- 
cgit v1.2.3


From 41ca28ab2abd76dc203e2c3a7cd609607cb927c3 Mon Sep 17 00:00:00 2001
From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Date: Mon, 10 Dec 2007 23:03:43 +0300
Subject: kref: add kref_set()

This adds kref_set() to the kref api for future use by people who really
know what they are doing with krefs...

From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kref.h |  1 +
 lib/kref.c           | 15 +++++++++++++--
 2 files changed, 14 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/kref.h b/include/linux/kref.h
index 6fee3539893f..5d185635786e 100644
--- a/include/linux/kref.h
+++ b/include/linux/kref.h
@@ -24,6 +24,7 @@ struct kref {
 	atomic_t refcount;
 };
 
+void kref_set(struct kref *kref, int num);
 void kref_init(struct kref *kref);
 void kref_get(struct kref *kref);
 int kref_put(struct kref *kref, void (*release) (struct kref *kref));
diff --git a/lib/kref.c b/lib/kref.c
index a6dc3ec328e0..9ecd6e865610 100644
--- a/lib/kref.c
+++ b/lib/kref.c
@@ -14,14 +14,24 @@
 #include <linux/kref.h>
 #include <linux/module.h>
 
+/**
+ * kref_set - initialize object and set refcount to requested number.
+ * @kref: object in question.
+ * @num: initial reference counter
+ */
+void kref_set(struct kref *kref, int num)
+{
+	atomic_set(&kref->refcount, num);
+	smp_mb();
+}
+
 /**
  * kref_init - initialize object.
  * @kref: object in question.
  */
 void kref_init(struct kref *kref)
 {
-	atomic_set(&kref->refcount,1);
-	smp_mb();
+	kref_set(kref, 1);
 }
 
 /**
@@ -61,6 +71,7 @@ int kref_put(struct kref *kref, void (*release)(struct kref *kref))
 	return 0;
 }
 
+EXPORT_SYMBOL(kref_set);
 EXPORT_SYMBOL(kref_init);
 EXPORT_SYMBOL(kref_get);
 EXPORT_SYMBOL(kref_put);
-- 
cgit v1.2.3


From 891f78ea833edd4a1e524e15bfe297a7a84d81a0 Mon Sep 17 00:00:00 2001
From: Tony Jones <tonyj@suse.de>
Date: Tue, 25 Sep 2007 02:03:03 +0200
Subject: DMA: Convert from class_device to device for DMA engine

Signed-off-by: Tony Jones <tonyj@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Cc: Shannon Nelson <shannon.nelson@intel.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/dma/dmaengine.c   | 43 ++++++++++++++++++++++---------------------
 include/linux/dmaengine.h |  3 ++-
 2 files changed, 24 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index d59b2f417306..bcf52df30339 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -41,12 +41,12 @@
  * the definition of dma_event_callback in dmaengine.h.
  *
  * Each device has a kref, which is initialized to 1 when the device is
- * registered. A kref_get is done for each class_device registered.  When the
- * class_device is released, the coresponding kref_put is done in the release
+ * registered. A kref_get is done for each device registered.  When the
+ * device is released, the coresponding kref_put is done in the release
  * method. Every time one of the device's channels is allocated to a client,
  * a kref_get occurs.  When the channel is freed, the coresponding kref_put
  * happens. The device's release function does a completion, so
- * unregister_device does a remove event, class_device_unregister, a kref_put
+ * unregister_device does a remove event, device_unregister, a kref_put
  * for the first reference, then waits on the completion for all other
  * references to finish.
  *
@@ -77,9 +77,9 @@ static LIST_HEAD(dma_client_list);
 
 /* --- sysfs implementation --- */
 
-static ssize_t show_memcpy_count(struct class_device *cd, char *buf)
+static ssize_t show_memcpy_count(struct device *dev, struct device_attribute *attr, char *buf)
 {
-	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	struct dma_chan *chan = to_dma_chan(dev);
 	unsigned long count = 0;
 	int i;
 
@@ -89,9 +89,10 @@ static ssize_t show_memcpy_count(struct class_device *cd, char *buf)
 	return sprintf(buf, "%lu\n", count);
 }
 
-static ssize_t show_bytes_transferred(struct class_device *cd, char *buf)
+static ssize_t show_bytes_transferred(struct device *dev, struct device_attribute *attr,
+				      char *buf)
 {
-	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	struct dma_chan *chan = to_dma_chan(dev);
 	unsigned long count = 0;
 	int i;
 
@@ -101,9 +102,9 @@ static ssize_t show_bytes_transferred(struct class_device *cd, char *buf)
 	return sprintf(buf, "%lu\n", count);
 }
 
-static ssize_t show_in_use(struct class_device *cd, char *buf)
+static ssize_t show_in_use(struct device *dev, struct device_attribute *attr, char *buf)
 {
-	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	struct dma_chan *chan = to_dma_chan(dev);
 	int in_use = 0;
 
 	if (unlikely(chan->slow_ref) &&
@@ -119,7 +120,7 @@ static ssize_t show_in_use(struct class_device *cd, char *buf)
 	return sprintf(buf, "%d\n", in_use);
 }
 
-static struct class_device_attribute dma_class_attrs[] = {
+static struct device_attribute dma_attrs[] = {
 	__ATTR(memcpy_count, S_IRUGO, show_memcpy_count, NULL),
 	__ATTR(bytes_transferred, S_IRUGO, show_bytes_transferred, NULL),
 	__ATTR(in_use, S_IRUGO, show_in_use, NULL),
@@ -128,16 +129,16 @@ static struct class_device_attribute dma_class_attrs[] = {
 
 static void dma_async_device_cleanup(struct kref *kref);
 
-static void dma_class_dev_release(struct class_device *cd)
+static void dma_dev_release(struct device *dev)
 {
-	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	struct dma_chan *chan = to_dma_chan(dev);
 	kref_put(&chan->device->refcount, dma_async_device_cleanup);
 }
 
 static struct class dma_devclass = {
-	.name            = "dma",
-	.class_dev_attrs = dma_class_attrs,
-	.release = dma_class_dev_release,
+	.name		= "dma",
+	.dev_attrs	= dma_attrs,
+	.dev_release	= dma_dev_release,
 };
 
 /* --- client and device registration --- */
@@ -377,12 +378,12 @@ int dma_async_device_register(struct dma_device *device)
 			continue;
 
 		chan->chan_id = chancnt++;
-		chan->class_dev.class = &dma_devclass;
-		chan->class_dev.dev = NULL;
-		snprintf(chan->class_dev.class_id, BUS_ID_SIZE, "dma%dchan%d",
+		chan->dev.class = &dma_devclass;
+		chan->dev.parent = NULL;
+		snprintf(chan->dev.bus_id, BUS_ID_SIZE, "dma%dchan%d",
 		         device->dev_id, chan->chan_id);
 
-		rc = class_device_register(&chan->class_dev);
+		rc = device_register(&chan->dev);
 		if (rc) {
 			chancnt--;
 			free_percpu(chan->local);
@@ -411,7 +412,7 @@ err_out:
 		if (chan->local == NULL)
 			continue;
 		kref_put(&device->refcount, dma_async_device_cleanup);
-		class_device_unregister(&chan->class_dev);
+		device_unregister(&chan->dev);
 		chancnt--;
 		free_percpu(chan->local);
 	}
@@ -445,7 +446,7 @@ void dma_async_device_unregister(struct dma_device *device)
 
 	list_for_each_entry(chan, &device->channels, device_node) {
 		dma_clients_notify_removed(chan);
-		class_device_unregister(&chan->class_dev);
+		device_unregister(&chan->dev);
 		dma_chan_release(chan);
 	}
 
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index a3b6035b6c86..55c9a6952f44 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -132,7 +132,7 @@ struct dma_chan {
 
 	/* sysfs */
 	int chan_id;
-	struct class_device class_dev;
+	struct device dev;
 
 	struct kref refcount;
 	int slow_ref;
@@ -142,6 +142,7 @@ struct dma_chan {
 	struct dma_chan_percpu *local;
 };
 
+#define to_dma_chan(p) container_of(p, struct dma_chan, dev)
 
 void dma_chan_cleanup(struct kref *kref);
 
-- 
cgit v1.2.3


From 6013c12be8313b3205b41912d965b03f3b06147d Mon Sep 17 00:00:00 2001
From: Tony Jones <tonyj@suse.de>
Date: Tue, 25 Sep 2007 02:03:03 +0200
Subject: pktcdvd: Convert from class_device to device for block/pktcdvd

struct class_device is going away, this converts the code to use struct
device instead.

Signed-off-by: Tony Jones <tonyj@suse.de>
Cc: Peter Osterlund <petero2@telia.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/block/pktcdvd.c | 16 +++++++---------
 include/linux/pktcdvd.h |  2 +-
 2 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 3535ef896677..17da6999bef0 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -301,18 +301,16 @@ static struct kobj_type kobj_pkt_type_wqueue = {
 static void pkt_sysfs_dev_new(struct pktcdvd_device *pd)
 {
 	if (class_pktcdvd) {
-		pd->clsdev = class_device_create(class_pktcdvd,
-					NULL, pd->pkt_dev,
-					NULL, "%s", pd->name);
-		if (IS_ERR(pd->clsdev))
-			pd->clsdev = NULL;
+		pd->dev = device_create(class_pktcdvd, NULL, pd->pkt_dev, "%s", pd->name);
+		if (IS_ERR(pd->dev))
+			pd->dev = NULL;
 	}
-	if (pd->clsdev) {
+	if (pd->dev) {
 		pd->kobj_stat = pkt_kobj_create(pd, "stat",
-					&pd->clsdev->kobj,
+					&pd->dev->kobj,
 					&kobj_pkt_type_stat);
 		pd->kobj_wqueue = pkt_kobj_create(pd, "write_queue",
-					&pd->clsdev->kobj,
+					&pd->dev->kobj,
 					&kobj_pkt_type_wqueue);
 	}
 }
@@ -322,7 +320,7 @@ static void pkt_sysfs_dev_remove(struct pktcdvd_device *pd)
 	pkt_kobj_remove(pd->kobj_stat);
 	pkt_kobj_remove(pd->kobj_wqueue);
 	if (class_pktcdvd)
-		class_device_destroy(class_pktcdvd, pd->pkt_dev);
+		device_destroy(class_pktcdvd, pd->pkt_dev);
 }
 
 
diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h
index 5ea4f05683f6..04b4d7330e6d 100644
--- a/include/linux/pktcdvd.h
+++ b/include/linux/pktcdvd.h
@@ -290,7 +290,7 @@ struct pktcdvd_device
 	int			write_congestion_off;
 	int			write_congestion_on;
 
-	struct class_device	*clsdev;	/* sysfs pktcdvd[0-7] class dev */
+	struct device		*dev;		/* sysfs pktcdvd[0-7] dev */
 	struct pktcdvd_kobj	*kobj_stat;	/* sysfs pktcdvd[0-7]/stat/     */
 	struct pktcdvd_kobj	*kobj_wqueue;	/* sysfs pktcdvd[0-7]/write_queue/ */
 
-- 
cgit v1.2.3


From 7dd817d083b6fc103b9ea4f2b4f4a1c6a09e29a0 Mon Sep 17 00:00:00 2001
From: Tony Jones <tonyj@suse.de>
Date: Tue, 25 Sep 2007 02:03:03 +0200
Subject: tifm: Convert from class_device to device for TI flash media

Signed-off-by: Tony Jones <tonyj@suse.de>
Cc: Alex Dubov <oakad@yahoo.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/misc/tifm_7xx1.c |  4 ++--
 drivers/misc/tifm_core.c | 24 ++++++++++++------------
 include/linux/tifm.h     |  2 +-
 3 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/drivers/misc/tifm_7xx1.c b/drivers/misc/tifm_7xx1.c
index 2d1b3df95c5b..54380da343a5 100644
--- a/drivers/misc/tifm_7xx1.c
+++ b/drivers/misc/tifm_7xx1.c
@@ -149,7 +149,7 @@ static void tifm_7xx1_switch_media(struct work_struct *work)
 	socket_change_set = fm->socket_change_set;
 	fm->socket_change_set = 0;
 
-	dev_dbg(fm->cdev.dev, "checking media set %x\n",
+	dev_dbg(fm->dev.parent, "checking media set %x\n",
 		socket_change_set);
 
 	if (!socket_change_set) {
@@ -164,7 +164,7 @@ static void tifm_7xx1_switch_media(struct work_struct *work)
 		if (sock) {
 			printk(KERN_INFO
 			       "%s : demand removing card from socket %u:%u\n",
-			       fm->cdev.class_id, fm->id, cnt);
+			       fm->dev.bus_id, fm->id, cnt);
 			fm->sockets[cnt] = NULL;
 			sock_addr = sock->addr;
 			spin_unlock_irqrestore(&fm->lock, flags);
diff --git a/drivers/misc/tifm_core.c b/drivers/misc/tifm_core.c
index 8f77949f93dd..97544052e768 100644
--- a/drivers/misc/tifm_core.c
+++ b/drivers/misc/tifm_core.c
@@ -160,16 +160,16 @@ static struct bus_type tifm_bus_type = {
 	.resume    = tifm_device_resume
 };
 
-static void tifm_free(struct class_device *cdev)
+static void tifm_free(struct device *dev)
 {
-	struct tifm_adapter *fm = container_of(cdev, struct tifm_adapter, cdev);
+	struct tifm_adapter *fm = container_of(dev, struct tifm_adapter, dev);
 
 	kfree(fm);
 }
 
 static struct class tifm_adapter_class = {
 	.name    = "tifm_adapter",
-	.release = tifm_free
+	.dev_release = tifm_free
 };
 
 struct tifm_adapter *tifm_alloc_adapter(unsigned int num_sockets,
@@ -180,9 +180,9 @@ struct tifm_adapter *tifm_alloc_adapter(unsigned int num_sockets,
 	fm = kzalloc(sizeof(struct tifm_adapter)
 		     + sizeof(struct tifm_dev*) * num_sockets, GFP_KERNEL);
 	if (fm) {
-		fm->cdev.class = &tifm_adapter_class;
-		fm->cdev.dev = dev;
-		class_device_initialize(&fm->cdev);
+		fm->dev.class = &tifm_adapter_class;
+		fm->dev.parent = dev;
+		device_initialize(&fm->dev);
 		spin_lock_init(&fm->lock);
 		fm->num_sockets = num_sockets;
 	}
@@ -203,8 +203,8 @@ int tifm_add_adapter(struct tifm_adapter *fm)
 	if (rc)
 		return rc;
 
-	snprintf(fm->cdev.class_id, BUS_ID_SIZE, "tifm%u", fm->id);
-	rc = class_device_add(&fm->cdev);
+	snprintf(fm->dev.bus_id, BUS_ID_SIZE, "tifm%u", fm->id);
+	rc = device_add(&fm->dev);
 	if (rc) {
 		spin_lock(&tifm_adapter_lock);
 		idr_remove(&tifm_adapter_idr, fm->id);
@@ -228,13 +228,13 @@ void tifm_remove_adapter(struct tifm_adapter *fm)
 	spin_lock(&tifm_adapter_lock);
 	idr_remove(&tifm_adapter_idr, fm->id);
 	spin_unlock(&tifm_adapter_lock);
-	class_device_del(&fm->cdev);
+	device_del(&fm->dev);
 }
 EXPORT_SYMBOL(tifm_remove_adapter);
 
 void tifm_free_adapter(struct tifm_adapter *fm)
 {
-	class_device_put(&fm->cdev);
+	put_device(&fm->dev);
 }
 EXPORT_SYMBOL(tifm_free_adapter);
 
@@ -261,9 +261,9 @@ struct tifm_dev *tifm_alloc_device(struct tifm_adapter *fm, unsigned int id,
 		sock->card_event = tifm_dummy_event;
 		sock->data_event = tifm_dummy_event;
 
-		sock->dev.parent = fm->cdev.dev;
+		sock->dev.parent = fm->dev.parent;
 		sock->dev.bus = &tifm_bus_type;
-		sock->dev.dma_mask = fm->cdev.dev->dma_mask;
+		sock->dev.dma_mask = fm->dev.parent->dma_mask;
 		sock->dev.release = tifm_free_device;
 
 		snprintf(sock->dev.bus_id, BUS_ID_SIZE,
diff --git a/include/linux/tifm.h b/include/linux/tifm.h
index 6b3a31805c72..2096b76d0cee 100644
--- a/include/linux/tifm.h
+++ b/include/linux/tifm.h
@@ -120,7 +120,7 @@ struct tifm_adapter {
 	struct completion   *finish_me;
 
 	struct work_struct  media_switcher;
-	struct class_device cdev;
+	struct device	    dev;
 
 	void                (*eject)(struct tifm_adapter *fm,
 				     struct tifm_dev *sock);
-- 
cgit v1.2.3


From 7b8712e563df4fefc25d3107fa3fb3abb7331ff4 Mon Sep 17 00:00:00 2001
From: Emil Medve <Emilian.Medve@Freescale.com>
Date: Tue, 30 Oct 2007 14:37:14 -0500
Subject: driver core: Make the dev_*() family of macros in device.h complete

Removed duplicates defined elsewhere

Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/chips/isp1301_omap.c |  6 ------
 drivers/isdn/gigaset/gigaset.h   |  6 ------
 include/linux/device.h           | 24 +++++++++++++++---------
 3 files changed, 15 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c
index b767603a07ba..ebfbb2947ae6 100644
--- a/drivers/i2c/chips/isp1301_omap.c
+++ b/drivers/i2c/chips/isp1301_omap.c
@@ -259,12 +259,6 @@ static inline const char *state_name(struct isp1301 *isp)
 	return state_string(isp->otg.state);
 }
 
-#ifdef	VERBOSE
-#define	dev_vdbg			dev_dbg
-#else
-#define	dev_vdbg(dev, fmt, arg...)	do{}while(0)
-#endif
-
 /*-------------------------------------------------------------------------*/
 
 /* NOTE:  some of this ISP1301 setup is specific to H2 boards;
diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h
index a0317abaeb11..02bdaf22d7ea 100644
--- a/drivers/isdn/gigaset/gigaset.h
+++ b/drivers/isdn/gigaset/gigaset.h
@@ -106,12 +106,6 @@ enum debuglevel {
 					 activated */
 };
 
-/* missing from linux/device.h ... */
-#ifndef dev_notice
-#define dev_notice(dev, format, arg...)		\
-	dev_printk(KERN_NOTICE , dev , format , ## arg)
-#endif
-
 /* Kernel message macros for situations where dev_printk and friends cannot be
  * used for lack of reliable access to a device structure.
  * linux/usb.h already contains these but in an obsolete form which clutters
diff --git a/include/linux/device.h b/include/linux/device.h
index cf4ae5c5d193..dbbbe89e7267 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -565,6 +565,21 @@ extern const char *dev_driver_string(struct device *dev);
 #define dev_printk(level, dev, format, arg...)	\
 	printk(level "%s %s: " format , dev_driver_string(dev) , (dev)->bus_id , ## arg)
 
+#define dev_emerg(dev, format, arg...)		\
+	dev_printk(KERN_EMERG , dev , format , ## arg)
+#define dev_alert(dev, format, arg...)		\
+	dev_printk(KERN_ALERT , dev , format , ## arg)
+#define dev_crit(dev, format, arg...)		\
+	dev_printk(KERN_CRIT , dev , format , ## arg)
+#define dev_err(dev, format, arg...)		\
+	dev_printk(KERN_ERR , dev , format , ## arg)
+#define dev_warn(dev, format, arg...)		\
+	dev_printk(KERN_WARNING , dev , format , ## arg)
+#define dev_notice(dev, format, arg...)		\
+	dev_printk(KERN_NOTICE , dev , format , ## arg)
+#define dev_info(dev, format, arg...)		\
+	dev_printk(KERN_INFO , dev , format , ## arg)
+
 #ifdef DEBUG
 #define dev_dbg(dev, format, arg...)		\
 	dev_printk(KERN_DEBUG , dev , format , ## arg)
@@ -586,15 +601,6 @@ dev_vdbg(struct device * dev, const char * fmt, ...)
 }
 #endif
 
-#define dev_err(dev, format, arg...)		\
-	dev_printk(KERN_ERR , dev , format , ## arg)
-#define dev_info(dev, format, arg...)		\
-	dev_printk(KERN_INFO , dev , format , ## arg)
-#define dev_warn(dev, format, arg...)		\
-	dev_printk(KERN_WARNING , dev , format , ## arg)
-#define dev_notice(dev, format, arg...)		\
-	dev_printk(KERN_NOTICE , dev , format , ## arg)
-
 /* Create alias, so I can be autoloaded. */
 #define MODULE_ALIAS_CHARDEV(major,minor) \
 	MODULE_ALIAS("char-major-" __stringify(major) "-" __stringify(minor))
-- 
cgit v1.2.3


From 18041f4775688af073d9b3ab0ffc262c1847e60b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Dec 2007 21:31:08 -0800
Subject: kobject: make kobject_cleanup be static

No one except the kobject core calls it so make the function static.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 2 --
 lib/kobject.c           | 9 ++++-----
 2 files changed, 4 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 4a0d27f475d7..2d19a079ee79 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -79,8 +79,6 @@ static inline const char * kobject_name(const struct kobject * kobj)
 }
 
 extern void kobject_init(struct kobject *);
-extern void kobject_cleanup(struct kobject *);
-
 extern int __must_check kobject_add(struct kobject *);
 extern void kobject_del(struct kobject *);
 
diff --git a/lib/kobject.c b/lib/kobject.c
index 4a310e55a886..a152036db006 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -436,12 +436,11 @@ struct kobject * kobject_get(struct kobject * kobj)
 	return kobj;
 }
 
-/**
- *	kobject_cleanup - free kobject resources. 
- *	@kobj:	object.
+/*
+ * kobject_cleanup - free kobject resources.
+ * @kobj: object to cleanup
  */
-
-void kobject_cleanup(struct kobject * kobj)
+static void kobject_cleanup(struct kobject *kobj)
 {
 	struct kobj_type * t = get_ktype(kobj);
 	struct kset * s = kobj->kset;
-- 
cgit v1.2.3


From e86000d042d23904bbb609af2f8618a541cf129b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Dec 2007 21:31:08 -0800
Subject: kobject: add kobject_init_ng function

This is what the kobject_init function is going to become.

Add this to the kernel and then we can convert the tree over to use it.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  1 +
 lib/kobject.c           | 42 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 2d19a079ee79..bdf4f7c45f19 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -79,6 +79,7 @@ static inline const char * kobject_name(const struct kobject * kobj)
 }
 
 extern void kobject_init(struct kobject *);
+extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
 extern int __must_check kobject_add(struct kobject *);
 extern void kobject_del(struct kobject *);
 
diff --git a/lib/kobject.c b/lib/kobject.c
index a152036db006..60586bcc7a71 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -282,6 +282,48 @@ int kobject_set_name(struct kobject *kobj, const char *fmt, ...)
 }
 EXPORT_SYMBOL(kobject_set_name);
 
+/**
+ * kobject_init_ng - initialize a kobject structure
+ * @kobj: pointer to the kobject to initialize
+ * @ktype: pointer to the ktype for this kobject.
+ *
+ * This function will properly initialize a kobject such that it can then
+ * be passed to the kobject_add() call.
+ *
+ * After this function is called, the kobject MUST be cleaned up by a call
+ * to kobject_put(), not by a call to kfree directly to ensure that all of
+ * the memory is cleaned up properly.
+ */
+void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype)
+{
+	char *err_str;
+
+	if (!kobj) {
+		err_str = "invalid kobject pointer!";
+		goto error;
+	}
+	if (!ktype) {
+		err_str = "must have a ktype to be initialized properly!\n";
+		goto error;
+	}
+	if (atomic_read(&kobj->kref.refcount)) {
+		/* do not error out as sometimes we can recover */
+		printk(KERN_ERR "kobject: reference count is already set, "
+		       "something is seriously wrong.\n");
+		dump_stack();
+	}
+
+	kref_init(&kobj->kref);
+	INIT_LIST_HEAD(&kobj->entry);
+	kobj->ktype = ktype;
+	return;
+
+error:
+	printk(KERN_ERR "kobject: %s\n", err_str);
+	dump_stack();
+}
+EXPORT_SYMBOL(kobject_init_ng);
+
 /**
  *	kobject_rename - change the name of an object
  *	@kobj:	object in question.
-- 
cgit v1.2.3


From 244f6cee9a928103132a722292bfa0eb84114b07 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Dec 2007 21:31:08 -0800
Subject: kobject: add kobject_add_ng function

This is what the kobject_add function is going to become.

Add this to the kernel and then we can convert the tree over to use it.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  3 +++
 lib/kobject.c           | 66 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 69 insertions(+)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index bdf4f7c45f19..57eea4cb940d 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -81,6 +81,9 @@ static inline const char * kobject_name(const struct kobject * kobj)
 extern void kobject_init(struct kobject *);
 extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
 extern int __must_check kobject_add(struct kobject *);
+extern int __must_check kobject_add_ng(struct kobject *kobj,
+				       struct kobject *parent,
+				       const char *fmt, ...);
 extern void kobject_del(struct kobject *);
 
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
diff --git a/lib/kobject.c b/lib/kobject.c
index 60586bcc7a71..329fd1126b3f 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -324,6 +324,72 @@ error:
 }
 EXPORT_SYMBOL(kobject_init_ng);
 
+static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
+			    const char *fmt, va_list vargs)
+{
+	va_list aq;
+	int retval;
+
+	va_copy(aq, vargs);
+	retval = kobject_set_name_vargs(kobj, fmt, aq);
+	va_end(aq);
+	if (retval) {
+		printk(KERN_ERR "kobject: can not set name properly!\n");
+		return retval;
+	}
+	kobj->parent = parent;
+	return kobject_add(kobj);
+}
+
+/**
+ * kobject_add_ng - the main kobject add function
+ * @kobj: the kobject to add
+ * @parent: pointer to the parent of the kobject.
+ * @fmt: format to name the kobject with.
+ *
+ * The kobject name is set and added to the kobject hierarchy in this
+ * function.
+ *
+ * If @parent is set, then the parent of the @kobj will be set to it.
+ * If @parent is NULL, then the parent of the @kobj will be set to the
+ * kobject associted with the kset assigned to this kobject.  If no kset
+ * is assigned to the kobject, then the kobject will be located in the
+ * root of the sysfs tree.
+ *
+ * If this function returns an error, kobject_put() must be called to
+ * properly clean up the memory associated with the object.
+ *
+ * If the function is successful, the only way to properly clean up the
+ * memory is with a call to kobject_del(), in which case, a call to
+ * kobject_put() is not necessary (kobject_del() does the final
+ * kobject_put() to call the release function in the ktype's release
+ * pointer.)
+ *
+ * Under no instance should the kobject that is passed to this function
+ * be directly freed with a call to kfree(), that can leak memory.
+ *
+ * Note, no uevent will be created with this call, the caller should set
+ * up all of the necessary sysfs files for the object and then call
+ * kobject_uevent() with the UEVENT_ADD parameter to ensure that
+ * userspace is properly notified of this kobject's creation.
+ */
+int kobject_add_ng(struct kobject *kobj, struct kobject *parent,
+		   const char *fmt, ...)
+{
+	va_list args;
+	int retval;
+
+	if (!kobj)
+		return -EINVAL;
+
+	va_start(args, fmt);
+	retval = kobject_add_varg(kobj, parent, fmt, args);
+	va_end(args);
+
+	return retval;
+}
+EXPORT_SYMBOL(kobject_add_ng);
+
 /**
  *	kobject_rename - change the name of an object
  *	@kobj:	object in question.
-- 
cgit v1.2.3


From c11c4154e7ff4cebfadad849b1e22689d759c3f4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Dec 2007 21:31:08 -0800
Subject: kobject: add kobject_init_and_add function

Also add a kobject_init_and_add function which bundles up what a lot of
the current callers want to do all at once, and it properly handles the
memory usages, unlike kobject_register();

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  5 +++++
 lib/kobject.c           | 27 +++++++++++++++++++++++++++
 2 files changed, 32 insertions(+)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 57eea4cb940d..e2b8c3dae425 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -84,6 +84,11 @@ extern int __must_check kobject_add(struct kobject *);
 extern int __must_check kobject_add_ng(struct kobject *kobj,
 				       struct kobject *parent,
 				       const char *fmt, ...);
+extern int __must_check kobject_init_and_add(struct kobject *kobj,
+					     struct kobj_type *ktype,
+					     struct kobject *parent,
+					     const char *fmt, ...);
+
 extern void kobject_del(struct kobject *);
 
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
diff --git a/lib/kobject.c b/lib/kobject.c
index 329fd1126b3f..8f249408b2ec 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -390,6 +390,33 @@ int kobject_add_ng(struct kobject *kobj, struct kobject *parent,
 }
 EXPORT_SYMBOL(kobject_add_ng);
 
+/**
+ * kobject_init_and_add - initialize a kobject structure and add it to the kobject hierarchy
+ * @kobj: pointer to the kobject to initialize
+ * @ktype: pointer to the ktype for this kobject.
+ * @parent: pointer to the parent of this kobject.
+ * @fmt: the name of the kobject.
+ *
+ * This function combines the call to kobject_init_ng() and
+ * kobject_add_ng().  The same type of error handling after a call to
+ * kobject_add_ng() and kobject lifetime rules are the same here.
+ */
+int kobject_init_and_add(struct kobject *kobj, struct kobj_type *ktype,
+			 struct kobject *parent, const char *fmt, ...)
+{
+	va_list args;
+	int retval;
+
+	kobject_init_ng(kobj, ktype);
+
+	va_start(args, fmt);
+	retval = kobject_add_varg(kobj, parent, fmt, args);
+	va_end(args);
+
+	return retval;
+}
+EXPORT_SYMBOL_GPL(kobject_init_and_add);
+
 /**
  *	kobject_rename - change the name of an object
  *	@kobj:	object in question.
-- 
cgit v1.2.3


From 3514faca19a6fdc209734431c509631ea92b094e Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Oct 2007 10:11:44 -0600
Subject: kobject: remove struct kobj_type from struct kset

We don't need a "default" ktype for a kset.  We should set this
explicitly every time for each kset.  This change is needed so that we
can make ksets dynamic, and cleans up one of the odd, undocumented
assumption that the kset/kobject/ktype model has.

This patch is based on a lot of help from Kay Sievers.

Nasty bug in the block code was found by Dave Young
<hidave.darkstar@gmail.com>

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Dave Young <hidave.darkstar@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/powerpc/platforms/pseries/power.c |  2 +-
 arch/s390/hypfs/inode.c                |  4 ++--
 arch/s390/kernel/ipl.c                 |  8 ++++----
 block/genhd.c                          |  5 +++--
 drivers/acpi/bus.c                     |  2 +-
 drivers/base/bus.c                     |  5 +++--
 drivers/base/class.c                   |  8 +++++---
 drivers/base/core.c                    |  5 +++--
 drivers/base/firmware.c                |  5 +++--
 drivers/base/hypervisor.c              |  2 +-
 drivers/base/sys.c                     |  3 ++-
 drivers/edac/edac_mc_sysfs.c           |  2 +-
 drivers/firmware/edd.c                 |  5 +++--
 drivers/firmware/efivars.c             |  9 +++++----
 drivers/parisc/pdc_stable.c            |  9 +++++----
 drivers/pci/hotplug/pci_hotplug_core.c |  7 ++++---
 drivers/pci/hotplug/rpadlpar_sysfs.c   |  1 -
 drivers/uio/uio.c                      |  2 +-
 fs/configfs/mount.c                    |  4 ++--
 fs/debugfs/inode.c                     |  4 ++--
 fs/dlm/lockspace.c                     |  6 ++----
 fs/ecryptfs/main.c                     |  4 ++--
 fs/fuse/inode.c                        |  8 ++++----
 fs/gfs2/locking/dlm/sysfs.c            |  6 ++----
 fs/gfs2/sys.c                          |  6 ++----
 fs/namespace.c                         |  2 +-
 fs/ocfs2/cluster/masklog.c             |  2 +-
 fs/ocfs2/cluster/sys.c                 |  2 +-
 fs/sysfs/file.c                        |  4 +---
 include/linux/kobject.h                | 15 ++++-----------
 kernel/ksysfs.c                        |  2 +-
 kernel/module.c                        |  2 +-
 kernel/params.c                        |  9 +++++----
 kernel/power/main.c                    |  2 +-
 mm/slub.c                              |  5 +++--
 security/inode.c                       |  4 ++--
 36 files changed, 84 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/pseries/power.c b/arch/powerpc/platforms/pseries/power.c
index 73e69023d90a..08d7a5007167 100644
--- a/arch/powerpc/platforms/pseries/power.c
+++ b/arch/powerpc/platforms/pseries/power.c
@@ -57,7 +57,7 @@ static struct subsys_attribute auto_poweron_attr = {
 };
 
 #ifndef CONFIG_PM
-decl_subsys(power,NULL,NULL);
+decl_subsys(power, NULL);
 
 static struct attribute *g[] = {
         &auto_poweron_attr.attr,
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index 5245717295b8..c022ccc04d41 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -490,7 +490,7 @@ static struct super_operations hypfs_s_ops = {
 	.show_options	= hypfs_show_options,
 };
 
-static decl_subsys(s390, NULL, NULL);
+static decl_subsys(s390, NULL);
 
 static int __init hypfs_init(void)
 {
@@ -506,7 +506,7 @@ static int __init hypfs_init(void)
 			goto fail_diag;
 		}
 	}
-	kobj_set_kset_s(&s390_subsys, hypervisor_subsys);
+	s390_subsys.kobj.kset = &hypervisor_subsys;
 	rc = subsystem_register(&s390_subsys);
 	if (rc)
 		goto fail_sysfs;
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index ce0856d32500..cae793af5423 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -418,7 +418,7 @@ static struct attribute_group ipl_unknown_attr_group = {
 	.attrs = ipl_unknown_attrs,
 };
 
-static decl_subsys(ipl, NULL, NULL);
+static decl_subsys(ipl, NULL);
 
 /*
  * reipl section
@@ -590,7 +590,7 @@ static ssize_t reipl_type_store(struct kset *kset, const char *buf,
 static struct subsys_attribute reipl_type_attr =
 		__ATTR(reipl_type, 0644, reipl_type_show, reipl_type_store);
 
-static decl_subsys(reipl, NULL, NULL);
+static decl_subsys(reipl, NULL);
 
 /*
  * dump section
@@ -685,13 +685,13 @@ static ssize_t dump_type_store(struct kset *kset, const char *buf,
 static struct subsys_attribute dump_type_attr =
 		__ATTR(dump_type, 0644, dump_type_show, dump_type_store);
 
-static decl_subsys(dump, NULL, NULL);
+static decl_subsys(dump, NULL);
 
 /*
  * Shutdown actions section
  */
 
-static decl_subsys(shutdown_actions, NULL, NULL);
+static decl_subsys(shutdown_actions, NULL);
 
 /* on panic */
 
diff --git a/block/genhd.c b/block/genhd.c
index f2ac914160d1..32227b7ecd17 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -584,7 +584,7 @@ static struct kset_uevent_ops block_uevent_ops = {
 	.uevent		= block_uevent,
 };
 
-decl_subsys(block, &ktype_block, &block_uevent_ops);
+decl_subsys(block, &block_uevent_ops);
 
 /*
  * aggregate disk stat collector.  Uses the same stats that the sysfs
@@ -721,7 +721,8 @@ struct gendisk *alloc_disk_node(int minors, int node_id)
 			}
 		}
 		disk->minors = minors;
-		kobj_set_kset_s(disk,block_subsys);
+		disk->kobj.kset = &block_subsys;
+		disk->kobj.ktype = &ktype_block;
 		kobject_init(&disk->kobj);
 		rand_initialize_disk(disk);
 		INIT_WORK(&disk->async_notify,
diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index f4487c38d9f2..7c172d9d7acf 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -743,7 +743,7 @@ static int __init acpi_bus_init(void)
 	return -ENODEV;
 }
 
-decl_subsys(acpi, NULL, NULL);
+decl_subsys(acpi, NULL);
 
 static int __init acpi_init(void)
 {
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 9a19b071c573..630956037e18 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -166,7 +166,7 @@ static struct kset_uevent_ops bus_uevent_ops = {
 	.filter = bus_uevent_filter,
 };
 
-static decl_subsys(bus, &bus_ktype, &bus_uevent_ops);
+static decl_subsys(bus, &bus_uevent_ops);
 
 
 #ifdef CONFIG_HOTPLUG
@@ -639,6 +639,7 @@ int bus_add_driver(struct device_driver *drv)
 	if (error)
 		goto out_put_bus;
 	drv->kobj.kset = &bus->drivers;
+	drv->kobj.ktype = &driver_ktype;
 	error = kobject_register(&drv->kobj);
 	if (error)
 		goto out_put_bus;
@@ -851,6 +852,7 @@ int bus_register(struct bus_type * bus)
 		goto out;
 
 	bus->subsys.kobj.kset = &bus_subsys;
+	bus->subsys.kobj.ktype = &bus_ktype;
 
 	retval = subsystem_register(&bus->subsys);
 	if (retval)
@@ -868,7 +870,6 @@ int bus_register(struct bus_type * bus)
 
 	kobject_set_name(&bus->drivers.kobj, "drivers");
 	bus->drivers.kobj.parent = &bus->subsys.kobj;
-	bus->drivers.ktype = &driver_ktype;
 	retval = kset_register(&bus->drivers);
 	if (retval)
 		goto bus_drivers_fail;
diff --git a/drivers/base/class.c b/drivers/base/class.c
index a863bb091e11..8ad98924cddb 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -71,7 +71,7 @@ static struct kobj_type class_ktype = {
 };
 
 /* Hotplug events for classes go to the class_obj subsys */
-static decl_subsys(class, &class_ktype, NULL);
+static decl_subsys(class, NULL);
 
 
 int class_create_file(struct class * cls, const struct class_attribute * attr)
@@ -150,6 +150,7 @@ int class_register(struct class * cls)
 		return error;
 
 	cls->subsys.kobj.kset = &class_subsys;
+	cls->subsys.kobj.ktype = &class_ktype;
 
 	error = subsystem_register(&cls->subsys);
 	if (!error) {
@@ -452,7 +453,7 @@ static struct kset_uevent_ops class_uevent_ops = {
 	.uevent =	class_uevent,
 };
 
-static decl_subsys(class_obj, &class_device_ktype, &class_uevent_ops);
+static decl_subsys(class_obj, &class_uevent_ops);
 
 
 static int class_device_add_attrs(struct class_device * cd)
@@ -537,7 +538,8 @@ static struct class_device_attribute class_uevent_attr =
 
 void class_device_initialize(struct class_device *class_dev)
 {
-	kobj_set_kset_s(class_dev, class_obj_subsys);
+	class_dev->kobj.kset = &class_obj_subsys;
+	class_dev->kobj.ktype = &class_device_ktype;
 	kobject_init(&class_dev->kobj);
 	INIT_LIST_HEAD(&class_dev->node);
 }
diff --git a/drivers/base/core.c b/drivers/base/core.c
index ce6b64c489ad..c8f2ac03d46d 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -405,7 +405,7 @@ static struct device_attribute devt_attr =
  *	devices_subsys - structure to be registered with kobject core.
  */
 
-decl_subsys(devices, &device_ktype, &device_uevent_ops);
+decl_subsys(devices, &device_uevent_ops);
 
 
 /**
@@ -525,7 +525,8 @@ static void klist_children_put(struct klist_node *n)
 
 void device_initialize(struct device *dev)
 {
-	kobj_set_kset_s(dev, devices_subsys);
+	dev->kobj.kset = &devices_subsys;
+	dev->kobj.ktype = &device_ktype;
 	kobject_init(&dev->kobj);
 	klist_init(&dev->klist_children, klist_children_get,
 		   klist_children_put);
diff --git a/drivers/base/firmware.c b/drivers/base/firmware.c
index 90c862932169..336be0450d54 100644
--- a/drivers/base/firmware.c
+++ b/drivers/base/firmware.c
@@ -15,11 +15,12 @@
 
 #include "base.h"
 
-static decl_subsys(firmware, NULL, NULL);
+static decl_subsys(firmware, NULL);
 
 int firmware_register(struct kset *s)
 {
-	kobj_set_kset_s(s, firmware_subsys);
+	s->kobj.kset = &firmware_subsys;
+	s->kobj.ktype = NULL;
 	return subsystem_register(s);
 }
 
diff --git a/drivers/base/hypervisor.c b/drivers/base/hypervisor.c
index 7080b413ddc9..14e75e9ec783 100644
--- a/drivers/base/hypervisor.c
+++ b/drivers/base/hypervisor.c
@@ -11,7 +11,7 @@
 
 #include "base.h"
 
-decl_subsys(hypervisor, NULL, NULL);
+decl_subsys(hypervisor, NULL);
 EXPORT_SYMBOL_GPL(hypervisor_subsys);
 
 int __init hypervisor_init(void)
diff --git a/drivers/base/sys.c b/drivers/base/sys.c
index ac7ff6d0c6e5..7cf19fc318da 100644
--- a/drivers/base/sys.c
+++ b/drivers/base/sys.c
@@ -131,7 +131,7 @@ EXPORT_SYMBOL_GPL(sysdev_class_remove_file);
 /*
  * declare system_subsys
  */
-static decl_subsys(system, &ktype_sysdev_class, NULL);
+static decl_subsys(system, NULL);
 
 int sysdev_class_register(struct sysdev_class * cls)
 {
@@ -139,6 +139,7 @@ int sysdev_class_register(struct sysdev_class * cls)
 		 kobject_name(&cls->kset.kobj));
 	INIT_LIST_HEAD(&cls->drivers);
 	cls->kset.kobj.parent = &system_subsys.kobj;
+	cls->kset.kobj.ktype = &ktype_sysdev_class;
 	cls->kset.kobj.kset = &system_subsys;
 	return kset_register(&cls->kset);
 }
diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 3706b2bc0987..905fcd73c26e 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -744,7 +744,6 @@ static struct kobj_type ktype_mc_set_attribs = {
  */
 static struct kset mc_kset = {
 	.kobj = {.ktype = &ktype_mc_set_attribs },
-	.ktype = &ktype_mci,
 };
 
 
@@ -767,6 +766,7 @@ int edac_mc_register_sysfs_main_kobj(struct mem_ctl_info *mci)
 
 	/* this instance become part of the mc_kset */
 	kobj_mci->kset = &mc_kset;
+	kobj_mci->ktype = &ktype_mci;
 
 	/* set the name of the mc<id> object */
 	err = kobject_set_name(kobj_mci, "mc%d", mci->mc_idx);
diff --git a/drivers/firmware/edd.c b/drivers/firmware/edd.c
index 6942e065e609..fc567fad3f7c 100644
--- a/drivers/firmware/edd.c
+++ b/drivers/firmware/edd.c
@@ -631,7 +631,7 @@ static struct kobj_type edd_ktype = {
 	.default_attrs	= def_attrs,
 };
 
-static decl_subsys(edd, &edd_ktype, NULL);
+static decl_subsys(edd, NULL);
 
 
 /**
@@ -723,7 +723,8 @@ edd_device_register(struct edd_device *edev, int i)
 	edd_dev_set_info(edev, i);
 	kobject_set_name(&edev->kobj, "int13_dev%02x",
 			 0x80 + i);
-	kobj_set_kset_s(edev,edd_subsys);
+	edev->kobj.kset = &edd_subsys;
+	edev->kobj.ktype = &edd_ktype;
 	error = kobject_register(&edev->kobj);
 	if (!error)
 		edd_populate_dir(edev);
diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c
index 858a7b95933b..06ecdb9f6013 100644
--- a/drivers/firmware/efivars.c
+++ b/drivers/firmware/efivars.c
@@ -583,8 +583,8 @@ static struct subsys_attribute *efi_subsys_attrs[] = {
 	NULL,	/* maybe more in the future? */
 };
 
-static decl_subsys(vars, &efivar_ktype, NULL);
-static decl_subsys(efi, NULL, NULL);
+static decl_subsys(vars, NULL);
+static decl_subsys(efi, NULL);
 
 /*
  * efivar_create_sysfs_entry()
@@ -629,7 +629,8 @@ efivar_create_sysfs_entry(unsigned long variable_name_size,
 	efi_guid_unparse(vendor_guid, short_name + strlen(short_name));
 
 	kobject_set_name(&new_efivar->kobj, "%s", short_name);
-	kobj_set_kset_s(new_efivar, vars_subsys);
+	new_efivar->kobj.kset = &vars_subsys;
+	new_efivar->kobj.ktype = &efivar_ktype;
 	i = kobject_register(&new_efivar->kobj);
 	if (i) {
 		kfree(short_name);
@@ -687,7 +688,7 @@ efivars_init(void)
 		goto out_free;
 	}
 
-	kobj_set_kset_s(&vars_subsys, efi_subsys);
+	vars_subsys.kobj.kset = &efi_subsys;
 
 	error = subsystem_register(&vars_subsys);
 
diff --git a/drivers/parisc/pdc_stable.c b/drivers/parisc/pdc_stable.c
index ebb09e98d215..1382be64cc3f 100644
--- a/drivers/parisc/pdc_stable.c
+++ b/drivers/parisc/pdc_stable.c
@@ -964,8 +964,8 @@ static struct subsys_attribute *pdcs_subsys_attrs[] = {
 	NULL,
 };
 
-static decl_subsys(paths, &ktype_pdcspath, NULL);
-static decl_subsys(stable, NULL, NULL);
+static decl_subsys(paths, NULL);
+static decl_subsys(stable, NULL);
 
 /**
  * pdcs_register_pathentries - Prepares path entries kobjects for sysfs usage.
@@ -997,7 +997,8 @@ pdcs_register_pathentries(void)
 
 		if ((err = kobject_set_name(&entry->kobj, "%s", entry->name)))
 			return err;
-		kobj_set_kset_s(entry, paths_subsys);
+		entry->kobj.kset = &paths_subsys;
+		entry->kobj.ktype = &ktype_pdcspath;
 		if ((err = kobject_register(&entry->kobj)))
 			return err;
 		
@@ -1072,7 +1073,7 @@ pdc_stable_init(void)
 			error = subsys_create_file(&stable_subsys, attr);
 	
 	/* register the paths subsys as a subsystem of stable subsys */
-	kobj_set_kset_s(&paths_subsys, stable_subsys);
+	paths_subsys.kobj.kset = &stable_subsys;
 	if ((rc = subsystem_register(&paths_subsys)))
 		goto fail_subsysreg;
 
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index 01c351c176ac..ce1cff0fdeca 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -96,7 +96,7 @@ static struct kobj_type hotplug_slot_ktype = {
 	.release = &hotplug_slot_release,
 };
 
-decl_subsys_name(pci_hotplug_slots, slots, &hotplug_slot_ktype, NULL);
+decl_subsys_name(pci_hotplug_slots, slots, NULL);
 
 /* these strings match up with the values in pci_bus_speed */
 static char *pci_bus_speed_strings[] = {
@@ -633,7 +633,8 @@ int pci_hp_register (struct hotplug_slot *slot)
 	}
 
 	kobject_set_name(&slot->kobj, "%s", slot->name);
-	kobj_set_kset_s(slot, pci_hotplug_slots_subsys);
+	slot->kobj.kset = &pci_hotplug_slots_subsys;
+	slot->kobj.ktype = &hotplug_slot_ktype;
 
 	/* this can fail if we have already registered a slot with the same name */
 	if (kobject_register(&slot->kobj)) {
@@ -701,7 +702,7 @@ static int __init pci_hotplug_init (void)
 {
 	int result;
 
-	kobj_set_kset_s(&pci_hotplug_slots_subsys, pci_bus_type.subsys);
+	pci_hotplug_slots_subsys.kobj.kset = &pci_bus_type.subsys;
 	result = subsystem_register(&pci_hotplug_slots_subsys);
 	if (result) {
 		err("Register subsys with error %d\n", result);
diff --git a/drivers/pci/hotplug/rpadlpar_sysfs.c b/drivers/pci/hotplug/rpadlpar_sysfs.c
index a080fedf0332..76090937c758 100644
--- a/drivers/pci/hotplug/rpadlpar_sysfs.c
+++ b/drivers/pci/hotplug/rpadlpar_sysfs.c
@@ -131,7 +131,6 @@ struct kobj_type ktype_dlpar_io = {
 struct kset dlpar_io_kset = {
 	.kobj = {.ktype = &ktype_dlpar_io,
 		 .parent = &pci_hotplug_slots_subsys.kobj},
-	.ktype = &ktype_dlpar_io,
 };
 
 int dlpar_sysfs_init(void)
diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 865f32b63b5c..606aae7490ab 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -160,7 +160,7 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 		if (!map_found) {
 			map_found = 1;
 			kobject_set_name(&idev->map_attr_kset.kobj,"maps");
-			idev->map_attr_kset.ktype = &map_attr_type;
+			idev->map_attr_kset.kobj.ktype = &map_attr_type;
 			idev->map_attr_kset.kobj.parent = &idev->dev->kobj;
 			ret = kset_register(&idev->map_attr_kset);
 			if (ret)
diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index 3bf0278ea843..374ddbd6648d 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -128,7 +128,7 @@ void configfs_release_fs(void)
 }
 
 
-static decl_subsys(config, NULL, NULL);
+static decl_subsys(config, NULL);
 
 static int __init configfs_init(void)
 {
@@ -140,7 +140,7 @@ static int __init configfs_init(void)
 	if (!configfs_dir_cachep)
 		goto out;
 
-	kobj_set_kset_s(&config_subsys, kernel_subsys);
+	config_subsys.kobj.kset = &kernel_subsys;
 	err = subsystem_register(&config_subsys);
 	if (err) {
 		kmem_cache_destroy(configfs_dir_cachep);
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 6a713b33992f..f7f13516fc1a 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -426,13 +426,13 @@ exit:
 }
 EXPORT_SYMBOL_GPL(debugfs_rename);
 
-static decl_subsys(debug, NULL, NULL);
+static decl_subsys(debug, NULL);
 
 static int __init debugfs_init(void)
 {
 	int retval;
 
-	kobj_set_kset_s(&debug_subsys, kernel_subsys);
+	debug_subsys.kobj.kset = &kernel_subsys;
 	retval = subsystem_register(&debug_subsys);
 	if (retval)
 		return retval;
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 6353a8384520..18e4a17b9bee 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -166,9 +166,7 @@ static struct kobj_type dlm_ktype = {
 	.release       = lockspace_kobj_release,
 };
 
-static struct kset dlm_kset = {
-	.ktype  = &dlm_ktype,
-};
+static struct kset dlm_kset;
 
 static int kobject_setup(struct dlm_ls *ls)
 {
@@ -228,7 +226,7 @@ int dlm_lockspace_init(void)
 	spin_lock_init(&lslist_lock);
 
 	kobject_set_name(&dlm_kset.kobj, "dlm");
-	kobj_set_kset_s(&dlm_kset, kernel_subsys);
+	dlm_kset.kobj.kset = &kernel_subsys;
 	error = kset_register(&dlm_kset);
 	if (error)
 		printk("dlm_lockspace_init: cannot register kset %d\n", error);
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index f9f32472c505..fe2f44fa17cc 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -734,7 +734,7 @@ static int ecryptfs_init_kmem_caches(void)
 	return 0;
 }
 
-static decl_subsys(ecryptfs, NULL, NULL);
+static decl_subsys(ecryptfs, NULL);
 
 static ssize_t version_show(struct kset *kset, char *buff)
 {
@@ -798,6 +798,7 @@ static int do_sysfs_registration(void)
 {
 	int rc;
 
+	ecryptfs_subsys.kobj.kset = &fs_subsys;
 	rc = subsystem_register(&ecryptfs_subsys);
 	if (rc) {
 		printk(KERN_ERR
@@ -845,7 +846,6 @@ static int __init ecryptfs_init(void)
 		printk(KERN_ERR "Failed to register filesystem\n");
 		goto out_free_kmem_caches;
 	}
-	kobj_set_kset_s(&ecryptfs_subsys, fs_subsys);
 	rc = do_sysfs_registration();
 	if (rc) {
 		printk(KERN_ERR "sysfs registration failed\n");
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 84f9f7dfdf5b..f5e4182c482e 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -744,8 +744,8 @@ static inline void unregister_fuseblk(void)
 }
 #endif
 
-static decl_subsys(fuse, NULL, NULL);
-static decl_subsys(connections, NULL, NULL);
+static decl_subsys(fuse, NULL);
+static decl_subsys(connections, NULL);
 
 static void fuse_inode_init_once(struct kmem_cache *cachep, void *foo)
 {
@@ -795,12 +795,12 @@ static int fuse_sysfs_init(void)
 {
 	int err;
 
-	kobj_set_kset_s(&fuse_subsys, fs_subsys);
+	fuse_subsys.kobj.kset = &fs_subsys;
 	err = subsystem_register(&fuse_subsys);
 	if (err)
 		goto out_err;
 
-	kobj_set_kset_s(&connections_subsys, fuse_subsys);
+	connections_subsys.kobj.kset = &fuse_subsys;
 	err = subsystem_register(&connections_subsys);
 	if (err)
 		goto out_fuse_unregister;
diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c
index ae9e6a25fe2b..93e66b22757f 100644
--- a/fs/gfs2/locking/dlm/sysfs.c
+++ b/fs/gfs2/locking/dlm/sysfs.c
@@ -189,9 +189,7 @@ static struct kobj_type gdlm_ktype = {
 	.sysfs_ops     = &gdlm_attr_ops,
 };
 
-static struct kset gdlm_kset = {
-	.ktype  = &gdlm_ktype,
-};
+static struct kset gdlm_kset;
 
 int gdlm_kobject_setup(struct gdlm_ls *ls, struct kobject *fskobj)
 {
@@ -224,7 +222,7 @@ int gdlm_sysfs_init(void)
 	int error;
 
 	kobject_set_name(&gdlm_kset.kobj, "lock_dlm");
-	kobj_set_kset_s(&gdlm_kset, kernel_subsys);
+	gdlm_kset.kobj.kset = &kernel_subsys;
 	error = kset_register(&gdlm_kset);
 	if (error)
 		printk("lock_dlm: cannot register kset %d\n", error);
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 06e0b7768d97..d7fa54443f0c 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -221,9 +221,7 @@ static struct kobj_type gfs2_ktype = {
 	.sysfs_ops     = &gfs2_attr_ops,
 };
 
-static struct kset gfs2_kset = {
-	.ktype  = &gfs2_ktype,
-};
+static struct kset gfs2_kset;
 
 /*
  * display struct lm_lockstruct fields
@@ -551,7 +549,7 @@ int gfs2_sys_init(void)
 	gfs2_sys_margs = NULL;
 	spin_lock_init(&gfs2_sys_margs_lock);
 	kobject_set_name(&gfs2_kset.kobj, "gfs2");
-	kobj_set_kset_s(&gfs2_kset, fs_subsys);
+	gfs2_kset.kobj.kset = &fs_subsys;
 	return kset_register(&gfs2_kset);
 }
 
diff --git a/fs/namespace.c b/fs/namespace.c
index 06083885b21e..a4a3f70e7e26 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -41,7 +41,7 @@ static struct kmem_cache *mnt_cache __read_mostly;
 static struct rw_semaphore namespace_sem;
 
 /* /sys/fs */
-decl_subsys(fs, NULL, NULL);
+decl_subsys(fs, NULL);
 EXPORT_SYMBOL_GPL(fs_subsys);
 
 static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry)
diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c
index a4882c8df945..dead319932b3 100644
--- a/fs/ocfs2/cluster/masklog.c
+++ b/fs/ocfs2/cluster/masklog.c
@@ -157,7 +157,7 @@ int mlog_sys_init(struct kset *o2cb_subsys)
 	mlog_attr_ptrs[i] = NULL;
 
 	kobject_set_name(&mlog_kset.kobj, "logmask");
-	kobj_set_kset_s(&mlog_kset, *o2cb_subsys);
+	mlog_kset.kobj.kset = o2cb_subsys;
 	return kset_register(&mlog_kset);
 }
 
diff --git a/fs/ocfs2/cluster/sys.c b/fs/ocfs2/cluster/sys.c
index 64f6f378fd09..880d0138bb0a 100644
--- a/fs/ocfs2/cluster/sys.c
+++ b/fs/ocfs2/cluster/sys.c
@@ -72,7 +72,7 @@ static struct kobj_type o2cb_subsys_type = {
 };
 
 /* gives us o2cb_subsys */
-static decl_subsys(o2cb, NULL, NULL);
+static decl_subsys(o2cb, NULL);
 
 static ssize_t
 o2cb_show(struct kobject * kobj, struct attribute * attr, char * buffer)
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 09a0611b3364..387a63662793 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -365,9 +365,7 @@ static int sysfs_open_file(struct inode *inode, struct file *file)
 	/* if the kobject has no ktype, then we assume that it is a subsystem
 	 * itself, and use ops for it.
 	 */
-	if (kobj->kset && kobj->kset->ktype)
-		ops = kobj->kset->ktype->sysfs_ops;
-	else if (kobj->ktype)
+	if (kobj->ktype)
 		ops = kobj->ktype->sysfs_ops;
 	else
 		ops = &subsys_sysfs_ops;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index e2b8c3dae425..5031565ab30d 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -135,7 +135,6 @@ struct kset_uevent_ops {
  * define the attribute callbacks and other common events that happen to
  * a kobject.
  *
- * @ktype: the struct kobj_type for this specific kset
  * @list: the list of all kobjects for this kset
  * @list_lock: a lock for iterating over the kobjects
  * @kobj: the embedded kobject for this kset (recursion, isn't it fun...)
@@ -145,7 +144,6 @@ struct kset_uevent_ops {
  * desired.
  */
 struct kset {
-	struct kobj_type	*ktype;
 	struct list_head	list;
 	spinlock_t		list_lock;
 	struct kobject		kobj;
@@ -173,12 +171,9 @@ static inline void kset_put(struct kset * k)
 	kobject_put(&k->kobj);
 }
 
-static inline struct kobj_type * get_ktype(struct kobject * k)
+static inline struct kobj_type *get_ktype(struct kobject *kobj)
 {
-	if (k->kset && k->kset->ktype)
-		return k->kset->ktype;
-	else 
-		return k->ktype;
+	return kobj->ktype;
 }
 
 extern struct kobject * kset_find_obj(struct kset *, const char *);
@@ -191,16 +186,14 @@ extern struct kobject * kset_find_obj(struct kset *, const char *);
 #define set_kset_name(str)	.kset = { .kobj = { .k_name = str } }
 
 
-#define decl_subsys(_name,_type,_uevent_ops) \
+#define decl_subsys(_name,_uevent_ops) \
 struct kset _name##_subsys = { \
 	.kobj = { .k_name = __stringify(_name) }, \
-	.ktype = _type, \
 	.uevent_ops =_uevent_ops, \
 }
-#define decl_subsys_name(_varname,_name,_type,_uevent_ops) \
+#define decl_subsys_name(_varname,_name,_uevent_ops) \
 struct kset _varname##_subsys = { \
 	.kobj = { .k_name = __stringify(_name) }, \
-	.ktype = _type, \
 	.uevent_ops =_uevent_ops, \
 }
 
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index 65daa5373ca6..094e2bc101a8 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -94,7 +94,7 @@ static struct bin_attribute notes_attr = {
 	.read = &notes_read,
 };
 
-decl_subsys(kernel, NULL, NULL);
+decl_subsys(kernel, NULL);
 EXPORT_SYMBOL_GPL(kernel_subsys);
 
 static struct attribute * kernel_attrs[] = {
diff --git a/kernel/module.c b/kernel/module.c
index c2e3e2e98801..68df79738b3b 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1223,7 +1223,7 @@ int mod_sysfs_init(struct module *mod)
 	err = kobject_set_name(&mod->mkobj.kobj, "%s", mod->name);
 	if (err)
 		goto out;
-	kobj_set_kset_s(&mod->mkobj, module_subsys);
+	mod->mkobj.kobj.kset = &module_subsys;
 	mod->mkobj.mod = mod;
 
 	kobject_init(&mod->mkobj.kobj);
diff --git a/kernel/params.c b/kernel/params.c
index 7686417ee00e..9f051824097d 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -30,6 +30,8 @@
 #define DEBUGP(fmt, a...)
 #endif
 
+static struct kobj_type module_ktype;
+
 static inline char dash2underscore(char c)
 {
 	if (c == '-')
@@ -560,7 +562,8 @@ static void __init kernel_param_sysfs_setup(const char *name,
 	BUG_ON(!mk);
 
 	mk->mod = THIS_MODULE;
-	kobj_set_kset_s(mk, module_subsys);
+	mk->kobj.kset = &module_subsys;
+	mk->kobj.ktype = &module_ktype;
 	kobject_set_name(&mk->kobj, name);
 	kobject_init(&mk->kobj);
 	ret = kobject_add(&mk->kobj);
@@ -679,8 +682,6 @@ static struct sysfs_ops module_sysfs_ops = {
 	.store = module_attr_store,
 };
 
-static struct kobj_type module_ktype;
-
 static int uevent_filter(struct kset *kset, struct kobject *kobj)
 {
 	struct kobj_type *ktype = get_ktype(kobj);
@@ -694,7 +695,7 @@ static struct kset_uevent_ops module_uevent_ops = {
 	.filter = uevent_filter,
 };
 
-decl_subsys(module, &module_ktype, &module_uevent_ops);
+decl_subsys(module, &module_uevent_ops);
 int module_sysfs_initialized;
 
 static void module_release(struct kobject *kobj)
diff --git a/kernel/power/main.c b/kernel/power/main.c
index f71c9504a5c5..1ef31c91ce0e 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -276,7 +276,7 @@ EXPORT_SYMBOL(pm_suspend);
 
 #endif /* CONFIG_SUSPEND */
 
-decl_subsys(power,NULL,NULL);
+decl_subsys(power, NULL);
 
 
 /**
diff --git a/mm/slub.c b/mm/slub.c
index 474945ecd89d..40bdf41035e5 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3962,7 +3962,7 @@ static struct kset_uevent_ops slab_uevent_ops = {
 	.filter = uevent_filter,
 };
 
-static decl_subsys(slab, &slab_ktype, &slab_uevent_ops);
+static decl_subsys(slab, &slab_uevent_ops);
 
 #define ID_STR_LENGTH 64
 
@@ -4025,8 +4025,9 @@ static int sysfs_slab_add(struct kmem_cache *s)
 		name = create_unique_id(s);
 	}
 
-	kobj_set_kset_s(s, slab_subsys);
 	kobject_set_name(&s->kobj, name);
+	s->kobj.kset = &slab_subsys;
+	s->kobj.ktype = &slab_ktype;
 	kobject_init(&s->kobj);
 	err = kobject_add(&s->kobj);
 	if (err)
diff --git a/security/inode.c b/security/inode.c
index b28a8acae34d..9e42f5f705b2 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -315,13 +315,13 @@ void securityfs_remove(struct dentry *dentry)
 }
 EXPORT_SYMBOL_GPL(securityfs_remove);
 
-static decl_subsys(security, NULL, NULL);
+static decl_subsys(security, NULL);
 
 static int __init securityfs_init(void)
 {
 	int retval;
 
-	kobj_set_kset_s(&security_subsys, kernel_subsys);
+	security_subsys.kobj.kset = &kernel_subsys;
 	retval = subsystem_register(&security_subsys);
 	if (retval)
 		return retval;
-- 
cgit v1.2.3


From 12d03da7c19366268bdbc9fb0cd08d719c0cc283 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Oct 2007 10:11:44 -0600
Subject: kobject: remove kobj_set_kset_s as no one is using it anymore

What a confusing name for a macro...

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 18 ------------------
 1 file changed, 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 5031565ab30d..0b97b3a53917 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -202,24 +202,6 @@ extern struct kset kernel_subsys;
 /* The global /sys/hypervisor/ subsystem  */
 extern struct kset hypervisor_subsys;
 
-/*
- * Helpers for setting the kset of registered objects.
- * Often, a registered object belongs to a kset embedded in a 
- * subsystem. These do no magic, just make the resulting code
- * easier to follow. 
- */
-
-/**
- *	kobj_set_kset_s(obj,subsys) - set kset for embedded kobject.
- *	@obj:		ptr to some object type.
- *	@subsys:	a subsystem object (not a ptr).
- *
- *	Can be used for any object type with an embedded ->kobj.
- */
-
-#define kobj_set_kset_s(obj,subsys) \
-	(obj)->kobj.kset = &(subsys)
-
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
 
-- 
cgit v1.2.3


From b727c702896f88d2ff6c3e03bd011d7c3dffe3e1 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 27 Sep 2007 14:48:53 -0700
Subject: kset: add kset_create_and_add function

Now ksets can be dynamically created on the fly, no static definitions
are required.  Thanks to Miklos for hints on how to make this work
better for the callers.

And thanks to Kay for finding some stupid bugs in my original version
and pointing out that we need to handle the fact that kobject's can have
a kset as a parent and to handle that properly in kobject_add().

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  4 ++-
 lib/kobject.c           | 92 ++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 94 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 0b97b3a53917..f91aeb745664 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -150,11 +150,13 @@ struct kset {
 	struct kset_uevent_ops	*uevent_ops;
 };
 
-
 extern void kset_init(struct kset * k);
 extern int __must_check kset_add(struct kset * k);
 extern int __must_check kset_register(struct kset * k);
 extern void kset_unregister(struct kset * k);
+extern struct kset * __must_check kset_create_and_add(const char *name,
+						struct kset_uevent_ops *u,
+						struct kobject *parent_kobj);
 
 static inline struct kset * to_kset(struct kobject * kobj)
 {
diff --git a/lib/kobject.c b/lib/kobject.c
index 8f249408b2ec..4fb27ba28807 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -186,8 +186,15 @@ int kobject_add(struct kobject * kobj)
 	if (kobj->kset) {
 		spin_lock(&kobj->kset->list_lock);
 
-		if (!parent)
+		if (!parent) {
 			parent = kobject_get(&kobj->kset->kobj);
+			/*
+			 * If the kset is our parent, get a second
+			 * reference, we drop both the kset and the
+			 * parent ref on cleanup
+			 */
+			kobject_get(parent);
+		}
 
 		list_add_tail(&kobj->entry,&kobj->kset->list);
 		spin_unlock(&kobj->kset->list_lock);
@@ -787,6 +794,89 @@ int subsys_create_file(struct kset *s, struct subsys_attribute *a)
 	return error;
 }
 
+static void kset_release(struct kobject *kobj)
+{
+	struct kset *kset = container_of(kobj, struct kset, kobj);
+	pr_debug("kset %s: now freed\n", kobject_name(kobj));
+	kfree(kset);
+}
+
+static struct kobj_type kset_type = {
+	.release = kset_release,
+};
+
+/**
+ * kset_create - create a struct kset dynamically
+ *
+ * @name: the name for the kset
+ * @uevent_ops: a struct kset_uevent_ops for the kset
+ * @parent_kobj: the parent kobject of this kset, if any.
+ *
+ * This function creates a kset structure dynamically.  This structure can
+ * then be registered with the system and show up in sysfs with a call to
+ * kset_register().  When you are finished with this structure, if
+ * kset_register() has been called, call kset_unregister() and the
+ * structure will be dynamically freed when it is no longer being used.
+ *
+ * If the kset was not able to be created, NULL will be returned.
+ */
+static struct kset *kset_create(const char *name,
+				struct kset_uevent_ops *uevent_ops,
+				struct kobject *parent_kobj)
+{
+	struct kset *kset;
+
+	kset = kzalloc(sizeof(*kset), GFP_KERNEL);
+	if (!kset)
+		return NULL;
+	kobject_set_name(&kset->kobj, name);
+	kset->uevent_ops = uevent_ops;
+	kset->kobj.parent = parent_kobj;
+
+	/*
+	 * The kobject of this kset will have a type of kset_type and belong to
+	 * no kset itself.  That way we can properly free it when it is
+	 * finished being used.
+	 */
+	kset->kobj.ktype = &kset_type;
+	kset->kobj.kset = NULL;
+
+	return kset;
+}
+
+/**
+ * kset_create_and_add - create a struct kset dynamically and add it to sysfs
+ *
+ * @name: the name for the kset
+ * @uevent_ops: a struct kset_uevent_ops for the kset
+ * @parent_kobj: the parent kobject of this kset, if any.
+ *
+ * This function creates a kset structure dynamically and registers it
+ * with sysfs.  When you are finished with this structure, call
+ * kset_unregister() and the structure will be dynamically freed when it
+ * is no longer being used.
+ *
+ * If the kset was not able to be created, NULL will be returned.
+ */
+struct kset *kset_create_and_add(const char *name,
+				 struct kset_uevent_ops *uevent_ops,
+				 struct kobject *parent_kobj)
+{
+	struct kset *kset;
+	int error;
+
+	kset = kset_create(name, uevent_ops, parent_kobj);
+	if (!kset)
+		return NULL;
+	error = kset_register(kset);
+	if (error) {
+		kfree(kset);
+		return NULL;
+	}
+	return kset;
+}
+EXPORT_SYMBOL_GPL(kset_create_and_add);
+
 EXPORT_SYMBOL(kobject_init);
 EXPORT_SYMBOL(kobject_register);
 EXPORT_SYMBOL(kobject_unregister);
-- 
cgit v1.2.3


From 3f9e3ee9dc3605e5c593b5d708494571fb0d3970 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 5 Nov 2007 13:16:15 -0800
Subject: kobject: add kobject_create_and_add function

This lets users create dynamic kobjects much easier.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  3 ++
 lib/kobject.c           | 81 ++++++++++++++++++++++++++++++++++++++-----------
 2 files changed, 66 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index f91aeb745664..33e7a6142a75 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -91,6 +91,9 @@ extern int __must_check kobject_init_and_add(struct kobject *kobj,
 
 extern void kobject_del(struct kobject *);
 
+extern struct kobject * __must_check kobject_create_and_add(const char *name,
+						struct kobject *parent);
+
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 extern int __must_check kobject_move(struct kobject *, struct kobject *);
 
diff --git a/lib/kobject.c b/lib/kobject.c
index 4fb27ba28807..98422a3eeffc 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -619,18 +619,69 @@ void kobject_put(struct kobject * kobj)
 		kref_put(&kobj->kref, kobject_release);
 }
 
-
-static void dir_release(struct kobject *kobj)
+static void dynamic_kobj_release(struct kobject *kobj)
 {
+	pr_debug("%s: freeing %s\n", __FUNCTION__, kobject_name(kobj));
 	kfree(kobj);
 }
 
-static struct kobj_type dir_ktype = {
-	.release	= dir_release,
-	.sysfs_ops	= NULL,
-	.default_attrs	= NULL,
+static struct kobj_type dynamic_kobj_ktype = {
+	.release = dynamic_kobj_release,
 };
 
+/*
+ * kobject_create - create a struct kobject dynamically
+ *
+ * This function creates a kobject structure dynamically and sets it up
+ * to be a "dynamic" kobject with a default release function set up.
+ *
+ * If the kobject was not able to be created, NULL will be returned.
+ */
+static struct kobject *kobject_create(void)
+{
+	struct kobject *kobj;
+
+	kobj = kzalloc(sizeof(*kobj), GFP_KERNEL);
+	if (!kobj)
+		return NULL;
+
+	kobject_init_ng(kobj, &dynamic_kobj_ktype);
+	return kobj;
+}
+
+/**
+ * kobject_create_and_add - create a struct kobject dynamically and register it with sysfs
+ *
+ * @name: the name for the kset
+ * @parent: the parent kobject of this kobject, if any.
+ *
+ * This function creates a kset structure dynamically and registers it
+ * with sysfs.  When you are finished with this structure, call
+ * kobject_unregister() and the structure will be dynamically freed when
+ * it is no longer being used.
+ *
+ * If the kobject was not able to be created, NULL will be returned.
+ */
+struct kobject *kobject_create_and_add(const char *name, struct kobject *parent)
+{
+	struct kobject *kobj;
+	int retval;
+
+	kobj = kobject_create();
+	if (!kobj)
+		return NULL;
+
+	retval = kobject_add_ng(kobj, parent, "%s", name);
+	if (retval) {
+		printk(KERN_WARNING "%s: kobject_add error: %d\n",
+		       __FUNCTION__, retval);
+		kobject_put(kobj);
+		kobj = NULL;
+	}
+	return kobj;
+}
+EXPORT_SYMBOL_GPL(kobject_create_and_add);
+
 /**
  *	kobject_kset_add_dir - add sub directory of object.
  *	@kset:		kset the directory is belongs to.
@@ -645,23 +696,17 @@ struct kobject *kobject_kset_add_dir(struct kset *kset,
 	struct kobject *k;
 	int ret;
 
-	if (!parent)
-		return NULL;
-
-	k = kzalloc(sizeof(*k), GFP_KERNEL);
+	k = kobject_create();
 	if (!k)
 		return NULL;
 
 	k->kset = kset;
-	k->parent = parent;
-	k->ktype = &dir_ktype;
-	kobject_set_name(k, name);
-	ret = kobject_register(k);
+	ret = kobject_add_ng(k, parent, "%s", name);
 	if (ret < 0) {
-		printk(KERN_WARNING "%s: kobject_register error: %d\n",
+		printk(KERN_WARNING "%s: kobject_add error: %d\n",
 			__func__, ret);
-		kobject_del(k);
-		return NULL;
+		kobject_put(k);
+		k = NULL;
 	}
 
 	return k;
@@ -676,7 +721,7 @@ struct kobject *kobject_kset_add_dir(struct kset *kset,
  */
 struct kobject *kobject_add_dir(struct kobject *parent, const char *name)
 {
-	return kobject_kset_add_dir(NULL, parent, name);
+	return kobject_create_and_add(name, parent);
 }
 
 /**
-- 
cgit v1.2.3


From 4ff6abff832fbc6cb1d769f6106c841bc2b09f63 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 5 Nov 2007 22:24:43 -0800
Subject: kobject: get rid of kobject_add_dir

kobject_create_and_add is the same as kobject_add_dir, so drop
kobject_add_dir.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c     |  3 ++-
 fs/partitions/check.c   |  6 +++---
 include/linux/kobject.h |  1 -
 kernel/module.c         |  6 +++---
 lib/kobject.c           | 12 ------------
 5 files changed, 8 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index c8f2ac03d46d..992eba3289bd 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -562,7 +562,8 @@ static struct kobject *virtual_device_parent(struct device *dev)
 	static struct kobject *virtual_dir = NULL;
 
 	if (!virtual_dir)
-		virtual_dir = kobject_add_dir(&devices_subsys.kobj, "virtual");
+		virtual_dir = kobject_create_and_add("virtual",
+						     &devices_subsys.kobj);
 
 	return virtual_dir;
 }
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 722e12e5acc7..69685bb51c62 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -335,7 +335,7 @@ static inline void partition_sysfs_add_subdir(struct hd_struct *p)
 	struct kobject *k;
 
 	k = kobject_get(&p->kobj);
-	p->holder_dir = kobject_add_dir(k, "holders");
+	p->holder_dir = kobject_create_and_add("holders", k);
 	kobject_put(k);
 }
 
@@ -344,8 +344,8 @@ static inline void disk_sysfs_add_subdirs(struct gendisk *disk)
 	struct kobject *k;
 
 	k = kobject_get(&disk->kobj);
-	disk->holder_dir = kobject_add_dir(k, "holders");
-	disk->slave_dir = kobject_add_dir(k, "slaves");
+	disk->holder_dir = kobject_create_and_add("holders", k);
+	disk->slave_dir = kobject_create_and_add("slaves", k);
 	kobject_put(k);
 }
 
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 33e7a6142a75..7b09136fb211 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -105,7 +105,6 @@ extern void kobject_put(struct kobject *);
 
 extern struct kobject *kobject_kset_add_dir(struct kset *kset,
 					    struct kobject *, const char *);
-extern struct kobject *kobject_add_dir(struct kobject *, const char *);
 
 extern char * kobject_get_path(struct kobject *, gfp_t);
 
diff --git a/kernel/module.c b/kernel/module.c
index 68df79738b3b..55142775c581 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1122,7 +1122,7 @@ static void add_notes_attrs(struct module *mod, unsigned int nsect,
 		++loaded;
 	}
 
-	notes_attrs->dir = kobject_add_dir(&mod->mkobj.kobj, "notes");
+	notes_attrs->dir = kobject_create_and_add("notes", &mod->mkobj.kobj);
 	if (!notes_attrs->dir)
 		goto out;
 
@@ -1243,7 +1243,7 @@ int mod_sysfs_setup(struct module *mod,
 	if (err)
 		goto out;
 
-	mod->holders_dir = kobject_add_dir(&mod->mkobj.kobj, "holders");
+	mod->holders_dir = kobject_create_and_add("holders", &mod->mkobj.kobj);
 	if (!mod->holders_dir) {
 		err = -ENOMEM;
 		goto out_unreg;
@@ -2521,7 +2521,7 @@ static void module_create_drivers_dir(struct module_kobject *mk)
 	if (!mk || mk->drivers_dir)
 		return;
 
-	mk->drivers_dir = kobject_add_dir(&mk->kobj, "drivers");
+	mk->drivers_dir = kobject_create_and_add("drivers", &mk->kobj);
 }
 
 void module_add_driver(struct module *mod, struct device_driver *drv)
diff --git a/lib/kobject.c b/lib/kobject.c
index 98422a3eeffc..96b61d9a9284 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -712,18 +712,6 @@ struct kobject *kobject_kset_add_dir(struct kset *kset,
 	return k;
 }
 
-/**
- *	kobject_add_dir - add sub directory of object.
- *	@parent:	object in which a directory is created.
- *	@name:	directory name.
- *
- *	Add a plain directory object as child of given object.
- */
-struct kobject *kobject_add_dir(struct kobject *parent, const char *name)
-{
-	return kobject_create_and_add(name, parent);
-}
-
 /**
  *	kset_init - initialize a kset for use
  *	@k:	kset 
-- 
cgit v1.2.3


From 43968d2f1648f4dc92437dc0363a3e88377445b3 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 5 Nov 2007 22:24:43 -0800
Subject: kobject: get rid of kobject_kset_add_dir

kobject_kset_add_dir is only called in one place so remove it and use
kobject_create() instead.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c     | 16 ++++++++++++++--
 include/linux/kobject.h |  4 +---
 lib/kobject.c           | 37 +++++--------------------------------
 3 files changed, 20 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 992eba3289bd..7762ee86697d 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -571,6 +571,8 @@ static struct kobject *virtual_device_parent(struct device *dev)
 static struct kobject * get_device_parent(struct device *dev,
 					  struct device *parent)
 {
+	int retval;
+
 	if (dev->class) {
 		struct kobject *kobj = NULL;
 		struct kobject *parent_kobj;
@@ -600,8 +602,18 @@ static struct kobject * get_device_parent(struct device *dev,
 			return kobj;
 
 		/* or create a new class-directory at the parent device */
-		return kobject_kset_add_dir(&dev->class->class_dirs,
-					    parent_kobj, dev->class->name);
+		k = kobject_create();
+		if (!k)
+			return NULL;
+		k->kset = &dev->class->class_dirs;
+		retval = kobject_add_ng(k, parent_kobj, "%s", dev->class->name);
+		if (retval < 0) {
+			kobject_put(k);
+			return NULL;
+		}
+		/* Do not emit a uevent, as it's not needed for this
+		 * "class glue" directory. */
+		return k;
 	}
 
 	if (parent)
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 7b09136fb211..718b48811288 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -91,6 +91,7 @@ extern int __must_check kobject_init_and_add(struct kobject *kobj,
 
 extern void kobject_del(struct kobject *);
 
+extern struct kobject * __must_check kobject_create(void);
 extern struct kobject * __must_check kobject_create_and_add(const char *name,
 						struct kobject *parent);
 
@@ -103,9 +104,6 @@ extern void kobject_unregister(struct kobject *);
 extern struct kobject * kobject_get(struct kobject *);
 extern void kobject_put(struct kobject *);
 
-extern struct kobject *kobject_kset_add_dir(struct kset *kset,
-					    struct kobject *, const char *);
-
 extern char * kobject_get_path(struct kobject *, gfp_t);
 
 struct kobj_type {
diff --git a/lib/kobject.c b/lib/kobject.c
index 96b61d9a9284..67c3d38d48f0 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -629,15 +629,18 @@ static struct kobj_type dynamic_kobj_ktype = {
 	.release = dynamic_kobj_release,
 };
 
-/*
+/**
  * kobject_create - create a struct kobject dynamically
  *
  * This function creates a kobject structure dynamically and sets it up
  * to be a "dynamic" kobject with a default release function set up.
  *
  * If the kobject was not able to be created, NULL will be returned.
+ * The kobject structure returned from here must be cleaned up with a
+ * call to kobject_put() and not kfree(), as kobject_init_ng() has
+ * already been called on this structure.
  */
-static struct kobject *kobject_create(void)
+struct kobject *kobject_create(void)
 {
 	struct kobject *kobj;
 
@@ -682,36 +685,6 @@ struct kobject *kobject_create_and_add(const char *name, struct kobject *parent)
 }
 EXPORT_SYMBOL_GPL(kobject_create_and_add);
 
-/**
- *	kobject_kset_add_dir - add sub directory of object.
- *	@kset:		kset the directory is belongs to.
- *	@parent:	object in which a directory is created.
- *	@name:	directory name.
- *
- *	Add a plain directory object as child of given object.
- */
-struct kobject *kobject_kset_add_dir(struct kset *kset,
-				     struct kobject *parent, const char *name)
-{
-	struct kobject *k;
-	int ret;
-
-	k = kobject_create();
-	if (!k)
-		return NULL;
-
-	k->kset = kset;
-	ret = kobject_add_ng(k, parent, "%s", name);
-	if (ret < 0) {
-		printk(KERN_WARNING "%s: kobject_add error: %d\n",
-			__func__, ret);
-		kobject_put(k);
-		k = NULL;
-	}
-
-	return k;
-}
-
 /**
  *	kset_init - initialize a kset for use
  *	@k:	kset 
-- 
cgit v1.2.3


From 00d2666623368ffd39afc875ff8a2eead2a0436c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 29 Oct 2007 14:17:23 -0600
Subject: kobject: convert main fs kobject to use kobject_create

This also renames fs_subsys to fs_kobj to catch all current users with a
build error instead of a build warning which can easily be missed.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/ecryptfs/main.c |  2 +-
 fs/fuse/inode.c    |  2 +-
 fs/gfs2/sys.c      |  2 +-
 fs/namespace.c     | 11 +++++------
 include/linux/fs.h |  2 +-
 5 files changed, 9 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 4750d82c3db9..bdeac3877a84 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -798,7 +798,7 @@ static int do_sysfs_registration(void)
 {
 	int rc;
 
-	ecryptfs_kset = kset_create_and_add("ecryptfs", NULL, &fs_subsys.kobj);
+	ecryptfs_kset = kset_create_and_add("ecryptfs", NULL, fs_kobj);
 	if (!ecryptfs_kset) {
 		printk(KERN_ERR "Unable to create ecryptfs kset\n");
 		rc = -ENOMEM;
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 92118066f1d6..e6e23a2ad4b3 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -795,7 +795,7 @@ static int fuse_sysfs_init(void)
 {
 	int err;
 
-	fuse_kobj = kobject_create_and_add("fuse", &fs_subsys.kobj);
+	fuse_kobj = kobject_create_and_add("fuse", fs_kobj);
 	if (!fuse_kobj) {
 		err = -ENOMEM;
 		goto out_err;
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index d7fa54443f0c..a0bdc4a3acf9 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -549,7 +549,7 @@ int gfs2_sys_init(void)
 	gfs2_sys_margs = NULL;
 	spin_lock_init(&gfs2_sys_margs_lock);
 	kobject_set_name(&gfs2_kset.kobj, "gfs2");
-	gfs2_kset.kobj.kset = &fs_subsys;
+	gfs2_kset.kobj.parent = fs_kobj;
 	return kset_register(&gfs2_kset);
 }
 
diff --git a/fs/namespace.c b/fs/namespace.c
index a4a3f70e7e26..61bf376e29e8 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -41,8 +41,8 @@ static struct kmem_cache *mnt_cache __read_mostly;
 static struct rw_semaphore namespace_sem;
 
 /* /sys/fs */
-decl_subsys(fs, NULL);
-EXPORT_SYMBOL_GPL(fs_subsys);
+struct kobject *fs_kobj;
+EXPORT_SYMBOL_GPL(fs_kobj);
 
 static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry)
 {
@@ -1861,10 +1861,9 @@ void __init mnt_init(void)
 	if (err)
 		printk(KERN_WARNING "%s: sysfs_init error: %d\n",
 			__FUNCTION__, err);
-	err = subsystem_register(&fs_subsys);
-	if (err)
-		printk(KERN_WARNING "%s: subsystem_register error: %d\n",
-			__FUNCTION__, err);
+	fs_kobj = kobject_create_and_add("fs", NULL);
+	if (!fs_kobj)
+		printk(KERN_WARNING "%s: kobj create error\n", __FUNCTION__);
 	init_rootfs();
 	init_mount_tree();
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index b3ec4a496d64..21398a5d688d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1476,7 +1476,7 @@ extern void drop_collected_mounts(struct vfsmount *);
 extern int vfs_statfs(struct dentry *, struct kstatfs *);
 
 /* /sys/fs */
-extern struct kset fs_subsys;
+extern struct kobject *fs_kobj;
 
 #define FLOCK_VERIFY_READ  1
 #define FLOCK_VERIFY_WRITE 2
-- 
cgit v1.2.3


From 81ace5cd8fcb55e144f496af40d4275b03252456 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 29 Oct 2007 23:22:26 -0500
Subject: kset: convert pci hotplug to use kset_create_and_add

This also renames pci_hotplug_slots_subsys to pcis_hotplug_slots_kset
catch all current users with a build error instead of a build warning
which can easily be missed.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_ibm.c      |  4 ++--
 drivers/pci/hotplug/pci_hotplug_core.c | 23 +++++++++++------------
 drivers/pci/hotplug/rpadlpar_sysfs.c   |  2 +-
 include/linux/pci_hotplug.h            |  2 +-
 4 files changed, 15 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c
index 47d26b65e99a..750ebd7a4c10 100644
--- a/drivers/pci/hotplug/acpiphp_ibm.c
+++ b/drivers/pci/hotplug/acpiphp_ibm.c
@@ -429,7 +429,7 @@ static int __init ibm_acpiphp_init(void)
 	int retval = 0;
 	acpi_status status;
 	struct acpi_device *device;
-	struct kobject *sysdir = &pci_hotplug_slots_subsys.kobj;
+	struct kobject *sysdir = &pci_hotplug_slots_kset->kobj;
 
 	dbg("%s\n", __FUNCTION__);
 
@@ -476,7 +476,7 @@ init_return:
 static void __exit ibm_acpiphp_exit(void)
 {
 	acpi_status status;
-	struct kobject *sysdir = &pci_hotplug_slots_subsys.kobj;
+	struct kobject *sysdir = &pci_hotplug_slots_kset->kobj;
 
 	dbg("%s\n", __FUNCTION__);
 
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index ce1cff0fdeca..175e0c8599e3 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -61,7 +61,7 @@ static int debug;
 
 static LIST_HEAD(pci_hotplug_slot_list);
 
-struct kset pci_hotplug_slots_subsys;
+struct kset *pci_hotplug_slots_kset;
 
 static ssize_t hotplug_slot_attr_show(struct kobject *kobj,
 		struct attribute *attr, char *buf)
@@ -96,8 +96,6 @@ static struct kobj_type hotplug_slot_ktype = {
 	.release = &hotplug_slot_release,
 };
 
-decl_subsys_name(pci_hotplug_slots, slots, NULL);
-
 /* these strings match up with the values in pci_bus_speed */
 static char *pci_bus_speed_strings[] = {
 	"33 MHz PCI",		/* 0x00 */
@@ -633,7 +631,7 @@ int pci_hp_register (struct hotplug_slot *slot)
 	}
 
 	kobject_set_name(&slot->kobj, "%s", slot->name);
-	slot->kobj.kset = &pci_hotplug_slots_subsys;
+	slot->kobj.kset = pci_hotplug_slots_kset;
 	slot->kobj.ktype = &hotplug_slot_ktype;
 
 	/* this can fail if we have already registered a slot with the same name */
@@ -702,10 +700,11 @@ static int __init pci_hotplug_init (void)
 {
 	int result;
 
-	pci_hotplug_slots_subsys.kobj.kset = &pci_bus_type.subsys;
-	result = subsystem_register(&pci_hotplug_slots_subsys);
-	if (result) {
-		err("Register subsys with error %d\n", result);
+	pci_hotplug_slots_kset = kset_create_and_add("slots", NULL,
+						     &pci_bus_type.subsys.kobj);
+	if (!pci_hotplug_slots_kset) {
+		result = -ENOMEM;
+		err("Register subsys error\n");
 		goto exit;
 	}
 	result = cpci_hotplug_init(debug);
@@ -716,9 +715,9 @@ static int __init pci_hotplug_init (void)
 
 	info (DRIVER_DESC " version: " DRIVER_VERSION "\n");
 	goto exit;
-	
+
 err_subsys:
-	subsystem_unregister(&pci_hotplug_slots_subsys);
+	kset_unregister(pci_hotplug_slots_kset);
 exit:
 	return result;
 }
@@ -726,7 +725,7 @@ exit:
 static void __exit pci_hotplug_exit (void)
 {
 	cpci_hotplug_exit();
-	subsystem_unregister(&pci_hotplug_slots_subsys);
+	kset_unregister(pci_hotplug_slots_kset);
 }
 
 module_init(pci_hotplug_init);
@@ -738,7 +737,7 @@ MODULE_LICENSE("GPL");
 module_param(debug, bool, 0644);
 MODULE_PARM_DESC(debug, "Debugging mode enabled or not");
 
-EXPORT_SYMBOL_GPL(pci_hotplug_slots_subsys);
+EXPORT_SYMBOL_GPL(pci_hotplug_slots_kset);
 EXPORT_SYMBOL_GPL(pci_hp_register);
 EXPORT_SYMBOL_GPL(pci_hp_deregister);
 EXPORT_SYMBOL_GPL(pci_hp_change_slot_info);
diff --git a/drivers/pci/hotplug/rpadlpar_sysfs.c b/drivers/pci/hotplug/rpadlpar_sysfs.c
index 76090937c758..5c3ddb60a002 100644
--- a/drivers/pci/hotplug/rpadlpar_sysfs.c
+++ b/drivers/pci/hotplug/rpadlpar_sysfs.c
@@ -130,7 +130,7 @@ struct kobj_type ktype_dlpar_io = {
 
 struct kset dlpar_io_kset = {
 	.kobj = {.ktype = &ktype_dlpar_io,
-		 .parent = &pci_hotplug_slots_subsys.kobj},
+		 .parent = &pci_hotplug_slots_kset->kobj},
 };
 
 int dlpar_sysfs_init(void)
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index ab4cb6ecd47c..8f67e8f2a3cc 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -174,7 +174,7 @@ extern int pci_hp_register		(struct hotplug_slot *slot);
 extern int pci_hp_deregister		(struct hotplug_slot *slot);
 extern int __must_check pci_hp_change_slot_info	(struct hotplug_slot *slot,
 						 struct hotplug_slot_info *info);
-extern struct kset pci_hotplug_slots_subsys;
+extern struct kset *pci_hotplug_slots_kset;
 
 /* PCI Setting Record (Type 0) */
 struct hpp_type0 {
-- 
cgit v1.2.3


From e5e38a86c0bbe8475543f10f0a48393a45df5182 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 29 Oct 2007 23:22:26 -0500
Subject: kset: remove decl_subsys_name

The last user of this macro (pci hotplug core) is now switched over to
using a dynamic kset, so this macro is no longer needed at all.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 718b48811288..390ae14b73e8 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -193,11 +193,6 @@ struct kset _name##_subsys = { \
 	.kobj = { .k_name = __stringify(_name) }, \
 	.uevent_ops =_uevent_ops, \
 }
-#define decl_subsys_name(_varname,_name,_uevent_ops) \
-struct kset _varname##_subsys = { \
-	.kobj = { .k_name = __stringify(_name) }, \
-	.uevent_ops =_uevent_ops, \
-}
 
 /* The global /sys/kernel/ subsystem for people to chain off of */
 extern struct kset kernel_subsys;
-- 
cgit v1.2.3


From bd35b93d8049ab47b5bfaf6b10ba39badf21d1c3 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 29 Oct 2007 20:13:17 +0100
Subject: kset: convert kernel_subsys to use kset_create

Dynamically create the kset instead of declaring it statically.  We also
rename kernel_subsys to kernel_kset to catch all users of this symbol
with a build error instead of an easy-to-ignore build warning.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/configfs/mount.c         |  2 +-
 fs/debugfs/inode.c          |  2 +-
 fs/dlm/lockspace.c          |  2 +-
 fs/gfs2/locking/dlm/sysfs.c |  3 +--
 include/linux/kobject.h     |  4 ++--
 kernel/ksysfs.c             | 42 ++++++++++++++++++++++++++++++------------
 kernel/user.c               |  4 ++--
 security/inode.c            |  2 +-
 8 files changed, 39 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index 13300466464b..c4ee7f05de8b 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -140,7 +140,7 @@ static int __init configfs_init(void)
 	if (!configfs_dir_cachep)
 		goto out;
 
-	config_kobj = kobject_create_and_add("config", &kernel_subsys.kobj);
+	config_kobj = kobject_create_and_add("config", &kernel_kset->kobj);
 	if (!config_kobj) {
 		kmem_cache_destroy(configfs_dir_cachep);
 		configfs_dir_cachep = NULL;
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 667214200b03..5ce92c3d3b59 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -432,7 +432,7 @@ static int __init debugfs_init(void)
 {
 	int retval;
 
-	debug_kobj = kobject_create_and_add("debug", &kernel_subsys.kobj);
+	debug_kobj = kobject_create_and_add("debug", &kernel_kset->kobj);
 	if (!debug_kobj)
 		return -EINVAL;
 
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 83a9c4dd5114..0828beb2d35d 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -223,7 +223,7 @@ int dlm_lockspace_init(void)
 	INIT_LIST_HEAD(&lslist);
 	spin_lock_init(&lslist_lock);
 
-	dlm_kset = kset_create_and_add("dlm", NULL, &kernel_subsys.kobj);
+	dlm_kset = kset_create_and_add("dlm", NULL, &kernel_kset->kobj);
 	if (!dlm_kset) {
 		printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__);
 		return -ENOMEM;
diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c
index 0a8614088ec6..1a92b6f7bc10 100644
--- a/fs/gfs2/locking/dlm/sysfs.c
+++ b/fs/gfs2/locking/dlm/sysfs.c
@@ -219,8 +219,7 @@ void gdlm_kobject_release(struct gdlm_ls *ls)
 
 int gdlm_sysfs_init(void)
 {
-	gdlm_kset = kset_create_and_add("lock_dlm", NULL,
-					&kernel_subsys.kobj);
+	gdlm_kset = kset_create_and_add("lock_dlm", NULL, &kernel_kset->kobj);
 	if (!gdlm_kset) {
 		printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__);
 		return -ENOMEM;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 390ae14b73e8..bd741e86c11e 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -194,8 +194,8 @@ struct kset _name##_subsys = { \
 	.uevent_ops =_uevent_ops, \
 }
 
-/* The global /sys/kernel/ subsystem for people to chain off of */
-extern struct kset kernel_subsys;
+/* The global /sys/kernel/ kset for people to chain off of */
+extern struct kset *kernel_kset;
 /* The global /sys/hypervisor/ subsystem  */
 extern struct kset hypervisor_subsys;
 
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index 094e2bc101a8..cf02d4ba9add 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -94,8 +94,8 @@ static struct bin_attribute notes_attr = {
 	.read = &notes_read,
 };
 
-decl_subsys(kernel, NULL);
-EXPORT_SYMBOL_GPL(kernel_subsys);
+struct kset *kernel_kset;
+EXPORT_SYMBOL_GPL(kernel_kset);
 
 static struct attribute * kernel_attrs[] = {
 #if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
@@ -116,24 +116,42 @@ static struct attribute_group kernel_attr_group = {
 
 static int __init ksysfs_init(void)
 {
-	int error = subsystem_register(&kernel_subsys);
-	if (!error)
-		error = sysfs_create_group(&kernel_subsys.kobj,
-					   &kernel_attr_group);
+	int error;
 
-	if (!error && notes_size > 0) {
+	kernel_kset = kset_create_and_add("kernel", NULL, NULL);
+	if (!kernel_kset) {
+		error = -ENOMEM;
+		goto exit;
+	}
+	error = sysfs_create_group(&kernel_kset->kobj, &kernel_attr_group);
+	if (error)
+		goto kset_exit;
+
+	if (notes_size > 0) {
 		notes_attr.size = notes_size;
-		error = sysfs_create_bin_file(&kernel_subsys.kobj,
-					      &notes_attr);
+		error = sysfs_create_bin_file(&kernel_kset->kobj, &notes_attr);
+		if (error)
+			goto group_exit;
 	}
 
 	/*
 	 * Create "/sys/kernel/uids" directory and corresponding root user's
 	 * directory under it.
 	 */
-	if (!error)
-		error = uids_kobject_init();
-
+	error = uids_kobject_init();
+	if (error)
+		goto notes_exit;
+
+	return 0;
+
+notes_exit:
+	if (notes_size > 0)
+		sysfs_remove_bin_file(&kernel_kset->kobj, &notes_attr);
+group_exit:
+	sysfs_remove_group(&kernel_kset->kobj, &kernel_attr_group);
+kset_exit:
+	kset_unregister(kernel_kset);
+exit:
 	return error;
 }
 
diff --git a/kernel/user.c b/kernel/user.c
index 8320a87f3e5a..80f1116b8fcd 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -198,8 +198,8 @@ int __init uids_kobject_init(void)
 	int error;
 
 	/* create under /sys/kernel dir */
-	uids_kobject.parent = &kernel_subsys.kobj;
-	uids_kobject.kset = &kernel_subsys;
+	uids_kobject.parent = &kernel_kset->kobj;
+	uids_kobject.kset = kernel_kset;
 	kobject_set_name(&uids_kobject, "uids");
 	kobject_init(&uids_kobject);
 
diff --git a/security/inode.c b/security/inode.c
index dfc5978d4298..dbe040ac0549 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -321,7 +321,7 @@ static int __init securityfs_init(void)
 {
 	int retval;
 
-	security_kobj = kobject_create_and_add("security", &kernel_subsys.kobj);
+  	security_kobj = kobject_create_and_add("security", &kernel_kset->kobj);
 	if (!security_kobj)
 		return -EINVAL;
 
-- 
cgit v1.2.3


From 2d72fc00a1fb055e6127ccd30cac3f0eafaa98d0 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 09:29:06 -0600
Subject: kobject: convert /sys/hypervisor to use kobject_create

We don't need a kset here, a simple kobject will do just fine, so
dynamically create the kobject and use it.

We also rename hypervisor_subsys to hypervisor_kset to catch all users
of the variable.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/s390/hypfs/inode.c   |  2 +-
 drivers/base/hypervisor.c | 12 ++++++++----
 include/linux/kobject.h   |  4 ++--
 3 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index c022ccc04d41..b0ad479e7487 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -506,7 +506,7 @@ static int __init hypfs_init(void)
 			goto fail_diag;
 		}
 	}
-	s390_subsys.kobj.kset = &hypervisor_subsys;
+	s390_subsys.kobj.parent = hypervisor_kobj;
 	rc = subsystem_register(&s390_subsys);
 	if (rc)
 		goto fail_sysfs;
diff --git a/drivers/base/hypervisor.c b/drivers/base/hypervisor.c
index 14e75e9ec783..6428cba3aadd 100644
--- a/drivers/base/hypervisor.c
+++ b/drivers/base/hypervisor.c
@@ -2,19 +2,23 @@
  * hypervisor.c - /sys/hypervisor subsystem.
  *
  * Copyright (C) IBM Corp. 2006
+ * Copyright (C) 2007 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (C) 2007 Novell Inc.
  *
  * This file is released under the GPLv2
  */
 
 #include <linux/kobject.h>
 #include <linux/device.h>
-
 #include "base.h"
 
-decl_subsys(hypervisor, NULL);
-EXPORT_SYMBOL_GPL(hypervisor_subsys);
+struct kobject *hypervisor_kobj;
+EXPORT_SYMBOL_GPL(hypervisor_kobj);
 
 int __init hypervisor_init(void)
 {
-	return subsystem_register(&hypervisor_subsys);
+	hypervisor_kobj = kobject_create_and_add("hypervisor", NULL);
+	if (!hypervisor_kobj)
+		return -ENOMEM;
+	return 0;
 }
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index bd741e86c11e..f2483f6fd639 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -196,8 +196,8 @@ struct kset _name##_subsys = { \
 
 /* The global /sys/kernel/ kset for people to chain off of */
 extern struct kset *kernel_kset;
-/* The global /sys/hypervisor/ subsystem  */
-extern struct kset hypervisor_subsys;
+/* The global /sys/hypervisor/ kobject for people to chain off of */
+extern struct kobject *hypervisor_kobj;
 
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
-- 
cgit v1.2.3


From 7405c1e15edfe43b137bfbc5882f1af34d6d414d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 10:39:50 -0700
Subject: kset: convert /sys/module to use kset_create

Dynamically create the kset instead of declaring it statically.  We also
rename module_subsys to module_kset to catch all users of the variable.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/module.h |  4 +++-
 kernel/module.c        |  7 +++----
 kernel/params.c        | 29 +++++++++--------------------
 3 files changed, 15 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/module.h b/include/linux/module.h
index 2cbc0b87e329..fbe930b9b69c 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -574,7 +574,9 @@ struct device_driver;
 #ifdef CONFIG_SYSFS
 struct module;
 
-extern struct kset module_subsys;
+extern struct kset *module_kset;
+extern struct kobj_type module_ktype;
+extern int module_sysfs_initialized;
 
 int mod_sysfs_init(struct module *mod);
 int mod_sysfs_setup(struct module *mod,
diff --git a/kernel/module.c b/kernel/module.c
index 55142775c581..d03fcd9d652c 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -47,8 +47,6 @@
 #include <asm/cacheflush.h>
 #include <linux/license.h>
 
-extern int module_sysfs_initialized;
-
 #if 0
 #define DEBUGP printk
 #else
@@ -1223,7 +1221,8 @@ int mod_sysfs_init(struct module *mod)
 	err = kobject_set_name(&mod->mkobj.kobj, "%s", mod->name);
 	if (err)
 		goto out;
-	mod->mkobj.kobj.kset = &module_subsys;
+	mod->mkobj.kobj.kset = module_kset;
+	mod->mkobj.kobj.ktype = &module_ktype;
 	mod->mkobj.mod = mod;
 
 	kobject_init(&mod->mkobj.kobj);
@@ -2539,7 +2538,7 @@ void module_add_driver(struct module *mod, struct device_driver *drv)
 		struct kobject *mkobj;
 
 		/* Lookup built-in module entry in /sys/modules */
-		mkobj = kset_find_obj(&module_subsys, drv->mod_name);
+		mkobj = kset_find_obj(module_kset, drv->mod_name);
 		if (mkobj) {
 			mk = container_of(mkobj, struct module_kobject, kobj);
 			/* remember our module structure */
diff --git a/kernel/params.c b/kernel/params.c
index 9f051824097d..97e092312155 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -30,8 +30,6 @@
 #define DEBUGP(fmt, a...)
 #endif
 
-static struct kobj_type module_ktype;
-
 static inline char dash2underscore(char c)
 {
 	if (c == '-')
@@ -562,7 +560,7 @@ static void __init kernel_param_sysfs_setup(const char *name,
 	BUG_ON(!mk);
 
 	mk->mod = THIS_MODULE;
-	mk->kobj.kset = &module_subsys;
+	mk->kobj.kset = module_kset;
 	mk->kobj.ktype = &module_ktype;
 	kobject_set_name(&mk->kobj, name);
 	kobject_init(&mk->kobj);
@@ -695,7 +693,7 @@ static struct kset_uevent_ops module_uevent_ops = {
 	.filter = uevent_filter,
 };
 
-decl_subsys(module, &module_uevent_ops);
+struct kset *module_kset;
 int module_sysfs_initialized;
 
 static void module_release(struct kobject *kobj)
@@ -707,7 +705,7 @@ static void module_release(struct kobject *kobj)
 	 */
 }
 
-static struct kobj_type module_ktype = {
+struct kobj_type module_ktype = {
 	.sysfs_ops =	&module_sysfs_ops,
 	.release =	module_release,
 };
@@ -717,13 +715,11 @@ static struct kobj_type module_ktype = {
  */
 static int __init param_sysfs_init(void)
 {
-	int ret;
-
-	ret = subsystem_register(&module_subsys);
-	if (ret < 0) {
-		printk(KERN_WARNING "%s (%d): subsystem_register error: %d\n",
-			__FILE__, __LINE__, ret);
-		return ret;
+	module_kset = kset_create_and_add("module", &module_uevent_ops, NULL);
+	if (!module_kset) {
+		printk(KERN_WARNING "%s (%d): error creating kset\n",
+			__FILE__, __LINE__);
+		return -ENOMEM;
 	}
 	module_sysfs_initialized = 1;
 
@@ -733,14 +729,7 @@ static int __init param_sysfs_init(void)
 }
 subsys_initcall(param_sysfs_init);
 
-#else
-#if 0
-static struct sysfs_ops module_sysfs_ops = {
-	.show = NULL,
-	.store = NULL,
-};
-#endif
-#endif
+#endif /* CONFIG_SYSFS */
 
 EXPORT_SYMBOL(param_set_byte);
 EXPORT_SYMBOL(param_get_byte);
-- 
cgit v1.2.3


From 039a5dcd2fc45188a2d522df630db4f7ef903a0f Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 10:39:50 -0700
Subject: kset: convert /sys/power to use kset_create

Dynamically create the kset instead of declaring it statically.  We also
rename power_subsys to power_kset to catch all users of the variable and
we properly export it so that people don't have to guess that it really
is present in the system.

The pseries code is wierd, why is it createing /sys/power if CONFIG_PM
is disabled?  Oh well, stupid big boxes ignoring config options...

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/mach-omap1/pm.c               |  3 +--
 arch/powerpc/platforms/pseries/power.c | 14 ++++++--------
 include/linux/kobject.h                |  2 ++
 kernel/power/disk.c                    |  2 +-
 kernel/power/main.c                    | 11 +++++------
 kernel/power/power.h                   |  2 --
 6 files changed, 15 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-omap1/pm.c b/arch/arm/mach-omap1/pm.c
index 3bf01e28df33..402113c72981 100644
--- a/arch/arm/mach-omap1/pm.c
+++ b/arch/arm/mach-omap1/pm.c
@@ -97,7 +97,6 @@ static struct subsys_attribute sleep_while_idle_attr = {
 	.store  = omap_pm_sleep_while_idle_store,
 };
 
-extern struct kset power_subsys;
 static void (*omap_sram_idle)(void) = NULL;
 static void (*omap_sram_suspend)(unsigned long r0, unsigned long r1) = NULL;
 
@@ -726,7 +725,7 @@ static int __init omap_pm_init(void)
 	omap_pm_init_proc();
 #endif
 
-	error = subsys_create_file(&power_subsys, &sleep_while_idle_attr);
+	error = subsys_create_file(power_kset, &sleep_while_idle_attr);
 	if (error)
 		printk(KERN_ERR "subsys_create_file failed: %d\n", error);
 
diff --git a/arch/powerpc/platforms/pseries/power.c b/arch/powerpc/platforms/pseries/power.c
index 08d7a5007167..c36febe7ce7d 100644
--- a/arch/powerpc/platforms/pseries/power.c
+++ b/arch/powerpc/platforms/pseries/power.c
@@ -57,7 +57,7 @@ static struct subsys_attribute auto_poweron_attr = {
 };
 
 #ifndef CONFIG_PM
-decl_subsys(power, NULL);
+struct kset *power_kset;
 
 static struct attribute *g[] = {
         &auto_poweron_attr.attr,
@@ -70,18 +70,16 @@ static struct attribute_group attr_group = {
 
 static int __init pm_init(void)
 {
-        int error = subsystem_register(&power_subsys);
-        if (!error)
-                error = sysfs_create_group(&power_subsys.kobj, &attr_group);
-        return error;
+	power_kset = kset_create_and_add("power", NULL, NULL);
+	if (!power_kset)
+		return -ENOMEM;
+	return sysfs_create_group(&power_kset->kobj, &attr_group);
 }
 core_initcall(pm_init);
 #else
-extern struct kset power_subsys;
-
 static int __init apo_pm_init(void)
 {
-	return (subsys_create_file(&power_subsys, &auto_poweron_attr));
+	return (subsys_create_file(power_kset, &auto_poweron_attr));
 }
 __initcall(apo_pm_init);
 #endif
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index f2483f6fd639..a6dd669cda9d 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -198,6 +198,8 @@ struct kset _name##_subsys = { \
 extern struct kset *kernel_kset;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
 extern struct kobject *hypervisor_kobj;
+/* The global /sys/power/ kset for people to chain off of */
+extern struct kset *power_kset;
 
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 05b64790fe83..c3f0e61365dd 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -708,7 +708,7 @@ static struct attribute_group attr_group = {
 
 static int __init pm_disk_init(void)
 {
-	return sysfs_create_group(&power_subsys.kobj, &attr_group);
+	return sysfs_create_group(&power_kset->kobj, &attr_group);
 }
 
 core_initcall(pm_disk_init);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 1ef31c91ce0e..dce2d76d66de 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -276,8 +276,7 @@ EXPORT_SYMBOL(pm_suspend);
 
 #endif /* CONFIG_SUSPEND */
 
-decl_subsys(power, NULL);
-
+struct kset *power_kset;
 
 /**
  *	state - control system power state.
@@ -386,10 +385,10 @@ static struct attribute_group attr_group = {
 
 static int __init pm_init(void)
 {
-	int error = subsystem_register(&power_subsys);
-	if (!error)
-		error = sysfs_create_group(&power_subsys.kobj,&attr_group);
-	return error;
+	power_kset = kset_create_and_add("power", NULL, NULL);
+	if (!power_kset)
+		return -ENOMEM;
+	return sysfs_create_group(&power_kset->kobj, &attr_group);
 }
 
 core_initcall(pm_init);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 195dc4611764..1083e6b188ab 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -63,8 +63,6 @@ static struct subsys_attribute _name##_attr = {	\
 	.store	= _name##_store,		\
 }
 
-extern struct kset power_subsys;
-
 /* Preferred image size in bytes (default 500 MB) */
 extern unsigned long image_size;
 extern int in_suspend;
-- 
cgit v1.2.3


From 3d8995963dfec66ef6270e729bf75903e9043f9d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 13:31:26 -0700
Subject: kset: convert struct bus_device->devices to use kset_create

Dynamically create the kset instead of declaring it statically.

Having 3 static kobjects in one structure is not only foolish, but ripe
for nasty race conditions if handled improperly.  We also rename the
field to catch any potential users of it (not that there should be
outside of the driver core...)

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 19 ++++++++++---------
 include/linux/device.h |  2 +-
 2 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index e3b101077804..b23eeb2d4ead 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -449,7 +449,7 @@ int bus_add_device(struct device * dev)
 		error = device_add_attrs(bus, dev);
 		if (error)
 			goto out_put;
-		error = sysfs_create_link(&bus->devices.kobj,
+		error = sysfs_create_link(&bus->devices_kset->kobj,
 						&dev->kobj, dev->bus_id);
 		if (error)
 			goto out_id;
@@ -466,7 +466,7 @@ int bus_add_device(struct device * dev)
 out_deprecated:
 	sysfs_remove_link(&dev->kobj, "subsystem");
 out_subsys:
-	sysfs_remove_link(&bus->devices.kobj, dev->bus_id);
+	sysfs_remove_link(&bus->devices_kset->kobj, dev->bus_id);
 out_id:
 	device_remove_attrs(bus, dev);
 out_put:
@@ -512,7 +512,7 @@ void bus_remove_device(struct device * dev)
 	if (dev->bus) {
 		sysfs_remove_link(&dev->kobj, "subsystem");
 		remove_deprecated_bus_links(dev);
-		sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id);
+		sysfs_remove_link(&dev->bus->devices_kset->kobj, dev->bus_id);
 		device_remove_attrs(dev->bus, dev);
 		if (dev->is_registered) {
 			dev->is_registered = 0;
@@ -862,11 +862,12 @@ int bus_register(struct bus_type * bus)
 	if (retval)
 		goto bus_uevent_fail;
 
-	kobject_set_name(&bus->devices.kobj, "devices");
-	bus->devices.kobj.parent = &bus->subsys.kobj;
-	retval = kset_register(&bus->devices);
-	if (retval)
+	bus->devices_kset = kset_create_and_add("devices", NULL,
+						&bus->subsys.kobj);
+	if (!bus->devices_kset) {
+		retval = -ENOMEM;
 		goto bus_devices_fail;
+	}
 
 	kobject_set_name(&bus->drivers.kobj, "drivers");
 	bus->drivers.kobj.parent = &bus->subsys.kobj;
@@ -894,7 +895,7 @@ bus_attrs_fail:
 bus_probe_files_fail:
 	kset_unregister(&bus->drivers);
 bus_drivers_fail:
-	kset_unregister(&bus->devices);
+	kset_unregister(bus->devices_kset);
 bus_devices_fail:
 	bus_remove_file(bus, &bus_attr_uevent);
 bus_uevent_fail:
@@ -916,7 +917,7 @@ void bus_unregister(struct bus_type * bus)
 	bus_remove_attrs(bus);
 	remove_probe_files(bus);
 	kset_unregister(&bus->drivers);
-	kset_unregister(&bus->devices);
+	kset_unregister(bus->devices_kset);
 	bus_remove_file(bus, &bus_attr_uevent);
 	subsystem_unregister(&bus->subsys);
 }
diff --git a/include/linux/device.h b/include/linux/device.h
index dbbbe89e7267..82c277771376 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -55,7 +55,7 @@ struct bus_type {
 
 	struct kset		subsys;
 	struct kset		drivers;
-	struct kset		devices;
+	struct kset		*devices_kset;
 	struct klist		klist_devices;
 	struct klist		klist_drivers;
 
-- 
cgit v1.2.3


From 6dcec2511ff55b4abaca7ad3433011a7c04c2430 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 13:31:26 -0700
Subject: kset: convert struct bus_device->drivers to use kset_create

Dynamically create the kset instead of declaring it statically.

Having 3 static kobjects in one structure is not only foolish, but ripe
for nasty race conditions if handled improperly.  We also rename the
field to catch any potential users of it (not that there should be
outside of the driver core...)

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 15 ++++++++-------
 drivers/base/driver.c  |  2 +-
 include/linux/device.h |  2 +-
 3 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index b23eeb2d4ead..6796d3e46053 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -638,7 +638,7 @@ int bus_add_driver(struct device_driver *drv)
 	error = kobject_set_name(&drv->kobj, "%s", drv->name);
 	if (error)
 		goto out_put_bus;
-	drv->kobj.kset = &bus->drivers;
+	drv->kobj.kset = bus->drivers_kset;
 	drv->kobj.ktype = &driver_ktype;
 	error = kobject_register(&drv->kobj);
 	if (error)
@@ -869,11 +869,12 @@ int bus_register(struct bus_type * bus)
 		goto bus_devices_fail;
 	}
 
-	kobject_set_name(&bus->drivers.kobj, "drivers");
-	bus->drivers.kobj.parent = &bus->subsys.kobj;
-	retval = kset_register(&bus->drivers);
-	if (retval)
+	bus->drivers_kset = kset_create_and_add("drivers", NULL,
+						&bus->subsys.kobj);
+	if (!bus->drivers_kset) {
+		retval = -ENOMEM;
 		goto bus_drivers_fail;
+	}
 
 	klist_init(&bus->klist_devices, klist_devices_get, klist_devices_put);
 	klist_init(&bus->klist_drivers, NULL, NULL);
@@ -893,7 +894,7 @@ int bus_register(struct bus_type * bus)
 bus_attrs_fail:
 	remove_probe_files(bus);
 bus_probe_files_fail:
-	kset_unregister(&bus->drivers);
+	kset_unregister(bus->drivers_kset);
 bus_drivers_fail:
 	kset_unregister(bus->devices_kset);
 bus_devices_fail:
@@ -916,7 +917,7 @@ void bus_unregister(struct bus_type * bus)
 	pr_debug("bus %s: unregistering\n", bus->name);
 	bus_remove_attrs(bus);
 	remove_probe_files(bus);
-	kset_unregister(&bus->drivers);
+	kset_unregister(bus->drivers_kset);
 	kset_unregister(bus->devices_kset);
 	bus_remove_file(bus, &bus_attr_uevent);
 	subsystem_unregister(&bus->subsys);
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index eb11475293ed..1c9770dfb801 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -185,7 +185,7 @@ void driver_unregister(struct device_driver * drv)
  */
 struct device_driver *driver_find(const char *name, struct bus_type *bus)
 {
-	struct kobject *k = kset_find_obj(&bus->drivers, name);
+	struct kobject *k = kset_find_obj(bus->drivers_kset, name);
 	if (k)
 		return to_drv(k);
 	return NULL;
diff --git a/include/linux/device.h b/include/linux/device.h
index 82c277771376..110ace0dec3f 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -54,7 +54,7 @@ struct bus_type {
 	struct module		* owner;
 
 	struct kset		subsys;
-	struct kset		drivers;
+	struct kset		*drivers_kset;
 	struct kset		*devices_kset;
 	struct klist		klist_devices;
 	struct klist		klist_drivers;
-- 
cgit v1.2.3


From 23b5212cc7422f475b82124334b64277b5b43013 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 2 Nov 2007 13:47:53 +0100
Subject: Driver Core: add kobj_attribute handling

Add kobj_sysfs_ops to replace subsys_sysfs_ops. There is no
need for special kset operations, we want to be able to use
simple attribute operations at any kobject, not only ksets.

The whole concept of any default sysfs attribute operations
will go away with the upcoming removal of subsys_sysfs_ops.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 10 ++++++++++
 lib/kobject.c           | 29 +++++++++++++++++++++++++++++
 2 files changed, 39 insertions(+)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index a6dd669cda9d..e694261de90f 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -126,6 +126,16 @@ struct kset_uevent_ops {
 		      struct kobj_uevent_env *env);
 };
 
+struct kobj_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct kobject *kobj, struct kobj_attribute *attr,
+			char *buf);
+	ssize_t (*store)(struct kobject *kobj, struct kobj_attribute *attr,
+			 const char *buf, size_t count);
+};
+
+extern struct sysfs_ops kobj_sysfs_ops;
+
 /**
  * struct kset - a set of kobjects of a specific type, belonging to a specific subsystem.
  *
diff --git a/lib/kobject.c b/lib/kobject.c
index 67c3d38d48f0..1c343fe4ba63 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -697,6 +697,35 @@ void kset_init(struct kset * k)
 	spin_lock_init(&k->list_lock);
 }
 
+/* default kobject attribute operations */
+static ssize_t kobj_attr_show(struct kobject *kobj, struct attribute *attr,
+			      char *buf)
+{
+	struct kobj_attribute *kattr;
+	ssize_t ret = -EIO;
+
+	kattr = container_of(attr, struct kobj_attribute, attr);
+	if (kattr->show)
+		ret = kattr->show(kobj, kattr, buf);
+	return ret;
+}
+
+static ssize_t kobj_attr_store(struct kobject *kobj, struct attribute *attr,
+			       const char *buf, size_t count)
+{
+	struct kobj_attribute *kattr;
+	ssize_t ret = -EIO;
+
+	kattr = container_of(attr, struct kobj_attribute, attr);
+	if (kattr->store)
+		ret = kattr->store(kobj, kattr, buf, count);
+	return ret;
+}
+
+struct sysfs_ops kobj_sysfs_ops = {
+	.show	= kobj_attr_show,
+	.store	= kobj_attr_store,
+};
 
 /**
  *	kset_add - add a kset object to the hierarchy.
-- 
cgit v1.2.3


From eb41d9465cdafee45e0cb30f3b7338646221908e Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 2 Nov 2007 13:47:53 +0100
Subject: fix struct user_info export's sysfs interaction

Clean up the use of ksets and kobjects. Kobjects are instances of
objects (like struct user_info), ksets are collections of objects of a
similar type (like the uids directory containing the user_info directories).
So, use kobjects for the user_info directories, and a kset for the "uids"
directory.

On object cleanup, the final kobject_put() was missing.

Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/sched.h |   9 +----
 kernel/ksysfs.c       |   7 +---
 kernel/user.c         | 104 +++++++++++++++++++++++++-------------------------
 3 files changed, 55 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index cc14656f8682..d6eacda765ca 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -552,18 +552,13 @@ struct user_struct {
 #ifdef CONFIG_FAIR_USER_SCHED
 	struct task_group *tg;
 #ifdef CONFIG_SYSFS
-	struct kset kset;
-	struct subsys_attribute user_attr;
+	struct kobject kobj;
 	struct work_struct work;
 #endif
 #endif
 };
 
-#ifdef CONFIG_FAIR_USER_SCHED
-extern int uids_kobject_init(void);
-#else
-static inline int uids_kobject_init(void) { return 0; }
-#endif
+extern int uids_sysfs_init(void);
 
 extern struct user_struct *find_user(uid_t);
 
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index dd0f9e7f3414..45e646585605 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -141,11 +141,8 @@ static int __init ksysfs_init(void)
 			goto group_exit;
 	}
 
-	/*
-	 * Create "/sys/kernel/uids" directory and corresponding root user's
-	 * directory under it.
-	 */
-	error = uids_kobject_init();
+	/* create the /sys/kernel/uids/ directory */
+	error = uids_sysfs_init();
 	if (error)
 		goto notes_exit;
 
diff --git a/kernel/user.c b/kernel/user.c
index 80f1116b8fcd..5a106f3fdf05 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -115,7 +115,7 @@ static void sched_switch_user(struct task_struct *p) { }
 
 #if defined(CONFIG_FAIR_USER_SCHED) && defined(CONFIG_SYSFS)
 
-static struct kobject uids_kobject; /* represents /sys/kernel/uids directory */
+static struct kset *uids_kset; /* represents the /sys/kernel/uids/ directory */
 static DEFINE_MUTEX(uids_mutex);
 
 static inline void uids_mutex_lock(void)
@@ -128,86 +128,84 @@ static inline void uids_mutex_unlock(void)
 	mutex_unlock(&uids_mutex);
 }
 
-/* return cpu shares held by the user */
-static ssize_t cpu_shares_show(struct kset *kset, char *buffer)
+/* uid directory attributes */
+static ssize_t cpu_shares_show(struct kobject *kobj,
+			       struct kobj_attribute *attr,
+			       char *buf)
 {
-	struct user_struct *up = container_of(kset, struct user_struct, kset);
+	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
 
-	return sprintf(buffer, "%lu\n", sched_group_shares(up->tg));
+	return sprintf(buf, "%lu\n", sched_group_shares(up->tg));
 }
 
-/* modify cpu shares held by the user */
-static ssize_t cpu_shares_store(struct kset *kset, const char *buffer,
-				size_t size)
+static ssize_t cpu_shares_store(struct kobject *kobj,
+				struct kobj_attribute *attr,
+				const char *buf, size_t size)
 {
-	struct user_struct *up = container_of(kset, struct user_struct, kset);
+	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
 	unsigned long shares;
 	int rc;
 
-	sscanf(buffer, "%lu", &shares);
+	sscanf(buf, "%lu", &shares);
 
 	rc = sched_group_set_shares(up->tg, shares);
 
 	return (rc ? rc : size);
 }
 
-static void user_attr_init(struct subsys_attribute *sa, char *name, int mode)
+static struct kobj_attribute cpu_share_attr =
+	__ATTR(cpu_share, 0644, cpu_shares_show, cpu_shares_store);
+
+/* default attributes per uid directory */
+static struct attribute *uids_attributes[] = {
+	&cpu_share_attr.attr,
+	NULL
+};
+
+/* the lifetime of user_struct is not managed by the core (now) */
+static void uids_release(struct kobject *kobj)
 {
-	sa->attr.name = name;
-	sa->attr.mode = mode;
-	sa->show = cpu_shares_show;
-	sa->store = cpu_shares_store;
+	return;
 }
 
-/* Create "/sys/kernel/uids/<uid>" directory and
- *  "/sys/kernel/uids/<uid>/cpu_share" file for this user.
- */
-static int user_kobject_create(struct user_struct *up)
+static struct kobj_type uids_ktype = {
+	.sysfs_ops = &kobj_sysfs_ops,
+	.default_attrs = uids_attributes,
+	.release = uids_release,
+};
+
+/* create /sys/kernel/uids/<uid>/cpu_share file for this user */
+static int uids_user_create(struct user_struct *up)
 {
-	struct kset *kset = &up->kset;
-	struct kobject *kobj = &kset->kobj;
+	struct kobject *kobj = &up->kobj;
 	int error;
 
-	memset(kset, 0, sizeof(struct kset));
-	kobj->parent = &uids_kobject;	/* create under /sys/kernel/uids dir */
-	kobject_set_name(kobj, "%d", up->uid);
-	kset_init(kset);
-	user_attr_init(&up->user_attr, "cpu_share", 0644);
-
+	memset(kobj, 0, sizeof(struct kobject));
+	kobj->ktype = &uids_ktype;
+	kobj->kset = uids_kset;
+	kobject_init(kobj);
+	kobject_set_name(&up->kobj, "%d", up->uid);
 	error = kobject_add(kobj);
 	if (error)
 		goto done;
 
-	error = sysfs_create_file(kobj, &up->user_attr.attr);
-	if (error)
-		kobject_del(kobj);
-
 	kobject_uevent(kobj, KOBJ_ADD);
-
 done:
 	return error;
 }
 
-/* create these in sysfs filesystem:
+/* create these entries in sysfs:
  * 	"/sys/kernel/uids" directory
  * 	"/sys/kernel/uids/0" directory (for root user)
  * 	"/sys/kernel/uids/0/cpu_share" file (for root user)
  */
-int __init uids_kobject_init(void)
+int __init uids_sysfs_init(void)
 {
-	int error;
+	uids_kset = kset_create_and_add("uids", NULL, &kernel_kset->kobj);
+	if (!uids_kset)
+		return -ENOMEM;
 
-	/* create under /sys/kernel dir */
-	uids_kobject.parent = &kernel_kset->kobj;
-	uids_kobject.kset = kernel_kset;
-	kobject_set_name(&uids_kobject, "uids");
-	kobject_init(&uids_kobject);
-
-	error = kobject_add(&uids_kobject);
-	if (!error)
-		error = user_kobject_create(&root_user);
-
-	return error;
+	return uids_user_create(&root_user);
 }
 
 /* work function to remove sysfs directory for a user and free up
@@ -216,7 +214,6 @@ int __init uids_kobject_init(void)
 static void remove_user_sysfs_dir(struct work_struct *w)
 {
 	struct user_struct *up = container_of(w, struct user_struct, work);
-	struct kobject *kobj = &up->kset.kobj;
 	unsigned long flags;
 	int remove_user = 0;
 
@@ -238,9 +235,9 @@ static void remove_user_sysfs_dir(struct work_struct *w)
 	if (!remove_user)
 		goto done;
 
-	sysfs_remove_file(kobj, &up->user_attr.attr);
-	kobject_uevent(kobj, KOBJ_REMOVE);
-	kobject_del(kobj);
+	kobject_uevent(&up->kobj, KOBJ_REMOVE);
+	kobject_del(&up->kobj);
+	kobject_put(&up->kobj);
 
 	sched_destroy_user(up);
 	key_put(up->uid_keyring);
@@ -267,7 +264,8 @@ static inline void free_user(struct user_struct *up, unsigned long flags)
 
 #else	/* CONFIG_FAIR_USER_SCHED && CONFIG_SYSFS */
 
-static inline int user_kobject_create(struct user_struct *up) { return 0; }
+int uids_sysfs_init(void) { return 0; }
+static inline int uids_user_create(struct user_struct *up) { return 0; }
 static inline void uids_mutex_lock(void) { }
 static inline void uids_mutex_unlock(void) { }
 
@@ -324,7 +322,7 @@ struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 	struct hlist_head *hashent = uidhashentry(ns, uid);
 	struct user_struct *up;
 
-	/* Make uid_hash_find() + user_kobject_create() + uid_hash_insert()
+	/* Make uid_hash_find() + uids_user_create() + uid_hash_insert()
 	 * atomic.
 	 */
 	uids_mutex_lock();
@@ -370,7 +368,7 @@ struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 			return NULL;
 		}
 
-		if (user_kobject_create(new)) {
+		if (uids_user_create(new)) {
 			sched_destroy_user(new);
 			key_put(new->uid_keyring);
 			key_put(new->session_keyring);
-- 
cgit v1.2.3


From 9e5f7f9abe18a4f134585a2de016974cbda80539 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 2 Nov 2007 13:20:40 -0700
Subject: firmware: export firmware_kset so that people can use that instead of
 the braindead firmware_register interface

Needed for future firmware subsystem cleanups.

In the end, the firmware_register/unregister functions will be deleted
entirely, but we need this symbol so that subsystems can migrate over.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Matt Domsch <Matt_Domsch@dell.com>
Cc: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/firmware.c | 3 ++-
 include/linux/kobject.h | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/firmware.c b/drivers/base/firmware.c
index 6a4e494042f6..c7f635b11dfd 100644
--- a/drivers/base/firmware.c
+++ b/drivers/base/firmware.c
@@ -15,7 +15,8 @@
 
 #include "base.h"
 
-static struct kset *firmware_kset;
+struct kset *firmware_kset;
+EXPORT_SYMBOL_GPL(firmware_kset);
 
 int firmware_register(struct kset *s)
 {
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index e694261de90f..29dc444e3361 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -210,6 +210,8 @@ extern struct kset *kernel_kset;
 extern struct kobject *hypervisor_kobj;
 /* The global /sys/power/ kset for people to chain off of */
 extern struct kset *power_kset;
+/* The global /sys/firmware/ kset for people to chain off of */
+extern struct kset *firmware_kset;
 
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
-- 
cgit v1.2.3


From 000f2a4d8cfc1e1cbc0aa98136015e7ae7719b46 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 2 Nov 2007 13:47:53 +0100
Subject: Driver Core: kill subsys_attribute and default sysfs ops

Remove the no longer needed subsys_attributes, they are all converted to
the more sensical kobj_attributes.

There is no longer a magic fallback in sysfs attribute operations, all
kobjects which create simple attributes need explicitely a ktype
assigned, which tells the core what was intended here.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/sysfs/file.c         | 63 ++++++++-----------------------------------------
 include/linux/kobject.h |  9 -------
 lib/kobject.c           | 21 -----------------
 3 files changed, 10 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 387a63662793..8acf82bba44c 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -20,43 +20,6 @@
 
 #include "sysfs.h"
 
-#define to_sattr(a) container_of(a,struct subsys_attribute, attr)
-
-/*
- * Subsystem file operations.
- * These operations allow subsystems to have files that can be 
- * read/written. 
- */
-static ssize_t 
-subsys_attr_show(struct kobject * kobj, struct attribute * attr, char * page)
-{
-	struct kset *kset = to_kset(kobj);
-	struct subsys_attribute * sattr = to_sattr(attr);
-	ssize_t ret = -EIO;
-
-	if (sattr->show)
-		ret = sattr->show(kset, page);
-	return ret;
-}
-
-static ssize_t 
-subsys_attr_store(struct kobject * kobj, struct attribute * attr, 
-		  const char * page, size_t count)
-{
-	struct kset *kset = to_kset(kobj);
-	struct subsys_attribute * sattr = to_sattr(attr);
-	ssize_t ret = -EIO;
-
-	if (sattr->store)
-		ret = sattr->store(kset, page, count);
-	return ret;
-}
-
-static struct sysfs_ops subsys_sysfs_ops = {
-	.show	= subsys_attr_show,
-	.store	= subsys_attr_store,
-};
-
 /*
  * There's one sysfs_buffer for each open file and one
  * sysfs_open_dirent for each sysfs_dirent with one or more open
@@ -354,29 +317,23 @@ static int sysfs_open_file(struct inode *inode, struct file *file)
 {
 	struct sysfs_dirent *attr_sd = file->f_path.dentry->d_fsdata;
 	struct kobject *kobj = attr_sd->s_parent->s_dir.kobj;
-	struct sysfs_buffer * buffer;
-	struct sysfs_ops * ops = NULL;
-	int error;
+	struct sysfs_buffer *buffer;
+	struct sysfs_ops *ops;
+	int error = -EACCES;
 
 	/* need attr_sd for attr and ops, its parent for kobj */
 	if (!sysfs_get_active_two(attr_sd))
 		return -ENODEV;
 
-	/* if the kobject has no ktype, then we assume that it is a subsystem
-	 * itself, and use ops for it.
-	 */
-	if (kobj->ktype)
+	/* every kobject with an attribute needs a ktype assigned */
+	if (kobj->ktype && kobj->ktype->sysfs_ops)
 		ops = kobj->ktype->sysfs_ops;
-	else
-		ops = &subsys_sysfs_ops;
-
-	error = -EACCES;
-
-	/* No sysfs operations, either from having no subsystem,
-	 * or the subsystem have no operations.
-	 */
-	if (!ops)
+	else {
+		printk(KERN_ERR "missing sysfs attribute operations for "
+		       "kobject: %s\n", kobject_name(kobj));
+		WARN_ON(1);
 		goto err_out;
+	}
 
 	/* File needs write support.
 	 * The inode's perms must say it's ok, 
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 29dc444e3361..29841bb5badb 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -216,15 +216,6 @@ extern struct kset *firmware_kset;
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
 
-struct subsys_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct kset *, char *);
-	ssize_t (*store)(struct kset *, const char *, size_t);
-};
-
-extern int __must_check subsys_create_file(struct kset *,
-					struct subsys_attribute *);
-
 #if defined(CONFIG_HOTPLUG)
 int kobject_uevent(struct kobject *kobj, enum kobject_action action);
 int kobject_uevent_env(struct kobject *kobj, enum kobject_action action,
diff --git a/lib/kobject.c b/lib/kobject.c
index 99f6354a5751..c742ac25228a 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -810,26 +810,6 @@ void subsystem_unregister(struct kset *s)
 	kset_unregister(s);
 }
 
-/**
- *	subsystem_create_file - export sysfs attribute file.
- *	@s:	subsystem.
- *	@a:	subsystem attribute descriptor.
- */
-
-int subsys_create_file(struct kset *s, struct subsys_attribute *a)
-{
-	int error = 0;
-
-	if (!s || !a)
-		return -EINVAL;
-
-	if (kset_get(s)) {
-		error = sysfs_create_file(&s->kobj, &a->attr);
-		kset_put(s);
-	}
-	return error;
-}
-
 static void kset_release(struct kobject *kobj)
 {
 	struct kset *kset = container_of(kobj, struct kset, kobj);
@@ -927,4 +907,3 @@ EXPORT_SYMBOL(kset_unregister);
 
 EXPORT_SYMBOL(subsystem_register);
 EXPORT_SYMBOL(subsystem_unregister);
-EXPORT_SYMBOL(subsys_create_file);
-- 
cgit v1.2.3


From 99e0d2fc6c60ca44f56203eeda9fc0e07b508f06 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 2 Nov 2007 16:19:59 -0700
Subject: kobject: convert /sys/firmware/acpi/ to use kobject_create

We don't need a kset here, a simple kobject will do just fine, so
dynamically create the kobject and use it.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Len Brown <lenb@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acpi/bus.c      | 11 ++++++-----
 drivers/acpi/system.c   |  2 +-
 include/acpi/acpi_bus.h |  2 +-
 3 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index 7c172d9d7acf..e550da684a47 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -743,7 +743,7 @@ static int __init acpi_bus_init(void)
 	return -ENODEV;
 }
 
-decl_subsys(acpi, NULL);
+struct kobject *acpi_kobj;
 
 static int __init acpi_init(void)
 {
@@ -755,10 +755,11 @@ static int __init acpi_init(void)
 		return -ENODEV;
 	}
 
-	result = firmware_register(&acpi_subsys);
-	if (result < 0)
-		printk(KERN_WARNING "%s: firmware_register error: %d\n",
-			__FUNCTION__, result);
+	acpi_kobj = kobject_create_and_add("acpi", &firmware_kset->kobj);
+	if (!acpi_kobj) {
+		printk(KERN_WARNING "%s: kset create error\n", __FUNCTION__);
+		acpi_kobj = NULL;
+	}
 
 	result = acpi_bus_init();
 
diff --git a/drivers/acpi/system.c b/drivers/acpi/system.c
index edee2806e37b..c22b93aab7fe 100644
--- a/drivers/acpi/system.c
+++ b/drivers/acpi/system.c
@@ -135,7 +135,7 @@ static int acpi_system_sysfs_init(void)
 	int table_index = 0;
 	int result;
 
-	tables_kobj.parent = &acpi_subsys.kobj;
+	tables_kobj.parent = acpi_kobj;
 	kobject_set_name(&tables_kobj, "tables");
 	result = kobject_register(&tables_kobj);
 	if (result)
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 7b74b60a68a4..fb7171b1bd22 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -319,7 +319,7 @@ struct acpi_bus_event {
 	u32 data;
 };
 
-extern struct kset acpi_subsys;
+extern struct kobject *acpi_kobj;
 extern int acpi_bus_generate_netlink_event(const char*, const char*, u8, int);
 /*
  * External Functions
-- 
cgit v1.2.3


From 15f2f9b3a9db65aaf908fe7ee17bbe262ae3550f Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 2 Nov 2007 16:19:59 -0700
Subject: firmware: remove firmware_(un)register()

These functions are no longer called or needed, so we can remove them.

As I rewrote the whole firmware.c file, add my copyright.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/firmware.c | 19 ++-----------------
 include/linux/device.h  |  5 -----
 2 files changed, 2 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/base/firmware.c b/drivers/base/firmware.c
index c7f635b11dfd..9efff481f5dc 100644
--- a/drivers/base/firmware.c
+++ b/drivers/base/firmware.c
@@ -3,11 +3,11 @@
  *
  * Copyright (c) 2002-3 Patrick Mochel
  * Copyright (c) 2002-3 Open Source Development Labs
+ * Copyright (c) 2007 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (c) 2007 Novell Inc.
  *
  * This file is released under the GPLv2
- *
  */
-
 #include <linux/kobject.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -18,18 +18,6 @@
 struct kset *firmware_kset;
 EXPORT_SYMBOL_GPL(firmware_kset);
 
-int firmware_register(struct kset *s)
-{
-	s->kobj.kset = firmware_kset;
-	s->kobj.ktype = NULL;
-	return subsystem_register(s);
-}
-
-void firmware_unregister(struct kset *s)
-{
-	subsystem_unregister(s);
-}
-
 int __init firmware_init(void)
 {
 	firmware_kset = kset_create_and_add("firmware", NULL, NULL);
@@ -37,6 +25,3 @@ int __init firmware_init(void)
 		return -ENOMEM;
 	return 0;
 }
-
-EXPORT_SYMBOL_GPL(firmware_register);
-EXPORT_SYMBOL_GPL(firmware_unregister);
diff --git a/include/linux/device.h b/include/linux/device.h
index 110ace0dec3f..a3b3ff15fc87 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -555,11 +555,6 @@ extern void device_shutdown(void);
 /* drivers/base/sys.c */
 extern void sysdev_shutdown(void);
 
-
-/* drivers/base/firmware.c */
-extern int __must_check firmware_register(struct kset *);
-extern void firmware_unregister(struct kset *);
-
 /* debugging and troubleshooting/diagnostic helpers. */
 extern const char *dev_driver_string(struct device *dev);
 #define dev_printk(level, dev, format, arg...)	\
-- 
cgit v1.2.3


From f62ed9e33b3ccff54d66b08f82d11940bb9e269b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 5 Nov 2007 13:16:15 -0800
Subject: firmware: change firmware_kset to firmware_kobj

There is no firmware "subsystem" it's just a directory in /sys that
other portions of the kernel want to hook into.  So make it a kobject
not a kset to help alivate anyone who tries to do some odd kset-like
things with this.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/s390/kernel/ipl.c      | 8 ++++----
 drivers/acpi/bus.c          | 2 +-
 drivers/base/firmware.c     | 8 ++++----
 drivers/firmware/edd.c      | 2 +-
 drivers/firmware/efivars.c  | 2 +-
 drivers/parisc/pdc_stable.c | 2 +-
 include/linux/kobject.h     | 4 ++--
 7 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index c8179fc30baa..b97694fa62ec 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -855,7 +855,7 @@ static int __init ipl_init(void)
 {
 	int rc;
 
-	ipl_kset = kset_create_and_add("ipl", NULL, &firmware_kset->kobj);
+	ipl_kset = kset_create_and_add("ipl", NULL, firmware_kobj);
 	if (!ipl_kset)
 		return -ENOMEM;
 	switch (ipl_info.type) {
@@ -974,7 +974,7 @@ static int __init reipl_init(void)
 {
 	int rc;
 
-	reipl_kset = kset_create_and_add("reipl", NULL, &firmware_kset->kobj);
+	reipl_kset = kset_create_and_add("reipl", NULL, firmware_kobj);
 	if (!reipl_kset)
 		return -ENOMEM;
 	rc = sysfs_create_file(&reipl_kset->kobj, &reipl_type_attr.attr);
@@ -1063,7 +1063,7 @@ static int __init dump_init(void)
 {
 	int rc;
 
-	dump_kset = kset_create_and_add("dump", NULL, &firmware_kset->kobj);
+	dump_kset = kset_create_and_add("dump", NULL, firmware_kobj);
 	if (!dump_kset)
 		return -ENOMEM;
 	rc = sysfs_create_file(&dump_kset->kobj, &dump_type_attr);
@@ -1086,7 +1086,7 @@ static int __init shutdown_actions_init(void)
 	int rc;
 
 	shutdown_actions_kset = kset_create_and_add("shutdown_actions", NULL,
-						    &firmware_kset->kobj);
+						    firmware_kobj);
 	if (!shutdown_actions_kset)
 		return -ENOMEM;
 	rc = sysfs_create_file(&shutdown_actions_kset->kobj, &on_panic_attr);
diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index e550da684a47..1b4cf984b081 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -755,7 +755,7 @@ static int __init acpi_init(void)
 		return -ENODEV;
 	}
 
-	acpi_kobj = kobject_create_and_add("acpi", &firmware_kset->kobj);
+	acpi_kobj = kobject_create_and_add("acpi", firmware_kobj);
 	if (!acpi_kobj) {
 		printk(KERN_WARNING "%s: kset create error\n", __FUNCTION__);
 		acpi_kobj = NULL;
diff --git a/drivers/base/firmware.c b/drivers/base/firmware.c
index 9efff481f5dc..113815556809 100644
--- a/drivers/base/firmware.c
+++ b/drivers/base/firmware.c
@@ -15,13 +15,13 @@
 
 #include "base.h"
 
-struct kset *firmware_kset;
-EXPORT_SYMBOL_GPL(firmware_kset);
+struct kobject *firmware_kobj;
+EXPORT_SYMBOL_GPL(firmware_kobj);
 
 int __init firmware_init(void)
 {
-	firmware_kset = kset_create_and_add("firmware", NULL, NULL);
-	if (!firmware_kset)
+	firmware_kobj = kobject_create_and_add("firmware", NULL);
+	if (!firmware_kobj)
 		return -ENOMEM;
 	return 0;
 }
diff --git a/drivers/firmware/edd.c b/drivers/firmware/edd.c
index f07f37047cd1..ddcc95793068 100644
--- a/drivers/firmware/edd.c
+++ b/drivers/firmware/edd.c
@@ -756,7 +756,7 @@ edd_init(void)
 		return 1;
 	}
 
-	edd_kset = kset_create_and_add("edd", NULL, &firmware_kset->kobj);
+	edd_kset = kset_create_and_add("edd", NULL, firmware_kobj);
 	if (!edd_kset)
 		return -ENOMEM;
 
diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c
index e17cd8133548..d1ad48190afc 100644
--- a/drivers/firmware/efivars.c
+++ b/drivers/firmware/efivars.c
@@ -668,7 +668,7 @@ efivars_init(void)
 	/*
 	 * For now we'll register the efi subsys within this driver
 	 */
-	efi_kset = kset_create_and_add("efi", NULL, &firmware_kset->kobj);
+	efi_kset = kset_create_and_add("efi", NULL, firmware_kobj);
 	if (!efi_kset) {
 		printk(KERN_ERR "efivars: Firmware registration failed.\n");
 		error = -ENOMEM;
diff --git a/drivers/parisc/pdc_stable.c b/drivers/parisc/pdc_stable.c
index 444483405abf..ef1a353e5540 100644
--- a/drivers/parisc/pdc_stable.c
+++ b/drivers/parisc/pdc_stable.c
@@ -1059,7 +1059,7 @@ pdc_stable_init(void)
 	pdcs_osid = (u16)(result >> 16);
 
 	/* For now we'll register the stable kset within this driver */
-	stable_kset = kset_create_and_add("stable", NULL, &firmware_kset->kobj);
+	stable_kset = kset_create_and_add("stable", NULL, firmware_kobj);
 	if (!stable_kset) {
 		rc = -ENOMEM;
 		goto fail_firmreg;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 29841bb5badb..673623f18464 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -210,8 +210,8 @@ extern struct kset *kernel_kset;
 extern struct kobject *hypervisor_kobj;
 /* The global /sys/power/ kset for people to chain off of */
 extern struct kset *power_kset;
-/* The global /sys/firmware/ kset for people to chain off of */
-extern struct kset *firmware_kset;
+/* The global /sys/firmware/ kobject for people to chain off of */
+extern struct kobject *firmware_kobj;
 
 extern int __must_check subsystem_register(struct kset *);
 extern void subsystem_unregister(struct kset *);
-- 
cgit v1.2.3


From 5c03c7ab886859eb195440dbb6ccb8c30c4e84cc Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 2 Nov 2007 16:19:59 -0700
Subject: kset: remove decl_subsys macro

This macro is no longer used.  ksets should be created dynamically with
a call to kset_create_and_add() not declared statically.

Yes, there are 5 remaining static struct kset usages in the kernel tree,
but they will be fixed up soon.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c    | 11 +++++++++--
 include/linux/kobject.h |  6 ------
 2 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index d8a92c650b43..304f90eb9b00 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -453,8 +453,15 @@ static struct kset_uevent_ops class_uevent_ops = {
 	.uevent =	class_uevent,
 };
 
-static decl_subsys(class_obj, &class_uevent_ops);
-
+/*
+ * DO NOT copy how this is created, kset_create_and_add() should be
+ * called, but this is a hold-over from the old-way and will be deleted
+ * entirely soon.
+ */
+static struct kset class_obj_subsys = {
+	.kobj = { .k_name = "class_obj", },
+	.uevent_ops = &class_uevent_ops,
+};
 
 static int class_device_add_attrs(struct class_device * cd)
 {
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 673623f18464..9da3523e4a65 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -198,12 +198,6 @@ extern struct kobject * kset_find_obj(struct kset *, const char *);
 #define set_kset_name(str)	.kset = { .kobj = { .k_name = str } }
 
 
-#define decl_subsys(_name,_uevent_ops) \
-struct kset _name##_subsys = { \
-	.kobj = { .k_name = __stringify(_name) }, \
-	.uevent_ops =_uevent_ops, \
-}
-
 /* The global /sys/kernel/ kset for people to chain off of */
 extern struct kset *kernel_kset;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
-- 
cgit v1.2.3


From 0ff21e46630abce11fdaaffabd72bbd4eed5ac2c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 6 Nov 2007 10:36:58 -0800
Subject: kobject: convert kernel_kset to be a kobject

kernel_kset does not need to be a kset, but a much simpler kobject now
that we have kobj_attributes.

We also rename kernel_kset to kernel_kobj to catch all users of this
symbol with a build error instead of an easy-to-ignore build warning.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/configfs/mount.c         |  2 +-
 fs/debugfs/inode.c          |  2 +-
 fs/dlm/lockspace.c          |  2 +-
 fs/gfs2/locking/dlm/sysfs.c |  2 +-
 include/linux/kobject.h     |  4 ++--
 kernel/ksysfs.c             | 18 +++++++++---------
 kernel/user.c               |  2 +-
 mm/slub.c                   |  3 +--
 security/inode.c            |  2 +-
 9 files changed, 18 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index c4ee7f05de8b..54bf0db0d4b0 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -140,7 +140,7 @@ static int __init configfs_init(void)
 	if (!configfs_dir_cachep)
 		goto out;
 
-	config_kobj = kobject_create_and_add("config", &kernel_kset->kobj);
+	config_kobj = kobject_create_and_add("config", kernel_kobj);
 	if (!config_kobj) {
 		kmem_cache_destroy(configfs_dir_cachep);
 		configfs_dir_cachep = NULL;
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 5ce92c3d3b59..97f6381c36c2 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -432,7 +432,7 @@ static int __init debugfs_init(void)
 {
 	int retval;
 
-	debug_kobj = kobject_create_and_add("debug", &kernel_kset->kobj);
+	debug_kobj = kobject_create_and_add("debug", kernel_kobj);
 	if (!debug_kobj)
 		return -EINVAL;
 
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 0828beb2d35d..e64b0dc664f3 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -223,7 +223,7 @@ int dlm_lockspace_init(void)
 	INIT_LIST_HEAD(&lslist);
 	spin_lock_init(&lslist_lock);
 
-	dlm_kset = kset_create_and_add("dlm", NULL, &kernel_kset->kobj);
+	dlm_kset = kset_create_and_add("dlm", NULL, kernel_kobj);
 	if (!dlm_kset) {
 		printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__);
 		return -ENOMEM;
diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c
index 1a92b6f7bc10..e5a4fbf7265f 100644
--- a/fs/gfs2/locking/dlm/sysfs.c
+++ b/fs/gfs2/locking/dlm/sysfs.c
@@ -219,7 +219,7 @@ void gdlm_kobject_release(struct gdlm_ls *ls)
 
 int gdlm_sysfs_init(void)
 {
-	gdlm_kset = kset_create_and_add("lock_dlm", NULL, &kernel_kset->kobj);
+	gdlm_kset = kset_create_and_add("lock_dlm", NULL, kernel_kobj);
 	if (!gdlm_kset) {
 		printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__);
 		return -ENOMEM;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 9da3523e4a65..0930efdcc094 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -198,8 +198,8 @@ extern struct kobject * kset_find_obj(struct kset *, const char *);
 #define set_kset_name(str)	.kset = { .kobj = { .k_name = str } }
 
 
-/* The global /sys/kernel/ kset for people to chain off of */
-extern struct kset *kernel_kset;
+/* The global /sys/kernel/ kobject for people to chain off of */
+extern struct kobject *kernel_kobj;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
 extern struct kobject *hypervisor_kobj;
 /* The global /sys/power/ kset for people to chain off of */
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index 45e646585605..1081aff5fb9e 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -101,8 +101,8 @@ static struct bin_attribute notes_attr = {
 	.read = &notes_read,
 };
 
-struct kset *kernel_kset;
-EXPORT_SYMBOL_GPL(kernel_kset);
+struct kobject *kernel_kobj;
+EXPORT_SYMBOL_GPL(kernel_kobj);
 
 static struct attribute * kernel_attrs[] = {
 #if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
@@ -125,18 +125,18 @@ static int __init ksysfs_init(void)
 {
 	int error;
 
-	kernel_kset = kset_create_and_add("kernel", NULL, NULL);
-	if (!kernel_kset) {
+	kernel_kobj = kobject_create_and_add("kernel", NULL);
+	if (!kernel_kobj) {
 		error = -ENOMEM;
 		goto exit;
 	}
-	error = sysfs_create_group(&kernel_kset->kobj, &kernel_attr_group);
+	error = sysfs_create_group(kernel_kobj, &kernel_attr_group);
 	if (error)
 		goto kset_exit;
 
 	if (notes_size > 0) {
 		notes_attr.size = notes_size;
-		error = sysfs_create_bin_file(&kernel_kset->kobj, &notes_attr);
+		error = sysfs_create_bin_file(kernel_kobj, &notes_attr);
 		if (error)
 			goto group_exit;
 	}
@@ -150,11 +150,11 @@ static int __init ksysfs_init(void)
 
 notes_exit:
 	if (notes_size > 0)
-		sysfs_remove_bin_file(&kernel_kset->kobj, &notes_attr);
+		sysfs_remove_bin_file(kernel_kobj, &notes_attr);
 group_exit:
-	sysfs_remove_group(&kernel_kset->kobj, &kernel_attr_group);
+	sysfs_remove_group(kernel_kobj, &kernel_attr_group);
 kset_exit:
-	kset_unregister(kernel_kset);
+	kobject_unregister(kernel_kobj);
 exit:
 	return error;
 }
diff --git a/kernel/user.c b/kernel/user.c
index 5a106f3fdf05..7f17e6e8fd65 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -201,7 +201,7 @@ done:
  */
 int __init uids_sysfs_init(void)
 {
-	uids_kset = kset_create_and_add("uids", NULL, &kernel_kset->kobj);
+	uids_kset = kset_create_and_add("uids", NULL, kernel_kobj);
 	if (!uids_kset)
 		return -ENOMEM;
 
diff --git a/mm/slub.c b/mm/slub.c
index b6c79462157e..d26177fb293b 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -4091,8 +4091,7 @@ static int __init slab_sysfs_init(void)
 	struct kmem_cache *s;
 	int err;
 
-	slab_kset = kset_create_and_add("slab", &slab_uevent_ops,
-					&kernel_kset->kobj);
+	slab_kset = kset_create_and_add("slab", &slab_uevent_ops, kernel_kobj);
 	if (!slab_kset) {
 		printk(KERN_ERR "Cannot register slab subsystem.\n");
 		return -ENOSYS;
diff --git a/security/inode.c b/security/inode.c
index dbe040ac0549..def0cc1b07f2 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -321,7 +321,7 @@ static int __init securityfs_init(void)
 {
 	int retval;
 
-  	security_kobj = kobject_create_and_add("security", &kernel_kset->kobj);
+	security_kobj = kobject_create_and_add("security", kernel_kobj);
 	if (!security_kobj)
 		return -EINVAL;
 
-- 
cgit v1.2.3


From 2fb9113b974c3c7c43e76647bd5077238e274e1c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 6 Nov 2007 15:03:30 -0800
Subject: kobject: remove subsystem_(un)register functions

These functions are no longer used and are the last remants of the old
subsystem crap.  So delete them for good.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c      |  6 +++---
 drivers/base/class.c    |  4 ++--
 include/linux/kobject.h |  3 ---
 lib/kobject.c           | 13 -------------
 4 files changed, 5 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 6796d3e46053..871607b7c876 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -854,7 +854,7 @@ int bus_register(struct bus_type * bus)
 	bus->subsys.kobj.kset = bus_kset;
 	bus->subsys.kobj.ktype = &bus_ktype;
 
-	retval = subsystem_register(&bus->subsys);
+	retval = kset_register(&bus->subsys);
 	if (retval)
 		goto out;
 
@@ -900,7 +900,7 @@ bus_drivers_fail:
 bus_devices_fail:
 	bus_remove_file(bus, &bus_attr_uevent);
 bus_uevent_fail:
-	subsystem_unregister(&bus->subsys);
+	kset_unregister(&bus->subsys);
 out:
 	return retval;
 }
@@ -920,7 +920,7 @@ void bus_unregister(struct bus_type * bus)
 	kset_unregister(bus->drivers_kset);
 	kset_unregister(bus->devices_kset);
 	bus_remove_file(bus, &bus_attr_uevent);
-	subsystem_unregister(&bus->subsys);
+	kset_unregister(&bus->subsys);
 }
 
 int bus_register_notifier(struct bus_type *bus, struct notifier_block *nb)
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 304f90eb9b00..3ffcda753e78 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -152,7 +152,7 @@ int class_register(struct class * cls)
 	cls->subsys.kobj.kset = class_kset;
 	cls->subsys.kobj.ktype = &class_ktype;
 
-	error = subsystem_register(&cls->subsys);
+	error = kset_register(&cls->subsys);
 	if (!error) {
 		error = add_class_attrs(class_get(cls));
 		class_put(cls);
@@ -164,7 +164,7 @@ void class_unregister(struct class * cls)
 {
 	pr_debug("device class '%s': unregistering\n", cls->name);
 	remove_class_attrs(cls);
-	subsystem_unregister(&cls->subsys);
+	kset_unregister(&cls->subsys);
 }
 
 static void class_create_release(struct class *cls)
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 0930efdcc094..78c851b4e67e 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -207,9 +207,6 @@ extern struct kset *power_kset;
 /* The global /sys/firmware/ kobject for people to chain off of */
 extern struct kobject *firmware_kobj;
 
-extern int __must_check subsystem_register(struct kset *);
-extern void subsystem_unregister(struct kset *);
-
 #if defined(CONFIG_HOTPLUG)
 int kobject_uevent(struct kobject *kobj, enum kobject_action action);
 int kobject_uevent_env(struct kobject *kobj, enum kobject_action action,
diff --git a/lib/kobject.c b/lib/kobject.c
index c742ac25228a..7919c32a3a15 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -800,16 +800,6 @@ struct kobject * kset_find_obj(struct kset * kset, const char * name)
 	return ret;
 }
 
-int subsystem_register(struct kset *s)
-{
-	return kset_register(s);
-}
-
-void subsystem_unregister(struct kset *s)
-{
-	kset_unregister(s);
-}
-
 static void kset_release(struct kobject *kobj)
 {
 	struct kset *kset = container_of(kobj, struct kset, kobj);
@@ -904,6 +894,3 @@ EXPORT_SYMBOL(kobject_del);
 
 EXPORT_SYMBOL(kset_register);
 EXPORT_SYMBOL(kset_unregister);
-
-EXPORT_SYMBOL(subsystem_register);
-EXPORT_SYMBOL(subsystem_unregister);
-- 
cgit v1.2.3


From d76e15fb20eeb7632ef38876a884fe3508b2c01d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 27 Nov 2007 11:28:26 -0800
Subject: driver core: make /sys/power a kobject

/sys/power should not be a kset, that's overkill.  This patch renames it
to power_kset and fixes up all usages of it in the tree.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/mach-omap1/pm.c               |  2 +-
 arch/powerpc/platforms/pseries/power.c | 10 +++++-----
 include/linux/kobject.h                |  4 ++--
 kernel/power/disk.c                    |  2 +-
 kernel/power/main.c                    |  8 ++++----
 5 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-omap1/pm.c b/arch/arm/mach-omap1/pm.c
index 63edafb59490..d9805e3d9304 100644
--- a/arch/arm/mach-omap1/pm.c
+++ b/arch/arm/mach-omap1/pm.c
@@ -719,7 +719,7 @@ static int __init omap_pm_init(void)
 	omap_pm_init_proc();
 #endif
 
-	error = sysfs_create_file(&power_kset->kobj, &sleep_while_idle_attr);
+	error = sysfs_create_file(power_kobj, &sleep_while_idle_attr);
 	if (error)
 		printk(KERN_ERR "sysfs_create_file failed: %d\n", error);
 
diff --git a/arch/powerpc/platforms/pseries/power.c b/arch/powerpc/platforms/pseries/power.c
index 90706cf840fd..e95fc1594c84 100644
--- a/arch/powerpc/platforms/pseries/power.c
+++ b/arch/powerpc/platforms/pseries/power.c
@@ -53,7 +53,7 @@ static struct kobj_attribute auto_poweron_attr =
 	__ATTR(auto_poweron, 0644, auto_poweron_show, auto_poweron_store);
 
 #ifndef CONFIG_PM
-struct kset *power_kset;
+struct kobject *power_kobj;
 
 static struct attribute *g[] = {
         &auto_poweron_attr.attr,
@@ -66,16 +66,16 @@ static struct attribute_group attr_group = {
 
 static int __init pm_init(void)
 {
-	power_kset = kset_create_and_add("power", NULL, NULL);
-	if (!power_kset)
+	power_kobj = kobject_create_and_add("power", NULL);
+	if (!power_kobj)
 		return -ENOMEM;
-	return sysfs_create_group(&power_kset->kobj, &attr_group);
+	return sysfs_create_group(power_kobj, &attr_group);
 }
 core_initcall(pm_init);
 #else
 static int __init apo_pm_init(void)
 {
-	return (sysfs_create_file(&power_kset->kobj, &auto_poweron_attr));
+	return (sysfs_create_file(power_kobj, &auto_poweron_attr));
 }
 __initcall(apo_pm_init);
 #endif
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 78c851b4e67e..bb6868475edb 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -202,8 +202,8 @@ extern struct kobject * kset_find_obj(struct kset *, const char *);
 extern struct kobject *kernel_kobj;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
 extern struct kobject *hypervisor_kobj;
-/* The global /sys/power/ kset for people to chain off of */
-extern struct kset *power_kset;
+/* The global /sys/power/ kobject for people to chain off of */
+extern struct kobject *power_kobj;
 /* The global /sys/firmware/ kobject for people to chain off of */
 extern struct kobject *firmware_kobj;
 
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index ef5aa2ca0ab0..b138b431e271 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -714,7 +714,7 @@ static struct attribute_group attr_group = {
 
 static int __init pm_disk_init(void)
 {
-	return sysfs_create_group(&power_kset->kobj, &attr_group);
+	return sysfs_create_group(power_kobj, &attr_group);
 }
 
 core_initcall(pm_disk_init);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index b8139493b856..efc08360e627 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -276,7 +276,7 @@ EXPORT_SYMBOL(pm_suspend);
 
 #endif /* CONFIG_SUSPEND */
 
-struct kset *power_kset;
+struct kobject *power_kobj;
 
 /**
  *	state - control system power state.
@@ -389,10 +389,10 @@ static struct attribute_group attr_group = {
 
 static int __init pm_init(void)
 {
-	power_kset = kset_create_and_add("power", NULL, NULL);
-	if (!power_kset)
+	power_kobj = kobject_create_and_add("power", NULL);
+	if (!power_kobj)
 		return -ENOMEM;
-	return sysfs_create_group(&power_kset->kobj, &attr_group);
+	return sysfs_create_group(power_kobj, &attr_group);
 }
 
 core_initcall(pm_init);
-- 
cgit v1.2.3


From 81e7c6a636c81d9eeaeaa732bfbace44535fab00 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 4 Dec 2007 22:41:54 +0000
Subject: UIO: fix kobject usage

The uio kobject code is "wierd".  This patch should hopefully fix it up
to be sane and not leak memory anymore.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/uio/uio.c          | 91 ++++++++++++++++++++++++----------------------
 include/linux/uio_driver.h |  6 ++-
 2 files changed, 52 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 606aae7490ab..acc387de988f 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -34,7 +34,7 @@ struct uio_device {
 	wait_queue_head_t	wait;
 	int			vma_count;
 	struct uio_info		*info;
-	struct kset 		map_attr_kset;
+	struct kobject		*map_dir;
 };
 
 static int uio_major;
@@ -51,47 +51,48 @@ static struct uio_class {
  * attributes
  */
 
-static struct attribute attr_addr = {
-	.name  = "addr",
-	.mode  = S_IRUGO,
+struct uio_map {
+	struct kobject kobj;
+	struct uio_mem *mem;
 };
+#define to_map(map) container_of(map, struct uio_map, kobj)
 
-static struct attribute attr_size = {
-	.name  = "size",
-	.mode  = S_IRUGO,
-};
 
-static struct attribute* map_attrs[] = {
-	&attr_addr, &attr_size, NULL
-};
-
-static ssize_t map_attr_show(struct kobject *kobj, struct attribute *attr,
+static ssize_t map_attr_show(struct kobject *kobj, struct kobj_attribute *attr,
 			     char *buf)
 {
-	struct uio_mem *mem = container_of(kobj, struct uio_mem, kobj);
+	struct uio_map *map = to_map(kobj);
+	struct uio_mem *mem = map->mem;
 
-	if (strncmp(attr->name,"addr",4) == 0)
+	if (strncmp(attr->attr.name, "addr", 4) == 0)
 		return sprintf(buf, "0x%lx\n", mem->addr);
 
-	if (strncmp(attr->name,"size",4) == 0)
+	if (strncmp(attr->attr.name, "size", 4) == 0)
 		return sprintf(buf, "0x%lx\n", mem->size);
 
 	return -ENODEV;
 }
 
-static void map_attr_release(struct kobject *kobj)
-{
-	/* TODO ??? */
-}
+static struct kobj_attribute attr_attribute =
+	__ATTR(addr, S_IRUGO, map_attr_show, NULL);
+static struct kobj_attribute size_attribute =
+	__ATTR(size, S_IRUGO, map_attr_show, NULL);
 
-static struct sysfs_ops map_attr_ops = {
-	.show  = map_attr_show,
+static struct attribute *attrs[] = {
+	&attr_attribute.attr,
+	&size_attribute.attr,
+	NULL,	/* need to NULL terminate the list of attributes */
 };
 
+static void map_release(struct kobject *kobj)
+{
+	struct uio_map *map = to_map(kobj);
+	kfree(map);
+}
+
 static struct kobj_type map_attr_type = {
-	.release	= map_attr_release,
-	.sysfs_ops	= &map_attr_ops,
-	.default_attrs	= map_attrs,
+	.release	= map_release,
+	.default_attrs	= attrs,
 };
 
 static ssize_t show_name(struct device *dev,
@@ -148,6 +149,7 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 	int mi;
 	int map_found = 0;
 	struct uio_mem *mem;
+	struct uio_map *map;
 
 	ret = sysfs_create_group(&idev->dev->kobj, &uio_attr_grp);
 	if (ret)
@@ -159,31 +161,34 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 			break;
 		if (!map_found) {
 			map_found = 1;
-			kobject_set_name(&idev->map_attr_kset.kobj,"maps");
-			idev->map_attr_kset.kobj.ktype = &map_attr_type;
-			idev->map_attr_kset.kobj.parent = &idev->dev->kobj;
-			ret = kset_register(&idev->map_attr_kset);
-			if (ret)
-				goto err_remove_group;
+			idev->map_dir = kobject_create_and_add("maps",
+							&idev->dev->kobj);
+			if (!idev->map_dir)
+				goto err;
 		}
-		kobject_init(&mem->kobj);
-		kobject_set_name(&mem->kobj,"map%d",mi);
-		mem->kobj.parent = &idev->map_attr_kset.kobj;
-		mem->kobj.kset = &idev->map_attr_kset;
-		ret = kobject_add(&mem->kobj);
+		map = kzalloc(sizeof(*map), GFP_KERNEL);
+		if (!map)
+			goto err;
+		kobject_init_ng(&map->kobj, &map_attr_type);
+		map->mem = mem;
+		mem->map = map;
+		ret = kobject_add_ng(&map->kobj, idev->map_dir, "map%d", mi);
+		if (ret)
+			goto err;
+		ret = kobject_uevent(&map->kobj, KOBJ_ADD);
 		if (ret)
-			goto err_remove_maps;
+			goto err;
 	}
 
 	return 0;
 
-err_remove_maps:
+err:
 	for (mi--; mi>=0; mi--) {
 		mem = &idev->info->mem[mi];
-		kobject_unregister(&mem->kobj);
+		map = mem->map;
+		kobject_unregister(&map->kobj);
 	}
-	kset_unregister(&idev->map_attr_kset); /* Needed ? */
-err_remove_group:
+	kobject_unregister(idev->map_dir);
 	sysfs_remove_group(&idev->dev->kobj, &uio_attr_grp);
 err_group:
 	dev_err(idev->dev, "error creating sysfs files (%d)\n", ret);
@@ -198,9 +203,9 @@ static void uio_dev_del_attributes(struct uio_device *idev)
 		mem = &idev->info->mem[mi];
 		if (mem->size == 0)
 			break;
-		kobject_unregister(&mem->kobj);
+		kobject_unregister(&mem->map->kobj);
 	}
-	kset_unregister(&idev->map_attr_kset);
+	kobject_unregister(idev->map_dir);
 	sysfs_remove_group(&idev->dev->kobj, &uio_attr_grp);
 }
 
diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index 44c28e94df50..973386d439da 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -18,20 +18,22 @@
 #include <linux/fs.h>
 #include <linux/interrupt.h>
 
+struct uio_map;
+
 /**
  * struct uio_mem - description of a UIO memory region
- * @kobj:		kobject for this mapping
  * @addr:		address of the device's memory
  * @size:		size of IO
  * @memtype:		type of memory addr points to
  * @internal_addr:	ioremap-ped version of addr, for driver internal use
+ * @map:		for use by the UIO core only.
  */
 struct uio_mem {
-	struct kobject		kobj;
 	unsigned long		addr;
 	unsigned long		size;
 	int			memtype;
 	void __iomem		*internal_addr;
+	struct uio_map		*map;
 };
 
 #define MAX_UIO_MAPS 	5
-- 
cgit v1.2.3


From 35be0681983752116a8161ad3b30e830963108a4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 17 Dec 2007 15:54:39 -0400
Subject: Kobject: change drivers/infiniband to use kobject_init_and_add

Stop using kobject_register, as this way we can control the sending of
the uevent properly, after everything is properly initialized.

Cc: Roland Dreier <rolandd@cisco.com>
Cc: Sean Hefty <mshefty@ichips.intel.com>
Cc: Hal Rosenstock <hal.rosenstock@gmail.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/infiniband/core/sysfs.c | 35 +++++++++--------------------------
 include/rdma/ib_verbs.h         |  2 +-
 2 files changed, 10 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index 3d4050681325..aa81129704cf 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -508,19 +508,10 @@ static int add_port(struct ib_device *device, int port_num)
 
 	p->ibdev      = device;
 	p->port_num   = port_num;
-	p->kobj.ktype = &port_type;
 
-	p->kobj.parent = kobject_get(&device->ports_parent);
-	if (!p->kobj.parent) {
-		ret = -EBUSY;
-		goto err;
-	}
-
-	ret = kobject_set_name(&p->kobj, "%d", port_num);
-	if (ret)
-		goto err_put;
-
-	ret = kobject_register(&p->kobj);
+	ret = kobject_init_and_add(&p->kobj, &port_type,
+				   kobject_get(device->ports_parent),
+				   "%d", port_num);
 	if (ret)
 		goto err_put;
 
@@ -549,6 +540,7 @@ static int add_port(struct ib_device *device, int port_num)
 
 	list_add_tail(&p->kobj.entry, &device->port_list);
 
+	kobject_uevent(&p->kobj, KOBJ_ADD);
 	return 0;
 
 err_free_pkey:
@@ -570,9 +562,7 @@ err_remove_pma:
 	sysfs_remove_group(&p->kobj, &pma_group);
 
 err_put:
-	kobject_put(&device->ports_parent);
-
-err:
+	kobject_put(device->ports_parent);
 	kfree(p);
 	return ret;
 }
@@ -694,16 +684,9 @@ int ib_device_register_sysfs(struct ib_device *device)
 			goto err_unregister;
 	}
 
-	device->ports_parent.parent = kobject_get(&class_dev->kobj);
-	if (!device->ports_parent.parent) {
-		ret = -EBUSY;
-		goto err_unregister;
-	}
-	ret = kobject_set_name(&device->ports_parent, "ports");
-	if (ret)
-		goto err_put;
-	ret = kobject_register(&device->ports_parent);
-	if (ret)
+	device->ports_parent = kobject_create_and_add("ports",
+					kobject_get(&class_dev->kobj));
+	if (!device->ports_parent)
 		goto err_put;
 
 	if (device->node_type == RDMA_NODE_IB_SWITCH) {
@@ -758,7 +741,7 @@ void ib_device_unregister_sysfs(struct ib_device *device)
 		kobject_unregister(p);
 	}
 
-	kobject_unregister(&device->ports_parent);
+	kobject_unregister(device->ports_parent);
 	class_device_unregister(&device->class_dev);
 }
 
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 11f39606e7d9..cfbd38fe2998 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -1026,7 +1026,7 @@ struct ib_device {
 
 	struct module               *owner;
 	struct class_device          class_dev;
-	struct kobject               ports_parent;
+	struct kobject               *ports_parent;
 	struct list_head             port_list;
 
 	enum {
-- 
cgit v1.2.3


From cc972e896b303f453f5893ecf8eca0d0e395ab64 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 13:31:26 -0700
Subject: driver core: remove owner field from struct bus_type

This isn't used by anything in the driver core, and by no one in the 204
different usages of it in the kernel tree.  Remove this field so no one
gets any idea that it is needed to be used.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/device.h b/include/linux/device.h
index a3b3ff15fc87..313e0b32bc09 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -51,7 +51,6 @@ extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct bus_type {
 	const char		* name;
-	struct module		* owner;
 
 	struct kset		subsys;
 	struct kset		*drivers_kset;
-- 
cgit v1.2.3


From 0fed80f7a63abd7168907267af69ee31f6bcf301 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 19:41:16 -0700
Subject: driver core: add way to get to bus kset

This allows an easier way to get to the kset associated with a struct
bus_type (you have three to choose from...)  This will make it easier to
move these fields to be dynamic in a future patch.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c                     | 6 ++++++
 drivers/pci/hotplug/pci_hotplug_core.c | 5 ++++-
 include/linux/device.h                 | 2 ++
 3 files changed, 12 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 871607b7c876..8335a1079b0e 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -935,6 +935,12 @@ int bus_unregister_notifier(struct bus_type *bus, struct notifier_block *nb)
 }
 EXPORT_SYMBOL_GPL(bus_unregister_notifier);
 
+struct kset *bus_get_kset(struct bus_type *bus)
+{
+	return &bus->subsys;
+}
+EXPORT_SYMBOL_GPL(bus_get_kset);
+
 int __init buses_init(void)
 {
 	bus_kset = kset_create_and_add("bus", &bus_uevent_ops, NULL);
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index 0f05e6a68b3b..3606d5b52a70 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -699,9 +699,12 @@ int __must_check pci_hp_change_slot_info(struct hotplug_slot *slot,
 static int __init pci_hotplug_init (void)
 {
 	int result;
+	struct kset *pci_bus_kset;
+
+	pci_bus_kset = bus_get_kset(&pci_bus_type);
 
 	pci_hotplug_slots_kset = kset_create_and_add("slots", NULL,
-						     &pci_bus_type.subsys.kobj);
+						     &pci_bus_kset->kobj);
 	if (!pci_hotplug_slots_kset) {
 		result = -ENOMEM;
 		err("Register subsys error\n");
diff --git a/include/linux/device.h b/include/linux/device.h
index 313e0b32bc09..3cc13c32314a 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -117,6 +117,8 @@ extern int bus_unregister_notifier(struct bus_type *bus,
 #define BUS_NOTIFY_UNBIND_DRIVER	0x00000004 /* driver about to be
 						      unbound */
 
+extern struct kset *bus_get_kset(struct bus_type *bus);
+
 struct device_driver {
 	const char		* name;
 	struct bus_type		* bus;
-- 
cgit v1.2.3


From b249072ee6897fe4f8d461c7bb4b926223263c28 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 19:41:16 -0700
Subject: driver core: add way to get to bus device klist

This allows an easier way to get to the device klist associated with a
struct bus_type (you have three to choose from...)  This will make it
easier to move these fields to be dynamic in a future patch.

The only user of this is the PCI core which horribly abuses this
interface to rearrange the order of the pci devices.  This should be
done using the existing bus device walking functions, but that's left
for future patches.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     |  6 ++++++
 drivers/pci/probe.c    | 11 +++++++----
 include/linux/device.h |  1 +
 3 files changed, 14 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 8335a1079b0e..9c9027b2c44e 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -941,6 +941,12 @@ struct kset *bus_get_kset(struct bus_type *bus)
 }
 EXPORT_SYMBOL_GPL(bus_get_kset);
 
+struct klist *bus_get_device_klist(struct bus_type *bus)
+{
+	return &bus->klist_devices;
+}
+EXPORT_SYMBOL_GPL(bus_get_device_klist);
+
 int __init buses_init(void)
 {
 	bus_kset = kset_create_and_add("bus", &bus_uevent_ops, NULL);
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index c5ca3134513a..5fd585293e79 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -1210,16 +1210,19 @@ static void __init pci_sort_breadthfirst_klist(void)
 	struct klist_node *n;
 	struct device *dev;
 	struct pci_dev *pdev;
+	struct klist *device_klist;
 
-	spin_lock(&pci_bus_type.klist_devices.k_lock);
-	list_for_each_safe(pos, tmp, &pci_bus_type.klist_devices.k_list) {
+	device_klist = bus_get_device_klist(&pci_bus_type);
+
+	spin_lock(&device_klist->k_lock);
+	list_for_each_safe(pos, tmp, &device_klist->k_list) {
 		n = container_of(pos, struct klist_node, n_node);
 		dev = container_of(n, struct device, knode_bus);
 		pdev = to_pci_dev(dev);
 		pci_insertion_sort_klist(pdev, &sorted_devices);
 	}
-	list_splice(&sorted_devices, &pci_bus_type.klist_devices.k_list);
-	spin_unlock(&pci_bus_type.klist_devices.k_lock);
+	list_splice(&sorted_devices, &device_klist->k_list);
+	spin_unlock(&device_klist->k_lock);
 }
 
 static void __init pci_insertion_sort_devices(struct pci_dev *a, struct list_head *list)
diff --git a/include/linux/device.h b/include/linux/device.h
index 3cc13c32314a..62e695bd3c93 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -118,6 +118,7 @@ extern int bus_unregister_notifier(struct bus_type *bus,
 						      unbound */
 
 extern struct kset *bus_get_kset(struct bus_type *bus);
+extern struct klist *bus_get_device_klist(struct bus_type *bus);
 
 struct device_driver {
 	const char		* name;
-- 
cgit v1.2.3


From c6f7e72a3f4641095ade9ded287d910c980c6148 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 1 Nov 2007 19:41:16 -0700
Subject: driver core: remove fields from struct bus_type

struct bus_type is static everywhere in the kernel.  This moves the
kobject in the structure out of it, and a bunch of other private only to
the driver core fields are now moved to a private structure.  This lets
us dynamically create the backing kobject properly and gives us the
chance to be able to document to users exactly how to use the struct
bus_type as there are no fields they can improperly access.

Thanks to Kay for the build fixes on this patch.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h     |  30 ++++++++++++-
 drivers/base/bus.c      | 116 +++++++++++++++++++++++++++---------------------
 drivers/base/core.c     |   6 +--
 drivers/base/dd.c       |   4 +-
 drivers/base/driver.c   |   2 +-
 drivers/base/platform.c |   4 +-
 include/linux/device.h  |  12 +----
 7 files changed, 104 insertions(+), 70 deletions(-)

(limited to 'include')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 7e309a49a711..ca6d273064f8 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -1,6 +1,34 @@
 
-/* initialisation functions */
+/**
+ * struct bus_type_private - structure to hold the private to the driver core portions of the bus_type structure.
+ *
+ * @subsys - the struct kset that defines this bus.  This is the main kobject
+ * @drivers_kset - the list of drivers associated with this bus
+ * @devices_kset - the list of devices associated with this bus
+ * @klist_devices - the klist to iterate over the @devices_kset
+ * @klist_drivers - the klist to iterate over the @drivers_kset
+ * @bus_notifier - the bus notifier list for anything that cares about things
+ * on this bus.
+ * @bus - pointer back to the struct bus_type that this structure is associated
+ * with.
+ *
+ * This structure is the one that is the actual kobject allowing struct
+ * bus_type to be statically allocated safely.  Nothing outside of the driver
+ * core should ever touch these fields.
+ */
+struct bus_type_private {
+	struct kset subsys;
+	struct kset *drivers_kset;
+	struct kset *devices_kset;
+	struct klist klist_devices;
+	struct klist klist_drivers;
+	struct blocking_notifier_head bus_notifier;
+	unsigned int drivers_autoprobe:1;
+	struct bus_type *bus;
+};
+
 
+/* initialisation functions */
 extern int devices_init(void);
 extern int buses_init(void);
 extern int classes_init(void);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 9c9027b2c44e..04d3850ff4b7 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -17,7 +17,7 @@
 #include "power/power.h"
 
 #define to_bus_attr(_attr) container_of(_attr, struct bus_attribute, attr)
-#define to_bus(obj) container_of(obj, struct bus_type, subsys.kobj)
+#define to_bus(obj) container_of(obj, struct bus_type_private, subsys.kobj)
 
 /*
  * sysfs bindings for drivers
@@ -32,13 +32,17 @@ static int __must_check bus_rescan_devices_helper(struct device *dev,
 
 static struct bus_type *bus_get(struct bus_type *bus)
 {
-	return bus ? container_of(kset_get(&bus->subsys),
-				struct bus_type, subsys) : NULL;
+	if (bus) {
+		kset_get(&bus->p->subsys);
+		return bus;
+	}
+	return NULL;
 }
 
 static void bus_put(struct bus_type *bus)
 {
-	kset_put(&bus->subsys);
+	if (bus)
+		kset_put(&bus->p->subsys);
 }
 
 static ssize_t
@@ -104,11 +108,11 @@ static ssize_t
 bus_attr_show(struct kobject * kobj, struct attribute * attr, char * buf)
 {
 	struct bus_attribute * bus_attr = to_bus_attr(attr);
-	struct bus_type * bus = to_bus(kobj);
+	struct bus_type_private *bus_priv = to_bus(kobj);
 	ssize_t ret = 0;
 
 	if (bus_attr->show)
-		ret = bus_attr->show(bus, buf);
+		ret = bus_attr->show(bus_priv->bus, buf);
 	return ret;
 }
 
@@ -117,11 +121,11 @@ bus_attr_store(struct kobject * kobj, struct attribute * attr,
 	       const char * buf, size_t count)
 {
 	struct bus_attribute * bus_attr = to_bus_attr(attr);
-	struct bus_type * bus = to_bus(kobj);
+	struct bus_type_private *bus_priv = to_bus(kobj);
 	ssize_t ret = 0;
 
 	if (bus_attr->store)
-		ret = bus_attr->store(bus, buf, count);
+		ret = bus_attr->store(bus_priv->bus, buf, count);
 	return ret;
 }
 
@@ -134,7 +138,7 @@ int bus_create_file(struct bus_type * bus, struct bus_attribute * attr)
 {
 	int error;
 	if (bus_get(bus)) {
-		error = sysfs_create_file(&bus->subsys.kobj, &attr->attr);
+		error = sysfs_create_file(&bus->p->subsys.kobj, &attr->attr);
 		bus_put(bus);
 	} else
 		error = -EINVAL;
@@ -144,7 +148,7 @@ int bus_create_file(struct bus_type * bus, struct bus_attribute * attr)
 void bus_remove_file(struct bus_type * bus, struct bus_attribute * attr)
 {
 	if (bus_get(bus)) {
-		sysfs_remove_file(&bus->subsys.kobj, &attr->attr);
+		sysfs_remove_file(&bus->p->subsys.kobj, &attr->attr);
 		bus_put(bus);
 	}
 }
@@ -237,16 +241,16 @@ static DRIVER_ATTR(bind, S_IWUSR, NULL, driver_bind);
 
 static ssize_t show_drivers_autoprobe(struct bus_type *bus, char *buf)
 {
-	return sprintf(buf, "%d\n", bus->drivers_autoprobe);
+	return sprintf(buf, "%d\n", bus->p->drivers_autoprobe);
 }
 
 static ssize_t store_drivers_autoprobe(struct bus_type *bus,
 				       const char *buf, size_t count)
 {
 	if (buf[0] == '0')
-		bus->drivers_autoprobe = 0;
+		bus->p->drivers_autoprobe = 0;
 	else
-		bus->drivers_autoprobe = 1;
+		bus->p->drivers_autoprobe = 1;
 	return count;
 }
 
@@ -300,7 +304,7 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start,
 	if (!bus)
 		return -EINVAL;
 
-	klist_iter_init_node(&bus->klist_devices, &i,
+	klist_iter_init_node(&bus->p->klist_devices, &i,
 			     (start ? &start->knode_bus : NULL));
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
@@ -333,7 +337,7 @@ struct device * bus_find_device(struct bus_type *bus,
 	if (!bus)
 		return NULL;
 
-	klist_iter_init_node(&bus->klist_devices, &i,
+	klist_iter_init_node(&bus->p->klist_devices, &i,
 			     (start ? &start->knode_bus : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
@@ -379,7 +383,7 @@ int bus_for_each_drv(struct bus_type * bus, struct device_driver * start,
 	if (!bus)
 		return -EINVAL;
 
-	klist_iter_init_node(&bus->klist_drivers, &i,
+	klist_iter_init_node(&bus->p->klist_drivers, &i,
 			     start ? &start->knode_bus : NULL);
 	while ((drv = next_driver(&i)) && !error)
 		error = fn(drv, data);
@@ -420,7 +424,7 @@ static void device_remove_attrs(struct bus_type * bus, struct device * dev)
 static int make_deprecated_bus_links(struct device *dev)
 {
 	return sysfs_create_link(&dev->kobj,
-				 &dev->bus->subsys.kobj, "bus");
+				 &dev->bus->p->subsys.kobj, "bus");
 }
 
 static void remove_deprecated_bus_links(struct device *dev)
@@ -449,12 +453,12 @@ int bus_add_device(struct device * dev)
 		error = device_add_attrs(bus, dev);
 		if (error)
 			goto out_put;
-		error = sysfs_create_link(&bus->devices_kset->kobj,
+		error = sysfs_create_link(&bus->p->devices_kset->kobj,
 						&dev->kobj, dev->bus_id);
 		if (error)
 			goto out_id;
 		error = sysfs_create_link(&dev->kobj,
-				&dev->bus->subsys.kobj, "subsystem");
+				&dev->bus->p->subsys.kobj, "subsystem");
 		if (error)
 			goto out_subsys;
 		error = make_deprecated_bus_links(dev);
@@ -466,7 +470,7 @@ int bus_add_device(struct device * dev)
 out_deprecated:
 	sysfs_remove_link(&dev->kobj, "subsystem");
 out_subsys:
-	sysfs_remove_link(&bus->devices_kset->kobj, dev->bus_id);
+	sysfs_remove_link(&bus->p->devices_kset->kobj, dev->bus_id);
 out_id:
 	device_remove_attrs(bus, dev);
 out_put:
@@ -488,11 +492,11 @@ void bus_attach_device(struct device * dev)
 
 	if (bus) {
 		dev->is_registered = 1;
-		if (bus->drivers_autoprobe)
+		if (bus->p->drivers_autoprobe)
 			ret = device_attach(dev);
 		WARN_ON(ret < 0);
 		if (ret >= 0)
-			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
+			klist_add_tail(&dev->knode_bus, &bus->p->klist_devices);
 		else
 			dev->is_registered = 0;
 	}
@@ -512,7 +516,7 @@ void bus_remove_device(struct device * dev)
 	if (dev->bus) {
 		sysfs_remove_link(&dev->kobj, "subsystem");
 		remove_deprecated_bus_links(dev);
-		sysfs_remove_link(&dev->bus->devices_kset->kobj, dev->bus_id);
+		sysfs_remove_link(&dev->bus->p->devices_kset->kobj, dev->bus_id);
 		device_remove_attrs(dev->bus, dev);
 		if (dev->is_registered) {
 			dev->is_registered = 0;
@@ -638,18 +642,18 @@ int bus_add_driver(struct device_driver *drv)
 	error = kobject_set_name(&drv->kobj, "%s", drv->name);
 	if (error)
 		goto out_put_bus;
-	drv->kobj.kset = bus->drivers_kset;
+	drv->kobj.kset = bus->p->drivers_kset;
 	drv->kobj.ktype = &driver_ktype;
 	error = kobject_register(&drv->kobj);
 	if (error)
 		goto out_put_bus;
 
-	if (drv->bus->drivers_autoprobe) {
+	if (drv->bus->p->drivers_autoprobe) {
 		error = driver_attach(drv);
 		if (error)
 			goto out_unregister;
 	}
-	klist_add_tail(&drv->knode_bus, &bus->klist_drivers);
+	klist_add_tail(&drv->knode_bus, &bus->p->klist_drivers);
 	module_add_driver(drv->owner, drv);
 
 	error = driver_create_file(drv, &driver_attr_uevent);
@@ -828,7 +832,7 @@ static ssize_t bus_uevent_store(struct bus_type *bus,
 	enum kobject_action action;
 
 	if (kobject_action_type(buf, count, &action) == 0)
-		kobject_uevent(&bus->subsys.kobj, action);
+		kobject_uevent(&bus->p->subsys.kobj, action);
 	return count;
 }
 static BUS_ATTR(uevent, S_IWUSR, NULL, bus_uevent_store);
@@ -844,17 +848,26 @@ static BUS_ATTR(uevent, S_IWUSR, NULL, bus_uevent_store);
 int bus_register(struct bus_type * bus)
 {
 	int retval;
+	struct bus_type_private *priv;
+
+	priv = kzalloc(sizeof(struct bus_type_private), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->bus = bus;
+	bus->p = priv;
 
-	BLOCKING_INIT_NOTIFIER_HEAD(&bus->bus_notifier);
+	BLOCKING_INIT_NOTIFIER_HEAD(&priv->bus_notifier);
 
-	retval = kobject_set_name(&bus->subsys.kobj, "%s", bus->name);
+	retval = kobject_set_name(&priv->subsys.kobj, "%s", bus->name);
 	if (retval)
 		goto out;
 
-	bus->subsys.kobj.kset = bus_kset;
-	bus->subsys.kobj.ktype = &bus_ktype;
+	priv->subsys.kobj.kset = bus_kset;
+	priv->subsys.kobj.ktype = &bus_ktype;
+	priv->drivers_autoprobe = 1;
 
-	retval = kset_register(&bus->subsys);
+	retval = kset_register(&priv->subsys);
 	if (retval)
 		goto out;
 
@@ -862,24 +875,23 @@ int bus_register(struct bus_type * bus)
 	if (retval)
 		goto bus_uevent_fail;
 
-	bus->devices_kset = kset_create_and_add("devices", NULL,
-						&bus->subsys.kobj);
-	if (!bus->devices_kset) {
+	priv->devices_kset = kset_create_and_add("devices", NULL,
+						 &priv->subsys.kobj);
+	if (!priv->devices_kset) {
 		retval = -ENOMEM;
 		goto bus_devices_fail;
 	}
 
-	bus->drivers_kset = kset_create_and_add("drivers", NULL,
-						&bus->subsys.kobj);
-	if (!bus->drivers_kset) {
+	priv->drivers_kset = kset_create_and_add("drivers", NULL,
+						 &priv->subsys.kobj);
+	if (!priv->drivers_kset) {
 		retval = -ENOMEM;
 		goto bus_drivers_fail;
 	}
 
-	klist_init(&bus->klist_devices, klist_devices_get, klist_devices_put);
-	klist_init(&bus->klist_drivers, NULL, NULL);
+	klist_init(&priv->klist_devices, klist_devices_get, klist_devices_put);
+	klist_init(&priv->klist_drivers, NULL, NULL);
 
-	bus->drivers_autoprobe = 1;
 	retval = add_probe_files(bus);
 	if (retval)
 		goto bus_probe_files_fail;
@@ -894,13 +906,14 @@ int bus_register(struct bus_type * bus)
 bus_attrs_fail:
 	remove_probe_files(bus);
 bus_probe_files_fail:
-	kset_unregister(bus->drivers_kset);
+	kset_unregister(bus->p->drivers_kset);
 bus_drivers_fail:
-	kset_unregister(bus->devices_kset);
+	kset_unregister(bus->p->devices_kset);
 bus_devices_fail:
 	bus_remove_file(bus, &bus_attr_uevent);
 bus_uevent_fail:
-	kset_unregister(&bus->subsys);
+	kset_unregister(&bus->p->subsys);
+	kfree(bus->p);
 out:
 	return retval;
 }
@@ -917,33 +930,34 @@ void bus_unregister(struct bus_type * bus)
 	pr_debug("bus %s: unregistering\n", bus->name);
 	bus_remove_attrs(bus);
 	remove_probe_files(bus);
-	kset_unregister(bus->drivers_kset);
-	kset_unregister(bus->devices_kset);
+	kset_unregister(bus->p->drivers_kset);
+	kset_unregister(bus->p->devices_kset);
 	bus_remove_file(bus, &bus_attr_uevent);
-	kset_unregister(&bus->subsys);
+	kset_unregister(&bus->p->subsys);
+	kfree(bus->p);
 }
 
 int bus_register_notifier(struct bus_type *bus, struct notifier_block *nb)
 {
-	return blocking_notifier_chain_register(&bus->bus_notifier, nb);
+	return blocking_notifier_chain_register(&bus->p->bus_notifier, nb);
 }
 EXPORT_SYMBOL_GPL(bus_register_notifier);
 
 int bus_unregister_notifier(struct bus_type *bus, struct notifier_block *nb)
 {
-	return blocking_notifier_chain_unregister(&bus->bus_notifier, nb);
+	return blocking_notifier_chain_unregister(&bus->p->bus_notifier, nb);
 }
 EXPORT_SYMBOL_GPL(bus_unregister_notifier);
 
 struct kset *bus_get_kset(struct bus_type *bus)
 {
-	return &bus->subsys;
+	return &bus->p->subsys;
 }
 EXPORT_SYMBOL_GPL(bus_get_kset);
 
 struct klist *bus_get_device_klist(struct bus_type *bus)
 {
-	return &bus->klist_devices;
+	return &bus->p->klist_devices;
 }
 EXPORT_SYMBOL_GPL(bus_get_device_klist);
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index beb351600675..414a480e10a3 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -769,7 +769,7 @@ int device_add(struct device *dev)
 
 	/* notify clients of device entry (new way) */
 	if (dev->bus)
-		blocking_notifier_call_chain(&dev->bus->bus_notifier,
+		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_ADD_DEVICE, dev);
 
 	error = device_create_file(dev, &uevent_attr);
@@ -820,7 +820,7 @@ int device_add(struct device *dev)
 	dpm_sysfs_remove(dev);
  PMError:
 	if (dev->bus)
-		blocking_notifier_call_chain(&dev->bus->bus_notifier,
+		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_DEL_DEVICE, dev);
 	device_remove_attrs(dev);
  AttrsError:
@@ -999,7 +999,7 @@ void device_del(struct device * dev)
 	if (platform_notify_remove)
 		platform_notify_remove(dev);
 	if (dev->bus)
-		blocking_notifier_call_chain(&dev->bus->bus_notifier,
+		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_DEL_DEVICE, dev);
 	kobject_uevent(&dev->kobj, KOBJ_REMOVE);
 	kobject_del(&dev->kobj);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 7ac474db88c5..7bf0e674c975 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -38,7 +38,7 @@ static void driver_bound(struct device *dev)
 		 dev->bus_id, dev->driver->name);
 
 	if (dev->bus)
-		blocking_notifier_call_chain(&dev->bus->bus_notifier,
+		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_BOUND_DRIVER, dev);
 
 	klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices);
@@ -296,7 +296,7 @@ static void __device_release_driver(struct device * dev)
 		klist_remove(&dev->knode_driver);
 
 		if (dev->bus)
-			blocking_notifier_call_chain(&dev->bus->bus_notifier,
+			blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 						     BUS_NOTIFY_UNBIND_DRIVER,
 						     dev);
 
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 1c9770dfb801..f94be40646d2 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -185,7 +185,7 @@ void driver_unregister(struct device_driver * drv)
  */
 struct device_driver *driver_find(const char *name, struct bus_type *bus)
 {
-	struct kobject *k = kset_find_obj(bus->drivers_kset, name);
+	struct kobject *k = kset_find_obj(bus->p->drivers_kset, name);
 	if (k)
 		return to_drv(k);
 	return NULL;
diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index fb5609241482..d56a05f94f67 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -497,12 +497,12 @@ int __init_or_module platform_driver_probe(struct platform_driver *drv,
 	 * if the probe was successful, and make sure any forced probes of
 	 * new devices fail.
 	 */
-	spin_lock(&platform_bus_type.klist_drivers.k_lock);
+	spin_lock(&platform_bus_type.p->klist_drivers.k_lock);
 	drv->probe = NULL;
 	if (code == 0 && list_empty(&drv->driver.klist_devices.k_list))
 		retval = -ENODEV;
 	drv->driver.probe = platform_drv_probe_fail;
-	spin_unlock(&platform_bus_type.klist_drivers.k_lock);
+	spin_unlock(&platform_bus_type.p->klist_drivers.k_lock);
 
 	if (code != retval)
 		platform_driver_unregister(drv);
diff --git a/include/linux/device.h b/include/linux/device.h
index 62e695bd3c93..3f24bf46d298 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -35,6 +35,7 @@ struct device_driver;
 struct class;
 struct class_device;
 struct bus_type;
+struct bus_type_private;
 
 struct bus_attribute {
 	struct attribute	attr;
@@ -51,15 +52,6 @@ extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct bus_type {
 	const char		* name;
-
-	struct kset		subsys;
-	struct kset		*drivers_kset;
-	struct kset		*devices_kset;
-	struct klist		klist_devices;
-	struct klist		klist_drivers;
-
-	struct blocking_notifier_head bus_notifier;
-
 	struct bus_attribute	* bus_attrs;
 	struct device_attribute	* dev_attrs;
 	struct driver_attribute	* drv_attrs;
@@ -75,7 +67,7 @@ struct bus_type {
 	int (*resume_early)(struct device * dev);
 	int (*resume)(struct device * dev);
 
-	unsigned int drivers_autoprobe:1;
+	struct bus_type_private *p;
 };
 
 extern int __must_check bus_register(struct bus_type * bus);
-- 
cgit v1.2.3


From 57c745340a60c51d2b9af3d4dcf7e0ede284855b Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Wed, 5 Dec 2007 12:50:23 +0100
Subject: driver core: Introduce default attribute groups.

This is lot like default attributes for devices (and indeed,
a lot of the code is lifted from there).

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/driver.c  | 42 +++++++++++++++++++++++++++++++++++++++++-
 include/linux/device.h |  1 +
 2 files changed, 42 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index f94be40646d2..e3b58407fedc 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -142,6 +142,37 @@ void put_driver(struct device_driver * drv)
 	kobject_put(&drv->kobj);
 }
 
+static int driver_add_groups(struct device_driver *drv,
+			     struct attribute_group **groups)
+{
+	int error = 0;
+	int i;
+
+	if (groups) {
+		for (i = 0; groups[i]; i++) {
+			error = sysfs_create_group(&drv->kobj, groups[i]);
+			if (error) {
+				while (--i >= 0)
+					sysfs_remove_group(&drv->kobj,
+							   groups[i]);
+				break;
+			}
+		}
+	}
+	return error;
+}
+
+static void driver_remove_groups(struct device_driver *drv,
+				 struct attribute_group **groups)
+{
+	int i;
+
+	if (groups)
+		for (i = 0; groups[i]; i++)
+			sysfs_remove_group(&drv->kobj, groups[i]);
+}
+
+
 /**
  *	driver_register - register driver with bus
  *	@drv:	driver to register
@@ -152,13 +183,21 @@ void put_driver(struct device_driver * drv)
  */
 int driver_register(struct device_driver * drv)
 {
+	int ret;
+
 	if ((drv->bus->probe && drv->probe) ||
 	    (drv->bus->remove && drv->remove) ||
 	    (drv->bus->shutdown && drv->shutdown)) {
 		printk(KERN_WARNING "Driver '%s' needs updating - please use bus_type methods\n", drv->name);
 	}
 	klist_init(&drv->klist_devices, NULL, NULL);
-	return bus_add_driver(drv);
+	ret = bus_add_driver(drv);
+	if (ret)
+		return ret;
+	ret = driver_add_groups(drv, drv->groups);
+	if (ret)
+		bus_remove_driver(drv);
+	return ret;
 }
 
 /**
@@ -170,6 +209,7 @@ int driver_register(struct device_driver * drv)
 
 void driver_unregister(struct device_driver * drv)
 {
+	driver_remove_groups(drv, drv->groups);
 	bus_remove_driver(drv);
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 3f24bf46d298..d974dda4aa51 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -129,6 +129,7 @@ struct device_driver {
 	void	(*shutdown)	(struct device * dev);
 	int	(*suspend)	(struct device * dev, pm_message_t state);
 	int	(*resume)	(struct device * dev);
+	struct attribute_group **groups;
 };
 
 
-- 
cgit v1.2.3


From cbe9c595f1de2e2a98403be2c14bfbc2486e84c4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 19 Dec 2007 15:54:39 -0400
Subject: Driver: add driver_add_kobj for looney iseries_veth driver

The iseries driver wants to hang kobjects off of its driver, so, to
preserve backwards compatibility, we need to add a call to the driver
core to allow future changes to work properly.

Hopefully no one uses this function in the future and the iseries_veth
driver authors come to their senses so I can remove this hack...

Cc: Dave Larson <larson1@us.ibm.com>
Cc: Santiago Leon <santil@us.ibm.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/driver.c      | 24 ++++++++++++++++++++++++
 drivers/net/iseries_veth.c |  2 +-
 include/linux/device.h     |  4 ++++
 3 files changed, 29 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index e3b58407fedc..633ae1d70e14 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -123,6 +123,30 @@ void driver_remove_file(struct device_driver * drv, struct driver_attribute * at
 }
 
 
+/**
+ * driver_add_kobj - add a kobject below the specified driver
+ *
+ * You really don't want to do this, this is only here due to one looney
+ * iseries driver, go poke those developers if you are annoyed about
+ * this...
+ */
+int driver_add_kobj(struct device_driver *drv, struct kobject *kobj,
+		    const char *fmt, ...)
+{
+	va_list args;
+	char *name;
+
+	va_start(args, fmt);
+	name = kvasprintf(GFP_KERNEL, fmt, args);
+	va_end(args);
+
+	if (!name)
+		return -ENOMEM;
+
+	return kobject_add_ng(kobj, &drv->kobj, "%s", name);
+}
+EXPORT_SYMBOL_GPL(driver_add_kobj);
+
 /**
  *	get_driver - increment driver reference count.
  *	@drv:	driver.
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index 90ff4ec5f6fc..1a8299acd3f1 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -1705,7 +1705,7 @@ static int __init veth_module_init(void)
 
 		kobj = &veth_cnx[i]->kobject;
 		/* If the add failes, complain but otherwise continue */
-		if (0 != kobject_add_ng(kobj, &veth_driver.driver.kobj,
+		if (0 != driver_add_kobj(&veth_driver.driver, kobj,
 					"cnx%.2d", veth_cnx[i]->remote_lp))
 			veth_error("cnx %d: Failed adding to sysfs.\n", i);
 	}
diff --git a/include/linux/device.h b/include/linux/device.h
index d974dda4aa51..721ee318d57b 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -156,6 +156,10 @@ extern int __must_check driver_create_file(struct device_driver *,
 					struct driver_attribute *);
 extern void driver_remove_file(struct device_driver *, struct driver_attribute *);
 
+extern int __must_check driver_add_kobj(struct device_driver *drv,
+					struct kobject *kobj,
+					const char *fmt, ...);
+
 extern int __must_check driver_for_each_device(struct device_driver * drv,
 		struct device *start, void *data,
 		int (*fn)(struct device *, void *));
-- 
cgit v1.2.3


From c63469a3985a9771c18a916b8d42845d044ea0b1 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 28 Nov 2007 12:23:18 -0800
Subject: Driver core: move the driver specific module code into the driver
 core

The module driver specific code should belong in the driver core, not in
the kernel/ directory.  So move this code.  This is done in preparation
for some struct device_driver rework that should be confined to the
driver core code only.

This also lets us keep from exporting these functions, as no external
code should ever be calling it.

Thanks to Andrew Morton for the !CONFIG_MODULES fix.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/Makefile  |  1 +
 drivers/base/base.h    |  9 +++++
 drivers/base/module.c  | 94 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/module.h | 15 --------
 kernel/module.c        | 87 ----------------------------------------------
 5 files changed, 104 insertions(+), 102 deletions(-)
 create mode 100644 drivers/base/module.c

(limited to 'include')

diff --git a/drivers/base/Makefile b/drivers/base/Makefile
index b39ea3f59c9b..ff2696823f8d 100644
--- a/drivers/base/Makefile
+++ b/drivers/base/Makefile
@@ -11,6 +11,7 @@ obj-$(CONFIG_FW_LOADER)	+= firmware_class.o
 obj-$(CONFIG_NUMA)	+= node.o
 obj-$(CONFIG_MEMORY_HOTPLUG_SPARSE) += memory.o
 obj-$(CONFIG_SMP)	+= topology.o
+obj-$(CONFIG_MODULES)	+= module.o
 obj-$(CONFIG_SYS_HYPERVISOR) += hypervisor.o
 
 ifeq ($(CONFIG_DEBUG_DRIVER),y)
diff --git a/drivers/base/base.h b/drivers/base/base.h
index ca6d273064f8..05472360f6a2 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -73,3 +73,12 @@ extern char *make_class_name(const char *name, struct kobject *kobj);
 extern int devres_release_all(struct device *dev);
 
 extern struct kset *devices_kset;
+
+#ifdef CONFIG_MODULES
+extern void module_add_driver(struct module *mod, struct device_driver *drv);
+extern void module_remove_driver(struct device_driver *drv);
+#else
+static inline void module_add_driver(struct module *mod,
+				     struct device_driver *drv) { }
+static inline void module_remove_driver(struct device_driver *drv) { }
+#endif
diff --git a/drivers/base/module.c b/drivers/base/module.c
new file mode 100644
index 000000000000..cad07be5de1a
--- /dev/null
+++ b/drivers/base/module.c
@@ -0,0 +1,94 @@
+/*
+ * module.c - module sysfs fun for drivers
+ *
+ * This file is released under the GPLv2
+ *
+ */
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include "base.h"
+
+static char *make_driver_name(struct device_driver *drv)
+{
+	char *driver_name;
+
+	driver_name = kmalloc(strlen(drv->name) + strlen(drv->bus->name) + 2,
+			      GFP_KERNEL);
+	if (!driver_name)
+		return NULL;
+
+	sprintf(driver_name, "%s:%s", drv->bus->name, drv->name);
+	return driver_name;
+}
+
+static void module_create_drivers_dir(struct module_kobject *mk)
+{
+	if (!mk || mk->drivers_dir)
+		return;
+
+	mk->drivers_dir = kobject_create_and_add("drivers", &mk->kobj);
+}
+
+void module_add_driver(struct module *mod, struct device_driver *drv)
+{
+	char *driver_name;
+	int no_warn;
+	struct module_kobject *mk = NULL;
+
+	if (!drv)
+		return;
+
+	if (mod)
+		mk = &mod->mkobj;
+	else if (drv->mod_name) {
+		struct kobject *mkobj;
+
+		/* Lookup built-in module entry in /sys/modules */
+		mkobj = kset_find_obj(module_kset, drv->mod_name);
+		if (mkobj) {
+			mk = container_of(mkobj, struct module_kobject, kobj);
+			/* remember our module structure */
+			drv->mkobj = mk;
+			/* kset_find_obj took a reference */
+			kobject_put(mkobj);
+		}
+	}
+
+	if (!mk)
+		return;
+
+	/* Don't check return codes; these calls are idempotent */
+	no_warn = sysfs_create_link(&drv->kobj, &mk->kobj, "module");
+	driver_name = make_driver_name(drv);
+	if (driver_name) {
+		module_create_drivers_dir(mk);
+		no_warn = sysfs_create_link(mk->drivers_dir, &drv->kobj,
+					    driver_name);
+		kfree(driver_name);
+	}
+}
+
+void module_remove_driver(struct device_driver *drv)
+{
+	struct module_kobject *mk = NULL;
+	char *driver_name;
+
+	if (!drv)
+		return;
+
+	sysfs_remove_link(&drv->kobj, "module");
+
+	if (drv->owner)
+		mk = &drv->owner->mkobj;
+	else if (drv->mkobj)
+		mk = drv->mkobj;
+	if (mk && mk->drivers_dir) {
+		driver_name = make_driver_name(drv);
+		if (driver_name) {
+			sysfs_remove_link(mk->drivers_dir, driver_name);
+			kfree(driver_name);
+		}
+	}
+}
diff --git a/include/linux/module.h b/include/linux/module.h
index fbe930b9b69c..c97bdb7eb957 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -609,21 +609,6 @@ static inline void module_remove_modinfo_attrs(struct module *mod)
 
 #endif /* CONFIG_SYSFS */
 
-#if defined(CONFIG_SYSFS) && defined(CONFIG_MODULES)
-
-void module_add_driver(struct module *mod, struct device_driver *drv);
-void module_remove_driver(struct device_driver *drv);
-
-#else /* not both CONFIG_SYSFS && CONFIG_MODULES */
-
-static inline void module_add_driver(struct module *mod, struct device_driver *drv)
-{ }
-
-static inline void module_remove_driver(struct device_driver *drv)
-{ }
-
-#endif
-
 #define symbol_request(x) try_then_request_module(symbol_get(x), "symbol:" #x)
 
 /* BELOW HERE ALL THESE ARE OBSOLETE AND WILL VANISH */
diff --git a/kernel/module.c b/kernel/module.c
index d03fcd9d652c..dc4d3f5ce820 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2501,93 +2501,6 @@ void print_modules(void)
 	printk("\n");
 }
 
-#ifdef CONFIG_SYSFS
-static char *make_driver_name(struct device_driver *drv)
-{
-	char *driver_name;
-
-	driver_name = kmalloc(strlen(drv->name) + strlen(drv->bus->name) + 2,
-			      GFP_KERNEL);
-	if (!driver_name)
-		return NULL;
-
-	sprintf(driver_name, "%s:%s", drv->bus->name, drv->name);
-	return driver_name;
-}
-
-static void module_create_drivers_dir(struct module_kobject *mk)
-{
-	if (!mk || mk->drivers_dir)
-		return;
-
-	mk->drivers_dir = kobject_create_and_add("drivers", &mk->kobj);
-}
-
-void module_add_driver(struct module *mod, struct device_driver *drv)
-{
-	char *driver_name;
-	int no_warn;
-	struct module_kobject *mk = NULL;
-
-	if (!drv)
-		return;
-
-	if (mod)
-		mk = &mod->mkobj;
-	else if (drv->mod_name) {
-		struct kobject *mkobj;
-
-		/* Lookup built-in module entry in /sys/modules */
-		mkobj = kset_find_obj(module_kset, drv->mod_name);
-		if (mkobj) {
-			mk = container_of(mkobj, struct module_kobject, kobj);
-			/* remember our module structure */
-			drv->mkobj = mk;
-			/* kset_find_obj took a reference */
-			kobject_put(mkobj);
-		}
-	}
-
-	if (!mk)
-		return;
-
-	/* Don't check return codes; these calls are idempotent */
-	no_warn = sysfs_create_link(&drv->kobj, &mk->kobj, "module");
-	driver_name = make_driver_name(drv);
-	if (driver_name) {
-		module_create_drivers_dir(mk);
-		no_warn = sysfs_create_link(mk->drivers_dir, &drv->kobj,
-					    driver_name);
-		kfree(driver_name);
-	}
-}
-EXPORT_SYMBOL(module_add_driver);
-
-void module_remove_driver(struct device_driver *drv)
-{
-	struct module_kobject *mk = NULL;
-	char *driver_name;
-
-	if (!drv)
-		return;
-
-	sysfs_remove_link(&drv->kobj, "module");
-
-	if (drv->owner)
-		mk = &drv->owner->mkobj;
-	else if (drv->mkobj)
-		mk = drv->mkobj;
-	if (mk && mk->drivers_dir) {
-		driver_name = make_driver_name(drv);
-		if (driver_name) {
-			sysfs_remove_link(mk->drivers_dir, driver_name);
-			kfree(driver_name);
-		}
-	}
-}
-EXPORT_SYMBOL(module_remove_driver);
-#endif
-
 #ifdef CONFIG_MODVERSIONS
 /* Generate the signature for struct module here, too, for modversions. */
 void struct_module(struct module *mod) { return; }
-- 
cgit v1.2.3


From e5dd12784617f0f1fae5f96a7fac1ec4c49fadbe Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 28 Nov 2007 15:59:15 -0800
Subject: Driver core: move the static kobject out of struct driver

This patch removes the kobject, and a few other driver-core-only fields
out of struct driver and into the driver core only.  Now drivers can be
safely create on the stack or statically (like they currently are.)

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h     |  8 ++++++
 drivers/base/bus.c      | 71 +++++++++++++++++++++++++++----------------------
 drivers/base/dd.c       | 24 ++++++++---------
 drivers/base/driver.c   | 40 ++++++++++++++++++----------
 drivers/base/module.c   | 12 ++++-----
 drivers/base/platform.c |  2 +-
 include/linux/device.h  | 16 +++++------
 7 files changed, 99 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 05472360f6a2..3b0f395552db 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -27,6 +27,14 @@ struct bus_type_private {
 	struct bus_type *bus;
 };
 
+struct driver_private {
+	struct kobject kobj;
+	struct klist klist_devices;
+	struct klist_node knode_bus;
+	struct module_kobject *mkobj;
+	struct device_driver *driver;
+};
+#define to_driver(obj) container_of(obj, struct driver_private, kobj)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 04d3850ff4b7..aa0c986c323a 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -3,6 +3,8 @@
  *
  * Copyright (c) 2002-3 Patrick Mochel
  * Copyright (c) 2002-3 Open Source Development Labs
+ * Copyright (c) 2007 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (c) 2007 Novell Inc.
  *
  * This file is released under the GPLv2
  *
@@ -24,7 +26,6 @@
  */
 
 #define to_drv_attr(_attr) container_of(_attr, struct driver_attribute, attr)
-#define to_driver(obj) container_of(obj, struct device_driver, kobj)
 
 
 static int __must_check bus_rescan_devices_helper(struct device *dev,
@@ -49,11 +50,11 @@ static ssize_t
 drv_attr_show(struct kobject * kobj, struct attribute * attr, char * buf)
 {
 	struct driver_attribute * drv_attr = to_drv_attr(attr);
-	struct device_driver * drv = to_driver(kobj);
+	struct driver_private *drv_priv = to_driver(kobj);
 	ssize_t ret = -EIO;
 
 	if (drv_attr->show)
-		ret = drv_attr->show(drv, buf);
+		ret = drv_attr->show(drv_priv->driver, buf);
 	return ret;
 }
 
@@ -62,11 +63,11 @@ drv_attr_store(struct kobject * kobj, struct attribute * attr,
 	       const char * buf, size_t count)
 {
 	struct driver_attribute * drv_attr = to_drv_attr(attr);
-	struct device_driver * drv = to_driver(kobj);
+	struct driver_private *drv_priv = to_driver(kobj);
 	ssize_t ret = -EIO;
 
 	if (drv_attr->store)
-		ret = drv_attr->store(drv, buf, count);
+		ret = drv_attr->store(drv_priv->driver, buf, count);
 	return ret;
 }
 
@@ -75,22 +76,12 @@ static struct sysfs_ops driver_sysfs_ops = {
 	.store	= drv_attr_store,
 };
 
-
-static void driver_release(struct kobject * kobj)
+static void driver_release(struct kobject *kobj)
 {
-	/*
-	 * Yes this is an empty release function, it is this way because struct
-	 * device is always a static object, not a dynamic one.  Yes, this is
-	 * not nice and bad, but remember, drivers are code, reference counted
-	 * by the module count, not a device, which is really data.  And yes,
-	 * in the future I do want to have all drivers be created dynamically,
-	 * and am working toward that goal, but it will take a bit longer...
-	 *
-	 * But do not let this example give _anyone_ the idea that they can
-	 * create a release function without any code in it at all, to do that
-	 * is almost always wrong.  If you have any questions about this,
-	 * please send an email to <greg@kroah.com>
-	 */
+	struct driver_private *drv_priv = to_driver(kobj);
+
+	pr_debug("%s: freeing %s\n", __FUNCTION__, kobject_name(kobj));
+	kfree(drv_priv);
 }
 
 static struct kobj_type driver_ktype = {
@@ -350,7 +341,13 @@ struct device * bus_find_device(struct bus_type *bus,
 static struct device_driver * next_driver(struct klist_iter * i)
 {
 	struct klist_node * n = klist_next(i);
-	return n ? container_of(n, struct device_driver, knode_bus) : NULL;
+	struct driver_private *drv_priv;
+
+	if (n) {
+		drv_priv = container_of(n, struct driver_private, knode_bus);
+		return drv_priv->driver;
+	}
+	return NULL;
 }
 
 /**
@@ -384,7 +381,7 @@ int bus_for_each_drv(struct bus_type * bus, struct device_driver * start,
 		return -EINVAL;
 
 	klist_iter_init_node(&bus->p->klist_drivers, &i,
-			     start ? &start->knode_bus : NULL);
+			     start ? &start->p->knode_bus : NULL);
 	while ((drv = next_driver(&i)) && !error)
 		error = fn(drv, data);
 	klist_iter_exit(&i);
@@ -620,7 +617,7 @@ static ssize_t driver_uevent_store(struct device_driver *drv,
 	enum kobject_action action;
 
 	if (kobject_action_type(buf, count, &action) == 0)
-		kobject_uevent(&drv->kobj, action);
+		kobject_uevent(&drv->p->kobj, action);
 	return count;
 }
 static DRIVER_ATTR(uevent, S_IWUSR, NULL, driver_uevent_store);
@@ -632,19 +629,29 @@ static DRIVER_ATTR(uevent, S_IWUSR, NULL, driver_uevent_store);
  */
 int bus_add_driver(struct device_driver *drv)
 {
-	struct bus_type * bus = bus_get(drv->bus);
+	struct bus_type *bus;
+	struct driver_private *priv;
 	int error = 0;
 
+	bus = bus_get(drv->bus);
 	if (!bus)
 		return -EINVAL;
 
 	pr_debug("bus %s: add driver %s\n", bus->name, drv->name);
-	error = kobject_set_name(&drv->kobj, "%s", drv->name);
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	error = kobject_set_name(&priv->kobj, "%s", drv->name);
 	if (error)
 		goto out_put_bus;
-	drv->kobj.kset = bus->p->drivers_kset;
-	drv->kobj.ktype = &driver_ktype;
-	error = kobject_register(&drv->kobj);
+	priv->kobj.kset = bus->p->drivers_kset;
+	priv->kobj.ktype = &driver_ktype;
+	klist_init(&priv->klist_devices, NULL, NULL);
+	priv->driver = drv;
+	drv->p = priv;
+	error = kobject_register(&priv->kobj);
 	if (error)
 		goto out_put_bus;
 
@@ -653,7 +660,7 @@ int bus_add_driver(struct device_driver *drv)
 		if (error)
 			goto out_unregister;
 	}
-	klist_add_tail(&drv->knode_bus, &bus->p->klist_drivers);
+	klist_add_tail(&priv->knode_bus, &bus->p->klist_drivers);
 	module_add_driver(drv->owner, drv);
 
 	error = driver_create_file(drv, &driver_attr_uevent);
@@ -676,7 +683,7 @@ int bus_add_driver(struct device_driver *drv)
 
 	return error;
 out_unregister:
-	kobject_unregister(&drv->kobj);
+	kobject_unregister(&priv->kobj);
 out_put_bus:
 	bus_put(bus);
 	return error;
@@ -699,11 +706,11 @@ void bus_remove_driver(struct device_driver * drv)
 	remove_bind_files(drv);
 	driver_remove_attrs(drv->bus, drv);
 	driver_remove_file(drv, &driver_attr_uevent);
-	klist_remove(&drv->knode_bus);
+	klist_remove(&drv->p->knode_bus);
 	pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name);
 	driver_detach(drv);
 	module_remove_driver(drv);
-	kobject_unregister(&drv->kobj);
+	kobject_unregister(&drv->p->kobj);
 	bus_put(drv->bus);
 }
 
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 7bf0e674c975..87a348ce818c 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -11,6 +11,8 @@
  *
  *	Copyright (c) 2002-5 Patrick Mochel
  *	Copyright (c) 2002-3 Open Source Development Labs
+ *	Copyright (c) 2007 Greg Kroah-Hartman <gregkh@suse.de>
+ *	Copyright (c) 2007 Novell Inc.
  *
  *	This file is released under the GPLv2
  */
@@ -23,8 +25,6 @@
 #include "base.h"
 #include "power/power.h"
 
-#define to_drv(node) container_of(node, struct device_driver, kobj.entry)
-
 
 static void driver_bound(struct device *dev)
 {
@@ -41,20 +41,20 @@ static void driver_bound(struct device *dev)
 		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_BOUND_DRIVER, dev);
 
-	klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices);
+	klist_add_tail(&dev->knode_driver, &dev->driver->p->klist_devices);
 }
 
 static int driver_sysfs_add(struct device *dev)
 {
 	int ret;
 
-	ret = sysfs_create_link(&dev->driver->kobj, &dev->kobj,
+	ret = sysfs_create_link(&dev->driver->p->kobj, &dev->kobj,
 			  kobject_name(&dev->kobj));
 	if (ret == 0) {
-		ret = sysfs_create_link(&dev->kobj, &dev->driver->kobj,
+		ret = sysfs_create_link(&dev->kobj, &dev->driver->p->kobj,
 					"driver");
 		if (ret)
-			sysfs_remove_link(&dev->driver->kobj,
+			sysfs_remove_link(&dev->driver->p->kobj,
 					kobject_name(&dev->kobj));
 	}
 	return ret;
@@ -65,7 +65,7 @@ static void driver_sysfs_remove(struct device *dev)
 	struct device_driver *drv = dev->driver;
 
 	if (drv) {
-		sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj));
+		sysfs_remove_link(&drv->p->kobj, kobject_name(&dev->kobj));
 		sysfs_remove_link(&dev->kobj, "driver");
 	}
 }
@@ -339,15 +339,15 @@ void driver_detach(struct device_driver * drv)
 	struct device * dev;
 
 	for (;;) {
-		spin_lock(&drv->klist_devices.k_lock);
-		if (list_empty(&drv->klist_devices.k_list)) {
-			spin_unlock(&drv->klist_devices.k_lock);
+		spin_lock(&drv->p->klist_devices.k_lock);
+		if (list_empty(&drv->p->klist_devices.k_list)) {
+			spin_unlock(&drv->p->klist_devices.k_lock);
 			break;
 		}
-		dev = list_entry(drv->klist_devices.k_list.prev,
+		dev = list_entry(drv->p->klist_devices.k_list.prev,
 				struct device, knode_driver.n_node);
 		get_device(dev);
-		spin_unlock(&drv->klist_devices.k_lock);
+		spin_unlock(&drv->p->klist_devices.k_lock);
 
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 633ae1d70e14..5aacff208f21 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -3,6 +3,8 @@
  *
  * Copyright (c) 2002-3 Patrick Mochel
  * Copyright (c) 2002-3 Open Source Development Labs
+ * Copyright (c) 2007 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (c) 2007 Novell Inc.
  *
  * This file is released under the GPLv2
  *
@@ -15,7 +17,6 @@
 #include "base.h"
 
 #define to_dev(node) container_of(node, struct device, driver_list)
-#define to_drv(obj) container_of(obj, struct device_driver, kobj)
 
 
 static struct device * next_device(struct klist_iter * i)
@@ -44,7 +45,7 @@ int driver_for_each_device(struct device_driver * drv, struct device * start,
 	if (!drv)
 		return -EINVAL;
 
-	klist_iter_init_node(&drv->klist_devices, &i,
+	klist_iter_init_node(&drv->p->klist_devices, &i,
 			     start ? &start->knode_driver : NULL);
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
@@ -80,7 +81,7 @@ struct device * driver_find_device(struct device_driver *drv,
 	if (!drv)
 		return NULL;
 
-	klist_iter_init_node(&drv->klist_devices, &i,
+	klist_iter_init_node(&drv->p->klist_devices, &i,
 			     (start ? &start->knode_driver : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
@@ -100,7 +101,7 @@ int driver_create_file(struct device_driver * drv, struct driver_attribute * att
 {
 	int error;
 	if (get_driver(drv)) {
-		error = sysfs_create_file(&drv->kobj, &attr->attr);
+		error = sysfs_create_file(&drv->p->kobj, &attr->attr);
 		put_driver(drv);
 	} else
 		error = -EINVAL;
@@ -117,7 +118,7 @@ int driver_create_file(struct device_driver * drv, struct driver_attribute * att
 void driver_remove_file(struct device_driver * drv, struct driver_attribute * attr)
 {
 	if (get_driver(drv)) {
-		sysfs_remove_file(&drv->kobj, &attr->attr);
+		sysfs_remove_file(&drv->p->kobj, &attr->attr);
 		put_driver(drv);
 	}
 }
@@ -143,7 +144,7 @@ int driver_add_kobj(struct device_driver *drv, struct kobject *kobj,
 	if (!name)
 		return -ENOMEM;
 
-	return kobject_add_ng(kobj, &drv->kobj, "%s", name);
+	return kobject_add_ng(kobj, &drv->p->kobj, "%s", name);
 }
 EXPORT_SYMBOL_GPL(driver_add_kobj);
 
@@ -153,7 +154,15 @@ EXPORT_SYMBOL_GPL(driver_add_kobj);
  */
 struct device_driver * get_driver(struct device_driver * drv)
 {
-	return drv ? to_drv(kobject_get(&drv->kobj)) : NULL;
+	if (drv) {
+		struct driver_private *priv;
+		struct kobject *kobj;
+
+		kobj = kobject_get(&drv->p->kobj);
+		priv = to_driver(kobj);
+		return priv->driver;
+	}
+	return NULL;
 }
 
 
@@ -163,7 +172,7 @@ struct device_driver * get_driver(struct device_driver * drv)
  */
 void put_driver(struct device_driver * drv)
 {
-	kobject_put(&drv->kobj);
+	kobject_put(&drv->p->kobj);
 }
 
 static int driver_add_groups(struct device_driver *drv,
@@ -174,10 +183,10 @@ static int driver_add_groups(struct device_driver *drv,
 
 	if (groups) {
 		for (i = 0; groups[i]; i++) {
-			error = sysfs_create_group(&drv->kobj, groups[i]);
+			error = sysfs_create_group(&drv->p->kobj, groups[i]);
 			if (error) {
 				while (--i >= 0)
-					sysfs_remove_group(&drv->kobj,
+					sysfs_remove_group(&drv->p->kobj,
 							   groups[i]);
 				break;
 			}
@@ -193,7 +202,7 @@ static void driver_remove_groups(struct device_driver *drv,
 
 	if (groups)
 		for (i = 0; groups[i]; i++)
-			sysfs_remove_group(&drv->kobj, groups[i]);
+			sysfs_remove_group(&drv->p->kobj, groups[i]);
 }
 
 
@@ -214,7 +223,6 @@ int driver_register(struct device_driver * drv)
 	    (drv->bus->shutdown && drv->shutdown)) {
 		printk(KERN_WARNING "Driver '%s' needs updating - please use bus_type methods\n", drv->name);
 	}
-	klist_init(&drv->klist_devices, NULL, NULL);
 	ret = bus_add_driver(drv);
 	if (ret)
 		return ret;
@@ -250,8 +258,12 @@ void driver_unregister(struct device_driver * drv)
 struct device_driver *driver_find(const char *name, struct bus_type *bus)
 {
 	struct kobject *k = kset_find_obj(bus->p->drivers_kset, name);
-	if (k)
-		return to_drv(k);
+	struct driver_private *priv;
+
+	if (k) {
+		priv = to_driver(k);
+		return priv->driver;
+	}
 	return NULL;
 }
 
diff --git a/drivers/base/module.c b/drivers/base/module.c
index cad07be5de1a..103be9cacb05 100644
--- a/drivers/base/module.c
+++ b/drivers/base/module.c
@@ -50,7 +50,7 @@ void module_add_driver(struct module *mod, struct device_driver *drv)
 		if (mkobj) {
 			mk = container_of(mkobj, struct module_kobject, kobj);
 			/* remember our module structure */
-			drv->mkobj = mk;
+			drv->p->mkobj = mk;
 			/* kset_find_obj took a reference */
 			kobject_put(mkobj);
 		}
@@ -60,11 +60,11 @@ void module_add_driver(struct module *mod, struct device_driver *drv)
 		return;
 
 	/* Don't check return codes; these calls are idempotent */
-	no_warn = sysfs_create_link(&drv->kobj, &mk->kobj, "module");
+	no_warn = sysfs_create_link(&drv->p->kobj, &mk->kobj, "module");
 	driver_name = make_driver_name(drv);
 	if (driver_name) {
 		module_create_drivers_dir(mk);
-		no_warn = sysfs_create_link(mk->drivers_dir, &drv->kobj,
+		no_warn = sysfs_create_link(mk->drivers_dir, &drv->p->kobj,
 					    driver_name);
 		kfree(driver_name);
 	}
@@ -78,12 +78,12 @@ void module_remove_driver(struct device_driver *drv)
 	if (!drv)
 		return;
 
-	sysfs_remove_link(&drv->kobj, "module");
+	sysfs_remove_link(&drv->p->kobj, "module");
 
 	if (drv->owner)
 		mk = &drv->owner->mkobj;
-	else if (drv->mkobj)
-		mk = drv->mkobj;
+	else if (drv->p->mkobj)
+		mk = drv->p->mkobj;
 	if (mk && mk->drivers_dir) {
 		driver_name = make_driver_name(drv);
 		if (driver_name) {
diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index d56a05f94f67..bdd59e8358f0 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -499,7 +499,7 @@ int __init_or_module platform_driver_probe(struct platform_driver *drv,
 	 */
 	spin_lock(&platform_bus_type.p->klist_drivers.k_lock);
 	drv->probe = NULL;
-	if (code == 0 && list_empty(&drv->driver.klist_devices.k_list))
+	if (code == 0 && list_empty(&drv->driver.p->klist_devices.k_list))
 		retval = -ENODEV;
 	drv->driver.probe = platform_drv_probe_fail;
 	spin_unlock(&platform_bus_type.p->klist_drivers.k_lock);
diff --git a/include/linux/device.h b/include/linux/device.h
index 721ee318d57b..92ba3a874627 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -32,6 +32,7 @@
 
 struct device;
 struct device_driver;
+struct driver_private;
 struct class;
 struct class_device;
 struct bus_type;
@@ -113,16 +114,11 @@ extern struct kset *bus_get_kset(struct bus_type *bus);
 extern struct klist *bus_get_device_klist(struct bus_type *bus);
 
 struct device_driver {
-	const char		* name;
-	struct bus_type		* bus;
-
-	struct kobject		kobj;
-	struct klist		klist_devices;
-	struct klist_node	knode_bus;
+	const char		*name;
+	struct bus_type		*bus;
 
-	struct module		* owner;
-	const char 		* mod_name;	/* used for built-in modules */
-	struct module_kobject	* mkobj;
+	struct module		*owner;
+	const char 		*mod_name;	/* used for built-in modules */
 
 	int	(*probe)	(struct device * dev);
 	int	(*remove)	(struct device * dev);
@@ -130,6 +126,8 @@ struct device_driver {
 	int	(*suspend)	(struct device * dev, pm_message_t state);
 	int	(*resume)	(struct device * dev);
 	struct attribute_group **groups;
+
+	struct driver_private *p;
 };
 
 
-- 
cgit v1.2.3


From edfaa7c36574f1bf09c65ad602412db9da5f96bf Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Mon, 21 May 2007 22:08:01 +0200
Subject: Driver core: convert block from raw kobjects to core devices

This moves the block devices to /sys/class/block. It will create a
flat list of all block devices, with the disks and partitions in one
directory. For compatibility /sys/block is created and contains symlinks
to the disks.

  /sys/class/block
  |-- sda -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda
  |-- sda1 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda1
  |-- sda10 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda10
  |-- sda5 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda5
  |-- sda6 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda6
  |-- sda7 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda7
  |-- sda8 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda8
  |-- sda9 -> ../../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda/sda9
  `-- sr0 -> ../../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sr0

  /sys/block/
  |-- sda -> ../devices/pci0000:00/0000:00:1f.2/host0/target0:0:0/0:0:0:0/block/sda
  `-- sr0 -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sr0

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 block/genhd.c              | 416 ++++++++++++++++++++-------------------------
 block/ll_rw_blk.c          |   4 +-
 drivers/base/class.c       |   7 +
 drivers/base/core.c        |  20 ++-
 drivers/block/aoe/aoeblk.c |  51 +++---
 drivers/block/nbd.c        |  15 +-
 drivers/ide/ide-probe.c    |   2 +-
 drivers/md/dm.c            |   4 +-
 drivers/md/md.c            |   8 +-
 fs/block_dev.c             |   8 +-
 fs/partitions/check.c      | 315 ++++++++++++----------------------
 include/linux/genhd.h      |  37 ++--
 init/do_mounts.c           | 108 +-----------
 13 files changed, 385 insertions(+), 610 deletions(-)

(limited to 'include')

diff --git a/block/genhd.c b/block/genhd.c
index 69aa7389d484..5e4ab4b37d9f 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -17,9 +17,10 @@
 #include <linux/buffer_head.h>
 #include <linux/mutex.h>
 
-struct kset *block_kset;
-static struct kset_uevent_ops block_uevent_ops;
-static DEFINE_MUTEX(block_subsys_lock);
+static DEFINE_MUTEX(block_class_lock);
+#ifndef CONFIG_SYSFS_DEPRECATED
+struct kobject *block_depr;
+#endif
 
 /*
  * Can be deleted altogether. Later.
@@ -38,19 +39,17 @@ static inline int major_to_index(int major)
 }
 
 #ifdef CONFIG_PROC_FS
-
 void blkdev_show(struct seq_file *f, off_t offset)
 {
 	struct blk_major_name *dp;
 
 	if (offset < BLKDEV_MAJOR_HASH_SIZE) {
-		mutex_lock(&block_subsys_lock);
+		mutex_lock(&block_class_lock);
 		for (dp = major_names[offset]; dp; dp = dp->next)
 			seq_printf(f, "%3d %s\n", dp->major, dp->name);
-		mutex_unlock(&block_subsys_lock);
+		mutex_unlock(&block_class_lock);
 	}
 }
-
 #endif /* CONFIG_PROC_FS */
 
 int register_blkdev(unsigned int major, const char *name)
@@ -58,7 +57,7 @@ int register_blkdev(unsigned int major, const char *name)
 	struct blk_major_name **n, *p;
 	int index, ret = 0;
 
-	mutex_lock(&block_subsys_lock);
+	mutex_lock(&block_class_lock);
 
 	/* temporary */
 	if (major == 0) {
@@ -103,7 +102,7 @@ int register_blkdev(unsigned int major, const char *name)
 		kfree(p);
 	}
 out:
-	mutex_unlock(&block_subsys_lock);
+	mutex_unlock(&block_class_lock);
 	return ret;
 }
 
@@ -115,7 +114,7 @@ void unregister_blkdev(unsigned int major, const char *name)
 	struct blk_major_name *p = NULL;
 	int index = major_to_index(major);
 
-	mutex_lock(&block_subsys_lock);
+	mutex_lock(&block_class_lock);
 	for (n = &major_names[index]; *n; n = &(*n)->next)
 		if ((*n)->major == major)
 			break;
@@ -125,7 +124,7 @@ void unregister_blkdev(unsigned int major, const char *name)
 		p = *n;
 		*n = p->next;
 	}
-	mutex_unlock(&block_subsys_lock);
+	mutex_unlock(&block_class_lock);
 	kfree(p);
 }
 
@@ -138,29 +137,30 @@ static struct kobj_map *bdev_map;
  * range must be nonzero
  * The hash chain is sorted on range, so that subranges can override.
  */
-void blk_register_region(dev_t dev, unsigned long range, struct module *module,
+void blk_register_region(dev_t devt, unsigned long range, struct module *module,
 			 struct kobject *(*probe)(dev_t, int *, void *),
 			 int (*lock)(dev_t, void *), void *data)
 {
-	kobj_map(bdev_map, dev, range, module, probe, lock, data);
+	kobj_map(bdev_map, devt, range, module, probe, lock, data);
 }
 
 EXPORT_SYMBOL(blk_register_region);
 
-void blk_unregister_region(dev_t dev, unsigned long range)
+void blk_unregister_region(dev_t devt, unsigned long range)
 {
-	kobj_unmap(bdev_map, dev, range);
+	kobj_unmap(bdev_map, devt, range);
 }
 
 EXPORT_SYMBOL(blk_unregister_region);
 
-static struct kobject *exact_match(dev_t dev, int *part, void *data)
+static struct kobject *exact_match(dev_t devt, int *part, void *data)
 {
 	struct gendisk *p = data;
-	return &p->kobj;
+
+	return &p->dev.kobj;
 }
 
-static int exact_lock(dev_t dev, void *data)
+static int exact_lock(dev_t devt, void *data)
 {
 	struct gendisk *p = data;
 
@@ -195,8 +195,6 @@ void unlink_gendisk(struct gendisk *disk)
 			      disk->minors);
 }
 
-#define to_disk(obj) container_of(obj,struct gendisk,kobj)
-
 /**
  * get_gendisk - get partitioning information for a given device
  * @dev: device to get partitioning information for
@@ -204,10 +202,12 @@ void unlink_gendisk(struct gendisk *disk)
  * This function gets the structure containing partitioning
  * information for the given device @dev.
  */
-struct gendisk *get_gendisk(dev_t dev, int *part)
+struct gendisk *get_gendisk(dev_t devt, int *part)
 {
-	struct kobject *kobj = kobj_lookup(bdev_map, dev, part);
-	return  kobj ? to_disk(kobj) : NULL;
+	struct kobject *kobj = kobj_lookup(bdev_map, devt, part);
+	struct device *dev = kobj_to_dev(kobj);
+
+	return  kobj ? dev_to_disk(dev) : NULL;
 }
 
 /*
@@ -217,13 +217,17 @@ struct gendisk *get_gendisk(dev_t dev, int *part)
  */
 void __init printk_all_partitions(void)
 {
-	int n;
+	struct device *dev;
 	struct gendisk *sgp;
+	char buf[BDEVNAME_SIZE];
+	int n;
 
-	mutex_lock(&block_subsys_lock);
+	mutex_lock(&block_class_lock);
 	/* For each block device... */
-	list_for_each_entry(sgp, &block_kset->list, kobj.entry) {
-		char buf[BDEVNAME_SIZE];
+	list_for_each_entry(dev, &block_class.devices, node) {
+		if (dev->type != &disk_type)
+			continue;
+		sgp = dev_to_disk(dev);
 		/*
 		 * Don't show empty devices or things that have been surpressed
 		 */
@@ -256,38 +260,46 @@ void __init printk_all_partitions(void)
 				sgp->major, n + 1 + sgp->first_minor,
 				(unsigned long long)sgp->part[n]->nr_sects >> 1,
 				disk_name(sgp, n + 1, buf));
-		} /* partition subloop */
-	} /* Block device loop */
+		}
+	}
 
-	mutex_unlock(&block_subsys_lock);
-	return;
+	mutex_unlock(&block_class_lock);
 }
 
 #ifdef CONFIG_PROC_FS
 /* iterator */
 static void *part_start(struct seq_file *part, loff_t *pos)
 {
-	struct list_head *p;
-	loff_t l = *pos;
+	loff_t k = *pos;
+	struct device *dev;
 
-	mutex_lock(&block_subsys_lock);
-	list_for_each(p, &block_kset->list)
-		if (!l--)
-			return list_entry(p, struct gendisk, kobj.entry);
+	mutex_lock(&block_class_lock);
+	list_for_each_entry(dev, &block_class.devices, node) {
+		if (dev->type != &disk_type)
+			continue;
+		if (!k--)
+			return dev_to_disk(dev);
+	}
 	return NULL;
 }
 
 static void *part_next(struct seq_file *part, void *v, loff_t *pos)
 {
-	struct list_head *p = ((struct gendisk *)v)->kobj.entry.next;
+	struct gendisk *gp = v;
+	struct device *dev;
 	++*pos;
-	return p==&block_kset->list ? NULL :
-		list_entry(p, struct gendisk, kobj.entry);
+	list_for_each_entry(dev, &gp->dev.node, node) {
+		if (&dev->node == &block_class.devices)
+			return NULL;
+		if (dev->type == &disk_type)
+			return dev_to_disk(dev);
+	}
+	return NULL;
 }
 
 static void part_stop(struct seq_file *part, void *v)
 {
-	mutex_unlock(&block_subsys_lock);
+	mutex_unlock(&block_class_lock);
 }
 
 static int show_partition(struct seq_file *part, void *v)
@@ -296,7 +308,7 @@ static int show_partition(struct seq_file *part, void *v)
 	int n;
 	char buf[BDEVNAME_SIZE];
 
-	if (&sgp->kobj.entry == block_kset->list.next)
+	if (&sgp->dev.node == block_class.devices.next)
 		seq_puts(part, "major minor  #blocks  name\n\n");
 
 	/* Don't show non-partitionable removeable devices or empty devices */
@@ -326,109 +338,81 @@ static int show_partition(struct seq_file *part, void *v)
 }
 
 struct seq_operations partitions_op = {
-	.start =part_start,
-	.next =	part_next,
-	.stop =	part_stop,
-	.show =	show_partition
+	.start	= part_start,
+	.next	= part_next,
+	.stop	= part_stop,
+	.show	= show_partition
 };
 #endif
 
 
 extern int blk_dev_init(void);
 
-static struct kobject *base_probe(dev_t dev, int *part, void *data)
+static struct kobject *base_probe(dev_t devt, int *part, void *data)
 {
-	if (request_module("block-major-%d-%d", MAJOR(dev), MINOR(dev)) > 0)
+	if (request_module("block-major-%d-%d", MAJOR(devt), MINOR(devt)) > 0)
 		/* Make old-style 2.4 aliases work */
-		request_module("block-major-%d", MAJOR(dev));
+		request_module("block-major-%d", MAJOR(devt));
 	return NULL;
 }
 
 static int __init genhd_device_init(void)
 {
-	bdev_map = kobj_map_init(base_probe, &block_subsys_lock);
+	class_register(&block_class);
+	bdev_map = kobj_map_init(base_probe, &block_class_lock);
 	blk_dev_init();
-	block_kset = kset_create_and_add("block", &block_uevent_ops, NULL);
-	if (!block_kset) {
-		printk(KERN_WARNING "%s: kset_create error\n", __FUNCTION__);
-		return -ENOMEM;
-	}
+
+#ifndef CONFIG_SYSFS_DEPRECATED
+	/* create top-level block dir */
+	block_depr = kobject_create_and_add("block", NULL);
+#endif
 	return 0;
 }
 
 subsys_initcall(genhd_device_init);
 
-
-
-/*
- * kobject & sysfs bindings for block devices
- */
-static ssize_t disk_attr_show(struct kobject *kobj, struct attribute *attr,
-			      char *page)
+static ssize_t disk_range_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
 {
-	struct gendisk *disk = to_disk(kobj);
-	struct disk_attribute *disk_attr =
-		container_of(attr,struct disk_attribute,attr);
-	ssize_t ret = -EIO;
+	struct gendisk *disk = dev_to_disk(dev);
 
-	if (disk_attr->show)
-		ret = disk_attr->show(disk,page);
-	return ret;
+	return sprintf(buf, "%d\n", disk->minors);
 }
 
-static ssize_t disk_attr_store(struct kobject * kobj, struct attribute * attr,
-			       const char *page, size_t count)
+static ssize_t disk_removable_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
 {
-	struct gendisk *disk = to_disk(kobj);
-	struct disk_attribute *disk_attr =
-		container_of(attr,struct disk_attribute,attr);
-	ssize_t ret = 0;
+	struct gendisk *disk = dev_to_disk(dev);
 
-	if (disk_attr->store)
-		ret = disk_attr->store(disk, page, count);
-	return ret;
+	return sprintf(buf, "%d\n",
+		       (disk->flags & GENHD_FL_REMOVABLE ? 1 : 0));
 }
 
-static struct sysfs_ops disk_sysfs_ops = {
-	.show	= &disk_attr_show,
-	.store	= &disk_attr_store,
-};
-
-static ssize_t disk_uevent_store(struct gendisk * disk,
-				 const char *buf, size_t count)
+static ssize_t disk_size_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
-	kobject_uevent(&disk->kobj, KOBJ_ADD);
-	return count;
-}
-static ssize_t disk_dev_read(struct gendisk * disk, char *page)
-{
-	dev_t base = MKDEV(disk->major, disk->first_minor); 
-	return print_dev_t(page, base);
-}
-static ssize_t disk_range_read(struct gendisk * disk, char *page)
-{
-	return sprintf(page, "%d\n", disk->minors);
-}
-static ssize_t disk_removable_read(struct gendisk * disk, char *page)
-{
-	return sprintf(page, "%d\n",
-		       (disk->flags & GENHD_FL_REMOVABLE ? 1 : 0));
+	struct gendisk *disk = dev_to_disk(dev);
 
+	return sprintf(buf, "%llu\n", (unsigned long long)get_capacity(disk));
 }
-static ssize_t disk_size_read(struct gendisk * disk, char *page)
-{
-	return sprintf(page, "%llu\n", (unsigned long long)get_capacity(disk));
-}
-static ssize_t disk_capability_read(struct gendisk *disk, char *page)
+
+static ssize_t disk_capability_show(struct device *dev,
+				    struct device_attribute *attr, char *buf)
 {
-	return sprintf(page, "%x\n", disk->flags);
+	struct gendisk *disk = dev_to_disk(dev);
+
+	return sprintf(buf, "%x\n", disk->flags);
 }
-static ssize_t disk_stats_read(struct gendisk * disk, char *page)
+
+static ssize_t disk_stat_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
+	struct gendisk *disk = dev_to_disk(dev);
+
 	preempt_disable();
 	disk_round_stats(disk);
 	preempt_enable();
-	return sprintf(page,
+	return sprintf(buf,
 		"%8lu %8lu %8llu %8u "
 		"%8lu %8lu %8llu %8u "
 		"%8u %8u %8u"
@@ -445,40 +429,21 @@ static ssize_t disk_stats_read(struct gendisk * disk, char *page)
 		jiffies_to_msecs(disk_stat_read(disk, io_ticks)),
 		jiffies_to_msecs(disk_stat_read(disk, time_in_queue)));
 }
-static struct disk_attribute disk_attr_uevent = {
-	.attr = {.name = "uevent", .mode = S_IWUSR },
-	.store	= disk_uevent_store
-};
-static struct disk_attribute disk_attr_dev = {
-	.attr = {.name = "dev", .mode = S_IRUGO },
-	.show	= disk_dev_read
-};
-static struct disk_attribute disk_attr_range = {
-	.attr = {.name = "range", .mode = S_IRUGO },
-	.show	= disk_range_read
-};
-static struct disk_attribute disk_attr_removable = {
-	.attr = {.name = "removable", .mode = S_IRUGO },
-	.show	= disk_removable_read
-};
-static struct disk_attribute disk_attr_size = {
-	.attr = {.name = "size", .mode = S_IRUGO },
-	.show	= disk_size_read
-};
-static struct disk_attribute disk_attr_capability = {
-	.attr = {.name = "capability", .mode = S_IRUGO },
-	.show	= disk_capability_read
-};
-static struct disk_attribute disk_attr_stat = {
-	.attr = {.name = "stat", .mode = S_IRUGO },
-	.show	= disk_stats_read
-};
 
 #ifdef CONFIG_FAIL_MAKE_REQUEST
+static ssize_t disk_fail_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct gendisk *disk = dev_to_disk(dev);
+
+	return sprintf(buf, "%d\n", disk->flags & GENHD_FL_FAIL ? 1 : 0);
+}
 
-static ssize_t disk_fail_store(struct gendisk * disk,
+static ssize_t disk_fail_store(struct device *dev,
+			       struct device_attribute *attr,
 			       const char *buf, size_t count)
 {
+	struct gendisk *disk = dev_to_disk(dev);
 	int i;
 
 	if (count > 0 && sscanf(buf, "%d", &i) > 0) {
@@ -490,134 +455,100 @@ static ssize_t disk_fail_store(struct gendisk * disk,
 
 	return count;
 }
-static ssize_t disk_fail_read(struct gendisk * disk, char *page)
-{
-	return sprintf(page, "%d\n", disk->flags & GENHD_FL_FAIL ? 1 : 0);
-}
-static struct disk_attribute disk_attr_fail = {
-	.attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR },
-	.store	= disk_fail_store,
-	.show	= disk_fail_read
-};
 
 #endif
 
-static struct attribute * default_attrs[] = {
-	&disk_attr_uevent.attr,
-	&disk_attr_dev.attr,
-	&disk_attr_range.attr,
-	&disk_attr_removable.attr,
-	&disk_attr_size.attr,
-	&disk_attr_stat.attr,
-	&disk_attr_capability.attr,
+static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL);
+static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL);
+static DEVICE_ATTR(size, S_IRUGO, disk_size_show, NULL);
+static DEVICE_ATTR(capability, S_IRUGO, disk_capability_show, NULL);
+static DEVICE_ATTR(stat, S_IRUGO, disk_stat_show, NULL);
 #ifdef CONFIG_FAIL_MAKE_REQUEST
-	&disk_attr_fail.attr,
+static struct device_attribute dev_attr_fail =
+	__ATTR(make-it-fail, S_IRUGO|S_IWUSR, disk_fail_show, disk_fail_store);
 #endif
-	NULL,
+
+static struct attribute *disk_attrs[] = {
+	&dev_attr_range.attr,
+	&dev_attr_removable.attr,
+	&dev_attr_size.attr,
+	&dev_attr_capability.attr,
+	&dev_attr_stat.attr,
+#ifdef CONFIG_FAIL_MAKE_REQUEST
+	&dev_attr_fail.attr,
+#endif
+	NULL
+};
+
+static struct attribute_group disk_attr_group = {
+	.attrs = disk_attrs,
+};
+
+static struct attribute_group *disk_attr_groups[] = {
+	&disk_attr_group,
+	NULL
 };
 
-static void disk_release(struct kobject * kobj)
+static void disk_release(struct device *dev)
 {
-	struct gendisk *disk = to_disk(kobj);
+	struct gendisk *disk = dev_to_disk(dev);
+
 	kfree(disk->random);
 	kfree(disk->part);
 	free_disk_stats(disk);
 	kfree(disk);
 }
-
-static struct kobj_type ktype_block = {
-	.release	= disk_release,
-	.sysfs_ops	= &disk_sysfs_ops,
-	.default_attrs	= default_attrs,
+struct class block_class = {
+	.name		= "block",
 };
 
-extern struct kobj_type ktype_part;
-
-static int block_uevent_filter(struct kset *kset, struct kobject *kobj)
-{
-	struct kobj_type *ktype = get_ktype(kobj);
-
-	return ((ktype == &ktype_block) || (ktype == &ktype_part));
-}
-
-static int block_uevent(struct kset *kset, struct kobject *kobj,
-			struct kobj_uevent_env *env)
-{
-	struct kobj_type *ktype = get_ktype(kobj);
-	struct device *physdev;
-	struct gendisk *disk;
-	struct hd_struct *part;
-
-	if (ktype == &ktype_block) {
-		disk = container_of(kobj, struct gendisk, kobj);
-		add_uevent_var(env, "MINOR=%u", disk->first_minor);
-	} else if (ktype == &ktype_part) {
-		disk = container_of(kobj->parent, struct gendisk, kobj);
-		part = container_of(kobj, struct hd_struct, kobj);
-		add_uevent_var(env, "MINOR=%u",
-			       disk->first_minor + part->partno);
-	} else
-		return 0;
-
-	add_uevent_var(env, "MAJOR=%u", disk->major);
-
-	/* add physical device, backing this device  */
-	physdev = disk->driverfs_dev;
-	if (physdev) {
-		char *path = kobject_get_path(&physdev->kobj, GFP_KERNEL);
-
-		add_uevent_var(env, "PHYSDEVPATH=%s", path);
-		kfree(path);
-
-		if (physdev->bus)
-			add_uevent_var(env, "PHYSDEVBUS=%s", physdev->bus->name);
-
-		if (physdev->driver)
-			add_uevent_var(env, physdev->driver->name);
-	}
-
-	return 0;
-}
-
-static struct kset_uevent_ops block_uevent_ops = {
-	.filter		= block_uevent_filter,
-	.uevent		= block_uevent,
+struct device_type disk_type = {
+	.name		= "disk",
+	.groups		= disk_attr_groups,
+	.release	= disk_release,
 };
 
 /*
  * aggregate disk stat collector.  Uses the same stats that the sysfs
  * entries do, above, but makes them available through one seq_file.
- * Watching a few disks may be efficient through sysfs, but watching
- * all of them will be more efficient through this interface.
  *
  * The output looks suspiciously like /proc/partitions with a bunch of
  * extra fields.
  */
 
-/* iterator */
 static void *diskstats_start(struct seq_file *part, loff_t *pos)
 {
 	loff_t k = *pos;
-	struct list_head *p;
+	struct device *dev;
 
-	mutex_lock(&block_subsys_lock);
-	list_for_each(p, &block_kset->list)
+	mutex_lock(&block_class_lock);
+	list_for_each_entry(dev, &block_class.devices, node) {
+		if (dev->type != &disk_type)
+			continue;
 		if (!k--)
-			return list_entry(p, struct gendisk, kobj.entry);
+			return dev_to_disk(dev);
+	}
 	return NULL;
 }
 
 static void *diskstats_next(struct seq_file *part, void *v, loff_t *pos)
 {
-	struct list_head *p = ((struct gendisk *)v)->kobj.entry.next;
+	struct gendisk *gp = v;
+	struct device *dev;
+
 	++*pos;
-	return p==&block_kset->list ? NULL :
-		list_entry(p, struct gendisk, kobj.entry);
+	list_for_each_entry(dev, &gp->dev.node, node) {
+		if (&dev->node == &block_class.devices)
+			return NULL;
+		if (dev->type == &disk_type)
+			return dev_to_disk(dev);
+	}
+	return NULL;
 }
 
 static void diskstats_stop(struct seq_file *part, void *v)
 {
-	mutex_unlock(&block_subsys_lock);
+	mutex_unlock(&block_class_lock);
 }
 
 static int diskstats_show(struct seq_file *s, void *v)
@@ -627,7 +558,7 @@ static int diskstats_show(struct seq_file *s, void *v)
 	int n = 0;
 
 	/*
-	if (&sgp->kobj.entry == block_kset->list.next)
+	if (&gp->dev.kobj.entry == block_class.devices.next)
 		seq_puts(s,	"major minor name"
 				"     rio rmerge rsect ruse wio wmerge "
 				"wsect wuse running use aveq"
@@ -681,7 +612,7 @@ static void media_change_notify_thread(struct work_struct *work)
 	 * set enviroment vars to indicate which event this is for
 	 * so that user space will know to go check the media status.
 	 */
-	kobject_uevent_env(&gd->kobj, KOBJ_CHANGE, envp);
+	kobject_uevent_env(&gd->dev.kobj, KOBJ_CHANGE, envp);
 	put_device(gd->driverfs_dev);
 }
 
@@ -692,6 +623,25 @@ void genhd_media_change_notify(struct gendisk *disk)
 }
 EXPORT_SYMBOL_GPL(genhd_media_change_notify);
 
+dev_t blk_lookup_devt(const char *name)
+{
+	struct device *dev;
+	dev_t devt = MKDEV(0, 0);
+
+	mutex_lock(&block_class_lock);
+	list_for_each_entry(dev, &block_class.devices, node) {
+		if (strcmp(dev->bus_id, name) == 0) {
+			devt = dev->devt;
+			break;
+		}
+	}
+	mutex_unlock(&block_class_lock);
+
+	return devt;
+}
+
+EXPORT_SYMBOL(blk_lookup_devt);
+
 struct gendisk *alloc_disk(int minors)
 {
 	return alloc_disk_node(minors, -1);
@@ -719,10 +669,10 @@ struct gendisk *alloc_disk_node(int minors, int node_id)
 			}
 		}
 		disk->minors = minors;
-		disk->kobj.kset = block_kset;
-		disk->kobj.ktype = &ktype_block;
-		kobject_init(&disk->kobj);
 		rand_initialize_disk(disk);
+		disk->dev.class = &block_class;
+		disk->dev.type = &disk_type;
+		device_initialize(&disk->dev);
 		INIT_WORK(&disk->async_notify,
 			media_change_notify_thread);
 	}
@@ -742,7 +692,7 @@ struct kobject *get_disk(struct gendisk *disk)
 	owner = disk->fops->owner;
 	if (owner && !try_module_get(owner))
 		return NULL;
-	kobj = kobject_get(&disk->kobj);
+	kobj = kobject_get(&disk->dev.kobj);
 	if (kobj == NULL) {
 		module_put(owner);
 		return NULL;
@@ -756,7 +706,7 @@ EXPORT_SYMBOL(get_disk);
 void put_disk(struct gendisk *disk)
 {
 	if (disk)
-		kobject_put(&disk->kobj);
+		kobject_put(&disk->dev.kobj);
 }
 
 EXPORT_SYMBOL(put_disk);
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 8b919940b2ab..3887b2a33ed0 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -4182,7 +4182,7 @@ int blk_register_queue(struct gendisk *disk)
 	if (!q || !q->request_fn)
 		return -ENXIO;
 
-	q->kobj.parent = kobject_get(&disk->kobj);
+	q->kobj.parent = kobject_get(&disk->dev.kobj);
 
 	ret = kobject_add(&q->kobj);
 	if (ret < 0)
@@ -4209,6 +4209,6 @@ void blk_unregister_queue(struct gendisk *disk)
 
 		kobject_uevent(&q->kobj, KOBJ_REMOVE);
 		kobject_del(&q->kobj);
-		kobject_put(&disk->kobj);
+		kobject_put(&disk->dev.kobj);
 	}
 }
diff --git a/drivers/base/class.c b/drivers/base/class.c
index ba6745b0fd2f..624b3316e938 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -17,6 +17,7 @@
 #include <linux/kdev_t.h>
 #include <linux/err.h>
 #include <linux/slab.h>
+#include <linux/genhd.h>
 #include "base.h"
 
 #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr)
@@ -149,7 +150,13 @@ int class_register(struct class * cls)
 	if (error)
 		return error;
 
+#ifdef CONFIG_SYSFS_DEPRECATED
+	/* let the block class directory show up in the root of sysfs */
+	if (cls != &block_class)
+		cls->subsys.kobj.kset = class_kset;
+#else
 	cls->subsys.kobj.kset = class_kset;
+#endif
 	cls->subsys.kobj.ktype = &class_ktype;
 
 	error = kset_register(&cls->subsys);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 13cae18936c5..06e8738ab263 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -671,14 +671,15 @@ static int device_add_class_symlinks(struct device *dev)
 
 #ifdef CONFIG_SYSFS_DEPRECATED
 	/* stacked class devices need a symlink in the class directory */
-	if (dev->kobj.parent != &dev->class->subsys.kobj) {
+	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	    dev->type != &part_type) {
 		error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj,
 					  dev->bus_id);
 		if (error)
 			goto out_subsys;
 	}
 
-	if (dev->parent) {
+	if (dev->parent && dev->type != &part_type) {
 		struct device *parent = dev->parent;
 		char *class_name;
 
@@ -707,10 +708,11 @@ static int device_add_class_symlinks(struct device *dev)
 	return 0;
 
 out_device:
-	if (dev->parent)
+	if (dev->parent && dev->type != &part_type)
 		sysfs_remove_link(&dev->kobj, "device");
 out_busid:
-	if (dev->kobj.parent != &dev->class->subsys.kobj)
+	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	    dev->type != &part_type)
 		sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
 #else
 	/* link in the class directory pointing to the device */
@@ -719,7 +721,7 @@ out_busid:
 	if (error)
 		goto out_subsys;
 
-	if (dev->parent) {
+	if (dev->parent && dev->type != &part_type) {
 		error = sysfs_create_link(&dev->kobj, &dev->parent->kobj,
 					  "device");
 		if (error)
@@ -743,7 +745,7 @@ static void device_remove_class_symlinks(struct device *dev)
 		return;
 
 #ifdef CONFIG_SYSFS_DEPRECATED
-	if (dev->parent) {
+	if (dev->parent && dev->type != &part_type) {
 		char *class_name;
 
 		class_name = make_class_name(dev->class->name, &dev->kobj);
@@ -754,10 +756,11 @@ static void device_remove_class_symlinks(struct device *dev)
 		sysfs_remove_link(&dev->kobj, "device");
 	}
 
-	if (dev->kobj.parent != &dev->class->subsys.kobj)
+	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	    dev->type != &part_type)
 		sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
 #else
-	if (dev->parent)
+	if (dev->parent && dev->type != &part_type)
 		sysfs_remove_link(&dev->kobj, "device");
 
 	sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
@@ -925,6 +928,7 @@ struct device * get_device(struct device * dev)
  */
 void put_device(struct device * dev)
 {
+	/* might_sleep(); */
 	if (dev)
 		kobject_put(&dev->kobj);
 }
diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
index ad00b3d94711..826d12381e21 100644
--- a/drivers/block/aoe/aoeblk.c
+++ b/drivers/block/aoe/aoeblk.c
@@ -15,8 +15,10 @@
 
 static struct kmem_cache *buf_pool_cache;
 
-static ssize_t aoedisk_show_state(struct gendisk * disk, char *page)
+static ssize_t aoedisk_show_state(struct device *dev,
+				  struct device_attribute *attr, char *page)
 {
+	struct gendisk *disk = dev_to_disk(dev);
 	struct aoedev *d = disk->private_data;
 
 	return snprintf(page, PAGE_SIZE,
@@ -26,50 +28,47 @@ static ssize_t aoedisk_show_state(struct gendisk * disk, char *page)
 			(d->nopen && !(d->flags & DEVFL_UP)) ? ",closewait" : "");
 	/* I'd rather see nopen exported so we can ditch closewait */
 }
-static ssize_t aoedisk_show_mac(struct gendisk * disk, char *page)
+static ssize_t aoedisk_show_mac(struct device *dev,
+				struct device_attribute *attr, char *page)
 {
+	struct gendisk *disk = dev_to_disk(dev);
 	struct aoedev *d = disk->private_data;
 
 	return snprintf(page, PAGE_SIZE, "%012llx\n",
 			(unsigned long long)mac_addr(d->addr));
 }
-static ssize_t aoedisk_show_netif(struct gendisk * disk, char *page)
+static ssize_t aoedisk_show_netif(struct device *dev,
+				  struct device_attribute *attr, char *page)
 {
+	struct gendisk *disk = dev_to_disk(dev);
 	struct aoedev *d = disk->private_data;
 
 	return snprintf(page, PAGE_SIZE, "%s\n", d->ifp->name);
 }
 /* firmware version */
-static ssize_t aoedisk_show_fwver(struct gendisk * disk, char *page)
+static ssize_t aoedisk_show_fwver(struct device *dev,
+				  struct device_attribute *attr, char *page)
 {
+	struct gendisk *disk = dev_to_disk(dev);
 	struct aoedev *d = disk->private_data;
 
 	return snprintf(page, PAGE_SIZE, "0x%04x\n", (unsigned int) d->fw_ver);
 }
 
-static struct disk_attribute disk_attr_state = {
-	.attr = {.name = "state", .mode = S_IRUGO },
-	.show = aoedisk_show_state
-};
-static struct disk_attribute disk_attr_mac = {
-	.attr = {.name = "mac", .mode = S_IRUGO },
-	.show = aoedisk_show_mac
-};
-static struct disk_attribute disk_attr_netif = {
-	.attr = {.name = "netif", .mode = S_IRUGO },
-	.show = aoedisk_show_netif
-};
-static struct disk_attribute disk_attr_fwver = {
-	.attr = {.name = "firmware-version", .mode = S_IRUGO },
-	.show = aoedisk_show_fwver
+static DEVICE_ATTR(state, S_IRUGO, aoedisk_show_state, NULL);
+static DEVICE_ATTR(mac, S_IRUGO, aoedisk_show_mac, NULL);
+static DEVICE_ATTR(netif, S_IRUGO, aoedisk_show_netif, NULL);
+static struct device_attribute dev_attr_firmware_version = {
+	.attr = { .name = "firmware-version", .mode = S_IRUGO, .owner = THIS_MODULE },
+	.show = aoedisk_show_fwver,
 };
 
 static struct attribute *aoe_attrs[] = {
-	&disk_attr_state.attr,
-	&disk_attr_mac.attr,
-	&disk_attr_netif.attr,
-	&disk_attr_fwver.attr,
-	NULL
+	&dev_attr_state.attr,
+	&dev_attr_mac.attr,
+	&dev_attr_netif.attr,
+	&dev_attr_firmware_version.attr,
+	NULL,
 };
 
 static const struct attribute_group attr_group = {
@@ -79,12 +78,12 @@ static const struct attribute_group attr_group = {
 static int
 aoedisk_add_sysfs(struct aoedev *d)
 {
-	return sysfs_create_group(&d->gd->kobj, &attr_group);
+	return sysfs_create_group(&d->gd->dev.kobj, &attr_group);
 }
 void
 aoedisk_rm_sysfs(struct aoedev *d)
 {
-	sysfs_remove_group(&d->gd->kobj, &attr_group);
+	sysfs_remove_group(&d->gd->dev.kobj, &attr_group);
 }
 
 static int
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index b4c0888aedc3..ba9b17e507e0 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -375,14 +375,17 @@ harderror:
 	return NULL;
 }
 
-static ssize_t pid_show(struct gendisk *disk, char *page)
+static ssize_t pid_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
 {
-	return sprintf(page, "%ld\n",
+	struct gendisk *disk = dev_to_disk(dev);
+
+	return sprintf(buf, "%ld\n",
 		(long) ((struct nbd_device *)disk->private_data)->pid);
 }
 
-static struct disk_attribute pid_attr = {
-	.attr = { .name = "pid", .mode = S_IRUGO },
+static struct device_attribute pid_attr = {
+	.attr = { .name = "pid", .mode = S_IRUGO, .owner = THIS_MODULE },
 	.show = pid_show,
 };
 
@@ -394,7 +397,7 @@ static int nbd_do_it(struct nbd_device *lo)
 	BUG_ON(lo->magic != LO_MAGIC);
 
 	lo->pid = current->pid;
-	ret = sysfs_create_file(&lo->disk->kobj, &pid_attr.attr);
+	ret = sysfs_create_file(&lo->disk->dev.kobj, &pid_attr.attr);
 	if (ret) {
 		printk(KERN_ERR "nbd: sysfs_create_file failed!");
 		return ret;
@@ -403,7 +406,7 @@ static int nbd_do_it(struct nbd_device *lo)
 	while ((req = nbd_read_stat(lo)) != NULL)
 		nbd_end_request(req);
 
-	sysfs_remove_file(&lo->disk->kobj, &pid_attr.attr);
+	sysfs_remove_file(&lo->disk->dev.kobj, &pid_attr.attr);
 	return 0;
 }
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 2994523be7bf..0cb3d2bb3ab9 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1173,7 +1173,7 @@ static struct kobject *exact_match(dev_t dev, int *part, void *data)
 {
 	struct gendisk *p = data;
 	*part &= (1 << PARTN_BITS) - 1;
-	return &p->kobj;
+	return &p->dev.kobj;
 }
 
 static int exact_lock(dev_t dev, void *data)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 88c0fd657825..f2d24eb3208c 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1109,7 +1109,7 @@ static void event_callback(void *context)
 	list_splice_init(&md->uevent_list, &uevents);
 	spin_unlock_irqrestore(&md->uevent_lock, flags);
 
-	dm_send_uevents(&uevents, &md->disk->kobj);
+	dm_send_uevents(&uevents, &md->disk->dev.kobj);
 
 	atomic_inc(&md->event_nr);
 	wake_up(&md->eventq);
@@ -1530,7 +1530,7 @@ out:
  *---------------------------------------------------------------*/
 void dm_kobject_uevent(struct mapped_device *md)
 {
-	kobject_uevent(&md->disk->kobj, KOBJ_CHANGE);
+	kobject_uevent(&md->disk->dev.kobj, KOBJ_CHANGE);
 }
 
 uint32_t dm_next_uevent_seq(struct mapped_device *md)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index c5030863d004..f79efb359215 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1396,9 +1396,9 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		goto fail;
 
 	if (rdev->bdev->bd_part)
-		ko = &rdev->bdev->bd_part->kobj;
+		ko = &rdev->bdev->bd_part->dev.kobj;
 	else
-		ko = &rdev->bdev->bd_disk->kobj;
+		ko = &rdev->bdev->bd_disk->dev.kobj;
 	if ((err = sysfs_create_link(&rdev->kobj, ko, "block"))) {
 		kobject_del(&rdev->kobj);
 		goto fail;
@@ -3083,7 +3083,7 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	add_disk(disk);
 	mddev->gendisk = disk;
 	mutex_unlock(&disks_mutex);
-	error = kobject_init_and_add(&mddev->kobj, &md_ktype, &disk->kobj,
+	error = kobject_init_and_add(&mddev->kobj, &md_ktype, &disk->dev.kobj,
 				     "%s", "md");
 	if (error)
 		printk(KERN_WARNING "md: cannot register %s/md - name in use\n",
@@ -3361,7 +3361,7 @@ static int do_md_run(mddev_t * mddev)
 
 	mddev->changed = 1;
 	md_new_event(mddev);
-	kobject_uevent(&mddev->gendisk->kobj, KOBJ_CHANGE);
+	kobject_uevent(&mddev->gendisk->dev.kobj, KOBJ_CHANGE);
 	return 0;
 }
 
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 993f78c55221..e48a630ae266 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -738,9 +738,9 @@ EXPORT_SYMBOL(bd_release);
 static struct kobject *bdev_get_kobj(struct block_device *bdev)
 {
 	if (bdev->bd_contains != bdev)
-		return kobject_get(&bdev->bd_part->kobj);
+		return kobject_get(&bdev->bd_part->dev.kobj);
 	else
-		return kobject_get(&bdev->bd_disk->kobj);
+		return kobject_get(&bdev->bd_disk->dev.kobj);
 }
 
 static struct kobject *bdev_get_holder(struct block_device *bdev)
@@ -1176,7 +1176,7 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
 				ret = -ENXIO;
 				goto out_first;
 			}
-			kobject_get(&p->kobj);
+			kobject_get(&p->dev.kobj);
 			bdev->bd_part = p;
 			bd_set_size(bdev, (loff_t) p->nr_sects << 9);
 		}
@@ -1299,7 +1299,7 @@ static int __blkdev_put(struct block_device *bdev, int for_part)
 		module_put(owner);
 
 		if (bdev->bd_contains != bdev) {
-			kobject_put(&bdev->bd_part->kobj);
+			kobject_put(&bdev->bd_part->dev.kobj);
 			bdev->bd_part = NULL;
 		}
 		bdev->bd_disk = NULL;
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 9184215f3ef3..97f3f5f064ee 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -195,96 +195,45 @@ check_partition(struct gendisk *hd, struct block_device *bdev)
 	return ERR_PTR(res);
 }
 
-/*
- * sysfs bindings for partitions
- */
-
-struct part_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct hd_struct *,char *);
-	ssize_t (*store)(struct hd_struct *,const char *, size_t);
-};
-
-static ssize_t 
-part_attr_show(struct kobject * kobj, struct attribute * attr, char * page)
+static ssize_t part_start_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
 {
-	struct hd_struct * p = container_of(kobj,struct hd_struct,kobj);
-	struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr);
-	ssize_t ret = 0;
-	if (part_attr->show)
-		ret = part_attr->show(p, page);
-	return ret;
-}
-static ssize_t
-part_attr_store(struct kobject * kobj, struct attribute * attr,
-		const char *page, size_t count)
-{
-	struct hd_struct * p = container_of(kobj,struct hd_struct,kobj);
-	struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr);
-	ssize_t ret = 0;
+	struct hd_struct *p = dev_to_part(dev);
 
-	if (part_attr->store)
-		ret = part_attr->store(p, page, count);
-	return ret;
+	return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect);
 }
 
-static struct sysfs_ops part_sysfs_ops = {
-	.show	=	part_attr_show,
-	.store	=	part_attr_store,
-};
-
-static ssize_t part_uevent_store(struct hd_struct * p,
-				 const char *page, size_t count)
+static ssize_t part_size_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
-	kobject_uevent(&p->kobj, KOBJ_ADD);
-	return count;
+	struct hd_struct *p = dev_to_part(dev);
+	return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects);
 }
-static ssize_t part_dev_read(struct hd_struct * p, char *page)
-{
-	struct gendisk *disk = container_of(p->kobj.parent,struct gendisk,kobj);
-	dev_t dev = MKDEV(disk->major, disk->first_minor + p->partno); 
-	return print_dev_t(page, dev);
-}
-static ssize_t part_start_read(struct hd_struct * p, char *page)
-{
-	return sprintf(page, "%llu\n",(unsigned long long)p->start_sect);
-}
-static ssize_t part_size_read(struct hd_struct * p, char *page)
-{
-	return sprintf(page, "%llu\n",(unsigned long long)p->nr_sects);
-}
-static ssize_t part_stat_read(struct hd_struct * p, char *page)
+
+static ssize_t part_stat_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
 {
-	return sprintf(page, "%8u %8llu %8u %8llu\n",
+	struct hd_struct *p = dev_to_part(dev);
+
+	return sprintf(buf, "%8u %8llu %8u %8llu\n",
 		       p->ios[0], (unsigned long long)p->sectors[0],
 		       p->ios[1], (unsigned long long)p->sectors[1]);
 }
-static struct part_attribute part_attr_uevent = {
-	.attr = {.name = "uevent", .mode = S_IWUSR },
-	.store	= part_uevent_store
-};
-static struct part_attribute part_attr_dev = {
-	.attr = {.name = "dev", .mode = S_IRUGO },
-	.show	= part_dev_read
-};
-static struct part_attribute part_attr_start = {
-	.attr = {.name = "start", .mode = S_IRUGO },
-	.show	= part_start_read
-};
-static struct part_attribute part_attr_size = {
-	.attr = {.name = "size", .mode = S_IRUGO },
-	.show	= part_size_read
-};
-static struct part_attribute part_attr_stat = {
-	.attr = {.name = "stat", .mode = S_IRUGO },
-	.show	= part_stat_read
-};
 
 #ifdef CONFIG_FAIL_MAKE_REQUEST
+static ssize_t part_fail_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct hd_struct *p = dev_to_part(dev);
 
-static ssize_t part_fail_store(struct hd_struct * p,
+	return sprintf(buf, "%d\n", p->make_it_fail);
+}
+
+static ssize_t part_fail_store(struct device *dev,
+			       struct device_attribute *attr,
 			       const char *buf, size_t count)
 {
+	struct hd_struct *p = dev_to_part(dev);
 	int i;
 
 	if (count > 0 && sscanf(buf, "%d", &i) > 0)
@@ -292,49 +241,52 @@ static ssize_t part_fail_store(struct hd_struct * p,
 
 	return count;
 }
-static ssize_t part_fail_read(struct hd_struct * p, char *page)
-{
-	return sprintf(page, "%d\n", p->make_it_fail);
-}
-static struct part_attribute part_attr_fail = {
-	.attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR },
-	.store	= part_fail_store,
-	.show	= part_fail_read
-};
+#endif
 
+static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL);
+static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
+static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL);
+#ifdef CONFIG_FAIL_MAKE_REQUEST
+static struct device_attribute dev_attr_fail =
+	__ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store);
 #endif
 
-static struct attribute * default_attrs[] = {
-	&part_attr_uevent.attr,
-	&part_attr_dev.attr,
-	&part_attr_start.attr,
-	&part_attr_size.attr,
-	&part_attr_stat.attr,
+static struct attribute *part_attrs[] = {
+	&dev_attr_start.attr,
+	&dev_attr_size.attr,
+	&dev_attr_stat.attr,
 #ifdef CONFIG_FAIL_MAKE_REQUEST
-	&part_attr_fail.attr,
+	&dev_attr_fail.attr,
 #endif
-	NULL,
+	NULL
 };
 
-extern struct kset *block_kset;
+static struct attribute_group part_attr_group = {
+	.attrs = part_attrs,
+};
 
-static void part_release(struct kobject *kobj)
+static struct attribute_group *part_attr_groups[] = {
+	&part_attr_group,
+	NULL
+};
+
+static void part_release(struct device *dev)
 {
-	struct hd_struct * p = container_of(kobj,struct hd_struct,kobj);
+	struct hd_struct *p = dev_to_part(dev);
 	kfree(p);
 }
 
-struct kobj_type ktype_part = {
+struct device_type part_type = {
+	.name		= "partition",
+	.groups		= part_attr_groups,
 	.release	= part_release,
-	.default_attrs	= default_attrs,
-	.sysfs_ops	= &part_sysfs_ops,
 };
 
 static inline void partition_sysfs_add_subdir(struct hd_struct *p)
 {
 	struct kobject *k;
 
-	k = kobject_get(&p->kobj);
+	k = kobject_get(&p->dev.kobj);
 	p->holder_dir = kobject_create_and_add("holders", k);
 	kobject_put(k);
 }
@@ -343,7 +295,7 @@ static inline void disk_sysfs_add_subdirs(struct gendisk *disk)
 {
 	struct kobject *k;
 
-	k = kobject_get(&disk->kobj);
+	k = kobject_get(&disk->dev.kobj);
 	disk->holder_dir = kobject_create_and_add("holders", k);
 	disk->slave_dir = kobject_create_and_add("slaves", k);
 	kobject_put(k);
@@ -352,6 +304,7 @@ static inline void disk_sysfs_add_subdirs(struct gendisk *disk)
 void delete_partition(struct gendisk *disk, int part)
 {
 	struct hd_struct *p = disk->part[part-1];
+
 	if (!p)
 		return;
 	if (!p->nr_sects)
@@ -361,113 +314,55 @@ void delete_partition(struct gendisk *disk, int part)
 	p->nr_sects = 0;
 	p->ios[0] = p->ios[1] = 0;
 	p->sectors[0] = p->sectors[1] = 0;
-	sysfs_remove_link(&p->kobj, "subsystem");
 	kobject_unregister(p->holder_dir);
-	kobject_uevent(&p->kobj, KOBJ_REMOVE);
-	kobject_del(&p->kobj);
-	kobject_put(&p->kobj);
+	device_del(&p->dev);
+	put_device(&p->dev);
 }
 
 void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags)
 {
 	struct hd_struct *p;
+	int err;
 
 	p = kzalloc(sizeof(*p), GFP_KERNEL);
 	if (!p)
 		return;
-	
+
 	p->start_sect = start;
 	p->nr_sects = len;
 	p->partno = part;
 	p->policy = disk->policy;
 
-	if (isdigit(disk->kobj.k_name[strlen(disk->kobj.k_name)-1]))
-		kobject_set_name(&p->kobj, "%sp%d",
-				 kobject_name(&disk->kobj), part);
+	if (isdigit(disk->dev.bus_id[strlen(disk->dev.bus_id)-1]))
+		snprintf(p->dev.bus_id, BUS_ID_SIZE,
+		"%sp%d", disk->dev.bus_id, part);
 	else
-		kobject_set_name(&p->kobj, "%s%d",
-				 kobject_name(&disk->kobj),part);
-	p->kobj.parent = &disk->kobj;
-	p->kobj.ktype = &ktype_part;
-	kobject_init(&p->kobj);
-	kobject_add(&p->kobj);
-	if (!disk->part_uevent_suppress)
-		kobject_uevent(&p->kobj, KOBJ_ADD);
-	sysfs_create_link(&p->kobj, &block_kset->kobj, "subsystem");
+		snprintf(p->dev.bus_id, BUS_ID_SIZE,
+			 "%s%d", disk->dev.bus_id, part);
+
+	device_initialize(&p->dev);
+	p->dev.devt = MKDEV(disk->major, disk->first_minor + part);
+	p->dev.class = &block_class;
+	p->dev.type = &part_type;
+	p->dev.parent = &disk->dev;
+	disk->part[part-1] = p;
+
+	/* delay uevent until 'holders' subdir is created */
+	p->dev.uevent_suppress = 1;
+	device_add(&p->dev);
+	partition_sysfs_add_subdir(p);
+	p->dev.uevent_suppress = 0;
 	if (flags & ADDPART_FLAG_WHOLEDISK) {
 		static struct attribute addpartattr = {
 			.name = "whole_disk",
 			.mode = S_IRUSR | S_IRGRP | S_IROTH,
 		};
-
-		sysfs_create_file(&p->kobj, &addpartattr);
+		err = sysfs_create_file(&p->dev.kobj, &addpartattr);
 	}
-	partition_sysfs_add_subdir(p);
-	disk->part[part-1] = p;
-}
 
-static char *make_block_name(struct gendisk *disk)
-{
-	char *name;
-	static char *block_str = "block:";
-	int size;
-	char *s;
-
-	size = strlen(block_str) + strlen(disk->disk_name) + 1;
-	name = kmalloc(size, GFP_KERNEL);
-	if (!name)
-		return NULL;
-	strcpy(name, block_str);
-	strcat(name, disk->disk_name);
-	/* ewww... some of these buggers have / in name... */
-	s = strchr(name, '/');
-	if (s)
-		*s = '!';
-	return name;
-}
-
-static int disk_sysfs_symlinks(struct gendisk *disk)
-{
-	struct device *target = get_device(disk->driverfs_dev);
-	int err;
-	char *disk_name = NULL;
-
-	if (target) {
-		disk_name = make_block_name(disk);
-		if (!disk_name) {
-			err = -ENOMEM;
-			goto err_out;
-		}
-
-		err = sysfs_create_link(&disk->kobj, &target->kobj, "device");
-		if (err)
-			goto err_out_disk_name;
-
-		err = sysfs_create_link(&target->kobj, &disk->kobj, disk_name);
-		if (err)
-			goto err_out_dev_link;
-	}
-
-	err = sysfs_create_link(&disk->kobj, &block_kset->kobj,
-				"subsystem");
-	if (err)
-		goto err_out_disk_name_lnk;
-
-	kfree(disk_name);
-
-	return 0;
-
-err_out_disk_name_lnk:
-	if (target) {
-		sysfs_remove_link(&target->kobj, disk_name);
-err_out_dev_link:
-		sysfs_remove_link(&disk->kobj, "device");
-err_out_disk_name:
-		kfree(disk_name);
-err_out:
-		put_device(target);
-	}
-	return err;
+	/* suppress uevent if the disk supresses it */
+	if (!disk->dev.uevent_suppress)
+		kobject_uevent(&p->dev.kobj, KOBJ_ADD);
 }
 
 /* Not exported, helper to add_disk(). */
@@ -479,19 +374,29 @@ void register_disk(struct gendisk *disk)
 	struct hd_struct *p;
 	int err;
 
-	kobject_set_name(&disk->kobj, "%s", disk->disk_name);
-	/* ewww... some of these buggers have / in name... */
-	s = strchr(disk->kobj.k_name, '/');
+	disk->dev.parent = disk->driverfs_dev;
+	disk->dev.devt = MKDEV(disk->major, disk->first_minor);
+
+	strlcpy(disk->dev.bus_id, disk->disk_name, KOBJ_NAME_LEN);
+	/* ewww... some of these buggers have / in the name... */
+	s = strchr(disk->dev.bus_id, '/');
 	if (s)
 		*s = '!';
-	if ((err = kobject_add(&disk->kobj)))
+
+	/* delay uevents, until we scanned partition table */
+	disk->dev.uevent_suppress = 1;
+
+	if (device_add(&disk->dev))
 		return;
-	err = disk_sysfs_symlinks(disk);
+#ifndef CONFIG_SYSFS_DEPRECATED
+	err = sysfs_create_link(block_depr, &disk->dev.kobj,
+				kobject_name(&disk->dev.kobj));
 	if (err) {
-		kobject_del(&disk->kobj);
+		device_del(&disk->dev);
 		return;
 	}
- 	disk_sysfs_add_subdirs(disk);
+#endif
+	disk_sysfs_add_subdirs(disk);
 
 	/* No minors to use for partitions */
 	if (disk->minors == 1)
@@ -505,25 +410,23 @@ void register_disk(struct gendisk *disk)
 	if (!bdev)
 		goto exit;
 
-	/* scan partition table, but suppress uevents */
 	bdev->bd_invalidated = 1;
-	disk->part_uevent_suppress = 1;
 	err = blkdev_get(bdev, FMODE_READ, 0);
-	disk->part_uevent_suppress = 0;
 	if (err < 0)
 		goto exit;
 	blkdev_put(bdev);
 
 exit:
-	/* announce disk after possible partitions are already created */
-	kobject_uevent(&disk->kobj, KOBJ_ADD);
+	/* announce disk after possible partitions are created */
+	disk->dev.uevent_suppress = 0;
+	kobject_uevent(&disk->dev.kobj, KOBJ_ADD);
 
 	/* announce possible partitions */
 	for (i = 1; i < disk->minors; i++) {
 		p = disk->part[i-1];
 		if (!p || !p->nr_sects)
 			continue;
-		kobject_uevent(&p->kobj, KOBJ_ADD);
+		kobject_uevent(&p->dev.kobj, KOBJ_ADD);
 	}
 }
 
@@ -602,19 +505,11 @@ void del_gendisk(struct gendisk *disk)
 	disk_stat_set_all(disk, 0);
 	disk->stamp = 0;
 
-	kobject_uevent(&disk->kobj, KOBJ_REMOVE);
 	kobject_unregister(disk->holder_dir);
 	kobject_unregister(disk->slave_dir);
-	if (disk->driverfs_dev) {
-		char *disk_name = make_block_name(disk);
-		sysfs_remove_link(&disk->kobj, "device");
-		if (disk_name) {
-			sysfs_remove_link(&disk->driverfs_dev->kobj, disk_name);
-			kfree(disk_name);
-		}
-		put_device(disk->driverfs_dev);
-		disk->driverfs_dev = NULL;
-	}
-	sysfs_remove_link(&disk->kobj, "subsystem");
-	kobject_del(&disk->kobj);
+	disk->driverfs_dev = NULL;
+#ifndef CONFIG_SYSFS_DEPRECATED
+	sysfs_remove_link(block_depr, disk->dev.bus_id);
+#endif
+	device_del(&disk->dev);
 }
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index a47b8025d399..1dbea0ac5693 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -10,9 +10,19 @@
  */
 
 #include <linux/types.h>
+#include <linux/kdev_t.h>
 
 #ifdef CONFIG_BLOCK
 
+#define kobj_to_dev(k) container_of(k, struct device, kobj)
+#define dev_to_disk(device) container_of(device, struct gendisk, dev)
+#define dev_to_part(device) container_of(device, struct hd_struct, dev)
+
+extern struct device_type disk_type;
+extern struct device_type part_type;
+extern struct kobject *block_depr;
+extern struct class block_class;
+
 enum {
 /* These three have identical behaviour; use the second one if DOS FDISK gets
    confused about extended/logical partitions starting past cylinder 1023. */
@@ -84,7 +94,7 @@ struct partition {
 struct hd_struct {
 	sector_t start_sect;
 	sector_t nr_sects;
-	struct kobject kobj;
+	struct device dev;
 	struct kobject *holder_dir;
 	unsigned ios[2], sectors[2];	/* READs and WRITEs */
 	int policy, partno;
@@ -117,15 +127,14 @@ struct gendisk {
                                          * disks that can't be partitioned. */
 	char disk_name[32];		/* name of major driver */
 	struct hd_struct **part;	/* [indexed by minor] */
-	int part_uevent_suppress;
 	struct block_device_operations *fops;
 	struct request_queue *queue;
 	void *private_data;
 	sector_t capacity;
 
 	int flags;
-	struct device *driverfs_dev;
-	struct kobject kobj;
+	struct device *driverfs_dev;  // FIXME: remove
+	struct device dev;
 	struct kobject *holder_dir;
 	struct kobject *slave_dir;
 
@@ -143,13 +152,6 @@ struct gendisk {
 	struct work_struct async_notify;
 };
 
-/* Structure for sysfs attributes on block devices */
-struct disk_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct gendisk *, char *);
-	ssize_t (*store)(struct gendisk *, const char *, size_t);
-};
-
 /* 
  * Macros to operate on percpu disk statistics:
  *
@@ -411,7 +413,8 @@ struct unixware_disklabel {
 #define ADDPART_FLAG_RAID	1
 #define ADDPART_FLAG_WHOLEDISK	2
 
-char *disk_name (struct gendisk *hd, int part, char *buf);
+extern dev_t blk_lookup_devt(const char *name);
+extern char *disk_name (struct gendisk *hd, int part, char *buf);
 
 extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev);
 extern void add_partition(struct gendisk *, int, sector_t, sector_t, int);
@@ -423,12 +426,12 @@ extern struct gendisk *alloc_disk(int minors);
 extern struct kobject *get_disk(struct gendisk *disk);
 extern void put_disk(struct gendisk *disk);
 extern void genhd_media_change_notify(struct gendisk *disk);
-extern void blk_register_region(dev_t dev, unsigned long range,
+extern void blk_register_region(dev_t devt, unsigned long range,
 			struct module *module,
 			struct kobject *(*probe)(dev_t, int *, void *),
 			int (*lock)(dev_t, void *),
 			void *data);
-extern void blk_unregister_region(dev_t dev, unsigned long range);
+extern void blk_unregister_region(dev_t devt, unsigned long range);
 
 static inline struct block_device *bdget_disk(struct gendisk *disk, int index)
 {
@@ -441,6 +444,12 @@ static inline struct block_device *bdget_disk(struct gendisk *disk, int index)
 
 static inline void printk_all_partitions(void) { }
 
+static inline dev_t blk_lookup_devt(const char *name)
+{
+	dev_t devt = MKDEV(0, 0);
+	return devt;
+}
+
 #endif /* CONFIG_BLOCK */
 
 #endif
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 4efa1e5385e3..2ae5b8462399 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -55,69 +55,6 @@ static int __init readwrite(char *str)
 __setup("ro", readonly);
 __setup("rw", readwrite);
 
-static dev_t try_name(char *name, int part)
-{
-	char path[64];
-	char buf[32];
-	int range;
-	dev_t res;
-	char *s;
-	int len;
-	int fd;
-	unsigned int maj, min;
-
-	/* read device number from .../dev */
-
-	sprintf(path, "/sys/block/%s/dev", name);
-	fd = sys_open(path, 0, 0);
-	if (fd < 0)
-		goto fail;
-	len = sys_read(fd, buf, 32);
-	sys_close(fd);
-	if (len <= 0 || len == 32 || buf[len - 1] != '\n')
-		goto fail;
-	buf[len - 1] = '\0';
-	if (sscanf(buf, "%u:%u", &maj, &min) == 2) {
-		/*
-		 * Try the %u:%u format -- see print_dev_t()
-		 */
-		res = MKDEV(maj, min);
-		if (maj != MAJOR(res) || min != MINOR(res))
-			goto fail;
-	} else {
-		/*
-		 * Nope.  Try old-style "0321"
-		 */
-		res = new_decode_dev(simple_strtoul(buf, &s, 16));
-		if (*s)
-			goto fail;
-	}
-
-	/* if it's there and we are not looking for a partition - that's it */
-	if (!part)
-		return res;
-
-	/* otherwise read range from .../range */
-	sprintf(path, "/sys/block/%s/range", name);
-	fd = sys_open(path, 0, 0);
-	if (fd < 0)
-		goto fail;
-	len = sys_read(fd, buf, 32);
-	sys_close(fd);
-	if (len <= 0 || len == 32 || buf[len - 1] != '\n')
-		goto fail;
-	buf[len - 1] = '\0';
-	range = simple_strtoul(buf, &s, 10);
-	if (*s)
-		goto fail;
-
-	/* if partition is within range - we got it */
-	if (part < range)
-		return res + part;
-fail:
-	return 0;
-}
-
 /*
  *	Convert a name into device number.  We accept the following variants:
  *
@@ -129,12 +66,10 @@ fail:
  *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
  *	   used when disk name of partitioned disk ends on a digit.
  *
- *	If name doesn't have fall into the categories above, we return 0.
- *	Sysfs is used to check if something is a disk name - it has
- *	all known disks under bus/block/devices.  If the disk name
- *	contains slashes, name of sysfs node has them replaced with
- *	bangs.  try_name() does the actual checks, assuming that sysfs
- *	is mounted on rootfs /sys.
+ *	If name doesn't have fall into the categories above, we return (0,0).
+ *	block_class is used to check if something is a disk name. If the disk
+ *	name contains slashes, the device name has them replaced with
+ *	bangs.
  */
 
 dev_t name_to_dev_t(char *name)
@@ -142,13 +77,6 @@ dev_t name_to_dev_t(char *name)
 	char s[32];
 	char *p;
 	dev_t res = 0;
-	int part;
-
-#ifdef CONFIG_SYSFS
-	int mkdir_err = sys_mkdir("/sys", 0700);
-	if (sys_mount("sysfs", "/sys", "sysfs", 0, NULL) < 0)
-		goto out;
-#endif
 
 	if (strncmp(name, "/dev/", 5) != 0) {
 		unsigned maj, min;
@@ -164,6 +92,7 @@ dev_t name_to_dev_t(char *name)
 		}
 		goto done;
 	}
+
 	name += 5;
 	res = Root_NFS;
 	if (strcmp(name, "nfs") == 0)
@@ -178,35 +107,14 @@ dev_t name_to_dev_t(char *name)
 	for (p = s; *p; p++)
 		if (*p == '/')
 			*p = '!';
-	res = try_name(s, 0);
+	res = blk_lookup_devt(s);
 	if (res)
 		goto done;
 
-	while (p > s && isdigit(p[-1]))
-		p--;
-	if (p == s || !*p || *p == '0')
-		goto fail;
-	part = simple_strtoul(p, NULL, 10);
-	*p = '\0';
-	res = try_name(s, part);
-	if (res)
-		goto done;
-
-	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		goto fail;
-	p[-1] = '\0';
-	res = try_name(s, part);
+fail:
+	return 0;
 done:
-#ifdef CONFIG_SYSFS
-	sys_umount("/sys", 0);
-out:
-	if (!mkdir_err)
-		sys_rmdir("/sys");
-#endif
 	return res;
-fail:
-	res = 0;
-	goto done;
 }
 
 static int __init root_dev_setup(char *line)
-- 
cgit v1.2.3


From 9e7bbccd0290e720e0874443932869c55f63d5a8 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 17 Dec 2007 23:05:35 -0700
Subject: Kobject: remove kobject_add() as no one uses it anymore

The old kobject_add() function is on longer in use, so let us remove it
from the public scope (kset mess in the kobject.c file still uses it,
but that can be cleaned up later very simply.)

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  1 -
 lib/kobject.c           | 22 ++++++++--------------
 2 files changed, 8 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index bb6868475edb..8b0aa715fa2f 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -80,7 +80,6 @@ static inline const char * kobject_name(const struct kobject * kobj)
 
 extern void kobject_init(struct kobject *);
 extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
-extern int __must_check kobject_add(struct kobject *);
 extern int __must_check kobject_add_ng(struct kobject *kobj,
 				       struct kobject *parent,
 				       const char *fmt, ...);
diff --git a/lib/kobject.c b/lib/kobject.c
index 493e991abb1b..d04789fa4da9 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -144,7 +144,7 @@ void kobject_init(struct kobject * kobj)
  *	Remove the kobject from the kset list and decrement
  *	its parent's refcount.
  *	This is separated out, so we can use it in both 
- *	kobject_del() and kobject_add() on error.
+ *	kobject_del() and kobject_add_internal() on error.
  */
 
 static void unlink(struct kobject * kobj)
@@ -161,12 +161,7 @@ static void unlink(struct kobject * kobj)
 	kobject_put(parent);
 }
 
-/**
- *	kobject_add - add an object to the hierarchy.
- *	@kobj:	object.
- */
-
-int kobject_add(struct kobject * kobj)
+static int kobject_add_internal(struct kobject *kobj)
 {
 	int error = 0;
 	struct kobject * parent;
@@ -215,13 +210,13 @@ int kobject_add(struct kobject * kobj)
 
 		/* be noisy on error issues */
 		if (error == -EEXIST)
-			printk(KERN_ERR "kobject_add failed for %s with "
+			printk(KERN_ERR "%s failed for %s with "
 			       "-EEXIST, don't try to register things with "
 			       "the same name in the same directory.\n",
-			       kobject_name(kobj));
+			       __FUNCTION__, kobject_name(kobj));
 		else
-			printk(KERN_ERR "kobject_add failed for %s (%d)\n",
-			       kobject_name(kobj), error);
+			printk(KERN_ERR "%s failed for %s (%d)\n",
+			       __FUNCTION__, kobject_name(kobj), error);
 		dump_stack();
 	}
 
@@ -351,7 +346,7 @@ static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
 		return retval;
 	}
 	kobj->parent = parent;
-	return kobject_add(kobj);
+	return kobject_add_internal(kobj);
 }
 
 /**
@@ -742,7 +737,7 @@ struct sysfs_ops kobj_sysfs_ops = {
 
 int kset_add(struct kset * k)
 {
-	return kobject_add(&k->kobj);
+	return kobject_add_internal(&k->kobj);
 }
 
 
@@ -897,7 +892,6 @@ EXPORT_SYMBOL(kobject_register);
 EXPORT_SYMBOL(kobject_unregister);
 EXPORT_SYMBOL(kobject_get);
 EXPORT_SYMBOL(kobject_put);
-EXPORT_SYMBOL(kobject_add);
 EXPORT_SYMBOL(kobject_del);
 
 EXPORT_SYMBOL(kset_register);
-- 
cgit v1.2.3


From b2d6db5878a0832659ed58476357eea2db915550 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 17 Dec 2007 23:05:35 -0700
Subject: Kobject: rename kobject_add_ng() to kobject_add()

Now that the old kobject_add() function is gone, rename kobject_add_ng()
to kobject_add() to clean up the namespace.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 block/elevator.c           |  2 +-
 block/ll_rw_blk.c          |  4 ++--
 drivers/base/class.c       |  4 ++--
 drivers/base/core.c        |  6 +++---
 drivers/base/driver.c      |  2 +-
 drivers/md/md.c            |  2 +-
 drivers/net/iseries_veth.c |  2 +-
 drivers/uio/uio.c          |  2 +-
 include/linux/kobject.h    |  6 +++---
 lib/kobject.c              | 14 +++++++-------
 10 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/block/elevator.c b/block/elevator.c
index 5445c3c2ee8a..645469a4f49f 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -929,7 +929,7 @@ int elv_register_queue(struct request_queue *q)
 	elevator_t *e = q->elevator;
 	int error;
 
-	error = kobject_add_ng(&e->kobj, &q->kobj, "%s", "iosched");
+	error = kobject_add(&e->kobj, &q->kobj, "%s", "iosched");
 	if (!error) {
 		struct elv_fs_entry *attr = e->elevator_type->elevator_attrs;
 		if (attr) {
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 8054b7d8e07c..234dd3de1824 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -4180,8 +4180,8 @@ int blk_register_queue(struct gendisk *disk)
 	if (!q || !q->request_fn)
 		return -ENXIO;
 
-	ret = kobject_add_ng(&q->kobj, kobject_get(&disk->dev.kobj),
-			     "%s", "queue");
+	ret = kobject_add(&q->kobj, kobject_get(&disk->dev.kobj),
+			  "%s", "queue");
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 624b3316e938..8e3cba224384 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -586,8 +586,8 @@ int class_device_add(struct class_device *class_dev)
 	else
 		class_dev->kobj.parent = &parent_class->subsys.kobj;
 
-	error = kobject_add_ng(&class_dev->kobj, class_dev->kobj.parent,
-			       "%s", class_dev->class_id);
+	error = kobject_add(&class_dev->kobj, class_dev->kobj.parent,
+			    "%s", class_dev->class_id);
 	if (error)
 		goto out2;
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 06e8738ab263..e88170293ca0 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -602,7 +602,7 @@ static struct kobject *get_device_parent(struct device *dev,
 		if (!k)
 			return NULL;
 		k->kset = &dev->class->class_dirs;
-		retval = kobject_add_ng(k, parent_kobj, "%s", dev->class->name);
+		retval = kobject_add(k, parent_kobj, "%s", dev->class->name);
 		if (retval < 0) {
 			kobject_put(k);
 			return NULL;
@@ -776,7 +776,7 @@ static void device_remove_class_symlinks(struct device *dev)
  *	This is part 2 of device_register(), though may be called
  *	separately _iff_ device_initialize() has been called separately.
  *
- *	This adds it to the kobject hierarchy via kobject_add_ng(), adds it
+ *	This adds it to the kobject hierarchy via kobject_add(), adds it
  *	to the global and sibling lists for the device, then
  *	adds it to the other relevant subsystems of the driver model.
  */
@@ -807,7 +807,7 @@ int device_add(struct device *dev)
 		goto Error;
 
 	/* first, register with generic layer. */
-	error = kobject_add_ng(&dev->kobj, dev->kobj.parent, "%s", dev->bus_id);
+	error = kobject_add(&dev->kobj, dev->kobj.parent, "%s", dev->bus_id);
 	if (error)
 		goto Error;
 
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 5aacff208f21..94b697a9b4e0 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -144,7 +144,7 @@ int driver_add_kobj(struct device_driver *drv, struct kobject *kobj,
 	if (!name)
 		return -ENOMEM;
 
-	return kobject_add_ng(kobj, &drv->p->kobj, "%s", name);
+	return kobject_add(kobj, &drv->p->kobj, "%s", name);
 }
 EXPORT_SYMBOL_GPL(driver_add_kobj);
 
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 7ae9740c483d..989d8549f988 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1389,7 +1389,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 	rdev->mddev = mddev;
 	printk(KERN_INFO "md: bind<%s>\n", b);
 
-	if ((err = kobject_add_ng(&rdev->kobj, &mddev->kobj, "dev-%s", b)))
+	if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b)))
 		goto fail;
 
 	if (rdev->bdev->bd_part)
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index 1a8299acd3f1..ee15667d6135 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -1084,7 +1084,7 @@ static struct net_device * __init veth_probe_one(int vlan,
 	}
 
 	kobject_init_ng(&port->kobject, &veth_port_ktype);
-	if (0 != kobject_add_ng(&port->kobject, &dev->dev.kobj, "veth_port"))
+	if (0 != kobject_add(&port->kobject, &dev->dev.kobj, "veth_port"))
 		veth_error("Failed adding port for %s to sysfs.\n", dev->name);
 
 	veth_info("%s attached to iSeries vlan %d (LPAR map = 0x%.4X)\n",
diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index acc387de988f..1ec2d31f2639 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -172,7 +172,7 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 		kobject_init_ng(&map->kobj, &map_attr_type);
 		map->mem = mem;
 		mem->map = map;
-		ret = kobject_add_ng(&map->kobj, idev->map_dir, "map%d", mi);
+		ret = kobject_add(&map->kobj, idev->map_dir, "map%d", mi);
 		if (ret)
 			goto err;
 		ret = kobject_uevent(&map->kobj, KOBJ_ADD);
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 8b0aa715fa2f..84c5afd58899 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -80,9 +80,9 @@ static inline const char * kobject_name(const struct kobject * kobj)
 
 extern void kobject_init(struct kobject *);
 extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
-extern int __must_check kobject_add_ng(struct kobject *kobj,
-				       struct kobject *parent,
-				       const char *fmt, ...);
+extern int __must_check kobject_add(struct kobject *kobj,
+				    struct kobject *parent,
+				    const char *fmt, ...);
 extern int __must_check kobject_init_and_add(struct kobject *kobj,
 					     struct kobj_type *ktype,
 					     struct kobject *parent,
diff --git a/lib/kobject.c b/lib/kobject.c
index d04789fa4da9..359e114790cb 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -350,7 +350,7 @@ static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
 }
 
 /**
- * kobject_add_ng - the main kobject add function
+ * kobject_add - the main kobject add function
  * @kobj: the kobject to add
  * @parent: pointer to the parent of the kobject.
  * @fmt: format to name the kobject with.
@@ -381,8 +381,8 @@ static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
  * kobject_uevent() with the UEVENT_ADD parameter to ensure that
  * userspace is properly notified of this kobject's creation.
  */
-int kobject_add_ng(struct kobject *kobj, struct kobject *parent,
-		   const char *fmt, ...)
+int kobject_add(struct kobject *kobj, struct kobject *parent,
+		const char *fmt, ...)
 {
 	va_list args;
 	int retval;
@@ -396,7 +396,7 @@ int kobject_add_ng(struct kobject *kobj, struct kobject *parent,
 
 	return retval;
 }
-EXPORT_SYMBOL(kobject_add_ng);
+EXPORT_SYMBOL(kobject_add);
 
 /**
  * kobject_init_and_add - initialize a kobject structure and add it to the kobject hierarchy
@@ -406,8 +406,8 @@ EXPORT_SYMBOL(kobject_add_ng);
  * @fmt: the name of the kobject.
  *
  * This function combines the call to kobject_init_ng() and
- * kobject_add_ng().  The same type of error handling after a call to
- * kobject_add_ng() and kobject lifetime rules are the same here.
+ * kobject_add().  The same type of error handling after a call to
+ * kobject_add() and kobject lifetime rules are the same here.
  */
 int kobject_init_and_add(struct kobject *kobj, struct kobj_type *ktype,
 			 struct kobject *parent, const char *fmt, ...)
@@ -677,7 +677,7 @@ struct kobject *kobject_create_and_add(const char *name, struct kobject *parent)
 	if (!kobj)
 		return NULL;
 
-	retval = kobject_add_ng(kobj, parent, "%s", name);
+	retval = kobject_add(kobj, parent, "%s", name);
 	if (retval) {
 		printk(KERN_WARNING "%s: kobject_add error: %d\n",
 		       __FUNCTION__, retval);
-- 
cgit v1.2.3


From e1543ddf739b22a8c4218716ad50c26b3e147403 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 17 Dec 2007 23:05:35 -0700
Subject: Kobject: remove kobject_init() as no one uses it anymore

The old kobject_init() function is on longer in use, so let us remove it
from the public scope (kset mess in the kobject.c file still uses it,
but that can be cleaned up later very simply.)

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  1 -
 lib/kobject.c           | 11 +++--------
 2 files changed, 3 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 84c5afd58899..53458b674fae 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -78,7 +78,6 @@ static inline const char * kobject_name(const struct kobject * kobj)
 	return kobj->k_name;
 }
 
-extern void kobject_init(struct kobject *);
 extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
 extern int __must_check kobject_add(struct kobject *kobj,
 				    struct kobject *parent,
diff --git a/lib/kobject.c b/lib/kobject.c
index 359e114790cb..10d977b6e69d 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -124,11 +124,7 @@ char *kobject_get_path(struct kobject *kobj, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL_GPL(kobject_get_path);
 
-/**
- *	kobject_init - initialize object.
- *	@kobj:	object in question.
- */
-void kobject_init(struct kobject * kobj)
+static void kobject_init_internal(struct kobject * kobj)
 {
 	if (!kobj)
 		return;
@@ -232,7 +228,7 @@ int kobject_register(struct kobject * kobj)
 {
 	int error = -EINVAL;
 	if (kobj) {
-		kobject_init(kobj);
+		kobject_init_internal(kobj);
 		error = kobject_add(kobj);
 		if (!error)
 			kobject_uevent(kobj, KOBJ_ADD);
@@ -695,7 +691,7 @@ EXPORT_SYMBOL_GPL(kobject_create_and_add);
 
 void kset_init(struct kset * k)
 {
-	kobject_init(&k->kobj);
+	kobject_init_internal(&k->kobj);
 	INIT_LIST_HEAD(&k->list);
 	spin_lock_init(&k->list_lock);
 }
@@ -887,7 +883,6 @@ struct kset *kset_create_and_add(const char *name,
 }
 EXPORT_SYMBOL_GPL(kset_create_and_add);
 
-EXPORT_SYMBOL(kobject_init);
 EXPORT_SYMBOL(kobject_register);
 EXPORT_SYMBOL(kobject_unregister);
 EXPORT_SYMBOL(kobject_get);
-- 
cgit v1.2.3


From f9cb074bff8e762ef24c44678a5a7d907f82fbeb Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 17 Dec 2007 23:05:35 -0700
Subject: Kobject: rename kobject_init_ng() to kobject_init()

Now that the old kobject_init() function is gone, rename
kobject_init_ng() to kobject_init() to clean up the namespace.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 block/elevator.c           |  2 +-
 block/ll_rw_blk.c          |  2 +-
 drivers/base/class.c       |  2 +-
 drivers/base/core.c        |  2 +-
 drivers/md/md.c            |  2 +-
 drivers/net/iseries_veth.c |  4 ++--
 drivers/uio/uio.c          |  2 +-
 fs/char_dev.c              |  4 ++--
 include/linux/kobject.h    |  2 +-
 lib/kobject.c              | 14 +++++++-------
 10 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/block/elevator.c b/block/elevator.c
index 645469a4f49f..f9736fbdab03 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -185,7 +185,7 @@ static elevator_t *elevator_alloc(struct request_queue *q,
 
 	eq->ops = &e->ops;
 	eq->elevator_type = e;
-	kobject_init_ng(&eq->kobj, &elv_ktype);
+	kobject_init(&eq->kobj, &elv_ktype);
 	mutex_init(&eq->sysfs_lock);
 
 	eq->hash = kmalloc_node(sizeof(struct hlist_head) * ELV_HASH_ENTRIES,
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 234dd3de1824..5ccec8aa964b 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -1862,7 +1862,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 
 	init_timer(&q->unplug_timer);
 
-	kobject_init_ng(&q->kobj, &queue_ktype);
+	kobject_init(&q->kobj, &queue_ktype);
 
 	mutex_init(&q->sysfs_lock);
 
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 8e3cba224384..61fd26cc9f0e 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -553,7 +553,7 @@ static struct class_device_attribute class_uevent_attr =
 void class_device_initialize(struct class_device *class_dev)
 {
 	class_dev->kobj.kset = &class_obj_subsys;
-	kobject_init_ng(&class_dev->kobj, &class_device_ktype);
+	kobject_init(&class_dev->kobj, &class_device_ktype);
 	INIT_LIST_HEAD(&class_dev->node);
 }
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index e88170293ca0..675a719dcdd2 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -525,7 +525,7 @@ static void klist_children_put(struct klist_node *n)
 void device_initialize(struct device *dev)
 {
 	dev->kobj.kset = devices_kset;
-	kobject_init_ng(&dev->kobj, &device_ktype);
+	kobject_init(&dev->kobj, &device_ktype);
 	klist_init(&dev->klist_children, klist_children_get,
 		   klist_children_put);
 	INIT_LIST_HEAD(&dev->dma_pools);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 989d8549f988..ae800ba061a6 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2033,7 +2033,7 @@ static mdk_rdev_t *md_import_device(dev_t newdev, int super_format, int super_mi
 	if (err)
 		goto abort_free;
 
-	kobject_init_ng(&rdev->kobj, &rdev_ktype);
+	kobject_init(&rdev->kobj, &rdev_ktype);
 
 	rdev->desc_nr = -1;
 	rdev->saved_raid_disk = -1;
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index ee15667d6135..419861cbc65e 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -844,7 +844,7 @@ static int veth_init_connection(u8 rlp)
 
 	/* This gets us 1 reference, which is held on behalf of the driver
 	 * infrastructure. It's released at module unload. */
-	kobject_init_ng(&cnx->kobject, &veth_lpar_connection_ktype);
+	kobject_init(&cnx->kobject, &veth_lpar_connection_ktype);
 
 	msgs = kcalloc(VETH_NUMBUFFERS, sizeof(struct veth_msg), GFP_KERNEL);
 	if (! msgs) {
@@ -1083,7 +1083,7 @@ static struct net_device * __init veth_probe_one(int vlan,
 		return NULL;
 	}
 
-	kobject_init_ng(&port->kobject, &veth_port_ktype);
+	kobject_init(&port->kobject, &veth_port_ktype);
 	if (0 != kobject_add(&port->kobject, &dev->dev.kobj, "veth_port"))
 		veth_error("Failed adding port for %s to sysfs.\n", dev->name);
 
diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 1ec2d31f2639..f352731add64 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -169,7 +169,7 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 		map = kzalloc(sizeof(*map), GFP_KERNEL);
 		if (!map)
 			goto err;
-		kobject_init_ng(&map->kobj, &map_attr_type);
+		kobject_init(&map->kobj, &map_attr_type);
 		map->mem = mem;
 		mem->map = map;
 		ret = kobject_add(&map->kobj, idev->map_dir, "map%d", mi);
diff --git a/fs/char_dev.c b/fs/char_dev.c
index b2dd5a036631..2c7a8b5b4598 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -511,7 +511,7 @@ struct cdev *cdev_alloc(void)
 	struct cdev *p = kzalloc(sizeof(struct cdev), GFP_KERNEL);
 	if (p) {
 		INIT_LIST_HEAD(&p->list);
-		kobject_init_ng(&p->kobj, &ktype_cdev_dynamic);
+		kobject_init(&p->kobj, &ktype_cdev_dynamic);
 	}
 	return p;
 }
@@ -528,7 +528,7 @@ void cdev_init(struct cdev *cdev, const struct file_operations *fops)
 {
 	memset(cdev, 0, sizeof *cdev);
 	INIT_LIST_HEAD(&cdev->list);
-	kobject_init_ng(&cdev->kobj, &ktype_cdev_default);
+	kobject_init(&cdev->kobj, &ktype_cdev_default);
 	cdev->ops = fops;
 }
 
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 53458b674fae..d9d8c368f044 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -78,7 +78,7 @@ static inline const char * kobject_name(const struct kobject * kobj)
 	return kobj->k_name;
 }
 
-extern void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype);
+extern void kobject_init(struct kobject *kobj, struct kobj_type *ktype);
 extern int __must_check kobject_add(struct kobject *kobj,
 				    struct kobject *parent,
 				    const char *fmt, ...);
diff --git a/lib/kobject.c b/lib/kobject.c
index 10d977b6e69d..4cc231c86225 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -287,7 +287,7 @@ int kobject_set_name(struct kobject *kobj, const char *fmt, ...)
 EXPORT_SYMBOL(kobject_set_name);
 
 /**
- * kobject_init_ng - initialize a kobject structure
+ * kobject_init - initialize a kobject structure
  * @kobj: pointer to the kobject to initialize
  * @ktype: pointer to the ktype for this kobject.
  *
@@ -298,7 +298,7 @@ EXPORT_SYMBOL(kobject_set_name);
  * to kobject_put(), not by a call to kfree directly to ensure that all of
  * the memory is cleaned up properly.
  */
-void kobject_init_ng(struct kobject *kobj, struct kobj_type *ktype)
+void kobject_init(struct kobject *kobj, struct kobj_type *ktype)
 {
 	char *err_str;
 
@@ -326,7 +326,7 @@ error:
 	printk(KERN_ERR "kobject: %s\n", err_str);
 	dump_stack();
 }
-EXPORT_SYMBOL(kobject_init_ng);
+EXPORT_SYMBOL(kobject_init);
 
 static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
 			    const char *fmt, va_list vargs)
@@ -401,7 +401,7 @@ EXPORT_SYMBOL(kobject_add);
  * @parent: pointer to the parent of this kobject.
  * @fmt: the name of the kobject.
  *
- * This function combines the call to kobject_init_ng() and
+ * This function combines the call to kobject_init() and
  * kobject_add().  The same type of error handling after a call to
  * kobject_add() and kobject lifetime rules are the same here.
  */
@@ -411,7 +411,7 @@ int kobject_init_and_add(struct kobject *kobj, struct kobj_type *ktype,
 	va_list args;
 	int retval;
 
-	kobject_init_ng(kobj, ktype);
+	kobject_init(kobj, ktype);
 
 	va_start(args, fmt);
 	retval = kobject_add_varg(kobj, parent, fmt, args);
@@ -636,7 +636,7 @@ static struct kobj_type dynamic_kobj_ktype = {
  *
  * If the kobject was not able to be created, NULL will be returned.
  * The kobject structure returned from here must be cleaned up with a
- * call to kobject_put() and not kfree(), as kobject_init_ng() has
+ * call to kobject_put() and not kfree(), as kobject_init() has
  * already been called on this structure.
  */
 struct kobject *kobject_create(void)
@@ -647,7 +647,7 @@ struct kobject *kobject_create(void)
 	if (!kobj)
 		return NULL;
 
-	kobject_init_ng(kobj, &dynamic_kobj_ktype);
+	kobject_init(kobj, &dynamic_kobj_ktype);
 	return kobj;
 }
 
-- 
cgit v1.2.3


From 6d06adfaf82d154023141ddc0c9de18b6a49090b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 19 Dec 2007 11:26:50 -0800
Subject: Kobject: remove kobject_register()

The function is no longer used by anyone in the kernel, and it prevents
the proper sending of the kobject uevent after the needed files are set
up by the caller.  kobject_init_and_add() can be used in its place.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  1 -
 lib/kobject.c           | 18 ------------------
 2 files changed, 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index d9d8c368f044..259084758617 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -96,7 +96,6 @@ extern struct kobject * __must_check kobject_create_and_add(const char *name,
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 extern int __must_check kobject_move(struct kobject *, struct kobject *);
 
-extern int __must_check kobject_register(struct kobject *);
 extern void kobject_unregister(struct kobject *);
 
 extern struct kobject * kobject_get(struct kobject *);
diff --git a/lib/kobject.c b/lib/kobject.c
index 4cc231c86225..3326281c96b6 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -219,23 +219,6 @@ static int kobject_add_internal(struct kobject *kobj)
 	return error;
 }
 
-/**
- *	kobject_register - initialize and add an object.
- *	@kobj:	object in question.
- */
-
-int kobject_register(struct kobject * kobj)
-{
-	int error = -EINVAL;
-	if (kobj) {
-		kobject_init_internal(kobj);
-		error = kobject_add(kobj);
-		if (!error)
-			kobject_uevent(kobj, KOBJ_ADD);
-	}
-	return error;
-}
-
 /**
  * kobject_set_name_vargs - Set the name of an kobject
  * @kobj: struct kobject to set the name of
@@ -883,7 +866,6 @@ struct kset *kset_create_and_add(const char *name,
 }
 EXPORT_SYMBOL_GPL(kset_create_and_add);
 
-EXPORT_SYMBOL(kobject_register);
 EXPORT_SYMBOL(kobject_unregister);
 EXPORT_SYMBOL(kobject_get);
 EXPORT_SYMBOL(kobject_put);
-- 
cgit v1.2.3


From 12e339ac6e31a34fe42396aec8fb1c0b43caf61e Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 9 Apr 2002 12:14:34 -0700
Subject: Kset: remove kset_add function

No one is calling this anymore, so just remove it and hard-code the one
internal-use of it.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  1 -
 lib/kobject.c           | 13 +------------
 2 files changed, 1 insertion(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 259084758617..63967da073af 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -158,7 +158,6 @@ struct kset {
 };
 
 extern void kset_init(struct kset * k);
-extern int __must_check kset_add(struct kset * k);
 extern int __must_check kset_register(struct kset * k);
 extern void kset_unregister(struct kset * k);
 extern struct kset * __must_check kset_create_and_add(const char *name,
diff --git a/lib/kobject.c b/lib/kobject.c
index 3326281c96b6..c321f1910cc2 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -709,17 +709,6 @@ struct sysfs_ops kobj_sysfs_ops = {
 	.store	= kobj_attr_store,
 };
 
-/**
- *	kset_add - add a kset object to the hierarchy.
- *	@k:	kset.
- */
-
-int kset_add(struct kset * k)
-{
-	return kobject_add_internal(&k->kobj);
-}
-
-
 /**
  *	kset_register - initialize and add a kset.
  *	@k:	kset.
@@ -733,7 +722,7 @@ int kset_register(struct kset * k)
 		return -EINVAL;
 
 	kset_init(k);
-	err = kset_add(k);
+	err = kobject_add_internal(&k->kobj);
 	if (err)
 		return err;
 	kobject_uevent(&k->kobj, KOBJ_ADD);
-- 
cgit v1.2.3


From 0f4dafc0563c6c49e17fe14b3f5f356e4c4b8806 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Wed, 19 Dec 2007 01:40:42 +0100
Subject: Kobject: auto-cleanup on final unref

We save the current state in the object itself, so we can do proper
cleanup when the last reference is dropped.

If the initial reference is dropped, the object will be removed from
sysfs if needed, if an "add" event was sent, "remove" will be send, and
the allocated resources are released.

This allows us to clean up some driver core usage as well as allowing us
to do other such changes to the rest of the kernel.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c     |  32 ++-------
 include/linux/kobject.h |   5 ++
 lib/kobject.c           | 170 +++++++++++++++++++++++++++---------------------
 lib/kobject_uevent.c    |  11 ++++
 4 files changed, 119 insertions(+), 99 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 675a719dcdd2..d5d542db96fd 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -576,8 +576,8 @@ static struct kobject *get_device_parent(struct device *dev,
 
 		/*
 		 * If we have no parent, we live in "virtual".
-		 * Class-devices with a bus-device as parent, live
-		 * in a class-directory to prevent namespace collisions.
+		 * Class-devices with a non class-device as parent, live
+		 * in a "glue" directory to prevent namespace collisions.
 		 */
 		if (parent == NULL)
 			parent_kobj = virtual_device_parent(dev);
@@ -607,8 +607,7 @@ static struct kobject *get_device_parent(struct device *dev,
 			kobject_put(k);
 			return NULL;
 		}
-		/* Do not emit a uevent, as it's not needed for this
-		 * "class glue" directory. */
+		/* do not emit an uevent for this simple "glue" directory */
 		return k;
 	}
 
@@ -619,30 +618,13 @@ static struct kobject *get_device_parent(struct device *dev,
 
 static void cleanup_device_parent(struct device *dev)
 {
-	struct device *d;
-	int other = 0;
+	struct kobject *glue_dir = dev->kobj.parent;
 
-	if (!dev->class)
-		return;
-
-	/* see if we live in a parent class directory */
-	if (dev->kobj.parent->kset != &dev->class->class_dirs)
+	/* see if we live in a "glue" directory */
+	if (!dev->class || glue_dir->kset != &dev->class->class_dirs)
 		return;
 
-	/* if we are the last child of our class, delete the directory */
-	down(&dev->class->sem);
-	list_for_each_entry(d, &dev->class->devices, node) {
-		if (d == dev)
-			continue;
-		if (d->kobj.parent == dev->kobj.parent) {
-			other = 1;
-			break;
-		}
-	}
-	if (!other)
-		kobject_del(dev->kobj.parent);
-	kobject_put(dev->kobj.parent);
-	up(&dev->class->sem);
+	kobject_put(glue_dir);
 }
 #endif
 
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 63967da073af..be03ce83f9cc 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -68,6 +68,11 @@ struct kobject {
 	struct kset		* kset;
 	struct kobj_type	* ktype;
 	struct sysfs_dirent	* sd;
+	unsigned int state_initialized:1;
+	unsigned int state_name_set:1;
+	unsigned int state_in_sysfs:1;
+	unsigned int state_add_uevent_sent:1;
+	unsigned int state_remove_uevent_sent:1;
 };
 
 extern int kobject_set_name(struct kobject *, const char *, ...)
diff --git a/lib/kobject.c b/lib/kobject.c
index c321f1910cc2..4fce5ca42c2e 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -124,85 +124,74 @@ char *kobject_get_path(struct kobject *kobj, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL_GPL(kobject_get_path);
 
-static void kobject_init_internal(struct kobject * kobj)
+/* add the kobject to its kset's list */
+static void kobj_kset_join(struct kobject *kobj)
 {
-	if (!kobj)
+	if (!kobj->kset)
 		return;
-	kref_init(&kobj->kref);
-	INIT_LIST_HEAD(&kobj->entry);
+
+	kset_get(kobj->kset);
+	spin_lock(&kobj->kset->list_lock);
+	list_add_tail(&kobj->entry, &kobj->kset->list);
+	spin_unlock(&kobj->kset->list_lock);
 }
 
+/* remove the kobject from its kset's list */
+static void kobj_kset_leave(struct kobject *kobj)
+{
+	if (!kobj->kset)
+		return;
 
-/**
- *	unlink - remove kobject from kset list.
- *	@kobj:	kobject.
- *
- *	Remove the kobject from the kset list and decrement
- *	its parent's refcount.
- *	This is separated out, so we can use it in both 
- *	kobject_del() and kobject_add_internal() on error.
- */
+	spin_lock(&kobj->kset->list_lock);
+	list_del_init(&kobj->entry);
+	spin_unlock(&kobj->kset->list_lock);
+	kset_put(kobj->kset);
+}
 
-static void unlink(struct kobject * kobj)
+static void kobject_init_internal(struct kobject * kobj)
 {
-	struct kobject *parent = kobj->parent;
-
-	if (kobj->kset) {
-		spin_lock(&kobj->kset->list_lock);
-		list_del_init(&kobj->entry);
-		spin_unlock(&kobj->kset->list_lock);
-	}
-	kobj->parent = NULL;
-	kobject_put(kobj);
-	kobject_put(parent);
+	if (!kobj)
+		return;
+	kref_init(&kobj->kref);
+	INIT_LIST_HEAD(&kobj->entry);
 }
 
+
 static int kobject_add_internal(struct kobject *kobj)
 {
 	int error = 0;
 	struct kobject * parent;
 
-	if (!(kobj = kobject_get(kobj)))
+	if (!kobj)
 		return -ENOENT;
-	if (!kobj->k_name)
-		kobject_set_name(kobj, "NO_NAME");
-	if (!*kobj->k_name) {
-		pr_debug("kobject (%p) attempted to be registered with no "
+
+	if (!kobj->k_name || !kobj->k_name[0]) {
+		pr_debug("kobject: (%p): attempted to be registered with empty "
 			 "name!\n", kobj);
 		WARN_ON(1);
-		kobject_put(kobj);
 		return -EINVAL;
 	}
-	parent = kobject_get(kobj->parent);
 
-	pr_debug("kobject: '%s' (%p): %s: parent: '%s', set: '%s'\n",
-		 kobject_name(kobj), kobj, __FUNCTION__,
-		 parent ? kobject_name(parent) : "<NULL>",
-		 kobj->kset ? kobject_name(&kobj->kset->kobj) : "<NULL>" );
+	parent = kobject_get(kobj->parent);
 
+	/* join kset if set, use it as parent if we do not already have one */
 	if (kobj->kset) {
-		kobj->kset = kset_get(kobj->kset);
-
-		if (!parent) {
+		if (!parent)
 			parent = kobject_get(&kobj->kset->kobj);
-			/*
-			 * If the kset is our parent, get a second
-			 * reference, we drop both the kset and the
-			 * parent ref on cleanup
-			 */
-			kobject_get(parent);
-		}
-
-		spin_lock(&kobj->kset->list_lock);
-		list_add_tail(&kobj->entry, &kobj->kset->list);
-		spin_unlock(&kobj->kset->list_lock);
+		kobj_kset_join(kobj);
 		kobj->parent = parent;
 	}
 
+	pr_debug("kobject: '%s' (%p): %s: parent: '%s', set: '%s'\n",
+		 kobject_name(kobj), kobj, __FUNCTION__,
+		 parent ? kobject_name(parent) : "<NULL>",
+		 kobj->kset ? kobject_name(&kobj->kset->kobj) : "<NULL>" );
+
 	error = create_dir(kobj);
 	if (error) {
-		/* unlink does the kobject_put() for us */
-		unlink(kobj);
+		kobj_kset_leave(kobj);
+		kobject_put(parent);
+		kobj->parent = NULL;
 
 		/* be noisy on error issues */
 		if (error == -EEXIST)
@@ -214,7 +203,8 @@ static int kobject_add_internal(struct kobject *kobj)
 			printk(KERN_ERR "%s failed for %s (%d)\n",
 			       __FUNCTION__, kobject_name(kobj), error);
 		dump_stack();
-	}
+	} else
+		kobj->state_in_sysfs = 1;
 
 	return error;
 }
@@ -238,11 +228,13 @@ static int kobject_set_name_vargs(struct kobject *kobj, const char *fmt,
 	if (!name)
 		return -ENOMEM;
 
+
 	/* Free the old name, if necessary. */
 	kfree(kobj->k_name);
 
 	/* Now, set the new name */
 	kobj->k_name = name;
+	kobj->state_name_set = 1;
 
 	return 0;
 }
@@ -293,20 +285,25 @@ void kobject_init(struct kobject *kobj, struct kobj_type *ktype)
 		err_str = "must have a ktype to be initialized properly!\n";
 		goto error;
 	}
-	if (atomic_read(&kobj->kref.refcount)) {
+	if (kobj->state_initialized) {
 		/* do not error out as sometimes we can recover */
-		printk(KERN_ERR "kobject: reference count is already set, "
-		       "something is seriously wrong.\n");
+		printk(KERN_ERR "kobject (%p): tried to init an initialized "
+		       "object, something is seriously wrong.\n", kobj);
 		dump_stack();
 	}
 
 	kref_init(&kobj->kref);
 	INIT_LIST_HEAD(&kobj->entry);
 	kobj->ktype = ktype;
+	kobj->state_name_set = 0;
+	kobj->state_in_sysfs = 0;
+	kobj->state_add_uevent_sent = 0;
+	kobj->state_remove_uevent_sent = 0;
+	kobj->state_initialized = 1;
 	return;
 
 error:
-	printk(KERN_ERR "kobject: %s\n", err_str);
+	printk(KERN_ERR "kobject (%p): %s\n", kobj, err_str);
 	dump_stack();
 }
 EXPORT_SYMBOL(kobject_init);
@@ -345,17 +342,10 @@ static int kobject_add_varg(struct kobject *kobj, struct kobject *parent,
  *
  * If this function returns an error, kobject_put() must be called to
  * properly clean up the memory associated with the object.
- *
- * If the function is successful, the only way to properly clean up the
- * memory is with a call to kobject_del(), in which case, a call to
- * kobject_put() is not necessary (kobject_del() does the final
- * kobject_put() to call the release function in the ktype's release
- * pointer.)
- *
  * Under no instance should the kobject that is passed to this function
  * be directly freed with a call to kfree(), that can leak memory.
  *
- * Note, no uevent will be created with this call, the caller should set
+ * Note, no "add" uevent will be created with this call, the caller should set
  * up all of the necessary sysfs files for the object and then call
  * kobject_uevent() with the UEVENT_ADD parameter to ensure that
  * userspace is properly notified of this kobject's creation.
@@ -369,6 +359,13 @@ int kobject_add(struct kobject *kobj, struct kobject *parent,
 	if (!kobj)
 		return -EINVAL;
 
+	if (!kobj->state_initialized) {
+		printk(KERN_ERR "kobject '%s' (%p): tried to add an "
+		       "uninitialized object, something is seriously wrong.\n",
+		       kobject_name(kobj), kobj);
+		dump_stack();
+		return -EINVAL;
+	}
 	va_start(args, fmt);
 	retval = kobject_add_varg(kobj, parent, fmt, args);
 	va_end(args);
@@ -527,8 +524,12 @@ void kobject_del(struct kobject * kobj)
 {
 	if (!kobj)
 		return;
+
 	sysfs_remove_dir(kobj);
-	unlink(kobj);
+	kobj->state_in_sysfs = 0;
+	kobj_kset_leave(kobj);
+	kobject_put(kobj->parent);
+	kobj->parent = NULL;
 }
 
 /**
@@ -565,21 +566,43 @@ struct kobject * kobject_get(struct kobject * kobj)
  */
 static void kobject_cleanup(struct kobject *kobj)
 {
-	struct kobj_type * t = get_ktype(kobj);
-	struct kset * s = kobj->kset;
+	struct kobj_type *t = get_ktype(kobj);
 	const char *name = kobj->k_name;
+	int name_set = kobj->state_name_set;
 
 	pr_debug("kobject: '%s' (%p): %s\n",
 		 kobject_name(kobj), kobj, __FUNCTION__);
+
+	if (t && !t->release)
+		pr_debug("kobject: '%s' (%p): does not have a release() "
+			 "function, it is broken and must be fixed.\n",
+			 kobject_name(kobj), kobj);
+
+	/* send "remove" if the caller did not do it but sent "add" */
+	if (kobj->state_add_uevent_sent && !kobj->state_remove_uevent_sent) {
+		pr_debug("kobject: '%s' (%p): auto cleanup 'remove' event\n",
+			 kobject_name(kobj), kobj);
+		kobject_uevent(kobj, KOBJ_REMOVE);
+	}
+
+	/* remove from sysfs if the caller did not do it */
+	if (kobj->state_in_sysfs) {
+		pr_debug("kobject: '%s' (%p): auto cleanup kobject_del\n",
+			 kobject_name(kobj), kobj);
+		kobject_del(kobj);
+	}
+
 	if (t && t->release) {
+		pr_debug("kobject: '%s' (%p): calling ktype release\n",
+			 kobject_name(kobj), kobj);
 		t->release(kobj);
-		/* If we have a release function, we can guess that this was
-		 * not a statically allocated kobject, so we should be safe to
-		 * free the name */
+	}
+
+	/* free name if we allocated it */
+	if (name_set && name) {
+		pr_debug("kobject: '%s': free name\n", name);
 		kfree(name);
 	}
-	if (s)
-		kset_put(s);
 }
 
 static void kobject_release(struct kref *kref)
@@ -601,8 +624,7 @@ void kobject_put(struct kobject * kobj)
 
 static void dynamic_kobj_release(struct kobject *kobj)
 {
-	pr_debug("kobject: '%s' (%p): %s\n",
-		 kobject_name(kobj), kobj, __FUNCTION__);
+	pr_debug("kobject: (%p): %s\n", kobj, __FUNCTION__);
 	kfree(kobj);
 }
 
diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c
index 51dc4d287add..b021e67c4294 100644
--- a/lib/kobject_uevent.c
+++ b/lib/kobject_uevent.c
@@ -180,6 +180,17 @@ int kobject_uevent_env(struct kobject *kobj, enum kobject_action action,
 		}
 	}
 
+	/*
+	 * Mark "add" and "remove" events in the object to ensure proper
+	 * events to userspace during automatic cleanup. If the object did
+	 * send an "add" event, "remove" will automatically generated by
+	 * the core, if not already done by the caller.
+	 */
+	if (action == KOBJ_ADD)
+		kobj->state_add_uevent_sent = 1;
+	else if (action == KOBJ_REMOVE)
+		kobj->state_remove_uevent_sent = 1;
+
 	/* we will send an event, so request a new sequence number */
 	spin_lock(&sequence_lock);
 	seq = ++uevent_seqnum;
-- 
cgit v1.2.3


From 528a4bf1d5ffed310d26fc1d82d45c02949f71cf Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 20 Dec 2007 08:13:05 -0800
Subject: Kobject: remove kobject_unregister() as no one uses it anymore

There are no in-kernel users of kobject_unregister() so it should be
removed.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  2 --
 lib/kobject.c           | 17 -----------------
 2 files changed, 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index be03ce83f9cc..504ac0eb4412 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -101,8 +101,6 @@ extern struct kobject * __must_check kobject_create_and_add(const char *name,
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 extern int __must_check kobject_move(struct kobject *, struct kobject *);
 
-extern void kobject_unregister(struct kobject *);
-
 extern struct kobject * kobject_get(struct kobject *);
 extern void kobject_put(struct kobject *);
 
diff --git a/lib/kobject.c b/lib/kobject.c
index 462946ee3e64..a0773734545c 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -532,22 +532,6 @@ void kobject_del(struct kobject * kobj)
 	kobj->parent = NULL;
 }
 
-/**
- *	kobject_unregister - remove object from hierarchy and decrement refcount.
- *	@kobj:	object going away.
- */
-
-void kobject_unregister(struct kobject * kobj)
-{
-	if (!kobj)
-		return;
-	pr_debug("kobject: '%s' (%p): %s\n",
-		 kobject_name(kobj), kobj, __FUNCTION__);
-	kobject_uevent(kobj, KOBJ_REMOVE);
-	kobject_del(kobj);
-	kobject_put(kobj);
-}
-
 /**
  *	kobject_get - increment refcount for object.
  *	@kobj:	object.
@@ -877,7 +861,6 @@ struct kset *kset_create_and_add(const char *name,
 }
 EXPORT_SYMBOL_GPL(kset_create_and_add);
 
-EXPORT_SYMBOL(kobject_unregister);
 EXPORT_SYMBOL(kobject_get);
 EXPORT_SYMBOL(kobject_put);
 EXPORT_SYMBOL(kobject_del);
-- 
cgit v1.2.3


From af5ca3f4ec5cc4432a42a73b050dd8898ce8fd00 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Thu, 20 Dec 2007 02:09:39 +0100
Subject: Driver core: change sysdev classes to use dynamic kobject names

All kobjects require a dynamically allocated name now. We no longer
need to keep track if the name is statically assigned, we can just
unconditionally free() all kobject names on cleanup.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/kernel/time.c                   |  4 ++--
 arch/arm/mach-integrator/integrator_ap.c |  2 +-
 arch/arm/mach-pxa/cm-x270.c              |  2 +-
 arch/arm/mach-pxa/lpd270.c               |  2 +-
 arch/arm/mach-pxa/lubbock.c              |  2 +-
 arch/arm/mach-pxa/mainstone.c            |  2 +-
 arch/arm/mach-s3c2410/s3c2410.c          |  2 +-
 arch/arm/mach-s3c2412/s3c2412.c          |  2 +-
 arch/arm/mach-s3c2440/mach-osiris.c      |  2 +-
 arch/arm/mach-s3c2443/s3c2443.c          |  2 +-
 arch/arm/mach-sa1100/irq.c               |  2 +-
 arch/arm/oprofile/common.c               |  2 +-
 arch/arm/plat-omap/gpio.c                |  2 +-
 arch/arm/plat-s3c24xx/dma.c              |  2 +-
 arch/arm/plat-s3c24xx/s3c244x.c          |  4 ++--
 arch/avr32/kernel/time.c                 |  2 +-
 arch/mips/kernel/i8259.c                 |  2 +-
 arch/powerpc/platforms/cell/spu_base.c   |  2 +-
 arch/powerpc/platforms/powermac/pic.c    |  2 +-
 arch/powerpc/sysdev/ipic.c               |  2 +-
 arch/powerpc/sysdev/mpic.c               |  2 +-
 arch/powerpc/sysdev/qe_lib/qe_ic.c       |  2 +-
 arch/ppc/syslib/ipic.c                   |  2 +-
 arch/ppc/syslib/open_pic.c               |  2 +-
 arch/ppc/syslib/open_pic2.c              |  2 +-
 arch/s390/kernel/time.c                  |  2 +-
 arch/sh/drivers/dma/dma-sysfs.c          |  2 +-
 arch/sh/kernel/time.c                    |  2 +-
 arch/x86/kernel/apic_32.c                |  2 +-
 arch/x86/kernel/apic_64.c                |  2 +-
 arch/x86/kernel/cpu/mcheck/mce_64.c      |  2 +-
 arch/x86/kernel/i8237.c                  |  2 +-
 arch/x86/kernel/i8259_32.c               |  2 +-
 arch/x86/kernel/i8259_64.c               |  2 +-
 arch/x86/kernel/io_apic_32.c             |  2 +-
 arch/x86/kernel/io_apic_64.c             |  2 +-
 arch/x86/kernel/nmi_32.c                 |  2 +-
 arch/x86/kernel/nmi_64.c                 |  2 +-
 arch/x86/oprofile/nmi_int.c              |  2 +-
 drivers/acpi/pci_link.c                  |  2 +-
 drivers/base/class.c                     |  2 +-
 drivers/base/cpu.c                       |  2 +-
 drivers/base/memory.c                    |  2 +-
 drivers/base/node.c                      |  2 +-
 drivers/base/sys.c                       |  1 +
 drivers/edac/edac_module.c               |  2 +-
 drivers/kvm/kvm_main.c                   |  2 +-
 drivers/macintosh/via-pmu.c              |  2 +-
 drivers/scsi/libsas/sas_scsi_host.c      |  2 +-
 include/linux/kobject.h                  | 13 ++-----------
 include/linux/sysdev.h                   |  1 +
 kernel/rtmutex-tester.c                  |  2 +-
 kernel/time/clocksource.c                |  2 +-
 kernel/time/timekeeping.c                |  2 +-
 lib/kobject.c                            | 14 +++++---------
 55 files changed, 62 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index 1533d3ecd7a0..f6f3689a86ee 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -195,7 +195,7 @@ static int leds_shutdown(struct sys_device *dev)
 }
 
 static struct sysdev_class leds_sysclass = {
-	set_kset_name("leds"),
+	.name		= "leds",
 	.shutdown	= leds_shutdown,
 	.suspend	= leds_suspend,
 	.resume		= leds_resume,
@@ -369,7 +369,7 @@ static int timer_resume(struct sys_device *dev)
 #endif
 
 static struct sysdev_class timer_sysclass = {
-	set_kset_name("timer"),
+	.name		= "timer",
 	.suspend	= timer_suspend,
 	.resume		= timer_resume,
 };
diff --git a/arch/arm/mach-integrator/integrator_ap.c b/arch/arm/mach-integrator/integrator_ap.c
index 72280754354d..df37e93c6fc9 100644
--- a/arch/arm/mach-integrator/integrator_ap.c
+++ b/arch/arm/mach-integrator/integrator_ap.c
@@ -214,7 +214,7 @@ static int irq_resume(struct sys_device *dev)
 #endif
 
 static struct sysdev_class irq_class = {
-	set_kset_name("irq"),
+	.name		= "irq",
 	.suspend	= irq_suspend,
 	.resume		= irq_resume,
 };
diff --git a/arch/arm/mach-pxa/cm-x270.c b/arch/arm/mach-pxa/cm-x270.c
index 177664ccb2e2..a16349272f54 100644
--- a/arch/arm/mach-pxa/cm-x270.c
+++ b/arch/arm/mach-pxa/cm-x270.c
@@ -566,7 +566,7 @@ static int cmx270_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class cmx270_pm_sysclass = {
-	set_kset_name("pm"),
+	.name = "pm",
 	.resume = cmx270_resume,
 	.suspend = cmx270_suspend,
 };
diff --git a/arch/arm/mach-pxa/lpd270.c b/arch/arm/mach-pxa/lpd270.c
index 26116440a7c9..78ebad063cba 100644
--- a/arch/arm/mach-pxa/lpd270.c
+++ b/arch/arm/mach-pxa/lpd270.c
@@ -122,7 +122,7 @@ static int lpd270_irq_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class lpd270_irq_sysclass = {
-	set_kset_name("cpld_irq"),
+	.name = "cpld_irq",
 	.resume = lpd270_irq_resume,
 };
 
diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c
index 011a1a72b61c..1d3112dc629e 100644
--- a/arch/arm/mach-pxa/lubbock.c
+++ b/arch/arm/mach-pxa/lubbock.c
@@ -126,7 +126,7 @@ static int lubbock_irq_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class lubbock_irq_sysclass = {
-	set_kset_name("cpld_irq"),
+	.name = "cpld_irq",
 	.resume = lubbock_irq_resume,
 };
 
diff --git a/arch/arm/mach-pxa/mainstone.c b/arch/arm/mach-pxa/mainstone.c
index a4bc3483cbb3..41d8c6cea62b 100644
--- a/arch/arm/mach-pxa/mainstone.c
+++ b/arch/arm/mach-pxa/mainstone.c
@@ -120,7 +120,7 @@ static int mainstone_irq_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class mainstone_irq_sysclass = {
-	set_kset_name("cpld_irq"),
+	.name = "cpld_irq",
 	.resume = mainstone_irq_resume,
 };
 
diff --git a/arch/arm/mach-s3c2410/s3c2410.c b/arch/arm/mach-s3c2410/s3c2410.c
index e580303cb0ab..0e7991940f81 100644
--- a/arch/arm/mach-s3c2410/s3c2410.c
+++ b/arch/arm/mach-s3c2410/s3c2410.c
@@ -100,7 +100,7 @@ void __init s3c2410_init_clocks(int xtal)
 }
 
 struct sysdev_class s3c2410_sysclass = {
-	set_kset_name("s3c2410-core"),
+	.name = "s3c2410-core",
 };
 
 static struct sys_device s3c2410_sysdev = {
diff --git a/arch/arm/mach-s3c2412/s3c2412.c b/arch/arm/mach-s3c2412/s3c2412.c
index 4f92a1562d77..265cd3f567a3 100644
--- a/arch/arm/mach-s3c2412/s3c2412.c
+++ b/arch/arm/mach-s3c2412/s3c2412.c
@@ -196,7 +196,7 @@ void __init s3c2412_init_clocks(int xtal)
 */
 
 struct sysdev_class s3c2412_sysclass = {
-	set_kset_name("s3c2412-core"),
+	.name = "s3c2412-core",
 };
 
 static int __init s3c2412_core_init(void)
diff --git a/arch/arm/mach-s3c2440/mach-osiris.c b/arch/arm/mach-s3c2440/mach-osiris.c
index c326983f4a8f..78af7664988b 100644
--- a/arch/arm/mach-s3c2440/mach-osiris.c
+++ b/arch/arm/mach-s3c2440/mach-osiris.c
@@ -312,7 +312,7 @@ static int osiris_pm_resume(struct sys_device *sd)
 #endif
 
 static struct sysdev_class osiris_pm_sysclass = {
-	set_kset_name("mach-osiris"),
+	.name		= "mach-osiris",
 	.suspend	= osiris_pm_suspend,
 	.resume		= osiris_pm_resume,
 };
diff --git a/arch/arm/mach-s3c2443/s3c2443.c b/arch/arm/mach-s3c2443/s3c2443.c
index 8d8117158d23..9ce490560af9 100644
--- a/arch/arm/mach-s3c2443/s3c2443.c
+++ b/arch/arm/mach-s3c2443/s3c2443.c
@@ -43,7 +43,7 @@ static struct map_desc s3c2443_iodesc[] __initdata = {
 };
 
 struct sysdev_class s3c2443_sysclass = {
-	set_kset_name("s3c2443-core"),
+	.name = "s3c2443-core",
 };
 
 static struct sys_device s3c2443_sysdev = {
diff --git a/arch/arm/mach-sa1100/irq.c b/arch/arm/mach-sa1100/irq.c
index edf3347d9c5b..3dc17d7bf38e 100644
--- a/arch/arm/mach-sa1100/irq.c
+++ b/arch/arm/mach-sa1100/irq.c
@@ -283,7 +283,7 @@ static int sa1100irq_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class sa1100irq_sysclass = {
-	set_kset_name("sa11x0-irq"),
+	.name		= "sa11x0-irq",
 	.suspend	= sa1100irq_suspend,
 	.resume		= sa1100irq_resume,
 };
diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index a9de727c9327..0a5cf3a6438b 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -96,7 +96,7 @@ static int op_arm_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class oprofile_sysclass = {
-	set_kset_name("oprofile"),
+	.name		= "oprofile",
 	.resume		= op_arm_resume,
 	.suspend	= op_arm_suspend,
 };
diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c
index 6097753394ad..b2a87b8ef673 100644
--- a/arch/arm/plat-omap/gpio.c
+++ b/arch/arm/plat-omap/gpio.c
@@ -1455,7 +1455,7 @@ static int omap_gpio_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class omap_gpio_sysclass = {
-	set_kset_name("gpio"),
+	.name		= "gpio",
 	.suspend	= omap_gpio_suspend,
 	.resume		= omap_gpio_resume,
 };
diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c
index 29696e46ed65..aae1b9cbaf44 100644
--- a/arch/arm/plat-s3c24xx/dma.c
+++ b/arch/arm/plat-s3c24xx/dma.c
@@ -1265,7 +1265,7 @@ static int s3c2410_dma_resume(struct sys_device *dev)
 #endif /* CONFIG_PM */
 
 struct sysdev_class dma_sysclass = {
-	set_kset_name("s3c24xx-dma"),
+	.name		= "s3c24xx-dma",
 	.suspend	= s3c2410_dma_suspend,
 	.resume		= s3c2410_dma_resume,
 };
diff --git a/arch/arm/plat-s3c24xx/s3c244x.c b/arch/arm/plat-s3c24xx/s3c244x.c
index 3444b13afac5..f197bb3a2366 100644
--- a/arch/arm/plat-s3c24xx/s3c244x.c
+++ b/arch/arm/plat-s3c24xx/s3c244x.c
@@ -151,13 +151,13 @@ static int s3c244x_resume(struct sys_device *dev)
 /* Since the S3C2442 and S3C2440 share  items, put both sysclasses here */
 
 struct sysdev_class s3c2440_sysclass = {
-	set_kset_name("s3c2440-core"),
+	.name		= "s3c2440-core",
 	.suspend	= s3c244x_suspend,
 	.resume		= s3c244x_resume
 };
 
 struct sysdev_class s3c2442_sysclass = {
-	set_kset_name("s3c2442-core"),
+	.name		= "s3c2442-core",
 	.suspend	= s3c244x_suspend,
 	.resume		= s3c244x_resume
 };
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index 7014a3571ec0..36a46c3ae308 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -214,7 +214,7 @@ void __init time_init(void)
 }
 
 static struct sysdev_class timer_class = {
-	set_kset_name("timer"),
+	.name = "timer",
 };
 
 static struct sys_device timer_device = {
diff --git a/arch/mips/kernel/i8259.c b/arch/mips/kernel/i8259.c
index 471013577108..197d7977de35 100644
--- a/arch/mips/kernel/i8259.c
+++ b/arch/mips/kernel/i8259.c
@@ -238,7 +238,7 @@ static int i8259A_shutdown(struct sys_device *dev)
 }
 
 static struct sysdev_class i8259_sysdev_class = {
-	set_kset_name("i8259"),
+	.name = "i8259",
 	.resume = i8259A_resume,
 	.shutdown = i8259A_shutdown,
 };
diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c
index c83c3e3f5178..a08862203643 100644
--- a/arch/powerpc/platforms/cell/spu_base.c
+++ b/arch/powerpc/platforms/cell/spu_base.c
@@ -459,7 +459,7 @@ static int spu_shutdown(struct sys_device *sysdev)
 }
 
 static struct sysdev_class spu_sysdev_class = {
-	set_kset_name("spu"),
+	.name = "spu",
 	.shutdown = spu_shutdown,
 };
 
diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c
index 999f5e160897..84c0d4ef76a2 100644
--- a/arch/powerpc/platforms/powermac/pic.c
+++ b/arch/powerpc/platforms/powermac/pic.c
@@ -663,7 +663,7 @@ static int pmacpic_resume(struct sys_device *sysdev)
 #endif /* CONFIG_PM && CONFIG_PPC32 */
 
 static struct sysdev_class pmacpic_sysclass = {
-	set_kset_name("pmac_pic"),
+	.name = "pmac_pic",
 };
 
 static struct sys_device device_pmacpic = {
diff --git a/arch/powerpc/sysdev/ipic.c b/arch/powerpc/sysdev/ipic.c
index 05a56e55804c..e898ff4d2b97 100644
--- a/arch/powerpc/sysdev/ipic.c
+++ b/arch/powerpc/sysdev/ipic.c
@@ -725,7 +725,7 @@ unsigned int ipic_get_irq(void)
 }
 
 static struct sysdev_class ipic_sysclass = {
-	set_kset_name("ipic"),
+	.name = "ipic",
 };
 
 static struct sys_device device_ipic = {
diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index e47938899a92..212a94f5d34b 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -1584,7 +1584,7 @@ static struct sysdev_class mpic_sysclass = {
 	.resume = mpic_resume,
 	.suspend = mpic_suspend,
 #endif
-	set_kset_name("mpic"),
+	.name = "mpic",
 };
 
 static int mpic_init_sys(void)
diff --git a/arch/powerpc/sysdev/qe_lib/qe_ic.c b/arch/powerpc/sysdev/qe_lib/qe_ic.c
index e1c0fd6dbc1a..f59444d3be75 100644
--- a/arch/powerpc/sysdev/qe_lib/qe_ic.c
+++ b/arch/powerpc/sysdev/qe_lib/qe_ic.c
@@ -483,7 +483,7 @@ int qe_ic_set_high_priority(unsigned int virq, unsigned int priority, int high)
 }
 
 static struct sysdev_class qe_ic_sysclass = {
-	set_kset_name("qe_ic"),
+	.name = "qe_ic",
 };
 
 static struct sys_device device_qe_ic = {
diff --git a/arch/ppc/syslib/ipic.c b/arch/ppc/syslib/ipic.c
index 9192777d0f78..4f163e20939e 100644
--- a/arch/ppc/syslib/ipic.c
+++ b/arch/ppc/syslib/ipic.c
@@ -614,7 +614,7 @@ int ipic_get_irq(void)
 }
 
 static struct sysdev_class ipic_sysclass = {
-	set_kset_name("ipic"),
+	.name = "ipic",
 };
 
 static struct sys_device device_ipic = {
diff --git a/arch/ppc/syslib/open_pic.c b/arch/ppc/syslib/open_pic.c
index 18ec94733293..da36522d327a 100644
--- a/arch/ppc/syslib/open_pic.c
+++ b/arch/ppc/syslib/open_pic.c
@@ -1043,7 +1043,7 @@ int openpic_resume(struct sys_device *sysdev)
 #endif /* CONFIG_PM */
 
 static struct sysdev_class openpic_sysclass = {
-	set_kset_name("openpic"),
+	.name = "openpic",
 };
 
 static struct sys_device device_openpic = {
diff --git a/arch/ppc/syslib/open_pic2.c b/arch/ppc/syslib/open_pic2.c
index d585207f9f77..449075a04798 100644
--- a/arch/ppc/syslib/open_pic2.c
+++ b/arch/ppc/syslib/open_pic2.c
@@ -666,7 +666,7 @@ int openpic2_resume(struct sys_device *sysdev)
 
 /* HACK ALERT */
 static struct sysdev_class openpic2_sysclass = {
-	set_kset_name("openpic2"),
+	.name = "openpic2",
 };
 
 static struct sys_device device_openpic2 = {
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 22b800ce2126..3bbac1293be4 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -1145,7 +1145,7 @@ static void etr_work_fn(struct work_struct *work)
  * Sysfs interface functions
  */
 static struct sysdev_class etr_sysclass = {
-	set_kset_name("etr")
+	.name	= "etr",
 };
 
 static struct sys_device etr_port0_dev = {
diff --git a/arch/sh/drivers/dma/dma-sysfs.c b/arch/sh/drivers/dma/dma-sysfs.c
index eebcd4768bbf..51b57c0d1a3c 100644
--- a/arch/sh/drivers/dma/dma-sysfs.c
+++ b/arch/sh/drivers/dma/dma-sysfs.c
@@ -19,7 +19,7 @@
 #include <asm/dma.h>
 
 static struct sysdev_class dma_sysclass = {
-	set_kset_name("dma"),
+	.name = "dma",
 };
 EXPORT_SYMBOL(dma_sysclass);
 
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index a3a67d151e52..2bc04bfee738 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -174,7 +174,7 @@ int timer_resume(struct sys_device *dev)
 #endif
 
 static struct sysdev_class timer_sysclass = {
-	set_kset_name("timer"),
+	.name	 = "timer",
 	.suspend = timer_suspend,
 	.resume	 = timer_resume,
 };
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index edb5108e5d0e..a56c782653be 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1530,7 +1530,7 @@ static int lapic_resume(struct sys_device *dev)
  */
 
 static struct sysdev_class lapic_sysclass = {
-	set_kset_name("lapic"),
+	.name		= "lapic",
 	.resume		= lapic_resume,
 	.suspend	= lapic_suspend,
 };
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index f28ccb588fba..fa6cdee6d303 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -639,7 +639,7 @@ static int lapic_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class lapic_sysclass = {
-	set_kset_name("lapic"),
+	.name		= "lapic",
 	.resume		= lapic_resume,
 	.suspend	= lapic_suspend,
 };
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 4b21d29fb5aa..242e8668dbeb 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -745,7 +745,7 @@ static void mce_restart(void)
 
 static struct sysdev_class mce_sysclass = {
 	.resume = mce_resume,
-	set_kset_name("machinecheck"),
+	.name = "machinecheck",
 };
 
 DEFINE_PER_CPU(struct sys_device, device_mce);
diff --git a/arch/x86/kernel/i8237.c b/arch/x86/kernel/i8237.c
index 29313832df0c..dbd6c1d1b638 100644
--- a/arch/x86/kernel/i8237.c
+++ b/arch/x86/kernel/i8237.c
@@ -51,7 +51,7 @@ static int i8237A_suspend(struct sys_device *dev, pm_message_t state)
 }
 
 static struct sysdev_class i8237_sysdev_class = {
-	set_kset_name("i8237"),
+	.name = "i8237",
 	.suspend = i8237A_suspend,
 	.resume = i8237A_resume,
 };
diff --git a/arch/x86/kernel/i8259_32.c b/arch/x86/kernel/i8259_32.c
index f634fc715c99..5f3496d01984 100644
--- a/arch/x86/kernel/i8259_32.c
+++ b/arch/x86/kernel/i8259_32.c
@@ -258,7 +258,7 @@ static int i8259A_shutdown(struct sys_device *dev)
 }
 
 static struct sysdev_class i8259_sysdev_class = {
-	set_kset_name("i8259"),
+	.name = "i8259",
 	.suspend = i8259A_suspend,
 	.resume = i8259A_resume,
 	.shutdown = i8259A_shutdown,
diff --git a/arch/x86/kernel/i8259_64.c b/arch/x86/kernel/i8259_64.c
index 3f27ea0b9816..ba6d57286f56 100644
--- a/arch/x86/kernel/i8259_64.c
+++ b/arch/x86/kernel/i8259_64.c
@@ -370,7 +370,7 @@ static int i8259A_shutdown(struct sys_device *dev)
 }
 
 static struct sysdev_class i8259_sysdev_class = {
-	set_kset_name("i8259"),
+	.name = "i8259",
 	.suspend = i8259A_suspend,
 	.resume = i8259A_resume,
 	.shutdown = i8259A_shutdown,
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index a6b1490e00c4..ab77f1905469 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -2401,7 +2401,7 @@ static int ioapic_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class ioapic_sysdev_class = {
-	set_kset_name("ioapic"),
+	.name = "ioapic",
 	.suspend = ioapic_suspend,
 	.resume = ioapic_resume,
 };
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index cbac1670c7c3..23a3ac06a23e 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1850,7 +1850,7 @@ static int ioapic_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class ioapic_sysdev_class = {
-	set_kset_name("ioapic"),
+	.name = "ioapic",
 	.suspend = ioapic_suspend,
 	.resume = ioapic_resume,
 };
diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 852db2906921..4f4bfd3a88b6 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -176,7 +176,7 @@ static int lapic_nmi_resume(struct sys_device *dev)
 
 
 static struct sysdev_class nmi_sysclass = {
-	set_kset_name("lapic_nmi"),
+	.name		= "lapic_nmi",
 	.resume		= lapic_nmi_resume,
 	.suspend	= lapic_nmi_suspend,
 };
diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 4253c4e8849c..c3d1476b6a11 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -211,7 +211,7 @@ static int lapic_nmi_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class nmi_sysclass = {
-	set_kset_name("lapic_nmi"),
+	.name		= "lapic_nmi",
 	.resume		= lapic_nmi_resume,
 	.suspend	= lapic_nmi_suspend,
 };
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 944bbcdd2b8d..c8ab79ef4276 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -51,7 +51,7 @@ static int nmi_resume(struct sys_device *dev)
 
 
 static struct sysdev_class oprofile_sysclass = {
-	set_kset_name("oprofile"),
+	.name		= "oprofile",
 	.resume		= nmi_resume,
 	.suspend	= nmi_suspend,
 };
diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c
index c9f526e55392..5400ea173f6f 100644
--- a/drivers/acpi/pci_link.c
+++ b/drivers/acpi/pci_link.c
@@ -911,7 +911,7 @@ __setup("acpi_irq_balance", acpi_irq_balance_set);
 
 /* FIXME: we will remove this interface after all drivers call pci_disable_device */
 static struct sysdev_class irqrouter_sysdev_class = {
-	set_kset_name("irqrouter"),
+	.name = "irqrouter",
 	.resume = irqrouter_resume,
 };
 
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 61fd26cc9f0e..b962a76875d2 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -466,7 +466,6 @@ static struct kset_uevent_ops class_uevent_ops = {
  * entirely soon.
  */
 static struct kset class_obj_subsys = {
-	.kobj = { .k_name = "class_obj", },
 	.uevent_ops = &class_uevent_ops,
 };
 
@@ -872,6 +871,7 @@ int __init classes_init(void)
 	/* ick, this is ugly, the things we go through to keep from showing up
 	 * in sysfs... */
 	kset_init(&class_obj_subsys);
+	kobject_set_name(&class_obj_subsys.kobj, "class_obj");
 	if (!class_obj_subsys.kobj.parent)
 		class_obj_subsys.kobj.parent = &class_obj_subsys.kobj;
 	return 0;
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 40545071e3c9..c5885f5ce0ac 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -14,7 +14,7 @@
 #include "base.h"
 
 struct sysdev_class cpu_sysdev_class = {
-	set_kset_name("cpu"),
+	.name = "cpu",
 };
 EXPORT_SYMBOL(cpu_sysdev_class);
 
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 7868707c7eda..7ae413fdd5fc 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -26,7 +26,7 @@
 #define MEMORY_CLASS_NAME	"memory"
 
 static struct sysdev_class memory_sysdev_class = {
-	set_kset_name(MEMORY_CLASS_NAME),
+	.name = MEMORY_CLASS_NAME,
 };
 
 static const char *memory_uevent_name(struct kset *kset, struct kobject *kobj)
diff --git a/drivers/base/node.c b/drivers/base/node.c
index 88eeed72b5d6..e59861f18ce5 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -15,7 +15,7 @@
 #include <linux/device.h>
 
 static struct sysdev_class node_class = {
-	set_kset_name("node"),
+	.name = "node",
 };
 
 
diff --git a/drivers/base/sys.c b/drivers/base/sys.c
index e666441dd76b..2f79c55acdcc 100644
--- a/drivers/base/sys.c
+++ b/drivers/base/sys.c
@@ -136,6 +136,7 @@ int sysdev_class_register(struct sysdev_class * cls)
 	cls->kset.kobj.parent = &system_kset->kobj;
 	cls->kset.kobj.ktype = &ktype_sysdev_class;
 	cls->kset.kobj.kset = system_kset;
+	kobject_set_name(&cls->kset.kobj, cls->name);
 	return kset_register(&cls->kset);
 }
 
diff --git a/drivers/edac/edac_module.c b/drivers/edac/edac_module.c
index e0c4a4086055..7e1374afd967 100644
--- a/drivers/edac/edac_module.c
+++ b/drivers/edac/edac_module.c
@@ -31,7 +31,7 @@ struct workqueue_struct *edac_workqueue;
  *	need to export to other files in this modules
  */
 static struct sysdev_class edac_class = {
-	set_kset_name("edac"),
+	.name = "edac",
 };
 static int edac_class_valid;
 
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 47c10b8f89b3..c0f372f1d761 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -3451,7 +3451,7 @@ static int kvm_resume(struct sys_device *dev)
 }
 
 static struct sysdev_class kvm_sysdev_class = {
-	set_kset_name("kvm"),
+	.name = "kvm",
 	.suspend = kvm_suspend,
 	.resume = kvm_resume,
 };
diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c
index 6123c70153d3..ac420b17e16f 100644
--- a/drivers/macintosh/via-pmu.c
+++ b/drivers/macintosh/via-pmu.c
@@ -2796,7 +2796,7 @@ static int pmu_sys_resume(struct sys_device *sysdev)
 #endif /* CONFIG_PM_SLEEP && CONFIG_PPC32 */
 
 static struct sysdev_class pmu_sysclass = {
-	set_kset_name("pmu"),
+	.name = "pmu",
 };
 
 static struct sys_device device_pmu = {
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 7663841eb4cf..a3fdc57e2673 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -464,7 +464,7 @@ int sas_eh_bus_reset_handler(struct scsi_cmnd *cmd)
 	res = sas_phy_reset(phy, 1);
 	if (res)
 		SAS_DPRINTK("Bus reset of %s failed 0x%x\n",
-			    phy->dev.kobj.k_name,
+			    kobject_name(&phy->dev.kobj),
 			    res);
 	if (res == TMF_RESP_FUNC_SUCC || res == TMF_RESP_FUNC_COMPLETE)
 		return SUCCESS;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 504ac0eb4412..4adbe1d83081 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -61,7 +61,7 @@ enum kobject_action {
 };
 
 struct kobject {
-	const char		* k_name;
+	const char		*name;
 	struct kref		kref;
 	struct list_head	entry;
 	struct kobject		* parent;
@@ -69,7 +69,6 @@ struct kobject {
 	struct kobj_type	* ktype;
 	struct sysfs_dirent	* sd;
 	unsigned int state_initialized:1;
-	unsigned int state_name_set:1;
 	unsigned int state_in_sysfs:1;
 	unsigned int state_add_uevent_sent:1;
 	unsigned int state_remove_uevent_sent:1;
@@ -80,7 +79,7 @@ extern int kobject_set_name(struct kobject *, const char *, ...)
 
 static inline const char * kobject_name(const struct kobject * kobj)
 {
-	return kobj->k_name;
+	return kobj->name;
 }
 
 extern void kobject_init(struct kobject *kobj, struct kobj_type *ktype);
@@ -189,14 +188,6 @@ static inline struct kobj_type *get_ktype(struct kobject *kobj)
 
 extern struct kobject * kset_find_obj(struct kset *, const char *);
 
-
-/*
- * Use this when initializing an embedded kset with no other 
- * fields to initialize.
- */
-#define set_kset_name(str)	.kset = { .kobj = { .k_name = str } }
-
-
 /* The global /sys/kernel/ kobject for people to chain off of */
 extern struct kobject *kernel_kobj;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
diff --git a/include/linux/sysdev.h b/include/linux/sysdev.h
index e285746588d6..f752e73bf977 100644
--- a/include/linux/sysdev.h
+++ b/include/linux/sysdev.h
@@ -29,6 +29,7 @@
 struct sys_device;
 
 struct sysdev_class {
+	const char *name;
 	struct list_head	drivers;
 
 	/* Default operations for these types of devices */
diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c
index e3055ba69159..092e4c620af9 100644
--- a/kernel/rtmutex-tester.c
+++ b/kernel/rtmutex-tester.c
@@ -394,7 +394,7 @@ static SYSDEV_ATTR(status, 0600, sysfs_test_status, NULL);
 static SYSDEV_ATTR(command, 0600, NULL, sysfs_test_command);
 
 static struct sysdev_class rttest_sysclass = {
-	set_kset_name("rttest"),
+	.name = "rttest",
 };
 
 static int init_test_thread(int id)
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index c8a9d13874df..8d6125ad2cf0 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -441,7 +441,7 @@ static SYSDEV_ATTR(available_clocksource, 0600,
 		   sysfs_show_available_clocksources, NULL);
 
 static struct sysdev_class clocksource_sysclass = {
-	set_kset_name("clocksource"),
+	.name = "clocksource",
 };
 
 static struct sys_device device_clocksource = {
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index e5e466b27598..ab46ae8c062b 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -335,9 +335,9 @@ static int timekeeping_suspend(struct sys_device *dev, pm_message_t state)
 
 /* sysfs resume/suspend bits for timekeeping */
 static struct sysdev_class timekeeping_sysclass = {
+	.name		= "timekeeping",
 	.resume		= timekeeping_resume,
 	.suspend	= timekeeping_suspend,
-	set_kset_name("timekeeping"),
 };
 
 static struct sys_device device_timer = {
diff --git a/lib/kobject.c b/lib/kobject.c
index a0773734545c..8dc32454661d 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -165,7 +165,7 @@ static int kobject_add_internal(struct kobject *kobj)
 	if (!kobj)
 		return -ENOENT;
 
-	if (!kobj->k_name || !kobj->k_name[0]) {
+	if (!kobj->name || !kobj->name[0]) {
 		pr_debug("kobject: (%p): attempted to be registered with empty "
 			 "name!\n", kobj);
 		WARN_ON(1);
@@ -228,13 +228,11 @@ static int kobject_set_name_vargs(struct kobject *kobj, const char *fmt,
 	if (!name)
 		return -ENOMEM;
 
-
 	/* Free the old name, if necessary. */
-	kfree(kobj->k_name);
+	kfree(kobj->name);
 
 	/* Now, set the new name */
-	kobj->k_name = name;
-	kobj->state_name_set = 1;
+	kobj->name = name;
 
 	return 0;
 }
@@ -295,7 +293,6 @@ void kobject_init(struct kobject *kobj, struct kobj_type *ktype)
 	kref_init(&kobj->kref);
 	INIT_LIST_HEAD(&kobj->entry);
 	kobj->ktype = ktype;
-	kobj->state_name_set = 0;
 	kobj->state_in_sysfs = 0;
 	kobj->state_add_uevent_sent = 0;
 	kobj->state_remove_uevent_sent = 0;
@@ -551,8 +548,7 @@ struct kobject * kobject_get(struct kobject * kobj)
 static void kobject_cleanup(struct kobject *kobj)
 {
 	struct kobj_type *t = get_ktype(kobj);
-	const char *name = kobj->k_name;
-	int name_set = kobj->state_name_set;
+	const char *name = kobj->name;
 
 	pr_debug("kobject: '%s' (%p): %s\n",
 		 kobject_name(kobj), kobj, __FUNCTION__);
@@ -583,7 +579,7 @@ static void kobject_cleanup(struct kobject *kobj)
 	}
 
 	/* free name if we allocated it */
-	if (name_set && name) {
+	if (name) {
 		pr_debug("kobject: '%s': free name\n", name);
 		kfree(name);
 	}
-- 
cgit v1.2.3


From ae72cddb2338bc36b991674a56a7bf70ae104d9e Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 11 Jan 2008 17:24:53 +1100
Subject: Driver Core: constify the name passed to
 platform_device_register_simple

This name is just passed to platform_device_alloc which has its parameter
declared const.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/platform.c         | 2 +-
 include/linux/platform_device.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index bdd59e8358f0..48d5db4f92ee 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -360,7 +360,7 @@ EXPORT_SYMBOL_GPL(platform_device_unregister);
  *	the Linux driver model.  In particular, when such drivers are built
  *	as modules, they can't be "hotplugged".
  */
-struct platform_device *platform_device_register_simple(char *name, int id,
+struct platform_device *platform_device_register_simple(const char *name, int id,
 							struct resource *res, unsigned int num)
 {
 	struct platform_device *pdev;
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index e80804316cdb..3261681c82a4 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -35,7 +35,7 @@ extern struct resource *platform_get_resource_byname(struct platform_device *, u
 extern int platform_get_irq_byname(struct platform_device *, char *);
 extern int platform_add_devices(struct platform_device **, int);
 
-extern struct platform_device *platform_device_register_simple(char *, int id,
+extern struct platform_device *platform_device_register_simple(const char *, int id,
 					struct resource *, unsigned int);
 
 extern struct platform_device *platform_device_alloc(const char *name, int id);
-- 
cgit v1.2.3


From fd04897bb20be29d60f7e426a053545aebeaa61a Mon Sep 17 00:00:00 2001
From: Dave Young <hidave.darkstar@gmail.com>
Date: Tue, 22 Jan 2008 15:27:08 +0800
Subject: Driver Core: add class iteration api

Add the following class iteration functions for driver use:
	class_for_each_device
	class_find_device
	class_for_each_child
	class_find_child

Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 133 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h |   9 +++-
 2 files changed, 140 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index b962a76875d2..9f737ff0fc71 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -809,6 +809,139 @@ void class_device_put(struct class_device *class_dev)
 		kobject_put(&class_dev->kobj);
 }
 
+/**
+ * class_for_each_device - device iterator
+ * @class: the class we're iterating
+ * @data: data for the callback
+ * @fn: function to be called for each device
+ *
+ * Iterate over @class's list of devices, and call @fn for each,
+ * passing it @data.
+ *
+ * We check the return of @fn each time. If it returns anything
+ * other than 0, we break out and return that value.
+ *
+ * Note, we hold class->sem in this function, so it can not be
+ * re-acquired in @fn, otherwise it will self-deadlocking. For
+ * example, calls to add or remove class members would be verboten.
+ */
+int class_for_each_device(struct class *class, void *data,
+			   int (*fn)(struct device *, void *))
+{
+	struct device *dev;
+	int error = 0;
+
+	if (!class)
+		return -EINVAL;
+	down(&class->sem);
+	list_for_each_entry(dev, &class->devices, node) {
+		dev = get_device(dev);
+		if (dev) {
+			error = fn(dev, data);
+			put_device(dev);
+		} else
+			error = -ENODEV;
+		if (error)
+			break;
+	}
+	up(&class->sem);
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(class_for_each_device);
+
+/**
+ * class_find_device - device iterator for locating a particular device
+ * @class: the class we're iterating
+ * @data: data for the match function
+ * @match: function to check device
+ *
+ * This is similar to the class_for_each_dev() function above, but it
+ * returns a reference to a device that is 'found' for later use, as
+ * determined by the @match callback.
+ *
+ * The callback should return 0 if the device doesn't match and non-zero
+ * if it does.  If the callback returns non-zero, this function will
+ * return to the caller and not iterate over any more devices.
+
+ * Note, you will need to drop the reference with put_device() after use.
+ *
+ * We hold class->sem in this function, so it can not be
+ * re-acquired in @match, otherwise it will self-deadlocking. For
+ * example, calls to add or remove class members would be verboten.
+ */
+struct device *class_find_device(struct class *class, void *data,
+				   int (*match)(struct device *, void *))
+{
+	struct device *dev;
+	int found = 0;
+
+	if (!class)
+		return NULL;
+
+	down(&class->sem);
+	list_for_each_entry(dev, &class->devices, node) {
+		dev = get_device(dev);
+		if (dev) {
+			if (match(dev, data)) {
+				found = 1;
+				break;
+			} else
+				put_device(dev);
+		} else
+			break;
+	}
+	up(&class->sem);
+
+	return found ? dev : NULL;
+}
+EXPORT_SYMBOL_GPL(class_find_device);
+
+/**
+ * class_find_child - device iterator for locating a particular class_device
+ * @class: the class we're iterating
+ * @data: data for the match function
+ * @match: function to check class_device
+ *
+ * This function returns a reference to a class_device that is 'found' for
+ * later use, as determined by the @match callback.
+ *
+ * The callback should return 0 if the class_device doesn't match and non-zero
+ * if it does.  If the callback returns non-zero, this function will
+ * return to the caller and not iterate over any more class_devices.
+ *
+ * Note, you will need to drop the reference with class_device_put() after use.
+ *
+ * We hold class->sem in this function, so it can not be
+ * re-acquired in @match, otherwise it will self-deadlocking. For
+ * example, calls to add or remove class members would be verboten.
+ */
+struct class_device *class_find_child(struct class *class, void *data,
+				   int (*match)(struct class_device *, void *))
+{
+	struct class_device *dev;
+	int found = 0;
+
+	if (!class)
+		return NULL;
+
+	down(&class->sem);
+	list_for_each_entry(dev, &class->children, node) {
+		dev = class_device_get(dev);
+		if (dev) {
+			if (match(dev, data)) {
+				found = 1;
+				break;
+			} else
+				class_device_put(dev);
+		} else
+			break;
+	}
+	up(&class->sem);
+
+	return found ? dev : NULL;
+}
+EXPORT_SYMBOL_GPL(class_find_child);
 
 int class_interface_register(struct class_interface *class_intf)
 {
diff --git a/include/linux/device.h b/include/linux/device.h
index 92ba3a874627..cdaf57bf4d19 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -177,8 +177,7 @@ struct class {
 	struct list_head	devices;
 	struct list_head	interfaces;
 	struct kset		class_dirs;
-	struct semaphore	sem;	/* locks both the children and interfaces lists */
-
+	struct semaphore	sem; /* locks children, devices, interfaces */
 	struct class_attribute		* class_attrs;
 	struct class_device_attribute	* class_dev_attrs;
 	struct device_attribute		* dev_attrs;
@@ -196,6 +195,12 @@ struct class {
 
 extern int __must_check class_register(struct class *);
 extern void class_unregister(struct class *);
+extern int class_for_each_device(struct class *class, void *data,
+				 int (*fn)(struct device *dev, void *data));
+extern struct device *class_find_device(struct class *class, void *data,
+					int (*match)(struct device *, void *));
+extern struct class_device *class_find_child(struct class *class, void *data,
+				   int (*match)(struct class_device *, void *));
 
 
 struct class_attribute {
-- 
cgit v1.2.3


From d462943afee8bff610258a82dba666e8ab72c9c8 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 24 Jan 2008 21:04:46 -0800
Subject: Driver core: fix coding style issues in device.h

Finally clean up the odd spaces and other mess in device.h

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h | 283 +++++++++++++++++++++++++------------------------
 1 file changed, 145 insertions(+), 138 deletions(-)

(limited to 'include')

diff --git a/include/linux/device.h b/include/linux/device.h
index cdaf57bf4d19..1880208964d6 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -25,7 +25,8 @@
 #include <asm/device.h>
 
 #define DEVICE_NAME_SIZE	50
-#define DEVICE_NAME_HALF	__stringify(20)	/* Less than half to accommodate slop */
+/* DEVICE_NAME_HALF is really less than half to accommodate slop */
+#define DEVICE_NAME_HALF	__stringify(20)
 #define DEVICE_ID_SIZE		32
 #define BUS_ID_SIZE		KOBJ_NAME_LEN
 
@@ -40,52 +41,53 @@ struct bus_type_private;
 
 struct bus_attribute {
 	struct attribute	attr;
-	ssize_t (*show)(struct bus_type *, char * buf);
-	ssize_t (*store)(struct bus_type *, const char * buf, size_t count);
+	ssize_t (*show)(struct bus_type *bus, char *buf);
+	ssize_t (*store)(struct bus_type *bus, const char *buf, size_t count);
 };
 
-#define BUS_ATTR(_name,_mode,_show,_store)	\
-struct bus_attribute bus_attr_##_name = __ATTR(_name,_mode,_show,_store)
+#define BUS_ATTR(_name, _mode, _show, _store)	\
+struct bus_attribute bus_attr_##_name = __ATTR(_name, _mode, _show, _store)
 
 extern int __must_check bus_create_file(struct bus_type *,
 					struct bus_attribute *);
 extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct bus_type {
-	const char		* name;
-	struct bus_attribute	* bus_attrs;
-	struct device_attribute	* dev_attrs;
-	struct driver_attribute	* drv_attrs;
-
-	int		(*match)(struct device * dev, struct device_driver * drv);
-	int		(*uevent)(struct device *dev, struct kobj_uevent_env *env);
-	int		(*probe)(struct device * dev);
-	int		(*remove)(struct device * dev);
-	void		(*shutdown)(struct device * dev);
-
-	int (*suspend)(struct device * dev, pm_message_t state);
-	int (*suspend_late)(struct device * dev, pm_message_t state);
-	int (*resume_early)(struct device * dev);
-	int (*resume)(struct device * dev);
+	const char		*name;
+	struct bus_attribute	*bus_attrs;
+	struct device_attribute	*dev_attrs;
+	struct driver_attribute	*drv_attrs;
+
+	int (*match)(struct device *dev, struct device_driver *drv);
+	int (*uevent)(struct device *dev, struct kobj_uevent_env *env);
+	int (*probe)(struct device *dev);
+	int (*remove)(struct device *dev);
+	void (*shutdown)(struct device *dev);
+
+	int (*suspend)(struct device *dev, pm_message_t state);
+	int (*suspend_late)(struct device *dev, pm_message_t state);
+	int (*resume_early)(struct device *dev);
+	int (*resume)(struct device *dev);
 
 	struct bus_type_private *p;
 };
 
-extern int __must_check bus_register(struct bus_type * bus);
-extern void bus_unregister(struct bus_type * bus);
+extern int __must_check bus_register(struct bus_type *bus);
+extern void bus_unregister(struct bus_type *bus);
 
-extern int __must_check bus_rescan_devices(struct bus_type * bus);
+extern int __must_check bus_rescan_devices(struct bus_type *bus);
 
 /* iterator helpers for buses */
 
-int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data,
-		     int (*fn)(struct device *, void *));
-struct device * bus_find_device(struct bus_type *bus, struct device *start,
-				void *data, int (*match)(struct device *, void *));
+int bus_for_each_dev(struct bus_type *bus, struct device *start, void *data,
+		     int (*fn)(struct device *dev, void *data));
+struct device *bus_find_device(struct bus_type *bus, struct device *start,
+			       void *data,
+			       int (*match)(struct device *dev, void *data));
 
 int __must_check bus_for_each_drv(struct bus_type *bus,
-		struct device_driver *start, void *data,
-		int (*fn)(struct device_driver *, void *));
+				  struct device_driver *start, void *data,
+				  int (*fn)(struct device_driver *, void *));
 
 /*
  * Bus notifiers: Get notified of addition/removal of devices
@@ -120,57 +122,63 @@ struct device_driver {
 	struct module		*owner;
 	const char 		*mod_name;	/* used for built-in modules */
 
-	int	(*probe)	(struct device * dev);
-	int	(*remove)	(struct device * dev);
-	void	(*shutdown)	(struct device * dev);
-	int	(*suspend)	(struct device * dev, pm_message_t state);
-	int	(*resume)	(struct device * dev);
+	int (*probe) (struct device *dev);
+	int (*remove) (struct device *dev);
+	void (*shutdown) (struct device *dev);
+	int (*suspend) (struct device *dev, pm_message_t state);
+	int (*resume) (struct device *dev);
 	struct attribute_group **groups;
 
 	struct driver_private *p;
 };
 
 
-extern int __must_check driver_register(struct device_driver * drv);
-extern void driver_unregister(struct device_driver * drv);
+extern int __must_check driver_register(struct device_driver *drv);
+extern void driver_unregister(struct device_driver *drv);
 
-extern struct device_driver * get_driver(struct device_driver * drv);
-extern void put_driver(struct device_driver * drv);
-extern struct device_driver *driver_find(const char *name, struct bus_type *bus);
+extern struct device_driver *get_driver(struct device_driver *drv);
+extern void put_driver(struct device_driver *drv);
+extern struct device_driver *driver_find(const char *name,
+					 struct bus_type *bus);
 extern int driver_probe_done(void);
 
 /* sysfs interface for exporting driver attributes */
 
 struct driver_attribute {
-	struct attribute	attr;
-	ssize_t (*show)(struct device_driver *, char * buf);
-	ssize_t (*store)(struct device_driver *, const char * buf, size_t count);
+	struct attribute attr;
+	ssize_t (*show)(struct device_driver *driver, char *buf);
+	ssize_t (*store)(struct device_driver *driver, const char *buf,
+			 size_t count);
 };
 
-#define DRIVER_ATTR(_name,_mode,_show,_store)	\
-struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store)
+#define DRIVER_ATTR(_name, _mode, _show, _store)	\
+struct driver_attribute driver_attr_##_name =		\
+	__ATTR(_name, _mode, _show, _store)
 
-extern int __must_check driver_create_file(struct device_driver *,
-					struct driver_attribute *);
-extern void driver_remove_file(struct device_driver *, struct driver_attribute *);
+extern int __must_check driver_create_file(struct device_driver *driver,
+					   struct driver_attribute *attr);
+extern void driver_remove_file(struct device_driver *driver,
+			       struct driver_attribute *attr);
 
 extern int __must_check driver_add_kobj(struct device_driver *drv,
 					struct kobject *kobj,
 					const char *fmt, ...);
 
-extern int __must_check driver_for_each_device(struct device_driver * drv,
-		struct device *start, void *data,
-		int (*fn)(struct device *, void *));
-struct device * driver_find_device(struct device_driver *drv,
-				   struct device *start, void *data,
-				   int (*match)(struct device *, void *));
+extern int __must_check driver_for_each_device(struct device_driver *drv,
+					       struct device *start,
+					       void *data,
+					       int (*fn)(struct device *dev,
+							 void *));
+struct device *driver_find_device(struct device_driver *drv,
+				  struct device *start, void *data,
+				  int (*match)(struct device *dev, void *data));
 
 /*
  * device classes
  */
 struct class {
-	const char		* name;
-	struct module		* owner;
+	const char		*name;
+	struct module		*owner;
 
 	struct kset		subsys;
 	struct list_head	children;
@@ -178,23 +186,23 @@ struct class {
 	struct list_head	interfaces;
 	struct kset		class_dirs;
 	struct semaphore	sem; /* locks children, devices, interfaces */
-	struct class_attribute		* class_attrs;
-	struct class_device_attribute	* class_dev_attrs;
-	struct device_attribute		* dev_attrs;
+	struct class_attribute		*class_attrs;
+	struct class_device_attribute	*class_dev_attrs;
+	struct device_attribute		*dev_attrs;
 
-	int	(*uevent)(struct class_device *dev, struct kobj_uevent_env *env);
-	int	(*dev_uevent)(struct device *dev, struct kobj_uevent_env *env);
+	int (*uevent)(struct class_device *dev, struct kobj_uevent_env *env);
+	int (*dev_uevent)(struct device *dev, struct kobj_uevent_env *env);
 
-	void	(*release)(struct class_device *dev);
-	void	(*class_release)(struct class *class);
-	void	(*dev_release)(struct device *dev);
+	void (*release)(struct class_device *dev);
+	void (*class_release)(struct class *class);
+	void (*dev_release)(struct device *dev);
 
-	int	(*suspend)(struct device *, pm_message_t state);
-	int	(*resume)(struct device *);
+	int (*suspend)(struct device *dev, pm_message_t state);
+	int (*resume)(struct device *dev);
 };
 
-extern int __must_check class_register(struct class *);
-extern void class_unregister(struct class *);
+extern int __must_check class_register(struct class *class);
+extern void class_unregister(struct class *class);
 extern int class_for_each_device(struct class *class, void *data,
 				 int (*fn)(struct device *dev, void *data));
 extern struct device *class_find_device(struct class *class, void *data,
@@ -204,27 +212,28 @@ extern struct class_device *class_find_child(struct class *class, void *data,
 
 
 struct class_attribute {
-	struct attribute	attr;
-	ssize_t (*show)(struct class *, char * buf);
-	ssize_t (*store)(struct class *, const char * buf, size_t count);
+	struct attribute attr;
+	ssize_t (*show)(struct class *class, char *buf);
+	ssize_t (*store)(struct class *class, const char *buf, size_t count);
 };
 
-#define CLASS_ATTR(_name,_mode,_show,_store)			\
-struct class_attribute class_attr_##_name = __ATTR(_name,_mode,_show,_store) 
+#define CLASS_ATTR(_name, _mode, _show, _store)			\
+struct class_attribute class_attr_##_name = __ATTR(_name, _mode, _show, _store)
 
-extern int __must_check class_create_file(struct class *,
-					const struct class_attribute *);
-extern void class_remove_file(struct class *, const struct class_attribute *);
+extern int __must_check class_create_file(struct class *class,
+					  const struct class_attribute *attr);
+extern void class_remove_file(struct class *class,
+			      const struct class_attribute *attr);
 
 struct class_device_attribute {
-	struct attribute	attr;
-	ssize_t (*show)(struct class_device *, char * buf);
-	ssize_t (*store)(struct class_device *, const char * buf, size_t count);
+	struct attribute attr;
+	ssize_t (*show)(struct class_device *, char *buf);
+	ssize_t (*store)(struct class_device *, const char *buf, size_t count);
 };
 
-#define CLASS_DEVICE_ATTR(_name,_mode,_show,_store)		\
+#define CLASS_DEVICE_ATTR(_name, _mode, _show, _store)		\
 struct class_device_attribute class_device_attr_##_name = 	\
-	__ATTR(_name,_mode,_show,_store)
+	__ATTR(_name, _mode, _show, _store)
 
 extern int __must_check class_device_create_file(struct class_device *,
 				    const struct class_device_attribute *);
@@ -257,26 +266,24 @@ struct class_device {
 	struct list_head	node;
 
 	struct kobject		kobj;
-	struct class		* class;	/* required */
-	dev_t			devt;		/* dev_t, creates the sysfs "dev" */
-	struct device		* dev;		/* not necessary, but nice to have */
-	void			* class_data;	/* class-specific data */
-	struct class_device	*parent;	/* parent of this child device, if there is one */
-	struct attribute_group  ** groups;	/* optional groups */
-
-	void	(*release)(struct class_device *dev);
-	int	(*uevent)(struct class_device *dev, struct kobj_uevent_env *env);
-	char	class_id[BUS_ID_SIZE];	/* unique to this class */
+	struct class		*class;
+	dev_t			devt;
+	struct device		*dev;
+	void			*class_data;
+	struct class_device	*parent;
+	struct attribute_group  **groups;
+
+	void (*release)(struct class_device *dev);
+	int (*uevent)(struct class_device *dev, struct kobj_uevent_env *env);
+	char class_id[BUS_ID_SIZE];
 };
 
-static inline void *
-class_get_devdata (struct class_device *dev)
+static inline void *class_get_devdata(struct class_device *dev)
 {
 	return dev->class_data;
 }
 
-static inline void
-class_set_devdata (struct class_device *dev, void *data)
+static inline void class_set_devdata(struct class_device *dev, void *data)
 {
 	dev->class_data = data;
 }
@@ -288,10 +295,10 @@ extern void class_device_initialize(struct class_device *);
 extern int __must_check class_device_add(struct class_device *);
 extern void class_device_del(struct class_device *);
 
-extern struct class_device * class_device_get(struct class_device *);
+extern struct class_device *class_device_get(struct class_device *);
 extern void class_device_put(struct class_device *);
 
-extern void class_device_remove_file(struct class_device *, 
+extern void class_device_remove_file(struct class_device *,
 				     const struct class_device_attribute *);
 extern int __must_check class_device_create_bin_file(struct class_device *,
 					struct bin_attribute *);
@@ -318,7 +325,7 @@ extern struct class_device *class_device_create(struct class *cls,
 						dev_t devt,
 						struct device *device,
 						const char *fmt, ...)
-					__attribute__((format(printf,5,6)));
+					__attribute__((format(printf, 5, 6)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
 /*
@@ -335,8 +342,8 @@ struct device_type {
 	struct attribute_group **groups;
 	int (*uevent)(struct device *dev, struct kobj_uevent_env *env);
 	void (*release)(struct device *dev);
-	int (*suspend)(struct device * dev, pm_message_t state);
-	int (*resume)(struct device * dev);
+	int (*suspend)(struct device *dev, pm_message_t state);
+	int (*resume)(struct device *dev);
 };
 
 /* interface for exporting device attributes */
@@ -348,18 +355,19 @@ struct device_attribute {
 			 const char *buf, size_t count);
 };
 
-#define DEVICE_ATTR(_name,_mode,_show,_store) \
-struct device_attribute dev_attr_##_name = __ATTR(_name,_mode,_show,_store)
+#define DEVICE_ATTR(_name, _mode, _show, _store) \
+struct device_attribute dev_attr_##_name = __ATTR(_name, _mode, _show, _store)
 
 extern int __must_check device_create_file(struct device *device,
-					struct device_attribute * entry);
-extern void device_remove_file(struct device * dev, struct device_attribute * attr);
+					   struct device_attribute *entry);
+extern void device_remove_file(struct device *dev,
+			       struct device_attribute *attr);
 extern int __must_check device_create_bin_file(struct device *dev,
 					       struct bin_attribute *attr);
 extern void device_remove_bin_file(struct device *dev,
 				   struct bin_attribute *attr);
 extern int device_schedule_callback_owner(struct device *dev,
-		void (*func)(struct device *), struct module *owner);
+		void (*func)(struct device *dev), struct module *owner);
 
 /* This is a macro to avoid include problems with THIS_MODULE */
 #define device_schedule_callback(dev, func)			\
@@ -370,21 +378,21 @@ typedef void (*dr_release_t)(struct device *dev, void *res);
 typedef int (*dr_match_t)(struct device *dev, void *res, void *match_data);
 
 #ifdef CONFIG_DEBUG_DEVRES
-extern void * __devres_alloc(dr_release_t release, size_t size, gfp_t gfp,
+extern void *__devres_alloc(dr_release_t release, size_t size, gfp_t gfp,
 			     const char *name);
 #define devres_alloc(release, size, gfp) \
 	__devres_alloc(release, size, gfp, #release)
 #else
-extern void * devres_alloc(dr_release_t release, size_t size, gfp_t gfp);
+extern void *devres_alloc(dr_release_t release, size_t size, gfp_t gfp);
 #endif
 extern void devres_free(void *res);
 extern void devres_add(struct device *dev, void *res);
-extern void * devres_find(struct device *dev, dr_release_t release,
-			  dr_match_t match, void *match_data);
-extern void * devres_get(struct device *dev, void *new_res,
+extern void *devres_find(struct device *dev, dr_release_t release,
 			 dr_match_t match, void *match_data);
-extern void * devres_remove(struct device *dev, dr_release_t release,
-			    dr_match_t match, void *match_data);
+extern void *devres_get(struct device *dev, void *new_res,
+			dr_match_t match, void *match_data);
+extern void *devres_remove(struct device *dev, dr_release_t release,
+			   dr_match_t match, void *match_data);
 extern int devres_destroy(struct device *dev, dr_release_t release,
 			  dr_match_t match, void *match_data);
 
@@ -401,7 +409,7 @@ extern void devm_kfree(struct device *dev, void *p);
 
 struct device {
 	struct klist		klist_children;
-	struct klist_node	knode_parent;		/* node in sibling list */
+	struct klist_node	knode_parent;	/* node in sibling list */
 	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device		*parent;
@@ -416,7 +424,7 @@ struct device {
 					 * its driver.
 					 */
 
-	struct bus_type	* bus;		/* type of bus device is on */
+	struct bus_type	*bus;		/* type of bus device is on */
 	struct device_driver *driver;	/* which driver has allocated this
 					   device */
 	void		*driver_data;	/* data private to the driver */
@@ -447,10 +455,10 @@ struct device {
 	/* class_device migration path */
 	struct list_head	node;
 	struct class		*class;
-	dev_t			devt;		/* dev_t, creates the sysfs "dev" */
+	dev_t			devt;	/* dev_t, creates the sysfs "dev" */
 	struct attribute_group	**groups;	/* optional groups */
 
-	void	(*release)(struct device * dev);
+	void	(*release)(struct device *dev);
 };
 
 #ifdef CONFIG_NUMA
@@ -472,14 +480,12 @@ static inline void set_dev_node(struct device *dev, int node)
 }
 #endif
 
-static inline void *
-dev_get_drvdata (struct device *dev)
+static inline void *dev_get_drvdata(struct device *dev)
 {
 	return dev->driver_data;
 }
 
-static inline void
-dev_set_drvdata (struct device *dev, void *data)
+static inline void dev_set_drvdata(struct device *dev, void *data)
 {
 	dev->driver_data = data;
 }
@@ -494,15 +500,15 @@ void driver_init(void);
 /*
  * High level routines for use by the bus drivers
  */
-extern int __must_check device_register(struct device * dev);
-extern void device_unregister(struct device * dev);
-extern void device_initialize(struct device * dev);
-extern int __must_check device_add(struct device * dev);
-extern void device_del(struct device * dev);
-extern int device_for_each_child(struct device *, void *,
-		     int (*fn)(struct device *, void *));
-extern struct device *device_find_child(struct device *, void *data,
-					int (*match)(struct device *, void *));
+extern int __must_check device_register(struct device *dev);
+extern void device_unregister(struct device *dev);
+extern void device_initialize(struct device *dev);
+extern int __must_check device_add(struct device *dev);
+extern void device_del(struct device *dev);
+extern int device_for_each_child(struct device *dev, void *data,
+		     int (*fn)(struct device *dev, void *data));
+extern struct device *device_find_child(struct device *dev, void *data,
+				int (*match)(struct device *dev, void *data));
 extern int device_rename(struct device *dev, char *new_name);
 extern int device_move(struct device *dev, struct device *new_parent);
 
@@ -511,8 +517,8 @@ extern int device_move(struct device *dev, struct device *new_parent);
  * for information on use.
  */
 extern int __must_check device_bind_driver(struct device *dev);
-extern void device_release_driver(struct device * dev);
-extern int  __must_check device_attach(struct device * dev);
+extern void device_release_driver(struct device *dev);
+extern int  __must_check device_attach(struct device *dev);
 extern int __must_check driver_attach(struct device_driver *drv);
 extern int __must_check device_reprobe(struct device *dev);
 
@@ -521,7 +527,7 @@ extern int __must_check device_reprobe(struct device *dev);
  */
 extern struct device *device_create(struct class *cls, struct device *parent,
 				    dev_t devt, const char *fmt, ...)
-				    __attribute__((format(printf,4,5)));
+				    __attribute__((format(printf, 4, 5)));
 extern void device_destroy(struct class *cls, dev_t devt);
 #ifdef CONFIG_PM_SLEEP
 extern void destroy_suspended_device(struct class *cls, dev_t devt);
@@ -538,17 +544,17 @@ static inline void destroy_suspended_device(struct class *cls, dev_t devt)
  * know about.
  */
 /* Notify platform of device discovery */
-extern int (*platform_notify)(struct device * dev);
+extern int (*platform_notify)(struct device *dev);
 
-extern int (*platform_notify_remove)(struct device * dev);
+extern int (*platform_notify_remove)(struct device *dev);
 
 
 /**
  * get_device - atomically increment the reference count for the device.
  *
  */
-extern struct device * get_device(struct device * dev);
-extern void put_device(struct device * dev);
+extern struct device *get_device(struct device *dev);
+extern void put_device(struct device *dev);
 
 
 /* drivers/base/power/shutdown.c */
@@ -560,7 +566,8 @@ extern void sysdev_shutdown(void);
 /* debugging and troubleshooting/diagnostic helpers. */
 extern const char *dev_driver_string(struct device *dev);
 #define dev_printk(level, dev, format, arg...)	\
-	printk(level "%s %s: " format , dev_driver_string(dev) , (dev)->bus_id , ## arg)
+	printk(level "%s %s: " format , dev_driver_string(dev) , \
+	       (dev)->bus_id , ## arg)
 
 #define dev_emerg(dev, format, arg...)		\
 	dev_printk(KERN_EMERG , dev , format , ## arg)
@@ -582,7 +589,7 @@ extern const char *dev_driver_string(struct device *dev);
 	dev_printk(KERN_DEBUG , dev , format , ## arg)
 #else
 static inline int __attribute__ ((format (printf, 2, 3)))
-dev_dbg(struct device * dev, const char * fmt, ...)
+dev_dbg(struct device *dev, const char *fmt, ...)
 {
 	return 0;
 }
@@ -592,7 +599,7 @@ dev_dbg(struct device * dev, const char * fmt, ...)
 #define dev_vdbg	dev_dbg
 #else
 static inline int __attribute__ ((format (printf, 2, 3)))
-dev_vdbg(struct device * dev, const char * fmt, ...)
+dev_vdbg(struct device *dev, const char *fmt, ...)
 {
 	return 0;
 }
-- 
cgit v1.2.3


From 79a6ee42fd81be9abc6bdab08f932875924b26a5 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 24 Jan 2008 21:27:06 -0800
Subject: Kobject: fix coding style issues in kobject.h

Finally clean up the odd spaces and other mess in kobject.h

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 67 +++++++++++++++++++++++++------------------------
 1 file changed, 34 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 4adbe1d83081..caa3f411f15d 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -3,15 +3,14 @@
  *
  * Copyright (c) 2002-2003 Patrick Mochel
  * Copyright (c) 2002-2003 Open Source Development Labs
- * Copyright (c) 2006-2007 Greg Kroah-Hartman <greg@kroah.com>
- * Copyright (c) 2006-2007 Novell Inc.
+ * Copyright (c) 2006-2008 Greg Kroah-Hartman <greg@kroah.com>
+ * Copyright (c) 2006-2008 Novell Inc.
  *
  * This file is released under the GPLv2.
  *
- * 
  * Please read Documentation/kobject.txt before using the kobject
  * interface, ESPECIALLY the parts about reference counts and object
- * destructors. 
+ * destructors.
  */
 
 #ifndef _KOBJECT_H_
@@ -64,20 +63,20 @@ struct kobject {
 	const char		*name;
 	struct kref		kref;
 	struct list_head	entry;
-	struct kobject		* parent;
-	struct kset		* kset;
-	struct kobj_type	* ktype;
-	struct sysfs_dirent	* sd;
+	struct kobject		*parent;
+	struct kset		*kset;
+	struct kobj_type	*ktype;
+	struct sysfs_dirent	*sd;
 	unsigned int state_initialized:1;
 	unsigned int state_in_sysfs:1;
 	unsigned int state_add_uevent_sent:1;
 	unsigned int state_remove_uevent_sent:1;
 };
 
-extern int kobject_set_name(struct kobject *, const char *, ...)
-	__attribute__((format(printf,2,3)));
+extern int kobject_set_name(struct kobject *kobj, const char *name, ...)
+			    __attribute__((format(printf, 2, 3)));
 
-static inline const char * kobject_name(const struct kobject * kobj)
+static inline const char *kobject_name(const struct kobject *kobj)
 {
 	return kobj->name;
 }
@@ -91,7 +90,7 @@ extern int __must_check kobject_init_and_add(struct kobject *kobj,
 					     struct kobject *parent,
 					     const char *fmt, ...);
 
-extern void kobject_del(struct kobject *);
+extern void kobject_del(struct kobject *kobj);
 
 extern struct kobject * __must_check kobject_create(void);
 extern struct kobject * __must_check kobject_create_and_add(const char *name,
@@ -100,15 +99,15 @@ extern struct kobject * __must_check kobject_create_and_add(const char *name,
 extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 extern int __must_check kobject_move(struct kobject *, struct kobject *);
 
-extern struct kobject * kobject_get(struct kobject *);
-extern void kobject_put(struct kobject *);
+extern struct kobject *kobject_get(struct kobject *kobj);
+extern void kobject_put(struct kobject *kobj);
 
-extern char * kobject_get_path(struct kobject *, gfp_t);
+extern char *kobject_get_path(struct kobject *kobj, gfp_t flag);
 
 struct kobj_type {
-	void (*release)(struct kobject *);
-	struct sysfs_ops	* sysfs_ops;
-	struct attribute	** default_attrs;
+	void (*release)(struct kobject *kobj);
+	struct sysfs_ops *sysfs_ops;
+	struct attribute **default_attrs;
 };
 
 struct kobj_uevent_env {
@@ -153,30 +152,30 @@ extern struct sysfs_ops kobj_sysfs_ops;
  * desired.
  */
 struct kset {
-	struct list_head	list;
-	spinlock_t		list_lock;
-	struct kobject		kobj;
-	struct kset_uevent_ops	*uevent_ops;
+	struct list_head list;
+	spinlock_t list_lock;
+	struct kobject kobj;
+	struct kset_uevent_ops *uevent_ops;
 };
 
-extern void kset_init(struct kset * k);
-extern int __must_check kset_register(struct kset * k);
-extern void kset_unregister(struct kset * k);
+extern void kset_init(struct kset *kset);
+extern int __must_check kset_register(struct kset *kset);
+extern void kset_unregister(struct kset *kset);
 extern struct kset * __must_check kset_create_and_add(const char *name,
 						struct kset_uevent_ops *u,
 						struct kobject *parent_kobj);
 
-static inline struct kset * to_kset(struct kobject * kobj)
+static inline struct kset *to_kset(struct kobject *kobj)
 {
-	return kobj ? container_of(kobj,struct kset,kobj) : NULL;
+	return kobj ? container_of(kobj, struct kset, kobj) : NULL;
 }
 
-static inline struct kset * kset_get(struct kset * k)
+static inline struct kset *kset_get(struct kset *k)
 {
 	return k ? to_kset(kobject_get(&k->kobj)) : NULL;
 }
 
-static inline void kset_put(struct kset * k)
+static inline void kset_put(struct kset *k)
 {
 	kobject_put(&k->kobj);
 }
@@ -186,7 +185,7 @@ static inline struct kobj_type *get_ktype(struct kobject *kobj)
 	return kobj->ktype;
 }
 
-extern struct kobject * kset_find_obj(struct kset *, const char *);
+extern struct kobject *kset_find_obj(struct kset *, const char *);
 
 /* The global /sys/kernel/ kobject for people to chain off of */
 extern struct kobject *kernel_kobj;
@@ -208,18 +207,20 @@ int add_uevent_var(struct kobj_uevent_env *env, const char *format, ...)
 int kobject_action_type(const char *buf, size_t count,
 			enum kobject_action *type);
 #else
-static inline int kobject_uevent(struct kobject *kobj, enum kobject_action action)
+static inline int kobject_uevent(struct kobject *kobj,
+				 enum kobject_action action)
 { return 0; }
 static inline int kobject_uevent_env(struct kobject *kobj,
 				      enum kobject_action action,
 				      char *envp[])
 { return 0; }
 
-static inline int add_uevent_var(struct kobj_uevent_env *env, const char *format, ...)
+static inline int add_uevent_var(struct kobj_uevent_env *env,
+				 const char *format, ...)
 { return 0; }
 
 static inline int kobject_action_type(const char *buf, size_t count,
-			enum kobject_action *type)
+				      enum kobject_action *type)
 { return -EINVAL; }
 #endif
 
-- 
cgit v1.2.3


From 13b54a50525a9685065684e1e11258d27dd27bdf Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Tue, 27 Nov 2007 13:50:45 +0100
Subject: [AVR32] Enable debugging only when needed

Keep track of processes being debugged (including the kernel itself)
and turn the OCD system on and off as appropriate. Since enabling
debugging turns off some optimizations in the CPU core, this fixes the
issue that enabling KProbes support or simply running a program under
gdbserver will reduce system performance significantly until the next
reboot.

The CPU performance will still be reduced for all processes while a
process is being debugged, but this is a lot better than reducing the
performance forever.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
---
 arch/avr32/kernel/Makefile      |   2 +-
 arch/avr32/kernel/kprobes.c     |   5 +-
 arch/avr32/kernel/ocd.c         | 163 ++++++++++++++++++++++++++++++++++++++++
 arch/avr32/kernel/process.c     |   5 +-
 arch/avr32/kernel/ptrace.c      |   5 +-
 include/asm-avr32/ocd.h         |   5 ++
 include/asm-avr32/ptrace.h      |  13 +++-
 include/asm-avr32/thread_info.h |   1 +
 8 files changed, 186 insertions(+), 13 deletions(-)
 create mode 100644 arch/avr32/kernel/ocd.c

(limited to 'include')

diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile
index 2d6d48f35f69..bc224a4e39fe 100644
--- a/arch/avr32/kernel/Makefile
+++ b/arch/avr32/kernel/Makefile
@@ -6,7 +6,7 @@ extra-y				:= head.o vmlinux.lds
 
 obj-$(CONFIG_SUBARCH_AVR32B)	+= entry-avr32b.o
 obj-y				+= syscall_table.o syscall-stubs.o irq.o
-obj-y				+= setup.o traps.o semaphore.o ptrace.o
+obj-y				+= setup.o traps.o semaphore.o ocd.o ptrace.o
 obj-y				+= signal.o sys_avr32.o process.o time.o
 obj-y				+= init_task.o switch_to.o cpu.o
 obj-$(CONFIG_MODULES)		+= module.o avr32_ksyms.o
diff --git a/arch/avr32/kernel/kprobes.c b/arch/avr32/kernel/kprobes.c
index 799ba89b07a8..f820e9f25520 100644
--- a/arch/avr32/kernel/kprobes.c
+++ b/arch/avr32/kernel/kprobes.c
@@ -48,6 +48,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 void __kprobes arch_arm_kprobe(struct kprobe *p)
 {
 	pr_debug("arming kprobe at %p\n", p->addr);
+	ocd_enable(NULL);
 	*p->addr = BREAKPOINT_INSTRUCTION;
 	flush_icache_range((unsigned long)p->addr,
 			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
@@ -56,6 +57,7 @@ void __kprobes arch_arm_kprobe(struct kprobe *p)
 void __kprobes arch_disarm_kprobe(struct kprobe *p)
 {
 	pr_debug("disarming kprobe at %p\n", p->addr);
+	ocd_disable(NULL);
 	*p->addr = p->opcode;
 	flush_icache_range((unsigned long)p->addr,
 			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
@@ -260,9 +262,6 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 
 int __init arch_init_kprobes(void)
 {
-	printk("KPROBES: Enabling monitor mode (MM|DBE)...\n");
-	ocd_write(DC, (1 << OCD_DC_MM_BIT) | (1 << OCD_DC_DBE_BIT));
-
 	/* TODO: Register kretprobe trampoline */
 	return 0;
 }
diff --git a/arch/avr32/kernel/ocd.c b/arch/avr32/kernel/ocd.c
new file mode 100644
index 000000000000..c4f023294d75
--- /dev/null
+++ b/arch/avr32/kernel/ocd.c
@@ -0,0 +1,163 @@
+/*
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+
+#include <asm/ocd.h>
+
+static long ocd_count;
+static spinlock_t ocd_lock;
+
+/**
+ * ocd_enable - enable on-chip debugging
+ * @child: task to be debugged
+ *
+ * If @child is non-NULL, ocd_enable() first checks if debugging has
+ * already been enabled for @child, and if it has, does nothing.
+ *
+ * If @child is NULL (e.g. when debugging the kernel), or debugging
+ * has not already been enabled for it, ocd_enable() increments the
+ * reference count and enables the debugging hardware.
+ */
+void ocd_enable(struct task_struct *child)
+{
+	u32 dc;
+
+	if (child)
+		pr_debug("ocd_enable: child=%s [%u]\n",
+				child->comm, child->pid);
+	else
+		pr_debug("ocd_enable (no child)\n");
+
+	if (!child || !test_and_set_tsk_thread_flag(child, TIF_DEBUG)) {
+		spin_lock(&ocd_lock);
+		ocd_count++;
+		dc = ocd_read(DC);
+		dc |= (1 << OCD_DC_MM_BIT) | (1 << OCD_DC_DBE_BIT);
+		ocd_write(DC, dc);
+		spin_unlock(&ocd_lock);
+	}
+}
+
+/**
+ * ocd_disable - disable on-chip debugging
+ * @child: task that was being debugged, but isn't anymore
+ *
+ * If @child is non-NULL, ocd_disable() checks if debugging is enabled
+ * for @child, and if it isn't, does nothing.
+ *
+ * If @child is NULL (e.g. when debugging the kernel), or debugging is
+ * enabled, ocd_disable() decrements the reference count, and if it
+ * reaches zero, disables the debugging hardware.
+ */
+void ocd_disable(struct task_struct *child)
+{
+	u32 dc;
+
+	if (!child)
+		pr_debug("ocd_disable (no child)\n");
+	else if (test_tsk_thread_flag(child, TIF_DEBUG))
+		pr_debug("ocd_disable: child=%s [%u]\n",
+				child->comm, child->pid);
+
+	if (!child || test_and_clear_tsk_thread_flag(child, TIF_DEBUG)) {
+		spin_lock(&ocd_lock);
+		ocd_count--;
+
+		WARN_ON(ocd_count < 0);
+
+		if (ocd_count <= 0) {
+			dc = ocd_read(DC);
+			dc &= ~((1 << OCD_DC_MM_BIT) | (1 << OCD_DC_DBE_BIT));
+			ocd_write(DC, dc);
+		}
+		spin_unlock(&ocd_lock);
+	}
+}
+
+#ifdef CONFIG_DEBUG_FS
+#include <linux/debugfs.h>
+#include <linux/module.h>
+
+static struct dentry *ocd_debugfs_root;
+static struct dentry *ocd_debugfs_DC;
+static struct dentry *ocd_debugfs_DS;
+static struct dentry *ocd_debugfs_count;
+
+static u64 ocd_DC_get(void *data)
+{
+	return ocd_read(DC);
+}
+static void ocd_DC_set(void *data, u64 val)
+{
+	ocd_write(DC, val);
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_DC, ocd_DC_get, ocd_DC_set, "0x%08llx\n");
+
+static u64 ocd_DS_get(void *data)
+{
+	return ocd_read(DS);
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_DS, ocd_DS_get, NULL, "0x%08llx\n");
+
+static u64 ocd_count_get(void *data)
+{
+	return ocd_count;
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_count, ocd_count_get, NULL, "%lld\n");
+
+static void ocd_debugfs_init(void)
+{
+	struct dentry *root;
+
+	root = debugfs_create_dir("ocd", NULL);
+	if (IS_ERR(root) || !root)
+		goto err_root;
+	ocd_debugfs_root = root;
+
+	ocd_debugfs_DC = debugfs_create_file("DC", S_IRUSR | S_IWUSR,
+				root, NULL, &fops_DC);
+	if (!ocd_debugfs_DC)
+		goto err_DC;
+
+	ocd_debugfs_DS = debugfs_create_file("DS", S_IRUSR, root,
+				NULL, &fops_DS);
+	if (!ocd_debugfs_DS)
+		goto err_DS;
+
+	ocd_debugfs_count = debugfs_create_file("count", S_IRUSR, root,
+				NULL, &fops_count);
+	if (!ocd_debugfs_count)
+		goto err_count;
+
+	return;
+
+err_count:
+	debugfs_remove(ocd_debugfs_DS);
+err_DS:
+	debugfs_remove(ocd_debugfs_DC);
+err_DC:
+	debugfs_remove(ocd_debugfs_root);
+err_root:
+	printk(KERN_WARNING "OCD: Failed to create debugfs entries\n");
+}
+#else
+static inline void ocd_debugfs_init(void)
+{
+
+}
+#endif
+
+static int __init ocd_init(void)
+{
+	spin_lock_init(&ocd_lock);
+	ocd_debugfs_init();
+	return 0;
+}
+arch_initcall(ocd_init);
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index 9d6dac8af7a2..eaaa69bbdc38 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -103,7 +103,7 @@ EXPORT_SYMBOL(kernel_thread);
  */
 void exit_thread(void)
 {
-	/* nothing to do */
+	ocd_disable(current);
 }
 
 void flush_thread(void)
@@ -345,6 +345,9 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 	p->thread.cpu_context.ksp = (unsigned long)childregs;
 	p->thread.cpu_context.pc = (unsigned long)ret_from_fork;
 
+	if ((clone_flags & CLONE_PTRACE) && test_thread_flag(TIF_DEBUG))
+		ocd_enable(p);
+
 	return 0;
 }
 
diff --git a/arch/avr32/kernel/ptrace.c b/arch/avr32/kernel/ptrace.c
index 002369e44093..1fed38fcf594 100644
--- a/arch/avr32/kernel/ptrace.c
+++ b/arch/avr32/kernel/ptrace.c
@@ -58,6 +58,7 @@ void ptrace_disable(struct task_struct *child)
 {
 	clear_tsk_thread_flag(child, TIF_SINGLE_STEP);
 	clear_tsk_thread_flag(child, TIF_BREAKPOINT);
+	ocd_disable(child);
 }
 
 /*
@@ -144,10 +145,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 {
 	int ret;
 
-	pr_debug("ptrace: Enabling monitor mode...\n");
-	ocd_write(DC, ocd_read(DC) | (1 << OCD_DC_MM_BIT)
-			| (1 << OCD_DC_DBE_BIT));
-
 	switch (request) {
 	/* Read the word at location addr in the child process */
 	case PTRACE_PEEKTEXT:
diff --git a/include/asm-avr32/ocd.h b/include/asm-avr32/ocd.h
index 996405e0393f..6bef09490235 100644
--- a/include/asm-avr32/ocd.h
+++ b/include/asm-avr32/ocd.h
@@ -533,6 +533,11 @@ static inline void __ocd_write(unsigned int reg, unsigned long value)
 #define ocd_read(reg)			__ocd_read(OCD_##reg)
 #define ocd_write(reg, value)		__ocd_write(OCD_##reg, value)
 
+struct task_struct;
+
+void ocd_enable(struct task_struct *child);
+void ocd_disable(struct task_struct *child);
+
 #endif /* !__ASSEMBLER__ */
 
 #endif /* __ASM_AVR32_OCD_H */
diff --git a/include/asm-avr32/ptrace.h b/include/asm-avr32/ptrace.h
index 8c5dba5e33df..9e2d44f4e0fe 100644
--- a/include/asm-avr32/ptrace.h
+++ b/include/asm-avr32/ptrace.h
@@ -121,7 +121,15 @@ struct pt_regs {
 };
 
 #ifdef __KERNEL__
-# define user_mode(regs) (((regs)->sr & MODE_MASK) == MODE_USER)
+
+#include <asm/ocd.h>
+
+#define arch_ptrace_attach(child)       ocd_enable(child)
+
+#define user_mode(regs)                 (((regs)->sr & MODE_MASK) == MODE_USER)
+#define instruction_pointer(regs)       ((regs)->pc)
+#define profile_pc(regs)                instruction_pointer(regs)
+
 extern void show_regs (struct pt_regs *);
 
 static __inline__ int valid_user_regs(struct pt_regs *regs)
@@ -141,9 +149,6 @@ static __inline__ int valid_user_regs(struct pt_regs *regs)
 	return 0;
 }
 
-#define instruction_pointer(regs) ((regs)->pc)
-
-#define profile_pc(regs) instruction_pointer(regs)
 
 #endif /* __KERNEL__ */
 
diff --git a/include/asm-avr32/thread_info.h b/include/asm-avr32/thread_info.h
index 184b574289b4..07049f6c0d41 100644
--- a/include/asm-avr32/thread_info.h
+++ b/include/asm-avr32/thread_info.h
@@ -88,6 +88,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_MEMDIE		6
 #define TIF_RESTORE_SIGMASK	7	/* restore signal mask in do_signal */
 #define TIF_CPU_GOING_TO_SLEEP	8	/* CPU is entering sleep 0 mode */
+#define TIF_DEBUG		30	/* debugging enabled */
 #define TIF_USERSPACE		31      /* true if FS sets userspace */
 
 #define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
-- 
cgit v1.2.3


From 281ef58ccf62eaa6c4e4b7e4c0a3ee6b52e84e5b Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Fri, 7 Dec 2007 10:21:02 +0100
Subject: [AVR32] Provide more CPU information in /proc/cpuinfo and dmesg

Add the following fields to /proc/cpuinfo:
  * chip type and revision (from the JTAG chip id)
  * cpu MHz (from clk_get_rate())
  * features (from the CONFIG0 register)

Also rename "cpu family" to "cpu arch" and "cpu type" to "cpu core" to
remove some ambiguity.

Show chip type and revision at bootup, and clarify that the other
kinds of IDs that we're already printing are for the cpu core and
architecture. Rename "AP7000" to "AP7" since that's the name of the
core.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
---
 arch/avr32/kernel/cpu.c       | 94 +++++++++++++++++++++++++++++++------------
 include/asm-avr32/processor.h | 14 +++++++
 2 files changed, 82 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/avr32/kernel/cpu.c b/arch/avr32/kernel/cpu.c
index 2714cf6452b5..14610019216e 100644
--- a/arch/avr32/kernel/cpu.c
+++ b/arch/avr32/kernel/cpu.c
@@ -13,6 +13,7 @@
 #include <linux/percpu.h>
 #include <linux/param.h>
 #include <linux/errno.h>
+#include <linux/clk.h>
 
 #include <asm/setup.h>
 #include <asm/sysreg.h>
@@ -187,9 +188,20 @@ static int __init topology_init(void)
 
 subsys_initcall(topology_init);
 
+struct chip_id_map {
+	u16	mid;
+	u16	pn;
+	const char *name;
+};
+
+static const struct chip_id_map chip_names[] = {
+	{ .mid = 0x1f, .pn = 0x1e82, .name = "AT32AP700x" },
+};
+#define NR_CHIP_NAMES ARRAY_SIZE(chip_names)
+
 static const char *cpu_names[] = {
 	"Morgan",
-	"AP7000",
+	"AP7",
 };
 #define NR_CPU_NAMES ARRAY_SIZE(cpu_names)
 
@@ -206,12 +218,32 @@ static const char *mmu_types[] = {
 	"MPU"
 };
 
+static const char *cpu_feature_flags[] = {
+	"rmw", "dsp", "simd", "ocd", "perfctr", "java", "fpu",
+};
+
+static const char *get_chip_name(struct avr32_cpuinfo *cpu)
+{
+	unsigned int i;
+	unsigned int mid = avr32_get_manufacturer_id(cpu);
+	unsigned int pn = avr32_get_product_number(cpu);
+
+	for (i = 0; i < NR_CHIP_NAMES; i++) {
+		if (chip_names[i].mid == mid && chip_names[i].pn == pn)
+			return chip_names[i].name;
+	}
+
+	return "(unknown)";
+}
+
 void __init setup_processor(void)
 {
 	unsigned long config0, config1;
 	unsigned long features;
 	unsigned cpu_id, cpu_rev, arch_id, arch_rev, mmu_type;
+	unsigned device_id;
 	unsigned tmp;
+	unsigned i;
 
 	config0 = sysreg_read(CONFIG0);
 	config1 = sysreg_read(CONFIG1);
@@ -221,11 +253,14 @@ void __init setup_processor(void)
 	arch_rev = SYSREG_BFEXT(AR, config0);
 	mmu_type = SYSREG_BFEXT(MMUT, config0);
 
+	device_id = ocd_read(DID);
+
 	boot_cpu_data.arch_type = arch_id;
 	boot_cpu_data.cpu_type = cpu_id;
 	boot_cpu_data.arch_revision = arch_rev;
 	boot_cpu_data.cpu_revision = cpu_rev;
 	boot_cpu_data.tlb_config = mmu_type;
+	boot_cpu_data.device_id = device_id;
 
 	tmp = SYSREG_BFEXT(ILSZ, config1);
 	if (tmp) {
@@ -247,41 +282,34 @@ void __init setup_processor(void)
 		return;
 	}
 
-	printk ("CPU: %s [%02x] revision %d (%s revision %d)\n",
+	printk ("CPU: %s chip revision %c\n", get_chip_name(&boot_cpu_data),
+			avr32_get_chip_revision(&boot_cpu_data) + 'A');
+	printk ("CPU: %s [%02x] core revision %d (%s arch revision %d)\n",
 		cpu_names[cpu_id], cpu_id, cpu_rev,
 		arch_names[arch_id], arch_rev);
 	printk ("CPU: MMU configuration: %s\n", mmu_types[mmu_type]);
 
 	printk ("CPU: features:");
 	features = 0;
-	if (config0 & SYSREG_BIT(CONFIG0_R)) {
+	if (config0 & SYSREG_BIT(CONFIG0_R))
 		features |= AVR32_FEATURE_RMW;
-		printk(" rmw");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_D)) {
+	if (config0 & SYSREG_BIT(CONFIG0_D))
 		features |= AVR32_FEATURE_DSP;
-		printk(" dsp");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_S)) {
+	if (config0 & SYSREG_BIT(CONFIG0_S))
 		features |= AVR32_FEATURE_SIMD;
-		printk(" simd");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_O)) {
+	if (config0 & SYSREG_BIT(CONFIG0_O))
 		features |= AVR32_FEATURE_OCD;
-		printk(" ocd");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_P)) {
+	if (config0 & SYSREG_BIT(CONFIG0_P))
 		features |= AVR32_FEATURE_PCTR;
-		printk(" perfctr");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_J)) {
+	if (config0 & SYSREG_BIT(CONFIG0_J))
 		features |= AVR32_FEATURE_JAVA;
-		printk(" java");
-	}
-	if (config0 & SYSREG_BIT(CONFIG0_F)) {
+	if (config0 & SYSREG_BIT(CONFIG0_F))
 		features |= AVR32_FEATURE_FPU;
-		printk(" fpu");
-	}
+
+	for (i = 0; i < ARRAY_SIZE(cpu_feature_flags); i++)
+		if (features & (1 << i))
+			printk(" %s", cpu_feature_flags[i]);
+
 	printk("\n");
 	boot_cpu_data.features = features;
 }
@@ -291,6 +319,8 @@ static int c_show(struct seq_file *m, void *v)
 {
 	unsigned int icache_size, dcache_size;
 	unsigned int cpu = smp_processor_id();
+	unsigned int freq;
+	unsigned int i;
 
 	icache_size = boot_cpu_data.icache.ways *
 		boot_cpu_data.icache.sets *
@@ -301,15 +331,21 @@ static int c_show(struct seq_file *m, void *v)
 
 	seq_printf(m, "processor\t: %d\n", cpu);
 
+	seq_printf(m, "chip type\t: %s revision %c\n",
+			get_chip_name(&boot_cpu_data),
+			avr32_get_chip_revision(&boot_cpu_data) + 'A');
 	if (boot_cpu_data.arch_type < NR_ARCH_NAMES)
-		seq_printf(m, "cpu family\t: %s revision %d\n",
+		seq_printf(m, "cpu arch\t: %s revision %d\n",
 			   arch_names[boot_cpu_data.arch_type],
 			   boot_cpu_data.arch_revision);
 	if (boot_cpu_data.cpu_type < NR_CPU_NAMES)
-		seq_printf(m, "cpu type\t: %s revision %d\n",
+		seq_printf(m, "cpu core\t: %s revision %d\n",
 			   cpu_names[boot_cpu_data.cpu_type],
 			   boot_cpu_data.cpu_revision);
 
+	freq = (clk_get_rate(boot_cpu_data.clk) + 500) / 1000;
+	seq_printf(m, "cpu MHz\t\t: %u.%03u\n", freq / 1000, freq % 1000);
+
 	seq_printf(m, "i-cache\t\t: %dK (%u ways x %u sets x %u)\n",
 		   icache_size >> 10,
 		   boot_cpu_data.icache.ways,
@@ -320,7 +356,13 @@ static int c_show(struct seq_file *m, void *v)
 		   boot_cpu_data.dcache.ways,
 		   boot_cpu_data.dcache.sets,
 		   boot_cpu_data.dcache.linesz);
-	seq_printf(m, "bogomips\t: %lu.%02lu\n",
+
+	seq_printf(m, "features\t:");
+	for (i = 0; i < ARRAY_SIZE(cpu_feature_flags); i++)
+		if (boot_cpu_data.features & (1 << i))
+			seq_printf(m, " %s", cpu_feature_flags[i]);
+
+	seq_printf(m, "\nbogomips\t: %lu.%02lu\n",
 		   boot_cpu_data.loops_per_jiffy / (500000/HZ),
 		   (boot_cpu_data.loops_per_jiffy / (5000/HZ)) % 100);
 
diff --git a/include/asm-avr32/processor.h b/include/asm-avr32/processor.h
index a52576b25afe..4212551c1cd9 100644
--- a/include/asm-avr32/processor.h
+++ b/include/asm-avr32/processor.h
@@ -57,11 +57,25 @@ struct avr32_cpuinfo {
 	unsigned short cpu_revision;
 	enum tlb_config tlb_config;
 	unsigned long features;
+	u32 device_id;
 
 	struct cache_info icache;
 	struct cache_info dcache;
 };
 
+static inline unsigned int avr32_get_manufacturer_id(struct avr32_cpuinfo *cpu)
+{
+	return (cpu->device_id >> 1) & 0x7f;
+}
+static inline unsigned int avr32_get_product_number(struct avr32_cpuinfo *cpu)
+{
+	return (cpu->device_id >> 12) & 0xffff;
+}
+static inline unsigned int avr32_get_chip_revision(struct avr32_cpuinfo *cpu)
+{
+	return (cpu->device_id >> 28) & 0x0f;
+}
+
 extern struct avr32_cpuinfo boot_cpu_data;
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3


From 438ff3f3cc33833206a48492e9d6674e4e82bed8 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 29 Oct 2007 15:28:07 +0100
Subject: [AVR32] Add support for AT32AP7001 and AT32AP7002

These are derivatives of the AT32AP7000 chip, which means that most of
the code stays the same. Rename a few files, functions, definitions
and config symbols to reflect that they apply to all AP700x chips, and
exclude some platform devices from chips where they aren't present.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
---
 arch/avr32/Kconfig                         |   21 +-
 arch/avr32/Makefile                        |    2 +-
 arch/avr32/boards/atngw100/setup.c         |    2 +-
 arch/avr32/configs/atngw100_defconfig      |    9 +-
 arch/avr32/configs/atstk1002_defconfig     |    8 +-
 arch/avr32/mach-at32ap/Kconfig             |   12 +-
 arch/avr32/mach-at32ap/Makefile            |    4 +-
 arch/avr32/mach-at32ap/at32ap7000.c        | 1730 ---------------------------
 arch/avr32/mach-at32ap/at32ap700x.c        | 1742 ++++++++++++++++++++++++++++
 drivers/watchdog/Kconfig                   |    2 +-
 include/asm-avr32/arch-at32ap/at32ap7000.h |   35 -
 include/asm-avr32/arch-at32ap/at32ap700x.h |   35 +
 include/asm-avr32/arch-at32ap/cpu.h        |    2 +-
 include/asm-avr32/arch-at32ap/io.h         |    4 +-
 14 files changed, 1819 insertions(+), 1789 deletions(-)
 delete mode 100644 arch/avr32/mach-at32ap/at32ap7000.c
 create mode 100644 arch/avr32/mach-at32ap/at32ap700x.c
 delete mode 100644 include/asm-avr32/arch-at32ap/at32ap7000.h
 create mode 100644 include/asm-avr32/arch-at32ap/at32ap700x.h

(limited to 'include')

diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index 767a19cfa423..e37e6b92658d 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -84,13 +84,26 @@ config PLATFORM_AT32AP
 	select MMU
 	select PERFORMANCE_COUNTERS
 
+config CPU_AT32AP700X
+	bool
+	select PLATFORM_AT32AP
+
 choice
 	prompt "AVR32 CPU type"
 	default CPU_AT32AP7000
 
 config CPU_AT32AP7000
 	bool "AT32AP7000"
-	select PLATFORM_AT32AP
+	select CPU_AT32AP700X
+
+config CPU_AT32AP7001
+	bool "AT32AP7001"
+	select CPU_AT32AP700X
+
+config CPU_AT32AP7002
+	bool "AT32AP7002"
+	select CPU_AT32AP700X
+
 endchoice
 
 #
@@ -126,15 +139,15 @@ source "arch/avr32/mach-at32ap/Kconfig"
 
 config LOAD_ADDRESS
 	hex
-	default 0x10000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+	default 0x10000000 if LOADER_U_BOOT=y && CPU_AT32AP700X=y
 
 config ENTRY_ADDRESS
 	hex
-	default 0x90000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+	default 0x90000000 if LOADER_U_BOOT=y && CPU_AT32AP700X=y
 
 config PHYS_OFFSET
 	hex
-	default 0x10000000 if CPU_AT32AP7000=y
+	default 0x10000000 if CPU_AT32AP700X=y
 
 source "kernel/Kconfig.preempt"
 
diff --git a/arch/avr32/Makefile b/arch/avr32/Makefile
index f75d52cd2a4c..17a3529341dd 100644
--- a/arch/avr32/Makefile
+++ b/arch/avr32/Makefile
@@ -16,7 +16,7 @@ KBUILD_AFLAGS	+= -mrelax -mno-pic
 CFLAGS_MODULE	+= -mno-relax
 LDFLAGS_vmlinux	+= --relax
 
-cpuflags-$(CONFIG_CPU_AT32AP7000)	+= -mcpu=ap7000
+cpuflags-$(CONFIG_PLATFORM_AT32AP)	+= -march=ap
 
 KBUILD_CFLAGS	+= $(cpuflags-y)
 KBUILD_AFLAGS	+= $(cpuflags-y)
diff --git a/arch/avr32/boards/atngw100/setup.c b/arch/avr32/boards/atngw100/setup.c
index 52987c81d668..a398be284966 100644
--- a/arch/avr32/boards/atngw100/setup.c
+++ b/arch/avr32/boards/atngw100/setup.c
@@ -20,7 +20,7 @@
 #include <asm/io.h>
 #include <asm/setup.h>
 
-#include <asm/arch/at32ap7000.h>
+#include <asm/arch/at32ap700x.h>
 #include <asm/arch/board.h>
 #include <asm/arch/init.h>
 #include <asm/arch/portmux.h>
diff --git a/arch/avr32/configs/atngw100_defconfig b/arch/avr32/configs/atngw100_defconfig
index b799a68ffd97..374c17bcc05e 100644
--- a/arch/avr32/configs/atngw100_defconfig
+++ b/arch/avr32/configs/atngw100_defconfig
@@ -111,7 +111,10 @@ CONFIG_SUBARCH_AVR32B=y
 CONFIG_MMU=y
 CONFIG_PERFORMANCE_COUNTERS=y
 CONFIG_PLATFORM_AT32AP=y
+CONFIG_CPU_AT32AP700X=y
 CONFIG_CPU_AT32AP7000=y
+# CONFIG_CPU_AT32AP7001 is not set
+# CONFIG_CPU_AT32AP7002 is not set
 # CONFIG_BOARD_ATSTK1000 is not set
 CONFIG_BOARD_ATNGW100=y
 CONFIG_LOADER_U_BOOT=y
@@ -119,9 +122,9 @@ CONFIG_LOADER_U_BOOT=y
 #
 # Atmel AVR32 AP options
 #
-# CONFIG_AP7000_32_BIT_SMC is not set
-CONFIG_AP7000_16_BIT_SMC=y
-# CONFIG_AP7000_8_BIT_SMC is not set
+# CONFIG_AP700X_32_BIT_SMC is not set
+CONFIG_AP700X_16_BIT_SMC=y
+# CONFIG_AP700X_8_BIT_SMC is not set
 CONFIG_LOAD_ADDRESS=0x10000000
 CONFIG_ENTRY_ADDRESS=0x90000000
 CONFIG_PHYS_OFFSET=0x10000000
diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig
index 3b977fdbaa78..7614c5465cd6 100644
--- a/arch/avr32/configs/atstk1002_defconfig
+++ b/arch/avr32/configs/atstk1002_defconfig
@@ -114,6 +114,8 @@ CONFIG_MMU=y
 CONFIG_PERFORMANCE_COUNTERS=y
 CONFIG_PLATFORM_AT32AP=y
 CONFIG_CPU_AT32AP7000=y
+# CONFIG_CPU_AT32AP7001 is not set
+# CONFIG_CPU_AT32AP7002 is not set
 CONFIG_BOARD_ATSTK1002=y
 CONFIG_BOARD_ATSTK1000=y
 # CONFIG_BOARD_ATNGW100 is not set
@@ -122,9 +124,9 @@ CONFIG_LOADER_U_BOOT=y
 #
 # Atmel AVR32 AP options
 #
-# CONFIG_AP7000_32_BIT_SMC is not set
-CONFIG_AP7000_16_BIT_SMC=y
-# CONFIG_AP7000_8_BIT_SMC is not set
+# CONFIG_AP700X_32_BIT_SMC is not set
+CONFIG_AP700X_16_BIT_SMC=y
+# CONFIG_AP700X_8_BIT_SMC is not set
 CONFIG_LOAD_ADDRESS=0x10000000
 CONFIG_ENTRY_ADDRESS=0x90000000
 CONFIG_PHYS_OFFSET=0x10000000
diff --git a/arch/avr32/mach-at32ap/Kconfig b/arch/avr32/mach-at32ap/Kconfig
index eb307838457b..a7bbcc82058e 100644
--- a/arch/avr32/mach-at32ap/Kconfig
+++ b/arch/avr32/mach-at32ap/Kconfig
@@ -3,9 +3,9 @@ if PLATFORM_AT32AP
 menu "Atmel AVR32 AP options"
 
 choice
-	prompt "AT32AP7000 static memory bus width"
-	depends on CPU_AT32AP7000
-	default AP7000_16_BIT_SMC
+	prompt "AT32AP700x static memory bus width"
+	depends on CPU_AT32AP700X
+	default AP700X_16_BIT_SMC
 	help
 	  Define the width of the AP7000 external static memory interface.
 	  This is used to determine how to mangle the address and/or data
@@ -15,13 +15,13 @@ choice
 	  width for all chip selects, excluding the flash (which is using
 	  raw access and is thus not affected by any of this.)
 
-config AP7000_32_BIT_SMC
+config AP700X_32_BIT_SMC
 	bool "32 bit"
 
-config AP7000_16_BIT_SMC
+config AP700X_16_BIT_SMC
 	bool "16 bit"
 
-config AP7000_8_BIT_SMC
+config AP700X_8_BIT_SMC
 	bool "8 bit"
 
 endchoice
diff --git a/arch/avr32/mach-at32ap/Makefile b/arch/avr32/mach-at32ap/Makefile
index a8b445046e3e..5e9f8217befc 100644
--- a/arch/avr32/mach-at32ap/Makefile
+++ b/arch/avr32/mach-at32ap/Makefile
@@ -1,4 +1,4 @@
 obj-y				+= at32ap.o clock.o intc.o extint.o pio.o hsmc.o
-obj-$(CONFIG_CPU_AT32AP7000)	+= at32ap7000.o
-obj-$(CONFIG_CPU_AT32AP7000)	+= time-tc.o
+obj-$(CONFIG_CPU_AT32AP700X)	+= at32ap700x.o
+obj-$(CONFIG_CPU_AT32AP700X)	+= time-tc.o
 obj-$(CONFIG_CPU_FREQ_AT32AP)	+= cpufreq.o
diff --git a/arch/avr32/mach-at32ap/at32ap7000.c b/arch/avr32/mach-at32ap/at32ap7000.c
deleted file mode 100644
index 7c4388f4f17f..000000000000
--- a/arch/avr32/mach-at32ap/at32ap7000.c
+++ /dev/null
@@ -1,1730 +0,0 @@
-/*
- * Copyright (C) 2005-2006 Atmel Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/clk.h>
-#include <linux/fb.h>
-#include <linux/init.h>
-#include <linux/platform_device.h>
-#include <linux/dma-mapping.h>
-#include <linux/spi/spi.h>
-
-#include <asm/io.h>
-
-#include <asm/arch/at32ap7000.h>
-#include <asm/arch/board.h>
-#include <asm/arch/portmux.h>
-
-#include <video/atmel_lcdc.h>
-
-#include "clock.h"
-#include "hmatrix.h"
-#include "pio.h"
-#include "pm.h"
-
-
-#define PBMEM(base)					\
-	{						\
-		.start		= base,			\
-		.end		= base + 0x3ff,		\
-		.flags		= IORESOURCE_MEM,	\
-	}
-#define IRQ(num)					\
-	{						\
-		.start		= num,			\
-		.end		= num,			\
-		.flags		= IORESOURCE_IRQ,	\
-	}
-#define NAMED_IRQ(num, _name)				\
-	{						\
-		.start		= num,			\
-		.end		= num,			\
-		.name		= _name,		\
-		.flags		= IORESOURCE_IRQ,	\
-	}
-
-/* REVISIT these assume *every* device supports DMA, but several
- * don't ... tc, smc, pio, rtc, watchdog, pwm, ps2, and more.
- */
-#define DEFINE_DEV(_name, _id)					\
-static u64 _name##_id##_dma_mask = DMA_32BIT_MASK;		\
-static struct platform_device _name##_id##_device = {		\
-	.name		= #_name,				\
-	.id		= _id,					\
-	.dev		= {					\
-		.dma_mask = &_name##_id##_dma_mask,		\
-		.coherent_dma_mask = DMA_32BIT_MASK,		\
-	},							\
-	.resource	= _name##_id##_resource,		\
-	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
-}
-#define DEFINE_DEV_DATA(_name, _id)				\
-static u64 _name##_id##_dma_mask = DMA_32BIT_MASK;		\
-static struct platform_device _name##_id##_device = {		\
-	.name		= #_name,				\
-	.id		= _id,					\
-	.dev		= {					\
-		.dma_mask = &_name##_id##_dma_mask,		\
-		.platform_data	= &_name##_id##_data,		\
-		.coherent_dma_mask = DMA_32BIT_MASK,		\
-	},							\
-	.resource	= _name##_id##_resource,		\
-	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
-}
-
-#define select_peripheral(pin, periph, flags)			\
-	at32_select_periph(GPIO_PIN_##pin, GPIO_##periph, flags)
-
-#define DEV_CLK(_name, devname, bus, _index)			\
-static struct clk devname##_##_name = {				\
-	.name		= #_name,				\
-	.dev		= &devname##_device.dev,		\
-	.parent		= &bus##_clk,				\
-	.mode		= bus##_clk_mode,			\
-	.get_rate	= bus##_clk_get_rate,			\
-	.index		= _index,				\
-}
-
-static DEFINE_SPINLOCK(pm_lock);
-
-unsigned long at32ap7000_osc_rates[3] = {
-	[0] = 32768,
-	/* FIXME: these are ATSTK1002-specific */
-	[1] = 20000000,
-	[2] = 12000000,
-};
-
-static unsigned long osc_get_rate(struct clk *clk)
-{
-	return at32ap7000_osc_rates[clk->index];
-}
-
-static unsigned long pll_get_rate(struct clk *clk, unsigned long control)
-{
-	unsigned long div, mul, rate;
-
-	if (!(control & PM_BIT(PLLEN)))
-		return 0;
-
-	div = PM_BFEXT(PLLDIV, control) + 1;
-	mul = PM_BFEXT(PLLMUL, control) + 1;
-
-	rate = clk->parent->get_rate(clk->parent);
-	rate = (rate + div / 2) / div;
-	rate *= mul;
-
-	return rate;
-}
-
-static unsigned long pll0_get_rate(struct clk *clk)
-{
-	u32 control;
-
-	control = pm_readl(PLL0);
-
-	return pll_get_rate(clk, control);
-}
-
-static unsigned long pll1_get_rate(struct clk *clk)
-{
-	u32 control;
-
-	control = pm_readl(PLL1);
-
-	return pll_get_rate(clk, control);
-}
-
-/*
- * The AT32AP7000 has five primary clock sources: One 32kHz
- * oscillator, two crystal oscillators and two PLLs.
- */
-static struct clk osc32k = {
-	.name		= "osc32k",
-	.get_rate	= osc_get_rate,
-	.users		= 1,
-	.index		= 0,
-};
-static struct clk osc0 = {
-	.name		= "osc0",
-	.get_rate	= osc_get_rate,
-	.users		= 1,
-	.index		= 1,
-};
-static struct clk osc1 = {
-	.name		= "osc1",
-	.get_rate	= osc_get_rate,
-	.index		= 2,
-};
-static struct clk pll0 = {
-	.name		= "pll0",
-	.get_rate	= pll0_get_rate,
-	.parent		= &osc0,
-};
-static struct clk pll1 = {
-	.name		= "pll1",
-	.get_rate	= pll1_get_rate,
-	.parent		= &osc0,
-};
-
-/*
- * The main clock can be either osc0 or pll0.  The boot loader may
- * have chosen one for us, so we don't really know which one until we
- * have a look at the SM.
- */
-static struct clk *main_clock;
-
-/*
- * Synchronous clocks are generated from the main clock. The clocks
- * must satisfy the constraint
- *   fCPU >= fHSB >= fPB
- * i.e. each clock must not be faster than its parent.
- */
-static unsigned long bus_clk_get_rate(struct clk *clk, unsigned int shift)
-{
-	return main_clock->get_rate(main_clock) >> shift;
-};
-
-static void cpu_clk_mode(struct clk *clk, int enabled)
-{
-	unsigned long flags;
-	u32 mask;
-
-	spin_lock_irqsave(&pm_lock, flags);
-	mask = pm_readl(CPU_MASK);
-	if (enabled)
-		mask |= 1 << clk->index;
-	else
-		mask &= ~(1 << clk->index);
-	pm_writel(CPU_MASK, mask);
-	spin_unlock_irqrestore(&pm_lock, flags);
-}
-
-static unsigned long cpu_clk_get_rate(struct clk *clk)
-{
-	unsigned long cksel, shift = 0;
-
-	cksel = pm_readl(CKSEL);
-	if (cksel & PM_BIT(CPUDIV))
-		shift = PM_BFEXT(CPUSEL, cksel) + 1;
-
-	return bus_clk_get_rate(clk, shift);
-}
-
-static long cpu_clk_set_rate(struct clk *clk, unsigned long rate, int apply)
-{
-	u32 control;
-	unsigned long parent_rate, child_div, actual_rate, div;
-
-	parent_rate = clk->parent->get_rate(clk->parent);
-	control = pm_readl(CKSEL);
-
-	if (control & PM_BIT(HSBDIV))
-		child_div = 1 << (PM_BFEXT(HSBSEL, control) + 1);
-	else
-		child_div = 1;
-
-	if (rate > 3 * (parent_rate / 4) || child_div == 1) {
-		actual_rate = parent_rate;
-		control &= ~PM_BIT(CPUDIV);
-	} else {
-		unsigned int cpusel;
-		div = (parent_rate + rate / 2) / rate;
-		if (div > child_div)
-			div = child_div;
-		cpusel = (div > 1) ? (fls(div) - 2) : 0;
-		control = PM_BIT(CPUDIV) | PM_BFINS(CPUSEL, cpusel, control);
-		actual_rate = parent_rate / (1 << (cpusel + 1));
-	}
-
-	pr_debug("clk %s: new rate %lu (actual rate %lu)\n",
-			clk->name, rate, actual_rate);
-
-	if (apply)
-		pm_writel(CKSEL, control);
-
-	return actual_rate;
-}
-
-static void hsb_clk_mode(struct clk *clk, int enabled)
-{
-	unsigned long flags;
-	u32 mask;
-
-	spin_lock_irqsave(&pm_lock, flags);
-	mask = pm_readl(HSB_MASK);
-	if (enabled)
-		mask |= 1 << clk->index;
-	else
-		mask &= ~(1 << clk->index);
-	pm_writel(HSB_MASK, mask);
-	spin_unlock_irqrestore(&pm_lock, flags);
-}
-
-static unsigned long hsb_clk_get_rate(struct clk *clk)
-{
-	unsigned long cksel, shift = 0;
-
-	cksel = pm_readl(CKSEL);
-	if (cksel & PM_BIT(HSBDIV))
-		shift = PM_BFEXT(HSBSEL, cksel) + 1;
-
-	return bus_clk_get_rate(clk, shift);
-}
-
-static void pba_clk_mode(struct clk *clk, int enabled)
-{
-	unsigned long flags;
-	u32 mask;
-
-	spin_lock_irqsave(&pm_lock, flags);
-	mask = pm_readl(PBA_MASK);
-	if (enabled)
-		mask |= 1 << clk->index;
-	else
-		mask &= ~(1 << clk->index);
-	pm_writel(PBA_MASK, mask);
-	spin_unlock_irqrestore(&pm_lock, flags);
-}
-
-static unsigned long pba_clk_get_rate(struct clk *clk)
-{
-	unsigned long cksel, shift = 0;
-
-	cksel = pm_readl(CKSEL);
-	if (cksel & PM_BIT(PBADIV))
-		shift = PM_BFEXT(PBASEL, cksel) + 1;
-
-	return bus_clk_get_rate(clk, shift);
-}
-
-static void pbb_clk_mode(struct clk *clk, int enabled)
-{
-	unsigned long flags;
-	u32 mask;
-
-	spin_lock_irqsave(&pm_lock, flags);
-	mask = pm_readl(PBB_MASK);
-	if (enabled)
-		mask |= 1 << clk->index;
-	else
-		mask &= ~(1 << clk->index);
-	pm_writel(PBB_MASK, mask);
-	spin_unlock_irqrestore(&pm_lock, flags);
-}
-
-static unsigned long pbb_clk_get_rate(struct clk *clk)
-{
-	unsigned long cksel, shift = 0;
-
-	cksel = pm_readl(CKSEL);
-	if (cksel & PM_BIT(PBBDIV))
-		shift = PM_BFEXT(PBBSEL, cksel) + 1;
-
-	return bus_clk_get_rate(clk, shift);
-}
-
-static struct clk cpu_clk = {
-	.name		= "cpu",
-	.get_rate	= cpu_clk_get_rate,
-	.set_rate	= cpu_clk_set_rate,
-	.users		= 1,
-};
-static struct clk hsb_clk = {
-	.name		= "hsb",
-	.parent		= &cpu_clk,
-	.get_rate	= hsb_clk_get_rate,
-};
-static struct clk pba_clk = {
-	.name		= "pba",
-	.parent		= &hsb_clk,
-	.mode		= hsb_clk_mode,
-	.get_rate	= pba_clk_get_rate,
-	.index		= 1,
-};
-static struct clk pbb_clk = {
-	.name		= "pbb",
-	.parent		= &hsb_clk,
-	.mode		= hsb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.users		= 1,
-	.index		= 2,
-};
-
-/* --------------------------------------------------------------------
- *  Generic Clock operations
- * -------------------------------------------------------------------- */
-
-static void genclk_mode(struct clk *clk, int enabled)
-{
-	u32 control;
-
-	control = pm_readl(GCCTRL(clk->index));
-	if (enabled)
-		control |= PM_BIT(CEN);
-	else
-		control &= ~PM_BIT(CEN);
-	pm_writel(GCCTRL(clk->index), control);
-}
-
-static unsigned long genclk_get_rate(struct clk *clk)
-{
-	u32 control;
-	unsigned long div = 1;
-
-	control = pm_readl(GCCTRL(clk->index));
-	if (control & PM_BIT(DIVEN))
-		div = 2 * (PM_BFEXT(DIV, control) + 1);
-
-	return clk->parent->get_rate(clk->parent) / div;
-}
-
-static long genclk_set_rate(struct clk *clk, unsigned long rate, int apply)
-{
-	u32 control;
-	unsigned long parent_rate, actual_rate, div;
-
-	parent_rate = clk->parent->get_rate(clk->parent);
-	control = pm_readl(GCCTRL(clk->index));
-
-	if (rate > 3 * parent_rate / 4) {
-		actual_rate = parent_rate;
-		control &= ~PM_BIT(DIVEN);
-	} else {
-		div = (parent_rate + rate) / (2 * rate) - 1;
-		control = PM_BFINS(DIV, div, control) | PM_BIT(DIVEN);
-		actual_rate = parent_rate / (2 * (div + 1));
-	}
-
-	dev_dbg(clk->dev, "clk %s: new rate %lu (actual rate %lu)\n",
-		clk->name, rate, actual_rate);
-
-	if (apply)
-		pm_writel(GCCTRL(clk->index), control);
-
-	return actual_rate;
-}
-
-int genclk_set_parent(struct clk *clk, struct clk *parent)
-{
-	u32 control;
-
-	dev_dbg(clk->dev, "clk %s: new parent %s (was %s)\n",
-		clk->name, parent->name, clk->parent->name);
-
-	control = pm_readl(GCCTRL(clk->index));
-
-	if (parent == &osc1 || parent == &pll1)
-		control |= PM_BIT(OSCSEL);
-	else if (parent == &osc0 || parent == &pll0)
-		control &= ~PM_BIT(OSCSEL);
-	else
-		return -EINVAL;
-
-	if (parent == &pll0 || parent == &pll1)
-		control |= PM_BIT(PLLSEL);
-	else
-		control &= ~PM_BIT(PLLSEL);
-
-	pm_writel(GCCTRL(clk->index), control);
-	clk->parent = parent;
-
-	return 0;
-}
-
-static void __init genclk_init_parent(struct clk *clk)
-{
-	u32 control;
-	struct clk *parent;
-
-	BUG_ON(clk->index > 7);
-
-	control = pm_readl(GCCTRL(clk->index));
-	if (control & PM_BIT(OSCSEL))
-		parent = (control & PM_BIT(PLLSEL)) ? &pll1 : &osc1;
-	else
-		parent = (control & PM_BIT(PLLSEL)) ? &pll0 : &osc0;
-
-	clk->parent = parent;
-}
-
-/* --------------------------------------------------------------------
- *  System peripherals
- * -------------------------------------------------------------------- */
-static struct resource at32_pm0_resource[] = {
-	{
-		.start	= 0xfff00000,
-		.end	= 0xfff0007f,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(20),
-};
-
-static struct resource at32ap700x_rtc0_resource[] = {
-	{
-		.start	= 0xfff00080,
-		.end	= 0xfff000af,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(21),
-};
-
-static struct resource at32_wdt0_resource[] = {
-	{
-		.start	= 0xfff000b0,
-		.end	= 0xfff000cf,
-		.flags	= IORESOURCE_MEM,
-	},
-};
-
-static struct resource at32_eic0_resource[] = {
-	{
-		.start	= 0xfff00100,
-		.end	= 0xfff0013f,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(19),
-};
-
-DEFINE_DEV(at32_pm, 0);
-DEFINE_DEV(at32ap700x_rtc, 0);
-DEFINE_DEV(at32_wdt, 0);
-DEFINE_DEV(at32_eic, 0);
-
-/*
- * Peripheral clock for PM, RTC, WDT and EIC. PM will ensure that this
- * is always running.
- */
-static struct clk at32_pm_pclk = {
-	.name		= "pclk",
-	.dev		= &at32_pm0_device.dev,
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.users		= 1,
-	.index		= 0,
-};
-
-static struct resource intc0_resource[] = {
-	PBMEM(0xfff00400),
-};
-struct platform_device at32_intc0_device = {
-	.name		= "intc",
-	.id		= 0,
-	.resource	= intc0_resource,
-	.num_resources	= ARRAY_SIZE(intc0_resource),
-};
-DEV_CLK(pclk, at32_intc0, pbb, 1);
-
-static struct clk ebi_clk = {
-	.name		= "ebi",
-	.parent		= &hsb_clk,
-	.mode		= hsb_clk_mode,
-	.get_rate	= hsb_clk_get_rate,
-	.users		= 1,
-};
-static struct clk hramc_clk = {
-	.name		= "hramc",
-	.parent		= &hsb_clk,
-	.mode		= hsb_clk_mode,
-	.get_rate	= hsb_clk_get_rate,
-	.users		= 1,
-	.index		= 3,
-};
-
-static struct resource smc0_resource[] = {
-	PBMEM(0xfff03400),
-};
-DEFINE_DEV(smc, 0);
-DEV_CLK(pclk, smc0, pbb, 13);
-DEV_CLK(mck, smc0, hsb, 0);
-
-static struct platform_device pdc_device = {
-	.name		= "pdc",
-	.id		= 0,
-};
-DEV_CLK(hclk, pdc, hsb, 4);
-DEV_CLK(pclk, pdc, pba, 16);
-
-static struct clk pico_clk = {
-	.name		= "pico",
-	.parent		= &cpu_clk,
-	.mode		= cpu_clk_mode,
-	.get_rate	= cpu_clk_get_rate,
-	.users		= 1,
-};
-
-static struct resource dmaca0_resource[] = {
-	{
-		.start	= 0xff200000,
-		.end	= 0xff20ffff,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(2),
-};
-DEFINE_DEV(dmaca, 0);
-DEV_CLK(hclk, dmaca0, hsb, 10);
-
-/* --------------------------------------------------------------------
- * HMATRIX
- * -------------------------------------------------------------------- */
-
-static struct clk hmatrix_clk = {
-	.name		= "hmatrix_clk",
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.index		= 2,
-	.users		= 1,
-};
-#define HMATRIX_BASE	((void __iomem *)0xfff00800)
-
-#define hmatrix_readl(reg)					\
-	__raw_readl((HMATRIX_BASE) + HMATRIX_##reg)
-#define hmatrix_writel(reg,value)				\
-	__raw_writel((value), (HMATRIX_BASE) + HMATRIX_##reg)
-
-/*
- * Set bits in the HMATRIX Special Function Register (SFR) used by the
- * External Bus Interface (EBI). This can be used to enable special
- * features like CompactFlash support, NAND Flash support, etc. on
- * certain chipselects.
- */
-static inline void set_ebi_sfr_bits(u32 mask)
-{
-	u32 sfr;
-
-	clk_enable(&hmatrix_clk);
-	sfr = hmatrix_readl(SFR4);
-	sfr |= mask;
-	hmatrix_writel(SFR4, sfr);
-	clk_disable(&hmatrix_clk);
-}
-
-/* --------------------------------------------------------------------
- *  System Timer/Counter (TC)
- * -------------------------------------------------------------------- */
-static struct resource at32_systc0_resource[] = {
-	PBMEM(0xfff00c00),
-	IRQ(22),
-};
-struct platform_device at32_systc0_device = {
-	.name		= "systc",
-	.id		= 0,
-	.resource	= at32_systc0_resource,
-	.num_resources	= ARRAY_SIZE(at32_systc0_resource),
-};
-DEV_CLK(pclk, at32_systc0, pbb, 3);
-
-/* --------------------------------------------------------------------
- *  PIO
- * -------------------------------------------------------------------- */
-
-static struct resource pio0_resource[] = {
-	PBMEM(0xffe02800),
-	IRQ(13),
-};
-DEFINE_DEV(pio, 0);
-DEV_CLK(mck, pio0, pba, 10);
-
-static struct resource pio1_resource[] = {
-	PBMEM(0xffe02c00),
-	IRQ(14),
-};
-DEFINE_DEV(pio, 1);
-DEV_CLK(mck, pio1, pba, 11);
-
-static struct resource pio2_resource[] = {
-	PBMEM(0xffe03000),
-	IRQ(15),
-};
-DEFINE_DEV(pio, 2);
-DEV_CLK(mck, pio2, pba, 12);
-
-static struct resource pio3_resource[] = {
-	PBMEM(0xffe03400),
-	IRQ(16),
-};
-DEFINE_DEV(pio, 3);
-DEV_CLK(mck, pio3, pba, 13);
-
-static struct resource pio4_resource[] = {
-	PBMEM(0xffe03800),
-	IRQ(17),
-};
-DEFINE_DEV(pio, 4);
-DEV_CLK(mck, pio4, pba, 14);
-
-void __init at32_add_system_devices(void)
-{
-	platform_device_register(&at32_pm0_device);
-	platform_device_register(&at32_intc0_device);
-	platform_device_register(&at32ap700x_rtc0_device);
-	platform_device_register(&at32_wdt0_device);
-	platform_device_register(&at32_eic0_device);
-	platform_device_register(&smc0_device);
-	platform_device_register(&pdc_device);
-	platform_device_register(&dmaca0_device);
-
-	platform_device_register(&at32_systc0_device);
-
-	platform_device_register(&pio0_device);
-	platform_device_register(&pio1_device);
-	platform_device_register(&pio2_device);
-	platform_device_register(&pio3_device);
-	platform_device_register(&pio4_device);
-}
-
-/* --------------------------------------------------------------------
- *  USART
- * -------------------------------------------------------------------- */
-
-static struct atmel_uart_data atmel_usart0_data = {
-	.use_dma_tx	= 1,
-	.use_dma_rx	= 1,
-};
-static struct resource atmel_usart0_resource[] = {
-	PBMEM(0xffe00c00),
-	IRQ(6),
-};
-DEFINE_DEV_DATA(atmel_usart, 0);
-DEV_CLK(usart, atmel_usart0, pba, 3);
-
-static struct atmel_uart_data atmel_usart1_data = {
-	.use_dma_tx	= 1,
-	.use_dma_rx	= 1,
-};
-static struct resource atmel_usart1_resource[] = {
-	PBMEM(0xffe01000),
-	IRQ(7),
-};
-DEFINE_DEV_DATA(atmel_usart, 1);
-DEV_CLK(usart, atmel_usart1, pba, 4);
-
-static struct atmel_uart_data atmel_usart2_data = {
-	.use_dma_tx	= 1,
-	.use_dma_rx	= 1,
-};
-static struct resource atmel_usart2_resource[] = {
-	PBMEM(0xffe01400),
-	IRQ(8),
-};
-DEFINE_DEV_DATA(atmel_usart, 2);
-DEV_CLK(usart, atmel_usart2, pba, 5);
-
-static struct atmel_uart_data atmel_usart3_data = {
-	.use_dma_tx	= 1,
-	.use_dma_rx	= 1,
-};
-static struct resource atmel_usart3_resource[] = {
-	PBMEM(0xffe01800),
-	IRQ(9),
-};
-DEFINE_DEV_DATA(atmel_usart, 3);
-DEV_CLK(usart, atmel_usart3, pba, 6);
-
-static inline void configure_usart0_pins(void)
-{
-	select_peripheral(PA(8),  PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PA(9),  PERIPH_B, 0);	/* TXD	*/
-}
-
-static inline void configure_usart1_pins(void)
-{
-	select_peripheral(PA(17), PERIPH_A, 0);	/* RXD	*/
-	select_peripheral(PA(18), PERIPH_A, 0);	/* TXD	*/
-}
-
-static inline void configure_usart2_pins(void)
-{
-	select_peripheral(PB(26), PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PB(27), PERIPH_B, 0);	/* TXD	*/
-}
-
-static inline void configure_usart3_pins(void)
-{
-	select_peripheral(PB(18), PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PB(17), PERIPH_B, 0);	/* TXD	*/
-}
-
-static struct platform_device *__initdata at32_usarts[4];
-
-void __init at32_map_usart(unsigned int hw_id, unsigned int line)
-{
-	struct platform_device *pdev;
-
-	switch (hw_id) {
-	case 0:
-		pdev = &atmel_usart0_device;
-		configure_usart0_pins();
-		break;
-	case 1:
-		pdev = &atmel_usart1_device;
-		configure_usart1_pins();
-		break;
-	case 2:
-		pdev = &atmel_usart2_device;
-		configure_usart2_pins();
-		break;
-	case 3:
-		pdev = &atmel_usart3_device;
-		configure_usart3_pins();
-		break;
-	default:
-		return;
-	}
-
-	if (PXSEG(pdev->resource[0].start) == P4SEG) {
-		/* Addresses in the P4 segment are permanently mapped 1:1 */
-		struct atmel_uart_data *data = pdev->dev.platform_data;
-		data->regs = (void __iomem *)pdev->resource[0].start;
-	}
-
-	pdev->id = line;
-	at32_usarts[line] = pdev;
-}
-
-struct platform_device *__init at32_add_device_usart(unsigned int id)
-{
-	platform_device_register(at32_usarts[id]);
-	return at32_usarts[id];
-}
-
-struct platform_device *atmel_default_console_device;
-
-void __init at32_setup_serial_console(unsigned int usart_id)
-{
-	atmel_default_console_device = at32_usarts[usart_id];
-}
-
-/* --------------------------------------------------------------------
- *  Ethernet
- * -------------------------------------------------------------------- */
-
-static struct eth_platform_data macb0_data;
-static struct resource macb0_resource[] = {
-	PBMEM(0xfff01800),
-	IRQ(25),
-};
-DEFINE_DEV_DATA(macb, 0);
-DEV_CLK(hclk, macb0, hsb, 8);
-DEV_CLK(pclk, macb0, pbb, 6);
-
-static struct eth_platform_data macb1_data;
-static struct resource macb1_resource[] = {
-	PBMEM(0xfff01c00),
-	IRQ(26),
-};
-DEFINE_DEV_DATA(macb, 1);
-DEV_CLK(hclk, macb1, hsb, 9);
-DEV_CLK(pclk, macb1, pbb, 7);
-
-struct platform_device *__init
-at32_add_device_eth(unsigned int id, struct eth_platform_data *data)
-{
-	struct platform_device *pdev;
-
-	switch (id) {
-	case 0:
-		pdev = &macb0_device;
-
-		select_peripheral(PC(3),  PERIPH_A, 0);	/* TXD0	*/
-		select_peripheral(PC(4),  PERIPH_A, 0);	/* TXD1	*/
-		select_peripheral(PC(7),  PERIPH_A, 0);	/* TXEN	*/
-		select_peripheral(PC(8),  PERIPH_A, 0);	/* TXCK */
-		select_peripheral(PC(9),  PERIPH_A, 0);	/* RXD0	*/
-		select_peripheral(PC(10), PERIPH_A, 0);	/* RXD1	*/
-		select_peripheral(PC(13), PERIPH_A, 0);	/* RXER	*/
-		select_peripheral(PC(15), PERIPH_A, 0);	/* RXDV	*/
-		select_peripheral(PC(16), PERIPH_A, 0);	/* MDC	*/
-		select_peripheral(PC(17), PERIPH_A, 0);	/* MDIO	*/
-
-		if (!data->is_rmii) {
-			select_peripheral(PC(0),  PERIPH_A, 0);	/* COL	*/
-			select_peripheral(PC(1),  PERIPH_A, 0);	/* CRS	*/
-			select_peripheral(PC(2),  PERIPH_A, 0);	/* TXER	*/
-			select_peripheral(PC(5),  PERIPH_A, 0);	/* TXD2	*/
-			select_peripheral(PC(6),  PERIPH_A, 0);	/* TXD3 */
-			select_peripheral(PC(11), PERIPH_A, 0);	/* RXD2	*/
-			select_peripheral(PC(12), PERIPH_A, 0);	/* RXD3	*/
-			select_peripheral(PC(14), PERIPH_A, 0);	/* RXCK	*/
-			select_peripheral(PC(18), PERIPH_A, 0);	/* SPD	*/
-		}
-		break;
-
-	case 1:
-		pdev = &macb1_device;
-
-		select_peripheral(PD(13), PERIPH_B, 0);		/* TXD0	*/
-		select_peripheral(PD(14), PERIPH_B, 0);		/* TXD1	*/
-		select_peripheral(PD(11), PERIPH_B, 0);		/* TXEN	*/
-		select_peripheral(PD(12), PERIPH_B, 0);		/* TXCK */
-		select_peripheral(PD(10), PERIPH_B, 0);		/* RXD0	*/
-		select_peripheral(PD(6),  PERIPH_B, 0);		/* RXD1	*/
-		select_peripheral(PD(5),  PERIPH_B, 0);		/* RXER	*/
-		select_peripheral(PD(4),  PERIPH_B, 0);		/* RXDV	*/
-		select_peripheral(PD(3),  PERIPH_B, 0);		/* MDC	*/
-		select_peripheral(PD(2),  PERIPH_B, 0);		/* MDIO	*/
-
-		if (!data->is_rmii) {
-			select_peripheral(PC(19), PERIPH_B, 0);	/* COL	*/
-			select_peripheral(PC(23), PERIPH_B, 0);	/* CRS	*/
-			select_peripheral(PC(26), PERIPH_B, 0);	/* TXER	*/
-			select_peripheral(PC(27), PERIPH_B, 0);	/* TXD2	*/
-			select_peripheral(PC(28), PERIPH_B, 0);	/* TXD3 */
-			select_peripheral(PC(29), PERIPH_B, 0);	/* RXD2	*/
-			select_peripheral(PC(30), PERIPH_B, 0);	/* RXD3	*/
-			select_peripheral(PC(24), PERIPH_B, 0);	/* RXCK	*/
-			select_peripheral(PD(15), PERIPH_B, 0);	/* SPD	*/
-		}
-		break;
-
-	default:
-		return NULL;
-	}
-
-	memcpy(pdev->dev.platform_data, data, sizeof(struct eth_platform_data));
-	platform_device_register(pdev);
-
-	return pdev;
-}
-
-/* --------------------------------------------------------------------
- *  SPI
- * -------------------------------------------------------------------- */
-static struct resource atmel_spi0_resource[] = {
-	PBMEM(0xffe00000),
-	IRQ(3),
-};
-DEFINE_DEV(atmel_spi, 0);
-DEV_CLK(spi_clk, atmel_spi0, pba, 0);
-
-static struct resource atmel_spi1_resource[] = {
-	PBMEM(0xffe00400),
-	IRQ(4),
-};
-DEFINE_DEV(atmel_spi, 1);
-DEV_CLK(spi_clk, atmel_spi1, pba, 1);
-
-static void __init
-at32_spi_setup_slaves(unsigned int bus_num, struct spi_board_info *b,
-		      unsigned int n, const u8 *pins)
-{
-	unsigned int pin, mode;
-
-	for (; n; n--, b++) {
-		b->bus_num = bus_num;
-		if (b->chip_select >= 4)
-			continue;
-		pin = (unsigned)b->controller_data;
-		if (!pin) {
-			pin = pins[b->chip_select];
-			b->controller_data = (void *)pin;
-		}
-		mode = AT32_GPIOF_OUTPUT;
-		if (!(b->mode & SPI_CS_HIGH))
-			mode |= AT32_GPIOF_HIGH;
-		at32_select_gpio(pin, mode);
-	}
-}
-
-struct platform_device *__init
-at32_add_device_spi(unsigned int id, struct spi_board_info *b, unsigned int n)
-{
-	/*
-	 * Manage the chipselects as GPIOs, normally using the same pins
-	 * the SPI controller expects; but boards can use other pins.
-	 */
-	static u8 __initdata spi0_pins[] =
-		{ GPIO_PIN_PA(3), GPIO_PIN_PA(4),
-		  GPIO_PIN_PA(5), GPIO_PIN_PA(20), };
-	static u8 __initdata spi1_pins[] =
-		{ GPIO_PIN_PB(2), GPIO_PIN_PB(3),
-		  GPIO_PIN_PB(4), GPIO_PIN_PA(27), };
-	struct platform_device *pdev;
-
-	switch (id) {
-	case 0:
-		pdev = &atmel_spi0_device;
-		select_peripheral(PA(0),  PERIPH_A, 0);	/* MISO	 */
-		select_peripheral(PA(1),  PERIPH_A, 0);	/* MOSI	 */
-		select_peripheral(PA(2),  PERIPH_A, 0);	/* SCK	 */
-		at32_spi_setup_slaves(0, b, n, spi0_pins);
-		break;
-
-	case 1:
-		pdev = &atmel_spi1_device;
-		select_peripheral(PB(0),  PERIPH_B, 0);	/* MISO  */
-		select_peripheral(PB(1),  PERIPH_B, 0);	/* MOSI  */
-		select_peripheral(PB(5),  PERIPH_B, 0);	/* SCK   */
-		at32_spi_setup_slaves(1, b, n, spi1_pins);
-		break;
-
-	default:
-		return NULL;
-	}
-
-	spi_register_board_info(b, n);
-	platform_device_register(pdev);
-	return pdev;
-}
-
-/* --------------------------------------------------------------------
- *  TWI
- * -------------------------------------------------------------------- */
-static struct resource atmel_twi0_resource[] __initdata = {
-	PBMEM(0xffe00800),
-	IRQ(5),
-};
-static struct clk atmel_twi0_pclk = {
-	.name		= "twi_pclk",
-	.parent		= &pba_clk,
-	.mode		= pba_clk_mode,
-	.get_rate	= pba_clk_get_rate,
-	.index		= 2,
-};
-
-struct platform_device *__init at32_add_device_twi(unsigned int id)
-{
-	struct platform_device *pdev;
-
-	if (id != 0)
-		return NULL;
-
-	pdev = platform_device_alloc("atmel_twi", id);
-	if (!pdev)
-		return NULL;
-
-	if (platform_device_add_resources(pdev, atmel_twi0_resource,
-				ARRAY_SIZE(atmel_twi0_resource)))
-		goto err_add_resources;
-
-	select_peripheral(PA(6),  PERIPH_A, 0);	/* SDA	*/
-	select_peripheral(PA(7),  PERIPH_A, 0);	/* SDL	*/
-
-	atmel_twi0_pclk.dev = &pdev->dev;
-
-	platform_device_add(pdev);
-	return pdev;
-
-err_add_resources:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- * MMC
- * -------------------------------------------------------------------- */
-static struct resource atmel_mci0_resource[] __initdata = {
-	PBMEM(0xfff02400),
-	IRQ(28),
-};
-static struct clk atmel_mci0_pclk = {
-	.name		= "mci_clk",
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.index		= 9,
-};
-
-struct platform_device *__init at32_add_device_mci(unsigned int id)
-{
-	struct platform_device *pdev;
-
-	if (id != 0)
-		return NULL;
-
-	pdev = platform_device_alloc("atmel_mci", id);
-	if (!pdev)
-		return NULL;
-
-	if (platform_device_add_resources(pdev, atmel_mci0_resource,
-				ARRAY_SIZE(atmel_mci0_resource)))
-		goto err_add_resources;
-
-	select_peripheral(PA(10), PERIPH_A, 0);	/* CLK	 */
-	select_peripheral(PA(11), PERIPH_A, 0);	/* CMD	 */
-	select_peripheral(PA(12), PERIPH_A, 0);	/* DATA0 */
-	select_peripheral(PA(13), PERIPH_A, 0);	/* DATA1 */
-	select_peripheral(PA(14), PERIPH_A, 0);	/* DATA2 */
-	select_peripheral(PA(15), PERIPH_A, 0);	/* DATA3 */
-
-	atmel_mci0_pclk.dev = &pdev->dev;
-
-	platform_device_add(pdev);
-	return pdev;
-
-err_add_resources:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- *  LCDC
- * -------------------------------------------------------------------- */
-static struct atmel_lcdfb_info atmel_lcdfb0_data;
-static struct resource atmel_lcdfb0_resource[] = {
-	{
-		.start		= 0xff000000,
-		.end		= 0xff000fff,
-		.flags		= IORESOURCE_MEM,
-	},
-	IRQ(1),
-	{
-		/* Placeholder for pre-allocated fb memory */
-		.start		= 0x00000000,
-		.end		= 0x00000000,
-		.flags		= 0,
-	},
-};
-DEFINE_DEV_DATA(atmel_lcdfb, 0);
-DEV_CLK(hck1, atmel_lcdfb0, hsb, 7);
-static struct clk atmel_lcdfb0_pixclk = {
-	.name		= "lcdc_clk",
-	.dev		= &atmel_lcdfb0_device.dev,
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 7,
-};
-
-struct platform_device *__init
-at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
-		     unsigned long fbmem_start, unsigned long fbmem_len)
-{
-	struct platform_device *pdev;
-	struct atmel_lcdfb_info *info;
-	struct fb_monspecs *monspecs;
-	struct fb_videomode *modedb;
-	unsigned int modedb_size;
-
-	/*
-	 * Do a deep copy of the fb data, monspecs and modedb. Make
-	 * sure all allocations are done before setting up the
-	 * portmux.
-	 */
-	monspecs = kmemdup(data->default_monspecs,
-			   sizeof(struct fb_monspecs), GFP_KERNEL);
-	if (!monspecs)
-		return NULL;
-
-	modedb_size = sizeof(struct fb_videomode) * monspecs->modedb_len;
-	modedb = kmemdup(monspecs->modedb, modedb_size, GFP_KERNEL);
-	if (!modedb)
-		goto err_dup_modedb;
-	monspecs->modedb = modedb;
-
-	switch (id) {
-	case 0:
-		pdev = &atmel_lcdfb0_device;
-		select_peripheral(PC(19), PERIPH_A, 0);	/* CC	  */
-		select_peripheral(PC(20), PERIPH_A, 0);	/* HSYNC  */
-		select_peripheral(PC(21), PERIPH_A, 0);	/* PCLK	  */
-		select_peripheral(PC(22), PERIPH_A, 0);	/* VSYNC  */
-		select_peripheral(PC(23), PERIPH_A, 0);	/* DVAL	  */
-		select_peripheral(PC(24), PERIPH_A, 0);	/* MODE	  */
-		select_peripheral(PC(25), PERIPH_A, 0);	/* PWR	  */
-		select_peripheral(PC(26), PERIPH_A, 0);	/* DATA0  */
-		select_peripheral(PC(27), PERIPH_A, 0);	/* DATA1  */
-		select_peripheral(PC(28), PERIPH_A, 0);	/* DATA2  */
-		select_peripheral(PC(29), PERIPH_A, 0);	/* DATA3  */
-		select_peripheral(PC(30), PERIPH_A, 0);	/* DATA4  */
-		select_peripheral(PC(31), PERIPH_A, 0);	/* DATA5  */
-		select_peripheral(PD(0),  PERIPH_A, 0);	/* DATA6  */
-		select_peripheral(PD(1),  PERIPH_A, 0);	/* DATA7  */
-		select_peripheral(PD(2),  PERIPH_A, 0);	/* DATA8  */
-		select_peripheral(PD(3),  PERIPH_A, 0);	/* DATA9  */
-		select_peripheral(PD(4),  PERIPH_A, 0);	/* DATA10 */
-		select_peripheral(PD(5),  PERIPH_A, 0);	/* DATA11 */
-		select_peripheral(PD(6),  PERIPH_A, 0);	/* DATA12 */
-		select_peripheral(PD(7),  PERIPH_A, 0);	/* DATA13 */
-		select_peripheral(PD(8),  PERIPH_A, 0);	/* DATA14 */
-		select_peripheral(PD(9),  PERIPH_A, 0);	/* DATA15 */
-		select_peripheral(PD(10), PERIPH_A, 0);	/* DATA16 */
-		select_peripheral(PD(11), PERIPH_A, 0);	/* DATA17 */
-		select_peripheral(PD(12), PERIPH_A, 0);	/* DATA18 */
-		select_peripheral(PD(13), PERIPH_A, 0);	/* DATA19 */
-		select_peripheral(PD(14), PERIPH_A, 0);	/* DATA20 */
-		select_peripheral(PD(15), PERIPH_A, 0);	/* DATA21 */
-		select_peripheral(PD(16), PERIPH_A, 0);	/* DATA22 */
-		select_peripheral(PD(17), PERIPH_A, 0);	/* DATA23 */
-
-		clk_set_parent(&atmel_lcdfb0_pixclk, &pll0);
-		clk_set_rate(&atmel_lcdfb0_pixclk, clk_get_rate(&pll0));
-		break;
-
-	default:
-		goto err_invalid_id;
-	}
-
-	if (fbmem_len) {
-		pdev->resource[2].start = fbmem_start;
-		pdev->resource[2].end = fbmem_start + fbmem_len - 1;
-		pdev->resource[2].flags = IORESOURCE_MEM;
-	}
-
-	info = pdev->dev.platform_data;
-	memcpy(info, data, sizeof(struct atmel_lcdfb_info));
-	info->default_monspecs = monspecs;
-
-	platform_device_register(pdev);
-	return pdev;
-
-err_invalid_id:
-	kfree(modedb);
-err_dup_modedb:
-	kfree(monspecs);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- *  SSC
- * -------------------------------------------------------------------- */
-static struct resource ssc0_resource[] = {
-	PBMEM(0xffe01c00),
-	IRQ(10),
-};
-DEFINE_DEV(ssc, 0);
-DEV_CLK(pclk, ssc0, pba, 7);
-
-static struct resource ssc1_resource[] = {
-	PBMEM(0xffe02000),
-	IRQ(11),
-};
-DEFINE_DEV(ssc, 1);
-DEV_CLK(pclk, ssc1, pba, 8);
-
-static struct resource ssc2_resource[] = {
-	PBMEM(0xffe02400),
-	IRQ(12),
-};
-DEFINE_DEV(ssc, 2);
-DEV_CLK(pclk, ssc2, pba, 9);
-
-struct platform_device *__init
-at32_add_device_ssc(unsigned int id, unsigned int flags)
-{
-	struct platform_device *pdev;
-
-	switch (id) {
-	case 0:
-		pdev = &ssc0_device;
-		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PA(21), PERIPH_A, 0);	/* RF */
-		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PA(22), PERIPH_A, 0);	/* RK */
-		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PA(23), PERIPH_A, 0);	/* TK */
-		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PA(24), PERIPH_A, 0);	/* TF */
-		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PA(25), PERIPH_A, 0);	/* TD */
-		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PA(26), PERIPH_A, 0);	/* RD */
-		break;
-	case 1:
-		pdev = &ssc1_device;
-		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PA(0), PERIPH_B, 0);	/* RF */
-		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PA(1), PERIPH_B, 0);	/* RK */
-		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PA(2), PERIPH_B, 0);	/* TK */
-		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PA(3), PERIPH_B, 0);	/* TF */
-		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PA(4), PERIPH_B, 0);	/* TD */
-		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PA(5), PERIPH_B, 0);	/* RD */
-		break;
-	case 2:
-		pdev = &ssc2_device;
-		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PB(13), PERIPH_A, 0);	/* TD */
-		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PB(14), PERIPH_A, 0);	/* RD */
-		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PB(15), PERIPH_A, 0);	/* TK */
-		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PB(16), PERIPH_A, 0);	/* TF */
-		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PB(17), PERIPH_A, 0);	/* RF */
-		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PB(18), PERIPH_A, 0);	/* RK */
-		break;
-	default:
-		return NULL;
-	}
-
-	platform_device_register(pdev);
-	return pdev;
-}
-
-/* --------------------------------------------------------------------
- *  USB Device Controller
- * -------------------------------------------------------------------- */
-static struct resource usba0_resource[] __initdata = {
-	{
-		.start		= 0xff300000,
-		.end		= 0xff3fffff,
-		.flags		= IORESOURCE_MEM,
-	}, {
-		.start		= 0xfff03000,
-		.end		= 0xfff033ff,
-		.flags		= IORESOURCE_MEM,
-	},
-	IRQ(31),
-};
-static struct clk usba0_pclk = {
-	.name		= "pclk",
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.index		= 12,
-};
-static struct clk usba0_hclk = {
-	.name		= "hclk",
-	.parent		= &hsb_clk,
-	.mode		= hsb_clk_mode,
-	.get_rate	= hsb_clk_get_rate,
-	.index		= 6,
-};
-
-struct platform_device *__init
-at32_add_device_usba(unsigned int id, struct usba_platform_data *data)
-{
-	struct platform_device *pdev;
-
-	if (id != 0)
-		return NULL;
-
-	pdev = platform_device_alloc("atmel_usba_udc", 0);
-	if (!pdev)
-		return NULL;
-
-	if (platform_device_add_resources(pdev, usba0_resource,
-					  ARRAY_SIZE(usba0_resource)))
-		goto out_free_pdev;
-
-	if (data) {
-		if (platform_device_add_data(pdev, data, sizeof(*data)))
-			goto out_free_pdev;
-
-		if (data->vbus_pin != GPIO_PIN_NONE)
-			at32_select_gpio(data->vbus_pin, 0);
-	}
-
-	usba0_pclk.dev = &pdev->dev;
-	usba0_hclk.dev = &pdev->dev;
-
-	platform_device_add(pdev);
-
-	return pdev;
-
-out_free_pdev:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- * IDE / CompactFlash
- * -------------------------------------------------------------------- */
-static struct resource at32_smc_cs4_resource[] __initdata = {
-	{
-		.start	= 0x04000000,
-		.end	= 0x07ffffff,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(~0UL), /* Magic IRQ will be overridden */
-};
-static struct resource at32_smc_cs5_resource[] __initdata = {
-	{
-		.start	= 0x20000000,
-		.end	= 0x23ffffff,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(~0UL), /* Magic IRQ will be overridden */
-};
-
-static int __init at32_init_ide_or_cf(struct platform_device *pdev,
-		unsigned int cs, unsigned int extint)
-{
-	static unsigned int extint_pin_map[4] __initdata = {
-		GPIO_PIN_PB(25),
-		GPIO_PIN_PB(26),
-		GPIO_PIN_PB(27),
-		GPIO_PIN_PB(28),
-	};
-	static bool common_pins_initialized __initdata = false;
-	unsigned int extint_pin;
-	int ret;
-
-	if (extint >= ARRAY_SIZE(extint_pin_map))
-		return -EINVAL;
-	extint_pin = extint_pin_map[extint];
-
-	switch (cs) {
-	case 4:
-		ret = platform_device_add_resources(pdev,
-				at32_smc_cs4_resource,
-				ARRAY_SIZE(at32_smc_cs4_resource));
-		if (ret)
-			return ret;
-
-		select_peripheral(PE(21), PERIPH_A, 0); /* NCS4   -> OE_N  */
-		set_ebi_sfr_bits(HMATRIX_BIT(CS4A));
-		break;
-	case 5:
-		ret = platform_device_add_resources(pdev,
-				at32_smc_cs5_resource,
-				ARRAY_SIZE(at32_smc_cs5_resource));
-		if (ret)
-			return ret;
-
-		select_peripheral(PE(22), PERIPH_A, 0); /* NCS5   -> OE_N  */
-		set_ebi_sfr_bits(HMATRIX_BIT(CS5A));
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	if (!common_pins_initialized) {
-		select_peripheral(PE(19), PERIPH_A, 0);	/* CFCE1  -> CS0_N */
-		select_peripheral(PE(20), PERIPH_A, 0);	/* CFCE2  -> CS1_N */
-		select_peripheral(PE(23), PERIPH_A, 0); /* CFRNW  -> DIR   */
-		select_peripheral(PE(24), PERIPH_A, 0); /* NWAIT  <- IORDY */
-		common_pins_initialized = true;
-	}
-
-	at32_select_periph(extint_pin, GPIO_PERIPH_A, AT32_GPIOF_DEGLITCH);
-
-	pdev->resource[1].start = EIM_IRQ_BASE + extint;
-	pdev->resource[1].end = pdev->resource[1].start;
-
-	return 0;
-}
-
-struct platform_device *__init
-at32_add_device_ide(unsigned int id, unsigned int extint,
-		    struct ide_platform_data *data)
-{
-	struct platform_device *pdev;
-
-	pdev = platform_device_alloc("at32_ide", id);
-	if (!pdev)
-		goto fail;
-
-	if (platform_device_add_data(pdev, data,
-				sizeof(struct ide_platform_data)))
-		goto fail;
-
-	if (at32_init_ide_or_cf(pdev, data->cs, extint))
-		goto fail;
-
-	platform_device_add(pdev);
-	return pdev;
-
-fail:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-struct platform_device *__init
-at32_add_device_cf(unsigned int id, unsigned int extint,
-		    struct cf_platform_data *data)
-{
-	struct platform_device *pdev;
-
-	pdev = platform_device_alloc("at32_cf", id);
-	if (!pdev)
-		goto fail;
-
-	if (platform_device_add_data(pdev, data,
-				sizeof(struct cf_platform_data)))
-		goto fail;
-
-	if (at32_init_ide_or_cf(pdev, data->cs, extint))
-		goto fail;
-
-	if (data->detect_pin != GPIO_PIN_NONE)
-		at32_select_gpio(data->detect_pin, AT32_GPIOF_DEGLITCH);
-	if (data->reset_pin != GPIO_PIN_NONE)
-		at32_select_gpio(data->reset_pin, 0);
-	if (data->vcc_pin != GPIO_PIN_NONE)
-		at32_select_gpio(data->vcc_pin, 0);
-	/* READY is used as extint, so we can't select it as gpio */
-
-	platform_device_add(pdev);
-	return pdev;
-
-fail:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- * AC97C
- * -------------------------------------------------------------------- */
-static struct resource atmel_ac97c0_resource[] __initdata = {
-	PBMEM(0xfff02800),
-	IRQ(29),
-};
-static struct clk atmel_ac97c0_pclk = {
-	.name		= "pclk",
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.index		= 10,
-};
-
-struct platform_device *__init at32_add_device_ac97c(unsigned int id)
-{
-	struct platform_device *pdev;
-
-	if (id != 0)
-		return NULL;
-
-	pdev = platform_device_alloc("atmel_ac97c", id);
-	if (!pdev)
-		return NULL;
-
-	if (platform_device_add_resources(pdev, atmel_ac97c0_resource,
-				ARRAY_SIZE(atmel_ac97c0_resource)))
-		goto err_add_resources;
-
-	select_peripheral(PB(20), PERIPH_B, 0);	/* SYNC	*/
-	select_peripheral(PB(21), PERIPH_B, 0);	/* SDO	*/
-	select_peripheral(PB(22), PERIPH_B, 0);	/* SDI	*/
-	select_peripheral(PB(23), PERIPH_B, 0);	/* SCLK	*/
-
-	atmel_ac97c0_pclk.dev = &pdev->dev;
-
-	platform_device_add(pdev);
-	return pdev;
-
-err_add_resources:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- * ABDAC
- * -------------------------------------------------------------------- */
-static struct resource abdac0_resource[] __initdata = {
-	PBMEM(0xfff02000),
-	IRQ(27),
-};
-static struct clk abdac0_pclk = {
-	.name		= "pclk",
-	.parent		= &pbb_clk,
-	.mode		= pbb_clk_mode,
-	.get_rate	= pbb_clk_get_rate,
-	.index		= 8,
-};
-static struct clk abdac0_sample_clk = {
-	.name		= "sample_clk",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 6,
-};
-
-struct platform_device *__init at32_add_device_abdac(unsigned int id)
-{
-	struct platform_device *pdev;
-
-	if (id != 0)
-		return NULL;
-
-	pdev = platform_device_alloc("abdac", id);
-	if (!pdev)
-		return NULL;
-
-	if (platform_device_add_resources(pdev, abdac0_resource,
-				ARRAY_SIZE(abdac0_resource)))
-		goto err_add_resources;
-
-	select_peripheral(PB(20), PERIPH_A, 0);	/* DATA1	*/
-	select_peripheral(PB(21), PERIPH_A, 0);	/* DATA0	*/
-	select_peripheral(PB(22), PERIPH_A, 0);	/* DATAN1	*/
-	select_peripheral(PB(23), PERIPH_A, 0);	/* DATAN0	*/
-
-	abdac0_pclk.dev = &pdev->dev;
-	abdac0_sample_clk.dev = &pdev->dev;
-
-	platform_device_add(pdev);
-	return pdev;
-
-err_add_resources:
-	platform_device_put(pdev);
-	return NULL;
-}
-
-/* --------------------------------------------------------------------
- *  GCLK
- * -------------------------------------------------------------------- */
-static struct clk gclk0 = {
-	.name		= "gclk0",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 0,
-};
-static struct clk gclk1 = {
-	.name		= "gclk1",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 1,
-};
-static struct clk gclk2 = {
-	.name		= "gclk2",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 2,
-};
-static struct clk gclk3 = {
-	.name		= "gclk3",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 3,
-};
-static struct clk gclk4 = {
-	.name		= "gclk4",
-	.mode		= genclk_mode,
-	.get_rate	= genclk_get_rate,
-	.set_rate	= genclk_set_rate,
-	.set_parent	= genclk_set_parent,
-	.index		= 4,
-};
-
-struct clk *at32_clock_list[] = {
-	&osc32k,
-	&osc0,
-	&osc1,
-	&pll0,
-	&pll1,
-	&cpu_clk,
-	&hsb_clk,
-	&pba_clk,
-	&pbb_clk,
-	&at32_pm_pclk,
-	&at32_intc0_pclk,
-	&hmatrix_clk,
-	&ebi_clk,
-	&hramc_clk,
-	&smc0_pclk,
-	&smc0_mck,
-	&pdc_hclk,
-	&pdc_pclk,
-	&dmaca0_hclk,
-	&pico_clk,
-	&pio0_mck,
-	&pio1_mck,
-	&pio2_mck,
-	&pio3_mck,
-	&pio4_mck,
-	&at32_systc0_pclk,
-	&atmel_usart0_usart,
-	&atmel_usart1_usart,
-	&atmel_usart2_usart,
-	&atmel_usart3_usart,
-	&macb0_hclk,
-	&macb0_pclk,
-	&macb1_hclk,
-	&macb1_pclk,
-	&atmel_spi0_spi_clk,
-	&atmel_spi1_spi_clk,
-	&atmel_twi0_pclk,
-	&atmel_mci0_pclk,
-	&atmel_lcdfb0_hck1,
-	&atmel_lcdfb0_pixclk,
-	&ssc0_pclk,
-	&ssc1_pclk,
-	&ssc2_pclk,
-	&usba0_hclk,
-	&usba0_pclk,
-	&atmel_ac97c0_pclk,
-	&abdac0_pclk,
-	&abdac0_sample_clk,
-	&gclk0,
-	&gclk1,
-	&gclk2,
-	&gclk3,
-	&gclk4,
-};
-unsigned int at32_nr_clocks = ARRAY_SIZE(at32_clock_list);
-
-void __init at32_portmux_init(void)
-{
-	at32_init_pio(&pio0_device);
-	at32_init_pio(&pio1_device);
-	at32_init_pio(&pio2_device);
-	at32_init_pio(&pio3_device);
-	at32_init_pio(&pio4_device);
-}
-
-void __init at32_clock_init(void)
-{
-	u32 cpu_mask = 0, hsb_mask = 0, pba_mask = 0, pbb_mask = 0;
-	int i;
-
-	if (pm_readl(MCCTRL) & PM_BIT(PLLSEL)) {
-		main_clock = &pll0;
-		cpu_clk.parent = &pll0;
-	} else {
-		main_clock = &osc0;
-		cpu_clk.parent = &osc0;
-	}
-
-	if (pm_readl(PLL0) & PM_BIT(PLLOSC))
-		pll0.parent = &osc1;
-	if (pm_readl(PLL1) & PM_BIT(PLLOSC))
-		pll1.parent = &osc1;
-
-	genclk_init_parent(&gclk0);
-	genclk_init_parent(&gclk1);
-	genclk_init_parent(&gclk2);
-	genclk_init_parent(&gclk3);
-	genclk_init_parent(&gclk4);
-	genclk_init_parent(&atmel_lcdfb0_pixclk);
-	genclk_init_parent(&abdac0_sample_clk);
-
-	/*
-	 * Turn on all clocks that have at least one user already, and
-	 * turn off everything else. We only do this for module
-	 * clocks, and even though it isn't particularly pretty to
-	 * check the address of the mode function, it should do the
-	 * trick...
-	 */
-	for (i = 0; i < ARRAY_SIZE(at32_clock_list); i++) {
-		struct clk *clk = at32_clock_list[i];
-
-		if (clk->users == 0)
-			continue;
-
-		if (clk->mode == &cpu_clk_mode)
-			cpu_mask |= 1 << clk->index;
-		else if (clk->mode == &hsb_clk_mode)
-			hsb_mask |= 1 << clk->index;
-		else if (clk->mode == &pba_clk_mode)
-			pba_mask |= 1 << clk->index;
-		else if (clk->mode == &pbb_clk_mode)
-			pbb_mask |= 1 << clk->index;
-	}
-
-	pm_writel(CPU_MASK, cpu_mask);
-	pm_writel(HSB_MASK, hsb_mask);
-	pm_writel(PBA_MASK, pba_mask);
-	pm_writel(PBB_MASK, pbb_mask);
-}
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
new file mode 100644
index 000000000000..9386e1f82fb8
--- /dev/null
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -0,0 +1,1742 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/fb.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/spi/spi.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/at32ap700x.h>
+#include <asm/arch/board.h>
+#include <asm/arch/portmux.h>
+
+#include <video/atmel_lcdc.h>
+
+#include "clock.h"
+#include "hmatrix.h"
+#include "pio.h"
+#include "pm.h"
+
+
+#define PBMEM(base)					\
+	{						\
+		.start		= base,			\
+		.end		= base + 0x3ff,		\
+		.flags		= IORESOURCE_MEM,	\
+	}
+#define IRQ(num)					\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+#define NAMED_IRQ(num, _name)				\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.name		= _name,		\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+
+/* REVISIT these assume *every* device supports DMA, but several
+ * don't ... tc, smc, pio, rtc, watchdog, pwm, ps2, and more.
+ */
+#define DEFINE_DEV(_name, _id)					\
+static u64 _name##_id##_dma_mask = DMA_32BIT_MASK;		\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.dev		= {					\
+		.dma_mask = &_name##_id##_dma_mask,		\
+		.coherent_dma_mask = DMA_32BIT_MASK,		\
+	},							\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+#define DEFINE_DEV_DATA(_name, _id)				\
+static u64 _name##_id##_dma_mask = DMA_32BIT_MASK;		\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.dev		= {					\
+		.dma_mask = &_name##_id##_dma_mask,		\
+		.platform_data	= &_name##_id##_data,		\
+		.coherent_dma_mask = DMA_32BIT_MASK,		\
+	},							\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+
+#define select_peripheral(pin, periph, flags)			\
+	at32_select_periph(GPIO_PIN_##pin, GPIO_##periph, flags)
+
+#define DEV_CLK(_name, devname, bus, _index)			\
+static struct clk devname##_##_name = {				\
+	.name		= #_name,				\
+	.dev		= &devname##_device.dev,		\
+	.parent		= &bus##_clk,				\
+	.mode		= bus##_clk_mode,			\
+	.get_rate	= bus##_clk_get_rate,			\
+	.index		= _index,				\
+}
+
+static DEFINE_SPINLOCK(pm_lock);
+
+unsigned long at32ap7000_osc_rates[3] = {
+	[0] = 32768,
+	/* FIXME: these are ATSTK1002-specific */
+	[1] = 20000000,
+	[2] = 12000000,
+};
+
+static unsigned long osc_get_rate(struct clk *clk)
+{
+	return at32ap7000_osc_rates[clk->index];
+}
+
+static unsigned long pll_get_rate(struct clk *clk, unsigned long control)
+{
+	unsigned long div, mul, rate;
+
+	if (!(control & PM_BIT(PLLEN)))
+		return 0;
+
+	div = PM_BFEXT(PLLDIV, control) + 1;
+	mul = PM_BFEXT(PLLMUL, control) + 1;
+
+	rate = clk->parent->get_rate(clk->parent);
+	rate = (rate + div / 2) / div;
+	rate *= mul;
+
+	return rate;
+}
+
+static unsigned long pll0_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = pm_readl(PLL0);
+
+	return pll_get_rate(clk, control);
+}
+
+static unsigned long pll1_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = pm_readl(PLL1);
+
+	return pll_get_rate(clk, control);
+}
+
+/*
+ * The AT32AP7000 has five primary clock sources: One 32kHz
+ * oscillator, two crystal oscillators and two PLLs.
+ */
+static struct clk osc32k = {
+	.name		= "osc32k",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 0,
+};
+static struct clk osc0 = {
+	.name		= "osc0",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 1,
+};
+static struct clk osc1 = {
+	.name		= "osc1",
+	.get_rate	= osc_get_rate,
+	.index		= 2,
+};
+static struct clk pll0 = {
+	.name		= "pll0",
+	.get_rate	= pll0_get_rate,
+	.parent		= &osc0,
+};
+static struct clk pll1 = {
+	.name		= "pll1",
+	.get_rate	= pll1_get_rate,
+	.parent		= &osc0,
+};
+
+/*
+ * The main clock can be either osc0 or pll0.  The boot loader may
+ * have chosen one for us, so we don't really know which one until we
+ * have a look at the SM.
+ */
+static struct clk *main_clock;
+
+/*
+ * Synchronous clocks are generated from the main clock. The clocks
+ * must satisfy the constraint
+ *   fCPU >= fHSB >= fPB
+ * i.e. each clock must not be faster than its parent.
+ */
+static unsigned long bus_clk_get_rate(struct clk *clk, unsigned int shift)
+{
+	return main_clock->get_rate(main_clock) >> shift;
+};
+
+static void cpu_clk_mode(struct clk *clk, int enabled)
+{
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&pm_lock, flags);
+	mask = pm_readl(CPU_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	pm_writel(CPU_MASK, mask);
+	spin_unlock_irqrestore(&pm_lock, flags);
+}
+
+static unsigned long cpu_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = pm_readl(CKSEL);
+	if (cksel & PM_BIT(CPUDIV))
+		shift = PM_BFEXT(CPUSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static long cpu_clk_set_rate(struct clk *clk, unsigned long rate, int apply)
+{
+	u32 control;
+	unsigned long parent_rate, child_div, actual_rate, div;
+
+	parent_rate = clk->parent->get_rate(clk->parent);
+	control = pm_readl(CKSEL);
+
+	if (control & PM_BIT(HSBDIV))
+		child_div = 1 << (PM_BFEXT(HSBSEL, control) + 1);
+	else
+		child_div = 1;
+
+	if (rate > 3 * (parent_rate / 4) || child_div == 1) {
+		actual_rate = parent_rate;
+		control &= ~PM_BIT(CPUDIV);
+	} else {
+		unsigned int cpusel;
+		div = (parent_rate + rate / 2) / rate;
+		if (div > child_div)
+			div = child_div;
+		cpusel = (div > 1) ? (fls(div) - 2) : 0;
+		control = PM_BIT(CPUDIV) | PM_BFINS(CPUSEL, cpusel, control);
+		actual_rate = parent_rate / (1 << (cpusel + 1));
+	}
+
+	pr_debug("clk %s: new rate %lu (actual rate %lu)\n",
+			clk->name, rate, actual_rate);
+
+	if (apply)
+		pm_writel(CKSEL, control);
+
+	return actual_rate;
+}
+
+static void hsb_clk_mode(struct clk *clk, int enabled)
+{
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&pm_lock, flags);
+	mask = pm_readl(HSB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	pm_writel(HSB_MASK, mask);
+	spin_unlock_irqrestore(&pm_lock, flags);
+}
+
+static unsigned long hsb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = pm_readl(CKSEL);
+	if (cksel & PM_BIT(HSBDIV))
+		shift = PM_BFEXT(HSBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pba_clk_mode(struct clk *clk, int enabled)
+{
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&pm_lock, flags);
+	mask = pm_readl(PBA_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	pm_writel(PBA_MASK, mask);
+	spin_unlock_irqrestore(&pm_lock, flags);
+}
+
+static unsigned long pba_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = pm_readl(CKSEL);
+	if (cksel & PM_BIT(PBADIV))
+		shift = PM_BFEXT(PBASEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pbb_clk_mode(struct clk *clk, int enabled)
+{
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&pm_lock, flags);
+	mask = pm_readl(PBB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	pm_writel(PBB_MASK, mask);
+	spin_unlock_irqrestore(&pm_lock, flags);
+}
+
+static unsigned long pbb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = pm_readl(CKSEL);
+	if (cksel & PM_BIT(PBBDIV))
+		shift = PM_BFEXT(PBBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static struct clk cpu_clk = {
+	.name		= "cpu",
+	.get_rate	= cpu_clk_get_rate,
+	.set_rate	= cpu_clk_set_rate,
+	.users		= 1,
+};
+static struct clk hsb_clk = {
+	.name		= "hsb",
+	.parent		= &cpu_clk,
+	.get_rate	= hsb_clk_get_rate,
+};
+static struct clk pba_clk = {
+	.name		= "pba",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pba_clk_get_rate,
+	.index		= 1,
+};
+static struct clk pbb_clk = {
+	.name		= "pbb",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.users		= 1,
+	.index		= 2,
+};
+
+/* --------------------------------------------------------------------
+ *  Generic Clock operations
+ * -------------------------------------------------------------------- */
+
+static void genclk_mode(struct clk *clk, int enabled)
+{
+	u32 control;
+
+	control = pm_readl(GCCTRL(clk->index));
+	if (enabled)
+		control |= PM_BIT(CEN);
+	else
+		control &= ~PM_BIT(CEN);
+	pm_writel(GCCTRL(clk->index), control);
+}
+
+static unsigned long genclk_get_rate(struct clk *clk)
+{
+	u32 control;
+	unsigned long div = 1;
+
+	control = pm_readl(GCCTRL(clk->index));
+	if (control & PM_BIT(DIVEN))
+		div = 2 * (PM_BFEXT(DIV, control) + 1);
+
+	return clk->parent->get_rate(clk->parent) / div;
+}
+
+static long genclk_set_rate(struct clk *clk, unsigned long rate, int apply)
+{
+	u32 control;
+	unsigned long parent_rate, actual_rate, div;
+
+	parent_rate = clk->parent->get_rate(clk->parent);
+	control = pm_readl(GCCTRL(clk->index));
+
+	if (rate > 3 * parent_rate / 4) {
+		actual_rate = parent_rate;
+		control &= ~PM_BIT(DIVEN);
+	} else {
+		div = (parent_rate + rate) / (2 * rate) - 1;
+		control = PM_BFINS(DIV, div, control) | PM_BIT(DIVEN);
+		actual_rate = parent_rate / (2 * (div + 1));
+	}
+
+	dev_dbg(clk->dev, "clk %s: new rate %lu (actual rate %lu)\n",
+		clk->name, rate, actual_rate);
+
+	if (apply)
+		pm_writel(GCCTRL(clk->index), control);
+
+	return actual_rate;
+}
+
+int genclk_set_parent(struct clk *clk, struct clk *parent)
+{
+	u32 control;
+
+	dev_dbg(clk->dev, "clk %s: new parent %s (was %s)\n",
+		clk->name, parent->name, clk->parent->name);
+
+	control = pm_readl(GCCTRL(clk->index));
+
+	if (parent == &osc1 || parent == &pll1)
+		control |= PM_BIT(OSCSEL);
+	else if (parent == &osc0 || parent == &pll0)
+		control &= ~PM_BIT(OSCSEL);
+	else
+		return -EINVAL;
+
+	if (parent == &pll0 || parent == &pll1)
+		control |= PM_BIT(PLLSEL);
+	else
+		control &= ~PM_BIT(PLLSEL);
+
+	pm_writel(GCCTRL(clk->index), control);
+	clk->parent = parent;
+
+	return 0;
+}
+
+static void __init genclk_init_parent(struct clk *clk)
+{
+	u32 control;
+	struct clk *parent;
+
+	BUG_ON(clk->index > 7);
+
+	control = pm_readl(GCCTRL(clk->index));
+	if (control & PM_BIT(OSCSEL))
+		parent = (control & PM_BIT(PLLSEL)) ? &pll1 : &osc1;
+	else
+		parent = (control & PM_BIT(PLLSEL)) ? &pll0 : &osc0;
+
+	clk->parent = parent;
+}
+
+/* --------------------------------------------------------------------
+ *  System peripherals
+ * -------------------------------------------------------------------- */
+static struct resource at32_pm0_resource[] = {
+	{
+		.start	= 0xfff00000,
+		.end	= 0xfff0007f,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(20),
+};
+
+static struct resource at32ap700x_rtc0_resource[] = {
+	{
+		.start	= 0xfff00080,
+		.end	= 0xfff000af,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(21),
+};
+
+static struct resource at32_wdt0_resource[] = {
+	{
+		.start	= 0xfff000b0,
+		.end	= 0xfff000cf,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+static struct resource at32_eic0_resource[] = {
+	{
+		.start	= 0xfff00100,
+		.end	= 0xfff0013f,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(19),
+};
+
+DEFINE_DEV(at32_pm, 0);
+DEFINE_DEV(at32ap700x_rtc, 0);
+DEFINE_DEV(at32_wdt, 0);
+DEFINE_DEV(at32_eic, 0);
+
+/*
+ * Peripheral clock for PM, RTC, WDT and EIC. PM will ensure that this
+ * is always running.
+ */
+static struct clk at32_pm_pclk = {
+	.name		= "pclk",
+	.dev		= &at32_pm0_device.dev,
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.users		= 1,
+	.index		= 0,
+};
+
+static struct resource intc0_resource[] = {
+	PBMEM(0xfff00400),
+};
+struct platform_device at32_intc0_device = {
+	.name		= "intc",
+	.id		= 0,
+	.resource	= intc0_resource,
+	.num_resources	= ARRAY_SIZE(intc0_resource),
+};
+DEV_CLK(pclk, at32_intc0, pbb, 1);
+
+static struct clk ebi_clk = {
+	.name		= "ebi",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+};
+static struct clk hramc_clk = {
+	.name		= "hramc",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+	.index		= 3,
+};
+
+static struct resource smc0_resource[] = {
+	PBMEM(0xfff03400),
+};
+DEFINE_DEV(smc, 0);
+DEV_CLK(pclk, smc0, pbb, 13);
+DEV_CLK(mck, smc0, hsb, 0);
+
+static struct platform_device pdc_device = {
+	.name		= "pdc",
+	.id		= 0,
+};
+DEV_CLK(hclk, pdc, hsb, 4);
+DEV_CLK(pclk, pdc, pba, 16);
+
+static struct clk pico_clk = {
+	.name		= "pico",
+	.parent		= &cpu_clk,
+	.mode		= cpu_clk_mode,
+	.get_rate	= cpu_clk_get_rate,
+	.users		= 1,
+};
+
+static struct resource dmaca0_resource[] = {
+	{
+		.start	= 0xff200000,
+		.end	= 0xff20ffff,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(2),
+};
+DEFINE_DEV(dmaca, 0);
+DEV_CLK(hclk, dmaca0, hsb, 10);
+
+/* --------------------------------------------------------------------
+ * HMATRIX
+ * -------------------------------------------------------------------- */
+
+static struct clk hmatrix_clk = {
+	.name		= "hmatrix_clk",
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.index		= 2,
+	.users		= 1,
+};
+#define HMATRIX_BASE	((void __iomem *)0xfff00800)
+
+#define hmatrix_readl(reg)					\
+	__raw_readl((HMATRIX_BASE) + HMATRIX_##reg)
+#define hmatrix_writel(reg,value)				\
+	__raw_writel((value), (HMATRIX_BASE) + HMATRIX_##reg)
+
+/*
+ * Set bits in the HMATRIX Special Function Register (SFR) used by the
+ * External Bus Interface (EBI). This can be used to enable special
+ * features like CompactFlash support, NAND Flash support, etc. on
+ * certain chipselects.
+ */
+static inline void set_ebi_sfr_bits(u32 mask)
+{
+	u32 sfr;
+
+	clk_enable(&hmatrix_clk);
+	sfr = hmatrix_readl(SFR4);
+	sfr |= mask;
+	hmatrix_writel(SFR4, sfr);
+	clk_disable(&hmatrix_clk);
+}
+
+/* --------------------------------------------------------------------
+ *  System Timer/Counter (TC)
+ * -------------------------------------------------------------------- */
+static struct resource at32_systc0_resource[] = {
+	PBMEM(0xfff00c00),
+	IRQ(22),
+};
+struct platform_device at32_systc0_device = {
+	.name		= "systc",
+	.id		= 0,
+	.resource	= at32_systc0_resource,
+	.num_resources	= ARRAY_SIZE(at32_systc0_resource),
+};
+DEV_CLK(pclk, at32_systc0, pbb, 3);
+
+/* --------------------------------------------------------------------
+ *  PIO
+ * -------------------------------------------------------------------- */
+
+static struct resource pio0_resource[] = {
+	PBMEM(0xffe02800),
+	IRQ(13),
+};
+DEFINE_DEV(pio, 0);
+DEV_CLK(mck, pio0, pba, 10);
+
+static struct resource pio1_resource[] = {
+	PBMEM(0xffe02c00),
+	IRQ(14),
+};
+DEFINE_DEV(pio, 1);
+DEV_CLK(mck, pio1, pba, 11);
+
+static struct resource pio2_resource[] = {
+	PBMEM(0xffe03000),
+	IRQ(15),
+};
+DEFINE_DEV(pio, 2);
+DEV_CLK(mck, pio2, pba, 12);
+
+static struct resource pio3_resource[] = {
+	PBMEM(0xffe03400),
+	IRQ(16),
+};
+DEFINE_DEV(pio, 3);
+DEV_CLK(mck, pio3, pba, 13);
+
+static struct resource pio4_resource[] = {
+	PBMEM(0xffe03800),
+	IRQ(17),
+};
+DEFINE_DEV(pio, 4);
+DEV_CLK(mck, pio4, pba, 14);
+
+void __init at32_add_system_devices(void)
+{
+	platform_device_register(&at32_pm0_device);
+	platform_device_register(&at32_intc0_device);
+	platform_device_register(&at32ap700x_rtc0_device);
+	platform_device_register(&at32_wdt0_device);
+	platform_device_register(&at32_eic0_device);
+	platform_device_register(&smc0_device);
+	platform_device_register(&pdc_device);
+	platform_device_register(&dmaca0_device);
+
+	platform_device_register(&at32_systc0_device);
+
+	platform_device_register(&pio0_device);
+	platform_device_register(&pio1_device);
+	platform_device_register(&pio2_device);
+	platform_device_register(&pio3_device);
+	platform_device_register(&pio4_device);
+}
+
+/* --------------------------------------------------------------------
+ *  USART
+ * -------------------------------------------------------------------- */
+
+static struct atmel_uart_data atmel_usart0_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+static struct resource atmel_usart0_resource[] = {
+	PBMEM(0xffe00c00),
+	IRQ(6),
+};
+DEFINE_DEV_DATA(atmel_usart, 0);
+DEV_CLK(usart, atmel_usart0, pba, 3);
+
+static struct atmel_uart_data atmel_usart1_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+static struct resource atmel_usart1_resource[] = {
+	PBMEM(0xffe01000),
+	IRQ(7),
+};
+DEFINE_DEV_DATA(atmel_usart, 1);
+DEV_CLK(usart, atmel_usart1, pba, 4);
+
+static struct atmel_uart_data atmel_usart2_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+static struct resource atmel_usart2_resource[] = {
+	PBMEM(0xffe01400),
+	IRQ(8),
+};
+DEFINE_DEV_DATA(atmel_usart, 2);
+DEV_CLK(usart, atmel_usart2, pba, 5);
+
+static struct atmel_uart_data atmel_usart3_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+static struct resource atmel_usart3_resource[] = {
+	PBMEM(0xffe01800),
+	IRQ(9),
+};
+DEFINE_DEV_DATA(atmel_usart, 3);
+DEV_CLK(usart, atmel_usart3, pba, 6);
+
+static inline void configure_usart0_pins(void)
+{
+	select_peripheral(PA(8),  PERIPH_B, 0);	/* RXD	*/
+	select_peripheral(PA(9),  PERIPH_B, 0);	/* TXD	*/
+}
+
+static inline void configure_usart1_pins(void)
+{
+	select_peripheral(PA(17), PERIPH_A, 0);	/* RXD	*/
+	select_peripheral(PA(18), PERIPH_A, 0);	/* TXD	*/
+}
+
+static inline void configure_usart2_pins(void)
+{
+	select_peripheral(PB(26), PERIPH_B, 0);	/* RXD	*/
+	select_peripheral(PB(27), PERIPH_B, 0);	/* TXD	*/
+}
+
+static inline void configure_usart3_pins(void)
+{
+	select_peripheral(PB(18), PERIPH_B, 0);	/* RXD	*/
+	select_peripheral(PB(17), PERIPH_B, 0);	/* TXD	*/
+}
+
+static struct platform_device *__initdata at32_usarts[4];
+
+void __init at32_map_usart(unsigned int hw_id, unsigned int line)
+{
+	struct platform_device *pdev;
+
+	switch (hw_id) {
+	case 0:
+		pdev = &atmel_usart0_device;
+		configure_usart0_pins();
+		break;
+	case 1:
+		pdev = &atmel_usart1_device;
+		configure_usart1_pins();
+		break;
+	case 2:
+		pdev = &atmel_usart2_device;
+		configure_usart2_pins();
+		break;
+	case 3:
+		pdev = &atmel_usart3_device;
+		configure_usart3_pins();
+		break;
+	default:
+		return;
+	}
+
+	if (PXSEG(pdev->resource[0].start) == P4SEG) {
+		/* Addresses in the P4 segment are permanently mapped 1:1 */
+		struct atmel_uart_data *data = pdev->dev.platform_data;
+		data->regs = (void __iomem *)pdev->resource[0].start;
+	}
+
+	pdev->id = line;
+	at32_usarts[line] = pdev;
+}
+
+struct platform_device *__init at32_add_device_usart(unsigned int id)
+{
+	platform_device_register(at32_usarts[id]);
+	return at32_usarts[id];
+}
+
+struct platform_device *atmel_default_console_device;
+
+void __init at32_setup_serial_console(unsigned int usart_id)
+{
+	atmel_default_console_device = at32_usarts[usart_id];
+}
+
+/* --------------------------------------------------------------------
+ *  Ethernet
+ * -------------------------------------------------------------------- */
+
+#ifdef CONFIG_CPU_AT32AP7000
+static struct eth_platform_data macb0_data;
+static struct resource macb0_resource[] = {
+	PBMEM(0xfff01800),
+	IRQ(25),
+};
+DEFINE_DEV_DATA(macb, 0);
+DEV_CLK(hclk, macb0, hsb, 8);
+DEV_CLK(pclk, macb0, pbb, 6);
+
+static struct eth_platform_data macb1_data;
+static struct resource macb1_resource[] = {
+	PBMEM(0xfff01c00),
+	IRQ(26),
+};
+DEFINE_DEV_DATA(macb, 1);
+DEV_CLK(hclk, macb1, hsb, 9);
+DEV_CLK(pclk, macb1, pbb, 7);
+
+struct platform_device *__init
+at32_add_device_eth(unsigned int id, struct eth_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &macb0_device;
+
+		select_peripheral(PC(3),  PERIPH_A, 0);	/* TXD0	*/
+		select_peripheral(PC(4),  PERIPH_A, 0);	/* TXD1	*/
+		select_peripheral(PC(7),  PERIPH_A, 0);	/* TXEN	*/
+		select_peripheral(PC(8),  PERIPH_A, 0);	/* TXCK */
+		select_peripheral(PC(9),  PERIPH_A, 0);	/* RXD0	*/
+		select_peripheral(PC(10), PERIPH_A, 0);	/* RXD1	*/
+		select_peripheral(PC(13), PERIPH_A, 0);	/* RXER	*/
+		select_peripheral(PC(15), PERIPH_A, 0);	/* RXDV	*/
+		select_peripheral(PC(16), PERIPH_A, 0);	/* MDC	*/
+		select_peripheral(PC(17), PERIPH_A, 0);	/* MDIO	*/
+
+		if (!data->is_rmii) {
+			select_peripheral(PC(0),  PERIPH_A, 0);	/* COL	*/
+			select_peripheral(PC(1),  PERIPH_A, 0);	/* CRS	*/
+			select_peripheral(PC(2),  PERIPH_A, 0);	/* TXER	*/
+			select_peripheral(PC(5),  PERIPH_A, 0);	/* TXD2	*/
+			select_peripheral(PC(6),  PERIPH_A, 0);	/* TXD3 */
+			select_peripheral(PC(11), PERIPH_A, 0);	/* RXD2	*/
+			select_peripheral(PC(12), PERIPH_A, 0);	/* RXD3	*/
+			select_peripheral(PC(14), PERIPH_A, 0);	/* RXCK	*/
+			select_peripheral(PC(18), PERIPH_A, 0);	/* SPD	*/
+		}
+		break;
+
+	case 1:
+		pdev = &macb1_device;
+
+		select_peripheral(PD(13), PERIPH_B, 0);		/* TXD0	*/
+		select_peripheral(PD(14), PERIPH_B, 0);		/* TXD1	*/
+		select_peripheral(PD(11), PERIPH_B, 0);		/* TXEN	*/
+		select_peripheral(PD(12), PERIPH_B, 0);		/* TXCK */
+		select_peripheral(PD(10), PERIPH_B, 0);		/* RXD0	*/
+		select_peripheral(PD(6),  PERIPH_B, 0);		/* RXD1	*/
+		select_peripheral(PD(5),  PERIPH_B, 0);		/* RXER	*/
+		select_peripheral(PD(4),  PERIPH_B, 0);		/* RXDV	*/
+		select_peripheral(PD(3),  PERIPH_B, 0);		/* MDC	*/
+		select_peripheral(PD(2),  PERIPH_B, 0);		/* MDIO	*/
+
+		if (!data->is_rmii) {
+			select_peripheral(PC(19), PERIPH_B, 0);	/* COL	*/
+			select_peripheral(PC(23), PERIPH_B, 0);	/* CRS	*/
+			select_peripheral(PC(26), PERIPH_B, 0);	/* TXER	*/
+			select_peripheral(PC(27), PERIPH_B, 0);	/* TXD2	*/
+			select_peripheral(PC(28), PERIPH_B, 0);	/* TXD3 */
+			select_peripheral(PC(29), PERIPH_B, 0);	/* RXD2	*/
+			select_peripheral(PC(30), PERIPH_B, 0);	/* RXD3	*/
+			select_peripheral(PC(24), PERIPH_B, 0);	/* RXCK	*/
+			select_peripheral(PD(15), PERIPH_B, 0);	/* SPD	*/
+		}
+		break;
+
+	default:
+		return NULL;
+	}
+
+	memcpy(pdev->dev.platform_data, data, sizeof(struct eth_platform_data));
+	platform_device_register(pdev);
+
+	return pdev;
+}
+#endif
+
+/* --------------------------------------------------------------------
+ *  SPI
+ * -------------------------------------------------------------------- */
+static struct resource atmel_spi0_resource[] = {
+	PBMEM(0xffe00000),
+	IRQ(3),
+};
+DEFINE_DEV(atmel_spi, 0);
+DEV_CLK(spi_clk, atmel_spi0, pba, 0);
+
+static struct resource atmel_spi1_resource[] = {
+	PBMEM(0xffe00400),
+	IRQ(4),
+};
+DEFINE_DEV(atmel_spi, 1);
+DEV_CLK(spi_clk, atmel_spi1, pba, 1);
+
+static void __init
+at32_spi_setup_slaves(unsigned int bus_num, struct spi_board_info *b,
+		      unsigned int n, const u8 *pins)
+{
+	unsigned int pin, mode;
+
+	for (; n; n--, b++) {
+		b->bus_num = bus_num;
+		if (b->chip_select >= 4)
+			continue;
+		pin = (unsigned)b->controller_data;
+		if (!pin) {
+			pin = pins[b->chip_select];
+			b->controller_data = (void *)pin;
+		}
+		mode = AT32_GPIOF_OUTPUT;
+		if (!(b->mode & SPI_CS_HIGH))
+			mode |= AT32_GPIOF_HIGH;
+		at32_select_gpio(pin, mode);
+	}
+}
+
+struct platform_device *__init
+at32_add_device_spi(unsigned int id, struct spi_board_info *b, unsigned int n)
+{
+	/*
+	 * Manage the chipselects as GPIOs, normally using the same pins
+	 * the SPI controller expects; but boards can use other pins.
+	 */
+	static u8 __initdata spi0_pins[] =
+		{ GPIO_PIN_PA(3), GPIO_PIN_PA(4),
+		  GPIO_PIN_PA(5), GPIO_PIN_PA(20), };
+	static u8 __initdata spi1_pins[] =
+		{ GPIO_PIN_PB(2), GPIO_PIN_PB(3),
+		  GPIO_PIN_PB(4), GPIO_PIN_PA(27), };
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &atmel_spi0_device;
+		select_peripheral(PA(0),  PERIPH_A, 0);	/* MISO	 */
+		select_peripheral(PA(1),  PERIPH_A, 0);	/* MOSI	 */
+		select_peripheral(PA(2),  PERIPH_A, 0);	/* SCK	 */
+		at32_spi_setup_slaves(0, b, n, spi0_pins);
+		break;
+
+	case 1:
+		pdev = &atmel_spi1_device;
+		select_peripheral(PB(0),  PERIPH_B, 0);	/* MISO  */
+		select_peripheral(PB(1),  PERIPH_B, 0);	/* MOSI  */
+		select_peripheral(PB(5),  PERIPH_B, 0);	/* SCK   */
+		at32_spi_setup_slaves(1, b, n, spi1_pins);
+		break;
+
+	default:
+		return NULL;
+	}
+
+	spi_register_board_info(b, n);
+	platform_device_register(pdev);
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  TWI
+ * -------------------------------------------------------------------- */
+static struct resource atmel_twi0_resource[] __initdata = {
+	PBMEM(0xffe00800),
+	IRQ(5),
+};
+static struct clk atmel_twi0_pclk = {
+	.name		= "twi_pclk",
+	.parent		= &pba_clk,
+	.mode		= pba_clk_mode,
+	.get_rate	= pba_clk_get_rate,
+	.index		= 2,
+};
+
+struct platform_device *__init at32_add_device_twi(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	if (id != 0)
+		return NULL;
+
+	pdev = platform_device_alloc("atmel_twi", id);
+	if (!pdev)
+		return NULL;
+
+	if (platform_device_add_resources(pdev, atmel_twi0_resource,
+				ARRAY_SIZE(atmel_twi0_resource)))
+		goto err_add_resources;
+
+	select_peripheral(PA(6),  PERIPH_A, 0);	/* SDA	*/
+	select_peripheral(PA(7),  PERIPH_A, 0);	/* SDL	*/
+
+	atmel_twi0_pclk.dev = &pdev->dev;
+
+	platform_device_add(pdev);
+	return pdev;
+
+err_add_resources:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+/* --------------------------------------------------------------------
+ * MMC
+ * -------------------------------------------------------------------- */
+static struct resource atmel_mci0_resource[] __initdata = {
+	PBMEM(0xfff02400),
+	IRQ(28),
+};
+static struct clk atmel_mci0_pclk = {
+	.name		= "mci_clk",
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.index		= 9,
+};
+
+struct platform_device *__init at32_add_device_mci(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	if (id != 0)
+		return NULL;
+
+	pdev = platform_device_alloc("atmel_mci", id);
+	if (!pdev)
+		return NULL;
+
+	if (platform_device_add_resources(pdev, atmel_mci0_resource,
+				ARRAY_SIZE(atmel_mci0_resource)))
+		goto err_add_resources;
+
+	select_peripheral(PA(10), PERIPH_A, 0);	/* CLK	 */
+	select_peripheral(PA(11), PERIPH_A, 0);	/* CMD	 */
+	select_peripheral(PA(12), PERIPH_A, 0);	/* DATA0 */
+	select_peripheral(PA(13), PERIPH_A, 0);	/* DATA1 */
+	select_peripheral(PA(14), PERIPH_A, 0);	/* DATA2 */
+	select_peripheral(PA(15), PERIPH_A, 0);	/* DATA3 */
+
+	atmel_mci0_pclk.dev = &pdev->dev;
+
+	platform_device_add(pdev);
+	return pdev;
+
+err_add_resources:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+/* --------------------------------------------------------------------
+ *  LCDC
+ * -------------------------------------------------------------------- */
+#if defined(CONFIG_CPU_AT32AP7000) || defined(CONFIG_CPU_AT32AP7002)
+static struct atmel_lcdfb_info atmel_lcdfb0_data;
+static struct resource atmel_lcdfb0_resource[] = {
+	{
+		.start		= 0xff000000,
+		.end		= 0xff000fff,
+		.flags		= IORESOURCE_MEM,
+	},
+	IRQ(1),
+	{
+		/* Placeholder for pre-allocated fb memory */
+		.start		= 0x00000000,
+		.end		= 0x00000000,
+		.flags		= 0,
+	},
+};
+DEFINE_DEV_DATA(atmel_lcdfb, 0);
+DEV_CLK(hck1, atmel_lcdfb0, hsb, 7);
+static struct clk atmel_lcdfb0_pixclk = {
+	.name		= "lcdc_clk",
+	.dev		= &atmel_lcdfb0_device.dev,
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 7,
+};
+
+struct platform_device *__init
+at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
+		     unsigned long fbmem_start, unsigned long fbmem_len)
+{
+	struct platform_device *pdev;
+	struct atmel_lcdfb_info *info;
+	struct fb_monspecs *monspecs;
+	struct fb_videomode *modedb;
+	unsigned int modedb_size;
+
+	/*
+	 * Do a deep copy of the fb data, monspecs and modedb. Make
+	 * sure all allocations are done before setting up the
+	 * portmux.
+	 */
+	monspecs = kmemdup(data->default_monspecs,
+			   sizeof(struct fb_monspecs), GFP_KERNEL);
+	if (!monspecs)
+		return NULL;
+
+	modedb_size = sizeof(struct fb_videomode) * monspecs->modedb_len;
+	modedb = kmemdup(monspecs->modedb, modedb_size, GFP_KERNEL);
+	if (!modedb)
+		goto err_dup_modedb;
+	monspecs->modedb = modedb;
+
+	switch (id) {
+	case 0:
+		pdev = &atmel_lcdfb0_device;
+		select_peripheral(PC(19), PERIPH_A, 0);	/* CC	  */
+		select_peripheral(PC(20), PERIPH_A, 0);	/* HSYNC  */
+		select_peripheral(PC(21), PERIPH_A, 0);	/* PCLK	  */
+		select_peripheral(PC(22), PERIPH_A, 0);	/* VSYNC  */
+		select_peripheral(PC(23), PERIPH_A, 0);	/* DVAL	  */
+		select_peripheral(PC(24), PERIPH_A, 0);	/* MODE	  */
+		select_peripheral(PC(25), PERIPH_A, 0);	/* PWR	  */
+		select_peripheral(PC(26), PERIPH_A, 0);	/* DATA0  */
+		select_peripheral(PC(27), PERIPH_A, 0);	/* DATA1  */
+		select_peripheral(PC(28), PERIPH_A, 0);	/* DATA2  */
+		select_peripheral(PC(29), PERIPH_A, 0);	/* DATA3  */
+		select_peripheral(PC(30), PERIPH_A, 0);	/* DATA4  */
+		select_peripheral(PC(31), PERIPH_A, 0);	/* DATA5  */
+		select_peripheral(PD(0),  PERIPH_A, 0);	/* DATA6  */
+		select_peripheral(PD(1),  PERIPH_A, 0);	/* DATA7  */
+		select_peripheral(PD(2),  PERIPH_A, 0);	/* DATA8  */
+		select_peripheral(PD(3),  PERIPH_A, 0);	/* DATA9  */
+		select_peripheral(PD(4),  PERIPH_A, 0);	/* DATA10 */
+		select_peripheral(PD(5),  PERIPH_A, 0);	/* DATA11 */
+		select_peripheral(PD(6),  PERIPH_A, 0);	/* DATA12 */
+		select_peripheral(PD(7),  PERIPH_A, 0);	/* DATA13 */
+		select_peripheral(PD(8),  PERIPH_A, 0);	/* DATA14 */
+		select_peripheral(PD(9),  PERIPH_A, 0);	/* DATA15 */
+		select_peripheral(PD(10), PERIPH_A, 0);	/* DATA16 */
+		select_peripheral(PD(11), PERIPH_A, 0);	/* DATA17 */
+		select_peripheral(PD(12), PERIPH_A, 0);	/* DATA18 */
+		select_peripheral(PD(13), PERIPH_A, 0);	/* DATA19 */
+		select_peripheral(PD(14), PERIPH_A, 0);	/* DATA20 */
+		select_peripheral(PD(15), PERIPH_A, 0);	/* DATA21 */
+		select_peripheral(PD(16), PERIPH_A, 0);	/* DATA22 */
+		select_peripheral(PD(17), PERIPH_A, 0);	/* DATA23 */
+
+		clk_set_parent(&atmel_lcdfb0_pixclk, &pll0);
+		clk_set_rate(&atmel_lcdfb0_pixclk, clk_get_rate(&pll0));
+		break;
+
+	default:
+		goto err_invalid_id;
+	}
+
+	if (fbmem_len) {
+		pdev->resource[2].start = fbmem_start;
+		pdev->resource[2].end = fbmem_start + fbmem_len - 1;
+		pdev->resource[2].flags = IORESOURCE_MEM;
+	}
+
+	info = pdev->dev.platform_data;
+	memcpy(info, data, sizeof(struct atmel_lcdfb_info));
+	info->default_monspecs = monspecs;
+
+	platform_device_register(pdev);
+	return pdev;
+
+err_invalid_id:
+	kfree(modedb);
+err_dup_modedb:
+	kfree(monspecs);
+	return NULL;
+}
+#endif
+
+/* --------------------------------------------------------------------
+ *  SSC
+ * -------------------------------------------------------------------- */
+static struct resource ssc0_resource[] = {
+	PBMEM(0xffe01c00),
+	IRQ(10),
+};
+DEFINE_DEV(ssc, 0);
+DEV_CLK(pclk, ssc0, pba, 7);
+
+static struct resource ssc1_resource[] = {
+	PBMEM(0xffe02000),
+	IRQ(11),
+};
+DEFINE_DEV(ssc, 1);
+DEV_CLK(pclk, ssc1, pba, 8);
+
+static struct resource ssc2_resource[] = {
+	PBMEM(0xffe02400),
+	IRQ(12),
+};
+DEFINE_DEV(ssc, 2);
+DEV_CLK(pclk, ssc2, pba, 9);
+
+struct platform_device *__init
+at32_add_device_ssc(unsigned int id, unsigned int flags)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &ssc0_device;
+		if (flags & ATMEL_SSC_RF)
+			select_peripheral(PA(21), PERIPH_A, 0);	/* RF */
+		if (flags & ATMEL_SSC_RK)
+			select_peripheral(PA(22), PERIPH_A, 0);	/* RK */
+		if (flags & ATMEL_SSC_TK)
+			select_peripheral(PA(23), PERIPH_A, 0);	/* TK */
+		if (flags & ATMEL_SSC_TF)
+			select_peripheral(PA(24), PERIPH_A, 0);	/* TF */
+		if (flags & ATMEL_SSC_TD)
+			select_peripheral(PA(25), PERIPH_A, 0);	/* TD */
+		if (flags & ATMEL_SSC_RD)
+			select_peripheral(PA(26), PERIPH_A, 0);	/* RD */
+		break;
+	case 1:
+		pdev = &ssc1_device;
+		if (flags & ATMEL_SSC_RF)
+			select_peripheral(PA(0), PERIPH_B, 0);	/* RF */
+		if (flags & ATMEL_SSC_RK)
+			select_peripheral(PA(1), PERIPH_B, 0);	/* RK */
+		if (flags & ATMEL_SSC_TK)
+			select_peripheral(PA(2), PERIPH_B, 0);	/* TK */
+		if (flags & ATMEL_SSC_TF)
+			select_peripheral(PA(3), PERIPH_B, 0);	/* TF */
+		if (flags & ATMEL_SSC_TD)
+			select_peripheral(PA(4), PERIPH_B, 0);	/* TD */
+		if (flags & ATMEL_SSC_RD)
+			select_peripheral(PA(5), PERIPH_B, 0);	/* RD */
+		break;
+	case 2:
+		pdev = &ssc2_device;
+		if (flags & ATMEL_SSC_TD)
+			select_peripheral(PB(13), PERIPH_A, 0);	/* TD */
+		if (flags & ATMEL_SSC_RD)
+			select_peripheral(PB(14), PERIPH_A, 0);	/* RD */
+		if (flags & ATMEL_SSC_TK)
+			select_peripheral(PB(15), PERIPH_A, 0);	/* TK */
+		if (flags & ATMEL_SSC_TF)
+			select_peripheral(PB(16), PERIPH_A, 0);	/* TF */
+		if (flags & ATMEL_SSC_RF)
+			select_peripheral(PB(17), PERIPH_A, 0);	/* RF */
+		if (flags & ATMEL_SSC_RK)
+			select_peripheral(PB(18), PERIPH_A, 0);	/* RK */
+		break;
+	default:
+		return NULL;
+	}
+
+	platform_device_register(pdev);
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  USB Device Controller
+ * -------------------------------------------------------------------- */
+static struct resource usba0_resource[] __initdata = {
+	{
+		.start		= 0xff300000,
+		.end		= 0xff3fffff,
+		.flags		= IORESOURCE_MEM,
+	}, {
+		.start		= 0xfff03000,
+		.end		= 0xfff033ff,
+		.flags		= IORESOURCE_MEM,
+	},
+	IRQ(31),
+};
+static struct clk usba0_pclk = {
+	.name		= "pclk",
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.index		= 12,
+};
+static struct clk usba0_hclk = {
+	.name		= "hclk",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.index		= 6,
+};
+
+struct platform_device *__init
+at32_add_device_usba(unsigned int id, struct usba_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	if (id != 0)
+		return NULL;
+
+	pdev = platform_device_alloc("atmel_usba_udc", 0);
+	if (!pdev)
+		return NULL;
+
+	if (platform_device_add_resources(pdev, usba0_resource,
+					  ARRAY_SIZE(usba0_resource)))
+		goto out_free_pdev;
+
+	if (data) {
+		if (platform_device_add_data(pdev, data, sizeof(*data)))
+			goto out_free_pdev;
+
+		if (data->vbus_pin != GPIO_PIN_NONE)
+			at32_select_gpio(data->vbus_pin, 0);
+	}
+
+	usba0_pclk.dev = &pdev->dev;
+	usba0_hclk.dev = &pdev->dev;
+
+	platform_device_add(pdev);
+
+	return pdev;
+
+out_free_pdev:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+/* --------------------------------------------------------------------
+ * IDE / CompactFlash
+ * -------------------------------------------------------------------- */
+#if defined(CONFIG_CPU_AT32AP7000) || defined(CONFIG_CPU_AT32AP7001)
+static struct resource at32_smc_cs4_resource[] __initdata = {
+	{
+		.start	= 0x04000000,
+		.end	= 0x07ffffff,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(~0UL), /* Magic IRQ will be overridden */
+};
+static struct resource at32_smc_cs5_resource[] __initdata = {
+	{
+		.start	= 0x20000000,
+		.end	= 0x23ffffff,
+		.flags	= IORESOURCE_MEM,
+	},
+	IRQ(~0UL), /* Magic IRQ will be overridden */
+};
+
+static int __init at32_init_ide_or_cf(struct platform_device *pdev,
+		unsigned int cs, unsigned int extint)
+{
+	static unsigned int extint_pin_map[4] __initdata = {
+		GPIO_PIN_PB(25),
+		GPIO_PIN_PB(26),
+		GPIO_PIN_PB(27),
+		GPIO_PIN_PB(28),
+	};
+	static bool common_pins_initialized __initdata = false;
+	unsigned int extint_pin;
+	int ret;
+
+	if (extint >= ARRAY_SIZE(extint_pin_map))
+		return -EINVAL;
+	extint_pin = extint_pin_map[extint];
+
+	switch (cs) {
+	case 4:
+		ret = platform_device_add_resources(pdev,
+				at32_smc_cs4_resource,
+				ARRAY_SIZE(at32_smc_cs4_resource));
+		if (ret)
+			return ret;
+
+		select_peripheral(PE(21), PERIPH_A, 0); /* NCS4   -> OE_N  */
+		set_ebi_sfr_bits(HMATRIX_BIT(CS4A));
+		break;
+	case 5:
+		ret = platform_device_add_resources(pdev,
+				at32_smc_cs5_resource,
+				ARRAY_SIZE(at32_smc_cs5_resource));
+		if (ret)
+			return ret;
+
+		select_peripheral(PE(22), PERIPH_A, 0); /* NCS5   -> OE_N  */
+		set_ebi_sfr_bits(HMATRIX_BIT(CS5A));
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (!common_pins_initialized) {
+		select_peripheral(PE(19), PERIPH_A, 0);	/* CFCE1  -> CS0_N */
+		select_peripheral(PE(20), PERIPH_A, 0);	/* CFCE2  -> CS1_N */
+		select_peripheral(PE(23), PERIPH_A, 0); /* CFRNW  -> DIR   */
+		select_peripheral(PE(24), PERIPH_A, 0); /* NWAIT  <- IORDY */
+		common_pins_initialized = true;
+	}
+
+	at32_select_periph(extint_pin, GPIO_PERIPH_A, AT32_GPIOF_DEGLITCH);
+
+	pdev->resource[1].start = EIM_IRQ_BASE + extint;
+	pdev->resource[1].end = pdev->resource[1].start;
+
+	return 0;
+}
+
+struct platform_device *__init
+at32_add_device_ide(unsigned int id, unsigned int extint,
+		    struct ide_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	pdev = platform_device_alloc("at32_ide", id);
+	if (!pdev)
+		goto fail;
+
+	if (platform_device_add_data(pdev, data,
+				sizeof(struct ide_platform_data)))
+		goto fail;
+
+	if (at32_init_ide_or_cf(pdev, data->cs, extint))
+		goto fail;
+
+	platform_device_add(pdev);
+	return pdev;
+
+fail:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+struct platform_device *__init
+at32_add_device_cf(unsigned int id, unsigned int extint,
+		    struct cf_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	pdev = platform_device_alloc("at32_cf", id);
+	if (!pdev)
+		goto fail;
+
+	if (platform_device_add_data(pdev, data,
+				sizeof(struct cf_platform_data)))
+		goto fail;
+
+	if (at32_init_ide_or_cf(pdev, data->cs, extint))
+		goto fail;
+
+	if (data->detect_pin != GPIO_PIN_NONE)
+		at32_select_gpio(data->detect_pin, AT32_GPIOF_DEGLITCH);
+	if (data->reset_pin != GPIO_PIN_NONE)
+		at32_select_gpio(data->reset_pin, 0);
+	if (data->vcc_pin != GPIO_PIN_NONE)
+		at32_select_gpio(data->vcc_pin, 0);
+	/* READY is used as extint, so we can't select it as gpio */
+
+	platform_device_add(pdev);
+	return pdev;
+
+fail:
+	platform_device_put(pdev);
+	return NULL;
+}
+#endif
+
+/* --------------------------------------------------------------------
+ * AC97C
+ * -------------------------------------------------------------------- */
+static struct resource atmel_ac97c0_resource[] __initdata = {
+	PBMEM(0xfff02800),
+	IRQ(29),
+};
+static struct clk atmel_ac97c0_pclk = {
+	.name		= "pclk",
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.index		= 10,
+};
+
+struct platform_device *__init at32_add_device_ac97c(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	if (id != 0)
+		return NULL;
+
+	pdev = platform_device_alloc("atmel_ac97c", id);
+	if (!pdev)
+		return NULL;
+
+	if (platform_device_add_resources(pdev, atmel_ac97c0_resource,
+				ARRAY_SIZE(atmel_ac97c0_resource)))
+		goto err_add_resources;
+
+	select_peripheral(PB(20), PERIPH_B, 0);	/* SYNC	*/
+	select_peripheral(PB(21), PERIPH_B, 0);	/* SDO	*/
+	select_peripheral(PB(22), PERIPH_B, 0);	/* SDI	*/
+	select_peripheral(PB(23), PERIPH_B, 0);	/* SCLK	*/
+
+	atmel_ac97c0_pclk.dev = &pdev->dev;
+
+	platform_device_add(pdev);
+	return pdev;
+
+err_add_resources:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+/* --------------------------------------------------------------------
+ * ABDAC
+ * -------------------------------------------------------------------- */
+static struct resource abdac0_resource[] __initdata = {
+	PBMEM(0xfff02000),
+	IRQ(27),
+};
+static struct clk abdac0_pclk = {
+	.name		= "pclk",
+	.parent		= &pbb_clk,
+	.mode		= pbb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.index		= 8,
+};
+static struct clk abdac0_sample_clk = {
+	.name		= "sample_clk",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 6,
+};
+
+struct platform_device *__init at32_add_device_abdac(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	if (id != 0)
+		return NULL;
+
+	pdev = platform_device_alloc("abdac", id);
+	if (!pdev)
+		return NULL;
+
+	if (platform_device_add_resources(pdev, abdac0_resource,
+				ARRAY_SIZE(abdac0_resource)))
+		goto err_add_resources;
+
+	select_peripheral(PB(20), PERIPH_A, 0);	/* DATA1	*/
+	select_peripheral(PB(21), PERIPH_A, 0);	/* DATA0	*/
+	select_peripheral(PB(22), PERIPH_A, 0);	/* DATAN1	*/
+	select_peripheral(PB(23), PERIPH_A, 0);	/* DATAN0	*/
+
+	abdac0_pclk.dev = &pdev->dev;
+	abdac0_sample_clk.dev = &pdev->dev;
+
+	platform_device_add(pdev);
+	return pdev;
+
+err_add_resources:
+	platform_device_put(pdev);
+	return NULL;
+}
+
+/* --------------------------------------------------------------------
+ *  GCLK
+ * -------------------------------------------------------------------- */
+static struct clk gclk0 = {
+	.name		= "gclk0",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 0,
+};
+static struct clk gclk1 = {
+	.name		= "gclk1",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 1,
+};
+static struct clk gclk2 = {
+	.name		= "gclk2",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 2,
+};
+static struct clk gclk3 = {
+	.name		= "gclk3",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 3,
+};
+static struct clk gclk4 = {
+	.name		= "gclk4",
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 4,
+};
+
+struct clk *at32_clock_list[] = {
+	&osc32k,
+	&osc0,
+	&osc1,
+	&pll0,
+	&pll1,
+	&cpu_clk,
+	&hsb_clk,
+	&pba_clk,
+	&pbb_clk,
+	&at32_pm_pclk,
+	&at32_intc0_pclk,
+	&hmatrix_clk,
+	&ebi_clk,
+	&hramc_clk,
+	&smc0_pclk,
+	&smc0_mck,
+	&pdc_hclk,
+	&pdc_pclk,
+	&dmaca0_hclk,
+	&pico_clk,
+	&pio0_mck,
+	&pio1_mck,
+	&pio2_mck,
+	&pio3_mck,
+	&pio4_mck,
+	&at32_systc0_pclk,
+	&atmel_usart0_usart,
+	&atmel_usart1_usart,
+	&atmel_usart2_usart,
+	&atmel_usart3_usart,
+#if defined(CONFIG_CPU_AT32AP7000)
+	&macb0_hclk,
+	&macb0_pclk,
+	&macb1_hclk,
+	&macb1_pclk,
+#endif
+	&atmel_spi0_spi_clk,
+	&atmel_spi1_spi_clk,
+	&atmel_twi0_pclk,
+	&atmel_mci0_pclk,
+#if defined(CONFIG_CPU_AT32AP7000) || defined(CONFIG_CPU_AT32AP7002)
+	&atmel_lcdfb0_hck1,
+	&atmel_lcdfb0_pixclk,
+#endif
+	&ssc0_pclk,
+	&ssc1_pclk,
+	&ssc2_pclk,
+	&usba0_hclk,
+	&usba0_pclk,
+	&atmel_ac97c0_pclk,
+	&abdac0_pclk,
+	&abdac0_sample_clk,
+	&gclk0,
+	&gclk1,
+	&gclk2,
+	&gclk3,
+	&gclk4,
+};
+unsigned int at32_nr_clocks = ARRAY_SIZE(at32_clock_list);
+
+void __init at32_portmux_init(void)
+{
+	at32_init_pio(&pio0_device);
+	at32_init_pio(&pio1_device);
+	at32_init_pio(&pio2_device);
+	at32_init_pio(&pio3_device);
+	at32_init_pio(&pio4_device);
+}
+
+void __init at32_clock_init(void)
+{
+	u32 cpu_mask = 0, hsb_mask = 0, pba_mask = 0, pbb_mask = 0;
+	int i;
+
+	if (pm_readl(MCCTRL) & PM_BIT(PLLSEL)) {
+		main_clock = &pll0;
+		cpu_clk.parent = &pll0;
+	} else {
+		main_clock = &osc0;
+		cpu_clk.parent = &osc0;
+	}
+
+	if (pm_readl(PLL0) & PM_BIT(PLLOSC))
+		pll0.parent = &osc1;
+	if (pm_readl(PLL1) & PM_BIT(PLLOSC))
+		pll1.parent = &osc1;
+
+	genclk_init_parent(&gclk0);
+	genclk_init_parent(&gclk1);
+	genclk_init_parent(&gclk2);
+	genclk_init_parent(&gclk3);
+	genclk_init_parent(&gclk4);
+#if defined(CONFIG_CPU_AT32AP7000) || defined(CONFIG_CPU_AT32AP7002)
+	genclk_init_parent(&atmel_lcdfb0_pixclk);
+#endif
+	genclk_init_parent(&abdac0_sample_clk);
+
+	/*
+	 * Turn on all clocks that have at least one user already, and
+	 * turn off everything else. We only do this for module
+	 * clocks, and even though it isn't particularly pretty to
+	 * check the address of the mode function, it should do the
+	 * trick...
+	 */
+	for (i = 0; i < ARRAY_SIZE(at32_clock_list); i++) {
+		struct clk *clk = at32_clock_list[i];
+
+		if (clk->users == 0)
+			continue;
+
+		if (clk->mode == &cpu_clk_mode)
+			cpu_mask |= 1 << clk->index;
+		else if (clk->mode == &hsb_clk_mode)
+			hsb_mask |= 1 << clk->index;
+		else if (clk->mode == &pba_clk_mode)
+			pba_mask |= 1 << clk->index;
+		else if (clk->mode == &pbb_clk_mode)
+			pbb_mask |= 1 << clk->index;
+	}
+
+	pm_writel(CPU_MASK, cpu_mask);
+	pm_writel(HSB_MASK, hsb_mask);
+	pm_writel(PBA_MASK, pba_mask);
+	pm_writel(PBB_MASK, pbb_mask);
+}
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 52dff40ec192..fbd61127b9d9 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -223,7 +223,7 @@ config DAVINCI_WATCHDOG
 
 config AT32AP700X_WDT
 	tristate "AT32AP700x watchdog"
-	depends on CPU_AT32AP7000
+	depends on CPU_AT32AP700X
 	help
 	  Watchdog timer embedded into AT32AP700x devices. This will reboot
 	  your system when the timeout is reached.
diff --git a/include/asm-avr32/arch-at32ap/at32ap7000.h b/include/asm-avr32/arch-at32ap/at32ap7000.h
deleted file mode 100644
index 3914d7b94ff4..000000000000
--- a/include/asm-avr32/arch-at32ap/at32ap7000.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * Pin definitions for AT32AP7000.
- *
- * Copyright (C) 2006 Atmel Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#ifndef __ASM_ARCH_AT32AP7000_H__
-#define __ASM_ARCH_AT32AP7000_H__
-
-#define GPIO_PERIPH_A	0
-#define GPIO_PERIPH_B	1
-
-#define NR_GPIO_CONTROLLERS	4
-
-/*
- * Pin numbers identifying specific GPIO pins on the chip. They can
- * also be converted to IRQ numbers by passing them through
- * gpio_to_irq().
- */
-#define GPIO_PIOA_BASE	(0)
-#define GPIO_PIOB_BASE	(GPIO_PIOA_BASE + 32)
-#define GPIO_PIOC_BASE	(GPIO_PIOB_BASE + 32)
-#define GPIO_PIOD_BASE	(GPIO_PIOC_BASE + 32)
-#define GPIO_PIOE_BASE	(GPIO_PIOD_BASE + 32)
-
-#define GPIO_PIN_PA(N)	(GPIO_PIOA_BASE + (N))
-#define GPIO_PIN_PB(N)	(GPIO_PIOB_BASE + (N))
-#define GPIO_PIN_PC(N)	(GPIO_PIOC_BASE + (N))
-#define GPIO_PIN_PD(N)	(GPIO_PIOD_BASE + (N))
-#define GPIO_PIN_PE(N)	(GPIO_PIOE_BASE + (N))
-
-#endif /* __ASM_ARCH_AT32AP7000_H__ */
diff --git a/include/asm-avr32/arch-at32ap/at32ap700x.h b/include/asm-avr32/arch-at32ap/at32ap700x.h
new file mode 100644
index 000000000000..99684d6f3967
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/at32ap700x.h
@@ -0,0 +1,35 @@
+/*
+ * Pin definitions for AT32AP7000.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_ARCH_AT32AP700X_H__
+#define __ASM_ARCH_AT32AP700X_H__
+
+#define GPIO_PERIPH_A	0
+#define GPIO_PERIPH_B	1
+
+#define NR_GPIO_CONTROLLERS	4
+
+/*
+ * Pin numbers identifying specific GPIO pins on the chip. They can
+ * also be converted to IRQ numbers by passing them through
+ * gpio_to_irq().
+ */
+#define GPIO_PIOA_BASE	(0)
+#define GPIO_PIOB_BASE	(GPIO_PIOA_BASE + 32)
+#define GPIO_PIOC_BASE	(GPIO_PIOB_BASE + 32)
+#define GPIO_PIOD_BASE	(GPIO_PIOC_BASE + 32)
+#define GPIO_PIOE_BASE	(GPIO_PIOD_BASE + 32)
+
+#define GPIO_PIN_PA(N)	(GPIO_PIOA_BASE + (N))
+#define GPIO_PIN_PB(N)	(GPIO_PIOB_BASE + (N))
+#define GPIO_PIN_PC(N)	(GPIO_PIOC_BASE + (N))
+#define GPIO_PIN_PD(N)	(GPIO_PIOD_BASE + (N))
+#define GPIO_PIN_PE(N)	(GPIO_PIOE_BASE + (N))
+
+#endif /* __ASM_ARCH_AT32AP700X_H__ */
diff --git a/include/asm-avr32/arch-at32ap/cpu.h b/include/asm-avr32/arch-at32ap/cpu.h
index a762f42cbb71..0dc20261c1ea 100644
--- a/include/asm-avr32/arch-at32ap/cpu.h
+++ b/include/asm-avr32/arch-at32ap/cpu.h
@@ -14,7 +14,7 @@
  * Only AT32AP7000 is defined for now. We can identify the specific
  * chip at runtime, but I'm not sure if it's really worth it.
  */
-#ifdef CONFIG_CPU_AT32AP7000
+#ifdef CONFIG_CPU_AT32AP700X
 # define cpu_is_at32ap7000()	(1)
 #else
 # define cpu_is_at32ap7000()	(0)
diff --git a/include/asm-avr32/arch-at32ap/io.h b/include/asm-avr32/arch-at32ap/io.h
index ee59e401f041..4ec6abc68ea3 100644
--- a/include/asm-avr32/arch-at32ap/io.h
+++ b/include/asm-avr32/arch-at32ap/io.h
@@ -4,7 +4,7 @@
 /* For "bizarre" halfword swapping */
 #include <linux/byteorder/swabb.h>
 
-#if defined(CONFIG_AP7000_32_BIT_SMC)
+#if defined(CONFIG_AP700X_32_BIT_SMC)
 # define __swizzle_addr_b(addr)	(addr ^ 3UL)
 # define __swizzle_addr_w(addr)	(addr ^ 2UL)
 # define __swizzle_addr_l(addr)	(addr)
@@ -14,7 +14,7 @@
 # define __mem_ioswabb(a, x)	(x)
 # define __mem_ioswabw(a, x)	swab16(x)
 # define __mem_ioswabl(a, x)	swab32(x)
-#elif defined(CONFIG_AP7000_16_BIT_SMC)
+#elif defined(CONFIG_AP700X_16_BIT_SMC)
 # define __swizzle_addr_b(addr)	(addr ^ 1UL)
 # define __swizzle_addr_w(addr)	(addr)
 # define __swizzle_addr_l(addr)	(addr)
-- 
cgit v1.2.3


From e7ba176b47db2ed53f258a6b4fe9d9fc6fa437a9 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Wed, 10 Oct 2007 14:58:29 +0200
Subject: [AVR32] NMI debugging

Change the NMI handler to use the die notifier chain to signal anyone
who cares. Add a simple "nmi debugger" which hooks into this chain and
that may dump registers, task state, etc. when it happens.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
---
 Documentation/kernel-parameters.txt |  5 +++
 arch/avr32/Kconfig                  | 10 +++++
 arch/avr32/kernel/Makefile          |  1 +
 arch/avr32/kernel/irq.c             | 11 +++++
 arch/avr32/kernel/nmi_debug.c       | 82 +++++++++++++++++++++++++++++++++++++
 arch/avr32/kernel/traps.c           | 21 ++++++++--
 arch/avr32/mach-at32ap/at32ap700x.c |  1 +
 arch/avr32/mach-at32ap/extint.c     | 39 ++++++++++++++----
 include/asm-avr32/irq.h             |  5 +++
 include/asm-avr32/kdebug.h          |  1 +
 10 files changed, 166 insertions(+), 10 deletions(-)
 create mode 100644 arch/avr32/kernel/nmi_debug.c

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index c4178778e7fd..17fc60e32443 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -34,6 +34,7 @@ parameter is applicable:
 	ALSA	ALSA sound support is enabled.
 	APIC	APIC support is enabled.
 	APM	Advanced Power Management support is enabled.
+	AVR32	AVR32 architecture is enabled.
 	AX25	Appropriate AX.25 support is enabled.
 	BLACKFIN Blackfin architecture is enabled.
 	DRM	Direct Rendering Management support is enabled.
@@ -1123,6 +1124,10 @@ and is between 256 and 4096 characters. It is defined in the file
 			of returning the full 64-bit number.
 			The default is to return 64-bit inode numbers.
 
+	nmi_debug=	[KNL,AVR32] Specify one or more actions to take
+			when a NMI is triggered.
+			Format: [state][,regs][,debounce][,die]
+
 	nmi_watchdog=	[KNL,BUGS=X86-32] Debugging features for SMP kernels
 
 	no387		[BUGS=X86-32] Tells the kernel to use the 387 maths
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index 516015b3293b..e34e2c9c94cb 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -170,6 +170,16 @@ config OWNERSHIP_TRACE
 	  enabling Nexus-compliant debuggers to keep track of the PID of the
 	  currently executing task.
 
+config NMI_DEBUGGING
+	bool "NMI Debugging"
+	default n
+	help
+	  Say Y here and pass the nmi_debug command-line parameter to
+	  the kernel to turn on NMI debugging. Depending on the value
+	  of the nmi_debug option, various pieces of information will
+	  be dumped to the console when a Non-Maskable Interrupt
+	  happens.
+
 # FPU emulation goes here
 
 source "kernel/Kconfig.hz"
diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile
index bc224a4e39fe..e4b6d122b033 100644
--- a/arch/avr32/kernel/Makefile
+++ b/arch/avr32/kernel/Makefile
@@ -12,3 +12,4 @@ obj-y				+= init_task.o switch_to.o cpu.o
 obj-$(CONFIG_MODULES)		+= module.o avr32_ksyms.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+obj-$(CONFIG_NMI_DEBUGGING)	+= nmi_debug.o
diff --git a/arch/avr32/kernel/irq.c b/arch/avr32/kernel/irq.c
index 61f2de266f62..a8e767d836aa 100644
--- a/arch/avr32/kernel/irq.c
+++ b/arch/avr32/kernel/irq.c
@@ -25,6 +25,17 @@ void ack_bad_irq(unsigned int irq)
 	printk("unexpected IRQ %u\n", irq);
 }
 
+/* May be overridden by platform code */
+int __weak nmi_enable(void)
+{
+	return -ENOSYS;
+}
+
+void __weak nmi_disable(void)
+{
+
+}
+
 #ifdef CONFIG_PROC_FS
 int show_interrupts(struct seq_file *p, void *v)
 {
diff --git a/arch/avr32/kernel/nmi_debug.c b/arch/avr32/kernel/nmi_debug.c
new file mode 100644
index 000000000000..3414b8566c29
--- /dev/null
+++ b/arch/avr32/kernel/nmi_debug.c
@@ -0,0 +1,82 @@
+/*
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/delay.h>
+#include <linux/kdebug.h>
+#include <linux/notifier.h>
+#include <linux/sched.h>
+
+#include <asm/irq.h>
+
+enum nmi_action {
+	NMI_SHOW_STATE	= 1 << 0,
+	NMI_SHOW_REGS	= 1 << 1,
+	NMI_DIE		= 1 << 2,
+	NMI_DEBOUNCE	= 1 << 3,
+};
+
+static unsigned long nmi_actions;
+
+static int nmi_debug_notify(struct notifier_block *self,
+		unsigned long val, void *data)
+{
+	struct die_args *args = data;
+
+	if (likely(val != DIE_NMI))
+		return NOTIFY_DONE;
+
+	if (nmi_actions & NMI_SHOW_STATE)
+		show_state();
+	if (nmi_actions & NMI_SHOW_REGS)
+		show_regs(args->regs);
+	if (nmi_actions & NMI_DEBOUNCE)
+		mdelay(10);
+	if (nmi_actions & NMI_DIE)
+		return NOTIFY_BAD;
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block nmi_debug_nb = {
+	.notifier_call = nmi_debug_notify,
+};
+
+static int __init nmi_debug_setup(char *str)
+{
+	char *p, *sep;
+
+	register_die_notifier(&nmi_debug_nb);
+	if (nmi_enable()) {
+		printk(KERN_WARNING "Unable to enable NMI.\n");
+		return 0;
+	}
+
+	if (*str != '=')
+		return 0;
+
+	for (p = str + 1; *p; p = sep + 1) {
+		sep = strchr(p, ',');
+		if (sep)
+			*sep = 0;
+		if (strcmp(p, "state") == 0)
+			nmi_actions |= NMI_SHOW_STATE;
+		else if (strcmp(p, "regs") == 0)
+			nmi_actions |= NMI_SHOW_REGS;
+		else if (strcmp(p, "debounce") == 0)
+			nmi_actions |= NMI_DEBOUNCE;
+		else if (strcmp(p, "die") == 0)
+			nmi_actions |= NMI_DIE;
+		else
+			printk(KERN_WARNING "NMI: Unrecognized action `%s'\n",
+				p);
+		if (!sep)
+			break;
+	}
+
+	return 0;
+}
+__setup("nmi_debug", nmi_debug_setup);
diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c
index 870c075e6314..cf6f686d9b0b 100644
--- a/arch/avr32/kernel/traps.c
+++ b/arch/avr32/kernel/traps.c
@@ -9,6 +9,7 @@
 #include <linux/bug.h>
 #include <linux/init.h>
 #include <linux/kallsyms.h>
+#include <linux/kdebug.h>
 #include <linux/module.h>
 #include <linux/notifier.h>
 #include <linux/sched.h>
@@ -107,9 +108,23 @@ void _exception(long signr, struct pt_regs *regs, int code,
 
 asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs)
 {
-	printk(KERN_ALERT "Got Non-Maskable Interrupt, dumping regs\n");
-	show_regs_log_lvl(regs, KERN_ALERT);
-	show_stack_log_lvl(current, regs->sp, regs, KERN_ALERT);
+	int ret;
+
+	nmi_enter();
+
+	ret = notify_die(DIE_NMI, "NMI", regs, 0, ecr, SIGINT);
+	switch (ret) {
+	case NOTIFY_OK:
+	case NOTIFY_STOP:
+		return;
+	case NOTIFY_BAD:
+		die("Fatal Non-Maskable Interrupt", regs, SIGINT);
+	default:
+		break;
+	}
+
+	printk(KERN_ALERT "Got NMI, but nobody cared. Disabling...\n");
+	nmi_disable();
 }
 
 asmlinkage void do_critical_exception(unsigned long ecr, struct pt_regs *regs)
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 9386e1f82fb8..14e61f05e1f6 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -13,6 +13,7 @@
 #include <linux/spi/spi.h>
 
 #include <asm/io.h>
+#include <asm/irq.h>
 
 #include <asm/arch/at32ap700x.h>
 #include <asm/arch/board.h>
diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c
index f5bfd4c81fe7..e108e7bba8c0 100644
--- a/arch/avr32/mach-at32ap/extint.c
+++ b/arch/avr32/mach-at32ap/extint.c
@@ -26,16 +26,10 @@
 #define EIC_MODE				0x0014
 #define EIC_EDGE				0x0018
 #define EIC_LEVEL				0x001c
-#define EIC_TEST				0x0020
 #define EIC_NMIC				0x0024
 
-/* Bitfields in TEST */
-#define EIC_TESTEN_OFFSET			31
-#define EIC_TESTEN_SIZE				1
-
 /* Bitfields in NMIC */
-#define EIC_EN_OFFSET				0
-#define EIC_EN_SIZE				1
+#define EIC_NMIC_ENABLE				(1 << 0)
 
 /* Bit manipulation macros */
 #define EIC_BIT(name)					\
@@ -63,6 +57,9 @@ struct eic {
 	unsigned int first_irq;
 };
 
+static struct eic *nmi_eic;
+static bool nmi_enabled;
+
 static void eic_ack_irq(unsigned int irq)
 {
 	struct eic *eic = get_irq_chip_data(irq);
@@ -174,6 +171,24 @@ static void demux_eic_irq(unsigned int irq, struct irq_desc *desc)
 	}
 }
 
+int nmi_enable(void)
+{
+	nmi_enabled = true;
+
+	if (nmi_eic)
+		eic_writel(nmi_eic, NMIC, EIC_NMIC_ENABLE);
+
+	return 0;
+}
+
+void nmi_disable(void)
+{
+	if (nmi_eic)
+		eic_writel(nmi_eic, NMIC, 0);
+
+	nmi_enabled = false;
+}
+
 static int __init eic_probe(struct platform_device *pdev)
 {
 	struct eic *eic;
@@ -230,6 +245,16 @@ static int __init eic_probe(struct platform_device *pdev)
 	set_irq_chained_handler(int_irq, demux_eic_irq);
 	set_irq_data(int_irq, eic);
 
+	if (pdev->id == 0) {
+		nmi_eic = eic;
+		if (nmi_enabled)
+			/*
+			 * Someone tried to enable NMI before we were
+			 * ready. Do it now.
+			 */
+			nmi_enable();
+	}
+
 	dev_info(&pdev->dev,
 		 "External Interrupt Controller at 0x%p, IRQ %u\n",
 		 eic->regs, int_irq);
diff --git a/include/asm-avr32/irq.h b/include/asm-avr32/irq.h
index 83e6549d7783..9315724c0596 100644
--- a/include/asm-avr32/irq.h
+++ b/include/asm-avr32/irq.h
@@ -11,4 +11,9 @@
 
 #define irq_canonicalize(i)	(i)
 
+#ifndef __ASSEMBLER__
+int nmi_enable(void);
+void nmi_disable(void);
+#endif
+
 #endif /* __ASM_AVR32_IOCTLS_H */
diff --git a/include/asm-avr32/kdebug.h b/include/asm-avr32/kdebug.h
index fd7e99046b2f..ca4f9542365a 100644
--- a/include/asm-avr32/kdebug.h
+++ b/include/asm-avr32/kdebug.h
@@ -5,6 +5,7 @@
 enum die_val {
 	DIE_BREAKPOINT,
 	DIE_SSTEP,
+	DIE_NMI,
 };
 
 #endif /* __ASM_AVR32_KDEBUG_H */
-- 
cgit v1.2.3


From 366ca51f30de1cbb5b356c70b7bb22051c558e41 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Fri, 18 Jan 2008 10:47:01 -0600
Subject: [SCSI] libsas: abstract STP task status into a function

Break out the frame processor for STP tasks from aic94xx so they can
be shared by other SAS HBA's

Original patch from Jeff Garzik <jeff@garzik.org>

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/aic94xx/aic94xx_task.c | 20 ++------------------
 drivers/scsi/libsas/Makefile        |  3 ++-
 drivers/scsi/libsas/sas_task.c      | 36 ++++++++++++++++++++++++++++++++++++
 include/scsi/libsas.h               |  4 ++++
 4 files changed, 44 insertions(+), 19 deletions(-)
 create mode 100644 drivers/scsi/libsas/sas_task.c

(limited to 'include')

diff --git a/drivers/scsi/aic94xx/aic94xx_task.c b/drivers/scsi/aic94xx/aic94xx_task.c
index 663fdef38c52..965d4bb999d9 100644
--- a/drivers/scsi/aic94xx/aic94xx_task.c
+++ b/drivers/scsi/aic94xx/aic94xx_task.c
@@ -192,24 +192,8 @@ static void asd_get_response_tasklet(struct asd_ascb *ascb,
 			r + 16 + sizeof(struct ssp_frame_hdr);
 
 		ts->residual = le32_to_cpu(*(__le32 *)r);
-		ts->resp = SAS_TASK_COMPLETE;
-		if (iu->datapres == 0)
-			ts->stat = iu->status;
-		else if (iu->datapres == 1)
-			ts->stat = iu->resp_data[3];
-		else if (iu->datapres == 2) {
-			ts->stat = SAM_CHECK_COND;
-			ts->buf_valid_size = min((u32) SAS_STATUS_BUF_SIZE,
-					 be32_to_cpu(iu->sense_data_len));
-			memcpy(ts->buf, iu->sense_data, ts->buf_valid_size);
-			if (iu->status != SAM_CHECK_COND) {
-				ASD_DPRINTK("device %llx sent sense data, but "
-					    "stat(0x%x) is not CHECK_CONDITION"
-					    "\n",
-					    SAS_ADDR(task->dev->sas_addr),
-					    iu->status);
-			}
-		}
+
+		sas_ssp_task_response(&asd_ha->pcidev->dev, task, iu);
 	}  else {
 		struct ata_task_resp *resp = (void *) &ts->buf[0];
 
diff --git a/drivers/scsi/libsas/Makefile b/drivers/scsi/libsas/Makefile
index 60d6e93dd949..1ad1323c60fa 100644
--- a/drivers/scsi/libsas/Makefile
+++ b/drivers/scsi/libsas/Makefile
@@ -33,6 +33,7 @@ libsas-y +=  sas_init.o     \
 		sas_dump.o     \
 		sas_discover.o \
 		sas_expander.o \
-		sas_scsi_host.o
+		sas_scsi_host.o \
+		sas_task.o
 libsas-$(CONFIG_SCSI_SAS_ATA) +=	sas_ata.o
 libsas-$(CONFIG_SCSI_SAS_HOST_SMP) +=	sas_host_smp.o
\ No newline at end of file
diff --git a/drivers/scsi/libsas/sas_task.c b/drivers/scsi/libsas/sas_task.c
new file mode 100644
index 000000000000..594524d5bfa1
--- /dev/null
+++ b/drivers/scsi/libsas/sas_task.c
@@ -0,0 +1,36 @@
+#include <linux/kernel.h>
+#include <scsi/sas.h>
+#include <scsi/libsas.h>
+
+/* fill task_status_struct based on SSP response frame */
+void sas_ssp_task_response(struct device *dev, struct sas_task *task,
+			   struct ssp_response_iu *iu)
+{
+	struct task_status_struct *tstat = &task->task_status;
+
+	tstat->resp = SAS_TASK_COMPLETE;
+
+	if (iu->datapres == 0)
+		tstat->stat = iu->status;
+	else if (iu->datapres == 1)
+		tstat->stat = iu->resp_data[3];
+	else if (iu->datapres == 2) {
+		tstat->stat = SAM_CHECK_COND;
+		tstat->buf_valid_size =
+			min_t(int, SAS_STATUS_BUF_SIZE,
+			      be32_to_cpu(iu->sense_data_len));
+		memcpy(tstat->buf, iu->sense_data, tstat->buf_valid_size);
+
+		if (iu->status != SAM_CHECK_COND)
+			dev_printk(KERN_WARNING, dev,
+				   "dev %llx sent sense data, but "
+				   "stat(%x) is not CHECK CONDITION\n",
+				   SAS_ADDR(task->dev->sas_addr),
+				   iu->status);
+	}
+	else
+		/* when datapres contains corrupt/unknown value... */
+		tstat->stat = SAM_CHECK_COND;
+}
+EXPORT_SYMBOL_GPL(sas_ssp_task_response);
+
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index a075f130303c..3ffd6b582a97 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -672,4 +672,8 @@ extern int sas_ioctl(struct scsi_device *sdev, int cmd, void __user *arg);
 
 extern int sas_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
 			   struct request *req);
+
+extern void sas_ssp_task_response(struct device *dev, struct sas_task *task,
+				  struct ssp_response_iu *iu);
+
 #endif /* _SASLIB_H_ */
-- 
cgit v1.2.3


From 6b2d7700266b9402e12824e11e0099ae6a4a6a79 Mon Sep 17 00:00:00 2001
From: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Date: Fri, 25 Jan 2008 21:08:00 +0100
Subject: sched: group scheduler, fix fairness of cpu bandwidth allocation for
 task groups

The current load balancing scheme isn't good enough for precise
group fairness.

For example: on a 8-cpu system, I created 3 groups as under:

	a = 8 tasks (cpu.shares = 1024)
	b = 4 tasks (cpu.shares = 1024)
	c = 3 tasks (cpu.shares = 1024)

a, b and c are task groups that have equal weight. We would expect each
of the groups to receive 33.33% of cpu bandwidth under a fair scheduler.

This is what I get with the latest scheduler git tree:

Signed-off-by: Ingo Molnar <mingo@elte.hu>
--------------------------------------------------------------------------------
Col1  | Col2    | Col3  |  Col4
------|---------|-------|-------------------------------------------------------
a     | 277.676 | 57.8% | 54.1%  54.1%  54.1%  54.2%  56.7%  62.2%  62.8% 64.5%
b     | 116.108 | 24.2% | 47.4%  48.1%  48.7%  49.3%
c     |  86.326 | 18.0% | 47.5%  47.9%  48.5%
--------------------------------------------------------------------------------

Explanation of o/p:

Col1 -> Group name
Col2 -> Cumulative execution time (in seconds) received by all tasks of that
	group in a 60sec window across 8 cpus
Col3 -> CPU bandwidth received by the group in the 60sec window, expressed in
        percentage. Col3 data is derived as:
		Col3 = 100 * Col2 / (NR_CPUS * 60)
Col4 -> CPU bandwidth received by each individual task of the group.
		Col4 = 100 * cpu_time_recd_by_task / 60

[I can share the test case that produces a similar o/p if reqd]

The deviation from desired group fairness is as below:

	a = +24.47%
	b = -9.13%
	c = -15.33%

which is quite high.

After the patch below is applied, here are the results:

--------------------------------------------------------------------------------
Col1  | Col2    | Col3  |  Col4
------|---------|-------|-------------------------------------------------------
a     | 163.112 | 34.0% | 33.2%  33.4%  33.5%  33.5%  33.7%  34.4%  34.8% 35.3%
b     | 156.220 | 32.5% | 63.3%  64.5%  66.1%  66.5%
c     | 160.653 | 33.5% | 85.8%  90.6%  91.4%
--------------------------------------------------------------------------------

Deviation from desired group fairness is as below:

	a = +0.67%
	b = -0.83%
	c = +0.17%

which is far better IMO. Most of other runs have yielded a deviation within
+-2% at the most, which is good.

Why do we see bad (group) fairness with current scheuler?
=========================================================

Currently cpu's weight is just the summation of individual task weights.
This can yield incorrect results. For ex: consider three groups as below
on a 2-cpu system:

	CPU0	CPU1
---------------------------
	A (10)  B(5)
		C(5)
---------------------------

Group A has 10 tasks, all on CPU0, Group B and C have 5 tasks each all
of which are on CPU1. Each task has the same weight (NICE_0_LOAD =
1024).

The current scheme would yield a cpu weight of 10240 (10*1024) for each cpu and
the load balancer will think both CPUs are perfectly balanced and won't
move around any tasks. This, however, would yield this bandwidth:

	A = 50%
	B = 25%
	C = 25%

which is not the desired result.

What's changing in the patch?
=============================

	- How cpu weights are calculated when CONFIF_FAIR_GROUP_SCHED is
	  defined (see below)
	- API Change
		- Two tunables introduced in sysfs (under SCHED_DEBUG) to
		  control the frequency at which the load balance monitor
		  thread runs.

The basic change made in this patch is how cpu weight (rq->load.weight) is
calculated. Its now calculated as the summation of group weights on a cpu,
rather than summation of task weights. Weight exerted by a group on a
cpu is dependent on the shares allocated to it and also the number of
tasks the group has on that cpu compared to the total number of
(runnable) tasks the group has in the system.

Let,
	W(K,i)  = Weight of group K on cpu i
	T(K,i)  = Task load present in group K's cfs_rq on cpu i
	T(K)    = Total task load of group K across various cpus
	S(K) 	= Shares allocated to group K
	NRCPUS	= Number of online cpus in the scheduler domain to
	 	  which group K is assigned.

Then,
	W(K,i) = S(K) * NRCPUS * T(K,i) / T(K)

A load balance monitor thread is created at bootup, which periodically
runs and adjusts group's weight on each cpu. To avoid its overhead, two
min/max tunables are introduced (under SCHED_DEBUG) to control the rate
at which it runs.

Fixes from: Peter Zijlstra <a.p.zijlstra@chello.nl>

- don't start the load_balance_monitor when there is only a single cpu.
- rename the kthread because its currently longer than TASK_COMM_LEN

Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |   4 +
 kernel/sched.c        | 270 +++++++++++++++++++++++++++++++++++++++++++++++---
 kernel/sched_fair.c   |  84 ++++++++++------
 kernel/sysctl.c       |  18 ++++
 4 files changed, 331 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index d6eacda765ca..288245f83bd4 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1453,6 +1453,10 @@ extern unsigned int sysctl_sched_child_runs_first;
 extern unsigned int sysctl_sched_features;
 extern unsigned int sysctl_sched_migration_cost;
 extern unsigned int sysctl_sched_nr_migrate;
+#if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
+extern unsigned int sysctl_sched_min_bal_int_shares;
+extern unsigned int sysctl_sched_max_bal_int_shares;
+#endif
 
 int sched_nr_latency_handler(struct ctl_table *table, int write,
 		struct file *file, void __user *buffer, size_t *length,
diff --git a/kernel/sched.c b/kernel/sched.c
index d9585f15043f..86e55a9c2de6 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -168,7 +168,43 @@ struct task_group {
 	struct sched_entity **se;
 	/* runqueue "owned" by this group on each cpu */
 	struct cfs_rq **cfs_rq;
+
+	/*
+	 * shares assigned to a task group governs how much of cpu bandwidth
+	 * is allocated to the group. The more shares a group has, the more is
+	 * the cpu bandwidth allocated to it.
+	 *
+	 * For ex, lets say that there are three task groups, A, B and C which
+	 * have been assigned shares 1000, 2000 and 3000 respectively. Then,
+	 * cpu bandwidth allocated by the scheduler to task groups A, B and C
+	 * should be:
+	 *
+	 *	Bw(A) = 1000/(1000+2000+3000) * 100 = 16.66%
+	 *	Bw(B) = 2000/(1000+2000+3000) * 100 = 33.33%
+	 * 	Bw(C) = 3000/(1000+2000+3000) * 100 = 50%
+	 *
+	 * The weight assigned to a task group's schedulable entities on every
+	 * cpu (task_group.se[a_cpu]->load.weight) is derived from the task
+	 * group's shares. For ex: lets say that task group A has been
+	 * assigned shares of 1000 and there are two CPUs in a system. Then,
+	 *
+	 *  tg_A->se[0]->load.weight = tg_A->se[1]->load.weight = 1000;
+	 *
+	 * Note: It's not necessary that each of a task's group schedulable
+	 * 	 entity have the same weight on all CPUs. If the group
+	 * 	 has 2 of its tasks on CPU0 and 1 task on CPU1, then a
+	 * 	 better distribution of weight could be:
+	 *
+	 *	tg_A->se[0]->load.weight = 2/3 * 2000 = 1333
+	 *	tg_A->se[1]->load.weight = 1/2 * 2000 =  667
+	 *
+	 * rebalance_shares() is responsible for distributing the shares of a
+	 * task groups like this among the group's schedulable entities across
+	 * cpus.
+	 *
+	 */
 	unsigned long shares;
+
 	struct rcu_head rcu;
 };
 
@@ -188,6 +224,14 @@ static DEFINE_MUTEX(task_group_mutex);
 /* doms_cur_mutex serializes access to doms_cur[] array */
 static DEFINE_MUTEX(doms_cur_mutex);
 
+#ifdef CONFIG_SMP
+/* kernel thread that runs rebalance_shares() periodically */
+static struct task_struct *lb_monitor_task;
+static int load_balance_monitor(void *unused);
+#endif
+
+static void set_se_shares(struct sched_entity *se, unsigned long shares);
+
 /* Default task group.
  *	Every task in system belong to this group at bootup.
  */
@@ -202,6 +246,8 @@ struct task_group init_task_group = {
 # define INIT_TASK_GROUP_LOAD	NICE_0_LOAD
 #endif
 
+#define MIN_GROUP_SHARES       2
+
 static int init_task_group_load = INIT_TASK_GROUP_LOAD;
 
 /* return group to which a task belongs */
@@ -6736,6 +6782,21 @@ void __init sched_init_smp(void)
 	if (set_cpus_allowed(current, non_isolated_cpus) < 0)
 		BUG();
 	sched_init_granularity();
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	if (nr_cpu_ids == 1)
+		return;
+
+	lb_monitor_task = kthread_create(load_balance_monitor, NULL,
+					 "group_balance");
+	if (!IS_ERR(lb_monitor_task)) {
+		lb_monitor_task->flags |= PF_NOFREEZE;
+		wake_up_process(lb_monitor_task);
+	} else {
+		printk(KERN_ERR "Could not create load balance monitor thread"
+			"(error = %ld) \n", PTR_ERR(lb_monitor_task));
+	}
+#endif
 }
 #else
 void __init sched_init_smp(void)
@@ -6988,6 +7049,157 @@ void set_curr_task(int cpu, struct task_struct *p)
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 
+#ifdef CONFIG_SMP
+/*
+ * distribute shares of all task groups among their schedulable entities,
+ * to reflect load distrbution across cpus.
+ */
+static int rebalance_shares(struct sched_domain *sd, int this_cpu)
+{
+	struct cfs_rq *cfs_rq;
+	struct rq *rq = cpu_rq(this_cpu);
+	cpumask_t sdspan = sd->span;
+	int balanced = 1;
+
+	/* Walk thr' all the task groups that we have */
+	for_each_leaf_cfs_rq(rq, cfs_rq) {
+		int i;
+		unsigned long total_load = 0, total_shares;
+		struct task_group *tg = cfs_rq->tg;
+
+		/* Gather total task load of this group across cpus */
+		for_each_cpu_mask(i, sdspan)
+			total_load += tg->cfs_rq[i]->load.weight;
+
+		/* Nothing to do if this group has no load  */
+		if (!total_load)
+			continue;
+
+		/*
+		 * tg->shares represents the number of cpu shares the task group
+		 * is eligible to hold on a single cpu. On N cpus, it is
+		 * eligible to hold (N * tg->shares) number of cpu shares.
+		 */
+		total_shares = tg->shares * cpus_weight(sdspan);
+
+		/*
+		 * redistribute total_shares across cpus as per the task load
+		 * distribution.
+		 */
+		for_each_cpu_mask(i, sdspan) {
+			unsigned long local_load, local_shares;
+
+			local_load = tg->cfs_rq[i]->load.weight;
+			local_shares = (local_load * total_shares) / total_load;
+			if (!local_shares)
+				local_shares = MIN_GROUP_SHARES;
+			if (local_shares == tg->se[i]->load.weight)
+				continue;
+
+			spin_lock_irq(&cpu_rq(i)->lock);
+			set_se_shares(tg->se[i], local_shares);
+			spin_unlock_irq(&cpu_rq(i)->lock);
+			balanced = 0;
+		}
+	}
+
+	return balanced;
+}
+
+/*
+ * How frequently should we rebalance_shares() across cpus?
+ *
+ * The more frequently we rebalance shares, the more accurate is the fairness
+ * of cpu bandwidth distribution between task groups. However higher frequency
+ * also implies increased scheduling overhead.
+ *
+ * sysctl_sched_min_bal_int_shares represents the minimum interval between
+ * consecutive calls to rebalance_shares() in the same sched domain.
+ *
+ * sysctl_sched_max_bal_int_shares represents the maximum interval between
+ * consecutive calls to rebalance_shares() in the same sched domain.
+ *
+ * These settings allows for the appropriate tradeoff between accuracy of
+ * fairness and the associated overhead.
+ *
+ */
+
+/* default: 8ms, units: milliseconds */
+const_debug unsigned int sysctl_sched_min_bal_int_shares = 8;
+
+/* default: 128ms, units: milliseconds */
+const_debug unsigned int sysctl_sched_max_bal_int_shares = 128;
+
+/* kernel thread that runs rebalance_shares() periodically */
+static int load_balance_monitor(void *unused)
+{
+	unsigned int timeout = sysctl_sched_min_bal_int_shares;
+	struct sched_param schedparm;
+	int ret;
+
+	/*
+	 * We don't want this thread's execution to be limited by the shares
+	 * assigned to default group (init_task_group). Hence make it run
+	 * as a SCHED_RR RT task at the lowest priority.
+	 */
+	schedparm.sched_priority = 1;
+	ret = sched_setscheduler(current, SCHED_RR, &schedparm);
+	if (ret)
+		printk(KERN_ERR "Couldn't set SCHED_RR policy for load balance"
+				" monitor thread (error = %d) \n", ret);
+
+	while (!kthread_should_stop()) {
+		int i, cpu, balanced = 1;
+
+		/* Prevent cpus going down or coming up */
+		lock_cpu_hotplug();
+		/* lockout changes to doms_cur[] array */
+		lock_doms_cur();
+		/*
+		 * Enter a rcu read-side critical section to safely walk rq->sd
+		 * chain on various cpus and to walk task group list
+		 * (rq->leaf_cfs_rq_list) in rebalance_shares().
+		 */
+		rcu_read_lock();
+
+		for (i = 0; i < ndoms_cur; i++) {
+			cpumask_t cpumap = doms_cur[i];
+			struct sched_domain *sd = NULL, *sd_prev = NULL;
+
+			cpu = first_cpu(cpumap);
+
+			/* Find the highest domain at which to balance shares */
+			for_each_domain(cpu, sd) {
+				if (!(sd->flags & SD_LOAD_BALANCE))
+					continue;
+				sd_prev = sd;
+			}
+
+			sd = sd_prev;
+			/* sd == NULL? No load balance reqd in this domain */
+			if (!sd)
+				continue;
+
+			balanced &= rebalance_shares(sd, cpu);
+		}
+
+		rcu_read_unlock();
+
+		unlock_doms_cur();
+		unlock_cpu_hotplug();
+
+		if (!balanced)
+			timeout = sysctl_sched_min_bal_int_shares;
+		else if (timeout < sysctl_sched_max_bal_int_shares)
+			timeout *= 2;
+
+		msleep_interruptible(timeout);
+	}
+
+	return 0;
+}
+#endif	/* CONFIG_SMP */
+
 /* allocate runqueue etc for a new task group */
 struct task_group *sched_create_group(void)
 {
@@ -7144,47 +7356,77 @@ done:
 	task_rq_unlock(rq, &flags);
 }
 
+/* rq->lock to be locked by caller */
 static void set_se_shares(struct sched_entity *se, unsigned long shares)
 {
 	struct cfs_rq *cfs_rq = se->cfs_rq;
 	struct rq *rq = cfs_rq->rq;
 	int on_rq;
 
-	spin_lock_irq(&rq->lock);
+	if (!shares)
+		shares = MIN_GROUP_SHARES;
 
 	on_rq = se->on_rq;
-	if (on_rq)
+	if (on_rq) {
 		dequeue_entity(cfs_rq, se, 0);
+		dec_cpu_load(rq, se->load.weight);
+	}
 
 	se->load.weight = shares;
 	se->load.inv_weight = div64_64((1ULL<<32), shares);
 
-	if (on_rq)
+	if (on_rq) {
 		enqueue_entity(cfs_rq, se, 0);
-
-	spin_unlock_irq(&rq->lock);
+		inc_cpu_load(rq, se->load.weight);
+	}
 }
 
 int sched_group_set_shares(struct task_group *tg, unsigned long shares)
 {
 	int i;
-
-	/*
-	 * A weight of 0 or 1 can cause arithmetics problems.
-	 * (The default weight is 1024 - so there's no practical
-	 *  limitation from this.)
-	 */
-	if (shares < 2)
-		shares = 2;
+	struct cfs_rq *cfs_rq;
+	struct rq *rq;
 
 	lock_task_group_list();
 	if (tg->shares == shares)
 		goto done;
 
+	if (shares < MIN_GROUP_SHARES)
+		shares = MIN_GROUP_SHARES;
+
+	/*
+	 * Prevent any load balance activity (rebalance_shares,
+	 * load_balance_fair) from referring to this group first,
+	 * by taking it off the rq->leaf_cfs_rq_list on each cpu.
+	 */
+	for_each_possible_cpu(i) {
+		cfs_rq = tg->cfs_rq[i];
+		list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+	}
+
+	/* wait for any ongoing reference to this group to finish */
+	synchronize_sched();
+
+	/*
+	 * Now we are free to modify the group's share on each cpu
+	 * w/o tripping rebalance_share or load_balance_fair.
+	 */
 	tg->shares = shares;
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		spin_lock_irq(&cpu_rq(i)->lock);
 		set_se_shares(tg->se[i], shares);
+		spin_unlock_irq(&cpu_rq(i)->lock);
+	}
 
+	/*
+	 * Enable load balance activity on this group, by inserting it back on
+	 * each cpu's rq->leaf_cfs_rq_list.
+	 */
+	for_each_possible_cpu(i) {
+		rq = cpu_rq(i);
+		cfs_rq = tg->cfs_rq[i];
+		list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+	}
 done:
 	unlock_task_group_list();
 	return 0;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 30ae9c2a2861..5c208e090ae4 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -707,6 +707,8 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 	return se->parent;
 }
 
+#define GROUP_IMBALANCE_PCT	20
+
 #else	/* CONFIG_FAIR_GROUP_SCHED */
 
 #define for_each_sched_entity(se) \
@@ -967,25 +969,6 @@ static struct task_struct *load_balance_next_fair(void *arg)
 	return __load_balance_iterator(cfs_rq, cfs_rq->rb_load_balance_curr);
 }
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
-static int cfs_rq_best_prio(struct cfs_rq *cfs_rq)
-{
-	struct sched_entity *curr;
-	struct task_struct *p;
-
-	if (!cfs_rq->nr_running)
-		return MAX_PRIO;
-
-	curr = cfs_rq->curr;
-	if (!curr)
-		curr = __pick_next_entity(cfs_rq);
-
-	p = task_of(curr);
-
-	return p->prio;
-}
-#endif
-
 static unsigned long
 load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		  unsigned long max_load_move,
@@ -995,28 +978,45 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 	struct cfs_rq *busy_cfs_rq;
 	long rem_load_move = max_load_move;
 	struct rq_iterator cfs_rq_iterator;
+	unsigned long load_moved;
 
 	cfs_rq_iterator.start = load_balance_start_fair;
 	cfs_rq_iterator.next = load_balance_next_fair;
 
 	for_each_leaf_cfs_rq(busiest, busy_cfs_rq) {
 #ifdef CONFIG_FAIR_GROUP_SCHED
-		struct cfs_rq *this_cfs_rq;
-		long imbalance;
-		unsigned long maxload;
+		struct cfs_rq *this_cfs_rq = busy_cfs_rq->tg->cfs_rq[this_cpu];
+		unsigned long maxload, task_load, group_weight;
+		unsigned long thisload, per_task_load;
+		struct sched_entity *se = busy_cfs_rq->tg->se[busiest->cpu];
 
-		this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu);
+		task_load = busy_cfs_rq->load.weight;
+		group_weight = se->load.weight;
 
-		imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight;
-		/* Don't pull if this_cfs_rq has more load than busy_cfs_rq */
-		if (imbalance <= 0)
+		/*
+		 * 'group_weight' is contributed by tasks of total weight
+		 * 'task_load'. To move 'rem_load_move' worth of weight only,
+		 * we need to move a maximum task load of:
+		 *
+		 * 	maxload = (remload / group_weight) * task_load;
+		 */
+		maxload = (rem_load_move * task_load) / group_weight;
+
+		if (!maxload || !task_load)
 			continue;
 
-		/* Don't pull more than imbalance/2 */
-		imbalance /= 2;
-		maxload = min(rem_load_move, imbalance);
+		per_task_load = task_load / busy_cfs_rq->nr_running;
+		/*
+		 * balance_tasks will try to forcibly move atleast one task if
+		 * possible (because of SCHED_LOAD_SCALE_FUZZ). Avoid that if
+		 * maxload is less than GROUP_IMBALANCE_FUZZ% the per_task_load.
+		 */
+		 if (100 * maxload < GROUP_IMBALANCE_PCT * per_task_load)
+			continue;
 
-		*this_best_prio = cfs_rq_best_prio(this_cfs_rq);
+		/* Disable priority-based load balance */
+		*this_best_prio = 0;
+		thisload = this_cfs_rq->load.weight;
 #else
 # define maxload rem_load_move
 #endif
@@ -1025,11 +1025,33 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		 * load_balance_[start|next]_fair iterators
 		 */
 		cfs_rq_iterator.arg = busy_cfs_rq;
-		rem_load_move -= balance_tasks(this_rq, this_cpu, busiest,
+		load_moved = balance_tasks(this_rq, this_cpu, busiest,
 					       maxload, sd, idle, all_pinned,
 					       this_best_prio,
 					       &cfs_rq_iterator);
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+		/*
+		 * load_moved holds the task load that was moved. The
+		 * effective (group) weight moved would be:
+		 * 	load_moved_eff = load_moved/task_load * group_weight;
+		 */
+		load_moved = (group_weight * load_moved) / task_load;
+
+		/* Adjust shares on both cpus to reflect load_moved */
+		group_weight -= load_moved;
+		set_se_shares(se, group_weight);
+
+		se = busy_cfs_rq->tg->se[this_cpu];
+		if (!thisload)
+			group_weight = load_moved;
+		else
+			group_weight = se->load.weight + load_moved;
+		set_se_shares(se, group_weight);
+#endif
+
+		rem_load_move -= load_moved;
+
 		if (rem_load_move <= 0)
 			break;
 	}
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index c68f68dcc605..c95f3ed34474 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -309,6 +309,24 @@ static struct ctl_table kern_table[] = {
 		.mode		= 644,
 		.proc_handler	= &proc_dointvec,
 	},
+#if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
+	{
+		.ctl_name       = CTL_UNNUMBERED,
+		.procname       = "sched_min_bal_int_shares",
+		.data           = &sysctl_sched_min_bal_int_shares,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = &proc_dointvec,
+	},
+	{
+		.ctl_name       = CTL_UNNUMBERED,
+		.procname       = "sched_max_bal_int_shares",
+		.data           = &sysctl_sched_max_bal_int_shares,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = &proc_dointvec,
+	},
+#endif
 #endif
 	{
 		.ctl_name	= CTL_UNNUMBERED,
-- 
cgit v1.2.3


From d221938c049f4845da13c8593132595a6b9222a8 Mon Sep 17 00:00:00 2001
From: Gautham R Shenoy <ego@in.ibm.com>
Date: Fri, 25 Jan 2008 21:08:01 +0100
Subject: cpu-hotplug: refcount based cpu hotplug

This patch implements a Refcount + Waitqueue based model for
cpu-hotplug.

Now, a thread which wants to prevent cpu-hotplug, will bump up a global
refcount and the thread which wants to perform a cpu-hotplug operation
will block till the global refcount goes to zero.

The readers, if any, during an ongoing cpu-hotplug operation are blocked
until the cpu-hotplug operation is over.

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Paul Jackson <pj@sgi.com> [For !CONFIG_HOTPLUG_CPU ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpu.h |   3 ++
 init/main.c         |   1 +
 kernel/cpu.c        | 152 ++++++++++++++++++++++++++++++++++++++--------------
 3 files changed, 115 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 92f2029a34f3..a40247e4d462 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -83,6 +83,9 @@ static inline void unregister_cpu_notifier(struct notifier_block *nb)
 
 #endif /* CONFIG_SMP */
 extern struct sysdev_class cpu_sysdev_class;
+extern void cpu_hotplug_init(void);
+extern void cpu_maps_update_begin(void);
+extern void cpu_maps_update_done(void);
 
 #ifdef CONFIG_HOTPLUG_CPU
 /* Stop CPUs going up and down. */
diff --git a/init/main.c b/init/main.c
index 80b04b6c5157..f287ca5862b9 100644
--- a/init/main.c
+++ b/init/main.c
@@ -607,6 +607,7 @@ asmlinkage void __init start_kernel(void)
 	vfs_caches_init_early();
 	cpuset_init_early();
 	mem_init();
+	cpu_hotplug_init();
 	kmem_cache_init();
 	setup_per_cpu_pageset();
 	numa_policy_init();
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 6b3a0c15144f..656dc3fcbbae 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -15,9 +15,8 @@
 #include <linux/stop_machine.h>
 #include <linux/mutex.h>
 
-/* This protects CPUs going up and down... */
+/* Serializes the updates to cpu_online_map, cpu_present_map */
 static DEFINE_MUTEX(cpu_add_remove_lock);
-static DEFINE_MUTEX(cpu_bitmask_lock);
 
 static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
 
@@ -26,52 +25,123 @@ static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
  */
 static int cpu_hotplug_disabled;
 
-#ifdef CONFIG_HOTPLUG_CPU
+static struct {
+	struct task_struct *active_writer;
+	struct mutex lock; /* Synchronizes accesses to refcount, */
+	/*
+	 * Also blocks the new readers during
+	 * an ongoing cpu hotplug operation.
+	 */
+	int refcount;
+	wait_queue_head_t writer_queue;
+} cpu_hotplug;
 
-/* Crappy recursive lock-takers in cpufreq! Complain loudly about idiots */
-static struct task_struct *recursive;
-static int recursive_depth;
+#define writer_exists() (cpu_hotplug.active_writer != NULL)
+
+void __init cpu_hotplug_init(void)
+{
+	cpu_hotplug.active_writer = NULL;
+	mutex_init(&cpu_hotplug.lock);
+	cpu_hotplug.refcount = 0;
+	init_waitqueue_head(&cpu_hotplug.writer_queue);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
 
 void lock_cpu_hotplug(void)
 {
-	struct task_struct *tsk = current;
-
-	if (tsk == recursive) {
-		static int warnings = 10;
-		if (warnings) {
-			printk(KERN_ERR "Lukewarm IQ detected in hotplug locking\n");
-			WARN_ON(1);
-			warnings--;
-		}
-		recursive_depth++;
+	might_sleep();
+	if (cpu_hotplug.active_writer == current)
 		return;
-	}
-	mutex_lock(&cpu_bitmask_lock);
-	recursive = tsk;
+	mutex_lock(&cpu_hotplug.lock);
+	cpu_hotplug.refcount++;
+	mutex_unlock(&cpu_hotplug.lock);
+
 }
 EXPORT_SYMBOL_GPL(lock_cpu_hotplug);
 
 void unlock_cpu_hotplug(void)
 {
-	WARN_ON(recursive != current);
-	if (recursive_depth) {
-		recursive_depth--;
+	if (cpu_hotplug.active_writer == current)
 		return;
-	}
-	recursive = NULL;
-	mutex_unlock(&cpu_bitmask_lock);
+	mutex_lock(&cpu_hotplug.lock);
+	cpu_hotplug.refcount--;
+
+	if (unlikely(writer_exists()) && !cpu_hotplug.refcount)
+		wake_up(&cpu_hotplug.writer_queue);
+
+	mutex_unlock(&cpu_hotplug.lock);
+
 }
 EXPORT_SYMBOL_GPL(unlock_cpu_hotplug);
 
 #endif	/* CONFIG_HOTPLUG_CPU */
 
+/*
+ * The following two API's must be used when attempting
+ * to serialize the updates to cpu_online_map, cpu_present_map.
+ */
+void cpu_maps_update_begin(void)
+{
+	mutex_lock(&cpu_add_remove_lock);
+}
+
+void cpu_maps_update_done(void)
+{
+	mutex_unlock(&cpu_add_remove_lock);
+}
+
+/*
+ * This ensures that the hotplug operation can begin only when the
+ * refcount goes to zero.
+ *
+ * Note that during a cpu-hotplug operation, the new readers, if any,
+ * will be blocked by the cpu_hotplug.lock
+ *
+ * Since cpu_maps_update_begin is always called after invoking
+ * cpu_maps_update_begin, we can be sure that only one writer is active.
+ *
+ * Note that theoretically, there is a possibility of a livelock:
+ * - Refcount goes to zero, last reader wakes up the sleeping
+ *   writer.
+ * - Last reader unlocks the cpu_hotplug.lock.
+ * - A new reader arrives at this moment, bumps up the refcount.
+ * - The writer acquires the cpu_hotplug.lock finds the refcount
+ *   non zero and goes to sleep again.
+ *
+ * However, this is very difficult to achieve in practice since
+ * lock_cpu_hotplug() not an api which is called all that often.
+ *
+ */
+static void cpu_hotplug_begin(void)
+{
+	DECLARE_WAITQUEUE(wait, current);
+
+	mutex_lock(&cpu_hotplug.lock);
+
+	cpu_hotplug.active_writer = current;
+	add_wait_queue_exclusive(&cpu_hotplug.writer_queue, &wait);
+	while (cpu_hotplug.refcount) {
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&cpu_hotplug.lock);
+		schedule();
+		mutex_lock(&cpu_hotplug.lock);
+	}
+	remove_wait_queue_locked(&cpu_hotplug.writer_queue, &wait);
+}
+
+static void cpu_hotplug_done(void)
+{
+	cpu_hotplug.active_writer = NULL;
+	mutex_unlock(&cpu_hotplug.lock);
+}
 /* Need to know about CPUs going up/down? */
 int __cpuinit register_cpu_notifier(struct notifier_block *nb)
 {
 	int ret;
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	ret = raw_notifier_chain_register(&cpu_chain, nb);
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return ret;
 }
 
@@ -81,9 +151,9 @@ EXPORT_SYMBOL(register_cpu_notifier);
 
 void unregister_cpu_notifier(struct notifier_block *nb)
 {
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	raw_notifier_chain_unregister(&cpu_chain, nb);
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 }
 EXPORT_SYMBOL(unregister_cpu_notifier);
 
@@ -147,6 +217,7 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 	if (!cpu_online(cpu))
 		return -EINVAL;
 
+	cpu_hotplug_begin();
 	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
 	err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod,
 					hcpu, -1, &nr_calls);
@@ -166,9 +237,7 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 	cpu_clear(cpu, tmp);
 	set_cpus_allowed(current, tmp);
 
-	mutex_lock(&cpu_bitmask_lock);
 	p = __stop_machine_run(take_cpu_down, &tcd_param, cpu);
-	mutex_unlock(&cpu_bitmask_lock);
 
 	if (IS_ERR(p) || cpu_online(cpu)) {
 		/* CPU didn't die: tell everyone.  Can't complain. */
@@ -203,6 +272,7 @@ out_allowed:
 	set_cpus_allowed(current, old_allowed);
 out_release:
 	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
+	cpu_hotplug_done();
 	return err;
 }
 
@@ -210,13 +280,13 @@ int cpu_down(unsigned int cpu)
 {
 	int err = 0;
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	if (cpu_hotplug_disabled)
 		err = -EBUSY;
 	else
 		err = _cpu_down(cpu, 0);
 
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return err;
 }
 #endif /*CONFIG_HOTPLUG_CPU*/
@@ -231,6 +301,7 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen)
 	if (cpu_online(cpu) || !cpu_present(cpu))
 		return -EINVAL;
 
+	cpu_hotplug_begin();
 	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
 	ret = __raw_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE | mod, hcpu,
 							-1, &nr_calls);
@@ -243,9 +314,7 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen)
 	}
 
 	/* Arch-specific enabling code. */
-	mutex_lock(&cpu_bitmask_lock);
 	ret = __cpu_up(cpu);
-	mutex_unlock(&cpu_bitmask_lock);
 	if (ret != 0)
 		goto out_notify;
 	BUG_ON(!cpu_online(cpu));
@@ -258,6 +327,7 @@ out_notify:
 		__raw_notifier_call_chain(&cpu_chain,
 				CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL);
 	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
+	cpu_hotplug_done();
 
 	return ret;
 }
@@ -275,13 +345,13 @@ int __cpuinit cpu_up(unsigned int cpu)
 		return -EINVAL;
 	}
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	if (cpu_hotplug_disabled)
 		err = -EBUSY;
 	else
 		err = _cpu_up(cpu, 0);
 
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return err;
 }
 
@@ -292,7 +362,7 @@ int disable_nonboot_cpus(void)
 {
 	int cpu, first_cpu, error = 0;
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	first_cpu = first_cpu(cpu_online_map);
 	/* We take down all of the non-boot CPUs in one shot to avoid races
 	 * with the userspace trying to use the CPU hotplug at the same time
@@ -319,7 +389,7 @@ int disable_nonboot_cpus(void)
 	} else {
 		printk(KERN_ERR "Non-boot CPUs are not disabled\n");
 	}
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return error;
 }
 
@@ -328,7 +398,7 @@ void enable_nonboot_cpus(void)
 	int cpu, error;
 
 	/* Allow everyone to use the CPU hotplug again */
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	cpu_hotplug_disabled = 0;
 	if (cpus_empty(frozen_cpus))
 		goto out;
@@ -344,6 +414,6 @@ void enable_nonboot_cpus(void)
 	}
 	cpus_clear(frozen_cpus);
 out:
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 }
 #endif /* CONFIG_PM_SLEEP_SMP */
-- 
cgit v1.2.3


From 86ef5c9a8edd78e6bf92879f32329d89b2d55b5a Mon Sep 17 00:00:00 2001
From: Gautham R Shenoy <ego@in.ibm.com>
Date: Fri, 25 Jan 2008 21:08:02 +0100
Subject: cpu-hotplug: replace lock_cpu_hotplug() with get_online_cpus()

Replace all lock_cpu_hotplug/unlock_cpu_hotplug from the kernel and use
get_online_cpus and put_online_cpus instead as it highlights the
refcount semantics in these operations.

The new API guarantees protection against the cpu-hotplug operation, but
it doesn't guarantee serialized access to any of the local data
structures. Hence the changes needs to be reviewed.

In case of pseries_add_processor/pseries_remove_processor, use
cpu_maps_update_begin()/cpu_maps_update_done() as we're modifying the
cpu_present_map there.

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 Documentation/cpu-hotplug.txt                | 11 ++++++-----
 arch/mips/kernel/mips-mt-fpaff.c             | 10 +++++-----
 arch/powerpc/platforms/pseries/hotplug-cpu.c |  8 ++++----
 arch/powerpc/platforms/pseries/rtasd.c       |  8 ++++----
 arch/x86/kernel/cpu/mtrr/main.c              |  8 ++++----
 arch/x86/kernel/microcode.c                  | 16 ++++++++--------
 drivers/lguest/x86/core.c                    |  8 ++++----
 drivers/s390/char/sclp_config.c              |  4 ++--
 include/linux/cpu.h                          |  8 ++++----
 kernel/cpu.c                                 | 10 +++++-----
 kernel/cpuset.c                              | 14 +++++++-------
 kernel/rcutorture.c                          |  6 +++---
 kernel/sched.c                               |  4 ++--
 kernel/stop_machine.c                        |  4 ++--
 net/core/flow.c                              |  4 ++--
 15 files changed, 62 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index a741f658a3c9..fb94f5a71b68 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -109,12 +109,13 @@ Never use anything other than cpumask_t to represent bitmap of CPUs.
 	for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask.
 
 	#include <linux/cpu.h>
-	lock_cpu_hotplug() and unlock_cpu_hotplug():
+	get_online_cpus() and put_online_cpus():
 
-The above calls are used to inhibit cpu hotplug operations. While holding the
-cpucontrol mutex, cpu_online_map will not change. If you merely need to avoid
-cpus going away, you could also use preempt_disable() and preempt_enable()
-for those sections. Just remember the critical section cannot call any
+The above calls are used to inhibit cpu hotplug operations. While the
+cpu_hotplug.refcount is non zero, the cpu_online_map will not change.
+If you merely need to avoid cpus going away, you could also use
+preempt_disable() and preempt_enable() for those sections.
+Just remember the critical section cannot call any
 function that can sleep or schedule this process away. The preempt_disable()
 will work as long as stop_machine_run() is used to take a cpu down.
 
diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c
index 892665bb12b1..bb4f00c0cbe9 100644
--- a/arch/mips/kernel/mips-mt-fpaff.c
+++ b/arch/mips/kernel/mips-mt-fpaff.c
@@ -58,13 +58,13 @@ asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len,
 	if (copy_from_user(&new_mask, user_mask_ptr, sizeof(new_mask)))
 		return -EFAULT;
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	p = find_process_by_pid(pid);
 	if (!p) {
 		read_unlock(&tasklist_lock);
-		unlock_cpu_hotplug();
+		put_online_cpus();
 		return -ESRCH;
 	}
 
@@ -106,7 +106,7 @@ asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len,
 
 out_unlock:
 	put_task_struct(p);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 	return retval;
 }
 
@@ -125,7 +125,7 @@ asmlinkage long mipsmt_sys_sched_getaffinity(pid_t pid, unsigned int len,
 	if (len < real_len)
 		return -EINVAL;
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	retval = -ESRCH;
@@ -140,7 +140,7 @@ asmlinkage long mipsmt_sys_sched_getaffinity(pid_t pid, unsigned int len,
 
 out_unlock:
 	read_unlock(&tasklist_lock);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 	if (retval)
 		return retval;
 	if (copy_to_user(user_mask_ptr, &mask, real_len))
diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
index 412e6b42986f..c4ad54e0f288 100644
--- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
+++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
@@ -153,7 +153,7 @@ static int pseries_add_processor(struct device_node *np)
 	for (i = 0; i < nthreads; i++)
 		cpu_set(i, tmp);
 
-	lock_cpu_hotplug();
+	cpu_maps_update_begin();
 
 	BUG_ON(!cpus_subset(cpu_present_map, cpu_possible_map));
 
@@ -190,7 +190,7 @@ static int pseries_add_processor(struct device_node *np)
 	}
 	err = 0;
 out_unlock:
-	unlock_cpu_hotplug();
+	cpu_maps_update_done();
 	return err;
 }
 
@@ -211,7 +211,7 @@ static void pseries_remove_processor(struct device_node *np)
 
 	nthreads = len / sizeof(u32);
 
-	lock_cpu_hotplug();
+	cpu_maps_update_begin();
 	for (i = 0; i < nthreads; i++) {
 		for_each_present_cpu(cpu) {
 			if (get_hard_smp_processor_id(cpu) != intserv[i])
@@ -225,7 +225,7 @@ static void pseries_remove_processor(struct device_node *np)
 			printk(KERN_WARNING "Could not find cpu to remove "
 			       "with physical id 0x%x\n", intserv[i]);
 	}
-	unlock_cpu_hotplug();
+	cpu_maps_update_done();
 }
 
 static int pseries_smp_notifier(struct notifier_block *nb,
diff --git a/arch/powerpc/platforms/pseries/rtasd.c b/arch/powerpc/platforms/pseries/rtasd.c
index 73401c820110..e3078ce41518 100644
--- a/arch/powerpc/platforms/pseries/rtasd.c
+++ b/arch/powerpc/platforms/pseries/rtasd.c
@@ -382,7 +382,7 @@ static void do_event_scan_all_cpus(long delay)
 {
 	int cpu;
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	cpu = first_cpu(cpu_online_map);
 	for (;;) {
 		set_cpus_allowed(current, cpumask_of_cpu(cpu));
@@ -390,15 +390,15 @@ static void do_event_scan_all_cpus(long delay)
 		set_cpus_allowed(current, CPU_MASK_ALL);
 
 		/* Drop hotplug lock, and sleep for the specified delay */
-		unlock_cpu_hotplug();
+		put_online_cpus();
 		msleep_interruptible(delay);
-		lock_cpu_hotplug();
+		get_online_cpus();
 
 		cpu = next_cpu(cpu, cpu_online_map);
 		if (cpu == NR_CPUS)
 			break;
 	}
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 static int rtasd(void *unused)
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 3b20613325dc..beb45c9c0835 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -349,7 +349,7 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 	replace = -1;
 
 	/* No CPU hotplug when we change MTRR entries */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	/*  Search for existing MTRR  */
 	mutex_lock(&mtrr_mutex);
 	for (i = 0; i < num_var_ranges; ++i) {
@@ -405,7 +405,7 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 	error = i;
  out:
 	mutex_unlock(&mtrr_mutex);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 	return error;
 }
 
@@ -495,7 +495,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 
 	max = num_var_ranges;
 	/* No CPU hotplug when we change MTRR entries */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	mutex_lock(&mtrr_mutex);
 	if (reg < 0) {
 		/*  Search for existing MTRR  */
@@ -536,7 +536,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 	error = reg;
  out:
 	mutex_unlock(&mtrr_mutex);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 	return error;
 }
 /**
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
index 09c315214a5e..40cfd5488719 100644
--- a/arch/x86/kernel/microcode.c
+++ b/arch/x86/kernel/microcode.c
@@ -436,7 +436,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_
 		return -EINVAL;
 	}
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	mutex_lock(&microcode_mutex);
 
 	user_buffer = (void __user *) buf;
@@ -447,7 +447,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_
 		ret = (ssize_t)len;
 
 	mutex_unlock(&microcode_mutex);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 	return ret;
 }
@@ -658,14 +658,14 @@ static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz)
 
 		old = current->cpus_allowed;
 
-		lock_cpu_hotplug();
+		get_online_cpus();
 		set_cpus_allowed(current, cpumask_of_cpu(cpu));
 
 		mutex_lock(&microcode_mutex);
 		if (uci->valid)
 			err = cpu_request_microcode(cpu);
 		mutex_unlock(&microcode_mutex);
-		unlock_cpu_hotplug();
+		put_online_cpus();
 		set_cpus_allowed(current, old);
 	}
 	if (err)
@@ -817,9 +817,9 @@ static int __init microcode_init (void)
 		return PTR_ERR(microcode_pdev);
 	}
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 	if (error) {
 		microcode_dev_exit();
 		platform_device_unregister(microcode_pdev);
@@ -839,9 +839,9 @@ static void __exit microcode_exit (void)
 
 	unregister_hotcpu_notifier(&mc_cpu_notifier);
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 	sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 	platform_device_unregister(microcode_pdev);
 }
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 482aec2a9631..96d0fd07c57d 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -459,7 +459,7 @@ void __init lguest_arch_host_init(void)
 
 	/* We don't need the complexity of CPUs coming and going while we're
 	 * doing this. */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	if (cpu_has_pge) { /* We have a broader idea of "global". */
 		/* Remember that this was originally set (for cleanup). */
 		cpu_had_pge = 1;
@@ -469,20 +469,20 @@ void __init lguest_arch_host_init(void)
 		/* Turn off the feature in the global feature set. */
 		clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
 	}
-	unlock_cpu_hotplug();
+	put_online_cpus();
 };
 /*:*/
 
 void __exit lguest_arch_host_fini(void)
 {
 	/* If we had PGE before we started, turn it back on now. */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	if (cpu_had_pge) {
 		set_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
 		/* adjust_pge's argument "1" means set PGE. */
 		on_each_cpu(adjust_pge, (void *)1, 0, 1);
 	}
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 
diff --git a/drivers/s390/char/sclp_config.c b/drivers/s390/char/sclp_config.c
index 5322e5e54a98..9dc77f14fa52 100644
--- a/drivers/s390/char/sclp_config.c
+++ b/drivers/s390/char/sclp_config.c
@@ -29,12 +29,12 @@ static void sclp_cpu_capability_notify(struct work_struct *work)
 	struct sys_device *sysdev;
 
 	printk(KERN_WARNING TAG "cpu capability changed.\n");
-	lock_cpu_hotplug();
+	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		sysdev = get_cpu_sysdev(cpu);
 		kobject_uevent(&sysdev->kobj, KOBJ_CHANGE);
 	}
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 static void sclp_conf_receiver_fn(struct evbuf_header *evbuf)
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index a40247e4d462..3a3ff1c5cbef 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -100,8 +100,8 @@ static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex)
 	mutex_unlock(cpu_hp_mutex);
 }
 
-extern void lock_cpu_hotplug(void);
-extern void unlock_cpu_hotplug(void);
+extern void get_online_cpus(void);
+extern void put_online_cpus(void);
 #define hotcpu_notifier(fn, pri) {				\
 	static struct notifier_block fn##_nb =			\
 		{ .notifier_call = fn, .priority = pri };	\
@@ -118,8 +118,8 @@ static inline void cpuhotplug_mutex_lock(struct mutex *cpu_hp_mutex)
 static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex)
 { }
 
-#define lock_cpu_hotplug()	do { } while (0)
-#define unlock_cpu_hotplug()	do { } while (0)
+#define get_online_cpus()	do { } while (0)
+#define put_online_cpus()	do { } while (0)
 #define hotcpu_notifier(fn, pri)	do { (void)(fn); } while (0)
 /* These aren't inline functions due to a GCC bug. */
 #define register_hotcpu_notifier(nb)	({ (void)(nb); 0; })
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 656dc3fcbbae..b0c4152995f8 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -48,7 +48,7 @@ void __init cpu_hotplug_init(void)
 
 #ifdef CONFIG_HOTPLUG_CPU
 
-void lock_cpu_hotplug(void)
+void get_online_cpus(void)
 {
 	might_sleep();
 	if (cpu_hotplug.active_writer == current)
@@ -58,9 +58,9 @@ void lock_cpu_hotplug(void)
 	mutex_unlock(&cpu_hotplug.lock);
 
 }
-EXPORT_SYMBOL_GPL(lock_cpu_hotplug);
+EXPORT_SYMBOL_GPL(get_online_cpus);
 
-void unlock_cpu_hotplug(void)
+void put_online_cpus(void)
 {
 	if (cpu_hotplug.active_writer == current)
 		return;
@@ -73,7 +73,7 @@ void unlock_cpu_hotplug(void)
 	mutex_unlock(&cpu_hotplug.lock);
 
 }
-EXPORT_SYMBOL_GPL(unlock_cpu_hotplug);
+EXPORT_SYMBOL_GPL(put_online_cpus);
 
 #endif	/* CONFIG_HOTPLUG_CPU */
 
@@ -110,7 +110,7 @@ void cpu_maps_update_done(void)
  *   non zero and goes to sleep again.
  *
  * However, this is very difficult to achieve in practice since
- * lock_cpu_hotplug() not an api which is called all that often.
+ * get_online_cpus() not an api which is called all that often.
  *
  */
 static void cpu_hotplug_begin(void)
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 50f5dc463688..cfaf6419d817 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -537,10 +537,10 @@ static int cpusets_overlap(struct cpuset *a, struct cpuset *b)
  *
  * Call with cgroup_mutex held.  May take callback_mutex during
  * call due to the kfifo_alloc() and kmalloc() calls.  May nest
- * a call to the lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
+ * a call to the get_online_cpus()/put_online_cpus() pair.
  * Must not be called holding callback_mutex, because we must not
- * call lock_cpu_hotplug() while holding callback_mutex.  Elsewhere
- * the kernel nests callback_mutex inside lock_cpu_hotplug() calls.
+ * call get_online_cpus() while holding callback_mutex.  Elsewhere
+ * the kernel nests callback_mutex inside get_online_cpus() calls.
  * So the reverse nesting would risk an ABBA deadlock.
  *
  * The three key local variables below are:
@@ -691,9 +691,9 @@ restart:
 
 rebuild:
 	/* Have scheduler rebuild sched domains */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	partition_sched_domains(ndoms, doms);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 done:
 	if (q && !IS_ERR(q))
@@ -1617,10 +1617,10 @@ static struct cgroup_subsys_state *cpuset_create(
  *
  * If the cpuset being removed has its flag 'sched_load_balance'
  * enabled, then simulate turning sched_load_balance off, which
- * will call rebuild_sched_domains().  The lock_cpu_hotplug()
+ * will call rebuild_sched_domains().  The get_online_cpus()
  * call in rebuild_sched_domains() must not be made while holding
  * callback_mutex.  Elsewhere the kernel nests callback_mutex inside
- * lock_cpu_hotplug() calls.  So the reverse nesting would risk an
+ * get_online_cpus() calls.  So the reverse nesting would risk an
  * ABBA deadlock.
  */
 
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index c3e165c2318f..fd599829e72a 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -726,11 +726,11 @@ static void rcu_torture_shuffle_tasks(void)
 	cpumask_t tmp_mask = CPU_MASK_ALL;
 	int i;
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 
 	/* No point in shuffling if there is only one online CPU (ex: UP) */
 	if (num_online_cpus() == 1) {
-		unlock_cpu_hotplug();
+		put_online_cpus();
 		return;
 	}
 
@@ -762,7 +762,7 @@ static void rcu_torture_shuffle_tasks(void)
 	else
 		rcu_idle_cpu--;
 
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 /* Shuffle tasks across CPUs, with the intent of allowing each CPU in the
diff --git a/kernel/sched.c b/kernel/sched.c
index 86e55a9c2de6..672aa68bfeac 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7152,7 +7152,7 @@ static int load_balance_monitor(void *unused)
 		int i, cpu, balanced = 1;
 
 		/* Prevent cpus going down or coming up */
-		lock_cpu_hotplug();
+		get_online_cpus();
 		/* lockout changes to doms_cur[] array */
 		lock_doms_cur();
 		/*
@@ -7186,7 +7186,7 @@ static int load_balance_monitor(void *unused)
 		rcu_read_unlock();
 
 		unlock_doms_cur();
-		unlock_cpu_hotplug();
+		put_online_cpus();
 
 		if (!balanced)
 			timeout = sysctl_sched_min_bal_int_shares;
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 319821ef78af..51b5ee53571a 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -203,13 +203,13 @@ int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu)
 	int ret;
 
 	/* No CPUs can come up or down during this. */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	p = __stop_machine_run(fn, data, cpu);
 	if (!IS_ERR(p))
 		ret = kthread_stop(p);
 	else
 		ret = PTR_ERR(p);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 	return ret;
 }
diff --git a/net/core/flow.c b/net/core/flow.c
index 3ed2b4b1d6d4..6489f4e24ecf 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -293,7 +293,7 @@ void flow_cache_flush(void)
 	static DEFINE_MUTEX(flow_flush_sem);
 
 	/* Don't want cpus going down or up during this. */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	mutex_lock(&flow_flush_sem);
 	atomic_set(&info.cpuleft, num_online_cpus());
 	init_completion(&info.completion);
@@ -305,7 +305,7 @@ void flow_cache_flush(void)
 
 	wait_for_completion(&info.completion);
 	mutex_unlock(&flow_flush_sem);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 static void __devinit flow_cache_cpu_prepare(int cpu)
-- 
cgit v1.2.3


From 95402b3829010fe1e208f44e4a158ccade88969a Mon Sep 17 00:00:00 2001
From: Gautham R Shenoy <ego@in.ibm.com>
Date: Fri, 25 Jan 2008 21:08:02 +0100
Subject: cpu-hotplug: replace per-subsystem mutexes with get_online_cpus()

This patch converts the known per-subsystem mutexes to get_online_cpus
put_online_cpus. It also eliminates the CPU_LOCK_ACQUIRE and
CPU_LOCK_RELEASE hotplug notification events.

Signed-off-by: Gautham  R Shenoy <ego@in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/notifier.h |  4 +---
 kernel/cpu.c             |  4 ----
 kernel/sched.c           | 25 +++++++++----------------
 kernel/workqueue.c       | 35 +++++++++++++++--------------------
 mm/slab.c                | 18 +++++++++++-------
 5 files changed, 36 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index 0c40cc0b4a36..5dfbc684ce7d 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -207,9 +207,7 @@ static inline int notifier_to_errno(int ret)
 #define CPU_DOWN_PREPARE	0x0005 /* CPU (unsigned)v going down */
 #define CPU_DOWN_FAILED		0x0006 /* CPU (unsigned)v NOT going down */
 #define CPU_DEAD		0x0007 /* CPU (unsigned)v dead */
-#define CPU_LOCK_ACQUIRE	0x0008 /* Acquire all hotcpu locks */
-#define CPU_LOCK_RELEASE	0x0009 /* Release all hotcpu locks */
-#define CPU_DYING		0x000A /* CPU (unsigned)v not running any task,
+#define CPU_DYING		0x0008 /* CPU (unsigned)v not running any task,
 				        * not handling interrupts, soon dead */
 
 /* Used for CPU hotplug events occuring while tasks are frozen due to a suspend
diff --git a/kernel/cpu.c b/kernel/cpu.c
index b0c4152995f8..e0d3a4f56ecb 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -218,7 +218,6 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 		return -EINVAL;
 
 	cpu_hotplug_begin();
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
 	err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod,
 					hcpu, -1, &nr_calls);
 	if (err == NOTIFY_BAD) {
@@ -271,7 +270,6 @@ out_thread:
 out_allowed:
 	set_cpus_allowed(current, old_allowed);
 out_release:
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
 	cpu_hotplug_done();
 	return err;
 }
@@ -302,7 +300,6 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen)
 		return -EINVAL;
 
 	cpu_hotplug_begin();
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
 	ret = __raw_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE | mod, hcpu,
 							-1, &nr_calls);
 	if (ret == NOTIFY_BAD) {
@@ -326,7 +323,6 @@ out_notify:
 	if (ret != 0)
 		__raw_notifier_call_chain(&cpu_chain,
 				CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL);
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
 	cpu_hotplug_done();
 
 	return ret;
diff --git a/kernel/sched.c b/kernel/sched.c
index 672aa68bfeac..c0e2db683e29 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -439,7 +439,6 @@ struct rq {
 };
 
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
-static DEFINE_MUTEX(sched_hotcpu_mutex);
 
 static inline void check_preempt_curr(struct rq *rq, struct task_struct *p)
 {
@@ -4546,13 +4545,13 @@ long sched_setaffinity(pid_t pid, cpumask_t new_mask)
 	struct task_struct *p;
 	int retval;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	p = find_process_by_pid(pid);
 	if (!p) {
 		read_unlock(&tasklist_lock);
-		mutex_unlock(&sched_hotcpu_mutex);
+		put_online_cpus();
 		return -ESRCH;
 	}
 
@@ -4592,7 +4591,7 @@ long sched_setaffinity(pid_t pid, cpumask_t new_mask)
 	}
 out_unlock:
 	put_task_struct(p);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 	return retval;
 }
 
@@ -4649,7 +4648,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 	struct task_struct *p;
 	int retval;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	retval = -ESRCH;
@@ -4665,7 +4664,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 
 out_unlock:
 	read_unlock(&tasklist_lock);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 
 	return retval;
 }
@@ -5625,9 +5624,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	struct rq *rq;
 
 	switch (action) {
-	case CPU_LOCK_ACQUIRE:
-		mutex_lock(&sched_hotcpu_mutex);
-		break;
 
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
@@ -5697,9 +5693,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		spin_unlock_irq(&rq->lock);
 		break;
 #endif
-	case CPU_LOCK_RELEASE:
-		mutex_unlock(&sched_hotcpu_mutex);
-		break;
 	}
 	return NOTIFY_OK;
 }
@@ -6655,10 +6648,10 @@ static int arch_reinit_sched_domains(void)
 {
 	int err;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	detach_destroy_domains(&cpu_online_map);
 	err = arch_init_sched_domains(&cpu_online_map);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 
 	return err;
 }
@@ -6769,12 +6762,12 @@ void __init sched_init_smp(void)
 {
 	cpumask_t non_isolated_cpus;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	arch_init_sched_domains(&cpu_online_map);
 	cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map);
 	if (cpus_empty(non_isolated_cpus))
 		cpu_set(smp_processor_id(), non_isolated_cpus);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 	/* XXX: Theoretical race here - CPU may be hotplugged now */
 	hotcpu_notifier(update_sched_domains, 0);
 
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 8db0b597509e..52db48e7f6e7 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -67,9 +67,8 @@ struct workqueue_struct {
 #endif
 };
 
-/* All the per-cpu workqueues on the system, for hotplug cpu to add/remove
-   threads to each one as cpus come/go. */
-static DEFINE_MUTEX(workqueue_mutex);
+/* Serializes the accesses to the list of workqueues. */
+static DEFINE_SPINLOCK(workqueue_lock);
 static LIST_HEAD(workqueues);
 
 static int singlethread_cpu __read_mostly;
@@ -592,8 +591,6 @@ EXPORT_SYMBOL(schedule_delayed_work_on);
  * Returns zero on success.
  * Returns -ve errno on failure.
  *
- * Appears to be racy against CPU hotplug.
- *
  * schedule_on_each_cpu() is very slow.
  */
 int schedule_on_each_cpu(work_func_t func)
@@ -605,7 +602,7 @@ int schedule_on_each_cpu(work_func_t func)
 	if (!works)
 		return -ENOMEM;
 
-	preempt_disable();		/* CPU hotplug */
+	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		struct work_struct *work = per_cpu_ptr(works, cpu);
 
@@ -613,8 +610,8 @@ int schedule_on_each_cpu(work_func_t func)
 		set_bit(WORK_STRUCT_PENDING, work_data_bits(work));
 		__queue_work(per_cpu_ptr(keventd_wq->cpu_wq, cpu), work);
 	}
-	preempt_enable();
 	flush_workqueue(keventd_wq);
+	put_online_cpus();
 	free_percpu(works);
 	return 0;
 }
@@ -750,8 +747,10 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 		err = create_workqueue_thread(cwq, singlethread_cpu);
 		start_workqueue_thread(cwq, -1);
 	} else {
-		mutex_lock(&workqueue_mutex);
+		get_online_cpus();
+		spin_lock(&workqueue_lock);
 		list_add(&wq->list, &workqueues);
+		spin_unlock(&workqueue_lock);
 
 		for_each_possible_cpu(cpu) {
 			cwq = init_cpu_workqueue(wq, cpu);
@@ -760,7 +759,7 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 			err = create_workqueue_thread(cwq, cpu);
 			start_workqueue_thread(cwq, cpu);
 		}
-		mutex_unlock(&workqueue_mutex);
+		put_online_cpus();
 	}
 
 	if (err) {
@@ -775,7 +774,7 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu)
 {
 	/*
 	 * Our caller is either destroy_workqueue() or CPU_DEAD,
-	 * workqueue_mutex protects cwq->thread
+	 * get_online_cpus() protects cwq->thread.
 	 */
 	if (cwq->thread == NULL)
 		return;
@@ -810,9 +809,11 @@ void destroy_workqueue(struct workqueue_struct *wq)
 	struct cpu_workqueue_struct *cwq;
 	int cpu;
 
-	mutex_lock(&workqueue_mutex);
+	get_online_cpus();
+	spin_lock(&workqueue_lock);
 	list_del(&wq->list);
-	mutex_unlock(&workqueue_mutex);
+	spin_unlock(&workqueue_lock);
+	put_online_cpus();
 
 	for_each_cpu_mask(cpu, *cpu_map) {
 		cwq = per_cpu_ptr(wq->cpu_wq, cpu);
@@ -835,13 +836,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 	action &= ~CPU_TASKS_FROZEN;
 
 	switch (action) {
-	case CPU_LOCK_ACQUIRE:
-		mutex_lock(&workqueue_mutex);
-		return NOTIFY_OK;
-
-	case CPU_LOCK_RELEASE:
-		mutex_unlock(&workqueue_mutex);
-		return NOTIFY_OK;
 
 	case CPU_UP_PREPARE:
 		cpu_set(cpu, cpu_populated_map);
@@ -854,7 +848,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 		case CPU_UP_PREPARE:
 			if (!create_workqueue_thread(cwq, cpu))
 				break;
-			printk(KERN_ERR "workqueue for %i failed\n", cpu);
+			printk(KERN_ERR "workqueue [%s] for %i failed\n",
+				wq->name, cpu);
 			return NOTIFY_BAD;
 
 		case CPU_ONLINE:
diff --git a/mm/slab.c b/mm/slab.c
index ff31261fd24f..40c00dacbe4b 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -730,8 +730,7 @@ static inline void init_lock_keys(void)
 #endif
 
 /*
- * 1. Guard access to the cache-chain.
- * 2. Protect sanity of cpu_online_map against cpu hotplug events
+ * Guard access to the cache-chain.
  */
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
@@ -1331,12 +1330,11 @@ static int __cpuinit cpuup_callback(struct notifier_block *nfb,
 	int err = 0;
 
 	switch (action) {
-	case CPU_LOCK_ACQUIRE:
-		mutex_lock(&cache_chain_mutex);
-		break;
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
+		mutex_lock(&cache_chain_mutex);
 		err = cpuup_prepare(cpu);
+		mutex_unlock(&cache_chain_mutex);
 		break;
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
@@ -1373,9 +1371,8 @@ static int __cpuinit cpuup_callback(struct notifier_block *nfb,
 #endif
 	case CPU_UP_CANCELED:
 	case CPU_UP_CANCELED_FROZEN:
+		mutex_lock(&cache_chain_mutex);
 		cpuup_canceled(cpu);
-		break;
-	case CPU_LOCK_RELEASE:
 		mutex_unlock(&cache_chain_mutex);
 		break;
 	}
@@ -2170,6 +2167,7 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 	 * We use cache_chain_mutex to ensure a consistent view of
 	 * cpu_online_map as well.  Please see cpuup_callback
 	 */
+	get_online_cpus();
 	mutex_lock(&cache_chain_mutex);
 
 	list_for_each_entry(pc, &cache_chain, next) {
@@ -2396,6 +2394,7 @@ oops:
 		panic("kmem_cache_create(): failed to create slab `%s'\n",
 		      name);
 	mutex_unlock(&cache_chain_mutex);
+	put_online_cpus();
 	return cachep;
 }
 EXPORT_SYMBOL(kmem_cache_create);
@@ -2547,9 +2546,11 @@ int kmem_cache_shrink(struct kmem_cache *cachep)
 	int ret;
 	BUG_ON(!cachep || in_interrupt());
 
+	get_online_cpus();
 	mutex_lock(&cache_chain_mutex);
 	ret = __cache_shrink(cachep);
 	mutex_unlock(&cache_chain_mutex);
+	put_online_cpus();
 	return ret;
 }
 EXPORT_SYMBOL(kmem_cache_shrink);
@@ -2575,6 +2576,7 @@ void kmem_cache_destroy(struct kmem_cache *cachep)
 	BUG_ON(!cachep || in_interrupt());
 
 	/* Find the cache in the chain of caches. */
+	get_online_cpus();
 	mutex_lock(&cache_chain_mutex);
 	/*
 	 * the chain is never empty, cache_cache is never destroyed
@@ -2584,6 +2586,7 @@ void kmem_cache_destroy(struct kmem_cache *cachep)
 		slab_error(cachep, "Can't free all objects");
 		list_add(&cachep->next, &cache_chain);
 		mutex_unlock(&cache_chain_mutex);
+		put_online_cpus();
 		return;
 	}
 
@@ -2592,6 +2595,7 @@ void kmem_cache_destroy(struct kmem_cache *cachep)
 
 	__kmem_cache_destroy(cachep);
 	mutex_unlock(&cache_chain_mutex);
+	put_online_cpus();
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
-- 
cgit v1.2.3


From d0d23b5432fe61229dd3641c5e94d4130bc4e61b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:02 +0100
Subject: cpu-hotplug: fix build on !CONFIG_SMP

fix build on !CONFIG_SMP.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpu.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 3a3ff1c5cbef..0be8d65bc3c8 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -71,19 +71,25 @@ static inline void unregister_cpu_notifier(struct notifier_block *nb)
 
 int cpu_up(unsigned int cpu);
 
+extern void cpu_hotplug_init(void);
+
 #else
 
 static inline int register_cpu_notifier(struct notifier_block *nb)
 {
 	return 0;
 }
+
 static inline void unregister_cpu_notifier(struct notifier_block *nb)
 {
 }
 
+static inline void cpu_hotplug_init(void)
+{
+}
+
 #endif /* CONFIG_SMP */
 extern struct sysdev_class cpu_sysdev_class;
-extern void cpu_hotplug_init(void);
 extern void cpu_maps_update_begin(void);
 extern void cpu_maps_update_done(void);
 
-- 
cgit v1.2.3


From 82a1fcb90287052aabfa235e7ffc693ea003fe69 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:02 +0100
Subject: softlockup: automatically detect hung TASK_UNINTERRUPTIBLE tasks

this patch extends the soft-lockup detector to automatically
detect hung TASK_UNINTERRUPTIBLE tasks. Such hung tasks are
printed the following way:

 ------------------>
 INFO: task prctl:3042 blocked for more than 120 seconds.
 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message
 prctl         D fd5e3793     0  3042   2997
        f6050f38 00000046 00000001 fd5e3793 00000009 c06d8264 c06dae80 00000286
        f6050f40 f6050f00 f7d34d90 f7d34fc8 c1e1be80 00000001 f6050000 00000000
        f7e92d00 00000286 f6050f18 c0489d1a f6050f40 00006605 00000000 c0133a5b
 Call Trace:
  [<c04883a5>] schedule_timeout+0x6d/0x8b
  [<c04883d8>] schedule_timeout_uninterruptible+0x15/0x17
  [<c0133a76>] msleep+0x10/0x16
  [<c0138974>] sys_prctl+0x30/0x1e2
  [<c0104c52>] sysenter_past_esp+0x5f/0xa5
  =======================
 2 locks held by prctl/3042:
 #0:  (&sb->s_type->i_mutex_key#5){--..}, at: [<c0197d11>] do_fsync+0x38/0x7a
 #1:  (jbd_handle){--..}, at: [<c01ca3d2>] journal_start+0xc7/0xe9
 <------------------

the current default timeout is 120 seconds. Such messages are printed
up to 10 times per bootup. If the system has crashed already then the
messages are not printed.

if lockdep is enabled then all held locks are printed as well.

this feature is a natural extension to the softlockup-detector (kernel
locked up without scheduling) and to the NMI watchdog (kernel locked up
with IRQs disabled).

[ Gautham R Shenoy <ego@in.ibm.com>: CPU hotplug fixes. ]
[ Andrew Morton <akpm@linux-foundation.org>: build warning fix. ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 include/linux/debug_locks.h |   5 ++
 include/linux/sched.h       |  10 ++++
 kernel/fork.c               |   5 ++
 kernel/lockdep.c            |  12 ++++-
 kernel/sched.c              |   4 +-
 kernel/softlockup.c         | 114 ++++++++++++++++++++++++++++++++++++++++----
 kernel/sysctl.c             |  27 +++++++++++
 7 files changed, 164 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h
index 1678a5de7013..f4a5871767f5 100644
--- a/include/linux/debug_locks.h
+++ b/include/linux/debug_locks.h
@@ -47,6 +47,7 @@ struct task_struct;
 
 #ifdef CONFIG_LOCKDEP
 extern void debug_show_all_locks(void);
+extern void __debug_show_held_locks(struct task_struct *task);
 extern void debug_show_held_locks(struct task_struct *task);
 extern void debug_check_no_locks_freed(const void *from, unsigned long len);
 extern void debug_check_no_locks_held(struct task_struct *task);
@@ -55,6 +56,10 @@ static inline void debug_show_all_locks(void)
 {
 }
 
+static inline void __debug_show_held_locks(struct task_struct *task)
+{
+}
+
 static inline void debug_show_held_locks(struct task_struct *task)
 {
 }
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 288245f83bd4..0846f1f9e196 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -258,12 +258,17 @@ extern void account_process_tick(struct task_struct *task, int user);
 extern void update_process_times(int user);
 extern void scheduler_tick(void);
 
+extern void sched_show_task(struct task_struct *p);
+
 #ifdef CONFIG_DETECT_SOFTLOCKUP
 extern void softlockup_tick(void);
 extern void spawn_softlockup_task(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
 extern int softlockup_thresh;
+extern unsigned long sysctl_hung_task_check_count;
+extern unsigned long sysctl_hung_task_timeout_secs;
+extern long sysctl_hung_task_warnings;
 #else
 static inline void softlockup_tick(void)
 {
@@ -1041,6 +1046,11 @@ struct task_struct {
 /* ipc stuff */
 	struct sysv_sem sysvsem;
 #endif
+#ifdef CONFIG_DETECT_SOFTLOCKUP
+/* hung task detection */
+	unsigned long last_switch_timestamp;
+	unsigned long last_switch_count;
+#endif
 /* CPU-specific state of this task */
 	struct thread_struct thread;
 /* filesystem information */
diff --git a/kernel/fork.c b/kernel/fork.c
index 8dd8ff281009..09c0b90a69cc 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1059,6 +1059,11 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->prev_utime = cputime_zero;
 	p->prev_stime = cputime_zero;
 
+#ifdef CONFIG_DETECT_SOFTLOCKUP
+	p->last_switch_count = 0;
+	p->last_switch_timestamp = 0;
+#endif
+
 #ifdef CONFIG_TASK_XACCT
 	p->rchar = 0;		/* I/O counter: bytes read */
 	p->wchar = 0;		/* I/O counter: bytes written */
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index e2c07ece367d..3574379f4d62 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -3206,7 +3206,11 @@ retry:
 
 EXPORT_SYMBOL_GPL(debug_show_all_locks);
 
-void debug_show_held_locks(struct task_struct *task)
+/*
+ * Careful: only use this function if you are sure that
+ * the task cannot run in parallel!
+ */
+void __debug_show_held_locks(struct task_struct *task)
 {
 	if (unlikely(!debug_locks)) {
 		printk("INFO: lockdep is turned off.\n");
@@ -3214,6 +3218,12 @@ void debug_show_held_locks(struct task_struct *task)
 	}
 	lockdep_print_held_locks(task);
 }
+EXPORT_SYMBOL_GPL(__debug_show_held_locks);
+
+void debug_show_held_locks(struct task_struct *task)
+{
+		__debug_show_held_locks(task);
+}
 
 EXPORT_SYMBOL_GPL(debug_show_held_locks);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index c0e2db683e29..5b3d46574eeb 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4945,7 +4945,7 @@ out_unlock:
 
 static const char stat_nam[] = "RSDTtZX";
 
-static void show_task(struct task_struct *p)
+void sched_show_task(struct task_struct *p)
 {
 	unsigned long free = 0;
 	unsigned state;
@@ -4998,7 +4998,7 @@ void show_state_filter(unsigned long state_filter)
 		 */
 		touch_nmi_watchdog();
 		if (!state_filter || (p->state & state_filter))
-			show_task(p);
+			sched_show_task(p);
 	} while_each_thread(g, p);
 
 	touch_all_softlockup_watchdogs();
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 11df812263c8..02f0ad534441 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -8,6 +8,7 @@
  */
 #include <linux/mm.h>
 #include <linux/cpu.h>
+#include <linux/nmi.h>
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/freezer.h>
@@ -24,7 +25,7 @@ static DEFINE_PER_CPU(unsigned long, print_timestamp);
 static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
 
 static int did_panic;
-int softlockup_thresh = 10;
+int softlockup_thresh = 60;
 
 static int
 softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
@@ -45,7 +46,7 @@ static struct notifier_block panic_block = {
  */
 static unsigned long get_timestamp(int this_cpu)
 {
-	return cpu_clock(this_cpu) >> 30;  /* 2^30 ~= 10^9 */
+	return cpu_clock(this_cpu) >> 30LL;  /* 2^30 ~= 10^9 */
 }
 
 void touch_softlockup_watchdog(void)
@@ -100,11 +101,7 @@ void softlockup_tick(void)
 
 	now = get_timestamp(this_cpu);
 
-	/* Wake up the high-prio watchdog task every second: */
-	if (now > (touch_timestamp + 1))
-		wake_up_process(per_cpu(watchdog_task, this_cpu));
-
-	/* Warn about unreasonable 10+ seconds delays: */
+	/* Warn about unreasonable delays: */
 	if (now <= (touch_timestamp + softlockup_thresh))
 		return;
 
@@ -121,12 +118,94 @@ void softlockup_tick(void)
 	spin_unlock(&print_lock);
 }
 
+/*
+ * Have a reasonable limit on the number of tasks checked:
+ */
+unsigned long sysctl_hung_task_check_count = 1024;
+
+/*
+ * Zero means infinite timeout - no checking done:
+ */
+unsigned long sysctl_hung_task_timeout_secs = 120;
+
+long sysctl_hung_task_warnings = 10;
+
+/*
+ * Only do the hung-tasks check on one CPU:
+ */
+static int check_cpu __read_mostly = -1;
+
+static void check_hung_task(struct task_struct *t, unsigned long now)
+{
+	unsigned long switch_count = t->nvcsw + t->nivcsw;
+
+	if (t->flags & PF_FROZEN)
+		return;
+
+	if (switch_count != t->last_switch_count || !t->last_switch_timestamp) {
+		t->last_switch_count = switch_count;
+		t->last_switch_timestamp = now;
+		return;
+	}
+	if ((long)(now - t->last_switch_timestamp) <
+					sysctl_hung_task_timeout_secs)
+		return;
+	if (sysctl_hung_task_warnings < 0)
+		return;
+	sysctl_hung_task_warnings--;
+
+	/*
+	 * Ok, the task did not get scheduled for more than 2 minutes,
+	 * complain:
+	 */
+	printk(KERN_ERR "INFO: task %s:%d blocked for more than "
+			"%ld seconds.\n", t->comm, t->pid,
+			sysctl_hung_task_timeout_secs);
+	printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
+			" disables this message.\n");
+	sched_show_task(t);
+	__debug_show_held_locks(t);
+
+	t->last_switch_timestamp = now;
+	touch_nmi_watchdog();
+}
+
+/*
+ * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for
+ * a really long time (120 seconds). If that happens, print out
+ * a warning.
+ */
+static void check_hung_uninterruptible_tasks(int this_cpu)
+{
+	int max_count = sysctl_hung_task_check_count;
+	unsigned long now = get_timestamp(this_cpu);
+	struct task_struct *g, *t;
+
+	/*
+	 * If the system crashed already then all bets are off,
+	 * do not report extra hung tasks:
+	 */
+	if ((tainted & TAINT_DIE) || did_panic)
+		return;
+
+	read_lock(&tasklist_lock);
+	do_each_thread(g, t) {
+		if (!--max_count)
+			break;
+		if (t->state & TASK_UNINTERRUPTIBLE)
+			check_hung_task(t, now);
+	} while_each_thread(g, t);
+
+	read_unlock(&tasklist_lock);
+}
+
 /*
  * The watchdog thread - runs every second and touches the timestamp.
  */
 static int watchdog(void *__bind_cpu)
 {
 	struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+	int this_cpu = (long)__bind_cpu;
 
 	sched_setscheduler(current, SCHED_FIFO, &param);
 
@@ -135,13 +214,18 @@ static int watchdog(void *__bind_cpu)
 
 	/*
 	 * Run briefly once per second to reset the softlockup timestamp.
-	 * If this gets delayed for more than 10 seconds then the
+	 * If this gets delayed for more than 60 seconds then the
 	 * debug-printout triggers in softlockup_tick().
 	 */
 	while (!kthread_should_stop()) {
-		set_current_state(TASK_INTERRUPTIBLE);
 		touch_softlockup_watchdog();
-		schedule();
+		msleep_interruptible(10000);
+
+		if (this_cpu != check_cpu)
+			continue;
+
+		if (sysctl_hung_task_timeout_secs)
+			check_hung_uninterruptible_tasks(this_cpu);
 	}
 
 	return 0;
@@ -171,6 +255,7 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		break;
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
+		check_cpu = any_online_cpu(cpu_online_map);
 		wake_up_process(per_cpu(watchdog_task, hotcpu));
 		break;
 #ifdef CONFIG_HOTPLUG_CPU
@@ -181,6 +266,15 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		/* Unbind so it can run.  Fall thru. */
 		kthread_bind(per_cpu(watchdog_task, hotcpu),
 			     any_online_cpu(cpu_online_map));
+	case CPU_DOWN_PREPARE:
+	case CPU_DOWN_PREPARE_FROZEN:
+		if (hotcpu == check_cpu) {
+			cpumask_t temp_cpu_online_map = cpu_online_map;
+
+			cpu_clear(hotcpu, temp_cpu_online_map);
+			check_cpu = any_online_cpu(temp_cpu_online_map);
+		}
+		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 		p = per_cpu(watchdog_task, hotcpu);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index c95f3ed34474..96f31c1bc4f0 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -753,6 +753,33 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &one,
 		.extra2		= &sixty,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_check_count",
+		.data		= &sysctl_hung_task_check_count,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_timeout_secs",
+		.data		= &sysctl_hung_task_timeout_secs,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_warnings",
+		.data		= &sysctl_hung_task_warnings,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
 #endif
 #ifdef CONFIG_COMPAT
 	{
-- 
cgit v1.2.3


From 73fe6aae84400e2b475e2a1dc4e8592cd3ed6e69 Mon Sep 17 00:00:00 2001
From: Gregory Haskins <ghaskins@novell.com>
Date: Fri, 25 Jan 2008 21:08:07 +0100
Subject: sched: add RT-balance cpu-weight

Some RT tasks (particularly kthreads) are bound to one specific CPU.
It is fairly common for two or more bound tasks to get queued up at the
same time.  Consider, for instance, softirq_timer and softirq_sched.  A
timer goes off in an ISR which schedules softirq_thread to run at RT50.
Then the timer handler determines that it's time to smp-rebalance the
system so it schedules softirq_sched to run.  So we are in a situation
where we have two RT50 tasks queued, and the system will go into
rt-overload condition to request other CPUs for help.

This causes two problems in the current code:

1) If a high-priority bound task and a low-priority unbounded task queue
   up behind the running task, we will fail to ever relocate the unbounded
   task because we terminate the search on the first unmovable task.

2) We spend precious futile cycles in the fast-path trying to pull
   overloaded tasks over.  It is therefore optimial to strive to avoid the
   overhead all together if we can cheaply detect the condition before
   overload even occurs.

This patch tries to achieve this optimization by utilizing the hamming
weight of the task->cpus_allowed mask.  A weight of 1 indicates that
the task cannot be migrated.  We will then utilize this information to
skip non-migratable tasks and to eliminate uncessary rebalance attempts.

We introduce a per-rq variable to count the number of migratable tasks
that are currently running.  We only go into overload if we have more
than one rt task, AND at least one of them is migratable.

In addition, we introduce a per-task variable to cache the cpus_allowed
weight, since the hamming calculation is probably relatively expensive.
We only update the cached value when the mask is updated which should be
relatively infrequent, especially compared to scheduling frequency
in the fast path.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/init_task.h |  1 +
 include/linux/sched.h     |  2 ++
 kernel/fork.c             |  1 +
 kernel/sched.c            |  9 ++++++++-
 kernel/sched_rt.c         | 50 ++++++++++++++++++++++++++++++++++++++++++-----
 5 files changed, 57 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index cae35b6b9aec..572c65bcc80f 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -130,6 +130,7 @@ extern struct group_info init_groups;
 	.normal_prio	= MAX_PRIO-20,					\
 	.policy		= SCHED_NORMAL,					\
 	.cpus_allowed	= CPU_MASK_ALL,					\
+	.nr_cpus_allowed = NR_CPUS,					\
 	.mm		= NULL,						\
 	.active_mm	= &init_mm,					\
 	.run_list	= LIST_HEAD_INIT(tsk.run_list),			\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 0846f1f9e196..b07a2cf76401 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -847,6 +847,7 @@ struct sched_class {
 	void (*set_curr_task) (struct rq *rq);
 	void (*task_tick) (struct rq *rq, struct task_struct *p);
 	void (*task_new) (struct rq *rq, struct task_struct *p);
+	void (*set_cpus_allowed)(struct task_struct *p, cpumask_t *newmask);
 };
 
 struct load_weight {
@@ -956,6 +957,7 @@ struct task_struct {
 
 	unsigned int policy;
 	cpumask_t cpus_allowed;
+	int nr_cpus_allowed;
 	unsigned int time_slice;
 
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
diff --git a/kernel/fork.c b/kernel/fork.c
index 09c0b90a69cc..930c51865ab4 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1242,6 +1242,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 * parent's CPU). This avoids alot of nasty races.
 	 */
 	p->cpus_allowed = current->cpus_allowed;
+	p->nr_cpus_allowed = current->nr_cpus_allowed;
 	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) ||
 			!cpu_online(task_cpu(p))))
 		set_task_cpu(p, smp_processor_id());
diff --git a/kernel/sched.c b/kernel/sched.c
index 357d3a084de8..66e99b419b31 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -343,6 +343,7 @@ struct rt_rq {
 	int rt_load_balance_idx;
 	struct list_head *rt_load_balance_head, *rt_load_balance_curr;
 	unsigned long rt_nr_running;
+	unsigned long rt_nr_migratory;
 	/* highest queued rt task prio */
 	int highest_prio;
 };
@@ -5144,7 +5145,13 @@ int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
 		goto out;
 	}
 
-	p->cpus_allowed = new_mask;
+	if (p->sched_class->set_cpus_allowed)
+		p->sched_class->set_cpus_allowed(p, &new_mask);
+	else {
+		p->cpus_allowed    = new_mask;
+		p->nr_cpus_allowed = cpus_weight(new_mask);
+	}
+
 	/* Can the task run on the task's current CPU? If so, we're done */
 	if (cpu_isset(task_cpu(p), new_mask))
 		goto out;
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index c492fd2b2eec..ae4995c09aac 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -33,6 +33,14 @@ static inline void rt_clear_overload(struct rq *rq)
 	atomic_dec(&rto_count);
 	cpu_clear(rq->cpu, rt_overload_mask);
 }
+
+static void update_rt_migration(struct rq *rq)
+{
+	if (rq->rt.rt_nr_migratory && (rq->rt.rt_nr_running > 1))
+		rt_set_overload(rq);
+	else
+		rt_clear_overload(rq);
+}
 #endif /* CONFIG_SMP */
 
 /*
@@ -65,8 +73,10 @@ static inline void inc_rt_tasks(struct task_struct *p, struct rq *rq)
 #ifdef CONFIG_SMP
 	if (p->prio < rq->rt.highest_prio)
 		rq->rt.highest_prio = p->prio;
-	if (rq->rt.rt_nr_running > 1)
-		rt_set_overload(rq);
+	if (p->nr_cpus_allowed > 1)
+		rq->rt.rt_nr_migratory++;
+
+	update_rt_migration(rq);
 #endif /* CONFIG_SMP */
 }
 
@@ -88,8 +98,10 @@ static inline void dec_rt_tasks(struct task_struct *p, struct rq *rq)
 		} /* otherwise leave rq->highest prio alone */
 	} else
 		rq->rt.highest_prio = MAX_RT_PRIO;
-	if (rq->rt.rt_nr_running < 2)
-		rt_clear_overload(rq);
+	if (p->nr_cpus_allowed > 1)
+		rq->rt.rt_nr_migratory--;
+
+	update_rt_migration(rq);
 #endif /* CONFIG_SMP */
 }
 
@@ -182,7 +194,8 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep);
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)))
+	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
+	    (p->nr_cpus_allowed > 1))
 		return 1;
 	return 0;
 }
@@ -584,6 +597,32 @@ move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
 	/* don't touch RT tasks */
 	return 0;
 }
+static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
+{
+	int weight = cpus_weight(*new_mask);
+
+	BUG_ON(!rt_task(p));
+
+	/*
+	 * Update the migration status of the RQ if we have an RT task
+	 * which is running AND changing its weight value.
+	 */
+	if (p->se.on_rq && (weight != p->nr_cpus_allowed)) {
+		struct rq *rq = task_rq(p);
+
+		if ((p->nr_cpus_allowed <= 1) && (weight > 1))
+			rq->rt.rt_nr_migratory++;
+		else if((p->nr_cpus_allowed > 1) && (weight <= 1)) {
+			BUG_ON(!rq->rt.rt_nr_migratory);
+			rq->rt.rt_nr_migratory--;
+		}
+
+		update_rt_migration(rq);
+	}
+
+	p->cpus_allowed    = *new_mask;
+	p->nr_cpus_allowed = weight;
+}
 #else /* CONFIG_SMP */
 # define schedule_tail_balance_rt(rq)	do { } while (0)
 # define schedule_balance_rt(rq, prev)	do { } while (0)
@@ -637,6 +676,7 @@ const struct sched_class rt_sched_class = {
 #ifdef CONFIG_SMP
 	.load_balance		= load_balance_rt,
 	.move_one_task		= move_one_task_rt,
+	.set_cpus_allowed       = set_cpus_allowed_rt,
 #endif
 
 	.set_curr_task          = set_curr_task_rt,
-- 
cgit v1.2.3


From e7693a362ec84bb5b6fd441d8a8b4b9d568a7a0c Mon Sep 17 00:00:00 2001
From: Gregory Haskins <ghaskins@novell.com>
Date: Fri, 25 Jan 2008 21:08:09 +0100
Subject: sched: de-SCHED_OTHER-ize the RT path

The current wake-up code path tries to determine if it can optimize the
wake-up to "this_cpu" by computing load calculations.  The problem is that
these calculations are only relevant to SCHED_OTHER tasks where load is king.
For RT tasks, priority is king.  So the load calculation is completely wasted
bandwidth.

Therefore, we create a new sched_class interface to help with
pre-wakeup routing decisions and move the load calculation as a function
of CFS task's class.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h   |   1 +
 kernel/sched.c          | 167 ++++++++----------------------------------------
 kernel/sched_fair.c     | 148 ++++++++++++++++++++++++++++++++++++++++++
 kernel/sched_idletask.c |   9 +++
 kernel/sched_rt.c       |  10 +++
 5 files changed, 195 insertions(+), 140 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index b07a2cf76401..6fbbf38555ac 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -827,6 +827,7 @@ struct sched_class {
 	void (*enqueue_task) (struct rq *rq, struct task_struct *p, int wakeup);
 	void (*dequeue_task) (struct rq *rq, struct task_struct *p, int sleep);
 	void (*yield_task) (struct rq *rq);
+	int  (*select_task_rq)(struct task_struct *p, int sync);
 
 	void (*check_preempt_curr) (struct rq *rq, struct task_struct *p);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 66e99b419b31..3344ba776b97 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -960,6 +960,13 @@ static inline void dec_cpu_load(struct rq *rq, unsigned long load)
 	update_load_sub(&rq->load, load);
 }
 
+#ifdef CONFIG_SMP
+static unsigned long source_load(int cpu, int type);
+static unsigned long target_load(int cpu, int type);
+static unsigned long cpu_avg_load_per_task(int cpu);
+static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
+#endif /* CONFIG_SMP */
+
 #include "sched_stats.h"
 #include "sched_idletask.c"
 #include "sched_fair.c"
@@ -1118,7 +1125,7 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 /*
  * Is this task likely cache-hot:
  */
-static inline int
+static int
 task_hot(struct task_struct *p, u64 now, struct sched_domain *sd)
 {
 	s64 delta;
@@ -1343,7 +1350,7 @@ static unsigned long target_load(int cpu, int type)
 /*
  * Return the average load per task on the cpu's run queue
  */
-static inline unsigned long cpu_avg_load_per_task(int cpu)
+static unsigned long cpu_avg_load_per_task(int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 	unsigned long total = weighted_cpuload(cpu);
@@ -1500,58 +1507,6 @@ static int sched_balance_self(int cpu, int flag)
 
 #endif /* CONFIG_SMP */
 
-/*
- * wake_idle() will wake a task on an idle cpu if task->cpu is
- * not idle and an idle cpu is available.  The span of cpus to
- * search starts with cpus closest then further out as needed,
- * so we always favor a closer, idle cpu.
- *
- * Returns the CPU we should wake onto.
- */
-#if defined(ARCH_HAS_SCHED_WAKE_IDLE)
-static int wake_idle(int cpu, struct task_struct *p)
-{
-	cpumask_t tmp;
-	struct sched_domain *sd;
-	int i;
-
-	/*
-	 * If it is idle, then it is the best cpu to run this task.
-	 *
-	 * This cpu is also the best, if it has more than one task already.
-	 * Siblings must be also busy(in most cases) as they didn't already
-	 * pickup the extra load from this cpu and hence we need not check
-	 * sibling runqueue info. This will avoid the checks and cache miss
-	 * penalities associated with that.
-	 */
-	if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1)
-		return cpu;
-
-	for_each_domain(cpu, sd) {
-		if (sd->flags & SD_WAKE_IDLE) {
-			cpus_and(tmp, sd->span, p->cpus_allowed);
-			for_each_cpu_mask(i, tmp) {
-				if (idle_cpu(i)) {
-					if (i != task_cpu(p)) {
-						schedstat_inc(p,
-							se.nr_wakeups_idle);
-					}
-					return i;
-				}
-			}
-		} else {
-			break;
-		}
-	}
-	return cpu;
-}
-#else
-static inline int wake_idle(int cpu, struct task_struct *p)
-{
-	return cpu;
-}
-#endif
-
 /***
  * try_to_wake_up - wake up a thread
  * @p: the to-be-woken-up thread
@@ -1573,8 +1528,6 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	long old_state;
 	struct rq *rq;
 #ifdef CONFIG_SMP
-	struct sched_domain *sd, *this_sd = NULL;
-	unsigned long load, this_load;
 	int new_cpu;
 #endif
 
@@ -1594,90 +1547,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	if (unlikely(task_running(rq, p)))
 		goto out_activate;
 
-	new_cpu = cpu;
-
-	schedstat_inc(rq, ttwu_count);
-	if (cpu == this_cpu) {
-		schedstat_inc(rq, ttwu_local);
-		goto out_set_cpu;
-	}
-
-	for_each_domain(this_cpu, sd) {
-		if (cpu_isset(cpu, sd->span)) {
-			schedstat_inc(sd, ttwu_wake_remote);
-			this_sd = sd;
-			break;
-		}
-	}
-
-	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
-		goto out_set_cpu;
-
-	/*
-	 * Check for affine wakeup and passive balancing possibilities.
-	 */
-	if (this_sd) {
-		int idx = this_sd->wake_idx;
-		unsigned int imbalance;
-
-		imbalance = 100 + (this_sd->imbalance_pct - 100) / 2;
-
-		load = source_load(cpu, idx);
-		this_load = target_load(this_cpu, idx);
-
-		new_cpu = this_cpu; /* Wake to this CPU if we can */
-
-		if (this_sd->flags & SD_WAKE_AFFINE) {
-			unsigned long tl = this_load;
-			unsigned long tl_per_task;
-
-			/*
-			 * Attract cache-cold tasks on sync wakeups:
-			 */
-			if (sync && !task_hot(p, rq->clock, this_sd))
-				goto out_set_cpu;
-
-			schedstat_inc(p, se.nr_wakeups_affine_attempts);
-			tl_per_task = cpu_avg_load_per_task(this_cpu);
-
-			/*
-			 * If sync wakeup then subtract the (maximum possible)
-			 * effect of the currently running task from the load
-			 * of the current CPU:
-			 */
-			if (sync)
-				tl -= current->se.load.weight;
-
-			if ((tl <= load &&
-				tl + target_load(cpu, idx) <= tl_per_task) ||
-			       100*(tl + p->se.load.weight) <= imbalance*load) {
-				/*
-				 * This domain has SD_WAKE_AFFINE and
-				 * p is cache cold in this domain, and
-				 * there is no bad imbalance.
-				 */
-				schedstat_inc(this_sd, ttwu_move_affine);
-				schedstat_inc(p, se.nr_wakeups_affine);
-				goto out_set_cpu;
-			}
-		}
-
-		/*
-		 * Start passive balancing when half the imbalance_pct
-		 * limit is reached.
-		 */
-		if (this_sd->flags & SD_WAKE_BALANCE) {
-			if (imbalance*this_load <= 100*load) {
-				schedstat_inc(this_sd, ttwu_move_balance);
-				schedstat_inc(p, se.nr_wakeups_passive);
-				goto out_set_cpu;
-			}
-		}
-	}
-
-	new_cpu = cpu; /* Could not wake to this_cpu. Wake to cpu instead */
-out_set_cpu:
-	new_cpu = wake_idle(new_cpu, p);
+	new_cpu = p->sched_class->select_task_rq(p, sync);
 	if (new_cpu != cpu) {
 		set_task_cpu(p, new_cpu);
 		task_rq_unlock(rq, &flags);
@@ -1693,6 +1563,23 @@ out_set_cpu:
 		cpu = task_cpu(p);
 	}
 
+#ifdef CONFIG_SCHEDSTATS
+	schedstat_inc(rq, ttwu_count);
+	if (cpu == this_cpu)
+		schedstat_inc(rq, ttwu_local);
+	else {
+		struct sched_domain *sd;
+		for_each_domain(this_cpu, sd) {
+			if (cpu_isset(cpu, sd->span)) {
+				schedstat_inc(sd, ttwu_wake_remote);
+				break;
+			}
+		}
+	}
+
+#endif
+
+
 out_activate:
 #endif /* CONFIG_SMP */
 	schedstat_inc(p, se.nr_wakeups);
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 5c208e090ae4..f881fc5e035c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -860,6 +860,151 @@ static void yield_task_fair(struct rq *rq)
 	se->vruntime = rightmost->vruntime + 1;
 }
 
+/*
+ * wake_idle() will wake a task on an idle cpu if task->cpu is
+ * not idle and an idle cpu is available.  The span of cpus to
+ * search starts with cpus closest then further out as needed,
+ * so we always favor a closer, idle cpu.
+ *
+ * Returns the CPU we should wake onto.
+ */
+#if defined(ARCH_HAS_SCHED_WAKE_IDLE)
+static int wake_idle(int cpu, struct task_struct *p)
+{
+	cpumask_t tmp;
+	struct sched_domain *sd;
+	int i;
+
+	/*
+	 * If it is idle, then it is the best cpu to run this task.
+	 *
+	 * This cpu is also the best, if it has more than one task already.
+	 * Siblings must be also busy(in most cases) as they didn't already
+	 * pickup the extra load from this cpu and hence we need not check
+	 * sibling runqueue info. This will avoid the checks and cache miss
+	 * penalities associated with that.
+	 */
+	if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1)
+		return cpu;
+
+	for_each_domain(cpu, sd) {
+		if (sd->flags & SD_WAKE_IDLE) {
+			cpus_and(tmp, sd->span, p->cpus_allowed);
+			for_each_cpu_mask(i, tmp) {
+				if (idle_cpu(i)) {
+					if (i != task_cpu(p)) {
+						schedstat_inc(p,
+						       se.nr_wakeups_idle);
+					}
+					return i;
+				}
+			}
+		} else {
+			break;
+		}
+	}
+	return cpu;
+}
+#else
+static inline int wake_idle(int cpu, struct task_struct *p)
+{
+	return cpu;
+}
+#endif
+
+#ifdef CONFIG_SMP
+static int select_task_rq_fair(struct task_struct *p, int sync)
+{
+	int cpu, this_cpu;
+	struct rq *rq;
+	struct sched_domain *sd, *this_sd = NULL;
+	int new_cpu;
+
+	cpu      = task_cpu(p);
+	rq       = task_rq(p);
+	this_cpu = smp_processor_id();
+	new_cpu  = cpu;
+
+	for_each_domain(this_cpu, sd) {
+		if (cpu_isset(cpu, sd->span)) {
+			this_sd = sd;
+			break;
+		}
+	}
+
+	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
+		goto out_set_cpu;
+
+	/*
+	 * Check for affine wakeup and passive balancing possibilities.
+	 */
+	if (this_sd) {
+		int idx = this_sd->wake_idx;
+		unsigned int imbalance;
+		unsigned long load, this_load;
+
+		imbalance = 100 + (this_sd->imbalance_pct - 100) / 2;
+
+		load = source_load(cpu, idx);
+		this_load = target_load(this_cpu, idx);
+
+		new_cpu = this_cpu; /* Wake to this CPU if we can */
+
+		if (this_sd->flags & SD_WAKE_AFFINE) {
+			unsigned long tl = this_load;
+			unsigned long tl_per_task;
+
+			/*
+			 * Attract cache-cold tasks on sync wakeups:
+			 */
+			if (sync && !task_hot(p, rq->clock, this_sd))
+				goto out_set_cpu;
+
+			schedstat_inc(p, se.nr_wakeups_affine_attempts);
+			tl_per_task = cpu_avg_load_per_task(this_cpu);
+
+			/*
+			 * If sync wakeup then subtract the (maximum possible)
+			 * effect of the currently running task from the load
+			 * of the current CPU:
+			 */
+			if (sync)
+				tl -= current->se.load.weight;
+
+			if ((tl <= load &&
+				tl + target_load(cpu, idx) <= tl_per_task) ||
+			       100*(tl + p->se.load.weight) <= imbalance*load) {
+				/*
+				 * This domain has SD_WAKE_AFFINE and
+				 * p is cache cold in this domain, and
+				 * there is no bad imbalance.
+				 */
+				schedstat_inc(this_sd, ttwu_move_affine);
+				schedstat_inc(p, se.nr_wakeups_affine);
+				goto out_set_cpu;
+			}
+		}
+
+		/*
+		 * Start passive balancing when half the imbalance_pct
+		 * limit is reached.
+		 */
+		if (this_sd->flags & SD_WAKE_BALANCE) {
+			if (imbalance*this_load <= 100*load) {
+				schedstat_inc(this_sd, ttwu_move_balance);
+				schedstat_inc(p, se.nr_wakeups_passive);
+				goto out_set_cpu;
+			}
+		}
+	}
+
+	new_cpu = cpu; /* Could not wake to this_cpu. Wake to cpu instead */
+out_set_cpu:
+	return wake_idle(new_cpu, p);
+}
+#endif /* CONFIG_SMP */
+
+
 /*
  * Preempt the current task with a newly woken task if needed:
  */
@@ -1153,6 +1298,9 @@ static const struct sched_class fair_sched_class = {
 	.enqueue_task		= enqueue_task_fair,
 	.dequeue_task		= dequeue_task_fair,
 	.yield_task		= yield_task_fair,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_fair,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_wakeup,
 
diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c
index bf9c25c15b8b..ca5374860aef 100644
--- a/kernel/sched_idletask.c
+++ b/kernel/sched_idletask.c
@@ -5,6 +5,12 @@
  *  handled in sched_fair.c)
  */
 
+#ifdef CONFIG_SMP
+static int select_task_rq_idle(struct task_struct *p, int sync)
+{
+	return task_cpu(p); /* IDLE tasks as never migrated */
+}
+#endif /* CONFIG_SMP */
 /*
  * Idle tasks are unconditionally rescheduled:
  */
@@ -72,6 +78,9 @@ const struct sched_class idle_sched_class = {
 
 	/* dequeue is not valid, we print a debug message there: */
 	.dequeue_task		= dequeue_task_idle,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_idle,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_curr_idle,
 
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index b788e35ffd3f..5de1aebdbd1b 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -150,6 +150,13 @@ yield_task_rt(struct rq *rq)
 	requeue_task_rt(rq, rq->curr);
 }
 
+#ifdef CONFIG_SMP
+static int select_task_rq_rt(struct task_struct *p, int sync)
+{
+	return task_cpu(p);
+}
+#endif /* CONFIG_SMP */
+
 /*
  * Preempt the current task with a newly woken task if needed:
  */
@@ -667,6 +674,9 @@ const struct sched_class rt_sched_class = {
 	.enqueue_task		= enqueue_task_rt,
 	.dequeue_task		= dequeue_task_rt,
 	.yield_task		= yield_task_rt,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_rt,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_curr_rt,
 
-- 
cgit v1.2.3


From 57d885fea0da0e9541d7730a9e1dcf734981a173 Mon Sep 17 00:00:00 2001
From: Gregory Haskins <ghaskins@novell.com>
Date: Fri, 25 Jan 2008 21:08:18 +0100
Subject: sched: add sched-domain roots

We add the notion of a root-domain which will be used later to rescope
global variables to per-domain variables.  Each exclusive cpuset
essentially defines an island domain by fully partitioning the member cpus
from any other cpuset.  However, we currently still maintain some
policy/state as global variables which transcend all cpusets.  Consider,
for instance, rt-overload state.

Whenever a new exclusive cpuset is created, we also create a new
root-domain object and move each cpu member to the root-domain's span.
By default the system creates a single root-domain with all cpus as
members (mimicking the global state we have today).

We add some plumbing for storing class specific data in our root-domain.
Whenever a RQ is switching root-domains (because of repartitioning) we
give each sched_class the opportunity to remove any state from its old
domain and add state to the new one.  This logic doesn't have any clients
yet but it will later in the series.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
CC: Christoph Lameter <clameter@sgi.com>
CC: Paul Jackson <pj@sgi.com>
CC: Simon Derr <simon.derr@bull.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |   3 ++
 kernel/sched.c        | 121 ++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 121 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6fbbf38555ac..2e69f19369e4 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -849,6 +849,9 @@ struct sched_class {
 	void (*task_tick) (struct rq *rq, struct task_struct *p);
 	void (*task_new) (struct rq *rq, struct task_struct *p);
 	void (*set_cpus_allowed)(struct task_struct *p, cpumask_t *newmask);
+
+	void (*join_domain)(struct rq *rq);
+	void (*leave_domain)(struct rq *rq);
 };
 
 struct load_weight {
diff --git a/kernel/sched.c b/kernel/sched.c
index 36bd8ff2a669..34b7d721d735 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -349,6 +349,28 @@ struct rt_rq {
 	int overloaded;
 };
 
+#ifdef CONFIG_SMP
+
+/*
+ * We add the notion of a root-domain which will be used to define per-domain
+ * variables.  Each exclusive cpuset essentially defines an island domain by
+ * fully partitioning the member cpus from any other cpuset.  Whenever a new
+ * exclusive cpuset is created, we also create and attach a new root-domain
+ * object.
+ *
+ * By default the system creates a single root-domain with all cpus as
+ * members (mimicking the global state we have today).
+ */
+struct root_domain {
+	atomic_t refcount;
+	cpumask_t span;
+	cpumask_t online;
+};
+
+static struct root_domain def_root_domain;
+
+#endif
+
 /*
  * This is the main, per-CPU runqueue data structure.
  *
@@ -406,6 +428,7 @@ struct rq {
 	atomic_t nr_iowait;
 
 #ifdef CONFIG_SMP
+	struct root_domain  *rd;
 	struct sched_domain *sd;
 
 	/* For active balancing */
@@ -5550,6 +5573,15 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	case CPU_ONLINE_FROZEN:
 		/* Strictly unnecessary, as first user will wake it. */
 		wake_up_process(cpu_rq(cpu)->migration_thread);
+
+		/* Update our root-domain */
+		rq = cpu_rq(cpu);
+		spin_lock_irqsave(&rq->lock, flags);
+		if (rq->rd) {
+			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			cpu_set(cpu, rq->rd->online);
+		}
+		spin_unlock_irqrestore(&rq->lock, flags);
 		break;
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -5600,6 +5632,17 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		}
 		spin_unlock_irq(&rq->lock);
 		break;
+
+	case CPU_DOWN_PREPARE:
+		/* Update our root-domain */
+		rq = cpu_rq(cpu);
+		spin_lock_irqsave(&rq->lock, flags);
+		if (rq->rd) {
+			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			cpu_clear(cpu, rq->rd->online);
+		}
+		spin_unlock_irqrestore(&rq->lock, flags);
+		break;
 #endif
 	}
 	return NOTIFY_OK;
@@ -5788,11 +5831,69 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	return 1;
 }
 
+static void rq_attach_root(struct rq *rq, struct root_domain *rd)
+{
+	unsigned long flags;
+	const struct sched_class *class;
+
+	spin_lock_irqsave(&rq->lock, flags);
+
+	if (rq->rd) {
+		struct root_domain *old_rd = rq->rd;
+
+		for (class = sched_class_highest; class; class = class->next)
+			if (class->leave_domain)
+				class->leave_domain(rq);
+
+		if (atomic_dec_and_test(&old_rd->refcount))
+			kfree(old_rd);
+	}
+
+	atomic_inc(&rd->refcount);
+	rq->rd = rd;
+
+	for (class = sched_class_highest; class; class = class->next)
+		if (class->join_domain)
+			class->join_domain(rq);
+
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static void init_rootdomain(struct root_domain *rd, const cpumask_t *map)
+{
+	memset(rd, 0, sizeof(*rd));
+
+	rd->span = *map;
+	cpus_and(rd->online, rd->span, cpu_online_map);
+}
+
+static void init_defrootdomain(void)
+{
+	cpumask_t cpus = CPU_MASK_ALL;
+
+	init_rootdomain(&def_root_domain, &cpus);
+	atomic_set(&def_root_domain.refcount, 1);
+}
+
+static struct root_domain *alloc_rootdomain(const cpumask_t *map)
+{
+	struct root_domain *rd;
+
+	rd = kmalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return NULL;
+
+	init_rootdomain(rd, map);
+
+	return rd;
+}
+
 /*
  * Attach the domain 'sd' to 'cpu' as its base domain.  Callers must
  * hold the hotplug lock.
  */
-static void cpu_attach_domain(struct sched_domain *sd, int cpu)
+static void cpu_attach_domain(struct sched_domain *sd,
+			      struct root_domain *rd, int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
@@ -5817,6 +5918,7 @@ static void cpu_attach_domain(struct sched_domain *sd, int cpu)
 
 	sched_domain_debug(sd, cpu);
 
+	rq_attach_root(rq, rd);
 	rcu_assign_pointer(rq->sd, sd);
 }
 
@@ -6185,6 +6287,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 static int build_sched_domains(const cpumask_t *cpu_map)
 {
 	int i;
+	struct root_domain *rd;
 #ifdef CONFIG_NUMA
 	struct sched_group **sched_group_nodes = NULL;
 	int sd_allnodes = 0;
@@ -6201,6 +6304,12 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 	sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes;
 #endif
 
+	rd = alloc_rootdomain(cpu_map);
+	if (!rd) {
+		printk(KERN_WARNING "Cannot alloc root domain\n");
+		return -ENOMEM;
+	}
+
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
 	 */
@@ -6417,7 +6526,7 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 #else
 		sd = &per_cpu(phys_domains, i);
 #endif
-		cpu_attach_domain(sd, i);
+		cpu_attach_domain(sd, rd, i);
 	}
 
 	return 0;
@@ -6475,7 +6584,7 @@ static void detach_destroy_domains(const cpumask_t *cpu_map)
 	unregister_sched_domain_sysctl();
 
 	for_each_cpu_mask(i, *cpu_map)
-		cpu_attach_domain(NULL, i);
+		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
 	arch_destroy_sched_domains(cpu_map);
 }
@@ -6727,6 +6836,10 @@ void __init sched_init(void)
 	int highest_cpu = 0;
 	int i, j;
 
+#ifdef CONFIG_SMP
+	init_defrootdomain();
+#endif
+
 	for_each_possible_cpu(i) {
 		struct rt_prio_array *array;
 		struct rq *rq;
@@ -6765,6 +6878,8 @@ void __init sched_init(void)
 			rq->cpu_load[j] = 0;
 #ifdef CONFIG_SMP
 		rq->sd = NULL;
+		rq->rd = NULL;
+		rq_attach_root(rq, &def_root_domain);
 		rq->active_balance = 0;
 		rq->next_balance = jiffies;
 		rq->push_cpu = 0;
-- 
cgit v1.2.3


From 52d853431e8d9dc17ba94792123a3fe2bc039831 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:20 +0100
Subject: sched: reactivate fork balancing

reactivate fork balancing.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/topology.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/linux/topology.h b/include/linux/topology.h
index 47729f18bfdf..b6c073d8b709 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -103,6 +103,7 @@
 	.forkexec_idx		= 0,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_IDLE		\
@@ -134,6 +135,7 @@
 	.forkexec_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_IDLE		\
@@ -165,6 +167,7 @@
 	.forkexec_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_NEWIDLE	\
+				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
 				| SD_WAKE_AFFINE	\
 				| BALANCE_FOR_PKG_POWER,\
-- 
cgit v1.2.3


From 32525d022ad52a5c14e80e130260431e16e294b6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:20 +0100
Subject: sched: whitespace cleanups in topology.h

whitespace cleanups in topology.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/topology.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/topology.h b/include/linux/topology.h
index b6c073d8b709..2352f46160d3 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -5,7 +5,7 @@
  *
  * Copyright (C) 2002, IBM Corp.
  *
- * All rights reserved.          
+ * All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
-- 
cgit v1.2.3


From 9a897c5a6701bcb6f099f7ca20194999102729fd Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 25 Jan 2008 21:08:22 +0100
Subject: sched: RT-balance, replace hooks with pre/post schedule and wakeup
 methods

To make the main sched.c code more agnostic to the schedule classes.
Instead of having specific hooks in the schedule code for the RT class
balancing. They are replaced with a pre_schedule, post_schedule
and task_wake_up methods. These methods may be used by any of the classes
but currently, only the sched_rt class implements them.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  3 +++
 kernel/sched.c        | 20 ++++++++++++++++----
 kernel/sched_rt.c     | 17 +++++++----------
 3 files changed, 26 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2e69f19369e4..c67d2c2f0111 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -843,6 +843,9 @@ struct sched_class {
 	int (*move_one_task) (struct rq *this_rq, int this_cpu,
 			      struct rq *busiest, struct sched_domain *sd,
 			      enum cpu_idle_type idle);
+	void (*pre_schedule) (struct rq *this_rq, struct task_struct *task);
+	void (*post_schedule) (struct rq *this_rq);
+	void (*task_wake_up) (struct rq *this_rq, struct task_struct *task);
 #endif
 
 	void (*set_curr_task) (struct rq *rq);
diff --git a/kernel/sched.c b/kernel/sched.c
index 9d6fb731559b..2368a0d882e3 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1625,7 +1625,10 @@ out_activate:
 
 out_running:
 	p->state = TASK_RUNNING;
-	wakeup_balance_rt(rq, p);
+#ifdef CONFIG_SMP
+	if (p->sched_class->task_wake_up)
+		p->sched_class->task_wake_up(rq, p);
+#endif
 out:
 	task_rq_unlock(rq, &flags);
 
@@ -1748,7 +1751,10 @@ void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		inc_nr_running(p, rq);
 	}
 	check_preempt_curr(rq, p);
-	wakeup_balance_rt(rq, p);
+#ifdef CONFIG_SMP
+	if (p->sched_class->task_wake_up)
+		p->sched_class->task_wake_up(rq, p);
+#endif
 	task_rq_unlock(rq, &flags);
 }
 
@@ -1869,7 +1875,10 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	prev_state = prev->state;
 	finish_arch_switch(prev);
 	finish_lock_switch(rq, prev);
-	schedule_tail_balance_rt(rq);
+#ifdef CONFIG_SMP
+	if (current->sched_class->post_schedule)
+		current->sched_class->post_schedule(rq);
+#endif
 
 	fire_sched_in_preempt_notifiers(current);
 	if (mm)
@@ -3638,7 +3647,10 @@ need_resched_nonpreemptible:
 		switch_count = &prev->nvcsw;
 	}
 
-	schedule_balance_rt(rq, prev);
+#ifdef CONFIG_SMP
+	if (prev->sched_class->pre_schedule)
+		prev->sched_class->pre_schedule(rq, prev);
+#endif
 
 	if (unlikely(!rq->nr_running))
 		idle_balance(cpu, rq);
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 3ea0cae513d2..a5a45104603a 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -689,14 +689,14 @@ static int pull_rt_task(struct rq *this_rq)
 	return ret;
 }
 
-static void schedule_balance_rt(struct rq *rq, struct task_struct *prev)
+static void pre_schedule_rt(struct rq *rq, struct task_struct *prev)
 {
 	/* Try to pull RT tasks here if we lower this rq's prio */
 	if (unlikely(rt_task(prev)) && rq->rt.highest_prio > prev->prio)
 		pull_rt_task(rq);
 }
 
-static void schedule_tail_balance_rt(struct rq *rq)
+static void post_schedule_rt(struct rq *rq)
 {
 	/*
 	 * If we have more than one rt_task queued, then
@@ -713,10 +713,9 @@ static void schedule_tail_balance_rt(struct rq *rq)
 }
 
 
-static void wakeup_balance_rt(struct rq *rq, struct task_struct *p)
+static void task_wake_up_rt(struct rq *rq, struct task_struct *p)
 {
-	if (unlikely(rt_task(p)) &&
-	    !task_running(rq, p) &&
+	if (!task_running(rq, p) &&
 	    (p->prio >= rq->rt.highest_prio) &&
 	    rq->rt.overloaded)
 		push_rt_tasks(rq);
@@ -780,11 +779,6 @@ static void leave_domain_rt(struct rq *rq)
 	if (rq->rt.overloaded)
 		rt_clear_overload(rq);
 }
-
-#else /* CONFIG_SMP */
-# define schedule_tail_balance_rt(rq)	do { } while (0)
-# define schedule_balance_rt(rq, prev)	do { } while (0)
-# define wakeup_balance_rt(rq, p)	do { } while (0)
 #endif /* CONFIG_SMP */
 
 static void task_tick_rt(struct rq *rq, struct task_struct *p)
@@ -840,6 +834,9 @@ const struct sched_class rt_sched_class = {
 	.set_cpus_allowed       = set_cpus_allowed_rt,
 	.join_domain            = join_domain_rt,
 	.leave_domain           = leave_domain_rt,
+	.pre_schedule		= pre_schedule_rt,
+	.post_schedule		= post_schedule_rt,
+	.task_wake_up		= task_wake_up_rt,
 #endif
 
 	.set_curr_task          = set_curr_task_rt,
-- 
cgit v1.2.3


From cb46984504048db946cd551c261df4e70d59a8ea Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 25 Jan 2008 21:08:22 +0100
Subject: sched: RT-balance, add new methods to sched_class

Dmitry Adamushko found that the current implementation of the RT
balancing code left out changes to the sched_setscheduler and
rt_mutex_setprio.

This patch addresses this issue by adding methods to the schedule classes
to handle being switched out of (switched_from) and being switched into
(switched_to) a sched_class. Also a method for changing of priorities
is also added (prio_changed).

This patch also removes some duplicate logic between rt_mutex_setprio and
sched_setscheduler.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h   |  7 ++++
 kernel/sched.c          | 42 +++++++++++------------
 kernel/sched_fair.c     | 39 ++++++++++++++++++++++
 kernel/sched_idletask.c | 31 +++++++++++++++++
 kernel/sched_rt.c       | 89 +++++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 186 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c67d2c2f0111..f2044e707004 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -855,6 +855,13 @@ struct sched_class {
 
 	void (*join_domain)(struct rq *rq);
 	void (*leave_domain)(struct rq *rq);
+
+	void (*switched_from) (struct rq *this_rq, struct task_struct *task,
+			       int running);
+	void (*switched_to) (struct rq *this_rq, struct task_struct *task,
+			     int running);
+	void (*prio_changed) (struct rq *this_rq, struct task_struct *task,
+			     int oldprio, int running);
 };
 
 struct load_weight {
diff --git a/kernel/sched.c b/kernel/sched.c
index 2368a0d882e3..5834c7fb79a5 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1152,6 +1152,18 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 #endif
 }
 
+static inline void check_class_changed(struct rq *rq, struct task_struct *p,
+				       const struct sched_class *prev_class,
+				       int oldprio, int running)
+{
+	if (prev_class != p->sched_class) {
+		if (prev_class->switched_from)
+			prev_class->switched_from(rq, p, running);
+		p->sched_class->switched_to(rq, p, running);
+	} else
+		p->sched_class->prio_changed(rq, p, oldprio, running);
+}
+
 #ifdef CONFIG_SMP
 
 /*
@@ -4017,6 +4029,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	unsigned long flags;
 	int oldprio, on_rq, running;
 	struct rq *rq;
+	const struct sched_class *prev_class = p->sched_class;
 
 	BUG_ON(prio < 0 || prio > MAX_PRIO);
 
@@ -4042,18 +4055,10 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	if (on_rq) {
 		if (running)
 			p->sched_class->set_curr_task(rq);
+
 		enqueue_task(rq, p, 0);
-		/*
-		 * Reschedule if we are currently running on this runqueue and
-		 * our priority decreased, or if we are not currently running on
-		 * this runqueue and our priority is higher than the current's
-		 */
-		if (running) {
-			if (p->prio > oldprio)
-				resched_task(rq->curr);
-		} else {
-			check_preempt_curr(rq, p);
-		}
+
+		check_class_changed(rq, p, prev_class, oldprio, running);
 	}
 	task_rq_unlock(rq, &flags);
 }
@@ -4253,6 +4258,7 @@ int sched_setscheduler(struct task_struct *p, int policy,
 {
 	int retval, oldprio, oldpolicy = -1, on_rq, running;
 	unsigned long flags;
+	const struct sched_class *prev_class = p->sched_class;
 	struct rq *rq;
 
 	/* may grab non-irq protected spin_locks */
@@ -4346,18 +4352,10 @@ recheck:
 	if (on_rq) {
 		if (running)
 			p->sched_class->set_curr_task(rq);
+
 		activate_task(rq, p, 0);
-		/*
-		 * Reschedule if we are currently running on this runqueue and
-		 * our priority decreased, or if we are not currently running on
-		 * this runqueue and our priority is higher than the current's
-		 */
-		if (running) {
-			if (p->prio > oldprio)
-				resched_task(rq->curr);
-		} else {
-			check_preempt_curr(rq, p);
-		}
+
+		check_class_changed(rq, p, prev_class, oldprio, running);
 	}
 	__task_rq_unlock(rq);
 	spin_unlock_irqrestore(&p->pi_lock, flags);
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 10aa6e1ae3dd..dfa18d55561d 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1280,6 +1280,42 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
 	resched_task(rq->curr);
 }
 
+/*
+ * Priority of the task has changed. Check to see if we preempt
+ * the current task.
+ */
+static void prio_changed_fair(struct rq *rq, struct task_struct *p,
+			      int oldprio, int running)
+{
+	/*
+	 * Reschedule if we are currently running on this runqueue and
+	 * our priority decreased, or if we are not currently running on
+	 * this runqueue and our priority is higher than the current's
+	 */
+	if (running) {
+		if (p->prio > oldprio)
+			resched_task(rq->curr);
+	} else
+		check_preempt_curr(rq, p);
+}
+
+/*
+ * We switched to the sched_fair class.
+ */
+static void switched_to_fair(struct rq *rq, struct task_struct *p,
+			     int running)
+{
+	/*
+	 * We were most likely switched from sched_rt, so
+	 * kick off the schedule if running, otherwise just see
+	 * if we can still preempt the current task.
+	 */
+	if (running)
+		resched_task(rq->curr);
+	else
+		check_preempt_curr(rq, p);
+}
+
 /* Account for a task changing its policy or group.
  *
  * This routine is mostly called to set cfs_rq->curr field when a task
@@ -1318,6 +1354,9 @@ static const struct sched_class fair_sched_class = {
 	.set_curr_task          = set_curr_task_fair,
 	.task_tick		= task_tick_fair,
 	.task_new		= task_new_fair,
+
+	.prio_changed		= prio_changed_fair,
+	.switched_to		= switched_to_fair,
 };
 
 #ifdef CONFIG_SCHED_DEBUG
diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c
index ca5374860aef..ef7a2661fa10 100644
--- a/kernel/sched_idletask.c
+++ b/kernel/sched_idletask.c
@@ -69,6 +69,33 @@ static void set_curr_task_idle(struct rq *rq)
 {
 }
 
+static void switched_to_idle(struct rq *rq, struct task_struct *p,
+			     int running)
+{
+	/* Can this actually happen?? */
+	if (running)
+		resched_task(rq->curr);
+	else
+		check_preempt_curr(rq, p);
+}
+
+static void prio_changed_idle(struct rq *rq, struct task_struct *p,
+			      int oldprio, int running)
+{
+	/* This can happen for hot plug CPUS */
+
+	/*
+	 * Reschedule if we are currently running on this runqueue and
+	 * our priority decreased, or if we are not currently running on
+	 * this runqueue and our priority is higher than the current's
+	 */
+	if (running) {
+		if (p->prio > oldprio)
+			resched_task(rq->curr);
+	} else
+		check_preempt_curr(rq, p);
+}
+
 /*
  * Simple, special scheduling class for the per-CPU idle tasks:
  */
@@ -94,5 +121,9 @@ const struct sched_class idle_sched_class = {
 
 	.set_curr_task          = set_curr_task_idle,
 	.task_tick		= task_tick_idle,
+
+	.prio_changed		= prio_changed_idle,
+	.switched_to		= switched_to_idle,
+
 	/* no .task_new for idle tasks */
 };
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index a5a45104603a..57fa3d96847b 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -779,7 +779,92 @@ static void leave_domain_rt(struct rq *rq)
 	if (rq->rt.overloaded)
 		rt_clear_overload(rq);
 }
+
+/*
+ * When switch from the rt queue, we bring ourselves to a position
+ * that we might want to pull RT tasks from other runqueues.
+ */
+static void switched_from_rt(struct rq *rq, struct task_struct *p,
+			   int running)
+{
+	/*
+	 * If there are other RT tasks then we will reschedule
+	 * and the scheduling of the other RT tasks will handle
+	 * the balancing. But if we are the last RT task
+	 * we may need to handle the pulling of RT tasks
+	 * now.
+	 */
+	if (!rq->rt.rt_nr_running)
+		pull_rt_task(rq);
+}
+#endif /* CONFIG_SMP */
+
+/*
+ * When switching a task to RT, we may overload the runqueue
+ * with RT tasks. In this case we try to push them off to
+ * other runqueues.
+ */
+static void switched_to_rt(struct rq *rq, struct task_struct *p,
+			   int running)
+{
+	int check_resched = 1;
+
+	/*
+	 * If we are already running, then there's nothing
+	 * that needs to be done. But if we are not running
+	 * we may need to preempt the current running task.
+	 * If that current running task is also an RT task
+	 * then see if we can move to another run queue.
+	 */
+	if (!running) {
+#ifdef CONFIG_SMP
+		if (rq->rt.overloaded && push_rt_task(rq) &&
+		    /* Don't resched if we changed runqueues */
+		    rq != task_rq(p))
+			check_resched = 0;
+#endif /* CONFIG_SMP */
+		if (check_resched && p->prio < rq->curr->prio)
+			resched_task(rq->curr);
+	}
+}
+
+/*
+ * Priority of the task has changed. This may cause
+ * us to initiate a push or pull.
+ */
+static void prio_changed_rt(struct rq *rq, struct task_struct *p,
+			    int oldprio, int running)
+{
+	if (running) {
+#ifdef CONFIG_SMP
+		/*
+		 * If our priority decreases while running, we
+		 * may need to pull tasks to this runqueue.
+		 */
+		if (oldprio < p->prio)
+			pull_rt_task(rq);
+		/*
+		 * If there's a higher priority task waiting to run
+		 * then reschedule.
+		 */
+		if (p->prio > rq->rt.highest_prio)
+			resched_task(p);
+#else
+		/* For UP simply resched on drop of prio */
+		if (oldprio < p->prio)
+			resched_task(p);
 #endif /* CONFIG_SMP */
+	} else {
+		/*
+		 * This task is not running, but if it is
+		 * greater than the current running task
+		 * then reschedule.
+		 */
+		if (p->prio < rq->curr->prio)
+			resched_task(rq->curr);
+	}
+}
+
 
 static void task_tick_rt(struct rq *rq, struct task_struct *p)
 {
@@ -837,8 +922,12 @@ const struct sched_class rt_sched_class = {
 	.pre_schedule		= pre_schedule_rt,
 	.post_schedule		= post_schedule_rt,
 	.task_wake_up		= task_wake_up_rt,
+	.switched_from		= switched_from_rt,
 #endif
 
 	.set_curr_task          = set_curr_task_rt,
 	.task_tick		= task_tick_rt,
+
+	.prio_changed		= prio_changed_rt,
+	.switched_to		= switched_to_rt,
 };
-- 
cgit v1.2.3


From c2d727aa2ff17a1c8e5ed1e5e231bb8579b27e82 Mon Sep 17 00:00:00 2001
From: Dipankar Sarma <dipankar@in.ibm.com>
Date: Fri, 25 Jan 2008 21:08:23 +0100
Subject: Preempt-RCU: Use softirq instead of tasklets for

This patch makes RCU use softirq instead of tasklets.

It also adds a memory barrier after raising the softirq
inorder to ensure that the cpu sees the most recently updated
value of rcu->cur while processing callbacks.
The discussion of the related theoretical race pointed out
by James Huang can be found here --> http://lkml.org/lkml/2007/11/20/603

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Reviewed-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/interrupt.h |  1 +
 kernel/rcupdate.c         | 25 +++++++++++++++++--------
 2 files changed, 18 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 2306920fa388..c3db4a00f1fa 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -256,6 +256,7 @@ enum
 #ifdef CONFIG_HIGH_RES_TIMERS
 	HRTIMER_SOFTIRQ,
 #endif
+	RCU_SOFTIRQ, 	/* Preferable RCU should always be the last softirq */
 };
 
 /* softirq mask and active fields moved to irq_cpustat_t in
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index f2c1a04e9b18..4dfa0b792efa 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -73,8 +73,6 @@ static struct rcu_ctrlblk rcu_bh_ctrlblk = {
 DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
 DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
 
-/* Fake initialization required by compiler */
-static DEFINE_PER_CPU(struct tasklet_struct, rcu_tasklet) = {NULL};
 static int blimit = 10;
 static int qhimark = 10000;
 static int qlowmark = 100;
@@ -231,6 +229,18 @@ void rcu_barrier(void)
 }
 EXPORT_SYMBOL_GPL(rcu_barrier);
 
+/* Raises the softirq for processing rcu_callbacks. */
+static inline void raise_rcu_softirq(void)
+{
+	raise_softirq(RCU_SOFTIRQ);
+	/*
+	 * The smp_mb() here is required to ensure that this cpu's
+	 * __rcu_process_callbacks() reads the most recently updated
+	 * value of rcu->cur.
+	 */
+	smp_mb();
+}
+
 /*
  * Invoke the completed RCU callbacks. They are expected to be in
  * a per-cpu list.
@@ -260,7 +270,7 @@ static void rcu_do_batch(struct rcu_data *rdp)
 	if (!rdp->donelist)
 		rdp->donetail = &rdp->donelist;
 	else
-		tasklet_schedule(&per_cpu(rcu_tasklet, rdp->cpu));
+		raise_rcu_softirq();
 }
 
 /*
@@ -412,7 +422,6 @@ static void rcu_offline_cpu(int cpu)
 					&per_cpu(rcu_bh_data, cpu));
 	put_cpu_var(rcu_data);
 	put_cpu_var(rcu_bh_data);
-	tasklet_kill_immediate(&per_cpu(rcu_tasklet, cpu), cpu);
 }
 
 #else
@@ -424,7 +433,7 @@ static void rcu_offline_cpu(int cpu)
 #endif
 
 /*
- * This does the RCU processing work from tasklet context. 
+ * This does the RCU processing work from softirq context.
  */
 static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
 					struct rcu_data *rdp)
@@ -469,7 +478,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
 		rcu_do_batch(rdp);
 }
 
-static void rcu_process_callbacks(unsigned long unused)
+static void rcu_process_callbacks(struct softirq_action *unused)
 {
 	__rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data));
 	__rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data));
@@ -533,7 +542,7 @@ void rcu_check_callbacks(int cpu, int user)
 		rcu_bh_qsctr_inc(cpu);
 	} else if (!in_softirq())
 		rcu_bh_qsctr_inc(cpu);
-	tasklet_schedule(&per_cpu(rcu_tasklet, cpu));
+	raise_rcu_softirq();
 }
 
 static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp,
@@ -556,7 +565,7 @@ static void __cpuinit rcu_online_cpu(int cpu)
 
 	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
 	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
-	tasklet_init(&per_cpu(rcu_tasklet, cpu), rcu_process_callbacks, 0UL);
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
 }
 
 static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
-- 
cgit v1.2.3


From 01c1c660f4b8086cad7a62345fd04290f3d82c8f Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Fri, 25 Jan 2008 21:08:24 +0100
Subject: Preempt-RCU: reorganize RCU code into rcuclassic.c and rcupdate.c

This patch re-organizes the RCU code to enable multiple implementations
of RCU. Users of RCU continues to include rcupdate.h and the
RCU interfaces remain the same. This is in preparation for
subsequently merging the preemptible RCU implementation.

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/rcuclassic.h | 161 +++++++++++++
 include/linux/rcupdate.h   | 168 ++++---------
 kernel/Makefile            |   2 +-
 kernel/rcuclassic.c        | 576 +++++++++++++++++++++++++++++++++++++++++++++
 kernel/rcupdate.c          | 575 ++------------------------------------------
 5 files changed, 812 insertions(+), 670 deletions(-)
 create mode 100644 include/linux/rcuclassic.h
 create mode 100644 kernel/rcuclassic.c

(limited to 'include')

diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
new file mode 100644
index 000000000000..2b8b045a51d5
--- /dev/null
+++ b/include/linux/rcuclassic.h
@@ -0,0 +1,161 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion (classic version)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2001
+ *
+ * Author: Dipankar Sarma <dipankar@in.ibm.com>
+ *
+ * Based on the original work by Paul McKenney <paulmck@us.ibm.com>
+ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen.
+ * Papers:
+ * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf
+ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001)
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU
+ *
+ */
+
+#ifndef __LINUX_RCUCLASSIC_H
+#define __LINUX_RCUCLASSIC_H
+
+#ifdef __KERNEL__
+
+#include <linux/cache.h>
+#include <linux/spinlock.h>
+#include <linux/threads.h>
+#include <linux/percpu.h>
+#include <linux/cpumask.h>
+#include <linux/seqlock.h>
+
+
+/* Global control variables for rcupdate callback mechanism. */
+struct rcu_ctrlblk {
+	long	cur;		/* Current batch number.                      */
+	long	completed;	/* Number of the last completed batch         */
+	int	next_pending;	/* Is the next batch already waiting?         */
+
+	int	signaled;
+
+	spinlock_t	lock	____cacheline_internodealigned_in_smp;
+	cpumask_t	cpumask; /* CPUs that need to switch in order    */
+				 /* for current batch to proceed.        */
+} ____cacheline_internodealigned_in_smp;
+
+/* Is batch a before batch b ? */
+static inline int rcu_batch_before(long a, long b)
+{
+	return (a - b) < 0;
+}
+
+/* Is batch a after batch b ? */
+static inline int rcu_batch_after(long a, long b)
+{
+	return (a - b) > 0;
+}
+
+/*
+ * Per-CPU data for Read-Copy UPdate.
+ * nxtlist - new callbacks are added here
+ * curlist - current batch for which quiescent cycle started if any
+ */
+struct rcu_data {
+	/* 1) quiescent state handling : */
+	long		quiescbatch;     /* Batch # for grace period */
+	int		passed_quiesc;	 /* User-mode/idle loop etc. */
+	int		qs_pending;	 /* core waits for quiesc state */
+
+	/* 2) batch handling */
+	long  	       	batch;           /* Batch # for current RCU batch */
+	struct rcu_head *nxtlist;
+	struct rcu_head **nxttail;
+	long            qlen; 	 	 /* # of queued callbacks */
+	struct rcu_head *curlist;
+	struct rcu_head **curtail;
+	struct rcu_head *donelist;
+	struct rcu_head **donetail;
+	long		blimit;		 /* Upper limit on a processed batch */
+	int cpu;
+	struct rcu_head barrier;
+};
+
+DECLARE_PER_CPU(struct rcu_data, rcu_data);
+DECLARE_PER_CPU(struct rcu_data, rcu_bh_data);
+
+/*
+ * Increment the quiescent state counter.
+ * The counter is a bit degenerated: We do not need to know
+ * how many quiescent states passed, just if there was at least
+ * one since the start of the grace period. Thus just a flag.
+ */
+static inline void rcu_qsctr_inc(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+	rdp->passed_quiesc = 1;
+}
+static inline void rcu_bh_qsctr_inc(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu);
+	rdp->passed_quiesc = 1;
+}
+
+extern int rcu_pending(int cpu);
+extern int rcu_needs_cpu(int cpu);
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+extern struct lockdep_map rcu_lock_map;
+# define rcu_read_acquire()	\
+			lock_acquire(&rcu_lock_map, 0, 0, 2, 1, _THIS_IP_)
+# define rcu_read_release()	lock_release(&rcu_lock_map, 1, _THIS_IP_)
+#else
+# define rcu_read_acquire()	do { } while (0)
+# define rcu_read_release()	do { } while (0)
+#endif
+
+#define __rcu_read_lock() \
+	do { \
+		preempt_disable(); \
+		__acquire(RCU); \
+		rcu_read_acquire(); \
+	} while (0)
+#define __rcu_read_unlock() \
+	do { \
+		rcu_read_release(); \
+		__release(RCU); \
+		preempt_enable(); \
+	} while (0)
+#define __rcu_read_lock_bh() \
+	do { \
+		local_bh_disable(); \
+		__acquire(RCU_BH); \
+		rcu_read_acquire(); \
+	} while (0)
+#define __rcu_read_unlock_bh() \
+	do { \
+		rcu_read_release(); \
+		__release(RCU_BH); \
+		local_bh_enable(); \
+	} while (0)
+
+#define __synchronize_sched() synchronize_rcu()
+
+extern void __rcu_init(void);
+extern void rcu_check_callbacks(int cpu, int user);
+extern void rcu_restart_cpu(int cpu);
+
+#endif /* __KERNEL__ */
+#endif /* __LINUX_RCUCLASSIC_H */
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index cc24a01df940..12aa13e13150 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -15,7 +15,7 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  *
- * Copyright (C) IBM Corporation, 2001
+ * Copyright IBM Corporation, 2001
  *
  * Author: Dipankar Sarma <dipankar@in.ibm.com>
  * 
@@ -53,96 +53,14 @@ struct rcu_head {
 	void (*func)(struct rcu_head *head);
 };
 
+#include <linux/rcuclassic.h>
+
 #define RCU_HEAD_INIT 	{ .next = NULL, .func = NULL }
 #define RCU_HEAD(head) struct rcu_head head = RCU_HEAD_INIT
 #define INIT_RCU_HEAD(ptr) do { \
        (ptr)->next = NULL; (ptr)->func = NULL; \
 } while (0)
 
-
-
-/* Global control variables for rcupdate callback mechanism. */
-struct rcu_ctrlblk {
-	long	cur;		/* Current batch number.                      */
-	long	completed;	/* Number of the last completed batch         */
-	int	next_pending;	/* Is the next batch already waiting?         */
-
-	int	signaled;
-
-	spinlock_t	lock	____cacheline_internodealigned_in_smp;
-	cpumask_t	cpumask; /* CPUs that need to switch in order    */
-	                         /* for current batch to proceed.        */
-} ____cacheline_internodealigned_in_smp;
-
-/* Is batch a before batch b ? */
-static inline int rcu_batch_before(long a, long b)
-{
-        return (a - b) < 0;
-}
-
-/* Is batch a after batch b ? */
-static inline int rcu_batch_after(long a, long b)
-{
-        return (a - b) > 0;
-}
-
-/*
- * Per-CPU data for Read-Copy UPdate.
- * nxtlist - new callbacks are added here
- * curlist - current batch for which quiescent cycle started if any
- */
-struct rcu_data {
-	/* 1) quiescent state handling : */
-	long		quiescbatch;     /* Batch # for grace period */
-	int		passed_quiesc;	 /* User-mode/idle loop etc. */
-	int		qs_pending;	 /* core waits for quiesc state */
-
-	/* 2) batch handling */
-	long  	       	batch;           /* Batch # for current RCU batch */
-	struct rcu_head *nxtlist;
-	struct rcu_head **nxttail;
-	long            qlen; 	 	 /* # of queued callbacks */
-	struct rcu_head *curlist;
-	struct rcu_head **curtail;
-	struct rcu_head *donelist;
-	struct rcu_head **donetail;
-	long		blimit;		 /* Upper limit on a processed batch */
-	int cpu;
-	struct rcu_head barrier;
-};
-
-DECLARE_PER_CPU(struct rcu_data, rcu_data);
-DECLARE_PER_CPU(struct rcu_data, rcu_bh_data);
-
-/*
- * Increment the quiescent state counter.
- * The counter is a bit degenerated: We do not need to know
- * how many quiescent states passed, just if there was at least
- * one since the start of the grace period. Thus just a flag.
- */
-static inline void rcu_qsctr_inc(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	rdp->passed_quiesc = 1;
-}
-static inline void rcu_bh_qsctr_inc(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu);
-	rdp->passed_quiesc = 1;
-}
-
-extern int rcu_pending(int cpu);
-extern int rcu_needs_cpu(int cpu);
-
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-extern struct lockdep_map rcu_lock_map;
-# define rcu_read_acquire()	lock_acquire(&rcu_lock_map, 0, 0, 2, 1, _THIS_IP_)
-# define rcu_read_release()	lock_release(&rcu_lock_map, 1, _THIS_IP_)
-#else
-# define rcu_read_acquire()	do { } while (0)
-# define rcu_read_release()	do { } while (0)
-#endif
-
 /**
  * rcu_read_lock - mark the beginning of an RCU read-side critical section.
  *
@@ -172,24 +90,13 @@ extern struct lockdep_map rcu_lock_map;
  *
  * It is illegal to block while in an RCU read-side critical section.
  */
-#define rcu_read_lock() \
-	do { \
-		preempt_disable(); \
-		__acquire(RCU); \
-		rcu_read_acquire(); \
-	} while(0)
+#define rcu_read_lock() __rcu_read_lock()
 
 /**
  * rcu_read_unlock - marks the end of an RCU read-side critical section.
  *
  * See rcu_read_lock() for more information.
  */
-#define rcu_read_unlock() \
-	do { \
-		rcu_read_release(); \
-		__release(RCU); \
-		preempt_enable(); \
-	} while(0)
 
 /*
  * So where is rcu_write_lock()?  It does not exist, as there is no
@@ -200,6 +107,7 @@ extern struct lockdep_map rcu_lock_map;
  * used as well.  RCU does not care how the writers keep out of each
  * others' way, as long as they do so.
  */
+#define rcu_read_unlock() __rcu_read_unlock()
 
 /**
  * rcu_read_lock_bh - mark the beginning of a softirq-only RCU critical section
@@ -212,24 +120,14 @@ extern struct lockdep_map rcu_lock_map;
  * can use just rcu_read_lock().
  *
  */
-#define rcu_read_lock_bh() \
-	do { \
-		local_bh_disable(); \
-		__acquire(RCU_BH); \
-		rcu_read_acquire(); \
-	} while(0)
+#define rcu_read_lock_bh() __rcu_read_lock_bh()
 
 /*
  * rcu_read_unlock_bh - marks the end of a softirq-only RCU critical section
  *
  * See rcu_read_lock_bh() for more information.
  */
-#define rcu_read_unlock_bh() \
-	do { \
-		rcu_read_release(); \
-		__release(RCU_BH); \
-		local_bh_enable(); \
-	} while(0)
+#define rcu_read_unlock_bh() __rcu_read_unlock_bh()
 
 /*
  * Prevent the compiler from merging or refetching accesses.  The compiler
@@ -293,21 +191,53 @@ extern struct lockdep_map rcu_lock_map;
  * In "classic RCU", these two guarantees happen to be one and
  * the same, but can differ in realtime RCU implementations.
  */
-#define synchronize_sched() synchronize_rcu()
+#define synchronize_sched() __synchronize_sched()
+
+/**
+ * call_rcu - Queue an RCU callback for invocation after a grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed.  RCU read-side critical
+ * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
+ * and may be nested.
+ */
+extern void call_rcu(struct rcu_head *head,
+			      void (*func)(struct rcu_head *head));
+
+/**
+ * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed. call_rcu_bh() assumes
+ * that the read-side critical sections end on completion of a softirq
+ * handler. This means that read-side critical sections in process
+ * context must not be interrupted by softirqs. This interface is to be
+ * used when most of the read-side critical sections are in softirq context.
+ * RCU read-side critical sections are delimited by :
+ *  - rcu_read_lock() and  rcu_read_unlock(), if in interrupt context.
+ *  OR
+ *  - rcu_read_lock_bh() and rcu_read_unlock_bh(), if in process context.
+ *  These may be nested.
+ */
+extern void call_rcu_bh(struct rcu_head *head,
+			void (*func)(struct rcu_head *head));
+
+/* Exported common interfaces */
+extern void synchronize_rcu(void);
+extern void rcu_barrier(void);
 
+/* Internal to kernel */
 extern void rcu_init(void);
 extern void rcu_check_callbacks(int cpu, int user);
-extern void rcu_restart_cpu(int cpu);
+
 extern long rcu_batches_completed(void);
 extern long rcu_batches_completed_bh(void);
 
-/* Exported interfaces */
-extern void FASTCALL(call_rcu(struct rcu_head *head, 
-				void (*func)(struct rcu_head *head)));
-extern void FASTCALL(call_rcu_bh(struct rcu_head *head,
-				void (*func)(struct rcu_head *head)));
-extern void synchronize_rcu(void);
-extern void rcu_barrier(void);
-
 #endif /* __KERNEL__ */
 #endif /* __LINUX_RCUPDATE_H */
diff --git a/kernel/Makefile b/kernel/Makefile
index dfa96956dae0..def5dd6097a0 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -6,7 +6,7 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    exit.o itimer.o time.o softirq.o resource.o \
 	    sysctl.o capability.o ptrace.o timer.o user.o user_namespace.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o \
-	    rcupdate.o extable.o params.o posix-timers.o \
+	    rcupdate.o rcuclassic.o extable.o params.o posix-timers.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
 	    hrtimer.o rwsem.o latency.o nsproxy.o srcu.o \
 	    utsname.o notifier.o
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
new file mode 100644
index 000000000000..18369e3386e2
--- /dev/null
+++ b/kernel/rcuclassic.c
@@ -0,0 +1,576 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2001
+ *
+ * Authors: Dipankar Sarma <dipankar@in.ibm.com>
+ *	    Manfred Spraul <manfred@colorfullife.com>
+ *
+ * Based on the original work by Paul McKenney <paulmck@us.ibm.com>
+ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen.
+ * Papers:
+ * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf
+ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001)
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+/* #include <linux/rcupdate.h> @@@ */
+#include <linux/cpu.h>
+#include <linux/mutex.h>
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static struct lock_class_key rcu_lock_key;
+struct lockdep_map rcu_lock_map =
+	STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key);
+EXPORT_SYMBOL_GPL(rcu_lock_map);
+#endif
+
+
+/* Definition for rcupdate control block. */
+static struct rcu_ctrlblk rcu_ctrlblk = {
+	.cur = -300,
+	.completed = -300,
+	.lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock),
+	.cpumask = CPU_MASK_NONE,
+};
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
+	.cur = -300,
+	.completed = -300,
+	.lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock),
+	.cpumask = CPU_MASK_NONE,
+};
+
+DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
+DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
+
+static int blimit = 10;
+static int qhimark = 10000;
+static int qlowmark = 100;
+
+#ifdef CONFIG_SMP
+static void force_quiescent_state(struct rcu_data *rdp,
+			struct rcu_ctrlblk *rcp)
+{
+	int cpu;
+	cpumask_t cpumask;
+	set_need_resched();
+	if (unlikely(!rcp->signaled)) {
+		rcp->signaled = 1;
+		/*
+		 * Don't send IPI to itself. With irqs disabled,
+		 * rdp->cpu is the current cpu.
+		 */
+		cpumask = rcp->cpumask;
+		cpu_clear(rdp->cpu, cpumask);
+		for_each_cpu_mask(cpu, cpumask)
+			smp_send_reschedule(cpu);
+	}
+}
+#else
+static inline void force_quiescent_state(struct rcu_data *rdp,
+			struct rcu_ctrlblk *rcp)
+{
+	set_need_resched();
+}
+#endif
+
+/**
+ * call_rcu - Queue an RCU callback for invocation after a grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed.  RCU read-side critical
+ * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
+ * and may be nested.
+ */
+void call_rcu(struct rcu_head *head,
+				void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = &__get_cpu_var(rcu_data);
+	*rdp->nxttail = head;
+	rdp->nxttail = &head->next;
+	if (unlikely(++rdp->qlen > qhimark)) {
+		rdp->blimit = INT_MAX;
+		force_quiescent_state(rdp, &rcu_ctrlblk);
+	}
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+/**
+ * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed. call_rcu_bh() assumes
+ * that the read-side critical sections end on completion of a softirq
+ * handler. This means that read-side critical sections in process
+ * context must not be interrupted by softirqs. This interface is to be
+ * used when most of the read-side critical sections are in softirq context.
+ * RCU read-side critical sections are delimited by rcu_read_lock() and
+ * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh()
+ * and rcu_read_unlock_bh(), if in process context. These may be nested.
+ */
+void call_rcu_bh(struct rcu_head *head,
+				void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = &__get_cpu_var(rcu_bh_data);
+	*rdp->nxttail = head;
+	rdp->nxttail = &head->next;
+
+	if (unlikely(++rdp->qlen > qhimark)) {
+		rdp->blimit = INT_MAX;
+		force_quiescent_state(rdp, &rcu_bh_ctrlblk);
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu_bh);
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed(void)
+{
+	return rcu_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed);
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed_bh(void)
+{
+	return rcu_bh_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
+
+/* Raises the softirq for processing rcu_callbacks. */
+static inline void raise_rcu_softirq(void)
+{
+	raise_softirq(RCU_SOFTIRQ);
+	/*
+	 * The smp_mb() here is required to ensure that this cpu's
+	 * __rcu_process_callbacks() reads the most recently updated
+	 * value of rcu->cur.
+	 */
+	smp_mb();
+}
+
+/*
+ * Invoke the completed RCU callbacks. They are expected to be in
+ * a per-cpu list.
+ */
+static void rcu_do_batch(struct rcu_data *rdp)
+{
+	struct rcu_head *next, *list;
+	int count = 0;
+
+	list = rdp->donelist;
+	while (list) {
+		next = list->next;
+		prefetch(next);
+		list->func(list);
+		list = next;
+		if (++count >= rdp->blimit)
+			break;
+	}
+	rdp->donelist = list;
+
+	local_irq_disable();
+	rdp->qlen -= count;
+	local_irq_enable();
+	if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark)
+		rdp->blimit = blimit;
+
+	if (!rdp->donelist)
+		rdp->donetail = &rdp->donelist;
+	else
+		raise_rcu_softirq();
+}
+
+/*
+ * Grace period handling:
+ * The grace period handling consists out of two steps:
+ * - A new grace period is started.
+ *   This is done by rcu_start_batch. The start is not broadcasted to
+ *   all cpus, they must pick this up by comparing rcp->cur with
+ *   rdp->quiescbatch. All cpus are recorded  in the
+ *   rcu_ctrlblk.cpumask bitmap.
+ * - All cpus must go through a quiescent state.
+ *   Since the start of the grace period is not broadcasted, at least two
+ *   calls to rcu_check_quiescent_state are required:
+ *   The first call just notices that a new grace period is running. The
+ *   following calls check if there was a quiescent state since the beginning
+ *   of the grace period. If so, it updates rcu_ctrlblk.cpumask. If
+ *   the bitmap is empty, then the grace period is completed.
+ *   rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace
+ *   period (if necessary).
+ */
+/*
+ * Register a new batch of callbacks, and start it up if there is currently no
+ * active batch and the batch to be registered has not already occurred.
+ * Caller must hold rcu_ctrlblk.lock.
+ */
+static void rcu_start_batch(struct rcu_ctrlblk *rcp)
+{
+	if (rcp->next_pending &&
+			rcp->completed == rcp->cur) {
+		rcp->next_pending = 0;
+		/*
+		 * next_pending == 0 must be visible in
+		 * __rcu_process_callbacks() before it can see new value of cur.
+		 */
+		smp_wmb();
+		rcp->cur++;
+
+		/*
+		 * Accessing nohz_cpu_mask before incrementing rcp->cur needs a
+		 * Barrier  Otherwise it can cause tickless idle CPUs to be
+		 * included in rcp->cpumask, which will extend graceperiods
+		 * unnecessarily.
+		 */
+		smp_mb();
+		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
+
+		rcp->signaled = 0;
+	}
+}
+
+/*
+ * cpu went through a quiescent state since the beginning of the grace period.
+ * Clear it from the cpu mask and complete the grace period if it was the last
+ * cpu. Start another grace period if someone has further entries pending
+ */
+static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp)
+{
+	cpu_clear(cpu, rcp->cpumask);
+	if (cpus_empty(rcp->cpumask)) {
+		/* batch completed ! */
+		rcp->completed = rcp->cur;
+		rcu_start_batch(rcp);
+	}
+}
+
+/*
+ * Check if the cpu has gone through a quiescent state (say context
+ * switch). If so and if it already hasn't done so in this RCU
+ * quiescent cycle, then indicate that it has done so.
+ */
+static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp,
+					struct rcu_data *rdp)
+{
+	if (rdp->quiescbatch != rcp->cur) {
+		/* start new grace period: */
+		rdp->qs_pending = 1;
+		rdp->passed_quiesc = 0;
+		rdp->quiescbatch = rcp->cur;
+		return;
+	}
+
+	/* Grace period already completed for this cpu?
+	 * qs_pending is checked instead of the actual bitmap to avoid
+	 * cacheline trashing.
+	 */
+	if (!rdp->qs_pending)
+		return;
+
+	/*
+	 * Was there a quiescent state since the beginning of the grace
+	 * period? If no, then exit and wait for the next call.
+	 */
+	if (!rdp->passed_quiesc)
+		return;
+	rdp->qs_pending = 0;
+
+	spin_lock(&rcp->lock);
+	/*
+	 * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync
+	 * during cpu startup. Ignore the quiescent state.
+	 */
+	if (likely(rdp->quiescbatch == rcp->cur))
+		cpu_quiet(rdp->cpu, rcp);
+
+	spin_unlock(&rcp->lock);
+}
+
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing
+ * locking requirements, the list it's pulling from has to belong to a cpu
+ * which is dead and hence not processing interrupts.
+ */
+static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list,
+				struct rcu_head **tail)
+{
+	local_irq_disable();
+	*this_rdp->nxttail = list;
+	if (list)
+		this_rdp->nxttail = tail;
+	local_irq_enable();
+}
+
+static void __rcu_offline_cpu(struct rcu_data *this_rdp,
+				struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
+{
+	/* if the cpu going offline owns the grace period
+	 * we can block indefinitely waiting for it, so flush
+	 * it here
+	 */
+	spin_lock_bh(&rcp->lock);
+	if (rcp->cur != rcp->completed)
+		cpu_quiet(rdp->cpu, rcp);
+	spin_unlock_bh(&rcp->lock);
+	rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail);
+	rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail);
+	rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail);
+}
+
+static void rcu_offline_cpu(int cpu)
+{
+	struct rcu_data *this_rdp = &get_cpu_var(rcu_data);
+	struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data);
+
+	__rcu_offline_cpu(this_rdp, &rcu_ctrlblk,
+					&per_cpu(rcu_data, cpu));
+	__rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk,
+					&per_cpu(rcu_bh_data, cpu));
+	put_cpu_var(rcu_data);
+	put_cpu_var(rcu_bh_data);
+}
+
+#else
+
+static void rcu_offline_cpu(int cpu)
+{
+}
+
+#endif
+
+/*
+ * This does the RCU processing work from softirq context.
+ */
+static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
+					struct rcu_data *rdp)
+{
+	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) {
+		*rdp->donetail = rdp->curlist;
+		rdp->donetail = rdp->curtail;
+		rdp->curlist = NULL;
+		rdp->curtail = &rdp->curlist;
+	}
+
+	if (rdp->nxtlist && !rdp->curlist) {
+		local_irq_disable();
+		rdp->curlist = rdp->nxtlist;
+		rdp->curtail = rdp->nxttail;
+		rdp->nxtlist = NULL;
+		rdp->nxttail = &rdp->nxtlist;
+		local_irq_enable();
+
+		/*
+		 * start the next batch of callbacks
+		 */
+
+		/* determine batch number */
+		rdp->batch = rcp->cur + 1;
+		/* see the comment and corresponding wmb() in
+		 * the rcu_start_batch()
+		 */
+		smp_rmb();
+
+		if (!rcp->next_pending) {
+			/* and start it/schedule start if it's a new batch */
+			spin_lock(&rcp->lock);
+			rcp->next_pending = 1;
+			rcu_start_batch(rcp);
+			spin_unlock(&rcp->lock);
+		}
+	}
+
+	rcu_check_quiescent_state(rcp, rdp);
+	if (rdp->donelist)
+		rcu_do_batch(rdp);
+}
+
+static void rcu_process_callbacks(struct softirq_action *unused)
+{
+	__rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data));
+	__rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data));
+}
+
+static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
+{
+	/* This cpu has pending rcu entries and the grace period
+	 * for them has completed.
+	 */
+	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch))
+		return 1;
+
+	/* This cpu has no pending entries, but there are new entries */
+	if (!rdp->curlist && rdp->nxtlist)
+		return 1;
+
+	/* This cpu has finished callbacks to invoke */
+	if (rdp->donelist)
+		return 1;
+
+	/* The rcu core waits for a quiescent state from the cpu */
+	if (rdp->quiescbatch != rcp->cur || rdp->qs_pending)
+		return 1;
+
+	/* nothing to do */
+	return 0;
+}
+
+/*
+ * Check to see if there is any immediate RCU-related work to be done
+ * by the current CPU, returning 1 if so.  This function is part of the
+ * RCU implementation; it is -not- an exported member of the RCU API.
+ */
+int rcu_pending(int cpu)
+{
+	return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) ||
+		__rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu));
+}
+
+/*
+ * Check to see if any future RCU-related work will need to be done
+ * by the current CPU, even if none need be done immediately, returning
+ * 1 if so.  This function is part of the RCU implementation; it is -not-
+ * an exported member of the RCU API.
+ */
+int rcu_needs_cpu(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+	struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu);
+
+	return (!!rdp->curlist || !!rdp_bh->curlist || rcu_pending(cpu));
+}
+
+void rcu_check_callbacks(int cpu, int user)
+{
+	if (user ||
+	    (idle_cpu(cpu) && !in_softirq() &&
+				hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
+		rcu_qsctr_inc(cpu);
+		rcu_bh_qsctr_inc(cpu);
+	} else if (!in_softirq())
+		rcu_bh_qsctr_inc(cpu);
+	raise_rcu_softirq();
+}
+
+static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp,
+						struct rcu_data *rdp)
+{
+	memset(rdp, 0, sizeof(*rdp));
+	rdp->curtail = &rdp->curlist;
+	rdp->nxttail = &rdp->nxtlist;
+	rdp->donetail = &rdp->donelist;
+	rdp->quiescbatch = rcp->completed;
+	rdp->qs_pending = 0;
+	rdp->cpu = cpu;
+	rdp->blimit = blimit;
+}
+
+static void __cpuinit rcu_online_cpu(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+	struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu);
+
+	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
+	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+}
+
+static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
+				unsigned long action, void *hcpu)
+{
+	long cpu = (long)hcpu;
+
+	switch (action) {
+	case CPU_UP_PREPARE:
+	case CPU_UP_PREPARE_FROZEN:
+		rcu_online_cpu(cpu);
+		break;
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		rcu_offline_cpu(cpu);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata rcu_nb = {
+	.notifier_call	= rcu_cpu_notify,
+};
+
+/*
+ * Initializes rcu mechanism.  Assumed to be called early.
+ * That is before local timer(SMP) or jiffie timer (uniproc) is setup.
+ * Note that rcu_qsctr and friends are implicitly
+ * initialized due to the choice of ``0'' for RCU_CTR_INVALID.
+ */
+void __init __rcu_init(void)
+{
+	rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
+			(void *)(long)smp_processor_id());
+	/* Register notifier for non-boot CPUs */
+	register_cpu_notifier(&rcu_nb);
+}
+
+module_param(blimit, int, 0);
+module_param(qhimark, int, 0);
+module_param(qlowmark, int, 0);
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index 4dfa0b792efa..0ccd0095ebdc 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -15,7 +15,7 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  *
- * Copyright (C) IBM Corporation, 2001
+ * Copyright IBM Corporation, 2001
  *
  * Authors: Dipankar Sarma <dipankar@in.ibm.com>
  *	    Manfred Spraul <manfred@colorfullife.com>
@@ -35,163 +35,57 @@
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/smp.h>
-#include <linux/rcupdate.h>
 #include <linux/interrupt.h>
 #include <linux/sched.h>
 #include <asm/atomic.h>
 #include <linux/bitops.h>
-#include <linux/module.h>
 #include <linux/completion.h>
-#include <linux/moduleparam.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/mutex.h>
+#include <linux/module.h>
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-static struct lock_class_key rcu_lock_key;
-struct lockdep_map rcu_lock_map =
-	STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key);
-
-EXPORT_SYMBOL_GPL(rcu_lock_map);
-#endif
-
-/* Definition for rcupdate control block. */
-static struct rcu_ctrlblk rcu_ctrlblk = {
-	.cur = -300,
-	.completed = -300,
-	.lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
-};
-static struct rcu_ctrlblk rcu_bh_ctrlblk = {
-	.cur = -300,
-	.completed = -300,
-	.lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
+struct rcu_synchronize {
+	struct rcu_head head;
+	struct completion completion;
 };
 
-DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
-DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
-
-static int blimit = 10;
-static int qhimark = 10000;
-static int qlowmark = 100;
-
+static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL};
 static atomic_t rcu_barrier_cpu_count;
 static DEFINE_MUTEX(rcu_barrier_mutex);
 static struct completion rcu_barrier_completion;
 
-#ifdef CONFIG_SMP
-static void force_quiescent_state(struct rcu_data *rdp,
-			struct rcu_ctrlblk *rcp)
-{
-	int cpu;
-	cpumask_t cpumask;
-	set_need_resched();
-	if (unlikely(!rcp->signaled)) {
-		rcp->signaled = 1;
-		/*
-		 * Don't send IPI to itself. With irqs disabled,
-		 * rdp->cpu is the current cpu.
-		 */
-		cpumask = rcp->cpumask;
-		cpu_clear(rdp->cpu, cpumask);
-		for_each_cpu_mask(cpu, cpumask)
-			smp_send_reschedule(cpu);
-	}
-}
-#else
-static inline void force_quiescent_state(struct rcu_data *rdp,
-			struct rcu_ctrlblk *rcp)
+/* Because of FASTCALL declaration of complete, we use this wrapper */
+static void wakeme_after_rcu(struct rcu_head  *head)
 {
-	set_need_resched();
+	struct rcu_synchronize *rcu;
+
+	rcu = container_of(head, struct rcu_synchronize, head);
+	complete(&rcu->completion);
 }
-#endif
 
 /**
- * call_rcu - Queue an RCU callback for invocation after a grace period.
- * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * synchronize_rcu - wait until a grace period has elapsed.
  *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
+ * Control will return to the caller some time after a full grace
+ * period has elapsed, in other words after all currently executing RCU
  * read-side critical sections have completed.  RCU read-side critical
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
-void fastcall call_rcu(struct rcu_head *head,
-				void (*func)(struct rcu_head *rcu))
-{
-	unsigned long flags;
-	struct rcu_data *rdp;
-
-	head->func = func;
-	head->next = NULL;
-	local_irq_save(flags);
-	rdp = &__get_cpu_var(rcu_data);
-	*rdp->nxttail = head;
-	rdp->nxttail = &head->next;
-	if (unlikely(++rdp->qlen > qhimark)) {
-		rdp->blimit = INT_MAX;
-		force_quiescent_state(rdp, &rcu_ctrlblk);
-	}
-	local_irq_restore(flags);
-}
-
-/**
- * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
- * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
- *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed. call_rcu_bh() assumes
- * that the read-side critical sections end on completion of a softirq
- * handler. This means that read-side critical sections in process
- * context must not be interrupted by softirqs. This interface is to be
- * used when most of the read-side critical sections are in softirq context.
- * RCU read-side critical sections are delimited by rcu_read_lock() and
- * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh()
- * and rcu_read_unlock_bh(), if in process context. These may be nested.
- */
-void fastcall call_rcu_bh(struct rcu_head *head,
-				void (*func)(struct rcu_head *rcu))
+void synchronize_rcu(void)
 {
-	unsigned long flags;
-	struct rcu_data *rdp;
-
-	head->func = func;
-	head->next = NULL;
-	local_irq_save(flags);
-	rdp = &__get_cpu_var(rcu_bh_data);
-	*rdp->nxttail = head;
-	rdp->nxttail = &head->next;
-
-	if (unlikely(++rdp->qlen > qhimark)) {
-		rdp->blimit = INT_MAX;
-		force_quiescent_state(rdp, &rcu_bh_ctrlblk);
-	}
-
-	local_irq_restore(flags);
-}
+	struct rcu_synchronize rcu;
 
-/*
- * Return the number of RCU batches processed thus far.  Useful
- * for debug and statistics.
- */
-long rcu_batches_completed(void)
-{
-	return rcu_ctrlblk.completed;
-}
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished */
+	call_rcu(&rcu.head, wakeme_after_rcu);
 
-/*
- * Return the number of RCU batches processed thus far.  Useful
- * for debug and statistics.
- */
-long rcu_batches_completed_bh(void)
-{
-	return rcu_bh_ctrlblk.completed;
+	/* Wait for it */
+	wait_for_completion(&rcu.completion);
 }
+EXPORT_SYMBOL_GPL(synchronize_rcu);
 
 static void rcu_barrier_callback(struct rcu_head *notused)
 {
@@ -205,10 +99,8 @@ static void rcu_barrier_callback(struct rcu_head *notused)
 static void rcu_barrier_func(void *notused)
 {
 	int cpu = smp_processor_id();
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_head *head;
+	struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu);
 
-	head = &rdp->barrier;
 	atomic_inc(&rcu_barrier_cpu_count);
 	call_rcu(head, rcu_barrier_callback);
 }
@@ -229,425 +121,8 @@ void rcu_barrier(void)
 }
 EXPORT_SYMBOL_GPL(rcu_barrier);
 
-/* Raises the softirq for processing rcu_callbacks. */
-static inline void raise_rcu_softirq(void)
-{
-	raise_softirq(RCU_SOFTIRQ);
-	/*
-	 * The smp_mb() here is required to ensure that this cpu's
-	 * __rcu_process_callbacks() reads the most recently updated
-	 * value of rcu->cur.
-	 */
-	smp_mb();
-}
-
-/*
- * Invoke the completed RCU callbacks. They are expected to be in
- * a per-cpu list.
- */
-static void rcu_do_batch(struct rcu_data *rdp)
-{
-	struct rcu_head *next, *list;
-	int count = 0;
-
-	list = rdp->donelist;
-	while (list) {
-		next = list->next;
-		prefetch(next);
-		list->func(list);
-		list = next;
-		if (++count >= rdp->blimit)
-			break;
-	}
-	rdp->donelist = list;
-
-	local_irq_disable();
-	rdp->qlen -= count;
-	local_irq_enable();
-	if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark)
-		rdp->blimit = blimit;
-
-	if (!rdp->donelist)
-		rdp->donetail = &rdp->donelist;
-	else
-		raise_rcu_softirq();
-}
-
-/*
- * Grace period handling:
- * The grace period handling consists out of two steps:
- * - A new grace period is started.
- *   This is done by rcu_start_batch. The start is not broadcasted to
- *   all cpus, they must pick this up by comparing rcp->cur with
- *   rdp->quiescbatch. All cpus are recorded  in the
- *   rcu_ctrlblk.cpumask bitmap.
- * - All cpus must go through a quiescent state.
- *   Since the start of the grace period is not broadcasted, at least two
- *   calls to rcu_check_quiescent_state are required:
- *   The first call just notices that a new grace period is running. The
- *   following calls check if there was a quiescent state since the beginning
- *   of the grace period. If so, it updates rcu_ctrlblk.cpumask. If
- *   the bitmap is empty, then the grace period is completed.
- *   rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace
- *   period (if necessary).
- */
-/*
- * Register a new batch of callbacks, and start it up if there is currently no
- * active batch and the batch to be registered has not already occurred.
- * Caller must hold rcu_ctrlblk.lock.
- */
-static void rcu_start_batch(struct rcu_ctrlblk *rcp)
-{
-	if (rcp->next_pending &&
-			rcp->completed == rcp->cur) {
-		rcp->next_pending = 0;
-		/*
-		 * next_pending == 0 must be visible in
-		 * __rcu_process_callbacks() before it can see new value of cur.
-		 */
-		smp_wmb();
-		rcp->cur++;
-
-		/*
-		 * Accessing nohz_cpu_mask before incrementing rcp->cur needs a
-		 * Barrier  Otherwise it can cause tickless idle CPUs to be
-		 * included in rcp->cpumask, which will extend graceperiods
-		 * unnecessarily.
-		 */
-		smp_mb();
-		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
-
-		rcp->signaled = 0;
-	}
-}
-
-/*
- * cpu went through a quiescent state since the beginning of the grace period.
- * Clear it from the cpu mask and complete the grace period if it was the last
- * cpu. Start another grace period if someone has further entries pending
- */
-static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp)
-{
-	cpu_clear(cpu, rcp->cpumask);
-	if (cpus_empty(rcp->cpumask)) {
-		/* batch completed ! */
-		rcp->completed = rcp->cur;
-		rcu_start_batch(rcp);
-	}
-}
-
-/*
- * Check if the cpu has gone through a quiescent state (say context
- * switch). If so and if it already hasn't done so in this RCU
- * quiescent cycle, then indicate that it has done so.
- */
-static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp,
-					struct rcu_data *rdp)
-{
-	if (rdp->quiescbatch != rcp->cur) {
-		/* start new grace period: */
-		rdp->qs_pending = 1;
-		rdp->passed_quiesc = 0;
-		rdp->quiescbatch = rcp->cur;
-		return;
-	}
-
-	/* Grace period already completed for this cpu?
-	 * qs_pending is checked instead of the actual bitmap to avoid
-	 * cacheline trashing.
-	 */
-	if (!rdp->qs_pending)
-		return;
-
-	/* 
-	 * Was there a quiescent state since the beginning of the grace
-	 * period? If no, then exit and wait for the next call.
-	 */
-	if (!rdp->passed_quiesc)
-		return;
-	rdp->qs_pending = 0;
-
-	spin_lock(&rcp->lock);
-	/*
-	 * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync
-	 * during cpu startup. Ignore the quiescent state.
-	 */
-	if (likely(rdp->quiescbatch == rcp->cur))
-		cpu_quiet(rdp->cpu, rcp);
-
-	spin_unlock(&rcp->lock);
-}
-
-
-#ifdef CONFIG_HOTPLUG_CPU
-
-/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing
- * locking requirements, the list it's pulling from has to belong to a cpu
- * which is dead and hence not processing interrupts.
- */
-static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list,
-				struct rcu_head **tail)
-{
-	local_irq_disable();
-	*this_rdp->nxttail = list;
-	if (list)
-		this_rdp->nxttail = tail;
-	local_irq_enable();
-}
-
-static void __rcu_offline_cpu(struct rcu_data *this_rdp,
-				struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
-{
-	/* if the cpu going offline owns the grace period
-	 * we can block indefinitely waiting for it, so flush
-	 * it here
-	 */
-	spin_lock_bh(&rcp->lock);
-	if (rcp->cur != rcp->completed)
-		cpu_quiet(rdp->cpu, rcp);
-	spin_unlock_bh(&rcp->lock);
-	rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail);
-	rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail);
-	rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail);
-}
-
-static void rcu_offline_cpu(int cpu)
-{
-	struct rcu_data *this_rdp = &get_cpu_var(rcu_data);
-	struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data);
-
-	__rcu_offline_cpu(this_rdp, &rcu_ctrlblk,
-					&per_cpu(rcu_data, cpu));
-	__rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk,
-					&per_cpu(rcu_bh_data, cpu));
-	put_cpu_var(rcu_data);
-	put_cpu_var(rcu_bh_data);
-}
-
-#else
-
-static void rcu_offline_cpu(int cpu)
-{
-}
-
-#endif
-
-/*
- * This does the RCU processing work from softirq context.
- */
-static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
-					struct rcu_data *rdp)
-{
-	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) {
-		*rdp->donetail = rdp->curlist;
-		rdp->donetail = rdp->curtail;
-		rdp->curlist = NULL;
-		rdp->curtail = &rdp->curlist;
-	}
-
-	if (rdp->nxtlist && !rdp->curlist) {
-		local_irq_disable();
-		rdp->curlist = rdp->nxtlist;
-		rdp->curtail = rdp->nxttail;
-		rdp->nxtlist = NULL;
-		rdp->nxttail = &rdp->nxtlist;
-		local_irq_enable();
-
-		/*
-		 * start the next batch of callbacks
-		 */
-
-		/* determine batch number */
-		rdp->batch = rcp->cur + 1;
-		/* see the comment and corresponding wmb() in
-		 * the rcu_start_batch()
-		 */
-		smp_rmb();
-
-		if (!rcp->next_pending) {
-			/* and start it/schedule start if it's a new batch */
-			spin_lock(&rcp->lock);
-			rcp->next_pending = 1;
-			rcu_start_batch(rcp);
-			spin_unlock(&rcp->lock);
-		}
-	}
-
-	rcu_check_quiescent_state(rcp, rdp);
-	if (rdp->donelist)
-		rcu_do_batch(rdp);
-}
-
-static void rcu_process_callbacks(struct softirq_action *unused)
-{
-	__rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data));
-	__rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data));
-}
-
-static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
-{
-	/* This cpu has pending rcu entries and the grace period
-	 * for them has completed.
-	 */
-	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch))
-		return 1;
-
-	/* This cpu has no pending entries, but there are new entries */
-	if (!rdp->curlist && rdp->nxtlist)
-		return 1;
-
-	/* This cpu has finished callbacks to invoke */
-	if (rdp->donelist)
-		return 1;
-
-	/* The rcu core waits for a quiescent state from the cpu */
-	if (rdp->quiescbatch != rcp->cur || rdp->qs_pending)
-		return 1;
-
-	/* nothing to do */
-	return 0;
-}
-
-/*
- * Check to see if there is any immediate RCU-related work to be done
- * by the current CPU, returning 1 if so.  This function is part of the
- * RCU implementation; it is -not- an exported member of the RCU API.
- */
-int rcu_pending(int cpu)
-{
-	return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) ||
-		__rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu));
-}
-
-/*
- * Check to see if any future RCU-related work will need to be done
- * by the current CPU, even if none need be done immediately, returning
- * 1 if so.  This function is part of the RCU implementation; it is -not-
- * an exported member of the RCU API.
- */
-int rcu_needs_cpu(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu);
-
-	return (!!rdp->curlist || !!rdp_bh->curlist || rcu_pending(cpu));
-}
-
-void rcu_check_callbacks(int cpu, int user)
-{
-	if (user || 
-	    (idle_cpu(cpu) && !in_softirq() && 
-				hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
-		rcu_qsctr_inc(cpu);
-		rcu_bh_qsctr_inc(cpu);
-	} else if (!in_softirq())
-		rcu_bh_qsctr_inc(cpu);
-	raise_rcu_softirq();
-}
-
-static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp,
-						struct rcu_data *rdp)
-{
-	memset(rdp, 0, sizeof(*rdp));
-	rdp->curtail = &rdp->curlist;
-	rdp->nxttail = &rdp->nxtlist;
-	rdp->donetail = &rdp->donelist;
-	rdp->quiescbatch = rcp->completed;
-	rdp->qs_pending = 0;
-	rdp->cpu = cpu;
-	rdp->blimit = blimit;
-}
-
-static void __cpuinit rcu_online_cpu(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu);
-
-	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
-	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
-	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
-}
-
-static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
-				unsigned long action, void *hcpu)
-{
-	long cpu = (long)hcpu;
-	switch (action) {
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		rcu_online_cpu(cpu);
-		break;
-	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
-		rcu_offline_cpu(cpu);
-		break;
-	default:
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-static struct notifier_block __cpuinitdata rcu_nb = {
-	.notifier_call	= rcu_cpu_notify,
-};
-
-/*
- * Initializes rcu mechanism.  Assumed to be called early.
- * That is before local timer(SMP) or jiffie timer (uniproc) is setup.
- * Note that rcu_qsctr and friends are implicitly
- * initialized due to the choice of ``0'' for RCU_CTR_INVALID.
- */
 void __init rcu_init(void)
 {
-	rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
-			(void *)(long)smp_processor_id());
-	/* Register notifier for non-boot CPUs */
-	register_cpu_notifier(&rcu_nb);
-}
-
-struct rcu_synchronize {
-	struct rcu_head head;
-	struct completion completion;
-};
-
-/* Because of FASTCALL declaration of complete, we use this wrapper */
-static void wakeme_after_rcu(struct rcu_head  *head)
-{
-	struct rcu_synchronize *rcu;
-
-	rcu = container_of(head, struct rcu_synchronize, head);
-	complete(&rcu->completion);
-}
-
-/**
- * synchronize_rcu - wait until a grace period has elapsed.
- *
- * Control will return to the caller some time after a full grace
- * period has elapsed, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
- * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
- * and may be nested.
- *
- * If your read-side code is not protected by rcu_read_lock(), do -not-
- * use synchronize_rcu().
- */
-void synchronize_rcu(void)
-{
-	struct rcu_synchronize rcu;
-
-	init_completion(&rcu.completion);
-	/* Will wake me after RCU finished */
-	call_rcu(&rcu.head, wakeme_after_rcu);
-
-	/* Wait for it */
-	wait_for_completion(&rcu.completion);
+	__rcu_init();
 }
 
-module_param(blimit, int, 0);
-module_param(qhimark, int, 0);
-module_param(qlowmark, int, 0);
-EXPORT_SYMBOL_GPL(rcu_batches_completed);
-EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
-EXPORT_SYMBOL_GPL(call_rcu);
-EXPORT_SYMBOL_GPL(call_rcu_bh);
-EXPORT_SYMBOL_GPL(synchronize_rcu);
-- 
cgit v1.2.3


From e260be673a15b6125068270e0216a3bfbfc12f87 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Fri, 25 Jan 2008 21:08:24 +0100
Subject: Preempt-RCU: implementation

This patch implements a new version of RCU which allows its read-side
critical sections to be preempted. It uses a set of counter pairs
to keep track of the read-side critical sections and flips them
when all tasks exit read-side critical section. The details
of this implementation can be found in this paper -

	http://www.rdrop.com/users/paulmck/RCU/OLSrtRCU.2006.08.11a.pdf

and the article-

	http://lwn.net/Articles/253651/

This patch was developed as a part of the -rt kernel development and
meant to provide better latencies when read-side critical sections of
RCU don't disable preemption.  As a consequence of keeping track of RCU
readers, the readers have a slight overhead (optimizations in the paper).
This implementation co-exists with the "classic" RCU implementations
and can be switched to at compiler.

Also includes RCU tracing summarized in debugfs.

[ akpm@linux-foundation.org: build fixes on non-preempt architectures ]

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com>
Reviewed-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 fs/Kconfig                       |   1 -
 include/linux/rcuclassic.h       |   3 +
 include/linux/rcupdate.h         |  11 +-
 include/linux/rcupreempt.h       |  86 +++++
 include/linux/rcupreempt_trace.h |  99 +++++
 include/linux/sched.h            |   5 +
 init/Kconfig                     |  28 ++
 kernel/Kconfig.preempt           |  10 +
 kernel/Makefile                  |   7 +-
 kernel/fork.c                    |   4 +
 kernel/rcuclassic.c              |   1 -
 kernel/rcupreempt.c              | 816 +++++++++++++++++++++++++++++++++++++++
 kernel/rcupreempt_trace.c        | 330 ++++++++++++++++
 13 files changed, 1394 insertions(+), 7 deletions(-)
 create mode 100644 include/linux/rcupreempt.h
 create mode 100644 include/linux/rcupreempt_trace.h
 create mode 100644 kernel/rcupreempt.c
 create mode 100644 kernel/rcupreempt_trace.c

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index 781b47d2f9f2..b4799efaf9e8 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -2130,4 +2130,3 @@ source "fs/nls/Kconfig"
 source "fs/dlm/Kconfig"
 
 endmenu
-
diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
index 2b8b045a51d5..4d6624260b4c 100644
--- a/include/linux/rcuclassic.h
+++ b/include/linux/rcuclassic.h
@@ -157,5 +157,8 @@ extern void __rcu_init(void);
 extern void rcu_check_callbacks(int cpu, int user);
 extern void rcu_restart_cpu(int cpu);
 
+extern long rcu_batches_completed(void);
+extern long rcu_batches_completed_bh(void);
+
 #endif /* __KERNEL__ */
 #endif /* __LINUX_RCUCLASSIC_H */
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 12aa13e13150..d32c14de270e 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -53,7 +53,11 @@ struct rcu_head {
 	void (*func)(struct rcu_head *head);
 };
 
+#ifdef CONFIG_CLASSIC_RCU
 #include <linux/rcuclassic.h>
+#else /* #ifdef CONFIG_CLASSIC_RCU */
+#include <linux/rcupreempt.h>
+#endif /* #else #ifdef CONFIG_CLASSIC_RCU */
 
 #define RCU_HEAD_INIT 	{ .next = NULL, .func = NULL }
 #define RCU_HEAD(head) struct rcu_head head = RCU_HEAD_INIT
@@ -231,13 +235,12 @@ extern void call_rcu_bh(struct rcu_head *head,
 /* Exported common interfaces */
 extern void synchronize_rcu(void);
 extern void rcu_barrier(void);
+extern long rcu_batches_completed(void);
+extern long rcu_batches_completed_bh(void);
 
 /* Internal to kernel */
 extern void rcu_init(void);
-extern void rcu_check_callbacks(int cpu, int user);
-
-extern long rcu_batches_completed(void);
-extern long rcu_batches_completed_bh(void);
+extern int rcu_needs_cpu(int cpu);
 
 #endif /* __KERNEL__ */
 #endif /* __LINUX_RCUPDATE_H */
diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
new file mode 100644
index 000000000000..ece8eb3e4151
--- /dev/null
+++ b/include/linux/rcupreempt.h
@@ -0,0 +1,86 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion (RT implementation)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2006
+ *
+ * Author:  Paul McKenney <paulmck@us.ibm.com>
+ *
+ * Based on the original work by Paul McKenney <paul.mckenney@us.ibm.com>
+ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen.
+ * Papers:
+ * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf
+ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001)
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU
+ *
+ */
+
+#ifndef __LINUX_RCUPREEMPT_H
+#define __LINUX_RCUPREEMPT_H
+
+#ifdef __KERNEL__
+
+#include <linux/cache.h>
+#include <linux/spinlock.h>
+#include <linux/threads.h>
+#include <linux/percpu.h>
+#include <linux/cpumask.h>
+#include <linux/seqlock.h>
+
+#define rcu_qsctr_inc(cpu)
+#define rcu_bh_qsctr_inc(cpu)
+#define call_rcu_bh(head, rcu) call_rcu(head, rcu)
+
+extern void __rcu_read_lock(void);
+extern void __rcu_read_unlock(void);
+extern int rcu_pending(int cpu);
+extern int rcu_needs_cpu(int cpu);
+
+#define __rcu_read_lock_bh()	{ rcu_read_lock(); local_bh_disable(); }
+#define __rcu_read_unlock_bh()	{ local_bh_enable(); rcu_read_unlock(); }
+
+extern void __synchronize_sched(void);
+
+extern void __rcu_init(void);
+extern void rcu_check_callbacks(int cpu, int user);
+extern void rcu_restart_cpu(int cpu);
+extern long rcu_batches_completed(void);
+
+/*
+ * Return the number of RCU batches processed thus far. Useful for debug
+ * and statistic. The _bh variant is identifcal to straight RCU
+ */
+static inline long rcu_batches_completed_bh(void)
+{
+	return rcu_batches_completed();
+}
+
+#ifdef CONFIG_RCU_TRACE
+struct rcupreempt_trace;
+extern long *rcupreempt_flipctr(int cpu);
+extern long rcupreempt_data_completed(void);
+extern int rcupreempt_flip_flag(int cpu);
+extern int rcupreempt_mb_flag(int cpu);
+extern char *rcupreempt_try_flip_state_name(void);
+extern struct rcupreempt_trace *rcupreempt_trace_cpu(int cpu);
+#endif
+
+struct softirq_action;
+
+#endif /* __KERNEL__ */
+#endif /* __LINUX_RCUPREEMPT_H */
diff --git a/include/linux/rcupreempt_trace.h b/include/linux/rcupreempt_trace.h
new file mode 100644
index 000000000000..21cd6b2a5c42
--- /dev/null
+++ b/include/linux/rcupreempt_trace.h
@@ -0,0 +1,99 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion (RT implementation)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2006
+ *
+ * Author:  Paul McKenney <paulmck@us.ibm.com>
+ *
+ * Based on the original work by Paul McKenney <paul.mckenney@us.ibm.com>
+ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen.
+ * Papers:
+ * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf
+ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001)
+ *
+ * For detailed explanation of the Preemptible Read-Copy Update mechanism see -
+ * 		 http://lwn.net/Articles/253651/
+ */
+
+#ifndef __LINUX_RCUPREEMPT_TRACE_H
+#define __LINUX_RCUPREEMPT_TRACE_H
+
+#ifdef __KERNEL__
+#include <linux/types.h>
+#include <linux/kernel.h>
+
+#include <asm/atomic.h>
+
+/*
+ * PREEMPT_RCU data structures.
+ */
+
+struct rcupreempt_trace {
+	long		next_length;
+	long		next_add;
+	long		wait_length;
+	long		wait_add;
+	long		done_length;
+	long		done_add;
+	long		done_remove;
+	atomic_t	done_invoked;
+	long		rcu_check_callbacks;
+	atomic_t	rcu_try_flip_1;
+	atomic_t	rcu_try_flip_e1;
+	long		rcu_try_flip_i1;
+	long		rcu_try_flip_ie1;
+	long		rcu_try_flip_g1;
+	long		rcu_try_flip_a1;
+	long		rcu_try_flip_ae1;
+	long		rcu_try_flip_a2;
+	long		rcu_try_flip_z1;
+	long		rcu_try_flip_ze1;
+	long		rcu_try_flip_z2;
+	long		rcu_try_flip_m1;
+	long		rcu_try_flip_me1;
+	long		rcu_try_flip_m2;
+};
+
+#ifdef CONFIG_RCU_TRACE
+#define RCU_TRACE(fn, arg) 	fn(arg);
+#else
+#define RCU_TRACE(fn, arg)
+#endif
+
+extern void rcupreempt_trace_move2done(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_move2wait(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_e1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_i1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_ie1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_g1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_a1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_ae1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_a2(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_z1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_ze1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_z2(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_m1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_me1(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_try_flip_m2(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_check_callbacks(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_done_remove(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_invoke(struct rcupreempt_trace *trace);
+extern void rcupreempt_trace_next_add(struct rcupreempt_trace *trace);
+
+#endif /* __KERNEL__ */
+#endif /* __LINUX_RCUPREEMPT_TRACE_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f2044e707004..72e1b8ecfbe1 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -974,6 +974,11 @@ struct task_struct {
 	int nr_cpus_allowed;
 	unsigned int time_slice;
 
+#ifdef CONFIG_PREEMPT_RCU
+	int rcu_read_lock_nesting;
+	int rcu_flipctr_idx;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
+
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
 	struct sched_info sched_info;
 #endif
diff --git a/init/Kconfig b/init/Kconfig
index f5becd2a12f6..0eda68f0ad54 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -763,3 +763,31 @@ source "block/Kconfig"
 
 config PREEMPT_NOTIFIERS
 	bool
+
+choice
+	prompt "RCU implementation type:"
+	default CLASSIC_RCU
+
+config CLASSIC_RCU
+	bool "Classic RCU"
+	help
+	  This option selects the classic RCU implementation that is
+	  designed for best read-side performance on non-realtime
+	  systems.
+
+	  Say Y if you are unsure.
+
+config PREEMPT_RCU
+	bool "Preemptible RCU"
+	depends on PREEMPT
+	help
+	  This option reduces the latency of the kernel by making certain
+	  RCU sections preemptible. Normally RCU code is non-preemptible, if
+	  this option is selected then read-only RCU sections become
+	  preemptible. This helps latency, but may expose bugs due to
+	  now-naive assumptions about each RCU read-side critical section
+	  remaining on a given CPU through its execution.
+
+	  Say N if you are unsure.
+
+endchoice
diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
index c64ce9c14207..61fa116efcde 100644
--- a/kernel/Kconfig.preempt
+++ b/kernel/Kconfig.preempt
@@ -63,3 +63,13 @@ config PREEMPT_BKL
 	  Say Y here if you are building a kernel for a desktop system.
 	  Say N if you are unsure.
 
+config RCU_TRACE
+	bool "Enable tracing for RCU - currently stats in debugfs"
+	select DEBUG_FS
+	default y
+	help
+	  This option provides tracing in RCU which presents stats
+	  in debugfs for debugging RCU implementation.
+
+	  Say Y here if you want to enable RCU tracing
+	  Say N if you are unsure.
diff --git a/kernel/Makefile b/kernel/Makefile
index def5dd6097a0..68755cd9a7e4 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -6,7 +6,7 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    exit.o itimer.o time.o softirq.o resource.o \
 	    sysctl.o capability.o ptrace.o timer.o user.o user_namespace.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o \
-	    rcupdate.o rcuclassic.o extable.o params.o posix-timers.o \
+	    rcupdate.o extable.o params.o posix-timers.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
 	    hrtimer.o rwsem.o latency.o nsproxy.o srcu.o \
 	    utsname.o notifier.o
@@ -52,6 +52,11 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
 obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
+obj-$(CONFIG_CLASSIC_RCU) += rcuclassic.o
+obj-$(CONFIG_PREEMPT_RCU) += rcupreempt.o
+ifeq ($(CONFIG_PREEMPT_RCU),y)
+obj-$(CONFIG_RCU_TRACE) += rcupreempt_trace.o
+endif
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
diff --git a/kernel/fork.c b/kernel/fork.c
index 930c51865ab4..9f8ef32cbc7a 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1045,6 +1045,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	copy_flags(clone_flags, p);
 	INIT_LIST_HEAD(&p->children);
 	INIT_LIST_HEAD(&p->sibling);
+#ifdef CONFIG_PREEMPT_RCU
+	p->rcu_read_lock_nesting = 0;
+	p->rcu_flipctr_idx = 0;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
 	p->vfork_done = NULL;
 	spin_lock_init(&p->alloc_lock);
 
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index ce0cf16cab67..f4ffbd0f306f 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -45,7 +45,6 @@
 #include <linux/moduleparam.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
-/* #include <linux/rcupdate.h> @@@ */
 #include <linux/cpu.h>
 #include <linux/mutex.h>
 
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
new file mode 100644
index 000000000000..a5aabb1677f8
--- /dev/null
+++ b/kernel/rcupreempt.c
@@ -0,0 +1,816 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion, realtime implementation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2006
+ *
+ * Authors: Paul E. McKenney <paulmck@us.ibm.com>
+ *		With thanks to Esben Nielsen, Bill Huey, and Ingo Molnar
+ *		for pushing me away from locks and towards counters, and
+ *		to Suparna Bhattacharya for pushing me completely away
+ *		from atomic instructions on the read side.
+ *
+ * Papers:  http://www.rdrop.com/users/paulmck/RCU
+ *
+ * Design Document: http://lwn.net/Articles/253651/
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU/ *.txt
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/rcupdate.h>
+#include <linux/cpu.h>
+#include <linux/random.h>
+#include <linux/delay.h>
+#include <linux/byteorder/swabb.h>
+#include <linux/cpumask.h>
+#include <linux/rcupreempt_trace.h>
+
+/*
+ * Macro that prevents the compiler from reordering accesses, but does
+ * absolutely -nothing- to prevent CPUs from reordering.  This is used
+ * only to mediate communication between mainline code and hardware
+ * interrupt and NMI handlers.
+ */
+#define ACCESS_ONCE(x) (*(volatile typeof(x) *)&(x))
+
+/*
+ * PREEMPT_RCU data structures.
+ */
+
+/*
+ * GP_STAGES specifies the number of times the state machine has
+ * to go through the all the rcu_try_flip_states (see below)
+ * in a single Grace Period.
+ *
+ * GP in GP_STAGES stands for Grace Period ;)
+ */
+#define GP_STAGES    2
+struct rcu_data {
+	spinlock_t	lock;		/* Protect rcu_data fields. */
+	long		completed;	/* Number of last completed batch. */
+	int		waitlistcount;
+	struct tasklet_struct rcu_tasklet;
+	struct rcu_head *nextlist;
+	struct rcu_head **nexttail;
+	struct rcu_head *waitlist[GP_STAGES];
+	struct rcu_head **waittail[GP_STAGES];
+	struct rcu_head *donelist;
+	struct rcu_head **donetail;
+	long rcu_flipctr[2];
+#ifdef CONFIG_RCU_TRACE
+	struct rcupreempt_trace trace;
+#endif /* #ifdef CONFIG_RCU_TRACE */
+};
+
+/*
+ * States for rcu_try_flip() and friends.
+ */
+
+enum rcu_try_flip_states {
+
+	/*
+	 * Stay here if nothing is happening. Flip the counter if somthing
+	 * starts happening. Denoted by "I"
+	 */
+	rcu_try_flip_idle_state,
+
+	/*
+	 * Wait here for all CPUs to notice that the counter has flipped. This
+	 * prevents the old set of counters from ever being incremented once
+	 * we leave this state, which in turn is necessary because we cannot
+	 * test any individual counter for zero -- we can only check the sum.
+	 * Denoted by "A".
+	 */
+	rcu_try_flip_waitack_state,
+
+	/*
+	 * Wait here for the sum of the old per-CPU counters to reach zero.
+	 * Denoted by "Z".
+	 */
+	rcu_try_flip_waitzero_state,
+
+	/*
+	 * Wait here for each of the other CPUs to execute a memory barrier.
+	 * This is necessary to ensure that these other CPUs really have
+	 * completed executing their RCU read-side critical sections, despite
+	 * their CPUs wildly reordering memory. Denoted by "M".
+	 */
+	rcu_try_flip_waitmb_state,
+};
+
+struct rcu_ctrlblk {
+	spinlock_t	fliplock;	/* Protect state-machine transitions. */
+	long		completed;	/* Number of last completed batch. */
+	enum rcu_try_flip_states rcu_try_flip_state; /* The current state of
+							the rcu state machine */
+};
+
+static DEFINE_PER_CPU(struct rcu_data, rcu_data);
+static struct rcu_ctrlblk rcu_ctrlblk = {
+	.fliplock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.fliplock),
+	.completed = 0,
+	.rcu_try_flip_state = rcu_try_flip_idle_state,
+};
+
+
+#ifdef CONFIG_RCU_TRACE
+static char *rcu_try_flip_state_names[] =
+	{ "idle", "waitack", "waitzero", "waitmb" };
+#endif /* #ifdef CONFIG_RCU_TRACE */
+
+/*
+ * Enum and per-CPU flag to determine when each CPU has seen
+ * the most recent counter flip.
+ */
+
+enum rcu_flip_flag_values {
+	rcu_flip_seen,		/* Steady/initial state, last flip seen. */
+				/* Only GP detector can update. */
+	rcu_flipped		/* Flip just completed, need confirmation. */
+				/* Only corresponding CPU can update. */
+};
+static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_flip_flag_values, rcu_flip_flag)
+								= rcu_flip_seen;
+
+/*
+ * Enum and per-CPU flag to determine when each CPU has executed the
+ * needed memory barrier to fence in memory references from its last RCU
+ * read-side critical section in the just-completed grace period.
+ */
+
+enum rcu_mb_flag_values {
+	rcu_mb_done,		/* Steady/initial state, no mb()s required. */
+				/* Only GP detector can update. */
+	rcu_mb_needed		/* Flip just completed, need an mb(). */
+				/* Only corresponding CPU can update. */
+};
+static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_mb_flag_values, rcu_mb_flag)
+								= rcu_mb_done;
+
+/*
+ * RCU_DATA_ME: find the current CPU's rcu_data structure.
+ * RCU_DATA_CPU: find the specified CPU's rcu_data structure.
+ */
+#define RCU_DATA_ME()		(&__get_cpu_var(rcu_data))
+#define RCU_DATA_CPU(cpu)	(&per_cpu(rcu_data, cpu))
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is not
+ * cached in a local variable, but where the CPU number is so cached.
+ */
+#define RCU_TRACE_CPU(f, cpu) RCU_TRACE(f, &(RCU_DATA_CPU(cpu)->trace));
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is not
+ * cached in a local variable.
+ */
+#define RCU_TRACE_ME(f) RCU_TRACE(f, &(RCU_DATA_ME()->trace));
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is pointed
+ * to by a local variable.
+ */
+#define RCU_TRACE_RDP(f, rdp) RCU_TRACE(f, &((rdp)->trace));
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed(void)
+{
+	return rcu_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed);
+
+EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
+
+void __rcu_read_lock(void)
+{
+	int idx;
+	struct task_struct *t = current;
+	int nesting;
+
+	nesting = ACCESS_ONCE(t->rcu_read_lock_nesting);
+	if (nesting != 0) {
+
+		/* An earlier rcu_read_lock() covers us, just count it. */
+
+		t->rcu_read_lock_nesting = nesting + 1;
+
+	} else {
+		unsigned long flags;
+
+		/*
+		 * We disable interrupts for the following reasons:
+		 * - If we get scheduling clock interrupt here, and we
+		 *   end up acking the counter flip, it's like a promise
+		 *   that we will never increment the old counter again.
+		 *   Thus we will break that promise if that
+		 *   scheduling clock interrupt happens between the time
+		 *   we pick the .completed field and the time that we
+		 *   increment our counter.
+		 *
+		 * - We don't want to be preempted out here.
+		 *
+		 * NMIs can still occur, of course, and might themselves
+		 * contain rcu_read_lock().
+		 */
+
+		local_irq_save(flags);
+
+		/*
+		 * Outermost nesting of rcu_read_lock(), so increment
+		 * the current counter for the current CPU.  Use volatile
+		 * casts to prevent the compiler from reordering.
+		 */
+
+		idx = ACCESS_ONCE(rcu_ctrlblk.completed) & 0x1;
+		ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])++;
+
+		/*
+		 * Now that the per-CPU counter has been incremented, we
+		 * are protected from races with rcu_read_lock() invoked
+		 * from NMI handlers on this CPU.  We can therefore safely
+		 * increment the nesting counter, relieving further NMIs
+		 * of the need to increment the per-CPU counter.
+		 */
+
+		ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting + 1;
+
+		/*
+		 * Now that we have preventing any NMIs from storing
+		 * to the ->rcu_flipctr_idx, we can safely use it to
+		 * remember which counter to decrement in the matching
+		 * rcu_read_unlock().
+		 */
+
+		ACCESS_ONCE(t->rcu_flipctr_idx) = idx;
+		local_irq_restore(flags);
+	}
+}
+EXPORT_SYMBOL_GPL(__rcu_read_lock);
+
+void __rcu_read_unlock(void)
+{
+	int idx;
+	struct task_struct *t = current;
+	int nesting;
+
+	nesting = ACCESS_ONCE(t->rcu_read_lock_nesting);
+	if (nesting > 1) {
+
+		/*
+		 * We are still protected by the enclosing rcu_read_lock(),
+		 * so simply decrement the counter.
+		 */
+
+		t->rcu_read_lock_nesting = nesting - 1;
+
+	} else {
+		unsigned long flags;
+
+		/*
+		 * Disable local interrupts to prevent the grace-period
+		 * detection state machine from seeing us half-done.
+		 * NMIs can still occur, of course, and might themselves
+		 * contain rcu_read_lock() and rcu_read_unlock().
+		 */
+
+		local_irq_save(flags);
+
+		/*
+		 * Outermost nesting of rcu_read_unlock(), so we must
+		 * decrement the current counter for the current CPU.
+		 * This must be done carefully, because NMIs can
+		 * occur at any point in this code, and any rcu_read_lock()
+		 * and rcu_read_unlock() pairs in the NMI handlers
+		 * must interact non-destructively with this code.
+		 * Lots of volatile casts, and -very- careful ordering.
+		 *
+		 * Changes to this code, including this one, must be
+		 * inspected, validated, and tested extremely carefully!!!
+		 */
+
+		/*
+		 * First, pick up the index.
+		 */
+
+		idx = ACCESS_ONCE(t->rcu_flipctr_idx);
+
+		/*
+		 * Now that we have fetched the counter index, it is
+		 * safe to decrement the per-task RCU nesting counter.
+		 * After this, any interrupts or NMIs will increment and
+		 * decrement the per-CPU counters.
+		 */
+		ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting - 1;
+
+		/*
+		 * It is now safe to decrement this task's nesting count.
+		 * NMIs that occur after this statement will route their
+		 * rcu_read_lock() calls through this "else" clause, and
+		 * will thus start incrementing the per-CPU counter on
+		 * their own.  They will also clobber ->rcu_flipctr_idx,
+		 * but that is OK, since we have already fetched it.
+		 */
+
+		ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])--;
+		local_irq_restore(flags);
+	}
+}
+EXPORT_SYMBOL_GPL(__rcu_read_unlock);
+
+/*
+ * If a global counter flip has occurred since the last time that we
+ * advanced callbacks, advance them.  Hardware interrupts must be
+ * disabled when calling this function.
+ */
+static void __rcu_advance_callbacks(struct rcu_data *rdp)
+{
+	int cpu;
+	int i;
+	int wlc = 0;
+
+	if (rdp->completed != rcu_ctrlblk.completed) {
+		if (rdp->waitlist[GP_STAGES - 1] != NULL) {
+			*rdp->donetail = rdp->waitlist[GP_STAGES - 1];
+			rdp->donetail = rdp->waittail[GP_STAGES - 1];
+			RCU_TRACE_RDP(rcupreempt_trace_move2done, rdp);
+		}
+		for (i = GP_STAGES - 2; i >= 0; i--) {
+			if (rdp->waitlist[i] != NULL) {
+				rdp->waitlist[i + 1] = rdp->waitlist[i];
+				rdp->waittail[i + 1] = rdp->waittail[i];
+				wlc++;
+			} else {
+				rdp->waitlist[i + 1] = NULL;
+				rdp->waittail[i + 1] =
+					&rdp->waitlist[i + 1];
+			}
+		}
+		if (rdp->nextlist != NULL) {
+			rdp->waitlist[0] = rdp->nextlist;
+			rdp->waittail[0] = rdp->nexttail;
+			wlc++;
+			rdp->nextlist = NULL;
+			rdp->nexttail = &rdp->nextlist;
+			RCU_TRACE_RDP(rcupreempt_trace_move2wait, rdp);
+		} else {
+			rdp->waitlist[0] = NULL;
+			rdp->waittail[0] = &rdp->waitlist[0];
+		}
+		rdp->waitlistcount = wlc;
+		rdp->completed = rcu_ctrlblk.completed;
+	}
+
+	/*
+	 * Check to see if this CPU needs to report that it has seen
+	 * the most recent counter flip, thereby declaring that all
+	 * subsequent rcu_read_lock() invocations will respect this flip.
+	 */
+
+	cpu = raw_smp_processor_id();
+	if (per_cpu(rcu_flip_flag, cpu) == rcu_flipped) {
+		smp_mb();  /* Subsequent counter accesses must see new value */
+		per_cpu(rcu_flip_flag, cpu) = rcu_flip_seen;
+		smp_mb();  /* Subsequent RCU read-side critical sections */
+			   /*  seen -after- acknowledgement. */
+	}
+}
+
+/*
+ * Get here when RCU is idle.  Decide whether we need to
+ * move out of idle state, and return non-zero if so.
+ * "Straightforward" approach for the moment, might later
+ * use callback-list lengths, grace-period duration, or
+ * some such to determine when to exit idle state.
+ * Might also need a pre-idle test that does not acquire
+ * the lock, but let's get the simple case working first...
+ */
+
+static int
+rcu_try_flip_idle(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_i1);
+	if (!rcu_pending(smp_processor_id())) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_ie1);
+		return 0;
+	}
+
+	/*
+	 * Do the flip.
+	 */
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_g1);
+	rcu_ctrlblk.completed++;  /* stands in for rcu_try_flip_g2 */
+
+	/*
+	 * Need a memory barrier so that other CPUs see the new
+	 * counter value before they see the subsequent change of all
+	 * the rcu_flip_flag instances to rcu_flipped.
+	 */
+
+	smp_mb();	/* see above block comment. */
+
+	/* Now ask each CPU for acknowledgement of the flip. */
+
+	for_each_possible_cpu(cpu)
+		per_cpu(rcu_flip_flag, cpu) = rcu_flipped;
+
+	return 1;
+}
+
+/*
+ * Wait for CPUs to acknowledge the flip.
+ */
+
+static int
+rcu_try_flip_waitack(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_a1);
+	for_each_possible_cpu(cpu)
+		if (per_cpu(rcu_flip_flag, cpu) != rcu_flip_seen) {
+			RCU_TRACE_ME(rcupreempt_trace_try_flip_ae1);
+			return 0;
+		}
+
+	/*
+	 * Make sure our checks above don't bleed into subsequent
+	 * waiting for the sum of the counters to reach zero.
+	 */
+
+	smp_mb();	/* see above block comment. */
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_a2);
+	return 1;
+}
+
+/*
+ * Wait for collective ``last'' counter to reach zero,
+ * then tell all CPUs to do an end-of-grace-period memory barrier.
+ */
+
+static int
+rcu_try_flip_waitzero(void)
+{
+	int cpu;
+	int lastidx = !(rcu_ctrlblk.completed & 0x1);
+	int sum = 0;
+
+	/* Check to see if the sum of the "last" counters is zero. */
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_z1);
+	for_each_possible_cpu(cpu)
+		sum += RCU_DATA_CPU(cpu)->rcu_flipctr[lastidx];
+	if (sum != 0) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_ze1);
+		return 0;
+	}
+
+	/*
+	 * This ensures that the other CPUs see the call for
+	 * memory barriers -after- the sum to zero has been
+	 * detected here
+	 */
+	smp_mb();  /*  ^^^^^^^^^^^^ */
+
+	/* Call for a memory barrier from each CPU. */
+	for_each_possible_cpu(cpu)
+		per_cpu(rcu_mb_flag, cpu) = rcu_mb_needed;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_z2);
+	return 1;
+}
+
+/*
+ * Wait for all CPUs to do their end-of-grace-period memory barrier.
+ * Return 0 once all CPUs have done so.
+ */
+
+static int
+rcu_try_flip_waitmb(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_m1);
+	for_each_possible_cpu(cpu)
+		if (per_cpu(rcu_mb_flag, cpu) != rcu_mb_done) {
+			RCU_TRACE_ME(rcupreempt_trace_try_flip_me1);
+			return 0;
+		}
+
+	smp_mb(); /* Ensure that the above checks precede any following flip. */
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_m2);
+	return 1;
+}
+
+/*
+ * Attempt a single flip of the counters.  Remember, a single flip does
+ * -not- constitute a grace period.  Instead, the interval between
+ * at least GP_STAGES consecutive flips is a grace period.
+ *
+ * If anyone is nuts enough to run this CONFIG_PREEMPT_RCU implementation
+ * on a large SMP, they might want to use a hierarchical organization of
+ * the per-CPU-counter pairs.
+ */
+static void rcu_try_flip(void)
+{
+	unsigned long flags;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_1);
+	if (unlikely(!spin_trylock_irqsave(&rcu_ctrlblk.fliplock, flags))) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_e1);
+		return;
+	}
+
+	/*
+	 * Take the next transition(s) through the RCU grace-period
+	 * flip-counter state machine.
+	 */
+
+	switch (rcu_ctrlblk.rcu_try_flip_state) {
+	case rcu_try_flip_idle_state:
+		if (rcu_try_flip_idle())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitack_state;
+		break;
+	case rcu_try_flip_waitack_state:
+		if (rcu_try_flip_waitack())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitzero_state;
+		break;
+	case rcu_try_flip_waitzero_state:
+		if (rcu_try_flip_waitzero())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitmb_state;
+		break;
+	case rcu_try_flip_waitmb_state:
+		if (rcu_try_flip_waitmb())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_idle_state;
+	}
+	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
+}
+
+/*
+ * Check to see if this CPU needs to do a memory barrier in order to
+ * ensure that any prior RCU read-side critical sections have committed
+ * their counter manipulations and critical-section memory references
+ * before declaring the grace period to be completed.
+ */
+static void rcu_check_mb(int cpu)
+{
+	if (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed) {
+		smp_mb();  /* Ensure RCU read-side accesses are visible. */
+		per_cpu(rcu_mb_flag, cpu) = rcu_mb_done;
+	}
+}
+
+void rcu_check_callbacks(int cpu, int user)
+{
+	unsigned long flags;
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	rcu_check_mb(cpu);
+	if (rcu_ctrlblk.completed == rdp->completed)
+		rcu_try_flip();
+	spin_lock_irqsave(&rdp->lock, flags);
+	RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp);
+	__rcu_advance_callbacks(rdp);
+	if (rdp->donelist == NULL) {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+	} else {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+		raise_softirq(RCU_SOFTIRQ);
+	}
+}
+
+/*
+ * Needed by dynticks, to make sure all RCU processing has finished
+ * when we go idle:
+ */
+void rcu_advance_callbacks(int cpu, int user)
+{
+	unsigned long flags;
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	if (rcu_ctrlblk.completed == rdp->completed) {
+		rcu_try_flip();
+		if (rcu_ctrlblk.completed == rdp->completed)
+			return;
+	}
+	spin_lock_irqsave(&rdp->lock, flags);
+	RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp);
+	__rcu_advance_callbacks(rdp);
+	spin_unlock_irqrestore(&rdp->lock, flags);
+}
+
+static void rcu_process_callbacks(struct softirq_action *unused)
+{
+	unsigned long flags;
+	struct rcu_head *next, *list;
+	struct rcu_data *rdp = RCU_DATA_ME();
+
+	spin_lock_irqsave(&rdp->lock, flags);
+	list = rdp->donelist;
+	if (list == NULL) {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+		return;
+	}
+	rdp->donelist = NULL;
+	rdp->donetail = &rdp->donelist;
+	RCU_TRACE_RDP(rcupreempt_trace_done_remove, rdp);
+	spin_unlock_irqrestore(&rdp->lock, flags);
+	while (list) {
+		next = list->next;
+		list->func(list);
+		list = next;
+		RCU_TRACE_ME(rcupreempt_trace_invoke);
+	}
+}
+
+void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = RCU_DATA_ME();
+	spin_lock(&rdp->lock);
+	__rcu_advance_callbacks(rdp);
+	*rdp->nexttail = head;
+	rdp->nexttail = &head->next;
+	RCU_TRACE_RDP(rcupreempt_trace_next_add, rdp);
+	spin_unlock(&rdp->lock);
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+/*
+ * Wait until all currently running preempt_disable() code segments
+ * (including hardware-irq-disable segments) complete.  Note that
+ * in -rt this does -not- necessarily result in all currently executing
+ * interrupt -handlers- having completed.
+ */
+void __synchronize_sched(void)
+{
+	cpumask_t oldmask;
+	int cpu;
+
+	if (sched_getaffinity(0, &oldmask) < 0)
+		oldmask = cpu_possible_map;
+	for_each_online_cpu(cpu) {
+		sched_setaffinity(0, cpumask_of_cpu(cpu));
+		schedule();
+	}
+	sched_setaffinity(0, oldmask);
+}
+EXPORT_SYMBOL_GPL(__synchronize_sched);
+
+/*
+ * Check to see if any future RCU-related work will need to be done
+ * by the current CPU, even if none need be done immediately, returning
+ * 1 if so.  Assumes that notifiers would take care of handling any
+ * outstanding requests from the RCU core.
+ *
+ * This function is part of the RCU implementation; it is -not-
+ * an exported member of the RCU API.
+ */
+int rcu_needs_cpu(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	return (rdp->donelist != NULL ||
+		!!rdp->waitlistcount ||
+		rdp->nextlist != NULL);
+}
+
+int rcu_pending(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	/* The CPU has at least one callback queued somewhere. */
+
+	if (rdp->donelist != NULL ||
+	    !!rdp->waitlistcount ||
+	    rdp->nextlist != NULL)
+		return 1;
+
+	/* The RCU core needs an acknowledgement from this CPU. */
+
+	if ((per_cpu(rcu_flip_flag, cpu) == rcu_flipped) ||
+	    (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed))
+		return 1;
+
+	/* This CPU has fallen behind the global grace-period number. */
+
+	if (rdp->completed != rcu_ctrlblk.completed)
+		return 1;
+
+	/* Nothing needed from this CPU. */
+
+	return 0;
+}
+
+void __init __rcu_init(void)
+{
+	int cpu;
+	int i;
+	struct rcu_data *rdp;
+
+	printk(KERN_NOTICE "Preemptible RCU implementation.\n");
+	for_each_possible_cpu(cpu) {
+		rdp = RCU_DATA_CPU(cpu);
+		spin_lock_init(&rdp->lock);
+		rdp->completed = 0;
+		rdp->waitlistcount = 0;
+		rdp->nextlist = NULL;
+		rdp->nexttail = &rdp->nextlist;
+		for (i = 0; i < GP_STAGES; i++) {
+			rdp->waitlist[i] = NULL;
+			rdp->waittail[i] = &rdp->waitlist[i];
+		}
+		rdp->donelist = NULL;
+		rdp->donetail = &rdp->donelist;
+		rdp->rcu_flipctr[0] = 0;
+		rdp->rcu_flipctr[1] = 0;
+	}
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+}
+
+/*
+ * Deprecated, use synchronize_rcu() or synchronize_sched() instead.
+ */
+void synchronize_kernel(void)
+{
+	synchronize_rcu();
+}
+
+#ifdef CONFIG_RCU_TRACE
+long *rcupreempt_flipctr(int cpu)
+{
+	return &RCU_DATA_CPU(cpu)->rcu_flipctr[0];
+}
+EXPORT_SYMBOL_GPL(rcupreempt_flipctr);
+
+int rcupreempt_flip_flag(int cpu)
+{
+	return per_cpu(rcu_flip_flag, cpu);
+}
+EXPORT_SYMBOL_GPL(rcupreempt_flip_flag);
+
+int rcupreempt_mb_flag(int cpu)
+{
+	return per_cpu(rcu_mb_flag, cpu);
+}
+EXPORT_SYMBOL_GPL(rcupreempt_mb_flag);
+
+char *rcupreempt_try_flip_state_name(void)
+{
+	return rcu_try_flip_state_names[rcu_ctrlblk.rcu_try_flip_state];
+}
+EXPORT_SYMBOL_GPL(rcupreempt_try_flip_state_name);
+
+struct rcupreempt_trace *rcupreempt_trace_cpu(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	return &rdp->trace;
+}
+EXPORT_SYMBOL_GPL(rcupreempt_trace_cpu);
+
+#endif /* #ifdef RCU_TRACE */
diff --git a/kernel/rcupreempt_trace.c b/kernel/rcupreempt_trace.c
new file mode 100644
index 000000000000..49ac4947af24
--- /dev/null
+++ b/kernel/rcupreempt_trace.c
@@ -0,0 +1,330 @@
+/*
+ * Read-Copy Update tracing for realtime implementation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2006
+ *
+ * Papers:  http://www.rdrop.com/users/paulmck/RCU
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU/ *.txt
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/rcupdate.h>
+#include <linux/cpu.h>
+#include <linux/mutex.h>
+#include <linux/rcupreempt_trace.h>
+#include <linux/debugfs.h>
+
+static struct mutex rcupreempt_trace_mutex;
+static char *rcupreempt_trace_buf;
+#define RCUPREEMPT_TRACE_BUF_SIZE 4096
+
+void rcupreempt_trace_move2done(struct rcupreempt_trace *trace)
+{
+	trace->done_length += trace->wait_length;
+	trace->done_add += trace->wait_length;
+	trace->wait_length = 0;
+}
+void rcupreempt_trace_move2wait(struct rcupreempt_trace *trace)
+{
+	trace->wait_length += trace->next_length;
+	trace->wait_add += trace->next_length;
+	trace->next_length = 0;
+}
+void rcupreempt_trace_try_flip_1(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->rcu_try_flip_1);
+}
+void rcupreempt_trace_try_flip_e1(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->rcu_try_flip_e1);
+}
+void rcupreempt_trace_try_flip_i1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_i1++;
+}
+void rcupreempt_trace_try_flip_ie1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ie1++;
+}
+void rcupreempt_trace_try_flip_g1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_g1++;
+}
+void rcupreempt_trace_try_flip_a1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_a1++;
+}
+void rcupreempt_trace_try_flip_ae1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ae1++;
+}
+void rcupreempt_trace_try_flip_a2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_a2++;
+}
+void rcupreempt_trace_try_flip_z1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_z1++;
+}
+void rcupreempt_trace_try_flip_ze1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ze1++;
+}
+void rcupreempt_trace_try_flip_z2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_z2++;
+}
+void rcupreempt_trace_try_flip_m1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_m1++;
+}
+void rcupreempt_trace_try_flip_me1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_me1++;
+}
+void rcupreempt_trace_try_flip_m2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_m2++;
+}
+void rcupreempt_trace_check_callbacks(struct rcupreempt_trace *trace)
+{
+	trace->rcu_check_callbacks++;
+}
+void rcupreempt_trace_done_remove(struct rcupreempt_trace *trace)
+{
+	trace->done_remove += trace->done_length;
+	trace->done_length = 0;
+}
+void rcupreempt_trace_invoke(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->done_invoked);
+}
+void rcupreempt_trace_next_add(struct rcupreempt_trace *trace)
+{
+	trace->next_add++;
+	trace->next_length++;
+}
+
+static void rcupreempt_trace_sum(struct rcupreempt_trace *sp)
+{
+	struct rcupreempt_trace *cp;
+	int cpu;
+
+	memset(sp, 0, sizeof(*sp));
+	for_each_possible_cpu(cpu) {
+		cp = rcupreempt_trace_cpu(cpu);
+		sp->next_length += cp->next_length;
+		sp->next_add += cp->next_add;
+		sp->wait_length += cp->wait_length;
+		sp->wait_add += cp->wait_add;
+		sp->done_length += cp->done_length;
+		sp->done_add += cp->done_add;
+		sp->done_remove += cp->done_remove;
+		atomic_set(&sp->done_invoked, atomic_read(&cp->done_invoked));
+		sp->rcu_check_callbacks += cp->rcu_check_callbacks;
+		atomic_set(&sp->rcu_try_flip_1,
+			   atomic_read(&cp->rcu_try_flip_1));
+		atomic_set(&sp->rcu_try_flip_e1,
+			   atomic_read(&cp->rcu_try_flip_e1));
+		sp->rcu_try_flip_i1 += cp->rcu_try_flip_i1;
+		sp->rcu_try_flip_ie1 += cp->rcu_try_flip_ie1;
+		sp->rcu_try_flip_g1 += cp->rcu_try_flip_g1;
+		sp->rcu_try_flip_a1 += cp->rcu_try_flip_a1;
+		sp->rcu_try_flip_ae1 += cp->rcu_try_flip_ae1;
+		sp->rcu_try_flip_a2 += cp->rcu_try_flip_a2;
+		sp->rcu_try_flip_z1 += cp->rcu_try_flip_z1;
+		sp->rcu_try_flip_ze1 += cp->rcu_try_flip_ze1;
+		sp->rcu_try_flip_z2 += cp->rcu_try_flip_z2;
+		sp->rcu_try_flip_m1 += cp->rcu_try_flip_m1;
+		sp->rcu_try_flip_me1 += cp->rcu_try_flip_me1;
+		sp->rcu_try_flip_m2 += cp->rcu_try_flip_m2;
+	}
+}
+
+static ssize_t rcustats_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	struct rcupreempt_trace trace;
+	ssize_t bcount;
+	int cnt = 0;
+
+	rcupreempt_trace_sum(&trace);
+	mutex_lock(&rcupreempt_trace_mutex);
+	snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+		 "ggp=%ld rcc=%ld\n",
+		 rcu_batches_completed(),
+		 trace.rcu_check_callbacks);
+	snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+		 "na=%ld nl=%ld wa=%ld wl=%ld da=%ld dl=%ld dr=%ld di=%d\n"
+		 "1=%d e1=%d i1=%ld ie1=%ld g1=%ld a1=%ld ae1=%ld a2=%ld\n"
+		 "z1=%ld ze1=%ld z2=%ld m1=%ld me1=%ld m2=%ld\n",
+
+		 trace.next_add, trace.next_length,
+		 trace.wait_add, trace.wait_length,
+		 trace.done_add, trace.done_length,
+		 trace.done_remove, atomic_read(&trace.done_invoked),
+		 atomic_read(&trace.rcu_try_flip_1),
+		 atomic_read(&trace.rcu_try_flip_e1),
+		 trace.rcu_try_flip_i1, trace.rcu_try_flip_ie1,
+		 trace.rcu_try_flip_g1,
+		 trace.rcu_try_flip_a1, trace.rcu_try_flip_ae1,
+			 trace.rcu_try_flip_a2,
+		 trace.rcu_try_flip_z1, trace.rcu_try_flip_ze1,
+			 trace.rcu_try_flip_z2,
+		 trace.rcu_try_flip_m1, trace.rcu_try_flip_me1,
+			trace.rcu_try_flip_m2);
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static ssize_t rcugp_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	long oldgp = rcu_batches_completed();
+	ssize_t bcount;
+
+	mutex_lock(&rcupreempt_trace_mutex);
+	synchronize_rcu();
+	snprintf(rcupreempt_trace_buf, RCUPREEMPT_TRACE_BUF_SIZE,
+		"oldggp=%ld  newggp=%ld\n", oldgp, rcu_batches_completed());
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static ssize_t rcuctrs_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	int cnt = 0;
+	int cpu;
+	int f = rcu_batches_completed() & 0x1;
+	ssize_t bcount;
+
+	mutex_lock(&rcupreempt_trace_mutex);
+
+	cnt += snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE,
+				"CPU last cur F M\n");
+	for_each_online_cpu(cpu) {
+		long *flipctr = rcupreempt_flipctr(cpu);
+		cnt += snprintf(&rcupreempt_trace_buf[cnt],
+				RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+					"%3d %4ld %3ld %d %d\n",
+			       cpu,
+			       flipctr[!f],
+			       flipctr[f],
+			       rcupreempt_flip_flag(cpu),
+			       rcupreempt_mb_flag(cpu));
+	}
+	cnt += snprintf(&rcupreempt_trace_buf[cnt],
+			RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+			"ggp = %ld, state = %s\n",
+			rcu_batches_completed(),
+			rcupreempt_try_flip_state_name());
+	cnt += snprintf(&rcupreempt_trace_buf[cnt],
+			RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+			"\n");
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static struct file_operations rcustats_fops = {
+	.owner = THIS_MODULE,
+	.read = rcustats_read,
+};
+
+static struct file_operations rcugp_fops = {
+	.owner = THIS_MODULE,
+	.read = rcugp_read,
+};
+
+static struct file_operations rcuctrs_fops = {
+	.owner = THIS_MODULE,
+	.read = rcuctrs_read,
+};
+
+static struct dentry *rcudir, *statdir, *ctrsdir, *gpdir;
+static int rcupreempt_debugfs_init(void)
+{
+	rcudir = debugfs_create_dir("rcu", NULL);
+	if (!rcudir)
+		goto out;
+	statdir = debugfs_create_file("rcustats", 0444, rcudir,
+						NULL, &rcustats_fops);
+	if (!statdir)
+		goto free_out;
+
+	gpdir = debugfs_create_file("rcugp", 0444, rcudir, NULL, &rcugp_fops);
+	if (!gpdir)
+		goto free_out;
+
+	ctrsdir = debugfs_create_file("rcuctrs", 0444, rcudir,
+						NULL, &rcuctrs_fops);
+	if (!ctrsdir)
+		goto free_out;
+	return 0;
+free_out:
+	if (statdir)
+		debugfs_remove(statdir);
+	if (gpdir)
+		debugfs_remove(gpdir);
+	debugfs_remove(rcudir);
+out:
+	return 1;
+}
+
+static int __init rcupreempt_trace_init(void)
+{
+	mutex_init(&rcupreempt_trace_mutex);
+	rcupreempt_trace_buf = kmalloc(RCUPREEMPT_TRACE_BUF_SIZE, GFP_KERNEL);
+	if (!rcupreempt_trace_buf)
+		return 1;
+	return rcupreempt_debugfs_init();
+}
+
+static void __exit rcupreempt_trace_cleanup(void)
+{
+	debugfs_remove(statdir);
+	debugfs_remove(gpdir);
+	debugfs_remove(ctrsdir);
+	debugfs_remove(rcudir);
+	kfree(rcupreempt_trace_buf);
+}
+
+
+module_init(rcupreempt_trace_init);
+module_exit(rcupreempt_trace_cleanup);
-- 
cgit v1.2.3


From fa717060f1ab7eb6570f2fb49136f838fc9195a9 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:27 +0100
Subject: sched: sched_rt_entity

Move the task_struct members specific to rt scheduling together.
A future optimization could be to put sched_entity and sched_rt_entity
into a union.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/init_task.h |  5 +++--
 include/linux/sched.h     |  8 ++++++--
 kernel/sched.c            |  2 +-
 kernel/sched_rt.c         | 20 ++++++++++----------
 mm/oom_kill.c             |  2 +-
 5 files changed, 21 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 572c65bcc80f..ee65d87bedb7 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -133,9 +133,10 @@ extern struct group_info init_groups;
 	.nr_cpus_allowed = NR_CPUS,					\
 	.mm		= NULL,						\
 	.active_mm	= &init_mm,					\
-	.run_list	= LIST_HEAD_INIT(tsk.run_list),			\
+	.rt		= {						\
+		.run_list	= LIST_HEAD_INIT(tsk.rt.run_list),	\
+		.time_slice	= HZ, },				\
 	.ioprio		= 0,						\
-	.time_slice	= HZ,						\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
 	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
 	.ptrace_list	= LIST_HEAD_INIT(tsk.ptrace_list),		\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 72e1b8ecfbe1..a06d09ebd5c6 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -929,6 +929,11 @@ struct sched_entity {
 #endif
 };
 
+struct sched_rt_entity {
+	struct list_head run_list;
+	unsigned int time_slice;
+};
+
 struct task_struct {
 	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
 	void *stack;
@@ -945,9 +950,9 @@ struct task_struct {
 #endif
 
 	int prio, static_prio, normal_prio;
-	struct list_head run_list;
 	const struct sched_class *sched_class;
 	struct sched_entity se;
+	struct sched_rt_entity rt;
 
 #ifdef CONFIG_PREEMPT_NOTIFIERS
 	/* list of struct preempt_notifier: */
@@ -972,7 +977,6 @@ struct task_struct {
 	unsigned int policy;
 	cpumask_t cpus_allowed;
 	int nr_cpus_allowed;
-	unsigned int time_slice;
 
 #ifdef CONFIG_PREEMPT_RCU
 	int rcu_read_lock_nesting;
diff --git a/kernel/sched.c b/kernel/sched.c
index 02d468844a91..c2cedd09d895 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1685,7 +1685,7 @@ static void __sched_fork(struct task_struct *p)
 	p->se.wait_max			= 0;
 #endif
 
-	INIT_LIST_HEAD(&p->run_list);
+	INIT_LIST_HEAD(&p->rt.run_list);
 	p->se.on_rq = 0;
 
 #ifdef CONFIG_PREEMPT_NOTIFIERS
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 9affb3c9d3db..29963af782ae 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -111,7 +111,7 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
 {
 	struct rt_prio_array *array = &rq->rt.active;
 
-	list_add_tail(&p->run_list, array->queue + p->prio);
+	list_add_tail(&p->rt.run_list, array->queue + p->prio);
 	__set_bit(p->prio, array->bitmap);
 	inc_cpu_load(rq, p->se.load.weight);
 
@@ -127,7 +127,7 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 
 	update_curr_rt(rq);
 
-	list_del(&p->run_list);
+	list_del(&p->rt.run_list);
 	if (list_empty(array->queue + p->prio))
 		__clear_bit(p->prio, array->bitmap);
 	dec_cpu_load(rq, p->se.load.weight);
@@ -143,7 +143,7 @@ static void requeue_task_rt(struct rq *rq, struct task_struct *p)
 {
 	struct rt_prio_array *array = &rq->rt.active;
 
-	list_move_tail(&p->run_list, array->queue + p->prio);
+	list_move_tail(&p->rt.run_list, array->queue + p->prio);
 }
 
 static void
@@ -212,7 +212,7 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 		return NULL;
 
 	queue = array->queue + idx;
-	next = list_entry(queue->next, struct task_struct, run_list);
+	next = list_entry(queue->next, struct task_struct, rt.run_list);
 
 	next->se.exec_start = rq->clock;
 
@@ -261,14 +261,14 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 	queue = array->queue + idx;
 	BUG_ON(list_empty(queue));
 
-	next = list_entry(queue->next, struct task_struct, run_list);
+	next = list_entry(queue->next, struct task_struct, rt.run_list);
 	if (unlikely(pick_rt_task(rq, next, cpu)))
 		goto out;
 
 	if (queue->next->next != queue) {
 		/* same prio task */
 		next = list_entry(queue->next->next, struct task_struct,
-				  run_list);
+				  rt.run_list);
 		if (pick_rt_task(rq, next, cpu))
 			goto out;
 	}
@@ -282,7 +282,7 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 	queue = array->queue + idx;
 	BUG_ON(list_empty(queue));
 
-	list_for_each_entry(next, queue, run_list) {
+	list_for_each_entry(next, queue, rt.run_list) {
 		if (pick_rt_task(rq, next, cpu))
 			goto out;
 	}
@@ -846,16 +846,16 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p)
 	if (p->policy != SCHED_RR)
 		return;
 
-	if (--p->time_slice)
+	if (--p->rt.time_slice)
 		return;
 
-	p->time_slice = DEF_TIMESLICE;
+	p->rt.time_slice = DEF_TIMESLICE;
 
 	/*
 	 * Requeue to the end of queue if we are not the only element
 	 * on the queue:
 	 */
-	if (p->run_list.prev != p->run_list.next) {
+	if (p->rt.run_list.prev != p->rt.run_list.next) {
 		requeue_task_rt(rq, p);
 		set_tsk_need_resched(p);
 	}
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 91a081a82f55..96473b482099 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -286,7 +286,7 @@ static void __oom_kill_task(struct task_struct *p, int verbose)
 	 * all the memory it needs. That way it should be able to
 	 * exit() and clear out its resources quickly...
 	 */
-	p->time_slice = HZ;
+	p->rt.time_slice = HZ;
 	set_tsk_thread_flag(p, TIF_MEMDIE);
 
 	force_sig(SIGKILL, p);
-- 
cgit v1.2.3


From 78f2c7db6068fd6ef75b8c120f04a388848eacb5 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:27 +0100
Subject: sched: SCHED_FIFO/SCHED_RR watchdog timer

Introduce a new rlimit that allows the user to set a runtime timeout on
real-time tasks their slice. Once this limit is exceeded the task will receive
SIGXCPU.

So it measures runtime since the last sleep.

Input and ideas by Thomas Gleixner and Lennart Poettering.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Lennart Poettering <mzxreary@0pointer.de>
CC: Michael Kerrisk <mtk.manpages@googlemail.com>
CC: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-generic/resource.h |  5 +++--
 include/linux/sched.h          |  1 +
 kernel/posix-cpu-timers.c      | 29 +++++++++++++++++++++++++++++
 kernel/sched_rt.c              | 30 ++++++++++++++++++++++++++++++
 4 files changed, 63 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/resource.h b/include/asm-generic/resource.h
index a4a22cc35898..587566f95f6c 100644
--- a/include/asm-generic/resource.h
+++ b/include/asm-generic/resource.h
@@ -44,8 +44,8 @@
 #define RLIMIT_NICE		13	/* max nice prio allowed to raise to
 					   0-39 for nice level 19 .. -20 */
 #define RLIMIT_RTPRIO		14	/* maximum realtime priority */
-
-#define RLIM_NLIMITS		15
+#define RLIMIT_RTTIME		15	/* timeout for RT tasks in us */
+#define RLIM_NLIMITS		16
 
 /*
  * SuS says limits have to be unsigned.
@@ -86,6 +86,7 @@
 	[RLIMIT_MSGQUEUE]	= {   MQ_BYTES_MAX,   MQ_BYTES_MAX },	\
 	[RLIMIT_NICE]		= { 0, 0 },				\
 	[RLIMIT_RTPRIO]		= { 0, 0 },				\
+	[RLIMIT_RTTIME]		= {  RLIM_INFINITY,  RLIM_INFINITY },	\
 }
 
 #endif	/* __KERNEL__ */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index a06d09ebd5c6..fe3f8fbc614e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -932,6 +932,7 @@ struct sched_entity {
 struct sched_rt_entity {
 	struct list_head run_list;
 	unsigned int time_slice;
+	unsigned long timeout;
 };
 
 struct task_struct {
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 68c96376e84a..2c076b36c4f6 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -967,6 +967,7 @@ static void check_thread_timers(struct task_struct *tsk,
 {
 	int maxfire;
 	struct list_head *timers = tsk->cpu_timers;
+	struct signal_struct *const sig = tsk->signal;
 
 	maxfire = 20;
 	tsk->it_prof_expires = cputime_zero;
@@ -1011,6 +1012,34 @@ static void check_thread_timers(struct task_struct *tsk,
 		t->firing = 1;
 		list_move_tail(&t->entry, firing);
 	}
+
+	/*
+	 * Check for the special case thread timers.
+	 */
+	if (sig->rlim[RLIMIT_RTTIME].rlim_cur != RLIM_INFINITY) {
+		unsigned long hard = sig->rlim[RLIMIT_RTTIME].rlim_max;
+		unsigned long *soft = &sig->rlim[RLIMIT_RTTIME].rlim_cur;
+
+		if (tsk->rt.timeout > DIV_ROUND_UP(hard, USEC_PER_SEC/HZ)) {
+			/*
+			 * At the hard limit, we just die.
+			 * No need to calculate anything else now.
+			 */
+			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
+			return;
+		}
+		if (tsk->rt.timeout > DIV_ROUND_UP(*soft, USEC_PER_SEC/HZ)) {
+			/*
+			 * At the soft limit, send a SIGXCPU every second.
+			 */
+			if (sig->rlim[RLIMIT_RTTIME].rlim_cur
+			    < sig->rlim[RLIMIT_RTTIME].rlim_max) {
+				sig->rlim[RLIMIT_RTTIME].rlim_cur +=
+								USEC_PER_SEC;
+			}
+			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
+		}
+	}
 }
 
 /*
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 29963af782ae..f350f7b15158 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -116,6 +116,9 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
 	inc_cpu_load(rq, p->se.load.weight);
 
 	inc_rt_tasks(p, rq);
+
+	if (wakeup)
+		p->rt.timeout = 0;
 }
 
 /*
@@ -834,11 +837,38 @@ static void prio_changed_rt(struct rq *rq, struct task_struct *p,
 	}
 }
 
+static void watchdog(struct rq *rq, struct task_struct *p)
+{
+	unsigned long soft, hard;
+
+	if (!p->signal)
+		return;
+
+	soft = p->signal->rlim[RLIMIT_RTTIME].rlim_cur;
+	hard = p->signal->rlim[RLIMIT_RTTIME].rlim_max;
+
+	if (soft != RLIM_INFINITY) {
+		unsigned long next;
+
+		p->rt.timeout++;
+		next = DIV_ROUND_UP(min(soft, hard), USEC_PER_SEC/HZ);
+		if (next > p->rt.timeout) {
+			u64 next_time = p->se.sum_exec_runtime;
+
+			next_time += next * (NSEC_PER_SEC/HZ);
+			if (p->it_sched_expires > next_time)
+				p->it_sched_expires = next_time;
+		} else
+			p->it_sched_expires = p->se.sum_exec_runtime;
+	}
+}
 
 static void task_tick_rt(struct rq *rq, struct task_struct *p)
 {
 	update_curr_rt(rq);
 
+	watchdog(rq, p);
+
 	/*
 	 * RR tasks need a special form of timeslice management.
 	 * FIFO tasks have no timeslices.
-- 
cgit v1.2.3


From 02b67cc3ba36bdba351d6c3a00593f4ec550d9d3 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 25 Jan 2008 21:08:28 +0100
Subject: sched: do not do cond_resched() when CONFIG_PREEMPT

Why do we even have cond_resched when real preemption
is on? It seems to be a waste of space and time.

remove cond_resched with CONFIG_PREEMPT on.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/kernel.h |  4 ++--
 include/linux/sched.h  | 13 ++++++++++++-
 kernel/sched.c         |  6 ++++--
 3 files changed, 18 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 94bc99656963..a7283c9beadf 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -105,8 +105,8 @@ struct user;
  * supposed to.
  */
 #ifdef CONFIG_PREEMPT_VOLUNTARY
-extern int cond_resched(void);
-# define might_resched() cond_resched()
+extern int _cond_resched(void);
+# define might_resched() _cond_resched()
 #else
 # define might_resched() do { } while (0)
 #endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index fe3f8fbc614e..7907845c2348 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1885,7 +1885,18 @@ static inline int need_resched(void)
  * cond_resched_lock() will drop the spinlock before scheduling,
  * cond_resched_softirq() will enable bhs before scheduling.
  */
-extern int cond_resched(void);
+#ifdef CONFIG_PREEMPT
+static inline int cond_resched(void)
+{
+	return 0;
+}
+#else
+extern int _cond_resched(void);
+static inline int cond_resched(void)
+{
+	return _cond_resched();
+}
+#endif
 extern int cond_resched_lock(spinlock_t * lock);
 extern int cond_resched_softirq(void);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index b9ee0f4db66a..6ee37602a6d8 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4678,7 +4678,8 @@ static void __cond_resched(void)
 	} while (need_resched());
 }
 
-int __sched cond_resched(void)
+#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PREEMPT_VOLUNTARY)
+int __sched _cond_resched(void)
 {
 	if (need_resched() && !(preempt_count() & PREEMPT_ACTIVE) &&
 					system_state == SYSTEM_RUNNING) {
@@ -4687,7 +4688,8 @@ int __sched cond_resched(void)
 	}
 	return 0;
 }
-EXPORT_SYMBOL(cond_resched);
+EXPORT_SYMBOL(_cond_resched);
+#endif
 
 /*
  * cond_resched_lock() - if a reschedule is pending, drop the given lock,
-- 
cgit v1.2.3


From 8f4d37ec073c17e2d4aa8851df5837d798606d6f Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:29 +0100
Subject: sched: high-res preemption tick

Use HR-timers (when available) to deliver an accurate preemption tick.

The regular scheduler tick that runs at 1/HZ can be too coarse when nice
level are used. The fairness system will still keep the cpu utilisation 'fair'
by then delaying the task that got an excessive amount of CPU time but try to
minimize this by delivering preemption points spot-on.

The average frequency of this extra interrupt is sched_latency / nr_latency.
Which need not be higher than 1/HZ, its just that the distribution within the
sched_latency period is important.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/entry_64.S       |   6 +-
 arch/x86/kernel/signal_32.c      |   3 +
 arch/x86/kernel/signal_64.c      |   3 +
 include/asm-x86/thread_info_32.h |   2 +
 include/asm-x86/thread_info_64.h |   5 +
 include/linux/hrtimer.h          |   9 ++
 include/linux/sched.h            |   3 +-
 kernel/Kconfig.hz                |   2 +
 kernel/sched.c                   | 210 +++++++++++++++++++++++++++++++++++++--
 kernel/sched_fair.c              |  69 ++++++++++++-
 kernel/sched_idletask.c          |   2 +-
 kernel/sched_rt.c                |   2 +-
 12 files changed, 295 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 3a058bb16409..e70f3881d7e4 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -283,7 +283,7 @@ sysret_careful:
 sysret_signal:
 	TRACE_IRQS_ON
 	sti
-	testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
+	testl $_TIF_DO_NOTIFY_MASK,%edx
 	jz    1f
 
 	/* Really a signal */
@@ -377,7 +377,7 @@ int_very_careful:
 	jmp int_restore_rest
 	
 int_signal:
-	testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
+	testl $_TIF_DO_NOTIFY_MASK,%edx
 	jz 1f
 	movq %rsp,%rdi		# &ptregs -> arg1
 	xorl %esi,%esi		# oldset -> arg2
@@ -603,7 +603,7 @@ retint_careful:
 	jmp retint_check
 	
 retint_signal:
-	testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
+	testl $_TIF_DO_NOTIFY_MASK,%edx
 	jz    retint_swapgs
 	TRACE_IRQS_ON
 	sti
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 9bdd83022f5f..20f29e4c1d33 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -658,6 +658,9 @@ void do_notify_resume(struct pt_regs *regs, void *_unused,
 	/* deal with pending signal delivery */
 	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
 		do_signal(regs);
+
+	if (thread_info_flags & _TIF_HRTICK_RESCHED)
+		hrtick_resched();
 	
 	clear_thread_flag(TIF_IRET);
 }
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index ab086b0357fc..38d806467c0f 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -480,6 +480,9 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 	/* deal with pending signal delivery */
 	if (thread_info_flags & (_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK))
 		do_signal(regs);
+
+	if (thread_info_flags & _TIF_HRTICK_RESCHED)
+		hrtick_resched();
 }
 
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index 22a8cbcd35e2..ef58fd2a6eb0 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -132,6 +132,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_SYSCALL_AUDIT	6	/* syscall auditing active */
 #define TIF_SECCOMP		7	/* secure computing */
 #define TIF_RESTORE_SIGMASK	8	/* restore signal mask in do_signal() */
+#define TIF_HRTICK_RESCHED	9	/* reprogram hrtick timer */
 #define TIF_MEMDIE		16
 #define TIF_DEBUG		17	/* uses debug registers */
 #define TIF_IO_BITMAP		18	/* uses I/O bitmap */
@@ -147,6 +148,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
 #define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
+#define _TIF_HRTICK_RESCHED	(1<<TIF_HRTICK_RESCHED)
 #define _TIF_DEBUG		(1<<TIF_DEBUG)
 #define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index beae2bfb62ca..7f6ee68f0002 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -115,6 +115,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_SECCOMP		8	/* secure computing */
 #define TIF_RESTORE_SIGMASK	9	/* restore signal mask in do_signal */
 #define TIF_MCE_NOTIFY		10	/* notify userspace of an MCE */
+#define TIF_HRTICK_RESCHED	11	/* reprogram hrtick timer */
 /* 16 free */
 #define TIF_IA32		17	/* 32bit process */ 
 #define TIF_FORK		18	/* ret_from_fork */
@@ -133,6 +134,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
 #define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_MCE_NOTIFY		(1<<TIF_MCE_NOTIFY)
+#define _TIF_HRTICK_RESCHED	(1<<TIF_HRTICK_RESCHED)
 #define _TIF_IA32		(1<<TIF_IA32)
 #define _TIF_FORK		(1<<TIF_FORK)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
@@ -146,6 +148,9 @@ static inline struct thread_info *stack_thread_info(void)
 /* work to do on any return to user space */
 #define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
 
+#define _TIF_DO_NOTIFY_MASK \
+	(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY|_TIF_HRTICK_RESCHED)
+
 /* flags to check in __switch_to() */
 #define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP)
 
diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 7a9398e19704..ecc8e2685e2b 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -217,6 +217,11 @@ static inline ktime_t hrtimer_cb_get_time(struct hrtimer *timer)
 	return timer->base->get_time();
 }
 
+static inline int hrtimer_is_hres_active(struct hrtimer *timer)
+{
+	return timer->base->cpu_base->hres_active;
+}
+
 /*
  * The resolution of the clocks. The resolution value is returned in
  * the clock_getres() system call to give application programmers an
@@ -248,6 +253,10 @@ static inline ktime_t hrtimer_cb_get_time(struct hrtimer *timer)
 	return timer->base->softirq_time;
 }
 
+static inline int hrtimer_is_hres_active(struct hrtimer *timer)
+{
+	return 0;
+}
 #endif
 
 extern ktime_t ktime_get(void);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 7907845c2348..43e0339d65fc 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -257,6 +257,7 @@ extern void trap_init(void);
 extern void account_process_tick(struct task_struct *task, int user);
 extern void update_process_times(int user);
 extern void scheduler_tick(void);
+extern void hrtick_resched(void);
 
 extern void sched_show_task(struct task_struct *p);
 
@@ -849,7 +850,7 @@ struct sched_class {
 #endif
 
 	void (*set_curr_task) (struct rq *rq);
-	void (*task_tick) (struct rq *rq, struct task_struct *p);
+	void (*task_tick) (struct rq *rq, struct task_struct *p, int queued);
 	void (*task_new) (struct rq *rq, struct task_struct *p);
 	void (*set_cpus_allowed)(struct task_struct *p, cpumask_t *newmask);
 
diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz
index 4af15802ccd4..526128a2e622 100644
--- a/kernel/Kconfig.hz
+++ b/kernel/Kconfig.hz
@@ -54,3 +54,5 @@ config HZ
 	default 300 if HZ_300
 	default 1000 if HZ_1000
 
+config SCHED_HRTICK
+	def_bool HIGH_RES_TIMERS && X86
diff --git a/kernel/sched.c b/kernel/sched.c
index 6ee37602a6d8..17f93d3eda91 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -65,6 +65,7 @@
 #include <linux/reciprocal_div.h>
 #include <linux/unistd.h>
 #include <linux/pagemap.h>
+#include <linux/hrtimer.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>
@@ -451,6 +452,12 @@ struct rq {
 	struct list_head migration_queue;
 #endif
 
+#ifdef CONFIG_SCHED_HRTICK
+	unsigned long hrtick_flags;
+	ktime_t hrtick_expire;
+	struct hrtimer hrtick_timer;
+#endif
+
 #ifdef CONFIG_SCHEDSTATS
 	/* latency stats */
 	struct sched_info rq_sched_info;
@@ -572,6 +579,8 @@ enum {
 	SCHED_FEAT_START_DEBIT		= 4,
 	SCHED_FEAT_TREE_AVG		= 8,
 	SCHED_FEAT_APPROX_AVG		= 16,
+	SCHED_FEAT_HRTICK		= 32,
+	SCHED_FEAT_DOUBLE_TICK		= 64,
 };
 
 const_debug unsigned int sysctl_sched_features =
@@ -579,7 +588,9 @@ const_debug unsigned int sysctl_sched_features =
 		SCHED_FEAT_WAKEUP_PREEMPT	* 1 |
 		SCHED_FEAT_START_DEBIT		* 1 |
 		SCHED_FEAT_TREE_AVG		* 0 |
-		SCHED_FEAT_APPROX_AVG		* 0;
+		SCHED_FEAT_APPROX_AVG		* 0 |
+		SCHED_FEAT_HRTICK		* 1 |
+		SCHED_FEAT_DOUBLE_TICK		* 0;
 
 #define sched_feat(x) (sysctl_sched_features & SCHED_FEAT_##x)
 
@@ -796,6 +807,173 @@ void sched_clock_idle_wakeup_event(u64 delta_ns)
 }
 EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
 
+static void __resched_task(struct task_struct *p, int tif_bit);
+
+static inline void resched_task(struct task_struct *p)
+{
+	__resched_task(p, TIF_NEED_RESCHED);
+}
+
+#ifdef CONFIG_SCHED_HRTICK
+/*
+ * Use HR-timers to deliver accurate preemption points.
+ *
+ * Its all a bit involved since we cannot program an hrt while holding the
+ * rq->lock. So what we do is store a state in in rq->hrtick_* and ask for a
+ * reschedule event.
+ *
+ * When we get rescheduled we reprogram the hrtick_timer outside of the
+ * rq->lock.
+ */
+static inline void resched_hrt(struct task_struct *p)
+{
+	__resched_task(p, TIF_HRTICK_RESCHED);
+}
+
+static inline void resched_rq(struct rq *rq)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rq->lock, flags);
+	resched_task(rq->curr);
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+enum {
+	HRTICK_SET,		/* re-programm hrtick_timer */
+	HRTICK_RESET,		/* not a new slice */
+};
+
+/*
+ * Use hrtick when:
+ *  - enabled by features
+ *  - hrtimer is actually high res
+ */
+static inline int hrtick_enabled(struct rq *rq)
+{
+	if (!sched_feat(HRTICK))
+		return 0;
+	return hrtimer_is_hres_active(&rq->hrtick_timer);
+}
+
+/*
+ * Called to set the hrtick timer state.
+ *
+ * called with rq->lock held and irqs disabled
+ */
+static void hrtick_start(struct rq *rq, u64 delay, int reset)
+{
+	assert_spin_locked(&rq->lock);
+
+	/*
+	 * preempt at: now + delay
+	 */
+	rq->hrtick_expire =
+		ktime_add_ns(rq->hrtick_timer.base->get_time(), delay);
+	/*
+	 * indicate we need to program the timer
+	 */
+	__set_bit(HRTICK_SET, &rq->hrtick_flags);
+	if (reset)
+		__set_bit(HRTICK_RESET, &rq->hrtick_flags);
+
+	/*
+	 * New slices are called from the schedule path and don't need a
+	 * forced reschedule.
+	 */
+	if (reset)
+		resched_hrt(rq->curr);
+}
+
+static void hrtick_clear(struct rq *rq)
+{
+	if (hrtimer_active(&rq->hrtick_timer))
+		hrtimer_cancel(&rq->hrtick_timer);
+}
+
+/*
+ * Update the timer from the possible pending state.
+ */
+static void hrtick_set(struct rq *rq)
+{
+	ktime_t time;
+	int set, reset;
+	unsigned long flags;
+
+	WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
+
+	spin_lock_irqsave(&rq->lock, flags);
+	set = __test_and_clear_bit(HRTICK_SET, &rq->hrtick_flags);
+	reset = __test_and_clear_bit(HRTICK_RESET, &rq->hrtick_flags);
+	time = rq->hrtick_expire;
+	clear_thread_flag(TIF_HRTICK_RESCHED);
+	spin_unlock_irqrestore(&rq->lock, flags);
+
+	if (set) {
+		hrtimer_start(&rq->hrtick_timer, time, HRTIMER_MODE_ABS);
+		if (reset && !hrtimer_active(&rq->hrtick_timer))
+			resched_rq(rq);
+	} else
+		hrtick_clear(rq);
+}
+
+/*
+ * High-resolution timer tick.
+ * Runs from hardirq context with interrupts disabled.
+ */
+static enum hrtimer_restart hrtick(struct hrtimer *timer)
+{
+	struct rq *rq = container_of(timer, struct rq, hrtick_timer);
+
+	WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
+
+	spin_lock(&rq->lock);
+	__update_rq_clock(rq);
+	rq->curr->sched_class->task_tick(rq, rq->curr, 1);
+	spin_unlock(&rq->lock);
+
+	return HRTIMER_NORESTART;
+}
+
+static inline void init_rq_hrtick(struct rq *rq)
+{
+	rq->hrtick_flags = 0;
+	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	rq->hrtick_timer.function = hrtick;
+	rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+}
+
+void hrtick_resched(void)
+{
+	struct rq *rq;
+	unsigned long flags;
+
+	if (!test_thread_flag(TIF_HRTICK_RESCHED))
+		return;
+
+	local_irq_save(flags);
+	rq = cpu_rq(smp_processor_id());
+	hrtick_set(rq);
+	local_irq_restore(flags);
+}
+#else
+static inline void hrtick_clear(struct rq *rq)
+{
+}
+
+static inline void hrtick_set(struct rq *rq)
+{
+}
+
+static inline void init_rq_hrtick(struct rq *rq)
+{
+}
+
+void hrtick_resched(void)
+{
+}
+#endif
+
 /*
  * resched_task - mark a task 'to be rescheduled now'.
  *
@@ -809,16 +987,16 @@ EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
 #define tsk_is_polling(t) test_tsk_thread_flag(t, TIF_POLLING_NRFLAG)
 #endif
 
-static void resched_task(struct task_struct *p)
+static void __resched_task(struct task_struct *p, int tif_bit)
 {
 	int cpu;
 
 	assert_spin_locked(&task_rq(p)->lock);
 
-	if (unlikely(test_tsk_thread_flag(p, TIF_NEED_RESCHED)))
+	if (unlikely(test_tsk_thread_flag(p, tif_bit)))
 		return;
 
-	set_tsk_thread_flag(p, TIF_NEED_RESCHED);
+	set_tsk_thread_flag(p, tif_bit);
 
 	cpu = task_cpu(p);
 	if (cpu == smp_processor_id())
@@ -841,10 +1019,10 @@ static void resched_cpu(int cpu)
 	spin_unlock_irqrestore(&rq->lock, flags);
 }
 #else
-static inline void resched_task(struct task_struct *p)
+static void __resched_task(struct task_struct *p, int tif_bit)
 {
 	assert_spin_locked(&task_rq(p)->lock);
-	set_tsk_need_resched(p);
+	set_tsk_thread_flag(p, tif_bit);
 }
 #endif
 
@@ -3497,7 +3675,7 @@ void scheduler_tick(void)
 	rq->tick_timestamp = rq->clock;
 	update_cpu_load(rq);
 	if (curr != rq->idle) /* FIXME: needed? */
-		curr->sched_class->task_tick(rq, curr);
+		curr->sched_class->task_tick(rq, curr, 0);
 	spin_unlock(&rq->lock);
 
 #ifdef CONFIG_SMP
@@ -3643,6 +3821,8 @@ need_resched_nonpreemptible:
 
 	schedule_debug(prev);
 
+	hrtick_clear(rq);
+
 	/*
 	 * Do the rq-clock update outside the rq lock:
 	 */
@@ -3680,14 +3860,20 @@ need_resched_nonpreemptible:
 		++*switch_count;
 
 		context_switch(rq, prev, next); /* unlocks the rq */
+		/*
+		 * the context switch might have flipped the stack from under
+		 * us, hence refresh the local variables.
+		 */
+		cpu = smp_processor_id();
+		rq = cpu_rq(cpu);
 	} else
 		spin_unlock_irq(&rq->lock);
 
-	if (unlikely(reacquire_kernel_lock(current) < 0)) {
-		cpu = smp_processor_id();
-		rq = cpu_rq(cpu);
+	hrtick_set(rq);
+
+	if (unlikely(reacquire_kernel_lock(current) < 0))
 		goto need_resched_nonpreemptible;
-	}
+
 	preempt_enable_no_resched();
 	if (unlikely(test_thread_flag(TIF_NEED_RESCHED)))
 		goto need_resched;
@@ -6913,6 +7099,8 @@ void __init sched_init(void)
 		rq->rt.overloaded = 0;
 		rq_attach_root(rq, &def_root_domain);
 #endif
+		init_rq_hrtick(rq);
+
 		atomic_set(&rq->nr_iowait, 0);
 
 		array = &rq->rt.active;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index dfa18d55561d..3dab1ff83c4f 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -642,13 +642,29 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev)
 	cfs_rq->curr = NULL;
 }
 
-static void entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
+static void
+entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued)
 {
 	/*
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
 
+#ifdef CONFIG_SCHED_HRTICK
+	/*
+	 * queued ticks are scheduled to match the slice, so don't bother
+	 * validating it and just reschedule.
+	 */
+	if (queued)
+		return resched_task(rq_of(cfs_rq)->curr);
+	/*
+	 * don't let the period tick interfere with the hrtick preemption
+	 */
+	if (!sched_feat(DOUBLE_TICK) &&
+			hrtimer_active(&rq_of(cfs_rq)->hrtick_timer))
+		return;
+#endif
+
 	if (cfs_rq->nr_running > 1 || !sched_feat(WAKEUP_PREEMPT))
 		check_preempt_tick(cfs_rq, curr);
 }
@@ -754,6 +770,43 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 
 #endif	/* CONFIG_FAIR_GROUP_SCHED */
 
+#ifdef CONFIG_SCHED_HRTICK
+static void hrtick_start_fair(struct rq *rq, struct task_struct *p)
+{
+	int requeue = rq->curr == p;
+	struct sched_entity *se = &p->se;
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+	WARN_ON(task_rq(p) != rq);
+
+	if (hrtick_enabled(rq) && cfs_rq->nr_running > 1) {
+		u64 slice = sched_slice(cfs_rq, se);
+		u64 ran = se->sum_exec_runtime - se->prev_sum_exec_runtime;
+		s64 delta = slice - ran;
+
+		if (delta < 0) {
+			if (rq->curr == p)
+				resched_task(p);
+			return;
+		}
+
+		/*
+		 * Don't schedule slices shorter than 10000ns, that just
+		 * doesn't make sense. Rely on vruntime for fairness.
+		 */
+		if (!requeue)
+			delta = max(10000LL, delta);
+
+		hrtick_start(rq, delta, requeue);
+	}
+}
+#else
+static inline void
+hrtick_start_fair(struct rq *rq, struct task_struct *p)
+{
+}
+#endif
+
 /*
  * The enqueue_task method is called before nr_running is
  * increased. Here we update the fair scheduling stats and
@@ -782,6 +835,8 @@ static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
 	 */
 	if (incload)
 		inc_cpu_load(rq, topse->load.weight);
+
+	hrtick_start_fair(rq, rq->curr);
 }
 
 /*
@@ -814,6 +869,8 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int sleep)
 	 */
 	if (decload)
 		dec_cpu_load(rq, topse->load.weight);
+
+	hrtick_start_fair(rq, rq->curr);
 }
 
 /*
@@ -1049,6 +1106,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p)
 
 static struct task_struct *pick_next_task_fair(struct rq *rq)
 {
+	struct task_struct *p;
 	struct cfs_rq *cfs_rq = &rq->cfs;
 	struct sched_entity *se;
 
@@ -1060,7 +1118,10 @@ static struct task_struct *pick_next_task_fair(struct rq *rq)
 		cfs_rq = group_cfs_rq(se);
 	} while (cfs_rq);
 
-	return task_of(se);
+	p = task_of(se);
+	hrtick_start_fair(rq, p);
+
+	return p;
 }
 
 /*
@@ -1235,14 +1296,14 @@ move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 /*
  * scheduler tick hitting a task of our scheduling class:
  */
-static void task_tick_fair(struct rq *rq, struct task_struct *curr)
+static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 {
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &curr->se;
 
 	for_each_sched_entity(se) {
 		cfs_rq = cfs_rq_of(se);
-		entity_tick(cfs_rq, se);
+		entity_tick(cfs_rq, se, queued);
 	}
 }
 
diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c
index ef7a2661fa10..2bcafa375633 100644
--- a/kernel/sched_idletask.c
+++ b/kernel/sched_idletask.c
@@ -61,7 +61,7 @@ move_one_task_idle(struct rq *this_rq, int this_cpu, struct rq *busiest,
 }
 #endif
 
-static void task_tick_idle(struct rq *rq, struct task_struct *curr)
+static void task_tick_idle(struct rq *rq, struct task_struct *curr, int queued)
 {
 }
 
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index f350f7b15158..83fbbcb8019e 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -863,7 +863,7 @@ static void watchdog(struct rq *rq, struct task_struct *p)
 	}
 }
 
-static void task_tick_rt(struct rq *rq, struct task_struct *p)
+static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 {
 	update_curr_rt(rq);
 
-- 
cgit v1.2.3


From fa85ae2418e6843953107cd6a06f645752829bc0 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:29 +0100
Subject: sched: rt time limit

Very simple time limit on the realtime scheduling classes.
Allow the rq's realtime class to consume sched_rt_ratio of every
sched_rt_period slice. If the class exceeds this quota the fair class
will preempt the realtime class.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  2 ++
 kernel/sched.c        | 70 ++++++++++++++++++++++++++++++++++++---------------
 kernel/sched_rt.c     | 53 ++++++++++++++++++++++++++++++++++++++
 kernel/sysctl.c       | 18 ++++++++++++-
 4 files changed, 122 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 43e0339d65fc..d5ea144df836 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1490,6 +1490,8 @@ extern unsigned int sysctl_sched_child_runs_first;
 extern unsigned int sysctl_sched_features;
 extern unsigned int sysctl_sched_migration_cost;
 extern unsigned int sysctl_sched_nr_migrate;
+extern unsigned int sysctl_sched_rt_period;
+extern unsigned int sysctl_sched_rt_ratio;
 #if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
 extern unsigned int sysctl_sched_min_bal_int_shares;
 extern unsigned int sysctl_sched_max_bal_int_shares;
diff --git a/kernel/sched.c b/kernel/sched.c
index 17f93d3eda91..e9a7beee9b79 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -342,13 +342,14 @@ struct cfs_rq {
 /* Real-Time classes' related field in a runqueue: */
 struct rt_rq {
 	struct rt_prio_array active;
-	int rt_load_balance_idx;
-	struct list_head *rt_load_balance_head, *rt_load_balance_curr;
 	unsigned long rt_nr_running;
+#ifdef CONFIG_SMP
 	unsigned long rt_nr_migratory;
-	/* highest queued rt task prio */
-	int highest_prio;
+	int highest_prio; /* highest queued rt task prio */
 	int overloaded;
+#endif
+	u64 rt_time;
+	u64 rt_throttled;
 };
 
 #ifdef CONFIG_SMP
@@ -415,6 +416,7 @@ struct rq {
 	struct list_head leaf_cfs_rq_list;
 #endif
 	struct rt_rq rt;
+	u64 rt_period_expire;
 
 	/*
 	 * This is part of a global counter where only the total sum
@@ -600,6 +602,21 @@ const_debug unsigned int sysctl_sched_features =
  */
 const_debug unsigned int sysctl_sched_nr_migrate = 32;
 
+/*
+ * period over which we measure -rt task cpu usage in ms.
+ * default: 1s
+ */
+const_debug unsigned int sysctl_sched_rt_period = 1000;
+
+#define SCHED_RT_FRAC_SHIFT	16
+#define SCHED_RT_FRAC		(1UL << SCHED_RT_FRAC_SHIFT)
+
+/*
+ * ratio of time -rt tasks may consume.
+ * default: 100%
+ */
+const_debug unsigned int sysctl_sched_rt_ratio = SCHED_RT_FRAC;
+
 /*
  * For kernel-internal use: high-speed (but slightly incorrect) per-cpu
  * clock constructed from sched_clock():
@@ -3674,8 +3691,8 @@ void scheduler_tick(void)
 		rq->clock = next_tick;
 	rq->tick_timestamp = rq->clock;
 	update_cpu_load(rq);
-	if (curr != rq->idle) /* FIXME: needed? */
-		curr->sched_class->task_tick(rq, curr, 0);
+	curr->sched_class->task_tick(rq, curr, 0);
+	update_sched_rt_period(rq);
 	spin_unlock(&rq->lock);
 
 #ifdef CONFIG_SMP
@@ -7041,6 +7058,29 @@ static void init_cfs_rq(struct cfs_rq *cfs_rq, struct rq *rq)
 	cfs_rq->min_vruntime = (u64)(-(1LL << 20));
 }
 
+static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq)
+{
+	struct rt_prio_array *array;
+	int i;
+
+	array = &rt_rq->active;
+	for (i = 0; i < MAX_RT_PRIO; i++) {
+		INIT_LIST_HEAD(array->queue + i);
+		__clear_bit(i, array->bitmap);
+	}
+	/* delimiter for bitsearch: */
+	__set_bit(MAX_RT_PRIO, array->bitmap);
+
+#ifdef CONFIG_SMP
+	rt_rq->rt_nr_migratory = 0;
+	rt_rq->highest_prio = MAX_RT_PRIO;
+	rt_rq->overloaded = 0;
+#endif
+
+	rt_rq->rt_time = 0;
+	rt_rq->rt_throttled = 0;
+}
+
 void __init sched_init(void)
 {
 	int highest_cpu = 0;
@@ -7051,7 +7091,6 @@ void __init sched_init(void)
 #endif
 
 	for_each_possible_cpu(i) {
-		struct rt_prio_array *array;
 		struct rq *rq;
 
 		rq = cpu_rq(i);
@@ -7083,6 +7122,8 @@ void __init sched_init(void)
 		}
 		init_task_group.shares = init_task_group_load;
 #endif
+		init_rt_rq(&rq->rt, rq);
+		rq->rt_period_expire = 0;
 
 		for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
 			rq->cpu_load[j] = 0;
@@ -7095,22 +7136,11 @@ void __init sched_init(void)
 		rq->cpu = i;
 		rq->migration_thread = NULL;
 		INIT_LIST_HEAD(&rq->migration_queue);
-		rq->rt.highest_prio = MAX_RT_PRIO;
-		rq->rt.overloaded = 0;
 		rq_attach_root(rq, &def_root_domain);
 #endif
 		init_rq_hrtick(rq);
-
 		atomic_set(&rq->nr_iowait, 0);
-
-		array = &rq->rt.active;
-		for (j = 0; j < MAX_RT_PRIO; j++) {
-			INIT_LIST_HEAD(array->queue + j);
-			__clear_bit(j, array->bitmap);
-		}
 		highest_cpu = i;
-		/* delimiter for bitsearch: */
-		__set_bit(MAX_RT_PRIO, array->bitmap);
 	}
 
 	set_load_weight(&init_task);
@@ -7282,7 +7312,7 @@ void set_curr_task(int cpu, struct task_struct *p)
 #ifdef CONFIG_SMP
 /*
  * distribute shares of all task groups among their schedulable entities,
- * to reflect load distrbution across cpus.
+ * to reflect load distribution across cpus.
  */
 static int rebalance_shares(struct sched_domain *sd, int this_cpu)
 {
@@ -7349,7 +7379,7 @@ static int rebalance_shares(struct sched_domain *sd, int this_cpu)
  * sysctl_sched_max_bal_int_shares represents the maximum interval between
  * consecutive calls to rebalance_shares() in the same sched domain.
  *
- * These settings allows for the appropriate tradeoff between accuracy of
+ * These settings allows for the appropriate trade-off between accuracy of
  * fairness and the associated overhead.
  *
  */
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 83fbbcb8019e..fd10d965aa06 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -45,6 +45,50 @@ static void update_rt_migration(struct rq *rq)
 }
 #endif /* CONFIG_SMP */
 
+static int sched_rt_ratio_exceeded(struct rq *rq, struct rt_rq *rt_rq)
+{
+	u64 period, ratio;
+
+	if (sysctl_sched_rt_ratio == SCHED_RT_FRAC)
+		return 0;
+
+	if (rt_rq->rt_throttled)
+		return 1;
+
+	period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
+	ratio = (period * sysctl_sched_rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+	if (rt_rq->rt_time > ratio) {
+		rt_rq->rt_throttled = rq->clock + period - rt_rq->rt_time;
+		return 1;
+	}
+
+	return 0;
+}
+
+static void update_sched_rt_period(struct rq *rq)
+{
+	while (rq->clock > rq->rt_period_expire) {
+		u64 period, ratio;
+
+		period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
+		ratio = (period * sysctl_sched_rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+		rq->rt.rt_time -= min(rq->rt.rt_time, ratio);
+		rq->rt_period_expire += period;
+	}
+
+	/*
+	 * When the rt throttle is expired, let them rip.
+	 * (XXX: use hrtick when available)
+	 */
+	if (rq->rt.rt_throttled && rq->clock > rq->rt.rt_throttled) {
+		rq->rt.rt_throttled = 0;
+		if (!sched_rt_ratio_exceeded(rq, &rq->rt))
+			resched_task(rq->curr);
+	}
+}
+
 /*
  * Update the current task's runtime statistics. Skip current tasks that
  * are not in our scheduling class.
@@ -66,6 +110,11 @@ static void update_curr_rt(struct rq *rq)
 	curr->se.sum_exec_runtime += delta_exec;
 	curr->se.exec_start = rq->clock;
 	cpuacct_charge(curr, delta_exec);
+
+	rq->rt.rt_time += delta_exec;
+	update_sched_rt_period(rq);
+	if (sched_rt_ratio_exceeded(rq, &rq->rt))
+		resched_task(curr);
 }
 
 static inline void inc_rt_tasks(struct task_struct *p, struct rq *rq)
@@ -208,8 +257,12 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 	struct rt_prio_array *array = &rq->rt.active;
 	struct task_struct *next;
 	struct list_head *queue;
+	struct rt_rq *rt_rq = &rq->rt;
 	int idx;
 
+	if (sched_rt_ratio_exceeded(rq, rt_rq))
+		return NULL;
+
 	idx = sched_find_first_bit(array->bitmap);
 	if (idx >= MAX_RT_PRIO)
 		return NULL;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 96f31c1bc4f0..3afbd25f43eb 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -306,7 +306,23 @@ static struct ctl_table kern_table[] = {
 		.procname	= "sched_nr_migrate",
 		.data		= &sysctl_sched_nr_migrate,
 		.maxlen		= sizeof(unsigned int),
-		.mode		= 644,
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "sched_rt_period_ms",
+		.data		= &sysctl_sched_rt_period,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "sched_rt_ratio",
+		.data		= &sysctl_sched_rt_ratio,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
 #if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
-- 
cgit v1.2.3


From 6f505b16425a51270058e4a93441fe64de3dd435 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:30 +0100
Subject: sched: rt group scheduling

Extend group scheduling to also cover the realtime classes. It uses the time
limiting introduced by the previous patch to allow multiple realtime groups.

The hard time limit is required to keep behaviour deterministic.

The algorithms used make the realtime scheduler O(tg), linear scaling wrt the
number of task groups. This is the worst case behaviour I can't seem to get out
of, the avg. case of the algorithms can be improved, I focused on correctness
and worst case.

[ akpm@linux-foundation.org: move side-effects out of BUG_ON(). ]

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/init_task.h |   5 +-
 include/linux/sched.h     |  10 +-
 kernel/fork.c             |   2 +-
 kernel/sched.c            | 283 +++++++++++++++++++---------
 kernel/sched_rt.c         | 455 ++++++++++++++++++++++++++++++++++------------
 5 files changed, 549 insertions(+), 206 deletions(-)

(limited to 'include')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index ee65d87bedb7..796019b22b6f 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -130,12 +130,13 @@ extern struct group_info init_groups;
 	.normal_prio	= MAX_PRIO-20,					\
 	.policy		= SCHED_NORMAL,					\
 	.cpus_allowed	= CPU_MASK_ALL,					\
-	.nr_cpus_allowed = NR_CPUS,					\
 	.mm		= NULL,						\
 	.active_mm	= &init_mm,					\
 	.rt		= {						\
 		.run_list	= LIST_HEAD_INIT(tsk.rt.run_list),	\
-		.time_slice	= HZ, },				\
+		.time_slice	= HZ, 					\
+		.nr_cpus_allowed = NR_CPUS,				\
+	},								\
 	.ioprio		= 0,						\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
 	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index d5ea144df836..04eecbf0241e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -934,6 +934,15 @@ struct sched_rt_entity {
 	struct list_head run_list;
 	unsigned int time_slice;
 	unsigned long timeout;
+	int nr_cpus_allowed;
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct sched_rt_entity	*parent;
+	/* rq on which this entity is (to be) queued: */
+	struct rt_rq		*rt_rq;
+	/* rq "owned" by this entity/group: */
+	struct rt_rq		*my_q;
+#endif
 };
 
 struct task_struct {
@@ -978,7 +987,6 @@ struct task_struct {
 
 	unsigned int policy;
 	cpumask_t cpus_allowed;
-	int nr_cpus_allowed;
 
 #ifdef CONFIG_PREEMPT_RCU
 	int rcu_read_lock_nesting;
diff --git a/kernel/fork.c b/kernel/fork.c
index 9f8ef32cbc7a..0c969f4fade0 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1246,7 +1246,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 * parent's CPU). This avoids alot of nasty races.
 	 */
 	p->cpus_allowed = current->cpus_allowed;
-	p->nr_cpus_allowed = current->nr_cpus_allowed;
+	p->rt.nr_cpus_allowed = current->rt.nr_cpus_allowed;
 	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) ||
 			!cpu_online(task_cpu(p))))
 		set_task_cpu(p, smp_processor_id());
diff --git a/kernel/sched.c b/kernel/sched.c
index e9a7beee9b79..5ea2c533b432 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -161,6 +161,8 @@ struct rt_prio_array {
 
 struct cfs_rq;
 
+static LIST_HEAD(task_groups);
+
 /* task group related information */
 struct task_group {
 #ifdef CONFIG_FAIR_CGROUP_SCHED
@@ -171,6 +173,11 @@ struct task_group {
 	/* runqueue "owned" by this group on each cpu */
 	struct cfs_rq **cfs_rq;
 
+	struct sched_rt_entity **rt_se;
+	struct rt_rq **rt_rq;
+
+	unsigned int rt_ratio;
+
 	/*
 	 * shares assigned to a task group governs how much of cpu bandwidth
 	 * is allocated to the group. The more shares a group has, the more is
@@ -208,6 +215,7 @@ struct task_group {
 	unsigned long shares;
 
 	struct rcu_head rcu;
+	struct list_head list;
 };
 
 /* Default task group's sched entity on each cpu */
@@ -215,9 +223,15 @@ static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
 /* Default task group's cfs_rq on each cpu */
 static DEFINE_PER_CPU(struct cfs_rq, init_cfs_rq) ____cacheline_aligned_in_smp;
 
+static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity);
+static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp;
+
 static struct sched_entity *init_sched_entity_p[NR_CPUS];
 static struct cfs_rq *init_cfs_rq_p[NR_CPUS];
 
+static struct sched_rt_entity *init_sched_rt_entity_p[NR_CPUS];
+static struct rt_rq *init_rt_rq_p[NR_CPUS];
+
 /* task_group_mutex serializes add/remove of task groups and also changes to
  * a task group's cpu shares.
  */
@@ -240,6 +254,9 @@ static void set_se_shares(struct sched_entity *se, unsigned long shares);
 struct task_group init_task_group = {
 	.se	= init_sched_entity_p,
 	.cfs_rq = init_cfs_rq_p,
+
+	.rt_se	= init_sched_rt_entity_p,
+	.rt_rq	= init_rt_rq_p,
 };
 
 #ifdef CONFIG_FAIR_USER_SCHED
@@ -269,10 +286,13 @@ static inline struct task_group *task_group(struct task_struct *p)
 }
 
 /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu)
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 {
 	p->se.cfs_rq = task_group(p)->cfs_rq[cpu];
 	p->se.parent = task_group(p)->se[cpu];
+
+	p->rt.rt_rq  = task_group(p)->rt_rq[cpu];
+	p->rt.parent = task_group(p)->rt_se[cpu];
 }
 
 static inline void lock_task_group_list(void)
@@ -297,7 +317,7 @@ static inline void unlock_doms_cur(void)
 
 #else
 
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu) { }
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { }
 static inline void lock_task_group_list(void) { }
 static inline void unlock_task_group_list(void) { }
 static inline void lock_doms_cur(void) { }
@@ -343,13 +363,22 @@ struct cfs_rq {
 struct rt_rq {
 	struct rt_prio_array active;
 	unsigned long rt_nr_running;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	int highest_prio; /* highest queued rt task prio */
+#endif
 #ifdef CONFIG_SMP
 	unsigned long rt_nr_migratory;
-	int highest_prio; /* highest queued rt task prio */
 	int overloaded;
 #endif
+	int rt_throttled;
 	u64 rt_time;
-	u64 rt_throttled;
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct rq *rq;
+	struct list_head leaf_rt_rq_list;
+	struct task_group *tg;
+	struct sched_rt_entity *rt_se;
+#endif
 };
 
 #ifdef CONFIG_SMP
@@ -411,12 +440,14 @@ struct rq {
 	u64 nr_switches;
 
 	struct cfs_rq cfs;
+	struct rt_rq rt;
+	u64 rt_period_expire;
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* list of leaf cfs_rq on this cpu: */
 	struct list_head leaf_cfs_rq_list;
+	struct list_head leaf_rt_rq_list;
 #endif
-	struct rt_rq rt;
-	u64 rt_period_expire;
 
 	/*
 	 * This is part of a global counter where only the total sum
@@ -613,9 +644,9 @@ const_debug unsigned int sysctl_sched_rt_period = 1000;
 
 /*
  * ratio of time -rt tasks may consume.
- * default: 100%
+ * default: 95%
  */
-const_debug unsigned int sysctl_sched_rt_ratio = SCHED_RT_FRAC;
+const_debug unsigned int sysctl_sched_rt_ratio = 62259;
 
 /*
  * For kernel-internal use: high-speed (but slightly incorrect) per-cpu
@@ -1337,7 +1368,7 @@ unsigned long weighted_cpuload(const int cpu)
 
 static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 {
-	set_task_cfs_rq(p, cpu);
+	set_task_rq(p, cpu);
 #ifdef CONFIG_SMP
 	/*
 	 * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be
@@ -5281,7 +5312,7 @@ int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
 		p->sched_class->set_cpus_allowed(p, &new_mask);
 	else {
 		p->cpus_allowed = new_mask;
-		p->nr_cpus_allowed = cpus_weight(new_mask);
+		p->rt.nr_cpus_allowed = cpus_weight(new_mask);
 	}
 
 	/* Can the task run on the task's current CPU? If so, we're done */
@@ -7079,8 +7110,50 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq)
 
 	rt_rq->rt_time = 0;
 	rt_rq->rt_throttled = 0;
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	rt_rq->rq = rq;
+#endif
 }
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static void init_tg_cfs_entry(struct rq *rq, struct task_group *tg,
+		struct cfs_rq *cfs_rq, struct sched_entity *se,
+		int cpu, int add)
+{
+	tg->cfs_rq[cpu] = cfs_rq;
+	init_cfs_rq(cfs_rq, rq);
+	cfs_rq->tg = tg;
+	if (add)
+		list_add(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+
+	tg->se[cpu] = se;
+	se->cfs_rq = &rq->cfs;
+	se->my_q = cfs_rq;
+	se->load.weight = tg->shares;
+	se->load.inv_weight = div64_64(1ULL<<32, se->load.weight);
+	se->parent = NULL;
+}
+
+static void init_tg_rt_entry(struct rq *rq, struct task_group *tg,
+		struct rt_rq *rt_rq, struct sched_rt_entity *rt_se,
+		int cpu, int add)
+{
+	tg->rt_rq[cpu] = rt_rq;
+	init_rt_rq(rt_rq, rq);
+	rt_rq->tg = tg;
+	rt_rq->rt_se = rt_se;
+	if (add)
+		list_add(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list);
+
+	tg->rt_se[cpu] = rt_se;
+	rt_se->rt_rq = &rq->rt;
+	rt_se->my_q = rt_rq;
+	rt_se->parent = NULL;
+	INIT_LIST_HEAD(&rt_se->run_list);
+}
+#endif
+
 void __init sched_init(void)
 {
 	int highest_cpu = 0;
@@ -7090,6 +7163,10 @@ void __init sched_init(void)
 	init_defrootdomain();
 #endif
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	list_add(&init_task_group.list, &task_groups);
+#endif
+
 	for_each_possible_cpu(i) {
 		struct rq *rq;
 
@@ -7099,30 +7176,20 @@ void __init sched_init(void)
 		rq->nr_running = 0;
 		rq->clock = 1;
 		init_cfs_rq(&rq->cfs, rq);
+		init_rt_rq(&rq->rt, rq);
 #ifdef CONFIG_FAIR_GROUP_SCHED
-		INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
-		{
-			struct cfs_rq *cfs_rq = &per_cpu(init_cfs_rq, i);
-			struct sched_entity *se =
-					 &per_cpu(init_sched_entity, i);
-
-			init_cfs_rq_p[i] = cfs_rq;
-			init_cfs_rq(cfs_rq, rq);
-			cfs_rq->tg = &init_task_group;
-			list_add(&cfs_rq->leaf_cfs_rq_list,
-							 &rq->leaf_cfs_rq_list);
-
-			init_sched_entity_p[i] = se;
-			se->cfs_rq = &rq->cfs;
-			se->my_q = cfs_rq;
-			se->load.weight = init_task_group_load;
-			se->load.inv_weight =
-				 div64_64(1ULL<<32, init_task_group_load);
-			se->parent = NULL;
-		}
 		init_task_group.shares = init_task_group_load;
+		INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
+		init_tg_cfs_entry(rq, &init_task_group,
+				&per_cpu(init_cfs_rq, i),
+				&per_cpu(init_sched_entity, i), i, 1);
+
+		init_task_group.rt_ratio = sysctl_sched_rt_ratio; /* XXX */
+		INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
+		init_tg_rt_entry(rq, &init_task_group,
+				&per_cpu(init_rt_rq, i),
+				&per_cpu(init_sched_rt_entity, i), i, 1);
 #endif
-		init_rt_rq(&rq->rt, rq);
 		rq->rt_period_expire = 0;
 
 		for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
@@ -7460,12 +7527,36 @@ static int load_balance_monitor(void *unused)
 }
 #endif	/* CONFIG_SMP */
 
+static void free_sched_group(struct task_group *tg)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (tg->cfs_rq)
+			kfree(tg->cfs_rq[i]);
+		if (tg->se)
+			kfree(tg->se[i]);
+		if (tg->rt_rq)
+			kfree(tg->rt_rq[i]);
+		if (tg->rt_se)
+			kfree(tg->rt_se[i]);
+	}
+
+	kfree(tg->cfs_rq);
+	kfree(tg->se);
+	kfree(tg->rt_rq);
+	kfree(tg->rt_se);
+	kfree(tg);
+}
+
 /* allocate runqueue etc for a new task group */
 struct task_group *sched_create_group(void)
 {
 	struct task_group *tg;
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se;
+	struct rt_rq *rt_rq;
+	struct sched_rt_entity *rt_se;
 	struct rq *rq;
 	int i;
 
@@ -7479,100 +7570,89 @@ struct task_group *sched_create_group(void)
 	tg->se = kzalloc(sizeof(se) * NR_CPUS, GFP_KERNEL);
 	if (!tg->se)
 		goto err;
+	tg->rt_rq = kzalloc(sizeof(rt_rq) * NR_CPUS, GFP_KERNEL);
+	if (!tg->rt_rq)
+		goto err;
+	tg->rt_se = kzalloc(sizeof(rt_se) * NR_CPUS, GFP_KERNEL);
+	if (!tg->rt_se)
+		goto err;
+
+	tg->shares = NICE_0_LOAD;
+	tg->rt_ratio = 0; /* XXX */
 
 	for_each_possible_cpu(i) {
 		rq = cpu_rq(i);
 
-		cfs_rq = kmalloc_node(sizeof(struct cfs_rq), GFP_KERNEL,
-							 cpu_to_node(i));
+		cfs_rq = kmalloc_node(sizeof(struct cfs_rq),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
 		if (!cfs_rq)
 			goto err;
 
-		se = kmalloc_node(sizeof(struct sched_entity), GFP_KERNEL,
-							cpu_to_node(i));
+		se = kmalloc_node(sizeof(struct sched_entity),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
 		if (!se)
 			goto err;
 
-		memset(cfs_rq, 0, sizeof(struct cfs_rq));
-		memset(se, 0, sizeof(struct sched_entity));
+		rt_rq = kmalloc_node(sizeof(struct rt_rq),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+		if (!rt_rq)
+			goto err;
 
-		tg->cfs_rq[i] = cfs_rq;
-		init_cfs_rq(cfs_rq, rq);
-		cfs_rq->tg = tg;
+		rt_se = kmalloc_node(sizeof(struct sched_rt_entity),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+		if (!rt_se)
+			goto err;
 
-		tg->se[i] = se;
-		se->cfs_rq = &rq->cfs;
-		se->my_q = cfs_rq;
-		se->load.weight = NICE_0_LOAD;
-		se->load.inv_weight = div64_64(1ULL<<32, NICE_0_LOAD);
-		se->parent = NULL;
+		init_tg_cfs_entry(rq, tg, cfs_rq, se, i, 0);
+		init_tg_rt_entry(rq, tg, rt_rq, rt_se, i, 0);
 	}
 
-	tg->shares = NICE_0_LOAD;
-
 	lock_task_group_list();
 	for_each_possible_cpu(i) {
 		rq = cpu_rq(i);
 		cfs_rq = tg->cfs_rq[i];
 		list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+		rt_rq = tg->rt_rq[i];
+		list_add_rcu(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list);
 	}
+	list_add_rcu(&tg->list, &task_groups);
 	unlock_task_group_list();
 
 	return tg;
 
 err:
-	for_each_possible_cpu(i) {
-		if (tg->cfs_rq)
-			kfree(tg->cfs_rq[i]);
-		if (tg->se)
-			kfree(tg->se[i]);
-	}
-	kfree(tg->cfs_rq);
-	kfree(tg->se);
-	kfree(tg);
-
+	free_sched_group(tg);
 	return ERR_PTR(-ENOMEM);
 }
 
 /* rcu callback to free various structures associated with a task group */
-static void free_sched_group(struct rcu_head *rhp)
+static void free_sched_group_rcu(struct rcu_head *rhp)
 {
-	struct task_group *tg = container_of(rhp, struct task_group, rcu);
-	struct cfs_rq *cfs_rq;
-	struct sched_entity *se;
-	int i;
-
 	/* now it should be safe to free those cfs_rqs */
-	for_each_possible_cpu(i) {
-		cfs_rq = tg->cfs_rq[i];
-		kfree(cfs_rq);
-
-		se = tg->se[i];
-		kfree(se);
-	}
-
-	kfree(tg->cfs_rq);
-	kfree(tg->se);
-	kfree(tg);
+	free_sched_group(container_of(rhp, struct task_group, rcu));
 }
 
 /* Destroy runqueue etc associated with a task group */
 void sched_destroy_group(struct task_group *tg)
 {
 	struct cfs_rq *cfs_rq = NULL;
+	struct rt_rq *rt_rq = NULL;
 	int i;
 
 	lock_task_group_list();
 	for_each_possible_cpu(i) {
 		cfs_rq = tg->cfs_rq[i];
 		list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+		rt_rq = tg->rt_rq[i];
+		list_del_rcu(&rt_rq->leaf_rt_rq_list);
 	}
+	list_del_rcu(&tg->list);
 	unlock_task_group_list();
 
 	BUG_ON(!cfs_rq);
 
 	/* wait for possible concurrent references to cfs_rqs complete */
-	call_rcu(&tg->rcu, free_sched_group);
+	call_rcu(&tg->rcu, free_sched_group_rcu);
 }
 
 /* change task's runqueue when it moves between groups.
@@ -7588,11 +7668,6 @@ void sched_move_task(struct task_struct *tsk)
 
 	rq = task_rq_lock(tsk, &flags);
 
-	if (tsk->sched_class != &fair_sched_class) {
-		set_task_cfs_rq(tsk, task_cpu(tsk));
-		goto done;
-	}
-
 	update_rq_clock(rq);
 
 	running = task_current(rq, tsk);
@@ -7604,7 +7679,7 @@ void sched_move_task(struct task_struct *tsk)
 			tsk->sched_class->put_prev_task(rq, tsk);
 	}
 
-	set_task_cfs_rq(tsk, task_cpu(tsk));
+	set_task_rq(tsk, task_cpu(tsk));
 
 	if (on_rq) {
 		if (unlikely(running))
@@ -7612,7 +7687,6 @@ void sched_move_task(struct task_struct *tsk)
 		enqueue_task(rq, tsk, 0);
 	}
 
-done:
 	task_rq_unlock(rq, &flags);
 }
 
@@ -7697,6 +7771,31 @@ unsigned long sched_group_shares(struct task_group *tg)
 	return tg->shares;
 }
 
+/*
+ * Ensure the total rt_ratio <= sysctl_sched_rt_ratio
+ */
+int sched_group_set_rt_ratio(struct task_group *tg, unsigned long rt_ratio)
+{
+	struct task_group *tgi;
+	unsigned long total = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(tgi, &task_groups, list)
+		total += tgi->rt_ratio;
+	rcu_read_unlock();
+
+	if (total + rt_ratio - tg->rt_ratio > sysctl_sched_rt_ratio)
+		return -EINVAL;
+
+	tg->rt_ratio = rt_ratio;
+	return 0;
+}
+
+unsigned long sched_group_rt_ratio(struct task_group *tg)
+{
+	return tg->rt_ratio;
+}
+
 #endif	/* CONFIG_FAIR_GROUP_SCHED */
 
 #ifdef CONFIG_FAIR_CGROUP_SCHED
@@ -7772,12 +7871,30 @@ static u64 cpu_shares_read_uint(struct cgroup *cgrp, struct cftype *cft)
 	return (u64) tg->shares;
 }
 
+static int cpu_rt_ratio_write_uint(struct cgroup *cgrp, struct cftype *cftype,
+		u64 rt_ratio_val)
+{
+	return sched_group_set_rt_ratio(cgroup_tg(cgrp), rt_ratio_val);
+}
+
+static u64 cpu_rt_ratio_read_uint(struct cgroup *cgrp, struct cftype *cft)
+{
+	struct task_group *tg = cgroup_tg(cgrp);
+
+	return (u64) tg->rt_ratio;
+}
+
 static struct cftype cpu_files[] = {
 	{
 		.name = "shares",
 		.read_uint = cpu_shares_read_uint,
 		.write_uint = cpu_shares_write_uint,
 	},
+	{
+		.name = "rt_ratio",
+		.read_uint = cpu_rt_ratio_read_uint,
+		.write_uint = cpu_rt_ratio_write_uint,
+	},
 };
 
 static int cpu_cgroup_populate(struct cgroup_subsys *ss, struct cgroup *cont)
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index fd10d965aa06..1178257613ad 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -45,47 +45,167 @@ static void update_rt_migration(struct rq *rq)
 }
 #endif /* CONFIG_SMP */
 
-static int sched_rt_ratio_exceeded(struct rq *rq, struct rt_rq *rt_rq)
+static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se)
 {
+	return container_of(rt_se, struct task_struct, rt);
+}
+
+static inline int on_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return !list_empty(&rt_se->run_list);
+}
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+
+static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+{
+	if (!rt_rq->tg)
+		return SCHED_RT_FRAC;
+
+	return rt_rq->tg->rt_ratio;
+}
+
+#define for_each_leaf_rt_rq(rt_rq, rq) \
+	list_for_each_entry(rt_rq, &rq->leaf_rt_rq_list, leaf_rt_rq_list)
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return rt_rq->rq;
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	return rt_se->rt_rq;
+}
+
+#define for_each_sched_rt_entity(rt_se) \
+	for (; rt_se; rt_se = rt_se->parent)
+
+static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return rt_se->my_q;
+}
+
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se);
+static void dequeue_rt_entity(struct sched_rt_entity *rt_se);
+
+static void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+{
+	struct sched_rt_entity *rt_se = rt_rq->rt_se;
+
+	if (rt_se && !on_rt_rq(rt_se) && rt_rq->rt_nr_running) {
+		enqueue_rt_entity(rt_se);
+		resched_task(rq_of_rt_rq(rt_rq)->curr);
+	}
+}
+
+static void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+{
+	struct sched_rt_entity *rt_se = rt_rq->rt_se;
+
+	if (rt_se && on_rt_rq(rt_se))
+		dequeue_rt_entity(rt_se);
+}
+
+#else
+
+static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+{
+	return sysctl_sched_rt_ratio;
+}
+
+#define for_each_leaf_rt_rq(rt_rq, rq) \
+	for (rt_rq = &rq->rt; rt_rq; rt_rq = NULL)
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return container_of(rt_rq, struct rq, rt);
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	struct task_struct *p = rt_task_of(rt_se);
+	struct rq *rq = task_rq(p);
+
+	return &rq->rt;
+}
+
+#define for_each_sched_rt_entity(rt_se) \
+	for (; rt_se; rt_se = NULL)
+
+static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return NULL;
+}
+
+static inline void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+{
+}
+
+static inline void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+{
+}
+
+#endif
+
+static inline int rt_se_prio(struct sched_rt_entity *rt_se)
+{
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct rt_rq *rt_rq = group_rt_rq(rt_se);
+
+	if (rt_rq)
+		return rt_rq->highest_prio;
+#endif
+
+	return rt_task_of(rt_se)->prio;
+}
+
+static int sched_rt_ratio_exceeded(struct rt_rq *rt_rq)
+{
+	unsigned int rt_ratio = sched_rt_ratio(rt_rq);
 	u64 period, ratio;
 
-	if (sysctl_sched_rt_ratio == SCHED_RT_FRAC)
+	if (rt_ratio == SCHED_RT_FRAC)
 		return 0;
 
 	if (rt_rq->rt_throttled)
 		return 1;
 
 	period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
-	ratio = (period * sysctl_sched_rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+	ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
 
 	if (rt_rq->rt_time > ratio) {
-		rt_rq->rt_throttled = rq->clock + period - rt_rq->rt_time;
+		rt_rq->rt_throttled = 1;
+		sched_rt_ratio_dequeue(rt_rq);
 		return 1;
 	}
 
 	return 0;
 }
 
+static void __update_sched_rt_period(struct rt_rq *rt_rq, u64 period)
+{
+	unsigned long rt_ratio = sched_rt_ratio(rt_rq);
+	u64 ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+	rt_rq->rt_time -= min(rt_rq->rt_time, ratio);
+	if (rt_rq->rt_throttled) {
+		rt_rq->rt_throttled = 0;
+		sched_rt_ratio_enqueue(rt_rq);
+	}
+}
+
 static void update_sched_rt_period(struct rq *rq)
 {
-	while (rq->clock > rq->rt_period_expire) {
-		u64 period, ratio;
+	struct rt_rq *rt_rq;
+	u64 period;
 
+	while (rq->clock > rq->rt_period_expire) {
 		period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
-		ratio = (period * sysctl_sched_rt_ratio) >> SCHED_RT_FRAC_SHIFT;
-
-		rq->rt.rt_time -= min(rq->rt.rt_time, ratio);
 		rq->rt_period_expire += period;
-	}
 
-	/*
-	 * When the rt throttle is expired, let them rip.
-	 * (XXX: use hrtick when available)
-	 */
-	if (rq->rt.rt_throttled && rq->clock > rq->rt.rt_throttled) {
-		rq->rt.rt_throttled = 0;
-		if (!sched_rt_ratio_exceeded(rq, &rq->rt))
-			resched_task(rq->curr);
+		for_each_leaf_rt_rq(rt_rq, rq)
+			__update_sched_rt_period(rt_rq, period);
 	}
 }
 
@@ -96,6 +216,8 @@ static void update_sched_rt_period(struct rq *rq)
 static void update_curr_rt(struct rq *rq)
 {
 	struct task_struct *curr = rq->curr;
+	struct sched_rt_entity *rt_se = &curr->rt;
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	u64 delta_exec;
 
 	if (!task_has_rt_policy(curr))
@@ -111,95 +233,184 @@ static void update_curr_rt(struct rq *rq)
 	curr->se.exec_start = rq->clock;
 	cpuacct_charge(curr, delta_exec);
 
-	rq->rt.rt_time += delta_exec;
-	update_sched_rt_period(rq);
-	if (sched_rt_ratio_exceeded(rq, &rq->rt))
+	rt_rq->rt_time += delta_exec;
+	/*
+	 * might make it a tad more accurate:
+	 *
+	 * update_sched_rt_period(rq);
+	 */
+	if (sched_rt_ratio_exceeded(rt_rq))
 		resched_task(curr);
 }
 
-static inline void inc_rt_tasks(struct task_struct *p, struct rq *rq)
+static inline
+void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
-	WARN_ON(!rt_task(p));
-	rq->rt.rt_nr_running++;
+	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
+	rt_rq->rt_nr_running++;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	if (rt_se_prio(rt_se) < rt_rq->highest_prio)
+		rt_rq->highest_prio = rt_se_prio(rt_se);
+#endif
 #ifdef CONFIG_SMP
-	if (p->prio < rq->rt.highest_prio)
-		rq->rt.highest_prio = p->prio;
-	if (p->nr_cpus_allowed > 1)
+	if (rt_se->nr_cpus_allowed > 1) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
 		rq->rt.rt_nr_migratory++;
+	}
 
-	update_rt_migration(rq);
-#endif /* CONFIG_SMP */
+	update_rt_migration(rq_of_rt_rq(rt_rq));
+#endif
 }
 
-static inline void dec_rt_tasks(struct task_struct *p, struct rq *rq)
+static inline
+void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
-	WARN_ON(!rt_task(p));
-	WARN_ON(!rq->rt.rt_nr_running);
-	rq->rt.rt_nr_running--;
-#ifdef CONFIG_SMP
-	if (rq->rt.rt_nr_running) {
+	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
+	WARN_ON(!rt_rq->rt_nr_running);
+	rt_rq->rt_nr_running--;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	if (rt_rq->rt_nr_running) {
 		struct rt_prio_array *array;
 
-		WARN_ON(p->prio < rq->rt.highest_prio);
-		if (p->prio == rq->rt.highest_prio) {
+		WARN_ON(rt_se_prio(rt_se) < rt_rq->highest_prio);
+		if (rt_se_prio(rt_se) == rt_rq->highest_prio) {
 			/* recalculate */
-			array = &rq->rt.active;
-			rq->rt.highest_prio =
+			array = &rt_rq->active;
+			rt_rq->highest_prio =
 				sched_find_first_bit(array->bitmap);
 		} /* otherwise leave rq->highest prio alone */
 	} else
-		rq->rt.highest_prio = MAX_RT_PRIO;
-	if (p->nr_cpus_allowed > 1)
+		rt_rq->highest_prio = MAX_RT_PRIO;
+#endif
+#ifdef CONFIG_SMP
+	if (rt_se->nr_cpus_allowed > 1) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
 		rq->rt.rt_nr_migratory--;
+	}
 
-	update_rt_migration(rq);
+	update_rt_migration(rq_of_rt_rq(rt_rq));
 #endif /* CONFIG_SMP */
 }
 
-static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
+	struct rt_prio_array *array = &rt_rq->active;
+	struct rt_rq *group_rq = group_rt_rq(rt_se);
 
-	list_add_tail(&p->rt.run_list, array->queue + p->prio);
-	__set_bit(p->prio, array->bitmap);
-	inc_cpu_load(rq, p->se.load.weight);
+	if (group_rq && group_rq->rt_throttled)
+		return;
 
-	inc_rt_tasks(p, rq);
+	list_add_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+	__set_bit(rt_se_prio(rt_se), array->bitmap);
 
-	if (wakeup)
-		p->rt.timeout = 0;
+	inc_rt_tasks(rt_se, rt_rq);
+}
+
+static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
+{
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
+	struct rt_prio_array *array = &rt_rq->active;
+
+	list_del_init(&rt_se->run_list);
+	if (list_empty(array->queue + rt_se_prio(rt_se)))
+		__clear_bit(rt_se_prio(rt_se), array->bitmap);
+
+	dec_rt_tasks(rt_se, rt_rq);
+}
+
+/*
+ * Because the prio of an upper entry depends on the lower
+ * entries, we must remove entries top - down.
+ *
+ * XXX: O(1/2 h^2) because we can only walk up, not down the chain.
+ *      doesn't matter much for now, as h=2 for GROUP_SCHED.
+ */
+static void dequeue_rt_stack(struct task_struct *p)
+{
+	struct sched_rt_entity *rt_se, *top_se;
+
+	/*
+	 * dequeue all, top - down.
+	 */
+	do {
+		rt_se = &p->rt;
+		top_se = NULL;
+		for_each_sched_rt_entity(rt_se) {
+			if (on_rt_rq(rt_se))
+				top_se = rt_se;
+		}
+		if (top_se)
+			dequeue_rt_entity(top_se);
+	} while (top_se);
 }
 
 /*
  * Adding/removing a task to/from a priority array:
  */
+static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+{
+	struct sched_rt_entity *rt_se = &p->rt;
+
+	if (wakeup)
+		rt_se->timeout = 0;
+
+	dequeue_rt_stack(p);
+
+	/*
+	 * enqueue everybody, bottom - up.
+	 */
+	for_each_sched_rt_entity(rt_se)
+		enqueue_rt_entity(rt_se);
+
+	inc_cpu_load(rq, p->se.load.weight);
+}
+
 static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct sched_rt_entity *rt_se = &p->rt;
+	struct rt_rq *rt_rq;
 
 	update_curr_rt(rq);
 
-	list_del(&p->rt.run_list);
-	if (list_empty(array->queue + p->prio))
-		__clear_bit(p->prio, array->bitmap);
-	dec_cpu_load(rq, p->se.load.weight);
+	dequeue_rt_stack(p);
+
+	/*
+	 * re-enqueue all non-empty rt_rq entities.
+	 */
+	for_each_sched_rt_entity(rt_se) {
+		rt_rq = group_rt_rq(rt_se);
+		if (rt_rq && rt_rq->rt_nr_running)
+			enqueue_rt_entity(rt_se);
+	}
 
-	dec_rt_tasks(p, rq);
+	dec_cpu_load(rq, p->se.load.weight);
 }
 
 /*
  * Put task to the end of the run list without the overhead of dequeue
  * followed by enqueue.
  */
+static
+void requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se)
+{
+	struct rt_prio_array *array = &rt_rq->active;
+
+	list_move_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+}
+
 static void requeue_task_rt(struct rq *rq, struct task_struct *p)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct sched_rt_entity *rt_se = &p->rt;
+	struct rt_rq *rt_rq;
 
-	list_move_tail(&p->rt.run_list, array->queue + p->prio);
+	for_each_sched_rt_entity(rt_se) {
+		rt_rq = rt_rq_of_se(rt_se);
+		requeue_rt_entity(rt_rq, rt_se);
+	}
 }
 
-static void
-yield_task_rt(struct rq *rq)
+static void yield_task_rt(struct rq *rq)
 {
 	requeue_task_rt(rq, rq->curr);
 }
@@ -229,7 +440,7 @@ static int select_task_rq_rt(struct task_struct *p, int sync)
 	 * cold cache anyway.
 	 */
 	if (unlikely(rt_task(rq->curr)) &&
-	    (p->nr_cpus_allowed > 1)) {
+	    (p->rt.nr_cpus_allowed > 1)) {
 		int cpu = find_lowest_rq(p);
 
 		return (cpu == -1) ? task_cpu(p) : cpu;
@@ -252,27 +463,51 @@ static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p)
 		resched_task(rq->curr);
 }
 
-static struct task_struct *pick_next_task_rt(struct rq *rq)
+static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq,
+						   struct rt_rq *rt_rq)
 {
-	struct rt_prio_array *array = &rq->rt.active;
-	struct task_struct *next;
+	struct rt_prio_array *array = &rt_rq->active;
+	struct sched_rt_entity *next = NULL;
 	struct list_head *queue;
-	struct rt_rq *rt_rq = &rq->rt;
 	int idx;
 
-	if (sched_rt_ratio_exceeded(rq, rt_rq))
-		return NULL;
+	if (sched_rt_ratio_exceeded(rt_rq))
+		goto out;
 
 	idx = sched_find_first_bit(array->bitmap);
-	if (idx >= MAX_RT_PRIO)
-		return NULL;
+	BUG_ON(idx >= MAX_RT_PRIO);
 
 	queue = array->queue + idx;
-	next = list_entry(queue->next, struct task_struct, rt.run_list);
+	next = list_entry(queue->next, struct sched_rt_entity, run_list);
+ out:
+	return next;
+}
 
-	next->se.exec_start = rq->clock;
+static struct task_struct *pick_next_task_rt(struct rq *rq)
+{
+	struct sched_rt_entity *rt_se;
+	struct task_struct *p;
+	struct rt_rq *rt_rq;
 
-	return next;
+ retry:
+	rt_rq = &rq->rt;
+
+	if (unlikely(!rt_rq->rt_nr_running))
+		return NULL;
+
+	if (sched_rt_ratio_exceeded(rt_rq))
+		return NULL;
+
+	do {
+		rt_se = pick_next_rt_entity(rq, rt_rq);
+		if (unlikely(!rt_se))
+			goto retry;
+		rt_rq = group_rt_rq(rt_se);
+	} while (rt_rq);
+
+	p = rt_task_of(rt_se);
+	p->se.exec_start = rq->clock;
+	return p;
 }
 
 static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
@@ -282,6 +517,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 }
 
 #ifdef CONFIG_SMP
+
 /* Only try algorithms three times */
 #define RT_MAX_TRIES 3
 
@@ -292,7 +528,7 @@ static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
 	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
-	    (p->nr_cpus_allowed > 1))
+	    (p->rt.nr_cpus_allowed > 1))
 		return 1;
 	return 0;
 }
@@ -300,52 +536,33 @@ static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 /* Return the second highest RT task, NULL otherwise */
 static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 {
-	struct rt_prio_array *array = &rq->rt.active;
-	struct task_struct *next;
-	struct list_head *queue;
+	struct task_struct *next = NULL;
+	struct sched_rt_entity *rt_se;
+	struct rt_prio_array *array;
+	struct rt_rq *rt_rq;
 	int idx;
 
-	if (likely(rq->rt.rt_nr_running < 2))
-		return NULL;
-
-	idx = sched_find_first_bit(array->bitmap);
-	if (unlikely(idx >= MAX_RT_PRIO)) {
-		WARN_ON(1); /* rt_nr_running is bad */
-		return NULL;
-	}
-
-	queue = array->queue + idx;
-	BUG_ON(list_empty(queue));
-
-	next = list_entry(queue->next, struct task_struct, rt.run_list);
-	if (unlikely(pick_rt_task(rq, next, cpu)))
-		goto out;
-
-	if (queue->next->next != queue) {
-		/* same prio task */
-		next = list_entry(queue->next->next, struct task_struct,
-				  rt.run_list);
-		if (pick_rt_task(rq, next, cpu))
-			goto out;
-	}
-
- retry:
-	/* slower, but more flexible */
-	idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx+1);
-	if (unlikely(idx >= MAX_RT_PRIO))
-		return NULL;
-
-	queue = array->queue + idx;
-	BUG_ON(list_empty(queue));
-
-	list_for_each_entry(next, queue, rt.run_list) {
-		if (pick_rt_task(rq, next, cpu))
-			goto out;
+	for_each_leaf_rt_rq(rt_rq, rq) {
+		array = &rt_rq->active;
+		idx = sched_find_first_bit(array->bitmap);
+ next_idx:
+		if (idx >= MAX_RT_PRIO)
+			continue;
+		if (next && next->prio < idx)
+			continue;
+		list_for_each_entry(rt_se, array->queue + idx, run_list) {
+			struct task_struct *p = rt_task_of(rt_se);
+			if (pick_rt_task(rq, p, cpu)) {
+				next = p;
+				break;
+			}
+		}
+		if (!next) {
+			idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx+1);
+			goto next_idx;
+		}
 	}
 
-	goto retry;
-
- out:
 	return next;
 }
 
@@ -774,12 +991,12 @@ static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
 	 * Update the migration status of the RQ if we have an RT task
 	 * which is running AND changing its weight value.
 	 */
-	if (p->se.on_rq && (weight != p->nr_cpus_allowed)) {
+	if (p->se.on_rq && (weight != p->rt.nr_cpus_allowed)) {
 		struct rq *rq = task_rq(p);
 
-		if ((p->nr_cpus_allowed <= 1) && (weight > 1)) {
+		if ((p->rt.nr_cpus_allowed <= 1) && (weight > 1)) {
 			rq->rt.rt_nr_migratory++;
-		} else if ((p->nr_cpus_allowed > 1) && (weight <= 1)) {
+		} else if ((p->rt.nr_cpus_allowed > 1) && (weight <= 1)) {
 			BUG_ON(!rq->rt.rt_nr_migratory);
 			rq->rt.rt_nr_migratory--;
 		}
@@ -788,7 +1005,7 @@ static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
 	}
 
 	p->cpus_allowed    = *new_mask;
-	p->nr_cpus_allowed = weight;
+	p->rt.nr_cpus_allowed = weight;
 }
 
 /* Assumes rq->lock is held */
-- 
cgit v1.2.3


From 48d5e258216f1c7713633439beb98a38c7290649 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:31 +0100
Subject: sched: rt throttling vs no_hz

We need to teach no_hz about the rt throttling because its tick driven.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    |  2 ++
 kernel/sched.c           | 23 ++++++++++++++++++++++-
 kernel/sched_rt.c        | 30 ++++++++++++++++--------------
 kernel/time/tick-sched.c |  5 +++++
 4 files changed, 45 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 04eecbf0241e..acadcab89ef9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -230,6 +230,8 @@ static inline int select_nohz_load_balancer(int cpu)
 }
 #endif
 
+extern unsigned long rt_needs_cpu(int cpu);
+
 /*
  * Only dump TASK_* tasks. (0 for all tasks)
  */
diff --git a/kernel/sched.c b/kernel/sched.c
index 5ea2c533b432..22712b2e058a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -442,6 +442,7 @@ struct rq {
 	struct cfs_rq cfs;
 	struct rt_rq rt;
 	u64 rt_period_expire;
+	int rt_throttled;
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* list of leaf cfs_rq on this cpu: */
@@ -594,6 +595,23 @@ static void update_rq_clock(struct rq *rq)
 #define task_rq(p)		cpu_rq(task_cpu(p))
 #define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
 
+unsigned long rt_needs_cpu(int cpu)
+{
+	struct rq *rq = cpu_rq(cpu);
+	u64 delta;
+
+	if (!rq->rt_throttled)
+		return 0;
+
+	if (rq->clock > rq->rt_period_expire)
+		return 1;
+
+	delta = rq->rt_period_expire - rq->clock;
+	do_div(delta, NSEC_PER_SEC / HZ);
+
+	return (unsigned long)delta;
+}
+
 /*
  * Tunables that become constants when CONFIG_SCHED_DEBUG is off:
  */
@@ -7102,9 +7120,11 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq)
 	/* delimiter for bitsearch: */
 	__set_bit(MAX_RT_PRIO, array->bitmap);
 
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	rt_rq->highest_prio = MAX_RT_PRIO;
+#endif
 #ifdef CONFIG_SMP
 	rt_rq->rt_nr_migratory = 0;
-	rt_rq->highest_prio = MAX_RT_PRIO;
 	rt_rq->overloaded = 0;
 #endif
 
@@ -7191,6 +7211,7 @@ void __init sched_init(void)
 				&per_cpu(init_sched_rt_entity, i), i, 1);
 #endif
 		rq->rt_period_expire = 0;
+		rq->rt_throttled = 0;
 
 		for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
 			rq->cpu_load[j] = 0;
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 1144bf55669d..8bfdb3f8a52d 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -175,7 +175,11 @@ static int sched_rt_ratio_exceeded(struct rt_rq *rt_rq)
 	ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
 
 	if (rt_rq->rt_time > ratio) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+
+		rq->rt_throttled = 1;
 		rt_rq->rt_throttled = 1;
+
 		sched_rt_ratio_dequeue(rt_rq);
 		return 1;
 	}
@@ -183,18 +187,6 @@ static int sched_rt_ratio_exceeded(struct rt_rq *rt_rq)
 	return 0;
 }
 
-static void __update_sched_rt_period(struct rt_rq *rt_rq, u64 period)
-{
-	unsigned long rt_ratio = sched_rt_ratio(rt_rq);
-	u64 ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
-
-	rt_rq->rt_time -= min(rt_rq->rt_time, ratio);
-	if (rt_rq->rt_throttled) {
-		rt_rq->rt_throttled = 0;
-		sched_rt_ratio_enqueue(rt_rq);
-	}
-}
-
 static void update_sched_rt_period(struct rq *rq)
 {
 	struct rt_rq *rt_rq;
@@ -204,8 +196,18 @@ static void update_sched_rt_period(struct rq *rq)
 		period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
 		rq->rt_period_expire += period;
 
-		for_each_leaf_rt_rq(rt_rq, rq)
-			__update_sched_rt_period(rt_rq, period);
+		for_each_leaf_rt_rq(rt_rq, rq) {
+			unsigned long rt_ratio = sched_rt_ratio(rt_rq);
+			u64 ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+			rt_rq->rt_time -= min(rt_rq->rt_time, ratio);
+			if (rt_rq->rt_throttled) {
+				rt_rq->rt_throttled = 0;
+				sched_rt_ratio_enqueue(rt_rq);
+			}
+		}
+
+		rq->rt_throttled = 0;
 	}
 }
 
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index cb89fa8db110..5f9fb645b725 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -153,6 +153,7 @@ void tick_nohz_update_jiffies(void)
 void tick_nohz_stop_sched_tick(void)
 {
 	unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
+	unsigned long rt_jiffies;
 	struct tick_sched *ts;
 	ktime_t last_update, expires, now, delta;
 	struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
@@ -216,6 +217,10 @@ void tick_nohz_stop_sched_tick(void)
 	next_jiffies = get_next_timer_interrupt(last_jiffies);
 	delta_jiffies = next_jiffies - last_jiffies;
 
+	rt_jiffies = rt_needs_cpu(cpu);
+	if (rt_jiffies && rt_jiffies < delta_jiffies)
+		delta_jiffies = rt_jiffies;
+
 	if (rcu_needs_cpu(cpu))
 		delta_jiffies = 1;
 	/*
-- 
cgit v1.2.3


From d3d74453c34f8fd87674a8cf5b8a327c68f22e99 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 25 Jan 2008 21:08:31 +0100
Subject: hrtimer: fixup the HRTIMER_CB_IRQSAFE_NO_SOFTIRQ fallback

Currently all highres=off timers are run from softirq context, but
HRTIMER_CB_IRQSAFE_NO_SOFTIRQ timers expect to run from irq context.

Fix this up by splitting it similar to the highres=on case.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/hrtimer.h |   5 +-
 kernel/hrtimer.c        | 270 +++++++++++++++++++++++++-----------------------
 kernel/timer.c          |   3 +-
 3 files changed, 143 insertions(+), 135 deletions(-)

(limited to 'include')

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index ecc8e2685e2b..49067f14fac1 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -115,10 +115,8 @@ struct hrtimer {
 	enum hrtimer_restart		(*function)(struct hrtimer *);
 	struct hrtimer_clock_base	*base;
 	unsigned long			state;
-#ifdef CONFIG_HIGH_RES_TIMERS
 	enum hrtimer_cb_mode		cb_mode;
 	struct list_head		cb_entry;
-#endif
 #ifdef CONFIG_TIMER_STATS
 	void				*start_site;
 	char				start_comm[16];
@@ -194,10 +192,10 @@ struct hrtimer_cpu_base {
 	spinlock_t			lock;
 	struct lock_class_key		lock_key;
 	struct hrtimer_clock_base	clock_base[HRTIMER_MAX_CLOCK_BASES];
+	struct list_head		cb_pending;
 #ifdef CONFIG_HIGH_RES_TIMERS
 	ktime_t				expires_next;
 	int				hres_active;
-	struct list_head		cb_pending;
 	unsigned long			nr_events;
 #endif
 };
@@ -319,6 +317,7 @@ extern void hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
 
 /* Soft interrupt function to run the hrtimer queues: */
 extern void hrtimer_run_queues(void);
+extern void hrtimer_run_pending(void);
 
 /* Bootup initialization: */
 extern void __init hrtimers_init(void);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 9f850ca032b6..061ae28a36a0 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -325,6 +325,22 @@ unsigned long ktime_divns(const ktime_t kt, s64 div)
 }
 #endif /* BITS_PER_LONG >= 64 */
 
+/*
+ * Check, whether the timer is on the callback pending list
+ */
+static inline int hrtimer_cb_pending(const struct hrtimer *timer)
+{
+	return timer->state & HRTIMER_STATE_PENDING;
+}
+
+/*
+ * Remove a timer from the callback pending list
+ */
+static inline void hrtimer_remove_cb_pending(struct hrtimer *timer)
+{
+	list_del_init(&timer->cb_entry);
+}
+
 /* High resolution timer related functions */
 #ifdef CONFIG_HIGH_RES_TIMERS
 
@@ -493,22 +509,6 @@ void hres_timers_resume(void)
 	retrigger_next_event(NULL);
 }
 
-/*
- * Check, whether the timer is on the callback pending list
- */
-static inline int hrtimer_cb_pending(const struct hrtimer *timer)
-{
-	return timer->state & HRTIMER_STATE_PENDING;
-}
-
-/*
- * Remove a timer from the callback pending list
- */
-static inline void hrtimer_remove_cb_pending(struct hrtimer *timer)
-{
-	list_del_init(&timer->cb_entry);
-}
-
 /*
  * Initialize the high resolution related parts of cpu_base
  */
@@ -516,7 +516,6 @@ static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base)
 {
 	base->expires_next.tv64 = KTIME_MAX;
 	base->hres_active = 0;
-	INIT_LIST_HEAD(&base->cb_pending);
 }
 
 /*
@@ -524,7 +523,6 @@ static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base)
  */
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer)
 {
-	INIT_LIST_HEAD(&timer->cb_entry);
 }
 
 /*
@@ -618,10 +616,13 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 {
 	return 0;
 }
-static inline int hrtimer_cb_pending(struct hrtimer *timer) { return 0; }
-static inline void hrtimer_remove_cb_pending(struct hrtimer *timer) { }
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { }
+static inline int hrtimer_reprogram(struct hrtimer *timer,
+				    struct hrtimer_clock_base *base)
+{
+	return 0;
+}
 
 #endif /* CONFIG_HIGH_RES_TIMERS */
 
@@ -1001,6 +1002,7 @@ void hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
 		clock_id = CLOCK_MONOTONIC;
 
 	timer->base = &cpu_base->clock_base[clock_id];
+	INIT_LIST_HEAD(&timer->cb_entry);
 	hrtimer_init_timer_hres(timer);
 
 #ifdef CONFIG_TIMER_STATS
@@ -1030,6 +1032,85 @@ int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp)
 }
 EXPORT_SYMBOL_GPL(hrtimer_get_res);
 
+static void run_hrtimer_pending(struct hrtimer_cpu_base *cpu_base)
+{
+	spin_lock_irq(&cpu_base->lock);
+
+	while (!list_empty(&cpu_base->cb_pending)) {
+		enum hrtimer_restart (*fn)(struct hrtimer *);
+		struct hrtimer *timer;
+		int restart;
+
+		timer = list_entry(cpu_base->cb_pending.next,
+				   struct hrtimer, cb_entry);
+
+		timer_stats_account_hrtimer(timer);
+
+		fn = timer->function;
+		__remove_hrtimer(timer, timer->base, HRTIMER_STATE_CALLBACK, 0);
+		spin_unlock_irq(&cpu_base->lock);
+
+		restart = fn(timer);
+
+		spin_lock_irq(&cpu_base->lock);
+
+		timer->state &= ~HRTIMER_STATE_CALLBACK;
+		if (restart == HRTIMER_RESTART) {
+			BUG_ON(hrtimer_active(timer));
+			/*
+			 * Enqueue the timer, allow reprogramming of the event
+			 * device
+			 */
+			enqueue_hrtimer(timer, timer->base, 1);
+		} else if (hrtimer_active(timer)) {
+			/*
+			 * If the timer was rearmed on another CPU, reprogram
+			 * the event device.
+			 */
+			if (timer->base->first == &timer->node)
+				hrtimer_reprogram(timer, timer->base);
+		}
+	}
+	spin_unlock_irq(&cpu_base->lock);
+}
+
+static void __run_hrtimer(struct hrtimer *timer)
+{
+	struct hrtimer_clock_base *base = timer->base;
+	struct hrtimer_cpu_base *cpu_base = base->cpu_base;
+	enum hrtimer_restart (*fn)(struct hrtimer *);
+	int restart;
+
+	__remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0);
+	timer_stats_account_hrtimer(timer);
+
+	fn = timer->function;
+	if (timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ) {
+		/*
+		 * Used for scheduler timers, avoid lock inversion with
+		 * rq->lock and tasklist_lock.
+		 *
+		 * These timers are required to deal with enqueue expiry
+		 * themselves and are not allowed to migrate.
+		 */
+		spin_unlock(&cpu_base->lock);
+		restart = fn(timer);
+		spin_lock(&cpu_base->lock);
+	} else
+		restart = fn(timer);
+
+	/*
+	 * Note: We clear the CALLBACK bit after enqueue_hrtimer to avoid
+	 * reprogramming of the event hardware. This happens at the end of this
+	 * function anyway.
+	 */
+	if (restart != HRTIMER_NORESTART) {
+		BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
+		enqueue_hrtimer(timer, base, 0);
+	}
+	timer->state &= ~HRTIMER_STATE_CALLBACK;
+}
+
 #ifdef CONFIG_HIGH_RES_TIMERS
 
 /*
@@ -1063,9 +1144,7 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 		basenow = ktime_add(now, base->offset);
 
 		while ((node = base->first)) {
-			enum hrtimer_restart (*fn)(struct hrtimer *);
 			struct hrtimer *timer;
-			int restart;
 
 			timer = rb_entry(node, struct hrtimer, node);
 
@@ -1089,37 +1168,7 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 				continue;
 			}
 
-			__remove_hrtimer(timer, base,
-					 HRTIMER_STATE_CALLBACK, 0);
-			timer_stats_account_hrtimer(timer);
-
-			fn = timer->function;
-			if (timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ) {
-				/*
-				 * Used for scheduler timers, avoid lock
-				 * inversion with rq->lock and tasklist_lock.
-				 *
-				 * These timers are required to deal with
-				 * enqueue expiry themselves and are not
-				 * allowed to migrate.
-				 */
-				spin_unlock(&cpu_base->lock);
-				restart = fn(timer);
-				spin_lock(&cpu_base->lock);
-			} else
-				restart = fn(timer);
-
-			/*
-			 * Note: We clear the CALLBACK bit after
-			 * enqueue_hrtimer to avoid reprogramming of
-			 * the event hardware. This happens at the end
-			 * of this function anyway.
-			 */
-			if (restart != HRTIMER_NORESTART) {
-				BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
-				enqueue_hrtimer(timer, base, 0);
-			}
-			timer->state &= ~HRTIMER_STATE_CALLBACK;
+			__run_hrtimer(timer);
 		}
 		spin_unlock(&cpu_base->lock);
 		base++;
@@ -1140,52 +1189,41 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 
 static void run_hrtimer_softirq(struct softirq_action *h)
 {
-	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
-
-	spin_lock_irq(&cpu_base->lock);
-
-	while (!list_empty(&cpu_base->cb_pending)) {
-		enum hrtimer_restart (*fn)(struct hrtimer *);
-		struct hrtimer *timer;
-		int restart;
-
-		timer = list_entry(cpu_base->cb_pending.next,
-				   struct hrtimer, cb_entry);
+	run_hrtimer_pending(&__get_cpu_var(hrtimer_bases));
+}
 
-		timer_stats_account_hrtimer(timer);
+#endif	/* CONFIG_HIGH_RES_TIMERS */
 
-		fn = timer->function;
-		__remove_hrtimer(timer, timer->base, HRTIMER_STATE_CALLBACK, 0);
-		spin_unlock_irq(&cpu_base->lock);
+/*
+ * Called from timer softirq every jiffy, expire hrtimers:
+ *
+ * For HRT its the fall back code to run the softirq in the timer
+ * softirq context in case the hrtimer initialization failed or has
+ * not been done yet.
+ */
+void hrtimer_run_pending(void)
+{
+	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
 
-		restart = fn(timer);
+	if (hrtimer_hres_active())
+		return;
 
-		spin_lock_irq(&cpu_base->lock);
+	/*
+	 * This _is_ ugly: We have to check in the softirq context,
+	 * whether we can switch to highres and / or nohz mode. The
+	 * clocksource switch happens in the timer interrupt with
+	 * xtime_lock held. Notification from there only sets the
+	 * check bit in the tick_oneshot code, otherwise we might
+	 * deadlock vs. xtime_lock.
+	 */
+	if (tick_check_oneshot_change(!hrtimer_is_hres_enabled()))
+		hrtimer_switch_to_hres();
 
-		timer->state &= ~HRTIMER_STATE_CALLBACK;
-		if (restart == HRTIMER_RESTART) {
-			BUG_ON(hrtimer_active(timer));
-			/*
-			 * Enqueue the timer, allow reprogramming of the event
-			 * device
-			 */
-			enqueue_hrtimer(timer, timer->base, 1);
-		} else if (hrtimer_active(timer)) {
-			/*
-			 * If the timer was rearmed on another CPU, reprogram
-			 * the event device.
-			 */
-			if (timer->base->first == &timer->node)
-				hrtimer_reprogram(timer, timer->base);
-		}
-	}
-	spin_unlock_irq(&cpu_base->lock);
+	run_hrtimer_pending(cpu_base);
 }
 
-#endif	/* CONFIG_HIGH_RES_TIMERS */
-
 /*
- * Expire the per base hrtimer-queue:
+ * Called from hardirq context every jiffy
  */
 static inline void run_hrtimer_queue(struct hrtimer_cpu_base *cpu_base,
 				     int index)
@@ -1199,46 +1237,27 @@ static inline void run_hrtimer_queue(struct hrtimer_cpu_base *cpu_base,
 	if (base->get_softirq_time)
 		base->softirq_time = base->get_softirq_time();
 
-	spin_lock_irq(&cpu_base->lock);
+	spin_lock(&cpu_base->lock);
 
 	while ((node = base->first)) {
 		struct hrtimer *timer;
-		enum hrtimer_restart (*fn)(struct hrtimer *);
-		int restart;
 
 		timer = rb_entry(node, struct hrtimer, node);
 		if (base->softirq_time.tv64 <= timer->expires.tv64)
 			break;
 
-#ifdef CONFIG_HIGH_RES_TIMERS
-		WARN_ON_ONCE(timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ);
-#endif
-		timer_stats_account_hrtimer(timer);
-
-		fn = timer->function;
-		__remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0);
-		spin_unlock_irq(&cpu_base->lock);
-
-		restart = fn(timer);
-
-		spin_lock_irq(&cpu_base->lock);
-
-		timer->state &= ~HRTIMER_STATE_CALLBACK;
-		if (restart != HRTIMER_NORESTART) {
-			BUG_ON(hrtimer_active(timer));
-			enqueue_hrtimer(timer, base, 0);
+		if (timer->cb_mode == HRTIMER_CB_SOFTIRQ) {
+			__remove_hrtimer(timer, base, HRTIMER_STATE_PENDING, 0);
+			list_add_tail(&timer->cb_entry,
+					&base->cpu_base->cb_pending);
+			continue;
 		}
+
+		__run_hrtimer(timer);
 	}
-	spin_unlock_irq(&cpu_base->lock);
+	spin_unlock(&cpu_base->lock);
 }
 
-/*
- * Called from timer softirq every jiffy, expire hrtimers:
- *
- * For HRT its the fall back code to run the softirq in the timer
- * softirq context in case the hrtimer initialization failed or has
- * not been done yet.
- */
 void hrtimer_run_queues(void)
 {
 	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
@@ -1247,18 +1266,6 @@ void hrtimer_run_queues(void)
 	if (hrtimer_hres_active())
 		return;
 
-	/*
-	 * This _is_ ugly: We have to check in the softirq context,
-	 * whether we can switch to highres and / or nohz mode. The
-	 * clocksource switch happens in the timer interrupt with
-	 * xtime_lock held. Notification from there only sets the
-	 * check bit in the tick_oneshot code, otherwise we might
-	 * deadlock vs. xtime_lock.
-	 */
-	if (tick_check_oneshot_change(!hrtimer_is_hres_enabled()))
-		if (hrtimer_switch_to_hres())
-			return;
-
 	hrtimer_get_softirq_time(cpu_base);
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
@@ -1407,6 +1414,7 @@ static void __cpuinit init_hrtimers_cpu(int cpu)
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
 		cpu_base->clock_base[i].cpu_base = cpu_base;
 
+	INIT_LIST_HEAD(&cpu_base->cb_pending);
 	hrtimer_init_hres(cpu_base);
 }
 
diff --git a/kernel/timer.c b/kernel/timer.c
index 2a00c22203f3..f739dfb539ce 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -896,7 +896,7 @@ static void run_timer_softirq(struct softirq_action *h)
 {
 	tvec_base_t *base = __get_cpu_var(tvec_bases);
 
-	hrtimer_run_queues();
+	hrtimer_run_pending();
 
 	if (time_after_eq(jiffies, base->timer_jiffies))
 		__run_timers(base);
@@ -907,6 +907,7 @@ static void run_timer_softirq(struct softirq_action *h)
  */
 void run_local_timers(void)
 {
+	hrtimer_run_queues();
 	raise_softirq(TIMER_SOFTIRQ);
 	softlockup_tick();
 }
-- 
cgit v1.2.3


From 6478d8800b75253b2a934ddcb734e13ade023ad0 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:33 +0100
Subject: sched: remove the !PREEMPT_BKL code

remove the !PREEMPT_BKL code.

this removes 160 lines of legacy code.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/hardirq.h  |   6 +--
 include/linux/smp_lock.h |  14 +-----
 kernel/Kconfig.preempt   |   4 --
 kernel/sched.c           |  19 ++------
 lib/kernel_lock.c        | 123 -----------------------------------------------
 5 files changed, 5 insertions(+), 161 deletions(-)

(limited to 'include')

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 8d302298a161..2961ec788046 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -72,11 +72,7 @@
 #define in_softirq()		(softirq_count())
 #define in_interrupt()		(irq_count())
 
-#if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL)
-# define in_atomic()	((preempt_count() & ~PREEMPT_ACTIVE) != kernel_locked())
-#else
-# define in_atomic()	((preempt_count() & ~PREEMPT_ACTIVE) != 0)
-#endif
+#define in_atomic()		((preempt_count() & ~PREEMPT_ACTIVE) != 0)
 
 #ifdef CONFIG_PREEMPT
 # define PREEMPT_CHECK_OFFSET 1
diff --git a/include/linux/smp_lock.h b/include/linux/smp_lock.h
index 58962c51dee1..aab3a4cff4e1 100644
--- a/include/linux/smp_lock.h
+++ b/include/linux/smp_lock.h
@@ -17,22 +17,10 @@ extern void __lockfunc __release_kernel_lock(void);
 		__release_kernel_lock();	\
 } while (0)
 
-/*
- * Non-SMP kernels will never block on the kernel lock,
- * so we are better off returning a constant zero from
- * reacquire_kernel_lock() so that the compiler can see
- * it at compile-time.
- */
-#if defined(CONFIG_SMP) && !defined(CONFIG_PREEMPT_BKL)
-# define return_value_on_smp return
-#else
-# define return_value_on_smp
-#endif
-
 static inline int reacquire_kernel_lock(struct task_struct *task)
 {
 	if (unlikely(task->lock_depth >= 0))
-		return_value_on_smp __reacquire_kernel_lock();
+		return __reacquire_kernel_lock();
 	return 0;
 }
 
diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
index 4420ef427f83..0669b70fa6a3 100644
--- a/kernel/Kconfig.preempt
+++ b/kernel/Kconfig.preempt
@@ -52,10 +52,6 @@ config PREEMPT
 
 endchoice
 
-config PREEMPT_BKL
-	def_bool y
-	depends on SMP || PREEMPT
-
 config RCU_TRACE
 	bool "Enable tracing for RCU - currently stats in debugfs"
 	select DEBUG_FS
diff --git a/kernel/sched.c b/kernel/sched.c
index 22712b2e058a..629614ad0358 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3955,10 +3955,9 @@ EXPORT_SYMBOL(schedule);
 asmlinkage void __sched preempt_schedule(void)
 {
 	struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
 	struct task_struct *task = current;
 	int saved_lock_depth;
-#endif
+
 	/*
 	 * If there is a non-zero preempt_count or interrupts are disabled,
 	 * we do not want to preempt the current task. Just return..
@@ -3974,14 +3973,10 @@ asmlinkage void __sched preempt_schedule(void)
 		 * clear ->lock_depth so that schedule() doesnt
 		 * auto-release the semaphore:
 		 */
-#ifdef CONFIG_PREEMPT_BKL
 		saved_lock_depth = task->lock_depth;
 		task->lock_depth = -1;
-#endif
 		schedule();
-#ifdef CONFIG_PREEMPT_BKL
 		task->lock_depth = saved_lock_depth;
-#endif
 		sub_preempt_count(PREEMPT_ACTIVE);
 
 		/*
@@ -4002,10 +3997,9 @@ EXPORT_SYMBOL(preempt_schedule);
 asmlinkage void __sched preempt_schedule_irq(void)
 {
 	struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
 	struct task_struct *task = current;
 	int saved_lock_depth;
-#endif
+
 	/* Catch callers which need to be fixed */
 	BUG_ON(ti->preempt_count || !irqs_disabled());
 
@@ -4017,16 +4011,12 @@ asmlinkage void __sched preempt_schedule_irq(void)
 		 * clear ->lock_depth so that schedule() doesnt
 		 * auto-release the semaphore:
 		 */
-#ifdef CONFIG_PREEMPT_BKL
 		saved_lock_depth = task->lock_depth;
 		task->lock_depth = -1;
-#endif
 		local_irq_enable();
 		schedule();
 		local_irq_disable();
-#ifdef CONFIG_PREEMPT_BKL
 		task->lock_depth = saved_lock_depth;
-#endif
 		sub_preempt_count(PREEMPT_ACTIVE);
 
 		/*
@@ -5241,11 +5231,8 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
 	spin_unlock_irqrestore(&rq->lock, flags);
 
 	/* Set the preempt count _outside_ the spinlocks! */
-#if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL)
-	task_thread_info(idle)->preempt_count = (idle->lock_depth >= 0);
-#else
 	task_thread_info(idle)->preempt_count = 0;
-#endif
+
 	/*
 	 * The idle tasks have their own, simple scheduling class:
 	 */
diff --git a/lib/kernel_lock.c b/lib/kernel_lock.c
index f73e2f8c308f..812dbf00844b 100644
--- a/lib/kernel_lock.c
+++ b/lib/kernel_lock.c
@@ -9,7 +9,6 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 
-#ifdef CONFIG_PREEMPT_BKL
 /*
  * The 'big kernel semaphore'
  *
@@ -86,128 +85,6 @@ void __lockfunc unlock_kernel(void)
 		up(&kernel_sem);
 }
 
-#else
-
-/*
- * The 'big kernel lock'
- *
- * This spinlock is taken and released recursively by lock_kernel()
- * and unlock_kernel().  It is transparently dropped and reacquired
- * over schedule().  It is used to protect legacy code that hasn't
- * been migrated to a proper locking design yet.
- *
- * Don't use in new code.
- */
-static  __cacheline_aligned_in_smp DEFINE_SPINLOCK(kernel_flag);
-
-
-/*
- * Acquire/release the underlying lock from the scheduler.
- *
- * This is called with preemption disabled, and should
- * return an error value if it cannot get the lock and
- * TIF_NEED_RESCHED gets set.
- *
- * If it successfully gets the lock, it should increment
- * the preemption count like any spinlock does.
- *
- * (This works on UP too - _raw_spin_trylock will never
- * return false in that case)
- */
-int __lockfunc __reacquire_kernel_lock(void)
-{
-	while (!_raw_spin_trylock(&kernel_flag)) {
-		if (test_thread_flag(TIF_NEED_RESCHED))
-			return -EAGAIN;
-		cpu_relax();
-	}
-	preempt_disable();
-	return 0;
-}
-
-void __lockfunc __release_kernel_lock(void)
-{
-	_raw_spin_unlock(&kernel_flag);
-	preempt_enable_no_resched();
-}
-
-/*
- * These are the BKL spinlocks - we try to be polite about preemption. 
- * If SMP is not on (ie UP preemption), this all goes away because the
- * _raw_spin_trylock() will always succeed.
- */
-#ifdef CONFIG_PREEMPT
-static inline void __lock_kernel(void)
-{
-	preempt_disable();
-	if (unlikely(!_raw_spin_trylock(&kernel_flag))) {
-		/*
-		 * If preemption was disabled even before this
-		 * was called, there's nothing we can be polite
-		 * about - just spin.
-		 */
-		if (preempt_count() > 1) {
-			_raw_spin_lock(&kernel_flag);
-			return;
-		}
-
-		/*
-		 * Otherwise, let's wait for the kernel lock
-		 * with preemption enabled..
-		 */
-		do {
-			preempt_enable();
-			while (spin_is_locked(&kernel_flag))
-				cpu_relax();
-			preempt_disable();
-		} while (!_raw_spin_trylock(&kernel_flag));
-	}
-}
-
-#else
-
-/*
- * Non-preemption case - just get the spinlock
- */
-static inline void __lock_kernel(void)
-{
-	_raw_spin_lock(&kernel_flag);
-}
-#endif
-
-static inline void __unlock_kernel(void)
-{
-	/*
-	 * the BKL is not covered by lockdep, so we open-code the
-	 * unlocking sequence (and thus avoid the dep-chain ops):
-	 */
-	_raw_spin_unlock(&kernel_flag);
-	preempt_enable();
-}
-
-/*
- * Getting the big kernel lock.
- *
- * This cannot happen asynchronously, so we only need to
- * worry about other CPU's.
- */
-void __lockfunc lock_kernel(void)
-{
-	int depth = current->lock_depth+1;
-	if (likely(!depth))
-		__lock_kernel();
-	current->lock_depth = depth;
-}
-
-void __lockfunc unlock_kernel(void)
-{
-	BUG_ON(current->lock_depth < 0);
-	if (likely(--current->lock_depth < 0))
-		__unlock_kernel();
-}
-
-#endif
-
 EXPORT_SYMBOL(lock_kernel);
 EXPORT_SYMBOL(unlock_kernel);
 
-- 
cgit v1.2.3


From e118adef232e637a8f091c1ded2fbf44fcf3ecc8 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Fri, 25 Jan 2008 21:08:34 +0100
Subject: timers: don't #error on higher HZ values

For some crazy reason (trying to work around hw problem in i810) I wanted
to use HZ around 4000.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/jiffies.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h
index 8b080024bbc1..7ba9e47bf061 100644
--- a/include/linux/jiffies.h
+++ b/include/linux/jiffies.h
@@ -29,6 +29,12 @@
 # define SHIFT_HZ	9
 #elif HZ >= 768 && HZ < 1536
 # define SHIFT_HZ	10
+#elif HZ >= 1536 && HZ < 3072
+# define SHIFT_HZ	11
+#elif HZ >= 3072 && HZ < 6144
+# define SHIFT_HZ	12
+#elif HZ >= 6144 && HZ < 12288
+# define SHIFT_HZ	13
 #else
 # error You lose.
 #endif
-- 
cgit v1.2.3


From 9745512ce79de686df354dc70a8d1a74d801892d Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jan 2008 21:08:34 +0100
Subject: sched: latencytop support

LatencyTOP kernel infrastructure; it measures latencies in the
scheduler and tracks it system wide and per process.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/stacktrace.c |  27 +++++
 fs/proc/base.c               |  78 ++++++++++++++
 include/linux/latencytop.h   |  44 ++++++++
 include/linux/sched.h        |   5 +
 include/linux/stacktrace.h   |   3 +
 kernel/Makefile              |   1 +
 kernel/fork.c                |   1 +
 kernel/latencytop.c          | 239 +++++++++++++++++++++++++++++++++++++++++++
 kernel/sched_fair.c          |   8 +-
 kernel/sysctl.c              |  10 ++
 lib/Kconfig.debug            |  14 +++
 11 files changed, 429 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/latencytop.h
 create mode 100644 kernel/latencytop.c

(limited to 'include')

diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index 6fa6cf036c70..55771fd7e545 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -33,6 +33,19 @@ static void save_stack_address(void *data, unsigned long addr)
 		trace->entries[trace->nr_entries++] = addr;
 }
 
+static void save_stack_address_nosched(void *data, unsigned long addr)
+{
+	struct stack_trace *trace = (struct stack_trace *)data;
+	if (in_sched_functions(addr))
+		return;
+	if (trace->skip > 0) {
+		trace->skip--;
+		return;
+	}
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = addr;
+}
+
 static const struct stacktrace_ops save_stack_ops = {
 	.warning = save_stack_warning,
 	.warning_symbol = save_stack_warning_symbol,
@@ -40,6 +53,13 @@ static const struct stacktrace_ops save_stack_ops = {
 	.address = save_stack_address,
 };
 
+static const struct stacktrace_ops save_stack_ops_nosched = {
+	.warning = save_stack_warning,
+	.warning_symbol = save_stack_warning_symbol,
+	.stack = save_stack_stack,
+	.address = save_stack_address_nosched,
+};
+
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
  */
@@ -50,3 +70,10 @@ void save_stack_trace(struct stack_trace *trace)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
 EXPORT_SYMBOL(save_stack_trace);
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	dump_trace(tsk, NULL, NULL, &save_stack_ops_nosched, trace);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 7411bfb0b7cc..91fa8e6ce8ad 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -310,6 +310,77 @@ static int proc_pid_schedstat(struct task_struct *task, char *buffer)
 }
 #endif
 
+#ifdef CONFIG_LATENCYTOP
+static int lstats_show_proc(struct seq_file *m, void *v)
+{
+	int i;
+	struct task_struct *task = m->private;
+	seq_puts(m, "Latency Top version : v0.1\n");
+
+	for (i = 0; i < 32; i++) {
+		if (task->latency_record[i].backtrace[0]) {
+			int q;
+			seq_printf(m, "%i %li %li ",
+				task->latency_record[i].count,
+				task->latency_record[i].time,
+				task->latency_record[i].max);
+			for (q = 0; q < LT_BACKTRACEDEPTH; q++) {
+				char sym[KSYM_NAME_LEN];
+				char *c;
+				if (!task->latency_record[i].backtrace[q])
+					break;
+				if (task->latency_record[i].backtrace[q] == ULONG_MAX)
+					break;
+				sprint_symbol(sym, task->latency_record[i].backtrace[q]);
+				c = strchr(sym, '+');
+				if (c)
+					*c = 0;
+				seq_printf(m, "%s ", sym);
+			}
+			seq_printf(m, "\n");
+		}
+
+	}
+	return 0;
+}
+
+static int lstats_open(struct inode *inode, struct file *file)
+{
+	int ret;
+	struct seq_file *m;
+	struct task_struct *task = get_proc_task(inode);
+
+	ret = single_open(file, lstats_show_proc, NULL);
+	if (!ret) {
+		m = file->private_data;
+		m->private = task;
+	}
+	return ret;
+}
+
+static ssize_t lstats_write(struct file *file, const char __user *buf,
+			    size_t count, loff_t *offs)
+{
+	struct seq_file *m;
+	struct task_struct *task;
+
+	m = file->private_data;
+	task = m->private;
+	clear_all_latency_tracing(task);
+
+	return count;
+}
+
+static const struct file_operations proc_lstats_operations = {
+	.open		= lstats_open,
+	.read		= seq_read,
+	.write		= lstats_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+#endif
+
 /* The badness from the OOM killer */
 unsigned long badness(struct task_struct *p, unsigned long uptime);
 static int proc_oom_score(struct task_struct *task, char *buffer)
@@ -1020,6 +1091,7 @@ static const struct file_operations proc_fault_inject_operations = {
 };
 #endif
 
+
 #ifdef CONFIG_SCHED_DEBUG
 /*
  * Print out various scheduling related per-task fields:
@@ -2230,6 +2302,9 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_SCHEDSTATS
 	INF("schedstat",  S_IRUGO, pid_schedstat),
 #endif
+#ifdef CONFIG_LATENCYTOP
+	REG("latency",  S_IRUGO, lstats),
+#endif
 #ifdef CONFIG_PROC_PID_CPUSET
 	REG("cpuset",     S_IRUGO, cpuset),
 #endif
@@ -2555,6 +2630,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_SCHEDSTATS
 	INF("schedstat", S_IRUGO, pid_schedstat),
 #endif
+#ifdef CONFIG_LATENCYTOP
+	REG("latency",  S_IRUGO, lstats),
+#endif
 #ifdef CONFIG_PROC_PID_CPUSET
 	REG("cpuset",    S_IRUGO, cpuset),
 #endif
diff --git a/include/linux/latencytop.h b/include/linux/latencytop.h
new file mode 100644
index 000000000000..901c2d6377a8
--- /dev/null
+++ b/include/linux/latencytop.h
@@ -0,0 +1,44 @@
+/*
+ * latencytop.h: Infrastructure for displaying latency
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ */
+
+#ifndef _INCLUDE_GUARD_LATENCYTOP_H_
+#define _INCLUDE_GUARD_LATENCYTOP_H_
+
+#ifdef CONFIG_LATENCYTOP
+
+#define LT_SAVECOUNT		32
+#define LT_BACKTRACEDEPTH	12
+
+struct latency_record {
+	unsigned long	backtrace[LT_BACKTRACEDEPTH];
+	unsigned int	count;
+	unsigned long	time;
+	unsigned long	max;
+};
+
+
+struct task_struct;
+
+void account_scheduler_latency(struct task_struct *task, int usecs, int inter);
+
+void clear_all_latency_tracing(struct task_struct *p);
+
+#else
+
+static inline void
+account_scheduler_latency(struct task_struct *task, int usecs, int inter)
+{
+}
+
+static inline void clear_all_latency_tracing(struct task_struct *p)
+{
+}
+
+#endif
+
+#endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index acadcab89ef9..dfc76e172f3f 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -88,6 +88,7 @@ struct sched_param {
 #include <linux/hrtimer.h>
 #include <linux/task_io_accounting.h>
 #include <linux/kobject.h>
+#include <linux/latencytop.h>
 
 #include <asm/processor.h>
 
@@ -1220,6 +1221,10 @@ struct task_struct {
 	int make_it_fail;
 #endif
 	struct prop_local_single dirties;
+#ifdef CONFIG_LATENCYTOP
+	int latency_record_count;
+	struct latency_record latency_record[LT_SAVECOUNT];
+#endif
 };
 
 /*
diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h
index e7fa657d0c49..5da9794b2d78 100644
--- a/include/linux/stacktrace.h
+++ b/include/linux/stacktrace.h
@@ -9,10 +9,13 @@ struct stack_trace {
 };
 
 extern void save_stack_trace(struct stack_trace *trace);
+extern void save_stack_trace_tsk(struct task_struct *tsk,
+				struct stack_trace *trace);
 
 extern void print_stack_trace(struct stack_trace *trace, int spaces);
 #else
 # define save_stack_trace(trace)			do { } while (0)
+# define save_stack_trace_tsk(tsk, trace)		do { } while (0)
 # define print_stack_trace(trace, spaces)		do { } while (0)
 #endif
 
diff --git a/kernel/Makefile b/kernel/Makefile
index 68755cd9a7e4..390d42146267 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -62,6 +62,7 @@ obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
 obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
 obj-$(CONFIG_MARKERS) += marker.o
+obj-$(CONFIG_LATENCYTOP) += latencytop.o
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
diff --git a/kernel/fork.c b/kernel/fork.c
index 0c969f4fade0..39d22b3357de 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1205,6 +1205,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 #ifdef TIF_SYSCALL_EMU
 	clear_tsk_thread_flag(p, TIF_SYSCALL_EMU);
 #endif
+	clear_all_latency_tracing(p);
 
 	/* Our parent execution domain becomes current domain
 	   These must match for thread signalling to apply */
diff --git a/kernel/latencytop.c b/kernel/latencytop.c
new file mode 100644
index 000000000000..b4e3c85abe74
--- /dev/null
+++ b/kernel/latencytop.c
@@ -0,0 +1,239 @@
+/*
+ * latencytop.c: Latency display infrastructure
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#include <linux/latencytop.h>
+#include <linux/kallsyms.h>
+#include <linux/seq_file.h>
+#include <linux/notifier.h>
+#include <linux/spinlock.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/stacktrace.h>
+
+static DEFINE_SPINLOCK(latency_lock);
+
+#define MAXLR 128
+static struct latency_record latency_record[MAXLR];
+
+int latencytop_enabled;
+
+void clear_all_latency_tracing(struct task_struct *p)
+{
+	unsigned long flags;
+
+	if (!latencytop_enabled)
+		return;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	memset(&p->latency_record, 0, sizeof(p->latency_record));
+	p->latency_record_count = 0;
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static void clear_global_latency_tracing(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	memset(&latency_record, 0, sizeof(latency_record));
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static void __sched
+account_global_scheduler_latency(struct task_struct *tsk, struct latency_record *lat)
+{
+	int firstnonnull = MAXLR + 1;
+	int i;
+
+	if (!latencytop_enabled)
+		return;
+
+	/* skip kernel threads for now */
+	if (!tsk->mm)
+		return;
+
+	for (i = 0; i < MAXLR; i++) {
+		int q;
+		int same = 1;
+		/* Nothing stored: */
+		if (!latency_record[i].backtrace[0]) {
+			if (firstnonnull > i)
+				firstnonnull = i;
+			continue;
+		}
+		for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) {
+			if (latency_record[i].backtrace[q] !=
+				lat->backtrace[q])
+				same = 0;
+			if (same && lat->backtrace[q] == 0)
+				break;
+			if (same && lat->backtrace[q] == ULONG_MAX)
+				break;
+		}
+		if (same) {
+			latency_record[i].count++;
+			latency_record[i].time += lat->time;
+			if (lat->time > latency_record[i].max)
+				latency_record[i].max = lat->time;
+			return;
+		}
+	}
+
+	i = firstnonnull;
+	if (i >= MAXLR - 1)
+		return;
+
+	/* Allocted a new one: */
+	memcpy(&latency_record[i], lat, sizeof(struct latency_record));
+}
+
+static inline void store_stacktrace(struct task_struct *tsk, struct latency_record *lat)
+{
+	struct stack_trace trace;
+
+	memset(&trace, 0, sizeof(trace));
+	trace.max_entries = LT_BACKTRACEDEPTH;
+	trace.entries = &lat->backtrace[0];
+	trace.skip = 0;
+	save_stack_trace_tsk(tsk, &trace);
+}
+
+void __sched
+account_scheduler_latency(struct task_struct *tsk, int usecs, int inter)
+{
+	unsigned long flags;
+	int i, q;
+	struct latency_record lat;
+
+	if (!latencytop_enabled)
+		return;
+
+	/* Long interruptible waits are generally user requested... */
+	if (inter && usecs > 5000)
+		return;
+
+	memset(&lat, 0, sizeof(lat));
+	lat.count = 1;
+	lat.time = usecs;
+	lat.max = usecs;
+	store_stacktrace(tsk, &lat);
+
+	spin_lock_irqsave(&latency_lock, flags);
+
+	account_global_scheduler_latency(tsk, &lat);
+
+	/*
+	 * short term hack; if we're > 32 we stop; future we recycle:
+	 */
+	tsk->latency_record_count++;
+	if (tsk->latency_record_count >= LT_SAVECOUNT)
+		goto out_unlock;
+
+	for (i = 0; i < LT_SAVECOUNT ; i++) {
+		struct latency_record *mylat;
+		int same = 1;
+		mylat = &tsk->latency_record[i];
+		for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) {
+			if (mylat->backtrace[q] !=
+				lat.backtrace[q])
+				same = 0;
+			if (same && lat.backtrace[q] == 0)
+				break;
+			if (same && lat.backtrace[q] == ULONG_MAX)
+				break;
+		}
+		if (same) {
+			mylat->count++;
+			mylat->time += lat.time;
+			if (lat.time > mylat->max)
+				mylat->max = lat.time;
+			goto out_unlock;
+		}
+	}
+
+	/* Allocated a new one: */
+	i = tsk->latency_record_count;
+	memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record));
+
+out_unlock:
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static int lstats_show(struct seq_file *m, void *v)
+{
+	int i;
+
+	seq_puts(m, "Latency Top version : v0.1\n");
+
+	for (i = 0; i < MAXLR; i++) {
+		if (latency_record[i].backtrace[0]) {
+			int q;
+			seq_printf(m, "%i %li %li ",
+				latency_record[i].count,
+				latency_record[i].time,
+				latency_record[i].max);
+			for (q = 0; q < LT_BACKTRACEDEPTH; q++) {
+				char sym[KSYM_NAME_LEN];
+				char *c;
+				if (!latency_record[i].backtrace[q])
+					break;
+				if (latency_record[i].backtrace[q] == ULONG_MAX)
+					break;
+				sprint_symbol(sym, latency_record[i].backtrace[q]);
+				c = strchr(sym, '+');
+				if (c)
+					*c = 0;
+				seq_printf(m, "%s ", sym);
+			}
+			seq_printf(m, "\n");
+		}
+	}
+	return 0;
+}
+
+static ssize_t
+lstats_write(struct file *file, const char __user *buf, size_t count,
+	     loff_t *offs)
+{
+	clear_global_latency_tracing();
+
+	return count;
+}
+
+static int lstats_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, lstats_show, NULL);
+}
+
+static struct file_operations lstats_fops = {
+	.open		= lstats_open,
+	.read		= seq_read,
+	.write		= lstats_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int __init init_lstats_procfs(void)
+{
+	struct proc_dir_entry *pe;
+
+	pe = create_proc_entry("latency_stats", 0644, NULL);
+	if (!pe)
+		return -ENOMEM;
+
+	pe->proc_fops = &lstats_fops;
+
+	return 0;
+}
+__initcall(init_lstats_procfs);
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 3dab1ff83c4f..1b3b40ad7c54 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -20,6 +20,8 @@
  *  Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
  */
 
+#include <linux/latencytop.h>
+
 /*
  * Targeted preemption latency for CPU-bound tasks:
  * (default: 20ms * (1 + ilog(ncpus)), units: nanoseconds)
@@ -434,6 +436,7 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 #ifdef CONFIG_SCHEDSTATS
 	if (se->sleep_start) {
 		u64 delta = rq_of(cfs_rq)->clock - se->sleep_start;
+		struct task_struct *tsk = task_of(se);
 
 		if ((s64)delta < 0)
 			delta = 0;
@@ -443,9 +446,12 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 
 		se->sleep_start = 0;
 		se->sum_sleep_runtime += delta;
+
+		account_scheduler_latency(tsk, delta >> 10, 1);
 	}
 	if (se->block_start) {
 		u64 delta = rq_of(cfs_rq)->clock - se->block_start;
+		struct task_struct *tsk = task_of(se);
 
 		if ((s64)delta < 0)
 			delta = 0;
@@ -462,11 +468,11 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 		 * time that the task spent sleeping:
 		 */
 		if (unlikely(prof_on == SLEEP_PROFILING)) {
-			struct task_struct *tsk = task_of(se);
 
 			profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk),
 				     delta >> 20);
 		}
+		account_scheduler_latency(tsk, delta >> 10, 0);
 	}
 #endif
 }
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 3afbd25f43eb..5418ef61e16e 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -81,6 +81,7 @@ extern int compat_log;
 extern int maps_protect;
 extern int sysctl_stat_interval;
 extern int audit_argv_kb;
+extern int latencytop_enabled;
 
 /* Constants used for minimum and  maximum */
 #ifdef CONFIG_DETECT_SOFTLOCKUP
@@ -416,6 +417,15 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= &proc_dointvec_taint,
 	},
 #endif
+#ifdef CONFIG_LATENCYTOP
+	{
+		.procname	= "latencytop",
+		.data		= &latencytop_enabled,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+#endif
 #ifdef CONFIG_SECURITY_CAPABILITIES
 	{
 		.procname	= "cap-bound",
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index a60109307d32..14fb355e3caa 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -517,4 +517,18 @@ config FAULT_INJECTION_STACKTRACE_FILTER
 	help
 	  Provide stacktrace filter for fault-injection capabilities
 
+config LATENCYTOP
+	bool "Latency measuring infrastructure"
+	select FRAME_POINTER if !MIPS
+	select KALLSYMS
+	select KALLSYMS_ALL
+	select STACKTRACE
+	select SCHEDSTATS
+	select SCHED_DEBUG
+	depends on X86 || X86_64
+	help
+	  Enable this option if you want to use the LatencyTOP tool
+	  to find out which userspace is blocking on what kernel operations.
+
+
 source "samples/Kconfig"
-- 
cgit v1.2.3


From 90739081ef8d5495d50abba9c5d333be9acd872a Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jan 2008 21:08:34 +0100
Subject: softlockup: fix signedness

fix softlockup tunables signedness.

mark tunables read-mostly.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  4 ++--
 kernel/softlockup.c   | 10 +++++-----
 kernel/sysctl.c       | 16 ++++++++--------
 3 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index dfc76e172f3f..53534f90a96a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -269,10 +269,10 @@ extern void softlockup_tick(void);
 extern void spawn_softlockup_task(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
-extern int softlockup_thresh;
+extern unsigned long  softlockup_thresh;
 extern unsigned long sysctl_hung_task_check_count;
 extern unsigned long sysctl_hung_task_timeout_secs;
-extern long sysctl_hung_task_warnings;
+extern unsigned long sysctl_hung_task_warnings;
 #else
 static inline void softlockup_tick(void)
 {
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 02f0ad534441..c1d76552446e 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -24,8 +24,8 @@ static DEFINE_PER_CPU(unsigned long, touch_timestamp);
 static DEFINE_PER_CPU(unsigned long, print_timestamp);
 static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
 
-static int did_panic;
-int softlockup_thresh = 60;
+static int __read_mostly did_panic;
+unsigned long __read_mostly softlockup_thresh = 60;
 
 static int
 softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
@@ -121,14 +121,14 @@ void softlockup_tick(void)
 /*
  * Have a reasonable limit on the number of tasks checked:
  */
-unsigned long sysctl_hung_task_check_count = 1024;
+unsigned long __read_mostly sysctl_hung_task_check_count = 1024;
 
 /*
  * Zero means infinite timeout - no checking done:
  */
-unsigned long sysctl_hung_task_timeout_secs = 120;
+unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120;
 
-long sysctl_hung_task_warnings = 10;
+unsigned long __read_mostly sysctl_hung_task_warnings = 10;
 
 /*
  * Only do the hung-tasks check on one CPU:
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 5418ef61e16e..8e96558cb8f3 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -772,9 +772,9 @@ static struct ctl_table kern_table[] = {
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "softlockup_thresh",
 		.data		= &softlockup_thresh,
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_doulongvec_minmax,
 		.strategy	= &sysctl_intvec,
 		.extra1		= &one,
 		.extra2		= &sixty,
@@ -783,27 +783,27 @@ static struct ctl_table kern_table[] = {
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "hung_task_check_count",
 		.data		= &sysctl_hung_task_check_count,
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_doulongvec_minmax,
 		.strategy	= &sysctl_intvec,
 	},
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "hung_task_timeout_secs",
 		.data		= &sysctl_hung_task_timeout_secs,
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_doulongvec_minmax,
 		.strategy	= &sysctl_intvec,
 	},
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "hung_task_warnings",
 		.data		= &sysctl_hung_task_warnings,
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_doulongvec_minmax,
 		.strategy	= &sysctl_intvec,
 	},
 #endif
-- 
cgit v1.2.3


From 286100a6cf1c1f692e5f81d14b364ff12b7662f5 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 25 Jan 2008 21:08:34 +0100
Subject: sched, futex: detach sched.h and futex.h

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/futex.h | 6 +++++-
 include/linux/sched.h | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/futex.h b/include/linux/futex.h
index 92d420fe03f8..1a15f8e237a7 100644
--- a/include/linux/futex.h
+++ b/include/linux/futex.h
@@ -1,8 +1,12 @@
 #ifndef _LINUX_FUTEX_H
 #define _LINUX_FUTEX_H
 
-#include <linux/sched.h>
+#include <linux/compiler.h>
+#include <linux/types.h>
 
+struct inode;
+struct mm_struct;
+struct task_struct;
 union ktime;
 
 /* Second argument to futex syscall */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 53534f90a96a..734f6d8f6ed5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -78,7 +78,6 @@ struct sched_param {
 #include <linux/proportions.h>
 #include <linux/seccomp.h>
 #include <linux/rcupdate.h>
-#include <linux/futex.h>
 #include <linux/rtmutex.h>
 
 #include <linux/time.h>
@@ -94,6 +93,7 @@ struct sched_param {
 
 struct exec_domain;
 struct futex_pi_state;
+struct robust_list_head;
 struct bio;
 
 /*
-- 
cgit v1.2.3


From 6d082592b62689fb91578d0338d04a9f50991990 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jan 2008 21:08:35 +0100
Subject: sched: keep total / count stats in addition to the max for

Right now, the linux kernel (with scheduler statistics enabled) keeps track
of the maximum time a process is waiting to be scheduled. While the maximum
is a very useful metric, tracking average and total is equally useful
(at least for latencytop) to figure out the accumulated effect of scheduler
delays. The accumulated effect is important to judge the performance impact
of scheduler tuning/behavior.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 2 ++
 kernel/sched_debug.c  | 4 ++++
 kernel/sched_fair.c   | 3 +++
 3 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 734f6d8f6ed5..df5b24ee80b3 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -895,6 +895,8 @@ struct sched_entity {
 #ifdef CONFIG_SCHEDSTATS
 	u64			wait_start;
 	u64			wait_max;
+	u64			wait_count;
+	u64			wait_sum;
 
 	u64			sleep_start;
 	u64			sleep_max;
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index 9e5de098d471..4b5e24cf2f4a 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -300,6 +300,8 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	PN(se.exec_max);
 	PN(se.slice_max);
 	PN(se.wait_max);
+	PN(se.wait_sum);
+	P(se.wait_count);
 	P(sched_info.bkl_count);
 	P(se.nr_migrations);
 	P(se.nr_migrations_cold);
@@ -367,6 +369,8 @@ void proc_sched_set_task(struct task_struct *p)
 {
 #ifdef CONFIG_SCHEDSTATS
 	p->se.wait_max				= 0;
+	p->se.wait_sum				= 0;
+	p->se.wait_count			= 0;
 	p->se.sleep_max				= 0;
 	p->se.sum_sleep_runtime			= 0;
 	p->se.block_max				= 0;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 45ff4e9411e0..72e25c7a3a18 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -385,6 +385,9 @@ update_stats_wait_end(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
 	schedstat_set(se->wait_max, max(se->wait_max,
 			rq_of(cfs_rq)->clock - se->wait_start));
+	schedstat_set(se->wait_count, se->wait_count + 1);
+	schedstat_set(se->wait_sum, se->wait_sum +
+			rq_of(cfs_rq)->clock - se->wait_start);
 	schedstat_set(se->wait_start, 0);
 }
 
-- 
cgit v1.2.3


From 7f1711234e6a21c153e892758d9d82c333ab37ac Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Date: Thu, 18 Oct 2007 19:56:47 -0300
Subject: V4L/DVB (6384): Replace TDA9887_SET_CONFIG by TUNER_SET_CONFIG

Currently, the only tuner-specific device that allows special
configurations is tda9887. However, tea5767 also may require some
special configurations (for example, to specify a different Xtal freq).

This patch replaces TDA9887_SET_CONFIG by a more generic internal ioctl
(TUNER_SET_CONFIG). The newer one allows specifying what tuner is
appliable to a configuration set, and allows an arbitrary configuration
struct.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/dvb/dvb-core/dvb_frontend.h   |  3 +++
 drivers/media/video/bt8xx/bttv-cards.c      |  8 ++++++--
 drivers/media/video/cx88/cx88-i2c.c         | 10 ++++++++--
 drivers/media/video/em28xx/em28xx-i2c.c     |  8 +++++++-
 drivers/media/video/em28xx/em28xx-video.c   |  5 -----
 drivers/media/video/saa7134/saa7134-cards.c | 13 +++++++++++--
 drivers/media/video/saa7134/saa7134-dvb.c   | 19 +++++++++++++------
 drivers/media/video/saa7134/saa7134-i2c.c   | 13 ++++++++++---
 drivers/media/video/saa7134/saa7134-video.c | 12 ++++++++++--
 drivers/media/video/tuner-core.c            | 23 +++++++++++++++++++----
 drivers/media/video/tuner-simple.c          | 15 +++++++++++++--
 drivers/media/video/v4l2-common.c           |  2 +-
 include/media/v4l2-common.h                 |  7 ++++++-
 13 files changed, 107 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.h b/drivers/media/dvb/dvb-core/dvb_frontend.h
index a5262e852c82..c49fa049717e 100644
--- a/drivers/media/dvb/dvb-core/dvb_frontend.h
+++ b/drivers/media/dvb/dvb-core/dvb_frontend.h
@@ -84,6 +84,9 @@ struct dvb_tuner_ops {
 	/** This is support for demods like the mt352 - fills out the supplied buffer with what to write. */
 	int (*calc_regs)(struct dvb_frontend *fe, struct dvb_frontend_parameters *p, u8 *buf, int buf_len);
 
+	/** This is to allow setting tuner-specific configs */
+	int (*set_config)(struct dvb_frontend *fe, void *priv_cfg);
+
 	int (*get_frequency)(struct dvb_frontend *fe, u32 *frequency);
 	int (*get_bandwidth)(struct dvb_frontend *fe, u32 *bandwidth);
 
diff --git a/drivers/media/video/bt8xx/bttv-cards.c b/drivers/media/video/bt8xx/bttv-cards.c
index 585d1ef95afd..4f91f98db37b 100644
--- a/drivers/media/video/bt8xx/bttv-cards.c
+++ b/drivers/media/video/bt8xx/bttv-cards.c
@@ -3574,8 +3574,12 @@ void __devinit bttv_init_card2(struct bttv *btv)
 	}
 
 	if (btv->tda9887_conf) {
-		bttv_call_i2c_clients(btv, TDA9887_SET_CONFIG,
-							&btv->tda9887_conf);
+		struct v4l2_priv_tun_config tda9887_cfg;
+
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv = &btv->tda9887_conf;
+
+		bttv_call_i2c_clients(btv, TUNER_SET_CONFIG, &tda9887_cfg);
 	}
 
 	btv->svhs = bttv_tvcards[btv->c.type].svhs;
diff --git a/drivers/media/video/cx88/cx88-i2c.c b/drivers/media/video/cx88/cx88-i2c.c
index c8b1c50625f4..937497c86247 100644
--- a/drivers/media/video/cx88/cx88-i2c.c
+++ b/drivers/media/video/cx88/cx88-i2c.c
@@ -127,8 +127,14 @@ static int attach_inform(struct i2c_client *client)
 		}
 	}
 
-	if (core->board.tda9887_conf)
-		client->driver->command(client, TDA9887_SET_CONFIG, &core->board.tda9887_conf);
+	if (core->board.tda9887_conf) {
+		struct v4l2_priv_tun_config tda9887_cfg;
+
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv  = &core->board.tda9887_conf;
+
+		client->driver->command(client, TUNER_SET_CONFIG, &tda9887_cfg);
+	}
 	return 0;
 }
 
diff --git a/drivers/media/video/em28xx/em28xx-i2c.c b/drivers/media/video/em28xx/em28xx-i2c.c
index e3a4aa7a9df4..e48191fb1dde 100644
--- a/drivers/media/video/em28xx/em28xx-i2c.c
+++ b/drivers/media/video/em28xx/em28xx-i2c.c
@@ -421,6 +421,8 @@ static int attach_inform(struct i2c_client *client)
 		case 0x96:
 		case 0x94:
 		{
+			struct v4l2_priv_tun_config tda9887_cfg;
+
 			struct tuner_setup tun_setup;
 
 			tun_setup.mode_mask = T_ANALOG_TV | T_RADIO;
@@ -428,7 +430,11 @@ static int attach_inform(struct i2c_client *client)
 			tun_setup.addr = client->addr;
 
 			em28xx_i2c_call_clients(dev, TUNER_SET_TYPE_ADDR, &tun_setup);
-			em28xx_i2c_call_clients(dev, TDA9887_SET_CONFIG, &dev->tda9887_conf);
+
+			tda9887_cfg.tuner = TUNER_TDA9887;
+			tda9887_cfg.priv = &dev->tda9887_conf;
+			em28xx_i2c_call_clients(dev, TUNER_SET_CONFIG,
+						&tda9887_cfg);
 			break;
 		}
 		case 0x42:
diff --git a/drivers/media/video/em28xx/em28xx-video.c b/drivers/media/video/em28xx/em28xx-video.c
index 0906bc5766cc..bb0933f9d84c 100644
--- a/drivers/media/video/em28xx/em28xx-video.c
+++ b/drivers/media/video/em28xx/em28xx-video.c
@@ -186,11 +186,6 @@ static void em28xx_config_i2c(struct em28xx *dev)
 	f.frequency = 9076;	/* FIXME:remove magic number */
 	dev->ctl_freq = f.frequency;
 	em28xx_i2c_call_clients(dev, VIDIOC_S_FREQUENCY, &f);
-
-	/* configure tda9887 */
-
-
-/*	em28xx_i2c_call_clients(dev,VIDIOC_S_STD,&dev->tvnorm->id); */
 }
 
 /*
diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c
index 98c1b084a716..78d7d7059a9e 100644
--- a/drivers/media/video/saa7134/saa7134-cards.c
+++ b/drivers/media/video/saa7134/saa7134-cards.c
@@ -4570,8 +4570,17 @@ int saa7134_board_init2(struct saa7134_dev *dev)
 
 		printk(KERN_INFO "%s Tuner type is %d\n", dev->name, dev->tuner_type);
 		if (dev->tuner_type == TUNER_PHILIPS_FMD1216ME_MK3) {
-			dev->tda9887_conf = TDA9887_PRESENT | TDA9887_PORT1_ACTIVE | TDA9887_PORT2_ACTIVE;
-			saa7134_i2c_call_clients(dev,TDA9887_SET_CONFIG, &dev->tda9887_conf);
+			struct v4l2_priv_tun_config tda9887_cfg;
+
+			tda9887_cfg.tuner = TUNER_TDA9887;
+			tda9887_cfg.priv  = &dev->tda9887_conf;
+
+			dev->tda9887_conf = TDA9887_PRESENT      |
+					    TDA9887_PORT1_ACTIVE |
+					    TDA9887_PORT2_ACTIVE;
+
+			saa7134_i2c_call_clients(dev, TUNER_SET_CONFIG,
+						 &tda9887_cfg);
 		}
 
 		tun_setup.mode_mask = T_RADIO | T_ANALOG_TV | T_DIGITAL_TV;
diff --git a/drivers/media/video/saa7134/saa7134-dvb.c b/drivers/media/video/saa7134/saa7134-dvb.c
index e1ab099ec4c6..a9ca5730826f 100644
--- a/drivers/media/video/saa7134/saa7134-dvb.c
+++ b/drivers/media/video/saa7134/saa7134-dvb.c
@@ -1073,14 +1073,21 @@ static int dvb_init(struct saa7134_dev *dev)
 
 static int dvb_fini(struct saa7134_dev *dev)
 {
-	static int on  = TDA9887_PRESENT | TDA9887_PORT2_INACTIVE;
+	/* FIXME: I suspect that this code is bogus, since the entry for
+	   Pinnacle 300I DVB-T PAL already defines the proper init to allow
+	   the detection of mt2032 (TDA9887_PORT2_INACTIVE)
+	 */
+	if (dev->board == SAA7134_BOARD_PINNACLE_300I_DVBT_PAL) {
+		struct v4l2_priv_tun_config tda9887_cfg;
+		static int on  = TDA9887_PRESENT | TDA9887_PORT2_INACTIVE;
+
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv  = &on;
 
-	switch (dev->board) {
-	case SAA7134_BOARD_PINNACLE_300I_DVBT_PAL:
 		/* otherwise we don't detect the tuner on next insmod */
-		saa7134_i2c_call_clients(dev,TDA9887_SET_CONFIG,&on);
-		break;
-	};
+		saa7134_i2c_call_clients(dev, TUNER_SET_CONFIG, &tda9887_cfg);
+	}
+
 	videobuf_dvb_unregister(&dev->dvb);
 	return 0;
 }
diff --git a/drivers/media/video/saa7134/saa7134-i2c.c b/drivers/media/video/saa7134/saa7134-i2c.c
index 6deaad1a5480..800b397f8f16 100644
--- a/drivers/media/video/saa7134/saa7134-i2c.c
+++ b/drivers/media/video/saa7134/saa7134-i2c.c
@@ -323,7 +323,6 @@ static int attach_inform(struct i2c_client *client)
 {
 	struct saa7134_dev *dev = client->adapter->algo_data;
 	int tuner = dev->tuner_type;
-	int conf  = dev->tda9887_conf;
 	struct tuner_setup tun_setup;
 
 	d1printk( "%s i2c attach [addr=0x%x,client=%s]\n",
@@ -360,7 +359,6 @@ static int attach_inform(struct i2c_client *client)
 	}
 
 	if (tuner != UNSET) {
-
 		tun_setup.type = tuner;
 		tun_setup.addr = saa7134_boards[dev->board].tuner_addr;
 		tun_setup.config = saa7134_boards[dev->board].tuner_config;
@@ -372,9 +370,18 @@ static int attach_inform(struct i2c_client *client)
 
 			client->driver->command(client,TUNER_SET_TYPE_ADDR, &tun_setup);
 		}
+
+		if (tuner == TUNER_TDA9887) {
+			struct v4l2_priv_tun_config tda9887_cfg;
+
+			tda9887_cfg.tuner = TUNER_TDA9887;
+			tda9887_cfg.priv = &dev->tda9887_conf;
+
+			client->driver->command(client, TUNER_SET_CONFIG,
+						&tda9887_cfg);
+		}
 	}
 
-	client->driver->command(client, TDA9887_SET_CONFIG, &conf);
 
 	return 0;
 }
diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c
index 6396d9b5c063..2c1d56ab2f2a 100644
--- a/drivers/media/video/saa7134/saa7134-video.c
+++ b/drivers/media/video/saa7134/saa7134-video.c
@@ -1236,16 +1236,24 @@ static int set_control(struct saa7134_dev *dev, struct saa7134_fh *fh,
 		restart_overlay = 1;
 		break;
 	case V4L2_CID_PRIVATE_AUTOMUTE:
+	{
+		struct v4l2_priv_tun_config tda9887_cfg;
+
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv = &dev->tda9887_conf;
+
 		dev->ctl_automute = c->value;
 		if (dev->tda9887_conf) {
 			if (dev->ctl_automute)
 				dev->tda9887_conf |= TDA9887_AUTOMUTE;
 			else
 				dev->tda9887_conf &= ~TDA9887_AUTOMUTE;
-			saa7134_i2c_call_clients(dev, TDA9887_SET_CONFIG,
-						 &dev->tda9887_conf);
+
+			saa7134_i2c_call_clients(dev, TUNER_SET_CONFIG,
+						 &tda9887_cfg);
 		}
 		break;
+	}
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 9e99f3636d3d..d1d6c664bb09 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -889,14 +889,29 @@ static int tuner_command(struct i2c_client *client, unsigned int cmd, void *arg)
 			return 0;
 		}
 #endif
-	case TDA9887_SET_CONFIG:
-		if (t->type == TUNER_TDA9887) {
-			int *i = arg;
+	case TUNER_SET_CONFIG:
+	{
+		struct dvb_tuner_ops *fe_tuner_ops = &t->fe.ops.tuner_ops;
+		struct v4l2_priv_tun_config *cfg = arg;
+
+		if (t->type != cfg->tuner)
+			break;
 
-			t->tda9887_config = *i;
+		if (t->type == TUNER_TDA9887) {
+			t->tda9887_config = *(unsigned int *)cfg->priv;
 			set_freq(client, t->tv_freq);
+			break;
 		}
+
+		if (NULL == fe_tuner_ops->set_config) {
+			tuner_warn("Tuner frontend module has no way to "
+				   "set config\n");
+			break;
+		}
+		fe_tuner_ops->set_config(&t->fe, cfg->priv);
+
 		break;
+	}
 	/* --- v4l ioctls --- */
 	/* take care: bttv does userspace copying, we'll get a
 	   kernel pointer here... */
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index 7b93d3b1f4c6..eec13cb6cd6f 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -355,10 +355,14 @@ static int simple_set_tv_freq(struct dvb_frontend *fe,
 	}
 	priv->last_div = div;
 	if (t_params->has_tda9887) {
+		struct v4l2_priv_tun_config tda9887_cfg;
 		int config = 0;
 		int is_secam_l = (params->std & (V4L2_STD_SECAM_L | V4L2_STD_SECAM_LC)) &&
 			!(params->std & ~(V4L2_STD_SECAM_L | V4L2_STD_SECAM_LC));
 
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv  = &config;
+
 		if (params->std == V4L2_STD_SECAM_LC) {
 			if (t_params->port1_active ^ t_params->port1_invert_for_secam_lc)
 				config |= TDA9887_PORT1_ACTIVE;
@@ -391,7 +395,8 @@ static int simple_set_tv_freq(struct dvb_frontend *fe,
 		}
 		if (t_params->default_pll_gating_18)
 			config |= TDA9887_GATING_18;
-		i2c_clients_command(priv->i2c_props.adap, TDA9887_SET_CONFIG, &config);
+		i2c_clients_command(priv->i2c_props.adap, TUNER_SET_CONFIG,
+				    &tda9887_cfg);
 	}
 	tuner_dbg("tv 0x%02x 0x%02x 0x%02x 0x%02x\n",
 		  buffer[0],buffer[1],buffer[2],buffer[3]);
@@ -534,6 +539,11 @@ static int simple_set_radio_freq(struct dvb_frontend *fe,
 
 	if (t_params->has_tda9887) {
 		int config = 0;
+		struct v4l2_priv_tun_config tda9887_cfg;
+
+		tda9887_cfg.tuner = TUNER_TDA9887;
+		tda9887_cfg.priv = &config;
+
 		if (t_params->port1_active && !t_params->port1_fm_high_sensitivity)
 			config |= TDA9887_PORT1_ACTIVE;
 		if (t_params->port2_active && !t_params->port2_fm_high_sensitivity)
@@ -546,7 +556,8 @@ static int simple_set_radio_freq(struct dvb_frontend *fe,
 			config |= TDA9887_GAIN_NORMAL;
 		if (t_params->radio_if == 2)
 			config |= TDA9887_RIF_41_3;
-		i2c_clients_command(priv->i2c_props.adap, TDA9887_SET_CONFIG, &config);
+		i2c_clients_command(priv->i2c_props.adap, TUNER_SET_CONFIG,
+					&tda9887_cfg);
 	}
 	if (4 != (rc = tuner_i2c_xfer_send(&priv->i2c_props,buffer,4)))
 		tuner_warn("i2c i/o error: rc == %d (should be 4)\n",rc);
diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index 1141b4bf41ce..a4d68d3ba5e4 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -400,7 +400,7 @@ static const char *v4l2_int_ioctls[] = {
 
 	[_IOC_NR(TUNER_SET_TYPE_ADDR)]         = "TUNER_SET_TYPE_ADDR",
 	[_IOC_NR(TUNER_SET_STANDBY)]           = "TUNER_SET_STANDBY",
-	[_IOC_NR(TDA9887_SET_CONFIG)]          = "TDA9887_SET_CONFIG",
+	[_IOC_NR(TUNER_SET_CONFIG)]            = "TUNER_SET_CONFIG",
 
 	[_IOC_NR(VIDIOC_INT_S_TUNER_MODE)]     = "VIDIOC_INT_S_TUNER_MODE",
 	[_IOC_NR(VIDIOC_INT_RESET)]            = "VIDIOC_INT_RESET",
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index 181a40c46a52..c019c1e99895 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -116,6 +116,11 @@ struct v4l2_decode_vbi_line {
 	u32 type;		/* VBI service type (V4L2_SLICED_*). 0 if no service found */
 };
 
+struct v4l2_priv_tun_config {
+	int tuner;
+	void *priv;
+};
+
 /* audio ioctls */
 
 /* v4l device was opened in Radio mode, to be replaced by VIDIOC_INT_S_TUNER_MODE */
@@ -131,7 +136,7 @@ struct v4l2_decode_vbi_line {
 #define TUNER_SET_STANDBY            _IOW('d', 91, int)
 
 /* Sets tda9887 specific stuff, like port1, port2 and qss */
-#define TDA9887_SET_CONFIG           _IOW('d', 92, int)
+#define TUNER_SET_CONFIG           _IOW('d', 92, struct v4l2_priv_tun_config)
 
 /* Switch the tuner to a specific tuner mode. Replacement of AUDC_SET_RADIO */
 #define VIDIOC_INT_S_TUNER_MODE	     _IOW('d', 93, enum v4l2_tuner_type)
-- 
cgit v1.2.3


From 12466577853b0d057f4416f4c7020e544f3a4209 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Date: Wed, 24 Oct 2007 11:08:20 -0300
Subject: V4L/DVB (6422): Add the pending entries for xc2028/3028 based boards

Also replaces all occurrences of TUNER_XCEIVE_XC3028 to TUNER_XC2028.

Some work is still may be required to make sure that non-tm6000 drivers will
be capable of using tuner-xc2028.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.tuner  |  2 +-
 drivers/media/video/em28xx/em28xx-cards.c | 71 +++++++++++++++++++++++++++++++
 drivers/media/video/ivtv/ivtv-cards.c     |  6 +--
 drivers/media/video/ivtv/ivtv-driver.c    |  2 +-
 drivers/media/video/tuner-types.c         |  6 +--
 include/media/tuner.h                     |  2 +-
 6 files changed, 80 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner
index a88c02d23805..84c3ac7c33d9 100644
--- a/Documentation/video4linux/CARDLIST.tuner
+++ b/Documentation/video4linux/CARDLIST.tuner
@@ -69,7 +69,7 @@ tuner=67 - Philips TD1316 Hybrid Tuner
 tuner=68 - Philips TUV1236D ATSC/NTSC dual in
 tuner=69 - Tena TNF 5335 and similar models
 tuner=70 - Samsung TCPN 2121P30A
-tuner=71 - Xceive xc3028
+tuner=71 - Xceive xc2028/xc3028 tuner
 tuner=72 - Thomson FE6600
 tuner=73 - Samsung TCPG 6121P30A
 tuner=75 - Philips TEA5761 FM Radio
diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c
index 418ea8b7f85a..2d72de0e7564 100644
--- a/drivers/media/video/em28xx/em28xx-cards.c
+++ b/drivers/media/video/em28xx/em28xx-cards.c
@@ -156,6 +156,74 @@ struct em28xx_board em28xx_boards[] = {
 					MSP_DSP_IN_SCART, MSP_DSP_IN_SCART),
 		}},
 	},
+	[EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900] = {
+		.name         = "Hauppauge WinTV HVR 900",
+		.vchannels    = 3,
+		.norm         = VIDEO_MODE_PAL,
+		.tda9887_conf = TDA9887_PRESENT,
+		.tuner_type   = TUNER_XC2028,
+		.has_tuner    = 1,
+		.decoder      = EM28XX_TVP5150,
+		.input          = {{
+			.type     = EM28XX_VMUX_COMPOSITE1,
+			.vmux     = TVP5150_COMPOSITE1,
+			.amux     = 1,
+		},{
+			.type     = EM28XX_VMUX_TELEVISION,
+			.vmux     = TVP5150_COMPOSITE0,
+			.amux     = 0,
+		},{
+			.type     = EM28XX_VMUX_SVIDEO,
+			.vmux     = TVP5150_SVIDEO,
+			.amux     = 1,
+		}},
+	},
+	[EM2880_BOARD_TERRATEC_HYBRID_XS] = {
+		.name         = "Terratec Hybrid XS",
+		.vchannels    = 3,
+		.norm         = VIDEO_MODE_PAL,
+		.tda9887_conf = TDA9887_PRESENT,
+		.has_tuner    = 1,
+		.tuner_type   = TUNER_XC2028,
+		.decoder      = EM28XX_TVP5150,
+		.input          = {{
+			.type     = EM28XX_VMUX_TELEVISION,
+			.vmux     = TVP5150_COMPOSITE0,
+			.amux     = 0,
+		},{
+			.type     = EM28XX_VMUX_COMPOSITE1,
+			.vmux     = TVP5150_COMPOSITE1,
+			.amux     = 1,
+		},{
+			.type     = EM28XX_VMUX_SVIDEO,
+			.vmux     = TVP5150_SVIDEO,
+			.amux     = 1,
+		}},
+	},
+	/* maybe there's a reason behind it why Terratec sells the Hybrid XS as Prodigy XS with a
+	 * different PID, let's keep it separated for now maybe we'll need it lateron */
+	[EM2880_BOARD_TERRATEC_PRODIGY_XS] = {
+		.name         = "Terratec Prodigy XS",
+		.vchannels    = 3,
+		.norm         = VIDEO_MODE_PAL,
+		.tda9887_conf = TDA9887_PRESENT,
+		.has_tuner    = 1,
+		.tuner_type   = TUNER_XC2028,
+		.decoder      = EM28XX_TVP5150,
+		.input          = {{
+			.type     = EM28XX_VMUX_TELEVISION,
+			.vmux     = TVP5150_COMPOSITE0,
+			.amux     = 0,
+		},{
+			.type     = EM28XX_VMUX_COMPOSITE1,
+			.vmux     = TVP5150_COMPOSITE1,
+			.amux     = 1,
+		},{
+			.type     = EM28XX_VMUX_SVIDEO,
+			.vmux     = TVP5150_SVIDEO,
+			.amux     = 1,
+		}},
+	},
 	[EM2820_BOARD_MSI_VOX_USB_2] = {
 		.name		= "MSI VOX USB 2.0",
 		.vchannels	= 3,
@@ -274,6 +342,9 @@ struct usb_device_id em28xx_id_table [] = {
 	{ USB_DEVICE(0x2304, 0x0208), .driver_info = EM2820_BOARD_PINNACLE_USB_2 },
 	{ USB_DEVICE(0x2040, 0x4200), .driver_info = EM2820_BOARD_HAUPPAUGE_WINTV_USB_2 },
 	{ USB_DEVICE(0x2304, 0x0207), .driver_info = EM2820_BOARD_PINNACLE_DVC_90 },
+	{ USB_DEVICE(0x2040, 0x6500), .driver_info = EM2880_BOARD_HAUPPAUGE_WINTV_HVR_900 },
+	{ USB_DEVICE(0x0ccd, 0x0042), .driver_info = EM2880_BOARD_TERRATEC_HYBRID_XS },
+	{ USB_DEVICE(0x0ccd, 0x0047), .driver_info = EM2880_BOARD_TERRATEC_PRODIGY_XS },
 	{ },
 };
 
diff --git a/drivers/media/video/ivtv/ivtv-cards.c b/drivers/media/video/ivtv/ivtv-cards.c
index b6a8be622d3c..aaa114b5c268 100644
--- a/drivers/media/video/ivtv/ivtv-cards.c
+++ b/drivers/media/video/ivtv/ivtv-cards.c
@@ -848,7 +848,7 @@ static const struct ivtv_card ivtv_card_pg600v2 = {
 		{ IVTV_CARD_INPUT_LINE_IN1,   CX25840_AUDIO_SERIAL },
 	},
 	.tuners = {
-		{ .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 },
+		{ .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 },
 	},
 	.pci_list = ivtv_pci_pg600v2,
 };
@@ -879,7 +879,7 @@ static const struct ivtv_card ivtv_card_club3d = {
 		{ IVTV_CARD_INPUT_LINE_IN1,   CX25840_AUDIO_SERIAL },
 	},
 	.tuners = {
-		{ .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 },
+		{ .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 },
 	},
 	.pci_list = ivtv_pci_club3d,
 };
@@ -910,7 +910,7 @@ static const struct ivtv_card ivtv_card_avertv_mce116 = {
 	},
 	.gpio_init = { .direction = 0xe000, .initial_value = 0x4000 }, /* enable line-in */
 	.tuners = {
-		{ .std = V4L2_STD_ALL, .tuner = TUNER_XCEIVE_XC3028 },
+		{ .std = V4L2_STD_ALL, .tuner = TUNER_XC2028 },
 	},
 	.pci_list = ivtv_pci_avertv_mce116,
 };
diff --git a/drivers/media/video/ivtv/ivtv-driver.c b/drivers/media/video/ivtv/ivtv-driver.c
index 6d2dd8764f81..ec40453e46e4 100644
--- a/drivers/media/video/ivtv/ivtv-driver.c
+++ b/drivers/media/video/ivtv/ivtv-driver.c
@@ -829,7 +829,7 @@ static void ivtv_load_and_init_modules(struct ivtv *itv)
 	/* load modules */
 #ifndef CONFIG_VIDEO_TUNER
 	if (hw & IVTV_HW_TUNER) {
-		if (itv->options.tuner == TUNER_XCEIVE_XC3028) {
+		if (itv->options.tuner == TUNER_XC2028) {
 			IVTV_INFO("Xceive tuner not yet supported, only composite and S-Video inputs will be available\n");
 			itv->tunerid = 1;
 		}
diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c
index c6a7934bd5a6..ac363f019227 100644
--- a/drivers/media/video/tuner-types.c
+++ b/drivers/media/video/tuner-types.c
@@ -1452,9 +1452,9 @@ struct tunertype tuners[] = {
 		.params = tuner_samsung_tcpn_2121p30a_params,
 		.count  = ARRAY_SIZE(tuner_samsung_tcpn_2121p30a_params),
 	},
-	[TUNER_XCEIVE_XC3028] = { /* Xceive 3028 */
-		.name	= "Xceive xc3028",
-		/* see xc3028.c for details */
+	[TUNER_XC2028] = { /* Xceive 2028 */
+		.name   = "Xceive xc2028/xc3028 tuner",
+		/* see tuner-xc2028.c for details */
 	},
 	[TUNER_THOMSON_FE6600] = { /* Thomson PAL / DVB-T */
 		.name   = "Thomson FE6600",
diff --git a/include/media/tuner.h b/include/media/tuner.h
index c03dceb92605..d49392d90e56 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -117,7 +117,7 @@ extern int tuner_debug;
 #define TUNER_PHILIPS_TUV1236D		68	/* ATI HDTV Wonder */
 #define TUNER_TNF_5335MF                69	/* Sabrent Bt848   */
 #define TUNER_SAMSUNG_TCPN_2121P30A     70 	/* Hauppauge PVR-500MCE NTSC */
-#define TUNER_XCEIVE_XC3028		71
+#define TUNER_XC2028			71
 
 #define TUNER_THOMSON_FE6600		72	/* DViCO FusionHDTV DVB-T Hybrid */
 #define TUNER_SAMSUNG_TCPG_6121P30A     73 	/* Hauppauge PVR-500 PAL */
-- 
cgit v1.2.3


From 5bea1cd3871351d70cc7624af138f8aa68b7be77 Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@linuxtv.org>
Date: Mon, 22 Oct 2007 09:56:38 -0300
Subject: V4L/DVB (6435): tda8290: add support for NXP TDA18271 tuner and
 TDA8295 analog demod

Add basic support for NXP TDA8295 analog demod and TDA18271 tuner silicon.

TDA8295 + TDA8275a not yet tested.
TDA8290 + TDA18271 not yet supported.

Digital mode of TDA18271 not yet tested & needs more work.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Reviewed-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.tuner |    1 +
 drivers/media/Kconfig                    |    3 +-
 drivers/media/dvb/frontends/Kconfig      |    7 +
 drivers/media/dvb/frontends/Makefile     |    1 +
 drivers/media/dvb/frontends/tda18271.c   | 1054 ++++++++++++++++++++++++++++++
 drivers/media/dvb/frontends/tda18271.h   |   40 ++
 drivers/media/video/tda8290.c            |  319 ++++++++-
 drivers/media/video/tda8290.h            |    8 +
 drivers/media/video/tuner-core.c         |    5 +
 drivers/media/video/tuner-types.c        |    3 +
 drivers/media/video/tveeprom.c           |    2 +-
 include/media/tuner.h                    |    2 +
 12 files changed, 1437 insertions(+), 8 deletions(-)
 create mode 100644 drivers/media/dvb/frontends/tda18271.c
 create mode 100644 drivers/media/dvb/frontends/tda18271.h

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner
index 84c3ac7c33d9..ac47b48715dd 100644
--- a/Documentation/video4linux/CARDLIST.tuner
+++ b/Documentation/video4linux/CARDLIST.tuner
@@ -73,3 +73,4 @@ tuner=71 - Xceive xc2028/xc3028 tuner
 tuner=72 - Thomson FE6600
 tuner=73 - Samsung TCPG 6121P30A
 tuner=75 - Philips TEA5761 FM Radio
+tuner=76 - tda8295+18271
diff --git a/drivers/media/Kconfig b/drivers/media/Kconfig
index d363d0cae38b..e52229380984 100644
--- a/drivers/media/Kconfig
+++ b/drivers/media/Kconfig
@@ -105,9 +105,10 @@ config TUNER_MT20XX
 	  Say Y here to include support for the MT2032 / MT2050 tuner.
 
 config TUNER_TDA8290
-	tristate "TDA 8290+8275(a) tuner combo"
+	tristate "TDA 8290/8295 + 8275(a)/18271 tuner combo"
 	depends on I2C
 	select DVB_TDA827X
+	select DVB_TDA18271
 	default m if VIDEO_TUNER_CUSTOMIZE
 	help
 	  Say Y here to include support for Philips TDA8290+8275(a) tuner.
diff --git a/drivers/media/dvb/frontends/Kconfig b/drivers/media/dvb/frontends/Kconfig
index 59b9ed1f1aec..57178d6e1cf8 100644
--- a/drivers/media/dvb/frontends/Kconfig
+++ b/drivers/media/dvb/frontends/Kconfig
@@ -316,6 +316,13 @@ config DVB_TDA827X
 	help
 	  A DVB-T silicon tuner module. Say Y when you want to support this tuner.
 
+config DVB_TDA18271
+	tristate "NXP TDA18271 silicon tuner"
+	depends on I2C
+	default m if DVB_FE_CUSTOMISE
+	help
+	  A silicon tuner module. Say Y when you want to support this tuner.
+
 config DVB_TUNER_QT1010
 	tristate "Quantek QT1010 silicon tuner"
 	depends on DVB_CORE && I2C
diff --git a/drivers/media/dvb/frontends/Makefile b/drivers/media/dvb/frontends/Makefile
index 4b8ad1f132aa..457effcda5da 100644
--- a/drivers/media/dvb/frontends/Makefile
+++ b/drivers/media/dvb/frontends/Makefile
@@ -39,6 +39,7 @@ obj-$(CONFIG_DVB_ISL6421) += isl6421.o
 obj-$(CONFIG_DVB_TDA10086) += tda10086.o
 obj-$(CONFIG_DVB_TDA826X) += tda826x.o
 obj-$(CONFIG_DVB_TDA827X) += tda827x.o
+obj-$(CONFIG_DVB_TDA18271) += tda18271.o
 obj-$(CONFIG_DVB_TUNER_MT2060) += mt2060.o
 obj-$(CONFIG_DVB_TUNER_MT2266) += mt2266.o
 obj-$(CONFIG_DVB_TUNER_DIB0070) += dib0070.o
diff --git a/drivers/media/dvb/frontends/tda18271.c b/drivers/media/dvb/frontends/tda18271.c
new file mode 100644
index 000000000000..3395f2bda498
--- /dev/null
+++ b/drivers/media/dvb/frontends/tda18271.c
@@ -0,0 +1,1054 @@
+/*
+    tda18271.c - driver for the Philips / NXP TDA18271 silicon tuner
+
+    Copyright (C) 2007 Michael Krufky (mkrufky@linuxtv.org)
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#include <linux/i2c.h>
+#include <linux/delay.h>
+#include <linux/videodev2.h>
+
+#include "tda18271.h"
+
+static int debug;
+module_param(debug, int, 0644);
+MODULE_PARM_DESC(debug, "Turn on/off frontend debugging (default:off).");
+
+#define tuner_dbg(fmt, arg...) do {\
+	if (debug > 0) \
+		printk(KERN_DEBUG fmt, ##arg); } while (0)
+
+#define tuner_extra_dbg(fmt, arg...) do {\
+	if (debug > 1) \
+		printk(KERN_DEBUG fmt, ##arg); } while (0)
+
+#define R_ID     0x00	/* ID byte                */
+#define R_TM     0x01	/* Thermo byte            */
+#define R_PL     0x02	/* Power level byte       */
+#define R_EP1    0x03	/* Easy Prog byte 1       */
+#define R_EP2    0x04	/* Easy Prog byte 2       */
+#define R_EP3    0x05	/* Easy Prog byte 3       */
+#define R_EP4    0x06	/* Easy Prog byte 4       */
+#define R_EP5    0x07	/* Easy Prog byte 5       */
+#define R_CPD    0x08	/* Cal Post-Divider byte  */
+#define R_CD1    0x09	/* Cal Divider byte 1     */
+#define R_CD2    0x0a	/* Cal Divider byte 2     */
+#define R_CD3    0x0b	/* Cal Divider byte 3     */
+#define R_MPD    0x0c	/* Main Post-Divider byte */
+#define R_MD1    0x0d	/* Main Divider byte 1    */
+#define R_MD2    0x0e	/* Main Divider byte 2    */
+#define R_MD3    0x0f	/* Main Divider byte 3    */
+#define R_EB1    0x10	/* Extended byte 1        */
+#define R_EB2    0x11	/* Extended byte 2        */
+#define R_EB3    0x12	/* Extended byte 3        */
+#define R_EB4    0x13	/* Extended byte 4        */
+#define R_EB5    0x14	/* Extended byte 5        */
+#define R_EB6    0x15	/* Extended byte 6        */
+#define R_EB7    0x16	/* Extended byte 7        */
+#define R_EB8    0x17	/* Extended byte 8        */
+#define R_EB9    0x18	/* Extended byte 9        */
+#define R_EB10   0x19	/* Extended byte 10       */
+#define R_EB11   0x1a	/* Extended byte 11       */
+#define R_EB12   0x1b	/* Extended byte 12       */
+#define R_EB13   0x1c	/* Extended byte 13       */
+#define R_EB14   0x1d	/* Extended byte 14       */
+#define R_EB15   0x1e	/* Extended byte 15       */
+#define R_EB16   0x1f	/* Extended byte 16       */
+#define R_EB17   0x20	/* Extended byte 17       */
+#define R_EB18   0x21	/* Extended byte 18       */
+#define R_EB19   0x22	/* Extended byte 19       */
+#define R_EB20   0x23	/* Extended byte 20       */
+#define R_EB21   0x24	/* Extended byte 21       */
+#define R_EB22   0x25	/* Extended byte 22       */
+#define R_EB23   0x26	/* Extended byte 23       */
+
+struct tda18271_pll_map {
+	u32 lomax;
+	u8 pd; /* post div */
+	u8 d;  /*      div */
+};
+
+static struct tda18271_pll_map tda18271_main_pll[] = {
+	{ .lomax =  32000, .pd = 0x5f, .d = 0xf0 },
+	{ .lomax =  35000, .pd = 0x5e, .d = 0xe0 },
+	{ .lomax =  37000, .pd = 0x5d, .d = 0xd0 },
+	{ .lomax =  41000, .pd = 0x5c, .d = 0xc0 },
+	{ .lomax =  44000, .pd = 0x5b, .d = 0xb0 },
+	{ .lomax =  49000, .pd = 0x5a, .d = 0xa0 },
+	{ .lomax =  54000, .pd = 0x59, .d = 0x90 },
+	{ .lomax =  61000, .pd = 0x58, .d = 0x80 },
+	{ .lomax =  65000, .pd = 0x4f, .d = 0x78 },
+	{ .lomax =  70000, .pd = 0x4e, .d = 0x70 },
+	{ .lomax =  75000, .pd = 0x4d, .d = 0x68 },
+	{ .lomax =  82000, .pd = 0x4c, .d = 0x60 },
+	{ .lomax =  89000, .pd = 0x4b, .d = 0x58 },
+	{ .lomax =  98000, .pd = 0x4a, .d = 0x50 },
+	{ .lomax = 109000, .pd = 0x49, .d = 0x48 },
+	{ .lomax = 123000, .pd = 0x48, .d = 0x40 },
+	{ .lomax = 131000, .pd = 0x3f, .d = 0x3c },
+	{ .lomax = 141000, .pd = 0x3e, .d = 0x38 },
+	{ .lomax = 151000, .pd = 0x3d, .d = 0x34 },
+	{ .lomax = 164000, .pd = 0x3c, .d = 0x30 },
+	{ .lomax = 179000, .pd = 0x3b, .d = 0x2c },
+	{ .lomax = 197000, .pd = 0x3a, .d = 0x28 },
+	{ .lomax = 219000, .pd = 0x39, .d = 0x24 },
+	{ .lomax = 246000, .pd = 0x38, .d = 0x20 },
+	{ .lomax = 263000, .pd = 0x2f, .d = 0x1e },
+	{ .lomax = 282000, .pd = 0x2e, .d = 0x1c },
+	{ .lomax = 303000, .pd = 0x2d, .d = 0x1a },
+	{ .lomax = 329000, .pd = 0x2c, .d = 0x18 },
+	{ .lomax = 359000, .pd = 0x2b, .d = 0x16 },
+	{ .lomax = 395000, .pd = 0x2a, .d = 0x14 },
+	{ .lomax = 438000, .pd = 0x29, .d = 0x12 },
+	{ .lomax = 493000, .pd = 0x28, .d = 0x10 },
+	{ .lomax = 526000, .pd = 0x1f, .d = 0x0f },
+	{ .lomax = 564000, .pd = 0x1e, .d = 0x0e },
+	{ .lomax = 607000, .pd = 0x1d, .d = 0x0d },
+	{ .lomax = 658000, .pd = 0x1c, .d = 0x0c },
+	{ .lomax = 718000, .pd = 0x1b, .d = 0x0b },
+	{ .lomax = 790000, .pd = 0x1a, .d = 0x0a },
+	{ .lomax = 877000, .pd = 0x19, .d = 0x09 },
+	{ .lomax = 987000, .pd = 0x18, .d = 0x08 },
+	{ .lomax =      0, .pd = 0x00, .d = 0x00 }, /* end */
+};
+
+static struct tda18271_pll_map tda18271_cal_pll[] = {
+	{ .lomax =   33000, .pd = 0xdd, .d = 0xd0 },
+	{ .lomax =   36000, .pd = 0xdc, .d = 0xc0 },
+	{ .lomax =   40000, .pd = 0xdb, .d = 0xb0 },
+	{ .lomax =   44000, .pd = 0xda, .d = 0xa0 },
+	{ .lomax =   49000, .pd = 0xd9, .d = 0x90 },
+	{ .lomax =   55000, .pd = 0xd8, .d = 0x80 },
+	{ .lomax =   63000, .pd = 0xd3, .d = 0x70 },
+	{ .lomax =   67000, .pd = 0xcd, .d = 0x68 },
+	{ .lomax =   73000, .pd = 0xcc, .d = 0x60 },
+	{ .lomax =   80000, .pd = 0xcb, .d = 0x58 },
+	{ .lomax =   88000, .pd = 0xca, .d = 0x50 },
+	{ .lomax =   98000, .pd = 0xc9, .d = 0x48 },
+	{ .lomax =  110000, .pd = 0xc8, .d = 0x40 },
+	{ .lomax =  126000, .pd = 0xc3, .d = 0x38 },
+	{ .lomax =  135000, .pd = 0xbd, .d = 0x34 },
+	{ .lomax =  147000, .pd = 0xbc, .d = 0x30 },
+	{ .lomax =  160000, .pd = 0xbb, .d = 0x2c },
+	{ .lomax =  176000, .pd = 0xba, .d = 0x28 },
+	{ .lomax =  196000, .pd = 0xb9, .d = 0x24 },
+	{ .lomax =  220000, .pd = 0xb8, .d = 0x20 },
+	{ .lomax =  252000, .pd = 0xb3, .d = 0x1c },
+	{ .lomax =  271000, .pd = 0xad, .d = 0x1a },
+	{ .lomax =  294000, .pd = 0xac, .d = 0x18 },
+	{ .lomax =  321000, .pd = 0xab, .d = 0x16 },
+	{ .lomax =  353000, .pd = 0xaa, .d = 0x14 },
+	{ .lomax =  392000, .pd = 0xa9, .d = 0x12 },
+	{ .lomax =  441000, .pd = 0xa8, .d = 0x10 },
+	{ .lomax =  505000, .pd = 0xa3, .d = 0x0e },
+	{ .lomax =  543000, .pd = 0x9d, .d = 0x0d },
+	{ .lomax =  589000, .pd = 0x9c, .d = 0x0c },
+	{ .lomax =  642000, .pd = 0x9b, .d = 0x0b },
+	{ .lomax =  707000, .pd = 0x9a, .d = 0x0a },
+	{ .lomax =  785000, .pd = 0x99, .d = 0x09 },
+	{ .lomax =  883000, .pd = 0x98, .d = 0x08 },
+	{ .lomax = 1010000, .pd = 0x93, .d = 0x07 },
+	{ .lomax =       0, .pd = 0x00, .d = 0x00 }, /* end */
+};
+
+struct tda18271_map {
+	u32 rfmax;
+	u8  val;
+};
+
+static struct tda18271_map tda18271_bp_filter[] = {
+	{ .rfmax =  62000, .val = 0x00 },
+	{ .rfmax =  84000, .val = 0x01 },
+	{ .rfmax = 100000, .val = 0x02 },
+	{ .rfmax = 140000, .val = 0x03 },
+	{ .rfmax = 170000, .val = 0x04 },
+	{ .rfmax = 180000, .val = 0x05 },
+	{ .rfmax = 865000, .val = 0x06 },
+	{ .rfmax =      0, .val = 0x00 }, /* end */
+};
+
+static struct tda18271_map tda18271_km[] = {
+	{ .rfmax =  61100, .val = 0x74 },
+	{ .rfmax = 350000, .val = 0x40 },
+	{ .rfmax = 720000, .val = 0x30 },
+	{ .rfmax = 865000, .val = 0x40 },
+	{ .rfmax =      0, .val = 0x00 }, /* end */
+};
+
+static struct tda18271_map tda18271_rf_band[] = {
+	{ .rfmax =  47900, .val = 0x00 },
+	{ .rfmax =  61100, .val = 0x01 },
+/*	{ .rfmax = 152600, .val = 0x02 }, */
+	{ .rfmax = 121200, .val = 0x02 },
+	{ .rfmax = 164700, .val = 0x03 },
+	{ .rfmax = 203500, .val = 0x04 },
+	{ .rfmax = 457800, .val = 0x05 },
+	{ .rfmax = 865000, .val = 0x06 },
+	{ .rfmax =      0, .val = 0x00 }, /* end */
+};
+
+static struct tda18271_map tda18271_gain_taper[] = {
+	{ .rfmax =  45400, .val = 0x1f },
+	{ .rfmax =  45800, .val = 0x1e },
+	{ .rfmax =  46200, .val = 0x1d },
+	{ .rfmax =  46700, .val = 0x1c },
+	{ .rfmax =  47100, .val = 0x1b },
+	{ .rfmax =  47500, .val = 0x1a },
+	{ .rfmax =  47900, .val = 0x19 },
+	{ .rfmax =  49600, .val = 0x17 },
+	{ .rfmax =  51200, .val = 0x16 },
+	{ .rfmax =  52900, .val = 0x15 },
+	{ .rfmax =  54500, .val = 0x14 },
+	{ .rfmax =  56200, .val = 0x13 },
+	{ .rfmax =  57800, .val = 0x12 },
+	{ .rfmax =  59500, .val = 0x11 },
+	{ .rfmax =  61100, .val = 0x10 },
+	{ .rfmax =  67600, .val = 0x0d },
+	{ .rfmax =  74200, .val = 0x0c },
+	{ .rfmax =  80700, .val = 0x0b },
+	{ .rfmax =  87200, .val = 0x0a },
+	{ .rfmax =  93800, .val = 0x09 },
+	{ .rfmax = 100300, .val = 0x08 },
+	{ .rfmax = 106900, .val = 0x07 },
+	{ .rfmax = 113400, .val = 0x06 },
+	{ .rfmax = 119900, .val = 0x05 },
+	{ .rfmax = 126500, .val = 0x04 },
+	{ .rfmax = 133000, .val = 0x03 },
+	{ .rfmax = 139500, .val = 0x02 },
+	{ .rfmax = 146100, .val = 0x01 },
+	{ .rfmax = 152600, .val = 0x00 },
+	{ .rfmax = 154300, .val = 0x1f },
+	{ .rfmax = 156100, .val = 0x1e },
+	{ .rfmax = 157800, .val = 0x1d },
+	{ .rfmax = 159500, .val = 0x1c },
+	{ .rfmax = 161200, .val = 0x1b },
+	{ .rfmax = 163000, .val = 0x1a },
+	{ .rfmax = 164700, .val = 0x19 },
+	{ .rfmax = 170200, .val = 0x17 },
+	{ .rfmax = 175800, .val = 0x16 },
+	{ .rfmax = 181300, .val = 0x15 },
+	{ .rfmax = 186900, .val = 0x14 },
+	{ .rfmax = 192400, .val = 0x13 },
+	{ .rfmax = 198000, .val = 0x12 },
+	{ .rfmax = 203500, .val = 0x11 },
+	{ .rfmax = 216200, .val = 0x14 },
+	{ .rfmax = 228900, .val = 0x13 },
+	{ .rfmax = 241600, .val = 0x12 },
+	{ .rfmax = 254400, .val = 0x11 },
+	{ .rfmax = 267100, .val = 0x10 },
+	{ .rfmax = 279800, .val = 0x0f },
+	{ .rfmax = 292500, .val = 0x0e },
+	{ .rfmax = 305200, .val = 0x0d },
+	{ .rfmax = 317900, .val = 0x0c },
+	{ .rfmax = 330700, .val = 0x0b },
+	{ .rfmax = 343400, .val = 0x0a },
+	{ .rfmax = 356100, .val = 0x09 },
+	{ .rfmax = 368800, .val = 0x08 },
+	{ .rfmax = 381500, .val = 0x07 },
+	{ .rfmax = 394200, .val = 0x06 },
+	{ .rfmax = 406900, .val = 0x05 },
+	{ .rfmax = 419700, .val = 0x04 },
+	{ .rfmax = 432400, .val = 0x03 },
+	{ .rfmax = 445100, .val = 0x02 },
+	{ .rfmax = 457800, .val = 0x01 },
+	{ .rfmax = 476300, .val = 0x19 },
+	{ .rfmax = 494800, .val = 0x18 },
+	{ .rfmax = 513300, .val = 0x17 },
+	{ .rfmax = 531800, .val = 0x16 },
+	{ .rfmax = 550300, .val = 0x15 },
+	{ .rfmax = 568900, .val = 0x14 },
+	{ .rfmax = 587400, .val = 0x13 },
+	{ .rfmax = 605900, .val = 0x12 },
+	{ .rfmax = 624400, .val = 0x11 },
+	{ .rfmax = 642900, .val = 0x10 },
+	{ .rfmax = 661400, .val = 0x0f },
+	{ .rfmax = 679900, .val = 0x0e },
+	{ .rfmax = 698400, .val = 0x0d },
+	{ .rfmax = 716900, .val = 0x0c },
+	{ .rfmax = 735400, .val = 0x0b },
+	{ .rfmax = 753900, .val = 0x0a },
+	{ .rfmax = 772500, .val = 0x09 },
+	{ .rfmax = 791000, .val = 0x08 },
+	{ .rfmax = 809500, .val = 0x07 },
+	{ .rfmax = 828000, .val = 0x06 },
+	{ .rfmax = 846500, .val = 0x05 },
+	{ .rfmax = 865000, .val = 0x04 },
+	{ .rfmax =      0, .val = 0x00 }, /* end */
+};
+
+static struct tda18271_map tda18271_rf_cal[] = {
+	{ .rfmax = 41000, .val = 0x1e },
+	{ .rfmax = 43000, .val = 0x30 },
+	{ .rfmax = 45000, .val = 0x43 },
+	{ .rfmax = 46000, .val = 0x4d },
+	{ .rfmax = 47000, .val = 0x54 },
+	{ .rfmax = 47900, .val = 0x64 },
+	{ .rfmax = 49100, .val = 0x20 },
+	{ .rfmax = 50000, .val = 0x22 },
+	{ .rfmax = 51000, .val = 0x2a },
+	{ .rfmax = 53000, .val = 0x32 },
+	{ .rfmax = 55000, .val = 0x35 },
+	{ .rfmax = 56000, .val = 0x3c },
+	{ .rfmax = 57000, .val = 0x3f },
+	{ .rfmax = 58000, .val = 0x48 },
+	{ .rfmax = 59000, .val = 0x4d },
+	{ .rfmax = 60000, .val = 0x58 },
+	{ .rfmax = 61100, .val = 0x5f },
+	{ .rfmax =     0, .val = 0x00 }, /* end */
+};
+
+/*---------------------------------------------------------------------*/
+
+#define TDA18271_NUM_REGS 39
+
+#define TDA18271_ANALOG  0
+#define TDA18271_DIGITAL 1
+
+struct tda18271_priv {
+	u8 i2c_addr;
+	struct i2c_adapter *i2c_adap;
+	unsigned char tda18271_regs[TDA18271_NUM_REGS];
+	int mode;
+
+	u32 frequency;
+	u32 bandwidth;
+};
+
+/*---------------------------------------------------------------------*/
+
+static void tda18271_dump_regs(struct dvb_frontend *fe)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	unsigned char *regs = priv->tda18271_regs;
+
+	tuner_dbg("=== TDA18271 REG DUMP ===\n");
+	tuner_dbg("ID_BYTE            = 0x%x\n", 0xff & regs[R_ID]);
+	tuner_dbg("THERMO_BYTE        = 0x%x\n", 0xff & regs[R_TM]);
+	tuner_dbg("POWER_LEVEL_BYTE   = 0x%x\n", 0xff & regs[R_PL]);
+	tuner_dbg("EASY_PROG_BYTE_1   = 0x%x\n", 0xff & regs[R_EP1]);
+	tuner_dbg("EASY_PROG_BYTE_2   = 0x%x\n", 0xff & regs[R_EP2]);
+	tuner_dbg("EASY_PROG_BYTE_3   = 0x%x\n", 0xff & regs[R_EP3]);
+	tuner_dbg("EASY_PROG_BYTE_4   = 0x%x\n", 0xff & regs[R_EP4]);
+	tuner_dbg("EASY_PROG_BYTE_5   = 0x%x\n", 0xff & regs[R_EP5]);
+	tuner_dbg("CAL_POST_DIV_BYTE  = 0x%x\n", 0xff & regs[R_CPD]);
+	tuner_dbg("CAL_DIV_BYTE_1     = 0x%x\n", 0xff & regs[R_CD1]);
+	tuner_dbg("CAL_DIV_BYTE_2     = 0x%x\n", 0xff & regs[R_CD2]);
+	tuner_dbg("CAL_DIV_BYTE_3     = 0x%x\n", 0xff & regs[R_CD3]);
+	tuner_dbg("MAIN_POST_DIV_BYTE = 0x%x\n", 0xff & regs[R_MPD]);
+	tuner_dbg("MAIN_DIV_BYTE_1    = 0x%x\n", 0xff & regs[R_MD1]);
+	tuner_dbg("MAIN_DIV_BYTE_2    = 0x%x\n", 0xff & regs[R_MD2]);
+	tuner_dbg("MAIN_DIV_BYTE_3    = 0x%x\n", 0xff & regs[R_MD3]);
+}
+
+static void tda18271_read_regs(struct dvb_frontend *fe)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	unsigned char *regs = priv->tda18271_regs;
+	unsigned char buf = 0x00;
+	int ret;
+	struct i2c_msg msg[] = {
+		{ .addr = priv->i2c_addr, .flags = 0,
+		  .buf = &buf, .len = 1 },
+		{ .addr = priv->i2c_addr, .flags = I2C_M_RD,
+		  .buf = regs, .len = 16 }
+	};
+
+	if (fe->ops.i2c_gate_ctrl)
+		fe->ops.i2c_gate_ctrl(fe, 1);
+
+	/* read all registers */
+	ret = i2c_transfer(priv->i2c_adap, msg, 2);
+
+	if (fe->ops.i2c_gate_ctrl)
+		fe->ops.i2c_gate_ctrl(fe, 0);
+
+	if (ret != 2)
+		printk("ERROR: %s: i2c_transfer returned: %d\n",
+		       __FUNCTION__, ret);
+
+	if (debug > 1)
+		tda18271_dump_regs(fe);
+}
+
+static void tda18271_write_regs(struct dvb_frontend *fe, int idx, int len)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	unsigned char *regs = priv->tda18271_regs;
+	unsigned char buf[TDA18271_NUM_REGS+1];
+	struct i2c_msg msg = { .addr = priv->i2c_addr, .flags = 0,
+			       .buf = buf, .len = len+1 };
+	int i, ret;
+
+	BUG_ON((len == 0) || (idx+len > sizeof(buf)));
+
+	buf[0] = idx;
+	for (i = 1; i <= len; i++) {
+		buf[i] = regs[idx-1+i];
+	}
+
+	if (fe->ops.i2c_gate_ctrl)
+		fe->ops.i2c_gate_ctrl(fe, 1);
+
+	/* write registers */
+	ret = i2c_transfer(priv->i2c_adap, &msg, 1);
+
+	if (fe->ops.i2c_gate_ctrl)
+		fe->ops.i2c_gate_ctrl(fe, 0);
+
+	if (ret != 1)
+		printk(KERN_WARNING "ERROR: %s: i2c_transfer returned: %d\n",
+		       __FUNCTION__, ret);
+}
+
+/*---------------------------------------------------------------------*/
+
+static void tda18271_init_regs(struct dvb_frontend *fe)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	unsigned char *regs = priv->tda18271_regs;
+
+	tda18271_read_regs(fe);
+
+	/* test IR_CAL_OK to see if we need init */
+	if ((regs[R_EP1] & 0x08) != 0)
+		return;
+
+	printk(KERN_INFO "tda18271: initializing registers\n");
+
+	/* initialize registers */
+	regs[R_ID]   = 0x83;
+	regs[R_TM]   = 0x08;
+	regs[R_PL]   = 0x80;
+	regs[R_EP1]  = 0xc6;
+	regs[R_EP2]  = 0xdf;
+	regs[R_EP3]  = 0x16;
+	regs[R_EP4]  = 0x60;
+	regs[R_EP5]  = 0x80;
+	regs[R_CPD]  = 0x80;
+	regs[R_CD1]  = 0x00;
+	regs[R_CD2]  = 0x00;
+	regs[R_CD3]  = 0x00;
+	regs[R_MPD]  = 0x00;
+	regs[R_MD1]  = 0x00;
+	regs[R_MD2]  = 0x00;
+	regs[R_MD3]  = 0x00;
+	regs[R_EB1]  = 0xff;
+	regs[R_EB2]  = 0x01;
+	regs[R_EB3]  = 0x84;
+	regs[R_EB4]  = 0x41;
+	regs[R_EB5]  = 0x01;
+	regs[R_EB6]  = 0x84;
+	regs[R_EB7]  = 0x40;
+	regs[R_EB8]  = 0x07;
+	regs[R_EB9]  = 0x00;
+	regs[R_EB10] = 0x00;
+	regs[R_EB11] = 0x96;
+	regs[R_EB12] = 0x0f;
+	regs[R_EB13] = 0xc1;
+	regs[R_EB14] = 0x00;
+	regs[R_EB15] = 0x8f;
+	regs[R_EB16] = 0x00;
+	regs[R_EB17] = 0x00;
+	regs[R_EB18] = 0x00;
+	regs[R_EB19] = 0x00;
+	regs[R_EB20] = 0x20;
+	regs[R_EB21] = 0x33;
+	regs[R_EB22] = 0x48;
+	regs[R_EB23] = 0xb0;
+
+	tda18271_write_regs(fe, 0x00, TDA18271_NUM_REGS);
+	/* setup AGC1 & AGC2 */
+	regs[R_EB17] = 0x00;
+	tda18271_write_regs(fe, R_EB17, 1);
+	regs[R_EB17] = 0x03;
+	tda18271_write_regs(fe, R_EB17, 1);
+	regs[R_EB17] = 0x43;
+	tda18271_write_regs(fe, R_EB17, 1);
+	regs[R_EB17] = 0x4c;
+	tda18271_write_regs(fe, R_EB17, 1);
+
+	regs[R_EB20] = 0xa0;
+	tda18271_write_regs(fe, R_EB20, 1);
+	regs[R_EB20] = 0xa7;
+	tda18271_write_regs(fe, R_EB20, 1);
+	regs[R_EB20] = 0xe7;
+	tda18271_write_regs(fe, R_EB20, 1);
+	regs[R_EB20] = 0xec;
+	tda18271_write_regs(fe, R_EB20, 1);
+
+	/* image rejection calibration */
+
+	/* low-band */
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x81;
+	regs[R_CPD] = 0xcc;
+	regs[R_CD1] = 0x6c;
+	regs[R_CD2] = 0x00;
+	regs[R_CD3] = 0x00;
+	regs[R_MPD] = 0xcd;
+	regs[R_MD1] = 0x77;
+	regs[R_MD2] = 0x08;
+	regs[R_MD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 11);
+	msleep(5); /* pll locking */
+
+	regs[R_EP1] = 0xc6;
+	tda18271_write_regs(fe, R_EP1, 1);
+	msleep(5); /* wanted low measurement */
+
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x85;
+	regs[R_CPD] = 0xcb;
+	regs[R_CD1] = 0x66;
+	regs[R_CD2] = 0x70;
+	regs[R_CD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 7);
+	msleep(5); /* pll locking */
+
+	regs[R_EP2] = 0xdf;
+	tda18271_write_regs(fe, R_EP2, 1);
+	msleep(30); /* image low optimization completion */
+
+	/* mid-band */
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x82;
+	regs[R_CPD] = 0xa8;
+	regs[R_CD1] = 0x66;
+	regs[R_CD2] = 0x00;
+	regs[R_CD3] = 0x00;
+	regs[R_MPD] = 0xa9;
+	regs[R_MD1] = 0x73;
+	regs[R_MD2] = 0x1a;
+	regs[R_MD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 11);
+	msleep(5); /* pll locking */
+
+	regs[R_EP1] = 0xc6;
+	tda18271_write_regs(fe, R_EP1, 1);
+	msleep(5); /* wanted mid measurement */
+
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x86;
+	regs[R_CPD] = 0xa8;
+	regs[R_CD1] = 0x66;
+	regs[R_CD2] = 0xa0;
+	regs[R_CD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 7);
+	msleep(5); /* pll locking */
+
+	regs[R_EP2] = 0xdf;
+	tda18271_write_regs(fe, R_EP2, 1);
+	msleep(30); /* image mid optimization completion */
+
+	/* high-band */
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x83;
+	regs[R_CPD] = 0x98;
+	regs[R_CD1] = 0x65;
+	regs[R_CD2] = 0x00;
+	regs[R_CD3] = 0x00;
+	regs[R_MPD] = 0x99;
+	regs[R_MD1] = 0x71;
+	regs[R_MD2] = 0xcd;
+	regs[R_MD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 11);
+	msleep(5); /* pll locking */
+
+	regs[R_EP1] = 0xc6;
+	tda18271_write_regs(fe, R_EP1, 1);
+	msleep(5); /* wanted high measurement */
+
+	regs[R_EP3] = 0x1f;
+	regs[R_EP4] = 0x66;
+	regs[R_EP5] = 0x87;
+	regs[R_CPD] = 0x98;
+	regs[R_CD1] = 0x65;
+	regs[R_CD2] = 0x50;
+	regs[R_CD3] = 0x00;
+
+	tda18271_write_regs(fe, R_EP3, 7);
+	msleep(5); /* pll locking */
+
+	regs[R_EP2] = 0xdf;
+
+	tda18271_write_regs(fe, R_EP2, 1);
+	msleep(30); /* image high optimization completion */
+
+	regs[R_EP4] = 0x64;
+	tda18271_write_regs(fe, R_EP4, 1);
+
+	regs[R_EP1] = 0xc6;
+	tda18271_write_regs(fe, R_EP1, 1);
+}
+
+static int tda18271_tune(struct dvb_frontend *fe,
+			 u32 ifc, u32 freq, u32 bw, u8 std)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	unsigned char *regs = priv->tda18271_regs;
+	u32 div, N = 0;
+	int i;
+
+
+	tuner_dbg("%s: freq = %d, ifc = %d\n", __FUNCTION__, freq, ifc);
+
+	tda18271_init_regs(fe);
+	/* RF tracking filter calibration */
+
+	/* calculate BP_Filter */
+	i = 0;
+	while ((tda18271_bp_filter[i].rfmax * 1000) < freq) {
+		if (tda18271_bp_filter[i + 1].rfmax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("bp filter = 0x%x, i = %d\n",
+			tda18271_bp_filter[i].val, i);
+
+	regs[R_EP1]  &= ~0x07; /* clear bp filter bits */
+	regs[R_EP1]  |= tda18271_bp_filter[i].val;
+	tda18271_write_regs(fe, R_EP1, 1);
+
+	regs[R_EB4]  &= 0x07;
+	regs[R_EB4]  |= 0x60;
+	tda18271_write_regs(fe, R_EB4, 1);
+
+	regs[R_EB7]   = 0x60;
+	tda18271_write_regs(fe, R_EB7, 1);
+
+	regs[R_EB14]  = 0x00;
+	tda18271_write_regs(fe, R_EB14, 1);
+
+	regs[R_EB20]  = 0xcc;
+	tda18271_write_regs(fe, R_EB20, 1);
+
+	/* set CAL mode to RF tracking filter calibration */
+	regs[R_EB4]  |= 0x03;
+
+	/* calculate CAL PLL */
+
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		N = freq - 1250000;
+		break;
+	case TDA18271_DIGITAL:
+		N = freq + bw / 2;
+		break;
+	}
+
+	i = 0;
+	while ((tda18271_cal_pll[i].lomax * 1000) < N) {
+		if (tda18271_cal_pll[i + 1].lomax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("cal pll, pd = 0x%x, d = 0x%x, i = %d\n",
+			tda18271_cal_pll[i].pd, tda18271_cal_pll[i].d, i);
+
+	regs[R_CPD]   = tda18271_cal_pll[i].pd;
+
+	div =  ((tda18271_cal_pll[i].d * (N / 1000)) << 7) / 125;
+	regs[R_CD1]   = 0xff & (div >> 16);
+	regs[R_CD2]   = 0xff & (div >> 8);
+	regs[R_CD3]   = 0xff & div;
+
+	/* calculate MAIN PLL */
+
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		N = freq - 250000;
+		break;
+	case TDA18271_DIGITAL:
+		N = freq + bw / 2 + 1000000;
+		break;
+	}
+
+	i = 0;
+	while ((tda18271_main_pll[i].lomax * 1000) < N) {
+		if (tda18271_main_pll[i + 1].lomax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("main pll, pd = 0x%x, d = 0x%x, i = %d\n",
+			tda18271_main_pll[i].pd, tda18271_main_pll[i].d, i);
+
+	regs[R_MPD]   = (0x7f & tda18271_main_pll[i].pd);
+
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		regs[R_MPD]  &= ~0x08;
+		break;
+	case TDA18271_DIGITAL:
+		regs[R_MPD]  |=  0x08;
+		break;
+	}
+
+	div =  ((tda18271_main_pll[i].d * (N / 1000)) << 7) / 125;
+	regs[R_MD1]   = 0xff & (div >> 16);
+	regs[R_MD2]   = 0xff & (div >> 8);
+	regs[R_MD3]   = 0xff & div;
+
+	tda18271_write_regs(fe, R_EP3, 11);
+	msleep(5); /* RF tracking filter calibration initialization */
+
+	/* search for K,M,CO for RF Calibration */
+	i = 0;
+	while ((tda18271_km[i].rfmax * 1000) < freq) {
+		if (tda18271_km[i + 1].rfmax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("km = 0x%x, i = %d\n", tda18271_km[i].val, i);
+
+	regs[R_EB13] &= 0x83;
+	regs[R_EB13] |= tda18271_km[i].val;
+	tda18271_write_regs(fe, R_EB13, 1);
+
+	/* search for RF_BAND */
+	i = 0;
+	while ((tda18271_rf_band[i].rfmax * 1000) < freq) {
+		if (tda18271_rf_band[i + 1].rfmax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("rf band = 0x%x, i = %d\n",
+			tda18271_rf_band[i].val, i);
+
+	regs[R_EP2]  &= ~0xe0; /* clear rf band bits */
+	regs[R_EP2]  |= (tda18271_rf_band[i].val << 5);
+
+	/* search for Gain_Taper */
+	i = 0;
+	while ((tda18271_gain_taper[i].rfmax * 1000) < freq) {
+		if (tda18271_gain_taper[i + 1].rfmax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("gain taper = 0x%x, i = %d\n",
+			tda18271_gain_taper[i].val, i);
+
+	regs[R_EP2]  &= ~0x1f; /* clear gain taper bits */
+	regs[R_EP2]  |= tda18271_gain_taper[i].val;
+
+	tda18271_write_regs(fe, R_EP2, 1);
+	tda18271_write_regs(fe, R_EP1, 1);
+	tda18271_write_regs(fe, R_EP2, 1);
+	tda18271_write_regs(fe, R_EP1, 1);
+
+	regs[R_EB4]  &= 0x07;
+	regs[R_EB4]  |= 0x40;
+	tda18271_write_regs(fe, R_EB4, 1);
+
+	regs[R_EB7]   = 0x40;
+	tda18271_write_regs(fe, R_EB7, 1);
+	msleep(10);
+
+	regs[R_EB20]  = 0xec;
+	tda18271_write_regs(fe, R_EB20, 1);
+	msleep(60); /* RF tracking filter calibration completion */
+
+	regs[R_EP4]  &= ~0x03; /* set cal mode to normal */
+	tda18271_write_regs(fe, R_EP4, 1);
+
+	tda18271_write_regs(fe, R_EP1, 1);
+
+	/* RF tracking filer correction for VHF_Low band */
+	i = 0;
+	while ((tda18271_rf_cal[i].rfmax * 1000) < freq) {
+		if (tda18271_rf_cal[i].rfmax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("rf cal = 0x%x, i = %d\n", tda18271_rf_cal[i].val, i);
+
+	/* VHF_Low band only */
+	if (tda18271_rf_cal[i].rfmax != 0) {
+		regs[R_EB14]   = tda18271_rf_cal[i].val;
+		tda18271_write_regs(fe, R_EB14, 1);
+	}
+
+	/* Channel Configuration */
+
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		regs[R_EB22]  = 0x2c;
+		break;
+	case TDA18271_DIGITAL:
+		regs[R_EB22]  = 0x37;
+		break;
+	}
+	tda18271_write_regs(fe, R_EB22, 1);
+
+	regs[R_EP1]  |= 0x40; /* set dis power level on */
+
+	/* set standard */
+	regs[R_EP3]  &= ~0x1f; /* clear std bits */
+
+	/* see table 22 */
+	regs[R_EP3]  |= std;
+
+	/* TO DO:           *
+	 * ================ *
+	 * FM radio,   0x18 *
+	 * ATSC  6MHz, 0x1c *
+	 * DVB-T 6MHz, 0x1c *
+	 * DVB-T 7MHz, 0x1d *
+	 * DVB-T 8MHz, 0x1e *
+	 * QAM   6MHz, 0x1d *
+	 * QAM   8MHz, 0x1f */
+
+	regs[R_EP4]  &= ~0x03; /* set cal mode to normal */
+
+	regs[R_EP4]  &= ~0x1c; /* clear if level bits */
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		regs[R_MPD]  &= ~0x80; /* IF notch = 0 */
+		break;
+	case TDA18271_DIGITAL:
+		regs[R_EP4]  |= 0x04;
+		regs[R_MPD]  |= 0x80;
+		break;
+	}
+
+	regs[R_EP4]  &= ~0x80; /* turn this bit on only for fm */
+
+	/* FIXME: image rejection validity EP5[2:0] */
+
+	/* calculate MAIN PLL */
+	N = freq + ifc;
+
+	i = 0;
+	while ((tda18271_main_pll[i].lomax * 1000) < N) {
+		if (tda18271_main_pll[i + 1].lomax == 0)
+			break;
+		i++;
+	}
+	tuner_extra_dbg("main pll, pd = 0x%x, d = 0x%x, i = %d\n",
+			tda18271_main_pll[i].pd, tda18271_main_pll[i].d, i);
+
+	regs[R_MPD]   = (0x7f & tda18271_main_pll[i].pd);
+	switch (priv->mode) {
+	case TDA18271_ANALOG:
+		regs[R_MPD]  &= ~0x08;
+		break;
+	case TDA18271_DIGITAL:
+		regs[R_MPD]  |= 0x08;
+		break;
+	}
+
+	div =  ((tda18271_main_pll[i].d * (N / 1000)) << 7) / 125;
+	regs[R_MD1]   = 0xff & (div >> 16);
+	regs[R_MD2]   = 0xff & (div >> 8);
+	regs[R_MD3]   = 0xff & div;
+
+	tda18271_write_regs(fe, R_TM, 15);
+	msleep(5);
+	return 0;
+}
+
+/* ------------------------------------------------------------------ */
+
+static int tda18271_set_params(struct dvb_frontend *fe,
+			       struct dvb_frontend_parameters *params)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	u8 std;
+	u32 bw, sgIF = 0;
+
+	u32 freq = params->frequency;
+
+	priv->mode = TDA18271_DIGITAL;
+
+	/* see table 22 */
+	if (fe->ops.info.type == FE_ATSC) {
+		switch (params->u.vsb.modulation) {
+		case VSB_8:
+		case VSB_16:
+			std = 0x1b; /* device-specific (spec says 0x1c) */
+			sgIF = 5380000;
+			break;
+		case QAM_64:
+		case QAM_256:
+			std = 0x18; /* device-specific (spec says 0x1d) */
+			sgIF = 4000000;
+			break;
+		default:
+			printk(KERN_WARNING "%s: modulation not set!\n",
+			       __FUNCTION__);
+			return -EINVAL;
+		}
+		freq += 1750000; /* Adjust to center (+1.75MHZ) */
+		bw = 6000000;
+	} else if (fe->ops.info.type == FE_OFDM) {
+		switch (params->u.ofdm.bandwidth) {
+		case BANDWIDTH_6_MHZ:
+			std = 0x1c;
+			bw = 6000000;
+			break;
+		case BANDWIDTH_7_MHZ:
+			std = 0x1d;
+			bw = 7000000;
+			break;
+		case BANDWIDTH_8_MHZ:
+			std = 0x1e;
+			bw = 8000000;
+			break;
+		default:
+			printk(KERN_WARNING "%s: bandwidth not set!\n",
+			       __FUNCTION__);
+			return -EINVAL;
+		}
+	} else {
+		printk(KERN_WARNING "%s: modulation type not supported!\n",
+		       __FUNCTION__);
+		return -EINVAL;
+	}
+
+	return tda18271_tune(fe, sgIF, freq, bw, std);
+}
+
+static int tda18271_set_analog_params(struct dvb_frontend *fe,
+				      struct analog_parameters *params)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	u8 std;
+	unsigned int sgIF;
+	char *mode;
+
+	priv->mode = TDA18271_ANALOG;
+
+	/* see table 22 */
+	if (params->std & V4L2_STD_MN) {
+		std = 0x0d;
+		sgIF =  92;
+		mode = "MN";
+	} else if (params->std & V4L2_STD_B) {
+		std = 0x0e;
+		sgIF =  108;
+		mode = "B";
+	} else if (params->std & V4L2_STD_GH) {
+		std = 0x0f;
+		sgIF =  124;
+		mode = "GH";
+	} else if (params->std & V4L2_STD_PAL_I) {
+		std = 0x0f;
+		sgIF =  124;
+		mode = "I";
+	} else if (params->std & V4L2_STD_DK) {
+		std = 0x0f;
+		sgIF =  124;
+		mode = "DK";
+	} else if (params->std & V4L2_STD_SECAM_L) {
+		std = 0x0f;
+		sgIF =  124;
+		mode = "L";
+	} else if (params->std & V4L2_STD_SECAM_LC) {
+		std = 0x0f;
+		sgIF =  20;
+		mode = "LC";
+	} else {
+		std = 0x0f;
+		sgIF =  124;
+		mode = "xx";
+	}
+
+	if (params->mode == V4L2_TUNER_RADIO)
+		sgIF =  88; /* if frequency is 5.5 MHz */
+
+	tuner_dbg("setting tda18271 to system %s\n", mode);
+
+	return tda18271_tune(fe, sgIF * 62500, params->frequency * 62500,
+			     0, std);
+}
+
+static int tda18271_release(struct dvb_frontend *fe)
+{
+	kfree(fe->tuner_priv);
+	fe->tuner_priv = NULL;
+	return 0;
+}
+
+static int tda18271_get_frequency(struct dvb_frontend *fe, u32 *frequency)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	*frequency = priv->frequency;
+	return 0;
+}
+
+static int tda18271_get_bandwidth(struct dvb_frontend *fe, u32 *bandwidth)
+{
+	struct tda18271_priv *priv = fe->tuner_priv;
+	*bandwidth = priv->bandwidth;
+	return 0;
+}
+
+static struct dvb_tuner_ops tda18271_tuner_ops = {
+	.info = {
+		.name = "NXP TDA18271HD",
+		.frequency_min  =  45000000,
+		.frequency_max  = 864000000,
+		.frequency_step =     62500
+	},
+	.set_params        = tda18271_set_params,
+	.set_analog_params = tda18271_set_analog_params,
+	.release           = tda18271_release,
+	.get_frequency     = tda18271_get_frequency,
+	.get_bandwidth     = tda18271_get_bandwidth,
+};
+
+struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe, u8 addr,
+				     struct i2c_adapter *i2c)
+{
+	struct tda18271_priv *priv = NULL;
+
+	tuner_dbg("%s:\n", __FUNCTION__);
+	priv = kzalloc(sizeof(struct tda18271_priv), GFP_KERNEL);
+	if (priv == NULL)
+		return NULL;
+
+	priv->i2c_addr = addr;
+	priv->i2c_adap = i2c;
+
+	memcpy(&fe->ops.tuner_ops, &tda18271_tuner_ops,
+	       sizeof(struct dvb_tuner_ops));
+
+	fe->tuner_priv = priv;
+
+	return fe;
+}
+EXPORT_SYMBOL_GPL(tda18271_attach);
+MODULE_DESCRIPTION("NXP TDA18271HD analog / digital tuner driver");
+MODULE_AUTHOR("Michael Krufky <mkrufky@linuxtv.org>");
+MODULE_LICENSE("GPL");
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-basic-offset: 8
+ * End:
+ */
diff --git a/drivers/media/dvb/frontends/tda18271.h b/drivers/media/dvb/frontends/tda18271.h
new file mode 100644
index 000000000000..a8a19a7197f0
--- /dev/null
+++ b/drivers/media/dvb/frontends/tda18271.h
@@ -0,0 +1,40 @@
+/*
+    tda18271.h - header for the Philips / NXP TDA18271 silicon tuner
+
+    Copyright (C) 2007 Michael Krufky (mkrufky@linuxtv.org)
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#ifndef __TDA18271_H__
+#define __TDA18271_H__
+
+#include <linux/i2c.h>
+#include "dvb_frontend.h"
+
+#if defined(CONFIG_DVB_TDA18271) || (defined(CONFIG_DVB_TDA18271_MODULE) && defined(MODULE))
+extern struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe, u8 addr,
+					    struct i2c_adapter *i2c);
+#else
+static inline struct dvb_frontend *tda18271_attach(struct dvb_frontend *fe,
+						   u8 addr,
+						   struct i2c_adapter *i2c)
+{
+	printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__);
+	return NULL;
+}
+#endif
+
+#endif /* __TDA18271_H__ */
diff --git a/drivers/media/video/tda8290.c b/drivers/media/video/tda8290.c
index 5d30cbcd7361..a38ed9db6403 100644
--- a/drivers/media/video/tda8290.c
+++ b/drivers/media/video/tda8290.c
@@ -25,8 +25,9 @@
 #include <linux/videodev.h>
 #include "tda8290.h"
 #include "tda827x.h"
+#include "tda18271.h"
 
-static int tuner_debug = 0;
+static int tuner_debug;
 module_param_named(debug, tuner_debug, int, 0644);
 MODULE_PARM_DESC(debug, "enable verbose debug messages");
 
@@ -65,6 +66,34 @@ static void tda8290_i2c_bridge(struct tuner *t, int close)
 	}
 }
 
+static void tda8295_i2c_bridge(struct tuner *t, int close)
+{
+	struct tda8290_priv *priv = t->priv;
+
+	unsigned char  enable[2] = { 0x45, 0xc1 };
+	unsigned char disable[2] = { 0x46, 0x00 };
+	unsigned char buf[3] = { 0x45, 0x01, 0x00 };
+	unsigned char *msg;
+	if (close) {
+		msg = enable;
+		tuner_i2c_xfer_send(&priv->i2c_props, msg, 2);
+		/* let the bridge stabilize */
+		msleep(20);
+	} else {
+		msg = disable;
+		tuner_i2c_xfer_send(&priv->i2c_props, msg, 1);
+		tuner_i2c_xfer_recv(&priv->i2c_props, &msg[1], 1);
+
+		buf[2] = msg[1];
+		buf[2] &= ~0x04;
+		tuner_i2c_xfer_send(&priv->i2c_props, buf, 3);
+		msleep(5);
+
+		msg[1] |= 0x04;
+		tuner_i2c_xfer_send(&priv->i2c_props, msg, 2);
+	}
+}
+
 /*---------------------------------------------------------------------*/
 
 static void set_audio(struct tuner *t)
@@ -233,6 +262,153 @@ static void tda8290_set_freq(struct tuner *t, unsigned int freq)
 
 /*---------------------------------------------------------------------*/
 
+static void tda8295_power(struct tuner *t, int enable)
+{
+	struct tda8290_priv *priv = t->priv;
+	unsigned char buf[] = { 0x30, 0x00 }; /* clb_stdbt */
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1);
+
+	if (enable)
+		buf[1] = 0x01;
+	else
+		buf[1] = 0x03;
+
+	tuner_i2c_xfer_send(&priv->i2c_props, buf, 2);
+}
+
+static void tda8295_set_easy_mode(struct tuner *t, int enable)
+{
+	struct tda8290_priv *priv = t->priv;
+	unsigned char buf[] = { 0x01, 0x00 };
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1);
+
+	if (enable)
+		buf[1] = 0x01; /* rising edge sets regs 0x02 - 0x23 */
+	else
+		buf[1] = 0x00; /* reset active bit */
+
+	tuner_i2c_xfer_send(&priv->i2c_props, buf, 2);
+}
+
+static void tda8295_set_video_std(struct tuner *t)
+{
+	struct tda8290_priv *priv = t->priv;
+	unsigned char buf[] = { 0x00, priv->tda8290_easy_mode };
+
+	tuner_i2c_xfer_send(&priv->i2c_props, buf, 2);
+
+	tda8295_set_easy_mode(t, 1);
+	msleep(20);
+	tda8295_set_easy_mode(t, 0);
+}
+
+/*---------------------------------------------------------------------*/
+
+static void tda8295_agc1_out(struct tuner *t, int enable)
+{
+	struct tda8290_priv *priv = t->priv;
+	unsigned char buf[] = { 0x02, 0x00 }; /* DIV_FUNC */
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &buf[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &buf[1], 1);
+
+	if (enable)
+		buf[1] &= ~0x40;
+	else
+		buf[1] |= 0x40;
+
+	tuner_i2c_xfer_send(&priv->i2c_props, buf, 2);
+}
+
+static void tda8295_agc2_out(struct tuner *t, int enable)
+{
+	struct tda8290_priv *priv = t->priv;
+	unsigned char set_gpio_cf[]    = { 0x44, 0x00 };
+	unsigned char set_gpio_val[]   = { 0x46, 0x00 };
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &set_gpio_cf[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &set_gpio_cf[1], 1);
+	tuner_i2c_xfer_send(&priv->i2c_props, &set_gpio_val[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &set_gpio_val[1], 1);
+
+	set_gpio_cf[1] &= 0xf0; /* clear GPIO_0 bits 3-0 */
+
+	if (enable) {
+		set_gpio_cf[1]  |= 0x01; /* config GPIO_0 as Open Drain Out */
+		set_gpio_val[1] &= 0xfe; /* set GPIO_0 pin low */
+	}
+	tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_cf, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_val, 2);
+}
+
+static int tda8295_has_signal(struct tuner *t)
+{
+	struct tda8290_priv *priv = t->priv;
+
+	unsigned char hvpll_stat = 0x26;
+	unsigned char ret;
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &hvpll_stat, 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &ret, 1);
+	return (ret & 0x01) ? 65535 : 0;
+}
+
+/*---------------------------------------------------------------------*/
+
+static void tda8295_set_freq(struct tuner *t, unsigned int freq)
+{
+	struct tda8290_priv *priv = t->priv;
+	u16 ifc;
+
+	unsigned char blanking_mode[]     = { 0x1d, 0x00 };
+
+	struct analog_parameters params = {
+		.frequency = freq,
+		.mode      = t->mode,
+		.audmode   = t->audmode,
+		.std       = t->std
+	};
+
+	set_audio(t);
+
+	ifc = priv->cfg.sgIF; /* FIXME */
+
+	tuner_dbg("%s: ifc = %u, freq = %d\n", __FUNCTION__, ifc, freq);
+
+	tda8295_power(t, 1);
+	tda8295_agc1_out(t, 1);
+
+	tuner_i2c_xfer_send(&priv->i2c_props, &blanking_mode[0], 1);
+	tuner_i2c_xfer_recv(&priv->i2c_props, &blanking_mode[1], 1);
+
+	tda8295_set_video_std(t);
+
+	blanking_mode[1] = 0x03;
+	tuner_i2c_xfer_send(&priv->i2c_props, blanking_mode, 2);
+	msleep(20);
+
+	tda8295_i2c_bridge(t, 1);
+
+	if (t->fe.ops.tuner_ops.set_analog_params)
+		t->fe.ops.tuner_ops.set_analog_params(&t->fe, &params);
+
+	if (priv->cfg.agcf)
+		priv->cfg.agcf(&t->fe);
+
+	if (tda8295_has_signal(t))
+		tuner_dbg("tda8295 is locked\n");
+	else
+		tuner_dbg("tda8295 not locked, no signal?\n");
+
+	tda8295_i2c_bridge(t, 0);
+}
+
+/*---------------------------------------------------------------------*/
+
 static int tda8290_has_signal(struct tuner *t)
 {
 	struct tda8290_priv *priv = t->priv;
@@ -264,6 +440,13 @@ static void tda8290_standby(struct tuner *t)
 	tuner_i2c_xfer_send(&priv->i2c_props, tda8290_standby, 2);
 }
 
+static void tda8295_standby(struct tuner *t)
+{
+	tda8295_agc1_out(t, 0); /* Put AGC in tri-state */
+
+	tda8295_power(t, 0);
+}
+
 static void tda8290_init_if(struct tuner *t)
 {
 	struct tda8290_priv *priv = t->priv;
@@ -279,6 +462,35 @@ static void tda8290_init_if(struct tuner *t)
 	tuner_i2c_xfer_send(&priv->i2c_props, set_VS, 2);
 }
 
+static void tda8295_init_if(struct tuner *t)
+{
+	struct tda8290_priv *priv = t->priv;
+
+	static unsigned char set_adc_ctl[]       = { 0x33, 0x14 };
+	static unsigned char set_adc_ctl2[]      = { 0x34, 0x00 };
+	static unsigned char set_pll_reg6[]      = { 0x3e, 0x63 };
+	static unsigned char set_pll_reg0[]      = { 0x38, 0x23 };
+	static unsigned char set_pll_reg7[]      = { 0x3f, 0x01 };
+	static unsigned char set_pll_reg10[]     = { 0x42, 0x61 };
+	static unsigned char set_gpio_reg0[]     = { 0x44, 0x0b };
+
+	tda8295_power(t, 1);
+
+	tda8295_set_easy_mode(t, 0);
+	tda8295_set_video_std(t);
+
+	tuner_i2c_xfer_send(&priv->i2c_props, set_adc_ctl, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_adc_ctl2, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg6, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg0, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg7, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_pll_reg10, 2);
+	tuner_i2c_xfer_send(&priv->i2c_props, set_gpio_reg0, 2);
+
+	tda8295_agc1_out(t, 0);
+	tda8295_agc2_out(t, 0);
+}
+
 static void tda8290_init_tuner(struct tuner *t)
 {
 	struct tda8290_priv *priv = t->priv;
@@ -298,7 +510,7 @@ static void tda8290_init_tuner(struct tuner *t)
 
 /*---------------------------------------------------------------------*/
 
-static void tda8290_release(struct tuner *t)
+static void tda829x_release(struct tuner *t)
 {
 	if (t->fe.ops.tuner_ops.release)
 		t->fe.ops.tuner_ops.release(&t->fe);
@@ -312,7 +524,15 @@ static struct tuner_operations tda8290_tuner_ops = {
 	.set_radio_freq = tda8290_set_freq,
 	.has_signal     = tda8290_has_signal,
 	.standby        = tda8290_standby,
-	.release        = tda8290_release,
+	.release        = tda829x_release,
+};
+
+static struct tuner_operations tda8295_tuner_ops = {
+	.set_tv_freq    = tda8295_set_freq,
+	.set_radio_freq = tda8295_set_freq,
+	.has_signal     = tda8295_has_signal,
+	.standby        = tda8295_standby,
+	.release        = tda829x_release,
 };
 
 int tda8290_attach(struct tuner *t)
@@ -403,6 +623,95 @@ int tda8290_attach(struct tuner *t)
 	tda8290_init_if(t);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(tda8290_attach);
+
+int tda8295_attach(struct tuner *t)
+{
+	struct tda8290_priv *priv = NULL;
+	u8 data;
+	int i, ret, tuners_found;
+	u32 tuner_addrs;
+	struct i2c_msg msg = { .flags = I2C_M_RD, .buf = &data, .len = 1 };
+
+	priv = kzalloc(sizeof(struct tda8290_priv), GFP_KERNEL);
+	if (priv == NULL)
+		return -ENOMEM;
+	t->priv = priv;
+
+	priv->i2c_props.addr     = t->i2c.addr;
+	priv->i2c_props.adap     = t->i2c.adapter;
+
+	tda8295_i2c_bridge(t, 1);
+	/* probe for tuner chip */
+	tuners_found = 0;
+	tuner_addrs = 0;
+	for (i = 0x60; i <= 0x63; i++) {
+		msg.addr = i;
+		ret = i2c_transfer(priv->i2c_props.adap, &msg, 1);
+		if (ret == 1) {
+			tuners_found++;
+			tuner_addrs = (tuner_addrs << 8) + i;
+		}
+	}
+	/* if there is more than one tuner, we expect the right one is
+	   behind the bridge and we choose the highest address that doesn't
+	   give a response now
+	 */
+	tda8295_i2c_bridge(t, 0);
+	if (tuners_found > 1)
+		for (i = 0; i < tuners_found; i++) {
+			msg.addr = tuner_addrs  & 0xff;
+			ret = i2c_transfer(priv->i2c_props.adap, &msg, 1);
+			if (ret == 1)
+				tuner_addrs = tuner_addrs >> 8;
+			else
+				break;
+		}
+	if (tuner_addrs == 0) {
+		tuner_addrs = 0x60;
+		tuner_info("could not clearly identify tuner address, "
+			   "defaulting to %x\n", tuner_addrs);
+	} else {
+		tuner_addrs = tuner_addrs & 0xff;
+		tuner_info("setting tuner address to %x\n", tuner_addrs);
+	}
+	priv->tda827x_addr = tuner_addrs;
+	msg.addr = tuner_addrs;
+
+	tda8295_i2c_bridge(t, 1);
+	ret = i2c_transfer(priv->i2c_props.adap, &msg, 1);
+	tda8295_i2c_bridge(t, 0);
+	if (ret != 1)
+		tuner_warn("TDA827x access failed!\n");
+	if ((data & 0x3c) == 0) {
+		strlcpy(t->i2c.name, "tda8295+18271", sizeof(t->i2c.name));
+		tda18271_attach(&t->fe, priv->tda827x_addr,
+				priv->i2c_props.adap);
+		priv->tda827x_ver = 4;
+	} else {
+		strlcpy(t->i2c.name, "tda8295+75a", sizeof(t->i2c.name));
+		tda827x_attach(&t->fe, priv->tda827x_addr,
+			       priv->i2c_props.adap, &priv->cfg);
+
+		/* FIXME: tda827x module doesn't probe the tuner until
+		 * tda827x_initial_sleep is called
+		 */
+		if (t->fe.ops.tuner_ops.sleep)
+			t->fe.ops.tuner_ops.sleep(&t->fe);
+		priv->tda827x_ver = 2;
+	}
+	priv->tda827x_ver |= 1; /* signifies 8295 vs 8290 */
+	tuner_info("type set to %s\n", t->i2c.name);
+
+	memcpy(&t->ops, &tda8295_tuner_ops, sizeof(struct tuner_operations));
+
+	priv->cfg.tda827x_lpsel = 0;
+	t->mode = V4L2_TUNER_ANALOG_TV;
+
+	tda8295_init_if(t);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(tda8295_attach);
 
 int tda8290_probe(struct tuner *t)
 {
@@ -434,12 +743,10 @@ int tda8290_probe(struct tuner *t)
 	tuner_i2c_xfer_send(&i2c_props, restore_9886, 3);
 	return -1;
 }
-
 EXPORT_SYMBOL_GPL(tda8290_probe);
-EXPORT_SYMBOL_GPL(tda8290_attach);
 
 MODULE_DESCRIPTION("Philips TDA8290 + TDA8275 / TDA8275a tuner driver");
-MODULE_AUTHOR("Gerd Knorr, Hartmut Hackmann");
+MODULE_AUTHOR("Gerd Knorr, Hartmut Hackmann, Michael Krufky");
 MODULE_LICENSE("GPL");
 
 /*
diff --git a/drivers/media/video/tda8290.h b/drivers/media/video/tda8290.h
index 9b63e62b3a0f..dbbcb0f001e0 100644
--- a/drivers/media/video/tda8290.h
+++ b/drivers/media/video/tda8290.h
@@ -24,6 +24,7 @@
 extern int tda8290_probe(struct tuner *t);
 
 extern int tda8290_attach(struct tuner *t);
+extern int tda8295_attach(struct tuner *t);
 #else
 static inline int tda8290_probe(struct tuner *t)
 {
@@ -37,6 +38,13 @@ static inline int tda8290_attach(struct tuner *t)
 	       __FUNCTION__);
 	return -EINVAL;
 }
+
+static inline int tda8295_attach(struct tuner *t)
+{
+	printk(KERN_INFO "%s: not probed - driver disabled by Kconfig\n",
+	       __FUNCTION__);
+	return -EINVAL;
+}
 #endif
 
 #endif /* __TDA8290_H__ */
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 99558c708ae7..6ae8cb205d8e 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -278,6 +278,11 @@ static void set_type(struct i2c_client *c, unsigned int type,
 		tda8290_attach(t);
 		break;
 	}
+	case TUNER_PHILIPS_TDA8295:
+	{
+		tda8295_attach(t);
+		break;
+	}
 	case TUNER_TEA5767:
 		if (tea5767_attach(&t->fe, t->i2c.adapter, t->i2c.addr) == NULL) {
 			t->type = TUNER_ABSENT;
diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c
index ac363f019227..4a674c436cf4 100644
--- a/drivers/media/video/tuner-types.c
+++ b/drivers/media/video/tuner-types.c
@@ -1475,6 +1475,9 @@ struct tunertype tuners[] = {
 		.name   = "Philips TEA5761 FM Radio",
 		/* see tea5767.c for details */
 	},
+	[TUNER_PHILIPS_TDA8295] = { /* Philips PAL|NTSC */
+		.name   = "tda8295+18271",
+		/* see tda8290.c for details */ },
 };
 
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c
index 4b2c4034f5b3..27bfe6d3a33c 100644
--- a/drivers/media/video/tveeprom.c
+++ b/drivers/media/video/tveeprom.c
@@ -257,7 +257,7 @@ hauppauge_tuner[] =
 	{ TUNER_ABSENT,        "LG TAPQ_H702F"},
 	{ TUNER_ABSENT,        "TCL M09WPP_4N_E"},
 	{ TUNER_ABSENT,        "MaxLinear MXL5005_v2"},
-	{ TUNER_ABSENT,        "Philips 18271_8295"},
+	{ TUNER_PHILIPS_TDA8295, "Philips 18271_8295"},
 };
 
 static struct HAUPPAUGE_AUDIOIC
diff --git a/include/media/tuner.h b/include/media/tuner.h
index d49392d90e56..faacd2d50b72 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -124,6 +124,8 @@ extern int tuner_debug;
 #define TUNER_TDA9887                   74      /* This tuner should be used only internally */
 #define TUNER_TEA5761			75	/* Only FM Radio Tuner */
 
+#define TUNER_PHILIPS_TDA8295           76
+
 /* tv card specific */
 #define TDA9887_PRESENT 		(1<<0)
 #define TDA9887_PORT1_INACTIVE 		(1<<1)
-- 
cgit v1.2.3


From 8ffbc6559493c64d6194c92d856196fdaeb8a5fb Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Wed, 12 Sep 2007 08:32:50 -0300
Subject: V4L/DVB (6451): v4l2: add support for bus-based I2C drivers

Two new headers were added: one for I2C drivers that are only used
by V4L2 drivers converted to the new bus-based I2C API, and one that
can be used by both converted and unconverted drivers (at the expense of
some additional overhead).

To support the legacy I2C API a helper function was added to v4l2-common.c.

These headers take care of all the 'boilerplate' code that all V4L2 I2C drivers
have in common and will automatically support the bus-based I2C API introduced
in kernel 2.6.22.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/v4l2-common.c   |  31 +++++++++
 include/media/v4l2-common.h         |  11 +++
 include/media/v4l2-i2c-drv-legacy.h | 132 ++++++++++++++++++++++++++++++++++++
 include/media/v4l2-i2c-drv.h        |  61 +++++++++++++++++
 4 files changed, 235 insertions(+)
 create mode 100644 include/media/v4l2-i2c-drv-legacy.h
 create mode 100644 include/media/v4l2-i2c-drv.h

(limited to 'include')

diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index a4d68d3ba5e4..32607a612b18 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -1013,6 +1013,35 @@ int v4l2_chip_match_host(u32 match_type, u32 match_chip)
 
 /* ----------------------------------------------------------------- */
 
+/* Helper function for I2C legacy drivers */
+
+int v4l2_i2c_attach(struct i2c_adapter *adapter, int address, struct i2c_driver *driver,
+		const char *name, int (*probe)(struct i2c_client *))
+{
+	struct i2c_client *client;
+	int err;
+
+	client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL);
+	if (client == 0)
+		return -ENOMEM;
+
+	client->addr = address;
+	client->adapter = adapter;
+	client->driver = driver;
+	snprintf(client->name, sizeof(client->name) - 1, name);
+
+	err = probe(client);
+	if (err == 0) {
+		i2c_attach_client(client);
+	}
+	else {
+		kfree(client);
+	}
+	return err;
+}
+
+/* ----------------------------------------------------------------- */
+
 EXPORT_SYMBOL(v4l2_norm_to_name);
 EXPORT_SYMBOL(v4l2_video_std_construct);
 
@@ -1038,6 +1067,8 @@ EXPORT_SYMBOL(v4l2_chip_match_i2c_client);
 EXPORT_SYMBOL(v4l2_chip_ident_i2c_client);
 EXPORT_SYMBOL(v4l2_chip_match_host);
 
+EXPORT_SYMBOL(v4l2_i2c_attach);
+
 /*
  * Local variables:
  * c-basic-offset: 8
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index c019c1e99895..475d0d8275e0 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -104,6 +104,17 @@ int v4l2_chip_match_host(u32 id_type, u32 chip_id);
 
 /* ------------------------------------------------------------------------- */
 
+/* Helper function for I2C legacy drivers */
+
+struct i2c_driver;
+struct i2c_adapter;
+struct i2c_client;
+
+int v4l2_i2c_attach(struct i2c_adapter *adapter, int address, struct i2c_driver *driver,
+		const char *name, int (*probe)(struct i2c_client *));
+
+/* ------------------------------------------------------------------------- */
+
 /* Internal ioctls */
 
 /* VIDIOC_INT_DECODE_VBI_LINE */
diff --git a/include/media/v4l2-i2c-drv-legacy.h b/include/media/v4l2-i2c-drv-legacy.h
new file mode 100644
index 000000000000..c059b32844c1
--- /dev/null
+++ b/include/media/v4l2-i2c-drv-legacy.h
@@ -0,0 +1,132 @@
+/*
+ * v4l2-i2c-drv-legacy.h - contains I2C handling code that's identical
+ *		    for all V4L2 I2C drivers. Use this header if the
+ *		    I2C driver is used by both legacy drivers and
+ *		    drivers converted to the bus-based I2C API.
+ *
+ * Copyright (C) 2007 Hans Verkuil <hverkuil@xs4all.nl>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+struct v4l2_i2c_driver_data {
+	const char * const name;
+	int driverid;
+	int (*command)(struct i2c_client *client, unsigned int cmd, void *arg);
+	int (*probe)(struct i2c_client *client);
+	int (*remove)(struct i2c_client *client);
+	int (*suspend)(struct i2c_client *client, pm_message_t state);
+	int (*resume)(struct i2c_client *client);
+	int legacy_class;
+};
+
+static struct v4l2_i2c_driver_data v4l2_i2c_data;
+static struct i2c_client_address_data addr_data;
+static struct i2c_driver v4l2_i2c_driver_legacy;
+static char v4l2_i2c_drv_name_legacy[32];
+
+static int v4l2_i2c_drv_attach_legacy(struct i2c_adapter *adapter, int address, int kind)
+{
+	return v4l2_i2c_attach(adapter, address, &v4l2_i2c_driver_legacy,
+			v4l2_i2c_drv_name_legacy, v4l2_i2c_data.probe);
+}
+
+static int v4l2_i2c_drv_probe_legacy(struct i2c_adapter *adapter)
+{
+	if (adapter->class & v4l2_i2c_data.legacy_class)
+		return i2c_probe(adapter, &addr_data, v4l2_i2c_drv_attach_legacy);
+	return 0;
+}
+
+static int v4l2_i2c_drv_detach_legacy(struct i2c_client *client)
+{
+	int err = i2c_detach_client(client);
+
+	if (err)
+		return err;
+	if (v4l2_i2c_data.remove)
+		v4l2_i2c_data.remove(client);
+	kfree(client);
+
+	return 0;
+}
+
+static int v4l2_i2c_drv_suspend_helper(struct i2c_client *client, pm_message_t state)
+{
+	return v4l2_i2c_data.suspend ? v4l2_i2c_data.suspend(client, state) : 0;
+}
+
+static int v4l2_i2c_drv_resume_helper(struct i2c_client *client)
+{
+	return v4l2_i2c_data.resume ? v4l2_i2c_data.resume(client) : 0;
+}
+
+/* ----------------------------------------------------------------------- */
+
+/* i2c implementation */
+static struct i2c_driver v4l2_i2c_driver_legacy = {
+	.driver = {
+		.owner = THIS_MODULE,
+	},
+	.attach_adapter = v4l2_i2c_drv_probe_legacy,
+	.detach_client = v4l2_i2c_drv_detach_legacy,
+	.suspend = v4l2_i2c_drv_suspend_helper,
+	.resume  = v4l2_i2c_drv_resume_helper,
+};
+
+/* ----------------------------------------------------------------------- */
+
+/* i2c implementation */
+static struct i2c_driver v4l2_i2c_driver = {
+	.suspend 	= v4l2_i2c_drv_suspend_helper,
+	.resume  	= v4l2_i2c_drv_resume_helper,
+};
+
+static int __init v4l2_i2c_drv_init(void)
+{
+	int err;
+
+	strlcpy(v4l2_i2c_drv_name_legacy, v4l2_i2c_data.name, sizeof(v4l2_i2c_drv_name_legacy));
+	strlcat(v4l2_i2c_drv_name_legacy, "'", sizeof(v4l2_i2c_drv_name_legacy));
+
+	if (v4l2_i2c_data.legacy_class == 0)
+		v4l2_i2c_data.legacy_class = I2C_CLASS_TV_ANALOG;
+
+	v4l2_i2c_driver_legacy.driver.name = v4l2_i2c_drv_name_legacy;
+	v4l2_i2c_driver_legacy.id = v4l2_i2c_data.driverid;
+	v4l2_i2c_driver_legacy.command = v4l2_i2c_data.command;
+	err = i2c_add_driver(&v4l2_i2c_driver_legacy);
+
+	if (err)
+		return err;
+	v4l2_i2c_driver.driver.name = v4l2_i2c_data.name;
+	v4l2_i2c_driver.id = v4l2_i2c_data.driverid;
+	v4l2_i2c_driver.command = v4l2_i2c_data.command;
+	v4l2_i2c_driver.probe = v4l2_i2c_data.probe;
+	v4l2_i2c_driver.remove = v4l2_i2c_data.remove;
+	err = i2c_add_driver(&v4l2_i2c_driver);
+	if (err)
+		i2c_del_driver(&v4l2_i2c_driver_legacy);
+	return err;
+}
+
+static void __exit v4l2_i2c_drv_cleanup(void)
+{
+	i2c_del_driver(&v4l2_i2c_driver_legacy);
+	i2c_del_driver(&v4l2_i2c_driver);
+}
+
+module_init(v4l2_i2c_drv_init);
+module_exit(v4l2_i2c_drv_cleanup);
diff --git a/include/media/v4l2-i2c-drv.h b/include/media/v4l2-i2c-drv.h
new file mode 100644
index 000000000000..547db60d3640
--- /dev/null
+++ b/include/media/v4l2-i2c-drv.h
@@ -0,0 +1,61 @@
+/*
+ * v4l2-i2c-drv.h - contains I2C handling code that's identical for
+ *		    all V4L2 I2C drivers. Use this header if the
+ *		    I2C driver is only used by drivers converted
+ *		    to the bus-based I2C API.
+ *
+ * Copyright (C) 2007 Hans Verkuil <hverkuil@xs4all.nl>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+struct v4l2_i2c_driver_data {
+	const char * const name;
+	int driverid;
+	int (*command)(struct i2c_client *client, unsigned int cmd, void *arg);
+	int (*probe)(struct i2c_client *client);
+	int (*remove)(struct i2c_client *client);
+	int (*suspend)(struct i2c_client *client, pm_message_t state);
+	int (*resume)(struct i2c_client *client);
+	int legacy_class;
+};
+
+static struct v4l2_i2c_driver_data v4l2_i2c_data;
+static struct i2c_client_address_data addr_data;
+static struct i2c_driver v4l2_i2c_driver;
+
+
+/* Bus-based I2C implementation for kernels >= 2.6.22 */
+
+static int __init v4l2_i2c_drv_init(void)
+{
+	v4l2_i2c_driver.driver.name = v4l2_i2c_data.name;
+	v4l2_i2c_driver.id = v4l2_i2c_data.driverid;
+	v4l2_i2c_driver.command = v4l2_i2c_data.command;
+	v4l2_i2c_driver.probe = v4l2_i2c_data.probe;
+	v4l2_i2c_driver.remove = v4l2_i2c_data.remove;
+	v4l2_i2c_driver.suspend = v4l2_i2c_data.suspend;
+	v4l2_i2c_driver.resume = v4l2_i2c_data.resume;
+	return i2c_add_driver(&v4l2_i2c_driver);
+}
+
+
+static void __exit v4l2_i2c_drv_cleanup(void)
+{
+	i2c_del_driver(&v4l2_i2c_driver);
+}
+
+module_init(v4l2_i2c_drv_init);
+module_exit(v4l2_i2c_drv_cleanup);
-- 
cgit v1.2.3


From 88307eb3c69c80a705072e68463d8f72005fc027 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Fri, 14 Sep 2007 04:49:16 -0300
Subject: V4L/DVB (6460): v4l2-i2c-drv: add legacy_probe function pointer

Some devices do complicated tests whether the device can be probed or not.
Add a legacy_probe function pointer to support that.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/v4l2-i2c-drv-legacy.h | 6 ++++++
 include/media/v4l2-i2c-drv.h        | 1 +
 2 files changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/media/v4l2-i2c-drv-legacy.h b/include/media/v4l2-i2c-drv-legacy.h
index c059b32844c1..433c984eefa5 100644
--- a/include/media/v4l2-i2c-drv-legacy.h
+++ b/include/media/v4l2-i2c-drv-legacy.h
@@ -29,6 +29,7 @@ struct v4l2_i2c_driver_data {
 	int (*remove)(struct i2c_client *client);
 	int (*suspend)(struct i2c_client *client, pm_message_t state);
 	int (*resume)(struct i2c_client *client);
+	int (*legacy_probe)(struct i2c_adapter *adapter);
 	int legacy_class;
 };
 
@@ -45,6 +46,11 @@ static int v4l2_i2c_drv_attach_legacy(struct i2c_adapter *adapter, int address,
 
 static int v4l2_i2c_drv_probe_legacy(struct i2c_adapter *adapter)
 {
+	if (v4l2_i2c_data.legacy_probe) {
+		if (v4l2_i2c_data.legacy_probe(adapter))
+			return i2c_probe(adapter, &addr_data, v4l2_i2c_drv_attach_legacy);
+		return 0;
+	}
 	if (adapter->class & v4l2_i2c_data.legacy_class)
 		return i2c_probe(adapter, &addr_data, v4l2_i2c_drv_attach_legacy);
 	return 0;
diff --git a/include/media/v4l2-i2c-drv.h b/include/media/v4l2-i2c-drv.h
index 547db60d3640..bc61ab27779d 100644
--- a/include/media/v4l2-i2c-drv.h
+++ b/include/media/v4l2-i2c-drv.h
@@ -29,6 +29,7 @@ struct v4l2_i2c_driver_data {
 	int (*remove)(struct i2c_client *client);
 	int (*suspend)(struct i2c_client *client, pm_message_t state);
 	int (*resume)(struct i2c_client *client);
+	int (*legacy_probe)(struct i2c_adapter *adapter);
 	int legacy_class;
 };
 
-- 
cgit v1.2.3


From a2e521e85c04af3511af3ea8971efb4ef8bde533 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Mon, 17 Sep 2007 05:13:45 -0300
Subject: V4L/DVB (6466): v4l2-i2c-drv: first call remove, then detach client

The remove driver function expects that the client is still attached
to the driver, so do the detach after calling remove().

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/v4l2-i2c-drv-legacy.h | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/media/v4l2-i2c-drv-legacy.h b/include/media/v4l2-i2c-drv-legacy.h
index 433c984eefa5..241854229d6f 100644
--- a/include/media/v4l2-i2c-drv-legacy.h
+++ b/include/media/v4l2-i2c-drv-legacy.h
@@ -58,12 +58,14 @@ static int v4l2_i2c_drv_probe_legacy(struct i2c_adapter *adapter)
 
 static int v4l2_i2c_drv_detach_legacy(struct i2c_client *client)
 {
-	int err = i2c_detach_client(client);
+	int err;
 
-	if (err)
-		return err;
 	if (v4l2_i2c_data.remove)
 		v4l2_i2c_data.remove(client);
+
+	err = i2c_detach_client(client);
+	if (err)
+		return err;
 	kfree(client);
 
 	return 0;
-- 
cgit v1.2.3


From 0b5f4a12e628893624b478523fb4e2708c605710 Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@linuxtv.org>
Date: Sat, 27 Oct 2007 13:09:16 -0300
Subject: V4L/DVB (6469): tuner: remove TUNER_PHILIPS_TDA8295

TUNER_PHILIPS_TDA8290 will autodetect a TDA8290 or a TDA8295,
so we don't need this separate entry anymore.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.tuner | 3 +--
 drivers/media/video/tuner-core.c         | 1 -
 drivers/media/video/tuner-types.c        | 5 +----
 drivers/media/video/tveeprom.c           | 2 +-
 include/media/tuner.h                    | 2 --
 5 files changed, 3 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner
index ac47b48715dd..2211c8eac643 100644
--- a/Documentation/video4linux/CARDLIST.tuner
+++ b/Documentation/video4linux/CARDLIST.tuner
@@ -52,7 +52,7 @@ tuner=50 - TCL 2002N
 tuner=51 - Philips PAL/SECAM_D (FM 1256 I-H3)
 tuner=52 - Thomson DTT 7610 (ATSC/NTSC)
 tuner=53 - Philips FQ1286
-tuner=54 - tda8290+75
+tuner=54 - Philips/NXP TDA 8290/8295 + 8275/8275A/18271
 tuner=55 - TCL 2002MB
 tuner=56 - Philips PAL/SECAM multi (FQ1216AME MK4)
 tuner=57 - Philips FQ1236A MK4
@@ -73,4 +73,3 @@ tuner=71 - Xceive xc2028/xc3028 tuner
 tuner=72 - Thomson FE6600
 tuner=73 - Samsung TCPG 6121P30A
 tuner=75 - Philips TEA5761 FM Radio
-tuner=76 - tda8295+18271
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 6ab57ec3f5a3..600174239d09 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -295,7 +295,6 @@ static void set_type(struct i2c_client *c, unsigned int type,
 		microtune_attach(&t->fe, t->i2c.adapter, t->i2c.addr);
 		break;
 	case TUNER_PHILIPS_TDA8290:
-	case TUNER_PHILIPS_TDA8295:
 	{
 		tda829x_attach(t);
 		break;
diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c
index 4a674c436cf4..e2cd05a05802 100644
--- a/drivers/media/video/tuner-types.c
+++ b/drivers/media/video/tuner-types.c
@@ -1366,7 +1366,7 @@ struct tunertype tuners[] = {
 		.count  = ARRAY_SIZE(tuner_philips_fq1286_params),
 	},
 	[TUNER_PHILIPS_TDA8290] = { /* Philips PAL|NTSC */
-		.name   = "tda8290+75",
+		.name   = "Philips/NXP TDA 8290/8295 + 8275/8275A/18271",
 		/* see tda8290.c for details */ },
 	[TUNER_TCL_2002MB] = { /* TCL PAL */
 		.name   = "TCL 2002MB",
@@ -1475,9 +1475,6 @@ struct tunertype tuners[] = {
 		.name   = "Philips TEA5761 FM Radio",
 		/* see tea5767.c for details */
 	},
-	[TUNER_PHILIPS_TDA8295] = { /* Philips PAL|NTSC */
-		.name   = "tda8295+18271",
-		/* see tda8290.c for details */ },
 };
 
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c
index 27bfe6d3a33c..0faa14992353 100644
--- a/drivers/media/video/tveeprom.c
+++ b/drivers/media/video/tveeprom.c
@@ -257,7 +257,7 @@ hauppauge_tuner[] =
 	{ TUNER_ABSENT,        "LG TAPQ_H702F"},
 	{ TUNER_ABSENT,        "TCL M09WPP_4N_E"},
 	{ TUNER_ABSENT,        "MaxLinear MXL5005_v2"},
-	{ TUNER_PHILIPS_TDA8295, "Philips 18271_8295"},
+	{ TUNER_PHILIPS_TDA8290, "Philips 18271_8295"},
 };
 
 static struct HAUPPAUGE_AUDIOIC
diff --git a/include/media/tuner.h b/include/media/tuner.h
index faacd2d50b72..d49392d90e56 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -124,8 +124,6 @@ extern int tuner_debug;
 #define TUNER_TDA9887                   74      /* This tuner should be used only internally */
 #define TUNER_TEA5761			75	/* Only FM Radio Tuner */
 
-#define TUNER_PHILIPS_TDA8295           76
-
 /* tv card specific */
 #define TDA9887_PRESENT 		(1<<0)
 #define TDA9887_PORT1_INACTIVE 		(1<<1)
-- 
cgit v1.2.3


From 761dacd25d9e9be2358df21a0f086bbb4cf1c789 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 30 Oct 2007 05:41:25 -0300
Subject: V4L/DVB (6486): m52790: add new Mitsubishi A/V switch i2c driver

This driver is used by the ASUS Falcon2 cx23416-based cards.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/Kconfig     |   9 +++
 drivers/media/video/Makefile    |   1 +
 drivers/media/video/m52790.c    | 172 ++++++++++++++++++++++++++++++++++++++++
 include/media/m52790.h          |  93 ++++++++++++++++++++++
 include/media/v4l2-chip-ident.h |   3 +
 5 files changed, 278 insertions(+)
 create mode 100644 drivers/media/video/m52790.c
 create mode 100644 include/media/m52790.h

(limited to 'include')

diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index ea5be3711a7d..320f9e06a933 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -119,6 +119,15 @@ config VIDEO_CS53L32A
 	  To compile this driver as a module, choose M here: the
 	  module will be called cs53l32a.
 
+config VIDEO_M52790
+       tristate "Mitsubishi M52790 A/V switch"
+       depends on VIDEO_V4L2 && I2C && EXPERIMENTAL
+       ---help---
+	 Support for the Mitsubishi M52790 A/V switch.
+
+	 To compile this driver as a module, choose M here: the
+	 module will be called m52790.
+
 config VIDEO_TLV320AIC23B
 	tristate "Texas Instruments TLV320AIC23B audio codec"
 	depends on VIDEO_V4L2 && I2C && EXPERIMENTAL
diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile
index 8a2781821232..d8910d87acaa 100644
--- a/drivers/media/video/Makefile
+++ b/drivers/media/video/Makefile
@@ -67,6 +67,7 @@ obj-$(CONFIG_VIDEO_TVP5150) += tvp5150.o
 obj-$(CONFIG_VIDEO_PVRUSB2) += pvrusb2/
 obj-$(CONFIG_VIDEO_MSP3400) += msp3400.o
 obj-$(CONFIG_VIDEO_CS53L32A) += cs53l32a.o
+obj-$(CONFIG_VIDEO_M52790) += m52790.o
 obj-$(CONFIG_VIDEO_TLV320AIC23B) += tlv320aic23b.o
 obj-$(CONFIG_VIDEO_WM8775) += wm8775.o
 obj-$(CONFIG_VIDEO_WM8739) += wm8739.o
diff --git a/drivers/media/video/m52790.c b/drivers/media/video/m52790.c
new file mode 100644
index 000000000000..51ecbfbc1ae7
--- /dev/null
+++ b/drivers/media/video/m52790.c
@@ -0,0 +1,172 @@
+/*
+ * m52790 i2c ivtv driver.
+ * Copyright (C) 2007  Hans Verkuil
+ *
+ * A/V source switching Mitsubishi M52790SP/FP
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/ioctl.h>
+#include <asm/uaccess.h>
+#include <linux/i2c.h>
+#include <linux/i2c-id.h>
+#include <linux/videodev.h>
+#include <media/m52790.h>
+#include <media/v4l2-common.h>
+#include <media/v4l2-chip-ident.h>
+#include <media/v4l2-i2c-drv-legacy.h>
+
+MODULE_DESCRIPTION("i2c device driver for m52790 A/V switch");
+MODULE_AUTHOR("Hans Verkuil");
+MODULE_LICENSE("GPL");
+
+static unsigned short normal_i2c[] = { 0x90 >> 1, I2C_CLIENT_END };
+
+
+I2C_CLIENT_INSMOD;
+
+struct m52790_state {
+	u16 input;
+	u16 output;
+};
+
+/* ----------------------------------------------------------------------- */
+
+static int m52790_write(struct i2c_client *client)
+{
+	struct m52790_state *state = i2c_get_clientdata(client);
+	u8 sw1 = (state->input | state->output) & 0xff;
+	u8 sw2 = (state->input | state->output) >> 8;
+
+	return i2c_smbus_write_byte_data(client, sw1, sw2);
+}
+
+static int m52790_command(struct i2c_client *client, unsigned int cmd,
+			    void *arg)
+{
+	struct m52790_state *state = i2c_get_clientdata(client);
+	struct v4l2_routing *route = arg;
+
+	/* Note: audio and video are linked and cannot be switched separately.
+	   So audio and video routing commands are identical for this chip.
+	   In theory the video amplifier and audio modes could be handled
+	   separately for the output, but that seems to be overkill right now.
+	   The same holds for implementing an audio mute control, this is now
+	   part of the audio output routing. The normal case is that another
+	   chip takes care of the actual muting so making it part of the
+	   output routing seems to be the right thing to do for now. */
+	switch (cmd) {
+	case VIDIOC_INT_G_AUDIO_ROUTING:
+	case VIDIOC_INT_G_VIDEO_ROUTING:
+		route->input = state->input;
+		route->output = state->output;
+		break;
+
+	case VIDIOC_INT_S_AUDIO_ROUTING:
+	case VIDIOC_INT_S_VIDEO_ROUTING:
+		state->input = route->input;
+		state->output = route->output;
+		m52790_write(client);
+		break;
+
+#ifdef CONFIG_VIDEO_ADV_DEBUG
+	case VIDIOC_DBG_G_REGISTER:
+	case VIDIOC_DBG_S_REGISTER:
+	{
+		struct v4l2_register *reg = arg;
+
+		if (!v4l2_chip_match_i2c_client(client,
+					reg->match_type, reg->match_chip))
+			return -EINVAL;
+		if (!capable(CAP_SYS_ADMIN))
+			return -EPERM;
+		if (reg->reg != 0)
+			return -EINVAL;
+		if (cmd == VIDIOC_DBG_G_REGISTER)
+			reg->val = state->input | state->output;
+		else {
+			state->input = reg->val & 0x0303;
+			state->output = reg->val & ~0x0303;
+			m52790_write(client);
+		}
+		break;
+	}
+#endif
+
+	case VIDIOC_G_CHIP_IDENT:
+		return v4l2_chip_ident_i2c_client(client, arg,
+				V4L2_IDENT_M52790, 0);
+
+	case VIDIOC_LOG_STATUS:
+		v4l_info(client, "Switch 1: %02x\n",
+				(state->input | state->output) & 0xff);
+		v4l_info(client, "Switch 2: %02x\n",
+				(state->input | state->output) >> 8);
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/* ----------------------------------------------------------------------- */
+
+/* i2c implementation */
+
+static int m52790_probe(struct i2c_client *client)
+{
+	struct m52790_state *state;
+
+	/* Check if the adapter supports the needed features */
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_BYTE_DATA))
+		return -EIO;
+
+	snprintf(client->name, sizeof(client->name) - 1, "m52790");
+
+	v4l_info(client, "chip found @ 0x%x (%s)\n",
+			client->addr << 1, client->adapter->name);
+
+	state = kmalloc(sizeof(struct m52790_state), GFP_KERNEL);
+	if (state == NULL)
+		return -ENOMEM;
+
+	state->input = M52790_IN_TUNER;
+	state->output = M52790_OUT_STEREO;
+	i2c_set_clientdata(client, state);
+	m52790_write(client);
+	return 0;
+}
+
+static int m52790_remove(struct i2c_client *client)
+{
+	kfree(i2c_get_clientdata(client));
+	return 0;
+}
+
+/* ----------------------------------------------------------------------- */
+
+static struct v4l2_i2c_driver_data v4l2_i2c_data = {
+	.name = "m52790",
+	.driverid = I2C_DRIVERID_M52790,
+	.command = m52790_command,
+	.probe = m52790_probe,
+	.remove = m52790_remove,
+};
+
diff --git a/include/media/m52790.h b/include/media/m52790.h
new file mode 100644
index 000000000000..7ddffae31a67
--- /dev/null
+++ b/include/media/m52790.h
@@ -0,0 +1,93 @@
+/*
+    m52790.h - definition for m52790 inputs and outputs
+
+    Copyright (C) 2007 Hans Verkuil (hverkuil@xs4all.nl)
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#ifndef _M52790_H_
+#define _M52790_H_
+
+/* Input routing switch 1 */
+
+#define M52790_SW1_IN_MASK 	0x0003
+#define M52790_SW1_IN_TUNER 	0x0000
+#define M52790_SW1_IN_V2    	0x0001
+#define M52790_SW1_IN_V3    	0x0002
+#define M52790_SW1_IN_V4    	0x0003
+
+/* Selects component input instead of composite */
+#define M52790_SW1_YCMIX    	0x0004
+
+
+/* Input routing switch 2 */
+
+#define M52790_SW2_IN_MASK 	0x0300
+#define M52790_SW2_IN_TUNER 	0x0000
+#define M52790_SW2_IN_V2    	0x0100
+#define M52790_SW2_IN_V3    	0x0200
+#define M52790_SW2_IN_V4    	0x0300
+
+/* Selects component input instead of composite */
+#define M52790_SW2_YCMIX    	0x0400
+
+
+/* Output routing switch 1 */
+
+/* Enable 6dB amplifier for composite out */
+#define M52790_SW1_V_AMP    	0x0008
+
+/* Enable 6dB amplifier for component out */
+#define M52790_SW1_YC_AMP   	0x0010
+
+/* Audio output mode */
+#define M52790_SW1_AUDIO_MASK 	0x00c0
+#define M52790_SW1_AUDIO_MUTE 	0x0000
+#define M52790_SW1_AUDIO_R 	0x0040
+#define M52790_SW1_AUDIO_L 	0x0080
+#define M52790_SW1_AUDIO_STEREO 0x00c0
+
+
+/* Output routing switch 2 */
+
+/* Enable 6dB amplifier for composite out */
+#define M52790_SW2_V_AMP    	0x0800
+
+/* Enable 6dB amplifier for component out */
+#define M52790_SW2_YC_AMP   	0x1000
+
+/* Audio output mode */
+#define M52790_SW2_AUDIO_MASK 	0xc000
+#define M52790_SW2_AUDIO_MUTE 	0x0000
+#define M52790_SW2_AUDIO_R 	0x4000
+#define M52790_SW2_AUDIO_L 	0x8000
+#define M52790_SW2_AUDIO_STEREO 0xc000
+
+
+/* Common values */
+#define M52790_IN_TUNER (M52790_SW1_IN_TUNER | M52790_SW2_IN_TUNER)
+#define M52790_IN_V2    (M52790_SW1_IN_V2 | M52790_SW2_IN_V2)
+#define M52790_IN_V3    (M52790_SW1_IN_V3 | M52790_SW2_IN_V3)
+#define M52790_IN_V4    (M52790_SW1_IN_V4 | M52790_SW2_IN_V4)
+
+#define M52790_OUT_STEREO 	(M52790_SW1_AUDIO_STEREO | \
+				 M52790_SW2_AUDIO_STEREO)
+#define M52790_OUT_AMP_STEREO 	(M52790_SW1_AUDIO_STEREO | \
+				 M52790_SW1_V_AMP | \
+				 M52790_SW2_AUDIO_STEREO | \
+				 M52790_SW2_V_AMP)
+
+#endif
diff --git a/include/media/v4l2-chip-ident.h b/include/media/v4l2-chip-ident.h
index 8ae42c41dd08..2442c281e4fe 100644
--- a/include/media/v4l2-chip-ident.h
+++ b/include/media/v4l2-chip-ident.h
@@ -83,6 +83,9 @@ enum {
 	/* module upd64083: just ident 64083 */
 	V4L2_IDENT_UPD64083 = 64083,
 
+	/* module m52790: just ident 52790 */
+	V4L2_IDENT_M52790 = 52790,
+
 	/* module msp34xx: reserved range 34000-34999 */
 	V4L2_IDENT_MSP3400B = 34002,
 	V4L2_IDENT_MSP3410B = 34102,
-- 
cgit v1.2.3


From 05e997197e459a03df577ba49c34c1820957ee4a Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 30 Oct 2007 05:41:54 -0300
Subject: V4L/DVB (6487): i2c-id: add M52790 driver ID

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/linux/i2c-id.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index e18017d45758..adbdc6da39ca 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -125,6 +125,7 @@
 #define I2C_DRIVERID_LM4857 	92 	/* LM4857 Audio Amplifier */
 #define I2C_DRIVERID_VP27SMPX	93	/* Panasonic VP27s tuner internal MPX */
 #define I2C_DRIVERID_CS4270	94	/* Cirrus Logic 4270 audio codec */
+#define I2C_DRIVERID_M52790 	95      /* Mitsubishi M52790SP/FP AV switch */
 
 #define I2C_DRIVERID_I2CDEV	900
 #define I2C_DRIVERID_ARP        902    /* SMBus ARP Client              */
-- 
cgit v1.2.3


From c5d857d5d3a7cfe355d93916fc094035c85147f7 Mon Sep 17 00:00:00 2001
From: Marco Schluessler <marco@lordzodiac.de>
Date: Wed, 31 Oct 2007 00:44:22 -0300
Subject: V4L/DVB (6496): saa7146_vv.h: remove wrong include <linux/videodev.h>

remove wrong include <linux/videodev.h>

Signed-off-by: Marco Schluessler <marco@lordzodiac.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/saa7146_vv.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/media/saa7146_vv.h b/include/media/saa7146_vv.h
index e49f7e156061..89c442eb8849 100644
--- a/include/media/saa7146_vv.h
+++ b/include/media/saa7146_vv.h
@@ -1,7 +1,6 @@
 #ifndef __SAA7146_VV__
 #define __SAA7146_VV__
 
-#include <linux/videodev.h>
 #include <media/v4l2-common.h>
 #include <media/saa7146.h>
 #include <media/videobuf-dma-sg.h>
-- 
cgit v1.2.3


From 9695a49839353bdaff79fe27399be2eb72585223 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 5 Nov 2007 14:07:18 -0300
Subject: V4L/DVB (6574): common/ir-functions.c: make a function static

ir_rc5_decode() can become static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/common/ir-functions.c | 3 +--
 include/media/ir-common.h           | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/media/common/ir-functions.c b/drivers/media/common/ir-functions.c
index e7c3ab951a44..bb2a027b9483 100644
--- a/drivers/media/common/ir-functions.c
+++ b/drivers/media/common/ir-functions.c
@@ -258,7 +258,7 @@ int ir_decode_biphase(u32 *samples, int count, int low, int high)
  * saa7134 */
 
 /* decode raw bit pattern to RC5 code */
-u32 ir_rc5_decode(unsigned int code)
+static u32 ir_rc5_decode(unsigned int code)
 {
 	unsigned int org_code = code;
 	unsigned int pair;
@@ -371,7 +371,6 @@ EXPORT_SYMBOL_GPL(ir_dump_samples);
 EXPORT_SYMBOL_GPL(ir_decode_biphase);
 EXPORT_SYMBOL_GPL(ir_decode_pulsedistance);
 
-EXPORT_SYMBOL_GPL(ir_rc5_decode);
 EXPORT_SYMBOL_GPL(ir_rc5_timer_end);
 EXPORT_SYMBOL_GPL(ir_rc5_timer_keyup);
 
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index 7a785fa77212..b904d3257260 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -97,7 +97,6 @@ int  ir_dump_samples(u32 *samples, int count);
 int  ir_decode_biphase(u32 *samples, int count, int low, int high);
 int  ir_decode_pulsedistance(u32 *samples, int count, int low, int high);
 
-u32 ir_rc5_decode(unsigned int code);
 void ir_rc5_timer_end(unsigned long data);
 void ir_rc5_timer_keyup(unsigned long data);
 
-- 
cgit v1.2.3


From 2c94a674e059e89252d58da655efa4e798be4d48 Mon Sep 17 00:00:00 2001
From: Sakari Ailus <sakari.ailus@nokia.com>
Date: Tue, 30 Oct 2007 05:52:52 -0300
Subject: V4L/DVB (6580): Set slave's master before master's attach call.

V4L: Int if: Set slave's master before attach, remove master argument

The master also now gets its own pointer from slave's structure.

Signed-off-by: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/v4l2-int-device.c | 6 +++---
 include/media/v4l2-int-device.h       | 5 ++---
 2 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/v4l2-int-device.c b/drivers/media/video/v4l2-int-device.c
index 8b4ef530a3a8..a545dcaf857f 100644
--- a/drivers/media/video/v4l2-int-device.c
+++ b/drivers/media/video/v4l2-int-device.c
@@ -57,12 +57,12 @@ static void v4l2_int_device_try_attach_all(void)
 			if (!try_module_get(m->module))
 				continue;
 
-			if (m->u.master->attach(m, s)) {
+			s->u.slave->master = m;
+			if (m->u.master->attach(s)) {
+				s->u.slave->master = NULL;
 				module_put(m->module);
 				continue;
 			}
-
-			s->u.slave->master = m;
 		}
 	}
 }
diff --git a/include/media/v4l2-int-device.h b/include/media/v4l2-int-device.h
index 066ebfc4f983..c8b80e0f0651 100644
--- a/include/media/v4l2-int-device.h
+++ b/include/media/v4l2-int-device.h
@@ -44,9 +44,8 @@ enum v4l2_int_type {
 struct v4l2_int_device;
 
 struct v4l2_int_master {
-	int (*attach)(struct v4l2_int_device *master,
-		      struct v4l2_int_device *slave);
-	void (*detach)(struct v4l2_int_device *master);
+	int (*attach)(struct v4l2_int_device *slave);
+	void (*detach)(struct v4l2_int_device *slave);
 };
 
 typedef int (v4l2_int_ioctl_func)(struct v4l2_int_device *);
-- 
cgit v1.2.3


From 0fc0686e64d21a6abded96af2b1a895dfa8b2530 Mon Sep 17 00:00:00 2001
From: Brandon Philips <brandon@ifup.org>
Date: Tue, 6 Nov 2007 20:02:36 -0300
Subject: V4L/DVB (6600): V4L: videobuf: don't chew up namespace STATE_.*,
 convert to VIDEOBUF_

s/STATE_NEEDS_INIT/VIDEOBUF_NEEDS_INIT/g
s/STATE_PREPARED/VIDEOBUF_PREPARED/g
s/STATE_QUEUED/VIDEOBUF_QUEUED/g
s/STATE_ACTIVE/VIDEOBUF_ACTIVE/g
s/STATE_DONE/VIDEOBUF_DONE/g
s/STATE_ERROR/VIDEOBUF_ERROR/g
s/STATE_IDLE/VIDEOBUF_IDLE/g

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/common/saa7146_fops.c         |  8 ++---
 drivers/media/common/saa7146_vbi.c          | 10 +++---
 drivers/media/common/saa7146_video.c        |  8 ++---
 drivers/media/video/bt8xx/bttv-driver.c     | 26 +++++++--------
 drivers/media/video/bt8xx/bttv-risc.c       | 14 ++++----
 drivers/media/video/bt8xx/bttv-vbi.c        |  6 ++--
 drivers/media/video/cx23885/cx23885-core.c  | 18 +++++------
 drivers/media/video/cx88/cx88-alsa.c        |  2 +-
 drivers/media/video/cx88/cx88-core.c        |  4 +--
 drivers/media/video/cx88/cx88-mpeg.c        | 14 ++++----
 drivers/media/video/cx88/cx88-vbi.c         | 10 +++---
 drivers/media/video/cx88/cx88-video.c       | 22 ++++++-------
 drivers/media/video/saa7134/saa7134-core.c  |  8 ++---
 drivers/media/video/saa7134/saa7134-ts.c    |  8 ++---
 drivers/media/video/saa7134/saa7134-vbi.c   |  8 ++---
 drivers/media/video/saa7134/saa7134-video.c | 12 +++----
 drivers/media/video/videobuf-core.c         | 50 ++++++++++++++---------------
 drivers/media/video/videobuf-dvb.c          |  2 +-
 drivers/media/video/vivi.c                  | 24 +++++++-------
 include/media/videobuf-core.h               | 14 ++++----
 20 files changed, 134 insertions(+), 134 deletions(-)

(limited to 'include')

diff --git a/drivers/media/common/saa7146_fops.c b/drivers/media/common/saa7146_fops.c
index 67d1b1b1b254..b40bf2306fbc 100644
--- a/drivers/media/common/saa7146_fops.c
+++ b/drivers/media/common/saa7146_fops.c
@@ -61,7 +61,7 @@ void saa7146_dma_free(struct saa7146_dev *dev,struct videobuf_queue *q,
 	videobuf_waiton(&buf->vb,0,0);
 	videobuf_dma_unmap(q, dma);
 	videobuf_dma_free(dma);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 
@@ -83,7 +83,7 @@ int saa7146_buffer_queue(struct saa7146_dev *dev,
 		buf->activate(dev,buf,NULL);
 	} else {
 		list_add_tail(&buf->vb.queue,&q->queue);
-		buf->vb.state = STATE_QUEUED;
+		buf->vb.state = VIDEOBUF_QUEUED;
 		DEB_D(("adding buffer %p to queue. (active buffer present)\n", buf));
 	}
 	return 0;
@@ -174,7 +174,7 @@ void saa7146_buffer_timeout(unsigned long data)
 	spin_lock_irqsave(&dev->slock,flags);
 	if (q->curr) {
 		DEB_D(("timeout on %p\n", q->curr));
-		saa7146_buffer_finish(dev,q,STATE_ERROR);
+		saa7146_buffer_finish(dev,q,VIDEOBUF_ERROR);
 	}
 
 	/* we don't restart the transfer here like other drivers do. when
@@ -366,7 +366,7 @@ static unsigned int fops_poll(struct file *file, struct poll_table_struct *wait)
 	}
 
 	poll_wait(file, &buf->done, wait);
-	if (buf->state == STATE_DONE || buf->state == STATE_ERROR) {
+	if (buf->state == VIDEOBUF_DONE || buf->state == VIDEOBUF_ERROR) {
 		DEB_D(("poll succeeded!\n"));
 		return POLLIN|POLLRDNORM;
 	}
diff --git a/drivers/media/common/saa7146_vbi.c b/drivers/media/common/saa7146_vbi.c
index 6103484e4442..c32dda973e92 100644
--- a/drivers/media/common/saa7146_vbi.c
+++ b/drivers/media/common/saa7146_vbi.c
@@ -205,7 +205,7 @@ static int buffer_activate(struct saa7146_dev *dev,
 			   struct saa7146_buf *next)
 {
 	struct saa7146_vv *vv = dev->vv_data;
-	buf->vb.state = STATE_ACTIVE;
+	buf->vb.state = VIDEOBUF_ACTIVE;
 
 	DEB_VBI(("dev:%p, buf:%p, next:%p\n",dev,buf,next));
 	saa7146_set_vbi_capture(dev,buf,next);
@@ -238,7 +238,7 @@ static int buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,e
 	if (buf->vb.size != size)
 		saa7146_dma_free(dev,q,buf);
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
 
 		buf->vb.width  = llength;
@@ -257,7 +257,7 @@ static int buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,e
 		if (0 != err)
 			return err;
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->activate = buffer_activate;
 
 	return 0;
@@ -335,7 +335,7 @@ static void vbi_stop(struct saa7146_fh *fh, struct file *file)
 	saa7146_write(dev, MC1, MASK_20);
 
 	if (vv->vbi_q.curr) {
-		saa7146_buffer_finish(dev,&vv->vbi_q,STATE_DONE);
+		saa7146_buffer_finish(dev,&vv->vbi_q,VIDEOBUF_DONE);
 	}
 
 	videobuf_queue_cancel(&fh->vbi_q);
@@ -458,7 +458,7 @@ static void vbi_irq_done(struct saa7146_dev *dev, unsigned long status)
 		/* this must be += 2, one count for each field */
 		vv->vbi_fieldcount+=2;
 		vv->vbi_q.curr->vb.field_count = vv->vbi_fieldcount;
-		saa7146_buffer_finish(dev,&vv->vbi_q,STATE_DONE);
+		saa7146_buffer_finish(dev,&vv->vbi_q,VIDEOBUF_DONE);
 	} else {
 		DEB_VBI(("dev:%p\n",dev));
 	}
diff --git a/drivers/media/common/saa7146_video.c b/drivers/media/common/saa7146_video.c
index ae36d101006b..c31ab480d8e1 100644
--- a/drivers/media/common/saa7146_video.c
+++ b/drivers/media/common/saa7146_video.c
@@ -1235,7 +1235,7 @@ static int buffer_activate (struct saa7146_dev *dev,
 {
 	struct saa7146_vv *vv = dev->vv_data;
 
-	buf->vb.state = STATE_ACTIVE;
+	buf->vb.state = VIDEOBUF_ACTIVE;
 	saa7146_set_capture(dev,buf,next);
 
 	mod_timer(&vv->video_q.timeout, jiffies+BUFFER_TIMEOUT);
@@ -1281,7 +1281,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 		saa7146_dma_free(dev,q,buf);
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct saa7146_format *sfmt;
 
 		buf->vb.bytesperline  = fh->video_fmt.bytesperline;
@@ -1314,7 +1314,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 		if (err)
 			goto oops;
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->activate = buffer_activate;
 
 	return 0;
@@ -1453,7 +1453,7 @@ static void video_irq_done(struct saa7146_dev *dev, unsigned long st)
 
 	/* only finish the buffer if we have one... */
 	if( NULL != q->curr ) {
-		saa7146_buffer_finish(dev,q,STATE_DONE);
+		saa7146_buffer_finish(dev,q,VIDEOBUF_DONE);
 	}
 	saa7146_buffer_next(dev,q,0);
 
diff --git a/drivers/media/video/bt8xx/bttv-driver.c b/drivers/media/video/bt8xx/bttv-driver.c
index db0e4b78e7dd..eb2c12eb2fc8 100644
--- a/drivers/media/video/bt8xx/bttv-driver.c
+++ b/drivers/media/video/bt8xx/bttv-driver.c
@@ -1638,7 +1638,7 @@ bttv_switch_overlay(struct bttv *btv, struct bttv_fh *fh,
 
 	dprintk("switch_overlay: enter [new=%p]\n",new);
 	if (new)
-		new->vb.state = STATE_DONE;
+		new->vb.state = VIDEOBUF_DONE;
 	spin_lock_irqsave(&btv->s_lock,flags);
 	old = btv->screen;
 	btv->screen = new;
@@ -1749,7 +1749,7 @@ static int bttv_prepare_buffer(struct videobuf_queue *q,struct bttv *btv,
 	}
 
 	/* alloc risc memory */
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		redo_dma_risc = 1;
 		if (0 != (rc = videobuf_iolock(q,&buf->vb,&btv->fbuf)))
 			goto fail;
@@ -1759,7 +1759,7 @@ static int bttv_prepare_buffer(struct videobuf_queue *q,struct bttv *btv,
 		if (0 != (rc = bttv_buffer_risc(btv,buf)))
 			goto fail;
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	return 0;
 
  fail:
@@ -1798,7 +1798,7 @@ buffer_queue(struct videobuf_queue *q, struct videobuf_buffer *vb)
 	struct bttv_fh *fh = q->priv_data;
 	struct bttv    *btv = fh->btv;
 
-	buf->vb.state = STATE_QUEUED;
+	buf->vb.state = VIDEOBUF_QUEUED;
 	list_add_tail(&buf->vb.queue,&btv->capture);
 	if (!btv->curr.frame_irq) {
 		btv->loop_irq |= 1;
@@ -3102,8 +3102,8 @@ static unsigned int bttv_poll(struct file *file, poll_table *wait)
 	}
 
 	poll_wait(file, &buf->vb.done, wait);
-	if (buf->vb.state == STATE_DONE ||
-	    buf->vb.state == STATE_ERROR)
+	if (buf->vb.state == VIDEOBUF_DONE ||
+	    buf->vb.state == VIDEOBUF_ERROR)
 		return POLLIN|POLLRDNORM;
 	return 0;
 }
@@ -3699,20 +3699,20 @@ static void bttv_irq_timeout(unsigned long data)
 	bttv_set_dma(btv, 0);
 
 	/* wake up */
-	bttv_irq_wakeup_video(btv, &old, &new, STATE_ERROR);
-	bttv_irq_wakeup_vbi(btv, ovbi, STATE_ERROR);
+	bttv_irq_wakeup_video(btv, &old, &new, VIDEOBUF_ERROR);
+	bttv_irq_wakeup_vbi(btv, ovbi, VIDEOBUF_ERROR);
 
 	/* cancel all outstanding capture / vbi requests */
 	while (!list_empty(&btv->capture)) {
 		item = list_entry(btv->capture.next, struct bttv_buffer, vb.queue);
 		list_del(&item->vb.queue);
-		item->vb.state = STATE_ERROR;
+		item->vb.state = VIDEOBUF_ERROR;
 		wake_up(&item->vb.done);
 	}
 	while (!list_empty(&btv->vcapture)) {
 		item = list_entry(btv->vcapture.next, struct bttv_buffer, vb.queue);
 		list_del(&item->vb.queue);
-		item->vb.state = STATE_ERROR;
+		item->vb.state = VIDEOBUF_ERROR;
 		wake_up(&item->vb.done);
 	}
 
@@ -3735,7 +3735,7 @@ bttv_irq_wakeup_top(struct bttv *btv)
 
 	do_gettimeofday(&wakeup->vb.ts);
 	wakeup->vb.field_count = btv->field_count;
-	wakeup->vb.state = STATE_DONE;
+	wakeup->vb.state = VIDEOBUF_DONE;
 	wake_up(&wakeup->vb.done);
 	spin_unlock(&btv->s_lock);
 }
@@ -3784,7 +3784,7 @@ bttv_irq_switch_video(struct bttv *btv)
 	}
 
 	/* wake up finished buffers */
-	bttv_irq_wakeup_video(btv, &old, &new, STATE_DONE);
+	bttv_irq_wakeup_video(btv, &old, &new, VIDEOBUF_DONE);
 	spin_unlock(&btv->s_lock);
 }
 
@@ -3817,7 +3817,7 @@ bttv_irq_switch_vbi(struct bttv *btv)
 	bttv_buffer_activate_vbi(btv, new);
 	bttv_set_dma(btv, 0);
 
-	bttv_irq_wakeup_vbi(btv, old, STATE_DONE);
+	bttv_irq_wakeup_vbi(btv, old, VIDEOBUF_DONE);
 	spin_unlock(&btv->s_lock);
 }
 
diff --git a/drivers/media/video/bt8xx/bttv-risc.c b/drivers/media/video/bt8xx/bttv-risc.c
index 58986f1a5f1a..e5979f77504c 100644
--- a/drivers/media/video/bt8xx/bttv-risc.c
+++ b/drivers/media/video/bt8xx/bttv-risc.c
@@ -582,7 +582,7 @@ bttv_dma_free(struct videobuf_queue *q,struct bttv *btv, struct bttv_buffer *buf
 	videobuf_dma_free(dma);
 	btcx_riscmem_free(btv->c.pci,&buf->bottom);
 	btcx_riscmem_free(btv->c.pci,&buf->top);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 int
@@ -602,7 +602,7 @@ bttv_buffer_activate_vbi(struct bttv *btv,
 	if (vbi) {
 		unsigned int crop, vdelay;
 
-		vbi->vb.state = STATE_ACTIVE;
+		vbi->vb.state = VIDEOBUF_ACTIVE;
 		list_del(&vbi->vb.queue);
 
 		/* VDELAY is start of video, end of VBI capturing. */
@@ -644,12 +644,12 @@ bttv_buffer_activate_video(struct bttv *btv,
 	/* video capture */
 	if (NULL != set->top  &&  NULL != set->bottom) {
 		if (set->top == set->bottom) {
-			set->top->vb.state    = STATE_ACTIVE;
+			set->top->vb.state    = VIDEOBUF_ACTIVE;
 			if (set->top->vb.queue.next)
 				list_del(&set->top->vb.queue);
 		} else {
-			set->top->vb.state    = STATE_ACTIVE;
-			set->bottom->vb.state = STATE_ACTIVE;
+			set->top->vb.state    = VIDEOBUF_ACTIVE;
+			set->bottom->vb.state = VIDEOBUF_ACTIVE;
 			if (set->top->vb.queue.next)
 				list_del(&set->top->vb.queue);
 			if (set->bottom->vb.queue.next)
@@ -666,7 +666,7 @@ bttv_buffer_activate_video(struct bttv *btv,
 		btaor((set->top->btswap & 0x0a) | (set->bottom->btswap & 0x05),
 		      ~0x0f, BT848_COLOR_CTL);
 	} else if (NULL != set->top) {
-		set->top->vb.state  = STATE_ACTIVE;
+		set->top->vb.state  = VIDEOBUF_ACTIVE;
 		if (set->top->vb.queue.next)
 			list_del(&set->top->vb.queue);
 		bttv_apply_geo(btv, &set->top->geo,1);
@@ -677,7 +677,7 @@ bttv_buffer_activate_video(struct bttv *btv,
 		btaor(set->top->btformat & 0xff, ~0xff, BT848_COLOR_FMT);
 		btaor(set->top->btswap & 0x0f,   ~0x0f, BT848_COLOR_CTL);
 	} else if (NULL != set->bottom) {
-		set->bottom->vb.state = STATE_ACTIVE;
+		set->bottom->vb.state = VIDEOBUF_ACTIVE;
 		if (set->bottom->vb.queue.next)
 			list_del(&set->bottom->vb.queue);
 		bttv_apply_geo(btv, &set->bottom->geo,1);
diff --git a/drivers/media/video/bt8xx/bttv-vbi.c b/drivers/media/video/bt8xx/bttv-vbi.c
index 346ce019bdcb..b924f05e3b74 100644
--- a/drivers/media/video/bt8xx/bttv-vbi.c
+++ b/drivers/media/video/bt8xx/bttv-vbi.c
@@ -142,7 +142,7 @@ static int vbi_buffer_prepare(struct videobuf_queue *q,
 		redo_dma_risc = 1;
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		redo_dma_risc = 1;
 		if (0 != (rc = videobuf_iolock(q, &buf->vb, NULL)))
 			goto fail;
@@ -189,7 +189,7 @@ static int vbi_buffer_prepare(struct videobuf_queue *q,
 	/* For bttv_buffer_activate_vbi(). */
 	buf->geo.vdelay = min_vdelay;
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->vb.field = field;
 	dprintk("buf prepare %p: top=%p bottom=%p field=%s\n",
 		vb, &buf->top, &buf->bottom,
@@ -209,7 +209,7 @@ vbi_buffer_queue(struct videobuf_queue *q, struct videobuf_buffer *vb)
 	struct bttv_buffer *buf = container_of(vb,struct bttv_buffer,vb);
 
 	dprintk("queue %p\n",vb);
-	buf->vb.state = STATE_QUEUED;
+	buf->vb.state = VIDEOBUF_QUEUED;
 	list_add_tail(&buf->vb.queue,&btv->vcapture);
 	if (NULL == btv->cvbi) {
 		fh->btv->loop_irq |= 4;
diff --git a/drivers/media/video/cx23885/cx23885-core.c b/drivers/media/video/cx23885/cx23885-core.c
index 0c9c9af0efb7..f205ad6354e7 100644
--- a/drivers/media/video/cx23885/cx23885-core.c
+++ b/drivers/media/video/cx23885/cx23885-core.c
@@ -378,7 +378,7 @@ static void cx23885_wakeup(struct cx23885_tsport *port,
 		do_gettimeofday(&buf->vb.ts);
 		dprintk(2, "[%p/%d] wakeup reg=%d buf=%d\n", buf, buf->vb.i,
 			count, buf->count);
-		buf->vb.state = STATE_DONE;
+		buf->vb.state = VIDEOBUF_DONE;
 		list_del(&buf->vb.queue);
 		wake_up(&buf->vb.done);
 	}
@@ -972,7 +972,7 @@ void cx23885_free_buffer(struct videobuf_queue *q, struct cx23885_buffer *buf)
 	videobuf_dma_unmap(q, dma);
 	videobuf_dma_free(dma);
 	btcx_riscmem_free((struct pci_dev *)q->dev, &buf->risc);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 static int cx23885_start_dma(struct cx23885_tsport *port,
@@ -1075,7 +1075,7 @@ static int cx23885_restart_queue(struct cx23885_tsport *port,
 				list_del(&buf->vb.queue);
 				list_add_tail(&buf->vb.queue, &q->active);
 				cx23885_start_dma(port, q, buf);
-				buf->vb.state = STATE_ACTIVE;
+				buf->vb.state = VIDEOBUF_ACTIVE;
 				buf->count    = q->count++;
 				mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
 				dprintk(5, "[%p/%d] restart_queue - first active\n",
@@ -1086,7 +1086,7 @@ static int cx23885_restart_queue(struct cx23885_tsport *port,
 				   prev->fmt       == buf->fmt) {
 				list_del(&buf->vb.queue);
 				list_add_tail(&buf->vb.queue, &q->active);
-				buf->vb.state = STATE_ACTIVE;
+				buf->vb.state = VIDEOBUF_ACTIVE;
 				buf->count    = q->count++;
 				prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 				prev->risc.jmp[2] = cpu_to_le32(0); /* 64 bit bits 63-32 */
@@ -1123,7 +1123,7 @@ int cx23885_buf_prepare(struct videobuf_queue *q, struct cx23885_tsport *port,
 	if (0 != buf->vb.baddr  &&  buf->vb.bsize < size)
 		return -EINVAL;
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		buf->vb.width  = port->ts_packet_size;
 		buf->vb.height = port->ts_packet_count;
 		buf->vb.size   = size;
@@ -1135,7 +1135,7 @@ int cx23885_buf_prepare(struct videobuf_queue *q, struct cx23885_tsport *port,
 					videobuf_to_dma(&buf->vb)->sglist,
 					buf->vb.width, buf->vb.height);
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	return 0;
 
  fail:
@@ -1158,7 +1158,7 @@ void cx23885_buf_queue(struct cx23885_tsport *port, struct cx23885_buffer *buf)
 		dprintk( 1, "queue is empty - first active\n" );
 		list_add_tail(&buf->vb.queue, &cx88q->active);
 		cx23885_start_dma(port, cx88q, buf);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = cx88q->count++;
 		mod_timer(&cx88q->timeout, jiffies + BUFFER_TIMEOUT);
 		dprintk(1, "[%p/%d] %s - first active\n",
@@ -1168,7 +1168,7 @@ void cx23885_buf_queue(struct cx23885_tsport *port, struct cx23885_buffer *buf)
 		prev = list_entry(cx88q->active.prev, struct cx23885_buffer,
 				  vb.queue);
 		list_add_tail(&buf->vb.queue, &cx88q->active);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = cx88q->count++;
 		prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 		prev->risc.jmp[2] = cpu_to_le32(0); /* 64 bit bits 63-32 */
@@ -1192,7 +1192,7 @@ static void do_cancel_buffers(struct cx23885_tsport *port, char *reason,
 		buf = list_entry(q->active.next, struct cx23885_buffer,
 				 vb.queue);
 		list_del(&buf->vb.queue);
-		buf->vb.state = STATE_ERROR;
+		buf->vb.state = VIDEOBUF_ERROR;
 		wake_up(&buf->vb.done);
 		dprintk(1, "[%p/%d] %s - dma=0x%08lx\n",
 			buf, buf->vb.i, reason, (unsigned long)buf->risc.dma);
diff --git a/drivers/media/video/cx88/cx88-alsa.c b/drivers/media/video/cx88/cx88-alsa.c
index 40ffd7a5579a..8735227f7e47 100644
--- a/drivers/media/video/cx88/cx88-alsa.c
+++ b/drivers/media/video/cx88/cx88-alsa.c
@@ -417,7 +417,7 @@ static int snd_cx88_hw_params(struct snd_pcm_substream * substream,
 	buf->risc.jmp[0] = cpu_to_le32(RISC_JUMP|RISC_IRQ1|RISC_CNT_INC);
 	buf->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 
 	chip->buf = buf;
 	chip->dma_risc = dma;
diff --git a/drivers/media/video/cx88/cx88-core.c b/drivers/media/video/cx88/cx88-core.c
index 62e8dd24c5f5..01e2ac98970b 100644
--- a/drivers/media/video/cx88/cx88-core.c
+++ b/drivers/media/video/cx88/cx88-core.c
@@ -220,7 +220,7 @@ cx88_free_buffer(struct videobuf_queue *q, struct cx88_buffer *buf)
 	videobuf_dma_unmap(q, dma);
 	videobuf_dma_free(dma);
 	btcx_riscmem_free((struct pci_dev *)q->dev, &buf->risc);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 /* ------------------------------------------------------------------ */
@@ -538,7 +538,7 @@ void cx88_wakeup(struct cx88_core *core,
 		do_gettimeofday(&buf->vb.ts);
 		dprintk(2,"[%p/%d] wakeup reg=%d buf=%d\n",buf,buf->vb.i,
 			count, buf->count);
-		buf->vb.state = STATE_DONE;
+		buf->vb.state = VIDEOBUF_DONE;
 		list_del(&buf->vb.queue);
 		wake_up(&buf->vb.done);
 	}
diff --git a/drivers/media/video/cx88/cx88-mpeg.c b/drivers/media/video/cx88/cx88-mpeg.c
index 60ac4234d84c..339a88a64f54 100644
--- a/drivers/media/video/cx88/cx88-mpeg.c
+++ b/drivers/media/video/cx88/cx88-mpeg.c
@@ -195,7 +195,7 @@ static int cx8802_restart_queue(struct cx8802_dev    *dev,
 				list_del(&buf->vb.queue);
 				list_add_tail(&buf->vb.queue,&q->active);
 				cx8802_start_dma(dev, q, buf);
-				buf->vb.state = STATE_ACTIVE;
+				buf->vb.state = VIDEOBUF_ACTIVE;
 				buf->count    = q->count++;
 				mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
 				dprintk(1,"[%p/%d] restart_queue - first active\n",
@@ -206,7 +206,7 @@ static int cx8802_restart_queue(struct cx8802_dev    *dev,
 				   prev->fmt       == buf->fmt) {
 				list_del(&buf->vb.queue);
 				list_add_tail(&buf->vb.queue,&q->active);
-				buf->vb.state = STATE_ACTIVE;
+				buf->vb.state = VIDEOBUF_ACTIVE;
 				buf->count    = q->count++;
 				prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 				dprintk(1,"[%p/%d] restart_queue - move to active\n",
@@ -242,7 +242,7 @@ int cx8802_buf_prepare(struct videobuf_queue *q, struct cx8802_dev *dev,
 	if (0 != buf->vb.baddr  &&  buf->vb.bsize < size)
 		return -EINVAL;
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		buf->vb.width  = dev->ts_packet_size;
 		buf->vb.height = dev->ts_packet_count;
 		buf->vb.size   = size;
@@ -254,7 +254,7 @@ int cx8802_buf_prepare(struct videobuf_queue *q, struct cx8802_dev *dev,
 				     dma->sglist,
 				     buf->vb.width, buf->vb.height, 0);
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	return 0;
 
  fail:
@@ -276,7 +276,7 @@ void cx8802_buf_queue(struct cx8802_dev *dev, struct cx88_buffer *buf)
 		dprintk( 1, "queue is empty - first active\n" );
 		list_add_tail(&buf->vb.queue,&cx88q->active);
 		cx8802_start_dma(dev, cx88q, buf);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = cx88q->count++;
 		mod_timer(&cx88q->timeout, jiffies+BUFFER_TIMEOUT);
 		dprintk(1,"[%p/%d] %s - first active\n",
@@ -286,7 +286,7 @@ void cx8802_buf_queue(struct cx8802_dev *dev, struct cx88_buffer *buf)
 		dprintk( 1, "queue is not empty - append to active\n" );
 		prev = list_entry(cx88q->active.prev, struct cx88_buffer, vb.queue);
 		list_add_tail(&buf->vb.queue,&cx88q->active);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = cx88q->count++;
 		prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 		dprintk( 1, "[%p/%d] %s - append to active\n",
@@ -306,7 +306,7 @@ static void do_cancel_buffers(struct cx8802_dev *dev, char *reason, int restart)
 	while (!list_empty(&q->active)) {
 		buf = list_entry(q->active.next, struct cx88_buffer, vb.queue);
 		list_del(&buf->vb.queue);
-		buf->vb.state = STATE_ERROR;
+		buf->vb.state = VIDEOBUF_ERROR;
 		wake_up(&buf->vb.done);
 		dprintk(1,"[%p/%d] %s - dma=0x%08lx\n",
 			buf, buf->vb.i, reason, (unsigned long)buf->risc.dma);
diff --git a/drivers/media/video/cx88/cx88-vbi.c b/drivers/media/video/cx88/cx88-vbi.c
index babb08556406..d96ecfcf393a 100644
--- a/drivers/media/video/cx88/cx88-vbi.c
+++ b/drivers/media/video/cx88/cx88-vbi.c
@@ -130,7 +130,7 @@ void cx8800_vbi_timeout(unsigned long data)
 	while (!list_empty(&q->active)) {
 		buf = list_entry(q->active.next, struct cx88_buffer, vb.queue);
 		list_del(&buf->vb.queue);
-		buf->vb.state = STATE_ERROR;
+		buf->vb.state = VIDEOBUF_ERROR;
 		wake_up(&buf->vb.done);
 		printk("%s/0: [%p/%d] timeout - dma=0x%08lx\n", dev->core->name,
 		       buf, buf->vb.i, (unsigned long)buf->risc.dma);
@@ -168,7 +168,7 @@ vbi_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 	if (0 != buf->vb.baddr  &&  buf->vb.bsize < size)
 		return -EINVAL;
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
 		buf->vb.width  = VBI_LINE_LENGTH;
 		buf->vb.height = VBI_LINE_COUNT;
@@ -183,7 +183,7 @@ vbi_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 				 buf->vb.width, 0,
 				 buf->vb.height);
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	return 0;
 
  fail:
@@ -207,7 +207,7 @@ vbi_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 	if (list_empty(&q->active)) {
 		list_add_tail(&buf->vb.queue,&q->active);
 		cx8800_start_vbi_dma(dev, q, buf);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = q->count++;
 		mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
 		dprintk(2,"[%p/%d] vbi_queue - first active\n",
@@ -216,7 +216,7 @@ vbi_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 	} else {
 		prev = list_entry(q->active.prev, struct cx88_buffer, vb.queue);
 		list_add_tail(&buf->vb.queue,&q->active);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = q->count++;
 		prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 		dprintk(2,"[%p/%d] buffer_queue - append to active\n",
diff --git a/drivers/media/video/cx88/cx88-video.c b/drivers/media/video/cx88/cx88-video.c
index c84dafbdb991..6d5ea8ce9830 100644
--- a/drivers/media/video/cx88/cx88-video.c
+++ b/drivers/media/video/cx88/cx88-video.c
@@ -486,7 +486,7 @@ static int restart_video_queue(struct cx8800_dev    *dev,
 		if (NULL == prev) {
 			list_move_tail(&buf->vb.queue, &q->active);
 			start_video_dma(dev, q, buf);
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			buf->count    = q->count++;
 			mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
 			dprintk(2,"[%p/%d] restart_queue - first active\n",
@@ -496,7 +496,7 @@ static int restart_video_queue(struct cx8800_dev    *dev,
 			   prev->vb.height == buf->vb.height &&
 			   prev->fmt       == buf->fmt) {
 			list_move_tail(&buf->vb.queue, &q->active);
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			buf->count    = q->count++;
 			prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 			dprintk(2,"[%p/%d] restart_queue - move to active\n",
@@ -553,7 +553,7 @@ buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 		init_buffer = 1;
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		init_buffer = 1;
 		if (0 != (rc = videobuf_iolock(q,&buf->vb,NULL)))
 			goto fail;
@@ -601,7 +601,7 @@ buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 		fh->width, fh->height, fh->fmt->depth, fh->fmt->name,
 		(unsigned long)buf->risc.dma);
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	return 0;
 
  fail:
@@ -625,14 +625,14 @@ buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 
 	if (!list_empty(&q->queued)) {
 		list_add_tail(&buf->vb.queue,&q->queued);
-		buf->vb.state = STATE_QUEUED;
+		buf->vb.state = VIDEOBUF_QUEUED;
 		dprintk(2,"[%p/%d] buffer_queue - append to queued\n",
 			buf, buf->vb.i);
 
 	} else if (list_empty(&q->active)) {
 		list_add_tail(&buf->vb.queue,&q->active);
 		start_video_dma(dev, q, buf);
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		buf->count    = q->count++;
 		mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
 		dprintk(2,"[%p/%d] buffer_queue - first active\n",
@@ -644,7 +644,7 @@ buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 		    prev->vb.height == buf->vb.height &&
 		    prev->fmt       == buf->fmt) {
 			list_add_tail(&buf->vb.queue,&q->active);
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			buf->count    = q->count++;
 			prev->risc.jmp[1] = cpu_to_le32(buf->risc.dma);
 			dprintk(2,"[%p/%d] buffer_queue - append to active\n",
@@ -652,7 +652,7 @@ buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 
 		} else {
 			list_add_tail(&buf->vb.queue,&q->queued);
-			buf->vb.state = STATE_QUEUED;
+			buf->vb.state = VIDEOBUF_QUEUED;
 			dprintk(2,"[%p/%d] buffer_queue - first queued\n",
 				buf, buf->vb.i);
 		}
@@ -822,8 +822,8 @@ video_poll(struct file *file, struct poll_table_struct *wait)
 			return POLLERR;
 	}
 	poll_wait(file, &buf->vb.done, wait);
-	if (buf->vb.state == STATE_DONE ||
-	    buf->vb.state == STATE_ERROR)
+	if (buf->vb.state == VIDEOBUF_DONE ||
+	    buf->vb.state == VIDEOBUF_ERROR)
 		return POLLIN|POLLRDNORM;
 	return 0;
 }
@@ -1496,7 +1496,7 @@ static void cx8800_vid_timeout(unsigned long data)
 	while (!list_empty(&q->active)) {
 		buf = list_entry(q->active.next, struct cx88_buffer, vb.queue);
 		list_del(&buf->vb.queue);
-		buf->vb.state = STATE_ERROR;
+		buf->vb.state = VIDEOBUF_ERROR;
 		wake_up(&buf->vb.done);
 		printk("%s/0: [%p/%d] timeout - dma=0x%08lx\n", core->name,
 		       buf, buf->vb.i, (unsigned long)buf->risc.dma);
diff --git a/drivers/media/video/saa7134/saa7134-core.c b/drivers/media/video/saa7134/saa7134-core.c
index fc71843b9dc2..76410f5bdcdd 100644
--- a/drivers/media/video/saa7134/saa7134-core.c
+++ b/drivers/media/video/saa7134/saa7134-core.c
@@ -294,7 +294,7 @@ void saa7134_dma_free(struct videobuf_queue *q,struct saa7134_buf *buf)
 	videobuf_waiton(&buf->vb,0,0);
 	videobuf_dma_unmap(q, dma);
 	videobuf_dma_free(dma);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 /* ------------------------------------------------------------------ */
@@ -313,7 +313,7 @@ int saa7134_buffer_queue(struct saa7134_dev *dev,
 			buf->activate(dev,buf,NULL);
 		} else if (list_empty(&q->queue)) {
 			list_add_tail(&buf->vb.queue,&q->queue);
-			buf->vb.state = STATE_QUEUED;
+			buf->vb.state = VIDEOBUF_QUEUED;
 		} else {
 			next = list_entry(q->queue.next,struct saa7134_buf,
 					  vb.queue);
@@ -322,7 +322,7 @@ int saa7134_buffer_queue(struct saa7134_dev *dev,
 		}
 	} else {
 		list_add_tail(&buf->vb.queue,&q->queue);
-		buf->vb.state = STATE_QUEUED;
+		buf->vb.state = VIDEOBUF_QUEUED;
 	}
 	return 0;
 }
@@ -387,7 +387,7 @@ void saa7134_buffer_timeout(unsigned long data)
 	   try to start over with the next one. */
 	if (q->curr) {
 		dprintk("timeout on %p\n",q->curr);
-		saa7134_buffer_finish(dev,q,STATE_ERROR);
+		saa7134_buffer_finish(dev,q,VIDEOBUF_ERROR);
 	}
 	saa7134_buffer_next(dev,q);
 	spin_unlock_irqrestore(&dev->slock,flags);
diff --git a/drivers/media/video/saa7134/saa7134-ts.c b/drivers/media/video/saa7134/saa7134-ts.c
index 4b63ad3e8466..f1b8fcaeb43a 100644
--- a/drivers/media/video/saa7134/saa7134-ts.c
+++ b/drivers/media/video/saa7134/saa7134-ts.c
@@ -47,7 +47,7 @@ static int buffer_activate(struct saa7134_dev *dev,
 {
 
 	dprintk("buffer_activate [%p]",buf);
-	buf->vb.state = STATE_ACTIVE;
+	buf->vb.state = VIDEOBUF_ACTIVE;
 	buf->top_seen = 0;
 
 	if (NULL == next)
@@ -91,7 +91,7 @@ static int buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 		saa7134_dma_free(q,buf);
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
 
 		buf->vb.width  = llength;
@@ -121,7 +121,7 @@ static int buffer_prepare(struct videobuf_queue *q, struct videobuf_buffer *vb,
 	saa_writel(SAA7134_RS_PITCH(5),TS_PACKET_SIZE);
 	saa_writel(SAA7134_RS_CONTROL(5),control);
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->activate = buffer_activate;
 	buf->vb.field = field;
 	return 0;
@@ -242,7 +242,7 @@ void saa7134_irq_ts_done(struct saa7134_dev *dev, unsigned long status)
 			if ((status & 0x100000) != 0x100000)
 				goto done;
 		}
-		saa7134_buffer_finish(dev,&dev->ts_q,STATE_DONE);
+		saa7134_buffer_finish(dev,&dev->ts_q,VIDEOBUF_DONE);
 	}
 	saa7134_buffer_next(dev,&dev->ts_q);
 
diff --git a/drivers/media/video/saa7134/saa7134-vbi.c b/drivers/media/video/saa7134/saa7134-vbi.c
index 81a2aedeff5c..f0d5ed9c2b06 100644
--- a/drivers/media/video/saa7134/saa7134-vbi.c
+++ b/drivers/media/video/saa7134/saa7134-vbi.c
@@ -85,7 +85,7 @@ static int buffer_activate(struct saa7134_dev *dev,
 	unsigned long control,base;
 
 	dprintk("buffer_activate [%p]\n",buf);
-	buf->vb.state = STATE_ACTIVE;
+	buf->vb.state = VIDEOBUF_ACTIVE;
 	buf->top_seen = 0;
 
 	task_init(dev,buf,TASK_A);
@@ -136,7 +136,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 	if (buf->vb.size != size)
 		saa7134_dma_free(q,buf);
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
 
 		buf->vb.width  = llength;
@@ -154,7 +154,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 		if (err)
 			goto oops;
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->activate = buffer_activate;
 	buf->vb.field = field;
 	return 0;
@@ -240,7 +240,7 @@ void saa7134_irq_vbi_done(struct saa7134_dev *dev, unsigned long status)
 			goto done;
 
 		dev->vbi_q.curr->vb.field_count = dev->vbi_fieldcount;
-		saa7134_buffer_finish(dev,&dev->vbi_q,STATE_DONE);
+		saa7134_buffer_finish(dev,&dev->vbi_q,VIDEOBUF_DONE);
 	}
 	saa7134_buffer_next(dev,&dev->vbi_q);
 
diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c
index 86fe452842b3..9c160b2bca3d 100644
--- a/drivers/media/video/saa7134/saa7134-video.c
+++ b/drivers/media/video/saa7134/saa7134-video.c
@@ -945,7 +945,7 @@ static int buffer_activate(struct saa7134_dev *dev,
 	unsigned long bpl_uv,lines_uv,base2,base3,tmp; /* planar */
 
 	dprintk("buffer_activate buf=%p\n",buf);
-	buf->vb.state = STATE_ACTIVE;
+	buf->vb.state = VIDEOBUF_ACTIVE;
 	buf->top_seen = 0;
 
 	set_size(dev,TASK_A,buf->vb.width,buf->vb.height,
@@ -1054,7 +1054,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 		saa7134_dma_free(q,buf);
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
 
 		buf->vb.width  = fh->width;
@@ -1074,7 +1074,7 @@ static int buffer_prepare(struct videobuf_queue *q,
 		if (err)
 			goto oops;
 	}
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 	buf->activate = buffer_activate;
 	return 0;
 
@@ -1421,8 +1421,8 @@ video_poll(struct file *file, struct poll_table_struct *wait)
 		return POLLERR;
 
 	poll_wait(file, &buf->done, wait);
-	if (buf->state == STATE_DONE ||
-	    buf->state == STATE_ERROR)
+	if (buf->state == VIDEOBUF_DONE ||
+	    buf->state == VIDEOBUF_ERROR)
 		return POLLIN|POLLRDNORM;
 	return 0;
 }
@@ -2519,7 +2519,7 @@ void saa7134_irq_video_done(struct saa7134_dev *dev, unsigned long status)
 				goto done;
 		}
 		dev->video_q.curr->vb.field_count = dev->video_fieldcount;
-		saa7134_buffer_finish(dev,&dev->video_q,STATE_DONE);
+		saa7134_buffer_finish(dev,&dev->video_q,VIDEOBUF_DONE);
 	}
 	saa7134_buffer_next(dev,&dev->video_q);
 
diff --git a/drivers/media/video/videobuf-core.c b/drivers/media/video/videobuf-core.c
index c8a5cb57963b..26d1a5001738 100644
--- a/drivers/media/video/videobuf-core.c
+++ b/drivers/media/video/videobuf-core.c
@@ -68,14 +68,14 @@ int videobuf_waiton(struct videobuf_buffer *vb, int non_blocking, int intr)
 
 	MAGIC_CHECK(vb->magic,MAGIC_BUFFER);
 	add_wait_queue(&vb->done, &wait);
-	while (vb->state == STATE_ACTIVE || vb->state == STATE_QUEUED) {
+	while (vb->state == VIDEOBUF_ACTIVE || vb->state == VIDEOBUF_QUEUED) {
 		if (non_blocking) {
 			retval = -EAGAIN;
 			break;
 		}
 		set_current_state(intr  ? TASK_INTERRUPTIBLE
 					: TASK_UNINTERRUPTIBLE);
-		if (vb->state == STATE_ACTIVE || vb->state == STATE_QUEUED)
+		if (vb->state == VIDEOBUF_ACTIVE || vb->state == VIDEOBUF_QUEUED)
 			schedule();
 		set_current_state(TASK_RUNNING);
 		if (intr && signal_pending(current)) {
@@ -167,11 +167,11 @@ int videobuf_queue_is_busy(struct videobuf_queue *q)
 			dprintk(1,"busy: buffer #%d mapped\n",i);
 			return 1;
 		}
-		if (q->bufs[i]->state == STATE_QUEUED) {
+		if (q->bufs[i]->state == VIDEOBUF_QUEUED) {
 			dprintk(1,"busy: buffer #%d queued\n",i);
 			return 1;
 		}
-		if (q->bufs[i]->state == STATE_ACTIVE) {
+		if (q->bufs[i]->state == VIDEOBUF_ACTIVE) {
 			dprintk(1,"busy: buffer #%d avtive\n",i);
 			return 1;
 		}
@@ -191,9 +191,9 @@ void videobuf_queue_cancel(struct videobuf_queue *q)
 	for (i = 0; i < VIDEO_MAX_FRAME; i++) {
 		if (NULL == q->bufs[i])
 			continue;
-		if (q->bufs[i]->state == STATE_QUEUED) {
+		if (q->bufs[i]->state == VIDEOBUF_QUEUED) {
 			list_del(&q->bufs[i]->queue);
-			q->bufs[i]->state = STATE_ERROR;
+			q->bufs[i]->state = VIDEOBUF_ERROR;
 		}
 	}
 	if (q->irqlock)
@@ -259,17 +259,17 @@ static void videobuf_status(struct videobuf_queue *q, struct v4l2_buffer *b,
 		b->flags |= V4L2_BUF_FLAG_MAPPED;
 
 	switch (vb->state) {
-	case STATE_PREPARED:
-	case STATE_QUEUED:
-	case STATE_ACTIVE:
+	case VIDEOBUF_PREPARED:
+	case VIDEOBUF_QUEUED:
+	case VIDEOBUF_ACTIVE:
 		b->flags |= V4L2_BUF_FLAG_QUEUED;
 		break;
-	case STATE_DONE:
-	case STATE_ERROR:
+	case VIDEOBUF_DONE:
+	case VIDEOBUF_ERROR:
 		b->flags |= V4L2_BUF_FLAG_DONE;
 		break;
-	case STATE_NEEDS_INIT:
-	case STATE_IDLE:
+	case VIDEOBUF_NEEDS_INIT:
+	case VIDEOBUF_IDLE:
 		/* nothing */
 		break;
 	}
@@ -498,7 +498,7 @@ int videobuf_qbuf(struct videobuf_queue *q,
 		dprintk(1,"qbuf: memory type is wrong.\n");
 		goto done;
 	}
-	if (buf->state != STATE_NEEDS_INIT && buf->state != STATE_IDLE) {
+	if (buf->state != VIDEOBUF_NEEDS_INIT && buf->state != VIDEOBUF_IDLE) {
 		dprintk(1,"qbuf: buffer is already queued or active.\n");
 		goto done;
 	}
@@ -525,7 +525,7 @@ int videobuf_qbuf(struct videobuf_queue *q,
 			dprintk(1,"qbuf: buffer length is not enough\n");
 			goto done;
 		}
-		if (STATE_NEEDS_INIT != buf->state && buf->baddr != b->m.userptr)
+		if (VIDEOBUF_NEEDS_INIT != buf->state && buf->baddr != b->m.userptr)
 			q->ops->buf_release(q,buf);
 		buf->baddr = b->m.userptr;
 		break;
@@ -595,16 +595,16 @@ int videobuf_dqbuf(struct videobuf_queue *q,
 		goto done;
 	}
 	switch (buf->state) {
-	case STATE_ERROR:
+	case VIDEOBUF_ERROR:
 		dprintk(1,"dqbuf: state is error\n");
 		retval = -EIO;
 		CALL(q,sync,q, buf);
-		buf->state = STATE_IDLE;
+		buf->state = VIDEOBUF_IDLE;
 		break;
-	case STATE_DONE:
+	case VIDEOBUF_DONE:
 		dprintk(1,"dqbuf: state is done\n");
 		CALL(q,sync,q, buf);
-		buf->state = STATE_IDLE;
+		buf->state = VIDEOBUF_IDLE;
 		break;
 	default:
 		dprintk(1,"dqbuf: state invalid\n");
@@ -637,7 +637,7 @@ int videobuf_streamon(struct videobuf_queue *q)
 	if (q->irqlock)
 		spin_lock_irqsave(q->irqlock,flags);
 	list_for_each_entry(buf, &q->stream, stream)
-		if (buf->state == STATE_PREPARED)
+		if (buf->state == VIDEOBUF_PREPARED)
 			q->ops->buf_queue(q,buf);
 	if (q->irqlock)
 		spin_unlock_irqrestore(q->irqlock,flags);
@@ -704,7 +704,7 @@ static ssize_t videobuf_read_zerocopy(struct videobuf_queue *q,
 	retval = videobuf_waiton(q->read_buf,0,0);
 	if (0 == retval) {
 		CALL(q,sync,q,q->read_buf);
-		if (STATE_ERROR == q->read_buf->state)
+		if (VIDEOBUF_ERROR == q->read_buf->state)
 			retval = -EIO;
 		else
 			retval = q->read_buf->size;
@@ -778,7 +778,7 @@ ssize_t videobuf_read_one(struct videobuf_queue *q,
 
 	CALL(q,sync,q,q->read_buf);
 
-	if (STATE_ERROR == q->read_buf->state) {
+	if (VIDEOBUF_ERROR == q->read_buf->state) {
 		/* catch I/O errors */
 		q->ops->buf_release(q,q->read_buf);
 		kfree(q->read_buf);
@@ -931,7 +931,7 @@ ssize_t videobuf_read_stream(struct videobuf_queue *q,
 			break;
 		}
 
-		if (q->read_buf->state == STATE_DONE) {
+		if (q->read_buf->state == VIDEOBUF_DONE) {
 			rc = CALL (q,copy_stream, q, data + retval, count,
 					retval, vbihack, nonblocking);
 			if (rc < 0) {
@@ -999,8 +999,8 @@ unsigned int videobuf_poll_stream(struct file *file,
 
 	if (0 == rc) {
 		poll_wait(file, &buf->done, wait);
-		if (buf->state == STATE_DONE ||
-		    buf->state == STATE_ERROR)
+		if (buf->state == VIDEOBUF_DONE ||
+		    buf->state == VIDEOBUF_ERROR)
 			rc = POLLIN|POLLRDNORM;
 	}
 	mutex_unlock(&q->lock);
diff --git a/drivers/media/video/videobuf-dvb.c b/drivers/media/video/videobuf-dvb.c
index 880317e04a02..b73aba65d21d 100644
--- a/drivers/media/video/videobuf-dvb.c
+++ b/drivers/media/video/videobuf-dvb.c
@@ -67,7 +67,7 @@ static int videobuf_dvb_thread(void *data)
 
 		/* feed buffer data to demux */
 		dma=videobuf_to_dma(buf);
-		if (buf->state == STATE_DONE)
+		if (buf->state == VIDEOBUF_DONE)
 			dvb_dmx_swfilter(&dvb->demux, dma->vmalloc,
 					 buf->size);
 
diff --git a/drivers/media/video/vivi.c b/drivers/media/video/vivi.c
index 9b54ff9d2e36..fe9784a0cc21 100644
--- a/drivers/media/video/vivi.c
+++ b/drivers/media/video/vivi.c
@@ -370,7 +370,7 @@ static void vivi_fillbuff(struct vivi_dev *dev,struct vivi_buffer *buf)
 			(unsigned long)tmpbuf,pos);
 
 	/* Advice that buffer was filled */
-	buf->vb.state = STATE_DONE;
+	buf->vb.state = VIDEOBUF_DONE;
 	buf->vb.field_count++;
 	do_gettimeofday(&ts);
 	buf->vb.ts = ts;
@@ -522,7 +522,7 @@ static int restart_video_queue(struct vivi_dmaqueue *dma_q)
 		/* cancel all outstanding capture / vbi requests */
 		list_for_each_entry_safe(buf, prev, &dma_q->active, vb.queue) {
 			list_del(&buf->vb.queue);
-			buf->vb.state = STATE_ERROR;
+			buf->vb.state = VIDEOBUF_ERROR;
 			wake_up(&buf->vb.done);
 		}
 		mod_timer(&dma_q->timeout, jiffies+BUFFER_TIMEOUT);
@@ -543,7 +543,7 @@ static int restart_video_queue(struct vivi_dmaqueue *dma_q)
 			vivi_stop_thread(dma_q);
 			vivi_start_thread(dma_q);
 
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			mod_timer(&dma_q->timeout, jiffies+BUFFER_TIMEOUT);
 			dprintk(2,"[%p/%d] restart_queue - first active\n",
 				buf,buf->vb.i);
@@ -553,7 +553,7 @@ static int restart_video_queue(struct vivi_dmaqueue *dma_q)
 			   prev->fmt       == buf->fmt) {
 			list_del(&buf->vb.queue);
 			list_add_tail(&buf->vb.queue,&dma_q->active);
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			dprintk(2,"[%p/%d] restart_queue - move to active\n",
 				buf,buf->vb.i);
 		} else {
@@ -572,7 +572,7 @@ static void vivi_vid_timeout(unsigned long data)
 	while (!list_empty(&vidq->active)) {
 		buf = list_entry(vidq->active.next, struct vivi_buffer, vb.queue);
 		list_del(&buf->vb.queue);
-		buf->vb.state = STATE_ERROR;
+		buf->vb.state = VIDEOBUF_ERROR;
 		wake_up(&buf->vb.done);
 		printk("vivi/0: [%p/%d] timeout\n", buf, buf->vb.i);
 	}
@@ -610,7 +610,7 @@ static void free_buffer(struct videobuf_queue *vq, struct vivi_buffer *buf)
 
 	videobuf_waiton(&buf->vb,0,0);
 	videobuf_vmalloc_free(&buf->vb);
-	buf->vb.state = STATE_NEEDS_INIT;
+	buf->vb.state = VIDEOBUF_NEEDS_INIT;
 }
 
 #define norm_maxw() 1024
@@ -644,12 +644,12 @@ buffer_prepare(struct videobuf_queue *vq, struct videobuf_buffer *vb,
 		init_buffer = 1;
 	}
 
-	if (STATE_NEEDS_INIT == buf->vb.state) {
+	if (VIDEOBUF_NEEDS_INIT == buf->vb.state) {
 		if (0 != (rc = videobuf_iolock(vq,&buf->vb,NULL)))
 			goto fail;
 	}
 
-	buf->vb.state = STATE_PREPARED;
+	buf->vb.state = VIDEOBUF_PREPARED;
 
 	return 0;
 
@@ -670,13 +670,13 @@ buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 	if (!list_empty(&vidq->queued)) {
 		dprintk(1,"adding vb queue=0x%08lx\n",(unsigned long)&buf->vb.queue);
 		list_add_tail(&buf->vb.queue,&vidq->queued);
-		buf->vb.state = STATE_QUEUED;
+		buf->vb.state = VIDEOBUF_QUEUED;
 		dprintk(2,"[%p/%d] buffer_queue - append to queued\n",
 			buf, buf->vb.i);
 	} else if (list_empty(&vidq->active)) {
 		list_add_tail(&buf->vb.queue,&vidq->active);
 
-		buf->vb.state = STATE_ACTIVE;
+		buf->vb.state = VIDEOBUF_ACTIVE;
 		mod_timer(&vidq->timeout, jiffies+BUFFER_TIMEOUT);
 		dprintk(2,"[%p/%d] buffer_queue - first active\n",
 			buf, buf->vb.i);
@@ -688,13 +688,13 @@ buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 		    prev->vb.height == buf->vb.height &&
 		    prev->fmt       == buf->fmt) {
 			list_add_tail(&buf->vb.queue,&vidq->active);
-			buf->vb.state = STATE_ACTIVE;
+			buf->vb.state = VIDEOBUF_ACTIVE;
 			dprintk(2,"[%p/%d] buffer_queue - append to active\n",
 				buf, buf->vb.i);
 
 		} else {
 			list_add_tail(&buf->vb.queue,&vidq->queued);
-			buf->vb.state = STATE_QUEUED;
+			buf->vb.state = VIDEOBUF_QUEUED;
 			dprintk(2,"[%p/%d] buffer_queue - first queued\n",
 				buf, buf->vb.i);
 		}
diff --git a/include/media/videobuf-core.h b/include/media/videobuf-core.h
index 4fd5d0eaa935..567dfe22c07f 100644
--- a/include/media/videobuf-core.h
+++ b/include/media/videobuf-core.h
@@ -56,13 +56,13 @@ struct videobuf_mapping {
 };
 
 enum videobuf_state {
-	STATE_NEEDS_INIT = 0,
-	STATE_PREPARED   = 1,
-	STATE_QUEUED     = 2,
-	STATE_ACTIVE     = 3,
-	STATE_DONE       = 4,
-	STATE_ERROR      = 5,
-	STATE_IDLE       = 6,
+	VIDEOBUF_NEEDS_INIT = 0,
+	VIDEOBUF_PREPARED   = 1,
+	VIDEOBUF_QUEUED     = 2,
+	VIDEOBUF_ACTIVE     = 3,
+	VIDEOBUF_DONE       = 4,
+	VIDEOBUF_ERROR      = 5,
+	VIDEOBUF_IDLE       = 6,
 };
 
 struct videobuf_buffer {
-- 
cgit v1.2.3


From d6964aa8d4a418d69da1652121940fabc6b5d591 Mon Sep 17 00:00:00 2001
From: Brandon Philips <brandon@ifup.org>
Date: Tue, 6 Nov 2007 20:23:08 -0300
Subject: V4L/DVB (6603): V4L: videobuf: convert streaming and reading to
 bitfields

Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/videobuf-core.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/media/videobuf-core.h b/include/media/videobuf-core.h
index 567dfe22c07f..7aa7a7b64c1b 100644
--- a/include/media/videobuf-core.h
+++ b/include/media/videobuf-core.h
@@ -162,12 +162,12 @@ struct videobuf_queue {
 	struct videobuf_queue_ops  *ops;
 	struct videobuf_qtype_ops  *int_ops;
 
+	unsigned int               streaming:1;
+	unsigned int               reading:1;
 	/* capture via mmap() + ioctl(QBUF/DQBUF) */
-	unsigned int               streaming;
 	struct list_head           stream;
 
 	/* capture via read() */
-	unsigned int               reading;
 	unsigned int               read_off;
 	struct videobuf_buffer     *read_buf;
 
-- 
cgit v1.2.3


From 2a2bfbff3d867c62a49f43266ca993c9c2c2ff28 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 8 Dec 2007 07:43:14 -0300
Subject: V4L/DVB (6766): ivtv: remove i2c legacy support from drivers that no
 longer need it

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/m52790.c    | 4 ----
 drivers/media/video/saa7127.c   | 4 ----
 drivers/media/video/upd64031a.c | 4 ----
 drivers/media/video/upd64083.c  | 4 ----
 drivers/media/video/vp27smpx.c  | 4 ----
 drivers/media/video/wm8739.c    | 3 ---
 include/media/v4l2-i2c-drv.h    | 1 -
 7 files changed, 24 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/m52790.c b/drivers/media/video/m52790.c
index 51ecbfbc1ae7..4d2a52204ad4 100644
--- a/drivers/media/video/m52790.c
+++ b/drivers/media/video/m52790.c
@@ -36,10 +36,6 @@ MODULE_DESCRIPTION("i2c device driver for m52790 A/V switch");
 MODULE_AUTHOR("Hans Verkuil");
 MODULE_LICENSE("GPL");
 
-static unsigned short normal_i2c[] = { 0x90 >> 1, I2C_CLIENT_END };
-
-
-I2C_CLIENT_INSMOD;
 
 struct m52790_state {
 	u16 input;
diff --git a/drivers/media/video/saa7127.c b/drivers/media/video/saa7127.c
index 0262acde0888..a5fbfe025333 100644
--- a/drivers/media/video/saa7127.c
+++ b/drivers/media/video/saa7127.c
@@ -69,10 +69,6 @@ module_param(test_image, int, 0644);
 MODULE_PARM_DESC(debug, "debug level (0-2)");
 MODULE_PARM_DESC(test_image, "test_image (0-1)");
 
-static unsigned short normal_i2c[] = { 0x88 >> 1, I2C_CLIENT_END };
-
-
-I2C_CLIENT_INSMOD;
 
 /*
  * SAA7127 registers
diff --git a/drivers/media/video/upd64031a.c b/drivers/media/video/upd64031a.c
index 0318432d85f8..1b162f2dca95 100644
--- a/drivers/media/video/upd64031a.c
+++ b/drivers/media/video/upd64031a.c
@@ -49,10 +49,6 @@ module_param(debug, int, 0644);
 
 MODULE_PARM_DESC(debug, "Debug level (0-1)");
 
-static unsigned short normal_i2c[] = { 0x24 >> 1, 0x26 >> 1, I2C_CLIENT_END };
-
-
-I2C_CLIENT_INSMOD;
 
 enum {
 	R00 = 0, R01, R02, R03, R04,
diff --git a/drivers/media/video/upd64083.c b/drivers/media/video/upd64083.c
index 7e32c5b0c29d..0f9e86cfc628 100644
--- a/drivers/media/video/upd64083.c
+++ b/drivers/media/video/upd64083.c
@@ -39,10 +39,6 @@ module_param(debug, bool, 0644);
 
 MODULE_PARM_DESC(debug, "Debug level (0-1)");
 
-static unsigned short normal_i2c[] = { 0xb8 >> 1, 0xba >> 1, I2C_CLIENT_END };
-
-
-I2C_CLIENT_INSMOD;
 
 enum {
 	R00 = 0, R01, R02, R03, R04,
diff --git a/drivers/media/video/vp27smpx.c b/drivers/media/video/vp27smpx.c
index 97e24769e65b..cd98084ac090 100644
--- a/drivers/media/video/vp27smpx.c
+++ b/drivers/media/video/vp27smpx.c
@@ -36,10 +36,6 @@ MODULE_DESCRIPTION("vp27smpx driver");
 MODULE_AUTHOR("Hans Verkuil");
 MODULE_LICENSE("GPL");
 
-static unsigned short normal_i2c[] = { 0xb6 >> 1, I2C_CLIENT_END };
-
-
-I2C_CLIENT_INSMOD;
 
 /* ----------------------------------------------------------------------- */
 
diff --git a/drivers/media/video/wm8739.c b/drivers/media/video/wm8739.c
index ac4fee1d0251..1c30a5c8455c 100644
--- a/drivers/media/video/wm8739.c
+++ b/drivers/media/video/wm8739.c
@@ -37,15 +37,12 @@ MODULE_AUTHOR("T. Adachi, Hans Verkuil");
 MODULE_LICENSE("GPL");
 
 static int debug;
-static unsigned short normal_i2c[] = { 0x34 >> 1, 0x36 >> 1, I2C_CLIENT_END };
 
 module_param(debug, int, 0644);
 
 MODULE_PARM_DESC(debug, "Debug level (0-1)");
 
 
-I2C_CLIENT_INSMOD;
-
 /* ------------------------------------------------------------------------ */
 
 enum {
diff --git a/include/media/v4l2-i2c-drv.h b/include/media/v4l2-i2c-drv.h
index bc61ab27779d..d758b765e0ff 100644
--- a/include/media/v4l2-i2c-drv.h
+++ b/include/media/v4l2-i2c-drv.h
@@ -34,7 +34,6 @@ struct v4l2_i2c_driver_data {
 };
 
 static struct v4l2_i2c_driver_data v4l2_i2c_data;
-static struct i2c_client_address_data addr_data;
 static struct i2c_driver v4l2_i2c_driver;
 
 
-- 
cgit v1.2.3


From ab1660503ac3af7febfcf987648509b484d4feda Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@linuxtv.org>
Date: Sun, 9 Dec 2007 02:26:48 -0300
Subject: V4L/DVB (6785): tda8290: remove dependency on struct tuner

- remove dependency of tda8290 module on struct tuner
- move tuner_foo printk macros from tuner-driver.h into tuner-core.c
- clean up #includes of tuner-i2c.h / tuner-driver.h

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/tda8290.c      | 103 ++++++++++++++++++-------------------
 drivers/media/video/tda8290.h      |  24 ++++++---
 drivers/media/video/tda9887.c      |   1 +
 drivers/media/video/tuner-core.c   |  43 ++++++++++++++--
 drivers/media/video/tuner-driver.h |  27 ----------
 drivers/media/video/tuner-i2c.h    |   2 -
 include/media/tuner.h              |   2 -
 7 files changed, 109 insertions(+), 93 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/tda8290.c b/drivers/media/video/tda8290.c
index a39cb1f9e875..d0d13bc61250 100644
--- a/drivers/media/video/tda8290.c
+++ b/drivers/media/video/tda8290.c
@@ -23,12 +23,14 @@
 #include <linux/i2c.h>
 #include <linux/delay.h>
 #include <linux/videodev.h>
+#include "tuner-driver.h"
+#include "tuner-i2c.h"
 #include "tda8290.h"
 #include "tda827x.h"
 #include "tda18271.h"
 
-static int tuner_debug;
-module_param_named(debug, tuner_debug, int, 0644);
+static int debug;
+module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "enable verbose debug messages");
 
 #define PREFIX "tda8290"
@@ -50,8 +52,6 @@ struct tda8290_priv {
 #define TDA18271 16
 
 	struct tda827x_config cfg;
-
-	struct tuner *t;
 };
 
 /*---------------------------------------------------------------------*/
@@ -114,7 +114,6 @@ static void set_audio(struct dvb_frontend *fe,
 		      struct analog_parameters *params)
 {
 	struct tda8290_priv *priv = fe->analog_demod_priv;
-	struct tuner *t = priv->t;
 	char* mode;
 
 	if (params->std & V4L2_STD_MN) {
@@ -150,7 +149,6 @@ static void tda8290_set_params(struct dvb_frontend *fe,
 			       struct analog_parameters *params)
 {
 	struct tda8290_priv *priv = fe->analog_demod_priv;
-	struct tuner *t = priv->t;
 
 	unsigned char soft_reset[]  = { 0x00, 0x00 };
 	unsigned char easy_mode[]   = { 0x01, priv->tda8290_easy_mode };
@@ -176,7 +174,8 @@ static void tda8290_set_params(struct dvb_frontend *fe,
 
 	set_audio(fe, params);
 
-	tuner_dbg("tda827xa config is 0x%02x\n", t->config);
+	if (priv->cfg.config)
+		tuner_dbg("tda827xa config is 0x%02x\n", *priv->cfg.config);
 	tuner_i2c_xfer_send(&priv->i2c_props, easy_mode, 2);
 	tuner_i2c_xfer_send(&priv->i2c_props, agc_out_on, 2);
 	tuner_i2c_xfer_send(&priv->i2c_props, soft_reset, 2);
@@ -362,7 +361,6 @@ static void tda8295_set_params(struct dvb_frontend *fe,
 			       struct analog_parameters *params)
 {
 	struct tda8290_priv *priv = fe->analog_demod_priv;
-	struct tuner *t = priv->t;
 
 	unsigned char blanking_mode[]     = { 0x1d, 0x00 };
 
@@ -442,13 +440,13 @@ static void tda8295_standby(struct dvb_frontend *fe)
 static void tda8290_init_if(struct dvb_frontend *fe)
 {
 	struct tda8290_priv *priv = fe->analog_demod_priv;
-	struct tuner *t = priv->t;
 
 	unsigned char set_VS[] = { 0x30, 0x6F };
 	unsigned char set_GP00_CF[] = { 0x20, 0x01 };
 	unsigned char set_GP01_CF[] = { 0x20, 0x0B };
 
-	if ((t->config == 1) || (t->config == 2))
+	if ((priv->cfg.config) &&
+	    ((*priv->cfg.config == 1) || (*priv->cfg.config == 2)))
 		tuner_i2c_xfer_send(&priv->i2c_props, set_GP00_CF, 2);
 	else
 		tuner_i2c_xfer_send(&priv->i2c_props, set_GP01_CF, 2);
@@ -516,7 +514,6 @@ static int tda829x_find_tuner(struct dvb_frontend *fe)
 {
 	struct tda8290_priv *priv = fe->analog_demod_priv;
 	struct analog_tuner_ops *ops = fe->ops.analog_demod_ops;
-	struct tuner *t = priv->t;
 	int i, ret, tuners_found;
 	u32 tuner_addrs;
 	u8 data;
@@ -595,29 +592,6 @@ static int tda829x_find_tuner(struct dvb_frontend *fe)
 
 	ops->i2c_gate_ctrl(fe, 0);
 
-	switch (priv->ver) {
-	case TDA8290 | TDA8275:
-		strlcpy(t->i2c->name, "tda8290+75", sizeof(t->i2c->name));
-		break;
-	case TDA8295 | TDA8275:
-		strlcpy(t->i2c->name, "tda8295+75", sizeof(t->i2c->name));
-		break;
-	case TDA8290 | TDA8275A:
-		strlcpy(t->i2c->name, "tda8290+75a", sizeof(t->i2c->name));
-		break;
-	case TDA8295 | TDA8275A:
-		strlcpy(t->i2c->name, "tda8295+75a", sizeof(t->i2c->name));
-		break;
-	case TDA8290 | TDA18271:
-		strlcpy(t->i2c->name, "tda8290+18271", sizeof(t->i2c->name));
-		break;
-	case TDA8295 | TDA18271:
-		strlcpy(t->i2c->name, "tda8295+18271", sizeof(t->i2c->name));
-		break;
-	default:
-		return -EINVAL;
-	}
-
 	return 0;
 }
 
@@ -631,7 +605,7 @@ static int tda8290_probe(struct tuner_i2c_props *i2c_props)
 	tuner_i2c_xfer_recv(i2c_props, &tda8290_id[1], 1);
 
 	if (tda8290_id[1] == TDA8290_ID) {
-		if (tuner_debug)
+		if (debug)
 			printk(KERN_DEBUG "%s: tda8290 detected @ %d-%04x\n",
 			       __FUNCTION__, i2c_adapter_id(i2c_props->adap),
 			       i2c_props->addr);
@@ -651,7 +625,7 @@ static int tda8295_probe(struct tuner_i2c_props *i2c_props)
 	tuner_i2c_xfer_recv(i2c_props, &tda8295_id[1], 1);
 
 	if (tda8295_id[1] == TDA8295_ID) {
-		if (tuner_debug)
+		if (debug)
 			printk(KERN_DEBUG "%s: tda8295 detected @ %d-%04x\n",
 			       __FUNCTION__, i2c_adapter_id(i2c_props->adap),
 			       i2c_props->addr);
@@ -677,21 +651,24 @@ static struct analog_tuner_ops tda8295_tuner_ops = {
 	.i2c_gate_ctrl  = tda8295_i2c_bridge,
 };
 
-int tda829x_attach(struct tuner *t)
+struct dvb_frontend *tda829x_attach(struct dvb_frontend *fe,
+				    struct i2c_adapter *i2c_adap, u8 i2c_addr,
+				    struct tda829x_config *cfg)
 {
-	struct dvb_frontend *fe = &t->fe;
 	struct tda8290_priv *priv = NULL;
+	char *name;
 
 	priv = kzalloc(sizeof(struct tda8290_priv), GFP_KERNEL);
 	if (priv == NULL)
-		return -ENOMEM;
+		return NULL;
 	fe->analog_demod_priv = priv;
 
-	priv->i2c_props.addr     = t->i2c->addr;
-	priv->i2c_props.adap     = t->i2c->adapter;
-	priv->cfg.config         = &t->config;
-	priv->cfg.tuner_callback = t->tuner_callback;
-	priv->t = t;
+	priv->i2c_props.addr     = i2c_addr;
+	priv->i2c_props.adap     = i2c_adap;
+	if (cfg) {
+		priv->cfg.config         = cfg->lna_cfg;
+		priv->cfg.tuner_callback = cfg->tuner_callback;
+	}
 
 	if (tda8290_probe(&priv->i2c_props) == 0) {
 		priv->ver = TDA8290;
@@ -706,30 +683,50 @@ int tda829x_attach(struct tuner *t)
 	if (tda829x_find_tuner(fe) < 0)
 		goto fail;
 
+	switch (priv->ver) {
+	case TDA8290 | TDA8275:
+		name = "tda8290+75";
+		break;
+	case TDA8295 | TDA8275:
+		name = "tda8295+75";
+		break;
+	case TDA8290 | TDA8275A:
+		name = "tda8290+75a";
+		break;
+	case TDA8295 | TDA8275A:
+		name = "tda8295+75a";
+		break;
+	case TDA8290 | TDA18271:
+		name = "tda8290+18271";
+		break;
+	case TDA8295 | TDA18271:
+		name = "tda8295+18271";
+		break;
+	default:
+		goto fail;
+	}
+	tuner_info("type set to %s\n", name);
+
 	if (priv->ver & TDA8290) {
 		tda8290_init_tuner(fe);
 		tda8290_init_if(fe);
 	} else if (priv->ver & TDA8295)
 		tda8295_init_if(fe);
 
-	tuner_info("type set to %s\n", t->i2c->name);
-
-	t->mode = V4L2_TUNER_ANALOG_TV;
-
-	return 0;
+	return fe;
 
 fail:
 	tda829x_release(fe);
 	fe->ops.analog_demod_ops = NULL;
-	return -EINVAL;
+	return NULL;
 }
 EXPORT_SYMBOL_GPL(tda829x_attach);
 
-int tda829x_probe(struct tuner *t)
+int tda829x_probe(struct i2c_adapter *i2c_adap, u8 i2c_addr)
 {
 	struct tuner_i2c_props i2c_props = {
-		.adap = t->i2c->adapter,
-		.addr = t->i2c->addr
+		.adap = i2c_adap,
+		.addr = i2c_addr,
 	};
 
 	unsigned char soft_reset[]   = { 0x00, 0x00 };
diff --git a/drivers/media/video/tda8290.h b/drivers/media/video/tda8290.h
index 3a1f04520615..4c547577e7ca 100644
--- a/drivers/media/video/tda8290.h
+++ b/drivers/media/video/tda8290.h
@@ -18,24 +18,36 @@
 #define __TDA8290_H__
 
 #include <linux/i2c.h>
-#include "tuner-driver.h"
+#include "dvb_frontend.h"
+
+struct tda829x_config
+{
+	unsigned int *lna_cfg;
+	int (*tuner_callback) (void *dev, int command, int arg);
+};
 
 #if defined(CONFIG_TUNER_TDA8290) || (defined(CONFIG_TUNER_TDA8290_MODULE) && defined(MODULE))
-extern int tda829x_probe(struct tuner *t);
+extern int tda829x_probe(struct i2c_adapter *i2c_adap, u8 i2c_addr);
 
-extern int tda829x_attach(struct tuner *t);
+extern struct dvb_frontend *tda829x_attach(struct dvb_frontend *fe,
+					   struct i2c_adapter *i2c_adap,
+					   u8 i2c_addr,
+					   struct tda829x_config *cfg);
 #else
-static inline int tda829x_probe(struct tuner *t)
+static inline int tda829x_probe(struct i2c_adapter *i2c_adap, u8 i2c_addr)
 {
 	printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__);
 	return -EINVAL;
 }
 
-static inline int tda829x_attach(struct tuner *t)
+static inline struct dvb_frontend *tda829x_attach(struct dvb_frontend *fe,
+						  struct i2c_adapter *i2c_adap,
+						  u8 i2c_addr,
+						  struct tda829x_config *cfg)
 {
 	printk(KERN_INFO "%s: not probed - driver disabled by Kconfig\n",
 	       __FUNCTION__);
-	return -EINVAL;
+	return NULL;
 }
 #endif
 
diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c
index d506d4bf8253..98d45c4912bd 100644
--- a/drivers/media/video/tda9887.c
+++ b/drivers/media/video/tda9887.c
@@ -9,6 +9,7 @@
 #include <linux/videodev.h>
 #include <media/v4l2-common.h>
 #include <media/tuner.h>
+#include "tuner-i2c.h"
 #include "tda9887.h"
 
 
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 5f8bffc8209b..0cc190761e24 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -52,7 +52,34 @@ static unsigned int no_autodetect = 0;
 static unsigned int show_i2c = 0;
 
 /* insmod options used at runtime => read/write */
-int tuner_debug = 0;
+static int tuner_debug;
+
+#define tuner_warn(fmt, arg...) do {			\
+	printk(KERN_WARNING "%s %d-%04x: " fmt, PREFIX, \
+	       i2c_adapter_id(t->i2c->adapter),		\
+	       t->i2c->addr, ##arg);			\
+	 } while (0)
+
+#define tuner_info(fmt, arg...) do {			\
+	printk(KERN_INFO "%s %d-%04x: " fmt, PREFIX,	\
+	       i2c_adapter_id(t->i2c->adapter),		\
+	       t->i2c->addr, ##arg);			\
+	 } while (0)
+
+#define tuner_err(fmt, arg...) do {			\
+	printk(KERN_ERR "%s %d-%04x: " fmt, PREFIX,	\
+	       i2c_adapter_id(t->i2c->adapter),		\
+	       t->i2c->addr, ##arg);			\
+	 } while (0)
+
+#define tuner_dbg(fmt, arg...) do {				\
+	if (tuner_debug)					\
+		printk(KERN_DEBUG "%s %d-%04x: " fmt, PREFIX,	\
+		       i2c_adapter_id(t->i2c->adapter),		\
+		       t->i2c->addr, ##arg);			\
+	 } while (0)
+
+/* ------------------------------------------------------------------------ */
 
 static unsigned int tv_range[2] = { 44, 958 };
 static unsigned int radio_range[2] = { 65, 108 };
@@ -261,6 +288,15 @@ static void attach_simple_tuner(struct tuner *t)
 	simple_tuner_attach(&t->fe, t->i2c->adapter, t->i2c->addr, &cfg);
 }
 
+static void attach_tda829x(struct tuner *t)
+{
+	struct tda829x_config cfg = {
+		.lna_cfg        = &t->config,
+		.tuner_callback = t->tuner_callback,
+	};
+	tda829x_attach(&t->fe, t->i2c->adapter, t->i2c->addr, &cfg);
+}
+
 static void set_type(struct i2c_client *c, unsigned int type,
 		     unsigned int new_mode_mask, unsigned int new_config,
 		     int (*tuner_callback) (void *dev, int command,int arg))
@@ -303,7 +339,7 @@ static void set_type(struct i2c_client *c, unsigned int type,
 		break;
 	case TUNER_PHILIPS_TDA8290:
 	{
-		tda829x_attach(t);
+		attach_tda829x(t);
 		break;
 	}
 	case TUNER_TEA5767:
@@ -1045,7 +1081,8 @@ static int tuner_probe(struct i2c_client *client)
 		case 0x4b:
 			/* If chip is not tda8290, don't register.
 			   since it can be tda9887*/
-			if (tda829x_probe(t) == 0) {
+			if (tda829x_probe(t->i2c->adapter,
+					  t->i2c->addr) == 0) {
 				tuner_dbg("tda829x detected\n");
 			} else {
 				/* Default is being tda9887 */
diff --git a/drivers/media/video/tuner-driver.h b/drivers/media/video/tuner-driver.h
index 91913ee08a07..a215161d0241 100644
--- a/drivers/media/video/tuner-driver.h
+++ b/drivers/media/video/tuner-driver.h
@@ -24,13 +24,10 @@
 
 #include <linux/videodev2.h>
 #include <linux/i2c.h>
-#include "tuner-i2c.h"
 #include "dvb_frontend.h"
 
 extern unsigned const int tuner_count;
 
-struct tuner;
-
 struct analog_tuner_ops {
 	void (*set_params)(struct dvb_frontend *fe,
 			   struct analog_parameters *params);
@@ -69,28 +66,4 @@ struct tuner {
 	int (*tuner_callback) (void *dev, int command,int arg);
 };
 
-/* ------------------------------------------------------------------------ */
-
-#define tuner_warn(fmt, arg...) do {					\
-	printk(KERN_WARNING "%s %d-%04x: " fmt, PREFIX, 		\
-	       i2c_adapter_id(t->i2c->adapter), t->i2c->addr, ##arg);	\
-	 } while (0)
-
-#define tuner_info(fmt, arg...) do {					\
-	printk(KERN_INFO "%s %d-%04x: " fmt, PREFIX,			\
-	       i2c_adapter_id(t->i2c->adapter), t->i2c->addr , ##arg);	\
-	 } while (0)
-
-#define tuner_err(fmt, arg...) do {					\
-	printk(KERN_ERR "%s %d-%04x: " fmt, PREFIX,			\
-	       i2c_adapter_id(t->i2c->adapter), t->i2c->addr , ##arg);	\
-	 } while (0)
-
-#define tuner_dbg(fmt, arg...) do {					       \
-	extern int tuner_debug;						       \
-	if (tuner_debug)						       \
-		printk(KERN_DEBUG "%s %d-%04x: " fmt, PREFIX,		       \
-		       i2c_adapter_id(t->i2c->adapter), t->i2c->addr , ##arg); \
-	 } while (0)
-
 #endif /* __TUNER_DRIVER_H__ */
diff --git a/drivers/media/video/tuner-i2c.h b/drivers/media/video/tuner-i2c.h
index d7cf72c3fd71..de52e8ffd347 100644
--- a/drivers/media/video/tuner-i2c.h
+++ b/drivers/media/video/tuner-i2c.h
@@ -59,7 +59,6 @@ static inline int tuner_i2c_xfer_send_recv(struct tuner_i2c_props *props,
 	return (ret == 2) ? ilen : ret;
 }
 
-#ifndef __TUNER_DRIVER_H__
 #define tuner_warn(fmt, arg...) do {					\
 	printk(KERN_WARNING "%s %d-%04x: " fmt, PREFIX,			\
 			i2c_adapter_id(priv->i2c_props.adap),		\
@@ -84,6 +83,5 @@ static inline int tuner_i2c_xfer_send_recv(struct tuner_i2c_props *props,
 			i2c_adapter_id(priv->i2c_props.adap),		\
 			priv->i2c_props.addr , ##arg);			\
 	} while (0)
-#endif /* __TUNER_DRIVER_H__ */
 
 #endif /* __TUNER_I2C_H__ */
diff --git a/include/media/tuner.h b/include/media/tuner.h
index d49392d90e56..97be269afbee 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -24,8 +24,6 @@
 
 #include <linux/videodev2.h>
 
-extern int tuner_debug;
-
 #define ADDR_UNSET (255)
 
 #define TUNER_TEMIC_PAL			0        /* 4002 FH5 (3X 7756, 9483) */
-- 
cgit v1.2.3


From ea48c13ad0e5626b827bd7076c22df1a352e3983 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Wed, 12 Dec 2007 07:04:58 -0300
Subject: V4L/DVB (6861): cx2341x: command argument should be u32 instead of
 int

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/cx2341x.c                 | 2 +-
 drivers/media/video/cx88/cx88-blackbird.c     | 2 +-
 drivers/media/video/ivtv/ivtv-mailbox.c       | 2 +-
 drivers/media/video/ivtv/ivtv-mailbox.h       | 2 +-
 drivers/media/video/pvrusb2/pvrusb2-encoder.c | 2 +-
 include/media/cx2341x.h                       | 2 +-
 6 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/cx2341x.c b/drivers/media/video/cx2341x.c
index 890c8867af1c..c592899a2317 100644
--- a/drivers/media/video/cx2341x.c
+++ b/drivers/media/video/cx2341x.c
@@ -775,7 +775,7 @@ void cx2341x_fill_defaults(struct cx2341x_mpeg_params *p)
 EXPORT_SYMBOL(cx2341x_fill_defaults);
 
 static int cx2341x_api(void *priv, cx2341x_mbox_func func,
-		       int cmd, int args, ...)
+		       u32 cmd, int args, ...)
 {
 	u32 data[CX2341X_MBOX_MAX_DATA];
 	va_list vargs;
diff --git a/drivers/media/video/cx88/cx88-blackbird.c b/drivers/media/video/cx88/cx88-blackbird.c
index 5f79442924fd..a99e9d5950aa 100644
--- a/drivers/media/video/cx88/cx88-blackbird.c
+++ b/drivers/media/video/cx88/cx88-blackbird.c
@@ -307,7 +307,7 @@ static int register_read(struct cx88_core *core, u32 address, u32 *value)
 
 /* ------------------------------------------------------------------ */
 
-static int blackbird_mbox_func(void *priv, int command, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA])
+static int blackbird_mbox_func(void *priv, u32 command, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA])
 {
 	struct cx8802_dev *dev = priv;
 	unsigned long timeout;
diff --git a/drivers/media/video/ivtv/ivtv-mailbox.c b/drivers/media/video/ivtv/ivtv-mailbox.c
index b05436da7136..13a6c374d2db 100644
--- a/drivers/media/video/ivtv/ivtv-mailbox.c
+++ b/drivers/media/video/ivtv/ivtv-mailbox.c
@@ -333,7 +333,7 @@ int ivtv_api(struct ivtv *itv, int cmd, int args, u32 data[])
 	return (res == -EBUSY) ? ivtv_api_call(itv, cmd, args, data) : res;
 }
 
-int ivtv_api_func(void *priv, int cmd, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA])
+int ivtv_api_func(void *priv, u32 cmd, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA])
 {
 	return ivtv_api(priv, cmd, in, data);
 }
diff --git a/drivers/media/video/ivtv/ivtv-mailbox.h b/drivers/media/video/ivtv/ivtv-mailbox.h
index 71a54eef8fc7..6ef12091e3f3 100644
--- a/drivers/media/video/ivtv/ivtv-mailbox.h
+++ b/drivers/media/video/ivtv/ivtv-mailbox.h
@@ -28,6 +28,6 @@ void ivtv_api_get_data(struct ivtv_mailbox_data *mbox, int mb, u32 data[]);
 int ivtv_api(struct ivtv *itv, int cmd, int args, u32 data[]);
 int ivtv_vapi_result(struct ivtv *itv, u32 data[CX2341X_MBOX_MAX_DATA], int cmd, int args, ...);
 int ivtv_vapi(struct ivtv *itv, int cmd, int args, ...);
-int ivtv_api_func(void *priv, int cmd, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA]);
+int ivtv_api_func(void *priv, u32 cmd, int in, int out, u32 data[CX2341X_MBOX_MAX_DATA]);
 
 #endif
diff --git a/drivers/media/video/pvrusb2/pvrusb2-encoder.c b/drivers/media/video/pvrusb2/pvrusb2-encoder.c
index 52b39954f990..64062879981e 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-encoder.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-encoder.c
@@ -140,7 +140,7 @@ static int pvr2_encoder_read_words(struct pvr2_hdw *hdw,
    cx2341x.ko to write to our encoder (by handing it a pointer to this
    function).  For earlier kernels this doesn't really matter. */
 static int pvr2_encoder_cmd(void *ctxt,
-			    int cmd,
+			    u32 cmd,
 			    int arg_cnt_send,
 			    int arg_cnt_recv,
 			    u32 *argp)
diff --git a/include/media/cx2341x.h b/include/media/cx2341x.h
index af8071d7620d..5f4608e88476 100644
--- a/include/media/cx2341x.h
+++ b/include/media/cx2341x.h
@@ -83,7 +83,7 @@ struct cx2341x_mpeg_params {
 #define CX2341X_MBOX_MAX_DATA 16
 
 extern const u32 cx2341x_mpeg_ctrls[];
-typedef int (*cx2341x_mbox_func)(void *priv, int cmd, int in, int out,
+typedef int (*cx2341x_mbox_func)(void *priv, u32 cmd, int in, int out,
 		u32 data[CX2341X_MBOX_MAX_DATA]);
 int cx2341x_update(void *priv, cx2341x_mbox_func func,
 		const struct cx2341x_mpeg_params *old,
-- 
cgit v1.2.3


From 0b394def21e7d3bd02aeee5570473582ce7984ec Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 18 Dec 2007 19:27:31 -0300
Subject: V4L/DVB (6868): i2c-id.h: add I2C_DRIVERID_CS5345

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/linux/i2c-id.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index adbdc6da39ca..c7a51a196f51 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -126,6 +126,7 @@
 #define I2C_DRIVERID_VP27SMPX	93	/* Panasonic VP27s tuner internal MPX */
 #define I2C_DRIVERID_CS4270	94	/* Cirrus Logic 4270 audio codec */
 #define I2C_DRIVERID_M52790 	95      /* Mitsubishi M52790SP/FP AV switch */
+#define I2C_DRIVERID_CS5345	96	/* cs5345 audio processor	*/
 
 #define I2C_DRIVERID_I2CDEV	900
 #define I2C_DRIVERID_ARP        902    /* SMBus ARP Client              */
-- 
cgit v1.2.3


From 6fb377f85cb8c2c1580ce8b134c887a7b53c7aa9 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 18 Dec 2007 19:40:44 -0300
Subject: V4L/DVB (6869): cs5345: new i2c driver

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/Kconfig     |  10 +++
 drivers/media/video/Makefile    |   1 +
 drivers/media/video/cs5345.c    | 168 ++++++++++++++++++++++++++++++++++++++++
 include/media/cs5345.h          |  39 ++++++++++
 include/media/v4l2-chip-ident.h |   3 +
 5 files changed, 221 insertions(+)
 create mode 100644 drivers/media/video/cs5345.c
 create mode 100644 include/media/cs5345.h

(limited to 'include')

diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index 6c7835629bcd..7431f6da5cae 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -109,6 +109,16 @@ config VIDEO_MSP3400
 	  To compile this driver as a module, choose M here: the
 	  module will be called msp3400.
 
+config VIDEO_CS5345
+	tristate "Cirrus Logic CS5345 audio ADC"
+	depends on VIDEO_V4L2 && I2C && EXPERIMENTAL
+	---help---
+	  Support for the Cirrus Logic CS5345 24-bit, 192 kHz
+	  stereo A/D converter.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called cs5345.
+
 config VIDEO_CS53L32A
 	tristate "Cirrus Logic CS53L32A audio ADC"
 	depends on VIDEO_V4L2 && I2C
diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile
index a837e1ea518f..74a2a1c45fe3 100644
--- a/drivers/media/video/Makefile
+++ b/drivers/media/video/Makefile
@@ -66,6 +66,7 @@ obj-$(CONFIG_VIDEO_USBVISION) += usbvision/
 obj-$(CONFIG_VIDEO_TVP5150) += tvp5150.o
 obj-$(CONFIG_VIDEO_PVRUSB2) += pvrusb2/
 obj-$(CONFIG_VIDEO_MSP3400) += msp3400.o
+obj-$(CONFIG_VIDEO_CS5345) += cs5345.o
 obj-$(CONFIG_VIDEO_CS53L32A) += cs53l32a.o
 obj-$(CONFIG_VIDEO_M52790) += m52790.o
 obj-$(CONFIG_VIDEO_TLV320AIC23B) += tlv320aic23b.o
diff --git a/drivers/media/video/cs5345.c b/drivers/media/video/cs5345.c
new file mode 100644
index 000000000000..fae469ce16f5
--- /dev/null
+++ b/drivers/media/video/cs5345.c
@@ -0,0 +1,168 @@
+/*
+ * cs5345 Cirrus Logic 24-bit, 192 kHz Stereo Audio ADC
+ * Copyright (C) 2007 Hans Verkuil
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/i2c.h>
+#include <linux/videodev2.h>
+#include <media/v4l2-i2c-drv.h>
+#include <media/v4l2-common.h>
+#include <media/v4l2-chip-ident.h>
+
+MODULE_DESCRIPTION("i2c device driver for cs5345 Audio ADC");
+MODULE_AUTHOR("Hans Verkuil");
+MODULE_LICENSE("GPL");
+
+static int debug;
+
+module_param(debug, bool, 0644);
+
+MODULE_PARM_DESC(debug, "Debugging messages\n\t\t\t0=Off (default), 1=On");
+
+
+/* ----------------------------------------------------------------------- */
+
+static inline int cs5345_write(struct i2c_client *client, u8 reg, u8 value)
+{
+	return i2c_smbus_write_byte_data(client, reg, value);
+}
+
+static inline int cs5345_read(struct i2c_client *client, u8 reg)
+{
+	return i2c_smbus_read_byte_data(client, reg);
+}
+
+static int cs5345_command(struct i2c_client *client, unsigned cmd, void *arg)
+{
+	struct v4l2_routing *route = arg;
+	struct v4l2_control *ctrl = arg;
+
+	switch (cmd) {
+	case VIDIOC_INT_G_AUDIO_ROUTING:
+		route->input = cs5345_read(client, 0x09) & 7;
+		route->input |= cs5345_read(client, 0x05) & 0x70;
+		route->output = 0;
+		break;
+
+	case VIDIOC_INT_S_AUDIO_ROUTING:
+		if ((route->input & 0xf) > 6) {
+			v4l_err(client, "Invalid input %d.\n", route->input);
+			return -EINVAL;
+		}
+		cs5345_write(client, 0x09, route->input & 0xf);
+		cs5345_write(client, 0x05, route->input & 0xf0);
+		break;
+
+	case VIDIOC_G_CTRL:
+		if (ctrl->id == V4L2_CID_AUDIO_MUTE) {
+			ctrl->value = (cs5345_read(client, 0x04) & 0x08) != 0;
+			break;
+		}
+		if (ctrl->id != V4L2_CID_AUDIO_VOLUME)
+			return -EINVAL;
+		ctrl->value = cs5345_read(client, 0x07) & 0x3f;
+		if (ctrl->value >= 32)
+			ctrl->value = ctrl->value - 64;
+		break;
+
+	case VIDIOC_S_CTRL:
+		break;
+		if (ctrl->id == V4L2_CID_AUDIO_MUTE) {
+			cs5345_write(client, 0x04, ctrl->value ? 0x80 : 0);
+			break;
+		}
+		if (ctrl->id != V4L2_CID_AUDIO_VOLUME)
+			return -EINVAL;
+		if (ctrl->value > 24 || ctrl->value < -24)
+			return -EINVAL;
+		cs5345_write(client, 0x07, ((u8)ctrl->value) & 0x3f);
+		cs5345_write(client, 0x08, ((u8)ctrl->value) & 0x3f);
+		break;
+
+#ifdef CONFIG_VIDEO_ADV_DEBUG
+	case VIDIOC_DBG_G_REGISTER:
+	case VIDIOC_DBG_S_REGISTER:
+	{
+		struct v4l2_register *reg = arg;
+
+		if (!v4l2_chip_match_i2c_client(client,
+					reg->match_type, reg->match_chip))
+			return -EINVAL;
+		if (!capable(CAP_SYS_ADMIN))
+			return -EPERM;
+		if (cmd == VIDIOC_DBG_G_REGISTER)
+			reg->val = cs5345_read(client, reg->reg & 0x1f);
+		else
+			cs5345_write(client, reg->reg & 0x1f, reg->val & 0x1f);
+		break;
+	}
+#endif
+
+	case VIDIOC_G_CHIP_IDENT:
+		return v4l2_chip_ident_i2c_client(client,
+				arg, V4L2_IDENT_CS5345, 0);
+
+	case VIDIOC_LOG_STATUS:
+		{
+			u8 v = cs5345_read(client, 0x09) & 7;
+			u8 m = cs5345_read(client, 0x04);
+			int vol = cs5345_read(client, 0x08) & 0x3f;
+
+			v4l_info(client, "Input:  %d%s\n", v,
+				      (m & 0x80) ? " (muted)" : "");
+			if (vol >= 32)
+				vol = vol - 64;
+			v4l_info(client, "Volume: %d dB\n", vol);
+			break;
+		}
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/* ----------------------------------------------------------------------- */
+
+static int cs5345_probe(struct i2c_client *client)
+{
+	/* Check if the adapter supports the needed features */
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_BYTE_DATA))
+		return -EIO;
+
+	v4l_info(client, "chip found @ 0x%x (%s)\n",
+			client->addr << 1, client->adapter->name);
+
+	cs5345_write(client, 0x02, 0x00);
+	cs5345_write(client, 0x04, 0x01);
+	cs5345_write(client, 0x09, 0x01);
+	return 0;
+}
+
+/* ----------------------------------------------------------------------- */
+
+static struct v4l2_i2c_driver_data v4l2_i2c_data = {
+	.name = "cs5345",
+	.driverid = I2C_DRIVERID_CS5345,
+	.command = cs5345_command,
+	.probe = cs5345_probe,
+};
+
diff --git a/include/media/cs5345.h b/include/media/cs5345.h
new file mode 100644
index 000000000000..6ccae24e65ed
--- /dev/null
+++ b/include/media/cs5345.h
@@ -0,0 +1,39 @@
+/*
+    cs5345.h - definition for cs5345 inputs and outputs
+
+    Copyright (C) 2007 Hans Verkuil (hverkuil@xs4all.nl)
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#ifndef _CS5345_H_
+#define _CS5345_H_
+
+/* CS5345 HW inputs */
+#define CS5345_IN_MIC 0
+#define CS5345_IN_1   1
+#define CS5345_IN_2   2
+#define CS5345_IN_3   3
+#define CS5345_IN_4   4
+#define CS5345_IN_5   5
+#define CS5345_IN_6   6
+
+#define CS5345_MCLK_1   0x00
+#define CS5345_MCLK_1_5 0x10
+#define CS5345_MCLK_2   0x20
+#define CS5345_MCLK_3   0x30
+#define CS5345_MCLK_4   0x40
+
+#endif
diff --git a/include/media/v4l2-chip-ident.h b/include/media/v4l2-chip-ident.h
index 2442c281e4fe..032bb75f69c2 100644
--- a/include/media/v4l2-chip-ident.h
+++ b/include/media/v4l2-chip-ident.h
@@ -68,6 +68,9 @@ enum {
 	/* module vp27smpx: just ident 2700 */
 	V4L2_IDENT_VP27SMPX = 2700,
 
+	/* module cs5345: just ident 5345 */
+	V4L2_IDENT_CS5345 = 5345,
+
 	/* module wm8739: just ident 8739 */
 	V4L2_IDENT_WM8739 = 8739,
 
-- 
cgit v1.2.3


From 8574ac690eab730a3ff735177af1d52e45e36c8b Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@linuxtv.org>
Date: Mon, 24 Dec 2007 15:49:33 -0300
Subject: V4L/DVB (6921): include/media/v4l2-i2c-drv.h must #include
 <media/v4l2-common.h>

Fix the following compiler error:

v4l2-i2c-drv.h:72: error: implicit declaration of function 'v4l2_i2c_attach'

Also, prevent multiple inclusions of v4l2-i2c-drv.h

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/v4l2-i2c-drv.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/media/v4l2-i2c-drv.h b/include/media/v4l2-i2c-drv.h
index d758b765e0ff..9e4bab276915 100644
--- a/include/media/v4l2-i2c-drv.h
+++ b/include/media/v4l2-i2c-drv.h
@@ -21,6 +21,11 @@
  * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
+#ifndef __V4L2_I2C_DRV_H__
+#define __V4L2_I2C_DRV_H__
+
+#include <media/v4l2-common.h>
+
 struct v4l2_i2c_driver_data {
 	const char * const name;
 	int driverid;
@@ -59,3 +64,5 @@ static void __exit v4l2_i2c_drv_cleanup(void)
 
 module_init(v4l2_i2c_drv_init);
 module_exit(v4l2_i2c_drv_cleanup);
+
+#endif /* __V4L2_I2C_DRV_H__ */
-- 
cgit v1.2.3


From e8018c9e78b0ff4bb0290e46f4045fb4ea589ae8 Mon Sep 17 00:00:00 2001
From: "Andrey J. Melnikoff (TEMHOTA)" <temnota@kmv.ru>
Date: Mon, 7 Jan 2008 05:17:39 -0300
Subject: V4L/DVB (6973): Add Beholder TV 401/405/407/409/505/507/609/M6
 support

This patch updates cardlist for Beholder TV tuners:
old models (with GPIO ir) 401, 403, 405, 407, 409, 505, 507
and add support for 607, 609, M6 cards with new i2c-ir.

Signed-off-by: Igor Kuznetsov <igk72@yandex.ru>
Signed-off-by: Andrey J. Melnikov <temnota@kmv.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.saa7134  |  13 +
 drivers/media/common/ir-keymaps.c           | 104 ++++++
 drivers/media/video/ir-kbd-i2c.c            |   3 +-
 drivers/media/video/saa7134/saa7134-cards.c | 504 ++++++++++++++++++++++++++++
 drivers/media/video/saa7134/saa7134-i2c.c   |   2 +
 drivers/media/video/saa7134/saa7134-input.c |  60 ++++
 drivers/media/video/saa7134/saa7134.h       |  13 +
 include/media/ir-common.h                   |   1 +
 8 files changed, 699 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134
index 29ca16cdb713..387c35f2ecd7 100644
--- a/Documentation/video4linux/CARDLIST.saa7134
+++ b/Documentation/video4linux/CARDLIST.saa7134
@@ -116,3 +116,16 @@
 115 -> Sabrent PCMCIA TV-PCB05                  [0919:2003]
 116 -> 10MOONS TM300 TV Card                    [1131:2304]
 117 -> Avermedia Super 007                      [1461:f01d]
+118 -> Beholder BeholdTV 401                    [1131:0000,0000:4016]
+119 -> Beholder BeholdTV 403                    [0000:4036]
+120 -> Beholder BeholdTV 403 FM                 [1131:0000,0000:4037]
+121 -> Beholder BeholdTV 405                    [0000:4050]
+122 -> Beholder BeholdTV 405 FM                 [0000:4051]
+123 ->                                          [0000:4070]
+124 -> Beholder BeholdTV 407 FM                 [0000:4071]
+125 -> Beholder BeholdTV 409                    [0000:4090]
+126 -> Beholder BeholdTV 505 FM/RDS             [0000:5051,0000:505B,5ace:5050]
+127 -> Beholder BeholdTV 507 FM/RDS / BeholdTV 509 FM [0000:5071,0000:507B,5ace:5070,5ace:5090]
+128 -> Beholder BeholdTV Columbus TVFM          [0000:5201]
+129 -> Beholder BeholdTV 607 / BeholdTV 609     [5ace:6070,5ace:6071,5ace:6072,5ace:6073,5ace:6090,5ace:6091,5ace:6092,5ace:6093]
+130 -> Beholder BeholdTV M6 / BeholdTV M6 Extra [5ace:6190,5ace:6193]
diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c
index 42762dfb7387..ee2c69e3152a 100644
--- a/drivers/media/common/ir-keymaps.c
+++ b/drivers/media/common/ir-keymaps.c
@@ -1848,3 +1848,107 @@ IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE] = {
 };
 
 EXPORT_SYMBOL_GPL(ir_codes_fusionhdtv_mce);
+
+/*
+ * Igor Kuznetsov <igk72@ya.ru>
+ * Andrey J. Melnikov <temnota@kmv.ru>
+ *
+ * Keytable is used by BeholdTV 60x series, M6 series at
+ * least, and probably other cards too.
+ * The "ascii-art picture" below (in comments, first row
+ * is the keycode in hex, and subsequent row(s) shows
+ * the button labels (several variants when appropriate)
+ * helps to descide which keycodes to assign to the buttons.
+ */
+IR_KEYTAB_TYPE ir_codes_behold[IR_KEYTAB_SIZE] = {
+
+	/*  0x1c            0x12  *
+	 *  TV/FM          POWER  *
+	 *                        */
+	[ 0x1c ] = KEY_TUNER,	/*XXX KEY_TV KEY_RADIO */
+	[ 0x12 ] = KEY_POWER,
+
+	/*  0x01    0x02    0x03  *
+	 *   1       2       3    *
+	 *                        *
+	 *  0x04    0x05    0x06  *
+	 *   4       5       6    *
+	 *                        *
+	 *  0x07    0x08    0x09  *
+	 *   7       8       9    *
+	 *                        */
+	[ 0x01 ] = KEY_1,
+	[ 0x02 ] = KEY_2,
+	[ 0x03 ] = KEY_3,
+	[ 0x04 ] = KEY_4,
+	[ 0x05 ] = KEY_5,
+	[ 0x06 ] = KEY_6,
+	[ 0x07 ] = KEY_7,
+	[ 0x08 ] = KEY_8,
+	[ 0x09 ] = KEY_9,
+
+	/*  0x0a    0x00    0x17  *
+	 * RECALL    0      MODE  *
+	 *                        */
+	[ 0x0a ] = KEY_AGAIN,
+	[ 0x00 ] = KEY_0,
+	[ 0x17 ] = KEY_MODE,
+
+	/*  0x14          0x10    *
+	 * ASPECT      FULLSCREEN *
+	 *                        */
+	[ 0x14 ] = KEY_SCREEN,
+	[ 0x10 ] = KEY_ZOOM,
+
+	/*          0x0b          *
+	 *           Up           *
+	 *                        *
+	 *  0x18    0x16    0x0c  *
+	 *  Left     Ok     Right *
+	 *                        *
+	 *         0x015          *
+	 *         Down           *
+	 *                        */
+	[ 0x0b ] = KEY_CHANNELUP,	/*XXX KEY_UP */
+	[ 0x18 ] = KEY_VOLUMEDOWN,	/*XXX KEY_LEFT */
+	[ 0x16 ] = KEY_OK,		/*XXX KEY_ENTER */
+	[ 0x0c ] = KEY_VOLUMEUP,	/*XXX KEY_RIGHT */
+	[ 0x15 ] = KEY_CHANNELDOWN,	/*XXX KEY_DOWN */
+
+	/*  0x11            0x0d  *
+	 *  MUTE            INFO  *
+	 *                        */
+	[ 0x11 ] = KEY_MUTE,
+	[ 0x0d ] = KEY_INFO,
+
+	/*  0x0f    0x1b    0x1a  *
+	 * RECORD PLAY/PAUSE STOP *
+	 *                        *
+	 *  0x0e    0x1f    0x1e  *
+	 *TELETEXT  AUDIO  SOURCE *
+	 *           RED   YELLOW *
+	 *                        */
+	[ 0x0f ] = KEY_RECORD,
+	[ 0x1b ] = KEY_PLAYPAUSE,
+	[ 0x1a ] = KEY_STOP,
+	[ 0x0e ] = KEY_TEXT,
+	[ 0x1f ] = KEY_RED,	/*XXX KEY_AUDIO */
+	[ 0x1e ] = KEY_YELLOW,	/*XXX KEY_SOURCE */
+
+	/*  0x1d   0x13     0x19  *
+	 * SLEEP  PREVIEW   DVB   *
+	 *         GREEN    BLUE  *
+	 *                        */
+	[ 0x1d ] = KEY_SLEEP,
+	[ 0x13 ] = KEY_GREEN,
+	[ 0x19 ] = KEY_BLUE,	/*XXX KEY_SAT */
+
+	/*  0x58           0x5c   *
+	 * FREEZE        SNAPSHOT *
+	 *                        */
+	[ 0x58 ] = KEY_SLOW,
+	[ 0x5c ] = KEY_SAVE,
+
+};
+
+EXPORT_SYMBOL_GPL(ir_codes_behold);
diff --git a/drivers/media/video/ir-kbd-i2c.c b/drivers/media/video/ir-kbd-i2c.c
index 29779d8bf7fb..9851987b95fb 100644
--- a/drivers/media/video/ir-kbd-i2c.c
+++ b/drivers/media/video/ir-kbd-i2c.c
@@ -398,6 +398,7 @@ static int ir_attach(struct i2c_adapter *adap, int addr,
 	case 0x7a:
 	case 0x47:
 	case 0x71:
+	case 0x2d:
 		if (adap->id == I2C_HW_B_CX2388x) {
 			/* Handled by cx88-input */
 			name        = "CX2388x remote";
@@ -504,7 +505,7 @@ static int ir_probe(struct i2c_adapter *adap)
 	*/
 
 	static const int probe_bttv[] = { 0x1a, 0x18, 0x4b, 0x64, 0x30, -1};
-	static const int probe_saa7134[] = { 0x7a, 0x47, 0x71, -1 };
+	static const int probe_saa7134[] = { 0x7a, 0x47, 0x71, 0x2d, -1 };
 	static const int probe_em28XX[] = { 0x30, 0x47, -1 };
 	static const int probe_cx88[] = { 0x18, 0x6b, 0x71, -1 };
 	static const int probe_cx23885[] = { 0x6b, -1 };
diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c
index b29427aaa88d..6a35d88fdbb2 100644
--- a/drivers/media/video/saa7134/saa7134-cards.c
+++ b/drivers/media/video/saa7134/saa7134-cards.c
@@ -3562,6 +3562,329 @@ struct saa7134_board saa7134_boards[] = {
 			.tv     = 1,
 		}},
 	},
+	[SAA7134_BOARD_BEHOLD_401] = {
+		.name           = "Beholder BeholdTV 401",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FQ1216ME,
+		.radio_type     = UNSET,
+		.tuner_addr	= ADDR_UNSET,
+		.radio_addr	= ADDR_UNSET,
+		.inputs         = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		}},
+		.mute = {
+			.name = name_mute,
+			.amux = LINE1,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_403] = {
+		.name           = "Beholder BeholdTV 403",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FQ1216ME,
+		.radio_type     = UNSET,
+		.tuner_addr	= ADDR_UNSET,
+		.radio_addr	= ADDR_UNSET,
+		.inputs         = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		},{
+			.name   = name_comp1,
+			.vmux   = 1,
+			.amux   = LINE1,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		}},
+	},
+	[SAA7134_BOARD_BEHOLD_403FM] = {
+		.name           = "Beholder BeholdTV 403 FM",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FQ1216ME,
+		.radio_type     = UNSET,
+		.tuner_addr	= ADDR_UNSET,
+		.radio_addr	= ADDR_UNSET,
+		.inputs         = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		},{
+			.name   = name_comp1,
+			.vmux   = 1,
+			.amux   = LINE1,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_405] = {
+		.name           = "Beholder BeholdTV 405",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		},{
+			.name = name_comp1,
+			.vmux = 3,
+			.amux = LINE1,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		}},
+	},
+	[SAA7134_BOARD_BEHOLD_405FM] = {
+		/* Sergey <skiv@orel.ru> */
+		.name           = "Beholder BeholdTV 405 FM",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		},{
+			.name = name_comp1,
+			.vmux = 3,
+			.amux = LINE1,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_407FM] = {
+		.name 		= "Beholder BeholdTV 407 FM",
+		.audio_clock 	= 0x00187de7,
+		.tuner_type 	= TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type 	= UNSET,
+		.tuner_addr 	= ADDR_UNSET,
+		.radio_addr 	= ADDR_UNSET,
+		.tda9887_conf 	= TDA9887_PRESENT,
+		.gpiomask = 0xc0c000,
+		.inputs = {{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+			.gpio = 0xc0c000,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+			.gpio = 0xc0c000,
+		},{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv = 1,
+			.gpio = 0xc0c000,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+			.gpio = 0xc0c000,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_409] = {
+		.name           = "Beholder BeholdTV 409",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+	},
+	[SAA7134_BOARD_BEHOLD_505FM] = {
+		.name           = "Beholder BeholdTV 505 FM/RDS",
+		.audio_clock    = 0x00200000,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = LINE2,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+		.mute = {
+			.name = name_mute,
+			.amux = LINE1,
+		},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_507_9FM] = {
+		.name           = "Beholder BeholdTV 507 FM/RDS / BeholdTV 509 FM",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+			.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM] = {
+		.name           = "Beholder BeholdTV Columbus TVFM",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_ALPS_TSBE5_PAL,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_607_9FM] = {
+		/* Andrey Melnikoff <temnota@kmv.ru> */
+		.name           = "Beholder BeholdTV 607 / BeholdTV 609",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+	},
+	[SAA7134_BOARD_BEHOLD_M6] = {
+		/* Igor Kuznetsov <igk@igk.ru> */
+		/* Andrey Melnikoff <temnota@kmv.ru> */
+		.name           = "Beholder BeholdTV M6 / BeholdTV M6 Extra",
+		.audio_clock    = 0x00187de7,
+		.tuner_type     = TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 3,
+			.amux = TV,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 1,
+			.amux = LINE1,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE1,
+		}},
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+		},
+		.mpeg  = SAA7134_MPEG_EMPRESS,
+	},
 };
 
 const unsigned int saa7134_bcount = ARRAY_SIZE(saa7134_boards);
@@ -4326,6 +4649,174 @@ struct pci_device_id saa7134_pci_tbl[] = {
 		.subvendor    = 0x1461, /* Avermedia Technologies Inc */
 		.subdevice    = 0xf01d, /* AVerTV DVB-T Super 007 */
 		.driver_data  = SAA7134_BOARD_AVERMEDIA_SUPER_007,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x1131,
+		.subdevice    = 0x0000,
+		.driver_data  = SAA7134_BOARD_BEHOLD_401,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4016,
+		.driver_data  = SAA7134_BOARD_BEHOLD_401,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x1131,
+		.subdevice    = 0x0000,
+		.driver_data  = SAA7134_BOARD_BEHOLD_403FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4036,
+		.driver_data  = SAA7134_BOARD_BEHOLD_403,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4037,
+		.driver_data  = SAA7134_BOARD_BEHOLD_403FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4050,
+		.driver_data  = SAA7134_BOARD_BEHOLD_405,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4051,
+		.driver_data  = SAA7134_BOARD_BEHOLD_405FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4070,
+		.driver_data  = SAA7134_BOARD_BEHOLD_407,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4071,
+		.driver_data  = SAA7134_BOARD_BEHOLD_407FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x4090,
+		.driver_data  = SAA7134_BOARD_BEHOLD_409,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x5051,
+		.driver_data  = SAA7134_BOARD_BEHOLD_505FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x505B,
+		.driver_data  = SAA7134_BOARD_BEHOLD_505FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x5050,
+		.driver_data  = SAA7134_BOARD_BEHOLD_505FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x5071,
+		.driver_data  = SAA7134_BOARD_BEHOLD_507_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x507B,
+		.driver_data  = SAA7134_BOARD_BEHOLD_507_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x5070,
+		.driver_data  = SAA7134_BOARD_BEHOLD_507_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x5090,
+		.driver_data  = SAA7134_BOARD_BEHOLD_507_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x0000,
+		.subdevice    = 0x5201,
+		.driver_data  = SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6070,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6071,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6072,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6073,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6090,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6091,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6092,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6093,
+		.driver_data  = SAA7134_BOARD_BEHOLD_607_9FM,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6190,
+		.driver_data  = SAA7134_BOARD_BEHOLD_M6,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
+		.subvendor    = 0x5ace,
+		.subdevice    = 0x6193,
+		.driver_data  = SAA7134_BOARD_BEHOLD_M6,
 	},{
 		.vendor       = PCI_VENDOR_ID_PHILIPS,
 		.device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
@@ -4473,6 +4964,16 @@ int saa7134_board_init1(struct saa7134_dev *dev)
 	case SAA7134_BOARD_ENCORE_ENLTV:
 	case SAA7134_BOARD_ENCORE_ENLTV_FM:
 	case SAA7134_BOARD_10MOONSTVMASTER3:
+	case SAA7134_BOARD_BEHOLD_401:
+	case SAA7134_BOARD_BEHOLD_403:
+	case SAA7134_BOARD_BEHOLD_403FM:
+	case SAA7134_BOARD_BEHOLD_405:
+	case SAA7134_BOARD_BEHOLD_405FM:
+	case SAA7134_BOARD_BEHOLD_407:
+	case SAA7134_BOARD_BEHOLD_407FM:
+	case SAA7134_BOARD_BEHOLD_409:
+	case SAA7134_BOARD_BEHOLD_505FM:
+	case SAA7134_BOARD_BEHOLD_507_9FM:
 		dev->has_remote = SAA7134_REMOTE_GPIO;
 		break;
 	case SAA7134_BOARD_FLYDVBS_LR300:
@@ -4512,6 +5013,7 @@ int saa7134_board_init1(struct saa7134_dev *dev)
 		saa_andorl(SAA7134_GPIO_GPSTATUS0 >> 2, 0x08000000, 0x00000000);
 		break;
 	case SAA7134_BOARD_AVERMEDIA_CARDBUS:
+	case SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM:
 		/* power-up tuner chip */
 		saa_andorl(SAA7134_GPIO_GPMODE0 >> 2,   0xffffffff, 0xffffffff);
 		saa_andorl(SAA7134_GPIO_GPSTATUS0 >> 2, 0xffffffff, 0xffffffff);
@@ -4533,6 +5035,8 @@ int saa7134_board_init1(struct saa7134_dev *dev)
 	case SAA7134_BOARD_PINNACLE_PCTV_310i:
 	case SAA7134_BOARD_UPMOST_PURPLE_TV:
 	case SAA7134_BOARD_HAUPPAUGE_HVR1110:
+	case SAA7134_BOARD_BEHOLD_607_9FM:
+	case SAA7134_BOARD_BEHOLD_M6:
 		dev->has_remote = SAA7134_REMOTE_I2C;
 		break;
 	case SAA7134_BOARD_AVERMEDIA_A169_B:
diff --git a/drivers/media/video/saa7134/saa7134-i2c.c b/drivers/media/video/saa7134/saa7134-i2c.c
index 800b397f8f16..d3322c3018f2 100644
--- a/drivers/media/video/saa7134/saa7134-i2c.c
+++ b/drivers/media/video/saa7134/saa7134-i2c.c
@@ -334,6 +334,7 @@ static int attach_inform(struct i2c_client *client)
 		case 0x7a:
 		case 0x47:
 		case 0x71:
+		case 0x2d:
 		{
 			struct IR_i2c *ir = i2c_get_clientdata(client);
 			d1printk("%s i2c IR detected (%s).\n",
@@ -439,6 +440,7 @@ static char *i2c_devs[128] = {
 	[ 0xa0 >> 1 ] = "eeprom",
 	[ 0xc0 >> 1 ] = "tuner (analog)",
 	[ 0x86 >> 1 ] = "tda9887",
+	[ 0x5a >> 1 ] = "remote control",
 };
 
 static void do_i2c_scan(char *name, struct i2c_client *c)
diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c
index 8cfeb2bde90f..7f4212928e9f 100644
--- a/drivers/media/video/saa7134/saa7134-input.c
+++ b/drivers/media/video/saa7134/saa7134-input.c
@@ -52,6 +52,11 @@ module_param(repeat_period, int, 0644);
 MODULE_PARM_DESC(repeat_period, "repeat period between "
     "keypresses when key is down");
 
+static unsigned int disable_other_ir;
+module_param(disable_other_ir, int, 0644);
+MODULE_PARM_DESC(disable_other_ir, "disable full codes of "
+    "alternative remotes from other manufacturers");
+
 #define dprintk(fmt, arg...)	if (ir_debug) \
 	printk(KERN_DEBUG "%s/ir: " fmt, dev->name , ## arg)
 #define i2cdprintk(fmt, arg...)    if (ir_debug) \
@@ -154,6 +159,45 @@ static int get_key_hvr1110(struct IR_i2c *ir, u32 *ir_key, u32 *ir_raw)
 	return 1;
 }
 
+
+static int get_key_beholdm6xx(struct IR_i2c *ir, u32 *ir_key, u32 *ir_raw)
+{
+	unsigned char data[12];
+	u32 gpio;
+
+	struct saa7134_dev *dev = ir->c.adapter->algo_data;
+
+	/* rising SAA7134_GPIO_GPRESCAN reads the status */
+	saa_clearb(SAA7134_GPIO_GPMODE3, SAA7134_GPIO_GPRESCAN);
+	saa_setb(SAA7134_GPIO_GPMODE3, SAA7134_GPIO_GPRESCAN);
+
+	gpio = saa_readl(SAA7134_GPIO_GPSTATUS0 >> 2);
+
+	if (0x400000 &~ gpio)
+		return 0; /* No button press */
+
+	ir->c.addr = 0x5a >> 1;
+
+	if (12 != i2c_master_recv(&ir->c, data, 12)) {
+		i2cdprintk("read error\n");
+		return -EIO;
+	}
+	/* IR of this card normally decode signals NEC-standard from
+	 * - Sven IHOO MT 5.1R remote. xxyye718
+	 * - Sven DVD HD-10xx remote. xxyyf708
+	 * - BBK ...
+	 * - mayby others
+	 * So, skip not our, if disable full codes mode.
+	 */
+	if (data[10] != 0x6b && data[11] != 0x86 && disable_other_ir)
+		return 0;
+
+	*ir_key = data[9];
+	*ir_raw = data[9];
+
+	return 1;
+}
+
 void saa7134_input_irq(struct saa7134_dev *dev)
 {
 	struct card_ir *ir = dev->remote;
@@ -288,6 +332,16 @@ int saa7134_input_init1(struct saa7134_dev *dev)
 	case SAA7134_BOARD_MANLI_MTV001:
 	case SAA7134_BOARD_MANLI_MTV002:
 	case SAA7134_BOARD_BEHOLD_409FM:
+	case SAA7134_BOARD_BEHOLD_401:
+	case SAA7134_BOARD_BEHOLD_403:
+	case SAA7134_BOARD_BEHOLD_403FM:
+	case SAA7134_BOARD_BEHOLD_405:
+	case SAA7134_BOARD_BEHOLD_405FM:
+	case SAA7134_BOARD_BEHOLD_407:
+	case SAA7134_BOARD_BEHOLD_407FM:
+	case SAA7134_BOARD_BEHOLD_409:
+	case SAA7134_BOARD_BEHOLD_505FM:
+	case SAA7134_BOARD_BEHOLD_507_9FM:
 		ir_codes     = ir_codes_manli;
 		mask_keycode = 0x001f00;
 		mask_keyup   = 0x004000;
@@ -458,6 +512,12 @@ void saa7134_set_i2c_ir(struct saa7134_dev *dev, struct IR_i2c *ir)
 		ir->get_key   = get_key_hvr1110;
 		ir->ir_codes  = ir_codes_hauppauge_new;
 		break;
+	case SAA7134_BOARD_BEHOLD_607_9FM:
+	case SAA7134_BOARD_BEHOLD_M6:
+		snprintf(ir->c.name, sizeof(ir->c.name), "BeholdTV");
+		ir->get_key   = get_key_beholdm6xx;
+		ir->ir_codes  = ir_codes_behold;
+		break;
 	default:
 		dprintk("Shouldn't get here: Unknown board %x for I2C IR?\n",dev->board);
 		break;
diff --git a/drivers/media/video/saa7134/saa7134.h b/drivers/media/video/saa7134/saa7134.h
index 3e77aeb7974f..ce450304fb53 100644
--- a/drivers/media/video/saa7134/saa7134.h
+++ b/drivers/media/video/saa7134/saa7134.h
@@ -240,6 +240,19 @@ struct saa7134_format {
 #define SAA7134_BOARD_SABRENT_TV_PCB05     115
 #define SAA7134_BOARD_10MOONSTVMASTER3     116
 #define SAA7134_BOARD_AVERMEDIA_SUPER_007  117
+#define SAA7134_BOARD_BEHOLD_401  	118
+#define SAA7134_BOARD_BEHOLD_403  	119
+#define SAA7134_BOARD_BEHOLD_403FM	120
+#define SAA7134_BOARD_BEHOLD_405	121
+#define SAA7134_BOARD_BEHOLD_405FM	122
+#define SAA7134_BOARD_BEHOLD_407	123
+#define SAA7134_BOARD_BEHOLD_407FM	124
+#define SAA7134_BOARD_BEHOLD_409	125
+#define SAA7134_BOARD_BEHOLD_505FM	126
+#define SAA7134_BOARD_BEHOLD_507_9FM	127
+#define SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM 128
+#define SAA7134_BOARD_BEHOLD_607_9FM	129
+#define SAA7134_BOARD_BEHOLD_M6		130
 
 #define SAA7134_MAXBOARDS 8
 #define SAA7134_INPUT_MAX 8
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index b904d3257260..9c2fc09a7133 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -140,6 +140,7 @@ extern IR_KEYTAB_TYPE ir_codes_asus_pc39[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_encore_enltv[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_tt_1500[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_behold[IR_KEYTAB_SIZE];
 
 #endif
 
-- 
cgit v1.2.3


From d05051c82e0e8ff748e9c9a06a061bda3ad656e5 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Date: Thu, 10 Jan 2008 07:33:03 -0300
Subject: V4L/DVB (6997): Replace a very dirty hack on videobuf for a clean
 wait_event

In order to videobuf_iolock to work, mmap_mapper should be called first.
Otherwise, an OOPS is generated.

On some cases, .mmap file handler used to took some time to be called. On those
situations, mmap_mmapper() were called after iolock.

This patch properly waits for mmap_mapper to be called, otherwise generating an
error.

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/videobuf-core.c | 15 +++++++++++----
 include/media/videobuf-core.h       |  2 ++
 2 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/videobuf-core.c b/drivers/media/video/videobuf-core.c
index 43fe3f78c8df..c3adbd686ffe 100644
--- a/drivers/media/video/videobuf-core.c
+++ b/drivers/media/video/videobuf-core.c
@@ -98,13 +98,15 @@ int videobuf_iolock(struct videobuf_queue *q, struct videobuf_buffer *vb,
 	MAGIC_CHECK(vb->magic, MAGIC_BUFFER);
 	MAGIC_CHECK(q->int_ops->magic, MAGIC_QTYPE_OPS);
 
-	/* FIXME: This is required to avoid OOPS on some cases,
+	/* This is required to avoid OOPS on some cases,
 	   since mmap_mapper() method should be called before _iolock.
 	   On some cases, the mmap_mapper() is called only after scheduling.
-
-	   However, this way is just too dirty! Better to wait for some event.
 	 */
-	schedule_timeout(HZ);
+	wait_event_timeout(vb->done, q->is_mmapped, msecs_to_jiffies(100));
+	if (!q->is_mmapped) {
+		printk(KERN_ERR "Error: mmap_mapper() never called!\n");
+		return -EINVAL;
+	}
 
 	return CALL(q, iolock, q, vb, fbuf);
 }
@@ -300,7 +302,11 @@ static int __videobuf_mmap_free(struct videobuf_queue *q)
 
 	MAGIC_CHECK(q->int_ops->magic, MAGIC_QTYPE_OPS);
 
+
 	rc  = CALL(q, mmap_free, q);
+
+	q->is_mmapped = 0;
+
 	if (rc < 0)
 		return rc;
 
@@ -1022,6 +1028,7 @@ int videobuf_mmap_mapper(struct videobuf_queue *q,
 
 	mutex_lock(&q->lock);
 	retval = CALL(q, mmap_mapper, q, vma);
+	q->is_mmapped = 1;
 	mutex_unlock(&q->lock);
 
 	return retval;
diff --git a/include/media/videobuf-core.h b/include/media/videobuf-core.h
index 7aa7a7b64c1b..97f14d469595 100644
--- a/include/media/videobuf-core.h
+++ b/include/media/videobuf-core.h
@@ -164,6 +164,8 @@ struct videobuf_queue {
 
 	unsigned int               streaming:1;
 	unsigned int               reading:1;
+	unsigned int		   is_mmapped:1;
+
 	/* capture via mmap() + ioctl(QBUF/DQBUF) */
 	struct list_head           stream;
 
-- 
cgit v1.2.3


From f234081bc564c69eb0e2cd4e957ad1cbae4a6144 Mon Sep 17 00:00:00 2001
From: Steven Toth <stoth@hauppauge.com>
Date: Thu, 10 Jan 2008 01:22:39 -0300
Subject: V4L/DVB (7002): cx25840: Add basic CX23885 AVCore support

The cx23885/7/8 PCIe bridge has an internal AVCore modelled on
the cx2584x family. Many of the registers positions are identical
but some moved. The register values are also different because
the different bridges run at different clock rates.

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/cx25840/cx25840-audio.c    |  65 ++++++--
 drivers/media/video/cx25840/cx25840-core.c     | 219 ++++++++++++++++++++++---
 drivers/media/video/cx25840/cx25840-core.h     |   1 +
 drivers/media/video/cx25840/cx25840-firmware.c |   5 +
 include/media/cx25840.h                        |  19 +++
 5 files changed, 272 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/cx25840/cx25840-audio.c b/drivers/media/video/cx25840/cx25840-audio.c
index 51fc0af01578..d6421e1e8f6a 100644
--- a/drivers/media/video/cx25840/cx25840-audio.c
+++ b/drivers/media/video/cx25840/cx25840-audio.c
@@ -32,11 +32,17 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 
 	/* common for all inputs and rates */
 	/* SA_MCLK_SEL=1, SA_MCLK_DIV=0x10 */
-	cx25840_write(client, 0x127, 0x50);
+	if (!state->is_cx23885)
+		cx25840_write(client, 0x127, 0x50);
 
 	if (state->aud_input != CX25840_AUDIO_SERIAL) {
 		switch (freq) {
 		case 32000:
+			if (state->is_cx23885) {
+				/* We don't have register values
+				 * so avoid destroying registers. */
+				break;
+			}
 			/* VID_PLL and AUX_PLL */
 			cx25840_write4(client, 0x108, 0x1006040f);
 
@@ -53,6 +59,11 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 			break;
 
 		case 44100:
+			if (state->is_cx23885) {
+				/* We don't have register values
+				 * so avoid destroying registers. */
+				break;
+			}
 			/* VID_PLL and AUX_PLL */
 			cx25840_write4(client, 0x108, 0x1009040f);
 
@@ -69,6 +80,11 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 			break;
 
 		case 48000:
+			if (state->is_cx23885) {
+				/* We don't have register values
+				 * so avoid destroying registers. */
+				break;
+			}
 			/* VID_PLL and AUX_PLL */
 			cx25840_write4(client, 0x108, 0x100a040f);
 
@@ -87,6 +103,11 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 	} else {
 		switch (freq) {
 		case 32000:
+			if (state->is_cx23885) {
+				/* We don't have register values
+				 * so avoid destroying registers. */
+				break;
+			}
 			/* VID_PLL and AUX_PLL */
 			cx25840_write4(client, 0x108, 0x1e08040f);
 
@@ -109,6 +130,12 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 			break;
 
 		case 44100:
+			if (state->is_cx23885) {
+				/* We don't have register values
+				 * so avoid destroying registers. */
+				break;
+			}
+
 			/* VID_PLL and AUX_PLL */
 			cx25840_write4(client, 0x108, 0x1809040f);
 
@@ -128,22 +155,33 @@ static int set_audclk_freq(struct i2c_client *client, u32 freq)
 			break;
 
 		case 48000:
-			/* VID_PLL and AUX_PLL */
-			cx25840_write4(client, 0x108, 0x180a040f);
+			if (!state->is_cx23885) {
+				/* VID_PLL and AUX_PLL */
+				cx25840_write4(client, 0x108, 0x180a040f);
 
-			/* AUX_PLL_FRAC */
-			cx25840_write4(client, 0x110, 0x0098d6e5);
+				/* AUX_PLL_FRAC */
+				cx25840_write4(client, 0x110, 0x0098d6e5);
+			}
 
 			if (state->is_cx25836)
 				break;
 
-			/* src1_ctl = 0x08010000 */
-			cx25840_write4(client, 0x8f8, 0x08018000);
+			if (!state->is_cx23885) {
+				/* src1_ctl */
+				cx25840_write4(client, 0x8f8, 0x08018000);
 
-			/* src3/4/6_ctl = 0x08020000 */
-			cx25840_write4(client, 0x900, 0x08015555);
-			cx25840_write4(client, 0x904, 0x08015555);
-			cx25840_write4(client, 0x90c, 0x08015555);
+				/* src3/4/6_ctl */
+				cx25840_write4(client, 0x900, 0x08015555);
+				cx25840_write4(client, 0x904, 0x08015555);
+				cx25840_write4(client, 0x90c, 0x08015555);
+			} else {
+
+				cx25840_write4(client, 0x8f8, 0x0801867c);
+
+				cx25840_write4(client, 0x900, 0x08014faa);
+				cx25840_write4(client, 0x904, 0x08014faa);
+				cx25840_write4(client, 0x90c, 0x08014faa);
+			}
 			break;
 		}
 	}
@@ -188,6 +226,11 @@ void cx25840_audio_set_path(struct i2c_client *client)
 
 	/* deassert soft reset */
 	cx25840_and_or(client, 0x810, ~0x1, 0x00);
+
+	if (state->is_cx23885) {
+		/* Ensure the controller is running when we exit */
+		cx25840_and_or(client, 0x803, ~0x10, 0x10);
+	}
 }
 
 static int get_volume(struct i2c_client *client)
diff --git a/drivers/media/video/cx25840/cx25840-core.c b/drivers/media/video/cx25840/cx25840-core.c
index 0d3d24aff504..756a1eeb274e 100644
--- a/drivers/media/video/cx25840/cx25840-core.c
+++ b/drivers/media/video/cx25840/cx25840-core.c
@@ -13,6 +13,8 @@
  * NTSC sliced VBI support by Christopher Neufeld <television@cneufeld.ca>
  * with additional fixes by Hans Verkuil <hverkuil@xs4all.nl>.
  *
+ * CX23885 support by Steven Toth <stoth@hauppauge.com>.
+ *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
  * as published by the Free Software Foundation; either version 2
@@ -255,6 +257,96 @@ static void cx25840_initialize(struct i2c_client *client)
 	cx25840_and_or(client, 0x803, ~0x10, 0x10);
 }
 
+static void cx23885_initialize(struct i2c_client *client)
+{
+	DEFINE_WAIT(wait);
+	struct cx25840_state *state = i2c_get_clientdata(client);
+	struct workqueue_struct *q;
+
+	/* Internal Reset */
+	cx25840_and_or(client, 0x102, ~0x01, 0x01);
+	cx25840_and_or(client, 0x102, ~0x01, 0x00);
+
+	/* Stop microcontroller */
+	cx25840_and_or(client, 0x803, ~0x10, 0x00);
+
+	/* DIF in reset? */
+	cx25840_write(client, 0x398, 0);
+
+	/* Trust the default xtal, no division */
+	/* This changes for the cx23888 products */
+	cx25840_write(client, 0x2, 0x76);
+
+	/* Bring down the regulator for AUX clk */
+	cx25840_write(client, 0x1, 0x40);
+
+	/* Sys PLL frac */
+	cx25840_write4(client, 0x11c, 0x01d1744c);
+
+	/* Sys PLL int */
+	cx25840_write4(client, 0x118, 0x00000416);
+
+	/* Disable DIF bypass */
+	cx25840_write4(client, 0x33c, 0x00000001);
+
+	/* DIF Src phase inc */
+	cx25840_write4(client, 0x340, 0x0df7df83);
+
+	/* Vid PLL frac */
+	cx25840_write4(client, 0x10c, 0x01b6db7b);
+
+	/* Vid PLL int */
+	cx25840_write4(client, 0x108, 0x00000512);
+
+	/* Luma */
+	cx25840_write4(client, 0x414, 0x00107d12);
+
+	/* Chroma */
+	cx25840_write4(client, 0x420, 0x3d008282);
+
+	/* Aux PLL frac */
+	cx25840_write4(client, 0x114, 0x017dbf48);
+
+	/* Aux PLL int */
+	cx25840_write4(client, 0x110, 0x000a030e);
+
+	/* ADC2 input select */
+	cx25840_write(client, 0x102, 0x10);
+
+	/* VIN1 & VIN5 */
+	cx25840_write(client, 0x103, 0x11);
+
+	/* Enable format auto detect */
+	cx25840_write(client, 0x400, 0);
+	/* Fast subchroma lock */
+	/* White crush, Chroma AGC & Chroma Killer enabled */
+	cx25840_write(client, 0x401, 0xe8);
+
+	/* Select AFE clock pad output source */
+	cx25840_write(client, 0x144, 0x05);
+
+	/* Do the firmware load in a work handler to prevent.
+	   Otherwise the kernel is blocked waiting for the
+	   bit-banging i2c interface to finish uploading the
+	   firmware. */
+	INIT_WORK(&state->fw_work, cx25840_work_handler);
+	init_waitqueue_head(&state->fw_wait);
+	q = create_singlethread_workqueue("cx25840_fw");
+	prepare_to_wait(&state->fw_wait, &wait, TASK_UNINTERRUPTIBLE);
+	queue_work(q, &state->fw_work);
+	schedule();
+	finish_wait(&state->fw_wait, &wait);
+	destroy_workqueue(q);
+
+	cx25840_vbi_setup(client);
+
+	/* (re)set input */
+	set_input(client, state->vid_input, state->aud_input);
+
+	/* start microcontroller */
+	cx25840_and_or(client, 0x803, ~0x10, 0x10);
+}
+
 /* ----------------------------------------------------------------------- */
 
 static void input_change(struct i2c_client *client)
@@ -318,9 +410,22 @@ static int set_input(struct i2c_client *client, enum cx25840_video_input vid_inp
 			   vid_input <= CX25840_COMPOSITE8);
 	u8 reg;
 
-	v4l_dbg(1, cx25840_debug, client, "decoder set video input %d, audio input %d\n",
-			vid_input, aud_input);
+	v4l_dbg(1, cx25840_debug, client,
+		"decoder set video input %d, audio input %d\n",
+		vid_input, aud_input);
 
+	if (vid_input >= CX25840_VIN1_CH1) {
+		v4l_dbg(1, cx25840_debug, client, "vid_input 0x%x\n",
+			vid_input);
+		reg = vid_input & 0xff;
+		if ((vid_input & CX25840_SVIDEO_ON) == CX25840_SVIDEO_ON)
+			is_composite = 0;
+		else
+			is_composite = 1;
+
+		v4l_dbg(1, cx25840_debug, client, "mux cfg 0x%x comp=%d\n",
+			reg, is_composite);
+	} else
 	if (is_composite) {
 		reg = 0xf0 + (vid_input - CX25840_COMPOSITE1);
 	} else {
@@ -330,7 +435,8 @@ static int set_input(struct i2c_client *client, enum cx25840_video_input vid_inp
 		if ((vid_input & ~0xff0) ||
 		    luma < CX25840_SVIDEO_LUMA1 || luma > CX25840_SVIDEO_LUMA4 ||
 		    chroma < CX25840_SVIDEO_CHROMA4 || chroma > CX25840_SVIDEO_CHROMA8) {
-			v4l_err(client, "0x%04x is not a valid video input!\n", vid_input);
+			v4l_err(client, "0x%04x is not a valid video input!\n",
+				vid_input);
 			return -EINVAL;
 		}
 		reg = 0xf0 + ((luma - CX25840_SVIDEO_LUMA1) >> 4);
@@ -343,31 +449,49 @@ static int set_input(struct i2c_client *client, enum cx25840_video_input vid_inp
 		}
 	}
 
-	switch (aud_input) {
-	case CX25840_AUDIO_SERIAL:
-		/* do nothing, use serial audio input */
-		break;
-	case CX25840_AUDIO4: reg &= ~0x30; break;
-	case CX25840_AUDIO5: reg &= ~0x30; reg |= 0x10; break;
-	case CX25840_AUDIO6: reg &= ~0x30; reg |= 0x20; break;
-	case CX25840_AUDIO7: reg &= ~0xc0; break;
-	case CX25840_AUDIO8: reg &= ~0xc0; reg |= 0x40; break;
+	/* The caller has previously prepared the correct routing
+	 * configuration in reg (for the cx23885) so we have no
+	 * need to attempt to flip bits for earlier av decoders.
+	 */
+	if (!state->is_cx23885) {
+		switch (aud_input) {
+		case CX25840_AUDIO_SERIAL:
+			/* do nothing, use serial audio input */
+			break;
+		case CX25840_AUDIO4: reg &= ~0x30; break;
+		case CX25840_AUDIO5: reg &= ~0x30; reg |= 0x10; break;
+		case CX25840_AUDIO6: reg &= ~0x30; reg |= 0x20; break;
+		case CX25840_AUDIO7: reg &= ~0xc0; break;
+		case CX25840_AUDIO8: reg &= ~0xc0; reg |= 0x40; break;
 
-	default:
-		v4l_err(client, "0x%04x is not a valid audio input!\n", aud_input);
-		return -EINVAL;
+		default:
+			v4l_err(client, "0x%04x is not a valid audio input!\n",
+				aud_input);
+			return -EINVAL;
+		}
 	}
 
 	cx25840_write(client, 0x103, reg);
+
 	/* Set INPUT_MODE to Composite (0) or S-Video (1) */
 	cx25840_and_or(client, 0x401, ~0x6, is_composite ? 0 : 0x02);
-	/* Set CH_SEL_ADC2 to 1 if input comes from CH3 */
-	cx25840_and_or(client, 0x102, ~0x2, (reg & 0x80) == 0 ? 2 : 0);
-	/* Set DUAL_MODE_ADC2 to 1 if input comes from both CH2 and CH3 */
-	if ((reg & 0xc0) != 0xc0 && (reg & 0x30) != 0x30)
-		cx25840_and_or(client, 0x102, ~0x4, 4);
-	else
-		cx25840_and_or(client, 0x102, ~0x4, 0);
+
+	if (!state->is_cx23885) {
+		/* Set CH_SEL_ADC2 to 1 if input comes from CH3 */
+		cx25840_and_or(client, 0x102, ~0x2, (reg & 0x80) == 0 ? 2 : 0);
+		/* Set DUAL_MODE_ADC2 to 1 if input comes from both CH2&CH3 */
+		if ((reg & 0xc0) != 0xc0 && (reg & 0x30) != 0x30)
+			cx25840_and_or(client, 0x102, ~0x4, 4);
+		else
+			cx25840_and_or(client, 0x102, ~0x4, 0);
+	} else {
+		if (is_composite)
+			/* ADC2 input select channel 2 */
+			cx25840_and_or(client, 0x102, ~0x2, 0);
+		else
+			/* ADC2 input select channel 3 */
+			cx25840_and_or(client, 0x102, ~0x2, 2);
+	}
 
 	state->vid_input = vid_input;
 	state->aud_input = aud_input;
@@ -375,6 +499,25 @@ static int set_input(struct i2c_client *client, enum cx25840_video_input vid_inp
 		cx25840_audio_set_path(client);
 		input_change(client);
 	}
+
+	if (state->is_cx23885) {
+		/* Audio channel 1 src : Parallel 1 */
+		cx25840_write(client, 0x124, 0x03);
+
+		/* Select AFE clock pad output source */
+		cx25840_write(client, 0x144, 0x05);
+
+		/* I2S_IN_CTL: I2S_IN_SONY_MODE, LEFT SAMPLE on WS=1 */
+		cx25840_write(client, 0x914, 0xa0);
+
+		/* I2S_OUT_CTL:
+		 * I2S_IN_SONY_MODE, LEFT SAMPLE on WS=1
+		 * I2S_OUT_MASTER_MODE = Master
+		 */
+		cx25840_write(client, 0x918, 0xa0);
+		cx25840_write(client, 0x919, 0x01);
+	}
+
 	return 0;
 }
 
@@ -853,6 +996,8 @@ static int cx25840_command(struct i2c_client *client, unsigned int cmd,
 		state->is_initialized = 1;
 		if (state->is_cx25836)
 			cx25836_initialize(client);
+		else if (state->is_cx23885)
+			cx23885_initialize(client);
 		else
 			cx25840_initialize(client);
 	}
@@ -870,6 +1015,7 @@ static int cx25840_command(struct i2c_client *client, unsigned int cmd,
 			return -EINVAL;
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
+
 		if (cmd == VIDIOC_DBG_G_REGISTER)
 			reg->val = cx25840_read(client, reg->reg & 0x0fff);
 		else
@@ -886,14 +1032,26 @@ static int cx25840_command(struct i2c_client *client, unsigned int cmd,
 
 	case VIDIOC_STREAMON:
 		v4l_dbg(1, cx25840_debug, client, "enable output\n");
-		cx25840_write(client, 0x115, state->is_cx25836 ? 0x0c : 0x8c);
-		cx25840_write(client, 0x116, state->is_cx25836 ? 0x04 : 0x07);
+		if (state->is_cx23885) {
+			u8 v = (cx25840_read(client, 0x421) | 0x0b);
+			cx25840_write(client, 0x421, v);
+		} else {
+			cx25840_write(client, 0x115,
+				state->is_cx25836 ? 0x0c : 0x8c);
+			cx25840_write(client, 0x116,
+				state->is_cx25836 ? 0x04 : 0x07);
+		}
 		break;
 
 	case VIDIOC_STREAMOFF:
 		v4l_dbg(1, cx25840_debug, client, "disable output\n");
-		cx25840_write(client, 0x115, 0x00);
-		cx25840_write(client, 0x116, 0x00);
+		if (state->is_cx23885) {
+			u8 v = cx25840_read(client, 0x421) & ~(0x0b);
+			cx25840_write(client, 0x421, v);
+		} else {
+			cx25840_write(client, 0x115, 0x00);
+			cx25840_write(client, 0x116, 0x00);
+		}
 		break;
 
 	case VIDIOC_LOG_STATUS:
@@ -1056,6 +1214,8 @@ static int cx25840_command(struct i2c_client *client, unsigned int cmd,
 	case VIDIOC_INT_RESET:
 		if (state->is_cx25836)
 			cx25836_initialize(client);
+		else if (state->is_cx23885)
+			cx23885_initialize(client);
 		else
 			cx25840_initialize(client);
 		break;
@@ -1086,6 +1246,7 @@ static int cx25840_probe(struct i2c_client *client)
 
 	device_id = cx25840_read(client, 0x101) << 8;
 	device_id |= cx25840_read(client, 0x100);
+	v4l_dbg(1, cx25840_debug, client, "device_id = 0x%04x\n", device_id);
 
 	/* The high byte of the device ID should be
 	 * 0x83 for the cx2583x and 0x84 for the cx2584x */
@@ -1094,6 +1255,10 @@ static int cx25840_probe(struct i2c_client *client)
 	}
 	else if ((device_id & 0xff00) == 0x8400) {
 		id = V4L2_IDENT_CX25840 + ((device_id >> 4) & 0xf);
+	} else if (device_id == 0x0000) {
+		id = V4L2_IDENT_CX25836 + ((device_id >> 4) & 0xf) - 6;
+	} else if (device_id == 0x1313) {
+		id = V4L2_IDENT_CX25836 + ((device_id >> 4) & 0xf) - 6;
 	}
 	else {
 		v4l_dbg(1, cx25840_debug, client, "cx25840 not found\n");
@@ -1115,6 +1280,7 @@ static int cx25840_probe(struct i2c_client *client)
 	i2c_set_clientdata(client, state);
 	state->c = client;
 	state->is_cx25836 = ((device_id & 0xff00) == 0x8300);
+	state->is_cx23885 = (device_id == 0x0000) || (device_id == 0x1313);
 	state->vid_input = CX25840_COMPOSITE7;
 	state->aud_input = CX25840_AUDIO8;
 	state->audclk_freq = 48000;
@@ -1124,6 +1290,7 @@ static int cx25840_probe(struct i2c_client *client)
 	state->vbi_line_offset = 8;
 	state->id = id;
 	state->rev = device_id;
+
 	return 0;
 }
 
diff --git a/drivers/media/video/cx25840/cx25840-core.h b/drivers/media/video/cx25840/cx25840-core.h
index ea669b1f084d..95093edc9186 100644
--- a/drivers/media/video/cx25840/cx25840-core.h
+++ b/drivers/media/video/cx25840/cx25840-core.h
@@ -47,6 +47,7 @@ struct cx25840_state {
 	u32 id;
 	u32 rev;
 	int is_cx25836;
+	int is_cx23885;
 	int is_initialized;
 	wait_queue_head_t fw_wait;    /* wake up when the fw load is finished */
 	struct work_struct fw_work;   /* work entry for fw load */
diff --git a/drivers/media/video/cx25840/cx25840-firmware.c b/drivers/media/video/cx25840/cx25840-firmware.c
index e852024a5ea3..1ddf724a2c74 100644
--- a/drivers/media/video/cx25840/cx25840-firmware.c
+++ b/drivers/media/video/cx25840/cx25840-firmware.c
@@ -24,6 +24,7 @@
 #include "cx25840-core.h"
 
 #define FWFILE "v4l-cx25840.fw"
+#define FWFILE_CX23885 "v4l-cx23885-avcore-01.fw"
 
 /*
  * Mike Isely <isely@pobox.com> - The FWSEND parameter controls the
@@ -92,10 +93,14 @@ static int fw_write(struct i2c_client *client, u8 * data, int size)
 
 int cx25840_loadfw(struct i2c_client *client)
 {
+	struct cx25840_state *state = i2c_get_clientdata(client);
 	const struct firmware *fw = NULL;
 	u8 buffer[4], *ptr;
 	int size, send, retval;
 
+	if (state->is_cx23885)
+		firmware = FWFILE_CX23885;
+
 	if (request_firmware(&fw, firmware, FWDEV(client)) != 0) {
 		v4l_err(client, "unable to open firmware %s\n", firmware);
 		return -EINVAL;
diff --git a/include/media/cx25840.h b/include/media/cx25840.h
index 8e7e52d659a0..cd599ad29fb2 100644
--- a/include/media/cx25840.h
+++ b/include/media/cx25840.h
@@ -49,6 +49,25 @@ enum cx25840_video_input {
 	CX25840_SVIDEO2 = 0x620,
 	CX25840_SVIDEO3 = 0x730,
 	CX25840_SVIDEO4 = 0x840,
+
+	/* Allow frames to specify specific input configurations */
+	CX25840_VIN1_CH1  = 0x80000000,
+	CX25840_VIN2_CH1  = 0x80000001,
+	CX25840_VIN3_CH1  = 0x80000002,
+	CX25840_VIN4_CH1  = 0x80000003,
+	CX25840_VIN5_CH1  = 0x80000004,
+	CX25840_VIN6_CH1  = 0x80000005,
+	CX25840_VIN7_CH1  = 0x80000006,
+	CX25840_VIN8_CH1  = 0x80000007,
+	CX25840_VIN4_CH2  = 0x80000000,
+	CX25840_VIN5_CH2  = 0x80000010,
+	CX25840_VIN6_CH2  = 0x80000020,
+	CX25840_NONE_CH2  = 0x80000030,
+	CX25840_VIN7_CH3  = 0x80000000,
+	CX25840_VIN8_CH3  = 0x80000040,
+	CX25840_NONE0_CH3 = 0x80000080,
+	CX25840_NONE1_CH3 = 0x800000c0,
+	CX25840_SVIDEO_ON = 0x80000100,
 };
 
 enum cx25840_audio_input {
-- 
cgit v1.2.3


From 27c685a4b09b5e391023d769cddb97e4fcb3b9e1 Mon Sep 17 00:00:00 2001
From: Steven Toth <stoth@hauppauge.com>
Date: Sat, 5 Jan 2008 16:50:14 -0300
Subject: V4L/DVB (7042): xc5000: Tuner analog support

From Zhang: This an updated patch that adds analog support for
the xc5000 tuner driver. it was tested on a Pinnacle PCTV HD 800i
card (patches to follow).

Patch commited as-is, cleanup to follow ... Steve.

Signed-off-by: Chaogui Zhang <czhang1974@gmail.com>
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.tuner |   1 +
 drivers/media/dvb/frontends/xc5000.c     | 213 ++++++++++++++++++++++++++-----
 drivers/media/dvb/frontends/xc5000.h     |  12 +-
 drivers/media/video/tuner-core.c         |  20 +++
 drivers/media/video/tuner-types.c        |   4 +
 include/media/tuner.h                    |   1 +
 6 files changed, 213 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner
index 2211c8eac643..0e2394695bb8 100644
--- a/Documentation/video4linux/CARDLIST.tuner
+++ b/Documentation/video4linux/CARDLIST.tuner
@@ -73,3 +73,4 @@ tuner=71 - Xceive xc2028/xc3028 tuner
 tuner=72 - Thomson FE6600
 tuner=73 - Samsung TCPG 6121P30A
 tuner=75 - Philips TEA5761 FM Radio
+tuner=76 - Xceive 5000 tuner
diff --git a/drivers/media/dvb/frontends/xc5000.c b/drivers/media/dvb/frontends/xc5000.c
index a036530ee4b4..28048d5307a2 100644
--- a/drivers/media/dvb/frontends/xc5000.c
+++ b/drivers/media/dvb/frontends/xc5000.c
@@ -22,6 +22,7 @@
 
 #include <linux/module.h>
 #include <linux/moduleparam.h>
+#include <linux/videodev2.h>
 #include <linux/delay.h>
 #include <linux/dvb/frontend.h>
 #include <linux/i2c.h>
@@ -56,6 +57,10 @@ MODULE_PARM_DESC(debug, "Turn on/off debugging (default:off).");
 #define XC_RESULT_I2C_READ_FAILURE	3
 #define XC_RESULT_OUT_OF_RANGE		5
 
+/* Product id */
+#define XC_PRODUCT_ID_FW_NOT_LOADED	0x2000
+#define XC_PRODUCT_ID_FW_LOADED 	0x1388
+
 /* Registers */
 #define XREG_INIT         0x00
 #define XREG_VIDEO_MODE   0x01
@@ -122,7 +127,29 @@ typedef struct {
 } XC_TV_STANDARD;
 
 /* Tuner standards */
-#define DTV6	17
+#define MN_NTSC_PAL_BTSC	0
+#define MN_NTSC_PAL_A2		1
+#define MN_NTSC_PAL_EIAJ	2
+#define MN_NTSC_PAL_Mono	3
+#define BG_PAL_A2		4
+#define BG_PAL_NICAM		5
+#define BG_PAL_MONO		6
+#define I_PAL_NICAM		7
+#define I_PAL_NICAM_MONO	8
+#define DK_PAL_A2		9
+#define DK_PAL_NICAM		10
+#define DK_PAL_MONO		11
+#define DK_SECAM_A2DK1		12
+#define DK_SECAM_A2LDK3 	13
+#define DK_SECAM_A2MONO 	14
+#define L_SECAM_NICAM		15
+#define LC_SECAM_NICAM		16
+#define DTV6			17
+#define DTV8			18
+#define DTV7_8			19
+#define DTV7			20
+#define FM_Radio_INPUT2 	21
+#define FM_Radio_INPUT1 	22
 
 XC_TV_STANDARD XC5000_Standard[MAX_TV_STANDARD] = {
 	{"M/N-NTSC/PAL-BTSC", 0x0400, 0x8020},
@@ -184,12 +211,13 @@ static void xc5000_TunerReset(struct dvb_frontend *fe)
 
 	dprintk(1, "%s()\n", __FUNCTION__);
 
-	if (priv->cfg->tuner_reset) {
-		ret = priv->cfg->tuner_reset(fe);
+	if (priv->cfg->tuner_callback) {
+		ret = priv->cfg->tuner_callback(priv->cfg->video_dev,
+						XC5000_TUNER_RESET, 0);
 		if (ret)
 			printk(KERN_ERR "xc5000: reset failed\n");
 	} else
-		printk(KERN_ERR "xc5000: no tuner reset function, fatal\n");
+		printk(KERN_ERR "xc5000: no tuner reset callback function, fatal\n");
 }
 
 static int xc_write_reg(struct xc5000_priv *priv, u16 regAddr, u16 i2cData)
@@ -259,7 +287,6 @@ static int xc_load_i2c_sequence(struct dvb_frontend *fe, u8 i2c_sequence[])
 
 	index=0;
 	while ((i2c_sequence[index]!=0xFF) || (i2c_sequence[index+1]!=0xFF)) {
-
 		len = i2c_sequence[index]* 256 + i2c_sequence[index+1];
 		if (len == 0x0000) {
 			/* RESET command */
@@ -311,7 +338,7 @@ static int xc_SetTVStandard(struct xc5000_priv *priv,
 	u16 VideoMode, u16 AudioMode)
 {
 	int ret;
-	dprintk(1, "%s(%d,%d)\n", __FUNCTION__, VideoMode, AudioMode);
+	dprintk(1, "%s(0x%04x,0x%04x)\n", __FUNCTION__, VideoMode, AudioMode);
 	dprintk(1, "%s() Standard = %s\n",
 		__FUNCTION__,
 		XC5000_Standard[priv->video_standard].Name);
@@ -325,7 +352,11 @@ static int xc_SetTVStandard(struct xc5000_priv *priv,
 
 static int xc_shutdown(struct xc5000_priv *priv)
 {
-	return xc_write_reg(priv, XREG_POWER_DOWN, 0);
+	return 0;
+	/* Fixme: cannot bring tuner back alive once shutdown
+	 *        without reloading the driver modules.
+	 *    return xc_write_reg(priv, XREG_POWER_DOWN, 0);
+	 */
 }
 
 static int xc_SetSignalSource(struct xc5000_priv *priv, u16 rf_mode)
@@ -349,7 +380,7 @@ static int xc_set_RF_frequency(struct xc5000_priv *priv, u32 freq_hz)
 {
 	u16 freq_code;
 
-	dprintk(1, "%s(%d)\n", __FUNCTION__, freq_hz);
+	dprintk(1, "%s(%u)\n", __FUNCTION__, freq_hz);
 
 	if ((freq_hz > xc5000_tuner_ops.info.frequency_max) ||
 		(freq_hz < xc5000_tuner_ops.info.frequency_min))
@@ -457,7 +488,7 @@ static int xc_tune_channel(struct xc5000_priv *priv, u32 freq_hz)
 {
 	int found = 0;
 
-	dprintk(1, "%s(%d)\n", __FUNCTION__, freq_hz);
+	dprintk(1, "%s(%u)\n", __FUNCTION__, freq_hz);
 
 	if (xc_set_RF_frequency(priv, freq_hz) != XC_RESULT_SUCCESS)
 		return 0;
@@ -480,7 +511,7 @@ static int xc5000_readreg(struct xc5000_priv *priv, u16 reg, u16 *val)
 	};
 
 	if (i2c_transfer(priv->i2c, msg, 2) != 2) {
-		printk(KERN_WARNING "xc5000 I2C read failed\n");
+		printk(KERN_WARNING "xc5000: I2C read failed\n");
 		return -EREMOTEIO;
 	}
 
@@ -494,7 +525,7 @@ static int xc5000_writeregs(struct xc5000_priv *priv, u8 *buf, u8 len)
 		.flags = 0, .buf = buf, .len = len };
 
 	if (i2c_transfer(priv->i2c, &msg, 1) != 1) {
-		printk(KERN_ERR "xc5000 I2C write failed (len=%i)\n",
+		printk(KERN_ERR "xc5000: I2C write failed (len=%i)\n",
 			(int)len);
 		return -EREMOTEIO;
 	}
@@ -519,16 +550,11 @@ static int xc5000_fwupload(struct dvb_frontend* fe)
 	const struct firmware *fw;
 	int ret;
 
-	if (!priv->cfg->request_firmware) {
-		printk(KERN_ERR "xc5000: no firmware callback, fatal\n");
-		return -EIO;
-	}
-
 	/* request the firmware, this will block and timeout */
 	printk(KERN_INFO "xc5000: waiting for firmware upload (%s)...\n",
 		XC5000_DEFAULT_FIRMWARE);
 
-	ret = priv->cfg->request_firmware(fe, &fw, XC5000_DEFAULT_FIRMWARE);
+	ret = request_firmware(&fw, XC5000_DEFAULT_FIRMWARE, &priv->i2c->dev);
 	if (ret) {
 		printk(KERN_ERR "xc5000: Upload failed. (file not found?)\n");
 		ret = XC_RESULT_RESET_FAILURE;
@@ -601,7 +627,6 @@ static int xc5000_set_params(struct dvb_frontend *fe,
 
 	dprintk(1, "%s() frequency=%d (Hz)\n", __FUNCTION__, params->frequency);
 
-
 	switch(params->u.vsb.modulation) {
 	case VSB_8:
 	case VSB_16:
@@ -658,6 +683,93 @@ static int xc5000_set_params(struct dvb_frontend *fe,
 	return 0;
 }
 
+static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe);
+
+static int xc5000_set_analog_params(struct dvb_frontend *fe,
+	struct analog_parameters *params)
+{
+	struct xc5000_priv *priv = fe->tuner_priv;
+	int ret;
+
+	if(priv->fwloaded == 0)
+		xc_load_fw_and_init_tuner(fe);
+
+	dprintk(1, "%s() frequency=%d (in units of 62.5khz)\n",
+		__FUNCTION__, params->frequency);
+
+	priv->rf_mode = XC_RF_MODE_CABLE; /* Fix me: it could be air. */
+
+	/* params->frequency is in units of 62.5khz */
+	priv->freq_hz = params->frequency * 62500;
+
+	/* FIX ME: Some video standards may have several possible audio
+		   standards. We simply default to one of them here.
+	 */
+	if(params->std & V4L2_STD_MN) {
+		/* default to BTSC audio standard */
+		priv->video_standard = MN_NTSC_PAL_BTSC;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_PAL_BG) {
+		/* default to NICAM audio standard */
+		priv->video_standard = BG_PAL_NICAM;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_PAL_I) {
+		/* default to NICAM audio standard */
+		priv->video_standard = I_PAL_NICAM;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_PAL_DK) {
+		/* default to NICAM audio standard */
+		priv->video_standard = DK_PAL_NICAM;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_SECAM_DK) {
+		/* default to A2 DK1 audio standard */
+		priv->video_standard = DK_SECAM_A2DK1;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_SECAM_L) {
+		priv->video_standard = L_SECAM_NICAM;
+		goto tune_channel;
+	}
+
+	if(params->std & V4L2_STD_SECAM_LC) {
+		priv->video_standard = LC_SECAM_NICAM;
+		goto tune_channel;
+	}
+
+tune_channel:
+	ret = xc_SetSignalSource(priv, priv->rf_mode);
+	if (ret != XC_RESULT_SUCCESS) {
+	printk(KERN_ERR
+			"xc5000: xc_SetSignalSource(%d) failed\n",
+			priv->rf_mode);
+		return -EREMOTEIO;
+	}
+
+	ret = xc_SetTVStandard(priv,
+		XC5000_Standard[priv->video_standard].VideoMode,
+		XC5000_Standard[priv->video_standard].AudioMode);
+	if (ret != XC_RESULT_SUCCESS) {
+		printk(KERN_ERR "xc5000: xc_SetTVStandard failed\n");
+		return -EREMOTEIO;
+	}
+
+	xc_tune_channel(priv, priv->freq_hz);
+
+	if (debug)
+		xc_debug_dump(priv);
+
+	return 0;
+}
+
 static int xc5000_get_frequency(struct dvb_frontend *fe, u32 *freq)
 {
 	struct xc5000_priv *priv = fe->tuner_priv;
@@ -670,6 +782,7 @@ static int xc5000_get_bandwidth(struct dvb_frontend *fe, u32 *bw)
 {
 	struct xc5000_priv *priv = fe->tuner_priv;
 	dprintk(1, "%s()\n", __FUNCTION__);
+
 	*bw = priv->bandwidth;
 	return 0;
 }
@@ -691,13 +804,12 @@ static int xc5000_get_status(struct dvb_frontend *fe, u32 *status)
 static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe)
 {
 	struct xc5000_priv *priv = fe->tuner_priv;
-	int ret;
+	int ret = 0;
 
 	if (priv->fwloaded == 0) {
 		ret = xc5000_fwupload(fe);
 		if (ret != XC_RESULT_SUCCESS)
 			return ret;
-
 		priv->fwloaded = 1;
 	}
 
@@ -720,9 +832,27 @@ static int xc_load_fw_and_init_tuner(struct dvb_frontend *fe)
 static int xc5000_sleep(struct dvb_frontend *fe)
 {
 	struct xc5000_priv *priv = fe->tuner_priv;
+	int ret;
+
 	dprintk(1, "%s()\n", __FUNCTION__);
 
-	return xc_shutdown(priv);
+	/* On Pinnacle PCTV HD 800i, the tuner cannot be reinitialized
+	 * once shutdown without reloading the driver. Maybe I am not
+	 * doing something right.
+	 *
+	 */
+
+	ret = xc_shutdown(priv);
+	if(ret != XC_RESULT_SUCCESS) {
+		printk(KERN_ERR
+			"xc5000: %s() unable to shutdown tuner\n",
+			__FUNCTION__);
+		return -EREMOTEIO;
+	}
+	else {
+		/* priv->fwloaded = 0; */
+		return XC_RESULT_SUCCESS;
+	}
 }
 
 static int xc5000_init(struct dvb_frontend *fe)
@@ -757,14 +887,15 @@ static const struct dvb_tuner_ops xc5000_tuner_ops = {
 		.frequency_step =      50000,
 	},
 
-	.release       = xc5000_release,
-	.init          = xc5000_init,
-	.sleep         = xc5000_sleep,
+	.release	   = xc5000_release,
+	.init		   = xc5000_init,
+	.sleep		   = xc5000_sleep,
 
-	.set_params    = xc5000_set_params,
-	.get_frequency = xc5000_get_frequency,
-	.get_bandwidth = xc5000_get_bandwidth,
-	.get_status    = xc5000_get_status
+	.set_params	   = xc5000_set_params,
+	.set_analog_params = xc5000_set_analog_params,
+	.get_frequency	   = xc5000_get_frequency,
+	.get_bandwidth	   = xc5000_get_bandwidth,
+	.get_status	   = xc5000_get_status
 };
 
 struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe,
@@ -783,14 +914,33 @@ struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe,
 	priv->cfg = cfg;
 	priv->bandwidth = BANDWIDTH_6_MHZ;
 	priv->i2c = i2c;
-	priv->fwloaded = 0;
 
+	/* Check if firmware has been loaded. It is possible that another
+	   instance of the driver has loaded the firmware.
+	 */
 	if (xc5000_readreg(priv, XREG_PRODUCT_ID, &id) != 0) {
 		kfree(priv);
 		return NULL;
 	}
 
-	if ((id != 0x2000) && (id != 0x1388)) {
+	switch(id) {
+	case XC_PRODUCT_ID_FW_LOADED:
+		printk(KERN_INFO
+			"xc5000: Successfully identified at address 0x%02x\n",
+			cfg->i2c_address);
+		printk(KERN_INFO
+			"xc5000: Firmware has been loaded previously\n");
+		priv->fwloaded = 1;
+		break;
+	case XC_PRODUCT_ID_FW_NOT_LOADED:
+		printk(KERN_INFO
+			"xc5000: Successfully identified at address 0x%02x\n",
+			cfg->i2c_address);
+		printk(KERN_INFO
+			"xc5000: Firmware has not been loaded previously\n");
+		priv->fwloaded = 0;
+		break;
+	default:
 		printk(KERN_ERR
 			"xc5000: Device not found at addr 0x%02x (0x%x)\n",
 			cfg->i2c_address, id);
@@ -798,9 +948,6 @@ struct dvb_frontend * xc5000_attach(struct dvb_frontend *fe,
 		return NULL;
 	}
 
-	printk(KERN_INFO "xc5000: successfully identified at address 0x%02x\n",
-		cfg->i2c_address);
-
 	memcpy(&fe->ops.tuner_ops, &xc5000_tuner_ops,
 		sizeof(struct dvb_tuner_ops));
 
diff --git a/drivers/media/dvb/frontends/xc5000.h b/drivers/media/dvb/frontends/xc5000.h
index 941b31948133..85b2d438fa41 100644
--- a/drivers/media/dvb/frontends/xc5000.h
+++ b/drivers/media/dvb/frontends/xc5000.h
@@ -28,13 +28,15 @@ struct dvb_frontend;
 struct i2c_adapter;
 
 struct xc5000_config {
-	u8  i2c_address;
-	u32 if_khz;
-	int (*request_firmware)(struct dvb_frontend *fe,
-		const struct firmware **fw, char *name);
-	int (*tuner_reset)(struct dvb_frontend* fe);
+	u8   i2c_address;
+	u32  if_khz;
+	void *video_dev;
+	int  (*tuner_callback) (void *dev, int command, int arg);
 };
 
+/* xc5000 callback command */
+#define XC5000_TUNER_RESET		0
+
 #if defined(CONFIG_DVB_TUNER_XC5000) || defined(CONFIG_DVB_TUNER_XC5000_MODULE)
 extern struct dvb_frontend* xc5000_attach(struct dvb_frontend *fe,
 					  struct i2c_adapter *i2c,
diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c
index 35976e6cb1b8..16cdeeafeb6c 100644
--- a/drivers/media/video/tuner-core.c
+++ b/drivers/media/video/tuner-core.c
@@ -27,6 +27,7 @@
 #include "tuner-xc2028.h"
 #include "tuner-simple.h"
 #include "tda9887.h"
+#include "xc5000.h"
 
 #define UNSET (-1U)
 
@@ -329,6 +330,8 @@ static void attach_tda829x(struct tuner *t)
 	tda829x_attach(&t->fe, t->i2c->adapter, t->i2c->addr, &cfg);
 }
 
+static struct xc5000_config xc5000_cfg;
+
 static void set_type(struct i2c_client *c, unsigned int type,
 		     unsigned int new_mode_mask, unsigned int new_config,
 		     int (*tuner_callback) (void *dev, int command,int arg))
@@ -428,6 +431,23 @@ static void set_type(struct i2c_client *c, unsigned int type,
 	case TUNER_TDA9887:
 		tda9887_attach(&t->fe, t->i2c->adapter, t->i2c->addr);
 		break;
+	case TUNER_XC5000:
+		xc5000_cfg.i2c_address	  = t->i2c->addr;
+		xc5000_cfg.if_khz	  = 5380;
+		xc5000_cfg.video_dev	  = c->adapter->algo_data;
+		xc5000_cfg.tuner_callback = t->tuner_callback;
+		if (!xc5000_attach(&t->fe, t->i2c->adapter, &xc5000_cfg)) {
+			t->type = TUNER_ABSENT;
+			t->mode_mask = T_UNINITIALIZED;
+			return;
+		}
+		{
+		struct dvb_tuner_ops *xc_tuner_ops;
+		xc_tuner_ops = &t->fe.ops.tuner_ops;
+		if(xc_tuner_ops->init != NULL)
+			xc_tuner_ops->init(&t->fe);
+		}
+		break;
 	default:
 		attach_simple_tuner(t);
 		break;
diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c
index e2cd05a05802..883047f9c28c 100644
--- a/drivers/media/video/tuner-types.c
+++ b/drivers/media/video/tuner-types.c
@@ -1475,6 +1475,10 @@ struct tunertype tuners[] = {
 		.name   = "Philips TEA5761 FM Radio",
 		/* see tea5767.c for details */
 	},
+	[TUNER_XC5000] = { /* Xceive 5000 */
+		.name   = "Xceive 5000 tuner",
+		/* see xc5000.c for details */
+	},
 };
 
 unsigned const int tuner_count = ARRAY_SIZE(tuners);
diff --git a/include/media/tuner.h b/include/media/tuner.h
index 97be269afbee..1bf24a6ed8f1 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -121,6 +121,7 @@
 #define TUNER_SAMSUNG_TCPG_6121P30A     73 	/* Hauppauge PVR-500 PAL */
 #define TUNER_TDA9887                   74      /* This tuner should be used only internally */
 #define TUNER_TEA5761			75	/* Only FM Radio Tuner */
+#define TUNER_XC5000			76	/* Xceive Silicon Tuner */
 
 /* tv card specific */
 #define TDA9887_PRESENT 		(1<<0)
-- 
cgit v1.2.3


From 9121106a7f149f9702ed7c2ac0d86e9feb54ede8 Mon Sep 17 00:00:00 2001
From: Steven Toth <stoth@hauppauge.com>
Date: Tue, 22 Jan 2008 01:00:33 -0300
Subject: V4L/DVB (7058): IR corrections for the Pinnacle 800i

IR corrections for the Pinnacle 800i

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Chaogui Zhang <czhang1974@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/common/ir-keymaps.c     | 35 +++++++++++++++++++++++++++++++++++
 drivers/media/video/cx88/cx88-input.c |  6 +++++-
 include/media/ir-common.h             |  1 +
 3 files changed, 41 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c
index ee2c69e3152a..a4a937c90534 100644
--- a/drivers/media/common/ir-keymaps.c
+++ b/drivers/media/common/ir-keymaps.c
@@ -1849,6 +1849,41 @@ IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE] = {
 
 EXPORT_SYMBOL_GPL(ir_codes_fusionhdtv_mce);
 
+/* Pinnacle PCTV HD 800i mini remote */
+IR_KEYTAB_TYPE ir_codes_pinnacle_pctv_hd[IR_KEYTAB_SIZE] = {
+
+	[0x0f] = KEY_1,
+	[0x15] = KEY_2,
+	[0x10] = KEY_3,
+	[0x18] = KEY_4,
+	[0x1b] = KEY_5,
+	[0x1e] = KEY_6,
+	[0x11] = KEY_7,
+	[0x21] = KEY_8,
+	[0x12] = KEY_9,
+	[0x27] = KEY_0,
+
+	[0x24] = KEY_ZOOM,
+	[0x2a] = KEY_SUBTITLE,
+
+	[0x00] = KEY_MUTE,
+	[0x01] = KEY_ENTER,	/* Pinnacle Logo */
+	[0x39] = KEY_POWER,
+
+	[0x03] = KEY_VOLUMEUP,
+	[0x09] = KEY_VOLUMEDOWN,
+	[0x06] = KEY_CHANNELUP,
+	[0x0c] = KEY_CHANNELDOWN,
+
+	[0x2d] = KEY_REWIND,
+	[0x30] = KEY_PLAYPAUSE,
+	[0x33] = KEY_FASTFORWARD,
+	[0x3c] = KEY_STOP,
+	[0x36] = KEY_RECORD,
+	[0x3f] = KEY_EPG,	/* Labeled "?" */
+};
+EXPORT_SYMBOL_GPL(ir_codes_pinnacle_pctv_hd);
+
 /*
  * Igor Kuznetsov <igk72@ya.ru>
  * Andrey J. Melnikov <temnota@kmv.ru>
diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c
index f79a1c77414c..bb0911b4d2f6 100644
--- a/drivers/media/video/cx88/cx88-input.c
+++ b/drivers/media/video/cx88/cx88-input.c
@@ -224,7 +224,6 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci)
 	case CX88_BOARD_HAUPPAUGE_NOVASPLUS_S1:
 	case CX88_BOARD_HAUPPAUGE_HVR1100:
 	case CX88_BOARD_HAUPPAUGE_HVR3000:
-	case CX88_BOARD_PINNACLE_PCTV_HD_800i:
 		ir_codes = ir_codes_hauppauge_new;
 		ir_type = IR_TYPE_RC5;
 		ir->sampling = 1;
@@ -306,6 +305,11 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci)
 		ir->mask_keycode = 0xfa;
 		ir->polling = 50; /* ms */
 		break;
+	case CX88_BOARD_PINNACLE_PCTV_HD_800i:
+		ir_codes = ir_codes_pinnacle_pctv_hd;
+		ir_type = IR_TYPE_RC5;
+		ir->sampling = 1;
+		break;
 	}
 
 	if (NULL == ir_codes) {
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index 9c2fc09a7133..831547d79683 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -141,6 +141,7 @@ extern IR_KEYTAB_TYPE ir_codes_encore_enltv[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_tt_1500[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_fusionhdtv_mce[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_behold[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_pinnacle_pctv_hd[IR_KEYTAB_SIZE];
 
 #endif
 
-- 
cgit v1.2.3


From 1c029fd658baa2442e8e51dc9c819301cad95777 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:05 +0100
Subject: ide: remove ->dma_master field from ide_hwif_t (take 5)

* Convert cmd64x, hpt366 and pdc202xx_old host drivers to use
  pci_resource_start(hwif->pci_dev, 4) instead of hwif->dma_master.

* Remove no longer needed ->dma_master field from ide_hwif_t.

v2:
* Use the more readable 'hwif->dma_base - (hwif->channel * 8)' instead of
  pci_resource_start(hwif->pci_dev, 4).

v3:
* Use hwif->extra_base in hpt366/pdc20xx_old + some cosmetic fixups over v2
  (suggested by Sergei).

v4:
* Correct offsets in hpt3xxn_set_clock().

v5:
* Use hwif->extra_base in hpt366 for _real_ this time. (Noticed by Sergei)

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c          |  7 -------
 drivers/ide/ide.c              |  2 --
 drivers/ide/pci/cmd64x.c       |  8 +++++---
 drivers/ide/pci/hpt366.c       | 21 +++++++++++----------
 drivers/ide/pci/pdc202xx_old.c | 12 ++++++------
 include/linux/ide.h            |  1 -
 6 files changed, 22 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 4703837bf1fc..7568c3e81f2f 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -968,11 +968,6 @@ void ide_setup_dma(ide_hwif_t *hwif, unsigned long base, unsigned num_ports)
 
 	hwif->dma_base = base;
 
-	if (hwif->mate)
-		hwif->dma_master = hwif->channel ? hwif->mate->dma_base : base;
-	else
-		hwif->dma_master = base;
-
 	if (!(hwif->dma_command))
 		hwif->dma_command	= hwif->dma_base;
 	if (!(hwif->dma_vendor1))
@@ -1014,8 +1009,6 @@ void ide_setup_dma(ide_hwif_t *hwif, unsigned long base, unsigned num_ports)
 		       hwif->drives[1].name, (dma_stat & 0x40) ? "DMA" : "pio");
 	}
 	printk("\n");
-
-	BUG_ON(!hwif->dma_master);
 }
 
 EXPORT_SYMBOL_GPL(ide_setup_dma);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 54943da6e4e5..9ab5458fe950 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -468,7 +468,6 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 #endif
 
 	hwif->dma_base			= tmp_hwif->dma_base;
-	hwif->dma_master		= tmp_hwif->dma_master;
 	hwif->dma_command		= tmp_hwif->dma_command;
 	hwif->dma_vendor1		= tmp_hwif->dma_vendor1;
 	hwif->dma_status		= tmp_hwif->dma_status;
@@ -602,7 +601,6 @@ void ide_unregister(unsigned int index)
 		(void) ide_release_dma(hwif);
 
 		hwif->dma_base = 0;
-		hwif->dma_master = 0;
 		hwif->dma_command = 0;
 		hwif->dma_vendor1 = 0;
 		hwif->dma_status = 0;
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index bc553337b1be..f3613bac9dbe 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -333,14 +333,15 @@ static void cmd64x_set_dma_mode(ide_drive_t *drive, const u8 speed)
 static int cmd648_ide_dma_end (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
+	unsigned long base	= hwif->dma_base - (hwif->channel * 8);
 	int err			= __ide_dma_end(drive);
 	u8  irq_mask		= hwif->channel ? MRDMODE_INTR_CH1 :
 						  MRDMODE_INTR_CH0;
-	u8  mrdmode		= inb(hwif->dma_master + 0x01);
+	u8  mrdmode		= inb(base + 1);
 
 	/* clear the interrupt bit */
 	outb((mrdmode & ~(MRDMODE_INTR_CH0 | MRDMODE_INTR_CH1)) | irq_mask,
-	     hwif->dma_master + 0x01);
+	     base + 1);
 
 	return err;
 }
@@ -365,10 +366,11 @@ static int cmd64x_ide_dma_end (ide_drive_t *drive)
 static int cmd648_ide_dma_test_irq (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
+	unsigned long base	= hwif->dma_base - (hwif->channel * 8);
 	u8 irq_mask		= hwif->channel ? MRDMODE_INTR_CH1 :
 						  MRDMODE_INTR_CH0;
 	u8 dma_stat		= inb(hwif->dma_status);
-	u8 mrdmode		= inb(hwif->dma_master + 0x01);
+	u8 mrdmode		= inb(base + 1);
 
 #ifdef DEBUG
 	printk("%s: dma_stat: 0x%02x mrdmode: 0x%02x irq_mask: 0x%02x\n",
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index df14d692743c..d3826a66834d 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -894,32 +894,33 @@ static int hpt374_ide_dma_end(ide_drive_t *drive)
 
 static void hpt3xxn_set_clock(ide_hwif_t *hwif, u8 mode)
 {
-	u8 scr2 = inb(hwif->dma_master + 0x7b);
+	unsigned long base = hwif->extra_base;
+	u8 scr2 = inb(base + 0x6b);
 
 	if ((scr2 & 0x7f) == mode)
 		return;
 
 	/* Tristate the bus */
-	outb(0x80, hwif->dma_master + 0x73);
-	outb(0x80, hwif->dma_master + 0x77);
+	outb(0x80, base + 0x63);
+	outb(0x80, base + 0x67);
 
 	/* Switch clock and reset channels */
-	outb(mode, hwif->dma_master + 0x7b);
-	outb(0xc0, hwif->dma_master + 0x79);
+	outb(mode, base + 0x6b);
+	outb(0xc0, base + 0x69);
 
 	/*
 	 * Reset the state machines.
 	 * NOTE: avoid accidentally enabling the disabled channels.
 	 */
-	outb(inb(hwif->dma_master + 0x70) | 0x32, hwif->dma_master + 0x70);
-	outb(inb(hwif->dma_master + 0x74) | 0x32, hwif->dma_master + 0x74);
+	outb(inb(base + 0x60) | 0x32, base + 0x60);
+	outb(inb(base + 0x64) | 0x32, base + 0x64);
 
 	/* Complete reset */
-	outb(0x00, hwif->dma_master + 0x79);
+	outb(0x00, base + 0x69);
 
 	/* Reconnect channels to bus */
-	outb(0x00, hwif->dma_master + 0x73);
-	outb(0x00, hwif->dma_master + 0x77);
+	outb(0x00, base + 0x63);
+	outb(0x00, base + 0x67);
 }
 
 /**
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index e09742e2ba59..22c7a7533b69 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -162,7 +162,7 @@ static u8 pdc202xx_old_cable_detect (ide_hwif_t *hwif)
  */
 static void pdc_old_enable_66MHz_clock(ide_hwif_t *hwif)
 {
-	unsigned long clock_reg = hwif->dma_master + 0x11;
+	unsigned long clock_reg = hwif->extra_base + 0x01;
 	u8 clock = inb(clock_reg);
 
 	outb(clock | (hwif->channel ? 0x08 : 0x02), clock_reg);
@@ -170,7 +170,7 @@ static void pdc_old_enable_66MHz_clock(ide_hwif_t *hwif)
 
 static void pdc_old_disable_66MHz_clock(ide_hwif_t *hwif)
 {
-	unsigned long clock_reg = hwif->dma_master + 0x11;
+	unsigned long clock_reg = hwif->extra_base + 0x01;
 	u8 clock = inb(clock_reg);
 
 	outb(clock & ~(hwif->channel ? 0x08 : 0x02), clock_reg);
@@ -193,7 +193,7 @@ static void pdc202xx_old_ide_dma_start(ide_drive_t *drive)
 	if (drive->media != ide_disk || drive->addressing == 1) {
 		struct request *rq	= HWGROUP(drive)->rq;
 		ide_hwif_t *hwif	= HWIF(drive);
-		unsigned long high_16   = hwif->dma_master;
+		unsigned long high_16	= hwif->extra_base - 16;
 		unsigned long atapi_reg	= high_16 + (hwif->channel ? 0x24 : 0x20);
 		u32 word_count	= 0;
 		u8 clock = inb(high_16 + 0x11);
@@ -212,7 +212,7 @@ static int pdc202xx_old_ide_dma_end(ide_drive_t *drive)
 {
 	if (drive->media != ide_disk || drive->addressing == 1) {
 		ide_hwif_t *hwif	= HWIF(drive);
-		unsigned long high_16	= hwif->dma_master;
+		unsigned long high_16	= hwif->extra_base - 16;
 		unsigned long atapi_reg	= high_16 + (hwif->channel ? 0x24 : 0x20);
 		u8 clock		= 0;
 
@@ -228,7 +228,7 @@ static int pdc202xx_old_ide_dma_end(ide_drive_t *drive)
 static int pdc202xx_old_ide_dma_test_irq(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	unsigned long high_16	= hwif->dma_master;
+	unsigned long high_16	= hwif->extra_base - 16;
 	u8 dma_stat		= inb(hwif->dma_status);
 	u8 sc1d			= inb(high_16 + 0x001d);
 
@@ -271,7 +271,7 @@ static void pdc202xx_dma_timeout(ide_drive_t *drive)
 
 static void pdc202xx_reset_host (ide_hwif_t *hwif)
 {
-	unsigned long high_16	= hwif->dma_master;
+	unsigned long high_16	= hwif->extra_base - 16;
 	u8 udma_speed_flag	= inb(high_16 | 0x001f);
 
 	outb(udma_speed_flag | 0x10, high_16 | 0x001f);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9a6a41e7079f..1fb03b630676 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -766,7 +766,6 @@ typedef struct hwif_s {
 	int		rqsize;		/* max sectors per request */
 	int		irq;		/* our irq number */
 
-	unsigned long	dma_master;	/* reference base addr dmabase */
 	unsigned long	dma_base;	/* base addr for dma ports */
 	unsigned long	dma_command;	/* dma command register */
 	unsigned long	dma_vendor1;	/* dma vendor 1 register */
-- 
cgit v1.2.3


From cd2a2d969761c26542095c01324201ca0b3ee896 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:06 +0100
Subject: ide: remove task_ioreg_t typedef (take 2)

Remove task_ioreg_t typedef from the kernel code (but leave it
in <linux/hdreg.h> for #ifndef/#endif __KERNEL__ case).

While at it also move sata_ioreg_t typedef under #ifndef/#endif __KERNEL__.

v2:
Remove name of the second parameter from ide_execute_command() declaration.
(Noticed by Sergei).

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     | 20 ++++++++++----------
 drivers/ide/ide-iops.c     |  5 +++--
 drivers/ide/ide-taskfile.c |  4 ++--
 include/linux/hdreg.h      | 46 +++++++++++++++++++++++-----------------------
 include/linux/ide.h        |  7 ++++---
 5 files changed, 42 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index b1781908e1f2..322c4691836a 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -138,7 +138,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 	ide_hwif_t *hwif	= HWIF(drive);
 	unsigned int dma	= drive->using_dma;
 	u8 lba48		= (drive->addressing == 1) ? 1 : 0;
-	task_ioreg_t command	= WIN_NOP;
+	u8 command		= WIN_NOP;
 	ata_nsector_t		nsectors;
 
 	nsectors.all		= (u16) rq->nr_sectors;
@@ -162,7 +162,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 
 	if (drive->select.b.lba) {
 		if (lba48) {
-			task_ioreg_t tasklets[10];
+			u8 tasklets[10];
 
 			pr_debug("%s: LBA=0x%012llx\n", drive->name,
 					(unsigned long long)block);
@@ -171,16 +171,16 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 			tasklets[1] = 0;
 			tasklets[2] = nsectors.b.low;
 			tasklets[3] = nsectors.b.high;
-			tasklets[4] = (task_ioreg_t) block;
-			tasklets[5] = (task_ioreg_t) (block>>8);
-			tasklets[6] = (task_ioreg_t) (block>>16);
-			tasklets[7] = (task_ioreg_t) (block>>24);
+			tasklets[4] = (u8) block;
+			tasklets[5] = (u8)(block >>  8);
+			tasklets[6] = (u8)(block >> 16);
+			tasklets[7] = (u8)(block >> 24);
 			if (sizeof(block) == 4) {
-				tasklets[8] = (task_ioreg_t) 0;
-				tasklets[9] = (task_ioreg_t) 0;
+				tasklets[8] = 0;
+				tasklets[9] = 0;
 			} else {
-				tasklets[8] = (task_ioreg_t)((u64)block >> 32);
-				tasklets[9] = (task_ioreg_t)((u64)block >> 40);
+				tasklets[8] = (u8)((u64)block >> 32);
+				tasklets[9] = (u8)((u64)block >> 40);
 			}
 #ifdef DEBUG
 			printk("%s: 0x%02x%02x 0x%02x%02x%02x%02x%02x%02x\n",
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index bb9693dabe41..f4e8a0cf06e2 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -902,8 +902,9 @@ EXPORT_SYMBOL(ide_set_handler);
  *	handler and IRQ setup do not race. All IDE command kick off
  *	should go via this function or do equivalent locking.
  */
- 
-void ide_execute_command(ide_drive_t *drive, task_ioreg_t cmd, ide_handler_t *handler, unsigned timeout, ide_expiry_t *expiry)
+
+void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
+			 unsigned timeout, ide_expiry_t *expiry)
 {
 	unsigned long flags;
 	ide_hwgroup_t *hwgroup = HWGROUP(drive);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 2b60f1b0437e..91948a46cc6e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -519,8 +519,8 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	ide_task_t		args;
 	u8 *outbuf		= NULL;
 	u8 *inbuf		= NULL;
-	task_ioreg_t *argsptr	= args.tfRegister;
-	task_ioreg_t *hobsptr	= args.hobRegister;
+	u8 *argsptr		= args.tfRegister;
+	u8 *hobsptr		= args.hobRegister;
 	int err			= 0;
 	int tasksize		= sizeof(struct ide_task_request_s);
 	unsigned int taskin	= 0;
diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h
index 818c6afc1091..3bcb88560418 100644
--- a/include/linux/hdreg.h
+++ b/include/linux/hdreg.h
@@ -87,10 +87,10 @@
 #ifndef __KERNEL__
 #define IDE_TASKFILE_STD_OUT_FLAGS	0xFE
 #define IDE_HOB_STD_OUT_FLAGS		0x3C
-#endif
 
 typedef unsigned char task_ioreg_t;
 typedef unsigned long sata_ioreg_t;
+#endif
 
 typedef union ide_reg_valid_s {
 	unsigned all				: 16;
@@ -116,8 +116,8 @@ typedef union ide_reg_valid_s {
 } ide_reg_valid_t;
 
 typedef struct ide_task_request_s {
-	task_ioreg_t	io_ports[8];
-	task_ioreg_t	hob_ports[8];
+	__u8		io_ports[8];
+	__u8		hob_ports[8];
 	ide_reg_valid_t	out_flags;
 	ide_reg_valid_t	in_flags;
 	int		data_phase;
@@ -133,32 +133,32 @@ typedef struct ide_ioctl_request_s {
 } ide_ioctl_request_t;
 
 struct hd_drive_cmd_hdr {
-	task_ioreg_t command;
-	task_ioreg_t sector_number;
-	task_ioreg_t feature;
-	task_ioreg_t sector_count;
+	__u8 command;
+	__u8 sector_number;
+	__u8 feature;
+	__u8 sector_count;
 };
 
 typedef struct hd_drive_task_hdr {
-	task_ioreg_t data;
-	task_ioreg_t feature;
-	task_ioreg_t sector_count;
-	task_ioreg_t sector_number;
-	task_ioreg_t low_cylinder;
-	task_ioreg_t high_cylinder;
-	task_ioreg_t device_head;
-	task_ioreg_t command;
+	__u8 data;
+	__u8 feature;
+	__u8 sector_count;
+	__u8 sector_number;
+	__u8 low_cylinder;
+	__u8 high_cylinder;
+	__u8 device_head;
+	__u8 command;
 } task_struct_t;
 
 typedef struct hd_drive_hob_hdr {
-	task_ioreg_t data;
-	task_ioreg_t feature;
-	task_ioreg_t sector_count;
-	task_ioreg_t sector_number;
-	task_ioreg_t low_cylinder;
-	task_ioreg_t high_cylinder;
-	task_ioreg_t device_head;
-	task_ioreg_t control;
+	__u8 data;
+	__u8 feature;
+	__u8 sector_count;
+	__u8 sector_number;
+	__u8 low_cylinder;
+	__u8 high_cylinder;
+	__u8 device_head;
+	__u8 control;
 } hob_struct_t;
 
 #define TASKFILE_INVALID		0x7fff
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1fb03b630676..66a38f101175 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1019,7 +1019,8 @@ int ide_end_dequeued_request(ide_drive_t *drive, struct request *rq,
 
 extern void ide_set_handler (ide_drive_t *drive, ide_handler_t *handler, unsigned int timeout, ide_expiry_t *expiry);
 
-extern void ide_execute_command(ide_drive_t *, task_ioreg_t cmd, ide_handler_t *, unsigned int, ide_expiry_t *);
+void ide_execute_command(ide_drive_t *, u8, ide_handler_t *, unsigned int,
+			 ide_expiry_t *);
 
 ide_startstop_t __ide_error(ide_drive_t *, struct request *, u8, u8);
 
@@ -1068,8 +1069,8 @@ typedef struct ide_task_s {
  *	struct hd_drive_hob_hdr		hobf;
  *	hob_struct_t		hobf;
  */
-	task_ioreg_t		tfRegister[8];
-	task_ioreg_t		hobRegister[8];
+	u8			tfRegister[8];
+	u8			hobRegister[8];
 	ide_reg_valid_t		tf_out_flags;
 	ide_reg_valid_t		tf_in_flags;
 	int			data_phase;
-- 
cgit v1.2.3


From 650d841d9e053a618dd8ce753422f91b493cf2f6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:06 +0100
Subject: ide: add struct ide_taskfile (take 2)

* Don't set write-only ide_task_t.hobRegister[6] and ide_task_t.hobRegister[7]
  in idedisk_set_max_address_ext().

* Add struct ide_taskfile and use it in ide_task_t instead of tfRegister[]
  and hobRegister[].

* Remove no longer needed IDE_CONTROL_OFFSET_HOB define.

* Add #ifndef/#endif __KERNEL__ around definitions of {task,hob}_struct_t.

While at it:

* Use ATA_LBA define for LBA bit (0x40) as suggested by Tejun Heo.

v2:
* Add missing newlines. (Noticed by Sergei)

* Use ~ATA_LBA instead of 0xBF. (Noticed by Sergei)

* Use unnamed unions for error/feature and status/command.
  (Suggested by Sergei).

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c     |   8 +--
 drivers/ide/ide-disk.c     | 120 +++++++++++++++++++++------------------------
 drivers/ide/ide-io.c       |  61 ++++++++++++-----------
 drivers/ide/ide-iops.c     |  12 ++---
 drivers/ide/ide-lib.c      |   3 +-
 drivers/ide/ide-taskfile.c |  99 +++++++++++++++++--------------------
 include/linux/hdreg.h      |   4 +-
 include/linux/ide.h        |  43 ++++++++++++----
 8 files changed, 179 insertions(+), 171 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 899d56536e80..747c51889f7d 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -388,13 +388,7 @@ static int taskfile_load_raw(ide_drive_t *drive,
 	args.handler      = &task_no_data_intr;
 
 	/* convert gtf to IDE Taskfile */
-	args.tfRegister[1] = gtf->tfa[0];	/* 0x1f1 */
-	args.tfRegister[2] = gtf->tfa[1];	/* 0x1f2 */
-	args.tfRegister[3] = gtf->tfa[2];	/* 0x1f3 */
-	args.tfRegister[4] = gtf->tfa[3];	/* 0x1f4 */
-	args.tfRegister[5] = gtf->tfa[4];	/* 0x1f5 */
-	args.tfRegister[6] = gtf->tfa[5];	/* 0x1f6 */
-	args.tfRegister[7] = gtf->tfa[6];	/* 0x1f7 */
+	memcpy(&args.tf_array[7], &gtf->tfa, 7);
 
 	if (ide_noacpitfs) {
 		DEBPRINT("_GTF execution disabled\n");
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 322c4691836a..a4c4d4350560 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -310,23 +310,22 @@ static ide_startstop_t ide_do_rw_disk (ide_drive_t *drive, struct request *rq, s
 static unsigned long idedisk_read_native_max_address(ide_drive_t *drive)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 	unsigned long addr = 0;
 
 	/* Create IDE/ATA command request structure */
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_SELECT_OFFSET]	= 0x40;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_READ_NATIVE_MAX;
+	tf->device  = ATA_LBA;
+	tf->command = WIN_READ_NATIVE_MAX;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
 	/* submit command request */
 	ide_raw_taskfile(drive, &args, NULL);
 
 	/* if OK, compute maximum address value */
-	if ((args.tfRegister[IDE_STATUS_OFFSET] & 0x01) == 0) {
-		addr = ((args.tfRegister[IDE_SELECT_OFFSET] & 0x0f) << 24)
-		     | ((args.tfRegister[  IDE_HCYL_OFFSET]       ) << 16)
-		     | ((args.tfRegister[  IDE_LCYL_OFFSET]       ) <<  8)
-		     | ((args.tfRegister[IDE_SECTOR_OFFSET]       ));
+	if ((tf->status & 0x01) == 0) {
+		addr = ((tf->device & 0xf) << 24) |
+		       (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
 		addr++;	/* since the return value is (maxlba - 1), we add 1 */
 	}
 	return addr;
@@ -335,26 +334,24 @@ static unsigned long idedisk_read_native_max_address(ide_drive_t *drive)
 static unsigned long long idedisk_read_native_max_address_ext(ide_drive_t *drive)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 	unsigned long long addr = 0;
 
 	/* Create IDE/ATA command request structure */
 	memset(&args, 0, sizeof(ide_task_t));
-
-	args.tfRegister[IDE_SELECT_OFFSET]	= 0x40;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_READ_NATIVE_MAX_EXT;
+	tf->device  = ATA_LBA;
+	tf->command = WIN_READ_NATIVE_MAX_EXT;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
         /* submit command request */
         ide_raw_taskfile(drive, &args, NULL);
 
 	/* if OK, compute maximum address value */
-	if ((args.tfRegister[IDE_STATUS_OFFSET] & 0x01) == 0) {
-		u32 high = (args.hobRegister[IDE_HCYL_OFFSET] << 16) |
-			   (args.hobRegister[IDE_LCYL_OFFSET] <<  8) |
-			    args.hobRegister[IDE_SECTOR_OFFSET];
-		u32 low  = ((args.tfRegister[IDE_HCYL_OFFSET])<<16) |
-			   ((args.tfRegister[IDE_LCYL_OFFSET])<<8) |
-			    (args.tfRegister[IDE_SECTOR_OFFSET]);
+	if ((tf->status & 0x01) == 0) {
+		u32 high, low;
+
+		high = (tf->hob_lbah << 16) | (tf->hob_lbam << 8) | tf->hob_lbal;
+		low  = (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
 		addr = ((__u64)high << 24) | low;
 		addr++;	/* since the return value is (maxlba - 1), we add 1 */
 	}
@@ -368,26 +365,25 @@ static unsigned long long idedisk_read_native_max_address_ext(ide_drive_t *drive
 static unsigned long idedisk_set_max_address(ide_drive_t *drive, unsigned long addr_req)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 	unsigned long addr_set = 0;
 	
 	addr_req--;
 	/* Create IDE/ATA command request structure */
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_SECTOR_OFFSET]	= ((addr_req >>  0) & 0xff);
-	args.tfRegister[IDE_LCYL_OFFSET]	= ((addr_req >>  8) & 0xff);
-	args.tfRegister[IDE_HCYL_OFFSET]	= ((addr_req >> 16) & 0xff);
-	args.tfRegister[IDE_SELECT_OFFSET]	= ((addr_req >> 24) & 0x0f) | 0x40;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SET_MAX;
+	tf->lbal    =  (addr_req >>  0) & 0xff;
+	tf->lbam    =  (addr_req >>  8) & 0xff;
+	tf->lbah    =  (addr_req >> 16) & 0xff;
+	tf->device  = ((addr_req >> 24) & 0x0f) | ATA_LBA;
+	tf->command = WIN_SET_MAX;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
 	/* submit command request */
 	ide_raw_taskfile(drive, &args, NULL);
 	/* if OK, read new maximum address value */
-	if ((args.tfRegister[IDE_STATUS_OFFSET] & 0x01) == 0) {
-		addr_set = ((args.tfRegister[IDE_SELECT_OFFSET] & 0x0f) << 24)
-			 | ((args.tfRegister[  IDE_HCYL_OFFSET]       ) << 16)
-			 | ((args.tfRegister[  IDE_LCYL_OFFSET]       ) <<  8)
-			 | ((args.tfRegister[IDE_SECTOR_OFFSET]       ));
+	if ((tf->status & 0x01) == 0) {
+		addr_set = ((tf->device & 0xf) << 24) |
+			   (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
 		addr_set++;
 	}
 	return addr_set;
@@ -396,33 +392,30 @@ static unsigned long idedisk_set_max_address(ide_drive_t *drive, unsigned long a
 static unsigned long long idedisk_set_max_address_ext(ide_drive_t *drive, unsigned long long addr_req)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 	unsigned long long addr_set = 0;
 
 	addr_req--;
 	/* Create IDE/ATA command request structure */
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_SECTOR_OFFSET]	= ((addr_req >>  0) & 0xff);
-	args.tfRegister[IDE_LCYL_OFFSET]	= ((addr_req >>= 8) & 0xff);
-	args.tfRegister[IDE_HCYL_OFFSET]	= ((addr_req >>= 8) & 0xff);
-	args.tfRegister[IDE_SELECT_OFFSET]      = 0x40;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SET_MAX_EXT;
-	args.hobRegister[IDE_SECTOR_OFFSET]	= (addr_req >>= 8) & 0xff;
-	args.hobRegister[IDE_LCYL_OFFSET]	= (addr_req >>= 8) & 0xff;
-	args.hobRegister[IDE_HCYL_OFFSET]	= (addr_req >>= 8) & 0xff;
-	args.hobRegister[IDE_SELECT_OFFSET]	= 0x40;
-	args.hobRegister[IDE_CONTROL_OFFSET_HOB]= (drive->ctl|0x80);
+	tf->lbal     = (addr_req >>  0) & 0xff;
+	tf->lbam     = (addr_req >>= 8) & 0xff;
+	tf->lbah     = (addr_req >>= 8) & 0xff;
+	tf->device   = ATA_LBA;
+	tf->command  = WIN_SET_MAX_EXT;
+	tf->hob_lbal = (addr_req >>= 8) & 0xff;
+	tf->hob_lbam = (addr_req >>= 8) & 0xff;
+	tf->hob_lbah = (addr_req >>= 8) & 0xff;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
 	/* submit command request */
 	ide_raw_taskfile(drive, &args, NULL);
 	/* if OK, compute maximum address value */
-	if ((args.tfRegister[IDE_STATUS_OFFSET] & 0x01) == 0) {
-		u32 high = (args.hobRegister[IDE_HCYL_OFFSET] << 16) |
-			   (args.hobRegister[IDE_LCYL_OFFSET] <<  8) |
-			    args.hobRegister[IDE_SECTOR_OFFSET];
-		u32 low  = ((args.tfRegister[IDE_HCYL_OFFSET])<<16) |
-			   ((args.tfRegister[IDE_LCYL_OFFSET])<<8) |
-			    (args.tfRegister[IDE_SECTOR_OFFSET]);
+	if ((tf->status & 0x01) == 0) {
+		u32 high, low;
+
+		high = (tf->hob_lbah << 16) | (tf->hob_lbam << 8) | tf->hob_lbal;
+		low  = (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
 		addr_set = ((__u64)high << 24) | low;
 		addr_set++;
 	}
@@ -556,12 +549,13 @@ static sector_t idedisk_capacity (ide_drive_t *drive)
 static int smart_enable(ide_drive_t *drive)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_FEATURE_OFFSET]	= SMART_ENABLE;
-	args.tfRegister[IDE_LCYL_OFFSET]	= SMART_LCYL_PASS;
-	args.tfRegister[IDE_HCYL_OFFSET]	= SMART_HCYL_PASS;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SMART;
+	tf->feature = SMART_ENABLE;
+	tf->lbam    = SMART_LCYL_PASS;
+	tf->lbah    = SMART_HCYL_PASS;
+	tf->command = WIN_SMART;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
 	return ide_raw_taskfile(drive, &args, NULL);
@@ -570,13 +564,14 @@ static int smart_enable(ide_drive_t *drive)
 static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
 {
 	ide_task_t args;
+	struct ide_taskfile *tf = &args.tf;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_FEATURE_OFFSET]	= sub_cmd;
-	args.tfRegister[IDE_NSECTOR_OFFSET]	= 0x01;
-	args.tfRegister[IDE_LCYL_OFFSET]	= SMART_LCYL_PASS;
-	args.tfRegister[IDE_HCYL_OFFSET]	= SMART_HCYL_PASS;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SMART;
+	tf->feature = sub_cmd;
+	tf->nsect   = 0x01;
+	tf->lbam    = SMART_LCYL_PASS;
+	tf->lbah    = SMART_HCYL_PASS;
+	tf->command = WIN_SMART;
 	args.command_type			= IDE_DRIVE_TASK_IN;
 	args.data_phase				= TASKFILE_IN;
 	args.handler				= &task_in_intr;
@@ -753,9 +748,9 @@ static int write_cache(ide_drive_t *drive, int arg)
 
 	if (ide_id_has_flush_cache(drive->id)) {
 		memset(&args, 0, sizeof(ide_task_t));
-		args.tfRegister[IDE_FEATURE_OFFSET]	= (arg) ?
+		args.tf.feature = arg ?
 			SETFEATURES_EN_WCACHE : SETFEATURES_DIS_WCACHE;
-		args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SETFEATURES;
+		args.tf.command = WIN_SETFEATURES;
 		args.command_type		= IDE_DRIVE_TASK_NO_DATA;
 		args.handler			= &task_no_data_intr;
 		err = ide_raw_taskfile(drive, &args, NULL);
@@ -774,9 +769,9 @@ static int do_idedisk_flushcache (ide_drive_t *drive)
 
 	memset(&args, 0, sizeof(ide_task_t));
 	if (ide_id_has_flush_cache_ext(drive->id))
-		args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_FLUSH_CACHE_EXT;
+		args.tf.command = WIN_FLUSH_CACHE_EXT;
 	else
-		args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_FLUSH_CACHE;
+		args.tf.command = WIN_FLUSH_CACHE;
 	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
 	args.handler				= &task_no_data_intr;
 	return ide_raw_taskfile(drive, &args, NULL);
@@ -790,10 +785,9 @@ static int set_acoustic (ide_drive_t *drive, int arg)
 		return -EINVAL;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_FEATURE_OFFSET]	= (arg) ? SETFEATURES_EN_AAM :
-							  SETFEATURES_DIS_AAM;
-	args.tfRegister[IDE_NSECTOR_OFFSET]	= arg;
-	args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_SETFEATURES;
+	args.tf.feature = arg ? SETFEATURES_EN_AAM : SETFEATURES_DIS_AAM;
+	args.tf.nsect   = arg;
+	args.tf.command = WIN_SETFEATURES;
 	args.command_type = IDE_DRIVE_TASK_NO_DATA;
 	args.handler	  = &task_no_data_intr;
 	ide_raw_taskfile(drive, &args, NULL);
@@ -1057,7 +1051,7 @@ static int idedisk_open(struct inode *inode, struct file *filp)
 	if (drive->removable && idkp->openers == 1) {
 		ide_task_t args;
 		memset(&args, 0, sizeof(ide_task_t));
-		args.tfRegister[IDE_COMMAND_OFFSET] = WIN_DOORLOCK;
+		args.tf.command = WIN_DOORLOCK;
 		args.command_type = IDE_DRIVE_TASK_NO_DATA;
 		args.handler	  = &task_no_data_intr;
 		check_disk_change(inode->i_bdev);
@@ -1084,7 +1078,7 @@ static int idedisk_release(struct inode *inode, struct file *filp)
 	if (drive->removable && idkp->openers == 1) {
 		ide_task_t args;
 		memset(&args, 0, sizeof(ide_task_t));
-		args.tfRegister[IDE_COMMAND_OFFSET] = WIN_DOORUNLOCK;
+		args.tf.command = WIN_DOORUNLOCK;
 		args.command_type = IDE_DRIVE_TASK_NO_DATA;
 		args.handler	  = &task_no_data_intr;
 		if (drive->doorlocking && ide_raw_taskfile(drive, &args, NULL))
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index bb1b0a884cfc..dc984c5b0d13 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -189,15 +189,15 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 			return ide_stopped;
 		}
 		if (ide_id_has_flush_cache_ext(drive->id))
-			args->tfRegister[IDE_COMMAND_OFFSET] = WIN_FLUSH_CACHE_EXT;
+			args->tf.command = WIN_FLUSH_CACHE_EXT;
 		else
-			args->tfRegister[IDE_COMMAND_OFFSET] = WIN_FLUSH_CACHE;
+			args->tf.command = WIN_FLUSH_CACHE;
 		args->command_type = IDE_DRIVE_TASK_NO_DATA;
 		args->handler	   = &task_no_data_intr;
 		return do_rw_taskfile(drive, args);
 
 	case idedisk_pm_standby:	/* Suspend step 2 (standby) */
-		args->tfRegister[IDE_COMMAND_OFFSET] = WIN_STANDBYNOW1;
+		args->tf.command = WIN_STANDBYNOW1;
 		args->command_type = IDE_DRIVE_TASK_NO_DATA;
 		args->handler	   = &task_no_data_intr;
 		return do_rw_taskfile(drive, args);
@@ -214,7 +214,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 		return ide_stopped;
 
 	case idedisk_pm_idle:		/* Resume step 2 (idle) */
-		args->tfRegister[IDE_COMMAND_OFFSET] = WIN_IDLEIMMEDIATE;
+		args->tf.command = WIN_IDLEIMMEDIATE;
 		args->command_type = IDE_DRIVE_TASK_NO_DATA;
 		args->handler = task_no_data_intr;
 		return do_rw_taskfile(drive, args);
@@ -354,28 +354,31 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
 			
 		if (args) {
+			struct ide_taskfile *tf = &args->tf;
+
 			if (args->tf_in_flags.b.data) {
-				u16 data				= hwif->INW(IDE_DATA_REG);
-				args->tfRegister[IDE_DATA_OFFSET]	= (data) & 0xFF;
-				args->hobRegister[IDE_DATA_OFFSET]	= (data >> 8) & 0xFF;
+				u16 data = hwif->INW(IDE_DATA_REG);
+
+				tf->data = data & 0xff;
+				tf->hob_data = (data >> 8) & 0xff;
 			}
-			args->tfRegister[IDE_ERROR_OFFSET]   = err;
+			tf->error = err;
 			/* be sure we're looking at the low order bits */
 			hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
-			args->tfRegister[IDE_NSECTOR_OFFSET] = hwif->INB(IDE_NSECTOR_REG);
-			args->tfRegister[IDE_SECTOR_OFFSET]  = hwif->INB(IDE_SECTOR_REG);
-			args->tfRegister[IDE_LCYL_OFFSET]    = hwif->INB(IDE_LCYL_REG);
-			args->tfRegister[IDE_HCYL_OFFSET]    = hwif->INB(IDE_HCYL_REG);
-			args->tfRegister[IDE_SELECT_OFFSET]  = hwif->INB(IDE_SELECT_REG);
-			args->tfRegister[IDE_STATUS_OFFSET]  = stat;
+			tf->nsect  = hwif->INB(IDE_NSECTOR_REG);
+			tf->lbal   = hwif->INB(IDE_SECTOR_REG);
+			tf->lbam   = hwif->INB(IDE_LCYL_REG);
+			tf->lbah   = hwif->INB(IDE_HCYL_REG);
+			tf->device = hwif->INB(IDE_SELECT_REG);
+			tf->status = stat;
 
 			if (drive->addressing == 1) {
 				hwif->OUTB(drive->ctl|0x80, IDE_CONTROL_REG);
-				args->hobRegister[IDE_FEATURE_OFFSET]	= hwif->INB(IDE_FEATURE_REG);
-				args->hobRegister[IDE_NSECTOR_OFFSET]	= hwif->INB(IDE_NSECTOR_REG);
-				args->hobRegister[IDE_SECTOR_OFFSET]	= hwif->INB(IDE_SECTOR_REG);
-				args->hobRegister[IDE_LCYL_OFFSET]	= hwif->INB(IDE_LCYL_REG);
-				args->hobRegister[IDE_HCYL_OFFSET]	= hwif->INB(IDE_HCYL_REG);
+				tf->hob_feature = hwif->INB(IDE_FEATURE_REG);
+				tf->hob_nsect   = hwif->INB(IDE_NSECTOR_REG);
+				tf->hob_lbal    = hwif->INB(IDE_SECTOR_REG);
+				tf->hob_lbam    = hwif->INB(IDE_LCYL_REG);
+				tf->hob_lbah    = hwif->INB(IDE_HCYL_REG);
 			}
 		}
 	} else if (blk_pm_request(rq)) {
@@ -675,28 +678,28 @@ static ide_startstop_t drive_cmd_intr (ide_drive_t *drive)
 
 static void ide_init_specify_cmd(ide_drive_t *drive, ide_task_t *task)
 {
-	task->tfRegister[IDE_NSECTOR_OFFSET] = drive->sect;
-	task->tfRegister[IDE_SECTOR_OFFSET]  = drive->sect;
-	task->tfRegister[IDE_LCYL_OFFSET]    = drive->cyl;
-	task->tfRegister[IDE_HCYL_OFFSET]    = drive->cyl>>8;
-	task->tfRegister[IDE_SELECT_OFFSET]  = ((drive->head-1)|drive->select.all)&0xBF;
-	task->tfRegister[IDE_COMMAND_OFFSET] = WIN_SPECIFY;
+	task->tf.nsect   = drive->sect;
+	task->tf.lbal    = drive->sect;
+	task->tf.lbam    = drive->cyl;
+	task->tf.lbah    = drive->cyl >> 8;
+	task->tf.device  = ((drive->head - 1) | drive->select.all) & ~ATA_LBA;
+	task->tf.command = WIN_SPECIFY;
 
 	task->handler = &set_geometry_intr;
 }
 
 static void ide_init_restore_cmd(ide_drive_t *drive, ide_task_t *task)
 {
-	task->tfRegister[IDE_NSECTOR_OFFSET] = drive->sect;
-	task->tfRegister[IDE_COMMAND_OFFSET] = WIN_RESTORE;
+	task->tf.nsect   = drive->sect;
+	task->tf.command = WIN_RESTORE;
 
 	task->handler = &recal_intr;
 }
 
 static void ide_init_setmult_cmd(ide_drive_t *drive, ide_task_t *task)
 {
-	task->tfRegister[IDE_NSECTOR_OFFSET] = drive->mult_req;
-	task->tfRegister[IDE_COMMAND_OFFSET] = WIN_SETMULT;
+	task->tf.nsect   = drive->mult_req;
+	task->tf.command = WIN_SETMULT;
 
 	task->handler = &set_multmode_intr;
 }
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index f4e8a0cf06e2..4aac1cc7101d 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -642,9 +642,9 @@ no_80w:
 
 int ide_ata66_check (ide_drive_t *drive, ide_task_t *args)
 {
-	if ((args->tfRegister[IDE_COMMAND_OFFSET] == WIN_SETFEATURES) &&
-	    (args->tfRegister[IDE_SECTOR_OFFSET] > XFER_UDMA_2) &&
-	    (args->tfRegister[IDE_FEATURE_OFFSET] == SETFEATURES_XFER)) {
+	if (args->tf.command == WIN_SETFEATURES &&
+	    args->tf.lbal > XFER_UDMA_2 &&
+	    args->tf.feature == SETFEATURES_XFER) {
 		if (eighty_ninty_three(drive) == 0) {
 			printk(KERN_WARNING "%s: UDMA speeds >UDMA33 cannot "
 					    "be set\n", drive->name);
@@ -662,9 +662,9 @@ int ide_ata66_check (ide_drive_t *drive, ide_task_t *args)
  */
 int set_transfer (ide_drive_t *drive, ide_task_t *args)
 {
-	if ((args->tfRegister[IDE_COMMAND_OFFSET] == WIN_SETFEATURES) &&
-	    (args->tfRegister[IDE_SECTOR_OFFSET] >= XFER_SW_DMA_0) &&
-	    (args->tfRegister[IDE_FEATURE_OFFSET] == SETFEATURES_XFER) &&
+	if (args->tf.command == WIN_SETFEATURES &&
+	    args->tf.lbal >= XFER_SW_DMA_0 &&
+	    args->tf.feature == SETFEATURES_XFER &&
 	    (drive->id->dma_ultra ||
 	     drive->id->dma_mword ||
 	     drive->id->dma_1word))
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 062d3bcb2471..6dbf2af0d215 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -468,8 +468,7 @@ static void ide_dump_opcode(ide_drive_t *drive)
 	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *args = rq->special;
 		if (args) {
-			task_struct_t *tf = (task_struct_t *) args->tfRegister;
-			opcode = tf->command;
+			opcode = args->tf.command;
 			found = 1;
 		}
 	}
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 91948a46cc6e..5a1a9f7846b6 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -66,12 +66,13 @@ static void taskfile_output_data(ide_drive_t *drive, void *buffer, u32 wcount)
 int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 {
 	ide_task_t args;
+
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tfRegister[IDE_NSECTOR_OFFSET]	= 0x01;
+	args.tf.nsect = 0x01;
 	if (drive->media == ide_disk)
-		args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_IDENTIFY;
+		args.tf.command = WIN_IDENTIFY;
 	else
-		args.tfRegister[IDE_COMMAND_OFFSET]	= WIN_PIDENTIFY;
+		args.tf.command = WIN_PIDENTIFY;
 	args.command_type = IDE_DRIVE_TASK_IN;
 	args.data_phase   = TASKFILE_IN;
 	args.handler	  = &task_in_intr;
@@ -81,8 +82,7 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	task_struct_t *taskfile	= (task_struct_t *) task->tfRegister;
-	hob_struct_t *hobfile	= (hob_struct_t *) task->hobRegister;
+	struct ide_taskfile *tf = &task->tf;
 	u8 HIHI			= (drive->addressing == 1) ? 0xE0 : 0xEF;
 
 	/* ALL Command Block Executions SHALL clear nIEN, unless otherwise */
@@ -93,35 +93,35 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	SELECT_MASK(drive, 0);
 
 	if (drive->addressing == 1) {
-		hwif->OUTB(hobfile->feature, IDE_FEATURE_REG);
-		hwif->OUTB(hobfile->sector_count, IDE_NSECTOR_REG);
-		hwif->OUTB(hobfile->sector_number, IDE_SECTOR_REG);
-		hwif->OUTB(hobfile->low_cylinder, IDE_LCYL_REG);
-		hwif->OUTB(hobfile->high_cylinder, IDE_HCYL_REG);
+		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
+		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
+		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
+		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
 	}
 
-	hwif->OUTB(taskfile->feature, IDE_FEATURE_REG);
-	hwif->OUTB(taskfile->sector_count, IDE_NSECTOR_REG);
-	hwif->OUTB(taskfile->sector_number, IDE_SECTOR_REG);
-	hwif->OUTB(taskfile->low_cylinder, IDE_LCYL_REG);
-	hwif->OUTB(taskfile->high_cylinder, IDE_HCYL_REG);
+	hwif->OUTB(tf->feature, IDE_FEATURE_REG);
+	hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
+	hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
+	hwif->OUTB(tf->lbam, IDE_LCYL_REG);
+	hwif->OUTB(tf->lbah, IDE_HCYL_REG);
 
-	hwif->OUTB((taskfile->device_head & HIHI) | drive->select.all, IDE_SELECT_REG);
+	hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
 
 	if (task->handler != NULL) {
 		if (task->prehandler != NULL) {
-			hwif->OUTBSYNC(drive, taskfile->command, IDE_COMMAND_REG);
+			hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
 			ndelay(400);	/* FIXME */
 			return task->prehandler(drive, task->rq);
 		}
-		ide_execute_command(drive, taskfile->command, task->handler, WAIT_WORSTCASE, NULL);
+		ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
 		return ide_started;
 	}
 
 	if (!drive->using_dma)
 		return ide_stopped;
 
-	switch (taskfile->command) {
+	switch (tf->command) {
 		case WIN_WRITEDMA_ONCE:
 		case WIN_WRITEDMA:
 		case WIN_WRITEDMA_EXT:
@@ -130,7 +130,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 		case WIN_READDMA_EXT:
 		case WIN_IDENTIFY_DMA:
 			if (!hwif->dma_setup(drive)) {
-				hwif->dma_exec_cmd(drive, taskfile->command);
+				hwif->dma_exec_cmd(drive, tf->command);
 				hwif->dma_start(drive);
 				return ide_started;
 			}
@@ -483,7 +483,7 @@ static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long
 	 */
 	if (args->command_type != IDE_DRIVE_TASK_NO_DATA) {
 		if (data_size == 0)
-			rq.nr_sectors = (args->hobRegister[IDE_NSECTOR_OFFSET] << 8) | args->tfRegister[IDE_NSECTOR_OFFSET];
+			rq.nr_sectors = (args->tf.hob_nsect << 8) | args->tf.nsect;
 		else
 			rq.nr_sectors = data_size / SECTOR_SIZE;
 
@@ -519,8 +519,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	ide_task_t		args;
 	u8 *outbuf		= NULL;
 	u8 *inbuf		= NULL;
-	u8 *argsptr		= args.tfRegister;
-	u8 *hobsptr		= args.hobRegister;
 	int err			= 0;
 	int tasksize		= sizeof(struct ide_task_request_s);
 	unsigned int taskin	= 0;
@@ -572,9 +570,9 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	}
 
 	memset(&args, 0, sizeof(ide_task_t));
-	memcpy(argsptr, req_task->io_ports, HDIO_DRIVE_TASK_HDR_SIZE);
-	memcpy(hobsptr, req_task->hob_ports, HDIO_DRIVE_HOB_HDR_SIZE);
 
+	memcpy(&args.tf_array[0], req_task->hob_ports, HDIO_DRIVE_HOB_HDR_SIZE - 2);
+	memcpy(&args.tf_array[6], req_task->io_ports, HDIO_DRIVE_TASK_HDR_SIZE);
 	args.tf_in_flags  = req_task->in_flags;
 	args.tf_out_flags = req_task->out_flags;
 	args.data_phase   = req_task->data_phase;
@@ -628,8 +626,8 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			goto abort;
 	}
 
-	memcpy(req_task->io_ports, &(args.tfRegister), HDIO_DRIVE_TASK_HDR_SIZE);
-	memcpy(req_task->hob_ports, &(args.hobRegister), HDIO_DRIVE_HOB_HDR_SIZE);
+	memcpy(req_task->hob_ports, &args.tf_array[0], HDIO_DRIVE_HOB_HDR_SIZE - 2);
+	memcpy(req_task->io_ports, &args.tf_array[6], HDIO_DRIVE_TASK_HDR_SIZE);
 	req_task->in_flags  = args.tf_in_flags;
 	req_task->out_flags = args.tf_out_flags;
 
@@ -688,6 +686,7 @@ int ide_cmd_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	u8 xfer_rate = 0;
 	int argsize = 4;
 	ide_task_t tfargs;
+	struct ide_taskfile *tf = &tfargs.tf;
 
 	if (NULL == (void *) arg) {
 		struct request rq;
@@ -699,13 +698,10 @@ int ide_cmd_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 		return -EFAULT;
 
 	memset(&tfargs, 0, sizeof(ide_task_t));
-	tfargs.tfRegister[IDE_FEATURE_OFFSET] = args[2];
-	tfargs.tfRegister[IDE_NSECTOR_OFFSET] = args[3];
-	tfargs.tfRegister[IDE_SECTOR_OFFSET]  = args[1];
-	tfargs.tfRegister[IDE_LCYL_OFFSET]    = 0x00;
-	tfargs.tfRegister[IDE_HCYL_OFFSET]    = 0x00;
-	tfargs.tfRegister[IDE_SELECT_OFFSET]  = 0x00;
-	tfargs.tfRegister[IDE_COMMAND_OFFSET] = args[0];
+	tf->feature = args[2];
+	tf->nsect   = args[3];
+	tf->lbal    = args[1];
+	tf->command = args[0];
 
 	if (args[3]) {
 		argsize = 4 + (SECTOR_WORDS * 4 * args[3]);
@@ -767,8 +763,7 @@ int ide_task_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	task_struct_t *taskfile	= (task_struct_t *) task->tfRegister;
-	hob_struct_t *hobfile	= (hob_struct_t *) task->hobRegister;
+	struct ide_taskfile *tf = &task->tf;
 
 	if (task->data_phase == TASKFILE_MULTI_IN ||
 	    task->data_phase == TASKFILE_MULTI_OUT) {
@@ -798,34 +793,32 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
 	SELECT_MASK(drive, 0);
 
-	if (task->tf_out_flags.b.data) {
-		u16 data =  taskfile->data + (hobfile->data << 8);
-		hwif->OUTW(data, IDE_DATA_REG);
-	}
+	if (task->tf_out_flags.b.data)
+		hwif->OUTW((tf->hob_data << 8) | tf->data, IDE_DATA_REG);
 
 	/* (ks) send hob registers first */
 	if (task->tf_out_flags.b.nsector_hob)
-		hwif->OUTB(hobfile->sector_count, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
 	if (task->tf_out_flags.b.sector_hob)
-		hwif->OUTB(hobfile->sector_number, IDE_SECTOR_REG);
+		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
 	if (task->tf_out_flags.b.lcyl_hob)
-		hwif->OUTB(hobfile->low_cylinder, IDE_LCYL_REG);
+		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
 	if (task->tf_out_flags.b.hcyl_hob)
-		hwif->OUTB(hobfile->high_cylinder, IDE_HCYL_REG);
+		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
 
 	/* (ks) Send now the standard registers */
 	if (task->tf_out_flags.b.error_feature)
-		hwif->OUTB(taskfile->feature, IDE_FEATURE_REG);
+		hwif->OUTB(tf->feature, IDE_FEATURE_REG);
 	/* refers to number of sectors to transfer */
 	if (task->tf_out_flags.b.nsector)
-		hwif->OUTB(taskfile->sector_count, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
 	/* refers to sector offset or start sector */
 	if (task->tf_out_flags.b.sector)
-		hwif->OUTB(taskfile->sector_number, IDE_SECTOR_REG);
+		hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
 	if (task->tf_out_flags.b.lcyl)
-		hwif->OUTB(taskfile->low_cylinder, IDE_LCYL_REG);
+		hwif->OUTB(tf->lbam, IDE_LCYL_REG);
 	if (task->tf_out_flags.b.hcyl)
-		hwif->OUTB(taskfile->high_cylinder, IDE_HCYL_REG);
+		hwif->OUTB(tf->lbah, IDE_HCYL_REG);
 
         /*
 	 * (ks) In the flagged taskfile approch, we will use all specified
@@ -833,7 +826,7 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 	 * select bit (master/slave) in the drive_head register. We must make
 	 * sure that the desired drive is selected.
 	 */
-	hwif->OUTB(taskfile->device_head | drive->select.all, IDE_SELECT_REG);
+	hwif->OUTB(tf->device | drive->select.all, IDE_SELECT_REG);
 	switch(task->data_phase) {
 
    	        case TASKFILE_OUT_DMAQ:
@@ -844,7 +837,7 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 				break;
 
 			if (!hwif->dma_setup(drive)) {
-				hwif->dma_exec_cmd(drive, taskfile->command);
+				hwif->dma_exec_cmd(drive, tf->command);
 				hwif->dma_start(drive);
 				return ide_started;
 			}
@@ -856,11 +849,11 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 
 			/* Issue the command */
 			if (task->prehandler) {
-				hwif->OUTBSYNC(drive, taskfile->command, IDE_COMMAND_REG);
+				hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
 				ndelay(400);	/* FIXME */
 				return task->prehandler(drive, task->rq);
 			}
-			ide_execute_command(drive, taskfile->command, task->handler, WAIT_WORSTCASE, NULL);
+			ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
 			return ide_started;
 	}
 
diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h
index 3bcb88560418..df17bf767d98 100644
--- a/include/linux/hdreg.h
+++ b/include/linux/hdreg.h
@@ -117,7 +117,7 @@ typedef union ide_reg_valid_s {
 
 typedef struct ide_task_request_s {
 	__u8		io_ports[8];
-	__u8		hob_ports[8];
+	__u8		hob_ports[8]; /* bytes 6 and 7 are unused */
 	ide_reg_valid_t	out_flags;
 	ide_reg_valid_t	in_flags;
 	int		data_phase;
@@ -139,6 +139,7 @@ struct hd_drive_cmd_hdr {
 	__u8 sector_count;
 };
 
+#ifndef __KERNEL__
 typedef struct hd_drive_task_hdr {
 	__u8 data;
 	__u8 feature;
@@ -160,6 +161,7 @@ typedef struct hd_drive_hob_hdr {
 	__u8 device_head;
 	__u8 control;
 } hob_struct_t;
+#endif
 
 #define TASKFILE_INVALID		0x7fff
 #define TASKFILE_48			0x8000
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 66a38f101175..69b78bb39caf 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -103,8 +103,6 @@ typedef unsigned char	byte;	/* used everywhere */
 #define IDE_FEATURE_OFFSET	IDE_ERROR_OFFSET
 #define IDE_COMMAND_OFFSET	IDE_STATUS_OFFSET
 
-#define IDE_CONTROL_OFFSET_HOB	(7)
-
 #define IDE_DATA_REG		(HWIF(drive)->io_ports[IDE_DATA_OFFSET])
 #define IDE_ERROR_REG		(HWIF(drive)->io_ports[IDE_ERROR_OFFSET])
 #define IDE_NSECTOR_REG		(HWIF(drive)->io_ports[IDE_NSECTOR_OFFSET])
@@ -1062,15 +1060,40 @@ extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
  */
 extern int ide_wait_cmd(ide_drive_t *, u8, u8, u8, u8, u8 *);
 
+struct ide_taskfile {
+	u8	hob_data;	/*  0: high data byte (for TASKFILE IOCTL) */
+
+	u8	hob_feature;	/*  1-5: additional data to support LBA48 */
+	u8	hob_nsect;
+	u8	hob_lbal;
+	u8	hob_lbam;
+	u8	hob_lbah;
+
+	u8	data;		/*  6: low data byte (for TASKFILE IOCTL) */
+
+	union {			/*  7: */
+		u8 error;	/*   read:  error */
+		u8 feature;	/*  write: feature */
+	};
+
+	u8	nsect;		/*  8: number of sectors */
+	u8	lbal;		/*  9: LBA low */
+	u8	lbam;		/* 10: LBA mid */
+	u8	lbah;		/* 11: LBA high */
+
+	u8	device;		/* 12: device select */
+
+	union {			/* 13: */
+		u8 status;	/*  read: status  */
+		u8 command;	/* write: command */
+	};
+};
+
 typedef struct ide_task_s {
-/*
- *	struct hd_drive_task_hdr	tf;
- *	task_struct_t		tf;
- *	struct hd_drive_hob_hdr		hobf;
- *	hob_struct_t		hobf;
- */
-	u8			tfRegister[8];
-	u8			hobRegister[8];
+	union {
+		struct ide_taskfile	tf;
+		u8			tf_array[14];
+	};
 	ide_reg_valid_t		tf_out_flags;
 	ide_reg_valid_t		tf_in_flags;
 	int			data_phase;
-- 
cgit v1.2.3


From 9e42237f26cf517a3f682505f03a3a8d89b3b35d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:07 +0100
Subject: ide: add ide_tf_load() helper

Based on the earlier work by Tejun Heo.

* Add 'tf_flags' field (for taskfile flags) to ide_task_t.

* Add IDE_TFLAG_LBA48 taskfile flag for LBA48 taskfiles.

* Add IDE_TFLAG_NO_SELECT_MASK taskfile flag for __ide_do_rw_disk()
  which doesn't use SELECT_MASK() (looks like a bug but it requires
  some more investigation).

* Split off ide_tf_load() helper from do_rw_taskfile().

* Convert __ide_do_rw_disk() to use ide_tf_load().

There should be no functionality changes caused by this patch.

Cc: Tejun Heo <htejun@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     | 28 +++++------------------
 drivers/ide/ide-taskfile.c | 56 ++++++++++++++++++++++++++++------------------
 include/linux/ide.h        |  8 +++++++
 3 files changed, 48 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 20d357e9329d..6387222dd200 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -140,7 +140,8 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 	u8 lba48		= (drive->addressing == 1) ? 1 : 0;
 	u8 command		= WIN_NOP;
 	ata_nsector_t		nsectors;
-	struct ide_taskfile ltf, *tf = &ltf;
+	ide_task_t		task;
+	struct ide_taskfile	*tf = &task.tf;
 
 	nsectors.all		= (u16) rq->nr_sectors;
 
@@ -156,12 +157,8 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 		ide_map_sg(drive, rq);
 	}
 
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
-
-	/* FIXME: SELECT_MASK(drive, 0) ? */
-
-	memset(tf, 0, sizeof(*tf));
+	memset(&task, 0, sizeof(task));
+	task.tf_flags = IDE_TFLAG_NO_SELECT_MASK;  /* FIXME? */
 
 	if (drive->select.b.lba) {
 		if (lba48) {
@@ -185,6 +182,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 				tf->hob_lbah, tf->hob_lbam, tf->hob_lbal,
 				tf->lbah, tf->lbam, tf->lbal);
 #endif
+			task.tf_flags |= IDE_TFLAG_LBA48;
 		} else {
 			tf->nsect  = nsectors.b.low;
 			tf->lbal   = block;
@@ -208,21 +206,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 		tf->device = head;
 	}
 
-	if (drive->select.b.lba && lba48) {
-		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
-		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
-		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
-		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
-		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
-	}
-
-	hwif->OUTB(tf->feature, IDE_FEATURE_REG);
-	hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
-	hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
-	hwif->OUTB(tf->lbam, IDE_LCYL_REG);
-	hwif->OUTB(tf->lbah, IDE_HCYL_REG);
-
-	hwif->OUTB(tf->device | drive->select.all, IDE_SELECT_REG);
+	ide_tf_load(drive, &task);
 
 	if (dma) {
 		if (!hwif->dma_setup(drive)) {
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 5a1a9f7846b6..a79150e6be07 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -63,6 +63,37 @@ static void taskfile_output_data(ide_drive_t *drive, void *buffer, u32 wcount)
 	}
 }
 
+void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct ide_taskfile *tf = &task->tf;
+	u8 HIHI = (task->tf_flags & IDE_TFLAG_LBA48) ? 0xE0 : 0xEF;
+
+	if (IDE_CONTROL_REG)
+		hwif->OUTB(drive->ctl, IDE_CONTROL_REG); /* clear nIEN */
+
+	if ((task->tf_flags & IDE_TFLAG_NO_SELECT_MASK) == 0)
+		SELECT_MASK(drive, 0);
+
+	if (task->tf_flags & IDE_TFLAG_LBA48) {
+		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
+		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
+		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
+		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
+	}
+
+	hwif->OUTB(tf->feature, IDE_FEATURE_REG);
+	hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
+	hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
+	hwif->OUTB(tf->lbam, IDE_LCYL_REG);
+	hwif->OUTB(tf->lbah, IDE_HCYL_REG);
+
+	hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
+}
+
+EXPORT_SYMBOL_GPL(ide_tf_load);
+
 int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 {
 	ide_task_t args;
@@ -83,30 +114,11 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct ide_taskfile *tf = &task->tf;
-	u8 HIHI			= (drive->addressing == 1) ? 0xE0 : 0xEF;
 
-	/* ALL Command Block Executions SHALL clear nIEN, unless otherwise */
-	if (IDE_CONTROL_REG) {
-		/* clear nIEN */
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
-	}
-	SELECT_MASK(drive, 0);
+	if (drive->addressing == 1)
+		task->tf_flags |= IDE_TFLAG_LBA48;
 
-	if (drive->addressing == 1) {
-		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
-		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
-		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
-		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
-		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
-	}
-
-	hwif->OUTB(tf->feature, IDE_FEATURE_REG);
-	hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
-	hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
-	hwif->OUTB(tf->lbam, IDE_LCYL_REG);
-	hwif->OUTB(tf->lbah, IDE_HCYL_REG);
-
-	hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
+	ide_tf_load(drive, task);
 
 	if (task->handler != NULL) {
 		if (task->prehandler != NULL) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 69b78bb39caf..e25fd0b1dd79 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1060,6 +1060,11 @@ extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
  */
 extern int ide_wait_cmd(ide_drive_t *, u8, u8, u8, u8, u8 *);
 
+enum {
+	IDE_TFLAG_LBA48			= (1 << 0),
+	IDE_TFLAG_NO_SELECT_MASK	= (1 << 1),
+};
+
 struct ide_taskfile {
 	u8	hob_data;	/*  0: high data byte (for TASKFILE IOCTL) */
 
@@ -1094,6 +1099,7 @@ typedef struct ide_task_s {
 		struct ide_taskfile	tf;
 		u8			tf_array[14];
 	};
+	u8			tf_flags;
 	ide_reg_valid_t		tf_out_flags;
 	ide_reg_valid_t		tf_in_flags;
 	int			data_phase;
@@ -1104,6 +1110,8 @@ typedef struct ide_task_s {
 	void			*special;	/* valid_t generally */
 } ide_task_t;
 
+void ide_tf_load(ide_drive_t *, ide_task_t *);
+
 extern u32 ide_read_24(ide_drive_t *);
 
 extern void SELECT_DRIVE(ide_drive_t *);
-- 
cgit v1.2.3


From 9a3c49be5c5f7388eefb712be9a383904140532e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:07 +0100
Subject: ide: add ide_no_data_taskfile() helper

* Add ide_no_data_taskfile() helper and convert ide_raw_taskfile() w/ NO DATA
  protocol users to use it instead.

* Set ->data_phase explicitly in ide_no_data_taskfile()
  (TASKFILE_NO_DATA is defined as 0x0000).

* Unexport task_no_data_intr().

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c     |  7 ++-----
 drivers/ide/ide-disk.c     | 32 ++++++++------------------------
 drivers/ide/ide-taskfile.c | 12 ++++++++++--
 include/linux/ide.h        |  2 ++
 4 files changed, 22 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 747c51889f7d..f0a6a3d6d2d3 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -383,9 +383,6 @@ static int taskfile_load_raw(ide_drive_t *drive,
 	       gtf->tfa[3], gtf->tfa[4], gtf->tfa[5], gtf->tfa[6]);
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.command_type = IDE_DRIVE_TASK_NO_DATA;
-	args.data_phase   = TASKFILE_NO_DATA;
-	args.handler      = &task_no_data_intr;
 
 	/* convert gtf to IDE Taskfile */
 	memcpy(&args.tf_array[7], &gtf->tfa, 7);
@@ -395,9 +392,9 @@ static int taskfile_load_raw(ide_drive_t *drive,
 		return err;
 	}
 
-	err = ide_raw_taskfile(drive, &args, NULL);
+	err = ide_no_data_taskfile(drive, &args);
 	if (err)
-		printk(KERN_ERR "%s: ide_raw_taskfile failed: %u\n",
+		printk(KERN_ERR "%s: ide_no_data_taskfile failed: %u\n",
 		       __FUNCTION__, err);
 
 	return err;
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 6387222dd200..b534fe97d476 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -303,10 +303,8 @@ static u64 idedisk_read_native_max_address(ide_drive_t *drive, int lba48)
 	else
 		tf->command = WIN_READ_NATIVE_MAX;
 	tf->device  = ATA_LBA;
-	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
-	args.handler				= &task_no_data_intr;
 	/* submit command request */
-	ide_raw_taskfile(drive, &args, NULL);
+	ide_no_data_taskfile(drive, &args);
 
 	/* if OK, compute maximum address value */
 	if ((tf->status & 0x01) == 0) {
@@ -350,10 +348,8 @@ static u64 idedisk_set_max_address(ide_drive_t *drive, u64 addr_req, int lba48)
 		tf->command  = WIN_SET_MAX;
 	}
 	tf->device |= ATA_LBA;
-	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
-	args.handler				= &task_no_data_intr;
 	/* submit command request */
-	ide_raw_taskfile(drive, &args, NULL);
+	ide_no_data_taskfile(drive, &args);
 	/* if OK, compute maximum address value */
 	if ((tf->status & 0x01) == 0) {
 		u32 high, low;
@@ -500,9 +496,7 @@ static int smart_enable(ide_drive_t *drive)
 	tf->lbam    = SMART_LCYL_PASS;
 	tf->lbah    = SMART_HCYL_PASS;
 	tf->command = WIN_SMART;
-	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
-	args.handler				= &task_no_data_intr;
-	return ide_raw_taskfile(drive, &args, NULL);
+	return ide_no_data_taskfile(drive, &args);
 }
 
 static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
@@ -695,9 +689,7 @@ static int write_cache(ide_drive_t *drive, int arg)
 		args.tf.feature = arg ?
 			SETFEATURES_EN_WCACHE : SETFEATURES_DIS_WCACHE;
 		args.tf.command = WIN_SETFEATURES;
-		args.command_type		= IDE_DRIVE_TASK_NO_DATA;
-		args.handler			= &task_no_data_intr;
-		err = ide_raw_taskfile(drive, &args, NULL);
+		err = ide_no_data_taskfile(drive, &args);
 		if (err == 0)
 			drive->wcache = arg;
 	}
@@ -716,9 +708,7 @@ static int do_idedisk_flushcache (ide_drive_t *drive)
 		args.tf.command = WIN_FLUSH_CACHE_EXT;
 	else
 		args.tf.command = WIN_FLUSH_CACHE;
-	args.command_type			= IDE_DRIVE_TASK_NO_DATA;
-	args.handler				= &task_no_data_intr;
-	return ide_raw_taskfile(drive, &args, NULL);
+	return ide_no_data_taskfile(drive, &args);
 }
 
 static int set_acoustic (ide_drive_t *drive, int arg)
@@ -732,9 +722,7 @@ static int set_acoustic (ide_drive_t *drive, int arg)
 	args.tf.feature = arg ? SETFEATURES_EN_AAM : SETFEATURES_DIS_AAM;
 	args.tf.nsect   = arg;
 	args.tf.command = WIN_SETFEATURES;
-	args.command_type = IDE_DRIVE_TASK_NO_DATA;
-	args.handler	  = &task_no_data_intr;
-	ide_raw_taskfile(drive, &args, NULL);
+	ide_no_data_taskfile(drive, &args);
 	drive->acoustic = arg;
 	return 0;
 }
@@ -996,15 +984,13 @@ static int idedisk_open(struct inode *inode, struct file *filp)
 		ide_task_t args;
 		memset(&args, 0, sizeof(ide_task_t));
 		args.tf.command = WIN_DOORLOCK;
-		args.command_type = IDE_DRIVE_TASK_NO_DATA;
-		args.handler	  = &task_no_data_intr;
 		check_disk_change(inode->i_bdev);
 		/*
 		 * Ignore the return code from door_lock,
 		 * since the open() has already succeeded,
 		 * and the door_lock is irrelevant at this point.
 		 */
-		if (drive->doorlocking && ide_raw_taskfile(drive, &args, NULL))
+		if (drive->doorlocking && ide_no_data_taskfile(drive, &args))
 			drive->doorlocking = 0;
 	}
 	return 0;
@@ -1023,9 +1009,7 @@ static int idedisk_release(struct inode *inode, struct file *filp)
 		ide_task_t args;
 		memset(&args, 0, sizeof(ide_task_t));
 		args.tf.command = WIN_DOORUNLOCK;
-		args.command_type = IDE_DRIVE_TASK_NO_DATA;
-		args.handler	  = &task_no_data_intr;
-		if (drive->doorlocking && ide_raw_taskfile(drive, &args, NULL))
+		if (drive->doorlocking && ide_no_data_taskfile(drive, &args))
 			drive->doorlocking = 0;
 	}
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index a79150e6be07..7cb674f81315 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -229,8 +229,6 @@ ide_startstop_t task_no_data_intr (ide_drive_t *drive)
 	return ide_stopped;
 }
 
-EXPORT_SYMBOL(task_no_data_intr);
-
 static u8 wait_drive_not_busy(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -524,6 +522,16 @@ int ide_raw_taskfile (ide_drive_t *drive, ide_task_t *args, u8 *buf)
 
 EXPORT_SYMBOL(ide_raw_taskfile);
 
+int ide_no_data_taskfile(ide_drive_t *drive, ide_task_t *task)
+{
+	task->command_type = IDE_DRIVE_TASK_NO_DATA;
+	task->data_phase   = TASKFILE_NO_DATA;
+	task->handler      = task_no_data_intr;
+
+	return ide_raw_taskfile(drive, task, NULL);
+}
+EXPORT_SYMBOL_GPL(ide_no_data_taskfile);
+
 #ifdef CONFIG_IDE_TASK_IOCTL
 int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e25fd0b1dd79..11bfbc40c50a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1140,6 +1140,8 @@ extern ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
 
 extern int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *);
 
+int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
+
 int ide_taskfile_ioctl(ide_drive_t *, unsigned int, unsigned long);
 int ide_cmd_ioctl(ide_drive_t *, unsigned int, unsigned long);
 int ide_task_ioctl(ide_drive_t *, unsigned int, unsigned long);
-- 
cgit v1.2.3


From 74095a91ed02f6727b62d4416be00a041f2d7436 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:07 +0100
Subject: ide: use do_rw_taskfile() in flagged_taskfile()

Based on the earlier work by Tejun Heo.

* Move setting IDE_TFLAG_LBA48 taskfile flag from do_rw_taskfile()
  function to the callers.

* Add IDE_TFLAG_FLAGGED taskfile flag for flagged taskfiles coming
  from ide_taskfile_ioctl().  Check it instead of ->tf_out_flags.all.

* Add IDE_TFLAG_OUT_DATA taskfile flag to indicate the need to load
  IDE data register in ide_tf_load().

* Add IDE_TFLAG_OUT_* taskfile flags to indicate the need to load
  particular IDE taskfile registers in ide_tf_load().

* Update do_rw_taskfile() and ide_tf_load() users to set respective
  IDE_TFLAG_OUT_* taksfile flags.

* Add task_dma_ok() helper.

* Use IDE_TFLAG_FLAGGED taskfile flag to select HIHI mask in ide_tf_load().

* Use do_rw_taskfile() in flagged_taskfile().

* Remove no longer needed 'tf_out_flags' field from ide_task_t.

There should be no functionality changes caused by this patch.

Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     |   3 +-
 drivers/ide/ide-io.c       |  31 +++++---
 drivers/ide/ide-taskfile.c | 180 ++++++++++++++++++---------------------------
 include/linux/ide.h        |  25 ++++++-
 4 files changed, 119 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index b534fe97d476..6182c23d2020 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -159,6 +159,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_NO_SELECT_MASK;  /* FIXME? */
+	task.tf_flags |= IDE_TFLAG_OUT_TF;
 
 	if (drive->select.b.lba) {
 		if (lba48) {
@@ -182,7 +183,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 				tf->hob_lbah, tf->hob_lbam, tf->hob_lbal,
 				tf->lbah, tf->lbam, tf->lbal);
 #endif
-			task.tf_flags |= IDE_TFLAG_LBA48;
+			task.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
 		} else {
 			tf->nsect  = nsectors.b.low;
 			tf->lbal   = block;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index dc984c5b0d13..33458fe7f490 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -192,15 +192,11 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 			args->tf.command = WIN_FLUSH_CACHE_EXT;
 		else
 			args->tf.command = WIN_FLUSH_CACHE;
-		args->command_type = IDE_DRIVE_TASK_NO_DATA;
-		args->handler	   = &task_no_data_intr;
-		return do_rw_taskfile(drive, args);
+		goto out_do_tf;
 
 	case idedisk_pm_standby:	/* Suspend step 2 (standby) */
 		args->tf.command = WIN_STANDBYNOW1;
-		args->command_type = IDE_DRIVE_TASK_NO_DATA;
-		args->handler	   = &task_no_data_intr;
-		return do_rw_taskfile(drive, args);
+		goto out_do_tf;
 
 	case idedisk_pm_restore_pio:	/* Resume step 1 (restore PIO) */
 		ide_set_max_pio(drive);
@@ -215,9 +211,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 
 	case idedisk_pm_idle:		/* Resume step 2 (idle) */
 		args->tf.command = WIN_IDLEIMMEDIATE;
-		args->command_type = IDE_DRIVE_TASK_NO_DATA;
-		args->handler = task_no_data_intr;
-		return do_rw_taskfile(drive, args);
+		goto out_do_tf;
 
 	case ide_pm_restore_dma:	/* Resume step 3 (restore DMA) */
 		/*
@@ -236,6 +230,14 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 	}
 	pm->pm_step = ide_pm_state_completed;
 	return ide_stopped;
+
+out_do_tf:
+	args->tf_flags = IDE_TFLAG_OUT_TF;
+	if (drive->addressing == 1)
+		args->tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+	args->command_type = IDE_DRIVE_TASK_NO_DATA;
+	args->handler      = task_no_data_intr;
+	return do_rw_taskfile(drive, args);
 }
 
 /**
@@ -730,6 +732,10 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 		return ide_stopped;
 	}
 
+	args.tf_flags = IDE_TFLAG_OUT_TF;
+	if (drive->addressing == 1)
+		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+
 	do_rw_taskfile(drive, &args);
 
 	return ide_started;
@@ -883,8 +889,13 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 			break;
 		}
 
-		if (args->tf_out_flags.all != 0) 
+		if (args->tf_flags & IDE_TFLAG_FLAGGED)
 			return flagged_taskfile(drive, args);
+
+		args->tf_flags |= IDE_TFLAG_OUT_TF;
+		if (drive->addressing == 1)
+			args->tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+
 		return do_rw_taskfile(drive, args);
 	} else if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
 		u8 *args = rq->buffer;
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 7cb674f81315..a7668b459fe1 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -69,25 +69,39 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 	struct ide_taskfile *tf = &task->tf;
 	u8 HIHI = (task->tf_flags & IDE_TFLAG_LBA48) ? 0xE0 : 0xEF;
 
+	if (task->tf_flags & IDE_TFLAG_FLAGGED)
+		HIHI = 0xFF;
+
 	if (IDE_CONTROL_REG)
 		hwif->OUTB(drive->ctl, IDE_CONTROL_REG); /* clear nIEN */
 
 	if ((task->tf_flags & IDE_TFLAG_NO_SELECT_MASK) == 0)
 		SELECT_MASK(drive, 0);
 
-	if (task->tf_flags & IDE_TFLAG_LBA48) {
+	if (task->tf_flags & IDE_TFLAG_OUT_DATA)
+		hwif->OUTW((tf->hob_data << 8) | tf->data, IDE_DATA_REG);
+
+	if (task->tf_flags & IDE_TFLAG_OUT_HOB_FEATURE)
 		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_HOB_NSECT)
 		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAL)
 		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAM)
 		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAH)
 		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
-	}
 
-	hwif->OUTB(tf->feature, IDE_FEATURE_REG);
-	hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
-	hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
-	hwif->OUTB(tf->lbam, IDE_LCYL_REG);
-	hwif->OUTB(tf->lbah, IDE_HCYL_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_FEATURE)
+		hwif->OUTB(tf->feature, IDE_FEATURE_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_NSECT)
+		hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_LBAL)
+		hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_LBAM)
+		hwif->OUTB(tf->lbam, IDE_LCYL_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_LBAH)
+		hwif->OUTB(tf->lbah, IDE_HCYL_REG);
 
 	hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
 }
@@ -110,14 +124,30 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 	return ide_raw_taskfile(drive, &args, buf);
 }
 
+static int inline task_dma_ok(ide_task_t *task)
+{
+	if (task->tf_flags & IDE_TFLAG_FLAGGED)
+		return 1;
+
+	switch (task->tf.command) {
+		case WIN_WRITEDMA_ONCE:
+		case WIN_WRITEDMA:
+		case WIN_WRITEDMA_EXT:
+		case WIN_READDMA_ONCE:
+		case WIN_READDMA:
+		case WIN_READDMA_EXT:
+		case WIN_IDENTIFY_DMA:
+			return 1;
+	}
+
+	return 0;
+}
+
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct ide_taskfile *tf = &task->tf;
 
-	if (drive->addressing == 1)
-		task->tf_flags |= IDE_TFLAG_LBA48;
-
 	ide_tf_load(drive, task);
 
 	if (task->handler != NULL) {
@@ -130,26 +160,10 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 		return ide_started;
 	}
 
-	if (!drive->using_dma)
-		return ide_stopped;
-
-	switch (tf->command) {
-		case WIN_WRITEDMA_ONCE:
-		case WIN_WRITEDMA:
-		case WIN_WRITEDMA_EXT:
-		case WIN_READDMA_ONCE:
-		case WIN_READDMA:
-		case WIN_READDMA_EXT:
-		case WIN_IDENTIFY_DMA:
-			if (!hwif->dma_setup(drive)) {
-				hwif->dma_exec_cmd(drive, tf->command);
-				hwif->dma_start(drive);
-				return ide_started;
-			}
-			break;
-		default:
-			if (task->handler == NULL)
-				return ide_stopped;
+	if (task_dma_ok(task) && drive->using_dma && !hwif->dma_setup(drive)) {
+		hwif->dma_exec_cmd(drive, tf->command);
+		hwif->dma_start(drive);
+		return ide_started;
 	}
 
 	return ide_stopped;
@@ -373,7 +387,7 @@ static void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *task = rq->special;
 
-		if (task->tf_out_flags.all) {
+		if (task->tf_flags & IDE_TFLAG_FLAGGED) {
 			u8 err = drive->hwif->INB(IDE_ERROR_REG);
 			ide_end_drive_cmd(drive, stat, err);
 			return;
@@ -594,10 +608,36 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	memcpy(&args.tf_array[0], req_task->hob_ports, HDIO_DRIVE_HOB_HDR_SIZE - 2);
 	memcpy(&args.tf_array[6], req_task->io_ports, HDIO_DRIVE_TASK_HDR_SIZE);
 	args.tf_in_flags  = req_task->in_flags;
-	args.tf_out_flags = req_task->out_flags;
 	args.data_phase   = req_task->data_phase;
 	args.command_type = req_task->req_cmd;
 
+	if (req_task->out_flags.all) {
+		args.tf_flags |= IDE_TFLAG_FLAGGED;
+
+		if (req_task->out_flags.b.data)
+			args.tf_flags |= IDE_TFLAG_OUT_DATA;
+
+		if (req_task->out_flags.b.nsector_hob)
+			args.tf_flags |= IDE_TFLAG_OUT_HOB_NSECT;
+		if (req_task->out_flags.b.sector_hob)
+			args.tf_flags |= IDE_TFLAG_OUT_HOB_LBAL;
+		if (req_task->out_flags.b.lcyl_hob)
+			args.tf_flags |= IDE_TFLAG_OUT_HOB_LBAM;
+		if (req_task->out_flags.b.hcyl_hob)
+			args.tf_flags |= IDE_TFLAG_OUT_HOB_LBAH;
+
+		if (req_task->out_flags.b.error_feature)
+			args.tf_flags |= IDE_TFLAG_OUT_FEATURE;
+		if (req_task->out_flags.b.nsector)
+			args.tf_flags |= IDE_TFLAG_OUT_NSECT;
+		if (req_task->out_flags.b.sector)
+			args.tf_flags |= IDE_TFLAG_OUT_LBAL;
+		if (req_task->out_flags.b.lcyl)
+			args.tf_flags |= IDE_TFLAG_OUT_LBAM;
+		if (req_task->out_flags.b.hcyl)
+			args.tf_flags |= IDE_TFLAG_OUT_LBAH;
+	}
+
 	drive->io_32bit = 0;
 	switch(req_task->data_phase) {
 		case TASKFILE_OUT_DMAQ:
@@ -649,7 +689,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	memcpy(req_task->hob_ports, &args.tf_array[0], HDIO_DRIVE_HOB_HDR_SIZE - 2);
 	memcpy(req_task->io_ports, &args.tf_array[6], HDIO_DRIVE_TASK_HDR_SIZE);
 	req_task->in_flags  = args.tf_in_flags;
-	req_task->out_flags = args.tf_out_flags;
 
 	if (copy_to_user(buf, req_task, tasksize)) {
 		err = -EFAULT;
@@ -782,9 +821,6 @@ int ide_task_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
  */
 ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
-	struct ide_taskfile *tf = &task->tf;
-
 	if (task->data_phase == TASKFILE_MULTI_IN ||
 	    task->data_phase == TASKFILE_MULTI_OUT) {
 		if (!drive->mult_count) {
@@ -807,75 +843,5 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 			task->tf_in_flags.all |= (IDE_HOB_STD_IN_FLAGS  << 8);
         }
 
-	/* ALL Command Block Executions SHALL clear nIEN, unless otherwise */
-	if (IDE_CONTROL_REG)
-		/* clear nIEN */
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
-	SELECT_MASK(drive, 0);
-
-	if (task->tf_out_flags.b.data)
-		hwif->OUTW((tf->hob_data << 8) | tf->data, IDE_DATA_REG);
-
-	/* (ks) send hob registers first */
-	if (task->tf_out_flags.b.nsector_hob)
-		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
-	if (task->tf_out_flags.b.sector_hob)
-		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
-	if (task->tf_out_flags.b.lcyl_hob)
-		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
-	if (task->tf_out_flags.b.hcyl_hob)
-		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
-
-	/* (ks) Send now the standard registers */
-	if (task->tf_out_flags.b.error_feature)
-		hwif->OUTB(tf->feature, IDE_FEATURE_REG);
-	/* refers to number of sectors to transfer */
-	if (task->tf_out_flags.b.nsector)
-		hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
-	/* refers to sector offset or start sector */
-	if (task->tf_out_flags.b.sector)
-		hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
-	if (task->tf_out_flags.b.lcyl)
-		hwif->OUTB(tf->lbam, IDE_LCYL_REG);
-	if (task->tf_out_flags.b.hcyl)
-		hwif->OUTB(tf->lbah, IDE_HCYL_REG);
-
-        /*
-	 * (ks) In the flagged taskfile approch, we will use all specified
-	 * registers and the register value will not be changed, except the
-	 * select bit (master/slave) in the drive_head register. We must make
-	 * sure that the desired drive is selected.
-	 */
-	hwif->OUTB(tf->device | drive->select.all, IDE_SELECT_REG);
-	switch(task->data_phase) {
-
-   	        case TASKFILE_OUT_DMAQ:
-		case TASKFILE_OUT_DMA:
-		case TASKFILE_IN_DMAQ:
-		case TASKFILE_IN_DMA:
-			if (!drive->using_dma)
-				break;
-
-			if (!hwif->dma_setup(drive)) {
-				hwif->dma_exec_cmd(drive, tf->command);
-				hwif->dma_start(drive);
-				return ide_started;
-			}
-			break;
-
-	        default:
- 			if (task->handler == NULL)
-				return ide_stopped;
-
-			/* Issue the command */
-			if (task->prehandler) {
-				hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
-				ndelay(400);	/* FIXME */
-				return task->prehandler(drive, task->rq);
-			}
-			ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
-			return ide_started;
-	}
-
-	return ide_stopped;
+	return do_rw_taskfile(drive, task);
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 11bfbc40c50a..6dfee2a46f1b 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1063,6 +1063,28 @@ extern int ide_wait_cmd(ide_drive_t *, u8, u8, u8, u8, u8 *);
 enum {
 	IDE_TFLAG_LBA48			= (1 << 0),
 	IDE_TFLAG_NO_SELECT_MASK	= (1 << 1),
+	IDE_TFLAG_FLAGGED		= (1 << 2),
+	IDE_TFLAG_OUT_DATA		= (1 << 3),
+	IDE_TFLAG_OUT_HOB_FEATURE	= (1 << 4),
+	IDE_TFLAG_OUT_HOB_NSECT		= (1 << 5),
+	IDE_TFLAG_OUT_HOB_LBAL		= (1 << 6),
+	IDE_TFLAG_OUT_HOB_LBAM		= (1 << 7),
+	IDE_TFLAG_OUT_HOB_LBAH		= (1 << 8),
+	IDE_TFLAG_OUT_HOB		= IDE_TFLAG_OUT_HOB_FEATURE |
+					  IDE_TFLAG_OUT_HOB_NSECT |
+					  IDE_TFLAG_OUT_HOB_LBAL |
+					  IDE_TFLAG_OUT_HOB_LBAM |
+					  IDE_TFLAG_OUT_HOB_LBAH,
+	IDE_TFLAG_OUT_FEATURE		= (1 << 9),
+	IDE_TFLAG_OUT_NSECT		= (1 << 10),
+	IDE_TFLAG_OUT_LBAL		= (1 << 11),
+	IDE_TFLAG_OUT_LBAM		= (1 << 12),
+	IDE_TFLAG_OUT_LBAH		= (1 << 13),
+	IDE_TFLAG_OUT_TF		= IDE_TFLAG_OUT_FEATURE |
+					  IDE_TFLAG_OUT_NSECT |
+					  IDE_TFLAG_OUT_LBAL |
+					  IDE_TFLAG_OUT_LBAM |
+					  IDE_TFLAG_OUT_LBAH,
 };
 
 struct ide_taskfile {
@@ -1099,8 +1121,7 @@ typedef struct ide_task_s {
 		struct ide_taskfile	tf;
 		u8			tf_array[14];
 	};
-	u8			tf_flags;
-	ide_reg_valid_t		tf_out_flags;
+	u16			tf_flags;
 	ide_reg_valid_t		tf_in_flags;
 	int			data_phase;
 	int			command_type;
-- 
cgit v1.2.3


From 4ee06b7e677da4c75f2fcc5fd850543852d18bf2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:08 +0100
Subject: ide: remove stale ide.h "configuration options"

Remove stale ide.h "configuration options":

* INITIAL_MULT_COUNT - always defined to 0

* SUPPORT_SLOW_DATA_PORTS - unused

* OK_TO_RESET_CONTROLLER - always defined to 1

* DISABLE_IRQ_NOSYNC - always defined to 0

Leave SUPPORT_VLB_SYNC (defined to 0 for CRIS and FRV, otherwise to 1)
for now but disallow overriding it by <asm/ide.h>.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c            |  4 ----
 drivers/ide/ide-iops.c          |  9 ++-------
 drivers/ide/ide-probe.c         |  4 ++--
 include/asm-cris/arch-v10/ide.h |  5 -----
 include/asm-cris/arch-v32/ide.h |  5 -----
 include/asm-frv/ide.h           |  6 ------
 include/asm-powerpc/ide.h       |  3 ---
 include/linux/ide.h             | 27 ++++-----------------------
 8 files changed, 8 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 33458fe7f490..4ea8419feeec 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1469,12 +1469,8 @@ void ide_timer_expiry (unsigned long data)
 			 */
 			spin_unlock(&ide_lock);
 			hwif  = HWIF(drive);
-#if DISABLE_IRQ_NOSYNC
-			disable_irq_nosync(hwif->irq);
-#else
 			/* disable_irq_nosync ?? */
 			disable_irq(hwif->irq);
-#endif /* DISABLE_IRQ_NOSYNC */
 			/* local CPU only,
 			 * as if we were handling an interrupt */
 			local_irq_disable();
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 6a6d1c188030..617888048eed 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -1050,8 +1050,7 @@ static void ide_disk_pre_reset(ide_drive_t *drive)
 	drive->special.all = 0;
 	drive->special.b.set_geometry = legacy;
 	drive->special.b.recalibrate  = legacy;
-	if (OK_TO_RESET_CONTROLLER)
-		drive->mult_count = 0;
+	drive->mult_count = 0;
 	if (!drive->keep_settings && !drive->using_dma)
 		drive->mult_req = 0;
 	if (drive->mult_req != drive->mult_count)
@@ -1136,7 +1135,6 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	for (unit = 0; unit < MAX_DRIVES; ++unit)
 		pre_reset(&hwif->drives[unit]);
 
-#if OK_TO_RESET_CONTROLLER
 	if (!IDE_CONTROL_REG) {
 		spin_unlock_irqrestore(&ide_lock, flags);
 		return ide_stopped;
@@ -1173,11 +1171,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * state when the disks are reset this way. At least, the Winbond
 	 * 553 documentation says that
 	 */
-	if (hwif->resetproc != NULL) {
+	if (hwif->resetproc)
 		hwif->resetproc(drive);
-	}
-	
-#endif	/* OK_TO_RESET_CONTROLLER */
 
 	spin_unlock_irqrestore(&ide_lock, flags);
 	return ide_started;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 0dda7ac0d951..8e5d8dd315af 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -95,10 +95,10 @@ static void ide_disk_init_mult_count(ide_drive_t *drive)
 #ifdef CONFIG_IDEDISK_MULTI_MODE
 		id->multsect = ((id->max_multsect/2) > 1) ? id->max_multsect : 0;
 		id->multsect_valid = id->multsect ? 1 : 0;
-		drive->mult_req = id->multsect_valid ? id->max_multsect : INITIAL_MULT_COUNT;
+		drive->mult_req = id->multsect_valid ? id->max_multsect : 0;
 		drive->special.b.set_multmode = drive->mult_req ? 1 : 0;
 #else	/* original, pre IDE-NFG, per request of AC */
-		drive->mult_req = INITIAL_MULT_COUNT;
+		drive->mult_req = 0;
 		if (drive->mult_req > id->max_multsect)
 			drive->mult_req = id->max_multsect;
 		if (drive->mult_req || ((id->multsect_valid & 1) && id->multsect))
diff --git a/include/asm-cris/arch-v10/ide.h b/include/asm-cris/arch-v10/ide.h
index 78b301ed7b12..ea34e0d0a388 100644
--- a/include/asm-cris/arch-v10/ide.h
+++ b/include/asm-cris/arch-v10/ide.h
@@ -89,11 +89,6 @@ static inline void ide_init_default_hwifs(void)
 	}
 }
 
-/* some configuration options we don't need */
-
-#undef SUPPORT_VLB_SYNC
-#define SUPPORT_VLB_SYNC 0
-
 #endif /* __KERNEL__ */
 
 #endif /* __ASMCRIS_IDE_H */
diff --git a/include/asm-cris/arch-v32/ide.h b/include/asm-cris/arch-v32/ide.h
index 11296170d057..fb9c3627a5b4 100644
--- a/include/asm-cris/arch-v32/ide.h
+++ b/include/asm-cris/arch-v32/ide.h
@@ -48,11 +48,6 @@ static inline unsigned long ide_default_io_base(int index)
 	return REG_TYPE_CONV(unsigned long, reg_ata_rw_ctrl2, ctrl2);
 }
 
-/* some configuration options we don't need */
-
-#undef SUPPORT_VLB_SYNC
-#define SUPPORT_VLB_SYNC 0
-
 #define IDE_ARCH_ACK_INTR
 #define ide_ack_intr(hwif)	((hwif)->ack_intr(hwif))
 
diff --git a/include/asm-frv/ide.h b/include/asm-frv/ide.h
index f0bd2cb250c1..8c9a540d4344 100644
--- a/include/asm-frv/ide.h
+++ b/include/asm-frv/ide.h
@@ -18,12 +18,6 @@
 #include <asm/io.h>
 #include <asm/irq.h>
 
-#undef SUPPORT_SLOW_DATA_PORTS
-#define SUPPORT_SLOW_DATA_PORTS 0
-
-#undef SUPPORT_VLB_SYNC
-#define SUPPORT_VLB_SYNC 0
-
 #ifndef MAX_HWIFS
 #define MAX_HWIFS 8
 #endif
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index fd7f5a430f0a..6d50310ecaea 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -42,9 +42,6 @@ struct ide_machdep_calls {
 
 extern struct ide_machdep_calls ppc_ide_md;
 
-#undef	SUPPORT_SLOW_DATA_PORTS
-#define	SUPPORT_SLOW_DATA_PORTS	0
-
 #define IDE_ARCH_OBSOLETE_DEFAULTS
 
 static __inline__ int ide_default_irq(unsigned long base)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 6dfee2a46f1b..0b088f18d422 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -27,25 +27,10 @@
 #include <asm/semaphore.h>
 #include <asm/mutex.h>
 
-/******************************************************************************
- * IDE driver configuration options (play with these as desired):
- *
- * REALLY_SLOW_IO can be defined in ide.c and ide-cd.c, if necessary
- */
-#define INITIAL_MULT_COUNT	0	/* off=0; on=2,4,8,16,32, etc.. */
-
-#ifndef SUPPORT_SLOW_DATA_PORTS		/* 1 to support slow data ports */
-#define SUPPORT_SLOW_DATA_PORTS	1	/* 0 to reduce kernel size */
-#endif
-#ifndef SUPPORT_VLB_SYNC		/* 1 to support weird 32-bit chips */
-#define SUPPORT_VLB_SYNC	1	/* 0 to reduce kernel size */
-#endif
-#ifndef OK_TO_RESET_CONTROLLER		/* 1 needed for good error recovery */
-#define OK_TO_RESET_CONTROLLER	1	/* 0 for use with AH2372A/B interface */
-#endif
-
-#ifndef DISABLE_IRQ_NOSYNC
-#define DISABLE_IRQ_NOSYNC	0
+#if defined(CRIS) || defined(FRV)
+# define SUPPORT_VLB_SYNC 0
+#else
+# define SUPPORT_VLB_SYNC 1
 #endif
 
 /*
@@ -55,10 +40,6 @@
  
 #define IDE_NO_IRQ		(-1)
 
-/*
- *  "No user-serviceable parts" beyond this point  :)
- *****************************************************************************/
-
 typedef unsigned char	byte;	/* used everywhere */
 
 /*
-- 
cgit v1.2.3


From 807e35d695690011faa1ce3ad67dfc23c1e39bdc Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:10 +0100
Subject: ide: use ide_tf_load() in execute_drive_cmd()

* Add IDE_TFLAG_OUT_DEVICE taskfile flag to indicate the need of writing
  the Device register and handle it in ide_tf_load().

  Update ide_tf_load() and {do_rw,flagged}_taskfile() users accordingly.

* Use struct ide_taskfile and ide_tf_load() in execute_drive_cmd().

* Make the debugging code dump all taskfile registers for both
  REQ_ATA_TYPE_{CMD,TASK} requests and move it to ide_tf_load()
  so it also covers REQ_ATA_TYPE_TASKFILE requests.

There should be no functionality changes caused by this patch
(unless DEBUG is defined).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     |  2 +-
 drivers/ide/ide-io.c       | 60 +++++++++++++++++++---------------------------
 drivers/ide/ide-taskfile.c | 10 +++++++-
 include/linux/ide.h        |  1 +
 4 files changed, 36 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 6182c23d2020..3e03d0c1a475 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -159,7 +159,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_NO_SELECT_MASK;  /* FIXME? */
-	task.tf_flags |= IDE_TFLAG_OUT_TF;
+	task.tf_flags |= (IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE);
 
 	if (drive->select.b.lba) {
 		if (lba48) {
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index c5a7b3ac7c87..9ffab8c71e75 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -232,7 +232,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 	return ide_stopped;
 
 out_do_tf:
-	args->tf_flags = IDE_TFLAG_OUT_TF;
+	args->tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	if (drive->addressing == 1)
 		args->tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
 	args->command_type = IDE_DRIVE_TASK_NO_DATA;
@@ -710,7 +710,7 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 		return ide_stopped;
 	}
 
-	args.tf_flags = IDE_TFLAG_OUT_TF;
+	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	if (drive->addressing == 1)
 		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
 
@@ -849,6 +849,8 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	u8 *args = rq->buffer;
+	ide_task_t ltask;
+	struct ide_taskfile *tf = &ltask.tf;
 
 	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *task = rq->special;
@@ -869,6 +871,8 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 			break;
 		}
 
+		task->tf_flags |= IDE_TFLAG_OUT_DEVICE;
+
 		if (task->tf_flags & IDE_TFLAG_FLAGGED)
 			return flagged_taskfile(drive, task);
 
@@ -882,46 +886,32 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 	if (args == NULL)
 		goto done;
 
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG); /* clear nIEN */
-
-	SELECT_MASK(drive, 0);
-
+	memset(&ltask, 0, sizeof(ltask));
 	if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
 #ifdef DEBUG
- 		printk("%s: DRIVE_TASK_CMD ", drive->name);
- 		printk("cmd=0x%02x ", args[0]);
- 		printk("fr=0x%02x ", args[1]);
- 		printk("ns=0x%02x ", args[2]);
- 		printk("sc=0x%02x ", args[3]);
- 		printk("lcyl=0x%02x ", args[4]);
- 		printk("hcyl=0x%02x ", args[5]);
- 		printk("sel=0x%02x\n", args[6]);
+		printk("%s: DRIVE_TASK_CMD\n", drive->name);
 #endif
- 		hwif->OUTB(args[1], IDE_FEATURE_REG);
-		hwif->OUTB(args[2], IDE_NSECTOR_REG);
- 		hwif->OUTB(args[3], IDE_SECTOR_REG);
- 		hwif->OUTB(args[4], IDE_LCYL_REG);
- 		hwif->OUTB(args[5], IDE_HCYL_REG);
- 		hwif->OUTB((args[6] & 0xEF)|drive->select.all, IDE_SELECT_REG);
+		memcpy(&ltask.tf_array[7], &args[1], 6);
+		ltask.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	} else { /* rq->cmd_type == REQ_TYPE_ATA_CMD */
 #ifdef DEBUG
- 		printk("%s: DRIVE_CMD ", drive->name);
- 		printk("cmd=0x%02x ", args[0]);
- 		printk("sc=0x%02x ", args[1]);
- 		printk("fr=0x%02x ", args[2]);
- 		printk("xx=0x%02x\n", args[3]);
+		printk("%s: DRIVE_CMD\n", drive->name);
 #endif
-		hwif->OUTB(args[2], IDE_FEATURE_REG);
- 		if (args[0] == WIN_SMART) {
-			hwif->OUTB(args[3],IDE_NSECTOR_REG);
- 			hwif->OUTB(args[1],IDE_SECTOR_REG);
-			hwif->OUTB(0x4f, IDE_LCYL_REG);
-			hwif->OUTB(0xc2, IDE_HCYL_REG);
-		} else
-			hwif->OUTB(args[1], IDE_NSECTOR_REG);
+		tf->feature = args[2];
+		if (args[0] == WIN_SMART) {
+			tf->nsect = args[3];
+			tf->lbal  = args[1];
+			tf->lbam  = 0x4f;
+			tf->lbah  = 0xc2;
+			ltask.tf_flags = IDE_TFLAG_OUT_TF;
+		} else {
+			tf->nsect = args[1];
+			ltask.tf_flags = IDE_TFLAG_OUT_FEATURE |
+					 IDE_TFLAG_OUT_NSECT;
+		}
  	}
-
+	tf->command = args[0];
+	ide_tf_load(drive, &ltask);
 	ide_execute_command(drive, args[0], &drive_cmd_intr, WAIT_CMD, NULL);
 	return ide_started;
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index a7668b459fe1..8a5a10fdcfcb 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -72,6 +72,13 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_FLAGGED)
 		HIHI = 0xFF;
 
+#ifdef DEBUG
+	printk("%s: tf: feat 0x%02x nsect 0x%02x lbal 0x%02x "
+		"lbam 0x%02x lbah 0x%02x dev 0x%02x cmd 0x%02x\n",
+		drive->name, tf->feature, tf->nsect, tf->lbal,
+		tf->lbam, tf->lbah, tf->device, tf->command);
+#endif
+
 	if (IDE_CONTROL_REG)
 		hwif->OUTB(drive->ctl, IDE_CONTROL_REG); /* clear nIEN */
 
@@ -103,7 +110,8 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_OUT_LBAH)
 		hwif->OUTB(tf->lbah, IDE_HCYL_REG);
 
-	hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
+	if (task->tf_flags & IDE_TFLAG_OUT_DEVICE)
+		hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
 }
 
 EXPORT_SYMBOL_GPL(ide_tf_load);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 0b088f18d422..e1f652c6440c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1066,6 +1066,7 @@ enum {
 					  IDE_TFLAG_OUT_LBAL |
 					  IDE_TFLAG_OUT_LBAM |
 					  IDE_TFLAG_OUT_LBAH,
+	IDE_TFLAG_OUT_DEVICE		= (1 << 14),
 };
 
 struct ide_taskfile {
-- 
cgit v1.2.3


From 29ed2a5f8c4380959f18e9cbaff13bc61e09889c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:11 +0100
Subject: ide: remove REQ_TYPE_ATA_TASK

Based on the earlier work by Tejun Heo.

All users are gone so we can finally remove it.

Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 25 +------------------------
 drivers/ide/ide-lib.c  |  3 +--
 include/linux/blkdev.h |  1 -
 3 files changed, 2 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 73e021267f3b..d36df77c7519 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -332,22 +332,6 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			args[1] = err;
 			args[2] = hwif->INB(IDE_NSECTOR_REG);
 		}
-	} else if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
-		u8 *args = (u8 *) rq->buffer;
-		if (rq->errors == 0)
-			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
-
-		if (args) {
-			args[0] = stat;
-			args[1] = err;
-			/* be sure we're looking at the low order bits */
-			hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
-			args[2] = hwif->INB(IDE_NSECTOR_REG);
-			args[3] = hwif->INB(IDE_SECTOR_REG);
-			args[4] = hwif->INB(IDE_LCYL_REG);
-			args[5] = hwif->INB(IDE_HCYL_REG);
-			args[6] = hwif->INB(IDE_SELECT_REG);
-		}
 	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *args = (ide_task_t *) rq->special;
 		if (rq->errors == 0)
@@ -877,13 +861,7 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		goto done;
 
 	memset(&ltask, 0, sizeof(ltask));
-	if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
-#ifdef DEBUG
-		printk("%s: DRIVE_TASK_CMD\n", drive->name);
-#endif
-		memcpy(&ltask.tf_array[7], &args[1], 6);
-		ltask.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
-	} else { /* rq->cmd_type == REQ_TYPE_ATA_CMD */
+	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
 #ifdef DEBUG
 		printk("%s: DRIVE_CMD\n", drive->name);
 #endif
@@ -1011,7 +989,6 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 			ide_config_drive_speed(drive, drive->desired_speed);
 
 		if (rq->cmd_type == REQ_TYPE_ATA_CMD ||
-		    rq->cmd_type == REQ_TYPE_ATA_TASK ||
 		    rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
 			return execute_drive_cmd(drive, rq);
 		else if (blk_pm_request(rq)) {
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 6dbf2af0d215..3bae2c46924f 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -458,8 +458,7 @@ static void ide_dump_opcode(ide_drive_t *drive)
 	spin_unlock(&ide_lock);
 	if (!rq)
 		return;
-	if (rq->cmd_type == REQ_TYPE_ATA_CMD ||
-	    rq->cmd_type == REQ_TYPE_ATA_TASK) {
+	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
 		char *args = rq->buffer;
 		if (args) {
 			opcode = args[0];
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d18ee67b40f8..40ee1706caa3 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -144,7 +144,6 @@ enum rq_cmd_type_bits {
 	 * private REQ_LB opcodes to differentiate what type of request this is
 	 */
 	REQ_TYPE_ATA_CMD,
-	REQ_TYPE_ATA_TASK,
 	REQ_TYPE_ATA_TASKFILE,
 	REQ_TYPE_ATA_PC,
 };
-- 
cgit v1.2.3


From 6a2144146aa2e0eb60e48ba73ac0e1c51346edf6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:11 +0100
Subject: ide: CPU endianness doesn't matter for special_t

special_t is used only internally by the IDE subsystem (it isn't
related to hardware registers and isn't exported to the user-space).

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 11 -----------
 1 file changed, 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index e1f652c6440c..c6ffc2028fb5 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -306,23 +306,12 @@ static inline void ide_init_hwif_ports(hw_regs_t *hw,
 typedef union {
 	unsigned all			: 8;
 	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
 		unsigned set_geometry	: 1;
 		unsigned recalibrate	: 1;
 		unsigned set_multmode	: 1;
 		unsigned set_tune	: 1;
 		unsigned serviced	: 1;
 		unsigned reserved	: 3;
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned reserved	: 3;
-		unsigned serviced	: 1;
-		unsigned set_tune	: 1;
-		unsigned set_multmode	: 1;
-		unsigned recalibrate	: 1;
-		unsigned set_geometry	: 1;
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
 	} b;
 } special_t;
 
-- 
cgit v1.2.3


From 22c525b976778cce5bb6f8fdcc70046168c54b1a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:11 +0100
Subject: ide: remove ata_status_t and atapi_status_t

Remove ata_status_t (unused) and atapi_status_t.

While at it:
* replace 'HWIF(drive)' by 'drive->hwif' (or just 'hwif' where possible)

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 14 +++++++-------
 drivers/ide/ide-lib.c    | 21 +++++++++------------
 drivers/ide/ide-tape.c   | 34 +++++++++++++++++-----------------
 drivers/scsi/ide-scsi.c  |  9 ++++-----
 include/linux/ide.h      | 44 --------------------------------------------
 5 files changed, 37 insertions(+), 85 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 53f90257a8f0..7b94c7aff256 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -788,12 +788,12 @@ static void idefloppy_retry_pc (ide_drive_t *drive)
 static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	atapi_status_t status;
 	atapi_bcount_t bcount;
 	atapi_ireason_t ireason;
 	idefloppy_pc_t *pc = floppy->pc;
 	struct request *rq = pc->rq;
 	unsigned int temp;
+	u8 stat;
 
 	debug_log(KERN_INFO "ide-floppy: Reached %s interrupt handler\n",
 		__FUNCTION__);
@@ -809,16 +809,16 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	}
 
 	/* Clear the interrupt */
-	status.all = HWIF(drive)->INB(IDE_STATUS_REG);
+	stat = drive->hwif->INB(IDE_STATUS_REG);
 
-	if (!status.b.drq) {			/* No more interrupts */
+	if ((stat & DRQ_STAT) == 0) {		/* No more interrupts */
 		debug_log(KERN_INFO "Packet command completed, %d bytes "
 			"transferred\n", pc->actually_transferred);
 		clear_bit(PC_DMA_IN_PROGRESS, &pc->flags);
 
 		local_irq_enable_in_hardirq();
 
-		if (status.b.check || test_bit(PC_DMA_ERROR, &pc->flags)) {
+		if ((stat & ERR_STAT) || test_bit(PC_DMA_ERROR, &pc->flags)) {
 			/* Error detected */
 			debug_log(KERN_INFO "ide-floppy: %s: I/O error\n",
 				drive->name);
@@ -1632,14 +1632,14 @@ static int idefloppy_get_format_progress(ide_drive_t *drive, int __user *arg)
 		/* Else assume format_unit has finished, and we're
 		** at 0x10000 */
 	} else {
-		atapi_status_t status;
 		unsigned long flags;
+		u8 stat;
 
 		local_irq_save(flags);
-		status.all = HWIF(drive)->INB(IDE_STATUS_REG);
+		stat = drive->hwif->INB(IDE_STATUS_REG);
 		local_irq_restore(flags);
 
-		progress_indication = !status.b.dsc ? 0 : 0x10000;
+		progress_indication = ((stat & SEEK_STAT) == 0) ? 0 : 0x10000;
 	}
 	if (put_user(progress_indication, arg))
 		return (-EFAULT);
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 3bae2c46924f..dc7e539b4d0b 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -562,27 +562,24 @@ static u8 ide_dump_ata_status(ide_drive_t *drive, const char *msg, u8 stat)
 static u8 ide_dump_atapi_status(ide_drive_t *drive, const char *msg, u8 stat)
 {
 	unsigned long flags;
-
-	atapi_status_t status;
 	atapi_error_t error;
 
-	status.all = stat;
 	error.all = 0;
 	local_irq_save(flags);
 	printk("%s: %s: status=0x%02x { ", drive->name, msg, stat);
-	if (status.b.bsy)
+	if (stat & BUSY_STAT)
 		printk("Busy ");
 	else {
-		if (status.b.drdy)	printk("DriveReady ");
-		if (status.b.df)	printk("DeviceFault ");
-		if (status.b.dsc)	printk("SeekComplete ");
-		if (status.b.drq)	printk("DataRequest ");
-		if (status.b.corr)	printk("CorrectedError ");
-		if (status.b.idx)	printk("Index ");
-		if (status.b.check)	printk("Error ");
+		if (stat & READY_STAT)	printk("DriveReady ");
+		if (stat & WRERR_STAT)	printk("DeviceFault ");
+		if (stat & SEEK_STAT)	printk("SeekComplete ");
+		if (stat & DRQ_STAT)	printk("DataRequest ");
+		if (stat & ECC_STAT)	printk("CorrectedError ");
+		if (stat & INDEX_STAT)	printk("Index ");
+		if (stat & ERR_STAT)	printk("Error ");
 	}
 	printk("}\n");
-	if (status.b.check && !status.b.bsy) {
+	if ((stat & (BUSY_STAT|ERR_STAT)) == ERR_STAT) {
 		error.all = HWIF(drive)->INB(IDE_ERROR_REG);
 		printk("%s: %s: error=0x%02x { ", drive->name, msg, error.all);
 		if (error.b.ili)	printk("IllegalLengthIndication ");
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 90e902d233cf..c91039505436 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1848,15 +1848,14 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	atapi_status_t status;
 	atapi_bcount_t bcount;
 	atapi_ireason_t ireason;
 	idetape_pc_t *pc = tape->pc;
-
 	unsigned int temp;
 #if SIMULATE_ERRORS
 	static int error_sim_count = 0;
 #endif
+	u8 stat;
 
 #if IDETAPE_DEBUG_LOG
 	if (tape->debug_level >= 4)
@@ -1865,10 +1864,10 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 #endif /* IDETAPE_DEBUG_LOG */	
 
 	/* Clear the interrupt */
-	status.all = HWIF(drive)->INB(IDE_STATUS_REG);
+	stat = hwif->INB(IDE_STATUS_REG);
 
 	if (test_bit(PC_DMA_IN_PROGRESS, &pc->flags)) {
-		if (HWIF(drive)->ide_dma_end(drive) || status.b.check) {
+		if (hwif->ide_dma_end(drive) || (stat & ERR_STAT)) {
 			/*
 			 * A DMA error is sometimes expected. For example,
 			 * if the tape is crossing a filemark during a
@@ -1902,7 +1901,7 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 	}
 
 	/* No more interrupts */
-	if (!status.b.drq) {
+	if ((stat & DRQ_STAT) == 0) {
 #if IDETAPE_DEBUG_LOG
 		if (tape->debug_level >= 2)
 			printk(KERN_INFO "ide-tape: Packet command completed, %d bytes transferred\n", pc->actually_transferred);
@@ -1917,12 +1916,13 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 		    (++error_sim_count % 100) == 0) {
 			printk(KERN_INFO "ide-tape: %s: simulating error\n",
 				tape->name);
-			status.b.check = 1;
+			stat |= ERR_STAT;
 		}
 #endif
-		if (status.b.check && pc->c[0] == IDETAPE_REQUEST_SENSE_CMD)
-			status.b.check = 0;
-		if (status.b.check || test_bit(PC_DMA_ERROR, &pc->flags)) {	/* Error detected */
+		if ((stat & ERR_STAT) && pc->c[0] == IDETAPE_REQUEST_SENSE_CMD)
+			stat &= ~ERR_STAT;
+		if ((stat & ERR_STAT) || test_bit(PC_DMA_ERROR, &pc->flags)) {
+			/* Error detected */
 #if IDETAPE_DEBUG_LOG
 			if (tape->debug_level >= 1)
 				printk(KERN_INFO "ide-tape: %s: I/O error\n",
@@ -1941,7 +1941,7 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 		}
 		pc->error = 0;
 		if (test_bit(PC_WAIT_FOR_DSC, &pc->flags) &&
-		    !status.b.dsc) {
+		    (stat & SEEK_STAT) == 0) {
 			/* Media access command */
 			tape->dsc_polling_start = jiffies;
 			tape->dsc_polling_frequency = IDETAPE_DSC_MA_FAST;
@@ -2285,11 +2285,11 @@ static ide_startstop_t idetape_media_access_finished (ide_drive_t *drive)
 {
 	idetape_tape_t *tape = drive->driver_data;
 	idetape_pc_t *pc = tape->pc;
-	atapi_status_t status;
+	u8 stat;
 
-	status.all = HWIF(drive)->INB(IDE_STATUS_REG);
-	if (status.b.dsc) {
-		if (status.b.check) {
+	stat = drive->hwif->INB(IDE_STATUS_REG);
+	if (stat & SEEK_STAT) {
+		if (stat & ERR_STAT) {
 			/* Error detected */
 			if (pc->c[0] != IDETAPE_TEST_UNIT_READY_CMD)
 				printk(KERN_ERR "ide-tape: %s: I/O error, ",
@@ -2407,7 +2407,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	idetape_tape_t *tape = drive->driver_data;
 	idetape_pc_t *pc = NULL;
 	struct request *postponed_rq = tape->postponed_rq;
-	atapi_status_t status;
+	u8 stat;
 
 #if IDETAPE_DEBUG_LOG
 #if 0
@@ -2455,7 +2455,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	 * If the tape is still busy, postpone our request and service
 	 * the other device meanwhile.
 	 */
-	status.all = HWIF(drive)->INB(IDE_STATUS_REG);
+	stat = drive->hwif->INB(IDE_STATUS_REG);
 
 	if (!drive->dsc_overlap && !(rq->cmd[0] & REQ_IDETAPE_PC2))
 		set_bit(IDETAPE_IGNORE_DSC, &tape->flags);
@@ -2471,7 +2471,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 		tape->insert_speed = tape->insert_size / 1024 * HZ / (jiffies - tape->insert_time);
 	calculate_speeds(drive);
 	if (!test_and_clear_bit(IDETAPE_IGNORE_DSC, &tape->flags) &&
-	    !status.b.dsc) {
+	    (stat & SEEK_STAT) == 0) {
 		if (postponed_rq == NULL) {
 			tape->dsc_polling_start = jiffies;
 			tape->dsc_polling_frequency = tape->best_dsc_rw_frequency;
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 9706de9d98d5..bd2b56912472 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -398,11 +398,10 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 	idescsi_pc_t *pc=scsi->pc;
 	struct request *rq = pc->rq;
 	atapi_bcount_t bcount;
-	atapi_status_t status;
 	atapi_ireason_t ireason;
 	atapi_feature_t feature;
-
 	unsigned int temp;
+	u8 stat;
 
 #if IDESCSI_DEBUG_LOG
 	printk (KERN_INFO "ide-scsi: Reached idescsi_pc_intr interrupt handler\n");
@@ -427,14 +426,14 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 
 	feature.all = 0;
 	/* Clear the interrupt */
-	status.all = HWIF(drive)->INB(IDE_STATUS_REG);
+	stat = drive->hwif->INB(IDE_STATUS_REG);
 
-	if (!status.b.drq) {
+	if ((stat & DRQ_STAT) == 0) {
 		/* No more interrupts */
 		if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
 			printk (KERN_INFO "Packet command completed, %d bytes transferred\n", pc->actually_transferred);
 		local_irq_enable_in_hardirq();
-		if (status.b.check)
+		if (stat & ERR_STAT)
 			rq->errors++;
 		idescsi_end_request (drive, 1, 0);
 		return ide_stopped;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c6ffc2028fb5..921cf717f22a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -365,50 +365,6 @@ typedef union {
 	} b;
 } select_t, ata_select_t;
 
-/*
- * The ATA-IDE Status Register.
- * The ATAPI Status Register.
- *
- * check	: Error occurred
- * idx		: Index Error
- * corr		: Correctable error occurred
- * drq		: Data is request by the device
- * dsc		: Disk Seek Complete			: ata
- *		: Media access command finished		: atapi
- * df		: Device Fault				: ata
- *		: Reserved				: atapi
- * drdy		: Ready, Command Mode Capable		: ata
- *		: Ignored for ATAPI commands		: atapi
- * bsy		: Disk is Busy
- *		: The device has access to the command block
- */
-typedef union {
-	unsigned all			:8;
-	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-		unsigned check		:1;
-		unsigned idx		:1;
-		unsigned corr		:1;
-		unsigned drq		:1;
-		unsigned dsc		:1;
-		unsigned df		:1;
-		unsigned drdy		:1;
-		unsigned bsy		:1;
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned bsy		:1;
-		unsigned drdy		:1;
-		unsigned df		:1;
-		unsigned dsc		:1;
-		unsigned drq		:1;
-		unsigned corr           :1;
-		unsigned idx		:1;
-		unsigned check		:1;
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
-	} b;
-} ata_status_t, atapi_status_t;
-
 /*
  * ATAPI Feature Register
  *
-- 
cgit v1.2.3


From 0e38a66a1e69821ab57a06d5a7b11f0df9275bf4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:12 +0100
Subject: ide: remove atapi_error_t (take 2)

Remove atapi_error_t.

While at it:
* replace 'HWIF(drive)' by 'drive->hwif'

v2:
* Add {ILI,EOM,LFS}_ERR defines to <linux/hdreg.h>.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c |  3 +--
 drivers/ide/ide-lib.c    | 21 ++++++++++-----------
 drivers/ide/ide-tape.c   |  3 +--
 include/linux/hdreg.h    |  3 +++
 include/linux/ide.h      | 30 ------------------------------
 5 files changed, 15 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 7b94c7aff256..2a37a08ddd57 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -772,9 +772,8 @@ static void idefloppy_retry_pc (ide_drive_t *drive)
 {
 	idefloppy_pc_t *pc;
 	struct request *rq;
-	atapi_error_t error;
 
-	error.all = HWIF(drive)->INB(IDE_ERROR_REG);
+	(void)drive->hwif->INB(IDE_ERROR_REG);
 	pc = idefloppy_next_pc_storage(drive);
 	rq = idefloppy_next_rq_storage(drive);
 	idefloppy_create_request_sense_cmd(pc);
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index dc7e539b4d0b..562f5efae9c6 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -562,9 +562,8 @@ static u8 ide_dump_ata_status(ide_drive_t *drive, const char *msg, u8 stat)
 static u8 ide_dump_atapi_status(ide_drive_t *drive, const char *msg, u8 stat)
 {
 	unsigned long flags;
-	atapi_error_t error;
+	u8 err = 0;
 
-	error.all = 0;
 	local_irq_save(flags);
 	printk("%s: %s: status=0x%02x { ", drive->name, msg, stat);
 	if (stat & BUSY_STAT)
@@ -580,19 +579,19 @@ static u8 ide_dump_atapi_status(ide_drive_t *drive, const char *msg, u8 stat)
 	}
 	printk("}\n");
 	if ((stat & (BUSY_STAT|ERR_STAT)) == ERR_STAT) {
-		error.all = HWIF(drive)->INB(IDE_ERROR_REG);
-		printk("%s: %s: error=0x%02x { ", drive->name, msg, error.all);
-		if (error.b.ili)	printk("IllegalLengthIndication ");
-		if (error.b.eom)	printk("EndOfMedia ");
-		if (error.b.abrt)	printk("AbortedCommand ");
-		if (error.b.mcr)	printk("MediaChangeRequested ");
-		if (error.b.sense_key)	printk("LastFailedSense=0x%02x ",
-						error.b.sense_key);
+		err = drive->hwif->INB(IDE_ERROR_REG);
+		printk("%s: %s: error=0x%02x { ", drive->name, msg, err);
+		if (err & ILI_ERR)	printk("IllegalLengthIndication ");
+		if (err & EOM_ERR)	printk("EndOfMedia ");
+		if (err & ABRT_ERR)	printk("AbortedCommand ");
+		if (err & MCR_ERR)	printk("MediaChangeRequested ");
+		if (err & LFS_ERR)	printk("LastFailedSense=0x%02x ",
+						(err & LFS_ERR) >> 4);
 		printk("}\n");
 	}
 	ide_dump_opcode(drive);
 	local_irq_restore(flags);
-	return error.all;
+	return err;
 }
 
 /**
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index c91039505436..2c03f469f06e 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1808,9 +1808,8 @@ static ide_startstop_t idetape_retry_pc (ide_drive_t *drive)
 	idetape_tape_t *tape = drive->driver_data;
 	idetape_pc_t *pc;
 	struct request *rq;
-	atapi_error_t error;
 
-	error.all = HWIF(drive)->INB(IDE_ERROR_REG);
+	(void)drive->hwif->INB(IDE_ERROR_REG);
 	pc = idetape_next_pc_storage(drive);
 	rq = idetape_next_rq_storage(drive);
 	idetape_create_request_sense_cmd(pc);
diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h
index df17bf767d98..0521f1234f15 100644
--- a/include/linux/hdreg.h
+++ b/include/linux/hdreg.h
@@ -44,7 +44,9 @@
 
 /* Bits for HD_ERROR */
 #define MARK_ERR		0x01	/* Bad address mark */
+#define ILI_ERR			0x01	/* Illegal Length Indication (ATAPI) */
 #define TRK0_ERR		0x02	/* couldn't find track 0 */
+#define EOM_ERR			0x02	/* End Of Media (ATAPI) */
 #define ABRT_ERR		0x04	/* Command aborted */
 #define MCR_ERR			0x08	/* media change request */
 #define ID_ERR			0x10	/* ID field not found */
@@ -52,6 +54,7 @@
 #define ECC_ERR			0x40	/* Uncorrectable ECC error */
 #define BBD_ERR			0x80	/* pre-EIDE meaning:  block marked bad */
 #define ICRC_ERR		0x80	/* new meaning:  CRC error during transfer */
+#define LFS_ERR			0xf0	/* Last Failed Sense (ATAPI) */
 
 /* Bits of HD_NSECTOR */
 #define CD			0x01
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 921cf717f22a..0c1b0aaa2865 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -416,36 +416,6 @@ typedef union {
 	} b;
 } atapi_ireason_t;
 
-/*
- * The ATAPI error register.
- *
- * ili		: Illegal Length Indication
- * eom		: End Of Media Detected
- * abrt		: Aborted command - As defined by ATA
- * mcr		: Media Change Requested - As defined by ATA
- * sense_key	: Sense key of the last failed packet command
- */
-typedef union {
-	unsigned all			:8;
-	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-		unsigned ili		:1;
-		unsigned eom		:1;
-		unsigned abrt		:1;
-		unsigned mcr		:1;
-		unsigned sense_key	:4;
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned sense_key	:4;
-		unsigned mcr		:1;
-		unsigned abrt		:1;
-		unsigned eom		:1;
-		unsigned ili		:1;
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
-	} b;
-} atapi_error_t;
-
 /*
  * Status returned from various ide_ functions
  */
-- 
cgit v1.2.3


From e5f9f5a89a01abc2b9c09747452aeb9218d6bffd Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:12 +0100
Subject: ide: remove atapi_feature_t

Remove atapi_feature_t.

While at it:
* replace 'HWIF(drive)' by 'hwif'

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 10 +++++-----
 drivers/scsi/ide-scsi.c  | 11 ++++-------
 include/linux/ide.h      | 27 ---------------------------
 3 files changed, 9 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 2a37a08ddd57..95e302790001 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -1019,9 +1019,9 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	ide_hwif_t *hwif = drive->hwif;
-	atapi_feature_t feature;
 	atapi_bcount_t bcount;
 	ide_handler_t *pkt_xfer_routine;
+	u8 dma;
 
 	if (floppy->failed_pc == NULL &&
 	    pc->c[0] != IDEFLOPPY_REQUEST_SENSE_CMD)
@@ -1064,20 +1064,20 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 	if (test_and_clear_bit(PC_DMA_ERROR, &pc->flags))
 		ide_dma_off(drive);
 
-	feature.all = 0;
+	dma = 0;
 
 	if (test_bit(PC_DMA_RECOMMENDED, &pc->flags) && drive->using_dma)
-		feature.b.dma = !hwif->dma_setup(drive);
+		dma = !hwif->dma_setup(drive);
 
 	if (IDE_CONTROL_REG)
 		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
 	/* Use PIO/DMA */
-	HWIF(drive)->OUTB(feature.all, IDE_FEATURE_REG);
+	hwif->OUTB(dma, IDE_FEATURE_REG);
 	HWIF(drive)->OUTB(bcount.b.high, IDE_BCOUNTH_REG);
 	HWIF(drive)->OUTB(bcount.b.low, IDE_BCOUNTL_REG);
 	HWIF(drive)->OUTB(drive->select.all, IDE_SELECT_REG);
 
-	if (feature.b.dma) {	/* Begin DMA, if necessary */
+	if (dma) {	/* Begin DMA, if necessary */
 		set_bit(PC_DMA_IN_PROGRESS, &pc->flags);
 		hwif->dma_start(drive);
 	}
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index bd2b56912472..c009f2351346 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -399,7 +399,6 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 	struct request *rq = pc->rq;
 	atapi_bcount_t bcount;
 	atapi_ireason_t ireason;
-	atapi_feature_t feature;
 	unsigned int temp;
 	u8 stat;
 
@@ -424,7 +423,6 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 		(void) HWIF(drive)->ide_dma_end(drive);
 	}
 
-	feature.all = 0;
 	/* Clear the interrupt */
 	stat = drive->hwif->INB(IDE_STATUS_REG);
 
@@ -572,18 +570,17 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
 	ide_hwif_t *hwif = drive->hwif;
-	atapi_feature_t feature;
 	atapi_bcount_t bcount;
+	u8 dma = 0;
 
 	scsi->pc=pc;							/* Set the current packet command */
 	pc->actually_transferred=0;					/* We haven't transferred any data yet */
 	pc->current_position=pc->buffer;
 	bcount.all = min(pc->request_transfer, 63 * 1024);		/* Request to transfer the entire buffer at once */
 
-	feature.all = 0;
 	if (drive->using_dma && !idescsi_map_sg(drive, pc)) {
 		hwif->sg_mapped = 1;
-		feature.b.dma = !hwif->dma_setup(drive);
+		dma = !hwif->dma_setup(drive);
 		hwif->sg_mapped = 0;
 	}
 
@@ -591,11 +588,11 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 	if (IDE_CONTROL_REG)
 		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
 
-	HWIF(drive)->OUTB(feature.all, IDE_FEATURE_REG);
+	hwif->OUTB(dma, IDE_FEATURE_REG);
 	HWIF(drive)->OUTB(bcount.b.high, IDE_BCOUNTH_REG);
 	HWIF(drive)->OUTB(bcount.b.low, IDE_BCOUNTL_REG);
 
-	if (feature.b.dma)
+	if (dma)
 		set_bit(PC_DMA_OK, &pc->flags);
 
 	if (test_bit(IDESCSI_DRQ_INTERRUPT, &scsi->flags)) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 0c1b0aaa2865..a638dde17e77 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -365,33 +365,6 @@ typedef union {
 	} b;
 } select_t, ata_select_t;
 
-/*
- * ATAPI Feature Register
- *
- * dma		: Using DMA or PIO
- * reserved321	: Reserved
- * reserved654	: Reserved (Tag Type)
- * reserved7	: Reserved
- */
-typedef union {
-	unsigned all			:8;
-	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-		unsigned dma		:1;
-		unsigned reserved321	:3;
-		unsigned reserved654	:3;
-		unsigned reserved7	:1;
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned reserved7	:1;
-		unsigned reserved654	:3;
-		unsigned reserved321	:3;
-		unsigned dma		:1;
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
-	} b;
-} atapi_feature_t;
-
 /*
  * ATAPI Interrupt Reason Register.
  *
-- 
cgit v1.2.3


From 790d1239898d4f893112280decd344d90f43ee96 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:12 +0100
Subject: ide: remove ata_nsector_t, ata_data_t and atapi_bcount_t

Remove ata_nsector_t, ata_data_t (unused) and atapi_bcount_t.

While at it:
* replace 'HWIF(drive)' by 'hwif'

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c   | 12 +++++-------
 drivers/ide/ide-floppy.c | 37 ++++++++++++++++++-------------------
 drivers/ide/ide-tape.c   | 35 +++++++++++++++++++----------------
 drivers/scsi/ide-scsi.c  | 40 +++++++++++++++++++++++-----------------
 include/linux/ide.h      | 20 --------------------
 5 files changed, 65 insertions(+), 79 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index ffff96e6ab3e..747dc6023346 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -137,14 +137,12 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	unsigned int dma	= drive->using_dma;
+	u16 nsectors		= (u16)rq->nr_sectors;
 	u8 lba48		= (drive->addressing == 1) ? 1 : 0;
 	u8 command		= WIN_NOP;
-	ata_nsector_t		nsectors;
 	ide_task_t		task;
 	struct ide_taskfile	*tf = &task.tf;
 
-	nsectors.all		= (u16) rq->nr_sectors;
-
 	if ((hwif->host_flags & IDE_HFLAG_NO_LBA48_DMA) && lba48 && dma) {
 		if (block + rq->nr_sectors > 1ULL << 28)
 			dma = 0;
@@ -166,14 +164,14 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 			pr_debug("%s: LBA=0x%012llx\n", drive->name,
 					(unsigned long long)block);
 
-			tf->hob_nsect = nsectors.b.high;
+			tf->hob_nsect = (nsectors >> 8) & 0xff;
 			tf->hob_lbal  = (u8)(block >> 24);
 			if (sizeof(block) != 4) {
 				tf->hob_lbam = (u8)((u64)block >> 32);
 				tf->hob_lbah = (u8)((u64)block >> 40);
 			}
 
-			tf->nsect  = nsectors.b.low;
+			tf->nsect  = nsectors & 0xff;
 			tf->lbal   = (u8) block;
 			tf->lbam   = (u8)(block >>  8);
 			tf->lbah   = (u8)(block >> 16);
@@ -185,7 +183,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 #endif
 			task.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
 		} else {
-			tf->nsect  = nsectors.b.low;
+			tf->nsect  = nsectors & 0xff;
 			tf->lbal   = block;
 			tf->lbam   = block >>= 8;
 			tf->lbah   = block >>= 8;
@@ -200,7 +198,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 
 		pr_debug("%s: CHS=%u/%u/%u\n", drive->name, cyl, head, sect);
 
-		tf->nsect  = nsectors.b.low;
+		tf->nsect  = nsectors & 0xff;
 		tf->lbal   = sect;
 		tf->lbam   = cyl;
 		tf->lbah   = cyl >> 8;
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 95e302790001..239aebcfc359 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -787,11 +787,12 @@ static void idefloppy_retry_pc (ide_drive_t *drive)
 static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	atapi_bcount_t bcount;
+	ide_hwif_t *hwif = drive->hwif;
 	atapi_ireason_t ireason;
 	idefloppy_pc_t *pc = floppy->pc;
 	struct request *rq = pc->rq;
 	unsigned int temp;
+	u16 bcount;
 	u8 stat;
 
 	debug_log(KERN_INFO "ide-floppy: Reached %s interrupt handler\n",
@@ -848,8 +849,8 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	}
 
 	/* Get the number of bytes to transfer */
-	bcount.b.high = HWIF(drive)->INB(IDE_BCOUNTH_REG);
-	bcount.b.low = HWIF(drive)->INB(IDE_BCOUNTL_REG);
+	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
+		  hwif->INB(IDE_BCOUNTL_REG);
 	/* on this interrupt */
 	ireason.all = HWIF(drive)->INB(IDE_IREASON_REG);
 
@@ -867,13 +868,13 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	}
 	if (!test_bit(PC_WRITING, &pc->flags)) {
 		/* Reading - Check that we have enough space */
-		temp = pc->actually_transferred + bcount.all;
+		temp = pc->actually_transferred + bcount;
 		if (temp > pc->request_transfer) {
 			if (temp > pc->buffer_size) {
 				printk(KERN_ERR "ide-floppy: The floppy wants "
 					"to send us more data than expected "
 					"- discarding data\n");
-				idefloppy_discard_data(drive,bcount.all);
+				idefloppy_discard_data(drive, bcount);
 				BUG_ON(HWGROUP(drive)->handler != NULL);
 				ide_set_handler(drive,
 						&idefloppy_pc_intr,
@@ -889,23 +890,21 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	if (test_bit(PC_WRITING, &pc->flags)) {
 		if (pc->buffer != NULL)
 			/* Write the current buffer */
-			HWIF(drive)->atapi_output_bytes(drive,
-						pc->current_position,
-						bcount.all);
+			hwif->atapi_output_bytes(drive, pc->current_position,
+						 bcount);
 		else
-			idefloppy_output_buffers(drive, pc, bcount.all);
+			idefloppy_output_buffers(drive, pc, bcount);
 	} else {
 		if (pc->buffer != NULL)
 			/* Read the current buffer */
-			HWIF(drive)->atapi_input_bytes(drive,
-						pc->current_position,
-						bcount.all);
+			hwif->atapi_input_bytes(drive, pc->current_position,
+						bcount);
 		else
-			idefloppy_input_buffers(drive, pc, bcount.all);
+			idefloppy_input_buffers(drive, pc, bcount);
 	}
 	/* Update the current position */
-	pc->actually_transferred += bcount.all;
-	pc->current_position += bcount.all;
+	pc->actually_transferred += bcount;
+	pc->current_position += bcount;
 
 	BUG_ON(HWGROUP(drive)->handler != NULL);
 	ide_set_handler(drive, &idefloppy_pc_intr, IDEFLOPPY_WAIT_CMD, NULL);		/* And set the interrupt handler again */
@@ -1019,8 +1018,8 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	ide_hwif_t *hwif = drive->hwif;
-	atapi_bcount_t bcount;
 	ide_handler_t *pkt_xfer_routine;
+	u16 bcount;
 	u8 dma;
 
 	if (floppy->failed_pc == NULL &&
@@ -1059,7 +1058,7 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 	/* We haven't transferred any data yet */
 	pc->actually_transferred = 0;
 	pc->current_position = pc->buffer;
-	bcount.all = min(pc->request_transfer, 63 * 1024);
+	bcount = min(pc->request_transfer, 63 * 1024);
 
 	if (test_and_clear_bit(PC_DMA_ERROR, &pc->flags))
 		ide_dma_off(drive);
@@ -1073,8 +1072,8 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
 	/* Use PIO/DMA */
 	hwif->OUTB(dma, IDE_FEATURE_REG);
-	HWIF(drive)->OUTB(bcount.b.high, IDE_BCOUNTH_REG);
-	HWIF(drive)->OUTB(bcount.b.low, IDE_BCOUNTL_REG);
+	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
+	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
 	HWIF(drive)->OUTB(drive->select.all, IDE_SELECT_REG);
 
 	if (dma) {	/* Begin DMA, if necessary */
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 2c03f469f06e..4c24e185ccbf 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1847,13 +1847,13 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	atapi_bcount_t bcount;
 	atapi_ireason_t ireason;
 	idetape_pc_t *pc = tape->pc;
 	unsigned int temp;
 #if SIMULATE_ERRORS
 	static int error_sim_count = 0;
 #endif
+	u16 bcount;
 	u8 stat;
 
 #if IDETAPE_DEBUG_LOG
@@ -1962,8 +1962,8 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 		return ide_do_reset(drive);
 	}
 	/* Get the number of bytes to transfer on this interrupt. */
-	bcount.b.high = hwif->INB(IDE_BCOUNTH_REG);
-	bcount.b.low = hwif->INB(IDE_BCOUNTL_REG);
+	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
+		  hwif->INB(IDE_BCOUNTL_REG);
 
 	ireason.all = hwif->INB(IDE_IREASON_REG);
 
@@ -1981,11 +1981,11 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 	}
 	if (!test_bit(PC_WRITING, &pc->flags)) {
 		/* Reading - Check that we have enough space */
-		temp = pc->actually_transferred + bcount.all;
+		temp = pc->actually_transferred + bcount;
 		if (temp > pc->request_transfer) {
 			if (temp > pc->buffer_size) {
 				printk(KERN_ERR "ide-tape: The tape wants to send us more data than expected - discarding data\n");
-				idetape_discard_data(drive, bcount.all);
+				idetape_discard_data(drive, bcount);
 				ide_set_handler(drive, &idetape_pc_intr, IDETAPE_WAIT_CMD, NULL);
 				return ide_started;
 			}
@@ -1997,23 +1997,26 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 	}
 	if (test_bit(PC_WRITING, &pc->flags)) {
 		if (pc->bh != NULL)
-			idetape_output_buffers(drive, pc, bcount.all);
+			idetape_output_buffers(drive, pc, bcount);
 		else
 			/* Write the current buffer */
-			HWIF(drive)->atapi_output_bytes(drive, pc->current_position, bcount.all);
+			hwif->atapi_output_bytes(drive, pc->current_position,
+						 bcount);
 	} else {
 		if (pc->bh != NULL)
-			idetape_input_buffers(drive, pc, bcount.all);
+			idetape_input_buffers(drive, pc, bcount);
 		else
 			/* Read the current buffer */
-			HWIF(drive)->atapi_input_bytes(drive, pc->current_position, bcount.all);
+			hwif->atapi_input_bytes(drive, pc->current_position,
+						bcount);
 	}
 	/* Update the current position */
-	pc->actually_transferred += bcount.all;
-	pc->current_position += bcount.all;
+	pc->actually_transferred += bcount;
+	pc->current_position += bcount;
 #if IDETAPE_DEBUG_LOG
 	if (tape->debug_level >= 2)
-		printk(KERN_INFO "ide-tape: [cmd %x] transferred %d bytes on that interrupt\n", pc->c[0], bcount.all);
+		printk(KERN_INFO "ide-tape: [cmd %x] transferred %d bytes "
+				 "on that interrupt\n", pc->c[0], bcount);
 #endif
 	/* And set the interrupt handler again */
 	ide_set_handler(drive, &idetape_pc_intr, IDETAPE_WAIT_CMD, NULL);
@@ -2109,8 +2112,8 @@ static ide_startstop_t idetape_issue_packet_command (ide_drive_t *drive, idetape
 {
 	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	atapi_bcount_t bcount;
 	int dma_ok = 0;
+	u16 bcount;
 
 #if IDETAPE_DEBUG_BUGS
 	if (tape->pc->c[0] == IDETAPE_REQUEST_SENSE_CMD &&
@@ -2159,7 +2162,7 @@ static ide_startstop_t idetape_issue_packet_command (ide_drive_t *drive, idetape
 	pc->actually_transferred = 0;
 	pc->current_position = pc->buffer;
 	/* Request to transfer the entire buffer at once */
-	bcount.all = pc->request_transfer;
+	bcount = pc->request_transfer;
 
 	if (test_and_clear_bit(PC_DMA_ERROR, &pc->flags)) {
 		printk(KERN_WARNING "ide-tape: DMA disabled, "
@@ -2172,8 +2175,8 @@ static ide_startstop_t idetape_issue_packet_command (ide_drive_t *drive, idetape
 	if (IDE_CONTROL_REG)
 		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
 	hwif->OUTB(dma_ok ? 1 : 0, IDE_FEATURE_REG);	/* Use PIO/DMA */
-	hwif->OUTB(bcount.b.high, IDE_BCOUNTH_REG);
-	hwif->OUTB(bcount.b.low, IDE_BCOUNTL_REG);
+	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
+	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
 	hwif->OUTB(drive->select.all, IDE_SELECT_REG);
 	if (dma_ok)			/* Will begin DMA later */
 		set_bit(PC_DMA_IN_PROGRESS, &pc->flags);
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index c009f2351346..77e8a81228fc 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -395,11 +395,12 @@ static int idescsi_expiry(ide_drive_t *drive)
 static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	idescsi_pc_t *pc=scsi->pc;
+	ide_hwif_t *hwif = drive->hwif;
+	idescsi_pc_t *pc = scsi->pc;
 	struct request *rq = pc->rq;
-	atapi_bcount_t bcount;
 	atapi_ireason_t ireason;
 	unsigned int temp;
+	u16 bcount;
 	u8 stat;
 
 #if IDESCSI_DEBUG_LOG
@@ -436,8 +437,8 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 		idescsi_end_request (drive, 1, 0);
 		return ide_stopped;
 	}
-	bcount.b.low	= HWIF(drive)->INB(IDE_BCOUNTL_REG);
-	bcount.b.high	= HWIF(drive)->INB(IDE_BCOUNTH_REG);
+	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
+		  hwif->INB(IDE_BCOUNTL_REG);
 	ireason.all	= HWIF(drive)->INB(IDE_IREASON_REG);
 
 	if (ireason.b.cod) {
@@ -445,7 +446,7 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 		return ide_do_reset (drive);
 	}
 	if (ireason.b.io) {
-		temp = pc->actually_transferred + bcount.all;
+		temp = pc->actually_transferred + bcount;
 		if (temp > pc->request_transfer) {
 			if (temp > pc->buffer_size) {
 				printk(KERN_ERR "ide-scsi: The scsi wants to "
@@ -458,11 +459,13 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 						idescsi_input_buffers(drive, pc, temp);
 					else
 						drive->hwif->atapi_input_bytes(drive, pc->current_position, temp);
-					printk(KERN_ERR "ide-scsi: transferred %d of %d bytes\n", temp, bcount.all);
+					printk(KERN_ERR "ide-scsi: transferred"
+							" %d of %d bytes\n",
+							temp, bcount);
 				}
 				pc->actually_transferred += temp;
 				pc->current_position += temp;
-				idescsi_discard_data(drive, bcount.all - temp);
+				idescsi_discard_data(drive, bcount - temp);
 				ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
 				return ide_started;
 			}
@@ -474,19 +477,21 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 	if (ireason.b.io) {
 		clear_bit(PC_WRITING, &pc->flags);
 		if (pc->sg)
-			idescsi_input_buffers(drive, pc, bcount.all);
+			idescsi_input_buffers(drive, pc, bcount);
 		else
-			HWIF(drive)->atapi_input_bytes(drive, pc->current_position, bcount.all);
+			hwif->atapi_input_bytes(drive, pc->current_position,
+						bcount);
 	} else {
 		set_bit(PC_WRITING, &pc->flags);
 		if (pc->sg)
-			idescsi_output_buffers (drive, pc, bcount.all);
+			idescsi_output_buffers(drive, pc, bcount);
 		else
-			HWIF(drive)->atapi_output_bytes(drive, pc->current_position, bcount.all);
+			hwif->atapi_output_bytes(drive, pc->current_position,
+						 bcount);
 	}
 	/* Update the current position */
-	pc->actually_transferred += bcount.all;
-	pc->current_position += bcount.all;
+	pc->actually_transferred += bcount;
+	pc->current_position += bcount;
 
 	/* And set the interrupt handler again */
 	ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
@@ -570,13 +575,14 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
 	ide_hwif_t *hwif = drive->hwif;
-	atapi_bcount_t bcount;
+	u16 bcount;
 	u8 dma = 0;
 
 	scsi->pc=pc;							/* Set the current packet command */
 	pc->actually_transferred=0;					/* We haven't transferred any data yet */
 	pc->current_position=pc->buffer;
-	bcount.all = min(pc->request_transfer, 63 * 1024);		/* Request to transfer the entire buffer at once */
+	/* Request to transfer the entire buffer at once */
+	bcount = min(pc->request_transfer, 63 * 1024);
 
 	if (drive->using_dma && !idescsi_map_sg(drive, pc)) {
 		hwif->sg_mapped = 1;
@@ -589,8 +595,8 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
 
 	hwif->OUTB(dma, IDE_FEATURE_REG);
-	HWIF(drive)->OUTB(bcount.b.high, IDE_BCOUNTH_REG);
-	HWIF(drive)->OUTB(bcount.b.low, IDE_BCOUNTL_REG);
+	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
+	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
 
 	if (dma)
 		set_bit(PC_DMA_OK, &pc->flags);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index a638dde17e77..cf1a5aaebd9e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -315,26 +315,6 @@ typedef union {
 	} b;
 } special_t;
 
-/*
- * ATA DATA Register Special.
- * ATA NSECTOR Count Register().
- * ATAPI Byte Count Register.
- */
-typedef union {
-	unsigned all			:16;
-	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-		unsigned low		:8;	/* LSB */
-		unsigned high		:8;	/* MSB */
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned high		:8;	/* MSB */
-		unsigned low		:8;	/* LSB */
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
-	} b;
-} ata_nsector_t, ata_data_t, atapi_bcount_t;
-
 /*
  * ATA-IDE Select Register, aka Device-Head
  *
-- 
cgit v1.2.3


From 8e7657ae0f56c14882e53ffdae8055c2b1624de1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:12 +0100
Subject: ide: remove atapi_ireason_t (take 3)

Remove atapi_ireason_t.

While at it:
* replace 'HWIF(drive)' by 'drive->hwif' (or just 'hwif' where possible)

v2:
* v1 had CD and IO bits reversed in many places.

* Use CD and IO defines from <linux/hdreg.h>.

v3:
* Fix incorrect "(ireason & IO) == test_bit()". (Noticed by Sergei)

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 25 ++++++++++++-------------
 drivers/ide/ide-tape.c   | 27 +++++++++++++--------------
 drivers/scsi/ide-scsi.c  | 17 ++++++++---------
 include/linux/ide.h      | 24 ------------------------
 4 files changed, 33 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 239aebcfc359..830186fdcc74 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -788,12 +788,11 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	ide_hwif_t *hwif = drive->hwif;
-	atapi_ireason_t ireason;
 	idefloppy_pc_t *pc = floppy->pc;
 	struct request *rq = pc->rq;
 	unsigned int temp;
 	u16 bcount;
-	u8 stat;
+	u8 stat, ireason;
 
 	debug_log(KERN_INFO "ide-floppy: Reached %s interrupt handler\n",
 		__FUNCTION__);
@@ -852,18 +851,18 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
 		  hwif->INB(IDE_BCOUNTL_REG);
 	/* on this interrupt */
-	ireason.all = HWIF(drive)->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(IDE_IREASON_REG);
 
-	if (ireason.b.cod) {
+	if (ireason & CD) {
 		printk(KERN_ERR "ide-floppy: CoD != 0 in idefloppy_pc_intr\n");
 		return ide_do_reset(drive);
 	}
-	if (ireason.b.io == test_bit(PC_WRITING, &pc->flags)) {
+	if (((ireason & IO) == IO) == test_bit(PC_WRITING, &pc->flags)) {
 		/* Hopefully, we will never get here */
 		printk(KERN_ERR "ide-floppy: We wanted to %s, ",
-			ireason.b.io ? "Write":"Read");
+				(ireason & IO) ? "Write" : "Read");
 		printk(KERN_ERR "but the floppy wants us to %s !\n",
-			ireason.b.io ? "Read":"Write");
+				(ireason & IO) ? "Read" : "Write");
 		return ide_do_reset(drive);
 	}
 	if (!test_bit(PC_WRITING, &pc->flags)) {
@@ -920,15 +919,15 @@ static ide_startstop_t idefloppy_transfer_pc (ide_drive_t *drive)
 {
 	ide_startstop_t startstop;
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	atapi_ireason_t ireason;
+	u8 ireason;
 
 	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
 		printk(KERN_ERR "ide-floppy: Strange, packet command "
 				"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason.all = HWIF(drive)->INB(IDE_IREASON_REG);
-	if (!ireason.b.cod || ireason.b.io) {
+	ireason = drive->hwif->INB(IDE_IREASON_REG);
+	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-floppy: (IO,CoD) != (0,1) while "
 				"issuing a packet command\n");
 		return ide_do_reset(drive);
@@ -968,15 +967,15 @@ static ide_startstop_t idefloppy_transfer_pc1 (ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	ide_startstop_t startstop;
-	atapi_ireason_t ireason;
+	u8 ireason;
 
 	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
 		printk(KERN_ERR "ide-floppy: Strange, packet command "
 				"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason.all = HWIF(drive)->INB(IDE_IREASON_REG);
-	if (!ireason.b.cod || ireason.b.io) {
+	ireason = drive->hwif->INB(IDE_IREASON_REG);
+	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-floppy: (IO,CoD) != (0,1) "
 				"while issuing a packet command\n");
 		return ide_do_reset(drive);
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 4c24e185ccbf..3539131f23f4 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1847,14 +1847,13 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	atapi_ireason_t ireason;
 	idetape_pc_t *pc = tape->pc;
 	unsigned int temp;
 #if SIMULATE_ERRORS
 	static int error_sim_count = 0;
 #endif
 	u16 bcount;
-	u8 stat;
+	u8 stat, ireason;
 
 #if IDETAPE_DEBUG_LOG
 	if (tape->debug_level >= 4)
@@ -1965,18 +1964,18 @@ static ide_startstop_t idetape_pc_intr (ide_drive_t *drive)
 	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
 		  hwif->INB(IDE_BCOUNTL_REG);
 
-	ireason.all = hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(IDE_IREASON_REG);
 
-	if (ireason.b.cod) {
+	if (ireason & CD) {
 		printk(KERN_ERR "ide-tape: CoD != 0 in idetape_pc_intr\n");
 		return ide_do_reset(drive);
 	}
-	if (ireason.b.io == test_bit(PC_WRITING, &pc->flags)) {
+	if (((ireason & IO) == IO) == test_bit(PC_WRITING, &pc->flags)) {
 		/* Hopefully, we will never get here */
 		printk(KERN_ERR "ide-tape: We wanted to %s, ",
-			ireason.b.io ? "Write":"Read");
+				(ireason & IO) ? "Write" : "Read");
 		printk(KERN_ERR "ide-tape: but the tape wants us to %s !\n",
-			ireason.b.io ? "Read":"Write");
+				(ireason & IO) ? "Read" : "Write");
 		return ide_do_reset(drive);
 	}
 	if (!test_bit(PC_WRITING, &pc->flags)) {
@@ -2070,28 +2069,28 @@ static ide_startstop_t idetape_transfer_pc(ide_drive_t *drive)
 	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
 	idetape_pc_t *pc = tape->pc;
-	atapi_ireason_t ireason;
 	int retries = 100;
 	ide_startstop_t startstop;
+	u8 ireason;
 
 	if (ide_wait_stat(&startstop,drive,DRQ_STAT,BUSY_STAT,WAIT_READY)) {
 		printk(KERN_ERR "ide-tape: Strange, packet command initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason.all = hwif->INB(IDE_IREASON_REG);
-	while (retries-- && (!ireason.b.cod || ireason.b.io)) {
+	ireason = hwif->INB(IDE_IREASON_REG);
+	while (retries-- && ((ireason & CD) == 0 || (ireason & IO))) {
 		printk(KERN_ERR "ide-tape: (IO,CoD != (0,1) while issuing "
 				"a packet command, retrying\n");
 		udelay(100);
-		ireason.all = hwif->INB(IDE_IREASON_REG);
+		ireason = hwif->INB(IDE_IREASON_REG);
 		if (retries == 0) {
 			printk(KERN_ERR "ide-tape: (IO,CoD != (0,1) while "
 					"issuing a packet command, ignoring\n");
-			ireason.b.cod = 1;
-			ireason.b.io = 0;
+			ireason |= CD;
+			ireason &= ~IO;
 		}
 	}
-	if (!ireason.b.cod || ireason.b.io) {
+	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-tape: (IO,CoD) != (0,1) while issuing "
 				"a packet command\n");
 		return ide_do_reset(drive);
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 77e8a81228fc..ab7e8642cb88 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -398,10 +398,9 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 	ide_hwif_t *hwif = drive->hwif;
 	idescsi_pc_t *pc = scsi->pc;
 	struct request *rq = pc->rq;
-	atapi_ireason_t ireason;
 	unsigned int temp;
 	u16 bcount;
-	u8 stat;
+	u8 stat, ireason;
 
 #if IDESCSI_DEBUG_LOG
 	printk (KERN_INFO "ide-scsi: Reached idescsi_pc_intr interrupt handler\n");
@@ -439,13 +438,13 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 	}
 	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
 		  hwif->INB(IDE_BCOUNTL_REG);
-	ireason.all	= HWIF(drive)->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(IDE_IREASON_REG);
 
-	if (ireason.b.cod) {
+	if (ireason & CD) {
 		printk(KERN_ERR "ide-scsi: CoD != 0 in idescsi_pc_intr\n");
 		return ide_do_reset (drive);
 	}
-	if (ireason.b.io) {
+	if (ireason & IO) {
 		temp = pc->actually_transferred + bcount;
 		if (temp > pc->request_transfer) {
 			if (temp > pc->buffer_size) {
@@ -474,7 +473,7 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 #endif /* IDESCSI_DEBUG_LOG */
 		}
 	}
-	if (ireason.b.io) {
+	if (ireason & IO) {
 		clear_bit(PC_WRITING, &pc->flags);
 		if (pc->sg)
 			idescsi_input_buffers(drive, pc, bcount);
@@ -503,16 +502,16 @@ static ide_startstop_t idescsi_transfer_pc(ide_drive_t *drive)
 	ide_hwif_t *hwif = drive->hwif;
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
 	idescsi_pc_t *pc = scsi->pc;
-	atapi_ireason_t ireason;
 	ide_startstop_t startstop;
+	u8 ireason;
 
 	if (ide_wait_stat(&startstop,drive,DRQ_STAT,BUSY_STAT,WAIT_READY)) {
 		printk(KERN_ERR "ide-scsi: Strange, packet command "
 			"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason.all	= HWIF(drive)->INB(IDE_IREASON_REG);
-	if (!ireason.b.cod || ireason.b.io) {
+	ireason = hwif->INB(IDE_IREASON_REG);
+	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-scsi: (IO,CoD) != (0,1) while "
 				"issuing a packet command\n");
 		return ide_do_reset (drive);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index cf1a5aaebd9e..790ffa7f6949 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -345,30 +345,6 @@ typedef union {
 	} b;
 } select_t, ata_select_t;
 
-/*
- * ATAPI Interrupt Reason Register.
- *
- * cod		: Information transferred is command (1) or data (0)
- * io		: The device requests us to read (1) or write (0)
- * reserved	: Reserved
- */
-typedef union {
-	unsigned all			:8;
-	struct {
-#if defined(__LITTLE_ENDIAN_BITFIELD)
-		unsigned cod		:1;
-		unsigned io		:1;
-		unsigned reserved	:6;
-#elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned reserved	:6;
-		unsigned io		:1;
-		unsigned cod		:1;
-#else
-#error "Please fix <asm/byteorder.h>"
-#endif
-	} b;
-} atapi_ireason_t;
-
 /*
  * Status returned from various ide_ functions
  */
-- 
cgit v1.2.3


From 2fc573881957337c4ea1c84b92d2f27d076cad57 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:13 +0100
Subject: ide: add ide_pktcmd_tf_load() helper

Add ide_pktcmd_tf_load() helper and convert ATAPI device drivers to use it.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     | 10 ++--------
 drivers/ide/ide-floppy.c |  9 ++-------
 drivers/ide/ide-io.c     | 16 ++++++++++++++++
 drivers/ide/ide-tape.c   |  9 +++------
 drivers/scsi/ide-scsi.c  |  6 +-----
 include/linux/ide.h      |  2 ++
 6 files changed, 26 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 855a23c1c815..44b033ec0ab0 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -922,14 +922,8 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
 		info->dma = !hwif->dma_setup(drive);
 
 	/* Set up the controller registers. */
-	if (IDE_CONTROL_REG)
-		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
-	HWIF(drive)->OUTB(info->dma, IDE_FEATURE_REG);
-	HWIF(drive)->OUTB(0, IDE_IREASON_REG);
-	HWIF(drive)->OUTB(0, IDE_SECTOR_REG);
-
-	HWIF(drive)->OUTB(xferlen & 0xff, IDE_BCOUNTL_REG);
-	HWIF(drive)->OUTB(xferlen >> 8  , IDE_BCOUNTH_REG);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL |
+			   IDE_TFLAG_NO_SELECT_MASK, xferlen, info->dma);
  
 	if (CDROM_CONFIG_FLAGS (drive)->drq_interrupt) {
 		/* waiting for CDB interrupt, not DMA yet. */
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index d4fd064cc76c..ff8232ef9659 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -1067,13 +1067,8 @@ static ide_startstop_t idefloppy_issue_pc (ide_drive_t *drive, idefloppy_pc_t *p
 	if (test_bit(PC_DMA_RECOMMENDED, &pc->flags) && drive->using_dma)
 		dma = !hwif->dma_setup(drive);
 
-	if (IDE_CONTROL_REG)
-		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
-	/* Use PIO/DMA */
-	hwif->OUTB(dma, IDE_FEATURE_REG);
-	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
-	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
-	HWIF(drive)->OUTB(drive->select.all, IDE_SELECT_REG);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK |
+			   IDE_TFLAG_OUT_DEVICE, bcount, dma);
 
 	if (dma) {	/* Begin DMA, if necessary */
 		set_bit(PC_DMA_IN_PROGRESS, &pc->flags);
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index d36df77c7519..1ed7a8627cca 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1734,3 +1734,19 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 }
 
 EXPORT_SYMBOL(ide_do_drive_cmd);
+
+void ide_pktcmd_tf_load(ide_drive_t *drive, u32 tf_flags, u16 bcount, u8 dma)
+{
+	ide_task_t task;
+
+	memset(&task, 0, sizeof(task));
+	task.tf_flags = IDE_TFLAG_OUT_LBAH | IDE_TFLAG_OUT_LBAM |
+			IDE_TFLAG_OUT_FEATURE | tf_flags;
+	task.tf.feature = dma;		/* Use PIO/DMA */
+	task.tf.lbam    = bcount & 0xff;
+	task.tf.lbah    = (bcount >> 8) & 0xff;
+
+	ide_tf_load(drive, &task);
+}
+
+EXPORT_SYMBOL_GPL(ide_pktcmd_tf_load);
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index d1f2446739a4..3cbca3f4628a 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -2171,12 +2171,9 @@ static ide_startstop_t idetape_issue_packet_command (ide_drive_t *drive, idetape
 	if (test_bit(PC_DMA_RECOMMENDED, &pc->flags) && drive->using_dma)
 		dma_ok = !hwif->dma_setup(drive);
 
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
-	hwif->OUTB(dma_ok ? 1 : 0, IDE_FEATURE_REG);	/* Use PIO/DMA */
-	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
-	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
-	hwif->OUTB(drive->select.all, IDE_SELECT_REG);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK |
+			   IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
+
 	if (dma_ok)			/* Will begin DMA later */
 		set_bit(PC_DMA_IN_PROGRESS, &pc->flags);
 	if (test_bit(IDETAPE_DRQ_INTERRUPT, &tape->flags)) {
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 76f0ccd2968b..02e91893064d 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -590,12 +590,8 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 	}
 
 	SELECT_DRIVE(drive);
-	if (IDE_CONTROL_REG)
-		HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
 
-	hwif->OUTB(dma, IDE_FEATURE_REG);
-	hwif->OUTB(bcount & 0xff, IDE_BCOUNTL_REG);
-	hwif->OUTB((bcount >> 8) & 0xff, IDE_BCOUNTH_REG);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK, bcount, dma);
 
 	if (dma)
 		set_bit(PC_DMA_OK, &pc->flags);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 790ffa7f6949..20969eb1789a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -968,6 +968,8 @@ extern void QUIRK_LIST(ide_drive_t *);
 
 extern int drive_is_ready(ide_drive_t *);
 
+void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
+
 /*
  * taskfile io for disks for now...and builds request from ide_ioctl
  */
-- 
cgit v1.2.3


From cd3dbc99da337f2130f3cb2691fbb65c8bf22337 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:13 +0100
Subject: ide: remove QUIRK_LIST()

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-iops.c  | 6 ------
 drivers/ide/ide-probe.c | 5 ++++-
 include/linux/ide.h     | 1 -
 3 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 617888048eed..b6983b7b3b6c 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -189,12 +189,6 @@ void SELECT_MASK (ide_drive_t *drive, int mask)
 		HWIF(drive)->maskproc(drive, mask);
 }
 
-void QUIRK_LIST (ide_drive_t *drive)
-{
-	if (HWIF(drive)->quirkproc)
-		drive->quirk_list = HWIF(drive)->quirkproc(drive);
-}
-
 /*
  * Some localbus EIDE interfaces require a special access sequence
  * when using 32-bit I/O instructions to transfer data.  We call this
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 8e5d8dd315af..33e62d28465d 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -234,7 +234,10 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 
 	drive->media = ide_disk;
 	printk("%s DISK drive\n", (id->config == 0x848a) ? "CFA" : "ATA" );
-	QUIRK_LIST(drive);
+
+	if (hwif->quirkproc)
+		drive->quirk_list = hwif->quirkproc(drive);
+
 	return;
 
 err_misc:
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 20969eb1789a..7a144c9ee098 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -964,7 +964,6 @@ extern u32 ide_read_24(ide_drive_t *);
 extern void SELECT_DRIVE(ide_drive_t *);
 extern void SELECT_INTERRUPT(ide_drive_t *);
 extern void SELECT_MASK(ide_drive_t *, int);
-extern void QUIRK_LIST(ide_drive_t *);
 
 extern int drive_is_ready(ide_drive_t *);
 
-- 
cgit v1.2.3


From f919790f8c929ab1b392ad1a0c2e1b53337b5071 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:13 +0100
Subject: ide: remove SELECT_INTERRUPT()

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 5 ++++-
 drivers/ide/ide-iops.c | 8 --------
 include/linux/ide.h    | 1 -
 3 files changed, 4 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1ed7a8627cca..6b70ab9566d9 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1201,7 +1201,10 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 		    hwif != hwgroup->hwif &&
 		    hwif->io_ports[IDE_CONTROL_OFFSET]) {
 			/* set nIEN for previous hwif */
-			SELECT_INTERRUPT(drive);
+			if (hwif->intrproc)
+				hwif->intrproc(drive);
+			else
+				hwif->OUTB(drive->ctl | 2, IDE_CONTROL_REG);
 		}
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index b6983b7b3b6c..106454211cb8 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -175,14 +175,6 @@ void SELECT_DRIVE (ide_drive_t *drive)
 
 EXPORT_SYMBOL(SELECT_DRIVE);
 
-void SELECT_INTERRUPT (ide_drive_t *drive)
-{
-	if (HWIF(drive)->intrproc)
-		HWIF(drive)->intrproc(drive);
-	else
-		HWIF(drive)->OUTB(drive->ctl|2, IDE_CONTROL_REG);
-}
-
 void SELECT_MASK (ide_drive_t *drive, int mask)
 {
 	if (HWIF(drive)->maskproc)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 7a144c9ee098..e80351878b8b 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -962,7 +962,6 @@ void ide_tf_load(ide_drive_t *, ide_task_t *);
 extern u32 ide_read_24(ide_drive_t *);
 
 extern void SELECT_DRIVE(ide_drive_t *);
-extern void SELECT_INTERRUPT(ide_drive_t *);
 extern void SELECT_MASK(ide_drive_t *, int);
 
 extern int drive_is_ready(ide_drive_t *);
-- 
cgit v1.2.3


From 7299a3918442dc9a5abb71b9f65b1dd17637c8c0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:14 +0100
Subject: ide: remove hwif->intrproc

Given that:

* hpt366.c::hpt3xx_intrproc() is the only user of hwif->intrproc

* hpt366.c::hpt3xx_quirkproc() sets drive->quirk_list to 1 for quirky drives
  which is a value unique to hpt366 host driver

we can remove hwif->intproc and just check for drive->quirk_list == 1
in ide_do_request().

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c      |  9 +++++----
 drivers/ide/ide.c         |  1 -
 drivers/ide/pci/hpt366.c  | 10 ----------
 drivers/ide/pci/sgiioc4.c |  1 -
 include/linux/ide.h       |  2 --
 5 files changed, 5 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 6b70ab9566d9..6ee7458d34ed 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1200,10 +1200,11 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 		if (hwgroup->hwif->sharing_irq &&
 		    hwif != hwgroup->hwif &&
 		    hwif->io_ports[IDE_CONTROL_OFFSET]) {
-			/* set nIEN for previous hwif */
-			if (hwif->intrproc)
-				hwif->intrproc(drive);
-			else
+			/*
+			 * set nIEN for previous hwif, drives in the
+			 * quirk_list may not like intr setups/cleanups
+			 */
+			if (drive->quirk_list != 1)
 				hwif->OUTB(drive->ctl | 2, IDE_CONTROL_REG);
 		}
 		hwgroup->hwif = hwif;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 9ab5458fe950..4acd87e92cca 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -424,7 +424,6 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 	hwif->reset_poll		= tmp_hwif->reset_poll;
 	hwif->pre_reset			= tmp_hwif->pre_reset;
 	hwif->resetproc			= tmp_hwif->resetproc;
-	hwif->intrproc			= tmp_hwif->intrproc;
 	hwif->maskproc			= tmp_hwif->maskproc;
 	hwif->quirkproc			= tmp_hwif->quirkproc;
 	hwif->busproc			= tmp_hwif->busproc;
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index d3826a66834d..24d645751e0f 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -736,15 +736,6 @@ static int hpt3xx_quirkproc(ide_drive_t *drive)
 	return 0;
 }
 
-static void hpt3xx_intrproc(ide_drive_t *drive)
-{
-	if (drive->quirk_list)
-		return;
-
-	/* drives in the quirk_list may not like intr setups/cleanups */
-	outb(drive->ctl | 2, IDE_CONTROL_REG);
-}
-
 static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
@@ -1298,7 +1289,6 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
 	hwif->set_dma_mode	= &hpt3xx_set_mode;
 
 	hwif->quirkproc		= &hpt3xx_quirkproc;
-	hwif->intrproc		= &hpt3xx_intrproc;
 	hwif->maskproc		= &hpt3xx_maskproc;
 	hwif->busproc		= &hpt3xx_busproc;
 
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index de820aa58cd0..7e9dade5648d 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -582,7 +582,6 @@ ide_init_sgiioc4(ide_hwif_t * hwif)
 	hwif->pre_reset = NULL;	/* No HBA specific pre_set needed */
 	hwif->resetproc = &sgiioc4_resetproc;/* Reset DMA engine,
 						clear interrupts */
-	hwif->intrproc = NULL;	/* Enable or Disable interrupt from drive */
 	hwif->maskproc = &sgiioc4_maskproc;	/* Mask on/off NIEN register */
 	hwif->quirkproc = NULL;
 	hwif->busproc = NULL;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e80351878b8b..90f83b65eb8f 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -524,8 +524,6 @@ typedef struct hwif_s {
 	void	(*pre_reset)(ide_drive_t *);
 	/* routine to reset controller after a disk reset */
 	void	(*resetproc)(ide_drive_t *);
-	/* special interrupt handling for shared pci interrupts */
-	void	(*intrproc)(ide_drive_t *);
 	/* special host masking for drive selection */
 	void	(*maskproc)(ide_drive_t *, int);
 	/* check host's drive quirk list */
-- 
cgit v1.2.3


From ac026ff254b32915bb14ba97a23b4019d137f181 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:14 +0100
Subject: ide: remove 'command_type' field from ide_task_t

* Add 'data_buf' and 'nsect' variables in ide_taskfile_ioctl()
  to cache data buffer pointer and number of sectors to transfer
  (this allows us to have only one ide_diag_taskfile() call).

* Add IDE_TFLAG_WRITE taskfile flag and use it to check whether
  the REQ_RW request flag should be set.

* Move ->command_type handling from ide_diag_taskfile() to
  ide_taskfile_ioctl() and use ->req_cmd instead of ->command_type.

* Add 'nsect' parameter to ide_raw_taskfile().

* Merge ide_diag_taskfile() into ide_raw_taskfile().

* Initialize ->data_phase explicitly in idedisk_prepare_flush(),
  ide_start_power_step() and ide_disk_special().

* Remove no longer needed 'command_type' field from ide_task_t.

* Add #ifndef/#endif __KERNEL__ to <linux/hdreg.h> around no
  longer used by kernel IDE_DRIVE_TASK_* and TASKFILE_* defines.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     | 15 ++++----
 drivers/ide/ide-io.c       |  8 ++---
 drivers/ide/ide-taskfile.c | 89 +++++++++++++++++++++++-----------------------
 include/linux/hdreg.h      | 13 +++----
 include/linux/ide.h        |  5 ++-
 5 files changed, 64 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 747dc6023346..d9a4fe27685d 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -516,12 +516,11 @@ static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
 	tf->lbam    = SMART_LCYL_PASS;
 	tf->lbah    = SMART_HCYL_PASS;
 	tf->command = WIN_SMART;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
-	args.command_type			= IDE_DRIVE_TASK_IN;
-	args.data_phase				= TASKFILE_IN;
-	args.handler				= &task_in_intr;
+	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.data_phase	= TASKFILE_IN;
+	args.handler	= task_in_intr;
 	(void) smart_enable(drive);
-	return ide_raw_taskfile(drive, &args, buf);
+	return ide_raw_taskfile(drive, &args, buf, 1);
 }
 
 static int proc_idedisk_read_cache
@@ -607,9 +606,9 @@ static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
 		task.tf.command = WIN_FLUSH_CACHE_EXT;
 	else
 		task.tf.command = WIN_FLUSH_CACHE;
-	task.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
-	task.command_type = IDE_DRIVE_TASK_NO_DATA;
-	task.handler = task_no_data_intr;
+	task.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	task.data_phase	= TASKFILE_NO_DATA;
+	task.handler	= task_no_data_intr;
 
 	rq->cmd_type = REQ_TYPE_ATA_TASKFILE;
 	rq->cmd_flags |= REQ_SOFTBARRIER;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 6ee7458d34ed..f4f7e3db10ac 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -232,9 +232,9 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 	return ide_stopped;
 
 out_do_tf:
-	args->tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
-	args->command_type = IDE_DRIVE_TASK_NO_DATA;
-	args->handler      = task_no_data_intr;
+	args->tf_flags	 = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args->data_phase = TASKFILE_NO_DATA;
+	args->handler	 = task_no_data_intr;
 	return do_rw_taskfile(drive, args);
 }
 
@@ -672,7 +672,7 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 	ide_task_t args;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.command_type = IDE_DRIVE_TASK_NO_DATA;
+	args.data_phase = TASKFILE_NO_DATA;
 
 	if (s->b.set_geometry) {
 		s->b.set_geometry = 0;
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 7c8e9802898a..ff28449bbcaf 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -126,11 +126,10 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 		args.tf.command = WIN_IDENTIFY;
 	else
 		args.tf.command = WIN_PIDENTIFY;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
-	args.command_type = IDE_DRIVE_TASK_IN;
-	args.data_phase   = TASKFILE_IN;
-	args.handler	  = &task_in_intr;
-	return ide_raw_taskfile(drive, &args, buf);
+	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.data_phase	= TASKFILE_IN;
+	args.handler	= task_in_intr;
+	return ide_raw_taskfile(drive, &args, buf, 1);
 }
 
 static int inline task_dma_ok(ide_task_t *task)
@@ -499,7 +498,7 @@ ide_startstop_t pre_task_out_intr (ide_drive_t *drive, struct request *rq)
 }
 EXPORT_SYMBOL(pre_task_out_intr);
 
-static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long data_size, u8 *buf)
+int ide_raw_taskfile(ide_drive_t *drive, ide_task_t *task, u8 *buf, u16 nsect)
 {
 	struct request rq;
 
@@ -514,44 +513,26 @@ static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long
 	 * if we would find a solution to transfer any size.
 	 * To support special commands like READ LONG.
 	 */
-	if (args->command_type != IDE_DRIVE_TASK_NO_DATA) {
-		if (data_size == 0)
-			rq.nr_sectors = (args->tf.hob_nsect << 8) | args->tf.nsect;
-		else
-			rq.nr_sectors = data_size / SECTOR_SIZE;
-
-		if (!rq.nr_sectors) {
-			printk(KERN_ERR "%s: in/out command without data\n",
-					drive->name);
-			return -EFAULT;
-		}
+	rq.hard_nr_sectors = rq.nr_sectors = nsect;
+	rq.hard_cur_sectors = rq.current_nr_sectors = nsect;
 
-		rq.hard_nr_sectors = rq.nr_sectors;
-		rq.hard_cur_sectors = rq.current_nr_sectors = rq.nr_sectors;
+	if (task->tf_flags & IDE_TFLAG_WRITE)
+		rq.cmd_flags |= REQ_RW;
 
-		if (args->command_type == IDE_DRIVE_TASK_RAW_WRITE)
-			rq.cmd_flags |= REQ_RW;
-	}
+	rq.special = task;
+	task->rq = &rq;
 
-	rq.special = args;
-	args->rq = &rq;
 	return ide_do_drive_cmd(drive, &rq, ide_wait);
 }
 
-int ide_raw_taskfile (ide_drive_t *drive, ide_task_t *args, u8 *buf)
-{
-	return ide_diag_taskfile(drive, args, 0, buf);
-}
-
 EXPORT_SYMBOL(ide_raw_taskfile);
 
 int ide_no_data_taskfile(ide_drive_t *drive, ide_task_t *task)
 {
-	task->command_type = IDE_DRIVE_TASK_NO_DATA;
-	task->data_phase   = TASKFILE_NO_DATA;
-	task->handler      = task_no_data_intr;
+	task->data_phase = TASKFILE_NO_DATA;
+	task->handler    = task_no_data_intr;
 
-	return ide_raw_taskfile(drive, task, NULL);
+	return ide_raw_taskfile(drive, task, NULL, 0);
 }
 EXPORT_SYMBOL_GPL(ide_no_data_taskfile);
 
@@ -562,10 +543,12 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	ide_task_t		args;
 	u8 *outbuf		= NULL;
 	u8 *inbuf		= NULL;
+	u8 *data_buf		= NULL;
 	int err			= 0;
 	int tasksize		= sizeof(struct ide_task_request_s);
 	unsigned int taskin	= 0;
 	unsigned int taskout	= 0;
+	u16 nsect		= 0;
 	u8 io_32bit		= drive->io_32bit;
 	char __user *buf = (char __user *)arg;
 
@@ -618,7 +601,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	memcpy(&args.tf_array[6], req_task->io_ports, HDIO_DRIVE_TASK_HDR_SIZE);
 	args.tf_in_flags  = req_task->in_flags;
 	args.data_phase   = req_task->data_phase;
-	args.command_type = req_task->req_cmd;
 
 	args.tf_flags = IDE_TFLAG_OUT_DEVICE;
 	if (drive->addressing == 1)
@@ -657,14 +639,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	drive->io_32bit = 0;
 	switch(req_task->data_phase) {
-		case TASKFILE_OUT_DMAQ:
-		case TASKFILE_OUT_DMA:
-			err = ide_diag_taskfile(drive, &args, taskout, outbuf);
-			break;
-		case TASKFILE_IN_DMAQ:
-		case TASKFILE_IN_DMA:
-			err = ide_diag_taskfile(drive, &args, taskin, inbuf);
-			break;
 		case TASKFILE_MULTI_OUT:
 			if (!drive->mult_count) {
 				/* (hs): give up if multcount is not set */
@@ -678,7 +652,11 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 		case TASKFILE_OUT:
 			args.prehandler = &pre_task_out_intr;
 			args.handler = &task_out_intr;
-			err = ide_diag_taskfile(drive, &args, taskout, outbuf);
+			/* fall through */
+		case TASKFILE_OUT_DMAQ:
+		case TASKFILE_OUT_DMA:
+			nsect = taskout / SECTOR_SIZE;
+			data_buf = outbuf;
 			break;
 		case TASKFILE_MULTI_IN:
 			if (!drive->mult_count) {
@@ -692,17 +670,38 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			/* fall through */
 		case TASKFILE_IN:
 			args.handler = &task_in_intr;
-			err = ide_diag_taskfile(drive, &args, taskin, inbuf);
+			/* fall through */
+		case TASKFILE_IN_DMAQ:
+		case TASKFILE_IN_DMA:
+			nsect = taskin / SECTOR_SIZE;
+			data_buf = inbuf;
 			break;
 		case TASKFILE_NO_DATA:
 			args.handler = &task_no_data_intr;
-			err = ide_diag_taskfile(drive, &args, 0, NULL);
 			break;
 		default:
 			err = -EFAULT;
 			goto abort;
 	}
 
+	if (req_task->req_cmd == IDE_DRIVE_TASK_NO_DATA)
+		nsect = 0;
+	else if (!nsect) {
+		nsect = (args.tf.hob_nsect << 8) | args.tf.nsect;
+
+		if (!nsect) {
+			printk(KERN_ERR "%s: in/out command without data\n",
+					drive->name);
+			err = -EFAULT;
+			goto abort;
+		}
+	}
+
+	if (req_task->req_cmd == IDE_DRIVE_TASK_RAW_WRITE)
+		args.tf_flags |= IDE_TFLAG_WRITE;
+
+	err = ide_raw_taskfile(drive, &args, data_buf, nsect);
+
 	memcpy(req_task->hob_ports, &args.tf_array[0], HDIO_DRIVE_HOB_HDR_SIZE - 2);
 	memcpy(req_task->io_ports, &args.tf_array[6], HDIO_DRIVE_TASK_HDR_SIZE);
 	req_task->in_flags  = args.tf_in_flags;
diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h
index 0521f1234f15..ff43f8d6b5b3 100644
--- a/include/linux/hdreg.h
+++ b/include/linux/hdreg.h
@@ -73,13 +73,13 @@
 #define HDIO_DRIVE_HOB_HDR_SIZE		(8 * sizeof(__u8))
 #define HDIO_DRIVE_TASK_HDR_SIZE	(8 * sizeof(__u8))
 
-#define IDE_DRIVE_TASK_INVALID		-1
 #define IDE_DRIVE_TASK_NO_DATA		0
+#ifndef __KERNEL__
+#define IDE_DRIVE_TASK_INVALID		-1
 #define IDE_DRIVE_TASK_SET_XFER		1
-
 #define IDE_DRIVE_TASK_IN		2
-
 #define IDE_DRIVE_TASK_OUT		3
+#endif
 #define IDE_DRIVE_TASK_RAW_WRITE	4
 
 /*
@@ -166,9 +166,6 @@ typedef struct hd_drive_hob_hdr {
 } hob_struct_t;
 #endif
 
-#define TASKFILE_INVALID		0x7fff
-#define TASKFILE_48			0x8000
-
 #define TASKFILE_NO_DATA		0x0000
 
 #define TASKFILE_IN			0x0001
@@ -183,12 +180,16 @@ typedef struct hd_drive_hob_hdr {
 #define TASKFILE_IN_DMAQ		0x0080
 #define TASKFILE_OUT_DMAQ		0x0100
 
+#ifndef __KERNEL__
 #define TASKFILE_P_IN			0x0200
 #define TASKFILE_P_OUT			0x0400
 #define TASKFILE_P_IN_DMA		0x0800
 #define TASKFILE_P_OUT_DMA		0x1000
 #define TASKFILE_P_IN_DMAQ		0x2000
 #define TASKFILE_P_OUT_DMAQ		0x4000
+#define TASKFILE_48			0x8000
+#define TASKFILE_INVALID		0x7fff
+#endif
 
 /* ATA/ATAPI Commands pre T13 Spec */
 #define WIN_NOP				0x00
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 90f83b65eb8f..7485fc705ca4 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -909,6 +909,7 @@ enum {
 					  IDE_TFLAG_OUT_LBAM |
 					  IDE_TFLAG_OUT_LBAH,
 	IDE_TFLAG_OUT_DEVICE		= (1 << 14),
+	IDE_TFLAG_WRITE			= (1 << 15),
 };
 
 struct ide_taskfile {
@@ -948,7 +949,6 @@ typedef struct ide_task_s {
 	u16			tf_flags;
 	ide_reg_valid_t		tf_in_flags;
 	int			data_phase;
-	int			command_type;
 	ide_pre_handler_t	*prehandler;
 	ide_handler_t		*handler;
 	struct request		*rq;		/* copy of request */
@@ -983,8 +983,7 @@ extern ide_startstop_t task_no_data_intr(ide_drive_t *);
 extern ide_startstop_t task_in_intr(ide_drive_t *);
 extern ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
 
-extern int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *);
-
+int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *, u16);
 int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
 
 int ide_taskfile_ioctl(ide_drive_t *, unsigned int, unsigned long);
-- 
cgit v1.2.3


From 866e2ec9ce525de0e7c10d02ead8d85af27adffd Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:14 +0100
Subject: ide: remove 'tf_in_flags' field from ide_task_t

* Add IDE_TFLAG_IN_DATA taskfile flag to indicate the need of reading
  IDE_DATA_REG in ide_end_drive_cmd().

  Set the new flag in ide_taskfile_ioctl() if ->in_flags.b.data is set.

* Add IDE_TFLAG_FLAGGED_SET_IN_FLAGS taskfile flag to indicate the
  need of modifying ->in_flags in ide_taskfile_ioctl().

  Set the new flag in flagged_taskfile() and move the code modifying
  ->tf_in_flags to ide_taskfile_ioctl().

  While at it remove the bogus comment: ->tf_in_flags (except .b.data)
  have no effect on selection of registers to read.

* Remove no longer needed 'tf_in_flags' field from ide_task_t.

As the result we finally have the internals of HDIO_DRIVE_TASKFILE ioctl
separated from the core IDE code.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       |  2 +-
 drivers/ide/ide-taskfile.c | 29 +++++++++++++----------------
 include/linux/ide.h        |  5 +++--
 3 files changed, 17 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index f4f7e3db10ac..1112d8b049b9 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -340,7 +340,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 		if (args) {
 			struct ide_taskfile *tf = &args->tf;
 
-			if (args->tf_in_flags.b.data) {
+			if (args->tf_flags & IDE_TFLAG_IN_DATA) {
 				u16 data = hwif->INW(IDE_DATA_REG);
 
 				tf->data = data & 0xff;
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index ff28449bbcaf..03c4a0c55bf4 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -599,8 +599,8 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	memcpy(&args.tf_array[0], req_task->hob_ports, HDIO_DRIVE_HOB_HDR_SIZE - 2);
 	memcpy(&args.tf_array[6], req_task->io_ports, HDIO_DRIVE_TASK_HDR_SIZE);
-	args.tf_in_flags  = req_task->in_flags;
-	args.data_phase   = req_task->data_phase;
+
+	args.data_phase = req_task->data_phase;
 
 	args.tf_flags = IDE_TFLAG_OUT_DEVICE;
 	if (drive->addressing == 1)
@@ -637,6 +637,9 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			args.tf_flags |= IDE_TFLAG_OUT_HOB;
 	}
 
+	if (req_task->in_flags.b.data)
+		args.tf_flags |= IDE_TFLAG_IN_DATA;
+
 	drive->io_32bit = 0;
 	switch(req_task->data_phase) {
 		case TASKFILE_MULTI_OUT:
@@ -704,7 +707,13 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	memcpy(req_task->hob_ports, &args.tf_array[0], HDIO_DRIVE_HOB_HDR_SIZE - 2);
 	memcpy(req_task->io_ports, &args.tf_array[6], HDIO_DRIVE_TASK_HDR_SIZE);
-	req_task->in_flags  = args.tf_in_flags;
+
+	if ((args.tf_flags & IDE_TFLAG_FLAGGED_SET_IN_FLAGS) &&
+	    req_task->in_flags.all == 0) {
+		req_task->in_flags.all = IDE_TASKFILE_STD_IN_FLAGS;
+		if (drive->addressing == 1)
+			req_task->in_flags.all |= (IDE_HOB_STD_IN_FLAGS << 8);
+	}
 
 	if (copy_to_user(buf, req_task, tasksize)) {
 		err = -EFAULT;
@@ -846,19 +855,7 @@ ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
 		}
 	}
 
-	/*
-	 * (ks) Check taskfile in flags.
-	 * If set, then execute as it is defined.
-	 * If not set, then define default settings.
-	 * The default values are:
-	 *	read all taskfile registers (except data)
-	 *	read the hob registers (sector, nsector, lcyl, hcyl)
-	 */
-	if (task->tf_in_flags.all == 0) {
-		task->tf_in_flags.all = IDE_TASKFILE_STD_IN_FLAGS;
-		if (drive->addressing == 1)
-			task->tf_in_flags.all |= (IDE_HOB_STD_IN_FLAGS  << 8);
-        }
+	task->tf_flags |= IDE_TFLAG_FLAGGED_SET_IN_FLAGS;
 
 	return do_rw_taskfile(drive, task);
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 7485fc705ca4..c23ef2df2cbb 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -910,6 +910,8 @@ enum {
 					  IDE_TFLAG_OUT_LBAH,
 	IDE_TFLAG_OUT_DEVICE		= (1 << 14),
 	IDE_TFLAG_WRITE			= (1 << 15),
+	IDE_TFLAG_FLAGGED_SET_IN_FLAGS	= (1 << 16),
+	IDE_TFLAG_IN_DATA		= (1 << 17),
 };
 
 struct ide_taskfile {
@@ -946,8 +948,7 @@ typedef struct ide_task_s {
 		struct ide_taskfile	tf;
 		u8			tf_array[14];
 	};
-	u16			tf_flags;
-	ide_reg_valid_t		tf_in_flags;
+	u32			tf_flags;
 	int			data_phase;
 	ide_pre_handler_t	*prehandler;
 	ide_handler_t		*handler;
-- 
cgit v1.2.3


From 1edee60e9d994f2b9a79b1333be39790683541fe Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:15 +0100
Subject: ide: merge flagged_taskfile() into do_rw_taskfile()

Based on the earlier work by Tejun Heo.

task->data_phase == TASKFILE_MULTI_{IN,OUT} vs drive->mult_count == 0
check is needed also for ide_taskfile_ioctl() requests that don't have
IDE_TFLAG_FLAGGED taskfile flag set.

Cc: Tejun Heo <htejun@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       |  3 ---
 drivers/ide/ide-taskfile.c | 32 ++++++++++++--------------------
 include/linux/ide.h        |  5 -----
 3 files changed, 12 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1112d8b049b9..1af2cc4f864e 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -851,9 +851,6 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 			break;
 		}
 
-		if (task->tf_flags & IDE_TFLAG_FLAGGED)
-			return flagged_taskfile(drive, task);
-
 		return do_rw_taskfile(drive, task);
 	}
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 03c4a0c55bf4..1a34edb183ec 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -156,6 +156,18 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct ide_taskfile *tf = &task->tf;
 
+	if (task->data_phase == TASKFILE_MULTI_IN ||
+	    task->data_phase == TASKFILE_MULTI_OUT) {
+		if (!drive->mult_count) {
+			printk(KERN_ERR "%s: multimode not set!\n",
+					drive->name);
+			return ide_stopped;
+		}
+	}
+
+	if (task->tf_flags & IDE_TFLAG_FLAGGED)
+		task->tf_flags |= IDE_TFLAG_FLAGGED_SET_IN_FLAGS;
+
 	ide_tf_load(drive, task);
 
 	if (task->handler != NULL) {
@@ -839,23 +851,3 @@ int ide_task_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	return err;
 }
-
-/*
- * NOTICE: This is additions from IBM to provide a discrete interface,
- * for selective taskregister access operations.  Nice JOB Klaus!!!
- * Glad to be able to work and co-develop this with you and IBM.
- */
-ide_startstop_t flagged_taskfile (ide_drive_t *drive, ide_task_t *task)
-{
-	if (task->data_phase == TASKFILE_MULTI_IN ||
-	    task->data_phase == TASKFILE_MULTI_OUT) {
-		if (!drive->mult_count) {
-			printk(KERN_ERR "%s: multimode not set!\n", drive->name);
-			return ide_stopped;
-		}
-	}
-
-	task->tf_flags |= IDE_TFLAG_FLAGGED_SET_IN_FLAGS;
-
-	return do_rw_taskfile(drive, task);
-}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c23ef2df2cbb..849447572a6b 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -972,11 +972,6 @@ void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
  */
 extern ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
-/*
- * Special Flagged Register Validation Caller
- */
-extern ide_startstop_t flagged_taskfile(ide_drive_t *, ide_task_t *);
-
 extern ide_startstop_t set_multmode_intr(ide_drive_t *);
 extern ide_startstop_t set_geometry_intr(ide_drive_t *);
 extern ide_startstop_t recal_intr(ide_drive_t *);
-- 
cgit v1.2.3


From 10d90157c83d4b6743c9063c36f9e7f27aa254b6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:16 +0100
Subject: ide: convert do_rw_taskfile() to use ->data_phase

* Use task->data_phase in do_rw_taskfile() to decide what to do.

* task->prehandler is only used by TASKFILE[_MULTI]_OUT so just
  use pre_task_out_intr() directly and remove no longer needed
  'prehandler' field from ide_task_t.

* Remove no longer needed ide_pre_handler_t type.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-taskfile.c | 25 +++++++++++++------------
 include/linux/ide.h        |  2 --
 2 files changed, 13 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 1a34edb183ec..5f6d01a4222c 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -170,23 +170,25 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 
 	ide_tf_load(drive, task);
 
-	if (task->handler != NULL) {
-		if (task->prehandler != NULL) {
-			hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
-			ndelay(400);	/* FIXME */
-			return task->prehandler(drive, task->rq);
-		}
+	switch (task->data_phase) {
+	case TASKFILE_MULTI_OUT:
+	case TASKFILE_OUT:
+		hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
+		ndelay(400);	/* FIXME */
+		return pre_task_out_intr(drive, task->rq);
+	case TASKFILE_MULTI_IN:
+	case TASKFILE_IN:
+	case TASKFILE_NO_DATA:
 		ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
 		return ide_started;
-	}
-
-	if (task_dma_ok(task) && drive->using_dma && !hwif->dma_setup(drive)) {
+	default:
+		if (task_dma_ok(task) == 0 || drive->using_dma == 0 ||
+		    hwif->dma_setup(drive))
+			return ide_stopped;
 		hwif->dma_exec_cmd(drive, tf->command);
 		hwif->dma_start(drive);
 		return ide_started;
 	}
-
-	return ide_stopped;
 }
 
 /*
@@ -665,7 +667,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			}
 			/* fall through */
 		case TASKFILE_OUT:
-			args.prehandler = &pre_task_out_intr;
 			args.handler = &task_out_intr;
 			/* fall through */
 		case TASKFILE_OUT_DMAQ:
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 849447572a6b..2c28fb75915a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -626,7 +626,6 @@ typedef struct hwif_s {
 /*
  *  internal ide interrupt handler type
  */
-typedef ide_startstop_t (ide_pre_handler_t)(ide_drive_t *, struct request *);
 typedef ide_startstop_t (ide_handler_t)(ide_drive_t *);
 typedef int (ide_expiry_t)(ide_drive_t *);
 
@@ -950,7 +949,6 @@ typedef struct ide_task_s {
 	};
 	u32			tf_flags;
 	int			data_phase;
-	ide_pre_handler_t	*prehandler;
 	ide_handler_t		*handler;
 	struct request		*rq;		/* copy of request */
 	void			*special;	/* valid_t generally */
-- 
cgit v1.2.3


From 1192e528e064ebb9a578219731d2b0f78ca3c1ec Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:16 +0100
Subject: ide: use ->data_phase to set ->handler in do_rw_taskfile()

* Use ->data_phase to set ->handler in do_rw_taskfile() instead of
  setting ->handler in callers of ide_raw_taskfile()/do_rw_taskfile().

* Unexport task_no_data_intr() and make it static.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     |  2 --
 drivers/ide/ide-io.c       |  1 -
 drivers/ide/ide-taskfile.c | 16 ++++++++++------
 include/linux/ide.h        |  1 -
 4 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 424207e67f9c..a063957e9add 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -518,7 +518,6 @@ static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
 	tf->command = WIN_SMART;
 	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	args.data_phase	= TASKFILE_IN;
-	args.handler	= task_in_intr;
 	(void) smart_enable(drive);
 	return ide_raw_taskfile(drive, &args, buf, 1);
 }
@@ -608,7 +607,6 @@ static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
 		task.tf.command = WIN_FLUSH_CACHE;
 	task.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	task.data_phase	= TASKFILE_NO_DATA;
-	task.handler	= task_no_data_intr;
 
 	rq->cmd_type = REQ_TYPE_ATA_TASKFILE;
 	rq->cmd_flags |= REQ_SOFTBARRIER;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1af2cc4f864e..18ac1bd0811f 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -234,7 +234,6 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 out_do_tf:
 	args->tf_flags	 = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	args->data_phase = TASKFILE_NO_DATA;
-	args->handler	 = task_no_data_intr;
 	return do_rw_taskfile(drive, args);
 }
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 5f6d01a4222c..835465d61f70 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -128,7 +128,6 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 		args.tf.command = WIN_PIDENTIFY;
 	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
 	args.data_phase	= TASKFILE_IN;
-	args.handler	= task_in_intr;
 	return ide_raw_taskfile(drive, &args, buf, 1);
 }
 
@@ -151,6 +150,9 @@ static int inline task_dma_ok(ide_task_t *task)
 	return 0;
 }
 
+static ide_startstop_t task_no_data_intr(ide_drive_t *);
+static ide_startstop_t task_out_intr(ide_drive_t *);
+
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
@@ -173,12 +175,18 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
+		task->handler = task_out_intr;
 		hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
 	case TASKFILE_IN:
+		task->handler = task_in_intr;
+		/* fall-through */
 	case TASKFILE_NO_DATA:
+		/* WIN_{SPECIFY,RESTORE,SETMULT} use custom handlers */
+		if (task->handler == NULL)
+			task->handler = task_no_data_intr;
 		ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
 		return ide_started;
 	default:
@@ -248,7 +256,7 @@ ide_startstop_t recal_intr (ide_drive_t *drive)
 /*
  * Handler for commands without a data phase
  */
-ide_startstop_t task_no_data_intr (ide_drive_t *drive)
+static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 {
 	ide_task_t *args	= HWGROUP(drive)->rq->special;
 	ide_hwif_t *hwif	= HWIF(drive);
@@ -544,7 +552,6 @@ EXPORT_SYMBOL(ide_raw_taskfile);
 int ide_no_data_taskfile(ide_drive_t *drive, ide_task_t *task)
 {
 	task->data_phase = TASKFILE_NO_DATA;
-	task->handler    = task_no_data_intr;
 
 	return ide_raw_taskfile(drive, task, NULL, 0);
 }
@@ -667,7 +674,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			}
 			/* fall through */
 		case TASKFILE_OUT:
-			args.handler = &task_out_intr;
 			/* fall through */
 		case TASKFILE_OUT_DMAQ:
 		case TASKFILE_OUT_DMA:
@@ -685,7 +691,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			}
 			/* fall through */
 		case TASKFILE_IN:
-			args.handler = &task_in_intr;
 			/* fall through */
 		case TASKFILE_IN_DMAQ:
 		case TASKFILE_IN_DMA:
@@ -693,7 +698,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 			data_buf = inbuf;
 			break;
 		case TASKFILE_NO_DATA:
-			args.handler = &task_no_data_intr;
 			break;
 		default:
 			err = -EFAULT;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 2c28fb75915a..5d675f172037 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -973,7 +973,6 @@ extern ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 extern ide_startstop_t set_multmode_intr(ide_drive_t *);
 extern ide_startstop_t set_geometry_intr(ide_drive_t *);
 extern ide_startstop_t recal_intr(ide_drive_t *);
-extern ide_startstop_t task_no_data_intr(ide_drive_t *);
 extern ide_startstop_t task_in_intr(ide_drive_t *);
 extern ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
 
-- 
cgit v1.2.3


From 57d7366b78b74a9eef873e8212c03d8c2033a764 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:16 +0100
Subject: ide: remove 'handler' field from ide_task_t (take 2)

* Add IDE_TFLAG_CUSTOM_HANDLER taskfile flag and use it for internal requests
  which require custom handlers.  Check the flag in do_rw_taskfile() and set
  handler accordingly.

* Cleanup ide_init_{specify,restore,setmult}_cmd() and rename it to
  ide_tf_set_{specify,restore,setmult}_cmd().

* Make {set_geometry,recal,set_multmode}_intr() static.

* Remove no longer needed 'handler' field from ide_task_t.

v2:
* 'handler' in do_rw_taskfile() must be set to NULL initially.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       | 41 ++++++++++++++++++-----------------------
 drivers/ide/ide-taskfile.c | 28 +++++++++++++++++++---------
 include/linux/ide.h        |  5 +----
 3 files changed, 38 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 18ac1bd0811f..48c38b68bd36 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -637,32 +637,26 @@ static ide_startstop_t drive_cmd_intr (ide_drive_t *drive)
 	return ide_stopped;
 }
 
-static void ide_init_specify_cmd(ide_drive_t *drive, ide_task_t *task)
+static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
-	task->tf.nsect   = drive->sect;
-	task->tf.lbal    = drive->sect;
-	task->tf.lbam    = drive->cyl;
-	task->tf.lbah    = drive->cyl >> 8;
-	task->tf.device  = ((drive->head - 1) | drive->select.all) & ~ATA_LBA;
-	task->tf.command = WIN_SPECIFY;
-
-	task->handler = &set_geometry_intr;
+	tf->nsect   = drive->sect;
+	tf->lbal    = drive->sect;
+	tf->lbam    = drive->cyl;
+	tf->lbah    = drive->cyl >> 8;
+	tf->device  = ((drive->head - 1) | drive->select.all) & ~ATA_LBA;
+	tf->command = WIN_SPECIFY;
 }
 
-static void ide_init_restore_cmd(ide_drive_t *drive, ide_task_t *task)
+static void ide_tf_set_restore_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
-	task->tf.nsect   = drive->sect;
-	task->tf.command = WIN_RESTORE;
-
-	task->handler = &recal_intr;
+	tf->nsect   = drive->sect;
+	tf->command = WIN_RESTORE;
 }
 
-static void ide_init_setmult_cmd(ide_drive_t *drive, ide_task_t *task)
+static void ide_tf_set_setmult_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
-	task->tf.nsect   = drive->mult_req;
-	task->tf.command = WIN_SETMULT;
-
-	task->handler = &set_multmode_intr;
+	tf->nsect   = drive->mult_req;
+	tf->command = WIN_SETMULT;
 }
 
 static ide_startstop_t ide_disk_special(ide_drive_t *drive)
@@ -675,15 +669,15 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 
 	if (s->b.set_geometry) {
 		s->b.set_geometry = 0;
-		ide_init_specify_cmd(drive, &args);
+		ide_tf_set_specify_cmd(drive, &args.tf);
 	} else if (s->b.recalibrate) {
 		s->b.recalibrate = 0;
-		ide_init_restore_cmd(drive, &args);
+		ide_tf_set_restore_cmd(drive, &args.tf);
 	} else if (s->b.set_multmode) {
 		s->b.set_multmode = 0;
 		if (drive->mult_req > drive->id->max_multsect)
 			drive->mult_req = drive->id->max_multsect;
-		ide_init_setmult_cmd(drive, &args);
+		ide_tf_set_setmult_cmd(drive, &args.tf);
 	} else if (s->all) {
 		int special = s->all;
 		s->all = 0;
@@ -691,7 +685,8 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 		return ide_stopped;
 	}
 
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE |
+			IDE_TFLAG_CUSTOM_HANDLER;
 
 	do_rw_taskfile(drive, &args);
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 835465d61f70..236f91f11a73 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -151,12 +151,15 @@ static int inline task_dma_ok(ide_task_t *task)
 }
 
 static ide_startstop_t task_no_data_intr(ide_drive_t *);
-static ide_startstop_t task_out_intr(ide_drive_t *);
+static ide_startstop_t set_geometry_intr(ide_drive_t *);
+static ide_startstop_t recal_intr(ide_drive_t *);
+static ide_startstop_t set_multmode_intr(ide_drive_t *);
 
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct ide_taskfile *tf = &task->tf;
+	ide_handler_t *handler = NULL;
 
 	if (task->data_phase == TASKFILE_MULTI_IN ||
 	    task->data_phase == TASKFILE_MULTI_OUT) {
@@ -175,19 +178,26 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
-		task->handler = task_out_intr;
 		hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
 	case TASKFILE_IN:
-		task->handler = task_in_intr;
+		handler = task_in_intr;
 		/* fall-through */
 	case TASKFILE_NO_DATA:
+		if (handler == NULL)
+			handler = task_no_data_intr;
 		/* WIN_{SPECIFY,RESTORE,SETMULT} use custom handlers */
-		if (task->handler == NULL)
-			task->handler = task_no_data_intr;
-		ide_execute_command(drive, tf->command, task->handler, WAIT_WORSTCASE, NULL);
+		if (task->tf_flags & IDE_TFLAG_CUSTOM_HANDLER) {
+			switch (tf->command) {
+			case WIN_SPECIFY: handler = set_geometry_intr;	break;
+			case WIN_RESTORE: handler = recal_intr;		break;
+			case WIN_SETMULT: handler = set_multmode_intr;	break;
+			}
+		}
+		ide_execute_command(drive, tf->command, handler,
+				    WAIT_WORSTCASE, NULL);
 		return ide_started;
 	default:
 		if (task_dma_ok(task) == 0 || drive->using_dma == 0 ||
@@ -202,7 +212,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 /*
  * set_multmode_intr() is invoked on completion of a WIN_SETMULT cmd.
  */
-ide_startstop_t set_multmode_intr (ide_drive_t *drive)
+static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	u8 stat;
@@ -220,7 +230,7 @@ ide_startstop_t set_multmode_intr (ide_drive_t *drive)
 /*
  * set_geometry_intr() is invoked on completion of a WIN_SPECIFY cmd.
  */
-ide_startstop_t set_geometry_intr (ide_drive_t *drive)
+static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	int retries = 5;
@@ -243,7 +253,7 @@ ide_startstop_t set_geometry_intr (ide_drive_t *drive)
 /*
  * recal_intr() is invoked on completion of a WIN_RESTORE (recalibrate) cmd.
  */
-ide_startstop_t recal_intr (ide_drive_t *drive)
+static ide_startstop_t recal_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	u8 stat;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 5d675f172037..721c9d8f41a2 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -911,6 +911,7 @@ enum {
 	IDE_TFLAG_WRITE			= (1 << 15),
 	IDE_TFLAG_FLAGGED_SET_IN_FLAGS	= (1 << 16),
 	IDE_TFLAG_IN_DATA		= (1 << 17),
+	IDE_TFLAG_CUSTOM_HANDLER	= (1 << 18),
 };
 
 struct ide_taskfile {
@@ -949,7 +950,6 @@ typedef struct ide_task_s {
 	};
 	u32			tf_flags;
 	int			data_phase;
-	ide_handler_t		*handler;
 	struct request		*rq;		/* copy of request */
 	void			*special;	/* valid_t generally */
 } ide_task_t;
@@ -970,9 +970,6 @@ void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
  */
 extern ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
-extern ide_startstop_t set_multmode_intr(ide_drive_t *);
-extern ide_startstop_t set_geometry_intr(ide_drive_t *);
-extern ide_startstop_t recal_intr(ide_drive_t *);
 extern ide_startstop_t task_in_intr(ide_drive_t *);
 extern ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
 
-- 
cgit v1.2.3


From f6e29e35cc0f9facf2eb0b0454f9b09021b5aa6f Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:16 +0100
Subject: ide-disk: use do_rw_taskfile() (take 2)

* Add IDE_TFLAG_DMA_PIO_FALLBACK taskfile flag to indicate the need
  to skip loading taskfile registers in do_rw_taskfile().

* Export do_rw_taskfile().

* Convert __ide_do_rw_disk() to use do_rw_taskfile().

* Unexport ide_tf_load().

* Unexport {pre_task_out,task_in}_intr() and make it static.

* Remove incorrect comment about do_rw_taskfile() from <linux/ide.h>.

There should be no functionality changes caused by this patch.

v2:
* Add missing blk_fs_request() check to task_dma_ok() (for VDMA).

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c     | 29 ++++++++++-------------------
 drivers/ide/ide-taskfile.c | 16 ++++++++--------
 include/linux/ide.h        |  9 ++-------
 3 files changed, 20 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 97abc91db7d0..3d7127ba67a0 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -185,6 +185,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 	u8 lba48		= (drive->addressing == 1) ? 1 : 0;
 	ide_task_t		task;
 	struct ide_taskfile	*tf = &task.tf;
+	ide_startstop_t		rc;
 
 	if ((hwif->host_flags & IDE_HFLAG_NO_LBA48_DMA) && lba48 && dma) {
 		if (block + rq->nr_sectors > 1ULL << 28)
@@ -252,32 +253,22 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 		task.tf_flags |= IDE_TFLAG_WRITE;
 
 	ide_tf_set_cmd(drive, &task, dma);
+	if (!dma)
+		hwif->data_phase = task.data_phase;
+	task.rq = rq;
 
-	ide_tf_load(drive, &task);
+	rc = do_rw_taskfile(drive, &task);
 
-	if (dma) {
-		if (!hwif->dma_setup(drive)) {
-			hwif->dma_exec_cmd(drive, tf->command);
-			hwif->dma_start(drive);
-			return ide_started;
-		}
+	if (rc == ide_stopped && dma) {
 		/* fallback to PIO */
+		task.tf_flags |= IDE_TFLAG_DMA_PIO_FALLBACK;
 		ide_tf_set_cmd(drive, &task, 0);
+		hwif->data_phase = task.data_phase;
 		ide_init_sg_cmd(drive, rq);
+		rc = do_rw_taskfile(drive, &task);
 	}
 
-	hwif->data_phase = task.data_phase;
-
-	if (rq_data_dir(rq) == READ) {
-		ide_execute_command(drive, tf->command, &task_in_intr,
-				    WAIT_WORSTCASE, NULL);
-		return ide_started;
-	} else {
-		hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
-		ndelay(400);	/* FIXME */
-
-		return pre_task_out_intr(drive, rq);
-	}
+	return rc;
 }
 
 /*
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 236f91f11a73..2d63ea9ee61b 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -114,8 +114,6 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 		hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
 }
 
-EXPORT_SYMBOL_GPL(ide_tf_load);
-
 int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 {
 	ide_task_t args;
@@ -133,7 +131,7 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 
 static int inline task_dma_ok(ide_task_t *task)
 {
-	if (task->tf_flags & IDE_TFLAG_FLAGGED)
+	if (blk_fs_request(task->rq) || (task->tf_flags & IDE_TFLAG_FLAGGED))
 		return 1;
 
 	switch (task->tf.command) {
@@ -154,6 +152,8 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *);
 static ide_startstop_t set_geometry_intr(ide_drive_t *);
 static ide_startstop_t recal_intr(ide_drive_t *);
 static ide_startstop_t set_multmode_intr(ide_drive_t *);
+static ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
+static ide_startstop_t task_in_intr(ide_drive_t *);
 
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
@@ -173,7 +173,8 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_FLAGGED)
 		task->tf_flags |= IDE_TFLAG_FLAGGED_SET_IN_FLAGS;
 
-	ide_tf_load(drive, task);
+	if ((task->tf_flags & IDE_TFLAG_DMA_PIO_FALLBACK) == 0)
+		ide_tf_load(drive, task);
 
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
@@ -208,6 +209,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 		return ide_started;
 	}
 }
+EXPORT_SYMBOL_GPL(do_rw_taskfile);
 
 /*
  * set_multmode_intr() is invoked on completion of a WIN_SETMULT cmd.
@@ -446,7 +448,7 @@ static void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 /*
  * Handler for command with PIO data-in phase (Read/Read Multiple).
  */
-ide_startstop_t task_in_intr (ide_drive_t *drive)
+static ide_startstop_t task_in_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = HWGROUP(drive)->rq;
@@ -477,7 +479,6 @@ ide_startstop_t task_in_intr (ide_drive_t *drive)
 
 	return ide_started;
 }
-EXPORT_SYMBOL(task_in_intr);
 
 /*
  * Handler for command with PIO data-out phase (Write/Write Multiple).
@@ -507,7 +508,7 @@ static ide_startstop_t task_out_intr (ide_drive_t *drive)
 	return ide_started;
 }
 
-ide_startstop_t pre_task_out_intr (ide_drive_t *drive, struct request *rq)
+static ide_startstop_t pre_task_out_intr(ide_drive_t *drive, struct request *rq)
 {
 	ide_startstop_t startstop;
 
@@ -528,7 +529,6 @@ ide_startstop_t pre_task_out_intr (ide_drive_t *drive, struct request *rq)
 
 	return ide_started;
 }
-EXPORT_SYMBOL(pre_task_out_intr);
 
 int ide_raw_taskfile(ide_drive_t *drive, ide_task_t *task, u8 *buf, u16 nsect)
 {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 721c9d8f41a2..c333a7528d94 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -912,6 +912,7 @@ enum {
 	IDE_TFLAG_FLAGGED_SET_IN_FLAGS	= (1 << 16),
 	IDE_TFLAG_IN_DATA		= (1 << 17),
 	IDE_TFLAG_CUSTOM_HANDLER	= (1 << 18),
+	IDE_TFLAG_DMA_PIO_FALLBACK	= (1 << 19),
 };
 
 struct ide_taskfile {
@@ -965,13 +966,7 @@ extern int drive_is_ready(ide_drive_t *);
 
 void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
-/*
- * taskfile io for disks for now...and builds request from ide_ioctl
- */
-extern ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
-
-extern ide_startstop_t task_in_intr(ide_drive_t *);
-extern ide_startstop_t pre_task_out_intr(ide_drive_t *, struct request *);
+ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
 int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *, u16);
 int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
-- 
cgit v1.2.3


From c2b57cdc1d2976444d451a2a2e43e11b61ed0638 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:17 +0100
Subject: ide: add ide_tf_read() helper

* Factor out code reading taskfile registers from ide_end_drive_cmd()
  to the new ide_tf_read() helper.

* Add IDE_TFLAG_IN_* taskfile flags to indicate the need to load
  particular IDE taskfile register in ide_tf_read().

* Update ide_end_drive_cmd() to set respective IDE_TFLAG_IN_* taksfile flags.

* Add ide_get_lba_addr() for getting LBA sector address from taskfile struct.

* Factor out code getting sector address from ide_dump_ata_status()
  to the new ide_dump_sector() function.

* Convert ide_dump_sector() to use ide_tf_read() and ide_get_lba_addr().

* Remove no longer needed ide_read_24().

The only change in functionality caused by this patch is that
ide_dump_ata_status() no longer prints "high"/"low" parts of LBA48
sector address (of course LBA48 sector address is still printed).

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 68 ++++++++++++++++++++++++++++++++++---------------
 drivers/ide/ide-iops.c |  8 ------
 drivers/ide/ide-lib.c  | 69 +++++++++++++++++++++++++++-----------------------
 include/linux/ide.h    | 24 ++++++++++++++++--
 4 files changed, 106 insertions(+), 63 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 48c38b68bd36..e053e00a7058 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -297,6 +297,48 @@ static void ide_complete_pm_request (ide_drive_t *drive, struct request *rq)
 	spin_unlock_irqrestore(&ide_lock, flags);
 }
 
+void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct ide_taskfile *tf = &task->tf;
+
+	if (task->tf_flags & IDE_TFLAG_IN_DATA) {
+		u16 data = hwif->INW(IDE_DATA_REG);
+
+		tf->data = data & 0xff;
+		tf->hob_data = (data >> 8) & 0xff;
+	}
+
+	/* be sure we're looking at the low order bits */
+	hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
+
+	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
+		tf->nsect  = hwif->INB(IDE_NSECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
+		tf->lbal   = hwif->INB(IDE_SECTOR_REG);
+	if (task->tf_flags & IDE_TFLAG_IN_LBAM)
+		tf->lbam   = hwif->INB(IDE_LCYL_REG);
+	if (task->tf_flags & IDE_TFLAG_IN_LBAH)
+		tf->lbah   = hwif->INB(IDE_HCYL_REG);
+	if (task->tf_flags & IDE_TFLAG_IN_DEVICE)
+		tf->device = hwif->INB(IDE_SELECT_REG);
+
+	if (task->tf_flags & IDE_TFLAG_LBA48) {
+		hwif->OUTB(drive->ctl | 0x80, IDE_CONTROL_REG);
+
+		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
+			tf->hob_feature = hwif->INB(IDE_FEATURE_REG);
+		if (task->tf_flags & IDE_TFLAG_IN_HOB_NSECT)
+			tf->hob_nsect   = hwif->INB(IDE_NSECTOR_REG);
+		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAL)
+			tf->hob_lbal    = hwif->INB(IDE_SECTOR_REG);
+		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAM)
+			tf->hob_lbam    = hwif->INB(IDE_LCYL_REG);
+		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAH)
+			tf->hob_lbah    = hwif->INB(IDE_HCYL_REG);
+	}
+}
+
 /**
  *	ide_end_drive_cmd	-	end an explicit drive command
  *	@drive: command 
@@ -339,30 +381,14 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 		if (args) {
 			struct ide_taskfile *tf = &args->tf;
 
-			if (args->tf_flags & IDE_TFLAG_IN_DATA) {
-				u16 data = hwif->INW(IDE_DATA_REG);
-
-				tf->data = data & 0xff;
-				tf->hob_data = (data >> 8) & 0xff;
-			}
 			tf->error = err;
-			/* be sure we're looking at the low order bits */
-			hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
-			tf->nsect  = hwif->INB(IDE_NSECTOR_REG);
-			tf->lbal   = hwif->INB(IDE_SECTOR_REG);
-			tf->lbam   = hwif->INB(IDE_LCYL_REG);
-			tf->lbah   = hwif->INB(IDE_HCYL_REG);
-			tf->device = hwif->INB(IDE_SELECT_REG);
 			tf->status = stat;
 
-			if (args->tf_flags & IDE_TFLAG_LBA48) {
-				hwif->OUTB(drive->ctl|0x80, IDE_CONTROL_REG);
-				tf->hob_feature = hwif->INB(IDE_FEATURE_REG);
-				tf->hob_nsect   = hwif->INB(IDE_NSECTOR_REG);
-				tf->hob_lbal    = hwif->INB(IDE_SECTOR_REG);
-				tf->hob_lbam    = hwif->INB(IDE_LCYL_REG);
-				tf->hob_lbah    = hwif->INB(IDE_HCYL_REG);
-			}
+			args->tf_flags |= (IDE_TFLAG_IN_TF|IDE_TFLAG_IN_DEVICE);
+			if (args->tf_flags & IDE_TFLAG_LBA48)
+				args->tf_flags |= IDE_TFLAG_IN_HOB;
+
+			ide_tf_read(drive, args);
 		}
 	} else if (blk_pm_request(rq)) {
 		struct request_pm_state *pm = rq->data;
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 38d6b15d6c47..c97c0719ddf1 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -158,14 +158,6 @@ void default_hwif_mmiops (ide_hwif_t *hwif)
 
 EXPORT_SYMBOL(default_hwif_mmiops);
 
-u32 ide_read_24 (ide_drive_t *drive)
-{
-	u8 sect = HWIF(drive)->INB(IDE_SECTOR_REG);
-	u8 lcyl = HWIF(drive)->INB(IDE_LCYL_REG);
-	u8 hcyl = HWIF(drive)->INB(IDE_HCYL_REG);
-	return (hcyl<<16)|(lcyl<<8)|sect;
-}
-
 void SELECT_DRIVE (ide_drive_t *drive)
 {
 	if (HWIF(drive)->selectproc)
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 001085845a79..15736d4ce9b4 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -479,6 +479,42 @@ static void ide_dump_opcode(ide_drive_t *drive)
 		printk("0x%02x\n", opcode);
 }
 
+static u64 ide_get_lba_addr(struct ide_taskfile *tf, int lba48)
+{
+	u32 high, low;
+
+	if (lba48)
+		high = (tf->hob_lbah << 16) | (tf->hob_lbam << 8) |
+			tf->hob_lbal;
+	else
+		high = tf->device & 0xf;
+	low  = (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
+
+	return ((u64)high << 24) | low;
+}
+
+static void ide_dump_sector(ide_drive_t *drive)
+{
+	ide_task_t task;
+	struct ide_taskfile *tf = &task.tf;
+	int lba48 = (drive->addressing == 1) ? 1 : 0;
+
+	memset(&task, 0, sizeof(task));
+	if (lba48)
+		task.tf_flags = IDE_TFLAG_IN_LBA | IDE_TFLAG_IN_HOB_LBA |
+				IDE_TFLAG_LBA48;
+	else
+		task.tf_flags = IDE_TFLAG_IN_LBA | IDE_TFLAG_IN_DEVICE;
+
+	ide_tf_read(drive, &task);
+
+	if (lba48 || (tf->device & ATA_LBA))
+		printk(", LBAsect=%llu", ide_get_lba_addr(tf, lba48));
+	else
+		printk(", CHS=%d/%d/%d", (tf->lbah << 8) + tf->lbam,
+					 tf->device & 0xf, tf->lbal);
+}
+
 static u8 ide_dump_ata_status(ide_drive_t *drive, const char *msg, u8 stat)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -512,38 +548,7 @@ static u8 ide_dump_ata_status(ide_drive_t *drive, const char *msg, u8 stat)
 		printk("}");
 		if ((err & (BBD_ERR | ABRT_ERR)) == BBD_ERR ||
 		    (err & (ECC_ERR|ID_ERR|MARK_ERR))) {
-			if (drive->addressing == 1) {
-				__u64 sectors = 0;
-				u32 low = 0, high = 0;
-				hwif->OUTB(drive->ctl&~0x80, IDE_CONTROL_REG);
-				low = ide_read_24(drive);
-				hwif->OUTB(drive->ctl|0x80, IDE_CONTROL_REG);
-				high = ide_read_24(drive);
-				sectors = ((__u64)high << 24) | low;
-				printk(", LBAsect=%llu, high=%d, low=%d",
-				       (unsigned long long) sectors,
-				       high, low);
-			} else {
-				u8 sector, lcyl, hcyl, cur;
-
-				sector = hwif->INB(IDE_SECTOR_REG);
-				lcyl   = hwif->INB(IDE_LCYL_REG);
-				hcyl   = hwif->INB(IDE_HCYL_REG);
-				cur    = hwif->INB(IDE_SELECT_REG);
-
-				if (cur & 0x40) {	/* using LBA? */
-					printk(", LBAsect=%ld", (unsigned long)
-						((cur & 0xf) << 24) |
-						(hcyl << 16) |
-						(lcyl <<  8) |
-						sector);
-				} else {
-					printk(", CHS=%d/%d/%d",
-						(hcyl << 8) + lcyl,
-						cur & 0xf,
-						sector);
-				}
-			}
+			ide_dump_sector(drive);
 			if (HWGROUP(drive) && HWGROUP(drive)->rq)
 				printk(", sector=%llu",
 					(unsigned long long)HWGROUP(drive)->rq->sector);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c333a7528d94..67f98c096c03 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -913,6 +913,27 @@ enum {
 	IDE_TFLAG_IN_DATA		= (1 << 17),
 	IDE_TFLAG_CUSTOM_HANDLER	= (1 << 18),
 	IDE_TFLAG_DMA_PIO_FALLBACK	= (1 << 19),
+	IDE_TFLAG_IN_HOB_FEATURE	= (1 << 20),
+	IDE_TFLAG_IN_HOB_NSECT		= (1 << 21),
+	IDE_TFLAG_IN_HOB_LBAL		= (1 << 22),
+	IDE_TFLAG_IN_HOB_LBAM		= (1 << 23),
+	IDE_TFLAG_IN_HOB_LBAH		= (1 << 24),
+	IDE_TFLAG_IN_HOB_LBA		= IDE_TFLAG_IN_HOB_LBAL |
+					  IDE_TFLAG_IN_HOB_LBAM |
+					  IDE_TFLAG_IN_HOB_LBAH,
+	IDE_TFLAG_IN_HOB		= IDE_TFLAG_IN_HOB_FEATURE |
+					  IDE_TFLAG_IN_HOB_NSECT |
+					  IDE_TFLAG_IN_HOB_LBA,
+	IDE_TFLAG_IN_NSECT		= (1 << 25),
+	IDE_TFLAG_IN_LBAL		= (1 << 26),
+	IDE_TFLAG_IN_LBAM		= (1 << 27),
+	IDE_TFLAG_IN_LBAH		= (1 << 28),
+	IDE_TFLAG_IN_LBA		= IDE_TFLAG_IN_LBAL |
+					  IDE_TFLAG_IN_LBAM |
+					  IDE_TFLAG_IN_LBAH,
+	IDE_TFLAG_IN_TF			= IDE_TFLAG_IN_NSECT |
+					  IDE_TFLAG_IN_LBA,
+	IDE_TFLAG_IN_DEVICE		= (1 << 29),
 };
 
 struct ide_taskfile {
@@ -956,8 +977,7 @@ typedef struct ide_task_s {
 } ide_task_t;
 
 void ide_tf_load(ide_drive_t *, ide_task_t *);
-
-extern u32 ide_read_24(ide_drive_t *);
+void ide_tf_read(ide_drive_t *, ide_task_t *);
 
 extern void SELECT_DRIVE(ide_drive_t *);
 extern void SELECT_MASK(ide_drive_t *, int);
-- 
cgit v1.2.3


From a501633c7d44087e806597d3a213d735346edd51 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:17 +0100
Subject: ide-disk: use ide_get_lba_addr()

* Export ide_get_lba_addr().

* Convert idedisk_{read_native,set}_max_address() to use ide_get_lba_addr().

* Remove incorrect comment from idedisk_read_native_max_address()
  (noticed by Sergei).

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c | 26 ++++----------------------
 drivers/ide/ide-lib.c  |  3 ++-
 include/linux/ide.h    |  1 +
 3 files changed, 7 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 3d7127ba67a0..d8fdd865dea9 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -326,18 +326,9 @@ static u64 idedisk_read_native_max_address(ide_drive_t *drive, int lba48)
 	ide_no_data_taskfile(drive, &args);
 
 	/* if OK, compute maximum address value */
-	if ((tf->status & 0x01) == 0) {
-		u32 high, low;
+	if ((tf->status & 0x01) == 0)
+		addr = ide_get_lba_addr(tf, lba48) + 1;
 
-		if (lba48)
-			high = (tf->hob_lbah << 16) | (tf->hob_lbam << 8) |
-				tf->hob_lbal;
-		else
-			high = tf->device & 0xf;
-		low  = (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
-		addr = ((__u64)high << 24) | low;
-		addr++;	/* since the return value is (maxlba - 1), we add 1 */
-	}
 	return addr;
 }
 
@@ -373,18 +364,9 @@ static u64 idedisk_set_max_address(ide_drive_t *drive, u64 addr_req, int lba48)
 	/* submit command request */
 	ide_no_data_taskfile(drive, &args);
 	/* if OK, compute maximum address value */
-	if ((tf->status & 0x01) == 0) {
-		u32 high, low;
+	if ((tf->status & 0x01) == 0)
+		addr_set = ide_get_lba_addr(tf, lba48) + 1;
 
-		if (lba48)
-			high = (tf->hob_lbah << 16) | (tf->hob_lbam << 8) |
-				tf->hob_lbal;
-		else
-			high = tf->device & 0xf;
-		low  = (tf->lbah << 16) | (tf->lbam << 8) | tf->lbal;
-		addr_set = ((__u64)high << 24) | low;
-		addr_set++;
-	}
 	return addr_set;
 }
 
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index d7503c489e99..6b2e810cb9e0 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -479,7 +479,7 @@ static void ide_dump_opcode(ide_drive_t *drive)
 		printk("0x%02x\n", opcode);
 }
 
-static u64 ide_get_lba_addr(struct ide_taskfile *tf, int lba48)
+u64 ide_get_lba_addr(struct ide_taskfile *tf, int lba48)
 {
 	u32 high, low;
 
@@ -492,6 +492,7 @@ static u64 ide_get_lba_addr(struct ide_taskfile *tf, int lba48)
 
 	return ((u64)high << 24) | low;
 }
+EXPORT_SYMBOL_GPL(ide_get_lba_addr);
 
 static void ide_dump_sector(ide_drive_t *drive)
 {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 67f98c096c03..ecaa96e613dc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1238,6 +1238,7 @@ static inline int ide_dev_is_sata(struct hd_driveid *id)
 	return 0;
 }
 
+u64 ide_get_lba_addr(struct ide_taskfile *, int);
 u8 ide_dump_status(ide_drive_t *, const char *, u8);
 
 typedef struct ide_pio_timings_s {
-- 
cgit v1.2.3


From 3071a9d00b8684899d93f368e670c4de0293df29 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:17 +0100
Subject: ide: make 'extra' field in struct ide_port_info u8

The maximum value used currently for 'extra' field in struct ide_port_info
is 240.

Make 'extra' u8 so it packs nicely together with enablebits[] and 'chipset'
fields (ide_pci_enablebit_t is 3 bytes and hwif_chipset_t is 1 byte).

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index ecaa96e613dc..6f5b6b5e9c1d 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1111,7 +1111,7 @@ struct ide_port_info {
 	void			(*fixup)(ide_hwif_t *);
 	ide_pci_enablebit_t	enablebits[2];
 	hwif_chipset_t		chipset;
-	unsigned int		extra;
+	u8			extra;
 	u32			host_flags;
 	u8			pio_mask;
 	u8			swdma_mask;
-- 
cgit v1.2.3


From 4db90a145292327b95b03f6dcd3352327235cc36 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 25 Jan 2008 22:17:18 +0100
Subject: ide: add IDE_HFLAG_ABUSE_SET_DMA_MODE host flag

* Add IDE_HFLAG_ABUSE_SET_DMA_MODE host flag and use it to decide
  what to do with transfer modes < XFER_PIO_0 in ide_set_xfer_rate().

* Set IDE_HFLAG_ABUSE_SET_DMA_MODE in host drivers that need it
  (aec62xx, amd74xx, cs5520, cs5535, hpt34x, hpt366, pdc202xx_old,
  serverworks, tc86c001 and via82cxxx) and cleanup ->set_dma_mode
  methods in host drivers that don't (IDE core code guarantees that
  ->set_dma_mode will be called only for modes which are present
  in SWDMA/MWDMA/UDMA masks).

While at it:

* Add IDE_HFLAGS_HPT34X/HPT3XX/PDC202XX/SVWKS define in
  hpt34x/hpt366/pdc202xx_old/serverworks host driver.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c       |  2 --
 drivers/ide/cris/ide-cris.c    |  2 --
 drivers/ide/ide-lib.c          |  6 ++++++
 drivers/ide/mips/au1xxx-ide.c  |  2 --
 drivers/ide/pci/aec62xx.c      | 13 ++++++++++---
 drivers/ide/pci/alim15x3.c     |  3 ---
 drivers/ide/pci/amd74xx.c      |  1 +
 drivers/ide/pci/atiixp.c       |  3 ---
 drivers/ide/pci/cmd64x.c       |  2 --
 drivers/ide/pci/cs5520.c       |  1 +
 drivers/ide/pci/cs5530.c       |  2 --
 drivers/ide/pci/cs5535.c       |  2 +-
 drivers/ide/pci/hpt34x.c       | 12 +++++++-----
 drivers/ide/pci/hpt366.c       | 19 +++++++++++--------
 drivers/ide/pci/it8213.c       | 21 ++++-----------------
 drivers/ide/pci/pdc202xx_new.c | 38 ++++++++++++--------------------------
 drivers/ide/pci/pdc202xx_old.c | 12 +++++++-----
 drivers/ide/pci/piix.c         | 17 ++++-------------
 drivers/ide/pci/sc1200.c       |  2 --
 drivers/ide/pci/scc_pata.c     | 14 +-------------
 drivers/ide/pci/serverworks.c  | 17 ++++++++++-------
 drivers/ide/pci/siimage.c      | 29 ++++++++---------------------
 drivers/ide/pci/sis5513.c      | 23 ++++-------------------
 drivers/ide/pci/sl82c105.c     | 40 ++++++++++++++++------------------------
 drivers/ide/pci/slc90e66.c     | 14 ++------------
 drivers/ide/pci/tc86c001.c     |  3 ++-
 drivers/ide/pci/triflex.c      |  2 --
 drivers/ide/pci/via82cxxx.c    |  1 +
 drivers/ide/ppc/pmac.c         | 42 ++++++++++++------------------------------
 include/linux/ide.h            |  1 +
 30 files changed, 121 insertions(+), 225 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 93f71fcfc04d..673402f4a295 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -272,8 +272,6 @@ static void icside_set_dma_mode(ide_drive_t *drive, const u8 xfer_mode)
 	case XFER_SW_DMA_0:
 		cycle_time = 480;
 		break;
-	default:
-		return;
 	}
 
 	/*
diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c
index 476e0d65ed43..325e608d9e62 100644
--- a/drivers/ide/cris/ide-cris.c
+++ b/drivers/ide/cris/ide-cris.c
@@ -747,8 +747,6 @@ static void cris_set_dma_mode(ide_drive_t *drive, const u8 speed)
 			strobe = ATA_DMA2_STROBE;
 			hold = ATA_DMA2_HOLD;
 			break;
-		default:
-			return;
 	}
 
 	if (speed >= XFER_UDMA_0)
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 8649db33f67d..a3bd8e8ed6b0 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -441,6 +441,12 @@ int ide_set_xfer_rate(ide_drive_t *drive, u8 rate)
 	 * case could happen iff the transfer mode has already been set on
 	 * the device by ide-proc.c::set_xfer_rate()).
 	 */
+	if (rate < XFER_PIO_0) {
+		if (hwif->host_flags & IDE_HFLAG_ABUSE_SET_DMA_MODE)
+			return ide_set_dma_mode(drive, rate);
+		else
+			return ide_config_drive_speed(drive, rate);
+	}
 
 	return ide_set_dma_mode(drive, rate);
 }
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index a4ce3ba15d61..a4d0d4ca73d0 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -198,8 +198,6 @@ static void auide_set_dma_mode(ide_drive_t *drive, const u8 speed)
 
 		break;
 #endif
-	default:
-		return;
 	}
 
 	au_writel(mem_sttime,MEM_STTIME2);
diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 44268504ae43..7f4d1857d555 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -202,6 +202,7 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.host_flags	= IDE_HFLAG_SERIALIZE |
 				  IDE_HFLAG_NO_ATAPI_DMA |
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -211,6 +212,7 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_aec62xx,
 		.init_hwif	= init_hwif_aec62xx,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_NO_AUTODMA |
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -220,7 +222,8 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_aec62xx,
 		.init_hwif	= init_hwif_aec62xx,
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA,
+		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA4,
@@ -228,7 +231,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.name		= "AEC6280",
 		.init_chipset	= init_chipset_aec62xx,
 		.init_hwif	= init_hwif_aec62xx,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
+				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
@@ -237,7 +242,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_aec62xx,
 		.init_hwif	= init_hwif_aec62xx,
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
+				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index ce293936af4b..49aa82e412b6 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -402,9 +402,6 @@ static void ali_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	u8 tmpbyte		= 0x00;
 	int m5229_udma		= (hwif->channel) ? 0x57 : 0x56;
 
-	if (speed < XFER_PIO_0)
-		return;
-
 	if (speed == XFER_UDMA_6)
 		speed1 = 0x47;
 
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index 8d4125ec252c..cee51fdafcf6 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -266,6 +266,7 @@ static void __devinit init_hwif_amd74xx(ide_hwif_t *hwif)
 #define IDE_HFLAGS_AMD \
 	(IDE_HFLAG_PIO_NO_BLACKLIST | \
 	 IDE_HFLAG_PIO_NO_DOWNGRADE | \
+	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
 	 IDE_HFLAG_POST_SET_MODE | \
 	 IDE_HFLAG_IO_32BIT | \
 	 IDE_HFLAG_UNMASK_IRQS | \
diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c
index ef8e0164ef7a..5ae26564fb72 100644
--- a/drivers/ide/pci/atiixp.c
+++ b/drivers/ide/pci/atiixp.c
@@ -133,9 +133,6 @@ static void atiixp_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	u32 tmp32;
 	u16 tmp16;
 
-	if (speed < XFER_MW_DMA_0)
-		return;
-
 	spin_lock_irqsave(&atiixp_lock, flags);
 
 	save_mdma_mode[drive->dn] = 0;
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index f3613bac9dbe..0b1e9479f019 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -322,8 +322,6 @@ static void cmd64x_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	case XFER_MW_DMA_0:
 		program_cycle_times(drive, 480, 215);
 		break;
-	default:
-		return;
 	}
 
 	if (speed >= XFER_SW_DMA_0)
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index 0466462fd21b..d1a91bcb5b29 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -137,6 +137,7 @@ static void __devinit init_hwif_cs5520(ide_hwif_t *hwif)
 				  IDE_HFLAG_CS5520 |		\
 				  IDE_HFLAG_VDMA |		\
 				  IDE_HFLAG_NO_ATAPI_DMA |	\
+				  IDE_HFLAG_ABUSE_SET_DMA_MODE |\
 				  IDE_HFLAG_BOOTABLE,		\
 		.pio_mask	= ATA_PIO4,			\
 	}
diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index 547690395eee..df5966b33460 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -116,8 +116,6 @@ static void cs5530_set_dma_mode(ide_drive_t *drive, const u8 mode)
 		case XFER_MW_DMA_0:	timings = 0x00077771; break;
 		case XFER_MW_DMA_1:	timings = 0x00012121; break;
 		case XFER_MW_DMA_2:	timings = 0x00002020; break;
-		default:
-			return;
 	}
 	basereg = CS5530_BASEREG(drive->hwif);
 	reg = inl(basereg + 4);			/* get drive0 config register */
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index ddcbeba671e1..50b3d7791f55 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -190,7 +190,7 @@ static const struct ide_port_info cs5535_chipset __devinitdata = {
 	.name		= "CS5535",
 	.init_hwif	= init_hwif_cs5535,
 	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_POST_SET_MODE |
-			  IDE_HFLAG_BOOTABLE,
+			  IDE_HFLAG_ABUSE_SET_DMA_MODE | IDE_HFLAG_BOOTABLE,
 	.pio_mask	= ATA_PIO4,
 	.mwdma_mask	= ATA_MWDMA2,
 	.udma_mask	= ATA_UDMA4,
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index ae6307fae4f9..dfba0d13fcd3 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -129,14 +129,18 @@ static void __devinit init_hwif_hpt34x(ide_hwif_t *hwif)
 	hwif->set_dma_mode = &hpt34x_set_mode;
 }
 
+#define IDE_HFLAGS_HPT34X \
+	(IDE_HFLAG_NO_ATAPI_DMA | \
+	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
+	 IDE_HFLAG_NO_AUTODMA)
+
 static const struct ide_port_info hpt34x_chipsets[] __devinitdata = {
 	{ /* 0 */
 		.name		= "HPT343",
 		.init_chipset	= init_chipset_hpt34x,
 		.init_hwif	= init_hwif_hpt34x,
 		.extra		= 16,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_NO_AUTODMA,
+		.host_flags	= IDE_HFLAGS_HPT34X,
 		.pio_mask	= ATA_PIO5,
 	},
 	{ /* 1 */
@@ -144,9 +148,7 @@ static const struct ide_port_info hpt34x_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_hpt34x,
 		.init_hwif	= init_hwif_hpt34x,
 		.extra		= 16,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_NO_AUTODMA |
-				  IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT34X | IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO5,
 #ifdef CONFIG_HPT34X_AUTODMA
 		.swdma_mask	= ATA_SWDMA2,
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 24d645751e0f..3777fb8c8043 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1461,6 +1461,11 @@ static int __devinit hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2)
 	return 0;
 }
 
+#define IDE_HFLAGS_HPT3XX \
+	(IDE_HFLAG_NO_ATAPI_DMA | \
+	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
+	 IDE_HFLAG_OFF_BOARD)
+
 static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 	{	/* 0 */
 		.name		= "HPT36x",
@@ -1475,9 +1480,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		 */
 		.enablebits	= {{0x50,0x10,0x10}, {0x54,0x04,0x04}},
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_SINGLE |
-				  IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX | IDE_HFLAG_SINGLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	},{	/* 1 */
@@ -1487,7 +1490,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.init_dma	= init_dma_hpt366,
 		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	},{	/* 2 */
@@ -1497,7 +1500,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.init_dma	= init_dma_hpt366,
 		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	},{	/* 3 */
@@ -1507,7 +1510,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.init_dma	= init_dma_hpt366,
 		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	},{	/* 4 */
@@ -1518,7 +1521,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
 		.udma_mask	= ATA_UDMA5,
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	},{	/* 5 */
@@ -1528,7 +1531,7 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.init_dma	= init_dma_hpt366,
 		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
 		.extra		= 240,
-		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_HPT3XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 	}
diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 90b52ed37bfc..2a0f45c4f4c4 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -101,24 +101,11 @@ static void it8213_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	pci_read_config_byte(dev, 0x54, &reg54);
 	pci_read_config_byte(dev, 0x55, &reg55);
 
-	switch(speed) {
-		case XFER_UDMA_6:
-		case XFER_UDMA_4:
-		case XFER_UDMA_2:	u_speed = 2 << (drive->dn * 4); break;
-		case XFER_UDMA_5:
-		case XFER_UDMA_3:
-		case XFER_UDMA_1:	u_speed = 1 << (drive->dn * 4); break;
-		case XFER_UDMA_0:	u_speed = 0 << (drive->dn * 4); break;
-			break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_SW_DMA_2:
-			break;
-		default:
-			return;
-	}
-
 	if (speed >= XFER_UDMA_0) {
+		u8 udma = speed - XFER_UDMA_0;
+
+		u_speed = min_t(u8, 2 - (udma & 1), udma) << (drive->dn * 4);
+
 		if (!(reg48 & u_flag))
 			pci_write_config_byte(dev, 0x48, reg48 | u_flag);
 		if (speed >= XFER_UDMA_5) {
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 79ba8eff3644..ef4a99b99d1f 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -162,32 +162,18 @@ static void pdcnew_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	if (max_dma_rate(hwif->pci_dev) == 4) {
 		u8 mode = speed & 0x07;
 
-		switch (speed) {
-			case XFER_UDMA_6:
-			case XFER_UDMA_5:
-			case XFER_UDMA_4:
-			case XFER_UDMA_3:
-			case XFER_UDMA_2:
-			case XFER_UDMA_1:
-			case XFER_UDMA_0:
-				set_indexed_reg(hwif, 0x10 + adj,
-						udma_timings[mode].reg10);
-				set_indexed_reg(hwif, 0x11 + adj,
-						udma_timings[mode].reg11);
-				set_indexed_reg(hwif, 0x12 + adj,
-						udma_timings[mode].reg12);
-				break;
-			case XFER_MW_DMA_2:
-			case XFER_MW_DMA_1:
-			case XFER_MW_DMA_0:
-				set_indexed_reg(hwif, 0x0e + adj,
-						mwdma_timings[mode].reg0e);
-				set_indexed_reg(hwif, 0x0f + adj,
-						mwdma_timings[mode].reg0f);
-				break;
-			default:
-				printk(KERN_ERR "pdc202xx_new: "
-				       "Unknown speed %d ignored\n", speed);
+		if (speed >= XFER_UDMA_0) {
+			set_indexed_reg(hwif, 0x10 + adj,
+					udma_timings[mode].reg10);
+			set_indexed_reg(hwif, 0x11 + adj,
+					udma_timings[mode].reg11);
+			set_indexed_reg(hwif, 0x12 + adj,
+					udma_timings[mode].reg12);
+		} else {
+			set_indexed_reg(hwif, 0x0e + adj,
+					mwdma_timings[mode].reg0e);
+			set_indexed_reg(hwif, 0x0f + adj,
+					mwdma_timings[mode].reg0f);
 		}
 	} else if (speed == XFER_UDMA_2) {
 		/* Set tHOLD bit to 0 if using UDMA mode 2 */
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index 22c7a7533b69..67b2781e2213 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -375,6 +375,11 @@ static void __devinit pdc202ata4_fixup_irq(struct pci_dev *dev,
 	}
 }
 
+#define IDE_HFLAGS_PDC202XX \
+	(IDE_HFLAG_ERROR_STOPS_FIFO | \
+	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
+	 IDE_HFLAG_OFF_BOARD)
+
 #define DECLARE_PDC2026X_DEV(name_str, udma, extra_flags) \
 	{ \
 		.name		= name_str, \
@@ -382,9 +387,7 @@ static void __devinit pdc202ata4_fixup_irq(struct pci_dev *dev,
 		.init_hwif	= init_hwif_pdc202xx, \
 		.init_dma	= init_dma_pdc202xx, \
 		.extra		= 48, \
-		.host_flags	= IDE_HFLAG_ERROR_STOPS_FIFO | \
-				  extra_flags | \
-				  IDE_HFLAG_OFF_BOARD, \
+		.host_flags	= IDE_HFLAGS_PDC202XX | extra_flags, \
 		.pio_mask	= ATA_PIO4, \
 		.mwdma_mask	= ATA_MWDMA2, \
 		.udma_mask	= udma, \
@@ -397,8 +400,7 @@ static const struct ide_port_info pdc202xx_chipsets[] __devinitdata = {
 		.init_hwif	= init_hwif_pdc202xx,
 		.init_dma	= init_dma_pdc202xx,
 		.extra		= 16,
-		.host_flags	= IDE_HFLAG_ERROR_STOPS_FIFO |
-				  IDE_HFLAG_OFF_BOARD,
+		.host_flags	= IDE_HFLAGS_PDC202XX,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA2,
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index 27781d294cea..bd6d3f77d30c 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -203,20 +203,11 @@ static void piix_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	pci_read_config_byte(dev, 0x54, &reg54);
 	pci_read_config_byte(dev, 0x55, &reg55);
 
-	switch(speed) {
-		case XFER_UDMA_4:
-		case XFER_UDMA_2:	u_speed = 2 << (drive->dn * 4); break;
-		case XFER_UDMA_5:
-		case XFER_UDMA_3:
-		case XFER_UDMA_1:	u_speed = 1 << (drive->dn * 4); break;
-		case XFER_UDMA_0:	u_speed = 0 << (drive->dn * 4); break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_SW_DMA_2:	break;
-		default:		return;
-	}
-
 	if (speed >= XFER_UDMA_0) {
+		u8 udma = speed - XFER_UDMA_0;
+
+		u_speed = min_t(u8, 2 - (udma & 1), udma) << (drive->dn * 4);
+
 		if (!(reg48 & u_flag))
 			pci_write_config_byte(dev, 0x48, reg48 | u_flag);
 		if (speed == XFER_UDMA_5) {
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index f0029b364c57..569a8fe70d3e 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -185,8 +185,6 @@ static void sc1200_set_dma_mode(ide_drive_t *drive, const u8 mode)
 				case PCI_CLK_66:	timings = 0x00015151;	break;
 			}
 			break;
-		default:
-			return;
 	}
 
 	if (unit == 0) {			/* are we configuring drive0? */
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index ebb7132b9b84..24a85bbcd2a6 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -254,19 +254,7 @@ static void scc_set_dma_mode(ide_drive_t *drive, const u8 speed)
 		offset = 0; /* 100MHz */
 	}
 
-	switch (speed) {
-	case XFER_UDMA_6:
-	case XFER_UDMA_5:
-	case XFER_UDMA_4:
-	case XFER_UDMA_3:
-	case XFER_UDMA_2:
-	case XFER_UDMA_1:
-	case XFER_UDMA_0:
-		idx = speed - XFER_UDMA_0;
-		break;
-	default:
-		return;
-	}
+	idx = speed - XFER_UDMA_0;
 
 	jcactsel = JCACTSELtbl[offset][idx];
 	if (is_slave) {
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index a7280311357b..e9bd269547bb 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -366,12 +366,17 @@ static void __devinit init_hwif_svwks (ide_hwif_t *hwif)
 	}
 }
 
+#define IDE_HFLAGS_SVWKS \
+	(IDE_HFLAG_LEGACY_IRQS | \
+	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
+	 IDE_HFLAG_BOOTABLE)
+
 static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 	{	/* 0 */
 		.name		= "SvrWks OSB4",
 		.init_chipset	= init_chipset_svwks,
 		.init_hwif	= init_hwif_svwks,
-		.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_BOOTABLE,
+		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= 0x00, /* UDMA is problematic on OSB4 */
@@ -379,7 +384,7 @@ static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 		.name		= "SvrWks CSB5",
 		.init_chipset	= init_chipset_svwks,
 		.init_hwif	= init_hwif_svwks,
-		.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_BOOTABLE,
+		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
@@ -387,7 +392,7 @@ static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 		.name		= "SvrWks CSB6",
 		.init_chipset	= init_chipset_svwks,
 		.init_hwif	= init_hwif_svwks,
-		.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_BOOTABLE,
+		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
@@ -395,8 +400,7 @@ static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 		.name		= "SvrWks CSB6",
 		.init_chipset	= init_chipset_svwks,
 		.init_hwif	= init_hwif_svwks,
-		.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_SINGLE |
-				  IDE_HFLAG_BOOTABLE,
+		.host_flags	= IDE_HFLAGS_SVWKS | IDE_HFLAG_SINGLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
@@ -404,8 +408,7 @@ static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 		.name		= "SvrWks HT1000",
 		.init_chipset	= init_chipset_svwks,
 		.init_hwif	= init_hwif_svwks,
-		.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_SINGLE |
-				  IDE_HFLAG_BOOTABLE,
+		.host_flags	= IDE_HFLAGS_SVWKS | IDE_HFLAG_SINGLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 5709c252543b..7b45eaf5afd9 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -278,27 +278,14 @@ static void sil_set_dma_mode(ide_drive_t *drive, const u8 speed)
 
 	scsc = is_sata(hwif) ? 1 : scsc;
 
-	switch(speed) {
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_MW_DMA_0:
-			multi = dma[speed - XFER_MW_DMA_0];
-			mode |= ((unit) ? 0x20 : 0x02);
-			break;
-		case XFER_UDMA_6:
-		case XFER_UDMA_5:
-		case XFER_UDMA_4:
-		case XFER_UDMA_3:
-		case XFER_UDMA_2:
-		case XFER_UDMA_1:
-		case XFER_UDMA_0:
-			multi = dma[2];
-			ultra |= ((scsc) ? (ultra6[speed - XFER_UDMA_0]) :
-					   (ultra5[speed - XFER_UDMA_0]));
-			mode |= ((unit) ? 0x30 : 0x03);
-			break;
-		default:
-			return;
+	if (speed >= XFER_UDMA_0) {
+		multi = dma[2];
+		ultra |= (scsc ? ultra6[speed - XFER_UDMA_0] :
+				 ultra5[speed - XFER_UDMA_0]);
+		mode |= (unit ? 0x30 : 0x03);
+	} else {
+		multi = dma[speed - XFER_MW_DMA_0];
+		mode |= (unit ? 0x20 : 0x02);
 	}
 
 	if (hwif->mmio) {
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index 3f35386d9cad..85d36996e6af 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -351,25 +351,10 @@ static void sis_program_udma_timings(ide_drive_t *drive, const u8 mode)
 
 static void sis_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	/* Config chip for mode */
-	switch(speed) {
-		case XFER_UDMA_6:
-		case XFER_UDMA_5:
-		case XFER_UDMA_4:
-		case XFER_UDMA_3:
-		case XFER_UDMA_2:
-		case XFER_UDMA_1:
-		case XFER_UDMA_0:
-			sis_program_udma_timings(drive, speed);
-			break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_MW_DMA_0:
-			sis_program_timings(drive, speed);
-			break;
-		default:
-			break;
-	}
+	if (speed >= XFER_UDMA_0)
+		sis_program_udma_timings(drive, speed);
+	else
+		sis_program_timings(drive, speed);
 }
 
 static u8 sis5513_ata133_udma_filter(ide_drive_t *drive)
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index 147d783f7529..069f104fdcea 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -115,32 +115,24 @@ static void sl82c105_set_dma_mode(ide_drive_t *drive, const u8 speed)
  	DBG(("sl82c105_tune_chipset(drive:%s, speed:%s)\n",
 	     drive->name, ide_xfer_verbose(speed)));
 
-	switch (speed) {
-	case XFER_MW_DMA_2:
-	case XFER_MW_DMA_1:
-	case XFER_MW_DMA_0:
-		drv_ctrl = mwdma_timings[speed - XFER_MW_DMA_0];
+	drv_ctrl = mwdma_timings[speed - XFER_MW_DMA_0];
 
-		/*
-		 * Store the DMA timings so that we can actually program
-		 * them when DMA will be turned on...
-		 */
-		drive->drive_data &= 0x0000ffff;
-		drive->drive_data |= (unsigned long)drv_ctrl << 16;
+	/*
+	 * Store the DMA timings so that we can actually program
+	 * them when DMA will be turned on...
+	 */
+	drive->drive_data &= 0x0000ffff;
+	drive->drive_data |= (unsigned long)drv_ctrl << 16;
 
-		/*
-		 * If we are already using DMA, we just reprogram
-		 * the drive control register.
-		 */
-		if (drive->using_dma) {
-			struct pci_dev *dev	= HWIF(drive)->pci_dev;
-			int reg 		= 0x44 + drive->dn * 4;
-
-			pci_write_config_word(dev, reg, drv_ctrl);
-		}
-		break;
-	default:
-		return;
+	/*
+	 * If we are already using DMA, we just reprogram
+	 * the drive control register.
+	 */
+	if (drive->using_dma) {
+		struct pci_dev *dev	= HWIF(drive)->pci_dev;
+		int reg 		= 0x44 + drive->dn * 4;
+
+		pci_write_config_word(dev, reg, drv_ctrl);
 	}
 }
 
diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c
index eb4445b229ed..dbbb46819a2d 100644
--- a/drivers/ide/pci/slc90e66.c
+++ b/drivers/ide/pci/slc90e66.c
@@ -91,19 +91,9 @@ static void slc90e66_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	pci_read_config_word(dev, 0x48, &reg48);
 	pci_read_config_word(dev, 0x4a, &reg4a);
 
-	switch(speed) {
-		case XFER_UDMA_4:	u_speed = 4 << (drive->dn * 4); break;
-		case XFER_UDMA_3:	u_speed = 3 << (drive->dn * 4); break;
-		case XFER_UDMA_2:	u_speed = 2 << (drive->dn * 4); break;
-		case XFER_UDMA_1:	u_speed = 1 << (drive->dn * 4); break;
-		case XFER_UDMA_0:	u_speed = 0 << (drive->dn * 4); break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_SW_DMA_2:	break;
-		default:		return;
-	}
-
 	if (speed >= XFER_UDMA_0) {
+		u_speed = (speed - XFER_UDMA_0) << (drive->dn * 4);
+
 		if (!(reg48 & u_flag))
 			pci_write_config_word(dev, 0x48, reg48|u_flag);
 		/* FIXME: (reg4a & a_speed) ? */
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index a66ebd14664e..e1faf6c2fe16 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -222,7 +222,8 @@ static const struct ide_port_info tc86c001_chipset __devinitdata = {
 	.name		= "TC86C001",
 	.init_chipset	= init_chipset_tc86c001,
 	.init_hwif	= init_hwif_tc86c001,
-	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_OFF_BOARD,
+	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_OFF_BOARD |
+			  IDE_HFLAG_ABUSE_SET_DMA_MODE,
 	.pio_mask	= ATA_PIO4,
 	.mwdma_mask	= ATA_MWDMA2,
 	.udma_mask	= ATA_UDMA4,
diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c
index a227c41d23a3..ae52a96a1cf9 100644
--- a/drivers/ide/pci/triflex.c
+++ b/drivers/ide/pci/triflex.c
@@ -81,8 +81,6 @@ static void triflex_set_mode(ide_drive_t *drive, const u8 speed)
 		case XFER_PIO_0:
 			timing = 0x0808;
 			break;
-		default:
-			return;
 	}
 
 	triflex_timings &= ~(0xFFFF << (16 * unit));
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index a0d3c16b68ec..4b32c90f4896 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -439,6 +439,7 @@ static const struct ide_port_info via82cxxx_chipset __devinitdata = {
 	.enablebits	= { { 0x40, 0x02, 0x02 }, { 0x40, 0x01, 0x01 } },
 	.host_flags	= IDE_HFLAG_PIO_NO_BLACKLIST |
 			  IDE_HFLAG_PIO_NO_DOWNGRADE |
+			  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 			  IDE_HFLAG_POST_SET_MODE |
 			  IDE_HFLAG_IO_32BIT |
 			  IDE_HFLAG_BOOTABLE,
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 4559e29446e9..3dce80092fff 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -828,38 +828,20 @@ static void pmac_ide_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	tl[0] = *timings;
 	tl[1] = *timings2;
 
-	switch(speed) {
 #ifdef CONFIG_BLK_DEV_IDEDMA_PMAC
-		case XFER_UDMA_6:
-		case XFER_UDMA_5:
-		case XFER_UDMA_4:
-		case XFER_UDMA_3:
-		case XFER_UDMA_2:
-		case XFER_UDMA_1:
-		case XFER_UDMA_0:
-			if (pmif->kind == controller_kl_ata4)
-				ret = set_timings_udma_ata4(&tl[0], speed);
-			else if (pmif->kind == controller_un_ata6
-				 || pmif->kind == controller_k2_ata6)
-				ret = set_timings_udma_ata6(&tl[0], &tl[1], speed);
-			else if (pmif->kind == controller_sh_ata6)
-				ret = set_timings_udma_shasta(&tl[0], &tl[1], speed);
-			else
-				ret = 1;
-			break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:
-		case XFER_MW_DMA_0:
-			set_timings_mdma(drive, pmif->kind, &tl[0], &tl[1], speed);
-			break;
-		case XFER_SW_DMA_2:
-		case XFER_SW_DMA_1:
-		case XFER_SW_DMA_0:
-			return;
+	if (speed >= XFER_UDMA_0) {
+		if (pmif->kind == controller_kl_ata4)
+			ret = set_timings_udma_ata4(&tl[0], speed);
+		else if (pmif->kind == controller_un_ata6
+			 || pmif->kind == controller_k2_ata6)
+			ret = set_timings_udma_ata6(&tl[0], &tl[1], speed);
+		else if (pmif->kind == controller_sh_ata6)
+			ret = set_timings_udma_shasta(&tl[0], &tl[1], speed);
+		else
+			ret = -1;
+	} else
+		set_timings_mdma(drive, pmif->kind, &tl[0], &tl[1], speed);
 #endif /* CONFIG_BLK_DEV_IDEDMA_PMAC */
-		default:
-			ret = 1;
-	}
 	if (ret)
 		return;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 6f5b6b5e9c1d..1e4409937ec3 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1094,6 +1094,7 @@ enum {
 	IDE_HFLAG_IO_32BIT		= (1 << 24),
 	/* unmask IRQs */
 	IDE_HFLAG_UNMASK_IRQS		= (1 << 25),
+	IDE_HFLAG_ABUSE_SET_DMA_MODE	= (1 << 26),
 };
 
 #ifdef CONFIG_BLK_DEV_OFFBOARD
-- 
cgit v1.2.3


From 4fc8cd4919428f9b86f0b65e2f3245a1c186737f Mon Sep 17 00:00:00 2001
From: Sean Hefty <sean.hefty@intel.com>
Date: Tue, 27 Nov 2007 00:11:04 -0800
Subject: IB/mad: Report number of times a mad was retried

To allow ULPs to tune timeout values and capture retry statistics,
report the number of times that a mad send operation was retried.

For RMPP mads, report the total number of times that the any portion
(send window) of the send operation was retried.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/mad.c      | 9 +++++++--
 drivers/infiniband/core/mad_priv.h | 3 ++-
 drivers/infiniband/core/mad_rmpp.c | 2 +-
 include/rdma/ib_mad.h              | 4 +++-
 4 files changed, 13 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 649335a03312..5eace995836d 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -1100,7 +1100,9 @@ int ib_post_send_mad(struct ib_mad_send_buf *send_buf,
 		mad_send_wr->tid = ((struct ib_mad_hdr *) send_buf->mad)->tid;
 		/* Timeout will be updated after send completes */
 		mad_send_wr->timeout = msecs_to_jiffies(send_buf->timeout_ms);
-		mad_send_wr->retries = send_buf->retries;
+		mad_send_wr->max_retries = send_buf->retries;
+		mad_send_wr->retries_left = send_buf->retries;
+		send_buf->retries = 0;
 		/* Reference for work request to QP + response */
 		mad_send_wr->refcount = 1 + (mad_send_wr->timeout > 0);
 		mad_send_wr->status = IB_WC_SUCCESS;
@@ -2436,9 +2438,12 @@ static int retry_send(struct ib_mad_send_wr_private *mad_send_wr)
 {
 	int ret;
 
-	if (!mad_send_wr->retries--)
+	if (!mad_send_wr->retries_left)
 		return -ETIMEDOUT;
 
+	mad_send_wr->retries_left--;
+	mad_send_wr->send_buf.retries++;
+
 	mad_send_wr->timeout = msecs_to_jiffies(mad_send_wr->send_buf.timeout_ms);
 
 	if (mad_send_wr->mad_agent_priv->agent.rmpp_version) {
diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h
index 9be5cc00a3a9..8b75010016ec 100644
--- a/drivers/infiniband/core/mad_priv.h
+++ b/drivers/infiniband/core/mad_priv.h
@@ -131,7 +131,8 @@ struct ib_mad_send_wr_private {
 	struct ib_sge sg_list[IB_MAD_SEND_REQ_MAX_SG];
 	__be64 tid;
 	unsigned long timeout;
-	int retries;
+	int max_retries;
+	int retries_left;
 	int retry;
 	int refcount;
 	enum ib_wc_status status;
diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c
index d43bc62005b3..a5e2a310f312 100644
--- a/drivers/infiniband/core/mad_rmpp.c
+++ b/drivers/infiniband/core/mad_rmpp.c
@@ -684,7 +684,7 @@ static void process_rmpp_ack(struct ib_mad_agent_private *agent,
 
 	if (seg_num > mad_send_wr->last_ack) {
 		adjust_last_ack(mad_send_wr, seg_num);
-		mad_send_wr->retries = mad_send_wr->send_buf.retries;
+		mad_send_wr->retries_left = mad_send_wr->max_retries;
 	}
 	mad_send_wr->newwin = newwin;
 	if (mad_send_wr->last_ack == mad_send_wr->send_buf.seg_count) {
diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h
index 8ec3799e42e1..7228c056b9e9 100644
--- a/include/rdma/ib_mad.h
+++ b/include/rdma/ib_mad.h
@@ -230,7 +230,9 @@ struct ib_class_port_info
  * @seg_count: The number of RMPP segments allocated for this send.
  * @seg_size: Size of each RMPP segment.
  * @timeout_ms: Time to wait for a response.
- * @retries: Number of times to retry a request for a response.
+ * @retries: Number of times to retry a request for a response.  For MADs
+ *   using RMPP, this applies per window.  On completion, returns the number
+ *   of retries needed to complete the transfer.
  *
  * Users are responsible for initializing the MAD buffer itself, with the
  * exception of any RMPP header.  Additional segment buffer space allocated
-- 
cgit v1.2.3


From 88314e4dda1e158aabce76429ef4d017b48f8b92 Mon Sep 17 00:00:00 2001
From: Sean Hefty <sean.hefty@intel.com>
Date: Wed, 14 Nov 2007 00:29:50 -0800
Subject: RDMA/cma: add support for rdma_migrate_id()

This is based on user feedback from Doug Ledford at RedHat:

Events that occur on an rdma_cm_id are reported to userspace through an
event channel.  Connection request events are reported on the event
channel associated with the listen.  When the connection is accepted, a
new rdma_cm_id is created and automatically uses the listen event
channel.  This is suboptimal where the user only wants listen events on
that channel.

Additionally, it may be desirable to have events related to connection
establishment use a different event channel than those related to
already established connections.

Allow the user to migrate an rdma_cm_id between event channels. All
pending events associated with the rdma_cm_id are moved to the new event
channel.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/ucma.c | 92 ++++++++++++++++++++++++++++++++++++++++++
 include/rdma/rdma_user_cm.h    | 13 +++++-
 2 files changed, 104 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c
index 90d675ad9ec8..15937eb38aae 100644
--- a/drivers/infiniband/core/ucma.c
+++ b/drivers/infiniband/core/ucma.c
@@ -31,6 +31,7 @@
  */
 
 #include <linux/completion.h>
+#include <linux/file.h>
 #include <linux/mutex.h>
 #include <linux/poll.h>
 #include <linux/idr.h>
@@ -991,6 +992,96 @@ out:
 	return ret;
 }
 
+static void ucma_lock_files(struct ucma_file *file1, struct ucma_file *file2)
+{
+	/* Acquire mutex's based on pointer comparison to prevent deadlock. */
+	if (file1 < file2) {
+		mutex_lock(&file1->mut);
+		mutex_lock(&file2->mut);
+	} else {
+		mutex_lock(&file2->mut);
+		mutex_lock(&file1->mut);
+	}
+}
+
+static void ucma_unlock_files(struct ucma_file *file1, struct ucma_file *file2)
+{
+	if (file1 < file2) {
+		mutex_unlock(&file2->mut);
+		mutex_unlock(&file1->mut);
+	} else {
+		mutex_unlock(&file1->mut);
+		mutex_unlock(&file2->mut);
+	}
+}
+
+static void ucma_move_events(struct ucma_context *ctx, struct ucma_file *file)
+{
+	struct ucma_event *uevent, *tmp;
+
+	list_for_each_entry_safe(uevent, tmp, &ctx->file->event_list, list)
+		if (uevent->ctx == ctx)
+			list_move_tail(&uevent->list, &file->event_list);
+}
+
+static ssize_t ucma_migrate_id(struct ucma_file *new_file,
+			       const char __user *inbuf,
+			       int in_len, int out_len)
+{
+	struct rdma_ucm_migrate_id cmd;
+	struct rdma_ucm_migrate_resp resp;
+	struct ucma_context *ctx;
+	struct file *filp;
+	struct ucma_file *cur_file;
+	int ret = 0;
+
+	if (copy_from_user(&cmd, inbuf, sizeof(cmd)))
+		return -EFAULT;
+
+	/* Get current fd to protect against it being closed */
+	filp = fget(cmd.fd);
+	if (!filp)
+		return -ENOENT;
+
+	/* Validate current fd and prevent destruction of id. */
+	ctx = ucma_get_ctx(filp->private_data, cmd.id);
+	if (IS_ERR(ctx)) {
+		ret = PTR_ERR(ctx);
+		goto file_put;
+	}
+
+	cur_file = ctx->file;
+	if (cur_file == new_file) {
+		resp.events_reported = ctx->events_reported;
+		goto response;
+	}
+
+	/*
+	 * Migrate events between fd's, maintaining order, and avoiding new
+	 * events being added before existing events.
+	 */
+	ucma_lock_files(cur_file, new_file);
+	mutex_lock(&mut);
+
+	list_move_tail(&ctx->list, &new_file->ctx_list);
+	ucma_move_events(ctx, new_file);
+	ctx->file = new_file;
+	resp.events_reported = ctx->events_reported;
+
+	mutex_unlock(&mut);
+	ucma_unlock_files(cur_file, new_file);
+
+response:
+	if (copy_to_user((void __user *)(unsigned long)cmd.response,
+			 &resp, sizeof(resp)))
+		ret = -EFAULT;
+
+	ucma_put_ctx(ctx);
+file_put:
+	fput(filp);
+	return ret;
+}
+
 static ssize_t (*ucma_cmd_table[])(struct ucma_file *file,
 				   const char __user *inbuf,
 				   int in_len, int out_len) = {
@@ -1012,6 +1103,7 @@ static ssize_t (*ucma_cmd_table[])(struct ucma_file *file,
 	[RDMA_USER_CM_CMD_NOTIFY]	= ucma_notify,
 	[RDMA_USER_CM_CMD_JOIN_MCAST]	= ucma_join_multicast,
 	[RDMA_USER_CM_CMD_LEAVE_MCAST]	= ucma_leave_multicast,
+	[RDMA_USER_CM_CMD_MIGRATE_ID]	= ucma_migrate_id
 };
 
 static ssize_t ucma_write(struct file *filp, const char __user *buf,
diff --git a/include/rdma/rdma_user_cm.h b/include/rdma/rdma_user_cm.h
index 9749c1b34d00..c55705460b87 100644
--- a/include/rdma/rdma_user_cm.h
+++ b/include/rdma/rdma_user_cm.h
@@ -60,7 +60,8 @@ enum {
 	RDMA_USER_CM_CMD_SET_OPTION,
 	RDMA_USER_CM_CMD_NOTIFY,
 	RDMA_USER_CM_CMD_JOIN_MCAST,
-	RDMA_USER_CM_CMD_LEAVE_MCAST
+	RDMA_USER_CM_CMD_LEAVE_MCAST,
+	RDMA_USER_CM_CMD_MIGRATE_ID
 };
 
 /*
@@ -230,4 +231,14 @@ struct rdma_ucm_set_option {
 	__u32 optlen;
 };
 
+struct rdma_ucm_migrate_id {
+	__u64 response;
+	__u32 id;
+	__u32 fd;
+};
+
+struct rdma_ucm_migrate_resp {
+	__u32 events_reported;
+};
+
 #endif /* RDMA_USER_CM_H */
-- 
cgit v1.2.3


From a9e527e3f9f4510e9f3450ca3bc51bc3ef2854fd Mon Sep 17 00:00:00 2001
From: Rolf Manderscheid <rvm@obsidianresearch.com>
Date: Mon, 10 Dec 2007 13:38:41 -0700
Subject: IPoIB: improve IPv4/IPv6 to IB mcast mapping functions

An IPoIB subnet on an IB fabric that spans multiple IB subnets can't
use link-local scope in multicast GIDs.  The existing routines that
map IP/IPv6 multicast addresses into IB link-level addresses hard-code
the scope to link-local, and they also leave the partition key field
uninitialised.  This patch adds a parameter (the link-level broadcast
address) to the mapping routines, allowing them to initialise both the
scope and the P_Key appropriately, and fixes up the call sites.

The next step will be to add a way to configure the scope for an IPoIB
interface.

Signed-off-by: Rolf Manderscheid <rvm@obsidianresearch.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/cma.c                  |  4 +---
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c |  4 ----
 include/net/if_inet6.h                         | 11 +++++++----
 include/net/ip.h                               | 10 ++++++----
 net/ipv4/arp.c                                 |  2 +-
 net/ipv6/ndisc.c                               |  2 +-
 6 files changed, 16 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 312ec74f3d18..982836e69f55 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -2610,11 +2610,9 @@ static void cma_set_mgid(struct rdma_id_private *id_priv,
 		/* IPv6 address is an SA assigned MGID. */
 		memcpy(mgid, &sin6->sin6_addr, sizeof *mgid);
 	} else {
-		ip_ib_mc_map(sin->sin_addr.s_addr, mc_map);
+		ip_ib_mc_map(sin->sin_addr.s_addr, dev_addr->broadcast, mc_map);
 		if (id_priv->id.ps == RDMA_PS_UDP)
 			mc_map[7] = 0x01;	/* Use RDMA CM signature */
-		mc_map[8] = ib_addr_get_pkey(dev_addr) >> 8;
-		mc_map[9] = (unsigned char) ib_addr_get_pkey(dev_addr);
 		*mgid = *(union ib_gid *) (mc_map + 4);
 	}
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 858ada17f980..2628339e3a99 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -788,10 +788,6 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 
 		memcpy(mgid.raw, mclist->dmi_addr + 4, sizeof mgid);
 
-		/* Add in the P_Key */
-		mgid.raw[4] = (priv->pkey >> 8) & 0xff;
-		mgid.raw[5] = priv->pkey & 0xff;
-
 		mcast = __ipoib_mcast_find(dev, &mgid);
 		if (!mcast || test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) {
 			struct ipoib_mcast *nmcast;
diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index 448eccb20638..b24508abb850 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -269,18 +269,21 @@ static inline void ipv6_arcnet_mc_map(const struct in6_addr *addr, char *buf)
 	buf[0] = 0x00;
 }
 
-static inline void ipv6_ib_mc_map(struct in6_addr *addr, char *buf)
+static inline void ipv6_ib_mc_map(const struct in6_addr *addr,
+				  const unsigned char *broadcast, char *buf)
 {
+	unsigned char scope = broadcast[5] & 0xF;
+
 	buf[0]  = 0;		/* Reserved */
 	buf[1]  = 0xff;		/* Multicast QPN */
 	buf[2]  = 0xff;
 	buf[3]  = 0xff;
 	buf[4]  = 0xff;
-	buf[5]  = 0x12;		/* link local scope */
+	buf[5]  = 0x10 | scope;	/* scope from broadcast address */
 	buf[6]  = 0x60;		/* IPv6 signature */
 	buf[7]  = 0x1b;
-	buf[8]  = 0;		/* P_Key */
-	buf[9]  = 0;
+	buf[8]  = broadcast[8];	/* P_Key */
+	buf[9]  = broadcast[9];
 	memcpy(buf + 10, addr->s6_addr + 6, 10);
 }
 #endif
diff --git a/include/net/ip.h b/include/net/ip.h
index 840dd91b513b..50c8889b1b8d 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -266,20 +266,22 @@ static inline void ip_eth_mc_map(__be32 naddr, char *buf)
  *	Leave P_Key as 0 to be filled in by driver.
  */
 
-static inline void ip_ib_mc_map(__be32 naddr, char *buf)
+static inline void ip_ib_mc_map(__be32 naddr, const unsigned char *broadcast, char *buf)
 {
 	__u32 addr;
+	unsigned char scope = broadcast[5] & 0xF;
+
 	buf[0]  = 0;		/* Reserved */
 	buf[1]  = 0xff;		/* Multicast QPN */
 	buf[2]  = 0xff;
 	buf[3]  = 0xff;
 	addr    = ntohl(naddr);
 	buf[4]  = 0xff;
-	buf[5]  = 0x12;		/* link local scope */
+	buf[5]  = 0x10 | scope;	/* scope from broadcast address */
 	buf[6]  = 0x40;		/* IPv4 signature */
 	buf[7]  = 0x1b;
-	buf[8]  = 0;		/* P_Key */
-	buf[9]  = 0;
+	buf[8]  = broadcast[8];		/* P_Key */
+	buf[9]  = broadcast[9];
 	buf[10] = 0;
 	buf[11] = 0;
 	buf[12] = 0;
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 08174a2aa878..54a76b8b803a 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -211,7 +211,7 @@ int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir)
 		ip_tr_mc_map(addr, haddr);
 		return 0;
 	case ARPHRD_INFINIBAND:
-		ip_ib_mc_map(addr, haddr);
+		ip_ib_mc_map(addr, dev->broadcast, haddr);
 		return 0;
 	default:
 		if (dir) {
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 777ed733b2d7..85947eae5bf7 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -337,7 +337,7 @@ int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int d
 		ipv6_arcnet_mc_map(addr, buf);
 		return 0;
 	case ARPHRD_INFINIBAND:
-		ipv6_ib_mc_map(addr, buf);
+		ipv6_ib_mc_map(addr, dev->broadcast, buf);
 		return 0;
 	default:
 		if (dir) {
-- 
cgit v1.2.3


From d69a3ad6a0e47b2aa9b2b2ddfd385752132a4d34 Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Fri, 5 Oct 2007 14:31:44 -0700
Subject: dlm: Split lock mode and flag constants into a sharable header.

This allows others to use the DLM constants without being tied to the
function API of fs/dlm.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 include/linux/Kbuild         |   1 +
 include/linux/dlm.h          | 140 +------------------------------------
 include/linux/dlmconstants.h | 159 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 162 insertions(+), 138 deletions(-)
 create mode 100644 include/linux/dlmconstants.h

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index f30fa92a44a1..bd694f779346 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -49,6 +49,7 @@ header-y += comstats.h
 header-y += const.h
 header-y += cgroupstats.h
 header-y += cycx_cfm.h
+header-y += dlmconstants.h
 header-y += dlm_device.h
 header-y += dlm_netlink.h
 header-y += dm-ioctl.h
diff --git a/include/linux/dlm.h b/include/linux/dlm.h
index be9d278761e0..c743fbc769db 100644
--- a/include/linux/dlm.h
+++ b/include/linux/dlm.h
@@ -19,148 +19,12 @@
  * routines and structures to use DLM lockspaces
  */
 
-/*
- * Lock Modes
- */
+/* Lock levels and flags are here */
+#include <linux/dlmconstants.h>
 
-#define DLM_LOCK_IV		-1	/* invalid */
-#define DLM_LOCK_NL		0	/* null */
-#define DLM_LOCK_CR		1	/* concurrent read */
-#define DLM_LOCK_CW		2	/* concurrent write */
-#define DLM_LOCK_PR		3	/* protected read */
-#define DLM_LOCK_PW		4	/* protected write */
-#define DLM_LOCK_EX		5	/* exclusive */
-
-/*
- * Maximum size in bytes of a dlm_lock name
- */
 
 #define DLM_RESNAME_MAXLEN	64
 
-/*
- * Flags to dlm_lock
- *
- * DLM_LKF_NOQUEUE
- *
- * Do not queue the lock request on the wait queue if it cannot be granted
- * immediately.  If the lock cannot be granted because of this flag, DLM will
- * either return -EAGAIN from the dlm_lock call or will return 0 from
- * dlm_lock and -EAGAIN in the lock status block when the AST is executed.
- *
- * DLM_LKF_CANCEL
- *
- * Used to cancel a pending lock request or conversion.  A converting lock is
- * returned to its previously granted mode.
- *
- * DLM_LKF_CONVERT
- *
- * Indicates a lock conversion request.  For conversions the name and namelen
- * are ignored and the lock ID in the LKSB is used to identify the lock.
- *
- * DLM_LKF_VALBLK
- *
- * Requests DLM to return the current contents of the lock value block in the
- * lock status block.  When this flag is set in a lock conversion from PW or EX
- * modes, DLM assigns the value specified in the lock status block to the lock
- * value block of the lock resource.  The LVB is a DLM_LVB_LEN size array
- * containing application-specific information.
- *
- * DLM_LKF_QUECVT
- *
- * Force a conversion request to be queued, even if it is compatible with
- * the granted modes of other locks on the same resource.
- *
- * DLM_LKF_IVVALBLK
- *
- * Invalidate the lock value block.
- *
- * DLM_LKF_CONVDEADLK
- *
- * Allows the dlm to resolve conversion deadlocks internally by demoting the
- * granted mode of a converting lock to NL.  The DLM_SBF_DEMOTED flag is
- * returned for a conversion that's been effected by this.
- *
- * DLM_LKF_PERSISTENT
- *
- * Only relevant to locks originating in userspace.  A persistent lock will not
- * be removed if the process holding the lock exits.
- *
- * DLM_LKF_NODLCKWT
- *
- * Do not cancel the lock if it gets into conversion deadlock.
- * Exclude this lock from being monitored due to DLM_LSFL_TIMEWARN.
- *
- * DLM_LKF_NODLCKBLK
- *
- * net yet implemented
- *
- * DLM_LKF_EXPEDITE
- *
- * Used only with new requests for NL mode locks.  Tells the lock manager
- * to grant the lock, ignoring other locks in convert and wait queues.
- *
- * DLM_LKF_NOQUEUEBAST
- *
- * Send blocking AST's before returning -EAGAIN to the caller.  It is only
- * used along with the NOQUEUE flag.  Blocking AST's are not sent for failed
- * NOQUEUE requests otherwise.
- *
- * DLM_LKF_HEADQUE
- *
- * Add a lock to the head of the convert or wait queue rather than the tail.
- *
- * DLM_LKF_NOORDER
- *
- * Disregard the standard grant order rules and grant a lock as soon as it
- * is compatible with other granted locks.
- *
- * DLM_LKF_ORPHAN
- *
- * not yet implemented
- *
- * DLM_LKF_ALTPR
- *
- * If the requested mode cannot be granted immediately, try to grant the lock
- * in PR mode instead.  If this alternate mode is granted instead of the
- * requested mode, DLM_SBF_ALTMODE is returned in the lksb.
- *
- * DLM_LKF_ALTCW
- *
- * The same as ALTPR, but the alternate mode is CW.
- *
- * DLM_LKF_FORCEUNLOCK
- *
- * Unlock the lock even if it is converting or waiting or has sublocks.
- * Only really for use by the userland device.c code.
- *
- */
-
-#define DLM_LKF_NOQUEUE		0x00000001
-#define DLM_LKF_CANCEL		0x00000002
-#define DLM_LKF_CONVERT		0x00000004
-#define DLM_LKF_VALBLK		0x00000008
-#define DLM_LKF_QUECVT		0x00000010
-#define DLM_LKF_IVVALBLK	0x00000020
-#define DLM_LKF_CONVDEADLK	0x00000040
-#define DLM_LKF_PERSISTENT	0x00000080
-#define DLM_LKF_NODLCKWT	0x00000100
-#define DLM_LKF_NODLCKBLK	0x00000200
-#define DLM_LKF_EXPEDITE	0x00000400
-#define DLM_LKF_NOQUEUEBAST	0x00000800
-#define DLM_LKF_HEADQUE		0x00001000
-#define DLM_LKF_NOORDER		0x00002000
-#define DLM_LKF_ORPHAN		0x00004000
-#define DLM_LKF_ALTPR		0x00008000
-#define DLM_LKF_ALTCW		0x00010000
-#define DLM_LKF_FORCEUNLOCK	0x00020000
-#define DLM_LKF_TIMEOUT		0x00040000
-
-/*
- * Some return codes that are not in errno.h
- */
-
-#define DLM_ECANCEL		0x10001
-#define DLM_EUNLOCK		0x10002
 
 typedef void dlm_lockspace_t;
 
diff --git a/include/linux/dlmconstants.h b/include/linux/dlmconstants.h
new file mode 100644
index 000000000000..fddb3d3ff321
--- /dev/null
+++ b/include/linux/dlmconstants.h
@@ -0,0 +1,159 @@
+/******************************************************************************
+*******************************************************************************
+**
+**  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
+**  Copyright (C) 2004-2007 Red Hat, Inc.  All rights reserved.
+**
+**  This copyrighted material is made available to anyone wishing to use,
+**  modify, copy, or redistribute it subject to the terms and conditions
+**  of the GNU General Public License v.2.
+**
+*******************************************************************************
+******************************************************************************/
+
+#ifndef __DLMCONSTANTS_DOT_H__
+#define __DLMCONSTANTS_DOT_H__
+
+/*
+ * Constants used by DLM interface.
+ */
+
+/*
+ * Lock Modes
+ */
+
+#define DLM_LOCK_IV		(-1)	/* invalid */
+#define DLM_LOCK_NL		0	/* null */
+#define DLM_LOCK_CR		1	/* concurrent read */
+#define DLM_LOCK_CW		2	/* concurrent write */
+#define DLM_LOCK_PR		3	/* protected read */
+#define DLM_LOCK_PW		4	/* protected write */
+#define DLM_LOCK_EX		5	/* exclusive */
+
+
+/*
+ * Flags to dlm_lock
+ *
+ * DLM_LKF_NOQUEUE
+ *
+ * Do not queue the lock request on the wait queue if it cannot be granted
+ * immediately.  If the lock cannot be granted because of this flag, DLM will
+ * either return -EAGAIN from the dlm_lock call or will return 0 from
+ * dlm_lock and -EAGAIN in the lock status block when the AST is executed.
+ *
+ * DLM_LKF_CANCEL
+ *
+ * Used to cancel a pending lock request or conversion.  A converting lock is
+ * returned to its previously granted mode.
+ *
+ * DLM_LKF_CONVERT
+ *
+ * Indicates a lock conversion request.  For conversions the name and namelen
+ * are ignored and the lock ID in the LKSB is used to identify the lock.
+ *
+ * DLM_LKF_VALBLK
+ *
+ * Requests DLM to return the current contents of the lock value block in the
+ * lock status block.  When this flag is set in a lock conversion from PW or EX
+ * modes, DLM assigns the value specified in the lock status block to the lock
+ * value block of the lock resource.  The LVB is a DLM_LVB_LEN size array
+ * containing application-specific information.
+ *
+ * DLM_LKF_QUECVT
+ *
+ * Force a conversion request to be queued, even if it is compatible with
+ * the granted modes of other locks on the same resource.
+ *
+ * DLM_LKF_IVVALBLK
+ *
+ * Invalidate the lock value block.
+ *
+ * DLM_LKF_CONVDEADLK
+ *
+ * Allows the dlm to resolve conversion deadlocks internally by demoting the
+ * granted mode of a converting lock to NL.  The DLM_SBF_DEMOTED flag is
+ * returned for a conversion that's been effected by this.
+ *
+ * DLM_LKF_PERSISTENT
+ *
+ * Only relevant to locks originating in userspace.  A persistent lock will not
+ * be removed if the process holding the lock exits.
+ *
+ * DLM_LKF_NODLCKWT
+ *
+ * Do not cancel the lock if it gets into conversion deadlock.
+ * Exclude this lock from being monitored due to DLM_LSFL_TIMEWARN.
+ *
+ * DLM_LKF_NODLCKBLK
+ *
+ * net yet implemented
+ *
+ * DLM_LKF_EXPEDITE
+ *
+ * Used only with new requests for NL mode locks.  Tells the lock manager
+ * to grant the lock, ignoring other locks in convert and wait queues.
+ *
+ * DLM_LKF_NOQUEUEBAST
+ *
+ * Send blocking AST's before returning -EAGAIN to the caller.  It is only
+ * used along with the NOQUEUE flag.  Blocking AST's are not sent for failed
+ * NOQUEUE requests otherwise.
+ *
+ * DLM_LKF_HEADQUE
+ *
+ * Add a lock to the head of the convert or wait queue rather than the tail.
+ *
+ * DLM_LKF_NOORDER
+ *
+ * Disregard the standard grant order rules and grant a lock as soon as it
+ * is compatible with other granted locks.
+ *
+ * DLM_LKF_ORPHAN
+ *
+ * not yet implemented
+ *
+ * DLM_LKF_ALTPR
+ *
+ * If the requested mode cannot be granted immediately, try to grant the lock
+ * in PR mode instead.  If this alternate mode is granted instead of the
+ * requested mode, DLM_SBF_ALTMODE is returned in the lksb.
+ *
+ * DLM_LKF_ALTCW
+ *
+ * The same as ALTPR, but the alternate mode is CW.
+ *
+ * DLM_LKF_FORCEUNLOCK
+ *
+ * Unlock the lock even if it is converting or waiting or has sublocks.
+ * Only really for use by the userland device.c code.
+ *
+ */
+
+#define DLM_LKF_NOQUEUE		0x00000001
+#define DLM_LKF_CANCEL		0x00000002
+#define DLM_LKF_CONVERT		0x00000004
+#define DLM_LKF_VALBLK		0x00000008
+#define DLM_LKF_QUECVT		0x00000010
+#define DLM_LKF_IVVALBLK	0x00000020
+#define DLM_LKF_CONVDEADLK	0x00000040
+#define DLM_LKF_PERSISTENT	0x00000080
+#define DLM_LKF_NODLCKWT	0x00000100
+#define DLM_LKF_NODLCKBLK	0x00000200
+#define DLM_LKF_EXPEDITE	0x00000400
+#define DLM_LKF_NOQUEUEBAST	0x00000800
+#define DLM_LKF_HEADQUE		0x00001000
+#define DLM_LKF_NOORDER		0x00002000
+#define DLM_LKF_ORPHAN		0x00004000
+#define DLM_LKF_ALTPR		0x00008000
+#define DLM_LKF_ALTCW		0x00010000
+#define DLM_LKF_FORCEUNLOCK	0x00020000
+#define DLM_LKF_TIMEOUT		0x00040000
+
+/*
+ * Some return codes that are not in errno.h
+ */
+
+#define DLM_ECANCEL		0x10001
+#define DLM_EUNLOCK		0x10002
+
+#endif  /* __DLMCONSTANTS_DOT_H__ */
-- 
cgit v1.2.3


From 4e8e56c6713398f417317d449f50c08bf2756c66 Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Date: Sat, 26 Jan 2008 14:10:44 +0100
Subject: [S390] cio: Extend adapter interrupt interface.

From: Cornelia Huck <cornelia.huck@de.ibm.com>

Change the adapter interrupt interface in order to allow multiple
adapter interrupt handlers to be registered. Indicators are now
allocated by cio instead of the device driver.

The qdio parts have been
Acked-by: Ursula Braun <ubraun@linux.vnet.ibm.com>

Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 Documentation/DocBook/s390-drivers.tmpl |   1 +
 drivers/s390/cio/airq.c                 | 171 +++++++++++++++++++++-----------
 drivers/s390/cio/airq.h                 |  10 --
 drivers/s390/cio/cio.c                  |   2 +-
 drivers/s390/cio/cio.h                  |   1 +
 drivers/s390/cio/qdio.c                 |  35 ++++---
 include/asm-s390/airq.h                 |  19 ++++
 7 files changed, 154 insertions(+), 85 deletions(-)
 delete mode 100644 drivers/s390/cio/airq.h
 create mode 100644 include/asm-s390/airq.h

(limited to 'include')

diff --git a/Documentation/DocBook/s390-drivers.tmpl b/Documentation/DocBook/s390-drivers.tmpl
index 254e769282a4..3d2f31b99dd9 100644
--- a/Documentation/DocBook/s390-drivers.tmpl
+++ b/Documentation/DocBook/s390-drivers.tmpl
@@ -116,6 +116,7 @@
 !Iinclude/asm-s390/ccwdev.h
 !Edrivers/s390/cio/device.c
 !Edrivers/s390/cio/device_ops.c
+!Edrivers/s390/cio/airq.c
     </sect1>
     <sect1 id="cmf">
      <title>The channel-measurement facility</title>
diff --git a/drivers/s390/cio/airq.c b/drivers/s390/cio/airq.c
index 5287631fbfc8..b7a07a866291 100644
--- a/drivers/s390/cio/airq.c
+++ b/drivers/s390/cio/airq.c
@@ -1,12 +1,12 @@
 /*
  *  drivers/s390/cio/airq.c
- *   S/390 common I/O routines -- support for adapter interruptions
+ *    Support for adapter interruptions
  *
- *    Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH,
- *			      IBM Corporation
- *    Author(s): Ingo Adlung (adlung@de.ibm.com)
- *		 Cornelia Huck (cornelia.huck@de.ibm.com)
- *		 Arnd Bergmann (arndb@de.ibm.com)
+ *    Copyright IBM Corp. 1999,2007
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>
+ *		 Arnd Bergmann <arndb@de.ibm.com>
+ *		 Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
  */
 
 #include <linux/init.h>
@@ -14,72 +14,131 @@
 #include <linux/slab.h>
 #include <linux/rcupdate.h>
 
+#include <asm/airq.h>
+
+#include "cio.h"
 #include "cio_debug.h"
-#include "airq.h"
 
-static adapter_int_handler_t adapter_handler;
+#define NR_AIRQS		32
+#define NR_AIRQS_PER_WORD	sizeof(unsigned long)
+#define NR_AIRQ_WORDS		(NR_AIRQS / NR_AIRQS_PER_WORD)
 
-/*
- * register for adapter interrupts
- *
- * With HiperSockets the zSeries architecture provides for
- *  means of adapter interrups, pseudo I/O interrupts that are
- *  not tied to an I/O subchannel, but to an adapter. However,
- *  it doesn't disclose the info how to enable/disable them, but
- *  to recognize them only. Perhaps we should consider them
- *  being shared interrupts, and thus build a linked list
- *  of adapter handlers ... to be evaluated ...
- */
-int
-s390_register_adapter_interrupt (adapter_int_handler_t handler)
-{
-	int ret;
-	char dbf_txt[15];
+union indicator_t {
+	unsigned long word[NR_AIRQ_WORDS];
+	unsigned char byte[NR_AIRQS];
+} __attribute__((packed));
 
-	CIO_TRACE_EVENT (4, "rgaint");
+struct airq_t {
+	adapter_int_handler_t handler;
+	void *drv_data;
+};
 
-	if (handler == NULL)
-		ret = -EINVAL;
-	else
-		ret = (cmpxchg(&adapter_handler, NULL, handler) ? -EBUSY : 0);
-	if (!ret)
-		synchronize_sched();  /* Allow interrupts to complete. */
+static union indicator_t indicators;
+static struct airq_t *airqs[NR_AIRQS];
 
-	sprintf (dbf_txt, "ret:%d", ret);
-	CIO_TRACE_EVENT (4, dbf_txt);
+static int register_airq(struct airq_t *airq)
+{
+	int i;
 
-	return ret;
+	for (i = 0; i < NR_AIRQS; i++)
+		if (!cmpxchg(&airqs[i], NULL, airq))
+			return i;
+	return -ENOMEM;
 }
 
-int
-s390_unregister_adapter_interrupt (adapter_int_handler_t handler)
+/**
+ * s390_register_adapter_interrupt() - register adapter interrupt handler
+ * @handler: adapter handler to be registered
+ * @drv_data: driver data passed with each call to the handler
+ *
+ * Returns:
+ *  Pointer to the indicator to be used on success
+ *  ERR_PTR() if registration failed
+ */
+void *s390_register_adapter_interrupt(adapter_int_handler_t handler,
+				      void *drv_data)
 {
+	struct airq_t *airq;
+	char dbf_txt[16];
 	int ret;
-	char dbf_txt[15];
 
-	CIO_TRACE_EVENT (4, "urgaint");
-
-	if (handler == NULL)
-		ret = -EINVAL;
-	else {
-		adapter_handler = NULL;
-		synchronize_sched();  /* Allow interrupts to complete. */
-		ret = 0;
+	airq = kmalloc(sizeof(struct airq_t), GFP_KERNEL);
+	if (!airq) {
+		ret = -ENOMEM;
+		goto out;
 	}
-	sprintf (dbf_txt, "ret:%d", ret);
-	CIO_TRACE_EVENT (4, dbf_txt);
-
-	return ret;
+	airq->handler = handler;
+	airq->drv_data = drv_data;
+	ret = register_airq(airq);
+	if (ret < 0)
+		kfree(airq);
+out:
+	snprintf(dbf_txt, sizeof(dbf_txt), "rairq:%d", ret);
+	CIO_TRACE_EVENT(4, dbf_txt);
+	if (ret < 0)
+		return ERR_PTR(ret);
+	else
+		return &indicators.byte[ret];
 }
+EXPORT_SYMBOL(s390_register_adapter_interrupt);
 
-void
-do_adapter_IO (void)
+/**
+ * s390_unregister_adapter_interrupt - unregister adapter interrupt handler
+ * @ind: indicator for which the handler is to be unregistered
+ */
+void s390_unregister_adapter_interrupt(void *ind)
 {
-	CIO_TRACE_EVENT (6, "doaio");
+	struct airq_t *airq;
+	char dbf_txt[16];
+	int i;
 
-	if (adapter_handler)
-		(*adapter_handler) ();
+	i = (int) ((addr_t) ind) - ((addr_t) &indicators.byte[0]);
+	snprintf(dbf_txt, sizeof(dbf_txt), "urairq:%d", i);
+	CIO_TRACE_EVENT(4, dbf_txt);
+	indicators.byte[i] = 0;
+	airq = xchg(&airqs[i], NULL);
+	/*
+	 * Allow interrupts to complete. This will ensure that the airq handle
+	 * is no longer referenced by any interrupt handler.
+	 */
+	synchronize_sched();
+	kfree(airq);
 }
+EXPORT_SYMBOL(s390_unregister_adapter_interrupt);
+
+#define INDICATOR_MASK	(0xffUL << ((NR_AIRQS_PER_WORD - 1) * 8))
 
-EXPORT_SYMBOL (s390_register_adapter_interrupt);
-EXPORT_SYMBOL (s390_unregister_adapter_interrupt);
+void do_adapter_IO(void)
+{
+	int w;
+	int i;
+	unsigned long word;
+	struct airq_t *airq;
+
+	/*
+	 * Access indicator array in word-sized chunks to minimize storage
+	 * fetch operations.
+	 */
+	for (w = 0; w < NR_AIRQ_WORDS; w++) {
+		word = indicators.word[w];
+		i = w * NR_AIRQS_PER_WORD;
+		/*
+		 * Check bytes within word for active indicators.
+		 */
+		while (word) {
+			if (word & INDICATOR_MASK) {
+				airq = airqs[i];
+				if (likely(airq))
+					airq->handler(&indicators.byte[i],
+						      airq->drv_data);
+				else
+					/*
+					 * Reset ill-behaved indicator.
+					 */
+					indicators.byte[i] = 0;
+			}
+			word <<= 8;
+			i++;
+		}
+	}
+}
diff --git a/drivers/s390/cio/airq.h b/drivers/s390/cio/airq.h
deleted file mode 100644
index 7d6be3fdcd66..000000000000
--- a/drivers/s390/cio/airq.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef S390_AINTERRUPT_H
-#define S390_AINTERRUPT_H
-
-typedef	int (*adapter_int_handler_t)(void);
-
-extern int s390_register_adapter_interrupt(adapter_int_handler_t handler);
-extern int s390_unregister_adapter_interrupt(adapter_int_handler_t handler);
-extern void do_adapter_IO (void);
-
-#endif
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index d0bcebde3fa2..89ced3408138 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -23,7 +23,7 @@
 #include <asm/reset.h>
 #include <asm/ipl.h>
 #include <asm/chpid.h>
-#include "airq.h"
+#include <asm/airq.h>
 #include "cio.h"
 #include "css.h"
 #include "chsc.h"
diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h
index a6ef218defbe..d1483d65049c 100644
--- a/drivers/s390/cio/cio.h
+++ b/drivers/s390/cio/cio.h
@@ -99,6 +99,7 @@ extern int cio_get_options (struct subchannel *);
 extern int cio_modify (struct subchannel *);
 
 int cio_create_sch_lock(struct subchannel *);
+void do_adapter_IO(void);
 
 /* Use with care. */
 #ifdef CONFIG_CCW_CONSOLE
diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c
index 40a3208c7cf3..7a353db8564f 100644
--- a/drivers/s390/cio/qdio.c
+++ b/drivers/s390/cio/qdio.c
@@ -48,11 +48,11 @@
 #include <asm/debug.h>
 #include <asm/s390_rdev.h>
 #include <asm/qdio.h>
+#include <asm/airq.h>
 
 #include "cio.h"
 #include "css.h"
 #include "device.h"
-#include "airq.h"
 #include "qdio.h"
 #include "ioasm.h"
 #include "chsc.h"
@@ -96,7 +96,7 @@ static debug_info_t *qdio_dbf_slsb_in;
 static volatile struct qdio_q *tiq_list=NULL; /* volatile as it could change
 						 during a while loop */
 static DEFINE_SPINLOCK(ttiq_list_lock);
-static int register_thinint_result;
+static void *tiqdio_ind;
 static void tiqdio_tl(unsigned long);
 static DECLARE_TASKLET(tiqdio_tasklet,tiqdio_tl,0);
 
@@ -399,7 +399,7 @@ qdio_get_indicator(void)
 {
 	int i;
 
-	for (i=1;i<INDICATORS_PER_CACHELINE;i++)
+	for (i = 0; i < INDICATORS_PER_CACHELINE; i++)
 		if (!indicator_used[i]) {
 			indicator_used[i]=1;
 			return indicators+i;
@@ -1911,8 +1911,7 @@ qdio_fill_thresholds(struct qdio_irq *irq_ptr,
 	}
 }
 
-static int
-tiqdio_thinint_handler(void)
+static void tiqdio_thinint_handler(void *ind, void *drv_data)
 {
 	QDIO_DBF_TEXT4(0,trace,"thin_int");
 
@@ -1925,7 +1924,6 @@ tiqdio_thinint_handler(void)
 		tiqdio_clear_global_summary();
 
 	tiqdio_inbound_checks();
-	return 0;
 }
 
 static void
@@ -2445,7 +2443,7 @@ tiqdio_set_subchannel_ind(struct qdio_irq *irq_ptr, int reset_to_zero)
 		real_addr_dev_st_chg_ind=0;
 	} else {
 		real_addr_local_summary_bit=
-			virt_to_phys((volatile void *)indicators);
+			virt_to_phys((volatile void *)tiqdio_ind);
 		real_addr_dev_st_chg_ind=
 			virt_to_phys((volatile void *)irq_ptr->dev_st_chg_ind);
 	}
@@ -3740,23 +3738,25 @@ static void
 tiqdio_register_thinints(void)
 {
 	char dbf_text[20];
-	register_thinint_result=
-		s390_register_adapter_interrupt(&tiqdio_thinint_handler);
-	if (register_thinint_result) {
-		sprintf(dbf_text,"regthn%x",(register_thinint_result&0xff));
+
+	tiqdio_ind =
+		s390_register_adapter_interrupt(&tiqdio_thinint_handler, NULL);
+	if (IS_ERR(tiqdio_ind)) {
+		sprintf(dbf_text, "regthn%lx", PTR_ERR(tiqdio_ind));
 		QDIO_DBF_TEXT0(0,setup,dbf_text);
 		QDIO_PRINT_ERR("failed to register adapter handler " \
-			       "(rc=%i).\nAdapter interrupts might " \
+			       "(rc=%li).\nAdapter interrupts might " \
 			       "not work. Continuing.\n",
-			       register_thinint_result);
+			       PTR_ERR(tiqdio_ind));
+		tiqdio_ind = NULL;
 	}
 }
 
 static void
 tiqdio_unregister_thinints(void)
 {
-	if (!register_thinint_result)
-		s390_unregister_adapter_interrupt(&tiqdio_thinint_handler);
+	if (tiqdio_ind)
+		s390_unregister_adapter_interrupt(tiqdio_ind);
 }
 
 static int
@@ -3768,8 +3768,8 @@ qdio_get_qdio_memory(void)
 	for (i=1;i<INDICATORS_PER_CACHELINE;i++)
 		indicator_used[i]=0;
 	indicators = kzalloc(sizeof(__u32)*(INDICATORS_PER_CACHELINE),
-				   GFP_KERNEL);
-       	if (!indicators)
+			     GFP_KERNEL);
+	if (!indicators)
 		return -ENOMEM;
 	return 0;
 }
@@ -3780,7 +3780,6 @@ qdio_release_qdio_memory(void)
 	kfree(indicators);
 }
 
-
 static void
 qdio_unregister_dbf_views(void)
 {
diff --git a/include/asm-s390/airq.h b/include/asm-s390/airq.h
new file mode 100644
index 000000000000..41d028cb52a4
--- /dev/null
+++ b/include/asm-s390/airq.h
@@ -0,0 +1,19 @@
+/*
+ *  include/asm-s390/airq.h
+ *
+ *    Copyright IBM Corp. 2002,2007
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>
+ *		 Arnd Bergmann <arndb@de.ibm.com>
+ *		 Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#ifndef _ASM_S390_AIRQ_H
+#define _ASM_S390_AIRQ_H
+
+typedef void (*adapter_int_handler_t)(void *, void *);
+
+void *s390_register_adapter_interrupt(adapter_int_handler_t, void *);
+void s390_unregister_adapter_interrupt(void *);
+
+#endif /* _ASM_S390_AIRQ_H */
-- 
cgit v1.2.3


From 08d07968277cd898c88bf12b7720d89c02c4f139 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 26 Jan 2008 14:10:56 +0100
Subject: [S390] Standby cpu activation/deactivation.

Add a new interface so that cpus can be put into standby state and
configured state.
Only offline cpus can be put into standby state or configured state.
For that the new percpu sysfs attribute "configure" must be used.
To put a cpu in standby state a "0" must be written to the attribute.
In order to switch it into configured state a "1" must be written to
the attribute.
Only cpus in configured state can be brought online.
In addition this patch introduces a static mapping of physical to
logical cpus. As a result only the sysfs directories of present cpus
will be created. To scan for new cpus the new sysfs attribute "rescan"
must be used.
Writing to /sys/devices/system/cpu/rescan will trigger a rescan of
cpus and will create directories for new cpus.

On IPL only configured cpus will be used. And on reboot/shutdown all
cpus will remain in their current state (configured/standby).

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/early.c      |   3 +-
 arch/s390/kernel/setup.c      |   2 +-
 arch/s390/kernel/smp.c        | 401 ++++++++++++++++++++++++++++++++++--------
 drivers/s390/char/Makefile    |   2 +-
 drivers/s390/char/sclp.h      |   4 +-
 drivers/s390/char/sclp_cmd.c  | 319 +++++++++++++++++++++++++++++++++
 drivers/s390/char/sclp_info.c | 116 ------------
 include/asm-s390/sclp.h       |  21 ++-
 8 files changed, 675 insertions(+), 193 deletions(-)
 create mode 100644 drivers/s390/char/sclp_cmd.c
 delete mode 100644 drivers/s390/char/sclp_info.c

(limited to 'include')

diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 1b3af7dab816..c7cbb0114147 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -276,8 +276,9 @@ void __init startup_init(void)
 	create_kernel_nss();
 	sort_main_extable();
 	setup_lowcore_early();
-	sclp_readinfo_early();
+	sclp_read_info_early();
 	sclp_facilities_detect();
+	sclp_read_cpu_info_early();
 	memsize = sclp_memory_detect();
 #ifndef CONFIG_64BIT
 	/*
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 577aa7dd660e..d68a4025486e 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -920,7 +920,7 @@ setup_arch(char **cmdline_p)
 
 void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo)
 {
-   printk("cpu %d "
+   printk(KERN_INFO "cpu %d "
 #ifdef CONFIG_SMP
            "phys_idx=%d "
 #endif
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 264ea906db4c..66fe28930d82 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -42,6 +42,7 @@
 #include <asm/tlbflush.h>
 #include <asm/timer.h>
 #include <asm/lowcore.h>
+#include <asm/sclp.h>
 #include <asm/cpu.h>
 
 /*
@@ -58,6 +59,22 @@ EXPORT_SYMBOL(cpu_possible_map);
 
 static struct task_struct *current_set[NR_CPUS];
 
+static u8 smp_cpu_type;
+static int smp_use_sigp_detection;
+
+enum s390_cpu_state {
+	CPU_STATE_STANDBY,
+	CPU_STATE_CONFIGURED,
+};
+
+#ifdef CONFIG_HOTPLUG_CPU
+static DEFINE_MUTEX(smp_cpu_state_mutex);
+#endif
+static int smp_cpu_state[NR_CPUS];
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
+
 static void smp_ext_bitcall(int, ec_bit_sig);
 
 /*
@@ -355,6 +372,13 @@ void smp_ctl_clear_bit(int cr, int bit)
 }
 EXPORT_SYMBOL(smp_ctl_clear_bit);
 
+/*
+ * In early ipl state a temp. logically cpu number is needed, so the sigp
+ * functions can be used to sense other cpus. Since NR_CPUS is >= 2 on
+ * CONFIG_SMP and the ipl cpu is logical cpu 0, it must be 1.
+ */
+#define CPU_INIT_NO	1
+
 #if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE)
 
 /*
@@ -376,8 +400,9 @@ static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu)
 		return;
 	}
 	zfcpdump_save_areas[cpu] = alloc_bootmem(sizeof(union save_area));
-	__cpu_logical_map[1] = (__u16) phy_cpu;
-	while (signal_processor(1, sigp_stop_and_store_status) == sigp_busy)
+	__cpu_logical_map[CPU_INIT_NO] = (__u16) phy_cpu;
+	while (signal_processor(CPU_INIT_NO, sigp_stop_and_store_status) ==
+	       sigp_busy)
 		cpu_relax();
 	memcpy(zfcpdump_save_areas[cpu],
 	       (void *)(unsigned long) store_prefix() + SAVE_AREA_BASE,
@@ -397,32 +422,166 @@ static inline void smp_get_save_area(unsigned int cpu, unsigned int phy_cpu) { }
 
 #endif /* CONFIG_ZFCPDUMP || CONFIG_ZFCPDUMP_MODULE */
 
+static int cpu_stopped(int cpu)
+{
+	__u32 status;
+
+	/* Check for stopped state */
+	if (signal_processor_ps(&status, 0, cpu, sigp_sense) ==
+	    sigp_status_stored) {
+		if (status & 0x40)
+			return 1;
+	}
+	return 0;
+}
+
 /*
  * Lets check how many CPUs we have.
  */
-static unsigned int __init smp_count_cpus(void)
+static void __init smp_count_cpus(unsigned int *configured_cpus,
+				  unsigned int *standby_cpus)
 {
-	unsigned int cpu, num_cpus;
-	__u16 boot_cpu_addr;
+	unsigned int cpu;
+	struct sclp_cpu_info *info;
+	u16 boot_cpu_addr, cpu_addr;
 
-	/*
-	 * cpu 0 is the boot cpu. See smp_prepare_boot_cpu.
-	 */
 	boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr;
 	current_thread_info()->cpu = 0;
-	num_cpus = 1;
-	for (cpu = 0; cpu <= 65535; cpu++) {
-		if ((__u16) cpu == boot_cpu_addr)
+	*configured_cpus = 1;
+	*standby_cpus = 0;
+
+	info = alloc_bootmem_pages(sizeof(*info));
+	if (!info)
+		disabled_wait((unsigned long) __builtin_return_address(0));
+
+	/* Use sigp detection algorithm if sclp doesn't work. */
+	if (sclp_get_cpu_info(info)) {
+		smp_use_sigp_detection = 1;
+		for (cpu = 0; cpu <= 65535; cpu++) {
+			if (cpu == boot_cpu_addr)
+				continue;
+			__cpu_logical_map[CPU_INIT_NO] = cpu;
+			if (cpu_stopped(CPU_INIT_NO))
+				(*configured_cpus)++;
+		}
+		goto out;
+	}
+
+	if (info->has_cpu_type) {
+		for (cpu = 0; cpu < info->combined; cpu++) {
+			if (info->cpu[cpu].address == boot_cpu_addr) {
+				smp_cpu_type = info->cpu[cpu].type;
+				break;
+			}
+		}
+	}
+	/* Count cpus. */
+	for (cpu = 0; cpu < info->combined; cpu++) {
+		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
+			continue;
+		cpu_addr = info->cpu[cpu].address;
+		if (cpu_addr == boot_cpu_addr)
 			continue;
-		__cpu_logical_map[1] = (__u16) cpu;
-		if (signal_processor(1, sigp_sense) == sigp_not_operational)
+		__cpu_logical_map[CPU_INIT_NO] = cpu_addr;
+		if (!cpu_stopped(CPU_INIT_NO)) {
+			(*standby_cpus)++;
 			continue;
-		smp_get_save_area(num_cpus, cpu);
-		num_cpus++;
+		}
+		smp_get_save_area(*configured_cpus, cpu_addr);
+		(*configured_cpus)++;
 	}
-	printk("Detected %d CPU's\n", (int) num_cpus);
-	printk("Boot cpu address %2X\n", boot_cpu_addr);
-	return num_cpus;
+out:
+	printk(KERN_INFO "CPUs: %d configured, %d standby\n",
+	       *configured_cpus, *standby_cpus);
+	free_bootmem((unsigned long) info, sizeof(*info));
+}
+
+static int cpu_known(int cpu_id)
+{
+	int cpu;
+
+	for_each_present_cpu(cpu) {
+		if (__cpu_logical_map[cpu] == cpu_id)
+			return 1;
+	}
+	return 0;
+}
+
+static int smp_rescan_cpus_sigp(cpumask_t avail)
+{
+	int cpu_id, logical_cpu;
+
+	logical_cpu = first_cpu(avail);
+	if (logical_cpu == NR_CPUS)
+		return 0;
+	for (cpu_id = 0; cpu_id <= 65535; cpu_id++) {
+		if (cpu_known(cpu_id))
+			continue;
+		__cpu_logical_map[logical_cpu] = cpu_id;
+		if (!cpu_stopped(logical_cpu))
+			continue;
+		cpu_set(logical_cpu, cpu_present_map);
+		smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
+		logical_cpu = next_cpu(logical_cpu, avail);
+		if (logical_cpu == NR_CPUS)
+			break;
+	}
+	return 0;
+}
+
+static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail)
+{
+	struct sclp_cpu_info *info;
+	int cpu_id, logical_cpu, cpu;
+	int rc;
+
+	logical_cpu = first_cpu(avail);
+	if (logical_cpu == NR_CPUS)
+		return 0;
+	if (slab_is_available())
+		info = kmalloc(sizeof(*info), GFP_KERNEL);
+	else
+		info = alloc_bootmem(sizeof(*info));
+	if (!info)
+		return -ENOMEM;
+	rc = sclp_get_cpu_info(info);
+	if (rc)
+		goto out;
+	for (cpu = 0; cpu < info->combined; cpu++) {
+		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
+			continue;
+		cpu_id = info->cpu[cpu].address;
+		if (cpu_known(cpu_id))
+			continue;
+		__cpu_logical_map[logical_cpu] = cpu_id;
+		cpu_set(logical_cpu, cpu_present_map);
+		if (cpu >= info->configured)
+			smp_cpu_state[logical_cpu] = CPU_STATE_STANDBY;
+		else
+			smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
+		logical_cpu = next_cpu(logical_cpu, avail);
+		if (logical_cpu == NR_CPUS)
+			break;
+	}
+out:
+	if (slab_is_available())
+		kfree(info);
+	else
+		free_bootmem((unsigned long) info, sizeof(*info));
+	return rc;
+}
+
+static int smp_rescan_cpus(void)
+{
+	cpumask_t avail;
+
+	cpus_setall(avail);
+	cpus_and(avail, avail, cpu_possible_map);
+	cpus_andnot(avail, avail, cpu_present_map);
+	if (smp_use_sigp_detection)
+		return smp_rescan_cpus_sigp(avail);
+	else
+		return smp_rescan_cpus_sclp(avail);
 }
 
 /*
@@ -453,8 +612,6 @@ int __cpuinit start_secondary(void *cpuvoid)
 	return 0;
 }
 
-DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
-
 static void __init smp_create_idle(unsigned int cpu)
 {
 	struct task_struct *p;
@@ -470,37 +627,16 @@ static void __init smp_create_idle(unsigned int cpu)
 	spin_lock_init(&(&per_cpu(s390_idle, cpu))->lock);
 }
 
-static int cpu_stopped(int cpu)
-{
-	__u32 status;
-
-	/* Check for stopped state */
-	if (signal_processor_ps(&status, 0, cpu, sigp_sense) ==
-	    sigp_status_stored) {
-		if (status & 0x40)
-			return 1;
-	}
-	return 0;
-}
-
 /* Upping and downing of CPUs */
-
 int __cpu_up(unsigned int cpu)
 {
 	struct task_struct *idle;
 	struct _lowcore *cpu_lowcore;
 	struct stack_frame *sf;
 	sigp_ccode ccode;
-	int curr_cpu;
-
-	for (curr_cpu = 0; curr_cpu <= 65535; curr_cpu++) {
-		__cpu_logical_map[cpu] = (__u16) curr_cpu;
-		if (cpu_stopped(cpu))
-			break;
-	}
 
-	if (!cpu_stopped(cpu))
-		return -ENODEV;
+	if (smp_cpu_state[cpu] != CPU_STATE_CONFIGURED)
+		return -EIO;
 
 	ccode = signal_processor_p((__u32)(unsigned long)(lowcore_ptr[cpu]),
 				   cpu, sigp_set_prefix);
@@ -543,21 +679,18 @@ static unsigned int __initdata possible_cpus;
 
 void __init smp_setup_cpu_possible_map(void)
 {
-	unsigned int phy_cpus, pos_cpus, cpu;
-
-	phy_cpus = smp_count_cpus();
-	pos_cpus = min(phy_cpus + additional_cpus, (unsigned int) NR_CPUS);
+	unsigned int pos_cpus, cpu;
+	unsigned int configured_cpus, standby_cpus;
 
+	smp_count_cpus(&configured_cpus, &standby_cpus);
+	pos_cpus = min(configured_cpus + standby_cpus + additional_cpus,
+		       (unsigned int) NR_CPUS);
 	if (possible_cpus)
 		pos_cpus = min(possible_cpus, (unsigned int) NR_CPUS);
-
 	for (cpu = 0; cpu < pos_cpus; cpu++)
 		cpu_set(cpu, cpu_possible_map);
-
-	phy_cpus = min(phy_cpus, pos_cpus);
-
-	for (cpu = 0; cpu < phy_cpus; cpu++)
-		cpu_set(cpu, cpu_present_map);
+	cpu_present_map = cpumask_of_cpu(0);
+	smp_rescan_cpus();
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -612,7 +745,7 @@ void __cpu_die(unsigned int cpu)
 	/* Wait until target cpu is down */
 	while (!smp_cpu_not_running(cpu))
 		cpu_relax();
-	printk("Processor %d spun down\n", cpu);
+	printk(KERN_INFO "Processor %d spun down\n", cpu);
 }
 
 void cpu_die(void)
@@ -686,12 +819,12 @@ void __init smp_prepare_boot_cpu(void)
 	cpu_set(0, cpu_online_map);
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
 	current_set[0] = current;
+	smp_cpu_state[0] = CPU_STATE_CONFIGURED;
 	spin_lock_init(&(&__get_cpu_var(s390_idle))->lock);
 }
 
 void __init smp_cpus_done(unsigned int max_cpus)
 {
-	cpu_present_map = cpu_possible_map;
 }
 
 /*
@@ -705,7 +838,79 @@ int setup_profiling_timer(unsigned int multiplier)
 	return 0;
 }
 
-static DEFINE_PER_CPU(struct cpu, cpu_devices);
+#ifdef CONFIG_HOTPLUG_CPU
+static ssize_t cpu_configure_show(struct sys_device *dev, char *buf)
+{
+	ssize_t count;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	count = sprintf(buf, "%d\n", smp_cpu_state[dev->id]);
+	mutex_unlock(&smp_cpu_state_mutex);
+	return count;
+}
+
+static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf,
+				   size_t count)
+{
+	int cpu = dev->id;
+	int val, rc;
+	char delim;
+
+	if (sscanf(buf, "%d %c", &val, &delim) != 1)
+		return -EINVAL;
+	if (val != 0 && val != 1)
+		return -EINVAL;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	lock_cpu_hotplug();
+	rc = -EBUSY;
+	if (cpu_online(cpu))
+		goto out;
+	rc = 0;
+	switch (val) {
+	case 0:
+		if (smp_cpu_state[cpu] == CPU_STATE_CONFIGURED) {
+			rc = sclp_cpu_deconfigure(__cpu_logical_map[cpu]);
+			if (!rc)
+				smp_cpu_state[cpu] = CPU_STATE_STANDBY;
+		}
+		break;
+	case 1:
+		if (smp_cpu_state[cpu] == CPU_STATE_STANDBY) {
+			rc = sclp_cpu_configure(__cpu_logical_map[cpu]);
+			if (!rc)
+				smp_cpu_state[cpu] = CPU_STATE_CONFIGURED;
+		}
+		break;
+	default:
+		break;
+	}
+out:
+	unlock_cpu_hotplug();
+	mutex_unlock(&smp_cpu_state_mutex);
+	return rc ? rc : count;
+}
+static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store);
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static ssize_t show_cpu_address(struct sys_device *dev, char *buf)
+{
+	return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]);
+}
+static SYSDEV_ATTR(address, 0444, show_cpu_address, NULL);
+
+
+static struct attribute *cpu_common_attrs[] = {
+#ifdef CONFIG_HOTPLUG_CPU
+	&attr_configure.attr,
+#endif
+	&attr_address.attr,
+	NULL,
+};
+
+static struct attribute_group cpu_common_attr_group = {
+	.attrs = cpu_common_attrs,
+};
 
 static ssize_t show_capability(struct sys_device *dev, char *buf)
 {
@@ -750,15 +955,15 @@ static ssize_t show_idle_time(struct sys_device *dev, char *buf)
 }
 static SYSDEV_ATTR(idle_time_us, 0444, show_idle_time, NULL);
 
-static struct attribute *cpu_attrs[] = {
+static struct attribute *cpu_online_attrs[] = {
 	&attr_capability.attr,
 	&attr_idle_count.attr,
 	&attr_idle_time_us.attr,
 	NULL,
 };
 
-static struct attribute_group cpu_attr_group = {
-	.attrs = cpu_attrs,
+static struct attribute_group cpu_online_attr_group = {
+	.attrs = cpu_online_attrs,
 };
 
 static int __cpuinit smp_cpu_notify(struct notifier_block *self,
@@ -778,12 +983,12 @@ static int __cpuinit smp_cpu_notify(struct notifier_block *self,
 		idle->idle_time = 0;
 		idle->idle_count = 0;
 		spin_unlock_irq(&idle->lock);
-		if (sysfs_create_group(&s->kobj, &cpu_attr_group))
+		if (sysfs_create_group(&s->kobj, &cpu_online_attr_group))
 			return NOTIFY_BAD;
 		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
-		sysfs_remove_group(&s->kobj, &cpu_attr_group);
+		sysfs_remove_group(&s->kobj, &cpu_online_attr_group);
 		break;
 	}
 	return NOTIFY_OK;
@@ -793,6 +998,62 @@ static struct notifier_block __cpuinitdata smp_cpu_nb = {
 	.notifier_call = smp_cpu_notify,
 };
 
+static int smp_add_present_cpu(int cpu)
+{
+	struct cpu *c = &per_cpu(cpu_devices, cpu);
+	struct sys_device *s = &c->sysdev;
+	int rc;
+
+	c->hotpluggable = 1;
+	rc = register_cpu(c, cpu);
+	if (rc)
+		goto out;
+	rc = sysfs_create_group(&s->kobj, &cpu_common_attr_group);
+	if (rc)
+		goto out_cpu;
+	if (!cpu_online(cpu))
+		goto out;
+	rc = sysfs_create_group(&s->kobj, &cpu_online_attr_group);
+	if (!rc)
+		return 0;
+	sysfs_remove_group(&s->kobj, &cpu_common_attr_group);
+out_cpu:
+#ifdef CONFIG_HOTPLUG_CPU
+	unregister_cpu(c);
+#endif
+out:
+	return rc;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static ssize_t rescan_store(struct sys_device *dev, const char *buf,
+			    size_t count)
+{
+	cpumask_t newcpus;
+	int cpu;
+	int rc;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	lock_cpu_hotplug();
+	newcpus = cpu_present_map;
+	rc = smp_rescan_cpus();
+	if (rc)
+		goto out;
+	cpus_andnot(newcpus, cpu_present_map, newcpus);
+	for_each_cpu_mask(cpu, newcpus) {
+		rc = smp_add_present_cpu(cpu);
+		if (rc)
+			cpu_clear(cpu, cpu_present_map);
+	}
+	rc = 0;
+out:
+	unlock_cpu_hotplug();
+	mutex_unlock(&smp_cpu_state_mutex);
+	return rc ? rc : count;
+}
+static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store);
+#endif /* CONFIG_HOTPLUG_CPU */
+
 static int __init topology_init(void)
 {
 	int cpu;
@@ -800,16 +1061,14 @@ static int __init topology_init(void)
 
 	register_cpu_notifier(&smp_cpu_nb);
 
-	for_each_possible_cpu(cpu) {
-		struct cpu *c = &per_cpu(cpu_devices, cpu);
-		struct sys_device *s = &c->sysdev;
-
-		c->hotpluggable = 1;
-		register_cpu(c, cpu);
-		if (!cpu_online(cpu))
-			continue;
-		s = &c->sysdev;
-		rc = sysfs_create_group(&s->kobj, &cpu_attr_group);
+#ifdef CONFIG_HOTPLUG_CPU
+	rc = sysfs_create_file(&cpu_sysdev_class.kset.kobj,
+			       &attr_rescan.attr);
+	if (rc)
+		return rc;
+#endif
+	for_each_present_cpu(cpu) {
+		rc = smp_add_present_cpu(cpu);
 		if (rc)
 			return rc;
 	}
diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile
index bee3a3af691d..9317333ec149 100644
--- a/drivers/s390/char/Makefile
+++ b/drivers/s390/char/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-y += ctrlchar.o keyboard.o defkeymap.o sclp.o sclp_rw.o sclp_quiesce.o \
-	 sclp_info.o sclp_config.o sclp_chp.o sclp_cpi_sys.o
+	 sclp_cmd.o sclp_config.o sclp_chp.o sclp_cpi_sys.o
 
 obj-$(CONFIG_TN3270) += raw3270.o
 obj-$(CONFIG_TN3270_CONSOLE) += con3270.o
diff --git a/drivers/s390/char/sclp.h b/drivers/s390/char/sclp.h
index c7318a125852..aa8186d18aee 100644
--- a/drivers/s390/char/sclp.h
+++ b/drivers/s390/char/sclp.h
@@ -56,8 +56,6 @@ typedef unsigned int sclp_cmdw_t;
 #define SCLP_CMDW_READ_EVENT_DATA	0x00770005
 #define SCLP_CMDW_WRITE_EVENT_DATA	0x00760005
 #define SCLP_CMDW_WRITE_EVENT_MASK	0x00780005
-#define SCLP_CMDW_READ_SCP_INFO		0x00020001
-#define SCLP_CMDW_READ_SCP_INFO_FORCED	0x00120001
 
 #define GDS_ID_MDSMU		0x1310
 #define GDS_ID_MDSROUTEINFO	0x1311
@@ -83,6 +81,8 @@ extern u64 sclp_facilities;
 
 #define SCLP_HAS_CHP_INFO	(sclp_facilities & 0x8000000000000000ULL)
 #define SCLP_HAS_CHP_RECONFIG	(sclp_facilities & 0x2000000000000000ULL)
+#define SCLP_HAS_CPU_INFO	(sclp_facilities & 0x0800000000000000ULL)
+#define SCLP_HAS_CPU_RECONFIG	(sclp_facilities & 0x0400000000000000ULL)
 
 struct gds_subvector {
 	u8	length;
diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c
new file mode 100644
index 000000000000..ba004fd43c05
--- /dev/null
+++ b/drivers/s390/char/sclp_cmd.c
@@ -0,0 +1,319 @@
+/*
+ *  drivers/s390/char/sclp_cmd.c
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
+ */
+
+#include <linux/completion.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <asm/sclp.h>
+#include "sclp.h"
+
+#define TAG	"sclp_cmd: "
+
+#define SCLP_CMDW_READ_SCP_INFO		0x00020001
+#define SCLP_CMDW_READ_SCP_INFO_FORCED	0x00120001
+
+struct read_info_sccb {
+	struct	sccb_header header;	/* 0-7 */
+	u16	rnmax;			/* 8-9 */
+	u8	rnsize;			/* 10 */
+	u8	_reserved0[24 - 11];	/* 11-15 */
+	u8	loadparm[8];		/* 24-31 */
+	u8	_reserved1[48 - 32];	/* 32-47 */
+	u64	facilities;		/* 48-55 */
+	u8	_reserved2[84 - 56];	/* 56-83 */
+	u8	fac84;			/* 84 */
+	u8	_reserved3[91 - 85];	/* 85-90 */
+	u8	flags;			/* 91 */
+	u8	_reserved4[100 - 92];	/* 92-99 */
+	u32	rnsize2;		/* 100-103 */
+	u64	rnmax2;			/* 104-111 */
+	u8	_reserved5[4096 - 112];	/* 112-4095 */
+} __attribute__((packed, aligned(PAGE_SIZE)));
+
+static struct read_info_sccb __initdata early_read_info_sccb;
+static int __initdata early_read_info_sccb_valid;
+
+u64 sclp_facilities;
+static u8 sclp_fac84;
+
+static int __init sclp_cmd_sync_early(sclp_cmdw_t cmd, void *sccb)
+{
+	int rc;
+
+	__ctl_set_bit(0, 9);
+	rc = sclp_service_call(cmd, sccb);
+	if (rc)
+		goto out;
+	__load_psw_mask(PSW_BASE_BITS | PSW_MASK_EXT |
+			PSW_MASK_WAIT | PSW_DEFAULT_KEY);
+	local_irq_disable();
+out:
+	/* Contents of the sccb might have changed. */
+	barrier();
+	__ctl_clear_bit(0, 9);
+	return rc;
+}
+
+void __init sclp_read_info_early(void)
+{
+	int rc;
+	int i;
+	struct read_info_sccb *sccb;
+	sclp_cmdw_t commands[] = {SCLP_CMDW_READ_SCP_INFO_FORCED,
+				  SCLP_CMDW_READ_SCP_INFO};
+
+	sccb = &early_read_info_sccb;
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		do {
+			memset(sccb, 0, sizeof(*sccb));
+			sccb->header.length = sizeof(*sccb);
+			sccb->header.control_mask[2] = 0x80;
+			rc = sclp_cmd_sync_early(commands[i], sccb);
+		} while (rc == -EBUSY);
+
+		if (rc)
+			break;
+		if (sccb->header.response_code == 0x10) {
+			early_read_info_sccb_valid = 1;
+			break;
+		}
+		if (sccb->header.response_code != 0x1f0)
+			break;
+	}
+}
+
+void __init sclp_facilities_detect(void)
+{
+	if (!early_read_info_sccb_valid)
+		return;
+	sclp_facilities = early_read_info_sccb.facilities;
+	sclp_fac84 = early_read_info_sccb.fac84;
+}
+
+unsigned long long __init sclp_memory_detect(void)
+{
+	unsigned long long memsize;
+	struct read_info_sccb *sccb;
+
+	if (!early_read_info_sccb_valid)
+		return 0;
+	sccb = &early_read_info_sccb;
+	if (sccb->rnsize)
+		memsize = sccb->rnsize << 20;
+	else
+		memsize = sccb->rnsize2 << 20;
+	if (sccb->rnmax)
+		memsize *= sccb->rnmax;
+	else
+		memsize *= sccb->rnmax2;
+	return memsize;
+}
+
+/*
+ * This function will be called after sclp_memory_detect(), which gets called
+ * early from early.c code. Therefore the sccb should have valid contents.
+ */
+void __init sclp_get_ipl_info(struct sclp_ipl_info *info)
+{
+	struct read_info_sccb *sccb;
+
+	if (!early_read_info_sccb_valid)
+		return;
+	sccb = &early_read_info_sccb;
+	info->is_valid = 1;
+	if (sccb->flags & 0x2)
+		info->has_dump = 1;
+	memcpy(&info->loadparm, &sccb->loadparm, LOADPARM_LEN);
+}
+
+static void sclp_sync_callback(struct sclp_req *req, void *data)
+{
+	struct completion *completion = data;
+
+	complete(completion);
+}
+
+static int do_sync_request(sclp_cmdw_t cmd, void *sccb)
+{
+	struct completion completion;
+	struct sclp_req *request;
+	int rc;
+
+	request = kzalloc(sizeof(*request), GFP_KERNEL);
+	if (!request)
+		return -ENOMEM;
+	request->command = cmd;
+	request->sccb = sccb;
+	request->status = SCLP_REQ_FILLED;
+	request->callback = sclp_sync_callback;
+	request->callback_data = &completion;
+	init_completion(&completion);
+
+	/* Perform sclp request. */
+	rc = sclp_add_request(request);
+	if (rc)
+		goto out;
+	wait_for_completion(&completion);
+
+	/* Check response. */
+	if (request->status != SCLP_REQ_DONE) {
+		printk(KERN_WARNING TAG "sync request failed "
+		       "(cmd=0x%08x, status=0x%02x)\n", cmd, request->status);
+		rc = -EIO;
+	}
+out:
+	kfree(request);
+	return rc;
+}
+
+/*
+ * CPU configuration related functions.
+ */
+
+#define SCLP_CMDW_READ_CPU_INFO		0x00010001
+#define SCLP_CMDW_CONFIGURE_CPU		0x00110001
+#define SCLP_CMDW_DECONFIGURE_CPU	0x00100001
+
+struct read_cpu_info_sccb {
+	struct	sccb_header header;
+	u16	nr_configured;
+	u16	offset_configured;
+	u16	nr_standby;
+	u16	offset_standby;
+	u8	reserved[4096 - 16];
+} __attribute__((packed, aligned(PAGE_SIZE)));
+
+static struct read_cpu_info_sccb __initdata early_read_cpu_info_sccb;
+static struct sclp_cpu_info __initdata sclp_cpu_info;
+
+static void sclp_fill_cpu_info(struct sclp_cpu_info *info,
+			       struct read_cpu_info_sccb *sccb)
+{
+	char *page = (char *) sccb;
+
+	memset(info, 0, sizeof(*info));
+	info->configured = sccb->nr_configured;
+	info->standby = sccb->nr_standby;
+	info->combined = sccb->nr_configured + sccb->nr_standby;
+	info->has_cpu_type = sclp_fac84 & 0x1;
+	memcpy(&info->cpu, page + sccb->offset_configured,
+	       info->combined * sizeof(struct sclp_cpu_entry));
+}
+
+void __init sclp_read_cpu_info_early(void)
+{
+	int rc;
+	struct read_cpu_info_sccb *sccb;
+
+	if (!SCLP_HAS_CPU_INFO)
+		return;
+
+	sccb = &early_read_cpu_info_sccb;
+	do {
+		memset(sccb, 0, sizeof(*sccb));
+		sccb->header.length = sizeof(*sccb);
+		rc = sclp_cmd_sync_early(SCLP_CMDW_READ_CPU_INFO, sccb);
+	} while (rc == -EBUSY);
+
+	if (rc)
+		return;
+	if (sccb->header.response_code != 0x10)
+		return;
+	sclp_fill_cpu_info(&sclp_cpu_info, sccb);
+}
+
+static int __init sclp_get_cpu_info_early(struct sclp_cpu_info *info)
+{
+	if (!SCLP_HAS_CPU_INFO)
+		return -EOPNOTSUPP;
+	*info = sclp_cpu_info;
+	return 0;
+}
+
+static int sclp_get_cpu_info_late(struct sclp_cpu_info *info)
+{
+	int rc;
+	struct read_cpu_info_sccb *sccb;
+
+	if (!SCLP_HAS_CPU_INFO)
+		return -EOPNOTSUPP;
+	sccb = (struct read_cpu_info_sccb *)  __get_free_page(GFP_KERNEL
+							      | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	memset(sccb, 0, sizeof(*sccb));
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(SCLP_CMDW_READ_CPU_INFO, sccb);
+	if (rc)
+		goto out;
+	if (sccb->header.response_code != 0x0010) {
+		printk(KERN_WARNING TAG "readcpuinfo failed "
+		       "(response=0x%04x)\n", sccb->header.response_code);
+		rc = -EIO;
+		goto out;
+	}
+	sclp_fill_cpu_info(info, sccb);
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+
+int __init_refok sclp_get_cpu_info(struct sclp_cpu_info *info)
+{
+	if (slab_is_available())
+		return sclp_get_cpu_info_late(info);
+	return sclp_get_cpu_info_early(info);
+}
+
+struct cpu_configure_sccb {
+	struct sccb_header header;
+} __attribute__((packed, aligned(8)));
+
+static int do_cpu_configure(sclp_cmdw_t cmd)
+{
+	struct cpu_configure_sccb *sccb;
+	int rc;
+
+	if (!SCLP_HAS_CPU_RECONFIG)
+		return -EOPNOTSUPP;
+	/*
+	 * This is not going to cross a page boundary since we force
+	 * kmalloc to have a minimum alignment of 8 bytes on s390.
+	 */
+	sccb = kzalloc(sizeof(*sccb), GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(cmd, sccb);
+	if (rc)
+		goto out;
+	switch (sccb->header.response_code) {
+	case 0x0020:
+	case 0x0120:
+		break;
+	default:
+		printk(KERN_WARNING TAG "configure cpu failed (cmd=0x%08x, "
+		       "response=0x%04x)\n", cmd, sccb->header.response_code);
+		rc = -EIO;
+		break;
+	}
+out:
+	kfree(sccb);
+	return rc;
+}
+
+int sclp_cpu_configure(u8 cpu)
+{
+	return do_cpu_configure(SCLP_CMDW_CONFIGURE_CPU | cpu << 8);
+}
+
+int sclp_cpu_deconfigure(u8 cpu)
+{
+	return do_cpu_configure(SCLP_CMDW_DECONFIGURE_CPU | cpu << 8);
+}
diff --git a/drivers/s390/char/sclp_info.c b/drivers/s390/char/sclp_info.c
deleted file mode 100644
index a1136e052750..000000000000
--- a/drivers/s390/char/sclp_info.c
+++ /dev/null
@@ -1,116 +0,0 @@
-/*
- *  drivers/s390/char/sclp_info.c
- *
- *    Copyright IBM Corp. 2007
- *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
- */
-
-#include <linux/init.h>
-#include <linux/errno.h>
-#include <linux/string.h>
-#include <asm/sclp.h>
-#include "sclp.h"
-
-struct sclp_readinfo_sccb {
-	struct	sccb_header header;	/* 0-7 */
-	u16	rnmax;			/* 8-9 */
-	u8	rnsize;			/* 10 */
-	u8	_reserved0[24 - 11];	/* 11-23 */
-	u8	loadparm[8];		/* 24-31 */
-	u8	_reserved1[48 - 32];	/* 32-47 */
-	u64	facilities;		/* 48-55 */
-	u8	_reserved2[91 - 56];	/* 56-90 */
-	u8	flags;			/* 91 */
-	u8	_reserved3[100 - 92];	/* 92-99 */
-	u32	rnsize2;		/* 100-103 */
-	u64	rnmax2;			/* 104-111 */
-	u8	_reserved4[4096 - 112];	/* 112-4095 */
-} __attribute__((packed, aligned(4096)));
-
-static struct sclp_readinfo_sccb __initdata early_readinfo_sccb;
-static int __initdata early_readinfo_sccb_valid;
-
-u64 sclp_facilities;
-
-void __init sclp_readinfo_early(void)
-{
-	int ret;
-	int i;
-	struct sclp_readinfo_sccb *sccb;
-	sclp_cmdw_t commands[] = {SCLP_CMDW_READ_SCP_INFO_FORCED,
-				  SCLP_CMDW_READ_SCP_INFO};
-
-	/* Enable service signal subclass mask. */
-	__ctl_set_bit(0, 9);
-	sccb = &early_readinfo_sccb;
-	for (i = 0; i < ARRAY_SIZE(commands); i++) {
-		do {
-			memset(sccb, 0, sizeof(*sccb));
-			sccb->header.length = sizeof(*sccb);
-			sccb->header.control_mask[2] = 0x80;
-			ret = sclp_service_call(commands[i], sccb);
-		} while (ret == -EBUSY);
-
-		if (ret)
-			break;
-		__load_psw_mask(PSW_BASE_BITS | PSW_MASK_EXT |
-				PSW_MASK_WAIT | PSW_DEFAULT_KEY);
-		local_irq_disable();
-		/*
-		 * Contents of the sccb might have changed
-		 * therefore a barrier is needed.
-		 */
-		barrier();
-		if (sccb->header.response_code == 0x10) {
-			early_readinfo_sccb_valid = 1;
-			break;
-		}
-		if (sccb->header.response_code != 0x1f0)
-			break;
-	}
-	/* Disable service signal subclass mask again. */
-	__ctl_clear_bit(0, 9);
-}
-
-void __init sclp_facilities_detect(void)
-{
-	if (!early_readinfo_sccb_valid)
-		return;
-	sclp_facilities = early_readinfo_sccb.facilities;
-}
-
-unsigned long long __init sclp_memory_detect(void)
-{
-	unsigned long long memsize;
-	struct sclp_readinfo_sccb *sccb;
-
-	if (!early_readinfo_sccb_valid)
-		return 0;
-	sccb = &early_readinfo_sccb;
-	if (sccb->rnsize)
-		memsize = sccb->rnsize << 20;
-	else
-		memsize = sccb->rnsize2 << 20;
-	if (sccb->rnmax)
-		memsize *= sccb->rnmax;
-	else
-		memsize *= sccb->rnmax2;
-	return memsize;
-}
-
-/*
- * This function will be called after sclp_memory_detect(), which gets called
- * early from early.c code. Therefore the sccb should have valid contents.
- */
-void __init sclp_get_ipl_info(struct sclp_ipl_info *info)
-{
-	struct sclp_readinfo_sccb *sccb;
-
-	if (!early_readinfo_sccb_valid)
-		return;
-	sccb = &early_readinfo_sccb;
-	info->is_valid = 1;
-	if (sccb->flags & 0x2)
-		info->has_dump = 1;
-	memcpy(&info->loadparm, &sccb->loadparm, LOADPARM_LEN);
-}
diff --git a/include/asm-s390/sclp.h b/include/asm-s390/sclp.h
index cb9faf1ea5cf..b8c7695cd4c8 100644
--- a/include/asm-s390/sclp.h
+++ b/include/asm-s390/sclp.h
@@ -27,7 +27,26 @@ struct sclp_ipl_info {
 	char loadparm[LOADPARM_LEN];
 };
 
-void sclp_readinfo_early(void);
+struct sclp_cpu_entry {
+	u8 address;
+	u8 reserved0[13];
+	u8 type;
+	u8 reserved1;
+} __attribute__((packed));
+
+struct sclp_cpu_info {
+	unsigned int configured;
+	unsigned int standby;
+	unsigned int combined;
+	int has_cpu_type;
+	struct sclp_cpu_entry cpu[255];
+};
+
+int sclp_get_cpu_info(struct sclp_cpu_info *info);
+int sclp_cpu_configure(u8 cpu);
+int sclp_cpu_deconfigure(u8 cpu);
+void sclp_read_info_early(void);
+void sclp_read_cpu_info_early(void);
 void sclp_facilities_detect(void);
 unsigned long long sclp_memory_detect(void);
 int sclp_sdias_blk_count(void);
-- 
cgit v1.2.3


From 6f457e1a149eb39ee58d51913e8023fc27c52806 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 26 Jan 2008 14:10:58 +0100
Subject: [S390] Fix tlb flushing with idte.

The clear-by-asce operation of the idte instruction gets an asce
(address-space-control-element) as argument to specify which TLBs
need to get flushed. The current code passes a plain pointer to
the start of the pgd without the additional bits which would make
the pointer an asce. The current machines don't mind the difference
but a future model might want to use the designation type control
bits in the asce as a filter for the TLBs to flush.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/head64.S      |  2 +-
 include/asm-s390/mmu_context.h | 27 ++++++++++++---------------
 include/asm-s390/tlbflush.h    | 12 ++++++------
 3 files changed, 19 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index a87b1976d409..79dccd206a6e 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -157,7 +157,7 @@ startup_continue:
 	.long	0xb2b10000		# store facility list
 	tm	0xc8,0x08		# check bit for clearing-by-ASCE
 	bno	0f-.LPG1(%r13)
-	lhi	%r1,2094
+	lhi	%r1,2048
 	lhi	%r2,0
 	.long	0xb98e2001
 	oi	7(%r12),0x80		# set IDTE flag
diff --git a/include/asm-s390/mmu_context.h b/include/asm-s390/mmu_context.h
index 05b842126b99..a77d4ba3c8eb 100644
--- a/include/asm-s390/mmu_context.h
+++ b/include/asm-s390/mmu_context.h
@@ -12,10 +12,15 @@
 #include <asm/pgalloc.h>
 #include <asm-generic/mm_hooks.h>
 
-/*
- * get a new mmu context.. S390 don't know about contexts.
- */
-#define init_new_context(tsk,mm)        0
+static inline int init_new_context(struct task_struct *tsk,
+				   struct mm_struct *mm)
+{
+	mm->context = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS;
+#ifdef CONFIG_64BIT
+	mm->context |= _ASCE_TYPE_REGION3;
+#endif
+	return 0;
+}
 
 #define destroy_context(mm)             do { } while (0)
 
@@ -27,19 +32,11 @@
 
 static inline void update_mm(struct mm_struct *mm, struct task_struct *tsk)
 {
-	pgd_t *pgd = mm->pgd;
-	unsigned long asce_bits;
-
-	/* Calculate asce bits from the first pgd table entry. */
-	asce_bits = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS;
-#ifdef CONFIG_64BIT
-	asce_bits |= _ASCE_TYPE_REGION3;
-#endif
-	S390_lowcore.user_asce = asce_bits | __pa(pgd);
+	S390_lowcore.user_asce = mm->context | __pa(mm->pgd);
 	if (switch_amode) {
 		/* Load primary space page table origin. */
-		pgd_t *shadow_pgd = get_shadow_table(pgd) ? : pgd;
-		S390_lowcore.user_exec_asce = asce_bits | __pa(shadow_pgd);
+		pgd_t *shadow_pgd = get_shadow_table(mm->pgd) ? : mm->pgd;
+		S390_lowcore.user_exec_asce = mm->context | __pa(shadow_pgd);
 		asm volatile(LCTL_OPCODE" 1,1,%0\n"
 			     : : "m" (S390_lowcore.user_exec_asce) );
 	} else
diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index a69bd2490d52..a5e876539ec6 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -42,11 +42,11 @@ static inline void __tlb_flush_global(void)
 /*
  * Flush all tlb entries of a page table on all cpus.
  */
-static inline void __tlb_flush_idte(pgd_t *pgd)
+static inline void __tlb_flush_idte(unsigned long asce)
 {
 	asm volatile(
 		"	.insn	rrf,0xb98e0000,0,%0,%1,0"
-		: : "a" (2048), "a" (__pa(pgd) & PAGE_MASK) : "cc" );
+		: : "a" (2048), "a" (asce) : "cc" );
 }
 
 static inline void __tlb_flush_mm(struct mm_struct * mm)
@@ -61,11 +61,11 @@ static inline void __tlb_flush_mm(struct mm_struct * mm)
 	 * only ran on the local cpu.
 	 */
 	if (MACHINE_HAS_IDTE) {
-		pgd_t *shadow_pgd = get_shadow_table(mm->pgd);
+		pgd_t *shadow = get_shadow_table(mm->pgd);
 
-		if (shadow_pgd)
-			__tlb_flush_idte(shadow_pgd);
-		__tlb_flush_idte(mm->pgd);
+		if (shadow)
+			__tlb_flush_idte((unsigned long) shadow | mm->context);
+		__tlb_flush_idte((unsigned long) mm->pgd | mm->context);
 		return;
 	}
 	preempt_disable();
-- 
cgit v1.2.3


From 8ffd74a0924e4e04f6455eb2d2187a9564678d01 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 26 Jan 2008 14:10:59 +0100
Subject: [S390] Avoid warnings in tlblush.h

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/tlbflush.h | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index a5e876539ec6..70fa5ae58180 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -106,9 +106,23 @@ static inline void __tlb_flush_mm_cond(struct mm_struct * mm)
  */
 #define flush_tlb()				do { } while (0)
 #define flush_tlb_all()				do { } while (0)
-#define flush_tlb_mm(mm)			__tlb_flush_mm_cond(mm)
 #define flush_tlb_page(vma, addr)		do { } while (0)
-#define flush_tlb_range(vma, start, end)	__tlb_flush_mm_cond(mm)
-#define flush_tlb_kernel_range(start, end)	__tlb_flush_mm(&init_mm)
+
+static inline void flush_tlb_mm(struct mm_struct *mm)
+{
+	__tlb_flush_mm_cond(mm);
+}
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
+{
+	__tlb_flush_mm_cond(vma->vm_mm);
+}
+
+static inline void flush_tlb_kernel_range(unsigned long start,
+					  unsigned long end)
+{
+	__tlb_flush_mm(&init_mm);
+}
 
 #endif /* _S390_TLBFLUSH_H */
-- 
cgit v1.2.3


From 5fd9c6e214547a32d3da6ee4284c79004d667bc8 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:00 +0100
Subject: [S390] Change vmalloc defintions

Currently the vmalloc area starts at a dynamic address depending on
the memory size. There was also an 8MB security hole after the
physical memory to catch out-of-bounds accesses.
We can simplify the code by putting the vmalloc area explicitely at
the top of the kernel mapping and setting the vmalloc size to a fixed
value of 128MB/128GB for 31bit/64bit systems. Part of the vmalloc
area will be used for the vmem_map. This leaves an area of 96MB/1GB
for normal vmalloc allocations.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c   |  6 +++---
 arch/s390/mm/vmem.c        | 20 +++++++-------------
 include/asm-s390/pgtable.h | 46 ++++++++++++++++------------------------------
 3 files changed, 26 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index d68a4025486e..d071a81b62da 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -617,7 +617,7 @@ EXPORT_SYMBOL_GPL(real_memory_size);
 static void __init setup_memory_end(void)
 {
 	unsigned long memory_size;
-	unsigned long max_mem, max_phys;
+	unsigned long max_mem;
 	int i;
 
 #if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE)
@@ -625,10 +625,10 @@ static void __init setup_memory_end(void)
 		memory_end = ZFCPDUMP_HSA_SIZE;
 #endif
 	memory_size = 0;
-	max_phys = VMALLOC_END_INIT - VMALLOC_MIN_SIZE;
 	memory_end &= PAGE_MASK;
 
-	max_mem = memory_end ? min(max_phys, memory_end) : max_phys;
+	max_mem = memory_end ? min(VMALLOC_START, memory_end) : VMALLOC_START;
+	memory_end = min(max_mem, memory_end);
 
 	for (i = 0; i < MEMORY_CHUNKS; i++) {
 		struct mem_chunk *chunk = &memory_chunk[i];
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index fb9c5a85aa56..ee625c8c3b28 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -15,10 +15,6 @@
 #include <asm/setup.h>
 #include <asm/tlbflush.h>
 
-unsigned long vmalloc_end;
-EXPORT_SYMBOL(vmalloc_end);
-
-static struct page *vmem_map;
 static DEFINE_MUTEX(vmem_mutex);
 
 struct memory_segment {
@@ -188,8 +184,8 @@ static int vmem_add_mem_map(unsigned long start, unsigned long size)
 	pte_t  pte;
 	int ret = -ENOMEM;
 
-	map_start = vmem_map + PFN_DOWN(start);
-	map_end	= vmem_map + PFN_DOWN(start + size);
+	map_start = VMEM_MAP + PFN_DOWN(start);
+	map_end	= VMEM_MAP + PFN_DOWN(start + size);
 
 	start_addr = (unsigned long) map_start & PAGE_MASK;
 	end_addr = PFN_ALIGN((unsigned long) map_end);
@@ -254,7 +250,7 @@ static int insert_memory_segment(struct memory_segment *seg)
 {
 	struct memory_segment *tmp;
 
-	if (PFN_DOWN(seg->start + seg->size) > max_pfn ||
+	if (seg->start + seg->size >= VMALLOC_START ||
 	    seg->start + seg->size < seg->start)
 		return -ERANGE;
 
@@ -357,17 +353,15 @@ out:
 
 /*
  * map whole physical memory to virtual memory (identity mapping)
+ * we reserve enough space in the vmalloc area for vmemmap to hotplug
+ * additional memory segments.
  */
 void __init vmem_map_init(void)
 {
-	unsigned long map_size;
 	int i;
 
-	map_size = ALIGN(max_low_pfn, MAX_ORDER_NR_PAGES) * sizeof(struct page);
-	vmalloc_end = PFN_ALIGN(VMALLOC_END_INIT) - PFN_ALIGN(map_size);
-	vmem_map = (struct page *) vmalloc_end;
-	NODE_DATA(0)->node_mem_map = vmem_map;
-
+	BUILD_BUG_ON((unsigned long)VMEM_MAP + VMEM_MAP_SIZE > VMEM_MAP_MAX);
+	NODE_DATA(0)->node_mem_map = VMEM_MAP;
 	for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++)
 		vmem_add_mem(memory_chunk[i].addr, memory_chunk[i].size);
 }
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 1f530f8a6280..79b9eab1a0c7 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -104,41 +104,27 @@ extern char empty_zero_page[PAGE_SIZE];
 
 #ifndef __ASSEMBLY__
 /*
- * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 8MB value just means that there will be a 8MB "hole" after the
- * physical memory until the kernel virtual memory starts.  That means that
- * any out-of-bounds memory accesses will hopefully be caught.
- * The vmalloc() routines leaves a hole of 4kB between each vmalloced
- * area for the same reason. ;)
- * vmalloc area starts at 4GB to prevent syscall table entry exchanging
- * from modules.
- */
-extern unsigned long vmalloc_end;
-
-#ifdef CONFIG_64BIT
-#define VMALLOC_ADDR	(max(0x100000000UL, (unsigned long) high_memory))
-#else
-#define VMALLOC_ADDR	((unsigned long) high_memory)
-#endif
-#define VMALLOC_OFFSET	(8*1024*1024)
-#define VMALLOC_START	((VMALLOC_ADDR + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1))
-#define VMALLOC_END	vmalloc_end
-
-/*
- * We need some free virtual space to be able to do vmalloc.
- * VMALLOC_MIN_SIZE defines the minimum size of the vmalloc
- * area. On a machine with 2GB memory we make sure that we
- * have at least 128MB free space for vmalloc. On a machine
- * with 4TB we make sure we have at least 128GB.
+ * The vmalloc area will always be on the topmost area of the kernel
+ * mapping. We reserve 96MB (31bit) / 1GB (64bit) for vmalloc,
+ * which should be enough for any sane case.
+ * By putting vmalloc at the top, we maximise the gap between physical
+ * memory and vmalloc to catch misplaced memory accesses. As a side
+ * effect, this also makes sure that 64 bit module code cannot be used
+ * as system call address.
  */
 #ifndef __s390x__
-#define VMALLOC_MIN_SIZE	0x8000000UL
-#define VMALLOC_END_INIT	0x80000000UL
+#define VMALLOC_START	0x78000000UL
+#define VMALLOC_END	0x7e000000UL
+#define VMEM_MAP_MAX	0x80000000UL
 #else /* __s390x__ */
-#define VMALLOC_MIN_SIZE	0x2000000000UL
-#define VMALLOC_END_INIT	0x40000000000UL
+#define VMALLOC_START	0x3e000000000UL
+#define VMALLOC_END	0x3e040000000UL
+#define VMEM_MAP_MAX	0x40000000000UL
 #endif /* __s390x__ */
 
+#define VMEM_MAP	((struct page *) VMALLOC_END)
+#define VMEM_MAP_SIZE	((VMALLOC_START / PAGE_SIZE) * sizeof(struct page))
+
 /*
  * A 31 bit pagetable entry of S390 has following format:
  *  |   PFRA          |    |  OS  |
-- 
cgit v1.2.3


From 3b4beb31759765efdda9f9431aebfedf828bbfe0 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:03 +0100
Subject: [S390] Remove owner_pc member from raw_spinlock_t.

Used to contain the address of the holder of the lock. But since the
spinlock code is not inlined anymore all locks contain the same address
anyway. And since in addtition nobody complained about that for ages
its obviously unused. So remove it.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/lib/spinlock.c          | 12 ++++--------
 include/asm-s390/spinlock.h       | 19 ++++++-------------
 include/asm-s390/spinlock_types.h |  1 -
 3 files changed, 10 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c
index 8d76403fcf89..59c56c3d72d0 100644
--- a/arch/s390/lib/spinlock.c
+++ b/arch/s390/lib/spinlock.c
@@ -39,7 +39,7 @@ static inline void _raw_yield_cpu(int cpu)
 		_raw_yield();
 }
 
-void _raw_spin_lock_wait(raw_spinlock_t *lp, unsigned int pc)
+void _raw_spin_lock_wait(raw_spinlock_t *lp)
 {
 	int count = spin_retry;
 	unsigned int cpu = ~smp_processor_id();
@@ -53,15 +53,13 @@ void _raw_spin_lock_wait(raw_spinlock_t *lp, unsigned int pc)
 		}
 		if (__raw_spin_is_locked(lp))
 			continue;
-		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) {
-			lp->owner_pc = pc;
+		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
 			return;
-		}
 	}
 }
 EXPORT_SYMBOL(_raw_spin_lock_wait);
 
-int _raw_spin_trylock_retry(raw_spinlock_t *lp, unsigned int pc)
+int _raw_spin_trylock_retry(raw_spinlock_t *lp)
 {
 	unsigned int cpu = ~smp_processor_id();
 	int count;
@@ -69,10 +67,8 @@ int _raw_spin_trylock_retry(raw_spinlock_t *lp, unsigned int pc)
 	for (count = spin_retry; count > 0; count--) {
 		if (__raw_spin_is_locked(lp))
 			continue;
-		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) {
-			lp->owner_pc = pc;
+		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
 			return 1;
-		}
 	}
 	return 0;
 }
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index 3fd43826fd0b..c1d6e7e304e2 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -58,39 +58,32 @@ _raw_compare_and_swap(volatile unsigned int *lock,
 	do { while (__raw_spin_is_locked(lock)) \
 		 _raw_spin_relax(lock); } while (0)
 
-extern void _raw_spin_lock_wait(raw_spinlock_t *, unsigned int pc);
-extern int _raw_spin_trylock_retry(raw_spinlock_t *, unsigned int pc);
+extern void _raw_spin_lock_wait(raw_spinlock_t *);
+extern int _raw_spin_trylock_retry(raw_spinlock_t *);
 extern void _raw_spin_relax(raw_spinlock_t *lock);
 
 static inline void __raw_spin_lock(raw_spinlock_t *lp)
 {
-	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
 	int old;
 
 	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
-	if (likely(old == 0)) {
-		lp->owner_pc = pc;
+	if (likely(old == 0))
 		return;
-	}
-	_raw_spin_lock_wait(lp, pc);
+	_raw_spin_lock_wait(lp);
 }
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lp)
 {
-	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
 	int old;
 
 	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
-	if (likely(old == 0)) {
-		lp->owner_pc = pc;
+	if (likely(old == 0))
 		return 1;
-	}
-	return _raw_spin_trylock_retry(lp, pc);
+	return _raw_spin_trylock_retry(lp);
 }
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lp)
 {
-	lp->owner_pc = 0;
 	_raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0);
 }
 		
diff --git a/include/asm-s390/spinlock_types.h b/include/asm-s390/spinlock_types.h
index b7ac13f7aa37..654abc40de04 100644
--- a/include/asm-s390/spinlock_types.h
+++ b/include/asm-s390/spinlock_types.h
@@ -7,7 +7,6 @@
 
 typedef struct {
 	volatile unsigned int owner_cpu;
-	volatile unsigned int owner_pc;
 } __attribute__ ((aligned (4))) raw_spinlock_t;
 
 #define __RAW_SPIN_LOCK_UNLOCKED	{ 0 }
-- 
cgit v1.2.3


From 519580fc179be6d06a40edd918747bd5f0e1b899 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:04 +0100
Subject: [S390] Use new style spinlock initializer in __RWSEM_INITIALIZER.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/rwsem.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/rwsem.h b/include/asm-s390/rwsem.h
index 90f4eccaa290..9d2a17971805 100644
--- a/include/asm-s390/rwsem.h
+++ b/include/asm-s390/rwsem.h
@@ -91,8 +91,8 @@ struct rw_semaphore {
 #endif
 
 #define __RWSEM_INITIALIZER(name) \
-{ RWSEM_UNLOCKED_VALUE, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) \
-  __RWSEM_DEP_MAP_INIT(name) }
+ { RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait.lock), \
+   LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) }
 
 #define DECLARE_RWSEM(name) \
 	struct rw_semaphore name = __RWSEM_INITIALIZER(name)
-- 
cgit v1.2.3


From 48483b3290988952a593c6e66ca354c19f1a4350 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:05 +0100
Subject: [S390] Get rid of additional_cpus kernel parameter.

It caused only a lot of confusion. From now on cpu hotplug of up to
NR_CPUS will work by default. If somebody wants to limit that then
the possible_cpus parameter can be used.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 Documentation/cpu-hotplug.txt |   2 +-
 arch/s390/kernel/early.c      |   1 -
 arch/s390/kernel/setup.c      |   1 -
 arch/s390/kernel/smp.c        | 180 ++++++++++++++++++------------------------
 drivers/s390/char/sclp_cmd.c  |  46 +----------
 include/asm-s390/sclp.h       |   1 -
 include/asm-s390/smp.h        |   3 -
 7 files changed, 79 insertions(+), 155 deletions(-)

(limited to 'include')

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index fb94f5a71b68..ba0aacde94fb 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -50,7 +50,7 @@ additional_cpus=n (*)	Use this to limit hotpluggable cpus. This option sets
   			cpu_possible_map = cpu_present_map + additional_cpus
 
 (*) Option valid only for following architectures
-- x86_64, ia64, s390
+- x86_64, ia64
 
 ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
 to determine the number of potentially hot-pluggable cpus. The implementation
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index c7cbb0114147..9f7b73b180f0 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -278,7 +278,6 @@ void __init startup_init(void)
 	setup_lowcore_early();
 	sclp_read_info_early();
 	sclp_facilities_detect();
-	sclp_read_cpu_info_early();
 	memsize = sclp_memory_detect();
 #ifndef CONFIG_64BIT
 	/*
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 3a61bfc2c4fb..cbdf3fb05e81 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -922,7 +922,6 @@ setup_arch(char **cmdline_p)
 
         cpu_init();
         __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr;
-	smp_setup_cpu_possible_map();
 
 	/*
 	 * Setup capabilities (ELF_HWCAP & ELF_PLATFORM).
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 66fe28930d82..320e4e97bf52 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -54,7 +54,7 @@ EXPORT_SYMBOL(lowcore_ptr);
 cpumask_t cpu_online_map = CPU_MASK_NONE;
 EXPORT_SYMBOL(cpu_online_map);
 
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
+cpumask_t cpu_possible_map = CPU_MASK_ALL;
 EXPORT_SYMBOL(cpu_possible_map);
 
 static struct task_struct *current_set[NR_CPUS];
@@ -399,7 +399,7 @@ static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu)
 		       "kernel was compiled with NR_CPUS=%i\n", cpu, NR_CPUS);
 		return;
 	}
-	zfcpdump_save_areas[cpu] = alloc_bootmem(sizeof(union save_area));
+	zfcpdump_save_areas[cpu] = kmalloc(sizeof(union save_area), GFP_KERNEL);
 	__cpu_logical_map[CPU_INIT_NO] = (__u16) phy_cpu;
 	while (signal_processor(CPU_INIT_NO, sigp_stop_and_store_status) ==
 	       sigp_busy)
@@ -435,67 +435,6 @@ static int cpu_stopped(int cpu)
 	return 0;
 }
 
-/*
- * Lets check how many CPUs we have.
- */
-static void __init smp_count_cpus(unsigned int *configured_cpus,
-				  unsigned int *standby_cpus)
-{
-	unsigned int cpu;
-	struct sclp_cpu_info *info;
-	u16 boot_cpu_addr, cpu_addr;
-
-	boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr;
-	current_thread_info()->cpu = 0;
-	*configured_cpus = 1;
-	*standby_cpus = 0;
-
-	info = alloc_bootmem_pages(sizeof(*info));
-	if (!info)
-		disabled_wait((unsigned long) __builtin_return_address(0));
-
-	/* Use sigp detection algorithm if sclp doesn't work. */
-	if (sclp_get_cpu_info(info)) {
-		smp_use_sigp_detection = 1;
-		for (cpu = 0; cpu <= 65535; cpu++) {
-			if (cpu == boot_cpu_addr)
-				continue;
-			__cpu_logical_map[CPU_INIT_NO] = cpu;
-			if (cpu_stopped(CPU_INIT_NO))
-				(*configured_cpus)++;
-		}
-		goto out;
-	}
-
-	if (info->has_cpu_type) {
-		for (cpu = 0; cpu < info->combined; cpu++) {
-			if (info->cpu[cpu].address == boot_cpu_addr) {
-				smp_cpu_type = info->cpu[cpu].type;
-				break;
-			}
-		}
-	}
-	/* Count cpus. */
-	for (cpu = 0; cpu < info->combined; cpu++) {
-		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
-			continue;
-		cpu_addr = info->cpu[cpu].address;
-		if (cpu_addr == boot_cpu_addr)
-			continue;
-		__cpu_logical_map[CPU_INIT_NO] = cpu_addr;
-		if (!cpu_stopped(CPU_INIT_NO)) {
-			(*standby_cpus)++;
-			continue;
-		}
-		smp_get_save_area(*configured_cpus, cpu_addr);
-		(*configured_cpus)++;
-	}
-out:
-	printk(KERN_INFO "CPUs: %d configured, %d standby\n",
-	       *configured_cpus, *standby_cpus);
-	free_bootmem((unsigned long) info, sizeof(*info));
-}
-
 static int cpu_known(int cpu_id)
 {
 	int cpu;
@@ -529,7 +468,7 @@ static int smp_rescan_cpus_sigp(cpumask_t avail)
 	return 0;
 }
 
-static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail)
+static int smp_rescan_cpus_sclp(cpumask_t avail)
 {
 	struct sclp_cpu_info *info;
 	int cpu_id, logical_cpu, cpu;
@@ -538,10 +477,7 @@ static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail)
 	logical_cpu = first_cpu(avail);
 	if (logical_cpu == NR_CPUS)
 		return 0;
-	if (slab_is_available())
-		info = kmalloc(sizeof(*info), GFP_KERNEL);
-	else
-		info = alloc_bootmem(sizeof(*info));
+	info = kmalloc(sizeof(*info), GFP_KERNEL);
 	if (!info)
 		return -ENOMEM;
 	rc = sclp_get_cpu_info(info);
@@ -564,10 +500,7 @@ static int __init_refok smp_rescan_cpus_sclp(cpumask_t avail)
 			break;
 	}
 out:
-	if (slab_is_available())
-		kfree(info);
-	else
-		free_bootmem((unsigned long) info, sizeof(*info));
+	kfree(info);
 	return rc;
 }
 
@@ -575,15 +508,71 @@ static int smp_rescan_cpus(void)
 {
 	cpumask_t avail;
 
-	cpus_setall(avail);
-	cpus_and(avail, avail, cpu_possible_map);
-	cpus_andnot(avail, avail, cpu_present_map);
+	cpus_xor(avail, cpu_possible_map, cpu_present_map);
 	if (smp_use_sigp_detection)
 		return smp_rescan_cpus_sigp(avail);
 	else
 		return smp_rescan_cpus_sclp(avail);
 }
 
+static void __init smp_detect_cpus(void)
+{
+	unsigned int cpu, c_cpus, s_cpus;
+	struct sclp_cpu_info *info;
+	u16 boot_cpu_addr, cpu_addr;
+
+	c_cpus = 1;
+	s_cpus = 0;
+	boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr;
+	info = kmalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
+		panic("smp_detect_cpus failed to allocate memory\n");
+	/* Use sigp detection algorithm if sclp doesn't work. */
+	if (sclp_get_cpu_info(info)) {
+		smp_use_sigp_detection = 1;
+		for (cpu = 0; cpu <= 65535; cpu++) {
+			if (cpu == boot_cpu_addr)
+				continue;
+			__cpu_logical_map[CPU_INIT_NO] = cpu;
+			if (!cpu_stopped(CPU_INIT_NO))
+				continue;
+			smp_get_save_area(c_cpus, cpu);
+			c_cpus++;
+		}
+		goto out;
+	}
+
+	if (info->has_cpu_type) {
+		for (cpu = 0; cpu < info->combined; cpu++) {
+			if (info->cpu[cpu].address == boot_cpu_addr) {
+				smp_cpu_type = info->cpu[cpu].type;
+				break;
+			}
+		}
+	}
+
+	for (cpu = 0; cpu < info->combined; cpu++) {
+		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
+			continue;
+		cpu_addr = info->cpu[cpu].address;
+		if (cpu_addr == boot_cpu_addr)
+			continue;
+		__cpu_logical_map[CPU_INIT_NO] = cpu_addr;
+		if (!cpu_stopped(CPU_INIT_NO)) {
+			s_cpus++;
+			continue;
+		}
+		smp_get_save_area(c_cpus, cpu_addr);
+		c_cpus++;
+	}
+out:
+	kfree(info);
+	printk(KERN_INFO "CPUs: %d configured, %d standby\n", c_cpus, s_cpus);
+	lock_cpu_hotplug();
+	smp_rescan_cpus();
+	unlock_cpu_hotplug();
+}
+
 /*
  *	Activate a secondary processor.
  */
@@ -674,41 +663,20 @@ int __cpu_up(unsigned int cpu)
 	return 0;
 }
 
-static unsigned int __initdata additional_cpus;
-static unsigned int __initdata possible_cpus;
-
-void __init smp_setup_cpu_possible_map(void)
-{
-	unsigned int pos_cpus, cpu;
-	unsigned int configured_cpus, standby_cpus;
-
-	smp_count_cpus(&configured_cpus, &standby_cpus);
-	pos_cpus = min(configured_cpus + standby_cpus + additional_cpus,
-		       (unsigned int) NR_CPUS);
-	if (possible_cpus)
-		pos_cpus = min(possible_cpus, (unsigned int) NR_CPUS);
-	for (cpu = 0; cpu < pos_cpus; cpu++)
-		cpu_set(cpu, cpu_possible_map);
-	cpu_present_map = cpumask_of_cpu(0);
-	smp_rescan_cpus();
-}
-
-#ifdef CONFIG_HOTPLUG_CPU
-
-static int __init setup_additional_cpus(char *s)
-{
-	additional_cpus = simple_strtoul(s, NULL, 0);
-	return 0;
-}
-early_param("additional_cpus", setup_additional_cpus);
-
 static int __init setup_possible_cpus(char *s)
 {
-	possible_cpus = simple_strtoul(s, NULL, 0);
+	int pcpus, cpu;
+
+	pcpus = simple_strtoul(s, NULL, 0);
+	cpu_possible_map = cpumask_of_cpu(0);
+	for (cpu = 1; cpu < pcpus && cpu < NR_CPUS; cpu++)
+		cpu_set(cpu, cpu_possible_map);
 	return 0;
 }
 early_param("possible_cpus", setup_possible_cpus);
 
+#ifdef CONFIG_HOTPLUG_CPU
+
 int __cpu_disable(void)
 {
 	struct ec_creg_mask_parms cr_parms;
@@ -768,6 +736,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	unsigned int cpu;
 	int i;
 
+	smp_detect_cpus();
+
 	/* request the 0x1201 emergency signal external interrupt */
 	if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0)
 		panic("Couldn't request external interrupt 0x1201");
@@ -816,6 +786,8 @@ void __init smp_prepare_boot_cpu(void)
 {
 	BUG_ON(smp_processor_id() != 0);
 
+	current_thread_info()->cpu = 0;
+	cpu_set(0, cpu_present_map);
 	cpu_set(0, cpu_online_map);
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
 	current_set[0] = current;
diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c
index d7e6f4d65b78..b5c23396f8fe 100644
--- a/drivers/s390/char/sclp_cmd.c
+++ b/drivers/s390/char/sclp_cmd.c
@@ -191,9 +191,6 @@ struct read_cpu_info_sccb {
 	u8	reserved[4096 - 16];
 } __attribute__((packed, aligned(PAGE_SIZE)));
 
-static struct read_cpu_info_sccb __initdata early_read_cpu_info_sccb;
-static struct sclp_cpu_info __initdata sclp_cpu_info;
-
 static void sclp_fill_cpu_info(struct sclp_cpu_info *info,
 			       struct read_cpu_info_sccb *sccb)
 {
@@ -208,48 +205,16 @@ static void sclp_fill_cpu_info(struct sclp_cpu_info *info,
 	       info->combined * sizeof(struct sclp_cpu_entry));
 }
 
-void __init sclp_read_cpu_info_early(void)
-{
-	int rc;
-	struct read_cpu_info_sccb *sccb;
-
-	if (!SCLP_HAS_CPU_INFO)
-		return;
-
-	sccb = &early_read_cpu_info_sccb;
-	do {
-		memset(sccb, 0, sizeof(*sccb));
-		sccb->header.length = sizeof(*sccb);
-		rc = sclp_cmd_sync_early(SCLP_CMDW_READ_CPU_INFO, sccb);
-	} while (rc == -EBUSY);
-
-	if (rc)
-		return;
-	if (sccb->header.response_code != 0x10)
-		return;
-	sclp_fill_cpu_info(&sclp_cpu_info, sccb);
-}
-
-static int __init sclp_get_cpu_info_early(struct sclp_cpu_info *info)
-{
-	if (!SCLP_HAS_CPU_INFO)
-		return -EOPNOTSUPP;
-	*info = sclp_cpu_info;
-	return 0;
-}
-
-static int sclp_get_cpu_info_late(struct sclp_cpu_info *info)
+int sclp_get_cpu_info(struct sclp_cpu_info *info)
 {
 	int rc;
 	struct read_cpu_info_sccb *sccb;
 
 	if (!SCLP_HAS_CPU_INFO)
 		return -EOPNOTSUPP;
-	sccb = (struct read_cpu_info_sccb *)  __get_free_page(GFP_KERNEL
-							      | GFP_DMA);
+	sccb = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
 	if (!sccb)
 		return -ENOMEM;
-	memset(sccb, 0, sizeof(*sccb));
 	sccb->header.length = sizeof(*sccb);
 	rc = do_sync_request(SCLP_CMDW_READ_CPU_INFO, sccb);
 	if (rc)
@@ -266,13 +231,6 @@ out:
 	return rc;
 }
 
-int __init_refok sclp_get_cpu_info(struct sclp_cpu_info *info)
-{
-	if (slab_is_available())
-		return sclp_get_cpu_info_late(info);
-	return sclp_get_cpu_info_early(info);
-}
-
 struct cpu_configure_sccb {
 	struct sccb_header header;
 } __attribute__((packed, aligned(8)));
diff --git a/include/asm-s390/sclp.h b/include/asm-s390/sclp.h
index b8c7695cd4c8..b5f2843013a3 100644
--- a/include/asm-s390/sclp.h
+++ b/include/asm-s390/sclp.h
@@ -46,7 +46,6 @@ int sclp_get_cpu_info(struct sclp_cpu_info *info);
 int sclp_cpu_configure(u8 cpu);
 int sclp_cpu_deconfigure(u8 cpu);
 void sclp_read_info_early(void);
-void sclp_read_cpu_info_early(void);
 void sclp_facilities_detect(void);
 unsigned long long sclp_memory_detect(void);
 int sclp_sdias_blk_count(void);
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 07708c07701e..218454b91862 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -35,8 +35,6 @@ extern void machine_restart_smp(char *);
 extern void machine_halt_smp(void);
 extern void machine_power_off_smp(void);
 
-extern void smp_setup_cpu_possible_map(void);
-
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 /*
@@ -103,7 +101,6 @@ static inline void smp_send_stop(void)
 
 #define hard_smp_processor_id()		0
 #define smp_cpu_not_running(cpu)	1
-#define smp_setup_cpu_possible_map()	do { } while (0)
 #endif
 
 extern union save_area *zfcpdump_save_areas[NR_CPUS + 1];
-- 
cgit v1.2.3


From 99ca4e582d4a4088969681efff97be44d98421a1 Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:11 +0100
Subject: [S390] kernel: Shutdown Actions Interface

In case of a kernel panic it is currently possible to specify that a dump
should be created, the system should be rebooted or stopped. Virtual sysfs
files under the directory /sys/firmware/ are used for that configuration.
In addition to that, there are kernel parameters 'vmhalt', 'vmpoff'
and 'vmpanic', which can be used to specify z/VM commands, which are
automatically executed in case of halt, power off or a kernel panic.
This patch combines both functionalities and allows to specify the z/VM CP
commands also via sysfs attributes. In addition to that, it enhances the
existing handling of shutdown triggers (e.g. halt or panic) and associated
shutdown actions (e.g. dump or reipl) and makes it more flexible.

Signed-off-by: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ipl.c   | 953 ++++++++++++++++++++++++++++++-----------------
 arch/s390/kernel/setup.c | 103 +----
 arch/s390/kernel/smp.c   |  27 --
 include/asm-s390/ipl.h   |   4 +-
 4 files changed, 614 insertions(+), 473 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index b97694fa62ec..d73aff63725f 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -2,7 +2,7 @@
  *  arch/s390/kernel/ipl.c
  *    ipl/reipl/dump support for Linux on s390.
  *
- *    Copyright (C) IBM Corp. 2005,2006
+ *    Copyright IBM Corp. 2005,2007
  *    Author(s): Michael Holzheu <holzheu@de.ibm.com>
  *		 Heiko Carstens <heiko.carstens@de.ibm.com>
  *		 Volker Sameske <sameske@de.ibm.com>
@@ -31,6 +31,43 @@
 #define IPL_FCP_DUMP_STR	"fcp_dump"
 #define IPL_NSS_STR		"nss"
 
+#define DUMP_CCW_STR		"ccw"
+#define DUMP_FCP_STR		"fcp"
+#define DUMP_NONE_STR		"none"
+
+/*
+ * Four shutdown trigger types are supported:
+ * - panic
+ * - halt
+ * - power off
+ * - reipl
+ */
+#define ON_PANIC_STR		"on_panic"
+#define ON_HALT_STR		"on_halt"
+#define ON_POFF_STR		"on_poff"
+#define ON_REIPL_STR		"on_reboot"
+
+struct shutdown_action;
+struct shutdown_trigger {
+	char *name;
+	struct shutdown_action *action;
+};
+
+/*
+ * Five shutdown action types are supported:
+ */
+#define SHUTDOWN_ACTION_IPL_STR		"ipl"
+#define SHUTDOWN_ACTION_REIPL_STR	"reipl"
+#define SHUTDOWN_ACTION_DUMP_STR	"dump"
+#define SHUTDOWN_ACTION_VMCMD_STR	"vmcmd"
+#define SHUTDOWN_ACTION_STOP_STR	"stop"
+
+struct shutdown_action {
+	char *name;
+	void (*fn) (struct shutdown_trigger *trigger);
+	int (*init) (void);
+};
+
 static char *ipl_type_str(enum ipl_type type)
 {
 	switch (type) {
@@ -54,10 +91,6 @@ enum dump_type {
 	DUMP_TYPE_FCP	= 4,
 };
 
-#define DUMP_NONE_STR	 "none"
-#define DUMP_CCW_STR	 "ccw"
-#define DUMP_FCP_STR	 "fcp"
-
 static char *dump_type_str(enum dump_type type)
 {
 	switch (type) {
@@ -99,30 +132,6 @@ enum dump_method {
 	DUMP_METHOD_FCP_DIAG,
 };
 
-enum shutdown_action {
-	SHUTDOWN_REIPL,
-	SHUTDOWN_DUMP,
-	SHUTDOWN_STOP,
-};
-
-#define SHUTDOWN_REIPL_STR "reipl"
-#define SHUTDOWN_DUMP_STR  "dump"
-#define SHUTDOWN_STOP_STR  "stop"
-
-static char *shutdown_action_str(enum shutdown_action action)
-{
-	switch (action) {
-	case SHUTDOWN_REIPL:
-		return SHUTDOWN_REIPL_STR;
-	case SHUTDOWN_DUMP:
-		return SHUTDOWN_DUMP_STR;
-	case SHUTDOWN_STOP:
-		return SHUTDOWN_STOP_STR;
-	default:
-		return NULL;
-	}
-}
-
 static int diag308_set_works = 0;
 
 static int reipl_capabilities = IPL_TYPE_UNKNOWN;
@@ -140,8 +149,6 @@ static enum dump_method dump_method = DUMP_METHOD_NONE;
 static struct ipl_parameter_block *dump_block_fcp;
 static struct ipl_parameter_block *dump_block_ccw;
 
-static enum shutdown_action on_panic_action = SHUTDOWN_STOP;
-
 static struct sclp_ipl_info sclp_ipl_info;
 
 int diag308(unsigned long subcode, void *addr)
@@ -205,8 +212,8 @@ static ssize_t sys_##_prefix##_##_name##_store(struct kobject *kobj,	\
 		struct kobj_attribute *attr,				\
 		const char *buf, size_t len)				\
 {									\
-	if (sscanf(buf, _fmt_in, _value) != 1)				\
-		return -EINVAL;						\
+	strncpy(_value, buf, sizeof(_value) - 1);			\
+	strstrip(_value);						\
 	return len;							\
 }									\
 static struct kobj_attribute sys_##_prefix##_##_name##_attr =		\
@@ -245,33 +252,6 @@ static __init enum ipl_type get_ipl_type(void)
 	return IPL_TYPE_FCP;
 }
 
-void __init setup_ipl_info(void)
-{
-	ipl_info.type = get_ipl_type();
-	switch (ipl_info.type) {
-	case IPL_TYPE_CCW:
-		ipl_info.data.ccw.dev_id.devno = ipl_devno;
-		ipl_info.data.ccw.dev_id.ssid = 0;
-		break;
-	case IPL_TYPE_FCP:
-	case IPL_TYPE_FCP_DUMP:
-		ipl_info.data.fcp.dev_id.devno =
-			IPL_PARMBLOCK_START->ipl_info.fcp.devno;
-		ipl_info.data.fcp.dev_id.ssid = 0;
-		ipl_info.data.fcp.wwpn = IPL_PARMBLOCK_START->ipl_info.fcp.wwpn;
-		ipl_info.data.fcp.lun = IPL_PARMBLOCK_START->ipl_info.fcp.lun;
-		break;
-	case IPL_TYPE_NSS:
-		strncpy(ipl_info.data.nss.name, kernel_nss_name,
-			sizeof(ipl_info.data.nss.name));
-		break;
-	case IPL_TYPE_UNKNOWN:
-	default:
-		/* We have no info to copy */
-		break;
-	}
-}
-
 struct ipl_info ipl_info;
 EXPORT_SYMBOL_GPL(ipl_info);
 
@@ -428,8 +408,74 @@ static struct attribute_group ipl_unknown_attr_group = {
 
 static struct kset *ipl_kset;
 
+static int __init ipl_register_fcp_files(void)
+{
+	int rc;
+
+	rc = sysfs_create_group(&ipl_kset->kobj, &ipl_fcp_attr_group);
+	if (rc)
+		goto out;
+	rc = sysfs_create_bin_file(&ipl_kset->kobj, &ipl_parameter_attr);
+	if (rc)
+		goto out_ipl_parm;
+	rc = sysfs_create_bin_file(&ipl_kset->kobj, &ipl_scp_data_attr);
+	if (!rc)
+		goto out;
+
+	sysfs_remove_bin_file(&ipl_kset->kobj, &ipl_parameter_attr);
+
+out_ipl_parm:
+	sysfs_remove_group(&ipl_kset->kobj, &ipl_fcp_attr_group);
+out:
+	return rc;
+}
+
+static void ipl_run(struct shutdown_trigger *trigger)
+{
+	diag308(DIAG308_IPL, NULL);
+	if (MACHINE_IS_VM)
+		__cpcmd("IPL", NULL, 0, NULL);
+	else if (ipl_info.type == IPL_TYPE_CCW)
+		reipl_ccw_dev(&ipl_info.data.ccw.dev_id);
+}
+
+static int ipl_init(void)
+{
+	int rc;
+
+	ipl_kset = kset_create_and_add("ipl", NULL, firmware_kobj);
+	if (!ipl_kset) {
+		rc = -ENOMEM;
+		goto out;
+	}
+	switch (ipl_info.type) {
+	case IPL_TYPE_CCW:
+		rc = sysfs_create_group(&ipl_kset->kobj, &ipl_ccw_attr_group);
+		break;
+	case IPL_TYPE_FCP:
+	case IPL_TYPE_FCP_DUMP:
+		rc = ipl_register_fcp_files();
+		break;
+	case IPL_TYPE_NSS:
+		rc = sysfs_create_group(&ipl_kset->kobj, &ipl_nss_attr_group);
+		break;
+	default:
+		rc = sysfs_create_group(&ipl_kset->kobj,
+					&ipl_unknown_attr_group);
+		break;
+	}
+out:
+	if (rc)
+		panic("ipl_init failed: rc = %i\n", rc);
+
+	return 0;
+}
+
+static struct shutdown_action ipl_action = {SHUTDOWN_ACTION_IPL_STR, ipl_run,
+					    ipl_init};
+
 /*
- * reipl section
+ * reipl shutdown action: Reboot Linux on shutdown.
  */
 
 /* FCP reipl device attributes */
@@ -600,143 +646,11 @@ static ssize_t reipl_type_store(struct kobject *kobj,
 }
 
 static struct kobj_attribute reipl_type_attr =
-		__ATTR(reipl_type, 0644, reipl_type_show, reipl_type_store);
+	__ATTR(reipl_type, 0644, reipl_type_show, reipl_type_store);
 
 static struct kset *reipl_kset;
 
-/*
- * dump section
- */
-
-/* FCP dump device attributes */
-
-DEFINE_IPL_ATTR_RW(dump_fcp, wwpn, "0x%016llx\n", "%016llx\n",
-		   dump_block_fcp->ipl_info.fcp.wwpn);
-DEFINE_IPL_ATTR_RW(dump_fcp, lun, "0x%016llx\n", "%016llx\n",
-		   dump_block_fcp->ipl_info.fcp.lun);
-DEFINE_IPL_ATTR_RW(dump_fcp, bootprog, "%lld\n", "%lld\n",
-		   dump_block_fcp->ipl_info.fcp.bootprog);
-DEFINE_IPL_ATTR_RW(dump_fcp, br_lba, "%lld\n", "%lld\n",
-		   dump_block_fcp->ipl_info.fcp.br_lba);
-DEFINE_IPL_ATTR_RW(dump_fcp, device, "0.0.%04llx\n", "0.0.%llx\n",
-		   dump_block_fcp->ipl_info.fcp.devno);
-
-static struct attribute *dump_fcp_attrs[] = {
-	&sys_dump_fcp_device_attr.attr,
-	&sys_dump_fcp_wwpn_attr.attr,
-	&sys_dump_fcp_lun_attr.attr,
-	&sys_dump_fcp_bootprog_attr.attr,
-	&sys_dump_fcp_br_lba_attr.attr,
-	NULL,
-};
-
-static struct attribute_group dump_fcp_attr_group = {
-	.name  = IPL_FCP_STR,
-	.attrs = dump_fcp_attrs,
-};
-
-/* CCW dump device attributes */
-
-DEFINE_IPL_ATTR_RW(dump_ccw, device, "0.0.%04llx\n", "0.0.%llx\n",
-		   dump_block_ccw->ipl_info.ccw.devno);
-
-static struct attribute *dump_ccw_attrs[] = {
-	&sys_dump_ccw_device_attr.attr,
-	NULL,
-};
-
-static struct attribute_group dump_ccw_attr_group = {
-	.name  = IPL_CCW_STR,
-	.attrs = dump_ccw_attrs,
-};
-
-/* dump type */
-
-static int dump_set_type(enum dump_type type)
-{
-	if (!(dump_capabilities & type))
-		return -EINVAL;
-	switch(type) {
-	case DUMP_TYPE_CCW:
-		if (MACHINE_IS_VM)
-			dump_method = DUMP_METHOD_CCW_VM;
-		else if (diag308_set_works)
-			dump_method = DUMP_METHOD_CCW_DIAG;
-		else
-			dump_method = DUMP_METHOD_CCW_CIO;
-		break;
-	case DUMP_TYPE_FCP:
-		dump_method = DUMP_METHOD_FCP_DIAG;
-		break;
-	default:
-		dump_method = DUMP_METHOD_NONE;
-	}
-	dump_type = type;
-	return 0;
-}
-
-static ssize_t dump_type_show(struct kobject *kobj,
-			      struct kobj_attribute *attr, char *page)
-{
-	return sprintf(page, "%s\n", dump_type_str(dump_type));
-}
-
-static ssize_t dump_type_store(struct kobject *kobj,
-			       struct kobj_attribute *attr,
-			       const char *buf, size_t len)
-{
-	int rc = -EINVAL;
-
-	if (strncmp(buf, DUMP_NONE_STR, strlen(DUMP_NONE_STR)) == 0)
-		rc = dump_set_type(DUMP_TYPE_NONE);
-	else if (strncmp(buf, DUMP_CCW_STR, strlen(DUMP_CCW_STR)) == 0)
-		rc = dump_set_type(DUMP_TYPE_CCW);
-	else if (strncmp(buf, DUMP_FCP_STR, strlen(DUMP_FCP_STR)) == 0)
-		rc = dump_set_type(DUMP_TYPE_FCP);
-	return (rc != 0) ? rc : len;
-}
-
-static struct kobj_attribute dump_type_attr =
-		__ATTR(dump_type, 0644, dump_type_show, dump_type_store);
-
-static struct kset *dump_kset;
-
-/*
- * Shutdown actions section
- */
-
-static struct kset *shutdown_actions_kset;
-
-/* on panic */
-
-static ssize_t on_panic_show(struct kobject *kobj,
-			     struct kobj_attribute *attr, char *page)
-{
-	return sprintf(page, "%s\n", shutdown_action_str(on_panic_action));
-}
-
-static ssize_t on_panic_store(struct kobject *kobj,
-			      struct kobj_attribute *attr,
-			      const char *buf, size_t len)
-{
-	if (strncmp(buf, SHUTDOWN_REIPL_STR, strlen(SHUTDOWN_REIPL_STR)) == 0)
-		on_panic_action = SHUTDOWN_REIPL;
-	else if (strncmp(buf, SHUTDOWN_DUMP_STR,
-			 strlen(SHUTDOWN_DUMP_STR)) == 0)
-		on_panic_action = SHUTDOWN_DUMP;
-	else if (strncmp(buf, SHUTDOWN_STOP_STR,
-			 strlen(SHUTDOWN_STOP_STR)) == 0)
-		on_panic_action = SHUTDOWN_STOP;
-	else
-		return -EINVAL;
-
-	return len;
-}
-
-static struct kobj_attribute on_panic_attr =
-		__ATTR(on_panic, 0644, on_panic_show, on_panic_store);
-
-void do_reipl(void)
+void reipl_run(struct shutdown_trigger *trigger)
 {
 	struct ccw_dev_id devid;
 	static char buf[100];
@@ -787,119 +701,27 @@ void do_reipl(void)
 	default:
 		break;
 	}
-	signal_processor(smp_processor_id(), sigp_stop_and_store_status);
 }
 
-static void do_dump(void)
+static void __init reipl_probe(void)
 {
-	struct ccw_dev_id devid;
-	static char buf[100];
+	void *buffer;
 
-	switch (dump_method) {
-	case DUMP_METHOD_CCW_CIO:
-		smp_send_stop();
-		devid.devno = dump_block_ccw->ipl_info.ccw.devno;
-		devid.ssid  = 0;
-		reipl_ccw_dev(&devid);
-		break;
-	case DUMP_METHOD_CCW_VM:
-		smp_send_stop();
-		sprintf(buf, "STORE STATUS");
-		__cpcmd(buf, NULL, 0, NULL);
-		sprintf(buf, "IPL %X", dump_block_ccw->ipl_info.ccw.devno);
-		__cpcmd(buf, NULL, 0, NULL);
-		break;
-	case DUMP_METHOD_CCW_DIAG:
-		diag308(DIAG308_SET, dump_block_ccw);
-		diag308(DIAG308_DUMP, NULL);
-		break;
-	case DUMP_METHOD_FCP_DIAG:
-		diag308(DIAG308_SET, dump_block_fcp);
-		diag308(DIAG308_DUMP, NULL);
-		break;
-	case DUMP_METHOD_NONE:
-	default:
+	buffer = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!buffer)
 		return;
-	}
-	printk(KERN_EMERG "Dump failed!\n");
+	if (diag308(DIAG308_STORE, buffer) == DIAG308_RC_OK)
+		diag308_set_works = 1;
+	free_page((unsigned long)buffer);
 }
 
-/* init functions */
-
-static int __init ipl_register_fcp_files(void)
+static int __init reipl_nss_init(void)
 {
 	int rc;
 
-	rc = sysfs_create_group(&ipl_kset->kobj,
-				&ipl_fcp_attr_group);
-	if (rc)
-		goto out;
-	rc = sysfs_create_bin_file(&ipl_kset->kobj,
-				   &ipl_parameter_attr);
-	if (rc)
-		goto out_ipl_parm;
-	rc = sysfs_create_bin_file(&ipl_kset->kobj,
-				   &ipl_scp_data_attr);
-	if (!rc)
-		goto out;
-
-	sysfs_remove_bin_file(&ipl_kset->kobj, &ipl_parameter_attr);
-
-out_ipl_parm:
-	sysfs_remove_group(&ipl_kset->kobj, &ipl_fcp_attr_group);
-out:
-	return rc;
-}
-
-static int __init ipl_init(void)
-{
-	int rc;
-
-	ipl_kset = kset_create_and_add("ipl", NULL, firmware_kobj);
-	if (!ipl_kset)
-		return -ENOMEM;
-	switch (ipl_info.type) {
-	case IPL_TYPE_CCW:
-		rc = sysfs_create_group(&ipl_kset->kobj,
-					&ipl_ccw_attr_group);
-		break;
-	case IPL_TYPE_FCP:
-	case IPL_TYPE_FCP_DUMP:
-		rc = ipl_register_fcp_files();
-		break;
-	case IPL_TYPE_NSS:
-		rc = sysfs_create_group(&ipl_kset->kobj,
-					&ipl_nss_attr_group);
-		break;
-	default:
-		rc = sysfs_create_group(&ipl_kset->kobj,
-					&ipl_unknown_attr_group);
-		break;
-	}
-	if (rc)
-		kset_unregister(ipl_kset);
-	return rc;
-}
-
-static void __init reipl_probe(void)
-{
-	void *buffer;
-
-	buffer = (void *) get_zeroed_page(GFP_KERNEL);
-	if (!buffer)
-		return;
-	if (diag308(DIAG308_STORE, buffer) == DIAG308_RC_OK)
-		diag308_set_works = 1;
-	free_page((unsigned long)buffer);
-}
-
-static int __init reipl_nss_init(void)
-{
-	int rc;
-
-	if (!MACHINE_IS_VM)
-		return 0;
-	rc = sysfs_create_group(&reipl_kset->kobj, &reipl_nss_attr_group);
+	if (!MACHINE_IS_VM)
+		return 0;
+	rc = sysfs_create_group(&reipl_kset->kobj, &reipl_nss_attr_group);
 	if (rc)
 		return rc;
 	strncpy(reipl_nss_name, kernel_nss_name, NSS_NAME_SIZE + 1);
@@ -970,7 +792,7 @@ static int __init reipl_fcp_init(void)
 	return 0;
 }
 
-static int __init reipl_init(void)
+static int reipl_init(void)
 {
 	int rc;
 
@@ -997,6 +819,138 @@ static int __init reipl_init(void)
 	return 0;
 }
 
+static struct shutdown_action reipl_action = {SHUTDOWN_ACTION_REIPL_STR,
+					      reipl_run, reipl_init};
+
+/*
+ * dump shutdown action: Dump Linux on shutdown.
+ */
+
+/* FCP dump device attributes */
+
+DEFINE_IPL_ATTR_RW(dump_fcp, wwpn, "0x%016llx\n", "%016llx\n",
+		   dump_block_fcp->ipl_info.fcp.wwpn);
+DEFINE_IPL_ATTR_RW(dump_fcp, lun, "0x%016llx\n", "%016llx\n",
+		   dump_block_fcp->ipl_info.fcp.lun);
+DEFINE_IPL_ATTR_RW(dump_fcp, bootprog, "%lld\n", "%lld\n",
+		   dump_block_fcp->ipl_info.fcp.bootprog);
+DEFINE_IPL_ATTR_RW(dump_fcp, br_lba, "%lld\n", "%lld\n",
+		   dump_block_fcp->ipl_info.fcp.br_lba);
+DEFINE_IPL_ATTR_RW(dump_fcp, device, "0.0.%04llx\n", "0.0.%llx\n",
+		   dump_block_fcp->ipl_info.fcp.devno);
+
+static struct attribute *dump_fcp_attrs[] = {
+	&sys_dump_fcp_device_attr.attr,
+	&sys_dump_fcp_wwpn_attr.attr,
+	&sys_dump_fcp_lun_attr.attr,
+	&sys_dump_fcp_bootprog_attr.attr,
+	&sys_dump_fcp_br_lba_attr.attr,
+	NULL,
+};
+
+static struct attribute_group dump_fcp_attr_group = {
+	.name  = IPL_FCP_STR,
+	.attrs = dump_fcp_attrs,
+};
+
+/* CCW dump device attributes */
+
+DEFINE_IPL_ATTR_RW(dump_ccw, device, "0.0.%04llx\n", "0.0.%llx\n",
+		   dump_block_ccw->ipl_info.ccw.devno);
+
+static struct attribute *dump_ccw_attrs[] = {
+	&sys_dump_ccw_device_attr.attr,
+	NULL,
+};
+
+static struct attribute_group dump_ccw_attr_group = {
+	.name  = IPL_CCW_STR,
+	.attrs = dump_ccw_attrs,
+};
+
+/* dump type */
+
+static int dump_set_type(enum dump_type type)
+{
+	if (!(dump_capabilities & type))
+		return -EINVAL;
+	switch (type) {
+	case DUMP_TYPE_CCW:
+		if (MACHINE_IS_VM)
+			dump_method = DUMP_METHOD_CCW_VM;
+		else
+			dump_method = DUMP_METHOD_CCW_CIO;
+		break;
+	case DUMP_TYPE_FCP:
+		dump_method = DUMP_METHOD_FCP_DIAG;
+		break;
+	default:
+		dump_method = DUMP_METHOD_NONE;
+	}
+	dump_type = type;
+	return 0;
+}
+
+static ssize_t dump_type_show(struct kobject *kobj,
+			      struct kobj_attribute *attr, char *page)
+{
+	return sprintf(page, "%s\n", dump_type_str(dump_type));
+}
+
+static ssize_t dump_type_store(struct kobject *kobj,
+			       struct kobj_attribute *attr,
+			       const char *buf, size_t len)
+{
+	int rc = -EINVAL;
+
+	if (strncmp(buf, DUMP_NONE_STR, strlen(DUMP_NONE_STR)) == 0)
+		rc = dump_set_type(DUMP_TYPE_NONE);
+	else if (strncmp(buf, DUMP_CCW_STR, strlen(DUMP_CCW_STR)) == 0)
+		rc = dump_set_type(DUMP_TYPE_CCW);
+	else if (strncmp(buf, DUMP_FCP_STR, strlen(DUMP_FCP_STR)) == 0)
+		rc = dump_set_type(DUMP_TYPE_FCP);
+	return (rc != 0) ? rc : len;
+}
+
+static struct kobj_attribute dump_type_attr =
+	__ATTR(dump_type, 0644, dump_type_show, dump_type_store);
+
+static struct kset *dump_kset;
+
+static void dump_run(struct shutdown_trigger *trigger)
+{
+	struct ccw_dev_id devid;
+	static char buf[100];
+
+	switch (dump_method) {
+	case DUMP_METHOD_CCW_CIO:
+		smp_send_stop();
+		devid.devno = dump_block_ccw->ipl_info.ccw.devno;
+		devid.ssid  = 0;
+		reipl_ccw_dev(&devid);
+		break;
+	case DUMP_METHOD_CCW_VM:
+		smp_send_stop();
+		sprintf(buf, "STORE STATUS");
+		__cpcmd(buf, NULL, 0, NULL);
+		sprintf(buf, "IPL %X", dump_block_ccw->ipl_info.ccw.devno);
+		__cpcmd(buf, NULL, 0, NULL);
+		break;
+	case DUMP_METHOD_CCW_DIAG:
+		diag308(DIAG308_SET, dump_block_ccw);
+		diag308(DIAG308_DUMP, NULL);
+		break;
+	case DUMP_METHOD_FCP_DIAG:
+		diag308(DIAG308_SET, dump_block_fcp);
+		diag308(DIAG308_DUMP, NULL);
+		break;
+	case DUMP_METHOD_NONE:
+	default:
+		return;
+	}
+	printk(KERN_EMERG "Dump failed!\n");
+}
+
 static int __init dump_ccw_init(void)
 {
 	int rc;
@@ -1042,31 +996,14 @@ static int __init dump_fcp_init(void)
 	return 0;
 }
 
-#define SHUTDOWN_ON_PANIC_PRIO 0
-
-static int shutdown_on_panic_notify(struct notifier_block *self,
-				    unsigned long event, void *data)
-{
-	if (on_panic_action == SHUTDOWN_DUMP)
-		do_dump();
-	else if (on_panic_action == SHUTDOWN_REIPL)
-		do_reipl();
-	return NOTIFY_OK;
-}
-
-static struct notifier_block shutdown_on_panic_nb = {
-	.notifier_call = shutdown_on_panic_notify,
-	.priority = SHUTDOWN_ON_PANIC_PRIO
-};
-
-static int __init dump_init(void)
+static int dump_init(void)
 {
 	int rc;
 
 	dump_kset = kset_create_and_add("dump", NULL, firmware_kobj);
 	if (!dump_kset)
 		return -ENOMEM;
-	rc = sysfs_create_file(&dump_kset->kobj, &dump_type_attr);
+	rc = sysfs_create_file(&dump_kset->kobj, &dump_type_attr.attr);
 	if (rc) {
 		kset_unregister(dump_kset);
 		return rc;
@@ -1081,47 +1018,376 @@ static int __init dump_init(void)
 	return 0;
 }
 
-static int __init shutdown_actions_init(void)
+static struct shutdown_action dump_action = {SHUTDOWN_ACTION_DUMP_STR,
+					     dump_run, dump_init};
+
+/*
+ * vmcmd shutdown action: Trigger vm command on shutdown.
+ */
+
+static char vmcmd_on_reboot[128];
+static char vmcmd_on_panic[128];
+static char vmcmd_on_halt[128];
+static char vmcmd_on_poff[128];
+
+DEFINE_IPL_ATTR_STR_RW(vmcmd, on_reboot, "%s\n", "%s\n", vmcmd_on_reboot);
+DEFINE_IPL_ATTR_STR_RW(vmcmd, on_panic, "%s\n", "%s\n", vmcmd_on_panic);
+DEFINE_IPL_ATTR_STR_RW(vmcmd, on_halt, "%s\n", "%s\n", vmcmd_on_halt);
+DEFINE_IPL_ATTR_STR_RW(vmcmd, on_poff, "%s\n", "%s\n", vmcmd_on_poff);
+
+static struct attribute *vmcmd_attrs[] = {
+	&sys_vmcmd_on_reboot_attr.attr,
+	&sys_vmcmd_on_panic_attr.attr,
+	&sys_vmcmd_on_halt_attr.attr,
+	&sys_vmcmd_on_poff_attr.attr,
+	NULL,
+};
+
+static struct attribute_group vmcmd_attr_group = {
+	.attrs = vmcmd_attrs,
+};
+
+static struct kset *vmcmd_kset;
+
+static void vmcmd_run(struct shutdown_trigger *trigger)
 {
-	int rc;
+	char *cmd, *next_cmd;
+
+	if (strcmp(trigger->name, ON_REIPL_STR) == 0)
+		cmd = vmcmd_on_reboot;
+	else if (strcmp(trigger->name, ON_PANIC_STR) == 0)
+		cmd = vmcmd_on_panic;
+	else if (strcmp(trigger->name, ON_HALT_STR) == 0)
+		cmd = vmcmd_on_halt;
+	else if (strcmp(trigger->name, ON_POFF_STR) == 0)
+		cmd = vmcmd_on_poff;
+	else
+		return;
+
+	if (strlen(cmd) == 0)
+		return;
+	do {
+		next_cmd = strchr(cmd, '\n');
+		if (next_cmd) {
+			next_cmd[0] = 0;
+			next_cmd += 1;
+		}
+		__cpcmd(cmd, NULL, 0, NULL);
+		cmd = next_cmd;
+	} while (cmd != NULL);
+}
+
+static int vmcmd_init(void)
+{
+	if (!MACHINE_IS_VM)
+		return -ENOTSUPP;
+	vmcmd_kset = kset_create_and_add("vmcmd", NULL, firmware_kobj);
+	if (!vmcmd_kset)
+		return -ENOMEM;
+	return sysfs_create_group(&vmcmd_kset->kobj, &vmcmd_attr_group);
+}
+
+static struct shutdown_action vmcmd_action = {SHUTDOWN_ACTION_VMCMD_STR,
+					      vmcmd_run, vmcmd_init};
+
+/*
+ * stop shutdown action: Stop Linux on shutdown.
+ */
+
+static void stop_run(struct shutdown_trigger *trigger)
+{
+	signal_processor(smp_processor_id(), sigp_stop_and_store_status);
+	for (;;);
+}
+
+static struct shutdown_action stop_action = {SHUTDOWN_ACTION_STOP_STR,
+					     stop_run, NULL};
+
+/* action list */
+
+static struct shutdown_action *shutdown_actions_list[] = {
+	&ipl_action, &reipl_action, &dump_action, &vmcmd_action, &stop_action};
+#define SHUTDOWN_ACTIONS_COUNT (sizeof(shutdown_actions_list) / sizeof(void *))
+
+/*
+ * Trigger section
+ */
+
+static struct kset *shutdown_actions_kset;
+
+static int set_trigger(const char *buf, struct shutdown_trigger *trigger,
+		       size_t len)
+{
+	int i;
+	for (i = 0; i < SHUTDOWN_ACTIONS_COUNT; i++) {
+		if (!shutdown_actions_list[i])
+			continue;
+		if (strncmp(buf, shutdown_actions_list[i]->name,
+			    strlen(shutdown_actions_list[i]->name)) == 0) {
+			trigger->action = shutdown_actions_list[i];
+			return len;
+		}
+	}
+	return -EINVAL;
+}
+
+/* on reipl */
+
+static struct shutdown_trigger on_reboot_trigger = {ON_REIPL_STR,
+						    &reipl_action};
+
+static ssize_t on_reboot_show(struct kobject *kobj,
+			      struct kobj_attribute *attr, char *page)
+{
+	return sprintf(page, "%s\n", on_reboot_trigger.action->name);
+}
+
+static ssize_t on_reboot_store(struct kobject *kobj,
+			       struct kobj_attribute *attr,
+			       const char *buf, size_t len)
+{
+	return set_trigger(buf, &on_reboot_trigger, len);
+}
+
+static struct kobj_attribute on_reboot_attr =
+	__ATTR(on_reboot, 0644, on_reboot_show, on_reboot_store);
+
+static void do_machine_restart(char *__unused)
+{
+	smp_send_stop();
+	on_reboot_trigger.action->fn(&on_reboot_trigger);
+	reipl_run(NULL);
+}
+void (*_machine_restart)(char *command) = do_machine_restart;
+
+/* on panic */
+
+static struct shutdown_trigger on_panic_trigger = {ON_PANIC_STR, &stop_action};
+
+static ssize_t on_panic_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *page)
+{
+	return sprintf(page, "%s\n", on_panic_trigger.action->name);
+}
+
+static ssize_t on_panic_store(struct kobject *kobj,
+			      struct kobj_attribute *attr,
+			      const char *buf, size_t len)
+{
+	return set_trigger(buf, &on_panic_trigger, len);
+}
+
+static struct kobj_attribute on_panic_attr =
+	__ATTR(on_panic, 0644, on_panic_show, on_panic_store);
+
+static void do_panic(void)
+{
+	on_panic_trigger.action->fn(&on_panic_trigger);
+	stop_run(&on_panic_trigger);
+}
+
+/* on halt */
+
+static struct shutdown_trigger on_halt_trigger = {ON_HALT_STR, &stop_action};
+
+static ssize_t on_halt_show(struct kobject *kobj,
+			    struct kobj_attribute *attr, char *page)
+{
+	return sprintf(page, "%s\n", on_halt_trigger.action->name);
+}
+
+static ssize_t on_halt_store(struct kobject *kobj,
+			     struct kobj_attribute *attr,
+			     const char *buf, size_t len)
+{
+	return set_trigger(buf, &on_halt_trigger, len);
+}
+
+static struct kobj_attribute on_halt_attr =
+	__ATTR(on_halt, 0644, on_halt_show, on_halt_store);
 
+
+static void do_machine_halt(void)
+{
+	smp_send_stop();
+	on_halt_trigger.action->fn(&on_halt_trigger);
+	stop_run(&on_halt_trigger);
+}
+void (*_machine_halt)(void) = do_machine_halt;
+
+/* on power off */
+
+static struct shutdown_trigger on_poff_trigger = {ON_POFF_STR, &stop_action};
+
+static ssize_t on_poff_show(struct kobject *kobj,
+			    struct kobj_attribute *attr, char *page)
+{
+	return sprintf(page, "%s\n", on_poff_trigger.action->name);
+}
+
+static ssize_t on_poff_store(struct kobject *kobj,
+			     struct kobj_attribute *attr,
+			     const char *buf, size_t len)
+{
+	return set_trigger(buf, &on_poff_trigger, len);
+}
+
+static struct kobj_attribute on_poff_attr =
+	__ATTR(on_poff, 0644, on_poff_show, on_poff_store);
+
+
+static void do_machine_power_off(void)
+{
+	smp_send_stop();
+	on_poff_trigger.action->fn(&on_poff_trigger);
+	stop_run(&on_poff_trigger);
+}
+void (*_machine_power_off)(void) = do_machine_power_off;
+
+static void __init shutdown_triggers_init(void)
+{
 	shutdown_actions_kset = kset_create_and_add("shutdown_actions", NULL,
 						    firmware_kobj);
 	if (!shutdown_actions_kset)
-		return -ENOMEM;
-	rc = sysfs_create_file(&shutdown_actions_kset->kobj, &on_panic_attr);
-	if (rc) {
-		kset_unregister(shutdown_actions_kset);
-		return rc;
+		goto fail;
+	if (sysfs_create_file(&shutdown_actions_kset->kobj,
+			      &on_reboot_attr.attr))
+		goto fail;
+	if (sysfs_create_file(&shutdown_actions_kset->kobj,
+			      &on_panic_attr.attr))
+		goto fail;
+	if (sysfs_create_file(&shutdown_actions_kset->kobj,
+			      &on_halt_attr.attr))
+		goto fail;
+	if (sysfs_create_file(&shutdown_actions_kset->kobj,
+			      &on_poff_attr.attr))
+		goto fail;
+
+	return;
+fail:
+	panic("shutdown_triggers_init failed\n");
+}
+
+static void __init shutdown_actions_init(void)
+{
+	int i;
+
+	for (i = 0; i < SHUTDOWN_ACTIONS_COUNT; i++) {
+		if (!shutdown_actions_list[i]->init)
+			continue;
+		if (shutdown_actions_list[i]->init())
+			shutdown_actions_list[i] = NULL;
 	}
-	atomic_notifier_chain_register(&panic_notifier_list,
-				       &shutdown_on_panic_nb);
-	return 0;
 }
 
 static int __init s390_ipl_init(void)
 {
-	int rc;
-
-	sclp_get_ipl_info(&sclp_ipl_info);
 	reipl_probe();
-	rc = ipl_init();
-	if (rc)
-		return rc;
-	rc = reipl_init();
-	if (rc)
-		return rc;
-	rc = dump_init();
-	if (rc)
-		return rc;
-	rc = shutdown_actions_init();
-	if (rc)
-		return rc;
+	shutdown_actions_init();
+	shutdown_triggers_init();
 	return 0;
 }
 
 __initcall(s390_ipl_init);
 
+static void __init strncpy_skip_quote(char *dst, char *src, int n)
+{
+	int sx, dx;
+
+	dx = 0;
+	for (sx = 0; src[sx] != 0; sx++) {
+		if (src[sx] == '"')
+			continue;
+		dst[dx++] = src[sx];
+		if (dx >= n)
+			break;
+	}
+}
+
+static int __init vmcmd_on_reboot_setup(char *str)
+{
+	if (!MACHINE_IS_VM)
+		return 1;
+	strncpy_skip_quote(vmcmd_on_reboot, str, 127);
+	vmcmd_on_reboot[127] = 0;
+	on_reboot_trigger.action = &vmcmd_action;
+	return 1;
+}
+__setup("vmreboot=", vmcmd_on_reboot_setup);
+
+static int __init vmcmd_on_panic_setup(char *str)
+{
+	if (!MACHINE_IS_VM)
+		return 1;
+	strncpy_skip_quote(vmcmd_on_panic, str, 127);
+	vmcmd_on_panic[127] = 0;
+	on_panic_trigger.action = &vmcmd_action;
+	return 1;
+}
+__setup("vmpanic=", vmcmd_on_panic_setup);
+
+static int __init vmcmd_on_halt_setup(char *str)
+{
+	if (!MACHINE_IS_VM)
+		return 1;
+	strncpy_skip_quote(vmcmd_on_halt, str, 127);
+	vmcmd_on_halt[127] = 0;
+	on_halt_trigger.action = &vmcmd_action;
+	return 1;
+}
+__setup("vmhalt=", vmcmd_on_halt_setup);
+
+static int __init vmcmd_on_poff_setup(char *str)
+{
+	if (!MACHINE_IS_VM)
+		return 1;
+	strncpy_skip_quote(vmcmd_on_poff, str, 127);
+	vmcmd_on_poff[127] = 0;
+	on_poff_trigger.action = &vmcmd_action;
+	return 1;
+}
+__setup("vmpoff=", vmcmd_on_poff_setup);
+
+static int on_panic_notify(struct notifier_block *self,
+			   unsigned long event, void *data)
+{
+	do_panic();
+	return NOTIFY_OK;
+}
+
+static struct notifier_block on_panic_nb = {
+	.notifier_call = on_panic_notify,
+	.priority = 0,
+};
+
+void __init setup_ipl(void)
+{
+	ipl_info.type = get_ipl_type();
+	switch (ipl_info.type) {
+	case IPL_TYPE_CCW:
+		ipl_info.data.ccw.dev_id.devno = ipl_devno;
+		ipl_info.data.ccw.dev_id.ssid = 0;
+		break;
+	case IPL_TYPE_FCP:
+	case IPL_TYPE_FCP_DUMP:
+		ipl_info.data.fcp.dev_id.devno =
+			IPL_PARMBLOCK_START->ipl_info.fcp.devno;
+		ipl_info.data.fcp.dev_id.ssid = 0;
+		ipl_info.data.fcp.wwpn = IPL_PARMBLOCK_START->ipl_info.fcp.wwpn;
+		ipl_info.data.fcp.lun = IPL_PARMBLOCK_START->ipl_info.fcp.lun;
+		break;
+	case IPL_TYPE_NSS:
+		strncpy(ipl_info.data.nss.name, kernel_nss_name,
+			sizeof(ipl_info.data.nss.name));
+		break;
+	case IPL_TYPE_UNKNOWN:
+	default:
+		/* We have no info to copy */
+		break;
+	}
+	atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb);
+}
+
 void __init ipl_save_parameters(void)
 {
 	struct cio_iplinfo iplinfo;
@@ -1202,3 +1468,4 @@ void s390_reset_system(void)
 
 	do_reset_calls();
 }
+
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index cbdf3fb05e81..fcebf645618a 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -125,75 +125,6 @@ void __cpuinit cpu_init(void)
         enter_lazy_tlb(&init_mm, current);
 }
 
-/*
- * VM halt and poweroff setup routines
- */
-char vmhalt_cmd[128] = "";
-char vmpoff_cmd[128] = "";
-static char vmpanic_cmd[128] = "";
-
-static void strncpy_skip_quote(char *dst, char *src, int n)
-{
-        int sx, dx;
-
-        dx = 0;
-        for (sx = 0; src[sx] != 0; sx++) {
-                if (src[sx] == '"') continue;
-                dst[dx++] = src[sx];
-                if (dx >= n) break;
-        }
-}
-
-static int __init vmhalt_setup(char *str)
-{
-        strncpy_skip_quote(vmhalt_cmd, str, 127);
-        vmhalt_cmd[127] = 0;
-        return 1;
-}
-
-__setup("vmhalt=", vmhalt_setup);
-
-static int __init vmpoff_setup(char *str)
-{
-        strncpy_skip_quote(vmpoff_cmd, str, 127);
-        vmpoff_cmd[127] = 0;
-        return 1;
-}
-
-__setup("vmpoff=", vmpoff_setup);
-
-static int vmpanic_notify(struct notifier_block *self, unsigned long event,
-			  void *data)
-{
-	if (MACHINE_IS_VM && strlen(vmpanic_cmd) > 0)
-		cpcmd(vmpanic_cmd, NULL, 0, NULL);
-
-	return NOTIFY_OK;
-}
-
-#define PANIC_PRI_VMPANIC	0
-
-static struct notifier_block vmpanic_nb = {
-	.notifier_call = vmpanic_notify,
-	.priority = PANIC_PRI_VMPANIC
-};
-
-static int __init vmpanic_setup(char *str)
-{
-	static int register_done __initdata = 0;
-
-	strncpy_skip_quote(vmpanic_cmd, str, 127);
-	vmpanic_cmd[127] = 0;
-	if (!register_done) {
-		register_done = 1;
-		atomic_notifier_chain_register(&panic_notifier_list,
-					       &vmpanic_nb);
-	}
-	return 1;
-}
-
-__setup("vmpanic=", vmpanic_setup);
-
 /*
  * condev= and conmode= setup parameter.
  */
@@ -308,38 +239,6 @@ static void __init setup_zfcpdump(unsigned int console_devno)
 static inline void setup_zfcpdump(unsigned int console_devno) {}
 #endif /* CONFIG_ZFCPDUMP */
 
-#ifdef CONFIG_SMP
-void (*_machine_restart)(char *command) = machine_restart_smp;
-void (*_machine_halt)(void) = machine_halt_smp;
-void (*_machine_power_off)(void) = machine_power_off_smp;
-#else
-/*
- * Reboot, halt and power_off routines for non SMP.
- */
-static void do_machine_restart_nonsmp(char * __unused)
-{
-	do_reipl();
-}
-
-static void do_machine_halt_nonsmp(void)
-{
-        if (MACHINE_IS_VM && strlen(vmhalt_cmd) > 0)
-		__cpcmd(vmhalt_cmd, NULL, 0, NULL);
-        signal_processor(smp_processor_id(), sigp_stop_and_store_status);
-}
-
-static void do_machine_power_off_nonsmp(void)
-{
-        if (MACHINE_IS_VM && strlen(vmpoff_cmd) > 0)
-		__cpcmd(vmpoff_cmd, NULL, 0, NULL);
-        signal_processor(smp_processor_id(), sigp_stop_and_store_status);
-}
-
-void (*_machine_restart)(char *command) = do_machine_restart_nonsmp;
-void (*_machine_halt)(void) = do_machine_halt_nonsmp;
-void (*_machine_power_off)(void) = do_machine_power_off_nonsmp;
-#endif
-
  /*
  * Reboot, halt and power_off stubs. They just call _machine_restart,
  * _machine_halt or _machine_power_off. 
@@ -913,7 +812,7 @@ setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
-	setup_ipl_info();
+	setup_ipl();
 	setup_memory_end();
 	setup_addressing_mode();
 	setup_memory();
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 320e4e97bf52..d300a7fdf711 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -233,33 +233,6 @@ void smp_send_stop(void)
 	}
 }
 
-/*
- * Reboot, halt and power_off routines for SMP.
- */
-void machine_restart_smp(char *__unused)
-{
-	smp_send_stop();
-	do_reipl();
-}
-
-void machine_halt_smp(void)
-{
-	smp_send_stop();
-	if (MACHINE_IS_VM && strlen(vmhalt_cmd) > 0)
-		__cpcmd(vmhalt_cmd, NULL, 0, NULL);
-	signal_processor(smp_processor_id(), sigp_stop_and_store_status);
-	for (;;);
-}
-
-void machine_power_off_smp(void)
-{
-	smp_send_stop();
-	if (MACHINE_IS_VM && strlen(vmpoff_cmd) > 0)
-		__cpcmd(vmpoff_cmd, NULL, 0, NULL);
-	signal_processor(smp_processor_id(), sigp_stop_and_store_status);
-	for (;;);
-}
-
 /*
  * This is the main routine where commands issued by other
  * cpus are handled.
diff --git a/include/asm-s390/ipl.h b/include/asm-s390/ipl.h
index 2c40fd3a137f..d0dcc9c5f7c3 100644
--- a/include/asm-s390/ipl.h
+++ b/include/asm-s390/ipl.h
@@ -83,6 +83,8 @@ extern u32 dump_prefix_page;
 extern unsigned int zfcpdump_prefix_array[];
 
 extern void do_reipl(void);
+extern void do_halt(void);
+extern void do_poff(void);
 extern void ipl_save_parameters(void);
 
 enum {
@@ -118,7 +120,7 @@ struct ipl_info
 };
 
 extern struct ipl_info ipl_info;
-extern void setup_ipl_info(void);
+extern void setup_ipl(void);
 
 /*
  * DIAG 308 support
-- 
cgit v1.2.3


From 48657d223d403af676696d313b421368f5e2208a Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:17 +0100
Subject: [S390] Use diag308 subcodes 3 and 6 for reboot and dump when
 possible.

This patch fixes a problem with the following scenario:
 1. Linux booted from DASD "A"
 2. Reboot from DASD "B" using "/sys/firmware/reipl/ccw/device"
 3. Reboot DASD "B"
Without this patch in step 3 on newer s390 systems under LPAR instead of
DASD "B", DASD "A" will be booted. The reason is that in step 2 we use CCW
reipl and in step 3 we use DIAG308 (subcode 3) reipl. DIAG308 does not
notice the CCW reipl and still thinks that it has to reboot DASD "A".
Before applying this fix, ensure to have MCF RJ9967101E or z9 GA3 base driver
installed.

Signed-off-by: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ipl.c | 14 ++++++++------
 include/asm-s390/ipl.h |  4 ++++
 2 files changed, 12 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 14bdde9def40..db28cca81fef 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -595,7 +595,9 @@ static int reipl_set_type(enum ipl_type type)
 
 	switch(type) {
 	case IPL_TYPE_CCW:
-		if (MACHINE_IS_VM)
+		if (diag308_set_works)
+			reipl_method = REIPL_METHOD_CCW_DIAG;
+		else if (MACHINE_IS_VM)
 			reipl_method = REIPL_METHOD_CCW_VM;
 		else
 			reipl_method = REIPL_METHOD_CCW_CIO;
@@ -659,8 +661,6 @@ void reipl_run(struct shutdown_trigger *trigger)
 	switch (reipl_method) {
 	case REIPL_METHOD_CCW_CIO:
 		devid.devno = reipl_block_ccw->ipl_info.ccw.devno;
-		if (ipl_info.type == IPL_TYPE_CCW && devid.devno == ipl_devno)
-			diag308(DIAG308_IPL, NULL);
 		devid.ssid  = 0;
 		reipl_ccw_dev(&devid);
 		break;
@@ -745,6 +745,7 @@ static int __init reipl_ccw_init(void)
 	reipl_block_ccw->hdr.version = IPL_PARM_BLOCK_VERSION;
 	reipl_block_ccw->hdr.blk0_len = IPL_PARM_BLK0_CCW_LEN;
 	reipl_block_ccw->hdr.pbt = DIAG308_IPL_TYPE_CCW;
+	reipl_block_ccw->hdr.flags = DIAG308_FLAGS_LP_VALID;
 	/* check if read scp info worked and set loadparm */
 	if (sclp_ipl_info.is_valid)
 		memcpy(reipl_block_ccw->ipl_info.ccw.load_param,
@@ -753,8 +754,7 @@ static int __init reipl_ccw_init(void)
 		/* read scp info failed: set empty loadparm (EBCDIC blanks) */
 		memset(reipl_block_ccw->ipl_info.ccw.load_param, 0x40,
 		       LOADPARM_LEN);
-	/* FIXME: check for diag308_set_works when enabling diag ccw reipl */
-	if (!MACHINE_IS_VM)
+	if (!MACHINE_IS_VM && !diag308_set_works)
 		sys_reipl_ccw_loadparm_attr.attr.mode = S_IRUGO;
 	if (ipl_info.type == IPL_TYPE_CCW)
 		reipl_block_ccw->ipl_info.ccw.devno = ipl_devno;
@@ -876,7 +876,9 @@ static int dump_set_type(enum dump_type type)
 		return -EINVAL;
 	switch (type) {
 	case DUMP_TYPE_CCW:
-		if (MACHINE_IS_VM)
+		if (diag308_set_works)
+			dump_method = DUMP_METHOD_CCW_DIAG;
+		else if (MACHINE_IS_VM)
 			dump_method = DUMP_METHOD_CCW_VM;
 		else
 			dump_method = DUMP_METHOD_CCW_CIO;
diff --git a/include/asm-s390/ipl.h b/include/asm-s390/ipl.h
index d0dcc9c5f7c3..c1b2e50392bb 100644
--- a/include/asm-s390/ipl.h
+++ b/include/asm-s390/ipl.h
@@ -143,6 +143,10 @@ enum diag308_opt {
 	DIAG308_IPL_OPT_DUMP	= 0x20,
 };
 
+enum diag308_flags {
+	DIAG308_FLAGS_LP_VALID	= 0x80,
+};
+
 enum diag308_rc {
 	DIAG308_RC_OK	= 1,
 };
-- 
cgit v1.2.3


From 5800266a7811afeba205e771dcad79023331011b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sat, 26 Jan 2008 14:11:19 +0100
Subject: [S390] include/asm-s390/: Spelling fixes

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/cio.h    | 4 ++--
 include/asm-s390/dasd.h   | 2 +-
 include/asm-s390/qdio.h   | 2 +-
 include/asm-s390/zcrypt.h | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h
index 2f08c16e44ad..123b557c3ff4 100644
--- a/include/asm-s390/cio.h
+++ b/include/asm-s390/cio.h
@@ -24,8 +24,8 @@
  * @fmt: format
  * @pfch: prefetch
  * @isic: initial-status interruption control
- * @alcc: adress-limit checking control
- * @ssi: supress-suspended interruption
+ * @alcc: address-limit checking control
+ * @ssi: suppress-suspended interruption
  * @zcc: zero condition code
  * @ectl: extended control
  * @pno: path not operational
diff --git a/include/asm-s390/dasd.h b/include/asm-s390/dasd.h
index 604f68fa6f56..3f002e13d024 100644
--- a/include/asm-s390/dasd.h
+++ b/include/asm-s390/dasd.h
@@ -105,7 +105,7 @@ typedef struct dasd_information_t {
 } dasd_information_t;
 
 /*
- * Read Subsystem Data - Perfomance Statistics
+ * Read Subsystem Data - Performance Statistics
  */ 
 typedef struct dasd_rssd_perf_stats_t {
 	unsigned char  invalid:1;
diff --git a/include/asm-s390/qdio.h b/include/asm-s390/qdio.h
index 74db1dc10a7d..4b8ff55f680e 100644
--- a/include/asm-s390/qdio.h
+++ b/include/asm-s390/qdio.h
@@ -184,7 +184,7 @@ struct qdr {
 #endif /* QDIO_32_BIT */
 	unsigned long qiba;             /* queue-information-block address */
 	unsigned int  res8;             /* reserved */
-	unsigned int  qkey    :  4;     /* queue-informatio-block key */
+	unsigned int  qkey    :  4;	/* queue-information-block key */
 	unsigned int  res9    : 28;     /* reserved */
 /*	union _qd {*/ /* why this? */
 		struct qdesfmt0 qdf0[126];
diff --git a/include/asm-s390/zcrypt.h b/include/asm-s390/zcrypt.h
index a5dada617751..f228f1b86877 100644
--- a/include/asm-s390/zcrypt.h
+++ b/include/asm-s390/zcrypt.h
@@ -117,7 +117,7 @@ struct CPRBX {
 	unsigned char	padx004[16 - sizeof (char *)];
 	unsigned char *	req_extb;	/* request extension block 'addr'*/
 	unsigned char	padx005[16 - sizeof (char *)];
-	unsigned char *	rpl_extb;	/* reply extension block 'addres'*/
+	unsigned char *	rpl_extb;	/* reply extension block 'address'*/
 	unsigned short	ccp_rtcode;	/* server return code		 */
 	unsigned short	ccp_rscode;	/* server reason code		 */
 	unsigned int	mac_data_len;	/* Mac Data Length		 */
-- 
cgit v1.2.3


From 0ac30be461084f30ad6e22c6b91347e880ed41aa Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Sat, 26 Jan 2008 14:11:22 +0100
Subject: [S390] single-step cleanup

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ptrace.c | 15 ++++++---------
 arch/s390/kernel/signal.c | 20 +++++++++++++-------
 include/asm-s390/ptrace.h |  8 ++++++++
 3 files changed, 27 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 1d81bf9488ae..6e036bae9875 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -86,13 +86,13 @@ FixPerRegisters(struct task_struct *task)
 		per_info->control_regs.bits.storage_alt_space_ctl = 0;
 }
 
-static void set_single_step(struct task_struct *task)
+void user_enable_single_step(struct task_struct *task)
 {
 	task->thread.per_info.single_step = 1;
 	FixPerRegisters(task);
 }
 
-static void clear_single_step(struct task_struct *task)
+void user_disable_single_step(struct task_struct *task)
 {
 	task->thread.per_info.single_step = 0;
 	FixPerRegisters(task);
@@ -107,7 +107,7 @@ void
 ptrace_disable(struct task_struct *child)
 {
 	/* make sure the single step bit is not set. */
-	clear_single_step(child);
+	user_disable_single_step(child);
 }
 
 #ifndef CONFIG_64BIT
@@ -651,7 +651,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data)
 			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 		child->exit_code = data;
 		/* make sure the single step bit is not set. */
-		clear_single_step(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		return 0;
 
@@ -665,7 +665,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data)
 			return 0;
 		child->exit_code = SIGKILL;
 		/* make sure the single step bit is not set. */
-		clear_single_step(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		return 0;
 
@@ -675,10 +675,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data)
 			return -EIO;
 		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 		child->exit_code = data;
-		if (data)
-			set_tsk_thread_flag(child, TIF_SINGLE_STEP);
-		else
-			set_single_step(child);
+		user_enable_single_step(child);
 		/* give it a chance to run. */
 		wake_up_process(child);
 		return 0;
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index d264671c1b71..4449bf32cbf1 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -471,6 +471,7 @@ void do_signal(struct pt_regs *regs)
 
 	if (signr > 0) {
 		/* Whee!  Actually deliver the signal.  */
+		int ret;
 #ifdef CONFIG_COMPAT
 		if (test_thread_flag(TIF_31BIT)) {
 			extern int handle_signal32(unsigned long sig,
@@ -478,15 +479,12 @@ void do_signal(struct pt_regs *regs)
 						   siginfo_t *info,
 						   sigset_t *oldset,
 						   struct pt_regs *regs);
-			if (handle_signal32(
-				    signr, &ka, &info, oldset, regs) == 0) {
-				if (test_thread_flag(TIF_RESTORE_SIGMASK))
-					clear_thread_flag(TIF_RESTORE_SIGMASK);
-			}
-			return;
+			ret = handle_signal32(signr, &ka, &info, oldset, regs);
 	        }
+		else
 #endif
-		if (handle_signal(signr, &ka, &info, oldset, regs) == 0) {
+			ret = handle_signal(signr, &ka, &info, oldset, regs);
+		if (!ret) {
 			/*
 			 * A signal was successfully delivered; the saved
 			 * sigmask will have been stored in the signal frame,
@@ -495,6 +493,14 @@ void do_signal(struct pt_regs *regs)
 			 */
 			if (test_thread_flag(TIF_RESTORE_SIGMASK))
 				clear_thread_flag(TIF_RESTORE_SIGMASK);
+
+			/*
+			 * If we would have taken a single-step trap
+			 * for a normal instruction, act like we took
+			 * one for the handler setup.
+			 */
+			if (current->thread.per_info.single_step)
+				set_thread_flag(TIF_SINGLE_STEP);
 		}
 		return;
 	}
diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h
index 332ee73688fc..61f6952f2e35 100644
--- a/include/asm-s390/ptrace.h
+++ b/include/asm-s390/ptrace.h
@@ -465,6 +465,14 @@ struct user_regs_struct
 #ifdef __KERNEL__
 #define __ARCH_SYS_PTRACE	1
 
+/*
+ * These are defined as per linux/ptrace.h, which see.
+ */
+#define arch_has_single_step()	(1)
+struct task_struct;
+extern void user_enable_single_step(struct task_struct *);
+extern void user_disable_single_step(struct task_struct *);
+
 #define user_mode(regs) (((regs)->psw.mask & PSW_MASK_PSTATE) != 0)
 #define instruction_pointer(regs) ((regs)->psw.addr & PSW_ADDR_INSN)
 #define regs_return_value(regs)((regs)->gprs[2])
-- 
cgit v1.2.3


From 028fed8233b669e03def295a4e632b1b933720f9 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:25 +0100
Subject: [S390] Unused field / extern declaration in processor.h

Remove extern declaration of non-existent last_task_used_math and
remove unused field error_code from the thread_struct.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/processor.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index 21d40a19355e..c86b982aef5a 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -59,9 +59,6 @@ extern void s390_adjust_jiffies(void);
 extern void print_cpu_info(struct cpuinfo_S390 *);
 extern int get_cpu_capability(unsigned int *);
 
-/* Lazy FPU handling on uni-processor */
-extern struct task_struct *last_task_used_math;
-
 /*
  * User space process size: 2GB for 31 bit, 4TB for 64 bit.
  */
@@ -95,7 +92,6 @@ struct thread_struct {
         unsigned long ksp;              /* kernel stack pointer             */
 	mm_segment_t mm_segment;
         unsigned long prot_addr;        /* address of protection-excep.     */
-        unsigned int error_code;        /* error-code of last prog-excep.   */
         unsigned int trap_no;
         per_struct per_info;
 	/* Used to give failing instruction back to user for ieee exceptions */
-- 
cgit v1.2.3


From dab5209cd878c146d9f6923f061d1c2725ff210f Mon Sep 17 00:00:00 2001
From: Carsten Otte <cotte@de.ibm.com>
Date: Sat, 26 Jan 2008 14:11:27 +0100
Subject: [S390] add smp_call_function_mask

This patch adds the s390 variant for smp_call_function_mask(). The
implementation is pretty straight forward using the wrapper
__smp_call_function_map() which already takes a cpumask_t argument.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 27 +++++++++++++++++++++++++++
 include/asm-s390/smp.h |  2 ++
 2 files changed, 29 insertions(+)

(limited to 'include')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 040406dbe9aa..f66db7185d85 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -210,6 +210,33 @@ int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
 }
 EXPORT_SYMBOL(smp_call_function_single);
 
+/**
+ * smp_call_function_mask(): Run a function on a set of other CPUs.
+ * @mask: The set of cpus to run on.  Must not include the current cpu.
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @wait: If true, wait (atomically) until function has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code.
+ *
+ * If @wait is true, then returns once @func has returned; otherwise
+ * it returns just before the target cpu calls @func.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler.
+ */
+int
+smp_call_function_mask(cpumask_t mask,
+			void (*func)(void *), void *info,
+			int wait)
+{
+	preempt_disable();
+	__smp_call_function_map(func, info, 0, wait, mask);
+	preempt_enable();
+	return 0;
+}
+EXPORT_SYMBOL(smp_call_function_mask);
+
 void smp_send_stop(void)
 {
 	int cpu, rc;
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 218454b91862..c7b74326a527 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -90,6 +90,8 @@ extern void __cpu_die (unsigned int cpu);
 extern void cpu_die (void) __attribute__ ((noreturn));
 extern int __cpu_up (unsigned int cpu);
 
+extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
+	void *info, int wait);
 #endif
 
 #ifndef CONFIG_SMP
-- 
cgit v1.2.3


From 894cdde26b538c77b9943bc72f0570abf6e58e37 Mon Sep 17 00:00:00 2001
From: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Date: Sat, 26 Jan 2008 14:11:28 +0100
Subject: [S390] do local_irq_restore while spinning in spin_lock_irqsave.

In s390's spin_lock_irqsave, interrupts remain disabled while
spinning. In other architectures like x86 and powerpc, interrupts are
re-enabled while spinning if IRQ is not masked before spin_lock_irqsave
is called.

The following patch re-enables interrupts through local_irq_restore
while spinning for a lock acquisition.
This can improve system response.

[heiko.carstens@de.ibm.com: removed saving of pc]

Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/lib/spinlock.c    | 23 +++++++++++++++++++++++
 include/asm-s390/spinlock.h | 13 ++++++++++++-
 2 files changed, 35 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c
index 59c56c3d72d0..e41f4008afc5 100644
--- a/arch/s390/lib/spinlock.c
+++ b/arch/s390/lib/spinlock.c
@@ -59,6 +59,29 @@ void _raw_spin_lock_wait(raw_spinlock_t *lp)
 }
 EXPORT_SYMBOL(_raw_spin_lock_wait);
 
+void _raw_spin_lock_wait_flags(raw_spinlock_t *lp, unsigned long flags)
+{
+	int count = spin_retry;
+	unsigned int cpu = ~smp_processor_id();
+
+	local_irq_restore(flags);
+	while (1) {
+		if (count-- <= 0) {
+			unsigned int owner = lp->owner_cpu;
+			if (owner != 0)
+				_raw_yield_cpu(~owner);
+			count = spin_retry;
+		}
+		if (__raw_spin_is_locked(lp))
+			continue;
+		local_irq_disable();
+		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
+			return;
+		local_irq_restore(flags);
+	}
+}
+EXPORT_SYMBOL(_raw_spin_lock_wait_flags);
+
 int _raw_spin_trylock_retry(raw_spinlock_t *lp)
 {
 	unsigned int cpu = ~smp_processor_id();
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index c1d6e7e304e2..df84ae96915f 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -53,12 +53,12 @@ _raw_compare_and_swap(volatile unsigned int *lock,
  */
 
 #define __raw_spin_is_locked(x) ((x)->owner_cpu != 0)
-#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 #define __raw_spin_unlock_wait(lock) \
 	do { while (__raw_spin_is_locked(lock)) \
 		 _raw_spin_relax(lock); } while (0)
 
 extern void _raw_spin_lock_wait(raw_spinlock_t *);
+extern void _raw_spin_lock_wait_flags(raw_spinlock_t *, unsigned long flags);
 extern int _raw_spin_trylock_retry(raw_spinlock_t *);
 extern void _raw_spin_relax(raw_spinlock_t *lock);
 
@@ -72,6 +72,17 @@ static inline void __raw_spin_lock(raw_spinlock_t *lp)
 	_raw_spin_lock_wait(lp);
 }
 
+static inline void __raw_spin_lock_flags(raw_spinlock_t *lp,
+					 unsigned long flags)
+{
+	int old;
+
+	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
+	if (likely(old == 0))
+		return;
+	_raw_spin_lock_wait_flags(lp, flags);
+}
+
 static inline int __raw_spin_trylock(raw_spinlock_t *lp)
 {
 	int old;
-- 
cgit v1.2.3


From 4e9f9fd5148004b983b29e15de66918e71da56c0 Mon Sep 17 00:00:00 2001
From: Herbert Valerio Riedel <hvr@gnu.org>
Date: Mon, 26 Nov 2007 18:41:02 +0100
Subject: [ARM] 4668/1: ep93xx: implement new GPIO API

Implement new GPIO API for ep93xx platform as defined in Documentation/gpio.txt
and provide transitional __deprecated wrappers for the previous gpio_line_*
functions.

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                   |  1 +
 arch/arm/mach-ep93xx/core.c        | 63 +++++++++++++++++++++----------
 include/asm-arm/arch-ep93xx/gpio.h | 76 +++++++++++++++++++++++++++++++++-----
 3 files changed, 111 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index a04f507e7f2c..97e58a1863c2 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -217,6 +217,7 @@ config ARCH_EP93XX
 	bool "EP93xx-based"
 	select ARM_AMBA
 	select ARM_VIC
+	select GENERIC_GPIO
 	help
 	  This enables support for the Cirrus EP93xx series of CPUs.
 
diff --git a/arch/arm/mach-ep93xx/core.c b/arch/arm/mach-ep93xx/core.c
index 70b2c7801110..8a7340661377 100644
--- a/arch/arm/mach-ep93xx/core.c
+++ b/arch/arm/mach-ep93xx/core.c
@@ -188,7 +188,7 @@ static unsigned char data_direction_register_offset[8] = {
 	0x10, 0x14, 0x18, 0x1c, 0x24, 0x34, 0x3c, 0x44,
 };
 
-void gpio_line_config(int line, int direction)
+static void ep93xx_gpio_set_direction(unsigned line, int direction)
 {
 	unsigned int data_direction_register;
 	unsigned long flags;
@@ -219,39 +219,64 @@ void gpio_line_config(int line, int direction)
 	}
 	local_irq_restore(flags);
 }
+
+void __deprecated gpio_line_config(int line, int direction)
+{
+	ep93xx_gpio_set_direction(line, direction);
+}
 EXPORT_SYMBOL(gpio_line_config);
 
-int gpio_line_get(int line)
+int gpio_direction_input(unsigned gpio)
+{
+	if (gpio > EP93XX_GPIO_LINE_H(7))
+		return -EINVAL;
+
+	ep93xx_gpio_set_direction(gpio, GPIO_IN);
+
+	return 0;
+}
+EXPORT_SYMBOL(gpio_direction_input);
+
+int gpio_direction_output(unsigned gpio, int value)
+{
+	if (gpio > EP93XX_GPIO_LINE_H(7))
+		return -EINVAL;
+
+	gpio_set_value(gpio, value);
+	ep93xx_gpio_set_direction(gpio, GPIO_OUT);
+
+	return 0;
+}
+EXPORT_SYMBOL(gpio_direction_output);
+
+int gpio_get_value(unsigned gpio)
 {
 	unsigned int data_register;
 
-	data_register = EP93XX_GPIO_REG(data_register_offset[line >> 3]);
+	data_register = EP93XX_GPIO_REG(data_register_offset[gpio >> 3]);
 
-	return !!(__raw_readb(data_register) & (1 << (line & 7)));
+	return !!(__raw_readb(data_register) & (1 << (gpio & 7)));
 }
-EXPORT_SYMBOL(gpio_line_get);
+EXPORT_SYMBOL(gpio_get_value);
 
-void gpio_line_set(int line, int value)
+void gpio_set_value(unsigned gpio, int value)
 {
 	unsigned int data_register;
 	unsigned long flags;
 	unsigned char v;
 
-	data_register = EP93XX_GPIO_REG(data_register_offset[line >> 3]);
+	data_register = EP93XX_GPIO_REG(data_register_offset[gpio >> 3]);
 
 	local_irq_save(flags);
-	if (value == EP93XX_GPIO_HIGH) {
-		v = __raw_readb(data_register);
-		v |= 1 << (line & 7);
-		__raw_writeb(v, data_register);
-	} else if (value == EP93XX_GPIO_LOW) {
-		v = __raw_readb(data_register);
-		v &= ~(1 << (line & 7));
-		__raw_writeb(v, data_register);
-	}
+	v = __raw_readb(data_register);
+	if (value)
+		v |= 1 << (gpio & 7);
+	else
+		v &= ~(1 << (gpio & 7));
+	__raw_writeb(v, data_register);
 	local_irq_restore(flags);
 }
-EXPORT_SYMBOL(gpio_line_set);
+EXPORT_SYMBOL(gpio_set_value);
 
 
 /*************************************************************************
@@ -334,9 +359,9 @@ static int ep93xx_gpio_irq_type(unsigned int irq, unsigned int type)
 
 	line = irq - IRQ_EP93XX_GPIO(0);
 	if (line >= 0 && line < 16) {
-		gpio_line_config(line, GPIO_IN);
+		ep93xx_gpio_set_direction(line, GPIO_IN);
 	} else {
-		gpio_line_config(EP93XX_GPIO_LINE_F(line-16), GPIO_IN);
+		ep93xx_gpio_set_direction(EP93XX_GPIO_LINE_F(line-16), GPIO_IN);
 	}
 
 	port = line >> 3;
diff --git a/include/asm-arm/arch-ep93xx/gpio.h b/include/asm-arm/arch-ep93xx/gpio.h
index 1ee14a14cba0..fc1e57db5fac 100644
--- a/include/asm-arm/arch-ep93xx/gpio.h
+++ b/include/asm-arm/arch-ep93xx/gpio.h
@@ -5,16 +5,6 @@
 #ifndef __ASM_ARCH_GPIO_H
 #define __ASM_ARCH_GPIO_H
 
-#define GPIO_IN				0
-#define GPIO_OUT			1
-
-#define EP93XX_GPIO_LOW			0
-#define EP93XX_GPIO_HIGH		1
-
-extern void gpio_line_config(int line, int direction);
-extern int  gpio_line_get(int line);
-extern void gpio_line_set(int line, int value);
-
 /* GPIO port A.  */
 #define EP93XX_GPIO_LINE_A(x)		((x) + 0)
 #define EP93XX_GPIO_LINE_EGPIO0		EP93XX_GPIO_LINE_A(0)
@@ -103,5 +93,71 @@ extern void gpio_line_set(int line, int value);
 #define EP93XX_GPIO_LINE_DD6		EP93XX_GPIO_LINE_H(6)
 #define EP93XX_GPIO_LINE_DD7		EP93XX_GPIO_LINE_H(7)
 
+/* new generic GPIO API - see Documentation/gpio.txt */
+
+static inline int gpio_request(unsigned gpio, const char *label)
+{
+	if (gpio > EP93XX_GPIO_LINE_H(7))
+		return -EINVAL;
+	return 0;
+}
+
+static inline void gpio_free(unsigned gpio)
+{
+}
+
+int gpio_direction_input(unsigned gpio);
+int gpio_direction_output(unsigned gpio, int value);
+int gpio_get_value(unsigned gpio);
+void gpio_set_value(unsigned gpio, int value);
+
+#include <asm-generic/gpio.h> /* cansleep wrappers */
+
+/*
+ * Map GPIO A0..A7  (0..7)  to irq 64..71,
+ *          B0..B7  (7..15) to irq 72..79, and
+ *          F0..F7 (40..47) to irq 80..87.
+ */
+
+static inline int gpio_to_irq(unsigned gpio)
+{
+	if (gpio <= EP93XX_GPIO_LINE_EGPIO15)
+		return 64 + gpio;
+
+	if (gpio >= EP93XX_GPIO_LINE_F(0) && gpio <= EP93XX_GPIO_LINE_F(7))
+		return 80 + (gpio - EP93XX_GPIO_LINE_F(0));
+
+	return -EINVAL;
+}
+
+static inline int irq_to_gpio(unsigned irq)
+{
+	if (irq >= 64 && irq <= 79)
+		return irq - 64;
+
+	if (irq >= 80 && irq <= 87)
+		return (irq - 80) + EP93XX_GPIO_LINE_F(0);
+
+	return -EINVAL;
+}
+
+/* obsolete specific GPIO API */
+#define GPIO_IN				0
+#define GPIO_OUT			1
+
+#define EP93XX_GPIO_LOW			0
+#define EP93XX_GPIO_HIGH		1
+
+void __deprecated gpio_line_config(int line, int direction);
+
+static inline int  __deprecated gpio_line_get(int line)
+{
+	return gpio_get_value(line);
+}
+
+static inline void __deprecated gpio_line_set(int line, int value)
+{
+	gpio_set_value(line, value);
+}
 
 #endif
-- 
cgit v1.2.3


From 7ca722533979d47563e75a40c86c405153818b83 Mon Sep 17 00:00:00 2001
From: Herbert Valerio Riedel <hvr@gnu.org>
Date: Mon, 26 Nov 2007 18:45:59 +0100
Subject: [ARM] 4669/1: ep93xx: simplify GPIO code and cleanups

This patch renumbers the (virtual) GPIO line numbering to have all
irq-capable gpio lines <= EP93XX_GPIO_LINE_MAX_IRQ by swapping the
port f range with the port c range; This simplifies code such as

 #define IRQ_EP93XX_GPIO(x)  (64 + (((x) + (((x) >> 2) & 8)) & 0x1f))

or

 if (line >= 0 && line < 16) {
    /* Port A/B */
 } else if (line >= 40 && line < 48) {
    /* Port F */
 }

considerably; in addition to the renumbering this patch also
introduces macro constants EP93XX_GPIO_LINE_MAX_IRQ and
EP93XX_GPIO_LINE_MAX, and replaces most magic numbers by those and
invocations of gpio_to_irq()/irq_to_gpio().

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ep93xx/core.c        | 147 ++++++++++++++++++-------------------
 include/asm-arm/arch-ep93xx/gpio.h |  27 +++----
 include/asm-arm/arch-ep93xx/irqs.h |   6 --
 3 files changed, 82 insertions(+), 98 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-ep93xx/core.c b/arch/arm/mach-ep93xx/core.c
index 8a7340661377..58957d7a5763 100644
--- a/arch/arm/mach-ep93xx/core.c
+++ b/arch/arm/mach-ep93xx/core.c
@@ -157,35 +157,35 @@ static unsigned char gpio_int_enabled[3];
 static unsigned char gpio_int_type1[3];
 static unsigned char gpio_int_type2[3];
 
-static void update_gpio_int_params(int abf)
+/* Port ordering is: A B F */
+static const u8 int_type1_register_offset[3]	= { 0x90, 0xac, 0x4c };
+static const u8 int_type2_register_offset[3]	= { 0x94, 0xb0, 0x50 };
+static const u8 eoi_register_offset[3]		= { 0x98, 0xb4, 0x54 };
+static const u8 int_en_register_offset[3]	= { 0x9c, 0xb8, 0x5c };
+
+static void update_gpio_int_params(unsigned port)
 {
-	if (abf == 0) {
-		__raw_writeb(0, EP93XX_GPIO_A_INT_ENABLE);
-		__raw_writeb(gpio_int_type2[0], EP93XX_GPIO_A_INT_TYPE2);
-		__raw_writeb(gpio_int_type1[0], EP93XX_GPIO_A_INT_TYPE1);
-		__raw_writeb(gpio_int_unmasked[0] & gpio_int_enabled[0], EP93XX_GPIO_A_INT_ENABLE);
-	} else if (abf == 1) {
-		__raw_writeb(0, EP93XX_GPIO_B_INT_ENABLE);
-		__raw_writeb(gpio_int_type2[1], EP93XX_GPIO_B_INT_TYPE2);
-		__raw_writeb(gpio_int_type1[1], EP93XX_GPIO_B_INT_TYPE1);
-		__raw_writeb(gpio_int_unmasked[1] & gpio_int_enabled[1], EP93XX_GPIO_B_INT_ENABLE);
-	} else if (abf == 2) {
-		__raw_writeb(0, EP93XX_GPIO_F_INT_ENABLE);
-		__raw_writeb(gpio_int_type2[2], EP93XX_GPIO_F_INT_TYPE2);
-		__raw_writeb(gpio_int_type1[2], EP93XX_GPIO_F_INT_TYPE1);
-		__raw_writeb(gpio_int_unmasked[2] & gpio_int_enabled[2], EP93XX_GPIO_F_INT_ENABLE);
-	} else {
-		BUG();
-	}
-}
+	BUG_ON(port > 2);
 
+	__raw_writeb(0, EP93XX_GPIO_REG(int_en_register_offset[port]));
+
+	__raw_writeb(gpio_int_type2[port],
+		EP93XX_GPIO_REG(int_type2_register_offset[port]));
+
+	__raw_writeb(gpio_int_type1[port],
+		EP93XX_GPIO_REG(int_type1_register_offset[port]));
+
+	__raw_writeb(gpio_int_unmasked[port] & gpio_int_enabled[port],
+		EP93XX_GPIO_REG(int_en_register_offset[port]));
+}
 
-static unsigned char data_register_offset[8] = {
-	0x00, 0x04, 0x08, 0x0c, 0x20, 0x30, 0x38, 0x40,
+/* Port ordering is: A B F D E C G H */
+static const u8 data_register_offset[8] = {
+	0x00, 0x04, 0x30, 0x0c, 0x20, 0x08, 0x38, 0x40,
 };
 
-static unsigned char data_direction_register_offset[8] = {
-	0x10, 0x14, 0x18, 0x1c, 0x24, 0x34, 0x3c, 0x44,
+static const u8 data_direction_register_offset[8] = {
+	0x10, 0x14, 0x34, 0x1c, 0x24, 0x18, 0x3c, 0x44,
 };
 
 static void ep93xx_gpio_set_direction(unsigned line, int direction)
@@ -199,14 +199,10 @@ static void ep93xx_gpio_set_direction(unsigned line, int direction)
 
 	local_irq_save(flags);
 	if (direction == GPIO_OUT) {
-		if (line >= 0 && line < 16) {
-			/* Port A/B.  */
+		if (line >= 0 && line <= EP93XX_GPIO_LINE_MAX_IRQ) {
+			/* Port A/B/F */
 			gpio_int_unmasked[line >> 3] &= ~(1 << (line & 7));
 			update_gpio_int_params(line >> 3);
-		} else if (line >= 40 && line < 48) {
-			/* Port F.  */
-			gpio_int_unmasked[2] &= ~(1 << (line & 7));
-			update_gpio_int_params(2);
 		}
 
 		v = __raw_readb(data_direction_register);
@@ -228,7 +224,7 @@ EXPORT_SYMBOL(gpio_line_config);
 
 int gpio_direction_input(unsigned gpio)
 {
-	if (gpio > EP93XX_GPIO_LINE_H(7))
+	if (gpio > EP93XX_GPIO_LINE_MAX)
 		return -EINVAL;
 
 	ep93xx_gpio_set_direction(gpio, GPIO_IN);
@@ -239,7 +235,7 @@ EXPORT_SYMBOL(gpio_direction_input);
 
 int gpio_direction_output(unsigned gpio, int value)
 {
-	if (gpio > EP93XX_GPIO_LINE_H(7))
+	if (gpio > EP93XX_GPIO_LINE_MAX)
 		return -EINVAL;
 
 	gpio_set_value(gpio, value);
@@ -290,47 +286,50 @@ static void ep93xx_gpio_ab_irq_handler(unsigned int irq, struct irq_desc *desc)
 	status = __raw_readb(EP93XX_GPIO_A_INT_STATUS);
 	for (i = 0; i < 8; i++) {
 		if (status & (1 << i)) {
-			desc = irq_desc + IRQ_EP93XX_GPIO(0) + i;
-			desc_handle_irq(IRQ_EP93XX_GPIO(0) + i, desc);
+			int gpio_irq = gpio_to_irq(EP93XX_GPIO_LINE_A(0)) + i;
+			desc = irq_desc + gpio_irq;
+			desc_handle_irq(gpio_irq, desc);
 		}
 	}
 
 	status = __raw_readb(EP93XX_GPIO_B_INT_STATUS);
 	for (i = 0; i < 8; i++) {
 		if (status & (1 << i)) {
-			desc = irq_desc + IRQ_EP93XX_GPIO(8) + i;
-			desc_handle_irq(IRQ_EP93XX_GPIO(8) + i, desc);
+			int gpio_irq = gpio_to_irq(EP93XX_GPIO_LINE_B(0)) + i;
+			desc = irq_desc + gpio_irq;
+			desc_handle_irq(gpio_irq, desc);
 		}
 	}
 }
 
 static void ep93xx_gpio_f_irq_handler(unsigned int irq, struct irq_desc *desc)
 {
-	int gpio_irq = IRQ_EP93XX_GPIO(16) + (((irq + 1) & 7) ^ 4);
+	/*
+	 * map discontiguous hw irq range to continous sw irq range:
+	 *
+	 *  IRQ_EP93XX_GPIO{0..7}MUX -> gpio_to_irq(EP93XX_GPIO_LINE_F({0..7})
+	 */
+	int port_f_idx = ((irq + 1) & 7) ^ 4; /* {19..22,47..50} -> {0..7} */
+	int gpio_irq = gpio_to_irq(EP93XX_GPIO_LINE_F(0)) + port_f_idx;
 
 	desc_handle_irq(gpio_irq, irq_desc + gpio_irq);
 }
 
 static void ep93xx_gpio_irq_mask_ack(unsigned int irq)
 {
-	int line = irq - IRQ_EP93XX_GPIO(0);
+	int line = irq_to_gpio(irq);
 	int port = line >> 3;
+	int port_mask = 1 << (line & 7);
 
-	gpio_int_unmasked[port] &= ~(1 << (line & 7));
+	gpio_int_unmasked[port] &= ~port_mask;
 	update_gpio_int_params(port);
 
-	if (port == 0) {
-		__raw_writel(1 << (line & 7), EP93XX_GPIO_A_INT_ACK);
-	} else if (port == 1) {
-		__raw_writel(1 << (line & 7), EP93XX_GPIO_B_INT_ACK);
-	} else if (port == 2) {
-		__raw_writel(1 << (line & 7), EP93XX_GPIO_F_INT_ACK);
-	}
+	__raw_writeb(port_mask, EP93XX_GPIO_REG(eoi_register_offset[port]));
 }
 
 static void ep93xx_gpio_irq_mask(unsigned int irq)
 {
-	int line = irq - IRQ_EP93XX_GPIO(0);
+	int line = irq_to_gpio(irq);
 	int port = line >> 3;
 
 	gpio_int_unmasked[port] &= ~(1 << (line & 7));
@@ -339,7 +338,7 @@ static void ep93xx_gpio_irq_mask(unsigned int irq)
 
 static void ep93xx_gpio_irq_unmask(unsigned int irq)
 {
-	int line = irq - IRQ_EP93XX_GPIO(0);
+	int line = irq_to_gpio(irq);
 	int port = line >> 3;
 
 	gpio_int_unmasked[port] |= 1 << (line & 7);
@@ -354,37 +353,30 @@ static void ep93xx_gpio_irq_unmask(unsigned int irq)
  */
 static int ep93xx_gpio_irq_type(unsigned int irq, unsigned int type)
 {
-	int port;
-	int line;
+	const int gpio = irq_to_gpio(irq);
+	const int port = gpio >> 3;
+	const int port_mask = 1 << (gpio & 7);
 
-	line = irq - IRQ_EP93XX_GPIO(0);
-	if (line >= 0 && line < 16) {
-		ep93xx_gpio_set_direction(line, GPIO_IN);
-	} else {
-		ep93xx_gpio_set_direction(EP93XX_GPIO_LINE_F(line-16), GPIO_IN);
-	}
-
-	port = line >> 3;
-	line &= 7;
+	ep93xx_gpio_set_direction(gpio, GPIO_IN);
 
 	if (type & IRQT_RISING) {
-		gpio_int_enabled[port] |= 1 << line;
-		gpio_int_type1[port] |= 1 << line;
-		gpio_int_type2[port] |= 1 << line;
+		gpio_int_enabled[port] |= port_mask;
+		gpio_int_type1[port] |= port_mask;
+		gpio_int_type2[port] |= port_mask;
 	} else if (type & IRQT_FALLING) {
-		gpio_int_enabled[port] |= 1 << line;
-		gpio_int_type1[port] |= 1 << line;
-		gpio_int_type2[port] &= ~(1 << line);
+		gpio_int_enabled[port] |= port_mask;
+		gpio_int_type1[port] |= port_mask;
+		gpio_int_type2[port] &= ~port_mask;
 	} else if (type & IRQT_HIGH) {
-		gpio_int_enabled[port] |= 1 << line;
-		gpio_int_type1[port] &= ~(1 << line);
-		gpio_int_type2[port] |= 1 << line;
+		gpio_int_enabled[port] |= port_mask;
+		gpio_int_type1[port] &= ~port_mask;
+		gpio_int_type2[port] |= port_mask;
 	} else if (type & IRQT_LOW) {
-		gpio_int_enabled[port] |= 1 << line;
-		gpio_int_type1[port] &= ~(1 << line);
-		gpio_int_type2[port] &= ~(1 << line);
+		gpio_int_enabled[port] |= port_mask;
+		gpio_int_type1[port] &= ~port_mask;
+		gpio_int_type2[port] &= ~port_mask;
 	} else {
-		gpio_int_enabled[port] &= ~(1 << line);
+		gpio_int_enabled[port] &= ~port_mask;
 	}
 	update_gpio_int_params(port);
 
@@ -402,15 +394,16 @@ static struct irq_chip ep93xx_gpio_irq_chip = {
 
 void __init ep93xx_init_irq(void)
 {
-	int irq;
+	int gpio_irq;
 
 	vic_init((void *)EP93XX_VIC1_BASE, 0, EP93XX_VIC1_VALID_IRQ_MASK);
 	vic_init((void *)EP93XX_VIC2_BASE, 32, EP93XX_VIC2_VALID_IRQ_MASK);
 
-	for (irq = IRQ_EP93XX_GPIO(0); irq <= IRQ_EP93XX_GPIO(23); irq++) {
-		set_irq_chip(irq, &ep93xx_gpio_irq_chip);
-		set_irq_handler(irq, handle_level_irq);
-		set_irq_flags(irq, IRQF_VALID);
+	for (gpio_irq = gpio_to_irq(0);
+	     gpio_irq <= gpio_to_irq(EP93XX_GPIO_LINE_MAX_IRQ); ++gpio_irq) {
+		set_irq_chip(gpio_irq, &ep93xx_gpio_irq_chip);
+		set_irq_handler(gpio_irq, handle_level_irq);
+		set_irq_flags(gpio_irq, IRQF_VALID);
 	}
 
 	set_irq_chained_handler(IRQ_EP93XX_GPIO_AB, ep93xx_gpio_ab_irq_handler);
diff --git a/include/asm-arm/arch-ep93xx/gpio.h b/include/asm-arm/arch-ep93xx/gpio.h
index fc1e57db5fac..cebb64be7e4b 100644
--- a/include/asm-arm/arch-ep93xx/gpio.h
+++ b/include/asm-arm/arch-ep93xx/gpio.h
@@ -28,7 +28,7 @@
 #define EP93XX_GPIO_LINE_EGPIO15	EP93XX_GPIO_LINE_B(7)
 
 /* GPIO port C.  */
-#define EP93XX_GPIO_LINE_C(x)		((x) + 16)
+#define EP93XX_GPIO_LINE_C(x)		((x) + 40)
 #define EP93XX_GPIO_LINE_ROW0		EP93XX_GPIO_LINE_C(0)
 #define EP93XX_GPIO_LINE_ROW1		EP93XX_GPIO_LINE_C(1)
 #define EP93XX_GPIO_LINE_ROW2		EP93XX_GPIO_LINE_C(2)
@@ -61,7 +61,7 @@
 #define EP93XX_GPIO_LINE_IDEDA2		EP93XX_GPIO_LINE_E(7)
 
 /* GPIO port F.  */
-#define EP93XX_GPIO_LINE_F(x)		((x) + 40)
+#define EP93XX_GPIO_LINE_F(x)		((x) + 16)
 #define EP93XX_GPIO_LINE_WP		EP93XX_GPIO_LINE_F(0)
 #define EP93XX_GPIO_LINE_MCCD1		EP93XX_GPIO_LINE_F(1)
 #define EP93XX_GPIO_LINE_MCCD2		EP93XX_GPIO_LINE_F(2)
@@ -93,11 +93,17 @@
 #define EP93XX_GPIO_LINE_DD6		EP93XX_GPIO_LINE_H(6)
 #define EP93XX_GPIO_LINE_DD7		EP93XX_GPIO_LINE_H(7)
 
+/* maximum value for gpio line identifiers */
+#define EP93XX_GPIO_LINE_MAX		EP93XX_GPIO_LINE_H(7)
+
+/* maximum value for irq capable line identifiers */
+#define EP93XX_GPIO_LINE_MAX_IRQ	EP93XX_GPIO_LINE_F(7)
+
 /* new generic GPIO API - see Documentation/gpio.txt */
 
 static inline int gpio_request(unsigned gpio, const char *label)
 {
-	if (gpio > EP93XX_GPIO_LINE_H(7))
+	if (gpio > EP93XX_GPIO_LINE_MAX)
 		return -EINVAL;
 	return 0;
 }
@@ -116,29 +122,20 @@ void gpio_set_value(unsigned gpio, int value);
 /*
  * Map GPIO A0..A7  (0..7)  to irq 64..71,
  *          B0..B7  (7..15) to irq 72..79, and
- *          F0..F7 (40..47) to irq 80..87.
+ *          F0..F7 (16..24) to irq 80..87.
  */
 
 static inline int gpio_to_irq(unsigned gpio)
 {
-	if (gpio <= EP93XX_GPIO_LINE_EGPIO15)
+	if (gpio <= EP93XX_GPIO_LINE_MAX_IRQ)
 		return 64 + gpio;
 
-	if (gpio >= EP93XX_GPIO_LINE_F(0) && gpio <= EP93XX_GPIO_LINE_F(7))
-		return 80 + (gpio - EP93XX_GPIO_LINE_F(0));
-
 	return -EINVAL;
 }
 
 static inline int irq_to_gpio(unsigned irq)
 {
-	if (irq >= 64 && irq <= 79)
-		return irq - 64;
-
-	if (irq >= 80 && irq <= 87)
-		return (irq - 80) + EP93XX_GPIO_LINE_F(0);
-
-	return -EINVAL;
+	return irq - gpio_to_irq(0);
 }
 
 /* obsolete specific GPIO API */
diff --git a/include/asm-arm/arch-ep93xx/irqs.h b/include/asm-arm/arch-ep93xx/irqs.h
index 2a8c63638c5e..53d4a68bfc88 100644
--- a/include/asm-arm/arch-ep93xx/irqs.h
+++ b/include/asm-arm/arch-ep93xx/irqs.h
@@ -67,12 +67,6 @@
 #define IRQ_EP93XX_SAI			60
 #define EP93XX_VIC2_VALID_IRQ_MASK	0x1fffffff
 
-/*
- * Map GPIO A0..A7 to irq 64..71, B0..B7 to 72..79, and
- * F0..F7 to 80..87.
- */
-#define IRQ_EP93XX_GPIO(x)		(64 + (((x) + (((x) >> 2) & 8)) & 0x1f))
-
 #define NR_EP93XX_IRQS			(64 + 24)
 
 #define EP93XX_BOARD_IRQ(x)		(NR_EP93XX_IRQS + (x))
-- 
cgit v1.2.3


From 6331acd78f7916db16ec20b50d7838bd4944cd27 Mon Sep 17 00:00:00 2001
From: Herbert Valerio Riedel <hvr@gnu.org>
Date: Mon, 26 Nov 2007 18:50:42 +0100
Subject: [ARM] 4671/1: ep93xx: remove obsolete gpio_line_* operations

With the new GPIO methods in place the old gpio_line_* methods are redundant,
so this patch finally removes the old legacy gpio_line_* wrappers.

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ep93xx/core.c        |  9 +++------
 include/asm-arm/arch-ep93xx/gpio.h | 19 -------------------
 2 files changed, 3 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-ep93xx/core.c b/arch/arm/mach-ep93xx/core.c
index 2788da06417c..4cb3f742a6a1 100644
--- a/arch/arm/mach-ep93xx/core.c
+++ b/arch/arm/mach-ep93xx/core.c
@@ -189,6 +189,9 @@ static const u8 data_direction_register_offset[8] = {
 	0x10, 0x14, 0x34, 0x1c, 0x24, 0x18, 0x3c, 0x44,
 };
 
+#define GPIO_IN		0
+#define GPIO_OUT	1
+
 static void ep93xx_gpio_set_direction(unsigned line, int direction)
 {
 	unsigned int data_direction_register;
@@ -217,12 +220,6 @@ static void ep93xx_gpio_set_direction(unsigned line, int direction)
 	local_irq_restore(flags);
 }
 
-void __deprecated gpio_line_config(int line, int direction)
-{
-	ep93xx_gpio_set_direction(line, direction);
-}
-EXPORT_SYMBOL(gpio_line_config);
-
 int gpio_direction_input(unsigned gpio)
 {
 	if (gpio > EP93XX_GPIO_LINE_MAX)
diff --git a/include/asm-arm/arch-ep93xx/gpio.h b/include/asm-arm/arch-ep93xx/gpio.h
index cebb64be7e4b..9b1864bbd9a8 100644
--- a/include/asm-arm/arch-ep93xx/gpio.h
+++ b/include/asm-arm/arch-ep93xx/gpio.h
@@ -138,23 +138,4 @@ static inline int irq_to_gpio(unsigned irq)
 	return irq - gpio_to_irq(0);
 }
 
-/* obsolete specific GPIO API */
-#define GPIO_IN				0
-#define GPIO_OUT			1
-
-#define EP93XX_GPIO_LOW			0
-#define EP93XX_GPIO_HIGH		1
-
-void __deprecated gpio_line_config(int line, int direction);
-
-static inline int  __deprecated gpio_line_get(int line)
-{
-	return gpio_get_value(line);
-}
-
-static inline void __deprecated gpio_line_set(int line, int value)
-{
-	gpio_set_value(line, value);
-}
-
 #endif
-- 
cgit v1.2.3


From 20118ff97823822bf4d52ccb528ce2b5042c3057 Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Mon, 15 Oct 2007 14:27:41 +0100
Subject: [ARM] 4603/1: KS8695: debugfs interface to view pin state

This patch adds a debug interface (if CONFIG_DEBUG_FS is selected) to
display the basic configuration and current state of the GPIO pins on
the Kendin/Micrel KS8695 processor.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ks8695/gpio.c             | 83 +++++++++++++++++++++++++++++++++
 include/asm-arm/arch-ks8695/regs-gpio.h |  2 +
 2 files changed, 85 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-ks8695/gpio.c b/arch/arm/mach-ks8695/gpio.c
index b1aa3cb3d4a3..5e46191c0af9 100644
--- a/arch/arm/mach-ks8695/gpio.c
+++ b/arch/arm/mach-ks8695/gpio.c
@@ -20,6 +20,8 @@
 #include <linux/kernel.h>
 #include <linux/mm.h>
 #include <linux/init.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
 #include <linux/module.h>
 
 #include <asm/io.h>
@@ -216,3 +218,84 @@ int irq_to_gpio(unsigned int irq)
 	return (irq - KS8695_IRQ_EXTERN0);
 }
 EXPORT_SYMBOL(irq_to_gpio);
+
+
+/* .... Debug interface ..................................................... */
+
+#ifdef CONFIG_DEBUG_FS
+
+static int ks8695_gpio_show(struct seq_file *s, void *unused)
+{
+	unsigned int enable[] = { IOPC_IOEINT0EN, IOPC_IOEINT1EN, IOPC_IOEINT2EN, IOPC_IOEINT3EN, IOPC_IOTIM0EN, IOPC_IOTIM1EN };
+	unsigned int intmask[] = { IOPC_IOEINT0TM, IOPC_IOEINT1TM, IOPC_IOEINT2TM, IOPC_IOEINT3TM };
+	unsigned long mode, ctrl, data;
+	int i;
+
+	mode = __raw_readl(KS8695_GPIO_VA + KS8695_IOPM);
+	ctrl = __raw_readl(KS8695_GPIO_VA + KS8695_IOPC);
+	data = __raw_readl(KS8695_GPIO_VA + KS8695_IOPD);
+
+	seq_printf(s, "Pin\tI/O\tFunction\tState\n\n");
+
+	for (i = KS8695_GPIO_0; i <= KS8695_GPIO_15 ; i++) {
+		seq_printf(s, "%i:\t", i);
+
+		seq_printf(s, "%s\t", (mode & IOPM_(i)) ? "Output" : "Input");
+
+		if (i <= KS8695_GPIO_3) {
+			if (ctrl & enable[i]) {
+				seq_printf(s, "EXT%i ", i);
+
+				switch ((ctrl & intmask[i]) >> (4 * i)) {
+					case IOPC_TM_LOW:
+						seq_printf(s, "(Low)");		break;
+					case IOPC_TM_HIGH:
+						seq_printf(s, "(High)");	break;
+					case IOPC_TM_RISING:
+						seq_printf(s, "(Rising)");	break;
+					case IOPC_TM_FALLING:
+						seq_printf(s, "(Falling)");	break;
+					case IOPC_TM_EDGE:
+						seq_printf(s, "(Edges)");	break;
+				}
+			}
+			else
+				seq_printf(s, "GPIO\t");
+		}
+		else if (i <= KS8695_GPIO_5) {
+			if (ctrl & enable[i])
+				seq_printf(s, "TOUT%i\t", i - KS8695_GPIO_4);
+			else
+				seq_printf(s, "GPIO\t");
+		}
+		else
+			seq_printf(s, "GPIO\t");
+
+		seq_printf(s, "\t");
+
+		seq_printf(s, "%i\n", (data & IOPD_(i)) ? 1 : 0);
+	}
+	return 0;
+}
+
+static int ks8695_gpio_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, ks8695_gpio_show, NULL);
+}
+
+static const struct file_operations ks8695_gpio_operations = {
+	.open		= ks8695_gpio_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int __init ks8695_gpio_debugfs_init(void)
+{
+	/* /sys/kernel/debug/ks8695_gpio */
+	(void) debugfs_create_file("ks8695_gpio", S_IFREG | S_IRUGO, NULL, NULL, &ks8695_gpio_operations);
+	return 0;
+}
+postcore_initcall(ks8695_gpio_debugfs_init);
+
+#endif
diff --git a/include/asm-arm/arch-ks8695/regs-gpio.h b/include/asm-arm/arch-ks8695/regs-gpio.h
index 57fcf9fc82e4..6b95d77aea19 100644
--- a/include/asm-arm/arch-ks8695/regs-gpio.h
+++ b/include/asm-arm/arch-ks8695/regs-gpio.h
@@ -49,5 +49,7 @@
 #define IOPC_TM_FALLING		(4)		/* Falling Edge Detection */
 #define IOPC_TM_EDGE		(6)		/* Both Edge Detection */
 
+/* Port Data Register */
+#define IOPD_(x)		(1 << (x))	/* Signal Level of GPIO Pin x */
 
 #endif
-- 
cgit v1.2.3


From 3042102a28501510a409fe86962f20369e325cf2 Mon Sep 17 00:00:00 2001
From: Brian Swetland <swetland@google.com>
Date: Mon, 26 Nov 2007 04:11:43 -0800
Subject: [ARM] msm: core platform support for ARCH_MSM7X00A

- core header files for arch-msm
- Kconfig and Makefiles to enable ARCH_MSM7X00A builds
- MSM7X00A specific arch_idle
- peripheral iomap and irq number definitions

Signed-off-by: Brian Swetland <swetland@google.com>
---
 arch/arm/Kconfig                       |  11 ++++
 arch/arm/Makefile                      |   1 +
 arch/arm/mach-msm/Makefile             |   2 +
 arch/arm/mach-msm/Makefile.boot        |   3 +
 arch/arm/mach-msm/idle.S               |  36 ++++++++++++
 arch/arm/mach-msm/io.c                 |  85 +++++++++++++++++++++++++++
 arch/arm/mm/Kconfig                    |   3 +-
 include/asm-arm/arch-msm/board.h       |  37 ++++++++++++
 include/asm-arm/arch-msm/debug-macro.S |  40 +++++++++++++
 include/asm-arm/arch-msm/dma.h         |   1 +
 include/asm-arm/arch-msm/entry-macro.S |  38 ++++++++++++
 include/asm-arm/arch-msm/hardware.h    |  18 ++++++
 include/asm-arm/arch-msm/io.h          |  33 +++++++++++
 include/asm-arm/arch-msm/irqs.h        |  89 ++++++++++++++++++++++++++++
 include/asm-arm/arch-msm/memory.h      |  27 +++++++++
 include/asm-arm/arch-msm/msm_iomap.h   | 104 +++++++++++++++++++++++++++++++++
 include/asm-arm/arch-msm/system.h      |  23 ++++++++
 include/asm-arm/arch-msm/timex.h       |  20 +++++++
 include/asm-arm/arch-msm/uncompress.h  |  36 ++++++++++++
 include/asm-arm/arch-msm/vmalloc.h     |  22 +++++++
 20 files changed, 628 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-msm/Makefile
 create mode 100644 arch/arm/mach-msm/Makefile.boot
 create mode 100644 arch/arm/mach-msm/idle.S
 create mode 100644 arch/arm/mach-msm/io.c
 create mode 100644 include/asm-arm/arch-msm/board.h
 create mode 100644 include/asm-arm/arch-msm/debug-macro.S
 create mode 100644 include/asm-arm/arch-msm/dma.h
 create mode 100644 include/asm-arm/arch-msm/entry-macro.S
 create mode 100644 include/asm-arm/arch-msm/hardware.h
 create mode 100644 include/asm-arm/arch-msm/io.h
 create mode 100644 include/asm-arm/arch-msm/irqs.h
 create mode 100644 include/asm-arm/arch-msm/memory.h
 create mode 100644 include/asm-arm/arch-msm/msm_iomap.h
 create mode 100644 include/asm-arm/arch-msm/system.h
 create mode 100644 include/asm-arm/arch-msm/timex.h
 create mode 100644 include/asm-arm/arch-msm/uncompress.h
 create mode 100644 include/asm-arm/arch-msm/vmalloc.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index a04f507e7f2c..89c3811132c2 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -409,6 +409,17 @@ config ARCH_OMAP
 	help
 	  Support for TI's OMAP platform (OMAP1 and OMAP2).
 
+config ARCH_MSM7X00A
+	bool "Qualcomm MSM7X00A"
+	select GENERIC_TIME
+	select GENERIC_CLOCKEVENTS
+	help
+	  Support for Qualcomm MSM7X00A based systems.  This runs on the ARM11
+	  apps processor of the MSM7X00A and depends on a shared memory
+	  interface to the ARM9 modem processor which runs the baseband stack
+	  and controls some vital subsystems (clock and power control, etc).
+	  <http://www.cdmatech.com/products/msm7200_chipset_solution.jsp>
+
 endchoice
 
 source "arch/arm/mach-clps711x/Kconfig"
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 35e56c99ad1d..dd220d189843 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -139,6 +139,7 @@ endif
  machine-$(CONFIG_ARCH_KS8695)     := ks8695
   incdir-$(CONFIG_ARCH_MXC)	   := mxc
  machine-$(CONFIG_ARCH_MX3)	   := mx3
+ machine-$(CONFIG_ARCH_MSM7X00A)   := msm
 
 ifeq ($(CONFIG_ARCH_EBSA110),y)
 # This is what happens if you forget the IOCS16 line.
diff --git a/arch/arm/mach-msm/Makefile b/arch/arm/mach-msm/Makefile
new file mode 100644
index 000000000000..feb9e469df9a
--- /dev/null
+++ b/arch/arm/mach-msm/Makefile
@@ -0,0 +1,2 @@
+obj-y += io.o idle.o 
+
diff --git a/arch/arm/mach-msm/Makefile.boot b/arch/arm/mach-msm/Makefile.boot
new file mode 100644
index 000000000000..24dfbf8c07c4
--- /dev/null
+++ b/arch/arm/mach-msm/Makefile.boot
@@ -0,0 +1,3 @@
+  zreladdr-y		:= 0x10008000
+params_phys-y		:= 0x10000100
+initrd_phys-y		:= 0x10800000
diff --git a/arch/arm/mach-msm/idle.S b/arch/arm/mach-msm/idle.S
new file mode 100644
index 000000000000..2b1cb7f16943
--- /dev/null
+++ b/arch/arm/mach-msm/idle.S
@@ -0,0 +1,36 @@
+/* linux/include/asm-arm/arch-msm/idle.S
+ *
+ * Idle processing for MSM7K - work around bugs with SWFI.
+ *
+ * Copyright (c) 2007 QUALCOMM Incorporated.
+ * Copyright (C) 2007 Google, Inc. 
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */ 
+		
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+ENTRY(arch_idle)
+#ifdef CONFIG_MSM7X00A_IDLE
+	mrc     p15, 0, r1, c1, c0, 0    /* read current CR    */
+	bic     r0, r1, #(1 << 2)        /* clear dcache bit   */
+	bic     r0, r0, #(1 << 12)       /* clear icache bit   */
+	mcr     p15, 0, r0, c1, c0, 0    /* disable d/i cache  */
+
+	mov     r0, #0                   /* prepare wfi value  */
+	mcr     p15, 0, r0, c7, c10, 0   /* flush the cache    */
+	mcr     p15, 0, r0, c7, c10, 4   /* memory barrier     */
+	mcr     p15, 0, r0, c7, c0, 4    /* wait for interrupt */
+
+	mcr     p15, 0, r1, c1, c0, 0    /* restore d/i cache  */
+#endif
+	mov     pc, lr
diff --git a/arch/arm/mach-msm/io.c b/arch/arm/mach-msm/io.c
new file mode 100644
index 000000000000..c39edb994a88
--- /dev/null
+++ b/arch/arm/mach-msm/io.c
@@ -0,0 +1,85 @@
+/* arch/arm/mach-msm/io.c
+ *
+ * MSM7K io support
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/page.h>
+#include <asm/arch/msm_iomap.h>
+#include <asm/mach/map.h>
+
+#include <asm/arch/board.h>
+
+#define MSM_DEVICE(name) { \
+		.virtual = MSM_##name##_BASE, \
+		.pfn = __phys_to_pfn(MSM_##name##_PHYS), \
+		.length = MSM_##name##_SIZE, \
+		.type = MT_DEVICE_NONSHARED, \
+	 }
+
+static struct map_desc msm_io_desc[] __initdata = {
+	MSM_DEVICE(VIC),
+	MSM_DEVICE(CSR),
+	MSM_DEVICE(GPT),
+	MSM_DEVICE(DMOV),
+	MSM_DEVICE(UART1),
+	MSM_DEVICE(UART2),
+	MSM_DEVICE(UART3),
+	MSM_DEVICE(I2C),
+	MSM_DEVICE(GPIO1),
+	MSM_DEVICE(GPIO2),
+	MSM_DEVICE(HSUSB),
+	MSM_DEVICE(CLK_CTL),
+	MSM_DEVICE(PMDH),
+	MSM_DEVICE(EMDH),
+	MSM_DEVICE(MDP),
+	{
+		.virtual =  MSM_SHARED_RAM_BASE,
+		.pfn =      __phys_to_pfn(MSM_SHARED_RAM_PHYS),
+		.length =   MSM_SHARED_RAM_SIZE,
+		.type =     MT_DEVICE,
+	},
+};
+
+void __init msm_map_common_io(void)
+{
+	/* Make sure the peripheral register window is closed, since
+	 * we will use PTE flags (TEX[1]=1,B=0,C=1) to determine which
+	 * pages are peripheral interface or not.
+	 */
+	asm("mcr p15, 0, %0, c15, c2, 4" : : "r" (0));
+
+	iotable_init(msm_io_desc, ARRAY_SIZE(msm_io_desc));
+}
+
+void __iomem *
+__msm_ioremap(unsigned long phys_addr, size_t size, unsigned int mtype)
+{
+	if (mtype == MT_DEVICE) {
+		/* The peripherals in the 88000000 - D0000000 range
+		 * are only accessable by type MT_DEVICE_NONSHARED.
+		 * Adjust mtype as necessary to make this "just work."
+		 */
+		if ((phys_addr >= 0x88000000) && (phys_addr < 0xD0000000))
+			mtype = MT_DEVICE_NONSHARED;
+	}
+
+	return __arm_ioremap(phys_addr, size, mtype);
+}
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 7868f4dc1d00..bc2ca3c03a48 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -345,8 +345,9 @@ config CPU_XSC3
 # ARMv6
 config CPU_V6
 	bool "Support ARM V6 processor"
-	depends on ARCH_INTEGRATOR || MACH_REALVIEW_EB || ARCH_OMAP2 || ARCH_MX3
+	depends on ARCH_INTEGRATOR || MACH_REALVIEW_EB || ARCH_OMAP2 || ARCH_MX3 || ARCH_MSM7X00A
 	default y if ARCH_MX3
+	default y if ARCH_MSM7X00A
 	select CPU_32v6
 	select CPU_ABRT_EV6
 	select CPU_CACHE_V6
diff --git a/include/asm-arm/arch-msm/board.h b/include/asm-arm/arch-msm/board.h
new file mode 100644
index 000000000000..763051f8ba14
--- /dev/null
+++ b/include/asm-arm/arch-msm/board.h
@@ -0,0 +1,37 @@
+/* linux/include/asm-arm/arch-msm/board.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_BOARD_H
+#define __ASM_ARCH_MSM_BOARD_H
+
+#include <linux/types.h>
+
+/* platform device data structures */
+
+struct msm_mddi_platform_data
+{
+	void (*panel_power)(int on);
+	unsigned has_vsync_irq:1;
+};
+
+/* common init routines for use by arch/arm/mach-msm/board-*.c */
+
+void __init msm_add_devices(void);
+void __init msm_map_common_io(void);
+void __init msm_init_irq(void);
+void __init msm_init_gpio(void);
+
+#endif
diff --git a/include/asm-arm/arch-msm/debug-macro.S b/include/asm-arm/arch-msm/debug-macro.S
new file mode 100644
index 000000000000..393d5272e506
--- /dev/null
+++ b/include/asm-arm/arch-msm/debug-macro.S
@@ -0,0 +1,40 @@
+/* include/asm-arm/arch-msm7200/debug-macro.S
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <asm/hardware.h>
+#include <asm/arch/msm_iomap.h>
+
+	.macro	addruart,rx
+	@ see if the MMU is enabled and select appropriate base address
+	mrc	p15, 0, \rx, c1, c0
+	tst	\rx, #1
+	ldreq	\rx, =MSM_UART1_PHYS
+	ldrne	\rx, =MSM_UART1_BASE
+	.endm
+
+	.macro	senduart,rd,rx
+	str	\rd, [\rx, #0x0C]
+	.endm
+
+	.macro	waituart,rd,rx
+	@ wait for TX_READY
+1:	ldr	\rd, [\rx, #0x08]
+	tst	\rd, #0x04
+	beq	1b
+	.endm
+
+	.macro	busyuart,rd,rx
+	.endm
diff --git a/include/asm-arm/arch-msm/dma.h b/include/asm-arm/arch-msm/dma.h
new file mode 100644
index 000000000000..8b137891791f
--- /dev/null
+++ b/include/asm-arm/arch-msm/dma.h
@@ -0,0 +1 @@
+
diff --git a/include/asm-arm/arch-msm/entry-macro.S b/include/asm-arm/arch-msm/entry-macro.S
new file mode 100644
index 000000000000..ee24aece4cb0
--- /dev/null
+++ b/include/asm-arm/arch-msm/entry-macro.S
@@ -0,0 +1,38 @@
+/* include/asm-arm/arch-msm7200/entry-macro.S
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <asm/arch/msm_iomap.h>
+
+ 	.macro	disable_fiq
+	.endm
+
+	.macro	get_irqnr_preamble, base, tmp
+	@ enable imprecise aborts
+	cpsie	a
+	mov	\base, #MSM_VIC_BASE
+	.endm
+
+	.macro	arch_ret_to_user, tmp1, tmp2
+	.endm
+
+	.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
+	@ 0xD0 has irq# or old irq# if the irq has been handled
+	@ 0xD4 has irq# or -1 if none pending *but* if you just
+	@ read 0xD4 you never get the first irq for some reason
+	ldr	\irqnr, [\base, #0xD0]
+	ldr	\irqnr, [\base, #0xD4]
+	cmp	\irqnr, #0xffffffff
+	.endm
diff --git a/include/asm-arm/arch-msm/hardware.h b/include/asm-arm/arch-msm/hardware.h
new file mode 100644
index 000000000000..89af2b70182f
--- /dev/null
+++ b/include/asm-arm/arch-msm/hardware.h
@@ -0,0 +1,18 @@
+/* linux/include/asm-arm/arch-msm/hardware.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_HARDWARE_H
+
+#endif
diff --git a/include/asm-arm/arch-msm/io.h b/include/asm-arm/arch-msm/io.h
new file mode 100644
index 000000000000..4645ae26b62a
--- /dev/null
+++ b/include/asm-arm/arch-msm/io.h
@@ -0,0 +1,33 @@
+/* include/asm-arm/arch-msm/io.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARM_ARCH_IO_H
+#define __ASM_ARM_ARCH_IO_H
+
+#define IO_SPACE_LIMIT 0xffffffff
+
+#define __arch_ioremap __msm_ioremap
+#define __arch_iounmap __iounmap
+
+void __iomem *__msm_ioremap(unsigned long phys_addr, size_t size, unsigned int mtype);
+
+static inline void __iomem *__io(unsigned long addr)
+{
+	return (void __iomem *)addr;
+}
+#define __io(a)         __io(a)
+#define __mem_pci(a)    (a)
+
+#endif
diff --git a/include/asm-arm/arch-msm/irqs.h b/include/asm-arm/arch-msm/irqs.h
new file mode 100644
index 000000000000..565430cfaa7e
--- /dev/null
+++ b/include/asm-arm/arch-msm/irqs.h
@@ -0,0 +1,89 @@
+/* linux/include/asm-arm/arch-msm/irqs.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_IRQS_H
+
+/* MSM ARM11 Interrupt Numbers */
+/* See 80-VE113-1 A, pp219-221     */
+
+#define INT_A9_M2A_0         0
+#define INT_A9_M2A_1         1
+#define INT_A9_M2A_2         2
+#define INT_A9_M2A_3         3
+#define INT_A9_M2A_4         4
+#define INT_A9_M2A_5         5
+#define INT_A9_M2A_6         6
+#define INT_GP_TIMER_EXP     7
+#define INT_DEBUG_TIMER_EXP  8
+#define INT_UART1            9
+#define INT_UART2            10
+#define INT_UART3            11
+#define INT_UART1_RX         12
+#define INT_UART2_RX         13
+#define INT_UART3_RX         14
+#define INT_USB_OTG          15
+#define INT_MDDI_PRI         16
+#define INT_MDDI_EXT         17
+#define INT_MDDI_CLIENT      18
+#define INT_MDP              19
+#define INT_GRAPHICS         20
+#define INT_ADM_AARM         21
+#define INT_ADSP_A11         22
+#define INT_ADSP_A9_A11      23
+#define INT_SDC1_0           24
+#define INT_SDC1_1           25
+#define INT_SDC2_0           26
+#define INT_SDC2_1           27
+#define INT_KEYSENSE         28
+#define INT_TCHSCRN_SSBI     29
+#define INT_TCHSCRN1         30
+#define INT_TCHSCRN2         31
+
+#define INT_GPIO_GROUP1      (32 + 0)
+#define INT_GPIO_GROUP2      (32 + 1)
+#define INT_PWB_I2C          (32 + 2)
+#define INT_SOFTRESET        (32 + 3)
+#define INT_NAND_WR_ER_DONE  (32 + 4)
+#define INT_NAND_OP_DONE     (32 + 5)
+#define INT_PBUS_ARM11       (32 + 6)
+#define INT_AXI_MPU_SMI      (32 + 7)
+#define INT_AXI_MPU_EBI1     (32 + 8)
+#define INT_AD_HSSD          (32 + 9)
+#define INT_ARM11_PMU        (32 + 10)
+#define INT_ARM11_DMA        (32 + 11)
+#define INT_TSIF_IRQ         (32 + 12)
+#define INT_UART1DM_IRQ      (32 + 13)
+#define INT_UART1DM_RX       (32 + 14)
+#define INT_USB_HS           (32 + 15)
+#define INT_SDC3_0           (32 + 16)
+#define INT_SDC3_1           (32 + 17)
+#define INT_SDC4_0           (32 + 18)
+#define INT_SDC4_1           (32 + 19)
+#define INT_UART2DM_RX       (32 + 20)
+#define INT_UART2DM_IRQ      (32 + 21)
+
+/* 22-31 are reserved */
+
+#define MSM_IRQ_BIT(irq)     (1 << ((irq) & 31))
+
+#define NR_MSM_IRQS 64
+#define NR_GPIO_IRQS 122
+#define NR_BOARD_IRQS 64
+#define NR_IRQS (NR_MSM_IRQS + NR_GPIO_IRQS + NR_BOARD_IRQS)
+
+#define MSM_GPIO_TO_INT(n) (NR_MSM_IRQS + (n))
+
+#endif
diff --git a/include/asm-arm/arch-msm/memory.h b/include/asm-arm/arch-msm/memory.h
new file mode 100644
index 000000000000..b5ce0e9ac86d
--- /dev/null
+++ b/include/asm-arm/arch-msm/memory.h
@@ -0,0 +1,27 @@
+/* linux/include/asm-arm/arch-msm/memory.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MEMORY_H
+#define __ASM_ARCH_MEMORY_H
+
+/* physical offset of RAM */
+#define PHYS_OFFSET		UL(0x10000000)
+
+/* bus address and physical addresses are identical */
+#define __virt_to_bus(x)	__virt_to_phys(x)
+#define __bus_to_virt(x)	__phys_to_virt(x)
+
+#endif
+
diff --git a/include/asm-arm/arch-msm/msm_iomap.h b/include/asm-arm/arch-msm/msm_iomap.h
new file mode 100644
index 000000000000..b8955cc26fec
--- /dev/null
+++ b/include/asm-arm/arch-msm/msm_iomap.h
@@ -0,0 +1,104 @@
+/* linux/include/asm-arm/arch-msm/msm_iomap.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Brian Swetland <swetland@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ *
+ * The MSM peripherals are spread all over across 768MB of physical
+ * space, which makes just having a simple IO_ADDRESS macro to slide
+ * them into the right virtual location rough.  Instead, we will
+ * provide a master phys->virt mapping for peripherals here.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_IOMAP_H
+#define __ASM_ARCH_MSM_IOMAP_H
+
+#include <asm/sizes.h>
+
+/* Physical base address and size of peripherals.
+ * Ordered by the virtual base addresses they will be mapped at.
+ *
+ * MSM_VIC_BASE must be an value that can be loaded via a "mov"
+ * instruction, otherwise entry-macro.S will not compile.
+ *
+ * If you add or remove entries here, you'll want to edit the
+ * msm_io_desc array in arch/arm/mach-msm/io.c to reflect your
+ * changes.
+ *
+ */
+
+#define MSM_VIC_BASE          0xE0000000
+#define MSM_VIC_PHYS          0xC0000000
+#define MSM_VIC_SIZE          SZ_4K
+
+#define MSM_CSR_BASE          0xE0001000
+#define MSM_CSR_PHYS          0xC0100000
+#define MSM_CSR_SIZE          SZ_4K
+
+#define MSM_GPT_PHYS          MSM_CSR_PHYS
+#define MSM_GPT_BASE          MSM_CSR_BASE
+#define MSM_GPT_SIZE          SZ_4K
+
+#define MSM_DMOV_BASE         0xE0002000
+#define MSM_DMOV_PHYS         0xA9700000
+#define MSM_DMOV_SIZE         SZ_4K
+
+#define MSM_UART1_BASE        0xE0003000
+#define MSM_UART1_PHYS        0xA9A00000
+#define MSM_UART1_SIZE        SZ_4K
+
+#define MSM_UART2_BASE        0xE0004000
+#define MSM_UART2_PHYS        0xA9B00000
+#define MSM_UART2_SIZE        SZ_4K
+
+#define MSM_UART3_BASE        0xE0005000
+#define MSM_UART3_PHYS        0xA9C00000
+#define MSM_UART3_SIZE        SZ_4K
+
+#define MSM_I2C_BASE          0xE0006000
+#define MSM_I2C_PHYS          0xA9900000
+#define MSM_I2C_SIZE          SZ_4K
+
+#define MSM_GPIO1_BASE        0xE0007000
+#define MSM_GPIO1_PHYS        0xA9200000
+#define MSM_GPIO1_SIZE        SZ_4K
+
+#define MSM_GPIO2_BASE        0xE0008000
+#define MSM_GPIO2_PHYS        0xA9300000
+#define MSM_GPIO2_SIZE        SZ_4K
+
+#define MSM_HSUSB_BASE        0xE0009000
+#define MSM_HSUSB_PHYS        0xA0800000
+#define MSM_HSUSB_SIZE        SZ_4K
+
+#define MSM_CLK_CTL_BASE      0xE000A000
+#define MSM_CLK_CTL_PHYS      0xA8600000
+#define MSM_CLK_CTL_SIZE      SZ_4K
+
+#define MSM_PMDH_BASE         0xE000B000
+#define MSM_PMDH_PHYS         0xAA600000
+#define MSM_PMDH_SIZE         SZ_4K
+
+#define MSM_EMDH_BASE         0xE000C000
+#define MSM_EMDH_PHYS         0xAA700000
+#define MSM_EMDH_SIZE         SZ_4K
+
+#define MSM_MDP_BASE          0xE0010000
+#define MSM_MDP_PHYS          0xAA200000
+#define MSM_MDP_SIZE          0x000F0000
+
+#define MSM_SHARED_RAM_BASE   0xE0100000
+#define MSM_SHARED_RAM_PHYS   0x01F00000
+#define MSM_SHARED_RAM_SIZE   SZ_1M
+
+#endif
diff --git a/include/asm-arm/arch-msm/system.h b/include/asm-arm/arch-msm/system.h
new file mode 100644
index 000000000000..7c5544bdd0c7
--- /dev/null
+++ b/include/asm-arm/arch-msm/system.h
@@ -0,0 +1,23 @@
+/* linux/include/asm-arm/arch-msm/system.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <asm/hardware.h>
+
+void arch_idle(void);
+
+static inline void arch_reset(char mode)
+{
+	for (;;) ;  /* depends on IPC w/ other core */
+}
diff --git a/include/asm-arm/arch-msm/timex.h b/include/asm-arm/arch-msm/timex.h
new file mode 100644
index 000000000000..154b23fb3599
--- /dev/null
+++ b/include/asm-arm/arch-msm/timex.h
@@ -0,0 +1,20 @@
+/* linux/include/asm-arm/arch-msm/timex.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_TIMEX_H
+
+#define CLOCK_TICK_RATE		1000000
+
+#endif
diff --git a/include/asm-arm/arch-msm/uncompress.h b/include/asm-arm/arch-msm/uncompress.h
new file mode 100644
index 000000000000..e91ed786ffec
--- /dev/null
+++ b/include/asm-arm/arch-msm/uncompress.h
@@ -0,0 +1,36 @@
+/* linux/include/asm-arm/arch-msm/uncompress.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_UNCOMPRESS_H
+
+#include "hardware.h"
+
+static void putc(int c)
+{
+}
+
+static inline void flush(void)
+{
+}
+
+static inline void arch_decomp_setup(void)
+{
+}
+
+static inline void arch_decomp_wdog(void)
+{
+}
+
+#endif
diff --git a/include/asm-arm/arch-msm/vmalloc.h b/include/asm-arm/arch-msm/vmalloc.h
new file mode 100644
index 000000000000..60f8d910e825
--- /dev/null
+++ b/include/asm-arm/arch-msm/vmalloc.h
@@ -0,0 +1,22 @@
+/* linux/include/asm-arm/arch-msm/vmalloc.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __ASM_ARCH_MSM_VMALLOC_H
+#define __ASM_ARCH_MSM_VMALLOC_H
+
+#define VMALLOC_END	  (PAGE_OFFSET + 0x10000000)
+
+#endif
+
-- 
cgit v1.2.3


From bfe645adf1a79b873b528556523abb46f281a5dc Mon Sep 17 00:00:00 2001
From: Arve Hjønnevåg <arve@android.com>
Date: Mon, 26 Nov 2007 04:12:29 -0800
Subject: [ARM] msm: dma support for MSM7X00A

Signed-off-by: Brian Swetland <swetland@google.com>
---
 arch/arm/mach-msm/Makefile     |   2 +-
 arch/arm/mach-msm/dma.c        | 214 +++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-msm/dma.h | 150 +++++++++++++++++++++++++++++
 3 files changed, 365 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-msm/dma.c

(limited to 'include')

diff --git a/arch/arm/mach-msm/Makefile b/arch/arm/mach-msm/Makefile
index fcca853d6485..d12f23655850 100644
--- a/arch/arm/mach-msm/Makefile
+++ b/arch/arm/mach-msm/Makefile
@@ -1,4 +1,4 @@
-obj-y += io.o idle.o irq.o timer.o
+obj-y += io.o idle.o irq.o timer.o dma.o
 
 # Common code for board init
 obj-y += common.o
diff --git a/arch/arm/mach-msm/dma.c b/arch/arm/mach-msm/dma.c
new file mode 100644
index 000000000000..8b0f339b3274
--- /dev/null
+++ b/arch/arm/mach-msm/dma.c
@@ -0,0 +1,214 @@
+/* linux/arch/arm/mach-msm/dma.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <asm/io.h>
+#include <linux/interrupt.h>
+#include <asm/arch/dma.h>
+
+#define MSM_DMOV_CHANNEL_COUNT 16
+
+enum {
+	MSM_DMOV_PRINT_ERRORS = 1,
+	MSM_DMOV_PRINT_IO = 2,
+	MSM_DMOV_PRINT_FLOW = 4
+};
+
+static DEFINE_SPINLOCK(msm_dmov_lock);
+static struct msm_dmov_cmd active_command;
+static struct list_head ready_commands[MSM_DMOV_CHANNEL_COUNT];
+static struct list_head active_commands[MSM_DMOV_CHANNEL_COUNT];
+unsigned int msm_dmov_print_mask = MSM_DMOV_PRINT_ERRORS;
+
+#define MSM_DMOV_DPRINTF(mask, format, args...) \
+	do { \
+		if ((mask) & msm_dmov_print_mask) \
+			printk(KERN_ERR format, args); \
+	} while (0)
+#define PRINT_ERROR(format, args...) \
+	MSM_DMOV_DPRINTF(MSM_DMOV_PRINT_ERRORS, format, args);
+#define PRINT_IO(format, args...) \
+	MSM_DMOV_DPRINTF(MSM_DMOV_PRINT_IO, format, args);
+#define PRINT_FLOW(format, args...) \
+	MSM_DMOV_DPRINTF(MSM_DMOV_PRINT_FLOW, format, args);
+
+void msm_dmov_enqueue_cmd(unsigned id, struct msm_dmov_cmd *cmd)
+{
+	unsigned long irq_flags;
+	unsigned int status;
+
+	spin_lock_irqsave(&msm_dmov_lock, irq_flags);
+	status = readl(DMOV_STATUS(id));
+	if (list_empty(&ready_commands[id]) &&
+		(status & DMOV_STATUS_CMD_PTR_RDY)) {
+#if 0
+		if (list_empty(&active_commands[id])) {
+			PRINT_FLOW("msm_dmov_enqueue_cmd(%d), enable interrupt\n", id);
+			writel(DMOV_CONFIG_IRQ_EN, DMOV_CONFIG(id));
+		}
+#endif
+		PRINT_IO("msm_dmov_enqueue_cmd(%d), start command, status %x\n", id, status);
+		list_add_tail(&cmd->list, &active_commands[id]);
+		writel(cmd->cmdptr, DMOV_CMD_PTR(id));
+	} else {
+		if (list_empty(&active_commands[id]))
+			PRINT_ERROR("msm_dmov_enqueue_cmd(%d), error datamover stalled, status %x\n", id, status);
+
+		PRINT_IO("msm_dmov_enqueue_cmd(%d), enqueue command, status %x\n", id, status);
+		list_add_tail(&cmd->list, &ready_commands[id]);
+	}
+	spin_unlock_irqrestore(&msm_dmov_lock, irq_flags);
+}
+
+struct msm_dmov_exec_cmdptr_cmd {
+	struct msm_dmov_cmd dmov_cmd;
+	struct completion complete;
+	unsigned id;
+	unsigned int result;
+	unsigned int flush[6];
+};
+
+static void dmov_exec_cmdptr_complete_func(struct msm_dmov_cmd *_cmd, unsigned int result)
+{
+	struct msm_dmov_exec_cmdptr_cmd *cmd = container_of(_cmd, struct msm_dmov_exec_cmdptr_cmd, dmov_cmd);
+	cmd->result = result;
+	if (result != 0x80000002) {
+		cmd->flush[0] = readl(DMOV_FLUSH0(cmd->id));
+		cmd->flush[1] = readl(DMOV_FLUSH1(cmd->id));
+		cmd->flush[2] = readl(DMOV_FLUSH2(cmd->id));
+		cmd->flush[3] = readl(DMOV_FLUSH3(cmd->id));
+		cmd->flush[4] = readl(DMOV_FLUSH4(cmd->id));
+		cmd->flush[5] = readl(DMOV_FLUSH5(cmd->id));
+	}
+	complete(&cmd->complete);
+}
+
+int msm_dmov_exec_cmd(unsigned id, unsigned int cmdptr)
+{
+	struct msm_dmov_exec_cmdptr_cmd cmd;
+
+	PRINT_FLOW("dmov_exec_cmdptr(%d, %x)\n", id, cmdptr);
+
+	cmd.dmov_cmd.cmdptr = cmdptr;
+	cmd.dmov_cmd.complete_func = dmov_exec_cmdptr_complete_func;
+	cmd.id = id;
+	init_completion(&cmd.complete);
+
+	msm_dmov_enqueue_cmd(id, &cmd.dmov_cmd);
+	wait_for_completion(&cmd.complete);
+
+	if (cmd.result != 0x80000002) {
+		PRINT_ERROR("dmov_exec_cmdptr(%d): ERROR, result: %x\n", id, cmd.result);
+		PRINT_ERROR("dmov_exec_cmdptr(%d):  flush: %x %x %x %x\n",
+			id, cmd.flush[0], cmd.flush[1], cmd.flush[2], cmd.flush[3]);
+		return -EIO;
+	}
+	PRINT_FLOW("dmov_exec_cmdptr(%d, %x) done\n", id, cmdptr);
+	return 0;
+}
+
+
+static irqreturn_t msm_datamover_irq_handler(int irq, void *dev_id)
+{
+	unsigned int int_status, mask, id;
+	unsigned long irq_flags;
+	unsigned int ch_status;
+	unsigned int ch_result;
+	struct msm_dmov_cmd *cmd;
+
+	spin_lock_irqsave(&msm_dmov_lock, irq_flags);
+
+	int_status = readl(DMOV_ISR); /* read and clear interrupt */
+	PRINT_FLOW("msm_datamover_irq_handler: DMOV_ISR %x\n", int_status);
+
+	while (int_status) {
+		mask = int_status & -int_status;
+		id = fls(mask) - 1;
+		PRINT_FLOW("msm_datamover_irq_handler %08x %08x id %d\n", int_status, mask, id);
+		int_status &= ~mask;
+		ch_status = readl(DMOV_STATUS(id));
+		if (!(ch_status & DMOV_STATUS_RSLT_VALID)) {
+			PRINT_FLOW("msm_datamover_irq_handler id %d, result not valid %x\n", id, ch_status);
+			continue;
+		}
+		do {
+			ch_result = readl(DMOV_RSLT(id));
+			if (list_empty(&active_commands[id])) {
+				PRINT_ERROR("msm_datamover_irq_handler id %d, got result "
+					"with no active command, status %x, result %x\n",
+					id, ch_status, ch_result);
+				cmd = NULL;
+			} else
+				cmd = list_entry(active_commands[id].next, typeof(*cmd), list);
+			PRINT_FLOW("msm_datamover_irq_handler id %d, status %x, result %x\n", id, ch_status, ch_result);
+			if (ch_result & DMOV_RSLT_DONE) {
+				PRINT_FLOW("msm_datamover_irq_handler id %d, status %x\n",
+					id, ch_status);
+				PRINT_IO("msm_datamover_irq_handler id %d, got result "
+					"for %p, result %x\n", id, cmd, ch_result);
+				if (cmd) {
+					list_del(&cmd->list);
+					cmd->complete_func(cmd, ch_result);
+				}
+			}
+			if (ch_result & DMOV_RSLT_FLUSH) {
+				unsigned int flush0 = readl(DMOV_FLUSH0(id));
+				PRINT_FLOW("msm_datamover_irq_handler id %d, status %x\n", id, ch_status);
+				PRINT_FLOW("msm_datamover_irq_handler id %d, flush, result %x, flush0 %x\n", id, ch_result, flush0);
+				if (cmd) {
+					list_del(&cmd->list);
+					cmd->complete_func(cmd, ch_result);
+				}
+			}
+			if (ch_result & DMOV_RSLT_ERROR) {
+				unsigned int flush0 = readl(DMOV_FLUSH0(id));
+				PRINT_ERROR("msm_datamover_irq_handler id %d, status %x\n", id, ch_status);
+				PRINT_ERROR("msm_datamover_irq_handler id %d, error, result %x, flush0 %x\n", id, ch_result, flush0);
+				if (cmd) {
+					list_del(&cmd->list);
+					cmd->complete_func(cmd, ch_result);
+				}
+				/* this does not seem to work, once we get an error */
+				/* the datamover will no longer accept commands */
+				writel(0, DMOV_FLUSH0(id));
+			}
+			ch_status = readl(DMOV_STATUS(id));
+			PRINT_FLOW("msm_datamover_irq_handler id %d, status %x\n", id, ch_status);
+			if ((ch_status & DMOV_STATUS_CMD_PTR_RDY) && !list_empty(&ready_commands[id])) {
+				cmd = list_entry(ready_commands[id].next, typeof(*cmd), list);
+				list_del(&cmd->list);
+				list_add_tail(&cmd->list, &active_commands[id]);
+				PRINT_FLOW("msm_datamover_irq_handler id %d, start command\n", id);
+				writel(cmd->cmdptr, DMOV_CMD_PTR(id));
+			}
+		} while (ch_status & DMOV_STATUS_RSLT_VALID);
+		PRINT_FLOW("msm_datamover_irq_handler id %d, status %x\n", id, ch_status);
+	}
+	spin_unlock_irqrestore(&msm_dmov_lock, irq_flags);
+	return IRQ_HANDLED;
+}
+
+static int __init msm_init_datamover(void)
+{
+	int i;
+	for (i = 0; i < MSM_DMOV_CHANNEL_COUNT; i++) {
+		INIT_LIST_HEAD(&ready_commands[i]);
+		INIT_LIST_HEAD(&active_commands[i]);
+		writel(DMOV_CONFIG_IRQ_EN | DMOV_CONFIG_FORCE_TOP_PTR_RSLT | DMOV_CONFIG_FORCE_FLUSH_RSLT, DMOV_CONFIG(i));
+	}
+	return request_irq(INT_ADM_AARM, msm_datamover_irq_handler, 0, "msmdatamover", NULL);
+}
+
+arch_initcall(msm_init_datamover);
+
diff --git a/include/asm-arm/arch-msm/dma.h b/include/asm-arm/arch-msm/dma.h
index 8b137891791f..e4b565b27b35 100644
--- a/include/asm-arm/arch-msm/dma.h
+++ b/include/asm-arm/arch-msm/dma.h
@@ -1 +1,151 @@
+/* linux/include/asm-arm/arch-msm/dma.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
 
+#ifndef __ASM_ARCH_MSM_DMA_H
+
+#include <linux/list.h>
+#include <asm/arch/msm_iomap.h>
+
+struct msm_dmov_cmd {
+	struct list_head list;
+	unsigned int cmdptr;
+	void (*complete_func)(struct msm_dmov_cmd *cmd, unsigned int result);
+/*	void (*user_result_func)(struct msm_dmov_cmd *cmd); */
+};
+
+void msm_dmov_enqueue_cmd(unsigned id, struct msm_dmov_cmd *cmd);
+void msm_dmov_stop_cmd(unsigned id, struct msm_dmov_cmd *cmd);
+int msm_dmov_exec_cmd(unsigned id, unsigned int cmdptr);
+/* int msm_dmov_exec_cmd_etc(unsigned id, unsigned int cmdptr, int timeout, int interruptible); */
+
+
+
+#define DMOV_SD0(off, ch) (MSM_DMOV_BASE + 0x0000 + (off) + ((ch) << 2))
+#define DMOV_SD1(off, ch) (MSM_DMOV_BASE + 0x0400 + (off) + ((ch) << 2))
+#define DMOV_SD2(off, ch) (MSM_DMOV_BASE + 0x0800 + (off) + ((ch) << 2))
+#define DMOV_SD3(off, ch) (MSM_DMOV_BASE + 0x0C00 + (off) + ((ch) << 2))
+
+/* only security domain 3 is available to the ARM11
+ * SD0 -> mARM trusted, SD1 -> mARM nontrusted, SD2 -> aDSP, SD3 -> aARM
+ */
+
+#define DMOV_CMD_PTR(ch)      DMOV_SD3(0x000, ch)
+#define DMOV_CMD_LIST         (0 << 29) /* does not work */
+#define DMOV_CMD_PTR_LIST     (1 << 29) /* works */
+#define DMOV_CMD_INPUT_CFG    (2 << 29) /* untested */
+#define DMOV_CMD_OUTPUT_CFG   (3 << 29) /* untested */
+#define DMOV_CMD_ADDR(addr)   ((addr) >> 3)
+
+#define DMOV_RSLT(ch)         DMOV_SD3(0x040, ch)
+#define DMOV_RSLT_VALID       (1 << 31) /* 0 == host has empties result fifo */
+#define DMOV_RSLT_ERROR       (1 << 3)
+#define DMOV_RSLT_FLUSH       (1 << 2)
+#define DMOV_RSLT_DONE        (1 << 1)  /* top pointer done */
+#define DMOV_RSLT_USER        (1 << 0)  /* command with FR force result */
+
+#define DMOV_FLUSH0(ch)       DMOV_SD3(0x080, ch)
+#define DMOV_FLUSH1(ch)       DMOV_SD3(0x0C0, ch)
+#define DMOV_FLUSH2(ch)       DMOV_SD3(0x100, ch)
+#define DMOV_FLUSH3(ch)       DMOV_SD3(0x140, ch)
+#define DMOV_FLUSH4(ch)       DMOV_SD3(0x180, ch)
+#define DMOV_FLUSH5(ch)       DMOV_SD3(0x1C0, ch)
+
+#define DMOV_STATUS(ch)       DMOV_SD3(0x200, ch)
+#define DMOV_STATUS_RSLT_COUNT(n)    (((n) >> 29))
+#define DMOV_STATUS_CMD_COUNT(n)     (((n) >> 27) & 3)
+#define DMOV_STATUS_RSLT_VALID       (1 << 1)
+#define DMOV_STATUS_CMD_PTR_RDY      (1 << 0)
+
+#define DMOV_ISR              DMOV_SD3(0x380, 0)
+
+#define DMOV_CONFIG(ch)       DMOV_SD3(0x300, ch)
+#define DMOV_CONFIG_FORCE_TOP_PTR_RSLT (1 << 2)
+#define DMOV_CONFIG_FORCE_FLUSH_RSLT   (1 << 1)
+#define DMOV_CONFIG_IRQ_EN             (1 << 0)
+
+/* channel assignments */
+
+#define DMOV_NAND_CHAN        7
+#define DMOV_NAND_CRCI_CMD    5
+#define DMOV_NAND_CRCI_DATA   4
+
+#define DMOV_SDC1_CHAN        8
+#define DMOV_SDC1_CRCI        6
+
+#define DMOV_SDC2_CHAN        8
+#define DMOV_SDC2_CRCI        7
+
+#define DMOV_TSIF_CHAN        10
+#define DMOV_TSIF_CRCI        10
+
+#define DMOV_USB_CHAN         11
+
+/* no client rate control ifc (eg, ram) */
+#define DMOV_NONE_CRCI        0
+
+
+/* If the CMD_PTR register has CMD_PTR_LIST selected, the data mover
+ * is going to walk a list of 32bit pointers as described below.  Each
+ * pointer points to a *array* of dmov_s, etc structs.  The last pointer
+ * in the list is marked with CMD_PTR_LP.  The last struct in each array
+ * is marked with CMD_LC (see below).
+ */
+#define CMD_PTR_ADDR(addr)  ((addr) >> 3)
+#define CMD_PTR_LP          (1 << 31) /* last pointer */
+#define CMD_PTR_PT          (3 << 29) /* ? */
+
+/* Single Item Mode */
+typedef struct {
+	unsigned cmd;
+	unsigned src;
+	unsigned dst;
+	unsigned len;
+} dmov_s;
+
+/* Scatter/Gather Mode */
+typedef struct {
+	unsigned cmd;
+	unsigned src_dscr;
+	unsigned dst_dscr;
+	unsigned _reserved;
+} dmov_sg;
+
+/* bits for the cmd field of the above structures */
+
+#define CMD_LC      (1 << 31)  /* last command */
+#define CMD_FR      (1 << 22)  /* force result -- does not work? */
+#define CMD_OCU     (1 << 21)  /* other channel unblock */
+#define CMD_OCB     (1 << 20)  /* other channel block */
+#define CMD_TCB     (1 << 19)  /* ? */
+#define CMD_DAH     (1 << 18)  /* destination address hold -- does not work?*/
+#define CMD_SAH     (1 << 17)  /* source address hold -- does not work? */
+
+#define CMD_MODE_SINGLE     (0 << 0) /* dmov_s structure used */
+#define CMD_MODE_SG         (1 << 0) /* untested */
+#define CMD_MODE_IND_SG     (2 << 0) /* untested */
+#define CMD_MODE_BOX        (3 << 0) /* untested */
+
+#define CMD_DST_SWAP_BYTES  (1 << 14) /* exchange each byte n with byte n+1 */
+#define CMD_DST_SWAP_SHORTS (1 << 15) /* exchange each short n with short n+1 */
+#define CMD_DST_SWAP_WORDS  (1 << 16) /* exchange each word n with word n+1 */
+
+#define CMD_SRC_SWAP_BYTES  (1 << 11) /* exchange each byte n with byte n+1 */
+#define CMD_SRC_SWAP_SHORTS (1 << 12) /* exchange each short n with short n+1 */
+#define CMD_SRC_SWAP_WORDS  (1 << 13) /* exchange each word n with word n+1 */
+
+#define CMD_DST_CRCI(n)     (((n) & 15) << 7)
+#define CMD_SRC_CRCI(n)     (((n) & 15) << 3)
+
+#endif
-- 
cgit v1.2.3


From c98929c07a01c9ec2e1e5253456acc7168da8b66 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 22 Nov 2007 18:32:01 +0100
Subject: [ARM] 4582/2: Add support for the common VFP subarchitecture

This patch allows the VFP support code to run correctly on CPUs
compatible with the common VFP subarchitecture specification (Appendix
B in the ARM ARM v7-A and v7-R edition). It implements support for VFP
subarchitecture 2 while being backwards compatible with
subarchitecture 1.

On VFP subarchitecture 1, the arithmetic exceptions are asynchronous
(or imprecise as described in the old ARM ARM) unless the FPSCR.IXE
bit is 1. The exceptional instructions can be read from FPINST and
FPINST2 registers. With VFP subarchitecture 2, the arithmetic
exceptions can also be synchronous and marked by the FPEXC.DEX bit
(the FPEXC.EX bit is cleared). CPUs implementing the synchronous
arithmetic exceptions don't have the FPINST and FPINST2 registers and
accessing them would trigger and undefined exception.

Note that FPEXC.EX bit has an additional meaning on subarchitecture 1
- if it isn't set, there is no additional information in FPINST and
FPINST2 that needs to be saved at context switch or when lazy-loading
the VFP state of a different thread.

The patch also removes the clearing of the cumulative exception flags in
FPSCR when additional exceptions were raised. It is up to the user
application to clear these bits.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/vfp/vfphw.S     | 38 +++++++++----------
 arch/arm/vfp/vfpmodule.c | 98 +++++++++++++++++++++++++++---------------------
 include/asm-arm/vfp.h    | 30 +++++++--------
 3 files changed, 89 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/arch/arm/vfp/vfphw.S b/arch/arm/vfp/vfphw.S
index 0ac022f800a1..53d9f8e8fac3 100644
--- a/arch/arm/vfp/vfphw.S
+++ b/arch/arm/vfp/vfphw.S
@@ -100,10 +100,10 @@ vfp_support_entry:
 	cmp	r4, #0
 	beq	no_old_VFP_process
 	VFPFMRX	r5, FPSCR		@ current status
-	VFPFMRX	r6, FPINST		@ FPINST (always there, rev0 onwards)
-	tst	r1, #FPEXC_FPV2		@ is there an FPINST2 to read?
-	VFPFMRX	r8, FPINST2, NE		@ FPINST2 if needed - avoids reading
-					@ nonexistant reg on rev0
+	tst	r1, #FPEXC_EX		@ is there additional state to save?
+	VFPFMRX	r6, FPINST, NE		@ FPINST (only if FPEXC.EX is set)
+	tstne	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
+	VFPFMRX	r8, FPINST2, NE		@ FPINST2 if needed (and present)
 	VFPFSTMIA r4 			@ save the working registers
 	stmia	r4, {r1, r5, r6, r8}	@ save FPEXC, FPSCR, FPINST, FPINST2
 					@ and point r4 at the word at the
@@ -117,10 +117,10 @@ no_old_VFP_process:
 	VFPFLDMIA r10	 		@ reload the working registers while
 					@ FPEXC is in a safe state
 	ldmia	r10, {r1, r5, r6, r8}	@ load FPEXC, FPSCR, FPINST, FPINST2
-	tst	r1, #FPEXC_FPV2		@ is there an FPINST2 to write?
-	VFPFMXR	FPINST2, r8, NE		@ FPINST2 if needed - avoids writing
-					@ nonexistant reg on rev0
-	VFPFMXR	FPINST, r6
+	tst	r1, #FPEXC_EX		@ is there additional state to restore?
+	VFPFMXR	FPINST, r6, NE		@ restore FPINST (only if FPEXC.EX is set)
+	tstne	r1, #FPEXC_FP2V		@ is there an FPINST2 to write?
+	VFPFMXR	FPINST2, r8, NE		@ FPINST2 if needed (and present)
 	VFPFMXR	FPSCR, r5		@ restore status
 
 check_for_exception:
@@ -136,10 +136,14 @@ check_for_exception:
 
 
 look_for_VFP_exceptions:
-	tst	r1, #FPEXC_EX
+	@ Check for synchronous or asynchronous exception
+	tst	r1, #FPEXC_EX | FPEXC_DEX
 	bne	process_exception
+	@ On some implementations of the VFP subarch 1, setting FPSCR.IXE
+	@ causes all the CDP instructions to be bounced synchronously without
+	@ setting the FPEXC.EX bit
 	VFPFMRX	r5, FPSCR
-	tst	r5, #FPSCR_IXE		@ IXE doesn't set FPEXC_EX !
+	tst	r5, #FPSCR_IXE
 	bne	process_exception
 
 	@ Fall into hand on to next handler - appropriate coproc instr
@@ -150,10 +154,6 @@ look_for_VFP_exceptions:
 
 process_exception:
 	DBGSTR	"bounce"
-	sub	r2, r2, #4
-	str	r2, [sp, #S_PC]		@ retry the instruction on exit from
-					@ the imprecise exception handling in
-					@ the support code
 	mov	r2, sp			@ nothing stacked - regdump is at TOS
 	mov	lr, r9			@ setup for a return to the user code.
 
@@ -161,7 +161,7 @@ process_exception:
 	@   r0 holds the trigger instruction
 	@   r1 holds the FPEXC value
 	@   r2 pointer to register dump
-	b	VFP9_bounce		@ we have handled this - the support
+	b	VFP_bounce		@ we have handled this - the support
 					@ code will raise an exception if
 					@ required. If not, the user code will
 					@ retry the faulted instruction
@@ -175,10 +175,10 @@ vfp_save_state:
 	@ r1 - FPEXC
 	DBGSTR1	"save VFP state %p", r0
 	VFPFMRX	r2, FPSCR		@ current status
-	VFPFMRX	r3, FPINST		@ FPINST (always there, rev0 onwards)
-	tst	r1, #FPEXC_FPV2		@ is there an FPINST2 to read?
-	VFPFMRX	r12, FPINST2, NE	@ FPINST2 if needed - avoids reading
-					@ nonexistant reg on rev0
+	tst	r1, #FPEXC_EX		@ is there additional state to save?
+	VFPFMRX	r3, FPINST, NE		@ FPINST (only if FPEXC.EX is set)
+	tstne	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
+	VFPFMRX	r12, FPINST2, NE	@ FPINST2 if needed (and present)
 	VFPFSTMIA r0 			@ save the working registers
 	stmia	r0, {r1, r2, r3, r12}	@ save FPEXC, FPSCR, FPINST, FPINST2
 	mov	pc, lr
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index b4e210df92f2..32455c633f1c 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -125,13 +125,13 @@ void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
 	send_sig_info(SIGFPE, &info, current);
 }
 
-static void vfp_panic(char *reason)
+static void vfp_panic(char *reason, u32 inst)
 {
 	int i;
 
 	printk(KERN_ERR "VFP: Error: %s\n", reason);
 	printk(KERN_ERR "VFP: EXC 0x%08x SCR 0x%08x INST 0x%08x\n",
-		fmrx(FPEXC), fmrx(FPSCR), fmrx(FPINST));
+		fmrx(FPEXC), fmrx(FPSCR), inst);
 	for (i = 0; i < 32; i += 2)
 		printk(KERN_ERR "VFP: s%2u: 0x%08x s%2u: 0x%08x\n",
 		       i, vfp_get_float(i), i+1, vfp_get_float(i+1));
@@ -147,19 +147,16 @@ static void vfp_raise_exceptions(u32 exceptions, u32 inst, u32 fpscr, struct pt_
 	pr_debug("VFP: raising exceptions %08x\n", exceptions);
 
 	if (exceptions == VFP_EXCEPTION_ERROR) {
-		vfp_panic("unhandled bounce");
+		vfp_panic("unhandled bounce", inst);
 		vfp_raise_sigfpe(0, regs);
 		return;
 	}
 
 	/*
-	 * If any of the status flags are set, update the FPSCR.
+	 * Update the FPSCR with the additional exception flags.
 	 * Comparison instructions always return at least one of
 	 * these flags set.
 	 */
-	if (exceptions & (FPSCR_N|FPSCR_Z|FPSCR_C|FPSCR_V))
-		fpscr &= ~(FPSCR_N|FPSCR_Z|FPSCR_C|FPSCR_V);
-
 	fpscr |= exceptions;
 
 	fmxr(FPSCR, fpscr);
@@ -220,35 +217,64 @@ static u32 vfp_emulate_instruction(u32 inst, u32 fpscr, struct pt_regs *regs)
 /*
  * Package up a bounce condition.
  */
-void VFP9_bounce(u32 trigger, u32 fpexc, struct pt_regs *regs)
+void VFP_bounce(u32 trigger, u32 fpexc, struct pt_regs *regs)
 {
-	u32 fpscr, orig_fpscr, exceptions, inst;
+	u32 fpscr, orig_fpscr, fpsid, exceptions;
 
 	pr_debug("VFP: bounce: trigger %08x fpexc %08x\n", trigger, fpexc);
 
 	/*
-	 * Enable access to the VFP so we can handle the bounce.
+	 * At this point, FPEXC can have the following configuration:
+	 *
+	 *  EX DEX IXE
+	 *  0   1   x   - synchronous exception
+	 *  1   x   0   - asynchronous exception
+	 *  1   x   1   - sychronous on VFP subarch 1 and asynchronous on later
+	 *  0   0   1   - synchronous on VFP9 (non-standard subarch 1
+	 *                implementation), undefined otherwise
+	 *
+	 * Clear various bits and enable access to the VFP so we can
+	 * handle the bounce.
 	 */
-	fmxr(FPEXC, fpexc & ~(FPEXC_EX|FPEXC_FPV2|FPEXC_INV|FPEXC_UFC|FPEXC_OFC|FPEXC_IOC));
+	fmxr(FPEXC, fpexc & ~(FPEXC_EX|FPEXC_DEX|FPEXC_FP2V|FPEXC_VV|FPEXC_TRAP_MASK));
 
+	fpsid = fmrx(FPSID);
 	orig_fpscr = fpscr = fmrx(FPSCR);
 
 	/*
-	 * If we are running with inexact exceptions enabled, we need to
-	 * emulate the trigger instruction.  Note that as we're emulating
-	 * the trigger instruction, we need to increment PC.
+	 * Check for the special VFP subarch 1 and FPSCR.IXE bit case
 	 */
-	if (fpscr & FPSCR_IXE) {
-		regs->ARM_pc += 4;
+	if ((fpsid & FPSID_ARCH_MASK) == (1 << FPSID_ARCH_BIT)
+	    && (fpscr & FPSCR_IXE)) {
+		/*
+		 * Synchronous exception, emulate the trigger instruction
+		 */
 		goto emulate;
 	}
 
-	barrier();
+	if (fpexc & FPEXC_EX) {
+		/*
+		 * Asynchronous exception. The instruction is read from FPINST
+		 * and the interrupted instruction has to be restarted.
+		 */
+		trigger = fmrx(FPINST);
+		regs->ARM_pc -= 4;
+	} else if (!(fpexc & FPEXC_DEX)) {
+		/*
+		 * Illegal combination of bits. It can be caused by an
+		 * unallocated VFP instruction but with FPSCR.IXE set and not
+		 * on VFP subarch 1.
+		 */
+		 vfp_raise_exceptions(VFP_EXCEPTION_ERROR, trigger, fpscr, regs);
+		 return;
+	}
 
 	/*
-	 * Modify fpscr to indicate the number of iterations remaining
+	 * Modify fpscr to indicate the number of iterations remaining.
+	 * If FPEXC.EX is 0, FPEXC.DEX is 1 and the FPEXC.VV bit indicates
+	 * whether FPEXC.VECITR or FPSCR.LEN is used.
 	 */
-	if (fpexc & FPEXC_EX) {
+	if (fpexc & (FPEXC_EX | FPEXC_VV)) {
 		u32 len;
 
 		len = fpexc + (1 << FPEXC_LENGTH_BIT);
@@ -262,15 +288,15 @@ void VFP9_bounce(u32 trigger, u32 fpexc, struct pt_regs *regs)
 	 * FPEXC bounce reason, but this appears to be unreliable.
 	 * Emulate the bounced instruction instead.
 	 */
-	inst = fmrx(FPINST);
-	exceptions = vfp_emulate_instruction(inst, fpscr, regs);
+	exceptions = vfp_emulate_instruction(trigger, fpscr, regs);
 	if (exceptions)
-		vfp_raise_exceptions(exceptions, inst, orig_fpscr, regs);
+		vfp_raise_exceptions(exceptions, trigger, orig_fpscr, regs);
 
 	/*
-	 * If there isn't a second FP instruction, exit now.
+	 * If there isn't a second FP instruction, exit now. Note that
+	 * the FPEXC.FP2V bit is valid only if FPEXC.EX is 1.
 	 */
-	if (!(fpexc & FPEXC_FPV2))
+	if (fpexc ^ (FPEXC_EX | FPEXC_FP2V))
 		return;
 
 	/*
@@ -279,10 +305,9 @@ void VFP9_bounce(u32 trigger, u32 fpexc, struct pt_regs *regs)
 	 */
 	barrier();
 	trigger = fmrx(FPINST2);
-	orig_fpscr = fpscr = fmrx(FPSCR);
 
  emulate:
-	exceptions = vfp_emulate_instruction(trigger, fpscr, regs);
+	exceptions = vfp_emulate_instruction(trigger, orig_fpscr, regs);
 	if (exceptions)
 		vfp_raise_exceptions(exceptions, trigger, orig_fpscr, regs);
 }
@@ -306,16 +331,9 @@ static int __init vfp_init(void)
 {
 	unsigned int vfpsid;
 	unsigned int cpu_arch = cpu_architecture();
-	u32 access = 0;
 
-	if (cpu_arch >= CPU_ARCH_ARMv6) {
-		access = get_copro_access();
-
-		/*
-		 * Enable full access to VFP (cp10 and cp11)
-		 */
-		set_copro_access(access | CPACC_FULL(10) | CPACC_FULL(11));
-	}
+	if (cpu_arch >= CPU_ARCH_ARMv6)
+		vfp_enable(NULL);
 
 	/*
 	 * First check that there is a VFP that we can use.
@@ -329,15 +347,9 @@ static int __init vfp_init(void)
 	vfp_vector = vfp_null_entry;
 
 	printk(KERN_INFO "VFP support v0.3: ");
-	if (VFP_arch) {
+	if (VFP_arch)
 		printk("not present\n");
-
-		/*
-		 * Restore the copro access register.
-		 */
-		if (cpu_arch >= CPU_ARCH_ARMv6)
-			set_copro_access(access);
-	} else if (vfpsid & FPSID_NODOUBLE) {
+	else if (vfpsid & FPSID_NODOUBLE) {
 		printk("no double precision support\n");
 	} else {
 		smp_call_function(vfp_enable, NULL, 1, 1);
diff --git a/include/asm-arm/vfp.h b/include/asm-arm/vfp.h
index bd6be9d7f772..9d474d47b266 100644
--- a/include/asm-arm/vfp.h
+++ b/include/asm-arm/vfp.h
@@ -8,6 +8,8 @@
 #define FPSID			cr0
 #define FPSCR			cr1
 #define FPEXC			cr8
+#define FPINST			cr9
+#define FPINST2			cr10
 
 /* FPSID bits */
 #define FPSID_IMPLEMENTER_BIT	(24)
@@ -28,6 +30,19 @@
 /* FPEXC bits */
 #define FPEXC_EX		(1 << 31)
 #define FPEXC_EN		(1 << 30)
+#define FPEXC_DEX		(1 << 29)
+#define FPEXC_FP2V		(1 << 28)
+#define FPEXC_VV		(1 << 27)
+#define FPEXC_TFV		(1 << 26)
+#define FPEXC_LENGTH_BIT	(8)
+#define FPEXC_LENGTH_MASK	(7 << FPEXC_LENGTH_BIT)
+#define FPEXC_IDF		(1 << 7)
+#define FPEXC_IXF		(1 << 4)
+#define FPEXC_UFF		(1 << 3)
+#define FPEXC_OFF		(1 << 2)
+#define FPEXC_DZF		(1 << 1)
+#define FPEXC_IOF		(1 << 0)
+#define FPEXC_TRAP_MASK		(FPEXC_IDF|FPEXC_IXF|FPEXC_UFF|FPEXC_OFF|FPEXC_DZF|FPEXC_IOF)
 
 /* FPSCR bits */
 #define FPSCR_DEFAULT_NAN	(1<<25)
@@ -55,21 +70,6 @@
 #define FPSCR_IXC		(1<<4)
 #define FPSCR_IDC		(1<<7)
 
-/*
- * VFP9-S specific.
- */
-#define FPINST			cr9
-#define FPINST2			cr10
-
-/* FPEXC bits */
-#define FPEXC_FPV2		(1<<28)
-#define FPEXC_LENGTH_BIT	(8)
-#define FPEXC_LENGTH_MASK	(7 << FPEXC_LENGTH_BIT)
-#define FPEXC_INV		(1 << 7)
-#define FPEXC_UFC		(1 << 3)
-#define FPEXC_OFC		(1 << 2)
-#define FPEXC_IOC		(1 << 0)
-
 /* Bit patterns for decoding the packaged operation descriptors */
 #define VFPOPDESC_LENGTH_BIT	(9)
 #define VFPOPDESC_LENGTH_MASK	(0x07 << VFPOPDESC_LENGTH_BIT)
-- 
cgit v1.2.3


From 25ebee020bd34d1f4c5678538204f0b10bf9f6d5 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Tue, 25 Sep 2007 15:22:24 +0100
Subject: [ARM] 4583/1: ARMv7: Add VFPv3 support

This patch adds the support for VFPv3 (the kernel currently supports
VFPv2). The main difference is 32 double registers (compared to 16).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig            |  7 ++++++-
 arch/arm/vfp/vfp.h          |  4 ++++
 arch/arm/vfp/vfphw.S        | 22 ++++++++++++++++++----
 arch/arm/vfp/vfpinstr.h     |  6 +++---
 include/asm-arm/fpstate.h   |  9 +++++++--
 include/asm-arm/vfp.h       |  6 ++++++
 include/asm-arm/vfpmacros.h | 18 ++++++++++++++++--
 7 files changed, 60 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index a04f507e7f2c..f4eeb03bc6a9 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -951,7 +951,7 @@ config FPE_FASTFPE
 
 config VFP
 	bool "VFP-format floating point maths"
-	depends on CPU_V6 || CPU_ARM926T
+	depends on CPU_V6 || CPU_ARM926T || CPU_V7
 	help
 	  Say Y to include VFP support code in the kernel. This is needed
 	  if your hardware includes a VFP unit.
@@ -961,6 +961,11 @@ config VFP
 
 	  Say N if your target does not have VFP hardware.
 
+config VFPv3
+	bool
+	depends on VFP
+	default y if CPU_V7
+
 endmenu
 
 menu "Userspace binary formats"
diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h
index 791d0238c68f..c85860bad585 100644
--- a/arch/arm/vfp/vfp.h
+++ b/arch/arm/vfp/vfp.h
@@ -265,7 +265,11 @@ struct vfp_double {
  * which returns (double)0.0.  This is useful for the compare with
  * zero instructions.
  */
+#ifdef CONFIG_VFPv3
+#define VFP_REG_ZERO	32
+#else
 #define VFP_REG_ZERO	16
+#endif
 extern u64 vfp_get_double(unsigned int reg);
 extern void vfp_put_double(u64 val, unsigned int reg);
 
diff --git a/arch/arm/vfp/vfphw.S b/arch/arm/vfp/vfphw.S
index 53d9f8e8fac3..353f9e5c7919 100644
--- a/arch/arm/vfp/vfphw.S
+++ b/arch/arm/vfp/vfphw.S
@@ -99,12 +99,12 @@ vfp_support_entry:
 	DBGSTR1	"save old state %p", r4
 	cmp	r4, #0
 	beq	no_old_VFP_process
+	VFPFSTMIA r4, r5		@ save the working registers
 	VFPFMRX	r5, FPSCR		@ current status
 	tst	r1, #FPEXC_EX		@ is there additional state to save?
 	VFPFMRX	r6, FPINST, NE		@ FPINST (only if FPEXC.EX is set)
 	tstne	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
 	VFPFMRX	r8, FPINST2, NE		@ FPINST2 if needed (and present)
-	VFPFSTMIA r4 			@ save the working registers
 	stmia	r4, {r1, r5, r6, r8}	@ save FPEXC, FPSCR, FPINST, FPINST2
 					@ and point r4 at the word at the
 					@ start of the register dump
@@ -114,7 +114,7 @@ no_old_VFP_process:
 	DBGSTR1	"load state %p", r10
 	str	r10, [r3, r11, lsl #2]	@ update the last_VFP_context pointer
 					@ Load the saved state back into the VFP
-	VFPFLDMIA r10	 		@ reload the working registers while
+	VFPFLDMIA r10, r5		@ reload the working registers while
 					@ FPEXC is in a safe state
 	ldmia	r10, {r1, r5, r6, r8}	@ load FPEXC, FPSCR, FPINST, FPINST2
 	tst	r1, #FPEXC_EX		@ is there additional state to restore?
@@ -174,12 +174,12 @@ vfp_save_state:
 	@ r0 - save location
 	@ r1 - FPEXC
 	DBGSTR1	"save VFP state %p", r0
+	VFPFSTMIA r0, r2		@ save the working registers
 	VFPFMRX	r2, FPSCR		@ current status
 	tst	r1, #FPEXC_EX		@ is there additional state to save?
 	VFPFMRX	r3, FPINST, NE		@ FPINST (only if FPEXC.EX is set)
 	tstne	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
 	VFPFMRX	r12, FPINST2, NE	@ FPINST2 if needed (and present)
-	VFPFSTMIA r0 			@ save the working registers
 	stmia	r0, {r1, r2, r3, r12}	@ save FPEXC, FPSCR, FPINST, FPINST2
 	mov	pc, lr
 #endif
@@ -217,8 +217,15 @@ vfp_get_double:
 	fmrrd	r0, r1, d\dr
 	mov	pc, lr
 	.endr
+#ifdef CONFIG_VFPv3
+	@ d16 - d31 registers
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+	mrrc	p11, 3, r0, r1, c\dr	@ fmrrd	r0, r1, d\dr
+	mov	pc, lr
+	.endr
+#endif
 
-	@ virtual register 16 for compare with zero
+	@ virtual register 16 (or 32 if VFPv3) for compare with zero
 	mov	r0, #0
 	mov	r1, #0
 	mov	pc, lr
@@ -231,3 +238,10 @@ vfp_put_double:
 	fmdrr	d\dr, r0, r1
 	mov	pc, lr
 	.endr
+#ifdef CONFIG_VFPv3
+	@ d16 - d31 registers
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+	mcrr	p11, 3, r1, r2, c\dr	@ fmdrr	r1, r2, d\dr
+	mov	pc, lr
+	.endr
+#endif
diff --git a/arch/arm/vfp/vfpinstr.h b/arch/arm/vfp/vfpinstr.h
index 7f343a4beca0..15b95b5ab97e 100644
--- a/arch/arm/vfp/vfpinstr.h
+++ b/arch/arm/vfp/vfpinstr.h
@@ -52,11 +52,11 @@
 #define FEXT_TO_IDX(inst)	((inst & 0x000f0000) >> 15 | (inst & (1 << 7)) >> 7)
 
 #define vfp_get_sd(inst)	((inst & 0x0000f000) >> 11 | (inst & (1 << 22)) >> 22)
-#define vfp_get_dd(inst)	((inst & 0x0000f000) >> 12)
+#define vfp_get_dd(inst)	((inst & 0x0000f000) >> 12 | (inst & (1 << 22)) >> 18)
 #define vfp_get_sm(inst)	((inst & 0x0000000f) << 1 | (inst & (1 << 5)) >> 5)
-#define vfp_get_dm(inst)	((inst & 0x0000000f))
+#define vfp_get_dm(inst)	((inst & 0x0000000f) | (inst & (1 << 5)) >> 1)
 #define vfp_get_sn(inst)	((inst & 0x000f0000) >> 15 | (inst & (1 << 7)) >> 7)
-#define vfp_get_dn(inst)	((inst & 0x000f0000) >> 16)
+#define vfp_get_dn(inst)	((inst & 0x000f0000) >> 16 | (inst & (1 << 7)) >> 3)
 
 #define vfp_single(inst)	(((inst) & 0x0000f00) == 0xa00)
 
diff --git a/include/asm-arm/fpstate.h b/include/asm-arm/fpstate.h
index f31cda5a55ee..392eb5332323 100644
--- a/include/asm-arm/fpstate.h
+++ b/include/asm-arm/fpstate.h
@@ -17,14 +17,18 @@
 /*
  * VFP storage area has:
  *  - FPEXC, FPSCR, FPINST and FPINST2.
- *  - 16 double precision data registers
- *  - an implementation-dependant word of state for FLDMX/FSTMX
+ *  - 16 or 32 double precision data registers
+ *  - an implementation-dependant word of state for FLDMX/FSTMX (pre-ARMv6)
  * 
  *  FPEXC will always be non-zero once the VFP has been used in this process.
  */
 
 struct vfp_hard_struct {
+#ifdef CONFIG_VFPv3
+	__u64 fpregs[32];
+#else
 	__u64 fpregs[16];
+#endif
 #if __LINUX_ARM_ARCH__ < 6
 	__u32 fpmx_state;
 #endif
@@ -35,6 +39,7 @@ struct vfp_hard_struct {
 	 */
 	__u32 fpinst;
 	__u32 fpinst2;
+
 #ifdef CONFIG_SMP
 	__u32 cpu;
 #endif
diff --git a/include/asm-arm/vfp.h b/include/asm-arm/vfp.h
index 9d474d47b266..5f9a2cb3d452 100644
--- a/include/asm-arm/vfp.h
+++ b/include/asm-arm/vfp.h
@@ -7,6 +7,8 @@
 
 #define FPSID			cr0
 #define FPSCR			cr1
+#define MVFR1			cr6
+#define MVFR0			cr7
 #define FPEXC			cr8
 #define FPINST			cr9
 #define FPINST2			cr10
@@ -70,6 +72,10 @@
 #define FPSCR_IXC		(1<<4)
 #define FPSCR_IDC		(1<<7)
 
+/* MVFR0 bits */
+#define MVFR0_A_SIMD_BIT	(0)
+#define MVFR0_A_SIMD_MASK	(0xf << MVFR0_A_SIMD_BIT)
+
 /* Bit patterns for decoding the packaged operation descriptors */
 #define VFPOPDESC_LENGTH_BIT	(9)
 #define VFPOPDESC_LENGTH_MASK	(0x07 << VFPOPDESC_LENGTH_BIT)
diff --git a/include/asm-arm/vfpmacros.h b/include/asm-arm/vfpmacros.h
index 27fe028b4e72..cccb3892e73c 100644
--- a/include/asm-arm/vfpmacros.h
+++ b/include/asm-arm/vfpmacros.h
@@ -15,19 +15,33 @@
 	.endm
 
 	@ read all the working registers back into the VFP
-	.macro	VFPFLDMIA, base
+	.macro	VFPFLDMIA, base, tmp
 #if __LINUX_ARM_ARCH__ < 6
 	LDC	p11, cr0, [\base],#33*4		    @ FLDMIAX \base!, {d0-d15}
 #else
 	LDC	p11, cr0, [\base],#32*4		    @ FLDMIAD \base!, {d0-d15}
+#endif
+#ifdef CONFIG_VFPv3
+	VFPFMRX	\tmp, MVFR0			    @ Media and VFP Feature Register 0
+	and	\tmp, \tmp, #MVFR0_A_SIMD_MASK	    @ A_SIMD field
+	cmp	\tmp, #2			    @ 32 x 64bit registers?
+	ldceql	p11, cr0, [\base],#32*4		    @ FLDMIAD \base!, {d16-d31}
+	addne	\base, \base, #32*4		    @ step over unused register space
 #endif
 	.endm
 
 	@ write all the working registers out of the VFP
-	.macro	VFPFSTMIA, base
+	.macro	VFPFSTMIA, base, tmp
 #if __LINUX_ARM_ARCH__ < 6
 	STC	p11, cr0, [\base],#33*4		    @ FSTMIAX \base!, {d0-d15}
 #else
 	STC	p11, cr0, [\base],#32*4		    @ FSTMIAD \base!, {d0-d15}
+#endif
+#ifdef CONFIG_VFPv3
+	VFPFMRX	\tmp, MVFR0			    @ Media and VFP Feature Register 0
+	and	\tmp, \tmp, #MVFR0_A_SIMD_MASK	    @ A_SIMD field
+	cmp	\tmp, #2			    @ 32 x 64bit registers?
+	stceql	p11, cr0, [\base],#32*4		    @ FSTMIAD \base!, {d16-d31}
+	addne	\base, \base, #32*4		    @ step over unused register space
 #endif
 	.endm
-- 
cgit v1.2.3


From f1a6de9c7efc8e3fb535f1e61848f671f90b5cd5 Mon Sep 17 00:00:00 2001
From: Jan Altenberg <jan.altenberg@linutronix.de>
Date: Tue, 2 Oct 2007 13:26:43 -0700
Subject: [ARM] Remove at91_lcdc.h

include/asm-arm/arch-at91/at91_lcdc.h (which is still present in latest
git) has been superseeded by include/video/atmel_lcdc.h, so let's remove
it.

Signed-off-by: Jan Altenberg <jan.altenberg@linutronix.de>
Cc: Andrew Victor <andrew@sanpeople.com>
Cc: Nicolas Ferre <nicolas.ferre@rfo.atmel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-at91/at91_lcdc.h | 148 ----------------------------------
 1 file changed, 148 deletions(-)
 delete mode 100644 include/asm-arm/arch-at91/at91_lcdc.h

(limited to 'include')

diff --git a/include/asm-arm/arch-at91/at91_lcdc.h b/include/asm-arm/arch-at91/at91_lcdc.h
deleted file mode 100644
index ab040a40d37b..000000000000
--- a/include/asm-arm/arch-at91/at91_lcdc.h
+++ /dev/null
@@ -1,148 +0,0 @@
-/*
- * include/asm-arm/arch-at91/at91_lcdc.h
- *
- * LCD Controller (LCDC).
- * Based on AT91SAM9261 datasheet revision E.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#ifndef AT91_LCDC_H
-#define AT91_LCDC_H
-
-#define AT91_LCDC_DMABADDR1	0x00		/* DMA Base Address Register 1 */
-#define AT91_LCDC_DMABADDR2	0x04		/* DMA Base Address Register 2 */
-#define AT91_LCDC_DMAFRMPT1	0x08		/* DMA Frame Pointer Register 1 */
-#define AT91_LCDC_DMAFRMPT2	0x0c		/* DMA Frame Pointer Register 2 */
-#define AT91_LCDC_DMAFRMADD1	0x10		/* DMA Frame Address Register 1 */
-#define AT91_LCDC_DMAFRMADD2	0x14		/* DMA Frame Address Register 2 */
-
-#define AT91_LCDC_DMAFRMCFG	0x18		/* DMA Frame Configuration Register */
-#define		AT91_LCDC_FRSIZE	(0x7fffff <<  0)	/* Frame Size */
-#define		AT91_LCDC_BLENGTH	(0x7f     << 24)	/* Burst Length */
-
-#define AT91_LCDC_DMACON	0x1c		/* DMA Control Register */
-#define		AT91_LCDC_DMAEN		(0x1 << 0)	/* DMA Enable */
-#define		AT91_LCDC_DMARST	(0x1 << 1)	/* DMA Reset */
-#define		AT91_LCDC_DMABUSY	(0x1 << 2)	/* DMA Busy */
-
-#define AT91_LCDC_LCDCON1	0x0800		/* LCD Control Register 1 */
-#define		AT91_LCDC_BYPASS	(1     <<  0)	/* Bypass lcd_dotck divider */
-#define		AT91_LCDC_CLKVAL	(0x1ff << 12)	/* Clock Divider */
-#define		AT91_LCDC_LINCNT	(0x7ff << 21)	/* Line Counter */
-
-#define AT91_LCDC_LCDCON2	0x0804		/* LCD Control Register 2 */
-#define		AT91_LCDC_DISTYPE	(3 << 0)	/* Display Type */
-#define			AT91_LCDC_DISTYPE_STNMONO	(0 << 0)
-#define			AT91_LCDC_DISTYPE_STNCOLOR	(1 << 0)
-#define			AT91_LCDC_DISTYPE_TFT		(2 << 0)
-#define		AT91_LCDC_SCANMOD	(1 << 2)	/* Scan Mode */
-#define			AT91_LCDC_SCANMOD_SINGLE	(0 << 2)
-#define			AT91_LCDC_SCANMOD_DUAL		(1 << 2)
-#define		AT91_LCDC_IFWIDTH	(3 << 3)	/*Interface Width */
-#define			AT91_LCDC_IFWIDTH_4		(0 << 3)
-#define			AT91_LCDC_IFWIDTH_8		(1 << 3)
-#define			AT91_LCDC_IFWIDTH_16		(2 << 3)
-#define		AT91_LCDC_PIXELSIZE	(7 << 5)	/* Bits per pixel */
-#define			AT91_LCDC_PIXELSIZE_1		(0 << 5)
-#define			AT91_LCDC_PIXELSIZE_2		(1 << 5)
-#define			AT91_LCDC_PIXELSIZE_4		(2 << 5)
-#define			AT91_LCDC_PIXELSIZE_8		(3 << 5)
-#define			AT91_LCDC_PIXELSIZE_16		(4 << 5)
-#define			AT91_LCDC_PIXELSIZE_24		(5 << 5)
-#define		AT91_LCDC_INVVD		(1 << 8)	/* LCD Data polarity */
-#define			AT91_LCDC_INVVD_NORMAL		(0 << 8)
-#define			AT91_LCDC_INVVD_INVERTED	(1 << 8)
-#define		AT91_LCDC_INVFRAME	(1 << 9 )	/* LCD VSync polarity */
-#define			AT91_LCDC_INVFRAME_NORMAL	(0 << 9)
-#define			AT91_LCDC_INVFRAME_INVERTED	(1 << 9)
-#define		AT91_LCDC_INVLINE	(1 << 10)	/* LCD HSync polarity */
-#define			AT91_LCDC_INVLINE_NORMAL	(0 << 10)
-#define			AT91_LCDC_INVLINE_INVERTED	(1 << 10)
-#define		AT91_LCDC_INVCLK	(1 << 11)	/* LCD dotclk polarity */
-#define			AT91_LCDC_INVCLK_NORMAL		(0 << 11)
-#define			AT91_LCDC_INVCLK_INVERTED	(1 << 11)
-#define		AT91_LCDC_INVDVAL	(1 << 12)	/* LCD dval polarity */
-#define			AT91_LCDC_INVDVAL_NORMAL	(0 << 12)
-#define			AT91_LCDC_INVDVAL_INVERTED	(1 << 12)
-#define		AT91_LCDC_CLKMOD	(1 << 15)	/* LCD dotclk mode */
-#define			AT91_LCDC_CLKMOD_ACTIVEDISPLAY	(0 << 15)
-#define			AT91_LCDC_CLKMOD_ALWAYSACTIVE	(1 << 15)
-#define		AT91_LCDC_MEMOR		(1 << 31)	/* Memory Ordering Format */
-#define			AT91_LCDC_MEMOR_BIG		(0 << 31)
-#define			AT91_LCDC_MEMOR_LITTLE		(1 << 31)
-
-#define AT91_LCDC_TIM1		0x0808		/* LCD Timing Register 1 */
-#define		AT91_LCDC_VFP		(0xff <<  0)	/* Vertical Front Porch */
-#define		AT91_LCDC_VBP		(0xff <<  8)	/* Vertical Back Porch */
-#define		AT91_LCDC_VPW		(0x3f << 16)	/* Vertical Synchronization Pulse Width */
-#define		AT91_LCDC_VHDLY		(0xf  << 24)	/* Vertical to Horizontal Delay */
-
-#define AT91_LCDC_TIM2		0x080c		/* LCD Timing Register 2 */
-#define		AT91_LCDC_HBP		(0xff  <<  0)	/* Horizontal Back Porch */
-#define		AT91_LCDC_HPW		(0x3f  <<  8)	/* Horizontal Synchronization Pulse Width */
-#define		AT91_LCDC_HFP		(0x7ff << 21)	/* Horizontal Front Porch */
-
-#define AT91_LCDC_LCDFRMCFG	0x0810		/* LCD Frame Configuration Register */
-#define		AT91_LCDC_LINEVAL	(0x7ff <<  0)	/* Vertical Size of LCD Module */
-#define		AT91_LCDC_HOZVAL	(0x7ff << 21)	/* Horizontal Size of LCD Module */
-
-#define AT91_LCDC_FIFO		0x0814		/* LCD FIFO Register */
-#define		AT91_LCDC_FIFOTH	(0xffff)	/* FIFO Threshold */
-
-#define AT91_LCDC_DP1_2		0x081c		/* Dithering Pattern DP1_2 Register */
-#define AT91_LCDC_DP4_7		0x0820		/* Dithering Pattern DP4_7 Register */
-#define AT91_LCDC_DP3_5		0x0824		/* Dithering Pattern DP3_5 Register */
-#define AT91_LCDC_DP2_3		0x0828		/* Dithering Pattern DP2_3 Register */
-#define AT91_LCDC_DP5_7		0x082c		/* Dithering Pattern DP5_7 Register */
-#define AT91_LCDC_DP3_4		0x0830		/* Dithering Pattern DP3_4 Register */
-#define AT91_LCDC_DP4_5		0x0834		/* Dithering Pattern DP4_5 Register */
-#define AT91_LCDC_DP6_7		0x0838		/* Dithering Pattern DP6_7 Register */
-#define		AT91_LCDC_DP1_2_VAL	(0xff)
-#define		AT91_LCDC_DP4_7_VAL	(0xfffffff)
-#define		AT91_LCDC_DP3_5_VAL	(0xfffff)
-#define		AT91_LCDC_DP2_3_VAL	(0xfff)
-#define		AT91_LCDC_DP5_7_VAL	(0xfffffff)
-#define		AT91_LCDC_DP3_4_VAL	(0xffff)
-#define		AT91_LCDC_DP4_5_VAL	(0xfffff)
-#define		AT91_LCDC_DP6_7_VAL	(0xfffffff)
-
-#define AT91_LCDC_PWRCON	0x083c		/* Power Control Register */
-#define		AT91_LCDC_PWR		(1    <<  0)	/* LCD Module Power Control */
-#define		AT91_LCDC_GUARDT	(0x7f <<  1)	/* Delay in Frame Period */
-#define		AT91_LCDC_BUSY		(1    << 31)	/* LCD Busy */
-
-#define AT91_LCDC_CONTRAST_CTR	0x0840		/* Contrast Control Register */
-#define		AT91_LCDC_PS		(3 << 0)	/* Contrast Counter Prescaler */
-#define			AT91_LCDC_PS_DIV1		(0 << 0)
-#define			AT91_LCDC_PS_DIV2		(1 << 0)
-#define			AT91_LCDC_PS_DIV4		(2 << 0)
-#define			AT91_LCDC_PS_DIV8		(3 << 0)
-#define		AT91_LCDC_POL		(1 << 2)	/* Polarity of output Pulse */
-#define			AT91_LCDC_POL_NEGATIVE		(0 << 2)
-#define			AT91_LCDC_POL_POSITIVE		(1 << 2)
-#define		AT91_LCDC_ENA		(1 << 3)	/* PWM generator Control */
-#define			AT91_LCDC_ENA_PWMDISABLE	(0 << 3)
-#define			AT91_LCDC_ENA_PWMENABLE		(1 << 3)
-
-#define AT91_LCDC_CONTRAST_VAL	0x0844		/* Contrast Value Register */
-#define		AT91_LCDC_CVAL		(0xff)		/* PWM compare value */
-
-#define AT91_LCDC_IER		0x0848		/* Interrupt Enable Register */
-#define AT91_LCDC_IDR		0x084c		/* Interrupt Disable Register */
-#define AT91_LCDC_IMR		0x0850		/* Interrupt Mask Register */
-#define AT91_LCDC_ISR		0x0854		/* Interrupt Enable Register */
-#define AT91_LCDC_ICR		0x0858		/* Interrupt Clear Register */
-#define		AT91_LCDC_LNI		(1 << 0)	/* Line Interrupt */
-#define		AT91_LCDC_LSTLNI	(1 << 1)	/* Last Line Interrupt */
-#define		AT91_LCDC_EOFI		(1 << 2)	/* DMA End Of Frame Interrupt */
-#define		AT91_LCDC_UFLWI		(1 << 4)	/* FIFO Underflow Interrupt */
-#define		AT91_LCDC_OWRI		(1 << 5)	/* FIFO Overwrite Interrupt */
-#define		AT91_LCDC_MERI		(1 << 6)	/* DMA Memory Error Interrupt */
-
-#define AT91_LCDC_LUT_(n)	(0x0c00 + ((n)*4))	/* Palette Entry 0..255 */
-
-#endif
-- 
cgit v1.2.3


From d2936b1976a9c70d1cce4700c51f26d4b9495e9d Mon Sep 17 00:00:00 2001
From: Pavel Roskin <proski@gnu.org>
Date: Sun, 25 Nov 2007 02:12:39 +0100
Subject: [ARM] 4666/1: ixp4xx: fix sparse warnings in
 include/asm-arm/arch-ixp4xx/io.h

Don't lose __iomem in casts.  Use __force to cast __iomem addresses to
integers.  Use __force to cast u32 to __le32 and vice versa.

Signed-off-by: Pavel Roskin <proski@gnu.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp4xx/io.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-ixp4xx/io.h b/include/asm-arm/arch-ixp4xx/io.h
index eeeea90cd5a9..9c5d2357aff3 100644
--- a/include/asm-arm/arch-ixp4xx/io.h
+++ b/include/asm-arm/arch-ixp4xx/io.h
@@ -61,13 +61,13 @@ __ixp4xx_ioremap(unsigned long addr, size_t size, unsigned int mtype)
 	if((addr < PCIBIOS_MIN_MEM) || (addr > 0x4fffffff))
 		return __arm_ioremap(addr, size, mtype);
 
-	return (void *)addr;
+	return (void __iomem *)addr;
 }
 
 static inline void
 __ixp4xx_iounmap(void __iomem *addr)
 {
-	if ((u32)addr >= VMALLOC_START)
+	if ((__force u32)addr >= VMALLOC_START)
 		__iounmap(addr);
 }
 
@@ -141,9 +141,9 @@ __ixp4xx_writesw(volatile void __iomem *bus_addr, const u16 *vaddr, int count)
 static inline void 
 __ixp4xx_writel(u32 value, volatile void __iomem *p)
 {
-	u32 addr = (u32)p;
+	u32 addr = (__force u32)p;
 	if (addr >= VMALLOC_START) {
-		__raw_writel(value, addr);
+		__raw_writel(value, p);
 		return;
 	}
 
@@ -208,11 +208,11 @@ __ixp4xx_readsw(const volatile void __iomem *bus_addr, u16 *vaddr, u32 count)
 static inline unsigned long 
 __ixp4xx_readl(const volatile void __iomem *p)
 {
-	u32 addr = (u32)p;
+	u32 addr = (__force u32)p;
 	u32 data;
 
 	if (addr >= VMALLOC_START)
-		return __raw_readl(addr);
+		return __raw_readl(p);
 
 	if (ixp4xx_pci_read(addr, NP_CMD_MEMREAD, &data))
 		return 0xffffffff;
@@ -438,7 +438,7 @@ __ixp4xx_ioread32(const void __iomem *addr)
 		return	(unsigned int)__ixp4xx_inl(port & PIO_MASK);
 	else {
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
-		return le32_to_cpu(__raw_readl((u32)port));
+		return le32_to_cpu((__force __le32)__raw_readl(addr));
 #else
 		return (unsigned int)__ixp4xx_readl(addr);
 #endif
@@ -523,7 +523,7 @@ __ixp4xx_iowrite32(u32 value, void __iomem *addr)
 		__ixp4xx_outl(value, port & PIO_MASK);
 	else
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
-		__raw_writel(cpu_to_le32(value), port);
+		__raw_writel((u32 __force)cpu_to_le32(value), addr);
 #else
 		__ixp4xx_writel(value, addr);
 #endif
-- 
cgit v1.2.3


From 9ebbec27da6d7f4762b73985ac4929acf061d48b Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Tue, 20 Nov 2007 13:34:56 +0100
Subject: [ARM] 4657/1: AT91: Header definition update

Add definitions of registers / bits found on some AT91SAM9 processors

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-at91/at91_pmc.h           |  4 +++-
 include/asm-arm/arch-at91/at91_twi.h           | 11 +++++++++++
 include/asm-arm/arch-at91/at91sam9260_matrix.h |  2 +-
 3 files changed, 15 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-at91/at91_pmc.h b/include/asm-arm/arch-at91/at91_pmc.h
index 33ff5b6798ee..34804eaab4f2 100644
--- a/include/asm-arm/arch-at91/at91_pmc.h
+++ b/include/asm-arm/arch-at91/at91_pmc.h
@@ -37,7 +37,9 @@
 #define	AT91_PMC_PCDR		(AT91_PMC + 0x14)	/* Peripheral Clock Disable Register */
 #define	AT91_PMC_PCSR		(AT91_PMC + 0x18)	/* Peripheral Clock Status Register */
 
-#define	AT91_CKGR_MOR		(AT91_PMC + 0x20)	/* Main Oscillator Register */
+#define	AT91_CKGR_UCKR		(AT91_PMC + 0x1C)	/* UTMI Clock Register [SAM9RL only] */
+
+#define	AT91_CKGR_MOR		(AT91_PMC + 0x20)	/* Main Oscillator Register [not on SAM9RL] */
 #define		AT91_PMC_MOSCEN		(1    << 0)		/* Main Oscillator Enable */
 #define		AT91_PMC_OSCBYPASS	(1    << 1)		/* Oscillator Bypass [AT91SAM926x only] */
 #define		AT91_PMC_OSCOUNT	(0xff << 8)		/* Main Oscillator Start-up Time */
diff --git a/include/asm-arm/arch-at91/at91_twi.h b/include/asm-arm/arch-at91/at91_twi.h
index ca9a90733456..f9f2e3cd95c5 100644
--- a/include/asm-arm/arch-at91/at91_twi.h
+++ b/include/asm-arm/arch-at91/at91_twi.h
@@ -21,6 +21,8 @@
 #define		AT91_TWI_STOP		(1 <<  1)	/* Send a Stop Condition */
 #define		AT91_TWI_MSEN		(1 <<  2)	/* Master Transfer Enable */
 #define		AT91_TWI_MSDIS		(1 <<  3)	/* Master Transfer Disable */
+#define		AT91_TWI_SVEN		(1 <<  4)	/* Slave Transfer Enable [SAM9260 only] */
+#define		AT91_TWI_SVDIS		(1 <<  5)	/* Slave Transfer Disable [SAM9260 only] */
 #define		AT91_TWI_SWRST		(1 <<  7)	/* Software Reset */
 
 #define	AT91_TWI_MMR		0x04		/* Master Mode Register */
@@ -32,6 +34,9 @@
 #define		AT91_TWI_MREAD		(1    << 12)	/* Master Read Direction */
 #define		AT91_TWI_DADR		(0x7f << 16)	/* Device Address */
 
+#define	AT91_TWI_SMR		0x08		/* Slave Mode Register [SAM9260 only] */
+#define		AT91_TWI_SADR		(0x7f << 16)	/* Slave Address */
+
 #define	AT91_TWI_IADR		0x0c		/* Internal Address Register */
 
 #define	AT91_TWI_CWGR		0x10		/* Clock Waveform Generator Register */
@@ -43,9 +48,15 @@
 #define		AT91_TWI_TXCOMP		(1 <<  0)	/* Transmission Complete */
 #define		AT91_TWI_RXRDY		(1 <<  1)	/* Receive Holding Register Ready */
 #define		AT91_TWI_TXRDY		(1 <<  2)	/* Transmit Holding Register Ready */
+#define		AT91_TWI_SVREAD		(1 <<  3)	/* Slave Read [SAM9260 only] */
+#define		AT91_TWI_SVACC		(1 <<  4)	/* Slave Access [SAM9260 only] */
+#define		AT91_TWI_GACC		(1 <<  5)	/* General Call Access [SAM9260 only] */
 #define		AT91_TWI_OVRE		(1 <<  6)	/* Overrun Error [AT91RM9200 only] */
 #define		AT91_TWI_UNRE		(1 <<  7)	/* Underrun Error [AT91RM9200 only] */
 #define		AT91_TWI_NACK		(1 <<  8)	/* Not Acknowledged */
+#define		AT91_TWI_ARBLST		(1 <<  9)	/* Arbitration Lost [SAM9260 only] */
+#define		AT91_TWI_SCLWS		(1 << 10)	/* Clock Wait State [SAM9260 only] */
+#define		AT91_TWI_EOSACC		(1 << 11)	/* End of Slave Address [SAM9260 only] */
 
 #define	AT91_TWI_IER		0x24		/* Interrupt Enable Register */
 #define	AT91_TWI_IDR		0x28		/* Interrupt Disable Register */
diff --git a/include/asm-arm/arch-at91/at91sam9260_matrix.h b/include/asm-arm/arch-at91/at91sam9260_matrix.h
index aacb1e976422..a8e9fec6c735 100644
--- a/include/asm-arm/arch-at91/at91sam9260_matrix.h
+++ b/include/asm-arm/arch-at91/at91sam9260_matrix.h
@@ -67,7 +67,7 @@
 #define		AT91_MATRIX_CS4A		(1 << 4)	/* Chip Select 4 Assignment */
 #define			AT91_MATRIX_CS4A_SMC		(0 << 4)
 #define			AT91_MATRIX_CS4A_SMC_CF1	(1 << 4)
-#define		AT91_MATRIX_CS5A		(1 << 5 )	/* Chip Select 5 Assignment */
+#define		AT91_MATRIX_CS5A		(1 << 5)	/* Chip Select 5 Assignment */
 #define			AT91_MATRIX_CS5A_SMC		(0 << 5)
 #define			AT91_MATRIX_CS5A_SMC_CF2	(1 << 5)
 #define		AT91_MATRIX_DBPUC		(1 << 8)	/* Data Bus Pull-up Configuration */
-- 
cgit v1.2.3


From 156864f806baa4e1aa6eabd28ac45ecc92b31315 Mon Sep 17 00:00:00 2001
From: Marc Pignat <marc.pignat@hevs.ch>
Date: Mon, 3 Dec 2007 12:58:24 +0100
Subject: [ARM] 4688/1: at91: speed-up irq processing

Save N instructions on every interrupt processing (where N is the
number of interrupts processed in any one IRQ exception).

Signed-off-by: Marc Pignat <marc.pignat@hevs.ch>
Acked-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-at91/entry-macro.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-at91/entry-macro.S b/include/asm-arm/arch-at91/entry-macro.S
index cc1d850a0788..1005eee6219b 100644
--- a/include/asm-arm/arch-at91/entry-macro.S
+++ b/include/asm-arm/arch-at91/entry-macro.S
@@ -17,13 +17,13 @@
 	.endm
 
 	.macro  get_irqnr_preamble, base, tmp
+	ldr	\base, =(AT91_VA_BASE_SYS + AT91_AIC)		@ base virtual address of AIC peripheral
 	.endm
 
 	.macro  arch_ret_to_user, tmp1, tmp2
 	.endm
 
 	.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
-	ldr	\base, =(AT91_VA_BASE_SYS + AT91_AIC)		@ base virtual address of AIC peripheral
 	ldr	\irqnr, [\base, #(AT91_AIC_IVR - AT91_AIC)]	@ read IRQ vector register: de-asserts nIRQ to processor (and clears interrupt)
 	ldr	\irqstat, [\base, #(AT91_AIC_ISR - AT91_AIC)]	@ read interrupt source number
 	teq	\irqstat, #0					@ ISR is 0 when no current interrupt, or spurious interrupt
-- 
cgit v1.2.3


From e2920802351b3e01a3e70a26bb1bbb22f29da4cb Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Tue, 22 Jan 2008 11:43:26 +0100
Subject: [ARM] 4751/1: [AT91] ISI peripheral on SAM9263

Add support for the Image Sensor Interface (ISI) peripheral integrated
in the Atmel AT91SAM9263 processor.

Patch from MaLiK

Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/at91sam9263_devices.c | 50 ++++++++++++++++++++++++++++++++
 include/asm-arm/arch-at91/board.h        |  3 ++
 2 files changed, 53 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index ac329a98e959..d50a8dfb916d 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -731,6 +731,56 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  Image Sensor Interface
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_VIDEO_AT91_ISI) || defined(CONFIG_VIDEO_AT91_ISI_MODULE)
+
+struct resource isi_resources[] = {
+	[0] = {
+		.start	= AT91SAM9263_BASE_ISI,
+		.end	= AT91SAM9263_BASE_ISI + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9263_ID_ISI,
+		.end	= AT91SAM9263_ID_ISI,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9263_isi_device = {
+	.name		= "at91_isi",
+	.id		= -1,
+	.resource	= isi_resources,
+	.num_resources	= ARRAY_SIZE(isi_resources),
+};
+
+void __init at91_add_device_isi(void)
+{
+	at91_set_A_periph(AT91_PIN_PE0, 0);	/* ISI_D0 */
+	at91_set_A_periph(AT91_PIN_PE1, 0);	/* ISI_D1 */
+	at91_set_A_periph(AT91_PIN_PE2, 0);	/* ISI_D2 */
+	at91_set_A_periph(AT91_PIN_PE3, 0);	/* ISI_D3 */
+	at91_set_A_periph(AT91_PIN_PE4, 0);	/* ISI_D4 */
+	at91_set_A_periph(AT91_PIN_PE5, 0);	/* ISI_D5 */
+	at91_set_A_periph(AT91_PIN_PE6, 0);	/* ISI_D6 */
+	at91_set_A_periph(AT91_PIN_PE7, 0);	/* ISI_D7 */
+	at91_set_A_periph(AT91_PIN_PE8, 0);	/* ISI_PCK */
+	at91_set_A_periph(AT91_PIN_PE9, 0);	/* ISI_HSYNC */
+	at91_set_A_periph(AT91_PIN_PE10, 0);	/* ISI_VSYNC */
+	at91_set_B_periph(AT91_PIN_PE11, 0);	/* ISI_MCK (PCK3) */
+	at91_set_B_periph(AT91_PIN_PE12, 0);	/* ISI_PD8 */
+	at91_set_B_periph(AT91_PIN_PE13, 0);	/* ISI_PD9 */
+	at91_set_B_periph(AT91_PIN_PE14, 0);	/* ISI_PD10 */
+	at91_set_B_periph(AT91_PIN_PE15, 0);	/* ISI_PD11 */
+}
+#else
+void __init at91_add_device_isi(void) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  LEDs
  * -------------------------------------------------------------------- */
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index 79054965baa6..762148cc48c1 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -126,6 +126,9 @@ struct atmel_ac97_data {
 };
 extern void __init at91_add_device_ac97(struct atmel_ac97_data *data);
 
+ /* ISI */
+extern void __init at91_add_device_isi(void);
+
  /* LEDs */
 extern u8 at91_leds_cpu;
 extern u8 at91_leds_timer;
-- 
cgit v1.2.3


From 884f5a6a8dc8646d5904f166adf3c558be57e1ab Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Wed, 23 Jan 2008 09:11:13 +0100
Subject: [ARM] 4752/1: [AT91] RTT, RTC and WDT peripherals on SAM9

Add platform_device and initialization for the RTT (Real Time Timer)
and WDT (Watchdog) integrated in the Atmel AT91SAM9 processors.

For SAM9263, register both RTT peripherals.
   [From: David Brownell <dbrownell@users.sourceforge.net>]

Provide platform_resources for RTT peripherals
  [From: David Brownell <dbrownell@users.sourceforge.net>]

Add support for RTC peripheral on AT91SAM9RL (same RTC peripherals as
AT91RM9200)
  [From: David Brownell <dbrownell@users.sourceforge.net>]

Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/at91sam9260_devices.c | 47 ++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9261_devices.c | 47 ++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9263_devices.c | 63 +++++++++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9rl_devices.c  | 68 ++++++++++++++++++++++++++++++++
 include/asm-arm/arch-at91/at91_rtt.h     |  8 ++--
 5 files changed, 229 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index 3091bf47d8c9..1533556fb873 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -538,6 +538,51 @@ void __init at91_add_device_spi(struct spi_board_info *devices, int nr_devices)
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  RTT
+ * -------------------------------------------------------------------- */
+
+static struct resource rtt_resources[] = {
+	{
+		.start	= AT91_BASE_SYS + AT91_RTT,
+		.end	= AT91_BASE_SYS + AT91_RTT + SZ_16 - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91sam9260_rtt_device = {
+	.name		= "at91_rtt",
+	.id		= -1,
+	.resource	= rtt_resources,
+	.num_resources	= ARRAY_SIZE(rtt_resources),
+};
+
+static void __init at91_add_device_rtt(void)
+{
+	platform_device_register(&at91sam9260_rtt_device);
+}
+
+
+/* --------------------------------------------------------------------
+ *  Watchdog
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_AT91SAM9_WATCHDOG) || defined(CONFIG_AT91SAM9_WATCHDOG_MODULE)
+static struct platform_device at91sam9260_wdt_device = {
+	.name		= "at91_wdt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_watchdog(void)
+{
+	platform_device_register(&at91sam9260_wdt_device);
+}
+#else
+static void __init at91_add_device_watchdog(void) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  LEDs
  * -------------------------------------------------------------------- */
@@ -898,6 +943,8 @@ void __init at91_add_device_serial(void) {}
  */
 static int __init at91_add_standard_devices(void)
 {
+	at91_add_device_rtt();
+	at91_add_device_watchdog();
 	return 0;
 }
 
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index 1d03844a607a..17f3f6a24796 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -550,6 +550,51 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  RTT
+ * -------------------------------------------------------------------- */
+
+static struct resource rtt_resources[] = {
+	{
+		.start	= AT91_BASE_SYS + AT91_RTT,
+		.end	= AT91_BASE_SYS + AT91_RTT + SZ_16 - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91sam9261_rtt_device = {
+	.name		= "at91_rtt",
+	.id		= -1,
+	.resource	= rtt_resources,
+	.num_resources	= ARRAY_SIZE(rtt_resources),
+};
+
+static void __init at91_add_device_rtt(void)
+{
+	platform_device_register(&at91sam9261_rtt_device);
+}
+
+
+/* --------------------------------------------------------------------
+ *  Watchdog
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_AT91SAM9_WATCHDOG) || defined(CONFIG_AT91SAM9_WATCHDOG_MODULE)
+static struct platform_device at91sam9261_wdt_device = {
+	.name		= "at91_wdt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_watchdog(void)
+{
+	platform_device_register(&at91sam9261_wdt_device);
+}
+#else
+static void __init at91_add_device_watchdog(void) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  LEDs
  * -------------------------------------------------------------------- */
@@ -786,6 +831,8 @@ void __init at91_add_device_serial(void) {}
  */
 static int __init at91_add_standard_devices(void)
 {
+	at91_add_device_rtt();
+	at91_add_device_watchdog();
 	return 0;
 }
 
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index d50a8dfb916d..b4eb21769db9 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -781,6 +781,67 @@ void __init at91_add_device_isi(void) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  RTT
+ * -------------------------------------------------------------------- */
+
+static struct resource rtt0_resources[] = {
+	{
+		.start	= AT91_BASE_SYS + AT91_RTT0,
+		.end	= AT91_BASE_SYS + AT91_RTT0 + SZ_16 - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91sam9263_rtt0_device = {
+	.name		= "at91_rtt",
+	.id		= 0,
+	.resource	= rtt0_resources,
+	.num_resources	= ARRAY_SIZE(rtt0_resources),
+};
+
+static struct resource rtt1_resources[] = {
+	{
+		.start	= AT91_BASE_SYS + AT91_RTT1,
+		.end	= AT91_BASE_SYS + AT91_RTT1 + SZ_16 - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91sam9263_rtt1_device = {
+	.name		= "at91_rtt",
+	.id		= 1,
+	.resource	= rtt1_resources,
+	.num_resources	= ARRAY_SIZE(rtt1_resources),
+};
+
+static void __init at91_add_device_rtt(void)
+{
+	platform_device_register(&at91sam9263_rtt0_device);
+	platform_device_register(&at91sam9263_rtt1_device);
+}
+
+
+/* --------------------------------------------------------------------
+ *  Watchdog
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_AT91SAM9_WATCHDOG) || defined(CONFIG_AT91SAM9_WATCHDOG_MODULE)
+static struct platform_device at91sam9263_wdt_device = {
+	.name		= "at91_wdt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_watchdog(void)
+{
+	platform_device_register(&at91sam9263_wdt_device);
+}
+#else
+static void __init at91_add_device_watchdog(void) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  LEDs
  * -------------------------------------------------------------------- */
@@ -1021,6 +1082,8 @@ void __init at91_add_device_serial(void) {}
  */
 static int __init at91_add_standard_devices(void)
 {
+	at91_add_device_rtt();
+	at91_add_device_watchdog();
 	return 0;
 }
 
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index 2bd60a3dc623..19ca71d3faab 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -383,6 +383,71 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  RTC
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_RTC_DRV_AT91RM9200) || defined(CONFIG_RTC_DRV_AT91RM9200_MODULE)
+static struct platform_device at91sam9rl_rtc_device = {
+	.name		= "at91_rtc",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_rtc(void)
+{
+	platform_device_register(&at91sam9rl_rtc_device);
+}
+#else
+static void __init at91_add_device_rtc(void) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  RTT
+ * -------------------------------------------------------------------- */
+
+static struct resource rtt_resources[] = {
+	{
+		.start	= AT91_BASE_SYS + AT91_RTT,
+		.end	= AT91_BASE_SYS + AT91_RTT + SZ_16 - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91sam9rl_rtt_device = {
+	.name		= "at91_rtt",
+	.id		= -1,
+	.resource	= rtt_resources,
+	.num_resources	= ARRAY_SIZE(rtt_resources),
+};
+
+static void __init at91_add_device_rtt(void)
+{
+	platform_device_register(&at91sam9rl_rtt_device);
+}
+
+
+/* --------------------------------------------------------------------
+ *  Watchdog
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_AT91SAM9_WATCHDOG) || defined(CONFIG_AT91SAM9_WATCHDOG_MODULE)
+static struct platform_device at91sam9rl_wdt_device = {
+	.name		= "at91_wdt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_watchdog(void)
+{
+	platform_device_register(&at91sam9rl_wdt_device);
+}
+#else
+static void __init at91_add_device_watchdog(void) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  LEDs
  * -------------------------------------------------------------------- */
@@ -659,6 +724,9 @@ void __init at91_add_device_serial(void) {}
  */
 static int __init at91_add_standard_devices(void)
 {
+	at91_add_device_rtc();
+	at91_add_device_rtt();
+	at91_add_device_watchdog();
 	return 0;
 }
 
diff --git a/include/asm-arm/arch-at91/at91_rtt.h b/include/asm-arm/arch-at91/at91_rtt.h
index bae1103fbbb2..39a32633b275 100644
--- a/include/asm-arm/arch-at91/at91_rtt.h
+++ b/include/asm-arm/arch-at91/at91_rtt.h
@@ -13,19 +13,19 @@
 #ifndef AT91_RTT_H
 #define AT91_RTT_H
 
-#define AT91_RTT_MR		(AT91_RTT + 0x00)	/* Real-time Mode Register */
+#define AT91_RTT_MR		0x00			/* Real-time Mode Register */
 #define		AT91_RTT_RTPRES		(0xffff << 0)		/* Real-time Timer Prescaler Value */
 #define		AT91_RTT_ALMIEN		(1 << 16)		/* Alarm Interrupt Enable */
 #define		AT91_RTT_RTTINCIEN	(1 << 17)		/* Real Time Timer Increment Interrupt Enable */
 #define		AT91_RTT_RTTRST		(1 << 18)		/* Real Time Timer Restart */
 
-#define AT91_RTT_AR		(AT91_RTT + 0x04)	/* Real-time Alarm Register */
+#define AT91_RTT_AR		0x04			/* Real-time Alarm Register */
 #define		AT91_RTT_ALMV		(0xffffffff)		/* Alarm Value */
 
-#define AT91_RTT_VR		(AT91_RTT + 0x08)	/* Real-time Value Register */
+#define AT91_RTT_VR		0x08			/* Real-time Value Register */
 #define		AT91_RTT_CRTV		(0xffffffff)		/* Current Real-time Value */
 
-#define AT91_RTT_SR		(AT91_RTT + 0x0c)	/* Real-time Status Register */
+#define AT91_RTT_SR		0x0c			/* Real-time Status Register */
 #define		AT91_RTT_ALMS		(1 << 0)		/* Real-time Alarm Status */
 #define		AT91_RTT_RTTINC		(1 << 1)		/* Real-time Timer Increment */
 
-- 
cgit v1.2.3


From bfbc32663d4846039f88c0eccc1956587d89c042 Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Wed, 23 Jan 2008 09:18:06 +0100
Subject: [ARM] 4754/1: [AT91] SSC library support

Core support of the Atmel SSC library for all Atmel AT91 processors.

Based on David Brownell's initial patch for the AT91RM9200.

Signed-off-by: Andrew Victor <linux@maxim.org.za>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/at91rm9200_devices.c  | 173 +++++++++++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9260_devices.c |  79 ++++++++++++++
 arch/arm/mach-at91/at91sam9261_devices.c | 173 +++++++++++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9263_devices.c | 129 +++++++++++++++++++++++
 arch/arm/mach-at91/at91sam9rl_devices.c  | 129 +++++++++++++++++++++++
 include/asm-arm/arch-at91/board.h        |  17 +++
 6 files changed, 700 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-at91/at91rm9200_devices.c b/arch/arm/mach-at91/at91rm9200_devices.c
index a601c7c1670e..cebea4168778 100644
--- a/arch/arm/mach-at91/at91rm9200_devices.c
+++ b/arch/arm/mach-at91/at91rm9200_devices.c
@@ -635,6 +635,179 @@ void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc0_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc0_resources[] = {
+	[0] = {
+		.start	= AT91RM9200_BASE_SSC0,
+		.end	= AT91RM9200_BASE_SSC0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91RM9200_ID_SSC0,
+		.end	= AT91RM9200_ID_SSC0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91rm9200_ssc0_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc0_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc0_resources,
+	.num_resources	= ARRAY_SIZE(ssc0_resources),
+};
+
+static inline void configure_ssc0_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB0, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB1, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB2, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB3, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB4, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB5, 1);
+}
+
+static u64 ssc1_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc1_resources[] = {
+	[0] = {
+		.start	= AT91RM9200_BASE_SSC1,
+		.end	= AT91RM9200_BASE_SSC1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91RM9200_ID_SSC1,
+		.end	= AT91RM9200_ID_SSC1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91rm9200_ssc1_device = {
+	.name	= "ssc",
+	.id	= 1,
+	.dev	= {
+		.dma_mask		= &ssc1_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc1_resources,
+	.num_resources	= ARRAY_SIZE(ssc1_resources),
+};
+
+static inline void configure_ssc1_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB6, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB7, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB8, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB9, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB10, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB11, 1);
+}
+
+static u64 ssc2_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc2_resources[] = {
+	[0] = {
+		.start	= AT91RM9200_BASE_SSC2,
+		.end	= AT91RM9200_BASE_SSC2 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91RM9200_ID_SSC2,
+		.end	= AT91RM9200_ID_SSC2,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91rm9200_ssc2_device = {
+	.name	= "ssc",
+	.id	= 2,
+	.dev	= {
+		.dma_mask		= &ssc2_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc2_resources,
+	.num_resources	= ARRAY_SIZE(ssc2_resources),
+};
+
+static inline void configure_ssc2_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB12, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB13, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB14, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB15, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB16, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB17, 1);
+}
+
+/*
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91RM9200_ID_SSC0:
+		pdev = &at91rm9200_ssc0_device;
+		configure_ssc0_pins(pins);
+		at91_clock_associate("ssc0_clk", &pdev->dev, "ssc");
+		break;
+	case AT91RM9200_ID_SSC1:
+		pdev = &at91rm9200_ssc1_device;
+		configure_ssc1_pins(pins);
+		at91_clock_associate("ssc1_clk", &pdev->dev, "ssc");
+		break;
+	case AT91RM9200_ID_SSC2:
+		pdev = &at91rm9200_ssc2_device;
+		configure_ssc2_pins(pins);
+		at91_clock_associate("ssc2_clk", &pdev->dev, "ssc");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  UART
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index c34fde362ad4..2e3db137bc49 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -606,6 +606,85 @@ void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc_resources[] = {
+	[0] = {
+		.start	= AT91SAM9260_BASE_SSC,
+		.end	= AT91SAM9260_BASE_SSC + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9260_ID_SSC,
+		.end	= AT91SAM9260_ID_SSC,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9260_ssc_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc_resources,
+	.num_resources	= ARRAY_SIZE(ssc_resources),
+};
+
+static inline void configure_ssc_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB17, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB16, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB18, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB19, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB20, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB21, 1);
+}
+
+/*
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91SAM9260_ID_SSC:
+		pdev = &at91sam9260_ssc_device;
+		configure_ssc_pins(pins);
+		at91_clock_associate("ssc_clk", &pdev->dev, "pclk");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  UART
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index f7f02d15045d..266ccf01be53 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -618,6 +618,179 @@ void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc0_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc0_resources[] = {
+	[0] = {
+		.start	= AT91SAM9261_BASE_SSC0,
+		.end	= AT91SAM9261_BASE_SSC0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9261_ID_SSC0,
+		.end	= AT91SAM9261_ID_SSC0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9261_ssc0_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc0_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc0_resources,
+	.num_resources	= ARRAY_SIZE(ssc0_resources),
+};
+
+static inline void configure_ssc0_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB21, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB22, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB23, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB24, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB25, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB26, 1);
+}
+
+static u64 ssc1_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc1_resources[] = {
+	[0] = {
+		.start	= AT91SAM9261_BASE_SSC1,
+		.end	= AT91SAM9261_BASE_SSC1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9261_ID_SSC1,
+		.end	= AT91SAM9261_ID_SSC1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9261_ssc1_device = {
+	.name	= "ssc",
+	.id	= 1,
+	.dev	= {
+		.dma_mask		= &ssc1_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc1_resources,
+	.num_resources	= ARRAY_SIZE(ssc1_resources),
+};
+
+static inline void configure_ssc1_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_B_periph(AT91_PIN_PA17, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_B_periph(AT91_PIN_PA18, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_B_periph(AT91_PIN_PA19, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_B_periph(AT91_PIN_PA20, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_B_periph(AT91_PIN_PA21, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_B_periph(AT91_PIN_PA22, 1);
+}
+
+static u64 ssc2_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc2_resources[] = {
+	[0] = {
+		.start	= AT91SAM9261_BASE_SSC2,
+		.end	= AT91SAM9261_BASE_SSC2 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9261_ID_SSC2,
+		.end	= AT91SAM9261_ID_SSC2,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9261_ssc2_device = {
+	.name	= "ssc",
+	.id	= 2,
+	.dev	= {
+		.dma_mask		= &ssc2_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc2_resources,
+	.num_resources	= ARRAY_SIZE(ssc2_resources),
+};
+
+static inline void configure_ssc2_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_B_periph(AT91_PIN_PC25, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_B_periph(AT91_PIN_PC26, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_B_periph(AT91_PIN_PC27, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_B_periph(AT91_PIN_PC28, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_B_periph(AT91_PIN_PC29, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_B_periph(AT91_PIN_PC30, 1);
+}
+
+/*
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91SAM9261_ID_SSC0:
+		pdev = &at91sam9261_ssc0_device;
+		configure_ssc0_pins(pins);
+		at91_clock_associate("ssc0_clk", &pdev->dev, "pclk");
+		break;
+	case AT91SAM9261_ID_SSC1:
+		pdev = &at91sam9261_ssc1_device;
+		configure_ssc1_pins(pins);
+		at91_clock_associate("ssc1_clk", &pdev->dev, "pclk");
+		break;
+	case AT91SAM9261_ID_SSC2:
+		pdev = &at91sam9261_ssc2_device;
+		configure_ssc2_pins(pins);
+		at91_clock_associate("ssc2_clk", &pdev->dev, "pclk");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  UART
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index 9bdcf45defcf..f83ef62f2385 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -865,6 +865,135 @@ void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc0_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc0_resources[] = {
+	[0] = {
+		.start	= AT91SAM9263_BASE_SSC0,
+		.end	= AT91SAM9263_BASE_SSC0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9263_ID_SSC0,
+		.end	= AT91SAM9263_ID_SSC0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9263_ssc0_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc0_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc0_resources,
+	.num_resources	= ARRAY_SIZE(ssc0_resources),
+};
+
+static inline void configure_ssc0_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_B_periph(AT91_PIN_PB0, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_B_periph(AT91_PIN_PB1, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_B_periph(AT91_PIN_PB2, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_B_periph(AT91_PIN_PB3, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_B_periph(AT91_PIN_PB4, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_B_periph(AT91_PIN_PB5, 1);
+}
+
+static u64 ssc1_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc1_resources[] = {
+	[0] = {
+		.start	= AT91SAM9263_BASE_SSC1,
+		.end	= AT91SAM9263_BASE_SSC1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9263_ID_SSC1,
+		.end	= AT91SAM9263_ID_SSC1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9263_ssc1_device = {
+	.name	= "ssc",
+	.id	= 1,
+	.dev	= {
+		.dma_mask		= &ssc1_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc1_resources,
+	.num_resources	= ARRAY_SIZE(ssc1_resources),
+};
+
+static inline void configure_ssc1_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB6, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB7, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB8, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB9, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB10, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB11, 1);
+}
+
+/*
+ * Return the device node so that board init code can use it as the
+ * parent for the device node reflecting how it's used on this board.
+ *
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91SAM9263_ID_SSC0:
+		pdev = &at91sam9263_ssc0_device;
+		configure_ssc0_pins(pins);
+		at91_clock_associate("ssc0_clk", &pdev->dev, "pclk");
+		break;
+	case AT91SAM9263_ID_SSC1:
+		pdev = &at91sam9263_ssc1_device;
+		configure_ssc1_pins(pins);
+		at91_clock_associate("ssc1_clk", &pdev->dev, "pclk");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  UART
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index bec8066f9ba1..e886cdfe919d 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -471,6 +471,135 @@ void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
 #endif
 
 
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc0_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc0_resources[] = {
+	[0] = {
+		.start	= AT91SAM9RL_BASE_SSC0,
+		.end	= AT91SAM9RL_BASE_SSC0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9RL_ID_SSC0,
+		.end	= AT91SAM9RL_ID_SSC0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9rl_ssc0_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc0_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc0_resources,
+	.num_resources	= ARRAY_SIZE(ssc0_resources),
+};
+
+static inline void configure_ssc0_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PC0, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PC1, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PA15, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PA16, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_B_periph(AT91_PIN_PA10, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_B_periph(AT91_PIN_PA22, 1);
+}
+
+static u64 ssc1_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc1_resources[] = {
+	[0] = {
+		.start	= AT91SAM9RL_BASE_SSC1,
+		.end	= AT91SAM9RL_BASE_SSC1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91SAM9RL_ID_SSC1,
+		.end	= AT91SAM9RL_ID_SSC1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91sam9rl_ssc1_device = {
+	.name	= "ssc",
+	.id	= 1,
+	.dev	= {
+		.dma_mask		= &ssc1_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc1_resources,
+	.num_resources	= ARRAY_SIZE(ssc1_resources),
+};
+
+static inline void configure_ssc1_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_B_periph(AT91_PIN_PA29, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_B_periph(AT91_PIN_PA30, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_B_periph(AT91_PIN_PA13, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_B_periph(AT91_PIN_PA14, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_B_periph(AT91_PIN_PA9, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_B_periph(AT91_PIN_PA8, 1);
+}
+
+/*
+ * Return the device node so that board init code can use it as the
+ * parent for the device node reflecting how it's used on this board.
+ *
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91SAM9RL_ID_SSC0:
+		pdev = &at91sam9rl_ssc0_device;
+		configure_ssc0_pins(pins);
+		at91_clock_associate("ssc0_clk", &pdev->dev, "pclk");
+		break;
+	case AT91SAM9RL_ID_SSC1:
+		pdev = &at91sam9rl_ssc1_device;
+		configure_ssc1_pins(pins);
+		at91_clock_associate("ssc1_clk", &pdev->dev, "pclk");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
 /* --------------------------------------------------------------------
  *  UART
  * -------------------------------------------------------------------- */
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index 762148cc48c1..7fefe018833c 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -116,6 +116,23 @@ struct atmel_uart_data {
 };
 extern void __init at91_add_device_serial(void);
 
+/*
+ * SSC -- accessed through ssc_request(id).  Drivers don't bind to SSC
+ * platform devices.  Their SSC ID is part of their configuration data,
+ * along with information about which SSC signals they should use.
+ */
+#define ATMEL_SSC_TK	0x01
+#define ATMEL_SSC_TF	0x02
+#define ATMEL_SSC_TD	0x04
+#define ATMEL_SSC_TX	(ATMEL_SSC_TK | ATMEL_SSC_TF | ATMEL_SSC_TD)
+
+#define ATMEL_SSC_RK	0x10
+#define ATMEL_SSC_RF	0x20
+#define ATMEL_SSC_RD	0x40
+#define ATMEL_SSC_RX	(ATMEL_SSC_RK | ATMEL_SSC_RF | ATMEL_SSC_RD)
+
+extern void __init at91_add_device_ssc(unsigned id, unsigned pins);
+
  /* LCD Controller */
 struct atmel_lcdfb_info;
 extern void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data);
-- 
cgit v1.2.3


From c8f385a631ef1f49d67a3798ca40dec36ccdf07d Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Wed, 23 Jan 2008 09:25:15 +0100
Subject: [ARM] 4757/1: [AT91] UART initialization

Modify the UART initialization to allow the board-initialization code
to specify which pins are connected, and which pins should therefore
be initialized.

The current at91_init_serial() will continue to work as-is, but is
marked as "deprecated" and will be removed once the board-specific
files has been updated to use the new interface.

As in the AVR32 code, we assume that the TX and RX pins will always be
initialized.

Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/at91rm9200_devices.c  | 114 ++++++++++++++++++++++++------
 arch/arm/mach-at91/at91sam9260_devices.c | 115 ++++++++++++++++++++++++++-----
 arch/arm/mach-at91/at91sam9261_devices.c |  77 ++++++++++++++++++---
 arch/arm/mach-at91/at91sam9263_devices.c |  79 +++++++++++++++++----
 arch/arm/mach-at91/at91sam9rl_devices.c  |  99 ++++++++++++++++++++++----
 include/asm-arm/arch-at91/board.h        |  12 +++-
 6 files changed, 420 insertions(+), 76 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91/at91rm9200_devices.c b/arch/arm/mach-at91/at91rm9200_devices.c
index a2647265c214..23966229ea9a 100644
--- a/arch/arm/mach-at91/at91rm9200_devices.c
+++ b/arch/arm/mach-at91/at91rm9200_devices.c
@@ -884,17 +884,21 @@ static struct platform_device at91rm9200_uart0_device = {
 	.num_resources	= ARRAY_SIZE(uart0_resources),
 };
 
-static inline void configure_usart0_pins(void)
+static inline void configure_usart0_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA17, 1);		/* TXD0 */
 	at91_set_A_periph(AT91_PIN_PA18, 0);		/* RXD0 */
-	at91_set_A_periph(AT91_PIN_PA20, 0);		/* CTS0 */
 
-	/*
-	 * AT91RM9200 Errata #39 - RTS0 is not internally connected to PA21.
-	 *  We need to drive the pin manually.  Default is off (RTS is active low).
-	 */
-	at91_set_gpio_output(AT91_PIN_PA21, 1);
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA20, 0);	/* CTS0 */
+
+	if (pins & ATMEL_UART_RTS) {
+		/*
+		 * AT91RM9200 Errata #39 - RTS0 is not internally connected to PA21.
+		 *  We need to drive the pin manually.  Default is off (RTS is active low).
+		 */
+		at91_set_gpio_output(AT91_PIN_PA21, 1);
+	}
 }
 
 static struct resource uart1_resources[] = {
@@ -929,16 +933,23 @@ static struct platform_device at91rm9200_uart1_device = {
 	.num_resources	= ARRAY_SIZE(uart1_resources),
 };
 
-static inline void configure_usart1_pins(void)
+static inline void configure_usart1_pins(unsigned pins)
 {
-	at91_set_A_periph(AT91_PIN_PB18, 0);		/* RI1 */
-	at91_set_A_periph(AT91_PIN_PB19, 0);		/* DTR1 */
 	at91_set_A_periph(AT91_PIN_PB20, 1);		/* TXD1 */
 	at91_set_A_periph(AT91_PIN_PB21, 0);		/* RXD1 */
-	at91_set_A_periph(AT91_PIN_PB23, 0);		/* DCD1 */
-	at91_set_A_periph(AT91_PIN_PB24, 0);		/* CTS1 */
-	at91_set_A_periph(AT91_PIN_PB25, 0);		/* DSR1 */
-	at91_set_A_periph(AT91_PIN_PB26, 0);		/* RTS1 */
+
+	if (pins & ATMEL_UART_RI)
+		at91_set_A_periph(AT91_PIN_PB18, 0);	/* RI1 */
+	if (pins & ATMEL_UART_DTR)
+		at91_set_A_periph(AT91_PIN_PB19, 0);	/* DTR1 */
+	if (pins & ATMEL_UART_DCD)
+		at91_set_A_periph(AT91_PIN_PB23, 0);	/* DCD1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PB24, 0);	/* CTS1 */
+	if (pins & ATMEL_UART_DSR)
+		at91_set_A_periph(AT91_PIN_PB25, 0);	/* DSR1 */
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PB26, 0);	/* RTS1 */
 }
 
 static struct resource uart2_resources[] = {
@@ -973,10 +984,15 @@ static struct platform_device at91rm9200_uart2_device = {
 	.num_resources	= ARRAY_SIZE(uart2_resources),
 };
 
-static inline void configure_usart2_pins(void)
+static inline void configure_usart2_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA22, 0);		/* RXD2 */
 	at91_set_A_periph(AT91_PIN_PA23, 1);		/* TXD2 */
+
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PA30, 0);	/* CTS2 */
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PA31, 0);	/* RTS2 */
 }
 
 static struct resource uart3_resources[] = {
@@ -1011,16 +1027,21 @@ static struct platform_device at91rm9200_uart3_device = {
 	.num_resources	= ARRAY_SIZE(uart3_resources),
 };
 
-static inline void configure_usart3_pins(void)
+static inline void configure_usart3_pins(unsigned pins)
 {
 	at91_set_B_periph(AT91_PIN_PA5, 1);		/* TXD3 */
 	at91_set_B_periph(AT91_PIN_PA6, 0);		/* RXD3 */
+
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PB1, 0);	/* CTS3 */
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PB0, 0);	/* RTS3 */
 }
 
 static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
 struct platform_device *atmel_default_console_device;	/* the serial console device */
 
-void __init at91_init_serial(struct at91_uart_config *config)
+void __init __deprecated at91_init_serial(struct at91_uart_config *config)
 {
 	int i;
 
@@ -1028,22 +1049,22 @@ void __init at91_init_serial(struct at91_uart_config *config)
 	for (i = 0; i < config->nr_tty; i++) {
 		switch (config->tty_map[i]) {
 			case 0:
-				configure_usart0_pins();
+				configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91rm9200_uart0_device;
 				at91_clock_associate("usart0_clk", &at91rm9200_uart0_device.dev, "usart");
 				break;
 			case 1:
-				configure_usart1_pins();
+				configure_usart1_pins(ATMEL_UART_CTS | ATMEL_UART_RTS | ATMEL_UART_DSR | ATMEL_UART_DTR | ATMEL_UART_DCD | ATMEL_UART_RI);
 				at91_uarts[i] = &at91rm9200_uart1_device;
 				at91_clock_associate("usart1_clk", &at91rm9200_uart1_device.dev, "usart");
 				break;
 			case 2:
-				configure_usart2_pins();
+				configure_usart2_pins(0);
 				at91_uarts[i] = &at91rm9200_uart2_device;
 				at91_clock_associate("usart2_clk", &at91rm9200_uart2_device.dev, "usart");
 				break;
 			case 3:
-				configure_usart3_pins();
+				configure_usart3_pins(0);
 				at91_uarts[i] = &at91rm9200_uart3_device;
 				at91_clock_associate("usart3_clk", &at91rm9200_uart3_device.dev, "usart");
 				break;
@@ -1065,6 +1086,53 @@ void __init at91_init_serial(struct at91_uart_config *config)
 		printk(KERN_INFO "AT91: No default serial console defined.\n");
 }
 
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91rm9200_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91RM9200_ID_US0:
+			pdev = &at91rm9200_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91RM9200_ID_US1:
+			pdev = &at91rm9200_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91RM9200_ID_US2:
+			pdev = &at91rm9200_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		case AT91RM9200_ID_US3:
+			pdev = &at91rm9200_uart3_device;
+			configure_usart3_pins(pins);
+			at91_clock_associate("usart3_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
 void __init at91_add_device_serial(void)
 {
 	int i;
@@ -1075,7 +1143,9 @@ void __init at91_add_device_serial(void)
 	}
 }
 #else
-void __init at91_init_serial(struct at91_uart_config *config) {}
+void __init __deprecated at91_init_serial(struct at91_uart_config *config) {}
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
 void __init at91_add_device_serial(void) {}
 #endif
 
diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index 8acd62b75cb0..5dd06bdccb00 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -760,16 +760,23 @@ static struct platform_device at91sam9260_uart0_device = {
 	.num_resources	= ARRAY_SIZE(uart0_resources),
 };
 
-static inline void configure_usart0_pins(void)
+static inline void configure_usart0_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PB4, 1);		/* TXD0 */
 	at91_set_A_periph(AT91_PIN_PB5, 0);		/* RXD0 */
-	at91_set_A_periph(AT91_PIN_PB26, 0);		/* RTS0 */
-	at91_set_A_periph(AT91_PIN_PB27, 0);		/* CTS0 */
-	at91_set_A_periph(AT91_PIN_PB24, 0);		/* DTR0 */
-	at91_set_A_periph(AT91_PIN_PB22, 0);		/* DSR0 */
-	at91_set_A_periph(AT91_PIN_PB23, 0);		/* DCD0 */
-	at91_set_A_periph(AT91_PIN_PB25, 0);		/* RI0 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PB26, 0);	/* RTS0 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PB27, 0);	/* CTS0 */
+	if (pins & ATMEL_UART_DTR)
+		at91_set_A_periph(AT91_PIN_PB24, 0);	/* DTR0 */
+	if (pins & ATMEL_UART_DSR)
+		at91_set_A_periph(AT91_PIN_PB22, 0);	/* DSR0 */
+	if (pins & ATMEL_UART_DCD)
+		at91_set_A_periph(AT91_PIN_PB23, 0);	/* DCD0 */
+	if (pins & ATMEL_UART_RI)
+		at91_set_A_periph(AT91_PIN_PB25, 0);	/* RI0 */
 }
 
 static struct resource uart1_resources[] = {
@@ -804,12 +811,15 @@ static struct platform_device at91sam9260_uart1_device = {
 	.num_resources	= ARRAY_SIZE(uart1_resources),
 };
 
-static inline void configure_usart1_pins(void)
+static inline void configure_usart1_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PB6, 1);		/* TXD1 */
 	at91_set_A_periph(AT91_PIN_PB7, 0);		/* RXD1 */
-	at91_set_A_periph(AT91_PIN_PB28, 0);		/* RTS1 */
-	at91_set_A_periph(AT91_PIN_PB29, 0);		/* CTS1 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PB28, 0);	/* RTS1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PB29, 0);	/* CTS1 */
 }
 
 static struct resource uart2_resources[] = {
@@ -844,10 +854,15 @@ static struct platform_device at91sam9260_uart2_device = {
 	.num_resources	= ARRAY_SIZE(uart2_resources),
 };
 
-static inline void configure_usart2_pins(void)
+static inline void configure_usart2_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PB8, 1);		/* TXD2 */
 	at91_set_A_periph(AT91_PIN_PB9, 0);		/* RXD2 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PA4, 0);	/* RTS2 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA5, 0);	/* CTS2 */
 }
 
 static struct resource uart3_resources[] = {
@@ -882,10 +897,15 @@ static struct platform_device at91sam9260_uart3_device = {
 	.num_resources	= ARRAY_SIZE(uart3_resources),
 };
 
-static inline void configure_usart3_pins(void)
+static inline void configure_usart3_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PB10, 1);		/* TXD3 */
 	at91_set_A_periph(AT91_PIN_PB11, 0);		/* RXD3 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PC8, 0);	/* RTS3 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PC10, 0);	/* CTS3 */
 }
 
 static struct resource uart4_resources[] = {
@@ -967,7 +987,7 @@ static inline void configure_usart5_pins(void)
 static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
 struct platform_device *atmel_default_console_device;	/* the serial console device */
 
-void __init at91_init_serial(struct at91_uart_config *config)
+void __init __deprecated at91_init_serial(struct at91_uart_config *config)
 {
 	int i;
 
@@ -975,22 +995,22 @@ void __init at91_init_serial(struct at91_uart_config *config)
 	for (i = 0; i < config->nr_tty; i++) {
 		switch (config->tty_map[i]) {
 			case 0:
-				configure_usart0_pins();
+				configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS | ATMEL_UART_DSR | ATMEL_UART_DTR | ATMEL_UART_DCD | ATMEL_UART_RI);
 				at91_uarts[i] = &at91sam9260_uart0_device;
 				at91_clock_associate("usart0_clk", &at91sam9260_uart0_device.dev, "usart");
 				break;
 			case 1:
-				configure_usart1_pins();
+				configure_usart1_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9260_uart1_device;
 				at91_clock_associate("usart1_clk", &at91sam9260_uart1_device.dev, "usart");
 				break;
 			case 2:
-				configure_usart2_pins();
+				configure_usart2_pins(0);
 				at91_uarts[i] = &at91sam9260_uart2_device;
 				at91_clock_associate("usart2_clk", &at91sam9260_uart2_device.dev, "usart");
 				break;
 			case 3:
-				configure_usart3_pins();
+				configure_usart3_pins(0);
 				at91_uarts[i] = &at91sam9260_uart3_device;
 				at91_clock_associate("usart3_clk", &at91sam9260_uart3_device.dev, "usart");
 				break;
@@ -1022,6 +1042,63 @@ void __init at91_init_serial(struct at91_uart_config *config)
 		printk(KERN_INFO "AT91: No default serial console defined.\n");
 }
 
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91sam9260_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US0:
+			pdev = &at91sam9260_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US1:
+			pdev = &at91sam9260_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US2:
+			pdev = &at91sam9260_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US3:
+			pdev = &at91sam9260_uart3_device;
+			configure_usart3_pins(pins);
+			at91_clock_associate("usart3_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US4:
+			pdev = &at91sam9260_uart4_device;
+			configure_usart4_pins();
+			at91_clock_associate("usart4_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9260_ID_US5:
+			pdev = &at91sam9260_uart5_device;
+			configure_usart5_pins();
+			at91_clock_associate("usart5_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
 void __init at91_add_device_serial(void)
 {
 	int i;
@@ -1032,7 +1109,9 @@ void __init at91_add_device_serial(void)
 	}
 }
 #else
-void __init at91_init_serial(struct at91_uart_config *config) {}
+void __init __deprecated at91_init_serial(struct at91_uart_config *config) {}
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
 void __init at91_add_device_serial(void) {}
 #endif
 
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index ba84ba6f18fb..467a6431d71a 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -867,12 +867,15 @@ static struct platform_device at91sam9261_uart0_device = {
 	.num_resources	= ARRAY_SIZE(uart0_resources),
 };
 
-static inline void configure_usart0_pins(void)
+static inline void configure_usart0_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PC8, 1);		/* TXD0 */
 	at91_set_A_periph(AT91_PIN_PC9, 0);		/* RXD0 */
-	at91_set_A_periph(AT91_PIN_PC10, 0);		/* RTS0 */
-	at91_set_A_periph(AT91_PIN_PC11, 0);		/* CTS0 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PC10, 0);	/* RTS0 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PC11, 0);	/* CTS0 */
 }
 
 static struct resource uart1_resources[] = {
@@ -907,10 +910,15 @@ static struct platform_device at91sam9261_uart1_device = {
 	.num_resources	= ARRAY_SIZE(uart1_resources),
 };
 
-static inline void configure_usart1_pins(void)
+static inline void configure_usart1_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PC12, 1);		/* TXD1 */
 	at91_set_A_periph(AT91_PIN_PC13, 0);		/* RXD1 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PA12, 0);	/* RTS1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PA13, 0);	/* CTS1 */
 }
 
 static struct resource uart2_resources[] = {
@@ -945,16 +953,21 @@ static struct platform_device at91sam9261_uart2_device = {
 	.num_resources	= ARRAY_SIZE(uart2_resources),
 };
 
-static inline void configure_usart2_pins(void)
+static inline void configure_usart2_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PC15, 0);		/* RXD2 */
 	at91_set_A_periph(AT91_PIN_PC14, 1);		/* TXD2 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PA15, 0);	/* RTS2*/
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PA16, 0);	/* CTS2 */
 }
 
 static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
 struct platform_device *atmel_default_console_device;	/* the serial console device */
 
-void __init at91_init_serial(struct at91_uart_config *config)
+void __init __deprecated at91_init_serial(struct at91_uart_config *config)
 {
 	int i;
 
@@ -962,17 +975,17 @@ void __init at91_init_serial(struct at91_uart_config *config)
 	for (i = 0; i < config->nr_tty; i++) {
 		switch (config->tty_map[i]) {
 			case 0:
-				configure_usart0_pins();
+				configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9261_uart0_device;
 				at91_clock_associate("usart0_clk", &at91sam9261_uart0_device.dev, "usart");
 				break;
 			case 1:
-				configure_usart1_pins();
+				configure_usart1_pins(0);
 				at91_uarts[i] = &at91sam9261_uart1_device;
 				at91_clock_associate("usart1_clk", &at91sam9261_uart1_device.dev, "usart");
 				break;
 			case 2:
-				configure_usart2_pins();
+				configure_usart2_pins(0);
 				at91_uarts[i] = &at91sam9261_uart2_device;
 				at91_clock_associate("usart2_clk", &at91sam9261_uart2_device.dev, "usart");
 				break;
@@ -994,6 +1007,48 @@ void __init at91_init_serial(struct at91_uart_config *config)
 		printk(KERN_INFO "AT91: No default serial console defined.\n");
 }
 
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91sam9261_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91SAM9261_ID_US0:
+			pdev = &at91sam9261_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9261_ID_US1:
+			pdev = &at91sam9261_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9261_ID_US2:
+			pdev = &at91sam9261_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
 void __init at91_add_device_serial(void)
 {
 	int i;
@@ -1004,7 +1059,9 @@ void __init at91_add_device_serial(void)
 	}
 }
 #else
-void __init at91_init_serial(struct at91_uart_config *config) {}
+void __init __deprecated at91_init_serial(struct at91_uart_config *config) {}
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
 void __init at91_add_device_serial(void) {}
 #endif
 
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index c025f5c5ffaa..a9f35d0c1cb0 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -1071,12 +1071,15 @@ static struct platform_device at91sam9263_uart0_device = {
 	.num_resources	= ARRAY_SIZE(uart0_resources),
 };
 
-static inline void configure_usart0_pins(void)
+static inline void configure_usart0_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA26, 1);		/* TXD0 */
 	at91_set_A_periph(AT91_PIN_PA27, 0);		/* RXD0 */
-	at91_set_A_periph(AT91_PIN_PA28, 0);		/* RTS0 */
-	at91_set_A_periph(AT91_PIN_PA29, 0);		/* CTS0 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PA28, 0);	/* RTS0 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA29, 0);	/* CTS0 */
 }
 
 static struct resource uart1_resources[] = {
@@ -1111,12 +1114,15 @@ static struct platform_device at91sam9263_uart1_device = {
 	.num_resources	= ARRAY_SIZE(uart1_resources),
 };
 
-static inline void configure_usart1_pins(void)
+static inline void configure_usart1_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PD0, 1);		/* TXD1 */
 	at91_set_A_periph(AT91_PIN_PD1, 0);		/* RXD1 */
-	at91_set_B_periph(AT91_PIN_PD7, 0);		/* RTS1 */
-	at91_set_B_periph(AT91_PIN_PD8, 0);		/* CTS1 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PD7, 0);	/* RTS1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PD8, 0);	/* CTS1 */
 }
 
 static struct resource uart2_resources[] = {
@@ -1151,18 +1157,21 @@ static struct platform_device at91sam9263_uart2_device = {
 	.num_resources	= ARRAY_SIZE(uart2_resources),
 };
 
-static inline void configure_usart2_pins(void)
+static inline void configure_usart2_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PD2, 1);		/* TXD2 */
 	at91_set_A_periph(AT91_PIN_PD3, 0);		/* RXD2 */
-	at91_set_B_periph(AT91_PIN_PD5, 0);		/* RTS2 */
-	at91_set_B_periph(AT91_PIN_PD6, 0);		/* CTS2 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PD5, 0);	/* RTS2 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PD6, 0);	/* CTS2 */
 }
 
 static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
 struct platform_device *atmel_default_console_device;	/* the serial console device */
 
-void __init at91_init_serial(struct at91_uart_config *config)
+void __init __deprecated at91_init_serial(struct at91_uart_config *config)
 {
 	int i;
 
@@ -1170,17 +1179,17 @@ void __init at91_init_serial(struct at91_uart_config *config)
 	for (i = 0; i < config->nr_tty; i++) {
 		switch (config->tty_map[i]) {
 			case 0:
-				configure_usart0_pins();
+				configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9263_uart0_device;
 				at91_clock_associate("usart0_clk", &at91sam9263_uart0_device.dev, "usart");
 				break;
 			case 1:
-				configure_usart1_pins();
+				configure_usart1_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9263_uart1_device;
 				at91_clock_associate("usart1_clk", &at91sam9263_uart1_device.dev, "usart");
 				break;
 			case 2:
-				configure_usart2_pins();
+				configure_usart2_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9263_uart2_device;
 				at91_clock_associate("usart2_clk", &at91sam9263_uart2_device.dev, "usart");
 				break;
@@ -1202,6 +1211,48 @@ void __init at91_init_serial(struct at91_uart_config *config)
 		printk(KERN_INFO "AT91: No default serial console defined.\n");
 }
 
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91sam9263_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91SAM9263_ID_US0:
+			pdev = &at91sam9263_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9263_ID_US1:
+			pdev = &at91sam9263_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9263_ID_US2:
+			pdev = &at91sam9263_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
 void __init at91_add_device_serial(void)
 {
 	int i;
@@ -1213,6 +1264,8 @@ void __init at91_add_device_serial(void)
 }
 #else
 void __init at91_init_serial(struct at91_uart_config *config) {}
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
 void __init at91_add_device_serial(void) {}
 #endif
 
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index e886cdfe919d..a942b9e35560 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -676,12 +676,23 @@ static struct platform_device at91sam9rl_uart0_device = {
 	.num_resources	= ARRAY_SIZE(uart0_resources),
 };
 
-static inline void configure_usart0_pins(void)
+static inline void configure_usart0_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA6, 1);		/* TXD0 */
 	at91_set_A_periph(AT91_PIN_PA7, 0);		/* RXD0 */
-	at91_set_A_periph(AT91_PIN_PA9, 0);		/* RTS0 */
-	at91_set_A_periph(AT91_PIN_PA10, 0);		/* CTS0 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PA9, 0);	/* RTS0 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA10, 0);	/* CTS0 */
+	if (pins & ATMEL_UART_DSR)
+		at91_set_A_periph(AT91_PIN_PD14, 0);	/* DSR0 */
+	if (pins & ATMEL_UART_DTR)
+		at91_set_A_periph(AT91_PIN_PD15, 0);	/* DTR0 */
+	if (pins & ATMEL_UART_DCD)
+		at91_set_A_periph(AT91_PIN_PD16, 0);	/* DCD0 */
+	if (pins & ATMEL_UART_RI)
+		at91_set_A_periph(AT91_PIN_PD17, 0);	/* RI0 */
 }
 
 static struct resource uart1_resources[] = {
@@ -716,10 +727,15 @@ static struct platform_device at91sam9rl_uart1_device = {
 	.num_resources	= ARRAY_SIZE(uart1_resources),
 };
 
-static inline void configure_usart1_pins(void)
+static inline void configure_usart1_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA11, 1);		/* TXD1 */
 	at91_set_A_periph(AT91_PIN_PA12, 0);		/* RXD1 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PA18, 0);	/* RTS1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PA19, 0);	/* CTS1 */
 }
 
 static struct resource uart2_resources[] = {
@@ -754,10 +770,15 @@ static struct platform_device at91sam9rl_uart2_device = {
 	.num_resources	= ARRAY_SIZE(uart2_resources),
 };
 
-static inline void configure_usart2_pins(void)
+static inline void configure_usart2_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PA13, 1);		/* TXD2 */
 	at91_set_A_periph(AT91_PIN_PA14, 0);		/* RXD2 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PA29, 0);	/* RTS2 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA30, 0);	/* CTS2 */
 }
 
 static struct resource uart3_resources[] = {
@@ -792,16 +813,21 @@ static struct platform_device at91sam9rl_uart3_device = {
 	.num_resources	= ARRAY_SIZE(uart3_resources),
 };
 
-static inline void configure_usart3_pins(void)
+static inline void configure_usart3_pins(unsigned pins)
 {
 	at91_set_A_periph(AT91_PIN_PB0, 1);		/* TXD3 */
 	at91_set_A_periph(AT91_PIN_PB1, 0);		/* RXD3 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PD4, 0);	/* RTS3 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PD3, 0);	/* CTS3 */
 }
 
 static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
 struct platform_device *atmel_default_console_device;	/* the serial console device */
 
-void __init at91_init_serial(struct at91_uart_config *config)
+void __init __deprecated at91_init_serial(struct at91_uart_config *config)
 {
 	int i;
 
@@ -809,22 +835,22 @@ void __init at91_init_serial(struct at91_uart_config *config)
 	for (i = 0; i < config->nr_tty; i++) {
 		switch (config->tty_map[i]) {
 			case 0:
-				configure_usart0_pins();
+				configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
 				at91_uarts[i] = &at91sam9rl_uart0_device;
 				at91_clock_associate("usart0_clk", &at91sam9rl_uart0_device.dev, "usart");
 				break;
 			case 1:
-				configure_usart1_pins();
+				configure_usart1_pins(0);
 				at91_uarts[i] = &at91sam9rl_uart1_device;
 				at91_clock_associate("usart1_clk", &at91sam9rl_uart1_device.dev, "usart");
 				break;
 			case 2:
-				configure_usart2_pins();
+				configure_usart2_pins(0);
 				at91_uarts[i] = &at91sam9rl_uart2_device;
 				at91_clock_associate("usart2_clk", &at91sam9rl_uart2_device.dev, "usart");
 				break;
 			case 3:
-				configure_usart3_pins();
+				configure_usart3_pins(0);
 				at91_uarts[i] = &at91sam9rl_uart3_device;
 				at91_clock_associate("usart3_clk", &at91sam9rl_uart3_device.dev, "usart");
 				break;
@@ -846,6 +872,53 @@ void __init at91_init_serial(struct at91_uart_config *config)
 		printk(KERN_INFO "AT91: No default serial console defined.\n");
 }
 
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91sam9rl_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91SAM9RL_ID_US0:
+			pdev = &at91sam9rl_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9RL_ID_US1:
+			pdev = &at91sam9rl_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9RL_ID_US2:
+			pdev = &at91sam9rl_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		case AT91SAM9RL_ID_US3:
+			pdev = &at91sam9rl_uart3_device;
+			configure_usart3_pins(pins);
+			at91_clock_associate("usart3_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
 void __init at91_add_device_serial(void)
 {
 	int i;
@@ -856,7 +929,9 @@ void __init at91_add_device_serial(void)
 	}
 }
 #else
-void __init at91_init_serial(struct at91_uart_config *config) {}
+void __init __deprecated at91_init_serial(struct at91_uart_config *config) {}
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
 void __init at91_add_device_serial(void) {}
 #endif
 
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index 7fefe018833c..3f41ab28b37f 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -101,13 +101,23 @@ extern void __init at91_add_device_i2c(struct i2c_board_info *devices, int nr_de
 extern void __init at91_add_device_spi(struct spi_board_info *devices, int nr_devices);
 
  /* Serial */
+#define ATMEL_UART_CTS	0x01
+#define ATMEL_UART_RTS	0x02
+#define ATMEL_UART_DSR	0x04
+#define ATMEL_UART_DTR	0x08
+#define ATMEL_UART_DCD	0x10
+#define ATMEL_UART_RI	0x20
+
+extern void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins);
+extern void __init at91_set_serial_console(unsigned portnr);
+
 struct at91_uart_config {
 	unsigned short	console_tty;	/* tty number of serial console */
 	unsigned short	nr_tty;		/* number of serial tty's */
 	short		tty_map[];	/* map UART to tty number */
 };
 extern struct platform_device *atmel_default_console_device;
-extern void __init at91_init_serial(struct at91_uart_config *config);
+extern void __init __deprecated at91_init_serial(struct at91_uart_config *config);
 
 struct atmel_uart_data {
 	short		use_dma_tx;	/* use transmit DMA? */
-- 
cgit v1.2.3


From a04ff1af9723607f5901b79c559357e37cee6823 Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Wed, 23 Jan 2008 09:27:06 +0100
Subject: [ARM] 4758/1: [AT91] LEDs

Move the LED initialization code out of the various *_devices.c files,
and into leds.c.
Also add support for NEW_LEDs.

Patch from David Brownell.

Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/Makefile              |  2 +-
 arch/arm/mach-at91/at91rm9200_devices.c  | 22 -----------
 arch/arm/mach-at91/at91sam9260_devices.c | 22 -----------
 arch/arm/mach-at91/at91sam9261_devices.c | 22 -----------
 arch/arm/mach-at91/at91sam9263_devices.c | 22 -----------
 arch/arm/mach-at91/at91sam9rl_devices.c  | 22 -----------
 arch/arm/mach-at91/leds.c                | 68 +++++++++++++++++++++++++++++++-
 include/asm-arm/arch-at91/board.h        |  4 +-
 8 files changed, 70 insertions(+), 114 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
index a8a009c2cf7d..529ffb5f3f13 100644
--- a/arch/arm/mach-at91/Makefile
+++ b/arch/arm/mach-at91/Makefile
@@ -45,7 +45,7 @@ obj-$(CONFIG_MACH_AT91SAM9RLEK)	+= board-sam9rlek.o
 obj-$(CONFIG_MACH_AT91EB01)	+= board-eb01.o
 
 # Drivers
-obj-$(CONFIG_LEDS)		+= leds.o
+obj-y				+= leds.o
 
 # Power Management
 obj-$(CONFIG_PM)		+= pm.o
diff --git a/arch/arm/mach-at91/at91rm9200_devices.c b/arch/arm/mach-at91/at91rm9200_devices.c
index 23966229ea9a..ea9a952fb877 100644
--- a/arch/arm/mach-at91/at91rm9200_devices.c
+++ b/arch/arm/mach-at91/at91rm9200_devices.c
@@ -613,28 +613,6 @@ static void __init at91_add_device_watchdog(void) {}
 #endif
 
 
-/* --------------------------------------------------------------------
- *  LEDs
- * -------------------------------------------------------------------- */
-
-#if defined(CONFIG_LEDS)
-u8 at91_leds_cpu;
-u8 at91_leds_timer;
-
-void __init at91_init_leds(u8 cpu_led, u8 timer_led)
-{
-	/* Enable GPIO to access the LEDs */
-	at91_set_gpio_output(cpu_led, 1);
-	at91_set_gpio_output(timer_led, 1);
-
-	at91_leds_cpu	= cpu_led;
-	at91_leds_timer	= timer_led;
-}
-#else
-void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
-#endif
-
-
 /* --------------------------------------------------------------------
  *  SSC -- Synchronous Serial Controller
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index 5dd06bdccb00..105f8403860b 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -584,28 +584,6 @@ static void __init at91_add_device_watchdog(void) {}
 #endif
 
 
-/* --------------------------------------------------------------------
- *  LEDs
- * -------------------------------------------------------------------- */
-
-#if defined(CONFIG_LEDS)
-u8 at91_leds_cpu;
-u8 at91_leds_timer;
-
-void __init at91_init_leds(u8 cpu_led, u8 timer_led)
-{
-	/* Enable GPIO to access the LEDs */
-	at91_set_gpio_output(cpu_led, 1);
-	at91_set_gpio_output(timer_led, 1);
-
-	at91_leds_cpu	= cpu_led;
-	at91_leds_timer	= timer_led;
-}
-#else
-void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
-#endif
-
-
 /* --------------------------------------------------------------------
  *  SSC -- Synchronous Serial Controller
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index 467a6431d71a..57d7e3bd05db 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -596,28 +596,6 @@ static void __init at91_add_device_watchdog(void) {}
 #endif
 
 
-/* --------------------------------------------------------------------
- *  LEDs
- * -------------------------------------------------------------------- */
-
-#if defined(CONFIG_LEDS)
-u8 at91_leds_cpu;
-u8 at91_leds_timer;
-
-void __init at91_init_leds(u8 cpu_led, u8 timer_led)
-{
-	/* Enable GPIO to access the LEDs */
-	at91_set_gpio_output(cpu_led, 1);
-	at91_set_gpio_output(timer_led, 1);
-
-	at91_leds_cpu	= cpu_led;
-	at91_leds_timer	= timer_led;
-}
-#else
-void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
-#endif
-
-
 /* --------------------------------------------------------------------
  *  SSC -- Synchronous Serial Controller
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index a9f35d0c1cb0..0b12e1adcc8e 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -843,28 +843,6 @@ static void __init at91_add_device_watchdog(void) {}
 #endif
 
 
-/* --------------------------------------------------------------------
- *  LEDs
- * -------------------------------------------------------------------- */
-
-#if defined(CONFIG_LEDS)
-u8 at91_leds_cpu;
-u8 at91_leds_timer;
-
-void __init at91_init_leds(u8 cpu_led, u8 timer_led)
-{
-	/* Enable GPIO to access the LEDs */
-	at91_set_gpio_output(cpu_led, 1);
-	at91_set_gpio_output(timer_led, 1);
-
-	at91_leds_cpu	= cpu_led;
-	at91_leds_timer	= timer_led;
-}
-#else
-void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
-#endif
-
-
 /* --------------------------------------------------------------------
  *  SSC -- Synchronous Serial Controller
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index a942b9e35560..f43b5c33e45d 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -449,28 +449,6 @@ static void __init at91_add_device_watchdog(void) {}
 #endif
 
 
-/* --------------------------------------------------------------------
- *  LEDs
- * -------------------------------------------------------------------- */
-
-#if defined(CONFIG_LEDS)
-u8 at91_leds_cpu;
-u8 at91_leds_timer;
-
-void __init at91_init_leds(u8 cpu_led, u8 timer_led)
-{
-	/* Enable GPIO to access the LEDs */
-	at91_set_gpio_output(cpu_led, 1);
-	at91_set_gpio_output(timer_led, 1);
-
-	at91_leds_cpu	= cpu_led;
-	at91_leds_timer	= timer_led;
-}
-#else
-void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
-#endif
-
-
 /* --------------------------------------------------------------------
  *  SSC -- Synchronous Serial Controller
  * -------------------------------------------------------------------- */
diff --git a/arch/arm/mach-at91/leds.c b/arch/arm/mach-at91/leds.c
index 0d5144973988..9cdcda500fe8 100644
--- a/arch/arm/mach-at91/leds.c
+++ b/arch/arm/mach-at91/leds.c
@@ -14,11 +14,62 @@
 #include <linux/init.h>
 
 #include <asm/mach-types.h>
-#include <asm/leds.h>
 #include <asm/arch/board.h>
 #include <asm/arch/gpio.h>
 
 
+/* ------------------------------------------------------------------------- */
+
+#if defined(CONFIG_NEW_LEDS)
+
+#include <linux/platform_device.h>
+
+/*
+ * New cross-platform LED support.
+ */
+
+static struct gpio_led_platform_data led_data;
+
+static struct platform_device at91_leds = {
+	.name			= "leds-gpio",
+	.id			= -1,
+	.dev.platform_data	= &led_data,
+};
+
+void __init at91_gpio_leds(struct gpio_led *leds, int nr)
+{
+	int i;
+
+	if (!nr)
+		return;
+
+	for (i = 0; i < nr; i++)
+		at91_set_gpio_output(leds[i].gpio, leds[i].active_low);
+
+	led_data.leds = leds;
+	led_data.num_leds = nr;
+	platform_device_register(&at91_leds);
+}
+
+#else
+void __init at91_gpio_leds(struct gpio_led *leds, int nr) {}
+#endif
+
+
+/* ------------------------------------------------------------------------- */
+
+#if defined(CONFIG_LEDS)
+
+#include <asm/leds.h>
+
+/*
+ * Old ARM-specific LED framework; not fully functional when generic time is
+ * in use.
+ */
+
+static u8 at91_leds_cpu;
+static u8 at91_leds_timer;
+
 static inline void at91_led_on(unsigned int led)
 {
 	at91_set_gpio_value(led, 0);
@@ -93,3 +144,18 @@ static int __init leds_init(void)
 }
 
 __initcall(leds_init);
+
+
+void __init at91_init_leds(u8 cpu_led, u8 timer_led)
+{
+	/* Enable GPIO to access the LEDs */
+	at91_set_gpio_output(cpu_led, 1);
+	at91_set_gpio_output(timer_led, 1);
+
+	at91_leds_cpu	= cpu_led;
+	at91_leds_timer	= timer_led;
+}
+
+#else
+void __init at91_init_leds(u8 cpu_led, u8 timer_led) {}
+#endif
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index 3f41ab28b37f..e88ad8621c17 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -34,6 +34,7 @@
 #include <linux/mtd/partitions.h>
 #include <linux/device.h>
 #include <linux/i2c.h>
+#include <linux/leds.h>
 #include <linux/spi/spi.h>
 
  /* USB Device */
@@ -157,9 +158,8 @@ extern void __init at91_add_device_ac97(struct atmel_ac97_data *data);
 extern void __init at91_add_device_isi(void);
 
  /* LEDs */
-extern u8 at91_leds_cpu;
-extern u8 at91_leds_timer;
 extern void __init at91_init_leds(u8 cpu_led, u8 timer_led);
+extern void __init at91_gpio_leds(struct gpio_led *leds, int nr);
 
 /* FIXME: this needs a better location, but gets stuff building again */
 extern int at91_suspend_entering_slow_clock(void);
-- 
cgit v1.2.3


From 2b3b3516b6eeea1464e205b2dde9ebc9b7dd2ec8 Mon Sep 17 00:00:00 2001
From: Andrew Victor <linux@maxim.org.za>
Date: Thu, 24 Jan 2008 15:10:39 +0100
Subject: [ARM] 4764/1: [AT91] AT91CAP9 core support

Add support for Atmel's AT91CAP9 Customizable Microcontroller family.
  <http://www.atmel.com/products/AT91CAP/Default.asp>

Signed-off-by: Stelian Pop <stelian@popies.net>
Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                               |    4 +-
 arch/arm/mach-at91/Kconfig                     |    3 +
 arch/arm/mach-at91/Makefile                    |    1 +
 arch/arm/mach-at91/Makefile.boot               |    7 +-
 arch/arm/mach-at91/at91cap9.c                  |  365 ++++++++
 arch/arm/mach-at91/at91cap9_devices.c          | 1066 ++++++++++++++++++++++++
 arch/arm/mach-at91/clock.c                     |    2 +
 arch/arm/mach-at91/generic.h                   |    2 +
 arch/arm/mach-at91/pm.c                        |    5 +
 arch/arm/mm/Kconfig                            |    4 +-
 drivers/net/Kconfig                            |    2 +-
 drivers/usb/gadget/Kconfig                     |    2 +-
 drivers/video/Kconfig                          |    2 +-
 drivers/video/atmel_lcdfb.c                    |    2 +-
 include/asm-arm/arch-at91/at91_pmc.h           |    7 +-
 include/asm-arm/arch-at91/at91cap9.h           |  121 +++
 include/asm-arm/arch-at91/at91cap9_matrix.h    |  132 +++
 include/asm-arm/arch-at91/at91sam9263_matrix.h |    2 +-
 include/asm-arm/arch-at91/at91sam9rl_matrix.h  |    2 +-
 include/asm-arm/arch-at91/board.h              |    2 +-
 include/asm-arm/arch-at91/cpu.h                |    9 +-
 include/asm-arm/arch-at91/hardware.h           |    2 +
 include/asm-arm/arch-at91/timex.h              |    5 +
 include/asm-avr32/arch-at32ap/cpu.h            |    1 +
 24 files changed, 1735 insertions(+), 15 deletions(-)
 create mode 100644 arch/arm/mach-at91/at91cap9.c
 create mode 100644 arch/arm/mach-at91/at91cap9_devices.c
 create mode 100644 include/asm-arm/arch-at91/at91cap9.h
 create mode 100644 include/asm-arm/arch-at91/at91cap9_matrix.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 93e40b65fd0b..0bd479c47651 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -180,8 +180,8 @@ config ARCH_AT91
 	bool "Atmel AT91"
 	select GENERIC_GPIO
 	help
-	  This enables support for systems based on the Atmel AT91RM9200
-	  and AT91SAM9xxx processors.
+	  This enables support for systems based on the Atmel AT91RM9200,
+	  AT91SAM9 and AT91CAP9 processors.
 
 config ARCH_CLPS7500
 	bool "Cirrus CL-PS7500FE"
diff --git a/arch/arm/mach-at91/Kconfig b/arch/arm/mach-at91/Kconfig
index 214733e897e5..d12c616407d6 100644
--- a/arch/arm/mach-at91/Kconfig
+++ b/arch/arm/mach-at91/Kconfig
@@ -22,6 +22,9 @@ config ARCH_AT91SAM9263
 config ARCH_AT91SAM9RL
 	bool "AT91SAM9RL"
 
+config ARCH_AT91CAP9
+	bool "AT91CAP9"
+
 config ARCH_AT91X40
 	bool "AT91x40"
 
diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
index 529ffb5f3f13..10413c5d3f8a 100644
--- a/arch/arm/mach-at91/Makefile
+++ b/arch/arm/mach-at91/Makefile
@@ -15,6 +15,7 @@ obj-$(CONFIG_ARCH_AT91SAM9260)	+= at91sam9260.o at91sam926x_time.o at91sam9260_d
 obj-$(CONFIG_ARCH_AT91SAM9261)	+= at91sam9261.o at91sam926x_time.o at91sam9261_devices.o
 obj-$(CONFIG_ARCH_AT91SAM9263)	+= at91sam9263.o at91sam926x_time.o at91sam9263_devices.o
 obj-$(CONFIG_ARCH_AT91SAM9RL)	+= at91sam9rl.o at91sam926x_time.o at91sam9rl_devices.o
+obj-$(CONFIG_ARCH_AT91CAP9)	+= at91cap9.o at91sam926x_time.o at91cap9_devices.o
 obj-$(CONFIG_ARCH_AT91X40)	+= at91x40.o at91x40_time.o
 
 # AT91RM9200 board-specific support
diff --git a/arch/arm/mach-at91/Makefile.boot b/arch/arm/mach-at91/Makefile.boot
index e667dcc7cd34..071a2506a69f 100644
--- a/arch/arm/mach-at91/Makefile.boot
+++ b/arch/arm/mach-at91/Makefile.boot
@@ -3,7 +3,12 @@
 #   PARAMS_PHYS must be within 4MB of ZRELADDR
 #   INITRD_PHYS must be in RAM
 
+ifeq ($(CONFIG_ARCH_AT91CAP9),y)
+   zreladdr-y	:= 0x70008000
+params_phys-y	:= 0x70000100
+initrd_phys-y	:= 0x70410000
+else
    zreladdr-y	:= 0x20008000
 params_phys-y	:= 0x20000100
 initrd_phys-y	:= 0x20410000
-
+endif
diff --git a/arch/arm/mach-at91/at91cap9.c b/arch/arm/mach-at91/at91cap9.c
new file mode 100644
index 000000000000..48d27d8000b0
--- /dev/null
+++ b/arch/arm/mach-at91/at91cap9.c
@@ -0,0 +1,365 @@
+/*
+ * arch/arm/mach-at91/at91cap9.c
+ *
+ *  Copyright (C) 2007 Stelian Pop <stelian.pop@leadtechdesign.com>
+ *  Copyright (C) 2007 Lead Tech Design <www.leadtechdesign.com>
+ *  Copyright (C) 2007 Atmel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+#include <linux/module.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/arch/at91cap9.h>
+#include <asm/arch/at91_pmc.h>
+#include <asm/arch/at91_rstc.h>
+
+#include "generic.h"
+#include "clock.h"
+
+static struct map_desc at91cap9_io_desc[] __initdata = {
+	{
+		.virtual	= AT91_VA_BASE_SYS,
+		.pfn		= __phys_to_pfn(AT91_BASE_SYS),
+		.length		= SZ_16K,
+		.type		= MT_DEVICE,
+	}, {
+		.virtual	= AT91_IO_VIRT_BASE - AT91CAP9_SRAM_SIZE,
+		.pfn		= __phys_to_pfn(AT91CAP9_SRAM_BASE),
+		.length		= AT91CAP9_SRAM_SIZE,
+		.type		= MT_DEVICE,
+	},
+};
+
+/* --------------------------------------------------------------------
+ *  Clocks
+ * -------------------------------------------------------------------- */
+
+/*
+ * The peripheral clocks.
+ */
+static struct clk pioABCD_clk = {
+	.name		= "pioABCD_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_PIOABCD,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mpb0_clk = {
+	.name		= "mpb0_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MPB0,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mpb1_clk = {
+	.name		= "mpb1_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MPB1,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mpb2_clk = {
+	.name		= "mpb2_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MPB2,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mpb3_clk = {
+	.name		= "mpb3_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MPB3,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mpb4_clk = {
+	.name		= "mpb4_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MPB4,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk usart0_clk = {
+	.name		= "usart0_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_US0,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk usart1_clk = {
+	.name		= "usart1_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_US1,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk usart2_clk = {
+	.name		= "usart2_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_US2,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mmc0_clk = {
+	.name		= "mci0_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MCI0,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk mmc1_clk = {
+	.name		= "mci1_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_MCI1,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk can_clk = {
+	.name		= "can_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_CAN,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk twi_clk = {
+	.name		= "twi_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_TWI,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk spi0_clk = {
+	.name		= "spi0_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_SPI0,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk spi1_clk = {
+	.name		= "spi1_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_SPI1,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk ssc0_clk = {
+	.name		= "ssc0_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_SSC0,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk ssc1_clk = {
+	.name		= "ssc1_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_SSC1,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk ac97_clk = {
+	.name		= "ac97_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_AC97C,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk tcb_clk = {
+	.name		= "tcb_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_TCB,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk pwmc_clk = {
+	.name		= "pwmc_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_PWMC,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk macb_clk = {
+	.name		= "macb_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_EMAC,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk aestdes_clk = {
+	.name		= "aestdes_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_AESTDES,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk adc_clk = {
+	.name		= "adc_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_ADC,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk isi_clk = {
+	.name		= "isi_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_ISI,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk lcdc_clk = {
+	.name		= "lcdc_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_LCDC,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk dma_clk = {
+	.name		= "dma_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_DMA,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk udphs_clk = {
+	.name		= "udphs_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_UDPHS,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+static struct clk ohci_clk = {
+	.name		= "ohci_clk",
+	.pmc_mask	= 1 << AT91CAP9_ID_UHP,
+	.type		= CLK_TYPE_PERIPHERAL,
+};
+
+static struct clk *periph_clocks[] __initdata = {
+	&pioABCD_clk,
+	&mpb0_clk,
+	&mpb1_clk,
+	&mpb2_clk,
+	&mpb3_clk,
+	&mpb4_clk,
+	&usart0_clk,
+	&usart1_clk,
+	&usart2_clk,
+	&mmc0_clk,
+	&mmc1_clk,
+	&can_clk,
+	&twi_clk,
+	&spi0_clk,
+	&spi1_clk,
+	&ssc0_clk,
+	&ssc1_clk,
+	&ac97_clk,
+	&tcb_clk,
+	&pwmc_clk,
+	&macb_clk,
+	&aestdes_clk,
+	&adc_clk,
+	&isi_clk,
+	&lcdc_clk,
+	&dma_clk,
+	&udphs_clk,
+	&ohci_clk,
+	// irq0 .. irq1
+};
+
+/*
+ * The four programmable clocks.
+ * You must configure pin multiplexing to bring these signals out.
+ */
+static struct clk pck0 = {
+	.name		= "pck0",
+	.pmc_mask	= AT91_PMC_PCK0,
+	.type		= CLK_TYPE_PROGRAMMABLE,
+	.id		= 0,
+};
+static struct clk pck1 = {
+	.name		= "pck1",
+	.pmc_mask	= AT91_PMC_PCK1,
+	.type		= CLK_TYPE_PROGRAMMABLE,
+	.id		= 1,
+};
+static struct clk pck2 = {
+	.name		= "pck2",
+	.pmc_mask	= AT91_PMC_PCK2,
+	.type		= CLK_TYPE_PROGRAMMABLE,
+	.id		= 2,
+};
+static struct clk pck3 = {
+	.name		= "pck3",
+	.pmc_mask	= AT91_PMC_PCK3,
+	.type		= CLK_TYPE_PROGRAMMABLE,
+	.id		= 3,
+};
+
+static void __init at91cap9_register_clocks(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(periph_clocks); i++)
+		clk_register(periph_clocks[i]);
+
+	clk_register(&pck0);
+	clk_register(&pck1);
+	clk_register(&pck2);
+	clk_register(&pck3);
+}
+
+/* --------------------------------------------------------------------
+ *  GPIO
+ * -------------------------------------------------------------------- */
+
+static struct at91_gpio_bank at91cap9_gpio[] = {
+	{
+		.id		= AT91CAP9_ID_PIOABCD,
+		.offset		= AT91_PIOA,
+		.clock		= &pioABCD_clk,
+	}, {
+		.id		= AT91CAP9_ID_PIOABCD,
+		.offset		= AT91_PIOB,
+		.clock		= &pioABCD_clk,
+	}, {
+		.id		= AT91CAP9_ID_PIOABCD,
+		.offset		= AT91_PIOC,
+		.clock		= &pioABCD_clk,
+	}, {
+		.id		= AT91CAP9_ID_PIOABCD,
+		.offset		= AT91_PIOD,
+		.clock		= &pioABCD_clk,
+	}
+};
+
+static void at91cap9_reset(void)
+{
+	at91_sys_write(AT91_RSTC_CR, AT91_RSTC_KEY | AT91_RSTC_PROCRST | AT91_RSTC_PERRST);
+}
+
+/* --------------------------------------------------------------------
+ *  AT91CAP9 processor initialization
+ * -------------------------------------------------------------------- */
+
+void __init at91cap9_initialize(unsigned long main_clock)
+{
+	/* Map peripherals */
+	iotable_init(at91cap9_io_desc, ARRAY_SIZE(at91cap9_io_desc));
+
+	at91_arch_reset = at91cap9_reset;
+	at91_extern_irq = (1 << AT91CAP9_ID_IRQ0) | (1 << AT91CAP9_ID_IRQ1);
+
+	/* Init clock subsystem */
+	at91_clock_init(main_clock);
+
+	/* Register the processor-specific clocks */
+	at91cap9_register_clocks();
+
+	/* Register GPIO subsystem */
+	at91_gpio_init(at91cap9_gpio, 4);
+}
+
+/* --------------------------------------------------------------------
+ *  Interrupt initialization
+ * -------------------------------------------------------------------- */
+
+/*
+ * The default interrupt priority levels (0 = lowest, 7 = highest).
+ */
+static unsigned int at91cap9_default_irq_priority[NR_AIC_IRQS] __initdata = {
+	7,	/* Advanced Interrupt Controller (FIQ) */
+	7,	/* System Peripherals */
+	1,	/* Parallel IO Controller A, B, C and D */
+	0,	/* MP Block Peripheral 0 */
+	0,	/* MP Block Peripheral 1 */
+	0,	/* MP Block Peripheral 2 */
+	0,	/* MP Block Peripheral 3 */
+	0,	/* MP Block Peripheral 4 */
+	5,	/* USART 0 */
+	5,	/* USART 1 */
+	5,	/* USART 2 */
+	0,	/* Multimedia Card Interface 0 */
+	0,	/* Multimedia Card Interface 1 */
+	3,	/* CAN */
+	6,	/* Two-Wire Interface */
+	5,	/* Serial Peripheral Interface 0 */
+	5,	/* Serial Peripheral Interface 1 */
+	4,	/* Serial Synchronous Controller 0 */
+	4,	/* Serial Synchronous Controller 1 */
+	5,	/* AC97 Controller */
+	0,	/* Timer Counter 0, 1 and 2 */
+	0,	/* Pulse Width Modulation Controller */
+	3,	/* Ethernet */
+	0,	/* Advanced Encryption Standard, Triple DES*/
+	0,	/* Analog-to-Digital Converter */
+	0,	/* Image Sensor Interface */
+	3,	/* LCD Controller */
+	0,	/* DMA Controller */
+	2,	/* USB Device Port */
+	2,	/* USB Host port */
+	0,	/* Advanced Interrupt Controller (IRQ0) */
+	0,	/* Advanced Interrupt Controller (IRQ1) */
+};
+
+void __init at91cap9_init_interrupts(unsigned int priority[NR_AIC_IRQS])
+{
+	if (!priority)
+		priority = at91cap9_default_irq_priority;
+
+	/* Initialize the AIC interrupt controller */
+	at91_aic_init(priority);
+
+	/* Enable GPIO interrupts */
+	at91_gpio_irq_setup();
+}
diff --git a/arch/arm/mach-at91/at91cap9_devices.c b/arch/arm/mach-at91/at91cap9_devices.c
new file mode 100644
index 000000000000..c50fad9cd143
--- /dev/null
+++ b/arch/arm/mach-at91/at91cap9_devices.c
@@ -0,0 +1,1066 @@
+/*
+ * arch/arm/mach-at91/at91cap9_devices.c
+ *
+ *  Copyright (C) 2007 Stelian Pop <stelian.pop@leadtechdesign.com>
+ *  Copyright (C) 2007 Lead Tech Design <www.leadtechdesign.com>
+ *  Copyright (C) 2007 Atmel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
+#include <linux/mtd/physmap.h>
+
+#include <video/atmel_lcdc.h>
+
+#include <asm/arch/board.h>
+#include <asm/arch/gpio.h>
+#include <asm/arch/at91cap9.h>
+#include <asm/arch/at91sam926x_mc.h>
+#include <asm/arch/at91cap9_matrix.h>
+
+#include "generic.h"
+
+
+/* --------------------------------------------------------------------
+ *  USB Host
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_USB_OHCI_HCD) || defined(CONFIG_USB_OHCI_HCD_MODULE)
+static u64 ohci_dmamask = DMA_BIT_MASK(32);
+static struct at91_usbh_data usbh_data;
+
+static struct resource usbh_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_UHP_BASE,
+		.end	= AT91CAP9_UHP_BASE + SZ_1M - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_UHP,
+		.end	= AT91CAP9_ID_UHP,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91_usbh_device = {
+	.name		= "at91_ohci",
+	.id		= -1,
+	.dev		= {
+				.dma_mask		= &ohci_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &usbh_data,
+	},
+	.resource	= usbh_resources,
+	.num_resources	= ARRAY_SIZE(usbh_resources),
+};
+
+void __init at91_add_device_usbh(struct at91_usbh_data *data)
+{
+	int i;
+
+	if (!data)
+		return;
+
+	/* Enable VBus control for UHP ports */
+	for (i = 0; i < data->ports; i++) {
+		if (data->vbus_pin[i])
+			at91_set_gpio_output(data->vbus_pin[i], 0);
+	}
+
+	usbh_data = *data;
+	platform_device_register(&at91_usbh_device);
+}
+#else
+void __init at91_add_device_usbh(struct at91_usbh_data *data) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  Ethernet
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_MACB) || defined(CONFIG_MACB_MODULE)
+static u64 eth_dmamask = DMA_BIT_MASK(32);
+static struct at91_eth_data eth_data;
+
+static struct resource eth_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_EMAC,
+		.end	= AT91CAP9_BASE_EMAC + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_EMAC,
+		.end	= AT91CAP9_ID_EMAC,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_eth_device = {
+	.name		= "macb",
+	.id		= -1,
+	.dev		= {
+				.dma_mask		= &eth_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &eth_data,
+	},
+	.resource	= eth_resources,
+	.num_resources	= ARRAY_SIZE(eth_resources),
+};
+
+void __init at91_add_device_eth(struct at91_eth_data *data)
+{
+	if (!data)
+		return;
+
+	if (data->phy_irq_pin) {
+		at91_set_gpio_input(data->phy_irq_pin, 0);
+		at91_set_deglitch(data->phy_irq_pin, 1);
+	}
+
+	/* Pins used for MII and RMII */
+	at91_set_A_periph(AT91_PIN_PB21, 0);	/* ETXCK_EREFCK */
+	at91_set_A_periph(AT91_PIN_PB22, 0);	/* ERXDV */
+	at91_set_A_periph(AT91_PIN_PB25, 0);	/* ERX0 */
+	at91_set_A_periph(AT91_PIN_PB26, 0);	/* ERX1 */
+	at91_set_A_periph(AT91_PIN_PB27, 0);	/* ERXER */
+	at91_set_A_periph(AT91_PIN_PB28, 0);	/* ETXEN */
+	at91_set_A_periph(AT91_PIN_PB23, 0);	/* ETX0 */
+	at91_set_A_periph(AT91_PIN_PB24, 0);	/* ETX1 */
+	at91_set_A_periph(AT91_PIN_PB30, 0);	/* EMDIO */
+	at91_set_A_periph(AT91_PIN_PB29, 0);	/* EMDC */
+
+	if (!data->is_rmii) {
+		at91_set_B_periph(AT91_PIN_PC25, 0);	/* ECRS */
+		at91_set_B_periph(AT91_PIN_PC26, 0);	/* ECOL */
+		at91_set_B_periph(AT91_PIN_PC22, 0);	/* ERX2 */
+		at91_set_B_periph(AT91_PIN_PC23, 0);	/* ERX3 */
+		at91_set_B_periph(AT91_PIN_PC27, 0);	/* ERXCK */
+		at91_set_B_periph(AT91_PIN_PC20, 0);	/* ETX2 */
+		at91_set_B_periph(AT91_PIN_PC21, 0);	/* ETX3 */
+		at91_set_B_periph(AT91_PIN_PC24, 0);	/* ETXER */
+	}
+
+	eth_data = *data;
+	platform_device_register(&at91cap9_eth_device);
+}
+#else
+void __init at91_add_device_eth(struct at91_eth_data *data) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  MMC / SD
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_MMC_AT91) || defined(CONFIG_MMC_AT91_MODULE)
+static u64 mmc_dmamask = DMA_BIT_MASK(32);
+static struct at91_mmc_data mmc0_data, mmc1_data;
+
+static struct resource mmc0_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_MCI0,
+		.end	= AT91CAP9_BASE_MCI0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_MCI0,
+		.end	= AT91CAP9_ID_MCI0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_mmc0_device = {
+	.name		= "at91_mci",
+	.id		= 0,
+	.dev		= {
+				.dma_mask		= &mmc_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &mmc0_data,
+	},
+	.resource	= mmc0_resources,
+	.num_resources	= ARRAY_SIZE(mmc0_resources),
+};
+
+static struct resource mmc1_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_MCI1,
+		.end	= AT91CAP9_BASE_MCI1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_MCI1,
+		.end	= AT91CAP9_ID_MCI1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_mmc1_device = {
+	.name		= "at91_mci",
+	.id		= 1,
+	.dev		= {
+				.dma_mask		= &mmc_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &mmc1_data,
+	},
+	.resource	= mmc1_resources,
+	.num_resources	= ARRAY_SIZE(mmc1_resources),
+};
+
+void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data)
+{
+	if (!data)
+		return;
+
+	/* input/irq */
+	if (data->det_pin) {
+		at91_set_gpio_input(data->det_pin, 1);
+		at91_set_deglitch(data->det_pin, 1);
+	}
+	if (data->wp_pin)
+		at91_set_gpio_input(data->wp_pin, 1);
+	if (data->vcc_pin)
+		at91_set_gpio_output(data->vcc_pin, 0);
+
+	if (mmc_id == 0) {		/* MCI0 */
+		/* CLK */
+		at91_set_A_periph(AT91_PIN_PA2, 0);
+
+		/* CMD */
+		at91_set_A_periph(AT91_PIN_PA1, 1);
+
+		/* DAT0, maybe DAT1..DAT3 */
+		at91_set_A_periph(AT91_PIN_PA0, 1);
+		if (data->wire4) {
+			at91_set_A_periph(AT91_PIN_PA3, 1);
+			at91_set_A_periph(AT91_PIN_PA4, 1);
+			at91_set_A_periph(AT91_PIN_PA5, 1);
+		}
+
+		mmc0_data = *data;
+		at91_clock_associate("mci0_clk", &at91cap9_mmc1_device.dev, "mci_clk");
+		platform_device_register(&at91cap9_mmc0_device);
+	} else {			/* MCI1 */
+		/* CLK */
+		at91_set_A_periph(AT91_PIN_PA16, 0);
+
+		/* CMD */
+		at91_set_A_periph(AT91_PIN_PA17, 1);
+
+		/* DAT0, maybe DAT1..DAT3 */
+		at91_set_A_periph(AT91_PIN_PA18, 1);
+		if (data->wire4) {
+			at91_set_A_periph(AT91_PIN_PA19, 1);
+			at91_set_A_periph(AT91_PIN_PA20, 1);
+			at91_set_A_periph(AT91_PIN_PA21, 1);
+		}
+
+		mmc1_data = *data;
+		at91_clock_associate("mci1_clk", &at91cap9_mmc1_device.dev, "mci_clk");
+		platform_device_register(&at91cap9_mmc1_device);
+	}
+}
+#else
+void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  NAND / SmartMedia
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
+static struct at91_nand_data nand_data;
+
+#define NAND_BASE	AT91_CHIPSELECT_3
+
+static struct resource nand_resources[] = {
+	{
+		.start	= NAND_BASE,
+		.end	= NAND_BASE + SZ_256M - 1,
+		.flags	= IORESOURCE_MEM,
+	}
+};
+
+static struct platform_device at91cap9_nand_device = {
+	.name		= "at91_nand",
+	.id		= -1,
+	.dev		= {
+				.platform_data	= &nand_data,
+	},
+	.resource	= nand_resources,
+	.num_resources	= ARRAY_SIZE(nand_resources),
+};
+
+void __init at91_add_device_nand(struct at91_nand_data *data)
+{
+	unsigned long csa, mode;
+
+	if (!data)
+		return;
+
+	csa = at91_sys_read(AT91_MATRIX_EBICSA);
+	at91_sys_write(AT91_MATRIX_EBICSA, csa | AT91_MATRIX_EBI_CS3A_SMC_SMARTMEDIA | AT91_MATRIX_EBI_VDDIOMSEL_3_3V);
+
+	/* set the bus interface characteristics */
+	at91_sys_write(AT91_SMC_SETUP(3), AT91_SMC_NWESETUP_(2) | AT91_SMC_NCS_WRSETUP_(1)
+			| AT91_SMC_NRDSETUP_(2) | AT91_SMC_NCS_RDSETUP_(1));
+
+	at91_sys_write(AT91_SMC_PULSE(3), AT91_SMC_NWEPULSE_(4) | AT91_SMC_NCS_WRPULSE_(6)
+			| AT91_SMC_NRDPULSE_(4) | AT91_SMC_NCS_RDPULSE_(6));
+
+	at91_sys_write(AT91_SMC_CYCLE(3), AT91_SMC_NWECYCLE_(8) | AT91_SMC_NRDCYCLE_(8));
+
+	if (data->bus_width_16)
+		mode = AT91_SMC_DBW_16;
+	else
+		mode = AT91_SMC_DBW_8;
+	at91_sys_write(AT91_SMC_MODE(3), mode | AT91_SMC_READMODE | AT91_SMC_WRITEMODE | AT91_SMC_EXNWMODE_DISABLE | AT91_SMC_TDF_(1));
+
+	/* enable pin */
+	if (data->enable_pin)
+		at91_set_gpio_output(data->enable_pin, 1);
+
+	/* ready/busy pin */
+	if (data->rdy_pin)
+		at91_set_gpio_input(data->rdy_pin, 1);
+
+	/* card detect pin */
+	if (data->det_pin)
+		at91_set_gpio_input(data->det_pin, 1);
+
+	nand_data = *data;
+	platform_device_register(&at91cap9_nand_device);
+}
+#else
+void __init at91_add_device_nand(struct at91_nand_data *data) {}
+#endif
+
+/* --------------------------------------------------------------------
+ *  TWI (i2c)
+ * -------------------------------------------------------------------- */
+
+/*
+ * Prefer the GPIO code since the TWI controller isn't robust
+ * (gets overruns and underruns under load) and can only issue
+ * repeated STARTs in one scenario (the driver doesn't yet handle them).
+ */
+#if defined(CONFIG_I2C_GPIO) || defined(CONFIG_I2C_GPIO_MODULE)
+
+static struct i2c_gpio_platform_data pdata = {
+	.sda_pin		= AT91_PIN_PB4,
+	.sda_is_open_drain	= 1,
+	.scl_pin		= AT91_PIN_PB5,
+	.scl_is_open_drain	= 1,
+	.udelay			= 2,		/* ~100 kHz */
+};
+
+static struct platform_device at91cap9_twi_device = {
+	.name			= "i2c-gpio",
+	.id			= -1,
+	.dev.platform_data	= &pdata,
+};
+
+void __init at91_add_device_i2c(struct i2c_board_info *devices, int nr_devices)
+{
+	at91_set_GPIO_periph(AT91_PIN_PB4, 1);		/* TWD (SDA) */
+	at91_set_multi_drive(AT91_PIN_PB4, 1);
+
+	at91_set_GPIO_periph(AT91_PIN_PB5, 1);		/* TWCK (SCL) */
+	at91_set_multi_drive(AT91_PIN_PB5, 1);
+
+	i2c_register_board_info(0, devices, nr_devices);
+	platform_device_register(&at91cap9_twi_device);
+}
+
+#elif defined(CONFIG_I2C_AT91) || defined(CONFIG_I2C_AT91_MODULE)
+
+static struct resource twi_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_TWI,
+		.end	= AT91CAP9_BASE_TWI + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_TWI,
+		.end	= AT91CAP9_ID_TWI,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_twi_device = {
+	.name		= "at91_i2c",
+	.id		= -1,
+	.resource	= twi_resources,
+	.num_resources	= ARRAY_SIZE(twi_resources),
+};
+
+void __init at91_add_device_i2c(struct i2c_board_info *devices, int nr_devices)
+{
+	/* pins used for TWI interface */
+	at91_set_B_periph(AT91_PIN_PB4, 0);		/* TWD */
+	at91_set_multi_drive(AT91_PIN_PB4, 1);
+
+	at91_set_B_periph(AT91_PIN_PB5, 0);		/* TWCK */
+	at91_set_multi_drive(AT91_PIN_PB5, 1);
+
+	i2c_register_board_info(0, devices, nr_devices);
+	platform_device_register(&at91cap9_twi_device);
+}
+#else
+void __init at91_add_device_i2c(struct i2c_board_info *devices, int nr_devices) {}
+#endif
+
+/* --------------------------------------------------------------------
+ *  SPI
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_SPI_ATMEL) || defined(CONFIG_SPI_ATMEL_MODULE)
+static u64 spi_dmamask = DMA_BIT_MASK(32);
+
+static struct resource spi0_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_SPI0,
+		.end	= AT91CAP9_BASE_SPI0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_SPI0,
+		.end	= AT91CAP9_ID_SPI0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_spi0_device = {
+	.name		= "atmel_spi",
+	.id		= 0,
+	.dev		= {
+				.dma_mask		= &spi_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= spi0_resources,
+	.num_resources	= ARRAY_SIZE(spi0_resources),
+};
+
+static const unsigned spi0_standard_cs[4] = { AT91_PIN_PA5, AT91_PIN_PA3, AT91_PIN_PD0, AT91_PIN_PD1 };
+
+static struct resource spi1_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_SPI1,
+		.end	= AT91CAP9_BASE_SPI1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_SPI1,
+		.end	= AT91CAP9_ID_SPI1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_spi1_device = {
+	.name		= "atmel_spi",
+	.id		= 1,
+	.dev		= {
+				.dma_mask		= &spi_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= spi1_resources,
+	.num_resources	= ARRAY_SIZE(spi1_resources),
+};
+
+static const unsigned spi1_standard_cs[4] = { AT91_PIN_PB15, AT91_PIN_PB16, AT91_PIN_PB17, AT91_PIN_PB18 };
+
+void __init at91_add_device_spi(struct spi_board_info *devices, int nr_devices)
+{
+	int i;
+	unsigned long cs_pin;
+	short enable_spi0 = 0;
+	short enable_spi1 = 0;
+
+	/* Choose SPI chip-selects */
+	for (i = 0; i < nr_devices; i++) {
+		if (devices[i].controller_data)
+			cs_pin = (unsigned long) devices[i].controller_data;
+		else if (devices[i].bus_num == 0)
+			cs_pin = spi0_standard_cs[devices[i].chip_select];
+		else
+			cs_pin = spi1_standard_cs[devices[i].chip_select];
+
+		if (devices[i].bus_num == 0)
+			enable_spi0 = 1;
+		else
+			enable_spi1 = 1;
+
+		/* enable chip-select pin */
+		at91_set_gpio_output(cs_pin, 1);
+
+		/* pass chip-select pin to driver */
+		devices[i].controller_data = (void *) cs_pin;
+	}
+
+	spi_register_board_info(devices, nr_devices);
+
+	/* Configure SPI bus(es) */
+	if (enable_spi0) {
+		at91_set_B_periph(AT91_PIN_PA0, 0);	/* SPI0_MISO */
+		at91_set_B_periph(AT91_PIN_PA1, 0);	/* SPI0_MOSI */
+		at91_set_B_periph(AT91_PIN_PA2, 0);	/* SPI0_SPCK */
+
+		at91_clock_associate("spi0_clk", &at91cap9_spi0_device.dev, "spi_clk");
+		platform_device_register(&at91cap9_spi0_device);
+	}
+	if (enable_spi1) {
+		at91_set_A_periph(AT91_PIN_PB12, 0);	/* SPI1_MISO */
+		at91_set_A_periph(AT91_PIN_PB13, 0);	/* SPI1_MOSI */
+		at91_set_A_periph(AT91_PIN_PB14, 0);	/* SPI1_SPCK */
+
+		at91_clock_associate("spi1_clk", &at91cap9_spi1_device.dev, "spi_clk");
+		platform_device_register(&at91cap9_spi1_device);
+	}
+}
+#else
+void __init at91_add_device_spi(struct spi_board_info *devices, int nr_devices) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  RTT
+ * -------------------------------------------------------------------- */
+
+static struct platform_device at91cap9_rtt_device = {
+	.name		= "at91_rtt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_rtt(void)
+{
+	platform_device_register(&at91cap9_rtt_device);
+}
+
+
+/* --------------------------------------------------------------------
+ *  Watchdog
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_AT91SAM9_WATCHDOG) || defined(CONFIG_AT91SAM9_WATCHDOG_MODULE)
+static struct platform_device at91cap9_wdt_device = {
+	.name		= "at91_wdt",
+	.id		= -1,
+	.num_resources	= 0,
+};
+
+static void __init at91_add_device_watchdog(void)
+{
+	platform_device_register(&at91cap9_wdt_device);
+}
+#else
+static void __init at91_add_device_watchdog(void) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  AC97
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_SND_AT91_AC97) || defined(CONFIG_SND_AT91_AC97_MODULE)
+static u64 ac97_dmamask = DMA_BIT_MASK(32);
+static struct atmel_ac97_data ac97_data;
+
+static struct resource ac97_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_AC97C,
+		.end	= AT91CAP9_BASE_AC97C + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_AC97C,
+		.end	= AT91CAP9_ID_AC97C,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_ac97_device = {
+	.name		= "ac97c",
+	.id		= 1,
+	.dev		= {
+				.dma_mask		= &ac97_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &ac97_data,
+	},
+	.resource	= ac97_resources,
+	.num_resources	= ARRAY_SIZE(ac97_resources),
+};
+
+void __init at91_add_device_ac97(struct atmel_ac97_data *data)
+{
+	if (!data)
+		return;
+
+	at91_set_A_periph(AT91_PIN_PA6, 0);	/* AC97FS */
+	at91_set_A_periph(AT91_PIN_PA7, 0);	/* AC97CK */
+	at91_set_A_periph(AT91_PIN_PA8, 0);	/* AC97TX */
+	at91_set_A_periph(AT91_PIN_PA9, 0);	/* AC97RX */
+
+	/* reset */
+	if (data->reset_pin)
+		at91_set_gpio_output(data->reset_pin, 0);
+
+	ac97_data = *data;
+	platform_device_register(&at91cap9_ac97_device);
+}
+#else
+void __init at91_add_device_ac97(struct atmel_ac97_data *data) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  LCD Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_FB_ATMEL) || defined(CONFIG_FB_ATMEL_MODULE)
+static u64 lcdc_dmamask = DMA_BIT_MASK(32);
+static struct atmel_lcdfb_info lcdc_data;
+
+static struct resource lcdc_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_LCDC_BASE,
+		.end	= AT91CAP9_LCDC_BASE + SZ_4K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_LCDC,
+		.end	= AT91CAP9_ID_LCDC,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91_lcdc_device = {
+	.name		= "atmel_lcdfb",
+	.id		= 0,
+	.dev		= {
+				.dma_mask		= &lcdc_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &lcdc_data,
+	},
+	.resource	= lcdc_resources,
+	.num_resources	= ARRAY_SIZE(lcdc_resources),
+};
+
+void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
+{
+	if (!data)
+		return;
+
+	at91_set_A_periph(AT91_PIN_PC1, 0);	/* LCDHSYNC */
+	at91_set_A_periph(AT91_PIN_PC2, 0);	/* LCDDOTCK */
+	at91_set_A_periph(AT91_PIN_PC3, 0);	/* LCDDEN */
+	at91_set_B_periph(AT91_PIN_PB9, 0);	/* LCDCC */
+	at91_set_A_periph(AT91_PIN_PC6, 0);	/* LCDD2 */
+	at91_set_A_periph(AT91_PIN_PC7, 0);	/* LCDD3 */
+	at91_set_A_periph(AT91_PIN_PC8, 0);	/* LCDD4 */
+	at91_set_A_periph(AT91_PIN_PC9, 0);	/* LCDD5 */
+	at91_set_A_periph(AT91_PIN_PC10, 0);	/* LCDD6 */
+	at91_set_A_periph(AT91_PIN_PC11, 0);	/* LCDD7 */
+	at91_set_A_periph(AT91_PIN_PC14, 0);	/* LCDD10 */
+	at91_set_A_periph(AT91_PIN_PC15, 0);	/* LCDD11 */
+	at91_set_A_periph(AT91_PIN_PC16, 0);	/* LCDD12 */
+	at91_set_A_periph(AT91_PIN_PC17, 0);	/* LCDD13 */
+	at91_set_A_periph(AT91_PIN_PC18, 0);	/* LCDD14 */
+	at91_set_A_periph(AT91_PIN_PC19, 0);	/* LCDD15 */
+	at91_set_A_periph(AT91_PIN_PC22, 0);	/* LCDD18 */
+	at91_set_A_periph(AT91_PIN_PC23, 0);	/* LCDD19 */
+	at91_set_A_periph(AT91_PIN_PC24, 0);	/* LCDD20 */
+	at91_set_A_periph(AT91_PIN_PC25, 0);	/* LCDD21 */
+	at91_set_A_periph(AT91_PIN_PC26, 0);	/* LCDD22 */
+	at91_set_A_periph(AT91_PIN_PC27, 0);	/* LCDD23 */
+
+	lcdc_data = *data;
+	platform_device_register(&at91_lcdc_device);
+}
+#else
+void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  SSC -- Synchronous Serial Controller
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_ATMEL_SSC) || defined(CONFIG_ATMEL_SSC_MODULE)
+static u64 ssc0_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc0_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_SSC0,
+		.end	= AT91CAP9_BASE_SSC0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_SSC0,
+		.end	= AT91CAP9_ID_SSC0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_ssc0_device = {
+	.name	= "ssc",
+	.id	= 0,
+	.dev	= {
+		.dma_mask		= &ssc0_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc0_resources,
+	.num_resources	= ARRAY_SIZE(ssc0_resources),
+};
+
+static inline void configure_ssc0_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB0, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB1, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB2, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB3, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB4, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB5, 1);
+}
+
+static u64 ssc1_dmamask = DMA_BIT_MASK(32);
+
+static struct resource ssc1_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_SSC1,
+		.end	= AT91CAP9_BASE_SSC1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_SSC1,
+		.end	= AT91CAP9_ID_SSC1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device at91cap9_ssc1_device = {
+	.name	= "ssc",
+	.id	= 1,
+	.dev	= {
+		.dma_mask		= &ssc1_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+	.resource	= ssc1_resources,
+	.num_resources	= ARRAY_SIZE(ssc1_resources),
+};
+
+static inline void configure_ssc1_pins(unsigned pins)
+{
+	if (pins & ATMEL_SSC_TF)
+		at91_set_A_periph(AT91_PIN_PB6, 1);
+	if (pins & ATMEL_SSC_TK)
+		at91_set_A_periph(AT91_PIN_PB7, 1);
+	if (pins & ATMEL_SSC_TD)
+		at91_set_A_periph(AT91_PIN_PB8, 1);
+	if (pins & ATMEL_SSC_RD)
+		at91_set_A_periph(AT91_PIN_PB9, 1);
+	if (pins & ATMEL_SSC_RK)
+		at91_set_A_periph(AT91_PIN_PB10, 1);
+	if (pins & ATMEL_SSC_RF)
+		at91_set_A_periph(AT91_PIN_PB11, 1);
+}
+
+/*
+ * SSC controllers are accessed through library code, instead of any
+ * kind of all-singing/all-dancing driver.  For example one could be
+ * used by a particular I2S audio codec's driver, while another one
+ * on the same system might be used by a custom data capture driver.
+ */
+void __init at91_add_device_ssc(unsigned id, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	/*
+	 * NOTE: caller is responsible for passing information matching
+	 * "pins" to whatever will be using each particular controller.
+	 */
+	switch (id) {
+	case AT91CAP9_ID_SSC0:
+		pdev = &at91cap9_ssc0_device;
+		configure_ssc0_pins(pins);
+		at91_clock_associate("ssc0_clk", &pdev->dev, "ssc");
+		break;
+	case AT91CAP9_ID_SSC1:
+		pdev = &at91cap9_ssc1_device;
+		configure_ssc1_pins(pins);
+		at91_clock_associate("ssc1_clk", &pdev->dev, "ssc");
+		break;
+	default:
+		return;
+	}
+
+	platform_device_register(pdev);
+}
+
+#else
+void __init at91_add_device_ssc(unsigned id, unsigned pins) {}
+#endif
+
+
+/* --------------------------------------------------------------------
+ *  UART
+ * -------------------------------------------------------------------- */
+
+#if defined(CONFIG_SERIAL_ATMEL)
+static struct resource dbgu_resources[] = {
+	[0] = {
+		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
+		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91_ID_SYS,
+		.end	= AT91_ID_SYS,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct atmel_uart_data dbgu_data = {
+	.use_dma_tx	= 0,
+	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
+	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
+};
+
+static u64 dbgu_dmamask = DMA_BIT_MASK(32);
+
+static struct platform_device at91cap9_dbgu_device = {
+	.name		= "atmel_usart",
+	.id		= 0,
+	.dev		= {
+				.dma_mask		= &dbgu_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &dbgu_data,
+	},
+	.resource	= dbgu_resources,
+	.num_resources	= ARRAY_SIZE(dbgu_resources),
+};
+
+static inline void configure_dbgu_pins(void)
+{
+	at91_set_A_periph(AT91_PIN_PC30, 0);		/* DRXD */
+	at91_set_A_periph(AT91_PIN_PC31, 1);		/* DTXD */
+}
+
+static struct resource uart0_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_US0,
+		.end	= AT91CAP9_BASE_US0 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_US0,
+		.end	= AT91CAP9_ID_US0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct atmel_uart_data uart0_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+
+static u64 uart0_dmamask = DMA_BIT_MASK(32);
+
+static struct platform_device at91cap9_uart0_device = {
+	.name		= "atmel_usart",
+	.id		= 1,
+	.dev		= {
+				.dma_mask		= &uart0_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &uart0_data,
+	},
+	.resource	= uart0_resources,
+	.num_resources	= ARRAY_SIZE(uart0_resources),
+};
+
+static inline void configure_usart0_pins(unsigned pins)
+{
+	at91_set_A_periph(AT91_PIN_PA22, 1);		/* TXD0 */
+	at91_set_A_periph(AT91_PIN_PA23, 0);		/* RXD0 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_A_periph(AT91_PIN_PA24, 0);	/* RTS0 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_A_periph(AT91_PIN_PA25, 0);	/* CTS0 */
+}
+
+static struct resource uart1_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_US1,
+		.end	= AT91CAP9_BASE_US1 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_US1,
+		.end	= AT91CAP9_ID_US1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct atmel_uart_data uart1_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+
+static u64 uart1_dmamask = DMA_BIT_MASK(32);
+
+static struct platform_device at91cap9_uart1_device = {
+	.name		= "atmel_usart",
+	.id		= 2,
+	.dev		= {
+				.dma_mask		= &uart1_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &uart1_data,
+	},
+	.resource	= uart1_resources,
+	.num_resources	= ARRAY_SIZE(uart1_resources),
+};
+
+static inline void configure_usart1_pins(unsigned pins)
+{
+	at91_set_A_periph(AT91_PIN_PD0, 1);		/* TXD1 */
+	at91_set_A_periph(AT91_PIN_PD1, 0);		/* RXD1 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PD7, 0);	/* RTS1 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PD8, 0);	/* CTS1 */
+}
+
+static struct resource uart2_resources[] = {
+	[0] = {
+		.start	= AT91CAP9_BASE_US2,
+		.end	= AT91CAP9_BASE_US2 + SZ_16K - 1,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= AT91CAP9_ID_US2,
+		.end	= AT91CAP9_ID_US2,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct atmel_uart_data uart2_data = {
+	.use_dma_tx	= 1,
+	.use_dma_rx	= 1,
+};
+
+static u64 uart2_dmamask = DMA_BIT_MASK(32);
+
+static struct platform_device at91cap9_uart2_device = {
+	.name		= "atmel_usart",
+	.id		= 3,
+	.dev		= {
+				.dma_mask		= &uart2_dmamask,
+				.coherent_dma_mask	= DMA_BIT_MASK(32),
+				.platform_data		= &uart2_data,
+	},
+	.resource	= uart2_resources,
+	.num_resources	= ARRAY_SIZE(uart2_resources),
+};
+
+static inline void configure_usart2_pins(unsigned pins)
+{
+	at91_set_A_periph(AT91_PIN_PD2, 1);		/* TXD2 */
+	at91_set_A_periph(AT91_PIN_PD3, 0);		/* RXD2 */
+
+	if (pins & ATMEL_UART_RTS)
+		at91_set_B_periph(AT91_PIN_PD5, 0);	/* RTS2 */
+	if (pins & ATMEL_UART_CTS)
+		at91_set_B_periph(AT91_PIN_PD6, 0);	/* CTS2 */
+}
+
+static struct platform_device *at91_uarts[ATMEL_MAX_UART];	/* the UARTs to use */
+struct platform_device *atmel_default_console_device;	/* the serial console device */
+
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+		case 0:		/* DBGU */
+			pdev = &at91cap9_dbgu_device;
+			configure_dbgu_pins();
+			at91_clock_associate("mck", &pdev->dev, "usart");
+			break;
+		case AT91CAP9_ID_US0:
+			pdev = &at91cap9_uart0_device;
+			configure_usart0_pins(pins);
+			at91_clock_associate("usart0_clk", &pdev->dev, "usart");
+			break;
+		case AT91CAP9_ID_US1:
+			pdev = &at91cap9_uart1_device;
+			configure_usart1_pins(pins);
+			at91_clock_associate("usart1_clk", &pdev->dev, "usart");
+			break;
+		case AT91CAP9_ID_US2:
+			pdev = &at91cap9_uart2_device;
+			configure_usart2_pins(pins);
+			at91_clock_associate("usart2_clk", &pdev->dev, "usart");
+			break;
+		default:
+			return;
+	}
+	pdev->id = portnr;		/* update to mapped ID */
+
+	if (portnr < ATMEL_MAX_UART)
+		at91_uarts[portnr] = pdev;
+}
+
+void __init at91_set_serial_console(unsigned portnr)
+{
+	if (portnr < ATMEL_MAX_UART)
+		atmel_default_console_device = at91_uarts[portnr];
+	if (!atmel_default_console_device)
+		printk(KERN_INFO "AT91: No default serial console defined.\n");
+}
+
+void __init at91_add_device_serial(void)
+{
+	int i;
+
+	for (i = 0; i < ATMEL_MAX_UART; i++) {
+		if (at91_uarts[i])
+			platform_device_register(at91_uarts[i]);
+	}
+}
+#else
+void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins) {}
+void __init at91_set_serial_console(unsigned portnr) {}
+void __init at91_add_device_serial(void) {}
+#endif
+
+
+/* -------------------------------------------------------------------- */
+/*
+ * These devices are always present and don't need any board-specific
+ * setup.
+ */
+static int __init at91_add_standard_devices(void)
+{
+	at91_add_device_rtt();
+	at91_add_device_watchdog();
+	return 0;
+}
+
+arch_initcall(at91_add_standard_devices);
diff --git a/arch/arm/mach-at91/clock.c b/arch/arm/mach-at91/clock.c
index 57c3b647ce83..ec76eeaafd45 100644
--- a/arch/arm/mach-at91/clock.c
+++ b/arch/arm/mach-at91/clock.c
@@ -574,6 +574,8 @@ int __init at91_clock_init(unsigned long main_clock)
 	} else if (cpu_is_at91sam9260() || cpu_is_at91sam9261() || cpu_is_at91sam9263()) {
 		uhpck.pmc_mask = AT91SAM926x_PMC_UHP;
 		udpck.pmc_mask = AT91SAM926x_PMC_UDP;
+	} else if (cpu_is_at91cap9()) {
+		uhpck.pmc_mask = AT91CAP9_PMC_UHP;
 	}
 	at91_sys_write(AT91_CKGR_PLLBR, 0);
 
diff --git a/arch/arm/mach-at91/generic.h b/arch/arm/mach-at91/generic.h
index 77d4c0a37842..b5daf7f5e011 100644
--- a/arch/arm/mach-at91/generic.h
+++ b/arch/arm/mach-at91/generic.h
@@ -15,6 +15,7 @@ extern void __init at91sam9261_initialize(unsigned long main_clock);
 extern void __init at91sam9263_initialize(unsigned long main_clock);
 extern void __init at91sam9rl_initialize(unsigned long main_clock);
 extern void __init at91x40_initialize(unsigned long main_clock);
+extern void __init at91cap9_initialize(unsigned long main_clock);
 
  /* Interrupts */
 extern void __init at91rm9200_init_interrupts(unsigned int priority[]);
@@ -23,6 +24,7 @@ extern void __init at91sam9261_init_interrupts(unsigned int priority[]);
 extern void __init at91sam9263_init_interrupts(unsigned int priority[]);
 extern void __init at91sam9rl_init_interrupts(unsigned int priority[]);
 extern void __init at91x40_init_interrupts(unsigned int priority[]);
+extern void __init at91cap9_init_interrupts(unsigned int priority[]);
 extern void __init at91_aic_init(unsigned int priority[]);
 
  /* Timer */
diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
index 98cb61482917..4b120cc36135 100644
--- a/arch/arm/mach-at91/pm.c
+++ b/arch/arm/mach-at91/pm.c
@@ -80,6 +80,11 @@ static int at91_pm_verify_clocks(void)
 			pr_debug("AT91: PM - Suspend-to-RAM with USB still active\n");
 			return 0;
 		}
+	} else if (cpu_is_at91cap9()) {
+		if ((scsr & AT91CAP9_PMC_UHP) != 0) {
+			pr_debug("AT91: PM - Suspend-to-RAM with USB still active\n");
+			return 0;
+		}
 	}
 
 #ifdef CONFIG_AT91_PROGRAMMABLE_CLOCKS
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 7868f4dc1d00..1bf2a7bfe09d 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -171,8 +171,8 @@ config CPU_ARM925T
 # ARM926T
 config CPU_ARM926T
 	bool "Support ARM926T processor"
-	depends on ARCH_INTEGRATOR || ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || MACH_REALVIEW_EB || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_NS9XXX || ARCH_DAVINCI
-	default y if ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_NS9XXX || ARCH_DAVINCI
+	depends on ARCH_INTEGRATOR || ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || MACH_REALVIEW_EB || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
+	default y if ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
 	select CPU_32v5
 	select CPU_ABRT_EV5TJ
 	select CPU_CACHE_VIVT
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 9af05a2f4af3..a6728661c416 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -212,7 +212,7 @@ config MII
 
 config MACB
 	tristate "Atmel MACB support"
-	depends on AVR32 || ARCH_AT91SAM9260 || ARCH_AT91SAM9263
+	depends on AVR32 || ARCH_AT91SAM9260 || ARCH_AT91SAM9263 || ARCH_AT91CAP9
 	select PHYLIB
 	help
 	  The Atmel MACB ethernet interface is found on many AT32 and AT91
diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index f81d08d6538b..77a3759d6fc7 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -308,7 +308,7 @@ config USB_S3C2410_DEBUG
 
 config USB_GADGET_AT91
 	boolean "AT91 USB Device Port"
-	depends on ARCH_AT91 && !ARCH_AT91SAM9RL
+	depends on ARCH_AT91 && !ARCH_AT91SAM9RL && !ARCH_AT91CAP9
 	select USB_GADGET_SELECTED
 	help
 	   Many Atmel AT91 processors (such as the AT91RM2000) have a
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 5b3dbcfcda48..758435f8a6f8 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -889,7 +889,7 @@ config FB_S1D13XXX
 
 config FB_ATMEL
 	tristate "AT91/AT32 LCD Controller support"
-	depends on FB && (ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || AVR32)
+	depends on FB && (ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91CAP9 || AVR32)
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index 7c30cc8df71e..f8e711147501 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -30,7 +30,7 @@
 #define ATMEL_LCDC_CVAL_DEFAULT		0xc8
 #define ATMEL_LCDC_DMA_BURST_LEN	8
 
-#if defined(CONFIG_ARCH_AT91SAM9263)
+#if defined(CONFIG_ARCH_AT91SAM9263) || defined(CONFIG_ARCH_AT91CAP9)
 #define ATMEL_LCDC_FIFO_SIZE		2048
 #else
 #define ATMEL_LCDC_FIFO_SIZE		512
diff --git a/include/asm-arm/arch-at91/at91_pmc.h b/include/asm-arm/arch-at91/at91_pmc.h
index 34804eaab4f2..52cd8e5dabc9 100644
--- a/include/asm-arm/arch-at91/at91_pmc.h
+++ b/include/asm-arm/arch-at91/at91_pmc.h
@@ -25,6 +25,7 @@
 #define		AT91RM9200_PMC_MCKUDP	(1 <<  2)		/* USB Device Port Master Clock Automatic Disable on Suspend [AT91RM9200 only] */
 #define		AT91RM9200_PMC_UHP	(1 <<  4)		/* USB Host Port Clock [AT91RM9200 only] */
 #define		AT91SAM926x_PMC_UHP	(1 <<  6)		/* USB Host Port Clock [AT91SAM926x only] */
+#define		AT91CAP9_PMC_UHP	(1 <<  6)		/* USB Host Port Clock [AT91CAP9 only] */
 #define		AT91SAM926x_PMC_UDP	(1 <<  7)		/* USB Devcice Port Clock [AT91SAM926x only] */
 #define		AT91_PMC_PCK0		(1 <<  8)		/* Programmable Clock 0 */
 #define		AT91_PMC_PCK1		(1 <<  9)		/* Programmable Clock 1 */
@@ -37,7 +38,7 @@
 #define	AT91_PMC_PCDR		(AT91_PMC + 0x14)	/* Peripheral Clock Disable Register */
 #define	AT91_PMC_PCSR		(AT91_PMC + 0x18)	/* Peripheral Clock Status Register */
 
-#define	AT91_CKGR_UCKR		(AT91_PMC + 0x1C)	/* UTMI Clock Register [SAM9RL only] */
+#define	AT91_CKGR_UCKR		(AT91_PMC + 0x1C)	/* UTMI Clock Register [SAM9RL, CAP9] */
 
 #define	AT91_CKGR_MOR		(AT91_PMC + 0x20)	/* Main Oscillator Register [not on SAM9RL] */
 #define		AT91_PMC_MOSCEN		(1    << 0)		/* Main Oscillator Enable */
@@ -54,6 +55,10 @@
 #define		AT91_PMC_PLLCOUNT	(0x3f  <<  8)		/* PLL Counter */
 #define		AT91_PMC_OUT		(3     << 14)		/* PLL Clock Frequency Range */
 #define		AT91_PMC_MUL		(0x7ff << 16)		/* PLL Multiplier */
+#define		AT91_PMC_USBDIV		(3     << 28)		/* USB Divisor (PLLB only) */
+#define			AT91_PMC_USBDIV_1		(0 << 28)
+#define			AT91_PMC_USBDIV_2		(1 << 28)
+#define			AT91_PMC_USBDIV_4		(2 << 28)
 #define		AT91_PMC_USB96M		(1     << 28)		/* Divider by 2 Enable (PLLB only) */
 
 #define	AT91_PMC_MCKR		(AT91_PMC + 0x30)	/* Master Clock Register */
diff --git a/include/asm-arm/arch-at91/at91cap9.h b/include/asm-arm/arch-at91/at91cap9.h
new file mode 100644
index 000000000000..73e1fcf4a0aa
--- /dev/null
+++ b/include/asm-arm/arch-at91/at91cap9.h
@@ -0,0 +1,121 @@
+/*
+ * include/asm-arm/arch-at91/at91cap9.h
+ *
+ *  Copyright (C) 2007 Stelian Pop <stelian.pop@leadtechdesign.com>
+ *  Copyright (C) 2007 Lead Tech Design <www.leadtechdesign.com>
+ *  Copyright (C) 2007 Atmel Corporation.
+ *
+ * Common definitions.
+ * Based on AT91CAP9 datasheet revision B (Preliminary).
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91CAP9_H
+#define AT91CAP9_H
+
+/*
+ * Peripheral identifiers/interrupts.
+ */
+#define AT91_ID_FIQ		0	/* Advanced Interrupt Controller (FIQ) */
+#define AT91_ID_SYS		1	/* System Peripherals */
+#define AT91CAP9_ID_PIOABCD	2	/* Parallel IO Controller A, B, C and D */
+#define AT91CAP9_ID_MPB0	3	/* MP Block Peripheral 0 */
+#define AT91CAP9_ID_MPB1	4	/* MP Block Peripheral 1 */
+#define AT91CAP9_ID_MPB2	5	/* MP Block Peripheral 2 */
+#define AT91CAP9_ID_MPB3	6	/* MP Block Peripheral 3 */
+#define AT91CAP9_ID_MPB4	7	/* MP Block Peripheral 4 */
+#define AT91CAP9_ID_US0		8	/* USART 0 */
+#define AT91CAP9_ID_US1		9	/* USART 1 */
+#define AT91CAP9_ID_US2		10	/* USART 2 */
+#define AT91CAP9_ID_MCI0	11	/* Multimedia Card Interface 0 */
+#define AT91CAP9_ID_MCI1	12	/* Multimedia Card Interface 1 */
+#define AT91CAP9_ID_CAN		13	/* CAN */
+#define AT91CAP9_ID_TWI		14	/* Two-Wire Interface */
+#define AT91CAP9_ID_SPI0	15	/* Serial Peripheral Interface 0 */
+#define AT91CAP9_ID_SPI1	16	/* Serial Peripheral Interface 0 */
+#define AT91CAP9_ID_SSC0	17	/* Serial Synchronous Controller 0 */
+#define AT91CAP9_ID_SSC1	18	/* Serial Synchronous Controller 1 */
+#define AT91CAP9_ID_AC97C	19	/* AC97 Controller */
+#define AT91CAP9_ID_TCB		20	/* Timer Counter 0, 1 and 2 */
+#define AT91CAP9_ID_PWMC	21	/* Pulse Width Modulation Controller */
+#define AT91CAP9_ID_EMAC	22	/* Ethernet */
+#define AT91CAP9_ID_AESTDES	23	/* Advanced Encryption Standard, Triple DES */
+#define AT91CAP9_ID_ADC		24	/* Analog-to-Digital Converter */
+#define AT91CAP9_ID_ISI		25	/* Image Sensor Interface */
+#define AT91CAP9_ID_LCDC	26	/* LCD Controller */
+#define AT91CAP9_ID_DMA		27	/* DMA Controller */
+#define AT91CAP9_ID_UDPHS	28	/* USB High Speed Device Port */
+#define AT91CAP9_ID_UHP		29	/* USB Host Port */
+#define AT91CAP9_ID_IRQ0	30	/* Advanced Interrupt Controller (IRQ0) */
+#define AT91CAP9_ID_IRQ1	31	/* Advanced Interrupt Controller (IRQ1) */
+
+/*
+ * User Peripheral physical base addresses.
+ */
+#define AT91CAP9_BASE_UDPHS		0xfff78000
+#define AT91CAP9_BASE_TCB0		0xfff7c000
+#define AT91CAP9_BASE_TC0		0xfff7c000
+#define AT91CAP9_BASE_TC1		0xfff7c040
+#define AT91CAP9_BASE_TC2		0xfff7c080
+#define AT91CAP9_BASE_MCI0		0xfff80000
+#define AT91CAP9_BASE_MCI1		0xfff84000
+#define AT91CAP9_BASE_TWI		0xfff88000
+#define AT91CAP9_BASE_US0		0xfff8c000
+#define AT91CAP9_BASE_US1		0xfff90000
+#define AT91CAP9_BASE_US2		0xfff94000
+#define AT91CAP9_BASE_SSC0		0xfff98000
+#define AT91CAP9_BASE_SSC1		0xfff9c000
+#define AT91CAP9_BASE_AC97C		0xfffa0000
+#define AT91CAP9_BASE_SPI0		0xfffa4000
+#define AT91CAP9_BASE_SPI1		0xfffa8000
+#define AT91CAP9_BASE_CAN		0xfffac000
+#define AT91CAP9_BASE_PWMC		0xfffb8000
+#define AT91CAP9_BASE_EMAC		0xfffbc000
+#define AT91CAP9_BASE_ADC		0xfffc0000
+#define AT91CAP9_BASE_ISI		0xfffc4000
+#define AT91_BASE_SYS			0xffffe200
+
+/*
+ * System Peripherals (offset from AT91_BASE_SYS)
+ */
+#define AT91_ECC	(0xffffe200 - AT91_BASE_SYS)
+#define AT91_BCRAMC	(0xffffe400 - AT91_BASE_SYS)
+#define AT91_DDRSDRC	(0xffffe600 - AT91_BASE_SYS)
+#define AT91_SMC	(0xffffe800 - AT91_BASE_SYS)
+#define AT91_MATRIX	(0xffffea00 - AT91_BASE_SYS)
+#define AT91_CCFG	(0xffffeb10 - AT91_BASE_SYS)
+#define AT91_DMA	(0xffffec00 - AT91_BASE_SYS)
+#define AT91_DBGU	(0xffffee00 - AT91_BASE_SYS)
+#define AT91_AIC	(0xfffff000 - AT91_BASE_SYS)
+#define AT91_PIOA	(0xfffff200 - AT91_BASE_SYS)
+#define AT91_PIOB	(0xfffff400 - AT91_BASE_SYS)
+#define AT91_PIOC	(0xfffff600 - AT91_BASE_SYS)
+#define AT91_PIOD	(0xfffff800 - AT91_BASE_SYS)
+#define AT91_PMC	(0xfffffc00 - AT91_BASE_SYS)
+#define AT91_RSTC	(0xfffffd00 - AT91_BASE_SYS)
+#define AT91_SHDC	(0xfffffd10 - AT91_BASE_SYS)
+#define AT91_RTT	(0xfffffd20 - AT91_BASE_SYS)
+#define AT91_PIT	(0xfffffd30 - AT91_BASE_SYS)
+#define AT91_WDT	(0xfffffd40 - AT91_BASE_SYS)
+#define AT91_GPBR	(0xfffffd50 - AT91_BASE_SYS)
+
+/*
+ * Internal Memory.
+ */
+#define AT91CAP9_SRAM_BASE	0x00100000	/* Internal SRAM base address */
+#define AT91CAP9_SRAM_SIZE	(32 * SZ_1K)	/* Internal SRAM size (32Kb) */
+
+#define AT91CAP9_ROM_BASE	0x00400000	/* Internal ROM base address */
+#define AT91CAP9_ROM_SIZE	(32 * SZ_1K)	/* Internal ROM size (32Kb) */
+
+#define AT91CAP9_LCDC_BASE	0x00500000	/* LCD Controller */
+#define AT91CAP9_UDPHS_BASE	0x00600000	/* USB High Speed Device Port */
+#define AT91CAP9_UHP_BASE	0x00700000	/* USB Host controller */
+
+#define CONFIG_DRAM_BASE	AT91_CHIPSELECT_6
+
+#endif
diff --git a/include/asm-arm/arch-at91/at91cap9_matrix.h b/include/asm-arm/arch-at91/at91cap9_matrix.h
new file mode 100644
index 000000000000..a641686b6c3d
--- /dev/null
+++ b/include/asm-arm/arch-at91/at91cap9_matrix.h
@@ -0,0 +1,132 @@
+/*
+ * include/asm-arm/arch-at91/at91cap9_matrix.h
+ *
+ *  Copyright (C) 2007 Stelian Pop <stelian.pop@leadtechdesign.com>
+ *  Copyright (C) 2007 Lead Tech Design <www.leadtechdesign.com>
+ *  Copyright (C) 2006 Atmel Corporation.
+ *
+ * Memory Controllers (MATRIX, EBI) - System peripherals registers.
+ * Based on AT91CAP9 datasheet revision B (Preliminary).
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91CAP9_MATRIX_H
+#define AT91CAP9_MATRIX_H
+
+#define AT91_MATRIX_MCFG0	(AT91_MATRIX + 0x00)	/* Master Configuration Register 0 */
+#define AT91_MATRIX_MCFG1	(AT91_MATRIX + 0x04)	/* Master Configuration Register 1 */
+#define AT91_MATRIX_MCFG2	(AT91_MATRIX + 0x08)	/* Master Configuration Register 2 */
+#define AT91_MATRIX_MCFG3	(AT91_MATRIX + 0x0C)	/* Master Configuration Register 3 */
+#define AT91_MATRIX_MCFG4	(AT91_MATRIX + 0x10)	/* Master Configuration Register 4 */
+#define AT91_MATRIX_MCFG5	(AT91_MATRIX + 0x14)	/* Master Configuration Register 5 */
+#define AT91_MATRIX_MCFG6	(AT91_MATRIX + 0x18)	/* Master Configuration Register 6 */
+#define AT91_MATRIX_MCFG7	(AT91_MATRIX + 0x1C)	/* Master Configuration Register 7 */
+#define AT91_MATRIX_MCFG8	(AT91_MATRIX + 0x20)	/* Master Configuration Register 8 */
+#define AT91_MATRIX_MCFG9	(AT91_MATRIX + 0x24)	/* Master Configuration Register 9 */
+#define AT91_MATRIX_MCFG10	(AT91_MATRIX + 0x28)	/* Master Configuration Register 10 */
+#define AT91_MATRIX_MCFG11	(AT91_MATRIX + 0x2C)	/* Master Configuration Register 11 */
+#define		AT91_MATRIX_ULBT	(7 << 0)	/* Undefined Length Burst Type */
+#define			AT91_MATRIX_ULBT_INFINITE	(0 << 0)
+#define			AT91_MATRIX_ULBT_SINGLE		(1 << 0)
+#define			AT91_MATRIX_ULBT_FOUR		(2 << 0)
+#define			AT91_MATRIX_ULBT_EIGHT		(3 << 0)
+#define			AT91_MATRIX_ULBT_SIXTEEN	(4 << 0)
+
+#define AT91_MATRIX_SCFG0	(AT91_MATRIX + 0x40)	/* Slave Configuration Register 0 */
+#define AT91_MATRIX_SCFG1	(AT91_MATRIX + 0x44)	/* Slave Configuration Register 1 */
+#define AT91_MATRIX_SCFG2	(AT91_MATRIX + 0x48)	/* Slave Configuration Register 2 */
+#define AT91_MATRIX_SCFG3	(AT91_MATRIX + 0x4C)	/* Slave Configuration Register 3 */
+#define AT91_MATRIX_SCFG4	(AT91_MATRIX + 0x50)	/* Slave Configuration Register 4 */
+#define AT91_MATRIX_SCFG5	(AT91_MATRIX + 0x54)	/* Slave Configuration Register 5 */
+#define AT91_MATRIX_SCFG6	(AT91_MATRIX + 0x58)	/* Slave Configuration Register 6 */
+#define AT91_MATRIX_SCFG7	(AT91_MATRIX + 0x5C)	/* Slave Configuration Register 7 */
+#define AT91_MATRIX_SCFG8	(AT91_MATRIX + 0x60)	/* Slave Configuration Register 8 */
+#define AT91_MATRIX_SCFG9	(AT91_MATRIX + 0x64)	/* Slave Configuration Register 9 */
+#define		AT91_MATRIX_SLOT_CYCLE		(0xff << 0)	/* Maximum Number of Allowed Cycles for a Burst */
+#define		AT91_MATRIX_DEFMSTR_TYPE	(3    << 16)	/* Default Master Type */
+#define			AT91_MATRIX_DEFMSTR_TYPE_NONE	(0 << 16)
+#define			AT91_MATRIX_DEFMSTR_TYPE_LAST	(1 << 16)
+#define			AT91_MATRIX_DEFMSTR_TYPE_FIXED	(2 << 16)
+#define		AT91_MATRIX_FIXED_DEFMSTR	(0xf  << 18)	/* Fixed Index of Default Master */
+#define		AT91_MATRIX_ARBT		(3    << 24)	/* Arbitration Type */
+#define			AT91_MATRIX_ARBT_ROUND_ROBIN	(0 << 24)
+#define			AT91_MATRIX_ARBT_FIXED_PRIORITY	(1 << 24)
+
+#define AT91_MATRIX_PRAS0	(AT91_MATRIX + 0x80)	/* Priority Register A for Slave 0 */
+#define AT91_MATRIX_PRBS0	(AT91_MATRIX + 0x84)	/* Priority Register B for Slave 0 */
+#define AT91_MATRIX_PRAS1	(AT91_MATRIX + 0x88)	/* Priority Register A for Slave 1 */
+#define AT91_MATRIX_PRBS1	(AT91_MATRIX + 0x8C)	/* Priority Register B for Slave 1 */
+#define AT91_MATRIX_PRAS2	(AT91_MATRIX + 0x90)	/* Priority Register A for Slave 2 */
+#define AT91_MATRIX_PRBS2	(AT91_MATRIX + 0x94)	/* Priority Register B for Slave 2 */
+#define AT91_MATRIX_PRAS3	(AT91_MATRIX + 0x98)	/* Priority Register A for Slave 3 */
+#define AT91_MATRIX_PRBS3	(AT91_MATRIX + 0x9C)	/* Priority Register B for Slave 3 */
+#define AT91_MATRIX_PRAS4	(AT91_MATRIX + 0xA0)	/* Priority Register A for Slave 4 */
+#define AT91_MATRIX_PRBS4	(AT91_MATRIX + 0xA4)	/* Priority Register B for Slave 4 */
+#define AT91_MATRIX_PRAS5	(AT91_MATRIX + 0xA8)	/* Priority Register A for Slave 5 */
+#define AT91_MATRIX_PRBS5	(AT91_MATRIX + 0xAC)	/* Priority Register B for Slave 5 */
+#define AT91_MATRIX_PRAS6	(AT91_MATRIX + 0xB0)	/* Priority Register A for Slave 6 */
+#define AT91_MATRIX_PRBS6	(AT91_MATRIX + 0xB4)	/* Priority Register B for Slave 6 */
+#define AT91_MATRIX_PRAS7	(AT91_MATRIX + 0xB8)	/* Priority Register A for Slave 7 */
+#define AT91_MATRIX_PRBS7	(AT91_MATRIX + 0xBC)	/* Priority Register B for Slave 7 */
+#define AT91_MATRIX_PRAS8	(AT91_MATRIX + 0xC0)	/* Priority Register A for Slave 8 */
+#define AT91_MATRIX_PRBS8	(AT91_MATRIX + 0xC4)	/* Priority Register B for Slave 8 */
+#define AT91_MATRIX_PRAS9	(AT91_MATRIX + 0xC8)	/* Priority Register A for Slave 9 */
+#define AT91_MATRIX_PRBS9	(AT91_MATRIX + 0xCC)	/* Priority Register B for Slave 9 */
+#define		AT91_MATRIX_M0PR		(3 << 0)	/* Master 0 Priority */
+#define		AT91_MATRIX_M1PR		(3 << 4)	/* Master 1 Priority */
+#define		AT91_MATRIX_M2PR		(3 << 8)	/* Master 2 Priority */
+#define		AT91_MATRIX_M3PR		(3 << 12)	/* Master 3 Priority */
+#define		AT91_MATRIX_M4PR		(3 << 16)	/* Master 4 Priority */
+#define		AT91_MATRIX_M5PR		(3 << 20)	/* Master 5 Priority */
+#define		AT91_MATRIX_M6PR		(3 << 24)	/* Master 6 Priority */
+#define		AT91_MATRIX_M7PR		(3 << 28)	/* Master 7 Priority */
+#define		AT91_MATRIX_M8PR		(3 << 0)	/* Master 8 Priority (in Register B) */
+#define		AT91_MATRIX_M9PR		(3 << 4)	/* Master 9 Priority (in Register B) */
+#define		AT91_MATRIX_M10PR		(3 << 8)	/* Master 10 Priority (in Register B) */
+#define		AT91_MATRIX_M11PR		(3 << 12)	/* Master 11 Priority (in Register B) */
+
+#define AT91_MATRIX_MRCR	(AT91_MATRIX + 0x100)	/* Master Remap Control Register */
+#define		AT91_MATRIX_RCB0		(1 << 0)	/* Remap Command for AHB Master 0 (ARM926EJ-S Instruction Master) */
+#define		AT91_MATRIX_RCB1		(1 << 1)	/* Remap Command for AHB Master 1 (ARM926EJ-S Data Master) */
+#define		AT91_MATRIX_RCB2		(1 << 2)
+#define		AT91_MATRIX_RCB3		(1 << 3)
+#define		AT91_MATRIX_RCB4		(1 << 4)
+#define		AT91_MATRIX_RCB5		(1 << 5)
+#define		AT91_MATRIX_RCB6		(1 << 6)
+#define		AT91_MATRIX_RCB7		(1 << 7)
+#define		AT91_MATRIX_RCB8		(1 << 8)
+#define		AT91_MATRIX_RCB9		(1 << 9)
+#define		AT91_MATRIX_RCB10		(1 << 10)
+#define		AT91_MATRIX_RCB11		(1 << 11)
+
+#define AT91_MPBS0_SFR		(AT91_MATRIX + 0x114)	/* MPBlock Slave 0 Special Function Register */
+#define AT91_MPBS1_SFR		(AT91_MATRIX + 0x11C)	/* MPBlock Slave 1 Special Function Register */
+
+#define AT91_MATRIX_EBICSA	(AT91_MATRIX + 0x120)	/* EBI Chip Select Assignment Register */
+#define		AT91_MATRIX_EBI_CS1A		(1 << 1)	/* Chip Select 1 Assignment */
+#define			AT91_MATRIX_EBI_CS1A_SMC		(0 << 1)
+#define			AT91_MATRIX_EBI_CS1A_BCRAMC		(1 << 1)
+#define		AT91_MATRIX_EBI_CS3A		(1 << 3)	/* Chip Select 3 Assignment */
+#define			AT91_MATRIX_EBI_CS3A_SMC		(0 << 3)
+#define			AT91_MATRIX_EBI_CS3A_SMC_SMARTMEDIA	(1 << 3)
+#define		AT91_MATRIX_EBI_CS4A		(1 << 4)	/* Chip Select 4 Assignment */
+#define			AT91_MATRIX_EBI_CS4A_SMC		(0 << 4)
+#define			AT91_MATRIX_EBI_CS4A_SMC_CF1		(1 << 4)
+#define		AT91_MATRIX_EBI_CS5A		(1 << 5)	/* Chip Select 5 Assignment */
+#define			AT91_MATRIX_EBI_CS5A_SMC		(0 << 5)
+#define			AT91_MATRIX_EBI_CS5A_SMC_CF2		(1 << 5)
+#define		AT91_MATRIX_EBI_DBPUC		(1 << 8)	/* Data Bus Pull-up Configuration */
+#define		AT91_MATRIX_EBI_DQSPDC		(1 << 9)	/* Data Qualifier Strobe Pull-Down Configuration */
+#define		AT91_MATRIX_EBI_VDDIOMSEL	(1 << 16)	/* Memory voltage selection */
+#define			AT91_MATRIX_EBI_VDDIOMSEL_1_8V		(0 << 16)
+#define			AT91_MATRIX_EBI_VDDIOMSEL_3_3V		(1 << 16)
+
+#define AT91_MPBS2_SFR		(AT91_MATRIX + 0x12C)	/* MPBlock Slave 2 Special Function Register */
+#define AT91_MPBS3_SFR		(AT91_MATRIX + 0x130)	/* MPBlock Slave 3 Special Function Register */
+#define AT91_APB_SFR		(AT91_MATRIX + 0x134)	/* APB Bridge Special Function Register */
+
+#endif
diff --git a/include/asm-arm/arch-at91/at91sam9263_matrix.h b/include/asm-arm/arch-at91/at91sam9263_matrix.h
index 6fc6e4be624e..72f6e668e414 100644
--- a/include/asm-arm/arch-at91/at91sam9263_matrix.h
+++ b/include/asm-arm/arch-at91/at91sam9263_matrix.h
@@ -44,7 +44,7 @@
 #define			AT91_MATRIX_DEFMSTR_TYPE_NONE	(0 << 16)
 #define			AT91_MATRIX_DEFMSTR_TYPE_LAST	(1 << 16)
 #define			AT91_MATRIX_DEFMSTR_TYPE_FIXED	(2 << 16)
-#define		AT91_MATRIX_FIXED_DEFMSTR	(7    << 18)	/* Fixed Index of Default Master */
+#define		AT91_MATRIX_FIXED_DEFMSTR	(0xf  << 18)	/* Fixed Index of Default Master */
 #define		AT91_MATRIX_ARBT		(3    << 24)	/* Arbitration Type */
 #define			AT91_MATRIX_ARBT_ROUND_ROBIN	(0 << 24)
 #define			AT91_MATRIX_ARBT_FIXED_PRIORITY	(1 << 24)
diff --git a/include/asm-arm/arch-at91/at91sam9rl_matrix.h b/include/asm-arm/arch-at91/at91sam9rl_matrix.h
index b15f11b7c08d..84224174e6a1 100644
--- a/include/asm-arm/arch-at91/at91sam9rl_matrix.h
+++ b/include/asm-arm/arch-at91/at91sam9rl_matrix.h
@@ -38,7 +38,7 @@
 #define			AT91_MATRIX_DEFMSTR_TYPE_NONE	(0 << 16)
 #define			AT91_MATRIX_DEFMSTR_TYPE_LAST	(1 << 16)
 #define			AT91_MATRIX_DEFMSTR_TYPE_FIXED	(2 << 16)
-#define		AT91_MATRIX_FIXED_DEFMSTR	(7    << 18)	/* Fixed Index of Default Master */
+#define		AT91_MATRIX_FIXED_DEFMSTR	(0xf  << 18)	/* Fixed Index of Default Master */
 #define		AT91_MATRIX_ARBT		(3    << 24)	/* Arbitration Type */
 #define			AT91_MATRIX_ARBT_ROUND_ROBIN	(0 << 24)
 #define			AT91_MATRIX_ARBT_FIXED_PRIORITY	(1 << 24)
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index e88ad8621c17..55b07bd5316c 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -72,7 +72,7 @@ struct at91_eth_data {
 };
 extern void __init at91_add_device_eth(struct at91_eth_data *data);
 
-#if defined(CONFIG_ARCH_AT91SAM9260) || defined(CONFIG_ARCH_AT91SAM9263)
+#if defined(CONFIG_ARCH_AT91SAM9260) || defined(CONFIG_ARCH_AT91SAM9263) || defined(CONFIG_ARCH_AT91CAP9)
 #define eth_platform_data	at91_eth_data
 #endif
 
diff --git a/include/asm-arm/arch-at91/cpu.h b/include/asm-arm/arch-at91/cpu.h
index 080cbb401a87..7145166826a2 100644
--- a/include/asm-arm/arch-at91/cpu.h
+++ b/include/asm-arm/arch-at91/cpu.h
@@ -21,13 +21,13 @@
 #define ARCH_ID_AT91SAM9260	0x019803a0
 #define ARCH_ID_AT91SAM9261	0x019703a0
 #define ARCH_ID_AT91SAM9263	0x019607a0
+#define ARCH_ID_AT91SAM9RL64	0x019b03a0
+#define ARCH_ID_AT91CAP9	0x039A03A0
 
 #define ARCH_ID_AT91SAM9XE128	0x329973a0
 #define ARCH_ID_AT91SAM9XE256	0x329a93a0
 #define ARCH_ID_AT91SAM9XE512	0x329aa3a0
 
-#define ARCH_ID_AT91SAM9RL64	0x019b03a0
-
 #define ARCH_ID_AT91M40800	0x14080044
 #define ARCH_ID_AT91R40807	0x44080746
 #define ARCH_ID_AT91M40807	0x14080745
@@ -81,6 +81,11 @@ static inline unsigned long at91_arch_identify(void)
 #define cpu_is_at91sam9rl()	(0)
 #endif
 
+#ifdef CONFIG_ARCH_AT91CAP9
+#define cpu_is_at91cap9()	(at91_cpu_identify() == ARCH_ID_AT91CAP9)
+#else
+#define cpu_is_at91cap9()	(0)
+#endif
 
 /*
  * Since this is ARM, we will never run on any AVR32 CPU. But these
diff --git a/include/asm-arm/arch-at91/hardware.h b/include/asm-arm/arch-at91/hardware.h
index 8f1cdd38a969..2c826d8247a3 100644
--- a/include/asm-arm/arch-at91/hardware.h
+++ b/include/asm-arm/arch-at91/hardware.h
@@ -26,6 +26,8 @@
 #include <asm/arch/at91sam9263.h>
 #elif defined(CONFIG_ARCH_AT91SAM9RL)
 #include <asm/arch/at91sam9rl.h>
+#elif defined(CONFIG_ARCH_AT91CAP9)
+#include <asm/arch/at91cap9.h>
 #elif defined(CONFIG_ARCH_AT91X40)
 #include <asm/arch/at91x40.h>
 #else
diff --git a/include/asm-arm/arch-at91/timex.h b/include/asm-arm/arch-at91/timex.h
index a310698fb4da..f1933b0fa43f 100644
--- a/include/asm-arm/arch-at91/timex.h
+++ b/include/asm-arm/arch-at91/timex.h
@@ -42,6 +42,11 @@
 #define AT91SAM9_MASTER_CLOCK	100000000
 #define CLOCK_TICK_RATE		(AT91SAM9_MASTER_CLOCK/16)
 
+#elif defined(CONFIG_ARCH_AT91CAP9)
+
+#define AT91CAP9_MASTER_CLOCK	100000000
+#define CLOCK_TICK_RATE		(AT91CAP9_MASTER_CLOCK/16)
+
 #elif defined(CONFIG_ARCH_AT91X40)
 
 #define AT91X40_MASTER_CLOCK	40000000
diff --git a/include/asm-avr32/arch-at32ap/cpu.h b/include/asm-avr32/arch-at32ap/cpu.h
index 0dc20261c1ea..44d0bfa1f409 100644
--- a/include/asm-avr32/arch-at32ap/cpu.h
+++ b/include/asm-avr32/arch-at32ap/cpu.h
@@ -30,5 +30,6 @@
 #define cpu_is_at91sam9261()	(0)
 #define cpu_is_at91sam9263()	(0)
 #define cpu_is_at91sam9rl()	(0)
+#define cpu_is_at91cap9()	(0)
 
 #endif /* __ASM_ARCH_CPU_H */
-- 
cgit v1.2.3


From e50d64097b6e63278789ee3a4394d127bd6e4254 Mon Sep 17 00:00:00 2001
From: Assaf Hoffman <hoffman@marvell.com>
Date: Tue, 23 Oct 2007 15:14:41 -0400
Subject: [ARM] Marvell Feroceon CPU core support

The Feroceon is a family of independent ARMv5TE compliant CPU core
implementations, supporting a variable depth pipeline and out-of-order
execution.  The Feroceon is configurable with VFP support, and the
later models in the series are superscalar with up to two instructions
per clock cycle.

This patch adds the initial low-level cache/TLB handling for this core.

Signed-off-by: Assaf Hoffman <hoffman@marvell.com>
Reviewed-by: Tzachi Perelstein <tzachi@marvell.com>
Reviewed-by: Nicolas Pitre <nico@marvell.com>
Reviewed-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig             |   2 +-
 arch/arm/mm/Kconfig          |  16 +-
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-feroceon.S  | 479 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   8 +
 include/asm-arm/proc-fns.h   |   8 +
 6 files changed, 511 insertions(+), 3 deletions(-)
 create mode 100644 arch/arm/mm/proc-feroceon.S

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index a04f507e7f2c..fae3f164218f 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -951,7 +951,7 @@ config FPE_FASTFPE
 
 config VFP
 	bool "VFP-format floating point maths"
-	depends on CPU_V6 || CPU_ARM926T
+	depends on CPU_V6 || CPU_ARM926T || CPU_FEROCEON
 	help
 	  Say Y to include VFP support code in the kernel. This is needed
 	  if your hardware includes a VFP unit.
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 7868f4dc1d00..378fb744abe2 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -342,6 +342,18 @@ config CPU_XSC3
 	select CPU_TLB_V4WBI if MMU
 	select IO_36
 
+# Feroceon
+config CPU_FEROCEON
+	bool
+	depends on ARCH_ORION
+	default y
+	select CPU_32v5
+	select CPU_ABRT_EV5T
+	select CPU_CACHE_VIVT
+	select CPU_CP15_MMU
+	select CPU_COPY_V4WB if MMU
+	select CPU_TLB_V4WBI if MMU
+
 # ARMv6
 config CPU_V6
 	bool "Support ARM V6 processor"
@@ -538,7 +550,7 @@ comment "Processor Features"
 
 config ARM_THUMB
 	bool "Support Thumb user binaries"
-	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6 || CPU_V7
+	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6 || CPU_V7 || CPU_FEROCEON
 	default y
 	help
 	  Say Y if you want to include kernel support for running user space
@@ -600,7 +612,7 @@ config CPU_DCACHE_SIZE
 
 config CPU_DCACHE_WRITETHROUGH
 	bool "Force write through D-cache"
-	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020) && !CPU_DCACHE_DISABLE
+	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_FEROCEON) && !CPU_DCACHE_DISABLE
 	default y if CPU_ARM925T
 	help
 	  Say Y here to use the data cache in writethrough mode. Unless you
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 762702765fc3..44536a0b995a 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -68,6 +68,7 @@ obj-$(CONFIG_CPU_SA110)		+= proc-sa110.o
 obj-$(CONFIG_CPU_SA1100)	+= proc-sa1100.o
 obj-$(CONFIG_CPU_XSCALE)	+= proc-xscale.o
 obj-$(CONFIG_CPU_XSC3)		+= proc-xsc3.o
+obj-$(CONFIG_CPU_FEROCEON)	+= proc-feroceon.o
 obj-$(CONFIG_CPU_V6)		+= proc-v6.o
 obj-$(CONFIG_CPU_V7)		+= proc-v7.o
 
diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S
new file mode 100644
index 000000000000..8ad341af7906
--- /dev/null
+++ b/arch/arm/mm/proc-feroceon.S
@@ -0,0 +1,479 @@
+/*
+ *  linux/arch/arm/mm/proc-feroceon.S: MMU functions for Feroceon
+ *
+ *  Heavily based on proc-arm926.S
+ *  Maintainer: Assaf Hoffman <hoffman@marvell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/elf.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/ptrace.h>
+#include "proc-macros.S"
+
+/*
+ * This is the maximum size of an area which will be invalidated
+ * using the single invalidate entry instructions.  Anything larger
+ * than this, and we go for the whole cache.
+ *
+ * This value should be chosen such that we choose the cheapest
+ * alternative.
+ */
+#define CACHE_DLIMIT	16384
+
+/*
+ * the cache line size of the I and D cache
+ */
+#define CACHE_DLINESIZE	32
+
+	.text
+/*
+ * cpu_feroceon_proc_init()
+ */
+ENTRY(cpu_feroceon_proc_init)
+	mov	pc, lr
+
+/*
+ * cpu_feroceon_proc_fin()
+ */
+ENTRY(cpu_feroceon_proc_fin)
+	stmfd	sp!, {lr}
+	mov	ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+	msr	cpsr_c, ip
+	bl	feroceon_flush_kern_cache_all
+	mrc	p15, 0, r0, c1, c0, 0		@ ctrl register
+	bic	r0, r0, #0x1000			@ ...i............
+	bic	r0, r0, #0x000e			@ ............wca.
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	ldmfd	sp!, {pc}
+
+/*
+ * cpu_feroceon_reset(loc)
+ *
+ * Perform a soft reset of the system.  Put the CPU into the
+ * same state as it would be if it had been reset, and branch
+ * to what would be the reset vector.
+ *
+ * loc: location to jump to for soft reset
+ */
+	.align	5
+ENTRY(cpu_feroceon_reset)
+	mov	ip, #0
+	mcr	p15, 0, ip, c7, c7, 0		@ invalidate I,D caches
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+#ifdef CONFIG_MMU
+	mcr	p15, 0, ip, c8, c7, 0		@ invalidate I & D TLBs
+#endif
+	mrc	p15, 0, ip, c1, c0, 0		@ ctrl register
+	bic	ip, ip, #0x000f			@ ............wcam
+	bic	ip, ip, #0x1100			@ ...i...s........
+	mcr	p15, 0, ip, c1, c0, 0		@ ctrl register
+	mov	pc, r0
+
+/*
+ * cpu_feroceon_do_idle()
+ *
+ * Called with IRQs disabled
+ */
+	.align	10
+ENTRY(cpu_feroceon_do_idle)
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c10, 4		@ Drain write buffer
+	mcr	p15, 0, r0, c7, c0, 4		@ Wait for interrupt
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_all()
+ *
+ *	Clean and invalidate all cache entries in a particular
+ *	address space.
+ */
+ENTRY(feroceon_flush_user_cache_all)
+	/* FALLTHROUGH */
+
+/*
+ *	flush_kern_cache_all()
+ *
+ *	Clean and invalidate the entire cache.
+ */
+ENTRY(feroceon_flush_kern_cache_all)
+	mov	r2, #VM_EXEC
+	mov	ip, #0
+__flush_whole_cache:
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ invalidate D cache
+#else
+1:	mrc	p15, 0, r15, c7, c14, 3 	@ test,clean,invalidate
+	bne	1b
+#endif
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_range(start, end, flags)
+ *
+ *	Clean and invalidate a range of cache entries in the
+ *	specified address range.
+ *
+ *	- start	- start address (inclusive)
+ *	- end	- end address (exclusive)
+ *	- flags	- vm_flags describing address space
+ */
+ENTRY(feroceon_flush_user_cache_range)
+	mov	ip, #0
+	sub	r3, r1, r0			@ calculate total size
+	cmp	r3, #CACHE_DLIMIT
+	bgt	__flush_whole_cache
+1:	tst	r2, #VM_EXEC
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#else
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#endif
+	cmp	r0, r1
+	blo	1b
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	coherent_kern_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_coherent_kern_range)
+	/* FALLTHROUGH */
+
+/*
+ *	coherent_user_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_coherent_user_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	mcr	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_kern_dcache_page(void *page)
+ *
+ *	Ensure no D cache aliasing occurs, either with itself or
+ *	the I cache
+ *
+ *	- addr	- page aligned address
+ */
+ENTRY(feroceon_flush_kern_dcache_page)
+	add	r1, r0, #PAGE_SZ
+1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_inv_range(start, end)
+ *
+ *	Invalidate (discard) the specified virtual address range.
+ *	May not write back any entries.  If 'start' or 'end'
+ *	are not cache line aligned, those lines must be written
+ *	back.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ *
+ * (same as v4wb)
+ */
+ENTRY(feroceon_dma_inv_range)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	tst	r0, #CACHE_DLINESIZE - 1
+	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
+	tst	r1, #CACHE_DLINESIZE - 1
+	mcrne	p15, 0, r1, c7, c10, 1		@ clean D entry
+#endif
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_clean_range(start, end)
+ *
+ *	Clean the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ *
+ * (same as v4wb)
+ */
+ENTRY(feroceon_dma_clean_range)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_flush_range(start, end)
+ *
+ *	Clean and invalidate the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_dma_flush_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
+#else
+	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+#endif
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+ENTRY(feroceon_cache_fns)
+	.long	feroceon_flush_kern_cache_all
+	.long	feroceon_flush_user_cache_all
+	.long	feroceon_flush_user_cache_range
+	.long	feroceon_coherent_kern_range
+	.long	feroceon_coherent_user_range
+	.long	feroceon_flush_kern_dcache_page
+	.long	feroceon_dma_inv_range
+	.long	feroceon_dma_clean_range
+	.long	feroceon_dma_flush_range
+
+ENTRY(cpu_feroceon_dcache_clean_area)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	subs	r1, r1, #CACHE_DLINESIZE
+	bhi	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/* =============================== PageTable ============================== */
+
+/*
+ * cpu_feroceon_switch_mm(pgd)
+ *
+ * Set the translation base pointer to be as described by pgd.
+ *
+ * pgd: new page tables
+ */
+	.align	5
+ENTRY(cpu_feroceon_switch_mm)
+#ifdef CONFIG_MMU
+	mov	ip, #0
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ invalidate D cache
+#else
+@ && 'Clean & Invalidate whole DCache'
+1:	mrc	p15, 0, r15, c7, c14, 3 	@ test,clean,invalidate
+	bne	1b
+#endif
+	mcr	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mcr	p15, 0, r0, c2, c0, 0		@ load page table pointer
+	mcr	p15, 0, ip, c8, c7, 0		@ invalidate I & D TLBs
+#endif
+	mov	pc, lr
+
+/*
+ * cpu_feroceon_set_pte_ext(ptep, pte, ext)
+ *
+ * Set a PTE and flush it out
+ */
+	.align	5
+ENTRY(cpu_feroceon_set_pte_ext)
+#ifdef CONFIG_MMU
+	str	r1, [r0], #-2048		@ linux version
+
+	eor	r1, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
+
+	bic	r2, r1, #PTE_SMALL_AP_MASK
+	bic	r2, r2, #PTE_TYPE_MASK
+	orr	r2, r2, #PTE_TYPE_SMALL
+
+	tst	r1, #L_PTE_USER			@ User?
+	orrne	r2, r2, #PTE_SMALL_AP_URO_SRW
+
+	tst	r1, #L_PTE_WRITE | L_PTE_DIRTY	@ Write and Dirty?
+	orreq	r2, r2, #PTE_SMALL_AP_UNO_SRW
+
+	tst	r1, #L_PTE_PRESENT | L_PTE_YOUNG	@ Present and Young?
+	movne	r2, #0
+
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	eor	r3, r2, #0x0a			@ C & small page?
+	tst	r3, #0x0b
+	biceq	r2, r2, #4
+#endif
+	str	r2, [r0]			@ hardware version
+	mov	r0, r0
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+#endif
+	mov	pc, lr
+
+	__INIT
+
+	.type	__feroceon_setup, #function
+__feroceon_setup:
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c7		@ invalidate I,D caches on v4
+	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer on v4
+#ifdef CONFIG_MMU
+	mcr	p15, 0, r0, c8, c7		@ invalidate I,D TLBs on v4
+#endif
+
+
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r0, #4				@ disable write-back on caches explicitly
+	mcr	p15, 7, r0, c15, c0, 0
+#endif
+
+	adr	r5, feroceon_crval
+	ldmia	r5, {r5, r6}
+	mrc	p15, 0, r0, c1, c0		@ get control register v4
+	bic	r0, r0, r5
+	orr	r0, r0, r6
+#ifdef CONFIG_CPU_CACHE_ROUND_ROBIN
+	orr	r0, r0, #0x4000			@ .1.. .... .... ....
+#endif
+	mov	pc, lr
+	.size	__feroceon_setup, . - __feroceon_setup
+
+	/*
+	 *  R
+	 * .RVI ZFRS BLDP WCAM
+	 * .011 0001 ..11 0101
+	 *
+	 */
+	.type	feroceon_crval, #object
+feroceon_crval:
+	crval	clear=0x00007f3f, mmuset=0x00003135, ucset=0x00001134
+
+	__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+	.type	feroceon_processor_functions, #object
+feroceon_processor_functions:
+	.word	v5t_early_abort
+	.word	cpu_feroceon_proc_init
+	.word	cpu_feroceon_proc_fin
+	.word	cpu_feroceon_reset
+	.word	cpu_feroceon_do_idle
+	.word	cpu_feroceon_dcache_clean_area
+	.word	cpu_feroceon_switch_mm
+	.word	cpu_feroceon_set_pte_ext
+	.size	feroceon_processor_functions, . - feroceon_processor_functions
+
+	.section ".rodata"
+
+	.type	cpu_arch_name, #object
+cpu_arch_name:
+	.asciz	"armv5te"
+	.size	cpu_arch_name, . - cpu_arch_name
+
+	.type	cpu_elf_name, #object
+cpu_elf_name:
+	.asciz	"v5"
+	.size	cpu_elf_name, . - cpu_elf_name
+
+	.type	cpu_feroceon_name, #object
+cpu_feroceon_name:
+	.asciz	"Feroceon"
+	.size	cpu_feroceon_name, . - cpu_feroceon_name
+
+	.align
+
+	.section ".proc.info.init", #alloc, #execinstr
+
+	.type	__feroceon_proc_info,#object
+__feroceon_proc_info:
+	.long	0x56055310
+	.long	0xfffffff0
+	.long   PMD_TYPE_SECT | \
+		PMD_SECT_BUFFERABLE | \
+		PMD_SECT_CACHEABLE | \
+		PMD_BIT4 | \
+		PMD_SECT_AP_WRITE | \
+		PMD_SECT_AP_READ
+	.long   PMD_TYPE_SECT | \
+		PMD_BIT4 | \
+		PMD_SECT_AP_WRITE | \
+		PMD_SECT_AP_READ
+	b	__feroceon_setup
+	.long	cpu_arch_name
+	.long	cpu_elf_name
+	.long	HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
+	.long	cpu_feroceon_name
+	.long	feroceon_processor_functions
+	.long	v4wbi_tlb_fns
+	.long	v4wb_user_fns
+	.long	feroceon_cache_fns
+	.size	__feroceon_proc_info, . - __feroceon_proc_info
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index 6c1c968b2987..759a97b56eed 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -94,6 +94,14 @@
 # endif
 #endif
 
+#if defined(CONFIG_CPU_FEROCEON)
+# ifdef _CACHE
+#  define MULTI_CACHE 1
+# else
+#  define _CACHE feroceon
+# endif
+#endif
+
 #if defined(CONFIG_CPU_V6)
 //# ifdef _CACHE
 #  define MULTI_CACHE 1
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 5599d4e5e708..a4ce457199d3 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -185,6 +185,14 @@
 #   define CPU_NAME cpu_xsc3
 #  endif
 # endif
+# ifdef CONFIG_CPU_FEROCEON
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_feroceon
+#  endif
+# endif
 # ifdef CONFIG_CPU_V6
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From 585cf17561d3174a745bec49c422c1a621c95fc4 Mon Sep 17 00:00:00 2001
From: Tzachi Perelstein <tzachi@marvell.com>
Date: Tue, 23 Oct 2007 15:14:41 -0400
Subject: [ARM] basic support for the Marvell Orion SoC family

The Marvell Orion is a family of ARM SoCs with a DDR/DDR2 memory
controller, 10/100/1000 ethernet MAC, and USB 2.0 interfaces,
and, depending on the specific model, PCI-E interface, PCI-X
interface, SATA controllers, crypto unit, SPI interface, SDIO
interface, device bus, NAND controller, DMA engine and/or XOR
engine.

This contains the basic structure and architecture register definitions.

Signed-off-by: Tzachi Perelstein <tzachi@marvell.com>
Reviewed-by: Nicolas Pitre <nico@marvell.com>
Reviewed-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                         |   8 ++
 arch/arm/Makefile                        |   1 +
 arch/arm/mach-orion/Kconfig              |   7 ++
 arch/arm/mach-orion/Makefile             |   1 +
 arch/arm/mach-orion/Makefile.boot        |   3 +
 arch/arm/mach-orion/common.c             |  53 ++++++++++++
 arch/arm/mach-orion/common.h             |   9 ++
 include/asm-arm/arch-orion/debug-macro.S |  17 ++++
 include/asm-arm/arch-orion/dma.h         |   1 +
 include/asm-arm/arch-orion/entry-macro.S |  31 +++++++
 include/asm-arm/arch-orion/hardware.h    |  24 ++++++
 include/asm-arm/arch-orion/io.h          |  27 ++++++
 include/asm-arm/arch-orion/irqs.h        |  61 ++++++++++++++
 include/asm-arm/arch-orion/memory.h      |  15 ++++
 include/asm-arm/arch-orion/orion.h       | 140 +++++++++++++++++++++++++++++++
 include/asm-arm/arch-orion/system.h      |  31 +++++++
 include/asm-arm/arch-orion/timex.h       |  12 +++
 include/asm-arm/arch-orion/uncompress.h  |  44 ++++++++++
 include/asm-arm/arch-orion/vmalloc.h     |   5 ++
 19 files changed, 490 insertions(+)
 create mode 100644 arch/arm/mach-orion/Kconfig
 create mode 100644 arch/arm/mach-orion/Makefile
 create mode 100644 arch/arm/mach-orion/Makefile.boot
 create mode 100644 arch/arm/mach-orion/common.c
 create mode 100644 arch/arm/mach-orion/common.h
 create mode 100644 include/asm-arm/arch-orion/debug-macro.S
 create mode 100644 include/asm-arm/arch-orion/dma.h
 create mode 100644 include/asm-arm/arch-orion/entry-macro.S
 create mode 100644 include/asm-arm/arch-orion/hardware.h
 create mode 100644 include/asm-arm/arch-orion/io.h
 create mode 100644 include/asm-arm/arch-orion/irqs.h
 create mode 100644 include/asm-arm/arch-orion/memory.h
 create mode 100644 include/asm-arm/arch-orion/orion.h
 create mode 100644 include/asm-arm/arch-orion/system.h
 create mode 100644 include/asm-arm/arch-orion/timex.h
 create mode 100644 include/asm-arm/arch-orion/uncompress.h
 create mode 100644 include/asm-arm/arch-orion/vmalloc.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index fae3f164218f..036239150ff8 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -333,6 +333,12 @@ config ARCH_MXC
 	help
 	  Support for Freescale MXC/iMX-based family of processors
 
+config ARCH_ORION
+	bool "Marvell Orion"
+	depends on MMU
+	help
+	  Support for Marvell Orion System on Chip family.
+
 config ARCH_PNX4008
 	bool "Philips Nexperia PNX4008 Mobile"
 	help
@@ -441,6 +447,8 @@ source "arch/arm/mach-omap1/Kconfig"
 
 source "arch/arm/mach-omap2/Kconfig"
 
+source "arch/arm/mach-orion/Kconfig"
+
 source "arch/arm/plat-s3c24xx/Kconfig"
 source "arch/arm/plat-s3c/Kconfig"
 
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 35e56c99ad1d..2daa278149e0 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -139,6 +139,7 @@ endif
  machine-$(CONFIG_ARCH_KS8695)     := ks8695
   incdir-$(CONFIG_ARCH_MXC)	   := mxc
  machine-$(CONFIG_ARCH_MX3)	   := mx3
+ machine-$(CONFIG_ARCH_ORION)	   := orion
 
 ifeq ($(CONFIG_ARCH_EBSA110),y)
 # This is what happens if you forget the IOCS16 line.
diff --git a/arch/arm/mach-orion/Kconfig b/arch/arm/mach-orion/Kconfig
new file mode 100644
index 000000000000..c2af02125ce9
--- /dev/null
+++ b/arch/arm/mach-orion/Kconfig
@@ -0,0 +1,7 @@
+if ARCH_ORION
+
+menu "Orion Implementations"
+
+endmenu
+
+endif
diff --git a/arch/arm/mach-orion/Makefile b/arch/arm/mach-orion/Makefile
new file mode 100644
index 000000000000..69eb30187738
--- /dev/null
+++ b/arch/arm/mach-orion/Makefile
@@ -0,0 +1 @@
+obj-y  += common.o
diff --git a/arch/arm/mach-orion/Makefile.boot b/arch/arm/mach-orion/Makefile.boot
new file mode 100644
index 000000000000..67039c3e0c48
--- /dev/null
+++ b/arch/arm/mach-orion/Makefile.boot
@@ -0,0 +1,3 @@
+   zreladdr-y	:= 0x00008000
+params_phys-y	:= 0x00000100
+initrd_phys-y	:= 0x00800000
diff --git a/arch/arm/mach-orion/common.c b/arch/arm/mach-orion/common.c
new file mode 100644
index 000000000000..e04f3f07cac0
--- /dev/null
+++ b/arch/arm/mach-orion/common.c
@@ -0,0 +1,53 @@
+/*
+ * arch/arm/mach-orion/common.c
+ *
+ * Core functions for Marvell Orion System On Chip
+ *
+ * Maintainer: Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/arch/orion.h>
+#include "common.h"
+
+/*****************************************************************************
+ * I/O Address Mapping
+ ****************************************************************************/
+static struct map_desc orion_io_desc[] __initdata = {
+	{
+		.virtual	= ORION_REGS_BASE,
+		.pfn		= __phys_to_pfn(ORION_REGS_BASE),
+		.length		= ORION_REGS_SIZE,
+		.type		= MT_DEVICE
+	},
+	{
+		.virtual	= ORION_PCIE_IO_BASE,
+		.pfn		= __phys_to_pfn(ORION_PCIE_IO_BASE),
+		.length		= ORION_PCIE_IO_SIZE,
+		.type		= MT_DEVICE
+	},
+	{
+		.virtual	= ORION_PCI_IO_BASE,
+		.pfn		= __phys_to_pfn(ORION_PCI_IO_BASE),
+		.length		= ORION_PCI_IO_SIZE,
+		.type		= MT_DEVICE
+	},
+	{
+		.virtual	= ORION_PCIE_WA_BASE,
+		.pfn		= __phys_to_pfn(ORION_PCIE_WA_BASE),
+		.length		= ORION_PCIE_WA_SIZE,
+		.type		= MT_DEVICE
+	},
+};
+
+void __init orion_map_io(void)
+{
+	iotable_init(orion_io_desc, ARRAY_SIZE(orion_io_desc));
+}
diff --git a/arch/arm/mach-orion/common.h b/arch/arm/mach-orion/common.h
new file mode 100644
index 000000000000..224b44618a72
--- /dev/null
+++ b/arch/arm/mach-orion/common.h
@@ -0,0 +1,9 @@
+#ifndef __ARCH_ORION_COMMON_H__
+#define __ARCH_ORION_COMMON_H__
+
+/*
+ * Basic Orion init functions used early by machine-setup.
+ */
+void __init orion_map_io(void);
+
+#endif /* __ARCH_ORION_COMMON_H__ */
diff --git a/include/asm-arm/arch-orion/debug-macro.S b/include/asm-arm/arch-orion/debug-macro.S
new file mode 100644
index 000000000000..e2a80641f214
--- /dev/null
+++ b/include/asm-arm/arch-orion/debug-macro.S
@@ -0,0 +1,17 @@
+/*
+ * linux/include/asm-arm/arch-orion/debug-macro.S
+ *
+ * Debugging macro include header
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+	.macro  addruart,rx
+	mov   \rx, #0xf1000000
+	orr   \rx, \rx, #0x00012000
+	.endm
+
+#define UART_SHIFT	2
+#include <asm/hardware/debug-8250.S>
diff --git a/include/asm-arm/arch-orion/dma.h b/include/asm-arm/arch-orion/dma.h
new file mode 100644
index 000000000000..40a8c178f10d
--- /dev/null
+++ b/include/asm-arm/arch-orion/dma.h
@@ -0,0 +1 @@
+/* empty */
diff --git a/include/asm-arm/arch-orion/entry-macro.S b/include/asm-arm/arch-orion/entry-macro.S
new file mode 100644
index 000000000000..b76075a7e44b
--- /dev/null
+++ b/include/asm-arm/arch-orion/entry-macro.S
@@ -0,0 +1,31 @@
+/*
+ * include/asm-arm/arch-orion/entry-macro.S
+ *
+ * Low-level IRQ helper macros for Orion platforms
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <asm/arch/orion.h>
+
+	.macro  disable_fiq
+	.endm
+
+	.macro  arch_ret_to_user, tmp1, tmp2
+	.endm
+
+	.macro  get_irqnr_preamble, base, tmp
+	ldr	\base, =MAIN_IRQ_CAUSE
+	.endm
+
+	.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
+	ldr	\irqstat, [\base, #0]		@ main cause
+	ldr	\tmp, [\base, #(MAIN_IRQ_MASK - MAIN_IRQ_CAUSE)] @ main mask
+	mov	\irqnr, #0			@ default irqnr
+	@ find cause bits that are unmasked
+	ands	\irqstat, \irqstat, \tmp	@ clear Z flag if any
+	clzne	\irqnr,	\irqstat		@ calc irqnr
+	rsbne	\irqnr, \irqnr, #31
+	.endm
diff --git a/include/asm-arm/arch-orion/hardware.h b/include/asm-arm/arch-orion/hardware.h
new file mode 100644
index 000000000000..8a12d213fbdc
--- /dev/null
+++ b/include/asm-arm/arch-orion/hardware.h
@@ -0,0 +1,24 @@
+/*
+ * include/asm-arm/arch-orion/hardware.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#ifndef __ASM_ARCH_HARDWARE_H__
+#define __ASM_ARCH_HARDWARE_H__
+
+#include "orion.h"
+
+#define PCI_MEMORY_VADDR        ORION_PCI_SYS_MEM_BASE
+#define PCI_IO_VADDR            ORION_PCI_SYS_IO_BASE
+
+#define pcibios_assign_all_busses()  1
+
+#define PCIBIOS_MIN_IO  0x1000
+#define PCIBIOS_MIN_MEM 0x01000000
+#define PCIMEM_BASE     PCI_MEMORY_VADDR /* mem base for VGA */
+
+#endif  /* _ASM_ARCH_HARDWARE_H */
diff --git a/include/asm-arm/arch-orion/io.h b/include/asm-arm/arch-orion/io.h
new file mode 100644
index 000000000000..e0b8c39b9167
--- /dev/null
+++ b/include/asm-arm/arch-orion/io.h
@@ -0,0 +1,27 @@
+/*
+ * include/asm-arm/arch-orion/io.h
+ *
+ * Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __ASM_ARM_ARCH_IO_H
+#define __ASM_ARM_ARCH_IO_H
+
+#include "orion.h"
+
+#define IO_SPACE_LIMIT		0xffffffff
+#define IO_SPACE_REMAP		ORION_PCI_SYS_IO_BASE
+
+static inline void __iomem *__io(unsigned long addr)
+{
+	return (void __iomem *)addr;
+}
+
+#define __io(a)			__io(a)
+#define __mem_pci(a)		(a)
+
+#endif
diff --git a/include/asm-arm/arch-orion/irqs.h b/include/asm-arm/arch-orion/irqs.h
new file mode 100644
index 000000000000..eea65ca6076a
--- /dev/null
+++ b/include/asm-arm/arch-orion/irqs.h
@@ -0,0 +1,61 @@
+/*
+ * include/asm-arm/arch-orion/irqs.h
+ *
+ * IRQ definitions for Orion SoC
+ *
+ *  Maintainer: Tzachi Perelstein <tzachi@marvell.com>
+ *
+ *  This file is licensed under the terms of the GNU General Public
+ *  License version 2. This program is licensed "as is" without any
+ *  warranty of any kind, whether express or implied.
+ */
+
+#ifndef __ASM_ARCH_IRQS_H__
+#define __ASM_ARCH_IRQS_H__
+
+#include "orion.h"	/* need GPIO_MAX */
+
+/*
+ * Orion Main Interrupt Controller
+ */
+#define IRQ_ORION_BRIDGE	0
+#define IRQ_ORION_DOORBELL_H2C	1
+#define IRQ_ORION_DOORBELL_C2H	2
+#define IRQ_ORION_UART0		3
+#define IRQ_ORION_UART1		4
+#define IRQ_ORION_I2C		5
+#define IRQ_ORION_GPIO_0_7	6
+#define IRQ_ORION_GPIO_8_15	7
+#define IRQ_ORION_GPIO_16_23	8
+#define IRQ_ORION_GPIO_24_31	9
+#define IRQ_ORION_PCIE0_ERR	10
+#define IRQ_ORION_PCIE0_INT	11
+#define IRQ_ORION_USB1_CTRL	12
+#define IRQ_ORION_DEV_BUS_ERR	14
+#define IRQ_ORION_PCI_ERR	15
+#define IRQ_ORION_USB_BR_ERR	16
+#define IRQ_ORION_USB0_CTRL	17
+#define IRQ_ORION_ETH_RX	18
+#define IRQ_ORION_ETH_TX	19
+#define IRQ_ORION_ETH_MISC	20
+#define IRQ_ORION_ETH_SUM	21
+#define IRQ_ORION_ETH_ERR	22
+#define IRQ_ORION_IDMA_ERR	23
+#define IRQ_ORION_IDMA_0	24
+#define IRQ_ORION_IDMA_1	25
+#define IRQ_ORION_IDMA_2	26
+#define IRQ_ORION_IDMA_3	27
+#define IRQ_ORION_CESA		28
+#define IRQ_ORION_SATA		29
+#define IRQ_ORION_XOR0		30
+#define IRQ_ORION_XOR1		31
+
+/*
+ * Orion General Purpose Pins
+ */
+#define IRQ_ORION_GPIO_START	32
+#define NR_GPIO_IRQS		GPIO_MAX
+
+#define NR_IRQS			(IRQ_ORION_GPIO_START + NR_GPIO_IRQS)
+
+#endif /* __ASM_ARCH_IRQS_H__ */
diff --git a/include/asm-arm/arch-orion/memory.h b/include/asm-arm/arch-orion/memory.h
new file mode 100644
index 000000000000..d954dba87ced
--- /dev/null
+++ b/include/asm-arm/arch-orion/memory.h
@@ -0,0 +1,15 @@
+/*
+ * include/asm-arm/arch-orion/memory.h
+ *
+ * Marvell Orion memory definitions
+ */
+
+#ifndef __ASM_ARCH_MMU_H
+#define __ASM_ARCH_MMU_H
+
+#define PHYS_OFFSET	UL(0x00000000)
+
+#define __virt_to_bus(x)	__virt_to_phys(x)
+#define __bus_to_virt(x)	__phys_to_virt(x)
+
+#endif
diff --git a/include/asm-arm/arch-orion/orion.h b/include/asm-arm/arch-orion/orion.h
new file mode 100644
index 000000000000..061c03c5a17e
--- /dev/null
+++ b/include/asm-arm/arch-orion/orion.h
@@ -0,0 +1,140 @@
+/*
+ * include/asm-arm/arch-orion/orion.h
+ *
+ * Generic definitions of Orion SoC flavors:
+ *  Orion-1, Orion-NAS, Orion-VoIP, and Orion-2.
+ *
+ * Maintainer: Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __ASM_ARCH_ORION_H__
+#define __ASM_ARCH_ORION_H__
+
+/*******************************************************************************
+ * Orion Address Map
+ * Use the same mapping (1:1 virtual:physical) of internal registers and
+ * PCI system (PCI+PCIE) for all machines.
+ * Each machine defines the rest of its mapping (e.g. device bus flashes)
+ ******************************************************************************/
+#define ORION_REGS_BASE		0xf1000000
+#define ORION_REGS_SIZE		SZ_1M
+
+#define ORION_PCI_SYS_MEM_BASE	0xe0000000
+#define ORION_PCIE_MEM_BASE	ORION_PCI_SYS_MEM_BASE
+#define ORION_PCIE_MEM_SIZE	SZ_128M
+#define ORION_PCI_MEM_BASE	(ORION_PCIE_MEM_BASE + ORION_PCIE_MEM_SIZE)
+#define ORION_PCI_MEM_SIZE	SZ_128M
+
+#define ORION_PCI_SYS_IO_BASE	0xf2000000
+#define ORION_PCIE_IO_BASE	ORION_PCI_SYS_IO_BASE
+#define ORION_PCIE_IO_SIZE	SZ_1M
+#define ORION_PCIE_IO_REMAP	(ORION_PCIE_IO_BASE - ORION_PCI_SYS_IO_BASE)
+#define ORION_PCI_IO_BASE	(ORION_PCIE_IO_BASE + ORION_PCIE_IO_SIZE)
+#define ORION_PCI_IO_SIZE	SZ_1M
+#define ORION_PCI_IO_REMAP	(ORION_PCI_IO_BASE - ORION_PCI_SYS_IO_BASE)
+/* Relevant only for Orion-NAS */
+#define ORION_PCIE_WA_BASE	0xf0000000
+#define ORION_PCIE_WA_SIZE	SZ_16M
+
+/*******************************************************************************
+ * Supported Devices & Revisions
+ ******************************************************************************/
+/* Orion-NAS (88F5182) */
+#define MV88F5182_DEV_ID	0x5182
+#define MV88F5182_REV_A2	2
+/* Orion-2 (88F5281) */
+#define MV88F5281_DEV_ID	0x5281
+#define MV88F5281_REV_D1	5
+#define MV88F5281_REV_D2	6
+
+/*******************************************************************************
+ * Orion Registers Map
+ ******************************************************************************/
+#define ORION_DDR_REG_BASE	(ORION_REGS_BASE | 0x00000)
+#define ORION_DEV_BUS_REG_BASE	(ORION_REGS_BASE | 0x10000)
+#define ORION_BRIDGE_REG_BASE	(ORION_REGS_BASE | 0x20000)
+#define ORION_PCI_REG_BASE	(ORION_REGS_BASE | 0x30000)
+#define ORION_PCIE_REG_BASE	(ORION_REGS_BASE | 0x40000)
+#define ORION_USB0_REG_BASE	(ORION_REGS_BASE | 0x50000)
+#define ORION_ETH_REG_BASE	(ORION_REGS_BASE | 0x70000)
+#define ORION_SATA_REG_BASE	(ORION_REGS_BASE | 0x80000)
+#define ORION_USB1_REG_BASE	(ORION_REGS_BASE | 0xa0000)
+
+#define ORION_DDR_REG(x)	(ORION_DDR_REG_BASE | (x))
+#define ORION_DEV_BUS_REG(x)	(ORION_DEV_BUS_REG_BASE | (x))
+#define ORION_BRIDGE_REG(x)	(ORION_BRIDGE_REG_BASE | (x))
+#define ORION_PCI_REG(x)	(ORION_PCI_REG_BASE | (x))
+#define ORION_PCIE_REG(x)	(ORION_PCIE_REG_BASE | (x))
+#define ORION_USB0_REG(x)	(ORION_USB0_REG_BASE | (x))
+#define ORION_USB1_REG(x)	(ORION_USB1_REG_BASE | (x))
+#define ORION_ETH_REG(x)	(ORION_ETH_REG_BASE | (x))
+#define ORION_SATA_REG(x)	(ORION_SATA_REG_BASE | (x))
+
+/*******************************************************************************
+ * Device Bus Registers
+ ******************************************************************************/
+#define MPP_0_7_CTRL		ORION_DEV_BUS_REG(0x000)
+#define MPP_8_15_CTRL		ORION_DEV_BUS_REG(0x004)
+#define MPP_16_19_CTRL		ORION_DEV_BUS_REG(0x050)
+#define MPP_DEV_CTRL		ORION_DEV_BUS_REG(0x008)
+#define MPP_RESET_SAMPLE	ORION_DEV_BUS_REG(0x010)
+#define GPIO_OUT		ORION_DEV_BUS_REG(0x100)
+#define GPIO_IO_CONF		ORION_DEV_BUS_REG(0x104)
+#define GPIO_BLINK_EN		ORION_DEV_BUS_REG(0x108)
+#define GPIO_IN_POL		ORION_DEV_BUS_REG(0x10c)
+#define GPIO_DATA_IN		ORION_DEV_BUS_REG(0x110)
+#define GPIO_EDGE_CAUSE		ORION_DEV_BUS_REG(0x114)
+#define GPIO_EDGE_MASK		ORION_DEV_BUS_REG(0x118)
+#define GPIO_LEVEL_MASK		ORION_DEV_BUS_REG(0x11c)
+#define DEV_BANK_0_PARAM	ORION_DEV_BUS_REG(0x45c)
+#define DEV_BANK_1_PARAM	ORION_DEV_BUS_REG(0x460)
+#define DEV_BANK_2_PARAM	ORION_DEV_BUS_REG(0x464)
+#define DEV_BANK_BOOT_PARAM	ORION_DEV_BUS_REG(0x46c)
+#define DEV_BUS_CTRL		ORION_DEV_BUS_REG(0x4c0)
+#define DEV_BUS_INT_CAUSE	ORION_DEV_BUS_REG(0x4d0)
+#define DEV_BUS_INT_MASK	ORION_DEV_BUS_REG(0x4d4)
+#define I2C_BASE		ORION_DEV_BUS_REG(0x1000)
+#define UART0_BASE		ORION_DEV_BUS_REG(0x2000)
+#define UART1_BASE		ORION_DEV_BUS_REG(0x2100)
+#define GPIO_MAX		32
+
+/***************************************************************************
+ * Orion CPU Bridge Registers
+ **************************************************************************/
+#define CPU_CONF		ORION_BRIDGE_REG(0x100)
+#define CPU_CTRL		ORION_BRIDGE_REG(0x104)
+#define CPU_RESET_MASK		ORION_BRIDGE_REG(0x108)
+#define CPU_SOFT_RESET		ORION_BRIDGE_REG(0x10c)
+#define POWER_MNG_CTRL_REG	ORION_BRIDGE_REG(0x11C)
+#define BRIDGE_CAUSE		ORION_BRIDGE_REG(0x110)
+#define BRIDGE_MASK		ORION_BRIDGE_REG(0x114)
+#define MAIN_IRQ_CAUSE		ORION_BRIDGE_REG(0x200)
+#define MAIN_IRQ_MASK		ORION_BRIDGE_REG(0x204)
+#define TIMER_CTRL		ORION_BRIDGE_REG(0x300)
+#define TIMER_VAL(x)		ORION_BRIDGE_REG(0x314 + ((x) * 8))
+#define TIMER_VAL_RELOAD(x)	ORION_BRIDGE_REG(0x310 + ((x) * 8))
+
+#ifndef __ASSEMBLY__
+
+/*******************************************************************************
+ * Helpers to access Orion registers
+ ******************************************************************************/
+#include <asm/types.h>
+#include <asm/io.h>
+
+#define orion_read(r)		__raw_readl(r)
+#define orion_write(r, val)	__raw_writel(val, r)
+
+/*
+ * These are not preempt safe. Locks, if needed, must be taken care by caller.
+ */
+#define orion_setbits(r, mask)	orion_write((r), orion_read(r) | (mask))
+#define orion_clrbits(r, mask)	orion_write((r), orion_read(r) & ~(mask))
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_ARCH_ORION_H__ */
diff --git a/include/asm-arm/arch-orion/system.h b/include/asm-arm/arch-orion/system.h
new file mode 100644
index 000000000000..17704c68f90e
--- /dev/null
+++ b/include/asm-arm/arch-orion/system.h
@@ -0,0 +1,31 @@
+/*
+ * include/asm-arm/arch-orion/system.h
+ *
+ * Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __ASM_ARCH_SYSTEM_H
+#define __ASM_ARCH_SYSTEM_H
+
+#include <asm/arch/hardware.h>
+#include <asm/arch/orion.h>
+
+static inline void arch_idle(void)
+{
+	cpu_do_idle();
+}
+
+static inline void arch_reset(char mode)
+{
+	/*
+	 * Enable and issue soft reset
+	 */
+	orion_setbits(CPU_RESET_MASK, (1 << 2));
+	orion_setbits(CPU_SOFT_RESET, 1);
+}
+
+#endif
diff --git a/include/asm-arm/arch-orion/timex.h b/include/asm-arm/arch-orion/timex.h
new file mode 100644
index 000000000000..26c2c91eecf0
--- /dev/null
+++ b/include/asm-arm/arch-orion/timex.h
@@ -0,0 +1,12 @@
+/*
+ * include/asm-arm/arch-orion/timex.h
+ *
+ * Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#define ORION_TCLK		166666667
+#define CLOCK_TICK_RATE		ORION_TCLK
diff --git a/include/asm-arm/arch-orion/uncompress.h b/include/asm-arm/arch-orion/uncompress.h
new file mode 100644
index 000000000000..a1a222fb438c
--- /dev/null
+++ b/include/asm-arm/arch-orion/uncompress.h
@@ -0,0 +1,44 @@
+/*
+ * include/asm-arm/arch-orion/uncompress.h
+ *
+ * Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <asm/arch/orion.h>
+
+#define MV_UART_LSR 	((volatile unsigned char *)(UART0_BASE + 0x14))
+#define MV_UART_THR	((volatile unsigned char *)(UART0_BASE + 0x0))
+
+#define LSR_THRE	0x20
+
+static void putc(const char c)
+{
+	int j = 0x1000;
+	while (--j && !(*MV_UART_LSR & LSR_THRE))
+		barrier();
+	*MV_UART_THR = c;
+}
+
+static void flush(void)
+{
+}
+
+static void orion_early_putstr(const char *ptr)
+{
+	char c;
+	while ((c = *ptr++) != '\0') {
+		if (c == '\n')
+			putc('\r');
+		putc(c);
+	}
+}
+
+/*
+ * nothing to do
+ */
+#define arch_decomp_setup()
+#define arch_decomp_wdog()
diff --git a/include/asm-arm/arch-orion/vmalloc.h b/include/asm-arm/arch-orion/vmalloc.h
new file mode 100644
index 000000000000..23e2a102fe0c
--- /dev/null
+++ b/include/asm-arm/arch-orion/vmalloc.h
@@ -0,0 +1,5 @@
+/*
+ * include/asm-arm/arch-orion/vmalloc.h
+ */
+
+#define VMALLOC_END       0xf0000000
-- 
cgit v1.2.3


From 01af72e4e36fba66cd7cfc2a628efee866c346d1 Mon Sep 17 00:00:00 2001
From: Tzachi Perelstein <tzachi@marvell.com>
Date: Tue, 23 Oct 2007 15:14:42 -0400
Subject: [ARM] Orion: GPIO support

Signed-off-by: Tzachi Perelstein <tzachi@marvell.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
Acked-by: David Brownell <david-b@pacbell.net>
---
 arch/arm/Kconfig                  |   1 +
 arch/arm/mach-orion/Makefile      |   2 +-
 arch/arm/mach-orion/common.h      |   8 ++
 arch/arm/mach-orion/gpio.c        | 206 ++++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-orion/gpio.h |  27 +++++
 5 files changed, 243 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-orion/gpio.c
 create mode 100644 include/asm-arm/arch-orion/gpio.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 37aa3c606c0c..71e905ebcc05 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -337,6 +337,7 @@ config ARCH_ORION
 	bool "Marvell Orion"
 	depends on MMU
 	select PCI
+	select GENERIC_GPIO
 	help
 	  Support for Marvell Orion System on Chip family.
 
diff --git a/arch/arm/mach-orion/Makefile b/arch/arm/mach-orion/Makefile
index 1d4b5570067e..d6a6c86faafc 100644
--- a/arch/arm/mach-orion/Makefile
+++ b/arch/arm/mach-orion/Makefile
@@ -1 +1 @@
-obj-y  += common.o addr-map.o pci.o
+obj-y  += common.o addr-map.o pci.o gpio.o
diff --git a/arch/arm/mach-orion/common.h b/arch/arm/mach-orion/common.h
index 056043464ca1..a6a7af189b0e 100644
--- a/arch/arm/mach-orion/common.h
+++ b/arch/arm/mach-orion/common.h
@@ -53,4 +53,12 @@ struct pci_bus *orion_pci_sys_scan_bus(int nr, struct pci_sys_data *sys);
 int orion_pci_hw_rd_conf(u32 bus, u32 dev, u32 func, u32 where, u32 size, u32 *val);
 int orion_pci_hw_wr_conf(u32 bus, u32 dev, u32 func, u32 where, u32 size, u32 val);
 
+/*
+ * Valid GPIO pins according to MPP setup, used by machine-setup.
+ * (/mach-orion/gpio.c).
+ */
+
+void __init orion_gpio_set_valid_pins(u32 pins);
+void gpio_display(void);	/* debug */
+
 #endif /* __ARCH_ORION_COMMON_H__ */
diff --git a/arch/arm/mach-orion/gpio.c b/arch/arm/mach-orion/gpio.c
new file mode 100644
index 000000000000..af8553ccd230
--- /dev/null
+++ b/arch/arm/mach-orion/gpio.c
@@ -0,0 +1,206 @@
+/*
+ * arch/arm/mach-orion/gpio.c
+ *
+ * GPIO functions for Marvell Orion System On Chip
+ *
+ * Maintainer: Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/bitops.h>
+#include <asm/gpio.h>
+#include <asm/arch/orion.h>
+#include "common.h"
+
+static DEFINE_SPINLOCK(gpio_lock);
+static unsigned long gpio_valid[BITS_TO_LONGS(GPIO_MAX)];
+static const char *gpio_label[GPIO_MAX];  /* non null for allocated GPIOs */
+
+void __init orion_gpio_set_valid_pins(u32 pins)
+{
+	gpio_valid[0] = pins;
+}
+
+/*
+ * GENERIC_GPIO primitives
+ */
+int gpio_direction_input(unsigned pin)
+{
+	unsigned long flags;
+
+	if (pin >= GPIO_MAX || !test_bit(pin, gpio_valid)) {
+		pr_debug("%s: invalid GPIO %d\n", __FUNCTION__, pin);
+		return -EINVAL;
+	}
+
+	spin_lock_irqsave(&gpio_lock, flags);
+
+	/*
+	 * Some callers might have not used the gpio_request(),
+	 * so flag this pin as requested now.
+	 */
+	if (!gpio_label[pin])
+		gpio_label[pin] = "?";
+
+	orion_setbits(GPIO_IO_CONF, 1 << pin);
+
+	spin_unlock_irqrestore(&gpio_lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL(gpio_direction_input);
+
+int gpio_direction_output(unsigned pin, int value)
+{
+	unsigned long flags;
+	int mask;
+
+	if (pin >= GPIO_MAX || !test_bit(pin, gpio_valid)) {
+		pr_debug("%s: invalid GPIO %d\n", __FUNCTION__, pin);
+		return -EINVAL;
+	}
+
+	spin_lock_irqsave(&gpio_lock, flags);
+
+	/*
+	 * Some callers might have not used the gpio_request(),
+	 * so flag this pin as requested now.
+	 */
+	if (!gpio_label[pin])
+		gpio_label[pin] = "?";
+
+	mask = 1 << pin;
+	if (value)
+		orion_setbits(GPIO_OUT, mask);
+	else
+		orion_clrbits(GPIO_OUT, mask);
+	orion_clrbits(GPIO_IO_CONF, mask);
+
+	spin_unlock_irqrestore(&gpio_lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL(gpio_direction_output);
+
+int gpio_get_value(unsigned pin)
+{
+	int val, mask = 1 << pin;
+
+	if (orion_read(GPIO_IO_CONF) & mask)
+		val = orion_read(GPIO_DATA_IN);
+	else
+		val = orion_read(GPIO_OUT);
+
+	return val & mask;
+}
+EXPORT_SYMBOL(gpio_get_value);
+
+void gpio_set_value(unsigned pin, int value)
+{
+	unsigned long flags;
+	int mask = 1 << pin;
+
+	spin_lock_irqsave(&gpio_lock, flags);
+
+	if (value)
+		orion_setbits(GPIO_OUT, mask);
+	else
+		orion_clrbits(GPIO_OUT, mask);
+
+	spin_unlock_irqrestore(&gpio_lock, flags);
+}
+EXPORT_SYMBOL(gpio_set_value);
+
+int gpio_request(unsigned pin, const char *label)
+{
+	int ret = 0;
+	unsigned long flags;
+
+	if (pin >= GPIO_MAX || !test_bit(pin, gpio_valid)) {
+		pr_debug("%s: invalid GPIO %d\n", __FUNCTION__, pin);
+		return -EINVAL;
+	}
+
+	spin_lock_irqsave(&gpio_lock, flags);
+
+	if (gpio_label[pin]) {
+		pr_debug("%s: GPIO %d already used as %s\n",
+			 __FUNCTION__, pin, gpio_label[pin]);
+		ret = -EBUSY;
+	} else
+		gpio_label[pin] = label ? label : "?";
+
+	spin_unlock_irqrestore(&gpio_lock, flags);
+	return ret;
+}
+EXPORT_SYMBOL(gpio_request);
+
+void gpio_free(unsigned pin)
+{
+	if (pin >= GPIO_MAX || !test_bit(pin, gpio_valid)) {
+		pr_debug("%s: invalid GPIO %d\n", __FUNCTION__, pin);
+		return;
+	}
+
+	if (!gpio_label[pin])
+		pr_warning("%s: GPIO %d already freed\n", __FUNCTION__, pin);
+	else
+		gpio_label[pin] = NULL;
+}
+EXPORT_SYMBOL(gpio_free);
+
+/* Debug helper */
+void gpio_display(void)
+{
+	int i;
+
+	for (i = 0; i < GPIO_MAX; i++) {
+		printk(KERN_DEBUG "Pin-%d: ", i);
+
+		if (!test_bit(i, gpio_valid)) {
+			printk("non-GPIO\n");
+		} else if (!gpio_label[i]) {
+			printk("GPIO, free\n");
+		} else {
+			printk("GPIO, used by %s, ", gpio_label[i]);
+			if (orion_read(GPIO_IO_CONF) & (1 << i)) {
+				printk("input, active %s, level %s, edge %s\n",
+				((orion_read(GPIO_IN_POL) >> i) & 1) ? "low" : "high",
+				((orion_read(GPIO_LEVEL_MASK) >> i) & 1) ? "enabled" : "masked",
+				((orion_read(GPIO_EDGE_MASK) >> i) & 1) ? "enabled" : "masked");
+			} else {
+				printk("output, val=%d\n", (orion_read(GPIO_OUT) >> i) & 1);
+			}
+		}
+	}
+
+	printk(KERN_DEBUG "MPP_0_7_CTRL (0x%08x) = 0x%08x\n",
+				MPP_0_7_CTRL, orion_read(MPP_0_7_CTRL));
+	printk(KERN_DEBUG "MPP_8_15_CTRL (0x%08x) = 0x%08x\n",
+				MPP_8_15_CTRL, orion_read(MPP_8_15_CTRL));
+	printk(KERN_DEBUG "MPP_16_19_CTRL (0x%08x) = 0x%08x\n",
+				MPP_16_19_CTRL, orion_read(MPP_16_19_CTRL));
+	printk(KERN_DEBUG "MPP_DEV_CTRL (0x%08x) = 0x%08x\n",
+				MPP_DEV_CTRL, orion_read(MPP_DEV_CTRL));
+	printk(KERN_DEBUG "GPIO_OUT (0x%08x) = 0x%08x\n",
+				GPIO_OUT, orion_read(GPIO_OUT));
+	printk(KERN_DEBUG "GPIO_IO_CONF (0x%08x) = 0x%08x\n",
+				GPIO_IO_CONF, orion_read(GPIO_IO_CONF));
+	printk(KERN_DEBUG "GPIO_BLINK_EN (0x%08x) = 0x%08x\n",
+				GPIO_BLINK_EN, orion_read(GPIO_BLINK_EN));
+	printk(KERN_DEBUG "GPIO_IN_POL (0x%08x) = 0x%08x\n",
+				GPIO_IN_POL, orion_read(GPIO_IN_POL));
+	printk(KERN_DEBUG "GPIO_DATA_IN (0x%08x) = 0x%08x\n",
+				GPIO_DATA_IN, orion_read(GPIO_DATA_IN));
+	printk(KERN_DEBUG "GPIO_LEVEL_MASK (0x%08x) = 0x%08x\n",
+				GPIO_LEVEL_MASK, orion_read(GPIO_LEVEL_MASK));
+	printk(KERN_DEBUG "GPIO_EDGE_CAUSE (0x%08x) = 0x%08x\n",
+				GPIO_EDGE_CAUSE, orion_read(GPIO_EDGE_CAUSE));
+	printk(KERN_DEBUG "GPIO_EDGE_MASK (0x%08x) = 0x%08x\n",
+				GPIO_EDGE_MASK, orion_read(GPIO_EDGE_MASK));
+}
diff --git a/include/asm-arm/arch-orion/gpio.h b/include/asm-arm/arch-orion/gpio.h
new file mode 100644
index 000000000000..6d5848ed9a39
--- /dev/null
+++ b/include/asm-arm/arch-orion/gpio.h
@@ -0,0 +1,27 @@
+/*
+ * include/asm-arm/arch-orion/gpio.h
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+extern int gpio_request(unsigned pin, const char *label);
+extern void gpio_free(unsigned pin);
+extern int gpio_direction_input(unsigned pin);
+extern int gpio_direction_output(unsigned pin, int value);
+extern int gpio_get_value(unsigned pin);
+extern void gpio_set_value(unsigned pin, int value);
+extern void gpio_display(void);		/* debug */
+
+static inline int gpio_to_irq(int pin)
+{
+	return pin + IRQ_ORION_GPIO_START;
+}
+
+static inline int irq_to_gpio(int irq)
+{
+	return irq - IRQ_ORION_GPIO_START;
+}
+
+#include <asm-generic/gpio.h>		/* cansleep wrappers */
-- 
cgit v1.2.3


From b11e9e020c8c6cdd7e7cc6d5178cce2ad0ac0784 Mon Sep 17 00:00:00 2001
From: Herbert Valerio Riedel <hvr@gnu.org>
Date: Thu, 29 Nov 2007 15:19:56 +0100
Subject: [ARM] Orion: provide GPIO method for enabling hardware assisted
 blinking

This is a pre-requisite for implementing proper hardware accelerated
GPIO LED flashing, and since we want proper locking, it's sensible to provide
the orion specific orion_gpio_set_blink() implementation within
mach-orion/gpio.c. The functions orion_gpio_set_blink() and gpio_set_value()
implicitly turn off each others state.

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
Acked-by: Tzachi Perelstein <tzachi@marvell.com>
Acked-by: Nicolas Pitre <nico@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-orion/gpio.c        | 19 +++++++++++++++++++
 include/asm-arm/arch-orion/gpio.h |  1 +
 2 files changed, 20 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-orion/gpio.c b/arch/arm/mach-orion/gpio.c
index af8553ccd230..0418f5b1ebe7 100644
--- a/arch/arm/mach-orion/gpio.c
+++ b/arch/arm/mach-orion/gpio.c
@@ -76,6 +76,7 @@ int gpio_direction_output(unsigned pin, int value)
 		gpio_label[pin] = "?";
 
 	mask = 1 << pin;
+	orion_clrbits(GPIO_BLINK_EN, mask);
 	if (value)
 		orion_setbits(GPIO_OUT, mask);
 	else
@@ -107,6 +108,7 @@ void gpio_set_value(unsigned pin, int value)
 
 	spin_lock_irqsave(&gpio_lock, flags);
 
+	orion_clrbits(GPIO_BLINK_EN, mask);
 	if (value)
 		orion_setbits(GPIO_OUT, mask);
 	else
@@ -116,6 +118,23 @@ void gpio_set_value(unsigned pin, int value)
 }
 EXPORT_SYMBOL(gpio_set_value);
 
+void orion_gpio_set_blink(unsigned pin, int blink)
+{
+	unsigned long flags;
+	int mask = 1 << pin;
+
+	spin_lock_irqsave(&gpio_lock, flags);
+
+	orion_clrbits(GPIO_OUT, mask);
+	if (blink)
+		orion_setbits(GPIO_BLINK_EN, mask);
+	else
+		orion_clrbits(GPIO_BLINK_EN, mask);
+
+	spin_unlock_irqrestore(&gpio_lock, flags);
+}
+EXPORT_SYMBOL(orion_gpio_set_blink);
+
 int gpio_request(unsigned pin, const char *label)
 {
 	int ret = 0;
diff --git a/include/asm-arm/arch-orion/gpio.h b/include/asm-arm/arch-orion/gpio.h
index 6d5848ed9a39..d66284f9a14c 100644
--- a/include/asm-arm/arch-orion/gpio.h
+++ b/include/asm-arm/arch-orion/gpio.h
@@ -12,6 +12,7 @@ extern int gpio_direction_input(unsigned pin);
 extern int gpio_direction_output(unsigned pin, int value);
 extern int gpio_get_value(unsigned pin);
 extern void gpio_set_value(unsigned pin, int value);
+extern void orion_gpio_set_blink(unsigned pin, int blink);
 extern void gpio_display(void);		/* debug */
 
 static inline int gpio_to_irq(int pin)
-- 
cgit v1.2.3


From ca26f7d3ed3c841e561613a9ea2f44ca899e27de Mon Sep 17 00:00:00 2001
From: Tzachi Perelstein <tzachi@marvell.com>
Date: Tue, 23 Oct 2007 15:14:42 -0400
Subject: [ARM] Orion: platform device registration for UART, USB and NAND

Signed-off-by: Tzachi Perelstein <tzachi@marvell.com>
Reviewed-by: Nicolas Pitre <nico@marvell.com>
Reviewed-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-orion/common.c          | 123 ++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-orion/platform.h |  25 +++++++
 2 files changed, 148 insertions(+)
 create mode 100644 include/asm-arm/arch-orion/platform.h

(limited to 'include')

diff --git a/arch/arm/mach-orion/common.c b/arch/arm/mach-orion/common.c
index 69363cbcb63b..ab1000eeaefe 100644
--- a/arch/arm/mach-orion/common.c
+++ b/arch/arm/mach-orion/common.c
@@ -12,6 +12,8 @@
 
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/serial_8250.h>
 #include <asm/page.h>
 #include <asm/timex.h>
 #include <asm/mach/map.h>
@@ -53,6 +55,119 @@ void __init orion_map_io(void)
 	iotable_init(orion_io_desc, ARRAY_SIZE(orion_io_desc));
 }
 
+/*****************************************************************************
+ * UART
+ ****************************************************************************/
+
+static struct resource orion_uart_resources[] = {
+	{
+		.start		= UART0_BASE,
+		.end		= UART0_BASE + 0xff,
+		.flags		= IORESOURCE_MEM,
+	},
+	{
+		.start		= IRQ_ORION_UART0,
+		.end		= IRQ_ORION_UART0,
+		.flags		= IORESOURCE_IRQ,
+	},
+	{
+		.start		= UART1_BASE,
+		.end		= UART1_BASE + 0xff,
+		.flags		= IORESOURCE_MEM,
+	},
+	{
+		.start		= IRQ_ORION_UART1,
+		.end		= IRQ_ORION_UART1,
+		.flags		= IORESOURCE_IRQ,
+	},
+};
+
+static struct plat_serial8250_port orion_uart_data[] = {
+	{
+		.mapbase	= UART0_BASE,
+		.membase	= (char *)UART0_BASE,
+		.irq		= IRQ_ORION_UART0,
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+		.uartclk	= ORION_TCLK,
+	},
+	{
+		.mapbase	= UART1_BASE,
+		.membase	= (char *)UART1_BASE,
+		.irq		= IRQ_ORION_UART1,
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+		.uartclk	= ORION_TCLK,
+	},
+	{ },
+};
+
+static struct platform_device orion_uart = {
+	.name			= "serial8250",
+	.id			= PLAT8250_DEV_PLATFORM,
+	.dev			= {
+		.platform_data	= orion_uart_data,
+	},
+	.resource		= orion_uart_resources,
+	.num_resources		= ARRAY_SIZE(orion_uart_resources),
+};
+
+/*******************************************************************************
+ * USB Controller - 2 interfaces
+ ******************************************************************************/
+
+static struct resource orion_ehci0_resources[] = {
+	{
+		.start	= ORION_USB0_REG_BASE,
+		.end	= ORION_USB0_REG_BASE + SZ_4K,
+		.flags	= IORESOURCE_MEM,
+	},
+	{
+		.start	= IRQ_ORION_USB0_CTRL,
+		.end	= IRQ_ORION_USB0_CTRL,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct resource orion_ehci1_resources[] = {
+	{
+		.start	= ORION_USB1_REG_BASE,
+		.end	= ORION_USB1_REG_BASE + SZ_4K,
+		.flags	= IORESOURCE_MEM,
+	},
+	{
+		.start	= IRQ_ORION_USB1_CTRL,
+		.end	= IRQ_ORION_USB1_CTRL,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static u64 ehci_dmamask = 0xffffffffUL;
+
+static struct platform_device orion_ehci0 = {
+	.name		= "orion-ehci",
+	.id		= 0,
+	.dev		= {
+		.dma_mask		= &ehci_dmamask,
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.resource	= orion_ehci0_resources,
+	.num_resources	= ARRAY_SIZE(orion_ehci0_resources),
+};
+
+static struct platform_device orion_ehci1 = {
+	.name		= "orion-ehci",
+	.id		= 1,
+	.dev		= {
+		.dma_mask		= &ehci_dmamask,
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.resource	= orion_ehci1_resources,
+	.num_resources	= ARRAY_SIZE(orion_ehci1_resources),
+};
+
 /*****************************************************************************
  * General
  ****************************************************************************/
@@ -101,4 +216,12 @@ void __init orion_init(void)
 	orion_setup_pcie_wins();
 	if (dev == MV88F5182_DEV_ID)
 		orion_setup_sata_wins();
+
+	/*
+	 * REgister devices
+	 */
+	platform_device_register(&orion_uart);
+	platform_device_register(&orion_ehci0);
+	if (dev == MV88F5182_DEV_ID)
+		platform_device_register(&orion_ehci1);
 }
diff --git a/include/asm-arm/arch-orion/platform.h b/include/asm-arm/arch-orion/platform.h
new file mode 100644
index 000000000000..143c38e2fa0b
--- /dev/null
+++ b/include/asm-arm/arch-orion/platform.h
@@ -0,0 +1,25 @@
+/*
+ * asm-arm/arch-orion/platform.h
+ *
+ * Tzachi Perelstein <tzachi@marvell.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __ASM_ARCH_PLATFORM_H__
+#define __ASM_ARCH_PLATFORM_H__
+
+/*
+ * Device bus NAND private data
+ */
+struct orion_nand_data {
+	struct mtd_partition *parts;
+	u32 nr_parts;
+	u8 ale;		/* address line number connected to ALE */
+	u8 cle;		/* address line number connected to CLE */
+	u8 width;	/* buswidth */
+};
+
+#endif
-- 
cgit v1.2.3


From c9e3de941a1694aeab60a10bd39eb710c975010d Mon Sep 17 00:00:00 2001
From: Herbert Valerio Riedel <hvr@gnu.org>
Date: Sun, 11 Nov 2007 12:05:11 +0100
Subject: [ARM] Orion: MV88F5181 support bits

add MV88F5181 support bits required by D-link DNS-323 patch

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
Acked-by: Tzachi Perelstein <tzachi@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-orion/addr-map.c     | 4 +++-
 arch/arm/mach-orion/common.c       | 6 ++++++
 arch/arm/mach-orion/pci.c          | 2 +-
 include/asm-arm/arch-orion/orion.h | 3 +++
 4 files changed, 13 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-orion/addr-map.c b/arch/arm/mach-orion/addr-map.c
index ff6e010bccb2..488da3811a68 100644
--- a/arch/arm/mach-orion/addr-map.c
+++ b/arch/arm/mach-orion/addr-map.c
@@ -167,7 +167,9 @@ static int __init orion_cpu_win_can_remap(u32 win)
 	u32 dev, rev;
 
 	orion_pcie_id(&dev, &rev);
-	if ((dev == MV88F5281_DEV_ID && win < 4) || (dev == MV88F5182_DEV_ID && win < 2))
+	if ((dev == MV88F5281_DEV_ID && win < 4)
+	    || (dev == MV88F5182_DEV_ID && win < 2)
+	    || (dev == MV88F5181_DEV_ID && win < 2))
 		return 1;
 
 	return 0;
diff --git a/arch/arm/mach-orion/common.c b/arch/arm/mach-orion/common.c
index 12736fd60004..ff34827f82a5 100644
--- a/arch/arm/mach-orion/common.c
+++ b/arch/arm/mach-orion/common.c
@@ -237,6 +237,12 @@ static void orion_id(u32 *dev, u32 *rev, char **dev_name)
 		} else {
 			*dev_name = "MV88F5182-Rev-Unsupported";
 		}
+	} else if (*dev == MV88F5181_DEV_ID) {
+		if (*rev == MV88F5181_REV_B1) {
+			*dev_name = "MV88F5181-Rev-B1";
+		} else {
+			*dev_name = "MV88F5181-Rev-Unsupported";
+		}
 	} else {
 		*dev_name = "Device-Unknown";
 	}
diff --git a/arch/arm/mach-orion/pci.c b/arch/arm/mach-orion/pci.c
index cf569c647112..0498d7c69b30 100644
--- a/arch/arm/mach-orion/pci.c
+++ b/arch/arm/mach-orion/pci.c
@@ -154,7 +154,7 @@ static int orion_pcie_rd_conf(struct pci_bus *bus, u32 devfn, int where,
 			PCIE_CONF_REG(where) | PCIE_CONF_ADDR_EN);
 
 	orion_pcie_id(&dev, &rev);
-	if (dev == MV88F5182_DEV_ID) {
+	if (dev == MV88F5181_DEV_ID || dev == MV88F5182_DEV_ID) {
 		/* extended register space */
 		pcie_addr = ORION_PCIE_WA_BASE;
 		pcie_addr |= PCIE_CONF_BUS(bus->number) |
diff --git a/include/asm-arm/arch-orion/orion.h b/include/asm-arm/arch-orion/orion.h
index 061c03c5a17e..f787f752e58c 100644
--- a/include/asm-arm/arch-orion/orion.h
+++ b/include/asm-arm/arch-orion/orion.h
@@ -43,6 +43,9 @@
 /*******************************************************************************
  * Supported Devices & Revisions
  ******************************************************************************/
+/* Orion-1 (88F5181) */
+#define MV88F5181_DEV_ID	0x5181
+#define MV88F5181_REV_B1	3
 /* Orion-NAS (88F5182) */
 #define MV88F5182_DEV_ID	0x5182
 #define MV88F5182_REV_A2	2
-- 
cgit v1.2.3


From a0832798c05241f15e793805b6024919c07b8292 Mon Sep 17 00:00:00 2001
From: Tzachi Perelstein <tzachi@marvell.com>
Date: Mon, 12 Nov 2007 19:38:51 +0200
Subject: [I2C] Split mv643xx I2C platform support

The motivation for this change is to allow other chips, like the
Marvell Orion ARM SoC family, to use the existing i2c-mv64xxx driver.

Signed-off-by: Tzachi Perelstein <tzachi@marvell.com>
Acked-by: Nicolas Pitre <nico@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
Acked-by: Mark A. Greer <mgreer@mvista.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/busses/Kconfig       |  2 +-
 drivers/i2c/busses/i2c-mv64xxx.c | 31 +++++++++++++++++--------------
 include/linux/mv643xx.h          | 10 +---------
 include/linux/mv643xx_i2c.h      | 23 +++++++++++++++++++++++
 4 files changed, 42 insertions(+), 24 deletions(-)
 create mode 100644 include/linux/mv643xx_i2c.h

(limited to 'include')

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index c466c6cfc2e5..b148bf0ec6b7 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -648,7 +648,7 @@ config I2C_PCA_ISA
 
 config I2C_MV64XXX
 	tristate "Marvell mv64xxx I2C Controller"
-	depends on MV64X60 && EXPERIMENTAL
+	depends on (MV64X60 || ARCH_ORION) && EXPERIMENTAL
 	help
 	  If you say yes to this option, support will be included for the
 	  built-in I2C interface on the Marvell 64xxx line of host bridges.
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index bb7bf68a7fb6..cdd1ef99fcff 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -1,6 +1,6 @@
 /*
- * Driver for the i2c controller on the Marvell line of host bridges for MIPS
- * and PPC (e.g, gt642[46]0, mv643[46]0, mv644[46]0).
+ * Driver for the i2c controller on the Marvell line of host bridges
+ * (e.g, gt642[46]0, mv643[46]0, mv644[46]0, and Orion SoC family).
  *
  * Author: Mark A. Greer <mgreer@mvista.com>
  *
@@ -14,7 +14,7 @@
 #include <linux/spinlock.h>
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
-#include <linux/mv643xx.h>
+#include <linux/mv643xx_i2c.h>
 #include <linux/platform_device.h>
 
 #include <asm/io.h>
@@ -86,6 +86,7 @@ struct mv64xxx_i2c_data {
 	u32			cntl_bits;
 	void __iomem		*reg_base;
 	u32			reg_base_p;
+	u32			reg_size;
 	u32			addr1;
 	u32			addr2;
 	u32			bytes_left;
@@ -463,17 +464,20 @@ static int __devinit
 mv64xxx_i2c_map_regs(struct platform_device *pd,
 	struct mv64xxx_i2c_data *drv_data)
 {
-	struct resource	*r;
+	int size;
+	struct resource	*r = platform_get_resource(pd, IORESOURCE_MEM, 0);
 
-	if ((r = platform_get_resource(pd, IORESOURCE_MEM, 0)) &&
-		request_mem_region(r->start, MV64XXX_I2C_REG_BLOCK_SIZE,
-			drv_data->adapter.name)) {
+	if (!r)
+		return -ENODEV;
 
-		drv_data->reg_base = ioremap(r->start,
-			MV64XXX_I2C_REG_BLOCK_SIZE);
-		drv_data->reg_base_p = r->start;
-	} else
-		return -ENOMEM;
+	size = r->end - r->start + 1;
+
+	if (!request_mem_region(r->start, size, drv_data->adapter.name))
+		return -EBUSY;
+
+	drv_data->reg_base = ioremap(r->start, size);
+	drv_data->reg_base_p = r->start;
+	drv_data->reg_size = size;
 
 	return 0;
 }
@@ -483,8 +487,7 @@ mv64xxx_i2c_unmap_regs(struct mv64xxx_i2c_data *drv_data)
 {
 	if (drv_data->reg_base) {
 		iounmap(drv_data->reg_base);
-		release_mem_region(drv_data->reg_base_p,
-			MV64XXX_I2C_REG_BLOCK_SIZE);
+		release_mem_region(drv_data->reg_base_p, drv_data->reg_size);
 	}
 
 	drv_data->reg_base = NULL;
diff --git a/include/linux/mv643xx.h b/include/linux/mv643xx.h
index d2ae6185f03b..69327b7b4ce4 100644
--- a/include/linux/mv643xx.h
+++ b/include/linux/mv643xx.h
@@ -15,6 +15,7 @@
 
 #include <asm/types.h>
 #include <linux/mv643xx_eth.h>
+#include <linux/mv643xx_i2c.h>
 
 /****************************************/
 /* Processor Address Space              */
@@ -863,7 +864,6 @@
 /* I2C Registers                        */
 /****************************************/
 
-#define MV64XXX_I2C_CTLR_NAME					"mv64xxx_i2c"
 #define MV64XXX_I2C_OFFSET                                          0xc000
 #define MV64XXX_I2C_REG_BLOCK_SIZE                                  0x0020
 
@@ -968,14 +968,6 @@ struct mpsc_pdata {
 	u32	brg_clk_freq;
 };
 
-/* i2c Platform Device, Driver Data */
-struct mv64xxx_i2c_pdata {
-	u32	freq_m;
-	u32	freq_n;
-	u32	timeout;	/* In milliseconds */
-	u32	retries;
-};
-
 /* Watchdog Platform Device, Driver Data */
 #define	MV64x60_WDT_NAME			"mv64x60_wdt"
 
diff --git a/include/linux/mv643xx_i2c.h b/include/linux/mv643xx_i2c.h
new file mode 100644
index 000000000000..adb629b15477
--- /dev/null
+++ b/include/linux/mv643xx_i2c.h
@@ -0,0 +1,23 @@
+/*
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#ifndef _MV64XXX_I2C_H_
+#define _MV64XXX_I2C_H_
+
+#include <linux/types.h>
+
+#define MV64XXX_I2C_CTLR_NAME	"mv64xxx_i2c"
+
+/* i2c Platform Device, Driver Data */
+struct mv64xxx_i2c_pdata {
+	u32	freq_m;
+	u32	freq_n;
+	u32	timeout;	/* In milliseconds */
+	u32	retries;
+};
+
+#endif /*_MV64XXX_I2C_H_*/
-- 
cgit v1.2.3


From 2f0a8df40ff008822e5570b3323c56622cd92c95 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Thu, 22 Nov 2007 16:58:08 +0100
Subject: [I2C] i2c-mv64xxx: Don't set i2c_adapter.retries

I2C adapter drivers are supposed to handle retries on nack by themselves
if they do, so there's no point in setting .retries if they don't.

As this retry mechanism is going away (at least in its current form),
clean this up now so that we don't get build failures later.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Mark A. Greer <mgreer@mvista.com>
---
 arch/powerpc/sysdev/mv64x60_dev.c | 6 ------
 arch/ppc/syslib/mv64x60.c         | 1 -
 drivers/i2c/busses/i2c-mv64xxx.c  | 1 -
 include/linux/mv643xx_i2c.h       | 1 -
 4 files changed, 9 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/sysdev/mv64x60_dev.c b/arch/powerpc/sysdev/mv64x60_dev.c
index 548a32082e4a..4316f5a48a0f 100644
--- a/arch/powerpc/sysdev/mv64x60_dev.c
+++ b/arch/powerpc/sysdev/mv64x60_dev.c
@@ -361,12 +361,6 @@ static int __init mv64x60_i2c_device_setup(struct device_node *np, int id)
 	else
 		pdata.timeout = 1000;	/* 1 second */
 
-	prop = of_get_property(np, "retries", NULL);
-	if (prop)
-		pdata.retries = *prop;
-	else
-		pdata.retries = 1;
-
 	pdev = platform_device_alloc(MV64XXX_I2C_CTLR_NAME, id);
 	if (!pdev)
 		return -ENOMEM;
diff --git a/arch/ppc/syslib/mv64x60.c b/arch/ppc/syslib/mv64x60.c
index 2744b8a6f66a..90fe904d3614 100644
--- a/arch/ppc/syslib/mv64x60.c
+++ b/arch/ppc/syslib/mv64x60.c
@@ -411,7 +411,6 @@ static struct mv64xxx_i2c_pdata mv64xxx_i2c_pdata = {
 	.freq_m			= 8,
 	.freq_n			= 3,
 	.timeout		= 1000, /* Default timeout of 1 second */
-	.retries		= 1,
 };
 
 static struct resource mv64xxx_i2c_resources[] = {
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index cdd1ef99fcff..036e6a883e67 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -532,7 +532,6 @@ mv64xxx_i2c_probe(struct platform_device *pd)
 	drv_data->adapter.owner = THIS_MODULE;
 	drv_data->adapter.class = I2C_CLASS_HWMON;
 	drv_data->adapter.timeout = pdata->timeout;
-	drv_data->adapter.retries = pdata->retries;
 	drv_data->adapter.nr = pd->id;
 	platform_set_drvdata(pd, drv_data);
 	i2c_set_adapdata(&drv_data->adapter, drv_data);
diff --git a/include/linux/mv643xx_i2c.h b/include/linux/mv643xx_i2c.h
index adb629b15477..5db5152e9de5 100644
--- a/include/linux/mv643xx_i2c.h
+++ b/include/linux/mv643xx_i2c.h
@@ -17,7 +17,6 @@ struct mv64xxx_i2c_pdata {
 	u32	freq_m;
 	u32	freq_n;
 	u32	timeout;	/* In milliseconds */
-	u32	retries;
 };
 
 #endif /*_MV64XXX_I2C_H_*/
-- 
cgit v1.2.3


From 49db76eb5fd7d75babb4f3a5f30e86d1f8e82543 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 22 Nov 2007 17:57:30 +0100
Subject: [ARM] 4663/1: Only putc uncompressor output into FFUART if it was
 enabled by the bootloader

Also, use existing register and bit definitions instead of numbers.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-pxa/uncompress.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-pxa/uncompress.h b/include/asm-arm/arch-pxa/uncompress.h
index 178aa2e073ac..3faad53a6848 100644
--- a/include/asm-arm/arch-pxa/uncompress.h
+++ b/include/asm-arm/arch-pxa/uncompress.h
@@ -9,6 +9,9 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/serial_reg.h>
+#include <asm/arch/pxa-regs.h>
+
 #define FFUART		((volatile unsigned long *)0x40100000)
 #define BTUART		((volatile unsigned long *)0x40200000)
 #define STUART		((volatile unsigned long *)0x40700000)
@@ -19,9 +22,11 @@
 
 static inline void putc(char c)
 {
-	while (!(UART[5] & 0x20))
+	if (!(UART[UART_IER] & IER_UUE))
+		return;
+	while (!(UART[UART_LSR] & LSR_TDRQ))
 		barrier();
-	UART[0] = c;
+	UART[UART_TX] = c;
 }
 
 /*
-- 
cgit v1.2.3


From 88286450462216ca9b5c67c2175d75a5aebd5d0f Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Thu, 6 Dec 2007 17:56:42 +0800
Subject: [ARM] pxa: define "struct ssp_device" and add
 ssp_request()/ssp_free()

1. define "struct ssp_device" for SSP information, which is requested
   and released by function ssp_request()/ssp_free()

2. modify the ssp_init() and ssp_exit() to use the interface

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/ssp.c        | 296 +++++++++++++++++++++++++++++++++--------
 include/asm-arm/arch-pxa/ssp.h |  36 ++++-
 2 files changed, 273 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/ssp.c b/arch/arm/mach-pxa/ssp.c
index 422afee88169..be6fee78f439 100644
--- a/arch/arm/mach-pxa/ssp.c
+++ b/arch/arm/mach-pxa/ssp.c
@@ -32,39 +32,18 @@
 #include <linux/ioport.h>
 #include <linux/init.h>
 #include <linux/mutex.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/hardware.h>
 #include <asm/arch/ssp.h>
 #include <asm/arch/pxa-regs.h>
 
-#define PXA_SSP_PORTS 	3
-
 #define TIMEOUT 100000
 
-struct ssp_info_ {
-	int irq;
-	u32 clock;
-};
-
-/*
- * SSP port clock and IRQ settings
- */
-static const struct ssp_info_ ssp_info[PXA_SSP_PORTS] = {
-#if defined (CONFIG_PXA27x)
-	{IRQ_SSP,	CKEN_SSP1},
-	{IRQ_SSP2,	CKEN_SSP2},
-	{IRQ_SSP3,	CKEN_SSP3},
-#else
-	{IRQ_SSP,	CKEN_SSP},
-	{IRQ_NSSP,	CKEN_NSSP},
-	{IRQ_ASSP,	CKEN_ASSP},
-#endif
-};
-
-static DEFINE_MUTEX(mutex);
-static int use_count[PXA_SSP_PORTS] = {0, 0, 0};
-
 static irqreturn_t ssp_interrupt(int irq, void *dev_id)
 {
 	struct ssp_dev *dev = (struct ssp_dev*) dev_id;
@@ -256,44 +235,32 @@ int ssp_config(struct ssp_dev *dev, u32 mode, u32 flags, u32 psp_flags, u32 spee
  */
 int ssp_init(struct ssp_dev *dev, u32 port, u32 init_flags)
 {
+	struct ssp_device *ssp;
 	int ret;
 
-	if (port > PXA_SSP_PORTS || port == 0)
+	ssp = ssp_request(port, "SSP");
+	if (ssp == NULL)
 		return -ENODEV;
 
-	mutex_lock(&mutex);
-	if (use_count[port - 1]) {
-		mutex_unlock(&mutex);
-		return -EBUSY;
-	}
-	use_count[port - 1]++;
-
-	if (!request_mem_region(__PREG(SSCR0_P(port)), 0x2c, "SSP")) {
-		use_count[port - 1]--;
-		mutex_unlock(&mutex);
-		return -EBUSY;
-	}
+	dev->ssp = ssp;
 	dev->port = port;
 
 	/* do we need to get irq */
 	if (!(init_flags & SSP_NO_IRQ)) {
-		ret = request_irq(ssp_info[port-1].irq, ssp_interrupt,
+		ret = request_irq(ssp->irq, ssp_interrupt,
 				0, "SSP", dev);
 	    	if (ret)
 			goto out_region;
-	    	dev->irq = ssp_info[port-1].irq;
+		dev->irq = ssp->irq;
 	} else
 		dev->irq = 0;
 
 	/* turn on SSP port clock */
-	pxa_set_cken(ssp_info[port-1].clock, 1);
-	mutex_unlock(&mutex);
+	clk_enable(ssp->clk);
 	return 0;
 
 out_region:
-	release_mem_region(__PREG(SSCR0_P(port)), 0x2c);
-	use_count[port - 1]--;
-	mutex_unlock(&mutex);
+	ssp_free(ssp);
 	return ret;
 }
 
@@ -304,22 +271,239 @@ out_region:
  */
 void ssp_exit(struct ssp_dev *dev)
 {
-	mutex_lock(&mutex);
+	struct ssp_device *ssp = dev->ssp;
+
 	SSCR0_P(dev->port) &= ~SSCR0_SSE;
+	free_irq(dev->irq, dev);
+	clk_disable(ssp->clk);
+	ssp_free(ssp);
+}
+
+static DEFINE_MUTEX(ssp_lock);
+static LIST_HEAD(ssp_list);
+
+struct ssp_device *ssp_request(int port, const char *label)
+{
+	struct ssp_device *ssp = NULL;
+
+	mutex_lock(&ssp_lock);
 
-    	if (dev->port > PXA_SSP_PORTS || dev->port == 0) {
-		printk(KERN_WARNING "SSP: tried to close invalid port\n");
-		mutex_unlock(&mutex);
-		return;
+	list_for_each_entry(ssp, &ssp_list, node) {
+		if (ssp->port_id == port && ssp->use_count == 0) {
+			ssp->use_count++;
+			ssp->label = label;
+			break;
+		}
 	}
 
-	pxa_set_cken(ssp_info[dev->port-1].clock, 0);
-	if (dev->irq)
-		free_irq(dev->irq, dev);
-	release_mem_region(__PREG(SSCR0_P(dev->port)), 0x2c);
-	use_count[dev->port - 1]--;
-	mutex_unlock(&mutex);
+	mutex_unlock(&ssp_lock);
+
+	if (ssp->port_id != port)
+		return NULL;
+
+	return ssp;
+}
+EXPORT_SYMBOL(ssp_request);
+
+void ssp_free(struct ssp_device *ssp)
+{
+	mutex_lock(&ssp_lock);
+	if (ssp->use_count) {
+		ssp->use_count--;
+		ssp->label = NULL;
+	} else
+		dev_err(&ssp->pdev->dev, "device already free\n");
+	mutex_unlock(&ssp_lock);
 }
+EXPORT_SYMBOL(ssp_free);
+
+static int __devinit ssp_probe(struct platform_device *pdev, int type)
+{
+	struct resource *res;
+	struct ssp_device *ssp;
+	int ret = 0;
+
+	ssp = kzalloc(sizeof(struct ssp_device), GFP_KERNEL);
+	if (ssp == NULL) {
+		dev_err(&pdev->dev, "failed to allocate memory");
+		return -ENOMEM;
+	}
+
+	ssp->clk = clk_get(&pdev->dev, "SSPCLK");
+	if (IS_ERR(ssp->clk)) {
+		ret = PTR_ERR(ssp->clk);
+		goto err_free;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "no memory resource defined\n");
+		ret = -ENODEV;
+		goto err_free_clk;
+	}
+
+	res = request_mem_region(res->start, res->end - res->start + 1,
+			pdev->name);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "failed to request memory resource\n");
+		ret = -EBUSY;
+		goto err_free_clk;
+	}
+
+	ssp->phys_base = res->start;
+
+	ssp->mmio_base = ioremap(res->start, res->end - res->start + 1);
+	if (ssp->mmio_base == NULL) {
+		dev_err(&pdev->dev, "failed to ioremap() registers\n");
+		ret = -ENODEV;
+		goto err_free_mem;
+	}
+
+	ssp->irq = platform_get_irq(pdev, 0);
+	if (ssp->irq < 0) {
+		dev_err(&pdev->dev, "no IRQ resource defined\n");
+		ret = -ENODEV;
+		goto err_free_io;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "no SSP RX DRCMR defined\n");
+		ret = -ENODEV;
+		goto err_free_io;
+	}
+	ssp->drcmr_rx = res->start;
+
+	res = platform_get_resource(pdev, IORESOURCE_DMA, 1);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "no SSP TX DRCMR defined\n");
+		ret = -ENODEV;
+		goto err_free_io;
+	}
+	ssp->drcmr_tx = res->start;
+
+	/* PXA2xx/3xx SSP ports starts from 1 and the internal pdev->id
+	 * starts from 0, do a translation here
+	 */
+	ssp->port_id = pdev->id + 1;
+	ssp->use_count = 0;
+	ssp->type = type;
+
+	mutex_lock(&ssp_lock);
+	list_add(&ssp->node, &ssp_list);
+	mutex_unlock(&ssp_lock);
+
+	platform_set_drvdata(pdev, ssp);
+	return 0;
+
+err_free_io:
+	iounmap(ssp->mmio_base);
+err_free_mem:
+	release_mem_region(res->start, res->end - res->start + 1);
+err_free_clk:
+	clk_put(ssp->clk);
+err_free:
+	kfree(ssp);
+	return ret;
+}
+
+static int __devexit ssp_remove(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct ssp_device *ssp;
+
+	ssp = platform_get_drvdata(pdev);
+	if (ssp == NULL)
+		return -ENODEV;
+
+	iounmap(ssp->mmio_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(res->start, res->end - res->start + 1);
+
+	clk_put(ssp->clk);
+
+	mutex_lock(&ssp_lock);
+	list_del(&ssp->node);
+	mutex_unlock(&ssp_lock);
+
+	kfree(ssp);
+	return 0;
+}
+
+static int __devinit pxa25x_ssp_probe(struct platform_device *pdev)
+{
+	return ssp_probe(pdev, PXA25x_SSP);
+}
+
+static int __devinit pxa25x_nssp_probe(struct platform_device *pdev)
+{
+	return ssp_probe(pdev, PXA25x_NSSP);
+}
+
+static int __devinit pxa27x_ssp_probe(struct platform_device *pdev)
+{
+	return ssp_probe(pdev, PXA27x_SSP);
+}
+
+static struct platform_driver pxa25x_ssp_driver = {
+	.driver		= {
+		.name	= "pxa25x-ssp",
+	},
+	.probe		= pxa25x_ssp_probe,
+	.remove		= __devexit_p(ssp_remove),
+};
+
+static struct platform_driver pxa25x_nssp_driver = {
+	.driver		= {
+		.name	= "pxa25x-nssp",
+	},
+	.probe		= pxa25x_nssp_probe,
+	.remove		= __devexit_p(ssp_remove),
+};
+
+static struct platform_driver pxa27x_ssp_driver = {
+	.driver		= {
+		.name	= "pxa27x-ssp",
+	},
+	.probe		= pxa27x_ssp_probe,
+	.remove		= __devexit_p(ssp_remove),
+};
+
+static int __init pxa_ssp_init(void)
+{
+	int ret = 0;
+
+	ret = platform_driver_register(&pxa25x_ssp_driver);
+	if (ret) {
+		printk(KERN_ERR "failed to register pxa25x_ssp_driver");
+		return ret;
+	}
+
+	ret = platform_driver_register(&pxa25x_nssp_driver);
+	if (ret) {
+		printk(KERN_ERR "failed to register pxa25x_nssp_driver");
+		return ret;
+	}
+
+	ret = platform_driver_register(&pxa27x_ssp_driver);
+	if (ret) {
+		printk(KERN_ERR "failed to register pxa27x_ssp_driver");
+		return ret;
+	}
+
+	return ret;
+}
+
+static void __exit pxa_ssp_exit(void)
+{
+	platform_driver_unregister(&pxa25x_ssp_driver);
+	platform_driver_unregister(&pxa25x_nssp_driver);
+	platform_driver_unregister(&pxa27x_ssp_driver);
+}
+
+module_init(pxa_ssp_init);
+module_exit(pxa_ssp_exit);
 
 EXPORT_SYMBOL(ssp_write_word);
 EXPORT_SYMBOL(ssp_read_word);
diff --git a/include/asm-arm/arch-pxa/ssp.h b/include/asm-arm/arch-pxa/ssp.h
index ea200551a75f..a012882c9ee6 100644
--- a/include/asm-arm/arch-pxa/ssp.h
+++ b/include/asm-arm/arch-pxa/ssp.h
@@ -13,10 +13,37 @@
  *       PXA255     SSP, NSSP
  *       PXA26x     SSP, NSSP, ASSP
  *       PXA27x     SSP1, SSP2, SSP3
+ *       PXA3xx     SSP1, SSP2, SSP3, SSP4
  */
 
-#ifndef SSP_H
-#define SSP_H
+#ifndef __ASM_ARCH_SSP_H
+#define __ASM_ARCH_SSP_H
+
+#include <linux/list.h>
+
+enum pxa_ssp_type {
+	SSP_UNDEFINED = 0,
+	PXA25x_SSP,  /* pxa 210, 250, 255, 26x */
+	PXA25x_NSSP, /* pxa 255, 26x (including ASSP) */
+	PXA27x_SSP,
+};
+
+struct ssp_device {
+	struct platform_device *pdev;
+	struct list_head	node;
+
+	struct clk	*clk;
+	void __iomem	*mmio_base;
+	unsigned long	phys_base;
+
+	const char	*label;
+	int		port_id;
+	int		type;
+	int		use_count;
+	int		irq;
+	int		drcmr_rx;
+	int		drcmr_tx;
+};
 
 /*
  * SSP initialisation flags
@@ -31,6 +58,7 @@ struct ssp_state {
 };
 
 struct ssp_dev {
+	struct ssp_device *ssp;
 	u32 port;
 	u32 mode;
 	u32 flags;
@@ -50,4 +78,6 @@ int ssp_init(struct ssp_dev *dev, u32 port, u32 init_flags);
 int ssp_config(struct ssp_dev *dev, u32 mode, u32 flags, u32 psp_flags, u32 speed);
 void ssp_exit(struct ssp_dev *dev);
 
-#endif
+struct ssp_device *ssp_request(int port, const char *label);
+void ssp_free(struct ssp_device *);
+#endif /* __ASM_ARCH_SSP_H */
-- 
cgit v1.2.3


From 0aea1fd565857f002e873a506d67c92ff913f1af Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Wed, 21 Nov 2007 16:57:12 +0800
Subject: [ARM] pxa: move SSP register definitions from pxa-regs.h to
 regs-ssp.h

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/corgi_ssp.c       |   1 +
 arch/arm/mach-pxa/ssp.c             |   1 +
 drivers/spi/pxa2xx_spi.c            |   1 +
 include/asm-arm/arch-pxa/pxa-regs.h | 168 +---------------------------------
 include/asm-arm/arch-pxa/regs-ssp.h | 173 ++++++++++++++++++++++++++++++++++++
 5 files changed, 177 insertions(+), 167 deletions(-)
 create mode 100644 include/asm-arm/arch-pxa/regs-ssp.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/corgi_ssp.c b/arch/arm/mach-pxa/corgi_ssp.c
index 40dea3d5142b..efba65edcd51 100644
--- a/arch/arm/mach-pxa/corgi_ssp.c
+++ b/arch/arm/mach-pxa/corgi_ssp.c
@@ -21,6 +21,7 @@
 
 #include <asm/arch/ssp.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/regs-ssp.h>
 #include "sharpsl.h"
 
 static DEFINE_SPINLOCK(corgi_ssp_lock);
diff --git a/arch/arm/mach-pxa/ssp.c b/arch/arm/mach-pxa/ssp.c
index be6fee78f439..0225ee016f2d 100644
--- a/arch/arm/mach-pxa/ssp.c
+++ b/arch/arm/mach-pxa/ssp.c
@@ -41,6 +41,7 @@
 #include <asm/hardware.h>
 #include <asm/arch/ssp.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/regs-ssp.h>
 
 #define TIMEOUT 100000
 
diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index 1c2ab541d37d..0ac0f65f69a1 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -36,6 +36,7 @@
 
 #include <asm/arch/hardware.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/regs-ssp.h>
 #include <asm/arch/pxa2xx_spi.h>
 
 MODULE_AUTHOR("Stephen Street");
diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h
index 1bd398da07da..dd014712dfa5 100644
--- a/include/asm-arm/arch-pxa/pxa-regs.h
+++ b/include/asm-arm/arch-pxa/pxa-regs.h
@@ -1597,176 +1597,10 @@
 #define PWER_GPIO15	PWER_GPIO (15)	/* GPIO [15] wake-up enable        */
 #define PWER_RTC	0x80000000	/* RTC alarm wake-up enable        */
 
-
 /*
- * SSP Serial Port Registers
- * PXA250, PXA255, PXA26x and PXA27x SSP controllers are all slightly different.
- * PXA255, PXA26x and PXA27x have extra ports, registers and bits.
+ * SSP Serial Port Registers - see include/asm-arm/arch-pxa/regs-ssp.h
  */
 
- /* Common PXA2xx bits first */
-#define SSCR0_DSS	(0x0000000f)	/* Data Size Select (mask) */
-#define SSCR0_DataSize(x)  ((x) - 1)	/* Data Size Select [4..16] */
-#define SSCR0_FRF	(0x00000030)	/* FRame Format (mask) */
-#define SSCR0_Motorola	(0x0 << 4)	/* Motorola's Serial Peripheral Interface (SPI) */
-#define SSCR0_TI	(0x1 << 4)	/* Texas Instruments' Synchronous Serial Protocol (SSP) */
-#define SSCR0_National	(0x2 << 4)	/* National Microwire */
-#define SSCR0_ECS	(1 << 6)	/* External clock select */
-#define SSCR0_SSE	(1 << 7)	/* Synchronous Serial Port Enable */
-#if defined(CONFIG_PXA25x)
-#define SSCR0_SCR	(0x0000ff00)	/* Serial Clock Rate (mask) */
-#define SSCR0_SerClkDiv(x) ((((x) - 2)/2) << 8) /* Divisor [2..512] */
-#elif defined(CONFIG_PXA27x)
-#define SSCR0_SCR	(0x000fff00)	/* Serial Clock Rate (mask) */
-#define SSCR0_SerClkDiv(x) (((x) - 1) << 8) /* Divisor [1..4096] */
-#define SSCR0_EDSS	(1 << 20)	/* Extended data size select */
-#define SSCR0_NCS	(1 << 21)	/* Network clock select */
-#define SSCR0_RIM	(1 << 22)	/* Receive FIFO overrrun interrupt mask */
-#define SSCR0_TUM	(1 << 23)	/* Transmit FIFO underrun interrupt mask */
-#define SSCR0_FRDC	(0x07000000)	/* Frame rate divider control (mask) */
-#define SSCR0_SlotsPerFrm(x) (((x) - 1) << 24)	/* Time slots per frame [1..8] */
-#define SSCR0_ADC	(1 << 30)	/* Audio clock select */
-#define SSCR0_MOD	(1 << 31)	/* Mode (normal or network) */
-#endif
-
-#define SSCR1_RIE	(1 << 0)	/* Receive FIFO Interrupt Enable */
-#define SSCR1_TIE	(1 << 1)	/* Transmit FIFO Interrupt Enable */
-#define SSCR1_LBM	(1 << 2)	/* Loop-Back Mode */
-#define SSCR1_SPO	(1 << 3)	/* Motorola SPI SSPSCLK polarity setting */
-#define SSCR1_SPH	(1 << 4)	/* Motorola SPI SSPSCLK phase setting */
-#define SSCR1_MWDS	(1 << 5)	/* Microwire Transmit Data Size */
-#define SSCR1_TFT	(0x000003c0)	/* Transmit FIFO Threshold (mask) */
-#define SSCR1_TxTresh(x) (((x) - 1) << 6) /* level [1..16] */
-#define SSCR1_RFT	(0x00003c00)	/* Receive FIFO Threshold (mask) */
-#define SSCR1_RxTresh(x) (((x) - 1) << 10) /* level [1..16] */
-
-#define SSSR_TNF	(1 << 2)	/* Transmit FIFO Not Full */
-#define SSSR_RNE	(1 << 3)	/* Receive FIFO Not Empty */
-#define SSSR_BSY	(1 << 4)	/* SSP Busy */
-#define SSSR_TFS	(1 << 5)	/* Transmit FIFO Service Request */
-#define SSSR_RFS	(1 << 6)	/* Receive FIFO Service Request */
-#define SSSR_ROR	(1 << 7)	/* Receive FIFO Overrun */
-
-#define SSCR0_TIM		(1 << 23)	/* Transmit FIFO Under Run Interrupt Mask */
-#define SSCR0_RIM		(1 << 22)	/* Receive FIFO Over Run interrupt Mask */
-#define SSCR0_NCS		(1 << 21)	/* Network Clock Select */
-#define SSCR0_EDSS		(1 << 20)	/* Extended Data Size Select */
-
-/* extra bits in PXA255, PXA26x and PXA27x SSP ports */
-#define SSCR0_TISSP		(1 << 4)	/* TI Sync Serial Protocol */
-#define SSCR0_PSP		(3 << 4)	/* PSP - Programmable Serial Protocol */
-#define SSCR1_TTELP		(1 << 31)	/* TXD Tristate Enable Last Phase */
-#define SSCR1_TTE		(1 << 30)	/* TXD Tristate Enable */
-#define SSCR1_EBCEI		(1 << 29)	/* Enable Bit Count Error interrupt */
-#define SSCR1_SCFR		(1 << 28)	/* Slave Clock free Running */
-#define SSCR1_ECRA		(1 << 27)	/* Enable Clock Request A */
-#define SSCR1_ECRB		(1 << 26)	/* Enable Clock request B */
-#define SSCR1_SCLKDIR	(1 << 25)	/* Serial Bit Rate Clock Direction */
-#define SSCR1_SFRMDIR	(1 << 24)	/* Frame Direction */
-#define SSCR1_RWOT		(1 << 23)	/* Receive Without Transmit */
-#define SSCR1_TRAIL		(1 << 22)	/* Trailing Byte */
-#define SSCR1_TSRE		(1 << 21)	/* Transmit Service Request Enable */
-#define SSCR1_RSRE		(1 << 20)	/* Receive Service Request Enable */
-#define SSCR1_TINTE		(1 << 19)	/* Receiver Time-out Interrupt enable */
-#define SSCR1_PINTE		(1 << 18)	/* Peripheral Trailing Byte Interupt Enable */
-#define SSCR1_STRF		(1 << 15)	/* Select FIFO or EFWR */
-#define SSCR1_EFWR		(1 << 14)	/* Enable FIFO Write/Read */
-
-#define SSSR_BCE		(1 << 23)	/* Bit Count Error */
-#define SSSR_CSS		(1 << 22)	/* Clock Synchronisation Status */
-#define SSSR_TUR		(1 << 21)	/* Transmit FIFO Under Run */
-#define SSSR_EOC		(1 << 20)	/* End Of Chain */
-#define SSSR_TINT		(1 << 19)	/* Receiver Time-out Interrupt */
-#define SSSR_PINT		(1 << 18)	/* Peripheral Trailing Byte Interrupt */
-
-#define SSPSP_FSRT		(1 << 25)	/* Frame Sync Relative Timing */
-#define SSPSP_DMYSTOP(x)	((x) << 23)	/* Dummy Stop */
-#define SSPSP_SFRMWDTH(x)	((x) << 16)	/* Serial Frame Width */
-#define SSPSP_SFRMDLY(x)	((x) << 9)	/* Serial Frame Delay */
-#define SSPSP_DMYSTRT(x)	((x) << 7)	/* Dummy Start */
-#define SSPSP_STRTDLY(x)	((x) << 4)	/* Start Delay */
-#define SSPSP_ETDS			(1 << 3)	/* End of Transfer data State */
-#define SSPSP_SFRMP			(1 << 2)	/* Serial Frame Polarity */
-#define SSPSP_SCMODE(x)		((x) << 0)	/* Serial Bit Rate Clock Mode */
-
-#define SSACD_SCDB		(1 << 3)	/* SSPSYSCLK Divider Bypass */
-#define SSACD_ACPS(x)		((x) << 4)	/* Audio clock PLL select */
-#define SSACD_ACDS(x)		((x) << 0)	/* Audio clock divider select */
-
-#define SSCR0_P1	__REG(0x41000000)  /* SSP Port 1 Control Register 0 */
-#define SSCR1_P1	__REG(0x41000004)  /* SSP Port 1 Control Register 1 */
-#define SSSR_P1		__REG(0x41000008)  /* SSP Port 1 Status Register */
-#define SSITR_P1	__REG(0x4100000C)  /* SSP Port 1 Interrupt Test Register */
-#define SSDR_P1		__REG(0x41000010)  /* (Write / Read) SSP Port 1 Data Write Register/SSP Data Read Register */
-
-/* Support existing PXA25x drivers */
-#define SSCR0		SSCR0_P1  /* SSP Control Register 0 */
-#define SSCR1		SSCR1_P1  /* SSP Control Register 1 */
-#define SSSR		SSSR_P1	  /* SSP Status Register */
-#define SSITR		SSITR_P1  /* SSP Interrupt Test Register */
-#define SSDR		SSDR_P1	  /* (Write / Read) SSP Data Write Register/SSP Data Read Register */
-
-/* PXA27x ports */
-#if defined (CONFIG_PXA27x)
-#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
-#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
-#define SSTSA_P1	__REG(0x41000030)  /* SSP Port 1 Tx Timeslot Active */
-#define SSRSA_P1	__REG(0x41000034)  /* SSP Port 1 Rx Timeslot Active */
-#define SSTSS_P1	__REG(0x41000038)  /* SSP Port 1 Timeslot Status */
-#define SSACD_P1	__REG(0x4100003C)  /* SSP Port 1 Audio Clock Divider */
-#define SSCR0_P2	__REG(0x41700000)  /* SSP Port 2 Control Register 0 */
-#define SSCR1_P2	__REG(0x41700004)  /* SSP Port 2 Control Register 1 */
-#define SSSR_P2		__REG(0x41700008)  /* SSP Port 2 Status Register */
-#define SSITR_P2	__REG(0x4170000C)  /* SSP Port 2 Interrupt Test Register */
-#define SSDR_P2		__REG(0x41700010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
-#define SSTO_P2		__REG(0x41700028)  /* SSP Port 2 Time Out Register */
-#define SSPSP_P2	__REG(0x4170002C)  /* SSP Port 2 Programmable Serial Protocol */
-#define SSTSA_P2	__REG(0x41700030)  /* SSP Port 2 Tx Timeslot Active */
-#define SSRSA_P2	__REG(0x41700034)  /* SSP Port 2 Rx Timeslot Active */
-#define SSTSS_P2	__REG(0x41700038)  /* SSP Port 2 Timeslot Status */
-#define SSACD_P2	__REG(0x4170003C)  /* SSP Port 2 Audio Clock Divider */
-#define SSCR0_P3	__REG(0x41900000)  /* SSP Port 3 Control Register 0 */
-#define SSCR1_P3	__REG(0x41900004)  /* SSP Port 3 Control Register 1 */
-#define SSSR_P3		__REG(0x41900008)  /* SSP Port 3 Status Register */
-#define SSITR_P3	__REG(0x4190000C)  /* SSP Port 3 Interrupt Test Register */
-#define SSDR_P3		__REG(0x41900010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
-#define SSTO_P3		__REG(0x41900028)  /* SSP Port 3 Time Out Register */
-#define SSPSP_P3	__REG(0x4190002C)  /* SSP Port 3 Programmable Serial Protocol */
-#define SSTSA_P3	__REG(0x41900030)  /* SSP Port 3 Tx Timeslot Active */
-#define SSRSA_P3	__REG(0x41900034)  /* SSP Port 3 Rx Timeslot Active */
-#define SSTSS_P3	__REG(0x41900038)  /* SSP Port 3 Timeslot Status */
-#define SSACD_P3	__REG(0x4190003C)  /* SSP Port 3 Audio Clock Divider */
-#else /* PXA255 (only port 2) and PXA26x ports*/
-#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
-#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
-#define SSCR0_P2	__REG(0x41400000)  /* SSP Port 2 Control Register 0 */
-#define SSCR1_P2	__REG(0x41400004)  /* SSP Port 2 Control Register 1 */
-#define SSSR_P2		__REG(0x41400008)  /* SSP Port 2 Status Register */
-#define SSITR_P2	__REG(0x4140000C)  /* SSP Port 2 Interrupt Test Register */
-#define SSDR_P2		__REG(0x41400010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
-#define SSTO_P2		__REG(0x41400028)  /* SSP Port 2 Time Out Register */
-#define SSPSP_P2	__REG(0x4140002C)  /* SSP Port 2 Programmable Serial Protocol */
-#define SSCR0_P3	__REG(0x41500000)  /* SSP Port 3 Control Register 0 */
-#define SSCR1_P3	__REG(0x41500004)  /* SSP Port 3 Control Register 1 */
-#define SSSR_P3		__REG(0x41500008)  /* SSP Port 3 Status Register */
-#define SSITR_P3	__REG(0x4150000C)  /* SSP Port 3 Interrupt Test Register */
-#define SSDR_P3		__REG(0x41500010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
-#define SSTO_P3		__REG(0x41500028)  /* SSP Port 3 Time Out Register */
-#define SSPSP_P3	__REG(0x4150002C)  /* SSP Port 3 Programmable Serial Protocol */
-#endif
-
-#define SSCR0_P(x) (*(((x) == 1) ? &SSCR0_P1 : ((x) == 2) ? &SSCR0_P2 : ((x) == 3) ? &SSCR0_P3 : NULL))
-#define SSCR1_P(x) (*(((x) == 1) ? &SSCR1_P1 : ((x) == 2) ? &SSCR1_P2 : ((x) == 3) ? &SSCR1_P3 : NULL))
-#define SSSR_P(x) (*(((x) == 1) ? &SSSR_P1 : ((x) == 2) ? &SSSR_P2 : ((x) == 3) ? &SSSR_P3 : NULL))
-#define SSITR_P(x) (*(((x) == 1) ? &SSITR_P1 : ((x) == 2) ? &SSITR_P2 : ((x) == 3) ? &SSITR_P3 : NULL))
-#define SSDR_P(x) (*(((x) == 1) ? &SSDR_P1 : ((x) == 2) ? &SSDR_P2 : ((x) == 3) ? &SSDR_P3 : NULL))
-#define SSTO_P(x) (*(((x) == 1) ? &SSTO_P1 : ((x) == 2) ? &SSTO_P2 : ((x) == 3) ? &SSTO_P3 : NULL))
-#define SSPSP_P(x) (*(((x) == 1) ? &SSPSP_P1 : ((x) == 2) ? &SSPSP_P2 : ((x) == 3) ? &SSPSP_P3 : NULL))
-#define SSTSA_P(x) (*(((x) == 1) ? &SSTSA_P1 : ((x) == 2) ? &SSTSA_P2 : ((x) == 3) ? &SSTSA_P3 : NULL))
-#define SSRSA_P(x) (*(((x) == 1) ? &SSRSA_P1 : ((x) == 2) ? &SSRSA_P2 : ((x) == 3) ? &SSRSA_P3 : NULL))
-#define SSTSS_P(x) (*(((x) == 1) ? &SSTSS_P1 : ((x) == 2) ? &SSTSS_P2 : ((x) == 3) ? &SSTSS_P3 : NULL))
-#define SSACD_P(x) (*(((x) == 1) ? &SSACD_P1 : ((x) == 2) ? &SSACD_P2 : ((x) == 3) ? &SSACD_P3 : NULL))
-
 /*
  * MultiMediaCard (MMC) controller - see drivers/mmc/host/pxamci.h
  */
diff --git a/include/asm-arm/arch-pxa/regs-ssp.h b/include/asm-arm/arch-pxa/regs-ssp.h
new file mode 100644
index 000000000000..687ade109113
--- /dev/null
+++ b/include/asm-arm/arch-pxa/regs-ssp.h
@@ -0,0 +1,173 @@
+#ifndef __ASM_ARCH_REGS_SSP_H
+#define __ASM_ARCH_REGS_SSP_H
+
+/*
+ * SSP Serial Port Registers
+ * PXA250, PXA255, PXA26x and PXA27x SSP controllers are all slightly different.
+ * PXA255, PXA26x and PXA27x have extra ports, registers and bits.
+ */
+
+ /* Common PXA2xx bits first */
+#define SSCR0_DSS	(0x0000000f)	/* Data Size Select (mask) */
+#define SSCR0_DataSize(x)  ((x) - 1)	/* Data Size Select [4..16] */
+#define SSCR0_FRF	(0x00000030)	/* FRame Format (mask) */
+#define SSCR0_Motorola	(0x0 << 4)	/* Motorola's Serial Peripheral Interface (SPI) */
+#define SSCR0_TI	(0x1 << 4)	/* Texas Instruments' Synchronous Serial Protocol (SSP) */
+#define SSCR0_National	(0x2 << 4)	/* National Microwire */
+#define SSCR0_ECS	(1 << 6)	/* External clock select */
+#define SSCR0_SSE	(1 << 7)	/* Synchronous Serial Port Enable */
+#if defined(CONFIG_PXA25x)
+#define SSCR0_SCR	(0x0000ff00)	/* Serial Clock Rate (mask) */
+#define SSCR0_SerClkDiv(x) ((((x) - 2)/2) << 8) /* Divisor [2..512] */
+#elif defined(CONFIG_PXA27x)
+#define SSCR0_SCR	(0x000fff00)	/* Serial Clock Rate (mask) */
+#define SSCR0_SerClkDiv(x) (((x) - 1) << 8) /* Divisor [1..4096] */
+#define SSCR0_EDSS	(1 << 20)	/* Extended data size select */
+#define SSCR0_NCS	(1 << 21)	/* Network clock select */
+#define SSCR0_RIM	(1 << 22)	/* Receive FIFO overrrun interrupt mask */
+#define SSCR0_TUM	(1 << 23)	/* Transmit FIFO underrun interrupt mask */
+#define SSCR0_FRDC	(0x07000000)	/* Frame rate divider control (mask) */
+#define SSCR0_SlotsPerFrm(x) (((x) - 1) << 24)	/* Time slots per frame [1..8] */
+#define SSCR0_ADC	(1 << 30)	/* Audio clock select */
+#define SSCR0_MOD	(1 << 31)	/* Mode (normal or network) */
+#endif
+
+#define SSCR1_RIE	(1 << 0)	/* Receive FIFO Interrupt Enable */
+#define SSCR1_TIE	(1 << 1)	/* Transmit FIFO Interrupt Enable */
+#define SSCR1_LBM	(1 << 2)	/* Loop-Back Mode */
+#define SSCR1_SPO	(1 << 3)	/* Motorola SPI SSPSCLK polarity setting */
+#define SSCR1_SPH	(1 << 4)	/* Motorola SPI SSPSCLK phase setting */
+#define SSCR1_MWDS	(1 << 5)	/* Microwire Transmit Data Size */
+#define SSCR1_TFT	(0x000003c0)	/* Transmit FIFO Threshold (mask) */
+#define SSCR1_TxTresh(x) (((x) - 1) << 6) /* level [1..16] */
+#define SSCR1_RFT	(0x00003c00)	/* Receive FIFO Threshold (mask) */
+#define SSCR1_RxTresh(x) (((x) - 1) << 10) /* level [1..16] */
+
+#define SSSR_TNF	(1 << 2)	/* Transmit FIFO Not Full */
+#define SSSR_RNE	(1 << 3)	/* Receive FIFO Not Empty */
+#define SSSR_BSY	(1 << 4)	/* SSP Busy */
+#define SSSR_TFS	(1 << 5)	/* Transmit FIFO Service Request */
+#define SSSR_RFS	(1 << 6)	/* Receive FIFO Service Request */
+#define SSSR_ROR	(1 << 7)	/* Receive FIFO Overrun */
+
+#define SSCR0_TIM		(1 << 23)	/* Transmit FIFO Under Run Interrupt Mask */
+#define SSCR0_RIM		(1 << 22)	/* Receive FIFO Over Run interrupt Mask */
+#define SSCR0_NCS		(1 << 21)	/* Network Clock Select */
+#define SSCR0_EDSS		(1 << 20)	/* Extended Data Size Select */
+
+/* extra bits in PXA255, PXA26x and PXA27x SSP ports */
+#define SSCR0_TISSP		(1 << 4)	/* TI Sync Serial Protocol */
+#define SSCR0_PSP		(3 << 4)	/* PSP - Programmable Serial Protocol */
+#define SSCR1_TTELP		(1 << 31)	/* TXD Tristate Enable Last Phase */
+#define SSCR1_TTE		(1 << 30)	/* TXD Tristate Enable */
+#define SSCR1_EBCEI		(1 << 29)	/* Enable Bit Count Error interrupt */
+#define SSCR1_SCFR		(1 << 28)	/* Slave Clock free Running */
+#define SSCR1_ECRA		(1 << 27)	/* Enable Clock Request A */
+#define SSCR1_ECRB		(1 << 26)	/* Enable Clock request B */
+#define SSCR1_SCLKDIR		(1 << 25)	/* Serial Bit Rate Clock Direction */
+#define SSCR1_SFRMDIR		(1 << 24)	/* Frame Direction */
+#define SSCR1_RWOT		(1 << 23)	/* Receive Without Transmit */
+#define SSCR1_TRAIL		(1 << 22)	/* Trailing Byte */
+#define SSCR1_TSRE		(1 << 21)	/* Transmit Service Request Enable */
+#define SSCR1_RSRE		(1 << 20)	/* Receive Service Request Enable */
+#define SSCR1_TINTE		(1 << 19)	/* Receiver Time-out Interrupt enable */
+#define SSCR1_PINTE		(1 << 18)	/* Peripheral Trailing Byte Interupt Enable */
+#define SSCR1_STRF		(1 << 15)	/* Select FIFO or EFWR */
+#define SSCR1_EFWR		(1 << 14)	/* Enable FIFO Write/Read */
+
+#define SSSR_BCE		(1 << 23)	/* Bit Count Error */
+#define SSSR_CSS		(1 << 22)	/* Clock Synchronisation Status */
+#define SSSR_TUR		(1 << 21)	/* Transmit FIFO Under Run */
+#define SSSR_EOC		(1 << 20)	/* End Of Chain */
+#define SSSR_TINT		(1 << 19)	/* Receiver Time-out Interrupt */
+#define SSSR_PINT		(1 << 18)	/* Peripheral Trailing Byte Interrupt */
+
+#define SSPSP_FSRT		(1 << 25)	/* Frame Sync Relative Timing */
+#define SSPSP_DMYSTOP(x)	((x) << 23)	/* Dummy Stop */
+#define SSPSP_SFRMWDTH(x)	((x) << 16)	/* Serial Frame Width */
+#define SSPSP_SFRMDLY(x)	((x) << 9)	/* Serial Frame Delay */
+#define SSPSP_DMYSTRT(x)	((x) << 7)	/* Dummy Start */
+#define SSPSP_STRTDLY(x)	((x) << 4)	/* Start Delay */
+#define SSPSP_ETDS		(1 << 3)	/* End of Transfer data State */
+#define SSPSP_SFRMP		(1 << 2)	/* Serial Frame Polarity */
+#define SSPSP_SCMODE(x)		((x) << 0)	/* Serial Bit Rate Clock Mode */
+
+#define SSACD_SCDB		(1 << 3)	/* SSPSYSCLK Divider Bypass */
+#define SSACD_ACPS(x)		((x) << 4)	/* Audio clock PLL select */
+#define SSACD_ACDS(x)		((x) << 0)	/* Audio clock divider select */
+
+#define SSCR0_P1	__REG(0x41000000)  /* SSP Port 1 Control Register 0 */
+#define SSCR1_P1	__REG(0x41000004)  /* SSP Port 1 Control Register 1 */
+#define SSSR_P1		__REG(0x41000008)  /* SSP Port 1 Status Register */
+#define SSITR_P1	__REG(0x4100000C)  /* SSP Port 1 Interrupt Test Register */
+#define SSDR_P1		__REG(0x41000010)  /* (Write / Read) SSP Port 1 Data Write Register/SSP Data Read Register */
+
+/* Support existing PXA25x drivers */
+#define SSCR0		SSCR0_P1  /* SSP Control Register 0 */
+#define SSCR1		SSCR1_P1  /* SSP Control Register 1 */
+#define SSSR		SSSR_P1	  /* SSP Status Register */
+#define SSITR		SSITR_P1  /* SSP Interrupt Test Register */
+#define SSDR		SSDR_P1	  /* (Write / Read) SSP Data Write Register/SSP Data Read Register */
+
+/* PXA27x ports */
+#if defined (CONFIG_PXA27x)
+#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
+#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
+#define SSTSA_P1	__REG(0x41000030)  /* SSP Port 1 Tx Timeslot Active */
+#define SSRSA_P1	__REG(0x41000034)  /* SSP Port 1 Rx Timeslot Active */
+#define SSTSS_P1	__REG(0x41000038)  /* SSP Port 1 Timeslot Status */
+#define SSACD_P1	__REG(0x4100003C)  /* SSP Port 1 Audio Clock Divider */
+#define SSCR0_P2	__REG(0x41700000)  /* SSP Port 2 Control Register 0 */
+#define SSCR1_P2	__REG(0x41700004)  /* SSP Port 2 Control Register 1 */
+#define SSSR_P2		__REG(0x41700008)  /* SSP Port 2 Status Register */
+#define SSITR_P2	__REG(0x4170000C)  /* SSP Port 2 Interrupt Test Register */
+#define SSDR_P2		__REG(0x41700010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
+#define SSTO_P2		__REG(0x41700028)  /* SSP Port 2 Time Out Register */
+#define SSPSP_P2	__REG(0x4170002C)  /* SSP Port 2 Programmable Serial Protocol */
+#define SSTSA_P2	__REG(0x41700030)  /* SSP Port 2 Tx Timeslot Active */
+#define SSRSA_P2	__REG(0x41700034)  /* SSP Port 2 Rx Timeslot Active */
+#define SSTSS_P2	__REG(0x41700038)  /* SSP Port 2 Timeslot Status */
+#define SSACD_P2	__REG(0x4170003C)  /* SSP Port 2 Audio Clock Divider */
+#define SSCR0_P3	__REG(0x41900000)  /* SSP Port 3 Control Register 0 */
+#define SSCR1_P3	__REG(0x41900004)  /* SSP Port 3 Control Register 1 */
+#define SSSR_P3		__REG(0x41900008)  /* SSP Port 3 Status Register */
+#define SSITR_P3	__REG(0x4190000C)  /* SSP Port 3 Interrupt Test Register */
+#define SSDR_P3		__REG(0x41900010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
+#define SSTO_P3		__REG(0x41900028)  /* SSP Port 3 Time Out Register */
+#define SSPSP_P3	__REG(0x4190002C)  /* SSP Port 3 Programmable Serial Protocol */
+#define SSTSA_P3	__REG(0x41900030)  /* SSP Port 3 Tx Timeslot Active */
+#define SSRSA_P3	__REG(0x41900034)  /* SSP Port 3 Rx Timeslot Active */
+#define SSTSS_P3	__REG(0x41900038)  /* SSP Port 3 Timeslot Status */
+#define SSACD_P3	__REG(0x4190003C)  /* SSP Port 3 Audio Clock Divider */
+#else /* PXA255 (only port 2) and PXA26x ports*/
+#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
+#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
+#define SSCR0_P2	__REG(0x41400000)  /* SSP Port 2 Control Register 0 */
+#define SSCR1_P2	__REG(0x41400004)  /* SSP Port 2 Control Register 1 */
+#define SSSR_P2		__REG(0x41400008)  /* SSP Port 2 Status Register */
+#define SSITR_P2	__REG(0x4140000C)  /* SSP Port 2 Interrupt Test Register */
+#define SSDR_P2		__REG(0x41400010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
+#define SSTO_P2		__REG(0x41400028)  /* SSP Port 2 Time Out Register */
+#define SSPSP_P2	__REG(0x4140002C)  /* SSP Port 2 Programmable Serial Protocol */
+#define SSCR0_P3	__REG(0x41500000)  /* SSP Port 3 Control Register 0 */
+#define SSCR1_P3	__REG(0x41500004)  /* SSP Port 3 Control Register 1 */
+#define SSSR_P3		__REG(0x41500008)  /* SSP Port 3 Status Register */
+#define SSITR_P3	__REG(0x4150000C)  /* SSP Port 3 Interrupt Test Register */
+#define SSDR_P3		__REG(0x41500010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
+#define SSTO_P3		__REG(0x41500028)  /* SSP Port 3 Time Out Register */
+#define SSPSP_P3	__REG(0x4150002C)  /* SSP Port 3 Programmable Serial Protocol */
+#endif
+
+#define SSCR0_P(x) (*(((x) == 1) ? &SSCR0_P1 : ((x) == 2) ? &SSCR0_P2 : ((x) == 3) ? &SSCR0_P3 : NULL))
+#define SSCR1_P(x) (*(((x) == 1) ? &SSCR1_P1 : ((x) == 2) ? &SSCR1_P2 : ((x) == 3) ? &SSCR1_P3 : NULL))
+#define SSSR_P(x) (*(((x) == 1) ? &SSSR_P1 : ((x) == 2) ? &SSSR_P2 : ((x) == 3) ? &SSSR_P3 : NULL))
+#define SSITR_P(x) (*(((x) == 1) ? &SSITR_P1 : ((x) == 2) ? &SSITR_P2 : ((x) == 3) ? &SSITR_P3 : NULL))
+#define SSDR_P(x) (*(((x) == 1) ? &SSDR_P1 : ((x) == 2) ? &SSDR_P2 : ((x) == 3) ? &SSDR_P3 : NULL))
+#define SSTO_P(x) (*(((x) == 1) ? &SSTO_P1 : ((x) == 2) ? &SSTO_P2 : ((x) == 3) ? &SSTO_P3 : NULL))
+#define SSPSP_P(x) (*(((x) == 1) ? &SSPSP_P1 : ((x) == 2) ? &SSPSP_P2 : ((x) == 3) ? &SSPSP_P3 : NULL))
+#define SSTSA_P(x) (*(((x) == 1) ? &SSTSA_P1 : ((x) == 2) ? &SSTSA_P2 : ((x) == 3) ? &SSTSA_P3 : NULL))
+#define SSRSA_P(x) (*(((x) == 1) ? &SSRSA_P1 : ((x) == 2) ? &SSRSA_P2 : ((x) == 3) ? &SSRSA_P3 : NULL))
+#define SSTSS_P(x) (*(((x) == 1) ? &SSTSS_P1 : ((x) == 2) ? &SSTSS_P2 : ((x) == 3) ? &SSTSS_P3 : NULL))
+#define SSACD_P(x) (*(((x) == 1) ? &SSACD_P1 : ((x) == 2) ? &SSACD_P2 : ((x) == 3) ? &SSACD_P3 : NULL))
+
+#endif /* __ASM_ARCH_REGS_SSP_H */
-- 
cgit v1.2.3


From 3dcb00ea58f6b5dc62b89bbfd54353a06e6af921 Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Fri, 30 Nov 2007 18:26:56 +0800
Subject: [ARM] pxa: use __raw_writel()/__raw_readl() for ssp_xxxx()

1. change SSP register definitions from absolute virtual addresses to
   offsets

2. use __raw_writel()/__raw_readl() for functions of ssp_xxxx()

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/ssp.c             | 77 ++++++++++++++++++++------------
 include/asm-arm/arch-pxa/regs-ssp.h | 89 ++++++-------------------------------
 2 files changed, 63 insertions(+), 103 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/ssp.c b/arch/arm/mach-pxa/ssp.c
index 0225ee016f2d..b6d37fa7b85f 100644
--- a/arch/arm/mach-pxa/ssp.c
+++ b/arch/arm/mach-pxa/ssp.c
@@ -48,9 +48,11 @@
 static irqreturn_t ssp_interrupt(int irq, void *dev_id)
 {
 	struct ssp_dev *dev = (struct ssp_dev*) dev_id;
-	unsigned int status = SSSR_P(dev->port);
+	struct ssp_device *ssp = dev->ssp;
+	unsigned int status;
 
-	SSSR_P(dev->port) = status; /* clear status bits */
+	status = __raw_readl(ssp->mmio_base + SSSR);
+	__raw_writel(status, ssp->mmio_base + SSSR);
 
 	if (status & SSSR_ROR)
 		printk(KERN_WARNING "SSP(%d): receiver overrun\n", dev->port);
@@ -79,15 +81,16 @@ static irqreturn_t ssp_interrupt(int irq, void *dev_id)
  */
 int ssp_write_word(struct ssp_dev *dev, u32 data)
 {
+	struct ssp_device *ssp = dev->ssp;
 	int timeout = TIMEOUT;
 
-	while (!(SSSR_P(dev->port) & SSSR_TNF)) {
+	while (!(__raw_readl(ssp->mmio_base + SSSR) & SSSR_TNF)) {
 	        if (!--timeout)
 	        	return -ETIMEDOUT;
 		cpu_relax();
 	}
 
-	SSDR_P(dev->port) = data;
+	__raw_writel(data, ssp->mmio_base + SSDR);
 
 	return 0;
 }
@@ -109,15 +112,16 @@ int ssp_write_word(struct ssp_dev *dev, u32 data)
  */
 int ssp_read_word(struct ssp_dev *dev, u32 *data)
 {
+	struct ssp_device *ssp = dev->ssp;
 	int timeout = TIMEOUT;
 
-	while (!(SSSR_P(dev->port) & SSSR_RNE)) {
+	while (!(__raw_readl(ssp->mmio_base + SSSR) & SSSR_RNE)) {
 	        if (!--timeout)
 	        	return -ETIMEDOUT;
 		cpu_relax();
 	}
 
-	*data = SSDR_P(dev->port);
+	*data = __raw_readl(ssp->mmio_base + SSDR);
 	return 0;
 }
 
@@ -131,17 +135,18 @@ int ssp_read_word(struct ssp_dev *dev, u32 *data)
  */
 int ssp_flush(struct ssp_dev *dev)
 {
+	struct ssp_device *ssp = dev->ssp;
 	int timeout = TIMEOUT * 2;
 
 	do {
-		while (SSSR_P(dev->port) & SSSR_RNE) {
+		while (__raw_readl(ssp->mmio_base + SSSR) & SSSR_RNE) {
 		        if (!--timeout)
 		        	return -ETIMEDOUT;
-			(void) SSDR_P(dev->port);
+			(void)__raw_readl(ssp->mmio_base + SSDR);
 		}
 	        if (!--timeout)
 	        	return -ETIMEDOUT;
-	} while (SSSR_P(dev->port) & SSSR_BSY);
+	} while (__raw_readl(ssp->mmio_base + SSSR) & SSSR_BSY);
 
 	return 0;
 }
@@ -153,7 +158,12 @@ int ssp_flush(struct ssp_dev *dev)
  */
 void ssp_enable(struct ssp_dev *dev)
 {
-	SSCR0_P(dev->port) |= SSCR0_SSE;
+	struct ssp_device *ssp = dev->ssp;
+	uint32_t sscr0;
+
+	sscr0 = __raw_readl(ssp->mmio_base + SSCR0);
+	sscr0 |= SSCR0_SSE;
+	__raw_writel(sscr0, ssp->mmio_base + SSCR0);
 }
 
 /**
@@ -163,7 +173,12 @@ void ssp_enable(struct ssp_dev *dev)
  */
 void ssp_disable(struct ssp_dev *dev)
 {
-	SSCR0_P(dev->port) &= ~SSCR0_SSE;
+	struct ssp_device *ssp = dev->ssp;
+	uint32_t sscr0;
+
+	sscr0 = __raw_readl(ssp->mmio_base + SSCR0);
+	sscr0 &= ~SSCR0_SSE;
+	__raw_writel(sscr0, ssp->mmio_base + SSCR0);
 }
 
 /**
@@ -172,14 +187,16 @@ void ssp_disable(struct ssp_dev *dev)
  *
  * Save the configured SSP state for suspend.
  */
-void ssp_save_state(struct ssp_dev *dev, struct ssp_state *ssp)
+void ssp_save_state(struct ssp_dev *dev, struct ssp_state *state)
 {
-	ssp->cr0 = SSCR0_P(dev->port);
-	ssp->cr1 = SSCR1_P(dev->port);
-	ssp->to = SSTO_P(dev->port);
-	ssp->psp = SSPSP_P(dev->port);
+	struct ssp_device *ssp = dev->ssp;
+
+	state->cr0 = __raw_readl(ssp->mmio_base + SSCR0);
+	state->cr1 = __raw_readl(ssp->mmio_base + SSCR1);
+	state->to  = __raw_readl(ssp->mmio_base + SSTO);
+	state->psp = __raw_readl(ssp->mmio_base + SSPSP);
 
-	SSCR0_P(dev->port) &= ~SSCR0_SSE;
+	ssp_disable(dev);
 }
 
 /**
@@ -188,16 +205,18 @@ void ssp_save_state(struct ssp_dev *dev, struct ssp_state *ssp)
  *
  * Restore the SSP configuration saved previously by ssp_save_state.
  */
-void ssp_restore_state(struct ssp_dev *dev, struct ssp_state *ssp)
+void ssp_restore_state(struct ssp_dev *dev, struct ssp_state *state)
 {
-	SSSR_P(dev->port) = SSSR_ROR | SSSR_TUR | SSSR_BCE;
+	struct ssp_device *ssp = dev->ssp;
+	uint32_t sssr = SSSR_ROR | SSSR_TUR | SSSR_BCE;
 
-	SSCR0_P(dev->port) = ssp->cr0 & ~SSCR0_SSE;
-	SSCR1_P(dev->port) = ssp->cr1;
-	SSTO_P(dev->port) = ssp->to;
-	SSPSP_P(dev->port) = ssp->psp;
+	__raw_writel(sssr, ssp->mmio_base + SSSR);
 
-	SSCR0_P(dev->port) = ssp->cr0;
+	__raw_writel(state->cr0 & ~SSCR0_SSE, ssp->mmio_base + SSCR0);
+	__raw_writel(state->cr1, ssp->mmio_base + SSCR1);
+	__raw_writel(state->to,  ssp->mmio_base + SSTO);
+	__raw_writel(state->psp, ssp->mmio_base + SSPSP);
+	__raw_writel(state->cr0, ssp->mmio_base + SSCR0);
 }
 
 /**
@@ -211,15 +230,17 @@ void ssp_restore_state(struct ssp_dev *dev, struct ssp_state *ssp)
  */
 int ssp_config(struct ssp_dev *dev, u32 mode, u32 flags, u32 psp_flags, u32 speed)
 {
+	struct ssp_device *ssp = dev->ssp;
+
 	dev->mode = mode;
 	dev->flags = flags;
 	dev->psp_flags = psp_flags;
 	dev->speed = speed;
 
 	/* set up port type, speed, port settings */
-	SSCR0_P(dev->port) = (dev->speed | dev->mode);
-	SSCR1_P(dev->port) = dev->flags;
-	SSPSP_P(dev->port) = dev->psp_flags;
+	__raw_writel((dev->speed | dev->mode), ssp->mmio_base + SSCR0);
+	__raw_writel(dev->flags, ssp->mmio_base + SSCR1);
+	__raw_writel(dev->psp_flags, ssp->mmio_base + SSPSP);
 
 	return 0;
 }
@@ -274,7 +295,7 @@ void ssp_exit(struct ssp_dev *dev)
 {
 	struct ssp_device *ssp = dev->ssp;
 
-	SSCR0_P(dev->port) &= ~SSCR0_SSE;
+	ssp_disable(dev);
 	free_irq(dev->irq, dev);
 	clk_disable(ssp->clk);
 	ssp_free(ssp);
diff --git a/include/asm-arm/arch-pxa/regs-ssp.h b/include/asm-arm/arch-pxa/regs-ssp.h
index 687ade109113..991cb688db75 100644
--- a/include/asm-arm/arch-pxa/regs-ssp.h
+++ b/include/asm-arm/arch-pxa/regs-ssp.h
@@ -7,7 +7,20 @@
  * PXA255, PXA26x and PXA27x have extra ports, registers and bits.
  */
 
- /* Common PXA2xx bits first */
+#define SSCR0		(0x00)  /* SSP Control Register 0 */
+#define SSCR1		(0x04)  /* SSP Control Register 1 */
+#define SSSR		(0x08)  /* SSP Status Register */
+#define SSITR		(0x0C)  /* SSP Interrupt Test Register */
+#define SSDR		(0x10)  /* SSP Data Write/Data Read Register */
+
+#define SSTO		(0x28)  /* SSP Time Out Register */
+#define SSPSP		(0x2C)  /* SSP Programmable Serial Protocol */
+#define SSTSA		(0x30)  /* SSP Tx Timeslot Active */
+#define SSRSA		(0x34)  /* SSP Rx Timeslot Active */
+#define SSTSS		(0x38)  /* SSP Timeslot Status */
+#define SSACD		(0x3C)  /* SSP Audio Clock Divider */
+
+/* Common PXA2xx bits first */
 #define SSCR0_DSS	(0x0000000f)	/* Data Size Select (mask) */
 #define SSCR0_DataSize(x)  ((x) - 1)	/* Data Size Select [4..16] */
 #define SSCR0_FRF	(0x00000030)	/* FRame Format (mask) */
@@ -96,78 +109,4 @@
 #define SSACD_ACPS(x)		((x) << 4)	/* Audio clock PLL select */
 #define SSACD_ACDS(x)		((x) << 0)	/* Audio clock divider select */
 
-#define SSCR0_P1	__REG(0x41000000)  /* SSP Port 1 Control Register 0 */
-#define SSCR1_P1	__REG(0x41000004)  /* SSP Port 1 Control Register 1 */
-#define SSSR_P1		__REG(0x41000008)  /* SSP Port 1 Status Register */
-#define SSITR_P1	__REG(0x4100000C)  /* SSP Port 1 Interrupt Test Register */
-#define SSDR_P1		__REG(0x41000010)  /* (Write / Read) SSP Port 1 Data Write Register/SSP Data Read Register */
-
-/* Support existing PXA25x drivers */
-#define SSCR0		SSCR0_P1  /* SSP Control Register 0 */
-#define SSCR1		SSCR1_P1  /* SSP Control Register 1 */
-#define SSSR		SSSR_P1	  /* SSP Status Register */
-#define SSITR		SSITR_P1  /* SSP Interrupt Test Register */
-#define SSDR		SSDR_P1	  /* (Write / Read) SSP Data Write Register/SSP Data Read Register */
-
-/* PXA27x ports */
-#if defined (CONFIG_PXA27x)
-#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
-#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
-#define SSTSA_P1	__REG(0x41000030)  /* SSP Port 1 Tx Timeslot Active */
-#define SSRSA_P1	__REG(0x41000034)  /* SSP Port 1 Rx Timeslot Active */
-#define SSTSS_P1	__REG(0x41000038)  /* SSP Port 1 Timeslot Status */
-#define SSACD_P1	__REG(0x4100003C)  /* SSP Port 1 Audio Clock Divider */
-#define SSCR0_P2	__REG(0x41700000)  /* SSP Port 2 Control Register 0 */
-#define SSCR1_P2	__REG(0x41700004)  /* SSP Port 2 Control Register 1 */
-#define SSSR_P2		__REG(0x41700008)  /* SSP Port 2 Status Register */
-#define SSITR_P2	__REG(0x4170000C)  /* SSP Port 2 Interrupt Test Register */
-#define SSDR_P2		__REG(0x41700010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
-#define SSTO_P2		__REG(0x41700028)  /* SSP Port 2 Time Out Register */
-#define SSPSP_P2	__REG(0x4170002C)  /* SSP Port 2 Programmable Serial Protocol */
-#define SSTSA_P2	__REG(0x41700030)  /* SSP Port 2 Tx Timeslot Active */
-#define SSRSA_P2	__REG(0x41700034)  /* SSP Port 2 Rx Timeslot Active */
-#define SSTSS_P2	__REG(0x41700038)  /* SSP Port 2 Timeslot Status */
-#define SSACD_P2	__REG(0x4170003C)  /* SSP Port 2 Audio Clock Divider */
-#define SSCR0_P3	__REG(0x41900000)  /* SSP Port 3 Control Register 0 */
-#define SSCR1_P3	__REG(0x41900004)  /* SSP Port 3 Control Register 1 */
-#define SSSR_P3		__REG(0x41900008)  /* SSP Port 3 Status Register */
-#define SSITR_P3	__REG(0x4190000C)  /* SSP Port 3 Interrupt Test Register */
-#define SSDR_P3		__REG(0x41900010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
-#define SSTO_P3		__REG(0x41900028)  /* SSP Port 3 Time Out Register */
-#define SSPSP_P3	__REG(0x4190002C)  /* SSP Port 3 Programmable Serial Protocol */
-#define SSTSA_P3	__REG(0x41900030)  /* SSP Port 3 Tx Timeslot Active */
-#define SSRSA_P3	__REG(0x41900034)  /* SSP Port 3 Rx Timeslot Active */
-#define SSTSS_P3	__REG(0x41900038)  /* SSP Port 3 Timeslot Status */
-#define SSACD_P3	__REG(0x4190003C)  /* SSP Port 3 Audio Clock Divider */
-#else /* PXA255 (only port 2) and PXA26x ports*/
-#define SSTO_P1		__REG(0x41000028)  /* SSP Port 1 Time Out Register */
-#define SSPSP_P1	__REG(0x4100002C)  /* SSP Port 1 Programmable Serial Protocol */
-#define SSCR0_P2	__REG(0x41400000)  /* SSP Port 2 Control Register 0 */
-#define SSCR1_P2	__REG(0x41400004)  /* SSP Port 2 Control Register 1 */
-#define SSSR_P2		__REG(0x41400008)  /* SSP Port 2 Status Register */
-#define SSITR_P2	__REG(0x4140000C)  /* SSP Port 2 Interrupt Test Register */
-#define SSDR_P2		__REG(0x41400010)  /* (Write / Read) SSP Port 2 Data Write Register/SSP Data Read Register */
-#define SSTO_P2		__REG(0x41400028)  /* SSP Port 2 Time Out Register */
-#define SSPSP_P2	__REG(0x4140002C)  /* SSP Port 2 Programmable Serial Protocol */
-#define SSCR0_P3	__REG(0x41500000)  /* SSP Port 3 Control Register 0 */
-#define SSCR1_P3	__REG(0x41500004)  /* SSP Port 3 Control Register 1 */
-#define SSSR_P3		__REG(0x41500008)  /* SSP Port 3 Status Register */
-#define SSITR_P3	__REG(0x4150000C)  /* SSP Port 3 Interrupt Test Register */
-#define SSDR_P3		__REG(0x41500010)  /* (Write / Read) SSP Port 3 Data Write Register/SSP Data Read Register */
-#define SSTO_P3		__REG(0x41500028)  /* SSP Port 3 Time Out Register */
-#define SSPSP_P3	__REG(0x4150002C)  /* SSP Port 3 Programmable Serial Protocol */
-#endif
-
-#define SSCR0_P(x) (*(((x) == 1) ? &SSCR0_P1 : ((x) == 2) ? &SSCR0_P2 : ((x) == 3) ? &SSCR0_P3 : NULL))
-#define SSCR1_P(x) (*(((x) == 1) ? &SSCR1_P1 : ((x) == 2) ? &SSCR1_P2 : ((x) == 3) ? &SSCR1_P3 : NULL))
-#define SSSR_P(x) (*(((x) == 1) ? &SSSR_P1 : ((x) == 2) ? &SSSR_P2 : ((x) == 3) ? &SSSR_P3 : NULL))
-#define SSITR_P(x) (*(((x) == 1) ? &SSITR_P1 : ((x) == 2) ? &SSITR_P2 : ((x) == 3) ? &SSITR_P3 : NULL))
-#define SSDR_P(x) (*(((x) == 1) ? &SSDR_P1 : ((x) == 2) ? &SSDR_P2 : ((x) == 3) ? &SSDR_P3 : NULL))
-#define SSTO_P(x) (*(((x) == 1) ? &SSTO_P1 : ((x) == 2) ? &SSTO_P2 : ((x) == 3) ? &SSTO_P3 : NULL))
-#define SSPSP_P(x) (*(((x) == 1) ? &SSPSP_P1 : ((x) == 2) ? &SSPSP_P2 : ((x) == 3) ? &SSPSP_P3 : NULL))
-#define SSTSA_P(x) (*(((x) == 1) ? &SSTSA_P1 : ((x) == 2) ? &SSTSA_P2 : ((x) == 3) ? &SSTSA_P3 : NULL))
-#define SSRSA_P(x) (*(((x) == 1) ? &SSRSA_P1 : ((x) == 2) ? &SSRSA_P2 : ((x) == 3) ? &SSRSA_P3 : NULL))
-#define SSTSS_P(x) (*(((x) == 1) ? &SSTSS_P1 : ((x) == 2) ? &SSTSS_P2 : ((x) == 3) ? &SSTSS_P3 : NULL))
-#define SSACD_P(x) (*(((x) == 1) ? &SSACD_P1 : ((x) == 2) ? &SSACD_P2 : ((x) == 3) ? &SSACD_P3 : NULL))
-
 #endif /* __ASM_ARCH_REGS_SSP_H */
-- 
cgit v1.2.3


From 2f1a74e5a2de0459139b85af95e901448726c375 Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Wed, 21 Nov 2007 18:50:53 +0800
Subject: [ARM] pxa: make pxa2xx_spi driver use ssp_request()/ssp_free()

1. make pxa2xx_spi.c use ssp_request() and ssp_free() to get the common
   information of the designated SSP port.

2. remove those IRQ/memory request code, ssp_request() has done that for
   the driver

3. the SPI platform device is thus made psuedo, no resource (memory/IRQ)
   has to be defined, all will be retreived by ssp_request()

4. introduce ssp_get_clk_div() to handle controller difference in clock
   divisor setting

5. use clk_xxx() API for clock enable/disable, and clk_get_rate() to
   handle the different SSP clock frequency between different processors

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/lubbock.c           |  17 -----
 drivers/spi/Kconfig                   |   1 +
 drivers/spi/pxa2xx_spi.c              | 138 +++++++++++++---------------------
 include/asm-arm/arch-pxa/pxa2xx_spi.h |  24 ------
 4 files changed, 53 insertions(+), 127 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c
index 1d3112dc629e..ebb73f133486 100644
--- a/arch/arm/mach-pxa/lubbock.c
+++ b/arch/arm/mach-pxa/lubbock.c
@@ -206,30 +206,13 @@ static struct resource smc91x_resources[] = {
  * (to J5) and poking board registers (as done below).  Else it's only useful
  * for the temperature sensors.
  */
-static struct resource pxa_ssp_resources[] = {
-	[0] = {
-		.start	= __PREG(SSCR0_P(1)),
-		.end	= __PREG(SSCR0_P(1)) + 0x14,
-		.flags	= IORESOURCE_MEM,
-	},
-	[1] = {
-		.start	= IRQ_SSP,
-		.end	= IRQ_SSP,
-		.flags	= IORESOURCE_IRQ,
-	},
-};
-
 static struct pxa2xx_spi_master pxa_ssp_master_info = {
-	.ssp_type	= PXA25x_SSP,
-	.clock_enable	= CKEN_SSP,
 	.num_chipselect	= 0,
 };
 
 static struct platform_device pxa_ssp = {
 	.name		= "pxa2xx-spi",
 	.id		= 1,
-	.resource	= pxa_ssp_resources,
-	.num_resources	= ARRAY_SIZE(pxa_ssp_resources),
 	.dev = {
 		.platform_data	= &pxa_ssp_master_info,
 	},
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index abf05048c638..aaaea81e412a 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -153,6 +153,7 @@ config SPI_OMAP24XX
 config SPI_PXA2XX
 	tristate "PXA2xx SSP SPI master"
 	depends on SPI_MASTER && ARCH_PXA && EXPERIMENTAL
+	select PXA_SSP
 	help
 	  This enables using a PXA2xx SSP port as a SPI master controller.
 	  The driver can be configured to use any SSP port and additional
diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index 0ac0f65f69a1..eb817b8eb024 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -27,6 +27,7 @@
 #include <linux/spi/spi.h>
 #include <linux/workqueue.h>
 #include <linux/delay.h>
+#include <linux/clk.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
@@ -37,6 +38,7 @@
 #include <asm/arch/hardware.h>
 #include <asm/arch/pxa-regs.h>
 #include <asm/arch/regs-ssp.h>
+#include <asm/arch/ssp.h>
 #include <asm/arch/pxa2xx_spi.h>
 
 MODULE_AUTHOR("Stephen Street");
@@ -81,6 +83,9 @@ struct driver_data {
 	/* Driver model hookup */
 	struct platform_device *pdev;
 
+	/* SSP Info */
+	struct ssp_device *ssp;
+
 	/* SPI framework hookup */
 	enum pxa_ssp_type ssp_type;
 	struct spi_master *master;
@@ -779,6 +784,16 @@ int set_dma_burst_and_threshold(struct chip_data *chip, struct spi_device *spi,
 	return retval;
 }
 
+static unsigned int ssp_get_clk_div(struct ssp_device *ssp, int rate)
+{
+	unsigned long ssp_clk = clk_get_rate(ssp->clk);
+
+	if (ssp->type == PXA25x_SSP)
+		return ((ssp_clk / (2 * rate) - 1) & 0xff) << 8;
+	else
+		return ((ssp_clk / rate - 1) & 0xfff) << 8;
+}
+
 static void pump_transfers(unsigned long data)
 {
 	struct driver_data *drv_data = (struct driver_data *)data;
@@ -786,6 +801,7 @@ static void pump_transfers(unsigned long data)
 	struct spi_transfer *transfer = NULL;
 	struct spi_transfer *previous = NULL;
 	struct chip_data *chip = NULL;
+	struct ssp_device *ssp = drv_data->ssp;
 	void *reg = drv_data->ioaddr;
 	u32 clk_div = 0;
 	u8 bits = 0;
@@ -867,12 +883,7 @@ static void pump_transfers(unsigned long data)
 		if (transfer->bits_per_word)
 			bits = transfer->bits_per_word;
 
-		if (reg == SSP1_VIRT)
-			clk_div = SSP1_SerClkDiv(speed);
-		else if (reg == SSP2_VIRT)
-			clk_div = SSP2_SerClkDiv(speed);
-		else if (reg == SSP3_VIRT)
-			clk_div = SSP3_SerClkDiv(speed);
+		clk_div = ssp_get_clk_div(ssp, speed);
 
 		if (bits <= 8) {
 			drv_data->n_bytes = 1;
@@ -1075,6 +1086,7 @@ static int setup(struct spi_device *spi)
 	struct pxa2xx_spi_chip *chip_info = NULL;
 	struct chip_data *chip;
 	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
+	struct ssp_device *ssp = drv_data->ssp;
 	unsigned int clk_div;
 
 	if (!spi->bits_per_word)
@@ -1158,18 +1170,7 @@ static int setup(struct spi_device *spi)
 		}
 	}
 
-	if (drv_data->ioaddr == SSP1_VIRT)
-		clk_div = SSP1_SerClkDiv(spi->max_speed_hz);
-	else if (drv_data->ioaddr == SSP2_VIRT)
-		clk_div = SSP2_SerClkDiv(spi->max_speed_hz);
-	else if (drv_data->ioaddr == SSP3_VIRT)
-		clk_div = SSP3_SerClkDiv(spi->max_speed_hz);
-	else
-	{
-		dev_err(&spi->dev, "failed setup: unknown IO address=0x%p\n",
-			drv_data->ioaddr);
-		return -ENODEV;
-	}
+	clk_div = ssp_get_clk_div(ssp, spi->max_speed_hz);
 	chip->speed_hz = spi->max_speed_hz;
 
 	chip->cr0 = clk_div
@@ -1184,15 +1185,15 @@ static int setup(struct spi_device *spi)
 
 	/* NOTE:  PXA25x_SSP _could_ use external clocking ... */
 	if (drv_data->ssp_type != PXA25x_SSP)
-		dev_dbg(&spi->dev, "%d bits/word, %d Hz, mode %d\n",
+		dev_dbg(&spi->dev, "%d bits/word, %ld Hz, mode %d\n",
 				spi->bits_per_word,
-				(CLOCK_SPEED_HZ)
+				clk_get_rate(ssp->clk)
 					/ (1 + ((chip->cr0 & SSCR0_SCR) >> 8)),
 				spi->mode & 0x3);
 	else
-		dev_dbg(&spi->dev, "%d bits/word, %d Hz, mode %d\n",
+		dev_dbg(&spi->dev, "%d bits/word, %ld Hz, mode %d\n",
 				spi->bits_per_word,
-				(CLOCK_SPEED_HZ/2)
+				clk_get_rate(ssp->clk)
 					/ (1 + ((chip->cr0 & SSCR0_SCR) >> 8)),
 				spi->mode & 0x3);
 
@@ -1324,14 +1325,14 @@ static int __init pxa2xx_spi_probe(struct platform_device *pdev)
 	struct pxa2xx_spi_master *platform_info;
 	struct spi_master *master;
 	struct driver_data *drv_data = 0;
-	struct resource *memory_resource;
-	int irq;
+	struct ssp_device *ssp;
 	int status = 0;
 
 	platform_info = dev->platform_data;
 
-	if (platform_info->ssp_type == SSP_UNDEFINED) {
-		dev_err(&pdev->dev, "undefined SSP\n");
+	ssp = ssp_request(pdev->id, pdev->name);
+	if (ssp == NULL) {
+		dev_err(&pdev->dev, "failed to request SSP%d\n", pdev->id);
 		return -ENODEV;
 	}
 
@@ -1339,12 +1340,14 @@ static int __init pxa2xx_spi_probe(struct platform_device *pdev)
 	master = spi_alloc_master(dev, sizeof(struct driver_data) + 16);
 	if (!master) {
 		dev_err(&pdev->dev, "can not alloc spi_master\n");
+		ssp_free(ssp);
 		return -ENOMEM;
 	}
 	drv_data = spi_master_get_devdata(master);
 	drv_data->master = master;
 	drv_data->master_info = platform_info;
 	drv_data->pdev = pdev;
+	drv_data->ssp = ssp;
 
 	master->bus_num = pdev->id;
 	master->num_chipselect = platform_info->num_chipselect;
@@ -1352,21 +1355,13 @@ static int __init pxa2xx_spi_probe(struct platform_device *pdev)
 	master->setup = setup;
 	master->transfer = transfer;
 
-	drv_data->ssp_type = platform_info->ssp_type;
+	drv_data->ssp_type = ssp->type;
 	drv_data->null_dma_buf = (u32 *)ALIGN((u32)(drv_data +
 						sizeof(struct driver_data)), 8);
 
-	/* Setup register addresses */
-	memory_resource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!memory_resource) {
-		dev_err(&pdev->dev, "memory resources not defined\n");
-		status = -ENODEV;
-		goto out_error_master_alloc;
-	}
-
-	drv_data->ioaddr = (void *)io_p2v((unsigned long)(memory_resource->start));
-	drv_data->ssdr_physical = memory_resource->start + 0x00000010;
-	if (platform_info->ssp_type == PXA25x_SSP) {
+	drv_data->ioaddr = ssp->mmio_base;
+	drv_data->ssdr_physical = ssp->phys_base + SSDR;
+	if (ssp->type == PXA25x_SSP) {
 		drv_data->int_cr1 = SSCR1_TIE | SSCR1_RIE;
 		drv_data->dma_cr1 = 0;
 		drv_data->clear_sr = SSSR_ROR;
@@ -1378,15 +1373,7 @@ static int __init pxa2xx_spi_probe(struct platform_device *pdev)
 		drv_data->mask_sr = SSSR_TINT | SSSR_RFS | SSSR_TFS | SSSR_ROR;
 	}
 
-	/* Attach to IRQ */
-	irq = platform_get_irq(pdev, 0);
-	if (irq < 0) {
-		dev_err(&pdev->dev, "irq resource not defined\n");
-		status = -ENODEV;
-		goto out_error_master_alloc;
-	}
-
-	status = request_irq(irq, ssp_int, 0, dev->bus_id, drv_data);
+	status = request_irq(ssp->irq, ssp_int, 0, dev->bus_id, drv_data);
 	if (status < 0) {
 		dev_err(&pdev->dev, "can not get IRQ\n");
 		goto out_error_master_alloc;
@@ -1419,29 +1406,12 @@ static int __init pxa2xx_spi_probe(struct platform_device *pdev)
 			goto out_error_dma_alloc;
 		}
 
-		if (drv_data->ioaddr == SSP1_VIRT) {
-				DRCMRRXSSDR = DRCMR_MAPVLD
-						| drv_data->rx_channel;
-				DRCMRTXSSDR = DRCMR_MAPVLD
-						| drv_data->tx_channel;
-		} else if (drv_data->ioaddr == SSP2_VIRT) {
-				DRCMRRXSS2DR = DRCMR_MAPVLD
-						| drv_data->rx_channel;
-				DRCMRTXSS2DR = DRCMR_MAPVLD
-						| drv_data->tx_channel;
-		} else if (drv_data->ioaddr == SSP3_VIRT) {
-				DRCMRRXSS3DR = DRCMR_MAPVLD
-						| drv_data->rx_channel;
-				DRCMRTXSS3DR = DRCMR_MAPVLD
-						| drv_data->tx_channel;
-		} else {
-			dev_err(dev, "bad SSP type\n");
-			goto out_error_dma_alloc;
-		}
+		DRCMR(ssp->drcmr_rx) = DRCMR_MAPVLD | drv_data->rx_channel;
+		DRCMR(ssp->drcmr_tx) = DRCMR_MAPVLD | drv_data->tx_channel;
 	}
 
 	/* Enable SOC clock */
-	pxa_set_cken(platform_info->clock_enable, 1);
+	clk_enable(ssp->clk);
 
 	/* Load default SSP configuration */
 	write_SSCR0(0, drv_data->ioaddr);
@@ -1480,7 +1450,7 @@ out_error_queue_alloc:
 	destroy_queue(drv_data);
 
 out_error_clock_enabled:
-	pxa_set_cken(platform_info->clock_enable, 0);
+	clk_disable(ssp->clk);
 
 out_error_dma_alloc:
 	if (drv_data->tx_channel != -1)
@@ -1489,17 +1459,18 @@ out_error_dma_alloc:
 		pxa_free_dma(drv_data->rx_channel);
 
 out_error_irq_alloc:
-	free_irq(irq, drv_data);
+	free_irq(ssp->irq, drv_data);
 
 out_error_master_alloc:
 	spi_master_put(master);
+	ssp_free(ssp);
 	return status;
 }
 
 static int pxa2xx_spi_remove(struct platform_device *pdev)
 {
 	struct driver_data *drv_data = platform_get_drvdata(pdev);
-	int irq;
+	struct ssp_device *ssp = drv_data->ssp;
 	int status = 0;
 
 	if (!drv_data)
@@ -1521,28 +1492,21 @@ static int pxa2xx_spi_remove(struct platform_device *pdev)
 
 	/* Disable the SSP at the peripheral and SOC level */
 	write_SSCR0(0, drv_data->ioaddr);
-	pxa_set_cken(drv_data->master_info->clock_enable, 0);
+	clk_disable(ssp->clk);
 
 	/* Release DMA */
 	if (drv_data->master_info->enable_dma) {
-		if (drv_data->ioaddr == SSP1_VIRT) {
-			DRCMRRXSSDR = 0;
-			DRCMRTXSSDR = 0;
-		} else if (drv_data->ioaddr == SSP2_VIRT) {
-			DRCMRRXSS2DR = 0;
-			DRCMRTXSS2DR = 0;
-		} else if (drv_data->ioaddr == SSP3_VIRT) {
-			DRCMRRXSS3DR = 0;
-			DRCMRTXSS3DR = 0;
-		}
+		DRCMR(ssp->drcmr_rx) = 0;
+		DRCMR(ssp->drcmr_tx) = 0;
 		pxa_free_dma(drv_data->tx_channel);
 		pxa_free_dma(drv_data->rx_channel);
 	}
 
 	/* Release IRQ */
-	irq = platform_get_irq(pdev, 0);
-	if (irq >= 0)
-		free_irq(irq, drv_data);
+	free_irq(ssp->irq, drv_data);
+
+	/* Release SSP */
+	ssp_free(ssp);
 
 	/* Disconnect from the SPI framework */
 	spi_unregister_master(drv_data->master);
@@ -1577,6 +1541,7 @@ static int suspend_devices(struct device *dev, void *pm_message)
 static int pxa2xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	struct ssp_device *ssp = drv_data->ssp;
 	int status = 0;
 
 	/* Check all childern for current power state */
@@ -1589,7 +1554,7 @@ static int pxa2xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
 	if (status != 0)
 		return status;
 	write_SSCR0(0, drv_data->ioaddr);
-	pxa_set_cken(drv_data->master_info->clock_enable, 0);
+	clk_disable(ssp->clk);
 
 	return 0;
 }
@@ -1597,10 +1562,11 @@ static int pxa2xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
 static int pxa2xx_spi_resume(struct platform_device *pdev)
 {
 	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	struct ssp_device *ssp = drv_data->ssp;
 	int status = 0;
 
 	/* Enable the SSP clock */
-	pxa_set_cken(drv_data->master_info->clock_enable, 1);
+	clk_disable(ssp->clk);
 
 	/* Start the queue running */
 	status = start_queue(drv_data);
diff --git a/include/asm-arm/arch-pxa/pxa2xx_spi.h b/include/asm-arm/arch-pxa/pxa2xx_spi.h
index acc7ec7a84a1..3459fb26ce97 100644
--- a/include/asm-arm/arch-pxa/pxa2xx_spi.h
+++ b/include/asm-arm/arch-pxa/pxa2xx_spi.h
@@ -22,32 +22,8 @@
 #define PXA2XX_CS_ASSERT (0x01)
 #define PXA2XX_CS_DEASSERT (0x02)
 
-#if defined(CONFIG_PXA25x)
-#define CLOCK_SPEED_HZ 3686400
-#define SSP1_SerClkDiv(x) (((CLOCK_SPEED_HZ/2/(x+1))<<8)&0x0000ff00)
-#define SSP2_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
-#define SSP3_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
-#elif defined(CONFIG_PXA27x)
-#define CLOCK_SPEED_HZ 13000000
-#define SSP1_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
-#define SSP2_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
-#define SSP3_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
-#endif
-
-#define SSP1_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(1)))))
-#define SSP2_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(2)))))
-#define SSP3_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(3)))))
-
-enum pxa_ssp_type {
-	SSP_UNDEFINED = 0,
-	PXA25x_SSP,  /* pxa 210, 250, 255, 26x */
-	PXA25x_NSSP, /* pxa 255, 26x (including ASSP) */
-	PXA27x_SSP,
-};
-
 /* device.platform_data for SSP controller devices */
 struct pxa2xx_spi_master {
-	enum pxa_ssp_type ssp_type;
 	u32 clock_enable;
 	u16 num_chipselect;
 	u8 enable_dma;
-- 
cgit v1.2.3


From a333aeb73b45d2b6bbaaebd56f9e7e3a674ac039 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Fri, 21 Dec 2007 09:31:25 +0100
Subject: [ARM] 4718/1: Fix redefinition warnings in PXA uncompressor code

FFUART and friends are already defined as __REG(x) in pxa-regs.h.
Instead of redefining them here, we can just provide the __REG macro.
Including asm/arch/hardware.h is not an option because this physical
addresses are needed here.
This is a fix for the compiler warnings generated by 4663/1.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-pxa/uncompress.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-pxa/uncompress.h b/include/asm-arm/arch-pxa/uncompress.h
index 3faad53a6848..dadf4c20b622 100644
--- a/include/asm-arm/arch-pxa/uncompress.h
+++ b/include/asm-arm/arch-pxa/uncompress.h
@@ -12,10 +12,7 @@
 #include <linux/serial_reg.h>
 #include <asm/arch/pxa-regs.h>
 
-#define FFUART		((volatile unsigned long *)0x40100000)
-#define BTUART		((volatile unsigned long *)0x40200000)
-#define STUART		((volatile unsigned long *)0x40700000)
-#define HWUART		((volatile unsigned long *)0x41600000)
+#define __REG(x)	((volatile unsigned long *)x)
 
 #define UART		FFUART
 
-- 
cgit v1.2.3


From fafc9d3fa35530c1a14e6743c477d7398b431e74 Mon Sep 17 00:00:00 2001
From: Bridge Wu <bridge.wu@marvell.com>
Date: Fri, 21 Dec 2007 19:00:13 +0800
Subject: [ARM] pxa: mmc: add 1st host controller support for pxa3xx

This patchis to add the first mmc controller support for pxa3xx.
It's valid for pxa3[0|1|2]0.

On zylonite, the first controller supports two slots, this patch
only support the first one right now.

Signed-off-by: Bridge Wu <bridge.wu@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/devices.c         |  2 +-
 arch/arm/mach-pxa/pxa3xx.c          |  2 +
 arch/arm/mach-pxa/zylonite.c        | 86 +++++++++++++++++++++++++++++++++++++
 arch/arm/mach-pxa/zylonite_pxa300.c | 13 ++++++
 arch/arm/mach-pxa/zylonite_pxa320.c | 13 ++++++
 drivers/mmc/host/pxamci.h           |  2 +-
 include/asm-arm/arch-pxa/zylonite.h |  9 ++++
 7 files changed, 125 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/devices.c b/arch/arm/mach-pxa/devices.c
index 75949eb3b7bd..202d048f1cdd 100644
--- a/arch/arm/mach-pxa/devices.c
+++ b/arch/arm/mach-pxa/devices.c
@@ -51,7 +51,7 @@ static u64 pxamci_dmamask = 0xffffffffUL;
 
 struct platform_device pxa_device_mci = {
 	.name		= "pxa2xx-mci",
-	.id		= -1,
+	.id		= 0,
 	.dev		= {
 		.dma_mask = &pxamci_dmamask,
 		.coherent_dma_mask = 0xffffffff,
diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c
index c0483c3261d0..6271af303cb9 100644
--- a/arch/arm/mach-pxa/pxa3xx.c
+++ b/arch/arm/mach-pxa/pxa3xx.c
@@ -194,6 +194,8 @@ static struct clk pxa3xx_clks[] = {
 	PXA3xx_CKEN("SSPCLK", SSP2, 13000000, 0, &pxa27x_device_ssp2.dev),
 	PXA3xx_CKEN("SSPCLK", SSP3, 13000000, 0, &pxa27x_device_ssp3.dev),
 	PXA3xx_CKEN("SSPCLK", SSP4, 13000000, 0, &pxa3xx_device_ssp4.dev),
+
+	PXA3xx_CKEN("MMCCLK", MMC1, 19500000, 0, &pxa_device_mci.dev),
 };
 
 void __init pxa3xx_init_irq(void)
diff --git a/arch/arm/mach-pxa/zylonite.c b/arch/arm/mach-pxa/zylonite.c
index 743a87b2faa1..f72f37f6ff0c 100644
--- a/arch/arm/mach-pxa/zylonite.c
+++ b/arch/arm/mach-pxa/zylonite.c
@@ -25,9 +25,13 @@
 #include <asm/arch/gpio.h>
 #include <asm/arch/pxafb.h>
 #include <asm/arch/zylonite.h>
+#include <asm/arch/mmc.h>
 
 #include "generic.h"
 
+#define MAX_SLOTS	2
+struct platform_mmc_slot zylonite_mmc_slot[MAX_SLOTS];
+
 int gpio_backlight;
 int gpio_eth_irq;
 
@@ -156,6 +160,87 @@ static void __init zylonite_init_lcd(void)
 static inline void zylonite_init_lcd(void) {}
 #endif
 
+#if defined(CONFIG_MMC)
+static int zylonite_mci_ro(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+
+	return gpio_get_value(zylonite_mmc_slot[pdev->id].gpio_wp);
+}
+
+static int zylonite_mci_init(struct device *dev,
+			     irq_handler_t zylonite_detect_int,
+			     void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	int err, cd_irq, gpio_cd, gpio_wp;
+
+	cd_irq = gpio_to_irq(zylonite_mmc_slot[pdev->id].gpio_cd);
+	gpio_cd = zylonite_mmc_slot[pdev->id].gpio_cd;
+	gpio_wp = zylonite_mmc_slot[pdev->id].gpio_wp;
+
+	/*
+	 * setup GPIO for Zylonite MMC controller
+	 */
+	err = gpio_request(gpio_cd, "mmc card detect");
+	if (err)
+		goto err_request_cd;
+	gpio_direction_input(gpio_cd);
+
+	err = gpio_request(gpio_wp, "mmc write protect");
+	if (err)
+		goto err_request_wp;
+	gpio_direction_input(gpio_wp);
+
+	err = request_irq(cd_irq, zylonite_detect_int,
+			  IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
+			  "MMC card detect", data);
+	if (err) {
+		printk(KERN_ERR "%s: MMC/SD/SDIO: "
+				"can't request card detect IRQ\n", __func__);
+		goto err_request_irq;
+	}
+
+	return 0;
+
+err_request_irq:
+	gpio_free(gpio_wp);
+err_request_wp:
+	gpio_free(gpio_cd);
+err_request_cd:
+	return err;
+}
+
+static void zylonite_mci_exit(struct device *dev, void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	int cd_irq, gpio_cd, gpio_wp;
+
+	cd_irq = gpio_to_irq(zylonite_mmc_slot[pdev->id].gpio_cd);
+	gpio_cd = zylonite_mmc_slot[pdev->id].gpio_cd;
+	gpio_wp = zylonite_mmc_slot[pdev->id].gpio_wp;
+
+	free_irq(cd_irq, data);
+	gpio_free(gpio_cd);
+	gpio_free(gpio_wp);
+}
+
+static struct pxamci_platform_data zylonite_mci_platform_data = {
+	.detect_delay	= 20,
+	.ocr_mask	= MMC_VDD_32_33|MMC_VDD_33_34,
+	.init 		= zylonite_mci_init,
+	.exit		= zylonite_mci_exit,
+	.get_ro		= zylonite_mci_ro,
+};
+
+static void __init zylonite_init_mmc(void)
+{
+	pxa_set_mci_info(&zylonite_mci_platform_data);
+}
+#else
+static inline void zylonite_init_mmc(void) {}
+#endif
+
 static void __init zylonite_init(void)
 {
 	/* board-processor specific initialization */
@@ -171,6 +256,7 @@ static void __init zylonite_init(void)
 	platform_device_register(&smc91x_device);
 
 	zylonite_init_lcd();
+	zylonite_init_mmc();
 }
 
 MACHINE_START(ZYLONITE, "PXA3xx Platform Development Kit (aka Zylonite)")
diff --git a/arch/arm/mach-pxa/zylonite_pxa300.c b/arch/arm/mach-pxa/zylonite_pxa300.c
index 1832bc316501..cad92d480f2a 100644
--- a/arch/arm/mach-pxa/zylonite_pxa300.c
+++ b/arch/arm/mach-pxa/zylonite_pxa300.c
@@ -88,6 +88,15 @@ static mfp_cfg_t common_mfp_cfg[] __initdata = {
 	GPIO4_2_KP_MKOUT_5,
 	GPIO5_2_KP_MKOUT_6,
 	GPIO6_2_KP_MKOUT_7,
+
+	/* MMC1 */
+	GPIO3_MMC1_DAT0,
+	GPIO4_MMC1_DAT1,
+	GPIO5_MMC1_DAT2,
+	GPIO6_MMC1_DAT3,
+	GPIO7_MMC1_CLK,
+	GPIO8_MMC1_CMD,	/* CMD0 for slot 0 */
+	GPIO15_GPIO,	/* CMD1 default as GPIO for slot 0 */
 };
 
 static mfp_cfg_t pxa300_mfp_cfg[] __initdata = {
@@ -174,6 +183,10 @@ void __init zylonite_pxa300_init(void)
 
 		/* GPIO pin assignment */
 		gpio_backlight = mfp_to_gpio(MFP_PIN_GPIO20);
+
+		/* MMC card detect & write protect for controller 0 */
+		zylonite_mmc_slot[0].gpio_cd  = EXT_GPIO(0);
+		zylonite_mmc_slot[0].gpio_wp  = EXT_GPIO(2);
 	}
 
 	if (cpu_is_pxa300()) {
diff --git a/arch/arm/mach-pxa/zylonite_pxa320.c b/arch/arm/mach-pxa/zylonite_pxa320.c
index 94c715808b59..593f7bffb3b4 100644
--- a/arch/arm/mach-pxa/zylonite_pxa320.c
+++ b/arch/arm/mach-pxa/zylonite_pxa320.c
@@ -95,6 +95,15 @@ static mfp_cfg_t mfp_cfg[] __initdata = {
 	/* Ethernet */
 	GPIO4_nCS3,
 	GPIO90_GPIO,
+
+	/* MMC1 */
+	GPIO18_MMC1_DAT0,
+	GPIO19_MMC1_DAT1,
+	GPIO20_MMC1_DAT2,
+	GPIO21_MMC1_DAT3,
+	GPIO22_MMC1_CLK,
+	GPIO23_MMC1_CMD,/* CMD0 for slot 0 */
+	GPIO31_GPIO,	/* CMD1 default as GPIO for slot 0 */
 };
 
 #define NUM_LCD_DETECT_PINS	7
@@ -169,5 +178,9 @@ void __init zylonite_pxa320_init(void)
 		/* GPIO pin assignment */
 		gpio_backlight	= mfp_to_gpio(MFP_PIN_GPIO14);
 		gpio_eth_irq	= mfp_to_gpio(MFP_PIN_GPIO9);
+
+		/* MMC card detect & write protect for controller 0 */
+		zylonite_mmc_slot[0].gpio_cd  = mfp_to_gpio(MFP_PIN_GPIO1);
+		zylonite_mmc_slot[0].gpio_wp  = mfp_to_gpio(MFP_PIN_GPIO5);
 	}
 }
diff --git a/drivers/mmc/host/pxamci.h b/drivers/mmc/host/pxamci.h
index 748c7706f237..f6c2e2fcce37 100644
--- a/drivers/mmc/host/pxamci.h
+++ b/drivers/mmc/host/pxamci.h
@@ -68,7 +68,7 @@
 #define PRG_DONE		(1 << 1)
 #define DATA_TRAN_DONE		(1 << 0)
 
-#ifdef CONFIG_PXA27x
+#if defined(CONFIG_PXA27x) || defined(CONFIG_PXA3xx)
 #define MMC_I_MASK_ALL          0x00001fff
 #else
 #define MMC_I_MASK_ALL          0x0000007f
diff --git a/include/asm-arm/arch-pxa/zylonite.h b/include/asm-arm/arch-pxa/zylonite.h
index f58b59162b82..5f717d64ea7d 100644
--- a/include/asm-arm/arch-pxa/zylonite.h
+++ b/include/asm-arm/arch-pxa/zylonite.h
@@ -3,9 +3,18 @@
 
 #define ZYLONITE_ETH_PHYS	0x14000000
 
+#define EXT_GPIO(x)		(128 + (x))
+
 /* the following variables are processor specific and initialized
  * by the corresponding zylonite_pxa3xx_init()
  */
+struct platform_mmc_slot {
+	int gpio_cd;
+	int gpio_wp;
+};
+
+extern struct platform_mmc_slot zylonite_mmc_slot[];
+
 extern int gpio_backlight;
 extern int gpio_eth_irq;
 
-- 
cgit v1.2.3


From 8d33b05581d1bc66e2356957bb2739e177a9cc36 Mon Sep 17 00:00:00 2001
From: Bridge Wu <bridge.wu@marvell.com>
Date: Fri, 21 Dec 2007 19:15:36 +0800
Subject: [ARM] pxa: mmc: add 2nd host controller support for pxa3xx

This patch is to add the second mmc controller support for pxa3xx.
It's valid for pxa3[0|1|2]0.

On zylonite, the second controller has no slot.

Signed-off-by: Bridge Wu <bridge.wu@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/devices.c         | 40 +++++++++++++++++++++++++++++++++++++
 arch/arm/mach-pxa/devices.h         |  1 +
 arch/arm/mach-pxa/pxa3xx.c          |  1 +
 arch/arm/mach-pxa/zylonite.c        |  6 ++++++
 arch/arm/mach-pxa/zylonite_pxa300.c |  8 ++++++++
 arch/arm/mach-pxa/zylonite_pxa320.c |  8 ++++++++
 include/asm-arm/arch-pxa/mmc.h      |  1 +
 7 files changed, 65 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/devices.c b/arch/arm/mach-pxa/devices.c
index 202d048f1cdd..bc04e87ed3b4 100644
--- a/arch/arm/mach-pxa/devices.c
+++ b/arch/arm/mach-pxa/devices.c
@@ -549,4 +549,44 @@ struct platform_device pxa3xx_device_ssp4 = {
 	.resource	= pxa3xx_resource_ssp4,
 	.num_resources	= ARRAY_SIZE(pxa3xx_resource_ssp4),
 };
+
+static struct resource pxa3xx_resources_mci2[] = {
+	[0] = {
+		.start	= 0x42000000,
+		.end	= 0x42000fff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_MMC2,
+		.end	= IRQ_MMC2,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		.start	= 93,
+		.end	= 93,
+		.flags	= IORESOURCE_DMA,
+	},
+	[3] = {
+		.start	= 94,
+		.end	= 94,
+		.flags	= IORESOURCE_DMA,
+	},
+};
+
+struct platform_device pxa3xx_device_mci2 = {
+	.name		= "pxa2xx-mci",
+	.id		= 1,
+	.dev		= {
+		.dma_mask = &pxamci_dmamask,
+		.coherent_dma_mask =	0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(pxa3xx_resources_mci2),
+	.resource	= pxa3xx_resources_mci2,
+};
+
+void __init pxa3xx_set_mci2_info(struct pxamci_platform_data *info)
+{
+	pxa_register_device(&pxa3xx_device_mci2, info);
+}
+
 #endif /* CONFIG_PXA3xx */
diff --git a/arch/arm/mach-pxa/devices.h b/arch/arm/mach-pxa/devices.h
index e78b48f62b36..d78068cc8015 100644
--- a/arch/arm/mach-pxa/devices.h
+++ b/arch/arm/mach-pxa/devices.h
@@ -1,4 +1,5 @@
 extern struct platform_device pxa_device_mci;
+extern struct platform_device pxa3xx_device_mci2;
 extern struct platform_device pxa_device_udc;
 extern struct platform_device pxa_device_fb;
 extern struct platform_device pxa_device_ffuart;
diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c
index 6271af303cb9..124ad5eb7ab0 100644
--- a/arch/arm/mach-pxa/pxa3xx.c
+++ b/arch/arm/mach-pxa/pxa3xx.c
@@ -196,6 +196,7 @@ static struct clk pxa3xx_clks[] = {
 	PXA3xx_CKEN("SSPCLK", SSP4, 13000000, 0, &pxa3xx_device_ssp4.dev),
 
 	PXA3xx_CKEN("MMCCLK", MMC1, 19500000, 0, &pxa_device_mci.dev),
+	PXA3xx_CKEN("MMCCLK", MMC2, 19500000, 0, &pxa3xx_device_mci2.dev),
 };
 
 void __init pxa3xx_init_irq(void)
diff --git a/arch/arm/mach-pxa/zylonite.c b/arch/arm/mach-pxa/zylonite.c
index f72f37f6ff0c..33986036f146 100644
--- a/arch/arm/mach-pxa/zylonite.c
+++ b/arch/arm/mach-pxa/zylonite.c
@@ -233,9 +233,15 @@ static struct pxamci_platform_data zylonite_mci_platform_data = {
 	.get_ro		= zylonite_mci_ro,
 };
 
+static struct pxamci_platform_data zylonite_mci2_platform_data = {
+	.detect_delay	= 20,
+	.ocr_mask	= MMC_VDD_32_33|MMC_VDD_33_34,
+};
+
 static void __init zylonite_init_mmc(void)
 {
 	pxa_set_mci_info(&zylonite_mci_platform_data);
+	pxa3xx_set_mci2_info(&zylonite_mci2_platform_data);
 }
 #else
 static inline void zylonite_init_mmc(void) {}
diff --git a/arch/arm/mach-pxa/zylonite_pxa300.c b/arch/arm/mach-pxa/zylonite_pxa300.c
index cad92d480f2a..38c1ddc6cb1d 100644
--- a/arch/arm/mach-pxa/zylonite_pxa300.c
+++ b/arch/arm/mach-pxa/zylonite_pxa300.c
@@ -97,6 +97,14 @@ static mfp_cfg_t common_mfp_cfg[] __initdata = {
 	GPIO7_MMC1_CLK,
 	GPIO8_MMC1_CMD,	/* CMD0 for slot 0 */
 	GPIO15_GPIO,	/* CMD1 default as GPIO for slot 0 */
+
+	/* MMC2 */
+	GPIO9_MMC2_DAT0,
+	GPIO10_MMC2_DAT1,
+	GPIO11_MMC2_DAT2,
+	GPIO12_MMC2_DAT3,
+	GPIO13_MMC2_CLK,
+	GPIO14_MMC2_CMD,
 };
 
 static mfp_cfg_t pxa300_mfp_cfg[] __initdata = {
diff --git a/arch/arm/mach-pxa/zylonite_pxa320.c b/arch/arm/mach-pxa/zylonite_pxa320.c
index 593f7bffb3b4..1036da3315d8 100644
--- a/arch/arm/mach-pxa/zylonite_pxa320.c
+++ b/arch/arm/mach-pxa/zylonite_pxa320.c
@@ -104,6 +104,14 @@ static mfp_cfg_t mfp_cfg[] __initdata = {
 	GPIO22_MMC1_CLK,
 	GPIO23_MMC1_CMD,/* CMD0 for slot 0 */
 	GPIO31_GPIO,	/* CMD1 default as GPIO for slot 0 */
+
+	/* MMC2 */
+	GPIO24_MMC2_DAT0,
+	GPIO25_MMC2_DAT1,
+	GPIO26_MMC2_DAT2,
+	GPIO27_MMC2_DAT3,
+	GPIO28_MMC2_CLK,
+	GPIO29_MMC2_CMD,
 };
 
 #define NUM_LCD_DETECT_PINS	7
diff --git a/include/asm-arm/arch-pxa/mmc.h b/include/asm-arm/arch-pxa/mmc.h
index ef4f570381d1..dd30299a3c66 100644
--- a/include/asm-arm/arch-pxa/mmc.h
+++ b/include/asm-arm/arch-pxa/mmc.h
@@ -17,5 +17,6 @@ struct pxamci_platform_data {
 };
 
 extern void pxa_set_mci_info(struct pxamci_platform_data *info);
+extern void pxa3xx_set_mci2_info(struct pxamci_platform_data *info);
 
 #endif
-- 
cgit v1.2.3


From 5a1f21b1e546fafe840944f02949c1a8a8725132 Mon Sep 17 00:00:00 2001
From: Bridge Wu <bridge.wu@marvell.com>
Date: Fri, 21 Dec 2007 19:27:08 +0800
Subject: [ARM] pxa: mmc: add 3rd host controller support for pxa310

This patch is to add the third mmc controller support _only_
for pxa310.

On zylonite, the third controller support one slot.

Signed-off-by: Bridge Wu <bridge.wu@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/devices.c         | 39 +++++++++++++++++++++++++++++++++++++
 arch/arm/mach-pxa/devices.h         |  1 +
 arch/arm/mach-pxa/pxa3xx.c          |  1 +
 arch/arm/mach-pxa/zylonite.c        |  4 +++-
 arch/arm/mach-pxa/zylonite_pxa300.c | 12 ++++++++++++
 include/asm-arm/arch-pxa/mmc.h      |  1 +
 6 files changed, 57 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/devices.c b/arch/arm/mach-pxa/devices.c
index bc04e87ed3b4..061ec92eb1e2 100644
--- a/arch/arm/mach-pxa/devices.c
+++ b/arch/arm/mach-pxa/devices.c
@@ -589,4 +589,43 @@ void __init pxa3xx_set_mci2_info(struct pxamci_platform_data *info)
 	pxa_register_device(&pxa3xx_device_mci2, info);
 }
 
+static struct resource pxa3xx_resources_mci3[] = {
+	[0] = {
+		.start	= 0x42500000,
+		.end	= 0x42500fff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_MMC3,
+		.end	= IRQ_MMC3,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		.start	= 100,
+		.end	= 100,
+		.flags	= IORESOURCE_DMA,
+	},
+	[3] = {
+		.start	= 101,
+		.end	= 101,
+		.flags	= IORESOURCE_DMA,
+	},
+};
+
+struct platform_device pxa3xx_device_mci3 = {
+	.name		= "pxa2xx-mci",
+	.id		= 2,
+	.dev		= {
+		.dma_mask = &pxamci_dmamask,
+		.coherent_dma_mask = 0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(pxa3xx_resources_mci3),
+	.resource	= pxa3xx_resources_mci3,
+};
+
+void __init pxa3xx_set_mci3_info(struct pxamci_platform_data *info)
+{
+	pxa_register_device(&pxa3xx_device_mci3, info);
+}
+
 #endif /* CONFIG_PXA3xx */
diff --git a/arch/arm/mach-pxa/devices.h b/arch/arm/mach-pxa/devices.h
index d78068cc8015..96c7c8909068 100644
--- a/arch/arm/mach-pxa/devices.h
+++ b/arch/arm/mach-pxa/devices.h
@@ -1,5 +1,6 @@
 extern struct platform_device pxa_device_mci;
 extern struct platform_device pxa3xx_device_mci2;
+extern struct platform_device pxa3xx_device_mci3;
 extern struct platform_device pxa_device_udc;
 extern struct platform_device pxa_device_fb;
 extern struct platform_device pxa_device_ffuart;
diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c
index 124ad5eb7ab0..843ba9bf3df7 100644
--- a/arch/arm/mach-pxa/pxa3xx.c
+++ b/arch/arm/mach-pxa/pxa3xx.c
@@ -197,6 +197,7 @@ static struct clk pxa3xx_clks[] = {
 
 	PXA3xx_CKEN("MMCCLK", MMC1, 19500000, 0, &pxa_device_mci.dev),
 	PXA3xx_CKEN("MMCCLK", MMC2, 19500000, 0, &pxa3xx_device_mci2.dev),
+	PXA3xx_CKEN("MMCCLK", MMC3, 19500000, 0, &pxa3xx_device_mci3.dev),
 };
 
 void __init pxa3xx_init_irq(void)
diff --git a/arch/arm/mach-pxa/zylonite.c b/arch/arm/mach-pxa/zylonite.c
index 33986036f146..b49c69d024c7 100644
--- a/arch/arm/mach-pxa/zylonite.c
+++ b/arch/arm/mach-pxa/zylonite.c
@@ -29,7 +29,7 @@
 
 #include "generic.h"
 
-#define MAX_SLOTS	2
+#define MAX_SLOTS	3
 struct platform_mmc_slot zylonite_mmc_slot[MAX_SLOTS];
 
 int gpio_backlight;
@@ -242,6 +242,8 @@ static void __init zylonite_init_mmc(void)
 {
 	pxa_set_mci_info(&zylonite_mci_platform_data);
 	pxa3xx_set_mci2_info(&zylonite_mci2_platform_data);
+	if (cpu_is_pxa310())
+		pxa3xx_set_mci3_info(&zylonite_mci_platform_data);
 }
 #else
 static inline void zylonite_init_mmc(void) {}
diff --git a/arch/arm/mach-pxa/zylonite_pxa300.c b/arch/arm/mach-pxa/zylonite_pxa300.c
index 38c1ddc6cb1d..0e159a3024c0 100644
--- a/arch/arm/mach-pxa/zylonite_pxa300.c
+++ b/arch/arm/mach-pxa/zylonite_pxa300.c
@@ -133,6 +133,14 @@ static mfp_cfg_t pxa310_mfp_cfg[] __initdata = {
 	/* Ethernet */
 	GPIO2_nCS3,
 	GPIO102_GPIO,
+
+	/* MMC3 */
+	GPIO7_2_MMC3_DAT0,
+	GPIO8_2_MMC3_DAT1,
+	GPIO9_2_MMC3_DAT2,
+	GPIO10_2_MMC3_DAT3,
+	GPIO103_MMC3_CLK,
+	GPIO105_MMC3_CMD,
 };
 
 #define NUM_LCD_DETECT_PINS	7
@@ -205,5 +213,9 @@ void __init zylonite_pxa300_init(void)
 	if (cpu_is_pxa310()) {
 		pxa3xx_mfp_config(ARRAY_AND_SIZE(pxa310_mfp_cfg));
 		gpio_eth_irq = mfp_to_gpio(MFP_PIN_GPIO102);
+
+		/* MMC card detect & write protect for controller 2 */
+		zylonite_mmc_slot[2].gpio_cd = EXT_GPIO(30);
+		zylonite_mmc_slot[2].gpio_wp = EXT_GPIO(31);
 	}
 }
diff --git a/include/asm-arm/arch-pxa/mmc.h b/include/asm-arm/arch-pxa/mmc.h
index dd30299a3c66..6d1304c9270f 100644
--- a/include/asm-arm/arch-pxa/mmc.h
+++ b/include/asm-arm/arch-pxa/mmc.h
@@ -18,5 +18,6 @@ struct pxamci_platform_data {
 
 extern void pxa_set_mci_info(struct pxamci_platform_data *info);
 extern void pxa3xx_set_mci2_info(struct pxamci_platform_data *info);
+extern void pxa3xx_set_mci3_info(struct pxamci_platform_data *info);
 
 #endif
-- 
cgit v1.2.3


From 88d456386c2b3ac96e7170f5bc555b3c43f5a951 Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Tue, 25 Dec 2007 10:34:33 +0800
Subject: [ARM] pxa: make OHCI register definitions available to both pxa27x
 and pxa3xx

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-pxa/pxa-regs.h | 139 ++++++++++++++++++------------------
 1 file changed, 71 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h
index dd014712dfa5..f7809ea77396 100644
--- a/include/asm-arm/arch-pxa/pxa-regs.h
+++ b/include/asm-arm/arch-pxa/pxa-regs.h
@@ -1969,74 +1969,6 @@
 #define KPAS_SO         (0x1 << 31)
 #define KPASMKPx_SO     (0x1 << 31)
 
-/*
- * UHC: USB Host Controller (OHCI-like) register definitions
- */
-#define UHC_BASE_PHYS	(0x4C000000)
-#define UHCREV		__REG(0x4C000000) /* UHC HCI Spec Revision */
-#define UHCHCON		__REG(0x4C000004) /* UHC Host Control Register */
-#define UHCCOMS		__REG(0x4C000008) /* UHC Command Status Register */
-#define UHCINTS		__REG(0x4C00000C) /* UHC Interrupt Status Register */
-#define UHCINTE		__REG(0x4C000010) /* UHC Interrupt Enable */
-#define UHCINTD		__REG(0x4C000014) /* UHC Interrupt Disable */
-#define UHCHCCA		__REG(0x4C000018) /* UHC Host Controller Comm. Area */
-#define UHCPCED		__REG(0x4C00001C) /* UHC Period Current Endpt Descr */
-#define UHCCHED		__REG(0x4C000020) /* UHC Control Head Endpt Descr */
-#define UHCCCED		__REG(0x4C000024) /* UHC Control Current Endpt Descr */
-#define UHCBHED		__REG(0x4C000028) /* UHC Bulk Head Endpt Descr */
-#define UHCBCED		__REG(0x4C00002C) /* UHC Bulk Current Endpt Descr */
-#define UHCDHEAD	__REG(0x4C000030) /* UHC Done Head */
-#define UHCFMI		__REG(0x4C000034) /* UHC Frame Interval */
-#define UHCFMR		__REG(0x4C000038) /* UHC Frame Remaining */
-#define UHCFMN		__REG(0x4C00003C) /* UHC Frame Number */
-#define UHCPERS		__REG(0x4C000040) /* UHC Periodic Start */
-#define UHCLS		__REG(0x4C000044) /* UHC Low Speed Threshold */
-
-#define UHCRHDA		__REG(0x4C000048) /* UHC Root Hub Descriptor A */
-#define UHCRHDA_NOCP	(1 << 12)	/* No over current protection */
-
-#define UHCRHDB		__REG(0x4C00004C) /* UHC Root Hub Descriptor B */
-#define UHCRHS		__REG(0x4C000050) /* UHC Root Hub Status */
-#define UHCRHPS1	__REG(0x4C000054) /* UHC Root Hub Port 1 Status */
-#define UHCRHPS2	__REG(0x4C000058) /* UHC Root Hub Port 2 Status */
-#define UHCRHPS3	__REG(0x4C00005C) /* UHC Root Hub Port 3 Status */
-
-#define UHCSTAT		__REG(0x4C000060) /* UHC Status Register */
-#define UHCSTAT_UPS3	(1 << 16)	/* USB Power Sense Port3 */
-#define UHCSTAT_SBMAI	(1 << 15)	/* System Bus Master Abort Interrupt*/
-#define UHCSTAT_SBTAI	(1 << 14)	/* System Bus Target Abort Interrupt*/
-#define UHCSTAT_UPRI	(1 << 13)	/* USB Port Resume Interrupt */
-#define UHCSTAT_UPS2	(1 << 12)	/* USB Power Sense Port 2 */
-#define UHCSTAT_UPS1	(1 << 11)	/* USB Power Sense Port 1 */
-#define UHCSTAT_HTA	(1 << 10)	/* HCI Target Abort */
-#define UHCSTAT_HBA	(1 << 8)	/* HCI Buffer Active */
-#define UHCSTAT_RWUE	(1 << 7)	/* HCI Remote Wake Up Event */
-
-#define UHCHR           __REG(0x4C000064) /* UHC Reset Register */
-#define UHCHR_SSEP3	(1 << 11)	/* Sleep Standby Enable for Port3 */
-#define UHCHR_SSEP2	(1 << 10)	/* Sleep Standby Enable for Port2 */
-#define UHCHR_SSEP1	(1 << 9)	/* Sleep Standby Enable for Port1 */
-#define UHCHR_PCPL	(1 << 7)	/* Power control polarity low */
-#define UHCHR_PSPL	(1 << 6)	/* Power sense polarity low */
-#define UHCHR_SSE	(1 << 5)	/* Sleep Standby Enable */
-#define UHCHR_UIT	(1 << 4)	/* USB Interrupt Test */
-#define UHCHR_SSDC	(1 << 3)	/* Simulation Scale Down Clock */
-#define UHCHR_CGR	(1 << 2)	/* Clock Generation Reset */
-#define UHCHR_FHR	(1 << 1)	/* Force Host Controller Reset */
-#define UHCHR_FSBIR	(1 << 0)	/* Force System Bus Iface Reset */
-
-#define UHCHIE          __REG(0x4C000068) /* UHC Interrupt Enable Register*/
-#define UHCHIE_UPS3IE	(1 << 14)	/* Power Sense Port3 IntEn */
-#define UHCHIE_UPRIE	(1 << 13)	/* Port Resume IntEn */
-#define UHCHIE_UPS2IE	(1 << 12)	/* Power Sense Port2 IntEn */
-#define UHCHIE_UPS1IE	(1 << 11)	/* Power Sense Port1 IntEn */
-#define UHCHIE_TAIE	(1 << 10)	/* HCI Interface Transfer Abort
-					   Interrupt Enable*/
-#define UHCHIE_HBAIE	(1 << 8)	/* HCI Buffer Active IntEn */
-#define UHCHIE_RWIE	(1 << 7)	/* Remote Wake-up IntEn */
-
-#define UHCHIT          __REG(0x4C00006C) /* UHC Interrupt Test register */
-
 /* Camera Interface */
 #define CICR0		__REG(0x50000000)
 #define CICR1		__REG(0x50000004)
@@ -2184,6 +2116,77 @@
 
 #endif
 
+#if defined(CONFIG_PXA27x) || defined(CONFIG_PXA3xx)
+/*
+ * UHC: USB Host Controller (OHCI-like) register definitions
+ */
+#define UHC_BASE_PHYS	(0x4C000000)
+#define UHCREV		__REG(0x4C000000) /* UHC HCI Spec Revision */
+#define UHCHCON		__REG(0x4C000004) /* UHC Host Control Register */
+#define UHCCOMS		__REG(0x4C000008) /* UHC Command Status Register */
+#define UHCINTS		__REG(0x4C00000C) /* UHC Interrupt Status Register */
+#define UHCINTE		__REG(0x4C000010) /* UHC Interrupt Enable */
+#define UHCINTD		__REG(0x4C000014) /* UHC Interrupt Disable */
+#define UHCHCCA		__REG(0x4C000018) /* UHC Host Controller Comm. Area */
+#define UHCPCED		__REG(0x4C00001C) /* UHC Period Current Endpt Descr */
+#define UHCCHED		__REG(0x4C000020) /* UHC Control Head Endpt Descr */
+#define UHCCCED		__REG(0x4C000024) /* UHC Control Current Endpt Descr */
+#define UHCBHED		__REG(0x4C000028) /* UHC Bulk Head Endpt Descr */
+#define UHCBCED		__REG(0x4C00002C) /* UHC Bulk Current Endpt Descr */
+#define UHCDHEAD	__REG(0x4C000030) /* UHC Done Head */
+#define UHCFMI		__REG(0x4C000034) /* UHC Frame Interval */
+#define UHCFMR		__REG(0x4C000038) /* UHC Frame Remaining */
+#define UHCFMN		__REG(0x4C00003C) /* UHC Frame Number */
+#define UHCPERS		__REG(0x4C000040) /* UHC Periodic Start */
+#define UHCLS		__REG(0x4C000044) /* UHC Low Speed Threshold */
+
+#define UHCRHDA		__REG(0x4C000048) /* UHC Root Hub Descriptor A */
+#define UHCRHDA_NOCP	(1 << 12)	/* No over current protection */
+
+#define UHCRHDB		__REG(0x4C00004C) /* UHC Root Hub Descriptor B */
+#define UHCRHS		__REG(0x4C000050) /* UHC Root Hub Status */
+#define UHCRHPS1	__REG(0x4C000054) /* UHC Root Hub Port 1 Status */
+#define UHCRHPS2	__REG(0x4C000058) /* UHC Root Hub Port 2 Status */
+#define UHCRHPS3	__REG(0x4C00005C) /* UHC Root Hub Port 3 Status */
+
+#define UHCSTAT		__REG(0x4C000060) /* UHC Status Register */
+#define UHCSTAT_UPS3	(1 << 16)	/* USB Power Sense Port3 */
+#define UHCSTAT_SBMAI	(1 << 15)	/* System Bus Master Abort Interrupt*/
+#define UHCSTAT_SBTAI	(1 << 14)	/* System Bus Target Abort Interrupt*/
+#define UHCSTAT_UPRI	(1 << 13)	/* USB Port Resume Interrupt */
+#define UHCSTAT_UPS2	(1 << 12)	/* USB Power Sense Port 2 */
+#define UHCSTAT_UPS1	(1 << 11)	/* USB Power Sense Port 1 */
+#define UHCSTAT_HTA	(1 << 10)	/* HCI Target Abort */
+#define UHCSTAT_HBA	(1 << 8)	/* HCI Buffer Active */
+#define UHCSTAT_RWUE	(1 << 7)	/* HCI Remote Wake Up Event */
+
+#define UHCHR           __REG(0x4C000064) /* UHC Reset Register */
+#define UHCHR_SSEP3	(1 << 11)	/* Sleep Standby Enable for Port3 */
+#define UHCHR_SSEP2	(1 << 10)	/* Sleep Standby Enable for Port2 */
+#define UHCHR_SSEP1	(1 << 9)	/* Sleep Standby Enable for Port1 */
+#define UHCHR_PCPL	(1 << 7)	/* Power control polarity low */
+#define UHCHR_PSPL	(1 << 6)	/* Power sense polarity low */
+#define UHCHR_SSE	(1 << 5)	/* Sleep Standby Enable */
+#define UHCHR_UIT	(1 << 4)	/* USB Interrupt Test */
+#define UHCHR_SSDC	(1 << 3)	/* Simulation Scale Down Clock */
+#define UHCHR_CGR	(1 << 2)	/* Clock Generation Reset */
+#define UHCHR_FHR	(1 << 1)	/* Force Host Controller Reset */
+#define UHCHR_FSBIR	(1 << 0)	/* Force System Bus Iface Reset */
+
+#define UHCHIE          __REG(0x4C000068) /* UHC Interrupt Enable Register*/
+#define UHCHIE_UPS3IE	(1 << 14)	/* Power Sense Port3 IntEn */
+#define UHCHIE_UPRIE	(1 << 13)	/* Port Resume IntEn */
+#define UHCHIE_UPS2IE	(1 << 12)	/* Power Sense Port2 IntEn */
+#define UHCHIE_UPS1IE	(1 << 11)	/* Power Sense Port1 IntEn */
+#define UHCHIE_TAIE	(1 << 10)	/* HCI Interface Transfer Abort
+					   Interrupt Enable*/
+#define UHCHIE_HBAIE	(1 << 8)	/* HCI Buffer Active IntEn */
+#define UHCHIE_RWIE	(1 << 7)	/* Remote Wake-up IntEn */
+
+#define UHCHIT          __REG(0x4C00006C) /* UHC Interrupt Test register */
+
+#endif /* CONFIG_PXA27x || CONFIG_PXA3xx */
+
 /* PWRMODE register M field values */
 
 #define PWRMODE_IDLE		0x1
-- 
cgit v1.2.3


From 8785a8fbd5a1624dbabd7c782524450e902b722e Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 14 Jan 2008 17:02:33 +0000
Subject: [ARM] pxa: move memory controller registers into pxa2xx-regs.h

PXA3 has a different memory controller from PXA2 platforms.  Avoid
clashing definitions by moving the PXA2 definitions to pxa2xx-regs.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/cpu-pxa.c            |  1 +
 arch/arm/mach-pxa/lpd270.c             |  1 +
 arch/arm/mach-pxa/lubbock.c            |  1 +
 arch/arm/mach-pxa/mainstone.c          |  1 +
 arch/arm/mach-pxa/pxa27x.c             |  1 +
 arch/arm/mach-pxa/sleep.S              |  1 +
 drivers/pcmcia/pxa2xx_base.c           |  1 +
 include/asm-arm/arch-pxa/pxa-regs.h    | 63 -------------------------
 include/asm-arm/arch-pxa/pxa2xx-regs.h | 84 ++++++++++++++++++++++++++++++++++
 9 files changed, 91 insertions(+), 63 deletions(-)
 create mode 100644 include/asm-arm/arch-pxa/pxa2xx-regs.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/cpu-pxa.c b/arch/arm/mach-pxa/cpu-pxa.c
index 18d042bdf243..cbc583beedc8 100644
--- a/arch/arm/mach-pxa/cpu-pxa.c
+++ b/arch/arm/mach-pxa/cpu-pxa.c
@@ -39,6 +39,7 @@
 
 #include <asm/hardware.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 
 #ifdef DEBUG
 static unsigned int freq_debug;
diff --git a/arch/arm/mach-pxa/lpd270.c b/arch/arm/mach-pxa/lpd270.c
index 78ebad063cba..afa62ffe3ad5 100644
--- a/arch/arm/mach-pxa/lpd270.c
+++ b/arch/arm/mach-pxa/lpd270.c
@@ -38,6 +38,7 @@
 #include <asm/mach/flash.h>
 
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 #include <asm/arch/lpd270.h>
 #include <asm/arch/audio.h>
 #include <asm/arch/pxafb.h>
diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c
index 1b9290c0ed5e..a75594bf68b5 100644
--- a/arch/arm/mach-pxa/lubbock.c
+++ b/arch/arm/mach-pxa/lubbock.c
@@ -41,6 +41,7 @@
 #include <asm/hardware/sa1111.h>
 
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 #include <asm/arch/lubbock.h>
 #include <asm/arch/udc.h>
 #include <asm/arch/irda.h>
diff --git a/arch/arm/mach-pxa/mainstone.c b/arch/arm/mach-pxa/mainstone.c
index 237459bf67c6..06e13608a2c9 100644
--- a/arch/arm/mach-pxa/mainstone.c
+++ b/arch/arm/mach-pxa/mainstone.c
@@ -39,6 +39,7 @@
 #include <asm/mach/flash.h>
 
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 #include <asm/arch/mainstone.h>
 #include <asm/arch/audio.h>
 #include <asm/arch/pxafb.h>
diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c
index ec7597b1cb54..b9fa5ec0a954 100644
--- a/arch/arm/mach-pxa/pxa27x.c
+++ b/arch/arm/mach-pxa/pxa27x.c
@@ -21,6 +21,7 @@
 #include <asm/irq.h>
 #include <asm/arch/irqs.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 #include <asm/arch/ohci.h>
 #include <asm/arch/pm.h>
 #include <asm/arch/dma.h>
diff --git a/arch/arm/mach-pxa/sleep.S b/arch/arm/mach-pxa/sleep.S
index d0447723b73a..f33d0c4a96d6 100644
--- a/arch/arm/mach-pxa/sleep.S
+++ b/arch/arm/mach-pxa/sleep.S
@@ -16,6 +16,7 @@
 #include <asm/hardware.h>
 
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 
 #define MDREFR_KDIV	0x200a4000	// all banks
 #define CCCR_SLEEP	0x00000107	// L=7 2N=2 A=0 PPDIS=0 CPDIS=0
diff --git a/drivers/pcmcia/pxa2xx_base.c b/drivers/pcmcia/pxa2xx_base.c
index 874923fcb2f9..e439044d88f2 100644
--- a/drivers/pcmcia/pxa2xx_base.c
+++ b/drivers/pcmcia/pxa2xx_base.c
@@ -29,6 +29,7 @@
 #include <asm/irq.h>
 #include <asm/system.h>
 #include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx-regs.h>
 
 #include <pcmcia/cs_types.h>
 #include <pcmcia/ss.h>
diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h
index f7809ea77396..442494d71f12 100644
--- a/include/asm-arm/arch-pxa/pxa-regs.h
+++ b/include/asm-arm/arch-pxa/pxa-regs.h
@@ -1848,71 +1848,8 @@
 
 #define LDCMD_PAL	(1 << 26)	/* instructs DMA to load palette buffer */
 
-/*
- * Memory controller
- */
-
-#define MDCNFG		__REG(0x48000000)  /* SDRAM Configuration Register 0 */
-#define MDREFR		__REG(0x48000004)  /* SDRAM Refresh Control Register */
-#define MSC0		__REG(0x48000008)  /* Static Memory Control Register 0 */
-#define MSC1		__REG(0x4800000C)  /* Static Memory Control Register 1 */
-#define MSC2		__REG(0x48000010)  /* Static Memory Control Register 2 */
-#define MECR		__REG(0x48000014)  /* Expansion Memory (PCMCIA/Compact Flash) Bus Configuration */
-#define SXLCR		__REG(0x48000018)  /* LCR value to be written to SDRAM-Timing Synchronous Flash */
-#define SXCNFG		__REG(0x4800001C)  /* Synchronous Static Memory Control Register */
-#define SXMRS		__REG(0x48000024)  /* MRS value to be written to Synchronous Flash or SMROM */
-#define MCMEM0		__REG(0x48000028)  /* Card interface Common Memory Space Socket 0 Timing */
-#define MCMEM1		__REG(0x4800002C)  /* Card interface Common Memory Space Socket 1 Timing */
-#define MCATT0		__REG(0x48000030)  /* Card interface Attribute Space Socket 0 Timing Configuration */
-#define MCATT1		__REG(0x48000034)  /* Card interface Attribute Space Socket 1 Timing Configuration */
-#define MCIO0		__REG(0x48000038)  /* Card interface I/O Space Socket 0 Timing Configuration */
-#define MCIO1		__REG(0x4800003C)  /* Card interface I/O Space Socket 1 Timing Configuration */
-#define MDMRS		__REG(0x48000040)  /* MRS value to be written to SDRAM */
-#define BOOT_DEF	__REG(0x48000044)  /* Read-Only Boot-Time Register. Contains BOOT_SEL and PKG_SEL */
-
-/*
- * More handy macros for PCMCIA
- *
- * Arg is socket number
- */
-#define MCMEM(s)	__REG2(0x48000028, (s)<<2 )  /* Card interface Common Memory Space Socket s Timing */
-#define MCATT(s)	__REG2(0x48000030, (s)<<2 )  /* Card interface Attribute Space Socket s Timing Configuration */
-#define MCIO(s)		__REG2(0x48000038, (s)<<2 )  /* Card interface I/O Space Socket s Timing Configuration */
-
-/* MECR register defines */
-#define MECR_NOS	(1 << 0)	/* Number Of Sockets: 0 -> 1 sock, 1 -> 2 sock */
-#define MECR_CIT	(1 << 1)	/* Card Is There: 0 -> no card, 1 -> card inserted */
-
-#define MDREFR_K0DB4	(1 << 29)	/* SDCLK0 Divide by 4 Control/Status */
-#define MDREFR_K2FREE	(1 << 25)	/* SDRAM Free-Running Control */
-#define MDREFR_K1FREE	(1 << 24)	/* SDRAM Free-Running Control */
-#define MDREFR_K0FREE	(1 << 23)	/* SDRAM Free-Running Control */
-#define MDREFR_SLFRSH	(1 << 22)	/* SDRAM Self-Refresh Control/Status */
-#define MDREFR_APD	(1 << 20)	/* SDRAM/SSRAM Auto-Power-Down Enable */
-#define MDREFR_K2DB2	(1 << 19)	/* SDCLK2 Divide by 2 Control/Status */
-#define MDREFR_K2RUN	(1 << 18)	/* SDCLK2 Run Control/Status */
-#define MDREFR_K1DB2	(1 << 17)	/* SDCLK1 Divide by 2 Control/Status */
-#define MDREFR_K1RUN	(1 << 16)	/* SDCLK1 Run Control/Status */
-#define MDREFR_E1PIN	(1 << 15)	/* SDCKE1 Level Control/Status */
-#define MDREFR_K0DB2	(1 << 14)	/* SDCLK0 Divide by 2 Control/Status */
-#define MDREFR_K0RUN	(1 << 13)	/* SDCLK0 Run Control/Status */
-#define MDREFR_E0PIN	(1 << 12)	/* SDCKE0 Level Control/Status */
-
-
 #ifdef CONFIG_PXA27x
 
-#define ARB_CNTRL	__REG(0x48000048)  /* Arbiter Control Register */
-
-#define ARB_DMA_SLV_PARK	(1<<31)	   /* Be parked with DMA slave when idle */
-#define ARB_CI_PARK		(1<<30)	   /* Be parked with Camera Interface when idle */
-#define ARB_EX_MEM_PARK 	(1<<29)	   /* Be parked with external MEMC when idle */
-#define ARB_INT_MEM_PARK	(1<<28)	   /* Be parked with internal MEMC when idle */
-#define ARB_USB_PARK		(1<<27)	   /* Be parked with USB when idle */
-#define ARB_LCD_PARK		(1<<26)	   /* Be parked with LCD when idle */
-#define ARB_DMA_PARK		(1<<25)	   /* Be parked with DMA when idle */
-#define ARB_CORE_PARK		(1<<24)	   /* Be parked with core when idle */
-#define ARB_LOCK_FLAG		(1<<23)	   /* Only Locking masters gain access to the bus */
-
 /*
  * Keypad
  */
diff --git a/include/asm-arm/arch-pxa/pxa2xx-regs.h b/include/asm-arm/arch-pxa/pxa2xx-regs.h
new file mode 100644
index 000000000000..9553b54fa5bc
--- /dev/null
+++ b/include/asm-arm/arch-pxa/pxa2xx-regs.h
@@ -0,0 +1,84 @@
+/*
+ *  linux/include/asm-arm/arch-pxa/pxa2xx-regs.h
+ *
+ *  Taken from pxa-regs.h by Russell King
+ *
+ *  Author:	Nicolas Pitre
+ *  Copyright:	MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __PXA2XX_REGS_H
+#define __PXA2XX_REGS_H
+
+/*
+ * Memory controller
+ */
+
+#define MDCNFG		__REG(0x48000000)  /* SDRAM Configuration Register 0 */
+#define MDREFR		__REG(0x48000004)  /* SDRAM Refresh Control Register */
+#define MSC0		__REG(0x48000008)  /* Static Memory Control Register 0 */
+#define MSC1		__REG(0x4800000C)  /* Static Memory Control Register 1 */
+#define MSC2		__REG(0x48000010)  /* Static Memory Control Register 2 */
+#define MECR		__REG(0x48000014)  /* Expansion Memory (PCMCIA/Compact Flash) Bus Configuration */
+#define SXLCR		__REG(0x48000018)  /* LCR value to be written to SDRAM-Timing Synchronous Flash */
+#define SXCNFG		__REG(0x4800001C)  /* Synchronous Static Memory Control Register */
+#define SXMRS		__REG(0x48000024)  /* MRS value to be written to Synchronous Flash or SMROM */
+#define MCMEM0		__REG(0x48000028)  /* Card interface Common Memory Space Socket 0 Timing */
+#define MCMEM1		__REG(0x4800002C)  /* Card interface Common Memory Space Socket 1 Timing */
+#define MCATT0		__REG(0x48000030)  /* Card interface Attribute Space Socket 0 Timing Configuration */
+#define MCATT1		__REG(0x48000034)  /* Card interface Attribute Space Socket 1 Timing Configuration */
+#define MCIO0		__REG(0x48000038)  /* Card interface I/O Space Socket 0 Timing Configuration */
+#define MCIO1		__REG(0x4800003C)  /* Card interface I/O Space Socket 1 Timing Configuration */
+#define MDMRS		__REG(0x48000040)  /* MRS value to be written to SDRAM */
+#define BOOT_DEF	__REG(0x48000044)  /* Read-Only Boot-Time Register. Contains BOOT_SEL and PKG_SEL */
+
+/*
+ * More handy macros for PCMCIA
+ *
+ * Arg is socket number
+ */
+#define MCMEM(s)	__REG2(0x48000028, (s)<<2 )  /* Card interface Common Memory Space Socket s Timing */
+#define MCATT(s)	__REG2(0x48000030, (s)<<2 )  /* Card interface Attribute Space Socket s Timing Configuration */
+#define MCIO(s)		__REG2(0x48000038, (s)<<2 )  /* Card interface I/O Space Socket s Timing Configuration */
+
+/* MECR register defines */
+#define MECR_NOS	(1 << 0)	/* Number Of Sockets: 0 -> 1 sock, 1 -> 2 sock */
+#define MECR_CIT	(1 << 1)	/* Card Is There: 0 -> no card, 1 -> card inserted */
+
+#define MDREFR_K0DB4	(1 << 29)	/* SDCLK0 Divide by 4 Control/Status */
+#define MDREFR_K2FREE	(1 << 25)	/* SDRAM Free-Running Control */
+#define MDREFR_K1FREE	(1 << 24)	/* SDRAM Free-Running Control */
+#define MDREFR_K0FREE	(1 << 23)	/* SDRAM Free-Running Control */
+#define MDREFR_SLFRSH	(1 << 22)	/* SDRAM Self-Refresh Control/Status */
+#define MDREFR_APD	(1 << 20)	/* SDRAM/SSRAM Auto-Power-Down Enable */
+#define MDREFR_K2DB2	(1 << 19)	/* SDCLK2 Divide by 2 Control/Status */
+#define MDREFR_K2RUN	(1 << 18)	/* SDCLK2 Run Control/Status */
+#define MDREFR_K1DB2	(1 << 17)	/* SDCLK1 Divide by 2 Control/Status */
+#define MDREFR_K1RUN	(1 << 16)	/* SDCLK1 Run Control/Status */
+#define MDREFR_E1PIN	(1 << 15)	/* SDCKE1 Level Control/Status */
+#define MDREFR_K0DB2	(1 << 14)	/* SDCLK0 Divide by 2 Control/Status */
+#define MDREFR_K0RUN	(1 << 13)	/* SDCLK0 Run Control/Status */
+#define MDREFR_E0PIN	(1 << 12)	/* SDCKE0 Level Control/Status */
+
+
+#ifdef CONFIG_PXA27x
+
+#define ARB_CNTRL	__REG(0x48000048)  /* Arbiter Control Register */
+
+#define ARB_DMA_SLV_PARK	(1<<31)	   /* Be parked with DMA slave when idle */
+#define ARB_CI_PARK		(1<<30)	   /* Be parked with Camera Interface when idle */
+#define ARB_EX_MEM_PARK 	(1<<29)	   /* Be parked with external MEMC when idle */
+#define ARB_INT_MEM_PARK	(1<<28)	   /* Be parked with internal MEMC when idle */
+#define ARB_USB_PARK		(1<<27)	   /* Be parked with USB when idle */
+#define ARB_LCD_PARK		(1<<26)	   /* Be parked with LCD when idle */
+#define ARB_DMA_PARK		(1<<25)	   /* Be parked with DMA when idle */
+#define ARB_CORE_PARK		(1<<24)	   /* Be parked with core when idle */
+#define ARB_LOCK_FLAG		(1<<23)	   /* Only Locking masters gain access to the bus */
+
+#endif
+
+#endif
-- 
cgit v1.2.3


From 83b6759e72a83c344c7fe96a37543b4546f17131 Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Wed, 2 Jan 2008 16:45:06 +0800
Subject: [ARM] pxa: introduce a new file mfp-pxa3xx.h for common pin
 configurations

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-pxa/mfp-pxa300.h |   1 +
 include/asm-arm/arch-pxa/mfp-pxa320.h |   1 +
 include/asm-arm/arch-pxa/mfp-pxa3xx.h | 137 ++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-pxa/mfp.h        | 133 ---------------------------------
 4 files changed, 139 insertions(+), 133 deletions(-)
 create mode 100644 include/asm-arm/arch-pxa/mfp-pxa3xx.h

(limited to 'include')

diff --git a/include/asm-arm/arch-pxa/mfp-pxa300.h b/include/asm-arm/arch-pxa/mfp-pxa300.h
index a20996649889..bb410313556f 100644
--- a/include/asm-arm/arch-pxa/mfp-pxa300.h
+++ b/include/asm-arm/arch-pxa/mfp-pxa300.h
@@ -16,6 +16,7 @@
 #define __ASM_ARCH_MFP_PXA300_H
 
 #include <asm/arch/mfp.h>
+#include <asm/arch/mfp-pxa3xx.h>
 
 /* GPIO */
 #define GPIO46_GPIO		MFP_CFG(GPIO46, AF1)
diff --git a/include/asm-arm/arch-pxa/mfp-pxa320.h b/include/asm-arm/arch-pxa/mfp-pxa320.h
index 52deedcaf3bd..576aa46d90fc 100644
--- a/include/asm-arm/arch-pxa/mfp-pxa320.h
+++ b/include/asm-arm/arch-pxa/mfp-pxa320.h
@@ -16,6 +16,7 @@
 #define __ASM_ARCH_MFP_PXA320_H
 
 #include <asm/arch/mfp.h>
+#include <asm/arch/mfp-pxa3xx.h>
 
 /* GPIO */
 #define GPIO46_GPIO		MFP_CFG(GPIO46, AF0)
diff --git a/include/asm-arm/arch-pxa/mfp-pxa3xx.h b/include/asm-arm/arch-pxa/mfp-pxa3xx.h
new file mode 100644
index 000000000000..5da1857d5476
--- /dev/null
+++ b/include/asm-arm/arch-pxa/mfp-pxa3xx.h
@@ -0,0 +1,137 @@
+#ifndef __ASM_ARCH_MFP_PXA3XX_H
+#define __ASM_ARCH_MFP_PXA3XX_H
+
+/* PXA3xx common MFP configurations - processor specific ones defined
+ * in mfp-pxa300.h and mfp-pxa320.h
+ */
+#define GPIO0_GPIO		MFP_CFG(GPIO0, AF0)
+#define GPIO1_GPIO		MFP_CFG(GPIO1, AF0)
+#define GPIO2_GPIO		MFP_CFG(GPIO2, AF0)
+#define GPIO3_GPIO		MFP_CFG(GPIO3, AF0)
+#define GPIO4_GPIO		MFP_CFG(GPIO4, AF0)
+#define GPIO5_GPIO		MFP_CFG(GPIO5, AF0)
+#define GPIO6_GPIO		MFP_CFG(GPIO6, AF0)
+#define GPIO7_GPIO		MFP_CFG(GPIO7, AF0)
+#define GPIO8_GPIO		MFP_CFG(GPIO8, AF0)
+#define GPIO9_GPIO		MFP_CFG(GPIO9, AF0)
+#define GPIO10_GPIO		MFP_CFG(GPIO10, AF0)
+#define GPIO11_GPIO		MFP_CFG(GPIO11, AF0)
+#define GPIO12_GPIO		MFP_CFG(GPIO12, AF0)
+#define GPIO13_GPIO		MFP_CFG(GPIO13, AF0)
+#define GPIO14_GPIO		MFP_CFG(GPIO14, AF0)
+#define GPIO15_GPIO		MFP_CFG(GPIO15, AF0)
+#define GPIO16_GPIO		MFP_CFG(GPIO16, AF0)
+#define GPIO17_GPIO		MFP_CFG(GPIO17, AF0)
+#define GPIO18_GPIO		MFP_CFG(GPIO18, AF0)
+#define GPIO19_GPIO		MFP_CFG(GPIO19, AF0)
+#define GPIO20_GPIO		MFP_CFG(GPIO20, AF0)
+#define GPIO21_GPIO		MFP_CFG(GPIO21, AF0)
+#define GPIO22_GPIO		MFP_CFG(GPIO22, AF0)
+#define GPIO23_GPIO		MFP_CFG(GPIO23, AF0)
+#define GPIO24_GPIO		MFP_CFG(GPIO24, AF0)
+#define GPIO25_GPIO		MFP_CFG(GPIO25, AF0)
+#define GPIO26_GPIO		MFP_CFG(GPIO26, AF0)
+#define GPIO27_GPIO		MFP_CFG(GPIO27, AF0)
+#define GPIO28_GPIO		MFP_CFG(GPIO28, AF0)
+#define GPIO29_GPIO		MFP_CFG(GPIO29, AF0)
+#define GPIO30_GPIO		MFP_CFG(GPIO30, AF0)
+#define GPIO31_GPIO		MFP_CFG(GPIO31, AF0)
+#define GPIO32_GPIO		MFP_CFG(GPIO32, AF0)
+#define GPIO33_GPIO		MFP_CFG(GPIO33, AF0)
+#define GPIO34_GPIO		MFP_CFG(GPIO34, AF0)
+#define GPIO35_GPIO		MFP_CFG(GPIO35, AF0)
+#define GPIO36_GPIO		MFP_CFG(GPIO36, AF0)
+#define GPIO37_GPIO		MFP_CFG(GPIO37, AF0)
+#define GPIO38_GPIO		MFP_CFG(GPIO38, AF0)
+#define GPIO39_GPIO		MFP_CFG(GPIO39, AF0)
+#define GPIO40_GPIO		MFP_CFG(GPIO40, AF0)
+#define GPIO41_GPIO		MFP_CFG(GPIO41, AF0)
+#define GPIO42_GPIO		MFP_CFG(GPIO42, AF0)
+#define GPIO43_GPIO		MFP_CFG(GPIO43, AF0)
+#define GPIO44_GPIO		MFP_CFG(GPIO44, AF0)
+#define GPIO45_GPIO		MFP_CFG(GPIO45, AF0)
+
+#define GPIO47_GPIO		MFP_CFG(GPIO47, AF0)
+#define GPIO48_GPIO		MFP_CFG(GPIO48, AF0)
+
+#define GPIO53_GPIO		MFP_CFG(GPIO53, AF0)
+#define GPIO54_GPIO		MFP_CFG(GPIO54, AF0)
+#define GPIO55_GPIO		MFP_CFG(GPIO55, AF0)
+
+#define GPIO57_GPIO		MFP_CFG(GPIO57, AF0)
+
+#define GPIO63_GPIO		MFP_CFG(GPIO63, AF0)
+#define GPIO64_GPIO		MFP_CFG(GPIO64, AF0)
+#define GPIO65_GPIO		MFP_CFG(GPIO65, AF0)
+#define GPIO66_GPIO		MFP_CFG(GPIO66, AF0)
+#define GPIO67_GPIO		MFP_CFG(GPIO67, AF0)
+#define GPIO68_GPIO		MFP_CFG(GPIO68, AF0)
+#define GPIO69_GPIO		MFP_CFG(GPIO69, AF0)
+#define GPIO70_GPIO		MFP_CFG(GPIO70, AF0)
+#define GPIO71_GPIO		MFP_CFG(GPIO71, AF0)
+#define GPIO72_GPIO		MFP_CFG(GPIO72, AF0)
+#define GPIO73_GPIO		MFP_CFG(GPIO73, AF0)
+#define GPIO74_GPIO		MFP_CFG(GPIO74, AF0)
+#define GPIO75_GPIO		MFP_CFG(GPIO75, AF0)
+#define GPIO76_GPIO		MFP_CFG(GPIO76, AF0)
+#define GPIO77_GPIO		MFP_CFG(GPIO77, AF0)
+#define GPIO78_GPIO		MFP_CFG(GPIO78, AF0)
+#define GPIO79_GPIO		MFP_CFG(GPIO79, AF0)
+#define GPIO80_GPIO		MFP_CFG(GPIO80, AF0)
+#define GPIO81_GPIO		MFP_CFG(GPIO81, AF0)
+#define GPIO82_GPIO		MFP_CFG(GPIO82, AF0)
+#define GPIO83_GPIO		MFP_CFG(GPIO83, AF0)
+#define GPIO84_GPIO		MFP_CFG(GPIO84, AF0)
+#define GPIO85_GPIO		MFP_CFG(GPIO85, AF0)
+#define GPIO86_GPIO		MFP_CFG(GPIO86, AF0)
+#define GPIO87_GPIO		MFP_CFG(GPIO87, AF0)
+#define GPIO88_GPIO		MFP_CFG(GPIO88, AF0)
+#define GPIO89_GPIO		MFP_CFG(GPIO89, AF0)
+#define GPIO90_GPIO		MFP_CFG(GPIO90, AF0)
+#define GPIO91_GPIO		MFP_CFG(GPIO91, AF0)
+#define GPIO92_GPIO		MFP_CFG(GPIO92, AF0)
+#define GPIO93_GPIO		MFP_CFG(GPIO93, AF0)
+#define GPIO94_GPIO		MFP_CFG(GPIO94, AF0)
+#define GPIO95_GPIO		MFP_CFG(GPIO95, AF0)
+#define GPIO96_GPIO		MFP_CFG(GPIO96, AF0)
+#define GPIO97_GPIO		MFP_CFG(GPIO97, AF0)
+#define GPIO98_GPIO		MFP_CFG(GPIO98, AF0)
+#define GPIO99_GPIO		MFP_CFG(GPIO99, AF0)
+#define GPIO100_GPIO		MFP_CFG(GPIO100, AF0)
+#define GPIO101_GPIO		MFP_CFG(GPIO101, AF0)
+#define GPIO102_GPIO		MFP_CFG(GPIO102, AF0)
+#define GPIO103_GPIO		MFP_CFG(GPIO103, AF0)
+#define GPIO104_GPIO		MFP_CFG(GPIO104, AF0)
+#define GPIO105_GPIO		MFP_CFG(GPIO105, AF0)
+#define GPIO106_GPIO		MFP_CFG(GPIO106, AF0)
+#define GPIO107_GPIO		MFP_CFG(GPIO107, AF0)
+#define GPIO108_GPIO		MFP_CFG(GPIO108, AF0)
+#define GPIO109_GPIO		MFP_CFG(GPIO109, AF0)
+#define GPIO110_GPIO		MFP_CFG(GPIO110, AF0)
+#define GPIO111_GPIO		MFP_CFG(GPIO111, AF0)
+#define GPIO112_GPIO		MFP_CFG(GPIO112, AF0)
+#define GPIO113_GPIO		MFP_CFG(GPIO113, AF0)
+#define GPIO114_GPIO		MFP_CFG(GPIO114, AF0)
+#define GPIO115_GPIO		MFP_CFG(GPIO115, AF0)
+#define GPIO116_GPIO		MFP_CFG(GPIO116, AF0)
+#define GPIO117_GPIO		MFP_CFG(GPIO117, AF0)
+#define GPIO118_GPIO		MFP_CFG(GPIO118, AF0)
+#define GPIO119_GPIO		MFP_CFG(GPIO119, AF0)
+#define GPIO120_GPIO		MFP_CFG(GPIO120, AF0)
+#define GPIO121_GPIO		MFP_CFG(GPIO121, AF0)
+#define GPIO122_GPIO		MFP_CFG(GPIO122, AF0)
+#define GPIO123_GPIO		MFP_CFG(GPIO123, AF0)
+#define GPIO124_GPIO		MFP_CFG(GPIO124, AF0)
+#define GPIO125_GPIO		MFP_CFG(GPIO125, AF0)
+#define GPIO126_GPIO		MFP_CFG(GPIO126, AF0)
+#define GPIO127_GPIO		MFP_CFG(GPIO127, AF0)
+
+#define GPIO0_2_GPIO		MFP_CFG(GPIO0_2, AF0)
+#define GPIO1_2_GPIO		MFP_CFG(GPIO1_2, AF0)
+#define GPIO2_2_GPIO		MFP_CFG(GPIO2_2, AF0)
+#define GPIO3_2_GPIO		MFP_CFG(GPIO3_2, AF0)
+#define GPIO4_2_GPIO		MFP_CFG(GPIO4_2, AF0)
+#define GPIO5_2_GPIO		MFP_CFG(GPIO5_2, AF0)
+#define GPIO6_2_GPIO		MFP_CFG(GPIO6_2, AF0)
+
+#endif /* __ASM_ARCH_MFP_PXA3XX_H */
diff --git a/include/asm-arm/arch-pxa/mfp.h b/include/asm-arm/arch-pxa/mfp.h
index 03c508d94f0e..7de234b0cca7 100644
--- a/include/asm-arm/arch-pxa/mfp.h
+++ b/include/asm-arm/arch-pxa/mfp.h
@@ -376,139 +376,6 @@ typedef uint32_t mfp_cfg_t;
 	 (((MFP_LPM_##lpm) & 0x4) << 12) |\
 	 (((MFP_LPM_##lpm) & 0x8) << 10))
 
-/* common MFP configurations - processor specific ones defined
- * in mfp-pxa3xx.h
- */
-#define GPIO0_GPIO		MFP_CFG(GPIO0, AF0)
-#define GPIO1_GPIO		MFP_CFG(GPIO1, AF0)
-#define GPIO2_GPIO		MFP_CFG(GPIO2, AF0)
-#define GPIO3_GPIO		MFP_CFG(GPIO3, AF0)
-#define GPIO4_GPIO		MFP_CFG(GPIO4, AF0)
-#define GPIO5_GPIO		MFP_CFG(GPIO5, AF0)
-#define GPIO6_GPIO		MFP_CFG(GPIO6, AF0)
-#define GPIO7_GPIO		MFP_CFG(GPIO7, AF0)
-#define GPIO8_GPIO		MFP_CFG(GPIO8, AF0)
-#define GPIO9_GPIO		MFP_CFG(GPIO9, AF0)
-#define GPIO10_GPIO		MFP_CFG(GPIO10, AF0)
-#define GPIO11_GPIO		MFP_CFG(GPIO11, AF0)
-#define GPIO12_GPIO		MFP_CFG(GPIO12, AF0)
-#define GPIO13_GPIO		MFP_CFG(GPIO13, AF0)
-#define GPIO14_GPIO		MFP_CFG(GPIO14, AF0)
-#define GPIO15_GPIO		MFP_CFG(GPIO15, AF0)
-#define GPIO16_GPIO		MFP_CFG(GPIO16, AF0)
-#define GPIO17_GPIO		MFP_CFG(GPIO17, AF0)
-#define GPIO18_GPIO		MFP_CFG(GPIO18, AF0)
-#define GPIO19_GPIO		MFP_CFG(GPIO19, AF0)
-#define GPIO20_GPIO		MFP_CFG(GPIO20, AF0)
-#define GPIO21_GPIO		MFP_CFG(GPIO21, AF0)
-#define GPIO22_GPIO		MFP_CFG(GPIO22, AF0)
-#define GPIO23_GPIO		MFP_CFG(GPIO23, AF0)
-#define GPIO24_GPIO		MFP_CFG(GPIO24, AF0)
-#define GPIO25_GPIO		MFP_CFG(GPIO25, AF0)
-#define GPIO26_GPIO		MFP_CFG(GPIO26, AF0)
-#define GPIO27_GPIO		MFP_CFG(GPIO27, AF0)
-#define GPIO28_GPIO		MFP_CFG(GPIO28, AF0)
-#define GPIO29_GPIO		MFP_CFG(GPIO29, AF0)
-#define GPIO30_GPIO		MFP_CFG(GPIO30, AF0)
-#define GPIO31_GPIO		MFP_CFG(GPIO31, AF0)
-#define GPIO32_GPIO		MFP_CFG(GPIO32, AF0)
-#define GPIO33_GPIO		MFP_CFG(GPIO33, AF0)
-#define GPIO34_GPIO		MFP_CFG(GPIO34, AF0)
-#define GPIO35_GPIO		MFP_CFG(GPIO35, AF0)
-#define GPIO36_GPIO		MFP_CFG(GPIO36, AF0)
-#define GPIO37_GPIO		MFP_CFG(GPIO37, AF0)
-#define GPIO38_GPIO		MFP_CFG(GPIO38, AF0)
-#define GPIO39_GPIO		MFP_CFG(GPIO39, AF0)
-#define GPIO40_GPIO		MFP_CFG(GPIO40, AF0)
-#define GPIO41_GPIO		MFP_CFG(GPIO41, AF0)
-#define GPIO42_GPIO		MFP_CFG(GPIO42, AF0)
-#define GPIO43_GPIO		MFP_CFG(GPIO43, AF0)
-#define GPIO44_GPIO		MFP_CFG(GPIO44, AF0)
-#define GPIO45_GPIO		MFP_CFG(GPIO45, AF0)
-
-#define GPIO47_GPIO		MFP_CFG(GPIO47, AF0)
-#define GPIO48_GPIO		MFP_CFG(GPIO48, AF0)
-
-#define GPIO53_GPIO		MFP_CFG(GPIO53, AF0)
-#define GPIO54_GPIO		MFP_CFG(GPIO54, AF0)
-#define GPIO55_GPIO		MFP_CFG(GPIO55, AF0)
-
-#define GPIO57_GPIO		MFP_CFG(GPIO57, AF0)
-
-#define GPIO63_GPIO		MFP_CFG(GPIO63, AF0)
-#define GPIO64_GPIO		MFP_CFG(GPIO64, AF0)
-#define GPIO65_GPIO		MFP_CFG(GPIO65, AF0)
-#define GPIO66_GPIO		MFP_CFG(GPIO66, AF0)
-#define GPIO67_GPIO		MFP_CFG(GPIO67, AF0)
-#define GPIO68_GPIO		MFP_CFG(GPIO68, AF0)
-#define GPIO69_GPIO		MFP_CFG(GPIO69, AF0)
-#define GPIO70_GPIO		MFP_CFG(GPIO70, AF0)
-#define GPIO71_GPIO		MFP_CFG(GPIO71, AF0)
-#define GPIO72_GPIO		MFP_CFG(GPIO72, AF0)
-#define GPIO73_GPIO		MFP_CFG(GPIO73, AF0)
-#define GPIO74_GPIO		MFP_CFG(GPIO74, AF0)
-#define GPIO75_GPIO		MFP_CFG(GPIO75, AF0)
-#define GPIO76_GPIO		MFP_CFG(GPIO76, AF0)
-#define GPIO77_GPIO		MFP_CFG(GPIO77, AF0)
-#define GPIO78_GPIO		MFP_CFG(GPIO78, AF0)
-#define GPIO79_GPIO		MFP_CFG(GPIO79, AF0)
-#define GPIO80_GPIO		MFP_CFG(GPIO80, AF0)
-#define GPIO81_GPIO		MFP_CFG(GPIO81, AF0)
-#define GPIO82_GPIO		MFP_CFG(GPIO82, AF0)
-#define GPIO83_GPIO		MFP_CFG(GPIO83, AF0)
-#define GPIO84_GPIO		MFP_CFG(GPIO84, AF0)
-#define GPIO85_GPIO		MFP_CFG(GPIO85, AF0)
-#define GPIO86_GPIO		MFP_CFG(GPIO86, AF0)
-#define GPIO87_GPIO		MFP_CFG(GPIO87, AF0)
-#define GPIO88_GPIO		MFP_CFG(GPIO88, AF0)
-#define GPIO89_GPIO		MFP_CFG(GPIO89, AF0)
-#define GPIO90_GPIO		MFP_CFG(GPIO90, AF0)
-#define GPIO91_GPIO		MFP_CFG(GPIO91, AF0)
-#define GPIO92_GPIO		MFP_CFG(GPIO92, AF0)
-#define GPIO93_GPIO		MFP_CFG(GPIO93, AF0)
-#define GPIO94_GPIO		MFP_CFG(GPIO94, AF0)
-#define GPIO95_GPIO		MFP_CFG(GPIO95, AF0)
-#define GPIO96_GPIO		MFP_CFG(GPIO96, AF0)
-#define GPIO97_GPIO		MFP_CFG(GPIO97, AF0)
-#define GPIO98_GPIO		MFP_CFG(GPIO98, AF0)
-#define GPIO99_GPIO		MFP_CFG(GPIO99, AF0)
-#define GPIO100_GPIO		MFP_CFG(GPIO100, AF0)
-#define GPIO101_GPIO		MFP_CFG(GPIO101, AF0)
-#define GPIO102_GPIO		MFP_CFG(GPIO102, AF0)
-#define GPIO103_GPIO		MFP_CFG(GPIO103, AF0)
-#define GPIO104_GPIO		MFP_CFG(GPIO104, AF0)
-#define GPIO105_GPIO		MFP_CFG(GPIO105, AF0)
-#define GPIO106_GPIO		MFP_CFG(GPIO106, AF0)
-#define GPIO107_GPIO		MFP_CFG(GPIO107, AF0)
-#define GPIO108_GPIO		MFP_CFG(GPIO108, AF0)
-#define GPIO109_GPIO		MFP_CFG(GPIO109, AF0)
-#define GPIO110_GPIO		MFP_CFG(GPIO110, AF0)
-#define GPIO111_GPIO		MFP_CFG(GPIO111, AF0)
-#define GPIO112_GPIO		MFP_CFG(GPIO112, AF0)
-#define GPIO113_GPIO		MFP_CFG(GPIO113, AF0)
-#define GPIO114_GPIO		MFP_CFG(GPIO114, AF0)
-#define GPIO115_GPIO		MFP_CFG(GPIO115, AF0)
-#define GPIO116_GPIO		MFP_CFG(GPIO116, AF0)
-#define GPIO117_GPIO		MFP_CFG(GPIO117, AF0)
-#define GPIO118_GPIO		MFP_CFG(GPIO118, AF0)
-#define GPIO119_GPIO		MFP_CFG(GPIO119, AF0)
-#define GPIO120_GPIO		MFP_CFG(GPIO120, AF0)
-#define GPIO121_GPIO		MFP_CFG(GPIO121, AF0)
-#define GPIO122_GPIO		MFP_CFG(GPIO122, AF0)
-#define GPIO123_GPIO		MFP_CFG(GPIO123, AF0)
-#define GPIO124_GPIO		MFP_CFG(GPIO124, AF0)
-#define GPIO125_GPIO		MFP_CFG(GPIO125, AF0)
-#define GPIO126_GPIO		MFP_CFG(GPIO126, AF0)
-#define GPIO127_GPIO		MFP_CFG(GPIO127, AF0)
-
-#define GPIO0_2_GPIO		MFP_CFG(GPIO0_2, AF0)
-#define GPIO1_2_GPIO		MFP_CFG(GPIO1_2, AF0)
-#define GPIO2_2_GPIO		MFP_CFG(GPIO2_2, AF0)
-#define GPIO3_2_GPIO		MFP_CFG(GPIO3_2, AF0)
-#define GPIO4_2_GPIO		MFP_CFG(GPIO4_2, AF0)
-#define GPIO5_2_GPIO		MFP_CFG(GPIO5_2, AF0)
-#define GPIO6_2_GPIO		MFP_CFG(GPIO6_2, AF0)
-
 /*
  * each MFP pin will have a MFPR register, since the offset of the
  * register varies between processors, the processor specific code
-- 
cgit v1.2.3


From 0ad1fbc86045c2a27ff082c02344131be072699f Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Wed, 2 Jan 2008 17:34:27 +0800
Subject: [ARM] pxa: remove un-used pxa3xx_mfp_set_xxx() functions

pxa3xx_mfp_set_xxx() functions are originally provided for overwriting
MFP configurations performed by pxa3xx_mfp_config(), the usage of such
a dirtry trick is not recommended, since there is currently no user of
these functions, they are safely removed

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/mfp.c        | 111 -----------------------------------------
 include/asm-arm/arch-pxa/mfp.h |  14 ------
 2 files changed, 125 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/mfp.c b/arch/arm/mach-pxa/mfp.c
index 436f96574964..785ae0ddb84f 100644
--- a/arch/arm/mach-pxa/mfp.c
+++ b/arch/arm/mach-pxa/mfp.c
@@ -96,117 +96,6 @@ void pxa3xx_mfp_write(int mfp, unsigned long val)
 	spin_unlock_irqrestore(&mfp_spin_lock, flags);
 }
 
-void pxa3xx_mfp_set_afds(int mfp, int af, int ds)
-{
-	uint32_t mfpr_off, mfpr_val;
-	unsigned long flags;
-
-	BUG_ON(mfp >= MFP_PIN_MAX);
-
-	spin_lock_irqsave(&mfp_spin_lock, flags);
-	mfpr_off = mfp_table[mfp].mfpr_off;
-
-	mfpr_val = mfpr_readl(mfpr_off);
-	mfpr_val &= ~(MFPR_AF_MASK | MFPR_DRV_MASK);
-	mfpr_val |= (((af & 0x7) << MFPR_ALT_OFFSET) |
-		     ((ds & 0x7) << MFPR_DRV_OFFSET));
-
-	mfpr_writel(mfpr_off, mfpr_val);
-	mfpr_sync();
-
-	spin_unlock_irqrestore(&mfp_spin_lock, flags);
-}
-
-void pxa3xx_mfp_set_rdh(int mfp, int rdh)
-{
-	uint32_t mfpr_off, mfpr_val;
-	unsigned long flags;
-
-	BUG_ON(mfp >= MFP_PIN_MAX);
-
-	spin_lock_irqsave(&mfp_spin_lock, flags);
-
-	mfpr_off = mfp_table[mfp].mfpr_off;
-
-	mfpr_val = mfpr_readl(mfpr_off);
-	mfpr_val &= ~MFPR_RDH_MASK;
-
-	if (likely(rdh))
-		mfpr_val |= (1u << MFPR_SS_OFFSET);
-
-	mfpr_writel(mfpr_off, mfpr_val);
-	mfpr_sync();
-
-	spin_unlock_irqrestore(&mfp_spin_lock, flags);
-}
-
-void pxa3xx_mfp_set_lpm(int mfp, int lpm)
-{
-	uint32_t mfpr_off, mfpr_val;
-	unsigned long flags;
-
-	BUG_ON(mfp >= MFP_PIN_MAX);
-
-	spin_lock_irqsave(&mfp_spin_lock, flags);
-
-	mfpr_off = mfp_table[mfp].mfpr_off;
-	mfpr_val = mfpr_readl(mfpr_off);
-	mfpr_val &= ~MFPR_LPM_MASK;
-
-	if (lpm & 0x1) mfpr_val |= 1u << MFPR_SON_OFFSET;
-	if (lpm & 0x2) mfpr_val |= 1u << MFPR_SD_OFFSET;
-	if (lpm & 0x4) mfpr_val |= 1u << MFPR_PU_OFFSET;
-	if (lpm & 0x8) mfpr_val |= 1u << MFPR_PD_OFFSET;
-	if (lpm &0x10) mfpr_val |= 1u << MFPR_PS_OFFSET;
-
-	mfpr_writel(mfpr_off, mfpr_val);
-	mfpr_sync();
-
-	spin_unlock_irqrestore(&mfp_spin_lock, flags);
-}
-
-void pxa3xx_mfp_set_pull(int mfp, int pull)
-{
-	uint32_t mfpr_off, mfpr_val;
-	unsigned long flags;
-
-	BUG_ON(mfp >= MFP_PIN_MAX);
-
-	spin_lock_irqsave(&mfp_spin_lock, flags);
-
-	mfpr_off = mfp_table[mfp].mfpr_off;
-	mfpr_val = mfpr_readl(mfpr_off);
-	mfpr_val &= ~MFPR_PULL_MASK;
-	mfpr_val |= ((pull & 0x7u) << MFPR_PD_OFFSET);
-
-	mfpr_writel(mfpr_off, mfpr_val);
-	mfpr_sync();
-
-	spin_unlock_irqrestore(&mfp_spin_lock, flags);
-}
-
-void pxa3xx_mfp_set_edge(int mfp, int edge)
-{
-	uint32_t mfpr_off, mfpr_val;
-	unsigned long flags;
-
-	BUG_ON(mfp >= MFP_PIN_MAX);
-
-	spin_lock_irqsave(&mfp_spin_lock, flags);
-
-	mfpr_off = mfp_table[mfp].mfpr_off;
-	mfpr_val = mfpr_readl(mfpr_off);
-
-	mfpr_val &= ~MFPR_EDGE_MASK;
-	mfpr_val |= (edge & 0x3u) << MFPR_ERE_OFFSET;
-	mfpr_val |= (!edge & 0x1) << MFPR_EC_OFFSET;
-
-	mfpr_writel(mfpr_off, mfpr_val);
-	mfpr_sync();
-
-	spin_unlock_irqrestore(&mfp_spin_lock, flags);
-}
-
 void __init pxa3xx_mfp_init_addr(struct pxa3xx_mfp_addr_map *map)
 {
 	struct pxa3xx_mfp_addr_map *p;
diff --git a/include/asm-arm/arch-pxa/mfp.h b/include/asm-arm/arch-pxa/mfp.h
index 7de234b0cca7..60fbed1b2e89 100644
--- a/include/asm-arm/arch-pxa/mfp.h
+++ b/include/asm-arm/arch-pxa/mfp.h
@@ -418,20 +418,6 @@ struct pxa3xx_mfp_pin {
 unsigned long pxa3xx_mfp_read(int mfp);
 void pxa3xx_mfp_write(int mfp, unsigned long mfpr_val);
 
-/*
- * pxa3xx_mfp_set_afds - set MFP alternate function and drive strength
- * pxa3xx_mfp_set_rdh  - set MFP release delay hold on/off
- * pxa3xx_mfp_set_lpm  - set MFP low power mode state
- * pxa3xx_mfp_set_edge - set MFP edge detection in low power mode
- *
- * use these functions to override/change the default configuration
- * done by pxa3xx_mfp_set_config(s)
- */
-void pxa3xx_mfp_set_afds(int mfp, int af, int ds);
-void pxa3xx_mfp_set_rdh(int mfp, int rdh);
-void pxa3xx_mfp_set_lpm(int mfp, int lpm);
-void pxa3xx_mfp_set_edge(int mfp, int edge);
-
 /*
  * pxa3xx_mfp_config - configure the MFPR registers
  *
-- 
cgit v1.2.3


From 7f7c8a619253c83cf3b1071df3b001811d0c1a6c Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Thu, 3 Jan 2008 11:25:56 +0800
Subject: [ARM] pxa: make MFP configuration processor independent

There are two reasons for making the MFP configuration to be processor
independent, i.e. removing the relationship of configuration bits with
actual MFPR register settings:

   1. power management sometimes requires the MFP to be configured
      differently when in run mode or in low power mode

   2. for future integration of pxa{25x,27x} GPIO configurations

The modifications include:

1. introducing of processor independent MFP configuration bits, as
   defined in [include/asm-arm/arch-pxa/mfp.h]:

	bit  0.. 9 - MFP Pin Number (1024 Pins Maximum)
	bit 10..12 - Alternate Function Selection
	bit 13..15 - Drive Strength
	bit 16..18 - Low Power Mode State
	bit 19..20 - Low Power Mode Edge Detection
	bit 21..22 - Run Mode Pull State
	and so on,

2. moving the processor dependent code from mfp.h into mfp-pxa3xx.h

3. cleaning up of the MFPR bit definitions

4. mapping of processor independent MFP configuration into processor
   specific MFPR register settings is now totally encapsulated within
   pxa3xx_mfp_config()

5. using of "unsigned long" instead of invented type of "mfp_cfg_t"
   according to Documentation/CodingStyle Chapter 5, usage of this
   in platform code will be slowly removed in later patches

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/mfp.c               |  97 +++++++++++--
 include/asm-arm/arch-pxa/mfp-pxa3xx.h | 115 +++++++++++++++
 include/asm-arm/arch-pxa/mfp.h        | 262 +++++++++-------------------------
 3 files changed, 265 insertions(+), 209 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/mfp.c b/arch/arm/mach-pxa/mfp.c
index 785ae0ddb84f..c66b1cd7df64 100644
--- a/arch/arm/mach-pxa/mfp.c
+++ b/arch/arm/mach-pxa/mfp.c
@@ -20,6 +20,7 @@
 
 #include <asm/hardware.h>
 #include <asm/arch/mfp.h>
+#include <asm/arch/mfp-pxa3xx.h>
 
 /* mfp_spin_lock is used to ensure that MFP register configuration
  * (most likely a read-modify-write operation) is atomic, and that
@@ -28,43 +29,105 @@
 static DEFINE_SPINLOCK(mfp_spin_lock);
 
 static void __iomem *mfpr_mmio_base = (void __iomem *)&__REG(MFPR_BASE);
+
+struct pxa3xx_mfp_pin {
+	unsigned long	config;		/* -1 for not configured */
+	unsigned long	mfpr_off;	/* MFPRxx Register offset */
+	unsigned long	mfpr_run;	/* Run-Mode Register Value */
+	unsigned long	mfpr_lpm;	/* Low Power Mode Register Value */
+};
+
 static struct pxa3xx_mfp_pin mfp_table[MFP_PIN_MAX];
 
+/* mapping of MFP_LPM_* definitions to MFPR_LPM_* register bits */
+const static unsigned long mfpr_lpm[] = {
+	MFPR_LPM_INPUT,
+	MFPR_LPM_DRIVE_LOW,
+	MFPR_LPM_DRIVE_HIGH,
+	MFPR_LPM_PULL_LOW,
+	MFPR_LPM_PULL_HIGH,
+	MFPR_LPM_FLOAT,
+};
+
+/* mapping of MFP_PULL_* definitions to MFPR_PULL_* register bits */
+const static unsigned long mfpr_pull[] = {
+	MFPR_PULL_NONE,
+	MFPR_PULL_LOW,
+	MFPR_PULL_HIGH,
+	MFPR_PULL_BOTH,
+};
+
+/* mapping of MFP_LPM_EDGE_* definitions to MFPR_EDGE_* register bits */
+const static unsigned long mfpr_edge[] = {
+	MFPR_EDGE_NONE,
+	MFPR_EDGE_RISE,
+	MFPR_EDGE_FALL,
+	MFPR_EDGE_BOTH,
+};
+
 #define mfpr_readl(off)			\
 	__raw_readl(mfpr_mmio_base + (off))
 
 #define mfpr_writel(off, val)		\
 	__raw_writel(val, mfpr_mmio_base + (off))
 
+#define mfp_configured(p)	((p)->config != -1)
+
 /*
  * perform a read-back of any MFPR register to make sure the
  * previous writings are finished
  */
 #define mfpr_sync()	(void)__raw_readl(mfpr_mmio_base + 0)
 
-static inline void __mfp_config(int pin, unsigned long val)
+static inline void __mfp_config_run(struct pxa3xx_mfp_pin *p)
 {
-	unsigned long off = mfp_table[pin].mfpr_off;
+	if (mfp_configured(p))
+		mfpr_writel(p->mfpr_off, p->mfpr_run);
+}
 
-	mfp_table[pin].mfpr_val = val;
-	mfpr_writel(off, val);
+static inline void __mfp_config_lpm(struct pxa3xx_mfp_pin *p)
+{
+	if (mfp_configured(p) && p->mfpr_lpm != p->mfpr_run)
+		mfpr_writel(p->mfpr_off, p->mfpr_lpm);
 }
 
-void pxa3xx_mfp_config(mfp_cfg_t *mfp_cfgs, int num)
+void pxa3xx_mfp_config(unsigned long *mfp_cfgs, int num)
 {
-	int i, pin;
-	unsigned long val, flags;
-	mfp_cfg_t *mfp_cfg = mfp_cfgs;
+	unsigned long flags;
+	int i;
 
 	spin_lock_irqsave(&mfp_spin_lock, flags);
 
-	for (i = 0; i < num; i++, mfp_cfg++) {
-		pin = MFP_CFG_PIN(*mfp_cfg);
-		val = MFP_CFG_VAL(*mfp_cfg);
+	for (i = 0; i < num; i++, mfp_cfgs++) {
+		unsigned long tmp, c = *mfp_cfgs;
+		struct pxa3xx_mfp_pin *p;
+		int pin, af, drv, lpm, edge, pull;
 
+		pin = MFP_PIN(c);
 		BUG_ON(pin >= MFP_PIN_MAX);
-
-		__mfp_config(pin, val);
+		p = &mfp_table[pin];
+
+		af  = MFP_AF(c);
+		drv = MFP_DS(c);
+		lpm = MFP_LPM_STATE(c);
+		edge = MFP_LPM_EDGE(c);
+		pull = MFP_PULL(c);
+
+		/* run-mode pull settings will conflict with MFPR bits of
+		 * low power mode state,  calculate mfpr_run and mfpr_lpm
+		 * individually if pull != MFP_PULL_NONE
+		 */
+		tmp = MFPR_AF_SEL(af) | MFPR_DRIVE(drv);
+
+		if (likely(pull == MFP_PULL_NONE)) {
+			p->mfpr_run = tmp | mfpr_lpm[lpm] | mfpr_edge[edge];
+			p->mfpr_lpm = p->mfpr_run;
+		} else {
+			p->mfpr_lpm = tmp | mfpr_lpm[lpm] | mfpr_edge[edge];
+			p->mfpr_run = tmp | mfpr_pull[pull];
+		}
+
+		p->config = c; __mfp_config_run(p);
 	}
 
 	mfpr_sync();
@@ -110,7 +173,8 @@ void __init pxa3xx_mfp_init_addr(struct pxa3xx_mfp_addr_map *map)
 
 		do {
 			mfp_table[i].mfpr_off = offset;
-			mfp_table[i].mfpr_val = 0;
+			mfp_table[i].mfpr_run = 0;
+			mfp_table[i].mfpr_lpm = 0;
 			offset += 4; i++;
 		} while ((i <= p->end) && (p->end != -1));
 	}
@@ -120,5 +184,8 @@ void __init pxa3xx_mfp_init_addr(struct pxa3xx_mfp_addr_map *map)
 
 void __init pxa3xx_init_mfp(void)
 {
-	memset(mfp_table, 0, sizeof(mfp_table));
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mfp_table); i++)
+		mfp_table[i].config = -1;
 }
diff --git a/include/asm-arm/arch-pxa/mfp-pxa3xx.h b/include/asm-arm/arch-pxa/mfp-pxa3xx.h
index 5da1857d5476..1f6b35c015d0 100644
--- a/include/asm-arm/arch-pxa/mfp-pxa3xx.h
+++ b/include/asm-arm/arch-pxa/mfp-pxa3xx.h
@@ -1,6 +1,69 @@
 #ifndef __ASM_ARCH_MFP_PXA3XX_H
 #define __ASM_ARCH_MFP_PXA3XX_H
 
+#define MFPR_BASE	(0x40e10000)
+#define MFPR_SIZE	(PAGE_SIZE)
+
+/* MFPR register bit definitions */
+#define MFPR_PULL_SEL		(0x1 << 15)
+#define MFPR_PULLUP_EN		(0x1 << 14)
+#define MFPR_PULLDOWN_EN	(0x1 << 13)
+#define MFPR_SLEEP_SEL		(0x1 << 9)
+#define MFPR_SLEEP_OE_N		(0x1 << 7)
+#define MFPR_EDGE_CLEAR		(0x1 << 6)
+#define MFPR_EDGE_FALL_EN	(0x1 << 5)
+#define MFPR_EDGE_RISE_EN	(0x1 << 4)
+
+#define MFPR_SLEEP_DATA(x)	((x) << 8)
+#define MFPR_DRIVE(x)		(((x) & 0x7) << 10)
+#define MFPR_AF_SEL(x)		(((x) & 0x7) << 0)
+
+#define MFPR_EDGE_NONE		(0)
+#define MFPR_EDGE_RISE		(MFPR_EDGE_RISE_EN)
+#define MFPR_EDGE_FALL		(MFPR_EDGE_FALL_EN)
+#define MFPR_EDGE_BOTH		(MFPR_EDGE_RISE | MFPR_EDGE_FALL)
+
+/*
+ * Table that determines the low power modes outputs, with actual settings
+ * used in parentheses for don't-care values. Except for the float output,
+ * the configured driven and pulled levels match, so if there is a need for
+ * non-LPM pulled output, the same configuration could probably be used.
+ *
+ * Output value  sleep_oe_n  sleep_data  pullup_en  pulldown_en  pull_sel
+ *                 (bit 7)    (bit 8)    (bit 14)     (bit 13)   (bit 15)
+ *
+ * Input            0          X(0)        X(0)        X(0)       0
+ * Drive 0          0          0           0           X(1)       0
+ * Drive 1          0          1           X(1)        0	  0
+ * Pull hi (1)      1          X(1)        1           0	  0
+ * Pull lo (0)      1          X(0)        0           1	  0
+ * Z (float)        1          X(0)        0           0	  0
+ */
+#define MFPR_LPM_INPUT		(0)
+#define MFPR_LPM_DRIVE_LOW	(MFPR_SLEEP_DATA(0) | MFPR_PULLDOWN_EN)
+#define MFPR_LPM_DRIVE_HIGH    	(MFPR_SLEEP_DATA(1) | MFPR_PULLUP_EN)
+#define MFPR_LPM_PULL_LOW      	(MFPR_LPM_DRIVE_LOW  | MFPR_SLEEP_OE_N)
+#define MFPR_LPM_PULL_HIGH     	(MFPR_LPM_DRIVE_HIGH | MFPR_SLEEP_OE_N)
+#define MFPR_LPM_FLOAT         	(MFPR_SLEEP_OE_N)
+#define MFPR_LPM_MASK		(0xe080)
+
+/*
+ * The pullup and pulldown state of the MFP pin at run mode is by default
+ * determined by the selected alternate function. In case that some buggy
+ * devices need to override this default behavior,  the definitions below
+ * indicates the setting of corresponding MFPR bits
+ *
+ * Definition       pull_sel  pullup_en  pulldown_en
+ * MFPR_PULL_NONE       0         0        0
+ * MFPR_PULL_LOW        1         0        1
+ * MFPR_PULL_HIGH       1         1        0
+ * MFPR_PULL_BOTH       1         1        1
+ */
+#define MFPR_PULL_NONE		(0)
+#define MFPR_PULL_LOW		(MFPR_PULL_SEL | MFPR_PULLDOWN_EN)
+#define MFPR_PULL_BOTH		(MFPR_PULL_LOW | MFPR_PULLUP_EN)
+#define MFPR_PULL_HIGH		(MFPR_PULL_SEL | MFPR_PULLUP_EN)
+
 /* PXA3xx common MFP configurations - processor specific ones defined
  * in mfp-pxa300.h and mfp-pxa320.h
  */
@@ -134,4 +197,56 @@
 #define GPIO5_2_GPIO		MFP_CFG(GPIO5_2, AF0)
 #define GPIO6_2_GPIO		MFP_CFG(GPIO6_2, AF0)
 
+/*
+ * each MFP pin will have a MFPR register, since the offset of the
+ * register varies between processors, the processor specific code
+ * should initialize the pin offsets by pxa3xx_mfp_init_addr()
+ *
+ * pxa3xx_mfp_init_addr - accepts a table of "pxa3xx_mfp_addr_map"
+ * structure, which represents a range of MFP pins from "start" to
+ * "end", with the offset begining at "offset", to define a single
+ * pin, let "end" = -1
+ *
+ * use
+ *
+ * MFP_ADDR_X() to define a range of pins
+ * MFP_ADDR()   to define a single pin
+ * MFP_ADDR_END to signal the end of pin offset definitions
+ */
+struct pxa3xx_mfp_addr_map {
+	unsigned int	start;
+	unsigned int	end;
+	unsigned long	offset;
+};
+
+#define MFP_ADDR_X(start, end, offset) \
+	{ MFP_PIN_##start, MFP_PIN_##end, offset }
+
+#define MFP_ADDR(pin, offset) \
+	{ MFP_PIN_##pin, -1, offset }
+
+#define MFP_ADDR_END	{ MFP_PIN_INVALID, 0 }
+
+/*
+ * pxa3xx_mfp_read()/pxa3xx_mfp_write() - for direct read/write access
+ * to the MFPR register
+ */
+unsigned long pxa3xx_mfp_read(int mfp);
+void pxa3xx_mfp_write(int mfp, unsigned long mfpr_val);
+
+/*
+ * pxa3xx_mfp_config - configure the MFPR registers
+ *
+ * used by board specific initialization code
+ */
+void pxa3xx_mfp_config(unsigned long *mfp_cfgs, int num);
+
+/*
+ * pxa3xx_mfp_init_addr() - initialize the mapping between mfp pin
+ * index and MFPR register offset
+ *
+ * used by processor specific code
+ */
+void __init pxa3xx_mfp_init_addr(struct pxa3xx_mfp_addr_map *);
+void __init pxa3xx_init_mfp(void);
 #endif /* __ASM_ARCH_MFP_PXA3XX_H */
diff --git a/include/asm-arm/arch-pxa/mfp.h b/include/asm-arm/arch-pxa/mfp.h
index 60fbed1b2e89..02f6157396d3 100644
--- a/include/asm-arm/arch-pxa/mfp.h
+++ b/include/asm-arm/arch-pxa/mfp.h
@@ -16,9 +16,6 @@
 #ifndef __ASM_ARCH_MFP_H
 #define __ASM_ARCH_MFP_H
 
-#define MFPR_BASE	(0x40e10000)
-#define MFPR_SIZE	(PAGE_SIZE)
-
 #define mfp_to_gpio(m)	((m) % 128)
 
 /* list of all the configurable MFP pins */
@@ -216,115 +213,22 @@ enum {
 	MFP_PIN_MAX,
 };
 
-/*
- * Table that determines the low power modes outputs, with actual settings
- * used in parentheses for don't-care values. Except for the float output,
- * the configured driven and pulled levels match, so if there is a need for
- * non-LPM pulled output, the same configuration could probably be used.
- *
- * Output value  sleep_oe_n  sleep_data  pullup_en  pulldown_en  pull_sel
- *                 (bit 7)    (bit 8)    (bit 14d)   (bit 13d)
- *
- * Drive 0          0          0           0           X (1)      0
- * Drive 1          0          1           X (1)       0	  0
- * Pull hi (1)      1          X(1)        1           0	  0
- * Pull lo (0)      1          X(0)        0           1	  0
- * Z (float)        1          X(0)        0           0	  0
- */
-#define MFP_LPM_DRIVE_LOW	0x8
-#define MFP_LPM_DRIVE_HIGH    	0x6
-#define MFP_LPM_PULL_HIGH     	0x7
-#define MFP_LPM_PULL_LOW      	0x9
-#define MFP_LPM_FLOAT         	0x1
-#define MFP_LPM_PULL_NEITHER	0x0
-
-/*
- * The pullup and pulldown state of the MFP pin is by default determined by
- * selected alternate function. In case some buggy devices need to override
- * this default behavior,  pxa3xx_mfp_set_pull() can be invoked with one of
- * the following definition as the parameter.
- *
- * Definition       pull_sel  pullup_en  pulldown_en
- * MFP_PULL_HIGH        1         1        0
- * MFP_PULL_LOW         1         0        1
- * MFP_PULL_BOTH        1         1        1
- * MFP_PULL_NONE        1         0        0
- * MFP_PULL_DEFAULT     0         X        X
- *
- * NOTE: pxa3xx_mfp_set_pull() will modify the PULLUP_EN and PULLDOWN_EN
- * bits,  which will cause potential conflicts with the low power mode
- * setting, device drivers should take care of this
- */
-#define MFP_PULL_BOTH		(0x7u)
-#define MFP_PULL_HIGH		(0x6u)
-#define MFP_PULL_LOW		(0x5u)
-#define MFP_PULL_NONE		(0x4u)
-#define MFP_PULL_DEFAULT	(0x0u)
-
-#define MFP_AF0			(0)
-#define MFP_AF1			(1)
-#define MFP_AF2			(2)
-#define MFP_AF3			(3)
-#define MFP_AF4			(4)
-#define MFP_AF5			(5)
-#define MFP_AF6			(6)
-#define MFP_AF7			(7)
-
-#define MFP_DS01X		(0)
-#define MFP_DS02X		(1)
-#define MFP_DS03X		(2)
-#define MFP_DS04X		(3)
-#define MFP_DS06X		(4)
-#define MFP_DS08X		(5)
-#define MFP_DS10X		(6)
-#define MFP_DS12X		(7)
-
-#define MFP_EDGE_BOTH		0x3
-#define MFP_EDGE_RISE		0x2
-#define MFP_EDGE_FALL		0x1
-#define MFP_EDGE_NONE		0x0
-
-#define MFPR_AF_MASK		0x0007
-#define MFPR_DRV_MASK		0x1c00
-#define MFPR_RDH_MASK		0x0200
-#define MFPR_LPM_MASK		0xe180
-#define MFPR_PULL_MASK		0xe000
-#define MFPR_EDGE_MASK		0x0070
-
-#define MFPR_ALT_OFFSET		0
-#define MFPR_ERE_OFFSET		4
-#define MFPR_EFE_OFFSET		5
-#define MFPR_EC_OFFSET		6
-#define MFPR_SON_OFFSET		7
-#define MFPR_SD_OFFSET		8
-#define MFPR_SS_OFFSET		9
-#define MFPR_DRV_OFFSET		10
-#define MFPR_PD_OFFSET		13
-#define MFPR_PU_OFFSET		14
-#define MFPR_PS_OFFSET		15
-
-#define MFPR(af, drv, rdh, lpm, edge) \
-	(((af) & 0x7) | (((drv) & 0x7) << 10) |\
-	 (((rdh) & 0x1) << 9) |\
-	 (((lpm) & 0x3) << 7) |\
-	 (((lpm) & 0x4) << 12)|\
-	 (((lpm) & 0x8) << 10)|\
-	 ((!(edge)) << 6) |\
-	 (((edge) & 0x1) << 5) |\
-	 (((edge) & 0x2) << 3))
-
 /*
  * a possible MFP configuration is represented by a 32-bit integer
- * bit  0..15 - MFPR value (16-bit)
- * bit 16..31 - mfp pin index (used to obtain the MFPR offset)
+ *
+ * bit  0.. 9 - MFP Pin Number (1024 Pins Maximum)
+ * bit 10..12 - Alternate Function Selection
+ * bit 13..15 - Drive Strength
+ * bit 16..18 - Low Power Mode State
+ * bit 19..20 - Low Power Mode Edge Detection
+ * bit 21..22 - Run Mode Pull State
  *
  * to facilitate the definition, the following macros are provided
  *
- * MFPR_DEFAULT - default MFPR value, with
+ * MFP_CFG_DEFAULT - default MFP configuration value, with
  * 		  alternate function = 0,
- * 		  drive strength = fast 1mA (MFP_DS01X)
+ * 		  drive strength = fast 3mA (MFP_DS03X)
  * 		  low power mode = default
- * 		  release dalay hold = false (RDH bit)
  * 		  edge detection = none
  *
  * MFP_CFG	- default MFPR value with alternate function
@@ -334,104 +238,74 @@ enum {
  * 		  low power mode
  * MFP_CFG_X	- default MFPR value with alternate function,
  * 		  pin drive strength and low power mode
- *
- * use
- *
- * MFP_CFG_PIN	- to get the MFP pin index
- * MFP_CFG_VAL	- to get the corresponding MFPR value
  */
 
-typedef uint32_t mfp_cfg_t;
-
-#define MFP_CFG_PIN(mfp_cfg)	(((mfp_cfg) >> 16) & 0xffff)
-#define MFP_CFG_VAL(mfp_cfg)	((mfp_cfg) & 0xffff)
-
-/*
- * MFP register defaults to
- *   drive strength fast 3mA (010'b)
- *   edge detection logic disabled
- *   alternate function 0
- */
-#define MFPR_DEFAULT	(0x0840)
+typedef unsigned long mfp_cfg_t;
+
+#define MFP_PIN(x)		((x) & 0x3ff)
+
+#define MFP_AF0			(0x0 << 10)
+#define MFP_AF1			(0x1 << 10)
+#define MFP_AF2			(0x2 << 10)
+#define MFP_AF3			(0x3 << 10)
+#define MFP_AF4			(0x4 << 10)
+#define MFP_AF5			(0x5 << 10)
+#define MFP_AF6			(0x6 << 10)
+#define MFP_AF7			(0x7 << 10)
+#define MFP_AF_MASK		(0x7 << 10)
+#define MFP_AF(x)		(((x) >> 10) & 0x7)
+
+#define MFP_DS01X		(0x0 << 13)
+#define MFP_DS02X		(0x1 << 13)
+#define MFP_DS03X		(0x2 << 13)
+#define MFP_DS04X		(0x3 << 13)
+#define MFP_DS06X		(0x4 << 13)
+#define MFP_DS08X		(0x5 << 13)
+#define MFP_DS10X		(0x6 << 13)
+#define MFP_DS13X		(0x7 << 13)
+#define MFP_DS_MASK		(0x7 << 13)
+#define MFP_DS(x)		(((x) >> 13) & 0x7)
+
+#define MFP_LPM_INPUT		(0x0 << 16)
+#define MFP_LPM_DRIVE_LOW	(0x1 << 16)
+#define MFP_LPM_DRIVE_HIGH	(0x2 << 16)
+#define MFP_LPM_PULL_LOW	(0x3 << 16)
+#define MFP_LPM_PULL_HIGH	(0x4 << 16)
+#define MFP_LPM_FLOAT		(0x5 << 16)
+#define MFP_LPM_STATE_MASK	(0x7 << 16)
+#define MFP_LPM_STATE(x)	(((x) >> 16) & 0x7)
+
+#define MFP_LPM_EDGE_NONE	(0x0 << 19)
+#define MFP_LPM_EDGE_RISE	(0x1 << 19)
+#define MFP_LPM_EDGE_FALL	(0x2 << 19)
+#define MFP_LPM_EDGE_BOTH	(0x3 << 19)
+#define MFP_LPM_EDGE_MASK	(0x3 << 19)
+#define MFP_LPM_EDGE(x)		(((x) >> 19) & 0x3)
+
+#define MFP_PULL_NONE		(0x0 << 21)
+#define MFP_PULL_LOW		(0x1 << 21)
+#define MFP_PULL_HIGH		(0x2 << 21)
+#define MFP_PULL_BOTH		(0x3 << 21)
+#define MFP_PULL_MASK		(0x3 << 21)
+#define MFP_PULL(x)		(((x) >> 21) & 0x3)
+
+#define MFP_CFG_DEFAULT		(MFP_AF0 | MFP_DS03X | MFP_LPM_INPUT |\
+				 MFP_LPM_EDGE_NONE | MFP_PULL_NONE)
 
 #define MFP_CFG(pin, af)		\
-	((MFP_PIN_##pin << 16) | MFPR_DEFAULT | (MFP_##af))
+	((MFP_CFG_DEFAULT & ~MFP_AF_MASK) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af))
 
 #define MFP_CFG_DRV(pin, af, drv)	\
-	((MFP_PIN_##pin << 16) | (MFPR_DEFAULT & ~MFPR_DRV_MASK) |\
-	 ((MFP_##drv) << 10) | (MFP_##af))
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_DS_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_##drv))
 
 #define MFP_CFG_LPM(pin, af, lpm)	\
-	((MFP_PIN_##pin << 16) | (MFPR_DEFAULT & ~MFPR_LPM_MASK) |\
-	 (((MFP_LPM_##lpm) & 0x3) << 7)  |\
-	 (((MFP_LPM_##lpm) & 0x4) << 12) |\
-	 (((MFP_LPM_##lpm) & 0x8) << 10) |\
-	 (MFP_##af))
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_LPM_STATE_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_LPM_##lpm))
 
 #define MFP_CFG_X(pin, af, drv, lpm)	\
-	((MFP_PIN_##pin << 16) |\
-	 (MFPR_DEFAULT & ~(MFPR_DRV_MASK | MFPR_LPM_MASK)) |\
-	 ((MFP_##drv) << 10) | (MFP_##af) |\
-	 (((MFP_LPM_##lpm) & 0x3) << 7)  |\
-	 (((MFP_LPM_##lpm) & 0x4) << 12) |\
-	 (((MFP_LPM_##lpm) & 0x8) << 10))
-
-/*
- * each MFP pin will have a MFPR register, since the offset of the
- * register varies between processors, the processor specific code
- * should initialize the pin offsets by pxa3xx_mfp_init_addr()
- *
- * pxa3xx_mfp_init_addr - accepts a table of "pxa3xx_mfp_addr_map"
- * structure, which represents a range of MFP pins from "start" to
- * "end", with the offset begining at "offset", to define a single
- * pin, let "end" = -1
- *
- * use
- *
- * MFP_ADDR_X() to define a range of pins
- * MFP_ADDR()   to define a single pin
- * MFP_ADDR_END to signal the end of pin offset definitions
- */
-struct pxa3xx_mfp_addr_map {
-	unsigned int	start;
-	unsigned int	end;
-	unsigned long	offset;
-};
-
-#define MFP_ADDR_X(start, end, offset) \
-	{ MFP_PIN_##start, MFP_PIN_##end, offset }
-
-#define MFP_ADDR(pin, offset) \
-	{ MFP_PIN_##pin, -1, offset }
-
-#define MFP_ADDR_END	{ MFP_PIN_INVALID, 0 }
-
-struct pxa3xx_mfp_pin {
-	unsigned long	mfpr_off;	/* MFPRxx register offset */
-	unsigned long	mfpr_val;	/* MFPRxx register value */
-};
-
-/*
- * pxa3xx_mfp_read()/pxa3xx_mfp_write() - for direct read/write access
- * to the MFPR register
- */
-unsigned long pxa3xx_mfp_read(int mfp);
-void pxa3xx_mfp_write(int mfp, unsigned long mfpr_val);
-
-/*
- * pxa3xx_mfp_config - configure the MFPR registers
- *
- * used by board specific initialization code
- */
-void pxa3xx_mfp_config(mfp_cfg_t *mfp_cfgs, int num);
-
-/*
- * pxa3xx_mfp_init_addr() - initialize the mapping between mfp pin
- * index and MFPR register offset
- *
- * used by processor specific code
- */
-void __init pxa3xx_mfp_init_addr(struct pxa3xx_mfp_addr_map *);
-void __init pxa3xx_init_mfp(void);
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_DS_MASK | MFP_LPM_STATE_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_##drv | MFP_LPM_##lpm))
 
 #endif /* __ASM_ARCH_MFP_H */
-- 
cgit v1.2.3


From 7b5dea12346f98b0624e00ef585246bfec8b0959 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 7 Jan 2008 22:18:30 +0000
Subject: [ARM] pxa: Add PXA3 standby code hooked into the IRQ wake scheme

Wakeup sources on PXA3 are enabled at two levels.  First, the MFP
configuration has to be set to enable which edges a specific pin
will trigger a wakeup.  The pin also has to be routed to a functional
unit.  Lastly, the functional unit must be enabled as a wakeup source
in the appropriate AD*ER registers (AD2D0ER for standby resume.)

This doesn't fit well with the IRQ wake scheme - we currently do a
best effort conversion from IRQ numbers to functional unit wake enable
bits.  For instance, there's several USB client related enable bits but
there's no corresponding IRQs to determine which you'd want.  Conversely,
there's a single enable bit covering several functional units.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/pxa3xx.c             | 223 +++++++++++++++++++++++++++++++++
 arch/arm/mach-pxa/standby.S            |  80 ++++++++++++
 include/asm-arm/arch-pxa/pxa3xx-regs.h |  86 +++++++++++++
 3 files changed, 389 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c
index fcb2359b3867..0b2a15ed3999 100644
--- a/arch/arm/mach-pxa/pxa3xx.c
+++ b/arch/arm/mach-pxa/pxa3xx.c
@@ -19,6 +19,7 @@
 #include <linux/pm.h>
 #include <linux/platform_device.h>
 #include <linux/irq.h>
+#include <linux/io.h>
 
 #include <asm/hardware.h>
 #include <asm/arch/pxa3xx-regs.h>
@@ -201,6 +202,225 @@ static struct clk pxa3xx_clks[] = {
 	PXA3xx_CKEN("MMCCLK", MMC3, 19500000, 0, &pxa3xx_device_mci3.dev),
 };
 
+#ifdef CONFIG_PM
+#define SLEEP_SAVE_SIZE	4
+
+#define ISRAM_START	0x5c000000
+#define ISRAM_SIZE	SZ_256K
+
+static void __iomem *sram;
+static unsigned long wakeup_src;
+
+static void pxa3xx_cpu_pm_save(unsigned long *sleep_save)
+{
+	pr_debug("PM: CKENA=%08x CKENB=%08x\n", CKENA, CKENB);
+
+	if (CKENA & (1 << CKEN_USBH)) {
+		printk(KERN_ERR "PM: USB host clock not stopped?\n");
+		CKENA &= ~(1 << CKEN_USBH);
+	}
+//	CKENA |= 1 << (CKEN_ISC & 31);
+
+	/*
+	 * Low power modes require the HSIO2 clock to be enabled.
+	 */
+	CKENB |= 1 << (CKEN_HSIO2 & 31);
+}
+
+static void pxa3xx_cpu_pm_restore(unsigned long *sleep_save)
+{
+	CKENB &= ~(1 << (CKEN_HSIO2 & 31));
+}
+
+/*
+ * Enter a standby mode (S0D1C2 or S0D2C2).  Upon wakeup, the dynamic
+ * memory controller has to be reinitialised, so we place some code
+ * in the SRAM to perform this function.
+ *
+ * We disable FIQs across the standby - otherwise, we might receive a
+ * FIQ while the SDRAM is unavailable.
+ */
+static void pxa3xx_cpu_standby(unsigned int pwrmode)
+{
+	extern const char pm_enter_standby_start[], pm_enter_standby_end[];
+	void (*fn)(unsigned int) = (void __force *)(sram + 0x8000);
+
+	memcpy_toio(sram + 0x8000, pm_enter_standby_start,
+		    pm_enter_standby_end - pm_enter_standby_start);
+
+	AD2D0SR = ~0;
+	AD2D1SR = ~0;
+	AD2D0ER = wakeup_src;
+	AD2D1ER = 0;
+	ASCR = ASCR;
+	ARSR = ARSR;
+
+	local_fiq_disable();
+	fn(pwrmode);
+	local_fiq_enable();
+
+	AD2D0ER = 0;
+	AD2D1ER = 0;
+
+	printk("PM: AD2D0SR=%08x ASCR=%08x\n", AD2D0SR, ASCR);
+}
+
+static void pxa3xx_cpu_pm_enter(suspend_state_t state)
+{
+	/*
+	 * Don't sleep if no wakeup sources are defined
+	 */
+	if (wakeup_src == 0)
+		return;
+
+	switch (state) {
+	case PM_SUSPEND_STANDBY:
+		pxa3xx_cpu_standby(PXA3xx_PM_S0D2C2);
+		break;
+
+	case PM_SUSPEND_MEM:
+		break;
+	}
+}
+
+static int pxa3xx_cpu_pm_valid(suspend_state_t state)
+{
+	return state == PM_SUSPEND_MEM || state == PM_SUSPEND_STANDBY;
+}
+
+static struct pxa_cpu_pm_fns pxa3xx_cpu_pm_fns = {
+	.save_size	= SLEEP_SAVE_SIZE,
+	.save		= pxa3xx_cpu_pm_save,
+	.restore	= pxa3xx_cpu_pm_restore,
+	.valid		= pxa3xx_cpu_pm_valid,
+	.enter		= pxa3xx_cpu_pm_enter,
+};
+
+static void __init pxa3xx_init_pm(void)
+{
+	sram = ioremap(ISRAM_START, ISRAM_SIZE);
+	if (!sram) {
+		printk(KERN_ERR "Unable to map ISRAM: disabling standby/suspend\n");
+		return;
+	}
+
+	/*
+	 * Since we copy wakeup code into the SRAM, we need to ensure
+	 * that it is preserved over the low power modes.  Note: bit 8
+	 * is undocumented in the developer manual, but must be set.
+	 */
+	AD1R |= ADXR_L2 | ADXR_R0;
+	AD2R |= ADXR_L2 | ADXR_R0;
+	AD3R |= ADXR_L2 | ADXR_R0;
+
+	/*
+	 * Clear the resume enable registers.
+	 */
+	AD1D0ER = 0;
+	AD2D0ER = 0;
+	AD2D1ER = 0;
+	AD3ER = 0;
+
+	pxa_cpu_pm_fns = &pxa3xx_cpu_pm_fns;
+}
+
+static int pxa3xx_set_wake(unsigned int irq, unsigned int on)
+{
+	unsigned long flags, mask = 0;
+
+	switch (irq) {
+	case IRQ_SSP3:
+		mask = ADXER_MFP_WSSP3;
+		break;
+	case IRQ_MSL:
+		mask = ADXER_WMSL0;
+		break;
+	case IRQ_USBH2:
+	case IRQ_USBH1:
+		mask = ADXER_WUSBH;
+		break;
+	case IRQ_KEYPAD:
+		mask = ADXER_WKP;
+		break;
+	case IRQ_AC97:
+		mask = ADXER_MFP_WAC97;
+		break;
+	case IRQ_USIM:
+		mask = ADXER_WUSIM0;
+		break;
+	case IRQ_SSP2:
+		mask = ADXER_MFP_WSSP2;
+		break;
+	case IRQ_I2C:
+		mask = ADXER_MFP_WI2C;
+		break;
+	case IRQ_STUART:
+		mask = ADXER_MFP_WUART3;
+		break;
+	case IRQ_BTUART:
+		mask = ADXER_MFP_WUART2;
+		break;
+	case IRQ_FFUART:
+		mask = ADXER_MFP_WUART1;
+		break;
+	case IRQ_MMC:
+		mask = ADXER_MFP_WMMC1;
+		break;
+	case IRQ_SSP:
+		mask = ADXER_MFP_WSSP1;
+		break;
+	case IRQ_RTCAlrm:
+		mask = ADXER_WRTC;
+		break;
+	case IRQ_SSP4:
+		mask = ADXER_MFP_WSSP4;
+		break;
+	case IRQ_TSI:
+		mask = ADXER_WTSI;
+		break;
+	case IRQ_USIM2:
+		mask = ADXER_WUSIM1;
+		break;
+	case IRQ_MMC2:
+		mask = ADXER_MFP_WMMC2;
+		break;
+	case IRQ_NAND:
+		mask = ADXER_MFP_WFLASH;
+		break;
+	case IRQ_USB2:
+		mask = ADXER_WUSB2;
+		break;
+	case IRQ_WAKEUP0:
+		mask = ADXER_WEXTWAKE0;
+		break;
+	case IRQ_WAKEUP1:
+		mask = ADXER_WEXTWAKE1;
+		break;
+	case IRQ_MMC3:
+		mask = ADXER_MFP_GEN12;
+		break;
+	}
+
+	local_irq_save(flags);
+	if (on)
+		wakeup_src |= mask;
+	else
+		wakeup_src &= ~mask;
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static void pxa3xx_init_irq_pm(void)
+{
+	pxa_init_irq_set_wake(pxa3xx_set_wake);
+}
+
+#else
+static inline void pxa3xx_init_pm(void) {}
+static inline void pxa3xx_init_irq_pm(void) {}
+#endif
+
 void __init pxa3xx_init_irq(void)
 {
 	/* enable CP6 access */
@@ -212,6 +432,7 @@ void __init pxa3xx_init_irq(void)
 	pxa_init_irq_low();
 	pxa_init_irq_high();
 	pxa_init_irq_gpio(128);
+	pxa3xx_init_irq_pm();
 }
 
 /*
@@ -241,6 +462,8 @@ static int __init pxa3xx_init(void)
 		if ((ret = pxa_init_dma(32)))
 			return ret;
 
+		pxa3xx_init_pm();
+
 		return platform_add_devices(devices, ARRAY_SIZE(devices));
 	}
 	return 0;
diff --git a/arch/arm/mach-pxa/standby.S b/arch/arm/mach-pxa/standby.S
index 434a6ab0eca5..167412e6bec8 100644
--- a/arch/arm/mach-pxa/standby.S
+++ b/arch/arm/mach-pxa/standby.S
@@ -32,3 +32,83 @@ ENTRY(pxa_cpu_standby)
 	mov	pc, lr
 
 #endif
+
+#ifdef CONFIG_PXA3xx
+
+#define MDCNFG		0x0000
+#define MDCNFG_DMCEN	(1 << 30)
+#define DDR_HCAL	0x0060
+#define DDR_HCAL_HCRNG	0x1f
+#define DDR_HCAL_HCPROG	(1 << 28)
+#define DDR_HCAL_HCEN	(1 << 31)
+#define DMCIER		0x0070
+#define DMCIER_EDLP	(1 << 29)
+#define DMCISR		0x0078
+#define RCOMP		0x0100
+#define RCOMP_SWEVAL	(1 << 31)
+
+ENTRY(pm_enter_standby_start)
+	mov	r1, #0xf6000000		@ DMEMC_REG_BASE (MDCNFG)
+	add	r1, r1, #0x00100000
+
+	/*
+	 * Preload the TLB entry for accessing the dynamic memory
+	 * controller registers.  Note that page table lookups will
+	 * fail until the dynamic memory controller has been
+	 * reinitialised - and that includes MMU page table walks.
+	 * This also means that only the dynamic memory controller
+	 * can be reliably accessed in the code following standby.
+	 */
+	ldr	r2, [r1]		@ Dummy read MDCNFG
+
+	mcr	p14, 0, r0, c7, c0, 0
+	.rept	8
+	nop
+	.endr
+
+	ldr	r0, [r1, #DDR_HCAL]	@ Clear (and wait for) HCEN
+	bic	r0, r0, #DDR_HCAL_HCEN
+	str	r0, [r1, #DDR_HCAL]
+1:	ldr	r0, [r1, #DDR_HCAL]
+	tst	r0, #DDR_HCAL_HCEN
+	bne	1b
+
+	ldr	r0, [r1, #RCOMP]	@ Initiate RCOMP
+	orr	r0, r0, #RCOMP_SWEVAL
+	str	r0, [r1, #RCOMP]
+
+	mov	r0, #~0			@ Clear interrupts
+	str	r0, [r1, #DMCISR]
+
+	ldr	r0, [r1, #DMCIER]	@ set DMIER[EDLP]
+	orr	r0, r0, #DMCIER_EDLP
+	str	r0, [r1, #DMCIER]
+
+	ldr	r0, [r1, #DDR_HCAL]	@ clear HCRNG, set HCPROG, HCEN
+	bic	r0, r0, #DDR_HCAL_HCRNG
+	orr	r0, r0, #DDR_HCAL_HCEN | DDR_HCAL_HCPROG
+	str	r0, [r1, #DDR_HCAL]
+
+1:	ldr	r0, [r1, #DMCISR]
+	tst	r0, #DMCIER_EDLP
+	beq	1b
+
+	ldr	r0, [r1, #MDCNFG]	@ set MDCNFG[DMCEN]
+	orr	r0, r0, #MDCNFG_DMCEN
+	str	r0, [r1, #MDCNFG]
+1:	ldr	r0, [r1, #MDCNFG]
+	tst	r0, #MDCNFG_DMCEN
+	beq	1b
+
+	ldr	r0, [r1, #DDR_HCAL]	@ set DDR_HCAL[HCRNG]
+	orr	r0, r0, #2 @ HCRNG
+	str	r0, [r1, #DDR_HCAL]
+
+	ldr	r0, [r1, #DMCIER]	@ Clear the interrupt
+	bic	r0, r0, #0x20000000
+	str	r0, [r1, #DMCIER]
+
+	mov	pc, lr
+ENTRY(pm_enter_standby_end)
+
+#endif
diff --git a/include/asm-arm/arch-pxa/pxa3xx-regs.h b/include/asm-arm/arch-pxa/pxa3xx-regs.h
index 3900a0ca0bc0..66d54119757c 100644
--- a/include/asm-arm/arch-pxa/pxa3xx-regs.h
+++ b/include/asm-arm/arch-pxa/pxa3xx-regs.h
@@ -13,6 +13,92 @@
 #ifndef __ASM_ARCH_PXA3XX_REGS_H
 #define __ASM_ARCH_PXA3XX_REGS_H
 
+/*
+ * Slave Power Managment Unit
+ */
+#define ASCR		__REG(0x40f40000)	/* Application Subsystem Power Status/Configuration */
+#define ARSR		__REG(0x40f40004)	/* Application Subsystem Reset Status */
+#define AD3ER		__REG(0x40f40008)	/* Application Subsystem Wake-Up from D3 Enable */
+#define AD3SR		__REG(0x40f4000c)	/* Application Subsystem Wake-Up from D3 Status */
+#define AD2D0ER		__REG(0x40f40010)	/* Application Subsystem Wake-Up from D2 to D0 Enable */
+#define AD2D0SR		__REG(0x40f40014)	/* Application Subsystem Wake-Up from D2 to D0 Status */
+#define AD2D1ER		__REG(0x40f40018)	/* Application Subsystem Wake-Up from D2 to D1 Enable */
+#define AD2D1SR		__REG(0x40f4001c)	/* Application Subsystem Wake-Up from D2 to D1 Status */
+#define AD1D0ER		__REG(0x40f40020)	/* Application Subsystem Wake-Up from D1 to D0 Enable */
+#define AD1D0SR		__REG(0x40f40024)	/* Application Subsystem Wake-Up from D1 to D0 Status */
+#define AGENP		__REG(0x40f4002c)	/* Application Subsystem General Purpose */
+#define AD3R		__REG(0x40f40030)	/* Application Subsystem D3 Configuration */
+#define AD2R		__REG(0x40f40034)	/* Application Subsystem D2 Configuration */
+#define AD1R		__REG(0x40f40038)	/* Application Subsystem D1 Configuration */
+
+/*
+ * Application Subsystem Configuration bits.
+ */
+#define ASCR_RDH		(1 << 31)
+#define ASCR_D1S		(1 << 2)
+#define ASCR_D2S		(1 << 1)
+#define ASCR_D3S		(1 << 0)
+
+/*
+ * Application Reset Status bits.
+ */
+#define ARSR_GPR		(1 << 3)
+#define ARSR_LPMR		(1 << 2)
+#define ARSR_WDT		(1 << 1)
+#define ARSR_HWR		(1 << 0)
+
+/*
+ * Application Subsystem Wake-Up bits.
+ */
+#define ADXER_WRTC		(1 << 31)	/* RTC */
+#define ADXER_WOST		(1 << 30)	/* OS Timer */
+#define ADXER_WTSI		(1 << 29)	/* Touchscreen */
+#define ADXER_WUSBH		(1 << 28)	/* USB host */
+#define ADXER_WUSB2		(1 << 26)	/* USB client 2.0 */
+#define ADXER_WMSL0		(1 << 24)	/* MSL port 0*/
+#define ADXER_WDMUX3		(1 << 23)	/* USB EDMUX3 */
+#define ADXER_WDMUX2		(1 << 22)	/* USB EDMUX2 */
+#define ADXER_WKP		(1 << 21)	/* Keypad */
+#define ADXER_WUSIM1		(1 << 20)	/* USIM Port 1 */
+#define ADXER_WUSIM0		(1 << 19)	/* USIM Port 0 */
+#define ADXER_WOTG		(1 << 16)	/* USBOTG input */
+#define ADXER_MFP_WFLASH	(1 << 15)	/* MFP: Data flash busy */
+#define ADXER_MFP_GEN12		(1 << 14)	/* MFP: MMC3/GPIO/OST inputs */
+#define ADXER_MFP_WMMC2		(1 << 13)	/* MFP: MMC2 */
+#define ADXER_MFP_WMMC1		(1 << 12)	/* MFP: MMC1 */
+#define ADXER_MFP_WI2C		(1 << 11)	/* MFP: I2C */
+#define ADXER_MFP_WSSP4		(1 << 10)	/* MFP: SSP4 */
+#define ADXER_MFP_WSSP3		(1 << 9)	/* MFP: SSP3 */
+#define ADXER_MFP_WMAXTRIX	(1 << 8)	/* MFP: matrix keypad */
+#define ADXER_MFP_WUART3	(1 << 7)	/* MFP: UART3 */
+#define ADXER_MFP_WUART2	(1 << 6)	/* MFP: UART2 */
+#define ADXER_MFP_WUART1	(1 << 5)	/* MFP: UART1 */
+#define ADXER_MFP_WSSP2		(1 << 4)	/* MFP: SSP2 */
+#define ADXER_MFP_WSSP1		(1 << 3)	/* MFP: SSP1 */
+#define ADXER_MFP_WAC97		(1 << 2)	/* MFP: AC97 */
+#define ADXER_WEXTWAKE1		(1 << 1)	/* External Wake 1 */
+#define ADXER_WEXTWAKE0		(1 << 0)	/* External Wake 0 */
+
+/*
+ * AD3R/AD2R/AD1R bits.  R2-R5 are only defined for PXA320.
+ */
+#define ADXR_L2			(1 << 8)
+#define ADXR_R5			(1 << 5)
+#define ADXR_R4			(1 << 4)
+#define ADXR_R3			(1 << 3)
+#define ADXR_R2			(1 << 2)
+#define ADXR_R1			(1 << 1)
+#define ADXR_R0			(1 << 0)
+
+/*
+ * Values for PWRMODE CP15 register
+ */
+#define PXA3xx_PM_S3D4C4	0x07	/* aka deep sleep */
+#define PXA3xx_PM_S2D3C4	0x06	/* aka sleep */
+#define PXA3xx_PM_S0D2C2	0x03	/* aka standby */
+#define PXA3xx_PM_S0D1C2	0x02	/* aka LCD refresh */
+#define PXA3xx_PM_S0D0C1	0x01
+
 /*
  * Application Subsystem Clock
  */
-- 
cgit v1.2.3


From 35aa1df4328340f38edc46f00837f08d33d49f63 Mon Sep 17 00:00:00 2001
From: Quentin Barnes <qbarnes@gmail.com>
Date: Mon, 11 Jun 2007 22:20:10 +0000
Subject: ARM kprobes: instruction single-stepping support

This is the code implementing instruction single-stepping for kprobes
on ARM.

To get around the limitation of no Next-PC and no hardware single-
stepping, all kprobe'd instructions are split into three camps:
simulation, emulation, and rejected. "Simulated" instructions are
those instructions which behavior is reproduced by straight C code.
"Emulated" instructions are ones that are copied, slightly altered
and executed directly in the instruction slot to reproduce their
behavior.  "Rejected" instructions are ones that could be simulated,
but work hasn't been put into simulating them. These instructions
should be very rare, if not unencountered, in the kernel. If ever
needed, code could be added to simulate them.

One might wonder why this and the ptrace singlestep facility are not
sharing some code.  Both approaches are fundamentally different because
the ptrace code regains control after the stepped instruction by installing
a breakpoint after the instruction itself, and possibly at the location
where the instruction might be branching to, instead of simulating or
emulating the target instruction.

The ptrace approach isn't suitable for kprobes because the breakpoints
would have to be moved back, and the icache flushed, everytime the
probe is hit to let normal code execution resume, which would have a
significant performance impact. It is also racy on SMP since another
CPU could, with the right timing, sail through the probe point without
being caught.  Because ptrace single-stepping always result in a
different process to be scheduled, the concern for performance is much
less significant.

On the other hand, the kprobes approach isn't (currently) suitable for
ptrace because it has no provision for proper user space memory
protection and translation, and even if that was implemented, the gain
wouldn't be worth the added complexity in the ptrace path compared to
the current approach.

So, until kprobes does support user space, both kprobes and ptrace are
best kept independent and separate.

Signed-off-by: Quentin Barnes <qbarnes@gmail.com>
Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/kernel/Makefile         |    1 +
 arch/arm/kernel/kprobes-decode.c | 1529 ++++++++++++++++++++++++++++++++++++++
 include/asm-arm/kprobes.h        |   43 ++
 3 files changed, 1573 insertions(+)
 create mode 100644 arch/arm/kernel/kprobes-decode.c
 create mode 100644 include/asm-arm/kprobes.h

(limited to 'include')

diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 593b56509f4f..9e0eebacd3d9 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_ISA_DMA)		+= dma-isa.o
 obj-$(CONFIG_PCI)		+= bios32.o isa.o
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_KPROBES)		+= kprobes-decode.o
 obj-$(CONFIG_OABI_COMPAT)	+= sys_oabi-compat.o
 
 obj-$(CONFIG_CRUNCH)		+= crunch.o crunch-bits.o
diff --git a/arch/arm/kernel/kprobes-decode.c b/arch/arm/kernel/kprobes-decode.c
new file mode 100644
index 000000000000..d51bc8b60557
--- /dev/null
+++ b/arch/arm/kernel/kprobes-decode.c
@@ -0,0 +1,1529 @@
+/*
+ * arch/arm/kernel/kprobes-decode.c
+ *
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+/*
+ * We do not have hardware single-stepping on ARM, This
+ * effort is further complicated by the ARM not having a
+ * "next PC" register.  Instructions that change the PC
+ * can't be safely single-stepped in a MP environment, so
+ * we have a lot of work to do:
+ *
+ * In the prepare phase:
+ *   *) If it is an instruction that does anything
+ *      with the CPU mode, we reject it for a kprobe.
+ *      (This is out of laziness rather than need.  The
+ *      instructions could be simulated.)
+ *
+ *   *) Otherwise, decode the instruction rewriting its
+ *      registers to take fixed, ordered registers and
+ *      setting a handler for it to run the instruction.
+ *
+ * In the execution phase by an instruction's handler:
+ *
+ *   *) If the PC is written to by the instruction, the
+ *      instruction must be fully simulated in software.
+ *      If it is a conditional instruction, the handler
+ *      will use insn[0] to copy its condition code to
+ *	set r0 to 1 and insn[1] to "mov pc, lr" to return.
+ *
+ *   *) Otherwise, a modified form of the instruction is
+ *      directly executed.  Its handler calls the
+ *      instruction in insn[0].  In insn[1] is a
+ *      "mov pc, lr" to return.
+ *
+ *      Before calling, load up the reordered registers
+ *      from the original instruction's registers.  If one
+ *      of the original input registers is the PC, compute
+ *      and adjust the appropriate input register.
+ *
+ *	After call completes, copy the output registers to
+ *      the original instruction's original registers.
+ *
+ * We don't use a real breakpoint instruction since that
+ * would have us in the kernel go from SVC mode to SVC
+ * mode losing the link register.  Instead we use an
+ * undefined instruction.  To simplify processing, the
+ * undefined instruction used for kprobes must be reserved
+ * exclusively for kprobes use.
+ *
+ * TODO: ifdef out some instruction decoding based on architecture.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+
+#define sign_extend(x, signbit) ((x) | (0 - ((x) & (1 << (signbit)))))
+
+#define branch_displacement(insn) sign_extend(((insn) & 0xffffff) << 2, 25)
+
+#define PSR_fs	(PSR_f|PSR_s)
+
+#define KPROBE_RETURN_INSTRUCTION	0xe1a0f00e	/* mov pc, lr */
+#define SET_R0_TRUE_INSTRUCTION		0xe3a00001	/* mov	r0, #1 */
+
+#define	truecc_insn(insn)	(((insn) & 0xf0000000) | \
+				 (SET_R0_TRUE_INSTRUCTION & 0x0fffffff))
+
+typedef long (insn_0arg_fn_t)(void);
+typedef long (insn_1arg_fn_t)(long);
+typedef long (insn_2arg_fn_t)(long, long);
+typedef long (insn_3arg_fn_t)(long, long, long);
+typedef long (insn_4arg_fn_t)(long, long, long, long);
+typedef long long (insn_llret_0arg_fn_t)(void);
+typedef long long (insn_llret_3arg_fn_t)(long, long, long);
+typedef long long (insn_llret_4arg_fn_t)(long, long, long, long);
+
+union reg_pair {
+	long long	dr;
+#ifdef __LITTLE_ENDIAN
+	struct { long	r0, r1; };
+#else
+	struct { long	r1, r0; };
+#endif
+};
+
+/*
+ * For STR and STM instructions, an ARM core may choose to use either
+ * a +8 or a +12 displacement from the current instruction's address.
+ * Whichever value is chosen for a given core, it must be the same for
+ * both instructions and may not change.  This function measures it.
+ */
+
+static int str_pc_offset;
+
+static void __init find_str_pc_offset(void)
+{
+	int addr, scratch, ret;
+
+	__asm__ (
+		"sub	%[ret], pc, #4		\n\t"
+		"str	pc, %[addr]		\n\t"
+		"ldr	%[scr], %[addr]		\n\t"
+		"sub	%[ret], %[scr], %[ret]	\n\t"
+		: [ret] "=r" (ret), [scr] "=r" (scratch), [addr] "+m" (addr));
+
+	str_pc_offset = ret;
+}
+
+/*
+ * The insnslot_?arg_r[w]flags() functions below are to keep the
+ * msr -> *fn -> mrs instruction sequences indivisible so that
+ * the state of the CPSR flags aren't inadvertently modified
+ * just before or just after the call.
+ */
+
+static inline long __kprobes
+insnslot_0arg_rflags(long cpsr, insn_0arg_fn_t *fn)
+{
+	register long ret asm("r0");
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret)
+		: [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	return ret;
+}
+
+static inline long long __kprobes
+insnslot_llret_0arg_rflags(long cpsr, insn_llret_0arg_fn_t *fn)
+{
+	register long ret0 asm("r0");
+	register long ret1 asm("r1");
+	union reg_pair fnr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret0), "=r" (ret1)
+		: [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	fnr.r0 = ret0;
+	fnr.r1 = ret1;
+	return fnr.dr;
+}
+
+static inline long __kprobes
+insnslot_1arg_rflags(long r0, long cpsr, insn_1arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long ret asm("r0");
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret)
+		: "0" (rr0), [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_2arg_rflags(long r0, long r1, long cpsr, insn_2arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long ret asm("r0");
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret)
+		: "0" (rr0), "r" (rr1),
+		  [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_3arg_rflags(long r0, long r1, long r2, long cpsr, insn_3arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long ret asm("r0");
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret)
+		: "0" (rr0), "r" (rr1), "r" (rr2),
+		  [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	return ret;
+}
+
+static inline long long __kprobes
+insnslot_llret_3arg_rflags(long r0, long r1, long r2, long cpsr,
+			   insn_llret_3arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long ret0 asm("r0");
+	register long ret1 asm("r1");
+	union reg_pair fnr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret0), "=r" (ret1)
+		: "0" (rr0), "r" (rr1), "r" (rr2),
+		  [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	fnr.r0 = ret0;
+	fnr.r1 = ret1;
+	return fnr.dr;
+}
+
+static inline long __kprobes
+insnslot_4arg_rflags(long r0, long r1, long r2, long r3, long cpsr,
+		     insn_4arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long rr3 asm("r3") = r3;
+	register long ret asm("r0");
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		: "=r" (ret)
+		: "0" (rr0), "r" (rr1), "r" (rr2), "r" (rr3),
+		  [cpsr] "r" (cpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_1arg_rwflags(long r0, long *cpsr, insn_1arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long ret asm("r0");
+	long oldcpsr = *cpsr;
+	long newcpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: "=r" (ret), [newcpsr] "=r" (newcpsr)
+		: "0" (rr0), [oldcpsr] "r" (oldcpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	*cpsr = (oldcpsr & ~PSR_fs) | (newcpsr & PSR_fs);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_2arg_rwflags(long r0, long r1, long *cpsr, insn_2arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long ret asm("r0");
+	long oldcpsr = *cpsr;
+	long newcpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: "=r" (ret), [newcpsr] "=r" (newcpsr)
+		: "0" (rr0), "r" (rr1), [oldcpsr] "r" (oldcpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	*cpsr = (oldcpsr & ~PSR_fs) | (newcpsr & PSR_fs);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_3arg_rwflags(long r0, long r1, long r2, long *cpsr,
+		      insn_3arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long ret asm("r0");
+	long oldcpsr = *cpsr;
+	long newcpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: "=r" (ret), [newcpsr] "=r" (newcpsr)
+		: "0" (rr0), "r" (rr1), "r" (rr2),
+		  [oldcpsr] "r" (oldcpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	*cpsr = (oldcpsr & ~PSR_fs) | (newcpsr & PSR_fs);
+	return ret;
+}
+
+static inline long __kprobes
+insnslot_4arg_rwflags(long r0, long r1, long r2, long r3, long *cpsr,
+		      insn_4arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long rr3 asm("r3") = r3;
+	register long ret asm("r0");
+	long oldcpsr = *cpsr;
+	long newcpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: "=r" (ret), [newcpsr] "=r" (newcpsr)
+		: "0" (rr0), "r" (rr1), "r" (rr2), "r" (rr3),
+		  [oldcpsr] "r" (oldcpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	*cpsr = (oldcpsr & ~PSR_fs) | (newcpsr & PSR_fs);
+	return ret;
+}
+
+static inline long long __kprobes
+insnslot_llret_4arg_rwflags(long r0, long r1, long r2, long r3, long *cpsr,
+			    insn_llret_4arg_fn_t *fn)
+{
+	register long rr0 asm("r0") = r0;
+	register long rr1 asm("r1") = r1;
+	register long rr2 asm("r2") = r2;
+	register long rr3 asm("r3") = r3;
+	register long ret0 asm("r0");
+	register long ret1 asm("r1");
+	long oldcpsr = *cpsr;
+	long newcpsr;
+	union reg_pair fnr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"mov	lr, pc			\n\t"
+		"mov	pc, %[fn]		\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: "=r" (ret0), "=r" (ret1), [newcpsr] "=r" (newcpsr)
+		: "0" (rr0), "r" (rr1), "r" (rr2), "r" (rr3),
+		  [oldcpsr] "r" (oldcpsr), [fn] "r" (fn)
+		: "lr", "cc"
+	);
+	*cpsr = (oldcpsr & ~PSR_fs) | (newcpsr & PSR_fs);
+	fnr.r0 = ret0;
+	fnr.r1 = ret1;
+	return fnr.dr;
+}
+
+/*
+ * To avoid the complications of mimicing single-stepping on a
+ * processor without a Next-PC or a single-step mode, and to
+ * avoid having to deal with the side-effects of boosting, we
+ * simulate or emulate (almost) all ARM instructions.
+ *
+ * "Simulation" is where the instruction's behavior is duplicated in
+ * C code.  "Emulation" is where the original instruction is rewritten
+ * and executed, often by altering its registers.
+ *
+ * By having all behavior of the kprobe'd instruction completed before
+ * returning from the kprobe_handler(), all locks (scheduler and
+ * interrupt) can safely be released.  There is no need for secondary
+ * breakpoints, no race with MP or preemptable kernels, nor having to
+ * clean up resources counts at a later time impacting overall system
+ * performance.  By rewriting the instruction, only the minimum registers
+ * need to be loaded and saved back optimizing performance.
+ *
+ * Calling the insnslot_*_rwflags version of a function doesn't hurt
+ * anything even when the CPSR flags aren't updated by the
+ * instruction.  It's just a little slower in return for saving
+ * a little space by not having a duplicate function that doesn't
+ * update the flags.  (The same optimization can be said for
+ * instructions that do or don't perform register writeback)
+ * Also, instructions can either read the flags, only write the
+ * flags, or read and write the flags.  To save combinations
+ * rather than for sheer performance, flag functions just assume
+ * read and write of flags.
+ */
+
+static void __kprobes simulate_bbl(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	long iaddr = (long)p->addr;
+	int disp  = branch_displacement(insn);
+
+	if (!insnslot_1arg_rflags(0, regs->ARM_cpsr, i_fn))
+		return;
+
+	if (insn & (1 << 24))
+		regs->ARM_lr = iaddr + 4;
+
+	regs->ARM_pc = iaddr + 8 + disp;
+}
+
+static void __kprobes simulate_blx1(struct kprobe *p, struct pt_regs *regs)
+{
+	kprobe_opcode_t insn = p->opcode;
+	long iaddr = (long)p->addr;
+	int disp = branch_displacement(insn);
+
+	regs->ARM_lr = iaddr + 4;
+	regs->ARM_pc = iaddr + 8 + disp + ((insn >> 23) & 0x2);
+	regs->ARM_cpsr |= PSR_T_BIT;
+}
+
+static void __kprobes simulate_blx2bx(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rm = insn & 0xf;
+	long rmv = regs->uregs[rm];
+
+	if (!insnslot_1arg_rflags(0, regs->ARM_cpsr, i_fn))
+		return;
+
+	if (insn & (1 << 5))
+		regs->ARM_lr = (long)p->addr + 4;
+
+	regs->ARM_pc = rmv & ~0x1;
+	regs->ARM_cpsr &= ~PSR_T_BIT;
+	if (rmv & 0x1)
+		regs->ARM_cpsr |= PSR_T_BIT;
+}
+
+static void __kprobes simulate_ldm1stm1(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rn = (insn >> 16) & 0xf;
+	int lbit = insn & (1 << 20);
+	int wbit = insn & (1 << 21);
+	int ubit = insn & (1 << 23);
+	int pbit = insn & (1 << 24);
+	long *addr = (long *)regs->uregs[rn];
+	int reg_bit_vector;
+	int reg_count;
+
+	if (!insnslot_1arg_rflags(0, regs->ARM_cpsr, i_fn))
+		return;
+
+	reg_count = 0;
+	reg_bit_vector = insn & 0xffff;
+	while (reg_bit_vector) {
+		reg_bit_vector &= (reg_bit_vector - 1);
+		++reg_count;
+	}
+
+	if (!ubit)
+		addr -= reg_count;
+	addr += (!pbit ^ !ubit);
+
+	reg_bit_vector = insn & 0xffff;
+	while (reg_bit_vector) {
+		int reg = __ffs(reg_bit_vector);
+		reg_bit_vector &= (reg_bit_vector - 1);
+		if (lbit)
+			regs->uregs[reg] = *addr++;
+		else
+			*addr++ = regs->uregs[reg];
+	}
+
+	if (wbit) {
+		if (!ubit)
+			addr -= reg_count;
+		addr -= (!pbit ^ !ubit);
+		regs->uregs[rn] = (long)addr;
+	}
+}
+
+static void __kprobes simulate_stm1_pc(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+
+	if (!insnslot_1arg_rflags(0, regs->ARM_cpsr, i_fn))
+		return;
+
+	regs->ARM_pc = (long)p->addr + str_pc_offset;
+	simulate_ldm1stm1(p, regs);
+	regs->ARM_pc = (long)p->addr + 4;
+}
+
+static void __kprobes simulate_mov_ipsp(struct kprobe *p, struct pt_regs *regs)
+{
+	regs->uregs[12] = regs->uregs[13];
+}
+
+static void __kprobes emulate_ldcstc(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rn = (insn >> 16) & 0xf;
+	long rnv = regs->uregs[rn];
+
+	/* Save Rn in case of writeback. */
+	regs->uregs[rn] = insnslot_1arg_rflags(rnv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_ldrd(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_2arg_fn_t *i_fn = (insn_2arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;  /* rm may be invalid, don't care. */
+
+	/* Not following the C calling convention here, so need asm(). */
+	__asm__ __volatile__ (
+		"ldr	r0, %[rn]	\n\t"
+		"ldr	r1, %[rm]	\n\t"
+		"msr	cpsr_fs, %[cpsr]\n\t"
+		"mov	lr, pc		\n\t"
+		"mov	pc, %[i_fn]	\n\t"
+		"str	r0, %[rn]	\n\t"	/* in case of writeback */
+		"str	r2, %[rd0]	\n\t"
+		"str	r3, %[rd1]	\n\t"
+		: [rn]  "+m" (regs->uregs[rn]),
+		  [rd0] "=m" (regs->uregs[rd]),
+		  [rd1] "=m" (regs->uregs[rd+1])
+		: [rm]   "m" (regs->uregs[rm]),
+		  [cpsr] "r" (regs->ARM_cpsr),
+		  [i_fn] "r" (i_fn)
+		: "r0", "r1", "r2", "r3", "lr", "cc"
+	);
+}
+
+static void __kprobes emulate_strd(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_4arg_fn_t *i_fn = (insn_4arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm  = insn & 0xf;
+	long rnv = regs->uregs[rn];
+	long rmv = regs->uregs[rm];  /* rm/rmv may be invalid, don't care. */
+
+	regs->uregs[rn] = insnslot_4arg_rflags(rnv, rmv, regs->uregs[rd],
+					       regs->uregs[rd+1],
+					       regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_ldr(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_llret_3arg_fn_t *i_fn = (insn_llret_3arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	union reg_pair fnr;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+	long rdv;
+	long rnv  = regs->uregs[rn];
+	long rmv  = regs->uregs[rm]; /* rm/rmv may be invalid, don't care. */
+	long cpsr = regs->ARM_cpsr;
+
+	fnr.dr = insnslot_llret_3arg_rflags(rnv, 0, rmv, cpsr, i_fn);
+	regs->uregs[rn] = fnr.r0;  /* Save Rn in case of writeback. */
+	rdv = fnr.r1;
+
+	if (rd == 15) {
+#if __LINUX_ARM_ARCH__ >= 5
+		cpsr &= ~PSR_T_BIT;
+		if (rdv & 0x1)
+			cpsr |= PSR_T_BIT;
+		regs->ARM_cpsr = cpsr;
+		rdv &= ~0x1;
+#else
+		rdv &= ~0x2;
+#endif
+	}
+	regs->uregs[rd] = rdv;
+}
+
+static void __kprobes emulate_str(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_3arg_fn_t *i_fn = (insn_3arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	long iaddr = (long)p->addr;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+	long rdv = (rd == 15) ? iaddr + str_pc_offset : regs->uregs[rd];
+	long rnv = (rn == 15) ? iaddr +  8 : regs->uregs[rn];
+	long rmv = regs->uregs[rm];  /* rm/rmv may be invalid, don't care. */
+
+	/* Save Rn in case of writeback. */
+	regs->uregs[rn] =
+		insnslot_3arg_rflags(rnv, rdv, rmv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_mrrc(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_llret_0arg_fn_t *i_fn = (insn_llret_0arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	union reg_pair fnr;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+
+	fnr.dr = insnslot_llret_0arg_rflags(regs->ARM_cpsr, i_fn);
+	regs->uregs[rn] = fnr.r0;
+	regs->uregs[rd] = fnr.r1;
+}
+
+static void __kprobes emulate_mcrr(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_2arg_fn_t *i_fn = (insn_2arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	long rnv = regs->uregs[rn];
+	long rdv = regs->uregs[rd];
+
+	insnslot_2arg_rflags(rnv, rdv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_sat(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rm = insn & 0xf;
+	long rmv = regs->uregs[rm];
+
+	/* Writes Q flag */
+	regs->uregs[rd] = insnslot_1arg_rwflags(rmv, &regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_sel(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_2arg_fn_t *i_fn = (insn_2arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+	long rnv = regs->uregs[rn];
+	long rmv = regs->uregs[rm];
+
+	/* Reads GE bits */
+	regs->uregs[rd] = insnslot_2arg_rflags(rnv, rmv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_none(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_0arg_fn_t *i_fn = (insn_0arg_fn_t *)&p->ainsn.insn[0];
+
+	insnslot_0arg_rflags(regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_rd12(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_0arg_fn_t *i_fn = (insn_0arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+
+	regs->uregs[rd] = insnslot_0arg_rflags(regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_ird12(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int ird = (insn >> 12) & 0xf;
+
+	insnslot_1arg_rflags(regs->uregs[ird], regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_rn16(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rn = (insn >> 16) & 0xf;
+	long rnv = regs->uregs[rn];
+
+	insnslot_1arg_rflags(rnv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes emulate_rd12rm0(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rm = insn & 0xf;
+	long rmv = regs->uregs[rm];
+
+	regs->uregs[rd] = insnslot_1arg_rflags(rmv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_rd12rn16rm0_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_2arg_fn_t *i_fn = (insn_2arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+	long rnv = regs->uregs[rn];
+	long rmv = regs->uregs[rm];
+
+	regs->uregs[rd] =
+		insnslot_2arg_rwflags(rnv, rmv, &regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_rd16rn12rs8rm0_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_3arg_fn_t *i_fn = (insn_3arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 16) & 0xf;
+	int rn = (insn >> 12) & 0xf;
+	int rs = (insn >> 8) & 0xf;
+	int rm = insn & 0xf;
+	long rnv = regs->uregs[rn];
+	long rsv = regs->uregs[rs];
+	long rmv = regs->uregs[rm];
+
+	regs->uregs[rd] =
+		insnslot_3arg_rwflags(rnv, rsv, rmv, &regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_rd16rs8rm0_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_2arg_fn_t *i_fn = (insn_2arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 16) & 0xf;
+	int rs = (insn >> 8) & 0xf;
+	int rm = insn & 0xf;
+	long rsv = regs->uregs[rs];
+	long rmv = regs->uregs[rm];
+
+	regs->uregs[rd] =
+		insnslot_2arg_rwflags(rsv, rmv, &regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_rdhi16rdlo12rs8rm0_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_llret_4arg_fn_t *i_fn = (insn_llret_4arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	union reg_pair fnr;
+	int rdhi = (insn >> 16) & 0xf;
+	int rdlo = (insn >> 12) & 0xf;
+	int rs   = (insn >> 8) & 0xf;
+	int rm   = insn & 0xf;
+	long rsv = regs->uregs[rs];
+	long rmv = regs->uregs[rm];
+
+	fnr.dr = insnslot_llret_4arg_rwflags(regs->uregs[rdhi],
+					     regs->uregs[rdlo], rsv, rmv,
+					     &regs->ARM_cpsr, i_fn);
+	regs->uregs[rdhi] = fnr.r0;
+	regs->uregs[rdlo] = fnr.r1;
+}
+
+static void __kprobes
+emulate_alu_imm_rflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	long rnv = (rn == 15) ? (long)p->addr + 8 : regs->uregs[rn];
+
+	regs->uregs[rd] = insnslot_1arg_rflags(rnv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_alu_imm_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_1arg_fn_t *i_fn = (insn_1arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	long rnv = (rn == 15) ? (long)p->addr + 8 : regs->uregs[rn];
+
+	regs->uregs[rd] = insnslot_1arg_rwflags(rnv, &regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_alu_rflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_3arg_fn_t *i_fn = (insn_3arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	long ppc = (long)p->addr + 8;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;	/* rn/rnv/rs/rsv may be */
+	int rs = (insn >> 8) & 0xf;	/* invalid, don't care. */
+	int rm = insn & 0xf;
+	long rnv = (rn == 15) ? ppc : regs->uregs[rn];
+	long rmv = (rm == 15) ? ppc : regs->uregs[rm];
+	long rsv = regs->uregs[rs];
+
+	regs->uregs[rd] =
+		insnslot_3arg_rflags(rnv, rmv, rsv, regs->ARM_cpsr, i_fn);
+}
+
+static void __kprobes
+emulate_alu_rwflags(struct kprobe *p, struct pt_regs *regs)
+{
+	insn_3arg_fn_t *i_fn = (insn_3arg_fn_t *)&p->ainsn.insn[0];
+	kprobe_opcode_t insn = p->opcode;
+	long ppc = (long)p->addr + 8;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;	/* rn/rnv/rs/rsv may be */
+	int rs = (insn >> 8) & 0xf;	/* invalid, don't care. */
+	int rm = insn & 0xf;
+	long rnv = (rn == 15) ? ppc : regs->uregs[rn];
+	long rmv = (rm == 15) ? ppc : regs->uregs[rm];
+	long rsv = regs->uregs[rs];
+
+	regs->uregs[rd] =
+		insnslot_3arg_rwflags(rnv, rmv, rsv, &regs->ARM_cpsr, i_fn);
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_ldr_str(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	int ibit = (insn & (1 << 26)) ? 25 : 22;
+
+	insn &= 0xfff00fff;
+	insn |= 0x00001000;	/* Rn = r0, Rd = r1 */
+	if (insn & (1 << ibit)) {
+		insn &= ~0xf;
+		insn |= 2;	/* Rm = r2 */
+	}
+	asi->insn[0] = insn;
+	asi->insn_handler = (insn & (1 << 20)) ? emulate_ldr : emulate_str;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rd12rm0(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	insn &= 0xffff0ff0;	/* Rd = r0, Rm = r0 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rd12rm0;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rd12(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	insn &= 0xffff0fff;	/* Rd = r0 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rd12;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rd12rn16rm0_wflags(kprobe_opcode_t insn,
+				struct arch_specific_insn *asi)
+{
+	insn &= 0xfff00ff0;	/* Rd = r0, Rn = r0 */
+	insn |= 0x00000001;	/* Rm = r1 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rd12rn16rm0_rwflags;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rd16rs8rm0_wflags(kprobe_opcode_t insn,
+			       struct arch_specific_insn *asi)
+{
+	insn &= 0xfff0f0f0;	/* Rd = r0, Rs = r0 */
+	insn |= 0x00000001;	/* Rm = r1          */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rd16rs8rm0_rwflags;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rd16rn12rs8rm0_wflags(kprobe_opcode_t insn,
+				   struct arch_specific_insn *asi)
+{
+	insn &= 0xfff000f0;	/* Rd = r0, Rn = r0 */
+	insn |= 0x00000102;	/* Rs = r1, Rm = r2 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rd16rn12rs8rm0_rwflags;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+prep_emulate_rdhi16rdlo12rs8rm0_wflags(kprobe_opcode_t insn,
+				       struct arch_specific_insn *asi)
+{
+	insn &= 0xfff000f0;	/* RdHi = r0, RdLo = r1 */
+	insn |= 0x00001203;	/* Rs = r2, Rm = r3 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_rdhi16rdlo12rs8rm0_rwflags;
+	return INSN_GOOD;
+}
+
+/*
+ * For the instruction masking and comparisons in all the "space_*"
+ * functions below, Do _not_ rearrange the order of tests unless
+ * you're very, very sure of what you are doing.  For the sake of
+ * efficiency, the masks for some tests sometimes assume other test
+ * have been done prior to them so the number of patterns to test
+ * for an instruction set can be as broad as possible to reduce the
+ * number of tests needed.
+ */
+
+static enum kprobe_insn __kprobes
+space_1111(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* CPS mmod == 1 : 1111 0001 0000 xx10 xxxx xxxx xx0x xxxx */
+	/* RFE           : 1111 100x x0x1 xxxx xxxx 1010 xxxx xxxx */
+	/* SRS           : 1111 100x x1x0 1101 xxxx 0101 xxxx xxxx */
+	if ((insn & 0xfff30020) == 0xf1020000 ||
+	    (insn & 0xfe500f00) == 0xf8100a00 ||
+	    (insn & 0xfe5f0f00) == 0xf84d0500)
+		return INSN_REJECTED;
+
+	/* PLD : 1111 01x1 x101 xxxx xxxx xxxx xxxx xxxx : */
+	if ((insn & 0xfd700000) == 0xf4500000) {
+		insn &= 0xfff0ffff;	/* Rn = r0 */
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_rn16;
+		return INSN_GOOD;
+	}
+
+	/* BLX(1) : 1111 101x xxxx xxxx xxxx xxxx xxxx xxxx : */
+	if ((insn & 0xfe000000) == 0xfa000000) {
+		asi->insn_handler = simulate_blx1;
+		return INSN_GOOD_NO_SLOT;
+	}
+
+	/* SETEND : 1111 0001 0000 0001 xxxx xxxx 0000 xxxx */
+	/* CDP2   : 1111 1110 xxxx xxxx xxxx xxxx xxx0 xxxx */
+	if ((insn & 0xffff00f0) == 0xf1010000 ||
+	    (insn & 0xff000010) == 0xfe000000) {
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_none;
+		return INSN_GOOD;
+	}
+
+	/* MCRR2 : 1111 1100 0100 xxxx xxxx xxxx xxxx xxxx : (Rd != Rn) */
+	/* MRRC2 : 1111 1100 0101 xxxx xxxx xxxx xxxx xxxx : (Rd != Rn) */
+	if ((insn & 0xffe00000) == 0xfc400000) {
+		insn &= 0xfff00fff;	/* Rn = r0 */
+		insn |= 0x00001000;	/* Rd = r1 */
+		asi->insn[0] = insn;
+		asi->insn_handler =
+			(insn & (1 << 20)) ? emulate_mrrc : emulate_mcrr;
+		return INSN_GOOD;
+	}
+
+	/* LDC2 : 1111 110x xxx1 xxxx xxxx xxxx xxxx xxxx */
+	/* STC2 : 1111 110x xxx0 xxxx xxxx xxxx xxxx xxxx */
+	if ((insn & 0xfe000000) == 0xfc000000) {
+		insn &= 0xfff0ffff;      /* Rn = r0 */
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_ldcstc;
+		return INSN_GOOD;
+	}
+
+	/* MCR2 : 1111 1110 xxx0 xxxx xxxx xxxx xxx1 xxxx */
+	/* MRC2 : 1111 1110 xxx1 xxxx xxxx xxxx xxx1 xxxx */
+	insn &= 0xffff0fff;	/* Rd = r0 */
+	asi->insn[0]      = insn;
+	asi->insn_handler = (insn & (1 << 20)) ? emulate_rd12 : emulate_ird12;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_000x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* cccc 0001 0xx0 xxxx xxxx xxxx xxxx xxx0 xxxx */
+	if ((insn & 0x0f900010) == 0x01000000) {
+
+		/* BXJ  : cccc 0001 0010 xxxx xxxx xxxx 0010 xxxx */
+		/* MSR  : cccc 0001 0x10 xxxx xxxx xxxx 0000 xxxx */
+		if ((insn & 0x0ff000f0) == 0x01200020 ||
+		    (insn & 0x0fb000f0) == 0x01200000)
+			return INSN_REJECTED;
+
+		/* MRS : cccc 0001 0x00 xxxx xxxx xxxx 0000 xxxx */
+		if ((insn & 0x0fb00010) == 0x01000000)
+			return prep_emulate_rd12(insn, asi);
+
+		/* SMLALxy : cccc 0001 0100 xxxx xxxx xxxx 1xx0 xxxx */
+		if ((insn & 0x0ff00090) == 0x01400080)
+			return prep_emulate_rdhi16rdlo12rs8rm0_wflags(insn, asi);
+
+		/* SMULWy : cccc 0001 0010 xxxx xxxx xxxx 1x10 xxxx */
+		/* SMULxy : cccc 0001 0110 xxxx xxxx xxxx 1xx0 xxxx */
+		if ((insn & 0x0ff000b0) == 0x012000a0 ||
+		    (insn & 0x0ff00090) == 0x01600080)
+			return prep_emulate_rd16rs8rm0_wflags(insn, asi);
+
+		/* SMLAxy : cccc 0001 0000 xxxx xxxx xxxx 1xx0 xxxx : Q */
+		/* SMLAWy : cccc 0001 0010 xxxx xxxx xxxx 0x00 xxxx : Q */
+		return prep_emulate_rd16rn12rs8rm0_wflags(insn, asi);
+
+	}
+
+	/* cccc 0001 0xx0 xxxx xxxx xxxx xxxx 0xx1 xxxx */
+	else if ((insn & 0x0f900090) == 0x01000010) {
+
+		/* BKPT : 1110 0001 0010 xxxx xxxx xxxx 0111 xxxx */
+		if ((insn & 0xfff000f0) == 0xe1200070)
+			return INSN_REJECTED;
+
+		/* BLX(2) : cccc 0001 0010 xxxx xxxx xxxx 0011 xxxx */
+		/* BX     : cccc 0001 0010 xxxx xxxx xxxx 0001 xxxx */
+		if ((insn & 0x0ff000d0) == 0x01200010) {
+			asi->insn[0] = truecc_insn(insn);
+			asi->insn_handler = simulate_blx2bx;
+			return INSN_GOOD;
+		}
+
+		/* CLZ : cccc 0001 0110 xxxx xxxx xxxx 0001 xxxx */
+		if ((insn & 0x0ff000f0) == 0x01600010)
+			return prep_emulate_rd12rm0(insn, asi);
+
+		/* QADD    : cccc 0001 0000 xxxx xxxx xxxx 0101 xxxx :Q */
+		/* QSUB    : cccc 0001 0010 xxxx xxxx xxxx 0101 xxxx :Q */
+		/* QDADD   : cccc 0001 0100 xxxx xxxx xxxx 0101 xxxx :Q */
+		/* QDSUB   : cccc 0001 0110 xxxx xxxx xxxx 0101 xxxx :Q */
+		return prep_emulate_rd12rn16rm0_wflags(insn, asi);
+	}
+
+	/* cccc 0000 xxxx xxxx xxxx xxxx xxxx 1001 xxxx */
+	else if ((insn & 0x0f000090) == 0x00000090) {
+
+		/* MUL    : cccc 0000 0000 xxxx xxxx xxxx 1001 xxxx :   */
+		/* MULS   : cccc 0000 0001 xxxx xxxx xxxx 1001 xxxx :cc */
+		/* MLA    : cccc 0000 0010 xxxx xxxx xxxx 1001 xxxx :   */
+		/* MLAS   : cccc 0000 0011 xxxx xxxx xxxx 1001 xxxx :cc */
+		/* UMAAL  : cccc 0000 0100 xxxx xxxx xxxx 1001 xxxx :   */
+		/* UMULL  : cccc 0000 1000 xxxx xxxx xxxx 1001 xxxx :   */
+		/* UMULLS : cccc 0000 1001 xxxx xxxx xxxx 1001 xxxx :cc */
+		/* UMLAL  : cccc 0000 1010 xxxx xxxx xxxx 1001 xxxx :   */
+		/* UMLALS : cccc 0000 1011 xxxx xxxx xxxx 1001 xxxx :cc */
+		/* SMULL  : cccc 0000 1100 xxxx xxxx xxxx 1001 xxxx :   */
+		/* SMULLS : cccc 0000 1101 xxxx xxxx xxxx 1001 xxxx :cc */
+		/* SMLAL  : cccc 0000 1110 xxxx xxxx xxxx 1001 xxxx :   */
+		/* SMLALS : cccc 0000 1111 xxxx xxxx xxxx 1001 xxxx :cc */
+		if ((insn & 0x0fe000f0) == 0x00000090) {
+		       return prep_emulate_rd16rs8rm0_wflags(insn, asi);
+		} else if  ((insn & 0x0fe000f0) == 0x00200090) {
+		       return prep_emulate_rd16rn12rs8rm0_wflags(insn, asi);
+		} else {
+		       return prep_emulate_rdhi16rdlo12rs8rm0_wflags(insn, asi);
+		}
+	}
+
+	/* cccc 000x xxxx xxxx xxxx xxxx xxxx 1xx1 xxxx */
+	else if ((insn & 0x0e000090) == 0x00000090) {
+
+		/* SWP   : cccc 0001 0000 xxxx xxxx xxxx 1001 xxxx */
+		/* SWPB  : cccc 0001 0100 xxxx xxxx xxxx 1001 xxxx */
+		/* LDRD  : cccc 000x xxx0 xxxx xxxx xxxx 1101 xxxx */
+		/* STRD  : cccc 000x xxx0 xxxx xxxx xxxx 1111 xxxx */
+		/* STREX : cccc 0001 1000 xxxx xxxx xxxx 1001 xxxx */
+		/* LDREX : cccc 0001 1001 xxxx xxxx xxxx 1001 xxxx */
+		/* LDRH  : cccc 000x xxx1 xxxx xxxx xxxx 1011 xxxx */
+		/* STRH  : cccc 000x xxx0 xxxx xxxx xxxx 1011 xxxx */
+		/* LDRSB : cccc 000x xxx1 xxxx xxxx xxxx 1101 xxxx */
+		/* LDRSH : cccc 000x xxx1 xxxx xxxx xxxx 1111 xxxx */
+		if ((insn & 0x0fb000f0) == 0x01000090) {
+			/* SWP/SWPB */
+			return prep_emulate_rd12rn16rm0_wflags(insn, asi);
+		} else if ((insn & 0x0e1000d0) == 0x00000d0) {
+			/* STRD/LDRD */
+			insn &= 0xfff00fff;
+			insn |= 0x00002000;	/* Rn = r0, Rd = r2 */
+			if (insn & (1 << 22)) {
+				/* I bit */
+				insn &= ~0xf;
+				insn |= 1;	/* Rm = r1 */
+			}
+			asi->insn[0] = insn;
+			asi->insn_handler =
+				(insn & (1 << 5)) ? emulate_strd : emulate_ldrd;
+			return INSN_GOOD;
+		}
+
+		return prep_emulate_ldr_str(insn, asi);
+	}
+
+	/* cccc 000x xxxx xxxx xxxx xxxx xxxx xxxx xxxx */
+
+	/*
+	 * ALU op with S bit and Rd == 15 :
+	 * 	cccc 000x xxx1 xxxx 1111 xxxx xxxx xxxx
+	 */
+	if ((insn & 0x0e10f000) == 0x0010f000)
+		return INSN_REJECTED;
+
+	/*
+	 * "mov ip, sp" is the most common kprobe'd instruction by far.
+	 * Check and optimize for it explicitly.
+	 */
+	if (insn == 0xe1a0c00d) {
+		asi->insn_handler = simulate_mov_ipsp;
+		return INSN_GOOD_NO_SLOT;
+	}
+
+	/*
+	 * Data processing: Immediate-shift / Register-shift
+	 * ALU op : cccc 000x xxxx xxxx xxxx xxxx xxxx xxxx
+	 * CPY    : cccc 0001 1010 xxxx xxxx 0000 0000 xxxx
+	 * MOV    : cccc 0001 101x xxxx xxxx xxxx xxxx xxxx
+	 * *S (bit 20) updates condition codes
+	 * ADC/SBC/RSC reads the C flag
+	 */
+	insn &= 0xfff00ff0;	/* Rn = r0, Rd = r0 */
+	insn |= 0x00000001;	/* Rm = r1 */
+	if (insn & 0x010) {
+		insn &= 0xfffff0ff;     /* register shift */
+		insn |= 0x00000200;     /* Rs = r2 */
+	}
+	asi->insn[0] = insn;
+	asi->insn_handler = (insn & (1 << 20)) ?  /* S-bit */
+				emulate_alu_rwflags : emulate_alu_rflags;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_001x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/*
+	 * MSR   : cccc 0011 0x10 xxxx xxxx xxxx xxxx xxxx
+	 * Undef : cccc 0011 0x00 xxxx xxxx xxxx xxxx xxxx
+	 * ALU op with S bit and Rd == 15 :
+	 *	   cccc 001x xxx1 xxxx 1111 xxxx xxxx xxxx
+	 */
+	if ((insn & 0x0f900000) == 0x03200000 ||	/* MSR & Undef */
+	    (insn & 0x0e10f000) == 0x0210f000)		/* ALU s-bit, R15  */
+		return INSN_REJECTED;
+
+	/*
+	 * Data processing: 32-bit Immediate
+	 * ALU op : cccc 001x xxxx xxxx xxxx xxxx xxxx xxxx
+	 * MOV    : cccc 0011 101x xxxx xxxx xxxx xxxx xxxx
+	 * *S (bit 20) updates condition codes
+	 * ADC/SBC/RSC reads the C flag
+	 */
+	insn &= 0xfff00ff0;	/* Rn = r0, Rd = r0 */
+	asi->insn[0] = insn;
+	asi->insn_handler = (insn & (1 << 20)) ?  /* S-bit */
+			emulate_alu_imm_rwflags : emulate_alu_imm_rflags;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_0110__1(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* SEL : cccc 0110 1000 xxxx xxxx xxxx 1011 xxxx GE: !!! */
+	if ((insn & 0x0ff000f0) == 0x068000b0) {
+		insn &= 0xfff00ff0;	/* Rd = r0, Rn = r0 */
+		insn |= 0x00000001;	/* Rm = r1 */
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_sel;
+		return INSN_GOOD;
+	}
+
+	/* SSAT   : cccc 0110 101x xxxx xxxx xxxx xx01 xxxx :Q */
+	/* USAT   : cccc 0110 111x xxxx xxxx xxxx xx01 xxxx :Q */
+	/* SSAT16 : cccc 0110 1010 xxxx xxxx xxxx 0011 xxxx :Q */
+	/* USAT16 : cccc 0110 1110 xxxx xxxx xxxx 0011 xxxx :Q */
+	if ((insn & 0x0fa00030) == 0x06a00010 ||
+	    (insn & 0x0fb000f0) == 0x06a00030) {
+		insn &= 0xffff0ff0;	/* Rd = r0, Rm = r0 */
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_sat;
+		return INSN_GOOD;
+	}
+
+	/* REV    : cccc 0110 1011 xxxx xxxx xxxx 0011 xxxx */
+	/* REV16  : cccc 0110 1011 xxxx xxxx xxxx 1011 xxxx */
+	/* REVSH  : cccc 0110 1111 xxxx xxxx xxxx 1011 xxxx */
+	if ((insn & 0x0ff00070) == 0x06b00030 ||
+	    (insn & 0x0ff000f0) == 0x06f000b0)
+		return prep_emulate_rd12rm0(insn, asi);
+
+	/* SADD16    : cccc 0110 0001 xxxx xxxx xxxx 0001 xxxx :GE */
+	/* SADDSUBX  : cccc 0110 0001 xxxx xxxx xxxx 0011 xxxx :GE */
+	/* SSUBADDX  : cccc 0110 0001 xxxx xxxx xxxx 0101 xxxx :GE */
+	/* SSUB16    : cccc 0110 0001 xxxx xxxx xxxx 0111 xxxx :GE */
+	/* SADD8     : cccc 0110 0001 xxxx xxxx xxxx 1001 xxxx :GE */
+	/* SSUB8     : cccc 0110 0001 xxxx xxxx xxxx 1111 xxxx :GE */
+	/* QADD16    : cccc 0110 0010 xxxx xxxx xxxx 0001 xxxx :   */
+	/* QADDSUBX  : cccc 0110 0010 xxxx xxxx xxxx 0011 xxxx :   */
+	/* QSUBADDX  : cccc 0110 0010 xxxx xxxx xxxx 0101 xxxx :   */
+	/* QSUB16    : cccc 0110 0010 xxxx xxxx xxxx 0111 xxxx :   */
+	/* QADD8     : cccc 0110 0010 xxxx xxxx xxxx 1001 xxxx :   */
+	/* QSUB8     : cccc 0110 0010 xxxx xxxx xxxx 1111 xxxx :   */
+	/* SHADD16   : cccc 0110 0011 xxxx xxxx xxxx 0001 xxxx :   */
+	/* SHADDSUBX : cccc 0110 0011 xxxx xxxx xxxx 0011 xxxx :   */
+	/* SHSUBADDX : cccc 0110 0011 xxxx xxxx xxxx 0101 xxxx :   */
+	/* SHSUB16   : cccc 0110 0011 xxxx xxxx xxxx 0111 xxxx :   */
+	/* SHADD8    : cccc 0110 0011 xxxx xxxx xxxx 1001 xxxx :   */
+	/* SHSUB8    : cccc 0110 0011 xxxx xxxx xxxx 1111 xxxx :   */
+	/* UADD16    : cccc 0110 0101 xxxx xxxx xxxx 0001 xxxx :GE */
+	/* UADDSUBX  : cccc 0110 0101 xxxx xxxx xxxx 0011 xxxx :GE */
+	/* USUBADDX  : cccc 0110 0101 xxxx xxxx xxxx 0101 xxxx :GE */
+	/* USUB16    : cccc 0110 0101 xxxx xxxx xxxx 0111 xxxx :GE */
+	/* UADD8     : cccc 0110 0101 xxxx xxxx xxxx 1001 xxxx :GE */
+	/* USUB8     : cccc 0110 0101 xxxx xxxx xxxx 1111 xxxx :GE */
+	/* UQADD16   : cccc 0110 0110 xxxx xxxx xxxx 0001 xxxx :   */
+	/* UQADDSUBX : cccc 0110 0110 xxxx xxxx xxxx 0011 xxxx :   */
+	/* UQSUBADDX : cccc 0110 0110 xxxx xxxx xxxx 0101 xxxx :   */
+	/* UQSUB16   : cccc 0110 0110 xxxx xxxx xxxx 0111 xxxx :   */
+	/* UQADD8    : cccc 0110 0110 xxxx xxxx xxxx 1001 xxxx :   */
+	/* UQSUB8    : cccc 0110 0110 xxxx xxxx xxxx 1111 xxxx :   */
+	/* UHADD16   : cccc 0110 0111 xxxx xxxx xxxx 0001 xxxx :   */
+	/* UHADDSUBX : cccc 0110 0111 xxxx xxxx xxxx 0011 xxxx :   */
+	/* UHSUBADDX : cccc 0110 0111 xxxx xxxx xxxx 0101 xxxx :   */
+	/* UHSUB16   : cccc 0110 0111 xxxx xxxx xxxx 0111 xxxx :   */
+	/* UHADD8    : cccc 0110 0111 xxxx xxxx xxxx 1001 xxxx :   */
+	/* UHSUB8    : cccc 0110 0111 xxxx xxxx xxxx 1111 xxxx :   */
+	/* PKHBT     : cccc 0110 1000 xxxx xxxx xxxx x001 xxxx :   */
+	/* PKHTB     : cccc 0110 1000 xxxx xxxx xxxx x101 xxxx :   */
+	/* SXTAB16   : cccc 0110 1000 xxxx xxxx xxxx 0111 xxxx :   */
+	/* SXTB      : cccc 0110 1010 xxxx xxxx xxxx 0111 xxxx :   */
+	/* SXTAB     : cccc 0110 1010 xxxx xxxx xxxx 0111 xxxx :   */
+	/* SXTAH     : cccc 0110 1011 xxxx xxxx xxxx 0111 xxxx :   */
+	/* UXTAB16   : cccc 0110 1100 xxxx xxxx xxxx 0111 xxxx :   */
+	/* UXTAB     : cccc 0110 1110 xxxx xxxx xxxx 0111 xxxx :   */
+	/* UXTAH     : cccc 0110 1111 xxxx xxxx xxxx 0111 xxxx :   */
+	return prep_emulate_rd12rn16rm0_wflags(insn, asi);
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_0111__1(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* Undef : cccc 0111 1111 xxxx xxxx xxxx 1111 xxxx */
+	if ((insn & 0x0ff000f0) == 0x03f000f0)
+		return INSN_REJECTED;
+
+	/* USADA8 : cccc 0111 1000 xxxx xxxx xxxx 0001 xxxx */
+	/* USAD8  : cccc 0111 1000 xxxx 1111 xxxx 0001 xxxx */
+	if ((insn & 0x0ff000f0) == 0x07800010)
+		 return prep_emulate_rd16rn12rs8rm0_wflags(insn, asi);
+
+	/* SMLALD : cccc 0111 0100 xxxx xxxx xxxx 00x1 xxxx */
+	/* SMLSLD : cccc 0111 0100 xxxx xxxx xxxx 01x1 xxxx */
+	if ((insn & 0x0ff00090) == 0x07400010)
+		return prep_emulate_rdhi16rdlo12rs8rm0_wflags(insn, asi);
+
+	/* SMLAD  : cccc 0111 0000 xxxx xxxx xxxx 00x1 xxxx :Q */
+	/* SMLSD  : cccc 0111 0000 xxxx xxxx xxxx 01x1 xxxx :Q */
+	/* SMMLA  : cccc 0111 0101 xxxx xxxx xxxx 00x1 xxxx :  */
+	/* SMMLS  : cccc 0111 0101 xxxx xxxx xxxx 11x1 xxxx :  */
+	if ((insn & 0x0ff00090) == 0x07000010 ||
+	    (insn & 0x0ff000d0) == 0x07500010 ||
+	    (insn & 0x0ff000d0) == 0x075000d0)
+		return prep_emulate_rd16rn12rs8rm0_wflags(insn, asi);
+
+	/* SMUSD  : cccc 0111 0000 xxxx xxxx xxxx 01x1 xxxx :  */
+	/* SMUAD  : cccc 0111 0000 xxxx 1111 xxxx 00x1 xxxx :Q */
+	/* SMMUL  : cccc 0111 0101 xxxx 1111 xxxx 00x1 xxxx :  */
+	return prep_emulate_rd16rs8rm0_wflags(insn, asi);
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_01xx(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* LDR   : cccc 01xx x0x1 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRB  : cccc 01xx x1x1 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRBT : cccc 01x0 x111 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRT  : cccc 01x0 x011 xxxx xxxx xxxx xxxx xxxx */
+	/* STR   : cccc 01xx x0x0 xxxx xxxx xxxx xxxx xxxx */
+	/* STRB  : cccc 01xx x1x0 xxxx xxxx xxxx xxxx xxxx */
+	/* STRBT : cccc 01x0 x110 xxxx xxxx xxxx xxxx xxxx */
+	/* STRT  : cccc 01x0 x010 xxxx xxxx xxxx xxxx xxxx */
+	return prep_emulate_ldr_str(insn, asi);
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_100x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* LDM(2) : cccc 100x x101 xxxx 0xxx xxxx xxxx xxxx */
+	/* LDM(3) : cccc 100x x1x1 xxxx 1xxx xxxx xxxx xxxx */
+	if ((insn & 0x0e708000) == 0x85000000 ||
+	    (insn & 0x0e508000) == 0x85010000)
+		return INSN_REJECTED;
+
+	/* LDM(1) : cccc 100x x0x1 xxxx xxxx xxxx xxxx xxxx */
+	/* STM(1) : cccc 100x x0x0 xxxx xxxx xxxx xxxx xxxx */
+	asi->insn[0] = truecc_insn(insn);
+	asi->insn_handler = ((insn & 0x108000) == 0x008000) ? /* STM & R15 */
+				simulate_stm1_pc : simulate_ldm1stm1;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_101x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* B  : cccc 1010 xxxx xxxx xxxx xxxx xxxx xxxx */
+	/* BL : cccc 1011 xxxx xxxx xxxx xxxx xxxx xxxx */
+	asi->insn[0] = truecc_insn(insn);
+	asi->insn_handler = simulate_bbl;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_1100_010x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* MCRR : cccc 1100 0100 xxxx xxxx xxxx xxxx xxxx : (Rd!=Rn) */
+	/* MRRC : cccc 1100 0101 xxxx xxxx xxxx xxxx xxxx : (Rd!=Rn) */
+	insn &= 0xfff00fff;
+	insn |= 0x00001000;	/* Rn = r0, Rd = r1 */
+	asi->insn[0] = insn;
+	asi->insn_handler = (insn & (1 << 20)) ? emulate_mrrc : emulate_mcrr;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_110x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* LDC : cccc 110x xxx1 xxxx xxxx xxxx xxxx xxxx */
+	/* STC : cccc 110x xxx0 xxxx xxxx xxxx xxxx xxxx */
+	insn &= 0xfff0ffff;	/* Rn = r0 */
+	asi->insn[0] = insn;
+	asi->insn_handler = emulate_ldcstc;
+	return INSN_GOOD;
+}
+
+static enum kprobe_insn __kprobes
+space_cccc_111x(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	/* BKPT : 1110 0001 0010 xxxx xxxx xxxx 0111 xxxx */
+	/* SWI  : cccc 1111 xxxx xxxx xxxx xxxx xxxx xxxx */
+	if ((insn & 0xfff000f0) == 0xe1200070 ||
+	    (insn & 0x0f000000) == 0x0f000000)
+		return INSN_REJECTED;
+
+	/* CDP : cccc 1110 xxxx xxxx xxxx xxxx xxx0 xxxx */
+	if ((insn & 0x0f000010) == 0x0e000000) {
+		asi->insn[0] = insn;
+		asi->insn_handler = emulate_none;
+		return INSN_GOOD;
+	}
+
+	/* MCR : cccc 1110 xxx0 xxxx xxxx xxxx xxx1 xxxx */
+	/* MRC : cccc 1110 xxx1 xxxx xxxx xxxx xxx1 xxxx */
+	insn &= 0xffff0fff;	/* Rd = r0 */
+	asi->insn[0] = insn;
+	asi->insn_handler = (insn & (1 << 20)) ? emulate_rd12 : emulate_ird12;
+	return INSN_GOOD;
+}
+
+/* Return:
+ *   INSN_REJECTED     If instruction is one not allowed to kprobe,
+ *   INSN_GOOD         If instruction is supported and uses instruction slot,
+ *   INSN_GOOD_NO_SLOT If instruction is supported but doesn't use its slot.
+ *
+ * For instructions we don't want to kprobe (INSN_REJECTED return result):
+ *   These are generally ones that modify the processor state making
+ *   them "hard" to simulate such as switches processor modes or
+ *   make accesses in alternate modes.  Any of these could be simulated
+ *   if the work was put into it, but low return considering they
+ *   should also be very rare.
+ */
+enum kprobe_insn __kprobes
+arm_kprobe_decode_insn(kprobe_opcode_t insn, struct arch_specific_insn *asi)
+{
+	asi->insn[1] = KPROBE_RETURN_INSTRUCTION;
+
+	if ((insn & 0xf0000000) == 0xf0000000) {
+
+		return space_1111(insn, asi);
+
+	} else if ((insn & 0x0e000000) == 0x00000000) {
+
+		return space_cccc_000x(insn, asi);
+
+	} else if ((insn & 0x0e000000) == 0x02000000) {
+
+		return space_cccc_001x(insn, asi);
+
+	} else if ((insn & 0x0f000010) == 0x06000010) {
+
+		return space_cccc_0110__1(insn, asi);
+
+	} else if ((insn & 0x0f000010) == 0x07000010) {
+
+		return space_cccc_0111__1(insn, asi);
+
+	} else if ((insn & 0x0c000000) == 0x04000000) {
+
+		return space_cccc_01xx(insn, asi);
+
+	} else if ((insn & 0x0e000000) == 0x08000000) {
+
+		return space_cccc_100x(insn, asi);
+
+	} else if ((insn & 0x0e000000) == 0x0a000000) {
+
+		return space_cccc_101x(insn, asi);
+
+	} else if ((insn & 0x0fe00000) == 0x0c400000) {
+
+		return space_cccc_1100_010x(insn, asi);
+
+	} else if ((insn & 0x0e000000) == 0x0c400000) {
+
+		return space_cccc_110x(insn, asi);
+
+	}
+
+	return space_cccc_111x(insn, asi);
+}
+
+void __init arm_kprobe_decode_init(void)
+{
+	find_str_pc_offset();
+}
+
+
+/*
+ * All ARM instructions listed below.
+ *
+ * Instructions and their general purpose registers are given.
+ * If a particular register may not use R15, it is prefixed with a "!".
+ * If marked with a "*" means the value returned by reading R15
+ * is implementation defined.
+ *
+ * ADC/ADD/AND/BIC/CMN/CMP/EOR/MOV/MVN/ORR/RSB/RSC/SBC/SUB/TEQ
+ *     TST: Rd, Rn, Rm, !Rs
+ * BX: Rm
+ * BLX(2): !Rm
+ * BX: Rm (R15 legal, but discouraged)
+ * BXJ: !Rm,
+ * CLZ: !Rd, !Rm
+ * CPY: Rd, Rm
+ * LDC/2,STC/2 immediate offset & unindex: Rn
+ * LDC/2,STC/2 immediate pre/post-indexed: !Rn
+ * LDM(1/3): !Rn, register_list
+ * LDM(2): !Rn, !register_list
+ * LDR,STR,PLD immediate offset: Rd, Rn
+ * LDR,STR,PLD register offset: Rd, Rn, !Rm
+ * LDR,STR,PLD scaled register offset: Rd, !Rn, !Rm
+ * LDR,STR immediate pre/post-indexed: Rd, !Rn
+ * LDR,STR register pre/post-indexed: Rd, !Rn, !Rm
+ * LDR,STR scaled register pre/post-indexed: Rd, !Rn, !Rm
+ * LDRB,STRB immediate offset: !Rd, Rn
+ * LDRB,STRB register offset: !Rd, Rn, !Rm
+ * LDRB,STRB scaled register offset: !Rd, !Rn, !Rm
+ * LDRB,STRB immediate pre/post-indexed: !Rd, !Rn
+ * LDRB,STRB register pre/post-indexed: !Rd, !Rn, !Rm
+ * LDRB,STRB scaled register pre/post-indexed: !Rd, !Rn, !Rm
+ * LDRT,LDRBT,STRBT immediate pre/post-indexed: !Rd, !Rn
+ * LDRT,LDRBT,STRBT register pre/post-indexed: !Rd, !Rn, !Rm
+ * LDRT,LDRBT,STRBT scaled register pre/post-indexed: !Rd, !Rn, !Rm
+ * LDRH/SH/SB/D,STRH/SH/SB/D immediate offset: !Rd, Rn
+ * LDRH/SH/SB/D,STRH/SH/SB/D register offset: !Rd, Rn, !Rm
+ * LDRH/SH/SB/D,STRH/SH/SB/D immediate pre/post-indexed: !Rd, !Rn
+ * LDRH/SH/SB/D,STRH/SH/SB/D register pre/post-indexed: !Rd, !Rn, !Rm
+ * LDREX: !Rd, !Rn
+ * MCR/2: !Rd
+ * MCRR/2,MRRC/2: !Rd, !Rn
+ * MLA: !Rd, !Rn, !Rm, !Rs
+ * MOV: Rd
+ * MRC/2: !Rd (if Rd==15, only changes cond codes, not the register)
+ * MRS,MSR: !Rd
+ * MUL: !Rd, !Rm, !Rs
+ * PKH{BT,TB}: !Rd, !Rn, !Rm
+ * QDADD,[U]QADD/16/8/SUBX: !Rd, !Rm, !Rn
+ * QDSUB,[U]QSUB/16/8/ADDX: !Rd, !Rm, !Rn
+ * REV/16/SH: !Rd, !Rm
+ * RFE: !Rn
+ * {S,U}[H]ADD{16,8,SUBX},{S,U}[H]SUB{16,8,ADDX}: !Rd, !Rn, !Rm
+ * SEL: !Rd, !Rn, !Rm
+ * SMLA<x><y>,SMLA{D,W<y>},SMLSD,SMML{A,S}: !Rd, !Rn, !Rm, !Rs
+ * SMLAL<x><y>,SMLA{D,LD},SMLSLD,SMMULL,SMULW<y>: !RdHi, !RdLo, !Rm, !Rs
+ * SMMUL,SMUAD,SMUL<x><y>,SMUSD: !Rd, !Rm, !Rs
+ * SSAT/16: !Rd, !Rm
+ * STM(1/2): !Rn, register_list* (R15 in reg list not recommended)
+ * STRT immediate pre/post-indexed: Rd*, !Rn
+ * STRT register pre/post-indexed: Rd*, !Rn, !Rm
+ * STRT scaled register pre/post-indexed: Rd*, !Rn, !Rm
+ * STREX: !Rd, !Rn, !Rm
+ * SWP/B: !Rd, !Rn, !Rm
+ * {S,U}XTA{B,B16,H}: !Rd, !Rn, !Rm
+ * {S,U}XT{B,B16,H}: !Rd, !Rm
+ * UM{AA,LA,UL}L: !RdHi, !RdLo, !Rm, !Rs
+ * USA{D8,A8,T,T16}: !Rd, !Rm, !Rs
+ *
+ * May transfer control by writing R15 (possible mode changes or alternate
+ * mode accesses marked by "*"):
+ * ALU op (* with s-bit), B, BL, BKPT, BLX(1/2), BX, BXJ, CPS*, CPY,
+ * LDM(1), LDM(2/3)*, LDR, MOV, RFE*, SWI*
+ *
+ * Instructions that do not take general registers, nor transfer control:
+ * CDP/2, SETEND, SRS*
+ */
diff --git a/include/asm-arm/kprobes.h b/include/asm-arm/kprobes.h
new file mode 100644
index 000000000000..951322328793
--- /dev/null
+++ b/include/asm-arm/kprobes.h
@@ -0,0 +1,43 @@
+/*
+ * include/asm-arm/kprobes.h
+ *
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#ifndef _ARM_KPROBES_H
+#define _ARM_KPROBES_H
+
+#include <linux/types.h>
+#include <linux/ptrace.h>
+
+typedef u32 kprobe_opcode_t;
+
+struct kprobe;
+typedef void (kprobe_insn_handler_t)(struct kprobe *, struct pt_regs *);
+
+/* Architecture specific copy of original instruction. */
+struct arch_specific_insn {
+	kprobe_opcode_t		*insn;
+	kprobe_insn_handler_t	*insn_handler;
+};
+
+enum kprobe_insn {
+	INSN_REJECTED,
+	INSN_GOOD,
+	INSN_GOOD_NO_SLOT
+};
+
+enum kprobe_insn arm_kprobe_decode_insn(kprobe_opcode_t,
+					struct arch_specific_insn *);
+void __init arm_kprobe_decode_init(void);
+
+#endif /* _ARM_KPROBES_H */
-- 
cgit v1.2.3


From 24ba613c9d6cad315f484e658288db152f1dc447 Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Mon, 11 Jun 2007 22:20:10 +0000
Subject: ARM kprobes: core code

This is a full implementation of Kprobes including Jprobes and
Kretprobes support.

This ARM implementation does not follow the usual kprobes double-
exception model. The traditional model is where the initial kprobes
breakpoint calls kprobe_handler(), which returns from exception to
execute the instruction in its original context, then immediately
re-enters after a second breakpoint (or single-stepping exception)
into post_kprobe_handler(), each time the probe is hit..  The ARM
implementation only executes one kprobes exception per hit, so no
post_kprobe_handler() phase. All side-effects from the kprobe'd
instruction are resolved before returning from the initial exception.
As a result, all instructions are _always_ effectively boosted
regardless of the type of instruction, and even regardless of whether
or not there is a post-handler for the probe.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Signed-off-by: Quentin Barnes <qbarnes@gmail.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/kernel/Makefile  |   2 +-
 arch/arm/kernel/kprobes.c | 453 ++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/kprobes.h |  29 +++
 3 files changed, 483 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/kernel/kprobes.c

(limited to 'include')

diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 9e0eebacd3d9..faa761921153 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -19,7 +19,7 @@ obj-$(CONFIG_ISA_DMA)		+= dma-isa.o
 obj-$(CONFIG_PCI)		+= bios32.o isa.o
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
-obj-$(CONFIG_KPROBES)		+= kprobes-decode.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o kprobes-decode.o
 obj-$(CONFIG_OABI_COMPAT)	+= sys_oabi-compat.o
 
 obj-$(CONFIG_CRUNCH)		+= crunch.o crunch-bits.o
diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
new file mode 100644
index 000000000000..a18a8458e99d
--- /dev/null
+++ b/arch/arm/kernel/kprobes.c
@@ -0,0 +1,453 @@
+/*
+ * arch/arm/kernel/kprobes.c
+ *
+ * Kprobes on ARM
+ *
+ * Abhishek Sagar <sagar.abhishek@gmail.com>
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * Nicolas Pitre <nico@marvell.com>
+ * Copyright (C) 2007 Marvell Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/module.h>
+#include <linux/stringify.h>
+#include <asm/traps.h>
+#include <asm/cacheflush.h>
+
+/*
+ * This undefined instruction must be unique and
+ * reserved solely for kprobes' use.
+ */
+#define KPROBE_BREAKPOINT_INSTRUCTION	0xe7f001f8
+
+#define MIN_STACK_SIZE(addr) 				\
+	min((unsigned long)MAX_STACK_SIZE,		\
+	    (unsigned long)current_thread_info() + THREAD_START_SP - (addr))
+
+#define flush_insns(addr, cnt) 				\
+	flush_icache_range((unsigned long)(addr),	\
+			   (unsigned long)(addr) +	\
+			   sizeof(kprobe_opcode_t) * (cnt))
+
+/* Used as a marker in ARM_pc to note when we're in a jprobe. */
+#define JPROBE_MAGIC_ADDR		0xffffffff
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	kprobe_opcode_t insn;
+	kprobe_opcode_t tmp_insn[MAX_INSN_SIZE];
+	unsigned long addr = (unsigned long)p->addr;
+	int is;
+
+	if (addr & 0x3)
+		return -EINVAL;
+
+	insn = *p->addr;
+	p->opcode = insn;
+	p->ainsn.insn = tmp_insn;
+
+	switch (arm_kprobe_decode_insn(insn, &p->ainsn)) {
+	case INSN_REJECTED:	/* not supported */
+		return -EINVAL;
+
+	case INSN_GOOD:		/* instruction uses slot */
+		p->ainsn.insn = get_insn_slot();
+		if (!p->ainsn.insn)
+			return -ENOMEM;
+		for (is = 0; is < MAX_INSN_SIZE; ++is)
+			p->ainsn.insn[is] = tmp_insn[is];
+		flush_insns(&p->ainsn.insn, MAX_INSN_SIZE);
+		break;
+
+	case INSN_GOOD_NO_SLOT:	/* instruction doesn't need insn slot */
+		p->ainsn.insn = NULL;
+		break;
+	}
+
+	return 0;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	*p->addr = KPROBE_BREAKPOINT_INSTRUCTION;
+	flush_insns(p->addr, 1);
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	*p->addr = p->opcode;
+	flush_insns(p->addr, 1);
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	if (p->ainsn.insn) {
+		mutex_lock(&kprobe_mutex);
+		free_insn_slot(p->ainsn.insn, 0);
+		mutex_unlock(&kprobe_mutex);
+		p->ainsn.insn = NULL;
+	}
+}
+
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	kcb->prev_kprobe.kp = kprobe_running();
+	kcb->prev_kprobe.status = kcb->kprobe_status;
+}
+
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
+	kcb->kprobe_status = kcb->prev_kprobe.status;
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p)
+{
+	__get_cpu_var(current_kprobe) = p;
+}
+
+static void __kprobes singlestep(struct kprobe *p, struct pt_regs *regs,
+				 struct kprobe_ctlblk *kcb)
+{
+	regs->ARM_pc += 4;
+	p->ainsn.insn_handler(p, regs);
+}
+
+/*
+ * Called with IRQs disabled. IRQs must remain disabled from that point
+ * all the way until processing this kprobe is complete.  The current
+ * kprobes implementation cannot process more than one nested level of
+ * kprobe, and that level is reserved for user kprobe handlers, so we can't
+ * risk encountering a new kprobe in an interrupt handler.
+ */
+void __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p, *cur;
+	struct kprobe_ctlblk *kcb;
+	kprobe_opcode_t	*addr = (kprobe_opcode_t *)regs->ARM_pc;
+
+	kcb = get_kprobe_ctlblk();
+	cur = kprobe_running();
+	p = get_kprobe(addr);
+
+	if (p) {
+		if (cur) {
+			/* Kprobe is pending, so we're recursing. */
+			switch (kcb->kprobe_status) {
+			case KPROBE_HIT_ACTIVE:
+			case KPROBE_HIT_SSDONE:
+				/* A pre- or post-handler probe got us here. */
+				kprobes_inc_nmissed_count(p);
+				save_previous_kprobe(kcb);
+				set_current_kprobe(p);
+				kcb->kprobe_status = KPROBE_REENTER;
+				singlestep(p, regs, kcb);
+				restore_previous_kprobe(kcb);
+				break;
+			default:
+				/* impossible cases */
+				BUG();
+			}
+		} else {
+			set_current_kprobe(p);
+			kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+
+			/*
+			 * If we have no pre-handler or it returned 0, we
+			 * continue with normal processing.  If we have a
+			 * pre-handler and it returned non-zero, it prepped
+			 * for calling the break_handler below on re-entry,
+			 * so get out doing nothing more here.
+			 */
+			if (!p->pre_handler || !p->pre_handler(p, regs)) {
+				kcb->kprobe_status = KPROBE_HIT_SS;
+				singlestep(p, regs, kcb);
+				if (p->post_handler) {
+					kcb->kprobe_status = KPROBE_HIT_SSDONE;
+					p->post_handler(p, regs, 0);
+				}
+				reset_current_kprobe();
+			}
+		}
+	} else if (cur) {
+		/* We probably hit a jprobe.  Call its break handler. */
+		if (cur->break_handler && cur->break_handler(cur, regs)) {
+			kcb->kprobe_status = KPROBE_HIT_SS;
+			singlestep(cur, regs, kcb);
+			if (cur->post_handler) {
+				kcb->kprobe_status = KPROBE_HIT_SSDONE;
+				cur->post_handler(cur, regs, 0);
+			}
+		}
+		reset_current_kprobe();
+	} else {
+		/*
+		 * The probe was removed and a race is in progress.
+		 * There is nothing we can do about it.  Let's restart
+		 * the instruction.  By the time we can restart, the
+		 * real instruction will be there.
+		 */
+	}
+}
+
+static int kprobe_trap_handler(struct pt_regs *regs, unsigned int instr)
+{
+	kprobe_handler(regs);
+	return 0;
+}
+
+int __kprobes kprobe_fault_handler(struct pt_regs *regs, unsigned int fsr)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	switch (kcb->kprobe_status) {
+	case KPROBE_HIT_SS:
+	case KPROBE_REENTER:
+		/*
+		 * We are here because the instruction being single
+		 * stepped caused a page fault. We reset the current
+		 * kprobe and the PC to point back to the probe address
+		 * and allow the page fault handler to continue as a
+		 * normal page fault.
+		 */
+		regs->ARM_pc = (long)cur->addr;
+		if (kcb->kprobe_status == KPROBE_REENTER) {
+			restore_previous_kprobe(kcb);
+		} else {
+			reset_current_kprobe();
+		}
+		break;
+
+	case KPROBE_HIT_ACTIVE:
+	case KPROBE_HIT_SSDONE:
+		/*
+		 * We increment the nmissed count for accounting,
+		 * we can also use npre/npostfault count for accounting
+		 * these specific fault cases.
+		 */
+		kprobes_inc_nmissed_count(cur);
+
+		/*
+		 * We come here because instructions in the pre/post
+		 * handler caused the page_fault, this could happen
+		 * if handler tries to access user space by
+		 * copy_from_user(), get_user() etc. Let the
+		 * user-specified handler try to fix it.
+		 */
+		if (cur->fault_handler && cur->fault_handler(cur, regs, fsr))
+			return 1;
+		break;
+
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	/*
+	 * notify_die() is currently never called on ARM,
+	 * so this callback is currently empty.
+	 */
+	return NOTIFY_DONE;
+}
+
+/*
+ * When a retprobed function returns, trampoline_handler() is called,
+ * calling the kretprobe's handler. We construct a struct pt_regs to
+ * give a view of registers r0-r11 to the user return-handler.  This is
+ * not a complete pt_regs structure, but that should be plenty sufficient
+ * for kretprobe handlers which should normally be interested in r0 only
+ * anyway.
+ */
+static void __attribute__((naked)) __kprobes kretprobe_trampoline(void)
+{
+	__asm__ __volatile__ (
+		"stmdb	sp!, {r0 - r11}		\n\t"
+		"mov	r0, sp			\n\t"
+		"bl	trampoline_handler	\n\t"
+		"mov	lr, r0			\n\t"
+		"ldmia	sp!, {r0 - r11}		\n\t"
+		"mov	pc, lr			\n\t"
+		: : : "memory");
+}
+
+/* Called from kretprobe_trampoline */
+static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head, empty_rp;
+	struct hlist_node *node, *tmp;
+	unsigned long flags, orig_ret_address = 0;
+	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
+
+	INIT_HLIST_HEAD(&empty_rp);
+	spin_lock_irqsave(&kretprobe_lock, flags);
+	head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because multiple functions in the call path have
+	 * a return probe installed on them, and/or more than one return
+	 * probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		if (ri->rp && ri->rp->handler) {
+			__get_cpu_var(current_kprobe) = &ri->rp->kp;
+			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
+			ri->rp->handler(ri, regs);
+			__get_cpu_var(current_kprobe) = NULL;
+		}
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri, &empty_rp);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	kretprobe_assert(ri, orig_ret_address, trampoline_address);
+	spin_unlock_irqrestore(&kretprobe_lock, flags);
+
+	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
+		hlist_del(&ri->hlist);
+		kfree(ri);
+	}
+
+	return (void *)orig_ret_address;
+}
+
+/* Called with kretprobe_lock held. */
+void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
+				      struct pt_regs *regs)
+{
+	ri->ret_addr = (kprobe_opcode_t *)regs->ARM_lr;
+
+	/* Replace the return addr with trampoline addr. */
+	regs->ARM_lr = (unsigned long)&kretprobe_trampoline;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	long sp_addr = regs->ARM_sp;
+
+	kcb->jprobe_saved_regs = *regs;
+	memcpy(kcb->jprobes_stack, (void *)sp_addr, MIN_STACK_SIZE(sp_addr));
+	regs->ARM_pc = (long)jp->entry;
+	regs->ARM_cpsr |= PSR_I_BIT;
+	preempt_disable();
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	__asm__ __volatile__ (
+		/*
+		 * Setup an empty pt_regs. Fill SP and PC fields as
+		 * they're needed by longjmp_break_handler.
+		 */
+		"sub    sp, %0, %1		\n\t"
+		"ldr    r0, ="__stringify(JPROBE_MAGIC_ADDR)"\n\t"
+		"str    %0, [sp, %2]		\n\t"
+		"str    r0, [sp, %3]		\n\t"
+		"mov    r0, sp			\n\t"
+		"bl     kprobe_handler		\n\t"
+
+		/*
+		 * Return to the context saved by setjmp_pre_handler
+		 * and restored by longjmp_break_handler.
+		 */
+		"ldr	r0, [sp, %4]		\n\t"
+		"msr	cpsr_cxsf, r0		\n\t"
+		"ldmia	sp, {r0 - pc}		\n\t"
+		:
+		: "r" (kcb->jprobe_saved_regs.ARM_sp),
+		  "I" (sizeof(struct pt_regs)),
+		  "J" (offsetof(struct pt_regs, ARM_sp)),
+		  "J" (offsetof(struct pt_regs, ARM_pc)),
+		  "J" (offsetof(struct pt_regs, ARM_cpsr))
+		: "memory", "cc");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	long stack_addr = kcb->jprobe_saved_regs.ARM_sp;
+	long orig_sp = regs->ARM_sp;
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+
+	if (regs->ARM_pc == JPROBE_MAGIC_ADDR) {
+		if (orig_sp != stack_addr) {
+			struct pt_regs *saved_regs =
+				(struct pt_regs *)kcb->jprobe_saved_regs.ARM_sp;
+			printk("current sp %lx does not match saved sp %lx\n",
+			       orig_sp, stack_addr);
+			printk("Saved registers for jprobe %p\n", jp);
+			show_regs(saved_regs);
+			printk("Current registers\n");
+			show_regs(regs);
+			BUG();
+		}
+		*regs = kcb->jprobe_saved_regs;
+		memcpy((void *)stack_addr, kcb->jprobes_stack,
+		       MIN_STACK_SIZE(stack_addr));
+		preempt_enable_no_resched();
+		return 1;
+	}
+	return 0;
+}
+
+static struct undef_hook kprobes_break_hook = {
+	.instr_mask	= 0xffffffff,
+	.instr_val	= KPROBE_BREAKPOINT_INSTRUCTION,
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= SVC_MODE,
+	.fn		= kprobe_trap_handler,
+};
+
+int __init arch_init_kprobes()
+{
+	arm_kprobe_decode_init();
+	register_undef_hook(&kprobes_break_hook);
+	return 0;
+}
diff --git a/include/asm-arm/kprobes.h b/include/asm-arm/kprobes.h
index 951322328793..273f37413ee6 100644
--- a/include/asm-arm/kprobes.h
+++ b/include/asm-arm/kprobes.h
@@ -18,6 +18,16 @@
 
 #include <linux/types.h>
 #include <linux/ptrace.h>
+#include <linux/percpu.h>
+
+#define ARCH_SUPPORTS_KRETPROBES
+#define __ARCH_WANT_KPROBES_INSN_SLOT
+#define MAX_INSN_SIZE			2
+#define MAX_STACK_SIZE			64	/* 32 would probably be OK */
+
+#define regs_return_value(regs)		((regs)->ARM_r0)
+#define flush_insn_slot(p)		do { } while (0)
+#define kretprobe_blacklist_size	0
 
 typedef u32 kprobe_opcode_t;
 
@@ -30,6 +40,25 @@ struct arch_specific_insn {
 	kprobe_insn_handler_t	*insn_handler;
 };
 
+struct prev_kprobe {
+	struct kprobe *kp;
+	unsigned int status;
+};
+
+/* per-cpu kprobe control block */
+struct kprobe_ctlblk {
+	unsigned int kprobe_status;
+	struct prev_kprobe prev_kprobe;
+	struct pt_regs jprobe_saved_regs;
+	char jprobes_stack[MAX_STACK_SIZE];
+};
+
+void arch_remove_kprobe(struct kprobe *);
+
+int kprobe_fault_handler(struct pt_regs *regs, unsigned int fsr);
+int kprobe_exceptions_notify(struct notifier_block *self,
+			     unsigned long val, void *data);
+
 enum kprobe_insn {
 	INSN_REJECTED,
 	INSN_GOOD,
-- 
cgit v1.2.3


From 785d3cd286f0bf67d1bf692559b9ae5de12678f5 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Mon, 3 Dec 2007 15:27:56 -0500
Subject: ARM kprobes: prevent some functions involved with kprobes from being
 probed

Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/kernel/entry-armv.S  | 8 +++++++-
 arch/arm/kernel/kprobes.c     | 2 +-
 arch/arm/kernel/traps.c       | 9 ---------
 arch/arm/kernel/vmlinux.lds.S | 1 +
 arch/arm/mm/fault.c           | 4 ++--
 include/asm-arm/traps.h       | 9 +++++++++
 6 files changed, 20 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index 12ecc817be3f..3727537f5825 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -58,6 +58,12 @@
 
 	.endm
 
+#ifdef CONFIG_KPROBES
+	.section	.kprobes.text,"ax",%progbits
+#else
+	.text
+#endif
+
 /*
  * Invalid mode handlers
  */
@@ -562,7 +568,7 @@ do_fpe:
 	.data
 ENTRY(fp_enter)
 	.word	no_fp
-	.text
+	.previous
 
 no_fp:	mov	pc, lr
 
diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
index a18a8458e99d..450ee2cbfe17 100644
--- a/arch/arm/kernel/kprobes.c
+++ b/arch/arm/kernel/kprobes.c
@@ -55,7 +55,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	unsigned long addr = (unsigned long)p->addr;
 	int is;
 
-	if (addr & 0x3)
+	if (addr & 0x3 || in_exception_text(addr))
 		return -EINVAL;
 
 	insn = *p->addr;
diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index c34db4e868fa..65bb762b2d80 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -46,15 +46,6 @@ __setup("user_debug=", user_debug_setup);
 
 static void dump_mem(const char *str, unsigned long bottom, unsigned long top);
 
-static inline int in_exception_text(unsigned long ptr)
-{
-	extern char __exception_text_start[];
-	extern char __exception_text_end[];
-
-	return ptr >= (unsigned long)&__exception_text_start &&
-	       ptr < (unsigned long)&__exception_text_end;
-}
-
 void dump_backtrace_entry(unsigned long where, unsigned long from, unsigned long frame)
 {
 #ifdef CONFIG_KALLSYMS
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 5ff5406666b4..30f732c7fdb5 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -94,6 +94,7 @@ SECTIONS
 			TEXT_TEXT
 			SCHED_TEXT
 			LOCK_TEXT
+			KPROBES_TEXT
 #ifdef CONFIG_MMU
 			*(.fixup)
 #endif
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 1e7b1ed304d8..28ad7ab1c0cd 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -239,7 +239,7 @@ out:
 	return fault;
 }
 
-static int
+static int __kprobes
 do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 {
 	struct task_struct *tsk;
@@ -338,7 +338,7 @@ no_context:
  * interrupt or a critical region, and should only copy the information
  * from the master page table, nothing more.
  */
-static int
+static int __kprobes
 do_translation_fault(unsigned long addr, unsigned int fsr,
 		     struct pt_regs *regs)
 {
diff --git a/include/asm-arm/traps.h b/include/asm-arm/traps.h
index d4f34dc83eb0..f1541afcf85c 100644
--- a/include/asm-arm/traps.h
+++ b/include/asm-arm/traps.h
@@ -15,4 +15,13 @@ struct undef_hook {
 void register_undef_hook(struct undef_hook *hook);
 void unregister_undef_hook(struct undef_hook *hook);
 
+static inline int in_exception_text(unsigned long ptr)
+{
+	extern char __exception_text_start[];
+	extern char __exception_text_end[];
+
+	return ptr >= (unsigned long)&__exception_text_start &&
+	       ptr < (unsigned long)&__exception_text_end;
+}
+
 #endif
-- 
cgit v1.2.3


From 796969104cab0d454dbc792ad0d12a4f365a8564 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Mon, 3 Dec 2007 17:22:36 -0500
Subject: ARM kprobes: special hook for the kprobes breakpoint handler

The kprobes code is already able to cope with reentrant probes, so its
handler must be called outside of the region protected by undef_lock.

If ever this lock is released when handlers are called then this commit
could be reverted.

Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/kernel/kprobes.c |  8 +-------
 arch/arm/kernel/traps.c   | 12 ++++++++++++
 include/asm-arm/kprobes.h |  7 +++++++
 3 files changed, 20 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
index 450ee2cbfe17..a22a98c43ca5 100644
--- a/arch/arm/kernel/kprobes.c
+++ b/arch/arm/kernel/kprobes.c
@@ -26,12 +26,6 @@
 #include <asm/traps.h>
 #include <asm/cacheflush.h>
 
-/*
- * This undefined instruction must be unique and
- * reserved solely for kprobes' use.
- */
-#define KPROBE_BREAKPOINT_INSTRUCTION	0xe7f001f8
-
 #define MIN_STACK_SIZE(addr) 				\
 	min((unsigned long)MAX_STACK_SIZE,		\
 	    (unsigned long)current_thread_info() + THREAD_START_SP - (addr))
@@ -206,7 +200,7 @@ void __kprobes kprobe_handler(struct pt_regs *regs)
 	}
 }
 
-static int kprobe_trap_handler(struct pt_regs *regs, unsigned int instr)
+int kprobe_trap_handler(struct pt_regs *regs, unsigned int instr)
 {
 	kprobe_handler(regs);
 	return 0;
diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index 65bb762b2d80..5595fdd75e82 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -19,6 +19,7 @@
 #include <linux/kallsyms.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/kprobes.h>
 
 #include <asm/atomic.h>
 #include <asm/cacheflush.h>
@@ -313,6 +314,17 @@ asmlinkage void __exception do_undefinstr(struct pt_regs *regs)
 		get_user(instr, (u32 __user *)pc);
 	}
 
+#ifdef CONFIG_KPROBES
+	/*
+	 * It is possible to have recursive kprobes, so we can't call
+	 * the kprobe trap handler with the undef_lock held.
+	 */
+	if (instr == KPROBE_BREAKPOINT_INSTRUCTION && !user_mode(regs)) {
+		kprobe_trap_handler(regs, instr);
+		return;
+	}
+#endif
+
 	spin_lock_irqsave(&undef_lock, flags);
 	list_for_each_entry(hook, &undef_hook, node) {
 		if ((instr & hook->instr_mask) == hook->instr_val &&
diff --git a/include/asm-arm/kprobes.h b/include/asm-arm/kprobes.h
index 273f37413ee6..4e7bd32288ae 100644
--- a/include/asm-arm/kprobes.h
+++ b/include/asm-arm/kprobes.h
@@ -25,6 +25,12 @@
 #define MAX_INSN_SIZE			2
 #define MAX_STACK_SIZE			64	/* 32 would probably be OK */
 
+/*
+ * This undefined instruction must be unique and
+ * reserved solely for kprobes' use.
+ */
+#define KPROBE_BREAKPOINT_INSTRUCTION	0xe7f001f8
+
 #define regs_return_value(regs)		((regs)->ARM_r0)
 #define flush_insn_slot(p)		do { } while (0)
 #define kretprobe_blacklist_size	0
@@ -55,6 +61,7 @@ struct kprobe_ctlblk {
 
 void arch_remove_kprobe(struct kprobe *);
 
+int kprobe_trap_handler(struct pt_regs *regs, unsigned int instr);
 int kprobe_fault_handler(struct pt_regs *regs, unsigned int fsr);
 int kprobe_exceptions_notify(struct notifier_block *self,
 			     unsigned long val, void *data);
-- 
cgit v1.2.3


From 8704de8f296fcf6a4b2ff6bfd9a63974ad909b3e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:00 +0100
Subject: cy82c693: add ->set_dma_mode method

* Fix SWDMA/MWDMA masks in cy82c693_chipset.

* Add IDE_HFLAG_CY82C693 host flag and use it in ide_tune_dma() to
  check whether the DMA should be enabled even if ide_max_dma_mode()
  fails.

* Convert cy82c693_dma_enable() to become cy82c693_set_dma_mode()
  and remove no longer needed cy82c693_ide_dma_on().  Then set
  IDE_HFLAG_CY82C693 instead of IDE_HFLAG_TRUST_BIOS_FOR_DMA in
  cy82c693_chipset.

* Bump driver version.

As a result of this patch cy82c693 driver will configure and use DMA on
all SWDMA0-2 and MWDMA0-2 capable ATA devices instead of relying on BIOS.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c      | 15 ++++++++---
 drivers/ide/pci/cy82c693.c | 64 ++++++++--------------------------------------
 include/linux/ide.h        |  2 ++
 3 files changed, 24 insertions(+), 57 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 18c78ad2b31e..9d6dabbbf809 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -755,6 +755,7 @@ EXPORT_SYMBOL_GPL(ide_find_dma_mode);
 
 static int ide_tune_dma(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	u8 speed;
 
 	if (noautodma || drive->nodma || (drive->id->capability & 1) == 0)
@@ -767,15 +768,21 @@ static int ide_tune_dma(ide_drive_t *drive)
 	if (ide_id_dma_bug(drive))
 		return 0;
 
-	if (drive->hwif->host_flags & IDE_HFLAG_TRUST_BIOS_FOR_DMA)
+	if (hwif->host_flags & IDE_HFLAG_TRUST_BIOS_FOR_DMA)
 		return config_drive_for_dma(drive);
 
 	speed = ide_max_dma_mode(drive);
 
-	if (!speed)
-		return 0;
+	if (!speed) {
+		 /* is this really correct/needed? */
+		if ((hwif->host_flags & IDE_HFLAG_CY82C693) &&
+		    ide_dma_good_drive(drive))
+			return 1;
+		else
+			return 0;
+	}
 
-	if (drive->hwif->host_flags & IDE_HFLAG_NO_SET_MODE)
+	if (hwif->host_flags & IDE_HFLAG_NO_SET_MODE)
 		return 0;
 
 	if (ide_set_dma_mode(drive, speed))
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index e7466f2bee7b..3ec4c659a37d 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -1,5 +1,5 @@
 /*
- * linux/drivers/ide/pci/cy82c693.c		Version 0.43	Nov 7, 2007
+ * linux/drivers/ide/pci/cy82c693.c		Version 0.44	Nov 8, 2007
  *
  *  Copyright (C) 1998-2000 Andreas S. Krebs (akrebs@altavista.net), Maintainer
  *  Copyright (C) 1998-2002 Andre Hedrick <andre@linux-ide.org>, Integrator
@@ -176,14 +176,12 @@ static void compute_clocks (u8 pio, pio_clocks_t *p_pclk)
  * set DMA mode a specific channel for CY82C693
  */
 
-static void cy82c693_dma_enable (ide_drive_t *drive, int mode, int single)
+static void cy82c693_set_dma_mode(ide_drive_t *drive, const u8 mode)
 {
-	u8 index = 0, data = 0;
+	ide_hwif_t *hwif = drive->hwif;
+	u8 single = (mode & 0x10) >> 4, index = 0, data = 0;
 
-	if (mode>2)	/* make sure we set a valid mode */
-		mode = 2;
-
-	index = (HWIF(drive)->channel==0) ? CY82_INDEX_CHANNEL0 : CY82_INDEX_CHANNEL1;
+	index = hwif->channel ? CY82_INDEX_CHANNEL1 : CY82_INDEX_CHANNEL0;
 
 #if CY82C693_DEBUG_LOGS
 	/* for debug let's show the previous values */
@@ -196,7 +194,7 @@ static void cy82c693_dma_enable (ide_drive_t *drive, int mode, int single)
 		(data&0x3), ((data>>2)&1));
 #endif /* CY82C693_DEBUG_LOGS */
 
-	data = (u8)mode|(u8)(single<<2);
+	data = (mode & 3) | (single << 2);
 
 	outb(index, CY82_INDEX_PORT);
 	outb(data, CY82_DATA_PORT);
@@ -204,7 +202,7 @@ static void cy82c693_dma_enable (ide_drive_t *drive, int mode, int single)
 #if CY82C693_DEBUG_INFO
 	printk(KERN_INFO "%s (ch=%d, dev=%d): set DMA mode to %d (single=%d)\n",
 		drive->name, HWIF(drive)->channel, drive->select.b.unit,
-		mode, single);
+		mode & 3, single);
 #endif /* CY82C693_DEBUG_INFO */
 
 	/* 
@@ -227,42 +225,6 @@ static void cy82c693_dma_enable (ide_drive_t *drive, int mode, int single)
 #endif /* CY82C693_DEBUG_INFO */
 }
 
-/* 
- * used to set DMA mode for CY82C693 (single and multi modes)
- */
-static int cy82c693_ide_dma_on (ide_drive_t *drive)
-{
-	struct hd_driveid *id = drive->id;
-
-#if CY82C693_DEBUG_INFO
-	printk (KERN_INFO "dma_on: %s\n", drive->name);
-#endif /* CY82C693_DEBUG_INFO */
-
-	if (id != NULL) {		
-		/* Enable DMA on any drive that has DMA
-		 * (multi or single) enabled
-		 */
-		if (id->field_valid & 2) {	/* regular DMA */
-			int mmode, smode;
-
-			mmode = id->dma_mword & (id->dma_mword >> 8);
-			smode = id->dma_1word & (id->dma_1word >> 8);
-
-			mmode &= ATA_MWDMA2;
-			smode &= ATA_SWDMA2;
-
-			if (mmode != 0) {
-				/* enable multi */
-				cy82c693_dma_enable(drive, (mmode >> 1), 0);
-			} else if (smode != 0) {
-				/* enable single */
-				cy82c693_dma_enable(drive, (smode >> 1), 1);
-			}
-		}
-	}
-        return __ide_dma_on(drive);
-}
-
 static void cy82c693_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -429,11 +391,7 @@ static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev, const c
 static void __devinit init_hwif_cy82c693(ide_hwif_t *hwif)
 {
 	hwif->set_pio_mode = &cy82c693_set_pio_mode;
-
-	if (hwif->dma_base == 0)
-		return;
-
-	hwif->ide_dma_on = &cy82c693_ide_dma_on;
+	hwif->set_dma_mode = &cy82c693_set_dma_mode;
 }
 
 static void __devinit init_iops_cy82c693(ide_hwif_t *hwif)
@@ -454,11 +412,11 @@ static const struct ide_port_info cy82c693_chipset __devinitdata = {
 	.init_iops	= init_iops_cy82c693,
 	.init_hwif	= init_hwif_cy82c693,
 	.chipset	= ide_cy82c693,
-	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_TRUST_BIOS_FOR_DMA |
+	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_CY82C693 |
 			  IDE_HFLAG_BOOTABLE,
 	.pio_mask	= ATA_PIO4,
-	.swdma_mask	= ATA_SWDMA2_ONLY,
-	.mwdma_mask	= ATA_MWDMA2_ONLY,
+	.swdma_mask	= ATA_SWDMA2,
+	.mwdma_mask	= ATA_MWDMA2,
 };
 
 static int __devinit cy82c693_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1e4409937ec3..bf106d569cfc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1095,6 +1095,8 @@ enum {
 	/* unmask IRQs */
 	IDE_HFLAG_UNMASK_IRQS		= (1 << 25),
 	IDE_HFLAG_ABUSE_SET_DMA_MODE	= (1 << 26),
+	/* host is CY82C693 */
+	IDE_HFLAG_CY82C693		= (1 << 27),
 };
 
 #ifdef CONFIG_BLK_DEV_OFFBOARD
-- 
cgit v1.2.3


From 4a546e046d562bcd389149591fa5a534c8f832ca Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:01 +0100
Subject: ide: remove ->ide_dma_on and ->dma_off_quietly methods from
 ide_hwif_t

* Make ide_dma_off_quietly() and __ide_dma_on() always available.

* Drop "__" prefix from __ide_dma_on().

* Check for presence of ->dma_host_on instead of ->ide_dma_on.

* Convert all users of ->ide_dma_on and ->dma_off_quietly methods
  to use ide_dma_on() and ide_dma_off_quietly() instead.

* Remove no longer needed ->ide_dma_on and ->dma_off_quietly methods
  from ide_hwif_t.

* Make ide_dma_on() void.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c      | 16 ----------------
 drivers/ide/ide-dma.c         | 27 ++++++++++++---------------
 drivers/ide/ide-io.c          |  8 ++++----
 drivers/ide/ide-iops.c        | 10 +++++-----
 drivers/ide/ide-probe.c       |  2 +-
 drivers/ide/ide.c             |  4 +---
 drivers/ide/mips/au1xxx-ide.c | 16 ----------------
 drivers/ide/pci/sc1200.c      |  2 +-
 drivers/ide/pci/sgiioc4.c     | 19 -------------------
 drivers/ide/ppc/pmac.c        |  2 --
 include/linux/ide.h           |  8 ++++----
 11 files changed, 28 insertions(+), 86 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index d70442a37e34..3a8402bb5dc3 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -291,24 +291,10 @@ static void icside_dma_host_off(ide_drive_t *drive)
 {
 }
 
-static void icside_dma_off_quietly(ide_drive_t *drive)
-{
-	drive->using_dma = 0;
-	ide_toggle_bounce(drive, 0);
-}
-
 static void icside_dma_host_on(ide_drive_t *drive)
 {
 }
 
-static int icside_dma_on(ide_drive_t *drive)
-{
-	drive->using_dma = 1;
-	ide_toggle_bounce(drive, 1);
-
-	return 0;
-}
-
 static int icside_dma_end(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -425,9 +411,7 @@ static void icside_dma_init(ide_hwif_t *hwif)
 	hwif->set_dma_mode	= icside_set_dma_mode;
 
 	hwif->dma_host_off	= icside_dma_host_off;
-	hwif->dma_off_quietly	= icside_dma_off_quietly;
 	hwif->dma_host_on	= icside_dma_host_on;
-	hwif->ide_dma_on	= icside_dma_on;
 	hwif->dma_setup		= icside_dma_setup;
 	hwif->dma_exec_cmd	= icside_dma_exec_cmd;
 	hwif->dma_start		= icside_dma_start;
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 9d6dabbbf809..edd0018c4988 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -425,6 +425,7 @@ void ide_dma_host_off(ide_drive_t *drive)
 }
 
 EXPORT_SYMBOL(ide_dma_host_off);
+#endif /* CONFIG_BLK_DEV_IDEDMA_PCI */
 
 /**
  *	ide_dma_off_quietly	-	Generic DMA kill
@@ -442,7 +443,6 @@ void ide_dma_off_quietly(ide_drive_t *drive)
 }
 
 EXPORT_SYMBOL(ide_dma_off_quietly);
-#endif /* CONFIG_BLK_DEV_IDEDMA_PCI */
 
 /**
  *	ide_dma_off	-	disable DMA on a device
@@ -455,7 +455,7 @@ EXPORT_SYMBOL(ide_dma_off_quietly);
 void ide_dma_off(ide_drive_t *drive)
 {
 	printk(KERN_INFO "%s: DMA disabled\n", drive->name);
-	drive->hwif->dma_off_quietly(drive);
+	ide_dma_off_quietly(drive);
 }
 
 EXPORT_SYMBOL(ide_dma_off);
@@ -481,26 +481,26 @@ void ide_dma_host_on(ide_drive_t *drive)
 }
 
 EXPORT_SYMBOL(ide_dma_host_on);
+#endif
 
 /**
- *	__ide_dma_on		-	Enable DMA on a device
+ *	ide_dma_on		-	Enable DMA on a device
  *	@drive: drive to enable DMA on
  *
  *	Enable IDE DMA for a device on this IDE controller.
  */
- 
-int __ide_dma_on (ide_drive_t *drive)
+
+void ide_dma_on(ide_drive_t *drive)
 {
 	drive->using_dma = 1;
 	ide_toggle_bounce(drive, 1);
 
 	drive->hwif->dma_host_on(drive);
-
-	return 0;
 }
 
-EXPORT_SYMBOL(__ide_dma_on);
+EXPORT_SYMBOL(ide_dma_on);
 
+#ifdef CONFIG_BLK_DEV_IDEDMA_PCI
 /**
  *	ide_dma_setup	-	begin a DMA phase
  *	@drive: target device
@@ -827,7 +827,6 @@ err_out:
 
 int ide_set_dma(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	int rc;
 
 	/*
@@ -836,13 +835,15 @@ int ide_set_dma(ide_drive_t *drive)
 	 * things, if not checked and cleared.
 	 *   PARANOIA!!!
 	 */
-	hwif->dma_off_quietly(drive);
+	ide_dma_off_quietly(drive);
 
 	rc = ide_dma_check(drive);
 	if (rc)
 		return rc;
 
-	return hwif->ide_dma_on(drive);
+	ide_dma_on(drive);
+
+	return 0;
 }
 
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
@@ -979,12 +980,8 @@ void ide_setup_dma(ide_hwif_t *hwif, unsigned long base, unsigned num_ports)
 	if (!(hwif->dma_prdtable))
 		hwif->dma_prdtable	= (hwif->dma_base + 4);
 
-	if (!hwif->dma_off_quietly)
-		hwif->dma_off_quietly = &ide_dma_off_quietly;
 	if (!hwif->dma_host_off)
 		hwif->dma_host_off = &ide_dma_host_off;
-	if (!hwif->ide_dma_on)
-		hwif->ide_dma_on = &__ide_dma_on;
 	if (!hwif->dma_host_on)
 		hwif->dma_host_on = &ide_dma_host_on;
 	if (!hwif->dma_setup)
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 2711b5a6962d..b5a7d2578ab8 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -75,7 +75,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq,
 	 */
 	if (drive->state == DMA_PIO_RETRY && drive->retry_pio <= 3) {
 		drive->state = 0;
-		HWGROUP(drive)->hwif->ide_dma_on(drive);
+		ide_dma_on(drive);
 	}
 
 	if (!end_that_request_chunk(rq, uptodate, nr_bytes)) {
@@ -219,7 +219,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 		 * we could be smarter and check for current xfer_speed
 		 * in struct drive etc...
 		 */
-		if (drive->hwif->ide_dma_on == NULL)
+		if (drive->hwif->dma_host_on == NULL)
 			break;
 		/*
 		 * TODO: respect ->using_dma setting
@@ -787,7 +787,7 @@ static ide_startstop_t do_special (ide_drive_t *drive)
 
 			if (hwif->host_flags & IDE_HFLAG_SET_PIO_MODE_KEEP_DMA) {
 				if (keep_dma)
-					hwif->ide_dma_on(drive);
+					ide_dma_on(drive);
 			}
 		}
 
@@ -1334,7 +1334,7 @@ static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 	 */
 	drive->retry_pio++;
 	drive->state = DMA_PIO_RETRY;
-	hwif->dma_off_quietly(drive);
+	ide_dma_off_quietly(drive);
 
 	/*
 	 * un-busy drive etc (hwgroup->busy is cleared on return) and
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index e3e5e39f4906..76cb5f2bd4e6 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -742,7 +742,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 //		msleep(50);
 
 #ifdef CONFIG_BLK_DEV_IDEDMA
-	if (hwif->ide_dma_on)	/* check if host supports DMA */
+	if (hwif->dma_host_on)	/* check if host supports DMA */
 		hwif->dma_host_off(drive);
 #endif
 
@@ -801,8 +801,8 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 #ifdef CONFIG_BLK_DEV_IDEDMA
 	if (speed >= XFER_SW_DMA_0 || (hwif->host_flags & IDE_HFLAG_VDMA))
 		hwif->dma_host_on(drive);
-	else if (hwif->ide_dma_on)	/* check if host supports DMA */
-		hwif->dma_off_quietly(drive);
+	else if (hwif->dma_host_on)	/* check if host supports DMA */
+		ide_dma_off_quietly(drive);
 #endif
 
 	switch(speed) {
@@ -1012,10 +1012,10 @@ static void check_dma_crc(ide_drive_t *drive)
 {
 #ifdef CONFIG_BLK_DEV_IDEDMA
 	if (drive->crc_count) {
-		drive->hwif->dma_off_quietly(drive);
+		ide_dma_off_quietly(drive);
 		ide_set_xfer_rate(drive, ide_auto_reduce_xfer(drive));
 		if (drive->current_speed >= XFER_SW_DMA_0)
-			(void) HWIF(drive)->ide_dma_on(drive);
+			ide_dma_on(drive);
 	} else
 		ide_dma_off(drive);
 #endif
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 0379d1f697cf..b363a96607d1 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -833,7 +833,7 @@ static void probe_hwif(ide_hwif_t *hwif)
 
 			drive->nice1 = 1;
 
-			if (hwif->ide_dma_on)
+			if (hwif->dma_host_on)
 				ide_set_dma(drive);
 		}
 	}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index c6d4f630e18a..095ff34870da 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -437,8 +437,6 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 	hwif->dma_exec_cmd		= tmp_hwif->dma_exec_cmd;
 	hwif->dma_start			= tmp_hwif->dma_start;
 	hwif->ide_dma_end		= tmp_hwif->ide_dma_end;
-	hwif->ide_dma_on		= tmp_hwif->ide_dma_on;
-	hwif->dma_off_quietly		= tmp_hwif->dma_off_quietly;
 	hwif->ide_dma_test_irq		= tmp_hwif->ide_dma_test_irq;
 	hwif->ide_dma_clear_irq		= tmp_hwif->ide_dma_clear_irq;
 	hwif->dma_host_on		= tmp_hwif->dma_host_on;
@@ -836,7 +834,7 @@ int set_using_dma(ide_drive_t *drive, int arg)
 	if (!drive->id || !(drive->id->capability & 1))
 		goto out;
 
-	if (hwif->ide_dma_on == NULL)
+	if (hwif->dma_host_on == NULL)
 		goto out;
 
 	err = -EBUSY;
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 4fc032838051..4dfdca4ccbde 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -399,24 +399,10 @@ static void auide_dma_host_on(ide_drive_t *drive)
 {
 }
 
-static int auide_dma_on(ide_drive_t *drive)
-{
-	drive->using_dma = 1;
-	ide_toggle_bounce(drive, 1);
-
-	return 0;
-}
-
 static void auide_dma_host_off(ide_drive_t *drive)
 {
 }
 
-static void auide_dma_off_quietly(ide_drive_t *drive)
-{
-	drive->using_dma = 0;
-	ide_toggle_bounce(drive, 0);
-}
-
 static void auide_dma_lost_irq(ide_drive_t *drive)
 {
 	printk(KERN_ERR "%s: IRQ lost\n", drive->name);
@@ -684,7 +670,6 @@ static int au_ide_probe(struct device *dev)
 	hwif->set_dma_mode		= &auide_set_dma_mode;
 
 #ifdef CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA
-	hwif->dma_off_quietly		= &auide_dma_off_quietly;
 	hwif->dma_timeout		= &auide_dma_timeout;
 
 	hwif->mdma_filter		= &auide_mdma_filter;
@@ -697,7 +682,6 @@ static int au_ide_probe(struct device *dev)
 	hwif->dma_host_off		= &auide_dma_host_off;
 	hwif->dma_host_on		= &auide_dma_host_on;
 	hwif->dma_lost_irq		= &auide_dma_lost_irq;
-	hwif->ide_dma_on                = &auide_dma_on;
 #else /* !CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA */
 	hwif->channel                   = 0;
 	hwif->hold                      = 1;
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index fef20bd4aa78..8a94c3e8f7c4 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -220,7 +220,7 @@ static void sc1200_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	}
 	if (mode != -1) {
 		printk("SC1200: %s: changing (U)DMA mode\n", drive->name);
-		hwif->dma_off_quietly(drive);
+		ide_dma_off_quietly(drive);
 		if (ide_set_dma_mode(drive, mode) == 0)
 			hwif->dma_host_on(drive);
 		return;
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index fea56d3b3a3a..8c4e94bd4449 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -277,23 +277,6 @@ sgiioc4_ide_dma_end(ide_drive_t * drive)
 	return dma_stat;
 }
 
-static int
-sgiioc4_ide_dma_on(ide_drive_t * drive)
-{
-	drive->using_dma = 1;
-	ide_toggle_bounce(drive, 1);
-
-	return 0;
-}
-
-static void sgiioc4_dma_off_quietly(ide_drive_t *drive)
-{
-	drive->using_dma = 0;
-	ide_toggle_bounce(drive, 0);
-
-	drive->hwif->dma_host_off(drive);
-}
-
 static void sgiioc4_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
 }
@@ -598,8 +581,6 @@ ide_init_sgiioc4(ide_hwif_t * hwif)
 	hwif->dma_setup = &sgiioc4_ide_dma_setup;
 	hwif->dma_start = &sgiioc4_ide_dma_start;
 	hwif->ide_dma_end = &sgiioc4_ide_dma_end;
-	hwif->ide_dma_on = &sgiioc4_ide_dma_on;
-	hwif->dma_off_quietly = &sgiioc4_dma_off_quietly;
 	hwif->ide_dma_test_irq = &sgiioc4_ide_dma_test_irq;
 	hwif->dma_host_on = &sgiioc4_dma_host_on;
 	hwif->dma_host_off = &sgiioc4_dma_host_off;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 3dce80092fff..ca99b69cfac7 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1748,8 +1748,6 @@ pmac_ide_setup_dma(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 		return;
 	}
 
-	hwif->dma_off_quietly = &ide_dma_off_quietly;
-	hwif->ide_dma_on = &__ide_dma_on;
 	hwif->dma_setup = &pmac_ide_dma_setup;
 	hwif->dma_exec_cmd = &pmac_ide_dma_exec_cmd;
 	hwif->dma_start = &pmac_ide_dma_start;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index bf106d569cfc..140864d63aed 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -546,8 +546,6 @@ typedef struct hwif_s {
 	void (*dma_exec_cmd)(ide_drive_t *, u8);
 	void (*dma_start)(ide_drive_t *);
 	int (*ide_dma_end)(ide_drive_t *drive);
-	int (*ide_dma_on)(ide_drive_t *drive);
-	void (*dma_off_quietly)(ide_drive_t *drive);
 	int (*ide_dma_test_irq)(ide_drive_t *drive);
 	void (*ide_dma_clear_irq)(ide_drive_t *drive);
 	void (*dma_host_on)(ide_drive_t *drive);
@@ -1149,7 +1147,9 @@ static inline u8 ide_max_dma_mode(ide_drive_t *drive)
 	return ide_find_dma_mode(drive, XFER_UDMA_6);
 }
 
+void ide_dma_off_quietly(ide_drive_t *);
 void ide_dma_off(ide_drive_t *);
+void ide_dma_on(ide_drive_t *);
 int ide_set_dma(ide_drive_t *);
 ide_startstop_t ide_dma_intr(ide_drive_t *);
 
@@ -1161,9 +1161,7 @@ extern int ide_release_dma(ide_hwif_t *);
 extern void ide_setup_dma(ide_hwif_t *, unsigned long, unsigned int);
 
 void ide_dma_host_off(ide_drive_t *);
-void ide_dma_off_quietly(ide_drive_t *);
 void ide_dma_host_on(ide_drive_t *);
-extern int __ide_dma_on(ide_drive_t *);
 extern int ide_dma_setup(ide_drive_t *);
 extern void ide_dma_start(ide_drive_t *);
 extern int __ide_dma_end(ide_drive_t *);
@@ -1175,7 +1173,9 @@ extern void ide_dma_timeout(ide_drive_t *);
 static inline int ide_id_dma_bug(ide_drive_t *drive) { return 0; }
 static inline u8 ide_find_dma_mode(ide_drive_t *drive, u8 speed) { return 0; }
 static inline u8 ide_max_dma_mode(ide_drive_t *drive) { return 0; }
+static inline void ide_dma_off_quietly(ide_drive_t *drive) { ; }
 static inline void ide_dma_off(ide_drive_t *drive) { ; }
+static inline void ide_dma_on(ide_drive_t *drive) { ; }
 static inline void ide_dma_verbose(ide_drive_t *drive) { ; }
 static inline int ide_set_dma(ide_drive_t *drive) { return 1; }
 #endif /* CONFIG_BLK_DEV_IDEDMA */
-- 
cgit v1.2.3


From 15ce926ada545cb078711bd9a18c083c93fa01d7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:03 +0100
Subject: ide: merge ->dma_host_{on,off} methods into ->dma_host_set method

Merge ->dma_host_{on,off} methods into ->dma_host_set method
which takes 'int on' argument.

There should be no functionality changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c      |  9 ++------
 drivers/ide/cris/ide-cris.c   |  9 ++------
 drivers/ide/ide-dma.c         | 50 +++++++++++++------------------------------
 drivers/ide/ide-io.c          |  2 +-
 drivers/ide/ide-iops.c        |  8 +++----
 drivers/ide/ide-probe.c       |  2 +-
 drivers/ide/ide.c             |  5 ++---
 drivers/ide/mips/au1xxx-ide.c |  9 ++------
 drivers/ide/pci/cs5520.c      | 17 ++++-----------
 drivers/ide/pci/sc1200.c      |  2 +-
 drivers/ide/pci/sgiioc4.c     | 12 ++++-------
 drivers/ide/pci/trm290.c      |  9 ++------
 drivers/ide/ppc/pmac.c        |  9 ++------
 include/linux/ide.h           |  6 ++----
 14 files changed, 44 insertions(+), 105 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 3a8402bb5dc3..8a5c7205b77c 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -287,11 +287,7 @@ static void icside_set_dma_mode(ide_drive_t *drive, const u8 xfer_mode)
 		ide_xfer_verbose(xfer_mode), 2000 / drive->drive_data);
 }
 
-static void icside_dma_host_off(ide_drive_t *drive)
-{
-}
-
-static void icside_dma_host_on(ide_drive_t *drive)
+static void icside_dma_host_set(ide_drive_t *drive, int on)
 {
 }
 
@@ -410,8 +406,7 @@ static void icside_dma_init(ide_hwif_t *hwif)
 	hwif->dmatable_dma	= 0;
 	hwif->set_dma_mode	= icside_set_dma_mode;
 
-	hwif->dma_host_off	= icside_dma_host_off;
-	hwif->dma_host_on	= icside_dma_host_on;
+	hwif->dma_host_set	= icside_dma_host_set;
 	hwif->dma_setup		= icside_dma_setup;
 	hwif->dma_exec_cmd	= icside_dma_exec_cmd;
 	hwif->dma_start		= icside_dma_start;
diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c
index b0cd0326cf57..dcebc0299f5e 100644
--- a/drivers/ide/cris/ide-cris.c
+++ b/drivers/ide/cris/ide-cris.c
@@ -674,11 +674,7 @@ static void cris_ide_output_data (ide_drive_t *drive, void *, unsigned int);
 static void cris_atapi_input_bytes(ide_drive_t *drive, void *, unsigned int);
 static void cris_atapi_output_bytes(ide_drive_t *drive, void *, unsigned int);
 
-static void cris_dma_host_off(ide_drive_t *drive)
-{
-}
-
-static void cris_dma_host_on(ide_drive_t *drive)
+static void cris_dma_host_set(ide_drive_t *drive, int on)
 {
 }
 
@@ -792,6 +788,7 @@ init_e100_ide (void)
 		hwif->ata_output_data = &cris_ide_output_data;
 		hwif->atapi_input_bytes = &cris_atapi_input_bytes;
 		hwif->atapi_output_bytes = &cris_atapi_output_bytes;
+		hwif->dma_host_set = &cris_dma_host_set;
 		hwif->ide_dma_end = &cris_dma_end;
 		hwif->dma_setup = &cris_dma_setup;
 		hwif->dma_exec_cmd = &cris_dma_exec_cmd;
@@ -802,8 +799,6 @@ init_e100_ide (void)
 		hwif->OUTBSYNC = &cris_ide_outbsync;
 		hwif->INB = &cris_ide_inb;
 		hwif->INW = &cris_ide_inw;
-		hwif->dma_host_off = &cris_dma_host_off;
-		hwif->dma_host_on = &cris_dma_host_on;
 		hwif->cbl = ATA_CBL_PATA40;
 		hwif->host_flags |= IDE_HFLAG_NO_ATAPI_DMA;
 		hwif->pio_mask = ATA_PIO4,
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 780911e0537e..c9648b1ef220 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -408,23 +408,28 @@ static int dma_timer_expiry (ide_drive_t *drive)
 }
 
 /**
- *	ide_dma_host_off	-	Generic DMA kill
+ *	ide_dma_host_set	-	Enable/disable DMA on a host
  *	@drive: drive to control
  *
- *	Perform the generic IDE controller DMA off operation. This
- *	works for most IDE bus mastering controllers
+ *	Enable/disable DMA on an IDE controller following generic
+ *	bus-mastering IDE controller behaviour.
  */
 
-void ide_dma_host_off(ide_drive_t *drive)
+void ide_dma_host_set(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	u8 unit			= (drive->select.b.unit & 0x01);
 	u8 dma_stat		= hwif->INB(hwif->dma_status);
 
-	hwif->OUTB((dma_stat & ~(1<<(5+unit))), hwif->dma_status);
+	if (on)
+		dma_stat |= (1 << (5 + unit));
+	else
+		dma_stat &= ~(1 << (5 + unit));
+
+	hwif->OUTB(dma_stat, hwif->dma_status);
 }
 
-EXPORT_SYMBOL(ide_dma_host_off);
+EXPORT_SYMBOL_GPL(ide_dma_host_set);
 #endif /* CONFIG_BLK_DEV_IDEDMA_PCI */
 
 /**
@@ -439,7 +444,7 @@ void ide_dma_off_quietly(ide_drive_t *drive)
 	drive->using_dma = 0;
 	ide_toggle_bounce(drive, 0);
 
-	drive->hwif->dma_host_off(drive);
+	drive->hwif->dma_host_set(drive, 0);
 }
 
 EXPORT_SYMBOL(ide_dma_off_quietly);
@@ -460,29 +465,6 @@ void ide_dma_off(ide_drive_t *drive)
 
 EXPORT_SYMBOL(ide_dma_off);
 
-#ifdef CONFIG_BLK_DEV_IDEDMA_PCI
-/**
- *	ide_dma_host_on	-	Enable DMA on a host
- *	@drive: drive to enable for DMA
- *
- *	Enable DMA on an IDE controller following generic bus mastering
- *	IDE controller behaviour
- */
-
-void ide_dma_host_on(ide_drive_t *drive)
-{
-	if (1) {
-		ide_hwif_t *hwif	= HWIF(drive);
-		u8 unit			= (drive->select.b.unit & 0x01);
-		u8 dma_stat		= hwif->INB(hwif->dma_status);
-
-		hwif->OUTB((dma_stat|(1<<(5+unit))), hwif->dma_status);
-	}
-}
-
-EXPORT_SYMBOL(ide_dma_host_on);
-#endif
-
 /**
  *	ide_dma_on		-	Enable DMA on a device
  *	@drive: drive to enable DMA on
@@ -495,7 +477,7 @@ void ide_dma_on(ide_drive_t *drive)
 	drive->using_dma = 1;
 	ide_toggle_bounce(drive, 1);
 
-	drive->hwif->dma_host_on(drive);
+	drive->hwif->dma_host_set(drive, 1);
 }
 
 EXPORT_SYMBOL(ide_dma_on);
@@ -980,10 +962,8 @@ void ide_setup_dma(ide_hwif_t *hwif, unsigned long base, unsigned num_ports)
 	if (!(hwif->dma_prdtable))
 		hwif->dma_prdtable	= (hwif->dma_base + 4);
 
-	if (!hwif->dma_host_off)
-		hwif->dma_host_off = &ide_dma_host_off;
-	if (!hwif->dma_host_on)
-		hwif->dma_host_on = &ide_dma_host_on;
+	if (!hwif->dma_host_set)
+		hwif->dma_host_set = &ide_dma_host_set;
 	if (!hwif->dma_setup)
 		hwif->dma_setup = &ide_dma_setup;
 	if (!hwif->dma_exec_cmd)
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index b5a7d2578ab8..e37b09c81e3a 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -219,7 +219,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 		 * we could be smarter and check for current xfer_speed
 		 * in struct drive etc...
 		 */
-		if (drive->hwif->dma_host_on == NULL)
+		if (drive->hwif->dma_host_set == NULL)
 			break;
 		/*
 		 * TODO: respect ->using_dma setting
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index e30f67e09b9c..595a5cef41a2 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -742,8 +742,8 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 //		msleep(50);
 
 #ifdef CONFIG_BLK_DEV_IDEDMA
-	if (hwif->dma_host_on)	/* check if host supports DMA */
-		hwif->dma_host_off(drive);
+	if (hwif->dma_host_set)	/* check if host supports DMA */
+		hwif->dma_host_set(drive, 0);
 #endif
 
 	/* Skip setting PIO flow-control modes on pre-EIDE drives */
@@ -801,8 +801,8 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 #ifdef CONFIG_BLK_DEV_IDEDMA
 	if ((speed >= XFER_SW_DMA_0 || (hwif->host_flags & IDE_HFLAG_VDMA)) &&
 	    drive->using_dma)
-		hwif->dma_host_on(drive);
-	else if (hwif->dma_host_on)	/* check if host supports DMA */
+		hwif->dma_host_set(drive, 1);
+	else if (hwif->dma_host_set)	/* check if host supports DMA */
 		ide_dma_off_quietly(drive);
 #endif
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index b363a96607d1..fa95e79b9505 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -833,7 +833,7 @@ static void probe_hwif(ide_hwif_t *hwif)
 
 			drive->nice1 = 1;
 
-			if (hwif->dma_host_on)
+			if (hwif->dma_host_set)
 				ide_set_dma(drive);
 		}
 	}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 095ff34870da..7819fbd4d5fd 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -433,14 +433,13 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 	hwif->atapi_input_bytes		= tmp_hwif->atapi_input_bytes;
 	hwif->atapi_output_bytes	= tmp_hwif->atapi_output_bytes;
 
+	hwif->dma_host_set		= tmp_hwif->dma_host_set;
 	hwif->dma_setup			= tmp_hwif->dma_setup;
 	hwif->dma_exec_cmd		= tmp_hwif->dma_exec_cmd;
 	hwif->dma_start			= tmp_hwif->dma_start;
 	hwif->ide_dma_end		= tmp_hwif->ide_dma_end;
 	hwif->ide_dma_test_irq		= tmp_hwif->ide_dma_test_irq;
 	hwif->ide_dma_clear_irq		= tmp_hwif->ide_dma_clear_irq;
-	hwif->dma_host_on		= tmp_hwif->dma_host_on;
-	hwif->dma_host_off		= tmp_hwif->dma_host_off;
 	hwif->dma_lost_irq		= tmp_hwif->dma_lost_irq;
 	hwif->dma_timeout		= tmp_hwif->dma_timeout;
 
@@ -834,7 +833,7 @@ int set_using_dma(ide_drive_t *drive, int arg)
 	if (!drive->id || !(drive->id->capability & 1))
 		goto out;
 
-	if (hwif->dma_host_on == NULL)
+	if (hwif->dma_host_set == NULL)
 		goto out;
 
 	err = -EBUSY;
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 4dfdca4ccbde..27abff6f6ba2 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -395,11 +395,7 @@ static int auide_dma_test_irq(ide_drive_t *drive)
 	return 0;
 }
 
-static void auide_dma_host_on(ide_drive_t *drive)
-{
-}
-
-static void auide_dma_host_off(ide_drive_t *drive)
+static void auide_dma_host_set(ide_drive_t *drive, int on)
 {
 }
 
@@ -674,13 +670,12 @@ static int au_ide_probe(struct device *dev)
 
 	hwif->mdma_filter		= &auide_mdma_filter;
 
+	hwif->dma_host_set		= &auide_dma_host_set;
 	hwif->dma_exec_cmd              = &auide_dma_exec_cmd;
 	hwif->dma_start                 = &auide_dma_start;
 	hwif->ide_dma_end               = &auide_dma_end;
 	hwif->dma_setup                 = &auide_dma_setup;
 	hwif->ide_dma_test_irq          = &auide_dma_test_irq;
-	hwif->dma_host_off		= &auide_dma_host_off;
-	hwif->dma_host_on		= &auide_dma_host_on;
 	hwif->dma_lost_irq		= &auide_dma_lost_irq;
 #else /* !CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA */
 	hwif->channel                   = 0;
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index 2bd52af83d37..6ec00b8d7ec1 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -107,18 +107,10 @@ static void cs5520_set_dma_mode(ide_drive_t *drive, const u8 speed)
  *	ATAPI is harder so disable it for now using IDE_HFLAG_NO_ATAPI_DMA
  */
 
-static void cs5520_dma_host_on(ide_drive_t *drive)
+static void cs5520_dma_host_set(ide_drive_t *drive, int on)
 {
-	drive->vdma = 1;
-
-	ide_dma_host_on(drive);
-}
-
-static void cs5520_dma_host_off(ide_drive_t *drive)
-{
-	drive->vdma = 0;
-
-	ide_dma_host_off(drive);
+	drive->vdma = on;
+	ide_dma_host_set(drive, on);
 }
 
 static void __devinit init_hwif_cs5520(ide_hwif_t *hwif)
@@ -129,8 +121,7 @@ static void __devinit init_hwif_cs5520(ide_hwif_t *hwif)
 	if (hwif->dma_base == 0)
 		return;
 
-	hwif->dma_host_on  = &cs5520_dma_host_on;
-	hwif->dma_host_off = &cs5520_dma_host_off;
+	hwif->dma_host_set = &cs5520_dma_host_set;
 }
 
 #define DECLARE_CS_DEV(name_str)				\
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index 9303dfee7780..32fdf53379f5 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -222,7 +222,7 @@ static void sc1200_set_pio_mode(ide_drive_t *drive, const u8 pio)
 		printk("SC1200: %s: changing (U)DMA mode\n", drive->name);
 		ide_dma_off_quietly(drive);
 		if (ide_set_dma_mode(drive, mode) == 0 && drive->using_dma)
-			hwif->dma_host_on(drive);
+			hwif->dma_host_set(drive, 1);
 		return;
 	}
 
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 8c4e94bd4449..9fb35c528d54 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -288,13 +288,10 @@ sgiioc4_ide_dma_test_irq(ide_drive_t * drive)
 	return sgiioc4_checkirq(HWIF(drive));
 }
 
-static void sgiioc4_dma_host_on(ide_drive_t * drive)
+static void sgiioc4_dma_host_set(ide_drive_t *drive, int on)
 {
-}
-
-static void sgiioc4_dma_host_off(ide_drive_t * drive)
-{
-	sgiioc4_clearirq(drive);
+	if (!on)
+		sgiioc4_clearirq(drive);
 }
 
 static void
@@ -578,12 +575,11 @@ ide_init_sgiioc4(ide_hwif_t * hwif)
 
 	hwif->mwdma_mask = ATA_MWDMA2_ONLY;
 
+	hwif->dma_host_set = &sgiioc4_dma_host_set;
 	hwif->dma_setup = &sgiioc4_ide_dma_setup;
 	hwif->dma_start = &sgiioc4_ide_dma_start;
 	hwif->ide_dma_end = &sgiioc4_ide_dma_end;
 	hwif->ide_dma_test_irq = &sgiioc4_ide_dma_test_irq;
-	hwif->dma_host_on = &sgiioc4_dma_host_on;
-	hwif->dma_host_off = &sgiioc4_dma_host_off;
 	hwif->dma_lost_irq = &sgiioc4_dma_lost_irq;
 	hwif->dma_timeout = &ide_dma_timeout;
 }
diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index 0151d7fdfb8a..04cd893e1ab0 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -241,11 +241,7 @@ static int trm290_ide_dma_test_irq (ide_drive_t *drive)
 	return (status == 0x00ff);
 }
 
-static void trm290_dma_host_on(ide_drive_t *drive)
-{
-}
-
-static void trm290_dma_host_off(ide_drive_t *drive)
+static void trm290_dma_host_set(ide_drive_t *drive, int on)
 {
 }
 
@@ -289,8 +285,7 @@ static void __devinit init_hwif_trm290(ide_hwif_t *hwif)
 
 	ide_setup_dma(hwif, (hwif->config_data + 4) ^ (hwif->channel ? 0x0080 : 0x0000), 3);
 
-	hwif->dma_host_off	= &trm290_dma_host_off;
-	hwif->dma_host_on	= &trm290_dma_host_on;
+	hwif->dma_host_set	= &trm290_dma_host_set;
 	hwif->dma_setup 	= &trm290_dma_setup;
 	hwif->dma_exec_cmd	= &trm290_dma_exec_cmd;
 	hwif->dma_start 	= &trm290_dma_start;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ca99b69cfac7..6a4b0d479897 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1698,11 +1698,7 @@ pmac_ide_dma_test_irq (ide_drive_t *drive)
 	return 1;
 }
 
-static void pmac_ide_dma_host_off(ide_drive_t *drive)
-{
-}
-
-static void pmac_ide_dma_host_on(ide_drive_t *drive)
+static void pmac_ide_dma_host_set(ide_drive_t *drive, int on)
 {
 }
 
@@ -1748,13 +1744,12 @@ pmac_ide_setup_dma(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 		return;
 	}
 
+	hwif->dma_host_set = &pmac_ide_dma_host_set;
 	hwif->dma_setup = &pmac_ide_dma_setup;
 	hwif->dma_exec_cmd = &pmac_ide_dma_exec_cmd;
 	hwif->dma_start = &pmac_ide_dma_start;
 	hwif->ide_dma_end = &pmac_ide_dma_end;
 	hwif->ide_dma_test_irq = &pmac_ide_dma_test_irq;
-	hwif->dma_host_off = &pmac_ide_dma_host_off;
-	hwif->dma_host_on = &pmac_ide_dma_host_on;
 	hwif->dma_timeout = &ide_dma_timeout;
 	hwif->dma_lost_irq = &pmac_ide_dma_lost_irq;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 140864d63aed..ffb76d0d0814 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -542,14 +542,13 @@ typedef struct hwif_s {
 	void (*atapi_input_bytes)(ide_drive_t *, void *, u32);
 	void (*atapi_output_bytes)(ide_drive_t *, void *, u32);
 
+	void (*dma_host_set)(ide_drive_t *, int);
 	int (*dma_setup)(ide_drive_t *);
 	void (*dma_exec_cmd)(ide_drive_t *, u8);
 	void (*dma_start)(ide_drive_t *);
 	int (*ide_dma_end)(ide_drive_t *drive);
 	int (*ide_dma_test_irq)(ide_drive_t *drive);
 	void (*ide_dma_clear_irq)(ide_drive_t *drive);
-	void (*dma_host_on)(ide_drive_t *drive);
-	void (*dma_host_off)(ide_drive_t *drive);
 	void (*dma_lost_irq)(ide_drive_t *drive);
 	void (*dma_timeout)(ide_drive_t *drive);
 
@@ -1160,8 +1159,7 @@ extern void ide_destroy_dmatable(ide_drive_t *);
 extern int ide_release_dma(ide_hwif_t *);
 extern void ide_setup_dma(ide_hwif_t *, unsigned long, unsigned int);
 
-void ide_dma_host_off(ide_drive_t *);
-void ide_dma_host_on(ide_drive_t *);
+void ide_dma_host_set(ide_drive_t *, int);
 extern int ide_dma_setup(ide_drive_t *);
 extern void ide_dma_start(ide_drive_t *);
 extern int __ide_dma_end(ide_drive_t *);
-- 
cgit v1.2.3


From f01393e48c44e30f7c9a36c8b98a07b0232580fe Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:03 +0100
Subject: ide: merge ->fixup and ->quirkproc methods

* Assign drive->quirk_list in ->quirkproc implementations:
  - hpt366.c::hpt3xx_quirkproc()
  - pdc202xx_new.c::pdcnew_quirkproc()
  - pdc202xx_old.c::pdc202xx_quirkproc()

* Make ->quirkproc void.

* Move calling ->quirkproc from do_identify() to probe_hwif().

* Convert it821x_fixups() to it821x_quirkproc() in it821x.c.

* Convert siimage_fixup() to sil_quirkproc() in siimage.c, also remove
  no longer needed drive->present check from is_dev_seagate_sata().

* Convert ide_undecoded_slave() to accept 'drive' instead of 'hwif'
  as an argument.  Then convert ide_register_hw() to accept 'quirkproc'
  argument instead of 'fixup' one.

* Remove no longer needed ->fixup method.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c        | 20 ++++++++++----------
 drivers/ide/ide.c              |  8 +++-----
 drivers/ide/pci/hpt366.c       | 11 +++++++----
 drivers/ide/pci/it821x.c       | 37 ++++++++++++-------------------------
 drivers/ide/pci/pdc202xx_new.c | 11 +++++++----
 drivers/ide/pci/pdc202xx_old.c | 11 +++++++----
 drivers/ide/pci/siimage.c      | 15 +++++++--------
 drivers/ide/setup-pci.c        |  2 --
 include/linux/ide.h            | 10 ++++------
 9 files changed, 57 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index fa95e79b9505..c446e348e29a 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -235,9 +235,6 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	drive->media = ide_disk;
 	printk("%s DISK drive\n", (id->config == 0x848a) ? "CFA" : "ATA" );
 
-	if (hwif->quirkproc)
-		drive->quirk_list = hwif->quirkproc(drive);
-
 	return;
 
 err_misc:
@@ -676,19 +673,18 @@ out:
 
 /**
  *	ide_undecoded_slave	-	look for bad CF adapters
- *	@hwif: interface
+ *	@drive1: drive
  *
  *	Analyse the drives on the interface and attempt to decide if we
  *	have the same drive viewed twice. This occurs with crap CF adapters
  *	and PCMCIA sometimes.
  */
 
-void ide_undecoded_slave(ide_hwif_t *hwif)
+void ide_undecoded_slave(ide_drive_t *drive1)
 {
-	ide_drive_t *drive0 = &hwif->drives[0];
-	ide_drive_t *drive1 = &hwif->drives[1];
+	ide_drive_t *drive0 = &drive1->hwif->drives[0];
 
-	if (drive0->present == 0 || drive1->present == 0)
+	if ((drive1->dn & 1) == 0 || drive0->present == 0)
 		return;
 
 	/* If the models don't match they are not the same product */
@@ -817,8 +813,12 @@ static void probe_hwif(ide_hwif_t *hwif)
 		return;
 	}
 
-	if (hwif->fixup)
-		hwif->fixup(hwif);
+	for (unit = 0; unit < MAX_DRIVES; unit++) {
+		ide_drive_t *drive = &hwif->drives[unit];
+
+		if (drive->present && hwif->quirkproc)
+			hwif->quirkproc(drive);
+	}
 
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
 		ide_drive_t *drive = &hwif->drives[unit];
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 7819fbd4d5fd..0d7328e0fb96 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -414,8 +414,6 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 	hwif->cds			= tmp_hwif->cds;
 #endif
 
-	hwif->fixup			= tmp_hwif->fixup;
-
 	hwif->set_pio_mode		= tmp_hwif->set_pio_mode;
 	hwif->set_dma_mode		= tmp_hwif->set_dma_mode;
 	hwif->mdma_filter		= tmp_hwif->mdma_filter;
@@ -680,7 +678,7 @@ void ide_setup_ports (	hw_regs_t *hw,
 /**
  *	ide_register_hw		-	register IDE interface
  *	@hw: hardware registers
- *	@fixup: fixup function
+ *	@quirkproc: quirkproc function
  *	@initializing: set while initializing built-in drivers
  *	@hwifp: pointer to returned hwif
  *
@@ -690,7 +688,7 @@ void ide_setup_ports (	hw_regs_t *hw,
  *	Returns -1 on error.
  */
 
-int ide_register_hw(hw_regs_t *hw, void (*fixup)(ide_hwif_t *),
+int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
 		    int initializing, ide_hwif_t **hwifp)
 {
 	int index, retry = 1;
@@ -726,7 +724,7 @@ found:
 	memcpy(hwif->io_ports, hw->io_ports, sizeof(hwif->io_ports));
 	hwif->irq = hw->irq;
 	hwif->noprobe = 0;
-	hwif->fixup = fixup;
+	hwif->quirkproc = quirkproc;
 	hwif->chipset = hw->chipset;
 	hwif->gendev.parent = hw->dev;
 	hwif->ack_intr = hw->ack_intr;
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 3777fb8c8043..12685939a813 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -725,15 +725,18 @@ static void hpt3xx_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	hpt3xx_set_mode(drive, XFER_PIO_0 + pio);
 }
 
-static int hpt3xx_quirkproc(ide_drive_t *drive)
+static void hpt3xx_quirkproc(ide_drive_t *drive)
 {
 	struct hd_driveid *id	= drive->id;
 	const  char **list	= quirk_drives;
 
 	while (*list)
-		if (strstr(id->model, *list++))
-			return 1;
-	return 0;
+		if (strstr(id->model, *list++)) {
+			drive->quirk_list = 1;
+			return;
+		}
+
+	drive->quirk_list = 0;
 }
 
 static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index 99b7d763b6c7..e610a5340fdc 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -431,33 +431,29 @@ static u8 __devinit ata66_it821x(ide_hwif_t *hwif)
 }
 
 /**
- *	it821x_fixup	-	post init callback
- *	@hwif: interface
+ *	it821x_quirkproc	-	post init callback
+ *	@drive: drive
  *
- *	This callback is run after the drives have been probed but
+ *	This callback is run after the drive has been probed but
  *	before anything gets attached. It allows drivers to do any
  *	final tuning that is needed, or fixups to work around bugs.
  */
 
-static void __devinit it821x_fixups(ide_hwif_t *hwif)
+static void __devinit it821x_quirkproc(ide_drive_t *drive)
 {
-	struct it821x_dev *itdev = ide_get_hwifdata(hwif);
-	int i;
+	struct it821x_dev *itdev = ide_get_hwifdata(drive->hwif);
+	struct hd_driveid *id = drive->id;
+	u16 *idbits = (u16 *)drive->id;
 
-	if(!itdev->smart) {
+	if (!itdev->smart) {
 		/*
 		 *	If we are in pass through mode then not much
 		 *	needs to be done, but we do bother to clear the
 		 *	IRQ mask as we may well be in PIO (eg rev 0x10)
 		 *	for now and we know unmasking is safe on this chipset.
 		 */
-		for (i = 0; i < 2; i++) {
-			ide_drive_t *drive = &hwif->drives[i];
-			if(drive->present)
-				drive->unmask = 1;
-		}
-		return;
-	}
+		drive->unmask = 1;
+	} else {
 	/*
 	 *	Perform fixups on smart mode. We need to "lose" some
 	 *	capabilities the firmware lacks but does not filter, and
@@ -465,16 +461,6 @@ static void __devinit it821x_fixups(ide_hwif_t *hwif)
 	 *	in RAID mode.
 	 */
 
-	for(i = 0; i < 2; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
-		struct hd_driveid *id;
-		u16 *idbits;
-
-		if(!drive->present)
-			continue;
-		id = drive->id;
-		idbits = (u16 *)drive->id;
-
 		/* Check for RAID v native */
 		if(strstr(id->model, "Integrated Technology Express")) {
 			/* In raid mode the ident block is slightly buggy
@@ -537,6 +523,8 @@ static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 	struct it821x_dev *idev = kzalloc(sizeof(struct it821x_dev), GFP_KERNEL);
 	u8 conf;
 
+	hwif->quirkproc = &it821x_quirkproc;
+
 	if (idev == NULL) {
 		printk(KERN_ERR "it821x: out of memory, falling back to legacy behaviour.\n");
 		return;
@@ -633,7 +621,6 @@ static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const cha
 		.name		= name_str,		\
 		.init_chipset	= init_chipset_it821x,	\
 		.init_hwif	= init_hwif_it821x,	\
-		.fixup	 	= it821x_fixups,	\
 		.host_flags	= IDE_HFLAG_BOOTABLE,	\
 		.pio_mask	= ATA_PIO4,		\
 	}
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index ef4a99b99d1f..89d2363a1ebd 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -203,14 +203,17 @@ static u8 pdcnew_cable_detect(ide_hwif_t *hwif)
 		return ATA_CBL_PATA80;
 }
 
-static int pdcnew_quirkproc(ide_drive_t *drive)
+static void pdcnew_quirkproc(ide_drive_t *drive)
 {
 	const char **list, *model = drive->id->model;
 
 	for (list = pdc_quirk_drives; *list != NULL; list++)
-		if (strstr(model, *list) != NULL)
-			return 2;
-	return 0;
+		if (strstr(model, *list) != NULL) {
+			drive->quirk_list = 2;
+			return;
+		}
+
+	drive->quirk_list = 0;
 }
 
 static void pdcnew_reset(ide_drive_t *drive)
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index 67b2781e2213..3a1e081fe390 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -176,14 +176,17 @@ static void pdc_old_disable_66MHz_clock(ide_hwif_t *hwif)
 	outb(clock & ~(hwif->channel ? 0x08 : 0x02), clock_reg);
 }
 
-static int pdc202xx_quirkproc (ide_drive_t *drive)
+static void pdc202xx_quirkproc(ide_drive_t *drive)
 {
 	const char **list, *model = drive->id->model;
 
 	for (list = pdc_quirk_drives; *list != NULL; list++)
-		if (strstr(model, *list) != NULL)
-			return 2;
-	return 0;
+		if (strstr(model, *list) != NULL) {
+			drive->quirk_list = 2;
+			return;
+		}
+
+	drive->quirk_list = 0;
 }
 
 static void pdc202xx_old_ide_dma_start(ide_drive_t *drive)
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 7b45eaf5afd9..908f37b4e0ee 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -713,9 +713,6 @@ static int is_dev_seagate_sata(ide_drive_t *drive)
 	const char *s = &drive->id->model[0];
 	unsigned len;
 
-	if (!drive->present)
-		return 0;
-
 	len = strnlen(s, sizeof(drive->id->model));
 
 	if ((len > 4) && (!memcmp(s, "ST", 2))) {
@@ -730,18 +727,20 @@ static int is_dev_seagate_sata(ide_drive_t *drive)
 }
 
 /**
- *	siimage_fixup		-	post probe fixups
- *	@hwif: interface to fix up
+ *	sil_quirkproc		-	post probe fixups
+ *	@drive: drive
  *
  *	Called after drive probe we use this to decide whether the
  *	Seagate fixup must be applied. This used to be in init_iops but
  *	that can occur before we know what drives are present.
  */
 
-static void __devinit siimage_fixup(ide_hwif_t *hwif)
+static void __devinit sil_quirkproc(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
+
 	/* Try and raise the rqsize */
-	if (!is_sata(hwif) || !is_dev_seagate_sata(&hwif->drives[0]))
+	if (!is_sata(hwif) || !is_dev_seagate_sata(drive))
 		hwif->rqsize = 128;
 }
 
@@ -804,6 +803,7 @@ static void __devinit init_hwif_siimage(ide_hwif_t *hwif)
 
 	hwif->set_pio_mode = &sil_set_pio_mode;
 	hwif->set_dma_mode = &sil_set_dma_mode;
+	hwif->quirkproc = &sil_quirkproc;
 
 	if (sata) {
 		static int first = 1;
@@ -842,7 +842,6 @@ static void __devinit init_hwif_siimage(ide_hwif_t *hwif)
 		.init_chipset	= init_chipset_siimage,	\
 		.init_iops	= init_iops_siimage,	\
 		.init_hwif	= init_hwif_siimage,	\
-		.fixup		= siimage_fixup,	\
 		.host_flags	= IDE_HFLAG_BOOTABLE,	\
 		.pio_mask	= ATA_PIO4,		\
 		.mwdma_mask	= ATA_MWDMA2,		\
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index bbfdf7e0f182..d89f84d41b08 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -555,8 +555,6 @@ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d, int
 		    (d->host_flags & IDE_HFLAG_FORCE_LEGACY_IRQS))
 			hwif->irq = port ? 15 : 14;
 
-		hwif->fixup = d->fixup;
-
 		hwif->host_flags = d->host_flags;
 		hwif->pio_mask = d->pio_mask;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ffb76d0d0814..dd50a5c5ec10 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -199,7 +199,8 @@ typedef struct hw_regs_s {
 
 struct hwif_s * ide_find_port(unsigned long);
 
-int ide_register_hw(hw_regs_t *, void (*)(struct hwif_s *), int,
+struct ide_drive_s;
+int ide_register_hw(hw_regs_t *, void (*)(struct ide_drive_s *), int,
 		    struct hwif_s **);
 
 void ide_setup_ports(	hw_regs_t *hw,
@@ -527,15 +528,13 @@ typedef struct hwif_s {
 	/* special host masking for drive selection */
 	void	(*maskproc)(ide_drive_t *, int);
 	/* check host's drive quirk list */
-	int	(*quirkproc)(ide_drive_t *);
+	void	(*quirkproc)(ide_drive_t *);
 	/* driver soft-power interface */
 	int	(*busproc)(ide_drive_t *, int);
 #endif
 	u8 (*mdma_filter)(ide_drive_t *);
 	u8 (*udma_filter)(ide_drive_t *);
 
-	void (*fixup)(struct hwif_s *);
-
 	void (*ata_input_data)(ide_drive_t *, void *, u32);
 	void (*ata_output_data)(ide_drive_t *, void *, u32);
 
@@ -1108,7 +1107,6 @@ struct ide_port_info {
 	void			(*init_iops)(ide_hwif_t *);
 	void                    (*init_hwif)(ide_hwif_t *);
 	void			(*init_dma)(ide_hwif_t *, unsigned long);
-	void			(*fixup)(ide_hwif_t *);
 	ide_pci_enablebit_t	enablebits[2];
 	hwif_chipset_t		chipset;
 	u8			extra;
@@ -1203,7 +1201,7 @@ extern void ide_unregister (unsigned int index);
 void ide_register_region(struct gendisk *);
 void ide_unregister_region(struct gendisk *);
 
-void ide_undecoded_slave(ide_hwif_t *);
+void ide_undecoded_slave(ide_drive_t *);
 
 int ide_device_add(u8 idx[4]);
 
-- 
cgit v1.2.3


From 151575e4644f917d3a9f83c777ac3543284954f8 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:05 +0100
Subject: ide: remove ideprobe_init()

* Rename ide_device_add() to ide_device_add_all() and make it accept
  'u8 idx[MAX_HWIFS]' instead of 'u8 idx[4]' as an argument.

* Add ide_device_add() wrapper for ide_device_add_all().

* Convert ide_generic_init() to use ide_device_add_all().

* Remove no longer needed ideprobe_init().

There should be no functionality changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-generic.c |  8 ++++++-
 drivers/ide/ide-probe.c   | 54 ++++++++++++++---------------------------------
 include/linux/ide.h       |  3 +--
 3 files changed, 24 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 0f72b98d727f..bb30c29f6ec0 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -14,10 +14,16 @@
 
 static int __init ide_generic_init(void)
 {
+	u8 idx[MAX_HWIFS];
+	int i;
+
 	if (ide_hwifs[0].io_ports[IDE_DATA_OFFSET])
 		ide_get_lock(NULL, NULL); /* for atari only */
 
-	(void)ideprobe_init();
+	for (i = 0; i < MAX_HWIFS; i++)
+		idx[i] = ide_hwifs[i].present ? 0xff : i;
+
+	ide_device_add_all(idx);
 
 	if (ide_hwifs[0].io_ports[IDE_DATA_OFFSET])
 		ide_release_lock();	/* for atari only */
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index a7a1cd85f15d..1af94ac4892a 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1340,52 +1340,19 @@ static void hwif_register_devices(ide_hwif_t *hwif)
 	}
 }
 
-int ideprobe_init (void)
-{
-	unsigned int index;
-	int probe[MAX_HWIFS];
-
-	memset(probe, 0, MAX_HWIFS * sizeof(int));
-	for (index = 0; index < MAX_HWIFS; ++index)
-		probe[index] = !ide_hwifs[index].present;
-
-	for (index = 0; index < MAX_HWIFS; ++index)
-		if (probe[index])
-			probe_hwif(&ide_hwifs[index]);
-	for (index = 0; index < MAX_HWIFS; ++index)
-		if (probe[index])
-			hwif_init(&ide_hwifs[index]);
-	for (index = 0; index < MAX_HWIFS; ++index) {
-		if (probe[index]) {
-			ide_hwif_t *hwif = &ide_hwifs[index];
-			if (!hwif->present)
-				continue;
-			if (hwif->chipset == ide_unknown || hwif->chipset == ide_forced)
-				hwif->chipset = ide_generic;
-			hwif_register_devices(hwif);
-		}
-	}
-	for (index = 0; index < MAX_HWIFS; ++index)
-		if (probe[index])
-			ide_proc_register_port(&ide_hwifs[index]);
-	return 0;
-}
-
-EXPORT_SYMBOL_GPL(ideprobe_init);
-
-int ide_device_add(u8 idx[4])
+int ide_device_add_all(u8 idx[MAX_HWIFS])
 {
 	ide_hwif_t *hwif;
 	int i, rc = 0;
 
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < MAX_HWIFS; i++) {
 		if (idx[i] == 0xff)
 			continue;
 
 		probe_hwif(&ide_hwifs[idx[i]]);
 	}
 
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < MAX_HWIFS; i++) {
 		if (idx[i] == 0xff)
 			continue;
 
@@ -1399,7 +1366,7 @@ int ide_device_add(u8 idx[4])
 		}
 	}
 
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < MAX_HWIFS; i++) {
 		if (idx[i] == 0xff)
 			continue;
 
@@ -1413,12 +1380,23 @@ int ide_device_add(u8 idx[4])
 		}
 	}
 
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < MAX_HWIFS; i++) {
 		if (idx[i] != 0xff)
 			ide_proc_register_port(&ide_hwifs[idx[i]]);
 	}
 
 	return rc;
 }
+EXPORT_SYMBOL_GPL(ide_device_add_all);
+
+int ide_device_add(u8 idx[4])
+{
+	u8 idx_all[MAX_HWIFS];
+	int i;
 
+	for (i = 0; i < MAX_HWIFS; i++)
+		idx_all[i] = (i < 4) ? idx[i] : 0xff;
+
+	return ide_device_add_all(idx_all);
+}
 EXPORT_SYMBOL_GPL(ide_device_add);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index dd50a5c5ec10..d7c0f9a8bd9e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1011,8 +1011,6 @@ extern void do_ide_request(struct request_queue *);
 
 void ide_init_disk(struct gendisk *, ide_drive_t *);
 
-extern int ideprobe_init(void);
-
 #ifdef CONFIG_IDEPCI_PCIBUS_ORDER
 extern void ide_scan_pcibus(int scan_direction) __init;
 extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *owner, const char *mod_name);
@@ -1203,6 +1201,7 @@ void ide_unregister_region(struct gendisk *);
 
 void ide_undecoded_slave(ide_drive_t *);
 
+int ide_device_add_all(u8 idx[MAX_HWIFS]);
 int ide_device_add(u8 idx[4]);
 
 static inline void *ide_get_hwifdata (ide_hwif_t * hwif)
-- 
cgit v1.2.3


From b0d5bc27ce995adaafbc114b92fa76815025c94e Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Sat, 26 Jan 2008 20:13:05 +0100
Subject: ide: Fix build break caused by "ide: remove ideprobe_init()"

Fix build break of powerpc holly_defconfig:

In file included from arch/powerpc/platforms/embedded6xx/holly.c:24:
include/linux/ide.h:1206: error: 'CONFIG_IDE_MAX_HWIFS' undeclared here (not in a function)

There's no need to have a sized array in the prototype, might as well
turn it into a pointer.

It could probably be argued that large parts of the include file can be
covered under #ifdef CONFIG_IDE, but that's a larger undertaking.

Signed-off-by: Olof Johansson <olof@lixom.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c | 2 +-
 include/linux/ide.h     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 1af94ac4892a..18e9b82e132c 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1340,7 +1340,7 @@ static void hwif_register_devices(ide_hwif_t *hwif)
 	}
 }
 
-int ide_device_add_all(u8 idx[MAX_HWIFS])
+int ide_device_add_all(u8 *idx)
 {
 	ide_hwif_t *hwif;
 	int i, rc = 0;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index d7c0f9a8bd9e..ce9b16f38c08 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1201,7 +1201,7 @@ void ide_unregister_region(struct gendisk *);
 
 void ide_undecoded_slave(ide_drive_t *);
 
-int ide_device_add_all(u8 idx[MAX_HWIFS]);
+int ide_device_add_all(u8 *idx);
 int ide_device_add(u8 idx[4]);
 
 static inline void *ide_get_hwifdata (ide_hwif_t * hwif)
-- 
cgit v1.2.3


From 57c802e84f9c759c3d1794a9dbe81bc10444df62 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:05 +0100
Subject: ide: add ide_init_port_hw() helper

* Add ide_init_port_hw() helper.

* rapide.c: convert rapide_locate_hwif() to rapide_setup_ports()
  and use ide_init_port_hw().

* ide_platform.c: convert plat_ide_locate_hwif() to plat_ide_setup_ports()
  and use ide_init_port_hw().

* sgiioc4.c: use ide_init_port_hw().

* pmac.c: add 'hw_regs_t *hw' argument to pmac_ide_setup_device(),
  setup 'hw' in pmac_ide_{macio,pci}_attach() and use ide_init_port_hw()
  in pmac_ide_setup_device().

This patch is a preparation for the future changes in the IDE probing code.

There should be no functionality changes caused by this patch.

Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Anton Vorontsov <avorontsov@ru.mvista.com>
Cc: Jeremy Higdon <jeremy@sgi.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/rapide.c          | 34 ++++++++++++------------
 drivers/ide/ide.c                 | 19 +++++++++-----
 drivers/ide/legacy/ide_platform.c | 55 ++++++++++++++++++++-------------------
 drivers/ide/pci/sgiioc4.c         |  9 +++----
 drivers/ide/ppc/pmac.c            | 29 +++++++++++++--------
 include/linux/ide.h               |  1 +
 6 files changed, 81 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index c709d37ec095..0267467d1796 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -13,26 +13,18 @@
 
 #include <asm/ecard.h>
 
-static ide_hwif_t *
-rapide_locate_hwif(void __iomem *base, void __iomem *ctrl, unsigned int sz, int irq)
+static void rapide_setup_ports(hw_regs_t *hw, void __iomem *base,
+			       void __iomem *ctrl, unsigned int sz, int irq)
 {
 	unsigned long port = (unsigned long)base;
-	ide_hwif_t *hwif = ide_find_port(port);
 	int i;
 
-	if (hwif == NULL)
-		goto out;
-
 	for (i = IDE_DATA_OFFSET; i <= IDE_STATUS_OFFSET; i++) {
-		hwif->io_ports[i] = port;
+		hw->io_ports[i] = port;
 		port += sz;
 	}
-	hwif->io_ports[IDE_CONTROL_OFFSET] = (unsigned long)ctrl;
-	hwif->irq = irq;
-	hwif->mmio = 1;
-	default_hwif_mmiops(hwif);
-out:
-	return hwif;
+	hw->io_ports[IDE_CONTROL_OFFSET] = (unsigned long)ctrl;
+	hw->irq = irq;
 }
 
 static int __devinit
@@ -42,6 +34,7 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 	void __iomem *base;
 	int ret;
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
+	hw_regs_t hw;
 
 	ret = ecard_request_resources(ec);
 	if (ret)
@@ -53,12 +46,19 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 		goto release;
 	}
 
-	hwif = rapide_locate_hwif(base, base + 0x818, 1 << 6, ec->irq);
+	hwif = ide_find_port((unsigned long)base);
 	if (hwif) {
-		hwif->chipset = ide_generic;
+		memset(&hw, 0, sizeof(hw));
+		rapide_setup_ports(&hw, base, base + 0x818, 1 << 6, ec->irq);
+		hw.chipset = ide_generic;
+		hw.dev = &ec->dev;
+
+		ide_init_port_hw(hwif, &hw);
+
+		hwif->mmio = 1;
+		default_hwif_mmiops(hwif);
+
 		hwif->hwif_data = base;
-		hwif->gendev.parent = &ec->dev;
-		hwif->noprobe = 0;
 
 		idx[0] = hwif->index;
 
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 8ef521f66f87..98bd45e8c175 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -675,6 +675,17 @@ void ide_setup_ports (	hw_regs_t *hw,
  */
 }
 
+void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
+{
+	memcpy(hwif->io_ports, hw->io_ports, sizeof(hwif->io_ports));
+	hwif->irq = hw->irq;
+	hwif->noprobe = 0;
+	hwif->chipset = hw->chipset;
+	hwif->gendev.parent = hw->dev;
+	hwif->ack_intr = hw->ack_intr;
+}
+EXPORT_SYMBOL_GPL(ide_init_port_hw);
+
 /**
  *	ide_register_hw		-	register IDE interface
  *	@hw: hardware registers
@@ -729,13 +740,9 @@ found:
 	}
 	if (hwif->present)
 		return -1;
-	memcpy(hwif->io_ports, hw->io_ports, sizeof(hwif->io_ports));
-	hwif->irq = hw->irq;
-	hwif->noprobe = 0;
+
+	ide_init_port_hw(hwif, hw);
 	hwif->quirkproc = quirkproc;
-	hwif->chipset = hw->chipset;
-	hwif->gendev.parent = hw->dev;
-	hwif->ack_intr = hw->ack_intr;
 
 	if (initializing == 0) {
 		u8 idx[4] = { index, 0xff, 0xff, 0xff };
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index 7bb79f53dac8..69a0fb0e564f 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -28,39 +28,27 @@ static struct {
 	int index;
 } hwif_prop;
 
-static ide_hwif_t *__devinit plat_ide_locate_hwif(void __iomem *base,
-	    void __iomem *ctrl, struct pata_platform_info *pdata, int irq,
-	    int mmio)
+static void __devinit plat_ide_setup_ports(hw_regs_t *hw,
+					   void __iomem *base,
+					   void __iomem *ctrl,
+					   struct pata_platform_info *pdata,
+					   int irq)
 {
 	unsigned long port = (unsigned long)base;
-	ide_hwif_t *hwif = ide_find_port(port);
 	int i;
 
-	if (hwif == NULL)
-		goto out;
-
-	hwif->io_ports[IDE_DATA_OFFSET] = port;
+	hw->io_ports[IDE_DATA_OFFSET] = port;
 
 	port += (1 << pdata->ioport_shift);
 	for (i = IDE_ERROR_OFFSET; i <= IDE_STATUS_OFFSET;
 	     i++, port += (1 << pdata->ioport_shift))
-		hwif->io_ports[i] = port;
-
-	hwif->io_ports[IDE_CONTROL_OFFSET] = (unsigned long)ctrl;
+		hw->io_ports[i] = port;
 
-	hwif->irq = irq;
+	hw->io_ports[IDE_CONTROL_OFFSET] = (unsigned long)ctrl;
 
-	hwif->chipset = ide_generic;
+	hw->irq = irq;
 
-	if (mmio) {
-		hwif->mmio = 1;
-		default_hwif_mmiops(hwif);
-	}
-
-	hwif_prop.hwif = hwif;
-	hwif_prop.index = hwif->index;
-out:
-	return hwif;
+	hw->chipset = ide_generic;
 }
 
 static int __devinit plat_ide_probe(struct platform_device *pdev)
@@ -71,6 +59,7 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	int ret = 0;
 	int mmio = 0;
+	hw_regs_t hw;
 
 	pdata = pdev->dev.platform_data;
 
@@ -106,15 +95,27 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 			res_alt->start, res_alt->end - res_alt->start + 1);
 	}
 
-	hwif = plat_ide_locate_hwif(hwif_prop.plat_ide_mapbase,
-	         hwif_prop.plat_ide_alt_mapbase, pdata, res_irq->start, mmio);
-
+	hwif = ide_find_port((unsigned long)hwif_prop.plat_ide_mapbase);
 	if (!hwif) {
 		ret = -ENODEV;
 		goto out;
 	}
-	hwif->gendev.parent = &pdev->dev;
-	hwif->noprobe = 0;
+
+	memset(&hw, 0, sizeof(hw));
+	plat_ide_setup_ports(&hw, hwif_prop.plat_ide_mapbase,
+			     hwif_prop.plat_ide_alt_mapbase,
+			     pdata, res_irq->start);
+	hw.dev = &pdev->dev;
+
+	ide_init_port_hw(hwif, &hw);
+
+	if (mmio) {
+		hwif->mmio = 1;
+		default_hwif_mmiops(hwif);
+	}
+
+	hwif_prop.hwif = hwif;
+	hwif_prop.index = hwif->index;
 
 	idx[0] = hwif->index;
 
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index b188efcd3551..9e0be7d54980 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -636,14 +636,13 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	/* Initialize the IO registers */
 	memset(&hw, 0, sizeof(hw));
 	sgiioc4_init_hwif_ports(&hw, cmd_base, ctl, irqport);
-	memcpy(hwif->io_ports, hw.io_ports, sizeof(hwif->io_ports));
-	hwif->noprobe = 0;
+	hw.irq = dev->irq;
+	hw.chipset = ide_pci;
+	hw.dev = &dev->dev;
+	ide_init_port_hw(hwif, &hw);
 
-	hwif->irq = dev->irq;
-	hwif->chipset = ide_pci;
 	hwif->pci_dev = dev;
 	hwif->channel = 0;	/* Single Channel chip */
-	hwif->gendev.parent = &dev->dev;/* setup proper ancestral information */
 
 	/* The IOC4 uses MMIO rather than Port IO. */
 	default_hwif_mmiops(hwif);
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 6a4b0d479897..36e4b9570746 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1012,12 +1012,11 @@ pmac_ide_do_resume(ide_hwif_t *hwif)
  * rare machines unfortunately, but it's better this way.
  */
 static int
-pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
+pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif, hw_regs_t *hw)
 {
 	struct device_node *np = pmif->node;
 	const int *bidp;
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 
 	pmif->cable_80 = 0;
 	pmif->broken_dma = pmif->broken_dma_warn = 0;
@@ -1103,11 +1102,9 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 	/* Tell common code _not_ to mess with resources */
 	hwif->mmio = 1;
 	hwif->hwif_data = pmif;
-	memset(&hw, 0, sizeof(hw));
-	pmac_ide_init_hwif_ports(&hw, pmif->regbase, 0, &hwif->irq);
-	memcpy(hwif->io_ports, hw.io_ports, sizeof(hwif->io_ports));
-	hwif->chipset = ide_pmac;
-	hwif->noprobe = !hwif->io_ports[IDE_DATA_OFFSET] || pmif->mediabay;
+	hw->chipset = ide_pmac;
+	ide_init_port_hw(hwif, hw);
+	hwif->noprobe = pmif->mediabay;
 	hwif->hold = pmif->mediabay;
 	hwif->cbl = pmif->cable_80 ? ATA_CBL_PATA80 : ATA_CBL_PATA40;
 	hwif->drives[0].unmask = 1;
@@ -1163,6 +1160,7 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 	ide_hwif_t *hwif;
 	pmac_ide_hwif_t *pmif;
 	int i, rc;
+	hw_regs_t hw;
 
 	i = 0;
 	while (i < MAX_HWIFS && (ide_hwifs[i].io_ports[IDE_DATA_OFFSET] != 0
@@ -1205,7 +1203,6 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 	regbase = (unsigned long) base;
 
 	hwif->pci_dev = mdev->bus->pdev;
-	hwif->gendev.parent = &mdev->ofdev.dev;
 
 	pmif->mdev = mdev;
 	pmif->node = mdev->ofdev.node;
@@ -1223,7 +1220,12 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 #endif /* CONFIG_BLK_DEV_IDEDMA_PMAC */
 	dev_set_drvdata(&mdev->ofdev.dev, hwif);
 
-	rc = pmac_ide_setup_device(pmif, hwif);
+	memset(&hw, 0, sizeof(hw));
+	pmac_ide_init_hwif_ports(&hw, pmif->regbase, 0, NULL);
+	hw.irq = irq;
+	hw.dev = &mdev->ofdev.dev;
+
+	rc = pmac_ide_setup_device(pmif, hwif, &hw);
 	if (rc != 0) {
 		/* The inteface is released to the common IDE layer */
 		dev_set_drvdata(&mdev->ofdev.dev, NULL);
@@ -1282,6 +1284,7 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 	void __iomem *base;
 	unsigned long rbase, rlen;
 	int i, rc;
+	hw_regs_t hw;
 
 	np = pci_device_to_OF_node(pdev);
 	if (np == NULL) {
@@ -1315,7 +1318,6 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	hwif->pci_dev = pdev;
-	hwif->gendev.parent = &pdev->dev;
 	pmif->mdev = NULL;
 	pmif->node = np;
 
@@ -1332,7 +1334,12 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	pci_set_drvdata(pdev, hwif);
 
-	rc = pmac_ide_setup_device(pmif, hwif);
+	memset(&hw, 0, sizeof(hw));
+	pmac_ide_init_hwif_ports(&hw, pmif->regbase, 0, NULL);
+	hw.irq = pdev->irq;
+	hw.dev = &pdev->dev;
+
+	rc = pmac_ide_setup_device(pmif, hwif, &hw);
 	if (rc != 0) {
 		/* The inteface is released to the common IDE layer */
 		pci_set_drvdata(pdev, NULL);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ce9b16f38c08..de94a526ef9e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -198,6 +198,7 @@ typedef struct hw_regs_s {
 } hw_regs_t;
 
 struct hwif_s * ide_find_port(unsigned long);
+void ide_init_port_hw(struct hwif_s *, hw_regs_t *);
 
 struct ide_drive_s;
 int ide_register_hw(hw_regs_t *, void (*)(struct ide_drive_s *), int,
-- 
cgit v1.2.3


From cbb010c180294a5242a7681555c28737d9dd26ab Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:06 +0100
Subject: ide: drop 'initializing' argument from ide_register_hw()

* Rename init_hwif_data() to ide_init_port_data() and export it.

* For all users of ide_register_hw() with 'initializing' argument set
  hwif->present and hwif->hold are always zero so convert these host
  drivers to use ide_find_port()+ide_init_port_data()+ide_init_port_hw()
  instead (also no need for init_hwif_default() call since the setup
  done by it gets over-ridden by ide_init_port_hw() call).

* Drop 'initializing' argument from ide_register_hw().

Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/bast-ide.c     |  2 +-
 drivers/ide/arm/ide_arm.c      |  8 +++++++-
 drivers/ide/cris/ide-cris.c    |  4 +++-
 drivers/ide/h8300/ide-h8300.c  | 11 +++++++----
 drivers/ide/ide-pnp.c          | 11 +++++++----
 drivers/ide/ide.c              | 33 +++++++++++----------------------
 drivers/ide/legacy/buddha.c    | 11 ++++++++---
 drivers/ide/legacy/falconide.c | 14 +++++++++-----
 drivers/ide/legacy/gayle.c     | 10 +++++++---
 drivers/ide/legacy/ide-cs.c    |  2 +-
 drivers/ide/legacy/macide.c    | 29 ++++++++++++++---------------
 drivers/ide/legacy/q40ide.c    | 10 ++++++----
 drivers/ide/pci/delkin_cb.c    |  2 +-
 drivers/macintosh/mediabay.c   |  3 ++-
 include/linux/ide.h            |  3 ++-
 15 files changed, 86 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/bast-ide.c b/drivers/ide/arm/bast-ide.c
index 48db6167bb90..45bf9c825f2b 100644
--- a/drivers/ide/arm/bast-ide.c
+++ b/drivers/ide/arm/bast-ide.c
@@ -45,7 +45,7 @@ bastide_register(unsigned int base, unsigned int aux, int irq,
 	hw.io_ports[IDE_CONTROL_OFFSET] = aux + (6 * 0x20);
 	hw.irq = irq;
 
-	ide_register_hw(&hw, NULL, 0, hwif);
+	ide_register_hw(&hw, NULL, hwif);
 
 	return 0;
 }
diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c
index 8957cbadf5c2..1a03a2a285ef 100644
--- a/drivers/ide/arm/ide_arm.c
+++ b/drivers/ide/arm/ide_arm.c
@@ -26,10 +26,16 @@
 
 void __init ide_arm_init(void)
 {
+	ide_hwif_t *hwif;
 	hw_regs_t hw;
 
 	memset(&hw, 0, sizeof(hw));
 	ide_std_init_ports(&hw, IDE_ARM_IO, IDE_ARM_IO + 0x206);
 	hw.irq = IDE_ARM_IRQ;
-	ide_register_hw(&hw, NULL, 1, NULL);
+
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif) {
+		ide_init_port_data(hwif, hwif->index);
+		ide_init_port_hw(hwif, &hw);
+	}
 }
diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c
index dcebc0299f5e..7e33e2b42e92 100644
--- a/drivers/ide/cris/ide-cris.c
+++ b/drivers/ide/cris/ide-cris.c
@@ -777,9 +777,11 @@ init_e100_ide (void)
 		                ide_offsets,
 		                0, 0, cris_ide_ack_intr,
 		                ide_default_irq(0));
-		ide_register_hw(&hw, NULL, 1, &hwif);
+		hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
 		if (hwif == NULL)
 			continue;
+		ide_init_port_data(hwif, hwif->index);
+		ide_init_port_hw(hwif, &hw);
 		hwif->mmio = 1;
 		hwif->chipset = ide_etrax100;
 		hwif->set_pio_mode = &cris_set_pio_mode;
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 4a49b5c59acb..57d0d4ce8588 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -88,7 +88,7 @@ void __init h8300_ide_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
-	int idx;
+	int index;
 
 	if (!request_region(CONFIG_H8300_IDE_BASE, H8300_IDE_GAP*8, "ide-h8300"))
 		goto out_busy;
@@ -100,14 +100,17 @@ void __init h8300_ide_init(void)
 	hw_setup(&hw);
 
 	/* register if */
-	idx = ide_register_hw(&hw, NULL, 1, &hwif);
-	if (idx == -1) {
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif == NULL) {
 		printk(KERN_ERR "ide-h8300: IDE I/F register failed\n");
 		return;
 	}
 
+	index = hwif->index;
+	ide_init_port_data(hwif, index);
+	ide_init_port_hw(hwif, &hw);
 	hwif_setup(hwif);
-	printk(KERN_INFO "ide%d: H8/300 generic IDE interface\n", idx);
+	printk(KERN_INFO "ide%d: H8/300 generic IDE interface\n", index);
 	return;
 
 out_busy:
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index e245521af7b5..664bc489c550 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -31,7 +31,6 @@ static int idepnp_probe(struct pnp_dev * dev, const struct pnp_device_id *dev_id
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
-	int index;
 
 	if (!(pnp_port_valid(dev, 0) && pnp_port_valid(dev, 1) && pnp_irq_valid(dev, 0)))
 		return -1;
@@ -41,10 +40,14 @@ static int idepnp_probe(struct pnp_dev * dev, const struct pnp_device_id *dev_id
 				pnp_port_start(dev, 1));
 	hw.irq = pnp_irq(dev, 0);
 
-	index = ide_register_hw(&hw, NULL, 1, &hwif);
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif) {
+		u8 index = hwif->index;
+
+		ide_init_port_data(hwif, index);
+		ide_init_port_hw(hwif, &hw);
 
-	if (index != -1) {
-	    	printk(KERN_INFO "ide%d: generic PnP IDE interface\n", index);
+		printk(KERN_INFO "ide%d: generic PnP IDE interface\n", index);
 		pnp_set_drvdata(dev,hwif);
 		return 0;
 	}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 98bd45e8c175..3ec220b64d0b 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -116,7 +116,7 @@ EXPORT_SYMBOL(ide_hwifs);
 /*
  * Do not even *think* about calling this!
  */
-static void init_hwif_data(ide_hwif_t *hwif, unsigned int index)
+void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
 {
 	unsigned int unit;
 
@@ -159,6 +159,7 @@ static void init_hwif_data(ide_hwif_t *hwif, unsigned int index)
 		init_completion(&drive->gendev_rel_comp);
 	}
 }
+EXPORT_SYMBOL_GPL(ide_init_port_data);
 
 static void init_hwif_default(ide_hwif_t *hwif, unsigned int index)
 {
@@ -210,7 +211,7 @@ static void __init init_ide_data (void)
 	/* Initialise all interface structures */
 	for (index = 0; index < MAX_HWIFS; ++index) {
 		hwif = &ide_hwifs[index];
-		init_hwif_data(hwif, index);
+		ide_init_port_data(hwif, index);
 		init_hwif_default(hwif, index);
 #if !defined(CONFIG_PPC32) || !defined(CONFIG_PCI)
 		hwif->irq =
@@ -609,7 +610,7 @@ void ide_unregister(unsigned int index)
 	tmp_hwif = *hwif;
 
 	/* restore hwif data to pristine status */
-	init_hwif_data(hwif, index);
+	ide_init_port_data(hwif, index);
 	init_hwif_default(hwif, index);
 
 	ide_hwif_restore(hwif, &tmp_hwif);
@@ -690,29 +691,19 @@ EXPORT_SYMBOL_GPL(ide_init_port_hw);
  *	ide_register_hw		-	register IDE interface
  *	@hw: hardware registers
  *	@quirkproc: quirkproc function
- *	@initializing: set while initializing built-in drivers
  *	@hwifp: pointer to returned hwif
  *
  *	Register an IDE interface, specifying exactly the registers etc.
- *	Set init=1 iff calling before probes have taken place.
  *
  *	Returns -1 on error.
  */
 
 int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
-		    int initializing, ide_hwif_t **hwifp)
+		    ide_hwif_t **hwifp)
 {
 	int index, retry = 1;
 	ide_hwif_t *hwif;
-
-	if (initializing) {
-		hwif = ide_find_port(hw->io_ports[IDE_DATA_OFFSET]);
-		if (hwif) {
-			index = hwif->index;
-			goto found;
-		}
-		return -1;
-	}
+	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	do {
 		for (index = 0; index < MAX_HWIFS; ++index) {
@@ -735,7 +726,7 @@ found:
 	if (hwif->present)
 		ide_unregister(index);
 	else if (!hwif->hold) {
-		init_hwif_data(hwif, index);
+		ide_init_port_data(hwif, index);
 		init_hwif_default(hwif, index);
 	}
 	if (hwif->present)
@@ -744,16 +735,14 @@ found:
 	ide_init_port_hw(hwif, hw);
 	hwif->quirkproc = quirkproc;
 
-	if (initializing == 0) {
-		u8 idx[4] = { index, 0xff, 0xff, 0xff };
+	idx[0] = index;
 
-		ide_device_add(idx);
-	}
+	ide_device_add(idx);
 
 	if (hwifp)
 		*hwifp = hwif;
 
-	return (initializing || hwif->present) ? index : -1;
+	return hwif->present ? index : -1;
 }
 
 EXPORT_SYMBOL(ide_register_hw);
@@ -1076,7 +1065,7 @@ int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device
 			ide_init_hwif_ports(&hw, (unsigned long) args[0],
 					    (unsigned long) args[1], NULL);
 			hw.irq = args[2];
-			if (ide_register_hw(&hw, NULL, 0, NULL) == -1)
+			if (ide_register_hw(&hw, NULL, NULL) == -1)
 				return -EIO;
 			return 0;
 		}
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index 4a0be251a05f..8b9cb39c961e 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -147,7 +147,7 @@ void __init buddha_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
-	int i, index;
+	int i;
 
 	struct zorro_dev *z = NULL;
 	u_long buddha_board = 0;
@@ -213,8 +213,13 @@ fail_base2:
 						IRQ_AMIGA_PORTS);
 			}	
 
-			index = ide_register_hw(&hw, NULL, 1, &hwif);
-			if (index != -1) {
+			hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+			if (hwif) {
+				u8 index = hwif->index;
+
+				ide_init_port_data(hwif, index);
+				ide_init_port_hw(hwif, &hw);
+
 				hwif->mmio = 1;
 				printk("ide%d: ", index);
 				switch(type) {
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 7d7936f1b900..b861cfe25900 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -66,15 +66,19 @@ void __init falconide_init(void)
 {
     if (MACH_IS_ATARI && ATARIHW_PRESENT(IDE)) {
 	hw_regs_t hw;
-	int index;
 
 	ide_setup_ports(&hw, ATA_HD_BASE, falconide_offsets,
 			0, 0, NULL,
 //			falconide_iops,
 			IRQ_MFP_IDE);
-	index = ide_register_hw(&hw, NULL, 1, NULL);
 
-	if (index != -1)
-	    printk("ide%d: Falcon IDE interface\n", index);
-    }
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif) {
+		u8 index = hwif->index;
+
+		ide_init_port_data(hwif, index);
+		ide_init_port_hw(hwif, &hw);
+
+		printk("ide%d: Falcon IDE interface\n", index);
+	}
 }
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index 53331ee1e957..705d0b8a3f5d 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -133,7 +133,6 @@ found:
 	ide_ack_intr_t *ack_intr;
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
-	int index;
 	unsigned long phys_base, res_start, res_n;
 
 	if (a4000) {
@@ -165,8 +164,13 @@ found:
 //			&gayle_iops,
 			IRQ_AMIGA_PORTS);
 
-	index = ide_register_hw(&hw, NULL, 1, &hwif);
-	if (index != -1) {
+	hwif = ide_find_port(base);
+	if (hwif) {
+	    u8 index = hwif->index;
+
+	    ide_init_port_data(hwif, index);
+	    ide_init_port_hw(hwif, &hw);
+
 	    hwif->mmio = 1;
 	    switch (i) {
 		case 0:
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 03715c058664..f4ea15b32969 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -153,7 +153,7 @@ static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq
     hw.irq = irq;
     hw.chipset = ide_pci;
     hw.dev = &handle->dev;
-    return ide_register_hw(&hw, &ide_undecoded_slave, 0, NULL);
+    return ide_register_hw(&hw, &ide_undecoded_slave, NULL);
 }
 
 /*======================================================================
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index 5c6aa77c2370..1840fede5216 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -85,7 +85,6 @@ void __init macide_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
-	int index = -1;
 
 	switch (macintosh_config->ide_type) {
 	case MAC_IDE_QUADRA:
@@ -93,40 +92,40 @@ void __init macide_init(void)
 				0, 0, macide_ack_intr,
 //				quadra_ide_iops,
 				IRQ_NUBUS_F);
-		index = ide_register_hw(&hw, NULL, 1, &hwif);
 		break;
 	case MAC_IDE_PB:
 		ide_setup_ports(&hw, IDE_BASE, macide_offsets,
 				0, 0, macide_ack_intr,
 //				macide_pb_iops,
 				IRQ_NUBUS_C);
-		index = ide_register_hw(&hw, NULL, 1, &hwif);
 		break;
 	case MAC_IDE_BABOON:
 		ide_setup_ports(&hw, BABOON_BASE, macide_offsets,
 				0, 0, NULL,
 //				macide_baboon_iops,
 				IRQ_BABOON_1);
-		index = ide_register_hw(&hw, NULL, 1, &hwif);
-		if (index == -1) break;
-		if (macintosh_config->ident == MAC_MODEL_PB190) {
+		break;
+	default:
+		return;
+	}
 
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif) {
+		u8 index = hwif->index;
+
+		ide_init_port_data(hwif, index);
+		ide_init_port_hw(hwif, &hw);
+
+		if (macintosh_config->ide_type == MAC_IDE_BABOON &&
+		    macintosh_config->ident == MAC_MODEL_PB190) {
 			/* Fix breakage in ide-disk.c: drive capacity	*/
 			/* is not initialized for drives without a 	*/
 			/* hardware ID, and we can't get that without	*/
 			/* probing the drive which freezes a 190.	*/
-
-			ide_drive_t *drive = &ide_hwifs[index].drives[0];
+			ide_drive_t *drive = &hwif->drives[0];
 			drive->capacity64 = drive->cyl*drive->head*drive->sect;
-
 		}
-		break;
-
-	default:
-	    return;
-	}
 
-        if (index != -1) {
 		hwif->mmio = 1;
 		if (macintosh_config->ide_type == MAC_IDE_QUADRA)
 			printk(KERN_INFO "ide%d: Macintosh Quadra IDE interface\n", index);
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 6ea46a6723e2..31e54ffdfee4 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -115,7 +115,6 @@ void __init q40ide_init(void)
 {
     int i;
     ide_hwif_t *hwif;
-    int index;
     const char *name;
 
     if (!MACH_IS_Q40)
@@ -141,10 +140,13 @@ void __init q40ide_init(void)
 			0, NULL,
 //			m68kide_iops,
 			q40ide_default_irq(pcide_bases[i]));
-	index = ide_register_hw(&hw, NULL, 1, &hwif);
-	// **FIXME**
-	if (index != -1)
+
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	if (hwif) {
+		ide_init_port_data(hwif, hwif->index);
+		ide_init_port_hw(hwif, &hw);
 		hwif->mmio = 1;
+	}
     }
 }
 
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 83829081640a..26aa492071bb 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -80,7 +80,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 	hw.irq = dev->irq;
 	hw.chipset = ide_pci;		/* this enables IRQ sharing */
 
-	rc = ide_register_hw(&hw, &ide_undecoded_slave, 0, &hwif);
+	rc = ide_register_hw(&hw, &ide_undecoded_slave, &hwif);
 	if (rc < 0) {
 		printk(KERN_ERR "delkin_cb: ide_register_hw failed (%d)\n", rc);
 		pci_disable_device(dev);
diff --git a/drivers/macintosh/mediabay.c b/drivers/macintosh/mediabay.c
index 48d647abea46..eaba4a9b231e 100644
--- a/drivers/macintosh/mediabay.c
+++ b/drivers/macintosh/mediabay.c
@@ -563,7 +563,8 @@ static void media_bay_step(int i)
 				ide_init_hwif_ports(&hw, (unsigned long) bay->cd_base, (unsigned long) 0, NULL);
 				hw.irq = bay->cd_irq;
 				hw.chipset = ide_pmac;
-				bay->cd_index = ide_register_hw(&hw, NULL, 0, NULL);
+				bay->cd_index =
+					ide_register_hw(&hw, NULL, NULL);
 				pmu_resume();
 			}
 			if (bay->cd_index == -1) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index de94a526ef9e..9c037a0f2af2 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -198,10 +198,11 @@ typedef struct hw_regs_s {
 } hw_regs_t;
 
 struct hwif_s * ide_find_port(unsigned long);
+void ide_init_port_data(struct hwif_s *, unsigned int);
 void ide_init_port_hw(struct hwif_s *, hw_regs_t *);
 
 struct ide_drive_s;
-int ide_register_hw(hw_regs_t *, void (*)(struct ide_drive_s *), int,
+int ide_register_hw(hw_regs_t *, void (*)(struct ide_drive_s *),
 		    struct hwif_s **);
 
 void ide_setup_ports(	hw_regs_t *hw,
-- 
cgit v1.2.3


From ade2daf9c6e57845fe83a24e0a9fa1c03c6e91b1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:07 +0100
Subject: ide: make remaining built-in only IDE host drivers modular (take 2)

* Make remaining built-in only IDE host drivers modular, add ide-scan-pci.c
  file for probing PCI host drivers registered with IDE core (special case
  for built-in IDE and CONFIG_IDEPCI_PCIBUS_ORDER=y) and then take care of
  the ordering in which all IDE host drivers are probed when IDE is built-in
  during link time.

* Move probing of gayle, falconide, macide, q40ide and buddha (m68k arch
  specific) host drivers, before PCI ones (no PCI on m68k), ide-cris (cris
  arch specific), cmd640 (x86 arch specific) and pmac (ppc arch specific).

* Move probing of ide-cris (cris arch specific) host driver before cmd640
  (x86 arch specific).

* Move probing of mpc8xx (ppc specific) host driver before ide-pnp (depends
  on ISA and none of ppc platform that use mpc8xx supports ISA) and ide-h8300
  (h8300 arch specific).

* Add "probe_vlb" kernel parameter to cmd640 host driver and update
  Documentation/ide.txt accordingly.

* Make IDE_ARM config option visible so it can also be disabled if needed.

* Remove bogus comment from ide.c while at it.

v2:
* Fix two issues spotted by Sergei:
  - replace ENOMEM error value by ENOENT in ide-h8300 host driver
  - fix MODULE_PARM_DESC() in cmd640 host driver

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 Documentation/ide.txt          |   9 +--
 drivers/ide/Kconfig            |  26 ++++----
 drivers/ide/Makefile           |  58 ++++++++++--------
 drivers/ide/arm/Makefile       |   4 ++
 drivers/ide/arm/ide_arm.c      |   6 +-
 drivers/ide/cris/Makefile      |   2 +-
 drivers/ide/cris/ide-cris.c    |   7 ++-
 drivers/ide/h8300/Makefile     |   2 +
 drivers/ide/h8300/ide-h8300.c  |  10 +++-
 drivers/ide/ide-pnp.c          |   9 ++-
 drivers/ide/ide-scan-pci.c     |  11 ++++
 drivers/ide/ide.c              | 131 +----------------------------------------
 drivers/ide/legacy/Makefile    |  19 ++++--
 drivers/ide/legacy/ali14xx.c   |   5 +-
 drivers/ide/legacy/buddha.c    |   6 +-
 drivers/ide/legacy/dtc2278.c   |   5 +-
 drivers/ide/legacy/falconide.c |   7 ++-
 drivers/ide/legacy/gayle.c     |  10 +++-
 drivers/ide/legacy/ht6560b.c   |   5 +-
 drivers/ide/legacy/macide.c    |   8 ++-
 drivers/ide/legacy/q40ide.c    |   7 ++-
 drivers/ide/legacy/qd65xx.c    |   5 +-
 drivers/ide/legacy/umc8672.c   |   5 +-
 drivers/ide/pci/Makefile       |   4 ++
 drivers/ide/pci/cmd640.c       |   8 ++-
 drivers/ide/ppc/Makefile       |   3 +
 drivers/ide/ppc/mpc8xx.c       |   6 +-
 drivers/ide/ppc/pmac.c         |   2 +
 drivers/ide/setup-pci.c        |   9 +--
 include/linux/ide.h            |   3 +-
 30 files changed, 171 insertions(+), 221 deletions(-)
 create mode 100644 drivers/ide/h8300/Makefile
 create mode 100644 drivers/ide/ide-scan-pci.c
 create mode 100644 drivers/ide/ppc/Makefile

(limited to 'include')

diff --git a/Documentation/ide.txt b/Documentation/ide.txt
index 1d50f23a5cab..b29ccb43d6cc 100644
--- a/Documentation/ide.txt
+++ b/Documentation/ide.txt
@@ -30,7 +30,7 @@
 ***
 ***  The CMD640 is also used on some Vesa Local Bus (VLB) cards, and is *NOT*
 ***  automatically detected by Linux.  For safe, reliable operation with such
-***  interfaces, one *MUST* use the "ide0=cmd640_vlb" kernel option.
+***  interfaces, one *MUST* use the "cmd640.probe_vlb" kernel option.
 ***
 ***  Use of the "serialize" option is no longer necessary.
 
@@ -292,9 +292,6 @@ The following are valid ONLY on ide0, which usually corresponds
 to the first ATA interface found on the particular host, and the defaults for
 the base,ctl ports must not be altered.
 
- "ide0=cmd640_vlb"	: *REQUIRED* for VLB cards with the CMD640 chip
-			  (not for PCI -- automatically detected)
-
  "ide=doubler"		: probe/support IDE doublers on Amiga
 
 There may be more options than shown -- use the source, Luke!
@@ -310,6 +307,10 @@ i.e. to enable probing for ALI M14xx chipsets (ali14xx host driver) use:
 * "probe" module parameter when ali14xx driver is compiled as module
   ("modprobe ali14xx probe")
 
+Also for legacy CMD640 host driver (cmd640) you need to use "probe_vlb"
+kernel paremeter to enable probing for VLB version of the chipset (PCI ones
+are detected automatically).
+
 ================================================================================
 
 IDE ATAPI streaming tape driver
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 7c419e87a4a4..e92128a87f26 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -325,7 +325,7 @@ config BLK_DEV_PLATFORM
 	  If unsure, say N.
 
 config BLK_DEV_CMD640
-	bool "CMD640 chipset bugfix/support"
+	tristate "CMD640 chipset bugfix/support"
 	depends on X86
 	---help---
 	  The CMD-Technologies CMD640 IDE chip is used on many common 486 and
@@ -359,7 +359,7 @@ config BLK_DEV_CMD640_ENHANCED
 	  Otherwise say N.
 
 config BLK_DEV_IDEPNP
-	bool "PNP EIDE support"
+	tristate "PNP EIDE support"
 	depends on PNP
 	help
 	  If you have a PnP (Plug and Play) compatible EIDE card and
@@ -788,7 +788,7 @@ config BLK_DEV_CELLEB
 endif
 
 config BLK_DEV_IDE_PMAC
-	bool "Builtin PowerMac IDE support"
+	tristate "Builtin PowerMac IDE support"
 	depends on PPC_PMAC && IDE=y && BLK_DEV_IDE=y
 	help
 	  This driver provides support for the built-in IDE controller on
@@ -842,7 +842,9 @@ config BLK_DEV_IDE_AU1XXX_SEQTS_PER_RQ
        depends on BLK_DEV_IDE_AU1XXX
 
 config IDE_ARM
-	def_bool ARM && (ARCH_CLPS7500 || ARCH_RPC || ARCH_SHARK)
+	tristate "ARM IDE support"
+	depends on ARM && (ARCH_CLPS7500 || ARCH_RPC || ARCH_SHARK)
+	default y
 
 config BLK_DEV_IDE_ICSIDE
 	tristate "ICS IDE interface support"
@@ -874,7 +876,7 @@ config BLK_DEV_IDE_BAST
 	  Simtec BAST or the Thorcom VR1000
 
 config ETRAX_IDE
-	bool "ETRAX IDE support"
+	tristate "ETRAX IDE support"
 	depends on CRIS && BROKEN
 	select BLK_DEV_IDEDMA
 	help
@@ -908,14 +910,14 @@ config ETRAX_IDE_G27_RESET
 endchoice
 
 config IDE_H8300
-	bool "H8300 IDE support"
+	tristate "H8300 IDE support"
 	depends on H8300
 	default y
 	help
 	  Enables the H8300 IDE driver.
 
 config BLK_DEV_GAYLE
-	bool "Amiga Gayle IDE interface support"
+	tristate "Amiga Gayle IDE interface support"
 	depends on AMIGA
 	help
 	  This is the IDE driver for the Amiga Gayle IDE interface. It supports
@@ -946,7 +948,7 @@ config BLK_DEV_IDEDOUBLER
 	  runtime using the "ide=doubler" kernel boot parameter.
 
 config BLK_DEV_BUDDHA
-	bool "Buddha/Catweasel/X-Surf IDE interface support (EXPERIMENTAL)"
+	tristate "Buddha/Catweasel/X-Surf IDE interface support (EXPERIMENTAL)"
 	depends on ZORRO && EXPERIMENTAL
 	help
 	  This is the IDE driver for the IDE interfaces on the Buddha, 
@@ -958,7 +960,7 @@ config BLK_DEV_BUDDHA
 	  to one of its IDE interfaces.
 
 config BLK_DEV_FALCON_IDE
-	bool "Falcon IDE interface support"
+	tristate "Falcon IDE interface support"
 	depends on ATARI
 	help
 	  This is the IDE driver for the builtin IDE interface on the Atari
@@ -967,7 +969,7 @@ config BLK_DEV_FALCON_IDE
 	  interface.
 
 config BLK_DEV_MAC_IDE
-	bool "Macintosh Quadra/Powerbook IDE interface support"
+	tristate "Macintosh Quadra/Powerbook IDE interface support"
 	depends on MAC
 	help
 	  This is the IDE driver for the builtin IDE interface on some m68k
@@ -980,7 +982,7 @@ config BLK_DEV_MAC_IDE
 	  builtin IDE interface.
 
 config BLK_DEV_Q40IDE
-	bool "Q40/Q60 IDE interface support"
+	tristate "Q40/Q60 IDE interface support"
 	depends on Q40
 	help
 	  Enable the on-board IDE controller in the Q40/Q60.  This should
@@ -988,7 +990,7 @@ config BLK_DEV_Q40IDE
 	  drive subsystem through an expansion card.
 
 config BLK_DEV_MPC8xx_IDE
-	bool "MPC8xx IDE support"
+	tristate "MPC8xx IDE support"
 	depends on 8xx && (LWMON || IVMS8 || IVML24 || TQM8xxL) && IDE=y && BLK_DEV_IDE=y && !PPC_MERGE
 	help
 	  This option provides support for IDE on Motorola MPC8xx Systems.
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index b181fc672057..0d2da89d15cf 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -7,41 +7,37 @@
 # Note : at this point, these files are compiled on all systems.
 # In the future, some of these should be built conditionally.
 #
-# First come modules that register themselves with the core
+# link order is important here
 
 EXTRA_CFLAGS				+= -Idrivers/ide
 
-obj-$(CONFIG_BLK_DEV_IDE)		+= pci/
-
 ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o
 
-ide-core-$(CONFIG_BLK_DEV_CMD640)	+= pci/cmd640.o
-
-# Core IDE code - must come before legacy
+# core IDE code
 ide-core-$(CONFIG_BLK_DEV_IDEPCI)	+= setup-pci.o
 ide-core-$(CONFIG_BLK_DEV_IDEDMA)	+= ide-dma.o
 ide-core-$(CONFIG_IDE_PROC_FS)		+= ide-proc.o
-ide-core-$(CONFIG_BLK_DEV_IDEPNP)	+= ide-pnp.o
 ide-core-$(CONFIG_BLK_DEV_IDEACPI)	+= ide-acpi.o
 
-# built-in only drivers from arm/
-ide-core-$(CONFIG_IDE_ARM)		+= arm/ide_arm.o
+obj-$(CONFIG_BLK_DEV_IDE)		+= ide-core.o
 
-# built-in only drivers from legacy/
-ide-core-$(CONFIG_BLK_DEV_BUDDHA)	+= legacy/buddha.o
-ide-core-$(CONFIG_BLK_DEV_FALCON_IDE)	+= legacy/falconide.o
-ide-core-$(CONFIG_BLK_DEV_GAYLE)	+= legacy/gayle.o
-ide-core-$(CONFIG_BLK_DEV_MAC_IDE)	+= legacy/macide.o
-ide-core-$(CONFIG_BLK_DEV_Q40IDE)	+= legacy/q40ide.o
+ifeq ($(CONFIG_IDE_ARM), y)
+	ide-arm-core-y += arm/ide_arm.o
+	obj-y += ide-arm-core.o
+endif
 
-# built-in only drivers from ppc/
-ide-core-$(CONFIG_BLK_DEV_MPC8xx_IDE)	+= ppc/mpc8xx.o
-ide-core-$(CONFIG_BLK_DEV_IDE_PMAC)	+= ppc/pmac.o
+obj-$(CONFIG_BLK_DEV_IDE)		+= legacy/ pci/
 
-# built-in only drivers from h8300/
-ide-core-$(CONFIG_IDE_H8300)		+= h8300/ide-h8300.o
+obj-$(CONFIG_IDEPCI_PCIBUS_ORDER)	+= ide-scan-pci.o
 
-obj-$(CONFIG_BLK_DEV_IDE)		+= ide-core.o
+ifeq ($(CONFIG_BLK_DEV_CMD640), y)
+	cmd640-core-y += pci/cmd640.o
+	obj-y += cmd640-core.o
+endif
+
+obj-$(CONFIG_BLK_DEV_IDE)		+= cris/ ppc/
+obj-$(CONFIG_BLK_DEV_IDEPNP)		+= ide-pnp.o
+obj-$(CONFIG_IDE_H8300)			+= h8300/
 obj-$(CONFIG_IDE_GENERIC)		+= ide-generic.o
 
 obj-$(CONFIG_BLK_DEV_IDEDISK)		+= ide-disk.o
@@ -49,6 +45,20 @@ obj-$(CONFIG_BLK_DEV_IDECD)		+= ide-cd.o
 obj-$(CONFIG_BLK_DEV_IDETAPE)		+= ide-tape.o
 obj-$(CONFIG_BLK_DEV_IDEFLOPPY)		+= ide-floppy.o
 
-obj-$(CONFIG_BLK_DEV_IDE)		+= legacy/ arm/ mips/
-obj-$(CONFIG_BLK_DEV_HD)		+= legacy/
-obj-$(CONFIG_ETRAX_IDE)		+= cris/
+ifeq ($(CONFIG_BLK_DEV_IDECS), y)
+	ide-cs-core-y += legacy/ide-cs.o
+	obj-y += ide-cs-core.o
+endif
+
+ifeq ($(CONFIG_BLK_DEV_PLATFORM), y)
+	ide-platform-core-y += legacy/ide_platform.o
+	obj-y += ide-platform-core.o
+endif
+
+obj-$(CONFIG_BLK_DEV_IDE)		+= arm/ mips/
+
+# old hd driver must be last
+ifeq ($(CONFIG_BLK_DEV_HD), y)
+	hd-core-y += legacy/hd.o
+	obj-y += hd-core.o
+endif
diff --git a/drivers/ide/arm/Makefile b/drivers/ide/arm/Makefile
index 6a78f0755f26..5f63ad216862 100644
--- a/drivers/ide/arm/Makefile
+++ b/drivers/ide/arm/Makefile
@@ -3,4 +3,8 @@ obj-$(CONFIG_BLK_DEV_IDE_ICSIDE)	+= icside.o
 obj-$(CONFIG_BLK_DEV_IDE_RAPIDE)	+= rapide.o
 obj-$(CONFIG_BLK_DEV_IDE_BAST)		+= bast-ide.o
 
+ifeq ($(CONFIG_IDE_ARM), m)
+	obj-m += ide_arm.o
+endif
+
 EXTRA_CFLAGS	:= -Idrivers/ide
diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c
index a1b5ddab6a48..60f2497542c0 100644
--- a/drivers/ide/arm/ide_arm.c
+++ b/drivers/ide/arm/ide_arm.c
@@ -24,7 +24,7 @@
 # define IDE_ARM_IRQ	IRQ_HARDDISK
 #endif
 
-void __init ide_arm_init(void)
+static int __init ide_arm_init(void)
 {
 	ide_hwif_t *hwif;
 	hw_regs_t hw;
@@ -41,4 +41,8 @@ void __init ide_arm_init(void)
 
 		ide_device_add(idx);
 	}
+
+	return 0;
 }
+
+module_init(ide_arm_init);
diff --git a/drivers/ide/cris/Makefile b/drivers/ide/cris/Makefile
index 6176e8d6b2e6..20b95960531f 100644
--- a/drivers/ide/cris/Makefile
+++ b/drivers/ide/cris/Makefile
@@ -1,3 +1,3 @@
 EXTRA_CFLAGS				+= -Idrivers/ide
 
-obj-y					+= ide-cris.o
+obj-$(CONFIG_IDE_ETRAX)			+= ide-cris.o
diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c
index 924536297036..8c3294c4d23e 100644
--- a/drivers/ide/cris/ide-cris.c
+++ b/drivers/ide/cris/ide-cris.c
@@ -754,8 +754,7 @@ static void cris_set_dma_mode(ide_drive_t *drive, const u8 speed)
 		cris_ide_set_speed(TYPE_DMA, 0, strobe, hold);
 }
 
-void __init
-init_e100_ide (void)
+static int __init init_e100_ide(void)
 {
 	hw_regs_t hw;
 	int ide_offsets[IDE_NR_PORTS], h, i;
@@ -823,6 +822,8 @@ init_e100_ide (void)
 	cris_ide_set_speed(TYPE_UDMA, ATA_UDMA2_CYC, ATA_UDMA2_DVS, 0);
 
 	ide_device_add(idx);
+
+	return 0;
 }
 
 static cris_dma_descr_type mydescr __attribute__ ((__aligned__(16)));
@@ -1056,3 +1057,5 @@ static void cris_dma_start(ide_drive_t *drive)
 		LED_DISK_READ(1);
 	}
 }
+
+module_init(init_e100_ide);
diff --git a/drivers/ide/h8300/Makefile b/drivers/ide/h8300/Makefile
new file mode 100644
index 000000000000..5eba16f423f4
--- /dev/null
+++ b/drivers/ide/h8300/Makefile
@@ -0,0 +1,2 @@
+
+obj-$(CONFIG_IDE_H8300)			+= ide-h8300.o
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 9fa78e98d1ba..4f6d0191cf6c 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -84,7 +84,7 @@ static inline void hwif_setup(ide_hwif_t *hwif)
 	hwif->INSL  = NULL;
 }
 
-void __init h8300_ide_init(void)
+static int __init h8300_ide_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
@@ -104,7 +104,7 @@ void __init h8300_ide_init(void)
 	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
 	if (hwif == NULL) {
 		printk(KERN_ERR "ide-h8300: IDE I/F register failed\n");
-		return;
+		return -ENOENT;
 	}
 
 	index = hwif->index;
@@ -117,8 +117,12 @@ void __init h8300_ide_init(void)
 
 	ide_device_add(idx);
 
-	return;
+	return 0;
 
 out_busy:
 	printk(KERN_ERR "ide-h8300: IDE I/F resource already used.\n");
+
+	return -EBUSY;
 }
+
+module_init(h8300_ide_init);
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index 802efd4d9760..cbbb0f75be92 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -75,12 +75,15 @@ static struct pnp_driver idepnp_driver = {
 	.remove		= idepnp_remove,
 };
 
-void __init pnpide_init(void)
+static int __init pnpide_init(void)
 {
-	pnp_register_driver(&idepnp_driver);
+	return pnp_register_driver(&idepnp_driver);
 }
 
-void __exit pnpide_exit(void)
+static void __exit pnpide_exit(void)
 {
 	pnp_unregister_driver(&idepnp_driver);
 }
+
+module_init(pnpide_init);
+module_exit(pnpide_exit);
diff --git a/drivers/ide/ide-scan-pci.c b/drivers/ide/ide-scan-pci.c
new file mode 100644
index 000000000000..23015d89e735
--- /dev/null
+++ b/drivers/ide/ide-scan-pci.c
@@ -0,0 +1,11 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/ide.h>
+
+static int __init ide_scan_pci(void)
+{
+	return ide_scan_pcibus();
+}
+
+module_init(ide_scan_pci);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 6f99f5c90062..5f3e53ec5838 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -95,7 +95,7 @@ DEFINE_MUTEX(ide_cfg_mtx);
  __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
 
 #ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-static int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */
+int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */
 #endif
 
 int noautodma = 0;
@@ -178,8 +178,6 @@ static void init_hwif_default(ide_hwif_t *hwif, unsigned int index)
 #endif
 }
 
-extern void ide_arm_init(void);
-
 /*
  * init_ide_data() sets reasonable default values into all fields
  * of all instances of the hwifs and drives, but only on the first call.
@@ -1223,26 +1221,12 @@ static int __init match_parm (char *s, const char *keywords[], int vals[], int m
 	return 0;	/* zero = nothing matched */
 }
 
-#ifdef CONFIG_BLK_DEV_ALI14XX
 extern int probe_ali14xx;
-extern int ali14xx_init(void);
-#endif
-#ifdef CONFIG_BLK_DEV_UMC8672
 extern int probe_umc8672;
-extern int umc8672_init(void);
-#endif
-#ifdef CONFIG_BLK_DEV_DTC2278
 extern int probe_dtc2278;
-extern int dtc2278_init(void);
-#endif
-#ifdef CONFIG_BLK_DEV_HT6560B
 extern int probe_ht6560b;
-extern int ht6560b_init(void);
-#endif
-#ifdef CONFIG_BLK_DEV_QD65XX
 extern int probe_qd65xx;
-extern int qd65xx_init(void);
-#endif
+extern int cmd640_vlb;
 
 static int __initdata is_chipset_set[MAX_HWIFS];
 
@@ -1458,11 +1442,8 @@ static int __init ide_setup(char *s)
 #endif
 #ifdef CONFIG_BLK_DEV_CMD640
 			case -14: /* "cmd640_vlb" */
-			{
-				extern int cmd640_vlb; /* flag for cmd640.c */
 				cmd640_vlb = 1;
 				goto done;
-			}
 #endif
 #ifdef CONFIG_BLK_DEV_HT6560B
 			case -13: /* "ht6560b" */
@@ -1552,83 +1533,6 @@ done:
 	return 1;
 }
 
-extern void __init pnpide_init(void);
-extern void __exit pnpide_exit(void);
-extern void __init h8300_ide_init(void);
-extern void __init mpc8xx_ide_probe(void);
-
-/*
- * probe_for_hwifs() finds/initializes "known" IDE interfaces
- */
-static void __init probe_for_hwifs (void)
-{
-#ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-	ide_scan_pcibus(ide_scan_direction);
-#endif
-
-#ifdef CONFIG_ETRAX_IDE
-	{
-		extern void init_e100_ide(void);
-		init_e100_ide();
-	}
-#endif /* CONFIG_ETRAX_IDE */
-#ifdef CONFIG_BLK_DEV_CMD640
-	{
-		extern void ide_probe_for_cmd640x(void);
-		ide_probe_for_cmd640x();
-	}
-#endif /* CONFIG_BLK_DEV_CMD640 */
-#ifdef CONFIG_BLK_DEV_IDE_PMAC
-	{
-		extern int pmac_ide_probe(void);
-		(void)pmac_ide_probe();
-	}
-#endif /* CONFIG_BLK_DEV_IDE_PMAC */
-#ifdef CONFIG_BLK_DEV_GAYLE
-	{
-		extern void gayle_init(void);
-		gayle_init();
-	}
-#endif /* CONFIG_BLK_DEV_GAYLE */
-#ifdef CONFIG_BLK_DEV_FALCON_IDE
-	{
-		extern void falconide_init(void);
-		falconide_init();
-	}
-#endif /* CONFIG_BLK_DEV_FALCON_IDE */
-#ifdef CONFIG_BLK_DEV_MAC_IDE
-	{
-		extern void macide_init(void);
-		macide_init();
-	}
-#endif /* CONFIG_BLK_DEV_MAC_IDE */
-#ifdef CONFIG_BLK_DEV_Q40IDE
-	{
-		extern void q40ide_init(void);
-		q40ide_init();
-	}
-#endif /* CONFIG_BLK_DEV_Q40IDE */
-#ifdef CONFIG_BLK_DEV_BUDDHA
-	{
-		extern void buddha_init(void);
-		buddha_init();
-	}
-#endif /* CONFIG_BLK_DEV_BUDDHA */
-#ifdef CONFIG_BLK_DEV_IDEPNP
-	pnpide_init();
-#endif
-#ifdef CONFIG_H8300
-	h8300_ide_init();
-#endif
-#ifdef BLK_DEV_MPC8xx_IDE
-	mpc8xx_ide_probe();
-#endif
-}
-
-/*
- * Probe module
- */
-
 EXPORT_SYMBOL(ide_lock);
 
 static int ide_bus_match(struct device *dev, struct device_driver *drv)
@@ -1775,33 +1679,6 @@ static int __init ide_init(void)
 
 	proc_ide_create();
 
-#ifdef CONFIG_IDE_ARM
-	ide_arm_init();
-#endif
-#ifdef CONFIG_BLK_DEV_ALI14XX
-	if (probe_ali14xx)
-		(void)ali14xx_init();
-#endif
-#ifdef CONFIG_BLK_DEV_UMC8672
-	if (probe_umc8672)
-		(void)umc8672_init();
-#endif
-#ifdef CONFIG_BLK_DEV_DTC2278
-	if (probe_dtc2278)
-		(void)dtc2278_init();
-#endif
-#ifdef CONFIG_BLK_DEV_HT6560B
-	if (probe_ht6560b)
-		(void)ht6560b_init();
-#endif
-#ifdef CONFIG_BLK_DEV_QD65XX
-	if (probe_qd65xx)
-		(void)qd65xx_init();
-#endif
-
-	/* Probe for special PCI and other "known" interface chipsets. */
-	probe_for_hwifs();
-
 	return 0;
 }
 
@@ -1837,10 +1714,6 @@ void __exit cleanup_module (void)
 	for (index = 0; index < MAX_HWIFS; ++index)
 		ide_unregister(index);
 
-#ifdef CONFIG_BLK_DEV_IDEPNP
-	pnpide_exit();
-#endif
-
 	proc_ide_destroy();
 
 	bus_unregister(&ide_bus_type);
diff --git a/drivers/ide/legacy/Makefile b/drivers/ide/legacy/Makefile
index 409822349f10..7043ec7d1e05 100644
--- a/drivers/ide/legacy/Makefile
+++ b/drivers/ide/legacy/Makefile
@@ -1,15 +1,24 @@
 
+# link order is important here
+
 obj-$(CONFIG_BLK_DEV_ALI14XX)		+= ali14xx.o
+obj-$(CONFIG_BLK_DEV_UMC8672)		+= umc8672.o
 obj-$(CONFIG_BLK_DEV_DTC2278)		+= dtc2278.o
 obj-$(CONFIG_BLK_DEV_HT6560B)		+= ht6560b.o
 obj-$(CONFIG_BLK_DEV_QD65XX)		+= qd65xx.o
-obj-$(CONFIG_BLK_DEV_UMC8672)		+= umc8672.o
 
-obj-$(CONFIG_BLK_DEV_IDECS)		+= ide-cs.o
+obj-$(CONFIG_BLK_DEV_GAYLE)		+= gayle.o
+obj-$(CONFIG_BLK_DEV_FALCON_IDE)	+= falconide.o
+obj-$(CONFIG_BLK_DEV_MAC_IDE)		+= macide.o
+obj-$(CONFIG_BLK_DEV_Q40IDE)		+= q40ide.o
+obj-$(CONFIG_BLK_DEV_BUDDHA)		+= buddha.o
 
-obj-$(CONFIG_BLK_DEV_PLATFORM)		+= ide_platform.o
+ifeq ($(CONFIG_BLK_DEV_IDECS), m)
+	obj-m += ide-cs.o
+endif
 
-# Last of all
-obj-$(CONFIG_BLK_DEV_HD)		+= hd.o
+ifeq ($(CONFIG_BLK_DEV_PLATFORM), m)
+	obj-m += ide_platform.o
+endif
 
 EXTRA_CFLAGS	:= -Idrivers/ide
diff --git a/drivers/ide/legacy/ali14xx.c b/drivers/ide/legacy/ali14xx.c
index 38c3a6d63f30..5ec0be4cbad7 100644
--- a/drivers/ide/legacy/ali14xx.c
+++ b/drivers/ide/legacy/ali14xx.c
@@ -231,8 +231,7 @@ int probe_ali14xx = 0;
 module_param_named(probe, probe_ali14xx, bool, 0);
 MODULE_PARM_DESC(probe, "probe for ALI M14xx chipsets");
 
-/* Can be called directly from ide.c. */
-int __init ali14xx_init(void)
+static int __init ali14xx_init(void)
 {
 	if (probe_ali14xx == 0)
 		goto out;
@@ -248,9 +247,7 @@ out:
 	return -ENODEV;
 }
 
-#ifdef MODULE
 module_init(ali14xx_init);
-#endif
 
 MODULE_AUTHOR("see local file");
 MODULE_DESCRIPTION("support of ALI 14XX IDE chipsets");
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index ba64c4b9f918..e97766aef374 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -143,7 +143,7 @@ static int xsurf_ack_intr(ide_hwif_t *hwif)
      *  Probe for a Buddha or Catweasel IDE interface
      */
 
-void __init buddha_init(void)
+static int __init buddha_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
@@ -243,4 +243,8 @@ fail_base2:
 
 		ide_device_add(idx);
 	}
+
+	return 0;
 }
+
+module_init(buddha_init);
diff --git a/drivers/ide/legacy/dtc2278.c b/drivers/ide/legacy/dtc2278.c
index 24a845d45bd2..13eee6da2806 100644
--- a/drivers/ide/legacy/dtc2278.c
+++ b/drivers/ide/legacy/dtc2278.c
@@ -150,8 +150,7 @@ int probe_dtc2278 = 0;
 module_param_named(probe, probe_dtc2278, bool, 0);
 MODULE_PARM_DESC(probe, "probe for DTC2278xx chipsets");
 
-/* Can be called directly from ide.c. */
-int __init dtc2278_init(void)
+static int __init dtc2278_init(void)
 {
 	if (probe_dtc2278 == 0)
 		return -ENODEV;
@@ -163,9 +162,7 @@ int __init dtc2278_init(void)
 	return 0;
 }
 
-#ifdef MODULE
 module_init(dtc2278_init);
-#endif
 
 MODULE_AUTHOR("See Local File");
 MODULE_DESCRIPTION("support of DTC-2278 VLB IDE chipsets");
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index c1a84540beb1..dec2ef99c778 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -62,7 +62,7 @@ EXPORT_SYMBOL(falconide_intr_lock);
      *  Probe for a Falcon IDE interface
      */
 
-void __init falconide_init(void)
+static int __init falconide_init(void)
 {
     if (MACH_IS_ATARI && ATARIHW_PRESENT(IDE)) {
 	hw_regs_t hw;
@@ -84,4 +84,9 @@ void __init falconide_init(void)
 
 		ide_device_add(idx);
 	}
+    }
+
+    return 0;
 }
+
+module_init(falconide_init);
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index ec53dc9b483c..e21ef75c9055 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -110,13 +110,13 @@ static int gayle_ack_intr_a1200(ide_hwif_t *hwif)
      *  Probe for a Gayle IDE interface (and optionally for an IDE doubler)
      */
 
-void __init gayle_init(void)
+static int __init gayle_init(void)
 {
     int a4000, i;
     u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_AMIGA)
-	return;
+	return -ENODEV;
 
     if ((a4000 = AMIGAHW_PRESENT(A4000_IDE)) || AMIGAHW_PRESENT(A1200_IDE))
 	goto found;
@@ -126,7 +126,7 @@ void __init gayle_init(void)
 			  NULL))
 	goto found;
 #endif
-    return;
+    return -ENODEV;
 
 found:
     for (i = 0; i < GAYLE_NUM_PROBE_HWIFS; i++) {
@@ -191,4 +191,8 @@ found:
     }
 
     ide_device_add(idx);
+
+    return 0;
 }
+
+module_init(gayle_init);
diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index a4245d13f11b..8da5031a6d05 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -307,8 +307,7 @@ int probe_ht6560b = 0;
 module_param_named(probe, probe_ht6560b, bool, 0);
 MODULE_PARM_DESC(probe, "probe for HT6560B chipset");
 
-/* Can be called directly from ide.c. */
-int __init ht6560b_init(void)
+static int __init ht6560b_init(void)
 {
 	ide_hwif_t *hwif, *mate;
 	static u8 idx[4] = { 0, 1, 0xff, 0xff };
@@ -369,9 +368,7 @@ release_region:
 	return -ENODEV;
 }
 
-#ifdef MODULE
 module_init(ht6560b_init);
-#endif
 
 MODULE_AUTHOR("See Local File");
 MODULE_DESCRIPTION("HT-6560B EIDE-controller support");
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index c1b7881c280a..6b3e960350aa 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -81,7 +81,7 @@ int macide_ack_intr(ide_hwif_t* hwif)
  * Probe for a Macintosh IDE interface
  */
 
-void __init macide_init(void)
+static int __init macide_init(void)
 {
 	hw_regs_t hw;
 	ide_hwif_t *hwif;
@@ -106,7 +106,7 @@ void __init macide_init(void)
 				IRQ_BABOON_1);
 		break;
 	default:
-		return;
+		return -ENODEV;
 	}
 
 	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
@@ -139,4 +139,8 @@ void __init macide_init(void)
 
 		ide_device_add(idx);
 	}
+
+	return 0;
 }
+
+module_init(macide_init);
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 2082e9c6efd9..0154c91ee4b1 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -111,7 +111,7 @@ static const char *q40_ide_names[Q40IDE_NUM_HWIFS]={
  *  Probe for Q40 IDE interfaces
  */
 
-void __init q40ide_init(void)
+static int __init q40ide_init(void)
 {
     int i;
     ide_hwif_t *hwif;
@@ -119,7 +119,7 @@ void __init q40ide_init(void)
     u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_Q40)
-      return ;
+      return -ENODEV;
 
     for (i = 0; i < Q40IDE_NUM_HWIFS; i++) {
 	hw_regs_t hw;
@@ -153,5 +153,8 @@ void __init q40ide_init(void)
     }
 
     ide_device_add(idx);
+
+    return 0;
 }
 
+module_init(q40ide_init);
diff --git a/drivers/ide/legacy/qd65xx.c b/drivers/ide/legacy/qd65xx.c
index 912e73853faa..2bac4c1a6532 100644
--- a/drivers/ide/legacy/qd65xx.c
+++ b/drivers/ide/legacy/qd65xx.c
@@ -478,8 +478,7 @@ int probe_qd65xx = 0;
 module_param_named(probe, probe_qd65xx, bool, 0);
 MODULE_PARM_DESC(probe, "probe for QD65xx chipsets");
 
-/* Can be called directly from ide.c. */
-int __init qd65xx_init(void)
+static int __init qd65xx_init(void)
 {
 	if (probe_qd65xx == 0)
 		return -ENODEV;
@@ -492,9 +491,7 @@ int __init qd65xx_init(void)
 	return 0;
 }
 
-#ifdef MODULE
 module_init(qd65xx_init);
-#endif
 
 MODULE_AUTHOR("Samuel Thibault");
 MODULE_DESCRIPTION("support of qd65xx vlb ide chipset");
diff --git a/drivers/ide/legacy/umc8672.c b/drivers/ide/legacy/umc8672.c
index 79577b916874..a1ae1ae6699d 100644
--- a/drivers/ide/legacy/umc8672.c
+++ b/drivers/ide/legacy/umc8672.c
@@ -169,8 +169,7 @@ int probe_umc8672 = 0;
 module_param_named(probe, probe_umc8672, bool, 0);
 MODULE_PARM_DESC(probe, "probe for UMC8672 chipset");
 
-/* Can be called directly from ide.c. */
-int __init umc8672_init(void)
+static int __init umc8672_init(void)
 {
 	if (probe_umc8672 == 0)
 		goto out;
@@ -181,9 +180,7 @@ out:
 	return -ENODEV;;
 }
 
-#ifdef MODULE
 module_init(umc8672_init);
-#endif
 
 MODULE_AUTHOR("Wolfram Podien");
 MODULE_DESCRIPTION("Support for UMC 8672 IDE chipset");
diff --git a/drivers/ide/pci/Makefile b/drivers/ide/pci/Makefile
index 95d1ea8f1f14..94803253e8af 100644
--- a/drivers/ide/pci/Makefile
+++ b/drivers/ide/pci/Makefile
@@ -36,4 +36,8 @@ obj-$(CONFIG_BLK_DEV_VIA82CXXX)		+= via82cxxx.o
 # Must appear at the end of the block
 obj-$(CONFIG_BLK_DEV_GENERIC)          += generic.o
 
+ifeq ($(CONFIG_BLK_DEV_CMD640), m)
+	obj-m += cmd640.o
+endif
+
 EXTRA_CFLAGS	:= -Idrivers/ide
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index 5096e059ac5a..da3565e0071f 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -706,9 +706,9 @@ static int pci_conf2(void)
 }
 
 /*
- * Probe for a cmd640 chipset, and initialize it if found.  Called from ide.c
+ * Probe for a cmd640 chipset, and initialize it if found.
  */
-int __init ide_probe_for_cmd640x (void)
+static int __init cmd640x_init(void)
 {
 #ifdef CONFIG_BLK_DEV_CMD640_ENHANCED
 	int second_port_toggled = 0;
@@ -883,3 +883,7 @@ int __init ide_probe_for_cmd640x (void)
 	return 1;
 }
 
+module_param_named(probe_vlb, cmd640_vlb, bool, 0);
+MODULE_PARM_DESC(probe_vlb, "probe for VLB version of CMD640 chipset");
+
+module_init(cmd640x_init);
diff --git a/drivers/ide/ppc/Makefile b/drivers/ide/ppc/Makefile
new file mode 100644
index 000000000000..65af5848b28c
--- /dev/null
+++ b/drivers/ide/ppc/Makefile
@@ -0,0 +1,3 @@
+
+obj-$(CONFIG_BLK_DEV_IDE_PMAC)		+= pmac.o
+obj-$(CONFIG_BLK_DEV_MPC8xx_IDE)	+= mpc8xx.o
diff --git a/drivers/ide/ppc/mpc8xx.c b/drivers/ide/ppc/mpc8xx.c
index 8172e813b034..3fd5d45b5e0e 100644
--- a/drivers/ide/ppc/mpc8xx.c
+++ b/drivers/ide/ppc/mpc8xx.c
@@ -839,7 +839,7 @@ void m8xx_ide_init(void)
 	ppc_ide_md.ide_init_hwif        = m8xx_ide_init_hwif_ports;
 }
 
-void __init mpc8xx_ide_probe(void)
+static int __init mpc8xx_ide_probe(void)
 {
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
@@ -851,4 +851,8 @@ void __init mpc8xx_ide_probe(void)
 #endif
 
 	ide_device_add(idx);
+
+	return 0;
 }
+
+module_init(mpc8xx_ide_probe);
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 36e4b9570746..cd514743df96 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1786,3 +1786,5 @@ pmac_ide_setup_dma(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 }
 
 #endif /* CONFIG_BLK_DEV_IDEDMA_PMAC */
+
+module_init(pmac_ide_probe);
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index d89f84d41b08..63ef8aaa7b90 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -766,21 +766,20 @@ static int __init ide_scan_pcidev(struct pci_dev *dev)
 
 /**
  *	ide_scan_pcibus		-	perform the initial IDE driver scan
- *	@scan_direction: set for reverse order scanning
  *
  *	Perform the initial bus rather than driver ordered scan of the
  *	PCI drivers. After this all IDE pci handling becomes standard
  *	module ordering not traditionally ordered.
  */
- 	
-void __init ide_scan_pcibus (int scan_direction)
+
+int __init ide_scan_pcibus(void)
 {
 	struct pci_dev *dev = NULL;
 	struct pci_driver *d;
 	struct list_head *l, *n;
 
 	pre_init = 0;
-	if (!scan_direction)
+	if (!ide_scan_direction)
 		while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)))
 			ide_scan_pcidev(dev);
 	else
@@ -801,5 +800,7 @@ void __init ide_scan_pcibus (int scan_direction)
 			printk(KERN_ERR "%s: failed to register %s driver\n",
 					__FUNCTION__, d->driver.mod_name);
 	}
+
+	return 0;
 }
 #endif
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9c037a0f2af2..735737500f8f 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1014,7 +1014,8 @@ extern void do_ide_request(struct request_queue *);
 void ide_init_disk(struct gendisk *, ide_drive_t *);
 
 #ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-extern void ide_scan_pcibus(int scan_direction) __init;
+extern int ide_scan_direction;
+int __init ide_scan_pcibus(void);
 extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *owner, const char *mod_name);
 #define ide_pci_register_driver(d) __ide_pci_register_driver(d, THIS_MODULE, KBUILD_MODNAME)
 #else
-- 
cgit v1.2.3


From 81ca691981da718727281238b435dcf1528d2fda Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:08 +0100
Subject: ide: add ide_set_irq() inline helper

There should be no functionality changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       |  9 +++------
 drivers/ide/ide-iops.c     | 10 ++++------
 drivers/ide/ide-probe.c    | 12 ++++--------
 drivers/ide/ide-taskfile.c |  3 +--
 include/linux/ide.h        |  5 +++++
 5 files changed, 17 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index e37b09c81e3a..5b213dcaa5e6 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -939,8 +939,7 @@ static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 		if (rc)
 			printk(KERN_WARNING "%s: bus not ready on wakeup\n", drive->name);
 		SELECT_DRIVE(drive);
-		if (IDE_CONTROL_REG)
-			HWIF(drive)->OUTB(drive->ctl, IDE_CONTROL_REG);
+		ide_set_irq(drive, 1);
 		rc = ide_wait_not_busy(HWIF(drive), 100000);
 		if (rc)
 			printk(KERN_WARNING "%s: drive not ready on wakeup\n", drive->name);
@@ -1213,15 +1212,13 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 		}
 	again:
 		hwif = HWIF(drive);
-		if (hwgroup->hwif->sharing_irq &&
-		    hwif != hwgroup->hwif &&
-		    hwif->io_ports[IDE_CONTROL_OFFSET]) {
+		if (hwgroup->hwif->sharing_irq && hwif != hwgroup->hwif) {
 			/*
 			 * set nIEN for previous hwif, drives in the
 			 * quirk_list may not like intr setups/cleanups
 			 */
 			if (drive->quirk_list != 1)
-				hwif->OUTB(drive->ctl | 2, IDE_CONTROL_REG);
+				ide_set_irq(drive, 0);
 		}
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 595a5cef41a2..a26c9ca784a6 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -688,8 +688,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	 */
 
 	SELECT_MASK(drive, 1);
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl,IDE_CONTROL_REG);
+	ide_set_irq(drive, 1);
 	msleep(50);
 	hwif->OUTB(WIN_IDENTIFY, IDE_COMMAND_REG);
 	timeout = jiffies + WAIT_WORSTCASE;
@@ -772,13 +771,12 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	SELECT_DRIVE(drive);
 	SELECT_MASK(drive, 0);
 	udelay(1);
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl | 2, IDE_CONTROL_REG);
+	ide_set_irq(drive, 0);
 	hwif->OUTB(speed, IDE_NSECTOR_REG);
 	hwif->OUTB(SETFEATURES_XFER, IDE_FEATURE_REG);
 	hwif->OUTBSYNC(drive, WIN_SETFEATURES, IDE_COMMAND_REG);
-	if ((IDE_CONTROL_REG) && (drive->quirk_list == 2))
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
+	if (drive->quirk_list == 2)
+		ide_set_irq(drive, 1);
 
 	error = __ide_wait_stat(drive, drive->ready_stat,
 				BUSY_STAT|DRQ_STAT|ERR_STAT,
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 18e9b82e132c..9d9f1c6d602a 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -350,22 +350,19 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 	 * the irq handler isn't expecting.
 	 */
 	if (IDE_CONTROL_REG) {
-		u8 ctl = drive->ctl | 2;
 		if (!hwif->irq) {
 			autoprobe = 1;
 			cookie = probe_irq_on();
-			/* enable device irq */
-			ctl &= ~2;
 		}
-		hwif->OUTB(ctl, IDE_CONTROL_REG);
+		ide_set_irq(drive, autoprobe);
 	}
 
 	retval = actual_try_to_identify(drive, cmd);
 
 	if (autoprobe) {
 		int irq;
-		/* mask device irq */
-		hwif->OUTB(drive->ctl|2, IDE_CONTROL_REG);
+
+		ide_set_irq(drive, 0);
 		/* clear drive IRQ */
 		(void) hwif->INB(IDE_STATUS_REG);
 		udelay(5);
@@ -653,8 +650,7 @@ static int wait_hwif_ready(ide_hwif_t *hwif)
 		/* Ignore disks that we will not probe for later. */
 		if (!drive->noprobe || drive->present) {
 			SELECT_DRIVE(drive);
-			if (IDE_CONTROL_REG)
-				hwif->OUTB(drive->ctl, IDE_CONTROL_REG);
+			ide_set_irq(drive, 1);
 			mdelay(2);
 			rc = ide_wait_not_busy(hwif, 35000);
 			if (rc)
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 063e0eb6c9e4..c58edc86ed3e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -83,8 +83,7 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 		tf->hob_lbam, tf->hob_lbah);
 #endif
 
-	if (IDE_CONTROL_REG)
-		hwif->OUTB(drive->ctl, IDE_CONTROL_REG); /* clear nIEN */
+	ide_set_irq(drive, 1);
 
 	if ((task->tf_flags & IDE_TFLAG_NO_SELECT_MASK) == 0)
 		SELECT_MASK(drive, 0);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 735737500f8f..74f1ef9c6d94 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1302,4 +1302,9 @@ static inline ide_drive_t *ide_get_paired_drive(ide_drive_t *drive)
 	return &hwif->drives[(drive->dn ^ 1) & 1];
 }
 
+static inline void ide_set_irq(ide_drive_t *drive, int on)
+{
+	drive->hwif->OUTB(drive->ctl | (on ? 0 : 2), IDE_CONTROL_REG);
+}
+
 #endif /* _IDE_H */
-- 
cgit v1.2.3


From 9e47be0c97f7357b80e91dc0632e9cce2eb025e0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:09 +0100
Subject: ide: remove broken disk byte-swapping support

Remove broken disk byte-swapping support:
- it can cause a data corruption on SMP (or if using PREEMPT on UP)
- all data coming from disk are byte-swapped by taskfile_*_data() which
  results in incorrect identify data being reported by /proc/ide/ and IOCTLs
- "hdx=bswap/byteswap" kernel parameter has been broken on m68k host drivers
  (including Atari/Q40 ones) since 2.5.x days (because of 'hwif' zero-ing)
- byte-swapping is limited to PIO transfers (for working with TiVo disks on
  x86 machines using user-space solutions or dm-byteswap should result in
  much better performance because DMA can be used)

For previous discussions please see:

http://www.ussg.iu.edu/hypermail/linux/kernel/0201.0/0768.html
http://lkml.org/lkml/2004/2/28/111

[ I have dm-byteswap device mapper target if somebody is interested
  (patch is for 2.6.4 though but I'll dust it off if needed). ]

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 Documentation/ide.txt      |  4 ----
 drivers/ide/ide-disk.c     |  1 -
 drivers/ide/ide-taskfile.c | 32 ++------------------------------
 drivers/ide/ide.c          |  6 +-----
 include/linux/ide.h        |  1 -
 5 files changed, 3 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/Documentation/ide.txt b/Documentation/ide.txt
index b29ccb43d6cc..94e2e3b9e77f 100644
--- a/Documentation/ide.txt
+++ b/Documentation/ide.txt
@@ -244,10 +244,6 @@ Summary of ide driver parameters for kernel command line
 
  "hdx=nodma"		: disallow DMA
 
- "hdx=swapdata"		: when the drive is a disk, byte swap all data
-
- "hdx=bswap"		: same as above..........
-
  "hdx=scsi"		: the return of the ide-scsi flag, this is useful for
  			  allowing ide-floppy, ide-tape, and ide-cdrom|writers
  			  to use ide-scsi emulation on a device specific option.
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index c1fb75c387ea..041be43a62cb 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -761,7 +761,6 @@ static void idedisk_add_settings(ide_drive_t *drive)
 	ide_add_setting(drive,	"bios_head",	SETTING_RW,	TYPE_BYTE,	0,	255,			1,	1,	&drive->bios_head,	NULL);
 	ide_add_setting(drive,	"bios_sect",	SETTING_RW,	TYPE_BYTE,	0,	63,			1,	1,	&drive->bios_sect,	NULL);
 	ide_add_setting(drive,	"address",	SETTING_RW,	TYPE_BYTE,	0,	2,			1,	1,	&drive->addressing,	set_lba_addressing);
-	ide_add_setting(drive,	"bswap",	SETTING_READ,	TYPE_BYTE,	0,	1,			1,	1,	&drive->bswap,		NULL);
 	ide_add_setting(drive,	"multcount",	SETTING_RW,	TYPE_BYTE,	0,	id->max_multsect,	1,	1,	&drive->mult_count,	set_multcount);
 	ide_add_setting(drive,	"nowerr",	SETTING_RW,	TYPE_BYTE,	0,	1,			1,	1,	&drive->nowerr,		set_nowerr);
 	ide_add_setting(drive,	"lun",		SETTING_RW,	TYPE_INT,	0,	7,			1,	1,	&drive->lun,		NULL);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index c58edc86ed3e..3ecafab8f54e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -35,34 +35,6 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
-static void ata_bswap_data (void *buffer, int wcount)
-{
-	u16 *p = buffer;
-
-	while (wcount--) {
-		*p = *p << 8 | *p >> 8; p++;
-		*p = *p << 8 | *p >> 8; p++;
-	}
-}
-
-static void taskfile_input_data(ide_drive_t *drive, void *buffer, u32 wcount)
-{
-	HWIF(drive)->ata_input_data(drive, buffer, wcount);
-	if (drive->bswap)
-		ata_bswap_data(buffer, wcount);
-}
-
-static void taskfile_output_data(ide_drive_t *drive, void *buffer, u32 wcount)
-{
-	if (drive->bswap) {
-		ata_bswap_data(buffer, wcount);
-		HWIF(drive)->ata_output_data(drive, buffer, wcount);
-		ata_bswap_data(buffer, wcount);
-	} else {
-		HWIF(drive)->ata_output_data(drive, buffer, wcount);
-	}
-}
-
 void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif = drive->hwif;
@@ -352,9 +324,9 @@ static void ide_pio_sector(ide_drive_t *drive, unsigned int write)
 
 	/* do the actual data transfer */
 	if (write)
-		taskfile_output_data(drive, buf, SECTOR_WORDS);
+		hwif->ata_output_data(drive, buf, SECTOR_WORDS);
 	else
-		taskfile_input_data(drive, buf, SECTOR_WORDS);
+		hwif->ata_input_data(drive, buf, SECTOR_WORDS);
 
 	kunmap_atomic(buf, KM_BIO_SRC_IRQ);
 #ifdef CONFIG_HIGHMEM
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 5f3e53ec5838..52115ef1f01a 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -1303,7 +1303,7 @@ static int __init ide_setup(char *s)
 	if (s[0] == 'h' && s[1] == 'd' && s[2] >= 'a' && s[2] <= max_drive) {
 		const char *hd_words[] = {
 			"none", "noprobe", "nowerr", "cdrom", "nodma",
-			"autotune", "noautotune", "minus8", "swapdata", "bswap",
+			"autotune", "noautotune", "-8", "-9", "-10",
 			"noflush", "remap", "remap63", "scsi", NULL };
 		unit = s[2] - 'a';
 		hw   = unit / MAX_DRIVES;
@@ -1339,10 +1339,6 @@ static int __init ide_setup(char *s)
 			case -7: /* "noautotune" */
 				drive->autotune = IDE_TUNE_NOAUTO;
 				goto obsolete_option;
-			case -9: /* "swapdata" */
-			case -10: /* "bswap" */
-				drive->bswap = 1;
-				goto done;
 			case -11: /* noflush */
 				drive->noflush = 1;
 				goto done;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 74f1ef9c6d94..f94cf036bd95 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -394,7 +394,6 @@ typedef struct ide_drive_s {
 	u8	state;			/* retry state */
 	u8	waiting_for_dma;	/* dma currently in progress */
 	u8	unmask;			/* okay to unmask other irqs */
-	u8	bswap;			/* byte swap data */
 	u8	noflush;		/* don't attempt flushes */
 	u8	dsc_overlap;		/* DSC overlap */
 	u8	nice1;			/* give potential excess bandwidth */
-- 
cgit v1.2.3


From 35cf2b94d0ecb7034cfa05dd725721538bbb83fc Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 26 Jan 2008 20:13:10 +0100
Subject: ide: fix ->io_32bit race in ide_taskfile_ioctl()

In ide_taskfile_ioctl(), there was a race condition involving
drive->io_32bit.  It was cleared and restored during ioctl
requests but there was no synchronization with other requests.
So, other requests could execute with the altered ->io_32bit
setting or updated drive->io_32bit could be overwritten by
ide_taskfile_ioctl().

This patch adds IDE_TFLAG_IO_16BIT flag to indicate to
ide_pio_datablock() that 16-bit I/O is needed regardless of
drive->io_32bit settting.

Bart:
- ported it over recent IDE changes

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-taskfile.c | 17 ++++++++++++-----
 include/linux/ide.h        |  2 ++
 2 files changed, 14 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 3ecafab8f54e..b72a9aea7a51 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -346,9 +346,18 @@ static void ide_pio_multi(ide_drive_t *drive, unsigned int write)
 static void ide_pio_datablock(ide_drive_t *drive, struct request *rq,
 				     unsigned int write)
 {
+	u8 saved_io_32bit = drive->io_32bit;
+
 	if (rq->bio)	/* fs request */
 		rq->errors = 0;
 
+	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
+		ide_task_t *task = rq->special;
+
+		if (task->tf_flags & IDE_TFLAG_IO_16BIT)
+			drive->io_32bit = 0;
+	}
+
 	touch_softlockup_watchdog();
 
 	switch (drive->hwif->data_phase) {
@@ -360,6 +369,8 @@ static void ide_pio_datablock(ide_drive_t *drive, struct request *rq,
 		ide_pio_sector(drive, write);
 		break;
 	}
+
+	drive->io_32bit = saved_io_32bit;
 }
 
 static ide_startstop_t task_error(ide_drive_t *drive, struct request *rq,
@@ -555,7 +566,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	unsigned int taskin	= 0;
 	unsigned int taskout	= 0;
 	u16 nsect		= 0;
-	u8 io_32bit		= drive->io_32bit;
 	char __user *buf = (char __user *)arg;
 
 //	printk("IDE Taskfile ...\n");
@@ -608,7 +618,7 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	args.data_phase = req_task->data_phase;
 
-	args.tf_flags = IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_IO_16BIT | IDE_TFLAG_OUT_DEVICE;
 	if (drive->addressing == 1)
 		args.tf_flags |= IDE_TFLAG_LBA48;
 
@@ -646,7 +656,6 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	if (req_task->in_flags.b.data)
 		args.tf_flags |= IDE_TFLAG_IN_DATA;
 
-	drive->io_32bit = 0;
 	switch(req_task->data_phase) {
 		case TASKFILE_MULTI_OUT:
 			if (!drive->mult_count) {
@@ -742,8 +751,6 @@ abort:
 
 //	printk("IDE Taskfile ioctl ended. rc = %i\n", err);
 
-	drive->io_32bit = io_32bit;
-
 	return err;
 }
 #endif
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f94cf036bd95..c1a8b8bb93ab 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -931,6 +931,8 @@ enum {
 	IDE_TFLAG_IN_TF			= IDE_TFLAG_IN_NSECT |
 					  IDE_TFLAG_IN_LBA,
 	IDE_TFLAG_IN_DEVICE		= (1 << 29),
+	/* force 16-bit I/O operations */
+	IDE_TFLAG_IO_16BIT		= (1 << 30),
 };
 
 struct ide_taskfile {
-- 
cgit v1.2.3


From 657cc1a8f6cd6a9e2974cba3af9fccd8c25e06ad Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:10 +0100
Subject: ide: set IDE_TFLAG_IN_* flags before queuing/executing command

* Add IDE_TFLAG_{HOB,TF,DEVICE} defines.

* Set IDE_TFLAG_IN_* flags in {do_rw,ide_no_data,ide_raw}_taskfile() users.

* Remove no longer needed ->tf_flags setup from ide_end_drive_cmd().

There should be no functionality changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c     |  2 +-
 drivers/ide/ide-disk.c     | 24 ++++++++++++------------
 drivers/ide/ide-io.c       |  8 ++------
 drivers/ide/ide-taskfile.c |  9 +++++----
 include/linux/ide.h        |  6 ++++++
 5 files changed, 26 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index e0bb0cfa7bdd..e888fc35b27c 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -386,7 +386,7 @@ static int taskfile_load_raw(ide_drive_t *drive,
 
 	/* convert gtf to IDE Taskfile */
 	memcpy(&args.tf_array[7], &gtf->tfa, 7);
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 
 	if (ide_noacpitfs) {
 		DEBPRINT("_GTF execution disabled\n");
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 041be43a62cb..027bf4397592 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -201,7 +201,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_NO_SELECT_MASK;  /* FIXME? */
-	task.tf_flags |= (IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE);
+	task.tf_flags |= (IDE_TFLAG_TF | IDE_TFLAG_DEVICE);
 
 	if (drive->select.b.lba) {
 		if (lba48) {
@@ -220,7 +220,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 			tf->lbam   = (u8)(block >>  8);
 			tf->lbah   = (u8)(block >> 16);
 
-			task.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+			task.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_HOB);
 		} else {
 			tf->nsect  = nsectors & 0xff;
 			tf->lbal   = block;
@@ -314,9 +314,9 @@ static u64 idedisk_read_native_max_address(ide_drive_t *drive, int lba48)
 	else
 		tf->command = WIN_READ_NATIVE_MAX;
 	tf->device  = ATA_LBA;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	if (lba48)
-		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_HOB);
 	/* submit command request */
 	ide_no_data_taskfile(drive, &args);
 
@@ -353,9 +353,9 @@ static u64 idedisk_set_max_address(ide_drive_t *drive, u64 addr_req, int lba48)
 		tf->command  = WIN_SET_MAX;
 	}
 	tf->device |= ATA_LBA;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	if (lba48)
-		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_OUT_HOB);
+		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_HOB);
 	/* submit command request */
 	ide_no_data_taskfile(drive, &args);
 	/* if OK, compute maximum address value */
@@ -495,7 +495,7 @@ static int smart_enable(ide_drive_t *drive)
 	tf->lbam    = SMART_LCYL_PASS;
 	tf->lbah    = SMART_HCYL_PASS;
 	tf->command = WIN_SMART;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	return ide_no_data_taskfile(drive, &args);
 }
 
@@ -510,7 +510,7 @@ static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
 	tf->lbam    = SMART_LCYL_PASS;
 	tf->lbah    = SMART_HCYL_PASS;
 	tf->command = WIN_SMART;
-	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags	= IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	args.data_phase	= TASKFILE_IN;
 	(void) smart_enable(drive);
 	return ide_raw_taskfile(drive, &args, buf, 1);
@@ -689,7 +689,7 @@ static int write_cache(ide_drive_t *drive, int arg)
 		args.tf.feature = arg ?
 			SETFEATURES_EN_WCACHE : SETFEATURES_DIS_WCACHE;
 		args.tf.command = WIN_SETFEATURES;
-		args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+		args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 		err = ide_no_data_taskfile(drive, &args);
 		if (err == 0)
 			drive->wcache = arg;
@@ -709,7 +709,7 @@ static int do_idedisk_flushcache (ide_drive_t *drive)
 		args.tf.command = WIN_FLUSH_CACHE_EXT;
 	else
 		args.tf.command = WIN_FLUSH_CACHE;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	return ide_no_data_taskfile(drive, &args);
 }
 
@@ -724,7 +724,7 @@ static int set_acoustic (ide_drive_t *drive, int arg)
 	args.tf.feature = arg ? SETFEATURES_EN_AAM : SETFEATURES_DIS_AAM;
 	args.tf.nsect   = arg;
 	args.tf.command = WIN_SETFEATURES;
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	ide_no_data_taskfile(drive, &args);
 	drive->acoustic = arg;
 	return 0;
@@ -975,7 +975,7 @@ static int idedisk_set_doorlock(ide_drive_t *drive, int on)
 
 	memset(&task, 0, sizeof(task));
 	task.tf.command = on ? WIN_DOORLOCK : WIN_DOORUNLOCK;
-	task.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 
 	return ide_no_data_taskfile(drive, &task);
 }
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index f01b103e55f0..0f3e2f4f9c2c 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -231,7 +231,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 	return ide_stopped;
 
 out_do_tf:
-	args->tf_flags	 = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args->tf_flags	 = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	args->data_phase = TASKFILE_NO_DATA;
 	return do_rw_taskfile(drive, args);
 }
@@ -385,10 +385,6 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			tf->error = err;
 			tf->status = stat;
 
-			args->tf_flags |= (IDE_TFLAG_IN_TF|IDE_TFLAG_IN_DEVICE);
-			if (args->tf_flags & IDE_TFLAG_LBA48)
-				args->tf_flags |= IDE_TFLAG_IN_HOB;
-
 			ide_tf_read(drive, args);
 		}
 	} else if (blk_pm_request(rq)) {
@@ -712,7 +708,7 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 		return ide_stopped;
 	}
 
-	args.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE |
+	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE |
 			IDE_TFLAG_CUSTOM_HANDLER;
 
 	do_rw_taskfile(drive, &args);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index bc3d8aed9a8d..1f664ea57e58 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -99,7 +99,7 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 		args.tf.command = WIN_IDENTIFY;
 	else
 		args.tf.command = WIN_PIDENTIFY;
-	args.tf_flags	= IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags	= IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	args.data_phase	= TASKFILE_IN;
 	return ide_raw_taskfile(drive, &args, buf, 1);
 }
@@ -618,9 +618,10 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 
 	args.data_phase = req_task->data_phase;
 
-	args.tf_flags = IDE_TFLAG_IO_16BIT | IDE_TFLAG_OUT_DEVICE;
+	args.tf_flags = IDE_TFLAG_IO_16BIT | IDE_TFLAG_DEVICE |
+			IDE_TFLAG_IN_TF;
 	if (drive->addressing == 1)
-		args.tf_flags |= IDE_TFLAG_LBA48;
+		args.tf_flags |= (IDE_TFLAG_LBA48 | IDE_TFLAG_IN_HOB);
 
 	if (req_task->out_flags.all) {
 		args.tf_flags |= IDE_TFLAG_FLAGGED;
@@ -836,7 +837,7 @@ int ide_task_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
 	memset(&task, 0, sizeof(task));
 	memcpy(&task.tf_array[7], &args[1], 6);
 	task.tf.command = args[0];
-	task.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE;
+	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 
 	err = ide_no_data_taskfile(drive, &task);
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c1a8b8bb93ab..02493dbb156d 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -931,6 +931,12 @@ enum {
 	IDE_TFLAG_IN_TF			= IDE_TFLAG_IN_NSECT |
 					  IDE_TFLAG_IN_LBA,
 	IDE_TFLAG_IN_DEVICE		= (1 << 29),
+	IDE_TFLAG_HOB			= IDE_TFLAG_OUT_HOB |
+					  IDE_TFLAG_IN_HOB,
+	IDE_TFLAG_TF			= IDE_TFLAG_OUT_TF |
+					  IDE_TFLAG_IN_TF,
+	IDE_TFLAG_DEVICE		= IDE_TFLAG_OUT_DEVICE |
+					  IDE_TFLAG_IN_DEVICE,
 	/* force 16-bit I/O operations */
 	IDE_TFLAG_IO_16BIT		= (1 << 30),
 };
-- 
cgit v1.2.3


From 4d7a984bdcbdda69fc6b2a4a655415140270aa7b Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 26 Jan 2008 20:13:11 +0100
Subject: ide: task_end_request() fix

task_end_request() modified to always call ide_end_drive_cmd()
for taskfile requests.  Previously, ide_end_drive_cmd() was
called only when IDE_TFLAG_FLAGGED was set.  Also,
ide_dma_intr() is modified to use task_end_request().

Enables TASKFILE ioctls to get valid register outputs on
successful completion.

Bart:
- ported it over recent IDE changes

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c      |  8 +-------
 drivers/ide/ide-taskfile.c | 11 ++++-------
 include/linux/ide.h        |  2 ++
 3 files changed, 7 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index c9648b1ef220..5bf32038dc43 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -153,13 +153,7 @@ ide_startstop_t ide_dma_intr (ide_drive_t *drive)
 		if (!dma_stat) {
 			struct request *rq = HWGROUP(drive)->rq;
 
-			if (rq->rq_disk) {
-				ide_driver_t *drv;
-
-				drv = *(ide_driver_t **)rq->rq_disk->private_data;
-				drv->end_request(drive, 1, rq->nr_sectors);
-			} else
-				ide_end_request(drive, 1, rq->nr_sectors);
+			task_end_request(drive, rq, stat);
 			return ide_stopped;
 		}
 		printk(KERN_ERR "%s: dma_intr: bad DMA status (dma_stat=%x)\n", 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 3bbb438f4f92..17c2c0467297 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -408,16 +408,13 @@ static ide_startstop_t task_error(ide_drive_t *drive, struct request *rq,
 	return ide_error(drive, s, stat);
 }
 
-static void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
+void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 {
 	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
-		ide_task_t *task = rq->special;
+		u8 err = drive->hwif->INB(IDE_ERROR_REG);
 
-		if (task->tf_flags & IDE_TFLAG_FLAGGED) {
-			u8 err = drive->hwif->INB(IDE_ERROR_REG);
-			ide_end_drive_cmd(drive, stat, err);
-			return;
-		}
+		ide_end_drive_cmd(drive, stat, err);
+		return;
 	}
 
 	if (rq->rq_disk) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 02493dbb156d..6192564fcc5a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -993,6 +993,8 @@ void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
+void task_end_request(ide_drive_t *, struct request *, u8);
+
 int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *, u16);
 int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
 
-- 
cgit v1.2.3


From 4906f3b4cddc3e4d62955ed386598561f95602c0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:11 +0100
Subject: ide: kill DATA_READY define

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-taskfile.c | 4 ++--
 include/linux/ide.h        | 1 -
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 17c2c0467297..c34836c02b14 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -436,7 +436,7 @@ static ide_startstop_t task_in_intr(ide_drive_t *drive)
 	u8 stat = hwif->INB(IDE_STATUS_REG);
 
 	/* new way for dealing with premature shared PCI interrupts */
-	if (!OK_STAT(stat, DATA_READY, BAD_R_STAT)) {
+	if (!OK_STAT(stat, DRQ_STAT, BAD_R_STAT)) {
 		if (stat & (ERR_STAT | DRQ_STAT))
 			return task_error(drive, rq, __FUNCTION__, stat);
 		/* No data yet, so wait for another IRQ. */
@@ -493,7 +493,7 @@ static ide_startstop_t pre_task_out_intr(ide_drive_t *drive, struct request *rq)
 {
 	ide_startstop_t startstop;
 
-	if (ide_wait_stat(&startstop, drive, DATA_READY,
+	if (ide_wait_stat(&startstop, drive, DRQ_STAT,
 			  drive->bad_wstat, WAIT_DRQ)) {
 		printk(KERN_ERR "%s: no DRQ after issuing %sWRITE%s\n",
 				drive->name,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 6192564fcc5a..7615a6244a51 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -107,7 +107,6 @@ typedef unsigned char	byte;	/* used everywhere */
 #define BAD_W_STAT		(BAD_R_STAT  | WRERR_STAT)
 #define BAD_STAT		(BAD_R_STAT  | DRQ_STAT)
 #define DRIVE_READY		(READY_STAT  | SEEK_STAT)
-#define DATA_READY		(DRQ_STAT)
 
 #define BAD_CRC			(ABRT_ERR    | ICRC_ERR)
 
-- 
cgit v1.2.3


From 2624565caacedd740fce7803fe2c162842aa5df4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:11 +0100
Subject: ide: use wait_drive_not_busy() in drive_cmd_intr() (take 2)

Use wait_drive_not_busy() in drive_cmd_intr().

v2:
* Fix wait_drive_not_busy() comment (noticed by Sergei).

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       | 4 +---
 drivers/ide/ide-taskfile.c | 5 ++---
 include/linux/ide.h        | 2 ++
 3 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 0f3e2f4f9c2c..513a5685db2f 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -640,7 +640,6 @@ static ide_startstop_t drive_cmd_intr (ide_drive_t *drive)
 	ide_hwif_t *hwif = HWIF(drive);
 	u8 *args = (u8 *) rq->buffer;
 	u8 stat = hwif->INB(IDE_STATUS_REG);
-	int retries = 10;
 
 	local_irq_enable_in_hardirq();
 	if (rq->cmd_type == REQ_TYPE_ATA_CMD &&
@@ -649,8 +648,7 @@ static ide_startstop_t drive_cmd_intr (ide_drive_t *drive)
 		drive->io_32bit = 0;
 		hwif->ata_input_data(drive, &args[4], args[3] * SECTOR_WORDS);
 		drive->io_32bit = io_32bit;
-		while (((stat = hwif->INB(IDE_STATUS_REG)) & BUSY_STAT) && retries--)
-			udelay(100);
+		stat = wait_drive_not_busy(drive);
 	}
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index c34836c02b14..b559bece6e73 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -260,7 +260,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 	return ide_stopped;
 }
 
-static u8 wait_drive_not_busy(ide_drive_t *drive)
+u8 wait_drive_not_busy(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	int retries;
@@ -268,8 +268,7 @@ static u8 wait_drive_not_busy(ide_drive_t *drive)
 
 	/*
 	 * Last sector was transfered, wait until drive is ready.
-	 * This can take up to 10 usec, but we will wait max 1 ms
-	 * (drive_cmd_intr() waits that long).
+	 * This can take up to 10 usec, but we will wait max 1 ms.
 	 */
 	for (retries = 0; retries < 100; retries++) {
 		if ((stat = hwif->INB(IDE_STATUS_REG)) & BUSY_STAT)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 7615a6244a51..c889a6e91c5c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -994,6 +994,8 @@ ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
 void task_end_request(ide_drive_t *, struct request *, u8);
 
+u8 wait_drive_not_busy(ide_drive_t *);
+
 int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *, u16);
 int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
 
-- 
cgit v1.2.3


From 34f5d5ae35240a11846875d76eb935875ab0c366 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:12 +0100
Subject: ide: switch set_xfer_rate() to use REQ_TYPE_ATA_TASKFILE requests

Based on the earlier work by Tejun Heo.

Switch set_xfer_rate() to use REQ_TYPE_ATA_TASKFILE requests
and make ide_wait_cmd() static.

There should be no functionality changes caused by this patch.

Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-proc.c     | 12 +++++++++---
 drivers/ide/ide-taskfile.c |  3 ++-
 include/linux/ide.h        |  8 --------
 3 files changed, 11 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index a4007d30da52..aa663e7f46f2 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -346,14 +346,20 @@ static int ide_write_setting(ide_drive_t *drive, ide_settings_t *setting, int va
 
 static int set_xfer_rate (ide_drive_t *drive, int arg)
 {
+	ide_task_t task;
 	int err;
 
 	if (arg < 0 || arg > 70)
 		return -EINVAL;
 
-	err = ide_wait_cmd(drive,
-			WIN_SETFEATURES, (u8) arg,
-			SETFEATURES_XFER, 0, NULL);
+	memset(&task, 0, sizeof(task));
+	task.tf.command = WIN_SETFEATURES;
+	task.tf.feature = SETFEATURES_XFER;
+	task.tf.nsect   = (u8)arg;
+	task.tf_flags = IDE_TFLAG_OUT_FEATURE | IDE_TFLAG_OUT_NSECT |
+			IDE_TFLAG_IN_NSECT;
+
+	err = ide_no_data_taskfile(drive, &task);
 
 	if (!err && arg) {
 		ide_set_xfer_rate(drive, (u8) arg);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 94046509f004..a1796cf5835c 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -750,7 +750,8 @@ abort:
 }
 #endif
 
-int ide_wait_cmd (ide_drive_t *drive, u8 cmd, u8 nsect, u8 feature, u8 sectors, u8 *buf)
+static int ide_wait_cmd(ide_drive_t *drive, u8 cmd, u8 nsect, u8 feature,
+			u8 sectors, u8 *buf)
 {
 	struct request rq;
 	u8 buffer[4];
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c889a6e91c5c..27cb39de2ae2 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -870,14 +870,6 @@ extern int ide_do_drive_cmd(ide_drive_t *, struct request *, ide_action_t);
 
 extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
 
-/*
- * Issue ATA command and wait for completion.
- * Use for implementing commands in kernel
- *
- *  (ide_drive_t *drive, u8 cmd, u8 nsect, u8 feature, u8 sectors, u8 *buf)
- */
-extern int ide_wait_cmd(ide_drive_t *, u8, u8, u8, u8, u8 *);
-
 enum {
 	IDE_TFLAG_LBA48			= (1 << 0),
 	IDE_TFLAG_NO_SELECT_MASK	= (1 << 1),
-- 
cgit v1.2.3


From 7267c3377443322588cddaf457cf106839a60463 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Sat, 26 Jan 2008 20:13:13 +0100
Subject: ide: remove REQ_TYPE_ATA_CMD

Based on the earlier work by Tejun Heo.

All users are gone so we can finally remove it.

Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 98 +++-----------------------------------------------
 drivers/ide/ide-lib.c  | 25 ++++---------
 include/linux/blkdev.h |  1 -
 3 files changed, 11 insertions(+), 113 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index cad057d25a29..6f8f544392a8 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -354,7 +354,6 @@ void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
  
 void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 {
-	ide_hwif_t *hwif = HWIF(drive);
 	unsigned long flags;
 	struct request *rq;
 
@@ -362,19 +361,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 	rq = HWGROUP(drive)->rq;
 	spin_unlock_irqrestore(&ide_lock, flags);
 
-	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
-		u8 *args = (u8 *) rq->buffer;
-		if (rq->errors == 0)
-			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
-
-		if (args) {
-			args[0] = stat;
-			args[1] = err;
-			/* be sure we're looking at the low order bits */
-			hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
-			args[2] = hwif->INB(IDE_NSECTOR_REG);
-		}
-	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
+	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *args = (ide_task_t *) rq->special;
 		if (rq->errors == 0)
 			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
@@ -624,48 +611,6 @@ ide_startstop_t ide_abort(ide_drive_t *drive, const char *msg)
 		return __ide_abort(drive, rq);
 }
 
-/**
- *	drive_cmd_intr		- 	drive command completion interrupt
- *	@drive: drive the completion interrupt occurred on
- *
- *	drive_cmd_intr() is invoked on completion of a special DRIVE_CMD.
- *	We do any necessary data reading and then wait for the drive to
- *	go non busy. At that point we may read the error data and complete
- *	the request
- */
- 
-static ide_startstop_t drive_cmd_intr (ide_drive_t *drive)
-{
-	struct request *rq = HWGROUP(drive)->rq;
-	ide_hwif_t *hwif = HWIF(drive);
-	u8 *args = (u8 *)rq->buffer, pio_in = (args && args[3]) ? 1 : 0, stat;
-
-	if (pio_in) {
-		u8 io_32bit = drive->io_32bit;
-		stat = hwif->INB(IDE_STATUS_REG);
-		if (!OK_STAT(stat, DRQ_STAT, BAD_R_STAT)) {
-			if (stat & (ERR_STAT | DRQ_STAT))
-				return ide_error(drive, __FUNCTION__, stat);
-			ide_set_handler(drive, &drive_cmd_intr, WAIT_WORSTCASE,
-					NULL);
-			return ide_started;
-		}
-		drive->io_32bit = 0;
-		hwif->ata_input_data(drive, &args[4], args[3] * SECTOR_WORDS);
-		drive->io_32bit = io_32bit;
-		stat = wait_drive_not_busy(drive);
-	} else {
-		local_irq_enable_in_hardirq();
-		stat = hwif->INB(IDE_STATUS_REG);
-	}
-
-	if (!OK_STAT(stat, (pio_in ? 0 : READY_STAT), BAD_STAT))
-		return ide_error(drive, __FUNCTION__, stat);
-		/* calls ide_end_drive_cmd */
-	ide_end_drive_cmd(drive, stat, hwif->INB(IDE_ERROR_REG));
-	return ide_stopped;
-}
-
 static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
 	tf->nsect   = drive->sect;
@@ -851,16 +796,9 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		struct request *rq)
 {
 	ide_hwif_t *hwif = HWIF(drive);
-	u8 *args = rq->buffer;
-	ide_task_t ltask;
-	struct ide_taskfile *tf = &ltask.tf;
-
-	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
-		ide_task_t *task = rq->special;
- 
-		if (task == NULL)
-			goto done;
+	ide_task_t *task = rq->special;
 
+	if (task) {
 		hwif->data_phase = task->data_phase;
 
 		switch (hwif->data_phase) {
@@ -877,33 +815,6 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		return do_rw_taskfile(drive, task);
 	}
 
-	if (args == NULL)
-		goto done;
-
-	memset(&ltask, 0, sizeof(ltask));
-	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
-#ifdef DEBUG
-		printk("%s: DRIVE_CMD\n", drive->name);
-#endif
-		tf->feature = args[2];
-		if (args[0] == WIN_SMART) {
-			tf->nsect = args[3];
-			tf->lbal  = args[1];
-			tf->lbam  = 0x4f;
-			tf->lbah  = 0xc2;
-			ltask.tf_flags = IDE_TFLAG_OUT_TF;
-		} else {
-			tf->nsect = args[1];
-			ltask.tf_flags = IDE_TFLAG_OUT_FEATURE |
-					 IDE_TFLAG_OUT_NSECT;
-		}
- 	}
-	tf->command = args[0];
-	ide_tf_load(drive, &ltask);
-	ide_execute_command(drive, args[0], &drive_cmd_intr, WAIT_WORSTCASE, NULL);
-	return ide_started;
-
-done:
  	/*
  	 * NULL is actually a valid way of waiting for
  	 * all current requests to be flushed from the queue.
@@ -1007,8 +918,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		if (drive->current_speed == 0xff)
 			ide_config_drive_speed(drive, drive->desired_speed);
 
-		if (rq->cmd_type == REQ_TYPE_ATA_CMD ||
-		    rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
+		if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
 			return execute_drive_cmd(drive, rq);
 		else if (blk_pm_request(rq)) {
 			struct request_pm_state *pm = rq->data;
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index a3bd8e8ed6b0..9b44fbdfe41f 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -454,8 +454,7 @@ int ide_set_xfer_rate(ide_drive_t *drive, u8 rate)
 static void ide_dump_opcode(ide_drive_t *drive)
 {
 	struct request *rq;
-	u8 opcode = 0;
-	int found = 0;
+	ide_task_t *task = NULL;
 
 	spin_lock(&ide_lock);
 	rq = NULL;
@@ -464,25 +463,15 @@ static void ide_dump_opcode(ide_drive_t *drive)
 	spin_unlock(&ide_lock);
 	if (!rq)
 		return;
-	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
-		char *args = rq->buffer;
-		if (args) {
-			opcode = args[0];
-			found = 1;
-		}
-	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
-		ide_task_t *args = rq->special;
-		if (args) {
-			opcode = args->tf.command;
-			found = 1;
-		}
-	}
+
+	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
+		task = rq->special;
 
 	printk("ide: failed opcode was: ");
-	if (!found)
-		printk("unknown\n");
+	if (task == NULL)
+		printk(KERN_CONT "unknown\n");
 	else
-		printk("0x%02x\n", opcode);
+		printk(KERN_CONT "0x%02x\n", task->tf.command);
 }
 
 u64 ide_get_lba_addr(struct ide_taskfile *tf, int lba48)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index bd20a4e8663a..49b7a4c31a6d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -143,7 +143,6 @@ enum rq_cmd_type_bits {
 	 * use REQ_TYPE_SPECIAL and use rq->cmd[0] with the range of driver
 	 * private REQ_LB opcodes to differentiate what type of request this is
 	 */
-	REQ_TYPE_ATA_CMD,
 	REQ_TYPE_ATA_TASKFILE,
 	REQ_TYPE_ATA_PC,
 };
-- 
cgit v1.2.3


From eee87d3196c9a7ac3422f4298e2250ca68d791c1 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sun, 27 Jan 2008 18:14:45 +0100
Subject: i2c: the scheduled I2C RTC driver removal

This patch contains the scheduled removal of legacy I2C RTC drivers with
replacement drivers.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 Documentation/feature-removal-schedule.txt |   7 -
 arch/ppc/platforms/83xx/mpc834x_sys.c      |  20 --
 arch/ppc/platforms/85xx/tqm85xx.c          |  21 --
 arch/ppc/platforms/katana.c                |  21 --
 drivers/i2c/chips/Kconfig                  |  38 ---
 drivers/i2c/chips/Makefile                 |   3 -
 drivers/i2c/chips/ds1337.c                 | 410 ----------------------------
 drivers/i2c/chips/ds1374.c                 | 267 -------------------
 drivers/i2c/chips/m41t00.c                 | 413 -----------------------------
 include/linux/i2c-id.h                     |   2 -
 include/linux/m41t00.h                     |  50 ----
 11 files changed, 1252 deletions(-)
 delete mode 100644 drivers/i2c/chips/ds1337.c
 delete mode 100644 drivers/i2c/chips/ds1374.c
 delete mode 100644 drivers/i2c/chips/m41t00.c
 delete mode 100644 include/linux/m41t00.h

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 9b8291f4c211..2cd16deb7195 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -266,13 +266,6 @@ Who:  Tejun Heo <htejun@gmail.com>
 
 ---------------------------
 
-What:	Legacy RTC drivers (under drivers/i2c/chips)
-When:	November 2007
-Why:	Obsolete. We have a RTC subsystem with better drivers.
-Who:	Jean Delvare <khali@linux-fr.org>
-
----------------------------
-
 What:	iptables SAME target
 When:	1.1. 2008
 Files:	net/ipv4/netfilter/ipt_SAME.c, include/linux/netfilter_ipv4/ipt_SAME.h
diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c
index b84f8df325c4..cb0a7493ff64 100644
--- a/arch/ppc/platforms/83xx/mpc834x_sys.c
+++ b/arch/ppc/platforms/83xx/mpc834x_sys.c
@@ -224,26 +224,6 @@ mpc834x_sys_init_IRQ(void)
 	ipic_set_default_priority();
 }
 
-#if defined(CONFIG_I2C_MPC) && defined(CONFIG_SENSORS_DS1374)
-extern ulong	ds1374_get_rtc_time(void);
-extern int	ds1374_set_rtc_time(ulong);
-
-static int __init
-mpc834x_rtc_hookup(void)
-{
-	struct timespec	tv;
-
-	ppc_md.get_rtc_time = ds1374_get_rtc_time;
-	ppc_md.set_rtc_time = ds1374_set_rtc_time;
-
-	tv.tv_nsec = 0;
-	tv.tv_sec = (ppc_md.get_rtc_time)();
-	do_settimeofday(&tv);
-
-	return 0;
-}
-late_initcall(mpc834x_rtc_hookup);
-#endif
 static __inline__ void
 mpc834x_sys_set_bat(void)
 {
diff --git a/arch/ppc/platforms/85xx/tqm85xx.c b/arch/ppc/platforms/85xx/tqm85xx.c
index 4ee2bd156dc5..27ce389c1224 100644
--- a/arch/ppc/platforms/85xx/tqm85xx.c
+++ b/arch/ppc/platforms/85xx/tqm85xx.c
@@ -258,27 +258,6 @@ int tqm85xx_show_cpuinfo(struct seq_file *m)
 	return 0;
 }
 
-#if defined(CONFIG_I2C) && defined(CONFIG_SENSORS_DS1337)
-extern ulong ds1337_get_rtc_time(void);
-extern int ds1337_set_rtc_time(unsigned long nowtime);
-
-static int __init
-tqm85xx_rtc_hookup(void)
-{
-	struct timespec	tv;
-
-        ppc_md.set_rtc_time = ds1337_set_rtc_time;
-        ppc_md.get_rtc_time = ds1337_get_rtc_time;
-
-	tv.tv_nsec = 0;
-	tv.tv_sec = (ppc_md.get_rtc_time)();
-	do_settimeofday(&tv);
-
-	return 0;
-}
-late_initcall(tqm85xx_rtc_hookup);
-#endif
-
 #ifdef CONFIG_PCI
 /*
  * interrupt routing
diff --git a/arch/ppc/platforms/katana.c b/arch/ppc/platforms/katana.c
index 52f63e6f0856..fe6e88cdb1cd 100644
--- a/arch/ppc/platforms/katana.c
+++ b/arch/ppc/platforms/katana.c
@@ -838,27 +838,6 @@ katana_find_end_of_memory(void)
 	return bdp->bi_memsize;
 }
 
-#if defined(CONFIG_I2C_MV64XXX) && defined(CONFIG_SENSORS_M41T00)
-extern ulong	m41t00_get_rtc_time(void);
-extern int	m41t00_set_rtc_time(ulong);
-
-static int __init
-katana_rtc_hookup(void)
-{
-	struct timespec	tv;
-
-	ppc_md.get_rtc_time = m41t00_get_rtc_time;
-	ppc_md.set_rtc_time = m41t00_set_rtc_time;
-
-	tv.tv_nsec = 0;
-	tv.tv_sec = (ppc_md.get_rtc_time)();
-	do_settimeofday(&tv);
-
-	return 0;
-}
-late_initcall(katana_rtc_hookup);
-#endif
-
 #if defined(CONFIG_SERIAL_TEXT_DEBUG) && defined(CONFIG_SERIAL_MPSC_CONSOLE)
 static void __init
 katana_map_io(void)
diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index 17702b32eab0..fe3008e59467 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -4,32 +4,6 @@
 
 menu "Miscellaneous I2C Chip support"
 
-config SENSORS_DS1337
-	tristate "Dallas DS1337 and DS1339 Real Time Clock (DEPRECATED)"
-	depends on EXPERIMENTAL
-	help
-	  If you say yes here you get support for Dallas Semiconductor
-	  DS1337 and DS1339 real-time clock chips.
-
-	  This driver can also be built as a module.  If so, the module
-	  will be called ds1337.
-
-	  This driver is deprecated and will be dropped soon. Use
-	  rtc-ds1307 instead.
-
-config SENSORS_DS1374
-	tristate "Dallas DS1374 Real Time Clock (DEPRECATED)"
-	depends on EXPERIMENTAL
-	help
-	  If you say yes here you get support for Dallas Semiconductor
-	  DS1374 real-time clock chips.
-
-	  This driver can also be built as a module.  If so, the module
-	  will be called ds1374.
-
-	  This driver is deprecated and will be dropped soon. Use
-	  rtc-ds1374 instead.
-
 config DS1682
 	tristate "Dallas DS1682 Total Elapsed Time Recorder with Alarm"
 	depends on EXPERIMENTAL
@@ -130,18 +104,6 @@ config TPS65010
 	  This driver can also be built as a module.  If so, the module
 	  will be called tps65010.
 
-config SENSORS_M41T00
-	tristate "ST M41T00 RTC chip (DEPRECATED)"
-	depends on PPC32
-	help
-	  If you say yes here you get support for the ST M41T00 RTC chip.
-
-	  This driver can also be built as a module.  If so, the module
-	  will be called m41t00.
-
-	  This driver is deprecated and will be dropped soon. Use
-	  rtc-ds1307 or rtc-m41t80 instead.
-
 config SENSORS_MAX6875
 	tristate "Maxim MAX6875 Power supply supervisor"
 	depends on EXPERIMENTAL
diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile
index 50d4734b5038..501f00cea782 100644
--- a/drivers/i2c/chips/Makefile
+++ b/drivers/i2c/chips/Makefile
@@ -2,12 +2,9 @@
 # Makefile for miscellaneous I2C chip drivers.
 #
 
-obj-$(CONFIG_SENSORS_DS1337)	+= ds1337.o
-obj-$(CONFIG_SENSORS_DS1374)	+= ds1374.o
 obj-$(CONFIG_DS1682)		+= ds1682.o
 obj-$(CONFIG_SENSORS_EEPROM)	+= eeprom.o
 obj-$(CONFIG_SENSORS_MAX6875)	+= max6875.o
-obj-$(CONFIG_SENSORS_M41T00)	+= m41t00.o
 obj-$(CONFIG_SENSORS_PCA9539)	+= pca9539.o
 obj-$(CONFIG_SENSORS_PCF8574)	+= pcf8574.o
 obj-$(CONFIG_PCF8575)		+= pcf8575.o
diff --git a/drivers/i2c/chips/ds1337.c b/drivers/i2c/chips/ds1337.c
deleted file mode 100644
index ec17d6b684a2..000000000000
--- a/drivers/i2c/chips/ds1337.c
+++ /dev/null
@@ -1,410 +0,0 @@
-/*
- *  linux/drivers/i2c/chips/ds1337.c
- *
- *  Copyright (C) 2005 James Chapman <jchapman@katalix.com>
- *
- *	based on linux/drivers/acorn/char/pcf8583.c
- *  Copyright (C) 2000 Russell King
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * Driver for Dallas Semiconductor DS1337 and DS1339 real time clock chip
- */
-
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/i2c.h>
-#include <linux/string.h>
-#include <linux/rtc.h>		/* get the user-level API */
-#include <linux/bcd.h>
-#include <linux/list.h>
-
-/* Device registers */
-#define DS1337_REG_HOUR		2
-#define DS1337_REG_DAY		3
-#define DS1337_REG_DATE		4
-#define DS1337_REG_MONTH	5
-#define DS1337_REG_CONTROL	14
-#define DS1337_REG_STATUS	15
-
-/* FIXME - how do we export these interface constants? */
-#define DS1337_GET_DATE		0
-#define DS1337_SET_DATE		1
-
-/*
- * Functions declaration
- */
-static unsigned short normal_i2c[] = { 0x68, I2C_CLIENT_END };
-
-I2C_CLIENT_INSMOD_1(ds1337);
-
-static int ds1337_attach_adapter(struct i2c_adapter *adapter);
-static int ds1337_detect(struct i2c_adapter *adapter, int address, int kind);
-static void ds1337_init_client(struct i2c_client *client);
-static int ds1337_detach_client(struct i2c_client *client);
-static int ds1337_command(struct i2c_client *client, unsigned int cmd,
-			  void *arg);
-
-/*
- * Driver data (common to all clients)
- */
-static struct i2c_driver ds1337_driver = {
-	.driver = {
-		.name	= "ds1337",
-	},
-	.attach_adapter	= ds1337_attach_adapter,
-	.detach_client	= ds1337_detach_client,
-	.command	= ds1337_command,
-};
-
-/*
- * Client data (each client gets its own)
- */
-struct ds1337_data {
-	struct i2c_client client;
-	struct list_head list;
-};
-
-/*
- * Internal variables
- */
-static LIST_HEAD(ds1337_clients);
-
-static inline int ds1337_read(struct i2c_client *client, u8 reg, u8 *value)
-{
-	s32 tmp = i2c_smbus_read_byte_data(client, reg);
-
-	if (tmp < 0)
-		return -EIO;
-
-	*value = tmp;
-
-	return 0;
-}
-
-/*
- * Chip access functions
- */
-static int ds1337_get_datetime(struct i2c_client *client, struct rtc_time *dt)
-{
-	int result;
-	u8 buf[7];
-	u8 val;
-	struct i2c_msg msg[2];
-	u8 offs = 0;
-
-	if (!dt) {
-		dev_dbg(&client->dev, "%s: EINVAL: dt=NULL\n", __FUNCTION__);
-		return -EINVAL;
-	}
-
-	msg[0].addr = client->addr;
-	msg[0].flags = 0;
-	msg[0].len = 1;
-	msg[0].buf = &offs;
-
-	msg[1].addr = client->addr;
-	msg[1].flags = I2C_M_RD;
-	msg[1].len = sizeof(buf);
-	msg[1].buf = &buf[0];
-
-	result = i2c_transfer(client->adapter, msg, 2);
-
-	dev_dbg(&client->dev, "%s: [%d] %02x %02x %02x %02x %02x %02x %02x\n",
-		__FUNCTION__, result, buf[0], buf[1], buf[2], buf[3],
-		buf[4], buf[5], buf[6]);
-
-	if (result == 2) {
-		dt->tm_sec = BCD2BIN(buf[0]);
-		dt->tm_min = BCD2BIN(buf[1]);
-		val = buf[2] & 0x3f;
-		dt->tm_hour = BCD2BIN(val);
-		dt->tm_wday = BCD2BIN(buf[3]) - 1;
-		dt->tm_mday = BCD2BIN(buf[4]);
-		val = buf[5] & 0x7f;
-		dt->tm_mon = BCD2BIN(val) - 1;
-		dt->tm_year = BCD2BIN(buf[6]);
-		if (buf[5] & 0x80)
-			dt->tm_year += 100;
-
-		dev_dbg(&client->dev, "%s: secs=%d, mins=%d, "
-			"hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n",
-			__FUNCTION__, dt->tm_sec, dt->tm_min,
-			dt->tm_hour, dt->tm_mday,
-			dt->tm_mon, dt->tm_year, dt->tm_wday);
-
-		return 0;
-	}
-
-	dev_err(&client->dev, "error reading data! %d\n", result);
-	return -EIO;
-}
-
-static int ds1337_set_datetime(struct i2c_client *client, struct rtc_time *dt)
-{
-	int result;
-	u8 buf[8];
-	u8 val;
-	struct i2c_msg msg[1];
-
-	if (!dt) {
-		dev_dbg(&client->dev, "%s: EINVAL: dt=NULL\n", __FUNCTION__);
-		return -EINVAL;
-	}
-
-	dev_dbg(&client->dev, "%s: secs=%d, mins=%d, hours=%d, "
-		"mday=%d, mon=%d, year=%d, wday=%d\n", __FUNCTION__,
-		dt->tm_sec, dt->tm_min, dt->tm_hour,
-		dt->tm_mday, dt->tm_mon, dt->tm_year, dt->tm_wday);
-
-	buf[0] = 0;		/* reg offset */
-	buf[1] = BIN2BCD(dt->tm_sec);
-	buf[2] = BIN2BCD(dt->tm_min);
-	buf[3] = BIN2BCD(dt->tm_hour);
-	buf[4] = BIN2BCD(dt->tm_wday + 1);
-	buf[5] = BIN2BCD(dt->tm_mday);
-	buf[6] = BIN2BCD(dt->tm_mon + 1);
-	val = dt->tm_year;
-	if (val >= 100) {
-		val -= 100;
-		buf[6] |= (1 << 7);
-	}
-	buf[7] = BIN2BCD(val);
-
-	msg[0].addr = client->addr;
-	msg[0].flags = 0;
-	msg[0].len = sizeof(buf);
-	msg[0].buf = &buf[0];
-
-	result = i2c_transfer(client->adapter, msg, 1);
-	if (result == 1)
-		return 0;
-
-	dev_err(&client->dev, "error writing data! %d\n", result);
-	return -EIO;
-}
-
-static int ds1337_command(struct i2c_client *client, unsigned int cmd,
-			  void *arg)
-{
-	dev_dbg(&client->dev, "%s: cmd=%d\n", __FUNCTION__, cmd);
-
-	switch (cmd) {
-	case DS1337_GET_DATE:
-		return ds1337_get_datetime(client, arg);
-
-	case DS1337_SET_DATE:
-		return ds1337_set_datetime(client, arg);
-
-	default:
-		return -EINVAL;
-	}
-}
-
-/*
- * Public API for access to specific device. Useful for low-level
- * RTC access from kernel code.
- */
-int ds1337_do_command(int bus, int cmd, void *arg)
-{
-	struct list_head *walk;
-	struct list_head *tmp;
-	struct ds1337_data *data;
-
-	list_for_each_safe(walk, tmp, &ds1337_clients) {
-		data = list_entry(walk, struct ds1337_data, list);
-		if (data->client.adapter->nr == bus)
-			return ds1337_command(&data->client, cmd, arg);
-	}
-
-	return -ENODEV;
-}
-
-static int ds1337_attach_adapter(struct i2c_adapter *adapter)
-{
-	return i2c_probe(adapter, &addr_data, ds1337_detect);
-}
-
-/*
- * The following function does more than just detection. If detection
- * succeeds, it also registers the new chip.
- */
-static int ds1337_detect(struct i2c_adapter *adapter, int address, int kind)
-{
-	struct i2c_client *new_client;
-	struct ds1337_data *data;
-	int err = 0;
-	const char *name = "";
-
-	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA |
-				     I2C_FUNC_I2C))
-		goto exit;
-
-	if (!(data = kzalloc(sizeof(struct ds1337_data), GFP_KERNEL))) {
-		err = -ENOMEM;
-		goto exit;
-	}
-	INIT_LIST_HEAD(&data->list);
-
-	/* The common I2C client data is placed right before the
-	 * DS1337-specific data. 
-	 */
-	new_client = &data->client;
-	i2c_set_clientdata(new_client, data);
-	new_client->addr = address;
-	new_client->adapter = adapter;
-	new_client->driver = &ds1337_driver;
-	new_client->flags = 0;
-
-	/*
-	 * Now we do the remaining detection. A negative kind means that
-	 * the driver was loaded with no force parameter (default), so we
-	 * must both detect and identify the chip. A zero kind means that
-	 * the driver was loaded with the force parameter, the detection
-	 * step shall be skipped. A positive kind means that the driver
-	 * was loaded with the force parameter and a given kind of chip is
-	 * requested, so both the detection and the identification steps
-	 * are skipped.
-	 *
-	 * For detection, we read registers that are most likely to cause
-	 * detection failure, i.e. those that have more bits with fixed
-	 * or reserved values.
-	 */
-
-	/* Default to an DS1337 if forced */
-	if (kind == 0)
-		kind = ds1337;
-
-	if (kind < 0) {		/* detection and identification */
-		u8 data;
-
-		/* Check that status register bits 6-2 are zero */
-		if ((ds1337_read(new_client, DS1337_REG_STATUS, &data) < 0) ||
-		    (data & 0x7c))
-			goto exit_free;
-
-		/* Check for a valid day register value */
-		if ((ds1337_read(new_client, DS1337_REG_DAY, &data) < 0) ||
-		    (data == 0) || (data & 0xf8))
-			goto exit_free;
-
-		/* Check for a valid date register value */
-		if ((ds1337_read(new_client, DS1337_REG_DATE, &data) < 0) ||
-		    (data == 0) || (data & 0xc0) || ((data & 0x0f) > 9) ||
-		    (data >= 0x32))
-			goto exit_free;
-
-		/* Check for a valid month register value */
-		if ((ds1337_read(new_client, DS1337_REG_MONTH, &data) < 0) ||
-		    (data == 0) || (data & 0x60) || ((data & 0x0f) > 9) ||
-		    ((data >= 0x13) && (data <= 0x19)))
-			goto exit_free;
-
-		/* Check that control register bits 6-5 are zero */
-		if ((ds1337_read(new_client, DS1337_REG_CONTROL, &data) < 0) ||
-		    (data & 0x60))
-			goto exit_free;
-
-		kind = ds1337;
-	}
-
-	if (kind == ds1337)
-		name = "ds1337";
-
-	/* We can fill in the remaining client fields */
-	strlcpy(new_client->name, name, I2C_NAME_SIZE);
-
-	/* Tell the I2C layer a new client has arrived */
-	if ((err = i2c_attach_client(new_client)))
-		goto exit_free;
-
-	/* Initialize the DS1337 chip */
-	ds1337_init_client(new_client);
-
-	/* Add client to local list */
-	list_add(&data->list, &ds1337_clients);
-
-	return 0;
-
-exit_free:
-	kfree(data);
-exit:
-	return err;
-}
-
-static void ds1337_init_client(struct i2c_client *client)
-{
-	u8 status, control;
-
-	/* On some boards, the RTC isn't configured by boot firmware.
-	 * Handle that case by starting/configuring the RTC now.
-	 */
-	status = i2c_smbus_read_byte_data(client, DS1337_REG_STATUS);
-	control = i2c_smbus_read_byte_data(client, DS1337_REG_CONTROL);
-
-	if ((status & 0x80) || (control & 0x80)) {
-		/* RTC not running */
-		u8 buf[1+16];	/* First byte is interpreted as address */
-		struct i2c_msg msg[1];
-
-		dev_dbg(&client->dev, "%s: RTC not running!\n", __FUNCTION__);
-
-		/* Initialize all, including STATUS and CONTROL to zero */
-		memset(buf, 0, sizeof(buf));
-
-		/* Write valid values in the date/time registers */
-		buf[1+DS1337_REG_DAY] = 1;
-		buf[1+DS1337_REG_DATE] = 1;
-		buf[1+DS1337_REG_MONTH] = 1;
-
-		msg[0].addr = client->addr;
-		msg[0].flags = 0;
-		msg[0].len = sizeof(buf);
-		msg[0].buf = &buf[0];
-
-		i2c_transfer(client->adapter, msg, 1);
-	} else {
-		/* Running: ensure that device is set in 24-hour mode */
-		s32 val;
-
-		val = i2c_smbus_read_byte_data(client, DS1337_REG_HOUR);
-		if ((val >= 0) && (val & (1 << 6)))
-			i2c_smbus_write_byte_data(client, DS1337_REG_HOUR,
-						  val & 0x3f);
-	}
-}
-
-static int ds1337_detach_client(struct i2c_client *client)
-{
-	int err;
-	struct ds1337_data *data = i2c_get_clientdata(client);
-
-	if ((err = i2c_detach_client(client)))
-		return err;
-
-	list_del(&data->list);
-	kfree(data);
-	return 0;
-}
-
-static int __init ds1337_init(void)
-{
-	return i2c_add_driver(&ds1337_driver);
-}
-
-static void __exit ds1337_exit(void)
-{
-	i2c_del_driver(&ds1337_driver);
-}
-
-MODULE_AUTHOR("James Chapman <jchapman@katalix.com>");
-MODULE_DESCRIPTION("DS1337 RTC driver");
-MODULE_LICENSE("GPL");
-
-EXPORT_SYMBOL_GPL(ds1337_do_command);
-
-module_init(ds1337_init);
-module_exit(ds1337_exit);
diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c
deleted file mode 100644
index 8a2ff0c114d9..000000000000
--- a/drivers/i2c/chips/ds1374.c
+++ /dev/null
@@ -1,267 +0,0 @@
-/*
- * drivers/i2c/chips/ds1374.c
- *
- * I2C client/driver for the Maxim/Dallas DS1374 Real-Time Clock
- *
- * Author: Randy Vinson <rvinson@mvista.com>
- *
- * Based on the m41t00.c by Mark Greer <mgreer@mvista.com>
- *
- * 2005 (c) MontaVista Software, Inc. This file is licensed under
- * the terms of the GNU General Public License version 2. This program
- * is licensed "as is" without any warranty of any kind, whether express
- * or implied.
- */
-/*
- * This i2c client/driver wedges between the drivers/char/genrtc.c RTC
- * interface and the SMBus interface of the i2c subsystem.
- * It would be more efficient to use i2c msgs/i2c_transfer directly but, as
- * recommened in .../Documentation/i2c/writing-clients section
- * "Sending and receiving", using SMBus level communication is preferred.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/interrupt.h>
-#include <linux/i2c.h>
-#include <linux/rtc.h>
-#include <linux/bcd.h>
-#include <linux/mutex.h>
-#include <linux/workqueue.h>
-
-#define DS1374_REG_TOD0		0x00
-#define DS1374_REG_TOD1		0x01
-#define DS1374_REG_TOD2		0x02
-#define DS1374_REG_TOD3		0x03
-#define DS1374_REG_WDALM0	0x04
-#define DS1374_REG_WDALM1	0x05
-#define DS1374_REG_WDALM2	0x06
-#define DS1374_REG_CR		0x07
-#define DS1374_REG_SR		0x08
-#define DS1374_REG_SR_OSF	0x80
-#define DS1374_REG_TCR		0x09
-
-#define	DS1374_DRV_NAME		"ds1374"
-
-static DEFINE_MUTEX(ds1374_mutex);
-
-static struct i2c_driver ds1374_driver;
-static struct i2c_client *save_client;
-
-static unsigned short ignore[] = { I2C_CLIENT_END };
-static unsigned short normal_addr[] = { 0x68, I2C_CLIENT_END };
-
-static struct i2c_client_address_data addr_data = {
-	.normal_i2c = normal_addr,
-	.probe = ignore,
-	.ignore = ignore,
-};
-
-static ulong ds1374_read_rtc(void)
-{
-	ulong time = 0;
-	int reg = DS1374_REG_WDALM0;
-
-	while (reg--) {
-		s32 tmp;
-		if ((tmp = i2c_smbus_read_byte_data(save_client, reg)) < 0) {
-			dev_warn(&save_client->dev,
-				 "can't read from rtc chip\n");
-			return 0;
-		}
-		time = (time << 8) | (tmp & 0xff);
-	}
-	return time;
-}
-
-static void ds1374_write_rtc(ulong time)
-{
-	int reg;
-
-	for (reg = DS1374_REG_TOD0; reg < DS1374_REG_WDALM0; reg++) {
-		if (i2c_smbus_write_byte_data(save_client, reg, time & 0xff)
-		    < 0) {
-			dev_warn(&save_client->dev,
-				 "can't write to rtc chip\n");
-			break;
-		}
-		time = time >> 8;
-	}
-}
-
-static void ds1374_check_rtc_status(void)
-{
-	s32 tmp;
-
-	tmp = i2c_smbus_read_byte_data(save_client, DS1374_REG_SR);
-	if (tmp < 0) {
-		dev_warn(&save_client->dev,
-			 "can't read status from rtc chip\n");
-		return;
-	}
-	if (tmp & DS1374_REG_SR_OSF) {
-		dev_warn(&save_client->dev,
-			 "oscillator discontinuity flagged, time unreliable\n");
-		tmp &= ~DS1374_REG_SR_OSF;
-		tmp = i2c_smbus_write_byte_data(save_client, DS1374_REG_SR,
-						tmp & 0xff);
-		if (tmp < 0)
-			dev_warn(&save_client->dev,
-				 "can't clear discontinuity notification\n");
-	}
-}
-
-ulong ds1374_get_rtc_time(void)
-{
-	ulong t1, t2;
-	int limit = 10;		/* arbitrary retry limit */
-
-	mutex_lock(&ds1374_mutex);
-
-	/*
-	 * Since the reads are being performed one byte at a time using
-	 * the SMBus vs a 4-byte i2c transfer, there is a chance that a
-	 * carry will occur during the read. To detect this, 2 reads are
-	 * performed and compared.
-	 */
-	do {
-		t1 = ds1374_read_rtc();
-		t2 = ds1374_read_rtc();
-	} while (t1 != t2 && limit--);
-
-	mutex_unlock(&ds1374_mutex);
-
-	if (t1 != t2) {
-		dev_warn(&save_client->dev,
-			 "can't get consistent time from rtc chip\n");
-		t1 = 0;
-	}
-
-	return t1;
-}
-
-static ulong new_time;
-
-static void ds1374_set_work(struct work_struct *work)
-{
-	ulong t1, t2;
-	int limit = 10;		/* arbitrary retry limit */
-
-	t1 = new_time;
-
-	mutex_lock(&ds1374_mutex);
-
-	/*
-	 * Since the writes are being performed one byte at a time using
-	 * the SMBus vs a 4-byte i2c transfer, there is a chance that a
-	 * carry will occur during the write. To detect this, the write
-	 * value is read back and compared.
-	 */
-	do {
-		ds1374_write_rtc(t1);
-		t2 = ds1374_read_rtc();
-	} while (t1 != t2 && limit--);
-
-	mutex_unlock(&ds1374_mutex);
-
-	if (t1 != t2)
-		dev_warn(&save_client->dev,
-			 "can't confirm time set from rtc chip\n");
-}
-
-static struct workqueue_struct *ds1374_workqueue;
-
-static DECLARE_WORK(ds1374_work, ds1374_set_work);
-
-int ds1374_set_rtc_time(ulong nowtime)
-{
-	new_time = nowtime;
-
-	if (in_interrupt())
-		queue_work(ds1374_workqueue, &ds1374_work);
-	else
-		ds1374_set_work(NULL);
-
-	return 0;
-}
-
-/*
- *****************************************************************************
- *
- *	Driver Interface
- *
- *****************************************************************************
- */
-static int ds1374_probe(struct i2c_adapter *adap, int addr, int kind)
-{
-	struct i2c_client *client;
-	int rc;
-
-	client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL);
-	if (!client)
-		return -ENOMEM;
-
-	strncpy(client->name, DS1374_DRV_NAME, I2C_NAME_SIZE);
-	client->addr = addr;
-	client->adapter = adap;
-	client->driver = &ds1374_driver;
-
-	ds1374_workqueue = create_singlethread_workqueue("ds1374");
-	if (!ds1374_workqueue) {
-		kfree(client);
-		return -ENOMEM;	/* most expected reason */
-	}
-
-	if ((rc = i2c_attach_client(client)) != 0) {
-		kfree(client);
-		return rc;
-	}
-
-	save_client = client;
-
-	ds1374_check_rtc_status();
-
-	return 0;
-}
-
-static int ds1374_attach(struct i2c_adapter *adap)
-{
-	return i2c_probe(adap, &addr_data, ds1374_probe);
-}
-
-static int ds1374_detach(struct i2c_client *client)
-{
-	int rc;
-
-	if ((rc = i2c_detach_client(client)) == 0) {
-		kfree(i2c_get_clientdata(client));
-		destroy_workqueue(ds1374_workqueue);
-	}
-	return rc;
-}
-
-static struct i2c_driver ds1374_driver = {
-	.driver = {
-		.name	= DS1374_DRV_NAME,
-	},
-	.id = I2C_DRIVERID_DS1374,
-	.attach_adapter = ds1374_attach,
-	.detach_client = ds1374_detach,
-};
-
-static int __init ds1374_init(void)
-{
-	return i2c_add_driver(&ds1374_driver);
-}
-
-static void __exit ds1374_exit(void)
-{
-	i2c_del_driver(&ds1374_driver);
-}
-
-module_init(ds1374_init);
-module_exit(ds1374_exit);
-
-MODULE_AUTHOR("Randy Vinson <rvinson@mvista.com>");
-MODULE_DESCRIPTION("Maxim/Dallas DS1374 RTC I2C Client Driver");
-MODULE_LICENSE("GPL");
diff --git a/drivers/i2c/chips/m41t00.c b/drivers/i2c/chips/m41t00.c
deleted file mode 100644
index 3fcb646e2073..000000000000
--- a/drivers/i2c/chips/m41t00.c
+++ /dev/null
@@ -1,413 +0,0 @@
-/*
- * I2C client/driver for the ST M41T00 family of i2c rtc chips.
- *
- * Author: Mark A. Greer <mgreer@mvista.com>
- *
- * 2005, 2006 (c) MontaVista Software, Inc. This file is licensed under
- * the terms of the GNU General Public License version 2. This program
- * is licensed "as is" without any warranty of any kind, whether express
- * or implied.
- */
-/*
- * This i2c client/driver wedges between the drivers/char/genrtc.c RTC
- * interface and the SMBus interface of the i2c subsystem.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/interrupt.h>
-#include <linux/i2c.h>
-#include <linux/rtc.h>
-#include <linux/bcd.h>
-#include <linux/workqueue.h>
-#include <linux/platform_device.h>
-#include <linux/m41t00.h>
-#include <asm/time.h>
-#include <asm/rtc.h>
-
-static struct i2c_driver m41t00_driver;
-static struct i2c_client *save_client;
-
-static unsigned short ignore[] = { I2C_CLIENT_END };
-static unsigned short normal_addr[] = { I2C_CLIENT_END, I2C_CLIENT_END };
-
-static struct i2c_client_address_data addr_data = {
-	.normal_i2c	= normal_addr,
-	.probe		= ignore,
-	.ignore		= ignore,
-};
-
-struct m41t00_chip_info {
-	u8	type;
-	char	*name;
-	u8	read_limit;
-	u8	sec;		/* Offsets for chip regs */
-	u8	min;
-	u8	hour;
-	u8	day;
-	u8	mon;
-	u8	year;
-	u8	alarm_mon;
-	u8	alarm_hour;
-	u8	sqw;
-	u8	sqw_freq;
-};
-
-static struct m41t00_chip_info m41t00_chip_info_tbl[] = {
-	{
-		.type		= M41T00_TYPE_M41T00,
-		.name		= "m41t00",
-		.read_limit	= 5,
-		.sec		= 0,
-		.min		= 1,
-		.hour		= 2,
-		.day		= 4,
-		.mon		= 5,
-		.year		= 6,
-	},
-	{
-		.type		= M41T00_TYPE_M41T81,
-		.name		= "m41t81",
-		.read_limit	= 1,
-		.sec		= 1,
-		.min		= 2,
-		.hour		= 3,
-		.day		= 5,
-		.mon		= 6,
-		.year		= 7,
-		.alarm_mon	= 0xa,
-		.alarm_hour	= 0xc,
-		.sqw		= 0x13,
-	},
-	{
-		.type		= M41T00_TYPE_M41T85,
-		.name		= "m41t85",
-		.read_limit	= 1,
-		.sec		= 1,
-		.min		= 2,
-		.hour		= 3,
-		.day		= 5,
-		.mon		= 6,
-		.year		= 7,
-		.alarm_mon	= 0xa,
-		.alarm_hour	= 0xc,
-		.sqw		= 0x13,
-	},
-};
-static struct m41t00_chip_info *m41t00_chip;
-
-ulong
-m41t00_get_rtc_time(void)
-{
-	s32 sec, min, hour, day, mon, year;
-	s32 sec1, min1, hour1, day1, mon1, year1;
-	u8 reads = 0;
-	u8 buf[8], msgbuf[1] = { 0 }; /* offset into rtc's regs */
-	struct i2c_msg msgs[] = {
-		{
-			.addr	= save_client->addr,
-			.flags	= 0,
-			.len	= 1,
-			.buf	= msgbuf,
-		},
-		{
-			.addr	= save_client->addr,
-			.flags	= I2C_M_RD,
-			.len	= 8,
-			.buf	= buf,
-		},
-	};
-
-	sec = min = hour = day = mon = year = 0;
-
-	do {
-		if (i2c_transfer(save_client->adapter, msgs, 2) < 0)
-			goto read_err;
-
-		sec1 = sec;
-		min1 = min;
-		hour1 = hour;
-		day1 = day;
-		mon1 = mon;
-		year1 = year;
-
-		sec = buf[m41t00_chip->sec] & 0x7f;
-		min = buf[m41t00_chip->min] & 0x7f;
-		hour = buf[m41t00_chip->hour] & 0x3f;
-		day = buf[m41t00_chip->day] & 0x3f;
-		mon = buf[m41t00_chip->mon] & 0x1f;
-		year = buf[m41t00_chip->year];
-	} while ((++reads < m41t00_chip->read_limit) && ((sec != sec1)
-			|| (min != min1) || (hour != hour1) || (day != day1)
-			|| (mon != mon1) || (year != year1)));
-
-	if ((m41t00_chip->read_limit > 1) && ((sec != sec1) || (min != min1)
-			|| (hour != hour1) || (day != day1) || (mon != mon1)
-			|| (year != year1)))
-		goto read_err;
-
-	sec = BCD2BIN(sec);
-	min = BCD2BIN(min);
-	hour = BCD2BIN(hour);
-	day = BCD2BIN(day);
-	mon = BCD2BIN(mon);
-	year = BCD2BIN(year);
-
-	year += 1900;
-	if (year < 1970)
-		year += 100;
-
-	return mktime(year, mon, day, hour, min, sec);
-
-read_err:
-	dev_err(&save_client->dev, "m41t00_get_rtc_time: Read error\n");
-	return 0;
-}
-EXPORT_SYMBOL_GPL(m41t00_get_rtc_time);
-
-static void
-m41t00_set(void *arg)
-{
-	struct rtc_time	tm;
-	int nowtime = *(int *)arg;
-	s32 sec, min, hour, day, mon, year;
-	u8 wbuf[9], *buf = &wbuf[1], msgbuf[1] = { 0 };
-	struct i2c_msg msgs[] = {
-		{
-			.addr	= save_client->addr,
-			.flags	= 0,
-			.len	= 1,
-			.buf	= msgbuf,
-		},
-		{
-			.addr	= save_client->addr,
-			.flags	= I2C_M_RD,
-			.len	= 8,
-			.buf	= buf,
-		},
-	};
-
-	to_tm(nowtime, &tm);
-	tm.tm_year = (tm.tm_year - 1900) % 100;
-
-	sec = BIN2BCD(tm.tm_sec);
-	min = BIN2BCD(tm.tm_min);
-	hour = BIN2BCD(tm.tm_hour);
-	day = BIN2BCD(tm.tm_mday);
-	mon = BIN2BCD(tm.tm_mon);
-	year = BIN2BCD(tm.tm_year);
-
-	/* Read reg values into buf[0..7]/wbuf[1..8] */
-	if (i2c_transfer(save_client->adapter, msgs, 2) < 0) {
-		dev_err(&save_client->dev, "m41t00_set: Read error\n");
-		return;
-	}
-
-	wbuf[0] = 0; /* offset into rtc's regs */
-	buf[m41t00_chip->sec] = (buf[m41t00_chip->sec] & ~0x7f) | (sec & 0x7f);
-	buf[m41t00_chip->min] = (buf[m41t00_chip->min] & ~0x7f) | (min & 0x7f);
-	buf[m41t00_chip->hour] = (buf[m41t00_chip->hour] & ~0x3f) | (hour& 0x3f);
-	buf[m41t00_chip->day] = (buf[m41t00_chip->day] & ~0x3f) | (day & 0x3f);
-	buf[m41t00_chip->mon] = (buf[m41t00_chip->mon] & ~0x1f) | (mon & 0x1f);
-	buf[m41t00_chip->year] = year;
-
-	if (i2c_master_send(save_client, wbuf, 9) < 0)
-		dev_err(&save_client->dev, "m41t00_set: Write error\n");
-}
-
-static ulong new_time;
-/* well, isn't this API just _lovely_? */
-static void
-m41t00_barf(struct work_struct *unusable)
-{
-	m41t00_set(&new_time);
-}
-
-static struct workqueue_struct *m41t00_wq;
-static DECLARE_WORK(m41t00_work, m41t00_barf);
-
-int
-m41t00_set_rtc_time(ulong nowtime)
-{
-	new_time = nowtime;
-
-	if (in_interrupt())
-		queue_work(m41t00_wq, &m41t00_work);
-	else
-		m41t00_set(&new_time);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(m41t00_set_rtc_time);
-
-/*
- *****************************************************************************
- *
- *	platform_data Driver Interface
- *
- *****************************************************************************
- */
-static int __init
-m41t00_platform_probe(struct platform_device *pdev)
-{
-	struct m41t00_platform_data *pdata;
-	int i;
-
-	if (pdev && (pdata = pdev->dev.platform_data)) {
-		normal_addr[0] = pdata->i2c_addr;
-
-		for (i=0; i<ARRAY_SIZE(m41t00_chip_info_tbl); i++)
-			if (m41t00_chip_info_tbl[i].type == pdata->type) {
-				m41t00_chip = &m41t00_chip_info_tbl[i];
-				m41t00_chip->sqw_freq = pdata->sqw_freq;
-				return 0;
-			}
-	}
-	return -ENODEV;
-}
-
-static int __exit
-m41t00_platform_remove(struct platform_device *pdev)
-{
-	return 0;
-}
-
-static struct platform_driver m41t00_platform_driver = {
-	.probe  = m41t00_platform_probe,
-	.remove = m41t00_platform_remove,
-	.driver = {
-		.owner = THIS_MODULE,
-		.name  = M41T00_DRV_NAME,
-	},
-};
-
-/*
- *****************************************************************************
- *
- *	Driver Interface
- *
- *****************************************************************************
- */
-static int
-m41t00_probe(struct i2c_adapter *adap, int addr, int kind)
-{
-	struct i2c_client *client;
-	int rc;
-
-	if (!i2c_check_functionality(adap, I2C_FUNC_I2C
-			| I2C_FUNC_SMBUS_BYTE_DATA))
-		return 0;
-
-	client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL);
-	if (!client)
-		return -ENOMEM;
-
-	strlcpy(client->name, m41t00_chip->name, I2C_NAME_SIZE);
-	client->addr = addr;
-	client->adapter = adap;
-	client->driver = &m41t00_driver;
-
-	if ((rc = i2c_attach_client(client)))
-		goto attach_err;
-
-	if (m41t00_chip->type != M41T00_TYPE_M41T00) {
-		/* If asked, disable SQW, set SQW frequency & re-enable */
-		if (m41t00_chip->sqw_freq)
-			if (((rc = i2c_smbus_read_byte_data(client,
-					m41t00_chip->alarm_mon)) < 0)
-			 || ((rc = i2c_smbus_write_byte_data(client,
-					m41t00_chip->alarm_mon, rc & ~0x40)) <0)
-			 || ((rc = i2c_smbus_write_byte_data(client,
-					m41t00_chip->sqw,
-					m41t00_chip->sqw_freq)) < 0)
-			 || ((rc = i2c_smbus_write_byte_data(client,
-					m41t00_chip->alarm_mon, rc | 0x40)) <0))
-				goto sqw_err;
-
-		/* Make sure HT (Halt Update) bit is cleared */
-		if ((rc = i2c_smbus_read_byte_data(client,
-				m41t00_chip->alarm_hour)) < 0)
-			goto ht_err;
-
-		if (rc & 0x40)
-			if ((rc = i2c_smbus_write_byte_data(client,
-					m41t00_chip->alarm_hour, rc & ~0x40))<0)
-				goto ht_err;
-	}
-
-	/* Make sure ST (stop) bit is cleared */
-	if ((rc = i2c_smbus_read_byte_data(client, m41t00_chip->sec)) < 0)
-		goto st_err;
-
-	if (rc & 0x80)
-		if ((rc = i2c_smbus_write_byte_data(client, m41t00_chip->sec,
-				rc & ~0x80)) < 0)
-			goto st_err;
-
-	m41t00_wq = create_singlethread_workqueue(m41t00_chip->name);
-	save_client = client;
-	return 0;
-
-st_err:
-	dev_err(&client->dev, "m41t00_probe: Can't clear ST bit\n");
-	goto attach_err;
-ht_err:
-	dev_err(&client->dev, "m41t00_probe: Can't clear HT bit\n");
-	goto attach_err;
-sqw_err:
-	dev_err(&client->dev, "m41t00_probe: Can't set SQW Frequency\n");
-attach_err:
-	kfree(client);
-	return rc;
-}
-
-static int
-m41t00_attach(struct i2c_adapter *adap)
-{
-	return i2c_probe(adap, &addr_data, m41t00_probe);
-}
-
-static int
-m41t00_detach(struct i2c_client *client)
-{
-	int rc;
-
-	if ((rc = i2c_detach_client(client)) == 0) {
-		kfree(client);
-		destroy_workqueue(m41t00_wq);
-	}
-	return rc;
-}
-
-static struct i2c_driver m41t00_driver = {
-	.driver = {
-		.name	= M41T00_DRV_NAME,
-	},
-	.id		= I2C_DRIVERID_STM41T00,
-	.attach_adapter	= m41t00_attach,
-	.detach_client	= m41t00_detach,
-};
-
-static int __init
-m41t00_init(void)
-{
-	int rc;
-
-	if (!(rc = platform_driver_register(&m41t00_platform_driver)))
-		rc = i2c_add_driver(&m41t00_driver);
-	return rc;
-}
-
-static void __exit
-m41t00_exit(void)
-{
-	i2c_del_driver(&m41t00_driver);
-	platform_driver_unregister(&m41t00_platform_driver);
-}
-
-module_init(m41t00_init);
-module_exit(m41t00_exit);
-
-MODULE_AUTHOR("Mark A. Greer <mgreer@mvista.com>");
-MODULE_DESCRIPTION("ST Microelectronics M41T00 RTC I2C Client Driver");
-MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index c7a51a196f51..10c8edd7795c 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -83,7 +83,6 @@
 #define I2C_DRIVERID_SAA7114	49	/* video decoder		*/
 #define I2C_DRIVERID_ZR36120	50     /* Zoran 36120 video encoder	*/
 #define I2C_DRIVERID_24LC32A	51	/* Microchip 24LC32A 32k EEPROM	*/
-#define I2C_DRIVERID_STM41T00	52	/* real time clock		*/
 #define I2C_DRIVERID_UDA1342	53	/* UDA1342 audio codec		*/
 #define I2C_DRIVERID_ADV7170	54	/* video encoder		*/
 #define I2C_DRIVERID_MAX1617	56	/* temp sensor			*/
@@ -95,7 +94,6 @@
 #define I2C_DRIVERID_TDA7313	62	/* TDA7313 audio processor	*/
 #define I2C_DRIVERID_MAX6900	63	/* MAX6900 real-time clock	*/
 #define I2C_DRIVERID_SAA7114H	64	/* video decoder		*/
-#define I2C_DRIVERID_DS1374	65	/* DS1374 real time clock	*/
 #define I2C_DRIVERID_TDA9874	66	/* TV sound decoder		*/
 #define I2C_DRIVERID_SAA6752HS	67	/* MPEG2 encoder		*/
 #define I2C_DRIVERID_TVEEPROM	68	/* TV EEPROM			*/
diff --git a/include/linux/m41t00.h b/include/linux/m41t00.h
deleted file mode 100644
index b423360ca38e..000000000000
--- a/include/linux/m41t00.h
+++ /dev/null
@@ -1,50 +0,0 @@
-/*
- * Definitions for the ST M41T00 family of i2c rtc chips.
- *
- * Author: Mark A. Greer <mgreer@mvista.com>
- *
- * 2005, 2006 (c) MontaVista Software, Inc. This file is licensed under
- * the terms of the GNU General Public License version 2. This program
- * is licensed "as is" without any warranty of any kind, whether express
- * or implied.
- */
-
-#ifndef _M41T00_H
-#define _M41T00_H
-
-#define	M41T00_DRV_NAME		"m41t00"
-#define	M41T00_I2C_ADDR		0x68
-
-#define	M41T00_TYPE_M41T00	0
-#define	M41T00_TYPE_M41T81	81
-#define	M41T00_TYPE_M41T85	85
-
-struct m41t00_platform_data {
-	u8	type;
-	u8	i2c_addr;
-	u8	sqw_freq;
-};
-
-/* SQW output disabled, this is default value by power on */
-#define M41T00_SQW_DISABLE	(0)
-
-#define M41T00_SQW_32KHZ	(1<<4)		/* 32.768 KHz */
-#define M41T00_SQW_8KHZ		(2<<4)		/* 8.192 KHz */
-#define M41T00_SQW_4KHZ		(3<<4)		/* 4.096 KHz */
-#define M41T00_SQW_2KHZ		(4<<4)		/* 2.048 KHz */
-#define M41T00_SQW_1KHZ		(5<<4)		/* 1.024 KHz */
-#define M41T00_SQW_512HZ	(6<<4)		/* 512 Hz */
-#define M41T00_SQW_256HZ	(7<<4)		/* 256 Hz */
-#define M41T00_SQW_128HZ	(8<<4)		/* 128 Hz */
-#define M41T00_SQW_64HZ		(9<<4)		/* 64 Hz */
-#define M41T00_SQW_32HZ		(10<<4)		/* 32 Hz */
-#define M41T00_SQW_16HZ		(11<<4)		/* 16 Hz */
-#define M41T00_SQW_8HZ		(12<<4)		/* 8 Hz */
-#define M41T00_SQW_4HZ		(13<<4)		/* 4 Hz */
-#define M41T00_SQW_2HZ		(14<<4)		/* 2 Hz */
-#define M41T00_SQW_1HZ		(15<<4)		/* 1 Hz */
-
-extern ulong m41t00_get_rtc_time(void);
-extern int m41t00_set_rtc_time(ulong nowtime);
-
-#endif /* _M41T00_H */
-- 
cgit v1.2.3


From 7e8b99251be8b6f992baa88e3a6ba3c4ae01660b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sun, 27 Jan 2008 18:14:46 +0100
Subject: i2c: some overdue driver removal

This patch contains the overdue removal of three I2C drivers.

[JD: In fact only i2c-ixp4xx can be removed at the moment, the other two
platforms don't implement the generic GPIO layer yet.]

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 Documentation/feature-removal-schedule.txt |   8 --
 drivers/i2c/busses/Kconfig                 |  14 ---
 drivers/i2c/busses/Makefile                |   1 -
 drivers/i2c/busses/i2c-ixp4xx.c            | 178 -----------------------------
 include/asm-arm/arch-ixp4xx/platform.h     |  11 --
 include/linux/i2c-id.h                     |   1 -
 6 files changed, 213 deletions(-)
 delete mode 100644 drivers/i2c/busses/i2c-ixp4xx.c

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 2cd16deb7195..fb51d38e8b46 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -225,14 +225,6 @@ Who:	Len Brown <len.brown@intel.com>
 
 ---------------------------
 
-What:	i2c-ixp2000, i2c-ixp4xx and scx200_i2c drivers
-When:	September 2007
-Why:	Obsolete. The new i2c-gpio driver replaces all hardware-specific
-	I2C-over-GPIO drivers.
-Who:	Jean Delvare <khali@linux-fr.org>
-
----------------------------
-
 What:	'time' kernel boot parameter
 When:	January 2008
 Why:	replaced by 'printk.time=<value>' so that printk timestamps can be
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index c466c6cfc2e5..2cef92a0bd0e 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -259,20 +259,6 @@ config I2C_IOP3XX
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-iop3xx.
 
-config I2C_IXP4XX
-	tristate "IXP4xx GPIO-Based I2C Interface (DEPRECATED)"
-	depends on ARCH_IXP4XX
-	select I2C_ALGOBIT
-	help
-	  Say Y here if you have an Intel IXP4xx(420,421,422,425) based 
-	  system and are using GPIO lines for an I2C bus.
-
-	  This support is also available as a module. If so, the module
-	  will be called i2c-ixp4xx.
-
-	  This driver is deprecated and will be dropped soon. Use i2c-gpio
-	  instead.
-
 config I2C_IXP2000
 	tristate "IXP2000 GPIO-Based I2C Interface (DEPRECATED)"
 	depends on ARCH_IXP2000
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index 81d43c27cf93..ea7068f1eb6b 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -20,7 +20,6 @@ obj-$(CONFIG_I2C_I810)		+= i2c-i810.o
 obj-$(CONFIG_I2C_IBM_IIC)	+= i2c-ibm_iic.o
 obj-$(CONFIG_I2C_IOP3XX)	+= i2c-iop3xx.o
 obj-$(CONFIG_I2C_IXP2000)	+= i2c-ixp2000.o
-obj-$(CONFIG_I2C_IXP4XX)	+= i2c-ixp4xx.o
 obj-$(CONFIG_I2C_POWERMAC)	+= i2c-powermac.o
 obj-$(CONFIG_I2C_MPC)		+= i2c-mpc.o
 obj-$(CONFIG_I2C_MV64XXX)	+= i2c-mv64xxx.o
diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c
deleted file mode 100644
index 069ed7f3b395..000000000000
--- a/drivers/i2c/busses/i2c-ixp4xx.c
+++ /dev/null
@@ -1,178 +0,0 @@
-/*
- * drivers/i2c/busses/i2c-ixp4xx.c
- *
- * Intel's IXP4xx XScale NPU chipsets (IXP420, 421, 422, 425) do not have
- * an on board I2C controller but provide 16 GPIO pins that are often
- * used to create an I2C bus. This driver provides an i2c_adapter 
- * interface that plugs in under algo_bit and drives the GPIO pins
- * as instructed by the alogorithm driver.
- *
- * Author: Deepak Saxena <dsaxena@plexity.net>
- *
- * Copyright (c) 2003-2004 MontaVista Software Inc.
- *
- * This file is licensed under the terms of the GNU General Public 
- * License version 2. This program is licensed "as is" without any 
- * warranty of any kind, whether express or implied.
- *
- * NOTE: Since different platforms will use different GPIO pins for
- *       I2C, this driver uses an IXP4xx-specific platform_data
- *       pointer to pass the GPIO numbers to the driver. This 
- *       allows us to support all the different IXP4xx platforms
- *       w/o having to put #ifdefs in this driver.
- *
- *       See arch/arm/mach-ixp4xx/ixdp425.c for an example of building a 
- *       device list and filling in the ixp4xx_i2c_pins data structure 
- *       that is passed as the platform_data to this driver.
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/platform_device.h>
-#include <linux/module.h>
-#include <linux/i2c.h>
-#include <linux/i2c-algo-bit.h>
-
-#include <asm/hardware.h>	/* Pick up IXP4xx-specific bits */
-
-static inline int ixp4xx_scl_pin(void *data)
-{
-	return ((struct ixp4xx_i2c_pins*)data)->scl_pin;
-}
-
-static inline int ixp4xx_sda_pin(void *data)
-{
-	return ((struct ixp4xx_i2c_pins*)data)->sda_pin;
-}
-
-static void ixp4xx_bit_setscl(void *data, int val)
-{
-	gpio_line_set(ixp4xx_scl_pin(data), 0);
-	gpio_line_config(ixp4xx_scl_pin(data),
-		val ? IXP4XX_GPIO_IN : IXP4XX_GPIO_OUT );
-}
-
-static void ixp4xx_bit_setsda(void *data, int val)
-{
-	gpio_line_set(ixp4xx_sda_pin(data), 0);
-	gpio_line_config(ixp4xx_sda_pin(data),
-		val ? IXP4XX_GPIO_IN : IXP4XX_GPIO_OUT );
-}
-
-static int ixp4xx_bit_getscl(void *data)
-{
-	int scl;
-
-	gpio_line_config(ixp4xx_scl_pin(data), IXP4XX_GPIO_IN );
-	gpio_line_get(ixp4xx_scl_pin(data), &scl);
-
-	return scl;
-}	
-
-static int ixp4xx_bit_getsda(void *data)
-{
-	int sda;
-
-	gpio_line_config(ixp4xx_sda_pin(data), IXP4XX_GPIO_IN );
-	gpio_line_get(ixp4xx_sda_pin(data), &sda);
-
-	return sda;
-}	
-
-struct ixp4xx_i2c_data {
-	struct ixp4xx_i2c_pins *gpio_pins;
-	struct i2c_adapter adapter;
-	struct i2c_algo_bit_data algo_data;
-};
-
-static int ixp4xx_i2c_remove(struct platform_device *plat_dev)
-{
-	struct ixp4xx_i2c_data *drv_data = platform_get_drvdata(plat_dev);
-
-	platform_set_drvdata(plat_dev, NULL);
-
-	i2c_del_adapter(&drv_data->adapter);
-
-	kfree(drv_data);
-
-	return 0;
-}
-
-static int ixp4xx_i2c_probe(struct platform_device *plat_dev)
-{
-	int err;
-	struct ixp4xx_i2c_pins *gpio = plat_dev->dev.platform_data;
-	struct ixp4xx_i2c_data *drv_data = 
-		kzalloc(sizeof(struct ixp4xx_i2c_data), GFP_KERNEL);
-
-	if(!drv_data)
-		return -ENOMEM;
-
-	drv_data->gpio_pins = gpio;
-
-	/*
-	 * We could make a lot of these structures static, but
-	 * certain platforms may have multiple GPIO-based I2C
-	 * buses for various device domains, so we need per-device
-	 * algo_data->data. 
-	 */
-	drv_data->algo_data.data = gpio;
-	drv_data->algo_data.setsda = ixp4xx_bit_setsda;
-	drv_data->algo_data.setscl = ixp4xx_bit_setscl;
-	drv_data->algo_data.getsda = ixp4xx_bit_getsda;
-	drv_data->algo_data.getscl = ixp4xx_bit_getscl;
-	drv_data->algo_data.udelay = 10;
-	drv_data->algo_data.timeout = 100;
-
-	drv_data->adapter.id = I2C_HW_B_IXP4XX;
-	drv_data->adapter.class = I2C_CLASS_HWMON;
-	strlcpy(drv_data->adapter.name, plat_dev->dev.driver->name,
-		sizeof(drv_data->adapter.name));
-	drv_data->adapter.algo_data = &drv_data->algo_data;
-
-	drv_data->adapter.dev.parent = &plat_dev->dev;
-
-	gpio_line_config(gpio->scl_pin, IXP4XX_GPIO_IN);
-	gpio_line_config(gpio->sda_pin, IXP4XX_GPIO_IN);
-	gpio_line_set(gpio->scl_pin, 0);
-	gpio_line_set(gpio->sda_pin, 0);
-
-	err = i2c_bit_add_bus(&drv_data->adapter);
-	if (err) {
-		printk(KERN_ERR "ERROR: Could not install %s\n", plat_dev->dev.bus_id);
-
-		kfree(drv_data);
-		return err;
-	}
-
-	platform_set_drvdata(plat_dev, drv_data);
-
-	return 0;
-}
-
-static struct platform_driver ixp4xx_i2c_driver = {
-	.probe		= ixp4xx_i2c_probe,
-	.remove		= ixp4xx_i2c_remove,
-	.driver		= {
-		.name	= "IXP4XX-I2C",
-		.owner	= THIS_MODULE,
-	},
-};
-
-static int __init ixp4xx_i2c_init(void)
-{
-	return platform_driver_register(&ixp4xx_i2c_driver);
-}
-
-static void __exit ixp4xx_i2c_exit(void)
-{
-	platform_driver_unregister(&ixp4xx_i2c_driver);
-}
-
-module_init(ixp4xx_i2c_init);
-module_exit(ixp4xx_i2c_exit);
-
-MODULE_DESCRIPTION("GPIO-based I2C adapter for IXP4xx systems");
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Deepak Saxena <dsaxena@plexity.net>");
-
diff --git a/include/asm-arm/arch-ixp4xx/platform.h b/include/asm-arm/arch-ixp4xx/platform.h
index 2a44d3d67980..2ce28e3fd325 100644
--- a/include/asm-arm/arch-ixp4xx/platform.h
+++ b/include/asm-arm/arch-ixp4xx/platform.h
@@ -75,17 +75,6 @@ extern unsigned long ixp4xx_exp_bus_size;
 #define IXP4XX_PERIPHERAL_BUS_CLOCK 	(66) /* 66Mhzi APB BUS   */ 
 #define IXP4XX_UART_XTAL        	14745600
 
-/*
- * The IXP4xx chips do not have an I2C unit, so GPIO lines are just
- * used to 
- * Used as platform_data to provide GPIO pin information to the ixp42x
- * I2C driver.
- */
-struct ixp4xx_i2c_pins {
-	unsigned long sda_pin;
-	unsigned long scl_pin;
-};
-
 /*
  * This structure provide a means for the board setup code
  * to give information to th pata_ixp4xx driver. It is
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 10c8edd7795c..6bab140a4505 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -191,7 +191,6 @@
 #define I2C_HW_B_OMAHA		0x010014 /* Omaha I2C interface (ARM) */
 #define I2C_HW_B_GUIDE		0x010015 /* Guide bit-basher */
 #define I2C_HW_B_IXP2000	0x010016 /* GPIO on IXP2000 systems */
-#define I2C_HW_B_IXP4XX		0x010017 /* GPIO on IXP4XX systems */
 #define I2C_HW_B_S3VIA		0x010018 /* S3Via ProSavage adapter */
 #define I2C_HW_B_ZR36067	0x010019 /* Zoran-36057/36067 based boards */
 #define I2C_HW_B_PCILYNX	0x01001a /* TI PCILynx I2C adapter */
-- 
cgit v1.2.3


From bfb6df24facfde7ec6191edbba798777efb3c375 Mon Sep 17 00:00:00 2001
From: "Mark M. Hoffman" <mhoffman@lightlink.com>
Date: Sun, 27 Jan 2008 18:14:46 +0100
Subject: i2c: Constify client address data

This patch allows much of the I2C client address data to move from initdata
into text.

Signed-off-by: Mark M. Hoffman <mhoffman@lightlink.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/i2c-core.c              |   4 +-
 include/linux/i2c.h                 | 100 ++++++++++++++++--------------------
 include/media/v4l2-i2c-drv-legacy.h |   2 +-
 3 files changed, 46 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index b5e13e405e72..7788e1730063 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -978,7 +978,7 @@ static int i2c_probe_address(struct i2c_adapter *adapter, int addr, int kind,
 }
 
 int i2c_probe(struct i2c_adapter *adapter,
-	      struct i2c_client_address_data *address_data,
+	      const struct i2c_client_address_data *address_data,
 	      int (*found_proc) (struct i2c_adapter *, int, int))
 {
 	int i, err;
@@ -987,7 +987,7 @@ int i2c_probe(struct i2c_adapter *adapter,
 	/* Force entries are done first, and are not affected by ignore
 	   entries */
 	if (address_data->forces) {
-		unsigned short **forces = address_data->forces;
+		const unsigned short * const *forces = address_data->forces;
 		int kind;
 
 		for (kind = 0; forces[kind]; kind++) {
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index a100c9f8eb7c..37c14b08d61a 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -357,10 +357,10 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data)
  * command line
  */
 struct i2c_client_address_data {
-	unsigned short *normal_i2c;
-	unsigned short *probe;
-	unsigned short *ignore;
-	unsigned short **forces;
+	const unsigned short *normal_i2c;
+	const unsigned short *probe;
+	const unsigned short *ignore;
+	const unsigned short * const *forces;
 };
 
 /* Internal numbers to terminate lists */
@@ -405,7 +405,7 @@ extern void i2c_clients_command(struct i2c_adapter *adap,
  * specific address (unless a 'force' matched);
  */
 extern int i2c_probe(struct i2c_adapter *adapter,
-		struct i2c_client_address_data *address_data,
+		const struct i2c_client_address_data *address_data,
 		int (*found_proc) (struct i2c_adapter *, int, int));
 
 extern struct i2c_adapter* i2c_get_adapter(int id);
@@ -598,104 +598,93 @@ I2C_CLIENT_MODULE_PARM(probe, "List of adapter,address pairs to scan "	\
 		       "additionally");					\
 I2C_CLIENT_MODULE_PARM(ignore, "List of adapter,address pairs not to "	\
 		       "scan");						\
-static struct i2c_client_address_data addr_data = {			\
+const static struct i2c_client_address_data addr_data = {		\
 	.normal_i2c	= normal_i2c,					\
 	.probe		= probe,					\
 	.ignore		= ignore,					\
 	.forces		= forces,					\
 }
 
+#define I2C_CLIENT_FORCE_TEXT \
+	"List of adapter,address pairs to boldly assume to be present"
+
 /* These are the ones you want to use in your own drivers. Pick the one
    which matches the number of devices the driver differenciates between. */
-#define I2C_CLIENT_INSMOD \
-  I2C_CLIENT_MODULE_PARM(force, \
-                      "List of adapter,address pairs to boldly assume " \
-                      "to be present"); \
-	static unsigned short *forces[] = {				\
-			force,						\
-			NULL						\
-		};							\
+#define I2C_CLIENT_INSMOD						\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
+static const unsigned short * const forces[] = { force, NULL };		\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_1(chip1)					\
 enum chips { any_chip, chip1 };						\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
-static unsigned short *forces[] = { force, force_##chip1, NULL };	\
+static const unsigned short * const forces[] =	{ force,		\
+	force_##chip1, NULL };						\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_2(chip1, chip2)				\
 enum chips { any_chip, chip1, chip2 };					\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, NULL };		\
+static const unsigned short * const forces[] =	{ force,		\
+	force_##chip1, force_##chip2, NULL };				\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_3(chip1, chip2, chip3)			\
 enum chips { any_chip, chip1, chip2, chip3 };				\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    NULL };				\
+static const unsigned short * const forces[] =	{ force,		\
+	force_##chip1, force_##chip2, force_##chip3, NULL };		\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_4(chip1, chip2, chip3, chip4)			\
 enum chips { any_chip, chip1, chip2, chip3, chip4 };			\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip4);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    force_##chip4, NULL};		\
+static const unsigned short * const forces[] =	{ force,		\
+	force_##chip1, force_##chip2, force_##chip3,			\
+	force_##chip4, NULL};						\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_5(chip1, chip2, chip3, chip4, chip5)		\
 enum chips { any_chip, chip1, chip2, chip3, chip4, chip5 };		\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip4);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip5);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    force_##chip4, force_##chip5,	\
-				    NULL };				\
+static const unsigned short * const forces[] = { force,			\
+	force_##chip1, force_##chip2, force_##chip3,			\
+	force_##chip4, force_##chip5, NULL };				\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_6(chip1, chip2, chip3, chip4, chip5, chip6)	\
 enum chips { any_chip, chip1, chip2, chip3, chip4, chip5, chip6 };	\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip4);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip5);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip6);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    force_##chip4, force_##chip5,	\
-				    force_##chip6, NULL };		\
+static const unsigned short * const forces[] = { force,			\
+	force_##chip1, force_##chip2, force_##chip3,			\
+	force_##chip4, force_##chip5, force_##chip6, NULL };		\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_7(chip1, chip2, chip3, chip4, chip5, chip6, chip7) \
 enum chips { any_chip, chip1, chip2, chip3, chip4, chip5, chip6,	\
 	     chip7 };							\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
@@ -703,18 +692,16 @@ I2C_CLIENT_MODULE_PARM_FORCE(chip4);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip5);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip6);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip7);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    force_##chip4, force_##chip5,	\
-				    force_##chip6, force_##chip7,	\
-				    NULL };				\
+static const unsigned short * const forces[] = { force,			\
+	force_##chip1, force_##chip2, force_##chip3,			\
+	force_##chip4, force_##chip5, force_##chip6,			\
+	force_##chip7, NULL };						\
 I2C_CLIENT_INSMOD_COMMON
 
 #define I2C_CLIENT_INSMOD_8(chip1, chip2, chip3, chip4, chip5, chip6, chip7, chip8) \
 enum chips { any_chip, chip1, chip2, chip3, chip4, chip5, chip6,	\
 	     chip7, chip8 };						\
-I2C_CLIENT_MODULE_PARM(force, "List of adapter,address pairs to "	\
-		       "boldly assume to be present");			\
+I2C_CLIENT_MODULE_PARM(force, I2C_CLIENT_FORCE_TEXT);			\
 I2C_CLIENT_MODULE_PARM_FORCE(chip1);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip2);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip3);					\
@@ -723,11 +710,10 @@ I2C_CLIENT_MODULE_PARM_FORCE(chip5);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip6);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip7);					\
 I2C_CLIENT_MODULE_PARM_FORCE(chip8);					\
-static unsigned short *forces[] = { force, force_##chip1,		\
-				    force_##chip2, force_##chip3,	\
-				    force_##chip4, force_##chip5,	\
-				    force_##chip6, force_##chip7,	\
-				    force_##chip8, NULL };		\
+static const unsigned short * const forces[] = { force,			\
+	force_##chip1, force_##chip2, force_##chip3,			\
+	force_##chip4, force_##chip5, force_##chip6,			\
+	force_##chip7, force_##chip8, NULL };				\
 I2C_CLIENT_INSMOD_COMMON
 #endif /* __KERNEL__ */
 #endif /* _LINUX_I2C_H */
diff --git a/include/media/v4l2-i2c-drv-legacy.h b/include/media/v4l2-i2c-drv-legacy.h
index 241854229d6f..e7645578fc22 100644
--- a/include/media/v4l2-i2c-drv-legacy.h
+++ b/include/media/v4l2-i2c-drv-legacy.h
@@ -34,7 +34,7 @@ struct v4l2_i2c_driver_data {
 };
 
 static struct v4l2_i2c_driver_data v4l2_i2c_data;
-static struct i2c_client_address_data addr_data;
+static const struct i2c_client_address_data addr_data;
 static struct i2c_driver v4l2_i2c_driver_legacy;
 static char v4l2_i2c_drv_name_legacy[32];
 
-- 
cgit v1.2.3


From bdc511f438f6ca40307e06edda00331e6ac0f813 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Jan 2008 18:14:48 +0100
Subject: i2c: Use the driver model reference counting

Don't implement our own reference counting mechanism for i2c clients
when the driver model already has one.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: David Brownell <david-b@pacbell.net>
---
 drivers/i2c/i2c-core.c       | 46 ++------------------------------------------
 drivers/media/video/ks0127.c |  1 -
 include/linux/i2c.h          |  3 ---
 3 files changed, 2 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 7788e1730063..9b9d808578b9 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -696,8 +696,6 @@ int i2c_attach_client(struct i2c_client *client)
 	}
 	list_add_tail(&client->list,&adapter->clients);
 
-	client->usage_count = 0;
-
 	client->dev.parent = &client->adapter->dev;
 	client->dev.bus = &i2c_bus_type;
 
@@ -744,12 +742,6 @@ int i2c_detach_client(struct i2c_client *client)
 	struct i2c_adapter *adapter = client->adapter;
 	int res = 0;
 
-	if (client->usage_count > 0) {
-		dev_warn(&client->dev, "Client [%s] still busy, "
-			 "can't detach\n", client->name);
-		return -EBUSY;
-	}
-
 	if (adapter->client_unregister)  {
 		res = adapter->client_unregister(client);
 		if (res) {
@@ -772,50 +764,16 @@ int i2c_detach_client(struct i2c_client *client)
 }
 EXPORT_SYMBOL(i2c_detach_client);
 
-static int i2c_inc_use_client(struct i2c_client *client)
-{
-
-	if (!try_module_get(client->driver->driver.owner))
-		return -ENODEV;
-	if (!try_module_get(client->adapter->owner)) {
-		module_put(client->driver->driver.owner);
-		return -ENODEV;
-	}
-
-	return 0;
-}
-
-static void i2c_dec_use_client(struct i2c_client *client)
-{
-	module_put(client->driver->driver.owner);
-	module_put(client->adapter->owner);
-}
-
 int i2c_use_client(struct i2c_client *client)
 {
-	int ret;
-
-	ret = i2c_inc_use_client(client);
-	if (ret)
-		return ret;
-
-	client->usage_count++;
-
+	get_device(&client->dev);
 	return 0;
 }
 EXPORT_SYMBOL(i2c_use_client);
 
 int i2c_release_client(struct i2c_client *client)
 {
-	if (!client->usage_count) {
-		pr_debug("i2c-core: %s used one too many times\n",
-			 __FUNCTION__);
-		return -EPERM;
-	}
-
-	client->usage_count--;
-	i2c_dec_use_client(client);
-
+	put_device(&client->dev);
 	return 0;
 }
 EXPORT_SYMBOL(i2c_release_client);
diff --git a/drivers/media/video/ks0127.c b/drivers/media/video/ks0127.c
index b6cd21e6dab9..4895540be195 100644
--- a/drivers/media/video/ks0127.c
+++ b/drivers/media/video/ks0127.c
@@ -764,7 +764,6 @@ static struct i2c_client ks0127_client_tmpl =
 	.addr = 0,
 	.adapter = NULL,
 	.driver = &i2c_driver_ks0127,
-	.usage_count = 0
 };
 
 static int ks0127_found_proc(struct i2c_adapter *adapter, int addr, int kind)
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 37c14b08d61a..f7cd2f370c39 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -155,7 +155,6 @@ struct i2c_driver {
  *	generic enough to hide second-sourcing and compatible revisions.
  * @adapter: manages the bus segment hosting this I2C device
  * @driver: device's driver, hence pointer to access routines
- * @usage_count: counts current number of users of this client
  * @dev: Driver model device node for the slave.
  * @irq: indicates the IRQ generated by this device (if any)
  * @driver_name: Identifies new-style driver used with this device; also
@@ -175,8 +174,6 @@ struct i2c_client {
 	char name[I2C_NAME_SIZE];
 	struct i2c_adapter *adapter;	/* the adapter we sit on	*/
 	struct i2c_driver *driver;	/* and our access routines	*/
-	int usage_count;		/* How many accesses currently  */
-					/* to the client		*/
 	struct device dev;		/* the device structure		*/
 	int irq;			/* irq issued by device (or -1) */
 	char driver_name[KOBJ_NAME_LEN];
-- 
cgit v1.2.3


From e48d33193d94175f012c3ed606a1d1e574ed726a Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Jan 2008 18:14:48 +0100
Subject: i2c: Change prototypes of refcounting functions

Use more standard prototypes for i2c_use_client() and
i2c_release_client(). The former now returns a pointer to the client,
and the latter no longer returns anything. This matches what all other
subsystems do.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: David Brownell <david-b@pacbell.net>
---
 drivers/i2c/i2c-core.c     | 24 ++++++++++++++++++++----
 drivers/media/video/vino.c | 22 ++++------------------
 include/linux/i2c.h        |  7 ++-----
 3 files changed, 26 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 9b9d808578b9..f75344ac4663 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -764,17 +764,33 @@ int i2c_detach_client(struct i2c_client *client)
 }
 EXPORT_SYMBOL(i2c_detach_client);
 
-int i2c_use_client(struct i2c_client *client)
+/**
+ * i2c_use_client - increments the reference count of the i2c client structure
+ * @client: the client being referenced
+ *
+ * Each live reference to a client should be refcounted. The driver model does
+ * that automatically as part of driver binding, so that most drivers don't
+ * need to do this explicitly: they hold a reference until they're unbound
+ * from the device.
+ *
+ * A pointer to the client with the incremented reference counter is returned.
+ */
+struct i2c_client *i2c_use_client(struct i2c_client *client)
 {
 	get_device(&client->dev);
-	return 0;
+	return client;
 }
 EXPORT_SYMBOL(i2c_use_client);
 
-int i2c_release_client(struct i2c_client *client)
+/**
+ * i2c_release_client - release a use of the i2c client structure
+ * @client: the client being no longer referenced
+ *
+ * Must be called when a user of a client is finished with it.
+ */
+void i2c_release_client(struct i2c_client *client)
 {
 	put_device(&client->dev);
-	return 0;
 }
 EXPORT_SYMBOL(i2c_release_client);
 
diff --git a/drivers/media/video/vino.c b/drivers/media/video/vino.c
index 9a03dc82c6ca..5bb75294b5aa 100644
--- a/drivers/media/video/vino.c
+++ b/drivers/media/video/vino.c
@@ -2589,11 +2589,7 @@ static int vino_acquire_input(struct vino_channel_settings *vcs)
 	/* First try D1 and then SAA7191 */
 	if (vino_drvdata->camera.driver
 	    && (vino_drvdata->camera.owner == VINO_NO_CHANNEL)) {
-		if (i2c_use_client(vino_drvdata->camera.driver)) {
-			ret = -ENODEV;
-			goto out;
-		}
-
+		i2c_use_client(vino_drvdata->camera.driver);
 		vino_drvdata->camera.owner = vcs->channel;
 		vcs->input = VINO_INPUT_D1;
 		vcs->data_norm = VINO_DATA_NORM_D1;
@@ -2602,11 +2598,7 @@ static int vino_acquire_input(struct vino_channel_settings *vcs)
 		int input, data_norm;
 		int saa7191_input;
 
-		if (i2c_use_client(vino_drvdata->decoder.driver)) {
-			ret = -ENODEV;
-			goto out;
-		}
-
+		i2c_use_client(vino_drvdata->decoder.driver);
 		input = VINO_INPUT_COMPOSITE;
 
 		saa7191_input = vino_get_saa7191_input(input);
@@ -2688,10 +2680,7 @@ static int vino_set_input(struct vino_channel_settings *vcs, int input)
 		}
 
 		if (vino_drvdata->decoder.owner == VINO_NO_CHANNEL) {
-			if (i2c_use_client(vino_drvdata->decoder.driver)) {
-				ret = -ENODEV;
-				goto out;
-			}
+			i2c_use_client(vino_drvdata->decoder.driver);
 			vino_drvdata->decoder.owner = vcs->channel;
 		}
 
@@ -2759,10 +2748,7 @@ static int vino_set_input(struct vino_channel_settings *vcs, int input)
 		}
 
 		if (vino_drvdata->camera.owner == VINO_NO_CHANNEL) {
-			if (i2c_use_client(vino_drvdata->camera.driver)) {
-				ret = -ENODEV;
-				goto out;
-			}
+			i2c_use_client(vino_drvdata->camera.driver);
 			vino_drvdata->camera.owner = vcs->channel;
 		}
 
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index f7cd2f370c39..78a305bed8eb 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -386,11 +386,8 @@ static inline int i2c_add_driver(struct i2c_driver *driver)
 extern int i2c_attach_client(struct i2c_client *);
 extern int i2c_detach_client(struct i2c_client *);
 
-/* Should be used to make sure that client-struct is valid and that it
-   is okay to access the i2c-client.
-   returns -ENODEV if client has gone in the meantime */
-extern int i2c_use_client(struct i2c_client *);
-extern int i2c_release_client(struct i2c_client *);
+extern struct i2c_client *i2c_use_client(struct i2c_client *client);
+extern void i2c_release_client(struct i2c_client *client);
 
 /* call the i2c_client->command() of all attached clients with
  * the given arguments */
-- 
cgit v1.2.3


From 87c6c22945e5d68eb96dd1e5cb26185253cd5b9d Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Jan 2008 18:14:48 +0100
Subject: i2c: Drop redundant i2c_adapter.list

i2c_adapter.list is superfluous, this list duplicates the one
maintained by the driver core. Drop it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
---
 Documentation/feature-removal-schedule.txt |  9 ---------
 drivers/i2c/i2c-core.c                     | 24 ++++++++++--------------
 include/linux/i2c.h                        |  1 -
 3 files changed, 10 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 878ca34844e4..25370662cc5d 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -191,15 +191,6 @@ Who:	Kay Sievers <kay.sievers@suse.de>
 
 ---------------------------
 
-What:	i2c_adapter.list
-When:	July 2007
-Why:	Superfluous, this list duplicates the one maintained by the driver
-	core.
-Who:	Jean Delvare <khali@linux-fr.org>,
-	David Brownell <dbrownell@users.sourceforge.net>
-
----------------------------
-
 What:	ACPI procfs interface
 When:	July 2008
 Why:	ACPI sysfs conversion should be finished by January 2008.
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index f75344ac4663..3495ab4035ec 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -34,11 +34,11 @@
 #include <linux/mutex.h>
 #include <linux/completion.h>
 #include <asm/uaccess.h>
+#include <asm/semaphore.h>
 
 #include "i2c-core.h"
 
 
-static LIST_HEAD(adapters);
 static LIST_HEAD(drivers);
 static DEFINE_MUTEX(core_lists);
 static DEFINE_IDR(i2c_adapter_idr);
@@ -331,7 +331,6 @@ static int i2c_register_adapter(struct i2c_adapter *adap)
 	INIT_LIST_HEAD(&adap->clients);
 
 	mutex_lock(&core_lists);
-	list_add_tail(&adap->list, &adapters);
 
 	/* Add the adapter to the driver core.
 	 * If the parent pointer is not set up,
@@ -368,7 +367,6 @@ out_unlock:
 	return res;
 
 out_list:
-	list_del(&adap->list);
 	idr_remove(&i2c_adapter_idr, adap->nr);
 	goto out_unlock;
 }
@@ -473,7 +471,6 @@ EXPORT_SYMBOL_GPL(i2c_add_numbered_adapter);
 int i2c_del_adapter(struct i2c_adapter *adap)
 {
 	struct list_head  *item, *_n;
-	struct i2c_adapter *adap_from_list;
 	struct i2c_driver *driver;
 	struct i2c_client *client;
 	int res = 0;
@@ -481,11 +478,7 @@ int i2c_del_adapter(struct i2c_adapter *adap)
 	mutex_lock(&core_lists);
 
 	/* First make sure that this adapter was ever added */
-	list_for_each_entry(adap_from_list, &adapters, list) {
-		if (adap_from_list == adap)
-			break;
-	}
-	if (adap_from_list != adap) {
+	if (idr_find(&i2c_adapter_idr, adap->nr) != adap) {
 		pr_debug("i2c-core: attempting to delete unregistered "
 			 "adapter [%s]\n", adap->name);
 		res = -EINVAL;
@@ -529,7 +522,6 @@ int i2c_del_adapter(struct i2c_adapter *adap)
 	/* clean up the sysfs representation */
 	init_completion(&adap->dev_released);
 	device_unregister(&adap->dev);
-	list_del(&adap->list);
 
 	/* wait for sysfs to drop all references */
 	wait_for_completion(&adap->dev_released);
@@ -592,9 +584,12 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver)
 	if (driver->attach_adapter) {
 		struct i2c_adapter *adapter;
 
-		list_for_each_entry(adapter, &adapters, list) {
+		down(&i2c_adapter_class.sem);
+		list_for_each_entry(adapter, &i2c_adapter_class.devices,
+				    dev.node) {
 			driver->attach_adapter(adapter);
 		}
+		up(&i2c_adapter_class.sem);
 	}
 
 	mutex_unlock(&core_lists);
@@ -609,7 +604,7 @@ EXPORT_SYMBOL(i2c_register_driver);
  */
 void i2c_del_driver(struct i2c_driver *driver)
 {
-	struct list_head   *item1, *item2, *_n;
+	struct list_head   *item2, *_n;
 	struct i2c_client  *client;
 	struct i2c_adapter *adap;
 
@@ -623,8 +618,8 @@ void i2c_del_driver(struct i2c_driver *driver)
 	 * attached. If so, detach them to be able to kill the driver
 	 * afterwards.
 	 */
-	list_for_each(item1,&adapters) {
-		adap = list_entry(item1, struct i2c_adapter, list);
+	down(&i2c_adapter_class.sem);
+	list_for_each_entry(adap, &i2c_adapter_class.devices, dev.node) {
 		if (driver->detach_adapter) {
 			if (driver->detach_adapter(adap)) {
 				dev_err(&adap->dev, "detach_adapter failed "
@@ -648,6 +643,7 @@ void i2c_del_driver(struct i2c_driver *driver)
 			}
 		}
 	}
+	up(&i2c_adapter_class.sem);
 
  unregister:
 	driver_unregister(&driver->driver);
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 78a305bed8eb..5fa7e180501d 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -317,7 +317,6 @@ struct i2c_adapter {
 
 	int nr;
 	struct list_head clients;
-	struct list_head list;
 	char name[48];
 	struct completion dev_released;
 };
-- 
cgit v1.2.3


From 026526f5afcd421dce110f53e4c4e2b9e78753c2 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Jan 2008 18:14:49 +0100
Subject: i2c: Drop redundant i2c_driver.list

i2c_driver.list is superfluous, this list duplicates the one
maintained by the driver core. Drop it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
---
 drivers/i2c/i2c-core.c | 58 ++++++++++++++++++++++++++++++--------------------
 include/linux/i2c.h    |  1 -
 2 files changed, 35 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 3495ab4035ec..29d2a1b4b464 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -39,7 +39,6 @@
 #include "i2c-core.h"
 
 
-static LIST_HEAD(drivers);
 static DEFINE_MUTEX(core_lists);
 static DEFINE_IDR(i2c_adapter_idr);
 
@@ -320,11 +319,21 @@ static void i2c_scan_static_board_info(struct i2c_adapter *adapter)
 	mutex_unlock(&__i2c_board_lock);
 }
 
+static int i2c_do_add_adapter(struct device_driver *d, void *data)
+{
+	struct i2c_driver *driver = to_i2c_driver(d);
+	struct i2c_adapter *adap = data;
+
+	if (driver->attach_adapter) {
+		/* We ignore the return code; if it fails, too bad */
+		driver->attach_adapter(adap);
+	}
+	return 0;
+}
+
 static int i2c_register_adapter(struct i2c_adapter *adap)
 {
-	int res = 0;
-	struct list_head   *item;
-	struct i2c_driver  *driver;
+	int res = 0, dummy;
 
 	mutex_init(&adap->bus_lock);
 	mutex_init(&adap->clist_lock);
@@ -355,12 +364,8 @@ static int i2c_register_adapter(struct i2c_adapter *adap)
 		i2c_scan_static_board_info(adap);
 
 	/* let legacy drivers scan this bus for matching devices */
-	list_for_each(item,&drivers) {
-		driver = list_entry(item, struct i2c_driver, list);
-		if (driver->attach_adapter)
-			/* We ignore the return code; if it fails, too bad */
-			driver->attach_adapter(adap);
-	}
+	dummy = bus_for_each_drv(&i2c_bus_type, NULL, adap,
+				 i2c_do_add_adapter);
 
 out_unlock:
 	mutex_unlock(&core_lists);
@@ -460,6 +465,21 @@ retry:
 }
 EXPORT_SYMBOL_GPL(i2c_add_numbered_adapter);
 
+static int i2c_do_del_adapter(struct device_driver *d, void *data)
+{
+	struct i2c_driver *driver = to_i2c_driver(d);
+	struct i2c_adapter *adapter = data;
+	int res;
+
+	if (!driver->detach_adapter)
+		return 0;
+	res = driver->detach_adapter(adapter);
+	if (res)
+		dev_err(&adapter->dev, "detach_adapter failed (%d) "
+			"for driver [%s]\n", res, driver->driver.name);
+	return res;
+}
+
 /**
  * i2c_del_adapter - unregister I2C adapter
  * @adap: the adapter being unregistered
@@ -471,7 +491,6 @@ EXPORT_SYMBOL_GPL(i2c_add_numbered_adapter);
 int i2c_del_adapter(struct i2c_adapter *adap)
 {
 	struct list_head  *item, *_n;
-	struct i2c_driver *driver;
 	struct i2c_client *client;
 	int res = 0;
 
@@ -485,16 +504,11 @@ int i2c_del_adapter(struct i2c_adapter *adap)
 		goto out_unlock;
 	}
 
-	list_for_each(item,&drivers) {
-		driver = list_entry(item, struct i2c_driver, list);
-		if (driver->detach_adapter)
-			if ((res = driver->detach_adapter(adap))) {
-				dev_err(&adap->dev, "detach_adapter failed "
-					"for driver [%s]\n",
-					driver->driver.name);
-				goto out_unlock;
-			}
-	}
+	/* Tell drivers about this removal */
+	res = bus_for_each_drv(&i2c_bus_type, NULL, adap,
+			       i2c_do_del_adapter);
+	if (res)
+		goto out_unlock;
 
 	/* detach any active clients. This must be done first, because
 	 * it can fail; in which case we give up. */
@@ -577,7 +591,6 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver)
 
 	mutex_lock(&core_lists);
 
-	list_add_tail(&driver->list,&drivers);
 	pr_debug("i2c-core: driver [%s] registered\n", driver->driver.name);
 
 	/* legacy drivers scan i2c busses directly */
@@ -647,7 +660,6 @@ void i2c_del_driver(struct i2c_driver *driver)
 
  unregister:
 	driver_unregister(&driver->driver);
-	list_del(&driver->list);
 	pr_debug("i2c-core: driver [%s] unregistered\n", driver->driver.name);
 
 	mutex_unlock(&core_lists);
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 5fa7e180501d..0fc59efd80e4 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -140,7 +140,6 @@ struct i2c_driver {
 	int (*command)(struct i2c_client *client,unsigned int cmd, void *arg);
 
 	struct device_driver driver;
-	struct list_head list;
 };
 #define to_i2c_driver(d) container_of(d, struct i2c_driver, driver)
 
-- 
cgit v1.2.3


From 6d16bfb5e81d3925a7efb38b5cc3e0021b57d03a Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 27 Jan 2008 18:14:49 +0100
Subject: i2c/tps65010: move header to <linux/i2c/...>

Move the tps65010 header file from the OMAP arch directory to the
more generic <linux/i2c/...> directory, and remove the spurious
dependency of this driver on OMAP.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 arch/arm/mach-omap1/board-h2.c       |   2 +-
 arch/arm/mach-omap1/board-h3.c       |   2 +-
 arch/arm/mach-omap1/board-osk.c      |   3 +-
 arch/arm/mach-omap1/leds-osk.c       |   2 +-
 drivers/i2c/chips/Kconfig            |   4 -
 drivers/i2c/chips/isp1301_omap.c     |   2 +-
 drivers/i2c/chips/tps65010.c         |   2 +-
 drivers/mmc/host/omap.c              |   2 +-
 drivers/usb/host/ohci-omap.c         |   2 +-
 drivers/video/omap/lcd_h3.c          |   2 +-
 include/asm-arm/arch-omap/tps65010.h | 156 -----------------------------------
 include/linux/i2c/tps65010.h         | 156 +++++++++++++++++++++++++++++++++++
 12 files changed, 166 insertions(+), 169 deletions(-)
 delete mode 100644 include/asm-arm/arch-omap/tps65010.h
 create mode 100644 include/linux/i2c/tps65010.h

(limited to 'include')

diff --git a/arch/arm/mach-omap1/board-h2.c b/arch/arm/mach-omap1/board-h2.c
index 130681201c19..9393824cc150 100644
--- a/arch/arm/mach-omap1/board-h2.c
+++ b/arch/arm/mach-omap1/board-h2.c
@@ -27,6 +27,7 @@
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
 #include <linux/input.h>
+#include <linux/i2c/tps65010.h>
 
 #include <asm/hardware.h>
 #include <asm/gpio.h>
@@ -36,7 +37,6 @@
 #include <asm/mach/flash.h>
 #include <asm/mach/map.h>
 
-#include <asm/arch/tps65010.h>
 #include <asm/arch/mux.h>
 #include <asm/arch/tc.h>
 #include <asm/arch/irda.h>
diff --git a/arch/arm/mach-omap1/board-h3.c b/arch/arm/mach-omap1/board-h3.c
index 4f84ae273a1f..978cdab16535 100644
--- a/arch/arm/mach-omap1/board-h3.c
+++ b/arch/arm/mach-omap1/board-h3.c
@@ -26,6 +26,7 @@
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
 #include <linux/input.h>
+#include <linux/i2c/tps65010.h>
 
 #include <asm/setup.h>
 #include <asm/page.h>
@@ -37,7 +38,6 @@
 #include <asm/mach/flash.h>
 #include <asm/mach/map.h>
 
-#include <asm/arch/tps65010.h>
 #include <asm/arch/gpioexpander.h>
 #include <asm/arch/irqs.h>
 #include <asm/arch/mux.h>
diff --git a/arch/arm/mach-omap1/board-osk.c b/arch/arm/mach-omap1/board-osk.c
index 5db182da322b..5b575e687a2a 100644
--- a/arch/arm/mach-omap1/board-osk.c
+++ b/arch/arm/mach-omap1/board-osk.c
@@ -37,6 +37,8 @@
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
 
+#include <linux/i2c/tps65010.h>
+
 #include <asm/hardware.h>
 #include <asm/gpio.h>
 
@@ -46,7 +48,6 @@
 #include <asm/mach/flash.h>
 
 #include <asm/arch/usb.h>
-#include <asm/arch/tps65010.h>
 #include <asm/arch/mux.h>
 #include <asm/arch/tc.h>
 #include <asm/arch/common.h>
diff --git a/arch/arm/mach-omap1/leds-osk.c b/arch/arm/mach-omap1/leds-osk.c
index 86de303ecab2..6939d5e7569a 100644
--- a/arch/arm/mach-omap1/leds-osk.c
+++ b/arch/arm/mach-omap1/leds-osk.c
@@ -5,13 +5,13 @@
  */
 #include <linux/init.h>
 #include <linux/workqueue.h>
+#include <linux/i2c/tps65010.h>
 
 #include <asm/hardware.h>
 #include <asm/leds.h>
 #include <asm/system.h>
 
 #include <asm/arch/gpio.h>
-#include <asm/arch/tps65010.h>
 
 #include "leds.h"
 
diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index fe3008e59467..bd7082c2443d 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -88,12 +88,8 @@ config ISP1301_OMAP
 	  This driver can also be built as a module.  If so, the module
 	  will be called isp1301_omap.
 
-# NOTE:  This isn't really OMAP-specific, except for the current
-# interface location in  <include/asm-arm/arch-omap/tps65010.h>
-# and having mostly OMAP-specific board support
 config TPS65010
 	tristate "TPS6501x Power Management chips"
-	depends on ARCH_OMAP
 	default y if MACH_OMAP_H2 || MACH_OMAP_H3 || MACH_OMAP_OSK
 	help
 	  If you say yes here you get support for the TPS6501x series of
diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c
index ebfbb2947ae6..2a3160153f54 100644
--- a/drivers/i2c/chips/isp1301_omap.c
+++ b/drivers/i2c/chips/isp1301_omap.c
@@ -100,7 +100,7 @@ struct isp1301 {
 
 #if	defined(CONFIG_TPS65010) || defined(CONFIG_TPS65010_MODULE)
 
-#include <asm/arch/tps65010.h>
+#include <linux/i2c/tps65010.h>
 
 #else
 
diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c
index e320994b981c..4154a9108859 100644
--- a/drivers/i2c/chips/tps65010.c
+++ b/drivers/i2c/chips/tps65010.c
@@ -31,7 +31,7 @@
 #include <linux/seq_file.h>
 #include <linux/mutex.h>
 
-#include <asm/arch/tps65010.h>
+#include <linux/i2c/tps65010.h>
 
 /*-------------------------------------------------------------------------*/
 
diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c
index 971e18b91f4a..c9dfeb15b487 100644
--- a/drivers/mmc/host/omap.c
+++ b/drivers/mmc/host/omap.c
@@ -25,6 +25,7 @@
 #include <linux/mmc/card.h>
 #include <linux/clk.h>
 #include <linux/scatterlist.h>
+#include <linux/i2c/tps65010.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
@@ -35,7 +36,6 @@
 #include <asm/arch/dma.h>
 #include <asm/arch/mux.h>
 #include <asm/arch/fpga.h>
-#include <asm/arch/tps65010.h>
 
 #define	OMAP_MMC_REG_CMD	0x00
 #define	OMAP_MMC_REG_ARGL	0x04
diff --git a/drivers/usb/host/ohci-omap.c b/drivers/usb/host/ohci-omap.c
index 5cfa3d1c4413..74e1f4be10bb 100644
--- a/drivers/usb/host/ohci-omap.c
+++ b/drivers/usb/host/ohci-omap.c
@@ -47,7 +47,7 @@
 #endif
 
 #ifdef CONFIG_TPS65010
-#include <asm/arch/tps65010.h>
+#include <linux/i2c/tps65010.h>
 #else
 
 #define LOW	0
diff --git a/drivers/video/omap/lcd_h3.c b/drivers/video/omap/lcd_h3.c
index c604d935c188..31e978349a80 100644
--- a/drivers/video/omap/lcd_h3.c
+++ b/drivers/video/omap/lcd_h3.c
@@ -21,9 +21,9 @@
 
 #include <linux/module.h>
 #include <linux/platform_device.h>
+#include <linux/i2c/tps65010.h>
 
 #include <asm/arch/gpio.h>
-#include <asm/arch/tps65010.h>
 #include <asm/arch/omapfb.h>
 
 #define MODULE_NAME	"omapfb-lcd_h3"
diff --git a/include/asm-arm/arch-omap/tps65010.h b/include/asm-arm/arch-omap/tps65010.h
deleted file mode 100644
index b9aa2b3a3909..000000000000
--- a/include/asm-arm/arch-omap/tps65010.h
+++ /dev/null
@@ -1,156 +0,0 @@
-/* linux/include/asm-arm/arch-omap/tps65010.h
- *
- * Functions to access TPS65010 power management device.
- *
- * Copyright (C) 2004 Dirk Behme <dirk.behme@de.bosch.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
- *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
- * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
- * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
- * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * You should have received a copy of the  GNU General Public License along
- * with this program; if not, write  to the Free Software Foundation, Inc.,
- * 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#ifndef __ASM_ARCH_TPS65010_H
-#define __ASM_ARCH_TPS65010_H
-
-/*
- * ----------------------------------------------------------------------------
- * Registers, all 8 bits
- * ----------------------------------------------------------------------------
- */
-
-#define	TPS_CHGSTATUS		0x01
-#	define	TPS_CHG_USB		(1 << 7)
-#	define	TPS_CHG_AC		(1 << 6)
-#	define	TPS_CHG_THERM		(1 << 5)
-#	define	TPS_CHG_TERM		(1 << 4)
-#	define	TPS_CHG_TAPER_TMO	(1 << 3)
-#	define	TPS_CHG_CHG_TMO		(1 << 2)
-#	define	TPS_CHG_PRECHG_TMO	(1 << 1)
-#	define	TPS_CHG_TEMP_ERR	(1 << 0)
-#define	TPS_REGSTATUS		0x02
-#	define	TPS_REG_ONOFF		(1 << 7)
-#	define	TPS_REG_COVER		(1 << 6)
-#	define	TPS_REG_UVLO		(1 << 5)
-#	define	TPS_REG_NO_CHG		(1 << 4)	/* tps65013 */
-#	define	TPS_REG_PG_LD02		(1 << 3)
-#	define	TPS_REG_PG_LD01		(1 << 2)
-#	define	TPS_REG_PG_MAIN		(1 << 1)
-#	define	TPS_REG_PG_CORE		(1 << 0)
-#define	TPS_MASK1		0x03
-#define	TPS_MASK2		0x04
-#define	TPS_ACKINT1		0x05
-#define	TPS_ACKINT2		0x06
-#define	TPS_CHGCONFIG		0x07
-#	define	TPS_CHARGE_POR		(1 << 7)	/* 65010/65012 */
-#	define	TPS65013_AUA		(1 << 7)	/* 65011/65013 */
-#	define	TPS_CHARGE_RESET	(1 << 6)
-#	define	TPS_CHARGE_FAST		(1 << 5)
-#	define	TPS_CHARGE_CURRENT	(3 << 3)
-#	define	TPS_VBUS_500MA		(1 << 2)
-#	define	TPS_VBUS_CHARGING	(1 << 1)
-#	define	TPS_CHARGE_ENABLE	(1 << 0)
-#define	TPS_LED1_ON		0x08
-#define	TPS_LED1_PER		0x09
-#define	TPS_LED2_ON		0x0a
-#define	TPS_LED2_PER		0x0b
-#define	TPS_VDCDC1		0x0c
-#	define	TPS_ENABLE_LP		(1 << 3)
-#define	TPS_VDCDC2		0x0d
-#define	TPS_VREGS1		0x0e
-#	define	TPS_LDO2_ENABLE	(1 << 7)
-#	define	TPS_LDO2_OFF	(1 << 6)
-#	define	TPS_VLDO2_3_0V	(3 << 4)
-#	define	TPS_VLDO2_2_75V	(2 << 4)
-#	define	TPS_VLDO2_2_5V	(1 << 4)
-#	define	TPS_VLDO2_1_8V	(0 << 4)
-#	define	TPS_LDO1_ENABLE	(1 << 3)
-#	define	TPS_LDO1_OFF	(1 << 2)
-#	define	TPS_VLDO1_3_0V	(3 << 0)
-#	define	TPS_VLDO1_2_75V	(2 << 0)
-#	define	TPS_VLDO1_2_5V	(1 << 0)
-#	define	TPS_VLDO1_ADJ	(0 << 0)
-#define	TPS_MASK3		0x0f
-#define	TPS_DEFGPIO		0x10
-
-/*
- * ----------------------------------------------------------------------------
- * Macros used by exported functions
- * ----------------------------------------------------------------------------
- */
-
-#define LED1  1
-#define LED2  2
-#define OFF   0
-#define ON    1
-#define BLINK 2
-#define GPIO1 1
-#define GPIO2 2
-#define GPIO3 3
-#define GPIO4 4
-#define LOW   0
-#define HIGH  1
-
-/*
- * ----------------------------------------------------------------------------
- * Exported functions
- * ----------------------------------------------------------------------------
- */
-
-/* Draw from VBUS:
- *   0 mA -- DON'T DRAW (might supply power instead)
- * 100 mA -- usb unit load (slowest charge rate)
- * 500 mA -- usb high power (fast battery charge)
- */
-extern int tps65010_set_vbus_draw(unsigned mA);
-
-/* tps65010_set_gpio_out_value parameter:
- * gpio:  GPIO1, GPIO2, GPIO3 or GPIO4
- * value: LOW or HIGH
- */
-extern int tps65010_set_gpio_out_value(unsigned gpio, unsigned value);
-
-/* tps65010_set_led parameter:
- * led:  LED1 or LED2
- * mode: ON, OFF or BLINK
- */
-extern int tps65010_set_led(unsigned led, unsigned mode);
-
-/* tps65010_set_vib parameter:
- * value: ON or OFF
- */
-extern int tps65010_set_vib(unsigned value);
-
-/* tps65010_set_low_pwr parameter:
- * mode: ON or OFF
- */
-extern int tps65010_set_low_pwr(unsigned mode);
-
-/* tps65010_config_vregs1 parameter:
- * value to be written to VREGS1 register
- * Note: The complete register is written, set all bits you need
- */
-extern int tps65010_config_vregs1(unsigned value);
-
-/* tps65013_set_low_pwr parameter:
- * mode: ON or OFF
- */
-extern int tps65013_set_low_pwr(unsigned mode);
-
-#endif /*  __ASM_ARCH_TPS65010_H */
-
diff --git a/include/linux/i2c/tps65010.h b/include/linux/i2c/tps65010.h
new file mode 100644
index 000000000000..7021635ed6a0
--- /dev/null
+++ b/include/linux/i2c/tps65010.h
@@ -0,0 +1,156 @@
+/* linux/i2c/tps65010.h
+ *
+ * Functions to access TPS65010 power management device.
+ *
+ * Copyright (C) 2004 Dirk Behme <dirk.behme@de.bosch.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
+ * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
+ * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the  GNU General Public License along
+ * with this program; if not, write  to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __LINUX_I2C_TPS65010_H
+#define __LINUX_I2C_TPS65010_H
+
+/*
+ * ----------------------------------------------------------------------------
+ * Registers, all 8 bits
+ * ----------------------------------------------------------------------------
+ */
+
+#define	TPS_CHGSTATUS		0x01
+#	define	TPS_CHG_USB		(1 << 7)
+#	define	TPS_CHG_AC		(1 << 6)
+#	define	TPS_CHG_THERM		(1 << 5)
+#	define	TPS_CHG_TERM		(1 << 4)
+#	define	TPS_CHG_TAPER_TMO	(1 << 3)
+#	define	TPS_CHG_CHG_TMO		(1 << 2)
+#	define	TPS_CHG_PRECHG_TMO	(1 << 1)
+#	define	TPS_CHG_TEMP_ERR	(1 << 0)
+#define	TPS_REGSTATUS		0x02
+#	define	TPS_REG_ONOFF		(1 << 7)
+#	define	TPS_REG_COVER		(1 << 6)
+#	define	TPS_REG_UVLO		(1 << 5)
+#	define	TPS_REG_NO_CHG		(1 << 4)	/* tps65013 */
+#	define	TPS_REG_PG_LD02		(1 << 3)
+#	define	TPS_REG_PG_LD01		(1 << 2)
+#	define	TPS_REG_PG_MAIN		(1 << 1)
+#	define	TPS_REG_PG_CORE		(1 << 0)
+#define	TPS_MASK1		0x03
+#define	TPS_MASK2		0x04
+#define	TPS_ACKINT1		0x05
+#define	TPS_ACKINT2		0x06
+#define	TPS_CHGCONFIG		0x07
+#	define	TPS_CHARGE_POR		(1 << 7)	/* 65010/65012 */
+#	define	TPS65013_AUA		(1 << 7)	/* 65011/65013 */
+#	define	TPS_CHARGE_RESET	(1 << 6)
+#	define	TPS_CHARGE_FAST		(1 << 5)
+#	define	TPS_CHARGE_CURRENT	(3 << 3)
+#	define	TPS_VBUS_500MA		(1 << 2)
+#	define	TPS_VBUS_CHARGING	(1 << 1)
+#	define	TPS_CHARGE_ENABLE	(1 << 0)
+#define	TPS_LED1_ON		0x08
+#define	TPS_LED1_PER		0x09
+#define	TPS_LED2_ON		0x0a
+#define	TPS_LED2_PER		0x0b
+#define	TPS_VDCDC1		0x0c
+#	define	TPS_ENABLE_LP		(1 << 3)
+#define	TPS_VDCDC2		0x0d
+#define	TPS_VREGS1		0x0e
+#	define	TPS_LDO2_ENABLE	(1 << 7)
+#	define	TPS_LDO2_OFF	(1 << 6)
+#	define	TPS_VLDO2_3_0V	(3 << 4)
+#	define	TPS_VLDO2_2_75V	(2 << 4)
+#	define	TPS_VLDO2_2_5V	(1 << 4)
+#	define	TPS_VLDO2_1_8V	(0 << 4)
+#	define	TPS_LDO1_ENABLE	(1 << 3)
+#	define	TPS_LDO1_OFF	(1 << 2)
+#	define	TPS_VLDO1_3_0V	(3 << 0)
+#	define	TPS_VLDO1_2_75V	(2 << 0)
+#	define	TPS_VLDO1_2_5V	(1 << 0)
+#	define	TPS_VLDO1_ADJ	(0 << 0)
+#define	TPS_MASK3		0x0f
+#define	TPS_DEFGPIO		0x10
+
+/*
+ * ----------------------------------------------------------------------------
+ * Macros used by exported functions
+ * ----------------------------------------------------------------------------
+ */
+
+#define LED1  1
+#define LED2  2
+#define OFF   0
+#define ON    1
+#define BLINK 2
+#define GPIO1 1
+#define GPIO2 2
+#define GPIO3 3
+#define GPIO4 4
+#define LOW   0
+#define HIGH  1
+
+/*
+ * ----------------------------------------------------------------------------
+ * Exported functions
+ * ----------------------------------------------------------------------------
+ */
+
+/* Draw from VBUS:
+ *   0 mA -- DON'T DRAW (might supply power instead)
+ * 100 mA -- usb unit load (slowest charge rate)
+ * 500 mA -- usb high power (fast battery charge)
+ */
+extern int tps65010_set_vbus_draw(unsigned mA);
+
+/* tps65010_set_gpio_out_value parameter:
+ * gpio:  GPIO1, GPIO2, GPIO3 or GPIO4
+ * value: LOW or HIGH
+ */
+extern int tps65010_set_gpio_out_value(unsigned gpio, unsigned value);
+
+/* tps65010_set_led parameter:
+ * led:  LED1 or LED2
+ * mode: ON, OFF or BLINK
+ */
+extern int tps65010_set_led(unsigned led, unsigned mode);
+
+/* tps65010_set_vib parameter:
+ * value: ON or OFF
+ */
+extern int tps65010_set_vib(unsigned value);
+
+/* tps65010_set_low_pwr parameter:
+ * mode: ON or OFF
+ */
+extern int tps65010_set_low_pwr(unsigned mode);
+
+/* tps65010_config_vregs1 parameter:
+ * value to be written to VREGS1 register
+ * Note: The complete register is written, set all bits you need
+ */
+extern int tps65010_config_vregs1(unsigned value);
+
+/* tps65013_set_low_pwr parameter:
+ * mode: ON or OFF
+ */
+extern int tps65013_set_low_pwr(unsigned mode);
+
+#endif /*  __LINUX_I2C_TPS65010_H */
+
-- 
cgit v1.2.3


From 7bca0871ca332ad5373a0fd26886e3cfbafa822c Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Jan 2008 18:14:50 +0100
Subject: i2c: Discard unused driver IDs

Discard all I2C driver IDs that aren't used anywhere. That's not just a
couple of them, but more like 49 or one quarter of all defined IDs! And
this is just a first pass, next will come all IDs that are set but
never used, or used but never set.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 include/linux/i2c-id.h | 60 --------------------------------------------------
 1 file changed, 60 deletions(-)

(limited to 'include')

diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 6bab140a4505..f922b060158b 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -33,23 +33,13 @@
 
 #define I2C_DRIVERID_MSP3400	 1
 #define I2C_DRIVERID_TUNER	 2
-#define I2C_DRIVERID_VIDEOTEX	 3	/* please rename		*/
 #define I2C_DRIVERID_TDA8425	 4	/* stereo sound processor	*/
 #define I2C_DRIVERID_TEA6420	 5	/* audio matrix switch		*/
 #define I2C_DRIVERID_TEA6415C	 6	/* video matrix switch		*/
 #define I2C_DRIVERID_TDA9840	 7	/* stereo sound processor	*/
 #define I2C_DRIVERID_SAA7111A	 8	/* video input processor	*/
-#define I2C_DRIVERID_SAA5281	 9	/* videotext decoder		*/
-#define I2C_DRIVERID_SAA7112	10	/* video decoder, image scaler	*/
-#define I2C_DRIVERID_SAA7120	11	/* video encoder		*/
-#define I2C_DRIVERID_SAA7121	12	/* video encoder		*/
 #define I2C_DRIVERID_SAA7185B	13	/* video encoder		*/
-#define I2C_DRIVERID_CH7003	14	/* digital pc to tv encoder	*/
-#define I2C_DRIVERID_PCF8574A	15	/* i2c expander - 8 bit in/out	*/
-#define I2C_DRIVERID_PCF8582C	16	/* eeprom			*/
-#define I2C_DRIVERID_AT24Cxx	17	/* eeprom 1/2/4/8/16 K		*/
 #define I2C_DRIVERID_TEA6300	18	/* audio mixer			*/
-#define I2C_DRIVERID_BT829	19	/* pc to tv encoder		*/
 #define I2C_DRIVERID_TDA9850	20	/* audio mixer			*/
 #define I2C_DRIVERID_TDA9855	21	/* audio mixer			*/
 #define I2C_DRIVERID_SAA7110	22	/* video decoder		*/
@@ -60,40 +50,19 @@
 #define I2C_DRIVERID_TDA7432	27	/* Stereo sound processor	*/
 #define I2C_DRIVERID_TVMIXER    28      /* Mixer driver for tv cards    */
 #define I2C_DRIVERID_TVAUDIO    29      /* Generic TV sound driver      */
-#define I2C_DRIVERID_DPL3518    30      /* Dolby decoder chip           */
 #define I2C_DRIVERID_TDA9873    31      /* TV sound decoder chip        */
 #define I2C_DRIVERID_TDA9875    32      /* TV sound decoder chip        */
 #define I2C_DRIVERID_PIC16C54_PV9 33    /* Audio mux/ir receiver        */
-
-#define I2C_DRIVERID_SBATT      34     /* Smart Battery Device		*/
-#define I2C_DRIVERID_SBS        35     /* SB System Manager		*/
-#define I2C_DRIVERID_VES1893	36     /* VLSI DVB-S decoder		*/
-#define I2C_DRIVERID_VES1820	37     /* VLSI DVB-C decoder		*/
-#define I2C_DRIVERID_SAA7113	38     /* video decoder			*/
-#define I2C_DRIVERID_TDA8444	39     /* octuple 6-bit DAC             */
 #define I2C_DRIVERID_BT819	40     /* video decoder			*/
 #define I2C_DRIVERID_BT856	41     /* video encoder			*/
 #define I2C_DRIVERID_VPX3220	42     /* video decoder+vbi/vtxt	*/
-#define I2C_DRIVERID_DRP3510	43     /* ADR decoder (Astra Radio)	*/
-#define I2C_DRIVERID_SP5055	44     /* Satellite tuner		*/
-#define I2C_DRIVERID_STV0030	45     /* Multipurpose switch		*/
-#define I2C_DRIVERID_SAA7108	46     /* video decoder, image scaler   */
-#define I2C_DRIVERID_DS1307	47     /* DS1307 real time clock	*/
 #define I2C_DRIVERID_ADV7175	48     /* ADV 7175/7176 video encoder	*/
 #define I2C_DRIVERID_SAA7114	49	/* video decoder		*/
-#define I2C_DRIVERID_ZR36120	50     /* Zoran 36120 video encoder	*/
-#define I2C_DRIVERID_24LC32A	51	/* Microchip 24LC32A 32k EEPROM	*/
-#define I2C_DRIVERID_UDA1342	53	/* UDA1342 audio codec		*/
 #define I2C_DRIVERID_ADV7170	54	/* video encoder		*/
-#define I2C_DRIVERID_MAX1617	56	/* temp sensor			*/
 #define I2C_DRIVERID_SAA7191	57	/* video decoder		*/
 #define I2C_DRIVERID_INDYCAM	58	/* SGI IndyCam			*/
-#define I2C_DRIVERID_BT832	59	/* CMOS camera video processor	*/
-#define I2C_DRIVERID_TDA9887	60	/* TDA988x IF-PLL demodulator	*/
 #define I2C_DRIVERID_OVCAMCHIP	61	/* OmniVision CMOS image sens.	*/
-#define I2C_DRIVERID_TDA7313	62	/* TDA7313 audio processor	*/
 #define I2C_DRIVERID_MAX6900	63	/* MAX6900 real-time clock	*/
-#define I2C_DRIVERID_SAA7114H	64	/* video decoder		*/
 #define I2C_DRIVERID_TDA9874	66	/* TV sound decoder		*/
 #define I2C_DRIVERID_SAA6752HS	67	/* MPEG2 encoder		*/
 #define I2C_DRIVERID_TVEEPROM	68	/* TV EEPROM			*/
@@ -112,7 +81,6 @@
 #define I2C_DRIVERID_DS1672	81	/* Dallas/Maxim DS1672 RTC	*/
 #define I2C_DRIVERID_X1205	82	/* Xicor/Intersil X1205 RTC	*/
 #define I2C_DRIVERID_PCF8563	83	/* Philips PCF8563 RTC		*/
-#define I2C_DRIVERID_RS5C372	84	/* Ricoh RS5C372 RTC		*/
 #define I2C_DRIVERID_BT866	85	/* Conexant bt866 video encoder */
 #define I2C_DRIVERID_KS0127	86	/* Samsung ks0127 video decoder */
 #define I2C_DRIVERID_TLV320AIC23B 87	/* TI TLV320AIC23B audio codec  */
@@ -127,8 +95,6 @@
 #define I2C_DRIVERID_CS5345	96	/* cs5345 audio processor	*/
 
 #define I2C_DRIVERID_I2CDEV	900
-#define I2C_DRIVERID_ARP        902    /* SMBus ARP Client              */
-#define I2C_DRIVERID_ALERT      903    /* SMBus Alert Responder Client  */
 
 /* IDs --   Use DRIVERIDs 1000-1999 for sensors.
    These were originally in sensors.h in the lm_sensors package */
@@ -174,22 +140,15 @@
 
 /* --- Bit algorithm adapters						*/
 #define I2C_HW_B_LP		0x010000 /* Parallel port Philips style */
-#define I2C_HW_B_SER		0x010002 /* Serial line interface */
 #define I2C_HW_B_BT848		0x010005 /* BT848 video boards */
-#define I2C_HW_B_WNV		0x010006 /* Winnov Videums */
 #define I2C_HW_B_VIA		0x010007 /* Via vt82c586b */
 #define I2C_HW_B_HYDRA		0x010008 /* Apple Hydra Mac I/O */
 #define I2C_HW_B_G400		0x010009 /* Matrox G400 */
 #define I2C_HW_B_I810		0x01000a /* Intel I810 */
 #define I2C_HW_B_VOO		0x01000b /* 3dfx Voodoo 3 / Banshee */
-#define I2C_HW_B_PPORT		0x01000c /* Primitive parallel port adapter */
-#define I2C_HW_B_SAVG		0x01000d /* Savage 4 */
 #define I2C_HW_B_SCX200		0x01000e /* Nat'l Semi SCx200 I2C */
 #define I2C_HW_B_RIVA		0x010010 /* Riva based graphics cards */
 #define I2C_HW_B_IOC		0x010011 /* IOC bit-wiggling */
-#define I2C_HW_B_TSUNA		0x010012 /* DEC Tsunami chipset */
-#define I2C_HW_B_OMAHA		0x010014 /* Omaha I2C interface (ARM) */
-#define I2C_HW_B_GUIDE		0x010015 /* Guide bit-basher */
 #define I2C_HW_B_IXP2000	0x010016 /* GPIO on IXP2000 systems */
 #define I2C_HW_B_S3VIA		0x010018 /* S3Via ProSavage adapter */
 #define I2C_HW_B_ZR36067	0x010019 /* Zoran-36057/36067 based boards */
@@ -204,22 +163,11 @@
 #define I2C_HW_B_CX23885	0x010022 /* conexant 23885 based tv cards (bus1) */
 
 /* --- PCF 8584 based algorithms					*/
-#define I2C_HW_P_LP		0x020000 /* Parallel port interface */
-#define I2C_HW_P_ISA		0x020001 /* generic ISA Bus inteface card */
 #define I2C_HW_P_ELEK		0x020002 /* Elektor ISA Bus inteface card */
 
 /* --- PCA 9564 based algorithms */
 #define I2C_HW_A_ISA		0x1a0000 /* generic ISA Bus interface card */
 
-/* --- ACPI Embedded controller algorithms                              */
-#define I2C_HW_ACPI_EC          0x1f0000
-
-/* --- MPC824x PowerPC adapters						*/
-#define I2C_HW_MPC824X		0x100001 /* Motorola 8240 / 8245 */
-
-/* --- MPC8xx PowerPC adapters						*/
-#define I2C_HW_MPC8XX_EPON	0x110000 /* Eponymous MPC8xx I2C adapter */
-
 /* --- PowerPC on-chip adapters						*/
 #define I2C_HW_OCP		0x120000 /* IBM on-chip I2C adapter */
 
@@ -228,7 +176,6 @@
 
 /* --- SGI adapters							*/
 #define I2C_HW_SGI_VINO		0x160000
-#define I2C_HW_SGI_MACE		0x160001
 
 /* --- XSCALE on-chip adapters                          */
 #define I2C_HW_IOP3XX		0x140000
@@ -252,17 +199,10 @@
 #define I2C_HW_SMBUS_W9968CF	0x04000d
 #define I2C_HW_SMBUS_OV511	0x04000e /* OV511(+) USB 1.1 webcam ICs */
 #define I2C_HW_SMBUS_OV518	0x04000f /* OV518(+) USB 1.1 webcam ICs */
-#define I2C_HW_SMBUS_OV519	0x040010 /* OV519 USB 1.1 webcam IC */
 #define I2C_HW_SMBUS_OVFX2	0x040011 /* Cypress/OmniVision FX2 webcam */
 #define I2C_HW_SMBUS_CAFE	0x040012 /* Marvell 88ALP01 "CAFE" cam  */
 #define I2C_HW_SMBUS_ALI1563	0x040013
 
-/* --- ISA pseudo-adapter						*/
-#define I2C_HW_ISA		0x050000
-
-/* --- IPMB adapter						*/
-#define I2C_HW_IPMB		0x0c0000
-
 /* --- MCP107 adapter */
 #define I2C_HW_MPC107		0x0d0000
 
-- 
cgit v1.2.3


From b7a3670131c7662415fa799700fc0bdfe90a54b6 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Sun, 27 Jan 2008 18:14:50 +0100
Subject: i2c-pxa: Add polling transfer

Add polling I2C transfer implementation for PXA I2C. This is needed
for cases where I2C transactions have to occur at times interrups are
disabled.

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Acked-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 arch/arm/mach-pxa/pxa27x.c     |   6 ++
 drivers/i2c/busses/i2c-pxa.c   | 133 +++++++++++++++++++++++++++++++++++++----
 include/asm-arm/arch-pxa/i2c.h |   6 ++
 3 files changed, 133 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c
index 8e126e6b74c3..57efebdc4324 100644
--- a/arch/arm/mach-pxa/pxa27x.c
+++ b/arch/arm/mach-pxa/pxa27x.c
@@ -24,6 +24,7 @@
 #include <asm/arch/ohci.h>
 #include <asm/arch/pm.h>
 #include <asm/arch/dma.h>
+#include <asm/arch/i2c.h>
 
 #include "generic.h"
 #include "devices.h"
@@ -423,6 +424,11 @@ struct platform_device pxa27x_device_i2c_power = {
 	.num_resources	= ARRAY_SIZE(i2c_power_resources),
 };
 
+void __init pxa_set_i2c_power_info(struct i2c_pxa_platform_data *info)
+{
+	pxa27x_device_i2c_power.dev.platform_data = info;
+}
+
 static struct platform_device *devices[] __initdata = {
 	&pxa_device_mci,
 	&pxa_device_udc,
diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
index da3ecf5c591b..2598d29fd7a4 100644
--- a/drivers/i2c/busses/i2c-pxa.c
+++ b/drivers/i2c/busses/i2c-pxa.c
@@ -65,6 +65,7 @@ struct pxa_i2c {
 	unsigned long		iosize;
 
 	int			irq;
+	int			use_pio;
 };
 
 #define _IBMR(i2c)	((i2c)->reg_base + 0)
@@ -163,6 +164,7 @@ static void i2c_pxa_show_state(struct pxa_i2c *i2c, int lno, const char *fname)
 #define eedbg(lvl, x...) do { if ((lvl) < 1) { printk(KERN_DEBUG "" x); } } while(0)
 
 static void i2c_pxa_master_complete(struct pxa_i2c *i2c, int ret);
+static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
 
 static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
 {
@@ -554,6 +556,71 @@ static inline void i2c_pxa_stop_message(struct pxa_i2c *i2c)
 	writel(icr, _ICR(i2c));
 }
 
+static int i2c_pxa_pio_set_master(struct pxa_i2c *i2c)
+{
+	/* make timeout the same as for interrupt based functions */
+	long timeout = 2 * DEF_TIMEOUT;
+
+	/*
+	 * Wait for the bus to become free.
+	 */
+	while (timeout-- && readl(_ISR(i2c)) & (ISR_IBB | ISR_UB)) {
+		udelay(1000);
+		show_state(i2c);
+	}
+
+	if (timeout <= 0) {
+		show_state(i2c);
+		dev_err(&i2c->adap.dev,
+			"i2c_pxa: timeout waiting for bus free\n");
+		return I2C_RETRY;
+	}
+
+	/*
+	 * Set master mode.
+	 */
+	writel(readl(_ICR(i2c)) | ICR_SCLE, _ICR(i2c));
+
+	return 0;
+}
+
+static int i2c_pxa_do_pio_xfer(struct pxa_i2c *i2c,
+			       struct i2c_msg *msg, int num)
+{
+	unsigned long timeout = 500000; /* 5 seconds */
+	int ret = 0;
+
+	ret = i2c_pxa_pio_set_master(i2c);
+	if (ret)
+		goto out;
+
+	i2c->msg = msg;
+	i2c->msg_num = num;
+	i2c->msg_idx = 0;
+	i2c->msg_ptr = 0;
+	i2c->irqlogidx = 0;
+
+	i2c_pxa_start_message(i2c);
+
+	while (timeout-- && i2c->msg_num > 0) {
+		i2c_pxa_handler(0, i2c);
+		udelay(10);
+	}
+
+	i2c_pxa_stop_message(i2c);
+
+	/*
+	 * We place the return code in i2c->msg_idx.
+	 */
+	ret = i2c->msg_idx;
+
+out:
+	if (timeout == 0)
+		i2c_pxa_scream_blue_murder(i2c, "timeout");
+
+	return ret;
+}
+
 /*
  * We are protected by the adapter bus mutex.
  */
@@ -610,6 +677,35 @@ static int i2c_pxa_do_xfer(struct pxa_i2c *i2c, struct i2c_msg *msg, int num)
 	return ret;
 }
 
+static int i2c_pxa_pio_xfer(struct i2c_adapter *adap,
+			    struct i2c_msg msgs[], int num)
+{
+	struct pxa_i2c *i2c = adap->algo_data;
+	int ret, i;
+
+	/* If the I2C controller is disabled we need to reset it
+	  (probably due to a suspend/resume destroying state). We do
+	  this here as we can then avoid worrying about resuming the
+	  controller before its users. */
+	if (!(readl(_ICR(i2c)) & ICR_IUE))
+		i2c_pxa_reset(i2c);
+
+	for (i = adap->retries; i >= 0; i--) {
+		ret = i2c_pxa_do_pio_xfer(i2c, msgs, num);
+		if (ret != I2C_RETRY)
+			goto out;
+
+		if (i2c_debug)
+			dev_dbg(&adap->dev, "Retrying transmission\n");
+		udelay(100);
+	}
+	i2c_pxa_scream_blue_murder(i2c, "exhausted retries");
+	ret = -EREMOTEIO;
+ out:
+	i2c_pxa_set_slave(i2c, ret);
+	return ret;
+}
+
 /*
  * i2c_pxa_master_complete - complete the message and wake up.
  */
@@ -621,7 +717,8 @@ static void i2c_pxa_master_complete(struct pxa_i2c *i2c, int ret)
 	i2c->msg_num = 0;
 	if (ret)
 		i2c->msg_idx = ret;
-	wake_up(&i2c->wait);
+	if (!i2c->use_pio)
+		wake_up(&i2c->wait);
 }
 
 static void i2c_pxa_irq_txempty(struct pxa_i2c *i2c, u32 isr)
@@ -840,6 +937,11 @@ static const struct i2c_algorithm i2c_pxa_algorithm = {
 	.functionality	= i2c_pxa_functionality,
 };
 
+static const struct i2c_algorithm i2c_pxa_pio_algorithm = {
+	.master_xfer	= i2c_pxa_pio_xfer,
+	.functionality	= i2c_pxa_functionality,
+};
+
 static void i2c_pxa_enable(struct platform_device *dev)
 {
 	if (cpu_is_pxa27x()) {
@@ -890,7 +992,6 @@ static int i2c_pxa_probe(struct platform_device *dev)
 	}
 
 	i2c->adap.owner   = THIS_MODULE;
-	i2c->adap.algo    = &i2c_pxa_algorithm;
 	i2c->adap.retries = 5;
 
 	spin_lock_init(&i2c->lock);
@@ -927,20 +1028,26 @@ static int i2c_pxa_probe(struct platform_device *dev)
 	clk_enable(i2c->clk);
 	i2c_pxa_enable(dev);
 
-	ret = request_irq(irq, i2c_pxa_handler, IRQF_DISABLED,
-			  i2c->adap.name, i2c);
-	if (ret)
-		goto ereqirq;
+	if (plat) {
+		i2c->adap.class = plat->class;
+		i2c->use_pio = plat->use_pio;
+	}
+
+	if (i2c->use_pio) {
+		i2c->adap.algo = &i2c_pxa_pio_algorithm;
+	} else {
+		i2c->adap.algo = &i2c_pxa_algorithm;
+		ret = request_irq(irq, i2c_pxa_handler, IRQF_DISABLED,
+				  i2c->adap.name, i2c);
+		if (ret)
+			goto ereqirq;
+	}
 
 	i2c_pxa_reset(i2c);
 
 	i2c->adap.algo_data = i2c;
 	i2c->adap.dev.parent = &dev->dev;
 
-	if (plat) {
-		i2c->adap.class = plat->class;
-	}
-
 	/*
 	 * If "dev->id" is negative we consider it as zero.
 	 * The reason to do so is to avoid sysfs names that only make
@@ -966,7 +1073,8 @@ static int i2c_pxa_probe(struct platform_device *dev)
 	return 0;
 
 eadapt:
-	free_irq(irq, i2c);
+	if (!i2c->use_pio)
+		free_irq(irq, i2c);
 ereqirq:
 	clk_disable(i2c->clk);
 	i2c_pxa_disable(dev);
@@ -986,7 +1094,8 @@ static int i2c_pxa_remove(struct platform_device *dev)
 	platform_set_drvdata(dev, NULL);
 
 	i2c_del_adapter(&i2c->adap);
-	free_irq(i2c->irq, i2c);
+	if (!i2c->use_pio)
+		free_irq(i2c->irq, i2c);
 
 	clk_disable(i2c->clk);
 	clk_put(i2c->clk);
diff --git a/include/asm-arm/arch-pxa/i2c.h b/include/asm-arm/arch-pxa/i2c.h
index e404b233d8a8..80596b013443 100644
--- a/include/asm-arm/arch-pxa/i2c.h
+++ b/include/asm-arm/arch-pxa/i2c.h
@@ -65,7 +65,13 @@ struct i2c_pxa_platform_data {
 	unsigned int		slave_addr;
 	struct i2c_slave_client	*slave;
 	unsigned int		class;
+	int			use_pio;
 };
 
 extern void pxa_set_i2c_info(struct i2c_pxa_platform_data *info);
+
+#ifdef CONFIG_PXA27x
+extern void pxa_set_i2c_power_info(struct i2c_pxa_platform_data *info);
+#endif
+
 #endif
-- 
cgit v1.2.3


From 9b766b814d6a5f31ca1e9da1ebc08164b9352941 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 27 Jan 2008 18:14:51 +0100
Subject: i2c: Stop using the redundant client list

The i2c_adapter.clients list of i2c_client nodes duplicates driver
model state.  This patch starts removing that list, letting us remove
most existing users of those i2c-core lists.

 * The core I2C code now iterates over the driver model's list instead
   of the i2c-internal one in some places where it's safe:
      - Passing a command/ioctl to each client, a mechanims
        used almost exclusively by DVB adapters;
      - Device address checking, in both i2c-core and i2c-dev.

 * Provide i2c_verify_client() to use with driver model iterators.

 * Flag the relevant i2c_adapter and i2c_client fields as deprecated,
   to help prevent new users from appearing.

For the moment the list needs to stick around, since some issues show
up when deleting devices created by legacy I2C drivers.  (They don't
follow standard driver model rules.  Removing those devices can cause
self-deadlocks.)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/i2c-core.c | 78 +++++++++++++++++++++++++++++---------------------
 drivers/i2c/i2c-dev.c  | 29 ++++++++-----------
 include/linux/i2c.h    |  8 ++++--
 3 files changed, 63 insertions(+), 52 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index ddd1b83f44d4..c8bbd672226b 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -199,6 +199,25 @@ static struct bus_type i2c_bus_type = {
 	.resume		= i2c_device_resume,
 };
 
+
+/**
+ * i2c_verify_client - return parameter as i2c_client, or NULL
+ * @dev: device, probably from some driver model iterator
+ *
+ * When traversing the driver model tree, perhaps using driver model
+ * iterators like @device_for_each_child(), you can't assume very much
+ * about the nodes you find.  Use this function to avoid oopses caused
+ * by wrongly treating some non-I2C device as an i2c_client.
+ */
+struct i2c_client *i2c_verify_client(struct device *dev)
+{
+	return (dev->bus == &i2c_bus_type)
+			? to_i2c_client(dev)
+			: NULL;
+}
+EXPORT_SYMBOL(i2c_verify_client);
+
+
 /**
  * i2c_new_device - instantiate an i2c device for use with a new style driver
  * @adap: the adapter managing the device
@@ -670,28 +689,19 @@ EXPORT_SYMBOL(i2c_del_driver);
 
 /* ------------------------------------------------------------------------- */
 
-static int __i2c_check_addr(struct i2c_adapter *adapter, unsigned int addr)
+static int __i2c_check_addr(struct device *dev, void *addrp)
 {
-	struct list_head   *item;
-	struct i2c_client  *client;
+	struct i2c_client	*client = i2c_verify_client(dev);
+	int			addr = *(int *)addrp;
 
-	list_for_each(item,&adapter->clients) {
-		client = list_entry(item, struct i2c_client, list);
-		if (client->addr == addr)
-			return -EBUSY;
-	}
+	if (client && client->addr == addr)
+		return -EBUSY;
 	return 0;
 }
 
 static int i2c_check_addr(struct i2c_adapter *adapter, int addr)
 {
-	int rval;
-
-	mutex_lock(&adapter->clist_lock);
-	rval = __i2c_check_addr(adapter, addr);
-	mutex_unlock(&adapter->clist_lock);
-
-	return rval;
+	return device_for_each_child(&adapter->dev, &addr, __i2c_check_addr);
 }
 
 int i2c_attach_client(struct i2c_client *client)
@@ -700,7 +710,7 @@ int i2c_attach_client(struct i2c_client *client)
 	int res = 0;
 
 	mutex_lock(&adapter->clist_lock);
-	if (__i2c_check_addr(client->adapter, client->addr)) {
+	if (i2c_check_addr(client->adapter, client->addr)) {
 		res = -EBUSY;
 		goto out_unlock;
 	}
@@ -804,24 +814,28 @@ void i2c_release_client(struct i2c_client *client)
 }
 EXPORT_SYMBOL(i2c_release_client);
 
+struct i2c_cmd_arg {
+	unsigned	cmd;
+	void		*arg;
+};
+
+static int i2c_cmd(struct device *dev, void *_arg)
+{
+	struct i2c_client	*client = i2c_verify_client(dev);
+	struct i2c_cmd_arg	*arg = _arg;
+
+	if (client && client->driver && client->driver->command)
+		client->driver->command(client, arg->cmd, arg->arg);
+	return 0;
+}
+
 void i2c_clients_command(struct i2c_adapter *adap, unsigned int cmd, void *arg)
 {
-	struct list_head  *item;
-	struct i2c_client *client;
+	struct i2c_cmd_arg	cmd_arg;
 
-	mutex_lock(&adap->clist_lock);
-	list_for_each(item,&adap->clients) {
-		client = list_entry(item, struct i2c_client, list);
-		if (!try_module_get(client->driver->driver.owner))
-			continue;
-		if (NULL != client->driver->command) {
-			mutex_unlock(&adap->clist_lock);
-			client->driver->command(client,cmd,arg);
-			mutex_lock(&adap->clist_lock);
-		}
-		module_put(client->driver->driver.owner);
-       }
-       mutex_unlock(&adap->clist_lock);
+	cmd_arg.cmd = cmd;
+	cmd_arg.arg = arg;
+	device_for_each_child(&adap->dev, &cmd_arg, i2c_cmd);
 }
 EXPORT_SYMBOL(i2c_clients_command);
 
@@ -1087,7 +1101,7 @@ i2c_new_probed_device(struct i2c_adapter *adap,
 		}
 
 		/* Check address availability */
-		if (__i2c_check_addr(adap, addr_list[i])) {
+		if (i2c_check_addr(adap, addr_list[i])) {
 			dev_dbg(&adap->dev, "Address 0x%02x already in "
 				"use, not probing\n", addr_list[i]);
 			continue;
diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index df540d5dfaf4..393e679d9faa 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -182,27 +182,22 @@ static ssize_t i2cdev_write (struct file *file, const char __user *buf, size_t c
 	return ret;
 }
 
+static int i2cdev_check(struct device *dev, void *addrp)
+{
+	struct i2c_client *client = i2c_verify_client(dev);
+
+	if (!client || client->addr != *(unsigned int *)addrp)
+		return 0;
+
+	return dev->driver ? -EBUSY : 0;
+}
+
 /* This address checking function differs from the one in i2c-core
    in that it considers an address with a registered device, but no
-   bound driver, as NOT busy. */
+   driver bound to it, as NOT busy. */
 static int i2cdev_check_addr(struct i2c_adapter *adapter, unsigned int addr)
 {
-	struct list_head *item;
-	struct i2c_client *client;
-	int res = 0;
-
-	mutex_lock(&adapter->clist_lock);
-	list_for_each(item, &adapter->clients) {
-		client = list_entry(item, struct i2c_client, list);
-		if (client->addr == addr) {
-			if (client->driver)
-				res = -EBUSY;
-			break;
-		}
-	}
-	mutex_unlock(&adapter->clist_lock);
-
-	return res;
+	return device_for_each_child(&adapter->dev, &addr, i2cdev_check);
 }
 
 static int i2cdev_ioctl(struct inode *inode, struct file *file,
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 0fc59efd80e4..731928ae972c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -158,7 +158,7 @@ struct i2c_driver {
  * @irq: indicates the IRQ generated by this device (if any)
  * @driver_name: Identifies new-style driver used with this device; also
  *	used as the module name for hotplug/coldplug modprobe support.
- * @list: list of active/busy clients
+ * @list: list of active/busy clients (DEPRECATED)
  * @released: used to synchronize client releases & detaches and references
  *
  * An i2c_client identifies a single device (i.e. chip) connected to an
@@ -176,11 +176,13 @@ struct i2c_client {
 	struct device dev;		/* the device structure		*/
 	int irq;			/* irq issued by device (or -1) */
 	char driver_name[KOBJ_NAME_LEN];
-	struct list_head list;
+	struct list_head list;		/* DEPRECATED */
 	struct completion released;
 };
 #define to_i2c_client(d) container_of(d, struct i2c_client, dev)
 
+extern struct i2c_client *i2c_verify_client(struct device *dev);
+
 static inline struct i2c_client *kobj_to_i2c_client(struct kobject *kobj)
 {
 	struct device * const dev = container_of(kobj, struct device, kobj);
@@ -315,7 +317,7 @@ struct i2c_adapter {
 	struct device dev;		/* the adapter device */
 
 	int nr;
-	struct list_head clients;
+	struct list_head clients;	/* DEPRECATED */
 	char name[48];
 	struct completion dev_released;
 };
-- 
cgit v1.2.3


From e9f1373b643887f63878d1169b310c9acc534cd5 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 27 Jan 2008 18:14:52 +0100
Subject: i2c: Add i2c_new_dummy() utility

This adds a i2c_new_dummy() primitive to help work with devices
that consume multiple addresses, which include many I2C eeproms
and at least one RTC.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/i2c-core.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++++-
 include/linux/i2c.h    |  6 +++++
 2 files changed, 64 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index cd3fcb85ca7f..96da22e9a5a4 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -296,6 +296,50 @@ void i2c_unregister_device(struct i2c_client *client)
 EXPORT_SYMBOL_GPL(i2c_unregister_device);
 
 
+static int dummy_nop(struct i2c_client *client)
+{
+	return 0;
+}
+
+static struct i2c_driver dummy_driver = {
+	.driver.name	= "dummy",
+	.probe		= dummy_nop,
+	.remove		= dummy_nop,
+};
+
+/**
+ * i2c_new_dummy - return a new i2c device bound to a dummy driver
+ * @adapter: the adapter managing the device
+ * @address: seven bit address to be used
+ * @type: optional label used for i2c_client.name
+ * Context: can sleep
+ *
+ * This returns an I2C client bound to the "dummy" driver, intended for use
+ * with devices that consume multiple addresses.  Examples of such chips
+ * include various EEPROMS (like 24c04 and 24c08 models).
+ *
+ * These dummy devices have two main uses.  First, most I2C and SMBus calls
+ * except i2c_transfer() need a client handle; the dummy will be that handle.
+ * And second, this prevents the specified address from being bound to a
+ * different driver.
+ *
+ * This returns the new i2c client, which should be saved for later use with
+ * i2c_unregister_device(); or NULL to indicate an error.
+ */
+struct i2c_client *
+i2c_new_dummy(struct i2c_adapter *adapter, u16 address, const char *type)
+{
+	struct i2c_board_info info = {
+		.driver_name	= "dummy",
+		.addr		= address,
+	};
+
+	if (type)
+		strlcpy(info.type, type, sizeof info.type);
+	return i2c_new_device(adapter, &info);
+}
+EXPORT_SYMBOL_GPL(i2c_new_dummy);
+
 /* ------------------------------------------------------------------------- */
 
 /* I2C bus adapters -- one roots each I2C or SMBUS segment */
@@ -841,11 +885,24 @@ static int __init i2c_init(void)
 	retval = bus_register(&i2c_bus_type);
 	if (retval)
 		return retval;
-	return class_register(&i2c_adapter_class);
+	retval = class_register(&i2c_adapter_class);
+	if (retval)
+		goto bus_err;
+	retval = i2c_add_driver(&dummy_driver);
+	if (retval)
+		goto class_err;
+	return 0;
+
+class_err:
+	class_unregister(&i2c_adapter_class);
+bus_err:
+	bus_unregister(&i2c_bus_type);
+	return retval;
 }
 
 static void __exit i2c_exit(void)
 {
+	i2c_del_driver(&dummy_driver);
 	class_unregister(&i2c_adapter_class);
 	bus_unregister(&i2c_bus_type);
 }
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 731928ae972c..76014f8f3c60 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -259,6 +259,12 @@ i2c_new_probed_device(struct i2c_adapter *adap,
 		      struct i2c_board_info *info,
 		      unsigned short const *addr_list);
 
+/* For devices that use several addresses, use i2c_new_dummy() to make
+ * client handles for the extra addresses.
+ */
+extern struct i2c_client *
+i2c_new_dummy(struct i2c_adapter *adap, u16 address, const char *type);
+
 extern void i2c_unregister_device(struct i2c_client *);
 
 /* Mainboard arch_initcall() code should register all its I2C devices.
-- 
cgit v1.2.3


From 1f9ffc049d7a88c8489b883b6fc0a25185062002 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Sun, 27 Jan 2008 10:29:20 -0800
Subject: Driver core: add bus_find_device_by_name function

The driver core, and some other parts of the kernel just want to find a
device based on a name for a specific bus.  Give them a simple wrapper
to prevent them from having to always roll their own.

This will be used in the PPC patch later in this series.

Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 41 +++++++++++++++++++++++++++++------------
 include/linux/device.h |  3 +++
 2 files changed, 32 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index f484495b2ad1..055989e94799 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -163,15 +163,6 @@ static struct kset *bus_kset;
 
 #ifdef CONFIG_HOTPLUG
 /* Manually detach a device from its associated driver. */
-static int driver_helper(struct device *dev, void *data)
-{
-	const char *name = data;
-
-	if (strcmp(name, dev->bus_id) == 0)
-		return 1;
-	return 0;
-}
-
 static ssize_t driver_unbind(struct device_driver *drv,
 			     const char *buf, size_t count)
 {
@@ -179,7 +170,7 @@ static ssize_t driver_unbind(struct device_driver *drv,
 	struct device *dev;
 	int err = -ENODEV;
 
-	dev = bus_find_device(bus, NULL, (void *)buf, driver_helper);
+	dev = bus_find_device_by_name(bus, NULL, buf);
 	if (dev && dev->driver == drv) {
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
@@ -206,7 +197,7 @@ static ssize_t driver_bind(struct device_driver *drv,
 	struct device *dev;
 	int err = -ENODEV;
 
-	dev = bus_find_device(bus, NULL, (void *)buf, driver_helper);
+	dev = bus_find_device_by_name(bus, NULL, buf);
 	if (dev && dev->driver == NULL) {
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
@@ -250,7 +241,7 @@ static ssize_t store_drivers_probe(struct bus_type *bus,
 {
 	struct device *dev;
 
-	dev = bus_find_device(bus, NULL, (void *)buf, driver_helper);
+	dev = bus_find_device_by_name(bus, NULL, buf);
 	if (!dev)
 		return -ENODEV;
 	if (bus_rescan_devices_helper(dev, NULL) != 0)
@@ -338,6 +329,32 @@ struct device *bus_find_device(struct bus_type *bus,
 }
 EXPORT_SYMBOL_GPL(bus_find_device);
 
+static int match_name(struct device *dev, void *data)
+{
+	const char *name = data;
+
+	if (strcmp(name, dev->bus_id) == 0)
+		return 1;
+	return 0;
+}
+
+/**
+ * bus_find_device_by_name - device iterator for locating a particular device of a specific name
+ * @bus: bus type
+ * @start: Device to begin with
+ * @name: name of the device to match
+ *
+ * This is similar to the bus_find_device() function above, but it handles
+ * searching by a name automatically, no need to write another strcmp matching
+ * function.
+ */
+struct device *bus_find_device_by_name(struct bus_type *bus,
+				       struct device *start, const char *name)
+{
+	return bus_find_device(bus, start, (void *)name, match_name);
+}
+EXPORT_SYMBOL_GPL(bus_find_device_by_name);
+
 static struct device_driver *next_driver(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
diff --git a/include/linux/device.h b/include/linux/device.h
index 1880208964d6..db375be333c7 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -84,6 +84,9 @@ int bus_for_each_dev(struct bus_type *bus, struct device *start, void *data,
 struct device *bus_find_device(struct bus_type *bus, struct device *start,
 			       void *data,
 			       int (*match)(struct device *dev, void *data));
+struct device *bus_find_device_by_name(struct bus_type *bus,
+				       struct device *start,
+				       const char *name);
 
 int __must_check bus_for_each_drv(struct bus_type *bus,
 				  struct device_driver *start, void *data,
-- 
cgit v1.2.3


From 5a668651bf0da3891c46ea2cfcac227ded783a5a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:42:46 +0900
Subject: sh: Split out PXSEG segmentation per-CPU family.

The CPU family abstraction already exists, so move out the PXSEG
definition for each one. SH-2A already has this special cased,
and SH-5 will as well.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/addrspace.h          | 31 +++++++++++++------------------
 include/asm-sh/cpu-sh2/addrspace.h  |  7 +++++--
 include/asm-sh/cpu-sh2a/addrspace.h | 11 ++++++++++-
 include/asm-sh/cpu-sh3/addrspace.h  |  7 +++++--
 include/asm-sh/cpu-sh4/addrspace.h  |  6 ++++++
 5 files changed, 39 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index b860218e402e..cb9da9992fa0 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -9,24 +9,13 @@
  */
 #ifndef __ASM_SH_ADDRSPACE_H
 #define __ASM_SH_ADDRSPACE_H
+
 #ifdef __KERNEL__
 
 #include <asm/cpu/addrspace.h>
 
-/* Memory segments (32bit Privileged mode addresses)  */
-#ifndef CONFIG_CPU_SH2A
-#define P0SEG		0x00000000
-#define P1SEG		0x80000000
-#define P2SEG		0xa0000000
-#define P3SEG		0xc0000000
-#define P4SEG		0xe0000000
-#else
-#define P0SEG		0x00000000
-#define P1SEG		0x00000000
-#define P2SEG		0x20000000
-#define P3SEG		0x00000000
-#define P4SEG 		0x80000000
-#endif
+/* If this CPU supports segmentation, hook up the helpers */
+#ifdef P1SEG
 
 /* Returns the privileged segment base of a given address  */
 #define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
@@ -37,10 +26,16 @@
 /*
  * Map an address to a certain privileged segment
  */
-#define P1SEGADDR(a)	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P1SEG))
-#define P2SEGADDR(a)	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P2SEG))
-#define P3SEGADDR(a)	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P3SEG))
-#define P4SEGADDR(a)	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P4SEG))
+#define P1SEGADDR(a)	\
+	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P1SEG))
+#define P2SEGADDR(a)	\
+	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P2SEG))
+#define P3SEGADDR(a)	\
+	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P3SEG))
+#define P4SEGADDR(a)	\
+	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P4SEG))
+
+#endif /* P1SEG */
 
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_ADDRSPACE_H */
diff --git a/include/asm-sh/cpu-sh2/addrspace.h b/include/asm-sh/cpu-sh2/addrspace.h
index 8706c903c5a0..2b9ab93efa4e 100644
--- a/include/asm-sh/cpu-sh2/addrspace.h
+++ b/include/asm-sh/cpu-sh2/addrspace.h
@@ -10,7 +10,10 @@
 #ifndef __ASM_CPU_SH2_ADDRSPACE_H
 #define __ASM_CPU_SH2_ADDRSPACE_H
 
-/* Should fill here */
+#define P0SEG		0x00000000
+#define P1SEG		0x80000000
+#define P2SEG		0xa0000000
+#define P3SEG		0xc0000000
+#define P4SEG		0xe0000000
 
 #endif /* __ASM_CPU_SH2_ADDRSPACE_H */
-
diff --git a/include/asm-sh/cpu-sh2a/addrspace.h b/include/asm-sh/cpu-sh2a/addrspace.h
index 3d2e9aa21522..795ddd6856a3 100644
--- a/include/asm-sh/cpu-sh2a/addrspace.h
+++ b/include/asm-sh/cpu-sh2a/addrspace.h
@@ -1 +1,10 @@
-#include <asm/cpu-sh2/addrspace.h>
+#ifndef __ASM_SH_CPU_SH2A_ADDRSPACE_H
+#define __ASM_SH_CPU_SH2A_ADDRSPACE_H
+
+#define P0SEG		0x00000000
+#define P1SEG		0x00000000
+#define P2SEG		0x20000000
+#define P3SEG		0x00000000
+#define P4SEG		0x80000000
+
+#endif /* __ASM_SH_CPU_SH2A_ADDRSPACE_H */
diff --git a/include/asm-sh/cpu-sh3/addrspace.h b/include/asm-sh/cpu-sh3/addrspace.h
index 872e9e1b548c..0f94726c7d62 100644
--- a/include/asm-sh/cpu-sh3/addrspace.h
+++ b/include/asm-sh/cpu-sh3/addrspace.h
@@ -10,7 +10,10 @@
 #ifndef __ASM_CPU_SH3_ADDRSPACE_H
 #define __ASM_CPU_SH3_ADDRSPACE_H
 
-/* Should fill here */
+#define P0SEG		0x00000000
+#define P1SEG		0x80000000
+#define P2SEG		0xa0000000
+#define P3SEG		0xc0000000
+#define P4SEG		0xe0000000
 
 #endif /* __ASM_CPU_SH3_ADDRSPACE_H */
-
diff --git a/include/asm-sh/cpu-sh4/addrspace.h b/include/asm-sh/cpu-sh4/addrspace.h
index bb2e1b03060c..a3fa733c1c7d 100644
--- a/include/asm-sh/cpu-sh4/addrspace.h
+++ b/include/asm-sh/cpu-sh4/addrspace.h
@@ -10,6 +10,12 @@
 #ifndef __ASM_CPU_SH4_ADDRSPACE_H
 #define __ASM_CPU_SH4_ADDRSPACE_H
 
+#define P0SEG		0x00000000
+#define P1SEG		0x80000000
+#define P2SEG		0xa0000000
+#define P3SEG		0xc0000000
+#define P4SEG		0xe0000000
+
 /* Detailed P4SEG  */
 #define P4SEG_STORE_QUE	(P4SEG)
 #define P4SEG_IC_ADDR	0xf0000000
-- 
cgit v1.2.3


From 8d5fb297cc8f9f7de2840864e497bc38330abba6 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:44:09 +0900
Subject: sh: Split out cache status bits per-CPU family.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cache.h          | 5 -----
 include/asm-sh/cpu-sh2/cache.h  | 5 +++++
 include/asm-sh/cpu-sh2a/cache.h | 5 +++++
 include/asm-sh/cpu-sh3/cache.h  | 5 +++++
 include/asm-sh/cpu-sh4/cache.h  | 5 +++++
 5 files changed, 20 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/cache.h b/include/asm-sh/cache.h
index 01e5cf51ba9b..083419f47c65 100644
--- a/include/asm-sh/cache.h
+++ b/include/asm-sh/cache.h
@@ -12,11 +12,6 @@
 #include <linux/init.h>
 #include <asm/cpu/cache.h>
 
-#define SH_CACHE_VALID		1
-#define SH_CACHE_UPDATED	2
-#define SH_CACHE_COMBINED	4
-#define SH_CACHE_ASSOC		8
-
 #define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
 
 #define __read_mostly __attribute__((__section__(".data.read_mostly")))
diff --git a/include/asm-sh/cpu-sh2/cache.h b/include/asm-sh/cpu-sh2/cache.h
index f02ba7a672b2..66388ce16c30 100644
--- a/include/asm-sh/cpu-sh2/cache.h
+++ b/include/asm-sh/cpu-sh2/cache.h
@@ -12,6 +12,11 @@
 
 #define L1_CACHE_SHIFT	4
 
+#define SH_CACHE_VALID		1
+#define SH_CACHE_UPDATED	2
+#define SH_CACHE_COMBINED	4
+#define SH_CACHE_ASSOC		8
+
 #if defined(CONFIG_CPU_SUBTYPE_SH7619)
 #define CCR1		0xffffffec
 #define CCR		CCR1
diff --git a/include/asm-sh/cpu-sh2a/cache.h b/include/asm-sh/cpu-sh2a/cache.h
index 3e4b9e480982..d88774169b58 100644
--- a/include/asm-sh/cpu-sh2a/cache.h
+++ b/include/asm-sh/cpu-sh2a/cache.h
@@ -12,6 +12,11 @@
 
 #define L1_CACHE_SHIFT	4
 
+#define SH_CACHE_VALID		1
+#define SH_CACHE_UPDATED	2
+#define SH_CACHE_COMBINED	4
+#define SH_CACHE_ASSOC		8
+
 #define CCR1		0xfffc1000
 #define CCR2		0xfffc1004
 
diff --git a/include/asm-sh/cpu-sh3/cache.h b/include/asm-sh/cpu-sh3/cache.h
index 255016fc91f0..77dd45d82414 100644
--- a/include/asm-sh/cpu-sh3/cache.h
+++ b/include/asm-sh/cpu-sh3/cache.h
@@ -12,6 +12,11 @@
 
 #define L1_CACHE_SHIFT	4
 
+#define SH_CACHE_VALID		1
+#define SH_CACHE_UPDATED	2
+#define SH_CACHE_COMBINED	4
+#define SH_CACHE_ASSOC		8
+
 #define CCR		0xffffffec	/* Address of Cache Control Register */
 
 #define CCR_CACHE_CE	0x01	/* Cache Enable */
diff --git a/include/asm-sh/cpu-sh4/cache.h b/include/asm-sh/cpu-sh4/cache.h
index f92b20a0983d..1c61ebf5c8e3 100644
--- a/include/asm-sh/cpu-sh4/cache.h
+++ b/include/asm-sh/cpu-sh4/cache.h
@@ -12,6 +12,11 @@
 
 #define L1_CACHE_SHIFT	5
 
+#define SH_CACHE_VALID		1
+#define SH_CACHE_UPDATED	2
+#define SH_CACHE_COMBINED	4
+#define SH_CACHE_ASSOC		8
+
 #define CCR		0xff00001c	/* Address of Cache Control Register */
 #define CCR_CACHE_OCE	0x0001	/* Operand Cache Enable */
 #define CCR_CACHE_WT	0x0002	/* Write-Through (for P0,U0,P3) (else writeback)*/
-- 
cgit v1.2.3


From 34cd6d3ab47284dbebea03585ead6a0d5671b75c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:54:58 +0900
Subject: sh: Add SH-5 support to asm/module.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/module.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/module.h b/include/asm-sh/module.h
index 118d5a2b228f..46eccd331660 100644
--- a/include/asm-sh/module.h
+++ b/include/asm-sh/module.h
@@ -20,6 +20,8 @@ struct mod_arch_specific {
 #  define MODULE_PROC_FAMILY "SH3LE "
 # elif defined  CONFIG_CPU_SH4
 #  define MODULE_PROC_FAMILY "SH4LE "
+# elif defined  CONFIG_CPU_SH5
+#  define MODULE_PROC_FAMILY "SH5LE "
 # else
 #  error unknown processor family
 # endif
@@ -30,6 +32,8 @@ struct mod_arch_specific {
 #  define MODULE_PROC_FAMILY "SH3BE "
 # elif defined  CONFIG_CPU_SH4
 #  define MODULE_PROC_FAMILY "SH4BE "
+# elif defined  CONFIG_CPU_SH5
+#  define MODULE_PROC_FAMILY "SH5BE "
 # else
 #  error unknown processor family
 # endif
-- 
cgit v1.2.3


From 5a4a5bd127c147aaa16aefef856f4cb28e92cec1 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:55:39 +0900
Subject: sh: Fix up fixmap location for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/fixmap.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/fixmap.h b/include/asm-sh/fixmap.h
index 8a566177ad96..09463cd9bbb9 100644
--- a/include/asm-sh/fixmap.h
+++ b/include/asm-sh/fixmap.h
@@ -73,7 +73,11 @@ extern void __set_fixmap(enum fixed_addresses idx,
  * the start of the fixmap, and leave one page empty
  * at the top of mem..
  */
+#ifdef CONFIG_SUPERH32
 #define FIXADDR_TOP	(P4SEG - PAGE_SIZE)
+#else
+#define FIXADDR_TOP	(0xff000000 - PAGE_SIZE)
+#endif
 #define FIXADDR_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
 #define FIXADDR_START	(FIXADDR_TOP - FIXADDR_SIZE)
 
-- 
cgit v1.2.3


From e1cd93ea44cb55969e122758c8cc2ddfe21b74b3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:58:00 +0900
Subject: sh: BUGFLAG_WARNING needs GENERIC_BUG.

Move the HAVE_ARCH_BUG/HAVE_ARCH_WARN_ON definitions underneath
CONFIG_GENERIC_BUG. This is needed for BUGFLAG_WARNING usage.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/bug.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/bug.h b/include/asm-sh/bug.h
index a78d482e8b2f..afb1c2723ec7 100644
--- a/include/asm-sh/bug.h
+++ b/include/asm-sh/bug.h
@@ -3,7 +3,7 @@
 
 #define TRAPA_BUG_OPCODE	0xc33e	/* trapa #0x3e */
 
-#ifdef CONFIG_BUG
+#ifdef CONFIG_GENERIC_BUG
 #define HAVE_ARCH_BUG
 #define HAVE_ARCH_WARN_ON
 
@@ -77,7 +77,7 @@ struct pt_regs;
 /* arch/sh/kernel/traps.c */
 void handle_BUG(struct pt_regs *);
 
-#endif /* CONFIG_BUG */
+#endif /* CONFIG_GENERIC_BUG */
 
 #include <asm-generic/bug.h>
 
-- 
cgit v1.2.3


From a096a7e4762f685364df5ca03394eb63bbdb93df Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 18:58:52 +0900
Subject: sh: Add addrspace.h segmentation stub for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/addrspace.h | 6 ++++++
 1 file changed, 6 insertions(+)
 create mode 100644 include/asm-sh/cpu-sh5/addrspace.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/addrspace.h b/include/asm-sh/cpu-sh5/addrspace.h
new file mode 100644
index 000000000000..844c5210e8ed
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/addrspace.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_SH_CPU_SH5_ADDRSPACE_H
+#define __ASM_SH_CPU_SH5_ADDRSPACE_H
+
+/* No segmentation.. */
+
+#endif /* __ASM_SH_CPU_SH5_ADDRSPACE_H */
-- 
cgit v1.2.3


From d752542ade337f059d12c59c4bc9c312befa1f1e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 19:00:27 +0900
Subject: sh: Add cache definitions for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/cache.h |  94 ++++++++++++++++++++++++++++
 include/asm-sh64/cache.h       | 139 -----------------------------------------
 2 files changed, 94 insertions(+), 139 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh5/cache.h
 delete mode 100644 include/asm-sh64/cache.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/cache.h b/include/asm-sh/cpu-sh5/cache.h
new file mode 100644
index 000000000000..2d1f9c28b923
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/cache.h
@@ -0,0 +1,94 @@
+#ifndef __ASM_SH64_CACHE_H
+#define __ASM_SH64_CACHE_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/cache.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003, 2004  Paul Mundt
+ *
+ */
+
+#define L1_CACHE_SHIFT		5
+
+/* Valid and Dirty bits */
+#define SH_CACHE_VALID		(1LL<<0)
+#define SH_CACHE_UPDATED	(1LL<<57)
+
+/* Cache flags */
+#define SH_CACHE_MODE_WT	(1LL<<0)
+#define SH_CACHE_MODE_WB	(1LL<<1)
+
+/*
+ * Control Registers.
+ */
+#define ICCR_BASE	0x01600000	/* Instruction Cache Control Register */
+#define ICCR_REG0	0		/* Register 0 offset */
+#define ICCR_REG1	1		/* Register 1 offset */
+#define ICCR0		ICCR_BASE+ICCR_REG0
+#define ICCR1		ICCR_BASE+ICCR_REG1
+
+#define ICCR0_OFF	0x0		/* Set ICACHE off */
+#define ICCR0_ON	0x1		/* Set ICACHE on */
+#define ICCR0_ICI	0x2		/* Invalidate all in IC */
+
+#define ICCR1_NOLOCK	0x0		/* Set No Locking */
+
+#define OCCR_BASE	0x01E00000	/* Operand Cache Control Register */
+#define OCCR_REG0	0		/* Register 0 offset */
+#define OCCR_REG1	1		/* Register 1 offset */
+#define OCCR0		OCCR_BASE+OCCR_REG0
+#define OCCR1		OCCR_BASE+OCCR_REG1
+
+#define OCCR0_OFF	0x0		/* Set OCACHE off */
+#define OCCR0_ON	0x1		/* Set OCACHE on */
+#define OCCR0_OCI	0x2		/* Invalidate all in OC */
+#define OCCR0_WT	0x4		/* Set OCACHE in WT Mode */
+#define OCCR0_WB	0x0		/* Set OCACHE in WB Mode */
+
+#define OCCR1_NOLOCK	0x0		/* Set No Locking */
+
+/*
+ * SH-5
+ * A bit of description here, for neff=32.
+ *
+ *                               |<--- tag  (19 bits) --->|
+ * +-----------------------------+-----------------+------+----------+------+
+ * |                             |                 | ways |set index |offset|
+ * +-----------------------------+-----------------+------+----------+------+
+ *                                ^                 2 bits   8 bits   5 bits
+ *                                +- Bit 31
+ *
+ * Cacheline size is based on offset: 5 bits = 32 bytes per line
+ * A cache line is identified by a tag + set but OCACHETAG/ICACHETAG
+ * have a broader space for registers. These are outlined by
+ * CACHE_?C_*_STEP below.
+ *
+ */
+
+/* Instruction cache */
+#define CACHE_IC_ADDRESS_ARRAY 0x01000000
+
+/* Operand Cache */
+#define CACHE_OC_ADDRESS_ARRAY 0x01800000
+
+/* These declarations relate to cache 'synonyms' in the operand cache.  A
+   'synonym' occurs where effective address bits overlap between those used for
+   indexing the cache sets and those passed to the MMU for translation.  In the
+   case of SH5-101 & SH5-103, only bit 12 is affected for 4k pages. */
+
+#define CACHE_OC_N_SYNBITS  1               /* Number of synonym bits */
+#define CACHE_OC_SYN_SHIFT  12
+/* Mask to select synonym bit(s) */
+#define CACHE_OC_SYN_MASK   (((1UL<<CACHE_OC_N_SYNBITS)-1)<<CACHE_OC_SYN_SHIFT)
+
+/*
+ * Instruction cache can't be invalidated based on physical addresses.
+ * No Instruction Cache defines required, then.
+ */
+
+#endif /* __ASM_SH64_CACHE_H */
diff --git a/include/asm-sh64/cache.h b/include/asm-sh64/cache.h
deleted file mode 100644
index a4f36f0036e1..000000000000
--- a/include/asm-sh64/cache.h
+++ /dev/null
@@ -1,139 +0,0 @@
-#ifndef __ASM_SH64_CACHE_H
-#define __ASM_SH64_CACHE_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/cache.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- *
- */
-#include <asm/cacheflush.h>
-
-#define L1_CACHE_SHIFT		5
-/* bytes per L1 cache line */
-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
-#define L1_CACHE_ALIGN_MASK	(~(L1_CACHE_BYTES - 1))
-#define L1_CACHE_ALIGN(x)	(((x)+(L1_CACHE_BYTES - 1)) & L1_CACHE_ALIGN_MASK)
-#define L1_CACHE_SIZE_BYTES	(L1_CACHE_BYTES << 10)
-
-#ifdef MODULE
-#define __cacheline_aligned __attribute__((__aligned__(L1_CACHE_BYTES)))
-#else
-#define __cacheline_aligned					\
-  __attribute__((__aligned__(L1_CACHE_BYTES),			\
-		 __section__(".data.cacheline_aligned")))
-#endif
-
-/*
- * Control Registers.
- */
-#define ICCR_BASE	0x01600000	/* Instruction Cache Control Register */
-#define ICCR_REG0	0		/* Register 0 offset */
-#define ICCR_REG1	1		/* Register 1 offset */
-#define ICCR0		ICCR_BASE+ICCR_REG0
-#define ICCR1		ICCR_BASE+ICCR_REG1
-
-#define ICCR0_OFF	0x0		/* Set ICACHE off */
-#define ICCR0_ON	0x1		/* Set ICACHE on */
-#define ICCR0_ICI	0x2		/* Invalidate all in IC */
-
-#define ICCR1_NOLOCK	0x0		/* Set No Locking */
-
-#define OCCR_BASE	0x01E00000	/* Operand Cache Control Register */
-#define OCCR_REG0	0		/* Register 0 offset */
-#define OCCR_REG1	1		/* Register 1 offset */
-#define OCCR0		OCCR_BASE+OCCR_REG0
-#define OCCR1		OCCR_BASE+OCCR_REG1
-
-#define OCCR0_OFF	0x0		/* Set OCACHE off */
-#define OCCR0_ON	0x1		/* Set OCACHE on */
-#define OCCR0_OCI	0x2		/* Invalidate all in OC */
-#define OCCR0_WT	0x4		/* Set OCACHE in WT Mode */
-#define OCCR0_WB	0x0		/* Set OCACHE in WB Mode */
-
-#define OCCR1_NOLOCK	0x0		/* Set No Locking */
-
-
-/*
- * SH-5
- * A bit of description here, for neff=32.
- *
- *                               |<--- tag  (19 bits) --->|
- * +-----------------------------+-----------------+------+----------+------+
- * |                             |                 | ways |set index |offset|
- * +-----------------------------+-----------------+------+----------+------+
- *                                ^                 2 bits   8 bits   5 bits
- *                                +- Bit 31
- *
- * Cacheline size is based on offset: 5 bits = 32 bytes per line
- * A cache line is identified by a tag + set but OCACHETAG/ICACHETAG
- * have a broader space for registers. These are outlined by
- * CACHE_?C_*_STEP below.
- *
- */
-
-/* Valid and Dirty bits */
-#define SH_CACHE_VALID		(1LL<<0)
-#define SH_CACHE_UPDATED	(1LL<<57)
-
-/* Cache flags */
-#define SH_CACHE_MODE_WT	(1LL<<0)
-#define SH_CACHE_MODE_WB	(1LL<<1)
-
-#ifndef __ASSEMBLY__
-
-/*
- * Cache information structure.
- *
- * Defined for both I and D cache, per-processor.
- */
-struct cache_info {
-	unsigned int ways;
-	unsigned int sets;
-	unsigned int linesz;
-
-	unsigned int way_shift;
-	unsigned int entry_shift;
-	unsigned int set_shift;
-	unsigned int way_step_shift;
-	unsigned int asid_shift;
-
-	unsigned int way_ofs;
-
-	unsigned int asid_mask;
-	unsigned int idx_mask;
-	unsigned int epn_mask;
-
-	unsigned long flags;
-};
-
-#endif /* __ASSEMBLY__ */
-
-/* Instruction cache */
-#define CACHE_IC_ADDRESS_ARRAY 0x01000000
-
-/* Operand Cache */
-#define CACHE_OC_ADDRESS_ARRAY 0x01800000
-
-/* These declarations relate to cache 'synonyms' in the operand cache.  A
-   'synonym' occurs where effective address bits overlap between those used for
-   indexing the cache sets and those passed to the MMU for translation.  In the
-   case of SH5-101 & SH5-103, only bit 12 is affected for 4k pages. */
-
-#define CACHE_OC_N_SYNBITS  1               /* Number of synonym bits */
-#define CACHE_OC_SYN_SHIFT  12
-/* Mask to select synonym bit(s) */
-#define CACHE_OC_SYN_MASK   (((1UL<<CACHE_OC_N_SYNBITS)-1)<<CACHE_OC_SYN_SHIFT)
-
-
-/*
- * Instruction cache can't be invalidated based on physical addresses.
- * No Instruction Cache defines required, then.
- */
-
-#endif /* __ASM_SH64_CACHE_H */
-- 
cgit v1.2.3


From 0fa70efbd45a6c4fd946fa71c0a609f2c509d07c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Nov 2007 19:08:28 +0900
Subject: sh: Correct SH-5 instruction size value.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 4faa2fb88616..288abeb5476f 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -255,8 +255,10 @@ static inline void *set_exception_table_evt(unsigned int evt, void *handler)
  */
 #ifdef CONFIG_CPU_SH2A
 extern unsigned int instruction_size(unsigned int insn);
-#else
+#elif defined(CONFIG_SUPERH32)
 #define instruction_size(insn)	(2)
+#else
+#define instruction_size(insn)	(4)
 #endif
 
 /* XXX
-- 
cgit v1.2.3


From 3b9e78868d000ca10b740c465df9236b04d29718 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 12:56:06 +0900
Subject: sh: Add in cacheflush and DMA headers for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/cacheflush.h | 34 +++++++++++++++++++++++++
 include/asm-sh/cpu-sh5/dma.h        |  6 +++++
 include/asm-sh64/cacheflush.h       | 50 -------------------------------------
 3 files changed, 40 insertions(+), 50 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh5/cacheflush.h
 create mode 100644 include/asm-sh/cpu-sh5/dma.h
 delete mode 100644 include/asm-sh64/cacheflush.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/cacheflush.h b/include/asm-sh/cpu-sh5/cacheflush.h
new file mode 100644
index 000000000000..847374b6526f
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/cacheflush.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_SH64_CACHEFLUSH_H
+#define __ASM_SH64_CACHEFLUSH_H
+
+#ifndef __ASSEMBLY__
+
+#include <asm/page.h>
+
+struct vm_area_struct;
+struct page;
+struct mm_struct;
+
+extern void flush_cache_all(void);
+extern void flush_cache_mm(struct mm_struct *mm);
+extern void flush_cache_sigtramp(unsigned long start, unsigned long end);
+extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+			      unsigned long end);
+extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
+extern void flush_dcache_page(struct page *pg);
+extern void flush_icache_range(unsigned long start, unsigned long end);
+extern void flush_icache_user_range(struct vm_area_struct *vma,
+				    struct page *page, unsigned long addr,
+				    int len);
+
+#define flush_cache_dup_mm(mm)	flush_cache_mm(mm)
+
+#define flush_dcache_mmap_lock(mapping)		do { } while (0)
+#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
+
+#define flush_icache_page(vma, page)	do { } while (0)
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_SH64_CACHEFLUSH_H */
+
diff --git a/include/asm-sh/cpu-sh5/dma.h b/include/asm-sh/cpu-sh5/dma.h
new file mode 100644
index 000000000000..7bf6bb3d35ed
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/dma.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_SH_CPU_SH5_DMA_H
+#define __ASM_SH_CPU_SH5_DMA_H
+
+/* Nothing yet */
+
+#endif /* __ASM_SH_CPU_SH5_DMA_H */
diff --git a/include/asm-sh64/cacheflush.h b/include/asm-sh64/cacheflush.h
deleted file mode 100644
index 1e53a47bdc97..000000000000
--- a/include/asm-sh64/cacheflush.h
+++ /dev/null
@@ -1,50 +0,0 @@
-#ifndef __ASM_SH64_CACHEFLUSH_H
-#define __ASM_SH64_CACHEFLUSH_H
-
-#ifndef __ASSEMBLY__
-
-#include <asm/page.h>
-
-struct vm_area_struct;
-struct page;
-struct mm_struct;
-
-extern void flush_cache_all(void);
-extern void flush_cache_mm(struct mm_struct *mm);
-extern void flush_cache_sigtramp(unsigned long start, unsigned long end);
-extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
-			      unsigned long end);
-extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
-extern void flush_dcache_page(struct page *pg);
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void flush_icache_user_range(struct vm_area_struct *vma,
-				    struct page *page, unsigned long addr,
-				    int len);
-
-#define flush_cache_dup_mm(mm)	flush_cache_mm(mm)
-
-#define flush_dcache_mmap_lock(mapping)		do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
-
-#define flush_cache_vmap(start, end)		flush_cache_all()
-#define flush_cache_vunmap(start, end)		flush_cache_all()
-
-#define flush_icache_page(vma, page)	do { } while (0)
-
-#define copy_to_user_page(vma, page, vaddr, dst, src, len) \
-	do {							\
-		flush_cache_page(vma, vaddr, page_to_pfn(page));\
-		memcpy(dst, src, len);				\
-		flush_icache_user_range(vma, page, vaddr, len);	\
-	} while (0)
-
-#define copy_from_user_page(vma, page, vaddr, dst, src, len) \
-	do {							\
-		flush_cache_page(vma, vaddr, page_to_pfn(page));\
-		memcpy(dst, src, len);				\
-	} while (0)
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* __ASM_SH64_CACHEFLUSH_H */
-
-- 
cgit v1.2.3


From da06b8d0545a1bf95b9060bf301d6de3400fafd6 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 12:58:12 +0900
Subject: sh: Add SH-5 support to io.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/io.h | 38 ++++++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 6ed34d8eac5f..556aabe844c5 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -243,12 +243,20 @@ static inline void ctrl_outl(unsigned int b, unsigned long addr)
 
 static inline void ctrl_delay(void)
 {
+#ifdef P2SEG
 	ctrl_inw(P2SEG);
+#endif
 }
 
 #define IO_SPACE_LIMIT 0xffffffff
 
-#ifdef CONFIG_MMU
+#if !defined(CONFIG_MMU)
+#define virt_to_phys(address)	((unsigned long)(address))
+#define phys_to_virt(address)	((void *)(address))
+#elif defined(CONFIG_SUPERH64)
+#define virt_to_phys(address)	(__pa(address))
+#define phys_to_virt(address)	(__va(address))
+#else
 /*
  * Change virtual addresses to physical addresses and vv.
  * These are trivial on the 1:1 Linux/SuperH mapping
@@ -262,28 +270,24 @@ static inline void *phys_to_virt(unsigned long address)
 {
 	return (void *)P1SEGADDR(address);
 }
-#else
-#define phys_to_virt(address)	((void *)(address))
-#define virt_to_phys(address)	((unsigned long)(address))
 #endif
 
 /*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the x86 architecture, we just read/write the
- * memory location directly.
+ * On 32-bit SH, we traditionally have the whole physical address space
+ * mapped at all times (as MIPS does), so "ioremap()" and "iounmap()" do
+ * not need to do anything but place the address in the proper segment.
+ * This is true for P1 and P2 addresses, as well as some P3 ones.
+ * However, most of the P3 addresses and newer cores using extended
+ * addressing need to map through page tables, so the ioremap()
+ * implementation becomes a bit more complicated.
  *
- * On SH, we traditionally have the whole physical address space mapped
- * at all times (as MIPS does), so "ioremap()" and "iounmap()" do not
- * need to do anything but place the address in the proper segment. This
- * is true for P1 and P2 addresses, as well as some P3 ones. However,
- * most of the P3 addresses and newer cores using extended addressing
- * need to map through page tables, so the ioremap() implementation
- * becomes a bit more complicated. See arch/sh/mm/ioremap.c for
- * additional notes on this.
+ * See arch/sh/mm/ioremap.c for additional notes on this.
  *
  * We cheat a bit and always return uncachable areas until we've fixed
  * the drivers to handle caching properly.
+ *
+ * On the SH-5 the concept of segmentation in the 1:1 PXSEG sense simply
+ * doesn't exist, so everything must go through page tables.
  */
 #ifdef CONFIG_MMU
 void __iomem *__ioremap(unsigned long offset, unsigned long size,
@@ -297,6 +301,7 @@ void __iounmap(void __iomem *addr);
 static inline void __iomem *
 __ioremap_mode(unsigned long offset, unsigned long size, unsigned long flags)
 {
+#ifdef CONFIG_SUPERH32
 	unsigned long last_addr = offset + size - 1;
 
 	/*
@@ -311,6 +316,7 @@ __ioremap_mode(unsigned long offset, unsigned long size, unsigned long flags)
 
 		return (void __iomem *)P2SEGADDR(offset);
 	}
+#endif
 
 	return __ioremap(offset, size, flags);
 }
-- 
cgit v1.2.3


From 7960a1d02b00fd5dfa5c2d9b957e4e5f6ec23997 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 13:00:30 +0900
Subject: sh: Split out asm/string.h for sh32 and sh64.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/string.h    | 136 ++-------------------------------------------
 include/asm-sh/string_32.h | 131 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/string_64.h |  21 +++++++
 include/asm-sh64/string.h  |  21 -------
 4 files changed, 157 insertions(+), 152 deletions(-)
 create mode 100644 include/asm-sh/string_32.h
 create mode 100644 include/asm-sh/string_64.h
 delete mode 100644 include/asm-sh64/string.h

(limited to 'include')

diff --git a/include/asm-sh/string.h b/include/asm-sh/string.h
index 55f8db6bc1d7..8c1ea21dc0ae 100644
--- a/include/asm-sh/string.h
+++ b/include/asm-sh/string.h
@@ -1,131 +1,5 @@
-#ifndef __ASM_SH_STRING_H
-#define __ASM_SH_STRING_H
-
-#ifdef __KERNEL__
-
-/*
- * Copyright (C) 1999 Niibe Yutaka
- * But consider these trivial functions to be public domain.
- */
-
-#define __HAVE_ARCH_STRCPY
-static inline char *strcpy(char *__dest, const char *__src)
-{
-	register char *__xdest = __dest;
-	unsigned long __dummy;
-
-	__asm__ __volatile__("1:\n\t"
-			     "mov.b	@%1+, %2\n\t"
-			     "mov.b	%2, @%0\n\t"
-			     "cmp/eq	#0, %2\n\t"
-			     "bf/s	1b\n\t"
-			     " add	#1, %0\n\t"
-			     : "=r" (__dest), "=r" (__src), "=&z" (__dummy)
-			     : "0" (__dest), "1" (__src)
-			     : "memory", "t");
-
-	return __xdest;
-}
-
-#define __HAVE_ARCH_STRNCPY
-static inline char *strncpy(char *__dest, const char *__src, size_t __n)
-{
-	register char *__xdest = __dest;
-	unsigned long __dummy;
-
-	if (__n == 0)
-		return __xdest;
-
-	__asm__ __volatile__(
-		"1:\n"
-		"mov.b	@%1+, %2\n\t"
-		"mov.b	%2, @%0\n\t"
-		"cmp/eq	#0, %2\n\t"
-		"bt/s	2f\n\t"
-		" cmp/eq	%5,%1\n\t"
-		"bf/s	1b\n\t"
-		" add	#1, %0\n"
-		"2:"
-		: "=r" (__dest), "=r" (__src), "=&z" (__dummy)
-		: "0" (__dest), "1" (__src), "r" (__src+__n)
-		: "memory", "t");
-
-	return __xdest;
-}
-
-#define __HAVE_ARCH_STRCMP
-static inline int strcmp(const char *__cs, const char *__ct)
-{
-	register int __res;
-	unsigned long __dummy;
-
-	__asm__ __volatile__(
-		"mov.b	@%1+, %3\n"
-		"1:\n\t"
-		"mov.b	@%0+, %2\n\t"
-		"cmp/eq #0, %3\n\t"
-		"bt	2f\n\t"
-		"cmp/eq %2, %3\n\t"
-		"bt/s	1b\n\t"
-		" mov.b	@%1+, %3\n\t"
-		"add	#-2, %1\n\t"
-		"mov.b	@%1, %3\n\t"
-		"sub	%3, %2\n"
-		"2:"
-		: "=r" (__cs), "=r" (__ct), "=&r" (__res), "=&z" (__dummy)
-		: "0" (__cs), "1" (__ct)
-		: "t");
-
-	return __res;
-}
-
-#define __HAVE_ARCH_STRNCMP
-static inline int strncmp(const char *__cs, const char *__ct, size_t __n)
-{
-	register int __res;
-	unsigned long __dummy;
-
-	if (__n == 0)
-		return 0;
-
-	__asm__ __volatile__(
-		"mov.b	@%1+, %3\n"
-		"1:\n\t"
-		"mov.b	@%0+, %2\n\t"
-		"cmp/eq %6, %0\n\t"
-		"bt/s	2f\n\t"
-		" cmp/eq #0, %3\n\t"
-		"bt/s	3f\n\t"
-		" cmp/eq %3, %2\n\t"
-		"bt/s	1b\n\t"
-		" mov.b	@%1+, %3\n\t"
-		"add	#-2, %1\n\t"
-		"mov.b	@%1, %3\n"
-		"2:\n\t"
-		"sub	%3, %2\n"
-		"3:"
-		:"=r" (__cs), "=r" (__ct), "=&r" (__res), "=&z" (__dummy)
-		: "0" (__cs), "1" (__ct), "r" (__cs+__n)
-		: "t");
-
-	return __res;
-}
-
-#define __HAVE_ARCH_MEMSET
-extern void *memset(void *__s, int __c, size_t __count);
-
-#define __HAVE_ARCH_MEMCPY
-extern void *memcpy(void *__to, __const__ void *__from, size_t __n);
-
-#define __HAVE_ARCH_MEMMOVE
-extern void *memmove(void *__dest, __const__ void *__src, size_t __n);
-
-#define __HAVE_ARCH_MEMCHR
-extern void *memchr(const void *__s, int __c, size_t __n);
-
-#define __HAVE_ARCH_STRLEN
-extern size_t strlen(const char *);
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH_STRING_H */
+#ifdef CONFIG_SUPERH32
+# include "string_32.h"
+#else
+# include "string_64.h"
+#endif
diff --git a/include/asm-sh/string_32.h b/include/asm-sh/string_32.h
new file mode 100644
index 000000000000..55f8db6bc1d7
--- /dev/null
+++ b/include/asm-sh/string_32.h
@@ -0,0 +1,131 @@
+#ifndef __ASM_SH_STRING_H
+#define __ASM_SH_STRING_H
+
+#ifdef __KERNEL__
+
+/*
+ * Copyright (C) 1999 Niibe Yutaka
+ * But consider these trivial functions to be public domain.
+ */
+
+#define __HAVE_ARCH_STRCPY
+static inline char *strcpy(char *__dest, const char *__src)
+{
+	register char *__xdest = __dest;
+	unsigned long __dummy;
+
+	__asm__ __volatile__("1:\n\t"
+			     "mov.b	@%1+, %2\n\t"
+			     "mov.b	%2, @%0\n\t"
+			     "cmp/eq	#0, %2\n\t"
+			     "bf/s	1b\n\t"
+			     " add	#1, %0\n\t"
+			     : "=r" (__dest), "=r" (__src), "=&z" (__dummy)
+			     : "0" (__dest), "1" (__src)
+			     : "memory", "t");
+
+	return __xdest;
+}
+
+#define __HAVE_ARCH_STRNCPY
+static inline char *strncpy(char *__dest, const char *__src, size_t __n)
+{
+	register char *__xdest = __dest;
+	unsigned long __dummy;
+
+	if (__n == 0)
+		return __xdest;
+
+	__asm__ __volatile__(
+		"1:\n"
+		"mov.b	@%1+, %2\n\t"
+		"mov.b	%2, @%0\n\t"
+		"cmp/eq	#0, %2\n\t"
+		"bt/s	2f\n\t"
+		" cmp/eq	%5,%1\n\t"
+		"bf/s	1b\n\t"
+		" add	#1, %0\n"
+		"2:"
+		: "=r" (__dest), "=r" (__src), "=&z" (__dummy)
+		: "0" (__dest), "1" (__src), "r" (__src+__n)
+		: "memory", "t");
+
+	return __xdest;
+}
+
+#define __HAVE_ARCH_STRCMP
+static inline int strcmp(const char *__cs, const char *__ct)
+{
+	register int __res;
+	unsigned long __dummy;
+
+	__asm__ __volatile__(
+		"mov.b	@%1+, %3\n"
+		"1:\n\t"
+		"mov.b	@%0+, %2\n\t"
+		"cmp/eq #0, %3\n\t"
+		"bt	2f\n\t"
+		"cmp/eq %2, %3\n\t"
+		"bt/s	1b\n\t"
+		" mov.b	@%1+, %3\n\t"
+		"add	#-2, %1\n\t"
+		"mov.b	@%1, %3\n\t"
+		"sub	%3, %2\n"
+		"2:"
+		: "=r" (__cs), "=r" (__ct), "=&r" (__res), "=&z" (__dummy)
+		: "0" (__cs), "1" (__ct)
+		: "t");
+
+	return __res;
+}
+
+#define __HAVE_ARCH_STRNCMP
+static inline int strncmp(const char *__cs, const char *__ct, size_t __n)
+{
+	register int __res;
+	unsigned long __dummy;
+
+	if (__n == 0)
+		return 0;
+
+	__asm__ __volatile__(
+		"mov.b	@%1+, %3\n"
+		"1:\n\t"
+		"mov.b	@%0+, %2\n\t"
+		"cmp/eq %6, %0\n\t"
+		"bt/s	2f\n\t"
+		" cmp/eq #0, %3\n\t"
+		"bt/s	3f\n\t"
+		" cmp/eq %3, %2\n\t"
+		"bt/s	1b\n\t"
+		" mov.b	@%1+, %3\n\t"
+		"add	#-2, %1\n\t"
+		"mov.b	@%1, %3\n"
+		"2:\n\t"
+		"sub	%3, %2\n"
+		"3:"
+		:"=r" (__cs), "=r" (__ct), "=&r" (__res), "=&z" (__dummy)
+		: "0" (__cs), "1" (__ct), "r" (__cs+__n)
+		: "t");
+
+	return __res;
+}
+
+#define __HAVE_ARCH_MEMSET
+extern void *memset(void *__s, int __c, size_t __count);
+
+#define __HAVE_ARCH_MEMCPY
+extern void *memcpy(void *__to, __const__ void *__from, size_t __n);
+
+#define __HAVE_ARCH_MEMMOVE
+extern void *memmove(void *__dest, __const__ void *__src, size_t __n);
+
+#define __HAVE_ARCH_MEMCHR
+extern void *memchr(const void *__s, int __c, size_t __n);
+
+#define __HAVE_ARCH_STRLEN
+extern size_t strlen(const char *);
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_SH_STRING_H */
diff --git a/include/asm-sh/string_64.h b/include/asm-sh/string_64.h
new file mode 100644
index 000000000000..8a7357366ce8
--- /dev/null
+++ b/include/asm-sh/string_64.h
@@ -0,0 +1,21 @@
+#ifndef __ASM_SH64_STRING_H
+#define __ASM_SH64_STRING_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/string.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ *
+ * Empty on purpose. ARCH SH64 ASM libs are out of the current project scope.
+ *
+ */
+
+#define __HAVE_ARCH_MEMCPY
+
+extern void *memcpy(void *dest, const void *src, size_t count);
+
+#endif
diff --git a/include/asm-sh64/string.h b/include/asm-sh64/string.h
deleted file mode 100644
index 8a7357366ce8..000000000000
--- a/include/asm-sh64/string.h
+++ /dev/null
@@ -1,21 +0,0 @@
-#ifndef __ASM_SH64_STRING_H
-#define __ASM_SH64_STRING_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/string.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- * Empty on purpose. ARCH SH64 ASM libs are out of the current project scope.
- *
- */
-
-#define __HAVE_ARCH_MEMCPY
-
-extern void *memcpy(void *dest, const void *src, size_t count);
-
-#endif
-- 
cgit v1.2.3


From f64ee87614e80ca270de0b80c5164ab05f4f1d98 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 13:34:36 +0900
Subject: sh: Split out irqflags.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/registers.h | 106 +++++++++++++++++++++++++++++++++++++
 include/asm-sh/irqflags.h          |  97 ++-------------------------------
 include/asm-sh/irqflags_32.h       |  99 ++++++++++++++++++++++++++++++++++
 include/asm-sh/irqflags_64.h       |  85 +++++++++++++++++++++++++++++
 include/asm-sh64/registers.h       | 106 -------------------------------------
 5 files changed, 294 insertions(+), 199 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh5/registers.h
 create mode 100644 include/asm-sh/irqflags_32.h
 create mode 100644 include/asm-sh/irqflags_64.h
 delete mode 100644 include/asm-sh64/registers.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/registers.h b/include/asm-sh/cpu-sh5/registers.h
new file mode 100644
index 000000000000..7eec666acf84
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/registers.h
@@ -0,0 +1,106 @@
+#ifndef __ASM_SH64_REGISTERS_H
+#define __ASM_SH64_REGISTERS_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/registers.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2004  Richard Curnow
+ */
+
+#ifdef __ASSEMBLY__
+/* =====================================================================
+**
+** Section 1: acts on assembly sources pre-processed by GPP ( <source.S>).
+**	      Assigns symbolic names to control & target registers.
+*/
+
+/*
+ * Define some useful aliases for control registers.
+ */
+#define SR	cr0
+#define SSR	cr1
+#define PSSR	cr2
+			/* cr3 UNDEFINED */
+#define INTEVT	cr4
+#define EXPEVT	cr5
+#define PEXPEVT	cr6
+#define TRA	cr7
+#define SPC	cr8
+#define PSPC	cr9
+#define RESVEC	cr10
+#define VBR	cr11
+			/* cr12 UNDEFINED */
+#define TEA	cr13
+			/* cr14-cr15 UNDEFINED */
+#define DCR	cr16
+#define KCR0	cr17
+#define KCR1	cr18
+			/* cr19-cr31 UNDEFINED */
+			/* cr32-cr61 RESERVED */
+#define CTC	cr62
+#define USR	cr63
+
+/*
+ * ABI dependent registers (general purpose set)
+ */
+#define RET	r2
+#define ARG1	r2
+#define ARG2	r3
+#define ARG3	r4
+#define ARG4	r5
+#define ARG5	r6
+#define ARG6	r7
+#define SP	r15
+#define LINK	r18
+#define ZERO	r63
+
+/*
+ * Status register defines: used only by assembly sources (and
+ * 			    syntax independednt)
+ */
+#define SR_RESET_VAL	0x0000000050008000
+#define SR_HARMLESS	0x00000000500080f0	/* Write ignores for most */
+#define SR_ENABLE_FPU	0xffffffffffff7fff	/* AND with this */
+
+#if defined (CONFIG_SH64_SR_WATCH)
+#define SR_ENABLE_MMU	0x0000000084000000	/* OR with this */
+#else
+#define SR_ENABLE_MMU	0x0000000080000000	/* OR with this */
+#endif
+
+#define SR_UNBLOCK_EXC	0xffffffffefffffff	/* AND with this */
+#define SR_BLOCK_EXC	0x0000000010000000	/* OR with this */
+
+#else	/* Not __ASSEMBLY__ syntax */
+
+/*
+** Stringify reg. name
+*/
+#define __str(x)  #x
+
+/* Stringify control register names for use in inline assembly */
+#define __SR __str(SR)
+#define __SSR __str(SSR)
+#define __PSSR __str(PSSR)
+#define __INTEVT __str(INTEVT)
+#define __EXPEVT __str(EXPEVT)
+#define __PEXPEVT __str(PEXPEVT)
+#define __TRA __str(TRA)
+#define __SPC __str(SPC)
+#define __PSPC __str(PSPC)
+#define __RESVEC __str(RESVEC)
+#define __VBR __str(VBR)
+#define __TEA __str(TEA)
+#define __DCR __str(DCR)
+#define __KCR0 __str(KCR0)
+#define __KCR1 __str(KCR1)
+#define __CTC __str(CTC)
+#define __USR __str(USR)
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_SH64_REGISTERS_H */
diff --git a/include/asm-sh/irqflags.h b/include/asm-sh/irqflags.h
index 9dedc1b693e3..46e71da5be6b 100644
--- a/include/asm-sh/irqflags.h
+++ b/include/asm-sh/irqflags.h
@@ -1,81 +1,11 @@
 #ifndef __ASM_SH_IRQFLAGS_H
 #define __ASM_SH_IRQFLAGS_H
 
-static inline void raw_local_irq_enable(void)
-{
-	unsigned long __dummy0, __dummy1;
-
-	__asm__ __volatile__ (
-		"stc	sr, %0\n\t"
-		"and	%1, %0\n\t"
-#ifdef CONFIG_CPU_HAS_SR_RB
-		"stc	r6_bank, %1\n\t"
-		"or	%1, %0\n\t"
+#ifdef CONFIG_SUPERH32
+#include "irqflags_32.h"
+#else
+#include "irqflags_64.h"
 #endif
-		"ldc	%0, sr\n\t"
-		: "=&r" (__dummy0), "=r" (__dummy1)
-		: "1" (~0x000000f0)
-		: "memory"
-	);
-}
-
-static inline void raw_local_irq_disable(void)
-{
-	unsigned long flags;
-
-	__asm__ __volatile__ (
-		"stc	sr, %0\n\t"
-		"or	#0xf0, %0\n\t"
-		"ldc	%0, sr\n\t"
-		: "=&z" (flags)
-		: /* no inputs */
-		: "memory"
-	);
-}
-
-static inline void set_bl_bit(void)
-{
-	unsigned long __dummy0, __dummy1;
-
-	__asm__ __volatile__ (
-		"stc	sr, %0\n\t"
-		"or	%2, %0\n\t"
-		"and	%3, %0\n\t"
-		"ldc	%0, sr\n\t"
-		: "=&r" (__dummy0), "=r" (__dummy1)
-		: "r" (0x10000000), "r" (0xffffff0f)
-		: "memory"
-	);
-}
-
-static inline void clear_bl_bit(void)
-{
-	unsigned long __dummy0, __dummy1;
-
-	__asm__ __volatile__ (
-		"stc	sr, %0\n\t"
-		"and	%2, %0\n\t"
-		"ldc	%0, sr\n\t"
-		: "=&r" (__dummy0), "=r" (__dummy1)
-		: "1" (~0x10000000)
-		: "memory"
-	);
-}
-
-static inline unsigned long __raw_local_save_flags(void)
-{
-	unsigned long flags;
-
-	__asm__ __volatile__ (
-		"stc	sr, %0\n\t"
-		"and	#0xf0, %0\n\t"
-		: "=&z" (flags)
-		: /* no inputs */
-		: "memory"
-	);
-
-	return flags;
-}
 
 #define raw_local_save_flags(flags) \
 		do { (flags) = __raw_local_save_flags(); } while (0)
@@ -92,25 +22,6 @@ static inline int raw_irqs_disabled(void)
 	return raw_irqs_disabled_flags(flags);
 }
 
-static inline unsigned long __raw_local_irq_save(void)
-{
-	unsigned long flags, __dummy;
-
-	__asm__ __volatile__ (
-		"stc	sr, %1\n\t"
-		"mov	%1, %0\n\t"
-		"or	#0xf0, %0\n\t"
-		"ldc	%0, sr\n\t"
-		"mov	%1, %0\n\t"
-		"and	#0xf0, %0\n\t"
-		: "=&z" (flags), "=&r" (__dummy)
-		: /* no inputs */
-		: "memory"
-	);
-
-	return flags;
-}
-
 #define raw_local_irq_save(flags) \
 		do { (flags) = __raw_local_irq_save(); } while (0)
 
diff --git a/include/asm-sh/irqflags_32.h b/include/asm-sh/irqflags_32.h
new file mode 100644
index 000000000000..60218f541340
--- /dev/null
+++ b/include/asm-sh/irqflags_32.h
@@ -0,0 +1,99 @@
+#ifndef __ASM_SH_IRQFLAGS_32_H
+#define __ASM_SH_IRQFLAGS_32_H
+
+static inline void raw_local_irq_enable(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"and	%1, %0\n\t"
+#ifdef CONFIG_CPU_HAS_SR_RB
+		"stc	r6_bank, %1\n\t"
+		"or	%1, %0\n\t"
+#endif
+		"ldc	%0, sr\n\t"
+		: "=&r" (__dummy0), "=r" (__dummy1)
+		: "1" (~0x000000f0)
+		: "memory"
+	);
+}
+
+static inline void raw_local_irq_disable(void)
+{
+	unsigned long flags;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"or	#0xf0, %0\n\t"
+		"ldc	%0, sr\n\t"
+		: "=&z" (flags)
+		: /* no inputs */
+		: "memory"
+	);
+}
+
+static inline void set_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"or	%2, %0\n\t"
+		"and	%3, %0\n\t"
+		"ldc	%0, sr\n\t"
+		: "=&r" (__dummy0), "=r" (__dummy1)
+		: "r" (0x10000000), "r" (0xffffff0f)
+		: "memory"
+	);
+}
+
+static inline void clear_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"and	%2, %0\n\t"
+		"ldc	%0, sr\n\t"
+		: "=&r" (__dummy0), "=r" (__dummy1)
+		: "1" (~0x10000000)
+		: "memory"
+	);
+}
+
+static inline unsigned long __raw_local_save_flags(void)
+{
+	unsigned long flags;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"and	#0xf0, %0\n\t"
+		: "=&z" (flags)
+		: /* no inputs */
+		: "memory"
+	);
+
+	return flags;
+}
+
+static inline unsigned long __raw_local_irq_save(void)
+{
+	unsigned long flags, __dummy;
+
+	__asm__ __volatile__ (
+		"stc	sr, %1\n\t"
+		"mov	%1, %0\n\t"
+		"or	#0xf0, %0\n\t"
+		"ldc	%0, sr\n\t"
+		"mov	%1, %0\n\t"
+		"and	#0xf0, %0\n\t"
+		: "=&z" (flags), "=&r" (__dummy)
+		: /* no inputs */
+		: "memory"
+	);
+
+	return flags;
+}
+
+#endif /* __ASM_SH_IRQFLAGS_32_H */
diff --git a/include/asm-sh/irqflags_64.h b/include/asm-sh/irqflags_64.h
new file mode 100644
index 000000000000..4f6b8a56e7bd
--- /dev/null
+++ b/include/asm-sh/irqflags_64.h
@@ -0,0 +1,85 @@
+#ifndef __ASM_SH_IRQFLAGS_64_H
+#define __ASM_SH_IRQFLAGS_64_H
+
+#include <asm/cpu/registers.h>
+
+#define SR_MASK_LL	0x00000000000000f0LL
+#define SR_BL_LL	0x0000000010000000LL
+
+static inline void raw_local_irq_enable(void)
+{
+	unsigned long long __dummy0, __dummy1 = ~SR_MASK_LL;
+
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "and	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy0)
+			     : "r" (__dummy1));
+}
+
+static inline void raw_local_irq_disable(void)
+{
+	unsigned long long __dummy0, __dummy1 = SR_MASK_LL;
+
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "or	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy0)
+			     : "r" (__dummy1));
+}
+
+static inline void set_bl_bit(void)
+{
+	unsigned long long __dummy0, __dummy1 = SR_BL_LL;
+
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "or	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy0)
+			     : "r" (__dummy1));
+
+}
+
+static inline void clear_bl_bit(void)
+{
+	unsigned long long __dummy0, __dummy1 = ~SR_BL_LL;
+
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "and	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy0)
+			     : "r" (__dummy1));
+}
+
+static inline unsigned long __raw_local_save_flags(void)
+{
+	unsigned long long __dummy = SR_MASK_LL;
+	unsigned long flags;
+
+	__asm__ __volatile__ (
+		"getcon	" __SR ", %0\n\t"
+		"and	%0, %1, %0"
+		: "=&r" (flags)
+		: "r" (__dummy));
+
+	return flags;
+}
+
+static inline unsigned long __raw_local_irq_save(void)
+{
+	unsigned long long __dummy0, __dummy1 = SR_MASK_LL;
+	unsigned long flags;
+
+	__asm__ __volatile__ (
+		"getcon	" __SR ", %1\n\t"
+		"or	%1, r63, %0\n\t"
+		"or	%1, %2, %1\n\t"
+		"putcon	%1, " __SR "\n\t"
+		"and	%0, %2, %0"
+		: "=&r" (flags), "=&r" (__dummy0)
+		: "r" (__dummy1));
+
+	return flags;
+}
+
+#endif /* __ASM_SH_IRQFLAGS_64_H */
diff --git a/include/asm-sh64/registers.h b/include/asm-sh64/registers.h
deleted file mode 100644
index 7eec666acf84..000000000000
--- a/include/asm-sh64/registers.h
+++ /dev/null
@@ -1,106 +0,0 @@
-#ifndef __ASM_SH64_REGISTERS_H
-#define __ASM_SH64_REGISTERS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/registers.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2004  Richard Curnow
- */
-
-#ifdef __ASSEMBLY__
-/* =====================================================================
-**
-** Section 1: acts on assembly sources pre-processed by GPP ( <source.S>).
-**	      Assigns symbolic names to control & target registers.
-*/
-
-/*
- * Define some useful aliases for control registers.
- */
-#define SR	cr0
-#define SSR	cr1
-#define PSSR	cr2
-			/* cr3 UNDEFINED */
-#define INTEVT	cr4
-#define EXPEVT	cr5
-#define PEXPEVT	cr6
-#define TRA	cr7
-#define SPC	cr8
-#define PSPC	cr9
-#define RESVEC	cr10
-#define VBR	cr11
-			/* cr12 UNDEFINED */
-#define TEA	cr13
-			/* cr14-cr15 UNDEFINED */
-#define DCR	cr16
-#define KCR0	cr17
-#define KCR1	cr18
-			/* cr19-cr31 UNDEFINED */
-			/* cr32-cr61 RESERVED */
-#define CTC	cr62
-#define USR	cr63
-
-/*
- * ABI dependent registers (general purpose set)
- */
-#define RET	r2
-#define ARG1	r2
-#define ARG2	r3
-#define ARG3	r4
-#define ARG4	r5
-#define ARG5	r6
-#define ARG6	r7
-#define SP	r15
-#define LINK	r18
-#define ZERO	r63
-
-/*
- * Status register defines: used only by assembly sources (and
- * 			    syntax independednt)
- */
-#define SR_RESET_VAL	0x0000000050008000
-#define SR_HARMLESS	0x00000000500080f0	/* Write ignores for most */
-#define SR_ENABLE_FPU	0xffffffffffff7fff	/* AND with this */
-
-#if defined (CONFIG_SH64_SR_WATCH)
-#define SR_ENABLE_MMU	0x0000000084000000	/* OR with this */
-#else
-#define SR_ENABLE_MMU	0x0000000080000000	/* OR with this */
-#endif
-
-#define SR_UNBLOCK_EXC	0xffffffffefffffff	/* AND with this */
-#define SR_BLOCK_EXC	0x0000000010000000	/* OR with this */
-
-#else	/* Not __ASSEMBLY__ syntax */
-
-/*
-** Stringify reg. name
-*/
-#define __str(x)  #x
-
-/* Stringify control register names for use in inline assembly */
-#define __SR __str(SR)
-#define __SSR __str(SSR)
-#define __PSSR __str(PSSR)
-#define __INTEVT __str(INTEVT)
-#define __EXPEVT __str(EXPEVT)
-#define __PEXPEVT __str(PEXPEVT)
-#define __TRA __str(TRA)
-#define __SPC __str(SPC)
-#define __PSPC __str(PSPC)
-#define __RESVEC __str(RESVEC)
-#define __VBR __str(VBR)
-#define __TEA __str(TEA)
-#define __DCR __str(DCR)
-#define __KCR0 __str(KCR0)
-#define __KCR1 __str(KCR1)
-#define __CTC __str(CTC)
-#define __USR __str(USR)
-
-#endif /* __ASSEMBLY__ */
-#endif /* __ASM_SH64_REGISTERS_H */
-- 
cgit v1.2.3


From 114f132975ae8db3caa0c3420dc1083bae8d3757 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 13:40:50 +0900
Subject: sh: SH-5 version of current_thread_info().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/thread_info.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 1f7e1deb8d92..d49ee9d868e5 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -74,8 +74,10 @@ register unsigned long current_stack_pointer asm("r15") __attribute_used__;
 static inline struct thread_info *current_thread_info(void)
 {
 	struct thread_info *ti;
-#ifdef CONFIG_CPU_HAS_SR_RB
-	__asm__("stc	r7_bank, %0" : "=r" (ti));
+#if defined(CONFIG_SUPERH64)
+	__asm__ __volatile__ ("getcon	cr17, %0" : "=r" (ti));
+#elif defined(CONFIG_CPU_HAS_SR_RB)
+	__asm__ __volatile__ ("stc	r7_bank, %0" : "=r" (ti));
 #else
 	unsigned long __dummy;
 
-- 
cgit v1.2.3


From 7a65eaf4885d1d0afeec45239eaf9208a3235b51 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 13:58:44 +0900
Subject: sh: SH-5 byteorder routines.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/byteorder.h | 45 ++++++++++++++++++++++++++++++---------------
 1 file changed, 30 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/byteorder.h b/include/asm-sh/byteorder.h
index bff2b1382e01..0eb9904b6545 100644
--- a/include/asm-sh/byteorder.h
+++ b/include/asm-sh/byteorder.h
@@ -3,40 +3,55 @@
 
 /*
  * Copyright (C) 1999  Niibe Yutaka
+ * Copyright (C) 2000, 2001  Paolo Alberelli
  */
-
-#include <asm/types.h>
 #include <linux/compiler.h>
+#include <linux/types.h>
 
-static __inline__ __attribute_const__ __u32 ___arch__swab32(__u32 x)
+static inline __attribute_const__ __u32 ___arch__swab32(__u32 x)
 {
-	__asm__("swap.b	%0, %0\n\t"
-		"swap.w %0, %0\n\t"
-		"swap.b %0, %0"
+	__asm__(
+#ifdef CONFIG_SUPERH32
+		"swap.b		%0, %0\n\t"
+		"swap.w		%0, %0\n\t"
+		"swap.b		%0, %0"
+#else
+		"byterev	%0, %0\n\t"
+		"shari		%0, 32, %0"
+#endif
 		: "=r" (x)
 		: "0" (x));
+
 	return x;
 }
 
-static __inline__ __attribute_const__ __u16 ___arch__swab16(__u16 x)
+static inline __attribute_const__ __u16 ___arch__swab16(__u16 x)
 {
-	__asm__("swap.b %0, %0"
+	__asm__(
+#ifdef CONFIG_SUPERH32
+		"swap.b		%0, %0"
+#else
+		"byterev	%0, %0\n\t"
+		"shari		%0, 32, %0"
+
+#endif
 		: "=r" (x)
 		:  "0" (x));
+
 	return x;
 }
 
-static inline __u64 ___arch__swab64(__u64 val) 
-{ 
-	union { 
+static inline __u64 ___arch__swab64(__u64 val)
+{
+	union {
 		struct { __u32 a,b; } s;
 		__u64 u;
 	} v, w;
 	v.u = val;
-	w.s.b = ___arch__swab32(v.s.a); 
-	w.s.a = ___arch__swab32(v.s.b); 
-	return w.u;	
-} 
+	w.s.b = ___arch__swab32(v.s.a);
+	w.s.a = ___arch__swab32(v.s.b);
+	return w.u;
+}
 
 #define __arch__swab64(x) ___arch__swab64(x)
 #define __arch__swab32(x) ___arch__swab32(x)
-- 
cgit v1.2.3


From cdcc970829e81da3445346cb71b2ea264c9952b9 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 16:37:18 +0900
Subject: sh: Move in the SH-5 mmu_context headers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/mmu_context.h |  27 +++++
 include/asm-sh/mmu_context.h         |  87 ++++-----------
 include/asm-sh/mmu_context_32.h      |  47 ++++++++
 include/asm-sh/mmu_context_64.h      |  75 +++++++++++++
 include/asm-sh64/mmu_context.h       | 208 -----------------------------------
 5 files changed, 173 insertions(+), 271 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh5/mmu_context.h
 create mode 100644 include/asm-sh/mmu_context_32.h
 create mode 100644 include/asm-sh/mmu_context_64.h
 delete mode 100644 include/asm-sh64/mmu_context.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/mmu_context.h b/include/asm-sh/cpu-sh5/mmu_context.h
new file mode 100644
index 000000000000..df857fc09960
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/mmu_context.h
@@ -0,0 +1,27 @@
+#ifndef __ASM_SH_CPU_SH5_MMU_CONTEXT_H
+#define __ASM_SH_CPU_SH5_MMU_CONTEXT_H
+
+/* Common defines */
+#define TLB_STEP	0x00000010
+#define TLB_PTEH	0x00000000
+#define TLB_PTEL	0x00000008
+
+/* PTEH defines */
+#define PTEH_ASID_SHIFT	2
+#define PTEH_VALID	0x0000000000000001
+#define PTEH_SHARED	0x0000000000000002
+#define PTEH_MATCH_ASID	0x00000000000003ff
+
+#ifndef __ASSEMBLY__
+/* This has to be a common function because the next location to fill
+ * information is shared. */
+extern void __do_tlb_refill(unsigned long address, unsigned long long is_text_not_data, pte_t *pte);
+
+/* Profiling counter. */
+#ifdef CONFIG_SH64_PROC_TLB
+extern unsigned long long calls_to_do_fast_page_fault;
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_SH_CPU_SH5_MMU_CONTEXT_H */
diff --git a/include/asm-sh/mmu_context.h b/include/asm-sh/mmu_context.h
index 199662bb35c6..fe58d00b250c 100644
--- a/include/asm-sh/mmu_context.h
+++ b/include/asm-sh/mmu_context.h
@@ -1,13 +1,13 @@
 /*
  * Copyright (C) 1999 Niibe Yutaka
- * Copyright (C) 2003 - 2006 Paul Mundt
+ * Copyright (C) 2003 - 2007 Paul Mundt
  *
  * ASID handling idea taken from MIPS implementation.
  */
 #ifndef __ASM_SH_MMU_CONTEXT_H
 #define __ASM_SH_MMU_CONTEXT_H
-#ifdef __KERNEL__
 
+#ifdef __KERNEL__
 #include <asm/cpu/mmu_context.h>
 #include <asm/tlbflush.h>
 #include <asm/uaccess.h>
@@ -19,7 +19,6 @@
  *    (a) TLB cache version (or round, cycle whatever expression you like)
  *    (b) ASID (Address Space IDentifier)
  */
-
 #define MMU_CONTEXT_ASID_MASK		0x000000ff
 #define MMU_CONTEXT_VERSION_MASK	0xffffff00
 #define MMU_CONTEXT_FIRST_VERSION	0x00000100
@@ -28,10 +27,11 @@
 /* ASID is 8-bit value, so it can't be 0x100 */
 #define MMU_NO_ASID			0x100
 
-#define cpu_context(cpu, mm)	((mm)->context.id[cpu])
-#define cpu_asid(cpu, mm)	(cpu_context((cpu), (mm)) & \
-				 MMU_CONTEXT_ASID_MASK)
 #define asid_cache(cpu)		(cpu_data[cpu].asid_cache)
+#define cpu_context(cpu, mm)	((mm)->context.id[cpu])
+
+#define cpu_asid(cpu, mm)	\
+	(cpu_context((cpu), (mm)) & MMU_CONTEXT_ASID_MASK)
 
 /*
  * Virtual Page Number mask
@@ -39,6 +39,12 @@
 #define MMU_VPN_MASK	0xfffff000
 
 #ifdef CONFIG_MMU
+#if defined(CONFIG_SUPERH32)
+#include "mmu_context_32.h"
+#else
+#include "mmu_context_64.h"
+#endif
+
 /*
  * Get MMU context if needed.
  */
@@ -59,6 +65,14 @@ static inline void get_mmu_context(struct mm_struct *mm, unsigned int cpu)
 		 */
 		flush_tlb_all();
 
+#ifdef CONFIG_SUPERH64
+		/*
+		 * The SH-5 cache uses the ASIDs, requiring both the I and D
+		 * cache to be flushed when the ASID is exhausted. Weak.
+		 */
+		flush_cache_all();
+#endif
+
 		/*
 		 * Fix version; Note that we avoid version #0
 		 * to distingush NO_CONTEXT.
@@ -85,39 +99,6 @@ static inline int init_new_context(struct task_struct *tsk,
 	return 0;
 }
 
-/*
- * Destroy context related info for an mm_struct that is about
- * to be put to rest.
- */
-static inline void destroy_context(struct mm_struct *mm)
-{
-	/* Do nothing */
-}
-
-static inline void set_asid(unsigned long asid)
-{
-	unsigned long __dummy;
-
-	__asm__ __volatile__ ("mov.l	%2, %0\n\t"
-			      "and	%3, %0\n\t"
-			      "or	%1, %0\n\t"
-			      "mov.l	%0, %2"
-			      : "=&r" (__dummy)
-			      : "r" (asid), "m" (__m(MMU_PTEH)),
-			        "r" (0xffffff00));
-}
-
-static inline unsigned long get_asid(void)
-{
-	unsigned long asid;
-
-	__asm__ __volatile__ ("mov.l	%1, %0"
-			      : "=r" (asid)
-			      : "m" (__m(MMU_PTEH)));
-	asid &= MMU_CONTEXT_ASID_MASK;
-	return asid;
-}
-
 /*
  * After we have set current->mm to a new value, this activates
  * the context for the new mm so we see the new mappings.
@@ -128,17 +109,6 @@ static inline void activate_context(struct mm_struct *mm, unsigned int cpu)
 	set_asid(cpu_asid(cpu, mm));
 }
 
-/* MMU_TTB is used for optimizing the fault handling. */
-static inline void set_TTB(pgd_t *pgd)
-{
-	ctrl_outl((unsigned long)pgd, MMU_TTB);
-}
-
-static inline pgd_t *get_TTB(void)
-{
-	return (pgd_t *)ctrl_inl(MMU_TTB);
-}
-
 static inline void switch_mm(struct mm_struct *prev,
 			     struct mm_struct *next,
 			     struct task_struct *tsk)
@@ -153,17 +123,7 @@ static inline void switch_mm(struct mm_struct *prev,
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask))
 			activate_context(next, cpu);
 }
-
-#define deactivate_mm(tsk,mm)	do { } while (0)
-
-#define activate_mm(prev, next) \
-	switch_mm((prev),(next),NULL)
-
-static inline void
-enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
-{
-}
-#else /* !CONFIG_MMU */
+#else
 #define get_mmu_context(mm)		do { } while (0)
 #define init_new_context(tsk,mm)	(0)
 #define destroy_context(mm)		do { } while (0)
@@ -173,10 +133,11 @@ enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 #define get_TTB()			(0)
 #define activate_context(mm,cpu)	do { } while (0)
 #define switch_mm(prev,next,tsk)	do { } while (0)
+#endif /* CONFIG_MMU */
+
+#define activate_mm(prev, next)		switch_mm((prev),(next),NULL)
 #define deactivate_mm(tsk,mm)		do { } while (0)
-#define activate_mm(prev,next)		do { } while (0)
 #define enter_lazy_tlb(mm,tsk)		do { } while (0)
-#endif /* CONFIG_MMU */
 
 #if defined(CONFIG_CPU_SH3) || defined(CONFIG_CPU_SH4)
 /*
diff --git a/include/asm-sh/mmu_context_32.h b/include/asm-sh/mmu_context_32.h
new file mode 100644
index 000000000000..f4f9aebd68b7
--- /dev/null
+++ b/include/asm-sh/mmu_context_32.h
@@ -0,0 +1,47 @@
+#ifndef __ASM_SH_MMU_CONTEXT_32_H
+#define __ASM_SH_MMU_CONTEXT_32_H
+
+/*
+ * Destroy context related info for an mm_struct that is about
+ * to be put to rest.
+ */
+static inline void destroy_context(struct mm_struct *mm)
+{
+	/* Do nothing */
+}
+
+static inline void set_asid(unsigned long asid)
+{
+	unsigned long __dummy;
+
+	__asm__ __volatile__ ("mov.l	%2, %0\n\t"
+			      "and	%3, %0\n\t"
+			      "or	%1, %0\n\t"
+			      "mov.l	%0, %2"
+			      : "=&r" (__dummy)
+			      : "r" (asid), "m" (__m(MMU_PTEH)),
+			        "r" (0xffffff00));
+}
+
+static inline unsigned long get_asid(void)
+{
+	unsigned long asid;
+
+	__asm__ __volatile__ ("mov.l	%1, %0"
+			      : "=r" (asid)
+			      : "m" (__m(MMU_PTEH)));
+	asid &= MMU_CONTEXT_ASID_MASK;
+	return asid;
+}
+
+/* MMU_TTB is used for optimizing the fault handling. */
+static inline void set_TTB(pgd_t *pgd)
+{
+	ctrl_outl((unsigned long)pgd, MMU_TTB);
+}
+
+static inline pgd_t *get_TTB(void)
+{
+	return (pgd_t *)ctrl_inl(MMU_TTB);
+}
+#endif /* __ASM_SH_MMU_CONTEXT_32_H */
diff --git a/include/asm-sh/mmu_context_64.h b/include/asm-sh/mmu_context_64.h
new file mode 100644
index 000000000000..020be744b088
--- /dev/null
+++ b/include/asm-sh/mmu_context_64.h
@@ -0,0 +1,75 @@
+#ifndef __ASM_SH_MMU_CONTEXT_64_H
+#define __ASM_SH_MMU_CONTEXT_64_H
+
+/*
+ * sh64-specific mmu_context interface.
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003 - 2007  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <asm/cpu/registers.h>
+#include <asm/cacheflush.h>
+
+#define SR_ASID_MASK		0xffffffffff00ffffULL
+#define SR_ASID_SHIFT		16
+
+/*
+ * Destroy context related info for an mm_struct that is about
+ * to be put to rest.
+ */
+static inline void destroy_context(struct mm_struct *mm)
+{
+	/* Well, at least free TLB entries */
+	flush_tlb_mm(mm);
+}
+
+static inline unsigned long get_asid(void)
+{
+	unsigned long long sr;
+
+	asm volatile ("getcon   " __SR ", %0\n\t"
+		      : "=r" (sr));
+
+	sr = (sr >> SR_ASID_SHIFT) & MMU_CONTEXT_ASID_MASK;
+	return (unsigned long) sr;
+}
+
+/* Set ASID into SR */
+static inline void set_asid(unsigned long asid)
+{
+	unsigned long long sr, pc;
+
+	asm volatile ("getcon	" __SR ", %0" : "=r" (sr));
+
+	sr = (sr & SR_ASID_MASK) | (asid << SR_ASID_SHIFT);
+
+	/*
+	 * It is possible that this function may be inlined and so to avoid
+	 * the assembler reporting duplicate symbols we make use of the
+	 * gas trick of generating symbols using numerics and forward
+	 * reference.
+	 */
+	asm volatile ("movi	1, %1\n\t"
+		      "shlli	%1, 28, %1\n\t"
+		      "or	%0, %1, %1\n\t"
+		      "putcon	%1, " __SR "\n\t"
+		      "putcon	%0, " __SSR "\n\t"
+		      "movi	1f, %1\n\t"
+		      "ori	%1, 1 , %1\n\t"
+		      "putcon	%1, " __SPC "\n\t"
+		      "rte\n"
+		      "1:\n\t"
+		      : "=r" (sr), "=r" (pc) : "0" (sr));
+}
+
+/* No spare register to twiddle, so use a software cache */
+extern pgd_t *mmu_pdtp_cache;
+
+#define set_TTB(pgd)	(mmu_pdtp_cache = (pgd))
+#define get_TTB()	(mmu_pdtp_cache)
+
+#endif /* __ASM_SH_MMU_CONTEXT_64_H */
diff --git a/include/asm-sh64/mmu_context.h b/include/asm-sh64/mmu_context.h
deleted file mode 100644
index 507bf72bb8e1..000000000000
--- a/include/asm-sh64/mmu_context.h
+++ /dev/null
@@ -1,208 +0,0 @@
-#ifndef __ASM_SH64_MMU_CONTEXT_H
-#define __ASM_SH64_MMU_CONTEXT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/mmu_context.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- * ASID handling idea taken from MIPS implementation.
- *
- */
-
-#ifndef __ASSEMBLY__
-
-/*
- * Cache of MMU context last used.
- *
- * The MMU "context" consists of two things:
- *   (a) TLB cache version (or cycle, top 24 bits of mmu_context_cache)
- *   (b) ASID (Address Space IDentifier, bottom 8 bits of mmu_context_cache)
- */
-extern unsigned long mmu_context_cache;
-
-#include <asm/page.h>
-#include <asm-generic/mm_hooks.h>
-
-/* Current mm's pgd */
-extern pgd_t *mmu_pdtp_cache;
-
-#define SR_ASID_MASK		0xffffffffff00ffffULL
-#define SR_ASID_SHIFT		16
-
-#define MMU_CONTEXT_ASID_MASK		0x000000ff
-#define MMU_CONTEXT_VERSION_MASK	0xffffff00
-#define MMU_CONTEXT_FIRST_VERSION	0x00000100
-#define NO_CONTEXT			0
-
-/* ASID is 8-bit value, so it can't be 0x100 */
-#define MMU_NO_ASID			0x100
-
-
-/*
- * Virtual Page Number mask
- */
-#define MMU_VPN_MASK	0xfffff000
-
-static inline void
-get_new_mmu_context(struct mm_struct *mm)
-{
-	extern void flush_tlb_all(void);
-	extern void flush_cache_all(void);
-
-	unsigned long mc = ++mmu_context_cache;
-
-	if (!(mc & MMU_CONTEXT_ASID_MASK)) {
-		/* We exhaust ASID of this version.
-		   Flush all TLB and start new cycle. */
-		flush_tlb_all();
-		/* We have to flush all caches as ASIDs are
-                   used in cache */
-		flush_cache_all();
-		/* Fix version if needed.
-		   Note that we avoid version #0/asid #0 to distingush NO_CONTEXT. */
-		if (!mc)
-			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
-	}
-	mm->context = mc;
-}
-
-/*
- * Get MMU context if needed.
- */
-static __inline__ void
-get_mmu_context(struct mm_struct *mm)
-{
-	if (mm) {
-		unsigned long mc = mmu_context_cache;
-		/* Check if we have old version of context.
-		   If it's old, we need to get new context with new version. */
-		if ((mm->context ^ mc) & MMU_CONTEXT_VERSION_MASK)
-			get_new_mmu_context(mm);
-	}
-}
-
-/*
- * Initialize the context related info for a new mm_struct
- * instance.
- */
-static inline int init_new_context(struct task_struct *tsk,
-					struct mm_struct *mm)
-{
-	mm->context = NO_CONTEXT;
-
-	return 0;
-}
-
-/*
- * Destroy context related info for an mm_struct that is about
- * to be put to rest.
- */
-static inline void destroy_context(struct mm_struct *mm)
-{
-	extern void flush_tlb_mm(struct mm_struct *mm);
-
-	/* Well, at least free TLB entries */
-	flush_tlb_mm(mm);
-}
-
-#endif	/* __ASSEMBLY__ */
-
-/* Common defines */
-#define TLB_STEP	0x00000010
-#define TLB_PTEH	0x00000000
-#define TLB_PTEL	0x00000008
-
-/* PTEH defines */
-#define PTEH_ASID_SHIFT	2
-#define PTEH_VALID	0x0000000000000001
-#define PTEH_SHARED	0x0000000000000002
-#define PTEH_MATCH_ASID	0x00000000000003ff
-
-#ifndef __ASSEMBLY__
-/* This has to be a common function because the next location to fill
- * information is shared. */
-extern void __do_tlb_refill(unsigned long address, unsigned long long is_text_not_data, pte_t *pte);
-
-/* Profiling counter. */
-#ifdef CONFIG_SH64_PROC_TLB
-extern unsigned long long calls_to_do_fast_page_fault;
-#endif
-
-static inline unsigned long get_asid(void)
-{
-	unsigned long long sr;
-
-	asm volatile ("getcon   " __SR ", %0\n\t"
-		      : "=r" (sr));
-
-	sr = (sr >> SR_ASID_SHIFT) & MMU_CONTEXT_ASID_MASK;
-	return (unsigned long) sr;
-}
-
-/* Set ASID into SR */
-static inline void set_asid(unsigned long asid)
-{
-	unsigned long long sr, pc;
-
-	asm volatile ("getcon	" __SR ", %0" : "=r" (sr));
-
-	sr = (sr & SR_ASID_MASK) | (asid << SR_ASID_SHIFT);
-
-	/*
-	 * It is possible that this function may be inlined and so to avoid
-	 * the assembler reporting duplicate symbols we make use of the gas trick
-	 * of generating symbols using numerics and forward reference.
-	 */
-	asm volatile ("movi	1, %1\n\t"
-		      "shlli	%1, 28, %1\n\t"
-		      "or	%0, %1, %1\n\t"
-		      "putcon	%1, " __SR "\n\t"
-		      "putcon	%0, " __SSR "\n\t"
-		      "movi	1f, %1\n\t"
-		      "ori	%1, 1 , %1\n\t"
-		      "putcon	%1, " __SPC "\n\t"
-		      "rte\n"
-		      "1:\n\t"
-		      : "=r" (sr), "=r" (pc) : "0" (sr));
-}
-
-/*
- * After we have set current->mm to a new value, this activates
- * the context for the new mm so we see the new mappings.
- */
-static __inline__ void activate_context(struct mm_struct *mm)
-{
-	get_mmu_context(mm);
-	set_asid(mm->context & MMU_CONTEXT_ASID_MASK);
-}
-
-
-static __inline__ void switch_mm(struct mm_struct *prev,
-				 struct mm_struct *next,
-				 struct task_struct *tsk)
-{
-	if (prev != next) {
-		mmu_pdtp_cache = next->pgd;
-		activate_context(next);
-	}
-}
-
-#define deactivate_mm(tsk,mm)	do { } while (0)
-
-#define activate_mm(prev, next) \
-	switch_mm((prev),(next),NULL)
-
-static inline void
-enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
-{
-}
-
-#endif	/* __ASSEMBLY__ */
-
-#endif /* __ASM_SH64_MMU_CONTEXT_H */
-- 
cgit v1.2.3


From 33f242ed11ce6b5fbe73fe4ece7ef4bc2f4e2851 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 16:57:27 +0900
Subject: sh: SH-5 pt_regs.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/ptrace.h | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/ptrace.h b/include/asm-sh/ptrace.h
index b9789c8b4d15..a83a7b45ba6f 100644
--- a/include/asm-sh/ptrace.h
+++ b/include/asm-sh/ptrace.h
@@ -5,7 +5,16 @@
  * Copyright (C) 1999, 2000  Niibe Yutaka
  *
  */
-
+#if defined(__SH5__) || defined(CONFIG_SUPERH64)
+struct pt_regs {
+	unsigned long long pc;
+	unsigned long long sr;
+	unsigned long long syscall_nr;
+	unsigned long long regs[63];
+	unsigned long long tregs[8];
+	unsigned long long pad[2];
+};
+#else
 /*
  * GCC defines register number like this:
  * -----------------------------
@@ -28,7 +37,7 @@
 
 #define REG_PR		17
 #define REG_SR		18
-#define REG_GBR      	19
+#define REG_GBR		19
 #define REG_MACH	20
 #define REG_MACL	21
 
@@ -80,10 +89,14 @@ struct pt_dspregs {
 
 #define	PTRACE_GETDSPREGS	55
 #define	PTRACE_SETDSPREGS	56
+#endif
 
 #ifdef __KERNEL__
-#define user_mode(regs) (((regs)->sr & 0x40000000)==0)
-#define instruction_pointer(regs) ((regs)->pc)
+#include <asm/addrspace.h>
+
+#define user_mode(regs)			(((regs)->sr & 0x40000000)==0)
+#define instruction_pointer(regs)	((regs)->pc)
+
 extern void show_regs(struct pt_regs *);
 
 #ifdef CONFIG_SH_DSP
@@ -100,10 +113,13 @@ static inline unsigned long profile_pc(struct pt_regs *regs)
 {
 	unsigned long pc = instruction_pointer(regs);
 
-	if (pc >= 0xa0000000UL && pc < 0xc0000000UL)
+#ifdef P2SEG
+	if (pc >= P2SEG && pc < P3SEG)
 		pc -= 0x20000000;
+#endif
+
 	return pc;
 }
-#endif
+#endif /* __KERNEL__ */
 
 #endif /* __ASM_SH_PTRACE_H */
-- 
cgit v1.2.3


From af3c7dfe822b598a2f977098101ed8b63cf0fdd1 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 9 Nov 2007 17:08:54 +0900
Subject: sh: Split out processor.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile_64    |  21 ++++
 include/asm-sh/processor.h    | 247 +-----------------------------------
 include/asm-sh/processor_32.h | 248 ++++++++++++++++++++++++++++++++++++
 include/asm-sh/processor_64.h | 277 ++++++++++++++++++++++++++++++++++++++++
 include/asm-sh64/processor.h  | 287 ------------------------------------------
 5 files changed, 552 insertions(+), 528 deletions(-)
 create mode 100644 arch/sh/kernel/Makefile_64
 create mode 100644 include/asm-sh/processor_32.h
 create mode 100644 include/asm-sh/processor_64.h
 delete mode 100644 include/asm-sh64/processor.h

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
new file mode 100644
index 000000000000..7dd7de9ff3c0
--- /dev/null
+++ b/arch/sh/kernel/Makefile_64
@@ -0,0 +1,21 @@
+extra-y	:= head.o init_task.o vmlinux.lds
+
+obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
+	   ptrace.o semaphore.o setup.o signal.o sys_sh.o syscalls.o \
+	   time.o topology.o traps.o
+
+obj-y				+= cpu/ timers/
+obj-$(CONFIG_VSYSCALL)		+= vsyscall/
+obj-$(CONFIG_SMP)		+= smp.o
+obj-$(CONFIG_CF_ENABLER)	+= cf-enabler.o
+obj-$(CONFIG_SH_STANDARD_BIOS)	+= sh_bios.o
+obj-$(CONFIG_SH_KGDB)		+= kgdb_stub.o kgdb_jmp.o
+obj-$(CONFIG_SH_CPU_FREQ)	+= cpufreq.o
+obj-$(CONFIG_MODULES)		+= sh_ksyms.o module.o
+obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
+obj-$(CONFIG_PM)		+= pm.o
+obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+
+EXTRA_CFLAGS += -Werror
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index fda68480f377..bf01f486c3e1 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -1,31 +1,5 @@
-/*
- * include/asm-sh/processor.h
- *
- * Copyright (C) 1999, 2000  Niibe Yutaka
- * Copyright (C) 2002, 2003  Paul Mundt
- */
-
 #ifndef __ASM_SH_PROCESSOR_H
 #define __ASM_SH_PROCESSOR_H
-#ifdef __KERNEL__
-
-#include <linux/compiler.h>
-#include <asm/page.h>
-#include <asm/types.h>
-#include <asm/cache.h>
-#include <asm/ptrace.h>
-#include <asm/cpu-features.h>
-
-/*
- * Default implementation of macro that returns current
- * instruction pointer ("program counter").
- */
-#define current_text_addr() ({ void *pc; __asm__("mova	1f, %0\n1:":"=z" (pc)); pc; })
-
-/* Core Processor Version Register */
-#define CCN_PVR		0xff000030
-#define CCN_CVR		0xff000040
-#define CCN_PRR		0xff000044
 
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
@@ -57,229 +31,20 @@ enum cpu_type {
 	/* SH4AL-DSP types */
 	CPU_SH7343, CPU_SH7722,
 
+	/* SH-5 types */
+        CPU_SH5_101, CPU_SH5_103,
+
 	/* Unknown subtype */
 	CPU_SH_NONE
 };
 
-struct sh_cpuinfo {
-	unsigned int type;
-	unsigned long loops_per_jiffy;
-	unsigned long asid_cache;
-
-	struct cache_info icache;	/* Primary I-cache */
-	struct cache_info dcache;	/* Primary D-cache */
-	struct cache_info scache;	/* Secondary cache */
-
-	unsigned long flags;
-} __attribute__ ((aligned(L1_CACHE_BYTES)));
-
-extern struct sh_cpuinfo cpu_data[];
-#define boot_cpu_data cpu_data[0]
-#define current_cpu_data cpu_data[smp_processor_id()]
-#define raw_current_cpu_data cpu_data[raw_smp_processor_id()]
-
-/*
- * User space process size: 2GB.
- *
- * Since SH7709 and SH7750 have "area 7", we can't use 0x7c000000--0x7fffffff
- */
-#define TASK_SIZE	0x7c000000UL
-
-/* This decides where the kernel will search for a free chunk of vm
- * space during mmap's.
- */
-#define TASK_UNMAPPED_BASE	(TASK_SIZE / 3)
-
-/*
- * Bit of SR register
- *
- * FD-bit:
- *     When it's set, it means the processor doesn't have right to use FPU,
- *     and it results exception when the floating operation is executed.
- *
- * IMASK-bit:
- *     Interrupt level mask
- */
-#define SR_FD		0x00008000
-#define SR_DSP		0x00001000
-#define SR_IMASK	0x000000f0
-
-/*
- * FPU structure and data
- */
-
-struct sh_fpu_hard_struct {
-	unsigned long fp_regs[16];
-	unsigned long xfp_regs[16];
-	unsigned long fpscr;
-	unsigned long fpul;
-
-	long status; /* software status information */
-};
-
-/* Dummy fpu emulator  */
-struct sh_fpu_soft_struct {
-	unsigned long fp_regs[16];
-	unsigned long xfp_regs[16];
-	unsigned long fpscr;
-	unsigned long fpul;
-
-	unsigned char lookahead;
-	unsigned long entry_pc;
-};
-
-union sh_fpu_union {
-	struct sh_fpu_hard_struct hard;
-	struct sh_fpu_soft_struct soft;
-};
-
-struct thread_struct {
-	/* Saved registers when thread is descheduled */
-	unsigned long sp;
-	unsigned long pc;
-
-	/* Hardware debugging registers */
-	unsigned long ubc_pc;
-
-	/* floating point info */
-	union sh_fpu_union fpu;
-};
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-/* Count of active tasks with UBC settings */
-extern int ubc_usercnt;
-
-#define INIT_THREAD  {						\
-	.sp = sizeof(init_stack) + (long) &init_stack,		\
-}
-
-/*
- * Do necessary setup to start up a newly executed thread.
- */
-#define start_thread(regs, new_pc, new_sp)	 \
-	set_fs(USER_DS);			 \
-	regs->pr = 0;				 \
-	regs->sr = SR_FD;	/* User mode. */ \
-	regs->pc = new_pc;			 \
-	regs->regs[15] = new_sp
-
-/* Forward declaration, a strange C thing */
-struct task_struct;
-struct mm_struct;
-
-/* Free all resources held by a thread. */
-extern void release_thread(struct task_struct *);
-
-/* Prepare to copy thread state - unlazy all lazy status */
-#define prepare_to_copy(tsk)	do { } while (0)
-
-/*
- * create a kernel thread without removing it from tasklists
- */
-extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
-
-/* Copy and release all segment info associated with a VM */
-#define copy_segments(p, mm)	do { } while(0)
-#define release_segments(mm)	do { } while(0)
-
-/*
- * FPU lazy state save handling.
- */
-
-static __inline__ void disable_fpu(void)
-{
-	unsigned long __dummy;
-
-	/* Set FD flag in SR */
-	__asm__ __volatile__("stc	sr, %0\n\t"
-			     "or	%1, %0\n\t"
-			     "ldc	%0, sr"
-			     : "=&r" (__dummy)
-			     : "r" (SR_FD));
-}
-
-static __inline__ void enable_fpu(void)
-{
-	unsigned long __dummy;
-
-	/* Clear out FD flag in SR */
-	__asm__ __volatile__("stc	sr, %0\n\t"
-			     "and	%1, %0\n\t"
-			     "ldc	%0, sr"
-			     : "=&r" (__dummy)
-			     : "r" (~SR_FD));
-}
-
-static __inline__ void release_fpu(struct pt_regs *regs)
-{
-	regs->sr |= SR_FD;
-}
-
-static __inline__ void grab_fpu(struct pt_regs *regs)
-{
-	regs->sr &= ~SR_FD;
-}
-
-extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs);
-
-#define unlazy_fpu(tsk, regs) do {			\
-	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {	\
-		save_fpu(tsk, regs);			\
-	}						\
-} while (0)
-
-#define clear_fpu(tsk, regs) do {				\
-	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {		\
-		clear_tsk_thread_flag(tsk, TIF_USEDFPU);	\
-		release_fpu(regs);				\
-	}							\
-} while (0)
-
-/* Double presision, NANS as NANS, rounding to nearest, no exceptions */
-#define FPSCR_INIT  0x00080000
-
-#define	FPSCR_CAUSE_MASK	0x0001f000	/* Cause bits */
-#define	FPSCR_FLAG_MASK		0x0000007c	/* Flag bits */
-
-/*
- * Return saved PC of a blocked thread.
- */
-#define thread_saved_pc(tsk)	(tsk->thread.pc)
-
-void show_trace(struct task_struct *tsk, unsigned long *sp,
-		struct pt_regs *regs);
-extern unsigned long get_wchan(struct task_struct *p);
-
-#define KSTK_EIP(tsk)  (task_pt_regs(tsk)->pc)
-#define KSTK_ESP(tsk)  (task_pt_regs(tsk)->regs[15])
-
-#define cpu_sleep()	__asm__ __volatile__ ("sleep" : : : "memory")
-#define cpu_relax()	barrier()
-
-#if defined(CONFIG_CPU_SH2A) || defined(CONFIG_CPU_SH3) || \
-    defined(CONFIG_CPU_SH4)
-#define PREFETCH_STRIDE		L1_CACHE_BYTES
-#define ARCH_HAS_PREFETCH
-#define ARCH_HAS_PREFETCHW
-static inline void prefetch(void *x)
-{
-	__asm__ __volatile__ ("pref @%0\n\t" : : "r" (x) : "memory");
-}
-
-#define prefetchw(x)	prefetch(x)
-#endif
-
-#ifdef CONFIG_VSYSCALL
-extern int vsyscall_init(void);
+#ifdef CONFIG_SUPERH32
+# include "processor_32.h"
 #else
-#define vsyscall_init() do { } while (0)
+# include "processor_64.h"
 #endif
 
 /* arch/sh/kernel/setup.c */
 const char *get_cpu_subtype(struct sh_cpuinfo *c);
 
-#endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_H */
diff --git a/include/asm-sh/processor_32.h b/include/asm-sh/processor_32.h
new file mode 100644
index 000000000000..e10d0ee0c22b
--- /dev/null
+++ b/include/asm-sh/processor_32.h
@@ -0,0 +1,248 @@
+/*
+ * include/asm-sh/processor.h
+ *
+ * Copyright (C) 1999, 2000  Niibe Yutaka
+ * Copyright (C) 2002, 2003  Paul Mundt
+ */
+
+#ifndef __ASM_SH_PROCESSOR_32_H
+#define __ASM_SH_PROCESSOR_32_H
+#ifdef __KERNEL__
+
+#include <linux/compiler.h>
+#include <asm/page.h>
+#include <asm/types.h>
+#include <asm/cache.h>
+#include <asm/ptrace.h>
+#include <asm/cpu-features.h>
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr() ({ void *pc; __asm__("mova	1f, %0\n1:":"=z" (pc)); pc; })
+
+/* Core Processor Version Register */
+#define CCN_PVR		0xff000030
+#define CCN_CVR		0xff000040
+#define CCN_PRR		0xff000044
+
+struct sh_cpuinfo {
+	unsigned int type;
+	unsigned long loops_per_jiffy;
+	unsigned long asid_cache;
+
+	struct cache_info icache;	/* Primary I-cache */
+	struct cache_info dcache;	/* Primary D-cache */
+	struct cache_info scache;	/* Secondary cache */
+
+	unsigned long flags;
+} __attribute__ ((aligned(L1_CACHE_BYTES)));
+
+extern struct sh_cpuinfo cpu_data[];
+#define boot_cpu_data cpu_data[0]
+#define current_cpu_data cpu_data[smp_processor_id()]
+#define raw_current_cpu_data cpu_data[raw_smp_processor_id()]
+
+/*
+ * User space process size: 2GB.
+ *
+ * Since SH7709 and SH7750 have "area 7", we can't use 0x7c000000--0x7fffffff
+ */
+#define TASK_SIZE	0x7c000000UL
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE	(TASK_SIZE / 3)
+
+/*
+ * Bit of SR register
+ *
+ * FD-bit:
+ *     When it's set, it means the processor doesn't have right to use FPU,
+ *     and it results exception when the floating operation is executed.
+ *
+ * IMASK-bit:
+ *     Interrupt level mask
+ */
+#define SR_FD		0x00008000
+#define SR_DSP		0x00001000
+#define SR_IMASK	0x000000f0
+
+/*
+ * FPU structure and data
+ */
+
+struct sh_fpu_hard_struct {
+	unsigned long fp_regs[16];
+	unsigned long xfp_regs[16];
+	unsigned long fpscr;
+	unsigned long fpul;
+
+	long status; /* software status information */
+};
+
+/* Dummy fpu emulator  */
+struct sh_fpu_soft_struct {
+	unsigned long fp_regs[16];
+	unsigned long xfp_regs[16];
+	unsigned long fpscr;
+	unsigned long fpul;
+
+	unsigned char lookahead;
+	unsigned long entry_pc;
+};
+
+union sh_fpu_union {
+	struct sh_fpu_hard_struct hard;
+	struct sh_fpu_soft_struct soft;
+};
+
+struct thread_struct {
+	/* Saved registers when thread is descheduled */
+	unsigned long sp;
+	unsigned long pc;
+
+	/* Hardware debugging registers */
+	unsigned long ubc_pc;
+
+	/* floating point info */
+	union sh_fpu_union fpu;
+};
+
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
+/* Count of active tasks with UBC settings */
+extern int ubc_usercnt;
+
+#define INIT_THREAD  {						\
+	.sp = sizeof(init_stack) + (long) &init_stack,		\
+}
+
+/*
+ * Do necessary setup to start up a newly executed thread.
+ */
+#define start_thread(regs, new_pc, new_sp)	 \
+	set_fs(USER_DS);			 \
+	regs->pr = 0;				 \
+	regs->sr = SR_FD;	/* User mode. */ \
+	regs->pc = new_pc;			 \
+	regs->regs[15] = new_sp
+
+/* Forward declaration, a strange C thing */
+struct task_struct;
+struct mm_struct;
+
+/* Free all resources held by a thread. */
+extern void release_thread(struct task_struct *);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+#define prepare_to_copy(tsk)	do { } while (0)
+
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+/* Copy and release all segment info associated with a VM */
+#define copy_segments(p, mm)	do { } while(0)
+#define release_segments(mm)	do { } while(0)
+
+/*
+ * FPU lazy state save handling.
+ */
+
+static __inline__ void disable_fpu(void)
+{
+	unsigned long __dummy;
+
+	/* Set FD flag in SR */
+	__asm__ __volatile__("stc	sr, %0\n\t"
+			     "or	%1, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy)
+			     : "r" (SR_FD));
+}
+
+static __inline__ void enable_fpu(void)
+{
+	unsigned long __dummy;
+
+	/* Clear out FD flag in SR */
+	__asm__ __volatile__("stc	sr, %0\n\t"
+			     "and	%1, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy)
+			     : "r" (~SR_FD));
+}
+
+static __inline__ void release_fpu(struct pt_regs *regs)
+{
+	regs->sr |= SR_FD;
+}
+
+static __inline__ void grab_fpu(struct pt_regs *regs)
+{
+	regs->sr &= ~SR_FD;
+}
+
+extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs);
+
+#define unlazy_fpu(tsk, regs) do {			\
+	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {	\
+		save_fpu(tsk, regs);			\
+	}						\
+} while (0)
+
+#define clear_fpu(tsk, regs) do {				\
+	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {		\
+		clear_tsk_thread_flag(tsk, TIF_USEDFPU);	\
+		release_fpu(regs);				\
+	}							\
+} while (0)
+
+/* Double presision, NANS as NANS, rounding to nearest, no exceptions */
+#define FPSCR_INIT  0x00080000
+
+#define	FPSCR_CAUSE_MASK	0x0001f000	/* Cause bits */
+#define	FPSCR_FLAG_MASK		0x0000007c	/* Flag bits */
+
+/*
+ * Return saved PC of a blocked thread.
+ */
+#define thread_saved_pc(tsk)	(tsk->thread.pc)
+
+void show_trace(struct task_struct *tsk, unsigned long *sp,
+		struct pt_regs *regs);
+extern unsigned long get_wchan(struct task_struct *p);
+
+#define KSTK_EIP(tsk)  (task_pt_regs(tsk)->pc)
+#define KSTK_ESP(tsk)  (task_pt_regs(tsk)->regs[15])
+
+#define cpu_sleep()	__asm__ __volatile__ ("sleep" : : : "memory")
+#define cpu_relax()	barrier()
+
+#if defined(CONFIG_CPU_SH2A) || defined(CONFIG_CPU_SH3) || \
+    defined(CONFIG_CPU_SH4)
+#define PREFETCH_STRIDE		L1_CACHE_BYTES
+#define ARCH_HAS_PREFETCH
+#define ARCH_HAS_PREFETCHW
+static inline void prefetch(void *x)
+{
+	__asm__ __volatile__ ("pref @%0\n\t" : : "r" (x) : "memory");
+}
+
+#define prefetchw(x)	prefetch(x)
+#endif
+
+#ifdef CONFIG_VSYSCALL
+extern int vsyscall_init(void);
+#else
+#define vsyscall_init() do { } while (0)
+#endif
+
+#endif /* __KERNEL__ */
+#endif /* __ASM_SH_PROCESSOR_32_H */
diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
new file mode 100644
index 000000000000..5430e437b915
--- /dev/null
+++ b/include/asm-sh/processor_64.h
@@ -0,0 +1,277 @@
+#ifndef __ASM_SH_PROCESSOR_64_H
+#define __ASM_SH_PROCESSOR_64_H
+
+/*
+ * include/asm-sh/processor_64.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003  Paul Mundt
+ * Copyright (C) 2004  Richard Curnow
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#ifndef __ASSEMBLY__
+
+#include <linux/compiler.h>
+#include <asm/page.h>
+#include <asm/types.h>
+#include <asm/cache.h>
+#include <asm/cpu/registers.h>
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr() ({ \
+void *pc; \
+unsigned long long __dummy = 0; \
+__asm__("gettr	tr0, %1\n\t" \
+	"pta	4, tr0\n\t" \
+	"gettr	tr0, %0\n\t" \
+	"ptabs	%1, tr0\n\t"	\
+	:"=r" (pc), "=r" (__dummy) \
+	: "1" (__dummy)); \
+pc; })
+
+/*
+ * TLB information structure
+ *
+ * Defined for both I and D tlb, per-processor.
+ */
+struct tlb_info {
+	unsigned long long next;
+	unsigned long long first;
+	unsigned long long last;
+
+	unsigned int entries;
+	unsigned int step;
+
+	unsigned long flags;
+};
+
+struct sh_cpuinfo {
+	enum cpu_type type;
+	unsigned long loops_per_jiffy;
+
+	char	hard_math;
+
+	unsigned long *pgd_quick;
+	unsigned long *pmd_quick;
+	unsigned long *pte_quick;
+	unsigned long pgtable_cache_sz;
+	unsigned int cpu_clock, master_clock, bus_clock, module_clock;
+
+	/* Cache info */
+	struct cache_info icache;
+	struct cache_info dcache;
+
+	/* TLB info */
+	struct tlb_info itlb;
+	struct tlb_info dtlb;
+};
+
+extern struct sh_cpuinfo boot_cpu_data;
+
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+
+#endif
+
+/*
+ * User space process size: 2GB - 4k.
+ */
+#define TASK_SIZE	0x7ffff000UL
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE	(TASK_SIZE / 3)
+
+/*
+ * Bit of SR register
+ *
+ * FD-bit:
+ *     When it's set, it means the processor doesn't have right to use FPU,
+ *     and it results exception when the floating operation is executed.
+ *
+ * IMASK-bit:
+ *     Interrupt level mask
+ *
+ * STEP-bit:
+ *     Single step bit
+ *
+ */
+#define SR_FD    0x00008000
+
+#if defined(CONFIG_SH64_SR_WATCH)
+#define SR_MMU   0x84000000
+#else
+#define SR_MMU   0x80000000
+#endif
+
+#define SR_IMASK 0x000000f0
+#define SR_SSTEP 0x08000000
+
+#ifndef __ASSEMBLY__
+
+/*
+ * FPU structure and data : require 8-byte alignment as we need to access it
+   with fld.p, fst.p
+ */
+
+struct sh_fpu_hard_struct {
+	unsigned long fp_regs[64];
+	unsigned int fpscr;
+	/* long status; * software status information */
+};
+
+#if 0
+/* Dummy fpu emulator  */
+struct sh_fpu_soft_struct {
+	unsigned long long fp_regs[32];
+	unsigned int fpscr;
+	unsigned char lookahead;
+	unsigned long entry_pc;
+};
+#endif
+
+union sh_fpu_union {
+	struct sh_fpu_hard_struct hard;
+	/* 'hard' itself only produces 32 bit alignment, yet we need
+	   to access it using 64 bit load/store as well. */
+	unsigned long long alignment_dummy;
+};
+
+struct thread_struct {
+	unsigned long sp;
+	unsigned long pc;
+	/* This stores the address of the pt_regs built during a context
+	   switch, or of the register save area built for a kernel mode
+	   exception.  It is used for backtracing the stack of a sleeping task
+	   or one that traps in kernel mode. */
+        struct pt_regs *kregs;
+	/* This stores the address of the pt_regs constructed on entry from
+	   user mode.  It is a fixed value over the lifetime of a process, or
+	   NULL for a kernel thread. */
+	struct pt_regs *uregs;
+
+	unsigned long trap_no, error_code;
+	unsigned long address;
+	/* Hardware debugging registers may come here */
+
+	/* floating point info */
+	union sh_fpu_union fpu;
+};
+
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
+#define INIT_MMAP \
+{ &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, NULL, NULL }
+
+extern  struct pt_regs fake_swapper_regs;
+
+#define INIT_THREAD  {				\
+	.sp		= sizeof(init_stack) +	\
+			  (long) &init_stack,	\
+	.pc		= 0,			\
+        .kregs		= &fake_swapper_regs,	\
+	.uregs	        = NULL,			\
+	.trap_no	= 0,			\
+	.error_code	= 0,			\
+	.address	= 0,			\
+	.fpu		= { { { 0, } }, }	\
+}
+
+/*
+ * Do necessary setup to start up a newly executed thread.
+ */
+#define SR_USER (SR_MMU | SR_FD)
+
+#define start_thread(regs, new_pc, new_sp)			\
+	set_fs(USER_DS);					\
+	regs->sr = SR_USER;	/* User mode. */		\
+	regs->pc = new_pc - 4;	/* Compensate syscall exit */	\
+	regs->pc |= 1;		/* Set SHmedia ! */		\
+	regs->regs[18] = 0;					\
+	regs->regs[15] = new_sp
+
+/* Forward declaration, a strange C thing */
+struct task_struct;
+struct mm_struct;
+
+/* Free all resources held by a thread. */
+extern void release_thread(struct task_struct *);
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+
+/* Copy and release all segment info associated with a VM */
+#define copy_segments(p, mm)	do { } while (0)
+#define release_segments(mm)	do { } while (0)
+#define forget_segments()	do { } while (0)
+#define prepare_to_copy(tsk)	do { } while (0)
+/*
+ * FPU lazy state save handling.
+ */
+
+static inline void release_fpu(void)
+{
+	unsigned long long __dummy;
+
+	/* Set FD flag in SR */
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "or	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy)
+			     : "r" (SR_FD));
+}
+
+static inline void grab_fpu(void)
+{
+	unsigned long long __dummy;
+
+	/* Clear out FD flag in SR */
+	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
+			     "and	%0, %1, %0\n\t"
+			     "putcon	%0, " __SR "\n\t"
+			     : "=&r" (__dummy)
+			     : "r" (~SR_FD));
+}
+
+/* Round to nearest, no exceptions on inexact, overflow, underflow,
+   zero-divide, invalid.  Configure option for whether to flush denorms to
+   zero, or except if a denorm is encountered.  */
+#if defined(CONFIG_SH64_FPU_DENORM_FLUSH)
+#define FPSCR_INIT  0x00040000
+#else
+#define FPSCR_INIT  0x00000000
+#endif
+
+/* Save the current FP regs */
+void fpsave(struct sh_fpu_hard_struct *fpregs);
+
+/* Initialise the FP state of a task */
+void fpinit(struct sh_fpu_hard_struct *fpregs);
+
+extern struct task_struct *last_task_used_math;
+
+/*
+ * Return saved PC of a blocked thread.
+ */
+#define thread_saved_pc(tsk)	(tsk->thread.pc)
+
+extern unsigned long get_wchan(struct task_struct *p);
+
+#define KSTK_EIP(tsk)  ((tsk)->thread.pc)
+#define KSTK_ESP(tsk)  ((tsk)->thread.sp)
+
+#define cpu_relax()	barrier()
+
+#endif	/* __ASSEMBLY__ */
+#endif /* __ASM_SH_PROCESSOR_64_H */
diff --git a/include/asm-sh64/processor.h b/include/asm-sh64/processor.h
deleted file mode 100644
index eb2bee4b47b9..000000000000
--- a/include/asm-sh64/processor.h
+++ /dev/null
@@ -1,287 +0,0 @@
-#ifndef __ASM_SH64_PROCESSOR_H
-#define __ASM_SH64_PROCESSOR_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/processor.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- * Copyright (C) 2004  Richard Curnow
- *
- */
-
-#include <asm/page.h>
-
-#ifndef __ASSEMBLY__
-
-#include <asm/types.h>
-#include <asm/cache.h>
-#include <asm/registers.h>
-#include <linux/threads.h>
-#include <linux/compiler.h>
-
-/*
- * Default implementation of macro that returns current
- * instruction pointer ("program counter").
- */
-#define current_text_addr() ({ \
-void *pc; \
-unsigned long long __dummy = 0; \
-__asm__("gettr	tr0, %1\n\t" \
-	"pta	4, tr0\n\t" \
-	"gettr	tr0, %0\n\t" \
-	"ptabs	%1, tr0\n\t"	\
-	:"=r" (pc), "=r" (__dummy) \
-	: "1" (__dummy)); \
-pc; })
-
-/*
- *  CPU type and hardware bug flags. Kept separately for each CPU.
- */
-enum cpu_type {
-	CPU_SH5_101,
-	CPU_SH5_103,
-	CPU_SH_NONE
-};
-
-/*
- * TLB information structure
- *
- * Defined for both I and D tlb, per-processor.
- */
-struct tlb_info {
-	unsigned long long next;
-	unsigned long long first;
-	unsigned long long last;
-
-	unsigned int entries;
-	unsigned int step;
-
-	unsigned long flags;
-};
-
-struct sh_cpuinfo {
-	enum cpu_type type;
-	unsigned long loops_per_jiffy;
-
-	char	hard_math;
-
-	unsigned long *pgd_quick;
-	unsigned long *pmd_quick;
-	unsigned long *pte_quick;
-	unsigned long pgtable_cache_sz;
-	unsigned int cpu_clock, master_clock, bus_clock, module_clock;
-
-	/* Cache info */
-	struct cache_info icache;
-	struct cache_info dcache;
-
-	/* TLB info */
-	struct tlb_info itlb;
-	struct tlb_info dtlb;
-};
-
-extern struct sh_cpuinfo boot_cpu_data;
-
-#define cpu_data (&boot_cpu_data)
-#define current_cpu_data boot_cpu_data
-
-#endif
-
-/*
- * User space process size: 2GB - 4k.
- */
-#define TASK_SIZE	0x7ffff000UL
-
-/* This decides where the kernel will search for a free chunk of vm
- * space during mmap's.
- */
-#define TASK_UNMAPPED_BASE	(TASK_SIZE / 3)
-
-/*
- * Bit of SR register
- *
- * FD-bit:
- *     When it's set, it means the processor doesn't have right to use FPU,
- *     and it results exception when the floating operation is executed.
- *
- * IMASK-bit:
- *     Interrupt level mask
- *
- * STEP-bit:
- *     Single step bit
- *
- */
-#define SR_FD    0x00008000
-
-#if defined(CONFIG_SH64_SR_WATCH)
-#define SR_MMU   0x84000000
-#else
-#define SR_MMU   0x80000000
-#endif
-
-#define SR_IMASK 0x000000f0
-#define SR_SSTEP 0x08000000
-
-#ifndef __ASSEMBLY__
-
-/*
- * FPU structure and data : require 8-byte alignment as we need to access it
-   with fld.p, fst.p
- */
-
-struct sh_fpu_hard_struct {
-	unsigned long fp_regs[64];
-	unsigned int fpscr;
-	/* long status; * software status information */
-};
-
-#if 0
-/* Dummy fpu emulator  */
-struct sh_fpu_soft_struct {
-	unsigned long long fp_regs[32];
-	unsigned int fpscr;
-	unsigned char lookahead;
-	unsigned long entry_pc;
-};
-#endif
-
-union sh_fpu_union {
-	struct sh_fpu_hard_struct hard;
-	/* 'hard' itself only produces 32 bit alignment, yet we need
-	   to access it using 64 bit load/store as well. */
-	unsigned long long alignment_dummy;
-};
-
-struct thread_struct {
-	unsigned long sp;
-	unsigned long pc;
-	/* This stores the address of the pt_regs built during a context
-	   switch, or of the register save area built for a kernel mode
-	   exception.  It is used for backtracing the stack of a sleeping task
-	   or one that traps in kernel mode. */
-        struct pt_regs *kregs;
-	/* This stores the address of the pt_regs constructed on entry from
-	   user mode.  It is a fixed value over the lifetime of a process, or
-	   NULL for a kernel thread. */
-	struct pt_regs *uregs;
-
-	unsigned long trap_no, error_code;
-	unsigned long address;
-	/* Hardware debugging registers may come here */
-
-	/* floating point info */
-	union sh_fpu_union fpu;
-};
-
-#define INIT_MMAP \
-{ &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, NULL, NULL }
-
-extern  struct pt_regs fake_swapper_regs;
-
-#define INIT_THREAD  {				\
-	.sp		= sizeof(init_stack) +	\
-			  (long) &init_stack,	\
-	.pc		= 0,			\
-        .kregs		= &fake_swapper_regs,	\
-	.uregs	        = NULL,			\
-	.trap_no	= 0,			\
-	.error_code	= 0,			\
-	.address	= 0,			\
-	.fpu		= { { { 0, } }, }	\
-}
-
-/*
- * Do necessary setup to start up a newly executed thread.
- */
-#define SR_USER (SR_MMU | SR_FD)
-
-#define start_thread(regs, new_pc, new_sp) 	 		\
-	set_fs(USER_DS);			 		\
-	regs->sr = SR_USER;	/* User mode. */ 		\
-	regs->pc = new_pc - 4;	/* Compensate syscall exit */	\
-	regs->pc |= 1;		/* Set SHmedia ! */		\
-	regs->regs[18] = 0;   		 	 		\
-	regs->regs[15] = new_sp
-
-/* Forward declaration, a strange C thing */
-struct task_struct;
-struct mm_struct;
-
-/* Free all resources held by a thread. */
-extern void release_thread(struct task_struct *);
-/*
- * create a kernel thread without removing it from tasklists
- */
-extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
-
-
-/* Copy and release all segment info associated with a VM */
-#define copy_segments(p, mm)	do { } while (0)
-#define release_segments(mm)	do { } while (0)
-#define forget_segments()	do { } while (0)
-#define prepare_to_copy(tsk)	do { } while (0)
-/*
- * FPU lazy state save handling.
- */
-
-static inline void release_fpu(void)
-{
-	unsigned long long __dummy;
-
-	/* Set FD flag in SR */
-	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
-			     "or	%0, %1, %0\n\t"
-			     "putcon	%0, " __SR "\n\t"
-			     : "=&r" (__dummy)
-			     : "r" (SR_FD));
-}
-
-static inline void grab_fpu(void)
-{
-	unsigned long long __dummy;
-
-	/* Clear out FD flag in SR */
-	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
-			     "and	%0, %1, %0\n\t"
-			     "putcon	%0, " __SR "\n\t"
-			     : "=&r" (__dummy)
-			     : "r" (~SR_FD));
-}
-
-/* Round to nearest, no exceptions on inexact, overflow, underflow,
-   zero-divide, invalid.  Configure option for whether to flush denorms to
-   zero, or except if a denorm is encountered.  */
-#if defined(CONFIG_SH64_FPU_DENORM_FLUSH)
-#define FPSCR_INIT  0x00040000
-#else
-#define FPSCR_INIT  0x00000000
-#endif
-
-/* Save the current FP regs */
-void fpsave(struct sh_fpu_hard_struct *fpregs);
-
-/* Initialise the FP state of a task */
-void fpinit(struct sh_fpu_hard_struct *fpregs);
-
-extern struct task_struct *last_task_used_math;
-
-/*
- * Return saved PC of a blocked thread.
- */
-#define thread_saved_pc(tsk)	(tsk->thread.pc)
-
-extern unsigned long get_wchan(struct task_struct *p);
-
-#define KSTK_EIP(tsk)  ((tsk)->thread.pc)
-#define KSTK_ESP(tsk)  ((tsk)->thread.sp)
-
-#define cpu_relax()	barrier()
-
-#endif	/* __ASSEMBLY__ */
-#endif /* __ASM_SH64_PROCESSOR_H */
-
-- 
cgit v1.2.3


From 36bcd39dbca824daffe16d607ae574b6edc7d31a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 19:16:55 +0900
Subject: sh: Split out 29-bit and 32-bit physical mode definitions.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/Kconfig       | 11 +++++++++++
 arch/sh/mm/Makefile      |  2 +-
 include/asm-sh/pgtable.h | 17 +++++++++++++++++
 3 files changed, 29 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index a7bbc2cbbf8c..8192c872a69a 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -46,9 +46,20 @@ config MEMORY_SIZE
 	  as 0x00400000 which was the default value before this became
 	  configurable.
 
+# Physical addressing modes
+
+config 29BIT
+	def_bool !32BIT
+	depends on SUPERH32
+
 config 32BIT
+	bool
+	default y if CPU_SH5
+
+config PMB
 	bool "Support 32-bit physical addressing through PMB"
 	depends on MMU && (CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785)
+	select 32BIT
 	default y
 	help
 	  If you say Y here, physical addressing will be extended to
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index aa44607f072d..d35caccd0aaa 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -31,7 +31,7 @@ endif
 endif
 
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
-obj-$(CONFIG_32BIT)		+= pmb.o
+obj-$(CONFIG_PMB)		+= pmb.o
 obj-$(CONFIG_NUMA)		+= numa.o
 
 EXTRA_CFLAGS += -Werror
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 8f1e8be8d15d..6ab3ba82d255 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -28,6 +28,23 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 
 #endif /* !__ASSEMBLY__ */
 
+/*
+ * Effective and physical address definitions, to aid with sign
+ * extension.
+ */
+#define NEFF		32
+#define	NEFF_SIGN	(1LL << (NEFF - 1))
+#define	NEFF_MASK	(-1LL << NEFF)
+
+#ifdef CONFIG_29BIT
+#define NPHYS		29
+#else
+#define NPHYS		32
+#endif
+
+#define	NPHYS_SIGN	(1LL << (NPHYS - 1))
+#define	NPHYS_MASK	(-1LL << NPHYS)
+
 /*
  * traditional two-level paging structure
  */
-- 
cgit v1.2.3


From a62a3861e0adfd2612372883b5b1fc05a5182796 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 19:46:31 +0900
Subject: sh: Split out system.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h    |  97 ++---------------------
 include/asm-sh/system_32.h |  97 +++++++++++++++++++++++
 include/asm-sh/system_64.h |  39 ++++++++++
 include/asm-sh64/system.h  | 190 ---------------------------------------------
 4 files changed, 144 insertions(+), 279 deletions(-)
 create mode 100644 include/asm-sh/system_32.h
 create mode 100644 include/asm-sh/system_64.h
 delete mode 100644 include/asm-sh64/system.h

(limited to 'include')

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 288abeb5476f..0cfa96aa5844 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -12,60 +12,9 @@
 #include <asm/types.h>
 #include <asm/ptrace.h>
 
-struct task_struct *__switch_to(struct task_struct *prev,
-				struct task_struct *next);
-
 #define AT_VECTOR_SIZE_ARCH 1 /* entries in ARCH_DLINFO */
-/*
- *	switch_to() should switch tasks to task nr n, first
- */
-
-#define switch_to(prev, next, last) do {				\
- struct task_struct *__last;						\
- register unsigned long *__ts1 __asm__ ("r1") = &prev->thread.sp;	\
- register unsigned long *__ts2 __asm__ ("r2") = &prev->thread.pc;	\
- register unsigned long *__ts4 __asm__ ("r4") = (unsigned long *)prev;	\
- register unsigned long *__ts5 __asm__ ("r5") = (unsigned long *)next;	\
- register unsigned long *__ts6 __asm__ ("r6") = &next->thread.sp;	\
- register unsigned long __ts7 __asm__ ("r7") = next->thread.pc;		\
- __asm__ __volatile__ (".balign 4\n\t" 					\
-		       "stc.l	gbr, @-r15\n\t" 			\
-		       "sts.l	pr, @-r15\n\t" 				\
-		       "mov.l	r8, @-r15\n\t" 				\
-		       "mov.l	r9, @-r15\n\t" 				\
-		       "mov.l	r10, @-r15\n\t" 			\
-		       "mov.l	r11, @-r15\n\t" 			\
-		       "mov.l	r12, @-r15\n\t" 			\
-		       "mov.l	r13, @-r15\n\t" 			\
-		       "mov.l	r14, @-r15\n\t" 			\
-		       "mov.l	r15, @r1	! save SP\n\t"		\
-		       "mov.l	@r6, r15	! change to new stack\n\t" \
-		       "mova	1f, %0\n\t" 				\
-		       "mov.l	%0, @r2		! save PC\n\t" 		\
-		       "mov.l	2f, %0\n\t" 				\
-		       "jmp	@%0		! call __switch_to\n\t" \
-		       " lds	r7, pr		!  with return to new PC\n\t" \
-		       ".balign	4\n"					\
-		       "2:\n\t"						\
-		       ".long	__switch_to\n"				\
-		       "1:\n\t"						\
-		       "mov.l	@r15+, r14\n\t"				\
-		       "mov.l	@r15+, r13\n\t"				\
-		       "mov.l	@r15+, r12\n\t"				\
-		       "mov.l	@r15+, r11\n\t"				\
-		       "mov.l	@r15+, r10\n\t"				\
-		       "mov.l	@r15+, r9\n\t"				\
-		       "mov.l	@r15+, r8\n\t"				\
-		       "lds.l	@r15+, pr\n\t"				\
-		       "ldc.l	@r15+, gbr\n\t"				\
-		       : "=z" (__last)					\
-		       : "r" (__ts1), "r" (__ts2), "r" (__ts4), 	\
-			 "r" (__ts5), "r" (__ts6), "r" (__ts7) 		\
-		       : "r3", "t");					\
-	last = __last;							\
-} while (0)
 
-#ifdef CONFIG_CPU_SH4A
+#if defined(CONFIG_CPU_SH4A) || defined(CONFIG_CPU_SH5)
 #define __icbi()			\
 {					\
 	unsigned long __addr;		\
@@ -91,7 +40,7 @@ struct task_struct *__switch_to(struct task_struct *prev,
  * Historically we have only done this type of barrier for the MMUCR, but
  * it's also necessary for the CCR, so we make it generic here instead.
  */
-#ifdef CONFIG_CPU_SH4A
+#if defined(CONFIG_CPU_SH4A) || defined(CONFIG_CPU_SH5)
 #define mb()		__asm__ __volatile__ ("synco": : :"memory")
 #define rmb()		mb()
 #define wmb()		__asm__ __volatile__ ("synco": : :"memory")
@@ -119,42 +68,6 @@ struct task_struct *__switch_to(struct task_struct *prev,
 
 #define set_mb(var, value) do { (void)xchg(&var, value); } while (0)
 
-/*
- * Jump to P2 area.
- * When handling TLB or caches, we need to do it from P2 area.
- */
-#define jump_to_P2()			\
-do {					\
-	unsigned long __dummy;		\
-	__asm__ __volatile__(		\
-		"mov.l	1f, %0\n\t"	\
-		"or	%1, %0\n\t"	\
-		"jmp	@%0\n\t"	\
-		" nop\n\t" 		\
-		".balign 4\n"		\
-		"1:	.long 2f\n"	\
-		"2:"			\
-		: "=&r" (__dummy)	\
-		: "r" (0x20000000));	\
-} while (0)
-
-/*
- * Back to P1 area.
- */
-#define back_to_P1()					\
-do {							\
-	unsigned long __dummy;				\
-	ctrl_barrier();					\
-	__asm__ __volatile__(				\
-		"mov.l	1f, %0\n\t"			\
-		"jmp	@%0\n\t"			\
-		" nop\n\t"				\
-		".balign 4\n"				\
-		"1:	.long 2f\n"			\
-		"2:"					\
-		: "=&r" (__dummy));			\
-} while (0)
-
 static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
 {
 	unsigned long flags, retval;
@@ -281,4 +194,10 @@ asmlinkage void bug_trap_handler(unsigned long r4, unsigned long r5,
 
 #define arch_align_stack(x) (x)
 
+#ifdef CONFIG_SUPERH32
+# include "system_32.h"
+#else
+# include "system_64.h"
+#endif
+
 #endif
diff --git a/include/asm-sh/system_32.h b/include/asm-sh/system_32.h
new file mode 100644
index 000000000000..ad37e8d5f31e
--- /dev/null
+++ b/include/asm-sh/system_32.h
@@ -0,0 +1,97 @@
+#ifndef __ASM_SH_SYSTEM_32_H
+#define __ASM_SH_SYSTEM_32_H
+
+#include <linux/types.h>
+
+struct task_struct *__switch_to(struct task_struct *prev,
+				struct task_struct *next);
+
+/*
+ *	switch_to() should switch tasks to task nr n, first
+ */
+#define switch_to(prev, next, last)				\
+do {								\
+	register u32 *__ts1 __asm__ ("r1") = &prev->thread.sp;	\
+	register u32 *__ts2 __asm__ ("r2") = &prev->thread.pc;	\
+	register u32 *__ts4 __asm__ ("r4") = (u32 *)prev;	\
+	register u32 *__ts5 __asm__ ("r5") = (u32 *)next;	\
+	register u32 *__ts6 __asm__ ("r6") = &next->thread.sp;	\
+	register u32 __ts7 __asm__ ("r7") = next->thread.pc;	\
+	struct task_struct *__last;				\
+								\
+	__asm__ __volatile__ (					\
+		".balign 4\n\t"					\
+		"stc.l	gbr, @-r15\n\t"				\
+		"sts.l	pr, @-r15\n\t"				\
+		"mov.l	r8, @-r15\n\t"				\
+		"mov.l	r9, @-r15\n\t"				\
+		"mov.l	r10, @-r15\n\t"				\
+		"mov.l	r11, @-r15\n\t"				\
+		"mov.l	r12, @-r15\n\t"				\
+		"mov.l	r13, @-r15\n\t"				\
+		"mov.l	r14, @-r15\n\t"				\
+		"mov.l	r15, @r1\t! save SP\n\t"		\
+		"mov.l	@r6, r15\t! change to new stack\n\t"	\
+		"mova	1f, %0\n\t"				\
+		"mov.l	%0, @r2\t! save PC\n\t"			\
+		"mov.l	2f, %0\n\t"				\
+		"jmp	@%0\t! call __switch_to\n\t"		\
+		" lds	r7, pr\t!  with return to new PC\n\t"	\
+		".balign	4\n"				\
+		"2:\n\t"					\
+		".long	__switch_to\n"				\
+		"1:\n\t"					\
+		"mov.l	@r15+, r14\n\t"				\
+		"mov.l	@r15+, r13\n\t"				\
+		"mov.l	@r15+, r12\n\t"				\
+		"mov.l	@r15+, r11\n\t"				\
+		"mov.l	@r15+, r10\n\t"				\
+		"mov.l	@r15+, r9\n\t"				\
+		"mov.l	@r15+, r8\n\t"				\
+		"lds.l	@r15+, pr\n\t"				\
+		"ldc.l	@r15+, gbr\n\t"				\
+		: "=z" (__last)					\
+		: "r" (__ts1), "r" (__ts2), "r" (__ts4),	\
+		  "r" (__ts5), "r" (__ts6), "r" (__ts7)		\
+		: "r3", "t");					\
+								\
+	last = __last;						\
+} while (0)
+
+/*
+ * Jump to P2 area.
+ * When handling TLB or caches, we need to do it from P2 area.
+ */
+#define jump_to_P2()			\
+do {					\
+	unsigned long __dummy;		\
+	__asm__ __volatile__(		\
+		"mov.l	1f, %0\n\t"	\
+		"or	%1, %0\n\t"	\
+		"jmp	@%0\n\t"	\
+		" nop\n\t"		\
+		".balign 4\n"		\
+		"1:	.long 2f\n"	\
+		"2:"			\
+		: "=&r" (__dummy)	\
+		: "r" (0x20000000));	\
+} while (0)
+
+/*
+ * Back to P1 area.
+ */
+#define back_to_P1()					\
+do {							\
+	unsigned long __dummy;				\
+	ctrl_barrier();					\
+	__asm__ __volatile__(				\
+		"mov.l	1f, %0\n\t"			\
+		"jmp	@%0\n\t"			\
+		" nop\n\t"				\
+		".balign 4\n"				\
+		"1:	.long 2f\n"			\
+		"2:"					\
+		: "=&r" (__dummy));			\
+} while (0)
+
+#endif /* __ASM_SH_SYSTEM_32_H */
diff --git a/include/asm-sh/system_64.h b/include/asm-sh/system_64.h
new file mode 100644
index 000000000000..0e466e991f7d
--- /dev/null
+++ b/include/asm-sh/system_64.h
@@ -0,0 +1,39 @@
+#ifndef __ASM_SH_SYSTEM_64_H
+#define __ASM_SH_SYSTEM_64_H
+
+/*
+ * include/asm-sh/system_64.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003  Paul Mundt
+ * Copyright (C) 2004  Richard Curnow
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <asm/processor.h>
+
+/*
+ *	switch_to() should switch tasks to task nr n, first
+ */
+struct task_struct *sh64_switch_to(struct task_struct *prev,
+				   struct thread_struct *prev_thread,
+				   struct task_struct *next,
+				   struct thread_struct *next_thread);
+
+#define switch_to(prev,next,last)				\
+do {								\
+	if (last_task_used_math != next) {			\
+		struct pt_regs *regs = next->thread.uregs;	\
+		if (regs) regs->sr |= SR_FD;			\
+	}							\
+	last = sh64_switch_to(prev, &prev->thread, next,	\
+			      &next->thread);			\
+} while (0)
+
+/* No segmentation.. */
+#define jump_to_P2()	do { } while (0)
+#define back_to_P1()	do { } while (0)
+
+#endif /* __ASM_SH_SYSTEM_64_H */
diff --git a/include/asm-sh64/system.h b/include/asm-sh64/system.h
deleted file mode 100644
index be2a15ffcc55..000000000000
--- a/include/asm-sh64/system.h
+++ /dev/null
@@ -1,190 +0,0 @@
-#ifndef __ASM_SH64_SYSTEM_H
-#define __ASM_SH64_SYSTEM_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/system.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- * Copyright (C) 2004  Richard Curnow
- *
- */
-
-#include <asm/registers.h>
-#include <asm/processor.h>
-
-/*
- *	switch_to() should switch tasks to task nr n, first
- */
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-extern struct task_struct *sh64_switch_to(struct task_struct *prev,
-					  struct thread_struct *prev_thread,
-					  struct task_struct *next,
-					  struct thread_struct *next_thread);
-
-#define switch_to(prev,next,last) \
-	do {\
-		if (last_task_used_math != next) {\
-			struct pt_regs *regs = next->thread.uregs;\
-			if (regs) regs->sr |= SR_FD;\
-		}\
-		last = sh64_switch_to(prev, &prev->thread, next, &next->thread);\
-	} while(0)
-
-#define nop() __asm__ __volatile__ ("nop")
-
-#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
-
-extern void __xchg_called_with_bad_pointer(void);
-
-#define mb()	__asm__ __volatile__ ("synco": : :"memory")
-#define rmb()	mb()
-#define wmb()	__asm__ __volatile__ ("synco": : :"memory")
-#define read_barrier_depends()	do { } while (0)
-
-#ifdef CONFIG_SMP
-#define smp_mb()	mb()
-#define smp_rmb()	rmb()
-#define smp_wmb()	wmb()
-#define smp_read_barrier_depends()	read_barrier_depends()
-#else
-#define smp_mb()	barrier()
-#define smp_rmb()	barrier()
-#define smp_wmb()	barrier()
-#define smp_read_barrier_depends()	do { } while (0)
-#endif /* CONFIG_SMP */
-
-#define set_mb(var, value) do { (void)xchg(&var, value); } while (0)
-
-/* Interrupt Control */
-#ifndef HARD_CLI
-#define SR_MASK_L 0x000000f0L
-#define SR_MASK_LL 0x00000000000000f0LL
-#else
-#define SR_MASK_L 0x10000000L
-#define SR_MASK_LL 0x0000000010000000LL
-#endif
-
-static __inline__ void local_irq_enable(void)
-{
-	/* cli/sti based on SR.BL */
-	unsigned long long __dummy0, __dummy1=~SR_MASK_LL;
-
-	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
-			     "and	%0, %1, %0\n\t"
-			     "putcon	%0, " __SR "\n\t"
-			     : "=&r" (__dummy0)
-			     : "r" (__dummy1));
-}
-
-static __inline__ void local_irq_disable(void)
-{
-	/* cli/sti based on SR.BL */
-	unsigned long long __dummy0, __dummy1=SR_MASK_LL;
-	__asm__ __volatile__("getcon	" __SR ", %0\n\t"
-			     "or	%0, %1, %0\n\t"
-			     "putcon	%0, " __SR "\n\t"
-			     : "=&r" (__dummy0)
-			     : "r" (__dummy1));
-}
-
-#define local_save_flags(x) 						\
-(__extension__ ({	unsigned long long __dummy=SR_MASK_LL;		\
-	__asm__ __volatile__(						\
-		"getcon	" __SR ", %0\n\t"				\
-		"and	%0, %1, %0"					\
-		: "=&r" (x)						\
-		: "r" (__dummy));}))
-
-#define local_irq_save(x)						\
-(__extension__ ({	unsigned long long __d2=SR_MASK_LL, __d1;	\
-	__asm__ __volatile__(          	         			\
-		"getcon	" __SR ", %1\n\t" 				\
-		"or	%1, r63, %0\n\t"				\
-		"or	%1, %2, %1\n\t"					\
-		"putcon	%1, " __SR "\n\t"    				\
-		"and	%0, %2, %0"    					\
-		: "=&r" (x), "=&r" (__d1)				\
-		: "r" (__d2));}));
-
-#define local_irq_restore(x) do { 					\
-	if ( ((x) & SR_MASK_L) == 0 )		/* dropping to 0 ? */	\
-		local_irq_enable();		/* yes...re-enable */	\
-} while (0)
-
-#define irqs_disabled()			\
-({					\
-	unsigned long flags;		\
-	local_save_flags(flags);	\
-	(flags != 0);			\
-})
-
-static inline unsigned long xchg_u32(volatile int * m, unsigned long val)
-{
-	unsigned long flags, retval;
-
-	local_irq_save(flags);
-	retval = *m;
-	*m = val;
-	local_irq_restore(flags);
-	return retval;
-}
-
-static inline unsigned long xchg_u8(volatile unsigned char * m, unsigned long val)
-{
-	unsigned long flags, retval;
-
-	local_irq_save(flags);
-	retval = *m;
-	*m = val & 0xff;
-	local_irq_restore(flags);
-	return retval;
-}
-
-static __inline__ unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
-{
-	switch (size) {
-	case 4:
-		return xchg_u32(ptr, x);
-		break;
-	case 1:
-		return xchg_u8(ptr, x);
-		break;
-	}
-	__xchg_called_with_bad_pointer();
-	return x;
-}
-
-/* XXX
- * disable hlt during certain critical i/o operations
- */
-#define HAVE_DISABLE_HLT
-void disable_hlt(void);
-void enable_hlt(void);
-
-
-#define smp_mb()        barrier()
-#define smp_rmb()       barrier()
-#define smp_wmb()       barrier()
-
-#ifdef CONFIG_SH_ALPHANUMERIC
-/* This is only used for debugging. */
-extern void print_seg(char *file,int line);
-#define PLS() print_seg(__FILE__,__LINE__)
-#else	/* CONFIG_SH_ALPHANUMERIC */
-#define PLS()
-#endif	/* CONFIG_SH_ALPHANUMERIC */
-
-#define PL() printk("@ <%s,%s:%d>\n",__FILE__,__FUNCTION__,__LINE__)
-
-#define arch_align_stack(x) (x)
-
-#endif /* __ASM_SH64_SYSTEM_H */
-- 
cgit v1.2.3


From c0acca6789281650134cfbbe00fc461e39440446 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 19:54:16 +0900
Subject: sh: SH-5 also uses the ASID cache.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/processor_64.h | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
index 5430e437b915..6ad23387d7ba 100644
--- a/include/asm-sh/processor_64.h
+++ b/include/asm-sh/processor_64.h
@@ -54,18 +54,14 @@ struct tlb_info {
 struct sh_cpuinfo {
 	enum cpu_type type;
 	unsigned long loops_per_jiffy;
+	unsigned long asid_cache;
 
-	char	hard_math;
-
-	unsigned long *pgd_quick;
-	unsigned long *pmd_quick;
-	unsigned long *pte_quick;
-	unsigned long pgtable_cache_sz;
 	unsigned int cpu_clock, master_clock, bus_clock, module_clock;
 
 	/* Cache info */
 	struct cache_info icache;
 	struct cache_info dcache;
+	struct cache_info scache;
 
 	/* TLB info */
 	struct tlb_info itlb;
-- 
cgit v1.2.3


From 9b01bd9ee6408846c0553c03fb4b864353a845c9 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 19:55:50 +0900
Subject: sh: Split out uaccess.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/uaccess.h    | 564 +------------------------------------------
 include/asm-sh/uaccess_32.h | 575 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/uaccess_64.h | 316 ++++++++++++++++++++++++
 include/asm-sh64/uaccess.h  | 316 ------------------------
 4 files changed, 894 insertions(+), 877 deletions(-)
 create mode 100644 include/asm-sh/uaccess_32.h
 create mode 100644 include/asm-sh/uaccess_64.h
 delete mode 100644 include/asm-sh64/uaccess.h

(limited to 'include')

diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 77c391fa93d6..ff24ce95b238 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -1,563 +1,5 @@
-/* $Id: uaccess.h,v 1.11 2003/10/13 07:21:20 lethal Exp $
- *
- * User space memory access functions
- *
- * Copyright (C) 1999, 2002  Niibe Yutaka
- * Copyright (C) 2003  Paul Mundt
- *
- *  Based on:
- *     MIPS implementation version 1.15 by
- *              Copyright (C) 1996, 1997, 1998 by Ralf Baechle
- *     and i386 version.
- */
-#ifndef __ASM_SH_UACCESS_H
-#define __ASM_SH_UACCESS_H
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-
-#define VERIFY_READ    0
-#define VERIFY_WRITE   1
-
-/*
- * The fs value determines whether argument validity checking should be
- * performed or not.  If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons (Data Segment Register?), these macros are misnamed.
- */
-
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-
-#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFFUL)
-#define USER_DS		MAKE_MM_SEG(PAGE_OFFSET)
-
-#define segment_eq(a,b)	((a).seg == (b).seg)
-
-#define get_ds()	(KERNEL_DS)
-
-#if !defined(CONFIG_MMU)
-/* NOMMU is always true */
-#define __addr_ok(addr) (1)
-
-static inline mm_segment_t get_fs(void)
-{
-	return USER_DS;
-}
-
-static inline void set_fs(mm_segment_t s)
-{
-}
-
-/*
- * __access_ok: Check if address with size is OK or not.
- *
- * If we don't have an MMU (or if its disabled) the only thing we really have
- * to look out for is if the address resides somewhere outside of what
- * available RAM we have.
- *
- * TODO: This check could probably also stand to be restricted somewhat more..
- * though it still does the Right Thing(tm) for the time being.
- */
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	return ((addr >= memory_start) && ((addr + size) < memory_end));
-}
-#else /* CONFIG_MMU */
-#define __addr_ok(addr) \
-	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
-
-#define get_fs()	(current_thread_info()->addr_limit)
-#define set_fs(x)	(current_thread_info()->addr_limit = (x))
-
-/*
- * __access_ok: Check if address with size is OK or not.
- *
- * Uhhuh, this needs 33-bit arithmetic. We have a carry..
- *
- * sum := addr + size;  carry? --> flag = true;
- * if (sum >= addr_limit) flag = true;
- */
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	unsigned long flag, sum;
-
-	__asm__("clrt\n\t"
-		"addc	%3, %1\n\t"
-		"movt	%0\n\t"
-		"cmp/hi	%4, %1\n\t"
-		"rotcl	%0"
-		:"=&r" (flag), "=r" (sum)
-		:"1" (addr), "r" (size),
-		 "r" (current_thread_info()->addr_limit.seg)
-		:"t");
-	return flag == 0;
-
-}
-#endif /* CONFIG_MMU */
-
-static inline int access_ok(int type, const void __user *p, unsigned long size)
-{
-	unsigned long addr = (unsigned long)p;
-	return __access_ok(addr, size);
-}
-
-/*
- * Uh, these should become the main single-value transfer routines ...
- * They automatically use the right size if we just have the right
- * pointer type ...
- *
- * As SuperH uses the same address space for kernel and user data, we
- * can just do these as direct assignments.
- *
- * Careful to not
- * (a) re-use the arguments for side effects (sizeof is ok)
- * (b) require any knowledge of processes at this stage
- */
-#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
-#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
-
-/*
- * The "__xxx" versions do not do address space checking, useful when
- * doing multiple accesses to the same area (the user has to do the
- * checks by hand with "access_ok()")
- */
-#define __put_user(x,ptr) \
-  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
-#define __get_user(x,ptr) \
-  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
-
-struct __large_struct { unsigned long buf[100]; };
-#define __m(x) (*(struct __large_struct __user *)(x))
-
-#define __get_user_size(x,ptr,size,retval)			\
-do {								\
-	retval = 0;						\
-	__chk_user_ptr(ptr);					\
-	switch (size) {						\
-	case 1:							\
-		__get_user_asm(x, ptr, retval, "b");		\
-		break;						\
-	case 2:							\
-		__get_user_asm(x, ptr, retval, "w");		\
-		break;						\
-	case 4:							\
-		__get_user_asm(x, ptr, retval, "l");		\
-		break;						\
-	default:						\
-		__get_user_unknown();				\
-		break;						\
-	}							\
-} while (0)
-
-#define __get_user_nocheck(x,ptr,size)				\
-({								\
-	long __gu_err, __gu_val;				\
-	__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
-	__gu_err;						\
-})
-
-#ifdef CONFIG_MMU
-#define __get_user_check(x,ptr,size)				\
-({								\
-	long __gu_err, __gu_val;				\
-	__chk_user_ptr(ptr);					\
-	switch (size) {						\
-	case 1:							\
-		__get_user_1(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	case 2:							\
-		__get_user_2(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	case 4:							\
-		__get_user_4(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	default:						\
-		__get_user_unknown();				\
-		break;						\
-	}							\
-								\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
-	__gu_err;						\
-})
-
-#define __get_user_1(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.b	@%2, %1\n\t"			\
-	"extu.b	%1, %1\n"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-
-#define __get_user_2(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.w	@%2, %1\n\t"			\
-	"extu.w	%1, %1\n"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-
-#define __get_user_4(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.l	@%2, %1\n\t"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-#else /* CONFIG_MMU */
-#define __get_user_check(x,ptr,size)					\
-({									\
-	long __gu_err, __gu_val;					\
-	if (__access_ok((unsigned long)(ptr), (size))) {		\
-		__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
-		(x) = (__typeof__(*(ptr)))__gu_val;			\
-	} else								\
-		__gu_err = -EFAULT;					\
-	__gu_err;							\
-})
-#endif
-
-#define __get_user_asm(x, addr, err, insn) \
-({ \
-__asm__ __volatile__( \
-	"1:\n\t" \
-	"mov." insn "	%2, %1\n\t" \
-	"mov	#0, %0\n" \
-	"2:\n" \
-	".section	.fixup,\"ax\"\n" \
-	"3:\n\t" \
-	"mov	#0, %1\n\t" \
-	"mov.l	4f, %0\n\t" \
-	"jmp	@%0\n\t" \
-	" mov	%3, %0\n" \
-	"4:	.long	2b\n\t" \
-	".previous\n" \
-	".section	__ex_table,\"a\"\n\t" \
-	".long	1b, 3b\n\t" \
-	".previous" \
-	:"=&r" (err), "=&r" (x) \
-	:"m" (__m(addr)), "i" (-EFAULT)); })
-
-extern void __get_user_unknown(void);
-
-#define __put_user_size(x,ptr,size,retval)		\
-do {							\
-	retval = 0;					\
-	__chk_user_ptr(ptr);				\
-	switch (size) {					\
-	case 1:						\
-		__put_user_asm(x, ptr, retval, "b");	\
-		break;					\
-	case 2:						\
-		__put_user_asm(x, ptr, retval, "w");	\
-		break;					\
-	case 4:						\
-		__put_user_asm(x, ptr, retval, "l");	\
-		break;					\
-	case 8:						\
-		__put_user_u64(x, ptr, retval);		\
-		break;					\
-	default:					\
-		__put_user_unknown();			\
-	}						\
-} while (0)
-
-#define __put_user_nocheck(x,ptr,size)			\
-({							\
-	long __pu_err;					\
-	__put_user_size((x),(ptr),(size),__pu_err);	\
-	__pu_err;					\
-})
-
-#define __put_user_check(x,ptr,size)				\
-({								\
-	long __pu_err = -EFAULT;				\
-	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
-								\
-	if (__access_ok((unsigned long)__pu_addr,size))		\
-		__put_user_size((x),__pu_addr,(size),__pu_err);	\
-	__pu_err;						\
-})
-
-#define __put_user_asm(x, addr, err, insn) \
-({ \
-__asm__ __volatile__( \
-	"1:\n\t" \
-	"mov." insn "	%1, %2\n\t" \
-	"mov	#0, %0\n" \
-	"2:\n" \
-	".section	.fixup,\"ax\"\n" \
-	"3:\n\t" \
-	"nop\n\t" \
-	"mov.l	4f, %0\n\t" \
-	"jmp	@%0\n\t" \
-	"mov	%3, %0\n" \
-	"4:	.long	2b\n\t" \
-	".previous\n" \
-	".section	__ex_table,\"a\"\n\t" \
-	".long	1b, 3b\n\t" \
-	".previous" \
-	:"=&r" (err) \
-	:"r" (x), "m" (__m(addr)), "i" (-EFAULT) \
-        :"memory"); })
-
-#if defined(__LITTLE_ENDIAN__)
-#define __put_user_u64(val,addr,retval) \
-({ \
-__asm__ __volatile__( \
-	"1:\n\t" \
-	"mov.l	%R1,%2\n\t" \
-	"mov.l	%S1,%T2\n\t" \
-	"mov	#0,%0\n" \
-	"2:\n" \
-	".section	.fixup,\"ax\"\n" \
-	"3:\n\t" \
-	"nop\n\t" \
-	"mov.l	4f,%0\n\t" \
-	"jmp	@%0\n\t" \
-	" mov	%3,%0\n" \
-	"4:	.long	2b\n\t" \
-	".previous\n" \
-	".section	__ex_table,\"a\"\n\t" \
-	".long	1b, 3b\n\t" \
-	".previous" \
-	: "=r" (retval) \
-	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
-        : "memory"); })
+#ifdef CONFIG_SUPERH32
+# include "uaccess_32.h"
 #else
-#define __put_user_u64(val,addr,retval) \
-({ \
-__asm__ __volatile__( \
-	"1:\n\t" \
-	"mov.l	%S1,%2\n\t" \
-	"mov.l	%R1,%T2\n\t" \
-	"mov	#0,%0\n" \
-	"2:\n" \
-	".section	.fixup,\"ax\"\n" \
-	"3:\n\t" \
-	"nop\n\t" \
-	"mov.l	4f,%0\n\t" \
-	"jmp	@%0\n\t" \
-	" mov	%3,%0\n" \
-	"4:	.long	2b\n\t" \
-	".previous\n" \
-	".section	__ex_table,\"a\"\n\t" \
-	".long	1b, 3b\n\t" \
-	".previous" \
-	: "=r" (retval) \
-	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
-        : "memory"); })
+# include "uaccess_64.h"
 #endif
-
-extern void __put_user_unknown(void);
-
-/* Generic arbitrary sized copy.  */
-/* Return the number of bytes NOT copied */
-__kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n);
-
-#define copy_to_user(to,from,n) ({ \
-void *__copy_to = (void *) (to); \
-__kernel_size_t __copy_size = (__kernel_size_t) (n); \
-__kernel_size_t __copy_res; \
-if(__copy_size && __access_ok((unsigned long)__copy_to, __copy_size)) { \
-__copy_res = __copy_user(__copy_to, (void *) (from), __copy_size); \
-} else __copy_res = __copy_size; \
-__copy_res; })
-
-#define copy_from_user(to,from,n) ({ \
-void *__copy_to = (void *) (to); \
-void *__copy_from = (void *) (from); \
-__kernel_size_t __copy_size = (__kernel_size_t) (n); \
-__kernel_size_t __copy_res; \
-if(__copy_size && __access_ok((unsigned long)__copy_from, __copy_size)) { \
-__copy_res = __copy_user(__copy_to, __copy_from, __copy_size); \
-} else __copy_res = __copy_size; \
-__copy_res; })
-
-static __always_inline unsigned long
-__copy_from_user(void *to, const void __user *from, unsigned long n)
-{
-	return __copy_user(to, (__force void *)from, n);
-}
-
-static __always_inline unsigned long __must_check
-__copy_to_user(void __user *to, const void *from, unsigned long n)
-{
-	return __copy_user((__force void *)to, from, n);
-}
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-/*
- * Clear the area and return remaining number of bytes
- * (on failure.  Usually it's 0.)
- */
-extern __kernel_size_t __clear_user(void *addr, __kernel_size_t size);
-
-#define clear_user(addr,n) ({ \
-void * __cl_addr = (addr); \
-unsigned long __cl_size = (n); \
-if (__cl_size && __access_ok(((unsigned long)(__cl_addr)), __cl_size)) \
-__cl_size = __clear_user(__cl_addr, __cl_size); \
-__cl_size; })
-
-static __inline__ int
-__strncpy_from_user(unsigned long __dest, unsigned long __user __src, int __count)
-{
-	__kernel_size_t res;
-	unsigned long __dummy, _d, _s;
-
-	__asm__ __volatile__(
-		"9:\n"
-		"mov.b	@%2+, %1\n\t"
-		"cmp/eq	#0, %1\n\t"
-		"bt/s	2f\n"
-		"1:\n"
-		"mov.b	%1, @%3\n\t"
-		"dt	%7\n\t"
-		"bf/s	9b\n\t"
-		" add	#1, %3\n\t"
-		"2:\n\t"
-		"sub	%7, %0\n"
-		"3:\n"
-		".section .fixup,\"ax\"\n"
-		"4:\n\t"
-		"mov.l	5f, %1\n\t"
-		"jmp	@%1\n\t"
-		" mov	%8, %0\n\t"
-		".balign 4\n"
-		"5:	.long 3b\n"
-		".previous\n"
-		".section __ex_table,\"a\"\n"
-		"	.balign 4\n"
-		"	.long 9b,4b\n"
-		".previous"
-		: "=r" (res), "=&z" (__dummy), "=r" (_s), "=r" (_d)
-		: "0" (__count), "2" (__src), "3" (__dest), "r" (__count),
-		  "i" (-EFAULT)
-		: "memory", "t");
-
-	return res;
-}
-
-#define strncpy_from_user(dest,src,count) ({ \
-unsigned long __sfu_src = (unsigned long) (src); \
-int __sfu_count = (int) (count); \
-long __sfu_res = -EFAULT; \
-if(__access_ok(__sfu_src, __sfu_count)) { \
-__sfu_res = __strncpy_from_user((unsigned long) (dest), __sfu_src, __sfu_count); \
-} __sfu_res; })
-
-/*
- * Return the size of a string (including the ending 0!)
- */
-static __inline__ long __strnlen_user(const char __user *__s, long __n)
-{
-	unsigned long res;
-	unsigned long __dummy;
-
-	__asm__ __volatile__(
-		"9:\n"
-		"cmp/eq	%4, %0\n\t"
-		"bt	2f\n"
-		"1:\t"
-		"mov.b	@(%0,%3), %1\n\t"
-		"tst	%1, %1\n\t"
-		"bf/s	9b\n\t"
-		" add	#1, %0\n"
-		"2:\n"
-		".section .fixup,\"ax\"\n"
-		"3:\n\t"
-		"mov.l	4f, %1\n\t"
-		"jmp	@%1\n\t"
-		" mov	#0, %0\n"
-		".balign 4\n"
-		"4:	.long 2b\n"
-		".previous\n"
-		".section __ex_table,\"a\"\n"
-		"	.balign 4\n"
-		"	.long 1b,3b\n"
-		".previous"
-		: "=z" (res), "=&r" (__dummy)
-		: "0" (0), "r" (__s), "r" (__n)
-		: "t");
-	return res;
-}
-
-static __inline__ long strnlen_user(const char __user *s, long n)
-{
-	if (!__addr_ok(s))
-		return 0;
-	else
-		return __strnlen_user(s, n);
-}
-
-#define strlen_user(str)	strnlen_user(str, ~0UL >> 1)
-
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-extern int fixup_exception(struct pt_regs *regs);
-
-#endif /* __ASM_SH_UACCESS_H */
diff --git a/include/asm-sh/uaccess_32.h b/include/asm-sh/uaccess_32.h
new file mode 100644
index 000000000000..f18a1a5c95c0
--- /dev/null
+++ b/include/asm-sh/uaccess_32.h
@@ -0,0 +1,575 @@
+/* $Id: uaccess.h,v 1.11 2003/10/13 07:21:20 lethal Exp $
+ *
+ * User space memory access functions
+ *
+ * Copyright (C) 1999, 2002  Niibe Yutaka
+ * Copyright (C) 2003  Paul Mundt
+ *
+ *  Based on:
+ *     MIPS implementation version 1.15 by
+ *              Copyright (C) 1996, 1997, 1998 by Ralf Baechle
+ *     and i386 version.
+ */
+#ifndef __ASM_SH_UACCESS_H
+#define __ASM_SH_UACCESS_H
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+
+#define VERIFY_READ    0
+#define VERIFY_WRITE   1
+
+/*
+ * The fs value determines whether argument validity checking should be
+ * performed or not.  If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are misnamed.
+ */
+
+#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
+
+#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFFUL)
+#define USER_DS		MAKE_MM_SEG(PAGE_OFFSET)
+
+#define segment_eq(a,b)	((a).seg == (b).seg)
+
+#define get_ds()	(KERNEL_DS)
+
+#if !defined(CONFIG_MMU)
+/* NOMMU is always true */
+#define __addr_ok(addr) (1)
+
+static inline mm_segment_t get_fs(void)
+{
+	return USER_DS;
+}
+
+static inline void set_fs(mm_segment_t s)
+{
+}
+
+/*
+ * __access_ok: Check if address with size is OK or not.
+ *
+ * If we don't have an MMU (or if its disabled) the only thing we really have
+ * to look out for is if the address resides somewhere outside of what
+ * available RAM we have.
+ *
+ * TODO: This check could probably also stand to be restricted somewhat more..
+ * though it still does the Right Thing(tm) for the time being.
+ */
+static inline int __access_ok(unsigned long addr, unsigned long size)
+{
+	return ((addr >= memory_start) && ((addr + size) < memory_end));
+}
+#else /* CONFIG_MMU */
+#define __addr_ok(addr) \
+	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
+
+#define get_fs()	(current_thread_info()->addr_limit)
+#define set_fs(x)	(current_thread_info()->addr_limit = (x))
+
+/*
+ * __access_ok: Check if address with size is OK or not.
+ *
+ * We do three checks:
+ * (1) is it user space?
+ * (2) addr + size --> carry?
+ * (3) addr + size >= 0x80000000  (PAGE_OFFSET)
+ *
+ * (1) (2) (3) | RESULT
+ *  0   0   0  |  ok
+ *  0   0   1  |  ok
+ *  0   1   0  |  bad
+ *  0   1   1  |  bad
+ *  1   0   0  |  ok
+ *  1   0   1  |  bad
+ *  1   1   0  |  bad
+ *  1   1   1  |  bad
+ */
+static inline int __access_ok(unsigned long addr, unsigned long size)
+{
+	unsigned long flag, tmp;
+
+	__asm__("stc	r7_bank, %0\n\t"
+		"mov.l	@(8,%0), %0\n\t"
+		"clrt\n\t"
+		"addc	%2, %1\n\t"
+		"and	%1, %0\n\t"
+		"rotcl	%0\n\t"
+		"rotcl	%0\n\t"
+		"and	#3, %0"
+		: "=&z" (flag), "=r" (tmp)
+		: "r" (addr), "1" (size)
+		: "t");
+
+	return flag == 0;
+}
+#endif /* CONFIG_MMU */
+
+static inline int access_ok(int type, const void __user *p, unsigned long size)
+{
+	unsigned long addr = (unsigned long)p;
+	return __access_ok(addr, size);
+}
+
+/*
+ * Uh, these should become the main single-value transfer routines ...
+ * They automatically use the right size if we just have the right
+ * pointer type ...
+ *
+ * As SuperH uses the same address space for kernel and user data, we
+ * can just do these as direct assignments.
+ *
+ * Careful to not
+ * (a) re-use the arguments for side effects (sizeof is ok)
+ * (b) require any knowledge of processes at this stage
+ */
+#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
+#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * The "__xxx" versions do not do address space checking, useful when
+ * doing multiple accesses to the same area (the user has to do the
+ * checks by hand with "access_ok()")
+ */
+#define __put_user(x,ptr) \
+  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+#define __get_user(x,ptr) \
+  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+struct __large_struct { unsigned long buf[100]; };
+#define __m(x) (*(struct __large_struct __user *)(x))
+
+#define __get_user_size(x,ptr,size,retval)			\
+do {								\
+	retval = 0;						\
+	__chk_user_ptr(ptr);					\
+	switch (size) {						\
+	case 1:							\
+		__get_user_asm(x, ptr, retval, "b");		\
+		break;						\
+	case 2:							\
+		__get_user_asm(x, ptr, retval, "w");		\
+		break;						\
+	case 4:							\
+		__get_user_asm(x, ptr, retval, "l");		\
+		break;						\
+	default:						\
+		__get_user_unknown();				\
+		break;						\
+	}							\
+} while (0)
+
+#define __get_user_nocheck(x,ptr,size)				\
+({								\
+	long __gu_err, __gu_val;				\
+	__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
+	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	__gu_err;						\
+})
+
+#ifdef CONFIG_MMU
+#define __get_user_check(x,ptr,size)				\
+({								\
+	long __gu_err, __gu_val;				\
+	__chk_user_ptr(ptr);					\
+	switch (size) {						\
+	case 1:							\
+		__get_user_1(__gu_val, (ptr), __gu_err);	\
+		break;						\
+	case 2:							\
+		__get_user_2(__gu_val, (ptr), __gu_err);	\
+		break;						\
+	case 4:							\
+		__get_user_4(__gu_val, (ptr), __gu_err);	\
+		break;						\
+	default:						\
+		__get_user_unknown();				\
+		break;						\
+	}							\
+								\
+	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	__gu_err;						\
+})
+
+#define __get_user_1(x,addr,err) ({		\
+__asm__("stc	r7_bank, %1\n\t"		\
+	"mov.l	@(8,%1), %1\n\t"		\
+	"and	%2, %1\n\t"			\
+	"cmp/pz	%1\n\t"				\
+	"bt/s	1f\n\t"				\
+	" mov	#0, %0\n\t"			\
+	"0:\n"					\
+	"mov	#-14, %0\n\t"			\
+	"bra	2f\n\t"				\
+	" mov	#0, %1\n"			\
+	"1:\n\t"				\
+	"mov.b	@%2, %1\n\t"			\
+	"extu.b	%1, %1\n"			\
+	"2:\n"					\
+	".section	__ex_table,\"a\"\n\t"	\
+	".long	1b, 0b\n\t"			\
+	".previous"				\
+	: "=&r" (err), "=&r" (x)		\
+	: "r" (addr)				\
+	: "t");					\
+})
+
+#define __get_user_2(x,addr,err) ({		\
+__asm__("stc	r7_bank, %1\n\t"		\
+	"mov.l	@(8,%1), %1\n\t"		\
+	"and	%2, %1\n\t"			\
+	"cmp/pz	%1\n\t"				\
+	"bt/s	1f\n\t"				\
+	" mov	#0, %0\n\t"			\
+	"0:\n"					\
+	"mov	#-14, %0\n\t"			\
+	"bra	2f\n\t"				\
+	" mov	#0, %1\n"			\
+	"1:\n\t"				\
+	"mov.w	@%2, %1\n\t"			\
+	"extu.w	%1, %1\n"			\
+	"2:\n"					\
+	".section	__ex_table,\"a\"\n\t"	\
+	".long	1b, 0b\n\t"			\
+	".previous"				\
+	: "=&r" (err), "=&r" (x)		\
+	: "r" (addr)				\
+	: "t");					\
+})
+
+#define __get_user_4(x,addr,err) ({		\
+__asm__("stc	r7_bank, %1\n\t"		\
+	"mov.l	@(8,%1), %1\n\t"		\
+	"and	%2, %1\n\t"			\
+	"cmp/pz	%1\n\t"				\
+	"bt/s	1f\n\t"				\
+	" mov	#0, %0\n\t"			\
+	"0:\n"					\
+	"mov	#-14, %0\n\t"			\
+	"bra	2f\n\t"				\
+	" mov	#0, %1\n"			\
+	"1:\n\t"				\
+	"mov.l	@%2, %1\n\t"			\
+	"2:\n"					\
+	".section	__ex_table,\"a\"\n\t"	\
+	".long	1b, 0b\n\t"			\
+	".previous"				\
+	: "=&r" (err), "=&r" (x)		\
+	: "r" (addr)				\
+	: "t");					\
+})
+#else /* CONFIG_MMU */
+#define __get_user_check(x,ptr,size)					\
+({									\
+	long __gu_err, __gu_val;					\
+	if (__access_ok((unsigned long)(ptr), (size))) {		\
+		__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
+		(x) = (__typeof__(*(ptr)))__gu_val;			\
+	} else								\
+		__gu_err = -EFAULT;					\
+	__gu_err;							\
+})
+#endif
+
+#define __get_user_asm(x, addr, err, insn) \
+({ \
+__asm__ __volatile__( \
+	"1:\n\t" \
+	"mov." insn "	%2, %1\n\t" \
+	"mov	#0, %0\n" \
+	"2:\n" \
+	".section	.fixup,\"ax\"\n" \
+	"3:\n\t" \
+	"mov	#0, %1\n\t" \
+	"mov.l	4f, %0\n\t" \
+	"jmp	@%0\n\t" \
+	" mov	%3, %0\n" \
+	"4:	.long	2b\n\t" \
+	".previous\n" \
+	".section	__ex_table,\"a\"\n\t" \
+	".long	1b, 3b\n\t" \
+	".previous" \
+	:"=&r" (err), "=&r" (x) \
+	:"m" (__m(addr)), "i" (-EFAULT)); })
+
+extern void __get_user_unknown(void);
+
+#define __put_user_size(x,ptr,size,retval)		\
+do {							\
+	retval = 0;					\
+	__chk_user_ptr(ptr);				\
+	switch (size) {					\
+	case 1:						\
+		__put_user_asm(x, ptr, retval, "b");	\
+		break;					\
+	case 2:						\
+		__put_user_asm(x, ptr, retval, "w");	\
+		break;					\
+	case 4:						\
+		__put_user_asm(x, ptr, retval, "l");	\
+		break;					\
+	case 8:						\
+		__put_user_u64(x, ptr, retval);		\
+		break;					\
+	default:					\
+		__put_user_unknown();			\
+	}						\
+} while (0)
+
+#define __put_user_nocheck(x,ptr,size)			\
+({							\
+	long __pu_err;					\
+	__put_user_size((x),(ptr),(size),__pu_err);	\
+	__pu_err;					\
+})
+
+#define __put_user_check(x,ptr,size)				\
+({								\
+	long __pu_err = -EFAULT;				\
+	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
+								\
+	if (__access_ok((unsigned long)__pu_addr,size))		\
+		__put_user_size((x),__pu_addr,(size),__pu_err);	\
+	__pu_err;						\
+})
+
+#define __put_user_asm(x, addr, err, insn) \
+({ \
+__asm__ __volatile__( \
+	"1:\n\t" \
+	"mov." insn "	%1, %2\n\t" \
+	"mov	#0, %0\n" \
+	"2:\n" \
+	".section	.fixup,\"ax\"\n" \
+	"3:\n\t" \
+	"nop\n\t" \
+	"mov.l	4f, %0\n\t" \
+	"jmp	@%0\n\t" \
+	"mov	%3, %0\n" \
+	"4:	.long	2b\n\t" \
+	".previous\n" \
+	".section	__ex_table,\"a\"\n\t" \
+	".long	1b, 3b\n\t" \
+	".previous" \
+	:"=&r" (err) \
+	:"r" (x), "m" (__m(addr)), "i" (-EFAULT) \
+        :"memory"); })
+
+#if defined(__LITTLE_ENDIAN__)
+#define __put_user_u64(val,addr,retval) \
+({ \
+__asm__ __volatile__( \
+	"1:\n\t" \
+	"mov.l	%R1,%2\n\t" \
+	"mov.l	%S1,%T2\n\t" \
+	"mov	#0,%0\n" \
+	"2:\n" \
+	".section	.fixup,\"ax\"\n" \
+	"3:\n\t" \
+	"nop\n\t" \
+	"mov.l	4f,%0\n\t" \
+	"jmp	@%0\n\t" \
+	" mov	%3,%0\n" \
+	"4:	.long	2b\n\t" \
+	".previous\n" \
+	".section	__ex_table,\"a\"\n\t" \
+	".long	1b, 3b\n\t" \
+	".previous" \
+	: "=r" (retval) \
+	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
+        : "memory"); })
+#else
+#define __put_user_u64(val,addr,retval) \
+({ \
+__asm__ __volatile__( \
+	"1:\n\t" \
+	"mov.l	%S1,%2\n\t" \
+	"mov.l	%R1,%T2\n\t" \
+	"mov	#0,%0\n" \
+	"2:\n" \
+	".section	.fixup,\"ax\"\n" \
+	"3:\n\t" \
+	"nop\n\t" \
+	"mov.l	4f,%0\n\t" \
+	"jmp	@%0\n\t" \
+	" mov	%3,%0\n" \
+	"4:	.long	2b\n\t" \
+	".previous\n" \
+	".section	__ex_table,\"a\"\n\t" \
+	".long	1b, 3b\n\t" \
+	".previous" \
+	: "=r" (retval) \
+	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
+        : "memory"); })
+#endif
+
+extern void __put_user_unknown(void);
+
+/* Generic arbitrary sized copy.  */
+/* Return the number of bytes NOT copied */
+__kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n);
+
+#define copy_to_user(to,from,n) ({ \
+void *__copy_to = (void *) (to); \
+__kernel_size_t __copy_size = (__kernel_size_t) (n); \
+__kernel_size_t __copy_res; \
+if(__copy_size && __access_ok((unsigned long)__copy_to, __copy_size)) { \
+__copy_res = __copy_user(__copy_to, (void *) (from), __copy_size); \
+} else __copy_res = __copy_size; \
+__copy_res; })
+
+#define copy_from_user(to,from,n) ({ \
+void *__copy_to = (void *) (to); \
+void *__copy_from = (void *) (from); \
+__kernel_size_t __copy_size = (__kernel_size_t) (n); \
+__kernel_size_t __copy_res; \
+if(__copy_size && __access_ok((unsigned long)__copy_from, __copy_size)) { \
+__copy_res = __copy_user(__copy_to, __copy_from, __copy_size); \
+} else __copy_res = __copy_size; \
+__copy_res; })
+
+static __always_inline unsigned long
+__copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	return __copy_user(to, (__force void *)from, n);
+}
+
+static __always_inline unsigned long __must_check
+__copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	return __copy_user((__force void *)to, from, n);
+}
+
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+/*
+ * Clear the area and return remaining number of bytes
+ * (on failure.  Usually it's 0.)
+ */
+extern __kernel_size_t __clear_user(void *addr, __kernel_size_t size);
+
+#define clear_user(addr,n) ({ \
+void * __cl_addr = (addr); \
+unsigned long __cl_size = (n); \
+if (__cl_size && __access_ok(((unsigned long)(__cl_addr)), __cl_size)) \
+__cl_size = __clear_user(__cl_addr, __cl_size); \
+__cl_size; })
+
+static __inline__ int
+__strncpy_from_user(unsigned long __dest, unsigned long __user __src, int __count)
+{
+	__kernel_size_t res;
+	unsigned long __dummy, _d, _s;
+
+	__asm__ __volatile__(
+		"9:\n"
+		"mov.b	@%2+, %1\n\t"
+		"cmp/eq	#0, %1\n\t"
+		"bt/s	2f\n"
+		"1:\n"
+		"mov.b	%1, @%3\n\t"
+		"dt	%7\n\t"
+		"bf/s	9b\n\t"
+		" add	#1, %3\n\t"
+		"2:\n\t"
+		"sub	%7, %0\n"
+		"3:\n"
+		".section .fixup,\"ax\"\n"
+		"4:\n\t"
+		"mov.l	5f, %1\n\t"
+		"jmp	@%1\n\t"
+		" mov	%8, %0\n\t"
+		".balign 4\n"
+		"5:	.long 3b\n"
+		".previous\n"
+		".section __ex_table,\"a\"\n"
+		"	.balign 4\n"
+		"	.long 9b,4b\n"
+		".previous"
+		: "=r" (res), "=&z" (__dummy), "=r" (_s), "=r" (_d)
+		: "0" (__count), "2" (__src), "3" (__dest), "r" (__count),
+		  "i" (-EFAULT)
+		: "memory", "t");
+
+	return res;
+}
+
+#define strncpy_from_user(dest,src,count) ({ \
+unsigned long __sfu_src = (unsigned long) (src); \
+int __sfu_count = (int) (count); \
+long __sfu_res = -EFAULT; \
+if(__access_ok(__sfu_src, __sfu_count)) { \
+__sfu_res = __strncpy_from_user((unsigned long) (dest), __sfu_src, __sfu_count); \
+} __sfu_res; })
+
+/*
+ * Return the size of a string (including the ending 0!)
+ */
+static __inline__ long __strnlen_user(const char __user *__s, long __n)
+{
+	unsigned long res;
+	unsigned long __dummy;
+
+	__asm__ __volatile__(
+		"9:\n"
+		"cmp/eq	%4, %0\n\t"
+		"bt	2f\n"
+		"1:\t"
+		"mov.b	@(%0,%3), %1\n\t"
+		"tst	%1, %1\n\t"
+		"bf/s	9b\n\t"
+		" add	#1, %0\n"
+		"2:\n"
+		".section .fixup,\"ax\"\n"
+		"3:\n\t"
+		"mov.l	4f, %1\n\t"
+		"jmp	@%1\n\t"
+		" mov	#0, %0\n"
+		".balign 4\n"
+		"4:	.long 2b\n"
+		".previous\n"
+		".section __ex_table,\"a\"\n"
+		"	.balign 4\n"
+		"	.long 1b,3b\n"
+		".previous"
+		: "=z" (res), "=&r" (__dummy)
+		: "0" (0), "r" (__s), "r" (__n)
+		: "t");
+	return res;
+}
+
+static __inline__ long strnlen_user(const char __user *s, long n)
+{
+	if (!__addr_ok(s))
+		return 0;
+	else
+		return __strnlen_user(s, n);
+}
+
+#define strlen_user(str)	strnlen_user(str, ~0UL >> 1)
+
+/*
+ * The exception table consists of pairs of addresses: the first is the
+ * address of an instruction that is allowed to fault, and the second is
+ * the address at which the program should continue.  No registers are
+ * modified, so it is entirely up to the continuation code to figure out
+ * what to do.
+ *
+ * All the routines below use bits of fixup code that are out of line
+ * with the main instruction path.  This means when everything is well,
+ * we don't even have to jump over them.  Further, they do not intrude
+ * on our cache or tlb entries.
+ */
+
+struct exception_table_entry
+{
+	unsigned long insn, fixup;
+};
+
+extern int fixup_exception(struct pt_regs *regs);
+
+#endif /* __ASM_SH_UACCESS_H */
diff --git a/include/asm-sh/uaccess_64.h b/include/asm-sh/uaccess_64.h
new file mode 100644
index 000000000000..644c67b65f94
--- /dev/null
+++ b/include/asm-sh/uaccess_64.h
@@ -0,0 +1,316 @@
+#ifndef __ASM_SH64_UACCESS_H
+#define __ASM_SH64_UACCESS_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/uaccess.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003, 2004  Paul Mundt
+ *
+ * User space memory access functions
+ *
+ * Copyright (C) 1999  Niibe Yutaka
+ *
+ *  Based on:
+ *     MIPS implementation version 1.15 by
+ *              Copyright (C) 1996, 1997, 1998 by Ralf Baechle
+ *     and i386 version.
+ *
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+
+#define VERIFY_READ    0
+#define VERIFY_WRITE   1
+
+/*
+ * The fs value determines whether argument validity checking should be
+ * performed or not.  If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are misnamed.
+ */
+
+#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
+
+#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFF)
+#define USER_DS		MAKE_MM_SEG(0x80000000)
+
+#define get_ds()	(KERNEL_DS)
+#define get_fs()        (current_thread_info()->addr_limit)
+#define set_fs(x)       (current_thread_info()->addr_limit=(x))
+
+#define segment_eq(a,b)	((a).seg == (b).seg)
+
+#define __addr_ok(addr) ((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
+
+/*
+ * Uhhuh, this needs 33-bit arithmetic. We have a carry..
+ *
+ * sum := addr + size;  carry? --> flag = true;
+ * if (sum >= addr_limit) flag = true;
+ */
+#define __range_ok(addr,size) (((unsigned long) (addr) + (size) < (current_thread_info()->addr_limit.seg)) ? 0 : 1)
+
+#define access_ok(type,addr,size) (__range_ok(addr,size) == 0)
+#define __access_ok(addr,size) (__range_ok(addr,size) == 0)
+
+/*
+ * Uh, these should become the main single-value transfer routines ...
+ * They automatically use the right size if we just have the right
+ * pointer type ...
+ *
+ * As MIPS uses the same address space for kernel and user data, we
+ * can just do these as direct assignments.
+ *
+ * Careful to not
+ * (a) re-use the arguments for side effects (sizeof is ok)
+ * (b) require any knowledge of processes at this stage
+ */
+#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
+#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * The "__xxx" versions do not do address space checking, useful when
+ * doing multiple accesses to the same area (the user has to do the
+ * checks by hand with "access_ok()")
+ */
+#define __put_user(x,ptr) __put_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __get_user(x,ptr) __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * The "xxx_ret" versions return constant specified in third argument, if
+ * something bad happens. These macros can be optimized for the
+ * case of just returning from the function xxx_ret is used.
+ */
+
+#define put_user_ret(x,ptr,ret) ({ \
+if (put_user(x,ptr)) return ret; })
+
+#define get_user_ret(x,ptr,ret) ({ \
+if (get_user(x,ptr)) return ret; })
+
+#define __put_user_ret(x,ptr,ret) ({ \
+if (__put_user(x,ptr)) return ret; })
+
+#define __get_user_ret(x,ptr,ret) ({ \
+if (__get_user(x,ptr)) return ret; })
+
+struct __large_struct { unsigned long buf[100]; };
+#define __m(x) (*(struct __large_struct *)(x))
+
+#define __get_user_size(x,ptr,size,retval)			\
+do {								\
+	retval = 0;						\
+	switch (size) {						\
+	case 1:							\
+		retval = __get_user_asm_b(x, ptr);		\
+		break;						\
+	case 2:							\
+		retval = __get_user_asm_w(x, ptr);		\
+		break;						\
+	case 4:							\
+		retval = __get_user_asm_l(x, ptr);		\
+		break;						\
+	case 8:							\
+		retval = __get_user_asm_q(x, ptr);		\
+		break;						\
+	default:						\
+		__get_user_unknown();				\
+		break;						\
+	}							\
+} while (0)
+
+#define __get_user_nocheck(x,ptr,size)				\
+({								\
+	long __gu_err, __gu_val;				\
+	__get_user_size((void *)&__gu_val, (long)(ptr),		\
+			(size), __gu_err);			\
+	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	__gu_err;						\
+})
+
+#define __get_user_check(x,ptr,size)				\
+({								\
+	long __gu_addr = (long)(ptr);				\
+	long __gu_err = -EFAULT, __gu_val;			\
+	if (__access_ok(__gu_addr, (size)))			\
+		__get_user_size((void *)&__gu_val, __gu_addr,	\
+				(size), __gu_err);		\
+	(x) = (__typeof__(*(ptr))) __gu_val;			\
+	__gu_err;						\
+})
+
+extern long __get_user_asm_b(void *, long);
+extern long __get_user_asm_w(void *, long);
+extern long __get_user_asm_l(void *, long);
+extern long __get_user_asm_q(void *, long);
+extern void __get_user_unknown(void);
+
+#define __put_user_size(x,ptr,size,retval)			\
+do {								\
+	retval = 0;						\
+	switch (size) {						\
+	case 1:							\
+		retval = __put_user_asm_b(x, ptr);		\
+		break;						\
+	case 2:							\
+		retval = __put_user_asm_w(x, ptr);		\
+		break;						\
+	case 4:							\
+		retval = __put_user_asm_l(x, ptr);		\
+		break;						\
+	case 8:							\
+		retval = __put_user_asm_q(x, ptr);		\
+		break;						\
+	default:						\
+		__put_user_unknown();				\
+	}							\
+} while (0)
+
+#define __put_user_nocheck(x,ptr,size)				\
+({								\
+	long __pu_err;						\
+	__typeof__(*(ptr)) __pu_val = (x);			\
+	__put_user_size((void *)&__pu_val, (long)(ptr), (size), __pu_err); \
+	__pu_err;						\
+})
+
+#define __put_user_check(x,ptr,size)				\
+({								\
+	long __pu_err = -EFAULT;				\
+	long __pu_addr = (long)(ptr);				\
+	__typeof__(*(ptr)) __pu_val = (x);			\
+								\
+	if (__access_ok(__pu_addr, (size)))			\
+		__put_user_size((void *)&__pu_val, __pu_addr, (size), __pu_err);\
+	__pu_err;						\
+})
+
+extern long __put_user_asm_b(void *, long);
+extern long __put_user_asm_w(void *, long);
+extern long __put_user_asm_l(void *, long);
+extern long __put_user_asm_q(void *, long);
+extern void __put_user_unknown(void);
+
+
+/* Generic arbitrary sized copy.  */
+/* Return the number of bytes NOT copied */
+/* XXX: should be such that: 4byte and the rest. */
+extern __kernel_size_t __copy_user(void *__to, const void *__from, __kernel_size_t __n);
+
+#define copy_to_user(to,from,n) ({ \
+void *__copy_to = (void *) (to); \
+__kernel_size_t __copy_size = (__kernel_size_t) (n); \
+__kernel_size_t __copy_res; \
+if(__copy_size && __access_ok((unsigned long)__copy_to, __copy_size)) { \
+__copy_res = __copy_user(__copy_to, (void *) (from), __copy_size); \
+} else __copy_res = __copy_size; \
+__copy_res; })
+
+#define copy_to_user_ret(to,from,n,retval) ({ \
+if (copy_to_user(to,from,n)) \
+	return retval; \
+})
+
+#define __copy_to_user(to,from,n)		\
+	__copy_user((void *)(to),		\
+		    (void *)(from), n)
+
+#define __copy_to_user_ret(to,from,n,retval) ({ \
+if (__copy_to_user(to,from,n)) \
+	return retval; \
+})
+
+#define copy_from_user(to,from,n) ({ \
+void *__copy_to = (void *) (to); \
+void *__copy_from = (void *) (from); \
+__kernel_size_t __copy_size = (__kernel_size_t) (n); \
+__kernel_size_t __copy_res; \
+if(__copy_size && __access_ok((unsigned long)__copy_from, __copy_size)) { \
+__copy_res = __copy_user(__copy_to, __copy_from, __copy_size); \
+} else __copy_res = __copy_size; \
+__copy_res; })
+
+#define copy_from_user_ret(to,from,n,retval) ({ \
+if (copy_from_user(to,from,n)) \
+	return retval; \
+})
+
+#define __copy_from_user(to,from,n)		\
+	__copy_user((void *)(to),		\
+		    (void *)(from), n)
+
+#define __copy_from_user_ret(to,from,n,retval) ({ \
+if (__copy_from_user(to,from,n)) \
+	return retval; \
+})
+
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+/* XXX: Not sure it works well..
+   should be such that: 4byte clear and the rest. */
+extern __kernel_size_t __clear_user(void *addr, __kernel_size_t size);
+
+#define clear_user(addr,n) ({ \
+void * __cl_addr = (addr); \
+unsigned long __cl_size = (n); \
+if (__cl_size && __access_ok(((unsigned long)(__cl_addr)), __cl_size)) \
+__cl_size = __clear_user(__cl_addr, __cl_size); \
+__cl_size; })
+
+extern int __strncpy_from_user(unsigned long __dest, unsigned long __src, int __count);
+
+#define strncpy_from_user(dest,src,count) ({ \
+unsigned long __sfu_src = (unsigned long) (src); \
+int __sfu_count = (int) (count); \
+long __sfu_res = -EFAULT; \
+if(__access_ok(__sfu_src, __sfu_count)) { \
+__sfu_res = __strncpy_from_user((unsigned long) (dest), __sfu_src, __sfu_count); \
+} __sfu_res; })
+
+#define strlen_user(str) strnlen_user(str, ~0UL >> 1)
+
+/*
+ * Return the size of a string (including the ending 0!)
+ */
+extern long __strnlen_user(const char *__s, long __n);
+
+static inline long strnlen_user(const char *s, long n)
+{
+	if (!__addr_ok(s))
+		return 0;
+	else
+		return __strnlen_user(s, n);
+}
+
+struct exception_table_entry
+{
+	unsigned long insn, fixup;
+};
+
+#define ARCH_HAS_SEARCH_EXTABLE
+
+/* If gcc inlines memset, it will use st.q instructions.  Therefore, we need
+   kmalloc allocations to be 8-byte aligned.  Without this, the alignment
+   becomes BYTE_PER_WORD i.e. only 4 (since sizeof(long)==sizeof(void*)==4 on
+   sh64 at the moment). */
+#define ARCH_KMALLOC_MINALIGN 8
+
+/*
+ * We want 8-byte alignment for the slab caches as well, otherwise we have
+ * the same BYTES_PER_WORD (sizeof(void *)) min align in kmem_cache_create().
+ */
+#define ARCH_SLAB_MINALIGN 8
+
+/* Returns 0 if exception not found and fixup.unit otherwise.  */
+extern unsigned long search_exception_table(unsigned long addr);
+extern const struct exception_table_entry *search_exception_tables (unsigned long addr);
+
+#endif /* __ASM_SH64_UACCESS_H */
diff --git a/include/asm-sh64/uaccess.h b/include/asm-sh64/uaccess.h
deleted file mode 100644
index 644c67b65f94..000000000000
--- a/include/asm-sh64/uaccess.h
+++ /dev/null
@@ -1,316 +0,0 @@
-#ifndef __ASM_SH64_UACCESS_H
-#define __ASM_SH64_UACCESS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/uaccess.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- *
- * User space memory access functions
- *
- * Copyright (C) 1999  Niibe Yutaka
- *
- *  Based on:
- *     MIPS implementation version 1.15 by
- *              Copyright (C) 1996, 1997, 1998 by Ralf Baechle
- *     and i386 version.
- *
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-
-#define VERIFY_READ    0
-#define VERIFY_WRITE   1
-
-/*
- * The fs value determines whether argument validity checking should be
- * performed or not.  If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons (Data Segment Register?), these macros are misnamed.
- */
-
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-
-#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFF)
-#define USER_DS		MAKE_MM_SEG(0x80000000)
-
-#define get_ds()	(KERNEL_DS)
-#define get_fs()        (current_thread_info()->addr_limit)
-#define set_fs(x)       (current_thread_info()->addr_limit=(x))
-
-#define segment_eq(a,b)	((a).seg == (b).seg)
-
-#define __addr_ok(addr) ((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
-
-/*
- * Uhhuh, this needs 33-bit arithmetic. We have a carry..
- *
- * sum := addr + size;  carry? --> flag = true;
- * if (sum >= addr_limit) flag = true;
- */
-#define __range_ok(addr,size) (((unsigned long) (addr) + (size) < (current_thread_info()->addr_limit.seg)) ? 0 : 1)
-
-#define access_ok(type,addr,size) (__range_ok(addr,size) == 0)
-#define __access_ok(addr,size) (__range_ok(addr,size) == 0)
-
-/*
- * Uh, these should become the main single-value transfer routines ...
- * They automatically use the right size if we just have the right
- * pointer type ...
- *
- * As MIPS uses the same address space for kernel and user data, we
- * can just do these as direct assignments.
- *
- * Careful to not
- * (a) re-use the arguments for side effects (sizeof is ok)
- * (b) require any knowledge of processes at this stage
- */
-#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
-#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
-
-/*
- * The "__xxx" versions do not do address space checking, useful when
- * doing multiple accesses to the same area (the user has to do the
- * checks by hand with "access_ok()")
- */
-#define __put_user(x,ptr) __put_user_nocheck((x),(ptr),sizeof(*(ptr)))
-#define __get_user(x,ptr) __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
-
-/*
- * The "xxx_ret" versions return constant specified in third argument, if
- * something bad happens. These macros can be optimized for the
- * case of just returning from the function xxx_ret is used.
- */
-
-#define put_user_ret(x,ptr,ret) ({ \
-if (put_user(x,ptr)) return ret; })
-
-#define get_user_ret(x,ptr,ret) ({ \
-if (get_user(x,ptr)) return ret; })
-
-#define __put_user_ret(x,ptr,ret) ({ \
-if (__put_user(x,ptr)) return ret; })
-
-#define __get_user_ret(x,ptr,ret) ({ \
-if (__get_user(x,ptr)) return ret; })
-
-struct __large_struct { unsigned long buf[100]; };
-#define __m(x) (*(struct __large_struct *)(x))
-
-#define __get_user_size(x,ptr,size,retval)			\
-do {								\
-	retval = 0;						\
-	switch (size) {						\
-	case 1:							\
-		retval = __get_user_asm_b(x, ptr);		\
-		break;						\
-	case 2:							\
-		retval = __get_user_asm_w(x, ptr);		\
-		break;						\
-	case 4:							\
-		retval = __get_user_asm_l(x, ptr);		\
-		break;						\
-	case 8:							\
-		retval = __get_user_asm_q(x, ptr);		\
-		break;						\
-	default:						\
-		__get_user_unknown();				\
-		break;						\
-	}							\
-} while (0)
-
-#define __get_user_nocheck(x,ptr,size)				\
-({								\
-	long __gu_err, __gu_val;				\
-	__get_user_size((void *)&__gu_val, (long)(ptr),		\
-			(size), __gu_err);			\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
-	__gu_err;						\
-})
-
-#define __get_user_check(x,ptr,size)				\
-({								\
-	long __gu_addr = (long)(ptr);				\
-	long __gu_err = -EFAULT, __gu_val;			\
-	if (__access_ok(__gu_addr, (size)))			\
-		__get_user_size((void *)&__gu_val, __gu_addr,	\
-				(size), __gu_err);		\
-	(x) = (__typeof__(*(ptr))) __gu_val;			\
-	__gu_err;						\
-})
-
-extern long __get_user_asm_b(void *, long);
-extern long __get_user_asm_w(void *, long);
-extern long __get_user_asm_l(void *, long);
-extern long __get_user_asm_q(void *, long);
-extern void __get_user_unknown(void);
-
-#define __put_user_size(x,ptr,size,retval)			\
-do {								\
-	retval = 0;						\
-	switch (size) {						\
-	case 1:							\
-		retval = __put_user_asm_b(x, ptr);		\
-		break;						\
-	case 2:							\
-		retval = __put_user_asm_w(x, ptr);		\
-		break;						\
-	case 4:							\
-		retval = __put_user_asm_l(x, ptr);		\
-		break;						\
-	case 8:							\
-		retval = __put_user_asm_q(x, ptr);		\
-		break;						\
-	default:						\
-		__put_user_unknown();				\
-	}							\
-} while (0)
-
-#define __put_user_nocheck(x,ptr,size)				\
-({								\
-	long __pu_err;						\
-	__typeof__(*(ptr)) __pu_val = (x);			\
-	__put_user_size((void *)&__pu_val, (long)(ptr), (size), __pu_err); \
-	__pu_err;						\
-})
-
-#define __put_user_check(x,ptr,size)				\
-({								\
-	long __pu_err = -EFAULT;				\
-	long __pu_addr = (long)(ptr);				\
-	__typeof__(*(ptr)) __pu_val = (x);			\
-								\
-	if (__access_ok(__pu_addr, (size)))			\
-		__put_user_size((void *)&__pu_val, __pu_addr, (size), __pu_err);\
-	__pu_err;						\
-})
-
-extern long __put_user_asm_b(void *, long);
-extern long __put_user_asm_w(void *, long);
-extern long __put_user_asm_l(void *, long);
-extern long __put_user_asm_q(void *, long);
-extern void __put_user_unknown(void);
-
-
-/* Generic arbitrary sized copy.  */
-/* Return the number of bytes NOT copied */
-/* XXX: should be such that: 4byte and the rest. */
-extern __kernel_size_t __copy_user(void *__to, const void *__from, __kernel_size_t __n);
-
-#define copy_to_user(to,from,n) ({ \
-void *__copy_to = (void *) (to); \
-__kernel_size_t __copy_size = (__kernel_size_t) (n); \
-__kernel_size_t __copy_res; \
-if(__copy_size && __access_ok((unsigned long)__copy_to, __copy_size)) { \
-__copy_res = __copy_user(__copy_to, (void *) (from), __copy_size); \
-} else __copy_res = __copy_size; \
-__copy_res; })
-
-#define copy_to_user_ret(to,from,n,retval) ({ \
-if (copy_to_user(to,from,n)) \
-	return retval; \
-})
-
-#define __copy_to_user(to,from,n)		\
-	__copy_user((void *)(to),		\
-		    (void *)(from), n)
-
-#define __copy_to_user_ret(to,from,n,retval) ({ \
-if (__copy_to_user(to,from,n)) \
-	return retval; \
-})
-
-#define copy_from_user(to,from,n) ({ \
-void *__copy_to = (void *) (to); \
-void *__copy_from = (void *) (from); \
-__kernel_size_t __copy_size = (__kernel_size_t) (n); \
-__kernel_size_t __copy_res; \
-if(__copy_size && __access_ok((unsigned long)__copy_from, __copy_size)) { \
-__copy_res = __copy_user(__copy_to, __copy_from, __copy_size); \
-} else __copy_res = __copy_size; \
-__copy_res; })
-
-#define copy_from_user_ret(to,from,n,retval) ({ \
-if (copy_from_user(to,from,n)) \
-	return retval; \
-})
-
-#define __copy_from_user(to,from,n)		\
-	__copy_user((void *)(to),		\
-		    (void *)(from), n)
-
-#define __copy_from_user_ret(to,from,n,retval) ({ \
-if (__copy_from_user(to,from,n)) \
-	return retval; \
-})
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-/* XXX: Not sure it works well..
-   should be such that: 4byte clear and the rest. */
-extern __kernel_size_t __clear_user(void *addr, __kernel_size_t size);
-
-#define clear_user(addr,n) ({ \
-void * __cl_addr = (addr); \
-unsigned long __cl_size = (n); \
-if (__cl_size && __access_ok(((unsigned long)(__cl_addr)), __cl_size)) \
-__cl_size = __clear_user(__cl_addr, __cl_size); \
-__cl_size; })
-
-extern int __strncpy_from_user(unsigned long __dest, unsigned long __src, int __count);
-
-#define strncpy_from_user(dest,src,count) ({ \
-unsigned long __sfu_src = (unsigned long) (src); \
-int __sfu_count = (int) (count); \
-long __sfu_res = -EFAULT; \
-if(__access_ok(__sfu_src, __sfu_count)) { \
-__sfu_res = __strncpy_from_user((unsigned long) (dest), __sfu_src, __sfu_count); \
-} __sfu_res; })
-
-#define strlen_user(str) strnlen_user(str, ~0UL >> 1)
-
-/*
- * Return the size of a string (including the ending 0!)
- */
-extern long __strnlen_user(const char *__s, long __n);
-
-static inline long strnlen_user(const char *s, long n)
-{
-	if (!__addr_ok(s))
-		return 0;
-	else
-		return __strnlen_user(s, n);
-}
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-#define ARCH_HAS_SEARCH_EXTABLE
-
-/* If gcc inlines memset, it will use st.q instructions.  Therefore, we need
-   kmalloc allocations to be 8-byte aligned.  Without this, the alignment
-   becomes BYTE_PER_WORD i.e. only 4 (since sizeof(long)==sizeof(void*)==4 on
-   sh64 at the moment). */
-#define ARCH_KMALLOC_MINALIGN 8
-
-/*
- * We want 8-byte alignment for the slab caches as well, otherwise we have
- * the same BYTES_PER_WORD (sizeof(void *)) min align in kmem_cache_create().
- */
-#define ARCH_SLAB_MINALIGN 8
-
-/* Returns 0 if exception not found and fixup.unit otherwise.  */
-extern unsigned long search_exception_table(unsigned long addr);
-extern const struct exception_table_entry *search_exception_tables (unsigned long addr);
-
-#endif /* __ASM_SH64_UACCESS_H */
-- 
cgit v1.2.3


From 01fed9311ab8a724283b3f456c12e573cb51d92b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 19:57:58 +0900
Subject: sh: Consolidate slab/kmalloc minalign values.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/page.h       | 14 ++++++++++++++
 include/asm-sh/uaccess_64.h | 12 ------------
 2 files changed, 14 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index d00a8fde7c7f..d0273dbce6be 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -157,8 +157,22 @@ typedef struct { unsigned long pgd; } pgd_t;
  * Slub defaults to 8-byte alignment, we're only interested in 4.
  * Slab defaults to BYTES_PER_WORD, which ends up being the same anyways.
  */
+#ifdef CONFIG_SUPERH32
 #define ARCH_KMALLOC_MINALIGN	4
 #define ARCH_SLAB_MINALIGN	4
+#else
+/* If gcc inlines memset, it will use st.q instructions.  Therefore, we need
+   kmalloc allocations to be 8-byte aligned.  Without this, the alignment
+   becomes BYTE_PER_WORD i.e. only 4 (since sizeof(long)==sizeof(void*)==4 on
+   sh64 at the moment). */
+#define ARCH_KMALLOC_MINALIGN	8
+
+/*
+ * We want 8-byte alignment for the slab caches as well, otherwise we have
+ * the same BYTES_PER_WORD (sizeof(void *)) min align in kmem_cache_create().
+ */
+#define ARCH_SLAB_MINALIGN	8
+#endif
 
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PAGE_H */
diff --git a/include/asm-sh/uaccess_64.h b/include/asm-sh/uaccess_64.h
index 644c67b65f94..24800a8045ce 100644
--- a/include/asm-sh/uaccess_64.h
+++ b/include/asm-sh/uaccess_64.h
@@ -297,18 +297,6 @@ struct exception_table_entry
 
 #define ARCH_HAS_SEARCH_EXTABLE
 
-/* If gcc inlines memset, it will use st.q instructions.  Therefore, we need
-   kmalloc allocations to be 8-byte aligned.  Without this, the alignment
-   becomes BYTE_PER_WORD i.e. only 4 (since sizeof(long)==sizeof(void*)==4 on
-   sh64 at the moment). */
-#define ARCH_KMALLOC_MINALIGN 8
-
-/*
- * We want 8-byte alignment for the slab caches as well, otherwise we have
- * the same BYTES_PER_WORD (sizeof(void *)) min align in kmem_cache_create().
- */
-#define ARCH_SLAB_MINALIGN 8
-
 /* Returns 0 if exception not found and fixup.unit otherwise.  */
 extern unsigned long search_exception_table(unsigned long addr);
 extern const struct exception_table_entry *search_exception_tables (unsigned long addr);
-- 
cgit v1.2.3


From 76168c21b78a0bd684d4687d14a2bd76bcf92762 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:01:51 +0900
Subject: sh: More SH-5 cpuinfo tidying.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/cache.h |  4 ++++
 include/asm-sh/processor.h     |  2 ++
 include/asm-sh/processor_32.h  |  1 -
 include/asm-sh/processor_64.h  | 10 ++++++----
 4 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/cache.h b/include/asm-sh/cpu-sh5/cache.h
index 2d1f9c28b923..7eaa8894649f 100644
--- a/include/asm-sh/cpu-sh5/cache.h
+++ b/include/asm-sh/cpu-sh5/cache.h
@@ -19,6 +19,10 @@
 #define SH_CACHE_VALID		(1LL<<0)
 #define SH_CACHE_UPDATED	(1LL<<57)
 
+/* Unimplemented compat bits.. */
+#define SH_CACHE_COMBINED	0
+#define SH_CACHE_ASSOC		0
+
 /* Cache flags */
 #define SH_CACHE_MODE_WT	(1LL<<0)
 #define SH_CACHE_MODE_WB	(1LL<<1)
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index bf01f486c3e1..b98c882d5036 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -1,6 +1,8 @@
 #ifndef __ASM_SH_PROCESSOR_H
 #define __ASM_SH_PROCESSOR_H
 
+#include <asm/cpu-features.h>
+
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
  *
diff --git a/include/asm-sh/processor_32.h b/include/asm-sh/processor_32.h
index e10d0ee0c22b..35040fe43e5e 100644
--- a/include/asm-sh/processor_32.h
+++ b/include/asm-sh/processor_32.h
@@ -14,7 +14,6 @@
 #include <asm/types.h>
 #include <asm/cache.h>
 #include <asm/ptrace.h>
-#include <asm/cpu-features.h>
 
 /*
  * Default implementation of macro that returns current
diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
index 6ad23387d7ba..ecd6b403f7db 100644
--- a/include/asm-sh/processor_64.h
+++ b/include/asm-sh/processor_64.h
@@ -66,12 +66,14 @@ struct sh_cpuinfo {
 	/* TLB info */
 	struct tlb_info itlb;
 	struct tlb_info dtlb;
-};
 
-extern struct sh_cpuinfo boot_cpu_data;
+	unsigned long flags;
+};
 
-#define cpu_data (&boot_cpu_data)
-#define current_cpu_data boot_cpu_data
+extern struct sh_cpuinfo cpu_data[];
+#define boot_cpu_data cpu_data[0]
+#define current_cpu_data cpu_data[smp_processor_id()]
+#define raw_current_cpu_data cpu_data[raw_smp_processor_id()]
 
 #endif
 
-- 
cgit v1.2.3


From f7a7b15344e2bccdd1c73d42685edfe7d43ec5b7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:07:57 +0900
Subject: sh: Move in the SH-5 signal trampoline impl.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile_32  |   2 +-
 arch/sh/kernel/Makefile_64  |   2 +-
 arch/sh/kernel/signal.c     | 629 ------------------------------------
 arch/sh/kernel/signal_32.c  | 629 ++++++++++++++++++++++++++++++++++++
 arch/sh/kernel/signal_64.c  | 754 ++++++++++++++++++++++++++++++++++++++++++++
 arch/sh64/kernel/signal.c   | 750 -------------------------------------------
 include/asm-sh/sigcontext.h |  13 +
 7 files changed, 1398 insertions(+), 1381 deletions(-)
 delete mode 100644 arch/sh/kernel/signal.c
 create mode 100644 arch/sh/kernel/signal_32.c
 create mode 100644 arch/sh/kernel/signal_64.c
 delete mode 100644 arch/sh64/kernel/signal.c

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 54b5a2da562e..bad6bf5ac478 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -5,7 +5,7 @@
 extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
-	   ptrace_32.o semaphore.o setup.o signal.o sys_sh.o syscalls.o \
+	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o syscalls.o \
 	   time.o topology.o traps.o
 
 obj-y				+= cpu/ timers/
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index 7dc995d6fdeb..5f2711ae5468 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -1,7 +1,7 @@
 extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
-	   ptrace_64.o semaphore.o setup.o signal.o sys_sh.o syscalls.o \
+	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh.o syscalls.o \
 	   time.o topology.o traps.o
 
 obj-y				+= cpu/ timers/
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
deleted file mode 100644
index ca754fd42437..000000000000
--- a/arch/sh/kernel/signal.c
+++ /dev/null
@@ -1,629 +0,0 @@
-/*
- *  linux/arch/sh/kernel/signal.c
- *
- *  Copyright (C) 1991, 1992  Linus Torvalds
- *
- *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
- *
- *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
- *
- */
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
-#include <linux/signal.h>
-#include <linux/errno.h>
-#include <linux/wait.h>
-#include <linux/ptrace.h>
-#include <linux/unistd.h>
-#include <linux/stddef.h>
-#include <linux/tty.h>
-#include <linux/elf.h>
-#include <linux/personality.h>
-#include <linux/binfmts.h>
-#include <linux/freezer.h>
-#include <linux/io.h>
-#include <asm/system.h>
-#include <asm/ucontext.h>
-#include <asm/uaccess.h>
-#include <asm/pgtable.h>
-#include <asm/cacheflush.h>
-
-#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
-
-/*
- * Atomically swap in the new signal mask, and wait for a signal.
- */
-asmlinkage int
-sys_sigsuspend(old_sigset_t mask,
-	       unsigned long r5, unsigned long r6, unsigned long r7,
-	       struct pt_regs __regs)
-{
-	mask &= _BLOCKABLE;
-	spin_lock_irq(&current->sighand->siglock);
-	current->saved_sigmask = current->blocked;
-	siginitset(&current->blocked, mask);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	current->state = TASK_INTERRUPTIBLE;
-	schedule();
-	set_thread_flag(TIF_RESTORE_SIGMASK);
-	return -ERESTARTNOHAND;
-}
-
-asmlinkage int
-sys_sigaction(int sig, const struct old_sigaction __user *act,
-	      struct old_sigaction __user *oact)
-{
-	struct k_sigaction new_ka, old_ka;
-	int ret;
-
-	if (act) {
-		old_sigset_t mask;
-		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
-		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
-		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
-			return -EFAULT;
-		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
-		__get_user(mask, &act->sa_mask);
-		siginitset(&new_ka.sa.sa_mask, mask);
-	}
-
-	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
-
-	if (!ret && oact) {
-		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
-		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
-			return -EFAULT;
-		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
-		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
-	}
-
-	return ret;
-}
-
-asmlinkage int
-sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
-		unsigned long r6, unsigned long r7,
-		struct pt_regs __regs)
-{
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
-
-	return do_sigaltstack(uss, uoss, regs->regs[15]);
-}
-
-
-/*
- * Do a signal return; undo the signal stack.
- */
-
-#define MOVW(n)	 (0x9300|((n)-2))	/* Move mem word at PC+n to R3 */
-#if defined(CONFIG_CPU_SH2)
-#define TRAP_NOARG 0xc320		/* Syscall w/no args (NR in R3) */
-#else
-#define TRAP_NOARG 0xc310		/* Syscall w/no args (NR in R3) */
-#endif
-#define OR_R0_R0 0x200b			/* or r0,r0 (insert to avoid hardware bug) */
-
-struct sigframe
-{
-	struct sigcontext sc;
-	unsigned long extramask[_NSIG_WORDS-1];
-	u16 retcode[8];
-};
-
-struct rt_sigframe
-{
-	struct siginfo info;
-	struct ucontext uc;
-	u16 retcode[8];
-};
-
-#ifdef CONFIG_SH_FPU
-static inline int restore_sigcontext_fpu(struct sigcontext __user *sc)
-{
-	struct task_struct *tsk = current;
-
-	if (!(current_cpu_data.flags & CPU_HAS_FPU))
-		return 0;
-
-	set_used_math();
-	return __copy_from_user(&tsk->thread.fpu.hard, &sc->sc_fpregs[0],
-				sizeof(long)*(16*2+2));
-}
-
-static inline int save_sigcontext_fpu(struct sigcontext __user *sc,
-				      struct pt_regs *regs)
-{
-	struct task_struct *tsk = current;
-
-	if (!(current_cpu_data.flags & CPU_HAS_FPU))
-		return 0;
-
-	if (!used_math()) {
-		__put_user(0, &sc->sc_ownedfp);
-		return 0;
-	}
-
-	__put_user(1, &sc->sc_ownedfp);
-
-	/* This will cause a "finit" to be triggered by the next
-	   attempted FPU operation by the 'current' process.
-	   */
-	clear_used_math();
-
-	unlazy_fpu(tsk, regs);
-	return __copy_to_user(&sc->sc_fpregs[0], &tsk->thread.fpu.hard,
-			      sizeof(long)*(16*2+2));
-}
-#endif /* CONFIG_SH_FPU */
-
-static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *r0_p)
-{
-	unsigned int err = 0;
-
-#define COPY(x)		err |= __get_user(regs->x, &sc->sc_##x)
-			COPY(regs[1]);
-	COPY(regs[2]);	COPY(regs[3]);
-	COPY(regs[4]);	COPY(regs[5]);
-	COPY(regs[6]);	COPY(regs[7]);
-	COPY(regs[8]);	COPY(regs[9]);
-	COPY(regs[10]);	COPY(regs[11]);
-	COPY(regs[12]);	COPY(regs[13]);
-	COPY(regs[14]);	COPY(regs[15]);
-	COPY(gbr);	COPY(mach);
-	COPY(macl);	COPY(pr);
-	COPY(sr);	COPY(pc);
-#undef COPY
-
-#ifdef CONFIG_SH_FPU
-	if (current_cpu_data.flags & CPU_HAS_FPU) {
-		int owned_fp;
-		struct task_struct *tsk = current;
-
-		regs->sr |= SR_FD; /* Release FPU */
-		clear_fpu(tsk, regs);
-		clear_used_math();
-		__get_user (owned_fp, &sc->sc_ownedfp);
-		if (owned_fp)
-			err |= restore_sigcontext_fpu(sc);
-	}
-#endif
-
-	regs->tra = -1;		/* disable syscall checks */
-	err |= __get_user(*r0_p, &sc->sc_regs[0]);
-	return err;
-}
-
-asmlinkage int sys_sigreturn(unsigned long r4, unsigned long r5,
-			     unsigned long r6, unsigned long r7,
-			     struct pt_regs __regs)
-{
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
-	struct sigframe __user *frame = (struct sigframe __user *)regs->regs[15];
-	sigset_t set;
-	int r0;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__get_user(set.sig[0], &frame->sc.oldmask)
-	    || (_NSIG_WORDS > 1
-		&& __copy_from_user(&set.sig[1], &frame->extramask,
-				    sizeof(frame->extramask))))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->sc, &r0))
-		goto badframe;
-	return r0;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}
-
-asmlinkage int sys_rt_sigreturn(unsigned long r4, unsigned long r5,
-				unsigned long r6, unsigned long r7,
-				struct pt_regs __regs)
-{
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
-	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)regs->regs[15];
-	sigset_t set;
-	stack_t st;
-	int r0;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &r0))
-		goto badframe;
-
-	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
-		goto badframe;
-	/* It is more difficult to avoid calling this function than to
-	   call it and ignore errors.  */
-	do_sigaltstack((const stack_t __user *)&st, NULL, (unsigned long)frame);
-
-	return r0;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}
-
-/*
- * Set up a signal frame.
- */
-
-static int
-setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
-		 unsigned long mask)
-{
-	int err = 0;
-
-#define COPY(x)		err |= __put_user(regs->x, &sc->sc_##x)
-	COPY(regs[0]);	COPY(regs[1]);
-	COPY(regs[2]);	COPY(regs[3]);
-	COPY(regs[4]);	COPY(regs[5]);
-	COPY(regs[6]);	COPY(regs[7]);
-	COPY(regs[8]);	COPY(regs[9]);
-	COPY(regs[10]);	COPY(regs[11]);
-	COPY(regs[12]);	COPY(regs[13]);
-	COPY(regs[14]);	COPY(regs[15]);
-	COPY(gbr);	COPY(mach);
-	COPY(macl);	COPY(pr);
-	COPY(sr);	COPY(pc);
-#undef COPY
-
-#ifdef CONFIG_SH_FPU
-	err |= save_sigcontext_fpu(sc, regs);
-#endif
-
-	/* non-iBCS2 extensions.. */
-	err |= __put_user(mask, &sc->oldmask);
-
-	return err;
-}
-
-/*
- * Determine which stack to use..
- */
-static inline void __user *
-get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
-{
-	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(sp) == 0)
-			sp = current->sas_ss_sp + current->sas_ss_size;
-	}
-
-	return (void __user *)((sp - frame_size) & -8ul);
-}
-
-/* These symbols are defined with the addresses in the vsyscall page.
-   See vsyscall-trapa.S.  */
-extern void __user __kernel_sigreturn;
-extern void __user __kernel_rt_sigreturn;
-
-static int setup_frame(int sig, struct k_sigaction *ka,
-			sigset_t *set, struct pt_regs *regs)
-{
-	struct sigframe __user *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
-
-	if (_NSIG_WORDS > 1)
-		err |= __copy_to_user(frame->extramask, &set->sig[1],
-				      sizeof(frame->extramask));
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		regs->pr = (unsigned long) ka->sa.sa_restorer;
-#ifdef CONFIG_VSYSCALL
-	} else if (likely(current->mm->context.vdso)) {
-		regs->pr = VDSO_SYM(&__kernel_sigreturn);
-#endif
-	} else {
-		/* Generate return code (system call to sigreturn) */
-		err |= __put_user(MOVW(7), &frame->retcode[0]);
-		err |= __put_user(TRAP_NOARG, &frame->retcode[1]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[2]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[3]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[4]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[5]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[6]);
-		err |= __put_user((__NR_sigreturn), &frame->retcode[7]);
-		regs->pr = (unsigned long) frame->retcode;
-	}
-
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up registers for signal handler */
-	regs->regs[15] = (unsigned long) frame;
-	regs->regs[4] = signal; /* Arg for signal handler */
-	regs->regs[5] = 0;
-	regs->regs[6] = (unsigned long) &frame->sc;
-	regs->pc = (unsigned long) ka->sa.sa_handler;
-
-	set_fs(USER_DS);
-
-	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		 current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
-
-	flush_cache_sigtramp(regs->pr);
-
-	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
-		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-
-	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
-}
-
-static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   sigset_t *set, struct pt_regs *regs)
-{
-	struct rt_sigframe __user *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= copy_siginfo_to_user(&frame->info, info);
-
-	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
-	err |= __put_user(0, &frame->uc.uc_link);
-	err |= __put_user((void *)current->sas_ss_sp,
-			  &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->regs[15]),
-			  &frame->uc.uc_stack.ss_flags);
-	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext,
-			        regs, set->sig[0]);
-	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		regs->pr = (unsigned long) ka->sa.sa_restorer;
-#ifdef CONFIG_VSYSCALL
-	} else if (likely(current->mm->context.vdso)) {
-		regs->pr = VDSO_SYM(&__kernel_rt_sigreturn);
-#endif
-	} else {
-		/* Generate return code (system call to rt_sigreturn) */
-		err |= __put_user(MOVW(7), &frame->retcode[0]);
-		err |= __put_user(TRAP_NOARG, &frame->retcode[1]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[2]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[3]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[4]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[5]);
-		err |= __put_user(OR_R0_R0, &frame->retcode[6]);
-		err |= __put_user((__NR_rt_sigreturn), &frame->retcode[7]);
-		regs->pr = (unsigned long) frame->retcode;
-	}
-
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up registers for signal handler */
-	regs->regs[15] = (unsigned long) frame;
-	regs->regs[4] = signal; /* Arg for signal handler */
-	regs->regs[5] = (unsigned long) &frame->info;
-	regs->regs[6] = (unsigned long) &frame->uc;
-	regs->pc = (unsigned long) ka->sa.sa_handler;
-
-	set_fs(USER_DS);
-
-	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		 current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
-
-	flush_cache_sigtramp(regs->pr);
-
-	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
-		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-
-	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
-}
-
-/*
- * OK, we're invoking a handler
- */
-
-static int
-handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
-	      sigset_t *oldset, struct pt_regs *regs, unsigned int save_r0)
-{
-	int ret;
-
-	/* Are we from a system call? */
-	if (regs->tra >= 0) {
-		/* If so, check system call restarting.. */
-		switch (regs->regs[0]) {
-			case -ERESTART_RESTARTBLOCK:
-			case -ERESTARTNOHAND:
-				regs->regs[0] = -EINTR;
-				break;
-
-			case -ERESTARTSYS:
-				if (!(ka->sa.sa_flags & SA_RESTART)) {
-					regs->regs[0] = -EINTR;
-					break;
-				}
-			/* fallthrough */
-			case -ERESTARTNOINTR:
-				regs->regs[0] = save_r0;
-				regs->pc -= instruction_size(
-						ctrl_inw(regs->pc - 4));
-				break;
-		}
-#ifdef CONFIG_GUSA
-	} else {
-		/* gUSA handling */
-		preempt_disable();
-
-		if (regs->regs[15] >= 0xc0000000) {
-			int offset = (int)regs->regs[15];
-
-			/* Reset stack pointer: clear critical region mark */
-			regs->regs[15] = regs->regs[1];
-			if (regs->pc < regs->regs[0])
-				/* Go to rewind point #1 */
-				regs->pc = regs->regs[0] + offset -
-					instruction_size(ctrl_inw(regs->pc-4));
-		}
-
-		preempt_enable_no_resched();
-#endif
-	}
-
-	/* Set up the stack frame */
-	if (ka->sa.sa_flags & SA_SIGINFO)
-		ret = setup_rt_frame(sig, ka, info, oldset, regs);
-	else
-		ret = setup_frame(sig, ka, oldset, regs);
-
-	if (ka->sa.sa_flags & SA_ONESHOT)
-		ka->sa.sa_handler = SIG_DFL;
-
-	if (ret == 0) {
-		spin_lock_irq(&current->sighand->siglock);
-		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-		if (!(ka->sa.sa_flags & SA_NODEFER))
-			sigaddset(&current->blocked,sig);
-		recalc_sigpending();
-		spin_unlock_irq(&current->sighand->siglock);
-	}
-
-	return ret;
-}
-
-/*
- * Note that 'init' is a special process: it doesn't get signals it doesn't
- * want to handle. Thus you cannot kill init even with a SIGKILL even by
- * mistake.
- *
- * Note that we go through the signals twice: once to check the signals that
- * the kernel can handle, and then we build all the user-level signal handling
- * stack-frames in one go after that.
- */
-static void do_signal(struct pt_regs *regs, unsigned int save_r0)
-{
-	siginfo_t info;
-	int signr;
-	struct k_sigaction ka;
-	sigset_t *oldset;
-
-	/*
-	 * We want the common case to go fast, which
-	 * is why we may in certain cases get here from
-	 * kernel mode. Just return without doing anything
-	 * if so.
-	 */
-	if (!user_mode(regs))
-		return;
-
-	if (try_to_freeze())
-		goto no_signal;
-
-	if (test_thread_flag(TIF_RESTORE_SIGMASK))
-		oldset = &current->saved_sigmask;
-	else
-		oldset = &current->blocked;
-
-	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
-	if (signr > 0) {
-		/* Whee!  Actually deliver the signal.  */
-		if (handle_signal(signr, &ka, &info, oldset,
-				  regs, save_r0) == 0) {
-			/* a signal was successfully delivered; the saved
-			 * sigmask will have been stored in the signal frame,
-			 * and will be restored by sigreturn, so we can simply
-			 * clear the TIF_RESTORE_SIGMASK flag */
-			if (test_thread_flag(TIF_RESTORE_SIGMASK))
-				clear_thread_flag(TIF_RESTORE_SIGMASK);
-		}
-
-		return;
-	}
-
- no_signal:
-	/* Did we come from a system call? */
-	if (regs->tra >= 0) {
-		/* Restart the system call - no handlers present */
-		if (regs->regs[0] == -ERESTARTNOHAND ||
-		    regs->regs[0] == -ERESTARTSYS ||
-		    regs->regs[0] == -ERESTARTNOINTR) {
-			regs->regs[0] = save_r0;
-			regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
-		} else if (regs->regs[0] == -ERESTART_RESTARTBLOCK) {
-			regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
-			regs->regs[3] = __NR_restart_syscall;
-		}
-	}
-
-	/* if there's no signal to deliver, we just put the saved sigmask
-	 * back */
-	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
-		clear_thread_flag(TIF_RESTORE_SIGMASK);
-		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
-	}
-}
-
-asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
-				 __u32 thread_info_flags)
-{
-	/* deal with pending signal delivery */
-	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
-		do_signal(regs, save_r0);
-}
diff --git a/arch/sh/kernel/signal_32.c b/arch/sh/kernel/signal_32.c
new file mode 100644
index 000000000000..ca754fd42437
--- /dev/null
+++ b/arch/sh/kernel/signal_32.c
@@ -0,0 +1,629 @@
+/*
+ *  linux/arch/sh/kernel/signal.c
+ *
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *
+ *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
+ *
+ *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
+ *
+ */
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <linux/tty.h>
+#include <linux/elf.h>
+#include <linux/personality.h>
+#include <linux/binfmts.h>
+#include <linux/freezer.h>
+#include <linux/io.h>
+#include <asm/system.h>
+#include <asm/ucontext.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+asmlinkage int
+sys_sigsuspend(old_sigset_t mask,
+	       unsigned long r5, unsigned long r6, unsigned long r7,
+	       struct pt_regs __regs)
+{
+	mask &= _BLOCKABLE;
+	spin_lock_irq(&current->sighand->siglock);
+	current->saved_sigmask = current->blocked;
+	siginitset(&current->blocked, mask);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
+}
+
+asmlinkage int
+sys_sigaction(int sig, const struct old_sigaction __user *act,
+	      struct old_sigaction __user *oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+
+	if (act) {
+		old_sigset_t mask;
+		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
+		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
+		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
+			return -EFAULT;
+		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		__get_user(mask, &act->sa_mask);
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
+		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
+		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
+			return -EFAULT;
+		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
+	}
+
+	return ret;
+}
+
+asmlinkage int
+sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+		unsigned long r6, unsigned long r7,
+		struct pt_regs __regs)
+{
+	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
+
+	return do_sigaltstack(uss, uoss, regs->regs[15]);
+}
+
+
+/*
+ * Do a signal return; undo the signal stack.
+ */
+
+#define MOVW(n)	 (0x9300|((n)-2))	/* Move mem word at PC+n to R3 */
+#if defined(CONFIG_CPU_SH2)
+#define TRAP_NOARG 0xc320		/* Syscall w/no args (NR in R3) */
+#else
+#define TRAP_NOARG 0xc310		/* Syscall w/no args (NR in R3) */
+#endif
+#define OR_R0_R0 0x200b			/* or r0,r0 (insert to avoid hardware bug) */
+
+struct sigframe
+{
+	struct sigcontext sc;
+	unsigned long extramask[_NSIG_WORDS-1];
+	u16 retcode[8];
+};
+
+struct rt_sigframe
+{
+	struct siginfo info;
+	struct ucontext uc;
+	u16 retcode[8];
+};
+
+#ifdef CONFIG_SH_FPU
+static inline int restore_sigcontext_fpu(struct sigcontext __user *sc)
+{
+	struct task_struct *tsk = current;
+
+	if (!(current_cpu_data.flags & CPU_HAS_FPU))
+		return 0;
+
+	set_used_math();
+	return __copy_from_user(&tsk->thread.fpu.hard, &sc->sc_fpregs[0],
+				sizeof(long)*(16*2+2));
+}
+
+static inline int save_sigcontext_fpu(struct sigcontext __user *sc,
+				      struct pt_regs *regs)
+{
+	struct task_struct *tsk = current;
+
+	if (!(current_cpu_data.flags & CPU_HAS_FPU))
+		return 0;
+
+	if (!used_math()) {
+		__put_user(0, &sc->sc_ownedfp);
+		return 0;
+	}
+
+	__put_user(1, &sc->sc_ownedfp);
+
+	/* This will cause a "finit" to be triggered by the next
+	   attempted FPU operation by the 'current' process.
+	   */
+	clear_used_math();
+
+	unlazy_fpu(tsk, regs);
+	return __copy_to_user(&sc->sc_fpregs[0], &tsk->thread.fpu.hard,
+			      sizeof(long)*(16*2+2));
+}
+#endif /* CONFIG_SH_FPU */
+
+static int
+restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *r0_p)
+{
+	unsigned int err = 0;
+
+#define COPY(x)		err |= __get_user(regs->x, &sc->sc_##x)
+			COPY(regs[1]);
+	COPY(regs[2]);	COPY(regs[3]);
+	COPY(regs[4]);	COPY(regs[5]);
+	COPY(regs[6]);	COPY(regs[7]);
+	COPY(regs[8]);	COPY(regs[9]);
+	COPY(regs[10]);	COPY(regs[11]);
+	COPY(regs[12]);	COPY(regs[13]);
+	COPY(regs[14]);	COPY(regs[15]);
+	COPY(gbr);	COPY(mach);
+	COPY(macl);	COPY(pr);
+	COPY(sr);	COPY(pc);
+#undef COPY
+
+#ifdef CONFIG_SH_FPU
+	if (current_cpu_data.flags & CPU_HAS_FPU) {
+		int owned_fp;
+		struct task_struct *tsk = current;
+
+		regs->sr |= SR_FD; /* Release FPU */
+		clear_fpu(tsk, regs);
+		clear_used_math();
+		__get_user (owned_fp, &sc->sc_ownedfp);
+		if (owned_fp)
+			err |= restore_sigcontext_fpu(sc);
+	}
+#endif
+
+	regs->tra = -1;		/* disable syscall checks */
+	err |= __get_user(*r0_p, &sc->sc_regs[0]);
+	return err;
+}
+
+asmlinkage int sys_sigreturn(unsigned long r4, unsigned long r5,
+			     unsigned long r6, unsigned long r7,
+			     struct pt_regs __regs)
+{
+	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
+	struct sigframe __user *frame = (struct sigframe __user *)regs->regs[15];
+	sigset_t set;
+	int r0;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__get_user(set.sig[0], &frame->sc.oldmask)
+	    || (_NSIG_WORDS > 1
+		&& __copy_from_user(&set.sig[1], &frame->extramask,
+				    sizeof(frame->extramask))))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->sc, &r0))
+		goto badframe;
+	return r0;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+asmlinkage int sys_rt_sigreturn(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs __regs)
+{
+	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
+	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)regs->regs[15];
+	sigset_t set;
+	stack_t st;
+	int r0;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &r0))
+		goto badframe;
+
+	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+		goto badframe;
+	/* It is more difficult to avoid calling this function than to
+	   call it and ignore errors.  */
+	do_sigaltstack((const stack_t __user *)&st, NULL, (unsigned long)frame);
+
+	return r0;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+/*
+ * Set up a signal frame.
+ */
+
+static int
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
+		 unsigned long mask)
+{
+	int err = 0;
+
+#define COPY(x)		err |= __put_user(regs->x, &sc->sc_##x)
+	COPY(regs[0]);	COPY(regs[1]);
+	COPY(regs[2]);	COPY(regs[3]);
+	COPY(regs[4]);	COPY(regs[5]);
+	COPY(regs[6]);	COPY(regs[7]);
+	COPY(regs[8]);	COPY(regs[9]);
+	COPY(regs[10]);	COPY(regs[11]);
+	COPY(regs[12]);	COPY(regs[13]);
+	COPY(regs[14]);	COPY(regs[15]);
+	COPY(gbr);	COPY(mach);
+	COPY(macl);	COPY(pr);
+	COPY(sr);	COPY(pc);
+#undef COPY
+
+#ifdef CONFIG_SH_FPU
+	err |= save_sigcontext_fpu(sc, regs);
+#endif
+
+	/* non-iBCS2 extensions.. */
+	err |= __put_user(mask, &sc->oldmask);
+
+	return err;
+}
+
+/*
+ * Determine which stack to use..
+ */
+static inline void __user *
+get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
+{
+	if (ka->sa.sa_flags & SA_ONSTACK) {
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
+	}
+
+	return (void __user *)((sp - frame_size) & -8ul);
+}
+
+/* These symbols are defined with the addresses in the vsyscall page.
+   See vsyscall-trapa.S.  */
+extern void __user __kernel_sigreturn;
+extern void __user __kernel_rt_sigreturn;
+
+static int setup_frame(int sig, struct k_sigaction *ka,
+			sigset_t *set, struct pt_regs *regs)
+{
+	struct sigframe __user *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
+
+	if (_NSIG_WORDS > 1)
+		err |= __copy_to_user(frame->extramask, &set->sig[1],
+				      sizeof(frame->extramask));
+
+	/* Set up to return from userspace.  If provided, use a stub
+	   already in userspace.  */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_sigreturn);
+#endif
+	} else {
+		/* Generate return code (system call to sigreturn) */
+		err |= __put_user(MOVW(7), &frame->retcode[0]);
+		err |= __put_user(TRAP_NOARG, &frame->retcode[1]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[2]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[3]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[4]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[5]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[6]);
+		err |= __put_user((__NR_sigreturn), &frame->retcode[7]);
+		regs->pr = (unsigned long) frame->retcode;
+	}
+
+	if (err)
+		goto give_sigsegv;
+
+	/* Set up registers for signal handler */
+	regs->regs[15] = (unsigned long) frame;
+	regs->regs[4] = signal; /* Arg for signal handler */
+	regs->regs[5] = 0;
+	regs->regs[6] = (unsigned long) &frame->sc;
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+
+	set_fs(USER_DS);
+
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
+
+	flush_cache_sigtramp(regs->pr);
+
+	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
+		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
+
+	return 0;
+
+give_sigsegv:
+	force_sigsegv(sig, current);
+	return -EFAULT;
+}
+
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			   sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	/* Create the ucontext.  */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(0, &frame->uc.uc_link);
+	err |= __put_user((void *)current->sas_ss_sp,
+			  &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->regs[15]),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext,
+			        regs, set->sig[0]);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	/* Set up to return from userspace.  If provided, use a stub
+	   already in userspace.  */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_rt_sigreturn);
+#endif
+	} else {
+		/* Generate return code (system call to rt_sigreturn) */
+		err |= __put_user(MOVW(7), &frame->retcode[0]);
+		err |= __put_user(TRAP_NOARG, &frame->retcode[1]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[2]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[3]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[4]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[5]);
+		err |= __put_user(OR_R0_R0, &frame->retcode[6]);
+		err |= __put_user((__NR_rt_sigreturn), &frame->retcode[7]);
+		regs->pr = (unsigned long) frame->retcode;
+	}
+
+	if (err)
+		goto give_sigsegv;
+
+	/* Set up registers for signal handler */
+	regs->regs[15] = (unsigned long) frame;
+	regs->regs[4] = signal; /* Arg for signal handler */
+	regs->regs[5] = (unsigned long) &frame->info;
+	regs->regs[6] = (unsigned long) &frame->uc;
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+
+	set_fs(USER_DS);
+
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
+
+	flush_cache_sigtramp(regs->pr);
+
+	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
+		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
+
+	return 0;
+
+give_sigsegv:
+	force_sigsegv(sig, current);
+	return -EFAULT;
+}
+
+/*
+ * OK, we're invoking a handler
+ */
+
+static int
+handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
+	      sigset_t *oldset, struct pt_regs *regs, unsigned int save_r0)
+{
+	int ret;
+
+	/* Are we from a system call? */
+	if (regs->tra >= 0) {
+		/* If so, check system call restarting.. */
+		switch (regs->regs[0]) {
+			case -ERESTART_RESTARTBLOCK:
+			case -ERESTARTNOHAND:
+				regs->regs[0] = -EINTR;
+				break;
+
+			case -ERESTARTSYS:
+				if (!(ka->sa.sa_flags & SA_RESTART)) {
+					regs->regs[0] = -EINTR;
+					break;
+				}
+			/* fallthrough */
+			case -ERESTARTNOINTR:
+				regs->regs[0] = save_r0;
+				regs->pc -= instruction_size(
+						ctrl_inw(regs->pc - 4));
+				break;
+		}
+#ifdef CONFIG_GUSA
+	} else {
+		/* gUSA handling */
+		preempt_disable();
+
+		if (regs->regs[15] >= 0xc0000000) {
+			int offset = (int)regs->regs[15];
+
+			/* Reset stack pointer: clear critical region mark */
+			regs->regs[15] = regs->regs[1];
+			if (regs->pc < regs->regs[0])
+				/* Go to rewind point #1 */
+				regs->pc = regs->regs[0] + offset -
+					instruction_size(ctrl_inw(regs->pc-4));
+		}
+
+		preempt_enable_no_resched();
+#endif
+	}
+
+	/* Set up the stack frame */
+	if (ka->sa.sa_flags & SA_SIGINFO)
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
+	else
+		ret = setup_frame(sig, ka, oldset, regs);
+
+	if (ka->sa.sa_flags & SA_ONESHOT)
+		ka->sa.sa_handler = SIG_DFL;
+
+	if (ret == 0) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+		if (!(ka->sa.sa_flags & SA_NODEFER))
+			sigaddset(&current->blocked,sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	return ret;
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Note that we go through the signals twice: once to check the signals that
+ * the kernel can handle, and then we build all the user-level signal handling
+ * stack-frames in one go after that.
+ */
+static void do_signal(struct pt_regs *regs, unsigned int save_r0)
+{
+	siginfo_t info;
+	int signr;
+	struct k_sigaction ka;
+	sigset_t *oldset;
+
+	/*
+	 * We want the common case to go fast, which
+	 * is why we may in certain cases get here from
+	 * kernel mode. Just return without doing anything
+	 * if so.
+	 */
+	if (!user_mode(regs))
+		return;
+
+	if (try_to_freeze())
+		goto no_signal;
+
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
+		oldset = &current->blocked;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+	if (signr > 0) {
+		/* Whee!  Actually deliver the signal.  */
+		if (handle_signal(signr, &ka, &info, oldset,
+				  regs, save_r0) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
+
+		return;
+	}
+
+ no_signal:
+	/* Did we come from a system call? */
+	if (regs->tra >= 0) {
+		/* Restart the system call - no handlers present */
+		if (regs->regs[0] == -ERESTARTNOHAND ||
+		    regs->regs[0] == -ERESTARTSYS ||
+		    regs->regs[0] == -ERESTARTNOINTR) {
+			regs->regs[0] = save_r0;
+			regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
+		} else if (regs->regs[0] == -ERESTART_RESTARTBLOCK) {
+			regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
+			regs->regs[3] = __NR_restart_syscall;
+		}
+	}
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	 * back */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
+				 __u32 thread_info_flags)
+{
+	/* deal with pending signal delivery */
+	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, save_r0);
+}
diff --git a/arch/sh/kernel/signal_64.c b/arch/sh/kernel/signal_64.c
new file mode 100644
index 000000000000..96ba81f324be
--- /dev/null
+++ b/arch/sh/kernel/signal_64.c
@@ -0,0 +1,754 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * arch/sh64/kernel/signal.c
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003  Paul Mundt
+ * Copyright (C) 2004  Richard Curnow
+ *
+ * Started from sh version.
+ *
+ */
+#include <linux/rwsem.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/personality.h>
+#include <linux/freezer.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <asm/ucontext.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+
+#define REG_RET 9
+#define REG_ARG1 2
+#define REG_ARG2 3
+#define REG_ARG3 4
+#define REG_SP 15
+#define REG_PR 18
+#define REF_REG_RET regs->regs[REG_RET]
+#define REF_REG_SP regs->regs[REG_SP]
+#define DEREF_REG_PR regs->regs[REG_PR]
+
+#define DEBUG_SIG 0
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+
+asmlinkage int
+sys_sigsuspend(old_sigset_t mask,
+	       unsigned long r3, unsigned long r4, unsigned long r5,
+	       unsigned long r6, unsigned long r7,
+	       struct pt_regs * regs)
+{
+	sigset_t saveset;
+
+	mask &= _BLOCKABLE;
+	spin_lock_irq(&current->sighand->siglock);
+	saveset = current->blocked;
+	siginitset(&current->blocked, mask);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	REF_REG_RET = -EINTR;
+	while (1) {
+		current->state = TASK_INTERRUPTIBLE;
+		schedule();
+		regs->pc += 4;    /* because sys_sigreturn decrements the pc */
+		if (do_signal(regs, &saveset)) {
+			/* pc now points at signal handler. Need to decrement
+			   it because entry.S will increment it. */
+			regs->pc -= 4;
+			return -EINTR;
+		}
+	}
+}
+
+asmlinkage int
+sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
+	          unsigned long r4, unsigned long r5, unsigned long r6,
+	          unsigned long r7,
+	          struct pt_regs * regs)
+{
+	sigset_t saveset, newset;
+
+	/* XXX: Don't preclude handling different sized sigset_t's.  */
+	if (sigsetsize != sizeof(sigset_t))
+		return -EINVAL;
+
+	if (copy_from_user(&newset, unewset, sizeof(newset)))
+		return -EFAULT;
+	sigdelsetmask(&newset, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	saveset = current->blocked;
+	current->blocked = newset;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	REF_REG_RET = -EINTR;
+	while (1) {
+		current->state = TASK_INTERRUPTIBLE;
+		schedule();
+		regs->pc += 4;    /* because sys_sigreturn decrements the pc */
+		if (do_signal(regs, &saveset)) {
+			/* pc now points at signal handler. Need to decrement
+			   it because entry.S will increment it. */
+			regs->pc -= 4;
+			return -EINTR;
+		}
+	}
+}
+
+asmlinkage int
+sys_sigaction(int sig, const struct old_sigaction __user *act,
+	      struct old_sigaction __user *oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+
+	if (act) {
+		old_sigset_t mask;
+		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
+		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
+		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
+			return -EFAULT;
+		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		__get_user(mask, &act->sa_mask);
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
+		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
+		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
+			return -EFAULT;
+		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
+	}
+
+	return ret;
+}
+
+asmlinkage int
+sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+	        unsigned long r4, unsigned long r5, unsigned long r6,
+	        unsigned long r7,
+	        struct pt_regs * regs)
+{
+	return do_sigaltstack(uss, uoss, REF_REG_SP);
+}
+
+
+/*
+ * Do a signal return; undo the signal stack.
+ */
+
+struct sigframe
+{
+	struct sigcontext sc;
+	unsigned long extramask[_NSIG_WORDS-1];
+	long long retcode[2];
+};
+
+struct rt_sigframe
+{
+	struct siginfo __user *pinfo;
+	void *puc;
+	struct siginfo info;
+	struct ucontext uc;
+	long long retcode[2];
+};
+
+#ifdef CONFIG_SH_FPU
+static inline int
+restore_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	int err = 0;
+	int fpvalid;
+
+	err |= __get_user (fpvalid, &sc->sc_fpvalid);
+	conditional_used_math(fpvalid);
+	if (! fpvalid)
+		return err;
+
+	if (current == last_task_used_math) {
+		last_task_used_math = NULL;
+		regs->sr |= SR_FD;
+	}
+
+	err |= __copy_from_user(&current->thread.fpu.hard, &sc->sc_fpregs[0],
+				(sizeof(long long) * 32) + (sizeof(int) * 1));
+
+	return err;
+}
+
+static inline int
+setup_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	int err = 0;
+	int fpvalid;
+
+	fpvalid = !!used_math();
+	err |= __put_user(fpvalid, &sc->sc_fpvalid);
+	if (! fpvalid)
+		return err;
+
+	if (current == last_task_used_math) {
+		grab_fpu();
+		fpsave(&current->thread.fpu.hard);
+		release_fpu();
+		last_task_used_math = NULL;
+		regs->sr |= SR_FD;
+	}
+
+	err |= __copy_to_user(&sc->sc_fpregs[0], &current->thread.fpu.hard,
+			      (sizeof(long long) * 32) + (sizeof(int) * 1));
+	clear_used_math();
+
+	return err;
+}
+#else
+static inline int
+restore_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	return 0;
+}
+static inline int
+setup_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	return 0;
+}
+#endif
+
+static int
+restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, long long *r2_p)
+{
+	unsigned int err = 0;
+        unsigned long long current_sr, new_sr;
+#define SR_MASK 0xffff8cfd
+
+#define COPY(x)		err |= __get_user(regs->x, &sc->sc_##x)
+
+	COPY(regs[0]);	COPY(regs[1]);	COPY(regs[2]);	COPY(regs[3]);
+	COPY(regs[4]);	COPY(regs[5]);	COPY(regs[6]);	COPY(regs[7]);
+	COPY(regs[8]);	COPY(regs[9]);  COPY(regs[10]);	COPY(regs[11]);
+	COPY(regs[12]);	COPY(regs[13]);	COPY(regs[14]);	COPY(regs[15]);
+	COPY(regs[16]);	COPY(regs[17]);	COPY(regs[18]);	COPY(regs[19]);
+	COPY(regs[20]);	COPY(regs[21]);	COPY(regs[22]);	COPY(regs[23]);
+	COPY(regs[24]);	COPY(regs[25]);	COPY(regs[26]);	COPY(regs[27]);
+	COPY(regs[28]);	COPY(regs[29]);	COPY(regs[30]);	COPY(regs[31]);
+	COPY(regs[32]);	COPY(regs[33]);	COPY(regs[34]);	COPY(regs[35]);
+	COPY(regs[36]);	COPY(regs[37]);	COPY(regs[38]);	COPY(regs[39]);
+	COPY(regs[40]);	COPY(regs[41]);	COPY(regs[42]);	COPY(regs[43]);
+	COPY(regs[44]);	COPY(regs[45]);	COPY(regs[46]);	COPY(regs[47]);
+	COPY(regs[48]);	COPY(regs[49]);	COPY(regs[50]);	COPY(regs[51]);
+	COPY(regs[52]);	COPY(regs[53]);	COPY(regs[54]);	COPY(regs[55]);
+	COPY(regs[56]);	COPY(regs[57]);	COPY(regs[58]);	COPY(regs[59]);
+	COPY(regs[60]);	COPY(regs[61]);	COPY(regs[62]);
+	COPY(tregs[0]);	COPY(tregs[1]);	COPY(tregs[2]);	COPY(tregs[3]);
+	COPY(tregs[4]);	COPY(tregs[5]);	COPY(tregs[6]);	COPY(tregs[7]);
+
+        /* Prevent the signal handler manipulating SR in a way that can
+           crash the kernel. i.e. only allow S, Q, M, PR, SZ, FR to be
+           modified */
+        current_sr = regs->sr;
+        err |= __get_user(new_sr, &sc->sc_sr);
+        regs->sr &= SR_MASK;
+        regs->sr |= (new_sr & ~SR_MASK);
+
+	COPY(pc);
+
+#undef COPY
+
+	/* Must do this last in case it sets regs->sr.fd (i.e. after rest of sr
+	 * has been restored above.) */
+	err |= restore_sigcontext_fpu(regs, sc);
+
+	regs->syscall_nr = -1;		/* disable syscall checks */
+	err |= __get_user(*r2_p, &sc->sc_regs[REG_RET]);
+	return err;
+}
+
+asmlinkage int sys_sigreturn(unsigned long r2, unsigned long r3,
+				   unsigned long r4, unsigned long r5,
+				   unsigned long r6, unsigned long r7,
+				   struct pt_regs * regs)
+{
+	struct sigframe __user *frame = (struct sigframe __user *) (long) REF_REG_SP;
+	sigset_t set;
+	long long ret;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__get_user(set.sig[0], &frame->sc.oldmask)
+	    || (_NSIG_WORDS > 1
+		&& __copy_from_user(&set.sig[1], &frame->extramask,
+				    sizeof(frame->extramask))))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->sc, &ret))
+		goto badframe;
+	regs->pc -= 4;
+
+	return (int) ret;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+asmlinkage int sys_rt_sigreturn(unsigned long r2, unsigned long r3,
+				unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs * regs)
+{
+	struct rt_sigframe __user *frame = (struct rt_sigframe __user *) (long) REF_REG_SP;
+	sigset_t set;
+	stack_t __user st;
+	long long ret;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ret))
+		goto badframe;
+	regs->pc -= 4;
+
+	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+		goto badframe;
+	/* It is more difficult to avoid calling this function than to
+	   call it and ignore errors.  */
+	do_sigaltstack(&st, NULL, REF_REG_SP);
+
+	return (int) ret;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+/*
+ * Set up a signal frame.
+ */
+
+static int
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
+		 unsigned long mask)
+{
+	int err = 0;
+
+	/* Do this first, otherwise is this sets sr->fd, that value isn't preserved. */
+	err |= setup_sigcontext_fpu(regs, sc);
+
+#define COPY(x)		err |= __put_user(regs->x, &sc->sc_##x)
+
+	COPY(regs[0]);	COPY(regs[1]);	COPY(regs[2]);	COPY(regs[3]);
+	COPY(regs[4]);	COPY(regs[5]);	COPY(regs[6]);	COPY(regs[7]);
+	COPY(regs[8]);	COPY(regs[9]);	COPY(regs[10]);	COPY(regs[11]);
+	COPY(regs[12]);	COPY(regs[13]);	COPY(regs[14]);	COPY(regs[15]);
+	COPY(regs[16]);	COPY(regs[17]);	COPY(regs[18]);	COPY(regs[19]);
+	COPY(regs[20]);	COPY(regs[21]);	COPY(regs[22]);	COPY(regs[23]);
+	COPY(regs[24]);	COPY(regs[25]);	COPY(regs[26]);	COPY(regs[27]);
+	COPY(regs[28]);	COPY(regs[29]);	COPY(regs[30]);	COPY(regs[31]);
+	COPY(regs[32]);	COPY(regs[33]);	COPY(regs[34]);	COPY(regs[35]);
+	COPY(regs[36]);	COPY(regs[37]);	COPY(regs[38]);	COPY(regs[39]);
+	COPY(regs[40]);	COPY(regs[41]);	COPY(regs[42]);	COPY(regs[43]);
+	COPY(regs[44]);	COPY(regs[45]);	COPY(regs[46]);	COPY(regs[47]);
+	COPY(regs[48]);	COPY(regs[49]);	COPY(regs[50]);	COPY(regs[51]);
+	COPY(regs[52]);	COPY(regs[53]);	COPY(regs[54]);	COPY(regs[55]);
+	COPY(regs[56]);	COPY(regs[57]);	COPY(regs[58]);	COPY(regs[59]);
+	COPY(regs[60]);	COPY(regs[61]);	COPY(regs[62]);
+	COPY(tregs[0]);	COPY(tregs[1]);	COPY(tregs[2]);	COPY(tregs[3]);
+	COPY(tregs[4]);	COPY(tregs[5]);	COPY(tregs[6]);	COPY(tregs[7]);
+	COPY(sr);	COPY(pc);
+
+#undef COPY
+
+	err |= __put_user(mask, &sc->oldmask);
+
+	return err;
+}
+
+/*
+ * Determine which stack to use..
+ */
+static inline void __user *
+get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
+{
+	if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+	return (void __user *)((sp - frame_size) & -8ul);
+}
+
+void sa_default_restorer(void);		/* See comments below */
+void sa_default_rt_restorer(void);	/* See comments below */
+
+static void setup_frame(int sig, struct k_sigaction *ka,
+			sigset_t *set, struct pt_regs *regs)
+{
+	struct sigframe __user *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->regs[REG_SP], sizeof(*frame));
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
+
+	/* Give up earlier as i386, in case */
+	if (err)
+		goto give_sigsegv;
+
+	if (_NSIG_WORDS > 1) {
+		err |= __copy_to_user(frame->extramask, &set->sig[1],
+				      sizeof(frame->extramask)); }
+
+	/* Give up earlier as i386, in case */
+	if (err)
+		goto give_sigsegv;
+
+	/* Set up to return from userspace.  If provided, use a stub
+	   already in userspace.  */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		DEREF_REG_PR = (unsigned long) ka->sa.sa_restorer | 0x1;
+
+		/*
+		 * On SH5 all edited pointers are subject to NEFF
+		 */
+		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
+        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
+	} else {
+		/*
+		 * Different approach on SH5.
+	         * . Endianness independent asm code gets placed in entry.S .
+		 *   This is limited to four ASM instructions corresponding
+		 *   to two long longs in size.
+		 * . err checking is done on the else branch only
+		 * . flush_icache_range() is called upon __put_user() only
+		 * . all edited pointers are subject to NEFF
+		 * . being code, linker turns ShMedia bit on, always
+		 *   dereference index -1.
+		 */
+		DEREF_REG_PR = (unsigned long) frame->retcode | 0x01;
+		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
+        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
+
+		if (__copy_to_user(frame->retcode,
+			(unsigned long long)sa_default_restorer & (~1), 16) != 0)
+			goto give_sigsegv;
+
+		/* Cohere the trampoline with the I-cache. */
+		flush_cache_sigtramp(DEREF_REG_PR-1, DEREF_REG_PR-1+16);
+	}
+
+	/*
+	 * Set up registers for signal handler.
+	 * All edited pointers are subject to NEFF.
+	 */
+	regs->regs[REG_SP] = (unsigned long) frame;
+	regs->regs[REG_SP] = (regs->regs[REG_SP] & NEFF_SIGN) ?
+        		 (regs->regs[REG_SP] | NEFF_MASK) : regs->regs[REG_SP];
+	regs->regs[REG_ARG1] = signal; /* Arg for signal handler */
+
+        /* FIXME:
+           The glibc profiling support for SH-5 needs to be passed a sigcontext
+           so it can retrieve the PC.  At some point during 2003 the glibc
+           support was changed to receive the sigcontext through the 2nd
+           argument, but there are still versions of libc.so in use that use
+           the 3rd argument.  Until libc.so is stabilised, pass the sigcontext
+           through both 2nd and 3rd arguments.
+        */
+
+	regs->regs[REG_ARG2] = (unsigned long long)(unsigned long)(signed long)&frame->sc;
+	regs->regs[REG_ARG3] = (unsigned long long)(unsigned long)(signed long)&frame->sc;
+
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+	regs->pc = (regs->pc & NEFF_SIGN) ? (regs->pc | NEFF_MASK) : regs->pc;
+
+	set_fs(USER_DS);
+
+#if DEBUG_SIG
+	/* Broken %016Lx */
+	printk("SIG deliver (#%d,%s:%d): sp=%p pc=%08Lx%08Lx link=%08Lx%08Lx\n",
+		signal,
+		current->comm, current->pid, frame,
+		regs->pc >> 32, regs->pc & 0xffffffff,
+		DEREF_REG_PR >> 32, DEREF_REG_PR & 0xffffffff);
+#endif
+
+	return;
+
+give_sigsegv:
+	force_sigsegv(sig, current);
+}
+
+static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			   sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0;
+	int signal;
+
+	frame = get_sigframe(ka, regs->regs[REG_SP], sizeof(*frame));
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		goto give_sigsegv;
+
+	signal = current_thread_info()->exec_domain
+		&& current_thread_info()->exec_domain->signal_invmap
+		&& sig < 32
+		? current_thread_info()->exec_domain->signal_invmap[sig]
+		: sig;
+
+	err |= __put_user(&frame->info, &frame->pinfo);
+	err |= __put_user(&frame->uc, &frame->puc);
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	/* Give up earlier as i386, in case */
+	if (err)
+		goto give_sigsegv;
+
+	/* Create the ucontext.  */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(0, &frame->uc.uc_link);
+	err |= __put_user((void *)current->sas_ss_sp,
+			  &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->regs[REG_SP]),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext,
+			        regs, set->sig[0]);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	/* Give up earlier as i386, in case */
+	if (err)
+		goto give_sigsegv;
+
+	/* Set up to return from userspace.  If provided, use a stub
+	   already in userspace.  */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		DEREF_REG_PR = (unsigned long) ka->sa.sa_restorer | 0x1;
+
+		/*
+		 * On SH5 all edited pointers are subject to NEFF
+		 */
+		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
+        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
+	} else {
+		/*
+		 * Different approach on SH5.
+	         * . Endianness independent asm code gets placed in entry.S .
+		 *   This is limited to four ASM instructions corresponding
+		 *   to two long longs in size.
+		 * . err checking is done on the else branch only
+		 * . flush_icache_range() is called upon __put_user() only
+		 * . all edited pointers are subject to NEFF
+		 * . being code, linker turns ShMedia bit on, always
+		 *   dereference index -1.
+		 */
+
+		DEREF_REG_PR = (unsigned long) frame->retcode | 0x01;
+		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
+        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
+
+		if (__copy_to_user(frame->retcode,
+			(unsigned long long)sa_default_rt_restorer & (~1), 16) != 0)
+			goto give_sigsegv;
+
+		flush_icache_range(DEREF_REG_PR-1, DEREF_REG_PR-1+15);
+	}
+
+	/*
+	 * Set up registers for signal handler.
+	 * All edited pointers are subject to NEFF.
+	 */
+	regs->regs[REG_SP] = (unsigned long) frame;
+	regs->regs[REG_SP] = (regs->regs[REG_SP] & NEFF_SIGN) ?
+        		 (regs->regs[REG_SP] | NEFF_MASK) : regs->regs[REG_SP];
+	regs->regs[REG_ARG1] = signal; /* Arg for signal handler */
+	regs->regs[REG_ARG2] = (unsigned long long)(unsigned long)(signed long)&frame->info;
+	regs->regs[REG_ARG3] = (unsigned long long)(unsigned long)(signed long)&frame->uc.uc_mcontext;
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+	regs->pc = (regs->pc & NEFF_SIGN) ? (regs->pc | NEFF_MASK) : regs->pc;
+
+	set_fs(USER_DS);
+
+#if DEBUG_SIG
+	/* Broken %016Lx */
+	printk("SIG deliver (#%d,%s:%d): sp=%p pc=%08Lx%08Lx link=%08Lx%08Lx\n",
+		signal,
+		current->comm, current->pid, frame,
+		regs->pc >> 32, regs->pc & 0xffffffff,
+		DEREF_REG_PR >> 32, DEREF_REG_PR & 0xffffffff);
+#endif
+
+	return;
+
+give_sigsegv:
+	force_sigsegv(sig, current);
+}
+
+/*
+ * OK, we're invoking a handler
+ */
+
+static void
+handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
+		sigset_t *oldset, struct pt_regs * regs)
+{
+	/* Are we from a system call? */
+	if (regs->syscall_nr >= 0) {
+		/* If so, check system call restarting.. */
+		switch (regs->regs[REG_RET]) {
+			case -ERESTART_RESTARTBLOCK:
+			case -ERESTARTNOHAND:
+				regs->regs[REG_RET] = -EINTR;
+				break;
+
+			case -ERESTARTSYS:
+				if (!(ka->sa.sa_flags & SA_RESTART)) {
+					regs->regs[REG_RET] = -EINTR;
+					break;
+				}
+			/* fallthrough */
+			case -ERESTARTNOINTR:
+				/* Decode syscall # */
+				regs->regs[REG_RET] = regs->syscall_nr;
+				regs->pc -= 4;
+		}
+	}
+
+	/* Set up the stack frame */
+	if (ka->sa.sa_flags & SA_SIGINFO)
+		setup_rt_frame(sig, ka, info, oldset, regs);
+	else
+		setup_frame(sig, ka, oldset, regs);
+
+	spin_lock_irq(&current->sighand->siglock);
+	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+	if (!(ka->sa.sa_flags & SA_NODEFER))
+		sigaddset(&current->blocked,sig);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Note that we go through the signals twice: once to check the signals that
+ * the kernel can handle, and then we build all the user-level signal handling
+ * stack-frames in one go after that.
+ */
+int do_signal(struct pt_regs *regs, sigset_t *oldset)
+{
+	siginfo_t info;
+	int signr;
+	struct k_sigaction ka;
+
+	/*
+	 * We want the common case to go fast, which
+	 * is why we may in certain cases get here from
+	 * kernel mode. Just return without doing anything
+	 * if so.
+	 */
+	if (!user_mode(regs))
+		return 1;
+
+	if (try_to_freeze())
+		goto no_signal;
+
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else if (!oldset)
+		oldset = &current->blocked;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, 0);
+
+	if (signr > 0) {
+		/* Whee!  Actually deliver the signal.  */
+		handle_signal(signr, &info, &ka, oldset, regs);
+
+		/*
+		 * If a signal was successfully delivered, the saved sigmask
+		 * is in its frame, and we can clear the TIF_RESTORE_SIGMASK
+		 * flag.
+		 */
+		if (test_thread_flag(TIF_RESTORE_SIGMASK))
+			clear_thread_flag(TIF_RESTORE_SIGMASK);
+
+		return 1;
+	}
+
+no_signal:
+	/* Did we come from a system call? */
+	if (regs->syscall_nr >= 0) {
+		/* Restart the system call - no handlers present */
+		switch (regs->regs[REG_RET]) {
+		case -ERESTARTNOHAND:
+		case -ERESTARTSYS:
+		case -ERESTARTNOINTR:
+			/* Decode Syscall # */
+			regs->regs[REG_RET] = regs->syscall_nr;
+			regs->pc -= 4;
+			break;
+
+		case -ERESTART_RESTARTBLOCK:
+			regs->regs[REG_RET] = __NR_restart_syscall;
+			regs->pc -= 4;
+			break;
+		}
+	}
+
+	/* No signal to deliver -- put the saved sigmask back */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+
+	return 0;
+}
diff --git a/arch/sh64/kernel/signal.c b/arch/sh64/kernel/signal.c
deleted file mode 100644
index 79fc48cf54c6..000000000000
--- a/arch/sh64/kernel/signal.c
+++ /dev/null
@@ -1,750 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/signal.c
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- * Copyright (C) 2004  Richard Curnow
- *
- * Started from sh version.
- *
- */
-#include <linux/rwsem.h>
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
-#include <linux/signal.h>
-#include <linux/errno.h>
-#include <linux/wait.h>
-#include <linux/personality.h>
-#include <linux/freezer.h>
-#include <linux/ptrace.h>
-#include <linux/unistd.h>
-#include <linux/stddef.h>
-#include <asm/ucontext.h>
-#include <asm/uaccess.h>
-#include <asm/pgtable.h>
-
-
-#define REG_RET 9
-#define REG_ARG1 2
-#define REG_ARG2 3
-#define REG_ARG3 4
-#define REG_SP 15
-#define REG_PR 18
-#define REF_REG_RET regs->regs[REG_RET]
-#define REF_REG_SP regs->regs[REG_SP]
-#define DEREF_REG_PR regs->regs[REG_PR]
-
-#define DEBUG_SIG 0
-
-#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
-
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
-
-/*
- * Atomically swap in the new signal mask, and wait for a signal.
- */
-
-asmlinkage int
-sys_sigsuspend(old_sigset_t mask,
-	       unsigned long r3, unsigned long r4, unsigned long r5,
-	       unsigned long r6, unsigned long r7,
-	       struct pt_regs * regs)
-{
-	sigset_t saveset;
-
-	mask &= _BLOCKABLE;
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	siginitset(&current->blocked, mask);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	REF_REG_RET = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		regs->pc += 4;    /* because sys_sigreturn decrements the pc */
-		if (do_signal(regs, &saveset)) {
-			/* pc now points at signal handler. Need to decrement
-			   it because entry.S will increment it. */
-			regs->pc -= 4;
-			return -EINTR;
-		}
-	}
-}
-
-asmlinkage int
-sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
-	          unsigned long r4, unsigned long r5, unsigned long r6,
-	          unsigned long r7,
-	          struct pt_regs * regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	REF_REG_RET = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		regs->pc += 4;    /* because sys_sigreturn decrements the pc */
-		if (do_signal(regs, &saveset)) {
-			/* pc now points at signal handler. Need to decrement
-			   it because entry.S will increment it. */
-			regs->pc -= 4;
-			return -EINTR;
-		}
-	}
-}
-
-asmlinkage int
-sys_sigaction(int sig, const struct old_sigaction __user *act,
-	      struct old_sigaction __user *oact)
-{
-	struct k_sigaction new_ka, old_ka;
-	int ret;
-
-	if (act) {
-		old_sigset_t mask;
-		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
-		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
-		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
-			return -EFAULT;
-		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
-		__get_user(mask, &act->sa_mask);
-		siginitset(&new_ka.sa.sa_mask, mask);
-	}
-
-	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
-
-	if (!ret && oact) {
-		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
-		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
-			return -EFAULT;
-		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
-		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
-	}
-
-	return ret;
-}
-
-asmlinkage int
-sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
-	        unsigned long r4, unsigned long r5, unsigned long r6,
-	        unsigned long r7,
-	        struct pt_regs * regs)
-{
-	return do_sigaltstack(uss, uoss, REF_REG_SP);
-}
-
-
-/*
- * Do a signal return; undo the signal stack.
- */
-
-struct sigframe
-{
-	struct sigcontext sc;
-	unsigned long extramask[_NSIG_WORDS-1];
-	long long retcode[2];
-};
-
-struct rt_sigframe
-{
-	struct siginfo __user *pinfo;
-	void *puc;
-	struct siginfo info;
-	struct ucontext uc;
-	long long retcode[2];
-};
-
-#ifdef CONFIG_SH_FPU
-static inline int
-restore_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
-{
-	int err = 0;
-	int fpvalid;
-
-	err |= __get_user (fpvalid, &sc->sc_fpvalid);
-	conditional_used_math(fpvalid);
-	if (! fpvalid)
-		return err;
-
-	if (current == last_task_used_math) {
-		last_task_used_math = NULL;
-		regs->sr |= SR_FD;
-	}
-
-	err |= __copy_from_user(&current->thread.fpu.hard, &sc->sc_fpregs[0],
-				(sizeof(long long) * 32) + (sizeof(int) * 1));
-
-	return err;
-}
-
-static inline int
-setup_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
-{
-	int err = 0;
-	int fpvalid;
-
-	fpvalid = !!used_math();
-	err |= __put_user(fpvalid, &sc->sc_fpvalid);
-	if (! fpvalid)
-		return err;
-
-	if (current == last_task_used_math) {
-		grab_fpu();
-		fpsave(&current->thread.fpu.hard);
-		release_fpu();
-		last_task_used_math = NULL;
-		regs->sr |= SR_FD;
-	}
-
-	err |= __copy_to_user(&sc->sc_fpregs[0], &current->thread.fpu.hard,
-			      (sizeof(long long) * 32) + (sizeof(int) * 1));
-	clear_used_math();
-
-	return err;
-}
-#else
-static inline int
-restore_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
-{}
-static inline int
-setup_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
-{}
-#endif
-
-static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, long long *r2_p)
-{
-	unsigned int err = 0;
-        unsigned long long current_sr, new_sr;
-#define SR_MASK 0xffff8cfd
-
-#define COPY(x)		err |= __get_user(regs->x, &sc->sc_##x)
-
-	COPY(regs[0]);	COPY(regs[1]);	COPY(regs[2]);	COPY(regs[3]);
-	COPY(regs[4]);	COPY(regs[5]);	COPY(regs[6]);	COPY(regs[7]);
-	COPY(regs[8]);	COPY(regs[9]);  COPY(regs[10]);	COPY(regs[11]);
-	COPY(regs[12]);	COPY(regs[13]);	COPY(regs[14]);	COPY(regs[15]);
-	COPY(regs[16]);	COPY(regs[17]);	COPY(regs[18]);	COPY(regs[19]);
-	COPY(regs[20]);	COPY(regs[21]);	COPY(regs[22]);	COPY(regs[23]);
-	COPY(regs[24]);	COPY(regs[25]);	COPY(regs[26]);	COPY(regs[27]);
-	COPY(regs[28]);	COPY(regs[29]);	COPY(regs[30]);	COPY(regs[31]);
-	COPY(regs[32]);	COPY(regs[33]);	COPY(regs[34]);	COPY(regs[35]);
-	COPY(regs[36]);	COPY(regs[37]);	COPY(regs[38]);	COPY(regs[39]);
-	COPY(regs[40]);	COPY(regs[41]);	COPY(regs[42]);	COPY(regs[43]);
-	COPY(regs[44]);	COPY(regs[45]);	COPY(regs[46]);	COPY(regs[47]);
-	COPY(regs[48]);	COPY(regs[49]);	COPY(regs[50]);	COPY(regs[51]);
-	COPY(regs[52]);	COPY(regs[53]);	COPY(regs[54]);	COPY(regs[55]);
-	COPY(regs[56]);	COPY(regs[57]);	COPY(regs[58]);	COPY(regs[59]);
-	COPY(regs[60]);	COPY(regs[61]);	COPY(regs[62]);
-	COPY(tregs[0]);	COPY(tregs[1]);	COPY(tregs[2]);	COPY(tregs[3]);
-	COPY(tregs[4]);	COPY(tregs[5]);	COPY(tregs[6]);	COPY(tregs[7]);
-
-        /* Prevent the signal handler manipulating SR in a way that can
-           crash the kernel. i.e. only allow S, Q, M, PR, SZ, FR to be
-           modified */
-        current_sr = regs->sr;
-        err |= __get_user(new_sr, &sc->sc_sr);
-        regs->sr &= SR_MASK;
-        regs->sr |= (new_sr & ~SR_MASK);
-
-	COPY(pc);
-
-#undef COPY
-
-	/* Must do this last in case it sets regs->sr.fd (i.e. after rest of sr
-	 * has been restored above.) */
-	err |= restore_sigcontext_fpu(regs, sc);
-
-	regs->syscall_nr = -1;		/* disable syscall checks */
-	err |= __get_user(*r2_p, &sc->sc_regs[REG_RET]);
-	return err;
-}
-
-asmlinkage int sys_sigreturn(unsigned long r2, unsigned long r3,
-				   unsigned long r4, unsigned long r5,
-				   unsigned long r6, unsigned long r7,
-				   struct pt_regs * regs)
-{
-	struct sigframe __user *frame = (struct sigframe __user *) (long) REF_REG_SP;
-	sigset_t set;
-	long long ret;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__get_user(set.sig[0], &frame->sc.oldmask)
-	    || (_NSIG_WORDS > 1
-		&& __copy_from_user(&set.sig[1], &frame->extramask,
-				    sizeof(frame->extramask))))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->sc, &ret))
-		goto badframe;
-	regs->pc -= 4;
-
-	return (int) ret;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}
-
-asmlinkage int sys_rt_sigreturn(unsigned long r2, unsigned long r3,
-				unsigned long r4, unsigned long r5,
-				unsigned long r6, unsigned long r7,
-				struct pt_regs * regs)
-{
-	struct rt_sigframe __user *frame = (struct rt_sigframe __user *) (long) REF_REG_SP;
-	sigset_t set;
-	stack_t __user st;
-	long long ret;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ret))
-		goto badframe;
-	regs->pc -= 4;
-
-	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
-		goto badframe;
-	/* It is more difficult to avoid calling this function than to
-	   call it and ignore errors.  */
-	do_sigaltstack(&st, NULL, REF_REG_SP);
-
-	return (int) ret;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}
-
-/*
- * Set up a signal frame.
- */
-
-static int
-setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
-		 unsigned long mask)
-{
-	int err = 0;
-
-	/* Do this first, otherwise is this sets sr->fd, that value isn't preserved. */
-	err |= setup_sigcontext_fpu(regs, sc);
-
-#define COPY(x)		err |= __put_user(regs->x, &sc->sc_##x)
-
-	COPY(regs[0]);	COPY(regs[1]);	COPY(regs[2]);	COPY(regs[3]);
-	COPY(regs[4]);	COPY(regs[5]);	COPY(regs[6]);	COPY(regs[7]);
-	COPY(regs[8]);	COPY(regs[9]);	COPY(regs[10]);	COPY(regs[11]);
-	COPY(regs[12]);	COPY(regs[13]);	COPY(regs[14]);	COPY(regs[15]);
-	COPY(regs[16]);	COPY(regs[17]);	COPY(regs[18]);	COPY(regs[19]);
-	COPY(regs[20]);	COPY(regs[21]);	COPY(regs[22]);	COPY(regs[23]);
-	COPY(regs[24]);	COPY(regs[25]);	COPY(regs[26]);	COPY(regs[27]);
-	COPY(regs[28]);	COPY(regs[29]);	COPY(regs[30]);	COPY(regs[31]);
-	COPY(regs[32]);	COPY(regs[33]);	COPY(regs[34]);	COPY(regs[35]);
-	COPY(regs[36]);	COPY(regs[37]);	COPY(regs[38]);	COPY(regs[39]);
-	COPY(regs[40]);	COPY(regs[41]);	COPY(regs[42]);	COPY(regs[43]);
-	COPY(regs[44]);	COPY(regs[45]);	COPY(regs[46]);	COPY(regs[47]);
-	COPY(regs[48]);	COPY(regs[49]);	COPY(regs[50]);	COPY(regs[51]);
-	COPY(regs[52]);	COPY(regs[53]);	COPY(regs[54]);	COPY(regs[55]);
-	COPY(regs[56]);	COPY(regs[57]);	COPY(regs[58]);	COPY(regs[59]);
-	COPY(regs[60]);	COPY(regs[61]);	COPY(regs[62]);
-	COPY(tregs[0]);	COPY(tregs[1]);	COPY(tregs[2]);	COPY(tregs[3]);
-	COPY(tregs[4]);	COPY(tregs[5]);	COPY(tregs[6]);	COPY(tregs[7]);
-	COPY(sr);	COPY(pc);
-
-#undef COPY
-
-	err |= __put_user(mask, &sc->oldmask);
-
-	return err;
-}
-
-/*
- * Determine which stack to use..
- */
-static inline void __user *
-get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
-{
-	if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp))
-		sp = current->sas_ss_sp + current->sas_ss_size;
-
-	return (void __user *)((sp - frame_size) & -8ul);
-}
-
-void sa_default_restorer(void);		/* See comments below */
-void sa_default_rt_restorer(void);	/* See comments below */
-
-static void setup_frame(int sig, struct k_sigaction *ka,
-			sigset_t *set, struct pt_regs *regs)
-{
-	struct sigframe __user *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs->regs[REG_SP], sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
-
-	/* Give up earlier as i386, in case */
-	if (err)
-		goto give_sigsegv;
-
-	if (_NSIG_WORDS > 1) {
-		err |= __copy_to_user(frame->extramask, &set->sig[1],
-				      sizeof(frame->extramask)); }
-
-	/* Give up earlier as i386, in case */
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		DEREF_REG_PR = (unsigned long) ka->sa.sa_restorer | 0x1;
-
-		/*
-		 * On SH5 all edited pointers are subject to NEFF
-		 */
-		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
-        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
-	} else {
-		/*
-		 * Different approach on SH5.
-	         * . Endianness independent asm code gets placed in entry.S .
-		 *   This is limited to four ASM instructions corresponding
-		 *   to two long longs in size.
-		 * . err checking is done on the else branch only
-		 * . flush_icache_range() is called upon __put_user() only
-		 * . all edited pointers are subject to NEFF
-		 * . being code, linker turns ShMedia bit on, always
-		 *   dereference index -1.
-		 */
-		DEREF_REG_PR = (unsigned long) frame->retcode | 0x01;
-		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
-        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
-
-		if (__copy_to_user(frame->retcode,
-			(unsigned long long)sa_default_restorer & (~1), 16) != 0)
-			goto give_sigsegv;
-
-		/* Cohere the trampoline with the I-cache. */
-		flush_cache_sigtramp(DEREF_REG_PR-1, DEREF_REG_PR-1+16);
-	}
-
-	/*
-	 * Set up registers for signal handler.
-	 * All edited pointers are subject to NEFF.
-	 */
-	regs->regs[REG_SP] = (unsigned long) frame;
-	regs->regs[REG_SP] = (regs->regs[REG_SP] & NEFF_SIGN) ?
-        		 (regs->regs[REG_SP] | NEFF_MASK) : regs->regs[REG_SP];
-	regs->regs[REG_ARG1] = signal; /* Arg for signal handler */
-
-        /* FIXME:
-           The glibc profiling support for SH-5 needs to be passed a sigcontext
-           so it can retrieve the PC.  At some point during 2003 the glibc
-           support was changed to receive the sigcontext through the 2nd
-           argument, but there are still versions of libc.so in use that use
-           the 3rd argument.  Until libc.so is stabilised, pass the sigcontext
-           through both 2nd and 3rd arguments.
-        */
-
-	regs->regs[REG_ARG2] = (unsigned long long)(unsigned long)(signed long)&frame->sc;
-	regs->regs[REG_ARG3] = (unsigned long long)(unsigned long)(signed long)&frame->sc;
-
-	regs->pc = (unsigned long) ka->sa.sa_handler;
-	regs->pc = (regs->pc & NEFF_SIGN) ? (regs->pc | NEFF_MASK) : regs->pc;
-
-	set_fs(USER_DS);
-
-#if DEBUG_SIG
-	/* Broken %016Lx */
-	printk("SIG deliver (#%d,%s:%d): sp=%p pc=%08Lx%08Lx link=%08Lx%08Lx\n",
-		signal,
-		current->comm, current->pid, frame,
-		regs->pc >> 32, regs->pc & 0xffffffff,
-		DEREF_REG_PR >> 32, DEREF_REG_PR & 0xffffffff);
-#endif
-
-	return;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-}
-
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   sigset_t *set, struct pt_regs *regs)
-{
-	struct rt_sigframe __user *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs->regs[REG_SP], sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= __put_user(&frame->info, &frame->pinfo);
-	err |= __put_user(&frame->uc, &frame->puc);
-	err |= copy_siginfo_to_user(&frame->info, info);
-
-	/* Give up earlier as i386, in case */
-	if (err)
-		goto give_sigsegv;
-
-	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
-	err |= __put_user(0, &frame->uc.uc_link);
-	err |= __put_user((void *)current->sas_ss_sp,
-			  &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->regs[REG_SP]),
-			  &frame->uc.uc_stack.ss_flags);
-	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext,
-			        regs, set->sig[0]);
-	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-
-	/* Give up earlier as i386, in case */
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		DEREF_REG_PR = (unsigned long) ka->sa.sa_restorer | 0x1;
-
-		/*
-		 * On SH5 all edited pointers are subject to NEFF
-		 */
-		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
-        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
-	} else {
-		/*
-		 * Different approach on SH5.
-	         * . Endianness independent asm code gets placed in entry.S .
-		 *   This is limited to four ASM instructions corresponding
-		 *   to two long longs in size.
-		 * . err checking is done on the else branch only
-		 * . flush_icache_range() is called upon __put_user() only
-		 * . all edited pointers are subject to NEFF
-		 * . being code, linker turns ShMedia bit on, always
-		 *   dereference index -1.
-		 */
-
-		DEREF_REG_PR = (unsigned long) frame->retcode | 0x01;
-		DEREF_REG_PR = (DEREF_REG_PR & NEFF_SIGN) ?
-        		 	(DEREF_REG_PR | NEFF_MASK) : DEREF_REG_PR;
-
-		if (__copy_to_user(frame->retcode,
-			(unsigned long long)sa_default_rt_restorer & (~1), 16) != 0)
-			goto give_sigsegv;
-
-		flush_icache_range(DEREF_REG_PR-1, DEREF_REG_PR-1+15);
-	}
-
-	/*
-	 * Set up registers for signal handler.
-	 * All edited pointers are subject to NEFF.
-	 */
-	regs->regs[REG_SP] = (unsigned long) frame;
-	regs->regs[REG_SP] = (regs->regs[REG_SP] & NEFF_SIGN) ?
-        		 (regs->regs[REG_SP] | NEFF_MASK) : regs->regs[REG_SP];
-	regs->regs[REG_ARG1] = signal; /* Arg for signal handler */
-	regs->regs[REG_ARG2] = (unsigned long long)(unsigned long)(signed long)&frame->info;
-	regs->regs[REG_ARG3] = (unsigned long long)(unsigned long)(signed long)&frame->uc.uc_mcontext;
-	regs->pc = (unsigned long) ka->sa.sa_handler;
-	regs->pc = (regs->pc & NEFF_SIGN) ? (regs->pc | NEFF_MASK) : regs->pc;
-
-	set_fs(USER_DS);
-
-#if DEBUG_SIG
-	/* Broken %016Lx */
-	printk("SIG deliver (#%d,%s:%d): sp=%p pc=%08Lx%08Lx link=%08Lx%08Lx\n",
-		signal,
-		current->comm, current->pid, frame,
-		regs->pc >> 32, regs->pc & 0xffffffff,
-		DEREF_REG_PR >> 32, DEREF_REG_PR & 0xffffffff);
-#endif
-
-	return;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-}
-
-/*
- * OK, we're invoking a handler
- */
-
-static void
-handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
-		sigset_t *oldset, struct pt_regs * regs)
-{
-	/* Are we from a system call? */
-	if (regs->syscall_nr >= 0) {
-		/* If so, check system call restarting.. */
-		switch (regs->regs[REG_RET]) {
-			case -ERESTART_RESTARTBLOCK:
-			case -ERESTARTNOHAND:
-				regs->regs[REG_RET] = -EINTR;
-				break;
-
-			case -ERESTARTSYS:
-				if (!(ka->sa.sa_flags & SA_RESTART)) {
-					regs->regs[REG_RET] = -EINTR;
-					break;
-				}
-			/* fallthrough */
-			case -ERESTARTNOINTR:
-				/* Decode syscall # */
-				regs->regs[REG_RET] = regs->syscall_nr;
-				regs->pc -= 4;
-		}
-	}
-
-	/* Set up the stack frame */
-	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
-	else
-		setup_frame(sig, ka, oldset, regs);
-
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked,sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-}
-
-/*
- * Note that 'init' is a special process: it doesn't get signals it doesn't
- * want to handle. Thus you cannot kill init even with a SIGKILL even by
- * mistake.
- *
- * Note that we go through the signals twice: once to check the signals that
- * the kernel can handle, and then we build all the user-level signal handling
- * stack-frames in one go after that.
- */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
-{
-	siginfo_t info;
-	int signr;
-	struct k_sigaction ka;
-
-	/*
-	 * We want the common case to go fast, which
-	 * is why we may in certain cases get here from
-	 * kernel mode. Just return without doing anything
-	 * if so.
-	 */
-	if (!user_mode(regs))
-		return 1;
-
-	if (try_to_freeze())
-		goto no_signal;
-
-	if (test_thread_flag(TIF_RESTORE_SIGMASK))
-		oldset = &current->saved_sigmask;
-	else if (!oldset)
-		oldset = &current->blocked;
-
-	signr = get_signal_to_deliver(&info, &ka, regs, 0);
-
-	if (signr > 0) {
-		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &info, &ka, oldset, regs);
-
-		/*
-		 * If a signal was successfully delivered, the saved sigmask
-		 * is in its frame, and we can clear the TIF_RESTORE_SIGMASK
-		 * flag.
-		 */
-		if (test_thread_flag(TIF_RESTORE_SIGMASK))
-			clear_thread_flag(TIF_RESTORE_SIGMASK);
-
-		return 1;
-	}
-
-no_signal:
-	/* Did we come from a system call? */
-	if (regs->syscall_nr >= 0) {
-		/* Restart the system call - no handlers present */
-		switch (regs->regs[REG_RET]) {
-		case -ERESTARTNOHAND:
-		case -ERESTARTSYS:
-		case -ERESTARTNOINTR:
-			/* Decode Syscall # */
-			regs->regs[REG_RET] = regs->syscall_nr;
-			regs->pc -= 4;
-			break;
-
-		case -ERESTART_RESTARTBLOCK:
-			regs->regs[REG_RET] = __NR_restart_syscall;
-			regs->pc -= 4;
-			break;
-		}
-	}
-
-	/* No signal to deliver -- put the saved sigmask back */
-	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
-		clear_thread_flag(TIF_RESTORE_SIGMASK);
-		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
-	}
-
-	return 0;
-}
diff --git a/include/asm-sh/sigcontext.h b/include/asm-sh/sigcontext.h
index eb8effba2e80..8583143fa28f 100644
--- a/include/asm-sh/sigcontext.h
+++ b/include/asm-sh/sigcontext.h
@@ -4,6 +4,18 @@
 struct sigcontext {
 	unsigned long	oldmask;
 
+#if defined(__SH5__) || defined(CONFIG_CPU_SH5)
+	/* CPU registers */
+	unsigned long long sc_regs[63];
+	unsigned long long sc_tregs[8];
+	unsigned long long sc_pc;
+	unsigned long long sc_sr;
+
+	/* FPU registers */
+	unsigned long long sc_fpregs[32];
+	unsigned int sc_fpscr;
+	unsigned int sc_fpvalid;
+#else
 	/* CPU registers */
 	unsigned long sc_regs[16];
 	unsigned long sc_pc;
@@ -21,6 +33,7 @@ struct sigcontext {
 	unsigned int sc_fpul;
 	unsigned int sc_ownedfp;
 #endif
+#endif
 };
 
 #endif /* __ASM_SH_SIGCONTEXT_H */
-- 
cgit v1.2.3


From 781c63e68ddc37f2fb5eed9b44b3075cd2b3ced3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:12:25 +0900
Subject: sh: timer.h stub for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/timer.h | 4 ++++
 1 file changed, 4 insertions(+)
 create mode 100644 include/asm-sh/cpu-sh5/timer.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/timer.h b/include/asm-sh/cpu-sh5/timer.h
new file mode 100644
index 000000000000..88da9b341a36
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/timer.h
@@ -0,0 +1,4 @@
+#ifndef __ASM_SH_CPU_SH5_TIMER_H
+#define __ASM_SH_CPU_SH5_TIMER_H
+
+#endif /* __ASM_SH_CPU_SH5_TIMER_H */
-- 
cgit v1.2.3


From 256b22ca66987c537064dc25b0b267966189b5ba Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:27:03 +0900
Subject: sh: Have SH-5 provide an {en,dis}able_fpu() impl.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/ptrace_64.c    |  8 ++++----
 arch/sh/kernel/traps_64.c     |  8 ++++----
 include/asm-sh/processor_64.h | 15 +++++++++++++--
 3 files changed, 21 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/ptrace_64.c b/arch/sh/kernel/ptrace_64.c
index 8a2d339cf760..14e7d5e5679a 100644
--- a/arch/sh/kernel/ptrace_64.c
+++ b/arch/sh/kernel/ptrace_64.c
@@ -74,9 +74,9 @@ get_fpu_long(struct task_struct *task, unsigned long addr)
 	}
 
 	if (last_task_used_math == task) {
-		grab_fpu();
+		enable_fpu();
 		fpsave(&task->thread.fpu.hard);
-		release_fpu();
+		disable_fpu();
 		last_task_used_math = 0;
 		regs->sr |= SR_FD;
 	}
@@ -110,9 +110,9 @@ put_fpu_long(struct task_struct *task, unsigned long addr, unsigned long data)
 		fpinit(&task->thread.fpu.hard);
 		set_stopped_child_used_math(task);
 	} else if (last_task_used_math == task) {
-		grab_fpu();
+		enable_fpu();
 		fpsave(&task->thread.fpu.hard);
-		release_fpu();
+		disable_fpu();
 		last_task_used_math = 0;
 		regs->sr |= SR_FD;
 	}
diff --git a/arch/sh/kernel/traps_64.c b/arch/sh/kernel/traps_64.c
index 742ce18b6820..0f4ea3ac3e0b 100644
--- a/arch/sh/kernel/traps_64.c
+++ b/arch/sh/kernel/traps_64.c
@@ -617,9 +617,9 @@ static int misaligned_fpu_load(struct pt_regs *regs,
 		   context switch the registers into memory so they can be
 		   indexed by register number. */
 		if (last_task_used_math == current) {
-			grab_fpu();
+			enable_fpu();
 			fpsave(&current->thread.fpu.hard);
-			release_fpu();
+			disable_fpu();
 			last_task_used_math = NULL;
 			regs->sr |= SR_FD;
 		}
@@ -690,9 +690,9 @@ static int misaligned_fpu_store(struct pt_regs *regs,
 		   context switch the registers into memory so they can be
 		   indexed by register number. */
 		if (last_task_used_math == current) {
-			grab_fpu();
+			enable_fpu();
 			fpsave(&current->thread.fpu.hard);
-			release_fpu();
+			disable_fpu();
 			last_task_used_math = NULL;
 			regs->sr |= SR_FD;
 		}
diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
index ecd6b403f7db..312fd73fb87e 100644
--- a/include/asm-sh/processor_64.h
+++ b/include/asm-sh/processor_64.h
@@ -18,6 +18,7 @@
 #include <asm/page.h>
 #include <asm/types.h>
 #include <asm/cache.h>
+#include <asm/ptrace.h>
 #include <asm/cpu/registers.h>
 
 /*
@@ -218,7 +219,7 @@ extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
  * FPU lazy state save handling.
  */
 
-static inline void release_fpu(void)
+static inline void disable_fpu(void)
 {
 	unsigned long long __dummy;
 
@@ -230,7 +231,7 @@ static inline void release_fpu(void)
 			     : "r" (SR_FD));
 }
 
-static inline void grab_fpu(void)
+static inline void enable_fpu(void)
 {
 	unsigned long long __dummy;
 
@@ -242,6 +243,16 @@ static inline void grab_fpu(void)
 			     : "r" (~SR_FD));
 }
 
+static inline void release_fpu(struct pt_regs *regs)
+{
+	regs->sr |= SR_FD;
+}
+
+static inline void grab_fpu(struct pt_regs *regs)
+{
+	regs->sr &= ~SR_FD;
+}
+
 /* Round to nearest, no exceptions on inexact, overflow, underflow,
    zero-divide, invalid.  Configure option for whether to flush denorms to
    zero, or except if a denorm is encountered.  */
-- 
cgit v1.2.3


From 959f7d587e236a2d218f527771f156c336409d11 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:35:53 +0900
Subject: sh: Move over the SH-5 head.S and tlb.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile_32 |   2 +-
 arch/sh/kernel/Makefile_64 |   2 +-
 arch/sh/kernel/head.S      | 120 ---------------
 arch/sh/kernel/head_32.S   | 120 +++++++++++++++
 arch/sh/kernel/head_64.S   | 367 ++++++++++++++++++++++++++++++++++++++++++++
 arch/sh64/kernel/head.S    | 372 ---------------------------------------------
 include/asm-sh/tlb.h       |  10 +-
 include/asm-sh/tlb_64.h    |  69 +++++++++
 include/asm-sh64/tlb.h     |  92 -----------
 9 files changed, 567 insertions(+), 587 deletions(-)
 delete mode 100644 arch/sh/kernel/head.S
 create mode 100644 arch/sh/kernel/head_32.S
 create mode 100644 arch/sh/kernel/head_64.S
 delete mode 100644 arch/sh64/kernel/head.S
 create mode 100644 include/asm-sh/tlb_64.h
 delete mode 100644 include/asm-sh64/tlb.h

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 29b44eb3b934..35be5e9674f8 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -2,7 +2,7 @@
 # Makefile for the Linux/SuperH kernel.
 #
 
-extra-y	:= head.o init_task.o vmlinux.lds
+extra-y	:= head_32.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
 	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o syscalls.o \
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index fb87d642d1c1..6e72ed4c07fb 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -1,4 +1,4 @@
-extra-y	:= head.o init_task.o vmlinux.lds
+extra-y	:= head_64.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
 	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh64.o syscalls.o \
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
deleted file mode 100644
index 3338239717f1..000000000000
--- a/arch/sh/kernel/head.S
+++ /dev/null
@@ -1,120 +0,0 @@
-/* $Id: head.S,v 1.7 2003/09/01 17:58:19 lethal Exp $
- *
- *  arch/sh/kernel/head.S
- *
- *  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Head.S contains the SH exception handlers and startup code.
- */
-#include <linux/linkage.h>
-#include <asm/thread_info.h>
-
-#ifdef CONFIG_CPU_SH4A
-#define SYNCO()		synco
-
-#define PREFI(label, reg)	\
-	mov.l	label, reg;	\
-	prefi	@reg
-#else
-#define SYNCO()
-#define PREFI(label, reg)
-#endif
-
-	.section	.empty_zero_page, "aw"
-ENTRY(empty_zero_page)
-	.long	1		/* MOUNT_ROOT_RDONLY */
-	.long	0		/* RAMDISK_FLAGS */
-	.long	0x0200		/* ORIG_ROOT_DEV */
-	.long	1		/* LOADER_TYPE */
-	.long	0x00360000	/* INITRD_START */
-	.long	0x000a0000	/* INITRD_SIZE */
-	.long	0
-1:
-	.skip	PAGE_SIZE - empty_zero_page - 1b
-
-	.section	.text.head, "ax"
-
-/*
- * Condition at the entry of _stext:
- *
- *   BSC has already been initialized.
- *   INTC may or may not be initialized.
- *   VBR may or may not be initialized.
- *   MMU may or may not be initialized.
- *   Cache may or may not be initialized.
- *   Hardware (including on-chip modules) may or may not be initialized. 
- *
- */
-ENTRY(_stext)
-	!			Initialize Status Register
-	mov.l	1f, r0		! MD=1, RB=0, BL=0, IMASK=0xF
-	ldc	r0, sr
-	!			Initialize global interrupt mask
-#ifdef CONFIG_CPU_HAS_SR_RB
-	mov	#0, r0
-	ldc	r0, r6_bank
-#endif
-	
-	/*
-	 * Prefetch if possible to reduce cache miss penalty.
-	 *
-	 * We do this early on for SH-4A as a micro-optimization,
-	 * as later on we will have speculative execution enabled
-	 * and this will become less of an issue.
-	 */
-	PREFI(5f, r0)
-	PREFI(6f, r0)
-
-	!
-	mov.l	2f, r0
-	mov	r0, r15		! Set initial r15 (stack pointer)
-#ifdef CONFIG_CPU_HAS_SR_RB
-	mov.l	7f, r0
-	ldc	r0, r7_bank	! ... and initial thread_info
-#endif
-	
-	!			Clear BSS area
-#ifdef CONFIG_SMP	
-	mov.l	3f, r0
-	cmp/eq	#0, r0		! skip clear if set to zero
-	bt	10f
-#endif
-	
-	mov.l	3f, r1
-	add	#4, r1
-	mov.l	4f, r2
-	mov	#0, r0
-9:	cmp/hs	r2, r1
-	bf/s	9b		! while (r1 < r2)
-	 mov.l	r0,@-r2
-
-10:		
-	!			Additional CPU initialization
-	mov.l	6f, r0
-	jsr	@r0
-	 nop
-
-	SYNCO()			! Wait for pending instructions..
-	
-	!			Start kernel
-	mov.l	5f, r0
-	jmp	@r0
-	 nop
-
-	.balign 4
-#if defined(CONFIG_CPU_SH2)
-1:	.long	0x000000F0		! IMASK=0xF
-#else
-1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
-#endif
-ENTRY(stack_start)
-2:	.long	init_thread_union+THREAD_SIZE
-3:	.long	__bss_start
-4:	.long	_end
-5:	.long	start_kernel
-6:	.long	sh_cpu_init
-7:	.long	init_thread_union
diff --git a/arch/sh/kernel/head_32.S b/arch/sh/kernel/head_32.S
new file mode 100644
index 000000000000..3338239717f1
--- /dev/null
+++ b/arch/sh/kernel/head_32.S
@@ -0,0 +1,120 @@
+/* $Id: head.S,v 1.7 2003/09/01 17:58:19 lethal Exp $
+ *
+ *  arch/sh/kernel/head.S
+ *
+ *  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Head.S contains the SH exception handlers and startup code.
+ */
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+
+#ifdef CONFIG_CPU_SH4A
+#define SYNCO()		synco
+
+#define PREFI(label, reg)	\
+	mov.l	label, reg;	\
+	prefi	@reg
+#else
+#define SYNCO()
+#define PREFI(label, reg)
+#endif
+
+	.section	.empty_zero_page, "aw"
+ENTRY(empty_zero_page)
+	.long	1		/* MOUNT_ROOT_RDONLY */
+	.long	0		/* RAMDISK_FLAGS */
+	.long	0x0200		/* ORIG_ROOT_DEV */
+	.long	1		/* LOADER_TYPE */
+	.long	0x00360000	/* INITRD_START */
+	.long	0x000a0000	/* INITRD_SIZE */
+	.long	0
+1:
+	.skip	PAGE_SIZE - empty_zero_page - 1b
+
+	.section	.text.head, "ax"
+
+/*
+ * Condition at the entry of _stext:
+ *
+ *   BSC has already been initialized.
+ *   INTC may or may not be initialized.
+ *   VBR may or may not be initialized.
+ *   MMU may or may not be initialized.
+ *   Cache may or may not be initialized.
+ *   Hardware (including on-chip modules) may or may not be initialized. 
+ *
+ */
+ENTRY(_stext)
+	!			Initialize Status Register
+	mov.l	1f, r0		! MD=1, RB=0, BL=0, IMASK=0xF
+	ldc	r0, sr
+	!			Initialize global interrupt mask
+#ifdef CONFIG_CPU_HAS_SR_RB
+	mov	#0, r0
+	ldc	r0, r6_bank
+#endif
+	
+	/*
+	 * Prefetch if possible to reduce cache miss penalty.
+	 *
+	 * We do this early on for SH-4A as a micro-optimization,
+	 * as later on we will have speculative execution enabled
+	 * and this will become less of an issue.
+	 */
+	PREFI(5f, r0)
+	PREFI(6f, r0)
+
+	!
+	mov.l	2f, r0
+	mov	r0, r15		! Set initial r15 (stack pointer)
+#ifdef CONFIG_CPU_HAS_SR_RB
+	mov.l	7f, r0
+	ldc	r0, r7_bank	! ... and initial thread_info
+#endif
+	
+	!			Clear BSS area
+#ifdef CONFIG_SMP	
+	mov.l	3f, r0
+	cmp/eq	#0, r0		! skip clear if set to zero
+	bt	10f
+#endif
+	
+	mov.l	3f, r1
+	add	#4, r1
+	mov.l	4f, r2
+	mov	#0, r0
+9:	cmp/hs	r2, r1
+	bf/s	9b		! while (r1 < r2)
+	 mov.l	r0,@-r2
+
+10:		
+	!			Additional CPU initialization
+	mov.l	6f, r0
+	jsr	@r0
+	 nop
+
+	SYNCO()			! Wait for pending instructions..
+	
+	!			Start kernel
+	mov.l	5f, r0
+	jmp	@r0
+	 nop
+
+	.balign 4
+#if defined(CONFIG_CPU_SH2)
+1:	.long	0x000000F0		! IMASK=0xF
+#else
+1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
+#endif
+ENTRY(stack_start)
+2:	.long	init_thread_union+THREAD_SIZE
+3:	.long	__bss_start
+4:	.long	_end
+5:	.long	start_kernel
+6:	.long	sh_cpu_init
+7:	.long	init_thread_union
diff --git a/arch/sh/kernel/head_64.S b/arch/sh/kernel/head_64.S
new file mode 100644
index 000000000000..8015af6d1da2
--- /dev/null
+++ b/arch/sh/kernel/head_64.S
@@ -0,0 +1,367 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * arch/sh64/kernel/head.S
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003, 2004  Paul Mundt
+ *
+ *
+ * benedict.gaster@superh.com:	 2nd May 2002
+ *    Moved definition of empty_zero_page to its own section allowing
+ *    it to be placed at an absolute address known at load time.
+ *
+ * lethal@linux-sh.org:          9th May 2003
+ *    Kill off GLOBAL_NAME() usage.
+ *
+ * lethal@linux-sh.org:          8th May 2004
+ *    Add early SCIF console DTLB mapping.
+ */
+#include <asm/page.h>
+#include <asm/cache.h>
+#include <asm/tlb.h>
+#include <asm/cpu/registers.h>
+#include <asm/cpu/mmu_context.h>
+#include <asm/thread_info.h>
+
+/*
+ * MMU defines: TLB boundaries.
+ */
+
+#define MMUIR_FIRST	ITLB_FIXED
+#define MMUIR_END	ITLB_LAST_VAR_UNRESTRICTED+TLB_STEP
+#define MMUIR_STEP	TLB_STEP
+
+#define MMUDR_FIRST	DTLB_FIXED
+#define MMUDR_END	DTLB_LAST_VAR_UNRESTRICTED+TLB_STEP
+#define MMUDR_STEP	TLB_STEP
+
+/* Safety check : CONFIG_CACHED_MEMORY_OFFSET has to be a multiple of 512Mb */
+#if (CONFIG_CACHED_MEMORY_OFFSET & ((1UL<<29)-1))
+#error "CONFIG_CACHED_MEMORY_OFFSET must be a multiple of 512Mb"
+#endif
+
+/*
+ * MMU defines: Fixed TLBs.
+ */
+/* Deal safely with the case where the base of RAM is not 512Mb aligned */
+
+#define ALIGN_512M_MASK (0xffffffffe0000000)
+#define ALIGNED_EFFECTIVE ((CONFIG_CACHED_MEMORY_OFFSET + CONFIG_MEMORY_START) & ALIGN_512M_MASK)
+#define ALIGNED_PHYSICAL (CONFIG_MEMORY_START & ALIGN_512M_MASK)
+
+#define MMUIR_TEXT_H	(0x0000000000000003 | ALIGNED_EFFECTIVE)
+			/* Enabled, Shared, ASID 0, Eff. Add. 0xA0000000 */
+
+#define MMUIR_TEXT_L	(0x000000000000009a | ALIGNED_PHYSICAL)
+			/* 512 Mb, Cacheable, Write-back, execute, Not User, Ph. Add. */
+
+#define MMUDR_CACHED_H	0x0000000000000003 | ALIGNED_EFFECTIVE
+			/* Enabled, Shared, ASID 0, Eff. Add. 0xA0000000 */
+#define MMUDR_CACHED_L	0x000000000000015a | ALIGNED_PHYSICAL
+			/* 512 Mb, Cacheable, Write-back, read/write, Not User, Ph. Add. */
+
+#ifdef CONFIG_ICACHE_DISABLED
+#define	ICCR0_INIT_VAL	ICCR0_OFF			/* ICACHE off */
+#else
+#define	ICCR0_INIT_VAL	ICCR0_ON | ICCR0_ICI		/* ICE + ICI */
+#endif
+#define	ICCR1_INIT_VAL	ICCR1_NOLOCK			/* No locking */
+
+#if defined (CONFIG_DCACHE_DISABLED)
+#define	OCCR0_INIT_VAL	OCCR0_OFF			   /* D-cache: off  */
+#elif defined (CONFIG_DCACHE_WRITE_THROUGH)
+#define	OCCR0_INIT_VAL	OCCR0_ON | OCCR0_OCI | OCCR0_WT	   /* D-cache: on,   */
+							   /* WT, invalidate */
+#elif defined (CONFIG_DCACHE_WRITE_BACK)
+#define	OCCR0_INIT_VAL	OCCR0_ON | OCCR0_OCI | OCCR0_WB	   /* D-cache: on,   */
+							   /* WB, invalidate */
+#else
+#error preprocessor flag CONFIG_DCACHE_... not recognized!
+#endif
+
+#define	OCCR1_INIT_VAL	OCCR1_NOLOCK			   /* No locking     */
+
+	.section	.empty_zero_page, "aw"
+	.global empty_zero_page
+
+empty_zero_page:
+	.long	1		/* MOUNT_ROOT_RDONLY */
+	.long	0		/* RAMDISK_FLAGS */
+	.long	0x0200		/* ORIG_ROOT_DEV */
+	.long	1		/* LOADER_TYPE */
+	.long	0x00800000	/* INITRD_START */
+	.long	0x00800000	/* INITRD_SIZE */
+	.long	0
+
+	.text
+	.balign 4096,0,4096
+
+	.section	.data, "aw"
+	.balign	PAGE_SIZE
+
+	.section	.data, "aw"
+	.balign	PAGE_SIZE
+
+	.global swapper_pg_dir
+swapper_pg_dir:
+	.space PAGE_SIZE, 0
+
+	.global empty_bad_page
+empty_bad_page:
+	.space PAGE_SIZE, 0
+
+	.global empty_bad_pte_table
+empty_bad_pte_table:
+	.space PAGE_SIZE, 0
+
+	.global	fpu_in_use
+fpu_in_use:	.quad	0
+
+
+	.section	.text.head, "ax"
+	.balign L1_CACHE_BYTES
+/*
+ * Condition at the entry of __stext:
+ * . Reset state:
+ *   . SR.FD    = 1		(FPU disabled)
+ *   . SR.BL    = 1		(Exceptions disabled)
+ *   . SR.MD    = 1		(Privileged Mode)
+ *   . SR.MMU   = 0		(MMU Disabled)
+ *   . SR.CD    = 0		(CTC User Visible)
+ *   . SR.IMASK = Undefined	(Interrupt Mask)
+ *
+ * Operations supposed to be performed by __stext:
+ * . prevent speculative fetch onto device memory while MMU is off
+ * . reflect as much as possible SH5 ABI (r15, r26, r27, r18)
+ * . first, save CPU state and set it to something harmless
+ * . any CPU detection and/or endianness settings (?)
+ * . initialize EMI/LMI (but not TMU/RTC/INTC/SCIF): TBD
+ * . set initial TLB entries for cached and uncached regions
+ *   (no fine granularity paging)
+ * . set initial cache state
+ * . enable MMU and caches
+ * . set CPU to a consistent state
+ *   . registers (including stack pointer and current/KCR0)
+ *   . NOT expecting to set Exception handling nor VBR/RESVEC/DCR
+ *     at this stage. This is all to later Linux initialization steps.
+ *   . initialize FPU
+ * . clear BSS
+ * . jump into start_kernel()
+ * . be prepared to hopeless start_kernel() returns.
+ *
+ */
+	.global _stext
+_stext:
+	/*
+	 * Prevent speculative fetch on device memory due to
+	 * uninitialized target registers.
+	 */
+	ptabs/u	ZERO, tr0
+	ptabs/u	ZERO, tr1
+	ptabs/u	ZERO, tr2
+	ptabs/u	ZERO, tr3
+	ptabs/u	ZERO, tr4
+	ptabs/u	ZERO, tr5
+	ptabs/u	ZERO, tr6
+	ptabs/u	ZERO, tr7
+	synci
+
+	/*
+	 * Read/Set CPU state. After this block:
+	 * r29 = Initial SR
+	 */
+	getcon	SR, r29
+	movi	SR_HARMLESS, r20
+	putcon	r20, SR
+
+	/*
+	 * Initialize EMI/LMI. To Be Done.
+	 */
+
+	/*
+	 * CPU detection and/or endianness settings (?). To Be Done.
+	 * Pure PIC code here, please ! Just save state into r30.
+         * After this block:
+	 * r30 = CPU type/Platform Endianness
+	 */
+
+	/*
+	 * Set initial TLB entries for cached and uncached regions.
+	 * Note: PTA/BLINK is PIC code, PTABS/BLINK isn't !
+	 */
+	/* Clear ITLBs */
+	pta	clear_ITLB, tr1
+	movi	MMUIR_FIRST, r21
+	movi	MMUIR_END, r22
+clear_ITLB:
+	putcfg	r21, 0, ZERO		/* Clear MMUIR[n].PTEH.V */
+	addi	r21, MMUIR_STEP, r21
+        bne	r21, r22, tr1
+
+	/* Clear DTLBs */
+	pta	clear_DTLB, tr1
+	movi	MMUDR_FIRST, r21
+	movi	MMUDR_END, r22
+clear_DTLB:
+	putcfg	r21, 0, ZERO		/* Clear MMUDR[n].PTEH.V */
+	addi	r21, MMUDR_STEP, r21
+        bne	r21, r22, tr1
+
+	/* Map one big (512Mb) page for ITLB */
+	movi	MMUIR_FIRST, r21
+	movi	MMUIR_TEXT_L, r22	/* PTEL first */
+	add.l	r22, r63, r22		/* Sign extend */
+	putcfg	r21, 1, r22		/* Set MMUIR[0].PTEL */
+	movi	MMUIR_TEXT_H, r22	/* PTEH last */
+	add.l	r22, r63, r22		/* Sign extend */
+	putcfg	r21, 0, r22		/* Set MMUIR[0].PTEH */
+
+	/* Map one big CACHED (512Mb) page for DTLB */
+	movi	MMUDR_FIRST, r21
+	movi	MMUDR_CACHED_L, r22	/* PTEL first */
+	add.l	r22, r63, r22		/* Sign extend */
+	putcfg	r21, 1, r22		/* Set MMUDR[0].PTEL */
+	movi	MMUDR_CACHED_H, r22	/* PTEH last */
+	add.l	r22, r63, r22		/* Sign extend */
+	putcfg	r21, 0, r22		/* Set MMUDR[0].PTEH */
+
+#ifdef CONFIG_EARLY_PRINTK
+	/*
+	 * Setup a DTLB translation for SCIF phys.
+	 */
+	addi    r21, MMUDR_STEP, r21
+	movi    0x0a03, r22	/* SCIF phys */
+	shori   0x0148, r22
+	putcfg  r21, 1, r22	/* PTEL first */
+	movi    0xfa03, r22	/* 0xfa030000, fixed SCIF virt */
+	shori   0x0003, r22
+	putcfg  r21, 0, r22	/* PTEH last */
+#endif
+
+	/*
+	 * Set cache behaviours.
+	 */
+	/* ICache */
+	movi	ICCR_BASE, r21
+	movi	ICCR0_INIT_VAL, r22
+	movi	ICCR1_INIT_VAL, r23
+	putcfg	r21, ICCR_REG0, r22
+	putcfg	r21, ICCR_REG1, r23
+
+	/* OCache */
+	movi	OCCR_BASE, r21
+	movi	OCCR0_INIT_VAL, r22
+	movi	OCCR1_INIT_VAL, r23
+	putcfg	r21, OCCR_REG0, r22
+	putcfg	r21, OCCR_REG1, r23
+
+
+	/*
+	 * Enable Caches and MMU. Do the first non-PIC jump.
+         * Now head.S global variables, constants and externs
+	 * can be used.
+	 */
+	getcon	SR, r21
+	movi	SR_ENABLE_MMU, r22
+	or	r21, r22, r21
+	putcon	r21, SSR
+	movi	hyperspace, r22
+	ori	r22, 1, r22	    /* Make it SHmedia, not required but..*/
+	putcon	r22, SPC
+	synco
+	rte			    /* And now go into the hyperspace ... */
+hyperspace:			    /* ... that's the next instruction !  */
+
+	/*
+	 * Set CPU to a consistent state.
+	 * r31 = FPU support flag
+	 * tr0/tr7 in use. Others give a chance to loop somewhere safe
+	 */
+	movi	start_kernel, r32
+	ori	r32, 1, r32
+
+	ptabs	r32, tr0		    /* r32 = _start_kernel address        */
+	pta/u	hopeless, tr1
+	pta/u	hopeless, tr2
+	pta/u	hopeless, tr3
+	pta/u	hopeless, tr4
+	pta/u	hopeless, tr5
+	pta/u	hopeless, tr6
+	pta/u	hopeless, tr7
+	gettr	tr1, r28			/* r28 = hopeless address */
+
+	/* Set initial stack pointer */
+	movi	init_thread_union, SP
+	putcon	SP, KCR0		/* Set current to init_task */
+	movi	THREAD_SIZE, r22	/* Point to the end */
+	add	SP, r22, SP
+
+	/*
+	 * Initialize FPU.
+	 * Keep FPU flag in r31. After this block:
+	 * r31 = FPU flag
+	 */
+	movi fpu_in_use, r31	/* Temporary */
+
+#ifdef CONFIG_SH_FPU
+	getcon	SR, r21
+	movi	SR_ENABLE_FPU, r22
+	and	r21, r22, r22
+	putcon	r22, SR			/* Try to enable */
+	getcon	SR, r22
+	xor	r21, r22, r21
+	shlri	r21, 15, r21		/* Supposedly 0/1 */
+	st.q	r31, 0 , r21		/* Set fpu_in_use */
+#else
+	movi	0, r21
+	st.q	r31, 0 , r21		/* Set fpu_in_use */
+#endif
+	or	r21, ZERO, r31		/* Set FPU flag at last */
+
+#ifndef CONFIG_SH_NO_BSS_INIT
+/* Don't clear BSS if running on slow platforms such as an RTL simulation,
+   remote memory via SHdebug link, etc.  For these the memory can be guaranteed
+   to be all zero on boot anyway. */
+	/*
+	 * Clear bss
+	 */
+	pta	clear_quad, tr1
+	movi	__bss_start, r22
+	movi	_end, r23
+clear_quad:
+	st.q	r22, 0, ZERO
+	addi	r22, 8, r22
+	bne	r22, r23, tr1		/* Both quad aligned, see vmlinux.lds.S */
+#endif
+	pta/u	hopeless, tr1
+
+	/* Say bye to head.S but be prepared to wrongly get back ... */
+	blink	tr0, LINK
+
+	/* If we ever get back here through LINK/tr1-tr7 */
+	pta/u	hopeless, tr7
+
+hopeless:
+	/*
+	 * Something's badly wrong here. Loop endlessly,
+         * there's nothing more we can do about it.
+	 *
+	 * Note on hopeless: it can be jumped into invariably
+	 * before or after jumping into hyperspace. The only
+	 * requirement is to be PIC called (PTA) before and
+	 * any way (PTA/PTABS) after. According to Virtual
+	 * to Physical mapping a simulator/emulator can easily
+	 * tell where we came here from just looking at hopeless
+	 * (PC) address.
+	 *
+	 * For debugging purposes:
+	 * (r28) hopeless/loop address
+	 * (r29) Original SR
+	 * (r30) CPU type/Platform endianness
+	 * (r31) FPU Support
+	 * (r32) _start_kernel address
+	 */
+	blink	tr7, ZERO
diff --git a/arch/sh64/kernel/head.S b/arch/sh64/kernel/head.S
deleted file mode 100644
index 186406d3ad9c..000000000000
--- a/arch/sh64/kernel/head.S
+++ /dev/null
@@ -1,372 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/head.S
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- *
- *
- * benedict.gaster@superh.com:	 2nd May 2002
- *    Moved definition of empty_zero_page to its own section allowing
- *    it to be placed at an absolute address known at load time.
- *
- * lethal@linux-sh.org:          9th May 2003
- *    Kill off GLOBAL_NAME() usage.
- *
- * lethal@linux-sh.org:          8th May 2004
- *    Add early SCIF console DTLB mapping.
- */
-
-
-#include <asm/page.h>
-#include <asm/mmu_context.h>
-#include <asm/cache.h>
-#include <asm/tlb.h>
-#include <asm/processor.h>
-#include <asm/registers.h>
-#include <asm/thread_info.h>
-
-/*
- * MMU defines: TLB boundaries.
- */
-
-#define MMUIR_FIRST	ITLB_FIXED
-#define MMUIR_END	ITLB_LAST_VAR_UNRESTRICTED+TLB_STEP
-#define MMUIR_STEP	TLB_STEP
-
-#define MMUDR_FIRST	DTLB_FIXED
-#define MMUDR_END	DTLB_LAST_VAR_UNRESTRICTED+TLB_STEP
-#define MMUDR_STEP	TLB_STEP
-
-/* Safety check : CONFIG_CACHED_MEMORY_OFFSET has to be a multiple of 512Mb */
-#if (CONFIG_CACHED_MEMORY_OFFSET & ((1UL<<29)-1))
-#error "CONFIG_CACHED_MEMORY_OFFSET must be a multiple of 512Mb"
-#endif
-
-/*
- * MMU defines: Fixed TLBs.
- */
-/* Deal safely with the case where the base of RAM is not 512Mb aligned */
-
-#define ALIGN_512M_MASK (0xffffffffe0000000)
-#define ALIGNED_EFFECTIVE ((CONFIG_CACHED_MEMORY_OFFSET + CONFIG_MEMORY_START) & ALIGN_512M_MASK)
-#define ALIGNED_PHYSICAL (CONFIG_MEMORY_START & ALIGN_512M_MASK)
-
-#define MMUIR_TEXT_H	(0x0000000000000003 | ALIGNED_EFFECTIVE)
-			/* Enabled, Shared, ASID 0, Eff. Add. 0xA0000000 */
-
-#define MMUIR_TEXT_L	(0x000000000000009a | ALIGNED_PHYSICAL)
-			/* 512 Mb, Cacheable, Write-back, execute, Not User, Ph. Add. */
-
-#define MMUDR_CACHED_H	0x0000000000000003 | ALIGNED_EFFECTIVE
-			/* Enabled, Shared, ASID 0, Eff. Add. 0xA0000000 */
-#define MMUDR_CACHED_L	0x000000000000015a | ALIGNED_PHYSICAL
-			/* 512 Mb, Cacheable, Write-back, read/write, Not User, Ph. Add. */
-
-#ifdef CONFIG_ICACHE_DISABLED
-#define	ICCR0_INIT_VAL	ICCR0_OFF			/* ICACHE off */
-#else
-#define	ICCR0_INIT_VAL	ICCR0_ON | ICCR0_ICI		/* ICE + ICI */
-#endif
-#define	ICCR1_INIT_VAL	ICCR1_NOLOCK			/* No locking */
-
-#if defined (CONFIG_DCACHE_DISABLED)
-#define	OCCR0_INIT_VAL	OCCR0_OFF			   /* D-cache: off  */
-#elif defined (CONFIG_DCACHE_WRITE_THROUGH)
-#define	OCCR0_INIT_VAL	OCCR0_ON | OCCR0_OCI | OCCR0_WT	   /* D-cache: on,   */
-							   /* WT, invalidate */
-#elif defined (CONFIG_DCACHE_WRITE_BACK)
-#define	OCCR0_INIT_VAL	OCCR0_ON | OCCR0_OCI | OCCR0_WB	   /* D-cache: on,   */
-							   /* WB, invalidate */
-#else
-#error preprocessor flag CONFIG_DCACHE_... not recognized!
-#endif
-
-#define	OCCR1_INIT_VAL	OCCR1_NOLOCK			   /* No locking     */
-
-	.section	.empty_zero_page, "aw"
-	.global empty_zero_page
-
-empty_zero_page:
-	.long	1		/* MOUNT_ROOT_RDONLY */
-	.long	0		/* RAMDISK_FLAGS */
-	.long	0x0200		/* ORIG_ROOT_DEV */
-	.long	1		/* LOADER_TYPE */
-	.long	0x00800000	/* INITRD_START */
-	.long	0x00800000	/* INITRD_SIZE */
-	.long	0
-
-	.text
-	.balign 4096,0,4096
-
-	.section	.data, "aw"
-	.balign	PAGE_SIZE
-
-	.section	.data, "aw"
-	.balign	PAGE_SIZE
-
-	.global swapper_pg_dir
-swapper_pg_dir:
-	.space PAGE_SIZE, 0
-
-	.global empty_bad_page
-empty_bad_page:
-	.space PAGE_SIZE, 0
-
-	.global empty_bad_pte_table
-empty_bad_pte_table:
-	.space PAGE_SIZE, 0
-
-	.global	fpu_in_use
-fpu_in_use:	.quad	0
-
-
-	.section	.text.head, "ax"
-	.balign L1_CACHE_BYTES
-/*
- * Condition at the entry of __stext:
- * . Reset state:
- *   . SR.FD    = 1		(FPU disabled)
- *   . SR.BL    = 1		(Exceptions disabled)
- *   . SR.MD    = 1		(Privileged Mode)
- *   . SR.MMU   = 0		(MMU Disabled)
- *   . SR.CD    = 0		(CTC User Visible)
- *   . SR.IMASK = Undefined	(Interrupt Mask)
- *
- * Operations supposed to be performed by __stext:
- * . prevent speculative fetch onto device memory while MMU is off
- * . reflect as much as possible SH5 ABI (r15, r26, r27, r18)
- * . first, save CPU state and set it to something harmless
- * . any CPU detection and/or endianness settings (?)
- * . initialize EMI/LMI (but not TMU/RTC/INTC/SCIF): TBD
- * . set initial TLB entries for cached and uncached regions
- *   (no fine granularity paging)
- * . set initial cache state
- * . enable MMU and caches
- * . set CPU to a consistent state
- *   . registers (including stack pointer and current/KCR0)
- *   . NOT expecting to set Exception handling nor VBR/RESVEC/DCR
- *     at this stage. This is all to later Linux initialization steps.
- *   . initialize FPU
- * . clear BSS
- * . jump into start_kernel()
- * . be prepared to hopeless start_kernel() returns.
- *
- */
-	.global _stext
-_stext:
-	/*
-	 * Prevent speculative fetch on device memory due to
-	 * uninitialized target registers.
-	 */
-	ptabs/u	ZERO, tr0
-	ptabs/u	ZERO, tr1
-	ptabs/u	ZERO, tr2
-	ptabs/u	ZERO, tr3
-	ptabs/u	ZERO, tr4
-	ptabs/u	ZERO, tr5
-	ptabs/u	ZERO, tr6
-	ptabs/u	ZERO, tr7
-	synci
-
-	/*
-	 * Read/Set CPU state. After this block:
-	 * r29 = Initial SR
-	 */
-	getcon	SR, r29
-	movi	SR_HARMLESS, r20
-	putcon	r20, SR
-
-	/*
-	 * Initialize EMI/LMI. To Be Done.
-	 */
-
-	/*
-	 * CPU detection and/or endianness settings (?). To Be Done.
-	 * Pure PIC code here, please ! Just save state into r30.
-         * After this block:
-	 * r30 = CPU type/Platform Endianness
-	 */
-
-	/*
-	 * Set initial TLB entries for cached and uncached regions.
-	 * Note: PTA/BLINK is PIC code, PTABS/BLINK isn't !
-	 */
-	/* Clear ITLBs */
-	pta	clear_ITLB, tr1
-	movi	MMUIR_FIRST, r21
-	movi	MMUIR_END, r22
-clear_ITLB:
-	putcfg	r21, 0, ZERO		/* Clear MMUIR[n].PTEH.V */
-	addi	r21, MMUIR_STEP, r21
-        bne	r21, r22, tr1
-
-	/* Clear DTLBs */
-	pta	clear_DTLB, tr1
-	movi	MMUDR_FIRST, r21
-	movi	MMUDR_END, r22
-clear_DTLB:
-	putcfg	r21, 0, ZERO		/* Clear MMUDR[n].PTEH.V */
-	addi	r21, MMUDR_STEP, r21
-        bne	r21, r22, tr1
-
-	/* Map one big (512Mb) page for ITLB */
-	movi	MMUIR_FIRST, r21
-	movi	MMUIR_TEXT_L, r22	/* PTEL first */
-	add.l	r22, r63, r22		/* Sign extend */
-	putcfg	r21, 1, r22		/* Set MMUIR[0].PTEL */
-	movi	MMUIR_TEXT_H, r22	/* PTEH last */
-	add.l	r22, r63, r22		/* Sign extend */
-	putcfg	r21, 0, r22		/* Set MMUIR[0].PTEH */
-
-	/* Map one big CACHED (512Mb) page for DTLB */
-	movi	MMUDR_FIRST, r21
-	movi	MMUDR_CACHED_L, r22	/* PTEL first */
-	add.l	r22, r63, r22		/* Sign extend */
-	putcfg	r21, 1, r22		/* Set MMUDR[0].PTEL */
-	movi	MMUDR_CACHED_H, r22	/* PTEH last */
-	add.l	r22, r63, r22		/* Sign extend */
-	putcfg	r21, 0, r22		/* Set MMUDR[0].PTEH */
-
-#ifdef CONFIG_EARLY_PRINTK
-	/*
-	 * Setup a DTLB translation for SCIF phys.
-	 */
-	addi    r21, MMUDR_STEP, r21
-	movi    0x0a03, r22	/* SCIF phys */
-	shori   0x0148, r22
-	putcfg  r21, 1, r22	/* PTEL first */
-	movi    0xfa03, r22	/* 0xfa030000, fixed SCIF virt */
-	shori   0x0003, r22
-	putcfg  r21, 0, r22	/* PTEH last */
-#endif
-
-	/*
-	 * Set cache behaviours.
-	 */
-	/* ICache */
-	movi	ICCR_BASE, r21
-	movi	ICCR0_INIT_VAL, r22
-	movi	ICCR1_INIT_VAL, r23
-	putcfg	r21, ICCR_REG0, r22
-	putcfg	r21, ICCR_REG1, r23
-
-	/* OCache */
-	movi	OCCR_BASE, r21
-	movi	OCCR0_INIT_VAL, r22
-	movi	OCCR1_INIT_VAL, r23
-	putcfg	r21, OCCR_REG0, r22
-	putcfg	r21, OCCR_REG1, r23
-
-
-	/*
-	 * Enable Caches and MMU. Do the first non-PIC jump.
-         * Now head.S global variables, constants and externs
-	 * can be used.
-	 */
-	getcon	SR, r21
-	movi	SR_ENABLE_MMU, r22
-	or	r21, r22, r21
-	putcon	r21, SSR
-	movi	hyperspace, r22
-	ori	r22, 1, r22	    /* Make it SHmedia, not required but..*/
-	putcon	r22, SPC
-	synco
-	rte			    /* And now go into the hyperspace ... */
-hyperspace:			    /* ... that's the next instruction !  */
-
-	/*
-	 * Set CPU to a consistent state.
-	 * r31 = FPU support flag
-	 * tr0/tr7 in use. Others give a chance to loop somewhere safe
-	 */
-	movi	start_kernel, r32
-	ori	r32, 1, r32
-
-	ptabs	r32, tr0		    /* r32 = _start_kernel address        */
-	pta/u	hopeless, tr1
-	pta/u	hopeless, tr2
-	pta/u	hopeless, tr3
-	pta/u	hopeless, tr4
-	pta/u	hopeless, tr5
-	pta/u	hopeless, tr6
-	pta/u	hopeless, tr7
-	gettr	tr1, r28			/* r28 = hopeless address */
-
-	/* Set initial stack pointer */
-	movi	init_thread_union, SP
-	putcon	SP, KCR0		/* Set current to init_task */
-	movi	THREAD_SIZE, r22	/* Point to the end */
-	add	SP, r22, SP
-
-	/*
-	 * Initialize FPU.
-	 * Keep FPU flag in r31. After this block:
-	 * r31 = FPU flag
-	 */
-	movi fpu_in_use, r31	/* Temporary */
-
-#ifdef CONFIG_SH_FPU
-	getcon	SR, r21
-	movi	SR_ENABLE_FPU, r22
-	and	r21, r22, r22
-	putcon	r22, SR			/* Try to enable */
-	getcon	SR, r22
-	xor	r21, r22, r21
-	shlri	r21, 15, r21		/* Supposedly 0/1 */
-	st.q	r31, 0 , r21		/* Set fpu_in_use */
-#else
-	movi	0, r21
-	st.q	r31, 0 , r21		/* Set fpu_in_use */
-#endif
-	or	r21, ZERO, r31		/* Set FPU flag at last */
-
-#ifndef CONFIG_SH_NO_BSS_INIT
-/* Don't clear BSS if running on slow platforms such as an RTL simulation,
-   remote memory via SHdebug link, etc.  For these the memory can be guaranteed
-   to be all zero on boot anyway. */
-	/*
-	 * Clear bss
-	 */
-	pta	clear_quad, tr1
-	movi	__bss_start, r22
-	movi	_end, r23
-clear_quad:
-	st.q	r22, 0, ZERO
-	addi	r22, 8, r22
-	bne	r22, r23, tr1		/* Both quad aligned, see vmlinux.lds.S */
-#endif
-	pta/u	hopeless, tr1
-
-	/* Say bye to head.S but be prepared to wrongly get back ... */
-	blink	tr0, LINK
-
-	/* If we ever get back here through LINK/tr1-tr7 */
-	pta/u	hopeless, tr7
-
-hopeless:
-	/*
-	 * Something's badly wrong here. Loop endlessly,
-         * there's nothing more we can do about it.
-	 *
-	 * Note on hopeless: it can be jumped into invariably
-	 * before or after jumping into hyperspace. The only
-	 * requirement is to be PIC called (PTA) before and
-	 * any way (PTA/PTABS) after. According to Virtual
-	 * to Physical mapping a simulator/emulator can easily
-	 * tell where we came here from just looking at hopeless
-	 * (PC) address.
-	 *
-	 * For debugging purposes:
-	 * (r28) hopeless/loop address
-	 * (r29) Original SR
-	 * (r30) CPU type/Platform endianness
-	 * (r31) FPU Support
-	 * (r32) _start_kernel address
-	 */
-	blink	tr7, ZERO
-
-
diff --git a/include/asm-sh/tlb.h b/include/asm-sh/tlb.h
index 53d185bcf872..56ad1fb888a2 100644
--- a/include/asm-sh/tlb.h
+++ b/include/asm-sh/tlb.h
@@ -1,6 +1,12 @@
 #ifndef __ASM_SH_TLB_H
 #define __ASM_SH_TLB_H
 
+#ifdef CONFIG_SUPERH64
+# include "tlb_64.h"
+#endif
+
+#ifndef __ASSEMBLY__
+
 #define tlb_start_vma(tlb, vma) \
 	flush_cache_range(vma, vma->vm_start, vma->vm_end)
 
@@ -15,4 +21,6 @@
 #define tlb_flush(tlb)				flush_tlb_mm((tlb)->mm)
 
 #include <asm-generic/tlb.h>
-#endif
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_SH_TLB_H */
diff --git a/include/asm-sh/tlb_64.h b/include/asm-sh/tlb_64.h
new file mode 100644
index 000000000000..0308e05fc57b
--- /dev/null
+++ b/include/asm-sh/tlb_64.h
@@ -0,0 +1,69 @@
+/*
+ * include/asm-sh/tlb_64.h
+ *
+ * Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#ifndef __ASM_SH_TLB_64_H
+#define __ASM_SH_TLB_64_H
+
+/* ITLB defines */
+#define ITLB_FIXED	0x00000000	/* First fixed ITLB, see head.S */
+#define ITLB_LAST_VAR_UNRESTRICTED	0x000003F0	/* Last ITLB */
+
+/* DTLB defines */
+#define DTLB_FIXED	0x00800000	/* First fixed DTLB, see head.S */
+#define DTLB_LAST_VAR_UNRESTRICTED	0x008003F0	/* Last DTLB */
+
+#ifndef __ASSEMBLY__
+
+/**
+ * for_each_dtlb_entry
+ *
+ * @tlb:	TLB entry
+ *
+ * Iterate over free (non-wired) DTLB entries
+ */
+#define for_each_dtlb_entry(tlb)		\
+	for (tlb  = cpu_data->dtlb.first;	\
+	     tlb <= cpu_data->dtlb.last;	\
+	     tlb += cpu_data->dtlb.step)
+
+/**
+ * for_each_itlb_entry
+ *
+ * @tlb:	TLB entry
+ *
+ * Iterate over free (non-wired) ITLB entries
+ */
+#define for_each_itlb_entry(tlb)		\
+	for (tlb  = cpu_data->itlb.first;	\
+	     tlb <= cpu_data->itlb.last;	\
+	     tlb += cpu_data->itlb.step)
+
+/**
+ * __flush_tlb_slot
+ *
+ * @slot:	Address of TLB slot.
+ *
+ * Flushes TLB slot @slot.
+ */
+static inline void __flush_tlb_slot(unsigned long long slot)
+{
+	__asm__ __volatile__ ("putcfg %0, 0, r63\n" : : "r" (slot));
+}
+
+/* arch/sh64/mm/tlb.c */
+int sh64_tlb_init(void);
+unsigned long long sh64_next_free_dtlb_entry(void);
+unsigned long long sh64_get_wired_dtlb_entry(void);
+int sh64_put_wired_dtlb_entry(unsigned long long entry);
+void sh64_setup_tlb_slot(unsigned long long config_addr, unsigned long eaddr,
+			 unsigned long asid, unsigned long paddr);
+void sh64_teardown_tlb_slot(unsigned long long config_addr);
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_SH_TLB_64_H */
diff --git a/include/asm-sh64/tlb.h b/include/asm-sh64/tlb.h
deleted file mode 100644
index 4979408bd88c..000000000000
--- a/include/asm-sh64/tlb.h
+++ /dev/null
@@ -1,92 +0,0 @@
-/*
- * include/asm-sh64/tlb.h
- *
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-#ifndef __ASM_SH64_TLB_H
-#define __ASM_SH64_TLB_H
-
-/*
- * Note! These are mostly unused, we just need the xTLB_LAST_VAR_UNRESTRICTED
- * for head.S! Once this limitation is gone, we can clean the rest of this up.
- */
-
-/* ITLB defines */
-#define ITLB_FIXED	0x00000000	/* First fixed ITLB, see head.S */
-#define ITLB_LAST_VAR_UNRESTRICTED	0x000003F0	/* Last ITLB */
-
-/* DTLB defines */
-#define DTLB_FIXED	0x00800000	/* First fixed DTLB, see head.S */
-#define DTLB_LAST_VAR_UNRESTRICTED	0x008003F0	/* Last DTLB */
-
-#ifndef __ASSEMBLY__
-
-/**
- * for_each_dtlb_entry
- *
- * @tlb:	TLB entry
- *
- * Iterate over free (non-wired) DTLB entries
- */
-#define for_each_dtlb_entry(tlb)		\
-	for (tlb  = cpu_data->dtlb.first;	\
-	     tlb <= cpu_data->dtlb.last;	\
-	     tlb += cpu_data->dtlb.step)
-
-/**
- * for_each_itlb_entry
- *
- * @tlb:	TLB entry
- *
- * Iterate over free (non-wired) ITLB entries
- */
-#define for_each_itlb_entry(tlb)		\
-	for (tlb  = cpu_data->itlb.first;	\
-	     tlb <= cpu_data->itlb.last;	\
-	     tlb += cpu_data->itlb.step)
-
-/**
- * __flush_tlb_slot
- *
- * @slot:	Address of TLB slot.
- *
- * Flushes TLB slot @slot.
- */
-static inline void __flush_tlb_slot(unsigned long long slot)
-{
-	__asm__ __volatile__ ("putcfg %0, 0, r63\n" : : "r" (slot));
-}
-
-/* arch/sh64/mm/tlb.c */
-extern int sh64_tlb_init(void);
-extern unsigned long long sh64_next_free_dtlb_entry(void);
-extern unsigned long long sh64_get_wired_dtlb_entry(void);
-extern int sh64_put_wired_dtlb_entry(unsigned long long entry);
-
-extern void sh64_setup_tlb_slot(unsigned long long config_addr, unsigned long eaddr, unsigned long asid, unsigned long paddr);
-extern void sh64_teardown_tlb_slot(unsigned long long config_addr);
-
-#define tlb_start_vma(tlb, vma) \
-	flush_cache_range(vma, vma->vm_start, vma->vm_end)
-
-#define tlb_end_vma(tlb, vma)	\
-	flush_tlb_range(vma, vma->vm_start, vma->vm_end)
-
-#define __tlb_remove_tlb_entry(tlb, pte, address)	do { } while (0)
-
-/*
- * Flush whole TLBs for MM
- */
-#define tlb_flush(tlb)		flush_tlb_mm((tlb)->mm)
-
-#include <asm-generic/tlb.h>
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* __ASM_SH64_TLB_H */
-
-- 
cgit v1.2.3


From 0468b4bb125542e75e39f08ebaa74a7daf845631 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:39:06 +0900
Subject: sh: Fix up VMALLOC_START for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pgtable.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 6ab3ba82d255..b0bb76a6864f 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -71,7 +71,11 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 
 #define PTE_PHYS_MASK		(0x20000000 - PAGE_SIZE)
 
+#ifdef CONFIG_SUPERH32
 #define VMALLOC_START	(P3SEG)
+#else
+#define VMALLOC_START	(0xf0000000)
+#endif
 #define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
 
 /*
-- 
cgit v1.2.3


From f3ef75b773b38fe2028fa4627cab3991e2c60180 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:39:32 +0900
Subject: sh: Nopped out p3_cache_init() on SH-5 also.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/cacheflush.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/cacheflush.h b/include/asm-sh/cpu-sh5/cacheflush.h
index 847374b6526f..45d12d6f89c2 100644
--- a/include/asm-sh/cpu-sh5/cacheflush.h
+++ b/include/asm-sh/cpu-sh5/cacheflush.h
@@ -27,6 +27,7 @@ extern void flush_icache_user_range(struct vm_area_struct *vma,
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 
 #define flush_icache_page(vma, page)	do { } while (0)
+#define p3_cache_init()			do { } while (0)
 
 #endif /* __ASSEMBLY__ */
 
-- 
cgit v1.2.3


From acb499f0ddfbea1176a70d93eb53943c325a073b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 10 Nov 2007 20:39:56 +0900
Subject: sh: Move vsyscall_init() defs up one level.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/processor.h    | 6 ++++++
 include/asm-sh/processor_32.h | 6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index b98c882d5036..76c4dc7021cb 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -49,4 +49,10 @@ enum cpu_type {
 /* arch/sh/kernel/setup.c */
 const char *get_cpu_subtype(struct sh_cpuinfo *c);
 
+#ifdef CONFIG_VSYSCALL
+int vsyscall_init(void);
+#else
+#define vsyscall_init() do { } while (0)
+#endif
+
 #endif /* __ASM_SH_PROCESSOR_H */
diff --git a/include/asm-sh/processor_32.h b/include/asm-sh/processor_32.h
index 35040fe43e5e..1ad74633c009 100644
--- a/include/asm-sh/processor_32.h
+++ b/include/asm-sh/processor_32.h
@@ -237,11 +237,5 @@ static inline void prefetch(void *x)
 #define prefetchw(x)	prefetch(x)
 #endif
 
-#ifdef CONFIG_VSYSCALL
-extern int vsyscall_init(void);
-#else
-#define vsyscall_init() do { } while (0)
-#endif
-
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_32_H */
-- 
cgit v1.2.3


From 8a7bcf0dd0d49fe8b0071adef0dfe8610abdffaa Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sun, 11 Nov 2007 17:07:06 +0900
Subject: sh: Add SH-5 support to the consistent DMA impl.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/consistent.c      | 46 +++++++++++++++++++++++++++-----------------
 include/asm-sh/dma-mapping.h |  4 ++--
 2 files changed, 30 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c
index e220c29a3c00..65ad30031ad7 100644
--- a/arch/sh/mm/consistent.c
+++ b/arch/sh/mm/consistent.c
@@ -1,7 +1,7 @@
 /*
  * arch/sh/mm/consistent.c
  *
- * Copyright (C) 2004  Paul Mundt
+ * Copyright (C) 2004 - 2007  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -16,7 +16,7 @@
 void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
 {
 	struct page *page, *end, *free;
-	void *ret;
+	void *ret, *vp;
 	int order;
 
 	size = PAGE_ALIGN(size);
@@ -28,13 +28,20 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
 	split_page(page, order);
 
 	ret = page_address(page);
-	memset(ret, 0, size);
 	*handle = virt_to_phys(ret);
 
+	vp = ioremap_nocache(*handle, size);
+	if (!vp) {
+		free_pages((unsigned long)ret, order);
+		return NULL;
+	}
+
+	memset(vp, 0, size);
+
 	/*
 	 * We must flush the cache before we pass it on to the device
 	 */
-	__flush_purge_region(ret, size);
+	dma_cache_sync(NULL, ret, size, DMA_BIDIRECTIONAL);
 
 	page = virt_to_page(ret);
 	free = page + (size >> PAGE_SHIFT);
@@ -47,24 +54,31 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
 		}
 	}
 
-	return P2SEGADDR(ret);
+	return vp;
 }
+EXPORT_SYMBOL(consistent_alloc);
 
-void consistent_free(void *vaddr, size_t size)
+void consistent_free(void *vaddr, size_t size, dma_addr_t dma_handle)
 {
-	unsigned long addr = P1SEGADDR((unsigned long)vaddr);
-	struct page *page=virt_to_page(addr);
-	int num_pages=(size+PAGE_SIZE-1) >> PAGE_SHIFT;
-	int i;
+	struct page *page;
+	unsigned long addr;
 
-	for(i=0;i<num_pages;i++) {
-		__free_page((page+i));
-	}
+	addr = (unsigned long)phys_to_virt((unsigned long)dma_handle);
+	page = virt_to_page(addr);
+
+	free_pages(addr, get_order(size));
+
+	iounmap(vaddr);
 }
+EXPORT_SYMBOL(consistent_free);
 
 void consistent_sync(void *vaddr, size_t size, int direction)
 {
-	void * p1addr = (void*) P1SEGADDR((unsigned long)vaddr);
+#ifdef CONFIG_CPU_SH5
+	void *p1addr = vaddr;
+#else
+	void *p1addr = (void*) P1SEGADDR((unsigned long)vaddr);
+#endif
 
 	switch (direction) {
 	case DMA_FROM_DEVICE:		/* invalidate only */
@@ -80,8 +94,4 @@ void consistent_sync(void *vaddr, size_t size, int direction)
 		BUG();
 	}
 }
-
-EXPORT_SYMBOL(consistent_alloc);
-EXPORT_SYMBOL(consistent_free);
 EXPORT_SYMBOL(consistent_sync);
-
diff --git a/include/asm-sh/dma-mapping.h b/include/asm-sh/dma-mapping.h
index fcea067f7a9c..20ae762e5258 100644
--- a/include/asm-sh/dma-mapping.h
+++ b/include/asm-sh/dma-mapping.h
@@ -10,7 +10,7 @@ extern struct bus_type pci_bus_type;
 
 /* arch/sh/mm/consistent.c */
 extern void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle);
-extern void consistent_free(void *vaddr, size_t size);
+extern void consistent_free(void *vaddr, size_t size, dma_addr_t handle);
 extern void consistent_sync(void *vaddr, size_t size, int direction);
 
 #define dma_supported(dev, mask)	(1)
@@ -50,7 +50,7 @@ static inline void dma_free_coherent(struct device *dev, size_t size,
 			return;
 	}
 
-	consistent_free(vaddr, size);
+	consistent_free(vaddr, size, dma_handle);
 }
 
 #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-- 
cgit v1.2.3


From 63e2c803326babe74033b3cbe6e8c550037c554f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sun, 11 Nov 2007 17:28:18 +0900
Subject: sh: Plug in SH-5 ffz()/__ffs() bitops.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/bitops.h | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/bitops.h b/include/asm-sh/bitops.h
index df805f20b267..a7bd81a7f064 100644
--- a/include/asm-sh/bitops.h
+++ b/include/asm-sh/bitops.h
@@ -105,6 +105,7 @@ static inline int test_and_change_bit(int nr, volatile void * addr)
 
 #include <asm-generic/bitops/non-atomic.h>
 
+#ifdef CONFIG_SUPERH32
 static inline unsigned long ffz(unsigned long word)
 {
 	unsigned long result;
@@ -138,6 +139,31 @@ static inline unsigned long __ffs(unsigned long word)
 		: "t");
 	return result;
 }
+#else
+static inline unsigned long ffz(unsigned long word)
+{
+	unsigned long result, __d2, __d3;
+
+        __asm__("gettr  tr0, %2\n\t"
+                "pta    $+32, tr0\n\t"
+                "andi   %1, 1, %3\n\t"
+                "beq    %3, r63, tr0\n\t"
+                "pta    $+4, tr0\n"
+                "0:\n\t"
+                "shlri.l        %1, 1, %1\n\t"
+                "addi   %0, 1, %0\n\t"
+                "andi   %1, 1, %3\n\t"
+                "beqi   %3, 1, tr0\n"
+                "1:\n\t"
+                "ptabs  %2, tr0\n\t"
+                : "=r" (result), "=r" (word), "=r" (__d2), "=r" (__d3)
+                : "0" (0L), "1" (word));
+
+	return result;
+}
+
+#include <asm-generic/bitops/__ffs.h>
+#endif
 
 #include <asm-generic/bitops/find.h>
 #include <asm-generic/bitops/ffs.h>
-- 
cgit v1.2.3


From fcfdd0f14f94d47f2f650a24cd111e11475ccada Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sun, 11 Nov 2007 17:36:13 +0900
Subject: sh: Split out checksum.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/sh_ksyms.c    |   2 +
 include/asm-sh/checksum.h    | 216 +------------------------------------------
 include/asm-sh/checksum_32.h | 215 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/checksum_64.h |  80 ++++++++++++++++
 include/asm-sh64/checksum.h  |  82 ----------------
 5 files changed, 300 insertions(+), 295 deletions(-)
 create mode 100644 include/asm-sh/checksum_32.h
 create mode 100644 include/asm-sh/checksum_64.h
 delete mode 100644 include/asm-sh64/checksum.h

(limited to 'include')

diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index e1a6de9088b5..bc5239ef88ce 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -141,7 +141,9 @@ EXPORT_SYMBOL(clear_user_page);
 #endif
 
 EXPORT_SYMBOL(csum_partial);
+#ifdef CONFIG_SUPERH32
 EXPORT_SYMBOL(csum_partial_copy_generic);
+#endif
 #ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
 #endif
diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h
index 4bc8357e8892..67496ab0ef04 100644
--- a/include/asm-sh/checksum.h
+++ b/include/asm-sh/checksum.h
@@ -1,215 +1,5 @@
-#ifndef __ASM_SH_CHECKSUM_H
-#define __ASM_SH_CHECKSUM_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1999 by Kaz Kojima & Niibe Yutaka
- */
-
-#include <linux/in6.h>
-
-/*
- * computes the checksum of a memory block at buff, length len,
- * and adds in "sum" (32-bit)
- *
- * returns a 32-bit number suitable for feeding into itself
- * or csum_tcpudp_magic
- *
- * this function must be called with even lengths, except
- * for the last fragment, which may be odd
- *
- * it's best to have buff aligned on a 32-bit boundary
- */
-asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum);
-
-/*
- * the same as csum_partial, but copies from src while it
- * checksums, and handles user-space pointer exceptions correctly, when needed.
- *
- * here even more important to align src and dst on a 32-bit (or even
- * better 64-bit) boundary
- */
-
-asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst,
-					    int len, __wsum sum,
-					    int *src_err_ptr, int *dst_err_ptr);
-
-/*
- *	Note: when you get a NULL pointer exception here this means someone
- *	passed in an incorrect kernel address to one of these functions.
- *
- *	If you use these functions directly please don't forget the
- *	access_ok().
- */
-static inline
-__wsum csum_partial_copy_nocheck(const void *src, void *dst,
-				 int len, __wsum sum)
-{
-	return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
-}
-
-static inline
-__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
-				   int len, __wsum sum, int *err_ptr)
-{
-	return csum_partial_copy_generic((__force const void *)src, dst,
-					len, sum, err_ptr, NULL);
-}
-
-/*
- *	Fold a partial checksum
- */
-
-static inline __sum16 csum_fold(__wsum sum)
-{
-	unsigned int __dummy;
-	__asm__("swap.w %0, %1\n\t"
-		"extu.w	%0, %0\n\t"
-		"extu.w	%1, %1\n\t"
-		"add	%1, %0\n\t"
-		"swap.w	%0, %1\n\t"
-		"add	%1, %0\n\t"
-		"not	%0, %0\n\t"
-		: "=r" (sum), "=&r" (__dummy)
-		: "0" (sum)
-		: "t");
-	return (__force __sum16)sum;
-}
-
-/*
- *	This is a version of ip_compute_csum() optimized for IP headers,
- *	which always checksum on 4 octet boundaries.
- *
- *      i386 version by Jorge Cwik <jorge@laser.satlink.net>, adapted
- *      for linux by * Arnt Gulbrandsen.
- */
-static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
-{
-	unsigned int sum, __dummy0, __dummy1;
-
-	__asm__ __volatile__(
-		"mov.l	@%1+, %0\n\t"
-		"mov.l	@%1+, %3\n\t"
-		"add	#-2, %2\n\t"
-		"clrt\n\t"
-		"1:\t"
-		"addc	%3, %0\n\t"
-		"movt	%4\n\t"
-		"mov.l	@%1+, %3\n\t"
-		"dt	%2\n\t"
-		"bf/s	1b\n\t"
-		" cmp/eq #1, %4\n\t"
-		"addc	%3, %0\n\t"
-		"addc	%2, %0"	    /* Here %2 is 0, add carry-bit */
-	/* Since the input registers which are loaded with iph and ihl
-	   are modified, we must also specify them as outputs, or gcc
-	   will assume they contain their original values. */
-	: "=r" (sum), "=r" (iph), "=r" (ihl), "=&r" (__dummy0), "=&z" (__dummy1)
-	: "1" (iph), "2" (ihl)
-	: "t");
-
-	return	csum_fold(sum);
-}
-
-static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
-					__wsum sum)
-{
-#ifdef __LITTLE_ENDIAN__
-	unsigned long len_proto = (proto + len) << 8;
+#ifdef CONFIG_SUPERH32
+# include "checksum_32.h"
 #else
-	unsigned long len_proto = proto + len;
+# include "checksum_64.h"
 #endif
-	__asm__("clrt\n\t"
-		"addc	%0, %1\n\t"
-		"addc	%2, %1\n\t"
-		"addc	%3, %1\n\t"
-		"movt	%0\n\t"
-		"add	%1, %0"
-		: "=r" (sum), "=r" (len_proto)
-		: "r" (daddr), "r" (saddr), "1" (len_proto), "0" (sum)
-		: "t");
-
-	return sum;
-}
-
-/*
- * computes the checksum of the TCP/UDP pseudo-header
- * returns a 16-bit checksum, already complemented
- */
-static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
-					__wsum sum)
-{
-	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
-}
-
-/*
- * this routine is used for miscellaneous IP-like checksums, mainly
- * in icmp.c
- */
-static inline __sum16 ip_compute_csum(const void *buff, int len)
-{
-    return csum_fold(csum_partial(buff, len, 0));
-}
-
-#define _HAVE_ARCH_IPV6_CSUM
-static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
-				      const struct in6_addr *daddr,
-				      __u32 len, unsigned short proto,
-				      __wsum sum)
-{
-	unsigned int __dummy;
-	__asm__("clrt\n\t"
-		"mov.l	@(0,%2), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(4,%2), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(8,%2), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(12,%2), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(0,%3), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(4,%3), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(8,%3), %1\n\t"
-		"addc	%1, %0\n\t"
-		"mov.l	@(12,%3), %1\n\t"
-		"addc	%1, %0\n\t"
-		"addc	%4, %0\n\t"
-		"addc	%5, %0\n\t"
-		"movt	%1\n\t"
-		"add	%1, %0\n"
-		: "=r" (sum), "=&r" (__dummy)
-		: "r" (saddr), "r" (daddr),
-		  "r" (htonl(len)), "r" (htonl(proto)), "0" (sum)
-		: "t");
-
-	return csum_fold(sum);
-}
-
-/*
- *	Copy and checksum to user
- */
-#define HAVE_CSUM_COPY_USER
-static inline __wsum csum_and_copy_to_user(const void *src,
-					   void __user *dst,
-					   int len, __wsum sum,
-					   int *err_ptr)
-{
-	if (access_ok(VERIFY_WRITE, dst, len))
-		return csum_partial_copy_generic((__force const void *)src,
-						dst, len, sum, NULL, err_ptr);
-
-	if (len)
-		*err_ptr = -EFAULT;
-
-	return (__force __wsum)-1; /* invalid checksum */
-}
-#endif /* __ASM_SH_CHECKSUM_H */
diff --git a/include/asm-sh/checksum_32.h b/include/asm-sh/checksum_32.h
new file mode 100644
index 000000000000..4bc8357e8892
--- /dev/null
+++ b/include/asm-sh/checksum_32.h
@@ -0,0 +1,215 @@
+#ifndef __ASM_SH_CHECKSUM_H
+#define __ASM_SH_CHECKSUM_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1999 by Kaz Kojima & Niibe Yutaka
+ */
+
+#include <linux/in6.h>
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum);
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums, and handles user-space pointer exceptions correctly, when needed.
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+
+asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst,
+					    int len, __wsum sum,
+					    int *src_err_ptr, int *dst_err_ptr);
+
+/*
+ *	Note: when you get a NULL pointer exception here this means someone
+ *	passed in an incorrect kernel address to one of these functions.
+ *
+ *	If you use these functions directly please don't forget the
+ *	access_ok().
+ */
+static inline
+__wsum csum_partial_copy_nocheck(const void *src, void *dst,
+				 int len, __wsum sum)
+{
+	return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
+}
+
+static inline
+__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
+				   int len, __wsum sum, int *err_ptr)
+{
+	return csum_partial_copy_generic((__force const void *)src, dst,
+					len, sum, err_ptr, NULL);
+}
+
+/*
+ *	Fold a partial checksum
+ */
+
+static inline __sum16 csum_fold(__wsum sum)
+{
+	unsigned int __dummy;
+	__asm__("swap.w %0, %1\n\t"
+		"extu.w	%0, %0\n\t"
+		"extu.w	%1, %1\n\t"
+		"add	%1, %0\n\t"
+		"swap.w	%0, %1\n\t"
+		"add	%1, %0\n\t"
+		"not	%0, %0\n\t"
+		: "=r" (sum), "=&r" (__dummy)
+		: "0" (sum)
+		: "t");
+	return (__force __sum16)sum;
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ *
+ *      i386 version by Jorge Cwik <jorge@laser.satlink.net>, adapted
+ *      for linux by * Arnt Gulbrandsen.
+ */
+static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
+{
+	unsigned int sum, __dummy0, __dummy1;
+
+	__asm__ __volatile__(
+		"mov.l	@%1+, %0\n\t"
+		"mov.l	@%1+, %3\n\t"
+		"add	#-2, %2\n\t"
+		"clrt\n\t"
+		"1:\t"
+		"addc	%3, %0\n\t"
+		"movt	%4\n\t"
+		"mov.l	@%1+, %3\n\t"
+		"dt	%2\n\t"
+		"bf/s	1b\n\t"
+		" cmp/eq #1, %4\n\t"
+		"addc	%3, %0\n\t"
+		"addc	%2, %0"	    /* Here %2 is 0, add carry-bit */
+	/* Since the input registers which are loaded with iph and ihl
+	   are modified, we must also specify them as outputs, or gcc
+	   will assume they contain their original values. */
+	: "=r" (sum), "=r" (iph), "=r" (ihl), "=&r" (__dummy0), "=&z" (__dummy1)
+	: "1" (iph), "2" (ihl)
+	: "t");
+
+	return	csum_fold(sum);
+}
+
+static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+					unsigned short len,
+					unsigned short proto,
+					__wsum sum)
+{
+#ifdef __LITTLE_ENDIAN__
+	unsigned long len_proto = (proto + len) << 8;
+#else
+	unsigned long len_proto = proto + len;
+#endif
+	__asm__("clrt\n\t"
+		"addc	%0, %1\n\t"
+		"addc	%2, %1\n\t"
+		"addc	%3, %1\n\t"
+		"movt	%0\n\t"
+		"add	%1, %0"
+		: "=r" (sum), "=r" (len_proto)
+		: "r" (daddr), "r" (saddr), "1" (len_proto), "0" (sum)
+		: "t");
+
+	return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+					unsigned short len,
+					unsigned short proto,
+					__wsum sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+static inline __sum16 ip_compute_csum(const void *buff, int len)
+{
+    return csum_fold(csum_partial(buff, len, 0));
+}
+
+#define _HAVE_ARCH_IPV6_CSUM
+static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
+				      const struct in6_addr *daddr,
+				      __u32 len, unsigned short proto,
+				      __wsum sum)
+{
+	unsigned int __dummy;
+	__asm__("clrt\n\t"
+		"mov.l	@(0,%2), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(4,%2), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(8,%2), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(12,%2), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(0,%3), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(4,%3), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(8,%3), %1\n\t"
+		"addc	%1, %0\n\t"
+		"mov.l	@(12,%3), %1\n\t"
+		"addc	%1, %0\n\t"
+		"addc	%4, %0\n\t"
+		"addc	%5, %0\n\t"
+		"movt	%1\n\t"
+		"add	%1, %0\n"
+		: "=r" (sum), "=&r" (__dummy)
+		: "r" (saddr), "r" (daddr),
+		  "r" (htonl(len)), "r" (htonl(proto)), "0" (sum)
+		: "t");
+
+	return csum_fold(sum);
+}
+
+/*
+ *	Copy and checksum to user
+ */
+#define HAVE_CSUM_COPY_USER
+static inline __wsum csum_and_copy_to_user(const void *src,
+					   void __user *dst,
+					   int len, __wsum sum,
+					   int *err_ptr)
+{
+	if (access_ok(VERIFY_WRITE, dst, len))
+		return csum_partial_copy_generic((__force const void *)src,
+						dst, len, sum, NULL, err_ptr);
+
+	if (len)
+		*err_ptr = -EFAULT;
+
+	return (__force __wsum)-1; /* invalid checksum */
+}
+#endif /* __ASM_SH_CHECKSUM_H */
diff --git a/include/asm-sh/checksum_64.h b/include/asm-sh/checksum_64.h
new file mode 100644
index 000000000000..ec79b92e6bdc
--- /dev/null
+++ b/include/asm-sh/checksum_64.h
@@ -0,0 +1,80 @@
+#ifndef __ASM_SH64_CHECKSUM_H
+#define __ASM_SH64_CHECKSUM_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/checksum.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ *
+ */
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum);
+
+/*
+ *	Note: when you get a NULL pointer exception here this means someone
+ *	passed in an incorrect kernel address to one of these functions.
+ *
+ *	If you use these functions directly please don't forget the
+ *	access_ok().
+ */
+
+
+__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len,
+				       __wsum sum);
+
+__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
+					 int len, __wsum sum, int *err_ptr);
+
+static inline __sum16 csum_fold(__wsum csum)
+{
+	u32 sum = (__force u32)csum;
+        sum = (sum & 0xffff) + (sum >> 16);
+        sum = (sum & 0xffff) + (sum >> 16);
+        return (__force __sum16)~sum;
+}
+
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
+
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+				 unsigned short len, unsigned short proto,
+				 __wsum sum);
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+						   unsigned short len,
+						   unsigned short proto,
+						   __wsum sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+static inline __sum16 ip_compute_csum(const void *buff, int len)
+{
+	return csum_fold(csum_partial(buff, len, 0));
+}
+
+#endif /* __ASM_SH64_CHECKSUM_H */
+
diff --git a/include/asm-sh64/checksum.h b/include/asm-sh64/checksum.h
deleted file mode 100644
index ba594ccb42e5..000000000000
--- a/include/asm-sh64/checksum.h
+++ /dev/null
@@ -1,82 +0,0 @@
-#ifndef __ASM_SH64_CHECKSUM_H
-#define __ASM_SH64_CHECKSUM_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/checksum.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <asm/registers.h>
-
-/*
- * computes the checksum of a memory block at buff, length len,
- * and adds in "sum" (32-bit)
- *
- * returns a 32-bit number suitable for feeding into itself
- * or csum_tcpudp_magic
- *
- * this function must be called with even lengths, except
- * for the last fragment, which may be odd
- *
- * it's best to have buff aligned on a 32-bit boundary
- */
-asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum);
-
-/*
- *	Note: when you get a NULL pointer exception here this means someone
- *	passed in an incorrect kernel address to one of these functions.
- *
- *	If you use these functions directly please don't forget the
- *	access_ok().
- */
-
-
-__wsum csum_partial_copy_nocheck(const void *src, void *dst, int len,
-				       __wsum sum);
-
-__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
-					 int len, __wsum sum, int *err_ptr);
-
-static inline __sum16 csum_fold(__wsum csum)
-{
-	u32 sum = (__force u32)csum;
-        sum = (sum & 0xffff) + (sum >> 16);
-        sum = (sum & 0xffff) + (sum >> 16);
-        return (__force __sum16)~sum;
-}
-
-__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
-
-__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-				 unsigned short len, unsigned short proto,
-				 __wsum sum);
-
-/*
- * computes the checksum of the TCP/UDP pseudo-header
- * returns a 16-bit checksum, already complemented
- */
-static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
-{
-	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
-}
-
-/*
- * this routine is used for miscellaneous IP-like checksums, mainly
- * in icmp.c
- */
-static inline __sum16 ip_compute_csum(const void *buff, int len)
-{
-	return csum_fold(csum_partial(buff, len, 0));
-}
-
-#endif /* __ASM_SH64_CHECKSUM_H */
-
-- 
cgit v1.2.3


From 1b6cf8175e52cbda111ab8b590da6a1d093e9954 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sun, 11 Nov 2007 17:46:49 +0900
Subject: sh: Split out syscall ABI for _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile_32   |   4 +-
 arch/sh/kernel/Makefile_64   |   4 +-
 arch/sh/kernel/syscalls.S    | 343 -----------------------------------
 arch/sh/kernel/syscalls_32.S | 343 +++++++++++++++++++++++++++++++++++
 arch/sh/kernel/syscalls_64.S | 381 +++++++++++++++++++++++++++++++++++++++
 arch/sh64/kernel/syscalls.S  | 381 ---------------------------------------
 include/asm-sh/unistd.h      | 379 +--------------------------------------
 include/asm-sh/unistd_32.h   | 376 ++++++++++++++++++++++++++++++++++++++
 include/asm-sh/unistd_64.h   | 417 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh64/unistd.h    | 417 -------------------------------------------
 10 files changed, 1525 insertions(+), 1520 deletions(-)
 delete mode 100644 arch/sh/kernel/syscalls.S
 create mode 100644 arch/sh/kernel/syscalls_32.S
 create mode 100644 arch/sh/kernel/syscalls_64.S
 delete mode 100644 arch/sh64/kernel/syscalls.S
 create mode 100644 include/asm-sh/unistd_32.h
 create mode 100644 include/asm-sh/unistd_64.h
 delete mode 100644 include/asm-sh64/unistd.h

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 35be5e9674f8..c5a85bfbd451 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -5,8 +5,8 @@
 extra-y	:= head_32.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
-	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o syscalls.o \
-	   time.o topology.o traps_32.o
+	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o \
+	   syscalls_32.o time.o topology.o traps_32.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index 6e72ed4c07fb..41b3efc25686 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -1,8 +1,8 @@
 extra-y	:= head_64.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
-	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh64.o syscalls.o \
-	   time.o topology.o traps_64.o
+	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh64.o \
+	   syscalls_64.o time.o topology.o traps_64.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
deleted file mode 100644
index 10bec45415ba..000000000000
--- a/arch/sh/kernel/syscalls.S
+++ /dev/null
@@ -1,343 +0,0 @@
-/*
- * arch/sh/kernel/syscalls.S
- *
- * System call table for SuperH
- *
- *  Copyright (C) 1999, 2000, 2002  Niibe Yutaka
- *  Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-#include <linux/sys.h>
-#include <linux/linkage.h>
-
-	.data
-ENTRY(sys_call_table)
-	.long sys_restart_syscall	/* 0  -  old "setup()" system call*/
-	.long sys_exit
-	.long sys_fork
-	.long sys_read
-	.long sys_write
-	.long sys_open		/* 5 */
-	.long sys_close
-	.long sys_waitpid
-	.long sys_creat
-	.long sys_link
-	.long sys_unlink		/* 10 */
-	.long sys_execve
-	.long sys_chdir
-	.long sys_time
-	.long sys_mknod
-	.long sys_chmod		/* 15 */
-	.long sys_lchown16
-	.long sys_ni_syscall	/* old break syscall holder */
-	.long sys_stat
-	.long sys_lseek
-	.long sys_getpid		/* 20 */
-	.long sys_mount
-	.long sys_oldumount
-	.long sys_setuid16
-	.long sys_getuid16
-	.long sys_stime		/* 25 */
-	.long sys_ptrace
-	.long sys_alarm
-	.long sys_fstat
-	.long sys_pause
-	.long sys_utime		/* 30 */
-	.long sys_ni_syscall	/* old stty syscall holder */
-	.long sys_ni_syscall	/* old gtty syscall holder */
-	.long sys_access
-	.long sys_nice
-	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
-	.long sys_sync
-	.long sys_kill
-	.long sys_rename
-	.long sys_mkdir
-	.long sys_rmdir		/* 40 */
-	.long sys_dup
-	.long sys_pipe
-	.long sys_times
-	.long sys_ni_syscall	/* old prof syscall holder */
-	.long sys_brk		/* 45 */
-	.long sys_setgid16
-	.long sys_getgid16
-	.long sys_signal
-	.long sys_geteuid16
-	.long sys_getegid16	/* 50 */
-	.long sys_acct
-	.long sys_umount		/* recycled never used phys() */
-	.long sys_ni_syscall	/* old lock syscall holder */
-	.long sys_ioctl
-	.long sys_fcntl		/* 55 */
-	.long sys_ni_syscall	/* old mpx syscall holder */
-	.long sys_setpgid
-	.long sys_ni_syscall	/* old ulimit syscall holder */
-	.long sys_ni_syscall	/* sys_olduname */
-	.long sys_umask		/* 60 */
-	.long sys_chroot
-	.long sys_ustat
-	.long sys_dup2
-	.long sys_getppid
-	.long sys_getpgrp		/* 65 */
-	.long sys_setsid
-	.long sys_sigaction
-	.long sys_sgetmask
-	.long sys_ssetmask
-	.long sys_setreuid16	/* 70 */
-	.long sys_setregid16
-	.long sys_sigsuspend
-	.long sys_sigpending
-	.long sys_sethostname
-	.long sys_setrlimit	/* 75 */
-	.long sys_old_getrlimit
-	.long sys_getrusage
-	.long sys_gettimeofday
-	.long sys_settimeofday
-	.long sys_getgroups16	/* 80 */
-	.long sys_setgroups16
-	.long sys_ni_syscall	/* sys_oldselect */
-	.long sys_symlink
-	.long sys_lstat
-	.long sys_readlink		/* 85 */
-	.long sys_uselib
-	.long sys_swapon
-	.long sys_reboot
-	.long old_readdir
-	.long old_mmap		/* 90 */
-	.long sys_munmap
-	.long sys_truncate
-	.long sys_ftruncate
-	.long sys_fchmod
-	.long sys_fchown16		/* 95 */
-	.long sys_getpriority
-	.long sys_setpriority
-	.long sys_ni_syscall	/* old profil syscall holder */
-	.long sys_statfs
-	.long sys_fstatfs		/* 100 */
-	.long sys_ni_syscall	/* ioperm */
-	.long sys_socketcall
-	.long sys_syslog
-	.long sys_setitimer
-	.long sys_getitimer	/* 105 */
-	.long sys_newstat
-	.long sys_newlstat
-	.long sys_newfstat
-	.long sys_uname
-	.long sys_ni_syscall	/* 110 */ /* iopl */
-	.long sys_vhangup
-	.long sys_ni_syscall	/* idle */
-	.long sys_ni_syscall	/* vm86old */
-	.long sys_wait4
-	.long sys_swapoff		/* 115 */
-	.long sys_sysinfo
-	.long sys_ipc
-	.long sys_fsync
-	.long sys_sigreturn
-	.long sys_clone		/* 120 */
-	.long sys_setdomainname
-	.long sys_newuname
-	.long sys_ni_syscall	/* sys_modify_ldt */
-	.long sys_adjtimex
-	.long sys_mprotect		/* 125 */
-	.long sys_sigprocmask
-	.long sys_ni_syscall	/* old "create_module" */
-	.long sys_init_module
-	.long sys_delete_module
-	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
-	.long sys_quotactl
-	.long sys_getpgid
-	.long sys_fchdir
-	.long sys_bdflush
-	.long sys_sysfs		/* 135 */
-	.long sys_personality
-	.long sys_ni_syscall	/* for afs_syscall */
-	.long sys_setfsuid16
-	.long sys_setfsgid16
-	.long sys_llseek		/* 140 */
-	.long sys_getdents
-	.long sys_select
-	.long sys_flock
-	.long sys_msync
-	.long sys_readv		/* 145 */
-	.long sys_writev
-	.long sys_getsid
-	.long sys_fdatasync
-	.long sys_sysctl
-	.long sys_mlock		/* 150 */
-	.long sys_munlock
-	.long sys_mlockall
-	.long sys_munlockall
-	.long sys_sched_setparam
-	.long sys_sched_getparam   /* 155 */
-	.long sys_sched_setscheduler
-	.long sys_sched_getscheduler
-	.long sys_sched_yield
-	.long sys_sched_get_priority_max
-	.long sys_sched_get_priority_min  /* 160 */
-	.long sys_sched_rr_get_interval
-	.long sys_nanosleep
-	.long sys_mremap
-	.long sys_setresuid16
-	.long sys_getresuid16	/* 165 */
-	.long sys_ni_syscall	/* vm86 */
-	.long sys_ni_syscall	/* old "query_module" */
-	.long sys_poll
-	.long sys_nfsservctl
-	.long sys_setresgid16	/* 170 */
-	.long sys_getresgid16
-	.long sys_prctl
-	.long sys_rt_sigreturn
-	.long sys_rt_sigaction
-	.long sys_rt_sigprocmask	/* 175 */
-	.long sys_rt_sigpending
-	.long sys_rt_sigtimedwait
-	.long sys_rt_sigqueueinfo
-	.long sys_rt_sigsuspend
-	.long sys_pread_wrapper	   /* 180 */
-	.long sys_pwrite_wrapper
-	.long sys_chown16
-	.long sys_getcwd
-	.long sys_capget
-	.long sys_capset           /* 185 */
-	.long sys_sigaltstack
-	.long sys_sendfile
-	.long sys_ni_syscall	/* streams1 */
-	.long sys_ni_syscall	/* streams2 */
-	.long sys_vfork            /* 190 */
-	.long sys_getrlimit
-	.long sys_mmap2
-	.long sys_truncate64
-	.long sys_ftruncate64
-	.long sys_stat64		/* 195 */
-	.long sys_lstat64
-	.long sys_fstat64
-	.long sys_lchown
-	.long sys_getuid
-	.long sys_getgid		/* 200 */
-	.long sys_geteuid
-	.long sys_getegid
-	.long sys_setreuid
-	.long sys_setregid
-	.long sys_getgroups	/* 205 */
-	.long sys_setgroups
-	.long sys_fchown
-	.long sys_setresuid
-	.long sys_getresuid
-	.long sys_setresgid	/* 210 */
-	.long sys_getresgid
-	.long sys_chown
-	.long sys_setuid
-	.long sys_setgid
-	.long sys_setfsuid		/* 215 */
-	.long sys_setfsgid
-	.long sys_pivot_root
-	.long sys_mincore
-	.long sys_madvise
-	.long sys_getdents64	/* 220 */
-	.long sys_fcntl64
-	.long sys_ni_syscall	/* reserved for TUX */
-	.long sys_ni_syscall	/* Reserved for Security */
-	.long sys_gettid
-	.long sys_readahead	/* 225 */
-	.long sys_setxattr
-	.long sys_lsetxattr
-	.long sys_fsetxattr
-	.long sys_getxattr
-	.long sys_lgetxattr	/* 230 */
-	.long sys_fgetxattr
-	.long sys_listxattr
-	.long sys_llistxattr
-	.long sys_flistxattr
-	.long sys_removexattr	/* 235 */
-	.long sys_lremovexattr
-	.long sys_fremovexattr
-	.long sys_tkill
-	.long sys_sendfile64
-	.long sys_futex		/* 240 */
-	.long sys_sched_setaffinity
-	.long sys_sched_getaffinity
-	.long sys_ni_syscall
-	.long sys_ni_syscall
-	.long sys_io_setup	/* 245 */
-	.long sys_io_destroy
-	.long sys_io_getevents
-	.long sys_io_submit
-	.long sys_io_cancel
-	.long sys_fadvise64	/* 250 */
-	.long sys_ni_syscall
-	.long sys_exit_group
-	.long sys_lookup_dcookie
-	.long sys_epoll_create
-	.long sys_epoll_ctl	/* 255 */
-	.long sys_epoll_wait
- 	.long sys_remap_file_pages
- 	.long sys_set_tid_address
- 	.long sys_timer_create
- 	.long sys_timer_settime		/* 260 */
- 	.long sys_timer_gettime
- 	.long sys_timer_getoverrun
- 	.long sys_timer_delete
- 	.long sys_clock_settime
- 	.long sys_clock_gettime		/* 265 */
- 	.long sys_clock_getres
- 	.long sys_clock_nanosleep
-	.long sys_statfs64
-	.long sys_fstatfs64
-	.long sys_tgkill		/* 270 */
-	.long sys_utimes
- 	.long sys_fadvise64_64_wrapper
-	.long sys_ni_syscall	/* Reserved for vserver */
-	.long sys_mbind
-	.long sys_get_mempolicy		/* 275 */
-	.long sys_set_mempolicy
-	.long sys_mq_open
-	.long sys_mq_unlink
-	.long sys_mq_timedsend
-	.long sys_mq_timedreceive       /* 280 */
-	.long sys_mq_notify
-	.long sys_mq_getsetattr
-	.long sys_kexec_load
-	.long sys_waitid
-	.long sys_add_key		/* 285 */
-	.long sys_request_key
-	.long sys_keyctl
-	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init		/* 290 */
-	.long sys_inotify_add_watch
-	.long sys_inotify_rm_watch
-	.long sys_ni_syscall
-	.long sys_migrate_pages
-	.long sys_openat		/* 295 */
-	.long sys_mkdirat
-	.long sys_mknodat
-	.long sys_fchownat
-	.long sys_futimesat
-	.long sys_fstatat64		/* 300 */
-	.long sys_unlinkat
-	.long sys_renameat
-	.long sys_linkat
-	.long sys_symlinkat
-	.long sys_readlinkat		/* 305 */
-	.long sys_fchmodat
-	.long sys_faccessat
-	.long sys_pselect6
-	.long sys_ppoll
-	.long sys_unshare		/* 310 */
-	.long sys_set_robust_list
-	.long sys_get_robust_list
-	.long sys_splice
-	.long sys_sync_file_range
-	.long sys_tee			/* 315 */
-	.long sys_vmsplice
-	.long sys_move_pages
-	.long sys_getcpu
-	.long sys_epoll_pwait
-	.long sys_utimensat		/* 320 */
-	.long sys_signalfd
-	.long sys_timerfd
-	.long sys_eventfd
-	.long sys_fallocate
diff --git a/arch/sh/kernel/syscalls_32.S b/arch/sh/kernel/syscalls_32.S
new file mode 100644
index 000000000000..10bec45415ba
--- /dev/null
+++ b/arch/sh/kernel/syscalls_32.S
@@ -0,0 +1,343 @@
+/*
+ * arch/sh/kernel/syscalls.S
+ *
+ * System call table for SuperH
+ *
+ *  Copyright (C) 1999, 2000, 2002  Niibe Yutaka
+ *  Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+#include <linux/sys.h>
+#include <linux/linkage.h>
+
+	.data
+ENTRY(sys_call_table)
+	.long sys_restart_syscall	/* 0  -  old "setup()" system call*/
+	.long sys_exit
+	.long sys_fork
+	.long sys_read
+	.long sys_write
+	.long sys_open		/* 5 */
+	.long sys_close
+	.long sys_waitpid
+	.long sys_creat
+	.long sys_link
+	.long sys_unlink		/* 10 */
+	.long sys_execve
+	.long sys_chdir
+	.long sys_time
+	.long sys_mknod
+	.long sys_chmod		/* 15 */
+	.long sys_lchown16
+	.long sys_ni_syscall	/* old break syscall holder */
+	.long sys_stat
+	.long sys_lseek
+	.long sys_getpid		/* 20 */
+	.long sys_mount
+	.long sys_oldumount
+	.long sys_setuid16
+	.long sys_getuid16
+	.long sys_stime		/* 25 */
+	.long sys_ptrace
+	.long sys_alarm
+	.long sys_fstat
+	.long sys_pause
+	.long sys_utime		/* 30 */
+	.long sys_ni_syscall	/* old stty syscall holder */
+	.long sys_ni_syscall	/* old gtty syscall holder */
+	.long sys_access
+	.long sys_nice
+	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
+	.long sys_sync
+	.long sys_kill
+	.long sys_rename
+	.long sys_mkdir
+	.long sys_rmdir		/* 40 */
+	.long sys_dup
+	.long sys_pipe
+	.long sys_times
+	.long sys_ni_syscall	/* old prof syscall holder */
+	.long sys_brk		/* 45 */
+	.long sys_setgid16
+	.long sys_getgid16
+	.long sys_signal
+	.long sys_geteuid16
+	.long sys_getegid16	/* 50 */
+	.long sys_acct
+	.long sys_umount		/* recycled never used phys() */
+	.long sys_ni_syscall	/* old lock syscall holder */
+	.long sys_ioctl
+	.long sys_fcntl		/* 55 */
+	.long sys_ni_syscall	/* old mpx syscall holder */
+	.long sys_setpgid
+	.long sys_ni_syscall	/* old ulimit syscall holder */
+	.long sys_ni_syscall	/* sys_olduname */
+	.long sys_umask		/* 60 */
+	.long sys_chroot
+	.long sys_ustat
+	.long sys_dup2
+	.long sys_getppid
+	.long sys_getpgrp		/* 65 */
+	.long sys_setsid
+	.long sys_sigaction
+	.long sys_sgetmask
+	.long sys_ssetmask
+	.long sys_setreuid16	/* 70 */
+	.long sys_setregid16
+	.long sys_sigsuspend
+	.long sys_sigpending
+	.long sys_sethostname
+	.long sys_setrlimit	/* 75 */
+	.long sys_old_getrlimit
+	.long sys_getrusage
+	.long sys_gettimeofday
+	.long sys_settimeofday
+	.long sys_getgroups16	/* 80 */
+	.long sys_setgroups16
+	.long sys_ni_syscall	/* sys_oldselect */
+	.long sys_symlink
+	.long sys_lstat
+	.long sys_readlink		/* 85 */
+	.long sys_uselib
+	.long sys_swapon
+	.long sys_reboot
+	.long old_readdir
+	.long old_mmap		/* 90 */
+	.long sys_munmap
+	.long sys_truncate
+	.long sys_ftruncate
+	.long sys_fchmod
+	.long sys_fchown16		/* 95 */
+	.long sys_getpriority
+	.long sys_setpriority
+	.long sys_ni_syscall	/* old profil syscall holder */
+	.long sys_statfs
+	.long sys_fstatfs		/* 100 */
+	.long sys_ni_syscall	/* ioperm */
+	.long sys_socketcall
+	.long sys_syslog
+	.long sys_setitimer
+	.long sys_getitimer	/* 105 */
+	.long sys_newstat
+	.long sys_newlstat
+	.long sys_newfstat
+	.long sys_uname
+	.long sys_ni_syscall	/* 110 */ /* iopl */
+	.long sys_vhangup
+	.long sys_ni_syscall	/* idle */
+	.long sys_ni_syscall	/* vm86old */
+	.long sys_wait4
+	.long sys_swapoff		/* 115 */
+	.long sys_sysinfo
+	.long sys_ipc
+	.long sys_fsync
+	.long sys_sigreturn
+	.long sys_clone		/* 120 */
+	.long sys_setdomainname
+	.long sys_newuname
+	.long sys_ni_syscall	/* sys_modify_ldt */
+	.long sys_adjtimex
+	.long sys_mprotect		/* 125 */
+	.long sys_sigprocmask
+	.long sys_ni_syscall	/* old "create_module" */
+	.long sys_init_module
+	.long sys_delete_module
+	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
+	.long sys_quotactl
+	.long sys_getpgid
+	.long sys_fchdir
+	.long sys_bdflush
+	.long sys_sysfs		/* 135 */
+	.long sys_personality
+	.long sys_ni_syscall	/* for afs_syscall */
+	.long sys_setfsuid16
+	.long sys_setfsgid16
+	.long sys_llseek		/* 140 */
+	.long sys_getdents
+	.long sys_select
+	.long sys_flock
+	.long sys_msync
+	.long sys_readv		/* 145 */
+	.long sys_writev
+	.long sys_getsid
+	.long sys_fdatasync
+	.long sys_sysctl
+	.long sys_mlock		/* 150 */
+	.long sys_munlock
+	.long sys_mlockall
+	.long sys_munlockall
+	.long sys_sched_setparam
+	.long sys_sched_getparam   /* 155 */
+	.long sys_sched_setscheduler
+	.long sys_sched_getscheduler
+	.long sys_sched_yield
+	.long sys_sched_get_priority_max
+	.long sys_sched_get_priority_min  /* 160 */
+	.long sys_sched_rr_get_interval
+	.long sys_nanosleep
+	.long sys_mremap
+	.long sys_setresuid16
+	.long sys_getresuid16	/* 165 */
+	.long sys_ni_syscall	/* vm86 */
+	.long sys_ni_syscall	/* old "query_module" */
+	.long sys_poll
+	.long sys_nfsservctl
+	.long sys_setresgid16	/* 170 */
+	.long sys_getresgid16
+	.long sys_prctl
+	.long sys_rt_sigreturn
+	.long sys_rt_sigaction
+	.long sys_rt_sigprocmask	/* 175 */
+	.long sys_rt_sigpending
+	.long sys_rt_sigtimedwait
+	.long sys_rt_sigqueueinfo
+	.long sys_rt_sigsuspend
+	.long sys_pread_wrapper	   /* 180 */
+	.long sys_pwrite_wrapper
+	.long sys_chown16
+	.long sys_getcwd
+	.long sys_capget
+	.long sys_capset           /* 185 */
+	.long sys_sigaltstack
+	.long sys_sendfile
+	.long sys_ni_syscall	/* streams1 */
+	.long sys_ni_syscall	/* streams2 */
+	.long sys_vfork            /* 190 */
+	.long sys_getrlimit
+	.long sys_mmap2
+	.long sys_truncate64
+	.long sys_ftruncate64
+	.long sys_stat64		/* 195 */
+	.long sys_lstat64
+	.long sys_fstat64
+	.long sys_lchown
+	.long sys_getuid
+	.long sys_getgid		/* 200 */
+	.long sys_geteuid
+	.long sys_getegid
+	.long sys_setreuid
+	.long sys_setregid
+	.long sys_getgroups	/* 205 */
+	.long sys_setgroups
+	.long sys_fchown
+	.long sys_setresuid
+	.long sys_getresuid
+	.long sys_setresgid	/* 210 */
+	.long sys_getresgid
+	.long sys_chown
+	.long sys_setuid
+	.long sys_setgid
+	.long sys_setfsuid		/* 215 */
+	.long sys_setfsgid
+	.long sys_pivot_root
+	.long sys_mincore
+	.long sys_madvise
+	.long sys_getdents64	/* 220 */
+	.long sys_fcntl64
+	.long sys_ni_syscall	/* reserved for TUX */
+	.long sys_ni_syscall	/* Reserved for Security */
+	.long sys_gettid
+	.long sys_readahead	/* 225 */
+	.long sys_setxattr
+	.long sys_lsetxattr
+	.long sys_fsetxattr
+	.long sys_getxattr
+	.long sys_lgetxattr	/* 230 */
+	.long sys_fgetxattr
+	.long sys_listxattr
+	.long sys_llistxattr
+	.long sys_flistxattr
+	.long sys_removexattr	/* 235 */
+	.long sys_lremovexattr
+	.long sys_fremovexattr
+	.long sys_tkill
+	.long sys_sendfile64
+	.long sys_futex		/* 240 */
+	.long sys_sched_setaffinity
+	.long sys_sched_getaffinity
+	.long sys_ni_syscall
+	.long sys_ni_syscall
+	.long sys_io_setup	/* 245 */
+	.long sys_io_destroy
+	.long sys_io_getevents
+	.long sys_io_submit
+	.long sys_io_cancel
+	.long sys_fadvise64	/* 250 */
+	.long sys_ni_syscall
+	.long sys_exit_group
+	.long sys_lookup_dcookie
+	.long sys_epoll_create
+	.long sys_epoll_ctl	/* 255 */
+	.long sys_epoll_wait
+ 	.long sys_remap_file_pages
+ 	.long sys_set_tid_address
+ 	.long sys_timer_create
+ 	.long sys_timer_settime		/* 260 */
+ 	.long sys_timer_gettime
+ 	.long sys_timer_getoverrun
+ 	.long sys_timer_delete
+ 	.long sys_clock_settime
+ 	.long sys_clock_gettime		/* 265 */
+ 	.long sys_clock_getres
+ 	.long sys_clock_nanosleep
+	.long sys_statfs64
+	.long sys_fstatfs64
+	.long sys_tgkill		/* 270 */
+	.long sys_utimes
+ 	.long sys_fadvise64_64_wrapper
+	.long sys_ni_syscall	/* Reserved for vserver */
+	.long sys_mbind
+	.long sys_get_mempolicy		/* 275 */
+	.long sys_set_mempolicy
+	.long sys_mq_open
+	.long sys_mq_unlink
+	.long sys_mq_timedsend
+	.long sys_mq_timedreceive       /* 280 */
+	.long sys_mq_notify
+	.long sys_mq_getsetattr
+	.long sys_kexec_load
+	.long sys_waitid
+	.long sys_add_key		/* 285 */
+	.long sys_request_key
+	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get
+	.long sys_inotify_init		/* 290 */
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_ni_syscall
+	.long sys_migrate_pages
+	.long sys_openat		/* 295 */
+	.long sys_mkdirat
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64		/* 300 */
+	.long sys_unlinkat
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
+	.long sys_faccessat
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare		/* 310 */
+	.long sys_set_robust_list
+	.long sys_get_robust_list
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee			/* 315 */
+	.long sys_vmsplice
+	.long sys_move_pages
+	.long sys_getcpu
+	.long sys_epoll_pwait
+	.long sys_utimensat		/* 320 */
+	.long sys_signalfd
+	.long sys_timerfd
+	.long sys_eventfd
+	.long sys_fallocate
diff --git a/arch/sh/kernel/syscalls_64.S b/arch/sh/kernel/syscalls_64.S
new file mode 100644
index 000000000000..abb94c05d07a
--- /dev/null
+++ b/arch/sh/kernel/syscalls_64.S
@@ -0,0 +1,381 @@
+/*
+ * arch/sh64/kernel/syscalls.S
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2004 - 2007  Paul Mundt
+ * Copyright (C) 2003, 2004 Richard Curnow
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/sys.h>
+
+	.section .data, "aw"
+	.balign 32
+
+/*
+ * System calls jump table
+ */
+	.globl  sys_call_table
+sys_call_table:
+	.long sys_restart_syscall	/* 0  -  old "setup()" system call  */
+	.long sys_exit
+	.long sys_fork
+	.long sys_read
+	.long sys_write
+	.long sys_open			/* 5 */
+	.long sys_close
+	.long sys_waitpid
+	.long sys_creat
+	.long sys_link
+	.long sys_unlink		/* 10 */
+	.long sys_execve
+	.long sys_chdir
+	.long sys_time
+	.long sys_mknod
+	.long sys_chmod			/* 15 */
+	.long sys_lchown16
+	.long sys_ni_syscall	/* old break syscall holder */
+	.long sys_stat
+	.long sys_lseek
+	.long sys_getpid		/* 20 */
+	.long sys_mount
+	.long sys_oldumount
+	.long sys_setuid16
+	.long sys_getuid16
+	.long sys_stime			/* 25 */
+	.long sh64_ptrace
+	.long sys_alarm
+	.long sys_fstat
+	.long sys_pause
+	.long sys_utime			/* 30 */
+	.long sys_ni_syscall	/* old stty syscall holder */
+	.long sys_ni_syscall	/* old gtty syscall holder */
+	.long sys_access
+	.long sys_nice
+	.long sys_ni_syscall		/* 35 */ /* old ftime syscall holder */
+	.long sys_sync
+	.long sys_kill
+	.long sys_rename
+	.long sys_mkdir
+	.long sys_rmdir			/* 40 */
+	.long sys_dup
+	.long sys_pipe
+	.long sys_times
+	.long sys_ni_syscall	/* old prof syscall holder */
+	.long sys_brk			/* 45 */
+	.long sys_setgid16
+	.long sys_getgid16
+	.long sys_signal
+	.long sys_geteuid16
+	.long sys_getegid16		/* 50 */
+	.long sys_acct
+	.long sys_umount		/* recycled never used phys( */
+	.long sys_ni_syscall	/* old lock syscall holder */
+	.long sys_ioctl
+	.long sys_fcntl			/* 55 */
+	.long sys_ni_syscall	/* old mpx syscall holder */
+	.long sys_setpgid
+	.long sys_ni_syscall	/* old ulimit syscall holder */
+	.long sys_ni_syscall	/* sys_olduname */
+	.long sys_umask			/* 60 */
+	.long sys_chroot
+	.long sys_ustat
+	.long sys_dup2
+	.long sys_getppid
+	.long sys_getpgrp		/* 65 */
+	.long sys_setsid
+	.long sys_sigaction
+	.long sys_sgetmask
+	.long sys_ssetmask
+	.long sys_setreuid16		/* 70 */
+	.long sys_setregid16
+	.long sys_sigsuspend
+	.long sys_sigpending
+	.long sys_sethostname
+	.long sys_setrlimit		/* 75 */
+	.long sys_old_getrlimit
+	.long sys_getrusage
+	.long sys_gettimeofday
+	.long sys_settimeofday
+	.long sys_getgroups16		/* 80 */
+	.long sys_setgroups16
+	.long sys_ni_syscall	/* sys_oldselect */
+	.long sys_symlink
+	.long sys_lstat
+	.long sys_readlink		/* 85 */
+	.long sys_uselib
+	.long sys_swapon
+	.long sys_reboot
+	.long old_readdir
+	.long old_mmap			/* 90 */
+	.long sys_munmap
+	.long sys_truncate
+	.long sys_ftruncate
+	.long sys_fchmod
+	.long sys_fchown16		/* 95 */
+	.long sys_getpriority
+	.long sys_setpriority
+	.long sys_ni_syscall	/* old profil syscall holder */
+	.long sys_statfs
+	.long sys_fstatfs		/* 100 */
+	.long sys_ni_syscall	/* ioperm */
+	.long sys_socketcall	/* Obsolete implementation of socket syscall */
+	.long sys_syslog
+	.long sys_setitimer
+	.long sys_getitimer		/* 105 */
+	.long sys_newstat
+	.long sys_newlstat
+	.long sys_newfstat
+	.long sys_uname
+	.long sys_ni_syscall		/* 110 */ /* iopl */
+	.long sys_vhangup
+	.long sys_ni_syscall	/* idle */
+	.long sys_ni_syscall	/* vm86old */
+	.long sys_wait4
+	.long sys_swapoff		/* 115 */
+	.long sys_sysinfo
+	.long sys_ipc		/* Obsolete ipc syscall implementation */
+	.long sys_fsync
+	.long sys_sigreturn
+	.long sys_clone			/* 120 */
+	.long sys_setdomainname
+	.long sys_newuname
+	.long sys_ni_syscall	/* sys_modify_ldt */
+	.long sys_adjtimex
+	.long sys_mprotect		/* 125 */
+	.long sys_sigprocmask
+	.long sys_ni_syscall		/* old "create_module" */
+	.long sys_init_module
+	.long sys_delete_module
+	.long sys_ni_syscall		/* 130: old "get_kernel_syms" */
+	.long sys_quotactl
+	.long sys_getpgid
+	.long sys_fchdir
+	.long sys_bdflush
+	.long sys_sysfs			/* 135 */
+	.long sys_personality
+	.long sys_ni_syscall	/* for afs_syscall */
+	.long sys_setfsuid16
+	.long sys_setfsgid16
+	.long sys_llseek		/* 140 */
+	.long sys_getdents
+	.long sys_select
+	.long sys_flock
+	.long sys_msync
+	.long sys_readv			/* 145 */
+	.long sys_writev
+	.long sys_getsid
+	.long sys_fdatasync
+	.long sys_sysctl
+	.long sys_mlock			/* 150 */
+	.long sys_munlock
+	.long sys_mlockall
+	.long sys_munlockall
+	.long sys_sched_setparam
+	.long sys_sched_getparam	/* 155 */
+	.long sys_sched_setscheduler
+	.long sys_sched_getscheduler
+	.long sys_sched_yield
+	.long sys_sched_get_priority_max
+	.long sys_sched_get_priority_min  /* 160 */
+	.long sys_sched_rr_get_interval
+	.long sys_nanosleep
+	.long sys_mremap
+	.long sys_setresuid16
+	.long sys_getresuid16		/* 165 */
+	.long sys_ni_syscall	/* vm86 */
+	.long sys_ni_syscall	/* old "query_module" */
+	.long sys_poll
+	.long sys_nfsservctl
+	.long sys_setresgid16		/* 170 */
+	.long sys_getresgid16
+	.long sys_prctl
+	.long sys_rt_sigreturn
+	.long sys_rt_sigaction
+	.long sys_rt_sigprocmask	/* 175 */
+	.long sys_rt_sigpending
+	.long sys_rt_sigtimedwait
+	.long sys_rt_sigqueueinfo
+	.long sys_rt_sigsuspend
+	.long sys_pread64		/* 180 */
+	.long sys_pwrite64
+	.long sys_chown16
+	.long sys_getcwd
+	.long sys_capget
+	.long sys_capset		/* 185 */
+	.long sys_sigaltstack
+	.long sys_sendfile
+	.long sys_ni_syscall	/* streams1 */
+	.long sys_ni_syscall	/* streams2 */
+	.long sys_vfork			/* 190 */
+	.long sys_getrlimit
+	.long sys_mmap2
+	.long sys_truncate64
+	.long sys_ftruncate64
+	.long sys_stat64		/* 195 */
+	.long sys_lstat64
+	.long sys_fstat64
+	.long sys_lchown
+	.long sys_getuid
+	.long sys_getgid		/* 200 */
+	.long sys_geteuid
+	.long sys_getegid
+	.long sys_setreuid
+	.long sys_setregid
+	.long sys_getgroups		/* 205 */
+	.long sys_setgroups
+	.long sys_fchown
+	.long sys_setresuid
+	.long sys_getresuid
+	.long sys_setresgid		/* 210 */
+	.long sys_getresgid
+	.long sys_chown
+	.long sys_setuid
+	.long sys_setgid
+	.long sys_setfsuid		/* 215 */
+	.long sys_setfsgid
+	.long sys_pivot_root
+	.long sys_mincore
+	.long sys_madvise
+	/* Broken-out socket family (maintain backwards compatibility in syscall
+	   numbering with 2.4) */
+	.long sys_socket		/* 220 */
+	.long sys_bind
+	.long sys_connect
+	.long sys_listen
+	.long sys_accept
+	.long sys_getsockname		/* 225 */
+	.long sys_getpeername
+	.long sys_socketpair
+	.long sys_send
+	.long sys_sendto
+	.long sys_recv			/* 230*/
+	.long sys_recvfrom
+	.long sys_shutdown
+	.long sys_setsockopt
+	.long sys_getsockopt
+	.long sys_sendmsg		/* 235 */
+	.long sys_recvmsg
+	/* Broken-out IPC family (maintain backwards compatibility in syscall
+	   numbering with 2.4) */
+	.long sys_semop
+	.long sys_semget
+	.long sys_semctl
+	.long sys_msgsnd		/* 240 */
+	.long sys_msgrcv
+	.long sys_msgget
+	.long sys_msgctl
+	.long sys_shmat
+	.long sys_shmdt			/* 245 */
+	.long sys_shmget
+	.long sys_shmctl
+	/* Rest of syscalls listed in 2.4 i386 unistd.h */
+	.long sys_getdents64
+	.long sys_fcntl64
+	.long sys_ni_syscall		/* 250 reserved for TUX */
+	.long sys_ni_syscall		/* Reserved for Security */
+	.long sys_gettid
+	.long sys_readahead
+	.long sys_setxattr
+	.long sys_lsetxattr		/* 255 */
+	.long sys_fsetxattr
+	.long sys_getxattr
+	.long sys_lgetxattr
+	.long sys_fgetxattr
+	.long sys_listxattr		/* 260 */
+	.long sys_llistxattr
+	.long sys_flistxattr
+	.long sys_removexattr
+	.long sys_lremovexattr
+	.long sys_fremovexattr  	/* 265 */
+	.long sys_tkill
+	.long sys_sendfile64
+	.long sys_futex
+	.long sys_sched_setaffinity
+	.long sys_sched_getaffinity	/* 270 */
+	.long sys_ni_syscall
+	.long sys_ni_syscall
+	.long sys_io_setup
+	.long sys_io_destroy
+	.long sys_io_getevents		/* 275 */
+	.long sys_io_submit
+	.long sys_io_cancel
+	.long sys_fadvise64
+	.long sys_ni_syscall
+	.long sys_exit_group		/* 280 */
+	/* Rest of new 2.6 syscalls */
+	.long sys_lookup_dcookie
+	.long sys_epoll_create
+	.long sys_epoll_ctl
+	.long sys_epoll_wait
+ 	.long sys_remap_file_pages	/* 285 */
+ 	.long sys_set_tid_address
+ 	.long sys_timer_create
+ 	.long sys_timer_settime
+ 	.long sys_timer_gettime
+ 	.long sys_timer_getoverrun	/* 290 */
+ 	.long sys_timer_delete
+ 	.long sys_clock_settime
+ 	.long sys_clock_gettime
+ 	.long sys_clock_getres
+ 	.long sys_clock_nanosleep	/* 295 */
+	.long sys_statfs64
+	.long sys_fstatfs64
+	.long sys_tgkill
+	.long sys_utimes
+ 	.long sys_fadvise64_64		/* 300 */
+	.long sys_ni_syscall	/* Reserved for vserver */
+	.long sys_ni_syscall	/* Reserved for mbind */
+	.long sys_ni_syscall	/* get_mempolicy */
+	.long sys_ni_syscall	/* set_mempolicy */
+	.long sys_mq_open		/* 305 */
+	.long sys_mq_unlink
+	.long sys_mq_timedsend
+	.long sys_mq_timedreceive
+	.long sys_mq_notify
+	.long sys_mq_getsetattr		/* 310 */
+	.long sys_ni_syscall	/* Reserved for kexec */
+	.long sys_waitid
+	.long sys_add_key
+	.long sys_request_key
+	.long sys_keyctl		/* 315 */
+	.long sys_ioprio_set
+	.long sys_ioprio_get
+	.long sys_inotify_init
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch	/* 320 */
+	.long sys_ni_syscall
+	.long sys_migrate_pages
+	.long sys_openat
+	.long sys_mkdirat
+	.long sys_mknodat		/* 325 */
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64
+	.long sys_unlinkat
+	.long sys_renameat		/* 330 */
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat
+	.long sys_fchmodat
+	.long sys_faccessat		/* 335 */
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare
+	.long sys_set_robust_list
+	.long sys_get_robust_list	/* 340 */
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee
+	.long sys_vmsplice
+	.long sys_move_pages		/* 345 */
+	.long sys_getcpu
+	.long sys_epoll_pwait
+	.long sys_utimensat
+	.long sys_signalfd
+	.long sys_timerfd		/* 350 */
+	.long sys_eventfd
+	.long sys_fallocate
diff --git a/arch/sh64/kernel/syscalls.S b/arch/sh64/kernel/syscalls.S
deleted file mode 100644
index abb94c05d07a..000000000000
--- a/arch/sh64/kernel/syscalls.S
+++ /dev/null
@@ -1,381 +0,0 @@
-/*
- * arch/sh64/kernel/syscalls.S
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2004 - 2007  Paul Mundt
- * Copyright (C) 2003, 2004 Richard Curnow
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-
-#include <linux/sys.h>
-
-	.section .data, "aw"
-	.balign 32
-
-/*
- * System calls jump table
- */
-	.globl  sys_call_table
-sys_call_table:
-	.long sys_restart_syscall	/* 0  -  old "setup()" system call  */
-	.long sys_exit
-	.long sys_fork
-	.long sys_read
-	.long sys_write
-	.long sys_open			/* 5 */
-	.long sys_close
-	.long sys_waitpid
-	.long sys_creat
-	.long sys_link
-	.long sys_unlink		/* 10 */
-	.long sys_execve
-	.long sys_chdir
-	.long sys_time
-	.long sys_mknod
-	.long sys_chmod			/* 15 */
-	.long sys_lchown16
-	.long sys_ni_syscall	/* old break syscall holder */
-	.long sys_stat
-	.long sys_lseek
-	.long sys_getpid		/* 20 */
-	.long sys_mount
-	.long sys_oldumount
-	.long sys_setuid16
-	.long sys_getuid16
-	.long sys_stime			/* 25 */
-	.long sh64_ptrace
-	.long sys_alarm
-	.long sys_fstat
-	.long sys_pause
-	.long sys_utime			/* 30 */
-	.long sys_ni_syscall	/* old stty syscall holder */
-	.long sys_ni_syscall	/* old gtty syscall holder */
-	.long sys_access
-	.long sys_nice
-	.long sys_ni_syscall		/* 35 */ /* old ftime syscall holder */
-	.long sys_sync
-	.long sys_kill
-	.long sys_rename
-	.long sys_mkdir
-	.long sys_rmdir			/* 40 */
-	.long sys_dup
-	.long sys_pipe
-	.long sys_times
-	.long sys_ni_syscall	/* old prof syscall holder */
-	.long sys_brk			/* 45 */
-	.long sys_setgid16
-	.long sys_getgid16
-	.long sys_signal
-	.long sys_geteuid16
-	.long sys_getegid16		/* 50 */
-	.long sys_acct
-	.long sys_umount		/* recycled never used phys( */
-	.long sys_ni_syscall	/* old lock syscall holder */
-	.long sys_ioctl
-	.long sys_fcntl			/* 55 */
-	.long sys_ni_syscall	/* old mpx syscall holder */
-	.long sys_setpgid
-	.long sys_ni_syscall	/* old ulimit syscall holder */
-	.long sys_ni_syscall	/* sys_olduname */
-	.long sys_umask			/* 60 */
-	.long sys_chroot
-	.long sys_ustat
-	.long sys_dup2
-	.long sys_getppid
-	.long sys_getpgrp		/* 65 */
-	.long sys_setsid
-	.long sys_sigaction
-	.long sys_sgetmask
-	.long sys_ssetmask
-	.long sys_setreuid16		/* 70 */
-	.long sys_setregid16
-	.long sys_sigsuspend
-	.long sys_sigpending
-	.long sys_sethostname
-	.long sys_setrlimit		/* 75 */
-	.long sys_old_getrlimit
-	.long sys_getrusage
-	.long sys_gettimeofday
-	.long sys_settimeofday
-	.long sys_getgroups16		/* 80 */
-	.long sys_setgroups16
-	.long sys_ni_syscall	/* sys_oldselect */
-	.long sys_symlink
-	.long sys_lstat
-	.long sys_readlink		/* 85 */
-	.long sys_uselib
-	.long sys_swapon
-	.long sys_reboot
-	.long old_readdir
-	.long old_mmap			/* 90 */
-	.long sys_munmap
-	.long sys_truncate
-	.long sys_ftruncate
-	.long sys_fchmod
-	.long sys_fchown16		/* 95 */
-	.long sys_getpriority
-	.long sys_setpriority
-	.long sys_ni_syscall	/* old profil syscall holder */
-	.long sys_statfs
-	.long sys_fstatfs		/* 100 */
-	.long sys_ni_syscall	/* ioperm */
-	.long sys_socketcall	/* Obsolete implementation of socket syscall */
-	.long sys_syslog
-	.long sys_setitimer
-	.long sys_getitimer		/* 105 */
-	.long sys_newstat
-	.long sys_newlstat
-	.long sys_newfstat
-	.long sys_uname
-	.long sys_ni_syscall		/* 110 */ /* iopl */
-	.long sys_vhangup
-	.long sys_ni_syscall	/* idle */
-	.long sys_ni_syscall	/* vm86old */
-	.long sys_wait4
-	.long sys_swapoff		/* 115 */
-	.long sys_sysinfo
-	.long sys_ipc		/* Obsolete ipc syscall implementation */
-	.long sys_fsync
-	.long sys_sigreturn
-	.long sys_clone			/* 120 */
-	.long sys_setdomainname
-	.long sys_newuname
-	.long sys_ni_syscall	/* sys_modify_ldt */
-	.long sys_adjtimex
-	.long sys_mprotect		/* 125 */
-	.long sys_sigprocmask
-	.long sys_ni_syscall		/* old "create_module" */
-	.long sys_init_module
-	.long sys_delete_module
-	.long sys_ni_syscall		/* 130: old "get_kernel_syms" */
-	.long sys_quotactl
-	.long sys_getpgid
-	.long sys_fchdir
-	.long sys_bdflush
-	.long sys_sysfs			/* 135 */
-	.long sys_personality
-	.long sys_ni_syscall	/* for afs_syscall */
-	.long sys_setfsuid16
-	.long sys_setfsgid16
-	.long sys_llseek		/* 140 */
-	.long sys_getdents
-	.long sys_select
-	.long sys_flock
-	.long sys_msync
-	.long sys_readv			/* 145 */
-	.long sys_writev
-	.long sys_getsid
-	.long sys_fdatasync
-	.long sys_sysctl
-	.long sys_mlock			/* 150 */
-	.long sys_munlock
-	.long sys_mlockall
-	.long sys_munlockall
-	.long sys_sched_setparam
-	.long sys_sched_getparam	/* 155 */
-	.long sys_sched_setscheduler
-	.long sys_sched_getscheduler
-	.long sys_sched_yield
-	.long sys_sched_get_priority_max
-	.long sys_sched_get_priority_min  /* 160 */
-	.long sys_sched_rr_get_interval
-	.long sys_nanosleep
-	.long sys_mremap
-	.long sys_setresuid16
-	.long sys_getresuid16		/* 165 */
-	.long sys_ni_syscall	/* vm86 */
-	.long sys_ni_syscall	/* old "query_module" */
-	.long sys_poll
-	.long sys_nfsservctl
-	.long sys_setresgid16		/* 170 */
-	.long sys_getresgid16
-	.long sys_prctl
-	.long sys_rt_sigreturn
-	.long sys_rt_sigaction
-	.long sys_rt_sigprocmask	/* 175 */
-	.long sys_rt_sigpending
-	.long sys_rt_sigtimedwait
-	.long sys_rt_sigqueueinfo
-	.long sys_rt_sigsuspend
-	.long sys_pread64		/* 180 */
-	.long sys_pwrite64
-	.long sys_chown16
-	.long sys_getcwd
-	.long sys_capget
-	.long sys_capset		/* 185 */
-	.long sys_sigaltstack
-	.long sys_sendfile
-	.long sys_ni_syscall	/* streams1 */
-	.long sys_ni_syscall	/* streams2 */
-	.long sys_vfork			/* 190 */
-	.long sys_getrlimit
-	.long sys_mmap2
-	.long sys_truncate64
-	.long sys_ftruncate64
-	.long sys_stat64		/* 195 */
-	.long sys_lstat64
-	.long sys_fstat64
-	.long sys_lchown
-	.long sys_getuid
-	.long sys_getgid		/* 200 */
-	.long sys_geteuid
-	.long sys_getegid
-	.long sys_setreuid
-	.long sys_setregid
-	.long sys_getgroups		/* 205 */
-	.long sys_setgroups
-	.long sys_fchown
-	.long sys_setresuid
-	.long sys_getresuid
-	.long sys_setresgid		/* 210 */
-	.long sys_getresgid
-	.long sys_chown
-	.long sys_setuid
-	.long sys_setgid
-	.long sys_setfsuid		/* 215 */
-	.long sys_setfsgid
-	.long sys_pivot_root
-	.long sys_mincore
-	.long sys_madvise
-	/* Broken-out socket family (maintain backwards compatibility in syscall
-	   numbering with 2.4) */
-	.long sys_socket		/* 220 */
-	.long sys_bind
-	.long sys_connect
-	.long sys_listen
-	.long sys_accept
-	.long sys_getsockname		/* 225 */
-	.long sys_getpeername
-	.long sys_socketpair
-	.long sys_send
-	.long sys_sendto
-	.long sys_recv			/* 230*/
-	.long sys_recvfrom
-	.long sys_shutdown
-	.long sys_setsockopt
-	.long sys_getsockopt
-	.long sys_sendmsg		/* 235 */
-	.long sys_recvmsg
-	/* Broken-out IPC family (maintain backwards compatibility in syscall
-	   numbering with 2.4) */
-	.long sys_semop
-	.long sys_semget
-	.long sys_semctl
-	.long sys_msgsnd		/* 240 */
-	.long sys_msgrcv
-	.long sys_msgget
-	.long sys_msgctl
-	.long sys_shmat
-	.long sys_shmdt			/* 245 */
-	.long sys_shmget
-	.long sys_shmctl
-	/* Rest of syscalls listed in 2.4 i386 unistd.h */
-	.long sys_getdents64
-	.long sys_fcntl64
-	.long sys_ni_syscall		/* 250 reserved for TUX */
-	.long sys_ni_syscall		/* Reserved for Security */
-	.long sys_gettid
-	.long sys_readahead
-	.long sys_setxattr
-	.long sys_lsetxattr		/* 255 */
-	.long sys_fsetxattr
-	.long sys_getxattr
-	.long sys_lgetxattr
-	.long sys_fgetxattr
-	.long sys_listxattr		/* 260 */
-	.long sys_llistxattr
-	.long sys_flistxattr
-	.long sys_removexattr
-	.long sys_lremovexattr
-	.long sys_fremovexattr  	/* 265 */
-	.long sys_tkill
-	.long sys_sendfile64
-	.long sys_futex
-	.long sys_sched_setaffinity
-	.long sys_sched_getaffinity	/* 270 */
-	.long sys_ni_syscall
-	.long sys_ni_syscall
-	.long sys_io_setup
-	.long sys_io_destroy
-	.long sys_io_getevents		/* 275 */
-	.long sys_io_submit
-	.long sys_io_cancel
-	.long sys_fadvise64
-	.long sys_ni_syscall
-	.long sys_exit_group		/* 280 */
-	/* Rest of new 2.6 syscalls */
-	.long sys_lookup_dcookie
-	.long sys_epoll_create
-	.long sys_epoll_ctl
-	.long sys_epoll_wait
- 	.long sys_remap_file_pages	/* 285 */
- 	.long sys_set_tid_address
- 	.long sys_timer_create
- 	.long sys_timer_settime
- 	.long sys_timer_gettime
- 	.long sys_timer_getoverrun	/* 290 */
- 	.long sys_timer_delete
- 	.long sys_clock_settime
- 	.long sys_clock_gettime
- 	.long sys_clock_getres
- 	.long sys_clock_nanosleep	/* 295 */
-	.long sys_statfs64
-	.long sys_fstatfs64
-	.long sys_tgkill
-	.long sys_utimes
- 	.long sys_fadvise64_64		/* 300 */
-	.long sys_ni_syscall	/* Reserved for vserver */
-	.long sys_ni_syscall	/* Reserved for mbind */
-	.long sys_ni_syscall	/* get_mempolicy */
-	.long sys_ni_syscall	/* set_mempolicy */
-	.long sys_mq_open		/* 305 */
-	.long sys_mq_unlink
-	.long sys_mq_timedsend
-	.long sys_mq_timedreceive
-	.long sys_mq_notify
-	.long sys_mq_getsetattr		/* 310 */
-	.long sys_ni_syscall	/* Reserved for kexec */
-	.long sys_waitid
-	.long sys_add_key
-	.long sys_request_key
-	.long sys_keyctl		/* 315 */
-	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init
-	.long sys_inotify_add_watch
-	.long sys_inotify_rm_watch	/* 320 */
-	.long sys_ni_syscall
-	.long sys_migrate_pages
-	.long sys_openat
-	.long sys_mkdirat
-	.long sys_mknodat		/* 325 */
-	.long sys_fchownat
-	.long sys_futimesat
-	.long sys_fstatat64
-	.long sys_unlinkat
-	.long sys_renameat		/* 330 */
-	.long sys_linkat
-	.long sys_symlinkat
-	.long sys_readlinkat
-	.long sys_fchmodat
-	.long sys_faccessat		/* 335 */
-	.long sys_pselect6
-	.long sys_ppoll
-	.long sys_unshare
-	.long sys_set_robust_list
-	.long sys_get_robust_list	/* 340 */
-	.long sys_splice
-	.long sys_sync_file_range
-	.long sys_tee
-	.long sys_vmsplice
-	.long sys_move_pages		/* 345 */
-	.long sys_getcpu
-	.long sys_epoll_pwait
-	.long sys_utimensat
-	.long sys_signalfd
-	.long sys_timerfd		/* 350 */
-	.long sys_eventfd
-	.long sys_fallocate
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index b182b1cb05fd..4b21f369c28c 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -1,376 +1,5 @@
-#ifndef __ASM_SH_UNISTD_H
-#define __ASM_SH_UNISTD_H
-
-/*
- * Copyright (C) 1999  Niibe Yutaka
- */
-
-/*
- * This file contains the system call numbers.
- */
-
-#define __NR_restart_syscall	  0
-#define __NR_exit		  1
-#define __NR_fork		  2
-#define __NR_read		  3
-#define __NR_write		  4
-#define __NR_open		  5
-#define __NR_close		  6
-#define __NR_waitpid		  7
-#define __NR_creat		  8
-#define __NR_link		  9
-#define __NR_unlink		 10
-#define __NR_execve		 11
-#define __NR_chdir		 12
-#define __NR_time		 13
-#define __NR_mknod		 14
-#define __NR_chmod		 15
-#define __NR_lchown		 16
-#define __NR_break		 17
-#define __NR_oldstat		 18
-#define __NR_lseek		 19
-#define __NR_getpid		 20
-#define __NR_mount		 21
-#define __NR_umount		 22
-#define __NR_setuid		 23
-#define __NR_getuid		 24
-#define __NR_stime		 25
-#define __NR_ptrace		 26
-#define __NR_alarm		 27
-#define __NR_oldfstat		 28
-#define __NR_pause		 29
-#define __NR_utime		 30
-#define __NR_stty		 31
-#define __NR_gtty		 32
-#define __NR_access		 33
-#define __NR_nice		 34
-#define __NR_ftime		 35
-#define __NR_sync		 36
-#define __NR_kill		 37
-#define __NR_rename		 38
-#define __NR_mkdir		 39
-#define __NR_rmdir		 40
-#define __NR_dup		 41
-#define __NR_pipe		 42
-#define __NR_times		 43
-#define __NR_prof		 44
-#define __NR_brk		 45
-#define __NR_setgid		 46
-#define __NR_getgid		 47
-#define __NR_signal		 48
-#define __NR_geteuid		 49
-#define __NR_getegid		 50
-#define __NR_acct		 51
-#define __NR_umount2		 52
-#define __NR_lock		 53
-#define __NR_ioctl		 54
-#define __NR_fcntl		 55
-#define __NR_mpx		 56
-#define __NR_setpgid		 57
-#define __NR_ulimit		 58
-#define __NR_oldolduname	 59
-#define __NR_umask		 60
-#define __NR_chroot		 61
-#define __NR_ustat		 62
-#define __NR_dup2		 63
-#define __NR_getppid		 64
-#define __NR_getpgrp		 65
-#define __NR_setsid		 66
-#define __NR_sigaction		 67
-#define __NR_sgetmask		 68
-#define __NR_ssetmask		 69
-#define __NR_setreuid		 70
-#define __NR_setregid		 71
-#define __NR_sigsuspend		 72
-#define __NR_sigpending		 73
-#define __NR_sethostname	 74
-#define __NR_setrlimit		 75
-#define __NR_getrlimit		 76	/* Back compatible 2Gig limited rlimit */
-#define __NR_getrusage		 77
-#define __NR_gettimeofday	 78
-#define __NR_settimeofday	 79
-#define __NR_getgroups		 80
-#define __NR_setgroups		 81
-#define __NR_select		 82
-#define __NR_symlink		 83
-#define __NR_oldlstat		 84
-#define __NR_readlink		 85
-#define __NR_uselib		 86
-#define __NR_swapon		 87
-#define __NR_reboot		 88
-#define __NR_readdir		 89
-#define __NR_mmap		 90
-#define __NR_munmap		 91
-#define __NR_truncate		 92
-#define __NR_ftruncate		 93
-#define __NR_fchmod		 94
-#define __NR_fchown		 95
-#define __NR_getpriority	 96
-#define __NR_setpriority	 97
-#define __NR_profil		 98
-#define __NR_statfs		 99
-#define __NR_fstatfs		100
-#define __NR_ioperm		101
-#define __NR_socketcall		102
-#define __NR_syslog		103
-#define __NR_setitimer		104
-#define __NR_getitimer		105
-#define __NR_stat		106
-#define __NR_lstat		107
-#define __NR_fstat		108
-#define __NR_olduname		109
-#define __NR_iopl		110
-#define __NR_vhangup		111
-#define __NR_idle		112
-#define __NR_vm86old		113
-#define __NR_wait4		114
-#define __NR_swapoff		115
-#define __NR_sysinfo		116
-#define __NR_ipc		117
-#define __NR_fsync		118
-#define __NR_sigreturn		119
-#define __NR_clone		120
-#define __NR_setdomainname	121
-#define __NR_uname		122
-#define __NR_modify_ldt		123
-#define __NR_adjtimex		124
-#define __NR_mprotect		125
-#define __NR_sigprocmask	126
-#define __NR_create_module	127
-#define __NR_init_module	128
-#define __NR_delete_module	129
-#define __NR_get_kernel_syms	130
-#define __NR_quotactl		131
-#define __NR_getpgid		132
-#define __NR_fchdir		133
-#define __NR_bdflush		134
-#define __NR_sysfs		135
-#define __NR_personality	136
-#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
-#define __NR_setfsuid		138
-#define __NR_setfsgid		139
-#define __NR__llseek		140
-#define __NR_getdents		141
-#define __NR__newselect		142
-#define __NR_flock		143
-#define __NR_msync		144
-#define __NR_readv		145
-#define __NR_writev		146
-#define __NR_getsid		147
-#define __NR_fdatasync		148
-#define __NR__sysctl		149
-#define __NR_mlock		150
-#define __NR_munlock		151
-#define __NR_mlockall		152
-#define __NR_munlockall		153
-#define __NR_sched_setparam		154
-#define __NR_sched_getparam		155
-#define __NR_sched_setscheduler		156
-#define __NR_sched_getscheduler		157
-#define __NR_sched_yield		158
-#define __NR_sched_get_priority_max	159
-#define __NR_sched_get_priority_min	160
-#define __NR_sched_rr_get_interval	161
-#define __NR_nanosleep		162
-#define __NR_mremap		163
-#define __NR_setresuid		164
-#define __NR_getresuid		165
-#define __NR_vm86		166
-#define __NR_query_module	167
-#define __NR_poll		168
-#define __NR_nfsservctl		169
-#define __NR_setresgid		170
-#define __NR_getresgid		171
-#define __NR_prctl              172
-#define __NR_rt_sigreturn	173
-#define __NR_rt_sigaction	174
-#define __NR_rt_sigprocmask	175
-#define __NR_rt_sigpending	176
-#define __NR_rt_sigtimedwait	177
-#define __NR_rt_sigqueueinfo	178
-#define __NR_rt_sigsuspend	179
-#define __NR_pread64		180
-#define __NR_pwrite64		181
-#define __NR_chown		182
-#define __NR_getcwd		183
-#define __NR_capget		184
-#define __NR_capset		185
-#define __NR_sigaltstack	186
-#define __NR_sendfile		187
-#define __NR_streams1		188	/* some people actually want it */
-#define __NR_streams2		189	/* some people actually want it */
-#define __NR_vfork		190
-#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
-#define __NR_mmap2		192
-#define __NR_truncate64		193
-#define __NR_ftruncate64	194
-#define __NR_stat64		195
-#define __NR_lstat64		196
-#define __NR_fstat64		197
-#define __NR_lchown32		198
-#define __NR_getuid32		199
-#define __NR_getgid32		200
-#define __NR_geteuid32		201
-#define __NR_getegid32		202
-#define __NR_setreuid32		203
-#define __NR_setregid32		204
-#define __NR_getgroups32	205
-#define __NR_setgroups32	206
-#define __NR_fchown32		207
-#define __NR_setresuid32	208
-#define __NR_getresuid32	209
-#define __NR_setresgid32	210
-#define __NR_getresgid32	211
-#define __NR_chown32		212
-#define __NR_setuid32		213
-#define __NR_setgid32		214
-#define __NR_setfsuid32		215
-#define __NR_setfsgid32		216
-#define __NR_pivot_root		217
-#define __NR_mincore		218
-#define __NR_madvise		219
-#define __NR_getdents64		220
-#define __NR_fcntl64		221
-/* 223 is unused */
-#define __NR_gettid		224
-#define __NR_readahead		225
-#define __NR_setxattr		226
-#define __NR_lsetxattr		227
-#define __NR_fsetxattr		228
-#define __NR_getxattr		229
-#define __NR_lgetxattr		230
-#define __NR_fgetxattr		231
-#define __NR_listxattr		232
-#define __NR_llistxattr		233
-#define __NR_flistxattr		234
-#define __NR_removexattr	235
-#define __NR_lremovexattr	236
-#define __NR_fremovexattr	237
-#define __NR_tkill		238
-#define __NR_sendfile64		239
-#define __NR_futex		240
-#define __NR_sched_setaffinity	241
-#define __NR_sched_getaffinity	242
-#define __NR_set_thread_area	243
-#define __NR_get_thread_area	244
-#define __NR_io_setup		245
-#define __NR_io_destroy		246
-#define __NR_io_getevents	247
-#define __NR_io_submit		248
-#define __NR_io_cancel		249
-#define __NR_fadvise64		250
-
-#define __NR_exit_group		252
-#define __NR_lookup_dcookie	253
-#define __NR_epoll_create	254
-#define __NR_epoll_ctl		255
-#define __NR_epoll_wait		256
-#define __NR_remap_file_pages	257
-#define __NR_set_tid_address	258
-#define __NR_timer_create	259
-#define __NR_timer_settime	(__NR_timer_create+1)
-#define __NR_timer_gettime	(__NR_timer_create+2)
-#define __NR_timer_getoverrun	(__NR_timer_create+3)
-#define __NR_timer_delete	(__NR_timer_create+4)
-#define __NR_clock_settime	(__NR_timer_create+5)
-#define __NR_clock_gettime	(__NR_timer_create+6)
-#define __NR_clock_getres	(__NR_timer_create+7)
-#define __NR_clock_nanosleep	(__NR_timer_create+8)
-#define __NR_statfs64		268
-#define __NR_fstatfs64		269
-#define __NR_tgkill		270
-#define __NR_utimes		271
-#define __NR_fadvise64_64	272
-#define __NR_vserver		273
-#define __NR_mbind              274
-#define __NR_get_mempolicy      275
-#define __NR_set_mempolicy      276
-#define __NR_mq_open            277
-#define __NR_mq_unlink          (__NR_mq_open+1)
-#define __NR_mq_timedsend       (__NR_mq_open+2)
-#define __NR_mq_timedreceive    (__NR_mq_open+3)
-#define __NR_mq_notify          (__NR_mq_open+4)
-#define __NR_mq_getsetattr      (__NR_mq_open+5)
-#define __NR_kexec_load		283
-#define __NR_waitid		284
-#define __NR_add_key		285
-#define __NR_request_key	286
-#define __NR_keyctl		287
-#define __NR_ioprio_set		288
-#define __NR_ioprio_get		289
-#define __NR_inotify_init	290
-#define __NR_inotify_add_watch	291
-#define __NR_inotify_rm_watch	292
-/* 293 is unused */
-#define __NR_migrate_pages	294
-#define __NR_openat		295
-#define __NR_mkdirat		296
-#define __NR_mknodat		297
-#define __NR_fchownat		298
-#define __NR_futimesat		299
-#define __NR_fstatat64		300
-#define __NR_unlinkat		301
-#define __NR_renameat		302
-#define __NR_linkat		303
-#define __NR_symlinkat		304
-#define __NR_readlinkat		305
-#define __NR_fchmodat		306
-#define __NR_faccessat		307
-#define __NR_pselect6		308
-#define __NR_ppoll		309
-#define __NR_unshare		310
-#define __NR_set_robust_list	311
-#define __NR_get_robust_list	312
-#define __NR_splice		313
-#define __NR_sync_file_range	314
-#define __NR_tee		315
-#define __NR_vmsplice		316
-#define __NR_move_pages		317
-#define __NR_getcpu		318
-#define __NR_epoll_pwait	319
-#define __NR_utimensat		320
-#define __NR_signalfd		321
-#define __NR_timerfd		322
-#define __NR_eventfd		323
-#define __NR_fallocate		324
-
-#define NR_syscalls 325
-
-#ifdef __KERNEL__
-
-#define __ARCH_WANT_IPC_PARSE_VERSION
-#define __ARCH_WANT_OLD_READDIR
-#define __ARCH_WANT_OLD_STAT
-#define __ARCH_WANT_STAT64
-#define __ARCH_WANT_SYS_ALARM
-#define __ARCH_WANT_SYS_GETHOSTNAME
-#define __ARCH_WANT_SYS_PAUSE
-#define __ARCH_WANT_SYS_SGETMASK
-#define __ARCH_WANT_SYS_SIGNAL
-#define __ARCH_WANT_SYS_TIME
-#define __ARCH_WANT_SYS_UTIME
-#define __ARCH_WANT_SYS_WAITPID
-#define __ARCH_WANT_SYS_SOCKETCALL
-#define __ARCH_WANT_SYS_FADVISE64
-#define __ARCH_WANT_SYS_GETPGRP
-#define __ARCH_WANT_SYS_LLSEEK
-#define __ARCH_WANT_SYS_NICE
-#define __ARCH_WANT_SYS_OLD_GETRLIMIT
-#define __ARCH_WANT_SYS_OLDUMOUNT
-#define __ARCH_WANT_SYS_SIGPENDING
-#define __ARCH_WANT_SYS_SIGPROCMASK
-#define __ARCH_WANT_SYS_RT_SIGACTION
-#define __ARCH_WANT_SYS_RT_SIGSUSPEND
-
-/*
- * "Conditional" syscalls
- *
- * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
- * but it doesn't work on all toolchains, so we just do it by hand
- */
-#ifndef cond_syscall
-#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall")
+#ifdef CONFIG_SUPERH32
+# include "unistd_32.h"
+#else
+# include "unistd_64.h"
 #endif
-
-#endif /* __KERNEL__ */
-#endif /* __ASM_SH_UNISTD_H */
diff --git a/include/asm-sh/unistd_32.h b/include/asm-sh/unistd_32.h
new file mode 100644
index 000000000000..b182b1cb05fd
--- /dev/null
+++ b/include/asm-sh/unistd_32.h
@@ -0,0 +1,376 @@
+#ifndef __ASM_SH_UNISTD_H
+#define __ASM_SH_UNISTD_H
+
+/*
+ * Copyright (C) 1999  Niibe Yutaka
+ */
+
+/*
+ * This file contains the system call numbers.
+ */
+
+#define __NR_restart_syscall	  0
+#define __NR_exit		  1
+#define __NR_fork		  2
+#define __NR_read		  3
+#define __NR_write		  4
+#define __NR_open		  5
+#define __NR_close		  6
+#define __NR_waitpid		  7
+#define __NR_creat		  8
+#define __NR_link		  9
+#define __NR_unlink		 10
+#define __NR_execve		 11
+#define __NR_chdir		 12
+#define __NR_time		 13
+#define __NR_mknod		 14
+#define __NR_chmod		 15
+#define __NR_lchown		 16
+#define __NR_break		 17
+#define __NR_oldstat		 18
+#define __NR_lseek		 19
+#define __NR_getpid		 20
+#define __NR_mount		 21
+#define __NR_umount		 22
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_ptrace		 26
+#define __NR_alarm		 27
+#define __NR_oldfstat		 28
+#define __NR_pause		 29
+#define __NR_utime		 30
+#define __NR_stty		 31
+#define __NR_gtty		 32
+#define __NR_access		 33
+#define __NR_nice		 34
+#define __NR_ftime		 35
+#define __NR_sync		 36
+#define __NR_kill		 37
+#define __NR_rename		 38
+#define __NR_mkdir		 39
+#define __NR_rmdir		 40
+#define __NR_dup		 41
+#define __NR_pipe		 42
+#define __NR_times		 43
+#define __NR_prof		 44
+#define __NR_brk		 45
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_signal		 48
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_acct		 51
+#define __NR_umount2		 52
+#define __NR_lock		 53
+#define __NR_ioctl		 54
+#define __NR_fcntl		 55
+#define __NR_mpx		 56
+#define __NR_setpgid		 57
+#define __NR_ulimit		 58
+#define __NR_oldolduname	 59
+#define __NR_umask		 60
+#define __NR_chroot		 61
+#define __NR_ustat		 62
+#define __NR_dup2		 63
+#define __NR_getppid		 64
+#define __NR_getpgrp		 65
+#define __NR_setsid		 66
+#define __NR_sigaction		 67
+#define __NR_sgetmask		 68
+#define __NR_ssetmask		 69
+#define __NR_setreuid		 70
+#define __NR_setregid		 71
+#define __NR_sigsuspend		 72
+#define __NR_sigpending		 73
+#define __NR_sethostname	 74
+#define __NR_setrlimit		 75
+#define __NR_getrlimit		 76	/* Back compatible 2Gig limited rlimit */
+#define __NR_getrusage		 77
+#define __NR_gettimeofday	 78
+#define __NR_settimeofday	 79
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_select		 82
+#define __NR_symlink		 83
+#define __NR_oldlstat		 84
+#define __NR_readlink		 85
+#define __NR_uselib		 86
+#define __NR_swapon		 87
+#define __NR_reboot		 88
+#define __NR_readdir		 89
+#define __NR_mmap		 90
+#define __NR_munmap		 91
+#define __NR_truncate		 92
+#define __NR_ftruncate		 93
+#define __NR_fchmod		 94
+#define __NR_fchown		 95
+#define __NR_getpriority	 96
+#define __NR_setpriority	 97
+#define __NR_profil		 98
+#define __NR_statfs		 99
+#define __NR_fstatfs		100
+#define __NR_ioperm		101
+#define __NR_socketcall		102
+#define __NR_syslog		103
+#define __NR_setitimer		104
+#define __NR_getitimer		105
+#define __NR_stat		106
+#define __NR_lstat		107
+#define __NR_fstat		108
+#define __NR_olduname		109
+#define __NR_iopl		110
+#define __NR_vhangup		111
+#define __NR_idle		112
+#define __NR_vm86old		113
+#define __NR_wait4		114
+#define __NR_swapoff		115
+#define __NR_sysinfo		116
+#define __NR_ipc		117
+#define __NR_fsync		118
+#define __NR_sigreturn		119
+#define __NR_clone		120
+#define __NR_setdomainname	121
+#define __NR_uname		122
+#define __NR_modify_ldt		123
+#define __NR_adjtimex		124
+#define __NR_mprotect		125
+#define __NR_sigprocmask	126
+#define __NR_create_module	127
+#define __NR_init_module	128
+#define __NR_delete_module	129
+#define __NR_get_kernel_syms	130
+#define __NR_quotactl		131
+#define __NR_getpgid		132
+#define __NR_fchdir		133
+#define __NR_bdflush		134
+#define __NR_sysfs		135
+#define __NR_personality	136
+#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
+#define __NR_setfsuid		138
+#define __NR_setfsgid		139
+#define __NR__llseek		140
+#define __NR_getdents		141
+#define __NR__newselect		142
+#define __NR_flock		143
+#define __NR_msync		144
+#define __NR_readv		145
+#define __NR_writev		146
+#define __NR_getsid		147
+#define __NR_fdatasync		148
+#define __NR__sysctl		149
+#define __NR_mlock		150
+#define __NR_munlock		151
+#define __NR_mlockall		152
+#define __NR_munlockall		153
+#define __NR_sched_setparam		154
+#define __NR_sched_getparam		155
+#define __NR_sched_setscheduler		156
+#define __NR_sched_getscheduler		157
+#define __NR_sched_yield		158
+#define __NR_sched_get_priority_max	159
+#define __NR_sched_get_priority_min	160
+#define __NR_sched_rr_get_interval	161
+#define __NR_nanosleep		162
+#define __NR_mremap		163
+#define __NR_setresuid		164
+#define __NR_getresuid		165
+#define __NR_vm86		166
+#define __NR_query_module	167
+#define __NR_poll		168
+#define __NR_nfsservctl		169
+#define __NR_setresgid		170
+#define __NR_getresgid		171
+#define __NR_prctl              172
+#define __NR_rt_sigreturn	173
+#define __NR_rt_sigaction	174
+#define __NR_rt_sigprocmask	175
+#define __NR_rt_sigpending	176
+#define __NR_rt_sigtimedwait	177
+#define __NR_rt_sigqueueinfo	178
+#define __NR_rt_sigsuspend	179
+#define __NR_pread64		180
+#define __NR_pwrite64		181
+#define __NR_chown		182
+#define __NR_getcwd		183
+#define __NR_capget		184
+#define __NR_capset		185
+#define __NR_sigaltstack	186
+#define __NR_sendfile		187
+#define __NR_streams1		188	/* some people actually want it */
+#define __NR_streams2		189	/* some people actually want it */
+#define __NR_vfork		190
+#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
+#define __NR_mmap2		192
+#define __NR_truncate64		193
+#define __NR_ftruncate64	194
+#define __NR_stat64		195
+#define __NR_lstat64		196
+#define __NR_fstat64		197
+#define __NR_lchown32		198
+#define __NR_getuid32		199
+#define __NR_getgid32		200
+#define __NR_geteuid32		201
+#define __NR_getegid32		202
+#define __NR_setreuid32		203
+#define __NR_setregid32		204
+#define __NR_getgroups32	205
+#define __NR_setgroups32	206
+#define __NR_fchown32		207
+#define __NR_setresuid32	208
+#define __NR_getresuid32	209
+#define __NR_setresgid32	210
+#define __NR_getresgid32	211
+#define __NR_chown32		212
+#define __NR_setuid32		213
+#define __NR_setgid32		214
+#define __NR_setfsuid32		215
+#define __NR_setfsgid32		216
+#define __NR_pivot_root		217
+#define __NR_mincore		218
+#define __NR_madvise		219
+#define __NR_getdents64		220
+#define __NR_fcntl64		221
+/* 223 is unused */
+#define __NR_gettid		224
+#define __NR_readahead		225
+#define __NR_setxattr		226
+#define __NR_lsetxattr		227
+#define __NR_fsetxattr		228
+#define __NR_getxattr		229
+#define __NR_lgetxattr		230
+#define __NR_fgetxattr		231
+#define __NR_listxattr		232
+#define __NR_llistxattr		233
+#define __NR_flistxattr		234
+#define __NR_removexattr	235
+#define __NR_lremovexattr	236
+#define __NR_fremovexattr	237
+#define __NR_tkill		238
+#define __NR_sendfile64		239
+#define __NR_futex		240
+#define __NR_sched_setaffinity	241
+#define __NR_sched_getaffinity	242
+#define __NR_set_thread_area	243
+#define __NR_get_thread_area	244
+#define __NR_io_setup		245
+#define __NR_io_destroy		246
+#define __NR_io_getevents	247
+#define __NR_io_submit		248
+#define __NR_io_cancel		249
+#define __NR_fadvise64		250
+
+#define __NR_exit_group		252
+#define __NR_lookup_dcookie	253
+#define __NR_epoll_create	254
+#define __NR_epoll_ctl		255
+#define __NR_epoll_wait		256
+#define __NR_remap_file_pages	257
+#define __NR_set_tid_address	258
+#define __NR_timer_create	259
+#define __NR_timer_settime	(__NR_timer_create+1)
+#define __NR_timer_gettime	(__NR_timer_create+2)
+#define __NR_timer_getoverrun	(__NR_timer_create+3)
+#define __NR_timer_delete	(__NR_timer_create+4)
+#define __NR_clock_settime	(__NR_timer_create+5)
+#define __NR_clock_gettime	(__NR_timer_create+6)
+#define __NR_clock_getres	(__NR_timer_create+7)
+#define __NR_clock_nanosleep	(__NR_timer_create+8)
+#define __NR_statfs64		268
+#define __NR_fstatfs64		269
+#define __NR_tgkill		270
+#define __NR_utimes		271
+#define __NR_fadvise64_64	272
+#define __NR_vserver		273
+#define __NR_mbind              274
+#define __NR_get_mempolicy      275
+#define __NR_set_mempolicy      276
+#define __NR_mq_open            277
+#define __NR_mq_unlink          (__NR_mq_open+1)
+#define __NR_mq_timedsend       (__NR_mq_open+2)
+#define __NR_mq_timedreceive    (__NR_mq_open+3)
+#define __NR_mq_notify          (__NR_mq_open+4)
+#define __NR_mq_getsetattr      (__NR_mq_open+5)
+#define __NR_kexec_load		283
+#define __NR_waitid		284
+#define __NR_add_key		285
+#define __NR_request_key	286
+#define __NR_keyctl		287
+#define __NR_ioprio_set		288
+#define __NR_ioprio_get		289
+#define __NR_inotify_init	290
+#define __NR_inotify_add_watch	291
+#define __NR_inotify_rm_watch	292
+/* 293 is unused */
+#define __NR_migrate_pages	294
+#define __NR_openat		295
+#define __NR_mkdirat		296
+#define __NR_mknodat		297
+#define __NR_fchownat		298
+#define __NR_futimesat		299
+#define __NR_fstatat64		300
+#define __NR_unlinkat		301
+#define __NR_renameat		302
+#define __NR_linkat		303
+#define __NR_symlinkat		304
+#define __NR_readlinkat		305
+#define __NR_fchmodat		306
+#define __NR_faccessat		307
+#define __NR_pselect6		308
+#define __NR_ppoll		309
+#define __NR_unshare		310
+#define __NR_set_robust_list	311
+#define __NR_get_robust_list	312
+#define __NR_splice		313
+#define __NR_sync_file_range	314
+#define __NR_tee		315
+#define __NR_vmsplice		316
+#define __NR_move_pages		317
+#define __NR_getcpu		318
+#define __NR_epoll_pwait	319
+#define __NR_utimensat		320
+#define __NR_signalfd		321
+#define __NR_timerfd		322
+#define __NR_eventfd		323
+#define __NR_fallocate		324
+
+#define NR_syscalls 325
+
+#ifdef __KERNEL__
+
+#define __ARCH_WANT_IPC_PARSE_VERSION
+#define __ARCH_WANT_OLD_READDIR
+#define __ARCH_WANT_OLD_STAT
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_SGETMASK
+#define __ARCH_WANT_SYS_SIGNAL
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_SOCKETCALL
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_NICE
+#define __ARCH_WANT_SYS_OLD_GETRLIMIT
+#define __ARCH_WANT_SYS_OLDUMOUNT
+#define __ARCH_WANT_SYS_SIGPENDING
+#define __ARCH_WANT_SYS_SIGPROCMASK
+#define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
+
+/*
+ * "Conditional" syscalls
+ *
+ * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
+ * but it doesn't work on all toolchains, so we just do it by hand
+ */
+#ifndef cond_syscall
+#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall")
+#endif
+
+#endif /* __KERNEL__ */
+#endif /* __ASM_SH_UNISTD_H */
diff --git a/include/asm-sh/unistd_64.h b/include/asm-sh/unistd_64.h
new file mode 100644
index 000000000000..1a5197f369b2
--- /dev/null
+++ b/include/asm-sh/unistd_64.h
@@ -0,0 +1,417 @@
+#ifndef __ASM_SH64_UNISTD_H
+#define __ASM_SH64_UNISTD_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/unistd.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003 - 2007 Paul Mundt
+ * Copyright (C) 2004  Sean McGoogan
+ *
+ * This file contains the system call numbers.
+ *
+ */
+
+#define __NR_restart_syscall	  0
+#define __NR_exit		  1
+#define __NR_fork		  2
+#define __NR_read		  3
+#define __NR_write		  4
+#define __NR_open		  5
+#define __NR_close		  6
+#define __NR_waitpid		  7
+#define __NR_creat		  8
+#define __NR_link		  9
+#define __NR_unlink		 10
+#define __NR_execve		 11
+#define __NR_chdir		 12
+#define __NR_time		 13
+#define __NR_mknod		 14
+#define __NR_chmod		 15
+#define __NR_lchown		 16
+#define __NR_break		 17
+#define __NR_oldstat		 18
+#define __NR_lseek		 19
+#define __NR_getpid		 20
+#define __NR_mount		 21
+#define __NR_umount		 22
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_ptrace		 26
+#define __NR_alarm		 27
+#define __NR_oldfstat		 28
+#define __NR_pause		 29
+#define __NR_utime		 30
+#define __NR_stty		 31
+#define __NR_gtty		 32
+#define __NR_access		 33
+#define __NR_nice		 34
+#define __NR_ftime		 35
+#define __NR_sync		 36
+#define __NR_kill		 37
+#define __NR_rename		 38
+#define __NR_mkdir		 39
+#define __NR_rmdir		 40
+#define __NR_dup		 41
+#define __NR_pipe		 42
+#define __NR_times		 43
+#define __NR_prof		 44
+#define __NR_brk		 45
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_signal		 48
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_acct		 51
+#define __NR_umount2		 52
+#define __NR_lock		 53
+#define __NR_ioctl		 54
+#define __NR_fcntl		 55
+#define __NR_mpx		 56
+#define __NR_setpgid		 57
+#define __NR_ulimit		 58
+#define __NR_oldolduname	 59
+#define __NR_umask		 60
+#define __NR_chroot		 61
+#define __NR_ustat		 62
+#define __NR_dup2		 63
+#define __NR_getppid		 64
+#define __NR_getpgrp		 65
+#define __NR_setsid		 66
+#define __NR_sigaction		 67
+#define __NR_sgetmask		 68
+#define __NR_ssetmask		 69
+#define __NR_setreuid		 70
+#define __NR_setregid		 71
+#define __NR_sigsuspend		 72
+#define __NR_sigpending		 73
+#define __NR_sethostname	 74
+#define __NR_setrlimit		 75
+#define __NR_getrlimit	 	 76	/* Back compatible 2Gig limited rlimit */
+#define __NR_getrusage		 77
+#define __NR_gettimeofday	 78
+#define __NR_settimeofday	 79
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_select		 82
+#define __NR_symlink		 83
+#define __NR_oldlstat		 84
+#define __NR_readlink		 85
+#define __NR_uselib		 86
+#define __NR_swapon		 87
+#define __NR_reboot		 88
+#define __NR_readdir		 89
+#define __NR_mmap		 90
+#define __NR_munmap		 91
+#define __NR_truncate		 92
+#define __NR_ftruncate		 93
+#define __NR_fchmod		 94
+#define __NR_fchown		 95
+#define __NR_getpriority	 96
+#define __NR_setpriority	 97
+#define __NR_profil		 98
+#define __NR_statfs		 99
+#define __NR_fstatfs		100
+#define __NR_ioperm		101
+#define __NR_socketcall		102	/* old implementation of socket systemcall */
+#define __NR_syslog		103
+#define __NR_setitimer		104
+#define __NR_getitimer		105
+#define __NR_stat		106
+#define __NR_lstat		107
+#define __NR_fstat		108
+#define __NR_olduname		109
+#define __NR_iopl		110
+#define __NR_vhangup		111
+#define __NR_idle		112
+#define __NR_vm86old		113
+#define __NR_wait4		114
+#define __NR_swapoff		115
+#define __NR_sysinfo		116
+#define __NR_ipc		117
+#define __NR_fsync		118
+#define __NR_sigreturn		119
+#define __NR_clone		120
+#define __NR_setdomainname	121
+#define __NR_uname		122
+#define __NR_modify_ldt		123
+#define __NR_adjtimex		124
+#define __NR_mprotect		125
+#define __NR_sigprocmask	126
+#define __NR_create_module	127
+#define __NR_init_module	128
+#define __NR_delete_module	129
+#define __NR_get_kernel_syms	130
+#define __NR_quotactl		131
+#define __NR_getpgid		132
+#define __NR_fchdir		133
+#define __NR_bdflush		134
+#define __NR_sysfs		135
+#define __NR_personality	136
+#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
+#define __NR_setfsuid		138
+#define __NR_setfsgid		139
+#define __NR__llseek		140
+#define __NR_getdents		141
+#define __NR__newselect		142
+#define __NR_flock		143
+#define __NR_msync		144
+#define __NR_readv		145
+#define __NR_writev		146
+#define __NR_getsid		147
+#define __NR_fdatasync		148
+#define __NR__sysctl		149
+#define __NR_mlock		150
+#define __NR_munlock		151
+#define __NR_mlockall		152
+#define __NR_munlockall		153
+#define __NR_sched_setparam		154
+#define __NR_sched_getparam		155
+#define __NR_sched_setscheduler		156
+#define __NR_sched_getscheduler		157
+#define __NR_sched_yield		158
+#define __NR_sched_get_priority_max	159
+#define __NR_sched_get_priority_min	160
+#define __NR_sched_rr_get_interval	161
+#define __NR_nanosleep		162
+#define __NR_mremap		163
+#define __NR_setresuid		164
+#define __NR_getresuid		165
+#define __NR_vm86		166
+#define __NR_query_module	167
+#define __NR_poll		168
+#define __NR_nfsservctl		169
+#define __NR_setresgid		170
+#define __NR_getresgid		171
+#define __NR_prctl              172
+#define __NR_rt_sigreturn	173
+#define __NR_rt_sigaction	174
+#define __NR_rt_sigprocmask	175
+#define __NR_rt_sigpending	176
+#define __NR_rt_sigtimedwait	177
+#define __NR_rt_sigqueueinfo	178
+#define __NR_rt_sigsuspend	179
+#define __NR_pread64		180
+#define __NR_pwrite64		181
+#define __NR_chown		182
+#define __NR_getcwd		183
+#define __NR_capget		184
+#define __NR_capset		185
+#define __NR_sigaltstack	186
+#define __NR_sendfile		187
+#define __NR_streams1		188	/* some people actually want it */
+#define __NR_streams2		189	/* some people actually want it */
+#define __NR_vfork		190
+#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
+#define __NR_mmap2		192
+#define __NR_truncate64		193
+#define __NR_ftruncate64	194
+#define __NR_stat64		195
+#define __NR_lstat64		196
+#define __NR_fstat64		197
+#define __NR_lchown32		198
+#define __NR_getuid32		199
+#define __NR_getgid32		200
+#define __NR_geteuid32		201
+#define __NR_getegid32		202
+#define __NR_setreuid32		203
+#define __NR_setregid32		204
+#define __NR_getgroups32	205
+#define __NR_setgroups32	206
+#define __NR_fchown32		207
+#define __NR_setresuid32	208
+#define __NR_getresuid32	209
+#define __NR_setresgid32	210
+#define __NR_getresgid32	211
+#define __NR_chown32		212
+#define __NR_setuid32		213
+#define __NR_setgid32		214
+#define __NR_setfsuid32		215
+#define __NR_setfsgid32		216
+#define __NR_pivot_root		217
+#define __NR_mincore		218
+#define __NR_madvise		219
+
+/* Non-multiplexed socket family */
+#define __NR_socket		220
+#define __NR_bind		221
+#define __NR_connect		222
+#define __NR_listen		223
+#define __NR_accept		224
+#define __NR_getsockname	225
+#define __NR_getpeername	226
+#define __NR_socketpair		227
+#define __NR_send		228
+#define __NR_sendto		229
+#define __NR_recv		230
+#define __NR_recvfrom		231
+#define __NR_shutdown		232
+#define __NR_setsockopt		233
+#define __NR_getsockopt		234
+#define __NR_sendmsg		235
+#define __NR_recvmsg		236
+
+/* Non-multiplexed IPC family */
+#define __NR_semop		237
+#define __NR_semget		238
+#define __NR_semctl		239
+#define __NR_msgsnd		240
+#define __NR_msgrcv		241
+#define __NR_msgget		242
+#define __NR_msgctl		243
+#if 0
+#define __NR_shmatcall		244
+#endif
+#define __NR_shmdt		245
+#define __NR_shmget		246
+#define __NR_shmctl		247
+
+#define __NR_getdents64		248
+#define __NR_fcntl64		249
+/* 223 is unused */
+#define __NR_gettid		252
+#define __NR_readahead		253
+#define __NR_setxattr		254
+#define __NR_lsetxattr		255
+#define __NR_fsetxattr		256
+#define __NR_getxattr		257
+#define __NR_lgetxattr		258
+#define __NR_fgetxattr		269
+#define __NR_listxattr		260
+#define __NR_llistxattr		261
+#define __NR_flistxattr		262
+#define __NR_removexattr	263
+#define __NR_lremovexattr	264
+#define __NR_fremovexattr	265
+#define __NR_tkill		266
+#define __NR_sendfile64		267
+#define __NR_futex		268
+#define __NR_sched_setaffinity	269
+#define __NR_sched_getaffinity	270
+#define __NR_set_thread_area	271
+#define __NR_get_thread_area	272
+#define __NR_io_setup		273
+#define __NR_io_destroy		274
+#define __NR_io_getevents	275
+#define __NR_io_submit		276
+#define __NR_io_cancel		277
+#define __NR_fadvise64		278
+#define __NR_exit_group		280
+
+#define __NR_lookup_dcookie	281
+#define __NR_epoll_create	282
+#define __NR_epoll_ctl		283
+#define __NR_epoll_wait		284
+#define __NR_remap_file_pages	285
+#define __NR_set_tid_address	286
+#define __NR_timer_create	287
+#define __NR_timer_settime	(__NR_timer_create+1)
+#define __NR_timer_gettime	(__NR_timer_create+2)
+#define __NR_timer_getoverrun	(__NR_timer_create+3)
+#define __NR_timer_delete	(__NR_timer_create+4)
+#define __NR_clock_settime	(__NR_timer_create+5)
+#define __NR_clock_gettime	(__NR_timer_create+6)
+#define __NR_clock_getres	(__NR_timer_create+7)
+#define __NR_clock_nanosleep	(__NR_timer_create+8)
+#define __NR_statfs64		296
+#define __NR_fstatfs64		297
+#define __NR_tgkill		298
+#define __NR_utimes		299
+#define __NR_fadvise64_64	300
+#define __NR_vserver		301
+#define __NR_mbind              302
+#define __NR_get_mempolicy      303
+#define __NR_set_mempolicy      304
+#define __NR_mq_open            305
+#define __NR_mq_unlink          (__NR_mq_open+1)
+#define __NR_mq_timedsend       (__NR_mq_open+2)
+#define __NR_mq_timedreceive    (__NR_mq_open+3)
+#define __NR_mq_notify          (__NR_mq_open+4)
+#define __NR_mq_getsetattr      (__NR_mq_open+5)
+#define __NR_kexec_load		311
+#define __NR_waitid		312
+#define __NR_add_key		313
+#define __NR_request_key	314
+#define __NR_keyctl		315
+#define __NR_ioprio_set		316
+#define __NR_ioprio_get		317
+#define __NR_inotify_init	318
+#define __NR_inotify_add_watch	319
+#define __NR_inotify_rm_watch	320
+/* 321 is unused */
+#define __NR_migrate_pages	322
+#define __NR_openat		323
+#define __NR_mkdirat		324
+#define __NR_mknodat		325
+#define __NR_fchownat		326
+#define __NR_futimesat		327
+#define __NR_fstatat64		328
+#define __NR_unlinkat		329
+#define __NR_renameat		330
+#define __NR_linkat		331
+#define __NR_symlinkat		332
+#define __NR_readlinkat		333
+#define __NR_fchmodat		334
+#define __NR_faccessat		335
+#define __NR_pselect6		336
+#define __NR_ppoll		337
+#define __NR_unshare		338
+#define __NR_set_robust_list	339
+#define __NR_get_robust_list	340
+#define __NR_splice		341
+#define __NR_sync_file_range	342
+#define __NR_tee		343
+#define __NR_vmsplice		344
+#define __NR_move_pages		345
+#define __NR_getcpu		346
+#define __NR_epoll_pwait	347
+#define __NR_utimensat		348
+#define __NR_signalfd		349
+#define __NR_timerfd		350
+#define __NR_eventfd		351
+#define __NR_fallocate		352
+
+#ifdef __KERNEL__
+
+#define NR_syscalls 353
+
+#define __ARCH_WANT_IPC_PARSE_VERSION
+#define __ARCH_WANT_OLD_READDIR
+#define __ARCH_WANT_OLD_STAT
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_SGETMASK
+#define __ARCH_WANT_SYS_SIGNAL
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_SOCKETCALL
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_NICE
+#define __ARCH_WANT_SYS_OLD_GETRLIMIT
+#define __ARCH_WANT_SYS_OLDUMOUNT
+#define __ARCH_WANT_SYS_SIGPENDING
+#define __ARCH_WANT_SYS_SIGPROCMASK
+#define __ARCH_WANT_SYS_RT_SIGACTION
+
+/*
+ * "Conditional" syscalls
+ *
+ * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
+ * but it doesn't work on all toolchains, so we just do it by hand
+ */
+#ifndef cond_syscall
+#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall")
+#endif
+
+#endif /* __KERNEL__ */
+#endif /* __ASM_SH64_UNISTD_H */
diff --git a/include/asm-sh64/unistd.h b/include/asm-sh64/unistd.h
deleted file mode 100644
index 1a5197f369b2..000000000000
--- a/include/asm-sh64/unistd.h
+++ /dev/null
@@ -1,417 +0,0 @@
-#ifndef __ASM_SH64_UNISTD_H
-#define __ASM_SH64_UNISTD_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/unistd.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003 - 2007 Paul Mundt
- * Copyright (C) 2004  Sean McGoogan
- *
- * This file contains the system call numbers.
- *
- */
-
-#define __NR_restart_syscall	  0
-#define __NR_exit		  1
-#define __NR_fork		  2
-#define __NR_read		  3
-#define __NR_write		  4
-#define __NR_open		  5
-#define __NR_close		  6
-#define __NR_waitpid		  7
-#define __NR_creat		  8
-#define __NR_link		  9
-#define __NR_unlink		 10
-#define __NR_execve		 11
-#define __NR_chdir		 12
-#define __NR_time		 13
-#define __NR_mknod		 14
-#define __NR_chmod		 15
-#define __NR_lchown		 16
-#define __NR_break		 17
-#define __NR_oldstat		 18
-#define __NR_lseek		 19
-#define __NR_getpid		 20
-#define __NR_mount		 21
-#define __NR_umount		 22
-#define __NR_setuid		 23
-#define __NR_getuid		 24
-#define __NR_stime		 25
-#define __NR_ptrace		 26
-#define __NR_alarm		 27
-#define __NR_oldfstat		 28
-#define __NR_pause		 29
-#define __NR_utime		 30
-#define __NR_stty		 31
-#define __NR_gtty		 32
-#define __NR_access		 33
-#define __NR_nice		 34
-#define __NR_ftime		 35
-#define __NR_sync		 36
-#define __NR_kill		 37
-#define __NR_rename		 38
-#define __NR_mkdir		 39
-#define __NR_rmdir		 40
-#define __NR_dup		 41
-#define __NR_pipe		 42
-#define __NR_times		 43
-#define __NR_prof		 44
-#define __NR_brk		 45
-#define __NR_setgid		 46
-#define __NR_getgid		 47
-#define __NR_signal		 48
-#define __NR_geteuid		 49
-#define __NR_getegid		 50
-#define __NR_acct		 51
-#define __NR_umount2		 52
-#define __NR_lock		 53
-#define __NR_ioctl		 54
-#define __NR_fcntl		 55
-#define __NR_mpx		 56
-#define __NR_setpgid		 57
-#define __NR_ulimit		 58
-#define __NR_oldolduname	 59
-#define __NR_umask		 60
-#define __NR_chroot		 61
-#define __NR_ustat		 62
-#define __NR_dup2		 63
-#define __NR_getppid		 64
-#define __NR_getpgrp		 65
-#define __NR_setsid		 66
-#define __NR_sigaction		 67
-#define __NR_sgetmask		 68
-#define __NR_ssetmask		 69
-#define __NR_setreuid		 70
-#define __NR_setregid		 71
-#define __NR_sigsuspend		 72
-#define __NR_sigpending		 73
-#define __NR_sethostname	 74
-#define __NR_setrlimit		 75
-#define __NR_getrlimit	 	 76	/* Back compatible 2Gig limited rlimit */
-#define __NR_getrusage		 77
-#define __NR_gettimeofday	 78
-#define __NR_settimeofday	 79
-#define __NR_getgroups		 80
-#define __NR_setgroups		 81
-#define __NR_select		 82
-#define __NR_symlink		 83
-#define __NR_oldlstat		 84
-#define __NR_readlink		 85
-#define __NR_uselib		 86
-#define __NR_swapon		 87
-#define __NR_reboot		 88
-#define __NR_readdir		 89
-#define __NR_mmap		 90
-#define __NR_munmap		 91
-#define __NR_truncate		 92
-#define __NR_ftruncate		 93
-#define __NR_fchmod		 94
-#define __NR_fchown		 95
-#define __NR_getpriority	 96
-#define __NR_setpriority	 97
-#define __NR_profil		 98
-#define __NR_statfs		 99
-#define __NR_fstatfs		100
-#define __NR_ioperm		101
-#define __NR_socketcall		102	/* old implementation of socket systemcall */
-#define __NR_syslog		103
-#define __NR_setitimer		104
-#define __NR_getitimer		105
-#define __NR_stat		106
-#define __NR_lstat		107
-#define __NR_fstat		108
-#define __NR_olduname		109
-#define __NR_iopl		110
-#define __NR_vhangup		111
-#define __NR_idle		112
-#define __NR_vm86old		113
-#define __NR_wait4		114
-#define __NR_swapoff		115
-#define __NR_sysinfo		116
-#define __NR_ipc		117
-#define __NR_fsync		118
-#define __NR_sigreturn		119
-#define __NR_clone		120
-#define __NR_setdomainname	121
-#define __NR_uname		122
-#define __NR_modify_ldt		123
-#define __NR_adjtimex		124
-#define __NR_mprotect		125
-#define __NR_sigprocmask	126
-#define __NR_create_module	127
-#define __NR_init_module	128
-#define __NR_delete_module	129
-#define __NR_get_kernel_syms	130
-#define __NR_quotactl		131
-#define __NR_getpgid		132
-#define __NR_fchdir		133
-#define __NR_bdflush		134
-#define __NR_sysfs		135
-#define __NR_personality	136
-#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
-#define __NR_setfsuid		138
-#define __NR_setfsgid		139
-#define __NR__llseek		140
-#define __NR_getdents		141
-#define __NR__newselect		142
-#define __NR_flock		143
-#define __NR_msync		144
-#define __NR_readv		145
-#define __NR_writev		146
-#define __NR_getsid		147
-#define __NR_fdatasync		148
-#define __NR__sysctl		149
-#define __NR_mlock		150
-#define __NR_munlock		151
-#define __NR_mlockall		152
-#define __NR_munlockall		153
-#define __NR_sched_setparam		154
-#define __NR_sched_getparam		155
-#define __NR_sched_setscheduler		156
-#define __NR_sched_getscheduler		157
-#define __NR_sched_yield		158
-#define __NR_sched_get_priority_max	159
-#define __NR_sched_get_priority_min	160
-#define __NR_sched_rr_get_interval	161
-#define __NR_nanosleep		162
-#define __NR_mremap		163
-#define __NR_setresuid		164
-#define __NR_getresuid		165
-#define __NR_vm86		166
-#define __NR_query_module	167
-#define __NR_poll		168
-#define __NR_nfsservctl		169
-#define __NR_setresgid		170
-#define __NR_getresgid		171
-#define __NR_prctl              172
-#define __NR_rt_sigreturn	173
-#define __NR_rt_sigaction	174
-#define __NR_rt_sigprocmask	175
-#define __NR_rt_sigpending	176
-#define __NR_rt_sigtimedwait	177
-#define __NR_rt_sigqueueinfo	178
-#define __NR_rt_sigsuspend	179
-#define __NR_pread64		180
-#define __NR_pwrite64		181
-#define __NR_chown		182
-#define __NR_getcwd		183
-#define __NR_capget		184
-#define __NR_capset		185
-#define __NR_sigaltstack	186
-#define __NR_sendfile		187
-#define __NR_streams1		188	/* some people actually want it */
-#define __NR_streams2		189	/* some people actually want it */
-#define __NR_vfork		190
-#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
-#define __NR_mmap2		192
-#define __NR_truncate64		193
-#define __NR_ftruncate64	194
-#define __NR_stat64		195
-#define __NR_lstat64		196
-#define __NR_fstat64		197
-#define __NR_lchown32		198
-#define __NR_getuid32		199
-#define __NR_getgid32		200
-#define __NR_geteuid32		201
-#define __NR_getegid32		202
-#define __NR_setreuid32		203
-#define __NR_setregid32		204
-#define __NR_getgroups32	205
-#define __NR_setgroups32	206
-#define __NR_fchown32		207
-#define __NR_setresuid32	208
-#define __NR_getresuid32	209
-#define __NR_setresgid32	210
-#define __NR_getresgid32	211
-#define __NR_chown32		212
-#define __NR_setuid32		213
-#define __NR_setgid32		214
-#define __NR_setfsuid32		215
-#define __NR_setfsgid32		216
-#define __NR_pivot_root		217
-#define __NR_mincore		218
-#define __NR_madvise		219
-
-/* Non-multiplexed socket family */
-#define __NR_socket		220
-#define __NR_bind		221
-#define __NR_connect		222
-#define __NR_listen		223
-#define __NR_accept		224
-#define __NR_getsockname	225
-#define __NR_getpeername	226
-#define __NR_socketpair		227
-#define __NR_send		228
-#define __NR_sendto		229
-#define __NR_recv		230
-#define __NR_recvfrom		231
-#define __NR_shutdown		232
-#define __NR_setsockopt		233
-#define __NR_getsockopt		234
-#define __NR_sendmsg		235
-#define __NR_recvmsg		236
-
-/* Non-multiplexed IPC family */
-#define __NR_semop		237
-#define __NR_semget		238
-#define __NR_semctl		239
-#define __NR_msgsnd		240
-#define __NR_msgrcv		241
-#define __NR_msgget		242
-#define __NR_msgctl		243
-#if 0
-#define __NR_shmatcall		244
-#endif
-#define __NR_shmdt		245
-#define __NR_shmget		246
-#define __NR_shmctl		247
-
-#define __NR_getdents64		248
-#define __NR_fcntl64		249
-/* 223 is unused */
-#define __NR_gettid		252
-#define __NR_readahead		253
-#define __NR_setxattr		254
-#define __NR_lsetxattr		255
-#define __NR_fsetxattr		256
-#define __NR_getxattr		257
-#define __NR_lgetxattr		258
-#define __NR_fgetxattr		269
-#define __NR_listxattr		260
-#define __NR_llistxattr		261
-#define __NR_flistxattr		262
-#define __NR_removexattr	263
-#define __NR_lremovexattr	264
-#define __NR_fremovexattr	265
-#define __NR_tkill		266
-#define __NR_sendfile64		267
-#define __NR_futex		268
-#define __NR_sched_setaffinity	269
-#define __NR_sched_getaffinity	270
-#define __NR_set_thread_area	271
-#define __NR_get_thread_area	272
-#define __NR_io_setup		273
-#define __NR_io_destroy		274
-#define __NR_io_getevents	275
-#define __NR_io_submit		276
-#define __NR_io_cancel		277
-#define __NR_fadvise64		278
-#define __NR_exit_group		280
-
-#define __NR_lookup_dcookie	281
-#define __NR_epoll_create	282
-#define __NR_epoll_ctl		283
-#define __NR_epoll_wait		284
-#define __NR_remap_file_pages	285
-#define __NR_set_tid_address	286
-#define __NR_timer_create	287
-#define __NR_timer_settime	(__NR_timer_create+1)
-#define __NR_timer_gettime	(__NR_timer_create+2)
-#define __NR_timer_getoverrun	(__NR_timer_create+3)
-#define __NR_timer_delete	(__NR_timer_create+4)
-#define __NR_clock_settime	(__NR_timer_create+5)
-#define __NR_clock_gettime	(__NR_timer_create+6)
-#define __NR_clock_getres	(__NR_timer_create+7)
-#define __NR_clock_nanosleep	(__NR_timer_create+8)
-#define __NR_statfs64		296
-#define __NR_fstatfs64		297
-#define __NR_tgkill		298
-#define __NR_utimes		299
-#define __NR_fadvise64_64	300
-#define __NR_vserver		301
-#define __NR_mbind              302
-#define __NR_get_mempolicy      303
-#define __NR_set_mempolicy      304
-#define __NR_mq_open            305
-#define __NR_mq_unlink          (__NR_mq_open+1)
-#define __NR_mq_timedsend       (__NR_mq_open+2)
-#define __NR_mq_timedreceive    (__NR_mq_open+3)
-#define __NR_mq_notify          (__NR_mq_open+4)
-#define __NR_mq_getsetattr      (__NR_mq_open+5)
-#define __NR_kexec_load		311
-#define __NR_waitid		312
-#define __NR_add_key		313
-#define __NR_request_key	314
-#define __NR_keyctl		315
-#define __NR_ioprio_set		316
-#define __NR_ioprio_get		317
-#define __NR_inotify_init	318
-#define __NR_inotify_add_watch	319
-#define __NR_inotify_rm_watch	320
-/* 321 is unused */
-#define __NR_migrate_pages	322
-#define __NR_openat		323
-#define __NR_mkdirat		324
-#define __NR_mknodat		325
-#define __NR_fchownat		326
-#define __NR_futimesat		327
-#define __NR_fstatat64		328
-#define __NR_unlinkat		329
-#define __NR_renameat		330
-#define __NR_linkat		331
-#define __NR_symlinkat		332
-#define __NR_readlinkat		333
-#define __NR_fchmodat		334
-#define __NR_faccessat		335
-#define __NR_pselect6		336
-#define __NR_ppoll		337
-#define __NR_unshare		338
-#define __NR_set_robust_list	339
-#define __NR_get_robust_list	340
-#define __NR_splice		341
-#define __NR_sync_file_range	342
-#define __NR_tee		343
-#define __NR_vmsplice		344
-#define __NR_move_pages		345
-#define __NR_getcpu		346
-#define __NR_epoll_pwait	347
-#define __NR_utimensat		348
-#define __NR_signalfd		349
-#define __NR_timerfd		350
-#define __NR_eventfd		351
-#define __NR_fallocate		352
-
-#ifdef __KERNEL__
-
-#define NR_syscalls 353
-
-#define __ARCH_WANT_IPC_PARSE_VERSION
-#define __ARCH_WANT_OLD_READDIR
-#define __ARCH_WANT_OLD_STAT
-#define __ARCH_WANT_STAT64
-#define __ARCH_WANT_SYS_ALARM
-#define __ARCH_WANT_SYS_GETHOSTNAME
-#define __ARCH_WANT_SYS_PAUSE
-#define __ARCH_WANT_SYS_SGETMASK
-#define __ARCH_WANT_SYS_SIGNAL
-#define __ARCH_WANT_SYS_TIME
-#define __ARCH_WANT_SYS_UTIME
-#define __ARCH_WANT_SYS_WAITPID
-#define __ARCH_WANT_SYS_SOCKETCALL
-#define __ARCH_WANT_SYS_FADVISE64
-#define __ARCH_WANT_SYS_GETPGRP
-#define __ARCH_WANT_SYS_LLSEEK
-#define __ARCH_WANT_SYS_NICE
-#define __ARCH_WANT_SYS_OLD_GETRLIMIT
-#define __ARCH_WANT_SYS_OLDUMOUNT
-#define __ARCH_WANT_SYS_SIGPENDING
-#define __ARCH_WANT_SYS_SIGPROCMASK
-#define __ARCH_WANT_SYS_RT_SIGACTION
-
-/*
- * "Conditional" syscalls
- *
- * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
- * but it doesn't work on all toolchains, so we just do it by hand
- */
-#ifndef cond_syscall
-#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall")
-#endif
-
-#endif /* __KERNEL__ */
-#endif /* __ASM_SH64_UNISTD_H */
-- 
cgit v1.2.3


From 343ac72248d360f1fae72176aca1117be19189ec Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sun, 11 Nov 2007 18:11:18 +0900
Subject: sh: Move over the SH-5 entry.S.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/Makefile     |    6 +
 arch/sh/kernel/cpu/sh5/Makefile |    1 +
 arch/sh/kernel/cpu/sh5/entry.S  | 2100 ++++++++++++++++++++++++++++++++++++++
 arch/sh64/kernel/entry.S        | 2102 ---------------------------------------
 include/asm-sh/processor.h      |   17 +-
 5 files changed, 2119 insertions(+), 2107 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh5/Makefile
 create mode 100644 arch/sh/kernel/cpu/sh5/entry.S
 delete mode 100644 arch/sh64/kernel/entry.S

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index d055a3ea6b4b..f471d242774e 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -6,8 +6,14 @@ obj-$(CONFIG_CPU_SH2)		= sh2/
 obj-$(CONFIG_CPU_SH2A)		= sh2a/
 obj-$(CONFIG_CPU_SH3)		= sh3/
 obj-$(CONFIG_CPU_SH4)		= sh4/
+obj-$(CONFIG_CPU_SH5)		= sh5/
+
+# Special cases for family ancestry.
+
 obj-$(CONFIG_CPU_SH4A)		+= sh4a/
 
+# Common interfaces.
+
 obj-$(CONFIG_UBC_WAKEUP)	+= ubc.o
 obj-$(CONFIG_SH_ADC)		+= adc.o
 
diff --git a/arch/sh/kernel/cpu/sh5/Makefile b/arch/sh/kernel/cpu/sh5/Makefile
new file mode 100644
index 000000000000..9778f9bdff3a
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh5/Makefile
@@ -0,0 +1 @@
+obj-y := entry.o
diff --git a/arch/sh/kernel/cpu/sh5/entry.S b/arch/sh/kernel/cpu/sh5/entry.S
new file mode 100644
index 000000000000..2f505a7cb5f9
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh5/entry.S
@@ -0,0 +1,2100 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * arch/sh64/kernel/entry.S
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2004, 2005  Paul Mundt
+ * Copyright (C) 2003, 2004 Richard Curnow
+ *
+ */
+#include <linux/errno.h>
+#include <linux/sys.h>
+#include <asm/cpu/registers.h>
+#include <asm/processor.h>
+#include <asm/unistd.h>
+#include <asm/thread_info.h>
+#include <asm/asm-offsets.h>
+
+/*
+ * SR fields.
+ */
+#define SR_ASID_MASK	0x00ff0000
+#define SR_FD_MASK	0x00008000
+#define SR_SS		0x08000000
+#define SR_BL		0x10000000
+#define SR_MD		0x40000000
+
+/*
+ * Event code.
+ */
+#define	EVENT_INTERRUPT		0
+#define	EVENT_FAULT_TLB		1
+#define	EVENT_FAULT_NOT_TLB	2
+#define	EVENT_DEBUG		3
+
+/* EXPEVT values */
+#define	RESET_CAUSE		0x20
+#define DEBUGSS_CAUSE		0x980
+
+/*
+ * Frame layout. Quad index.
+ */
+#define	FRAME_T(x)	FRAME_TBASE+(x*8)
+#define	FRAME_R(x)	FRAME_RBASE+(x*8)
+#define	FRAME_S(x)	FRAME_SBASE+(x*8)
+#define FSPC		0
+#define FSSR		1
+#define FSYSCALL_ID	2
+
+/* Arrange the save frame to be a multiple of 32 bytes long */
+#define FRAME_SBASE	0
+#define FRAME_RBASE	(FRAME_SBASE+(3*8))	/* SYSCALL_ID - SSR - SPC */
+#define FRAME_TBASE	(FRAME_RBASE+(63*8))	/* r0 - r62 */
+#define FRAME_PBASE	(FRAME_TBASE+(8*8))	/* tr0 -tr7 */
+#define	FRAME_SIZE	(FRAME_PBASE+(2*8))	/* pad0-pad1 */
+
+#define FP_FRAME_SIZE	FP_FRAME_BASE+(33*8)	/* dr0 - dr31 + fpscr */
+#define FP_FRAME_BASE	0
+
+#define	SAVED_R2	0*8
+#define	SAVED_R3	1*8
+#define	SAVED_R4	2*8
+#define	SAVED_R5	3*8
+#define	SAVED_R18	4*8
+#define	SAVED_R6	5*8
+#define	SAVED_TR0	6*8
+
+/* These are the registers saved in the TLB path that aren't saved in the first
+   level of the normal one. */
+#define	TLB_SAVED_R25	7*8
+#define	TLB_SAVED_TR1	8*8
+#define	TLB_SAVED_TR2	9*8
+#define	TLB_SAVED_TR3	10*8
+#define	TLB_SAVED_TR4	11*8
+/* Save R0/R1 : PT-migrating compiler currently dishounours -ffixed-r0 and -ffixed-r1 causing
+   breakage otherwise. */
+#define	TLB_SAVED_R0	12*8
+#define	TLB_SAVED_R1	13*8
+
+#define CLI()				\
+	getcon	SR, r6;			\
+	ori	r6, 0xf0, r6;		\
+	putcon	r6, SR;
+
+#define STI()				\
+	getcon	SR, r6;			\
+	andi	r6, ~0xf0, r6;		\
+	putcon	r6, SR;
+
+#ifdef CONFIG_PREEMPT
+#  define preempt_stop()	CLI()
+#else
+#  define preempt_stop()
+#  define resume_kernel		restore_all
+#endif
+
+	.section	.data, "aw"
+
+#define FAST_TLBMISS_STACK_CACHELINES 4
+#define FAST_TLBMISS_STACK_QUADWORDS (4*FAST_TLBMISS_STACK_CACHELINES)
+
+/* Register back-up area for all exceptions */
+	.balign	32
+	/* Allow for 16 quadwords to be pushed by fast tlbmiss handling
+	 * register saves etc. */
+	.fill FAST_TLBMISS_STACK_QUADWORDS, 8, 0x0
+/* This is 32 byte aligned by construction */
+/* Register back-up area for all exceptions */
+reg_save_area:
+	.quad	0
+	.quad	0
+	.quad	0
+	.quad	0
+
+	.quad	0
+	.quad	0
+	.quad	0
+	.quad	0
+
+	.quad	0
+	.quad	0
+	.quad	0
+	.quad	0
+
+	.quad	0
+	.quad   0
+
+/* Save area for RESVEC exceptions. We cannot use reg_save_area because of
+ * reentrancy. Note this area may be accessed via physical address.
+ * Align so this fits a whole single cache line, for ease of purging.
+ */
+	.balign 32,0,32
+resvec_save_area:
+	.quad	0
+	.quad	0
+	.quad	0
+	.quad	0
+	.quad	0
+	.balign 32,0,32
+
+/* Jump table of 3rd level handlers  */
+trap_jtable:
+	.long	do_exception_error		/* 0x000 */
+	.long	do_exception_error		/* 0x020 */
+	.long	tlb_miss_load				/* 0x040 */
+	.long	tlb_miss_store				/* 0x060 */
+	! ARTIFICIAL pseudo-EXPEVT setting
+	.long	do_debug_interrupt		/* 0x080 */
+	.long	tlb_miss_load				/* 0x0A0 */
+	.long	tlb_miss_store				/* 0x0C0 */
+	.long	do_address_error_load	/* 0x0E0 */
+	.long	do_address_error_store	/* 0x100 */
+#ifdef CONFIG_SH_FPU
+	.long	do_fpu_error		/* 0x120 */
+#else
+	.long	do_exception_error		/* 0x120 */
+#endif
+	.long	do_exception_error		/* 0x140 */
+	.long	system_call				/* 0x160 */
+	.long	do_reserved_inst		/* 0x180 */
+	.long	do_illegal_slot_inst	/* 0x1A0 */
+	.long	do_NMI			/* 0x1C0 */
+	.long	do_exception_error		/* 0x1E0 */
+	.rept 15
+		.long do_IRQ		/* 0x200 - 0x3C0 */
+	.endr
+	.long	do_exception_error		/* 0x3E0 */
+	.rept 32
+		.long do_IRQ		/* 0x400 - 0x7E0 */
+	.endr
+	.long	fpu_error_or_IRQA			/* 0x800 */
+	.long	fpu_error_or_IRQB			/* 0x820 */
+	.long	do_IRQ			/* 0x840 */
+	.long	do_IRQ			/* 0x860 */
+	.rept 6
+		.long do_exception_error	/* 0x880 - 0x920 */
+	.endr
+	.long	do_software_break_point	/* 0x940 */
+	.long	do_exception_error		/* 0x960 */
+	.long	do_single_step		/* 0x980 */
+
+	.rept 3
+		.long do_exception_error	/* 0x9A0 - 0x9E0 */
+	.endr
+	.long	do_IRQ			/* 0xA00 */
+	.long	do_IRQ			/* 0xA20 */
+	.long	itlb_miss_or_IRQ			/* 0xA40 */
+	.long	do_IRQ			/* 0xA60 */
+	.long	do_IRQ			/* 0xA80 */
+	.long	itlb_miss_or_IRQ			/* 0xAA0 */
+	.long	do_exception_error		/* 0xAC0 */
+	.long	do_address_error_exec	/* 0xAE0 */
+	.rept 8
+		.long do_exception_error	/* 0xB00 - 0xBE0 */
+	.endr
+	.rept 18
+		.long do_IRQ		/* 0xC00 - 0xE20 */
+	.endr
+
+	.section	.text64, "ax"
+
+/*
+ * --- Exception/Interrupt/Event Handling Section
+ */
+
+/*
+ * VBR and RESVEC blocks.
+ *
+ * First level handler for VBR-based exceptions.
+ *
+ * To avoid waste of space, align to the maximum text block size.
+ * This is assumed to be at most 128 bytes or 32 instructions.
+ * DO NOT EXCEED 32 instructions on the first level handlers !
+ *
+ * Also note that RESVEC is contained within the VBR block
+ * where the room left (1KB - TEXT_SIZE) allows placing
+ * the RESVEC block (at most 512B + TEXT_SIZE).
+ *
+ * So first (and only) level handler for RESVEC-based exceptions.
+ *
+ * Where the fault/interrupt is handled (not_a_tlb_miss, tlb_miss
+ * and interrupt) we are a lot tight with register space until
+ * saving onto the stack frame, which is done in handle_exception().
+ *
+ */
+
+#define	TEXT_SIZE 	128
+#define	BLOCK_SIZE 	1664 		/* Dynamic check, 13*128 */
+
+	.balign TEXT_SIZE
+LVBR_block:
+	.space	256, 0			/* Power-on class handler, */
+					/* not required here       */
+not_a_tlb_miss:
+	synco	/* TAKum03020 (but probably a good idea anyway.) */
+	/* Save original stack pointer into KCR1 */
+	putcon	SP, KCR1
+
+	/* Save other original registers into reg_save_area */
+        movi  reg_save_area, SP
+	st.q	SP, SAVED_R2, r2
+	st.q	SP, SAVED_R3, r3
+	st.q	SP, SAVED_R4, r4
+	st.q	SP, SAVED_R5, r5
+	st.q	SP, SAVED_R6, r6
+	st.q	SP, SAVED_R18, r18
+	gettr	tr0, r3
+	st.q	SP, SAVED_TR0, r3
+
+	/* Set args for Non-debug, Not a TLB miss class handler */
+	getcon	EXPEVT, r2
+	movi	ret_from_exception, r3
+	ori	r3, 1, r3
+	movi	EVENT_FAULT_NOT_TLB, r4
+	or	SP, ZERO, r5
+	getcon	KCR1, SP
+	pta	handle_exception, tr0
+	blink	tr0, ZERO
+
+	.balign 256
+	! VBR+0x200
+	nop
+	.balign 256
+	! VBR+0x300
+	nop
+	.balign 256
+	/*
+	 * Instead of the natural .balign 1024 place RESVEC here
+	 * respecting the final 1KB alignment.
+	 */
+	.balign TEXT_SIZE
+	/*
+	 * Instead of '.space 1024-TEXT_SIZE' place the RESVEC
+	 * block making sure the final alignment is correct.
+	 */
+tlb_miss:
+	synco	/* TAKum03020 (but probably a good idea anyway.) */
+	putcon	SP, KCR1
+	movi	reg_save_area, SP
+	/* SP is guaranteed 32-byte aligned. */
+	st.q	SP, TLB_SAVED_R0 , r0
+	st.q	SP, TLB_SAVED_R1 , r1
+	st.q	SP, SAVED_R2 , r2
+	st.q	SP, SAVED_R3 , r3
+	st.q	SP, SAVED_R4 , r4
+	st.q	SP, SAVED_R5 , r5
+	st.q	SP, SAVED_R6 , r6
+	st.q	SP, SAVED_R18, r18
+
+	/* Save R25 for safety; as/ld may want to use it to achieve the call to
+	 * the code in mm/tlbmiss.c */
+	st.q	SP, TLB_SAVED_R25, r25
+	gettr	tr0, r2
+	gettr	tr1, r3
+	gettr	tr2, r4
+	gettr	tr3, r5
+	gettr	tr4, r18
+	st.q	SP, SAVED_TR0 , r2
+	st.q	SP, TLB_SAVED_TR1 , r3
+	st.q	SP, TLB_SAVED_TR2 , r4
+	st.q	SP, TLB_SAVED_TR3 , r5
+	st.q	SP, TLB_SAVED_TR4 , r18
+
+	pt	do_fast_page_fault, tr0
+	getcon	SSR, r2
+	getcon	EXPEVT, r3
+	getcon	TEA, r4
+	shlri	r2, 30, r2
+	andi	r2, 1, r2	/* r2 = SSR.MD */
+	blink 	tr0, LINK
+
+	pt	fixup_to_invoke_general_handler, tr1
+
+	/* If the fast path handler fixed the fault, just drop through quickly
+	   to the restore code right away to return to the excepting context.
+	   */
+	beqi/u	r2, 0, tr1
+
+fast_tlb_miss_restore:
+	ld.q	SP, SAVED_TR0, r2
+	ld.q	SP, TLB_SAVED_TR1, r3
+	ld.q	SP, TLB_SAVED_TR2, r4
+
+	ld.q	SP, TLB_SAVED_TR3, r5
+	ld.q	SP, TLB_SAVED_TR4, r18
+
+	ptabs	r2, tr0
+	ptabs	r3, tr1
+	ptabs	r4, tr2
+	ptabs	r5, tr3
+	ptabs	r18, tr4
+
+	ld.q	SP, TLB_SAVED_R0, r0
+	ld.q	SP, TLB_SAVED_R1, r1
+	ld.q	SP, SAVED_R2, r2
+	ld.q	SP, SAVED_R3, r3
+	ld.q	SP, SAVED_R4, r4
+	ld.q	SP, SAVED_R5, r5
+	ld.q	SP, SAVED_R6, r6
+	ld.q	SP, SAVED_R18, r18
+	ld.q	SP, TLB_SAVED_R25, r25
+
+	getcon	KCR1, SP
+	rte
+	nop /* for safety, in case the code is run on sh5-101 cut1.x */
+
+fixup_to_invoke_general_handler:
+
+	/* OK, new method.  Restore stuff that's not expected to get saved into
+	   the 'first-level' reg save area, then just fall through to setting
+	   up the registers and calling the second-level handler. */
+
+	/* 2nd level expects r2,3,4,5,6,18,tr0 to be saved.  So we must restore
+	   r25,tr1-4 and save r6 to get into the right state.  */
+
+	ld.q	SP, TLB_SAVED_TR1, r3
+	ld.q	SP, TLB_SAVED_TR2, r4
+	ld.q	SP, TLB_SAVED_TR3, r5
+	ld.q	SP, TLB_SAVED_TR4, r18
+	ld.q	SP, TLB_SAVED_R25, r25
+
+	ld.q	SP, TLB_SAVED_R0, r0
+	ld.q	SP, TLB_SAVED_R1, r1
+
+	ptabs/u	r3, tr1
+	ptabs/u	r4, tr2
+	ptabs/u	r5, tr3
+	ptabs/u	r18, tr4
+
+	/* Set args for Non-debug, TLB miss class handler */
+	getcon	EXPEVT, r2
+	movi	ret_from_exception, r3
+	ori	r3, 1, r3
+	movi	EVENT_FAULT_TLB, r4
+	or	SP, ZERO, r5
+	getcon	KCR1, SP
+	pta	handle_exception, tr0
+	blink	tr0, ZERO
+
+/* NB TAKE GREAT CARE HERE TO ENSURE THAT THE INTERRUPT CODE
+   DOES END UP AT VBR+0x600 */
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
+	.balign 256
+	/* VBR + 0x600 */
+
+interrupt:
+	synco	/* TAKum03020 (but probably a good idea anyway.) */
+	/* Save original stack pointer into KCR1 */
+	putcon	SP, KCR1
+
+	/* Save other original registers into reg_save_area */
+        movi  reg_save_area, SP
+	st.q	SP, SAVED_R2, r2
+	st.q	SP, SAVED_R3, r3
+	st.q	SP, SAVED_R4, r4
+	st.q	SP, SAVED_R5, r5
+	st.q	SP, SAVED_R6, r6
+	st.q	SP, SAVED_R18, r18
+	gettr	tr0, r3
+	st.q	SP, SAVED_TR0, r3
+
+	/* Set args for interrupt class handler */
+	getcon	INTEVT, r2
+	movi	ret_from_irq, r3
+	ori	r3, 1, r3
+	movi	EVENT_INTERRUPT, r4
+	or	SP, ZERO, r5
+	getcon	KCR1, SP
+	pta	handle_exception, tr0
+	blink	tr0, ZERO
+	.balign	TEXT_SIZE		/* let's waste the bare minimum */
+
+LVBR_block_end:				/* Marker. Used for total checking */
+
+	.balign 256
+LRESVEC_block:
+	/* Panic handler. Called with MMU off. Possible causes/actions:
+	 * - Reset:		Jump to program start.
+	 * - Single Step:	Turn off Single Step & return.
+	 * - Others:		Call panic handler, passing PC as arg.
+	 *			(this may need to be extended...)
+	 */
+reset_or_panic:
+	synco	/* TAKum03020 (but probably a good idea anyway.) */
+	putcon	SP, DCR
+	/* First save r0-1 and tr0, as we need to use these */
+	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
+	st.q	SP, 0, r0
+	st.q	SP, 8, r1
+	gettr	tr0, r0
+	st.q	SP, 32, r0
+
+	/* Check cause */
+	getcon	EXPEVT, r0
+	movi	RESET_CAUSE, r1
+	sub	r1, r0, r1		/* r1=0 if reset */
+	movi	_stext-CONFIG_CACHED_MEMORY_OFFSET, r0
+	ori	r0, 1, r0
+	ptabs	r0, tr0
+	beqi	r1, 0, tr0		/* Jump to start address if reset */
+
+	getcon	EXPEVT, r0
+	movi	DEBUGSS_CAUSE, r1
+	sub	r1, r0, r1		/* r1=0 if single step */
+	pta	single_step_panic, tr0
+	beqi	r1, 0, tr0		/* jump if single step */
+
+	/* Now jump to where we save the registers. */
+	movi	panic_stash_regs-CONFIG_CACHED_MEMORY_OFFSET, r1
+	ptabs	r1, tr0
+	blink	tr0, r63
+
+single_step_panic:
+	/* We are in a handler with Single Step set. We need to resume the
+	 * handler, by turning on MMU & turning off Single Step. */
+	getcon	SSR, r0
+	movi	SR_MMU, r1
+	or	r0, r1, r0
+	movi	~SR_SS, r1
+	and	r0, r1, r0
+	putcon	r0, SSR
+	/* Restore EXPEVT, as the rte won't do this */
+	getcon	PEXPEVT, r0
+	putcon	r0, EXPEVT
+	/* Restore regs */
+	ld.q	SP, 32, r0
+	ptabs	r0, tr0
+	ld.q	SP, 0, r0
+	ld.q	SP, 8, r1
+	getcon	DCR, SP
+	synco
+	rte
+
+
+	.balign	256
+debug_exception:
+	synco	/* TAKum03020 (but probably a good idea anyway.) */
+	/*
+	 * Single step/software_break_point first level handler.
+	 * Called with MMU off, so the first thing we do is enable it
+	 * by doing an rte with appropriate SSR.
+	 */
+	putcon	SP, DCR
+	/* Save SSR & SPC, together with R0 & R1, as we need to use 2 regs. */
+	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
+
+	/* With the MMU off, we are bypassing the cache, so purge any
+         * data that will be made stale by the following stores.
+         */
+	ocbp	SP, 0
+	synco
+
+	st.q	SP, 0, r0
+	st.q	SP, 8, r1
+	getcon	SPC, r0
+	st.q	SP, 16, r0
+	getcon	SSR, r0
+	st.q	SP, 24, r0
+
+	/* Enable MMU, block exceptions, set priv mode, disable single step */
+	movi	SR_MMU | SR_BL | SR_MD, r1
+	or	r0, r1, r0
+	movi	~SR_SS, r1
+	and	r0, r1, r0
+	putcon	r0, SSR
+	/* Force control to debug_exception_2 when rte is executed */
+	movi	debug_exeception_2, r0
+	ori	r0, 1, r0      /* force SHmedia, just in case */
+	putcon	r0, SPC
+	getcon	DCR, SP
+	synco
+	rte
+debug_exeception_2:
+	/* Restore saved regs */
+	putcon	SP, KCR1
+	movi	resvec_save_area, SP
+	ld.q	SP, 24, r0
+	putcon	r0, SSR
+	ld.q	SP, 16, r0
+	putcon	r0, SPC
+	ld.q	SP, 0, r0
+	ld.q	SP, 8, r1
+
+	/* Save other original registers into reg_save_area */
+        movi  reg_save_area, SP
+	st.q	SP, SAVED_R2, r2
+	st.q	SP, SAVED_R3, r3
+	st.q	SP, SAVED_R4, r4
+	st.q	SP, SAVED_R5, r5
+	st.q	SP, SAVED_R6, r6
+	st.q	SP, SAVED_R18, r18
+	gettr	tr0, r3
+	st.q	SP, SAVED_TR0, r3
+
+	/* Set args for debug class handler */
+	getcon	EXPEVT, r2
+	movi	ret_from_exception, r3
+	ori	r3, 1, r3
+	movi	EVENT_DEBUG, r4
+	or	SP, ZERO, r5
+	getcon	KCR1, SP
+	pta	handle_exception, tr0
+	blink	tr0, ZERO
+
+	.balign	256
+debug_interrupt:
+	/* !!! WE COME HERE IN REAL MODE !!! */
+	/* Hook-up debug interrupt to allow various debugging options to be
+	 * hooked into its handler. */
+	/* Save original stack pointer into KCR1 */
+	synco
+	putcon	SP, KCR1
+	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
+	ocbp	SP, 0
+	ocbp	SP, 32
+	synco
+
+	/* Save other original registers into reg_save_area thru real addresses */
+	st.q	SP, SAVED_R2, r2
+	st.q	SP, SAVED_R3, r3
+	st.q	SP, SAVED_R4, r4
+	st.q	SP, SAVED_R5, r5
+	st.q	SP, SAVED_R6, r6
+	st.q	SP, SAVED_R18, r18
+	gettr	tr0, r3
+	st.q	SP, SAVED_TR0, r3
+
+	/* move (spc,ssr)->(pspc,pssr).  The rte will shift
+	   them back again, so that they look like the originals
+	   as far as the real handler code is concerned. */
+	getcon	spc, r6
+	putcon	r6, pspc
+	getcon	ssr, r6
+	putcon	r6, pssr
+
+	! construct useful SR for handle_exception
+	movi	3, r6
+	shlli	r6, 30, r6
+	getcon	sr, r18
+	or	r18, r6, r6
+	putcon	r6, ssr
+
+	! SSR is now the current SR with the MD and MMU bits set
+	! i.e. the rte will switch back to priv mode and put
+	! the mmu back on
+
+	! construct spc
+	movi	handle_exception, r18
+	ori	r18, 1, r18		! for safety (do we need this?)
+	putcon	r18, spc
+
+	/* Set args for Non-debug, Not a TLB miss class handler */
+
+	! EXPEVT==0x80 is unused, so 'steal' this value to put the
+	! debug interrupt handler in the vectoring table
+	movi	0x80, r2
+	movi	ret_from_exception, r3
+	ori	r3, 1, r3
+	movi	EVENT_FAULT_NOT_TLB, r4
+
+	or	SP, ZERO, r5
+	movi	CONFIG_CACHED_MEMORY_OFFSET, r6
+	add	r6, r5, r5
+	getcon	KCR1, SP
+
+	synco	! for safety
+	rte	! -> handle_exception, switch back to priv mode again
+
+LRESVEC_block_end:			/* Marker. Unused. */
+
+	.balign	TEXT_SIZE
+
+/*
+ * Second level handler for VBR-based exceptions. Pre-handler.
+ * In common to all stack-frame sensitive handlers.
+ *
+ * Inputs:
+ * (KCR0) Current [current task union]
+ * (KCR1) Original SP
+ * (r2)   INTEVT/EXPEVT
+ * (r3)   appropriate return address
+ * (r4)   Event (0 = interrupt, 1 = TLB miss fault, 2 = Not TLB miss fault, 3=debug)
+ * (r5)   Pointer to reg_save_area
+ * (SP)   Original SP
+ *
+ * Available registers:
+ * (r6)
+ * (r18)
+ * (tr0)
+ *
+ */
+handle_exception:
+	/* Common 2nd level handler. */
+
+	/* First thing we need an appropriate stack pointer */
+	getcon	SSR, r6
+	shlri	r6, 30, r6
+	andi	r6, 1, r6
+	pta	stack_ok, tr0
+	bne	r6, ZERO, tr0		/* Original stack pointer is fine */
+
+	/* Set stack pointer for user fault */
+	getcon	KCR0, SP
+	movi	THREAD_SIZE, r6		/* Point to the end */
+	add	SP, r6, SP
+
+stack_ok:
+
+/* DEBUG : check for underflow/overflow of the kernel stack */
+	pta	no_underflow, tr0
+	getcon  KCR0, r6
+	movi	1024, r18
+	add	r6, r18, r6
+	bge	SP, r6, tr0 	! ? below 1k from bottom of stack : danger zone
+
+/* Just panic to cause a crash. */
+bad_sp:
+	ld.b	r63, 0, r6
+	nop
+
+no_underflow:
+	pta	bad_sp, tr0
+	getcon	kcr0, r6
+	movi	THREAD_SIZE, r18
+	add	r18, r6, r6
+	bgt	SP, r6, tr0	! sp above the stack
+
+	/* Make some room for the BASIC frame. */
+	movi	-(FRAME_SIZE), r6
+	add	SP, r6, SP
+
+/* Could do this with no stalling if we had another spare register, but the
+   code below will be OK. */
+	ld.q	r5, SAVED_R2, r6
+	ld.q	r5, SAVED_R3, r18
+	st.q	SP, FRAME_R(2), r6
+	ld.q	r5, SAVED_R4, r6
+	st.q	SP, FRAME_R(3), r18
+	ld.q	r5, SAVED_R5, r18
+	st.q	SP, FRAME_R(4), r6
+	ld.q	r5, SAVED_R6, r6
+	st.q	SP, FRAME_R(5), r18
+	ld.q	r5, SAVED_R18, r18
+	st.q	SP, FRAME_R(6), r6
+	ld.q	r5, SAVED_TR0, r6
+	st.q	SP, FRAME_R(18), r18
+	st.q	SP, FRAME_T(0), r6
+
+	/* Keep old SP around */
+	getcon	KCR1, r6
+
+	/* Save the rest of the general purpose registers */
+	st.q	SP, FRAME_R(0), r0
+	st.q	SP, FRAME_R(1), r1
+	st.q	SP, FRAME_R(7), r7
+	st.q	SP, FRAME_R(8), r8
+	st.q	SP, FRAME_R(9), r9
+	st.q	SP, FRAME_R(10), r10
+	st.q	SP, FRAME_R(11), r11
+	st.q	SP, FRAME_R(12), r12
+	st.q	SP, FRAME_R(13), r13
+	st.q	SP, FRAME_R(14), r14
+
+	/* SP is somewhere else */
+	st.q	SP, FRAME_R(15), r6
+
+	st.q	SP, FRAME_R(16), r16
+	st.q	SP, FRAME_R(17), r17
+	/* r18 is saved earlier. */
+	st.q	SP, FRAME_R(19), r19
+	st.q	SP, FRAME_R(20), r20
+	st.q	SP, FRAME_R(21), r21
+	st.q	SP, FRAME_R(22), r22
+	st.q	SP, FRAME_R(23), r23
+	st.q	SP, FRAME_R(24), r24
+	st.q	SP, FRAME_R(25), r25
+	st.q	SP, FRAME_R(26), r26
+	st.q	SP, FRAME_R(27), r27
+	st.q	SP, FRAME_R(28), r28
+	st.q	SP, FRAME_R(29), r29
+	st.q	SP, FRAME_R(30), r30
+	st.q	SP, FRAME_R(31), r31
+	st.q	SP, FRAME_R(32), r32
+	st.q	SP, FRAME_R(33), r33
+	st.q	SP, FRAME_R(34), r34
+	st.q	SP, FRAME_R(35), r35
+	st.q	SP, FRAME_R(36), r36
+	st.q	SP, FRAME_R(37), r37
+	st.q	SP, FRAME_R(38), r38
+	st.q	SP, FRAME_R(39), r39
+	st.q	SP, FRAME_R(40), r40
+	st.q	SP, FRAME_R(41), r41
+	st.q	SP, FRAME_R(42), r42
+	st.q	SP, FRAME_R(43), r43
+	st.q	SP, FRAME_R(44), r44
+	st.q	SP, FRAME_R(45), r45
+	st.q	SP, FRAME_R(46), r46
+	st.q	SP, FRAME_R(47), r47
+	st.q	SP, FRAME_R(48), r48
+	st.q	SP, FRAME_R(49), r49
+	st.q	SP, FRAME_R(50), r50
+	st.q	SP, FRAME_R(51), r51
+	st.q	SP, FRAME_R(52), r52
+	st.q	SP, FRAME_R(53), r53
+	st.q	SP, FRAME_R(54), r54
+	st.q	SP, FRAME_R(55), r55
+	st.q	SP, FRAME_R(56), r56
+	st.q	SP, FRAME_R(57), r57
+	st.q	SP, FRAME_R(58), r58
+	st.q	SP, FRAME_R(59), r59
+	st.q	SP, FRAME_R(60), r60
+	st.q	SP, FRAME_R(61), r61
+	st.q	SP, FRAME_R(62), r62
+
+	/*
+	 * Save the S* registers.
+	 */
+	getcon	SSR, r61
+	st.q	SP, FRAME_S(FSSR), r61
+	getcon	SPC, r62
+	st.q	SP, FRAME_S(FSPC), r62
+	movi	-1, r62			/* Reset syscall_nr */
+	st.q	SP, FRAME_S(FSYSCALL_ID), r62
+
+	/* Save the rest of the target registers */
+	gettr	tr1, r6
+	st.q	SP, FRAME_T(1), r6
+	gettr	tr2, r6
+	st.q	SP, FRAME_T(2), r6
+	gettr	tr3, r6
+	st.q	SP, FRAME_T(3), r6
+	gettr	tr4, r6
+	st.q	SP, FRAME_T(4), r6
+	gettr	tr5, r6
+	st.q	SP, FRAME_T(5), r6
+	gettr	tr6, r6
+	st.q	SP, FRAME_T(6), r6
+	gettr	tr7, r6
+	st.q	SP, FRAME_T(7), r6
+
+	! setup FP so that unwinder can wind back through nested kernel mode
+	! exceptions
+	add	SP, ZERO, r14
+
+#ifdef CONFIG_POOR_MANS_STRACE
+	/* We've pushed all the registers now, so only r2-r4 hold anything
+	 * useful. Move them into callee save registers */
+	or	r2, ZERO, r28
+	or	r3, ZERO, r29
+	or	r4, ZERO, r30
+
+	/* Preserve r2 as the event code */
+	movi	evt_debug, r3
+	ori	r3, 1, r3
+	ptabs	r3, tr0
+
+	or	SP, ZERO, r6
+	getcon	TRA, r5
+	blink	tr0, LINK
+
+	or	r28, ZERO, r2
+	or	r29, ZERO, r3
+	or	r30, ZERO, r4
+#endif
+
+	/* For syscall and debug race condition, get TRA now */
+	getcon	TRA, r5
+
+	/* We are in a safe position to turn SR.BL off, but set IMASK=0xf
+	 * Also set FD, to catch FPU usage in the kernel.
+	 *
+	 * benedict.gaster@superh.com 29/07/2002
+	 *
+	 * On all SH5-101 revisions it is unsafe to raise the IMASK and at the
+	 * same time change BL from 1->0, as any pending interrupt of a level
+	 * higher than he previous value of IMASK will leak through and be
+	 * taken unexpectedly.
+	 *
+	 * To avoid this we raise the IMASK and then issue another PUTCON to
+	 * enable interrupts.
+         */
+	getcon	SR, r6
+	movi	SR_IMASK | SR_FD, r7
+	or	r6, r7, r6
+	putcon	r6, SR
+	movi	SR_UNBLOCK_EXC, r7
+	and	r6, r7, r6
+	putcon	r6, SR
+
+
+	/* Now call the appropriate 3rd level handler */
+	or	r3, ZERO, LINK
+	movi	trap_jtable, r3
+	shlri	r2, 3, r2
+	ldx.l	r2, r3, r3
+	shlri	r2, 2, r2
+	ptabs	r3, tr0
+	or	SP, ZERO, r3
+	blink	tr0, ZERO
+
+/*
+ * Second level handler for VBR-based exceptions. Post-handlers.
+ *
+ * Post-handlers for interrupts (ret_from_irq), exceptions
+ * (ret_from_exception) and common reentrance doors (restore_all
+ * to get back to the original context, ret_from_syscall loop to
+ * check kernel exiting).
+ *
+ * ret_with_reschedule and work_notifysig are an inner lables of
+ * the ret_from_syscall loop.
+ *
+ * In common to all stack-frame sensitive handlers.
+ *
+ * Inputs:
+ * (SP)   struct pt_regs *, original register's frame pointer (basic)
+ *
+ */
+	.global ret_from_irq
+ret_from_irq:
+#ifdef CONFIG_POOR_MANS_STRACE
+	pta	evt_debug_ret_from_irq, tr0
+	ori	SP, 0, r2
+	blink	tr0, LINK
+#endif
+	ld.q	SP, FRAME_S(FSSR), r6
+	shlri	r6, 30, r6
+	andi	r6, 1, r6
+	pta	resume_kernel, tr0
+	bne	r6, ZERO, tr0		/* no further checks */
+	STI()
+	pta	ret_with_reschedule, tr0
+	blink	tr0, ZERO		/* Do not check softirqs */
+
+	.global ret_from_exception
+ret_from_exception:
+	preempt_stop()
+
+#ifdef CONFIG_POOR_MANS_STRACE
+	pta	evt_debug_ret_from_exc, tr0
+	ori	SP, 0, r2
+	blink	tr0, LINK
+#endif
+
+	ld.q	SP, FRAME_S(FSSR), r6
+	shlri	r6, 30, r6
+	andi	r6, 1, r6
+	pta	resume_kernel, tr0
+	bne	r6, ZERO, tr0		/* no further checks */
+
+	/* Check softirqs */
+
+#ifdef CONFIG_PREEMPT
+	pta   ret_from_syscall, tr0
+	blink   tr0, ZERO
+
+resume_kernel:
+	pta	restore_all, tr0
+
+	getcon	KCR0, r6
+	ld.l	r6, TI_PRE_COUNT, r7
+	beq/u	r7, ZERO, tr0
+
+need_resched:
+	ld.l	r6, TI_FLAGS, r7
+	movi	(1 << TIF_NEED_RESCHED), r8
+	and	r8, r7, r8
+	bne	r8, ZERO, tr0
+
+	getcon	SR, r7
+	andi	r7, 0xf0, r7
+	bne	r7, ZERO, tr0
+
+	movi	((PREEMPT_ACTIVE >> 16) & 65535), r8
+	shori	(PREEMPT_ACTIVE & 65535), r8
+	st.l	r6, TI_PRE_COUNT, r8
+
+	STI()
+	movi	schedule, r7
+	ori	r7, 1, r7
+	ptabs	r7, tr1
+	blink	tr1, LINK
+
+	st.l	r6, TI_PRE_COUNT, ZERO
+	CLI()
+
+	pta	need_resched, tr1
+	blink	tr1, ZERO
+#endif
+
+	.global ret_from_syscall
+ret_from_syscall:
+
+ret_with_reschedule:
+	getcon	KCR0, r6		! r6 contains current_thread_info
+	ld.l	r6, TI_FLAGS, r7	! r7 contains current_thread_info->flags
+
+	! FIXME:!!!
+	! no handling of TIF_SYSCALL_TRACE yet!!
+
+	movi	_TIF_NEED_RESCHED, r8
+	and	r8, r7, r8
+	pta	work_resched, tr0
+	bne	r8, ZERO, tr0
+
+	pta	restore_all, tr1
+
+	movi	(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), r8
+	and	r8, r7, r8
+	pta	work_notifysig, tr0
+	bne	r8, ZERO, tr0
+
+	blink	tr1, ZERO
+
+work_resched:
+	pta	ret_from_syscall, tr0
+	gettr	tr0, LINK
+	movi	schedule, r6
+	ptabs	r6, tr0
+	blink	tr0, ZERO		/* Call schedule(), return on top */
+
+work_notifysig:
+	gettr	tr1, LINK
+
+	movi	do_signal, r6
+	ptabs	r6, tr0
+	or	SP, ZERO, r2
+	or	ZERO, ZERO, r3
+	blink	tr0, LINK	    /* Call do_signal(regs, 0), return here */
+
+restore_all:
+	/* Do prefetches */
+
+	ld.q	SP, FRAME_T(0), r6
+	ld.q	SP, FRAME_T(1), r7
+	ld.q	SP, FRAME_T(2), r8
+	ld.q	SP, FRAME_T(3), r9
+	ptabs	r6, tr0
+	ptabs	r7, tr1
+	ptabs	r8, tr2
+	ptabs	r9, tr3
+	ld.q	SP, FRAME_T(4), r6
+	ld.q	SP, FRAME_T(5), r7
+	ld.q	SP, FRAME_T(6), r8
+	ld.q	SP, FRAME_T(7), r9
+	ptabs	r6, tr4
+	ptabs	r7, tr5
+	ptabs	r8, tr6
+	ptabs	r9, tr7
+
+	ld.q	SP, FRAME_R(0), r0
+	ld.q	SP, FRAME_R(1), r1
+	ld.q	SP, FRAME_R(2), r2
+	ld.q	SP, FRAME_R(3), r3
+	ld.q	SP, FRAME_R(4), r4
+	ld.q	SP, FRAME_R(5), r5
+	ld.q	SP, FRAME_R(6), r6
+	ld.q	SP, FRAME_R(7), r7
+	ld.q	SP, FRAME_R(8), r8
+	ld.q	SP, FRAME_R(9), r9
+	ld.q	SP, FRAME_R(10), r10
+	ld.q	SP, FRAME_R(11), r11
+	ld.q	SP, FRAME_R(12), r12
+	ld.q	SP, FRAME_R(13), r13
+	ld.q	SP, FRAME_R(14), r14
+
+	ld.q	SP, FRAME_R(16), r16
+	ld.q	SP, FRAME_R(17), r17
+	ld.q	SP, FRAME_R(18), r18
+	ld.q	SP, FRAME_R(19), r19
+	ld.q	SP, FRAME_R(20), r20
+	ld.q	SP, FRAME_R(21), r21
+	ld.q	SP, FRAME_R(22), r22
+	ld.q	SP, FRAME_R(23), r23
+	ld.q	SP, FRAME_R(24), r24
+	ld.q	SP, FRAME_R(25), r25
+	ld.q	SP, FRAME_R(26), r26
+	ld.q	SP, FRAME_R(27), r27
+	ld.q	SP, FRAME_R(28), r28
+	ld.q	SP, FRAME_R(29), r29
+	ld.q	SP, FRAME_R(30), r30
+	ld.q	SP, FRAME_R(31), r31
+	ld.q	SP, FRAME_R(32), r32
+	ld.q	SP, FRAME_R(33), r33
+	ld.q	SP, FRAME_R(34), r34
+	ld.q	SP, FRAME_R(35), r35
+	ld.q	SP, FRAME_R(36), r36
+	ld.q	SP, FRAME_R(37), r37
+	ld.q	SP, FRAME_R(38), r38
+	ld.q	SP, FRAME_R(39), r39
+	ld.q	SP, FRAME_R(40), r40
+	ld.q	SP, FRAME_R(41), r41
+	ld.q	SP, FRAME_R(42), r42
+	ld.q	SP, FRAME_R(43), r43
+	ld.q	SP, FRAME_R(44), r44
+	ld.q	SP, FRAME_R(45), r45
+	ld.q	SP, FRAME_R(46), r46
+	ld.q	SP, FRAME_R(47), r47
+	ld.q	SP, FRAME_R(48), r48
+	ld.q	SP, FRAME_R(49), r49
+	ld.q	SP, FRAME_R(50), r50
+	ld.q	SP, FRAME_R(51), r51
+	ld.q	SP, FRAME_R(52), r52
+	ld.q	SP, FRAME_R(53), r53
+	ld.q	SP, FRAME_R(54), r54
+	ld.q	SP, FRAME_R(55), r55
+	ld.q	SP, FRAME_R(56), r56
+	ld.q	SP, FRAME_R(57), r57
+	ld.q	SP, FRAME_R(58), r58
+
+	getcon	SR, r59
+	movi	SR_BLOCK_EXC, r60
+	or	r59, r60, r59
+	putcon	r59, SR			/* SR.BL = 1, keep nesting out */
+	ld.q	SP, FRAME_S(FSSR), r61
+	ld.q	SP, FRAME_S(FSPC), r62
+	movi	SR_ASID_MASK, r60
+	and	r59, r60, r59
+	andc	r61, r60, r61		/* Clear out older ASID */
+	or	r59, r61, r61		/* Retain current ASID */
+	putcon	r61, SSR
+	putcon	r62, SPC
+
+	/* Ignore FSYSCALL_ID */
+
+	ld.q	SP, FRAME_R(59), r59
+	ld.q	SP, FRAME_R(60), r60
+	ld.q	SP, FRAME_R(61), r61
+	ld.q	SP, FRAME_R(62), r62
+
+	/* Last touch */
+	ld.q	SP, FRAME_R(15), SP
+	rte
+	nop
+
+/*
+ * Third level handlers for VBR-based exceptions. Adapting args to
+ * and/or deflecting to fourth level handlers.
+ *
+ * Fourth level handlers interface.
+ * Most are C-coded handlers directly pointed by the trap_jtable.
+ * (Third = Fourth level)
+ * Inputs:
+ * (r2)   fault/interrupt code, entry number (e.g. NMI = 14,
+ *	  IRL0-3 (0000) = 16, RTLBMISS = 2, SYSCALL = 11, etc ...)
+ * (r3)   struct pt_regs *, original register's frame pointer
+ * (r4)   Event (0 = interrupt, 1 = TLB miss fault, 2 = Not TLB miss fault)
+ * (r5)   TRA control register (for syscall/debug benefit only)
+ * (LINK) return address
+ * (SP)   = r3
+ *
+ * Kernel TLB fault handlers will get a slightly different interface.
+ * (r2)   struct pt_regs *, original register's frame pointer
+ * (r3)   writeaccess, whether it's a store fault as opposed to load fault
+ * (r4)   execaccess, whether it's a ITLB fault as opposed to DTLB fault
+ * (r5)   Effective Address of fault
+ * (LINK) return address
+ * (SP)   = r2
+ *
+ * fpu_error_or_IRQ? is a helper to deflect to the right cause.
+ *
+ */
+tlb_miss_load:
+	or	SP, ZERO, r2
+	or	ZERO, ZERO, r3		/* Read */
+	or	ZERO, ZERO, r4		/* Data */
+	getcon	TEA, r5
+	pta	call_do_page_fault, tr0
+	beq	ZERO, ZERO, tr0
+
+tlb_miss_store:
+	or	SP, ZERO, r2
+	movi	1, r3			/* Write */
+	or	ZERO, ZERO, r4		/* Data */
+	getcon	TEA, r5
+	pta	call_do_page_fault, tr0
+	beq	ZERO, ZERO, tr0
+
+itlb_miss_or_IRQ:
+	pta	its_IRQ, tr0
+	beqi/u	r4, EVENT_INTERRUPT, tr0
+	or	SP, ZERO, r2
+	or	ZERO, ZERO, r3		/* Read */
+	movi	1, r4			/* Text */
+	getcon	TEA, r5
+	/* Fall through */
+
+call_do_page_fault:
+	movi	do_page_fault, r6
+        ptabs	r6, tr0
+        blink	tr0, ZERO
+
+fpu_error_or_IRQA:
+	pta	its_IRQ, tr0
+	beqi/l	r4, EVENT_INTERRUPT, tr0
+#ifdef CONFIG_SH_FPU
+	movi	do_fpu_state_restore, r6
+#else
+	movi	do_exception_error, r6
+#endif
+	ptabs	r6, tr0
+	blink	tr0, ZERO
+
+fpu_error_or_IRQB:
+	pta	its_IRQ, tr0
+	beqi/l	r4, EVENT_INTERRUPT, tr0
+#ifdef CONFIG_SH_FPU
+	movi	do_fpu_state_restore, r6
+#else
+	movi	do_exception_error, r6
+#endif
+	ptabs	r6, tr0
+	blink	tr0, ZERO
+
+its_IRQ:
+	movi	do_IRQ, r6
+	ptabs	r6, tr0
+	blink	tr0, ZERO
+
+/*
+ * system_call/unknown_trap third level handler:
+ *
+ * Inputs:
+ * (r2)   fault/interrupt code, entry number (TRAP = 11)
+ * (r3)   struct pt_regs *, original register's frame pointer
+ * (r4)   Not used. Event (0=interrupt, 1=TLB miss fault, 2=Not TLB miss fault)
+ * (r5)   TRA Control Reg (0x00xyzzzz: x=1 SYSCALL, y = #args, z=nr)
+ * (SP)   = r3
+ * (LINK) return address: ret_from_exception
+ * (*r3)  Syscall parms: SC#, arg0, arg1, ..., arg5 in order (Saved r2/r7)
+ *
+ * Outputs:
+ * (*r3)  Syscall reply (Saved r2)
+ * (LINK) In case of syscall only it can be scrapped.
+ *        Common second level post handler will be ret_from_syscall.
+ *        Common (non-trace) exit point to that is syscall_ret (saving
+ *        result to r2). Common bad exit point is syscall_bad (returning
+ *        ENOSYS then saved to r2).
+ *
+ */
+
+unknown_trap:
+	/* Unknown Trap or User Trace */
+	movi	do_unknown_trapa, r6
+	ptabs	r6, tr0
+        ld.q    r3, FRAME_R(9), r2	/* r2 = #arg << 16 | syscall # */
+        andi    r2, 0x1ff, r2		/* r2 = syscall # */
+	blink	tr0, LINK
+
+	pta	syscall_ret, tr0
+	blink	tr0, ZERO
+
+        /* New syscall implementation*/
+system_call:
+	pta	unknown_trap, tr0
+        or      r5, ZERO, r4            /* TRA (=r5) -> r4 */
+        shlri   r4, 20, r4
+	bnei	r4, 1, tr0		/* unknown_trap if not 0x1yzzzz */
+
+        /* It's a system call */
+	st.q    r3, FRAME_S(FSYSCALL_ID), r5 	/* ID (0x1yzzzz) -> stack */
+	andi    r5, 0x1ff, r5			/* syscall # -> r5	  */
+
+	STI()
+
+	pta	syscall_allowed, tr0
+	movi	NR_syscalls - 1, r4	/* Last valid */
+	bgeu/l	r4, r5, tr0
+
+syscall_bad:
+	/* Return ENOSYS ! */
+	movi	-(ENOSYS), r2		/* Fall-through */
+
+	.global syscall_ret
+syscall_ret:
+	st.q	SP, FRAME_R(9), r2	/* Expecting SP back to BASIC frame */
+
+#ifdef CONFIG_POOR_MANS_STRACE
+	/* nothing useful in registers at this point */
+
+	movi	evt_debug2, r5
+	ori	r5, 1, r5
+	ptabs	r5, tr0
+	ld.q	SP, FRAME_R(9), r2
+	or	SP, ZERO, r3
+	blink	tr0, LINK
+#endif
+
+	ld.q	SP, FRAME_S(FSPC), r2
+	addi	r2, 4, r2		/* Move PC, being pre-execution event */
+	st.q	SP, FRAME_S(FSPC), r2
+	pta	ret_from_syscall, tr0
+	blink	tr0, ZERO
+
+
+/*  A different return path for ret_from_fork, because we now need
+ *  to call schedule_tail with the later kernels. Because prev is
+ *  loaded into r2 by switch_to() means we can just call it straight  away
+ */
+
+.global	ret_from_fork
+ret_from_fork:
+
+	movi	schedule_tail,r5
+	ori	r5, 1, r5
+	ptabs	r5, tr0
+	blink	tr0, LINK
+
+#ifdef CONFIG_POOR_MANS_STRACE
+	/* nothing useful in registers at this point */
+
+	movi	evt_debug2, r5
+	ori	r5, 1, r5
+	ptabs	r5, tr0
+	ld.q	SP, FRAME_R(9), r2
+	or	SP, ZERO, r3
+	blink	tr0, LINK
+#endif
+
+	ld.q	SP, FRAME_S(FSPC), r2
+	addi	r2, 4, r2		/* Move PC, being pre-execution event */
+	st.q	SP, FRAME_S(FSPC), r2
+	pta	ret_from_syscall, tr0
+	blink	tr0, ZERO
+
+
+
+syscall_allowed:
+	/* Use LINK to deflect the exit point, default is syscall_ret */
+	pta	syscall_ret, tr0
+	gettr	tr0, LINK
+	pta	syscall_notrace, tr0
+
+	getcon	KCR0, r2
+	ld.l	r2, TI_FLAGS, r4
+	movi	(1 << TIF_SYSCALL_TRACE), r6
+	and	r6, r4, r6
+	beq/l	r6, ZERO, tr0
+
+	/* Trace it by calling syscall_trace before and after */
+	movi	syscall_trace, r4
+	ptabs	r4, tr0
+	blink	tr0, LINK
+	/* Reload syscall number as r5 is trashed by syscall_trace */
+	ld.q	SP, FRAME_S(FSYSCALL_ID), r5
+	andi	r5, 0x1ff, r5
+
+	pta	syscall_ret_trace, tr0
+	gettr	tr0, LINK
+
+syscall_notrace:
+	/* Now point to the appropriate 4th level syscall handler */
+	movi	sys_call_table, r4
+	shlli	r5, 2, r5
+	ldx.l	r4, r5, r5
+	ptabs	r5, tr0
+
+	/* Prepare original args */
+	ld.q	SP, FRAME_R(2), r2
+	ld.q	SP, FRAME_R(3), r3
+	ld.q	SP, FRAME_R(4), r4
+	ld.q	SP, FRAME_R(5), r5
+	ld.q	SP, FRAME_R(6), r6
+	ld.q	SP, FRAME_R(7), r7
+
+	/* And now the trick for those syscalls requiring regs * ! */
+	or	SP, ZERO, r8
+
+	/* Call it */
+	blink	tr0, ZERO	/* LINK is already properly set */
+
+syscall_ret_trace:
+	/* We get back here only if under trace */
+	st.q	SP, FRAME_R(9), r2	/* Save return value */
+
+	movi	syscall_trace, LINK
+	ptabs	LINK, tr0
+	blink	tr0, LINK
+
+	/* This needs to be done after any syscall tracing */
+	ld.q	SP, FRAME_S(FSPC), r2
+	addi	r2, 4, r2	/* Move PC, being pre-execution event */
+	st.q	SP, FRAME_S(FSPC), r2
+
+	pta	ret_from_syscall, tr0
+	blink	tr0, ZERO		/* Resume normal return sequence */
+
+/*
+ * --- Switch to running under a particular ASID and return the previous ASID value
+ * --- The caller is assumed to have done a cli before calling this.
+ *
+ * Input r2 : new ASID
+ * Output r2 : old ASID
+ */
+
+	.global switch_and_save_asid
+switch_and_save_asid:
+	getcon	sr, r0
+	movi	255, r4
+	shlli 	r4, 16, r4	/* r4 = mask to select ASID */
+	and	r0, r4, r3	/* r3 = shifted old ASID */
+	andi	r2, 255, r2	/* mask down new ASID */
+	shlli	r2, 16, r2	/* align new ASID against SR.ASID */
+	andc	r0, r4, r0	/* efface old ASID from SR */
+	or	r0, r2, r0	/* insert the new ASID */
+	putcon	r0, ssr
+	movi	1f, r0
+	putcon	r0, spc
+	rte
+	nop
+1:
+	ptabs	LINK, tr0
+	shlri	r3, 16, r2	/* r2 = old ASID */
+	blink tr0, r63
+
+	.global	route_to_panic_handler
+route_to_panic_handler:
+	/* Switch to real mode, goto panic_handler, don't return.  Useful for
+	   last-chance debugging, e.g. if no output wants to go to the console.
+	   */
+
+	movi	panic_handler - CONFIG_CACHED_MEMORY_OFFSET, r1
+	ptabs	r1, tr0
+	pta	1f, tr1
+	gettr	tr1, r0
+	putcon	r0, spc
+	getcon	sr, r0
+	movi	1, r1
+	shlli	r1, 31, r1
+	andc	r0, r1, r0
+	putcon	r0, ssr
+	rte
+	nop
+1:	/* Now in real mode */
+	blink tr0, r63
+	nop
+
+	.global peek_real_address_q
+peek_real_address_q:
+	/* Two args:
+	   r2 : real mode address to peek
+	   r2(out) : result quadword
+
+	   This is provided as a cheapskate way of manipulating device
+	   registers for debugging (to avoid the need to onchip_remap the debug
+	   module, and to avoid the need to onchip_remap the watchpoint
+	   controller in a way that identity maps sufficient bits to avoid the
+	   SH5-101 cut2 silicon defect).
+
+	   This code is not performance critical
+	*/
+
+	add.l	r2, r63, r2	/* sign extend address */
+	getcon	sr, r0		/* r0 = saved original SR */
+	movi	1, r1
+	shlli	r1, 28, r1
+	or	r0, r1, r1	/* r0 with block bit set */
+	putcon	r1, sr		/* now in critical section */
+	movi	1, r36
+	shlli	r36, 31, r36
+	andc	r1, r36, r1	/* turn sr.mmu off in real mode section */
+
+	putcon	r1, ssr
+	movi	.peek0 - CONFIG_CACHED_MEMORY_OFFSET, r36 /* real mode target address */
+	movi	1f, r37		/* virtual mode return addr */
+	putcon	r36, spc
+
+	synco
+	rte
+	nop
+
+.peek0:	/* come here in real mode, don't touch caches!!
+           still in critical section (sr.bl==1) */
+	putcon	r0, ssr
+	putcon	r37, spc
+	/* Here's the actual peek.  If the address is bad, all bets are now off
+	 * what will happen (handlers invoked in real-mode = bad news) */
+	ld.q	r2, 0, r2
+	synco
+	rte	/* Back to virtual mode */
+	nop
+
+1:
+	ptabs	LINK, tr0
+	blink	tr0, r63
+
+	.global poke_real_address_q
+poke_real_address_q:
+	/* Two args:
+	   r2 : real mode address to poke
+	   r3 : quadword value to write.
+
+	   This is provided as a cheapskate way of manipulating device
+	   registers for debugging (to avoid the need to onchip_remap the debug
+	   module, and to avoid the need to onchip_remap the watchpoint
+	   controller in a way that identity maps sufficient bits to avoid the
+	   SH5-101 cut2 silicon defect).
+
+	   This code is not performance critical
+	*/
+
+	add.l	r2, r63, r2	/* sign extend address */
+	getcon	sr, r0		/* r0 = saved original SR */
+	movi	1, r1
+	shlli	r1, 28, r1
+	or	r0, r1, r1	/* r0 with block bit set */
+	putcon	r1, sr		/* now in critical section */
+	movi	1, r36
+	shlli	r36, 31, r36
+	andc	r1, r36, r1	/* turn sr.mmu off in real mode section */
+
+	putcon	r1, ssr
+	movi	.poke0-CONFIG_CACHED_MEMORY_OFFSET, r36 /* real mode target address */
+	movi	1f, r37		/* virtual mode return addr */
+	putcon	r36, spc
+
+	synco
+	rte
+	nop
+
+.poke0:	/* come here in real mode, don't touch caches!!
+           still in critical section (sr.bl==1) */
+	putcon	r0, ssr
+	putcon	r37, spc
+	/* Here's the actual poke.  If the address is bad, all bets are now off
+	 * what will happen (handlers invoked in real-mode = bad news) */
+	st.q	r2, 0, r3
+	synco
+	rte	/* Back to virtual mode */
+	nop
+
+1:
+	ptabs	LINK, tr0
+	blink	tr0, r63
+
+/*
+ * --- User Access Handling Section
+ */
+
+/*
+ * User Access support. It all moved to non inlined Assembler
+ * functions in here.
+ *
+ * __kernel_size_t __copy_user(void *__to, const void *__from,
+ *			       __kernel_size_t __n)
+ *
+ * Inputs:
+ * (r2)  target address
+ * (r3)  source address
+ * (r4)  size in bytes
+ *
+ * Ouputs:
+ * (*r2) target data
+ * (r2)  non-copied bytes
+ *
+ * If a fault occurs on the user pointer, bail out early and return the
+ * number of bytes not copied in r2.
+ * Strategy : for large blocks, call a real memcpy function which can
+ * move >1 byte at a time using unaligned ld/st instructions, and can
+ * manipulate the cache using prefetch + alloco to improve the speed
+ * further.  If a fault occurs in that function, just revert to the
+ * byte-by-byte approach used for small blocks; this is rare so the
+ * performance hit for that case does not matter.
+ *
+ * For small blocks it's not worth the overhead of setting up and calling
+ * the memcpy routine; do the copy a byte at a time.
+ *
+ */
+	.global	__copy_user
+__copy_user:
+	pta	__copy_user_byte_by_byte, tr1
+	movi	16, r0 ! this value is a best guess, should tune it by benchmarking
+	bge/u	r0, r4, tr1
+	pta copy_user_memcpy, tr0
+	addi	SP, -32, SP
+	/* Save arguments in case we have to fix-up unhandled page fault */
+	st.q	SP, 0, r2
+	st.q	SP, 8, r3
+	st.q	SP, 16, r4
+	st.q	SP, 24, r35 ! r35 is callee-save
+	/* Save LINK in a register to reduce RTS time later (otherwise
+	   ld SP,*,LINK;ptabs LINK;trn;blink trn,r63 becomes a critical path) */
+	ori	LINK, 0, r35
+	blink	tr0, LINK
+
+	/* Copy completed normally if we get back here */
+	ptabs	r35, tr0
+	ld.q	SP, 24, r35
+	/* don't restore r2-r4, pointless */
+	/* set result=r2 to zero as the copy must have succeeded. */
+	or	r63, r63, r2
+	addi	SP, 32, SP
+	blink	tr0, r63 ! RTS
+
+	.global __copy_user_fixup
+__copy_user_fixup:
+	/* Restore stack frame */
+	ori	r35, 0, LINK
+	ld.q	SP, 24, r35
+	ld.q	SP, 16, r4
+	ld.q	SP,  8, r3
+	ld.q	SP,  0, r2
+	addi	SP, 32, SP
+	/* Fall through to original code, in the 'same' state we entered with */
+
+/* The slow byte-by-byte method is used if the fast copy traps due to a bad
+   user address.  In that rare case, the speed drop can be tolerated. */
+__copy_user_byte_by_byte:
+	pta	___copy_user_exit, tr1
+	pta	___copy_user1, tr0
+	beq/u	r4, r63, tr1	/* early exit for zero length copy */
+	sub	r2, r3, r0
+	addi	r0, -1, r0
+
+___copy_user1:
+	ld.b	r3, 0, r5		/* Fault address 1 */
+
+	/* Could rewrite this to use just 1 add, but the second comes 'free'
+	   due to load latency */
+	addi	r3, 1, r3
+	addi	r4, -1, r4		/* No real fixup required */
+___copy_user2:
+	stx.b	r3, r0, r5		/* Fault address 2 */
+	bne     r4, ZERO, tr0
+
+___copy_user_exit:
+	or	r4, ZERO, r2
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+/*
+ * __kernel_size_t __clear_user(void *addr, __kernel_size_t size)
+ *
+ * Inputs:
+ * (r2)  target address
+ * (r3)  size in bytes
+ *
+ * Ouputs:
+ * (*r2) zero-ed target data
+ * (r2)  non-zero-ed bytes
+ */
+	.global	__clear_user
+__clear_user:
+	pta	___clear_user_exit, tr1
+	pta	___clear_user1, tr0
+	beq/u	r3, r63, tr1
+
+___clear_user1:
+	st.b	r2, 0, ZERO		/* Fault address */
+	addi	r2, 1, r2
+	addi	r3, -1, r3		/* No real fixup required */
+	bne     r3, ZERO, tr0
+
+___clear_user_exit:
+	or	r3, ZERO, r2
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+/*
+ * int __strncpy_from_user(unsigned long __dest, unsigned long __src,
+ *			   int __count)
+ *
+ * Inputs:
+ * (r2)  target address
+ * (r3)  source address
+ * (r4)  maximum size in bytes
+ *
+ * Ouputs:
+ * (*r2) copied data
+ * (r2)  -EFAULT (in case of faulting)
+ *       copied data (otherwise)
+ */
+	.global	__strncpy_from_user
+__strncpy_from_user:
+	pta	___strncpy_from_user1, tr0
+	pta	___strncpy_from_user_done, tr1
+	or	r4, ZERO, r5		/* r5 = original count */
+	beq/u	r4, r63, tr1		/* early exit if r4==0 */
+	movi	-(EFAULT), r6		/* r6 = reply, no real fixup */
+	or	ZERO, ZERO, r7		/* r7 = data, clear top byte of data */
+
+___strncpy_from_user1:
+	ld.b	r3, 0, r7		/* Fault address: only in reading */
+	st.b	r2, 0, r7
+	addi	r2, 1, r2
+	addi	r3, 1, r3
+	beq/u	ZERO, r7, tr1
+	addi	r4, -1, r4		/* return real number of copied bytes */
+	bne/l	ZERO, r4, tr0
+
+___strncpy_from_user_done:
+	sub	r5, r4, r6		/* If done, return copied */
+
+___strncpy_from_user_exit:
+	or	r6, ZERO, r2
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+/*
+ * extern long __strnlen_user(const char *__s, long __n)
+ *
+ * Inputs:
+ * (r2)  source address
+ * (r3)  source size in bytes
+ *
+ * Ouputs:
+ * (r2)  -EFAULT (in case of faulting)
+ *       string length (otherwise)
+ */
+	.global	__strnlen_user
+__strnlen_user:
+	pta	___strnlen_user_set_reply, tr0
+	pta	___strnlen_user1, tr1
+	or	ZERO, ZERO, r5		/* r5 = counter */
+	movi	-(EFAULT), r6		/* r6 = reply, no real fixup */
+	or	ZERO, ZERO, r7		/* r7 = data, clear top byte of data */
+	beq	r3, ZERO, tr0
+
+___strnlen_user1:
+	ldx.b	r2, r5, r7		/* Fault address: only in reading */
+	addi	r3, -1, r3		/* No real fixup */
+	addi	r5, 1, r5
+	beq	r3, ZERO, tr0
+	bne	r7, ZERO, tr1
+! The line below used to be active.  This meant led to a junk byte lying between each pair
+! of entries in the argv & envp structures in memory.  Whilst the program saw the right data
+! via the argv and envp arguments to main, it meant the 'flat' representation visible through
+! /proc/$pid/cmdline was corrupt, causing trouble with ps, for example.
+!	addi	r5, 1, r5		/* Include '\0' */
+
+___strnlen_user_set_reply:
+	or	r5, ZERO, r6		/* If done, return counter */
+
+___strnlen_user_exit:
+	or	r6, ZERO, r2
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+/*
+ * extern long __get_user_asm_?(void *val, long addr)
+ *
+ * Inputs:
+ * (r2)  dest address
+ * (r3)  source address (in User Space)
+ *
+ * Ouputs:
+ * (r2)  -EFAULT (faulting)
+ *       0 	 (not faulting)
+ */
+	.global	__get_user_asm_b
+__get_user_asm_b:
+	or	r2, ZERO, r4
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___get_user_asm_b1:
+	ld.b	r3, 0, r5		/* r5 = data */
+	st.b	r4, 0, r5
+	or	ZERO, ZERO, r2
+
+___get_user_asm_b_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__get_user_asm_w
+__get_user_asm_w:
+	or	r2, ZERO, r4
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___get_user_asm_w1:
+	ld.w	r3, 0, r5		/* r5 = data */
+	st.w	r4, 0, r5
+	or	ZERO, ZERO, r2
+
+___get_user_asm_w_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__get_user_asm_l
+__get_user_asm_l:
+	or	r2, ZERO, r4
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___get_user_asm_l1:
+	ld.l	r3, 0, r5		/* r5 = data */
+	st.l	r4, 0, r5
+	or	ZERO, ZERO, r2
+
+___get_user_asm_l_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__get_user_asm_q
+__get_user_asm_q:
+	or	r2, ZERO, r4
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___get_user_asm_q1:
+	ld.q	r3, 0, r5		/* r5 = data */
+	st.q	r4, 0, r5
+	or	ZERO, ZERO, r2
+
+___get_user_asm_q_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+/*
+ * extern long __put_user_asm_?(void *pval, long addr)
+ *
+ * Inputs:
+ * (r2)  kernel pointer to value
+ * (r3)  dest address (in User Space)
+ *
+ * Ouputs:
+ * (r2)  -EFAULT (faulting)
+ *       0 	 (not faulting)
+ */
+	.global	__put_user_asm_b
+__put_user_asm_b:
+	ld.b	r2, 0, r4		/* r4 = data */
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___put_user_asm_b1:
+	st.b	r3, 0, r4
+	or	ZERO, ZERO, r2
+
+___put_user_asm_b_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__put_user_asm_w
+__put_user_asm_w:
+	ld.w	r2, 0, r4		/* r4 = data */
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___put_user_asm_w1:
+	st.w	r3, 0, r4
+	or	ZERO, ZERO, r2
+
+___put_user_asm_w_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__put_user_asm_l
+__put_user_asm_l:
+	ld.l	r2, 0, r4		/* r4 = data */
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___put_user_asm_l1:
+	st.l	r3, 0, r4
+	or	ZERO, ZERO, r2
+
+___put_user_asm_l_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+
+	.global	__put_user_asm_q
+__put_user_asm_q:
+	ld.q	r2, 0, r4		/* r4 = data */
+	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
+
+___put_user_asm_q1:
+	st.q	r3, 0, r4
+	or	ZERO, ZERO, r2
+
+___put_user_asm_q_exit:
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
+panic_stash_regs:
+	/* The idea is : when we get an unhandled panic, we dump the registers
+	   to a known memory location, the just sit in a tight loop.
+	   This allows the human to look at the memory region through the GDB
+	   session (assuming the debug module's SHwy initiator isn't locked up
+	   or anything), to hopefully analyze the cause of the panic. */
+
+	/* On entry, former r15 (SP) is in DCR
+	   former r0  is at resvec_saved_area + 0
+	   former r1  is at resvec_saved_area + 8
+	   former tr0 is at resvec_saved_area + 32
+	   DCR is the only register whose value is lost altogether.
+	*/
+
+	movi	0xffffffff80000000, r0 ! phy of dump area
+	ld.q	SP, 0x000, r1	! former r0
+	st.q	r0,  0x000, r1
+	ld.q	SP, 0x008, r1	! former r1
+	st.q	r0,  0x008, r1
+	st.q	r0,  0x010, r2
+	st.q	r0,  0x018, r3
+	st.q	r0,  0x020, r4
+	st.q	r0,  0x028, r5
+	st.q	r0,  0x030, r6
+	st.q	r0,  0x038, r7
+	st.q	r0,  0x040, r8
+	st.q	r0,  0x048, r9
+	st.q	r0,  0x050, r10
+	st.q	r0,  0x058, r11
+	st.q	r0,  0x060, r12
+	st.q	r0,  0x068, r13
+	st.q	r0,  0x070, r14
+	getcon	dcr, r14
+	st.q	r0,  0x078, r14
+	st.q	r0,  0x080, r16
+	st.q	r0,  0x088, r17
+	st.q	r0,  0x090, r18
+	st.q	r0,  0x098, r19
+	st.q	r0,  0x0a0, r20
+	st.q	r0,  0x0a8, r21
+	st.q	r0,  0x0b0, r22
+	st.q	r0,  0x0b8, r23
+	st.q	r0,  0x0c0, r24
+	st.q	r0,  0x0c8, r25
+	st.q	r0,  0x0d0, r26
+	st.q	r0,  0x0d8, r27
+	st.q	r0,  0x0e0, r28
+	st.q	r0,  0x0e8, r29
+	st.q	r0,  0x0f0, r30
+	st.q	r0,  0x0f8, r31
+	st.q	r0,  0x100, r32
+	st.q	r0,  0x108, r33
+	st.q	r0,  0x110, r34
+	st.q	r0,  0x118, r35
+	st.q	r0,  0x120, r36
+	st.q	r0,  0x128, r37
+	st.q	r0,  0x130, r38
+	st.q	r0,  0x138, r39
+	st.q	r0,  0x140, r40
+	st.q	r0,  0x148, r41
+	st.q	r0,  0x150, r42
+	st.q	r0,  0x158, r43
+	st.q	r0,  0x160, r44
+	st.q	r0,  0x168, r45
+	st.q	r0,  0x170, r46
+	st.q	r0,  0x178, r47
+	st.q	r0,  0x180, r48
+	st.q	r0,  0x188, r49
+	st.q	r0,  0x190, r50
+	st.q	r0,  0x198, r51
+	st.q	r0,  0x1a0, r52
+	st.q	r0,  0x1a8, r53
+	st.q	r0,  0x1b0, r54
+	st.q	r0,  0x1b8, r55
+	st.q	r0,  0x1c0, r56
+	st.q	r0,  0x1c8, r57
+	st.q	r0,  0x1d0, r58
+	st.q	r0,  0x1d8, r59
+	st.q	r0,  0x1e0, r60
+	st.q	r0,  0x1e8, r61
+	st.q	r0,  0x1f0, r62
+	st.q	r0,  0x1f8, r63	! bogus, but for consistency's sake...
+
+	ld.q	SP, 0x020, r1  ! former tr0
+	st.q	r0,  0x200, r1
+	gettr	tr1, r1
+	st.q	r0,  0x208, r1
+	gettr	tr2, r1
+	st.q	r0,  0x210, r1
+	gettr	tr3, r1
+	st.q	r0,  0x218, r1
+	gettr	tr4, r1
+	st.q	r0,  0x220, r1
+	gettr	tr5, r1
+	st.q	r0,  0x228, r1
+	gettr	tr6, r1
+	st.q	r0,  0x230, r1
+	gettr	tr7, r1
+	st.q	r0,  0x238, r1
+
+	getcon	sr,  r1
+	getcon	ssr,  r2
+	getcon	pssr,  r3
+	getcon	spc,  r4
+	getcon	pspc,  r5
+	getcon	intevt,  r6
+	getcon	expevt,  r7
+	getcon	pexpevt,  r8
+	getcon	tra,  r9
+	getcon	tea,  r10
+	getcon	kcr0, r11
+	getcon	kcr1, r12
+	getcon	vbr,  r13
+	getcon	resvec,  r14
+
+	st.q	r0,  0x240, r1
+	st.q	r0,  0x248, r2
+	st.q	r0,  0x250, r3
+	st.q	r0,  0x258, r4
+	st.q	r0,  0x260, r5
+	st.q	r0,  0x268, r6
+	st.q	r0,  0x270, r7
+	st.q	r0,  0x278, r8
+	st.q	r0,  0x280, r9
+	st.q	r0,  0x288, r10
+	st.q	r0,  0x290, r11
+	st.q	r0,  0x298, r12
+	st.q	r0,  0x2a0, r13
+	st.q	r0,  0x2a8, r14
+
+	getcon	SPC,r2
+	getcon	SSR,r3
+	getcon	EXPEVT,r4
+	/* Prepare to jump to C - physical address */
+	movi	panic_handler-CONFIG_CACHED_MEMORY_OFFSET, r1
+	ori	r1, 1, r1
+	ptabs   r1, tr0
+	getcon	DCR, SP
+	blink	tr0, ZERO
+	nop
+	nop
+	nop
+	nop
+
+
+
+
+/*
+ * --- Signal Handling Section
+ */
+
+/*
+ * extern long long _sa_default_rt_restorer
+ * extern long long _sa_default_restorer
+ *
+ *		 or, better,
+ *
+ * extern void _sa_default_rt_restorer(void)
+ * extern void _sa_default_restorer(void)
+ *
+ * Code prototypes to do a sys_rt_sigreturn() or sys_sysreturn()
+ * from user space. Copied into user space by signal management.
+ * Both must be quad aligned and 2 quad long (4 instructions).
+ *
+ */
+	.balign 8
+	.global sa_default_rt_restorer
+sa_default_rt_restorer:
+	movi	0x10, r9
+	shori	__NR_rt_sigreturn, r9
+	trapa	r9
+	nop
+
+	.balign 8
+	.global sa_default_restorer
+sa_default_restorer:
+	movi	0x10, r9
+	shori	__NR_sigreturn, r9
+	trapa	r9
+	nop
+
+/*
+ * --- __ex_table Section
+ */
+
+/*
+ * User Access Exception Table.
+ */
+	.section	__ex_table,  "a"
+
+	.global asm_uaccess_start	/* Just a marker */
+asm_uaccess_start:
+
+	.long	___copy_user1, ___copy_user_exit
+	.long	___copy_user2, ___copy_user_exit
+	.long	___clear_user1, ___clear_user_exit
+	.long	___strncpy_from_user1, ___strncpy_from_user_exit
+	.long	___strnlen_user1, ___strnlen_user_exit
+	.long	___get_user_asm_b1, ___get_user_asm_b_exit
+	.long	___get_user_asm_w1, ___get_user_asm_w_exit
+	.long	___get_user_asm_l1, ___get_user_asm_l_exit
+	.long	___get_user_asm_q1, ___get_user_asm_q_exit
+	.long	___put_user_asm_b1, ___put_user_asm_b_exit
+	.long	___put_user_asm_w1, ___put_user_asm_w_exit
+	.long	___put_user_asm_l1, ___put_user_asm_l_exit
+	.long	___put_user_asm_q1, ___put_user_asm_q_exit
+
+	.global asm_uaccess_end		/* Just a marker */
+asm_uaccess_end:
+
+
+
+
+/*
+ * --- .text.init Section
+ */
+
+	.section	.text.init, "ax"
+
+/*
+ * void trap_init (void)
+ *
+ */
+	.global	trap_init
+trap_init:
+	addi	SP, -24, SP			/* Room to save r28/r29/r30 */
+	st.q	SP, 0, r28
+	st.q	SP, 8, r29
+	st.q	SP, 16, r30
+
+	/* Set VBR and RESVEC */
+	movi	LVBR_block, r19
+	andi	r19, -4, r19			/* reset MMUOFF + reserved */
+	/* For RESVEC exceptions we force the MMU off, which means we need the
+	   physical address. */
+	movi	LRESVEC_block-CONFIG_CACHED_MEMORY_OFFSET, r20
+	andi	r20, -4, r20			/* reset reserved */
+	ori	r20, 1, r20			/* set MMUOFF */
+	putcon	r19, VBR
+	putcon	r20, RESVEC
+
+	/* Sanity check */
+	movi	LVBR_block_end, r21
+	andi	r21, -4, r21
+	movi	BLOCK_SIZE, r29			/* r29 = expected size */
+	or	r19, ZERO, r30
+	add	r19, r29, r19
+
+	/*
+	 * Ugly, but better loop forever now than crash afterwards.
+	 * We should print a message, but if we touch LVBR or
+	 * LRESVEC blocks we should not be surprised if we get stuck
+	 * in trap_init().
+	 */
+	pta	trap_init_loop, tr1
+	gettr	tr1, r28			/* r28 = trap_init_loop */
+	sub	r21, r30, r30			/* r30 = actual size */
+
+	/*
+	 * VBR/RESVEC handlers overlap by being bigger than
+	 * allowed. Very bad. Just loop forever.
+	 * (r28) panic/loop address
+	 * (r29) expected size
+	 * (r30) actual size
+	 */
+trap_init_loop:
+	bne	r19, r21, tr1
+
+	/* Now that exception vectors are set up reset SR.BL */
+	getcon 	SR, r22
+	movi	SR_UNBLOCK_EXC, r23
+	and	r22, r23, r22
+	putcon	r22, SR
+
+	addi	SP, 24, SP
+	ptabs	LINK, tr0
+	blink	tr0, ZERO
+
diff --git a/arch/sh64/kernel/entry.S b/arch/sh64/kernel/entry.S
deleted file mode 100644
index 7013fcb6665c..000000000000
--- a/arch/sh64/kernel/entry.S
+++ /dev/null
@@ -1,2102 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/entry.S
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2004, 2005  Paul Mundt
- * Copyright (C) 2003, 2004 Richard Curnow
- *
- */
-
-#include <linux/errno.h>
-#include <linux/sys.h>
-
-#include <asm/processor.h>
-#include <asm/registers.h>
-#include <asm/unistd.h>
-#include <asm/thread_info.h>
-#include <asm/asm-offsets.h>
-
-/*
- * SR fields.
- */
-#define SR_ASID_MASK	0x00ff0000
-#define SR_FD_MASK	0x00008000
-#define SR_SS		0x08000000
-#define SR_BL		0x10000000
-#define SR_MD		0x40000000
-
-/*
- * Event code.
- */
-#define	EVENT_INTERRUPT		0
-#define	EVENT_FAULT_TLB		1
-#define	EVENT_FAULT_NOT_TLB	2
-#define	EVENT_DEBUG		3
-
-/* EXPEVT values */
-#define	RESET_CAUSE		0x20
-#define DEBUGSS_CAUSE		0x980
-
-/*
- * Frame layout. Quad index.
- */
-#define	FRAME_T(x)	FRAME_TBASE+(x*8)
-#define	FRAME_R(x)	FRAME_RBASE+(x*8)
-#define	FRAME_S(x)	FRAME_SBASE+(x*8)
-#define FSPC		0
-#define FSSR		1
-#define FSYSCALL_ID	2
-
-/* Arrange the save frame to be a multiple of 32 bytes long */
-#define FRAME_SBASE	0
-#define FRAME_RBASE	(FRAME_SBASE+(3*8))	/* SYSCALL_ID - SSR - SPC */
-#define FRAME_TBASE	(FRAME_RBASE+(63*8))	/* r0 - r62 */
-#define FRAME_PBASE	(FRAME_TBASE+(8*8))	/* tr0 -tr7 */
-#define	FRAME_SIZE	(FRAME_PBASE+(2*8))	/* pad0-pad1 */
-
-#define FP_FRAME_SIZE	FP_FRAME_BASE+(33*8)	/* dr0 - dr31 + fpscr */
-#define FP_FRAME_BASE	0
-
-#define	SAVED_R2	0*8
-#define	SAVED_R3	1*8
-#define	SAVED_R4	2*8
-#define	SAVED_R5	3*8
-#define	SAVED_R18	4*8
-#define	SAVED_R6	5*8
-#define	SAVED_TR0	6*8
-
-/* These are the registers saved in the TLB path that aren't saved in the first
-   level of the normal one. */
-#define	TLB_SAVED_R25	7*8
-#define	TLB_SAVED_TR1	8*8
-#define	TLB_SAVED_TR2	9*8
-#define	TLB_SAVED_TR3	10*8
-#define	TLB_SAVED_TR4	11*8
-/* Save R0/R1 : PT-migrating compiler currently dishounours -ffixed-r0 and -ffixed-r1 causing
-   breakage otherwise. */
-#define	TLB_SAVED_R0	12*8
-#define	TLB_SAVED_R1	13*8
-
-#define CLI()				\
-	getcon	SR, r6;			\
-	ori	r6, 0xf0, r6;		\
-	putcon	r6, SR;
-
-#define STI()				\
-	getcon	SR, r6;			\
-	andi	r6, ~0xf0, r6;		\
-	putcon	r6, SR;
-
-#ifdef CONFIG_PREEMPT
-#  define preempt_stop()	CLI()
-#else
-#  define preempt_stop()
-#  define resume_kernel		restore_all
-#endif
-
-	.section	.data, "aw"
-
-#define FAST_TLBMISS_STACK_CACHELINES 4
-#define FAST_TLBMISS_STACK_QUADWORDS (4*FAST_TLBMISS_STACK_CACHELINES)
-
-/* Register back-up area for all exceptions */
-	.balign	32
-	/* Allow for 16 quadwords to be pushed by fast tlbmiss handling
-	 * register saves etc. */
-	.fill FAST_TLBMISS_STACK_QUADWORDS, 8, 0x0
-/* This is 32 byte aligned by construction */
-/* Register back-up area for all exceptions */
-reg_save_area:
-	.quad	0
-	.quad	0
-	.quad	0
-	.quad	0
-
-	.quad	0
-	.quad	0
-	.quad	0
-	.quad	0
-
-	.quad	0
-	.quad	0
-	.quad	0
-	.quad	0
-
-	.quad	0
-	.quad   0
-
-/* Save area for RESVEC exceptions. We cannot use reg_save_area because of
- * reentrancy. Note this area may be accessed via physical address.
- * Align so this fits a whole single cache line, for ease of purging.
- */
-	.balign 32,0,32
-resvec_save_area:
-	.quad	0
-	.quad	0
-	.quad	0
-	.quad	0
-	.quad	0
-	.balign 32,0,32
-
-/* Jump table of 3rd level handlers  */
-trap_jtable:
-	.long	do_exception_error		/* 0x000 */
-	.long	do_exception_error		/* 0x020 */
-	.long	tlb_miss_load				/* 0x040 */
-	.long	tlb_miss_store				/* 0x060 */
-	! ARTIFICIAL pseudo-EXPEVT setting
-	.long	do_debug_interrupt		/* 0x080 */
-	.long	tlb_miss_load				/* 0x0A0 */
-	.long	tlb_miss_store				/* 0x0C0 */
-	.long	do_address_error_load	/* 0x0E0 */
-	.long	do_address_error_store	/* 0x100 */
-#ifdef CONFIG_SH_FPU
-	.long	do_fpu_error		/* 0x120 */
-#else
-	.long	do_exception_error		/* 0x120 */
-#endif
-	.long	do_exception_error		/* 0x140 */
-	.long	system_call				/* 0x160 */
-	.long	do_reserved_inst		/* 0x180 */
-	.long	do_illegal_slot_inst	/* 0x1A0 */
-	.long	do_NMI			/* 0x1C0 */
-	.long	do_exception_error		/* 0x1E0 */
-	.rept 15
-		.long do_IRQ		/* 0x200 - 0x3C0 */
-	.endr
-	.long	do_exception_error		/* 0x3E0 */
-	.rept 32
-		.long do_IRQ		/* 0x400 - 0x7E0 */
-	.endr
-	.long	fpu_error_or_IRQA			/* 0x800 */
-	.long	fpu_error_or_IRQB			/* 0x820 */
-	.long	do_IRQ			/* 0x840 */
-	.long	do_IRQ			/* 0x860 */
-	.rept 6
-		.long do_exception_error	/* 0x880 - 0x920 */
-	.endr
-	.long	do_software_break_point	/* 0x940 */
-	.long	do_exception_error		/* 0x960 */
-	.long	do_single_step		/* 0x980 */
-
-	.rept 3
-		.long do_exception_error	/* 0x9A0 - 0x9E0 */
-	.endr
-	.long	do_IRQ			/* 0xA00 */
-	.long	do_IRQ			/* 0xA20 */
-	.long	itlb_miss_or_IRQ			/* 0xA40 */
-	.long	do_IRQ			/* 0xA60 */
-	.long	do_IRQ			/* 0xA80 */
-	.long	itlb_miss_or_IRQ			/* 0xAA0 */
-	.long	do_exception_error		/* 0xAC0 */
-	.long	do_address_error_exec	/* 0xAE0 */
-	.rept 8
-		.long do_exception_error	/* 0xB00 - 0xBE0 */
-	.endr
-	.rept 18
-		.long do_IRQ		/* 0xC00 - 0xE20 */
-	.endr
-
-	.section	.text64, "ax"
-
-/*
- * --- Exception/Interrupt/Event Handling Section
- */
-
-/*
- * VBR and RESVEC blocks.
- *
- * First level handler for VBR-based exceptions.
- *
- * To avoid waste of space, align to the maximum text block size.
- * This is assumed to be at most 128 bytes or 32 instructions.
- * DO NOT EXCEED 32 instructions on the first level handlers !
- *
- * Also note that RESVEC is contained within the VBR block
- * where the room left (1KB - TEXT_SIZE) allows placing
- * the RESVEC block (at most 512B + TEXT_SIZE).
- *
- * So first (and only) level handler for RESVEC-based exceptions.
- *
- * Where the fault/interrupt is handled (not_a_tlb_miss, tlb_miss
- * and interrupt) we are a lot tight with register space until
- * saving onto the stack frame, which is done in handle_exception().
- *
- */
-
-#define	TEXT_SIZE 	128
-#define	BLOCK_SIZE 	1664 		/* Dynamic check, 13*128 */
-
-	.balign TEXT_SIZE
-LVBR_block:
-	.space	256, 0			/* Power-on class handler, */
-					/* not required here       */
-not_a_tlb_miss:
-	synco	/* TAKum03020 (but probably a good idea anyway.) */
-	/* Save original stack pointer into KCR1 */
-	putcon	SP, KCR1
-
-	/* Save other original registers into reg_save_area */
-        movi  reg_save_area, SP
-	st.q	SP, SAVED_R2, r2
-	st.q	SP, SAVED_R3, r3
-	st.q	SP, SAVED_R4, r4
-	st.q	SP, SAVED_R5, r5
-	st.q	SP, SAVED_R6, r6
-	st.q	SP, SAVED_R18, r18
-	gettr	tr0, r3
-	st.q	SP, SAVED_TR0, r3
-
-	/* Set args for Non-debug, Not a TLB miss class handler */
-	getcon	EXPEVT, r2
-	movi	ret_from_exception, r3
-	ori	r3, 1, r3
-	movi	EVENT_FAULT_NOT_TLB, r4
-	or	SP, ZERO, r5
-	getcon	KCR1, SP
-	pta	handle_exception, tr0
-	blink	tr0, ZERO
-
-	.balign 256
-	! VBR+0x200
-	nop
-	.balign 256
-	! VBR+0x300
-	nop
-	.balign 256
-	/*
-	 * Instead of the natural .balign 1024 place RESVEC here
-	 * respecting the final 1KB alignment.
-	 */
-	.balign TEXT_SIZE
-	/*
-	 * Instead of '.space 1024-TEXT_SIZE' place the RESVEC
-	 * block making sure the final alignment is correct.
-	 */
-tlb_miss:
-	synco	/* TAKum03020 (but probably a good idea anyway.) */
-	putcon	SP, KCR1
-	movi	reg_save_area, SP
-	/* SP is guaranteed 32-byte aligned. */
-	st.q	SP, TLB_SAVED_R0 , r0
-	st.q	SP, TLB_SAVED_R1 , r1
-	st.q	SP, SAVED_R2 , r2
-	st.q	SP, SAVED_R3 , r3
-	st.q	SP, SAVED_R4 , r4
-	st.q	SP, SAVED_R5 , r5
-	st.q	SP, SAVED_R6 , r6
-	st.q	SP, SAVED_R18, r18
-
-	/* Save R25 for safety; as/ld may want to use it to achieve the call to
-	 * the code in mm/tlbmiss.c */
-	st.q	SP, TLB_SAVED_R25, r25
-	gettr	tr0, r2
-	gettr	tr1, r3
-	gettr	tr2, r4
-	gettr	tr3, r5
-	gettr	tr4, r18
-	st.q	SP, SAVED_TR0 , r2
-	st.q	SP, TLB_SAVED_TR1 , r3
-	st.q	SP, TLB_SAVED_TR2 , r4
-	st.q	SP, TLB_SAVED_TR3 , r5
-	st.q	SP, TLB_SAVED_TR4 , r18
-
-	pt	do_fast_page_fault, tr0
-	getcon	SSR, r2
-	getcon	EXPEVT, r3
-	getcon	TEA, r4
-	shlri	r2, 30, r2
-	andi	r2, 1, r2	/* r2 = SSR.MD */
-	blink 	tr0, LINK
-
-	pt	fixup_to_invoke_general_handler, tr1
-
-	/* If the fast path handler fixed the fault, just drop through quickly
-	   to the restore code right away to return to the excepting context.
-	   */
-	beqi/u	r2, 0, tr1
-
-fast_tlb_miss_restore:
-	ld.q	SP, SAVED_TR0, r2
-	ld.q	SP, TLB_SAVED_TR1, r3
-	ld.q	SP, TLB_SAVED_TR2, r4
-
-	ld.q	SP, TLB_SAVED_TR3, r5
-	ld.q	SP, TLB_SAVED_TR4, r18
-
-	ptabs	r2, tr0
-	ptabs	r3, tr1
-	ptabs	r4, tr2
-	ptabs	r5, tr3
-	ptabs	r18, tr4
-
-	ld.q	SP, TLB_SAVED_R0, r0
-	ld.q	SP, TLB_SAVED_R1, r1
-	ld.q	SP, SAVED_R2, r2
-	ld.q	SP, SAVED_R3, r3
-	ld.q	SP, SAVED_R4, r4
-	ld.q	SP, SAVED_R5, r5
-	ld.q	SP, SAVED_R6, r6
-	ld.q	SP, SAVED_R18, r18
-	ld.q	SP, TLB_SAVED_R25, r25
-
-	getcon	KCR1, SP
-	rte
-	nop /* for safety, in case the code is run on sh5-101 cut1.x */
-
-fixup_to_invoke_general_handler:
-
-	/* OK, new method.  Restore stuff that's not expected to get saved into
-	   the 'first-level' reg save area, then just fall through to setting
-	   up the registers and calling the second-level handler. */
-
-	/* 2nd level expects r2,3,4,5,6,18,tr0 to be saved.  So we must restore
-	   r25,tr1-4 and save r6 to get into the right state.  */
-
-	ld.q	SP, TLB_SAVED_TR1, r3
-	ld.q	SP, TLB_SAVED_TR2, r4
-	ld.q	SP, TLB_SAVED_TR3, r5
-	ld.q	SP, TLB_SAVED_TR4, r18
-	ld.q	SP, TLB_SAVED_R25, r25
-
-	ld.q	SP, TLB_SAVED_R0, r0
-	ld.q	SP, TLB_SAVED_R1, r1
-
-	ptabs/u	r3, tr1
-	ptabs/u	r4, tr2
-	ptabs/u	r5, tr3
-	ptabs/u	r18, tr4
-
-	/* Set args for Non-debug, TLB miss class handler */
-	getcon	EXPEVT, r2
-	movi	ret_from_exception, r3
-	ori	r3, 1, r3
-	movi	EVENT_FAULT_TLB, r4
-	or	SP, ZERO, r5
-	getcon	KCR1, SP
-	pta	handle_exception, tr0
-	blink	tr0, ZERO
-
-/* NB TAKE GREAT CARE HERE TO ENSURE THAT THE INTERRUPT CODE
-   DOES END UP AT VBR+0x600 */
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-
-	.balign 256
-	/* VBR + 0x600 */
-
-interrupt:
-	synco	/* TAKum03020 (but probably a good idea anyway.) */
-	/* Save original stack pointer into KCR1 */
-	putcon	SP, KCR1
-
-	/* Save other original registers into reg_save_area */
-        movi  reg_save_area, SP
-	st.q	SP, SAVED_R2, r2
-	st.q	SP, SAVED_R3, r3
-	st.q	SP, SAVED_R4, r4
-	st.q	SP, SAVED_R5, r5
-	st.q	SP, SAVED_R6, r6
-	st.q	SP, SAVED_R18, r18
-	gettr	tr0, r3
-	st.q	SP, SAVED_TR0, r3
-
-	/* Set args for interrupt class handler */
-	getcon	INTEVT, r2
-	movi	ret_from_irq, r3
-	ori	r3, 1, r3
-	movi	EVENT_INTERRUPT, r4
-	or	SP, ZERO, r5
-	getcon	KCR1, SP
-	pta	handle_exception, tr0
-	blink	tr0, ZERO
-	.balign	TEXT_SIZE		/* let's waste the bare minimum */
-
-LVBR_block_end:				/* Marker. Used for total checking */
-
-	.balign 256
-LRESVEC_block:
-	/* Panic handler. Called with MMU off. Possible causes/actions:
-	 * - Reset:		Jump to program start.
-	 * - Single Step:	Turn off Single Step & return.
-	 * - Others:		Call panic handler, passing PC as arg.
-	 *			(this may need to be extended...)
-	 */
-reset_or_panic:
-	synco	/* TAKum03020 (but probably a good idea anyway.) */
-	putcon	SP, DCR
-	/* First save r0-1 and tr0, as we need to use these */
-	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
-	st.q	SP, 0, r0
-	st.q	SP, 8, r1
-	gettr	tr0, r0
-	st.q	SP, 32, r0
-
-	/* Check cause */
-	getcon	EXPEVT, r0
-	movi	RESET_CAUSE, r1
-	sub	r1, r0, r1		/* r1=0 if reset */
-	movi	_stext-CONFIG_CACHED_MEMORY_OFFSET, r0
-	ori	r0, 1, r0
-	ptabs	r0, tr0
-	beqi	r1, 0, tr0		/* Jump to start address if reset */
-
-	getcon	EXPEVT, r0
-	movi	DEBUGSS_CAUSE, r1
-	sub	r1, r0, r1		/* r1=0 if single step */
-	pta	single_step_panic, tr0
-	beqi	r1, 0, tr0		/* jump if single step */
-
-	/* Now jump to where we save the registers. */
-	movi	panic_stash_regs-CONFIG_CACHED_MEMORY_OFFSET, r1
-	ptabs	r1, tr0
-	blink	tr0, r63
-
-single_step_panic:
-	/* We are in a handler with Single Step set. We need to resume the
-	 * handler, by turning on MMU & turning off Single Step. */
-	getcon	SSR, r0
-	movi	SR_MMU, r1
-	or	r0, r1, r0
-	movi	~SR_SS, r1
-	and	r0, r1, r0
-	putcon	r0, SSR
-	/* Restore EXPEVT, as the rte won't do this */
-	getcon	PEXPEVT, r0
-	putcon	r0, EXPEVT
-	/* Restore regs */
-	ld.q	SP, 32, r0
-	ptabs	r0, tr0
-	ld.q	SP, 0, r0
-	ld.q	SP, 8, r1
-	getcon	DCR, SP
-	synco
-	rte
-
-
-	.balign	256
-debug_exception:
-	synco	/* TAKum03020 (but probably a good idea anyway.) */
-	/*
-	 * Single step/software_break_point first level handler.
-	 * Called with MMU off, so the first thing we do is enable it
-	 * by doing an rte with appropriate SSR.
-	 */
-	putcon	SP, DCR
-	/* Save SSR & SPC, together with R0 & R1, as we need to use 2 regs. */
-	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
-
-	/* With the MMU off, we are bypassing the cache, so purge any
-         * data that will be made stale by the following stores.
-         */
-	ocbp	SP, 0
-	synco
-
-	st.q	SP, 0, r0
-	st.q	SP, 8, r1
-	getcon	SPC, r0
-	st.q	SP, 16, r0
-	getcon	SSR, r0
-	st.q	SP, 24, r0
-
-	/* Enable MMU, block exceptions, set priv mode, disable single step */
-	movi	SR_MMU | SR_BL | SR_MD, r1
-	or	r0, r1, r0
-	movi	~SR_SS, r1
-	and	r0, r1, r0
-	putcon	r0, SSR
-	/* Force control to debug_exception_2 when rte is executed */
-	movi	debug_exeception_2, r0
-	ori	r0, 1, r0      /* force SHmedia, just in case */
-	putcon	r0, SPC
-	getcon	DCR, SP
-	synco
-	rte
-debug_exeception_2:
-	/* Restore saved regs */
-	putcon	SP, KCR1
-	movi	resvec_save_area, SP
-	ld.q	SP, 24, r0
-	putcon	r0, SSR
-	ld.q	SP, 16, r0
-	putcon	r0, SPC
-	ld.q	SP, 0, r0
-	ld.q	SP, 8, r1
-
-	/* Save other original registers into reg_save_area */
-        movi  reg_save_area, SP
-	st.q	SP, SAVED_R2, r2
-	st.q	SP, SAVED_R3, r3
-	st.q	SP, SAVED_R4, r4
-	st.q	SP, SAVED_R5, r5
-	st.q	SP, SAVED_R6, r6
-	st.q	SP, SAVED_R18, r18
-	gettr	tr0, r3
-	st.q	SP, SAVED_TR0, r3
-
-	/* Set args for debug class handler */
-	getcon	EXPEVT, r2
-	movi	ret_from_exception, r3
-	ori	r3, 1, r3
-	movi	EVENT_DEBUG, r4
-	or	SP, ZERO, r5
-	getcon	KCR1, SP
-	pta	handle_exception, tr0
-	blink	tr0, ZERO
-
-	.balign	256
-debug_interrupt:
-	/* !!! WE COME HERE IN REAL MODE !!! */
-	/* Hook-up debug interrupt to allow various debugging options to be
-	 * hooked into its handler. */
-	/* Save original stack pointer into KCR1 */
-	synco
-	putcon	SP, KCR1
-	movi	resvec_save_area-CONFIG_CACHED_MEMORY_OFFSET, SP
-	ocbp	SP, 0
-	ocbp	SP, 32
-	synco
-
-	/* Save other original registers into reg_save_area thru real addresses */
-	st.q	SP, SAVED_R2, r2
-	st.q	SP, SAVED_R3, r3
-	st.q	SP, SAVED_R4, r4
-	st.q	SP, SAVED_R5, r5
-	st.q	SP, SAVED_R6, r6
-	st.q	SP, SAVED_R18, r18
-	gettr	tr0, r3
-	st.q	SP, SAVED_TR0, r3
-
-	/* move (spc,ssr)->(pspc,pssr).  The rte will shift
-	   them back again, so that they look like the originals
-	   as far as the real handler code is concerned. */
-	getcon	spc, r6
-	putcon	r6, pspc
-	getcon	ssr, r6
-	putcon	r6, pssr
-
-	! construct useful SR for handle_exception
-	movi	3, r6
-	shlli	r6, 30, r6
-	getcon	sr, r18
-	or	r18, r6, r6
-	putcon	r6, ssr
-
-	! SSR is now the current SR with the MD and MMU bits set
-	! i.e. the rte will switch back to priv mode and put
-	! the mmu back on
-
-	! construct spc
-	movi	handle_exception, r18
-	ori	r18, 1, r18		! for safety (do we need this?)
-	putcon	r18, spc
-
-	/* Set args for Non-debug, Not a TLB miss class handler */
-
-	! EXPEVT==0x80 is unused, so 'steal' this value to put the
-	! debug interrupt handler in the vectoring table
-	movi	0x80, r2
-	movi	ret_from_exception, r3
-	ori	r3, 1, r3
-	movi	EVENT_FAULT_NOT_TLB, r4
-
-	or	SP, ZERO, r5
-	movi	CONFIG_CACHED_MEMORY_OFFSET, r6
-	add	r6, r5, r5
-	getcon	KCR1, SP
-
-	synco	! for safety
-	rte	! -> handle_exception, switch back to priv mode again
-
-LRESVEC_block_end:			/* Marker. Unused. */
-
-	.balign	TEXT_SIZE
-
-/*
- * Second level handler for VBR-based exceptions. Pre-handler.
- * In common to all stack-frame sensitive handlers.
- *
- * Inputs:
- * (KCR0) Current [current task union]
- * (KCR1) Original SP
- * (r2)   INTEVT/EXPEVT
- * (r3)   appropriate return address
- * (r4)   Event (0 = interrupt, 1 = TLB miss fault, 2 = Not TLB miss fault, 3=debug)
- * (r5)   Pointer to reg_save_area
- * (SP)   Original SP
- *
- * Available registers:
- * (r6)
- * (r18)
- * (tr0)
- *
- */
-handle_exception:
-	/* Common 2nd level handler. */
-
-	/* First thing we need an appropriate stack pointer */
-	getcon	SSR, r6
-	shlri	r6, 30, r6
-	andi	r6, 1, r6
-	pta	stack_ok, tr0
-	bne	r6, ZERO, tr0		/* Original stack pointer is fine */
-
-	/* Set stack pointer for user fault */
-	getcon	KCR0, SP
-	movi	THREAD_SIZE, r6		/* Point to the end */
-	add	SP, r6, SP
-
-stack_ok:
-
-/* DEBUG : check for underflow/overflow of the kernel stack */
-	pta	no_underflow, tr0
-	getcon  KCR0, r6
-	movi	1024, r18
-	add	r6, r18, r6
-	bge	SP, r6, tr0 	! ? below 1k from bottom of stack : danger zone
-
-/* Just panic to cause a crash. */
-bad_sp:
-	ld.b	r63, 0, r6
-	nop
-
-no_underflow:
-	pta	bad_sp, tr0
-	getcon	kcr0, r6
-	movi	THREAD_SIZE, r18
-	add	r18, r6, r6
-	bgt	SP, r6, tr0	! sp above the stack
-
-	/* Make some room for the BASIC frame. */
-	movi	-(FRAME_SIZE), r6
-	add	SP, r6, SP
-
-/* Could do this with no stalling if we had another spare register, but the
-   code below will be OK. */
-	ld.q	r5, SAVED_R2, r6
-	ld.q	r5, SAVED_R3, r18
-	st.q	SP, FRAME_R(2), r6
-	ld.q	r5, SAVED_R4, r6
-	st.q	SP, FRAME_R(3), r18
-	ld.q	r5, SAVED_R5, r18
-	st.q	SP, FRAME_R(4), r6
-	ld.q	r5, SAVED_R6, r6
-	st.q	SP, FRAME_R(5), r18
-	ld.q	r5, SAVED_R18, r18
-	st.q	SP, FRAME_R(6), r6
-	ld.q	r5, SAVED_TR0, r6
-	st.q	SP, FRAME_R(18), r18
-	st.q	SP, FRAME_T(0), r6
-
-	/* Keep old SP around */
-	getcon	KCR1, r6
-
-	/* Save the rest of the general purpose registers */
-	st.q	SP, FRAME_R(0), r0
-	st.q	SP, FRAME_R(1), r1
-	st.q	SP, FRAME_R(7), r7
-	st.q	SP, FRAME_R(8), r8
-	st.q	SP, FRAME_R(9), r9
-	st.q	SP, FRAME_R(10), r10
-	st.q	SP, FRAME_R(11), r11
-	st.q	SP, FRAME_R(12), r12
-	st.q	SP, FRAME_R(13), r13
-	st.q	SP, FRAME_R(14), r14
-
-	/* SP is somewhere else */
-	st.q	SP, FRAME_R(15), r6
-
-	st.q	SP, FRAME_R(16), r16
-	st.q	SP, FRAME_R(17), r17
-	/* r18 is saved earlier. */
-	st.q	SP, FRAME_R(19), r19
-	st.q	SP, FRAME_R(20), r20
-	st.q	SP, FRAME_R(21), r21
-	st.q	SP, FRAME_R(22), r22
-	st.q	SP, FRAME_R(23), r23
-	st.q	SP, FRAME_R(24), r24
-	st.q	SP, FRAME_R(25), r25
-	st.q	SP, FRAME_R(26), r26
-	st.q	SP, FRAME_R(27), r27
-	st.q	SP, FRAME_R(28), r28
-	st.q	SP, FRAME_R(29), r29
-	st.q	SP, FRAME_R(30), r30
-	st.q	SP, FRAME_R(31), r31
-	st.q	SP, FRAME_R(32), r32
-	st.q	SP, FRAME_R(33), r33
-	st.q	SP, FRAME_R(34), r34
-	st.q	SP, FRAME_R(35), r35
-	st.q	SP, FRAME_R(36), r36
-	st.q	SP, FRAME_R(37), r37
-	st.q	SP, FRAME_R(38), r38
-	st.q	SP, FRAME_R(39), r39
-	st.q	SP, FRAME_R(40), r40
-	st.q	SP, FRAME_R(41), r41
-	st.q	SP, FRAME_R(42), r42
-	st.q	SP, FRAME_R(43), r43
-	st.q	SP, FRAME_R(44), r44
-	st.q	SP, FRAME_R(45), r45
-	st.q	SP, FRAME_R(46), r46
-	st.q	SP, FRAME_R(47), r47
-	st.q	SP, FRAME_R(48), r48
-	st.q	SP, FRAME_R(49), r49
-	st.q	SP, FRAME_R(50), r50
-	st.q	SP, FRAME_R(51), r51
-	st.q	SP, FRAME_R(52), r52
-	st.q	SP, FRAME_R(53), r53
-	st.q	SP, FRAME_R(54), r54
-	st.q	SP, FRAME_R(55), r55
-	st.q	SP, FRAME_R(56), r56
-	st.q	SP, FRAME_R(57), r57
-	st.q	SP, FRAME_R(58), r58
-	st.q	SP, FRAME_R(59), r59
-	st.q	SP, FRAME_R(60), r60
-	st.q	SP, FRAME_R(61), r61
-	st.q	SP, FRAME_R(62), r62
-
-	/*
-	 * Save the S* registers.
-	 */
-	getcon	SSR, r61
-	st.q	SP, FRAME_S(FSSR), r61
-	getcon	SPC, r62
-	st.q	SP, FRAME_S(FSPC), r62
-	movi	-1, r62			/* Reset syscall_nr */
-	st.q	SP, FRAME_S(FSYSCALL_ID), r62
-
-	/* Save the rest of the target registers */
-	gettr	tr1, r6
-	st.q	SP, FRAME_T(1), r6
-	gettr	tr2, r6
-	st.q	SP, FRAME_T(2), r6
-	gettr	tr3, r6
-	st.q	SP, FRAME_T(3), r6
-	gettr	tr4, r6
-	st.q	SP, FRAME_T(4), r6
-	gettr	tr5, r6
-	st.q	SP, FRAME_T(5), r6
-	gettr	tr6, r6
-	st.q	SP, FRAME_T(6), r6
-	gettr	tr7, r6
-	st.q	SP, FRAME_T(7), r6
-
-	! setup FP so that unwinder can wind back through nested kernel mode
-	! exceptions
-	add	SP, ZERO, r14
-
-#ifdef CONFIG_POOR_MANS_STRACE
-	/* We've pushed all the registers now, so only r2-r4 hold anything
-	 * useful. Move them into callee save registers */
-	or	r2, ZERO, r28
-	or	r3, ZERO, r29
-	or	r4, ZERO, r30
-
-	/* Preserve r2 as the event code */
-	movi	evt_debug, r3
-	ori	r3, 1, r3
-	ptabs	r3, tr0
-
-	or	SP, ZERO, r6
-	getcon	TRA, r5
-	blink	tr0, LINK
-
-	or	r28, ZERO, r2
-	or	r29, ZERO, r3
-	or	r30, ZERO, r4
-#endif
-
-	/* For syscall and debug race condition, get TRA now */
-	getcon	TRA, r5
-
-	/* We are in a safe position to turn SR.BL off, but set IMASK=0xf
-	 * Also set FD, to catch FPU usage in the kernel.
-	 *
-	 * benedict.gaster@superh.com 29/07/2002
-	 *
-	 * On all SH5-101 revisions it is unsafe to raise the IMASK and at the
-	 * same time change BL from 1->0, as any pending interrupt of a level
-	 * higher than he previous value of IMASK will leak through and be
-	 * taken unexpectedly.
-	 *
-	 * To avoid this we raise the IMASK and then issue another PUTCON to
-	 * enable interrupts.
-         */
-	getcon	SR, r6
-	movi	SR_IMASK | SR_FD, r7
-	or	r6, r7, r6
-	putcon	r6, SR
-	movi	SR_UNBLOCK_EXC, r7
-	and	r6, r7, r6
-	putcon	r6, SR
-
-
-	/* Now call the appropriate 3rd level handler */
-	or	r3, ZERO, LINK
-	movi	trap_jtable, r3
-	shlri	r2, 3, r2
-	ldx.l	r2, r3, r3
-	shlri	r2, 2, r2
-	ptabs	r3, tr0
-	or	SP, ZERO, r3
-	blink	tr0, ZERO
-
-/*
- * Second level handler for VBR-based exceptions. Post-handlers.
- *
- * Post-handlers for interrupts (ret_from_irq), exceptions
- * (ret_from_exception) and common reentrance doors (restore_all
- * to get back to the original context, ret_from_syscall loop to
- * check kernel exiting).
- *
- * ret_with_reschedule and work_notifysig are an inner lables of
- * the ret_from_syscall loop.
- *
- * In common to all stack-frame sensitive handlers.
- *
- * Inputs:
- * (SP)   struct pt_regs *, original register's frame pointer (basic)
- *
- */
-	.global ret_from_irq
-ret_from_irq:
-#ifdef CONFIG_POOR_MANS_STRACE
-	pta	evt_debug_ret_from_irq, tr0
-	ori	SP, 0, r2
-	blink	tr0, LINK
-#endif
-	ld.q	SP, FRAME_S(FSSR), r6
-	shlri	r6, 30, r6
-	andi	r6, 1, r6
-	pta	resume_kernel, tr0
-	bne	r6, ZERO, tr0		/* no further checks */
-	STI()
-	pta	ret_with_reschedule, tr0
-	blink	tr0, ZERO		/* Do not check softirqs */
-
-	.global ret_from_exception
-ret_from_exception:
-	preempt_stop()
-
-#ifdef CONFIG_POOR_MANS_STRACE
-	pta	evt_debug_ret_from_exc, tr0
-	ori	SP, 0, r2
-	blink	tr0, LINK
-#endif
-
-	ld.q	SP, FRAME_S(FSSR), r6
-	shlri	r6, 30, r6
-	andi	r6, 1, r6
-	pta	resume_kernel, tr0
-	bne	r6, ZERO, tr0		/* no further checks */
-
-	/* Check softirqs */
-
-#ifdef CONFIG_PREEMPT
-	pta   ret_from_syscall, tr0
-	blink   tr0, ZERO
-
-resume_kernel:
-	pta	restore_all, tr0
-
-	getcon	KCR0, r6
-	ld.l	r6, TI_PRE_COUNT, r7
-	beq/u	r7, ZERO, tr0
-
-need_resched:
-	ld.l	r6, TI_FLAGS, r7
-	movi	(1 << TIF_NEED_RESCHED), r8
-	and	r8, r7, r8
-	bne	r8, ZERO, tr0
-
-	getcon	SR, r7
-	andi	r7, 0xf0, r7
-	bne	r7, ZERO, tr0
-
-	movi	((PREEMPT_ACTIVE >> 16) & 65535), r8
-	shori	(PREEMPT_ACTIVE & 65535), r8
-	st.l	r6, TI_PRE_COUNT, r8
-
-	STI()
-	movi	schedule, r7
-	ori	r7, 1, r7
-	ptabs	r7, tr1
-	blink	tr1, LINK
-
-	st.l	r6, TI_PRE_COUNT, ZERO
-	CLI()
-
-	pta	need_resched, tr1
-	blink	tr1, ZERO
-#endif
-
-	.global ret_from_syscall
-ret_from_syscall:
-
-ret_with_reschedule:
-	getcon	KCR0, r6		! r6 contains current_thread_info
-	ld.l	r6, TI_FLAGS, r7	! r7 contains current_thread_info->flags
-
-	! FIXME:!!!
-	! no handling of TIF_SYSCALL_TRACE yet!!
-
-	movi	_TIF_NEED_RESCHED, r8
-	and	r8, r7, r8
-	pta	work_resched, tr0
-	bne	r8, ZERO, tr0
-
-	pta	restore_all, tr1
-
-	movi	(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), r8
-	and	r8, r7, r8
-	pta	work_notifysig, tr0
-	bne	r8, ZERO, tr0
-
-	blink	tr1, ZERO
-
-work_resched:
-	pta	ret_from_syscall, tr0
-	gettr	tr0, LINK
-	movi	schedule, r6
-	ptabs	r6, tr0
-	blink	tr0, ZERO		/* Call schedule(), return on top */
-
-work_notifysig:
-	gettr	tr1, LINK
-
-	movi	do_signal, r6
-	ptabs	r6, tr0
-	or	SP, ZERO, r2
-	or	ZERO, ZERO, r3
-	blink	tr0, LINK	    /* Call do_signal(regs, 0), return here */
-
-restore_all:
-	/* Do prefetches */
-
-	ld.q	SP, FRAME_T(0), r6
-	ld.q	SP, FRAME_T(1), r7
-	ld.q	SP, FRAME_T(2), r8
-	ld.q	SP, FRAME_T(3), r9
-	ptabs	r6, tr0
-	ptabs	r7, tr1
-	ptabs	r8, tr2
-	ptabs	r9, tr3
-	ld.q	SP, FRAME_T(4), r6
-	ld.q	SP, FRAME_T(5), r7
-	ld.q	SP, FRAME_T(6), r8
-	ld.q	SP, FRAME_T(7), r9
-	ptabs	r6, tr4
-	ptabs	r7, tr5
-	ptabs	r8, tr6
-	ptabs	r9, tr7
-
-	ld.q	SP, FRAME_R(0), r0
-	ld.q	SP, FRAME_R(1), r1
-	ld.q	SP, FRAME_R(2), r2
-	ld.q	SP, FRAME_R(3), r3
-	ld.q	SP, FRAME_R(4), r4
-	ld.q	SP, FRAME_R(5), r5
-	ld.q	SP, FRAME_R(6), r6
-	ld.q	SP, FRAME_R(7), r7
-	ld.q	SP, FRAME_R(8), r8
-	ld.q	SP, FRAME_R(9), r9
-	ld.q	SP, FRAME_R(10), r10
-	ld.q	SP, FRAME_R(11), r11
-	ld.q	SP, FRAME_R(12), r12
-	ld.q	SP, FRAME_R(13), r13
-	ld.q	SP, FRAME_R(14), r14
-
-	ld.q	SP, FRAME_R(16), r16
-	ld.q	SP, FRAME_R(17), r17
-	ld.q	SP, FRAME_R(18), r18
-	ld.q	SP, FRAME_R(19), r19
-	ld.q	SP, FRAME_R(20), r20
-	ld.q	SP, FRAME_R(21), r21
-	ld.q	SP, FRAME_R(22), r22
-	ld.q	SP, FRAME_R(23), r23
-	ld.q	SP, FRAME_R(24), r24
-	ld.q	SP, FRAME_R(25), r25
-	ld.q	SP, FRAME_R(26), r26
-	ld.q	SP, FRAME_R(27), r27
-	ld.q	SP, FRAME_R(28), r28
-	ld.q	SP, FRAME_R(29), r29
-	ld.q	SP, FRAME_R(30), r30
-	ld.q	SP, FRAME_R(31), r31
-	ld.q	SP, FRAME_R(32), r32
-	ld.q	SP, FRAME_R(33), r33
-	ld.q	SP, FRAME_R(34), r34
-	ld.q	SP, FRAME_R(35), r35
-	ld.q	SP, FRAME_R(36), r36
-	ld.q	SP, FRAME_R(37), r37
-	ld.q	SP, FRAME_R(38), r38
-	ld.q	SP, FRAME_R(39), r39
-	ld.q	SP, FRAME_R(40), r40
-	ld.q	SP, FRAME_R(41), r41
-	ld.q	SP, FRAME_R(42), r42
-	ld.q	SP, FRAME_R(43), r43
-	ld.q	SP, FRAME_R(44), r44
-	ld.q	SP, FRAME_R(45), r45
-	ld.q	SP, FRAME_R(46), r46
-	ld.q	SP, FRAME_R(47), r47
-	ld.q	SP, FRAME_R(48), r48
-	ld.q	SP, FRAME_R(49), r49
-	ld.q	SP, FRAME_R(50), r50
-	ld.q	SP, FRAME_R(51), r51
-	ld.q	SP, FRAME_R(52), r52
-	ld.q	SP, FRAME_R(53), r53
-	ld.q	SP, FRAME_R(54), r54
-	ld.q	SP, FRAME_R(55), r55
-	ld.q	SP, FRAME_R(56), r56
-	ld.q	SP, FRAME_R(57), r57
-	ld.q	SP, FRAME_R(58), r58
-
-	getcon	SR, r59
-	movi	SR_BLOCK_EXC, r60
-	or	r59, r60, r59
-	putcon	r59, SR			/* SR.BL = 1, keep nesting out */
-	ld.q	SP, FRAME_S(FSSR), r61
-	ld.q	SP, FRAME_S(FSPC), r62
-	movi	SR_ASID_MASK, r60
-	and	r59, r60, r59
-	andc	r61, r60, r61		/* Clear out older ASID */
-	or	r59, r61, r61		/* Retain current ASID */
-	putcon	r61, SSR
-	putcon	r62, SPC
-
-	/* Ignore FSYSCALL_ID */
-
-	ld.q	SP, FRAME_R(59), r59
-	ld.q	SP, FRAME_R(60), r60
-	ld.q	SP, FRAME_R(61), r61
-	ld.q	SP, FRAME_R(62), r62
-
-	/* Last touch */
-	ld.q	SP, FRAME_R(15), SP
-	rte
-	nop
-
-/*
- * Third level handlers for VBR-based exceptions. Adapting args to
- * and/or deflecting to fourth level handlers.
- *
- * Fourth level handlers interface.
- * Most are C-coded handlers directly pointed by the trap_jtable.
- * (Third = Fourth level)
- * Inputs:
- * (r2)   fault/interrupt code, entry number (e.g. NMI = 14,
- *	  IRL0-3 (0000) = 16, RTLBMISS = 2, SYSCALL = 11, etc ...)
- * (r3)   struct pt_regs *, original register's frame pointer
- * (r4)   Event (0 = interrupt, 1 = TLB miss fault, 2 = Not TLB miss fault)
- * (r5)   TRA control register (for syscall/debug benefit only)
- * (LINK) return address
- * (SP)   = r3
- *
- * Kernel TLB fault handlers will get a slightly different interface.
- * (r2)   struct pt_regs *, original register's frame pointer
- * (r3)   writeaccess, whether it's a store fault as opposed to load fault
- * (r4)   execaccess, whether it's a ITLB fault as opposed to DTLB fault
- * (r5)   Effective Address of fault
- * (LINK) return address
- * (SP)   = r2
- *
- * fpu_error_or_IRQ? is a helper to deflect to the right cause.
- *
- */
-tlb_miss_load:
-	or	SP, ZERO, r2
-	or	ZERO, ZERO, r3		/* Read */
-	or	ZERO, ZERO, r4		/* Data */
-	getcon	TEA, r5
-	pta	call_do_page_fault, tr0
-	beq	ZERO, ZERO, tr0
-
-tlb_miss_store:
-	or	SP, ZERO, r2
-	movi	1, r3			/* Write */
-	or	ZERO, ZERO, r4		/* Data */
-	getcon	TEA, r5
-	pta	call_do_page_fault, tr0
-	beq	ZERO, ZERO, tr0
-
-itlb_miss_or_IRQ:
-	pta	its_IRQ, tr0
-	beqi/u	r4, EVENT_INTERRUPT, tr0
-	or	SP, ZERO, r2
-	or	ZERO, ZERO, r3		/* Read */
-	movi	1, r4			/* Text */
-	getcon	TEA, r5
-	/* Fall through */
-
-call_do_page_fault:
-	movi	do_page_fault, r6
-        ptabs	r6, tr0
-        blink	tr0, ZERO
-
-fpu_error_or_IRQA:
-	pta	its_IRQ, tr0
-	beqi/l	r4, EVENT_INTERRUPT, tr0
-#ifdef CONFIG_SH_FPU
-	movi	do_fpu_state_restore, r6
-#else
-	movi	do_exception_error, r6
-#endif
-	ptabs	r6, tr0
-	blink	tr0, ZERO
-
-fpu_error_or_IRQB:
-	pta	its_IRQ, tr0
-	beqi/l	r4, EVENT_INTERRUPT, tr0
-#ifdef CONFIG_SH_FPU
-	movi	do_fpu_state_restore, r6
-#else
-	movi	do_exception_error, r6
-#endif
-	ptabs	r6, tr0
-	blink	tr0, ZERO
-
-its_IRQ:
-	movi	do_IRQ, r6
-	ptabs	r6, tr0
-	blink	tr0, ZERO
-
-/*
- * system_call/unknown_trap third level handler:
- *
- * Inputs:
- * (r2)   fault/interrupt code, entry number (TRAP = 11)
- * (r3)   struct pt_regs *, original register's frame pointer
- * (r4)   Not used. Event (0=interrupt, 1=TLB miss fault, 2=Not TLB miss fault)
- * (r5)   TRA Control Reg (0x00xyzzzz: x=1 SYSCALL, y = #args, z=nr)
- * (SP)   = r3
- * (LINK) return address: ret_from_exception
- * (*r3)  Syscall parms: SC#, arg0, arg1, ..., arg5 in order (Saved r2/r7)
- *
- * Outputs:
- * (*r3)  Syscall reply (Saved r2)
- * (LINK) In case of syscall only it can be scrapped.
- *        Common second level post handler will be ret_from_syscall.
- *        Common (non-trace) exit point to that is syscall_ret (saving
- *        result to r2). Common bad exit point is syscall_bad (returning
- *        ENOSYS then saved to r2).
- *
- */
-
-unknown_trap:
-	/* Unknown Trap or User Trace */
-	movi	do_unknown_trapa, r6
-	ptabs	r6, tr0
-        ld.q    r3, FRAME_R(9), r2	/* r2 = #arg << 16 | syscall # */
-        andi    r2, 0x1ff, r2		/* r2 = syscall # */
-	blink	tr0, LINK
-
-	pta	syscall_ret, tr0
-	blink	tr0, ZERO
-
-        /* New syscall implementation*/
-system_call:
-	pta	unknown_trap, tr0
-        or      r5, ZERO, r4            /* TRA (=r5) -> r4 */
-        shlri   r4, 20, r4
-	bnei	r4, 1, tr0		/* unknown_trap if not 0x1yzzzz */
-
-        /* It's a system call */
-	st.q    r3, FRAME_S(FSYSCALL_ID), r5 	/* ID (0x1yzzzz) -> stack */
-	andi    r5, 0x1ff, r5			/* syscall # -> r5	  */
-
-	STI()
-
-	pta	syscall_allowed, tr0
-	movi	NR_syscalls - 1, r4	/* Last valid */
-	bgeu/l	r4, r5, tr0
-
-syscall_bad:
-	/* Return ENOSYS ! */
-	movi	-(ENOSYS), r2		/* Fall-through */
-
-	.global syscall_ret
-syscall_ret:
-	st.q	SP, FRAME_R(9), r2	/* Expecting SP back to BASIC frame */
-
-#ifdef CONFIG_POOR_MANS_STRACE
-	/* nothing useful in registers at this point */
-
-	movi	evt_debug2, r5
-	ori	r5, 1, r5
-	ptabs	r5, tr0
-	ld.q	SP, FRAME_R(9), r2
-	or	SP, ZERO, r3
-	blink	tr0, LINK
-#endif
-
-	ld.q	SP, FRAME_S(FSPC), r2
-	addi	r2, 4, r2		/* Move PC, being pre-execution event */
-	st.q	SP, FRAME_S(FSPC), r2
-	pta	ret_from_syscall, tr0
-	blink	tr0, ZERO
-
-
-/*  A different return path for ret_from_fork, because we now need
- *  to call schedule_tail with the later kernels. Because prev is
- *  loaded into r2 by switch_to() means we can just call it straight  away
- */
-
-.global	ret_from_fork
-ret_from_fork:
-
-	movi	schedule_tail,r5
-	ori	r5, 1, r5
-	ptabs	r5, tr0
-	blink	tr0, LINK
-
-#ifdef CONFIG_POOR_MANS_STRACE
-	/* nothing useful in registers at this point */
-
-	movi	evt_debug2, r5
-	ori	r5, 1, r5
-	ptabs	r5, tr0
-	ld.q	SP, FRAME_R(9), r2
-	or	SP, ZERO, r3
-	blink	tr0, LINK
-#endif
-
-	ld.q	SP, FRAME_S(FSPC), r2
-	addi	r2, 4, r2		/* Move PC, being pre-execution event */
-	st.q	SP, FRAME_S(FSPC), r2
-	pta	ret_from_syscall, tr0
-	blink	tr0, ZERO
-
-
-
-syscall_allowed:
-	/* Use LINK to deflect the exit point, default is syscall_ret */
-	pta	syscall_ret, tr0
-	gettr	tr0, LINK
-	pta	syscall_notrace, tr0
-
-	getcon	KCR0, r2
-	ld.l	r2, TI_FLAGS, r4
-	movi	(1 << TIF_SYSCALL_TRACE), r6
-	and	r6, r4, r6
-	beq/l	r6, ZERO, tr0
-
-	/* Trace it by calling syscall_trace before and after */
-	movi	syscall_trace, r4
-	ptabs	r4, tr0
-	blink	tr0, LINK
-	/* Reload syscall number as r5 is trashed by syscall_trace */
-	ld.q	SP, FRAME_S(FSYSCALL_ID), r5
-	andi	r5, 0x1ff, r5
-
-	pta	syscall_ret_trace, tr0
-	gettr	tr0, LINK
-
-syscall_notrace:
-	/* Now point to the appropriate 4th level syscall handler */
-	movi	sys_call_table, r4
-	shlli	r5, 2, r5
-	ldx.l	r4, r5, r5
-	ptabs	r5, tr0
-
-	/* Prepare original args */
-	ld.q	SP, FRAME_R(2), r2
-	ld.q	SP, FRAME_R(3), r3
-	ld.q	SP, FRAME_R(4), r4
-	ld.q	SP, FRAME_R(5), r5
-	ld.q	SP, FRAME_R(6), r6
-	ld.q	SP, FRAME_R(7), r7
-
-	/* And now the trick for those syscalls requiring regs * ! */
-	or	SP, ZERO, r8
-
-	/* Call it */
-	blink	tr0, ZERO	/* LINK is already properly set */
-
-syscall_ret_trace:
-	/* We get back here only if under trace */
-	st.q	SP, FRAME_R(9), r2	/* Save return value */
-
-	movi	syscall_trace, LINK
-	ptabs	LINK, tr0
-	blink	tr0, LINK
-
-	/* This needs to be done after any syscall tracing */
-	ld.q	SP, FRAME_S(FSPC), r2
-	addi	r2, 4, r2	/* Move PC, being pre-execution event */
-	st.q	SP, FRAME_S(FSPC), r2
-
-	pta	ret_from_syscall, tr0
-	blink	tr0, ZERO		/* Resume normal return sequence */
-
-/*
- * --- Switch to running under a particular ASID and return the previous ASID value
- * --- The caller is assumed to have done a cli before calling this.
- *
- * Input r2 : new ASID
- * Output r2 : old ASID
- */
-
-	.global switch_and_save_asid
-switch_and_save_asid:
-	getcon	sr, r0
-	movi	255, r4
-	shlli 	r4, 16, r4	/* r4 = mask to select ASID */
-	and	r0, r4, r3	/* r3 = shifted old ASID */
-	andi	r2, 255, r2	/* mask down new ASID */
-	shlli	r2, 16, r2	/* align new ASID against SR.ASID */
-	andc	r0, r4, r0	/* efface old ASID from SR */
-	or	r0, r2, r0	/* insert the new ASID */
-	putcon	r0, ssr
-	movi	1f, r0
-	putcon	r0, spc
-	rte
-	nop
-1:
-	ptabs	LINK, tr0
-	shlri	r3, 16, r2	/* r2 = old ASID */
-	blink tr0, r63
-
-	.global	route_to_panic_handler
-route_to_panic_handler:
-	/* Switch to real mode, goto panic_handler, don't return.  Useful for
-	   last-chance debugging, e.g. if no output wants to go to the console.
-	   */
-
-	movi	panic_handler - CONFIG_CACHED_MEMORY_OFFSET, r1
-	ptabs	r1, tr0
-	pta	1f, tr1
-	gettr	tr1, r0
-	putcon	r0, spc
-	getcon	sr, r0
-	movi	1, r1
-	shlli	r1, 31, r1
-	andc	r0, r1, r0
-	putcon	r0, ssr
-	rte
-	nop
-1:	/* Now in real mode */
-	blink tr0, r63
-	nop
-
-	.global peek_real_address_q
-peek_real_address_q:
-	/* Two args:
-	   r2 : real mode address to peek
-	   r2(out) : result quadword
-
-	   This is provided as a cheapskate way of manipulating device
-	   registers for debugging (to avoid the need to onchip_remap the debug
-	   module, and to avoid the need to onchip_remap the watchpoint
-	   controller in a way that identity maps sufficient bits to avoid the
-	   SH5-101 cut2 silicon defect).
-
-	   This code is not performance critical
-	*/
-
-	add.l	r2, r63, r2	/* sign extend address */
-	getcon	sr, r0		/* r0 = saved original SR */
-	movi	1, r1
-	shlli	r1, 28, r1
-	or	r0, r1, r1	/* r0 with block bit set */
-	putcon	r1, sr		/* now in critical section */
-	movi	1, r36
-	shlli	r36, 31, r36
-	andc	r1, r36, r1	/* turn sr.mmu off in real mode section */
-
-	putcon	r1, ssr
-	movi	.peek0 - CONFIG_CACHED_MEMORY_OFFSET, r36 /* real mode target address */
-	movi	1f, r37		/* virtual mode return addr */
-	putcon	r36, spc
-
-	synco
-	rte
-	nop
-
-.peek0:	/* come here in real mode, don't touch caches!!
-           still in critical section (sr.bl==1) */
-	putcon	r0, ssr
-	putcon	r37, spc
-	/* Here's the actual peek.  If the address is bad, all bets are now off
-	 * what will happen (handlers invoked in real-mode = bad news) */
-	ld.q	r2, 0, r2
-	synco
-	rte	/* Back to virtual mode */
-	nop
-
-1:
-	ptabs	LINK, tr0
-	blink	tr0, r63
-
-	.global poke_real_address_q
-poke_real_address_q:
-	/* Two args:
-	   r2 : real mode address to poke
-	   r3 : quadword value to write.
-
-	   This is provided as a cheapskate way of manipulating device
-	   registers for debugging (to avoid the need to onchip_remap the debug
-	   module, and to avoid the need to onchip_remap the watchpoint
-	   controller in a way that identity maps sufficient bits to avoid the
-	   SH5-101 cut2 silicon defect).
-
-	   This code is not performance critical
-	*/
-
-	add.l	r2, r63, r2	/* sign extend address */
-	getcon	sr, r0		/* r0 = saved original SR */
-	movi	1, r1
-	shlli	r1, 28, r1
-	or	r0, r1, r1	/* r0 with block bit set */
-	putcon	r1, sr		/* now in critical section */
-	movi	1, r36
-	shlli	r36, 31, r36
-	andc	r1, r36, r1	/* turn sr.mmu off in real mode section */
-
-	putcon	r1, ssr
-	movi	.poke0-CONFIG_CACHED_MEMORY_OFFSET, r36 /* real mode target address */
-	movi	1f, r37		/* virtual mode return addr */
-	putcon	r36, spc
-
-	synco
-	rte
-	nop
-
-.poke0:	/* come here in real mode, don't touch caches!!
-           still in critical section (sr.bl==1) */
-	putcon	r0, ssr
-	putcon	r37, spc
-	/* Here's the actual poke.  If the address is bad, all bets are now off
-	 * what will happen (handlers invoked in real-mode = bad news) */
-	st.q	r2, 0, r3
-	synco
-	rte	/* Back to virtual mode */
-	nop
-
-1:
-	ptabs	LINK, tr0
-	blink	tr0, r63
-
-/*
- * --- User Access Handling Section
- */
-
-/*
- * User Access support. It all moved to non inlined Assembler
- * functions in here.
- *
- * __kernel_size_t __copy_user(void *__to, const void *__from,
- *			       __kernel_size_t __n)
- *
- * Inputs:
- * (r2)  target address
- * (r3)  source address
- * (r4)  size in bytes
- *
- * Ouputs:
- * (*r2) target data
- * (r2)  non-copied bytes
- *
- * If a fault occurs on the user pointer, bail out early and return the
- * number of bytes not copied in r2.
- * Strategy : for large blocks, call a real memcpy function which can
- * move >1 byte at a time using unaligned ld/st instructions, and can
- * manipulate the cache using prefetch + alloco to improve the speed
- * further.  If a fault occurs in that function, just revert to the
- * byte-by-byte approach used for small blocks; this is rare so the
- * performance hit for that case does not matter.
- *
- * For small blocks it's not worth the overhead of setting up and calling
- * the memcpy routine; do the copy a byte at a time.
- *
- */
-	.global	__copy_user
-__copy_user:
-	pta	__copy_user_byte_by_byte, tr1
-	movi	16, r0 ! this value is a best guess, should tune it by benchmarking
-	bge/u	r0, r4, tr1
-	pta copy_user_memcpy, tr0
-	addi	SP, -32, SP
-	/* Save arguments in case we have to fix-up unhandled page fault */
-	st.q	SP, 0, r2
-	st.q	SP, 8, r3
-	st.q	SP, 16, r4
-	st.q	SP, 24, r35 ! r35 is callee-save
-	/* Save LINK in a register to reduce RTS time later (otherwise
-	   ld SP,*,LINK;ptabs LINK;trn;blink trn,r63 becomes a critical path) */
-	ori	LINK, 0, r35
-	blink	tr0, LINK
-
-	/* Copy completed normally if we get back here */
-	ptabs	r35, tr0
-	ld.q	SP, 24, r35
-	/* don't restore r2-r4, pointless */
-	/* set result=r2 to zero as the copy must have succeeded. */
-	or	r63, r63, r2
-	addi	SP, 32, SP
-	blink	tr0, r63 ! RTS
-
-	.global __copy_user_fixup
-__copy_user_fixup:
-	/* Restore stack frame */
-	ori	r35, 0, LINK
-	ld.q	SP, 24, r35
-	ld.q	SP, 16, r4
-	ld.q	SP,  8, r3
-	ld.q	SP,  0, r2
-	addi	SP, 32, SP
-	/* Fall through to original code, in the 'same' state we entered with */
-
-/* The slow byte-by-byte method is used if the fast copy traps due to a bad
-   user address.  In that rare case, the speed drop can be tolerated. */
-__copy_user_byte_by_byte:
-	pta	___copy_user_exit, tr1
-	pta	___copy_user1, tr0
-	beq/u	r4, r63, tr1	/* early exit for zero length copy */
-	sub	r2, r3, r0
-	addi	r0, -1, r0
-
-___copy_user1:
-	ld.b	r3, 0, r5		/* Fault address 1 */
-
-	/* Could rewrite this to use just 1 add, but the second comes 'free'
-	   due to load latency */
-	addi	r3, 1, r3
-	addi	r4, -1, r4		/* No real fixup required */
-___copy_user2:
-	stx.b	r3, r0, r5		/* Fault address 2 */
-	bne     r4, ZERO, tr0
-
-___copy_user_exit:
-	or	r4, ZERO, r2
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-/*
- * __kernel_size_t __clear_user(void *addr, __kernel_size_t size)
- *
- * Inputs:
- * (r2)  target address
- * (r3)  size in bytes
- *
- * Ouputs:
- * (*r2) zero-ed target data
- * (r2)  non-zero-ed bytes
- */
-	.global	__clear_user
-__clear_user:
-	pta	___clear_user_exit, tr1
-	pta	___clear_user1, tr0
-	beq/u	r3, r63, tr1
-
-___clear_user1:
-	st.b	r2, 0, ZERO		/* Fault address */
-	addi	r2, 1, r2
-	addi	r3, -1, r3		/* No real fixup required */
-	bne     r3, ZERO, tr0
-
-___clear_user_exit:
-	or	r3, ZERO, r2
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-/*
- * int __strncpy_from_user(unsigned long __dest, unsigned long __src,
- *			   int __count)
- *
- * Inputs:
- * (r2)  target address
- * (r3)  source address
- * (r4)  maximum size in bytes
- *
- * Ouputs:
- * (*r2) copied data
- * (r2)  -EFAULT (in case of faulting)
- *       copied data (otherwise)
- */
-	.global	__strncpy_from_user
-__strncpy_from_user:
-	pta	___strncpy_from_user1, tr0
-	pta	___strncpy_from_user_done, tr1
-	or	r4, ZERO, r5		/* r5 = original count */
-	beq/u	r4, r63, tr1		/* early exit if r4==0 */
-	movi	-(EFAULT), r6		/* r6 = reply, no real fixup */
-	or	ZERO, ZERO, r7		/* r7 = data, clear top byte of data */
-
-___strncpy_from_user1:
-	ld.b	r3, 0, r7		/* Fault address: only in reading */
-	st.b	r2, 0, r7
-	addi	r2, 1, r2
-	addi	r3, 1, r3
-	beq/u	ZERO, r7, tr1
-	addi	r4, -1, r4		/* return real number of copied bytes */
-	bne/l	ZERO, r4, tr0
-
-___strncpy_from_user_done:
-	sub	r5, r4, r6		/* If done, return copied */
-
-___strncpy_from_user_exit:
-	or	r6, ZERO, r2
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-/*
- * extern long __strnlen_user(const char *__s, long __n)
- *
- * Inputs:
- * (r2)  source address
- * (r3)  source size in bytes
- *
- * Ouputs:
- * (r2)  -EFAULT (in case of faulting)
- *       string length (otherwise)
- */
-	.global	__strnlen_user
-__strnlen_user:
-	pta	___strnlen_user_set_reply, tr0
-	pta	___strnlen_user1, tr1
-	or	ZERO, ZERO, r5		/* r5 = counter */
-	movi	-(EFAULT), r6		/* r6 = reply, no real fixup */
-	or	ZERO, ZERO, r7		/* r7 = data, clear top byte of data */
-	beq	r3, ZERO, tr0
-
-___strnlen_user1:
-	ldx.b	r2, r5, r7		/* Fault address: only in reading */
-	addi	r3, -1, r3		/* No real fixup */
-	addi	r5, 1, r5
-	beq	r3, ZERO, tr0
-	bne	r7, ZERO, tr1
-! The line below used to be active.  This meant led to a junk byte lying between each pair
-! of entries in the argv & envp structures in memory.  Whilst the program saw the right data
-! via the argv and envp arguments to main, it meant the 'flat' representation visible through
-! /proc/$pid/cmdline was corrupt, causing trouble with ps, for example.
-!	addi	r5, 1, r5		/* Include '\0' */
-
-___strnlen_user_set_reply:
-	or	r5, ZERO, r6		/* If done, return counter */
-
-___strnlen_user_exit:
-	or	r6, ZERO, r2
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-/*
- * extern long __get_user_asm_?(void *val, long addr)
- *
- * Inputs:
- * (r2)  dest address
- * (r3)  source address (in User Space)
- *
- * Ouputs:
- * (r2)  -EFAULT (faulting)
- *       0 	 (not faulting)
- */
-	.global	__get_user_asm_b
-__get_user_asm_b:
-	or	r2, ZERO, r4
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___get_user_asm_b1:
-	ld.b	r3, 0, r5		/* r5 = data */
-	st.b	r4, 0, r5
-	or	ZERO, ZERO, r2
-
-___get_user_asm_b_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__get_user_asm_w
-__get_user_asm_w:
-	or	r2, ZERO, r4
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___get_user_asm_w1:
-	ld.w	r3, 0, r5		/* r5 = data */
-	st.w	r4, 0, r5
-	or	ZERO, ZERO, r2
-
-___get_user_asm_w_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__get_user_asm_l
-__get_user_asm_l:
-	or	r2, ZERO, r4
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___get_user_asm_l1:
-	ld.l	r3, 0, r5		/* r5 = data */
-	st.l	r4, 0, r5
-	or	ZERO, ZERO, r2
-
-___get_user_asm_l_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__get_user_asm_q
-__get_user_asm_q:
-	or	r2, ZERO, r4
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___get_user_asm_q1:
-	ld.q	r3, 0, r5		/* r5 = data */
-	st.q	r4, 0, r5
-	or	ZERO, ZERO, r2
-
-___get_user_asm_q_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-/*
- * extern long __put_user_asm_?(void *pval, long addr)
- *
- * Inputs:
- * (r2)  kernel pointer to value
- * (r3)  dest address (in User Space)
- *
- * Ouputs:
- * (r2)  -EFAULT (faulting)
- *       0 	 (not faulting)
- */
-	.global	__put_user_asm_b
-__put_user_asm_b:
-	ld.b	r2, 0, r4		/* r4 = data */
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___put_user_asm_b1:
-	st.b	r3, 0, r4
-	or	ZERO, ZERO, r2
-
-___put_user_asm_b_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__put_user_asm_w
-__put_user_asm_w:
-	ld.w	r2, 0, r4		/* r4 = data */
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___put_user_asm_w1:
-	st.w	r3, 0, r4
-	or	ZERO, ZERO, r2
-
-___put_user_asm_w_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__put_user_asm_l
-__put_user_asm_l:
-	ld.l	r2, 0, r4		/* r4 = data */
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___put_user_asm_l1:
-	st.l	r3, 0, r4
-	or	ZERO, ZERO, r2
-
-___put_user_asm_l_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-
-	.global	__put_user_asm_q
-__put_user_asm_q:
-	ld.q	r2, 0, r4		/* r4 = data */
-	movi	-(EFAULT), r2		/* r2 = reply, no real fixup */
-
-___put_user_asm_q1:
-	st.q	r3, 0, r4
-	or	ZERO, ZERO, r2
-
-___put_user_asm_q_exit:
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
-panic_stash_regs:
-	/* The idea is : when we get an unhandled panic, we dump the registers
-	   to a known memory location, the just sit in a tight loop.
-	   This allows the human to look at the memory region through the GDB
-	   session (assuming the debug module's SHwy initiator isn't locked up
-	   or anything), to hopefully analyze the cause of the panic. */
-
-	/* On entry, former r15 (SP) is in DCR
-	   former r0  is at resvec_saved_area + 0
-	   former r1  is at resvec_saved_area + 8
-	   former tr0 is at resvec_saved_area + 32
-	   DCR is the only register whose value is lost altogether.
-	*/
-
-	movi	0xffffffff80000000, r0 ! phy of dump area
-	ld.q	SP, 0x000, r1	! former r0
-	st.q	r0,  0x000, r1
-	ld.q	SP, 0x008, r1	! former r1
-	st.q	r0,  0x008, r1
-	st.q	r0,  0x010, r2
-	st.q	r0,  0x018, r3
-	st.q	r0,  0x020, r4
-	st.q	r0,  0x028, r5
-	st.q	r0,  0x030, r6
-	st.q	r0,  0x038, r7
-	st.q	r0,  0x040, r8
-	st.q	r0,  0x048, r9
-	st.q	r0,  0x050, r10
-	st.q	r0,  0x058, r11
-	st.q	r0,  0x060, r12
-	st.q	r0,  0x068, r13
-	st.q	r0,  0x070, r14
-	getcon	dcr, r14
-	st.q	r0,  0x078, r14
-	st.q	r0,  0x080, r16
-	st.q	r0,  0x088, r17
-	st.q	r0,  0x090, r18
-	st.q	r0,  0x098, r19
-	st.q	r0,  0x0a0, r20
-	st.q	r0,  0x0a8, r21
-	st.q	r0,  0x0b0, r22
-	st.q	r0,  0x0b8, r23
-	st.q	r0,  0x0c0, r24
-	st.q	r0,  0x0c8, r25
-	st.q	r0,  0x0d0, r26
-	st.q	r0,  0x0d8, r27
-	st.q	r0,  0x0e0, r28
-	st.q	r0,  0x0e8, r29
-	st.q	r0,  0x0f0, r30
-	st.q	r0,  0x0f8, r31
-	st.q	r0,  0x100, r32
-	st.q	r0,  0x108, r33
-	st.q	r0,  0x110, r34
-	st.q	r0,  0x118, r35
-	st.q	r0,  0x120, r36
-	st.q	r0,  0x128, r37
-	st.q	r0,  0x130, r38
-	st.q	r0,  0x138, r39
-	st.q	r0,  0x140, r40
-	st.q	r0,  0x148, r41
-	st.q	r0,  0x150, r42
-	st.q	r0,  0x158, r43
-	st.q	r0,  0x160, r44
-	st.q	r0,  0x168, r45
-	st.q	r0,  0x170, r46
-	st.q	r0,  0x178, r47
-	st.q	r0,  0x180, r48
-	st.q	r0,  0x188, r49
-	st.q	r0,  0x190, r50
-	st.q	r0,  0x198, r51
-	st.q	r0,  0x1a0, r52
-	st.q	r0,  0x1a8, r53
-	st.q	r0,  0x1b0, r54
-	st.q	r0,  0x1b8, r55
-	st.q	r0,  0x1c0, r56
-	st.q	r0,  0x1c8, r57
-	st.q	r0,  0x1d0, r58
-	st.q	r0,  0x1d8, r59
-	st.q	r0,  0x1e0, r60
-	st.q	r0,  0x1e8, r61
-	st.q	r0,  0x1f0, r62
-	st.q	r0,  0x1f8, r63	! bogus, but for consistency's sake...
-
-	ld.q	SP, 0x020, r1  ! former tr0
-	st.q	r0,  0x200, r1
-	gettr	tr1, r1
-	st.q	r0,  0x208, r1
-	gettr	tr2, r1
-	st.q	r0,  0x210, r1
-	gettr	tr3, r1
-	st.q	r0,  0x218, r1
-	gettr	tr4, r1
-	st.q	r0,  0x220, r1
-	gettr	tr5, r1
-	st.q	r0,  0x228, r1
-	gettr	tr6, r1
-	st.q	r0,  0x230, r1
-	gettr	tr7, r1
-	st.q	r0,  0x238, r1
-
-	getcon	sr,  r1
-	getcon	ssr,  r2
-	getcon	pssr,  r3
-	getcon	spc,  r4
-	getcon	pspc,  r5
-	getcon	intevt,  r6
-	getcon	expevt,  r7
-	getcon	pexpevt,  r8
-	getcon	tra,  r9
-	getcon	tea,  r10
-	getcon	kcr0, r11
-	getcon	kcr1, r12
-	getcon	vbr,  r13
-	getcon	resvec,  r14
-
-	st.q	r0,  0x240, r1
-	st.q	r0,  0x248, r2
-	st.q	r0,  0x250, r3
-	st.q	r0,  0x258, r4
-	st.q	r0,  0x260, r5
-	st.q	r0,  0x268, r6
-	st.q	r0,  0x270, r7
-	st.q	r0,  0x278, r8
-	st.q	r0,  0x280, r9
-	st.q	r0,  0x288, r10
-	st.q	r0,  0x290, r11
-	st.q	r0,  0x298, r12
-	st.q	r0,  0x2a0, r13
-	st.q	r0,  0x2a8, r14
-
-	getcon	SPC,r2
-	getcon	SSR,r3
-	getcon	EXPEVT,r4
-	/* Prepare to jump to C - physical address */
-	movi	panic_handler-CONFIG_CACHED_MEMORY_OFFSET, r1
-	ori	r1, 1, r1
-	ptabs   r1, tr0
-	getcon	DCR, SP
-	blink	tr0, ZERO
-	nop
-	nop
-	nop
-	nop
-
-
-
-
-/*
- * --- Signal Handling Section
- */
-
-/*
- * extern long long _sa_default_rt_restorer
- * extern long long _sa_default_restorer
- *
- *		 or, better,
- *
- * extern void _sa_default_rt_restorer(void)
- * extern void _sa_default_restorer(void)
- *
- * Code prototypes to do a sys_rt_sigreturn() or sys_sysreturn()
- * from user space. Copied into user space by signal management.
- * Both must be quad aligned and 2 quad long (4 instructions).
- *
- */
-	.balign 8
-	.global sa_default_rt_restorer
-sa_default_rt_restorer:
-	movi	0x10, r9
-	shori	__NR_rt_sigreturn, r9
-	trapa	r9
-	nop
-
-	.balign 8
-	.global sa_default_restorer
-sa_default_restorer:
-	movi	0x10, r9
-	shori	__NR_sigreturn, r9
-	trapa	r9
-	nop
-
-/*
- * --- __ex_table Section
- */
-
-/*
- * User Access Exception Table.
- */
-	.section	__ex_table,  "a"
-
-	.global asm_uaccess_start	/* Just a marker */
-asm_uaccess_start:
-
-	.long	___copy_user1, ___copy_user_exit
-	.long	___copy_user2, ___copy_user_exit
-	.long	___clear_user1, ___clear_user_exit
-	.long	___strncpy_from_user1, ___strncpy_from_user_exit
-	.long	___strnlen_user1, ___strnlen_user_exit
-	.long	___get_user_asm_b1, ___get_user_asm_b_exit
-	.long	___get_user_asm_w1, ___get_user_asm_w_exit
-	.long	___get_user_asm_l1, ___get_user_asm_l_exit
-	.long	___get_user_asm_q1, ___get_user_asm_q_exit
-	.long	___put_user_asm_b1, ___put_user_asm_b_exit
-	.long	___put_user_asm_w1, ___put_user_asm_w_exit
-	.long	___put_user_asm_l1, ___put_user_asm_l_exit
-	.long	___put_user_asm_q1, ___put_user_asm_q_exit
-
-	.global asm_uaccess_end		/* Just a marker */
-asm_uaccess_end:
-
-
-
-
-/*
- * --- .text.init Section
- */
-
-	.section	.text.init, "ax"
-
-/*
- * void trap_init (void)
- *
- */
-	.global	trap_init
-trap_init:
-	addi	SP, -24, SP			/* Room to save r28/r29/r30 */
-	st.q	SP, 0, r28
-	st.q	SP, 8, r29
-	st.q	SP, 16, r30
-
-	/* Set VBR and RESVEC */
-	movi	LVBR_block, r19
-	andi	r19, -4, r19			/* reset MMUOFF + reserved */
-	/* For RESVEC exceptions we force the MMU off, which means we need the
-	   physical address. */
-	movi	LRESVEC_block-CONFIG_CACHED_MEMORY_OFFSET, r20
-	andi	r20, -4, r20			/* reset reserved */
-	ori	r20, 1, r20			/* set MMUOFF */
-	putcon	r19, VBR
-	putcon	r20, RESVEC
-
-	/* Sanity check */
-	movi	LVBR_block_end, r21
-	andi	r21, -4, r21
-	movi	BLOCK_SIZE, r29			/* r29 = expected size */
-	or	r19, ZERO, r30
-	add	r19, r29, r19
-
-	/*
-	 * Ugly, but better loop forever now than crash afterwards.
-	 * We should print a message, but if we touch LVBR or
-	 * LRESVEC blocks we should not be surprised if we get stuck
-	 * in trap_init().
-	 */
-	pta	trap_init_loop, tr1
-	gettr	tr1, r28			/* r28 = trap_init_loop */
-	sub	r21, r30, r30			/* r30 = actual size */
-
-	/*
-	 * VBR/RESVEC handlers overlap by being bigger than
-	 * allowed. Very bad. Just loop forever.
-	 * (r28) panic/loop address
-	 * (r29) expected size
-	 * (r30) actual size
-	 */
-trap_init_loop:
-	bne	r19, r21, tr1
-
-	/* Now that exception vectors are set up reset SR.BL */
-	getcon 	SR, r22
-	movi	SR_UNBLOCK_EXC, r23
-	and	r22, r23, r22
-	putcon	r22, SR
-
-	addi	SP, 24, SP
-	ptabs	LINK, tr0
-	blink	tr0, ZERO
-
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index 76c4dc7021cb..f3bd82e95898 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -3,6 +3,8 @@
 
 #include <asm/cpu-features.h>
 
+#ifndef __ASSEMBLY__
+
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
  *
@@ -40,11 +42,8 @@ enum cpu_type {
 	CPU_SH_NONE
 };
 
-#ifdef CONFIG_SUPERH32
-# include "processor_32.h"
-#else
-# include "processor_64.h"
-#endif
+/* Forward decl */
+struct sh_cpuinfo;
 
 /* arch/sh/kernel/setup.c */
 const char *get_cpu_subtype(struct sh_cpuinfo *c);
@@ -55,4 +54,12 @@ int vsyscall_init(void);
 #define vsyscall_init() do { } while (0)
 #endif
 
+#endif /* __ASSEMBLY__ */
+
+#ifdef CONFIG_SUPERH32
+# include "processor_32.h"
+#else
+# include "processor_64.h"
+#endif
+
 #endif /* __ASM_SH_PROCESSOR_H */
-- 
cgit v1.2.3


From 249cfea914002baac0af4b080306e6b820cd86b2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 19 Nov 2007 18:26:19 +0900
Subject: sh: Split out pgtable.h in to _32 and _64 variants.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/ioremap_64.c     |  40 ++--
 include/asm-sh/page.h       |  10 +-
 include/asm-sh/pgtable.h    | 491 +------------------------------------------
 include/asm-sh/pgtable_32.h | 473 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/pgtable_64.h | 300 +++++++++++++++++++++++++++
 include/asm-sh64/pgtable.h  | 496 --------------------------------------------
 6 files changed, 820 insertions(+), 990 deletions(-)
 create mode 100644 include/asm-sh/pgtable_32.h
 create mode 100644 include/asm-sh/pgtable_64.h
 delete mode 100644 include/asm-sh64/pgtable.h

(limited to 'include')

diff --git a/arch/sh/mm/ioremap_64.c b/arch/sh/mm/ioremap_64.c
index 95462a0f3120..e27d16519235 100644
--- a/arch/sh/mm/ioremap_64.c
+++ b/arch/sh/mm/ioremap_64.c
@@ -1,23 +1,24 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/mm/ioremap.c
+ * arch/sh/mm/ioremap_64.c
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
+ * Copyright (C) 2003 - 2007  Paul Mundt
  *
  * Mostly derived from arch/sh/mm/ioremap.c which, in turn is mostly
  * derived from arch/i386/mm/ioremap.c .
  *
  *   (C) Copyright 1995 1996 Linus Torvalds
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/vmalloc.h>
 #include <linux/ioport.h>
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/io.h>
+#include <linux/bootmem.h>
 #include <linux/proc_fs.h>
 #include <asm/page.h>
 #include <asm/pgalloc.h>
@@ -42,7 +43,8 @@ static unsigned long shmedia_ioremap(struct resource *, u32, int);
  * have to convert them into an offset in a page-aligned mapping, but the
  * caller shouldn't need to know that small detail.
  */
-void * __ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags)
+void *__ioremap(unsigned long phys_addr, unsigned long size,
+		unsigned long flags)
 {
 	void * addr;
 	struct vm_struct * area;
@@ -83,7 +85,7 @@ void * __ioremap(unsigned long phys_addr, unsigned long size, unsigned long flag
 }
 EXPORT_SYMBOL(__ioremap);
 
-void iounmap(void *addr)
+void __iounmap(void *addr)
 {
 	struct vm_struct *area;
 
@@ -96,7 +98,7 @@ void iounmap(void *addr)
 
 	kfree(area);
 }
-EXPORT_SYMBOL(iounmap);
+EXPORT_SYMBOL(__iounmap);
 
 static struct resource shmedia_iomap = {
 	.name	= "shmedia_iomap",
@@ -265,6 +267,7 @@ static __init_refok void *sh64_get_page(void)
 static void shmedia_mapioaddr(unsigned long pa, unsigned long va)
 {
 	pgd_t *pgdp;
+	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep, pte;
 	pgprot_t prot;
@@ -274,11 +277,17 @@ static void shmedia_mapioaddr(unsigned long pa, unsigned long va)
 
 	pgdp = pgd_offset_k(va);
 	if (pgd_none(*pgdp) || !pgd_present(*pgdp)) {
+		pudp = (pud_t *)sh64_get_page();
+		set_pgd(pgdp, __pgd((unsigned long)pudp | _KERNPG_TABLE));
+	}
+
+	pudp = pud_offset(pgdp, va);
+	if (pud_none(*pudp) || !pud_present(*pudp)) {
 		pmdp = (pmd_t *)sh64_get_page();
-		set_pgd(pgdp, __pgd((unsigned long)pmdp | _KERNPG_TABLE));
+		set_pud(pudp, __pud((unsigned long)pmdp | _KERNPG_TABLE));
 	}
 
-	pmdp = pmd_offset(pgdp, va);
+	pmdp = pmd_offset(pudp, va);
 	if (pmd_none(*pmdp) || !pmd_present(*pmdp) ) {
 		ptep = (pte_t *)sh64_get_page();
 		set_pmd(pmdp, __pmd((unsigned long)ptep + _PAGE_TABLE));
@@ -302,12 +311,19 @@ static void shmedia_mapioaddr(unsigned long pa, unsigned long va)
 static void shmedia_unmapioaddr(unsigned long vaddr)
 {
 	pgd_t *pgdp;
+	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;
 
 	pgdp = pgd_offset_k(vaddr);
-	pmdp = pmd_offset(pgdp, vaddr);
+	if (pgd_none(*pgdp) || pgd_bad(*pgdp))
+		return;
+
+	pudp = pud_offset(pgdp, vaddr);
+	if (pud_none(*pudp) || pud_bad(*pudp))
+		return;
 
+	pmdp = pmd_offset(pudp, vaddr);
 	if (pmd_none(*pmdp) || pmd_bad(*pmdp))
 		return;
 
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index d0273dbce6be..93a89841227f 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -96,12 +96,18 @@ typedef struct { unsigned long long pgd; } pgd_t;
 	((x).pte_low | ((unsigned long long)(x).pte_high << 32))
 #define __pte(x) \
 	({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; })
-#else
+#elif defined(CONFIG_SUPERH32)
 typedef struct { unsigned long pte_low; } pte_t;
 typedef struct { unsigned long pgprot; } pgprot_t;
 typedef struct { unsigned long pgd; } pgd_t;
 #define pte_val(x)	((x).pte_low)
-#define __pte(x) ((pte_t) { (x) } )
+#define __pte(x)	((pte_t) { (x) } )
+#else
+typedef struct { unsigned long long pte_low; } pte_t;
+typedef struct { unsigned long pgprot; } pgprot_t;
+typedef struct { unsigned long pgd; } pgd_t;
+#define pte_val(x)	((x).pte_low)
+#define __pte(x)	((pte_t) { (x) } )
 #endif
 
 #define pgd_val(x)	((x).pgd)
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index b0bb76a6864f..5a800c69e049 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -3,7 +3,7 @@
  * use the SuperH page table tree.
  *
  * Copyright (C) 1999 Niibe Yutaka
- * Copyright (C) 2002 - 2005 Paul Mundt
+ * Copyright (C) 2002 - 2007 Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General
  * Public License.  See the file "COPYING" in the main directory of this
@@ -78,278 +78,12 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #endif
 #define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
 
-/*
- * Linux PTEL encoding.
- *
- * Hardware and software bit definitions for the PTEL value (see below for
- * notes on SH-X2 MMUs and 64-bit PTEs):
- *
- * - Bits 0 and 7 are reserved on SH-3 (_PAGE_WT and _PAGE_SZ1 on SH-4).
- *
- * - Bit 1 is the SH-bit, but is unused on SH-3 due to an MMU bug (the
- *   hardware PTEL value can't have the SH-bit set when MMUCR.IX is set,
- *   which is the default in cpu-sh3/mmu_context.h:MMU_CONTROL_INIT).
- *
- *   In order to keep this relatively clean, do not use these for defining
- *   SH-3 specific flags until all of the other unused bits have been
- *   exhausted.
- *
- * - Bit 9 is reserved by everyone and used by _PAGE_PROTNONE.
- *
- * - Bits 10 and 11 are low bits of the PPN that are reserved on >= 4K pages.
- *   Bit 10 is used for _PAGE_ACCESSED, bit 11 remains unused.
- *
- * - Bits 31, 30, and 29 remain unused by everyone and can be used for future
- *   software flags, although care must be taken to update _PAGE_CLEAR_FLAGS.
- *
- * XXX: Leave the _PAGE_FILE and _PAGE_WT overhaul for a rainy day.
- *
- * SH-X2 MMUs and extended PTEs
- *
- * SH-X2 supports an extended mode TLB with split data arrays due to the
- * number of bits needed for PR and SZ (now EPR and ESZ) encodings. The PR and
- * SZ bit placeholders still exist in data array 1, but are implemented as
- * reserved bits, with the real logic existing in data array 2.
- *
- * The downside to this is that we can no longer fit everything in to a 32-bit
- * PTE encoding, so a 64-bit pte_t is necessary for these parts. On the plus
- * side, this gives us quite a few spare bits to play with for future usage.
- */
-/* Legacy and compat mode bits */
-#define	_PAGE_WT	0x001		/* WT-bit on SH-4, 0 on SH-3 */
-#define _PAGE_HW_SHARED	0x002		/* SH-bit  : shared among processes */
-#define _PAGE_DIRTY	0x004		/* D-bit   : page changed */
-#define _PAGE_CACHABLE	0x008		/* C-bit   : cachable */
-#define _PAGE_SZ0	0x010		/* SZ0-bit : Size of page */
-#define _PAGE_RW	0x020		/* PR0-bit : write access allowed */
-#define _PAGE_USER	0x040		/* PR1-bit : user space access allowed*/
-#define _PAGE_SZ1	0x080		/* SZ1-bit : Size of page (on SH-4) */
-#define _PAGE_PRESENT	0x100		/* V-bit   : page is valid */
-#define _PAGE_PROTNONE	0x200		/* software: if not present  */
-#define _PAGE_ACCESSED	0x400		/* software: page referenced */
-#define _PAGE_FILE	_PAGE_WT	/* software: pagecache or swap? */
-
-#define _PAGE_SZ_MASK	(_PAGE_SZ0 | _PAGE_SZ1)
-#define _PAGE_PR_MASK	(_PAGE_RW | _PAGE_USER)
-
-/* Extended mode bits */
-#define _PAGE_EXT_ESZ0		0x0010	/* ESZ0-bit: Size of page */
-#define _PAGE_EXT_ESZ1		0x0020	/* ESZ1-bit: Size of page */
-#define _PAGE_EXT_ESZ2		0x0040	/* ESZ2-bit: Size of page */
-#define _PAGE_EXT_ESZ3		0x0080	/* ESZ3-bit: Size of page */
-
-#define _PAGE_EXT_USER_EXEC	0x0100	/* EPR0-bit: User space executable */
-#define _PAGE_EXT_USER_WRITE	0x0200	/* EPR1-bit: User space writable */
-#define _PAGE_EXT_USER_READ	0x0400	/* EPR2-bit: User space readable */
-
-#define _PAGE_EXT_KERN_EXEC	0x0800	/* EPR3-bit: Kernel space executable */
-#define _PAGE_EXT_KERN_WRITE	0x1000	/* EPR4-bit: Kernel space writable */
-#define _PAGE_EXT_KERN_READ	0x2000	/* EPR5-bit: Kernel space readable */
-
-/* Wrapper for extended mode pgprot twiddling */
-#define _PAGE_EXT(x)		((unsigned long long)(x) << 32)
-
-/* software: moves to PTEA.TC (Timing Control) */
-#define _PAGE_PCC_AREA5	0x00000000	/* use BSC registers for area5 */
-#define _PAGE_PCC_AREA6	0x80000000	/* use BSC registers for area6 */
-
-/* software: moves to PTEA.SA[2:0] (Space Attributes) */
-#define _PAGE_PCC_IODYN 0x00000001	/* IO space, dynamically sized bus */
-#define _PAGE_PCC_IO8	0x20000000	/* IO space, 8 bit bus */
-#define _PAGE_PCC_IO16	0x20000001	/* IO space, 16 bit bus */
-#define _PAGE_PCC_COM8	0x40000000	/* Common Memory space, 8 bit bus */
-#define _PAGE_PCC_COM16	0x40000001	/* Common Memory space, 16 bit bus */
-#define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
-#define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
-
-/* Mask which drops unused bits from the PTEL value */
-#if defined(CONFIG_CPU_SH3)
-#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED| \
-				 _PAGE_FILE	| _PAGE_SZ1	| \
-				 _PAGE_HW_SHARED)
-#elif defined(CONFIG_X2TLB)
-/* Get rid of the legacy PR/SZ bits when using extended mode */
-#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | \
-				 _PAGE_FILE | _PAGE_PR_MASK | _PAGE_SZ_MASK)
-#else
-#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_FILE)
-#endif
-
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
-
-/* Hardware flags, page size encoding */
-#if defined(CONFIG_X2TLB)
-# if defined(CONFIG_PAGE_SIZE_4KB)
-#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ0)
-# elif defined(CONFIG_PAGE_SIZE_8KB)
-#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ1)
-# elif defined(CONFIG_PAGE_SIZE_64KB)
-#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ2)
-# endif
-#else
-# if defined(CONFIG_PAGE_SIZE_4KB)
-#  define _PAGE_FLAGS_HARD	_PAGE_SZ0
-# elif defined(CONFIG_PAGE_SIZE_64KB)
-#  define _PAGE_FLAGS_HARD	_PAGE_SZ1
-# endif
-#endif
-
-#if defined(CONFIG_X2TLB)
-# if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
-#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ2)
-# elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
-#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ0 | _PAGE_EXT_ESZ2)
-# elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
-#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ0 | _PAGE_EXT_ESZ1 | _PAGE_EXT_ESZ2)
-# elif defined(CONFIG_HUGETLB_PAGE_SIZE_4MB)
-#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ3)
-# elif defined(CONFIG_HUGETLB_PAGE_SIZE_64MB)
-#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ2 | _PAGE_EXT_ESZ3)
-# endif
+#if defined(CONFIG_SUPERH32)
+#include <asm/pgtable_32.h>
 #else
-# if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
-#  define _PAGE_SZHUGE	(_PAGE_SZ1)
-# elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
-#  define _PAGE_SZHUGE	(_PAGE_SZ0 | _PAGE_SZ1)
-# endif
-#endif
-
-/*
- * Stub out _PAGE_SZHUGE if we don't have a good definition for it,
- * to make pte_mkhuge() happy.
- */
-#ifndef _PAGE_SZHUGE
-# define _PAGE_SZHUGE	(_PAGE_FLAGS_HARD)
-#endif
-
-#define _PAGE_CHG_MASK \
-	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY)
-
-#ifndef __ASSEMBLY__
-
-#if defined(CONFIG_X2TLB) /* SH-X2 TLB */
-#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
-				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ  | \
-					   _PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_USER_READ  | \
-					   _PAGE_EXT_USER_WRITE))
-
-#define PAGE_EXECREAD	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
-				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_EXEC | \
-					   _PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_USER_EXEC | \
-					   _PAGE_EXT_USER_READ))
-
-#define PAGE_COPY	PAGE_EXECREAD
-
-#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
-				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_USER_READ))
-
-#define PAGE_WRITEONLY	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
-				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_USER_WRITE))
-
-#define PAGE_RWX	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
-				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_KERN_READ  | \
-					   _PAGE_EXT_KERN_EXEC  | \
-					   _PAGE_EXT_USER_WRITE | \
-					   _PAGE_EXT_USER_READ  | \
-					   _PAGE_EXT_USER_EXEC))
-
-#define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
-				 _PAGE_DIRTY | _PAGE_ACCESSED | \
-				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_KERN_EXEC))
-
-#define PAGE_KERNEL_NOCACHE \
-			__pgprot(_PAGE_PRESENT | _PAGE_DIRTY | \
-				 _PAGE_ACCESSED | _PAGE_HW_SHARED | \
-				 _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_KERN_EXEC))
-
-#define PAGE_KERNEL_RO	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
-				 _PAGE_DIRTY | _PAGE_ACCESSED | \
-				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_KERN_EXEC))
-
-#define PAGE_KERNEL_PCC(slot, type) \
-			__pgprot(_PAGE_PRESENT | _PAGE_DIRTY | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD | \
-				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
-					   _PAGE_EXT_KERN_WRITE | \
-					   _PAGE_EXT_KERN_EXEC) \
-				 (slot ? _PAGE_PCC_AREA5 : _PAGE_PCC_AREA6) | \
-				 (type))
-
-#elif defined(CONFIG_MMU) /* SH-X TLB */
-#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | \
-				 _PAGE_CACHABLE | _PAGE_ACCESSED | \
-				 _PAGE_FLAGS_HARD)
-
-#define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-
-#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-
-#define PAGE_EXECREAD	PAGE_READONLY
-#define PAGE_RWX	PAGE_SHARED
-#define PAGE_WRITEONLY	PAGE_SHARED
-
-#define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_CACHABLE | \
-				 _PAGE_DIRTY | _PAGE_ACCESSED | \
-				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
-
-#define PAGE_KERNEL_NOCACHE \
-			__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | \
-				 _PAGE_ACCESSED | _PAGE_HW_SHARED | \
-				 _PAGE_FLAGS_HARD)
-
-#define PAGE_KERNEL_RO	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
-				 _PAGE_DIRTY | _PAGE_ACCESSED | \
-				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
-
-#define PAGE_KERNEL_PCC(slot, type) \
-			__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | \
-				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD | \
-				 (slot ? _PAGE_PCC_AREA5 : _PAGE_PCC_AREA6) | \
-				 (type))
-#else /* no mmu */
-#define PAGE_NONE		__pgprot(0)
-#define PAGE_SHARED		__pgprot(0)
-#define PAGE_COPY		__pgprot(0)
-#define PAGE_EXECREAD		__pgprot(0)
-#define PAGE_RWX		__pgprot(0)
-#define PAGE_READONLY		__pgprot(0)
-#define PAGE_WRITEONLY		__pgprot(0)
-#define PAGE_KERNEL		__pgprot(0)
-#define PAGE_KERNEL_NOCACHE	__pgprot(0)
-#define PAGE_KERNEL_RO		__pgprot(0)
-
-#define PAGE_KERNEL_PCC(slot, type) \
-				__pgprot(0)
+#include <asm/pgtable_64.h>
 #endif
 
-#endif /* __ASSEMBLY__ */
-
 /*
  * SH-X and lower (legacy) SuperH parts (SH-3, SH-4, some SH-4A) can't do page
  * protection for execute, and considers it the same as a read. Also, write
@@ -378,208 +112,6 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define __S110	PAGE_RWX
 #define __S111	PAGE_RWX
 
-#ifndef __ASSEMBLY__
-
-/*
- * Certain architectures need to do special things when PTEs
- * within a page table are directly modified.  Thus, the following
- * hook is made available.
- */
-#ifdef CONFIG_X2TLB
-static inline void set_pte(pte_t *ptep, pte_t pte)
-{
-	ptep->pte_high = pte.pte_high;
-	smp_wmb();
-	ptep->pte_low = pte.pte_low;
-}
-#else
-#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
-#endif
-
-#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
-
-/*
- * (pmds are folded into pgds so this doesn't get actually called,
- * but the define is needed for a generic inline function.)
- */
-#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
-
-#define pte_pfn(x)		((unsigned long)(((x).pte_low >> PAGE_SHIFT)))
-
-#define pfn_pte(pfn, prot) \
-	__pte(((unsigned long long)(pfn) << PAGE_SHIFT) | pgprot_val(prot))
-#define pfn_pmd(pfn, prot) \
-	__pmd(((unsigned long long)(pfn) << PAGE_SHIFT) | pgprot_val(prot))
-
-#define pte_none(x)		(!pte_val(x))
-#define pte_present(x)		((x).pte_low & (_PAGE_PRESENT | _PAGE_PROTNONE))
-
-#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
-
-#define pmd_none(x)	(!pmd_val(x))
-#define pmd_present(x)	(pmd_val(x))
-#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
-#define	pmd_bad(x)	(pmd_val(x) & ~PAGE_MASK)
-
-#define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
-#define pte_page(x)	pfn_to_page(pte_pfn(x))
-
-/*
- * The following only work if pte_present() is true.
- * Undefined behaviour if not..
- */
-#define pte_not_present(pte)	(!((pte).pte_low & _PAGE_PRESENT))
-#define pte_dirty(pte)		((pte).pte_low & _PAGE_DIRTY)
-#define pte_young(pte)		((pte).pte_low & _PAGE_ACCESSED)
-#define pte_file(pte)		((pte).pte_low & _PAGE_FILE)
-
-#ifdef CONFIG_X2TLB
-#define pte_write(pte)		((pte).pte_high & _PAGE_EXT_USER_WRITE)
-#else
-#define pte_write(pte)		((pte).pte_low & _PAGE_RW)
-#endif
-
-#define PTE_BIT_FUNC(h,fn,op) \
-static inline pte_t pte_##fn(pte_t pte) { pte.pte_##h op; return pte; }
-
-#ifdef CONFIG_X2TLB
-/*
- * We cheat a bit in the SH-X2 TLB case. As the permission bits are
- * individually toggled (and user permissions are entirely decoupled from
- * kernel permissions), we attempt to couple them a bit more sanely here.
- */
-PTE_BIT_FUNC(high, wrprotect, &= ~_PAGE_EXT_USER_WRITE);
-PTE_BIT_FUNC(high, mkwrite, |= _PAGE_EXT_USER_WRITE | _PAGE_EXT_KERN_WRITE);
-PTE_BIT_FUNC(high, mkhuge, |= _PAGE_SZHUGE);
-#else
-PTE_BIT_FUNC(low, wrprotect, &= ~_PAGE_RW);
-PTE_BIT_FUNC(low, mkwrite, |= _PAGE_RW);
-PTE_BIT_FUNC(low, mkhuge, |= _PAGE_SZHUGE);
-#endif
-
-PTE_BIT_FUNC(low, mkclean, &= ~_PAGE_DIRTY);
-PTE_BIT_FUNC(low, mkdirty, |= _PAGE_DIRTY);
-PTE_BIT_FUNC(low, mkold, &= ~_PAGE_ACCESSED);
-PTE_BIT_FUNC(low, mkyoung, |= _PAGE_ACCESSED);
-
-/*
- * Macro and implementation to make a page protection as uncachable.
- */
-#define pgprot_writecombine(prot) \
-	__pgprot(pgprot_val(prot) & ~_PAGE_CACHABLE)
-
-#define pgprot_noncached	 pgprot_writecombine
-
-/*
- * Conversion functions: convert a page and protection to a page entry,
- * and a page entry and page directory to the page they refer to.
- *
- * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
- */
-#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
-
-static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
-{
-	pte.pte_low &= _PAGE_CHG_MASK;
-	pte.pte_low |= pgprot_val(newprot);
-
-#ifdef CONFIG_X2TLB
-	pte.pte_high |= pgprot_val(newprot) >> 32;
-#endif
-
-	return pte;
-}
-
-#define pmd_page_vaddr(pmd)	((unsigned long)pmd_val(pmd))
-#define pmd_page(pmd)		(virt_to_page(pmd_val(pmd)))
-
-/* to find an entry in a page-table-directory. */
-#define pgd_index(address)	(((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
-#define pgd_offset(mm, address)	((mm)->pgd+pgd_index(address))
-
-/* to find an entry in a kernel page-table-directory */
-#define pgd_offset_k(address)	pgd_offset(&init_mm, address)
-
-/* Find an entry in the third-level page table.. */
-#define pte_index(address)	((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
-#define pte_offset_map(dir, address)		pte_offset_kernel(dir, address)
-#define pte_offset_map_nested(dir, address)	pte_offset_kernel(dir, address)
-
-#define pte_unmap(pte)		do { } while (0)
-#define pte_unmap_nested(pte)	do { } while (0)
-
-#ifdef CONFIG_X2TLB
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %p(%08lx%08lx).\n", __FILE__, __LINE__, \
-	       &(e), (e).pte_high, (e).pte_low)
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %016llx.\n", __FILE__, __LINE__, pgd_val(e))
-#else
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
-#endif
-
-struct vm_area_struct;
-extern void update_mmu_cache(struct vm_area_struct * vma,
-			     unsigned long address, pte_t pte);
-
-/*
- * Encode and de-code a swap entry
- *
- * Constraints:
- *	_PAGE_FILE at bit 0
- *	_PAGE_PRESENT at bit 8
- *	_PAGE_PROTNONE at bit 9
- *
- * For the normal case, we encode the swap type into bits 0:7 and the
- * swap offset into bits 10:30. For the 64-bit PTE case, we keep the
- * preserved bits in the low 32-bits and use the upper 32 as the swap
- * offset (along with a 5-bit type), following the same approach as x86
- * PAE. This keeps the logic quite simple, and allows for a full 32
- * PTE_FILE_MAX_BITS, as opposed to the 29-bits we're constrained with
- * in the pte_low case.
- *
- * As is evident by the Alpha code, if we ever get a 64-bit unsigned
- * long (swp_entry_t) to match up with the 64-bit PTEs, this all becomes
- * much cleaner..
- *
- * NOTE: We should set ZEROs at the position of _PAGE_PRESENT
- *       and _PAGE_PROTNONE bits
- */
-#ifdef CONFIG_X2TLB
-#define __swp_type(x)			((x).val & 0x1f)
-#define __swp_offset(x)			((x).val >> 5)
-#define __swp_entry(type, offset)	((swp_entry_t){ (type) | (offset) << 5})
-#define __pte_to_swp_entry(pte)		((swp_entry_t){ (pte).pte_high })
-#define __swp_entry_to_pte(x)		((pte_t){ 0, (x).val })
-
-/*
- * Encode and decode a nonlinear file mapping entry
- */
-#define pte_to_pgoff(pte)		((pte).pte_high)
-#define pgoff_to_pte(off)		((pte_t) { _PAGE_FILE, (off) })
-
-#define PTE_FILE_MAX_BITS		32
-#else
-#define __swp_type(x)			((x).val & 0xff)
-#define __swp_offset(x)			((x).val >> 10)
-#define __swp_entry(type, offset)	((swp_entry_t){(type) | (offset) <<10})
-
-#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) >> 1 })
-#define __swp_entry_to_pte(x)		((pte_t) { (x).val << 1 })
-
-/*
- * Encode and decode a nonlinear file mapping entry
- */
-#define PTE_FILE_MAX_BITS	29
-#define pte_to_pgoff(pte)	(pte_val(pte) >> 1)
-#define pgoff_to_pte(off)	((pte_t) { ((off) << 1) | _PAGE_FILE })
-#endif
-
 typedef pte_t *pte_addr_t;
 
 #define kern_addr_valid(addr)	(1)
@@ -587,27 +119,26 @@ typedef pte_t *pte_addr_t;
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
-struct mm_struct;
+#define pte_pfn(x)		((unsigned long)(((x).pte_low >> PAGE_SHIFT)))
 
 /*
  * No page table caches to initialise
  */
 #define pgtable_cache_init()	do { } while (0)
 
-#ifndef CONFIG_MMU
-extern unsigned int kobjsize(const void *objp);
-#endif /* !CONFIG_MMU */
-
 #if !defined(CONFIG_CACHE_OFF) && (defined(CONFIG_CPU_SH4) || \
 	defined(CONFIG_SH7705_CACHE_32KB))
+struct mm_struct;
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-extern pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
+pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 #endif
 
+struct vm_area_struct;
+extern void update_mmu_cache(struct vm_area_struct * vma,
+			     unsigned long address, pte_t pte);
 extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
 extern void paging_init(void);
 
 #include <asm-generic/pgtable.h>
 
-#endif /* !__ASSEMBLY__ */
-#endif /* __ASM_SH_PAGE_H */
+#endif /* __ASM_SH_PGTABLE_H */
diff --git a/include/asm-sh/pgtable_32.h b/include/asm-sh/pgtable_32.h
new file mode 100644
index 000000000000..70303603e89d
--- /dev/null
+++ b/include/asm-sh/pgtable_32.h
@@ -0,0 +1,473 @@
+#ifndef __ASM_SH_PGTABLE_32_H
+#define __ASM_SH_PGTABLE_32_H
+
+/*
+ * Linux PTEL encoding.
+ *
+ * Hardware and software bit definitions for the PTEL value (see below for
+ * notes on SH-X2 MMUs and 64-bit PTEs):
+ *
+ * - Bits 0 and 7 are reserved on SH-3 (_PAGE_WT and _PAGE_SZ1 on SH-4).
+ *
+ * - Bit 1 is the SH-bit, but is unused on SH-3 due to an MMU bug (the
+ *   hardware PTEL value can't have the SH-bit set when MMUCR.IX is set,
+ *   which is the default in cpu-sh3/mmu_context.h:MMU_CONTROL_INIT).
+ *
+ *   In order to keep this relatively clean, do not use these for defining
+ *   SH-3 specific flags until all of the other unused bits have been
+ *   exhausted.
+ *
+ * - Bit 9 is reserved by everyone and used by _PAGE_PROTNONE.
+ *
+ * - Bits 10 and 11 are low bits of the PPN that are reserved on >= 4K pages.
+ *   Bit 10 is used for _PAGE_ACCESSED, bit 11 remains unused.
+ *
+ * - Bits 31, 30, and 29 remain unused by everyone and can be used for future
+ *   software flags, although care must be taken to update _PAGE_CLEAR_FLAGS.
+ *
+ * XXX: Leave the _PAGE_FILE and _PAGE_WT overhaul for a rainy day.
+ *
+ * SH-X2 MMUs and extended PTEs
+ *
+ * SH-X2 supports an extended mode TLB with split data arrays due to the
+ * number of bits needed for PR and SZ (now EPR and ESZ) encodings. The PR and
+ * SZ bit placeholders still exist in data array 1, but are implemented as
+ * reserved bits, with the real logic existing in data array 2.
+ *
+ * The downside to this is that we can no longer fit everything in to a 32-bit
+ * PTE encoding, so a 64-bit pte_t is necessary for these parts. On the plus
+ * side, this gives us quite a few spare bits to play with for future usage.
+ */
+/* Legacy and compat mode bits */
+#define	_PAGE_WT	0x001		/* WT-bit on SH-4, 0 on SH-3 */
+#define _PAGE_HW_SHARED	0x002		/* SH-bit  : shared among processes */
+#define _PAGE_DIRTY	0x004		/* D-bit   : page changed */
+#define _PAGE_CACHABLE	0x008		/* C-bit   : cachable */
+#define _PAGE_SZ0	0x010		/* SZ0-bit : Size of page */
+#define _PAGE_RW	0x020		/* PR0-bit : write access allowed */
+#define _PAGE_USER	0x040		/* PR1-bit : user space access allowed*/
+#define _PAGE_SZ1	0x080		/* SZ1-bit : Size of page (on SH-4) */
+#define _PAGE_PRESENT	0x100		/* V-bit   : page is valid */
+#define _PAGE_PROTNONE	0x200		/* software: if not present  */
+#define _PAGE_ACCESSED	0x400		/* software: page referenced */
+#define _PAGE_FILE	_PAGE_WT	/* software: pagecache or swap? */
+
+#define _PAGE_SZ_MASK	(_PAGE_SZ0 | _PAGE_SZ1)
+#define _PAGE_PR_MASK	(_PAGE_RW | _PAGE_USER)
+
+/* Extended mode bits */
+#define _PAGE_EXT_ESZ0		0x0010	/* ESZ0-bit: Size of page */
+#define _PAGE_EXT_ESZ1		0x0020	/* ESZ1-bit: Size of page */
+#define _PAGE_EXT_ESZ2		0x0040	/* ESZ2-bit: Size of page */
+#define _PAGE_EXT_ESZ3		0x0080	/* ESZ3-bit: Size of page */
+
+#define _PAGE_EXT_USER_EXEC	0x0100	/* EPR0-bit: User space executable */
+#define _PAGE_EXT_USER_WRITE	0x0200	/* EPR1-bit: User space writable */
+#define _PAGE_EXT_USER_READ	0x0400	/* EPR2-bit: User space readable */
+
+#define _PAGE_EXT_KERN_EXEC	0x0800	/* EPR3-bit: Kernel space executable */
+#define _PAGE_EXT_KERN_WRITE	0x1000	/* EPR4-bit: Kernel space writable */
+#define _PAGE_EXT_KERN_READ	0x2000	/* EPR5-bit: Kernel space readable */
+
+/* Wrapper for extended mode pgprot twiddling */
+#define _PAGE_EXT(x)		((unsigned long long)(x) << 32)
+
+/* software: moves to PTEA.TC (Timing Control) */
+#define _PAGE_PCC_AREA5	0x00000000	/* use BSC registers for area5 */
+#define _PAGE_PCC_AREA6	0x80000000	/* use BSC registers for area6 */
+
+/* software: moves to PTEA.SA[2:0] (Space Attributes) */
+#define _PAGE_PCC_IODYN 0x00000001	/* IO space, dynamically sized bus */
+#define _PAGE_PCC_IO8	0x20000000	/* IO space, 8 bit bus */
+#define _PAGE_PCC_IO16	0x20000001	/* IO space, 16 bit bus */
+#define _PAGE_PCC_COM8	0x40000000	/* Common Memory space, 8 bit bus */
+#define _PAGE_PCC_COM16	0x40000001	/* Common Memory space, 16 bit bus */
+#define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
+#define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
+
+/* Mask which drops unused bits from the PTEL value */
+#if defined(CONFIG_CPU_SH3)
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED| \
+				 _PAGE_FILE	| _PAGE_SZ1	| \
+				 _PAGE_HW_SHARED)
+#elif defined(CONFIG_X2TLB)
+/* Get rid of the legacy PR/SZ bits when using extended mode */
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | \
+				 _PAGE_FILE | _PAGE_PR_MASK | _PAGE_SZ_MASK)
+#else
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_FILE)
+#endif
+
+#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+
+/* Hardware flags, page size encoding */
+#if defined(CONFIG_X2TLB)
+# if defined(CONFIG_PAGE_SIZE_4KB)
+#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ0)
+# elif defined(CONFIG_PAGE_SIZE_8KB)
+#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ1)
+# elif defined(CONFIG_PAGE_SIZE_64KB)
+#  define _PAGE_FLAGS_HARD	_PAGE_EXT(_PAGE_EXT_ESZ2)
+# endif
+#else
+# if defined(CONFIG_PAGE_SIZE_4KB)
+#  define _PAGE_FLAGS_HARD	_PAGE_SZ0
+# elif defined(CONFIG_PAGE_SIZE_64KB)
+#  define _PAGE_FLAGS_HARD	_PAGE_SZ1
+# endif
+#endif
+
+#if defined(CONFIG_X2TLB)
+# if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
+#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ2)
+# elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
+#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ0 | _PAGE_EXT_ESZ2)
+# elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
+#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ0 | _PAGE_EXT_ESZ1 | _PAGE_EXT_ESZ2)
+# elif defined(CONFIG_HUGETLB_PAGE_SIZE_4MB)
+#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ3)
+# elif defined(CONFIG_HUGETLB_PAGE_SIZE_64MB)
+#  define _PAGE_SZHUGE	(_PAGE_EXT_ESZ2 | _PAGE_EXT_ESZ3)
+# endif
+#else
+# if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
+#  define _PAGE_SZHUGE	(_PAGE_SZ1)
+# elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
+#  define _PAGE_SZHUGE	(_PAGE_SZ0 | _PAGE_SZ1)
+# endif
+#endif
+
+/*
+ * Stub out _PAGE_SZHUGE if we don't have a good definition for it,
+ * to make pte_mkhuge() happy.
+ */
+#ifndef _PAGE_SZHUGE
+# define _PAGE_SZHUGE	(_PAGE_FLAGS_HARD)
+#endif
+
+#define _PAGE_CHG_MASK \
+	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY)
+
+#ifndef __ASSEMBLY__
+
+#if defined(CONFIG_X2TLB) /* SH-X2 TLB */
+#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
+
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
+				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ  | \
+					   _PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_USER_READ  | \
+					   _PAGE_EXT_USER_WRITE))
+
+#define PAGE_EXECREAD	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
+				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_EXEC | \
+					   _PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_USER_EXEC | \
+					   _PAGE_EXT_USER_READ))
+
+#define PAGE_COPY	PAGE_EXECREAD
+
+#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
+				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_USER_READ))
+
+#define PAGE_WRITEONLY	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
+				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_USER_WRITE))
+
+#define PAGE_RWX	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED | \
+				 _PAGE_CACHABLE | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_KERN_READ  | \
+					   _PAGE_EXT_KERN_EXEC  | \
+					   _PAGE_EXT_USER_WRITE | \
+					   _PAGE_EXT_USER_READ  | \
+					   _PAGE_EXT_USER_EXEC))
+
+#define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
+				 _PAGE_DIRTY | _PAGE_ACCESSED | \
+				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_KERN_EXEC))
+
+#define PAGE_KERNEL_NOCACHE \
+			__pgprot(_PAGE_PRESENT | _PAGE_DIRTY | \
+				 _PAGE_ACCESSED | _PAGE_HW_SHARED | \
+				 _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_KERN_EXEC))
+
+#define PAGE_KERNEL_RO	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
+				 _PAGE_DIRTY | _PAGE_ACCESSED | \
+				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_KERN_EXEC))
+
+#define PAGE_KERNEL_PCC(slot, type) \
+			__pgprot(_PAGE_PRESENT | _PAGE_DIRTY | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD | \
+				 _PAGE_EXT(_PAGE_EXT_KERN_READ | \
+					   _PAGE_EXT_KERN_WRITE | \
+					   _PAGE_EXT_KERN_EXEC) \
+				 (slot ? _PAGE_PCC_AREA5 : _PAGE_PCC_AREA6) | \
+				 (type))
+
+#elif defined(CONFIG_MMU) /* SH-X TLB */
+#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
+
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | \
+				 _PAGE_CACHABLE | _PAGE_ACCESSED | \
+				 _PAGE_FLAGS_HARD)
+
+#define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
+
+#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
+
+#define PAGE_EXECREAD	PAGE_READONLY
+#define PAGE_RWX	PAGE_SHARED
+#define PAGE_WRITEONLY	PAGE_SHARED
+
+#define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_CACHABLE | \
+				 _PAGE_DIRTY | _PAGE_ACCESSED | \
+				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
+
+#define PAGE_KERNEL_NOCACHE \
+			__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | \
+				 _PAGE_ACCESSED | _PAGE_HW_SHARED | \
+				 _PAGE_FLAGS_HARD)
+
+#define PAGE_KERNEL_RO	__pgprot(_PAGE_PRESENT | _PAGE_CACHABLE | \
+				 _PAGE_DIRTY | _PAGE_ACCESSED | \
+				 _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
+
+#define PAGE_KERNEL_PCC(slot, type) \
+			__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | \
+				 _PAGE_ACCESSED | _PAGE_FLAGS_HARD | \
+				 (slot ? _PAGE_PCC_AREA5 : _PAGE_PCC_AREA6) | \
+				 (type))
+#else /* no mmu */
+#define PAGE_NONE		__pgprot(0)
+#define PAGE_SHARED		__pgprot(0)
+#define PAGE_COPY		__pgprot(0)
+#define PAGE_EXECREAD		__pgprot(0)
+#define PAGE_RWX		__pgprot(0)
+#define PAGE_READONLY		__pgprot(0)
+#define PAGE_WRITEONLY		__pgprot(0)
+#define PAGE_KERNEL		__pgprot(0)
+#define PAGE_KERNEL_NOCACHE	__pgprot(0)
+#define PAGE_KERNEL_RO		__pgprot(0)
+
+#define PAGE_KERNEL_PCC(slot, type) \
+				__pgprot(0)
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#ifndef __ASSEMBLY__
+
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+#ifdef CONFIG_X2TLB
+static inline void set_pte(pte_t *ptep, pte_t pte)
+{
+	ptep->pte_high = pte.pte_high;
+	smp_wmb();
+	ptep->pte_low = pte.pte_low;
+}
+#else
+#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
+#endif
+
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define pfn_pte(pfn, prot) \
+	__pte(((unsigned long long)(pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot) \
+	__pmd(((unsigned long long)(pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
+#define pte_none(x)		(!pte_val(x))
+#define pte_present(x)		((x).pte_low & (_PAGE_PRESENT | _PAGE_PROTNONE))
+
+#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
+
+#define pmd_none(x)	(!pmd_val(x))
+#define pmd_present(x)	(pmd_val(x))
+#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
+#define	pmd_bad(x)	(pmd_val(x) & ~PAGE_MASK)
+
+#define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
+#define pte_page(x)	pfn_to_page(pte_pfn(x))
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+#define pte_not_present(pte)	(!((pte).pte_low & _PAGE_PRESENT))
+#define pte_dirty(pte)		((pte).pte_low & _PAGE_DIRTY)
+#define pte_young(pte)		((pte).pte_low & _PAGE_ACCESSED)
+#define pte_file(pte)		((pte).pte_low & _PAGE_FILE)
+
+#ifdef CONFIG_X2TLB
+#define pte_write(pte)		((pte).pte_high & _PAGE_EXT_USER_WRITE)
+#else
+#define pte_write(pte)		((pte).pte_low & _PAGE_RW)
+#endif
+
+#define PTE_BIT_FUNC(h,fn,op) \
+static inline pte_t pte_##fn(pte_t pte) { pte.pte_##h op; return pte; }
+
+#ifdef CONFIG_X2TLB
+/*
+ * We cheat a bit in the SH-X2 TLB case. As the permission bits are
+ * individually toggled (and user permissions are entirely decoupled from
+ * kernel permissions), we attempt to couple them a bit more sanely here.
+ */
+PTE_BIT_FUNC(high, wrprotect, &= ~_PAGE_EXT_USER_WRITE);
+PTE_BIT_FUNC(high, mkwrite, |= _PAGE_EXT_USER_WRITE | _PAGE_EXT_KERN_WRITE);
+PTE_BIT_FUNC(high, mkhuge, |= _PAGE_SZHUGE);
+#else
+PTE_BIT_FUNC(low, wrprotect, &= ~_PAGE_RW);
+PTE_BIT_FUNC(low, mkwrite, |= _PAGE_RW);
+PTE_BIT_FUNC(low, mkhuge, |= _PAGE_SZHUGE);
+#endif
+
+PTE_BIT_FUNC(low, mkclean, &= ~_PAGE_DIRTY);
+PTE_BIT_FUNC(low, mkdirty, |= _PAGE_DIRTY);
+PTE_BIT_FUNC(low, mkold, &= ~_PAGE_ACCESSED);
+PTE_BIT_FUNC(low, mkyoung, |= _PAGE_ACCESSED);
+
+/*
+ * Macro and implementation to make a page protection as uncachable.
+ */
+#define pgprot_writecombine(prot) \
+	__pgprot(pgprot_val(prot) & ~_PAGE_CACHABLE)
+
+#define pgprot_noncached	 pgprot_writecombine
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ *
+ * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
+ */
+#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	pte.pte_low &= _PAGE_CHG_MASK;
+	pte.pte_low |= pgprot_val(newprot);
+
+#ifdef CONFIG_X2TLB
+	pte.pte_high |= pgprot_val(newprot) >> 32;
+#endif
+
+	return pte;
+}
+
+#define pmd_page_vaddr(pmd)	((unsigned long)pmd_val(pmd))
+#define pmd_page(pmd)		(virt_to_page(pmd_val(pmd)))
+
+/* to find an entry in a page-table-directory. */
+#define pgd_index(address)	(((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define pgd_offset(mm, address)	((mm)->pgd+pgd_index(address))
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address)	pgd_offset(&init_mm, address)
+
+/* Find an entry in the third-level page table.. */
+#define pte_index(address)	((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, address) \
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
+#define pte_offset_map(dir, address)		pte_offset_kernel(dir, address)
+#define pte_offset_map_nested(dir, address)	pte_offset_kernel(dir, address)
+
+#define pte_unmap(pte)		do { } while (0)
+#define pte_unmap_nested(pte)	do { } while (0)
+
+#ifdef CONFIG_X2TLB
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %p(%08lx%08lx).\n", __FILE__, __LINE__, \
+	       &(e), (e).pte_high, (e).pte_low)
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %016llx.\n", __FILE__, __LINE__, pgd_val(e))
+#else
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+#endif
+
+/*
+ * Encode and de-code a swap entry
+ *
+ * Constraints:
+ *	_PAGE_FILE at bit 0
+ *	_PAGE_PRESENT at bit 8
+ *	_PAGE_PROTNONE at bit 9
+ *
+ * For the normal case, we encode the swap type into bits 0:7 and the
+ * swap offset into bits 10:30. For the 64-bit PTE case, we keep the
+ * preserved bits in the low 32-bits and use the upper 32 as the swap
+ * offset (along with a 5-bit type), following the same approach as x86
+ * PAE. This keeps the logic quite simple, and allows for a full 32
+ * PTE_FILE_MAX_BITS, as opposed to the 29-bits we're constrained with
+ * in the pte_low case.
+ *
+ * As is evident by the Alpha code, if we ever get a 64-bit unsigned
+ * long (swp_entry_t) to match up with the 64-bit PTEs, this all becomes
+ * much cleaner..
+ *
+ * NOTE: We should set ZEROs at the position of _PAGE_PRESENT
+ *       and _PAGE_PROTNONE bits
+ */
+#ifdef CONFIG_X2TLB
+#define __swp_type(x)			((x).val & 0x1f)
+#define __swp_offset(x)			((x).val >> 5)
+#define __swp_entry(type, offset)	((swp_entry_t){ (type) | (offset) << 5})
+#define __pte_to_swp_entry(pte)		((swp_entry_t){ (pte).pte_high })
+#define __swp_entry_to_pte(x)		((pte_t){ 0, (x).val })
+
+/*
+ * Encode and decode a nonlinear file mapping entry
+ */
+#define pte_to_pgoff(pte)		((pte).pte_high)
+#define pgoff_to_pte(off)		((pte_t) { _PAGE_FILE, (off) })
+
+#define PTE_FILE_MAX_BITS		32
+#else
+#define __swp_type(x)			((x).val & 0xff)
+#define __swp_offset(x)			((x).val >> 10)
+#define __swp_entry(type, offset)	((swp_entry_t){(type) | (offset) <<10})
+
+#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) >> 1 })
+#define __swp_entry_to_pte(x)		((pte_t) { (x).val << 1 })
+
+/*
+ * Encode and decode a nonlinear file mapping entry
+ */
+#define PTE_FILE_MAX_BITS	29
+#define pte_to_pgoff(pte)	(pte_val(pte) >> 1)
+#define pgoff_to_pte(off)	((pte_t) { ((off) << 1) | _PAGE_FILE })
+#endif
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_SH_PGTABLE_32_H */
diff --git a/include/asm-sh/pgtable_64.h b/include/asm-sh/pgtable_64.h
new file mode 100644
index 000000000000..d422111006f4
--- /dev/null
+++ b/include/asm-sh/pgtable_64.h
@@ -0,0 +1,300 @@
+#ifndef __ASM_SH64_PGTABLE_H
+#define __ASM_SH64_PGTABLE_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/pgtable.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003, 2004  Paul Mundt
+ * Copyright (C) 2003, 2004  Richard Curnow
+ *
+ * This file contains the functions and defines necessary to modify and use
+ * the SuperH page table tree.
+ */
+
+#include <linux/threads.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+
+/*
+ * Error outputs.
+ */
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %016Lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
+/*
+ * Table setting routines. Used within arch/mm only.
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+static __inline__ void set_pte(pte_t *pteptr, pte_t pteval)
+{
+	unsigned long long x = ((unsigned long long) pteval.pte_low);
+	unsigned long long *xp = (unsigned long long *) pteptr;
+	/*
+	 * Sign-extend based on NPHYS.
+	 */
+	*(xp) = (x & NPHYS_SIGN) ? (x | NPHYS_MASK) : x;
+}
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+
+static __inline__ void pmd_set(pmd_t *pmdp,pte_t *ptep)
+{
+	pmd_val(*pmdp) = (unsigned long) ptep;
+}
+
+/*
+ * PGD defines. Top level.
+ */
+
+/* To find an entry in a generic PGD. */
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define __pgd_offset(address) pgd_index(address)
+#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+
+/* To find an entry in a kernel PGD. */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+/*
+ * PMD level access routines. Same notes as above.
+ */
+#define _PMD_EMPTY		0x0
+/* Either the PMD is empty or present, it's not paged out */
+#define pmd_present(pmd_entry)	(pmd_val(pmd_entry) & _PAGE_PRESENT)
+#define pmd_clear(pmd_entry_p)	(set_pmd((pmd_entry_p), __pmd(_PMD_EMPTY)))
+#define pmd_none(pmd_entry)	(pmd_val((pmd_entry)) == _PMD_EMPTY)
+#define pmd_bad(pmd_entry)	((pmd_val(pmd_entry) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
+
+#define pmd_page_vaddr(pmd_entry) \
+	((unsigned long) __va(pmd_val(pmd_entry) & PAGE_MASK))
+
+#define pmd_page(pmd) \
+	(virt_to_page(pmd_val(pmd)))
+
+/* PMD to PTE dereferencing */
+#define pte_index(address) \
+		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+
+#define pte_offset_kernel(dir, addr) \
+		((pte_t *) ((pmd_val(*(dir))) & PAGE_MASK) + pte_index((addr)))
+
+#define pte_offset_map(dir,addr)	pte_offset_kernel(dir, addr)
+#define pte_offset_map_nested(dir,addr)	pte_offset_kernel(dir, addr)
+#define pte_unmap(pte)		do { } while (0)
+#define pte_unmap_nested(pte)	do { } while (0)
+
+#ifndef __ASSEMBLY__
+#define IOBASE_VADDR	0xff000000
+#define IOBASE_END	0xffffffff
+
+/*
+ * PTEL coherent flags.
+ * See Chapter 17 ST50 CPU Core Volume 1, Architecture.
+ */
+/* The bits that are required in the SH-5 TLB are placed in the h/w-defined
+   positions, to avoid expensive bit shuffling on every refill.  The remaining
+   bits are used for s/w purposes and masked out on each refill.
+
+   Note, the PTE slots are used to hold data of type swp_entry_t when a page is
+   swapped out.  Only the _PAGE_PRESENT flag is significant when the page is
+   swapped out, and it must be placed so that it doesn't overlap either the
+   type or offset fields of swp_entry_t.  For x86, offset is at [31:8] and type
+   at [6:1], with _PAGE_PRESENT at bit 0 for both pte_t and swp_entry_t.  This
+   scheme doesn't map to SH-5 because bit [0] controls cacheability.  So bit
+   [2] is used for _PAGE_PRESENT and the type field of swp_entry_t is split
+   into 2 pieces.  That is handled by SWP_ENTRY and SWP_TYPE below. */
+#define _PAGE_WT	0x001  /* CB0: if cacheable, 1->write-thru, 0->write-back */
+#define _PAGE_DEVICE	0x001  /* CB0: if uncacheable, 1->device (i.e. no write-combining or reordering at bus level) */
+#define _PAGE_CACHABLE	0x002  /* CB1: uncachable/cachable */
+#define _PAGE_PRESENT	0x004  /* software: page referenced */
+#define _PAGE_FILE	0x004  /* software: only when !present */
+#define _PAGE_SIZE0	0x008  /* SZ0-bit : size of page */
+#define _PAGE_SIZE1	0x010  /* SZ1-bit : size of page */
+#define _PAGE_SHARED	0x020  /* software: reflects PTEH's SH */
+#define _PAGE_READ	0x040  /* PR0-bit : read access allowed */
+#define _PAGE_EXECUTE	0x080  /* PR1-bit : execute access allowed */
+#define _PAGE_WRITE	0x100  /* PR2-bit : write access allowed */
+#define _PAGE_USER	0x200  /* PR3-bit : user space access allowed */
+#define _PAGE_DIRTY	0x400  /* software: page accessed in write */
+#define _PAGE_ACCESSED	0x800  /* software: page referenced */
+
+/* Mask which drops software flags */
+#define _PAGE_FLAGS_HARDWARE_MASK	0xfffffffffffff3dbLL
+
+/*
+ * HugeTLB support
+ */
+#if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
+#define _PAGE_SZHUGE	(_PAGE_SIZE0)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
+#define _PAGE_SZHUGE	(_PAGE_SIZE1)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512MB)
+#define _PAGE_SZHUGE	(_PAGE_SIZE0 | _PAGE_SIZE1)
+#endif
+
+/*
+ * Default flags for a Kernel page.
+ * This is fundametally also SHARED because the main use of this define
+ * (other than for PGD/PMD entries) is for the VMALLOC pool which is
+ * contextless.
+ *
+ * _PAGE_EXECUTE is required for modules
+ *
+ */
+#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \
+			 _PAGE_EXECUTE | \
+			 _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_DIRTY | \
+			 _PAGE_SHARED)
+
+/* Default flags for a User page */
+#define _PAGE_TABLE	(_KERNPG_TABLE | _PAGE_USER)
+
+#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+/*
+ * We have full permissions (Read/Write/Execute/Shared).
+ */
+#define _PAGE_COMMON	(_PAGE_PRESENT | _PAGE_USER | \
+			 _PAGE_CACHABLE | _PAGE_ACCESSED)
+
+#define PAGE_NONE	__pgprot(_PAGE_CACHABLE | _PAGE_ACCESSED)
+#define PAGE_SHARED	__pgprot(_PAGE_COMMON | _PAGE_READ | _PAGE_WRITE | \
+				 _PAGE_SHARED)
+#define PAGE_EXECREAD	__pgprot(_PAGE_COMMON | _PAGE_READ | _PAGE_EXECUTE)
+
+/*
+ * We need to include PAGE_EXECUTE in PAGE_COPY because it is the default
+ * protection mode for the stack.
+ */
+#define PAGE_COPY	PAGE_EXECREAD
+
+#define PAGE_READONLY	__pgprot(_PAGE_COMMON | _PAGE_READ)
+#define PAGE_WRITEONLY	__pgprot(_PAGE_COMMON | _PAGE_WRITE)
+#define PAGE_RWX	__pgprot(_PAGE_COMMON | _PAGE_READ | \
+				 _PAGE_WRITE | _PAGE_EXECUTE)
+#define PAGE_KERNEL	__pgprot(_KERNPG_TABLE)
+
+/* Make it a device mapping for maximum safety (e.g. for mapping device
+   registers into user-space via /dev/map).  */
+#define pgprot_noncached(x) __pgprot(((x).pgprot & ~(_PAGE_CACHABLE)) | _PAGE_DEVICE)
+#define pgprot_writecombine(prot) __pgprot(pgprot_val(prot) & ~_PAGE_CACHABLE)
+
+/*
+ * Handling allocation failures during page table setup.
+ */
+extern void __handle_bad_pmd_kernel(pmd_t * pmd);
+#define __handle_bad_pmd(x)	__handle_bad_pmd_kernel(x)
+
+/*
+ * PTE level access routines.
+ *
+ * Note1:
+ * It's the tree walk leaf. This is physical address to be stored.
+ *
+ * Note 2:
+ * Regarding the choice of _PTE_EMPTY:
+
+   We must choose a bit pattern that cannot be valid, whether or not the page
+   is present.  bit[2]==1 => present, bit[2]==0 => swapped out.  If swapped
+   out, bits [31:8], [6:3], [1:0] are under swapper control, so only bit[7] is
+   left for us to select.  If we force bit[7]==0 when swapped out, we could use
+   the combination bit[7,2]=2'b10 to indicate an empty PTE.  Alternatively, if
+   we force bit[7]==1 when swapped out, we can use all zeroes to indicate
+   empty.  This is convenient, because the page tables get cleared to zero
+   when they are allocated.
+
+ */
+#define _PTE_EMPTY	0x0
+#define pte_present(x)	(pte_val(x) & _PAGE_PRESENT)
+#define pte_clear(mm,addr,xp)	(set_pte_at(mm, addr, xp, __pte(_PTE_EMPTY)))
+#define pte_none(x)	(pte_val(x) == _PTE_EMPTY)
+
+/*
+ * Some definitions to translate between mem_map, PTEs, and page
+ * addresses:
+ */
+
+/*
+ * Given a PTE, return the index of the mem_map[] entry corresponding
+ * to the page frame the PTE. Get the absolute physical address, make
+ * a relative physical address and translate it to an index.
+ */
+#define pte_pagenr(x)		(((unsigned long) (pte_val(x)) - \
+				 __MEMORY_START) >> PAGE_SHIFT)
+
+/*
+ * Given a PTE, return the "struct page *".
+ */
+#define pte_page(x)		(mem_map + pte_pagenr(x))
+
+/*
+ * Return number of (down rounded) MB corresponding to x pages.
+ */
+#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
+
+
+/*
+ * The following have defined behavior only work if pte_present() is true.
+ */
+static inline int pte_dirty(pte_t pte){ return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte){ return pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE; }
+static inline int pte_write(pte_t pte){ return pte_val(pte) & _PAGE_WRITE; }
+
+static inline pte_t pte_wrprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_WRITE)); return pte; }
+static inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY)); return pte; }
+static inline pte_t pte_mkold(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED)); return pte; }
+static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_WRITE)); return pte; }
+static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
+static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
+static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; }
+
+
+/*
+ * Conversion functions: convert a page and protection to a page entry.
+ *
+ * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
+ */
+#define mk_pte(page,pgprot)							\
+({										\
+	pte_t __pte;								\
+										\
+	set_pte(&__pte, __pte((((page)-mem_map) << PAGE_SHIFT) | 		\
+		__MEMORY_START | pgprot_val((pgprot))));			\
+	__pte;									\
+})
+
+/*
+ * This takes a (absolute) physical page address that is used
+ * by the remapping functions
+ */
+#define mk_pte_phys(physpage, pgprot) \
+({ pte_t __pte; set_pte(&__pte, __pte(physpage | pgprot_val(pgprot))); __pte; })
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{ set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot))); return pte; }
+
+/* Encode and decode a swap entry */
+#define __swp_type(x)			(((x).val & 3) + (((x).val >> 1) & 0x3c))
+#define __swp_offset(x)			((x).val >> 8)
+#define __swp_entry(type, offset)	((swp_entry_t) { ((offset << 8) + ((type & 0x3c) << 1) + (type & 3)) })
+#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x)		((pte_t) { (x).val })
+
+/* Encode and decode a nonlinear file mapping entry */
+#define PTE_FILE_MAX_BITS		29
+#define pte_to_pgoff(pte)		(pte_val(pte))
+#define pgoff_to_pte(off)		((pte_t) { (off) | _PAGE_FILE })
+
+#endif /* !__ASSEMBLY__ */
+
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
+#endif /* __ASM_SH64_PGTABLE_H */
diff --git a/include/asm-sh64/pgtable.h b/include/asm-sh64/pgtable.h
deleted file mode 100644
index 3488fe32e436..000000000000
--- a/include/asm-sh64/pgtable.h
+++ /dev/null
@@ -1,496 +0,0 @@
-#ifndef __ASM_SH64_PGTABLE_H
-#define __ASM_SH64_PGTABLE_H
-
-#include <asm-generic/4level-fixup.h>
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/pgtable.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- * Copyright (C) 2003, 2004  Richard Curnow
- *
- * This file contains the functions and defines necessary to modify and use
- * the SuperH page table tree.
- */
-
-#ifndef __ASSEMBLY__
-#include <asm/processor.h>
-#include <asm/page.h>
-#include <linux/threads.h>
-
-struct vm_area_struct;
-
-extern void paging_init(void);
-
-/* We provide our own get_unmapped_area to avoid cache synonym issue */
-#define HAVE_ARCH_UNMAPPED_AREA
-
-/*
- * Basically we have the same two-level (which is the logical three level
- * Linux page table layout folded) page tables as the i386.
- */
-
-/*
- * ZERO_PAGE is a global shared page that is always zero: used
- * for zero-mapped memory areas etc..
- */
-extern unsigned char empty_zero_page[PAGE_SIZE];
-#define ZERO_PAGE(vaddr) (mem_map + MAP_NR(empty_zero_page))
-
-#endif /* !__ASSEMBLY__ */
-
-/*
- * NEFF and NPHYS related defines.
- * FIXME : These need to be model-dependent.  For now this is OK, SH5-101 and SH5-103
- * implement 32 bits effective and 32 bits physical.  But future implementations may
- * extend beyond this.
- */
-#define NEFF		32
-#define	NEFF_SIGN	(1LL << (NEFF - 1))
-#define	NEFF_MASK	(-1LL << NEFF)
-
-#define NPHYS		32
-#define	NPHYS_SIGN	(1LL << (NPHYS - 1))
-#define	NPHYS_MASK	(-1LL << NPHYS)
-
-/* Typically 2-level is sufficient up to 32 bits of virtual address space, beyond
-   that 3-level would be appropriate. */
-#if defined(CONFIG_SH64_PGTABLE_2_LEVEL)
-/* For 4k pages, this contains 512 entries, i.e. 9 bits worth of address. */
-#define PTRS_PER_PTE	((1<<PAGE_SHIFT)/sizeof(unsigned long long))
-#define PTE_MAGNITUDE	3	      /* sizeof(unsigned long long) magnit. */
-#define PTE_SHIFT	PAGE_SHIFT
-#define PTE_BITS	(PAGE_SHIFT - PTE_MAGNITUDE)
-
-/* top level: PMD. */
-#define PGDIR_SHIFT	(PTE_SHIFT + PTE_BITS)
-#define PGD_BITS	(NEFF - PGDIR_SHIFT)
-#define PTRS_PER_PGD	(1<<PGD_BITS)
-
-/* middle level: PMD. This doesn't do anything for the 2-level case. */
-#define PTRS_PER_PMD	(1)
-
-#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
-#define PGDIR_MASK	(~(PGDIR_SIZE-1))
-#define PMD_SHIFT	PGDIR_SHIFT
-#define PMD_SIZE	PGDIR_SIZE
-#define PMD_MASK	PGDIR_MASK
-
-#elif defined(CONFIG_SH64_PGTABLE_3_LEVEL)
-/*
- * three-level asymmetric paging structure: PGD is top level.
- * The asymmetry comes from 32-bit pointers and 64-bit PTEs.
- */
-/* bottom level: PTE. It's 9 bits = 512 pointers */
-#define PTRS_PER_PTE	((1<<PAGE_SHIFT)/sizeof(unsigned long long))
-#define PTE_MAGNITUDE	3	      /* sizeof(unsigned long long) magnit. */
-#define PTE_SHIFT	PAGE_SHIFT
-#define PTE_BITS	(PAGE_SHIFT - PTE_MAGNITUDE)
-
-/* middle level: PMD. It's 10 bits = 1024 pointers */
-#define PTRS_PER_PMD	((1<<PAGE_SHIFT)/sizeof(unsigned long long *))
-#define PMD_MAGNITUDE	2	      /* sizeof(unsigned long long *) magnit. */
-#define PMD_SHIFT	(PTE_SHIFT + PTE_BITS)
-#define PMD_BITS	(PAGE_SHIFT - PMD_MAGNITUDE)
-
-/* top level: PMD. It's 1 bit = 2 pointers */
-#define PGDIR_SHIFT	(PMD_SHIFT + PMD_BITS)
-#define PGD_BITS	(NEFF - PGDIR_SHIFT)
-#define PTRS_PER_PGD	(1<<PGD_BITS)
-
-#define PMD_SIZE	(1UL << PMD_SHIFT)
-#define PMD_MASK	(~(PMD_SIZE-1))
-#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
-#define PGDIR_MASK	(~(PGDIR_SIZE-1))
-
-#else
-#error "No defined number of page table levels"
-#endif
-
-/*
- * Error outputs.
- */
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %016Lx.\n", __FILE__, __LINE__, pte_val(e))
-#define pmd_ERROR(e) \
-	printk("%s:%d: bad pmd %08lx.\n", __FILE__, __LINE__, pmd_val(e))
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
-
-/*
- * Table setting routines. Used within arch/mm only.
- */
-#define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)
-#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
-
-static __inline__ void set_pte(pte_t *pteptr, pte_t pteval)
-{
-	unsigned long long x = ((unsigned long long) pteval.pte);
-	unsigned long long *xp = (unsigned long long *) pteptr;
-	/*
-	 * Sign-extend based on NPHYS.
-	 */
-	*(xp) = (x & NPHYS_SIGN) ? (x | NPHYS_MASK) : x;
-}
-#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
-
-static __inline__ void pmd_set(pmd_t *pmdp,pte_t *ptep)
-{
-	pmd_val(*pmdp) = (unsigned long) ptep;
-}
-
-/*
- * PGD defines. Top level.
- */
-
-/* To find an entry in a generic PGD. */
-#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
-#define __pgd_offset(address) pgd_index(address)
-#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
-
-/* To find an entry in a kernel PGD. */
-#define pgd_offset_k(address) pgd_offset(&init_mm, address)
-
-/*
- * PGD level access routines.
- *
- * Note1:
- * There's no need to use physical addresses since the tree walk is all
- * in performed in software, until the PTE translation.
- *
- * Note 2:
- * A PGD entry can be uninitialized (_PGD_UNUSED), generically bad,
- * clear (_PGD_EMPTY), present. When present, lower 3 nibbles contain
- * _KERNPG_TABLE. Being a kernel virtual pointer also bit 31 must
- * be 1. Assuming an arbitrary clear value of bit 31 set to 0 and
- * lower 3 nibbles set to 0xFFF (_PGD_EMPTY) any other value is a
- * bad pgd that must be notified via printk().
- *
- */
-#define _PGD_EMPTY		0x0
-
-#if defined(CONFIG_SH64_PGTABLE_2_LEVEL)
-static inline int pgd_none(pgd_t pgd)		{ return 0; }
-static inline int pgd_bad(pgd_t pgd)		{ return 0; }
-#define pgd_present(pgd) ((pgd_val(pgd) & _PAGE_PRESENT) ? 1 : 0)
-#define pgd_clear(xx)				do { } while(0)
-
-#elif defined(CONFIG_SH64_PGTABLE_3_LEVEL)
-#define pgd_present(pgd_entry)	(1)
-#define pgd_none(pgd_entry)	(pgd_val((pgd_entry)) == _PGD_EMPTY)
-/* TODO: Think later about what a useful definition of 'bad' would be now. */
-#define pgd_bad(pgd_entry)	(0)
-#define pgd_clear(pgd_entry_p)	(set_pgd((pgd_entry_p), __pgd(_PGD_EMPTY)))
-
-#endif
-
-
-#define pgd_page_vaddr(pgd_entry)	((unsigned long) (pgd_val(pgd_entry) & PAGE_MASK))
-#define pgd_page(pgd)	(virt_to_page(pgd_val(pgd)))
-
-
-/*
- * PMD defines. Middle level.
- */
-
-/* PGD to PMD dereferencing */
-#if defined(CONFIG_SH64_PGTABLE_2_LEVEL)
-static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
-{
-	return (pmd_t *) dir;
-}
-#elif defined(CONFIG_SH64_PGTABLE_3_LEVEL)
-#define __pmd_offset(address) \
-		(((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
-#define pmd_offset(dir, addr) \
-		((pmd_t *) ((pgd_val(*(dir))) & PAGE_MASK) + __pmd_offset((addr)))
-#endif
-
-/*
- * PMD level access routines. Same notes as above.
- */
-#define _PMD_EMPTY		0x0
-/* Either the PMD is empty or present, it's not paged out */
-#define pmd_present(pmd_entry)	(pmd_val(pmd_entry) & _PAGE_PRESENT)
-#define pmd_clear(pmd_entry_p)	(set_pmd((pmd_entry_p), __pmd(_PMD_EMPTY)))
-#define pmd_none(pmd_entry)	(pmd_val((pmd_entry)) == _PMD_EMPTY)
-#define pmd_bad(pmd_entry)	((pmd_val(pmd_entry) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
-
-#define pmd_page_vaddr(pmd_entry) \
-	((unsigned long) __va(pmd_val(pmd_entry) & PAGE_MASK))
-
-#define pmd_page(pmd) \
-	(virt_to_page(pmd_val(pmd)))
-
-/* PMD to PTE dereferencing */
-#define pte_index(address) \
-		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-
-#define pte_offset_kernel(dir, addr) \
-		((pte_t *) ((pmd_val(*(dir))) & PAGE_MASK) + pte_index((addr)))
-
-#define pte_offset_map(dir,addr)	pte_offset_kernel(dir, addr)
-#define pte_offset_map_nested(dir,addr)	pte_offset_kernel(dir, addr)
-#define pte_unmap(pte)		do { } while (0)
-#define pte_unmap_nested(pte)	do { } while (0)
-
-/* Round it up ! */
-#define USER_PTRS_PER_PGD	((TASK_SIZE+PGDIR_SIZE-1)/PGDIR_SIZE)
-#define FIRST_USER_ADDRESS	0
-
-#ifndef __ASSEMBLY__
-#define VMALLOC_END	0xff000000
-#define VMALLOC_START	0xf0000000
-#define VMALLOC_VMADDR(x) ((unsigned long)(x))
-
-#define IOBASE_VADDR	0xff000000
-#define IOBASE_END	0xffffffff
-
-/*
- * PTEL coherent flags.
- * See Chapter 17 ST50 CPU Core Volume 1, Architecture.
- */
-/* The bits that are required in the SH-5 TLB are placed in the h/w-defined
-   positions, to avoid expensive bit shuffling on every refill.  The remaining
-   bits are used for s/w purposes and masked out on each refill.
-
-   Note, the PTE slots are used to hold data of type swp_entry_t when a page is
-   swapped out.  Only the _PAGE_PRESENT flag is significant when the page is
-   swapped out, and it must be placed so that it doesn't overlap either the
-   type or offset fields of swp_entry_t.  For x86, offset is at [31:8] and type
-   at [6:1], with _PAGE_PRESENT at bit 0 for both pte_t and swp_entry_t.  This
-   scheme doesn't map to SH-5 because bit [0] controls cacheability.  So bit
-   [2] is used for _PAGE_PRESENT and the type field of swp_entry_t is split
-   into 2 pieces.  That is handled by SWP_ENTRY and SWP_TYPE below. */
-#define _PAGE_WT	0x001  /* CB0: if cacheable, 1->write-thru, 0->write-back */
-#define _PAGE_DEVICE	0x001  /* CB0: if uncacheable, 1->device (i.e. no write-combining or reordering at bus level) */
-#define _PAGE_CACHABLE	0x002  /* CB1: uncachable/cachable */
-#define _PAGE_PRESENT	0x004  /* software: page referenced */
-#define _PAGE_FILE	0x004  /* software: only when !present */
-#define _PAGE_SIZE0	0x008  /* SZ0-bit : size of page */
-#define _PAGE_SIZE1	0x010  /* SZ1-bit : size of page */
-#define _PAGE_SHARED	0x020  /* software: reflects PTEH's SH */
-#define _PAGE_READ	0x040  /* PR0-bit : read access allowed */
-#define _PAGE_EXECUTE	0x080  /* PR1-bit : execute access allowed */
-#define _PAGE_WRITE	0x100  /* PR2-bit : write access allowed */
-#define _PAGE_USER	0x200  /* PR3-bit : user space access allowed */
-#define _PAGE_DIRTY	0x400  /* software: page accessed in write */
-#define _PAGE_ACCESSED	0x800  /* software: page referenced */
-
-/* Mask which drops software flags */
-#define _PAGE_FLAGS_HARDWARE_MASK	0xfffffffffffff3dbLL
-
-/*
- * HugeTLB support
- */
-#if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
-#define _PAGE_SZHUGE	(_PAGE_SIZE0)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
-#define _PAGE_SZHUGE	(_PAGE_SIZE1)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512MB)
-#define _PAGE_SZHUGE	(_PAGE_SIZE0 | _PAGE_SIZE1)
-#endif
-
-/*
- * Default flags for a Kernel page.
- * This is fundametally also SHARED because the main use of this define
- * (other than for PGD/PMD entries) is for the VMALLOC pool which is
- * contextless.
- *
- * _PAGE_EXECUTE is required for modules
- *
- */
-#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \
-			 _PAGE_EXECUTE | \
-			 _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_DIRTY | \
-			 _PAGE_SHARED)
-
-/* Default flags for a User page */
-#define _PAGE_TABLE	(_KERNPG_TABLE | _PAGE_USER)
-
-#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
-
-#define PAGE_NONE	__pgprot(_PAGE_CACHABLE | _PAGE_ACCESSED)
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \
-				 _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_USER | \
-				 _PAGE_SHARED)
-/* We need to include PAGE_EXECUTE in PAGE_COPY because it is the default
- * protection mode for the stack. */
-#define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_USER | _PAGE_EXECUTE)
-#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_CACHABLE | \
-				 _PAGE_ACCESSED | _PAGE_USER)
-#define PAGE_KERNEL	__pgprot(_KERNPG_TABLE)
-
-
-/*
- * In ST50 we have full permissions (Read/Write/Execute/Shared).
- * Just match'em all. These are for mmap(), therefore all at least
- * User/Cachable/Present/Accessed. No point in making Fault on Write.
- */
-#define __MMAP_COMMON	(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | _PAGE_ACCESSED)
-       /* sxwr */
-#define __P000	__pgprot(__MMAP_COMMON)
-#define __P001	__pgprot(__MMAP_COMMON | _PAGE_READ)
-#define __P010	__pgprot(__MMAP_COMMON)
-#define __P011	__pgprot(__MMAP_COMMON | _PAGE_READ)
-#define __P100	__pgprot(__MMAP_COMMON | _PAGE_EXECUTE)
-#define __P101	__pgprot(__MMAP_COMMON | _PAGE_EXECUTE | _PAGE_READ)
-#define __P110	__pgprot(__MMAP_COMMON | _PAGE_EXECUTE)
-#define __P111	__pgprot(__MMAP_COMMON | _PAGE_EXECUTE | _PAGE_READ)
-
-#define __S000	__pgprot(__MMAP_COMMON | _PAGE_SHARED)
-#define __S001	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_READ)
-#define __S010	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_WRITE)
-#define __S011	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_READ | _PAGE_WRITE)
-#define __S100	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_EXECUTE)
-#define __S101	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_EXECUTE | _PAGE_READ)
-#define __S110	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_EXECUTE | _PAGE_WRITE)
-#define __S111	__pgprot(__MMAP_COMMON | _PAGE_SHARED | _PAGE_EXECUTE | _PAGE_READ | _PAGE_WRITE)
-
-/* Make it a device mapping for maximum safety (e.g. for mapping device
-   registers into user-space via /dev/map).  */
-#define pgprot_noncached(x) __pgprot(((x).pgprot & ~(_PAGE_CACHABLE)) | _PAGE_DEVICE)
-#define pgprot_writecombine(prot) __pgprot(pgprot_val(prot) & ~_PAGE_CACHABLE)
-
-/*
- * Handling allocation failures during page table setup.
- */
-extern void __handle_bad_pmd_kernel(pmd_t * pmd);
-#define __handle_bad_pmd(x)	__handle_bad_pmd_kernel(x)
-
-/*
- * PTE level access routines.
- *
- * Note1:
- * It's the tree walk leaf. This is physical address to be stored.
- *
- * Note 2:
- * Regarding the choice of _PTE_EMPTY:
-
-   We must choose a bit pattern that cannot be valid, whether or not the page
-   is present.  bit[2]==1 => present, bit[2]==0 => swapped out.  If swapped
-   out, bits [31:8], [6:3], [1:0] are under swapper control, so only bit[7] is
-   left for us to select.  If we force bit[7]==0 when swapped out, we could use
-   the combination bit[7,2]=2'b10 to indicate an empty PTE.  Alternatively, if
-   we force bit[7]==1 when swapped out, we can use all zeroes to indicate
-   empty.  This is convenient, because the page tables get cleared to zero
-   when they are allocated.
-
- */
-#define _PTE_EMPTY	0x0
-#define pte_present(x)	(pte_val(x) & _PAGE_PRESENT)
-#define pte_clear(mm,addr,xp)	(set_pte_at(mm, addr, xp, __pte(_PTE_EMPTY)))
-#define pte_none(x)	(pte_val(x) == _PTE_EMPTY)
-
-/*
- * Some definitions to translate between mem_map, PTEs, and page
- * addresses:
- */
-
-/*
- * Given a PTE, return the index of the mem_map[] entry corresponding
- * to the page frame the PTE. Get the absolute physical address, make
- * a relative physical address and translate it to an index.
- */
-#define pte_pagenr(x)		(((unsigned long) (pte_val(x)) - \
-				 __MEMORY_START) >> PAGE_SHIFT)
-
-/*
- * Given a PTE, return the "struct page *".
- */
-#define pte_page(x)		(mem_map + pte_pagenr(x))
-
-/*
- * Return number of (down rounded) MB corresponding to x pages.
- */
-#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
-
-
-/*
- * The following have defined behavior only work if pte_present() is true.
- */
-static inline int pte_dirty(pte_t pte){ return pte_val(pte) & _PAGE_DIRTY; }
-static inline int pte_young(pte_t pte){ return pte_val(pte) & _PAGE_ACCESSED; }
-static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE; }
-static inline int pte_write(pte_t pte){ return pte_val(pte) & _PAGE_WRITE; }
-
-static inline pte_t pte_wrprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_WRITE)); return pte; }
-static inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY)); return pte; }
-static inline pte_t pte_mkold(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED)); return pte; }
-static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_WRITE)); return pte; }
-static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
-static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
-static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_SZHUGE)); return pte; }
-
-
-/*
- * Conversion functions: convert a page and protection to a page entry.
- *
- * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
- */
-#define mk_pte(page,pgprot)							\
-({										\
-	pte_t __pte;								\
-										\
-	set_pte(&__pte, __pte((((page)-mem_map) << PAGE_SHIFT) | 		\
-		__MEMORY_START | pgprot_val((pgprot))));			\
-	__pte;									\
-})
-
-/*
- * This takes a (absolute) physical page address that is used
- * by the remapping functions
- */
-#define mk_pte_phys(physpage, pgprot) \
-({ pte_t __pte; set_pte(&__pte, __pte(physpage | pgprot_val(pgprot))); __pte; })
-
-static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
-{ set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot))); return pte; }
-
-typedef pte_t *pte_addr_t;
-#define pgtable_cache_init()	do { } while (0)
-
-extern void update_mmu_cache(struct vm_area_struct * vma,
-			     unsigned long address, pte_t pte);
-
-/* Encode and decode a swap entry */
-#define __swp_type(x)			(((x).val & 3) + (((x).val >> 1) & 0x3c))
-#define __swp_offset(x)			((x).val >> 8)
-#define __swp_entry(type, offset)	((swp_entry_t) { ((offset << 8) + ((type & 0x3c) << 1) + (type & 3)) })
-#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) })
-#define __swp_entry_to_pte(x)		((pte_t) { (x).val })
-
-/* Encode and decode a nonlinear file mapping entry */
-#define PTE_FILE_MAX_BITS		29
-#define pte_to_pgoff(pte)		(pte_val(pte))
-#define pgoff_to_pte(off)		((pte_t) { (off) | _PAGE_FILE })
-
-/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */
-#define PageSkip(page)		(0)
-#define kern_addr_valid(addr)	(1)
-
-#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
-		remap_pfn_range(vma, vaddr, pfn, size, prot)
-
-#endif /* !__ASSEMBLY__ */
-
-/*
- * No page table caches to initialise
- */
-#define pgtable_cache_init()    do { } while (0)
-
-#define pte_pfn(x)		(((unsigned long)((x).pte)) >> PAGE_SHIFT)
-#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
-#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
-
-extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
-
-#include <asm-generic/pgtable.h>
-
-#endif /* __ASM_SH64_PGTABLE_H */
-- 
cgit v1.2.3


From 99432700cf2f28976191ab546deffc0007ef4eb7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 14:49:36 +0900
Subject: sh: Tidy up lib64 udelay impl.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/lib64/udelay.c | 7 ++-----
 include/asm-sh/delay.h | 8 ++++++--
 2 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/sh/lib64/udelay.c b/arch/sh/lib64/udelay.c
index 327653914007..4c71a7428e60 100644
--- a/arch/sh/lib64/udelay.c
+++ b/arch/sh/lib64/udelay.c
@@ -13,8 +13,6 @@
 #include <linux/sched.h>
 #include <asm/param.h>
 
-extern unsigned long loops_per_jiffy;
-
 /*
  * Use only for very small delays (< 1 msec).
  *
@@ -49,11 +47,10 @@ void __ndelay(unsigned long long nsecs, unsigned long lpj)
 
 void udelay(unsigned long usecs)
 {
-	__udelay(usecs, loops_per_jiffy);
+	__udelay(usecs, cpu_data[raw_smp_processor_id()].loops_per_jiffy);
 }
 
 void ndelay(unsigned long nsecs)
 {
-	__ndelay(nsecs, loops_per_jiffy);
+	__ndelay(nsecs, cpu_data[raw_smp_processor_id()].loops_per_jiffy);
 }
-
diff --git a/include/asm-sh/delay.h b/include/asm-sh/delay.h
index db599b2a5a9c..031db84f2aa1 100644
--- a/include/asm-sh/delay.h
+++ b/include/asm-sh/delay.h
@@ -6,7 +6,7 @@
  *
  * Delay routines calling functions in arch/sh/lib/delay.c
  */
- 
+
 extern void __bad_udelay(void);
 extern void __bad_ndelay(void);
 
@@ -15,13 +15,17 @@ extern void __ndelay(unsigned long nsecs);
 extern void __const_udelay(unsigned long usecs);
 extern void __delay(unsigned long loops);
 
+#ifdef CONFIG_SUPERH32
 #define udelay(n) (__builtin_constant_p(n) ? \
 	((n) > 20000 ? __bad_udelay() : __const_udelay((n) * 0x10c6ul)) : \
 	__udelay(n))
 
-
 #define ndelay(n) (__builtin_constant_p(n) ? \
 	((n) > 20000 ? __bad_ndelay() : __const_udelay((n) * 5ul)) : \
 	__ndelay(n))
+#else
+extern void udelay(unsigned long usecs);
+extern void ndelay(unsigned long nsecs);
+#endif
 
 #endif /* __ASM_SH_DELAY_H */
-- 
cgit v1.2.3


From b542ad1161d3b1771ea6fa3e7cedd311a49f91c9 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 15:14:31 +0900
Subject: sh: Plug in the SHmedia ELF relocations.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/elf.h | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 12cc4b392bf0..5a1e920f0598 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -5,7 +5,7 @@
 #include <asm/ptrace.h>
 #include <asm/user.h>
 
-/* SH relocation types  */
+/* SH (particularly SHcompact) relocation types  */
 #define	R_SH_NONE		0
 #define	R_SH_DIR32		1
 #define	R_SH_REL32		2
@@ -43,6 +43,11 @@
 #define	R_SH_RELATIVE		165
 #define	R_SH_GOTOFF		166
 #define	R_SH_GOTPC		167
+/* SHmedia relocs */
+#define R_SH_IMM_LOW16		246
+#define R_SH_IMM_LOW16_PCREL	247
+#define R_SH_IMM_MEDLOW16	248
+#define R_SH_IMM_MEDLOW16_PCREL	249
 /* Keep this the last entry.  */
 #define	R_SH_NUM		256
 
@@ -103,12 +108,35 @@ typedef struct user_fpu_struct elf_fpregset_t;
 
 #define ELF_PLATFORM  (NULL)
 
+#ifdef __SH5__
+#define ELF_PLAT_INIT(_r, load_addr) \
+  do { _r->regs[0]=0; _r->regs[1]=0; _r->regs[2]=0; _r->regs[3]=0; \
+       _r->regs[4]=0; _r->regs[5]=0; _r->regs[6]=0; _r->regs[7]=0; \
+       _r->regs[8]=0; _r->regs[9]=0; _r->regs[10]=0; _r->regs[11]=0; \
+       _r->regs[12]=0; _r->regs[13]=0; _r->regs[14]=0; _r->regs[15]=0; \
+       _r->regs[16]=0; _r->regs[17]=0; _r->regs[18]=0; _r->regs[19]=0; \
+       _r->regs[20]=0; _r->regs[21]=0; _r->regs[22]=0; _r->regs[23]=0; \
+       _r->regs[24]=0; _r->regs[25]=0; _r->regs[26]=0; _r->regs[27]=0; \
+       _r->regs[28]=0; _r->regs[29]=0; _r->regs[30]=0; _r->regs[31]=0; \
+       _r->regs[32]=0; _r->regs[33]=0; _r->regs[34]=0; _r->regs[35]=0; \
+       _r->regs[36]=0; _r->regs[37]=0; _r->regs[38]=0; _r->regs[39]=0; \
+       _r->regs[40]=0; _r->regs[41]=0; _r->regs[42]=0; _r->regs[43]=0; \
+       _r->regs[44]=0; _r->regs[45]=0; _r->regs[46]=0; _r->regs[47]=0; \
+       _r->regs[48]=0; _r->regs[49]=0; _r->regs[50]=0; _r->regs[51]=0; \
+       _r->regs[52]=0; _r->regs[53]=0; _r->regs[54]=0; _r->regs[55]=0; \
+       _r->regs[56]=0; _r->regs[57]=0; _r->regs[58]=0; _r->regs[59]=0; \
+       _r->regs[60]=0; _r->regs[61]=0; _r->regs[62]=0; \
+       _r->tregs[0]=0; _r->tregs[1]=0; _r->tregs[2]=0; _r->tregs[3]=0; \
+       _r->tregs[4]=0; _r->tregs[5]=0; _r->tregs[6]=0; _r->tregs[7]=0; \
+       _r->sr = SR_FD | SR_MMU; } while (0)
+#else
 #define ELF_PLAT_INIT(_r, load_addr) \
   do { _r->regs[0]=0; _r->regs[1]=0; _r->regs[2]=0; _r->regs[3]=0; \
        _r->regs[4]=0; _r->regs[5]=0; _r->regs[6]=0; _r->regs[7]=0; \
        _r->regs[8]=0; _r->regs[9]=0; _r->regs[10]=0; _r->regs[11]=0; \
        _r->regs[12]=0; _r->regs[13]=0; _r->regs[14]=0; \
        _r->sr = SR_FD; } while (0)
+#endif
 
 #ifdef __KERNEL__
 #define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
-- 
cgit v1.2.3


From 379a95d1d2c3e3682e380084c40b6fc01e38fa1f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 16:51:28 +0900
Subject: sh: Tidy up various clear_page()/copy_page() definitions.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/lib64/page_clear.S |  4 ++--
 arch/sh/lib64/page_copy.S  | 56 ++++++++++++++++++++++------------------------
 arch/sh/mm/copy_page.S     |  6 ++---
 arch/sh/mm/init.c          | 15 -------------
 arch/sh/mm/pg-nommu.c      |  4 ++--
 include/asm-sh/page.h      | 12 ++--------
 6 files changed, 36 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/arch/sh/lib64/page_clear.S b/arch/sh/lib64/page_clear.S
index ac0111d669a3..007ab48ecc1c 100644
--- a/arch/sh/lib64/page_clear.S
+++ b/arch/sh/lib64/page_clear.S
@@ -25,8 +25,8 @@
 	.little
 
 	.balign 8
-	.global sh64_page_clear
-sh64_page_clear:
+	.global clear_page
+clear_page:
 	pta/l 1f, tr1
 	pta/l 2f, tr2
 	ptabs/l r18, tr0
diff --git a/arch/sh/lib64/page_copy.S b/arch/sh/lib64/page_copy.S
index e159c3cd2582..0ec6fca63b56 100644
--- a/arch/sh/lib64/page_copy.S
+++ b/arch/sh/lib64/page_copy.S
@@ -10,8 +10,8 @@
    of SH5-101 cut2 eval chip with Cayman board DDR memory.
 
    Parameters:
-   r2 : source effective address (start of page)
-   r3 : destination effective address (start of page)
+   r2 : destination effective address (start of page)
+   r3 : source effective address (start of page)
 
    Always copies 4096 bytes.
 
@@ -27,10 +27,10 @@
 	.little
 
 	.balign 8
-	.global sh64_page_copy
-sh64_page_copy:
+	.global copy_page
+copy_page:
 
-	/* Copy 4096 bytes worth of data from r2 to r3.
+	/* Copy 4096 bytes worth of data from r3 to r2.
 	   Do prefetches 4 lines ahead.
 	   Do alloco 2 lines ahead */
 
@@ -41,21 +41,21 @@ sh64_page_copy:
 
 #if 0
 	/* TAKum03020 */
-	ld.q r2, 0x00, r63
-	ld.q r2, 0x20, r63
-	ld.q r2, 0x40, r63
-	ld.q r2, 0x60, r63
+	ld.q r3, 0x00, r63
+	ld.q r3, 0x20, r63
+	ld.q r3, 0x40, r63
+	ld.q r3, 0x60, r63
 #endif
-	alloco r3, 0x00
+	alloco r2, 0x00
 	synco		! TAKum03020
-	alloco r3, 0x20
+	alloco r2, 0x20
 	synco		! TAKum03020
 
 	movi 3968, r6
-	add  r3, r6, r6
+	add  r2, r6, r6
 	addi r6, 64, r7
 	addi r7, 64, r8
-	sub r2, r3, r60
+	sub r3, r2, r60
 	addi r60, 8, r61
 	addi r61, 8, r62
 	addi r62, 8, r23
@@ -67,25 +67,23 @@ sh64_page_copy:
 1:
 #if 0
 	/* TAKum03020 */
-	bge/u r3, r6, tr2  ! skip prefetch for last 4 lines
-	ldx.q r3, r22, r63 ! prefetch 4 lines hence
+	bge/u r2, r6, tr2  ! skip prefetch for last 4 lines
+	ldx.q r2, r22, r63 ! prefetch 4 lines hence
 #endif
 2:
-	bge/u r3, r7, tr3  ! skip alloco for last 2 lines
-	alloco r3, 0x40    ! alloc destination line 2 lines ahead
+	bge/u r2, r7, tr3  ! skip alloco for last 2 lines
+	alloco r2, 0x40    ! alloc destination line 2 lines ahead
 	synco		! TAKum03020
 3:
-	ldx.q r3, r60, r36
-	ldx.q r3, r61, r37
-	ldx.q r3, r62, r38
-	ldx.q r3, r23, r39
-	st.q  r3,   0, r36
-	st.q  r3,   8, r37
-	st.q  r3,  16, r38
-	st.q  r3,  24, r39
-	addi r3, 32, r3
-	bgt/l r8, r3, tr1
+	ldx.q r2, r60, r36
+	ldx.q r2, r61, r37
+	ldx.q r2, r62, r38
+	ldx.q r2, r23, r39
+	st.q  r2,   0, r36
+	st.q  r2,   8, r37
+	st.q  r2,  16, r38
+	st.q  r2,  24, r39
+	addi r2, 32, r2
+	bgt/l r8, r2, tr1
 
 	blink tr0, r63	   ! return
-
-
diff --git a/arch/sh/mm/copy_page.S b/arch/sh/mm/copy_page.S
index 40685018b952..b879545fa28b 100644
--- a/arch/sh/mm/copy_page.S
+++ b/arch/sh/mm/copy_page.S
@@ -9,11 +9,11 @@
 #include <asm/page.h>
 
 /*
- * copy_page_slow
+ * copy_page
  * @to: P1 address
  * @from: P1 address
  *
- * void copy_page_slow(void *to, void *from)
+ * void copy_page(void *to, void *from)
  */
 
 /*
@@ -23,7 +23,7 @@
  * r10 --- to
  * r11 --- from
  */
-ENTRY(copy_page_slow)
+ENTRY(copy_page)
 	mov.l	r8,@-r15
 	mov.l	r10,@-r15
 	mov.l	r11,@-r15
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 5d19c8c7ab53..79c309780f95 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -24,9 +24,6 @@
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 pgd_t swapper_pg_dir[PTRS_PER_PGD];
 
-void (*copy_page)(void *from, void *to);
-void (*clear_page)(void *to);
-
 void show_mem(void)
 {
 	int total = 0, reserved = 0, free = 0;
@@ -203,18 +200,6 @@ void __init mem_init(void)
 	memset(empty_zero_page, 0, PAGE_SIZE);
 	__flush_wback_region(empty_zero_page, PAGE_SIZE);
 
-	/*
-	 * Setup wrappers for copy/clear_page(), these will get overridden
-	 * later in the boot process if a better method is available.
-	 */
-#ifdef CONFIG_MMU
-	copy_page = copy_page_slow;
-	clear_page = clear_page_slow;
-#else
-	copy_page = copy_page_nommu;
-	clear_page = clear_page_nommu;
-#endif
-
 	after_bootmem = 1;
 
 	codesize =  (unsigned long) &_etext - (unsigned long) &_text;
diff --git a/arch/sh/mm/pg-nommu.c b/arch/sh/mm/pg-nommu.c
index d15221beaa16..677dd57f0877 100644
--- a/arch/sh/mm/pg-nommu.c
+++ b/arch/sh/mm/pg-nommu.c
@@ -14,12 +14,12 @@
 #include <linux/string.h>
 #include <asm/page.h>
 
-void copy_page_nommu(void *to, void *from)
+void copy_page(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
 
-void clear_page_nommu(void *to)
+void clear_page(void *to)
 {
 	memset(to, 0, PAGE_SIZE);
 }
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 93a89841227f..e21b0d12e139 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -55,20 +55,12 @@
 
 #ifndef __ASSEMBLY__
 
-extern void (*clear_page)(void *to);
-extern void (*copy_page)(void *to, void *from);
-
 extern unsigned long shm_align_mask;
 extern unsigned long max_low_pfn, min_low_pfn;
 extern unsigned long memory_start, memory_end;
 
-#ifdef CONFIG_MMU
-extern void clear_page_slow(void *to);
-extern void copy_page_slow(void *to, void *from);
-#else
-extern void clear_page_nommu(void *to);
-extern void copy_page_nommu(void *to, void *from);
-#endif
+extern void clear_page(void *to);
+extern void copy_page(void *to, void *from);
 
 #if !defined(CONFIG_CACHE_OFF) && defined(CONFIG_MMU) && \
 	(defined(CONFIG_CPU_SH4) || defined(CONFIG_SH7705_CACHE_32KB))
-- 
cgit v1.2.3


From 5a4f7c66be981c6b5f44a4d66a14ea6ac9b7b6b0 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 18:08:06 +0900
Subject: sh: Share bug/debug traps across _32 and _64.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile_32     |  2 +-
 arch/sh/kernel/Makefile_64     |  2 +-
 arch/sh/kernel/cpu/sh2/entry.S |  2 +-
 arch/sh/kernel/process_32.c    | 46 -----------------------------
 arch/sh/kernel/process_64.c    | 15 ++++------
 arch/sh/kernel/traps.c         | 66 ++++++++++++++++++++++++++++++++++++++++++
 arch/sh/kernel/traps_32.c      | 25 +---------------
 include/asm-sh/ptrace.h        |  2 +-
 include/asm-sh/system.h        | 25 ++++++++++++----
 include/asm-sh/types.h         |  6 ++++
 10 files changed, 102 insertions(+), 89 deletions(-)
 create mode 100644 arch/sh/kernel/traps.c

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 34e8fa0d40cc..990ba74db0d6 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -6,7 +6,7 @@ extra-y	:= head_32.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
 	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o sys_sh32.o \
-	   syscalls_32.o time.o topology.o traps_32.o
+	   syscalls_32.o time.o topology.o traps.o traps_32.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index aeac4006df7b..10e3ae1c64b8 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -2,7 +2,7 @@ extra-y	:= head_64.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
 	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh.o sys_sh64.o \
-	   syscalls_64.o time.o topology.o traps_64.o
+	   syscalls_64.o time.o topology.o traps.o traps_64.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/cpu/sh2/entry.S b/arch/sh/kernel/cpu/sh2/entry.S
index ee8f1fe84b08..4ff2334b4a38 100644
--- a/arch/sh/kernel/cpu/sh2/entry.S
+++ b/arch/sh/kernel/cpu/sh2/entry.S
@@ -250,7 +250,7 @@ ENTRY(sh_bios_handler)
 1:	.long	gdb_vbr_vector
 #endif /* CONFIG_SH_STANDARD_BIOS */
 
-ENTRY(address_error_handler)
+ENTRY(address_error_trap_handler)
 	mov	r15,r4				! regs
 	add	#4,r4
 	mov	#OFF_PC,r0
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index c9291f462311..b48324867eee 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -482,49 +482,3 @@ asmlinkage void break_point_trap(void)
 
 	force_sig(SIGTRAP, current);
 }
-
-/*
- * Generic trap handler.
- */
-asmlinkage void debug_trap_handler(unsigned long r4, unsigned long r5,
-				   unsigned long r6, unsigned long r7,
-				   struct pt_regs __regs)
-{
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
-
-	/* Rewind */
-	regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
-
-	if (notify_die(DIE_TRAP, "debug trap", regs, 0, regs->tra & 0xff,
-		       SIGTRAP) == NOTIFY_STOP)
-		return;
-
-	force_sig(SIGTRAP, current);
-}
-
-/*
- * Special handler for BUG() traps.
- */
-asmlinkage void bug_trap_handler(unsigned long r4, unsigned long r5,
-				 unsigned long r6, unsigned long r7,
-				 struct pt_regs __regs)
-{
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
-
-	/* Rewind */
-	regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
-
-	if (notify_die(DIE_TRAP, "bug trap", regs, 0, TRAPA_BUG_OPCODE & 0xff,
-		       SIGTRAP) == NOTIFY_STOP)
-		return;
-
-#ifdef CONFIG_BUG
-	if (__kernel_text_address(instruction_pointer(regs))) {
-		u16 insn = *(u16 *)instruction_pointer(regs);
-		if (insn == TRAPA_BUG_OPCODE)
-			handle_BUG(regs);
-	}
-#endif
-
-	force_sig(SIGTRAP, current);
-}
diff --git a/arch/sh/kernel/process_64.c b/arch/sh/kernel/process_64.c
index 47415671da0c..92d01465eb87 100644
--- a/arch/sh/kernel/process_64.c
+++ b/arch/sh/kernel/process_64.c
@@ -1,12 +1,10 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
+ * arch/sh/kernel/process_64.c
  *
- * arch/sh64/kernel/process.c
+ * This file handles the architecture-dependent parts of process handling..
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
+ * Copyright (C) 2003 - 2007  Paul Mundt
  * Copyright (C) 2003, 2004 Richard Curnow
  *
  * Started from SH3/4 version:
@@ -15,10 +13,9 @@
  *   In turn started from i386 version:
  *     Copyright (C) 1995  Linus Torvalds
  *
- */
-
-/*
- * This file handles the architecture-dependent parts of process handling..
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/mm.h>
 #include <linux/fs.h>
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
new file mode 100644
index 000000000000..bf70827b17cc
--- /dev/null
+++ b/arch/sh/kernel/traps.c
@@ -0,0 +1,66 @@
+#include <linux/bug.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <linux/kdebug.h>
+#include <asm/system.h>
+
+#ifdef CONFIG_BUG
+static void handle_BUG(struct pt_regs *regs)
+{
+	enum bug_trap_type tt;
+	tt = report_bug(regs->pc, regs);
+	if (tt == BUG_TRAP_TYPE_WARN) {
+		regs->pc += instruction_size(regs->pc);
+		return;
+	}
+
+	die("Kernel BUG", regs, TRAPA_BUG_OPCODE & 0xff);
+}
+
+int is_valid_bugaddr(unsigned long addr)
+{
+	return addr >= PAGE_OFFSET;
+}
+#endif
+
+/*
+ * Generic trap handler.
+ */
+BUILD_TRAP_HANDLER(debug)
+{
+	TRAP_HANDLER_DECL;
+
+	/* Rewind */
+	regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
+
+	if (notify_die(DIE_TRAP, "debug trap", regs, 0, vec & 0xff,
+		       SIGTRAP) == NOTIFY_STOP)
+		return;
+
+	force_sig(SIGTRAP, current);
+}
+
+/*
+ * Special handler for BUG() traps.
+ */
+BUILD_TRAP_HANDLER(bug)
+{
+	TRAP_HANDLER_DECL;
+
+	/* Rewind */
+	regs->pc -= instruction_size(ctrl_inw(regs->pc - 4));
+
+	if (notify_die(DIE_TRAP, "bug trap", regs, 0, TRAPA_BUG_OPCODE & 0xff,
+		       SIGTRAP) == NOTIFY_STOP)
+		return;
+
+#ifdef CONFIG_BUG
+	if (__kernel_text_address(instruction_pointer(regs))) {
+		opcode_t insn = *(opcode_t *)instruction_pointer(regs);
+		if (insn == TRAPA_BUG_OPCODE)
+			handle_BUG(regs);
+	}
+#endif
+
+	force_sig(SIGTRAP, current);
+}
diff --git a/arch/sh/kernel/traps_32.c b/arch/sh/kernel/traps_32.c
index cf99111cb33f..0d05fb3c48e3 100644
--- a/arch/sh/kernel/traps_32.c
+++ b/arch/sh/kernel/traps_32.c
@@ -837,10 +837,6 @@ void *set_exception_table_vec(unsigned int vec, void *handler)
 	return old_handler;
 }
 
-extern asmlinkage void address_error_handler(unsigned long r4, unsigned long r5,
-					     unsigned long r6, unsigned long r7,
-					     struct pt_regs __regs);
-
 void __init trap_init(void)
 {
 	set_exception_table_vec(TRAP_RESERVED_INST, do_reserved_inst);
@@ -866,7 +862,7 @@ void __init trap_init(void)
 #endif
 
 #ifdef CONFIG_CPU_SH2
-	set_exception_table_vec(TRAP_ADDRESS_ERROR, address_error_handler);
+	set_exception_table_vec(TRAP_ADDRESS_ERROR, address_error_trap_handler);
 #endif
 #ifdef CONFIG_CPU_SH2A
 	set_exception_table_vec(TRAP_DIVZERO_ERROR, do_divide_error);
@@ -877,25 +873,6 @@ void __init trap_init(void)
 	per_cpu_trap_init();
 }
 
-#ifdef CONFIG_BUG
-void handle_BUG(struct pt_regs *regs)
-{
-	enum bug_trap_type tt;
-	tt = report_bug(regs->pc, regs);
-	if (tt == BUG_TRAP_TYPE_WARN) {
-		regs->pc += 2;
-		return;
-	}
-
-	die("Kernel BUG", regs, TRAPA_BUG_OPCODE & 0xff);
-}
-
-int is_valid_bugaddr(unsigned long addr)
-{
-	return addr >= PAGE_OFFSET;
-}
-#endif
-
 void show_trace(struct task_struct *tsk, unsigned long *sp,
 		struct pt_regs *regs)
 {
diff --git a/include/asm-sh/ptrace.h b/include/asm-sh/ptrace.h
index a83a7b45ba6f..8d6c92b3e770 100644
--- a/include/asm-sh/ptrace.h
+++ b/include/asm-sh/ptrace.h
@@ -95,7 +95,7 @@ struct pt_dspregs {
 #include <asm/addrspace.h>
 
 #define user_mode(regs)			(((regs)->sr & 0x40000000)==0)
-#define instruction_pointer(regs)	((regs)->pc)
+#define instruction_pointer(regs)	((unsigned long)(regs)->pc)
 
 extern void show_regs(struct pt_regs *);
 
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 0cfa96aa5844..6b02dfd587ea 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -185,12 +185,25 @@ void default_idle(void);
 void per_cpu_trap_init(void);
 
 asmlinkage void break_point_trap(void);
-asmlinkage void debug_trap_handler(unsigned long r4, unsigned long r5,
-				   unsigned long r6, unsigned long r7,
-				   struct pt_regs __regs);
-asmlinkage void bug_trap_handler(unsigned long r4, unsigned long r5,
-				 unsigned long r6, unsigned long r7,
-				 struct pt_regs __regs);
+
+#ifdef CONFIG_SUPERH32
+#define BUILD_TRAP_HANDLER(name)					\
+asmlinkage void name##_trap_handler(unsigned long r4, unsigned long r5,	\
+				    unsigned long r6, unsigned long r7,	\
+				    struct pt_regs __regs)
+
+#define TRAP_HANDLER_DECL				\
+	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);	\
+	unsigned int vec = regs->tra;
+#else
+#define BUILD_TRAP_HANDLER(name)	\
+asmlinkage void name##_trap_handler(unsigned int vec, struct pt_regs *regs)
+#define TRAP_HANDLER_DECL
+#endif
+
+BUILD_TRAP_HANDLER(address_error);
+BUILD_TRAP_HANDLER(debug);
+BUILD_TRAP_HANDLER(bug);
 
 #define arch_align_stack(x) (x)
 
diff --git a/include/asm-sh/types.h b/include/asm-sh/types.h
index 7ba69d9707ef..a6e1d4126e67 100644
--- a/include/asm-sh/types.h
+++ b/include/asm-sh/types.h
@@ -52,6 +52,12 @@ typedef unsigned long long u64;
 
 typedef u32 dma_addr_t;
 
+#ifdef CONFIG_SUPERH32
+typedef u16 opcode_t;
+#else
+typedef u32 opcode_t;
+#endif
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From ac490a4893d37279f704876d430c3683071398b2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 18:26:28 +0900
Subject: sh: Move quad-word real-address I/O defs to io.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/ptrace_64.c | 1 -
 include/asm-sh/io.h        | 5 +++++
 2 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/sh/kernel/ptrace_64.c b/arch/sh/kernel/ptrace_64.c
index 14e7d5e5679a..51bb546365cb 100644
--- a/arch/sh/kernel/ptrace_64.c
+++ b/arch/sh/kernel/ptrace_64.c
@@ -253,7 +253,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 asmlinkage int sh64_ptrace(long request, long pid, long addr, long data)
 {
-	extern void poke_real_address_q(unsigned long long addr, unsigned long long data);
 #define WPC_DBRMODE 0x0d104008
 	static int first_call = 1;
 
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 556aabe844c5..74305edceeeb 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -248,6 +248,11 @@ static inline void ctrl_delay(void)
 #endif
 }
 
+/* Quad-word real-mode I/O, don't ask.. */
+unsigned long long peek_real_address_q(unsigned long long addr);
+unsigned long long poke_real_address_q(unsigned long long addr,
+				       unsigned long long val);
+
 #define IO_SPACE_LIMIT 0xffffffff
 
 #if !defined(CONFIG_MMU)
-- 
cgit v1.2.3


From ffd25eb6cd1aa2bcdedf0c06be7e180022055cd1 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 20 Nov 2007 18:44:39 +0900
Subject: sh: Stubs for fpu disabled on SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/processor_64.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
index 312fd73fb87e..f54648229231 100644
--- a/include/asm-sh/processor_64.h
+++ b/include/asm-sh/processor_64.h
@@ -262,11 +262,16 @@ static inline void grab_fpu(struct pt_regs *regs)
 #define FPSCR_INIT  0x00000000
 #endif
 
+#ifdef CONFIG_SH_FPU
 /* Save the current FP regs */
 void fpsave(struct sh_fpu_hard_struct *fpregs);
 
 /* Initialise the FP state of a task */
 void fpinit(struct sh_fpu_hard_struct *fpregs);
+#else
+#define fpsave(fpregs)	do { } while (0)
+#define fpinit(fpregs)	do { } while (0)
+#endif
 
 extern struct task_struct *last_task_used_math;
 
-- 
cgit v1.2.3


From caff44e7db86243fd6163bc79e543123b0139a46 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 15:23:07 +0900
Subject: sh: Set HPAGE_SHIFT for 512MB hugetlb pages.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/page.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index e21b0d12e139..cdaacefe509b 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -45,6 +45,8 @@
 #define HPAGE_SHIFT	22
 #elif defined(CONFIG_HUGETLB_PAGE_SIZE_64MB)
 #define HPAGE_SHIFT	26
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512MB)
+#define HPAGE_SHIFT	29
 #endif
 
 #ifdef CONFIG_HUGETLB_PAGE
-- 
cgit v1.2.3


From 6deb6f9129591a2386f5c125d54a22ab78fc8b61 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 15:34:02 +0900
Subject: sh: Move PXSEG comments to addrspace.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/addrspace.h |  8 ++++++++
 include/asm-sh/page.h      | 16 ----------------
 2 files changed, 8 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index cb9da9992fa0..e7f2deb28e06 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -17,6 +17,14 @@
 /* If this CPU supports segmentation, hook up the helpers */
 #ifdef P1SEG
 
+/*
+   [ P0/U0 (virtual) ]		0x00000000     <------ User space
+   [ P1 (fixed)   cached ]	0x80000000     <------ Kernel space
+   [ P2 (fixed)  non-cachable]	0xA0000000     <------ Physical access
+   [ P3 (virtual) cached]	0xC0000000     <------ vmalloced area
+   [ P4 control   ]		0xE0000000
+ */
+
 /* Returns the privileged segment base of a given address  */
 #define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
 
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index cdaacefe509b..407bf5a14936 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -5,14 +5,6 @@
  * Copyright (C) 1999  Niibe Yutaka
  */
 
-/*
-   [ P0/U0 (virtual) ]		0x00000000     <------ User space
-   [ P1 (fixed)   cached ]	0x80000000     <------ Kernel space
-   [ P2 (fixed)  non-cachable]	0xA0000000     <------ Physical access
-   [ P3 (virtual) cached]	0xC0000000     <------ vmalloced area
-   [ P4 control   ]		0xE0000000
- */
-
 #ifdef __KERNEL__
 
 /* PAGE_SHIFT determines the page size */
@@ -115,14 +107,6 @@ typedef struct { unsigned long pgd; } pgd_t;
 /* to align the pointer to the (next) page boundary */
 #define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
 
-/*
- * IF YOU CHANGE THIS, PLEASE ALSO CHANGE
- *
- *	arch/sh/kernel/vmlinux.lds.S
- *
- * which has the same constant encoded..
- */
-
 #define __MEMORY_START		CONFIG_MEMORY_START
 #define __MEMORY_SIZE		CONFIG_MEMORY_SIZE
 
-- 
cgit v1.2.3


From 0b2328bbc08ee8f7377f9dccdd6fb0c707b34d2f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 15:58:24 +0900
Subject: sh: Fix up generic BUG build for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/bug.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/bug.h b/include/asm-sh/bug.h
index afb1c2723ec7..c01718040166 100644
--- a/include/asm-sh/bug.h
+++ b/include/asm-sh/bug.h
@@ -72,11 +72,6 @@ do {							\
 	unlikely(__ret_warn_on);				\
 })
 
-struct pt_regs;
-
-/* arch/sh/kernel/traps.c */
-void handle_BUG(struct pt_regs *);
-
 #endif /* CONFIG_GENERIC_BUG */
 
 #include <asm-generic/bug.h>
-- 
cgit v1.2.3


From 55183e9bb2c2ce43d88eaa575c2d6d4fd6d865a3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 16:19:12 +0900
Subject: sh: SH-5 uses a 64-bit PTE_MAGNITUDE, as X2 TLB.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pgtable.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 5a800c69e049..b4d7561cd9e0 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -49,7 +49,7 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
  * traditional two-level paging structure
  */
 /* PTE bits */
-#ifdef CONFIG_X2TLB
+#if defined(CONFIG_X2TLB) || defined(CONFIG_SUPERH64)
 # define PTE_MAGNITUDE	3	/* 64-bit PTEs on extended mode SH-X2 TLB */
 #else
 # define PTE_MAGNITUDE	2	/* 32-bit PTEs */
-- 
cgit v1.2.3


From c8eef8800f1c693a2de6374b1948c8ea5e0ad75f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 17:28:09 +0900
Subject: sh: Purge dead sh64 headers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh64/Kbuild              |   1 -
 include/asm-sh64/a.out.h             |  38 -------
 include/asm-sh64/atomic.h            | 158 ----------------------------
 include/asm-sh64/auxvec.h            |   4 -
 include/asm-sh64/bitops.h            | 155 ----------------------------
 include/asm-sh64/bug.h               |  19 ----
 include/asm-sh64/bugs.h              |  38 -------
 include/asm-sh64/byteorder.h         |  49 ---------
 include/asm-sh64/cpumask.h           |   6 --
 include/asm-sh64/cputime.h           |   6 --
 include/asm-sh64/current.h           |  28 -----
 include/asm-sh64/delay.h             |  11 --
 include/asm-sh64/device.h            |   7 --
 include/asm-sh64/div64.h             |   6 --
 include/asm-sh64/dma-mapping.h       | 194 -----------------------------------
 include/asm-sh64/dma.h               |  41 --------
 include/asm-sh64/elf.h               | 107 -------------------
 include/asm-sh64/emergency-restart.h |   6 --
 include/asm-sh64/errno.h             |   6 --
 include/asm-sh64/fb.h                |  19 ----
 include/asm-sh64/fcntl.h             |   1 -
 include/asm-sh64/futex.h             |   6 --
 include/asm-sh64/gpio.h              |   8 --
 include/asm-sh64/hardirq.h           |  18 ----
 include/asm-sh64/hw_irq.h            |  15 ---
 include/asm-sh64/ide.h               |  29 ------
 include/asm-sh64/ioctl.h             |   1 -
 include/asm-sh64/ioctls.h            | 116 ---------------------
 include/asm-sh64/ipcbuf.h            |  40 --------
 include/asm-sh64/irq_regs.h          |   1 -
 include/asm-sh64/kdebug.h            |   1 -
 include/asm-sh64/kmap_types.h        |   7 --
 include/asm-sh64/linkage.h           |   7 --
 include/asm-sh64/local.h             |   7 --
 include/asm-sh64/mc146818rtc.h       |   7 --
 include/asm-sh64/mman.h              |   6 --
 include/asm-sh64/mmu.h               |   7 --
 include/asm-sh64/module.h            |  20 ----
 include/asm-sh64/msgbuf.h            |  42 --------
 include/asm-sh64/mutex.h             |   9 --
 include/asm-sh64/namei.h             |  24 -----
 include/asm-sh64/page.h              | 119 ---------------------
 include/asm-sh64/param.h             |  42 --------
 include/asm-sh64/percpu.h            |   6 --
 include/asm-sh64/pgalloc.h           | 125 ----------------------
 include/asm-sh64/poll.h              |   8 --
 include/asm-sh64/posix_types.h       | 131 -----------------------
 include/asm-sh64/ptrace.h            |  35 -------
 include/asm-sh64/resource.h          |   6 --
 include/asm-sh64/scatterlist.h       |  37 -------
 include/asm-sh64/sci.h               |   1 -
 include/asm-sh64/sections.h          |   7 --
 include/asm-sh64/segment.h           |   6 --
 include/asm-sh64/semaphore-helper.h  | 101 ------------------
 include/asm-sh64/semaphore.h         | 119 ---------------------
 include/asm-sh64/sembuf.h            |  36 -------
 include/asm-sh64/serial.h            |  31 ------
 include/asm-sh64/setup.h             |  22 ----
 include/asm-sh64/shmbuf.h            |  53 ----------
 include/asm-sh64/shmparam.h          |  12 ---
 include/asm-sh64/sigcontext.h        |  30 ------
 include/asm-sh64/siginfo.h           |   6 --
 include/asm-sh64/signal.h            | 159 ----------------------------
 include/asm-sh64/smp.h               |  15 ---
 include/asm-sh64/socket.h            |   6 --
 include/asm-sh64/sockios.h           |  25 -----
 include/asm-sh64/spinlock.h          |  17 ---
 include/asm-sh64/stat.h              |  88 ----------------
 include/asm-sh64/statfs.h            |   6 --
 include/asm-sh64/termbits.h          |   6 --
 include/asm-sh64/termios.h           |  99 ------------------
 include/asm-sh64/timex.h             |  31 ------
 include/asm-sh64/tlbflush.h          |  27 -----
 include/asm-sh64/topology.h          |   6 --
 include/asm-sh64/types.h             |  74 -------------
 include/asm-sh64/ucontext.h          |  23 -----
 include/asm-sh64/unaligned.h         |  17 ---
 include/asm-sh64/user.h              |  70 -------------
 78 files changed, 2873 deletions(-)
 delete mode 100644 include/asm-sh64/Kbuild
 delete mode 100644 include/asm-sh64/a.out.h
 delete mode 100644 include/asm-sh64/atomic.h
 delete mode 100644 include/asm-sh64/auxvec.h
 delete mode 100644 include/asm-sh64/bitops.h
 delete mode 100644 include/asm-sh64/bug.h
 delete mode 100644 include/asm-sh64/bugs.h
 delete mode 100644 include/asm-sh64/byteorder.h
 delete mode 100644 include/asm-sh64/cpumask.h
 delete mode 100644 include/asm-sh64/cputime.h
 delete mode 100644 include/asm-sh64/current.h
 delete mode 100644 include/asm-sh64/delay.h
 delete mode 100644 include/asm-sh64/device.h
 delete mode 100644 include/asm-sh64/div64.h
 delete mode 100644 include/asm-sh64/dma-mapping.h
 delete mode 100644 include/asm-sh64/dma.h
 delete mode 100644 include/asm-sh64/elf.h
 delete mode 100644 include/asm-sh64/emergency-restart.h
 delete mode 100644 include/asm-sh64/errno.h
 delete mode 100644 include/asm-sh64/fb.h
 delete mode 100644 include/asm-sh64/fcntl.h
 delete mode 100644 include/asm-sh64/futex.h
 delete mode 100644 include/asm-sh64/gpio.h
 delete mode 100644 include/asm-sh64/hardirq.h
 delete mode 100644 include/asm-sh64/hw_irq.h
 delete mode 100644 include/asm-sh64/ide.h
 delete mode 100644 include/asm-sh64/ioctl.h
 delete mode 100644 include/asm-sh64/ioctls.h
 delete mode 100644 include/asm-sh64/ipcbuf.h
 delete mode 100644 include/asm-sh64/irq_regs.h
 delete mode 100644 include/asm-sh64/kdebug.h
 delete mode 100644 include/asm-sh64/kmap_types.h
 delete mode 100644 include/asm-sh64/linkage.h
 delete mode 100644 include/asm-sh64/local.h
 delete mode 100644 include/asm-sh64/mc146818rtc.h
 delete mode 100644 include/asm-sh64/mman.h
 delete mode 100644 include/asm-sh64/mmu.h
 delete mode 100644 include/asm-sh64/module.h
 delete mode 100644 include/asm-sh64/msgbuf.h
 delete mode 100644 include/asm-sh64/mutex.h
 delete mode 100644 include/asm-sh64/namei.h
 delete mode 100644 include/asm-sh64/page.h
 delete mode 100644 include/asm-sh64/param.h
 delete mode 100644 include/asm-sh64/percpu.h
 delete mode 100644 include/asm-sh64/pgalloc.h
 delete mode 100644 include/asm-sh64/poll.h
 delete mode 100644 include/asm-sh64/posix_types.h
 delete mode 100644 include/asm-sh64/ptrace.h
 delete mode 100644 include/asm-sh64/resource.h
 delete mode 100644 include/asm-sh64/scatterlist.h
 delete mode 100644 include/asm-sh64/sci.h
 delete mode 100644 include/asm-sh64/sections.h
 delete mode 100644 include/asm-sh64/segment.h
 delete mode 100644 include/asm-sh64/semaphore-helper.h
 delete mode 100644 include/asm-sh64/semaphore.h
 delete mode 100644 include/asm-sh64/sembuf.h
 delete mode 100644 include/asm-sh64/serial.h
 delete mode 100644 include/asm-sh64/setup.h
 delete mode 100644 include/asm-sh64/shmbuf.h
 delete mode 100644 include/asm-sh64/shmparam.h
 delete mode 100644 include/asm-sh64/sigcontext.h
 delete mode 100644 include/asm-sh64/siginfo.h
 delete mode 100644 include/asm-sh64/signal.h
 delete mode 100644 include/asm-sh64/smp.h
 delete mode 100644 include/asm-sh64/socket.h
 delete mode 100644 include/asm-sh64/sockios.h
 delete mode 100644 include/asm-sh64/spinlock.h
 delete mode 100644 include/asm-sh64/stat.h
 delete mode 100644 include/asm-sh64/statfs.h
 delete mode 100644 include/asm-sh64/termbits.h
 delete mode 100644 include/asm-sh64/termios.h
 delete mode 100644 include/asm-sh64/timex.h
 delete mode 100644 include/asm-sh64/tlbflush.h
 delete mode 100644 include/asm-sh64/topology.h
 delete mode 100644 include/asm-sh64/types.h
 delete mode 100644 include/asm-sh64/ucontext.h
 delete mode 100644 include/asm-sh64/unaligned.h
 delete mode 100644 include/asm-sh64/user.h

(limited to 'include')

diff --git a/include/asm-sh64/Kbuild b/include/asm-sh64/Kbuild
deleted file mode 100644
index c68e1680da01..000000000000
--- a/include/asm-sh64/Kbuild
+++ /dev/null
@@ -1 +0,0 @@
-include include/asm-generic/Kbuild.asm
diff --git a/include/asm-sh64/a.out.h b/include/asm-sh64/a.out.h
deleted file mode 100644
index 237ee4e5b72a..000000000000
--- a/include/asm-sh64/a.out.h
+++ /dev/null
@@ -1,38 +0,0 @@
-#ifndef __ASM_SH64_A_OUT_H
-#define __ASM_SH64_A_OUT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/a.out.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#ifdef __KERNEL__
-
-#define STACK_TOP	TASK_SIZE
-#define STACK_TOP_MAX	STACK_TOP
-
-#endif
-
-#endif /* __ASM_SH64_A_OUT_H */
diff --git a/include/asm-sh64/atomic.h b/include/asm-sh64/atomic.h
deleted file mode 100644
index 28f2ea9b567b..000000000000
--- a/include/asm-sh64/atomic.h
+++ /dev/null
@@ -1,158 +0,0 @@
-#ifndef __ASM_SH64_ATOMIC_H
-#define __ASM_SH64_ATOMIC_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/atomic.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-
-/*
- * Atomic operations that C can't guarantee us.  Useful for
- * resource counting etc..
- *
- */
-
-typedef struct { volatile int counter; } atomic_t;
-
-#define ATOMIC_INIT(i)	( (atomic_t) { (i) } )
-
-#define atomic_read(v)		((v)->counter)
-#define atomic_set(v,i)		((v)->counter = (i))
-
-#include <asm/system.h>
-
-/*
- * To get proper branch prediction for the main line, we must branch
- * forward to code at the end of this object's .text section, then
- * branch back to restart the operation.
- */
-
-static __inline__ void atomic_add(int i, atomic_t * v)
-{
-	unsigned long flags;
-
-	local_irq_save(flags);
-	*(long *)v += i;
-	local_irq_restore(flags);
-}
-
-static __inline__ void atomic_sub(int i, atomic_t *v)
-{
-	unsigned long flags;
-
-	local_irq_save(flags);
-	*(long *)v -= i;
-	local_irq_restore(flags);
-}
-
-static __inline__ int atomic_add_return(int i, atomic_t * v)
-{
-	unsigned long temp, flags;
-
-	local_irq_save(flags);
-	temp = *(long *)v;
-	temp += i;
-	*(long *)v = temp;
-	local_irq_restore(flags);
-
-	return temp;
-}
-
-#define atomic_add_negative(a, v)	(atomic_add_return((a), (v)) < 0)
-
-static __inline__ int atomic_sub_return(int i, atomic_t * v)
-{
-	unsigned long temp, flags;
-
-	local_irq_save(flags);
-	temp = *(long *)v;
-	temp -= i;
-	*(long *)v = temp;
-	local_irq_restore(flags);
-
-	return temp;
-}
-
-#define atomic_dec_return(v) atomic_sub_return(1,(v))
-#define atomic_inc_return(v) atomic_add_return(1,(v))
-
-/*
- * atomic_inc_and_test - increment and test
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-#define atomic_inc_and_test(v) (atomic_inc_return(v) == 0)
-
-#define atomic_sub_and_test(i,v) (atomic_sub_return((i), (v)) == 0)
-#define atomic_dec_and_test(v) (atomic_sub_return(1, (v)) == 0)
-
-#define atomic_inc(v) atomic_add(1,(v))
-#define atomic_dec(v) atomic_sub(1,(v))
-
-static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
-{
-	int ret;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ret = v->counter;
-	if (likely(ret == old))
-		v->counter = new;
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
-
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
-{
-	int ret;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ret = v->counter;
-	if (ret != u)
-		v->counter += a;
-	local_irq_restore(flags);
-
-	return ret != u;
-}
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
-static __inline__ void atomic_clear_mask(unsigned int mask, atomic_t *v)
-{
-	unsigned long flags;
-
-	local_irq_save(flags);
-	*(long *)v &= ~mask;
-	local_irq_restore(flags);
-}
-
-static __inline__ void atomic_set_mask(unsigned int mask, atomic_t *v)
-{
-	unsigned long flags;
-
-	local_irq_save(flags);
-	*(long *)v |= mask;
-	local_irq_restore(flags);
-}
-
-/* Atomic operations are already serializing on SH */
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-
-#include <asm-generic/atomic.h>
-#endif /* __ASM_SH64_ATOMIC_H */
diff --git a/include/asm-sh64/auxvec.h b/include/asm-sh64/auxvec.h
deleted file mode 100644
index 1ad5a44bdc76..000000000000
--- a/include/asm-sh64/auxvec.h
+++ /dev/null
@@ -1,4 +0,0 @@
-#ifndef __ASM_SH64_AUXVEC_H
-#define __ASM_SH64_AUXVEC_H
-
-#endif /* __ASM_SH64_AUXVEC_H */
diff --git a/include/asm-sh64/bitops.h b/include/asm-sh64/bitops.h
deleted file mode 100644
index 600c59efb4c2..000000000000
--- a/include/asm-sh64/bitops.h
+++ /dev/null
@@ -1,155 +0,0 @@
-#ifndef __ASM_SH64_BITOPS_H
-#define __ASM_SH64_BITOPS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/bitops.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- */
-
-#ifdef __KERNEL__
-
-#ifndef _LINUX_BITOPS_H
-#error only <linux/bitops.h> can be included directly
-#endif
-
-#include <linux/compiler.h>
-#include <asm/system.h>
-/* For __swab32 */
-#include <asm/byteorder.h>
-
-static __inline__ void set_bit(int nr, volatile void * addr)
-{
-	int	mask;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a |= mask;
-	local_irq_restore(flags);
-}
-
-/*
- * clear_bit() doesn't provide any barrier for the compiler.
- */
-#define smp_mb__before_clear_bit()	barrier()
-#define smp_mb__after_clear_bit()	barrier()
-static inline void clear_bit(int nr, volatile unsigned long *a)
-{
-	int	mask;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a &= ~mask;
-	local_irq_restore(flags);
-}
-
-static __inline__ void change_bit(int nr, volatile void * addr)
-{
-	int	mask;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a ^= mask;
-	local_irq_restore(flags);
-}
-
-static __inline__ int test_and_set_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a |= mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
-
-static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a &= ~mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
-
-static __inline__ int test_and_change_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a ^= mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
-
-#include <asm-generic/bitops/non-atomic.h>
-
-static __inline__ unsigned long ffz(unsigned long word)
-{
-	unsigned long result, __d2, __d3;
-
-        __asm__("gettr  tr0, %2\n\t"
-                "pta    $+32, tr0\n\t"
-                "andi   %1, 1, %3\n\t"
-                "beq    %3, r63, tr0\n\t"
-                "pta    $+4, tr0\n"
-                "0:\n\t"
-                "shlri.l        %1, 1, %1\n\t"
-                "addi   %0, 1, %0\n\t"
-                "andi   %1, 1, %3\n\t"
-                "beqi   %3, 1, tr0\n"
-                "1:\n\t"
-                "ptabs  %2, tr0\n\t"
-                : "=r" (result), "=r" (word), "=r" (__d2), "=r" (__d3)
-                : "0" (0L), "1" (word));
-
-	return result;
-}
-
-#include <asm-generic/bitops/__ffs.h>
-#include <asm-generic/bitops/find.h>
-#include <asm-generic/bitops/hweight.h>
-#include <asm-generic/bitops/lock.h>
-#include <asm-generic/bitops/sched.h>
-#include <asm-generic/bitops/ffs.h>
-#include <asm-generic/bitops/ext2-non-atomic.h>
-#include <asm-generic/bitops/ext2-atomic.h>
-#include <asm-generic/bitops/minix.h>
-#include <asm-generic/bitops/fls.h>
-#include <asm-generic/bitops/fls64.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH64_BITOPS_H */
diff --git a/include/asm-sh64/bug.h b/include/asm-sh64/bug.h
deleted file mode 100644
index f3a9c9248ef4..000000000000
--- a/include/asm-sh64/bug.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef __ASM_SH64_BUG_H
-#define __ASM_SH64_BUG_H
-
-#ifdef CONFIG_BUG
-/*
- * Tell the user there is some problem, then force a segfault (in process
- * context) or a panic (interrupt context).
- */
-#define BUG() do { \
-	printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
-	*(volatile int *)0 = 0; \
-} while (0)
-
-#define HAVE_ARCH_BUG
-#endif
-
-#include <asm-generic/bug.h>
-
-#endif /* __ASM_SH64_BUG_H */
diff --git a/include/asm-sh64/bugs.h b/include/asm-sh64/bugs.h
deleted file mode 100644
index 05554aaea672..000000000000
--- a/include/asm-sh64/bugs.h
+++ /dev/null
@@ -1,38 +0,0 @@
-#ifndef __ASM_SH64_BUGS_H
-#define __ASM_SH64_BUGS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/bugs.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-
-/*
- * This is included by init/main.c to check for architecture-dependent bugs.
- *
- * Needs:
- *	void check_bugs(void);
- */
-
-/*
- * I don't know of any Super-H bugs yet.
- */
-
-#include <asm/processor.h>
-
-static void __init check_bugs(void)
-{
-	extern char *get_cpu_subtype(void);
-	extern unsigned long loops_per_jiffy;
-
-	cpu_data->loops_per_jiffy = loops_per_jiffy;
-
-	printk("CPU: %s\n", get_cpu_subtype());
-}
-#endif /* __ASM_SH64_BUGS_H */
diff --git a/include/asm-sh64/byteorder.h b/include/asm-sh64/byteorder.h
deleted file mode 100644
index 7419d78820ee..000000000000
--- a/include/asm-sh64/byteorder.h
+++ /dev/null
@@ -1,49 +0,0 @@
-#ifndef __ASM_SH64_BYTEORDER_H
-#define __ASM_SH64_BYTEORDER_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/byteorder.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <asm/types.h>
-
-static inline __attribute_const__ __u32 ___arch__swab32(__u32 x)
-{
-	__asm__("byterev	%0, %0\n\t"
-		"shari		%0, 32, %0"
-		: "=r" (x)
-		: "0" (x));
-	return x;
-}
-
-static inline __attribute_const__ __u16 ___arch__swab16(__u16 x)
-{
-	__asm__("byterev	%0, %0\n\t"
-		"shari		%0, 48, %0"
-		: "=r" (x)
-		: "0" (x));
-	return x;
-}
-
-#define __arch__swab32(x) ___arch__swab32(x)
-#define __arch__swab16(x) ___arch__swab16(x)
-
-#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
-#  define __BYTEORDER_HAS_U64__
-#  define __SWAB_64_THRU_32__
-#endif
-
-#ifdef __LITTLE_ENDIAN__
-#include <linux/byteorder/little_endian.h>
-#else
-#include <linux/byteorder/big_endian.h>
-#endif
-
-#endif /* __ASM_SH64_BYTEORDER_H */
diff --git a/include/asm-sh64/cpumask.h b/include/asm-sh64/cpumask.h
deleted file mode 100644
index b7b105dbedaf..000000000000
--- a/include/asm-sh64/cpumask.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_CPUMASK_H
-#define __ASM_SH64_CPUMASK_H
-
-#include <asm-generic/cpumask.h>
-
-#endif /* __ASM_SH64_CPUMASK_H */
diff --git a/include/asm-sh64/cputime.h b/include/asm-sh64/cputime.h
deleted file mode 100644
index 0fd89da2aa86..000000000000
--- a/include/asm-sh64/cputime.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __SH64_CPUTIME_H
-#define __SH64_CPUTIME_H
-
-#include <asm-generic/cputime.h>
-
-#endif /* __SH64_CPUTIME_H */
diff --git a/include/asm-sh64/current.h b/include/asm-sh64/current.h
deleted file mode 100644
index 261224339d6f..000000000000
--- a/include/asm-sh64/current.h
+++ /dev/null
@@ -1,28 +0,0 @@
-#ifndef __ASM_SH64_CURRENT_H
-#define __ASM_SH64_CURRENT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/current.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-
-#include <linux/thread_info.h>
-
-struct task_struct;
-
-static __inline__ struct task_struct * get_current(void)
-{
-	return current_thread_info()->task;
-}
-
-#define current get_current()
-
-#endif /* __ASM_SH64_CURRENT_H */
-
diff --git a/include/asm-sh64/delay.h b/include/asm-sh64/delay.h
deleted file mode 100644
index 6ae31301a16a..000000000000
--- a/include/asm-sh64/delay.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef __ASM_SH64_DELAY_H
-#define __ASM_SH64_DELAY_H
-
-extern void __delay(int loops);
-extern void __udelay(unsigned long long usecs, unsigned long lpj);
-extern void __ndelay(unsigned long long nsecs, unsigned long lpj);
-extern void udelay(unsigned long usecs);
-extern void ndelay(unsigned long nsecs);
-
-#endif /* __ASM_SH64_DELAY_H */
-
diff --git a/include/asm-sh64/device.h b/include/asm-sh64/device.h
deleted file mode 100644
index d8f9872b0e2d..000000000000
--- a/include/asm-sh64/device.h
+++ /dev/null
@@ -1,7 +0,0 @@
-/*
- * Arch specific extensions to struct device
- *
- * This file is released under the GPLv2
- */
-#include <asm-generic/device.h>
-
diff --git a/include/asm-sh64/div64.h b/include/asm-sh64/div64.h
deleted file mode 100644
index f75869565e2e..000000000000
--- a/include/asm-sh64/div64.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_DIV64_H
-#define __ASM_SH64_DIV64_H
-
-#include <asm-generic/div64.h>
-
-#endif /* __ASM_SH64_DIV64_H */
diff --git a/include/asm-sh64/dma-mapping.h b/include/asm-sh64/dma-mapping.h
deleted file mode 100644
index 18f8dd642ac5..000000000000
--- a/include/asm-sh64/dma-mapping.h
+++ /dev/null
@@ -1,194 +0,0 @@
-#ifndef __ASM_SH_DMA_MAPPING_H
-#define __ASM_SH_DMA_MAPPING_H
-
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-#include <asm/io.h>
-
-struct pci_dev;
-extern void *consistent_alloc(struct pci_dev *hwdev, size_t size,
-				    dma_addr_t *dma_handle);
-extern void consistent_free(struct pci_dev *hwdev, size_t size,
-				  void *vaddr, dma_addr_t dma_handle);
-
-#define dma_supported(dev, mask)	(1)
-
-static inline int dma_set_mask(struct device *dev, u64 mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
-}
-
-static inline void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t flag)
-{
-	return consistent_alloc(NULL, size, dma_handle);
-}
-
-static inline void dma_free_coherent(struct device *dev, size_t size,
-		       void *vaddr, dma_addr_t dma_handle)
-{
-	consistent_free(NULL, size, vaddr, dma_handle);
-}
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-#define dma_is_consistent(d, h) (1)
-
-static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
-				  enum dma_data_direction dir)
-{
-	unsigned long start = (unsigned long) vaddr;
-	unsigned long s = start & L1_CACHE_ALIGN_MASK;
-	unsigned long e = (start + size) & L1_CACHE_ALIGN_MASK;
-
-	for (; s <= e; s += L1_CACHE_BYTES)
-		asm volatile ("ocbp	%0, 0" : : "r" (s));
-}
-
-static inline dma_addr_t dma_map_single(struct device *dev,
-					void *ptr, size_t size,
-					enum dma_data_direction dir)
-{
-#if defined(CONFIG_PCI) && !defined(CONFIG_SH_PCIDMA_NONCOHERENT)
-	if (dev->bus == &pci_bus_type)
-		return virt_to_phys(ptr);
-#endif
-	dma_cache_sync(dev, ptr, size, dir);
-
-	return virt_to_phys(ptr);
-}
-
-#define dma_unmap_single(dev, addr, size, dir)	do { } while (0)
-
-static inline int dma_map_sg(struct device *dev, struct scatterlist *sg,
-			     int nents, enum dma_data_direction dir)
-{
-	int i;
-
-	for (i = 0; i < nents; i++) {
-#if !defined(CONFIG_PCI) || defined(CONFIG_SH_PCIDMA_NONCOHERENT)
-		dma_cache_sync(dev, sg_virt(&sg[i]), sg[i].length, dir);
-#endif
-		sg[i].dma_address = sg_phys(&sg[i]);
-	}
-
-	return nents;
-}
-
-#define dma_unmap_sg(dev, sg, nents, dir)	do { } while (0)
-
-static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
-				      unsigned long offset, size_t size,
-				      enum dma_data_direction dir)
-{
-	return dma_map_single(dev, page_address(page) + offset, size, dir);
-}
-
-static inline void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
-				  size_t size, enum dma_data_direction dir)
-{
-	dma_unmap_single(dev, dma_address, size, dir);
-}
-
-static inline void dma_sync_single(struct device *dev, dma_addr_t dma_handle,
-				   size_t size, enum dma_data_direction dir)
-{
-#if defined(CONFIG_PCI) && !defined(CONFIG_SH_PCIDMA_NONCOHERENT)
-	if (dev->bus == &pci_bus_type)
-		return;
-#endif
-	dma_cache_sync(dev, phys_to_virt(dma_handle), size, dir);
-}
-
-static inline void dma_sync_single_range(struct device *dev,
-					 dma_addr_t dma_handle,
-					 unsigned long offset, size_t size,
-					 enum dma_data_direction dir)
-{
-#if defined(CONFIG_PCI) && !defined(CONFIG_SH_PCIDMA_NONCOHERENT)
-	if (dev->bus == &pci_bus_type)
-		return;
-#endif
-	dma_cache_sync(dev, phys_to_virt(dma_handle) + offset, size, dir);
-}
-
-static inline void dma_sync_sg(struct device *dev, struct scatterlist *sg,
-			       int nelems, enum dma_data_direction dir)
-{
-	int i;
-
-	for (i = 0; i < nelems; i++) {
-#if !defined(CONFIG_PCI) || defined(CONFIG_SH_PCIDMA_NONCOHERENT)
-		dma_cache_sync(dev, sg_virt(&sg[i]), sg[i].length, dir);
-#endif
-		sg[i].dma_address = sg_phys(&sg[i]);
-	}
-}
-
-static inline void dma_sync_single_for_cpu(struct device *dev,
-					   dma_addr_t dma_handle, size_t size,
-					   enum dma_data_direction dir)
-{
-	dma_sync_single(dev, dma_handle, size, dir);
-}
-
-static inline void dma_sync_single_for_device(struct device *dev,
-					   dma_addr_t dma_handle, size_t size,
-					   enum dma_data_direction dir)
-{
-	dma_sync_single(dev, dma_handle, size, dir);
-}
-
-static inline void dma_sync_single_range_for_cpu(struct device *dev,
-						 dma_addr_t dma_handle,
-						 unsigned long offset,
-						 size_t size,
-						 enum dma_data_direction direction)
-{
-	dma_sync_single_for_cpu(dev, dma_handle+offset, size, direction);
-}
-
-static inline void dma_sync_single_range_for_device(struct device *dev,
-						    dma_addr_t dma_handle,
-						    unsigned long offset,
-						    size_t size,
-						    enum dma_data_direction direction)
-{
-	dma_sync_single_for_device(dev, dma_handle+offset, size, direction);
-}
-
-static inline void dma_sync_sg_for_cpu(struct device *dev,
-				       struct scatterlist *sg, int nelems,
-				       enum dma_data_direction dir)
-{
-	dma_sync_sg(dev, sg, nelems, dir);
-}
-
-static inline void dma_sync_sg_for_device(struct device *dev,
-				       struct scatterlist *sg, int nelems,
-				       enum dma_data_direction dir)
-{
-	dma_sync_sg(dev, sg, nelems, dir);
-}
-
-static inline int dma_get_cache_alignment(void)
-{
-	/*
-	 * Each processor family will define its own L1_CACHE_SHIFT,
-	 * L1_CACHE_BYTES wraps to this, so this is always safe.
-	 */
-	return L1_CACHE_BYTES;
-}
-
-static inline int dma_mapping_error(dma_addr_t dma_addr)
-{
-	return dma_addr == 0;
-}
-
-#endif /* __ASM_SH_DMA_MAPPING_H */
-
diff --git a/include/asm-sh64/dma.h b/include/asm-sh64/dma.h
deleted file mode 100644
index e701f39470a2..000000000000
--- a/include/asm-sh64/dma.h
+++ /dev/null
@@ -1,41 +0,0 @@
-#ifndef __ASM_SH64_DMA_H
-#define __ASM_SH64_DMA_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/dma.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-
-#include <linux/mm.h>
-#include <asm/io.h>
-#include <asm/pgtable.h>
-
-#define MAX_DMA_CHANNELS	4
-
-/*
- * SH5 can DMA in any memory area.
- *
- * The static definition is dodgy because it should limit
- * the highest DMA-able address based on the actual
- * Physical memory available. This is actually performed
- * at run time in defining the memory allowed to DMA_ZONE.
- */
-#define MAX_DMA_ADDRESS		~(NPHYS_MASK)
-
-#define DMA_MODE_READ		0
-#define DMA_MODE_WRITE		1
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy 	(0)
-#endif
-
-#endif /* __ASM_SH64_DMA_H */
diff --git a/include/asm-sh64/elf.h b/include/asm-sh64/elf.h
deleted file mode 100644
index f994286e1998..000000000000
--- a/include/asm-sh64/elf.h
+++ /dev/null
@@ -1,107 +0,0 @@
-#ifndef __ASM_SH64_ELF_H
-#define __ASM_SH64_ELF_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/elf.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * ELF register definitions..
- */
-
-#include <asm/ptrace.h>
-#include <asm/user.h>
-#include <asm/byteorder.h>
-
-typedef unsigned long elf_greg_t;
-
-#define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t))
-typedef elf_greg_t elf_gregset_t[ELF_NGREG];
-
-typedef struct user_fpu_struct elf_fpregset_t;
-
-/*
- * This is used to ensure we don't load something for the wrong architecture.
- */
-#define elf_check_arch(x) ( (x)->e_machine == EM_SH )
-
-/*
- * These are used to set parameters in the core dumps.
- */
-#define ELF_CLASS	ELFCLASS32
-#ifdef __LITTLE_ENDIAN__
-#define ELF_DATA	ELFDATA2LSB
-#else
-#define ELF_DATA	ELFDATA2MSB
-#endif
-#define ELF_ARCH	EM_SH
-
-#define USE_ELF_CORE_DUMP
-#define ELF_EXEC_PAGESIZE	4096
-
-/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
-   use of this is to invoke "./ld.so someprog" to test out a new version of
-   the loader.  We need to make sure that it is out of the way of the program
-   that it will "exec", and that there is sufficient room for the brk.  */
-
-#define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
-
-#define	R_SH_DIR32		1
-#define	R_SH_REL32		2
-#define	R_SH_IMM_LOW16		246
-#define	R_SH_IMM_LOW16_PCREL	247
-#define	R_SH_IMM_MEDLOW16	248
-#define	R_SH_IMM_MEDLOW16_PCREL	249
-
-#define ELF_CORE_COPY_REGS(_dest,_regs)				\
-	memcpy((char *) &_dest, (char *) _regs,			\
-	       sizeof(struct pt_regs));
-
-/* This yields a mask that user programs can use to figure out what
-   instruction set this CPU supports.  This could be done in user space,
-   but it's not easy, and we've already done it here.  */
-
-#define ELF_HWCAP	(0)
-
-/* This yields a string that ld.so will use to load implementation
-   specific libraries for optimization.  This is more specific in
-   intent than poking at uname or /proc/cpuinfo.
-
-   For the moment, we have only optimizations for the Intel generations,
-   but that could change... */
-
-#define ELF_PLATFORM  (NULL)
-
-#define ELF_PLAT_INIT(_r, load_addr) \
-  do { _r->regs[0]=0; _r->regs[1]=0; _r->regs[2]=0; _r->regs[3]=0; \
-       _r->regs[4]=0; _r->regs[5]=0; _r->regs[6]=0; _r->regs[7]=0; \
-       _r->regs[8]=0; _r->regs[9]=0; _r->regs[10]=0; _r->regs[11]=0; \
-       _r->regs[12]=0; _r->regs[13]=0; _r->regs[14]=0; _r->regs[15]=0; \
-       _r->regs[16]=0; _r->regs[17]=0; _r->regs[18]=0; _r->regs[19]=0; \
-       _r->regs[20]=0; _r->regs[21]=0; _r->regs[22]=0; _r->regs[23]=0; \
-       _r->regs[24]=0; _r->regs[25]=0; _r->regs[26]=0; _r->regs[27]=0; \
-       _r->regs[28]=0; _r->regs[29]=0; _r->regs[30]=0; _r->regs[31]=0; \
-       _r->regs[32]=0; _r->regs[33]=0; _r->regs[34]=0; _r->regs[35]=0; \
-       _r->regs[36]=0; _r->regs[37]=0; _r->regs[38]=0; _r->regs[39]=0; \
-       _r->regs[40]=0; _r->regs[41]=0; _r->regs[42]=0; _r->regs[43]=0; \
-       _r->regs[44]=0; _r->regs[45]=0; _r->regs[46]=0; _r->regs[47]=0; \
-       _r->regs[48]=0; _r->regs[49]=0; _r->regs[50]=0; _r->regs[51]=0; \
-       _r->regs[52]=0; _r->regs[53]=0; _r->regs[54]=0; _r->regs[55]=0; \
-       _r->regs[56]=0; _r->regs[57]=0; _r->regs[58]=0; _r->regs[59]=0; \
-       _r->regs[60]=0; _r->regs[61]=0; _r->regs[62]=0; \
-       _r->tregs[0]=0; _r->tregs[1]=0; _r->tregs[2]=0; _r->tregs[3]=0; \
-       _r->tregs[4]=0; _r->tregs[5]=0; _r->tregs[6]=0; _r->tregs[7]=0; \
-       _r->sr = SR_FD | SR_MMU; } while (0)
-
-#ifdef __KERNEL__
-#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
-#endif
-
-#endif /* __ASM_SH64_ELF_H */
diff --git a/include/asm-sh64/emergency-restart.h b/include/asm-sh64/emergency-restart.h
deleted file mode 100644
index 108d8c48e42e..000000000000
--- a/include/asm-sh64/emergency-restart.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_EMERGENCY_RESTART_H
-#define _ASM_EMERGENCY_RESTART_H
-
-#include <asm-generic/emergency-restart.h>
-
-#endif /* _ASM_EMERGENCY_RESTART_H */
diff --git a/include/asm-sh64/errno.h b/include/asm-sh64/errno.h
deleted file mode 100644
index 57b46d4bdd41..000000000000
--- a/include/asm-sh64/errno.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_ERRNO_H
-#define __ASM_SH64_ERRNO_H
-
-#include <asm-generic/errno.h>
-
-#endif /* __ASM_SH64_ERRNO_H */
diff --git a/include/asm-sh64/fb.h b/include/asm-sh64/fb.h
deleted file mode 100644
index d92e99cd8c8a..000000000000
--- a/include/asm-sh64/fb.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef _ASM_FB_H_
-#define _ASM_FB_H_
-
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
-
-#endif /* _ASM_FB_H_ */
diff --git a/include/asm-sh64/fcntl.h b/include/asm-sh64/fcntl.h
deleted file mode 100644
index 744dd79b9d5d..000000000000
--- a/include/asm-sh64/fcntl.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sh/fcntl.h>
diff --git a/include/asm-sh64/futex.h b/include/asm-sh64/futex.h
deleted file mode 100644
index 6a332a9f099c..000000000000
--- a/include/asm-sh64/futex.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
-
-#include <asm-generic/futex.h>
-
-#endif
diff --git a/include/asm-sh64/gpio.h b/include/asm-sh64/gpio.h
deleted file mode 100644
index 6bc5a13d8415..000000000000
--- a/include/asm-sh64/gpio.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef __ASM_SH64_GPIO_H
-#define __ASM_SH64_GPIO_H
-
-/*
- * This is just a stub, so that every arch using sh-sci has a gpio.h
- */
-
-#endif /* __ASM_SH64_GPIO_H */
diff --git a/include/asm-sh64/hardirq.h b/include/asm-sh64/hardirq.h
deleted file mode 100644
index 555fd7a35108..000000000000
--- a/include/asm-sh64/hardirq.h
+++ /dev/null
@@ -1,18 +0,0 @@
-#ifndef __ASM_SH64_HARDIRQ_H
-#define __ASM_SH64_HARDIRQ_H
-
-#include <linux/threads.h>
-#include <linux/irq.h>
-
-/* entry.S is sensitive to the offsets of these fields */
-typedef struct {
-	unsigned int __softirq_pending;
-} ____cacheline_aligned irq_cpustat_t;
-
-#include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
-
-/* arch/sh64/kernel/irq.c */
-extern void ack_bad_irq(unsigned int irq);
-
-#endif /* __ASM_SH64_HARDIRQ_H */
-
diff --git a/include/asm-sh64/hw_irq.h b/include/asm-sh64/hw_irq.h
deleted file mode 100644
index ebb39089b0ac..000000000000
--- a/include/asm-sh64/hw_irq.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef __ASM_SH64_HW_IRQ_H
-#define __ASM_SH64_HW_IRQ_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/hw_irq.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#endif /* __ASM_SH64_HW_IRQ_H */
diff --git a/include/asm-sh64/ide.h b/include/asm-sh64/ide.h
deleted file mode 100644
index b6e31e8b9410..000000000000
--- a/include/asm-sh64/ide.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- *  linux/include/asm-sh64/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- *
- *  sh64 version by Richard Curnow & Paul Mundt
- */
-
-/*
- *  This file contains the sh64 architecture specific IDE code.
- */
-
-#ifndef __ASM_SH64_IDE_H
-#define __ASM_SH64_IDE_H
-
-#ifdef __KERNEL__
-
-
-/* Without this, the initialisation of PCI IDE cards end up calling
- * ide_init_hwif_ports, which won't work. */
-#ifdef CONFIG_BLK_DEV_IDEPCI
-#define ide_default_io_ctl(base)	(0)
-#endif
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH64_IDE_H */
diff --git a/include/asm-sh64/ioctl.h b/include/asm-sh64/ioctl.h
deleted file mode 100644
index b279fe06dfe5..000000000000
--- a/include/asm-sh64/ioctl.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/ioctl.h>
diff --git a/include/asm-sh64/ioctls.h b/include/asm-sh64/ioctls.h
deleted file mode 100644
index 6b0c04f63c57..000000000000
--- a/include/asm-sh64/ioctls.h
+++ /dev/null
@@ -1,116 +0,0 @@
-#ifndef __ASM_SH64_IOCTLS_H
-#define __ASM_SH64_IOCTLS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/ioctls.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2004  Richard Curnow
- *
- */
-
-#include <asm/ioctl.h>
-
-#define FIOCLEX		0x6601		/* _IO('f', 1) */
-#define FIONCLEX	0x6602		/* _IO('f', 2) */
-#define FIOASYNC	0x4004667d	/* _IOW('f', 125, int) */
-#define FIONBIO		0x4004667e	/* _IOW('f', 126, int) */
-#define FIONREAD	0x8004667f	/* _IOW('f', 127, int) */
-#define TIOCINQ		FIONREAD
-#define FIOQSIZE	0x80086680	/* _IOR('f', 128, loff_t) */
-
-#define TCGETS		0x5401
-#define TCSETS		0x5402
-#define TCSETSW		0x5403
-#define TCSETSF		0x5404
-
-#define TCGETA		0x80127417	/* _IOR('t', 23, struct termio) */
-#define TCSETA		0x40127418	/* _IOW('t', 24, struct termio) */
-#define TCSETAW		0x40127419	/* _IOW('t', 25, struct termio) */
-#define TCSETAF		0x4012741c	/* _IOW('t', 28, struct termio) */
-
-#define TCSBRK		0x741d		/* _IO('t', 29) */
-#define TCXONC		0x741e		/* _IO('t', 30) */
-#define TCFLSH		0x741f		/* _IO('t', 31) */
-
-#define TIOCSWINSZ	0x40087467	/* _IOW('t', 103, struct winsize) */
-#define TIOCGWINSZ	0x80087468	/* _IOR('t', 104, struct winsize) */
-#define	TIOCSTART	0x746e		/* _IO('t', 110)  start output, like ^Q */
-#define	TIOCSTOP	0x746f		/* _IO('t', 111)  stop output, like ^S */
-#define TIOCOUTQ        0x80047473	/* _IOR('t', 115, int) output queue size */
-
-#define TIOCSPGRP	0x40047476	/* _IOW('t', 118, int) */
-#define TIOCGPGRP	0x80047477	/* _IOR('t', 119, int) */
-
-#define TIOCEXCL	0x540c		/* _IO('T', 12) */
-#define TIOCNXCL	0x540d		/* _IO('T', 13) */
-#define TIOCSCTTY	0x540e		/* _IO('T', 14) */
-
-#define TIOCSTI		0x40015412	/* _IOW('T', 18, char) 0x5412 */
-#define TIOCMGET	0x80045415	/* _IOR('T', 21, unsigned int) 0x5415 */
-#define TIOCMBIS	0x40045416	/* _IOW('T', 22, unsigned int) 0x5416 */
-#define TIOCMBIC	0x40045417	/* _IOW('T', 23, unsigned int) 0x5417 */
-#define TIOCMSET	0x40045418	/* _IOW('T', 24, unsigned int) 0x5418 */
-
-#define TIOCM_LE	0x001
-#define TIOCM_DTR	0x002
-#define TIOCM_RTS	0x004
-#define TIOCM_ST	0x008
-#define TIOCM_SR	0x010
-#define TIOCM_CTS	0x020
-#define TIOCM_CAR	0x040
-#define TIOCM_RNG	0x080
-#define TIOCM_DSR	0x100
-#define TIOCM_CD	TIOCM_CAR
-#define TIOCM_RI	TIOCM_RNG
-
-#define TIOCGSOFTCAR	0x80045419	/* _IOR('T', 25, unsigned int) 0x5419 */
-#define TIOCSSOFTCAR	0x4004541a	/* _IOW('T', 26, unsigned int) 0x541A */
-#define TIOCLINUX	0x4004541c	/* _IOW('T', 28, char) 0x541C */
-#define TIOCCONS	0x541d		/* _IO('T', 29) */
-#define TIOCGSERIAL	0x803c541e	/* _IOR('T', 30, struct serial_struct) 0x541E */
-#define TIOCSSERIAL	0x403c541f	/* _IOW('T', 31, struct serial_struct) 0x541F */
-#define TIOCPKT		0x40045420	/* _IOW('T', 32, int) 0x5420 */
-
-#define TIOCPKT_DATA		 0
-#define TIOCPKT_FLUSHREAD	 1
-#define TIOCPKT_FLUSHWRITE	 2
-#define TIOCPKT_STOP		 4
-#define TIOCPKT_START		 8
-#define TIOCPKT_NOSTOP		16
-#define TIOCPKT_DOSTOP		32
-
-
-#define TIOCNOTTY	0x5422		/* _IO('T', 34) */
-#define TIOCSETD	0x40045423	/* _IOW('T', 35, int) 0x5423 */
-#define TIOCGETD	0x80045424	/* _IOR('T', 36, int) 0x5424 */
-#define TCSBRKP		0x40045424	/* _IOW('T', 37, int) 0x5425 */	/* Needed for POSIX tcsendbreak() */
-#define TIOCTTYGSTRUCT	0x8c105426	/* _IOR('T', 38, struct tty_struct) 0x5426 */ /* For debugging only */
-#define TIOCSBRK	0x5427		/* _IO('T', 39) */ /* BSD compatibility */
-#define TIOCCBRK	0x5428		/* _IO('T', 40) */ /* BSD compatibility */
-#define TIOCGSID	0x80045429	/* _IOR('T', 41, pid_t) 0x5429 */ /* Return the session ID of FD */
-#define TIOCGPTN	0x80045430	/* _IOR('T',0x30, unsigned int) 0x5430 Get Pty Number (of pty-mux device) */
-#define TIOCSPTLCK	0x40045431	/* _IOW('T',0x31, int) Lock/unlock Pty */
-
-#define TIOCSERCONFIG	0x5453		/* _IO('T', 83) */
-#define TIOCSERGWILD	0x80045454	/* _IOR('T', 84,  int) 0x5454 */
-#define TIOCSERSWILD	0x40045455	/* _IOW('T', 85,  int) 0x5455 */
-#define TIOCGLCKTRMIOS	0x5456
-#define TIOCSLCKTRMIOS	0x5457
-#define TIOCSERGSTRUCT	0x80d85458	/* _IOR('T', 88, struct async_struct) 0x5458 */ /* For debugging only */
-#define TIOCSERGETLSR   0x80045459	/* _IOR('T', 89, unsigned int) 0x5459 */ /* Get line status register */
-
-/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
-#define TIOCSER_TEMT    0x01	/* Transmitter physically empty */
-
-#define TIOCSERGETMULTI 0x80a8545a	/* _IOR('T', 90, struct serial_multiport_struct) 0x545A */ /* Get multiport config  */
-#define TIOCSERSETMULTI 0x40a8545b	/* _IOW('T', 91, struct serial_multiport_struct) 0x545B */ /* Set multiport config */
-
-#define TIOCMIWAIT	0x545c		/* _IO('T', 92) wait for a change on serial input line(s) */
-#define TIOCGICOUNT	0x545d		/* read serial port inline interrupt counts */
-
-#endif /* __ASM_SH64_IOCTLS_H */
diff --git a/include/asm-sh64/ipcbuf.h b/include/asm-sh64/ipcbuf.h
deleted file mode 100644
index c441e35299c0..000000000000
--- a/include/asm-sh64/ipcbuf.h
+++ /dev/null
@@ -1,40 +0,0 @@
-#ifndef __ASM_SH64_IPCBUF_H__
-#define __ASM_SH64_IPCBUF_H__
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/ipcbuf.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * The ipc64_perm structure for i386 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 32-bit mode_t and seq
- * - 2 miscellaneous 32-bit values
- */
-
-struct ipc64_perm
-{
-	__kernel_key_t		key;
-	__kernel_uid32_t	uid;
-	__kernel_gid32_t	gid;
-	__kernel_uid32_t	cuid;
-	__kernel_gid32_t	cgid;
-	__kernel_mode_t		mode;
-	unsigned short		__pad1;
-	unsigned short		seq;
-	unsigned short		__pad2;
-	unsigned long		__unused1;
-	unsigned long		__unused2;
-};
-
-#endif /* __ASM_SH64_IPCBUF_H__ */
diff --git a/include/asm-sh64/irq_regs.h b/include/asm-sh64/irq_regs.h
deleted file mode 100644
index 3dd9c0b70270..000000000000
--- a/include/asm-sh64/irq_regs.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/irq_regs.h>
diff --git a/include/asm-sh64/kdebug.h b/include/asm-sh64/kdebug.h
deleted file mode 100644
index 6ece1b037665..000000000000
--- a/include/asm-sh64/kdebug.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/kdebug.h>
diff --git a/include/asm-sh64/kmap_types.h b/include/asm-sh64/kmap_types.h
deleted file mode 100644
index 2ae7c7587919..000000000000
--- a/include/asm-sh64/kmap_types.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_SH64_KMAP_TYPES_H
-#define __ASM_SH64_KMAP_TYPES_H
-
-#include <asm-sh/kmap_types.h>
-
-#endif /* __ASM_SH64_KMAP_TYPES_H */
-
diff --git a/include/asm-sh64/linkage.h b/include/asm-sh64/linkage.h
deleted file mode 100644
index 1dd0e84a228d..000000000000
--- a/include/asm-sh64/linkage.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_SH64_LINKAGE_H
-#define __ASM_SH64_LINKAGE_H
-
-#include <asm-sh/linkage.h>
-
-#endif /* __ASM_SH64_LINKAGE_H */
-
diff --git a/include/asm-sh64/local.h b/include/asm-sh64/local.h
deleted file mode 100644
index d9bd95dd36e2..000000000000
--- a/include/asm-sh64/local.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_SH64_LOCAL_H
-#define __ASM_SH64_LOCAL_H
-
-#include <asm-generic/local.h>
-
-#endif /* __ASM_SH64_LOCAL_H */
-
diff --git a/include/asm-sh64/mc146818rtc.h b/include/asm-sh64/mc146818rtc.h
deleted file mode 100644
index 6cd3aec68dbe..000000000000
--- a/include/asm-sh64/mc146818rtc.h
+++ /dev/null
@@ -1,7 +0,0 @@
-/*
- * linux/include/asm-sh64/mc146818rtc.h
- *
-*/
-
-/* For now, an empty place-holder to get IDE to compile. */
-
diff --git a/include/asm-sh64/mman.h b/include/asm-sh64/mman.h
deleted file mode 100644
index a9be6d885c3e..000000000000
--- a/include/asm-sh64/mman.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_MMAN_H
-#define __ASM_SH64_MMAN_H
-
-#include <asm-sh/mman.h>
-
-#endif /* __ASM_SH64_MMAN_H */
diff --git a/include/asm-sh64/mmu.h b/include/asm-sh64/mmu.h
deleted file mode 100644
index ccd36d26615a..000000000000
--- a/include/asm-sh64/mmu.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __MMU_H
-#define __MMU_H
-
-/* Default "unsigned long" context */
-typedef unsigned long mm_context_t;
-
-#endif
diff --git a/include/asm-sh64/module.h b/include/asm-sh64/module.h
deleted file mode 100644
index c313650d3d93..000000000000
--- a/include/asm-sh64/module.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef __ASM_SH64_MODULE_H
-#define __ASM_SH64_MODULE_H
-/*
- * This file contains the SH architecture specific module code.
- */
-
-struct mod_arch_specific {
-	/* empty */
-};
-
-#define Elf_Shdr		Elf32_Shdr
-#define Elf_Sym			Elf32_Sym
-#define Elf_Ehdr		Elf32_Ehdr
-
-#define module_map(x)		vmalloc(x)
-#define module_unmap(x)		vfree(x)
-#define module_arch_init(x)	(0)
-#define arch_init_modules(x)	do { } while (0)
-
-#endif /* __ASM_SH64_MODULE_H */
diff --git a/include/asm-sh64/msgbuf.h b/include/asm-sh64/msgbuf.h
deleted file mode 100644
index cf0494ce0ba8..000000000000
--- a/include/asm-sh64/msgbuf.h
+++ /dev/null
@@ -1,42 +0,0 @@
-#ifndef __ASM_SH64_MSGBUF_H
-#define __ASM_SH64_MSGBUF_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/msgbuf.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * The msqid64_ds structure for i386 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct msqid64_ds {
-	struct ipc64_perm msg_perm;
-	__kernel_time_t msg_stime;	/* last msgsnd time */
-	unsigned long	__unused1;
-	__kernel_time_t msg_rtime;	/* last msgrcv time */
-	unsigned long	__unused2;
-	__kernel_time_t msg_ctime;	/* last change time */
-	unsigned long	__unused3;
-	unsigned long  msg_cbytes;	/* current number of bytes on queue */
-	unsigned long  msg_qnum;	/* number of messages in queue */
-	unsigned long  msg_qbytes;	/* max number of bytes on queue */
-	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
-	__kernel_pid_t msg_lrpid;	/* last receive pid */
-	unsigned long  __unused4;
-	unsigned long  __unused5;
-};
-
-#endif /* __ASM_SH64_MSGBUF_H */
diff --git a/include/asm-sh64/mutex.h b/include/asm-sh64/mutex.h
deleted file mode 100644
index 458c1f7fbc18..000000000000
--- a/include/asm-sh64/mutex.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/*
- * Pull in the generic implementation for the mutex fastpath.
- *
- * TODO: implement optimized primitives instead, or leave the generic
- * implementation in place, or pick the atomic_xchg() based generic
- * implementation. (see asm-generic/mutex-xchg.h for details)
- */
-
-#include <asm-generic/mutex-dec.h>
diff --git a/include/asm-sh64/namei.h b/include/asm-sh64/namei.h
deleted file mode 100644
index 99d759a805ce..000000000000
--- a/include/asm-sh64/namei.h
+++ /dev/null
@@ -1,24 +0,0 @@
-#ifndef __ASM_SH64_NAMEI_H
-#define __ASM_SH64_NAMEI_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/namei.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- * Included from linux/fs/namei.c
- *
- */
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __ASM_SH64_NAMEI_H */
diff --git a/include/asm-sh64/page.h b/include/asm-sh64/page.h
deleted file mode 100644
index 472089aefc60..000000000000
--- a/include/asm-sh64/page.h
+++ /dev/null
@@ -1,119 +0,0 @@
-#ifndef __ASM_SH64_PAGE_H
-#define __ASM_SH64_PAGE_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/page.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- *
- * benedict.gaster@superh.com 19th, 24th July 2002.
- *
- * Modified to take account of enabling for D-CACHE support.
- *
- */
-
-
-/* PAGE_SHIFT determines the page size */
-#define PAGE_SHIFT	12
-#ifdef __ASSEMBLY__
-#define PAGE_SIZE	4096
-#else
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
-#endif
-#define PAGE_MASK	(~(PAGE_SIZE-1))
-#define PTE_MASK	PAGE_MASK
-
-#if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
-#define HPAGE_SHIFT	16
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_1MB)
-#define HPAGE_SHIFT	20
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512MB)
-#define HPAGE_SHIFT	29
-#endif
-
-#ifdef CONFIG_HUGETLB_PAGE
-#define HPAGE_SIZE		(1UL << HPAGE_SHIFT)
-#define HPAGE_MASK		(~(HPAGE_SIZE-1))
-#define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT-PAGE_SHIFT)
-#define ARCH_HAS_SETCLEAR_HUGE_PTE
-#endif
-
-#ifdef __KERNEL__
-#ifndef __ASSEMBLY__
-
-extern struct page *mem_map;
-extern void sh64_page_clear(void *page);
-extern void sh64_page_copy(void *from, void *to);
-
-#define clear_page(page)               sh64_page_clear(page)
-#define copy_page(to,from)             sh64_page_copy(from, to)
-
-#if defined(CONFIG_DCACHE_DISABLED)
-
-#define clear_user_page(page, vaddr, pg)	clear_page(page)
-#define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
-
-#else
-
-extern void clear_user_page(void *to, unsigned long address, struct page *pg);
-extern void copy_user_page(void *to, void *from, unsigned long address, struct page *pg);
-
-#endif /* defined(CONFIG_DCACHE_DISABLED) */
-
-/*
- * These are used to make use of C type-checking..
- */
-typedef struct { unsigned long long pte; } pte_t;
-typedef struct { unsigned long pmd; } pmd_t;
-typedef struct { unsigned long pgd; } pgd_t;
-typedef struct { unsigned long pgprot; } pgprot_t;
-
-#define pte_val(x)	((x).pte)
-#define pmd_val(x)	((x).pmd)
-#define pgd_val(x)	((x).pgd)
-#define pgprot_val(x)	((x).pgprot)
-
-#define __pte(x) ((pte_t) { (x) } )
-#define __pmd(x) ((pmd_t) { (x) } )
-#define __pgd(x) ((pgd_t) { (x) } )
-#define __pgprot(x)	((pgprot_t) { (x) } )
-
-#endif /* !__ASSEMBLY__ */
-
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
-/*
- * Kconfig defined.
- */
-#define __MEMORY_START		(CONFIG_MEMORY_START)
-#define PAGE_OFFSET		(CONFIG_CACHED_MEMORY_OFFSET)
-
-#define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
-#define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
-#define MAP_NR(addr)		((__pa(addr)-__MEMORY_START) >> PAGE_SHIFT)
-#define VALID_PAGE(page)	((page - mem_map) < max_mapnr)
-
-#define phys_to_page(phys)	(mem_map + (((phys) - __MEMORY_START) >> PAGE_SHIFT))
-#define page_to_phys(page)	(((page - mem_map) << PAGE_SHIFT) + __MEMORY_START)
-
-/* PFN start number, because of __MEMORY_START */
-#define PFN_START		(__MEMORY_START >> PAGE_SHIFT)
-#define ARCH_PFN_OFFSET		(PFN_START)
-#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
-#define pfn_valid(pfn)		(((pfn) - PFN_START) < max_mapnr)
-#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
-
-#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
-				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
-
-#include <asm-generic/memory_model.h>
-#include <asm-generic/page.h>
-
-#endif /* __KERNEL__ */
-#endif /* __ASM_SH64_PAGE_H */
diff --git a/include/asm-sh64/param.h b/include/asm-sh64/param.h
deleted file mode 100644
index f409adb41540..000000000000
--- a/include/asm-sh64/param.h
+++ /dev/null
@@ -1,42 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/param.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-#ifndef __ASM_SH64_PARAM_H
-#define __ASM_SH64_PARAM_H
-
-
-#ifdef __KERNEL__
-# ifdef CONFIG_SH_WDT
-#  define HZ		1000		/* Needed for high-res WOVF */
-# else
-#  define HZ		100
-# endif
-# define USER_HZ	100		/* User interfaces are in "ticks" */
-# define CLOCKS_PER_SEC	(USER_HZ)	/* frequency at which times() counts */
-#endif
-
-#ifndef HZ
-#define HZ 100
-#endif
-
-#define EXEC_PAGESIZE	4096
-
-#ifndef NGROUPS
-#define NGROUPS		32
-#endif
-
-#ifndef NOGROUP
-#define NOGROUP		(-1)
-#endif
-
-#define MAXHOSTNAMELEN	64	/* max length of hostname */
-
-#endif /* __ASM_SH64_PARAM_H */
diff --git a/include/asm-sh64/percpu.h b/include/asm-sh64/percpu.h
deleted file mode 100644
index a01d16cd0e8c..000000000000
--- a/include/asm-sh64/percpu.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_PERCPU
-#define __ASM_SH64_PERCPU
-
-#include <asm-generic/percpu.h>
-
-#endif /* __ASM_SH64_PERCPU */
diff --git a/include/asm-sh64/pgalloc.h b/include/asm-sh64/pgalloc.h
deleted file mode 100644
index 6eccab770a6d..000000000000
--- a/include/asm-sh64/pgalloc.h
+++ /dev/null
@@ -1,125 +0,0 @@
-#ifndef __ASM_SH64_PGALLOC_H
-#define __ASM_SH64_PGALLOC_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/pgalloc.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003, 2004  Paul Mundt
- * Copyright (C) 2003, 2004  Richard Curnow
- *
- */
-
-#include <linux/mm.h>
-#include <linux/quicklist.h>
-#include <asm/page.h>
-
-static inline void pgd_init(unsigned long page)
-{
-	unsigned long *pgd = (unsigned long *)page;
-	extern pte_t empty_bad_pte_table[PTRS_PER_PTE];
-	int i;
-
-	for (i = 0; i < USER_PTRS_PER_PGD; i++)
-		pgd[i] = (unsigned long)empty_bad_pte_table;
-}
-
-/*
- * Allocate and free page tables. The xxx_kernel() versions are
- * used to allocate a kernel page table - this turns on ASN bits
- * if any.
- */
-
-static inline pgd_t *get_pgd_slow(void)
-{
-	unsigned int pgd_size = (USER_PTRS_PER_PGD * sizeof(pgd_t));
-	pgd_t *ret = kmalloc(pgd_size, GFP_KERNEL);
-	return ret;
-}
-
-static inline pgd_t *pgd_alloc(struct mm_struct *mm)
-{
-	return quicklist_alloc(0, GFP_KERNEL, NULL);
-}
-
-static inline void pgd_free(pgd_t *pgd)
-{
-	quicklist_free(0, NULL, pgd);
-}
-
-static inline struct page *pte_alloc_one(struct mm_struct *mm,
-					 unsigned long address)
-{
-	void *pg = quicklist_alloc(0, GFP_KERNEL, NULL);
-	return pg ? virt_to_page(pg) : NULL;
-}
-
-static inline void pte_free_kernel(pte_t *pte)
-{
-	quicklist_free(0, NULL, pte);
-}
-
-static inline void pte_free(struct page *pte)
-{
-	quicklist_free_page(0, NULL, pte);
-}
-
-static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
-					   unsigned long address)
-{
-	return quicklist_alloc(0, GFP_KERNEL, NULL);
-}
-
-#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
-
-/*
- * allocating and freeing a pmd is trivial: the 1-entry pmd is
- * inside the pgd, so has no extra memory associated with it.
- */
-
-#if defined(CONFIG_SH64_PGTABLE_2_LEVEL)
-
-#define pmd_alloc_one(mm, addr)		({ BUG(); ((pmd_t *)2); })
-#define pmd_free(x)			do { } while (0)
-#define pgd_populate(mm, pmd, pte)	BUG()
-#define __pte_free_tlb(tlb,pte)		tlb_remove_page((tlb),(pte))
-#define __pmd_free_tlb(tlb,pmd)		do { } while (0)
-
-#elif defined(CONFIG_SH64_PGTABLE_3_LEVEL)
-
-static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long address)
-{
-	return quicklist_alloc(0, GFP_KERNEL, NULL);
-}
-
-static inline void pmd_free(pmd_t *pmd)
-{
-	quicklist_free(0, NULL, pmd);
-}
-
-#define pgd_populate(mm, pgd, pmd)	pgd_set(pgd, pmd)
-#define __pmd_free_tlb(tlb,pmd)		pmd_free(pmd)
-
-#else
-#error "No defined page table size"
-#endif
-
-#define pmd_populate_kernel(mm, pmd, pte) \
-	set_pmd(pmd, __pmd(_PAGE_TABLE + (unsigned long) (pte)))
-
-static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
-				struct page *pte)
-{
-	set_pmd(pmd, __pmd(_PAGE_TABLE + (unsigned long) page_address (pte)));
-}
-
-static inline void check_pgt_cache(void)
-{
-	quicklist_trim(0, NULL, 25, 16);
-}
-
-#endif /* __ASM_SH64_PGALLOC_H */
diff --git a/include/asm-sh64/poll.h b/include/asm-sh64/poll.h
deleted file mode 100644
index ca2950267c53..000000000000
--- a/include/asm-sh64/poll.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef __ASM_SH64_POLL_H
-#define __ASM_SH64_POLL_H
-
-#include <asm-generic/poll.h>
-
-#undef POLLREMOVE
-
-#endif /* __ASM_SH64_POLL_H */
diff --git a/include/asm-sh64/posix_types.h b/include/asm-sh64/posix_types.h
deleted file mode 100644
index 0620317a6f0f..000000000000
--- a/include/asm-sh64/posix_types.h
+++ /dev/null
@@ -1,131 +0,0 @@
-#ifndef __ASM_SH64_POSIX_TYPES_H
-#define __ASM_SH64_POSIX_TYPES_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/posix_types.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is generally used by user-level software, so you need to
- * be a little careful about namespace pollution etc.  Also, we cannot
- * assume GCC is being used.
- */
-
-typedef unsigned long	__kernel_ino_t;
-typedef unsigned short	__kernel_mode_t;
-typedef unsigned short	__kernel_nlink_t;
-typedef long		__kernel_off_t;
-typedef int		__kernel_pid_t;
-typedef unsigned short	__kernel_ipc_pid_t;
-typedef unsigned short	__kernel_uid_t;
-typedef unsigned short	__kernel_gid_t;
-typedef long unsigned int	__kernel_size_t;
-typedef int		__kernel_ssize_t;
-typedef int		__kernel_ptrdiff_t;
-typedef long		__kernel_time_t;
-typedef long		__kernel_suseconds_t;
-typedef long		__kernel_clock_t;
-typedef int		__kernel_timer_t;
-typedef int		__kernel_clockid_t;
-typedef int		__kernel_daddr_t;
-typedef char *		__kernel_caddr_t;
-typedef unsigned short	__kernel_uid16_t;
-typedef unsigned short	__kernel_gid16_t;
-typedef unsigned int	__kernel_uid32_t;
-typedef unsigned int	__kernel_gid32_t;
-
-typedef unsigned short	__kernel_old_uid_t;
-typedef unsigned short	__kernel_old_gid_t;
-typedef unsigned short	__kernel_old_dev_t;
-
-#ifdef __GNUC__
-typedef long long	__kernel_loff_t;
-#endif
-
-typedef struct {
-#if defined(__KERNEL__) || defined(__USE_ALL)
-	int	val[2];
-#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
-	int	__val[2];
-#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
-} __kernel_fsid_t;
-
-#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2)
-
-#undef	__FD_SET
-static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
-{
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	__fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
-}
-
-#undef	__FD_CLR
-static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
-{
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	__fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
-}
-
-
-#undef	__FD_ISSET
-static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
-{
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
-}
-
-/*
- * This will unroll the loop for the normal constant case (8 ints,
- * for a 256-bit fd_set)
- */
-#undef	__FD_ZERO
-static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
-{
-	unsigned long *__tmp = __p->fds_bits;
-	int __i;
-
-	if (__builtin_constant_p(__FDSET_LONGS)) {
-		switch (__FDSET_LONGS) {
-		case 16:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			__tmp[ 4] = 0; __tmp[ 5] = 0;
-			__tmp[ 6] = 0; __tmp[ 7] = 0;
-			__tmp[ 8] = 0; __tmp[ 9] = 0;
-			__tmp[10] = 0; __tmp[11] = 0;
-			__tmp[12] = 0; __tmp[13] = 0;
-			__tmp[14] = 0; __tmp[15] = 0;
-			return;
-
-		case 8:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			__tmp[ 4] = 0; __tmp[ 5] = 0;
-			__tmp[ 6] = 0; __tmp[ 7] = 0;
-			return;
-
-		case 4:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			return;
-		}
-	}
-	__i = __FDSET_LONGS;
-	while (__i) {
-		__i--;
-		*__tmp = 0;
-		__tmp++;
-	}
-}
-
-#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
-
-#endif /* __ASM_SH64_POSIX_TYPES_H */
diff --git a/include/asm-sh64/ptrace.h b/include/asm-sh64/ptrace.h
deleted file mode 100644
index c424f80e3ae0..000000000000
--- a/include/asm-sh64/ptrace.h
+++ /dev/null
@@ -1,35 +0,0 @@
-#ifndef __ASM_SH64_PTRACE_H
-#define __ASM_SH64_PTRACE_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/ptrace.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * This struct defines the way the registers are stored on the
- * kernel stack during a system call or other kernel entry.
- */
-struct pt_regs {
-	unsigned long long pc;
-	unsigned long long sr;
-	unsigned long long syscall_nr;
-	unsigned long long regs[63];
-	unsigned long long tregs[8];
-	unsigned long long pad[2];
-};
-
-#ifdef __KERNEL__
-#define user_mode(regs) (((regs)->sr & 0x40000000)==0)
-#define instruction_pointer(regs) ((regs)->pc)
-#define profile_pc(regs) ((unsigned long)instruction_pointer(regs))
-extern void show_regs(struct pt_regs *);
-#endif
-
-#endif /* __ASM_SH64_PTRACE_H */
diff --git a/include/asm-sh64/resource.h b/include/asm-sh64/resource.h
deleted file mode 100644
index 8ff93944ae66..000000000000
--- a/include/asm-sh64/resource.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_RESOURCE_H
-#define __ASM_SH64_RESOURCE_H
-
-#include <asm-sh/resource.h>
-
-#endif /* __ASM_SH64_RESOURCE_H */
diff --git a/include/asm-sh64/scatterlist.h b/include/asm-sh64/scatterlist.h
deleted file mode 100644
index 7f729bbfce43..000000000000
--- a/include/asm-sh64/scatterlist.h
+++ /dev/null
@@ -1,37 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/scatterlist.h
- *
- * Copyright (C) 2003  Paul Mundt
- *
- */
-#ifndef __ASM_SH64_SCATTERLIST_H
-#define __ASM_SH64_SCATTERLIST_H
-
-#include <asm/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-    unsigned long sg_magic;
-#endif
-    unsigned long page_link;
-    unsigned int offset;/* for highmem, page offset */
-    dma_addr_t dma_address;
-    unsigned int length;
-};
-
-/* These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns, or alternatively stop on the first sg_dma_len(sg) which
- * is 0.
- */
-#define sg_dma_address(sg)	((sg)->dma_address)
-#define sg_dma_len(sg)		((sg)->length)
-
-#define ISA_DMA_THRESHOLD (0xffffffff)
-
-#endif /* !__ASM_SH64_SCATTERLIST_H */
diff --git a/include/asm-sh64/sci.h b/include/asm-sh64/sci.h
deleted file mode 100644
index 793c568b7820..000000000000
--- a/include/asm-sh64/sci.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sh/sci.h>
diff --git a/include/asm-sh64/sections.h b/include/asm-sh64/sections.h
deleted file mode 100644
index 897f36bcdf85..000000000000
--- a/include/asm-sh64/sections.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_SH64_SECTIONS_H
-#define __ASM_SH64_SECTIONS_H
-
-#include <asm-sh/sections.h>
-
-#endif /* __ASM_SH64_SECTIONS_H */
-
diff --git a/include/asm-sh64/segment.h b/include/asm-sh64/segment.h
deleted file mode 100644
index 92ac001fc483..000000000000
--- a/include/asm-sh64/segment.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_SEGMENT_H
-#define _ASM_SEGMENT_H
-
-/* Only here because we have some old header files that expect it.. */
-
-#endif /* _ASM_SEGMENT_H */
diff --git a/include/asm-sh64/semaphore-helper.h b/include/asm-sh64/semaphore-helper.h
deleted file mode 100644
index fcfafe263e86..000000000000
--- a/include/asm-sh64/semaphore-helper.h
+++ /dev/null
@@ -1,101 +0,0 @@
-#ifndef __ASM_SH64_SEMAPHORE_HELPER_H
-#define __ASM_SH64_SEMAPHORE_HELPER_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/semaphore-helper.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-#include <asm/errno.h>
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- * (C) Copyright 1999 Andrea Arcangeli
- */
-
-/*
- * These two _must_ execute atomically wrt each other.
- *
- * This is trivially done with load_locked/store_cond,
- * which we have.  Let the rest of the losers suck eggs.
- */
-static __inline__ void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc((atomic_t *)&sem->sleepers);
-}
-
-static __inline__ int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- *
- * We must undo the sem->count down_trylock() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->sleepers--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif /* __ASM_SH64_SEMAPHORE_HELPER_H */
diff --git a/include/asm-sh64/semaphore.h b/include/asm-sh64/semaphore.h
deleted file mode 100644
index f027cc14b55b..000000000000
--- a/include/asm-sh64/semaphore.h
+++ /dev/null
@@ -1,119 +0,0 @@
-#ifndef __ASM_SH64_SEMAPHORE_H
-#define __ASM_SH64_SEMAPHORE_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/semaphore.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- * SMP- and interrupt-safe semaphores.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * SuperH verison by Niibe Yutaka
- *  (Currently no asm implementation but generic C code...)
- *
- */
-
-#include <linux/linkage.h>
-#include <linux/spinlock.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-/*
- *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
- *
- * i'd rather use the more flexible initialization above, but sadly
- * GCC 2.7.2.3 emits a bogus warning. EGCS doesnt. Oh well.
- */
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-#if 0
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-#endif
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-static inline void down(struct semaphore * sem)
-{
-	if (atomic_dec_return(&sem->count) < 0)
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	int ret = 0;
-
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_trylock(sem);
-	return ret;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- */
-static inline void up(struct semaphore * sem)
-{
-	if (atomic_inc_return(&sem->count) <= 0)
-		__up(sem);
-}
-
-#endif /* __ASM_SH64_SEMAPHORE_H */
diff --git a/include/asm-sh64/sembuf.h b/include/asm-sh64/sembuf.h
deleted file mode 100644
index ec4d9f143577..000000000000
--- a/include/asm-sh64/sembuf.h
+++ /dev/null
@@ -1,36 +0,0 @@
-#ifndef __ASM_SH64_SEMBUF_H
-#define __ASM_SH64_SEMBUF_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/sembuf.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * The semid64_ds structure for i386 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct semid64_ds {
-	struct ipc64_perm sem_perm;		/* permissions .. see ipc.h */
-	__kernel_time_t	sem_otime;		/* last semop time */
-	unsigned long	__unused1;
-	__kernel_time_t	sem_ctime;		/* last change time */
-	unsigned long	__unused2;
-	unsigned long	sem_nsems;		/* no. of semaphores in array */
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-};
-
-#endif /* __ASM_SH64_SEMBUF_H */
diff --git a/include/asm-sh64/serial.h b/include/asm-sh64/serial.h
deleted file mode 100644
index e8d7b3f2da57..000000000000
--- a/include/asm-sh64/serial.h
+++ /dev/null
@@ -1,31 +0,0 @@
-/*
- * include/asm-sh64/serial.h
- *
- * Configuration details for 8250, 16450, 16550, etc. serial ports
- */
-
-#ifndef _ASM_SERIAL_H
-#define _ASM_SERIAL_H
-
-/*
- * This assumes you have a 1.8432 MHz clock for your UART.
- *
- * It'd be nice if someone built a serial card with a 24.576 MHz
- * clock, since the 16550A is capable of handling a top speed of 1.5
- * megabits/second; but this requires the faster clock.
- */
-#define BASE_BAUD ( 1843200 / 16 )
-
-#define RS_TABLE_SIZE  2
-
-#define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST)
-
-#define SERIAL_PORT_DFNS			\
-	/* UART CLK   PORT IRQ     FLAGS        */			\
-	{ 0, BASE_BAUD, 0x3F8, 4, STD_COM_FLAGS },	/* ttyS0 */	\
-	{ 0, BASE_BAUD, 0x2F8, 3, STD_COM_FLAGS }	/* ttyS1 */
-
-/* XXX: This should be moved ino irq.h */
-#define irq_cannonicalize(x) (x)
-
-#endif /* _ASM_SERIAL_H */
diff --git a/include/asm-sh64/setup.h b/include/asm-sh64/setup.h
deleted file mode 100644
index 5b07b14c2927..000000000000
--- a/include/asm-sh64/setup.h
+++ /dev/null
@@ -1,22 +0,0 @@
-#ifndef __ASM_SH64_SETUP_H
-#define __ASM_SH64_SETUP_H
-
-#define COMMAND_LINE_SIZE 256
-
-#ifdef __KERNEL__
-
-#define PARAM ((unsigned char *)empty_zero_page)
-#define MOUNT_ROOT_RDONLY (*(unsigned long *) (PARAM+0x000))
-#define RAMDISK_FLAGS (*(unsigned long *) (PARAM+0x004))
-#define ORIG_ROOT_DEV (*(unsigned long *) (PARAM+0x008))
-#define LOADER_TYPE (*(unsigned long *) (PARAM+0x00c))
-#define INITRD_START (*(unsigned long *) (PARAM+0x010))
-#define INITRD_SIZE (*(unsigned long *) (PARAM+0x014))
-
-#define COMMAND_LINE ((char *) (PARAM+256))
-#define COMMAND_LINE_SIZE 256
-
-#endif  /*  __KERNEL__  */
-
-#endif /* __ASM_SH64_SETUP_H */
-
diff --git a/include/asm-sh64/shmbuf.h b/include/asm-sh64/shmbuf.h
deleted file mode 100644
index 022f3494dd64..000000000000
--- a/include/asm-sh64/shmbuf.h
+++ /dev/null
@@ -1,53 +0,0 @@
-#ifndef __ASM_SH64_SHMBUF_H
-#define __ASM_SH64_SHMBUF_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/shmbuf.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/*
- * The shmid64_ds structure for i386 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct shmid64_ds {
-	struct ipc64_perm	shm_perm;	/* operation perms */
-	size_t			shm_segsz;	/* size of segment (bytes) */
-	__kernel_time_t		shm_atime;	/* last attach time */
-	unsigned long		__unused1;
-	__kernel_time_t		shm_dtime;	/* last detach time */
-	unsigned long		__unused2;
-	__kernel_time_t		shm_ctime;	/* last change time */
-	unsigned long		__unused3;
-	__kernel_pid_t		shm_cpid;	/* pid of creator */
-	__kernel_pid_t		shm_lpid;	/* pid of last operator */
-	unsigned long		shm_nattch;	/* no. of current attaches */
-	unsigned long		__unused4;
-	unsigned long		__unused5;
-};
-
-struct shminfo64 {
-	unsigned long	shmmax;
-	unsigned long	shmmin;
-	unsigned long	shmmni;
-	unsigned long	shmseg;
-	unsigned long	shmall;
-	unsigned long	__unused1;
-	unsigned long	__unused2;
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-};
-
-#endif /* __ASM_SH64_SHMBUF_H */
diff --git a/include/asm-sh64/shmparam.h b/include/asm-sh64/shmparam.h
deleted file mode 100644
index 1bb820c833ee..000000000000
--- a/include/asm-sh64/shmparam.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef __ASM_SH64_SHMPARAM_H
-#define __ASM_SH64_SHMPARAM_H
-
-/*
- * Set this to a sensible safe default, we'll work out the specifics for the
- * align mask from the cache descriptor at run-time.
- */
-#define	SHMLBA	0x4000
-
-#define __ARCH_FORCE_SHMLBA
-
-#endif /* __ASM_SH64_SHMPARAM_H */
diff --git a/include/asm-sh64/sigcontext.h b/include/asm-sh64/sigcontext.h
deleted file mode 100644
index 6293509d8cc1..000000000000
--- a/include/asm-sh64/sigcontext.h
+++ /dev/null
@@ -1,30 +0,0 @@
-#ifndef __ASM_SH64_SIGCONTEXT_H
-#define __ASM_SH64_SIGCONTEXT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/sigcontext.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-struct sigcontext {
-	unsigned long	oldmask;
-
-	/* CPU registers */
-	unsigned long long sc_regs[63];
-	unsigned long long sc_tregs[8];
-	unsigned long long sc_pc;
-	unsigned long long sc_sr;
-
-	/* FPU registers */
-	unsigned long long sc_fpregs[32];
-	unsigned int sc_fpscr;
-	unsigned int sc_fpvalid;
-};
-
-#endif /* __ASM_SH64_SIGCONTEXT_H */
diff --git a/include/asm-sh64/siginfo.h b/include/asm-sh64/siginfo.h
deleted file mode 100644
index 56ef1da534d7..000000000000
--- a/include/asm-sh64/siginfo.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_SIGINFO_H
-#define __ASM_SH64_SIGINFO_H
-
-#include <asm-generic/siginfo.h>
-
-#endif /* __ASM_SH64_SIGINFO_H */
diff --git a/include/asm-sh64/signal.h b/include/asm-sh64/signal.h
deleted file mode 100644
index 244e134730d9..000000000000
--- a/include/asm-sh64/signal.h
+++ /dev/null
@@ -1,159 +0,0 @@
-#ifndef __ASM_SH64_SIGNAL_H
-#define __ASM_SH64_SIGNAL_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/signal.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <linux/types.h>
-
-/* Avoid too many header ordering problems.  */
-struct siginfo;
-
-#define _NSIG		64
-#define _NSIG_BPW	32
-#define _NSIG_WORDS	(_NSIG / _NSIG_BPW)
-
-typedef unsigned long old_sigset_t;		/* at least 32 bits */
-
-typedef struct {
-	unsigned long sig[_NSIG_WORDS];
-} sigset_t;
-
-#define SIGHUP		 1
-#define SIGINT		 2
-#define SIGQUIT		 3
-#define SIGILL		 4
-#define SIGTRAP		 5
-#define SIGABRT		 6
-#define SIGIOT		 6
-#define SIGBUS		 7
-#define SIGFPE		 8
-#define SIGKILL		 9
-#define SIGUSR1		10
-#define SIGSEGV		11
-#define SIGUSR2		12
-#define SIGPIPE		13
-#define SIGALRM		14
-#define SIGTERM		15
-#define SIGSTKFLT	16
-#define SIGCHLD		17
-#define SIGCONT		18
-#define SIGSTOP		19
-#define SIGTSTP		20
-#define SIGTTIN		21
-#define SIGTTOU		22
-#define SIGURG		23
-#define SIGXCPU		24
-#define SIGXFSZ		25
-#define SIGVTALRM	26
-#define SIGPROF		27
-#define SIGWINCH	28
-#define SIGIO		29
-#define SIGPOLL		SIGIO
-/*
-#define SIGLOST		29
-*/
-#define SIGPWR		30
-#define SIGSYS		31
-#define	SIGUNUSED	31
-
-/* These should not be considered constants from userland.  */
-#define SIGRTMIN	32
-#define SIGRTMAX	(_NSIG-1)
-
-/*
- * SA_FLAGS values:
- *
- * SA_ONSTACK indicates that a registered stack_t will be used.
- * SA_RESTART flag to get restarting signals (which were the default long ago)
- * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop.
- * SA_RESETHAND clears the handler when the signal is delivered.
- * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies.
- * SA_NODEFER prevents the current signal from being masked in the handler.
- *
- * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single
- * Unix names RESETHAND and NODEFER respectively.
- */
-#define SA_NOCLDSTOP	0x00000001
-#define SA_NOCLDWAIT	0x00000002 /* not supported yet */
-#define SA_SIGINFO	0x00000004
-#define SA_ONSTACK	0x08000000
-#define SA_RESTART	0x10000000
-#define SA_NODEFER	0x40000000
-#define SA_RESETHAND	0x80000000
-
-#define SA_NOMASK	SA_NODEFER
-#define SA_ONESHOT	SA_RESETHAND
-
-#define SA_RESTORER	0x04000000
-
-/*
- * sigaltstack controls
- */
-#define SS_ONSTACK	1
-#define SS_DISABLE	2
-
-#define MINSIGSTKSZ	2048
-#define SIGSTKSZ	THREAD_SIZE
-
-#include <asm-generic/signal.h>
-
-#ifdef __KERNEL__
-struct old_sigaction {
-	__sighandler_t sa_handler;
-	old_sigset_t sa_mask;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-};
-
-struct sigaction {
-	__sighandler_t sa_handler;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-	sigset_t sa_mask;		/* mask last for extensibility */
-};
-
-struct k_sigaction {
-	struct sigaction sa;
-};
-#else
-/* Here we must cater to libcs that poke about in kernel headers.  */
-
-struct sigaction {
-	union {
-	  __sighandler_t _sa_handler;
-	  void (*_sa_sigaction)(int, struct siginfo *, void *);
-	} _u;
-	sigset_t sa_mask;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-};
-
-#define sa_handler	_u._sa_handler
-#define sa_sigaction	_u._sa_sigaction
-
-#endif /* __KERNEL__ */
-
-typedef struct sigaltstack {
-	void *ss_sp;
-	int ss_flags;
-	size_t ss_size;
-} stack_t;
-
-#ifdef __KERNEL__
-#include <asm/sigcontext.h>
-
-#define sigmask(sig)	(1UL << ((sig) - 1))
-#define ptrace_signal_deliver(regs, cookie) do { } while (0)
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH64_SIGNAL_H */
diff --git a/include/asm-sh64/smp.h b/include/asm-sh64/smp.h
deleted file mode 100644
index 4a4d0da39a84..000000000000
--- a/include/asm-sh64/smp.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef __ASM_SH64_SMP_H
-#define __ASM_SH64_SMP_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/smp.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#endif /* __ASM_SH64_SMP_H */
diff --git a/include/asm-sh64/socket.h b/include/asm-sh64/socket.h
deleted file mode 100644
index 1853f7246ab0..000000000000
--- a/include/asm-sh64/socket.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_SOCKET_H
-#define __ASM_SH64_SOCKET_H
-
-#include <asm-sh/socket.h>
-
-#endif /* __ASM_SH64_SOCKET_H */
diff --git a/include/asm-sh64/sockios.h b/include/asm-sh64/sockios.h
deleted file mode 100644
index 419e76f12f41..000000000000
--- a/include/asm-sh64/sockios.h
+++ /dev/null
@@ -1,25 +0,0 @@
-#ifndef __ASM_SH64_SOCKIOS_H
-#define __ASM_SH64_SOCKIOS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/sockios.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-/* Socket-level I/O control calls. */
-#define FIOGETOWN	_IOR('f', 123, int)
-#define FIOSETOWN 	_IOW('f', 124, int)
-
-#define SIOCATMARK	_IOR('s', 7, int)
-#define SIOCSPGRP	_IOW('s', 8, pid_t)
-#define SIOCGPGRP	_IOR('s', 9, pid_t)
-
-#define SIOCGSTAMP	_IOR('s', 100, struct timeval) /* Get stamp (timeval) */
-#define SIOCGSTAMPNS	_IOR('s', 101, struct timespec) /* Get stamp (timespec) */
-#endif /* __ASM_SH64_SOCKIOS_H */
diff --git a/include/asm-sh64/spinlock.h b/include/asm-sh64/spinlock.h
deleted file mode 100644
index 296b0c9b24a2..000000000000
--- a/include/asm-sh64/spinlock.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef __ASM_SH64_SPINLOCK_H
-#define __ASM_SH64_SPINLOCK_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/spinlock.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#error "No SMP on SH64"
-
-#endif /* __ASM_SH64_SPINLOCK_H */
diff --git a/include/asm-sh64/stat.h b/include/asm-sh64/stat.h
deleted file mode 100644
index 86f551b1987e..000000000000
--- a/include/asm-sh64/stat.h
+++ /dev/null
@@ -1,88 +0,0 @@
-#ifndef __ASM_SH64_STAT_H
-#define __ASM_SH64_STAT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/stat.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-struct __old_kernel_stat {
-	unsigned short st_dev;
-	unsigned short st_ino;
-	unsigned short st_mode;
-	unsigned short st_nlink;
-	unsigned short st_uid;
-	unsigned short st_gid;
-	unsigned short st_rdev;
-	unsigned long  st_size;
-	unsigned long  st_atime;
-	unsigned long  st_mtime;
-	unsigned long  st_ctime;
-};
-
-struct stat {
-	unsigned short st_dev;
-	unsigned short __pad1;
-	unsigned long st_ino;
-	unsigned short st_mode;
-	unsigned short st_nlink;
-	unsigned short st_uid;
-	unsigned short st_gid;
-	unsigned short st_rdev;
-	unsigned short __pad2;
-	unsigned long  st_size;
-	unsigned long  st_blksize;
-	unsigned long  st_blocks;
-	unsigned long  st_atime;
-	unsigned long  st_atime_nsec;
-	unsigned long  st_mtime;
-	unsigned long  st_mtime_nsec;
-	unsigned long  st_ctime;
-	unsigned long  st_ctime_nsec;
-	unsigned long  __unused4;
-	unsigned long  __unused5;
-};
-
-/* This matches struct stat64 in glibc2.1, hence the absolutely
- * insane amounts of padding around dev_t's.
- */
-struct stat64 {
-	unsigned short	st_dev;
-	unsigned char	__pad0[10];
-
-	unsigned long	st_ino;
-	unsigned int	st_mode;
-	unsigned int	st_nlink;
-
-	unsigned long	st_uid;
-	unsigned long	st_gid;
-
-	unsigned short	st_rdev;
-	unsigned char	__pad3[10];
-
-	long long	st_size;
-	unsigned long	st_blksize;
-
-	unsigned long	st_blocks;	/* Number 512-byte blocks allocated. */
-	unsigned long	__pad4;		/* future possible st_blocks high bits */
-
-	unsigned long	st_atime;
-	unsigned long	st_atime_nsec;
-
-	unsigned long	st_mtime;
-	unsigned long	st_mtime_nsec;
-
-	unsigned long	st_ctime;
-	unsigned long	st_ctime_nsec;	/* will be high 32 bits of ctime someday */
-
-	unsigned long	__unused1;
-	unsigned long	__unused2;
-};
-
-#endif /* __ASM_SH64_STAT_H */
diff --git a/include/asm-sh64/statfs.h b/include/asm-sh64/statfs.h
deleted file mode 100644
index 083fd79b2417..000000000000
--- a/include/asm-sh64/statfs.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_STATFS_H
-#define __ASM_SH64_STATFS_H
-
-#include <asm-generic/statfs.h>
-
-#endif /* __ASM_SH64_STATFS_H */
diff --git a/include/asm-sh64/termbits.h b/include/asm-sh64/termbits.h
deleted file mode 100644
index 86bde5ec1414..000000000000
--- a/include/asm-sh64/termbits.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_TERMBITS_H
-#define __ASM_SH64_TERMBITS_H
-
-#include <asm-sh/termbits.h>
-
-#endif /* __ASM_SH64_TERMBITS_H */
diff --git a/include/asm-sh64/termios.h b/include/asm-sh64/termios.h
deleted file mode 100644
index dc44e6ed3a7c..000000000000
--- a/include/asm-sh64/termios.h
+++ /dev/null
@@ -1,99 +0,0 @@
-#ifndef __ASM_SH64_TERMIOS_H
-#define __ASM_SH64_TERMIOS_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/termios.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <asm/termbits.h>
-#include <asm/ioctls.h>
-
-struct winsize {
-	unsigned short ws_row;
-	unsigned short ws_col;
-	unsigned short ws_xpixel;
-	unsigned short ws_ypixel;
-};
-
-#define NCC 8
-struct termio {
-	unsigned short c_iflag;		/* input mode flags */
-	unsigned short c_oflag;		/* output mode flags */
-	unsigned short c_cflag;		/* control mode flags */
-	unsigned short c_lflag;		/* local mode flags */
-	unsigned char c_line;		/* line discipline */
-	unsigned char c_cc[NCC];	/* control characters */
-};
-
-/* modem lines */
-#define TIOCM_LE	0x001
-#define TIOCM_DTR	0x002
-#define TIOCM_RTS	0x004
-#define TIOCM_ST	0x008
-#define TIOCM_SR	0x010
-#define TIOCM_CTS	0x020
-#define TIOCM_CAR	0x040
-#define TIOCM_RNG	0x080
-#define TIOCM_DSR	0x100
-#define TIOCM_CD	TIOCM_CAR
-#define TIOCM_RI	TIOCM_RNG
-#define TIOCM_OUT1	0x2000
-#define TIOCM_OUT2	0x4000
-#define TIOCM_LOOP	0x8000
-
-/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
-
-#ifdef __KERNEL__
-
-/*	intr=^C		quit=^\		erase=del	kill=^U
-	eof=^D		vtime=\0	vmin=\1		sxtc=\0
-	start=^Q	stop=^S		susp=^Z		eol=\0
-	reprint=^R	discard=^U	werase=^W	lnext=^V
-	eol2=\0
-*/
-#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
-
-/*
- * Translate a "termio" structure into a "termios". Ugh.
- */
-#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \
-	unsigned short __tmp; \
-	get_user(__tmp,&(termio)->x); \
-	*(unsigned short *) &(termios)->x = __tmp; \
-}
-
-#define user_termio_to_kernel_termios(termios, termio) \
-({ \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_iflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_oflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_cflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_lflag); \
-	copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \
-})
-
-/*
- * Translate a "termios" structure into a "termio". Ugh.
- */
-#define kernel_termios_to_user_termio(termio, termios) \
-({ \
-	put_user((termios)->c_iflag, &(termio)->c_iflag); \
-	put_user((termios)->c_oflag, &(termio)->c_oflag); \
-	put_user((termios)->c_cflag, &(termio)->c_cflag); \
-	put_user((termios)->c_lflag, &(termio)->c_lflag); \
-	put_user((termios)->c_line,  &(termio)->c_line); \
-	copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \
-})
-
-#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios))
-#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios))
-
-#endif	/* __KERNEL__ */
-
-#endif	/* __ASM_SH64_TERMIOS_H */
diff --git a/include/asm-sh64/timex.h b/include/asm-sh64/timex.h
deleted file mode 100644
index 163e2b62fe27..000000000000
--- a/include/asm-sh64/timex.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __ASM_SH64_TIMEX_H
-#define __ASM_SH64_TIMEX_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/timex.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- * sh-5 architecture timex specifications
- *
- */
-
-#define CLOCK_TICK_RATE	1193180 /* Underlying HZ */
-#define CLOCK_TICK_FACTOR	20	/* Factor of both 1000000 and CLOCK_TICK_RATE */
-
-typedef unsigned long cycles_t;
-
-static __inline__ cycles_t get_cycles (void)
-{
-	return 0;
-}
-
-#define vxtime_lock()		do {} while (0)
-#define vxtime_unlock()		do {} while (0)
-
-#endif /* __ASM_SH64_TIMEX_H */
diff --git a/include/asm-sh64/tlbflush.h b/include/asm-sh64/tlbflush.h
deleted file mode 100644
index 16a164a23754..000000000000
--- a/include/asm-sh64/tlbflush.h
+++ /dev/null
@@ -1,27 +0,0 @@
-#ifndef __ASM_SH64_TLBFLUSH_H
-#define __ASM_SH64_TLBFLUSH_H
-
-#include <asm/pgalloc.h>
-
-/*
- * TLB flushing:
- *
- *  - flush_tlb() flushes the current mm struct TLBs
- *  - flush_tlb_all() flushes all processes TLBs
- *  - flush_tlb_mm(mm) flushes the specified mm context TLB's
- *  - flush_tlb_page(vma, vmaddr) flushes one page
- *  - flush_tlb_range(mm, start, end) flushes a range of pages
- *
- */
-
-extern void flush_tlb(void);
-extern void flush_tlb_all(void);
-extern void flush_tlb_mm(struct mm_struct *mm);
-extern void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
-			    unsigned long end);
-extern void flush_tlb_page(struct vm_area_struct *vma, unsigned long page);
-
-extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
-
-#endif /* __ASM_SH64_TLBFLUSH_H */
-
diff --git a/include/asm-sh64/topology.h b/include/asm-sh64/topology.h
deleted file mode 100644
index 34211787345f..000000000000
--- a/include/asm-sh64/topology.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_SH64_TOPOLOGY_H
-#define __ASM_SH64_TOPOLOGY_H
-
-#include <asm-generic/topology.h>
-
-#endif /* __ASM_SH64_TOPOLOGY_H */
diff --git a/include/asm-sh64/types.h b/include/asm-sh64/types.h
deleted file mode 100644
index 2c7ad73b3883..000000000000
--- a/include/asm-sh64/types.h
+++ /dev/null
@@ -1,74 +0,0 @@
-#ifndef __ASM_SH64_TYPES_H
-#define __ASM_SH64_TYPES_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/types.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#ifndef __ASSEMBLY__
-
-typedef unsigned short umode_t;
-
-/*
- * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the
- * header files exported to user space
- */
-
-typedef __signed__ char __s8;
-typedef unsigned char __u8;
-
-typedef __signed__ short __s16;
-typedef unsigned short __u16;
-
-typedef __signed__ int __s32;
-typedef unsigned int __u32;
-
-#if defined(__GNUC__)
-__extension__ typedef __signed__ long long __s64;
-__extension__ typedef unsigned long long __u64;
-#endif
-
-#endif /* __ASSEMBLY__ */
-
-/*
- * These aren't exported outside the kernel to avoid name space clashes
- */
-#ifdef __KERNEL__
-
-#ifndef __ASSEMBLY__
-
-typedef __signed__ char s8;
-typedef unsigned char u8;
-
-typedef __signed__ short s16;
-typedef unsigned short u16;
-
-typedef __signed__ int s32;
-typedef unsigned int u32;
-
-typedef __signed__ long long s64;
-typedef unsigned long long u64;
-
-/* DMA addresses come in generic and 64-bit flavours.  */
-
-#ifdef CONFIG_HIGHMEM64G
-typedef u64 dma_addr_t;
-#else
-typedef u32 dma_addr_t;
-#endif
-typedef u64 dma64_addr_t;
-
-#endif /* __ASSEMBLY__ */
-
-#define BITS_PER_LONG 32
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH64_TYPES_H */
diff --git a/include/asm-sh64/ucontext.h b/include/asm-sh64/ucontext.h
deleted file mode 100644
index cf77a08551ca..000000000000
--- a/include/asm-sh64/ucontext.h
+++ /dev/null
@@ -1,23 +0,0 @@
-#ifndef __ASM_SH64_UCONTEXT_H
-#define __ASM_SH64_UCONTEXT_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/ucontext.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-struct ucontext {
-	unsigned long	  uc_flags;
-	struct ucontext  *uc_link;
-	stack_t		  uc_stack;
-	struct sigcontext uc_mcontext;
-	sigset_t	  uc_sigmask;	/* mask last for extensibility */
-};
-
-#endif /* __ASM_SH64_UCONTEXT_H */
diff --git a/include/asm-sh64/unaligned.h b/include/asm-sh64/unaligned.h
deleted file mode 100644
index 74481b186ae8..000000000000
--- a/include/asm-sh64/unaligned.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef __ASM_SH64_UNALIGNED_H
-#define __ASM_SH64_UNALIGNED_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/unaligned.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <asm-generic/unaligned.h>
-
-#endif /* __ASM_SH64_UNALIGNED_H */
diff --git a/include/asm-sh64/user.h b/include/asm-sh64/user.h
deleted file mode 100644
index eb3b33edd73e..000000000000
--- a/include/asm-sh64/user.h
+++ /dev/null
@@ -1,70 +0,0 @@
-#ifndef __ASM_SH64_USER_H
-#define __ASM_SH64_USER_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/user.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-#include <linux/types.h>
-#include <asm/ptrace.h>
-#include <asm/page.h>
-
-/*
- * Core file format: The core file is written in such a way that gdb
- * can understand it and provide useful information to the user (under
- * linux we use the `trad-core' bfd).  The file contents are as follows:
- *
- *  upage: 1 page consisting of a user struct that tells gdb
- *	what is present in the file.  Directly after this is a
- *	copy of the task_struct, which is currently not used by gdb,
- *	but it may come in handy at some point.  All of the registers
- *	are stored as part of the upage.  The upage should always be
- *	only one page long.
- *  data: The data segment follows next.  We use current->end_text to
- *	current->brk to pick up all of the user variables, plus any memory
- *	that may have been sbrk'ed.  No attempt is made to determine if a
- *	page is demand-zero or if a page is totally unused, we just cover
- *	the entire range.  All of the addresses are rounded in such a way
- *	that an integral number of pages is written.
- *  stack: We need the stack information in order to get a meaningful
- *	backtrace.  We need to write the data from usp to
- *	current->start_stack, so we round each of these in order to be able
- *	to write an integer number of pages.
- */
-
-struct user_fpu_struct {
-        unsigned long long fp_regs[32];
-	unsigned int fpscr;
-};
-
-struct user {
-	struct pt_regs	regs;			/* entire machine state */
-	struct user_fpu_struct fpu;	/* Math Co-processor registers  */
-	int u_fpvalid;		/* True if math co-processor being used */
-	size_t		u_tsize;		/* text size (pages) */
-	size_t		u_dsize;		/* data size (pages) */
-	size_t		u_ssize;		/* stack size (pages) */
-	unsigned long	start_code;		/* text starting address */
-	unsigned long	start_data;		/* data starting address */
-	unsigned long	start_stack;		/* stack starting address */
-	long int	signal;			/* signal causing core dump */
-	struct regs *	u_ar0;			/* help gdb find registers */
-	struct user_fpu_struct* u_fpstate;	/* Math Co-processor pointer */
-	unsigned long	magic;			/* identifies a core file */
-	char		u_comm[32];		/* user command name */
-};
-
-#define NBPG			PAGE_SIZE
-#define UPAGES			1
-#define HOST_TEXT_START_ADDR	(u.start_code)
-#define HOST_DATA_START_ADDR	(u.start_data)
-#define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
-
-#endif /* __ASM_SH64_USER_H */
-- 
cgit v1.2.3


From 17bfa6397875e6901c3cafdc711437041664d94c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 17:28:37 +0900
Subject: sh: Special layout for SH-5 stat.h and user.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/stat.h | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-sh/user.h |  7 ++++++
 2 files changed, 69 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/stat.h b/include/asm-sh/stat.h
index 6d6ad26e3a2a..e1810cc6e3da 100644
--- a/include/asm-sh/stat.h
+++ b/include/asm-sh/stat.h
@@ -15,6 +15,66 @@ struct __old_kernel_stat {
 	unsigned long  st_ctime;
 };
 
+#if defined(__SH5__) || defined(CONFIG_CPU_SH5)
+struct stat {
+	unsigned short st_dev;
+	unsigned short __pad1;
+	unsigned long st_ino;
+	unsigned short st_mode;
+	unsigned short st_nlink;
+	unsigned short st_uid;
+	unsigned short st_gid;
+	unsigned short st_rdev;
+	unsigned short __pad2;
+	unsigned long  st_size;
+	unsigned long  st_blksize;
+	unsigned long  st_blocks;
+	unsigned long  st_atime;
+	unsigned long  st_atime_nsec;
+	unsigned long  st_mtime;
+	unsigned long  st_mtime_nsec;
+	unsigned long  st_ctime;
+	unsigned long  st_ctime_nsec;
+	unsigned long  __unused4;
+	unsigned long  __unused5;
+};
+
+/* This matches struct stat64 in glibc2.1, hence the absolutely
+ * insane amounts of padding around dev_t's.
+ */
+struct stat64 {
+	unsigned short	st_dev;
+	unsigned char	__pad0[10];
+
+	unsigned long	st_ino;
+	unsigned int	st_mode;
+	unsigned int	st_nlink;
+
+	unsigned long	st_uid;
+	unsigned long	st_gid;
+
+	unsigned short	st_rdev;
+	unsigned char	__pad3[10];
+
+	long long	st_size;
+	unsigned long	st_blksize;
+
+	unsigned long	st_blocks;	/* Number 512-byte blocks allocated. */
+	unsigned long	__pad4;		/* future possible st_blocks high bits */
+
+	unsigned long	st_atime;
+	unsigned long	st_atime_nsec;
+
+	unsigned long	st_mtime;
+	unsigned long	st_mtime_nsec;
+
+	unsigned long	st_ctime;
+	unsigned long	st_ctime_nsec;	/* will be high 32 bits of ctime someday */
+
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+};
+#else
 struct stat {
 	unsigned long  st_dev;
 	unsigned long  st_ino;
@@ -67,11 +127,12 @@ struct stat64 {
 	unsigned long	st_mtime_nsec;
 
 	unsigned long	st_ctime;
-	unsigned long	st_ctime_nsec; 
+	unsigned long	st_ctime_nsec;
 
 	unsigned long long	st_ino;
 };
 
 #define STAT_HAVE_NSEC 1
+#endif
 
 #endif /* __ASM_SH_STAT_H */
diff --git a/include/asm-sh/user.h b/include/asm-sh/user.h
index d1b8511d9d9f..706b1c7baba9 100644
--- a/include/asm-sh/user.h
+++ b/include/asm-sh/user.h
@@ -27,12 +27,19 @@
  *	to write an integer number of pages.
  */
 
+#if defined(__SH5__) || defined(CONFIG_CPU_SH5)
+struct user fpu_struct {
+	unsigned long fp_regs[32];
+	unsigned int fpscr;
+};
+#else
 struct user_fpu_struct {
 	unsigned long fp_regs[16];
 	unsigned long xfp_regs[16];
 	unsigned long fpscr;
 	unsigned long fpul;
 };
+#endif
 
 struct user {
 	struct pt_regs	regs;			/* entire machine state */
-- 
cgit v1.2.3


From fb8e569c1d4f44a4632e2db95a27ed45734d4705 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 17:55:12 +0900
Subject: sh: Fix up user_fpu_struct typo for SH-5.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh64/kernel/asm-offsets.c |  33 ----------
 arch/sh64/kernel/init_task.c   |  46 --------------
 arch/sh64/kernel/semaphore.c   | 140 -----------------------------------------
 include/asm-sh/user.h          |   2 +-
 4 files changed, 1 insertion(+), 220 deletions(-)
 delete mode 100644 arch/sh64/kernel/asm-offsets.c
 delete mode 100644 arch/sh64/kernel/init_task.c
 delete mode 100644 arch/sh64/kernel/semaphore.c

(limited to 'include')

diff --git a/arch/sh64/kernel/asm-offsets.c b/arch/sh64/kernel/asm-offsets.c
deleted file mode 100644
index ca76537c16c0..000000000000
--- a/arch/sh64/kernel/asm-offsets.c
+++ /dev/null
@@ -1,33 +0,0 @@
-/*
- * This program is used to generate definitions needed by
- * assembly language modules.
- *
- * We use the technique used in the OSF Mach kernel code:
- * generate asm statements containing #defines,
- * compile this file to assembler, and then extract the
- * #defines from the assembly-language output.
- */
-
-#include <linux/stddef.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <asm/thread_info.h>
-
-#define DEFINE(sym, val) \
-        asm volatile("\n->" #sym " %0 " #val : : "i" (val))
-
-#define BLANK() asm volatile("\n->" : : )
-
-int main(void)
-{
-	/* offsets into the thread_info struct */
-	DEFINE(TI_TASK,		offsetof(struct thread_info, task));
-	DEFINE(TI_EXEC_DOMAIN,	offsetof(struct thread_info, exec_domain));
-	DEFINE(TI_FLAGS,	offsetof(struct thread_info, flags));
-	DEFINE(TI_PRE_COUNT,	offsetof(struct thread_info, preempt_count));
-	DEFINE(TI_CPU,		offsetof(struct thread_info, cpu));
-	DEFINE(TI_ADDR_LIMIT,	offsetof(struct thread_info, addr_limit));
-	DEFINE(TI_RESTART_BLOCK,offsetof(struct thread_info, restart_block));
-
-	return 0;
-}
diff --git a/arch/sh64/kernel/init_task.c b/arch/sh64/kernel/init_task.c
deleted file mode 100644
index deee8bfd3270..000000000000
--- a/arch/sh64/kernel/init_task.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/init_task.c
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-#include <linux/rwsem.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/init_task.h>
-#include <linux/mqueue.h>
-#include <linux/fs.h>
-#include <asm/uaccess.h>
-#include <asm/pgtable.h>
-
-static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
-static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
-static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
-struct mm_struct init_mm = INIT_MM(init_mm);
-
-struct pt_regs fake_swapper_regs;
-
-/*
- * Initial thread structure.
- *
- * We need to make sure that this is THREAD_SIZE-byte aligned due
- * to the way process stacks are handled. This is done by having a
- * special "init_task" linker map entry..
- */
-union thread_union init_thread_union
-	__attribute__((__section__(".data.init_task"))) =
-		{ INIT_THREAD_INFO(init_task) };
-
-/*
- * Initial task structure.
- *
- * All other task structs will be allocated on slabs in fork.c
- */
-struct task_struct init_task = INIT_TASK(init_task);
-
diff --git a/arch/sh64/kernel/semaphore.c b/arch/sh64/kernel/semaphore.c
deleted file mode 100644
index 72c16533436e..000000000000
--- a/arch/sh64/kernel/semaphore.c
+++ /dev/null
@@ -1,140 +0,0 @@
-/*
- * Just taken from alpha implementation.
- * This can't work well, perhaps.
- */
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/errno.h>
-#include <linux/rwsem.h>
-#include <linux/sched.h>
-#include <linux/wait.h>
-#include <linux/init.h>
-#include <asm/semaphore.h>
-#include <asm/semaphore-helper.h>
-
-spinlock_t semaphore_wake_lock;
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-#define DOWN_VAR				\
-	struct task_struct *tsk = current;	\
-	wait_queue_t wait;			\
-	init_waitqueue_entry(&wait, tsk);
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	tsk->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		tsk->state = (task_state);	\
-	}					\
-	tsk->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DOWN_VAR
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-	DOWN_VAR
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, tsk);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/include/asm-sh/user.h b/include/asm-sh/user.h
index 706b1c7baba9..1a4f43c75126 100644
--- a/include/asm-sh/user.h
+++ b/include/asm-sh/user.h
@@ -28,7 +28,7 @@
  */
 
 #if defined(__SH5__) || defined(CONFIG_CPU_SH5)
-struct user fpu_struct {
+struct user_fpu_struct {
 	unsigned long fp_regs[32];
 	unsigned int fpscr;
 };
-- 
cgit v1.2.3


From 0f2c15cecee0ff00e4bfa91dd20b33ccd9285360 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 18:06:34 +0900
Subject: sh: Add onchip remap prototypes, kill old sh64 io.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/io.h   |  17 ++++-
 include/asm-sh64/io.h | 196 --------------------------------------------------
 2 files changed, 16 insertions(+), 197 deletions(-)
 delete mode 100644 include/asm-sh64/io.h

(limited to 'include')

diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 74305edceeeb..a4e5f5573eee 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -191,6 +191,8 @@ __BUILD_MEMORY_STRING(w, u16)
 
 #define mmiowb()	wmb()	/* synco on SH-4A, otherwise a nop */
 
+#define IO_SPACE_LIMIT 0xffffffff
+
 /*
  * This function provides a method for the generic case where a board-specific
  * ioport_map simply needs to return the port + some arbitrary port base.
@@ -226,6 +228,11 @@ static inline unsigned int ctrl_inl(unsigned long addr)
 	return *(volatile unsigned long*)addr;
 }
 
+static inline unsigned long long ctrl_inq(unsigned long addr)
+{
+	return *(volatile unsigned long long*)addr;
+}
+
 static inline void ctrl_outb(unsigned char b, unsigned long addr)
 {
 	*(volatile unsigned char*)addr = b;
@@ -241,6 +248,11 @@ static inline void ctrl_outl(unsigned int b, unsigned long addr)
         *(volatile unsigned long*)addr = b;
 }
 
+static inline void ctrl_outq(unsigned long long b, unsigned long addr)
+{
+	*(volatile unsigned long long*)addr = b;
+}
+
 static inline void ctrl_delay(void)
 {
 #ifdef P2SEG
@@ -253,7 +265,10 @@ unsigned long long peek_real_address_q(unsigned long long addr);
 unsigned long long poke_real_address_q(unsigned long long addr,
 				       unsigned long long val);
 
-#define IO_SPACE_LIMIT 0xffffffff
+/* arch/sh/mm/ioremap_64.c */
+unsigned long onchip_remap(unsigned long addr, unsigned long size,
+			   const char *name);
+extern void onchip_unmap(unsigned long vaddr);
 
 #if !defined(CONFIG_MMU)
 #define virt_to_phys(address)	((unsigned long)(address))
diff --git a/include/asm-sh64/io.h b/include/asm-sh64/io.h
deleted file mode 100644
index 7bd7314d38c2..000000000000
--- a/include/asm-sh64/io.h
+++ /dev/null
@@ -1,196 +0,0 @@
-#ifndef __ASM_SH64_IO_H
-#define __ASM_SH64_IO_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/io.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- */
-
-/*
- * Convention:
- *    read{b,w,l}/write{b,w,l} are for PCI,
- *    while in{b,w,l}/out{b,w,l} are for ISA
- * These may (will) be platform specific function.
- *
- * In addition, we have
- *   ctrl_in{b,w,l}/ctrl_out{b,w,l} for SuperH specific I/O.
- * which are processor specific. Address should be the result of
- * onchip_remap();
- */
-
-#include <linux/compiler.h>
-#include <asm/cache.h>
-#include <asm/system.h>
-#include <asm/page.h>
-#include <asm-generic/iomap.h>
-
-/*
- * Nothing overly special here.. instead of doing the same thing
- * over and over again, we just define a set of sh64_in/out functions
- * with an implicit size. The traditional read{b,w,l}/write{b,w,l}
- * mess is wrapped to this, as are the SH-specific ctrl_in/out routines.
- */
-static inline unsigned char sh64_in8(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned char __force *)addr;
-}
-
-static inline unsigned short sh64_in16(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned short __force *)addr;
-}
-
-static inline unsigned int sh64_in32(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned int __force *)addr;
-}
-
-static inline unsigned long long sh64_in64(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned long long __force *)addr;
-}
-
-static inline void sh64_out8(unsigned char b, volatile void __iomem *addr)
-{
-	*(volatile unsigned char __force *)addr = b;
-	wmb();
-}
-
-static inline void sh64_out16(unsigned short b, volatile void __iomem *addr)
-{
-	*(volatile unsigned short __force *)addr = b;
-	wmb();
-}
-
-static inline void sh64_out32(unsigned int b, volatile void __iomem *addr)
-{
-	*(volatile unsigned int __force *)addr = b;
-	wmb();
-}
-
-static inline void sh64_out64(unsigned long long b, volatile void __iomem *addr)
-{
-	*(volatile unsigned long long __force *)addr = b;
-	wmb();
-}
-
-#define readb(addr)		sh64_in8(addr)
-#define readw(addr)		sh64_in16(addr)
-#define readl(addr)		sh64_in32(addr)
-#define readb_relaxed(addr)	sh64_in8(addr)
-#define readw_relaxed(addr)	sh64_in16(addr)
-#define readl_relaxed(addr)	sh64_in32(addr)
-
-#define writeb(b, addr)		sh64_out8(b, addr)
-#define writew(b, addr)		sh64_out16(b, addr)
-#define writel(b, addr)		sh64_out32(b, addr)
-
-#define ctrl_inb(addr)		sh64_in8(ioport_map(addr, 1))
-#define ctrl_inw(addr)		sh64_in16(ioport_map(addr, 2))
-#define ctrl_inl(addr)		sh64_in32(ioport_map(addr, 4))
-
-#define ctrl_outb(b, addr)	sh64_out8(b, ioport_map(addr, 1))
-#define ctrl_outw(b, addr)	sh64_out16(b, ioport_map(addr, 2))
-#define ctrl_outl(b, addr)	sh64_out32(b, ioport_map(addr, 4))
-
-#define ioread8(addr)		sh64_in8(addr)
-#define ioread16(addr)		sh64_in16(addr)
-#define ioread32(addr)		sh64_in32(addr)
-#define iowrite8(b, addr)	sh64_out8(b, addr)
-#define iowrite16(b, addr)	sh64_out16(b, addr)
-#define iowrite32(b, addr)	sh64_out32(b, addr)
-
-#define inb(addr)		ctrl_inb(addr)
-#define inw(addr)		ctrl_inw(addr)
-#define inl(addr)		ctrl_inl(addr)
-#define outb(b, addr)		ctrl_outb(b, addr)
-#define outw(b, addr)		ctrl_outw(b, addr)
-#define outl(b, addr)		ctrl_outl(b, addr)
-
-void outsw(unsigned long port, const void *addr, unsigned long count);
-void insw(unsigned long port, void *addr, unsigned long count);
-void outsl(unsigned long port, const void *addr, unsigned long count);
-void insl(unsigned long port, void *addr, unsigned long count);
-
-#define inb_p(addr)    inb(addr)
-#define inw_p(addr)    inw(addr)
-#define inl_p(addr)    inl(addr)
-#define outb_p(x,addr) outb(x,addr)
-#define outw_p(x,addr) outw(x,addr)
-#define outl_p(x,addr) outl(x,addr)
-
-#define __raw_readb		readb
-#define __raw_readw		readw
-#define __raw_readl		readl
-#define __raw_writeb		writeb
-#define __raw_writew		writew
-#define __raw_writel		writel
-
-void memcpy_toio(void __iomem *to, const void *from, long count);
-void memcpy_fromio(void *to, void __iomem *from, long count);
-
-#define mmiowb()
-
-#ifdef __KERNEL__
-
-#ifdef CONFIG_SH_CAYMAN
-extern unsigned long smsc_superio_virt;
-#endif
-#ifdef CONFIG_PCI
-extern unsigned long pciio_virt;
-#endif
-
-#define IO_SPACE_LIMIT 0xffffffff
-
-/*
- * Change virtual addresses to physical addresses and vv.
- * These are trivial on the 1:1 Linux/SuperH mapping
- */
-static inline unsigned long virt_to_phys(volatile void * address)
-{
-	return __pa(address);
-}
-
-static inline void * phys_to_virt(unsigned long address)
-{
-	return __va(address);
-}
-
-extern void * __ioremap(unsigned long phys_addr, unsigned long size,
-			unsigned long flags);
-
-static inline void * ioremap(unsigned long phys_addr, unsigned long size)
-{
-	return __ioremap(phys_addr, size, 1);
-}
-
-static inline void * ioremap_nocache (unsigned long phys_addr, unsigned long size)
-{
-	return __ioremap(phys_addr, size, 0);
-}
-
-extern void iounmap(void *addr);
-
-unsigned long onchip_remap(unsigned long addr, unsigned long size, const char* name);
-extern void onchip_unmap(unsigned long vaddr);
-
-/*
- * Convert a physical pointer to a virtual kernel pointer for /dev/mem
- * access
- */
-#define xlate_dev_mem_ptr(p)	__va(p)
-
-/*
- * Convert a virtual cached pointer to an uncached pointer
- */
-#define xlate_dev_kmem_ptr(p)	p
-
-#endif /* __KERNEL__ */
-#endif /* __ASM_SH64_IO_H */
-- 
cgit v1.2.3


From caead5ef34e5abdda8c5189cf698e0b863904701 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 22:53:34 +0900
Subject: sh: Kill off the last of the sh64 headers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh5/addrspace.h |   5 ++
 include/asm-sh/cpu-sh5/irq.h       | 144 +++++++++++++++++++++++++++++++++++++
 include/asm-sh/platform.h          |  64 +++++++++++++++++
 include/asm-sh64/cayman.h          |  20 ------
 include/asm-sh64/hardware.h        |  22 ------
 include/asm-sh64/irq.h             | 144 -------------------------------------
 include/asm-sh64/keyboard.h        |  70 ------------------
 include/asm-sh64/pci.h             | 102 --------------------------
 include/asm-sh64/platform.h        |  64 -----------------
 include/asm-sh64/thread_info.h     |  91 -----------------------
 10 files changed, 213 insertions(+), 513 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh5/irq.h
 create mode 100644 include/asm-sh/platform.h
 delete mode 100644 include/asm-sh64/cayman.h
 delete mode 100644 include/asm-sh64/hardware.h
 delete mode 100644 include/asm-sh64/irq.h
 delete mode 100644 include/asm-sh64/keyboard.h
 delete mode 100644 include/asm-sh64/pci.h
 delete mode 100644 include/asm-sh64/platform.h
 delete mode 100644 include/asm-sh64/thread_info.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh5/addrspace.h b/include/asm-sh/cpu-sh5/addrspace.h
index 844c5210e8ed..dc36b9a03af6 100644
--- a/include/asm-sh/cpu-sh5/addrspace.h
+++ b/include/asm-sh/cpu-sh5/addrspace.h
@@ -1,6 +1,11 @@
 #ifndef __ASM_SH_CPU_SH5_ADDRSPACE_H
 #define __ASM_SH_CPU_SH5_ADDRSPACE_H
 
+#define	PHYS_PERIPHERAL_BLOCK	0x09000000
+#define PHYS_DMAC_BLOCK		0x0e000000
+#define PHYS_PCI_BLOCK		0x60000000
+#define PHYS_EMI_BLOCK		0xff000000
+
 /* No segmentation.. */
 
 #endif /* __ASM_SH_CPU_SH5_ADDRSPACE_H */
diff --git a/include/asm-sh/cpu-sh5/irq.h b/include/asm-sh/cpu-sh5/irq.h
new file mode 100644
index 000000000000..5c9e6a873aeb
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/irq.h
@@ -0,0 +1,144 @@
+#ifndef __ASM_SH64_IRQ_H
+#define __ASM_SH64_IRQ_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/irq.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ *
+ */
+
+
+/*
+ * Encoded IRQs are not considered worth to be supported.
+ * Main reason is that there's no per-encoded-interrupt
+ * enable/disable mechanism (as there was in SH3/4).
+ * An all enabled/all disabled is worth only if there's
+ * a cascaded IC to disable/enable/ack on. Until such
+ * IC is available there's no such support.
+ *
+ * Presumably Encoded IRQs may use extra IRQs beyond 64,
+ * below. Some logic must be added to cope with IRQ_IRL?
+ * in an exclusive way.
+ *
+ * Priorities are set at Platform level, when IRQ_IRL0-3
+ * are set to 0 Encoding is allowed. Otherwise it's not
+ * allowed.
+ */
+
+/* Independent IRQs */
+#define IRQ_IRL0	0
+#define IRQ_IRL1	1
+#define IRQ_IRL2	2
+#define IRQ_IRL3	3
+
+#define IRQ_INTA	4
+#define IRQ_INTB	5
+#define IRQ_INTC	6
+#define IRQ_INTD	7
+
+#define IRQ_SERR	12
+#define IRQ_ERR		13
+#define IRQ_PWR3	14
+#define IRQ_PWR2	15
+#define IRQ_PWR1	16
+#define IRQ_PWR0	17
+
+#define IRQ_DMTE0	18
+#define IRQ_DMTE1	19
+#define IRQ_DMTE2	20
+#define IRQ_DMTE3	21
+#define IRQ_DAERR	22
+
+#define IRQ_TUNI0	32
+#define IRQ_TUNI1	33
+#define IRQ_TUNI2	34
+#define IRQ_TICPI2	35
+
+#define IRQ_ATI		36
+#define IRQ_PRI		37
+#define IRQ_CUI		38
+
+#define IRQ_ERI		39
+#define IRQ_RXI		40
+#define IRQ_BRI		41
+#define IRQ_TXI		42
+
+#define IRQ_ITI		63
+
+#define NR_INTC_IRQS	64
+
+#ifdef CONFIG_SH_CAYMAN
+#define NR_EXT_IRQS     32
+#define START_EXT_IRQS  64
+
+/* PCI bus 2 uses encoded external interrupts on the Cayman board */
+#define IRQ_P2INTA      (START_EXT_IRQS + (3*8) + 0)
+#define IRQ_P2INTB      (START_EXT_IRQS + (3*8) + 1)
+#define IRQ_P2INTC      (START_EXT_IRQS + (3*8) + 2)
+#define IRQ_P2INTD      (START_EXT_IRQS + (3*8) + 3)
+
+#define I8042_KBD_IRQ	(START_EXT_IRQS + 2)
+#define I8042_AUX_IRQ	(START_EXT_IRQS + 6)
+
+#define IRQ_CFCARD	(START_EXT_IRQS + 7)
+#define IRQ_PCMCIA	(0)
+
+#else
+#define NR_EXT_IRQS	0
+#endif
+
+#define NR_IRQS		(NR_INTC_IRQS+NR_EXT_IRQS)
+
+
+/* Default IRQs, fixed */
+#define TIMER_IRQ	IRQ_TUNI0
+#define RTC_IRQ		IRQ_CUI
+
+/* Default Priorities, Platform may choose differently */
+#define	NO_PRIORITY	0	/* Disabled */
+#define TIMER_PRIORITY	2
+#define RTC_PRIORITY	TIMER_PRIORITY
+#define SCIF_PRIORITY	3
+#define INTD_PRIORITY	3
+#define	IRL3_PRIORITY	4
+#define INTC_PRIORITY	6
+#define	IRL2_PRIORITY	7
+#define INTB_PRIORITY	9
+#define	IRL1_PRIORITY	10
+#define INTA_PRIORITY	12
+#define	IRL0_PRIORITY	13
+#define TOP_PRIORITY	15
+
+extern int intc_evt_to_irq[(0xE20/0x20)+1];
+int intc_irq_describe(char* p, int irq);
+
+#define irq_canonicalize(irq)	(irq)
+
+#ifdef CONFIG_SH_CAYMAN
+int cayman_irq_demux(int evt);
+int cayman_irq_describe(char* p, int irq);
+#define irq_demux(x) cayman_irq_demux(x)
+#define irq_describe(p, x) cayman_irq_describe(p, x)
+#else
+#define irq_demux(x) (intc_evt_to_irq[x])
+#define irq_describe(p, x) intc_irq_describe(p, x)
+#endif
+
+/*
+ * Function for "on chip support modules".
+ */
+
+/*
+ * SH-5 supports Priority based interrupts only.
+ * Interrupt priorities are defined at platform level.
+ */
+#define set_ipr_data(a, b, c, d)
+#define make_ipr_irq(a)
+#define make_imask_irq(a)
+
+#endif /* __ASM_SH64_IRQ_H */
diff --git a/include/asm-sh/platform.h b/include/asm-sh/platform.h
new file mode 100644
index 000000000000..bd0d9c405a80
--- /dev/null
+++ b/include/asm-sh/platform.h
@@ -0,0 +1,64 @@
+#ifndef __ASM_SH64_PLATFORM_H
+#define __ASM_SH64_PLATFORM_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/platform.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ *
+ * benedict.gaster@superh.com:	 3rd May 2002
+ *    Added support for ramdisk, removing statically linked romfs at the same time.
+ */
+
+#include <linux/ioport.h>
+#include <asm/irq.h>
+
+
+/*
+ * Platform definition structure.
+ */
+struct sh64_platform {
+	unsigned int readonly_rootfs;
+	unsigned int ramdisk_flags;
+	unsigned int initial_root_dev;
+	unsigned int loader_type;
+	unsigned int initrd_start;
+	unsigned int initrd_size;
+	unsigned int fpu_flags;
+	unsigned int io_res_count;
+	unsigned int kram_res_count;
+	unsigned int xram_res_count;
+	unsigned int rom_res_count;
+	struct resource *io_res_p;
+	struct resource *kram_res_p;
+	struct resource *xram_res_p;
+	struct resource *rom_res_p;
+};
+
+extern struct sh64_platform platform_parms;
+
+extern unsigned long long memory_start, memory_end;
+
+extern unsigned long long fpu_in_use;
+
+extern int platform_int_priority[NR_INTC_IRQS];
+
+#define FPU_FLAGS		(platform_parms.fpu_flags)
+#define STANDARD_IO_RESOURCES	(platform_parms.io_res_count)
+#define STANDARD_KRAM_RESOURCES	(platform_parms.kram_res_count)
+#define STANDARD_XRAM_RESOURCES	(platform_parms.xram_res_count)
+#define STANDARD_ROM_RESOURCES	(platform_parms.rom_res_count)
+
+/*
+ * Kernel Memory description, Respectively:
+ * code = last but one memory descriptor
+ * data = last memory descriptor
+ */
+#define code_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 2])
+#define data_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 1])
+
+#endif	/* __ASM_SH64_PLATFORM_H */
diff --git a/include/asm-sh64/cayman.h b/include/asm-sh64/cayman.h
deleted file mode 100644
index 7b6b96844842..000000000000
--- a/include/asm-sh64/cayman.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/cayman.h
- *
- * Cayman definitions
- *
- * Global defintions for the SH5 Cayman board
- *
- * Copyright (C) 2002 Stuart Menefy
- */
-
-
-/* Setup for the SMSC FDC37C935 / LAN91C100FD */
-#define SMSC_IRQ         IRQ_IRL1
-
-/* Setup for PCI Bus 2, which transmits interrupts via the EPLD */
-#define PCI2_IRQ         IRQ_IRL3
diff --git a/include/asm-sh64/hardware.h b/include/asm-sh64/hardware.h
deleted file mode 100644
index 931c1ad80847..000000000000
--- a/include/asm-sh64/hardware.h
+++ /dev/null
@@ -1,22 +0,0 @@
-#ifndef __ASM_SH64_HARDWARE_H
-#define __ASM_SH64_HARDWARE_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/hardware.h
- *
- * Copyright (C) 2002 Stuart Menefy
- * Copyright (C) 2003 Paul Mundt
- *
- * Defitions of the locations of registers in the physical address space.
- */
-
-#define	PHYS_PERIPHERAL_BLOCK	0x09000000
-#define PHYS_DMAC_BLOCK		0x0e000000
-#define PHYS_PCI_BLOCK		0x60000000
-#define PHYS_EMI_BLOCK		0xff000000
-
-#endif /* __ASM_SH64_HARDWARE_H */
diff --git a/include/asm-sh64/irq.h b/include/asm-sh64/irq.h
deleted file mode 100644
index 5c9e6a873aeb..000000000000
--- a/include/asm-sh64/irq.h
+++ /dev/null
@@ -1,144 +0,0 @@
-#ifndef __ASM_SH64_IRQ_H
-#define __ASM_SH64_IRQ_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/irq.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- */
-
-
-/*
- * Encoded IRQs are not considered worth to be supported.
- * Main reason is that there's no per-encoded-interrupt
- * enable/disable mechanism (as there was in SH3/4).
- * An all enabled/all disabled is worth only if there's
- * a cascaded IC to disable/enable/ack on. Until such
- * IC is available there's no such support.
- *
- * Presumably Encoded IRQs may use extra IRQs beyond 64,
- * below. Some logic must be added to cope with IRQ_IRL?
- * in an exclusive way.
- *
- * Priorities are set at Platform level, when IRQ_IRL0-3
- * are set to 0 Encoding is allowed. Otherwise it's not
- * allowed.
- */
-
-/* Independent IRQs */
-#define IRQ_IRL0	0
-#define IRQ_IRL1	1
-#define IRQ_IRL2	2
-#define IRQ_IRL3	3
-
-#define IRQ_INTA	4
-#define IRQ_INTB	5
-#define IRQ_INTC	6
-#define IRQ_INTD	7
-
-#define IRQ_SERR	12
-#define IRQ_ERR		13
-#define IRQ_PWR3	14
-#define IRQ_PWR2	15
-#define IRQ_PWR1	16
-#define IRQ_PWR0	17
-
-#define IRQ_DMTE0	18
-#define IRQ_DMTE1	19
-#define IRQ_DMTE2	20
-#define IRQ_DMTE3	21
-#define IRQ_DAERR	22
-
-#define IRQ_TUNI0	32
-#define IRQ_TUNI1	33
-#define IRQ_TUNI2	34
-#define IRQ_TICPI2	35
-
-#define IRQ_ATI		36
-#define IRQ_PRI		37
-#define IRQ_CUI		38
-
-#define IRQ_ERI		39
-#define IRQ_RXI		40
-#define IRQ_BRI		41
-#define IRQ_TXI		42
-
-#define IRQ_ITI		63
-
-#define NR_INTC_IRQS	64
-
-#ifdef CONFIG_SH_CAYMAN
-#define NR_EXT_IRQS     32
-#define START_EXT_IRQS  64
-
-/* PCI bus 2 uses encoded external interrupts on the Cayman board */
-#define IRQ_P2INTA      (START_EXT_IRQS + (3*8) + 0)
-#define IRQ_P2INTB      (START_EXT_IRQS + (3*8) + 1)
-#define IRQ_P2INTC      (START_EXT_IRQS + (3*8) + 2)
-#define IRQ_P2INTD      (START_EXT_IRQS + (3*8) + 3)
-
-#define I8042_KBD_IRQ	(START_EXT_IRQS + 2)
-#define I8042_AUX_IRQ	(START_EXT_IRQS + 6)
-
-#define IRQ_CFCARD	(START_EXT_IRQS + 7)
-#define IRQ_PCMCIA	(0)
-
-#else
-#define NR_EXT_IRQS	0
-#endif
-
-#define NR_IRQS		(NR_INTC_IRQS+NR_EXT_IRQS)
-
-
-/* Default IRQs, fixed */
-#define TIMER_IRQ	IRQ_TUNI0
-#define RTC_IRQ		IRQ_CUI
-
-/* Default Priorities, Platform may choose differently */
-#define	NO_PRIORITY	0	/* Disabled */
-#define TIMER_PRIORITY	2
-#define RTC_PRIORITY	TIMER_PRIORITY
-#define SCIF_PRIORITY	3
-#define INTD_PRIORITY	3
-#define	IRL3_PRIORITY	4
-#define INTC_PRIORITY	6
-#define	IRL2_PRIORITY	7
-#define INTB_PRIORITY	9
-#define	IRL1_PRIORITY	10
-#define INTA_PRIORITY	12
-#define	IRL0_PRIORITY	13
-#define TOP_PRIORITY	15
-
-extern int intc_evt_to_irq[(0xE20/0x20)+1];
-int intc_irq_describe(char* p, int irq);
-
-#define irq_canonicalize(irq)	(irq)
-
-#ifdef CONFIG_SH_CAYMAN
-int cayman_irq_demux(int evt);
-int cayman_irq_describe(char* p, int irq);
-#define irq_demux(x) cayman_irq_demux(x)
-#define irq_describe(p, x) cayman_irq_describe(p, x)
-#else
-#define irq_demux(x) (intc_evt_to_irq[x])
-#define irq_describe(p, x) intc_irq_describe(p, x)
-#endif
-
-/*
- * Function for "on chip support modules".
- */
-
-/*
- * SH-5 supports Priority based interrupts only.
- * Interrupt priorities are defined at platform level.
- */
-#define set_ipr_data(a, b, c, d)
-#define make_ipr_irq(a)
-#define make_imask_irq(a)
-
-#endif /* __ASM_SH64_IRQ_H */
diff --git a/include/asm-sh64/keyboard.h b/include/asm-sh64/keyboard.h
deleted file mode 100644
index 0b01c3beb2f8..000000000000
--- a/include/asm-sh64/keyboard.h
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- *  linux/include/asm-shmedia/keyboard.h
- *
- * Copied from i386 version:
- *    Created 3 Nov 1996 by Geert Uytterhoeven
- */
-
-/*
- *  This file contains the i386 architecture specific keyboard definitions
- */
-
-#ifndef __ASM_SH64_KEYBOARD_H
-#define __ASM_SH64_KEYBOARD_H
-
-#ifdef __KERNEL__
-
-#include <linux/kernel.h>
-#include <linux/ioport.h>
-#include <asm/io.h>
-
-#ifdef CONFIG_SH_CAYMAN
-#define KEYBOARD_IRQ			(START_EXT_IRQS + 2) /* SMSC SuperIO IRQ 1 */
-#endif
-#define DISABLE_KBD_DURING_INTERRUPTS	0
-
-extern int pckbd_setkeycode(unsigned int scancode, unsigned int keycode);
-extern int pckbd_getkeycode(unsigned int scancode);
-extern int pckbd_translate(unsigned char scancode, unsigned char *keycode,
-			   char raw_mode);
-extern char pckbd_unexpected_up(unsigned char keycode);
-extern void pckbd_leds(unsigned char leds);
-extern void pckbd_init_hw(void);
-
-#define kbd_setkeycode		pckbd_setkeycode
-#define kbd_getkeycode		pckbd_getkeycode
-#define kbd_translate		pckbd_translate
-#define kbd_unexpected_up	pckbd_unexpected_up
-#define kbd_leds		pckbd_leds
-#define kbd_init_hw		pckbd_init_hw
-
-/* resource allocation */
-#define kbd_request_region()
-#define kbd_request_irq(handler) request_irq(KEYBOARD_IRQ, handler, 0, \
-                                             "keyboard", NULL)
-
-/* How to access the keyboard macros on this platform.  */
-#define kbd_read_input() inb(KBD_DATA_REG)
-#define kbd_read_status() inb(KBD_STATUS_REG)
-#define kbd_write_output(val) outb(val, KBD_DATA_REG)
-#define kbd_write_command(val) outb(val, KBD_CNTL_REG)
-
-/* Some stoneage hardware needs delays after some operations.  */
-#define kbd_pause() do { } while(0)
-
-/*
- * Machine specific bits for the PS/2 driver
- */
-
-#ifdef CONFIG_SH_CAYMAN
-#define AUX_IRQ (START_EXT_IRQS + 6) /* SMSC SuperIO IRQ12 */
-#endif
-
-#define aux_request_irq(hand, dev_id)					\
-	request_irq(AUX_IRQ, hand, IRQF_SHARED, "PS2 Mouse", dev_id)
-
-#define aux_free_irq(dev_id) free_irq(AUX_IRQ, dev_id)
-
-#endif /* __KERNEL__ */
-#endif /* __ASM_SH64_KEYBOARD_H */
-
diff --git a/include/asm-sh64/pci.h b/include/asm-sh64/pci.h
deleted file mode 100644
index 18055dbbb4b5..000000000000
--- a/include/asm-sh64/pci.h
+++ /dev/null
@@ -1,102 +0,0 @@
-#ifndef __ASM_SH64_PCI_H
-#define __ASM_SH64_PCI_H
-
-#ifdef __KERNEL__
-
-#include <linux/dma-mapping.h>
-
-/* Can be used to override the logic in pci_scan_bus for skipping
-   already-configured bus numbers - to be used for buggy BIOSes
-   or architectures with incomplete PCI setup by the loader */
-
-#define pcibios_assign_all_busses()     1
-
-/*
- * These are currently the correct values for the STM overdrive board
- * We need some way of setting this on a board specific way, it will
- * not be the same on other boards I think
- */
-#if defined(CONFIG_CPU_SUBTYPE_SH5_101) || defined(CONFIG_CPU_SUBTYPE_SH5_103)
-#define PCIBIOS_MIN_IO          0x2000
-#define PCIBIOS_MIN_MEM         0x40000000
-#endif
-
-extern void pcibios_set_master(struct pci_dev *dev);
-
-/*
- * Set penalize isa irq function
- */
-static inline void pcibios_penalize_isa_irq(int irq, int active)
-{
-	/* We don't do dynamic PCI IRQ allocation */
-}
-
-/* Dynamic DMA mapping stuff.
- * SuperH has everything mapped statically like x86.
- */
-
-/* The PCI address space does equal the physical memory
- * address space.  The networking and block device layers use
- * this boolean for bounce buffer decisions.
- */
-#define PCI_DMA_BUS_IS_PHYS	(1)
-
-#include <linux/types.h>
-#include <linux/slab.h>
-#include <asm/scatterlist.h>
-#include <linux/string.h>
-#include <asm/io.h>
-
-/* pci_unmap_{single,page} being a nop depends upon the
- * configuration.
- */
-#ifdef CONFIG_SH_PCIDMA_NONCOHERENT
-#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)	\
-	dma_addr_t ADDR_NAME;
-#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)		\
-	__u32 LEN_NAME;
-#define pci_unmap_addr(PTR, ADDR_NAME)			\
-	((PTR)->ADDR_NAME)
-#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)		\
-	(((PTR)->ADDR_NAME) = (VAL))
-#define pci_unmap_len(PTR, LEN_NAME)			\
-	((PTR)->LEN_NAME)
-#define pci_unmap_len_set(PTR, LEN_NAME, VAL)		\
-	(((PTR)->LEN_NAME) = (VAL))
-#else
-#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
-#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
-#define pci_unmap_addr(PTR, ADDR_NAME)		(0)
-#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)	do { } while (0)
-#define pci_unmap_len(PTR, LEN_NAME)		(0)
-#define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
-#endif
-
-#ifdef CONFIG_PCI
-static inline void pci_dma_burst_advice(struct pci_dev *pdev,
-					enum pci_dma_burst_strategy *strat,
-					unsigned long *strategy_parameter)
-{
-	*strat = PCI_DMA_BURST_INFINITY;
-	*strategy_parameter = ~0UL;
-}
-#endif
-
-/* Board-specific fixup routines. */
-extern void pcibios_fixup(void);
-extern void pcibios_fixup_irqs(void);
-
-#ifdef CONFIG_PCI_AUTO
-extern int pciauto_assign_resources(int busno, struct pci_channel *hose);
-#endif
-
-#endif /* __KERNEL__ */
-
-/* generic pci stuff */
-#include <asm-generic/pci.h>
-
-/* generic DMA-mapping stuff */
-#include <asm-generic/pci-dma-compat.h>
-
-#endif /* __ASM_SH64_PCI_H */
-
diff --git a/include/asm-sh64/platform.h b/include/asm-sh64/platform.h
deleted file mode 100644
index bd0d9c405a80..000000000000
--- a/include/asm-sh64/platform.h
+++ /dev/null
@@ -1,64 +0,0 @@
-#ifndef __ASM_SH64_PLATFORM_H
-#define __ASM_SH64_PLATFORM_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/platform.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- * benedict.gaster@superh.com:	 3rd May 2002
- *    Added support for ramdisk, removing statically linked romfs at the same time.
- */
-
-#include <linux/ioport.h>
-#include <asm/irq.h>
-
-
-/*
- * Platform definition structure.
- */
-struct sh64_platform {
-	unsigned int readonly_rootfs;
-	unsigned int ramdisk_flags;
-	unsigned int initial_root_dev;
-	unsigned int loader_type;
-	unsigned int initrd_start;
-	unsigned int initrd_size;
-	unsigned int fpu_flags;
-	unsigned int io_res_count;
-	unsigned int kram_res_count;
-	unsigned int xram_res_count;
-	unsigned int rom_res_count;
-	struct resource *io_res_p;
-	struct resource *kram_res_p;
-	struct resource *xram_res_p;
-	struct resource *rom_res_p;
-};
-
-extern struct sh64_platform platform_parms;
-
-extern unsigned long long memory_start, memory_end;
-
-extern unsigned long long fpu_in_use;
-
-extern int platform_int_priority[NR_INTC_IRQS];
-
-#define FPU_FLAGS		(platform_parms.fpu_flags)
-#define STANDARD_IO_RESOURCES	(platform_parms.io_res_count)
-#define STANDARD_KRAM_RESOURCES	(platform_parms.kram_res_count)
-#define STANDARD_XRAM_RESOURCES	(platform_parms.xram_res_count)
-#define STANDARD_ROM_RESOURCES	(platform_parms.rom_res_count)
-
-/*
- * Kernel Memory description, Respectively:
- * code = last but one memory descriptor
- * data = last memory descriptor
- */
-#define code_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 2])
-#define data_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 1])
-
-#endif	/* __ASM_SH64_PLATFORM_H */
diff --git a/include/asm-sh64/thread_info.h b/include/asm-sh64/thread_info.h
deleted file mode 100644
index f6d5117c53af..000000000000
--- a/include/asm-sh64/thread_info.h
+++ /dev/null
@@ -1,91 +0,0 @@
-#ifndef __ASM_SH64_THREAD_INFO_H
-#define __ASM_SH64_THREAD_INFO_H
-
-/*
- * SuperH 5 version
- * Copyright (C) 2003  Paul Mundt
- */
-
-#ifdef __KERNEL__
-
-#ifndef __ASSEMBLY__
-#include <asm/registers.h>
-
-/*
- * low level task data that entry.S needs immediate access to
- * - this struct should fit entirely inside of one cache line
- * - this struct shares the supervisor stack pages
- * - if the contents of this structure are changed, the assembly constants must also be changed
- */
-struct thread_info {
-	struct task_struct	*task;		/* main task structure */
-	struct exec_domain	*exec_domain;	/* execution domain */
-	unsigned long		flags;		/* low level flags */
-	/* Put the 4 32-bit fields together to make asm offsetting easier. */
-	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
-	__u16			cpu;
-
-	mm_segment_t		addr_limit;
-	struct restart_block	restart_block;
-
-	__u8			supervisor_stack[0];
-};
-
-/*
- * macros/functions for gaining access to the thread information structure
- */
-#define INIT_THREAD_INFO(tsk)			\
-{						\
-	.task		= &tsk,			\
-	.exec_domain	= &default_exec_domain,	\
-	.flags		= 0,			\
-	.cpu		= 0,			\
-	.preempt_count	= 1,			\
-	.addr_limit     = KERNEL_DS,            \
-	.restart_block	= {			\
-		.fn = do_no_restart_syscall,	\
-	},					\
-}
-
-#define init_thread_info	(init_thread_union.thread_info)
-#define init_stack		(init_thread_union.stack)
-
-/* how to get the thread information struct from C */
-static inline struct thread_info *current_thread_info(void)
-{
-	struct thread_info *ti;
-
-	__asm__ __volatile__ ("getcon " __KCR0 ", %0\n\t" : "=r" (ti));
-
-	return ti;
-}
-
-/* thread information allocation */
-
-
-
-#define alloc_thread_info(ti) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
-
-#endif /* __ASSEMBLY__ */
-
-#define THREAD_SIZE  8192
-
-#define PREEMPT_ACTIVE		0x10000000
-
-/* thread information flags */
-#define TIF_SYSCALL_TRACE	0	/* syscall trace active */
-#define TIF_SIGPENDING		2	/* signal pending */
-#define TIF_NEED_RESCHED	3	/* rescheduling necessary */
-#define TIF_MEMDIE		4
-#define TIF_RESTORE_SIGMASK	5	/* Restore signal mask in do_signal */
-
-#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
-#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
-#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
-#define _TIF_MEMDIE		(1 << TIF_MEMDIE)
-#define _TIF_RESTORE_SIGMASK	(1 << TIF_RESTORE_SIGMASK)
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH64_THREAD_INFO_H */
-- 
cgit v1.2.3


From 18bc81319b438ae3266e1b2653ce874912dae891 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 21 Nov 2007 23:16:33 +0900
Subject: sh: Get the mach-cayman IRQ support building.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/irq/Makefile        |   1 +
 arch/sh/kernel/cpu/irq/intc-sh5.c      | 257 +++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh5/Makefile        |   4 -
 arch/sh/kernel/cpu/sh5/setup-sh5-101.c |  15 --
 arch/sh/mach-cayman/Makefile           |   8 +-
 arch/sh/mach-cayman/iomap.c            |  22 ---
 arch/sh/mach-cayman/irq.c              |  26 ++--
 arch/sh/mach-cayman/setup.c            |  94 +++---------
 arch/sh64/kernel/irq_intc.c            | 272 ---------------------------------
 include/asm-sh/cpu-sh5/irq.h           |  28 +---
 include/asm-sh/platform.h              |  64 --------
 11 files changed, 295 insertions(+), 496 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/irq/intc-sh5.c
 delete mode 100644 arch/sh/kernel/cpu/sh5/setup-sh5-101.c
 delete mode 100644 arch/sh/mach-cayman/iomap.c
 delete mode 100644 arch/sh64/kernel/irq_intc.c
 delete mode 100644 include/asm-sh/platform.h

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/irq/Makefile b/arch/sh/kernel/cpu/irq/Makefile
index 81d37e948b86..cc1836e47a5d 100644
--- a/arch/sh/kernel/cpu/irq/Makefile
+++ b/arch/sh/kernel/cpu/irq/Makefile
@@ -4,5 +4,6 @@
 obj-y	+= intc.o
 
 obj-$(CONFIG_SUPERH32)			+= imask.o
+obj-$(CONFIG_CPU_SH5)			+= intc-sh5.o
 obj-$(CONFIG_CPU_HAS_IPR_IRQ)		+= ipr.o
 obj-$(CONFIG_CPU_HAS_MASKREG_IRQ)	+= maskreg.o
diff --git a/arch/sh/kernel/cpu/irq/intc-sh5.c b/arch/sh/kernel/cpu/irq/intc-sh5.c
new file mode 100644
index 000000000000..49b845a31fff
--- /dev/null
+++ b/arch/sh/kernel/cpu/irq/intc-sh5.c
@@ -0,0 +1,257 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * arch/sh64/kernel/irq_intc.c
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003  Paul Mundt
+ *
+ * Interrupt Controller support for SH5 INTC.
+ * Per-interrupt selective. IRLM=0 (Fixed priority) is not
+ * supported being useless without a cascaded interrupt
+ * controller.
+ *
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+#include <asm/cpu/irq.h>
+#include <asm/page.h>
+
+/*
+ * Maybe the generic Peripheral block could move to a more
+ * generic include file. INTC Block will be defined here
+ * and only here to make INTC self-contained in a single
+ * file.
+ */
+#define	INTC_BLOCK_OFFSET	0x01000000
+
+/* Base */
+#define INTC_BASE		PHYS_PERIPHERAL_BLOCK + \
+				INTC_BLOCK_OFFSET
+
+/* Address */
+#define INTC_ICR_SET		(intc_virt + 0x0)
+#define INTC_ICR_CLEAR		(intc_virt + 0x8)
+#define INTC_INTPRI_0		(intc_virt + 0x10)
+#define INTC_INTSRC_0		(intc_virt + 0x50)
+#define INTC_INTSRC_1		(intc_virt + 0x58)
+#define INTC_INTREQ_0		(intc_virt + 0x60)
+#define INTC_INTREQ_1		(intc_virt + 0x68)
+#define INTC_INTENB_0		(intc_virt + 0x70)
+#define INTC_INTENB_1		(intc_virt + 0x78)
+#define INTC_INTDSB_0		(intc_virt + 0x80)
+#define INTC_INTDSB_1		(intc_virt + 0x88)
+
+#define INTC_ICR_IRLM		0x1
+#define	INTC_INTPRI_PREGS	8		/* 8 Priority Registers */
+#define	INTC_INTPRI_PPREG	8		/* 8 Priorities per Register */
+
+
+/*
+ * Mapper between the vector ordinal and the IRQ number
+ * passed to kernel/device drivers.
+ */
+int intc_evt_to_irq[(0xE20/0x20)+1] = {
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x000 - 0x0E0 */
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x100 - 0x1E0 */
+	 0,  0,  0,  0,  0,  1,  0,  0,	/* 0x200 - 0x2E0 */
+	 2,  0,  0,  3,  0,  0,  0, -1,	/* 0x300 - 0x3E0 */
+	32, 33, 34, 35, 36, 37, 38, -1,	/* 0x400 - 0x4E0 */
+	-1, -1, -1, 63, -1, -1, -1, -1,	/* 0x500 - 0x5E0 */
+	-1, -1, 18, 19, 20, 21, 22, -1,	/* 0x600 - 0x6E0 */
+	39, 40, 41, 42, -1, -1, -1, -1,	/* 0x700 - 0x7E0 */
+	 4,  5,  6,  7, -1, -1, -1, -1,	/* 0x800 - 0x8E0 */
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x900 - 0x9E0 */
+	12, 13, 14, 15, 16, 17, -1, -1,	/* 0xA00 - 0xAE0 */
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xB00 - 0xBE0 */
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xC00 - 0xCE0 */
+	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xD00 - 0xDE0 */
+	-1, -1				/* 0xE00 - 0xE20 */
+};
+
+/*
+ * Opposite mapper.
+ */
+static int IRQ_to_vectorN[NR_INTC_IRQS] = {
+	0x12, 0x15, 0x18, 0x1B, 0x40, 0x41, 0x42, 0x43, /*  0- 7 */
+	  -1,   -1,   -1,   -1, 0x50, 0x51, 0x52, 0x53,	/*  8-15 */
+	0x54, 0x55, 0x32, 0x33, 0x34, 0x35, 0x36,   -1, /* 16-23 */
+	  -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1, /* 24-31 */
+	0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x38,	/* 32-39 */
+        0x39, 0x3A, 0x3B,   -1,   -1,   -1,   -1,   -1, /* 40-47 */
+	  -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1, /* 48-55 */
+	  -1,   -1,   -1,   -1,   -1,   -1,   -1, 0x2B, /* 56-63 */
+
+};
+
+static unsigned long intc_virt;
+
+static unsigned int startup_intc_irq(unsigned int irq);
+static void shutdown_intc_irq(unsigned int irq);
+static void enable_intc_irq(unsigned int irq);
+static void disable_intc_irq(unsigned int irq);
+static void mask_and_ack_intc(unsigned int);
+static void end_intc_irq(unsigned int irq);
+
+static struct hw_interrupt_type intc_irq_type = {
+	.typename = "INTC",
+	.startup = startup_intc_irq,
+	.shutdown = shutdown_intc_irq,
+	.enable = enable_intc_irq,
+	.disable = disable_intc_irq,
+	.ack = mask_and_ack_intc,
+	.end = end_intc_irq
+};
+
+static int irlm;		/* IRL mode */
+
+static unsigned int startup_intc_irq(unsigned int irq)
+{
+	enable_intc_irq(irq);
+	return 0; /* never anything pending */
+}
+
+static void shutdown_intc_irq(unsigned int irq)
+{
+	disable_intc_irq(irq);
+}
+
+static void enable_intc_irq(unsigned int irq)
+{
+	unsigned long reg;
+	unsigned long bitmask;
+
+	if ((irq <= IRQ_IRL3) && (irlm == NO_PRIORITY))
+		printk("Trying to use straight IRL0-3 with an encoding platform.\n");
+
+	if (irq < 32) {
+		reg = INTC_INTENB_0;
+		bitmask = 1 << irq;
+	} else {
+		reg = INTC_INTENB_1;
+		bitmask = 1 << (irq - 32);
+	}
+
+	ctrl_outl(bitmask, reg);
+}
+
+static void disable_intc_irq(unsigned int irq)
+{
+	unsigned long reg;
+	unsigned long bitmask;
+
+	if (irq < 32) {
+		reg = INTC_INTDSB_0;
+		bitmask = 1 << irq;
+	} else {
+		reg = INTC_INTDSB_1;
+		bitmask = 1 << (irq - 32);
+	}
+
+	ctrl_outl(bitmask, reg);
+}
+
+static void mask_and_ack_intc(unsigned int irq)
+{
+	disable_intc_irq(irq);
+}
+
+static void end_intc_irq(unsigned int irq)
+{
+	enable_intc_irq(irq);
+}
+
+/* For future use, if we ever support IRLM=0) */
+void make_intc_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].chip = &intc_irq_type;
+	disable_intc_irq(irq);
+}
+
+#if defined(CONFIG_PROC_FS) && defined(CONFIG_SYSCTL)
+int intc_irq_describe(char* p, int irq)
+{
+	if (irq < NR_INTC_IRQS)
+		return sprintf(p, "(0x%3x)", IRQ_to_vectorN[irq]*0x20);
+	else
+		return 0;
+}
+#endif
+
+void __init plat_irq_setup(void)
+{
+        unsigned long long __dummy0, __dummy1=~0x00000000100000f0;
+	unsigned long reg;
+	unsigned long data;
+	int i;
+
+	intc_virt = onchip_remap(INTC_BASE, 1024, "INTC");
+	if (!intc_virt) {
+		panic("Unable to remap INTC\n");
+	}
+
+
+	/* Set default: per-line enable/disable, priority driven ack/eoi */
+	for (i = 0; i < NR_INTC_IRQS; i++) {
+		if (platform_int_priority[i] != NO_PRIORITY) {
+			irq_desc[i].chip = &intc_irq_type;
+		}
+	}
+
+
+	/* Disable all interrupts and set all priorities to 0 to avoid trouble */
+	ctrl_outl(-1, INTC_INTDSB_0);
+	ctrl_outl(-1, INTC_INTDSB_1);
+
+	for (reg = INTC_INTPRI_0, i = 0; i < INTC_INTPRI_PREGS; i++, reg += 8)
+		ctrl_outl( NO_PRIORITY, reg);
+
+
+	/* Set IRLM */
+	/* If all the priorities are set to 'no priority', then
+	 * assume we are using encoded mode.
+	 */
+	irlm = platform_int_priority[IRQ_IRL0] + platform_int_priority[IRQ_IRL1] + \
+		platform_int_priority[IRQ_IRL2] + platform_int_priority[IRQ_IRL3];
+
+	if (irlm == NO_PRIORITY) {
+		/* IRLM = 0 */
+		reg = INTC_ICR_CLEAR;
+		i = IRQ_INTA;
+		printk("Trying to use encoded IRL0-3. IRLs unsupported.\n");
+	} else {
+		/* IRLM = 1 */
+		reg = INTC_ICR_SET;
+		i = IRQ_IRL0;
+	}
+	ctrl_outl(INTC_ICR_IRLM, reg);
+
+	/* Set interrupt priorities according to platform description */
+	for (data = 0, reg = INTC_INTPRI_0; i < NR_INTC_IRQS; i++) {
+		data |= platform_int_priority[i] << ((i % INTC_INTPRI_PPREG) * 4);
+		if ((i % INTC_INTPRI_PPREG) == (INTC_INTPRI_PPREG - 1)) {
+			/* Upon the 7th, set Priority Register */
+			ctrl_outl(data, reg);
+			data = 0;
+			reg += 8;
+		}
+	}
+
+	/*
+	 * And now let interrupts come in.
+	 * sti() is not enough, we need to
+	 * lower priority, too.
+	 */
+        __asm__ __volatile__("getcon    " __SR ", %0\n\t"
+                             "and       %0, %1, %0\n\t"
+                             "putcon    %0, " __SR "\n\t"
+                             : "=&r" (__dummy0)
+                             : "r" (__dummy1));
+}
diff --git a/arch/sh/kernel/cpu/sh5/Makefile b/arch/sh/kernel/cpu/sh5/Makefile
index 0ef257b72e5d..8646363e9ded 100644
--- a/arch/sh/kernel/cpu/sh5/Makefile
+++ b/arch/sh/kernel/cpu/sh5/Makefile
@@ -5,7 +5,3 @@ obj-y := entry.o probe.o switchto.o
 
 obj-$(CONFIG_SH_FPU)		+= fpu.o
 obj-$(CONFIG_KALLSYMS)		+= unwind.o
-
-# CPU subtype setup
-obj-$(CONFIG_CPU_SUBTYPE_SH5_101)	+= setup-sh5-101.o
-obj-$(CONFIG_CPU_SUBTYPE_SH5_103)	+= setup-sh5-101.o
diff --git a/arch/sh/kernel/cpu/sh5/setup-sh5-101.c b/arch/sh/kernel/cpu/sh5/setup-sh5-101.c
deleted file mode 100644
index 3680012d7109..000000000000
--- a/arch/sh/kernel/cpu/sh5/setup-sh5-101.c
+++ /dev/null
@@ -1,15 +0,0 @@
-/*
- * SH5-101 Setup
- *
- *  Copyright (C) 2007  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#include <linux/init.h>
-
-void __init plat_irq_setup(void)
-{
-	/* do nothing - all IRL interrupts are handled by the board code */
-}
diff --git a/arch/sh/mach-cayman/Makefile b/arch/sh/mach-cayman/Makefile
index 67a2258bf8c4..489a8f867368 100644
--- a/arch/sh/mach-cayman/Makefile
+++ b/arch/sh/mach-cayman/Makefile
@@ -1,11 +1,5 @@
 #
 # Makefile for the Hitachi Cayman specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
-
-obj-y := setup.o irq.o iomap.o
+obj-y := setup.o irq.o
 obj-$(CONFIG_HEARTBEAT)	+= led.o
-
diff --git a/arch/sh/mach-cayman/iomap.c b/arch/sh/mach-cayman/iomap.c
deleted file mode 100644
index a5c645f02d57..000000000000
--- a/arch/sh/mach-cayman/iomap.c
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * arch/sh64/mach-cayman/iomap.c
- *
- * Cayman iomap interface
- *
- * Copyright (C) 2004  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#include <asm/io.h>
-#include <asm/cayman.h>
-
-void __iomem *ioport_map(unsigned long port, unsigned int len)
-{
-	if (port < 0x400)
-		return (void __iomem *)((port << 2) | smsc_superio_virt);
-
-	return (void __iomem *)port;
-}
-
diff --git a/arch/sh/mach-cayman/irq.c b/arch/sh/mach-cayman/irq.c
index aaad36d37d1f..30ec7bebfaf1 100644
--- a/arch/sh/mach-cayman/irq.c
+++ b/arch/sh/mach-cayman/irq.c
@@ -1,24 +1,26 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/irq_cayman.c
- *
- * SH-5 Cayman Interrupt Support
+ * arch/sh/mach-cayman/irq.c - SH-5 Cayman Interrupt Support
  *
  * This file handles the board specific parts of the Cayman interrupt system
  *
  * Copyright (C) 2002 Stuart Menefy
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
-#include <asm/irq.h>
-#include <asm/page.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include <linux/irq.h>
 #include <linux/interrupt.h>
 #include <linux/signal.h>
-#include <asm/cayman.h>
+#include <asm/cpu/irq.h>
+#include <asm/page.h>
+
+/* Setup for the SMSC FDC37C935 / LAN91C100FD */
+#define SMSC_IRQ         IRQ_IRL1
+
+/* Setup for PCI Bus 2, which transmits interrupts via the EPLD */
+#define PCI2_IRQ         IRQ_IRL3
 
 unsigned long epld_virt;
 
diff --git a/arch/sh/mach-cayman/setup.c b/arch/sh/mach-cayman/setup.c
index 726c520d7eb9..8c9fa472d8f5 100644
--- a/arch/sh/mach-cayman/setup.c
+++ b/arch/sh/mach-cayman/setup.c
@@ -1,28 +1,19 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/mach-cayman/setup.c
+ * arch/sh/mach-cayman/setup.c
  *
  * SH5 Cayman support
  *
- * This file handles the architecture-dependent parts of initialization
+ * Copyright (C) 2002  David J. Mckay & Benedict Gaster
+ * Copyright (C) 2003 - 2007  Paul Mundt
  *
- * Copyright David J. Mckay.
- * Needs major work!
- *
- * benedict.gaster@superh.com:	 3rd May 2002
- *    Added support for ramdisk, removing statically linked romfs at the same time.
- *
- * lethal@linux-sh.org:          15th May 2003
- *    Use the generic procfs cpuinfo interface, just return a valid board name.
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/init.h>
+#include <linux/io.h>
 #include <linux/kernel.h>
-#include <asm/platform.h>
-#include <asm/irq.h>
-#include <asm/io.h>
+#include <asm/cpu/irq.h>
 
 /*
  * Platform Dependent Interrupt Priorities.
@@ -96,42 +87,6 @@
 
 unsigned long smsc_superio_virt;
 
-/*
- * Platform dependent structures: maps and parms block.
- */
-struct resource io_resources[] = {
-	/* To be updated with external devices */
-};
-
-struct resource kram_resources[] = {
-	/* These must be last in the array */
-	{ .name = "Kernel code", .start = 0, .end = 0 },
-	/* These must be last in the array */
-	{ .name = "Kernel data", .start = 0, .end = 0 }
-};
-
-struct resource xram_resources[] = {
-	/* To be updated with external devices */
-};
-
-struct resource rom_resources[] = {
-	/* To be updated with external devices */
-};
-
-struct sh64_platform platform_parms = {
-	.readonly_rootfs =	1,
-	.initial_root_dev =	0x0100,
-	.loader_type =		1,
-	.io_res_p =		io_resources,
-	.io_res_count =		ARRAY_SIZE(io_resources),
-	.kram_res_p =		kram_resources,
-	.kram_res_count =	ARRAY_SIZE(kram_resources),
-	.xram_res_p =		xram_resources,
-	.xram_res_count =	ARRAY_SIZE(xram_resources),
-	.rom_res_p =		rom_resources,
-	.rom_res_count =	ARRAY_SIZE(rom_resources),
-};
-
 int platform_int_priority[NR_INTC_IRQS] = {
 	IR0, IR1, IR2, IR3, PCA, PCB, PCC, PCD,	/* IRQ  0- 7 */
 	RES, RES, RES, RES, SER, ERR, PW3, PW2,	/* IRQ  8-15 */
@@ -210,30 +165,23 @@ static int __init smsc_superio_setup(void)
 
 	return 0;
 }
-
-/* This is grotty, but, because kernel is always referenced on the link line
- * before any devices, this is safe.
- */
 __initcall(smsc_superio_setup);
 
-void __init platform_setup(void)
-{
-	/* Cayman platform leaves the decision to head.S, for now */
-	platform_parms.fpu_flags = fpu_in_use;
-}
-
-void __init platform_monitor(void)
+static void __iomem *cayman_ioport_map(unsigned long port, unsigned int len)
 {
-	/* Nothing yet .. */
-}
+	if (port < 0x400) {
+		extern unsigned long smsc_superio_virt;
+		return (void __iomem *)((port << 2) | smsc_superio_virt);
+	}
 
-void __init platform_reserve(void)
-{
-	/* Nothing yet .. */
+	return (void __iomem *)port;
 }
 
-const char *get_system_type(void)
-{
-	return "Hitachi Cayman";
-}
+extern void init_cayman_irq(void);
 
+static struct sh_machine_vector mv_cayman __initmv = {
+	.mv_name		= "Hitachi Cayman",
+	.mv_nr_irqs		= 64,
+	.mv_ioport_map		= cayman_ioport_map,
+	.mv_init_irq		= init_cayman_irq,
+};
diff --git a/arch/sh64/kernel/irq_intc.c b/arch/sh64/kernel/irq_intc.c
deleted file mode 100644
index 3b63a93198f2..000000000000
--- a/arch/sh64/kernel/irq_intc.c
+++ /dev/null
@@ -1,272 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/irq_intc.c
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2003  Paul Mundt
- *
- * Interrupt Controller support for SH5 INTC.
- * Per-interrupt selective. IRLM=0 (Fixed priority) is not
- * supported being useless without a cascaded interrupt
- * controller.
- *
- */
-
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/irq.h>
-#include <linux/kernel.h>
-#include <linux/stddef.h>
-#include <linux/bitops.h>       /* this includes also <asm/registers.h */
-                                /* which is required to remap register */
-                                /* names used into __asm__ blocks...   */
-
-#include <asm/hardware.h>
-#include <asm/platform.h>
-#include <asm/page.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-
-/*
- * Maybe the generic Peripheral block could move to a more
- * generic include file. INTC Block will be defined here
- * and only here to make INTC self-contained in a single
- * file.
- */
-#define	INTC_BLOCK_OFFSET	0x01000000
-
-/* Base */
-#define INTC_BASE		PHYS_PERIPHERAL_BLOCK + \
-				INTC_BLOCK_OFFSET
-
-/* Address */
-#define INTC_ICR_SET		(intc_virt + 0x0)
-#define INTC_ICR_CLEAR		(intc_virt + 0x8)
-#define INTC_INTPRI_0		(intc_virt + 0x10)
-#define INTC_INTSRC_0		(intc_virt + 0x50)
-#define INTC_INTSRC_1		(intc_virt + 0x58)
-#define INTC_INTREQ_0		(intc_virt + 0x60)
-#define INTC_INTREQ_1		(intc_virt + 0x68)
-#define INTC_INTENB_0		(intc_virt + 0x70)
-#define INTC_INTENB_1		(intc_virt + 0x78)
-#define INTC_INTDSB_0		(intc_virt + 0x80)
-#define INTC_INTDSB_1		(intc_virt + 0x88)
-
-#define INTC_ICR_IRLM		0x1
-#define	INTC_INTPRI_PREGS	8		/* 8 Priority Registers */
-#define	INTC_INTPRI_PPREG	8		/* 8 Priorities per Register */
-
-
-/*
- * Mapper between the vector ordinal and the IRQ number
- * passed to kernel/device drivers.
- */
-int intc_evt_to_irq[(0xE20/0x20)+1] = {
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x000 - 0x0E0 */
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x100 - 0x1E0 */
-	 0,  0,  0,  0,  0,  1,  0,  0,	/* 0x200 - 0x2E0 */
-	 2,  0,  0,  3,  0,  0,  0, -1,	/* 0x300 - 0x3E0 */
-	32, 33, 34, 35, 36, 37, 38, -1,	/* 0x400 - 0x4E0 */
-	-1, -1, -1, 63, -1, -1, -1, -1,	/* 0x500 - 0x5E0 */
-	-1, -1, 18, 19, 20, 21, 22, -1,	/* 0x600 - 0x6E0 */
-	39, 40, 41, 42, -1, -1, -1, -1,	/* 0x700 - 0x7E0 */
-	 4,  5,  6,  7, -1, -1, -1, -1,	/* 0x800 - 0x8E0 */
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0x900 - 0x9E0 */
-	12, 13, 14, 15, 16, 17, -1, -1,	/* 0xA00 - 0xAE0 */
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xB00 - 0xBE0 */
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xC00 - 0xCE0 */
-	-1, -1, -1, -1, -1, -1, -1, -1,	/* 0xD00 - 0xDE0 */
-	-1, -1				/* 0xE00 - 0xE20 */
-};
-
-/*
- * Opposite mapper.
- */
-static int IRQ_to_vectorN[NR_INTC_IRQS] = {
-	0x12, 0x15, 0x18, 0x1B, 0x40, 0x41, 0x42, 0x43, /*  0- 7 */
-	  -1,   -1,   -1,   -1, 0x50, 0x51, 0x52, 0x53,	/*  8-15 */
-	0x54, 0x55, 0x32, 0x33, 0x34, 0x35, 0x36,   -1, /* 16-23 */
-	  -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1, /* 24-31 */
-	0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x38,	/* 32-39 */
-        0x39, 0x3A, 0x3B,   -1,   -1,   -1,   -1,   -1, /* 40-47 */
-	  -1,   -1,   -1,   -1,   -1,   -1,   -1,   -1, /* 48-55 */
-	  -1,   -1,   -1,   -1,   -1,   -1,   -1, 0x2B, /* 56-63 */
-
-};
-
-static unsigned long intc_virt;
-
-static unsigned int startup_intc_irq(unsigned int irq);
-static void shutdown_intc_irq(unsigned int irq);
-static void enable_intc_irq(unsigned int irq);
-static void disable_intc_irq(unsigned int irq);
-static void mask_and_ack_intc(unsigned int);
-static void end_intc_irq(unsigned int irq);
-
-static struct hw_interrupt_type intc_irq_type = {
-	.typename = "INTC",
-	.startup = startup_intc_irq,
-	.shutdown = shutdown_intc_irq,
-	.enable = enable_intc_irq,
-	.disable = disable_intc_irq,
-	.ack = mask_and_ack_intc,
-	.end = end_intc_irq
-};
-
-static int irlm;		/* IRL mode */
-
-static unsigned int startup_intc_irq(unsigned int irq)
-{
-	enable_intc_irq(irq);
-	return 0; /* never anything pending */
-}
-
-static void shutdown_intc_irq(unsigned int irq)
-{
-	disable_intc_irq(irq);
-}
-
-static void enable_intc_irq(unsigned int irq)
-{
-	unsigned long reg;
-	unsigned long bitmask;
-
-	if ((irq <= IRQ_IRL3) && (irlm == NO_PRIORITY))
-		printk("Trying to use straight IRL0-3 with an encoding platform.\n");
-
-	if (irq < 32) {
-		reg = INTC_INTENB_0;
-		bitmask = 1 << irq;
-	} else {
-		reg = INTC_INTENB_1;
-		bitmask = 1 << (irq - 32);
-	}
-
-	ctrl_outl(bitmask, reg);
-}
-
-static void disable_intc_irq(unsigned int irq)
-{
-	unsigned long reg;
-	unsigned long bitmask;
-
-	if (irq < 32) {
-		reg = INTC_INTDSB_0;
-		bitmask = 1 << irq;
-	} else {
-		reg = INTC_INTDSB_1;
-		bitmask = 1 << (irq - 32);
-	}
-
-	ctrl_outl(bitmask, reg);
-}
-
-static void mask_and_ack_intc(unsigned int irq)
-{
-	disable_intc_irq(irq);
-}
-
-static void end_intc_irq(unsigned int irq)
-{
-	enable_intc_irq(irq);
-}
-
-/* For future use, if we ever support IRLM=0) */
-void make_intc_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &intc_irq_type;
-	disable_intc_irq(irq);
-}
-
-#if defined(CONFIG_PROC_FS) && defined(CONFIG_SYSCTL)
-int intc_irq_describe(char* p, int irq)
-{
-	if (irq < NR_INTC_IRQS)
-		return sprintf(p, "(0x%3x)", IRQ_to_vectorN[irq]*0x20);
-	else
-		return 0;
-}
-#endif
-
-void __init init_IRQ(void)
-{
-        unsigned long long __dummy0, __dummy1=~0x00000000100000f0;
-	unsigned long reg;
-	unsigned long data;
-	int i;
-
-	intc_virt = onchip_remap(INTC_BASE, 1024, "INTC");
-	if (!intc_virt) {
-		panic("Unable to remap INTC\n");
-	}
-
-
-	/* Set default: per-line enable/disable, priority driven ack/eoi */
-	for (i = 0; i < NR_INTC_IRQS; i++) {
-		if (platform_int_priority[i] != NO_PRIORITY) {
-			irq_desc[i].chip = &intc_irq_type;
-		}
-	}
-
-
-	/* Disable all interrupts and set all priorities to 0 to avoid trouble */
-	ctrl_outl(-1, INTC_INTDSB_0);
-	ctrl_outl(-1, INTC_INTDSB_1);
-
-	for (reg = INTC_INTPRI_0, i = 0; i < INTC_INTPRI_PREGS; i++, reg += 8)
-		ctrl_outl( NO_PRIORITY, reg);
-
-
-	/* Set IRLM */
-	/* If all the priorities are set to 'no priority', then
-	 * assume we are using encoded mode.
-	 */
-	irlm = platform_int_priority[IRQ_IRL0] + platform_int_priority[IRQ_IRL1] + \
-		platform_int_priority[IRQ_IRL2] + platform_int_priority[IRQ_IRL3];
-
-	if (irlm == NO_PRIORITY) {
-		/* IRLM = 0 */
-		reg = INTC_ICR_CLEAR;
-		i = IRQ_INTA;
-		printk("Trying to use encoded IRL0-3. IRLs unsupported.\n");
-	} else {
-		/* IRLM = 1 */
-		reg = INTC_ICR_SET;
-		i = IRQ_IRL0;
-	}
-	ctrl_outl(INTC_ICR_IRLM, reg);
-
-	/* Set interrupt priorities according to platform description */
-	for (data = 0, reg = INTC_INTPRI_0; i < NR_INTC_IRQS; i++) {
-		data |= platform_int_priority[i] << ((i % INTC_INTPRI_PPREG) * 4);
-		if ((i % INTC_INTPRI_PPREG) == (INTC_INTPRI_PPREG - 1)) {
-			/* Upon the 7th, set Priority Register */
-			ctrl_outl(data, reg);
-			data = 0;
-			reg += 8;
-		}
-	}
-
-#ifdef CONFIG_SH_CAYMAN
-	{
-		extern void init_cayman_irq(void);
-
-		init_cayman_irq();
-	}
-#endif
-
-	/*
-	 * And now let interrupts come in.
-	 * sti() is not enough, we need to
-	 * lower priority, too.
-	 */
-        __asm__ __volatile__("getcon    " __SR ", %0\n\t"
-                             "and       %0, %1, %0\n\t"
-                             "putcon    %0, " __SR "\n\t"
-                             : "=&r" (__dummy0)
-                             : "r" (__dummy1));
-}
diff --git a/include/asm-sh/cpu-sh5/irq.h b/include/asm-sh/cpu-sh5/irq.h
index 5c9e6a873aeb..f539baec8529 100644
--- a/include/asm-sh/cpu-sh5/irq.h
+++ b/include/asm-sh/cpu-sh5/irq.h
@@ -92,9 +92,6 @@
 #define NR_EXT_IRQS	0
 #endif
 
-#define NR_IRQS		(NR_INTC_IRQS+NR_EXT_IRQS)
-
-
 /* Default IRQs, fixed */
 #define TIMER_IRQ	IRQ_TUNI0
 #define RTC_IRQ		IRQ_CUI
@@ -116,29 +113,6 @@
 
 extern int intc_evt_to_irq[(0xE20/0x20)+1];
 int intc_irq_describe(char* p, int irq);
-
-#define irq_canonicalize(irq)	(irq)
-
-#ifdef CONFIG_SH_CAYMAN
-int cayman_irq_demux(int evt);
-int cayman_irq_describe(char* p, int irq);
-#define irq_demux(x) cayman_irq_demux(x)
-#define irq_describe(p, x) cayman_irq_describe(p, x)
-#else
-#define irq_demux(x) (intc_evt_to_irq[x])
-#define irq_describe(p, x) intc_irq_describe(p, x)
-#endif
-
-/*
- * Function for "on chip support modules".
- */
-
-/*
- * SH-5 supports Priority based interrupts only.
- * Interrupt priorities are defined at platform level.
- */
-#define set_ipr_data(a, b, c, d)
-#define make_ipr_irq(a)
-#define make_imask_irq(a)
+extern int platform_int_priority[NR_INTC_IRQS];
 
 #endif /* __ASM_SH64_IRQ_H */
diff --git a/include/asm-sh/platform.h b/include/asm-sh/platform.h
deleted file mode 100644
index bd0d9c405a80..000000000000
--- a/include/asm-sh/platform.h
+++ /dev/null
@@ -1,64 +0,0 @@
-#ifndef __ASM_SH64_PLATFORM_H
-#define __ASM_SH64_PLATFORM_H
-
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/platform.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
- * benedict.gaster@superh.com:	 3rd May 2002
- *    Added support for ramdisk, removing statically linked romfs at the same time.
- */
-
-#include <linux/ioport.h>
-#include <asm/irq.h>
-
-
-/*
- * Platform definition structure.
- */
-struct sh64_platform {
-	unsigned int readonly_rootfs;
-	unsigned int ramdisk_flags;
-	unsigned int initial_root_dev;
-	unsigned int loader_type;
-	unsigned int initrd_start;
-	unsigned int initrd_size;
-	unsigned int fpu_flags;
-	unsigned int io_res_count;
-	unsigned int kram_res_count;
-	unsigned int xram_res_count;
-	unsigned int rom_res_count;
-	struct resource *io_res_p;
-	struct resource *kram_res_p;
-	struct resource *xram_res_p;
-	struct resource *rom_res_p;
-};
-
-extern struct sh64_platform platform_parms;
-
-extern unsigned long long memory_start, memory_end;
-
-extern unsigned long long fpu_in_use;
-
-extern int platform_int_priority[NR_INTC_IRQS];
-
-#define FPU_FLAGS		(platform_parms.fpu_flags)
-#define STANDARD_IO_RESOURCES	(platform_parms.io_res_count)
-#define STANDARD_KRAM_RESOURCES	(platform_parms.kram_res_count)
-#define STANDARD_XRAM_RESOURCES	(platform_parms.xram_res_count)
-#define STANDARD_ROM_RESOURCES	(platform_parms.rom_res_count)
-
-/*
- * Kernel Memory description, Respectively:
- * code = last but one memory descriptor
- * data = last memory descriptor
- */
-#define code_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 2])
-#define data_resource (platform_parms.kram_res_p[STANDARD_KRAM_RESOURCES - 1])
-
-#endif	/* __ASM_SH64_PLATFORM_H */
-- 
cgit v1.2.3


From 29e0d209b38479e6817318a971082421a7d630b7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 22 Nov 2007 13:11:39 +0900
Subject: sh: Add SH-5 subtypes to check_bugs() for utsname.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/bugs.h | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index b66139ff73fc..e2d79567e2a9 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -48,9 +48,16 @@ static void __init check_bugs(void)
 		*p++ = 's';
 		*p++ = 'p';
 		break;
-	default:
-		*p++ = '?';
-		*p++ = '!';
+	case CPU_SH5_101 ... CPU_SH5_103:
+		*p++ = '6';
+		*p++ = '4';
+		break;
+	case CPU_SH_NONE:
+		/*
+		 * Specifically use CPU_SH_NONE rather than default:,
+		 * so we're able to have the compiler whine about
+		 * unhandled enumerations.
+		 */
 		break;
 	}
 
-- 
cgit v1.2.3


From b6d7b666097e79a8908e3c43fd55fd291a95e133 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 22 Nov 2007 16:29:10 +0900
Subject: sh: Get the SH-5 PCI support building.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  |   1 +
 arch/sh/drivers/pci/Makefile     |   2 +
 arch/sh/drivers/pci/ops-cayman.c |  94 ++++++++
 arch/sh/drivers/pci/ops-sh5.c    |  93 ++++++++
 arch/sh/drivers/pci/pci-sh5.c    | 452 +++++++--------------------------------
 arch/sh/drivers/pci/pci-sh5.h    |  10 +-
 arch/sh/drivers/pci/pci.c        |   2 +-
 include/asm-sh/pci.h             |   5 +-
 8 files changed, 275 insertions(+), 384 deletions(-)
 create mode 100644 arch/sh/drivers/pci/ops-cayman.c
 create mode 100644 arch/sh/drivers/pci/ops-sh5.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index c0797b4aadbb..1ad3ce540021 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -512,6 +512,7 @@ config SH_SIMULATOR
 config SH_CAYMAN
 	bool "Hitachi Cayman"
 	depends on CPU_SUBTYPE_SH5_101 || CPU_SUBTYPE_SH5_103
+	select SYS_SUPPORTS_PCI
 
 config SH_HARP
 	bool "ST50 Harp"
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index fba6b5ba0b3a..172e81db9530 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7751R)	+= pci-sh7751.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7785)	+= pci-sh7780.o ops-sh4.o
+obj-$(CONFIG_CPU_SH5)			+= pci-sh5.o ops-sh5.o
 
 obj-$(CONFIG_SH_DREAMCAST)		+= ops-dreamcast.o fixups-dreamcast.o \
 					   dma-dreamcast.o
@@ -20,3 +21,4 @@ obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
 obj-$(CONFIG_SH_LANDISK)		+= ops-landisk.o
 obj-$(CONFIG_SH_LBOX_RE2)		+= ops-lboxre2.o fixups-lboxre2.o
 obj-$(CONFIG_SH_7780_SOLUTION_ENGINE)	+= ops-se7780.o fixups-se7780.o
+obj-$(CONFIG_SH_CAYMAN)			+= ops-cayman.o
diff --git a/arch/sh/drivers/pci/ops-cayman.c b/arch/sh/drivers/pci/ops-cayman.c
new file mode 100644
index 000000000000..980275ffa30b
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-cayman.c
@@ -0,0 +1,94 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <asm/cpu/irq.h>
+#include "pci-sh5.h"
+
+static inline u8 bridge_swizzle(u8 pin, u8 slot)
+{
+	return (((pin - 1) + slot) % 4) + 1;
+}
+
+int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int result = -1;
+
+	/* The complication here is that the PCI IRQ lines from the Cayman's 2
+	   5V slots get into the CPU via a different path from the IRQ lines
+	   from the 3 3.3V slots.  Thus, we have to detect whether the card's
+	   interrupts go via the 5V or 3.3V path, i.e. the 'bridge swizzling'
+	   at the point where we cross from 5V to 3.3V is not the normal case.
+
+	   The added complication is that we don't know that the 5V slots are
+	   always bus 2, because a card containing a PCI-PCI bridge may be
+	   plugged into a 3.3V slot, and this changes the bus numbering.
+
+	   Also, the Cayman has an intermediate PCI bus that goes a custom
+	   expansion board header (and to the secondary bridge).  This bus has
+	   never been used in practice.
+
+	   The 1ary onboard PCI-PCI bridge is device 3 on bus 0
+	   The 2ary onboard PCI-PCI bridge is device 0 on the 2ary bus of
+	   the 1ary bridge.
+	   */
+
+	struct slot_pin {
+		int slot;
+		int pin;
+	} path[4];
+	int i=0;
+
+	while (dev->bus->number > 0) {
+
+		slot = path[i].slot = PCI_SLOT(dev->devfn);
+		pin = path[i].pin = bridge_swizzle(pin, slot);
+		dev = dev->bus->self;
+		i++;
+		if (i > 3) panic("PCI path to root bus too long!\n");
+	}
+
+	slot = PCI_SLOT(dev->devfn);
+	/* This is the slot on bus 0 through which the device is eventually
+	   reachable. */
+
+	/* Now work back up. */
+	if ((slot < 3) || (i == 0)) {
+		/* Bus 0 (incl. PCI-PCI bridge itself) : perform the final
+		   swizzle now. */
+		result = IRQ_INTA + bridge_swizzle(pin, slot) - 1;
+	} else {
+		i--;
+		slot = path[i].slot;
+		pin  = path[i].pin;
+		if (slot > 0) {
+			panic("PCI expansion bus device found - not handled!\n");
+		} else {
+			if (i > 0) {
+				/* 5V slots */
+				i--;
+				slot = path[i].slot;
+				pin  = path[i].pin;
+				/* 'pin' was swizzled earlier wrt slot, don't do it again. */
+				result = IRQ_P2INTA + (pin - 1);
+			} else {
+				/* IRQ for 2ary PCI-PCI bridge : unused */
+				result = -1;
+			}
+		}
+	}
+
+	return result;
+}
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh5_pci_ops, NULL, NULL, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+int __init pcibios_init_platform(void)
+{
+	return sh5pci_init(__pa(memory_start),
+			   __pa(memory_end) - __pa(memory_start));
+}
diff --git a/arch/sh/drivers/pci/ops-sh5.c b/arch/sh/drivers/pci/ops-sh5.c
new file mode 100644
index 000000000000..729e38a6fe07
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-sh5.c
@@ -0,0 +1,93 @@
+/*
+ * Support functions for the SH5 PCI hardware.
+ *
+ * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
+ * Copyright (C) 2003, 2004 Paul Mundt
+ * Copyright (C) 2004 Richard Curnow
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ */
+#include <linux/kernel.h>
+#include <linux/rwsem.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/types.h>
+#include <linux/irq.h>
+#include <asm/pci.h>
+#include <asm/io.h>
+#include "pci-sh5.h"
+
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	printk("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i=0; i<4; i++) {
+		struct resource *r = &d->resource[i];
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+char * __devinit pcibios_setup(char *str)
+{
+	return str;
+}
+
+static int sh5pci_read(struct pci_bus *bus, unsigned int devfn, int where,
+			int size, u32 *val)
+{
+	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
+
+	switch (size) {
+		case 1:
+			*val = (u8)SH5PCI_READ_BYTE(PDR + (where & 3));
+			break;
+		case 2:
+			*val = (u16)SH5PCI_READ_SHORT(PDR + (where & 2));
+			break;
+		case 4:
+			*val = SH5PCI_READ(PDR);
+			break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int sh5pci_write(struct pci_bus *bus, unsigned int devfn, int where,
+			 int size, u32 val)
+{
+	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
+
+	switch (size) {
+		case 1:
+			SH5PCI_WRITE_BYTE(PDR + (where & 3), (u8)val);
+			break;
+		case 2:
+			SH5PCI_WRITE_SHORT(PDR + (where & 2), (u16)val);
+			break;
+		case 4:
+			SH5PCI_WRITE(PDR, val);
+			break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops sh5_pci_ops = {
+	.read		= sh5pci_read,
+	.write		= sh5pci_write,
+};
diff --git a/arch/sh/drivers/pci/pci-sh5.c b/arch/sh/drivers/pci/pci-sh5.c
index b4d9534d2b0e..a00a4df8c02d 100644
--- a/arch/sh/drivers/pci/pci-sh5.c
+++ b/arch/sh/drivers/pci/pci-sh5.c
@@ -18,40 +18,14 @@
 #include <linux/pci.h>
 #include <linux/delay.h>
 #include <linux/types.h>
-#include <asm/pci.h>
 #include <linux/irq.h>
-
+#include <asm/cpu/irq.h>
+#include <asm/pci.h>
 #include <asm/io.h>
-#include <asm/hardware.h>
-#include "pci_sh5.h"
-
-static unsigned long pcicr_virt;
-unsigned long pciio_virt;
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	printk("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+#include "pci-sh5.h"
 
-char * __devinit pcibios_setup(char *str)
-{
-	return str;
-}
+unsigned long pcicr_virt;
+unsigned long PCI_IO_AREA;
 
 /* Rounds a number UP to the nearest power of two. Used for
  * sizing the PCI window.
@@ -79,31 +53,73 @@ static u32 __init r2p2(u32 num)
 	return tmp;
 }
 
-extern unsigned long long memory_start, memory_end;
+static irqreturn_t pcish5_err_irq(int irq, void *dev_id)
+{
+	struct pt_regs *regs = get_irq_regs();
+	unsigned pci_int, pci_air, pci_cir, pci_aint;
+
+	pci_int = SH5PCI_READ(INT);
+	pci_cir = SH5PCI_READ(CIR);
+	pci_air = SH5PCI_READ(AIR);
 
-int __init sh5pci_init(unsigned memStart, unsigned memSize)
+	if (pci_int) {
+		printk("PCI INTERRUPT (at %08llx)!\n", regs->pc);
+		printk("PCI INT -> 0x%x\n", pci_int & 0xffff);
+		printk("PCI AIR -> 0x%x\n", pci_air);
+		printk("PCI CIR -> 0x%x\n", pci_cir);
+		SH5PCI_WRITE(INT, ~0);
+	}
+
+	pci_aint = SH5PCI_READ(AINT);
+	if (pci_aint) {
+		printk("PCI ARB INTERRUPT!\n");
+		printk("PCI AINT -> 0x%x\n", pci_aint);
+		printk("PCI AIR -> 0x%x\n", pci_air);
+		printk("PCI CIR -> 0x%x\n", pci_cir);
+		SH5PCI_WRITE(AINT, ~0);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t pcish5_serr_irq(int irq, void *dev_id)
+{
+	printk("SERR IRQ\n");
+
+	return IRQ_NONE;
+}
+
+int __init sh5pci_init(unsigned long memStart, unsigned long memSize)
 {
 	u32 lsr0;
 	u32 uval;
 
+        if (request_irq(IRQ_ERR, pcish5_err_irq,
+                        IRQF_DISABLED, "PCI Error",NULL) < 0) {
+                printk(KERN_ERR "PCISH5: Cannot hook PCI_PERR interrupt\n");
+                return -EINVAL;
+        }
+
+        if (request_irq(IRQ_SERR, pcish5_serr_irq,
+                        IRQF_DISABLED, "PCI SERR interrupt", NULL) < 0) {
+                printk(KERN_ERR "PCISH5: Cannot hook PCI_SERR interrupt\n");
+                return -EINVAL;
+        }
+
 	pcicr_virt = onchip_remap(SH5PCI_ICR_BASE, 1024, "PCICR");
 	if (!pcicr_virt) {
 		panic("Unable to remap PCICR\n");
 	}
 
-	pciio_virt = onchip_remap(SH5PCI_IO_BASE, 0x10000, "PCIIO");
-	if (!pciio_virt) {
+	PCI_IO_AREA = onchip_remap(SH5PCI_IO_BASE, 0x10000, "PCIIO");
+	if (!PCI_IO_AREA) {
 		panic("Unable to remap PCIIO\n");
 	}
 
-	pr_debug("Register base addres is 0x%08lx\n", pcicr_virt);
-
 	/* Clear snoop registers */
         SH5PCI_WRITE(CSCR0, 0);
         SH5PCI_WRITE(CSCR1, 0);
 
-	pr_debug("Wrote to reg\n");
-
         /* Switch off interrupts */
         SH5PCI_WRITE(INTM,  0);
         SH5PCI_WRITE(AINTM, 0);
@@ -113,16 +129,17 @@ int __init sh5pci_init(unsigned memStart, unsigned memSize)
         uval = SH5PCI_READ(CR);
 
 	/* Set command Register */
-        SH5PCI_WRITE(CR, uval | CR_LOCK_MASK | CR_CFINT| CR_FTO | CR_PFE | CR_PFCS | CR_BMAM);
+        SH5PCI_WRITE(CR, uval | CR_LOCK_MASK | CR_CFINT| CR_FTO | CR_PFE |
+		     CR_PFCS | CR_BMAM);
 
 	uval=SH5PCI_READ(CR);
-        pr_debug("CR is actually 0x%08x\n",uval);
 
         /* Allow it to be a master */
 	/* NB - WE DISABLE I/O ACCESS to stop overlap */
         /* set WAIT bit to enable stepping, an attempt to improve stability */
 	SH5PCI_WRITE_SHORT(CSR_CMD,
-			    PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER | PCI_COMMAND_WAIT);
+			    PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
+			    PCI_COMMAND_WAIT);
 
         /*
         ** Set translation mapping memory in order to convert the address
@@ -143,32 +160,30 @@ int __init sh5pci_init(unsigned memStart, unsigned memSize)
 
         SH5PCI_WRITE(IOBR,0x0);
 
-	pr_debug("PCI:Writing 0x%08x to IOBR\n",0);
-
         /* Set up a 256K window. Totally pointless waste  of address space */
         SH5PCI_WRITE(IOBMR,0);
-	pr_debug("PCI:Writing 0x%08x to IOBMR\n",0);
 
-	/* The SH5 has a HUGE 256K I/O region, which breaks the PCI spec. Ideally,
-         * we would want to map the I/O region somewhere, but it is so big this is not
-         * that easy!
+	/* The SH5 has a HUGE 256K I/O region, which breaks the PCI spec.
+	 * Ideally, we would want to map the I/O region somewhere, but it
+	 * is so big this is not that easy!
          */
 	SH5PCI_WRITE(CSR_IBAR0,~0);
 	/* Set memory size value */
         memSize = memory_end - memory_start;
 
-        /* Now we set up the mbars so the PCI bus can see the memory of the machine */
-        if (memSize < (1024 * 1024)) {
-                printk(KERN_ERR "PCISH5: Ridiculous memory size of 0x%x?\n", memSize);
+	/* Now we set up the mbars so the PCI bus can see the memory of
+	 * the machine */
+	if (memSize < (1024 * 1024)) {
+                printk(KERN_ERR "PCISH5: Ridiculous memory size of 0x%lx?\n",
+		       memSize);
                 return -EINVAL;
         }
 
         /* Set LSR 0 */
-        lsr0 = (memSize > (512 * 1024 * 1024)) ? 0x1ff00001 : ((r2p2(memSize) - 0x100000) | 0x1);
+        lsr0 = (memSize > (512 * 1024 * 1024)) ? 0x1ff00001 :
+		((r2p2(memSize) - 0x100000) | 0x1);
         SH5PCI_WRITE(LSR0, lsr0);
 
-	pr_debug("PCI:Writing 0x%08x to LSR0\n",lsr0);
-
         /* Set MBAR 0 */
         SH5PCI_WRITE(CSR_MBAR0, memory_start);
         SH5PCI_WRITE(LAR0, memory_start);
@@ -177,334 +192,21 @@ int __init sh5pci_init(unsigned memStart, unsigned memSize)
         SH5PCI_WRITE(LAR1,0);
         SH5PCI_WRITE(LSR1,0);
 
-	pr_debug("PCI:Writing 0x%08llx to CSR_MBAR0\n",memory_start);
-	pr_debug("PCI:Writing 0x%08llx to LAR0\n",memory_start);
-
         /* Enable the PCI interrupts on the device */
         SH5PCI_WRITE(INTM,  ~0);
         SH5PCI_WRITE(AINTM, ~0);
         SH5PCI_WRITE(PINTM, ~0);
 
-	pr_debug("Switching on all error interrupts\n");
-
-        return(0);
-}
-
-static int sh5pci_read(struct pci_bus *bus, unsigned int devfn, int where,
-			int size, u32 *val)
-{
-	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
-
-	switch (size) {
-		case 1:
-			*val = (u8)SH5PCI_READ_BYTE(PDR + (where & 3));
-			break;
-		case 2:
-			*val = (u16)SH5PCI_READ_SHORT(PDR + (where & 2));
-			break;
-		case 4:
-			*val = SH5PCI_READ(PDR);
-			break;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int sh5pci_write(struct pci_bus *bus, unsigned int devfn, int where,
-			 int size, u32 val)
-{
-	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
-
-	switch (size) {
-		case 1:
-			SH5PCI_WRITE_BYTE(PDR + (where & 3), (u8)val);
-			break;
-		case 2:
-			SH5PCI_WRITE_SHORT(PDR + (where & 2), (u16)val);
-			break;
-		case 4:
-			SH5PCI_WRITE(PDR, val);
-			break;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops pci_config_ops = {
-	.read =		sh5pci_read,
-	.write =	sh5pci_write,
-};
-
-/* Everything hangs off this */
-static struct pci_bus *pci_root_bus;
-
-
-static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
-{
-	pr_debug("swizzle for dev %d on bus %d slot %d pin is %d\n",
-	         dev->devfn,dev->bus->number, PCI_SLOT(dev->devfn),*pin);
-	return PCI_SLOT(dev->devfn);
-}
-
-static inline u8 bridge_swizzle(u8 pin, u8 slot)
-{
-	return (((pin-1) + slot) % 4) + 1;
-}
-
-u8 __init common_swizzle(struct pci_dev *dev, u8 *pinp)
-{
-	if (dev->bus->number != 0) {
-		u8 pin = *pinp;
-		do {
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
-			/* Move up the chain of bridges. */
-			dev = dev->bus->self;
-		} while (dev->bus->self);
-		*pinp = pin;
-
-		/* The slot is the slot of the last bridge. */
-	}
-
-	return PCI_SLOT(dev->devfn);
-}
-
-/* This needs to be shunted out of here into the board specific bit */
-
-static int __init map_cayman_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int result = -1;
-
-	/* The complication here is that the PCI IRQ lines from the Cayman's 2
-	   5V slots get into the CPU via a different path from the IRQ lines
-	   from the 3 3.3V slots.  Thus, we have to detect whether the card's
-	   interrupts go via the 5V or 3.3V path, i.e. the 'bridge swizzling'
-	   at the point where we cross from 5V to 3.3V is not the normal case.
-
-	   The added complication is that we don't know that the 5V slots are
-	   always bus 2, because a card containing a PCI-PCI bridge may be
-	   plugged into a 3.3V slot, and this changes the bus numbering.
-
-	   Also, the Cayman has an intermediate PCI bus that goes a custom
-	   expansion board header (and to the secondary bridge).  This bus has
-	   never been used in practice.
-
-	   The 1ary onboard PCI-PCI bridge is device 3 on bus 0
-	   The 2ary onboard PCI-PCI bridge is device 0 on the 2ary bus of the 1ary bridge.
-	   */
-
-	struct slot_pin {
-		int slot;
-		int pin;
-	} path[4];
-	int i=0;
-
-	while (dev->bus->number > 0) {
-
-		slot = path[i].slot = PCI_SLOT(dev->devfn);
-		pin = path[i].pin = bridge_swizzle(pin, slot);
-		dev = dev->bus->self;
-		i++;
-		if (i > 3) panic("PCI path to root bus too long!\n");
-	}
-
-	slot = PCI_SLOT(dev->devfn);
-	/* This is the slot on bus 0 through which the device is eventually
-	   reachable. */
-
-	/* Now work back up. */
-	if ((slot < 3) || (i == 0)) {
-		/* Bus 0 (incl. PCI-PCI bridge itself) : perform the final
-		   swizzle now. */
-		result = IRQ_INTA + bridge_swizzle(pin, slot) - 1;
-	} else {
-		i--;
-		slot = path[i].slot;
-		pin  = path[i].pin;
-		if (slot > 0) {
-			panic("PCI expansion bus device found - not handled!\n");
-		} else {
-			if (i > 0) {
-				/* 5V slots */
-				i--;
-				slot = path[i].slot;
-				pin  = path[i].pin;
-				/* 'pin' was swizzled earlier wrt slot, don't do it again. */
-				result = IRQ_P2INTA + (pin - 1);
-			} else {
-				/* IRQ for 2ary PCI-PCI bridge : unused */
-				result = -1;
-			}
-		}
-	}
-
-	return result;
-}
-
-static irqreturn_t pcish5_err_irq(int irq, void *dev_id)
-{
-	struct pt_regs *regs = get_irq_regs();
-	unsigned pci_int, pci_air, pci_cir, pci_aint;
-
-	pci_int = SH5PCI_READ(INT);
-	pci_cir = SH5PCI_READ(CIR);
-	pci_air = SH5PCI_READ(AIR);
-
-	if (pci_int) {
-		printk("PCI INTERRUPT (at %08llx)!\n", regs->pc);
-		printk("PCI INT -> 0x%x\n", pci_int & 0xffff);
-		printk("PCI AIR -> 0x%x\n", pci_air);
-		printk("PCI CIR -> 0x%x\n", pci_cir);
-		SH5PCI_WRITE(INT, ~0);
-	}
-
-	pci_aint = SH5PCI_READ(AINT);
-	if (pci_aint) {
-		printk("PCI ARB INTERRUPT!\n");
-		printk("PCI AINT -> 0x%x\n", pci_aint);
-		printk("PCI AIR -> 0x%x\n", pci_air);
-		printk("PCI CIR -> 0x%x\n", pci_cir);
-		SH5PCI_WRITE(AINT, ~0);
-	}
-
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t pcish5_serr_irq(int irq, void *dev_id)
-{
-	printk("SERR IRQ\n");
-
-	return IRQ_NONE;
-}
-
-static void __init
-pcibios_size_bridge(struct pci_bus *bus, struct resource *ior,
-		    struct resource *memr)
-{
-	struct resource io_res, mem_res;
-	struct pci_dev *dev;
-	struct pci_dev *bridge = bus->self;
-	struct list_head *ln;
-
-	if (!bridge)
-		return;	/* host bridge, nothing to do */
-
-	/* set reasonable default locations for pcibios_align_resource */
-	io_res.start = PCIBIOS_MIN_IO;
-	mem_res.start = PCIBIOS_MIN_MEM;
-
-	io_res.end = io_res.start;
-	mem_res.end = mem_res.start;
-
-	/* Collect information about how our direct children are layed out. */
-	for (ln=bus->devices.next; ln != &bus->devices; ln=ln->next) {
-		int i;
-		dev = pci_dev_b(ln);
-
-		/* Skip bridges for now */
-		if (dev->class >> 8 == PCI_CLASS_BRIDGE_PCI)
-			continue;
-
-		for (i = 0; i < PCI_NUM_RESOURCES; i++) {
-			struct resource res;
-			unsigned long size;
-
-			memcpy(&res, &dev->resource[i], sizeof(res));
-			size = res.end - res.start + 1;
-
-			if (res.flags & IORESOURCE_IO) {
-				res.start = io_res.end;
-				pcibios_align_resource(dev, &res, size, 0);
-				io_res.end = res.start + size;
-			} else if (res.flags & IORESOURCE_MEM) {
-				res.start = mem_res.end;
-				pcibios_align_resource(dev, &res, size, 0);
-				mem_res.end = res.start + size;
-			}
-		}
-	}
-
-	/* And for all of the subordinate busses. */
-	for (ln=bus->children.next; ln != &bus->children; ln=ln->next)
-		pcibios_size_bridge(pci_bus_b(ln), &io_res, &mem_res);
-
-	/* turn the ending locations into sizes (subtract start) */
-	io_res.end -= io_res.start;
-	mem_res.end -= mem_res.start;
-
-	/* Align the sizes up by bridge rules */
-	io_res.end = ALIGN(io_res.end, 4*1024) - 1;
-	mem_res.end = ALIGN(mem_res.end, 1*1024*1024) - 1;
-
-	/* Adjust the bridge's allocation requirements */
-	bridge->resource[0].end = bridge->resource[0].start + io_res.end;
-	bridge->resource[1].end = bridge->resource[1].start + mem_res.end;
-
-	bridge->resource[PCI_BRIDGE_RESOURCES].end =
-	    bridge->resource[PCI_BRIDGE_RESOURCES].start + io_res.end;
-	bridge->resource[PCI_BRIDGE_RESOURCES+1].end =
-	    bridge->resource[PCI_BRIDGE_RESOURCES+1].start + mem_res.end;
-
-	/* adjust parent's resource requirements */
-	if (ior) {
-		ior->end = ALIGN(ior->end, 4*1024);
-		ior->end += io_res.end;
-	}
-
-	if (memr) {
-		memr->end = ALIGN(memr->end, 1*1024*1024);
-		memr->end += mem_res.end;
-	}
-}
-
-static void __init pcibios_size_bridges(void)
-{
-	struct resource io_res, mem_res;
-
-	memset(&io_res, 0, sizeof(io_res));
-	memset(&mem_res, 0, sizeof(mem_res));
-
-	pcibios_size_bridge(pci_root_bus, &io_res, &mem_res);
-}
-
-static int __init pcibios_init(void)
-{
-        if (request_irq(IRQ_ERR, pcish5_err_irq,
-                        IRQF_DISABLED, "PCI Error",NULL) < 0) {
-                printk(KERN_ERR "PCISH5: Cannot hook PCI_PERR interrupt\n");
-                return -EINVAL;
-        }
-
-        if (request_irq(IRQ_SERR, pcish5_serr_irq,
-                        IRQF_DISABLED, "PCI SERR interrupt", NULL) < 0) {
-                printk(KERN_ERR "PCISH5: Cannot hook PCI_SERR interrupt\n");
-                return -EINVAL;
-        }
-
-	/* The pci subsystem needs to know where memory is and how much
-	 * of it there is. I've simply made these globals. A better mechanism
-	 * is probably needed.
-	 */
-	sh5pci_init(__pa(memory_start),
-		     __pa(memory_end) - __pa(memory_start));
-
-	pci_root_bus = pci_scan_bus(0, &pci_config_ops, NULL);
-	pcibios_size_bridges();
-	pci_assign_unassigned_resources();
-	pci_fixup_irqs(no_swizzle, map_cayman_irq);
-
 	return 0;
 }
 
-subsys_initcall(pcibios_init);
-
 void __devinit pcibios_fixup_bus(struct pci_bus *bus)
 {
 	struct pci_dev *dev = bus->self;
 	int i;
 
-#if 1
-	if(dev) {
-		for(i=0; i<3; i++) {
+	if (dev) {
+		for (i= 0; i < 3; i++) {
 			bus->resource[i] =
 				&dev->resource[PCI_BRIDGE_RESOURCES+i];
 			bus->resource[i]->name = bus->name;
@@ -514,23 +216,13 @@ void __devinit pcibios_fixup_bus(struct pci_bus *bus)
 
 		/* For now, propagate host limits to the bus;
 		 * we'll adjust them later. */
-
-#if 1
 		bus->resource[0]->end = 64*1024 - 1 ;
 		bus->resource[1]->end = PCIBIOS_MIN_MEM+(256*1024*1024)-1;
 		bus->resource[0]->start = PCIBIOS_MIN_IO;
 		bus->resource[1]->start = PCIBIOS_MIN_MEM;
-#else
-		bus->resource[0]->end = 0;
-		bus->resource[1]->end = 0;
-		bus->resource[0]->start =0;
-		bus->resource[1]->start = 0;
-#endif
+
 		/* Turn off downstream PF memory address range by default */
 		bus->resource[2]->start = 1024*1024;
 		bus->resource[2]->end = bus->resource[2]->start - 1;
 	}
-#endif
-
 }
-
diff --git a/arch/sh/drivers/pci/pci-sh5.h b/arch/sh/drivers/pci/pci-sh5.h
index c71159dd04b9..7cff3fc04d30 100644
--- a/arch/sh/drivers/pci/pci-sh5.h
+++ b/arch/sh/drivers/pci/pci-sh5.h
@@ -6,6 +6,8 @@
  *
  * Definitions for the SH5 PCI hardware.
  */
+#ifndef __PCI_SH5_H
+#define __PCI_SH5_H
 
 /* Product ID */
 #define PCISH5_PID		0x350d
@@ -73,13 +75,12 @@
 #define PCISH5_ICR_CSR_MBAR0   0x014   /* First  Memory base address register */
 #define PCISH5_ICR_CSR_MBAR1   0x018   /* Second Memory base address register */
 
-
-
 /* Base address of registers */
 #define SH5PCI_ICR_BASE (PHYS_PCI_BLOCK + 0x00040000)
 #define SH5PCI_IO_BASE  (PHYS_PCI_BLOCK + 0x00800000)
 /* #define SH5PCI_VCR_BASE (P2SEG_PCICB_BLOCK + P2SEG)    */
 
+extern unsigned long pcicr_virt;
 /* Register selection macro */
 #define PCISH5_ICR_REG(x)                ( pcicr_virt + (PCISH5_ICR_##x))
 /* #define PCISH5_VCR_REG(x)                ( SH5PCI_VCR_BASE (PCISH5_VCR_##x)) */
@@ -104,4 +105,9 @@
 #define PCISH5_MEM_SIZCONV(x)		  (((x / 0x40000) - 1) << 18)
 #define PCISH5_IO_SIZCONV(x)		  (((x / 0x40000) - 1) << 18)
 
+extern struct pci_ops sh5_pci_ops;
+
+/* arch/sh/drivers/pci/pci-sh5.c */
+int sh5pci_init(unsigned long memStart, unsigned long memSize);
 
+#endif /* __PCI_SH5_H */
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index ccaba368ac9b..49b435c3a57a 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -71,7 +71,7 @@ subsys_initcall(pcibios_init);
  *  Called after each bus is probed, but before its children
  *  are examined.
  */
-void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+void __devinit __weak pcibios_fixup_bus(struct pci_bus *bus)
 {
 	pci_read_bridge_bases(bus);
 }
diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 2757ce096ff7..df1d383e18a5 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -38,9 +38,12 @@ extern struct pci_channel board_pci_channels[];
 #if defined(CONFIG_CPU_SUBTYPE_SH7780) || defined(CONFIG_CPU_SUBTYPE_SH7785)
 #define PCI_IO_AREA		0xFE400000
 #define PCI_IO_SIZE		0x00400000
+#elif defined(CONFIG_CPU_SH5)
+extern unsigned long PCI_IO_AREA;
+#define PCI_IO_SIZE		0x00010000
 #else
 #define PCI_IO_AREA		0xFE240000
-#define PCI_IO_SIZE		0X00040000
+#define PCI_IO_SIZE		0x00040000
 #endif
 
 #define PCI_MEM_SIZE		0x01000000
-- 
cgit v1.2.3


From 332fd57b92d26e2ac6112340b98e92bb76117a41 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 22 Nov 2007 17:30:50 +0900
Subject: sh: Bring the SH-5 FPU in line with the SH-4 FPU API.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh5/fpu.c  | 16 +++++++---------
 arch/sh/kernel/process_64.c   |  4 ++--
 arch/sh/kernel/ptrace_64.c    |  4 ++--
 arch/sh/kernel/signal_64.c    |  2 +-
 arch/sh/kernel/traps_64.c     |  4 ++--
 include/asm-sh/fpu.h          | 44 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/processor.h    |  2 +-
 include/asm-sh/processor_32.h | 26 -------------------------
 include/asm-sh/processor_64.h | 16 ----------------
 9 files changed, 59 insertions(+), 59 deletions(-)
 create mode 100644 include/asm-sh/fpu.h

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh5/fpu.c b/arch/sh/kernel/cpu/sh5/fpu.c
index d3f5e7468dfe..5a391b1a00e2 100644
--- a/arch/sh/kernel/cpu/sh5/fpu.c
+++ b/arch/sh/kernel/cpu/sh5/fpu.c
@@ -30,12 +30,12 @@
 
 static union sh_fpu_union init_fpuregs = {
 	.hard = {
-	  .fp_regs = { [0 ... 63] = sNAN32 },
-	  .fpscr = FPSCR_INIT
+		.fp_regs = { [0 ... 63] = sNAN32 },
+		.fpscr = FPSCR_INIT
 	}
 };
 
-inline void fpsave(struct sh_fpu_hard_struct *fpregs)
+void save_fpu(struct task_struct *tsk, struct pt_regs *regs)
 {
 	asm volatile("fst.p     %0, (0*8), fp0\n\t"
 		     "fst.p     %0, (1*8), fp2\n\t"
@@ -73,11 +73,10 @@ inline void fpsave(struct sh_fpu_hard_struct *fpregs)
 		     "fgetscr   fr63\n\t"
 		     "fst.s     %0, (32*8), fr63\n\t"
 		: /* no output */
-		: "r" (fpregs)
+		: "r" (&tsk->thread.fpu.hard)
 		: "memory");
 }
 
-
 static inline void
 fpload(struct sh_fpu_hard_struct *fpregs)
 {
@@ -153,10 +152,10 @@ do_fpu_state_restore(unsigned long ex, struct pt_regs *regs)
 		return;
 
 	enable_fpu();
-	if (last_task_used_math != NULL) {
+	if (last_task_used_math != NULL)
 		/* Other processes fpu state, save away */
-		fpsave(&last_task_used_math->thread.fpu.hard);
-        }
+		save_fpu(last_task_used_math, regs);
+
         last_task_used_math = current;
         if (used_math()) {
                 fpload(&current->thread.fpu.hard);
@@ -167,4 +166,3 @@ do_fpu_state_restore(unsigned long ex, struct pt_regs *regs)
         }
 	disable_fpu();
 }
-
diff --git a/arch/sh/kernel/process_64.c b/arch/sh/kernel/process_64.c
index 2dd97eecb44c..973dd1a3d292 100644
--- a/arch/sh/kernel/process_64.c
+++ b/arch/sh/kernel/process_64.c
@@ -480,7 +480,7 @@ int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu)
 	if (fpvalid) {
 		if (current == last_task_used_math) {
 			enable_fpu();
-			fpsave(&tsk->thread.fpu.hard);
+			save_fpu(tsk, regs);
 			disable_fpu();
 			last_task_used_math = 0;
 			regs->sr |= SR_FD;
@@ -507,7 +507,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 #ifdef CONFIG_SH_FPU
 	if(last_task_used_math == current) {
 		enable_fpu();
-		fpsave(&current->thread.fpu.hard);
+		save_fpu(current, regs);
 		disable_fpu();
 		last_task_used_math = NULL;
 		regs->sr |= SR_FD;
diff --git a/arch/sh/kernel/ptrace_64.c b/arch/sh/kernel/ptrace_64.c
index 51bb546365cb..e9cc6ebd2cea 100644
--- a/arch/sh/kernel/ptrace_64.c
+++ b/arch/sh/kernel/ptrace_64.c
@@ -75,7 +75,7 @@ get_fpu_long(struct task_struct *task, unsigned long addr)
 
 	if (last_task_used_math == task) {
 		enable_fpu();
-		fpsave(&task->thread.fpu.hard);
+		save_fpu(task, regs);
 		disable_fpu();
 		last_task_used_math = 0;
 		regs->sr |= SR_FD;
@@ -111,7 +111,7 @@ put_fpu_long(struct task_struct *task, unsigned long addr, unsigned long data)
 		set_stopped_child_used_math(task);
 	} else if (last_task_used_math == task) {
 		enable_fpu();
-		fpsave(&task->thread.fpu.hard);
+		save_fpu(task, regs);
 		disable_fpu();
 		last_task_used_math = 0;
 		regs->sr |= SR_FD;
diff --git a/arch/sh/kernel/signal_64.c b/arch/sh/kernel/signal_64.c
index 08f403e23662..922891960c33 100644
--- a/arch/sh/kernel/signal_64.c
+++ b/arch/sh/kernel/signal_64.c
@@ -212,7 +212,7 @@ setup_sigcontext_fpu(struct pt_regs *regs, struct sigcontext __user *sc)
 
 	if (current == last_task_used_math) {
 		enable_fpu();
-		fpsave(&current->thread.fpu.hard);
+		save_fpu(current, regs);
 		disable_fpu();
 		last_task_used_math = NULL;
 		regs->sr |= SR_FD;
diff --git a/arch/sh/kernel/traps_64.c b/arch/sh/kernel/traps_64.c
index 0f4ea3ac3e0b..b8020f26b635 100644
--- a/arch/sh/kernel/traps_64.c
+++ b/arch/sh/kernel/traps_64.c
@@ -618,7 +618,7 @@ static int misaligned_fpu_load(struct pt_regs *regs,
 		   indexed by register number. */
 		if (last_task_used_math == current) {
 			enable_fpu();
-			fpsave(&current->thread.fpu.hard);
+			save_fpu(current, regs);
 			disable_fpu();
 			last_task_used_math = NULL;
 			regs->sr |= SR_FD;
@@ -691,7 +691,7 @@ static int misaligned_fpu_store(struct pt_regs *regs,
 		   indexed by register number. */
 		if (last_task_used_math == current) {
 			enable_fpu();
-			fpsave(&current->thread.fpu.hard);
+			save_fpu(current, regs);
 			disable_fpu();
 			last_task_used_math = NULL;
 			regs->sr |= SR_FD;
diff --git a/include/asm-sh/fpu.h b/include/asm-sh/fpu.h
new file mode 100644
index 000000000000..33db698a6b4d
--- /dev/null
+++ b/include/asm-sh/fpu.h
@@ -0,0 +1,44 @@
+#ifndef __ASM_SH_FPU_H
+#define __ASM_SH_FPU_H
+
+#define SR_FD    0x00008000
+
+#ifndef __ASSEMBLY__
+#include <asm/ptrace.h>
+
+#ifdef CONFIG_SH_FPU
+static inline void release_fpu(struct pt_regs *regs)
+{
+	regs->sr |= SR_FD;
+}
+
+static inline void grab_fpu(struct pt_regs *regs)
+{
+	regs->sr &= ~SR_FD;
+}
+
+struct task_struct;
+
+extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs);
+#else
+#define release_fpu(regs)	do { } while (0)
+#define grab_fpu(regs)		do { } while (0)
+#define save_fpu(tsk, regs)	do { } while (0)
+#endif
+
+#define unlazy_fpu(tsk, regs) do {			\
+	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {	\
+		save_fpu(tsk, regs);			\
+	}						\
+} while (0)
+
+#define clear_fpu(tsk, regs) do {				\
+	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {		\
+		clear_tsk_thread_flag(tsk, TIF_USEDFPU);	\
+		release_fpu(regs);				\
+	}							\
+} while (0)
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_SH_FPU_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index f3bd82e95898..3ae2a1c223f2 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -2,9 +2,9 @@
 #define __ASM_SH_PROCESSOR_H
 
 #include <asm/cpu-features.h>
+#include <asm/fpu.h>
 
 #ifndef __ASSEMBLY__
-
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
  *
diff --git a/include/asm-sh/processor_32.h b/include/asm-sh/processor_32.h
index 1ad74633c009..a7edaa1a870c 100644
--- a/include/asm-sh/processor_32.h
+++ b/include/asm-sh/processor_32.h
@@ -65,7 +65,6 @@ extern struct sh_cpuinfo cpu_data[];
  * IMASK-bit:
  *     Interrupt level mask
  */
-#define SR_FD		0x00008000
 #define SR_DSP		0x00001000
 #define SR_IMASK	0x000000f0
 
@@ -178,31 +177,6 @@ static __inline__ void enable_fpu(void)
 			     : "r" (~SR_FD));
 }
 
-static __inline__ void release_fpu(struct pt_regs *regs)
-{
-	regs->sr |= SR_FD;
-}
-
-static __inline__ void grab_fpu(struct pt_regs *regs)
-{
-	regs->sr &= ~SR_FD;
-}
-
-extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs);
-
-#define unlazy_fpu(tsk, regs) do {			\
-	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {	\
-		save_fpu(tsk, regs);			\
-	}						\
-} while (0)
-
-#define clear_fpu(tsk, regs) do {				\
-	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {		\
-		clear_tsk_thread_flag(tsk, TIF_USEDFPU);	\
-		release_fpu(regs);				\
-	}							\
-} while (0)
-
 /* Double presision, NANS as NANS, rounding to nearest, no exceptions */
 #define FPSCR_INIT  0x00080000
 
diff --git a/include/asm-sh/processor_64.h b/include/asm-sh/processor_64.h
index f54648229231..99c22b14a85b 100644
--- a/include/asm-sh/processor_64.h
+++ b/include/asm-sh/processor_64.h
@@ -102,8 +102,6 @@ extern struct sh_cpuinfo cpu_data[];
  *     Single step bit
  *
  */
-#define SR_FD    0x00008000
-
 #if defined(CONFIG_SH64_SR_WATCH)
 #define SR_MMU   0x84000000
 #else
@@ -243,16 +241,6 @@ static inline void enable_fpu(void)
 			     : "r" (~SR_FD));
 }
 
-static inline void release_fpu(struct pt_regs *regs)
-{
-	regs->sr |= SR_FD;
-}
-
-static inline void grab_fpu(struct pt_regs *regs)
-{
-	regs->sr &= ~SR_FD;
-}
-
 /* Round to nearest, no exceptions on inexact, overflow, underflow,
    zero-divide, invalid.  Configure option for whether to flush denorms to
    zero, or except if a denorm is encountered.  */
@@ -263,13 +251,9 @@ static inline void grab_fpu(struct pt_regs *regs)
 #endif
 
 #ifdef CONFIG_SH_FPU
-/* Save the current FP regs */
-void fpsave(struct sh_fpu_hard_struct *fpregs);
-
 /* Initialise the FP state of a task */
 void fpinit(struct sh_fpu_hard_struct *fpregs);
 #else
-#define fpsave(fpregs)	do { } while (0)
 #define fpinit(fpregs)	do { } while (0)
 #endif
 
-- 
cgit v1.2.3


From b000659b1c07f91f0c73bf94bb8922fa740c0ef0 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 23 Nov 2007 14:02:20 +0900
Subject: sh: Fix up bug trap handler build for sh32.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 6b02dfd587ea..8b01fc4a56af 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -194,7 +194,8 @@ asmlinkage void name##_trap_handler(unsigned long r4, unsigned long r5,	\
 
 #define TRAP_HANDLER_DECL				\
 	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);	\
-	unsigned int vec = regs->tra;
+	unsigned int vec = regs->tra;			\
+	(void)vec;
 #else
 #define BUILD_TRAP_HANDLER(name)	\
 asmlinkage void name##_trap_handler(unsigned int vec, struct pt_regs *regs)
-- 
cgit v1.2.3


From 1322b9def91ab8e9e673b58a64e13d6effaaa652 Mon Sep 17 00:00:00 2001
From: Yuichi Nakamura <ynakam@hitachisoft.jp>
Date: Sat, 10 Nov 2007 19:21:34 +0900
Subject: sh: syscall audit support.

Support syscall auditing..

Signed-off-by: Yuichi Nakamura <ynakam@hitachisoft.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/entry-common.S |  8 ++++++--
 arch/sh/kernel/ptrace_32.c    | 21 +++++++++++++++++----
 include/asm-sh/thread_info.h  |  2 ++
 init/Kconfig                  |  2 +-
 4 files changed, 26 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/entry-common.S b/arch/sh/kernel/entry-common.S
index e0317ed080c3..397ac71d97f1 100644
--- a/arch/sh/kernel/entry-common.S
+++ b/arch/sh/kernel/entry-common.S
@@ -224,7 +224,7 @@ work_resched:
 syscall_exit_work:
 	! r0: current_thread_info->flags
 	! r8: current_thread_info
-	tst	#_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP, r0
+	tst	#_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP | _TIF_SYSCALL_AUDIT, r0
 	bt/s	work_pending
 	 tst	#_TIF_NEED_RESCHED, r0
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -234,6 +234,8 @@ syscall_exit_work:
 #endif
 	sti
 	! XXX setup arguments...
+	mov	r15, r4
+	mov	#1, r5
 	mov.l	4f, r0			! do_syscall_trace
 	jsr	@r0
 	 nop
@@ -244,6 +246,8 @@ syscall_exit_work:
 syscall_trace_entry:
 	!                     	Yes it is traced.
 	! XXX setup arguments...
+	mov     r15, r4
+	mov     #0, r5
 	mov.l	4f, r11		! Call do_syscall_trace which notifies
 	jsr	@r11	    	! superior (will chomp R[0-7])
 	 nop
@@ -366,7 +370,7 @@ ENTRY(system_call)
 	!
 	get_current_thread_info r8, r10
 	mov.l	@(TI_FLAGS,r8), r8
-	mov	#_TIF_SYSCALL_TRACE, r10
+	mov	#(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT), r10
 	tst	r10, r8
 	bf	syscall_trace_entry
 	!
diff --git a/arch/sh/kernel/ptrace_32.c b/arch/sh/kernel/ptrace_32.c
index ac725f0aeb72..ce0664a58b49 100644
--- a/arch/sh/kernel/ptrace_32.c
+++ b/arch/sh/kernel/ptrace_32.c
@@ -6,7 +6,7 @@
  *	edited by Linus Torvalds
  *
  * SuperH version:   Copyright (C) 1999, 2000  Kaz Kojima & Niibe Yutaka
- *
+ * Audit support: Yuichi Nakamura <ynakam@hitachisoft.jp>
  */
 #include <linux/kernel.h>
 #include <linux/sched.h>
@@ -19,6 +19,7 @@
 #include <linux/security.h>
 #include <linux/signal.h>
 #include <linux/io.h>
+#include <linux/audit.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -248,15 +249,20 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	return ret;
 }
 
-asmlinkage void do_syscall_trace(void)
+asmlinkage void do_syscall_trace(struct pt_regs *regs, int entryexit)
 {
 	struct task_struct *tsk = current;
 
+	if (unlikely(current->audit_context) && entryexit)
+		audit_syscall_exit(AUDITSC_RESULT(regs->regs[0]),
+				   regs->regs[0]);
+
 	if (!test_thread_flag(TIF_SYSCALL_TRACE) &&
 	    !test_thread_flag(TIF_SINGLESTEP))
-		return;
+		goto out;
 	if (!(tsk->ptrace & PT_PTRACED))
-		return;
+		goto out;
+
 	/* the 0x80 provides a way for the tracing parent to distinguish
 	   between a syscall stop and SIGTRAP delivery */
 	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) &&
@@ -271,4 +277,11 @@ asmlinkage void do_syscall_trace(void)
 		send_sig(tsk->exit_code, tsk, 1);
 		tsk->exit_code = 0;
 	}
+
+out:
+	if (unlikely(current->audit_context) && !entryexit)
+		audit_syscall_entry(AUDIT_ARCH_SH, regs->regs[3],
+				    regs->regs[4], regs->regs[5],
+				    regs->regs[6], regs->regs[7]);
+
 }
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index d49ee9d868e5..c6577d3dc46d 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -113,6 +113,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_NEED_RESCHED	2	/* rescheduling necessary */
 #define TIF_RESTORE_SIGMASK	3	/* restore signal mask in do_signal() */
 #define TIF_SINGLESTEP		4	/* singlestepping active */
+#define TIF_SYSCALL_AUDIT	5
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
@@ -123,6 +124,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
 #define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
+#define _TIF_SYSCALL_AUDIT		(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
diff --git a/init/Kconfig b/init/Kconfig
index 0eda68f0ad54..288444b4cd8a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -238,7 +238,7 @@ config AUDIT
 
 config AUDITSYSCALL
 	bool "Enable system-call auditing support"
-	depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64)
+	depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
 	default y if SECURITY_SELINUX
 	help
 	  Enable low-overhead system-call auditing infrastructure that
-- 
cgit v1.2.3


From 6d01f51086cf6c475470cdae67d2f45e5fb57833 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 26 Nov 2007 18:17:21 +0900
Subject: sh: Add SH7203 CPU support.

This adds support for the SH7203 (SH-2A) CPU.

Signed-off-by: Kieran Bingham <kbingham@mpc-data.co.uk>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                        |   8 +-
 arch/sh/Kconfig.debug                  |   1 +
 arch/sh/kernel/cpu/sh2a/Makefile       |   1 +
 arch/sh/kernel/cpu/sh2a/clock-sh7203.c |  89 +++++++++++
 arch/sh/kernel/cpu/sh2a/probe.c        |  19 ++-
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c | 281 +++++++++++++++++++++++++++++++++
 arch/sh/kernel/setup.c                 |   1 +
 arch/sh/kernel/timers/timer-cmt.c      |   2 +-
 drivers/serial/sh-sci.h                |   6 +-
 include/asm-sh/bugs.h                  |   2 +-
 include/asm-sh/cpu-sh2a/freq.h         |   2 -
 include/asm-sh/processor.h             |   2 +-
 12 files changed, 399 insertions(+), 15 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh2a/clock-sh7203.c
 create mode 100644 arch/sh/kernel/cpu/sh2a/setup-sh7203.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d99ba375cc87..d8f62b842656 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -158,6 +158,10 @@ config CPU_SUBTYPE_SH7619
 
 # SH-2A Processor Support
 
+config CPU_SUBTYPE_SH7203
+	bool "Support SH7203 processor"
+	select CPU_SH2A
+
 config CPU_SUBTYPE_SH7206
 	bool "Support SH7206 processor"
 	select CPU_SH2A
@@ -556,7 +560,7 @@ config SH_PCLK_FREQ
 	default "32000000" if CPU_SUBTYPE_SH7722
 	default "33333333" if CPU_SUBTYPE_SH7770 || \
 			      CPU_SUBTYPE_SH7760 || CPU_SUBTYPE_SH7705 || \
-			      CPU_SUBTYPE_SH7206
+			      CPU_SUBTYPE_SH7203 || CPU_SUBTYPE_SH7206
 	default "60000000" if CPU_SUBTYPE_SH7751 || CPU_SUBTYPE_SH7751R
 	default "66000000" if CPU_SUBTYPE_SH4_202
 	default "50000000"
@@ -567,7 +571,7 @@ config SH_PCLK_FREQ
 
 config SH_CLK_MD
 	int "CPU Mode Pin Setting"
-	depends on CPU_SUBTYPE_SH7619 || CPU_SUBTYPE_SH7206
+	depends on CPU_SH2
 	default 6 if CPU_SUBTYPE_SH7206
 	default 5 if CPU_SUBTYPE_SH7619
 	default 0
diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 89bc174f25c7..f7e362c9d76d 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -32,6 +32,7 @@ config EARLY_SCIF_CONSOLE_PORT
 	depends on EARLY_SCIF_CONSOLE
 	default "0xffe00000" if CPU_SUBTYPE_SH7780
 	default "0xffea0000" if CPU_SUBTYPE_SH7785
+	default "0xfffe8000" if CPU_SUBTYPE_SH7203
 	default "0xfffe9800" if CPU_SUBTYPE_SH7206
 	default "0xf8420000" if CPU_SUBTYPE_SH7619
 	default "0xa4400000" if CPU_SUBTYPE_SH7712 || CPU_SUBTYPE_SH7705
diff --git a/arch/sh/kernel/cpu/sh2a/Makefile b/arch/sh/kernel/cpu/sh2a/Makefile
index 965fa2572b23..5286eeff1cc3 100644
--- a/arch/sh/kernel/cpu/sh2a/Makefile
+++ b/arch/sh/kernel/cpu/sh2a/Makefile
@@ -7,3 +7,4 @@ obj-y	:= common.o probe.o opcode_helper.o
 common-y	+= $(addprefix ../sh2/, ex.o entry.o)
 
 obj-$(CONFIG_CPU_SUBTYPE_SH7206) += setup-sh7206.o clock-sh7206.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7203) += setup-sh7203.o clock-sh7203.o
diff --git a/arch/sh/kernel/cpu/sh2a/clock-sh7203.c b/arch/sh/kernel/cpu/sh2a/clock-sh7203.c
new file mode 100644
index 000000000000..3feb95a4fcbc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh2a/clock-sh7203.c
@@ -0,0 +1,89 @@
+/*
+ * arch/sh/kernel/cpu/sh2a/clock-sh7203.c
+ *
+ * SH7203 support for the clock framework
+ *
+ *  Copyright (C) 2007 Kieran Bingham (MPC-Data Ltd)
+ *
+ * Based on clock-sh7263.c
+ *  Copyright (C) 2006  Yoshinori Sato
+ *
+ * Based on clock-sh4.c
+ *  Copyright (C) 2005  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+const static int pll1rate[]={8,12,16,0};
+const static int pfc_divisors[]={1,2,3,4,6,8,12};
+#define ifc_divisors pfc_divisors
+
+#if (CONFIG_SH_CLK_MD == 0)
+#define PLL2 (1)
+#elif (CONFIG_SH_CLK_MD == 1)
+#define PLL2 (2)
+#elif (CONFIG_SH_CLK_MD == 2)
+#define PLL2 (4)
+#elif (CONFIG_SH_CLK_MD == 3)
+#define PLL2 (4)
+#else
+#error "Illegal Clock Mode!"
+#endif
+
+static void master_clk_init(struct clk *clk)
+{
+	clk->rate *= pll1rate[(ctrl_inw(FREQCR) >> 8) & 0x0003] * PLL2 ;
+}
+
+static struct clk_ops sh7203_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int idx = (ctrl_inw(FREQCR) & 0x0007);
+	clk->rate = clk->parent->rate / pfc_divisors[idx];
+}
+
+static struct clk_ops sh7203_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void bus_clk_recalc(struct clk *clk)
+{
+	int idx = (ctrl_inw(FREQCR) & 0x0007);
+	clk->rate = clk->parent->rate / pfc_divisors[idx-2];
+}
+
+static struct clk_ops sh7203_bus_clk_ops = {
+	.recalc		= bus_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	clk->rate = clk->parent->rate;
+}
+
+static struct clk_ops sh7203_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7203_clk_ops[] = {
+	&sh7203_master_clk_ops,
+	&sh7203_module_clk_ops,
+	&sh7203_bus_clk_ops,
+	&sh7203_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7203_clk_ops))
+		*ops = sh7203_clk_ops[idx];
+}
diff --git a/arch/sh/kernel/cpu/sh2a/probe.c b/arch/sh/kernel/cpu/sh2a/probe.c
index 6d02465704b9..64f10102b720 100644
--- a/arch/sh/kernel/cpu/sh2a/probe.c
+++ b/arch/sh/kernel/cpu/sh2a/probe.c
@@ -3,25 +3,33 @@
  *
  * CPU Subtype Probing for SH-2A.
  *
- * Copyright (C) 2004, 2005 Paul Mundt
+ * Copyright (C) 2004 - 2007  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-
 #include <linux/init.h>
 #include <asm/processor.h>
 #include <asm/cache.h>
 
 int __init detect_cpu_and_cache_system(void)
 {
-	/* Just SH7206 for now .. */
-	boot_cpu_data.type			= CPU_SH7206;
+	/* All SH-2A CPUs have support for 16 and 32-bit opcodes.. */
 	boot_cpu_data.flags			|= CPU_HAS_OP32;
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7203)
+	boot_cpu_data.type			= CPU_SH7203;
+	/* SH7203 has an FPU.. */
+	boot_cpu_data.flags			|= CPU_HAS_FPU;
+#elif defined(CONFIG_CPU_SUBTYPE_SH7206)
+	boot_cpu_data.type			= CPU_SH7206;
+	/* While SH7206 has a DSP.. */
+	boot_cpu_data.flags			|= CPU_HAS_DSP;
+#endif
+
 	boot_cpu_data.dcache.ways		= 4;
-	boot_cpu_data.dcache.way_incr	= (1 << 11);
+	boot_cpu_data.dcache.way_incr		= (1 << 11);
 	boot_cpu_data.dcache.sets		= 128;
 	boot_cpu_data.dcache.entry_shift	= 4;
 	boot_cpu_data.dcache.linesz		= L1_CACHE_BYTES;
@@ -37,4 +45,3 @@ int __init detect_cpu_and_cache_system(void)
 
 	return 0;
 }
-
diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
new file mode 100644
index 000000000000..06f27c8b8583
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
@@ -0,0 +1,281 @@
+/*
+ * SH7203 Setup
+ *
+ *  Copyright (C) 2007  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+enum {
+	UNUSED = 0,
+
+	/* interrupt sources */
+	IRQ0, IRQ1, IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, IRQ7,
+	PINT0, PINT1, PINT2, PINT3, PINT4, PINT5, PINT6, PINT7,
+	DMAC0_DEI, DMAC0_HEI, DMAC1_DEI, DMAC1_HEI,
+	DMAC2_DEI, DMAC2_HEI, DMAC3_DEI, DMAC3_HEI,
+	DMAC4_DEI, DMAC4_HEI, DMAC5_DEI, DMAC5_HEI,
+	DMAC6_DEI, DMAC6_HEI, DMAC7_DEI, DMAC7_HEI,
+	USB, LCDC, CMT0, CMT1, BSC, WDT,
+	MTU2_TGI0A, MTU2_TGI0B, MTU2_TGI0C, MTU2_TGI0D,
+	MTU2_TCI0V, MTU2_TGI0E, MTU2_TGI0F,
+	MTU2_TGI1A, MTU2_TGI1B, MTU2_TCI1V, MTU2_TCI1U,
+	MTU2_TGI2A, MTU2_TGI2B, MTU2_TCI2V, MTU2_TCI2U,
+	MTU2_TGI3A, MTU2_TGI3B, MTU2_TGI3C, MTU2_TGI3D, MTU2_TCI3V,
+	MTU2_TGI4A, MTU2_TGI4B, MTU2_TGI4C, MTU2_TGI4D, MTU2_TCI4V,
+	ADC_ADI,
+	IIC30_STPI, IIC30_NAKI, IIC30_RXI, IIC30_TXI, IIC30_TEI,
+	IIC31_STPI, IIC31_NAKI, IIC31_RXI, IIC31_TXI, IIC31_TEI,
+	IIC32_STPI, IIC32_NAKI, IIC32_RXI, IIC32_TXI, IIC32_TEI,
+	IIC33_STPI, IIC33_NAKI, IIC33_RXI, IIC33_TXI, IIC33_TEI,
+	SCIF0_BRI, SCIF0_ERI, SCIF0_RXI, SCIF0_TXI,
+	SCIF1_BRI, SCIF1_ERI, SCIF1_RXI, SCIF1_TXI,
+	SCIF2_BRI, SCIF2_ERI, SCIF2_RXI, SCIF2_TXI,
+	SCIF3_BRI, SCIF3_ERI, SCIF3_RXI, SCIF3_TXI,
+	SSU0_SSERI, SSU0_SSRXI, SSU0_SSTXI,
+	SSU1_SSERI, SSU1_SSRXI, SSU1_SSTXI,
+	SSI0_SSII, SSI1_SSII, SSI2_SSII, SSI3_SSII,
+	FLCTL_FLSTEI, FLCTL_FLTENDI, FLCTL_FLTREQ0I, FLCTL_FLTREQ1I,
+	RTC_ARM, RTC_PRD, RTC_CUP,
+	RCAN0_ERS, RCAN0_OVR, RCAN0_RM0, RCAN0_RM1, RCAN0_SLE,
+	RCAN1_ERS, RCAN1_OVR, RCAN1_RM0, RCAN1_RM1, RCAN1_SLE,
+
+	/* interrupt groups */
+	PINT, DMAC0, DMAC1, DMAC2, DMAC3, DMAC4, DMAC5, DMAC6, DMAC7,
+	MTU0_ABCD, MTU0_VEF, MTU1_AB, MTU1_VU, MTU2_AB, MTU2_VU,
+	MTU3_ABCD, MTU4_ABCD,
+	IIC30, IIC31, IIC32, IIC33, SCIF0, SCIF1, SCIF2, SCIF3,
+	SSU0, SSU1, FLCTL, RTC, RCAN0, RCAN1
+};
+
+static struct intc_vect vectors[] __initdata = {
+	INTC_IRQ(IRQ0, 64), INTC_IRQ(IRQ1, 65),
+	INTC_IRQ(IRQ2, 66), INTC_IRQ(IRQ3, 67),
+	INTC_IRQ(IRQ4, 68), INTC_IRQ(IRQ5, 69),
+	INTC_IRQ(IRQ6, 70), INTC_IRQ(IRQ7, 71),
+	INTC_IRQ(PINT0, 80), INTC_IRQ(PINT1, 81),
+	INTC_IRQ(PINT2, 82), INTC_IRQ(PINT3, 83),
+	INTC_IRQ(PINT4, 84), INTC_IRQ(PINT5, 85),
+	INTC_IRQ(PINT6, 86), INTC_IRQ(PINT7, 87),
+	INTC_IRQ(DMAC0_DEI, 108), INTC_IRQ(DMAC0_HEI, 109),
+	INTC_IRQ(DMAC1_DEI, 112), INTC_IRQ(DMAC1_HEI, 113),
+	INTC_IRQ(DMAC2_DEI, 116), INTC_IRQ(DMAC2_HEI, 117),
+	INTC_IRQ(DMAC3_DEI, 120), INTC_IRQ(DMAC3_HEI, 121),
+	INTC_IRQ(DMAC4_DEI, 124), INTC_IRQ(DMAC4_HEI, 125),
+	INTC_IRQ(DMAC5_DEI, 128), INTC_IRQ(DMAC5_HEI, 129),
+	INTC_IRQ(DMAC6_DEI, 132), INTC_IRQ(DMAC6_HEI, 133),
+	INTC_IRQ(DMAC7_DEI, 136), INTC_IRQ(DMAC7_HEI, 137),
+	INTC_IRQ(USB, 140), INTC_IRQ(LCDC, 141),
+	INTC_IRQ(CMT0, 142), INTC_IRQ(CMT1, 143),
+	INTC_IRQ(BSC, 144), INTC_IRQ(WDT, 145),
+	INTC_IRQ(MTU2_TGI0A, 146), INTC_IRQ(MTU2_TGI0B, 147),
+	INTC_IRQ(MTU2_TGI0C, 148), INTC_IRQ(MTU2_TGI0D, 149),
+	INTC_IRQ(MTU2_TCI0V, 150),
+	INTC_IRQ(MTU2_TGI0E, 151), INTC_IRQ(MTU2_TGI0F, 152),
+	INTC_IRQ(MTU2_TGI1A, 153), INTC_IRQ(MTU2_TGI1B, 154),
+	INTC_IRQ(MTU2_TCI1V, 155), INTC_IRQ(MTU2_TCI1U, 156),
+	INTC_IRQ(MTU2_TGI2A, 157), INTC_IRQ(MTU2_TGI2B, 158),
+	INTC_IRQ(MTU2_TCI2V, 159), INTC_IRQ(MTU2_TCI2U, 160),
+	INTC_IRQ(MTU2_TGI3A, 161), INTC_IRQ(MTU2_TGI3B, 162),
+	INTC_IRQ(MTU2_TGI3C, 163), INTC_IRQ(MTU2_TGI3D, 164),
+	INTC_IRQ(MTU2_TCI3V, 165),
+	INTC_IRQ(MTU2_TGI4A, 166), INTC_IRQ(MTU2_TGI4B, 167),
+	INTC_IRQ(MTU2_TGI4C, 168), INTC_IRQ(MTU2_TGI4D, 169),
+	INTC_IRQ(MTU2_TCI4V, 170),
+	INTC_IRQ(ADC_ADI, 171),
+	INTC_IRQ(IIC30_STPI, 172), INTC_IRQ(IIC30_NAKI, 173),
+	INTC_IRQ(IIC30_RXI, 174), INTC_IRQ(IIC30_TXI, 175),
+	INTC_IRQ(IIC30_TEI, 176),
+	INTC_IRQ(IIC31_STPI, 177), INTC_IRQ(IIC31_NAKI, 178),
+	INTC_IRQ(IIC31_RXI, 179), INTC_IRQ(IIC31_TXI, 180),
+	INTC_IRQ(IIC31_TEI, 181),
+	INTC_IRQ(IIC32_STPI, 182), INTC_IRQ(IIC32_NAKI, 183),
+	INTC_IRQ(IIC32_RXI, 184), INTC_IRQ(IIC32_TXI, 185),
+	INTC_IRQ(IIC32_TEI, 186),
+	INTC_IRQ(IIC33_STPI, 187), INTC_IRQ(IIC33_NAKI, 188),
+	INTC_IRQ(IIC33_RXI, 189), INTC_IRQ(IIC33_TXI, 190),
+	INTC_IRQ(IIC33_TEI, 191),
+	INTC_IRQ(SCIF0_BRI, 192), INTC_IRQ(SCIF0_ERI, 193),
+	INTC_IRQ(SCIF0_RXI, 194), INTC_IRQ(SCIF0_TXI, 195),
+	INTC_IRQ(SCIF1_BRI, 196), INTC_IRQ(SCIF1_ERI, 197),
+	INTC_IRQ(SCIF1_RXI, 198), INTC_IRQ(SCIF1_TXI, 199),
+	INTC_IRQ(SCIF2_BRI, 200), INTC_IRQ(SCIF2_ERI, 201),
+	INTC_IRQ(SCIF2_RXI, 202), INTC_IRQ(SCIF2_TXI, 203),
+	INTC_IRQ(SCIF3_BRI, 204), INTC_IRQ(SCIF3_ERI, 205),
+	INTC_IRQ(SCIF3_RXI, 206), INTC_IRQ(SCIF3_TXI, 207),
+	INTC_IRQ(SSU0_SSERI, 208), INTC_IRQ(SSU0_SSRXI, 209),
+	INTC_IRQ(SSU0_SSTXI, 210),
+	INTC_IRQ(SSU1_SSERI, 211), INTC_IRQ(SSU1_SSRXI, 212),
+	INTC_IRQ(SSU1_SSTXI, 213),
+	INTC_IRQ(SSI0_SSII, 214), INTC_IRQ(SSI1_SSII, 215),
+	INTC_IRQ(SSI2_SSII, 216), INTC_IRQ(SSI3_SSII, 217),
+	INTC_IRQ(FLCTL_FLSTEI, 224), INTC_IRQ(FLCTL_FLTENDI, 225),
+	INTC_IRQ(FLCTL_FLTREQ0I, 226), INTC_IRQ(FLCTL_FLTREQ1I, 227),
+	INTC_IRQ(RTC_ARM, 231), INTC_IRQ(RTC_PRD, 232),
+	INTC_IRQ(RTC_CUP, 233),
+	INTC_IRQ(RCAN0_ERS, 234), INTC_IRQ(RCAN0_OVR, 235),
+	INTC_IRQ(RCAN0_RM0, 236), INTC_IRQ(RCAN0_RM1, 237),
+	INTC_IRQ(RCAN0_SLE, 238),
+	INTC_IRQ(RCAN1_ERS, 239), INTC_IRQ(RCAN1_OVR, 240),
+	INTC_IRQ(RCAN1_RM0, 241), INTC_IRQ(RCAN1_RM1, 242),
+	INTC_IRQ(RCAN1_SLE, 243),
+};
+
+static struct intc_group groups[] __initdata = {
+	INTC_GROUP(PINT, PINT0, PINT1, PINT2, PINT3,
+		   PINT4, PINT5, PINT6, PINT7),
+	INTC_GROUP(DMAC0, DMAC0_DEI, DMAC0_HEI),
+	INTC_GROUP(DMAC1, DMAC1_DEI, DMAC1_HEI),
+	INTC_GROUP(DMAC2, DMAC2_DEI, DMAC2_HEI),
+	INTC_GROUP(DMAC3, DMAC3_DEI, DMAC3_HEI),
+	INTC_GROUP(DMAC4, DMAC4_DEI, DMAC4_HEI),
+	INTC_GROUP(DMAC5, DMAC5_DEI, DMAC5_HEI),
+	INTC_GROUP(DMAC6, DMAC6_DEI, DMAC6_HEI),
+	INTC_GROUP(DMAC7, DMAC7_DEI, DMAC7_HEI),
+	INTC_GROUP(MTU0_ABCD, MTU2_TGI0A, MTU2_TGI0B, MTU2_TGI0C, MTU2_TGI0D),
+	INTC_GROUP(MTU0_VEF, MTU2_TCI0V, MTU2_TGI0E, MTU2_TGI0F),
+	INTC_GROUP(MTU1_AB, MTU2_TGI1A, MTU2_TGI1B),
+	INTC_GROUP(MTU1_VU, MTU2_TCI1V, MTU2_TCI1U),
+	INTC_GROUP(MTU2_AB, MTU2_TGI2A, MTU2_TGI2B),
+	INTC_GROUP(MTU2_VU, MTU2_TCI2V, MTU2_TCI2U),
+	INTC_GROUP(MTU3_ABCD, MTU2_TGI3A, MTU2_TGI3B, MTU2_TGI3C, MTU2_TGI3D),
+	INTC_GROUP(MTU4_ABCD, MTU2_TGI4A, MTU2_TGI4B, MTU2_TGI4C, MTU2_TGI4D),
+	INTC_GROUP(IIC30, IIC30_STPI, IIC30_NAKI, IIC30_RXI, IIC30_TXI,
+		   IIC30_TEI),
+	INTC_GROUP(IIC31, IIC31_STPI, IIC31_NAKI, IIC31_RXI, IIC31_TXI,
+		   IIC31_TEI),
+	INTC_GROUP(IIC32, IIC32_STPI, IIC32_NAKI, IIC32_RXI, IIC32_TXI,
+		   IIC32_TEI),
+	INTC_GROUP(IIC33, IIC33_STPI, IIC33_NAKI, IIC33_RXI, IIC33_TXI,
+		   IIC33_TEI),
+	INTC_GROUP(SCIF0, SCIF0_BRI, SCIF0_ERI, SCIF0_RXI, SCIF0_TXI),
+	INTC_GROUP(SCIF1, SCIF1_BRI, SCIF1_ERI, SCIF1_RXI, SCIF1_TXI),
+	INTC_GROUP(SCIF2, SCIF2_BRI, SCIF2_ERI, SCIF2_RXI, SCIF2_TXI),
+	INTC_GROUP(SCIF3, SCIF3_BRI, SCIF3_ERI, SCIF3_RXI, SCIF3_TXI),
+	INTC_GROUP(SSU0, SSU0_SSERI, SSU0_SSRXI, SSU0_SSTXI),
+	INTC_GROUP(SSU1, SSU1_SSERI, SSU1_SSRXI, SSU1_SSTXI),
+	INTC_GROUP(FLCTL, FLCTL_FLSTEI, FLCTL_FLTENDI, FLCTL_FLTREQ0I,
+		   FLCTL_FLTREQ1I),
+	INTC_GROUP(RTC, RTC_ARM, RTC_PRD, RTC_CUP),
+	INTC_GROUP(RCAN0, RCAN0_ERS, RCAN0_OVR, RCAN0_RM0, RCAN0_RM1,
+		   RCAN0_SLE),
+	INTC_GROUP(RCAN1, RCAN1_ERS, RCAN1_OVR, RCAN1_RM0, RCAN1_RM1,
+		   RCAN1_SLE),
+};
+
+static struct intc_prio_reg prio_registers[] __initdata = {
+	{ 0xfffe0818, 0, 16, 4, /* IPR01 */ { IRQ0, IRQ1, IRQ2, IRQ3 } },
+	{ 0xfffe081a, 0, 16, 4, /* IPR02 */ { IRQ4, IRQ5, IRQ6, IRQ7 } },
+	{ 0xfffe0820, 0, 16, 4, /* IPR05 */ { PINT, 0, 0, 0 } },
+	{ 0xfffe0c00, 0, 16, 4, /* IPR06 */ { DMAC0, DMAC1, DMAC2, DMAC3 } },
+	{ 0xfffe0c02, 0, 16, 4, /* IPR07 */ { DMAC4, DMAC5, DMAC6, DMAC7 } },
+	{ 0xfffe0c04, 0, 16, 4, /* IPR08 */ { USB, LCDC, CMT0, CMT1 } },
+	{ 0xfffe0c06, 0, 16, 4, /* IPR09 */ { BSC, WDT, MTU0_ABCD, MTU0_VEF } },
+	{ 0xfffe0c08, 0, 16, 4, /* IPR10 */ { MTU1_AB, MTU1_VU, MTU2_AB,
+					      MTU2_VU } },
+	{ 0xfffe0c0a, 0, 16, 4, /* IPR11 */ { MTU3_ABCD, MTU2_TCI3V, MTU4_ABCD,
+					      MTU2_TCI4V } },
+	{ 0xfffe0c0c, 0, 16, 4, /* IPR12 */ { ADC_ADI, IIC30, IIC31, IIC32 } },
+	{ 0xfffe0c0e, 0, 16, 4, /* IPR13 */ { IIC33, SCIF0, SCIF1, SCIF2 } },
+	{ 0xfffe0c10, 0, 16, 4, /* IPR14 */ { SCIF3, SSU0, SSU1, SSI0_SSII } },
+	{ 0xfffe0c12, 0, 16, 4, /* IPR15 */ { SSI1_SSII, SSI2_SSII,
+					      SSI3_SSII, 0 } },
+	{ 0xfffe0c14, 0, 16, 4, /* IPR16 */ { FLCTL, 0, RTC, RCAN0 } },
+	{ 0xfffe0c16, 0, 16, 4, /* IPR17 */ { RCAN1, 0, 0, 0 } },
+};
+
+static struct intc_mask_reg mask_registers[] __initdata = {
+	{ 0xfffe0808, 0, 16, /* PINTER */
+	  { 0, 0, 0, 0, 0, 0, 0, 0,
+	    PINT7, PINT6, PINT5, PINT4, PINT3, PINT2, PINT1, PINT0 } },
+};
+
+static DECLARE_INTC_DESC(intc_desc, "sh7203", vectors, groups,
+			 NULL, mask_registers, prio_registers, NULL);
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfffe8000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		=  { 193, 194, 195, 192 },
+	}, {
+		.mapbase	= 0xfffe8800,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		=  { 197, 198, 199, 196 },
+	}, {
+		.mapbase	= 0xfffe9000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		=  { 201, 202, 203, 200 },
+	}, {
+		.mapbase	= 0xfffe9800,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		=  { 205, 206, 207, 204 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		.start	= 0xffff2000,
+		.end	= 0xffff2000 + 0x58 - 1,
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= 232,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= 233,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= 231,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static struct platform_device *sh7203_devices[] __initdata = {
+	&sci_device,
+	&rtc_device,
+};
+
+static int __init sh7203_devices_setup(void)
+{
+	return platform_add_devices(sh7203_devices,
+				    ARRAY_SIZE(sh7203_devices));
+}
+__initcall(sh7203_devices_setup);
+
+void __init plat_irq_setup(void)
+{
+	register_intc_controller(&intc_desc);
+}
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 1ec40091a4e8..c5a453fb19ca 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -294,6 +294,7 @@ void __init setup_arch(char **cmdline_p)
 }
 
 static const char *cpu_name[] = {
+	[CPU_SH7203]	= "SH7203",
 	[CPU_SH7206]	= "SH7206",	[CPU_SH7619]	= "SH7619",
 	[CPU_SH7705]	= "SH7705",	[CPU_SH7706]	= "SH7706",
 	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
diff --git a/arch/sh/kernel/timers/timer-cmt.c b/arch/sh/kernel/timers/timer-cmt.c
index 82de6895ade5..4828a53d81ea 100644
--- a/arch/sh/kernel/timers/timer-cmt.c
+++ b/arch/sh/kernel/timers/timer-cmt.c
@@ -31,7 +31,7 @@
 #define cmt_clock_enable() do {	ctrl_outb(ctrl_inb(STBCR3) & ~0x10, STBCR3); } while(0)
 #define CMT_CMCSR_INIT	0x0040
 #define CMT_CMCSR_CALIB	0x0000
-#elif defined(CONFIG_CPU_SUBTYPE_SH7206)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7203) || defined(CONFIG_CPU_SUBTYPE_SH7206)
 #define CMT_CMSTR	0xfffec000
 #define CMT_CMCSR_0	0xfffec002
 #define CMT_CMCNT_0	0xfffec004
diff --git a/drivers/serial/sh-sci.h b/drivers/serial/sh-sci.h
index d24621ce799a..cde06a481829 100644
--- a/drivers/serial/sh-sci.h
+++ b/drivers/serial/sh-sci.h
@@ -142,7 +142,8 @@
 # define SCIF_OPER	0x0001		/* Overrun error bit */
 # define SCSCR_INIT(port)	0x3a	/* TIE=0,RIE=0,TE=1,RE=1,REIE=1 */
 # define SCIF_ONLY
-#elif defined(CONFIG_CPU_SUBTYPE_SH7206)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7203) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7206)
 # define SCSPTR0 0xfffe8020 /* 16 bit SCIF */
 # define SCSPTR1 0xfffe8820 /* 16 bit SCIF */
 # define SCSPTR2 0xfffe9020 /* 16 bit SCIF */
@@ -617,7 +618,8 @@ static inline int sci_rxd_in(struct uart_port *port)
 		return ctrl_inw(SCSPTR5) & 0x0001 ? 1 : 0; /* SCIF */
 	return 1;
 }
-#elif defined(CONFIG_CPU_SUBTYPE_SH7206)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7203) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7206)
 static inline int sci_rxd_in(struct uart_port *port)
 {
 	if (port->mapbase == 0xfffe8000)
diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index e2d79567e2a9..94149c8a759c 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -25,7 +25,7 @@ static void __init check_bugs(void)
 	case CPU_SH7619:
 		*p++ = '2';
 		break;
-	case CPU_SH7206:
+	case CPU_SH7203 ... CPU_SH7206:
 		*p++ = '2';
 		*p++ = 'a';
 		break;
diff --git a/include/asm-sh/cpu-sh2a/freq.h b/include/asm-sh/cpu-sh2a/freq.h
index e518fff6d10f..830fd43b6cdc 100644
--- a/include/asm-sh/cpu-sh2a/freq.h
+++ b/include/asm-sh/cpu-sh2a/freq.h
@@ -10,9 +10,7 @@
 #ifndef __ASM_CPU_SH2A_FREQ_H
 #define __ASM_CPU_SH2A_FREQ_H
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7206)
 #define FREQCR	0xfffe0010
-#endif
 
 #endif /* __ASM_CPU_SH2A_FREQ_H */
 
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index 3ae2a1c223f2..70f083f07d71 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -17,7 +17,7 @@ enum cpu_type {
 	CPU_SH7619,
 
 	/* SH-2A types */
-	CPU_SH7206,
+	CPU_SH7203, CPU_SH7206,
 
 	/* SH-3 types */
 	CPU_SH7705, CPU_SH7706, CPU_SH7707,
-- 
cgit v1.2.3


From a8f67f4b4d4b74cd14d3540ade8657ebee543340 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 26 Nov 2007 19:54:02 +0900
Subject: sh: Add SH7263 CPU support.

This adds support for the SH7263 (SH-2A) CPU.

This particular CPU is a superset of SH7203, adding some additional
peripheral blocks and hooking up additional (reserved on SH7203)
vectors in the INTC block.

No visibly nasty surprises, yet..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                        |  7 +++++-
 arch/sh/Kconfig.debug                  |  2 +-
 arch/sh/kernel/cpu/sh2a/Makefile       |  1 +
 arch/sh/kernel/cpu/sh2a/probe.c        |  3 +++
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c | 42 ++++++++++++++++++++++++++++++++--
 arch/sh/kernel/setup.c                 |  2 +-
 arch/sh/kernel/timers/timer-cmt.c      |  4 +++-
 drivers/serial/sh-sci.h                |  6 +++--
 include/asm-sh/bugs.h                  |  2 +-
 include/asm-sh/processor.h             |  2 +-
 10 files changed, 61 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d8f62b842656..c18a5512ac82 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -166,6 +166,10 @@ config CPU_SUBTYPE_SH7206
 	bool "Support SH7206 processor"
 	select CPU_SH2A
 
+config CPU_SUBTYPE_SH7263
+	bool "Support SH7263 processor"
+	select CPU_SH2A
+
 # SH-3 Processor Support
 
 config CPU_SUBTYPE_SH7705
@@ -560,7 +564,8 @@ config SH_PCLK_FREQ
 	default "32000000" if CPU_SUBTYPE_SH7722
 	default "33333333" if CPU_SUBTYPE_SH7770 || \
 			      CPU_SUBTYPE_SH7760 || CPU_SUBTYPE_SH7705 || \
-			      CPU_SUBTYPE_SH7203 || CPU_SUBTYPE_SH7206
+			      CPU_SUBTYPE_SH7203 || CPU_SUBTYPE_SH7206 || \
+			      CPU_SUBTYPE_SH7263
 	default "60000000" if CPU_SUBTYPE_SH7751 || CPU_SUBTYPE_SH7751R
 	default "66000000" if CPU_SUBTYPE_SH4_202
 	default "50000000"
diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index f7e362c9d76d..b0dcb240d798 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -33,7 +33,7 @@ config EARLY_SCIF_CONSOLE_PORT
 	default "0xffe00000" if CPU_SUBTYPE_SH7780
 	default "0xffea0000" if CPU_SUBTYPE_SH7785
 	default "0xfffe8000" if CPU_SUBTYPE_SH7203
-	default "0xfffe9800" if CPU_SUBTYPE_SH7206
+	default "0xfffe9800" if CPU_SUBTYPE_SH7206 || CPU_SUBTYPE_SH7263
 	default "0xf8420000" if CPU_SUBTYPE_SH7619
 	default "0xa4400000" if CPU_SUBTYPE_SH7712 || CPU_SUBTYPE_SH7705
 	default "0xa4430000" if CPU_SUBTYPE_SH7720
diff --git a/arch/sh/kernel/cpu/sh2a/Makefile b/arch/sh/kernel/cpu/sh2a/Makefile
index 5286eeff1cc3..50e4d0ffdd6f 100644
--- a/arch/sh/kernel/cpu/sh2a/Makefile
+++ b/arch/sh/kernel/cpu/sh2a/Makefile
@@ -8,3 +8,4 @@ common-y	+= $(addprefix ../sh2/, ex.o entry.o)
 
 obj-$(CONFIG_CPU_SUBTYPE_SH7206) += setup-sh7206.o clock-sh7206.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7203) += setup-sh7203.o clock-sh7203.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7263) += setup-sh7203.o clock-sh7203.o
diff --git a/arch/sh/kernel/cpu/sh2a/probe.c b/arch/sh/kernel/cpu/sh2a/probe.c
index 64f10102b720..6910e2664468 100644
--- a/arch/sh/kernel/cpu/sh2a/probe.c
+++ b/arch/sh/kernel/cpu/sh2a/probe.c
@@ -22,6 +22,9 @@ int __init detect_cpu_and_cache_system(void)
 	boot_cpu_data.type			= CPU_SH7203;
 	/* SH7203 has an FPU.. */
 	boot_cpu_data.flags			|= CPU_HAS_FPU;
+#elif defined(CONFIG_CPU_SUBTYPE_SH7263)
+	boot_cpu_data.type			= CPU_SH7263;
+	boot_cpu_data.flags			|= CPU_HAS_FPU;
 #elif defined(CONFIG_CPU_SUBTYPE_SH7206)
 	boot_cpu_data.type			= CPU_SH7206;
 	/* While SH7206 has a DSP.. */
diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
index 06f27c8b8583..3518f9c37d93 100644
--- a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
+++ b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
@@ -1,5 +1,5 @@
 /*
- * SH7203 Setup
+ * SH7203 and SH7263 Setup
  *
  *  Copyright (C) 2007  Paul Mundt
  *
@@ -41,17 +41,27 @@ enum {
 	SSU0_SSERI, SSU0_SSRXI, SSU0_SSTXI,
 	SSU1_SSERI, SSU1_SSRXI, SSU1_SSTXI,
 	SSI0_SSII, SSI1_SSII, SSI2_SSII, SSI3_SSII,
+
+	/* ROM-DEC, SDHI, SRC, and IEB are SH7263 specific */
+	ROMDEC_ISY, ROMDEC_IERR, ROMDEC_IARG, ROMDEC_ISEC, ROMDEC_IBUF,
+	ROMDEC_IREADY,
+
 	FLCTL_FLSTEI, FLCTL_FLTENDI, FLCTL_FLTREQ0I, FLCTL_FLTREQ1I,
+
+	SDHI3, SDHI0, SDHI1,
+
 	RTC_ARM, RTC_PRD, RTC_CUP,
 	RCAN0_ERS, RCAN0_OVR, RCAN0_RM0, RCAN0_RM1, RCAN0_SLE,
 	RCAN1_ERS, RCAN1_OVR, RCAN1_RM0, RCAN1_RM1, RCAN1_SLE,
 
+	SRC_OVF, SRC_ODFI, SRC_IDEI, IEBI,
+
 	/* interrupt groups */
 	PINT, DMAC0, DMAC1, DMAC2, DMAC3, DMAC4, DMAC5, DMAC6, DMAC7,
 	MTU0_ABCD, MTU0_VEF, MTU1_AB, MTU1_VU, MTU2_AB, MTU2_VU,
 	MTU3_ABCD, MTU4_ABCD,
 	IIC30, IIC31, IIC32, IIC33, SCIF0, SCIF1, SCIF2, SCIF3,
-	SSU0, SSU1, FLCTL, RTC, RCAN0, RCAN1
+	SSU0, SSU1, ROMDEC, SDHI, FLCTL, RTC, RCAN0, RCAN1, SRC
 };
 
 static struct intc_vect vectors[] __initdata = {
@@ -125,6 +135,20 @@ static struct intc_vect vectors[] __initdata = {
 	INTC_IRQ(RCAN1_ERS, 239), INTC_IRQ(RCAN1_OVR, 240),
 	INTC_IRQ(RCAN1_RM0, 241), INTC_IRQ(RCAN1_RM1, 242),
 	INTC_IRQ(RCAN1_SLE, 243),
+
+	/* SH7263-specific trash */
+#ifdef CONFIG_CPU_SUBTYPE_SH7263
+	INTC_IRQ(ROMDEC_ISY, 218), INTC_IRQ(ROMDEC_IERR, 219),
+	INTC_IRQ(ROMDEC_IARG, 220), INTC_IRQ(ROMDEC_ISEC, 221),
+	INTC_IRQ(ROMDEC_IBUF, 222), INTC_IRQ(ROMDEC_IREADY, 223),
+
+	INTC_IRQ(SDHI3, 228), INTC_IRQ(SDHI0, 229), INTC_IRQ(SDHI1, 230),
+
+	INTC_IRQ(SRC_OVF, 244), INTC_IRQ(SRC_ODFI, 245),
+	INTC_IRQ(SRC_IDEI, 246),
+
+	INTC_IRQ(IEBI, 247),
+#endif
 };
 
 static struct intc_group groups[] __initdata = {
@@ -167,6 +191,13 @@ static struct intc_group groups[] __initdata = {
 		   RCAN0_SLE),
 	INTC_GROUP(RCAN1, RCAN1_ERS, RCAN1_OVR, RCAN1_RM0, RCAN1_RM1,
 		   RCAN1_SLE),
+
+#ifdef CONFIG_CPU_SUBTYPE_SH7263
+	INTC_GROUP(ROMDEC, ROMDEC_ISY, ROMDEC_IERR, ROMDEC_IARG,
+		   ROMDEC_ISEC, ROMDEC_IBUF, ROMDEC_IREADY),
+	INTC_GROUP(SDHI, SDHI3, SDHI0, SDHI1),
+	INTC_GROUP(SRC, SRC_OVF, SRC_ODFI, SRC_IDEI),
+#endif
 };
 
 static struct intc_prio_reg prio_registers[] __initdata = {
@@ -184,10 +215,17 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xfffe0c0c, 0, 16, 4, /* IPR12 */ { ADC_ADI, IIC30, IIC31, IIC32 } },
 	{ 0xfffe0c0e, 0, 16, 4, /* IPR13 */ { IIC33, SCIF0, SCIF1, SCIF2 } },
 	{ 0xfffe0c10, 0, 16, 4, /* IPR14 */ { SCIF3, SSU0, SSU1, SSI0_SSII } },
+#ifdef CONFIG_CPU_SUBTYPE_SH7203
 	{ 0xfffe0c12, 0, 16, 4, /* IPR15 */ { SSI1_SSII, SSI2_SSII,
 					      SSI3_SSII, 0 } },
 	{ 0xfffe0c14, 0, 16, 4, /* IPR16 */ { FLCTL, 0, RTC, RCAN0 } },
 	{ 0xfffe0c16, 0, 16, 4, /* IPR17 */ { RCAN1, 0, 0, 0 } },
+#else
+	{ 0xfffe0c12, 0, 16, 4, /* IPR15 */ { SSI1_SSII, SSI2_SSII,
+					      SSI3_SSII, ROMDEC } },
+	{ 0xfffe0c14, 0, 16, 4, /* IPR16 */ { FLCTL, SDHI, RTC, RCAN0 } },
+	{ 0xfffe0c16, 0, 16, 4, /* IPR17 */ { RCAN1, SRC, IEBI, 0 } },
+#endif
 };
 
 static struct intc_mask_reg mask_registers[] __initdata = {
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index c5a453fb19ca..6891cc93548e 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -294,7 +294,7 @@ void __init setup_arch(char **cmdline_p)
 }
 
 static const char *cpu_name[] = {
-	[CPU_SH7203]	= "SH7203",
+	[CPU_SH7203]	= "SH7203",	[CPU_SH7263]	= "SH7263",
 	[CPU_SH7206]	= "SH7206",	[CPU_SH7619]	= "SH7619",
 	[CPU_SH7705]	= "SH7705",	[CPU_SH7706]	= "SH7706",
 	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
diff --git a/arch/sh/kernel/timers/timer-cmt.c b/arch/sh/kernel/timers/timer-cmt.c
index 4828a53d81ea..499e07beebe2 100644
--- a/arch/sh/kernel/timers/timer-cmt.c
+++ b/arch/sh/kernel/timers/timer-cmt.c
@@ -31,7 +31,9 @@
 #define cmt_clock_enable() do {	ctrl_outb(ctrl_inb(STBCR3) & ~0x10, STBCR3); } while(0)
 #define CMT_CMCSR_INIT	0x0040
 #define CMT_CMCSR_CALIB	0x0000
-#elif defined(CONFIG_CPU_SUBTYPE_SH7203) || defined(CONFIG_CPU_SUBTYPE_SH7206)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7203) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7206) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7263)
 #define CMT_CMSTR	0xfffec000
 #define CMT_CMCSR_0	0xfffec002
 #define CMT_CMCNT_0	0xfffec004
diff --git a/drivers/serial/sh-sci.h b/drivers/serial/sh-sci.h
index cde06a481829..0187dccfe8c1 100644
--- a/drivers/serial/sh-sci.h
+++ b/drivers/serial/sh-sci.h
@@ -143,7 +143,8 @@
 # define SCSCR_INIT(port)	0x3a	/* TIE=0,RIE=0,TE=1,RE=1,REIE=1 */
 # define SCIF_ONLY
 #elif defined(CONFIG_CPU_SUBTYPE_SH7203) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7206)
+      defined(CONFIG_CPU_SUBTYPE_SH7206) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7263)
 # define SCSPTR0 0xfffe8020 /* 16 bit SCIF */
 # define SCSPTR1 0xfffe8820 /* 16 bit SCIF */
 # define SCSPTR2 0xfffe9020 /* 16 bit SCIF */
@@ -619,7 +620,8 @@ static inline int sci_rxd_in(struct uart_port *port)
 	return 1;
 }
 #elif defined(CONFIG_CPU_SUBTYPE_SH7203) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7206)
+      defined(CONFIG_CPU_SUBTYPE_SH7206) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7263)
 static inline int sci_rxd_in(struct uart_port *port)
 {
 	if (port->mapbase == 0xfffe8000)
diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index 94149c8a759c..9ce67fd51190 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -25,7 +25,7 @@ static void __init check_bugs(void)
 	case CPU_SH7619:
 		*p++ = '2';
 		break;
-	case CPU_SH7203 ... CPU_SH7206:
+	case CPU_SH7203 ... CPU_SH7263:
 		*p++ = '2';
 		*p++ = 'a';
 		break;
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index 70f083f07d71..bbbdaab8de5f 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -17,7 +17,7 @@ enum cpu_type {
 	CPU_SH7619,
 
 	/* SH-2A types */
-	CPU_SH7203, CPU_SH7206,
+	CPU_SH7203, CPU_SH7206, CPU_SH7263,
 
 	/* SH-3 types */
 	CPU_SH7705, CPU_SH7706, CPU_SH7707,
-- 
cgit v1.2.3


From 74d99a5e262229ee865f6f68528d10b82471ead6 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 26 Nov 2007 20:38:36 +0900
Subject: sh: SH-2A FPU support.

Signed-off-by: Kieran Bingham <kbingham@mpc-data.co.uk>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  |   2 +
 arch/sh/Makefile                 |   4 -
 arch/sh/kernel/cpu/sh2/entry.S   |  17 +-
 arch/sh/kernel/cpu/sh2a/Makefile |   2 +
 arch/sh/kernel/cpu/sh2a/fpu.c    | 633 +++++++++++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh3/ex.S      |   2 +-
 arch/sh/kernel/cpu/sh4/fpu.c     |  23 +-
 arch/sh/kernel/traps_32.c        |  13 +-
 include/asm-sh/fpu.h             |   2 +
 include/asm-sh/sigcontext.h      |   3 +-
 include/asm-sh/system.h          |   2 +
 11 files changed, 673 insertions(+), 30 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh2a/fpu.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index c18a5512ac82..2dc3b177193c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -161,6 +161,7 @@ config CPU_SUBTYPE_SH7619
 config CPU_SUBTYPE_SH7203
 	bool "Support SH7203 processor"
 	select CPU_SH2A
+	select CPU_HAS_FPU
 
 config CPU_SUBTYPE_SH7206
 	bool "Support SH7206 processor"
@@ -169,6 +170,7 @@ config CPU_SUBTYPE_SH7206
 config CPU_SUBTYPE_SH7263
 	bool "Support SH7263 processor"
 	select CPU_SH2A
+	select CPU_HAS_FPU
 
 # SH-3 Processor Support
 
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index f7cbc13fd2a0..292d8618248d 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -20,10 +20,6 @@ isa-$(CONFIG_CPU_SH4AL_DSP)		:= sh4al
 isa-$(CONFIG_CPU_SH5)			:= shmedia
 isa-$(CONFIG_SH_DSP)			:= $(isa-y)-dsp
 
-ifndef CONFIG_MMU
-isa-y			:= $(isa-y)-nommu
-endif
-
 ifndef CONFIG_SH_DSP
 ifndef CONFIG_SH_FPU
 isa-y			:= $(isa-y)-nofpu
diff --git a/arch/sh/kernel/cpu/sh2/entry.S b/arch/sh/kernel/cpu/sh2/entry.S
index 4ff2334b4a38..7a26569e7956 100644
--- a/arch/sh/kernel/cpu/sh2/entry.S
+++ b/arch/sh/kernel/cpu/sh2/entry.S
@@ -149,6 +149,14 @@ ENTRY(exception_handler)
 	mov	#32,r8
 	cmp/hs	r8,r9
 	bt	trap_entry	! 64 > vec >= 32  is trap
+
+#if defined(CONFIG_SH_FPU)
+	mov     #13,r8
+	cmp/eq  r8,r9
+	bt      10f             ! fpu
+	nop
+#endif
+
 	mov.l	4f,r8
 	mov	r9,r4
 	shll2	r9
@@ -158,6 +166,10 @@ ENTRY(exception_handler)
 	cmp/eq	r9,r8
 	bf	3f
 	mov.l	8f,r8		! unhandled exception
+#if defined(CONFIG_SH_FPU)
+10:
+	mov.l	9f, r8		! unhandled exception
+#endif
 3:
 	mov.l	5f,r10
 	jmp	@r8
@@ -177,7 +189,10 @@ interrupt_entry:
 6:	.long	ret_from_irq
 7:	.long	do_IRQ
 8:	.long	do_exception_error
-	
+#ifdef CONFIG_SH_FPU
+9:	.long	fpu_error_trap_handler
+#endif
+
 trap_entry:
 	mov	#0x30,r8
 	cmp/ge	r8,r9		! vector 0x20-0x2f is systemcall
diff --git a/arch/sh/kernel/cpu/sh2a/Makefile b/arch/sh/kernel/cpu/sh2a/Makefile
index 50e4d0ffdd6f..b279cdc3a233 100644
--- a/arch/sh/kernel/cpu/sh2a/Makefile
+++ b/arch/sh/kernel/cpu/sh2a/Makefile
@@ -6,6 +6,8 @@ obj-y	:= common.o probe.o opcode_helper.o
 
 common-y	+= $(addprefix ../sh2/, ex.o entry.o)
 
+obj-$(CONFIG_SH_FPU)	+= fpu.o
+
 obj-$(CONFIG_CPU_SUBTYPE_SH7206) += setup-sh7206.o clock-sh7206.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7203) += setup-sh7203.o clock-sh7203.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7263) += setup-sh7203.o clock-sh7203.o
diff --git a/arch/sh/kernel/cpu/sh2a/fpu.c b/arch/sh/kernel/cpu/sh2a/fpu.c
new file mode 100644
index 000000000000..ff99562456fb
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh2a/fpu.c
@@ -0,0 +1,633 @@
+/*
+ * Save/restore floating point context for signal handlers.
+ *
+ * Copyright (C) 1999, 2000  Kaz Kojima & Niibe Yutaka
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * FIXME! These routines can be optimized in big endian case.
+ */
+#include <linux/sched.h>
+#include <linux/signal.h>
+#include <asm/processor.h>
+#include <asm/io.h>
+
+/* The PR (precision) bit in the FP Status Register must be clear when
+ * an frchg instruction is executed, otherwise the instruction is undefined.
+ * Executing frchg with PR set causes a trap on some SH4 implementations.
+ */
+
+#define FPSCR_RCHG 0x00000000
+
+
+/*
+ * Save FPU registers onto task structure.
+ * Assume called with FPU enabled (SR.FD=0).
+ */
+void
+save_fpu(struct task_struct *tsk, struct pt_regs *regs)
+{
+	unsigned long dummy;
+
+	clear_tsk_thread_flag(tsk, TIF_USEDFPU);
+	enable_fpu();
+	asm volatile("sts.l	fpul, @-%0\n\t"
+		     "sts.l	fpscr, @-%0\n\t"
+		     "fmov.s	fr15, @-%0\n\t"
+		     "fmov.s	fr14, @-%0\n\t"
+		     "fmov.s	fr13, @-%0\n\t"
+		     "fmov.s	fr12, @-%0\n\t"
+		     "fmov.s	fr11, @-%0\n\t"
+		     "fmov.s	fr10, @-%0\n\t"
+		     "fmov.s	fr9, @-%0\n\t"
+		     "fmov.s	fr8, @-%0\n\t"
+		     "fmov.s	fr7, @-%0\n\t"
+		     "fmov.s	fr6, @-%0\n\t"
+		     "fmov.s	fr5, @-%0\n\t"
+		     "fmov.s	fr4, @-%0\n\t"
+		     "fmov.s	fr3, @-%0\n\t"
+		     "fmov.s	fr2, @-%0\n\t"
+		     "fmov.s	fr1, @-%0\n\t"
+		     "fmov.s	fr0, @-%0\n\t"
+		     "lds	%3, fpscr\n\t"
+		     : "=r" (dummy)
+		     : "0" ((char *)(&tsk->thread.fpu.hard.status)),
+		       "r" (FPSCR_RCHG),
+		       "r" (FPSCR_INIT)
+		     : "memory");
+
+	disable_fpu();
+	release_fpu(regs);
+}
+
+static void
+restore_fpu(struct task_struct *tsk)
+{
+	unsigned long dummy;
+
+	enable_fpu();
+	asm volatile("fmov.s	@%0+, fr0\n\t"
+		     "fmov.s	@%0+, fr1\n\t"
+		     "fmov.s	@%0+, fr2\n\t"
+		     "fmov.s	@%0+, fr3\n\t"
+		     "fmov.s	@%0+, fr4\n\t"
+		     "fmov.s	@%0+, fr5\n\t"
+		     "fmov.s	@%0+, fr6\n\t"
+		     "fmov.s	@%0+, fr7\n\t"
+		     "fmov.s	@%0+, fr8\n\t"
+		     "fmov.s	@%0+, fr9\n\t"
+		     "fmov.s	@%0+, fr10\n\t"
+		     "fmov.s	@%0+, fr11\n\t"
+		     "fmov.s	@%0+, fr12\n\t"
+		     "fmov.s	@%0+, fr13\n\t"
+		     "fmov.s	@%0+, fr14\n\t"
+		     "fmov.s	@%0+, fr15\n\t"
+		     "lds.l	@%0+, fpscr\n\t"
+		     "lds.l	@%0+, fpul\n\t"
+		     : "=r" (dummy)
+		     : "0" (&tsk->thread.fpu), "r" (FPSCR_RCHG)
+		     : "memory");
+	disable_fpu();
+}
+
+/*
+ * Load the FPU with signalling NANS.  This bit pattern we're using
+ * has the property that no matter wether considered as single or as
+ * double precission represents signaling NANS.
+ */
+
+static void
+fpu_init(void)
+{
+	enable_fpu();
+	asm volatile("lds	%0, fpul\n\t"
+		     "fsts	fpul, fr0\n\t"
+		     "fsts	fpul, fr1\n\t"
+		     "fsts	fpul, fr2\n\t"
+		     "fsts	fpul, fr3\n\t"
+		     "fsts	fpul, fr4\n\t"
+		     "fsts	fpul, fr5\n\t"
+		     "fsts	fpul, fr6\n\t"
+		     "fsts	fpul, fr7\n\t"
+		     "fsts	fpul, fr8\n\t"
+		     "fsts	fpul, fr9\n\t"
+		     "fsts	fpul, fr10\n\t"
+		     "fsts	fpul, fr11\n\t"
+		     "fsts	fpul, fr12\n\t"
+		     "fsts	fpul, fr13\n\t"
+		     "fsts	fpul, fr14\n\t"
+		     "fsts	fpul, fr15\n\t"
+		     "lds	%2, fpscr\n\t"
+		     : /* no output */
+		     : "r" (0), "r" (FPSCR_RCHG), "r" (FPSCR_INIT));
+	disable_fpu();
+}
+
+/*
+ *	Emulate arithmetic ops on denormalized number for some FPU insns.
+ */
+
+/* denormalized float * float */
+static int denormal_mulf(int hx, int hy)
+{
+	unsigned int ix, iy;
+	unsigned long long m, n;
+	int exp, w;
+
+	ix = hx & 0x7fffffff;
+	iy = hy & 0x7fffffff;
+	if (iy < 0x00800000 || ix == 0)
+		return ((hx ^ hy) & 0x80000000);
+
+	exp = (iy & 0x7f800000) >> 23;
+	ix &= 0x007fffff;
+	iy = (iy & 0x007fffff) | 0x00800000;
+	m = (unsigned long long)ix * iy;
+	n = m;
+	w = -1;
+	while (n) { n >>= 1; w++; }
+
+	/* FIXME: use guard bits */
+	exp += w - 126 - 46;
+	if (exp > 0)
+		ix = ((int) (m >> (w - 23)) & 0x007fffff) | (exp << 23);
+	else if (exp + 22 >= 0)
+		ix = (int) (m >> (w - 22 - exp)) & 0x007fffff;
+	else
+		ix = 0;
+
+	ix |= (hx ^ hy) & 0x80000000;
+	return ix;
+}
+
+/* denormalized double * double */
+static void mult64(unsigned long long x, unsigned long long y,
+		unsigned long long *highp, unsigned long long *lowp)
+{
+	unsigned long long sub0, sub1, sub2, sub3;
+	unsigned long long high, low;
+
+	sub0 = (x >> 32) * (unsigned long) (y >> 32);
+	sub1 = (x & 0xffffffffLL) * (unsigned long) (y >> 32);
+	sub2 = (x >> 32) * (unsigned long) (y & 0xffffffffLL);
+	sub3 = (x & 0xffffffffLL) * (unsigned long) (y & 0xffffffffLL);
+	low = sub3;
+	high = 0LL;
+	sub3 += (sub1 << 32);
+	if (low > sub3)
+		high++;
+	low = sub3;
+	sub3 += (sub2 << 32);
+	if (low > sub3)
+		high++;
+	low = sub3;
+	high += (sub1 >> 32) + (sub2 >> 32);
+	high += sub0;
+	*lowp = low;
+	*highp = high;
+}
+
+static inline long long rshift64(unsigned long long mh,
+		unsigned long long ml, int n)
+{
+	if (n >= 64)
+		return mh >> (n - 64);
+	return (mh << (64 - n)) | (ml >> n);
+}
+
+static long long denormal_muld(long long hx, long long hy)
+{
+	unsigned long long ix, iy;
+	unsigned long long mh, ml, nh, nl;
+	int exp, w;
+
+	ix = hx & 0x7fffffffffffffffLL;
+	iy = hy & 0x7fffffffffffffffLL;
+	if (iy < 0x0010000000000000LL || ix == 0)
+		return ((hx ^ hy) & 0x8000000000000000LL);
+
+	exp = (iy & 0x7ff0000000000000LL) >> 52;
+	ix &= 0x000fffffffffffffLL;
+	iy = (iy & 0x000fffffffffffffLL) | 0x0010000000000000LL;
+	mult64(ix, iy, &mh, &ml);
+	nh = mh;
+	nl = ml;
+	w = -1;
+	if (nh) {
+		while (nh) { nh >>= 1; w++;}
+		w += 64;
+	} else
+		while (nl) { nl >>= 1; w++;}
+
+	/* FIXME: use guard bits */
+	exp += w - 1022 - 52 * 2;
+	if (exp > 0)
+		ix = (rshift64(mh, ml, w - 52) & 0x000fffffffffffffLL)
+			| ((long long)exp << 52);
+	else if (exp + 51 >= 0)
+		ix = rshift64(mh, ml, w - 51 - exp) & 0x000fffffffffffffLL;
+	else
+		ix = 0;
+
+	ix |= (hx ^ hy) & 0x8000000000000000LL;
+	return ix;
+}
+
+/* ix - iy where iy: denormal and ix, iy >= 0 */
+static int denormal_subf1(unsigned int ix, unsigned int iy)
+{
+	int frac;
+	int exp;
+
+	if (ix < 0x00800000)
+		return ix - iy;
+
+	exp = (ix & 0x7f800000) >> 23;
+	if (exp - 1 > 31)
+		return ix;
+	iy >>= exp - 1;
+	if (iy == 0)
+		return ix;
+
+	frac = (ix & 0x007fffff) | 0x00800000;
+	frac -= iy;
+	while (frac < 0x00800000) {
+		if (--exp == 0)
+			return frac;
+		frac <<= 1;
+	}
+
+	return (exp << 23) | (frac & 0x007fffff);
+}
+
+/* ix + iy where iy: denormal and ix, iy >= 0 */
+static int denormal_addf1(unsigned int ix, unsigned int iy)
+{
+	int frac;
+	int exp;
+
+	if (ix < 0x00800000)
+		return ix + iy;
+
+	exp = (ix & 0x7f800000) >> 23;
+	if (exp - 1 > 31)
+		return ix;
+	iy >>= exp - 1;
+	if (iy == 0)
+	  return ix;
+
+	frac = (ix & 0x007fffff) | 0x00800000;
+	frac += iy;
+	if (frac >= 0x01000000) {
+		frac >>= 1;
+		++exp;
+	}
+
+	return (exp << 23) | (frac & 0x007fffff);
+}
+
+static int denormal_addf(int hx, int hy)
+{
+	unsigned int ix, iy;
+	int sign;
+
+	if ((hx ^ hy) & 0x80000000) {
+		sign = hx & 0x80000000;
+		ix = hx & 0x7fffffff;
+		iy = hy & 0x7fffffff;
+		if (iy < 0x00800000) {
+			ix = denormal_subf1(ix, iy);
+			if (ix < 0) {
+				ix = -ix;
+				sign ^= 0x80000000;
+			}
+		} else {
+			ix = denormal_subf1(iy, ix);
+			sign ^= 0x80000000;
+		}
+	} else {
+		sign = hx & 0x80000000;
+		ix = hx & 0x7fffffff;
+		iy = hy & 0x7fffffff;
+		if (iy < 0x00800000)
+			ix = denormal_addf1(ix, iy);
+		else
+			ix = denormal_addf1(iy, ix);
+	}
+
+	return sign | ix;
+}
+
+/* ix - iy where iy: denormal and ix, iy >= 0 */
+static long long denormal_subd1(unsigned long long ix, unsigned long long iy)
+{
+	long long frac;
+	int exp;
+
+	if (ix < 0x0010000000000000LL)
+		return ix - iy;
+
+	exp = (ix & 0x7ff0000000000000LL) >> 52;
+	if (exp - 1 > 63)
+		return ix;
+	iy >>= exp - 1;
+	if (iy == 0)
+		return ix;
+
+	frac = (ix & 0x000fffffffffffffLL) | 0x0010000000000000LL;
+	frac -= iy;
+	while (frac < 0x0010000000000000LL) {
+		if (--exp == 0)
+			return frac;
+		frac <<= 1;
+	}
+
+	return ((long long)exp << 52) | (frac & 0x000fffffffffffffLL);
+}
+
+/* ix + iy where iy: denormal and ix, iy >= 0 */
+static long long denormal_addd1(unsigned long long ix, unsigned long long iy)
+{
+	long long frac;
+	long long exp;
+
+	if (ix < 0x0010000000000000LL)
+		return ix + iy;
+
+	exp = (ix & 0x7ff0000000000000LL) >> 52;
+	if (exp - 1 > 63)
+		return ix;
+	iy >>= exp - 1;
+	if (iy == 0)
+	  return ix;
+
+	frac = (ix & 0x000fffffffffffffLL) | 0x0010000000000000LL;
+	frac += iy;
+	if (frac >= 0x0020000000000000LL) {
+		frac >>= 1;
+		++exp;
+	}
+
+	return (exp << 52) | (frac & 0x000fffffffffffffLL);
+}
+
+static long long denormal_addd(long long hx, long long hy)
+{
+	unsigned long long ix, iy;
+	long long sign;
+
+	if ((hx ^ hy) & 0x8000000000000000LL) {
+		sign = hx & 0x8000000000000000LL;
+		ix = hx & 0x7fffffffffffffffLL;
+		iy = hy & 0x7fffffffffffffffLL;
+		if (iy < 0x0010000000000000LL) {
+			ix = denormal_subd1(ix, iy);
+			if (ix < 0) {
+				ix = -ix;
+				sign ^= 0x8000000000000000LL;
+			}
+		} else {
+			ix = denormal_subd1(iy, ix);
+			sign ^= 0x8000000000000000LL;
+		}
+	} else {
+		sign = hx & 0x8000000000000000LL;
+		ix = hx & 0x7fffffffffffffffLL;
+		iy = hy & 0x7fffffffffffffffLL;
+		if (iy < 0x0010000000000000LL)
+			ix = denormal_addd1(ix, iy);
+		else
+			ix = denormal_addd1(iy, ix);
+	}
+
+	return sign | ix;
+}
+
+/**
+ *	denormal_to_double - Given denormalized float number,
+ *	                     store double float
+ *
+ *	@fpu: Pointer to sh_fpu_hard structure
+ *	@n: Index to FP register
+ */
+static void
+denormal_to_double (struct sh_fpu_hard_struct *fpu, int n)
+{
+	unsigned long du, dl;
+	unsigned long x = fpu->fpul;
+	int exp = 1023 - 126;
+
+	if (x != 0 && (x & 0x7f800000) == 0) {
+		du = (x & 0x80000000);
+		while ((x & 0x00800000) == 0) {
+			x <<= 1;
+			exp--;
+		}
+		x &= 0x007fffff;
+		du |= (exp << 20) | (x >> 3);
+		dl = x << 29;
+
+		fpu->fp_regs[n] = du;
+		fpu->fp_regs[n+1] = dl;
+	}
+}
+
+/**
+ *	ieee_fpe_handler - Handle denormalized number exception
+ *
+ *	@regs: Pointer to register structure
+ *
+ *	Returns 1 when it's handled (should not cause exception).
+ */
+static int
+ieee_fpe_handler (struct pt_regs *regs)
+{
+	unsigned short insn = *(unsigned short *) regs->pc;
+	unsigned short finsn;
+	unsigned long nextpc;
+	int nib[4] = {
+		(insn >> 12) & 0xf,
+		(insn >> 8) & 0xf,
+		(insn >> 4) & 0xf,
+		insn & 0xf};
+
+	if (nib[0] == 0xb ||
+	    (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb)) /* bsr & jsr */
+		regs->pr = regs->pc + 4;
+	if (nib[0] == 0xa || nib[0] == 0xb) { /* bra & bsr */
+		nextpc = regs->pc + 4 + ((short) ((insn & 0xfff) << 4) >> 3);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xd) { /* bt/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		else
+			nextpc = regs->pc + 4;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xf) { /* bf/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4;
+		else
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x4 && nib[3] == 0xb &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* jmp & jsr */
+		nextpc = regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x0 && nib[3] == 0x3 &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* braf & bsrf */
+		nextpc = regs->pc + 4 + regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (insn == 0x000b) { /* rts */
+		nextpc = regs->pr;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else {
+		nextpc = regs->pc + 2;
+		finsn = insn;
+	}
+
+#define FPSCR_FPU_ERROR (1 << 17)
+
+	if ((finsn & 0xf1ff) == 0xf0ad) { /* fcnvsd */
+		struct task_struct *tsk = current;
+
+		if ((tsk->thread.fpu.hard.fpscr & FPSCR_FPU_ERROR)) {
+			/* FPU error */
+			denormal_to_double (&tsk->thread.fpu.hard,
+					    (finsn >> 8) & 0xf);
+		} else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	} else if ((finsn & 0xf00f) == 0xf002) { /* fmul */
+		struct task_struct *tsk = current;
+		int fpscr;
+		int n, m, prec;
+		unsigned int hx, hy;
+
+		n = (finsn >> 8) & 0xf;
+		m = (finsn >> 4) & 0xf;
+		hx = tsk->thread.fpu.hard.fp_regs[n];
+		hy = tsk->thread.fpu.hard.fp_regs[m];
+		fpscr = tsk->thread.fpu.hard.fpscr;
+		prec = fpscr & (1 << 19);
+
+		if ((fpscr & FPSCR_FPU_ERROR)
+		     && (prec && ((hx & 0x7fffffff) < 0x00100000
+				   || (hy & 0x7fffffff) < 0x00100000))) {
+			long long llx, lly;
+
+			/* FPU error because of denormal */
+			llx = ((long long) hx << 32)
+			       | tsk->thread.fpu.hard.fp_regs[n+1];
+			lly = ((long long) hy << 32)
+			       | tsk->thread.fpu.hard.fp_regs[m+1];
+			if ((hx & 0x7fffffff) >= 0x00100000)
+				llx = denormal_muld(lly, llx);
+			else
+				llx = denormal_muld(llx, lly);
+			tsk->thread.fpu.hard.fp_regs[n] = llx >> 32;
+			tsk->thread.fpu.hard.fp_regs[n+1] = llx & 0xffffffff;
+		} else if ((fpscr & FPSCR_FPU_ERROR)
+		     && (!prec && ((hx & 0x7fffffff) < 0x00800000
+				   || (hy & 0x7fffffff) < 0x00800000))) {
+			/* FPU error because of denormal */
+			if ((hx & 0x7fffffff) >= 0x00800000)
+				hx = denormal_mulf(hy, hx);
+			else
+				hx = denormal_mulf(hx, hy);
+			tsk->thread.fpu.hard.fp_regs[n] = hx;
+		} else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	} else if ((finsn & 0xf00e) == 0xf000) { /* fadd, fsub */
+		struct task_struct *tsk = current;
+		int fpscr;
+		int n, m, prec;
+		unsigned int hx, hy;
+
+		n = (finsn >> 8) & 0xf;
+		m = (finsn >> 4) & 0xf;
+		hx = tsk->thread.fpu.hard.fp_regs[n];
+		hy = tsk->thread.fpu.hard.fp_regs[m];
+		fpscr = tsk->thread.fpu.hard.fpscr;
+		prec = fpscr & (1 << 19);
+
+		if ((fpscr & FPSCR_FPU_ERROR)
+		     && (prec && ((hx & 0x7fffffff) < 0x00100000
+				   || (hy & 0x7fffffff) < 0x00100000))) {
+			long long llx, lly;
+
+			/* FPU error because of denormal */
+			llx = ((long long) hx << 32)
+			       | tsk->thread.fpu.hard.fp_regs[n+1];
+			lly = ((long long) hy << 32)
+			       | tsk->thread.fpu.hard.fp_regs[m+1];
+			if ((finsn & 0xf00f) == 0xf000)
+				llx = denormal_addd(llx, lly);
+			else
+				llx = denormal_addd(llx, lly ^ (1LL << 63));
+			tsk->thread.fpu.hard.fp_regs[n] = llx >> 32;
+			tsk->thread.fpu.hard.fp_regs[n+1] = llx & 0xffffffff;
+		} else if ((fpscr & FPSCR_FPU_ERROR)
+		     && (!prec && ((hx & 0x7fffffff) < 0x00800000
+				   || (hy & 0x7fffffff) < 0x00800000))) {
+			/* FPU error because of denormal */
+			if ((finsn & 0xf00f) == 0xf000)
+				hx = denormal_addf(hx, hy);
+			else
+				hx = denormal_addf(hx, hy ^ 0x80000000);
+			tsk->thread.fpu.hard.fp_regs[n] = hx;
+		} else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	}
+
+	return 0;
+}
+
+BUILD_TRAP_HANDLER(fpu_error)
+{
+	struct task_struct *tsk = current;
+	TRAP_HANDLER_DECL;
+
+	save_fpu(tsk, regs);
+	if (ieee_fpe_handler(regs)) {
+		tsk->thread.fpu.hard.fpscr &=
+			~(FPSCR_CAUSE_MASK | FPSCR_FLAG_MASK);
+		grab_fpu(regs);
+		restore_fpu(tsk);
+		set_tsk_thread_flag(tsk, TIF_USEDFPU);
+		return;
+	}
+
+	force_sig(SIGFPE, tsk);
+}
+
+BUILD_TRAP_HANDLER(fpu_state_restore)
+{
+	struct task_struct *tsk = current;
+	TRAP_HANDLER_DECL;
+
+	grab_fpu(regs);
+	if (!user_mode(regs)) {
+		printk(KERN_ERR "BUG: FPU is used in kernel mode.\n");
+		return;
+	}
+
+	if (used_math()) {
+		/* Using the FPU again.  */
+		restore_fpu(tsk);
+	} else	{
+		/* First time FPU user.  */
+		fpu_init();
+		set_used_math();
+	}
+	set_tsk_thread_flag(tsk, TIF_USEDFPU);
+}
diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S
index b6abf38d3a8d..11b6d9c6edae 100644
--- a/arch/sh/kernel/cpu/sh3/ex.S
+++ b/arch/sh/kernel/cpu/sh3/ex.S
@@ -36,7 +36,7 @@ ENTRY(exception_handling_table)
 	.long	exception_error	! address error store	/* 100 */
 #endif
 #if defined(CONFIG_SH_FPU)
-	.long	do_fpu_error		/* 120 */
+	.long	fpu_error_trap_handler	/* 120 */
 #else
 	.long	exception_error		/* 120 */
 #endif
diff --git a/arch/sh/kernel/cpu/sh4/fpu.c b/arch/sh/kernel/cpu/sh4/fpu.c
index c5a4fc77fa06..e624180b4467 100644
--- a/arch/sh/kernel/cpu/sh4/fpu.c
+++ b/arch/sh/kernel/cpu/sh4/fpu.c
@@ -82,8 +82,8 @@ save_fpu(struct task_struct *tsk, struct pt_regs *regs)
 		       "r" (FPSCR_INIT)
 		     : "memory");
 
- 	disable_fpu();
- 	release_fpu(regs);
+	disable_fpu();
+	release_fpu(regs);
 }
 
 static void
@@ -91,7 +91,7 @@ restore_fpu(struct task_struct *tsk)
 {
 	unsigned long dummy;
 
- 	enable_fpu();
+	enable_fpu();
 	asm volatile("lds	%2, fpscr\n\t"
 		     "fmov.s	@%0+, fr0\n\t"
 		     "fmov.s	@%0+, fr1\n\t"
@@ -138,7 +138,7 @@ restore_fpu(struct task_struct *tsk)
 /*
  * Load the FPU with signalling NANS.  This bit pattern we're using
  * has the property that no matter wether considered as single or as
- * double precision represents signaling NANS.  
+ * double precision represents signaling NANS.
  */
 
 static void
@@ -184,7 +184,7 @@ fpu_init(void)
 		     "lds	%2, fpscr\n\t"
 		     : /* no output */
 		     : "r" (0), "r" (FPSCR_RCHG), "r" (FPSCR_INIT));
- 	disable_fpu();
+	disable_fpu();
 }
 
 /**
@@ -238,7 +238,6 @@ ieee_fpe_handler (struct pt_regs *regs)
 	if (nib[0] == 0xb ||
 	    (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb)) /* bsr & jsr */
 		regs->pr = regs->pc + 4;
-  
 	if (nib[0] == 0xa || nib[0] == 0xb) { /* bra & bsr */
 		nextpc = regs->pc + 4 + ((short) ((insn & 0xfff) << 4) >> 3);
 		finsn = *(unsigned short *) (regs->pc + 2);
@@ -293,12 +292,10 @@ ieee_fpe_handler (struct pt_regs *regs)
 	return 0;
 }
 
-asmlinkage void
-do_fpu_error(unsigned long r4, unsigned long r5, unsigned long r6,
-	     unsigned long r7, struct pt_regs __regs)
+BUILD_TRAP_HANDLER(fpu_error)
 {
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
 	struct task_struct *tsk = current;
+	TRAP_HANDLER_DECL;
 
 	if (ieee_fpe_handler(regs))
 		return;
@@ -308,12 +305,10 @@ do_fpu_error(unsigned long r4, unsigned long r5, unsigned long r6,
 	force_sig(SIGFPE, tsk);
 }
 
-asmlinkage void
-do_fpu_state_restore(unsigned long r4, unsigned long r5, unsigned long r6,
-		     unsigned long r7, struct pt_regs __regs)
+BUILD_TRAP_HANDLER(fpu_state_restore)
 {
-	struct pt_regs *regs = RELOC_HIDE(&__regs, 0);
 	struct task_struct *tsk = current;
+	TRAP_HANDLER_DECL;
 
 	grab_fpu(regs);
 	if (!user_mode(regs)) {
diff --git a/arch/sh/kernel/traps_32.c b/arch/sh/kernel/traps_32.c
index 0d05fb3c48e3..2e58f7a6b746 100644
--- a/arch/sh/kernel/traps_32.c
+++ b/arch/sh/kernel/traps_32.c
@@ -662,11 +662,6 @@ asmlinkage void do_divide_error(unsigned long r4, unsigned long r5,
 }
 #endif
 
-/* arch/sh/kernel/cpu/sh4/fpu.c */
-extern int do_fpu_inst(unsigned short, struct pt_regs *);
-extern asmlinkage void do_fpu_state_restore(unsigned long r4, unsigned long r5,
-		unsigned long r6, unsigned long r7, struct pt_regs __regs);
-
 asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5,
 				unsigned long r6, unsigned long r7,
 				struct pt_regs __regs)
@@ -853,11 +848,11 @@ void __init trap_init(void)
 	set_exception_table_evt(0x820, do_illegal_slot_inst);
 #elif defined(CONFIG_SH_FPU)
 #ifdef CONFIG_CPU_SUBTYPE_SHX3
-	set_exception_table_evt(0xd80, do_fpu_state_restore);
-	set_exception_table_evt(0xda0, do_fpu_state_restore);
+	set_exception_table_evt(0xd80, fpu_state_restore_trap_handler);
+	set_exception_table_evt(0xda0, fpu_state_restore_trap_handler);
 #else
-	set_exception_table_evt(0x800, do_fpu_state_restore);
-	set_exception_table_evt(0x820, do_fpu_state_restore);
+	set_exception_table_evt(0x800, fpu_state_restore_trap_handler);
+	set_exception_table_evt(0x820, fpu_state_restore_trap_handler);
 #endif
 #endif
 
diff --git a/include/asm-sh/fpu.h b/include/asm-sh/fpu.h
index 33db698a6b4d..f8429880a270 100644
--- a/include/asm-sh/fpu.h
+++ b/include/asm-sh/fpu.h
@@ -26,6 +26,8 @@ extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs);
 #define save_fpu(tsk, regs)	do { } while (0)
 #endif
 
+extern int do_fpu_inst(unsigned short, struct pt_regs *);
+
 #define unlazy_fpu(tsk, regs) do {			\
 	if (test_tsk_thread_flag(tsk, TIF_USEDFPU)) {	\
 		save_fpu(tsk, regs);			\
diff --git a/include/asm-sh/sigcontext.h b/include/asm-sh/sigcontext.h
index 8583143fa28f..8ce1435bc0bf 100644
--- a/include/asm-sh/sigcontext.h
+++ b/include/asm-sh/sigcontext.h
@@ -25,7 +25,8 @@ struct sigcontext {
 	unsigned long sc_mach;
 	unsigned long sc_macl;
 
-#if defined(__SH4__) || defined(CONFIG_CPU_SH4)
+#if defined(__SH4__) || defined(CONFIG_CPU_SH4) || \
+    defined(__SH2A__) || defined(CONFIG_CPU_SH2A)
 	/* FPU registers */
 	unsigned long sc_fpregs[16];
 	unsigned long sc_xfpregs[16];
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 8b01fc4a56af..ad3d2a636130 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -205,6 +205,8 @@ asmlinkage void name##_trap_handler(unsigned int vec, struct pt_regs *regs)
 BUILD_TRAP_HANDLER(address_error);
 BUILD_TRAP_HANDLER(debug);
 BUILD_TRAP_HANDLER(bug);
+BUILD_TRAP_HANDLER(fpu_error);
+BUILD_TRAP_HANDLER(fpu_state_restore);
 
 #define arch_align_stack(x) (x)
 
-- 
cgit v1.2.3


From eddeeb32fe303910c58c4e3c27fde4b6f1503350 Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Mon, 26 Nov 2007 21:32:40 +0900
Subject: sh: Invalidate the TLB after applying PMB mappings.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/pmb.c                     | 6 ++++++
 include/asm-sh/cpu-sh4/mmu_context.h | 2 ++
 2 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index 1d45b82f0a63..b632051d6ce5 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -27,6 +27,7 @@
 #include <asm/pgtable.h>
 #include <asm/mmu.h>
 #include <asm/io.h>
+#include <asm/mmu_context.h>
 
 #define NR_PMB_ENTRIES	16
 
@@ -329,6 +330,11 @@ static int __init pmb_init(void)
 	/* PMB.SE and UB[7] */
 	ctrl_outl((1 << 31) | (1 << 7), PMB_PASCR);
 
+	/* Flush out the TLB */
+	i =  ctrl_inl(MMUCR);
+	i |= MMUCR_TI;
+	ctrl_outl(i, MMUCR);
+
 	back_to_P1();
 
 	return 0;
diff --git a/include/asm-sh/cpu-sh4/mmu_context.h b/include/asm-sh/cpu-sh4/mmu_context.h
index 979acddc0f8e..fdd56e3e3a3b 100644
--- a/include/asm-sh/cpu-sh4/mmu_context.h
+++ b/include/asm-sh/cpu-sh4/mmu_context.h
@@ -22,6 +22,8 @@
 #define MMU_UTLB_ADDRESS_ARRAY	0xF6000000
 #define MMU_PAGE_ASSOC_BIT	0x80
 
+#define MMUCR_TI		(1<<2)
+
 #ifdef CONFIG_X2TLB
 #define MMUCR_ME		(1 << 7)
 #else
-- 
cgit v1.2.3


From 66d485b45a5493f6a2ca067c6f472e7b2ca342c2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 27 Nov 2007 15:57:30 +0900
Subject: sh: Bump up ARCH_KMALLOC_MINALIGN for DMA cases.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/page.h | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 407bf5a14936..bff635a078c8 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -138,22 +138,18 @@ typedef struct { unsigned long pgd; } pgd_t;
 #endif
 
 /*
- * Slub defaults to 8-byte alignment, we're only interested in 4.
- * Slab defaults to BYTES_PER_WORD, which ends up being the same anyways.
+ * Some drivers need to perform DMA into kmalloc'ed buffers
+ * and so we have to increase the kmalloc minalign for this.
  */
-#ifdef CONFIG_SUPERH32
-#define ARCH_KMALLOC_MINALIGN	4
-#define ARCH_SLAB_MINALIGN	4
-#else
-/* If gcc inlines memset, it will use st.q instructions.  Therefore, we need
-   kmalloc allocations to be 8-byte aligned.  Without this, the alignment
-   becomes BYTE_PER_WORD i.e. only 4 (since sizeof(long)==sizeof(void*)==4 on
-   sh64 at the moment). */
-#define ARCH_KMALLOC_MINALIGN	8
+#define ARCH_KMALLOC_MINALIGN	L1_CACHE_BYTES
 
+#ifdef CONFIG_SUPERH64
 /*
- * We want 8-byte alignment for the slab caches as well, otherwise we have
- * the same BYTES_PER_WORD (sizeof(void *)) min align in kmem_cache_create().
+ * While BYTES_PER_WORD == 4 on the current sh64 ABI, GCC will still
+ * happily generate {ld/st}.q pairs, requiring us to have 8-byte
+ * alignment to avoid traps. The kmalloc alignment is gauranteed by
+ * virtue of L1_CACHE_BYTES, requiring this to only be special cased
+ * for slab caches.
  */
 #define ARCH_SLAB_MINALIGN	8
 #endif
-- 
cgit v1.2.3


From 3ee7702903c346fc814bd7540ba37eebef75054d Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 28 Nov 2007 15:56:27 +0900
Subject: sh: CCR1->CCR renaming for SH-2 parts.

Avoid namespace collision with a CCR1 definition. The general
SH code always expects CCR anyways, so there's no point in keeping
the CCR1 naming around.

Fixes up synclink collisions:

drivers/char/pcmcia/synclink_cs.c:283:1: warning: "CCR1" redefined
In file included from include/asm/cache.h:13,
                 from include/asm/processor_32.h:15,
                 from include/asm/processor.h:60,
                 from include/linux/prefetch.h:14,
                 from include/linux/list.h:8,
                 from include/linux/module.h:9,
                 from drivers/char/pcmcia/synclink_cs.c:38:
include/asm/cpu/cache.h:21:1: warning: this is the location of the previous definition

Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh2/cache.h  | 3 +--
 include/asm-sh/cpu-sh2a/cache.h | 6 +-----
 2 files changed, 2 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh2/cache.h b/include/asm-sh/cpu-sh2/cache.h
index 66388ce16c30..4e0b16500686 100644
--- a/include/asm-sh/cpu-sh2/cache.h
+++ b/include/asm-sh/cpu-sh2/cache.h
@@ -18,8 +18,7 @@
 #define SH_CACHE_ASSOC		8
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7619)
-#define CCR1		0xffffffec
-#define CCR		CCR1
+#define CCR		0xffffffec
 
 #define CCR_CACHE_CE	0x01	/* Cache enable */
 #define CCR_CACHE_WT	0x06    /* CCR[bit1=1,bit2=1] */
diff --git a/include/asm-sh/cpu-sh2a/cache.h b/include/asm-sh/cpu-sh2a/cache.h
index d88774169b58..afe228b3f493 100644
--- a/include/asm-sh/cpu-sh2a/cache.h
+++ b/include/asm-sh/cpu-sh2a/cache.h
@@ -17,12 +17,9 @@
 #define SH_CACHE_COMBINED	4
 #define SH_CACHE_ASSOC		8
 
-#define CCR1		0xfffc1000
+#define CCR		0xfffc1000 /* CCR1 */
 #define CCR2		0xfffc1004
 
-/* CCR1 behaves more like the traditional CCR */
-#define CCR		CCR1
-
 /*
  * Most of the SH-2A CCR1 definitions resemble the SH-4 ones. All others not
  * listed here are reserved.
@@ -41,4 +38,3 @@
 #define CCR_CACHE_INVALIDATE	(CCR_CACHE_OCI | CCR_CACHE_ICI)
 
 #endif /* __ASM_CPU_SH2A_CACHE_H */
-
-- 
cgit v1.2.3


From c019fd8839503a91c556ae68d773e3bbb9b476e1 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 28 Nov 2007 19:14:23 +0900
Subject: rtc: rtc-sh: Split out the CPU defs to asm/cpu/.

With all of the different CPU types this was getting a but unwieldly.
Since sh64 is now integrated, we don't have to worry about multiple
architectures caring about the header definitions.

Split out the defs for each asm/cpu/ to make rtc-sh slightly less
visually offensive.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/rtc/rtc-sh.c          | 20 +-------------------
 include/asm-sh/cpu-sh2/rtc.h  |  8 ++++++++
 include/asm-sh/cpu-sh2a/rtc.h |  8 ++++++++
 include/asm-sh/cpu-sh3/rtc.h  |  8 ++++++++
 include/asm-sh/cpu-sh4/rtc.h  |  8 ++++++++
 include/asm-sh/cpu-sh5/rtc.h  |  8 ++++++++
 include/asm-sh/rtc.h          |  2 ++
 7 files changed, 43 insertions(+), 19 deletions(-)
 create mode 100644 include/asm-sh/cpu-sh2/rtc.h
 create mode 100644 include/asm-sh/cpu-sh2a/rtc.h
 create mode 100644 include/asm-sh/cpu-sh3/rtc.h
 create mode 100644 include/asm-sh/cpu-sh4/rtc.h
 create mode 100644 include/asm-sh/cpu-sh5/rtc.h

(limited to 'include')

diff --git a/drivers/rtc/rtc-sh.c b/drivers/rtc/rtc-sh.c
index af9bc57c8920..c1d6a1880ccf 100644
--- a/drivers/rtc/rtc-sh.c
+++ b/drivers/rtc/rtc-sh.c
@@ -26,25 +26,7 @@
 #include <asm/rtc.h>
 
 #define DRV_NAME	"sh-rtc"
-#define DRV_VERSION	"0.1.5"
-
-#ifdef CONFIG_CPU_SH2A
-#define rtc_reg_size		sizeof(u16)
-#define RTC_BIT_INVERTED	0
-#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
-#elif defined(CONFIG_CPU_SH3)
-#define rtc_reg_size		sizeof(u16)
-#define RTC_BIT_INVERTED	0	/* No bug on SH7708, SH7709A */
-#define RTC_DEF_CAPABILITIES	0UL
-#elif defined(CONFIG_CPU_SH4)
-#define rtc_reg_size		sizeof(u32)
-#define RTC_BIT_INVERTED	0x40	/* bug on SH7750, SH7750S */
-#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
-#elif defined(CONFIG_CPU_SH5)
-#define rtc_reg_size		sizeof(u32)
-#define RTC_BIT_INVERTED	0	/* The SH-5 RTC is surprisingly sane! */
-#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
-#endif
+#define DRV_VERSION	"0.1.6"
 
 #define RTC_REG(r)	((r) * rtc_reg_size)
 
diff --git a/include/asm-sh/cpu-sh2/rtc.h b/include/asm-sh/cpu-sh2/rtc.h
new file mode 100644
index 000000000000..39e2d6e94782
--- /dev/null
+++ b/include/asm-sh/cpu-sh2/rtc.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_CPU_SH2_RTC_H
+#define __ASM_SH_CPU_SH2_RTC_H
+
+#define rtc_reg_size		sizeof(u16)
+#define RTC_BIT_INVERTED	0
+#define RTC_DEF_CAPABILITIES	0UL
+
+#endif /* __ASM_SH_CPU_SH2_RTC_H */
diff --git a/include/asm-sh/cpu-sh2a/rtc.h b/include/asm-sh/cpu-sh2a/rtc.h
new file mode 100644
index 000000000000..afb511e2bed7
--- /dev/null
+++ b/include/asm-sh/cpu-sh2a/rtc.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_CPU_SH2A_RTC_H
+#define __ASM_SH_CPU_SH2A_RTC_H
+
+#define rtc_reg_size		sizeof(u16)
+#define RTC_BIT_INVERTED	0
+#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
+
+#endif /* __ASM_SH_CPU_SH2A_RTC_H */
diff --git a/include/asm-sh/cpu-sh3/rtc.h b/include/asm-sh/cpu-sh3/rtc.h
new file mode 100644
index 000000000000..319404aaee37
--- /dev/null
+++ b/include/asm-sh/cpu-sh3/rtc.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_CPU_SH3_RTC_H
+#define __ASM_SH_CPU_SH3_RTC_H
+
+#define rtc_reg_size		sizeof(u16)
+#define RTC_BIT_INVERTED	0	/* No bug on SH7708, SH7709A */
+#define RTC_DEF_CAPABILITIES	0UL
+
+#endif /* __ASM_SH_CPU_SH3_RTC_H */
diff --git a/include/asm-sh/cpu-sh4/rtc.h b/include/asm-sh/cpu-sh4/rtc.h
new file mode 100644
index 000000000000..f3d0f53275e4
--- /dev/null
+++ b/include/asm-sh/cpu-sh4/rtc.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_CPU_SH4_RTC_H
+#define __ASM_SH_CPU_SH4_RTC_H
+
+#define rtc_reg_size		sizeof(u32)
+#define RTC_BIT_INVERTED	0x40	/* bug on SH7750, SH7750S */
+#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
+
+#endif /* __ASM_SH_CPU_SH4_RTC_H */
diff --git a/include/asm-sh/cpu-sh5/rtc.h b/include/asm-sh/cpu-sh5/rtc.h
new file mode 100644
index 000000000000..12ea0ed144e1
--- /dev/null
+++ b/include/asm-sh/cpu-sh5/rtc.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_CPU_SH5_RTC_H
+#define __ASM_SH_CPU_SH5_RTC_H
+
+#define rtc_reg_size		sizeof(u32)
+#define RTC_BIT_INVERTED	0	/* The SH-5 RTC is surprisingly sane! */
+#define RTC_DEF_CAPABILITIES	RTC_CAP_4_DIGIT_YEAR
+
+#endif /* __ASM_SH_CPU_SH5_RTC_H */
diff --git a/include/asm-sh/rtc.h b/include/asm-sh/rtc.h
index 858da99d37e0..ec45ba8e11d9 100644
--- a/include/asm-sh/rtc.h
+++ b/include/asm-sh/rtc.h
@@ -11,4 +11,6 @@ struct sh_rtc_platform_info {
 	unsigned long capabilities;
 };
 
+#include <asm/cpu/rtc.h>
+
 #endif /* _ASM_RTC_H */
-- 
cgit v1.2.3


From a23ba43573a24c351640bc19c06c701798fe6e25 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 28 Nov 2007 20:19:38 +0900
Subject: sh: comment tidying for sh64->sh migration.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/cayman/led.c         |  2 +-
 arch/sh/boot/compressed/Makefile_64 | 10 +++++-----
 arch/sh/boot/compressed/misc_64.c   |  2 +-
 arch/sh/kernel/cpu/irq/intc-sh5.c   | 10 +++++-----
 arch/sh/kernel/cpu/sh5/entry.S      | 13 ++++++-------
 arch/sh/kernel/cpu/sh5/fpu.c        | 10 ++++------
 arch/sh/kernel/cpu/sh5/switchto.S   |  2 +-
 arch/sh/kernel/cpu/sh5/unwind.c     |  2 +-
 arch/sh/kernel/head_64.S            | 19 ++++---------------
 arch/sh/kernel/process_64.c         | 29 ++++++++++++++++-------------
 arch/sh/kernel/sh_ksyms_64.c        |  9 ++++-----
 arch/sh/kernel/signal_64.c          | 11 ++++-------
 arch/sh/kernel/syscalls_64.S        |  2 +-
 arch/sh/kernel/time_64.c            | 10 +++++-----
 arch/sh/kernel/traps_64.c           | 14 ++++----------
 arch/sh/lib64/c-checksum.c          |  5 +----
 arch/sh/lib64/dbg.c                 |  2 +-
 arch/sh/lib64/udelay.c              |  2 +-
 arch/sh/mm/cache-sh5.c              | 15 ++++++---------
 arch/sh/mm/extable_64.c             | 28 +++++++++++++++-------------
 arch/sh/mm/tlb-sh5.c                |  4 +---
 include/asm-sh/checksum_64.h        | 16 +++++++---------
 include/asm-sh/cpu-sh5/cache.h      | 15 +++++++--------
 include/asm-sh/cpu-sh5/cacheflush.h |  6 +++---
 include/asm-sh/cpu-sh5/irq.h        | 15 +++++++--------
 include/asm-sh/cpu-sh5/registers.h  | 16 ++++++++--------
 include/asm-sh/pgtable_64.h         | 19 +++++++++----------
 include/asm-sh/string_64.h          | 18 +++++++-----------
 include/asm-sh/uaccess_64.h         | 16 +++++++---------
 include/asm-sh/unistd_64.h          | 18 ++++++++----------
 30 files changed, 150 insertions(+), 190 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/cayman/led.c b/arch/sh/boards/cayman/led.c
index b4e122fd9502..a808eac4ecd6 100644
--- a/arch/sh/boards/cayman/led.c
+++ b/arch/sh/boards/cayman/led.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/mach-cayman/led.c
+ * arch/sh/boards/cayman/led.c
  *
  * Copyright (C) 2002 Stuart Menefy <stuart.menefy@st.com>
  *
diff --git a/arch/sh/boot/compressed/Makefile_64 b/arch/sh/boot/compressed/Makefile_64
index 0b48ae99922b..4334f2b86d8f 100644
--- a/arch/sh/boot/compressed/Makefile_64
+++ b/arch/sh/boot/compressed/Makefile_64
@@ -1,14 +1,14 @@
 #
-# linux/arch/sh64/boot/compressed/Makefile
+# arch/sh/boot/compressed/Makefile_64
 #
-# This file is subject to the terms and conditions of the GNU General Public
-# License.  See the file "COPYING" in the main directory of this archive
-# for more details.
+# create a compressed vmlinux image from the original vmlinux
 #
 # Copyright (C) 2002 Stuart Menefy
 # Copyright (C) 2004 Paul Mundt
 #
-# create a compressed vmlinux image from the original vmlinux
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
 #
 
 targets		:= vmlinux vmlinux.bin vmlinux.bin.gz \
diff --git a/arch/sh/boot/compressed/misc_64.c b/arch/sh/boot/compressed/misc_64.c
index aea00c53ce29..a006ef89b9dd 100644
--- a/arch/sh/boot/compressed/misc_64.c
+++ b/arch/sh/boot/compressed/misc_64.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/boot/compressed/misc.c
+ * arch/sh/boot/compressed/misc_64.c
  *
  * This is a collection of several routines from gzip-1.0.3
  * adapted for Linux.
diff --git a/arch/sh/kernel/cpu/irq/intc-sh5.c b/arch/sh/kernel/cpu/irq/intc-sh5.c
index 49b845a31fff..43ee7a9a4f0b 100644
--- a/arch/sh/kernel/cpu/irq/intc-sh5.c
+++ b/arch/sh/kernel/cpu/irq/intc-sh5.c
@@ -1,18 +1,18 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
+ * arch/sh/kernel/cpu/irq/intc-sh5.c
  *
- * arch/sh64/kernel/irq_intc.c
+ * Interrupt Controller support for SH5 INTC.
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003  Paul Mundt
  *
- * Interrupt Controller support for SH5 INTC.
  * Per-interrupt selective. IRLM=0 (Fixed priority) is not
  * supported being useless without a cascaded interrupt
  * controller.
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/init.h>
 #include <linux/interrupt.h>
diff --git a/arch/sh/kernel/cpu/sh5/entry.S b/arch/sh/kernel/cpu/sh5/entry.S
index dbf2d7684502..ba8750176d91 100644
--- a/arch/sh/kernel/cpu/sh5/entry.S
+++ b/arch/sh/kernel/cpu/sh5/entry.S
@@ -1,14 +1,13 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/entry.S
+ * arch/sh/kernel/cpu/sh5/entry.S
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
- * Copyright (C) 2004, 2005  Paul Mundt
- * Copyright (C) 2003, 2004 Richard Curnow
+ * Copyright (C) 2004 - 2007  Paul Mundt
+ * Copyright (C) 2003, 2004  Richard Curnow
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/errno.h>
 #include <linux/sys.h>
diff --git a/arch/sh/kernel/cpu/sh5/fpu.c b/arch/sh/kernel/cpu/sh5/fpu.c
index 5a391b1a00e2..30b76a94abf2 100644
--- a/arch/sh/kernel/cpu/sh5/fpu.c
+++ b/arch/sh/kernel/cpu/sh5/fpu.c
@@ -1,9 +1,5 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/fpu.c
+ * arch/sh/kernel/cpu/sh5/fpu.c
  *
  * Copyright (C) 2001  Manuela Cirronis, Paolo Alberelli
  * Copyright (C) 2002  STMicroelectronics Limited
@@ -12,8 +8,10 @@
  * Started from SH4 version:
  *   Copyright (C) 1999, 2000  Kaz Kojima & Niibe Yutaka
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/sched.h>
 #include <linux/signal.h>
 #include <asm/processor.h>
diff --git a/arch/sh/kernel/cpu/sh5/switchto.S b/arch/sh/kernel/cpu/sh5/switchto.S
index 45b2d90eed7d..45c351b0f1ba 100644
--- a/arch/sh/kernel/cpu/sh5/switchto.S
+++ b/arch/sh/kernel/cpu/sh5/switchto.S
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/kernel/switchto.S
+ * arch/sh/kernel/cpu/sh5/switchto.S
  *
  * sh64 context switch
  *
diff --git a/arch/sh/kernel/cpu/sh5/unwind.c b/arch/sh/kernel/cpu/sh5/unwind.c
index 1214c78e3584..119c20afd4e5 100644
--- a/arch/sh/kernel/cpu/sh5/unwind.c
+++ b/arch/sh/kernel/cpu/sh5/unwind.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/kernel/unwind.c
+ * arch/sh/kernel/cpu/sh5/unwind.c
  *
  * Copyright (C) 2004  Paul Mundt
  * Copyright (C) 2004  Richard Curnow
diff --git a/arch/sh/kernel/head_64.S b/arch/sh/kernel/head_64.S
index d3479287b10a..f42d4c0feb76 100644
--- a/arch/sh/kernel/head_64.S
+++ b/arch/sh/kernel/head_64.S
@@ -1,23 +1,12 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/head.S
+ * arch/sh/kernel/head_64.S
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003, 2004  Paul Mundt
  *
- *
- * benedict.gaster@superh.com:	 2nd May 2002
- *    Moved definition of empty_zero_page to its own section allowing
- *    it to be placed at an absolute address known at load time.
- *
- * lethal@linux-sh.org:          9th May 2003
- *    Kill off GLOBAL_NAME() usage.
- *
- * lethal@linux-sh.org:          8th May 2004
- *    Add early SCIF console DTLB mapping.
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <asm/page.h>
 #include <asm/cache.h>
diff --git a/arch/sh/kernel/process_64.c b/arch/sh/kernel/process_64.c
index 973dd1a3d292..cff3b7dc9c56 100644
--- a/arch/sh/kernel/process_64.c
+++ b/arch/sh/kernel/process_64.c
@@ -421,19 +421,22 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
  */
 void exit_thread(void)
 {
-	/* See arch/sparc/kernel/process.c for the precedent for doing this -- RPC.
-
-	   The SH-5 FPU save/restore approach relies on last_task_used_math
-	   pointing to a live task_struct.  When another task tries to use the
-	   FPU for the 1st time, the FPUDIS trap handling (see
-	   arch/sh64/kernel/fpu.c) will save the existing FPU state to the
-	   FP regs field within last_task_used_math before re-loading the new
-	   task's FPU state (or initialising it if the FPU has been used
-	   before).  So if last_task_used_math is stale, and its page has already been
-	   re-allocated for another use, the consequences are rather grim. Unless we
-	   null it here, there is no other path through which it would get safely
-	   nulled. */
-
+	/*
+	 * See arch/sparc/kernel/process.c for the precedent for doing
+	 * this -- RPC.
+	 *
+	 * The SH-5 FPU save/restore approach relies on
+	 * last_task_used_math pointing to a live task_struct.  When
+	 * another task tries to use the FPU for the 1st time, the FPUDIS
+	 * trap handling (see arch/sh/kernel/cpu/sh5/fpu.c) will save the
+	 * existing FPU state to the FP regs field within
+	 * last_task_used_math before re-loading the new task's FPU state
+	 * (or initialising it if the FPU has been used before).  So if
+	 * last_task_used_math is stale, and its page has already been
+	 * re-allocated for another use, the consequences are rather
+	 * grim. Unless we null it here, there is no other path through
+	 * which it would get safely nulled.
+	 */
 #ifdef CONFIG_SH_FPU
 	if (last_task_used_math == current) {
 		last_task_used_math = NULL;
diff --git a/arch/sh/kernel/sh_ksyms_64.c b/arch/sh/kernel/sh_ksyms_64.c
index fbc991c33f5a..8004c38d3d37 100644
--- a/arch/sh/kernel/sh_ksyms_64.c
+++ b/arch/sh/kernel/sh_ksyms_64.c
@@ -1,12 +1,11 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/sh_ksyms.c
+ * arch/sh/kernel/sh_ksyms_64.c
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/rwsem.h>
 #include <linux/module.h>
diff --git a/arch/sh/kernel/signal_64.c b/arch/sh/kernel/signal_64.c
index 922891960c33..80bde19d445b 100644
--- a/arch/sh/kernel/signal_64.c
+++ b/arch/sh/kernel/signal_64.c
@@ -1,16 +1,13 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/signal.c
+ * arch/sh/kernel/signal_64.c
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003  Paul Mundt
  * Copyright (C) 2004  Richard Curnow
  *
- * Started from sh version.
- *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/rwsem.h>
 #include <linux/sched.h>
diff --git a/arch/sh/kernel/syscalls_64.S b/arch/sh/kernel/syscalls_64.S
index abb94c05d07a..98a93efe3691 100644
--- a/arch/sh/kernel/syscalls_64.S
+++ b/arch/sh/kernel/syscalls_64.S
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/kernel/syscalls.S
+ * arch/sh/kernel/syscalls_64.S
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2004 - 2007  Paul Mundt
diff --git a/arch/sh/kernel/time_64.c b/arch/sh/kernel/time_64.c
index 81003af8115a..f819ba38a6ce 100644
--- a/arch/sh/kernel/time_64.c
+++ b/arch/sh/kernel/time_64.c
@@ -1,9 +1,5 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/time.c
+ * arch/sh/kernel/time_64.c
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003 - 2007  Paul Mundt
@@ -13,6 +9,10 @@
  *    Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
  *      Some code taken from i386 version.
  *      Copyright (C) 1991, 1992, 1995  Linus Torvalds
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/errno.h>
 #include <linux/rwsem.h>
diff --git a/arch/sh/kernel/traps_64.c b/arch/sh/kernel/traps_64.c
index b8020f26b635..c0b3c6f6edb5 100644
--- a/arch/sh/kernel/traps_64.c
+++ b/arch/sh/kernel/traps_64.c
@@ -1,19 +1,13 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/kernel/traps.c
+ * arch/sh/kernel/traps_64.c
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003, 2004  Paul Mundt
  * Copyright (C) 2003, 2004  Richard Curnow
  *
- */
-
-/*
- * 'Traps.c' handles hardware traps and faults after we have saved some
- * state in 'entry.S'.
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/sched.h>
 #include <linux/kernel.h>
diff --git a/arch/sh/lib64/c-checksum.c b/arch/sh/lib64/c-checksum.c
index 053137abd8a0..5dfbd8b5e558 100644
--- a/arch/sh/lib64/c-checksum.c
+++ b/arch/sh/lib64/c-checksum.c
@@ -1,12 +1,9 @@
 /*
- * arch/sh64/lib/c-checksum.c
+ * arch/sh/lib64/c-checksum.c
  *
  * This file contains network checksum routines that are better done
  * in an architecture-specific manner due to speed..
  */
-
-#undef DEBUG
-
 #include <linux/string.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
diff --git a/arch/sh/lib64/dbg.c b/arch/sh/lib64/dbg.c
index 97816e0baf19..75825ef6e084 100644
--- a/arch/sh/lib64/dbg.c
+++ b/arch/sh/lib64/dbg.c
@@ -2,7 +2,7 @@
 --
 -- Identity : Linux50 Debug Funcions
 --
--- File     : arch/sh64/lib/dbg.C
+-- File     : arch/sh/lib64/dbg.c
 --
 -- Copyright 2000, 2001 STMicroelectronics Limited.
 -- Copyright 2004 Richard Curnow (evt_debug etc)
diff --git a/arch/sh/lib64/udelay.c b/arch/sh/lib64/udelay.c
index 4c71a7428e60..23c7d17fb9f7 100644
--- a/arch/sh/lib64/udelay.c
+++ b/arch/sh/lib64/udelay.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/lib/udelay.c
+ * arch/sh/lib64/udelay.c
  *
  * Delay routines, using a pre-computed "loops_per_jiffy" value.
  *
diff --git a/arch/sh/mm/cache-sh5.c b/arch/sh/mm/cache-sh5.c
index 421487cfff4c..4617e3aeee73 100644
--- a/arch/sh/mm/cache-sh5.c
+++ b/arch/sh/mm/cache-sh5.c
@@ -1,18 +1,15 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/mm/cache.c
+ * arch/sh/mm/cache-sh5.c
  *
  * Original version Copyright (C) 2000, 2001  Paolo Alberelli
  * Second version Copyright (C) benedict.gaster@superh.com 2002
  * Third version Copyright Richard.Curnow@superh.com 2003
  * Hacks to third version Copyright (C) 2003 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
-/****************************************************************************/
-
 #include <linux/init.h>
 #include <linux/mman.h>
 #include <linux/mm.h>
@@ -146,7 +143,7 @@ int __init sh64_cache_init(void)
 /* The following group of functions deal with mapping and unmapping a temporary
    page into the DTLB slot that have been set aside for our exclusive use. */
 /* In order to accomplish this, we use the generic interface for adding and
-   removing a wired slot entry as defined in arch/sh64/mm/tlb.c */
+   removing a wired slot entry as defined in arch/sh/mm/tlb-sh5.c */
 /****************************************************************************/
 
 static unsigned long slot_own_flags;
diff --git a/arch/sh/mm/extable_64.c b/arch/sh/mm/extable_64.c
index a2e6e0563772..f05499688d88 100644
--- a/arch/sh/mm/extable_64.c
+++ b/arch/sh/mm/extable_64.c
@@ -1,14 +1,14 @@
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * arch/sh64/mm/extable.c
+ * arch/sh/mm/extable_64.c
  *
  * Copyright (C) 2003 Richard Curnow
  * Copyright (C) 2003, 2004  Paul Mundt
  *
  * Cloned from the 2.5 SH version..
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 #include <linux/rwsem.h>
 #include <linux/module.h>
@@ -21,13 +21,16 @@ static const struct exception_table_entry __copy_user_fixup_ex = {
 	.fixup = (unsigned long)&__copy_user_fixup,
 };
 
-/* Some functions that may trap due to a bad user-mode address have too many loads
-   and stores in them to make it at all practical to label each one and put them all in
-   the main exception table.
-
-   In particular, the fast memcpy routine is like this.  It's fix-up is just to fall back
-   to a slow byte-at-a-time copy, which is handled the conventional way.  So it's functionally
-   OK to just handle any trap occurring in the fast memcpy with that fixup. */
+/*
+ * Some functions that may trap due to a bad user-mode address have too
+ * many loads and stores in them to make it at all practical to label
+ * each one and put them all in the main exception table.
+ *
+ * In particular, the fast memcpy routine is like this.  It's fix-up is
+ * just to fall back to a slow byte-at-a-time copy, which is handled the
+ * conventional way.  So it's functionally OK to just handle any trap
+ * occurring in the fast memcpy with that fixup.
+ */
 static const struct exception_table_entry *check_exception_ranges(unsigned long addr)
 {
 	if ((addr >= (unsigned long)&copy_user_memcpy) &&
@@ -77,4 +80,3 @@ int fixup_exception(struct pt_regs *regs)
 
 	return 0;
 }
-
diff --git a/arch/sh/mm/tlb-sh5.c b/arch/sh/mm/tlb-sh5.c
index d517e7d70340..f34274a1ded3 100644
--- a/arch/sh/mm/tlb-sh5.c
+++ b/arch/sh/mm/tlb-sh5.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh64/mm/tlb.c
+ * arch/sh/mm/tlb-sh5.c
  *
  * Copyright (C) 2003  Paul Mundt <lethal@linux-sh.org>
  * Copyright (C) 2003  Richard Curnow <richard.curnow@superh.com>
@@ -7,7 +7,6 @@
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
- *
  */
 #include <linux/mm.h>
 #include <linux/init.h>
@@ -163,4 +162,3 @@ inline void sh64_setup_tlb_slot(unsigned long long config_addr,
  */
 inline void sh64_teardown_tlb_slot(unsigned long long config_addr)
 	__attribute__ ((alias("__flush_tlb_slot")));
-
diff --git a/include/asm-sh/checksum_64.h b/include/asm-sh/checksum_64.h
index ec79b92e6bdc..9c62a031a8f5 100644
--- a/include/asm-sh/checksum_64.h
+++ b/include/asm-sh/checksum_64.h
@@ -1,15 +1,14 @@
-#ifndef __ASM_SH64_CHECKSUM_H
-#define __ASM_SH64_CHECKSUM_H
+#ifndef __ASM_SH_CHECKSUM_64_H
+#define __ASM_SH_CHECKSUM_64_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/checksum.h
+ * include/asm-sh/checksum_64.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 
 /*
@@ -76,5 +75,4 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 	return csum_fold(csum_partial(buff, len, 0));
 }
 
-#endif /* __ASM_SH64_CHECKSUM_H */
-
+#endif /* __ASM_SH_CHECKSUM_64_H */
diff --git a/include/asm-sh/cpu-sh5/cache.h b/include/asm-sh/cpu-sh5/cache.h
index 7eaa8894649f..ed050ab526f2 100644
--- a/include/asm-sh/cpu-sh5/cache.h
+++ b/include/asm-sh/cpu-sh5/cache.h
@@ -1,16 +1,15 @@
-#ifndef __ASM_SH64_CACHE_H
-#define __ASM_SH64_CACHE_H
+#ifndef __ASM_SH_CPU_SH5_CACHE_H
+#define __ASM_SH_CPU_SH5_CACHE_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/cache.h
+ * include/asm-sh/cpu-sh5/cache.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003, 2004  Paul Mundt
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 
 #define L1_CACHE_SHIFT		5
@@ -95,4 +94,4 @@
  * No Instruction Cache defines required, then.
  */
 
-#endif /* __ASM_SH64_CACHE_H */
+#endif /* __ASM_SH_CPU_SH5_CACHE_H */
diff --git a/include/asm-sh/cpu-sh5/cacheflush.h b/include/asm-sh/cpu-sh5/cacheflush.h
index 45d12d6f89c2..98edb5b1da32 100644
--- a/include/asm-sh/cpu-sh5/cacheflush.h
+++ b/include/asm-sh/cpu-sh5/cacheflush.h
@@ -1,5 +1,5 @@
-#ifndef __ASM_SH64_CACHEFLUSH_H
-#define __ASM_SH64_CACHEFLUSH_H
+#ifndef __ASM_SH_CPU_SH5_CACHEFLUSH_H
+#define __ASM_SH_CPU_SH5_CACHEFLUSH_H
 
 #ifndef __ASSEMBLY__
 
@@ -31,5 +31,5 @@ extern void flush_icache_user_range(struct vm_area_struct *vma,
 
 #endif /* __ASSEMBLY__ */
 
-#endif /* __ASM_SH64_CACHEFLUSH_H */
+#endif /* __ASM_SH_CPU_SH5_CACHEFLUSH_H */
 
diff --git a/include/asm-sh/cpu-sh5/irq.h b/include/asm-sh/cpu-sh5/irq.h
index f539baec8529..f0f0756e6e84 100644
--- a/include/asm-sh/cpu-sh5/irq.h
+++ b/include/asm-sh/cpu-sh5/irq.h
@@ -1,15 +1,14 @@
-#ifndef __ASM_SH64_IRQ_H
-#define __ASM_SH64_IRQ_H
+#ifndef __ASM_SH_CPU_SH5_IRQ_H
+#define __ASM_SH_CPU_SH5_IRQ_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/irq.h
+ * include/asm-sh/cpu-sh5/irq.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 
 
@@ -115,4 +114,4 @@ extern int intc_evt_to_irq[(0xE20/0x20)+1];
 int intc_irq_describe(char* p, int irq);
 extern int platform_int_priority[NR_INTC_IRQS];
 
-#endif /* __ASM_SH64_IRQ_H */
+#endif /* __ASM_SH_CPU_SH5_IRQ_H */
diff --git a/include/asm-sh/cpu-sh5/registers.h b/include/asm-sh/cpu-sh5/registers.h
index 7eec666acf84..6664ea6f1566 100644
--- a/include/asm-sh/cpu-sh5/registers.h
+++ b/include/asm-sh/cpu-sh5/registers.h
@@ -1,15 +1,15 @@
-#ifndef __ASM_SH64_REGISTERS_H
-#define __ASM_SH64_REGISTERS_H
+#ifndef __ASM_SH_CPU_SH5_REGISTERS_H
+#define __ASM_SH_CPU_SH5_REGISTERS_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/registers.h
+ * include/asm-sh/cpu-sh5/registers.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2004  Richard Curnow
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 
 #ifdef __ASSEMBLY__
@@ -103,4 +103,4 @@
 #define __USR __str(USR)
 
 #endif /* __ASSEMBLY__ */
-#endif /* __ASM_SH64_REGISTERS_H */
+#endif /* __ASM_SH_CPU_SH5_REGISTERS_H */
diff --git a/include/asm-sh/pgtable_64.h b/include/asm-sh/pgtable_64.h
index d422111006f4..972211671c9a 100644
--- a/include/asm-sh/pgtable_64.h
+++ b/include/asm-sh/pgtable_64.h
@@ -1,21 +1,20 @@
-#ifndef __ASM_SH64_PGTABLE_H
-#define __ASM_SH64_PGTABLE_H
+#ifndef __ASM_SH_PGTABLE_64_H
+#define __ASM_SH_PGTABLE_64_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
+ * include/asm-sh/pgtable_64.h
  *
- * include/asm-sh64/pgtable.h
+ * This file contains the functions and defines necessary to modify and use
+ * the SuperH page table tree.
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003, 2004  Paul Mundt
  * Copyright (C) 2003, 2004  Richard Curnow
  *
- * This file contains the functions and defines necessary to modify and use
- * the SuperH page table tree.
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/threads.h>
 #include <asm/processor.h>
 #include <asm/page.h>
@@ -297,4 +296,4 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
 #define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
 
-#endif /* __ASM_SH64_PGTABLE_H */
+#endif /* __ASM_SH_PGTABLE_64_H */
diff --git a/include/asm-sh/string_64.h b/include/asm-sh/string_64.h
index 8a7357366ce8..aa1fef229c78 100644
--- a/include/asm-sh/string_64.h
+++ b/include/asm-sh/string_64.h
@@ -1,21 +1,17 @@
-#ifndef __ASM_SH64_STRING_H
-#define __ASM_SH64_STRING_H
+#ifndef __ASM_SH_STRING_64_H
+#define __ASM_SH_STRING_64_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/string.h
+ * include/asm-sh/string_64.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  *
- * Empty on purpose. ARCH SH64 ASM libs are out of the current project scope.
- *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
 
 #define __HAVE_ARCH_MEMCPY
-
 extern void *memcpy(void *dest, const void *src, size_t count);
 
-#endif
+#endif /* __ASM_SH_STRING_64_H */
diff --git a/include/asm-sh/uaccess_64.h b/include/asm-sh/uaccess_64.h
index 24800a8045ce..d54ec082d25a 100644
--- a/include/asm-sh/uaccess_64.h
+++ b/include/asm-sh/uaccess_64.h
@@ -1,12 +1,8 @@
-#ifndef __ASM_SH64_UACCESS_H
-#define __ASM_SH64_UACCESS_H
+#ifndef __ASM_SH_UACCESS_64_H
+#define __ASM_SH_UACCESS_64_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/uaccess.h
+ * include/asm-sh/uaccess_64.h
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003, 2004  Paul Mundt
@@ -20,8 +16,10 @@
  *              Copyright (C) 1996, 1997, 1998 by Ralf Baechle
  *     and i386 version.
  *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/errno.h>
 #include <linux/sched.h>
 
@@ -301,4 +299,4 @@ struct exception_table_entry
 extern unsigned long search_exception_table(unsigned long addr);
 extern const struct exception_table_entry *search_exception_tables (unsigned long addr);
 
-#endif /* __ASM_SH64_UACCESS_H */
+#endif /* __ASM_SH_UACCESS_64_H */
diff --git a/include/asm-sh/unistd_64.h b/include/asm-sh/unistd_64.h
index 1a5197f369b2..944511882cac 100644
--- a/include/asm-sh/unistd_64.h
+++ b/include/asm-sh/unistd_64.h
@@ -1,21 +1,19 @@
-#ifndef __ASM_SH64_UNISTD_H
-#define __ASM_SH64_UNISTD_H
+#ifndef __ASM_SH_UNISTD_64_H
+#define __ASM_SH_UNISTD_64_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
+ * include/asm-sh/unistd_64.h
  *
- * include/asm-sh64/unistd.h
+ * This file contains the system call numbers.
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003 - 2007 Paul Mundt
  * Copyright (C) 2004  Sean McGoogan
  *
- * This file contains the system call numbers.
- *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #define __NR_restart_syscall	  0
 #define __NR_exit		  1
 #define __NR_fork		  2
@@ -414,4 +412,4 @@
 #endif
 
 #endif /* __KERNEL__ */
-#endif /* __ASM_SH64_UNISTD_H */
+#endif /* __ASM_SH_UNISTD_64_H */
-- 
cgit v1.2.3


From 1efe4ce3ca126da77e450d5a83f7201949d76f62 Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 16:12:36 +0900
Subject: sh: GUSA atomic rollback support.

This implements kernel-level atomic rollback built on top of gUSA,
as an alternative non-IRQ based atomicity method. This is generally
a faster method for platforms that are lacking the LL/SC pairs that
SH-4A and later use, and is only supportable on legacy cores.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                |  12 ++-
 arch/sh/kernel/cpu/sh3/entry.S |  24 +++++-
 arch/sh/kernel/entry-common.S  |  19 -----
 arch/sh/kernel/process_32.c    |  19 -----
 arch/sh/kernel/signal_32.c     |  18 -----
 include/asm-sh/atomic-grb.h    | 169 +++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/atomic.h        |  10 ++-
 include/asm-sh/bitops-grb.h    | 169 +++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/bitops-irq.h    |  91 ++++++++++++++++++++++
 include/asm-sh/bitops.h        |  91 ++--------------------
 include/asm-sh/cmpxchg-grb.h   |  70 +++++++++++++++++
 include/asm-sh/cmpxchg-irq.h   |  40 ++++++++++
 include/asm-sh/system.h        |  40 ++--------
 13 files changed, 591 insertions(+), 181 deletions(-)
 create mode 100644 include/asm-sh/atomic-grb.h
 create mode 100644 include/asm-sh/bitops-grb.h
 create mode 100644 include/asm-sh/bitops-irq.h
 create mode 100644 include/asm-sh/cmpxchg-grb.h
 create mode 100644 include/asm-sh/cmpxchg-irq.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 2dc3b177193c..f645f8416f1c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -692,7 +692,7 @@ source "kernel/Kconfig.preempt"
 
 config GUSA
 	def_bool y
-	depends on !SMP
+	depends on !SMP && SUPERH32
 	help
 	  This enables support for gUSA (general UserSpace Atomicity).
 	  This is the default implementation for both UP and non-ll/sc
@@ -704,6 +704,16 @@ config GUSA
 	  This should only be disabled for special cases where alternate
 	  atomicity implementations exist.
 
+config GUSA_RB
+	bool "Implement atomic operations by roll-back (gRB) (EXPERIMENTAL)"
+	depends on GUSA && CPU_SH3 || (CPU_SH4 && !CPU_SH4A)
+	help
+	  Enabling this option will allow the kernel to implement some
+	  atomic operations using a software implemention of load-locked/
+	  store-conditional (LLSC). On machines which do not have hardware
+	  LLSC, this should be more efficient than the other alternative of
+	  disabling insterrupts around the atomic sequence.
+
 endmenu
 
 menu "Boot options"
diff --git a/arch/sh/kernel/cpu/sh3/entry.S b/arch/sh/kernel/cpu/sh3/entry.S
index 0d12a124055c..4004073f98cd 100644
--- a/arch/sh/kernel/cpu/sh3/entry.S
+++ b/arch/sh/kernel/cpu/sh3/entry.S
@@ -13,8 +13,9 @@
 #include <linux/linkage.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
-#include <asm/cpu/mmu_context.h>
 #include <asm/unistd.h>
+#include <asm/cpu/mmu_context.h>
+#include <asm/page.h>
 
 ! NOTE:
 ! GNU as (as of 2.9.1) changes bf/s into bt/s and bra, when the address
@@ -409,6 +410,27 @@ ENTRY(handle_exception)
 	! Using k0, k1 for scratch registers (r0_bank1, r1_bank),
 	! save all registers onto stack.
 	!
+
+#ifdef CONFIG_GUSA
+	! Check for roll back gRB (User and Kernel)
+	mov	r15, k0
+	shll	k0
+	bf/s	1f
+	 shll	k0
+	bf/s	1f
+	 stc	spc, k1
+	stc	r0_bank, k0
+	cmp/hs	k0, k1		! test k1 (saved PC) >= k0 (saved r0)
+	bt/s	2f
+	 stc	r1_bank, k1
+
+	add	#-2, k0
+	add	r15, k0
+	ldc	k0, spc		! PC = saved r0 + r15 - 2
+2:	mov	k1, r15		! SP = r1
+1:
+#endif
+
 	stc	ssr, k0		! Is it from kernel space?
 	shll	k0		! Check MD bit (bit30) by shifting it into...
 	shll	k0		!       ...the T bit
diff --git a/arch/sh/kernel/entry-common.S b/arch/sh/kernel/entry-common.S
index 397ac71d97f1..926b2e7b11c1 100644
--- a/arch/sh/kernel/entry-common.S
+++ b/arch/sh/kernel/entry-common.S
@@ -176,25 +176,6 @@ work_notifysig:
 	jmp	@r1
 	 lds	r0, pr
 work_resched:
-#if defined(CONFIG_GUSA) && !defined(CONFIG_PREEMPT)
-	! gUSA handling
-	mov.l	@(OFF_SP,r15), r0	! get user space stack pointer
-	mov	r0, r1
-	shll	r0
-	bf/s	1f
-	 shll	r0
-	bf/s	1f
-	 mov	#OFF_PC, r0
-	! 				  SP >= 0xc0000000 : gUSA mark
-	mov.l	@(r0,r15), r2		! get user space PC (program counter)
-	mov.l	@(OFF_R0,r15), r3	! end point
-	cmp/hs	r3, r2			! r2 >= r3? 
-	bt	1f
-	add	r3, r1			! rewind point #2
-	mov.l	r1, @(r0,r15)		! reset PC to rewind point #2
-	!
-1:
-#endif
 	mov.l	1f, r1
 	jsr	@r1				! schedule
 	 nop
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index b48324867eee..9ab1926b9d10 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -322,25 +322,6 @@ struct task_struct *__switch_to(struct task_struct *prev,
 	unlazy_fpu(prev, task_pt_regs(prev));
 #endif
 
-#if defined(CONFIG_GUSA) && defined(CONFIG_PREEMPT)
-	{
-		struct pt_regs *regs;
-
-		preempt_disable();
-		regs = task_pt_regs(prev);
-		if (user_mode(regs) && regs->regs[15] >= 0xc0000000) {
-			int offset = (int)regs->regs[15];
-
-			/* Reset stack pointer: clear critical region mark */
-			regs->regs[15] = regs->regs[1];
-			if (regs->pc < regs->regs[0])
-				/* Go to rewind point */
-				regs->pc = regs->regs[0] + offset;
-		}
-		preempt_enable_no_resched();
-	}
-#endif
-
 #ifdef CONFIG_MMU
 	/*
 	 * Restore the kernel mode register
diff --git a/arch/sh/kernel/signal_32.c b/arch/sh/kernel/signal_32.c
index ca754fd42437..f6b5fbfe75c4 100644
--- a/arch/sh/kernel/signal_32.c
+++ b/arch/sh/kernel/signal_32.c
@@ -507,24 +507,6 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 						ctrl_inw(regs->pc - 4));
 				break;
 		}
-#ifdef CONFIG_GUSA
-	} else {
-		/* gUSA handling */
-		preempt_disable();
-
-		if (regs->regs[15] >= 0xc0000000) {
-			int offset = (int)regs->regs[15];
-
-			/* Reset stack pointer: clear critical region mark */
-			regs->regs[15] = regs->regs[1];
-			if (regs->pc < regs->regs[0])
-				/* Go to rewind point #1 */
-				regs->pc = regs->regs[0] + offset -
-					instruction_size(ctrl_inw(regs->pc-4));
-		}
-
-		preempt_enable_no_resched();
-#endif
 	}
 
 	/* Set up the stack frame */
diff --git a/include/asm-sh/atomic-grb.h b/include/asm-sh/atomic-grb.h
new file mode 100644
index 000000000000..4c5b7dbfcedb
--- /dev/null
+++ b/include/asm-sh/atomic-grb.h
@@ -0,0 +1,169 @@
+#ifndef __ASM_SH_ATOMIC_GRB_H
+#define __ASM_SH_ATOMIC_GRB_H
+
+static inline void atomic_add(int i, atomic_t *v)
+{
+	int tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   add     %2,   %0      \n\t" /* add */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (i)
+		: "memory" , "r0", "r1");
+}
+
+static inline void atomic_sub(int i, atomic_t *v)
+{
+	int tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov     r15,  r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   sub     %2,   %0      \n\t" /* sub */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (i)
+		: "memory" , "r0", "r1");
+}
+
+static inline int atomic_add_return(int i, atomic_t *v)
+{
+	int tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   add     %2,   %0      \n\t" /* add */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (i)
+		: "memory" , "r0", "r1");
+
+	return tmp;
+}
+
+static inline int atomic_sub_return(int i, atomic_t *v)
+{
+	int tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   sub     %2,   %0      \n\t" /* sub */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (i)
+		: "memory", "r0", "r1");
+
+	return tmp;
+}
+
+static inline void atomic_clear_mask(unsigned int mask, atomic_t *v)
+{
+	int tmp;
+	unsigned int _mask = ~mask;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   and     %2,   %0      \n\t" /* add */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (_mask)
+		: "memory" , "r0", "r1");
+}
+
+static inline void atomic_set_mask(unsigned int mask, atomic_t *v)
+{
+	int tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   or      %2,   %0      \n\t" /* or */
+		"   mov.l   %0,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "+r"  (v)
+		: "r"   (mask)
+		: "memory" , "r0", "r1");
+}
+
+static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+{
+	int ret;
+
+	__asm__ __volatile__ (
+		"   .align 2		\n\t"
+		"   mova     1f,  r0	\n\t"
+		"   nop			\n\t"
+		"   mov     r15,  r1	\n\t"
+		"   mov    #-8,  r15	\n\t"
+		"   mov.l   @%1,  %0	\n\t"
+		"   cmp/eq   %2,  %0	\n\t"
+		"   bf	     1f		\n\t"
+		"   mov.l    %3, @%1	\n\t"
+		"1: mov      r1,  r15	\n\t"
+		: "=&r" (ret)
+		: "r" (v), "r" (old), "r" (new)
+		: "memory" , "r0", "r1" , "t");
+
+	return ret;
+}
+
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int ret;
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+		"   .align 2		\n\t"
+		"   mova    1f,   r0	\n\t"
+		"   nop			\n\t"
+		"   mov    r15,   r1	\n\t"
+		"   mov    #-12,  r15	\n\t"
+		"   mov.l  @%2,   %1	\n\t"
+		"   mov	    %1,   %0    \n\t"
+		"   cmp/eq  %4,   %0	\n\t"
+		"   bt/s    1f		\n\t"
+		"    add    %3,   %1	\n\t"
+		"   mov.l   %1,  @%2	\n\t"
+		"1: mov     r1,   r15	\n\t"
+		: "=&r" (ret), "=&r" (tmp)
+		: "r" (v), "r" (a), "r" (u)
+		: "memory" , "r0", "r1" , "t");
+
+	return ret != u;
+}
+#endif /* __ASM_SH_ATOMIC_GRB_H */
diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h
index e12570b9339d..c043ef003028 100644
--- a/include/asm-sh/atomic.h
+++ b/include/asm-sh/atomic.h
@@ -17,7 +17,9 @@ typedef struct { volatile int counter; } atomic_t;
 #include <linux/compiler.h>
 #include <asm/system.h>
 
-#ifdef CONFIG_CPU_SH4A
+#if defined(CONFIG_GUSA_RB)
+#include <asm/atomic-grb.h>
+#elif defined(CONFIG_CPU_SH4A)
 #include <asm/atomic-llsc.h>
 #else
 #include <asm/atomic-irq.h>
@@ -44,6 +46,7 @@ typedef struct { volatile int counter; } atomic_t;
 #define atomic_inc(v) atomic_add(1,(v))
 #define atomic_dec(v) atomic_sub(1,(v))
 
+#ifndef CONFIG_GUSA_RB
 static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
 {
 	int ret;
@@ -58,8 +61,6 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
 	return ret;
 }
 
-#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
-
 static inline int atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int ret;
@@ -73,6 +74,9 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u)
 
 	return ret != u;
 }
+#endif
+
+#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
 /* Atomic operations are already serializing on SH */
diff --git a/include/asm-sh/bitops-grb.h b/include/asm-sh/bitops-grb.h
new file mode 100644
index 000000000000..a5907b94395b
--- /dev/null
+++ b/include/asm-sh/bitops-grb.h
@@ -0,0 +1,169 @@
+#ifndef __ASM_SH_BITOPS_GRB_H
+#define __ASM_SH_BITOPS_GRB_H
+
+static inline void set_bit(int nr, volatile void * addr)
+{
+	int	mask;
+	volatile unsigned int *a = addr;
+	unsigned long tmp;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+                "   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+                "   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+                "   mov.l  @%1,   %0      \n\t" /* load  old value */
+                "   or      %2,   %0      \n\t" /* or */
+                "   mov.l   %0,   @%1     \n\t" /* store new value */
+                "1: mov     r1,   r15     \n\t" /* LOGOUT */
+                : "=&r" (tmp),
+                  "+r"  (a)
+                : "r"   (mask)
+                : "memory" , "r0", "r1");
+}
+
+static inline void clear_bit(int nr, volatile void * addr)
+{
+	int	mask;
+	volatile unsigned int *a = addr;
+        unsigned long tmp;
+
+	a += nr >> 5;
+        mask = ~(1 << (nr & 0x1f));
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+                "   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+                "   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+                "   mov.l  @%1,   %0      \n\t" /* load  old value */
+                "   and     %2,   %0      \n\t" /* and */
+                "   mov.l   %0,   @%1     \n\t" /* store new value */
+                "1: mov     r1,   r15     \n\t" /* LOGOUT */
+                : "=&r" (tmp),
+                  "+r"  (a)
+                : "r"   (mask)
+                : "memory" , "r0", "r1");
+}
+
+static inline void change_bit(int nr, volatile void * addr)
+{
+        int     mask;
+        volatile unsigned int *a = addr;
+        unsigned long tmp;
+
+        a += nr >> 5;
+        mask = 1 << (nr & 0x1f);
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+                "   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+                "   mov    #-6,   r15     \n\t" /* LOGIN: r15 = size */
+                "   mov.l  @%1,   %0      \n\t" /* load  old value */
+                "   xor     %2,   %0      \n\t" /* xor */
+                "   mov.l   %0,   @%1     \n\t" /* store new value */
+                "1: mov     r1,   r15     \n\t" /* LOGOUT */
+                : "=&r" (tmp),
+                  "+r"  (a)
+                : "r"   (mask)
+                : "memory" , "r0", "r1");
+}
+
+static inline int test_and_set_bit(int nr, volatile void * addr)
+{
+        int     mask, retval;
+	volatile unsigned int *a = addr;
+        unsigned long tmp;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+                "   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+                "   mov   #-14,   r15     \n\t" /* LOGIN: r15 = size */
+                "   mov.l  @%2,   %0      \n\t" /* load old value */
+                "   mov     %0,   %1      \n\t"
+                "   tst     %1,   %3      \n\t" /* T = ((*a & mask) == 0) */
+                "   mov    #-1,   %1      \n\t" /* retvat = -1 */
+                "   negc    %1,   %1      \n\t" /* retval = (mask & *a) != 0 */
+                "   or      %3,   %0      \n\t"
+                "   mov.l   %0,  @%2      \n\t" /* store new value */
+                "1: mov     r1,  r15      \n\t" /* LOGOUT */
+                : "=&r" (tmp),
+                  "=&r" (retval),
+                  "+r"  (a)
+                : "r"   (mask)
+                : "memory" , "r0", "r1" ,"t");
+
+        return retval;
+}
+
+static inline int test_and_clear_bit(int nr, volatile void * addr)
+{
+        int     mask, retval,not_mask;
+        volatile unsigned int *a = addr;
+        unsigned long tmp;
+
+        a += nr >> 5;
+        mask = 1 << (nr & 0x1f);
+
+	not_mask = ~mask;
+
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov   #-14,   r15     \n\t" /* LOGIN */
+		"   mov.l  @%2,   %0      \n\t" /* load old value */
+                "   mov     %0,   %1      \n\t" /* %1 = *a */
+                "   tst     %1,   %3      \n\t" /* T = ((*a & mask) == 0) */
+		"   mov    #-1,   %1      \n\t" /* retvat = -1 */
+                "   negc    %1,   %1      \n\t" /* retval = (mask & *a) != 0 */
+                "   and     %4,   %0      \n\t"
+                "   mov.l   %0,  @%2      \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (tmp),
+		  "=&r" (retval),
+		  "+r"  (a)
+		: "r"   (mask),
+		  "r"   (not_mask)
+		: "memory" , "r0", "r1", "t");
+
+        return retval;
+}
+
+static inline int test_and_change_bit(int nr, volatile void * addr)
+{
+        int     mask, retval;
+        volatile unsigned int *a = addr;
+        unsigned long tmp;
+
+        a += nr >> 5;
+        mask = 1 << (nr & 0x1f);
+
+        __asm__ __volatile__ (
+                "   .align 2              \n\t"
+                "   mova    1f,   r0      \n\t" /* r0 = end point */
+                "   mov    r15,   r1      \n\t" /* r1 = saved sp */
+                "   mov   #-14,   r15     \n\t" /* LOGIN */
+                "   mov.l  @%2,   %0      \n\t" /* load old value */
+                "   mov     %0,   %1      \n\t" /* %1 = *a */
+                "   tst     %1,   %3      \n\t" /* T = ((*a & mask) == 0) */
+                "   mov    #-1,   %1      \n\t" /* retvat = -1 */
+                "   negc    %1,   %1      \n\t" /* retval = (mask & *a) != 0 */
+                "   xor     %3,   %0      \n\t"
+                "   mov.l   %0,  @%2      \n\t" /* store new value */
+                "1: mov     r1,   r15     \n\t" /* LOGOUT */
+                : "=&r" (tmp),
+                  "=&r" (retval),
+                  "+r"  (a)
+                : "r"   (mask)
+                : "memory" , "r0", "r1", "t");
+
+        return retval;
+}
+#endif /* __ASM_SH_BITOPS_GRB_H */
diff --git a/include/asm-sh/bitops-irq.h b/include/asm-sh/bitops-irq.h
new file mode 100644
index 000000000000..653a12750584
--- /dev/null
+++ b/include/asm-sh/bitops-irq.h
@@ -0,0 +1,91 @@
+#ifndef __ASM_SH_BITOPS_IRQ_H
+#define __ASM_SH_BITOPS_IRQ_H
+
+static inline void set_bit(int nr, volatile void *addr)
+{
+	int	mask;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	*a |= mask;
+	local_irq_restore(flags);
+}
+
+static inline void clear_bit(int nr, volatile void *addr)
+{
+	int	mask;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	*a &= ~mask;
+	local_irq_restore(flags);
+}
+
+static inline void change_bit(int nr, volatile void *addr)
+{
+	int	mask;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	*a ^= mask;
+	local_irq_restore(flags);
+}
+
+static inline int test_and_set_bit(int nr, volatile void *addr)
+{
+	int	mask, retval;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	retval = (mask & *a) != 0;
+	*a |= mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+static inline int test_and_clear_bit(int nr, volatile void *addr)
+{
+	int	mask, retval;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	retval = (mask & *a) != 0;
+	*a &= ~mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+static inline int test_and_change_bit(int nr, volatile void *addr)
+{
+	int	mask, retval;
+	volatile unsigned int *a = addr;
+	unsigned long flags;
+
+	a += nr >> 5;
+	mask = 1 << (nr & 0x1f);
+	local_irq_save(flags);
+	retval = (mask & *a) != 0;
+	*a ^= mask;
+	local_irq_restore(flags);
+
+	return retval;
+}
+
+#endif /* __ASM_SH_BITOPS_IRQ_H */
diff --git a/include/asm-sh/bitops.h b/include/asm-sh/bitops.h
index a7bd81a7f064..b6ba5a60dec2 100644
--- a/include/asm-sh/bitops.h
+++ b/include/asm-sh/bitops.h
@@ -11,97 +11,18 @@
 /* For __swab32 */
 #include <asm/byteorder.h>
 
-static inline void set_bit(int nr, volatile void * addr)
-{
-	int	mask;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a |= mask;
-	local_irq_restore(flags);
-}
+#ifdef CONFIG_GUSA_RB
+#include <asm/bitops-grb.h>
+#else
+#include <asm/bitops-irq.h>
+#endif
+
 
 /*
  * clear_bit() doesn't provide any barrier for the compiler.
  */
 #define smp_mb__before_clear_bit()	barrier()
 #define smp_mb__after_clear_bit()	barrier()
-static inline void clear_bit(int nr, volatile void * addr)
-{
-	int	mask;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a &= ~mask;
-	local_irq_restore(flags);
-}
-
-static inline void change_bit(int nr, volatile void * addr)
-{
-	int	mask;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	*a ^= mask;
-	local_irq_restore(flags);
-}
-
-static inline int test_and_set_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a |= mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
-
-static inline int test_and_clear_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a &= ~mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
-
-static inline int test_and_change_bit(int nr, volatile void * addr)
-{
-	int	mask, retval;
-	volatile unsigned int *a = addr;
-	unsigned long flags;
-
-	a += nr >> 5;
-	mask = 1 << (nr & 0x1f);
-	local_irq_save(flags);
-	retval = (mask & *a) != 0;
-	*a ^= mask;
-	local_irq_restore(flags);
-
-	return retval;
-}
 
 #include <asm-generic/bitops/non-atomic.h>
 
diff --git a/include/asm-sh/cmpxchg-grb.h b/include/asm-sh/cmpxchg-grb.h
new file mode 100644
index 000000000000..e2681abe764f
--- /dev/null
+++ b/include/asm-sh/cmpxchg-grb.h
@@ -0,0 +1,70 @@
+#ifndef __ASM_SH_CMPXCHG_GRB_H
+#define __ASM_SH_CMPXCHG_GRB_H
+
+static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
+{
+	unsigned long retval;
+
+	__asm__ __volatile__ (
+		"   .align 2              \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   nop                   \n\t"
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-4,   r15     \n\t" /* LOGIN */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   mov.l   %2,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (retval),
+		  "+r"  (m)
+		: "r"   (val)
+		: "memory", "r0", "r1");
+
+	return retval;
+}
+
+static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
+{
+	unsigned long retval;
+
+	__asm__ __volatile__ (
+		"   .align  2             \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-6,   r15     \n\t" /* LOGIN */
+		"   mov.b  @%1,   %0      \n\t" /* load  old value */
+		"   extu.b  %0,   %0      \n\t" /* extend as unsigned */
+		"   mov.b   %2,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (retval),
+		  "+r"  (m)
+		: "r"   (val)
+		: "memory" , "r0", "r1");
+
+	return retval;
+}
+
+static inline unsigned long __cmpxchg_u32(volatile int *m, unsigned long old,
+					  unsigned long new)
+{
+	unsigned long retval;
+
+	__asm__ __volatile__ (
+		"   .align  2             \n\t"
+		"   mova    1f,   r0      \n\t" /* r0 = end point */
+		"   nop                   \n\t"
+		"   mov    r15,   r1      \n\t" /* r1 = saved sp */
+		"   mov    #-8,   r15     \n\t" /* LOGIN */
+		"   mov.l  @%1,   %0      \n\t" /* load  old value */
+		"   cmp/eq  %0,   %2      \n\t"
+		"   bf            1f      \n\t" /* if not equal */
+		"   mov.l   %2,   @%1     \n\t" /* store new value */
+		"1: mov     r1,   r15     \n\t" /* LOGOUT */
+		: "=&r" (retval),
+		  "+r"  (m)
+		: "r"   (new)
+		: "memory" , "r0", "r1", "t");
+
+	return retval;
+}
+
+#endif /* __ASM_SH_CMPXCHG_GRB_H */
diff --git a/include/asm-sh/cmpxchg-irq.h b/include/asm-sh/cmpxchg-irq.h
new file mode 100644
index 000000000000..43049ec0554b
--- /dev/null
+++ b/include/asm-sh/cmpxchg-irq.h
@@ -0,0 +1,40 @@
+#ifndef __ASM_SH_CMPXCHG_IRQ_H
+#define __ASM_SH_CMPXCHG_IRQ_H
+
+static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
+{
+	unsigned long flags, retval;
+
+	local_irq_save(flags);
+	retval = *m;
+	*m = val;
+	local_irq_restore(flags);
+	return retval;
+}
+
+static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
+{
+	unsigned long flags, retval;
+
+	local_irq_save(flags);
+	retval = *m;
+	*m = val & 0xff;
+	local_irq_restore(flags);
+	return retval;
+}
+
+static inline unsigned long __cmpxchg_u32(volatile int *m, unsigned long old,
+	unsigned long new)
+{
+	__u32 retval;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	retval = *m;
+	if (retval == old)
+		*m = new;
+	local_irq_restore(flags);       /* implies memory barrier  */
+	return retval;
+}
+
+#endif /* __ASM_SH_CMPXCHG_IRQ_H */
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index ad3d2a636130..969f3d4afe2a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -68,27 +68,11 @@
 
 #define set_mb(var, value) do { (void)xchg(&var, value); } while (0)
 
-static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
-{
-	unsigned long flags, retval;
-
-	local_irq_save(flags);
-	retval = *m;
-	*m = val;
-	local_irq_restore(flags);
-	return retval;
-}
-
-static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
-{
-	unsigned long flags, retval;
-
-	local_irq_save(flags);
-	retval = *m;
-	*m = val & 0xff;
-	local_irq_restore(flags);
-	return retval;
-}
+#ifdef CONFIG_GUSA_RB
+#include <asm/cmpxchg-grb.h>
+#else
+#include <asm/cmpxchg-irq.h>
+#endif
 
 extern void __xchg_called_with_bad_pointer(void);
 
@@ -115,20 +99,6 @@ extern void __xchg_called_with_bad_pointer(void);
 #define xchg(ptr,x)	\
 	((__typeof__(*(ptr)))__xchg((ptr),(unsigned long)(x), sizeof(*(ptr))))
 
-static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
-	unsigned long new)
-{
-	__u32 retval;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	retval = *m;
-	if (retval == old)
-		*m = new;
-	local_irq_restore(flags);       /* implies memory barrier  */
-	return retval;
-}
-
 /* This function doesn't exist, so you'll get a linker error
  * if something tries to do an invalid cmpxchg(). */
 extern void __cmpxchg_called_with_bad_pointer(void);
-- 
cgit v1.2.3


From cbaa118ecfd99fc5ed7adbd9c34a30e1c05e3c93 Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 17:06:36 +0900
Subject: sh: Preparation for uncached jumps through PMB.

Presently most of the 29-bit physical parts do P1/P2 segmentation
with a 1:1 cached/uncached mapping, jumping between the two to
control the caching behaviour. This provides the basic infrastructure
to maintain this behaviour on 32-bit physical parts that don't map
P1/P2 at all, using a shiny new linker section and corresponding
fixmap entry.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c       |  6 +++---
 arch/sh/kernel/cpu/sh3/probe.c  |  6 +++---
 arch/sh/kernel/vmlinux_32.lds.S |  9 +++++++++
 arch/sh/mm/cache-debugfs.c      |  9 +++++----
 arch/sh/mm/cache-sh4.c          | 14 +++++++-------
 arch/sh/mm/cache-sh7705.c       | 12 ++++++------
 arch/sh/mm/init.c               | 16 +++++++++++++++-
 arch/sh/mm/pmb.c                | 18 +++++++++---------
 arch/sh/mm/tlb-sh4.c            |  7 ++++---
 include/asm-sh/fixmap.h         |  1 +
 include/asm-sh/sections.h       |  1 +
 include/asm-sh/system.h         |  2 ++
 include/asm-sh/system_32.h      | 36 +++++++++++++++++++-----------------
 include/asm-sh/system_64.h      |  7 ++++---
 14 files changed, 88 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index fd1688e6c61c..0f0c76a842e4 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -64,11 +64,11 @@ static void __init speculative_execution_init(void)
  * Generic first-level cache init
  */
 #ifdef CONFIG_SUPERH32
-static void __init cache_init(void)
+static void __uses_jump_to_uncached cache_init(void)
 {
 	unsigned long ccr, flags;
 
-	jump_to_P2();
+	jump_to_uncached();
 	ccr = ctrl_inl(CCR);
 
 	/*
@@ -145,7 +145,7 @@ static void __init cache_init(void)
 #endif
 
 	ctrl_outl(flags, CCR);
-	back_to_P1();
+	back_to_cached();
 }
 #else
 #define cache_init()	do { } while (0)
diff --git a/arch/sh/kernel/cpu/sh3/probe.c b/arch/sh/kernel/cpu/sh3/probe.c
index bf579e061e09..22070e43e34d 100644
--- a/arch/sh/kernel/cpu/sh3/probe.c
+++ b/arch/sh/kernel/cpu/sh3/probe.c
@@ -16,11 +16,11 @@
 #include <asm/cache.h>
 #include <asm/io.h>
 
-int __init detect_cpu_and_cache_system(void)
+int __uses_jump_to_uncached detect_cpu_and_cache_system(void)
 {
 	unsigned long addr0, addr1, data0, data1, data2, data3;
 
-	jump_to_P2();
+	jump_to_uncached();
 	/*
 	 * Check if the entry shadows or not.
 	 * When shadowed, it's 128-entry system.
@@ -48,7 +48,7 @@ int __init detect_cpu_and_cache_system(void)
 	ctrl_outl(data0&~SH_CACHE_VALID, addr0);
 	ctrl_outl(data2&~SH_CACHE_VALID, addr1);
 
-	back_to_P1();
+	back_to_cached();
 
 	boot_cpu_data.dcache.ways		= 4;
 	boot_cpu_data.dcache.entry_shift	= 4;
diff --git a/arch/sh/kernel/vmlinux_32.lds.S b/arch/sh/kernel/vmlinux_32.lds.S
index 0956fb3681a3..50c69c18dced 100644
--- a/arch/sh/kernel/vmlinux_32.lds.S
+++ b/arch/sh/kernel/vmlinux_32.lds.S
@@ -43,6 +43,15 @@ SECTIONS
 	NOTES
 	RO_DATA(PAGE_SIZE)
 
+	/*
+	 * Code which must be executed uncached and the associated data
+	 */
+	. = ALIGN(PAGE_SIZE);
+	__uncached_start = .;
+	.uncached.text : { *(.uncached.text) }
+	.uncached.data : { *(.uncached.data) }
+	__uncached_end = .;
+
 	. = ALIGN(THREAD_SIZE);
 	.data : {			/* Data */
 		*(.data.init_task)
diff --git a/arch/sh/mm/cache-debugfs.c b/arch/sh/mm/cache-debugfs.c
index de6d2c9aa477..db6d950b6f5e 100644
--- a/arch/sh/mm/cache-debugfs.c
+++ b/arch/sh/mm/cache-debugfs.c
@@ -22,7 +22,8 @@ enum cache_type {
 	CACHE_TYPE_UNIFIED,
 };
 
-static int cache_seq_show(struct seq_file *file, void *iter)
+static int __uses_jump_to_uncached cache_seq_show(struct seq_file *file,
+						  void *iter)
 {
 	unsigned int cache_type = (unsigned int)file->private;
 	struct cache_info *cache;
@@ -34,11 +35,11 @@ static int cache_seq_show(struct seq_file *file, void *iter)
 	 * Go uncached immediately so we don't skew the results any
 	 * more than we already are..
 	 */
-	jump_to_P2();
+	jump_to_uncached();
 
 	ccr = ctrl_inl(CCR);
 	if ((ccr & CCR_CACHE_ENABLE) == 0) {
-		back_to_P1();
+		back_to_cached();
 
 		seq_printf(file, "disabled\n");
 		return 0;
@@ -104,7 +105,7 @@ static int cache_seq_show(struct seq_file *file, void *iter)
 		addrstart += cache->way_incr;
 	}
 
-	back_to_P1();
+	back_to_cached();
 
 	return 0;
 }
diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 226b190c5b9c..43d7ff6b6ec7 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -190,7 +190,7 @@ void flush_icache_range(unsigned long start, unsigned long end)
  * .. which happens to be the same behavior as flush_icache_range().
  * So, we simply flush out a line.
  */
-void flush_cache_sigtramp(unsigned long addr)
+void __uses_jump_to_uncached flush_cache_sigtramp(unsigned long addr)
 {
 	unsigned long v, index;
 	unsigned long flags;
@@ -205,13 +205,13 @@ void flush_cache_sigtramp(unsigned long addr)
 			(v & boot_cpu_data.icache.entry_mask);
 
 	local_irq_save(flags);
-	jump_to_P2();
+	jump_to_uncached();
 
 	for (i = 0; i < boot_cpu_data.icache.ways;
 	     i++, index += boot_cpu_data.icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
 
-	back_to_P1();
+	back_to_cached();
 	wmb();
 	local_irq_restore(flags);
 }
@@ -256,12 +256,12 @@ void flush_dcache_page(struct page *page)
 }
 
 /* TODO: Selective icache invalidation through IC address array.. */
-static inline void flush_icache_all(void)
+static inline void __uses_jump_to_uncached flush_icache_all(void)
 {
 	unsigned long flags, ccr;
 
 	local_irq_save(flags);
-	jump_to_P2();
+	jump_to_uncached();
 
 	/* Flush I-cache */
 	ccr = ctrl_inl(CCR);
@@ -269,11 +269,11 @@ static inline void flush_icache_all(void)
 	ctrl_outl(ccr, CCR);
 
 	/*
-	 * back_to_P1() will take care of the barrier for us, don't add
+	 * back_to_cached() will take care of the barrier for us, don't add
 	 * another one!
 	 */
 
-	back_to_P1();
+	back_to_cached();
 	local_irq_restore(flags);
 }
 
diff --git a/arch/sh/mm/cache-sh7705.c b/arch/sh/mm/cache-sh7705.c
index 4896d7376926..22dacc778823 100644
--- a/arch/sh/mm/cache-sh7705.c
+++ b/arch/sh/mm/cache-sh7705.c
@@ -71,7 +71,7 @@ void flush_icache_range(unsigned long start, unsigned long end)
 /*
  * Writeback&Invalidate the D-cache of the page
  */
-static void __flush_dcache_page(unsigned long phys)
+static void __uses_jump_to_uncached __flush_dcache_page(unsigned long phys)
 {
 	unsigned long ways, waysize, addrstart;
 	unsigned long flags;
@@ -92,7 +92,7 @@ static void __flush_dcache_page(unsigned long phys)
 	 * possible.
 	 */
 	local_irq_save(flags);
-	jump_to_P2();
+	jump_to_uncached();
 
 	ways = current_cpu_data.dcache.ways;
 	waysize = current_cpu_data.dcache.sets;
@@ -118,7 +118,7 @@ static void __flush_dcache_page(unsigned long phys)
 		addrstart += current_cpu_data.dcache.way_incr;
 	} while (--ways);
 
-	back_to_P1();
+	back_to_cached();
 	local_irq_restore(flags);
 }
 
@@ -132,15 +132,15 @@ void flush_dcache_page(struct page *page)
 		__flush_dcache_page(PHYSADDR(page_address(page)));
 }
 
-void flush_cache_all(void)
+void __uses_jump_to_uncached flush_cache_all(void)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
-	jump_to_P2();
+	jump_to_uncached();
 
 	cache_wback_all();
-	back_to_P1();
+	back_to_cached();
 	local_irq_restore(flags);
 }
 
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 79c309780f95..094225e0d722 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -23,6 +23,7 @@
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 pgd_t swapper_pg_dir[PTRS_PER_PGD];
+unsigned long cached_to_uncached = 0;
 
 void show_mem(void)
 {
@@ -99,7 +100,8 @@ static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 
 	set_pte(pte, pfn_pte(phys >> PAGE_SHIFT, prot));
 
-	flush_tlb_one(get_asid(), addr);
+	if (cached_to_uncached)
+		flush_tlb_one(get_asid(), addr);
 }
 
 /*
@@ -164,6 +166,18 @@ void __init paging_init(void)
 	}
 
 	free_area_init_nodes(max_zone_pfns);
+
+	/* Set up the uncached fixmap */
+	set_fixmap_nocache(FIX_UNCACHED, __pa(&__uncached_start));
+
+#ifdef CONFIG_29BIT
+	/*
+	 * Handle trivial transitions between cached and uncached
+	 * segments, making use of the 1:1 mapping relationship in
+	 * 512MB lowmem.
+	 */
+	cached_to_uncached = P2SEG - P1SEG;
+#endif
 }
 
 static struct kcore_list kcore_mem, kcore_vmalloc;
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index ef6ab39eaf65..ab81c602295f 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -163,18 +163,18 @@ repeat:
 	return 0;
 }
 
-int set_pmb_entry(struct pmb_entry *pmbe)
+int __uses_jump_to_uncached set_pmb_entry(struct pmb_entry *pmbe)
 {
 	int ret;
 
-	jump_to_P2();
+	jump_to_uncached();
 	ret = __set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
-	back_to_P1();
+	back_to_cached();
 
 	return ret;
 }
 
-void clear_pmb_entry(struct pmb_entry *pmbe)
+void __uses_jump_to_uncached clear_pmb_entry(struct pmb_entry *pmbe)
 {
 	unsigned int entry = pmbe->entry;
 	unsigned long addr;
@@ -188,7 +188,7 @@ void clear_pmb_entry(struct pmb_entry *pmbe)
 		     entry >= NR_PMB_ENTRIES))
 		return;
 
-	jump_to_P2();
+	jump_to_uncached();
 
 	/* Clear V-bit */
 	addr = mk_pmb_addr(entry);
@@ -197,7 +197,7 @@ void clear_pmb_entry(struct pmb_entry *pmbe)
 	addr = mk_pmb_data(entry);
 	ctrl_outl(ctrl_inl(addr) & ~PMB_V, addr);
 
-	back_to_P1();
+	back_to_cached();
 
 	clear_bit(entry, &pmb_map);
 }
@@ -302,7 +302,7 @@ static void pmb_cache_ctor(struct kmem_cache *cachep, void *pmb)
 	pmbe->entry = PMB_NO_ENTRY;
 }
 
-static int __init pmb_init(void)
+static int __uses_jump_to_uncached pmb_init(void)
 {
 	unsigned int nr_entries = ARRAY_SIZE(pmb_init_map);
 	unsigned int entry, i;
@@ -312,7 +312,7 @@ static int __init pmb_init(void)
 	pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry), 0,
 				      SLAB_PANIC, pmb_cache_ctor);
 
-	jump_to_P2();
+	jump_to_uncached();
 
 	/*
 	 * Ordering is important, P2 must be mapped in the PMB before we
@@ -335,7 +335,7 @@ static int __init pmb_init(void)
 	i |= MMUCR_TI;
 	ctrl_outl(i, MMUCR);
 
-	back_to_P1();
+	back_to_cached();
 
 	return 0;
 }
diff --git a/arch/sh/mm/tlb-sh4.c b/arch/sh/mm/tlb-sh4.c
index 2d1dd6044307..f0c7b7397fa6 100644
--- a/arch/sh/mm/tlb-sh4.c
+++ b/arch/sh/mm/tlb-sh4.c
@@ -79,7 +79,8 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	local_irq_restore(flags);
 }
 
-void local_flush_tlb_one(unsigned long asid, unsigned long page)
+void __uses_jump_to_uncached local_flush_tlb_one(unsigned long asid,
+						 unsigned long page)
 {
 	unsigned long addr, data;
 
@@ -91,7 +92,7 @@ void local_flush_tlb_one(unsigned long asid, unsigned long page)
 	 */
 	addr = MMU_UTLB_ADDRESS_ARRAY | MMU_PAGE_ASSOC_BIT;
 	data = page | asid; /* VALID bit is off */
-	jump_to_P2();
+	jump_to_uncached();
 	ctrl_outl(data, addr);
-	back_to_P1();
+	back_to_cached();
 }
diff --git a/include/asm-sh/fixmap.h b/include/asm-sh/fixmap.h
index 09463cd9bbb9..721fcc4d5e98 100644
--- a/include/asm-sh/fixmap.h
+++ b/include/asm-sh/fixmap.h
@@ -49,6 +49,7 @@ enum fixed_addresses {
 #define FIX_N_COLOURS 16
 	FIX_CMAP_BEGIN,
 	FIX_CMAP_END = FIX_CMAP_BEGIN + FIX_N_COLOURS,
+	FIX_UNCACHED,
 #ifdef CONFIG_HIGHMEM
 	FIX_KMAP_BEGIN,	/* reserved pte's for temporary kernel mappings */
 	FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
diff --git a/include/asm-sh/sections.h b/include/asm-sh/sections.h
index bd9cbc967c2a..8f8f4ad400df 100644
--- a/include/asm-sh/sections.h
+++ b/include/asm-sh/sections.h
@@ -4,6 +4,7 @@
 #include <asm-generic/sections.h>
 
 extern long __machvec_start, __machvec_end;
+extern char __uncached_start, __uncached_end;
 extern char _ebss[];
 
 #endif /* __ASM_SH_SECTIONS_H */
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 969f3d4afe2a..9bda8d063ecf 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -144,6 +144,8 @@ extern unsigned int instruction_size(unsigned int insn);
 #define instruction_size(insn)	(4)
 #endif
 
+extern unsigned long cached_to_uncached;
+
 /* XXX
  * disable hlt during certain critical i/o operations
  */
diff --git a/include/asm-sh/system_32.h b/include/asm-sh/system_32.h
index ad37e8d5f31e..e918bacd5ecf 100644
--- a/include/asm-sh/system_32.h
+++ b/include/asm-sh/system_32.h
@@ -58,29 +58,31 @@ do {								\
 	last = __last;						\
 } while (0)
 
+#define __uses_jump_to_uncached __attribute__ ((__section__ (".uncached.text")))
+
 /*
- * Jump to P2 area.
- * When handling TLB or caches, we need to do it from P2 area.
+ * Jump to uncached area.
+ * When handling TLB or caches, we need to do it from an uncached area.
  */
-#define jump_to_P2()			\
-do {					\
-	unsigned long __dummy;		\
-	__asm__ __volatile__(		\
-		"mov.l	1f, %0\n\t"	\
-		"or	%1, %0\n\t"	\
-		"jmp	@%0\n\t"	\
-		" nop\n\t"		\
-		".balign 4\n"		\
-		"1:	.long 2f\n"	\
-		"2:"			\
-		: "=&r" (__dummy)	\
-		: "r" (0x20000000));	\
+#define jump_to_uncached()			\
+do {						\
+	unsigned long __dummy;			\
+						\
+	__asm__ __volatile__(			\
+		"mova	1f, %0\n\t"		\
+		"add	%1, %0\n\t"		\
+		"jmp	@%0\n\t"		\
+		" nop\n\t"			\
+		".balign 4\n"			\
+		"1:"				\
+		: "=&z" (__dummy)		\
+		: "r" (cached_to_uncached));	\
 } while (0)
 
 /*
- * Back to P1 area.
+ * Back to cached area.
  */
-#define back_to_P1()					\
+#define back_to_cached()				\
 do {							\
 	unsigned long __dummy;				\
 	ctrl_barrier();					\
diff --git a/include/asm-sh/system_64.h b/include/asm-sh/system_64.h
index 0e466e991f7d..943acf5ea07c 100644
--- a/include/asm-sh/system_64.h
+++ b/include/asm-sh/system_64.h
@@ -32,8 +32,9 @@ do {								\
 			      &next->thread);			\
 } while (0)
 
-/* No segmentation.. */
-#define jump_to_P2()	do { } while (0)
-#define back_to_P1()	do { } while (0)
+#define __uses_jump_to_uncached
+
+#define jump_to_uncached()	do { } while (0)
+#define back_to_cached()	do { } while (0)
 
 #endif /* __ASM_SH_SYSTEM_64_H */
-- 
cgit v1.2.3


From d02b08f6e8b184ffef349e395210a5e82ff4f4bc Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 17:52:53 +0900
Subject: sh: Clean up places that make 29-bit physical assumptions.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boot/compressed/misc_32.c    |  5 +++-
 arch/sh/kernel/setup.c               |  4 +--
 arch/sh/kernel/vmlinux_32.lds.S      |  5 ++++
 include/asm-sh/addrspace.h           |  6 ++++-
 include/asm-sh/cpu-sh4/mmu_context.h |  8 +++++-
 include/asm-sh/io.h                  | 16 +-----------
 include/asm-sh/page.h                | 50 ++++++++++++++++++++++++++----------
 include/asm-sh/pgtable.h             |  8 +++++-
 include/asm-sh/pgtable_32.h          |  2 +-
 include/asm-sh/scatterlist.h         |  3 ++-
 10 files changed, 71 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boot/compressed/misc_32.c b/arch/sh/boot/compressed/misc_32.c
index df65e305acf7..adcea31e663e 100644
--- a/arch/sh/boot/compressed/misc_32.c
+++ b/arch/sh/boot/compressed/misc_32.c
@@ -230,7 +230,10 @@ long* stack_start = &user_stack[STACK_SIZE];
 void decompress_kernel(void)
 {
 	output_data = 0;
-	output_ptr = P2SEGADDR((unsigned long)&_text+PAGE_SIZE);
+	output_ptr = PHYSADDR((unsigned long)&_text+PAGE_SIZE);
+#ifdef CONFIG_29BIT
+	output_ptr |= P2SEG;
+#endif
 	free_mem_ptr = (unsigned long)&_end;
 	free_mem_end_ptr = free_mem_ptr + HEAP_SIZE;
 
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 7eb7fdcce0f8..f48ce8e5d0a8 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -82,7 +82,7 @@ static int __init early_parse_mem(char *p)
 {
 	unsigned long size;
 
-	memory_start = (unsigned long)PAGE_OFFSET+__MEMORY_START;
+	memory_start = (unsigned long)__va(__MEMORY_START);
 	size = memparse(p, &p);
 
 	if (size > __MEMORY_SIZE) {
@@ -254,7 +254,7 @@ void __init setup_arch(char **cmdline_p)
 	data_resource.start = virt_to_phys(_etext);
 	data_resource.end = virt_to_phys(_edata)-1;
 
-	memory_start = (unsigned long)PAGE_OFFSET+__MEMORY_START;
+	memory_start = (unsigned long)__va(__MEMORY_START);
 	if (!memory_end)
 		memory_end = memory_start + __MEMORY_SIZE;
 
diff --git a/arch/sh/kernel/vmlinux_32.lds.S b/arch/sh/kernel/vmlinux_32.lds.S
index 50c69c18dced..d549fac6d3e7 100644
--- a/arch/sh/kernel/vmlinux_32.lds.S
+++ b/arch/sh/kernel/vmlinux_32.lds.S
@@ -15,7 +15,12 @@ OUTPUT_ARCH(sh)
 ENTRY(_start)
 SECTIONS
 {
+#ifdef CONFIG_32BIT
+	. = CONFIG_PAGE_OFFSET + CONFIG_ZERO_PAGE_OFFSET;
+#else
 	. = CONFIG_PAGE_OFFSET + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
+#endif
+
 	_text = .;			/* Text and read-only data */
 
 	.empty_zero_page : {
diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index e7f2deb28e06..fa544fc38c23 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -31,6 +31,7 @@
 /* Returns the physical address of a PnSEG (n=1,2) address   */
 #define PHYSADDR(a)	(((unsigned long)(a)) & 0x1fffffff)
 
+#ifdef CONFIG_29BIT
 /*
  * Map an address to a certain privileged segment
  */
@@ -42,8 +43,11 @@
 	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P3SEG))
 #define P4SEGADDR(a)	\
 	((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) | P4SEG))
-
+#endif /* 29BIT */
 #endif /* P1SEG */
 
+/* Check if an address can be reached in 29 bits */
+#define IS_29BIT(a)	(((unsigned long)(a)) < 0x20000000)
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_ADDRSPACE_H */
diff --git a/include/asm-sh/cpu-sh4/mmu_context.h b/include/asm-sh/cpu-sh4/mmu_context.h
index fdd56e3e3a3b..9ea8eb27b18e 100644
--- a/include/asm-sh/cpu-sh4/mmu_context.h
+++ b/include/asm-sh/cpu-sh4/mmu_context.h
@@ -30,6 +30,12 @@
 #define MMUCR_ME		(0)
 #endif
 
+#if defined(CONFIG_32BIT) && defined(CONFIG_CPU_SUBTYPE_ST40)
+#define MMUCR_SE		(1 << 4)
+#else
+#define MMUCR_SE		(0)
+#endif
+
 #ifdef CONFIG_SH_STORE_QUEUES
 #define MMUCR_SQMD		(1 << 9)
 #else
@@ -37,7 +43,7 @@
 #endif
 
 #define MMU_NTLB_ENTRIES	64
-#define MMU_CONTROL_INIT	(0x05|MMUCR_SQMD|MMUCR_ME)
+#define MMU_CONTROL_INIT	(0x05|MMUCR_SQMD|MMUCR_ME|MMUCR_SE)
 
 #define MMU_ITLB_DATA_ARRAY	0xF3000000
 #define MMU_UTLB_DATA_ARRAY	0xF7000000
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index a4e5f5573eee..94900c089519 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -273,23 +273,9 @@ extern void onchip_unmap(unsigned long vaddr);
 #if !defined(CONFIG_MMU)
 #define virt_to_phys(address)	((unsigned long)(address))
 #define phys_to_virt(address)	((void *)(address))
-#elif defined(CONFIG_SUPERH64)
+#else
 #define virt_to_phys(address)	(__pa(address))
 #define phys_to_virt(address)	(__va(address))
-#else
-/*
- * Change virtual addresses to physical addresses and vv.
- * These are trivial on the 1:1 Linux/SuperH mapping
- */
-static inline unsigned long virt_to_phys(volatile void *address)
-{
-	return PHYSADDR(address);
-}
-
-static inline void *phys_to_virt(unsigned long address)
-{
-	return (void *)P1SEGADDR(address);
-}
 #endif
 
 /*
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index bff635a078c8..002e64a4f049 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -5,6 +5,8 @@
  * Copyright (C) 1999  Niibe Yutaka
  */
 
+#include <linux/const.h>
+
 #ifdef __KERNEL__
 
 /* PAGE_SHIFT determines the page size */
@@ -18,15 +20,13 @@
 # error "Bogus kernel page size?"
 #endif
 
-#ifdef __ASSEMBLY__
-#define PAGE_SIZE	(1 << PAGE_SHIFT)
-#else
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
-#endif
-
+#define PAGE_SIZE	(_AC(1, UL) << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
+/* to align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
+
 #if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
 #define HPAGE_SHIFT	16
 #elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
@@ -104,20 +104,44 @@ typedef struct { unsigned long pgd; } pgd_t;
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
+/*
+ * __MEMORY_START and SIZE are the physical addresses and size of RAM.
+ */
 #define __MEMORY_START		CONFIG_MEMORY_START
 #define __MEMORY_SIZE		CONFIG_MEMORY_SIZE
 
+/*
+ * PAGE_OFFSET is the virtual address of the start of kernel address
+ * space.
+ */
 #define PAGE_OFFSET		CONFIG_PAGE_OFFSET
-#define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
-#define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
-#define pfn_to_kaddr(pfn)	__va((pfn) << PAGE_SHIFT)
 
+/*
+ * Virtual to physical RAM address translation.
+ *
+ * In 29 bit mode, the physical offset of RAM from address 0 is visible in
+ * the kernel virtual address space, and thus we don't have to take
+ * this into account when translating. However in 32 bit mode this offset
+ * is not visible (it is part of the PMB mapping) and so needs to be
+ * added or subtracted as required.
+ */
+#ifdef CONFIG_32BIT
+#define __pa(x)	((unsigned long)(x)-PAGE_OFFSET+__MEMORY_START)
+#define __va(x)	((void *)((unsigned long)(x)+PAGE_OFFSET-__MEMORY_START))
+#else
+#define __pa(x)	((unsigned long)(x)-PAGE_OFFSET)
+#define __va(x)	((void *)((unsigned long)(x)+PAGE_OFFSET))
+#endif
+
+#define pfn_to_kaddr(pfn)	__va((pfn) << PAGE_SHIFT)
 #define page_to_phys(page)	(page_to_pfn(page) << PAGE_SHIFT)
 
-/* PFN start number, because of __MEMORY_START */
+/*
+ * PFN = physical frame number (ie PFN 0 == physical address 0)
+ * PFN_START is the PFN of the first page of RAM. By defining this we
+ * don't have struct page entries for the portion of address space
+ * between physical address 0 and the start of RAM.
+ */
 #define PFN_START		(__MEMORY_START >> PAGE_SHIFT)
 #define ARCH_PFN_OFFSET		(PFN_START)
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index b4d7561cd9e0..3df90f003e95 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -69,7 +69,13 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define USER_PTRS_PER_PGD	(TASK_SIZE/PGDIR_SIZE)
 #define FIRST_USER_ADDRESS	0
 
-#define PTE_PHYS_MASK		(0x20000000 - PAGE_SIZE)
+#ifdef CONFIG_32BIT
+#define PHYS_ADDR_MASK		0xffffffff
+#else
+#define PHYS_ADDR_MASK		0x1fffffff
+#endif
+
+#define PTE_PHYS_MASK		(PHYS_ADDR_MASK & PAGE_MASK)
 
 #ifdef CONFIG_SUPERH32
 #define VMALLOC_START	(P3SEG)
diff --git a/include/asm-sh/pgtable_32.h b/include/asm-sh/pgtable_32.h
index 70303603e89d..7efc95404c64 100644
--- a/include/asm-sh/pgtable_32.h
+++ b/include/asm-sh/pgtable_32.h
@@ -98,7 +98,7 @@
 #define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_FILE)
 #endif
 
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+#define _PAGE_FLAGS_HARDWARE_MASK	(PHYS_ADDR_MASK & ~(_PAGE_CLEAR_FLAGS))
 
 /* Hardware flags, page size encoding */
 #if defined(CONFIG_X2TLB)
diff --git a/include/asm-sh/scatterlist.h b/include/asm-sh/scatterlist.h
index a7d0d1856a99..35ef9c30f0de 100644
--- a/include/asm-sh/scatterlist.h
+++ b/include/asm-sh/scatterlist.h
@@ -1,6 +1,7 @@
 #ifndef __ASM_SH_SCATTERLIST_H
 #define __ASM_SH_SCATTERLIST_H
 
+#include <asm/pgtable.h>
 #include <asm/types.h>
 
 struct scatterlist {
@@ -13,7 +14,7 @@ struct scatterlist {
     unsigned int length;
 };
 
-#define ISA_DMA_THRESHOLD (0x1fffffff)
+#define ISA_DMA_THRESHOLD	PHYS_ADDR_MASK
 
 /* These macros should be used after a pci_map_sg call has been done
  * to get bus addresses of each of the SG entries and their lengths.
-- 
cgit v1.2.3


From fc55888f83c1c0ac09abe4680f9a94fc0662677f Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 18:05:18 +0900
Subject: sh: Document PTEL 31:29 use on PTEA-wielding parts.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pgtable_32.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/pgtable_32.h b/include/asm-sh/pgtable_32.h
index 7efc95404c64..3e3557c53c55 100644
--- a/include/asm-sh/pgtable_32.h
+++ b/include/asm-sh/pgtable_32.h
@@ -22,8 +22,9 @@
  * - Bits 10 and 11 are low bits of the PPN that are reserved on >= 4K pages.
  *   Bit 10 is used for _PAGE_ACCESSED, bit 11 remains unused.
  *
- * - Bits 31, 30, and 29 remain unused by everyone and can be used for future
- *   software flags, although care must be taken to update _PAGE_CLEAR_FLAGS.
+ * - On 29 bit platforms, bits 31 to 29 are used for the space attributes
+ *   and timing control which (together with bit 0) are moved into the
+ *   old-style PTEA on the parts that support it.
  *
  * XXX: Leave the _PAGE_FILE and _PAGE_WT overhaul for a rainy day.
  *
-- 
cgit v1.2.3


From 0fb19dcb64ec4bd9934eee26ce66417fe028ffd8 Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 18:16:23 +0900
Subject: sh: get_user fixes and nommu consolidation.

When a get_user(to, from++) is called the pointer increment is performed
after its first usage, in the specific after the __add_ok invokation.
This causes a wrong get_user return value, putting a wrong character
in the destination variable. This patch solves the problem using a new
temporary pointer.

Additionally this reworks the use of the register banks, allowing for
consolidation between the MMU and nommu implementations.

Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Signed-off-by: Giuseppe Condorelli <giuseppe.condorelli@st.com>
Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/uaccess_32.h | 253 +++++++++++++++++---------------------------
 1 file changed, 98 insertions(+), 155 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/uaccess_32.h b/include/asm-sh/uaccess_32.h
index f18a1a5c95c0..59a9f20c2dc7 100644
--- a/include/asm-sh/uaccess_32.h
+++ b/include/asm-sh/uaccess_32.h
@@ -73,37 +73,24 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 /*
  * __access_ok: Check if address with size is OK or not.
  *
- * We do three checks:
- * (1) is it user space?
- * (2) addr + size --> carry?
- * (3) addr + size >= 0x80000000  (PAGE_OFFSET)
+ * Uhhuh, this needs 33-bit arithmetic. We have a carry..
  *
- * (1) (2) (3) | RESULT
- *  0   0   0  |  ok
- *  0   0   1  |  ok
- *  0   1   0  |  bad
- *  0   1   1  |  bad
- *  1   0   0  |  ok
- *  1   0   1  |  bad
- *  1   1   0  |  bad
- *  1   1   1  |  bad
+ * sum := addr + size;  carry? --> flag = true;
+ * if (sum >= addr_limit) flag = true;
  */
 static inline int __access_ok(unsigned long addr, unsigned long size)
 {
-	unsigned long flag, tmp;
-
-	__asm__("stc	r7_bank, %0\n\t"
-		"mov.l	@(8,%0), %0\n\t"
-		"clrt\n\t"
-		"addc	%2, %1\n\t"
-		"and	%1, %0\n\t"
-		"rotcl	%0\n\t"
-		"rotcl	%0\n\t"
-		"and	#3, %0"
-		: "=&z" (flag), "=r" (tmp)
-		: "r" (addr), "1" (size)
-		: "t");
-
+	unsigned long flag, sum;
+
+	__asm__("clrt\n\t"
+		"addc	%3, %1\n\t"
+		"movt	%0\n\t"
+		"cmp/hi	%4, %1\n\t"
+		"rotcl	%0"
+		:"=&r" (flag), "=r" (sum)
+		:"1" (addr), "r" (size),
+		 "r" (current_thread_info()->addr_limit.seg)
+		:"t");
 	return flag == 0;
 }
 #endif /* CONFIG_MMU */
@@ -165,135 +152,47 @@ do {								\
 #define __get_user_nocheck(x,ptr,size)				\
 ({								\
 	long __gu_err, __gu_val;				\
-	__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
+	__typeof__(*(ptr)) *__pu_addr = (ptr);  \
+	__get_user_size(__gu_val, (__pu_addr), (size), __gu_err);	\
 	(x) = (__typeof__(*(ptr)))__gu_val;			\
 	__gu_err;						\
 })
 
-#ifdef CONFIG_MMU
-#define __get_user_check(x,ptr,size)				\
-({								\
-	long __gu_err, __gu_val;				\
-	__chk_user_ptr(ptr);					\
-	switch (size) {						\
-	case 1:							\
-		__get_user_1(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	case 2:							\
-		__get_user_2(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	case 4:							\
-		__get_user_4(__gu_val, (ptr), __gu_err);	\
-		break;						\
-	default:						\
-		__get_user_unknown();				\
-		break;						\
-	}							\
-								\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
-	__gu_err;						\
-})
-
-#define __get_user_1(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.b	@%2, %1\n\t"			\
-	"extu.b	%1, %1\n"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-
-#define __get_user_2(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.w	@%2, %1\n\t"			\
-	"extu.w	%1, %1\n"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-
-#define __get_user_4(x,addr,err) ({		\
-__asm__("stc	r7_bank, %1\n\t"		\
-	"mov.l	@(8,%1), %1\n\t"		\
-	"and	%2, %1\n\t"			\
-	"cmp/pz	%1\n\t"				\
-	"bt/s	1f\n\t"				\
-	" mov	#0, %0\n\t"			\
-	"0:\n"					\
-	"mov	#-14, %0\n\t"			\
-	"bra	2f\n\t"				\
-	" mov	#0, %1\n"			\
-	"1:\n\t"				\
-	"mov.l	@%2, %1\n\t"			\
-	"2:\n"					\
-	".section	__ex_table,\"a\"\n\t"	\
-	".long	1b, 0b\n\t"			\
-	".previous"				\
-	: "=&r" (err), "=&r" (x)		\
-	: "r" (addr)				\
-	: "t");					\
-})
-#else /* CONFIG_MMU */
 #define __get_user_check(x,ptr,size)					\
 ({									\
 	long __gu_err, __gu_val;					\
-	if (__access_ok((unsigned long)(ptr), (size))) {		\
-		__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
-		(x) = (__typeof__(*(ptr)))__gu_val;			\
-	} else								\
+	__typeof__(*(ptr)) *__pu_addr = (ptr);				\
+	__chk_user_ptr(__pu_addr);					\
+	if (likely(__addr_ok((unsigned long)(__pu_addr)))) {		\
+		__get_user_size(__gu_val, (__pu_addr), (size), __gu_err);\
+	} else {							\
 		__gu_err = -EFAULT;					\
+		__gu_val = 0;						\
+	}								\
+	(x) = (__typeof__(*(ptr)))__gu_val;				\
 	__gu_err;							\
 })
-#endif
 
 #define __get_user_asm(x, addr, err, insn) \
 ({ \
 __asm__ __volatile__( \
 	"1:\n\t" \
 	"mov." insn "	%2, %1\n\t" \
-	"mov	#0, %0\n" \
 	"2:\n" \
 	".section	.fixup,\"ax\"\n" \
 	"3:\n\t" \
 	"mov	#0, %1\n\t" \
 	"mov.l	4f, %0\n\t" \
 	"jmp	@%0\n\t" \
-	" mov	%3, %0\n" \
+	" mov	%3, %0\n\t" \
+	".balign	4\n" \
 	"4:	.long	2b\n\t" \
 	".previous\n" \
 	".section	__ex_table,\"a\"\n\t" \
 	".long	1b, 3b\n\t" \
 	".previous" \
 	:"=&r" (err), "=&r" (x) \
-	:"m" (__m(addr)), "i" (-EFAULT)); })
+	:"m" (__m(addr)), "i" (-EFAULT), "0" (err)); })
 
 extern void __get_user_unknown(void);
 
@@ -328,11 +227,13 @@ do {							\
 
 #define __put_user_check(x,ptr,size)				\
 ({								\
-	long __pu_err = -EFAULT;				\
+	long __pu_err;						\
 	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
 								\
-	if (__access_ok((unsigned long)__pu_addr,size))		\
+	if (likely(__addr_ok((unsigned long)__pu_addr)))	\
 		__put_user_size((x),__pu_addr,(size),__pu_err);	\
+	else							\
+		__pu_err = -EFAULT;				\
 	__pu_err;						\
 })
 
@@ -341,45 +242,43 @@ do {							\
 __asm__ __volatile__( \
 	"1:\n\t" \
 	"mov." insn "	%1, %2\n\t" \
-	"mov	#0, %0\n" \
 	"2:\n" \
 	".section	.fixup,\"ax\"\n" \
 	"3:\n\t" \
-	"nop\n\t" \
 	"mov.l	4f, %0\n\t" \
 	"jmp	@%0\n\t" \
-	"mov	%3, %0\n" \
+	" mov	%3, %0\n\t" \
+	".balign	4\n" \
 	"4:	.long	2b\n\t" \
 	".previous\n" \
 	".section	__ex_table,\"a\"\n\t" \
 	".long	1b, 3b\n\t" \
 	".previous" \
 	:"=&r" (err) \
-	:"r" (x), "m" (__m(addr)), "i" (-EFAULT) \
+	:"r" (x), "m" (__m(addr)), "i" (-EFAULT), "0" (err)	\
         :"memory"); })
 
-#if defined(__LITTLE_ENDIAN__)
+#if defined(CONFIG_CPU_LITTLE_ENDIAN)
 #define __put_user_u64(val,addr,retval) \
 ({ \
 __asm__ __volatile__( \
 	"1:\n\t" \
 	"mov.l	%R1,%2\n\t" \
 	"mov.l	%S1,%T2\n\t" \
-	"mov	#0,%0\n" \
 	"2:\n" \
 	".section	.fixup,\"ax\"\n" \
 	"3:\n\t" \
-	"nop\n\t" \
 	"mov.l	4f,%0\n\t" \
 	"jmp	@%0\n\t" \
-	" mov	%3,%0\n" \
+	" mov	%3,%0\n\t" \
+	".balign	4\n" \
 	"4:	.long	2b\n\t" \
 	".previous\n" \
 	".section	__ex_table,\"a\"\n\t" \
 	".long	1b, 3b\n\t" \
 	".previous" \
 	: "=r" (retval) \
-	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
+	: "r" (val), "m" (__m(addr)), "i" (-EFAULT), "0" (retval) \
         : "memory"); })
 #else
 #define __put_user_u64(val,addr,retval) \
@@ -388,21 +287,20 @@ __asm__ __volatile__( \
 	"1:\n\t" \
 	"mov.l	%S1,%2\n\t" \
 	"mov.l	%R1,%T2\n\t" \
-	"mov	#0,%0\n" \
 	"2:\n" \
 	".section	.fixup,\"ax\"\n" \
 	"3:\n\t" \
-	"nop\n\t" \
 	"mov.l	4f,%0\n\t" \
 	"jmp	@%0\n\t" \
-	" mov	%3,%0\n" \
+	" mov	%3,%0\n\t" \
+	".balign	4\n" \
 	"4:	.long	2b\n\t" \
 	".previous\n" \
 	".section	__ex_table,\"a\"\n\t" \
 	".long	1b, 3b\n\t" \
 	".previous" \
 	: "=r" (retval) \
-	: "r" (val), "m" (__m(addr)), "i" (-EFAULT) \
+	: "r" (val), "m" (__m(addr)), "i" (-EFAULT), "0" (retval) \
         : "memory"); })
 #endif
 
@@ -463,7 +361,7 @@ static __inline__ int
 __strncpy_from_user(unsigned long __dest, unsigned long __user __src, int __count)
 {
 	__kernel_size_t res;
-	unsigned long __dummy, _d, _s;
+	unsigned long __dummy, _d, _s, _c;
 
 	__asm__ __volatile__(
 		"9:\n"
@@ -472,17 +370,17 @@ __strncpy_from_user(unsigned long __dest, unsigned long __user __src, int __coun
 		"bt/s	2f\n"
 		"1:\n"
 		"mov.b	%1, @%3\n\t"
-		"dt	%7\n\t"
+		"dt	%4\n\t"
 		"bf/s	9b\n\t"
 		" add	#1, %3\n\t"
 		"2:\n\t"
-		"sub	%7, %0\n"
+		"sub	%4, %0\n"
 		"3:\n"
 		".section .fixup,\"ax\"\n"
 		"4:\n\t"
 		"mov.l	5f, %1\n\t"
 		"jmp	@%1\n\t"
-		" mov	%8, %0\n\t"
+		" mov	%9, %0\n\t"
 		".balign 4\n"
 		"5:	.long 3b\n"
 		".previous\n"
@@ -490,14 +388,32 @@ __strncpy_from_user(unsigned long __dest, unsigned long __user __src, int __coun
 		"	.balign 4\n"
 		"	.long 9b,4b\n"
 		".previous"
-		: "=r" (res), "=&z" (__dummy), "=r" (_s), "=r" (_d)
-		: "0" (__count), "2" (__src), "3" (__dest), "r" (__count),
+		: "=r" (res), "=&z" (__dummy), "=r" (_s), "=r" (_d), "=r"(_c)
+		: "0" (__count), "2" (__src), "3" (__dest), "4" (__count),
 		  "i" (-EFAULT)
 		: "memory", "t");
 
 	return res;
 }
 
+/**
+ * strncpy_from_user: - Copy a NUL terminated string from userspace.
+ * @dst:   Destination address, in kernel space.  This buffer must be at
+ *         least @count bytes long.
+ * @src:   Source address, in user space.
+ * @count: Maximum number of bytes to copy, including the trailing NUL.
+ *
+ * Copies a NUL-terminated string from userspace to kernel space.
+ *
+ * On success, returns the length of the string (not including the trailing
+ * NUL).
+ *
+ * If access to userspace fails, returns -EFAULT (some data may have been
+ * copied).
+ *
+ * If @count is smaller than the length of the string, copies @count bytes
+ * and returns @count.
+ */
 #define strncpy_from_user(dest,src,count) ({ \
 unsigned long __sfu_src = (unsigned long) (src); \
 int __sfu_count = (int) (count); \
@@ -507,7 +423,8 @@ __sfu_res = __strncpy_from_user((unsigned long) (dest), __sfu_src, __sfu_count);
 } __sfu_res; })
 
 /*
- * Return the size of a string (including the ending 0!)
+ * Return the size of a string (including the ending 0 even when we have
+ * exceeded the maximum string length).
  */
 static __inline__ long __strnlen_user(const char __user *__s, long __n)
 {
@@ -515,14 +432,13 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 	unsigned long __dummy;
 
 	__asm__ __volatile__(
-		"9:\n"
-		"cmp/eq	%4, %0\n\t"
-		"bt	2f\n"
 		"1:\t"
 		"mov.b	@(%0,%3), %1\n\t"
+		"cmp/eq	%4, %0\n\t"
+		"bt/s	2f\n\t"
+		" add	#1, %0\n\t"
 		"tst	%1, %1\n\t"
-		"bf/s	9b\n\t"
-		" add	#1, %0\n"
+		"bf	1b\n\t"
 		"2:\n"
 		".section .fixup,\"ax\"\n"
 		"3:\n\t"
@@ -542,6 +458,19 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 	return res;
 }
 
+/**
+ * strnlen_user: - Get the size of a string in user space.
+ * @s: The string to measure.
+ * @n: The maximum valid length
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * Get the size of a NUL-terminated string in user space.
+ *
+ * Returns the size of the string INCLUDING the terminating NUL.
+ * On exception, returns 0.
+ * If the string is too long, returns a value greater than @n.
+ */
 static __inline__ long strnlen_user(const char __user *s, long n)
 {
 	if (!__addr_ok(s))
@@ -550,6 +479,20 @@ static __inline__ long strnlen_user(const char __user *s, long n)
 		return __strnlen_user(s, n);
 }
 
+/**
+ * strlen_user: - Get the size of a string in user space.
+ * @str: The string to measure.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * Get the size of a NUL-terminated string in user space.
+ *
+ * Returns the size of the string INCLUDING the terminating NUL.
+ * On exception, returns 0.
+ *
+ * If there is a limit on the length of a valid string, you may wish to
+ * consider using strnlen_user() instead.
+ */
 #define strlen_user(str)	strnlen_user(str, ~0UL >> 1)
 
 /*
-- 
cgit v1.2.3


From c8c0a1aba9fa8f816dc8fb477ff816a5b700f0ea Mon Sep 17 00:00:00 2001
From: Stuart Menefy <stuart.menefy@st.com>
Date: Fri, 30 Nov 2007 18:42:27 +0900
Subject: sh: Support denormalization on SH-4 FPU.

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/Makefile    |   2 +-
 arch/sh/kernel/cpu/sh4/fpu.c       | 514 +++++++++++++--------
 arch/sh/kernel/cpu/sh4/softfloat.c | 892 +++++++++++++++++++++++++++++++++++++
 include/asm-sh/cpu-sh4/fpu.h       |  32 ++
 4 files changed, 1259 insertions(+), 181 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh4/softfloat.c
 create mode 100644 include/asm-sh/cpu-sh4/fpu.h

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index dadd6bffc128..d608557c7a3f 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -5,7 +5,7 @@
 obj-y	:= probe.o common.o
 common-y	+= $(addprefix ../sh3/, entry.o ex.o)
 
-obj-$(CONFIG_SH_FPU)			+= fpu.o
+obj-$(CONFIG_SH_FPU)			+= fpu.o softfloat.o
 obj-$(CONFIG_SH_STORE_QUEUES)		+= sq.o
 
 # CPU subtype setup
diff --git a/arch/sh/kernel/cpu/sh4/fpu.c b/arch/sh/kernel/cpu/sh4/fpu.c
index e624180b4467..817f9939cda6 100644
--- a/arch/sh/kernel/cpu/sh4/fpu.c
+++ b/arch/sh/kernel/cpu/sh4/fpu.c
@@ -1,7 +1,4 @@
-/* $Id: fpu.c,v 1.4 2004/01/13 05:52:11 kkojima Exp $
- *
- * linux/arch/sh/kernel/fpu.c
- *
+/*
  * Save/restore floating point context for signal handlers.
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -9,15 +6,16 @@
  * for more details.
  *
  * Copyright (C) 1999, 2000  Kaz Kojima & Niibe Yutaka
+ * Copyright (C) 2006  ST Microelectronics Ltd. (denorm support)
  *
- * FIXME! These routines can be optimized in big endian case.
+ * FIXME! These routines have not been tested for big endian case.
  */
-
 #include <linux/sched.h>
 #include <linux/signal.h>
+#include <linux/io.h>
+#include <asm/cpu/fpu.h>
 #include <asm/processor.h>
 #include <asm/system.h>
-#include <asm/io.h>
 
 /* The PR (precision) bit in the FP Status Register must be clear when
  * an frchg instruction is executed, otherwise the instruction is undefined.
@@ -25,113 +23,122 @@
  */
 
 #define FPSCR_RCHG 0x00000000
+extern unsigned long long float64_div(unsigned long long a,
+				      unsigned long long b);
+extern unsigned long int float32_div(unsigned long int a, unsigned long int b);
+extern unsigned long long float64_mul(unsigned long long a,
+				      unsigned long long b);
+extern unsigned long int float32_mul(unsigned long int a, unsigned long int b);
+extern unsigned long long float64_add(unsigned long long a,
+				      unsigned long long b);
+extern unsigned long int float32_add(unsigned long int a, unsigned long int b);
+extern unsigned long long float64_sub(unsigned long long a,
+				      unsigned long long b);
+extern unsigned long int float32_sub(unsigned long int a, unsigned long int b);
 
+static unsigned int fpu_exception_flags;
 
 /*
  * Save FPU registers onto task structure.
  * Assume called with FPU enabled (SR.FD=0).
  */
-void
-save_fpu(struct task_struct *tsk, struct pt_regs *regs)
+void save_fpu(struct task_struct *tsk, struct pt_regs *regs)
 {
 	unsigned long dummy;
 
 	clear_tsk_thread_flag(tsk, TIF_USEDFPU);
 	enable_fpu();
-	asm volatile("sts.l	fpul, @-%0\n\t"
-		     "sts.l	fpscr, @-%0\n\t"
-		     "lds	%2, fpscr\n\t"
-		     "frchg\n\t"
-		     "fmov.s	fr15, @-%0\n\t"
-		     "fmov.s	fr14, @-%0\n\t"
-		     "fmov.s	fr13, @-%0\n\t"
-		     "fmov.s	fr12, @-%0\n\t"
-		     "fmov.s	fr11, @-%0\n\t"
-		     "fmov.s	fr10, @-%0\n\t"
-		     "fmov.s	fr9, @-%0\n\t"
-		     "fmov.s	fr8, @-%0\n\t"
-		     "fmov.s	fr7, @-%0\n\t"
-		     "fmov.s	fr6, @-%0\n\t"
-		     "fmov.s	fr5, @-%0\n\t"
-		     "fmov.s	fr4, @-%0\n\t"
-		     "fmov.s	fr3, @-%0\n\t"
-		     "fmov.s	fr2, @-%0\n\t"
-		     "fmov.s	fr1, @-%0\n\t"
-		     "fmov.s	fr0, @-%0\n\t"
-		     "frchg\n\t"
-		     "fmov.s	fr15, @-%0\n\t"
-		     "fmov.s	fr14, @-%0\n\t"
-		     "fmov.s	fr13, @-%0\n\t"
-		     "fmov.s	fr12, @-%0\n\t"
-		     "fmov.s	fr11, @-%0\n\t"
-		     "fmov.s	fr10, @-%0\n\t"
-		     "fmov.s	fr9, @-%0\n\t"
-		     "fmov.s	fr8, @-%0\n\t"
-		     "fmov.s	fr7, @-%0\n\t"
-		     "fmov.s	fr6, @-%0\n\t"
-		     "fmov.s	fr5, @-%0\n\t"
-		     "fmov.s	fr4, @-%0\n\t"
-		     "fmov.s	fr3, @-%0\n\t"
-		     "fmov.s	fr2, @-%0\n\t"
-		     "fmov.s	fr1, @-%0\n\t"
-		     "fmov.s	fr0, @-%0\n\t"
-		     "lds	%3, fpscr\n\t"
-		     : "=r" (dummy)
-		     : "0" ((char *)(&tsk->thread.fpu.hard.status)),
-		       "r" (FPSCR_RCHG),
-		       "r" (FPSCR_INIT)
-		     : "memory");
+	asm volatile ("sts.l	fpul, @-%0\n\t"
+		      "sts.l	fpscr, @-%0\n\t"
+		      "lds	%2, fpscr\n\t"
+		      "frchg\n\t"
+		      "fmov.s	fr15, @-%0\n\t"
+		      "fmov.s	fr14, @-%0\n\t"
+		      "fmov.s	fr13, @-%0\n\t"
+		      "fmov.s	fr12, @-%0\n\t"
+		      "fmov.s	fr11, @-%0\n\t"
+		      "fmov.s	fr10, @-%0\n\t"
+		      "fmov.s	fr9, @-%0\n\t"
+		      "fmov.s	fr8, @-%0\n\t"
+		      "fmov.s	fr7, @-%0\n\t"
+		      "fmov.s	fr6, @-%0\n\t"
+		      "fmov.s	fr5, @-%0\n\t"
+		      "fmov.s	fr4, @-%0\n\t"
+		      "fmov.s	fr3, @-%0\n\t"
+		      "fmov.s	fr2, @-%0\n\t"
+		      "fmov.s	fr1, @-%0\n\t"
+		      "fmov.s	fr0, @-%0\n\t"
+		      "frchg\n\t"
+		      "fmov.s	fr15, @-%0\n\t"
+		      "fmov.s	fr14, @-%0\n\t"
+		      "fmov.s	fr13, @-%0\n\t"
+		      "fmov.s	fr12, @-%0\n\t"
+		      "fmov.s	fr11, @-%0\n\t"
+		      "fmov.s	fr10, @-%0\n\t"
+		      "fmov.s	fr9, @-%0\n\t"
+		      "fmov.s	fr8, @-%0\n\t"
+		      "fmov.s	fr7, @-%0\n\t"
+		      "fmov.s	fr6, @-%0\n\t"
+		      "fmov.s	fr5, @-%0\n\t"
+		      "fmov.s	fr4, @-%0\n\t"
+		      "fmov.s	fr3, @-%0\n\t"
+		      "fmov.s	fr2, @-%0\n\t"
+		      "fmov.s	fr1, @-%0\n\t"
+		      "fmov.s	fr0, @-%0\n\t"
+		      "lds	%3, fpscr\n\t":"=r" (dummy)
+		      :"0"((char *)(&tsk->thread.fpu.hard.status)),
+		      "r"(FPSCR_RCHG), "r"(FPSCR_INIT)
+		      :"memory");
 
 	disable_fpu();
 	release_fpu(regs);
 }
 
-static void
-restore_fpu(struct task_struct *tsk)
+static void restore_fpu(struct task_struct *tsk)
 {
 	unsigned long dummy;
 
 	enable_fpu();
-	asm volatile("lds	%2, fpscr\n\t"
-		     "fmov.s	@%0+, fr0\n\t"
-		     "fmov.s	@%0+, fr1\n\t"
-		     "fmov.s	@%0+, fr2\n\t"
-		     "fmov.s	@%0+, fr3\n\t"
-		     "fmov.s	@%0+, fr4\n\t"
-		     "fmov.s	@%0+, fr5\n\t"
-		     "fmov.s	@%0+, fr6\n\t"
-		     "fmov.s	@%0+, fr7\n\t"
-		     "fmov.s	@%0+, fr8\n\t"
-		     "fmov.s	@%0+, fr9\n\t"
-		     "fmov.s	@%0+, fr10\n\t"
-		     "fmov.s	@%0+, fr11\n\t"
-		     "fmov.s	@%0+, fr12\n\t"
-		     "fmov.s	@%0+, fr13\n\t"
-		     "fmov.s	@%0+, fr14\n\t"
-		     "fmov.s	@%0+, fr15\n\t"
-		     "frchg\n\t"
-		     "fmov.s	@%0+, fr0\n\t"
-		     "fmov.s	@%0+, fr1\n\t"
-		     "fmov.s	@%0+, fr2\n\t"
-		     "fmov.s	@%0+, fr3\n\t"
-		     "fmov.s	@%0+, fr4\n\t"
-		     "fmov.s	@%0+, fr5\n\t"
-		     "fmov.s	@%0+, fr6\n\t"
-		     "fmov.s	@%0+, fr7\n\t"
-		     "fmov.s	@%0+, fr8\n\t"
-		     "fmov.s	@%0+, fr9\n\t"
-		     "fmov.s	@%0+, fr10\n\t"
-		     "fmov.s	@%0+, fr11\n\t"
-		     "fmov.s	@%0+, fr12\n\t"
-		     "fmov.s	@%0+, fr13\n\t"
-		     "fmov.s	@%0+, fr14\n\t"
-		     "fmov.s	@%0+, fr15\n\t"
-		     "frchg\n\t"
-		     "lds.l	@%0+, fpscr\n\t"
-		     "lds.l	@%0+, fpul\n\t"
-		     : "=r" (dummy)
-		     : "0" (&tsk->thread.fpu), "r" (FPSCR_RCHG)
-		     : "memory");
+	asm volatile ("lds	%2, fpscr\n\t"
+		      "fmov.s	@%0+, fr0\n\t"
+		      "fmov.s	@%0+, fr1\n\t"
+		      "fmov.s	@%0+, fr2\n\t"
+		      "fmov.s	@%0+, fr3\n\t"
+		      "fmov.s	@%0+, fr4\n\t"
+		      "fmov.s	@%0+, fr5\n\t"
+		      "fmov.s	@%0+, fr6\n\t"
+		      "fmov.s	@%0+, fr7\n\t"
+		      "fmov.s	@%0+, fr8\n\t"
+		      "fmov.s	@%0+, fr9\n\t"
+		      "fmov.s	@%0+, fr10\n\t"
+		      "fmov.s	@%0+, fr11\n\t"
+		      "fmov.s	@%0+, fr12\n\t"
+		      "fmov.s	@%0+, fr13\n\t"
+		      "fmov.s	@%0+, fr14\n\t"
+		      "fmov.s	@%0+, fr15\n\t"
+		      "frchg\n\t"
+		      "fmov.s	@%0+, fr0\n\t"
+		      "fmov.s	@%0+, fr1\n\t"
+		      "fmov.s	@%0+, fr2\n\t"
+		      "fmov.s	@%0+, fr3\n\t"
+		      "fmov.s	@%0+, fr4\n\t"
+		      "fmov.s	@%0+, fr5\n\t"
+		      "fmov.s	@%0+, fr6\n\t"
+		      "fmov.s	@%0+, fr7\n\t"
+		      "fmov.s	@%0+, fr8\n\t"
+		      "fmov.s	@%0+, fr9\n\t"
+		      "fmov.s	@%0+, fr10\n\t"
+		      "fmov.s	@%0+, fr11\n\t"
+		      "fmov.s	@%0+, fr12\n\t"
+		      "fmov.s	@%0+, fr13\n\t"
+		      "fmov.s	@%0+, fr14\n\t"
+		      "fmov.s	@%0+, fr15\n\t"
+		      "frchg\n\t"
+		      "lds.l	@%0+, fpscr\n\t"
+		      "lds.l	@%0+, fpul\n\t"
+		      :"=r" (dummy)
+		      :"0"(&tsk->thread.fpu), "r"(FPSCR_RCHG)
+		      :"memory");
 	disable_fpu();
 }
 
@@ -141,61 +148,59 @@ restore_fpu(struct task_struct *tsk)
  * double precision represents signaling NANS.
  */
 
-static void
-fpu_init(void)
+static void fpu_init(void)
 {
 	enable_fpu();
-	asm volatile("lds	%0, fpul\n\t"
-		     "lds	%1, fpscr\n\t"
-		     "fsts	fpul, fr0\n\t"
-		     "fsts	fpul, fr1\n\t"
-		     "fsts	fpul, fr2\n\t"
-		     "fsts	fpul, fr3\n\t"
-		     "fsts	fpul, fr4\n\t"
-		     "fsts	fpul, fr5\n\t"
-		     "fsts	fpul, fr6\n\t"
-		     "fsts	fpul, fr7\n\t"
-		     "fsts	fpul, fr8\n\t"
-		     "fsts	fpul, fr9\n\t"
-		     "fsts	fpul, fr10\n\t"
-		     "fsts	fpul, fr11\n\t"
-		     "fsts	fpul, fr12\n\t"
-		     "fsts	fpul, fr13\n\t"
-		     "fsts	fpul, fr14\n\t"
-		     "fsts	fpul, fr15\n\t"
-		     "frchg\n\t"
-		     "fsts	fpul, fr0\n\t"
-		     "fsts	fpul, fr1\n\t"
-		     "fsts	fpul, fr2\n\t"
-		     "fsts	fpul, fr3\n\t"
-		     "fsts	fpul, fr4\n\t"
-		     "fsts	fpul, fr5\n\t"
-		     "fsts	fpul, fr6\n\t"
-		     "fsts	fpul, fr7\n\t"
-		     "fsts	fpul, fr8\n\t"
-		     "fsts	fpul, fr9\n\t"
-		     "fsts	fpul, fr10\n\t"
-		     "fsts	fpul, fr11\n\t"
-		     "fsts	fpul, fr12\n\t"
-		     "fsts	fpul, fr13\n\t"
-		     "fsts	fpul, fr14\n\t"
-		     "fsts	fpul, fr15\n\t"
-		     "frchg\n\t"
-		     "lds	%2, fpscr\n\t"
-		     : /* no output */
-		     : "r" (0), "r" (FPSCR_RCHG), "r" (FPSCR_INIT));
+	asm volatile (	"lds	%0, fpul\n\t"
+			"lds	%1, fpscr\n\t"
+			"fsts	fpul, fr0\n\t"
+			"fsts	fpul, fr1\n\t"
+			"fsts	fpul, fr2\n\t"
+			"fsts	fpul, fr3\n\t"
+			"fsts	fpul, fr4\n\t"
+			"fsts	fpul, fr5\n\t"
+			"fsts	fpul, fr6\n\t"
+			"fsts	fpul, fr7\n\t"
+			"fsts	fpul, fr8\n\t"
+			"fsts	fpul, fr9\n\t"
+			"fsts	fpul, fr10\n\t"
+			"fsts	fpul, fr11\n\t"
+			"fsts	fpul, fr12\n\t"
+			"fsts	fpul, fr13\n\t"
+			"fsts	fpul, fr14\n\t"
+			"fsts	fpul, fr15\n\t"
+			"frchg\n\t"
+			"fsts	fpul, fr0\n\t"
+			"fsts	fpul, fr1\n\t"
+			"fsts	fpul, fr2\n\t"
+			"fsts	fpul, fr3\n\t"
+			"fsts	fpul, fr4\n\t"
+			"fsts	fpul, fr5\n\t"
+			"fsts	fpul, fr6\n\t"
+			"fsts	fpul, fr7\n\t"
+			"fsts	fpul, fr8\n\t"
+			"fsts	fpul, fr9\n\t"
+			"fsts	fpul, fr10\n\t"
+			"fsts	fpul, fr11\n\t"
+			"fsts	fpul, fr12\n\t"
+			"fsts	fpul, fr13\n\t"
+			"fsts	fpul, fr14\n\t"
+			"fsts	fpul, fr15\n\t"
+			"frchg\n\t"
+			"lds	%2, fpscr\n\t"
+			:	/* no output */
+			:"r" (0), "r"(FPSCR_RCHG), "r"(FPSCR_INIT));
 	disable_fpu();
 }
 
 /**
- *	denormal_to_double - Given denormalized float number,
- *	                     store double float
+ *      denormal_to_double - Given denormalized float number,
+ *                           store double float
  *
- *	@fpu: Pointer to sh_fpu_hard structure
- *	@n: Index to FP register
+ *      @fpu: Pointer to sh_fpu_hard structure
+ *      @n: Index to FP register
  */
-static void
-denormal_to_double (struct sh_fpu_hard_struct *fpu, int n)
+static void denormal_to_double(struct sh_fpu_hard_struct *fpu, int n)
 {
 	unsigned long du, dl;
 	unsigned long x = fpu->fpul;
@@ -212,7 +217,7 @@ denormal_to_double (struct sh_fpu_hard_struct *fpu, int n)
 		dl = x << 29;
 
 		fpu->fp_regs[n] = du;
-		fpu->fp_regs[n+1] = dl;
+		fpu->fp_regs[n + 1] = dl;
 	}
 }
 
@@ -223,67 +228,191 @@ denormal_to_double (struct sh_fpu_hard_struct *fpu, int n)
  *
  *	Returns 1 when it's handled (should not cause exception).
  */
-static int
-ieee_fpe_handler (struct pt_regs *regs)
+static int ieee_fpe_handler(struct pt_regs *regs)
 {
-	unsigned short insn = *(unsigned short *) regs->pc;
+	unsigned short insn = *(unsigned short *)regs->pc;
 	unsigned short finsn;
 	unsigned long nextpc;
 	int nib[4] = {
 		(insn >> 12) & 0xf,
 		(insn >> 8) & 0xf,
 		(insn >> 4) & 0xf,
-		insn & 0xf};
-
-	if (nib[0] == 0xb ||
-	    (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb)) /* bsr & jsr */
-		regs->pr = regs->pc + 4;
-	if (nib[0] == 0xa || nib[0] == 0xb) { /* bra & bsr */
-		nextpc = regs->pc + 4 + ((short) ((insn & 0xfff) << 4) >> 3);
-		finsn = *(unsigned short *) (regs->pc + 2);
-	} else if (nib[0] == 0x8 && nib[1] == 0xd) { /* bt/s */
+		insn & 0xf
+	};
+
+	if (nib[0] == 0xb || (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb))
+		regs->pr = regs->pc + 4;  /* bsr & jsr */
+
+	if (nib[0] == 0xa || nib[0] == 0xb) {
+		/* bra & bsr */
+		nextpc = regs->pc + 4 + ((short)((insn & 0xfff) << 4) >> 3);
+		finsn = *(unsigned short *)(regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xd) {
+		/* bt/s */
 		if (regs->sr & 1)
-			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+			nextpc = regs->pc + 4 + ((char)(insn & 0xff) << 1);
 		else
 			nextpc = regs->pc + 4;
-		finsn = *(unsigned short *) (regs->pc + 2);
-	} else if (nib[0] == 0x8 && nib[1] == 0xf) { /* bf/s */
+		finsn = *(unsigned short *)(regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xf) {
+		/* bf/s */
 		if (regs->sr & 1)
 			nextpc = regs->pc + 4;
 		else
-			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
-		finsn = *(unsigned short *) (regs->pc + 2);
+			nextpc = regs->pc + 4 + ((char)(insn & 0xff) << 1);
+		finsn = *(unsigned short *)(regs->pc + 2);
 	} else if (nib[0] == 0x4 && nib[3] == 0xb &&
-		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* jmp & jsr */
+		   (nib[2] == 0x0 || nib[2] == 0x2)) {
+		/* jmp & jsr */
 		nextpc = regs->regs[nib[1]];
-		finsn = *(unsigned short *) (regs->pc + 2);
+		finsn = *(unsigned short *)(regs->pc + 2);
 	} else if (nib[0] == 0x0 && nib[3] == 0x3 &&
-		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* braf & bsrf */
+		   (nib[2] == 0x0 || nib[2] == 0x2)) {
+		/* braf & bsrf */
 		nextpc = regs->pc + 4 + regs->regs[nib[1]];
-		finsn = *(unsigned short *) (regs->pc + 2);
-	} else if (insn == 0x000b) { /* rts */
+		finsn = *(unsigned short *)(regs->pc + 2);
+	} else if (insn == 0x000b) {
+		/* rts */
 		nextpc = regs->pr;
-		finsn = *(unsigned short *) (regs->pc + 2);
+		finsn = *(unsigned short *)(regs->pc + 2);
 	} else {
 		nextpc = regs->pc + instruction_size(insn);
 		finsn = insn;
 	}
 
-	if ((finsn & 0xf1ff) == 0xf0ad) { /* fcnvsd */
+	if ((finsn & 0xf1ff) == 0xf0ad) {
+		/* fcnvsd */
 		struct task_struct *tsk = current;
 
 		save_fpu(tsk, regs);
-		if ((tsk->thread.fpu.hard.fpscr & (1 << 17))) {
+		if ((tsk->thread.fpu.hard.fpscr & FPSCR_CAUSE_ERROR))
 			/* FPU error */
-			denormal_to_double (&tsk->thread.fpu.hard,
-					    (finsn >> 8) & 0xf);
-			tsk->thread.fpu.hard.fpscr &=
-				~(FPSCR_CAUSE_MASK | FPSCR_FLAG_MASK);
-			grab_fpu(regs);
-			restore_fpu(tsk);
-			set_tsk_thread_flag(tsk, TIF_USEDFPU);
+			denormal_to_double(&tsk->thread.fpu.hard,
+					   (finsn >> 8) & 0xf);
+		else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	} else if ((finsn & 0xf00f) == 0xf002) {
+		/* fmul */
+		struct task_struct *tsk = current;
+		int fpscr;
+		int n, m, prec;
+		unsigned int hx, hy;
+
+		n = (finsn >> 8) & 0xf;
+		m = (finsn >> 4) & 0xf;
+		hx = tsk->thread.fpu.hard.fp_regs[n];
+		hy = tsk->thread.fpu.hard.fp_regs[m];
+		fpscr = tsk->thread.fpu.hard.fpscr;
+		prec = fpscr & FPSCR_DBL_PRECISION;
+
+		if ((fpscr & FPSCR_CAUSE_ERROR)
+		    && (prec && ((hx & 0x7fffffff) < 0x00100000
+				 || (hy & 0x7fffffff) < 0x00100000))) {
+			long long llx, lly;
+
+			/* FPU error because of denormal (doubles) */
+			llx = ((long long)hx << 32)
+			    | tsk->thread.fpu.hard.fp_regs[n + 1];
+			lly = ((long long)hy << 32)
+			    | tsk->thread.fpu.hard.fp_regs[m + 1];
+			llx = float64_mul(llx, lly);
+			tsk->thread.fpu.hard.fp_regs[n] = llx >> 32;
+			tsk->thread.fpu.hard.fp_regs[n + 1] = llx & 0xffffffff;
+		} else if ((fpscr & FPSCR_CAUSE_ERROR)
+			   && (!prec && ((hx & 0x7fffffff) < 0x00800000
+					 || (hy & 0x7fffffff) < 0x00800000))) {
+			/* FPU error because of denormal (floats) */
+			hx = float32_mul(hx, hy);
+			tsk->thread.fpu.hard.fp_regs[n] = hx;
+		} else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	} else if ((finsn & 0xf00e) == 0xf000) {
+		/* fadd, fsub */
+		struct task_struct *tsk = current;
+		int fpscr;
+		int n, m, prec;
+		unsigned int hx, hy;
+
+		n = (finsn >> 8) & 0xf;
+		m = (finsn >> 4) & 0xf;
+		hx = tsk->thread.fpu.hard.fp_regs[n];
+		hy = tsk->thread.fpu.hard.fp_regs[m];
+		fpscr = tsk->thread.fpu.hard.fpscr;
+		prec = fpscr & FPSCR_DBL_PRECISION;
+
+		if ((fpscr & FPSCR_CAUSE_ERROR)
+		    && (prec && ((hx & 0x7fffffff) < 0x00100000
+				 || (hy & 0x7fffffff) < 0x00100000))) {
+			long long llx, lly;
+
+			/* FPU error because of denormal (doubles) */
+			llx = ((long long)hx << 32)
+			    | tsk->thread.fpu.hard.fp_regs[n + 1];
+			lly = ((long long)hy << 32)
+			    | tsk->thread.fpu.hard.fp_regs[m + 1];
+			if ((finsn & 0xf00f) == 0xf000)
+				llx = float64_add(llx, lly);
+			else
+				llx = float64_sub(llx, lly);
+			tsk->thread.fpu.hard.fp_regs[n] = llx >> 32;
+			tsk->thread.fpu.hard.fp_regs[n + 1] = llx & 0xffffffff;
+		} else if ((fpscr & FPSCR_CAUSE_ERROR)
+			   && (!prec && ((hx & 0x7fffffff) < 0x00800000
+					 || (hy & 0x7fffffff) < 0x00800000))) {
+			/* FPU error because of denormal (floats) */
+			if ((finsn & 0xf00f) == 0xf000)
+				hx = float32_add(hx, hy);
+			else
+				hx = float32_sub(hx, hy);
+			tsk->thread.fpu.hard.fp_regs[n] = hx;
+		} else
+			return 0;
+
+		regs->pc = nextpc;
+		return 1;
+	} else if ((finsn & 0xf003) == 0xf003) {
+		/* fdiv */
+		struct task_struct *tsk = current;
+		int fpscr;
+		int n, m, prec;
+		unsigned int hx, hy;
+
+		n = (finsn >> 8) & 0xf;
+		m = (finsn >> 4) & 0xf;
+		hx = tsk->thread.fpu.hard.fp_regs[n];
+		hy = tsk->thread.fpu.hard.fp_regs[m];
+		fpscr = tsk->thread.fpu.hard.fpscr;
+		prec = fpscr & FPSCR_DBL_PRECISION;
+
+		if ((fpscr & FPSCR_CAUSE_ERROR)
+		    && (prec && ((hx & 0x7fffffff) < 0x00100000
+				 || (hy & 0x7fffffff) < 0x00100000))) {
+			long long llx, lly;
+
+			/* FPU error because of denormal (doubles) */
+			llx = ((long long)hx << 32)
+			    | tsk->thread.fpu.hard.fp_regs[n + 1];
+			lly = ((long long)hy << 32)
+			    | tsk->thread.fpu.hard.fp_regs[m + 1];
+
+			llx = float64_div(llx, lly);
+
+			tsk->thread.fpu.hard.fp_regs[n] = llx >> 32;
+			tsk->thread.fpu.hard.fp_regs[n + 1] = llx & 0xffffffff;
+		} else if ((fpscr & FPSCR_CAUSE_ERROR)
+			   && (!prec && ((hx & 0x7fffffff) < 0x00800000
+					 || (hy & 0x7fffffff) < 0x00800000))) {
+			/* FPU error because of denormal (floats) */
+			hx = float32_div(hx, hy);
+			tsk->thread.fpu.hard.fp_regs[n] = hx;
 		} else
-			force_sig(SIGFPE, tsk);
+			return 0;
 
 		regs->pc = nextpc;
 		return 1;
@@ -292,16 +421,41 @@ ieee_fpe_handler (struct pt_regs *regs)
 	return 0;
 }
 
+void float_raise(unsigned int flags)
+{
+	fpu_exception_flags |= flags;
+}
+
+int float_rounding_mode(void)
+{
+	struct task_struct *tsk = current;
+	int roundingMode = FPSCR_ROUNDING_MODE(tsk->thread.fpu.hard.fpscr);
+	return roundingMode;
+}
+
 BUILD_TRAP_HANDLER(fpu_error)
 {
 	struct task_struct *tsk = current;
 	TRAP_HANDLER_DECL;
 
-	if (ieee_fpe_handler(regs))
-		return;
-
-	regs->pc += 2;
 	save_fpu(tsk, regs);
+	fpu_exception_flags = 0;
+	if (ieee_fpe_handler(regs)) {
+		tsk->thread.fpu.hard.fpscr &=
+		    ~(FPSCR_CAUSE_MASK | FPSCR_FLAG_MASK);
+		tsk->thread.fpu.hard.fpscr |= fpu_exception_flags;
+		/* Set the FPSCR flag as well as cause bits - simply
+		 * replicate the cause */
+		tsk->thread.fpu.hard.fpscr |= (fpu_exception_flags >> 10);
+		grab_fpu(regs);
+		restore_fpu(tsk);
+		set_tsk_thread_flag(tsk, TIF_USEDFPU);
+		if ((((tsk->thread.fpu.hard.fpscr & FPSCR_ENABLE_MASK) >> 7) &
+		     (fpu_exception_flags >> 2)) == 0) {
+			return;
+		}
+	}
+
 	force_sig(SIGFPE, tsk);
 }
 
@@ -319,7 +473,7 @@ BUILD_TRAP_HANDLER(fpu_state_restore)
 	if (used_math()) {
 		/* Using the FPU again.  */
 		restore_fpu(tsk);
-	} else	{
+	} else {
 		/* First time FPU user.  */
 		fpu_init();
 		set_used_math();
diff --git a/arch/sh/kernel/cpu/sh4/softfloat.c b/arch/sh/kernel/cpu/sh4/softfloat.c
new file mode 100644
index 000000000000..7b2d337ee412
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/softfloat.c
@@ -0,0 +1,892 @@
+/*
+ * Floating point emulation support for subnormalised numbers on SH4
+ * architecture This file is derived from the SoftFloat IEC/IEEE
+ * Floating-point Arithmetic Package, Release 2 the original license of
+ * which is reproduced below.
+ *
+ * ========================================================================
+ *
+ * This C source file is part of the SoftFloat IEC/IEEE Floating-point
+ * Arithmetic Package, Release 2.
+ *
+ * Written by John R. Hauser.  This work was made possible in part by the
+ * International Computer Science Institute, located at Suite 600, 1947 Center
+ * Street, Berkeley, California 94704.  Funding was partially provided by the
+ * National Science Foundation under grant MIP-9311980.  The original version
+ * of this code was written as part of a project to build a fixed-point vector
+ * processor in collaboration with the University of California at Berkeley,
+ * overseen by Profs. Nelson Morgan and John Wawrzynek.  More information
+ * is available through the web page `http://HTTP.CS.Berkeley.EDU/~jhauser/
+ * arithmetic/softfloat.html'.
+ *
+ * THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE.  Although reasonable effort
+ * has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
+ * TIMES RESULT IN INCORRECT BEHAVIOR.  USE OF THIS SOFTWARE IS RESTRICTED TO
+ * PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY
+ * AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.
+ *
+ * Derivative works are acceptable, even for commercial purposes, so long as
+ * (1) they include prominent notice that the work is derivative, and (2) they
+ * include prominent notice akin to these three paragraphs for those parts of
+ * this code that are retained.
+ *
+ * ========================================================================
+ *
+ * SH4 modifications by Ismail Dhaoui <ismail.dhaoui@st.com>
+ * and Kamel Khelifi <kamel.khelifi@st.com>
+ */
+#include <linux/kernel.h>
+#include <asm/cpu/fpu.h>
+
+#define LIT64( a ) a##LL
+
+typedef char flag;
+typedef unsigned char uint8;
+typedef signed char int8;
+typedef int uint16;
+typedef int int16;
+typedef unsigned int uint32;
+typedef signed int int32;
+
+typedef unsigned long long int bits64;
+typedef signed long long int sbits64;
+
+typedef unsigned char bits8;
+typedef signed char sbits8;
+typedef unsigned short int bits16;
+typedef signed short int sbits16;
+typedef unsigned int bits32;
+typedef signed int sbits32;
+
+typedef unsigned long long int uint64;
+typedef signed long long int int64;
+
+typedef unsigned long int float32;
+typedef unsigned long long float64;
+
+extern void float_raise(unsigned int flags);	/* in fpu.c */
+extern int float_rounding_mode(void);	/* in fpu.c */
+
+inline bits64 extractFloat64Frac(float64 a);
+inline flag extractFloat64Sign(float64 a);
+inline int16 extractFloat64Exp(float64 a);
+inline int16 extractFloat32Exp(float32 a);
+inline flag extractFloat32Sign(float32 a);
+inline bits32 extractFloat32Frac(float32 a);
+inline float64 packFloat64(flag zSign, int16 zExp, bits64 zSig);
+inline void shift64RightJamming(bits64 a, int16 count, bits64 * zPtr);
+inline float32 packFloat32(flag zSign, int16 zExp, bits32 zSig);
+inline void shift32RightJamming(bits32 a, int16 count, bits32 * zPtr);
+float64 float64_sub(float64 a, float64 b);
+float32 float32_sub(float32 a, float32 b);
+float32 float32_add(float32 a, float32 b);
+float64 float64_add(float64 a, float64 b);
+float64 float64_div(float64 a, float64 b);
+float32 float32_div(float32 a, float32 b);
+float32 float32_mul(float32 a, float32 b);
+float64 float64_mul(float64 a, float64 b);
+inline void add128(bits64 a0, bits64 a1, bits64 b0, bits64 b1, bits64 * z0Ptr,
+		   bits64 * z1Ptr);
+inline void sub128(bits64 a0, bits64 a1, bits64 b0, bits64 b1, bits64 * z0Ptr,
+		   bits64 * z1Ptr);
+inline void mul64To128(bits64 a, bits64 b, bits64 * z0Ptr, bits64 * z1Ptr);
+
+static int8 countLeadingZeros32(bits32 a);
+static int8 countLeadingZeros64(bits64 a);
+static float64 normalizeRoundAndPackFloat64(flag zSign, int16 zExp,
+					    bits64 zSig);
+static float64 subFloat64Sigs(float64 a, float64 b, flag zSign);
+static float64 addFloat64Sigs(float64 a, float64 b, flag zSign);
+static float32 roundAndPackFloat32(flag zSign, int16 zExp, bits32 zSig);
+static float32 normalizeRoundAndPackFloat32(flag zSign, int16 zExp,
+					    bits32 zSig);
+static float64 roundAndPackFloat64(flag zSign, int16 zExp, bits64 zSig);
+static float32 subFloat32Sigs(float32 a, float32 b, flag zSign);
+static float32 addFloat32Sigs(float32 a, float32 b, flag zSign);
+static void normalizeFloat64Subnormal(bits64 aSig, int16 * zExpPtr,
+				      bits64 * zSigPtr);
+static bits64 estimateDiv128To64(bits64 a0, bits64 a1, bits64 b);
+static void normalizeFloat32Subnormal(bits32 aSig, int16 * zExpPtr,
+				      bits32 * zSigPtr);
+
+inline bits64 extractFloat64Frac(float64 a)
+{
+	return a & LIT64(0x000FFFFFFFFFFFFF);
+}
+
+inline flag extractFloat64Sign(float64 a)
+{
+	return a >> 63;
+}
+
+inline int16 extractFloat64Exp(float64 a)
+{
+	return (a >> 52) & 0x7FF;
+}
+
+inline int16 extractFloat32Exp(float32 a)
+{
+	return (a >> 23) & 0xFF;
+}
+
+inline flag extractFloat32Sign(float32 a)
+{
+	return a >> 31;
+}
+
+inline bits32 extractFloat32Frac(float32 a)
+{
+	return a & 0x007FFFFF;
+}
+
+inline float64 packFloat64(flag zSign, int16 zExp, bits64 zSig)
+{
+	return (((bits64) zSign) << 63) + (((bits64) zExp) << 52) + zSig;
+}
+
+inline void shift64RightJamming(bits64 a, int16 count, bits64 * zPtr)
+{
+	bits64 z;
+
+	if (count == 0) {
+		z = a;
+	} else if (count < 64) {
+		z = (a >> count) | ((a << ((-count) & 63)) != 0);
+	} else {
+		z = (a != 0);
+	}
+	*zPtr = z;
+}
+
+static int8 countLeadingZeros32(bits32 a)
+{
+	static const int8 countLeadingZerosHigh[] = {
+		8, 7, 6, 6, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 4, 4,
+		3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+		2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+		2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+		1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+		1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+		1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+		1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+	};
+	int8 shiftCount;
+
+	shiftCount = 0;
+	if (a < 0x10000) {
+		shiftCount += 16;
+		a <<= 16;
+	}
+	if (a < 0x1000000) {
+		shiftCount += 8;
+		a <<= 8;
+	}
+	shiftCount += countLeadingZerosHigh[a >> 24];
+	return shiftCount;
+
+}
+
+static int8 countLeadingZeros64(bits64 a)
+{
+	int8 shiftCount;
+
+	shiftCount = 0;
+	if (a < ((bits64) 1) << 32) {
+		shiftCount += 32;
+	} else {
+		a >>= 32;
+	}
+	shiftCount += countLeadingZeros32(a);
+	return shiftCount;
+
+}
+
+static float64 normalizeRoundAndPackFloat64(flag zSign, int16 zExp, bits64 zSig)
+{
+	int8 shiftCount;
+
+	shiftCount = countLeadingZeros64(zSig) - 1;
+	return roundAndPackFloat64(zSign, zExp - shiftCount,
+				   zSig << shiftCount);
+
+}
+
+static float64 subFloat64Sigs(float64 a, float64 b, flag zSign)
+{
+	int16 aExp, bExp, zExp;
+	bits64 aSig, bSig, zSig;
+	int16 expDiff;
+
+	aSig = extractFloat64Frac(a);
+	aExp = extractFloat64Exp(a);
+	bSig = extractFloat64Frac(b);
+	bExp = extractFloat64Exp(b);
+	expDiff = aExp - bExp;
+	aSig <<= 10;
+	bSig <<= 10;
+	if (0 < expDiff)
+		goto aExpBigger;
+	if (expDiff < 0)
+		goto bExpBigger;
+	if (aExp == 0) {
+		aExp = 1;
+		bExp = 1;
+	}
+	if (bSig < aSig)
+		goto aBigger;
+	if (aSig < bSig)
+		goto bBigger;
+	return packFloat64(float_rounding_mode() == FPSCR_RM_ZERO, 0, 0);
+      bExpBigger:
+	if (bExp == 0x7FF) {
+		return packFloat64(zSign ^ 1, 0x7FF, 0);
+	}
+	if (aExp == 0) {
+		++expDiff;
+	} else {
+		aSig |= LIT64(0x4000000000000000);
+	}
+	shift64RightJamming(aSig, -expDiff, &aSig);
+	bSig |= LIT64(0x4000000000000000);
+      bBigger:
+	zSig = bSig - aSig;
+	zExp = bExp;
+	zSign ^= 1;
+	goto normalizeRoundAndPack;
+      aExpBigger:
+	if (aExp == 0x7FF) {
+		return a;
+	}
+	if (bExp == 0) {
+		--expDiff;
+	} else {
+		bSig |= LIT64(0x4000000000000000);
+	}
+	shift64RightJamming(bSig, expDiff, &bSig);
+	aSig |= LIT64(0x4000000000000000);
+      aBigger:
+	zSig = aSig - bSig;
+	zExp = aExp;
+      normalizeRoundAndPack:
+	--zExp;
+	return normalizeRoundAndPackFloat64(zSign, zExp, zSig);
+
+}
+static float64 addFloat64Sigs(float64 a, float64 b, flag zSign)
+{
+	int16 aExp, bExp, zExp;
+	bits64 aSig, bSig, zSig;
+	int16 expDiff;
+
+	aSig = extractFloat64Frac(a);
+	aExp = extractFloat64Exp(a);
+	bSig = extractFloat64Frac(b);
+	bExp = extractFloat64Exp(b);
+	expDiff = aExp - bExp;
+	aSig <<= 9;
+	bSig <<= 9;
+	if (0 < expDiff) {
+		if (aExp == 0x7FF) {
+			return a;
+		}
+		if (bExp == 0) {
+			--expDiff;
+		} else {
+			bSig |= LIT64(0x2000000000000000);
+		}
+		shift64RightJamming(bSig, expDiff, &bSig);
+		zExp = aExp;
+	} else if (expDiff < 0) {
+		if (bExp == 0x7FF) {
+			return packFloat64(zSign, 0x7FF, 0);
+		}
+		if (aExp == 0) {
+			++expDiff;
+		} else {
+			aSig |= LIT64(0x2000000000000000);
+		}
+		shift64RightJamming(aSig, -expDiff, &aSig);
+		zExp = bExp;
+	} else {
+		if (aExp == 0x7FF) {
+			return a;
+		}
+		if (aExp == 0)
+			return packFloat64(zSign, 0, (aSig + bSig) >> 9);
+		zSig = LIT64(0x4000000000000000) + aSig + bSig;
+		zExp = aExp;
+		goto roundAndPack;
+	}
+	aSig |= LIT64(0x2000000000000000);
+	zSig = (aSig + bSig) << 1;
+	--zExp;
+	if ((sbits64) zSig < 0) {
+		zSig = aSig + bSig;
+		++zExp;
+	}
+      roundAndPack:
+	return roundAndPackFloat64(zSign, zExp, zSig);
+
+}
+
+inline float32 packFloat32(flag zSign, int16 zExp, bits32 zSig)
+{
+	return (((bits32) zSign) << 31) + (((bits32) zExp) << 23) + zSig;
+}
+
+inline void shift32RightJamming(bits32 a, int16 count, bits32 * zPtr)
+{
+	bits32 z;
+	if (count == 0) {
+		z = a;
+	} else if (count < 32) {
+		z = (a >> count) | ((a << ((-count) & 31)) != 0);
+	} else {
+		z = (a != 0);
+	}
+	*zPtr = z;
+}
+
+static float32 roundAndPackFloat32(flag zSign, int16 zExp, bits32 zSig)
+{
+	flag roundNearestEven;
+	int8 roundIncrement, roundBits;
+	flag isTiny;
+
+	/* SH4 has only 2 rounding modes - round to nearest and round to zero */
+	roundNearestEven = (float_rounding_mode() == FPSCR_RM_NEAREST);
+	roundIncrement = 0x40;
+	if (!roundNearestEven) {
+		roundIncrement = 0;
+	}
+	roundBits = zSig & 0x7F;
+	if (0xFD <= (bits16) zExp) {
+		if ((0xFD < zExp)
+		    || ((zExp == 0xFD)
+			&& ((sbits32) (zSig + roundIncrement) < 0))
+		    ) {
+			float_raise(FPSCR_CAUSE_OVERFLOW | FPSCR_CAUSE_INEXACT);
+			return packFloat32(zSign, 0xFF,
+					   0) - (roundIncrement == 0);
+		}
+		if (zExp < 0) {
+			isTiny = (zExp < -1)
+			    || (zSig + roundIncrement < 0x80000000);
+			shift32RightJamming(zSig, -zExp, &zSig);
+			zExp = 0;
+			roundBits = zSig & 0x7F;
+			if (isTiny && roundBits)
+				float_raise(FPSCR_CAUSE_UNDERFLOW);
+		}
+	}
+	if (roundBits)
+		float_raise(FPSCR_CAUSE_INEXACT);
+	zSig = (zSig + roundIncrement) >> 7;
+	zSig &= ~(((roundBits ^ 0x40) == 0) & roundNearestEven);
+	if (zSig == 0)
+		zExp = 0;
+	return packFloat32(zSign, zExp, zSig);
+
+}
+
+static float32 normalizeRoundAndPackFloat32(flag zSign, int16 zExp, bits32 zSig)
+{
+	int8 shiftCount;
+
+	shiftCount = countLeadingZeros32(zSig) - 1;
+	return roundAndPackFloat32(zSign, zExp - shiftCount,
+				   zSig << shiftCount);
+}
+
+static float64 roundAndPackFloat64(flag zSign, int16 zExp, bits64 zSig)
+{
+	flag roundNearestEven;
+	int16 roundIncrement, roundBits;
+	flag isTiny;
+
+	/* SH4 has only 2 rounding modes - round to nearest and round to zero */
+	roundNearestEven = (float_rounding_mode() == FPSCR_RM_NEAREST);
+	roundIncrement = 0x200;
+	if (!roundNearestEven) {
+		roundIncrement = 0;
+	}
+	roundBits = zSig & 0x3FF;
+	if (0x7FD <= (bits16) zExp) {
+		if ((0x7FD < zExp)
+		    || ((zExp == 0x7FD)
+			&& ((sbits64) (zSig + roundIncrement) < 0))
+		    ) {
+			float_raise(FPSCR_CAUSE_OVERFLOW | FPSCR_CAUSE_INEXACT);
+			return packFloat64(zSign, 0x7FF,
+					   0) - (roundIncrement == 0);
+		}
+		if (zExp < 0) {
+			isTiny = (zExp < -1)
+			    || (zSig + roundIncrement <
+				LIT64(0x8000000000000000));
+			shift64RightJamming(zSig, -zExp, &zSig);
+			zExp = 0;
+			roundBits = zSig & 0x3FF;
+			if (isTiny && roundBits)
+				float_raise(FPSCR_CAUSE_UNDERFLOW);
+		}
+	}
+	if (roundBits)
+		float_raise(FPSCR_CAUSE_INEXACT);
+	zSig = (zSig + roundIncrement) >> 10;
+	zSig &= ~(((roundBits ^ 0x200) == 0) & roundNearestEven);
+	if (zSig == 0)
+		zExp = 0;
+	return packFloat64(zSign, zExp, zSig);
+
+}
+
+static float32 subFloat32Sigs(float32 a, float32 b, flag zSign)
+{
+	int16 aExp, bExp, zExp;
+	bits32 aSig, bSig, zSig;
+	int16 expDiff;
+
+	aSig = extractFloat32Frac(a);
+	aExp = extractFloat32Exp(a);
+	bSig = extractFloat32Frac(b);
+	bExp = extractFloat32Exp(b);
+	expDiff = aExp - bExp;
+	aSig <<= 7;
+	bSig <<= 7;
+	if (0 < expDiff)
+		goto aExpBigger;
+	if (expDiff < 0)
+		goto bExpBigger;
+	if (aExp == 0) {
+		aExp = 1;
+		bExp = 1;
+	}
+	if (bSig < aSig)
+		goto aBigger;
+	if (aSig < bSig)
+		goto bBigger;
+	return packFloat32(float_rounding_mode() == FPSCR_RM_ZERO, 0, 0);
+      bExpBigger:
+	if (bExp == 0xFF) {
+		return packFloat32(zSign ^ 1, 0xFF, 0);
+	}
+	if (aExp == 0) {
+		++expDiff;
+	} else {
+		aSig |= 0x40000000;
+	}
+	shift32RightJamming(aSig, -expDiff, &aSig);
+	bSig |= 0x40000000;
+      bBigger:
+	zSig = bSig - aSig;
+	zExp = bExp;
+	zSign ^= 1;
+	goto normalizeRoundAndPack;
+      aExpBigger:
+	if (aExp == 0xFF) {
+		return a;
+	}
+	if (bExp == 0) {
+		--expDiff;
+	} else {
+		bSig |= 0x40000000;
+	}
+	shift32RightJamming(bSig, expDiff, &bSig);
+	aSig |= 0x40000000;
+      aBigger:
+	zSig = aSig - bSig;
+	zExp = aExp;
+      normalizeRoundAndPack:
+	--zExp;
+	return normalizeRoundAndPackFloat32(zSign, zExp, zSig);
+
+}
+
+static float32 addFloat32Sigs(float32 a, float32 b, flag zSign)
+{
+	int16 aExp, bExp, zExp;
+	bits32 aSig, bSig, zSig;
+	int16 expDiff;
+
+	aSig = extractFloat32Frac(a);
+	aExp = extractFloat32Exp(a);
+	bSig = extractFloat32Frac(b);
+	bExp = extractFloat32Exp(b);
+	expDiff = aExp - bExp;
+	aSig <<= 6;
+	bSig <<= 6;
+	if (0 < expDiff) {
+		if (aExp == 0xFF) {
+			return a;
+		}
+		if (bExp == 0) {
+			--expDiff;
+		} else {
+			bSig |= 0x20000000;
+		}
+		shift32RightJamming(bSig, expDiff, &bSig);
+		zExp = aExp;
+	} else if (expDiff < 0) {
+		if (bExp == 0xFF) {
+			return packFloat32(zSign, 0xFF, 0);
+		}
+		if (aExp == 0) {
+			++expDiff;
+		} else {
+			aSig |= 0x20000000;
+		}
+		shift32RightJamming(aSig, -expDiff, &aSig);
+		zExp = bExp;
+	} else {
+		if (aExp == 0xFF) {
+			return a;
+		}
+		if (aExp == 0)
+			return packFloat32(zSign, 0, (aSig + bSig) >> 6);
+		zSig = 0x40000000 + aSig + bSig;
+		zExp = aExp;
+		goto roundAndPack;
+	}
+	aSig |= 0x20000000;
+	zSig = (aSig + bSig) << 1;
+	--zExp;
+	if ((sbits32) zSig < 0) {
+		zSig = aSig + bSig;
+		++zExp;
+	}
+      roundAndPack:
+	return roundAndPackFloat32(zSign, zExp, zSig);
+
+}
+
+float64 float64_sub(float64 a, float64 b)
+{
+	flag aSign, bSign;
+
+	aSign = extractFloat64Sign(a);
+	bSign = extractFloat64Sign(b);
+	if (aSign == bSign) {
+		return subFloat64Sigs(a, b, aSign);
+	} else {
+		return addFloat64Sigs(a, b, aSign);
+	}
+
+}
+
+float32 float32_sub(float32 a, float32 b)
+{
+	flag aSign, bSign;
+
+	aSign = extractFloat32Sign(a);
+	bSign = extractFloat32Sign(b);
+	if (aSign == bSign) {
+		return subFloat32Sigs(a, b, aSign);
+	} else {
+		return addFloat32Sigs(a, b, aSign);
+	}
+
+}
+
+float32 float32_add(float32 a, float32 b)
+{
+	flag aSign, bSign;
+
+	aSign = extractFloat32Sign(a);
+	bSign = extractFloat32Sign(b);
+	if (aSign == bSign) {
+		return addFloat32Sigs(a, b, aSign);
+	} else {
+		return subFloat32Sigs(a, b, aSign);
+	}
+
+}
+
+float64 float64_add(float64 a, float64 b)
+{
+	flag aSign, bSign;
+
+	aSign = extractFloat64Sign(a);
+	bSign = extractFloat64Sign(b);
+	if (aSign == bSign) {
+		return addFloat64Sigs(a, b, aSign);
+	} else {
+		return subFloat64Sigs(a, b, aSign);
+	}
+}
+
+static void
+normalizeFloat64Subnormal(bits64 aSig, int16 * zExpPtr, bits64 * zSigPtr)
+{
+	int8 shiftCount;
+
+	shiftCount = countLeadingZeros64(aSig) - 11;
+	*zSigPtr = aSig << shiftCount;
+	*zExpPtr = 1 - shiftCount;
+}
+
+inline void add128(bits64 a0, bits64 a1, bits64 b0, bits64 b1, bits64 * z0Ptr,
+		   bits64 * z1Ptr)
+{
+	bits64 z1;
+
+	z1 = a1 + b1;
+	*z1Ptr = z1;
+	*z0Ptr = a0 + b0 + (z1 < a1);
+}
+
+inline void
+sub128(bits64 a0, bits64 a1, bits64 b0, bits64 b1, bits64 * z0Ptr,
+       bits64 * z1Ptr)
+{
+	*z1Ptr = a1 - b1;
+	*z0Ptr = a0 - b0 - (a1 < b1);
+}
+
+static bits64 estimateDiv128To64(bits64 a0, bits64 a1, bits64 b)
+{
+	bits64 b0, b1;
+	bits64 rem0, rem1, term0, term1;
+	bits64 z;
+	if (b <= a0)
+		return LIT64(0xFFFFFFFFFFFFFFFF);
+	b0 = b >> 32;
+	z = (b0 << 32 <= a0) ? LIT64(0xFFFFFFFF00000000) : (a0 / b0) << 32;
+	mul64To128(b, z, &term0, &term1);
+	sub128(a0, a1, term0, term1, &rem0, &rem1);
+	while (((sbits64) rem0) < 0) {
+		z -= LIT64(0x100000000);
+		b1 = b << 32;
+		add128(rem0, rem1, b0, b1, &rem0, &rem1);
+	}
+	rem0 = (rem0 << 32) | (rem1 >> 32);
+	z |= (b0 << 32 <= rem0) ? 0xFFFFFFFF : rem0 / b0;
+	return z;
+}
+
+inline void mul64To128(bits64 a, bits64 b, bits64 * z0Ptr, bits64 * z1Ptr)
+{
+	bits32 aHigh, aLow, bHigh, bLow;
+	bits64 z0, zMiddleA, zMiddleB, z1;
+
+	aLow = a;
+	aHigh = a >> 32;
+	bLow = b;
+	bHigh = b >> 32;
+	z1 = ((bits64) aLow) * bLow;
+	zMiddleA = ((bits64) aLow) * bHigh;
+	zMiddleB = ((bits64) aHigh) * bLow;
+	z0 = ((bits64) aHigh) * bHigh;
+	zMiddleA += zMiddleB;
+	z0 += (((bits64) (zMiddleA < zMiddleB)) << 32) + (zMiddleA >> 32);
+	zMiddleA <<= 32;
+	z1 += zMiddleA;
+	z0 += (z1 < zMiddleA);
+	*z1Ptr = z1;
+	*z0Ptr = z0;
+
+}
+
+static void normalizeFloat32Subnormal(bits32 aSig, int16 * zExpPtr,
+				      bits32 * zSigPtr)
+{
+	int8 shiftCount;
+
+	shiftCount = countLeadingZeros32(aSig) - 8;
+	*zSigPtr = aSig << shiftCount;
+	*zExpPtr = 1 - shiftCount;
+
+}
+
+float64 float64_div(float64 a, float64 b)
+{
+	flag aSign, bSign, zSign;
+	int16 aExp, bExp, zExp;
+	bits64 aSig, bSig, zSig;
+	bits64 rem0, rem1;
+	bits64 term0, term1;
+
+	aSig = extractFloat64Frac(a);
+	aExp = extractFloat64Exp(a);
+	aSign = extractFloat64Sign(a);
+	bSig = extractFloat64Frac(b);
+	bExp = extractFloat64Exp(b);
+	bSign = extractFloat64Sign(b);
+	zSign = aSign ^ bSign;
+	if (aExp == 0x7FF) {
+		if (bExp == 0x7FF) {
+		}
+		return packFloat64(zSign, 0x7FF, 0);
+	}
+	if (bExp == 0x7FF) {
+		return packFloat64(zSign, 0, 0);
+	}
+	if (bExp == 0) {
+		if (bSig == 0) {
+			if ((aExp | aSig) == 0) {
+				float_raise(FPSCR_CAUSE_INVALID);
+			}
+			return packFloat64(zSign, 0x7FF, 0);
+		}
+		normalizeFloat64Subnormal(bSig, &bExp, &bSig);
+	}
+	if (aExp == 0) {
+		if (aSig == 0)
+			return packFloat64(zSign, 0, 0);
+		normalizeFloat64Subnormal(aSig, &aExp, &aSig);
+	}
+	zExp = aExp - bExp + 0x3FD;
+	aSig = (aSig | LIT64(0x0010000000000000)) << 10;
+	bSig = (bSig | LIT64(0x0010000000000000)) << 11;
+	if (bSig <= (aSig + aSig)) {
+		aSig >>= 1;
+		++zExp;
+	}
+	zSig = estimateDiv128To64(aSig, 0, bSig);
+	if ((zSig & 0x1FF) <= 2) {
+		mul64To128(bSig, zSig, &term0, &term1);
+		sub128(aSig, 0, term0, term1, &rem0, &rem1);
+		while ((sbits64) rem0 < 0) {
+			--zSig;
+			add128(rem0, rem1, 0, bSig, &rem0, &rem1);
+		}
+		zSig |= (rem1 != 0);
+	}
+	return roundAndPackFloat64(zSign, zExp, zSig);
+
+}
+
+float32 float32_div(float32 a, float32 b)
+{
+	flag aSign, bSign, zSign;
+	int16 aExp, bExp, zExp;
+	bits32 aSig, bSig, zSig;
+
+	aSig = extractFloat32Frac(a);
+	aExp = extractFloat32Exp(a);
+	aSign = extractFloat32Sign(a);
+	bSig = extractFloat32Frac(b);
+	bExp = extractFloat32Exp(b);
+	bSign = extractFloat32Sign(b);
+	zSign = aSign ^ bSign;
+	if (aExp == 0xFF) {
+		if (bExp == 0xFF) {
+		}
+		return packFloat32(zSign, 0xFF, 0);
+	}
+	if (bExp == 0xFF) {
+		return packFloat32(zSign, 0, 0);
+	}
+	if (bExp == 0) {
+		if (bSig == 0) {
+			return packFloat32(zSign, 0xFF, 0);
+		}
+		normalizeFloat32Subnormal(bSig, &bExp, &bSig);
+	}
+	if (aExp == 0) {
+		if (aSig == 0)
+			return packFloat32(zSign, 0, 0);
+		normalizeFloat32Subnormal(aSig, &aExp, &aSig);
+	}
+	zExp = aExp - bExp + 0x7D;
+	aSig = (aSig | 0x00800000) << 7;
+	bSig = (bSig | 0x00800000) << 8;
+	if (bSig <= (aSig + aSig)) {
+		aSig >>= 1;
+		++zExp;
+	}
+	zSig = (((bits64) aSig) << 32) / bSig;
+	if ((zSig & 0x3F) == 0) {
+		zSig |= (((bits64) bSig) * zSig != ((bits64) aSig) << 32);
+	}
+	return roundAndPackFloat32(zSign, zExp, zSig);
+
+}
+
+float32 float32_mul(float32 a, float32 b)
+{
+	char aSign, bSign, zSign;
+	int aExp, bExp, zExp;
+	unsigned int aSig, bSig;
+	unsigned long long zSig64;
+	unsigned int zSig;
+
+	aSig = extractFloat32Frac(a);
+	aExp = extractFloat32Exp(a);
+	aSign = extractFloat32Sign(a);
+	bSig = extractFloat32Frac(b);
+	bExp = extractFloat32Exp(b);
+	bSign = extractFloat32Sign(b);
+	zSign = aSign ^ bSign;
+	if (aExp == 0) {
+		if (aSig == 0)
+			return packFloat32(zSign, 0, 0);
+		normalizeFloat32Subnormal(aSig, &aExp, &aSig);
+	}
+	if (bExp == 0) {
+		if (bSig == 0)
+			return packFloat32(zSign, 0, 0);
+		normalizeFloat32Subnormal(bSig, &bExp, &bSig);
+	}
+	if ((bExp == 0xff && bSig == 0) || (aExp == 0xff && aSig == 0))
+		return roundAndPackFloat32(zSign, 0xff, 0);
+
+	zExp = aExp + bExp - 0x7F;
+	aSig = (aSig | 0x00800000) << 7;
+	bSig = (bSig | 0x00800000) << 8;
+	shift64RightJamming(((unsigned long long)aSig) * bSig, 32, &zSig64);
+	zSig = zSig64;
+	if (0 <= (signed int)(zSig << 1)) {
+		zSig <<= 1;
+		--zExp;
+	}
+	return roundAndPackFloat32(zSign, zExp, zSig);
+
+}
+
+float64 float64_mul(float64 a, float64 b)
+{
+	char aSign, bSign, zSign;
+	int aExp, bExp, zExp;
+	unsigned long long int aSig, bSig, zSig0, zSig1;
+
+	aSig = extractFloat64Frac(a);
+	aExp = extractFloat64Exp(a);
+	aSign = extractFloat64Sign(a);
+	bSig = extractFloat64Frac(b);
+	bExp = extractFloat64Exp(b);
+	bSign = extractFloat64Sign(b);
+	zSign = aSign ^ bSign;
+
+	if (aExp == 0) {
+		if (aSig == 0)
+			return packFloat64(zSign, 0, 0);
+		normalizeFloat64Subnormal(aSig, &aExp, &aSig);
+	}
+	if (bExp == 0) {
+		if (bSig == 0)
+			return packFloat64(zSign, 0, 0);
+		normalizeFloat64Subnormal(bSig, &bExp, &bSig);
+	}
+	if ((aExp == 0x7ff && aSig == 0) || (bExp == 0x7ff && bSig == 0))
+		return roundAndPackFloat64(zSign, 0x7ff, 0);
+
+	zExp = aExp + bExp - 0x3FF;
+	aSig = (aSig | 0x0010000000000000LL) << 10;
+	bSig = (bSig | 0x0010000000000000LL) << 11;
+	mul64To128(aSig, bSig, &zSig0, &zSig1);
+	zSig0 |= (zSig1 != 0);
+	if (0 <= (signed long long int)(zSig0 << 1)) {
+		zSig0 <<= 1;
+		--zExp;
+	}
+	return roundAndPackFloat64(zSign, zExp, zSig0);
+}
diff --git a/include/asm-sh/cpu-sh4/fpu.h b/include/asm-sh/cpu-sh4/fpu.h
new file mode 100644
index 000000000000..febef7342528
--- /dev/null
+++ b/include/asm-sh/cpu-sh4/fpu.h
@@ -0,0 +1,32 @@
+/*
+ * linux/arch/sh/kernel/cpu/sh4/sh4_fpu.h
+ *
+ * Copyright (C) 2006 STMicroelectronics Limited
+ * Author: Carl Shaw <carl.shaw@st.com>
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License Version 2.  See linux/COPYING for more information.
+ *
+ * Definitions for SH4 FPU operations
+ */
+
+#ifndef __CPU_SH4_FPU_H
+#define __CPU_SH4_FPU_H
+
+#define FPSCR_ENABLE_MASK	0x00000f80UL
+
+#define FPSCR_FMOV_DOUBLE	(1<<1)
+
+#define FPSCR_CAUSE_INEXACT	(1<<12)
+#define FPSCR_CAUSE_UNDERFLOW	(1<<13)
+#define FPSCR_CAUSE_OVERFLOW	(1<<14)
+#define FPSCR_CAUSE_DIVZERO	(1<<15)
+#define FPSCR_CAUSE_INVALID	(1<<16)
+#define FPSCR_CAUSE_ERROR 	(1<<17)
+
+#define FPSCR_DBL_PRECISION	(1<<19)
+#define FPSCR_ROUNDING_MODE(x)	((x >> 20) & 3)
+#define FPSCR_RM_NEAREST	(0)
+#define FPSCR_RM_ZERO		(1)
+
+#endif
-- 
cgit v1.2.3


From cd01204b82933754a7276838656420477f64d4b8 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 10 Dec 2007 15:50:28 +0900
Subject: sh: Encode L1/L2 cache shape in auxvt.

This adds in the L1I/L1D/L2 cache shape support to their respective
entries in the ELF auxvt, based on the Alpha implementation. We use
this on the userspace libc side for calculating a tightly packed
SHMLBA amongst other things.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c | 29 ++++++++++++++++++++++++++++-
 arch/sh/kernel/setup.c    |  3 +++
 include/asm-sh/auxvec.h   | 12 ++++++++++++
 include/asm-sh/elf.h      | 19 ++++++++++++++++---
 include/asm-sh/system.h   |  2 +-
 5 files changed, 60 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 0f0c76a842e4..80a31329ead9 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -13,6 +13,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/log2.h>
 #include <asm/mmu_context.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
@@ -20,6 +21,7 @@
 #include <asm/system.h>
 #include <asm/cacheflush.h>
 #include <asm/cache.h>
+#include <asm/elf.h>
 #include <asm/io.h>
 #include <asm/smp.h>
 #ifdef CONFIG_SUPERH32
@@ -151,6 +153,27 @@ static void __uses_jump_to_uncached cache_init(void)
 #define cache_init()	do { } while (0)
 #endif
 
+#define CSHAPE(totalsize, linesize, assoc) \
+	((totalsize & ~0xff) | (linesize << 4) | assoc)
+
+#define CACHE_DESC_SHAPE(desc)	\
+	CSHAPE((desc).way_size * (desc).ways, ilog2((desc).linesz), (desc).ways)
+
+static void detect_cache_shape(void)
+{
+	l1d_cache_shape = CACHE_DESC_SHAPE(current_cpu_data.dcache);
+
+	if (current_cpu_data.dcache.flags & SH_CACHE_COMBINED)
+		l1i_cache_shape = l1d_cache_shape;
+	else
+		l1i_cache_shape = CACHE_DESC_SHAPE(current_cpu_data.icache);
+
+	if (current_cpu_data.flags & CPU_HAS_L2_CACHE)
+		l2_cache_shape = CACHE_DESC_SHAPE(current_cpu_data.scache);
+	else
+		l2_cache_shape = -1; /* No S-cache */
+}
+
 #ifdef CONFIG_SH_DSP
 static void __init release_dsp(void)
 {
@@ -237,11 +260,15 @@ asmlinkage void __cpuinit sh_cpu_init(void)
 	/* Init the cache */
 	cache_init();
 
-	if (raw_smp_processor_id() == 0)
+	if (raw_smp_processor_id() == 0) {
 		shm_align_mask = max_t(unsigned long,
 				       current_cpu_data.dcache.way_size - 1,
 				       PAGE_SIZE - 1);
 
+		/* Boot CPU sets the cache shape */
+		detect_cache_shape();
+	}
+
 	/* Disable the FPU */
 	if (fpu_disabled) {
 		printk("FPU Disabled\n");
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index f48ce8e5d0a8..9c105c827e86 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -26,6 +26,7 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/page.h>
+#include <asm/elf.h>
 #include <asm/sections.h>
 #include <asm/irq.h>
 #include <asm/setup.h>
@@ -78,6 +79,8 @@ EXPORT_SYMBOL(memory_start);
 unsigned long memory_end = 0;
 EXPORT_SYMBOL(memory_end);
 
+int l1i_cache_shape, l1d_cache_shape, l2_cache_shape;
+
 static int __init early_parse_mem(char *p)
 {
 	unsigned long size;
diff --git a/include/asm-sh/auxvec.h b/include/asm-sh/auxvec.h
index 1b6916e63e90..4069858f8131 100644
--- a/include/asm-sh/auxvec.h
+++ b/include/asm-sh/auxvec.h
@@ -15,4 +15,16 @@
 #define AT_SYSINFO_EHDR		33
 #endif
 
+/*
+ * More complete cache descriptions than AT_[DIU]CACHEBSIZE.  If the
+ * value is -1, then the cache doesn't exist.  Otherwise:
+ *
+ *    bit 0-3:	  Cache set-associativity; 0 means fully associative.
+ *    bit 4-7:	  Log2 of cacheline size.
+ *    bit 8-31:	  Size of the entire cache >> 8.
+ */
+#define AT_L1I_CACHESHAPE	34
+#define AT_L1D_CACHESHAPE	35
+#define AT_L2_CACHESHAPE	36
+
 #endif /* __ASM_SH_AUXVEC_H */
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 5a1e920f0598..61960408e6ef 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -161,12 +161,25 @@ extern void __kernel_vsyscall;
 #define VDSO_BASE		((unsigned long)current->mm->context.vdso)
 #define VDSO_SYM(x)		(VDSO_BASE + (unsigned long)(x))
 
+#define VSYSCALL_AUX_ENT					\
+	if (vdso_enabled)					\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_BASE);
+#else
+#define VSYSCALL_AUX_ENT
+#endif /* CONFIG_VSYSCALL */
+
+extern int l1i_cache_shape, l1d_cache_shape, l2_cache_shape;
+
 /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
 #define ARCH_DLINFO						\
 do {								\
-	if (vdso_enabled)					\
-		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_BASE);	\
+	/* Optional vsyscall entry */				\
+	VSYSCALL_AUX_ENT					\
+								\
+	/* Cache desc */					\
+	NEW_AUX_ENT(AT_L1I_CACHESHAPE, l1i_cache_shape);	\
+	NEW_AUX_ENT(AT_L1D_CACHESHAPE, l1d_cache_shape);	\
+	NEW_AUX_ENT(AT_L2_CACHESHAPE, l2_cache_shape);		\
 } while (0)
-#endif /* CONFIG_VSYSCALL */
 
 #endif /* __ASM_SH_ELF_H */
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 9bda8d063ecf..84592555ba2a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -12,7 +12,7 @@
 #include <asm/types.h>
 #include <asm/ptrace.h>
 
-#define AT_VECTOR_SIZE_ARCH 1 /* entries in ARCH_DLINFO */
+#define AT_VECTOR_SIZE_ARCH 4 /* entries in ARCH_DLINFO */
 
 #if defined(CONFIG_CPU_SH4A) || defined(CONFIG_CPU_SH5)
 #define __icbi()			\
-- 
cgit v1.2.3


From a9f1365e5e4963705f85a6381cbed74c5615dbac Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 10 Dec 2007 16:06:32 +0900
Subject: sh: Use utsname()->machine for ELF_PLATFORM.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/elf.h | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 61960408e6ef..9a95f8cb2e13 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -62,11 +62,6 @@ typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 
 typedef struct user_fpu_struct elf_fpregset_t;
 
-/*
- * This is used to ensure we don't load something for the wrong architecture.
- */
-#define elf_check_arch(x) ( (x)->e_machine == EM_SH )
-
 /*
  * These are used to set parameters in the core dumps.
  */
@@ -78,6 +73,12 @@ typedef struct user_fpu_struct elf_fpregset_t;
 #endif
 #define ELF_ARCH	EM_SH
 
+#ifdef __KERNEL__
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) ( (x)->e_machine == EM_SH )
+
 #define USE_ELF_CORE_DUMP
 #define ELF_EXEC_PAGESIZE	PAGE_SIZE
 
@@ -106,7 +107,7 @@ typedef struct user_fpu_struct elf_fpregset_t;
    For the moment, we have only optimizations for the Intel generations,
    but that could change... */
 
-#define ELF_PLATFORM  (NULL)
+#define ELF_PLATFORM	(utsname()->machine)
 
 #ifdef __SH5__
 #define ELF_PLAT_INIT(_r, load_addr) \
@@ -182,4 +183,5 @@ do {								\
 	NEW_AUX_ENT(AT_L2_CACHESHAPE, l2_cache_shape);		\
 } while (0)
 
+#endif /* __KERNEL__ */
 #endif /* __ASM_SH_ELF_H */
-- 
cgit v1.2.3


From 98c4ecde9da14fafbcf1b6c025b08342a327e644 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 10 Dec 2007 16:21:57 +0900
Subject: sh: Provide the FPSCR init through AT_FPUCW.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/auxvec.h |  6 ++++++
 include/asm-sh/elf.h    | 11 ++++++++++-
 include/asm-sh/system.h |  2 +-
 3 files changed, 17 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/auxvec.h b/include/asm-sh/auxvec.h
index 4069858f8131..a6b9d4f4859e 100644
--- a/include/asm-sh/auxvec.h
+++ b/include/asm-sh/auxvec.h
@@ -6,6 +6,12 @@
  * for more of them.
  */
 
+/*
+ * This entry gives some information about the FPU initialization
+ * performed by the kernel.
+ */
+#define AT_FPUCW		18	/* Used FPU control word.  */
+
 #ifdef CONFIG_VSYSCALL
 /*
  * Only define this in the vsyscall case, the entry point to
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 9a95f8cb2e13..e00a58cb476c 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -169,13 +169,22 @@ extern void __kernel_vsyscall;
 #define VSYSCALL_AUX_ENT
 #endif /* CONFIG_VSYSCALL */
 
+#ifdef CONFIG_SH_FPU
+#define FPU_AUX_ENT	NEW_AUX_ENT(AT_FPUCW, FPSCR_INIT)
+#else
+#define FPU_AUX_ENT
+#endif
+
 extern int l1i_cache_shape, l1d_cache_shape, l2_cache_shape;
 
 /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
 #define ARCH_DLINFO						\
 do {								\
+	/* Optional FPU initialization */			\
+	FPU_AUX_ENT;						\
+								\
 	/* Optional vsyscall entry */				\
-	VSYSCALL_AUX_ENT					\
+	VSYSCALL_AUX_ENT;					\
 								\
 	/* Cache desc */					\
 	NEW_AUX_ENT(AT_L1I_CACHESHAPE, l1i_cache_shape);	\
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 84592555ba2a..772cd1a0a674 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -12,7 +12,7 @@
 #include <asm/types.h>
 #include <asm/ptrace.h>
 
-#define AT_VECTOR_SIZE_ARCH 4 /* entries in ARCH_DLINFO */
+#define AT_VECTOR_SIZE_ARCH 5 /* entries in ARCH_DLINFO */
 
 #if defined(CONFIG_CPU_SH4A) || defined(CONFIG_CPU_SH5)
 #define __icbi()			\
-- 
cgit v1.2.3


From 543d5afe5b425ef25a865656bfb76083515dc1cf Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 10 Dec 2007 16:33:32 +0900
Subject: sh: Kill off superfluous __KERNEL__ check in asm/elf.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/elf.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index e00a58cb476c..05092da1aa59 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -89,7 +89,6 @@ typedef struct user_fpu_struct elf_fpregset_t;
 
 #define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
 
-
 #define ELF_CORE_COPY_REGS(_dest,_regs)				\
 	memcpy((char *) &_dest, (char *) _regs,			\
 	       sizeof(struct pt_regs));
@@ -139,7 +138,6 @@ typedef struct user_fpu_struct elf_fpregset_t;
        _r->sr = SR_FD; } while (0)
 #endif
 
-#ifdef __KERNEL__
 #define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
 struct task_struct;
 extern int dump_task_regs (struct task_struct *, elf_gregset_t *);
@@ -147,7 +145,6 @@ extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *);
 
 #define ELF_CORE_COPY_TASK_REGS(tsk, elf_regs) dump_task_regs(tsk, elf_regs)
 #define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
-#endif
 
 #ifdef CONFIG_VSYSCALL
 /* vDSO has arch_setup_additional_pages */
-- 
cgit v1.2.3


From 9acb98fb7ce948063a2269b4f8db83d6bef7e2b0 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 17 Dec 2007 10:52:11 +0900
Subject: sh: Stub in page_table_range_init() on nommu.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/tlb-nommu.c   | 6 ++++++
 include/asm-sh/pgtable.h | 2 ++
 2 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/arch/sh/mm/tlb-nommu.c b/arch/sh/mm/tlb-nommu.c
index 1ccca7c0532e..8842620604c3 100644
--- a/arch/sh/mm/tlb-nommu.c
+++ b/arch/sh/mm/tlb-nommu.c
@@ -9,6 +9,7 @@
  */
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <asm/pgtable.h>
 
 /*
  * Nothing too terribly exciting here ..
@@ -49,3 +50,8 @@ void update_mmu_cache(struct vm_area_struct * vma,
 {
 	BUG();
 }
+
+void __init page_table_range_init(unsigned long start, unsigned long end,
+				  pgd_t *pgd_base)
+{
+}
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 3df90f003e95..a4a8f8b93463 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -144,6 +144,8 @@ extern void update_mmu_cache(struct vm_area_struct * vma,
 			     unsigned long address, pte_t pte);
 extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
 extern void paging_init(void);
+extern void page_table_range_init(unsigned long start, unsigned long end,
+				  pgd_t *pgd);
 
 #include <asm-generic/pgtable.h>
 
-- 
cgit v1.2.3


From 830626caf8570675a15bcc9ec46fcb5bfbc71b0d Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 17 Dec 2007 10:52:37 +0900
Subject: sh: Fix up switch_to() type casts.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system_32.h | 94 +++++++++++++++++++++++-----------------------
 1 file changed, 47 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/system_32.h b/include/asm-sh/system_32.h
index e918bacd5ecf..7ff08d956ba8 100644
--- a/include/asm-sh/system_32.h
+++ b/include/asm-sh/system_32.h
@@ -9,53 +9,53 @@ struct task_struct *__switch_to(struct task_struct *prev,
 /*
  *	switch_to() should switch tasks to task nr n, first
  */
-#define switch_to(prev, next, last)				\
-do {								\
-	register u32 *__ts1 __asm__ ("r1") = &prev->thread.sp;	\
-	register u32 *__ts2 __asm__ ("r2") = &prev->thread.pc;	\
-	register u32 *__ts4 __asm__ ("r4") = (u32 *)prev;	\
-	register u32 *__ts5 __asm__ ("r5") = (u32 *)next;	\
-	register u32 *__ts6 __asm__ ("r6") = &next->thread.sp;	\
-	register u32 __ts7 __asm__ ("r7") = next->thread.pc;	\
-	struct task_struct *__last;				\
-								\
-	__asm__ __volatile__ (					\
-		".balign 4\n\t"					\
-		"stc.l	gbr, @-r15\n\t"				\
-		"sts.l	pr, @-r15\n\t"				\
-		"mov.l	r8, @-r15\n\t"				\
-		"mov.l	r9, @-r15\n\t"				\
-		"mov.l	r10, @-r15\n\t"				\
-		"mov.l	r11, @-r15\n\t"				\
-		"mov.l	r12, @-r15\n\t"				\
-		"mov.l	r13, @-r15\n\t"				\
-		"mov.l	r14, @-r15\n\t"				\
-		"mov.l	r15, @r1\t! save SP\n\t"		\
-		"mov.l	@r6, r15\t! change to new stack\n\t"	\
-		"mova	1f, %0\n\t"				\
-		"mov.l	%0, @r2\t! save PC\n\t"			\
-		"mov.l	2f, %0\n\t"				\
-		"jmp	@%0\t! call __switch_to\n\t"		\
-		" lds	r7, pr\t!  with return to new PC\n\t"	\
-		".balign	4\n"				\
-		"2:\n\t"					\
-		".long	__switch_to\n"				\
-		"1:\n\t"					\
-		"mov.l	@r15+, r14\n\t"				\
-		"mov.l	@r15+, r13\n\t"				\
-		"mov.l	@r15+, r12\n\t"				\
-		"mov.l	@r15+, r11\n\t"				\
-		"mov.l	@r15+, r10\n\t"				\
-		"mov.l	@r15+, r9\n\t"				\
-		"mov.l	@r15+, r8\n\t"				\
-		"lds.l	@r15+, pr\n\t"				\
-		"ldc.l	@r15+, gbr\n\t"				\
-		: "=z" (__last)					\
-		: "r" (__ts1), "r" (__ts2), "r" (__ts4),	\
-		  "r" (__ts5), "r" (__ts6), "r" (__ts7)		\
-		: "r3", "t");					\
-								\
-	last = __last;						\
+#define switch_to(prev, next, last)					\
+do {									\
+	register u32 *__ts1 __asm__ ("r1") = (u32 *)&prev->thread.sp;	\
+	register u32 *__ts2 __asm__ ("r2") = (u32 *)&prev->thread.pc;	\
+	register u32 *__ts4 __asm__ ("r4") = (u32 *)prev;		\
+	register u32 *__ts5 __asm__ ("r5") = (u32 *)next;		\
+	register u32 *__ts6 __asm__ ("r6") = (u32 *)&next->thread.sp;	\
+	register u32 __ts7 __asm__ ("r7") = next->thread.pc;		\
+	struct task_struct *__last;					\
+									\
+	__asm__ __volatile__ (						\
+		".balign 4\n\t"						\
+		"stc.l	gbr, @-r15\n\t"					\
+		"sts.l	pr, @-r15\n\t"					\
+		"mov.l	r8, @-r15\n\t"					\
+		"mov.l	r9, @-r15\n\t"					\
+		"mov.l	r10, @-r15\n\t"					\
+		"mov.l	r11, @-r15\n\t"					\
+		"mov.l	r12, @-r15\n\t"					\
+		"mov.l	r13, @-r15\n\t"					\
+		"mov.l	r14, @-r15\n\t"					\
+		"mov.l	r15, @r1\t! save SP\n\t"			\
+		"mov.l	@r6, r15\t! change to new stack\n\t"		\
+		"mova	1f, %0\n\t"					\
+		"mov.l	%0, @r2\t! save PC\n\t"				\
+		"mov.l	2f, %0\n\t"					\
+		"jmp	@%0\t! call __switch_to\n\t"			\
+		" lds	r7, pr\t!  with return to new PC\n\t"		\
+		".balign	4\n"					\
+		"2:\n\t"						\
+		".long	__switch_to\n"					\
+		"1:\n\t"						\
+		"mov.l	@r15+, r14\n\t"					\
+		"mov.l	@r15+, r13\n\t"					\
+		"mov.l	@r15+, r12\n\t"					\
+		"mov.l	@r15+, r11\n\t"					\
+		"mov.l	@r15+, r10\n\t"					\
+		"mov.l	@r15+, r9\n\t"					\
+		"mov.l	@r15+, r8\n\t"					\
+		"lds.l	@r15+, pr\n\t"					\
+		"ldc.l	@r15+, gbr\n\t"					\
+		: "=z" (__last)						\
+		: "r" (__ts1), "r" (__ts2), "r" (__ts4),		\
+		  "r" (__ts5), "r" (__ts6), "r" (__ts7)			\
+		: "r3", "t");						\
+									\
+	last = __last;							\
 } while (0)
 
 #define __uses_jump_to_uncached __attribute__ ((__section__ (".uncached.text")))
-- 
cgit v1.2.3


From e06712df7792165d12392bc1fdda6bf384e1313b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 17 Dec 2007 15:59:38 +0900
Subject: sh: Kill off pgtable.h from scatterlist.h.

Fixes up the mmc build.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/scatterlist.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/scatterlist.h b/include/asm-sh/scatterlist.h
index 35ef9c30f0de..2084d0373693 100644
--- a/include/asm-sh/scatterlist.h
+++ b/include/asm-sh/scatterlist.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_SH_SCATTERLIST_H
 #define __ASM_SH_SCATTERLIST_H
 
-#include <asm/pgtable.h>
 #include <asm/types.h>
 
 struct scatterlist {
-- 
cgit v1.2.3


From 003c6af6c6370cc3315ab828a0e6618f00713d96 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Mon, 17 Dec 2007 16:00:25 +0900
Subject: sh: Fix up binfmt_flat compile warnings.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/flat.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-sh/flat.h b/include/asm-sh/flat.h
index dc4f5950dafa..0cc800299e06 100644
--- a/include/asm-sh/flat.h
+++ b/include/asm-sh/flat.h
@@ -19,6 +19,6 @@
 #define	flat_get_addr_from_rp(rp, relval, flags, p)	get_unaligned(rp)
 #define	flat_put_addr_at_rp(rp, val, relval)	put_unaligned(val,rp)
 #define	flat_get_relocate_addr(rel)		(rel)
-#define	flat_set_persistent(relval, p)		0
+#define	flat_set_persistent(relval, p)		({ (void)p; 0; })
 
 #endif /* __ASM_SH_FLAT_H */
-- 
cgit v1.2.3


From 0095d58b4a91b9fb57aeb781909355b232517c64 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 18 Dec 2007 09:40:33 +0900
Subject: sh: include/asm-sh/: Spelling fixes.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/hd64461.h   | 28 ++++++++++++++--------------
 include/asm-sh/microdev.h  |  4 ++--
 include/asm-sh/voyagergx.h |  2 +-
 3 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/hd64461.h b/include/asm-sh/hd64461.h
index 342ca55a266a..8c1353baf00f 100644
--- a/include/asm-sh/hd64461.h
+++ b/include/asm-sh/hd64461.h
@@ -46,10 +46,10 @@
 /* CPU Data Bus Control Register */
 #define	HD64461_SCPUCR		(CONFIG_HD64461_IOBASE + 0x04)
 
-/* Base Adress Register */
+/* Base Address Register */
 #define	HD64461_LCDCBAR		(CONFIG_HD64461_IOBASE + 0x1000)
 
-/* Line increment adress */
+/* Line increment address */
 #define	HD64461_LCDCLOR		(CONFIG_HD64461_IOBASE + 0x1002)
 
 /* Controls LCD controller */
@@ -80,9 +80,9 @@
 #define	HD64461_LDR3		(CONFIG_HD64461_IOBASE + 0x101e)
 
 /* Palette Registers */
-#define	HD64461_CPTWAR		(CONFIG_HD64461_IOBASE + 0x1030)	/* Color Palette Write Adress Register */
+#define	HD64461_CPTWAR		(CONFIG_HD64461_IOBASE + 0x1030)	/* Color Palette Write Address Register */
 #define	HD64461_CPTWDR		(CONFIG_HD64461_IOBASE + 0x1032)	/* Color Palette Write Data Register */
-#define	HD64461_CPTRAR		(CONFIG_HD64461_IOBASE + 0x1034)	/* Color Palette Read Adress Register */
+#define	HD64461_CPTRAR		(CONFIG_HD64461_IOBASE + 0x1034)	/* Color Palette Read Address Register */
 #define	HD64461_CPTRDR		(CONFIG_HD64461_IOBASE + 0x1036)	/* Color Palette Read Data Register */
 
 #define	HD64461_GRDOR		(CONFIG_HD64461_IOBASE + 0x1040)	/* Display Resolution Offset Register */
@@ -97,8 +97,8 @@
 #define	HD64461_GRCFGR_COLORDEPTH8	0x01	/* Sets Colordepth 8 for Accelerator */
 
 /* Line Drawing Registers */
-#define	HD64461_LNSARH		(CONFIG_HD64461_IOBASE + 0x1046)	/* Line Start Adress Register (H) */
-#define	HD64461_LNSARL		(CONFIG_HD64461_IOBASE + 0x1048)	/* Line Start Adress Register (L) */
+#define	HD64461_LNSARH		(CONFIG_HD64461_IOBASE + 0x1046)	/* Line Start Address Register (H) */
+#define	HD64461_LNSARL		(CONFIG_HD64461_IOBASE + 0x1048)	/* Line Start Address Register (L) */
 #define	HD64461_LNAXLR		(CONFIG_HD64461_IOBASE + 0x104a)	/* Axis Pixel Length Register */
 #define	HD64461_LNDGR		(CONFIG_HD64461_IOBASE + 0x104c)	/* Diagonal Register */
 #define	HD64461_LNAXR		(CONFIG_HD64461_IOBASE + 0x104e)	/* Axial Register */
@@ -106,16 +106,16 @@
 #define	HD64461_LNMDR		(CONFIG_HD64461_IOBASE + 0x1052)	/* Line Mode Register */
 
 /* BitBLT Registers */
-#define	HD64461_BBTSSARH	(CONFIG_HD64461_IOBASE + 0x1054)	/* Source Start Adress Register (H) */
-#define	HD64461_BBTSSARL	(CONFIG_HD64461_IOBASE + 0x1056)	/* Source Start Adress Register (L) */
-#define	HD64461_BBTDSARH	(CONFIG_HD64461_IOBASE + 0x1058)	/* Destination Start Adress Register (H) */
-#define	HD64461_BBTDSARL	(CONFIG_HD64461_IOBASE + 0x105a)	/* Destination Start Adress Register (L) */
+#define	HD64461_BBTSSARH	(CONFIG_HD64461_IOBASE + 0x1054)	/* Source Start Address Register (H) */
+#define	HD64461_BBTSSARL	(CONFIG_HD64461_IOBASE + 0x1056)	/* Source Start Address Register (L) */
+#define	HD64461_BBTDSARH	(CONFIG_HD64461_IOBASE + 0x1058)	/* Destination Start Address Register (H) */
+#define	HD64461_BBTDSARL	(CONFIG_HD64461_IOBASE + 0x105a)	/* Destination Start Address Register (L) */
 #define	HD64461_BBTDWR		(CONFIG_HD64461_IOBASE + 0x105c)	/* Destination Block Width Register */
 #define	HD64461_BBTDHR		(CONFIG_HD64461_IOBASE + 0x105e)	/* Destination Block Height Register */
-#define	HD64461_BBTPARH		(CONFIG_HD64461_IOBASE + 0x1060)	/* Pattern Start Adress Register (H) */
-#define	HD64461_BBTPARL		(CONFIG_HD64461_IOBASE + 0x1062)	/* Pattern Start Adress Register (L) */
-#define	HD64461_BBTMARH		(CONFIG_HD64461_IOBASE + 0x1064)	/* Mask Start Adress Register (H) */
-#define	HD64461_BBTMARL		(CONFIG_HD64461_IOBASE + 0x1066)	/* Mask Start Adress Register (L) */
+#define	HD64461_BBTPARH		(CONFIG_HD64461_IOBASE + 0x1060)	/* Pattern Start Address Register (H) */
+#define	HD64461_BBTPARL		(CONFIG_HD64461_IOBASE + 0x1062)	/* Pattern Start Address Register (L) */
+#define	HD64461_BBTMARH		(CONFIG_HD64461_IOBASE + 0x1064)	/* Mask Start Address Register (H) */
+#define	HD64461_BBTMARL		(CONFIG_HD64461_IOBASE + 0x1066)	/* Mask Start Address Register (L) */
 #define	HD64461_BBTROPR		(CONFIG_HD64461_IOBASE + 0x1068)	/* ROP Register */
 #define	HD64461_BBTMDR		(CONFIG_HD64461_IOBASE + 0x106a)	/* BitBLT Mode Register */
 
diff --git a/include/asm-sh/microdev.h b/include/asm-sh/microdev.h
index 018332a9e590..1aed15856e11 100644
--- a/include/asm-sh/microdev.h
+++ b/include/asm-sh/microdev.h
@@ -17,7 +17,7 @@ extern void microdev_print_fpga_intc_status(void);
 /*
  * The following are useful macros for manipulating the interrupt
  * controller (INTC) on the CPU-board FPGA.  should be noted that there
- * is an INTC on the FPGA, and a seperate INTC on the SH4-202 core -
+ * is an INTC on the FPGA, and a separate INTC on the SH4-202 core -
  * these are two different things, both of which need to be prorammed to
  * correctly route - unfortunately, they have the same name and
  * abbreviations!
@@ -25,7 +25,7 @@ extern void microdev_print_fpga_intc_status(void);
 #define	MICRODEV_FPGA_INTC_BASE		0xa6110000ul				/* INTC base address on CPU-board FPGA */
 #define	MICRODEV_FPGA_INTENB_REG	(MICRODEV_FPGA_INTC_BASE+0ul)		/* Interrupt Enable Register on INTC on CPU-board FPGA */
 #define	MICRODEV_FPGA_INTDSB_REG	(MICRODEV_FPGA_INTC_BASE+8ul)		/* Interrupt Disable Register on INTC on CPU-board FPGA */
-#define	MICRODEV_FPGA_INTC_MASK(n)	(1ul<<(n))				/* Interupt mask to enable/disable INTC in CPU-board FPGA */
+#define	MICRODEV_FPGA_INTC_MASK(n)	(1ul<<(n))				/* Interrupt mask to enable/disable INTC in CPU-board FPGA */
 #define	MICRODEV_FPGA_INTPRI_REG(n)	(MICRODEV_FPGA_INTC_BASE+0x10+((n)/8)*8)/* Interrupt Priority Register on INTC on CPU-board FPGA */
 #define	MICRODEV_FPGA_INTPRI_LEVEL(n,x)	((x)<<(((n)%8)*4))			/* MICRODEV_FPGA_INTPRI_LEVEL(int_number, int_level) */
 #define	MICRODEV_FPGA_INTPRI_MASK(n)	(MICRODEV_FPGA_INTPRI_LEVEL((n),0xful))	/* Interrupt Priority Mask on INTC on CPU-board FPGA */
diff --git a/include/asm-sh/voyagergx.h b/include/asm-sh/voyagergx.h
index d825596562df..45b4547c74ac 100644
--- a/include/asm-sh/voyagergx.h
+++ b/include/asm-sh/voyagergx.h
@@ -213,7 +213,7 @@
 /* ----- Power mode 1 clock register -------------------------- */
 #define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
 
-/* ----- Power mode controll register ------------------------- */
+/* ----- Power mode control register ------------------------- */
 #define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
 
 /* ----- Miscellaneous Timing register ------------------------ */
-- 
cgit v1.2.3


From 7bff489a1496d363e9a640c00eb8970da15079dd Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 18 Dec 2007 09:43:15 +0900
Subject: sh: Always use CONFIG_HZ for HZ.

Currently the wdt forces HZ=1000 and sidesteps CONFIG_HZ completely. This
is a remnant from when HZ was hardcoded and before CONFIG_HZ was
introduced. Additionally, not all of the timers have this requirement
these days, so it's also an artificial limitation. Just kill it off and
use CONFIG_HZ directly.

Reported-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/param.h | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/param.h b/include/asm-sh/param.h
index 1012296e07ab..ae245afdfd6a 100644
--- a/include/asm-sh/param.h
+++ b/include/asm-sh/param.h
@@ -2,11 +2,7 @@
 #define __ASM_SH_PARAM_H
 
 #ifdef __KERNEL__
-# ifdef CONFIG_SH_WDT
-#  define HZ		1000		/* Needed for high-res WOVF */
-# else
-#  define HZ		CONFIG_HZ
-# endif
+# define HZ		CONFIG_HZ
 # define USER_HZ	100		/* User interfaces are in "ticks" */
 # define CLOCKS_PER_SEC	(USER_HZ)	/* frequency at which times() counts */
 #endif
-- 
cgit v1.2.3


From 31a49c4bf8f964b7a9897baa889916d71b51d9c1 Mon Sep 17 00:00:00 2001
From: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Date: Wed, 26 Dec 2007 11:45:06 +0900
Subject: sh: Add support for SH7721 CPU subtype.

Signed-off-by: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                       |  7 +++++++
 arch/sh/Kconfig.debug                 |  2 +-
 arch/sh/drivers/dma/Kconfig           |  2 +-
 arch/sh/drivers/dma/dma-sh.c          |  2 ++
 arch/sh/kernel/cpu/sh3/Makefile       |  1 +
 arch/sh/kernel/cpu/sh3/probe.c        |  3 +++
 arch/sh/kernel/cpu/sh3/setup-sh7720.c | 11 +++++++++--
 arch/sh/kernel/early_printk.c         | 12 ++++++++----
 arch/sh/kernel/setup.c                |  8 ++++----
 arch/sh/kernel/timers/timer-tmu.c     |  1 +
 drivers/serial/sh-sci.c               |  2 +-
 drivers/serial/sh-sci.h               | 21 ++++++++++++++-------
 include/asm-sh/cpu-sh3/cache.h        |  3 ++-
 include/asm-sh/cpu-sh3/dma.h          |  4 +++-
 include/asm-sh/cpu-sh3/gpio.h         |  3 ++-
 include/asm-sh/cpu-sh3/mmu_context.h  |  3 ++-
 include/asm-sh/cpu-sh3/timer.h        |  7 ++++---
 include/asm-sh/cpu-sh3/ubc.h          |  3 ++-
 include/asm-sh/processor.h            |  2 +-
 19 files changed, 68 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 538631ca10a6..d2f557719489 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -214,6 +214,13 @@ config CPU_SUBTYPE_SH7720
 	help
 	  Select SH7720 if you have a SH3-DSP SH7720 CPU.
 
+config CPU_SUBTYPE_SH7721
+	bool "Support SH7721 processor"
+	select CPU_SH3
+	select CPU_HAS_DSP
+	help
+	  Select SH7721 if you have a SH3-DSP SH7721 CPU.
+
 # SH-4 Processor Support
 
 config CPU_SUBTYPE_SH7750
diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 2881d8471a17..dde00d5c2478 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -35,7 +35,7 @@ config EARLY_SCIF_CONSOLE_PORT
 	default "0xfffe9800" if CPU_SUBTYPE_SH7206 || CPU_SUBTYPE_SH7263
 	default "0xf8420000" if CPU_SUBTYPE_SH7619
 	default "0xa4400000" if CPU_SUBTYPE_SH7712 || CPU_SUBTYPE_SH7705
-	default "0xa4430000" if CPU_SUBTYPE_SH7720
+	default "0xa4430000" if CPU_SUBTYPE_SH7720 || CPU_SUBTYPE_SH7721
 	default "0xffc30000" if CPU_SUBTYPE_SHX3
 	default "0xffe80000" if CPU_SH4
 	default "0x00000000"
diff --git a/arch/sh/drivers/dma/Kconfig b/arch/sh/drivers/dma/Kconfig
index 4e711a0c3dae..01936368b8b0 100644
--- a/arch/sh/drivers/dma/Kconfig
+++ b/arch/sh/drivers/dma/Kconfig
@@ -12,7 +12,7 @@ config SH_DMA
 config NR_ONCHIP_DMA_CHANNELS
 	int
 	depends on SH_DMA
-	default "6" if CPU_SUBTYPE_SH7720
+	default "6" if CPU_SUBTYPE_SH7720 || CPU_SUBTYPE_SH7721
 	default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R
 	default "12" if CPU_SUBTYPE_SH7780
 	default "4"
diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index 958bac1c585a..5c3359756a92 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -25,6 +25,7 @@ static int dmte_irq_map[] = {
 	DMTE2_IRQ,
 	DMTE3_IRQ,
 #if defined(CONFIG_CPU_SUBTYPE_SH7720)  ||	\
+    defined(CONFIG_CPU_SUBTYPE_SH7721)  ||	\
     defined(CONFIG_CPU_SUBTYPE_SH7751R) ||	\
     defined(CONFIG_CPU_SUBTYPE_SH7760)  ||	\
     defined(CONFIG_CPU_SUBTYPE_SH7709)  ||	\
@@ -203,6 +204,7 @@ static int sh_dmac_get_dma_residue(struct dma_channel *chan)
 }
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721) || \
     defined(CONFIG_CPU_SUBTYPE_SH7780)
 #define dmaor_read_reg()	ctrl_inw(DMAOR)
 #define dmaor_write_reg(data)	ctrl_outw(data, DMAOR)
diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index 646eb6933614..1afd05e62d1c 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SH7709)	+= setup-sh770x.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7710)	+= setup-sh7710.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7712)	+= setup-sh7710.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7720)	+= setup-sh7720.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7721)	+= setup-sh7720.o
 
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH3)			:= clock-sh3.o
diff --git a/arch/sh/kernel/cpu/sh3/probe.c b/arch/sh/kernel/cpu/sh3/probe.c
index 22070e43e34d..fcc80bb7bee7 100644
--- a/arch/sh/kernel/cpu/sh3/probe.c
+++ b/arch/sh/kernel/cpu/sh3/probe.c
@@ -84,6 +84,9 @@ int __uses_jump_to_uncached detect_cpu_and_cache_system(void)
 #if defined(CONFIG_CPU_SUBTYPE_SH7720)
 		boot_cpu_data.type = CPU_SH7720;
 #endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7721)
+		boot_cpu_data.type = CPU_SH7721;
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7705)
 		boot_cpu_data.type = CPU_SH7705;
 
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7720.c b/arch/sh/kernel/cpu/sh3/setup-sh7720.c
index a0929b8a95ae..00facd028a08 100644
--- a/arch/sh/kernel/cpu/sh3/setup-sh7720.c
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7720.c
@@ -127,8 +127,11 @@ static struct intc_vect vectors[] __initdata = {
 	INTC_VECT(USBF_SPD, 0x6e0),   INTC_VECT(DMAC1_DEI0, 0x800),
 	INTC_VECT(DMAC1_DEI1, 0x820), INTC_VECT(DMAC1_DEI2, 0x840),
 	INTC_VECT(DMAC1_DEI3, 0x860), INTC_VECT(LCDC, 0x900),
-	INTC_VECT(SSL, 0x980),        INTC_VECT(USBFI0, 0xa20),
-	INTC_VECT(USBFI1, 0xa40),     INTC_VECT(USBHI, 0xa60),
+#if defined(CONFIG_CPU_SUBTYPE_SH7720)
+	INTC_VECT(SSL, 0x980),
+#endif
+	INTC_VECT(USBFI0, 0xa20),     INTC_VECT(USBFI1, 0xa40),
+	INTC_VECT(USBHI, 0xa60),
 	INTC_VECT(DMAC2_DEI4, 0xb80), INTC_VECT(DMAC2_DEI5, 0xba0),
 	INTC_VECT(ADC, 0xbe0),        INTC_VECT(SCIF0, 0xc00),
 	INTC_VECT(SCIF1, 0xc20),      INTC_VECT(PINT07, 0xc80),
@@ -168,7 +171,11 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xA414FEE4UL, 0, 16, 4, /* IPRB */ { WDT, REF_RCMI, SIM, 0 } },
 	{ 0xA4140016UL, 0, 16, 4, /* IPRC */ { IRQ3, IRQ2, IRQ1, IRQ0 } },
 	{ 0xA4140018UL, 0, 16, 4, /* IPRD */ { USBF_SPD, TMU_SUNI, IRQ5, IRQ4 } },
+#if defined(CONFIG_CPU_SUBTYPE_SH7720)
 	{ 0xA414001AUL, 0, 16, 4, /* IPRE */ { DMAC1, 0, LCDC, SSL } },
+#else
+	{ 0xA414001AUL, 0, 16, 4, /* IPRE */ { DMAC1, 0, LCDC, 0 } },
+#endif
 	{ 0xA4080000UL, 0, 16, 4, /* IPRF */ { ADC, DMAC2, USBFI, CMT } },
 	{ 0xA4080002UL, 0, 16, 4, /* IPRG */ { SCIF0, SCIF1, 0, 0 } },
 	{ 0xA4080004UL, 0, 16, 4, /* IPRH */ { PINT07, PINT815, TPU, IIC } },
diff --git a/arch/sh/kernel/early_printk.c b/arch/sh/kernel/early_printk.c
index 2f30977558ad..957f25611543 100644
--- a/arch/sh/kernel/early_printk.c
+++ b/arch/sh/kernel/early_printk.c
@@ -63,7 +63,8 @@ static struct console bios_console = {
 #include <linux/serial_core.h>
 #include "../../../drivers/serial/sh-sci.h"
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7720)
+#if defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define EPK_SCSMR_VALUE 0x000
 #define EPK_SCBRR_VALUE 0x00C
 #define EPK_FIFO_SIZE 64
@@ -117,7 +118,8 @@ static struct console scif_console = {
 };
 
 #if !defined(CONFIG_SH_STANDARD_BIOS)
-#if defined(CONFIG_CPU_SUBTYPE_SH7720)
+#if defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 static void scif_sercon_init(char *s)
 {
 	sci_out(&scif_port, SCSCR, 0x0000);	/* clear TE and RE */
@@ -208,9 +210,11 @@ static int __init setup_early_printk(char *buf)
 	if (!strncmp(buf, "serial", 6)) {
 		early_console = &scif_console;
 
-#if (defined(CONFIG_CPU_SH4) || defined(CONFIG_CPU_SUBTYPE_SH7720)) && \
-    !defined(CONFIG_SH_STANDARD_BIOS)
+#if !defined(CONFIG_SH_STANDARD_BIOS)
+#if defined(CONFIG_CPU_SH4) || defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 		scif_sercon_init(buf + 6);
+#endif
 #endif
 	}
 #endif
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 9c105c827e86..48cd576754b5 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -314,10 +314,10 @@ static const char *cpu_name[] = {
 	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
 	[CPU_SH7709]	= "SH7709",	[CPU_SH7710]	= "SH7710",
 	[CPU_SH7712]	= "SH7712",	[CPU_SH7720]	= "SH7720",
-	[CPU_SH7729]	= "SH7729",	[CPU_SH7750]	= "SH7750",
-	[CPU_SH7750S]	= "SH7750S",	[CPU_SH7750R]	= "SH7750R",
-	[CPU_SH7751]	= "SH7751",	[CPU_SH7751R]	= "SH7751R",
-	[CPU_SH7760]	= "SH7760",
+	[CPU_SH7721]	= "SH7721",	[CPU_SH7729]	= "SH7729",
+	[CPU_SH7750]	= "SH7750",	[CPU_SH7750S]	= "SH7750S",
+	[CPU_SH7750R]	= "SH7750R",	[CPU_SH7751]	= "SH7751",
+	[CPU_SH7751R]	= "SH7751R",	[CPU_SH7760]	= "SH7760",
 	[CPU_SH4_202]	= "SH4-202",	[CPU_SH4_501]	= "SH4-501",
 	[CPU_SH7770]	= "SH7770",	[CPU_SH7780]	= "SH7780",
 	[CPU_SH7781]	= "SH7781",	[CPU_SH7343]	= "SH7343",
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index 628ec9a15e38..8935570008d2 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -174,6 +174,7 @@ static int tmu_timer_init(void)
 	tmu_timer_stop();
 
 #if !defined(CONFIG_CPU_SUBTYPE_SH7720) && \
+    !defined(CONFIG_CPU_SUBTYPE_SH7721) && \
     !defined(CONFIG_CPU_SUBTYPE_SH7760) && \
     !defined(CONFIG_CPU_SUBTYPE_SH7785) && \
     !defined(CONFIG_CPU_SUBTYPE_SHX3)
diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c
index 73440e26834b..6fbfd140b7e7 100644
--- a/drivers/serial/sh-sci.c
+++ b/drivers/serial/sh-sci.c
@@ -302,7 +302,7 @@ static void sci_init_pins_scif(struct uart_port* port, unsigned int cflag)
 	}
 	sci_out(port, SCFCR, fcr_val);
 }
-#elif defined(CONFIG_CPU_SUBTYPE_SH7720)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7720) || defined(CONFIG_CPU_SUBTYPE_SH7721)
 static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 {
 	unsigned int fcr_val = 0;
diff --git a/drivers/serial/sh-sci.h b/drivers/serial/sh-sci.h
index 0187dccfe8c1..85562040a6d3 100644
--- a/drivers/serial/sh-sci.h
+++ b/drivers/serial/sh-sci.h
@@ -46,7 +46,8 @@
  */
 # define SCSCR_INIT(port) (port->mapbase == SCIF2) ? 0xF3 : 0xF0
 # define SCIF_ONLY
-#elif defined(CONFIG_CPU_SUBTYPE_SH7720)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7721)
 # define SCSCR_INIT(port)  0x0030 /* TIE=0,RIE=0,TE=1,RE=1 */
 # define SCIF_ONLY
 #define SCIF_ORER    0x0200   /* overrun error bit */
@@ -216,7 +217,8 @@
 #define SCIF_DR    0x0001 /* 7705 SCIF, 7707 SCIF, 7709 SCIF, 7750 SCIF */
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7705) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define SCIF_ORER    0x0200
 #define SCIF_ERRORS ( SCIF_PER | SCIF_FER | SCIF_ER | SCIF_BRK | SCIF_ORER)
 #define SCIF_RFDC_MASK 0x007f
@@ -254,7 +256,8 @@
 # define SCxSR_PER(port)		SCIF_PER
 # define SCxSR_BRK(port)		SCIF_BRK
 #if defined(CONFIG_CPU_SUBTYPE_SH7705) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 # define SCxSR_RDxF_CLEAR(port)         (sci_in(port,SCxSR)&0xfffc)
 # define SCxSR_ERROR_CLEAR(port)        (sci_in(port,SCxSR)&0xfd73)
 # define SCxSR_TDxE_CLEAR(port)         (sci_in(port,SCxSR)&0xffdf)
@@ -363,7 +366,8 @@
 #define SCIF_FNS(name, sh3_scif_offset, sh3_scif_size, sh4_scif_offset, sh4_scif_size) \
 	  CPU_SCIF_FNS(name, sh4_scif_offset, sh4_scif_size)
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7720)
+      defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define SCIF_FNS(name, scif_offset, scif_size) \
   CPU_SCIF_FNS(name, scif_offset, scif_size)
 #else
@@ -390,7 +394,8 @@
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7705) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 
 SCIF_FNS(SCSMR,  0x00, 16)
 SCIF_FNS(SCBRR,  0x04,  8)
@@ -512,7 +517,8 @@ static inline void set_sh771x_scif_pfc(struct uart_port *port)
 		return;
 	}
 }
-#elif defined(CONFIG_CPU_SUBTYPE_SH7720)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7721)
 static inline int sci_rxd_in(struct uart_port *port)
 {
 	if (port->mapbase == 0xa4430000)
@@ -696,7 +702,8 @@ static inline int sci_rxd_in(struct uart_port *port)
     defined(CONFIG_CPU_SUBTYPE_SH7785)
 #define SCBRR_VALUE(bps, clk) ((clk+16*bps)/(16*bps)-1)
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7720)
+      defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define SCBRR_VALUE(bps, clk) (((clk*2)+16*bps)/(32*bps)-1)
 #elif defined(__H8300H__) || defined(__H8300S__)
 #define SCBRR_VALUE(bps) (((CONFIG_CPU_CLOCK*1000/32)/bps)-1)
diff --git a/include/asm-sh/cpu-sh3/cache.h b/include/asm-sh/cpu-sh3/cache.h
index 77dd45d82414..56bd838b7db4 100644
--- a/include/asm-sh/cpu-sh3/cache.h
+++ b/include/asm-sh/cpu-sh3/cache.h
@@ -33,7 +33,8 @@
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7705) || \
     defined(CONFIG_CPU_SUBTYPE_SH7710) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define CCR3	0xa40000b4
 #define CCR_CACHE_16KB  0x00010000
 #define CCR_CACHE_32KB	0x00020000
diff --git a/include/asm-sh/cpu-sh3/dma.h b/include/asm-sh/cpu-sh3/dma.h
index 54bfece328c2..092ff9d872c3 100644
--- a/include/asm-sh/cpu-sh3/dma.h
+++ b/include/asm-sh/cpu-sh3/dma.h
@@ -2,7 +2,9 @@
 #define __ASM_CPU_SH3_DMA_H
 
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7720) || defined(CONFIG_CPU_SUBTYPE_SH7709)
+#if defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7709)
 #define SH_DMAC_BASE	0xa4010020
 
 #define DMTE0_IRQ	48
diff --git a/include/asm-sh/cpu-sh3/gpio.h b/include/asm-sh/cpu-sh3/gpio.h
index 48770c1c7bdf..4e53eb314b8f 100644
--- a/include/asm-sh/cpu-sh3/gpio.h
+++ b/include/asm-sh/cpu-sh3/gpio.h
@@ -12,7 +12,8 @@
 #ifndef _CPU_SH3_GPIO_H
 #define _CPU_SH3_GPIO_H
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7720)
+#if defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 
 /* Control registers */
 #define PORT_PACR	0xA4050100UL
diff --git a/include/asm-sh/cpu-sh3/mmu_context.h b/include/asm-sh/cpu-sh3/mmu_context.h
index 16c2d63b7e39..ab09da73ce77 100644
--- a/include/asm-sh/cpu-sh3/mmu_context.h
+++ b/include/asm-sh/cpu-sh3/mmu_context.h
@@ -33,7 +33,8 @@
     defined(CONFIG_CPU_SUBTYPE_SH7709) || \
     defined(CONFIG_CPU_SUBTYPE_SH7710) || \
     defined(CONFIG_CPU_SUBTYPE_SH7712) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define INTEVT	0xa4000000	/* INTEVTE2(0xa4000000) */
 #else
 #define INTEVT	0xffffffd8
diff --git a/include/asm-sh/cpu-sh3/timer.h b/include/asm-sh/cpu-sh3/timer.h
index 7b795ac5477c..793acf12aa08 100644
--- a/include/asm-sh/cpu-sh3/timer.h
+++ b/include/asm-sh/cpu-sh3/timer.h
@@ -23,12 +23,13 @@
  * ---------------------------------------------------------------------------
  */
 
-#if  !defined(CONFIG_CPU_SUBTYPE_SH7720)
+#if  !defined(CONFIG_CPU_SUBTYPE_SH7720) && !defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define TMU_TOCR	0xfffffe90	/* Byte access */
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7710) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define TMU_012_TSTR	0xa412fe92	/* Byte access */
 
 #define TMU0_TCOR	0xa412fe94	/* Long access */
@@ -57,7 +58,7 @@
 #define TMU2_TCOR	0xfffffeac	/* Long access */
 #define TMU2_TCNT	0xfffffeb0	/* Long access */
 #define TMU2_TCR	0xfffffeb4	/* Word access */
-#if !defined(CONFIG_CPU_SUBTYPE_SH7720)
+#if !defined(CONFIG_CPU_SUBTYPE_SH7720) && !defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define TMU2_TCPR2	0xfffffeb8	/* Long access */
 #endif
 #endif
diff --git a/include/asm-sh/cpu-sh3/ubc.h b/include/asm-sh/cpu-sh3/ubc.h
index 18467c574534..4e6381d5ff7a 100644
--- a/include/asm-sh/cpu-sh3/ubc.h
+++ b/include/asm-sh/cpu-sh3/ubc.h
@@ -12,7 +12,8 @@
 #define __ASM_CPU_SH3_UBC_H
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7710) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7720)
+    defined(CONFIG_CPU_SUBTYPE_SH7720) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7721)
 #define UBC_BARA		0xa4ffffb0
 #define UBC_BAMRA		0xa4ffffb4
 #define UBC_BBRA		0xa4ffffb8
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index bbbdaab8de5f..f4a8da71ba45 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -23,7 +23,7 @@ enum cpu_type {
 	CPU_SH7705, CPU_SH7706, CPU_SH7707,
 	CPU_SH7708, CPU_SH7708S, CPU_SH7708R,
 	CPU_SH7709, CPU_SH7709A, CPU_SH7710, CPU_SH7712,
-	CPU_SH7720, CPU_SH7729,
+	CPU_SH7720, CPU_SH7721, CPU_SH7729,
 
 	/* SH-4 types */
 	CPU_SH7750, CPU_SH7750S, CPU_SH7750R, CPU_SH7751, CPU_SH7751R,
-- 
cgit v1.2.3


From 0465b9fb5f6b57f00a6f5bf2169e30e8f3c7d66c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 26 Dec 2007 18:37:16 +0900
Subject: sh: Fix get_user()/put_user() build error.

Fixes the build error caused by -Werror on gcc 3.x compilers:

arch/sh/kernel/signal_32.c: In function `sys_sigaction':
arch/sh/kernel/signal_32.c:66: warning: initialization discards qualifiers from pointer target type
arch/sh/kernel/signal_32.c:67: warning: initialization discards qualifiers from pointer target type
arch/sh/kernel/signal_32.c:69: warning: initialization discards qualifiers from pointer target type
arch/sh/kernel/signal_32.c:70: warning: initialization discards qualifiers from pointer target type

The mismatch in question was introduced by commit-id
9c5a4eec79b3eb8876d2e7fddfa1e040a7650e55.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/uaccess_32.h | 64 ++++++++++++++++++++-------------------------
 1 file changed, 28 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/uaccess_32.h b/include/asm-sh/uaccess_32.h
index 59a9f20c2dc7..b6082f3c1dc4 100644
--- a/include/asm-sh/uaccess_32.h
+++ b/include/asm-sh/uaccess_32.h
@@ -95,11 +95,9 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 }
 #endif /* CONFIG_MMU */
 
-static inline int access_ok(int type, const void __user *p, unsigned long size)
-{
-	unsigned long addr = (unsigned long)p;
-	return __access_ok(addr, size);
-}
+#define access_ok(type, addr, size)	\
+	(__chk_user_ptr(addr),		\
+	 __access_ok((unsigned long __force)(addr), (size)))
 
 /*
  * Uh, these should become the main single-value transfer routines ...
@@ -113,18 +111,16 @@ static inline int access_ok(int type, const void __user *p, unsigned long size)
  * (a) re-use the arguments for side effects (sizeof is ok)
  * (b) require any knowledge of processes at this stage
  */
-#define put_user(x,ptr)	__put_user_check((x),(ptr),sizeof(*(ptr)))
-#define get_user(x,ptr) __get_user_check((x),(ptr),sizeof(*(ptr)))
+#define put_user(x,ptr)		__put_user_check((x), (ptr), sizeof(*(ptr)))
+#define get_user(x,ptr)		__get_user_check((x), (ptr), sizeof(*(ptr)))
 
 /*
  * The "__xxx" versions do not do address space checking, useful when
  * doing multiple accesses to the same area (the user has to do the
  * checks by hand with "access_ok()")
  */
-#define __put_user(x,ptr) \
-  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
-#define __get_user(x,ptr) \
-  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __put_user(x,ptr)	__put_user_nocheck((x), (ptr), sizeof(*(ptr)))
+#define __get_user(x,ptr)	__get_user_nocheck((x), (ptr), sizeof(*(ptr)))
 
 struct __large_struct { unsigned long buf[100]; };
 #define __m(x) (*(struct __large_struct __user *)(x))
@@ -132,7 +128,6 @@ struct __large_struct { unsigned long buf[100]; };
 #define __get_user_size(x,ptr,size,retval)			\
 do {								\
 	retval = 0;						\
-	__chk_user_ptr(ptr);					\
 	switch (size) {						\
 	case 1:							\
 		__get_user_asm(x, ptr, retval, "b");		\
@@ -151,24 +146,22 @@ do {								\
 
 #define __get_user_nocheck(x,ptr,size)				\
 ({								\
-	long __gu_err, __gu_val;				\
-	__typeof__(*(ptr)) *__pu_addr = (ptr);  \
-	__get_user_size(__gu_val, (__pu_addr), (size), __gu_err);	\
+	long __gu_err;						\
+	unsigned long __gu_val;					\
+	const __typeof__(*(ptr)) __user *__gu_addr = (ptr);	\
+	__chk_user_ptr(ptr);					\
+	__get_user_size(__gu_val, __gu_addr, (size), __gu_err);	\
 	(x) = (__typeof__(*(ptr)))__gu_val;			\
 	__gu_err;						\
 })
 
 #define __get_user_check(x,ptr,size)					\
 ({									\
-	long __gu_err, __gu_val;					\
-	__typeof__(*(ptr)) *__pu_addr = (ptr);				\
-	__chk_user_ptr(__pu_addr);					\
-	if (likely(__addr_ok((unsigned long)(__pu_addr)))) {		\
-		__get_user_size(__gu_val, (__pu_addr), (size), __gu_err);\
-	} else {							\
-		__gu_err = -EFAULT;					\
-		__gu_val = 0;						\
-	}								\
+	long __gu_err = -EFAULT;					\
+	unsigned long __gu_val = 0;					\
+	const __typeof__(*(ptr)) *__gu_addr = (ptr);			\
+	if (likely(access_ok(VERIFY_READ, __gu_addr, (size))))		\
+		__get_user_size(__gu_val, __gu_addr, (size), __gu_err);	\
 	(x) = (__typeof__(*(ptr)))__gu_val;				\
 	__gu_err;							\
 })
@@ -199,7 +192,6 @@ extern void __get_user_unknown(void);
 #define __put_user_size(x,ptr,size,retval)		\
 do {							\
 	retval = 0;					\
-	__chk_user_ptr(ptr);				\
 	switch (size) {					\
 	case 1:						\
 		__put_user_asm(x, ptr, retval, "b");	\
@@ -218,22 +210,22 @@ do {							\
 	}						\
 } while (0)
 
-#define __put_user_nocheck(x,ptr,size)			\
-({							\
-	long __pu_err;					\
-	__put_user_size((x),(ptr),(size),__pu_err);	\
-	__pu_err;					\
+#define __put_user_nocheck(x,ptr,size)				\
+({								\
+	long __pu_err;						\
+	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
+	__chk_user_ptr(ptr);					\
+	__put_user_size((x), __pu_addr, (size), __pu_err);	\
+	__pu_err;						\
 })
 
 #define __put_user_check(x,ptr,size)				\
 ({								\
-	long __pu_err;						\
+	long __pu_err = -EFAULT;				\
 	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
-								\
-	if (likely(__addr_ok((unsigned long)__pu_addr)))	\
-		__put_user_size((x),__pu_addr,(size),__pu_err);	\
-	else							\
-		__pu_err = -EFAULT;				\
+	if (likely(access_ok(VERIFY_WRITE, __pu_addr, size)))	\
+		__put_user_size((x), __pu_addr, (size),		\
+				__pu_err);			\
 	__pu_err;						\
 })
 
-- 
cgit v1.2.3


From 7d740a066fb9c6681c2898c7977209725c9e552f Mon Sep 17 00:00:00 2001
From: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Date: Mon, 7 Jan 2008 14:40:07 +0900
Subject: sh: Add support for SH7763 CPU subtype.

Signed-off-by: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                        |   9 +-
 arch/sh/Kconfig.debug                  |   2 +-
 arch/sh/drivers/pci/Makefile           |   1 +
 arch/sh/drivers/pci/pci-sh4.h          |   4 +-
 arch/sh/drivers/pci/pci-sh7780.c       |   1 +
 arch/sh/drivers/pci/pci-sh7780.h       |   1 +
 arch/sh/kernel/cpu/sh4/probe.c         |   2 +
 arch/sh/kernel/cpu/sh4a/Makefile       |   2 +
 arch/sh/kernel/cpu/sh4a/clock-sh7763.c | 126 +++++++++++
 arch/sh/kernel/cpu/sh4a/setup-sh7763.c | 390 +++++++++++++++++++++++++++++++++
 arch/sh/kernel/setup.c                 |   8 +-
 drivers/serial/sh-sci.c                |   4 +-
 drivers/serial/sh-sci.h                |  19 +-
 include/asm-sh/bugs.h                  |   2 +-
 include/asm-sh/cpu-sh4/freq.h          |   3 +-
 include/asm-sh/processor.h             |   2 +-
 16 files changed, 564 insertions(+), 12 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh4a/clock-sh7763.c
 create mode 100644 arch/sh/kernel/cpu/sh4a/setup-sh7763.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d2f557719489..ddbd610d27e7 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -265,6 +265,12 @@ config CPU_SUBTYPE_SH4_202
 
 # SH-4A Processor Support
 
+config CPU_SUBTYPE_SH7763
+	bool "Support SH7763 processor"
+	select CPU_SH4A
+	help
+	  Select SH7763 if you have a SH4A SH7763(R5S77631) CPU.
+
 config CPU_SUBTYPE_SH7770
 	bool "Support SH7770 processor"
 	select CPU_SH4A
@@ -551,7 +557,8 @@ config SH_MTU2
 
 config SH_TIMER_IRQ
 	int
-	default "28" if CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785
+	default "28" if CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785 || \
+			CPU_SUBTYPE_SH7763
 	default "86" if CPU_SUBTYPE_SH7619
 	default "140" if CPU_SUBTYPE_SH7206
 	default "16"
diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index dde00d5c2478..f7c716166ce8 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -29,7 +29,7 @@ config EARLY_SCIF_CONSOLE
 config EARLY_SCIF_CONSOLE_PORT
 	hex
 	depends on EARLY_SCIF_CONSOLE
-	default "0xffe00000" if CPU_SUBTYPE_SH7780
+ 	default "0xffe00000" if CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7763
 	default "0xffea0000" if CPU_SUBTYPE_SH7785
 	default "0xfffe8000" if CPU_SUBTYPE_SH7203
 	default "0xfffe9800" if CPU_SUBTYPE_SH7206 || CPU_SUBTYPE_SH7263
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 172e81db9530..1086cf111f8b 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_PCI_AUTO)			+= pci-auto.o
 
 obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7751R)	+= pci-sh7751.o ops-sh4.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7763)	+= pci-sh7780.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o ops-sh4.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7785)	+= pci-sh7780.o ops-sh4.o
 obj-$(CONFIG_CPU_SH5)			+= pci-sh5.o ops-sh5.o
diff --git a/arch/sh/drivers/pci/pci-sh4.h b/arch/sh/drivers/pci/pci-sh4.h
index 1901c33cde6a..4925c79ea959 100644
--- a/arch/sh/drivers/pci/pci-sh4.h
+++ b/arch/sh/drivers/pci/pci-sh4.h
@@ -1,7 +1,9 @@
 #ifndef __PCI_SH4_H
 #define __PCI_SH4_H
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7780) || defined(CONFIG_CPU_SUBTYPE_SH7785)
+#if defined(CONFIG_CPU_SUBTYPE_SH7780) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7785) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7763)
 #include "pci-sh7780.h"
 #else
 #include "pci-sh7751.h"
diff --git a/arch/sh/drivers/pci/pci-sh7780.c b/arch/sh/drivers/pci/pci-sh7780.c
index e516087fb435..7d797f4de5e7 100644
--- a/arch/sh/drivers/pci/pci-sh7780.c
+++ b/arch/sh/drivers/pci/pci-sh7780.c
@@ -58,6 +58,7 @@ static int __init sh7780_pci_init(void)
 	id = pci_read_reg(SH7780_PCIVID);
 	if ((id & 0xffff) == SH7780_VENDOR_ID) {
 		switch ((id >> 16) & 0xffff) {
+		case SH7763_DEVICE_ID:
 		case SH7780_DEVICE_ID:
 		case SH7781_DEVICE_ID:
 		case SH7785_DEVICE_ID:
diff --git a/arch/sh/drivers/pci/pci-sh7780.h b/arch/sh/drivers/pci/pci-sh7780.h
index 1d069a859de2..97b2c98f05c4 100644
--- a/arch/sh/drivers/pci/pci-sh7780.h
+++ b/arch/sh/drivers/pci/pci-sh7780.h
@@ -16,6 +16,7 @@
 #define SH7780_VENDOR_ID	0x1912
 #define SH7781_DEVICE_ID	0x0001
 #define SH7780_DEVICE_ID	0x0002
+#define SH7763_DEVICE_ID	0x0004
 #define SH7785_DEVICE_ID	0x0007
 
 /* SH7780 Control Registers */
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index bc9c28a69bf1..f2b9238cda04 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -98,6 +98,8 @@ int __init detect_cpu_and_cache_system(void)
 	case 0x200A:
 		if (prr == 0x61)
 			boot_cpu_data.type = CPU_SH7781;
+		else if (prr == 0xa1)
+			boot_cpu_data.type = CPU_SH7763;
 		else
 			boot_cpu_data.type = CPU_SH7780;
 
diff --git a/arch/sh/kernel/cpu/sh4a/Makefile b/arch/sh/kernel/cpu/sh4a/Makefile
index 24539873943a..08ac6387bf17 100644
--- a/arch/sh/kernel/cpu/sh4a/Makefile
+++ b/arch/sh/kernel/cpu/sh4a/Makefile
@@ -3,6 +3,7 @@
 #
 
 # CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7763)	+= setup-sh7763.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7770)	+= setup-sh7770.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7785)	+= setup-sh7785.o
@@ -14,6 +15,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SHX3)		+= setup-shx3.o
 smp-$(CONFIG_CPU_SUBTYPE_SHX3)		:= smp-shx3.o
 
 # Primary on-chip clocks (common)
+clock-$(CONFIG_CPU_SUBTYPE_SH7763)	:= clock-sh7763.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7770)	:= clock-sh7770.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7780)	:= clock-sh7780.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7785)	:= clock-sh7785.o
diff --git a/arch/sh/kernel/cpu/sh4a/clock-sh7763.c b/arch/sh/kernel/cpu/sh4a/clock-sh7763.c
new file mode 100644
index 000000000000..45889d412c80
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4a/clock-sh7763.c
@@ -0,0 +1,126 @@
+/*
+ * arch/sh/kernel/cpu/sh4a/clock-sh7763.c
+ *
+ * SH7763 support for the clock framework
+ *
+ *  Copyright (C) 2005  Paul Mundt
+ *  Copyright (C) 2007  Yoshihiro Shimoda
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+static int bfc_divisors[] = { 1, 1, 1, 8, 1, 1, 1, 1 };
+static int p0fc_divisors[] = { 1, 1, 1, 8, 1, 1, 1, 1 };
+static int p1fc_divisors[] = { 1, 1, 1, 16, 1, 1, 1, 1 };
+static int cfc_divisors[] = { 1, 1, 4, 1, 1, 1, 1, 1 };
+
+static void master_clk_init(struct clk *clk)
+{
+	clk->rate *= p0fc_divisors[(ctrl_inl(FRQCR) >> 4) & 0x07];
+}
+
+static struct clk_ops sh7763_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int idx = ((ctrl_inl(FRQCR) >> 4) & 0x07);
+	clk->rate = clk->parent->rate / p0fc_divisors[idx];
+}
+
+static struct clk_ops sh7763_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void bus_clk_recalc(struct clk *clk)
+{
+	int idx = ((ctrl_inl(FRQCR) >> 16) & 0x07);
+	clk->rate = clk->parent->rate / bfc_divisors[idx];
+}
+
+static struct clk_ops sh7763_bus_clk_ops = {
+	.recalc		= bus_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	clk->rate = clk->parent->rate;
+}
+
+static struct clk_ops sh7763_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7763_clk_ops[] = {
+	&sh7763_master_clk_ops,
+	&sh7763_module_clk_ops,
+	&sh7763_bus_clk_ops,
+	&sh7763_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7763_clk_ops))
+		*ops = sh7763_clk_ops[idx];
+}
+
+static void shyway_clk_recalc(struct clk *clk)
+{
+	int idx = ((ctrl_inl(FRQCR) >> 20) & 0x07);
+	clk->rate = clk->parent->rate / cfc_divisors[idx];
+}
+
+static struct clk_ops sh7763_shyway_clk_ops = {
+	.recalc		= shyway_clk_recalc,
+};
+
+static struct clk sh7763_shyway_clk = {
+	.name		= "shyway_clk",
+	.flags		= CLK_ALWAYS_ENABLED,
+	.ops		= &sh7763_shyway_clk_ops,
+};
+
+/*
+ * Additional SH7763-specific on-chip clocks that aren't already part of the
+ * clock framework
+ */
+static struct clk *sh7763_onchip_clocks[] = {
+	&sh7763_shyway_clk,
+};
+
+static int __init sh7763_clk_init(void)
+{
+	struct clk *clk = clk_get(NULL, "master_clk");
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(sh7763_onchip_clocks); i++) {
+		struct clk *clkp = sh7763_onchip_clocks[i];
+
+		clkp->parent = clk;
+		clk_register(clkp);
+		clk_enable(clkp);
+	}
+
+	/*
+	 * Now that we have the rest of the clocks registered, we need to
+	 * force the parent clock to propagate so that these clocks will
+	 * automatically figure out their rate. We cheat by handing the
+	 * parent clock its current rate and forcing child propagation.
+	 */
+	clk_set_rate(clk, clk_get_rate(clk));
+
+	clk_put(clk);
+
+	return 0;
+}
+
+arch_initcall(sh7763_clk_init);
+
diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7763.c b/arch/sh/kernel/cpu/sh4a/setup-sh7763.c
new file mode 100644
index 000000000000..eabd5386812d
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4a/setup-sh7763.c
@@ -0,0 +1,390 @@
+/*
+ * SH7763 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *  Copyright (C) 2007  Yoshihiro Shimoda
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <linux/io.h>
+#include <asm/sci.h>
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		.start	= 0xffe80000,
+		.end	= 0xffe80000 + 0x58 - 1,
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= 21,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= 22,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= 20,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.mapbase	= 0xffe08000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct resource usb_ohci_resources[] = {
+	[0] = {
+		.start	= 0xffec8000,
+		.end	= 0xffec80ff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= 83,
+		.end	= 83,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static u64 usb_ohci_dma_mask = 0xffffffffUL;
+static struct platform_device usb_ohci_device = {
+	.name		= "sh_ohci",
+	.id		= -1,
+	.dev = {
+		.dma_mask		= &usb_ohci_dma_mask,
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(usb_ohci_resources),
+	.resource	= usb_ohci_resources,
+};
+
+static struct resource usbf_resources[] = {
+	[0] = {
+		.start	= 0xffec0000,
+		.end	= 0xffec00ff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= 84,
+		.end	= 84,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device usbf_device = {
+	.name		= "sh_udc",
+	.id		= -1,
+	.dev = {
+		.dma_mask		= NULL,
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(usbf_resources),
+	.resource	= usbf_resources,
+};
+
+static struct platform_device *sh7763_devices[] __initdata = {
+	&rtc_device,
+	&sci_device,
+	&usb_ohci_device,
+	&usbf_device,
+};
+
+static int __init sh7763_devices_setup(void)
+{
+	return platform_add_devices(sh7763_devices,
+				    ARRAY_SIZE(sh7763_devices));
+}
+__initcall(sh7763_devices_setup);
+
+enum {
+	UNUSED = 0,
+
+	/* interrupt sources */
+
+	IRL_LLLL, IRL_LLLH, IRL_LLHL, IRL_LLHH,
+	IRL_LHLL, IRL_LHLH, IRL_LHHL, IRL_LHHH,
+	IRL_HLLL, IRL_HLLH, IRL_HLHL, IRL_HLHH,
+	IRL_HHLL, IRL_HHLH, IRL_HHHL,
+
+	IRQ0, IRQ1, IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, IRQ7,
+	RTC_ATI, RTC_PRI, RTC_CUI,
+	WDT, TMU0, TMU1, TMU2, TMU2_TICPI,
+	HUDI, LCDC,
+	DMAC0_DMINT0, DMAC0_DMINT1, DMAC0_DMINT2, DMAC0_DMINT3, DMAC0_DMAE,
+	SCIF0_ERI, SCIF0_RXI, SCIF0_BRI, SCIF0_TXI,
+	DMAC0_DMINT4, DMAC0_DMINT5,
+	IIC0, IIC1,
+	CMT,
+	GEINT0, GEINT1, GEINT2,
+	HAC,
+	PCISERR, PCIINTA, PCIINTB, PCIINTC, PCIINTD,
+	PCIERR, PCIPWD3, PCIPWD2, PCIPWD1, PCIPWD0,
+	STIF0, STIF1,
+	SCIF1_ERI, SCIF1_RXI, SCIF1_BRI, SCIF1_TXI,
+	SIOF0, SIOF1, SIOF2,
+	USBH, USBFI0, USBFI1,
+	TPU, PCC,
+	MMCIF_FSTAT, MMCIF_TRAN, MMCIF_ERR, MMCIF_FRDY,
+	SIM_ERI, SIM_RXI, SIM_TXI, SIM_TEND,
+	TMU3, TMU4, TMU5, ADC, SSI0, SSI1, SSI2, SSI3,
+	SCIF2_ERI, SCIF2_RXI, SCIF2_BRI, SCIF2_TXI,
+	GPIO_CH0, GPIO_CH1, GPIO_CH2, GPIO_CH3,
+
+	/* interrupt groups */
+
+	TMU012, TMU345, RTC, DMAC, SCIF0, GETHER, PCIC5,
+	SCIF1, USBF, MMCIF, SIM, SCIF2, GPIO,
+};
+
+static struct intc_vect vectors[] __initdata = {
+	INTC_VECT(RTC_ATI, 0x480), INTC_VECT(RTC_PRI, 0x4a0),
+	INTC_VECT(RTC_CUI, 0x4c0),
+	INTC_VECT(WDT, 0x560), INTC_VECT(TMU0, 0x580),
+	INTC_VECT(TMU1, 0x5a0), INTC_VECT(TMU2, 0x5c0),
+	INTC_VECT(TMU2_TICPI, 0x5e0), INTC_VECT(HUDI, 0x600),
+	INTC_VECT(LCDC, 0x620),
+	INTC_VECT(DMAC0_DMINT0, 0x640), INTC_VECT(DMAC0_DMINT1, 0x660),
+	INTC_VECT(DMAC0_DMINT2, 0x680), INTC_VECT(DMAC0_DMINT3, 0x6a0),
+	INTC_VECT(DMAC0_DMAE, 0x6c0),
+	INTC_VECT(SCIF0_ERI, 0x700), INTC_VECT(SCIF0_RXI, 0x720),
+	INTC_VECT(SCIF0_BRI, 0x740), INTC_VECT(SCIF0_TXI, 0x760),
+	INTC_VECT(DMAC0_DMINT4, 0x780), INTC_VECT(DMAC0_DMINT5, 0x7a0),
+	INTC_VECT(IIC0, 0x8A0), INTC_VECT(IIC1, 0x8C0),
+	INTC_VECT(CMT, 0x900), INTC_VECT(GEINT0, 0x920),
+	INTC_VECT(GEINT1, 0x940), INTC_VECT(GEINT2, 0x960),
+	INTC_VECT(HAC, 0x980),
+	INTC_VECT(PCISERR, 0xa00), INTC_VECT(PCIINTA, 0xa20),
+	INTC_VECT(PCIINTB, 0xa40), INTC_VECT(PCIINTC, 0xa60),
+	INTC_VECT(PCIINTD, 0xa80), INTC_VECT(PCIERR, 0xaa0),
+	INTC_VECT(PCIPWD3, 0xac0), INTC_VECT(PCIPWD2, 0xae0),
+	INTC_VECT(PCIPWD1, 0xb00), INTC_VECT(PCIPWD0, 0xb20),
+	INTC_VECT(STIF0, 0xb40), INTC_VECT(STIF1, 0xb60),
+	INTC_VECT(SCIF1_ERI, 0xb80), INTC_VECT(SCIF1_RXI, 0xba0),
+	INTC_VECT(SCIF1_BRI, 0xbc0), INTC_VECT(SCIF1_TXI, 0xbe0),
+	INTC_VECT(SIOF0, 0xc00), INTC_VECT(SIOF1, 0xc20),
+	INTC_VECT(USBH, 0xc60), INTC_VECT(USBFI0, 0xc80),
+	INTC_VECT(USBFI1, 0xca0),
+	INTC_VECT(TPU, 0xcc0), INTC_VECT(PCC, 0xce0),
+	INTC_VECT(MMCIF_FSTAT, 0xd00), INTC_VECT(MMCIF_TRAN, 0xd20),
+	INTC_VECT(MMCIF_ERR, 0xd40), INTC_VECT(MMCIF_FRDY, 0xd60),
+	INTC_VECT(SIM_ERI, 0xd80), INTC_VECT(SIM_RXI, 0xda0),
+	INTC_VECT(SIM_TXI, 0xdc0), INTC_VECT(SIM_TEND, 0xde0),
+	INTC_VECT(TMU3, 0xe00), INTC_VECT(TMU4, 0xe20),
+	INTC_VECT(TMU5, 0xe40), INTC_VECT(ADC, 0xe60),
+	INTC_VECT(SSI0, 0xe80), INTC_VECT(SSI1, 0xea0),
+	INTC_VECT(SSI2, 0xec0), INTC_VECT(SSI3, 0xee0),
+	INTC_VECT(SCIF1_ERI, 0xf00), INTC_VECT(SCIF1_RXI, 0xf20),
+	INTC_VECT(SCIF1_BRI, 0xf40), INTC_VECT(SCIF1_TXI, 0xf60),
+	INTC_VECT(GPIO_CH0, 0xf80), INTC_VECT(GPIO_CH1, 0xfa0),
+	INTC_VECT(GPIO_CH2, 0xfc0), INTC_VECT(GPIO_CH3, 0xfe0),
+};
+
+static struct intc_group groups[] __initdata = {
+	INTC_GROUP(TMU012, TMU0, TMU1, TMU2, TMU2_TICPI),
+	INTC_GROUP(TMU345, TMU3, TMU4, TMU5),
+	INTC_GROUP(RTC, RTC_ATI, RTC_PRI, RTC_CUI),
+	INTC_GROUP(DMAC, DMAC0_DMINT0, DMAC0_DMINT1, DMAC0_DMINT2,
+		   DMAC0_DMINT3, DMAC0_DMINT4, DMAC0_DMINT5, DMAC0_DMAE),
+	INTC_GROUP(SCIF0, SCIF0_ERI, SCIF0_RXI, SCIF0_BRI, SCIF0_TXI),
+	INTC_GROUP(GETHER, GEINT0, GEINT1, GEINT2),
+	INTC_GROUP(PCIC5, PCIERR, PCIPWD3, PCIPWD2, PCIPWD1, PCIPWD0),
+	INTC_GROUP(SCIF1, SCIF1_ERI, SCIF1_RXI, SCIF1_BRI, SCIF1_TXI),
+	INTC_GROUP(USBF, USBFI0, USBFI1),
+	INTC_GROUP(MMCIF, MMCIF_FSTAT, MMCIF_TRAN, MMCIF_ERR, MMCIF_FRDY),
+	INTC_GROUP(SIM, SIM_ERI, SIM_RXI, SIM_TXI, SIM_TEND),
+	INTC_GROUP(SCIF2, SCIF2_ERI, SCIF2_RXI, SCIF2_BRI, SCIF2_TXI),
+	INTC_GROUP(GPIO, GPIO_CH0, GPIO_CH1, GPIO_CH2, GPIO_CH3),
+};
+
+static struct intc_prio priorities[] __initdata = {
+	INTC_PRIO(SCIF0, 3),
+	INTC_PRIO(SCIF1, 3),
+	INTC_PRIO(SCIF2, 3),
+};
+
+static struct intc_mask_reg mask_registers[] __initdata = {
+	{ 0xffd40038, 0xffd4003c, 32, /* INT2MSKR / INT2MSKCR */
+	  { 0, 0, 0, 0, 0, 0, GPIO, 0,
+	    SSI0, MMCIF, 0, SIOF0, PCIC5, PCIINTD, PCIINTC, PCIINTB,
+	    PCIINTA, PCISERR, HAC, CMT, 0, 0, 0, DMAC,
+	    HUDI, 0, WDT, SCIF1, SCIF0, RTC, TMU345, TMU012 } },
+	{ 0xffd400d0, 0xffd400d4, 32, /* INT2MSKR1 / INT2MSKCR1 */
+	  { 0, 0, 0, 0, 0, 0, SCIF2, USBF,
+	    0, 0, STIF1, STIF0, 0, 0, USBH, GETHER,
+	    PCC, 0, 0, ADC, TPU, SIM, SIOF2, SIOF1,
+	    LCDC, 0, IIC1, IIC0, SSI3, SSI2, SSI1, 0 } },
+};
+
+static struct intc_prio_reg prio_registers[] __initdata = {
+	{ 0xffd40000, 0, 32, 8, /* INT2PRI0 */ { TMU0, TMU1,
+						 TMU2, TMU2_TICPI } },
+	{ 0xffd40004, 0, 32, 8, /* INT2PRI1 */ { TMU3, TMU4, TMU5, RTC } },
+	{ 0xffd40008, 0, 32, 8, /* INT2PRI2 */ { SCIF0, SCIF1, WDT } },
+	{ 0xffd4000c, 0, 32, 8, /* INT2PRI3 */ { HUDI, DMAC, ADC } },
+	{ 0xffd40010, 0, 32, 8, /* INT2PRI4 */ { CMT, HAC,
+						 PCISERR, PCIINTA } },
+	{ 0xffd40014, 0, 32, 8, /* INT2PRI5 */ { PCIINTB, PCIINTC,
+						 PCIINTD, PCIC5 } },
+	{ 0xffd40018, 0, 32, 8, /* INT2PRI6 */ { SIOF0, USBF, MMCIF, SSI0 } },
+	{ 0xffd4001c, 0, 32, 8, /* INT2PRI7 */ { SCIF2, GPIO } },
+	{ 0xffd400a0, 0, 32, 8, /* INT2PRI8 */ { SSI3, SSI2, SSI1, 0 } },
+	{ 0xffd400a4, 0, 32, 8, /* INT2PRI9 */ { LCDC, 0, IIC1, IIC0 } },
+	{ 0xffd400a8, 0, 32, 8, /* INT2PRI10 */ { TPU, SIM, SIOF2, SIOF1 } },
+	{ 0xffd400ac, 0, 32, 8, /* INT2PRI11 */ { PCC } },
+	{ 0xffd400b0, 0, 32, 8, /* INT2PRI12 */ { 0, 0, USBH, GETHER } },
+	{ 0xffd400b4, 0, 32, 8, /* INT2PRI13 */ { 0, 0, STIF1, STIF0 } },
+};
+
+static DECLARE_INTC_DESC(intc_desc, "sh7763", vectors, groups, priorities,
+			 mask_registers, prio_registers, NULL);
+
+/* Support for external interrupt pins in IRQ mode */
+
+static struct intc_vect irq_vectors[] __initdata = {
+	INTC_VECT(IRQ0, 0x240), INTC_VECT(IRQ1, 0x280),
+	INTC_VECT(IRQ2, 0x2c0), INTC_VECT(IRQ3, 0x300),
+	INTC_VECT(IRQ4, 0x340), INTC_VECT(IRQ5, 0x380),
+	INTC_VECT(IRQ6, 0x3c0), INTC_VECT(IRQ7, 0x200),
+};
+
+static struct intc_mask_reg irq_mask_registers[] __initdata = {
+	{ 0xffd00044, 0xffd00064, 32, /* INTMSK0 / INTMSKCLR0 */
+	  { IRQ0, IRQ1, IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, IRQ7 } },
+};
+
+static struct intc_prio_reg irq_prio_registers[] __initdata = {
+	{ 0xffd00010, 0, 32, 4, /* INTPRI */ { IRQ0, IRQ1, IRQ2, IRQ3,
+					       IRQ4, IRQ5, IRQ6, IRQ7 } },
+};
+
+static struct intc_sense_reg irq_sense_registers[] __initdata = {
+	{ 0xffd0001c, 32, 2, /* ICR1 */   { IRQ0, IRQ1, IRQ2, IRQ3,
+					    IRQ4, IRQ5, IRQ6, IRQ7 } },
+};
+
+static DECLARE_INTC_DESC(intc_irq_desc, "sh7763-irq", irq_vectors,
+			 NULL, NULL, irq_mask_registers, irq_prio_registers,
+			 irq_sense_registers);
+
+/* External interrupt pins in IRL mode */
+
+static struct intc_vect irl_vectors[] __initdata = {
+	INTC_VECT(IRL_LLLL, 0x200), INTC_VECT(IRL_LLLH, 0x220),
+	INTC_VECT(IRL_LLHL, 0x240), INTC_VECT(IRL_LLHH, 0x260),
+	INTC_VECT(IRL_LHLL, 0x280), INTC_VECT(IRL_LHLH, 0x2a0),
+	INTC_VECT(IRL_LHHL, 0x2c0), INTC_VECT(IRL_LHHH, 0x2e0),
+	INTC_VECT(IRL_HLLL, 0x300), INTC_VECT(IRL_HLLH, 0x320),
+	INTC_VECT(IRL_HLHL, 0x340), INTC_VECT(IRL_HLHH, 0x360),
+	INTC_VECT(IRL_HHLL, 0x380), INTC_VECT(IRL_HHLH, 0x3a0),
+	INTC_VECT(IRL_HHHL, 0x3c0),
+};
+
+static struct intc_mask_reg irl3210_mask_registers[] __initdata = {
+	{ 0xffd40080, 0xffd40084, 32, /* INTMSK2 / INTMSKCLR2 */
+	  { IRL_LLLL, IRL_LLLH, IRL_LLHL, IRL_LLHH,
+	    IRL_LHLL, IRL_LHLH, IRL_LHHL, IRL_LHHH,
+	    IRL_HLLL, IRL_HLLH, IRL_HLHL, IRL_HLHH,
+	    IRL_HHLL, IRL_HHLH, IRL_HHHL, } },
+};
+
+static struct intc_mask_reg irl7654_mask_registers[] __initdata = {
+	{ 0xffd40080, 0xffd40084, 32, /* INTMSK2 / INTMSKCLR2 */
+	  { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	    IRL_LLLL, IRL_LLLH, IRL_LLHL, IRL_LLHH,
+	    IRL_LHLL, IRL_LHLH, IRL_LHHL, IRL_LHHH,
+	    IRL_HLLL, IRL_HLLH, IRL_HLHL, IRL_HLHH,
+	    IRL_HHLL, IRL_HHLH, IRL_HHHL, } },
+};
+
+static DECLARE_INTC_DESC(intc_irl7654_desc, "sh7763-irl7654", irl_vectors,
+			 NULL, NULL, irl7654_mask_registers, NULL, NULL);
+
+static DECLARE_INTC_DESC(intc_irl3210_desc, "sh7763-irl3210", irl_vectors,
+			 NULL, NULL, irl3210_mask_registers, NULL, NULL);
+
+#define INTC_ICR0	0xffd00000
+#define INTC_INTMSK0	0xffd00044
+#define INTC_INTMSK1	0xffd00048
+#define INTC_INTMSK2	0xffd40080
+#define INTC_INTMSKCLR1	0xffd00068
+#define INTC_INTMSKCLR2	0xffd40084
+
+void __init plat_irq_setup(void)
+{
+	/* disable IRQ7-0 */
+	ctrl_outl(0xff000000, INTC_INTMSK0);
+
+	/* disable IRL3-0 + IRL7-4 */
+	ctrl_outl(0xc0000000, INTC_INTMSK1);
+	ctrl_outl(0xfffefffe, INTC_INTMSK2);
+
+	register_intc_controller(&intc_desc);
+}
+
+void __init plat_irq_setup_pins(int mode)
+{
+	switch (mode) {
+	case IRQ_MODE_IRQ:
+		/* select IRQ mode for IRL3-0 + IRL7-4 */
+		ctrl_outl(ctrl_inl(INTC_ICR0) | 0x00c00000, INTC_ICR0);
+		register_intc_controller(&intc_irq_desc);
+		break;
+	case IRQ_MODE_IRL7654:
+		/* enable IRL7-4 but don't provide any masking */
+		ctrl_outl(0x40000000, INTC_INTMSKCLR1);
+		ctrl_outl(0x0000fffe, INTC_INTMSKCLR2);
+		break;
+	case IRQ_MODE_IRL3210:
+		/* enable IRL0-3 but don't provide any masking */
+		ctrl_outl(0x80000000, INTC_INTMSKCLR1);
+		ctrl_outl(0xfffe0000, INTC_INTMSKCLR2);
+		break;
+	case IRQ_MODE_IRL7654_MASK:
+		/* enable IRL7-4 and mask using cpu intc controller */
+		ctrl_outl(0x40000000, INTC_INTMSKCLR1);
+		register_intc_controller(&intc_irl7654_desc);
+		break;
+	case IRQ_MODE_IRL3210_MASK:
+		/* enable IRL0-3 and mask using cpu intc controller */
+		ctrl_outl(0x80000000, INTC_INTMSKCLR1);
+		register_intc_controller(&intc_irl3210_desc);
+		break;
+	default:
+		BUG();
+	}
+}
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 48cd576754b5..9cf1d2e998bf 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -319,10 +319,10 @@ static const char *cpu_name[] = {
 	[CPU_SH7750R]	= "SH7750R",	[CPU_SH7751]	= "SH7751",
 	[CPU_SH7751R]	= "SH7751R",	[CPU_SH7760]	= "SH7760",
 	[CPU_SH4_202]	= "SH4-202",	[CPU_SH4_501]	= "SH4-501",
-	[CPU_SH7770]	= "SH7770",	[CPU_SH7780]	= "SH7780",
-	[CPU_SH7781]	= "SH7781",	[CPU_SH7343]	= "SH7343",
-	[CPU_SH7785]	= "SH7785",	[CPU_SH7722]	= "SH7722",
-	[CPU_SHX3]	= "SH-X3",
+	[CPU_SH7763]	= "SH7763",	[CPU_SH7770]	= "SH7770",
+	[CPU_SH7780]	= "SH7780",	[CPU_SH7781]	= "SH7781",
+	[CPU_SH7343]	= "SH7343",	[CPU_SH7785]	= "SH7785",
+	[CPU_SH7722]	= "SH7722",	[CPU_SHX3]	= "SH-X3",
 	[CPU_SH5_101]	= "SH5-101",	[CPU_SH5_103]	= "SH5-103",
 	[CPU_SH_NONE]	= "Unknown"
 };
diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c
index 6fbfd140b7e7..ddf639144538 100644
--- a/drivers/serial/sh-sci.c
+++ b/drivers/serial/sh-sci.c
@@ -395,7 +395,8 @@ static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 	} else {
 #ifdef CONFIG_CPU_SUBTYPE_SH7343
 		/* Nothing */
-#elif defined(CONFIG_CPU_SUBTYPE_SH7780) || \
+#elif defined(CONFIG_CPU_SUBTYPE_SH7763) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7780) || \
       defined(CONFIG_CPU_SUBTYPE_SH7785) || \
       defined(CONFIG_CPU_SUBTYPE_SHX3)
 		ctrl_outw(0x0080, SCSPTR0); /* Set RTS = 1 */
@@ -408,6 +409,7 @@ static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7760) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7763) || \
     defined(CONFIG_CPU_SUBTYPE_SH7780) || \
     defined(CONFIG_CPU_SUBTYPE_SH7785)
 static inline int scif_txroom(struct uart_port *port)
diff --git a/drivers/serial/sh-sci.h b/drivers/serial/sh-sci.h
index 85562040a6d3..f5764ebcfe07 100644
--- a/drivers/serial/sh-sci.h
+++ b/drivers/serial/sh-sci.h
@@ -120,6 +120,12 @@
 # define SCSCR_INIT(port)          0x30 /* TIE=0,RIE=0,TE=1,RE=1 */
 # define SCI_ONLY
 # define H8300_SCI_DR(ch) *(volatile char *)(P1DR + h8300_sci_pins[ch].port)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7763)
+# define SCSPTR0 0xffe00024 /* 16 bit SCIF */
+# define SCSPTR1 0xffe08024 /* 16 bit SCIF */
+# define SCIF_ORER 0x0001  /* overrun error bit */
+# define SCSCR_INIT(port)	0x3a	/* TIE=0,RIE=0,TE=1,RE=1,REIE=1 */
+# define SCIF_ONLY
 #elif defined(CONFIG_CPU_SUBTYPE_SH7770)
 # define SCSPTR0 0xff923020 /* 16 bit SCIF */
 # define SCSPTR1 0xff924020 /* 16 bit SCIF */
@@ -419,6 +425,7 @@ SCIx_FNS(SCxSR,  0x08,  8, 0x10,  8, 0x08, 16, 0x10, 16, 0x04,  8)
 SCIx_FNS(SCxRDR, 0x0a,  8, 0x14,  8, 0x0A,  8, 0x14,  8, 0x05,  8)
 SCIF_FNS(SCFCR,                      0x0c,  8, 0x18, 16)
 #if defined(CONFIG_CPU_SUBTYPE_SH7760) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7763) || \
     defined(CONFIG_CPU_SUBTYPE_SH7780) || \
     defined(CONFIG_CPU_SUBTYPE_SH7785)
 SCIF_FNS(SCFDR,			     0x0e, 16, 0x1C, 16)
@@ -588,6 +595,15 @@ static inline int sci_rxd_in(struct uart_port *port)
 	int ch = (port->mapbase - SMR0) >> 3;
 	return (H8300_SCI_DR(ch) & h8300_sci_pins[ch].rx) ? 1 : 0;
 }
+#elif defined(CONFIG_CPU_SUBTYPE_SH7763)
+static inline int sci_rxd_in(struct uart_port *port)
+{
+	if (port->mapbase == 0xffe00000)
+		return ctrl_inw(SCSPTR0) & 0x0001 ? 1 : 0; /* SCIF */
+	if (port->mapbase == 0xffe08000)
+		return ctrl_inw(SCSPTR1) & 0x0001 ? 1 : 0; /* SCIF */
+	return 1;
+}
 #elif defined(CONFIG_CPU_SUBTYPE_SH7770)
 static inline int sci_rxd_in(struct uart_port *port)
 {
@@ -698,7 +714,8 @@ static inline int sci_rxd_in(struct uart_port *port)
  * -- Mitch Davis - 15 Jul 2000
  */
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7780) || \
+#if defined(CONFIG_CPU_SUBTYPE_SH7763) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7780) || \
     defined(CONFIG_CPU_SUBTYPE_SH7785)
 #define SCBRR_VALUE(bps, clk) ((clk+16*bps)/(16*bps)-1)
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index 9ce67fd51190..def8128b8b78 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -35,7 +35,7 @@ static void __init check_bugs(void)
 	case CPU_SH7750 ... CPU_SH4_501:
 		*p++ = '4';
 		break;
-	case CPU_SH7770 ... CPU_SHX3:
+	case CPU_SH7763 ... CPU_SHX3:
 		*p++ = '4';
 		*p++ = 'a';
 		break;
diff --git a/include/asm-sh/cpu-sh4/freq.h b/include/asm-sh/cpu-sh4/freq.h
index dc1d32a86374..1ac10b9a078f 100644
--- a/include/asm-sh/cpu-sh4/freq.h
+++ b/include/asm-sh/cpu-sh4/freq.h
@@ -16,7 +16,8 @@
 #define SCLKACR			0xa4150008
 #define SCLKBCR			0xa415000c
 #define IrDACLKCR		0xa4150010
-#elif defined(CONFIG_CPU_SUBTYPE_SH7780)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7763) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7780)
 #define	FRQCR			0xffc80000
 #elif defined(CONFIG_CPU_SUBTYPE_SH7785)
 #define FRQCR0			0xffc80000
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index f4a8da71ba45..c9b14161f73d 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -30,7 +30,7 @@ enum cpu_type {
 	CPU_SH7760, CPU_SH4_202, CPU_SH4_501,
 
 	/* SH-4A types */
-	CPU_SH7770, CPU_SH7780, CPU_SH7781, CPU_SH7785, CPU_SHX3,
+	CPU_SH7763, CPU_SH7770, CPU_SH7780, CPU_SH7781, CPU_SH7785, CPU_SHX3,
 
 	/* SH4AL-DSP types */
 	CPU_SH7343, CPU_SH7722,
-- 
cgit v1.2.3


From c3aa92afd0a6c253df974556b4a43c0a182d1fc4 Mon Sep 17 00:00:00 2001
From: Andrew Murray <amurray@mpc-data.co.uk>
Date: Tue, 8 Jan 2008 09:56:45 +0900
Subject: sh: sh7712 clock support

This patch provides specific clock support for the SH7712.

Signed-off-by: Andrew Murray <amurray@mpc-data.co.uk>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh3/Makefile       |  1 +
 arch/sh/kernel/cpu/sh3/clock-sh7712.c | 71 +++++++++++++++++++++++++++++++++++
 include/asm-sh/cpu-sh3/freq.h         |  5 +++
 3 files changed, 77 insertions(+)
 create mode 100644 arch/sh/kernel/cpu/sh3/clock-sh7712.c

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index 1afd05e62d1c..3ae4d9111f19 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -22,5 +22,6 @@ clock-$(CONFIG_CPU_SUBTYPE_SH7706)	:= clock-sh7706.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7709)	:= clock-sh7709.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7710)	:= clock-sh7710.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7720)	:= clock-sh7710.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7712)	:= clock-sh7712.o
 
 obj-y	+= $(clock-y)
diff --git a/arch/sh/kernel/cpu/sh3/clock-sh7712.c b/arch/sh/kernel/cpu/sh3/clock-sh7712.c
new file mode 100644
index 000000000000..54f54df51ef0
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/clock-sh7712.c
@@ -0,0 +1,71 @@
+/*
+ * arch/sh/kernel/cpu/sh3/clock-sh7712.c
+ *
+ * SH7712 support for the clock framework
+ *
+ *  Copyright (C) 2007  Andrew Murray <amurray@mpc-data.co.uk>
+ *
+ * Based on arch/sh/kernel/cpu/sh3/clock-sh3.c
+ *  Copyright (C) 2005  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+static int multipliers[] = { 1, 2, 3 };
+static int divisors[]    = { 1, 2, 3, 4, 6 };
+
+static void master_clk_init(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = (frqcr & 0x0300) >> 8;
+
+	clk->rate *= multipliers[idx];
+}
+
+static struct clk_ops sh7712_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = frqcr & 0x0007;
+
+	clk->rate = clk->parent->rate / divisors[idx];
+}
+
+static struct clk_ops sh7712_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = (frqcr & 0x0030) >> 4;
+
+	clk->rate = clk->parent->rate / divisors[idx];
+}
+
+static struct clk_ops sh7712_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7712_clk_ops[] = {
+	&sh7712_master_clk_ops,
+	&sh7712_module_clk_ops,
+	&sh7712_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7712_clk_ops))
+		*ops = sh7712_clk_ops[idx];
+}
+
diff --git a/include/asm-sh/cpu-sh3/freq.h b/include/asm-sh/cpu-sh3/freq.h
index 0a054b53b9de..53c62302b2e3 100644
--- a/include/asm-sh/cpu-sh3/freq.h
+++ b/include/asm-sh/cpu-sh3/freq.h
@@ -10,7 +10,12 @@
 #ifndef __ASM_CPU_SH3_FREQ_H
 #define __ASM_CPU_SH3_FREQ_H
 
+#ifdef CONFIG_CPU_SUBTYPE_SH7712
+#define FRQCR			0xA415FF80
+#else
 #define FRQCR			0xffffff80
+#endif
+
 #define MIN_DIVISOR_NR		0
 #define MAX_DIVISOR_NR		4
 
-- 
cgit v1.2.3


From e08d8aaeadac37c8d149b5d5bda6cf7a98d21e92 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 8 Jan 2008 18:05:03 +0900
Subject: sh: Fix posix_types.h userspace breakage from sh64 merge.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/Kbuild           |   5 ++
 include/asm-sh/posix_types.h    | 129 +++------------------------------------
 include/asm-sh/posix_types_32.h | 122 +++++++++++++++++++++++++++++++++++++
 include/asm-sh/posix_types_64.h | 131 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 265 insertions(+), 122 deletions(-)
 create mode 100644 include/asm-sh/posix_types_32.h
 create mode 100644 include/asm-sh/posix_types_64.h

(limited to 'include')

diff --git a/include/asm-sh/Kbuild b/include/asm-sh/Kbuild
index 76a8ccf254a5..43910cdf78a5 100644
--- a/include/asm-sh/Kbuild
+++ b/include/asm-sh/Kbuild
@@ -1,3 +1,8 @@
 include include/asm-generic/Kbuild.asm
 
 header-y += cpu-features.h
+
+unifdef-y += unistd_32.h
+unifdef-y += unistd_64.h
+unifdef-y += posix_types_32.h
+unifdef-y += posix_types_64.h
diff --git a/include/asm-sh/posix_types.h b/include/asm-sh/posix_types.h
index 0a3d2f54ab27..4b9d11c9fc77 100644
--- a/include/asm-sh/posix_types.h
+++ b/include/asm-sh/posix_types.h
@@ -1,122 +1,7 @@
-#ifndef __ASM_SH_POSIX_TYPES_H
-#define __ASM_SH_POSIX_TYPES_H
-
-/*
- * This file is generally used by user-level software, so you need to
- * be a little careful about namespace pollution etc.  Also, we cannot
- * assume GCC is being used.
- */
-
-typedef unsigned long	__kernel_ino_t;
-typedef unsigned short	__kernel_mode_t;
-typedef unsigned short	__kernel_nlink_t;
-typedef long		__kernel_off_t;
-typedef int		__kernel_pid_t;
-typedef unsigned short	__kernel_ipc_pid_t;
-typedef unsigned short	__kernel_uid_t;
-typedef unsigned short	__kernel_gid_t;
-typedef unsigned int	__kernel_size_t;
-typedef int		__kernel_ssize_t;
-typedef int		__kernel_ptrdiff_t;
-typedef long		__kernel_time_t;
-typedef long		__kernel_suseconds_t;
-typedef long		__kernel_clock_t;
-typedef int		__kernel_timer_t;
-typedef int		__kernel_clockid_t;
-typedef int		__kernel_daddr_t;
-typedef char *		__kernel_caddr_t;
-typedef unsigned short	__kernel_uid16_t;
-typedef unsigned short	__kernel_gid16_t;
-typedef unsigned int	__kernel_uid32_t;
-typedef unsigned int	__kernel_gid32_t;
-
-typedef unsigned short	__kernel_old_uid_t;
-typedef unsigned short	__kernel_old_gid_t;
-typedef unsigned short	__kernel_old_dev_t;
-
-#ifdef __GNUC__
-typedef long long	__kernel_loff_t;
-#endif
-
-typedef struct {
-#if defined(__KERNEL__) || defined(__USE_ALL)
-	int	val[2];
-#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
-	int	__val[2];
-#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
-} __kernel_fsid_t;
-
-#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2)
-
-#undef	__FD_SET
-static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
-{
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	__fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
-}
-
-#undef	__FD_CLR
-static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
-{
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	__fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
-}
-
-
-#undef	__FD_ISSET
-static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
-{ 
-	unsigned long __tmp = __fd / __NFDBITS;
-	unsigned long __rem = __fd % __NFDBITS;
-	return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
-}
-
-/*
- * This will unroll the loop for the normal constant case (8 ints,
- * for a 256-bit fd_set)
- */
-#undef	__FD_ZERO
-static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
-{
-	unsigned long *__tmp = __p->fds_bits;
-	int __i;
-
-	if (__builtin_constant_p(__FDSET_LONGS)) {
-		switch (__FDSET_LONGS) {
-		case 16:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			__tmp[ 4] = 0; __tmp[ 5] = 0;
-			__tmp[ 6] = 0; __tmp[ 7] = 0;
-			__tmp[ 8] = 0; __tmp[ 9] = 0;
-			__tmp[10] = 0; __tmp[11] = 0;
-			__tmp[12] = 0; __tmp[13] = 0;
-			__tmp[14] = 0; __tmp[15] = 0;
-			return;
-
-		case 8:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			__tmp[ 4] = 0; __tmp[ 5] = 0;
-			__tmp[ 6] = 0; __tmp[ 7] = 0;
-			return;
-
-		case 4:
-			__tmp[ 0] = 0; __tmp[ 1] = 0;
-			__tmp[ 2] = 0; __tmp[ 3] = 0;
-			return;
-		}
-	}
-	__i = __FDSET_LONGS;
-	while (__i) {
-		__i--;
-		*__tmp = 0;
-		__tmp++;
-	}
-}
-
-#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
-
-#endif /* __ASM_SH_POSIX_TYPES_H */
+#ifdef __KERNEL__
+# ifdef CONFIG_SUPERH32
+#  include "posix_types_32.h"
+# else
+#  include "posix_types_64.h"
+# endif
+#endif /* __KERNEL__ */
diff --git a/include/asm-sh/posix_types_32.h b/include/asm-sh/posix_types_32.h
new file mode 100644
index 000000000000..0a3d2f54ab27
--- /dev/null
+++ b/include/asm-sh/posix_types_32.h
@@ -0,0 +1,122 @@
+#ifndef __ASM_SH_POSIX_TYPES_H
+#define __ASM_SH_POSIX_TYPES_H
+
+/*
+ * This file is generally used by user-level software, so you need to
+ * be a little careful about namespace pollution etc.  Also, we cannot
+ * assume GCC is being used.
+ */
+
+typedef unsigned long	__kernel_ino_t;
+typedef unsigned short	__kernel_mode_t;
+typedef unsigned short	__kernel_nlink_t;
+typedef long		__kernel_off_t;
+typedef int		__kernel_pid_t;
+typedef unsigned short	__kernel_ipc_pid_t;
+typedef unsigned short	__kernel_uid_t;
+typedef unsigned short	__kernel_gid_t;
+typedef unsigned int	__kernel_size_t;
+typedef int		__kernel_ssize_t;
+typedef int		__kernel_ptrdiff_t;
+typedef long		__kernel_time_t;
+typedef long		__kernel_suseconds_t;
+typedef long		__kernel_clock_t;
+typedef int		__kernel_timer_t;
+typedef int		__kernel_clockid_t;
+typedef int		__kernel_daddr_t;
+typedef char *		__kernel_caddr_t;
+typedef unsigned short	__kernel_uid16_t;
+typedef unsigned short	__kernel_gid16_t;
+typedef unsigned int	__kernel_uid32_t;
+typedef unsigned int	__kernel_gid32_t;
+
+typedef unsigned short	__kernel_old_uid_t;
+typedef unsigned short	__kernel_old_gid_t;
+typedef unsigned short	__kernel_old_dev_t;
+
+#ifdef __GNUC__
+typedef long long	__kernel_loff_t;
+#endif
+
+typedef struct {
+#if defined(__KERNEL__) || defined(__USE_ALL)
+	int	val[2];
+#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+	int	__val[2];
+#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+} __kernel_fsid_t;
+
+#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2)
+
+#undef	__FD_SET
+static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	__fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
+}
+
+#undef	__FD_CLR
+static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	__fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
+}
+
+
+#undef	__FD_ISSET
+static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
+{ 
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
+}
+
+/*
+ * This will unroll the loop for the normal constant case (8 ints,
+ * for a 256-bit fd_set)
+ */
+#undef	__FD_ZERO
+static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
+{
+	unsigned long *__tmp = __p->fds_bits;
+	int __i;
+
+	if (__builtin_constant_p(__FDSET_LONGS)) {
+		switch (__FDSET_LONGS) {
+		case 16:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			__tmp[ 4] = 0; __tmp[ 5] = 0;
+			__tmp[ 6] = 0; __tmp[ 7] = 0;
+			__tmp[ 8] = 0; __tmp[ 9] = 0;
+			__tmp[10] = 0; __tmp[11] = 0;
+			__tmp[12] = 0; __tmp[13] = 0;
+			__tmp[14] = 0; __tmp[15] = 0;
+			return;
+
+		case 8:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			__tmp[ 4] = 0; __tmp[ 5] = 0;
+			__tmp[ 6] = 0; __tmp[ 7] = 0;
+			return;
+
+		case 4:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			return;
+		}
+	}
+	__i = __FDSET_LONGS;
+	while (__i) {
+		__i--;
+		*__tmp = 0;
+		__tmp++;
+	}
+}
+
+#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
+
+#endif /* __ASM_SH_POSIX_TYPES_H */
diff --git a/include/asm-sh/posix_types_64.h b/include/asm-sh/posix_types_64.h
new file mode 100644
index 000000000000..0620317a6f0f
--- /dev/null
+++ b/include/asm-sh/posix_types_64.h
@@ -0,0 +1,131 @@
+#ifndef __ASM_SH64_POSIX_TYPES_H
+#define __ASM_SH64_POSIX_TYPES_H
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * include/asm-sh64/posix_types.h
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003  Paul Mundt
+ *
+ * This file is generally used by user-level software, so you need to
+ * be a little careful about namespace pollution etc.  Also, we cannot
+ * assume GCC is being used.
+ */
+
+typedef unsigned long	__kernel_ino_t;
+typedef unsigned short	__kernel_mode_t;
+typedef unsigned short	__kernel_nlink_t;
+typedef long		__kernel_off_t;
+typedef int		__kernel_pid_t;
+typedef unsigned short	__kernel_ipc_pid_t;
+typedef unsigned short	__kernel_uid_t;
+typedef unsigned short	__kernel_gid_t;
+typedef long unsigned int	__kernel_size_t;
+typedef int		__kernel_ssize_t;
+typedef int		__kernel_ptrdiff_t;
+typedef long		__kernel_time_t;
+typedef long		__kernel_suseconds_t;
+typedef long		__kernel_clock_t;
+typedef int		__kernel_timer_t;
+typedef int		__kernel_clockid_t;
+typedef int		__kernel_daddr_t;
+typedef char *		__kernel_caddr_t;
+typedef unsigned short	__kernel_uid16_t;
+typedef unsigned short	__kernel_gid16_t;
+typedef unsigned int	__kernel_uid32_t;
+typedef unsigned int	__kernel_gid32_t;
+
+typedef unsigned short	__kernel_old_uid_t;
+typedef unsigned short	__kernel_old_gid_t;
+typedef unsigned short	__kernel_old_dev_t;
+
+#ifdef __GNUC__
+typedef long long	__kernel_loff_t;
+#endif
+
+typedef struct {
+#if defined(__KERNEL__) || defined(__USE_ALL)
+	int	val[2];
+#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+	int	__val[2];
+#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+} __kernel_fsid_t;
+
+#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2)
+
+#undef	__FD_SET
+static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	__fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
+}
+
+#undef	__FD_CLR
+static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	__fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
+}
+
+
+#undef	__FD_ISSET
+static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
+{
+	unsigned long __tmp = __fd / __NFDBITS;
+	unsigned long __rem = __fd % __NFDBITS;
+	return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
+}
+
+/*
+ * This will unroll the loop for the normal constant case (8 ints,
+ * for a 256-bit fd_set)
+ */
+#undef	__FD_ZERO
+static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
+{
+	unsigned long *__tmp = __p->fds_bits;
+	int __i;
+
+	if (__builtin_constant_p(__FDSET_LONGS)) {
+		switch (__FDSET_LONGS) {
+		case 16:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			__tmp[ 4] = 0; __tmp[ 5] = 0;
+			__tmp[ 6] = 0; __tmp[ 7] = 0;
+			__tmp[ 8] = 0; __tmp[ 9] = 0;
+			__tmp[10] = 0; __tmp[11] = 0;
+			__tmp[12] = 0; __tmp[13] = 0;
+			__tmp[14] = 0; __tmp[15] = 0;
+			return;
+
+		case 8:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			__tmp[ 4] = 0; __tmp[ 5] = 0;
+			__tmp[ 6] = 0; __tmp[ 7] = 0;
+			return;
+
+		case 4:
+			__tmp[ 0] = 0; __tmp[ 1] = 0;
+			__tmp[ 2] = 0; __tmp[ 3] = 0;
+			return;
+		}
+	}
+	__i = __FDSET_LONGS;
+	while (__i) {
+		__i--;
+		*__tmp = 0;
+		__tmp++;
+	}
+}
+
+#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
+
+#endif /* __ASM_SH64_POSIX_TYPES_H */
-- 
cgit v1.2.3


From 7f3edee81fbd49114c28057512906f169caa0bed Mon Sep 17 00:00:00 2001
From: Magnus Damm <damm@igel.co.jp>
Date: Thu, 10 Jan 2008 14:08:55 +0900
Subject: sh: intc - remove default interrupt priority tables

This patch removes interrupt priority tables from the intc code.
Optimal priority assignment varies with embedded application anyway,
so keeping the interrupt priority tables together with cpu-specific
code doesn't make sense.

The function intc_set_priority() should be used instead to set the
desired interrupt priority level.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/r7780rp/irq-r7780mp.c |  2 +-
 arch/sh/boards/renesas/r7780rp/irq-r7785rp.c |  2 +-
 arch/sh/boards/renesas/rts7751r2d/irq.c      |  4 ++--
 arch/sh/cchips/voyagergx/irq.c               |  2 +-
 arch/sh/kernel/cpu/irq/intc.c                | 31 ++++------------------------
 arch/sh/kernel/cpu/sh2/setup-sh7619.c        |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c       |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7206.c       |  2 +-
 arch/sh/kernel/cpu/sh3/setup-sh7705.c        | 10 ++-------
 arch/sh/kernel/cpu/sh3/setup-sh770x.c        | 11 ++--------
 arch/sh/kernel/cpu/sh3/setup-sh7710.c        | 16 ++------------
 arch/sh/kernel/cpu/sh3/setup-sh7720.c        | 14 ++-----------
 arch/sh/kernel/cpu/sh4/setup-sh7750.c        | 18 ++++++----------
 arch/sh/kernel/cpu/sh4/setup-sh7760.c        | 13 ++----------
 arch/sh/kernel/cpu/sh4a/setup-sh7722.c       | 10 +--------
 arch/sh/kernel/cpu/sh4a/setup-sh7780.c       | 13 ++++--------
 arch/sh/kernel/cpu/sh4a/setup-sh7785.c       | 19 +++++------------
 arch/sh/kernel/cpu/sh4a/setup-shx3.c         | 14 +++----------
 include/asm-sh/hw_irq.h                      | 12 +----------
 19 files changed, 42 insertions(+), 155 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/renesas/r7780rp/irq-r7780mp.c b/arch/sh/boards/renesas/r7780rp/irq-r7780mp.c
index 59b47fe061f9..1f8f073f27be 100644
--- a/arch/sh/boards/renesas/r7780rp/irq-r7780mp.c
+++ b/arch/sh/boards/renesas/r7780rp/irq-r7780mp.c
@@ -47,7 +47,7 @@ static unsigned char irl2irq[HL_NR_IRL] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "r7780mp", vectors,
-			 NULL, NULL, mask_registers, NULL, NULL);
+			 NULL, mask_registers, NULL, NULL);
 
 unsigned char * __init highlander_init_irq_r7780mp(void)
 {
diff --git a/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c b/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
index b2c6a84673bd..bbf18afc29a7 100644
--- a/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
+++ b/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
@@ -41,7 +41,7 @@ static unsigned char irl2irq[HL_NR_IRL] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "r7785rp", vectors,
-			 NULL, NULL, mask_registers, NULL, NULL);
+			 NULL, mask_registers, NULL, NULL);
 
 unsigned char * __init highlander_init_irq_r7785rp(void)
 {
diff --git a/arch/sh/boards/renesas/rts7751r2d/irq.c b/arch/sh/boards/renesas/rts7751r2d/irq.c
index 7cc2813adfe4..3fecd49e0d2a 100644
--- a/arch/sh/boards/renesas/rts7751r2d/irq.c
+++ b/arch/sh/boards/renesas/rts7751r2d/irq.c
@@ -71,7 +71,7 @@ static unsigned char irl2irq_r2d_1[R2D_NR_IRL] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_r2d_1, "r2d-1", vectors_r2d_1,
-			 NULL, NULL, mask_registers_r2d_1, NULL, NULL);
+			 NULL, mask_registers_r2d_1, NULL, NULL);
 
 #endif /* CONFIG_RTS7751R2D_1 */
 
@@ -109,7 +109,7 @@ static unsigned char irl2irq_r2d_plus[R2D_NR_IRL] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_r2d_plus, "r2d-plus", vectors_r2d_plus,
-			 NULL, NULL, mask_registers_r2d_plus, NULL, NULL);
+			 NULL, mask_registers_r2d_plus, NULL, NULL);
 
 #endif /* CONFIG_RTS7751R2D_PLUS */
 
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index ade303876841..e7e78c612fad 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -58,7 +58,7 @@ static struct intc_mask_reg mask_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "voyagergx", vectors,
-			 NULL, NULL, mask_registers, NULL, NULL);
+			 NULL, mask_registers, NULL, NULL);
 
 static unsigned int voyagergx_stat2irq[32] = {
 	IRQ_SM501_CI, IRQ_SM501_PV, IRQ_SM501_ZD, IRQ_SM501_2D,
diff --git a/arch/sh/kernel/cpu/irq/intc.c b/arch/sh/kernel/cpu/irq/intc.c
index 6ac018c15e03..84806b2027f8 100644
--- a/arch/sh/kernel/cpu/irq/intc.c
+++ b/arch/sh/kernel/cpu/irq/intc.c
@@ -335,31 +335,6 @@ static intc_enum __init intc_grp_id(struct intc_desc *desc,
 	return 0;
 }
 
-static unsigned int __init intc_prio_value(struct intc_desc *desc,
-					   intc_enum enum_id, int do_grps)
-{
-	struct intc_prio *p = desc->priorities;
-	unsigned int i;
-
-	for (i = 0; p && enum_id && i < desc->nr_priorities; i++) {
-		p = desc->priorities + i;
-
-		if (p->enum_id != enum_id)
-			continue;
-
-		return p->priority;
-	}
-
-	if (do_grps)
-		return intc_prio_value(desc, intc_grp_id(desc, enum_id), 0);
-
-	/* default to the lowest priority possible if no priority is set
-	 * - this needs to be at least 2 for 5-bit priorities on 7780
-	 */
-
-	return 2;
-}
-
 static unsigned int __init intc_mask_data(struct intc_desc *desc,
 					  struct intc_desc_int *d,
 					  intc_enum enum_id, int do_grps)
@@ -518,8 +493,10 @@ static void __init intc_register_irq(struct intc_desc *desc,
 				      handle_level_irq, "level");
 	set_irq_chip_data(irq, (void *)data[primary]);
 
-	/* record the desired priority level */
-	intc_prio_level[irq] = intc_prio_value(desc, enum_id, 1);
+	/* set priority level
+	 * - this needs to be at least 2 for 5-bit priorities on 7780
+	 */
+	intc_prio_level[irq] = 2;
 
 	/* enable secondary masking method if present */
 	if (data[!primary])
diff --git a/arch/sh/kernel/cpu/sh2/setup-sh7619.c b/arch/sh/kernel/cpu/sh2/setup-sh7619.c
index ec6adc3f306f..b230eb278cef 100644
--- a/arch/sh/kernel/cpu/sh2/setup-sh7619.c
+++ b/arch/sh/kernel/cpu/sh2/setup-sh7619.c
@@ -65,7 +65,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7619", vectors, groups,
-			 NULL, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 static struct plat_sci_port sci_platform_data[] = {
 	{
diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
index 3518f9c37d93..db6ef5cecde1 100644
--- a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
+++ b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
@@ -235,7 +235,7 @@ static struct intc_mask_reg mask_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7203", vectors, groups,
-			 NULL, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 
 static struct plat_sci_port sci_platform_data[] = {
 	{
diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7206.c b/arch/sh/kernel/cpu/sh2a/setup-sh7206.c
index bd745aa87222..a564425b905f 100644
--- a/arch/sh/kernel/cpu/sh2a/setup-sh7206.c
+++ b/arch/sh/kernel/cpu/sh2a/setup-sh7206.c
@@ -167,7 +167,7 @@ static struct intc_mask_reg mask_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7206", vectors, groups,
-			 NULL, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 
 static struct plat_sci_port sci_platform_data[] = {
 	{
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7705.c b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
index f6c65f2659e9..dd0a20a685f7 100644
--- a/arch/sh/kernel/cpu/sh3/setup-sh7705.c
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
@@ -66,12 +66,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(SCIF2, SCIF2_ERI, SCIF2_RXI, SCIF2_TXI),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(DMAC, 7),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(SCIF0, 3),
-};
-
 static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xfffffee2, 0, 16, 4, /* IPRA */ { TMU0, TMU1, TMU2, RTC } },
 	{ 0xfffffee4, 0, 16, 4, /* IPRB */ { WDT, REF_RCMI, 0, 0 } },
@@ -85,7 +79,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7705", vectors, groups,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 static struct intc_vect vectors_irq[] __initdata = {
 	INTC_VECT(IRQ0, 0x600), INTC_VECT(IRQ1, 0x620),
@@ -93,7 +87,7 @@ static struct intc_vect vectors_irq[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq, "sh7705-irq", vectors_irq, NULL,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 static struct plat_sci_port sci_platform_data[] = {
 	{
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh770x.c b/arch/sh/kernel/cpu/sh3/setup-sh770x.c
index 60b04b1f9453..969804bb523b 100644
--- a/arch/sh/kernel/cpu/sh3/setup-sh770x.c
+++ b/arch/sh/kernel/cpu/sh3/setup-sh770x.c
@@ -81,13 +81,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(SCIF2, SCIF2_ERI, SCIF2_RXI, SCIF2_BRI, SCIF2_TXI),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(DMAC, 7),
-	INTC_PRIO(SCI, 3),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(SCIF0, 3),
-};
-
 static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xfffffee2, 0, 16, 4, /* IPRA */ { TMU0, TMU1, TMU2, RTC } },
 	{ 0xfffffee4, 0, 16, 4, /* IPRB */ { WDT, REF, SCI, 0 } },
@@ -109,7 +102,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh770x", vectors, groups,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7707) || \
@@ -120,7 +113,7 @@ static struct intc_vect vectors_irq[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq, "sh770x-irq", vectors_irq, NULL,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 #endif
 
 static struct resource rtc_resources[] = {
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7710.c b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
index 84e5629fa841..0cc0e2bf135d 100644
--- a/arch/sh/kernel/cpu/sh3/setup-sh7710.c
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
@@ -73,18 +73,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(SIOF1, SIOF1_ERI, SIOF1_TXI, SIOF1_RXI, SIOF1_CCI),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(DMAC1, 7),
-	INTC_PRIO(DMAC2, 7),
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-	INTC_PRIO(SIOF0, 3),
-	INTC_PRIO(SIOF1, 3),
-	INTC_PRIO(EDMAC0, 5),
-	INTC_PRIO(EDMAC1, 5),
-	INTC_PRIO(EDMAC2, 5),
-};
-
 static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xfffffee2, 0, 16, 4, /* IPRA */ { TMU0, TMU1, TMU2, RTC } },
 	{ 0xfffffee4, 0, 16, 4, /* IPRB */ { WDT, REF, 0, 0 } },
@@ -101,7 +89,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7710", vectors, groups,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 static struct intc_vect vectors_irq[] __initdata = {
 	INTC_VECT(IRQ0, 0x600), INTC_VECT(IRQ1, 0x620),
@@ -109,7 +97,7 @@ static struct intc_vect vectors_irq[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq, "sh7710-irq", vectors_irq, NULL,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 static struct resource rtc_resources[] = {
 	[0] =	{
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7720.c b/arch/sh/kernel/cpu/sh3/setup-sh7720.c
index 00facd028a08..c00471a89212 100644
--- a/arch/sh/kernel/cpu/sh3/setup-sh7720.c
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7720.c
@@ -156,16 +156,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(MMC, MMCI0, MMCI1, MMCI2, MMCI3),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 2),
-	INTC_PRIO(SCIF1, 2),
-	INTC_PRIO(DMAC1, 1),
-	INTC_PRIO(DMAC2, 1),
-	INTC_PRIO(RTC, 2),
-	INTC_PRIO(TMU, 2),
-	INTC_PRIO(TPU, 2),
-};
-
 static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xA414FEE2UL, 0, 16, 4, /* IPRA */ { TMU0, TMU1, TMU2, RTC } },
 	{ 0xA414FEE4UL, 0, 16, 4, /* IPRB */ { WDT, REF_RCMI, SIM, 0 } },
@@ -184,7 +174,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7720", vectors, groups,
-		priorities, NULL, prio_registers, NULL);
+		NULL, prio_registers, NULL);
 
 static struct intc_sense_reg sense_registers[] __initdata = {
 	{ INTC_ICR1, 16, 2, { 0, 0, IRQ5, IRQ4, IRQ3, IRQ2, IRQ1, IRQ0 } },
@@ -197,7 +187,7 @@ static struct intc_vect vectors_irq[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_irq_desc, "sh7720-irq", vectors_irq,
-		NULL, priorities, NULL, prio_registers, sense_registers);
+		NULL, NULL, prio_registers, sense_registers);
 
 void __init plat_irq_setup_pins(int mode)
 {
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7750.c b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
index 523f68a9ce0e..ae3603aca615 100644
--- a/arch/sh/kernel/cpu/sh4/setup-sh7750.c
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
@@ -126,12 +126,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(REF, REF_RCMI, REF_ROVI),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF, 3),
-	INTC_PRIO(SCI1, 3),
-	INTC_PRIO(DMAC, 7),
-};
-
 static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xffd00004, 0, 16, 4, /* IPRA */ { TMU0, TMU1, TMU2, RTC } },
 	{ 0xffd00008, 0, 16, 4, /* IPRB */ { WDT, REF, SCI1, 0 } },
@@ -143,7 +137,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7750", vectors, groups,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 /* SH7750, SH7750S, SH7751 and SH7091 all have 4-channel DMA controllers */
 #if defined(CONFIG_CPU_SUBTYPE_SH7750) || \
@@ -163,7 +157,7 @@ static struct intc_group groups_dma4[] __initdata = {
 
 static DECLARE_INTC_DESC(intc_desc_dma4, "sh7750_dma4",
 			 vectors_dma4, groups_dma4,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 #endif
 
 /* SH7750R and SH7751R both have 8-channel DMA controllers */
@@ -184,7 +178,7 @@ static struct intc_group groups_dma8[] __initdata = {
 
 static DECLARE_INTC_DESC(intc_desc_dma8, "sh7750_dma8",
 			 vectors_dma8, groups_dma8,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 #endif
 
 /* SH7750R, SH7751 and SH7751R all have two extra timer channels */
@@ -205,7 +199,7 @@ static struct intc_mask_reg mask_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_tmu34, "sh7750_tmu34",
-			 vectors_tmu34, NULL, priorities,
+			 vectors_tmu34, NULL,
 			 mask_registers, prio_registers, NULL);
 #endif
 
@@ -216,7 +210,7 @@ static struct intc_vect vectors_irlm[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irlm, "sh7750_irlm", vectors_irlm, NULL,
-			 priorities, NULL, prio_registers, NULL);
+			 NULL, prio_registers, NULL);
 
 /* SH7751 and SH7751R both have PCI */
 #if defined(CONFIG_CPU_SUBTYPE_SH7751) || defined(CONFIG_CPU_SUBTYPE_SH7751R)
@@ -233,7 +227,7 @@ static struct intc_group groups_pci[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_pci, "sh7750_pci", vectors_pci, groups_pci,
-			 priorities, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7750) || \
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7760.c b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
index 7a898cb1d940..85f81579b97e 100644
--- a/arch/sh/kernel/cpu/sh4/setup-sh7760.c
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
@@ -92,15 +92,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(REF, REF_RCMI, REF_ROVI),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(SIM, 3),
-	INTC_PRIO(DMAC, 7),
-	INTC_PRIO(DMABRG, 13),
-};
-
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xfe080040, 0xfe080060, 32, /* INTMSK00 / INTMSKCLR00 */
 	  { IRQ4, IRQ5, IRQ6, IRQ7, 0, 0, HCAN20, HCAN21,
@@ -132,7 +123,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7760", vectors, groups,
-			 priorities, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 
 static struct intc_vect vectors_irq[] __initdata = {
 	INTC_VECT(IRL0, 0x240), INTC_VECT(IRL1, 0x2a0),
@@ -140,7 +131,7 @@ static struct intc_vect vectors_irq[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq, "sh7760-irq", vectors_irq, groups,
-			 priorities, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 
 static struct plat_sci_port sci_platform_data[] = {
 	{
diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c
index b9c6547c4a90..73c778d40d13 100644
--- a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c
+++ b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c
@@ -157,14 +157,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(SDHI, SDHI0, SDHI1, SDHI2, SDHI3),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(TMU0, 2),
-	INTC_PRIO(TMU1, 2),
-};
-
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xa4080080, 0xa40800c0, 8, /* IMR0 / IMCR0 */
 	  { } },
@@ -217,7 +209,7 @@ static struct intc_sense_reg sense_registers[] __initdata = {
 	  { IRQ0, IRQ1, IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, IRQ7 } },
 };
 
-static DECLARE_INTC_DESC(intc_desc, "sh7722", vectors, groups, priorities,
+static DECLARE_INTC_DESC(intc_desc, "sh7722", vectors, groups,
 			 mask_registers, prio_registers, sense_registers);
 
 void __init plat_irq_setup(void)
diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7780.c b/arch/sh/kernel/cpu/sh4a/setup-sh7780.c
index e8fd33ff0605..293004b526ff 100644
--- a/arch/sh/kernel/cpu/sh4a/setup-sh7780.c
+++ b/arch/sh/kernel/cpu/sh4a/setup-sh7780.c
@@ -168,11 +168,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(GPIO, GPIOI0, GPIOI1, GPIOI2, GPIOI3),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-};
-
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xffd40038, 0xffd4003c, 32, /* INT2MSKR / INT2MSKCR */
 	  { 0, 0, 0, 0, 0, 0, GPIO, FLCTL,
@@ -195,7 +190,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xffd4001c, 0, 32, 8, /* INT2PRI7 */ { FLCTL, GPIO } },
 };
 
-static DECLARE_INTC_DESC(intc_desc, "sh7780", vectors, groups, priorities,
+static DECLARE_INTC_DESC(intc_desc, "sh7780", vectors, groups,
 			 mask_registers, prio_registers, NULL);
 
 /* Support for external interrupt pins in IRQ mode */
@@ -223,7 +218,7 @@ static struct intc_sense_reg irq_sense_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_irq_desc, "sh7780-irq", irq_vectors,
-			 NULL, NULL, irq_mask_registers, irq_prio_registers,
+			 NULL, irq_mask_registers, irq_prio_registers,
 			 irq_sense_registers);
 
 /* External interrupt pins in IRL mode */
@@ -257,10 +252,10 @@ static struct intc_mask_reg irl7654_mask_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_irl7654_desc, "sh7780-irl7654", irl_vectors,
-			 NULL, NULL, irl7654_mask_registers, NULL, NULL);
+			 NULL, irl7654_mask_registers, NULL, NULL);
 
 static DECLARE_INTC_DESC(intc_irl3210_desc, "sh7780-irl3210", irl_vectors,
-			 NULL, NULL, irl3210_mask_registers, NULL, NULL);
+			 NULL, irl3210_mask_registers, NULL, NULL);
 
 #define INTC_ICR0	0xffd00000
 #define INTC_INTMSK0	0xffd00044
diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7785.c b/arch/sh/kernel/cpu/sh4a/setup-sh7785.c
index 39b215d6cee5..74b60e96cdf4 100644
--- a/arch/sh/kernel/cpu/sh4a/setup-sh7785.c
+++ b/arch/sh/kernel/cpu/sh4a/setup-sh7785.c
@@ -178,15 +178,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(GPIO, GPIOI0, GPIOI1, GPIOI2, GPIOI3),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(SCIF3, 3),
-	INTC_PRIO(SCIF4, 3),
-	INTC_PRIO(SCIF5, 3),
-};
-
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xffd00044, 0xffd00064, 32, /* INTMSK0 / INTMSKCLR0 */
 	  { IRQ0, IRQ1, IRQ2, IRQ3, IRQ4, IRQ5, IRQ6, IRQ7 } },
@@ -227,7 +218,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 	{ 0xffd40024, 0, 32, 8, /* INT2PRI9 */ { DU, GDTA, } },
 };
 
-static DECLARE_INTC_DESC(intc_desc, "sh7785", vectors, groups, priorities,
+static DECLARE_INTC_DESC(intc_desc, "sh7785", vectors, groups,
 			 mask_registers, prio_registers, NULL);
 
 /* Support for external interrupt pins in IRQ mode */
@@ -248,11 +239,11 @@ static struct intc_sense_reg sense_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq0123, "sh7785-irq0123", vectors_irq0123,
-			 NULL, NULL, mask_registers, prio_registers,
+			 NULL, mask_registers, prio_registers,
 			 sense_registers);
 
 static DECLARE_INTC_DESC(intc_desc_irq4567, "sh7785-irq4567", vectors_irq4567,
-			 NULL, NULL, mask_registers, prio_registers,
+			 NULL, mask_registers, prio_registers,
 			 sense_registers);
 
 /* External interrupt pins in IRL mode */
@@ -280,10 +271,10 @@ static struct intc_vect vectors_irl4567[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irl0123, "sh7785-irl0123", vectors_irl0123,
-			 NULL, NULL, mask_registers, NULL, NULL);
+			 NULL, mask_registers, NULL, NULL);
 
 static DECLARE_INTC_DESC(intc_desc_irl4567, "sh7785-irl4567", vectors_irl4567,
-			 NULL, NULL, mask_registers, NULL, NULL);
+			 NULL, mask_registers, NULL, NULL);
 
 #define INTC_ICR0	0xffd00000
 #define INTC_INTMSK0	0xffd00044
diff --git a/arch/sh/kernel/cpu/sh4a/setup-shx3.c b/arch/sh/kernel/cpu/sh4a/setup-shx3.c
index c6cdd7e3b049..4dc958b6b314 100644
--- a/arch/sh/kernel/cpu/sh4a/setup-shx3.c
+++ b/arch/sh/kernel/cpu/sh4a/setup-shx3.c
@@ -165,13 +165,6 @@ static struct intc_group groups[] __initdata = {
 	INTC_GROUP(DTU3, DTU3_TEND, DTU3_AE, DTU3_TMISS),
 };
 
-static struct intc_prio priorities[] __initdata = {
-	INTC_PRIO(SCIF0, 3),
-	INTC_PRIO(SCIF1, 3),
-	INTC_PRIO(SCIF2, 3),
-	INTC_PRIO(SCIF3, 3),
-};
-
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xfe410030, 0xfe410050, 32, /* CnINTMSK0 / CnINTMSKCLR0 */
 	  { IRQ0, IRQ1, IRQ2, IRQ3 } },
@@ -218,7 +211,7 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 	    INTICI3, INTICI2, INTICI1, INTICI0 }, INTC_SMP(4, 4) },
 };
 
-static DECLARE_INTC_DESC(intc_desc, "shx3", vectors, groups, priorities,
+static DECLARE_INTC_DESC(intc_desc, "shx3", vectors, groups,
 			 mask_registers, prio_registers, NULL);
 
 /* Support for external interrupt pins in IRQ mode */
@@ -232,8 +225,7 @@ static struct intc_sense_reg sense_registers[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irq, "shx3-irq", vectors_irq, groups,
-			 priorities, mask_registers, prio_registers,
-			 sense_registers);
+			 mask_registers, prio_registers, sense_registers);
 
 /* External interrupt pins in IRL mode */
 static struct intc_vect vectors_irl[] __initdata = {
@@ -248,7 +240,7 @@ static struct intc_vect vectors_irl[] __initdata = {
 };
 
 static DECLARE_INTC_DESC(intc_desc_irl, "shx3-irl", vectors_irl, groups,
-			 priorities, mask_registers, prio_registers, NULL);
+			 mask_registers, prio_registers, NULL);
 
 void __init plat_irq_setup_pins(int mode)
 {
diff --git a/include/asm-sh/hw_irq.h b/include/asm-sh/hw_irq.h
index cb0b6c9f7020..c958fdaa0095 100644
--- a/include/asm-sh/hw_irq.h
+++ b/include/asm-sh/hw_irq.h
@@ -33,13 +33,6 @@ struct intc_vect {
 #define INTC_VECT(enum_id, vect) { enum_id, vect }
 #define INTC_IRQ(enum_id, irq) INTC_VECT(enum_id, irq2evt(irq))
 
-struct intc_prio {
-	intc_enum enum_id;
-	unsigned char priority;
-};
-
-#define INTC_PRIO(enum_id, prio) { enum_id, prio }
-
 struct intc_group {
 	intc_enum enum_id;
 	intc_enum enum_ids[32];
@@ -79,8 +72,6 @@ struct intc_desc {
 	unsigned int nr_vectors;
 	struct intc_group *groups;
 	unsigned int nr_groups;
-	struct intc_prio *priorities;
-	unsigned int nr_priorities;
 	struct intc_mask_reg *mask_regs;
 	unsigned int nr_mask_regs;
 	struct intc_prio_reg *prio_regs;
@@ -92,10 +83,9 @@ struct intc_desc {
 
 #define _INTC_ARRAY(a) a, sizeof(a)/sizeof(*a)
 #define DECLARE_INTC_DESC(symbol, chipname, vectors, groups,		\
-	priorities, mask_regs, prio_regs, sense_regs)			\
+	mask_regs, prio_regs, sense_regs)				\
 struct intc_desc symbol __initdata = {					\
 	_INTC_ARRAY(vectors), _INTC_ARRAY(groups),			\
-	_INTC_ARRAY(priorities),					\
 	_INTC_ARRAY(mask_regs), _INTC_ARRAY(prio_regs),			\
 	_INTC_ARRAY(sense_regs),					\
 	chipname,							\
-- 
cgit v1.2.3


From 29ec6778a49d30c47345afdfaf1d03acff3cf656 Mon Sep 17 00:00:00 2001
From: Magnus Damm <damm@igel.co.jp>
Date: Tue, 15 Jan 2008 12:47:53 +0900
Subject: sh: remove voyagergx

This patch removes redundant irq handling code together with unused
consistent alloc code. R2D uart setup code is changed to use
sm501-regs.h and unused header files are removed.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                          |   1 -
 arch/sh/boards/renesas/rts7751r2d/irq.c   |   4 -
 arch/sh/boards/renesas/rts7751r2d/setup.c |  73 ++++---
 arch/sh/cchips/voyagergx/Makefile         |   9 -
 arch/sh/cchips/voyagergx/consistent.c     | 121 -----------
 arch/sh/cchips/voyagergx/irq.c            | 101 ---------
 arch/sh/cchips/voyagergx/setup.c          |  37 ----
 include/asm-sh/voyagergx.h                | 341 ------------------------------
 8 files changed, 39 insertions(+), 648 deletions(-)
 delete mode 100644 arch/sh/cchips/voyagergx/Makefile
 delete mode 100644 arch/sh/cchips/voyagergx/consistent.c
 delete mode 100644 arch/sh/cchips/voyagergx/irq.c
 delete mode 100644 arch/sh/cchips/voyagergx/setup.c
 delete mode 100644 include/asm-sh/voyagergx.h

(limited to 'include')

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 292d8618248d..e04b19293652 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -138,7 +138,6 @@ endif
 
 # Companion chips
 core-$(CONFIG_HD6446X_SERIES)	+= arch/sh/cchips/hd6446x/
-core-$(CONFIG_MFD_SM501)	+= arch/sh/cchips/voyagergx/
 
 cpuincdir-$(CONFIG_CPU_SH2)	:= cpu-sh2
 cpuincdir-$(CONFIG_CPU_SH2A)	:= cpu-sh2a
diff --git a/arch/sh/boards/renesas/rts7751r2d/irq.c b/arch/sh/boards/renesas/rts7751r2d/irq.c
index 3fecd49e0d2a..8e49f6e51247 100644
--- a/arch/sh/boards/renesas/rts7751r2d/irq.c
+++ b/arch/sh/boards/renesas/rts7751r2d/irq.c
@@ -13,7 +13,6 @@
 #include <linux/irq.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
-#include <asm/voyagergx.h>
 #include <asm/rts7751r2d.h>
 
 #define R2D_NR_IRL 13
@@ -153,7 +152,4 @@ void __init init_rts7751r2d_IRQ(void)
 	}
 
 	register_intc_controller(d);
-#ifdef CONFIG_MFD_SM501
-	setup_voyagergx_irq();
-#endif
 }
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index dc143c10cd18..8cb496617073 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -13,36 +13,13 @@
 #include <linux/pata_platform.h>
 #include <linux/serial_8250.h>
 #include <linux/sm501.h>
+#include <linux/sm501-regs.h>
 #include <linux/pm.h>
 #include <linux/fb.h>
 #include <asm/machvec.h>
 #include <asm/rts7751r2d.h>
-#include <asm/voyagergx.h>
 #include <asm/io.h>
 
-static void __init voyagergx_serial_init(void)
-{
-	unsigned long val;
-
-	/*
-	 * GPIO Control
-	 */
-	val = readl((void __iomem *)GPIO_MUX_HIGH);
-	val |= 0x00001fe0;
-	writel(val, (void __iomem *)GPIO_MUX_HIGH);
-
-	/*
-	 * Power Mode Gate
-	 */
-	val = readl((void __iomem *)POWER_MODE0_GATE);
-	val |= (POWER_MODE0_GATE_U0 | POWER_MODE0_GATE_U1);
-	writel(val, (void __iomem *)POWER_MODE0_GATE);
-
-	val = readl((void __iomem *)POWER_MODE1_GATE);
-	val |= (POWER_MODE1_GATE_U0 | POWER_MODE1_GATE_U1);
-	writel(val, (void __iomem *)POWER_MODE1_GATE);
-}
-
 static struct resource cf_ide_resources[] = {
 	[0] = {
 		.start	= PA_AREA5_IO + 0x1000,
@@ -94,11 +71,11 @@ static struct platform_device heartbeat_device = {
 #ifdef CONFIG_MFD_SM501
 static struct plat_serial8250_port uart_platform_data[] = {
 	{
-		.membase	= (void __iomem *)VOYAGER_UART_BASE,
-		.mapbase	= VOYAGER_UART_BASE,
+		.membase	= (void __iomem *)0xb3e30000,
+		.mapbase	= 0xb3e30000,
 		.iotype		= UPIO_MEM,
-		.irq		= IRQ_SM501_U0,
-		.flags		= UPF_BOOT_AUTOCONF | UPF_SKIP_TEST,
+		.irq		= IRQ_VOYAGER,
+		.flags		= UPF_BOOT_AUTOCONF | UPF_SHARE_IRQ,
 		.regshift	= 2,
 		.uartclk	= (9600 * 16),
 	},
@@ -125,7 +102,7 @@ static struct resource sm501_resources[] = {
 		.flags	= IORESOURCE_MEM,
 	},
 	[2]	= {
-		.start	= IRQ_SM501_CV,
+		.start	= IRQ_VOYAGER,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -167,7 +144,15 @@ static struct sm501_platdata_fb sm501_fb_pdata = {
 	.flags		= SM501_FBPD_SWAP_FB_ENDIAN,
 };
 
+static struct sm501_initdata sm501_initdata = {
+	.gpio_high	= {
+		.set	= 0x00001fe0,
+		.mask	= 0x0,
+	},
+};
+
 static struct sm501_platdata sm501_platform_data = {
+	.init		= &sm501_initdata,
 	.fb		= &sm501_fb_pdata,
 };
 
@@ -237,6 +222,7 @@ u8 rts7751r2d_readb(void __iomem *addr)
  */
 static void __init rts7751r2d_setup(char **cmdline_p)
 {
+	void __iomem *sm501_reg;
 	u16 ver = ctrl_inw(PA_VERREG);
 
 	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
@@ -247,7 +233,30 @@ static void __init rts7751r2d_setup(char **cmdline_p)
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
 
-	voyagergx_serial_init();
+	/* sm501 dram configuration:
+	 * ColSizeX = 11 - External Memory Column Size: 256 words.
+	 * APX = 1 - External Memory Active to Pre-Charge Delay: 7 clocks.
+	 * RstX = 1 - External Memory Reset: Normal.
+	 * Rfsh = 1 - Local Memory Refresh to Command Delay: 12 clocks.
+	 * BwC =  1 - Local Memory Block Write Cycle Time: 2 clocks.
+	 * BwP =  1 - Local Memory Block Write to Pre-Charge Delay: 1 clock.
+	 * AP = 1 - Internal Memory Active to Pre-Charge Delay: 7 clocks.
+	 * Rst = 1 - Internal Memory Reset: Normal.
+	 * RA = 1 - Internal Memory Remain in Active State: Do not remain.
+	 */
+
+	sm501_reg = (void __iomem *)0xb3e00000 + SM501_DRAM_CONTROL;
+	writel(readl(sm501_reg) | 0x00f107c0, sm501_reg);
+
+	/*
+	 * Power Mode Gate - Enable UART0
+	 */
+
+	sm501_reg = (void __iomem *)0xb3e00000 + SM501_POWER_MODE_0_GATE;
+	writel(readl(sm501_reg) | (1 << SM501_GATE_UART0), sm501_reg);
+
+	sm501_reg = (void __iomem *)0xb3e00000 + SM501_POWER_MODE_1_GATE;
+	writel(readl(sm501_reg) | (1 << SM501_GATE_UART0), sm501_reg);
 }
 
 /*
@@ -260,8 +269,4 @@ static struct sh_machine_vector mv_rts7751r2d __initmv = {
 	.mv_irq_demux		= rts7751r2d_irq_demux,
 	.mv_writeb		= rts7751r2d_writeb,
 	.mv_readb		= rts7751r2d_readb,
-#if defined(CONFIG_MFD_SM501) && defined(CONFIG_USB_OHCI_HCD)
-	.mv_consistent_alloc	= voyagergx_consistent_alloc,
-	.mv_consistent_free	= voyagergx_consistent_free,
-#endif
 };
diff --git a/arch/sh/cchips/voyagergx/Makefile b/arch/sh/cchips/voyagergx/Makefile
deleted file mode 100644
index f73963cb3744..000000000000
--- a/arch/sh/cchips/voyagergx/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-#
-# Makefile for VoyagerGX
-#
-
-obj-y	:= irq.o setup.o
-
-obj-$(CONFIG_USB_OHCI_HCD)	+= consistent.o
-
-EXTRA_CFLAGS += -Werror
diff --git a/arch/sh/cchips/voyagergx/consistent.c b/arch/sh/cchips/voyagergx/consistent.c
deleted file mode 100644
index 07e8b9c5a531..000000000000
--- a/arch/sh/cchips/voyagergx/consistent.c
+++ /dev/null
@@ -1,121 +0,0 @@
-/*
- * arch/sh/cchips/voyagergx/consistent.c
- *
- * Copyright (C) 2004  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#include <linux/mm.h>
-#include <linux/dma-mapping.h>
-#include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/types.h>
-#include <linux/module.h>
-#include <linux/device.h>
-#include <asm/io.h>
-
-
-struct voya_alloc_entry {
-	struct list_head list;
-	unsigned long ofs;
-	unsigned long len;
-};
-
-static DEFINE_SPINLOCK(voya_list_lock);
-static LIST_HEAD(voya_alloc_list);
-
-#define OHCI_SRAM_START	0xb0000000
-#define OHCI_HCCA_SIZE	0x100
-#define OHCI_SRAM_SIZE	0x10000
-
-#define VOYAGER_OHCI_NAME	"voyager-ohci"
-
-void *voyagergx_consistent_alloc(struct device *dev, size_t size,
-				 dma_addr_t *handle, gfp_t flag)
-{
-	struct list_head *list = &voya_alloc_list;
-	struct voya_alloc_entry *entry;
-	unsigned long start, end;
-	unsigned long flags;
-
-	/*
-	 * The SM501 contains an integrated 8051 with its own SRAM.
-	 * Devices within the cchip can all hook into the 8051 SRAM.
-	 * We presently use this for the OHCI.
-	 *
-	 * Everything else goes through consistent_alloc().
-	 */
-	if (!dev || strcmp(dev->driver->name, VOYAGER_OHCI_NAME))
-		return NULL;
-
-	start = OHCI_SRAM_START + OHCI_HCCA_SIZE;
-
-	entry = kmalloc(sizeof(struct voya_alloc_entry), GFP_ATOMIC);
-	if (!entry)
-		return ERR_PTR(-ENOMEM);
-
-	entry->len = (size + 15) & ~15;
-
-	/*
-	 * The basis for this allocator is dwmw2's malloc.. the
-	 * Matrox allocator :-)
-	 */
-	spin_lock_irqsave(&voya_list_lock, flags);
-	list_for_each(list, &voya_alloc_list) {
-		struct voya_alloc_entry *p;
-
-		p = list_entry(list, struct voya_alloc_entry, list);
-
-		if (p->ofs - start >= size)
-			goto out;
-
-		start = p->ofs + p->len;
-	}
-
-	end  = start + (OHCI_SRAM_SIZE  - OHCI_HCCA_SIZE);
-	list = &voya_alloc_list;
-
-	if (end - start >= size) {
-out:
-		entry->ofs = start;
-		list_add_tail(&entry->list, list);
-		spin_unlock_irqrestore(&voya_list_lock, flags);
-
-		*handle = start;
-		return (void *)start;
-	}
-
-	kfree(entry);
-	spin_unlock_irqrestore(&voya_list_lock, flags);
-
-	return ERR_PTR(-EINVAL);
-}
-
-int voyagergx_consistent_free(struct device *dev, size_t size,
-			      void *vaddr, dma_addr_t handle)
-{
-	struct voya_alloc_entry *entry;
-	unsigned long flags;
-
-	if (!dev || strcmp(dev->driver->name, VOYAGER_OHCI_NAME))
-		return -EINVAL;
-
-	spin_lock_irqsave(&voya_list_lock, flags);
-	list_for_each_entry(entry, &voya_alloc_list, list) {
-		if (entry->ofs != handle)
-			continue;
-
-		list_del(&entry->list);
-		kfree(entry);
-
-		break;
-	}
-	spin_unlock_irqrestore(&voya_list_lock, flags);
-
-	return 0;
-}
-
-EXPORT_SYMBOL(voyagergx_consistent_alloc);
-EXPORT_SYMBOL(voyagergx_consistent_free);
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
deleted file mode 100644
index e7e78c612fad..000000000000
--- a/arch/sh/cchips/voyagergx/irq.c
+++ /dev/null
@@ -1,101 +0,0 @@
-/* -------------------------------------------------------------------- */
-/* setup_voyagergx.c:                                                     */
-/* -------------------------------------------------------------------- */
-/*  This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-
-    Copyright 2003 (c) Lineo uSolutions,Inc.
-*/
-#include <linux/interrupt.h>
-#include <linux/init.h>
-#include <linux/io.h>
-#include <asm/voyagergx.h>
-#include <asm/rts7751r2d.h>
-
-enum {
-	UNUSED = 0,
-
-	/* voyager specific interrupt sources */
-	UP, G54, G53, G52, G51, G50, G49, G48,
-	I2C, PW, DMA, PCI, I2S, AC, US,
-	U1, U0, CV, MC, S1, S0,
-	UH, TWOD, ZD, PV, CI,
-};
-
-static struct intc_vect vectors[] __initdata = {
-	INTC_IRQ(UP, IRQ_SM501_UP), INTC_IRQ(G54, IRQ_SM501_G54),
-	INTC_IRQ(G53, IRQ_SM501_G53), INTC_IRQ(G52, IRQ_SM501_G52),
-	INTC_IRQ(G51, IRQ_SM501_G51), INTC_IRQ(G50, IRQ_SM501_G50),
-	INTC_IRQ(G49, IRQ_SM501_G49), INTC_IRQ(G48, IRQ_SM501_G48),
-	INTC_IRQ(I2C, IRQ_SM501_I2C), INTC_IRQ(PW, IRQ_SM501_PW),
-	INTC_IRQ(DMA, IRQ_SM501_DMA), INTC_IRQ(PCI, IRQ_SM501_PCI),
-	INTC_IRQ(I2S, IRQ_SM501_I2S), INTC_IRQ(AC, IRQ_SM501_AC),
-	INTC_IRQ(US, IRQ_SM501_US), INTC_IRQ(U1, IRQ_SM501_U1),
-	INTC_IRQ(U0, IRQ_SM501_U0), INTC_IRQ(CV, IRQ_SM501_CV),
-	INTC_IRQ(MC, IRQ_SM501_MC), INTC_IRQ(S1, IRQ_SM501_S1),
-	INTC_IRQ(S0, IRQ_SM501_S0), INTC_IRQ(UH, IRQ_SM501_UH),
-	INTC_IRQ(TWOD, IRQ_SM501_2D), INTC_IRQ(ZD, IRQ_SM501_ZD),
-	INTC_IRQ(PV, IRQ_SM501_PV), INTC_IRQ(CI, IRQ_SM501_CI),
-};
-
-static struct intc_mask_reg mask_registers[] __initdata = {
-	{ VOYAGER_INT_MASK, 0, 32, /* "Interrupt Mask", MMIO_base + 0x30 */
-	  { UP, G54, G53, G52, G51, G50, G49, G48,
-	    I2C, PW, 0, DMA, PCI, I2S, AC, US,
-	    0, 0, U1, U0, CV, MC, S1, S0,
-	    0, UH, 0, 0, TWOD, ZD, PV, CI } },
-};
-
-static DECLARE_INTC_DESC(intc_desc, "voyagergx", vectors,
-			 NULL, mask_registers, NULL, NULL);
-
-static unsigned int voyagergx_stat2irq[32] = {
-	IRQ_SM501_CI, IRQ_SM501_PV, IRQ_SM501_ZD, IRQ_SM501_2D,
-	0, 0, IRQ_SM501_UH, 0,
-	IRQ_SM501_S0, IRQ_SM501_S1, IRQ_SM501_MC, IRQ_SM501_CV,
-	IRQ_SM501_U0, IRQ_SM501_U1, 0, 0,
-	IRQ_SM501_US, IRQ_SM501_AC, IRQ_SM501_I2S, IRQ_SM501_PCI,
-	IRQ_SM501_DMA, 0, IRQ_SM501_PW, IRQ_SM501_I2C,
-	IRQ_SM501_G48, IRQ_SM501_G49, IRQ_SM501_G50, IRQ_SM501_G51,
-	IRQ_SM501_G52, IRQ_SM501_G53, IRQ_SM501_G54, IRQ_SM501_UP
-};
-
-static void voyagergx_irq_demux(unsigned int irq, struct irq_desc *desc)
-{
-	unsigned long intv = ctrl_inl(INT_STATUS);
-	struct irq_desc *ext_desc;
-	unsigned int ext_irq;
-	unsigned int k = 0;
-
-	while (intv) {
-		ext_irq = voyagergx_stat2irq[k];
-		if (ext_irq && (intv & 1)) {
-			ext_desc = irq_desc + ext_irq;
-			handle_level_irq(ext_irq, ext_desc);
-		}
-		intv >>= 1;
-		k++;
-	}
-}
-
-void __init setup_voyagergx_irq(void)
-{
-	printk(KERN_INFO "VoyagerGX on irq %d (mapped into %d to %d)\n",
-	       IRQ_VOYAGER,
-	       VOYAGER_IRQ_BASE,
-	       VOYAGER_IRQ_BASE + VOYAGER_IRQ_NUM - 1);
-
-	register_intc_controller(&intc_desc);
-	set_irq_chained_handler(IRQ_VOYAGER, voyagergx_irq_demux);
-}
diff --git a/arch/sh/cchips/voyagergx/setup.c b/arch/sh/cchips/voyagergx/setup.c
deleted file mode 100644
index 33f03027c193..000000000000
--- a/arch/sh/cchips/voyagergx/setup.c
+++ /dev/null
@@ -1,37 +0,0 @@
-/*
- * arch/sh/cchips/voyagergx/setup.c
- *
- * Setup routines for VoyagerGX cchip.
- *
- * Copyright (C) 2003 Lineo uSolutions, Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
- */
-#include <linux/init.h>
-#include <linux/module.h>
-#include <asm/io.h>
-#include <asm/voyagergx.h>
-
-static int __init setup_voyagergx(void)
-{
-	unsigned long val;
-
-	val = readl((void __iomem *)DRAM_CTRL);
-	val |= (DRAM_CTRL_CPU_COLUMN_SIZE_256	|
-		DRAM_CTRL_CPU_ACTIVE_PRECHARGE	|
-		DRAM_CTRL_CPU_RESET		|
-		DRAM_CTRL_REFRESH_COMMAND	|
-		DRAM_CTRL_BLOCK_WRITE_TIME	|
-		DRAM_CTRL_BLOCK_WRITE_PRECHARGE	|
-		DRAM_CTRL_ACTIVE_PRECHARGE	|
-		DRAM_CTRL_RESET			|
-		DRAM_CTRL_REMAIN_ACTIVE);
-	writel(val, (void __iomem *)DRAM_CTRL);
-
-	return 0;
-}
-
-module_init(setup_voyagergx);
diff --git a/include/asm-sh/voyagergx.h b/include/asm-sh/voyagergx.h
deleted file mode 100644
index 45b4547c74ac..000000000000
--- a/include/asm-sh/voyagergx.h
+++ /dev/null
@@ -1,341 +0,0 @@
-/* -------------------------------------------------------------------- */
-/* voyagergx.h	                                                      */
-/* -------------------------------------------------------------------- */
-/*  This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-
-    Copyright 2003 (c) Lineo uSolutions,Inc.
-*/
-/* -------------------------------------------------------------------- */
-
-#ifndef _VOYAGER_GX_REG_H
-#define _VOYAGER_GX_REG_H
-
-#define VOYAGER_BASE			0xb3e00000
-#define VOYAGER_USBH_BASE		(0x40000 + VOYAGER_BASE)
-#define VOYAGER_UART_BASE		(0x30000 + VOYAGER_BASE)
-#define	VOYAGER_AC97_BASE		(0xa0000 + VOYAGER_BASE)
-
-#define VOYAGER_IRQ_NUM			26
-#define VOYAGER_IRQ_BASE		200
-
-#define IRQ_SM501_UP			(VOYAGER_IRQ_BASE + 0)
-#define IRQ_SM501_G54			(VOYAGER_IRQ_BASE + 1)
-#define IRQ_SM501_G53			(VOYAGER_IRQ_BASE + 2)
-#define IRQ_SM501_G52			(VOYAGER_IRQ_BASE + 3)
-#define IRQ_SM501_G51			(VOYAGER_IRQ_BASE + 4)
-#define IRQ_SM501_G50			(VOYAGER_IRQ_BASE + 5)
-#define IRQ_SM501_G49			(VOYAGER_IRQ_BASE + 6)
-#define IRQ_SM501_G48			(VOYAGER_IRQ_BASE + 7)
-#define IRQ_SM501_I2C			(VOYAGER_IRQ_BASE + 8)
-#define IRQ_SM501_PW			(VOYAGER_IRQ_BASE + 9)
-#define IRQ_SM501_DMA			(VOYAGER_IRQ_BASE + 10)
-#define IRQ_SM501_PCI			(VOYAGER_IRQ_BASE + 11)
-#define IRQ_SM501_I2S			(VOYAGER_IRQ_BASE + 12)
-#define IRQ_SM501_AC			(VOYAGER_IRQ_BASE + 13)
-#define IRQ_SM501_US			(VOYAGER_IRQ_BASE + 14)
-#define IRQ_SM501_U1			(VOYAGER_IRQ_BASE + 15)
-#define IRQ_SM501_U0			(VOYAGER_IRQ_BASE + 16)
-#define IRQ_SM501_CV			(VOYAGER_IRQ_BASE + 17)
-#define IRQ_SM501_MC			(VOYAGER_IRQ_BASE + 18)
-#define IRQ_SM501_S1			(VOYAGER_IRQ_BASE + 19)
-#define IRQ_SM501_S0			(VOYAGER_IRQ_BASE + 20)
-#define IRQ_SM501_UH			(VOYAGER_IRQ_BASE + 21)
-#define IRQ_SM501_2D			(VOYAGER_IRQ_BASE + 22)
-#define IRQ_SM501_ZD			(VOYAGER_IRQ_BASE + 23)
-#define IRQ_SM501_PV			(VOYAGER_IRQ_BASE + 24)
-#define IRQ_SM501_CI			(VOYAGER_IRQ_BASE + 25)
-
-/* ----- MISC controle  register ------------------------------ */
-#define MISC_CTRL			(0x000004 + VOYAGER_BASE)
-#define MISC_CTRL_USBCLK_48		(3 << 28)
-#define MISC_CTRL_USBCLK_96		(2 << 28)
-#define MISC_CTRL_USBCLK_CRYSTAL	(1 << 28)
-
-/* ----- GPIO[31:0] register --------------------------------- */
-#define GPIO_MUX_LOW			(0x000008 + VOYAGER_BASE)
-#define GPIO_MUX_LOW_AC97		0x1F000000
-#define GPIO_MUX_LOW_8051		0x0000ffff
-#define GPIO_MUX_LOW_PWM		(1 << 29)
-
-/* ----- GPIO[63:32] register --------------------------------- */
-#define GPIO_MUX_HIGH			(0x00000C + VOYAGER_BASE)
-
-/* ----- DRAM controle  register ------------------------------- */
-#define DRAM_CTRL			(0x000010 + VOYAGER_BASE)
-#define DRAM_CTRL_EMBEDDED		(1 << 31)
-#define DRAM_CTRL_CPU_BURST_1		(0 << 28)
-#define DRAM_CTRL_CPU_BURST_2		(1 << 28)
-#define DRAM_CTRL_CPU_BURST_4		(2 << 28)
-#define DRAM_CTRL_CPU_BURST_8		(3 << 28)
-#define DRAM_CTRL_CPU_CAS_LATENCY	(1 << 27)
-#define DRAM_CTRL_CPU_SIZE_2		(0 << 24)
-#define DRAM_CTRL_CPU_SIZE_4		(1 << 24)
-#define DRAM_CTRL_CPU_SIZE_64		(4 << 24)
-#define DRAM_CTRL_CPU_SIZE_32		(5 << 24)
-#define DRAM_CTRL_CPU_SIZE_16		(6 << 24)
-#define DRAM_CTRL_CPU_SIZE_8		(7 << 24)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_1024	(0 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_512	(2 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_256	(3 << 22)
-#define DRAM_CTRL_CPU_ACTIVE_PRECHARGE	(1 << 21)
-#define DRAM_CTRL_CPU_RESET		(1 << 20)
-#define DRAM_CTRL_CPU_BANKS		(1 << 19)
-#define DRAM_CTRL_CPU_WRITE_PRECHARGE	(1 << 18)
-#define DRAM_CTRL_BLOCK_WRITE		(1 << 17)
-#define DRAM_CTRL_REFRESH_COMMAND	(1 << 16)
-#define DRAM_CTRL_SIZE_4		(0 << 13)
-#define DRAM_CTRL_SIZE_8		(1 << 13)
-#define DRAM_CTRL_SIZE_16		(2 << 13)
-#define DRAM_CTRL_SIZE_32		(3 << 13)
-#define DRAM_CTRL_SIZE_64		(4 << 13)
-#define DRAM_CTRL_SIZE_2		(5 << 13)
-#define DRAM_CTRL_COLUMN_SIZE_256	(0 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_512	(2 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_1024	(3 << 11)
-#define DRAM_CTRL_BLOCK_WRITE_TIME	(1 << 10)
-#define DRAM_CTRL_BLOCK_WRITE_PRECHARGE	(1 << 9)
-#define DRAM_CTRL_ACTIVE_PRECHARGE	(1 << 8)
-#define DRAM_CTRL_RESET			(1 << 7)
-#define DRAM_CTRL_REMAIN_ACTIVE		(1 << 6)
-#define DRAM_CTRL_BANKS			(1 << 1)
-#define DRAM_CTRL_WRITE_PRECHARGE	(1 << 0)
-
-/* ----- Arvitration control register -------------------------- */
-#define ARBITRATION_CTRL		(0x000014 + VOYAGER_BASE)
-#define ARBITRATION_CTRL_CPUMEM		(1 << 29)
-#define ARBITRATION_CTRL_INTMEM		(1 << 28)
-#define ARBITRATION_CTRL_USB_OFF	(0 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_1	(1 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_2	(2 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_3	(3 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_4	(4 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_5	(5 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_6	(6 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_7	(7 << 24)
-#define ARBITRATION_CTRL_PANEL_OFF	(0 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_1	(1 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_2	(2 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_3	(3 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_4	(4 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_5	(5 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_6	(6 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_7	(7 << 20)
-#define ARBITRATION_CTRL_ZVPORT_OFF	(0 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_1	(1 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_2	(2 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_3	(3 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_4	(4 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_5	(5 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_6	(6 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_7	(7 << 16)
-#define ARBITRATION_CTRL_CMD_INTPR_OFF	(0 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_1	(1 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_2	(2 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_3	(3 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_4	(4 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_5	(5 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_6	(6 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_7	(7 << 12)
-#define ARBITRATION_CTRL_DMA_OFF	(0 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_1	(1 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_2	(2 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_3	(3 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_4	(4 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_5	(5 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_6	(6 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_7	(7 << 8)
-#define ARBITRATION_CTRL_VIDEO_OFF	(0 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_1	(1 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_2	(2 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_3	(3 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_4	(4 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_5	(5 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_6	(6 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_7	(7 << 4)
-#define ARBITRATION_CTRL_CRT_OFF	(0 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_1	(1 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_2	(2 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_3	(3 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_4	(4 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_5	(5 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_6	(6 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_7	(7 << 0)
-
-/* ----- Command list status register -------------------------- */
-#define CMD_INTPR_STATUS		(0x000024 + VOYAGER_BASE)
-
-/* ----- Interrupt status register ----------------------------- */
-#define INT_STATUS			(0x00002c + VOYAGER_BASE)
-#define INT_STATUS_UH			(1 << 6)
-#define INT_STATUS_MC			(1 << 10)
-#define INT_STATUS_U0			(1 << 12)
-#define INT_STATUS_U1			(1 << 13)
-#define	INT_STATUS_AC			(1 << 17)
-
-/* ----- Interrupt mask register ------------------------------ */
-#define VOYAGER_INT_MASK		(0x000030 + VOYAGER_BASE)
-#define VOYAGER_INT_MASK_AC		(1 << 17)
-
-/* ----- Current Gate register ---------------------------------*/
-#define CURRENT_GATE			(0x000038 + VOYAGER_BASE)
-
-/* ----- Power mode 0 gate register --------------------------- */
-#define POWER_MODE0_GATE		(0x000040 + VOYAGER_BASE)
-#define POWER_MODE0_GATE_G		(1 << 6)
-#define POWER_MODE0_GATE_U0		(1 << 7)
-#define POWER_MODE0_GATE_U1		(1 << 8)
-#define POWER_MODE0_GATE_UH		(1 << 11)
-#define	POWER_MODE0_GATE_AC		(1 << 18)
-
-/* ----- Power mode 1 gate register --------------------------- */
-#define POWER_MODE1_GATE		(0x000048 + VOYAGER_BASE)
-#define POWER_MODE1_GATE_G		(1 << 6)
-#define POWER_MODE1_GATE_U0		(1 << 7)
-#define POWER_MODE1_GATE_U1		(1 << 8)
-#define POWER_MODE1_GATE_UH		(1 << 11)
-#define	POWER_MODE1_GATE_AC		(1 << 18)
-
-/* ----- Power mode 0 clock register -------------------------- */
-#define POWER_MODE0_CLOCK		(0x000044 + VOYAGER_BASE)
-
-/* ----- Power mode 1 clock register -------------------------- */
-#define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
-
-/* ----- Power mode control register ------------------------- */
-#define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
-
-/* ----- Miscellaneous Timing register ------------------------ */
-#define SYSTEM_DRAM_CTRL		(0x000068 + VOYAGER_BASE)
-
-/* ----- PWM register ------------------------------------------*/
-#define PWM_0				(0x010020 + VOYAGER_BASE)
-#define PWM_0_HC(x)			(((x)&0x0fff)<<20)
-#define PWM_0_LC(x)			(((x)&0x0fff)<<8 )
-#define PWM_0_CLK_DEV(x)		(((x)&0x000f)<<4 )
-#define PWM_0_EN			(1<<0)
-
-/* ----- I2C register ----------------------------------------- */
-#define I2C_BYTECOUNT			(0x010040 + VOYAGER_BASE)
-#define I2C_CONTROL			(0x010041 + VOYAGER_BASE)
-#define I2C_STATUS			(0x010042 + VOYAGER_BASE)
-#define I2C_RESET			(0x010042 + VOYAGER_BASE)
-#define I2C_SADDRESS			(0x010043 + VOYAGER_BASE)
-#define I2C_DATA			(0x010044 + VOYAGER_BASE)
-
-/* ----- Controle register bits ----------------------------------------- */
-#define I2C_CONTROL_E			(1 << 0)
-#define I2C_CONTROL_MODE		(1 << 1)
-#define I2C_CONTROL_STATUS		(1 << 2)
-#define I2C_CONTROL_INT			(1 << 4)
-#define I2C_CONTROL_INTACK		(1 << 5)
-#define I2C_CONTROL_REPEAT		(1 << 6)
-
-/* ----- Status register bits ----------------------------------------- */
-#define I2C_STATUS_BUSY			(1 << 0)
-#define I2C_STATUS_ACK			(1 << 1)
-#define I2C_STATUS_ERROR		(1 << 2)
-#define I2C_STATUS_COMPLETE		(1 << 3)
-
-/* ----- Reset register  ---------------------------------------------- */
-#define I2C_RESET_ERROR			(1 << 2)
-
-/* ----- transmission frequencies ------------------------------------- */
-#define I2C_SADDRESS_SELECT		(1 << 0)
-
-/* ----- Display Controll register ----------------------------------------- */
-#define PANEL_DISPLAY_CTRL		(0x080000 + VOYAGER_BASE)
-#define PANEL_DISPLAY_CTRL_BIAS         (1<<26)
-#define PANEL_PAN_CTRL			(0x080004 + VOYAGER_BASE)
-#define PANEL_COLOR_KEY			(0x080008 + VOYAGER_BASE)
-#define PANEL_FB_ADDRESS		(0x08000C + VOYAGER_BASE)
-#define PANEL_FB_WIDTH			(0x080010 + VOYAGER_BASE)
-#define PANEL_WINDOW_WIDTH		(0x080014 + VOYAGER_BASE)
-#define PANEL_WINDOW_HEIGHT		(0x080018 + VOYAGER_BASE)
-#define PANEL_PLANE_TL			(0x08001C + VOYAGER_BASE)
-#define PANEL_PLANE_BR			(0x080020 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_TOTAL		(0x080024 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_SYNC		(0x080028 + VOYAGER_BASE)
-#define PANEL_VERTICAL_TOTAL		(0x08002C + VOYAGER_BASE)
-#define PANEL_VERTICAL_SYNC		(0x080030 + VOYAGER_BASE)
-#define PANEL_CURRENT_LINE		(0x080034 + VOYAGER_BASE)
-#define VIDEO_DISPLAY_CTRL		(0x080040 + VOYAGER_BASE)
-#define VIDEO_FB_0_ADDRESS		(0x080044 + VOYAGER_BASE)
-#define VIDEO_FB_WIDTH			(0x080048 + VOYAGER_BASE)
-#define VIDEO_FB_0_LAST_ADDRESS		(0x08004C + VOYAGER_BASE)
-#define VIDEO_PLANE_TL			(0x080050 + VOYAGER_BASE)
-#define VIDEO_PLANE_BR			(0x080054 + VOYAGER_BASE)
-#define VIDEO_SCALE			(0x080058 + VOYAGER_BASE)
-#define VIDEO_INITIAL_SCALE		(0x08005C + VOYAGER_BASE)
-#define VIDEO_YUV_CONSTANTS		(0x080060 + VOYAGER_BASE)
-#define VIDEO_FB_1_ADDRESS		(0x080064 + VOYAGER_BASE)
-#define VIDEO_FB_1_LAST_ADDRESS		(0x080068 + VOYAGER_BASE)
-#define VIDEO_ALPHA_DISPLAY_CTRL	(0x080080 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_ADDRESS		(0x080084 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_WIDTH		(0x080088 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_LAST_ADDRESS	(0x08008C + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_TL		(0x080090 + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_BR		(0x080094 + VOYAGER_BASE)
-#define VIDEO_ALPHA_SCALE		(0x080098 + VOYAGER_BASE)
-#define VIDEO_ALPHA_INITIAL_SCALE	(0x08009C + VOYAGER_BASE)
-#define VIDEO_ALPHA_CHROMA_KEY		(0x0800A0 + VOYAGER_BASE)
-#define PANEL_HWC_ADDRESS		(0x0800F0 + VOYAGER_BASE)
-#define PANEL_HWC_LOCATION		(0x0800F4 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_12		(0x0800F8 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_3		(0x0800FC + VOYAGER_BASE)
-#define ALPHA_DISPLAY_CTRL		(0x080100 + VOYAGER_BASE)
-#define ALPHA_FB_ADDRESS		(0x080104 + VOYAGER_BASE)
-#define ALPHA_FB_WIDTH			(0x080108 + VOYAGER_BASE)
-#define ALPHA_PLANE_TL			(0x08010C + VOYAGER_BASE)
-#define ALPHA_PLANE_BR			(0x080110 + VOYAGER_BASE)
-#define ALPHA_CHROMA_KEY		(0x080114 + VOYAGER_BASE)
-#define CRT_DISPLAY_CTRL		(0x080200 + VOYAGER_BASE)
-#define CRT_FB_ADDRESS			(0x080204 + VOYAGER_BASE)
-#define CRT_FB_WIDTH			(0x080208 + VOYAGER_BASE)
-#define CRT_HORIZONTAL_TOTAL		(0x08020C + VOYAGER_BASE)
-#define CRT_HORIZONTAL_SYNC		(0x080210 + VOYAGER_BASE)
-#define CRT_VERTICAL_TOTAL		(0x080214 + VOYAGER_BASE)
-#define CRT_VERTICAL_SYNC		(0x080218 + VOYAGER_BASE)
-#define CRT_SIGNATURE_ANALYZER		(0x08021C + VOYAGER_BASE)
-#define CRT_CURRENT_LINE		(0x080220 + VOYAGER_BASE)
-#define CRT_MONITOR_DETECT		(0x080224 + VOYAGER_BASE)
-#define CRT_HWC_ADDRESS			(0x080230 + VOYAGER_BASE)
-#define CRT_HWC_LOCATION		(0x080234 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_12		(0x080238 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_3			(0x08023C + VOYAGER_BASE)
-#define CRT_PALETTE_RAM			(0x080400 + VOYAGER_BASE)
-#define PANEL_PALETTE_RAM		(0x080800 + VOYAGER_BASE)
-#define VIDEO_PALETTE_RAM		(0x080C00 + VOYAGER_BASE)
-
-/* ----- 8051 Controle register ----------------------------------------- */
-#define VOYAGER_8051_BASE		(0x000c0000 + VOYAGER_BASE)
-#define VOYAGER_8051_RESET		(0x000b0000 + VOYAGER_BASE)
-#define VOYAGER_8051_SELECT		(0x000b0004 + VOYAGER_BASE)
-#define VOYAGER_8051_CPU_INT		(0x000b000c + VOYAGER_BASE)
-
-/* ----- AC97 Controle register ----------------------------------------- */
-#define AC97_TX_SLOT0			(0x00000000 + VOYAGER_AC97_BASE)
-#define AC97_CONTROL_STATUS		(0x00000080 + VOYAGER_AC97_BASE)
-#define AC97C_READ			(1 << 19)
-#define AC97C_WD_BIT			(1 << 2)
-#define AC97C_INDEX_MASK		0x7f
-
-/* arch/sh/cchips/voyagergx/consistent.c */
-void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
-int voyagergx_consistent_free(struct device *, size_t, void *, dma_addr_t);
-
-/* arch/sh/cchips/voyagergx/irq.c */
-void setup_voyagergx_irq(void);
-
-#endif /* _VOYAGER_GX_REG_H */
-- 
cgit v1.2.3


From 03bbc0e6ba23700aea7fec801ac7e6c5a80f78f9 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 15 Jan 2008 16:55:55 +0900
Subject: sh: r7785rp: Hook up the rest of the HL7785 FPGA IRQ vectors.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/r7780rp/irq-r7785rp.c | 45 ++++++++++++++++++++--------
 include/asm-sh/r7780rp.h                     | 14 +++++++--
 2 files changed, 45 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c b/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
index bbf18afc29a7..af5ec74b2b1e 100644
--- a/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
+++ b/arch/sh/boards/renesas/r7780rp/irq-r7785rp.c
@@ -2,7 +2,7 @@
  * Renesas Solutions Highlander R7785RP Support.
  *
  * Copyright (C) 2002  Atom Create Engineering Co., Ltd.
- * Copyright (C) 2006  Paul Mundt
+ * Copyright (C) 2006 - 2008  Paul Mundt
  * Copyright (C) 2007  Magnus Damm
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -17,31 +17,52 @@
 enum {
 	UNUSED = 0,
 
-	/* board specific interrupt sources */
-	AX88796,          /* Ethernet controller */
-	CF,               /* Compact Flash */
+	/* FPGA specific interrupt sources */
+	CF,		/* Compact Flash */
+	SMBUS,		/* SMBUS */
+	TP,		/* Touch panel */
+	RTC,		/* RTC Alarm */
+	TH_ALERT,	/* Temperature sensor */
+	AX88796,	/* Ethernet controller */
+
+	/* external bus connector */
+	EXT0, EXT1, EXT2, EXT3, EXT4, EXT5, EXT6, EXT7,
 };
 
 static struct intc_vect vectors[] __initdata = {
 	INTC_IRQ(CF, IRQ_CF),
+	INTC_IRQ(SMBUS, IRQ_SMBUS),
+	INTC_IRQ(TP, IRQ_TP),
+	INTC_IRQ(RTC, IRQ_RTC),
+	INTC_IRQ(TH_ALERT, IRQ_TH_ALERT),
+
+	INTC_IRQ(EXT0, IRQ_EXT0), INTC_IRQ(EXT1, IRQ_EXT1),
+	INTC_IRQ(EXT2, IRQ_EXT2), INTC_IRQ(EXT3, IRQ_EXT3),
+
+	INTC_IRQ(EXT4, IRQ_EXT4), INTC_IRQ(EXT5, IRQ_EXT5),
+	INTC_IRQ(EXT6, IRQ_EXT6), INTC_IRQ(EXT7, IRQ_EXT7),
+
 	INTC_IRQ(AX88796, IRQ_AX88796),
 };
 
 static struct intc_mask_reg mask_registers[] __initdata = {
 	{ 0xa4000010, 0, 16, /* IRLMCR1 */
-	  { 0, 0, 0, 0, CF, AX88796, 0, 0,
-	    0, 0, 0, 0, 0, 0, 0, 0 } },
+	  { 0, 0, 0, 0, CF, AX88796, SMBUS, TP,
+	    RTC, 0, TH_ALERT, 0, 0, 0, 0, 0 } },
+	{ 0xa4000012, 0, 16, /* IRLMCR2 */
+	  { 0, 0, 0, 0, 0, 0, 0, 0,
+	    EXT7, EXT6, EXT5, EXT4, EXT3, EXT2, EXT1, EXT0 } },
 };
 
 static unsigned char irl2irq[HL_NR_IRL] __initdata = {
-	0, IRQ_CF, 0, 0,
-	0, 0, 0, 0,
-	0, 0, IRQ_AX88796, 0,
-	0, 0, 0,
+	0, IRQ_CF, IRQ_EXT4, IRQ_EXT5,
+	IRQ_EXT6, IRQ_EXT7, IRQ_SMBUS, IRQ_TP,
+	IRQ_RTC, IRQ_TH_ALERT, IRQ_AX88796, IRQ_EXT0,
+	IRQ_EXT1, IRQ_EXT2, IRQ_EXT3,
 };
 
 static DECLARE_INTC_DESC(intc_desc, "r7785rp", vectors,
-			 NULL, mask_registers, NULL, NULL);
+			 NULL, NULL, mask_registers, NULL, NULL);
 
 unsigned char * __init highlander_init_irq_r7785rp(void)
 {
@@ -58,7 +79,7 @@ unsigned char * __init highlander_init_irq_r7785rp(void)
 	ctrl_outw(0x7060, PA_IRLPRC);	/* FPGA IRLC */
 	ctrl_outw(0x0000, PA_IRLPRD);	/* FPGA IRLD */
 	ctrl_outw(0x4321, PA_IRLPRE);	/* FPGA IRLE */
-	ctrl_outw(0x0000, PA_IRLPRF);	/* FPGA IRLF */
+	ctrl_outw(0xdcba, PA_IRLPRF);	/* FPGA IRLF */
 
 	register_intc_controller(&intc_desc);
 	return irl2irq;
diff --git a/include/asm-sh/r7780rp.h b/include/asm-sh/r7780rp.h
index de37f933aa42..1c2d4d18fab2 100644
--- a/include/asm-sh/r7780rp.h
+++ b/include/asm-sh/r7780rp.h
@@ -195,8 +195,18 @@
 #ifndef IRQ_PSW
 #define IRQ_PSW			(HL_FPGA_IRQ_BASE + 2)
 #endif
-#define IRQ_EXT1		(HL_FPGA_IRQ_BASE + 3)
-#define IRQ_EXT4		(HL_FPGA_IRQ_BASE + 4)
+#define IRQ_EXT0		(HL_FPGA_IRQ_BASE + 3)
+#define IRQ_EXT1		(HL_FPGA_IRQ_BASE + 4)
+#define IRQ_EXT2		(HL_FPGA_IRQ_BASE + 5)
+#define IRQ_EXT3		(HL_FPGA_IRQ_BASE + 6)
+#define IRQ_EXT4		(HL_FPGA_IRQ_BASE + 7)
+#define IRQ_EXT5		(HL_FPGA_IRQ_BASE + 8)
+#define IRQ_EXT6		(HL_FPGA_IRQ_BASE + 9)
+#define IRQ_EXT7		(HL_FPGA_IRQ_BASE + 10)
+#define IRQ_SMBUS		(HL_FPGA_IRQ_BASE + 11)
+#define IRQ_TP			(HL_FPGA_IRQ_BASE + 12)
+#define IRQ_RTC			(HL_FPGA_IRQ_BASE + 13)
+#define IRQ_TH_ALERT		(HL_FPGA_IRQ_BASE + 14)
 
 void make_r7780rp_irq(unsigned int irq);
 
-- 
cgit v1.2.3


From a5350a9686efa65cbd2ad4677bcb9372c7ad05c7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Sat, 19 Jan 2008 16:06:01 +0900
Subject: sh: Kill off dead HS771RVoIP board support.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                             |   7 -
 arch/sh/Makefile                            |   1 -
 arch/sh/boards/renesas/hs7751rvoip/Kconfig  |  12 -
 arch/sh/boards/renesas/hs7751rvoip/Makefile |   8 -
 arch/sh/boards/renesas/hs7751rvoip/io.c     | 283 ---------
 arch/sh/boards/renesas/hs7751rvoip/irq.c    | 116 ----
 arch/sh/boards/renesas/hs7751rvoip/pci.c    | 149 -----
 arch/sh/boards/renesas/hs7751rvoip/setup.c  | 105 ----
 arch/sh/configs/hs7751rvoip_defconfig       | 908 ----------------------------
 arch/sh/drivers/pci/pci-auto.c              |   2 -
 arch/sh/tools/mach-types                    |   1 -
 include/asm-sh/hs7751rvoip.h                |  54 --
 12 files changed, 1646 deletions(-)
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/Kconfig
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/Makefile
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/io.c
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/irq.c
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/pci.c
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/setup.c
 delete mode 100644 arch/sh/configs/hs7751rvoip_defconfig
 delete mode 100644 include/asm-sh/hs7751rvoip.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index ddbd610d27e7..85bd9ac6f5e0 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -440,13 +440,6 @@ config SH_SECUREEDGE5410
 	  This includes both the OEM SecureEdge products as well as the
 	  SME product line.
 
-config SH_HS7751RVOIP
-	bool "HS7751RVOIP"
-	depends on CPU_SUBTYPE_SH7751R
-	help
-	  Select HS7751RVOIP if configuring for a Renesas Technology
-	  Sales VoIP board.
-
 config SH_7710VOIPGW
 	bool "SH7710-VOIP-GW"
 	depends on CPU_SUBTYPE_SH7710
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index e04b19293652..21cc649e1c12 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -112,7 +112,6 @@ machdir-$(CONFIG_SH_DREAMCAST)			+= dreamcast
 machdir-$(CONFIG_SH_MPC1211)			+= mpc1211
 machdir-$(CONFIG_SH_SH03)			+= sh03
 machdir-$(CONFIG_SH_SECUREEDGE5410)		+= snapgear
-machdir-$(CONFIG_SH_HS7751RVOIP)		+= renesas/hs7751rvoip
 machdir-$(CONFIG_SH_RTS7751R2D)			+= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		+= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			+= renesas/edosk7705
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Kconfig b/arch/sh/boards/renesas/hs7751rvoip/Kconfig
deleted file mode 100644
index 1743be477be5..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/Kconfig
+++ /dev/null
@@ -1,12 +0,0 @@
-if SH_HS7751RVOIP
-
-menu "HS7751RVoIP options"
-
-config HS7751RVOIP_CODEC
-	bool "Support VoIP Codec section"
-	help
-	  Selecting this option will support CODEC section.
-
-endmenu
-
-endif
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
deleted file mode 100644
index e626377c55ee..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ /dev/null
@@ -1,8 +0,0 @@
-#
-# Makefile for the HS7751RVoIP specific parts of the kernel
-#
-
-obj-y	 := setup.o io.o irq.o
-
-obj-$(CONFIG_PCI) += pci.o
-
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
deleted file mode 100644
index bb9aa0d62852..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ /dev/null
@@ -1,283 +0,0 @@
-/*
- * linux/arch/sh/boards/renesas/hs7751rvoip/io.c
- *
- * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
- * Based largely on io_se.c.
- *
- * I/O routine for Renesas Technology sales HS7751RVoIP
- *
- * Initial version only to support LAN access; some
- * placeholder code from io_hs7751rvoip.c left in with the
- * expectation of later SuperIO and PCMCIA access.
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/module.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-#include <asm/hs7751rvoip.h>
-#include <asm/addrspace.h>
-
-extern void *area6_io8_base;	/* Area 6 8bit I/O Base address */
-extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
-
-/*
- * The 7751R HS7751RVoIP uses the built-in PCI controller (PCIC)
- * of the 7751R processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */
-
-#define CODEC_IO_BASE	0x1000
-#define CODEC_IOMAP(a)	((unsigned long)area6_io8_base + ((a) - CODEC_IO_BASE))
-
-static inline unsigned long port2adr(unsigned int port)
-{
-	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
-		if (port == 0x3f6)
-			return ((unsigned long)area5_io16_base + 0x0c);
-		else
-			return ((unsigned long)area5_io16_base + 0x800 +
-				((port-0x1f0) << 1));
-	else
-		maybebadio((unsigned long)port);
-	return port;
-}
-
-/* The 7751R HS7751RVoIP seems to have everything hooked */
-/* up pretty normally (nothing on high-bytes only...) so this */
-/* shouldn't be needed */
-static inline int shifted_port(unsigned long port)
-{
-	/* For IDE registers, value is not shifted */
-	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
-		return 0;
-	else
-		return 1;
-}
-
-#if defined(CONFIG_HS7751RVOIP_CODEC)
-#define codec_port(port)	\
-	((CODEC_IO_BASE <= (port)) && ((port) < (CODEC_IO_BASE + 0x20)))
-#else
-#define codec_port(port)	(0)
-#endif
-
-/*
- * General outline: remap really low stuff [eventually] to SuperIO,
- * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
- * is mapped through the PCI IO window.  Stuff with high bits (PXSEG)
- * should be way beyond the window, and is used  w/o translation for
- * compatibility.
- */
-unsigned char hs7751rvoip_inb(unsigned long port)
-{
-	if (PXSEG(port))
-		return ctrl_inb(port);
-	else if (codec_port(port))
-		return ctrl_inb(CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		return ctrl_inb(pci_ioaddr(port));
-	else
-		return ctrl_inw(port2adr(port)) & 0xff;
-}
-
-unsigned char hs7751rvoip_inb_p(unsigned long port)
-{
-	unsigned char v;
-
-        if (PXSEG(port))
-		v = ctrl_inb(port);
-	else if (codec_port(port))
-		v = ctrl_inb(CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		v = ctrl_inb(pci_ioaddr(port));
-	else
-		v = ctrl_inw(port2adr(port)) & 0xff;
-	ctrl_delay();
-	return v;
-}
-
-unsigned short hs7751rvoip_inw(unsigned long port)
-{
-        if (PXSEG(port))
-		return ctrl_inw(port);
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		return ctrl_inw(pci_ioaddr(port));
-	else
-		maybebadio(port);
-	return 0;
-}
-
-unsigned int hs7751rvoip_inl(unsigned long port)
-{
-        if (PXSEG(port))
-		return ctrl_inl(port);
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		return ctrl_inl(pci_ioaddr(port));
-	else
-		maybebadio(port);
-	return 0;
-}
-
-void hs7751rvoip_outb(unsigned char value, unsigned long port)
-{
-
-        if (PXSEG(port))
-		ctrl_outb(value, port);
-	else if (codec_port(port))
-		ctrl_outb(value, CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		ctrl_outb(value, pci_ioaddr(port));
-	else
-		ctrl_outb(value, port2adr(port));
-}
-
-void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
-{
-        if (PXSEG(port))
-		ctrl_outb(value, port);
-	else if (codec_port(port))
-		ctrl_outb(value, CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		ctrl_outb(value, pci_ioaddr(port));
-	else
-		ctrl_outw(value, port2adr(port));
-
-	ctrl_delay();
-}
-
-void hs7751rvoip_outw(unsigned short value, unsigned long port)
-{
-        if (PXSEG(port))
-		ctrl_outw(value, port);
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		ctrl_outw(value, pci_ioaddr(port));
-	else
-		maybebadio(port);
-}
-
-void hs7751rvoip_outl(unsigned int value, unsigned long port)
-{
-        if (PXSEG(port))
-		ctrl_outl(value, port);
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		ctrl_outl(value, pci_ioaddr(port));
-	else
-		maybebadio(port);
-}
-
-void hs7751rvoip_insb(unsigned long port, void *addr, unsigned long count)
-{
-	u8 *buf = addr;
-
-	if (PXSEG(port))
-		while (count--)
-			*buf++ = ctrl_inb(port);
-	else if (codec_port(port))
-		while (count--)
-			*buf++ = ctrl_inb(CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
-
-		while (count--)
-			*buf++ = *bp;
-	} else {
-		volatile u16 *p = (volatile u16 *)port2adr(port);
-
-		while (count--)
-			*buf++ = *p & 0xff;
-	}
-}
-
-void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
-{
-	volatile u16 *p;
-	u16 *buf = addr;
-
-	if (PXSEG(port))
-		p = (volatile u16 *)port;
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		p = (volatile u16 *)pci_ioaddr(port);
-	else
-		p = (volatile u16 *)port2adr(port);
-	while (count--)
-		*buf++ = *p;
-}
-
-void hs7751rvoip_insl(unsigned long port, void *addr, unsigned long count)
-{
-
-	if (is_pci_ioaddr(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
-		u32 *buf = addr;
-
-		while (count--)
-			*buf++ = *p;
-	} else
-		maybebadio(port);
-}
-
-void hs7751rvoip_outsb(unsigned long port, const void *addr, unsigned long count)
-{
-	const u8 *buf = addr;
-
-	if (PXSEG(port))
-		while (count--)
-			ctrl_outb(*buf++, port);
-	else if (codec_port(port))
-		while (count--)
-			ctrl_outb(*buf++, CODEC_IOMAP(port));
-	else if (is_pci_ioaddr(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
-
-		while (count--)
-			*bp = *buf++;
-	} else {
-		volatile u16 *p = (volatile u16 *)port2adr(port);
-
-		while (count--)
-			*p = *buf++;
-	}
-}
-
-void hs7751rvoip_outsw(unsigned long port, const void *addr, unsigned long count)
-{
-	volatile u16 *p;
-	const u16 *buf = addr;
-
-	if (PXSEG(port))
-		p = (volatile u16 *)port;
-	else if (is_pci_ioaddr(port) || shifted_port(port))
-		p = (volatile u16 *)pci_ioaddr(port);
-	else
-		p = (volatile u16 *)port2adr(port);
-
-	while (count--)
-		*p = *buf++;
-}
-
-void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count)
-{
-	const u32 *buf = addr;
-
-	if (is_pci_ioaddr(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
-
-		while (count--)
-			*p = *buf++;
-	} else
-		maybebadio(port);
-}
-
-void __iomem *hs7751rvoip_ioport_map(unsigned long port, unsigned int size)
-{
-        if (PXSEG(port))
-                return (void __iomem *)port;
-	else if (unlikely(codec_port(port) && (size == 1)))
-		return (void __iomem *)CODEC_IOMAP(port);
-        else if (is_pci_ioaddr(port))
-                return (void __iomem *)pci_ioaddr(port);
-
-        return (void __iomem *)port2adr(port);
-}
diff --git a/arch/sh/boards/renesas/hs7751rvoip/irq.c b/arch/sh/boards/renesas/hs7751rvoip/irq.c
deleted file mode 100644
index e55c6686b21f..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/irq.c
+++ /dev/null
@@ -1,116 +0,0 @@
-/*
- * linux/arch/sh/boards/renesas/hs7751rvoip/irq.c
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Renesas Technology Sales HS7751RVoIP Support.
- *
- * Modified for HS7751RVoIP by
- * Atom Create Engineering Co., Ltd. 2002.
- * Lineo uSolutions, Inc. 2003.
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/interrupt.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/hs7751rvoip.h>
-
-static int mask_pos[] = {8, 9, 10, 11, 12, 13, 0, 1, 2, 3, 4, 5, 6, 7};
-
-static void enable_hs7751rvoip_irq(unsigned int irq);
-static void disable_hs7751rvoip_irq(unsigned int irq);
-
-/* shutdown is same as "disable" */
-#define shutdown_hs7751rvoip_irq disable_hs7751rvoip_irq
-
-static void ack_hs7751rvoip_irq(unsigned int irq);
-static void end_hs7751rvoip_irq(unsigned int irq);
-
-static unsigned int startup_hs7751rvoip_irq(unsigned int irq)
-{
-	enable_hs7751rvoip_irq(irq);
-	return 0; /* never anything pending */
-}
-
-static void disable_hs7751rvoip_irq(unsigned int irq)
-{
-	unsigned short val;
-	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
-
-	/* Set the priority in IPR to 0 */
-	val = ctrl_inw(IRLCNTR3);
-	val &= mask;
-	ctrl_outw(val, IRLCNTR3);
-}
-
-static void enable_hs7751rvoip_irq(unsigned int irq)
-{
-	unsigned short val;
-	unsigned short value = (0x0001 << mask_pos[irq]);
-
-	/* Set priority in IPR back to original value */
-	val = ctrl_inw(IRLCNTR3);
-	val |= value;
-	ctrl_outw(val, IRLCNTR3);
-}
-
-static void ack_hs7751rvoip_irq(unsigned int irq)
-{
-	disable_hs7751rvoip_irq(irq);
-}
-
-static void end_hs7751rvoip_irq(unsigned int irq)
-{
-	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-		enable_hs7751rvoip_irq(irq);
-}
-
-static struct hw_interrupt_type hs7751rvoip_irq_type = {
-	.typename =  "HS7751RVoIP IRQ",
-	.startup = startup_hs7751rvoip_irq,
-	.shutdown = shutdown_hs7751rvoip_irq,
-	.enable = enable_hs7751rvoip_irq,
-	.disable = disable_hs7751rvoip_irq,
-	.ack = ack_hs7751rvoip_irq,
-	.end = end_hs7751rvoip_irq,
-};
-
-static void make_hs7751rvoip_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &hs7751rvoip_irq_type;
-	disable_hs7751rvoip_irq(irq);
-}
-
-/*
- * Initialize IRQ setting
- */
-void __init init_hs7751rvoip_IRQ(void)
-{
-	int i;
-
-	/* IRL0=ON HOOK1
-	 * IRL1=OFF HOOK1
-	 * IRL2=ON HOOK2
-	 * IRL3=OFF HOOK2
-	 * IRL4=Ringing Detection
-	 * IRL5=CODEC
-	 * IRL6=Ethernet
-	 * IRL7=Ethernet Hub
-	 * IRL8=USB Communication
-	 * IRL9=USB Connection
-	 * IRL10=USB DMA
-	 * IRL11=CF Card
-	 * IRL12=PCMCIA
-	 * IRL13=PCI Slot
-	 */
-	ctrl_outw(0x9876, IRLCNTR1);
-	ctrl_outw(0xdcba, IRLCNTR2);
-	ctrl_outw(0x0050, IRLCNTR4);
-	ctrl_outw(0x4321, IRLCNTR5);
-
-	for (i=0; i<14; i++)
-		make_hs7751rvoip_irq(i);
-}
diff --git a/arch/sh/boards/renesas/hs7751rvoip/pci.c b/arch/sh/boards/renesas/hs7751rvoip/pci.c
deleted file mode 100644
index 1c0ddee30d21..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/pci.c
+++ /dev/null
@@ -1,149 +0,0 @@
-/*
- * linux/arch/sh/boards/renesas/hs7751rvoip/pci.c
- *
- * Author:  Ian DaSilva (idasilva@mvista.com)
- *
- * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * PCI initialization for the Renesas SH7751R HS7751RVoIP board
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/delay.h>
-#include <linux/pci.h>
-#include <linux/module.h>
-
-#include <asm/io.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-#include <asm/hs7751rvoip/hs7751rvoip.h>
-
-#define PCIMCR_MRSET_OFF	0xBFFFFFFF
-#define PCIMCR_RFSH_OFF		0xFFFFFFFB
-
-/*
- * Only long word accesses of the PCIC's internal local registers and the
- * configuration registers from the CPU is supported.
- */
-#define PCIC_WRITE(x,v) writel((v), PCI_REG(x))
-#define PCIC_READ(x) readl(PCI_REG(x))
-
-/*
- * Description:  This function sets up and initializes the pcic, sets
- * up the BARS, maps the DRAM into the address space etc, etc.
- */
-int __init pcibios_init_platform(void)
-{
-	unsigned long bcr1, wcr1, wcr2, wcr3, mcr;
-	unsigned short bcr2, bcr3;
-
-	/*
-	 * Initialize the slave bus controller on the pcic.  The values used
-	 * here should not be hardcoded, but they should be taken from the bsc
-	 * on the processor, to make this function as generic as possible.
-	 * (i.e. Another sbc may usr different SDRAM timing settings -- in order
-	 * for the pcic to work, its settings need to be exactly the same.)
-	 */
-	bcr1 = (*(volatile unsigned long *)(SH7751_BCR1));
-	bcr2 = (*(volatile unsigned short *)(SH7751_BCR2));
-	bcr3 = (*(volatile unsigned short *)(SH7751_BCR3));
-	wcr1 = (*(volatile unsigned long *)(SH7751_WCR1));
-	wcr2 = (*(volatile unsigned long *)(SH7751_WCR2));
-	wcr3 = (*(volatile unsigned long *)(SH7751_WCR3));
-	mcr = (*(volatile unsigned long *)(SH7751_MCR));
-
-	bcr1 = bcr1 | 0x00080000;  /* Enable Bit 19, BREQEN */
-	(*(volatile unsigned long *)(SH7751_BCR1)) = bcr1;
-
-	bcr1 = bcr1 | 0x40080000;  /* Enable Bit 19 BREQEN, set PCIC to slave */
-	PCIC_WRITE(SH7751_PCIBCR1, bcr1);	/* PCIC BCR1 */
-	PCIC_WRITE(SH7751_PCIBCR2, bcr2);	/* PCIC BCR2 */
-	PCIC_WRITE(SH7751_PCIBCR3, bcr3);	/* PCIC BCR3 */
-	PCIC_WRITE(SH7751_PCIWCR1, wcr1);	/* PCIC WCR1 */
-	PCIC_WRITE(SH7751_PCIWCR2, wcr2);	/* PCIC WCR2 */
-	PCIC_WRITE(SH7751_PCIWCR3, wcr3);	/* PCIC WCR3 */
-	mcr = (mcr & PCIMCR_MRSET_OFF) & PCIMCR_RFSH_OFF;
-	PCIC_WRITE(SH7751_PCIMCR, mcr);		/* PCIC MCR */
-
-	/* Enable all interrupts, so we know what to fix */
-	PCIC_WRITE(SH7751_PCIINTM, 0x0000c3ff);
-	PCIC_WRITE(SH7751_PCIAINTM, 0x0000380f);
-
-	/* Set up standard PCI config registers */
-	PCIC_WRITE(SH7751_PCICONF1, 0xFB900047); /* Bus Master, Mem & I/O access */
-	PCIC_WRITE(SH7751_PCICONF2, 0x00000000); /* PCI Class code & Revision ID */
-	PCIC_WRITE(SH7751_PCICONF4, 0xab000001); /* PCI I/O address (local regs) */
-	PCIC_WRITE(SH7751_PCICONF5, 0x0c000000); /* PCI MEM address (local RAM)  */
-	PCIC_WRITE(SH7751_PCICONF6, 0xd0000000); /* PCI MEM address (unused) */
-	PCIC_WRITE(SH7751_PCICONF11, 0x35051054); /* PCI Subsystem ID & Vendor ID */
-	PCIC_WRITE(SH7751_PCILSR0, 0x03f00000);	/* MEM (full 64M exposed) */
-	PCIC_WRITE(SH7751_PCILSR1, 0x00000000); /* MEM (unused) */
-	PCIC_WRITE(SH7751_PCILAR0, 0x0c000000); /* MEM (direct map from PCI) */
-	PCIC_WRITE(SH7751_PCILAR1, 0x00000000); /* MEM (unused) */
-
-	/* Now turn it on... */
-	PCIC_WRITE(SH7751_PCICR, 0xa5000001);
-
-	/*
-	 * Set PCIMBR and PCIIOBR here, assuming a single window
-	 * (16M MEM, 256K IO) is enough.  If a larger space is
-	 * needed, the readx/writex and inx/outx functions will
-	 * have to do more (e.g. setting registers for each call).
-	 */
-
-	/*
-	 * Set the MBR so PCI address is one-to-one with window,
-	 * meaning all calls go straight through... use ifdef to
-	 * catch erroneous assumption.
-	 */
-	BUG_ON(PCIBIOS_MIN_MEM != SH7751_PCI_MEMORY_BASE);
-
-	PCIC_WRITE(SH7751_PCIMBR, PCIBIOS_MIN_MEM);
-
-	/* Set IOBR for window containing area specified in pci.h */
-	PCIC_WRITE(SH7751_PCIIOBR, (PCIBIOS_MIN_IO & SH7751_PCIIOBR_MASK));
-
-	/* All done, may as well say so... */
-	printk("SH7751R PCI: Finished initialization of the PCI controller\n");
-
-	return 1;
-}
-
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
-{
-        switch (slot) {
-	case 0: return IRQ_PCISLOT;	/* PCI Extend slot */
-	case 1: return IRQ_PCMCIA;	/* PCI Cardbus Bridge */
-	case 2: return IRQ_PCIETH;	/* Realtek Ethernet controller */
-	case 3: return IRQ_PCIHUB;	/* Realtek Ethernet Hub controller */
-	default:
-		printk("PCI: Bad IRQ mapping request for slot %d\n", slot);
-		return -1;
-	}
-}
-
-static struct resource sh7751_io_resource = {
-	.name	= "SH7751_IO",
-	.start	= 0x4000,
-	.end	= 0x4000 + SH7751_PCI_IO_SIZE - 1,
-	.flags	= IORESOURCE_IO
-};
-
-static struct resource sh7751_mem_resource = {
-	.name	= "SH7751_mem",
-	.start	= SH7751_PCI_MEMORY_BASE,
-	.end	= SH7751_PCI_MEMORY_BASE + SH7751_PCI_MEM_SIZE - 1,
-	.flags	= IORESOURCE_MEM
-};
-
-extern struct pci_ops sh7751_pci_ops;
-
-struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
-	{ NULL, NULL, NULL, 0, 0 },
-};
-EXPORT_SYMBOL(board_pci_channels);
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
deleted file mode 100644
index c05625975f2c..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ /dev/null
@@ -1,105 +0,0 @@
-/*
- * Renesas Technology Sales HS7751RVoIP Support.
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Modified for HS7751RVoIP by
- * Atom Create Engineering Co., Ltd. 2002.
- * Lineo uSolutions, Inc. 2003.
- */
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/mm.h>
-#include <linux/pm.h>
-#include <asm/hs7751rvoip.h>
-#include <asm/io.h>
-#include <asm/machvec.h>
-
-static void hs7751rvoip_power_off(void)
-{
-	ctrl_outw(ctrl_inw(PA_OUTPORTR) & 0xffdf, PA_OUTPORTR);
-}
-
-void *area5_io8_base;
-void *area6_io8_base;
-void *area5_io16_base;
-void *area6_io16_base;
-
-static int __init hs7751rvoip_cf_init(void)
-{
-	pgprot_t prot;
-	unsigned long paddrbase;
-
-	/* open I/O area window */
-	paddrbase = virt_to_phys((void *)(PA_AREA5_IO+0x00000800));
-	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_COM16);
-	area5_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
-	if (!area5_io16_base) {
-		printk("allocate_cf_area : can't open CF I/O window!\n");
-		return -ENOMEM;
-	}
-
-	/* XXX : do we need attribute and common-memory area also? */
-
-	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
-	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_COM8);
-#else
-	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO8);
-#endif
-	area6_io8_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
-	if (!area6_io8_base) {
-		printk("allocate_cf_area : can't open CODEC I/O 8bit window!\n");
-		return -ENOMEM;
-	}
-	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
-	area6_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
-	if (!area6_io16_base) {
-		printk("allocate_cf_area : can't open CODEC I/O 16bit window!\n");
-		return -ENOMEM;
-	}
-
-	return 0;
-}
-device_initcall(hs7751rvoip_cf_init);
-
-/*
- * Initialize the board
- */
-static void __init hs7751rvoip_setup(char **cmdline_p)
-{
-	ctrl_outb(0xf0, PA_OUTPORTR);
-	pm_power_off = hs7751rvoip_power_off;
-
-	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
-}
-
-static struct sh_machine_vector mv_hs7751rvoip __initmv = {
-	.mv_name		= "HS7751RVoIP",
-	.mv_setup		= hs7751rvoip_setup,
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= hs7751rvoip_inb,
-	.mv_inw			= hs7751rvoip_inw,
-	.mv_inl			= hs7751rvoip_inl,
-	.mv_outb		= hs7751rvoip_outb,
-	.mv_outw		= hs7751rvoip_outw,
-	.mv_outl		= hs7751rvoip_outl,
-
-	.mv_inb_p		= hs7751rvoip_inb_p,
-	.mv_inw_p		= hs7751rvoip_inw,
-	.mv_inl_p		= hs7751rvoip_inl,
-	.mv_outb_p		= hs7751rvoip_outb_p,
-	.mv_outw_p		= hs7751rvoip_outw,
-	.mv_outl_p		= hs7751rvoip_outl,
-
-	.mv_insb		= hs7751rvoip_insb,
-	.mv_insw		= hs7751rvoip_insw,
-	.mv_insl		= hs7751rvoip_insl,
-	.mv_outsb		= hs7751rvoip_outsb,
-	.mv_outsw		= hs7751rvoip_outsw,
-	.mv_outsl		= hs7751rvoip_outsl,
-
-	.mv_init_irq		= init_hs7751rvoip_IRQ,
-	.mv_ioport_map		= hs7751rvoip_ioport_map,
-};
diff --git a/arch/sh/configs/hs7751rvoip_defconfig b/arch/sh/configs/hs7751rvoip_defconfig
deleted file mode 100644
index 5d9da5a02759..000000000000
--- a/arch/sh/configs/hs7751rvoip_defconfig
+++ /dev/null
@@ -1,908 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18
-# Tue Oct  3 13:04:52 2006
-#
-CONFIG_SUPERH=y
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-CONFIG_GENERIC_FIND_NEXT_BIT=y
-CONFIG_GENERIC_HWEIGHT=y
-CONFIG_GENERIC_HARDIRQS=y
-CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
-
-#
-# Code maturity level options
-#
-CONFIG_EXPERIMENTAL=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_LOCK_KERNEL=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-CONFIG_LOCALVERSION_AUTO=y
-CONFIG_SWAP=y
-CONFIG_SYSVIPC=y
-# CONFIG_IPC_NS is not set
-CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-# CONFIG_BSD_PROCESS_ACCT_V3 is not set
-# CONFIG_TASKSTATS is not set
-# CONFIG_UTS_NS is not set
-# CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
-# CONFIG_RELAY is not set
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_SYSCTL=y
-CONFIG_EMBEDDED=y
-CONFIG_UID16=y
-# CONFIG_SYSCTL_SYSCALL is not set
-# CONFIG_KALLSYMS is not set
-CONFIG_HOTPLUG=y
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-CONFIG_ELF_CORE=y
-CONFIG_BASE_FULL=y
-CONFIG_FUTEX=y
-CONFIG_EPOLL=y
-CONFIG_SHMEM=y
-CONFIG_SLAB=y
-CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_RT_MUTEXES=y
-# CONFIG_TINY_SHMEM is not set
-CONFIG_BASE_SMALL=0
-# CONFIG_SLOB is not set
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_MODULE_FORCE_UNLOAD is not set
-# CONFIG_MODVERSIONS is not set
-# CONFIG_MODULE_SRCVERSION_ALL is not set
-CONFIG_KMOD=y
-
-#
-# Block layer
-#
-CONFIG_BLOCK=y
-# CONFIG_LBD is not set
-# CONFIG_BLK_DEV_IO_TRACE is not set
-# CONFIG_LSF is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-CONFIG_IOSCHED_AS=y
-CONFIG_IOSCHED_DEADLINE=y
-CONFIG_IOSCHED_CFQ=y
-CONFIG_DEFAULT_AS=y
-# CONFIG_DEFAULT_DEADLINE is not set
-# CONFIG_DEFAULT_CFQ is not set
-# CONFIG_DEFAULT_NOOP is not set
-CONFIG_DEFAULT_IOSCHED="anticipatory"
-
-#
-# System type
-#
-# CONFIG_SH_SOLUTION_ENGINE is not set
-# CONFIG_SH_7751_SOLUTION_ENGINE is not set
-# CONFIG_SH_7300_SOLUTION_ENGINE is not set
-# CONFIG_SH_7343_SOLUTION_ENGINE is not set
-# CONFIG_SH_73180_SOLUTION_ENGINE is not set
-# CONFIG_SH_7751_SYSTEMH is not set
-# CONFIG_SH_HP6XX is not set
-# CONFIG_SH_EC3104 is not set
-# CONFIG_SH_SATURN is not set
-# CONFIG_SH_DREAMCAST is not set
-# CONFIG_SH_BIGSUR is not set
-# CONFIG_SH_MPC1211 is not set
-# CONFIG_SH_SH03 is not set
-# CONFIG_SH_SECUREEDGE5410 is not set
-CONFIG_SH_HS7751RVOIP=y
-# CONFIG_SH_7710VOIPGW is not set
-# CONFIG_SH_RTS7751R2D is not set
-# CONFIG_SH_R7780RP is not set
-# CONFIG_SH_EDOSK7705 is not set
-# CONFIG_SH_SH4202_MICRODEV is not set
-# CONFIG_SH_LANDISK is not set
-# CONFIG_SH_TITAN is not set
-# CONFIG_SH_SHMIN is not set
-# CONFIG_SH_UNKNOWN is not set
-
-#
-# Processor selection
-#
-CONFIG_CPU_SH4=y
-
-#
-# SH-2 Processor Support
-#
-# CONFIG_CPU_SUBTYPE_SH7604 is not set
-
-#
-# SH-3 Processor Support
-#
-# CONFIG_CPU_SUBTYPE_SH7300 is not set
-# CONFIG_CPU_SUBTYPE_SH7705 is not set
-# CONFIG_CPU_SUBTYPE_SH7706 is not set
-# CONFIG_CPU_SUBTYPE_SH7707 is not set
-# CONFIG_CPU_SUBTYPE_SH7708 is not set
-# CONFIG_CPU_SUBTYPE_SH7709 is not set
-# CONFIG_CPU_SUBTYPE_SH7710 is not set
-
-#
-# SH-4 Processor Support
-#
-# CONFIG_CPU_SUBTYPE_SH7750 is not set
-# CONFIG_CPU_SUBTYPE_SH7091 is not set
-# CONFIG_CPU_SUBTYPE_SH7750R is not set
-# CONFIG_CPU_SUBTYPE_SH7750S is not set
-# CONFIG_CPU_SUBTYPE_SH7751 is not set
-CONFIG_CPU_SUBTYPE_SH7751R=y
-# CONFIG_CPU_SUBTYPE_SH7760 is not set
-# CONFIG_CPU_SUBTYPE_SH4_202 is not set
-
-#
-# ST40 Processor Support
-#
-# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
-# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
-
-#
-# SH-4A Processor Support
-#
-# CONFIG_CPU_SUBTYPE_SH7770 is not set
-# CONFIG_CPU_SUBTYPE_SH7780 is not set
-
-#
-# SH4AL-DSP Processor Support
-#
-# CONFIG_CPU_SUBTYPE_SH73180 is not set
-# CONFIG_CPU_SUBTYPE_SH7343 is not set
-
-#
-# Memory management options
-#
-CONFIG_MMU=y
-CONFIG_PAGE_OFFSET=0x80000000
-CONFIG_MEMORY_START=0x0c000000
-CONFIG_MEMORY_SIZE=0x04000000
-CONFIG_VSYSCALL=y
-CONFIG_SELECT_MEMORY_MODEL=y
-CONFIG_FLATMEM_MANUAL=y
-# CONFIG_DISCONTIGMEM_MANUAL is not set
-# CONFIG_SPARSEMEM_MANUAL is not set
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-# CONFIG_SPARSEMEM_STATIC is not set
-CONFIG_SPLIT_PTLOCK_CPUS=4
-# CONFIG_RESOURCES_64BIT is not set
-
-#
-# Cache configuration
-#
-# CONFIG_SH_DIRECT_MAPPED is not set
-# CONFIG_SH_WRITETHROUGH is not set
-# CONFIG_SH_OCRAM is not set
-
-#
-# Processor features
-#
-CONFIG_CPU_LITTLE_ENDIAN=y
-CONFIG_SH_FPU=y
-# CONFIG_SH_DSP is not set
-# CONFIG_SH_STORE_QUEUES is not set
-CONFIG_CPU_HAS_INTEVT=y
-CONFIG_CPU_HAS_SR_RB=y
-
-#
-# Timer support
-#
-CONFIG_SH_TMU=y
-
-#
-# HS7751RVoIP options
-#
-CONFIG_HS7751RVOIP_CODEC=y
-CONFIG_SH_PCLK_FREQ=60000000
-
-#
-# CPU Frequency scaling
-#
-# CONFIG_CPU_FREQ is not set
-
-#
-# DMA support
-#
-# CONFIG_SH_DMA is not set
-
-#
-# Companion Chips
-#
-# CONFIG_HD6446X_SERIES is not set
-
-#
-# Kernel features
-#
-# CONFIG_HZ_100 is not set
-CONFIG_HZ_250=y
-# CONFIG_HZ_1000 is not set
-CONFIG_HZ=250
-# CONFIG_KEXEC is not set
-# CONFIG_SMP is not set
-# CONFIG_PREEMPT_NONE is not set
-# CONFIG_PREEMPT_VOLUNTARY is not set
-CONFIG_PREEMPT=y
-CONFIG_PREEMPT_BKL=y
-
-#
-# Boot options
-#
-CONFIG_ZERO_PAGE_OFFSET=0x00001000
-CONFIG_BOOT_LINK_OFFSET=0x00800000
-# CONFIG_UBC_WAKEUP is not set
-CONFIG_CMDLINE_BOOL=y
-CONFIG_CMDLINE="mem=64M console=ttySC1,115200 root=/dev/hda1"
-
-#
-# Bus options
-#
-# CONFIG_PCI is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_ELF=y
-# CONFIG_BINFMT_FLAT is not set
-# CONFIG_BINFMT_MISC is not set
-
-#
-# Power management options (EXPERIMENTAL)
-#
-# CONFIG_PM is not set
-
-#
-# Networking
-#
-CONFIG_NET=y
-
-#
-# Networking options
-#
-# CONFIG_NETDEBUG is not set
-CONFIG_PACKET=y
-CONFIG_PACKET_MMAP=y
-CONFIG_UNIX=y
-CONFIG_XFRM=y
-# CONFIG_XFRM_USER is not set
-# CONFIG_XFRM_SUB_POLICY is not set
-# CONFIG_NET_KEY is not set
-CONFIG_INET=y
-CONFIG_IP_MULTICAST=y
-CONFIG_IP_ADVANCED_ROUTER=y
-CONFIG_ASK_IP_FIB_HASH=y
-# CONFIG_IP_FIB_TRIE is not set
-CONFIG_IP_FIB_HASH=y
-# CONFIG_IP_MULTIPLE_TABLES is not set
-# CONFIG_IP_ROUTE_MULTIPATH is not set
-# CONFIG_IP_ROUTE_VERBOSE is not set
-# CONFIG_IP_PNP is not set
-# CONFIG_NET_IPIP is not set
-# CONFIG_NET_IPGRE is not set
-# CONFIG_IP_MROUTE is not set
-# CONFIG_ARPD is not set
-# CONFIG_SYN_COOKIES is not set
-# CONFIG_INET_AH is not set
-# CONFIG_INET_ESP is not set
-# CONFIG_INET_IPCOMP is not set
-# CONFIG_INET_XFRM_TUNNEL is not set
-# CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_DIAG=y
-CONFIG_INET_TCP_DIAG=y
-# CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_CUBIC=y
-CONFIG_DEFAULT_TCP_CONG="cubic"
-# CONFIG_IPV6 is not set
-# CONFIG_INET6_XFRM_TUNNEL is not set
-# CONFIG_INET6_TUNNEL is not set
-# CONFIG_NETWORK_SECMARK is not set
-# CONFIG_NETFILTER is not set
-
-#
-# DCCP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_DCCP is not set
-
-#
-# SCTP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_SCTP is not set
-
-#
-# TIPC Configuration (EXPERIMENTAL)
-#
-# CONFIG_TIPC is not set
-# CONFIG_ATM is not set
-# CONFIG_BRIDGE is not set
-# CONFIG_VLAN_8021Q is not set
-# CONFIG_DECNET is not set
-# CONFIG_LLC2 is not set
-# CONFIG_IPX is not set
-# CONFIG_ATALK is not set
-# CONFIG_X25 is not set
-# CONFIG_LAPB is not set
-# CONFIG_ECONET is not set
-# CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
-# CONFIG_NET_SCHED is not set
-
-#
-# Network testing
-#
-# CONFIG_NET_PKTGEN is not set
-# CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
-# CONFIG_BT is not set
-# CONFIG_IEEE80211 is not set
-
-#
-# Device Drivers
-#
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-CONFIG_FW_LOADER=m
-# CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
-# CONFIG_CONNECTOR is not set
-
-#
-# Memory Technology Devices (MTD)
-#
-# CONFIG_MTD is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_NBD is not set
-# CONFIG_BLK_DEV_RAM is not set
-# CONFIG_BLK_DEV_INITRD is not set
-# CONFIG_CDROM_PKTCDVD is not set
-# CONFIG_ATA_OVER_ETH is not set
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-CONFIG_IDE=y
-CONFIG_IDE_MAX_HWIFS=1
-CONFIG_BLK_DEV_IDE=y
-
-#
-# Please see Documentation/ide.txt for help/info on IDE drives
-#
-# CONFIG_BLK_DEV_IDE_SATA is not set
-CONFIG_BLK_DEV_IDEDISK=y
-# CONFIG_IDEDISK_MULTI_MODE is not set
-# CONFIG_BLK_DEV_IDECD is not set
-# CONFIG_BLK_DEV_IDETAPE is not set
-# CONFIG_BLK_DEV_IDEFLOPPY is not set
-# CONFIG_IDE_TASK_IOCTL is not set
-
-#
-# IDE chipset support/bugfixes
-#
-CONFIG_IDE_GENERIC=y
-# CONFIG_IDE_ARM is not set
-# CONFIG_BLK_DEV_IDEDMA is not set
-# CONFIG_IDEDMA_AUTO is not set
-# CONFIG_BLK_DEV_HD is not set
-
-#
-# SCSI device support
-#
-# CONFIG_RAID_ATTRS is not set
-# CONFIG_SCSI is not set
-# CONFIG_SCSI_NETLINK is not set
-
-#
-# Serial ATA (prod) and Parallel ATA (experimental) drivers
-#
-# CONFIG_ATA is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-
-#
-# I2O device support
-#
-
-#
-# Network device support
-#
-CONFIG_NETDEVICES=y
-# CONFIG_DUMMY is not set
-# CONFIG_BONDING is not set
-# CONFIG_EQUALIZER is not set
-# CONFIG_TUN is not set
-
-#
-# PHY device support
-#
-# CONFIG_PHYLIB is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
-CONFIG_NET_ETHERNET=y
-CONFIG_MII=y
-# CONFIG_STNIC is not set
-# CONFIG_SMC91X is not set
-
-#
-# Ethernet (1000 Mbit)
-#
-
-#
-# Ethernet (10000 Mbit)
-#
-
-#
-# Token Ring devices
-#
-
-#
-# Wireless LAN (non-hamradio)
-#
-# CONFIG_NET_RADIO is not set
-
-#
-# Wan interfaces
-#
-# CONFIG_WAN is not set
-# CONFIG_PPP is not set
-# CONFIG_SLIP is not set
-# CONFIG_SHAPER is not set
-# CONFIG_NETCONSOLE is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-# CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
-# CONFIG_PHONE is not set
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-# CONFIG_INPUT_FF_MEMLESS is not set
-
-#
-# Userland interfaces
-#
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-CONFIG_SERIO=y
-CONFIG_SERIO_I8042=y
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-# CONFIG_VT is not set
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-# CONFIG_SERIAL_8250 is not set
-
-#
-# Non-8250 serial port support
-#
-CONFIG_SERIAL_SH_SCI=y
-CONFIG_SERIAL_SH_SCI_NR_UARTS=2
-CONFIG_SERIAL_SH_SCI_CONSOLE=y
-CONFIG_SERIAL_CORE=y
-CONFIG_SERIAL_CORE_CONSOLE=y
-CONFIG_UNIX98_PTYS=y
-# CONFIG_LEGACY_PTYS is not set
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-CONFIG_HW_RANDOM=y
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-# CONFIG_TCG_TPM is not set
-# CONFIG_TELCLOCK is not set
-
-#
-# I2C support
-#
-# CONFIG_I2C is not set
-
-#
-# SPI support
-#
-# CONFIG_SPI is not set
-# CONFIG_SPI_MASTER is not set
-
-#
-# Dallas's 1-wire bus
-#
-
-#
-# Hardware Monitoring support
-#
-CONFIG_HWMON=y
-# CONFIG_HWMON_VID is not set
-# CONFIG_SENSORS_ABITUGURU is not set
-# CONFIG_SENSORS_F71805F is not set
-# CONFIG_SENSORS_VT1211 is not set
-# CONFIG_HWMON_DEBUG_CHIP is not set
-
-#
-# Misc devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-CONFIG_VIDEO_V4L2=y
-
-#
-# Digital Video Broadcasting Devices
-#
-# CONFIG_DVB is not set
-
-#
-# Graphics support
-#
-CONFIG_FIRMWARE_EDID=y
-# CONFIG_FB is not set
-# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-# CONFIG_USB_ARCH_HAS_HCD is not set
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# MMC/SD Card support
-#
-# CONFIG_MMC is not set
-
-#
-# LED devices
-#
-# CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
-# CONFIG_RTC_CLASS is not set
-
-#
-# DMA Engine support
-#
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# File systems
-#
-CONFIG_EXT2_FS=y
-# CONFIG_EXT2_FS_XATTR is not set
-# CONFIG_EXT2_FS_XIP is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-# CONFIG_XFS_FS is not set
-# CONFIG_OCFS2_FS is not set
-# CONFIG_MINIX_FS is not set
-# CONFIG_ROMFS_FS is not set
-CONFIG_INOTIFY=y
-CONFIG_INOTIFY_USER=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-# CONFIG_FUSE_FS is not set
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_PROC_KCORE=y
-CONFIG_PROC_SYSCTL=y
-CONFIG_SYSFS=y
-CONFIG_TMPFS=y
-# CONFIG_TMPFS_POSIX_ACL is not set
-# CONFIG_HUGETLBFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-# CONFIG_CONFIGFS_FS is not set
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_ADFS_FS is not set
-# CONFIG_AFFS_FS is not set
-# CONFIG_HFS_FS is not set
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_BEFS_FS is not set
-# CONFIG_BFS_FS is not set
-# CONFIG_EFS_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-# CONFIG_NFS_V3_ACL is not set
-CONFIG_NFS_V4=y
-CONFIG_NFS_DIRECTIO=y
-# CONFIG_NFSD is not set
-CONFIG_LOCKD=y
-CONFIG_LOCKD_V4=y
-CONFIG_NFS_COMMON=y
-CONFIG_SUNRPC=y
-CONFIG_SUNRPC_GSS=y
-CONFIG_RPCSEC_GSS_KRB5=y
-# CONFIG_RPCSEC_GSS_SPKM3 is not set
-# CONFIG_SMB_FS is not set
-# CONFIG_CIFS is not set
-# CONFIG_NCP_FS is not set
-# CONFIG_CODA_FS is not set
-# CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
-
-#
-# Partition Types
-#
-CONFIG_PARTITION_ADVANCED=y
-# CONFIG_ACORN_PARTITION is not set
-# CONFIG_OSF_PARTITION is not set
-# CONFIG_AMIGA_PARTITION is not set
-# CONFIG_ATARI_PARTITION is not set
-# CONFIG_MAC_PARTITION is not set
-CONFIG_MSDOS_PARTITION=y
-# CONFIG_BSD_DISKLABEL is not set
-# CONFIG_MINIX_SUBPARTITION is not set
-# CONFIG_SOLARIS_X86_PARTITION is not set
-# CONFIG_UNIXWARE_DISKLABEL is not set
-# CONFIG_LDM_PARTITION is not set
-# CONFIG_SGI_PARTITION is not set
-# CONFIG_ULTRIX_PARTITION is not set
-# CONFIG_SUN_PARTITION is not set
-# CONFIG_KARMA_PARTITION is not set
-# CONFIG_EFI_PARTITION is not set
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Profiling support
-#
-# CONFIG_PROFILING is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-CONFIG_ENABLE_MUST_CHECK=y
-# CONFIG_MAGIC_SYSRQ is not set
-# CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_KERNEL is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_DEBUG_BUGVERBOSE is not set
-# CONFIG_DEBUG_FS is not set
-# CONFIG_SH_STANDARD_BIOS is not set
-# CONFIG_EARLY_SCIF_CONSOLE is not set
-# CONFIG_KGDB is not set
-
-#
-# Security options
-#
-# CONFIG_KEYS is not set
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-CONFIG_CRYPTO=y
-CONFIG_CRYPTO_ALGAPI=y
-CONFIG_CRYPTO_BLKCIPHER=m
-CONFIG_CRYPTO_MANAGER=m
-# CONFIG_CRYPTO_HMAC is not set
-# CONFIG_CRYPTO_NULL is not set
-# CONFIG_CRYPTO_MD4 is not set
-CONFIG_CRYPTO_MD5=y
-# CONFIG_CRYPTO_SHA1 is not set
-# CONFIG_CRYPTO_SHA256 is not set
-# CONFIG_CRYPTO_SHA512 is not set
-# CONFIG_CRYPTO_WP512 is not set
-# CONFIG_CRYPTO_TGR192 is not set
-CONFIG_CRYPTO_ECB=m
-CONFIG_CRYPTO_CBC=m
-CONFIG_CRYPTO_DES=y
-# CONFIG_CRYPTO_BLOWFISH is not set
-# CONFIG_CRYPTO_TWOFISH is not set
-# CONFIG_CRYPTO_SERPENT is not set
-# CONFIG_CRYPTO_AES is not set
-# CONFIG_CRYPTO_CAST5 is not set
-# CONFIG_CRYPTO_CAST6 is not set
-# CONFIG_CRYPTO_TEA is not set
-# CONFIG_CRYPTO_ARC4 is not set
-# CONFIG_CRYPTO_KHAZAD is not set
-# CONFIG_CRYPTO_ANUBIS is not set
-# CONFIG_CRYPTO_DEFLATE is not set
-# CONFIG_CRYPTO_MICHAEL_MIC is not set
-# CONFIG_CRYPTO_CRC32C is not set
-# CONFIG_CRYPTO_TEST is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-# CONFIG_CRC16 is not set
-CONFIG_CRC32=y
-# CONFIG_LIBCRC32C is not set
-CONFIG_PLIST=y
diff --git a/arch/sh/drivers/pci/pci-auto.c b/arch/sh/drivers/pci/pci-auto.c
index 224e007736fb..ea404704ace8 100644
--- a/arch/sh/drivers/pci/pci-auto.c
+++ b/arch/sh/drivers/pci/pci-auto.c
@@ -516,10 +516,8 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 					 PCI_COMMAND, cmdstat | PCI_COMMAND_IO |
 					 PCI_COMMAND_MEMORY |
 					 PCI_COMMAND_MASTER);
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
 		early_write_config_byte(hose, top_bus, current_bus, pci_devfn,
 					PCI_LATENCY_TIMER, 0x80);
-#endif
 
 		/* Allocate PCI I/O and/or memory space */
 		pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_5);
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index af4b7aa9a6b2..25810670a0fa 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -29,7 +29,6 @@ HP6XX			SH_HP6XX
 DREAMCAST		SH_DREAMCAST
 MPC1211			SH_MPC1211
 SNAPGEAR		SH_SECUREEDGE5410
-HS7751RVOIP		SH_HS7751RVOIP
 EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
diff --git a/include/asm-sh/hs7751rvoip.h b/include/asm-sh/hs7751rvoip.h
deleted file mode 100644
index c4cff9d33927..000000000000
--- a/include/asm-sh/hs7751rvoip.h
+++ /dev/null
@@ -1,54 +0,0 @@
-#ifndef __ASM_SH_RENESAS_HS7751RVOIP_H
-#define __ASM_SH_RENESAS_HS7751RVOIP_H
-
-/*
- * linux/include/asm-sh/hs7751rvoip/hs7751rvoip.h
- *
- * Copyright (C) 2000  Atom Create Engineering Co., Ltd.
- *
- * Renesas Technology Sales HS7751RVoIP support
- */
-
-/* Box specific addresses.  */
-
-#define PA_BCR		0xa4000000	/* FPGA */
-#define PA_SLICCNTR1	0xa4000006	/* SLIC PIO Control 1 */
-#define PA_SLICCNTR2	0xa4000008	/* SLIC PIO Control 2 */
-#define PA_DMACNTR	0xa400000a	/* USB DMA Control */
-#define PA_INPORTR	0xa400000c	/* Input Port Register */
-#define PA_OUTPORTR	0xa400000e	/* Output Port Reguster */
-#define PA_VERREG	0xa4000014	/* FPGA Version Register */
-
-#define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
-
-#define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
-#define IRLCNTR2	(PA_BCR + 2)	/* Interrupt Control Register2 */
-#define IRLCNTR3	(PA_BCR + 4)	/* Interrupt Control Register3 */
-#define IRLCNTR4	(PA_BCR + 16)	/* Interrupt Control Register4 */
-#define IRLCNTR5	(PA_BCR + 18)	/* Interrupt Control Register5 */
-
-#define IRQ_PCIETH	6		/* PCI Ethernet IRQ */
-#define IRQ_PCIHUB	7		/* PCI Ethernet Hub IRQ */
-#define IRQ_USBCOM	8		/* USB Comunication IRQ */
-#define IRQ_USBCON	9		/* USB Connect IRQ */
-#define IRQ_USBDMA	10		/* USB DMA IRQ */
-#define IRQ_CFCARD	11		/* CF Card IRQ */
-#define IRQ_PCMCIA	12		/* PCMCIA IRQ */
-#define IRQ_PCISLOT	13		/* PCI Slot #1 IRQ */
-#define IRQ_ONHOOK1	0		/* ON HOOK1 IRQ */
-#define IRQ_OFFHOOK1	1		/* OFF HOOK1 IRQ */
-#define IRQ_ONHOOK2	2		/* ON HOOK2 IRQ */
-#define IRQ_OFFHOOK2	3		/* OFF HOOK2 IRQ */
-#define	IRQ_RINGING	4		/* Ringing IRQ */
-#define	IRQ_CODEC	5		/* CODEC IRQ */
-
-#define __IO_PREFIX	hs7751rvoip
-#include <asm/io_generic.h>
-
-/* arch/sh/boards/renesas/hs7751rvoip/irq.c */
-void init_hs7751rvoip_IRQ(void);
-
-/* arch/sh/boards/renesas/hs7751rvoip/io.c */
-void *hs7751rvoip_ioremap(unsigned long, unsigned long);
-
-#endif  /* __ASM_SH_RENESAS_HS7751RVOIP */
-- 
cgit v1.2.3


From 4862ec073975e28f432f164405e60fa6f5c9d071 Mon Sep 17 00:00:00 2001
From: Nicholas Beck <nbeck@mpc-data.co.uk>
Date: Wed, 23 Jan 2008 12:50:51 +0900
Subject: sh: Add support for SDK7780 board.

Add support for Renesas Technology Europe SDK7780 board.

Signed-off-by: Nicholas Beck <nbeck@mpc-data.co.uk>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                         |    9 +
 arch/sh/Makefile                        |    1 +
 arch/sh/boards/renesas/sdk7780/Kconfig  |   23 +
 arch/sh/boards/renesas/sdk7780/Makefile |    5 +
 arch/sh/boards/renesas/sdk7780/irq.c    |   46 +
 arch/sh/boards/renesas/sdk7780/setup.c  |  109 +++
 arch/sh/configs/sdk7780_defconfig       | 1394 +++++++++++++++++++++++++++++++
 arch/sh/drivers/pci/Makefile            |    1 +
 arch/sh/drivers/pci/fixups-sdk7780.c    |   59 ++
 arch/sh/drivers/pci/ops-sdk7780.c       |   73 ++
 include/asm-sh/sdk7780.h                |   81 ++
 11 files changed, 1801 insertions(+)
 create mode 100644 arch/sh/boards/renesas/sdk7780/Kconfig
 create mode 100644 arch/sh/boards/renesas/sdk7780/Makefile
 create mode 100644 arch/sh/boards/renesas/sdk7780/irq.c
 create mode 100644 arch/sh/boards/renesas/sdk7780/setup.c
 create mode 100644 arch/sh/configs/sdk7780_defconfig
 create mode 100644 arch/sh/drivers/pci/fixups-sdk7780.c
 create mode 100644 arch/sh/drivers/pci/ops-sdk7780.c
 create mode 100644 include/asm-sh/sdk7780.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index e94430b42474..1cd9c8fd927d 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -455,6 +455,14 @@ config SH_RTS7751R2D
 	  Select RTS7751R2D if configuring for a Renesas Technology
 	  Sales SH-Graphics board.
 
+config SH_SDK7780
+	bool "SDK7780R3"
+	depends on CPU_SUBTYPE_SH7780
+	select SYS_SUPPORTS_PCI
+	help
+	  Select SDK7780 if configuring for a Renesas SH7780 SDK7780R3
+	  evaluation board.
+
 config SH_HIGHLANDER
 	bool "Highlander"
 	depends on CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785
@@ -520,6 +528,7 @@ endmenu
 
 source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 source "arch/sh/boards/renesas/r7780rp/Kconfig"
+source "arch/sh/boards/renesas/sdk7780/Kconfig"
 source "arch/sh/boards/magicpanelr2/Kconfig"
 
 menu "Timer and clock configuration"
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 4cc71094efe7..17fc36186bf4 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -116,6 +116,7 @@ machdir-$(CONFIG_SH_RTS7751R2D)			+= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		+= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			+= renesas/edosk7705
 machdir-$(CONFIG_SH_HIGHLANDER)			+= renesas/r7780rp
+machdir-$(CONFIG_SH_SDK7780)			+= renesas/sdk7780
 machdir-$(CONFIG_SH_7710VOIPGW)			+= renesas/sh7710voipgw
 machdir-$(CONFIG_SH_X3PROTO)			+= renesas/x3proto
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		+= superh/microdev
diff --git a/arch/sh/boards/renesas/sdk7780/Kconfig b/arch/sh/boards/renesas/sdk7780/Kconfig
new file mode 100644
index 000000000000..e4f5b6985be1
--- /dev/null
+++ b/arch/sh/boards/renesas/sdk7780/Kconfig
@@ -0,0 +1,23 @@
+if SH_SDK7780
+
+choice
+	prompt "SDK7780 options"
+	default SH_SDK7780_BASE
+
+config SH_SDK7780_STANDALONE
+	bool "SDK7780 board support"
+	depends on CPU_SUBTYPE_SH7780
+	help
+	  Selecting this option will enable support for the
+	  standalone version of the SDK7780. If in doubt, say Y.
+
+config SH_SDK7780_BASE
+	bool "SDK7780 with base-board support"
+	depends on CPU_SUBTYPE_SH7780
+	help
+	  Selecting this option will enable support for the expansion
+	  baseboard devices. If in doubt, say Y.
+
+endchoice
+
+endif
diff --git a/arch/sh/boards/renesas/sdk7780/Makefile b/arch/sh/boards/renesas/sdk7780/Makefile
new file mode 100644
index 000000000000..3d8f0befc35d
--- /dev/null
+++ b/arch/sh/boards/renesas/sdk7780/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the SDK7780 specific parts of the kernel
+#
+obj-y	 := setup.o irq.o
+
diff --git a/arch/sh/boards/renesas/sdk7780/irq.c b/arch/sh/boards/renesas/sdk7780/irq.c
new file mode 100644
index 000000000000..87cdc578f6ff
--- /dev/null
+++ b/arch/sh/boards/renesas/sdk7780/irq.c
@@ -0,0 +1,46 @@
+/*
+ * linux/arch/sh/boards/renesas/sdk7780/irq.c
+ *
+ * Renesas Technology Europe SDK7780 Support.
+ *
+ * Copyright (C) 2008  Nicholas Beck <nbeck@mpc-data.co.uk>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <asm/sdk7780.h>
+
+enum {
+	UNUSED = 0,
+	/* board specific interrupt sources */
+	SMC91C111,	/* Ethernet controller */
+};
+
+static struct intc_vect fpga_vectors[] __initdata = {
+	INTC_IRQ(SMC91C111, IRQ_ETHERNET),
+};
+
+static struct intc_mask_reg fpga_mask_registers[] __initdata = {
+	{ 0, FPGA_IRQ0MR, 16,
+	  { 0, 0, 0, 0, 0, 0, 0, 0,
+	    0, 0, 0, SMC91C111, 0, 0, 0, 0 } },
+};
+
+static DECLARE_INTC_DESC(fpga_intc_desc, "sdk7780-irq", fpga_vectors,
+			 NULL, fpga_mask_registers, NULL, NULL);
+
+void __init init_sdk7780_IRQ(void)
+{
+	printk(KERN_INFO "Using SDK7780 interrupt controller.\n");
+
+	ctrl_outw(0xFFFF, FPGA_IRQ0MR);
+	/* Setup IRL 0-3 */
+	ctrl_outw(0x0003, FPGA_IMSR);
+	plat_irq_setup_pins(IRQ_MODE_IRL3210);
+
+	register_intc_controller(&fpga_intc_desc);
+}
diff --git a/arch/sh/boards/renesas/sdk7780/setup.c b/arch/sh/boards/renesas/sdk7780/setup.c
new file mode 100644
index 000000000000..5df32f201870
--- /dev/null
+++ b/arch/sh/boards/renesas/sdk7780/setup.c
@@ -0,0 +1,109 @@
+/*
+ * arch/sh/boards/renesas/sdk7780/setup.c
+ *
+ * Renesas Solutions SH7780 SDK Support
+ * Copyright (C) 2008 Nicholas Beck <nbeck@mpc-data.co.uk>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/platform_device.h>
+#include <linux/pata_platform.h>
+#include <asm/machvec.h>
+#include <asm/sdk7780.h>
+#include <asm/heartbeat.h>
+#include <asm/io.h>
+#include <asm/addrspace.h>
+
+#define GPIO_PECR        0xFFEA0008
+
+//* Heartbeat */
+static struct heartbeat_data heartbeat_data = {
+	.regsize = 16,
+};
+
+static struct resource heartbeat_resources[] = {
+	[0] = {
+		.start  = PA_LED,
+		.end    = PA_LED,
+		.flags  = IORESOURCE_MEM,
+	},
+};
+
+static struct platform_device heartbeat_device = {
+	.name           = "heartbeat",
+	.id             = -1,
+	.dev = {
+		.platform_data = &heartbeat_data,
+	},
+	.num_resources  = ARRAY_SIZE(heartbeat_resources),
+	.resource       = heartbeat_resources,
+};
+
+/* SMC91x */
+static struct resource smc91x_eth_resources[] = {
+	[0] = {
+		.name   = "smc91x-regs" ,
+		.start  = PA_LAN + 0x300,
+		.end    = PA_LAN + 0x300 + 0x10 ,
+		.flags  = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start  = IRQ_ETHERNET,
+		.end    = IRQ_ETHERNET,
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device smc91x_eth_device = {
+	.name           = "smc91x",
+	.id             = 0,
+	.dev = {
+		.dma_mask               = NULL,         /* don't use dma */
+		.coherent_dma_mask      = 0xffffffff,
+	},
+	.num_resources  = ARRAY_SIZE(smc91x_eth_resources),
+	.resource       = smc91x_eth_resources,
+};
+
+static struct platform_device *sdk7780_devices[] __initdata = {
+	&heartbeat_device,
+	&smc91x_eth_device,
+};
+
+static int __init sdk7780_devices_setup(void)
+{
+	return platform_add_devices(sdk7780_devices,
+		ARRAY_SIZE(sdk7780_devices));
+}
+device_initcall(sdk7780_devices_setup);
+
+static void __init sdk7780_setup(char **cmdline_p)
+{
+	u16 ver = ctrl_inw(FPGA_FPVERR);
+	u16 dateStamp = ctrl_inw(FPGA_FPDATER);
+
+	printk(KERN_INFO "Renesas Technology Europe SDK7780 support.\n");
+	printk(KERN_INFO "Board version: %d (revision %d), "
+			 "FPGA version: %d (revision %d), datestamp : %d\n",
+			 (ver >> 12) & 0xf, (ver >> 8) & 0xf,
+			 (ver >>  4) & 0xf, ver & 0xf,
+			 dateStamp);
+
+	/* Setup pin mux'ing for PCIC */
+	ctrl_outw(0x0000, GPIO_PECR);
+}
+
+/*
+ * The Machine Vector
+ */
+static struct sh_machine_vector mv_se7780 __initmv = {
+	.mv_name        = "Renesas SDK7780-R3" ,
+	.mv_setup		= sdk7780_setup,
+	.mv_nr_irqs		= 111,
+	.mv_init_irq	= init_sdk7780_IRQ,
+};
+
diff --git a/arch/sh/configs/sdk7780_defconfig b/arch/sh/configs/sdk7780_defconfig
new file mode 100644
index 000000000000..bb9bcd6591ab
--- /dev/null
+++ b/arch/sh/configs/sdk7780_defconfig
@@ -0,0 +1,1394 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.24-rc7
+# Tue Jan 22 11:34:03 2008
+#
+CONFIG_SUPERH=y
+CONFIG_SUPERH32=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_BUG=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+CONFIG_SYS_SUPPORTS_PCI=y
+CONFIG_STACKTRACE_SUPPORT=y
+CONFIG_LOCKDEP_SUPPORT=y
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_ARCH_NO_VIRT_TO_BUS=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION="_SDK7780"
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+CONFIG_SYSVIPC_SYSCTL=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_BSD_PROCESS_ACCT=y
+# CONFIG_BSD_PROCESS_ACCT_V3 is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
+# CONFIG_AUDIT is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=18
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_RELAY=y
+# CONFIG_BLK_DEV_INITRD is not set
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+CONFIG_UID16=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS=y
+CONFIG_KALLSYMS_ALL=y
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_ANON_INODES=y
+CONFIG_EPOLL=y
+CONFIG_SIGNALFD=y
+CONFIG_EVENTFD=y
+CONFIG_SHMEM=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_SLUB_DEBUG=y
+# CONFIG_SLAB is not set
+CONFIG_SLUB=y
+# CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
+CONFIG_RT_MUTEXES=y
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+CONFIG_BLOCK=y
+CONFIG_LBD=y
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+CONFIG_DEFAULT_AS=y
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="anticipatory"
+
+#
+# System type
+#
+CONFIG_CPU_SH4=y
+CONFIG_CPU_SH4A=y
+# CONFIG_CPU_SUBTYPE_SH7619 is not set
+# CONFIG_CPU_SUBTYPE_SH7203 is not set
+# CONFIG_CPU_SUBTYPE_SH7206 is not set
+# CONFIG_CPU_SUBTYPE_SH7263 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7706 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+# CONFIG_CPU_SUBTYPE_SH7710 is not set
+# CONFIG_CPU_SUBTYPE_SH7712 is not set
+# CONFIG_CPU_SUBTYPE_SH7720 is not set
+# CONFIG_CPU_SUBTYPE_SH7721 is not set
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+# CONFIG_CPU_SUBTYPE_SH7763 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+CONFIG_CPU_SUBTYPE_SH7780=y
+# CONFIG_CPU_SUBTYPE_SH7785 is not set
+# CONFIG_CPU_SUBTYPE_SHX3 is not set
+# CONFIG_CPU_SUBTYPE_SH7343 is not set
+# CONFIG_CPU_SUBTYPE_SH7722 is not set
+# CONFIG_CPU_SUBTYPE_SH5_101 is not set
+# CONFIG_CPU_SUBTYPE_SH5_103 is not set
+
+#
+# Memory management options
+#
+CONFIG_QUICKLIST=y
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x08000000
+CONFIG_MEMORY_SIZE=0x08000000
+CONFIG_29BIT=y
+# CONFIG_PMB is not set
+CONFIG_VSYSCALL=y
+CONFIG_ARCH_FLATMEM_ENABLE=y
+CONFIG_ARCH_SPARSEMEM_ENABLE=y
+CONFIG_ARCH_SPARSEMEM_DEFAULT=y
+CONFIG_MAX_ACTIVE_REGIONS=1
+CONFIG_ARCH_POPULATES_NODE_MAP=y
+CONFIG_ARCH_SELECT_MEMORY_MODEL=y
+CONFIG_PAGE_SIZE_4KB=y
+# CONFIG_PAGE_SIZE_8KB is not set
+# CONFIG_PAGE_SIZE_64KB is not set
+CONFIG_HUGETLB_PAGE_SIZE_64K=y
+# CONFIG_HUGETLB_PAGE_SIZE_256K is not set
+# CONFIG_HUGETLB_PAGE_SIZE_1MB is not set
+# CONFIG_HUGETLB_PAGE_SIZE_4MB is not set
+# CONFIG_HUGETLB_PAGE_SIZE_64MB is not set
+# CONFIG_HUGETLB_PAGE_SIZE_512MB is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+CONFIG_SPARSEMEM_STATIC=y
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+CONFIG_RESOURCES_64BIT=y
+CONFIG_ZONE_DMA_FLAG=0
+CONFIG_NR_QUICK=2
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+CONFIG_CACHE_WRITEBACK=y
+# CONFIG_CACHE_WRITETHROUGH is not set
+# CONFIG_CACHE_OFF is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_CPU_BIG_ENDIAN is not set
+CONFIG_SH_FPU=y
+CONFIG_SH_STORE_QUEUES=y
+# CONFIG_SPECULATIVE_EXECUTION is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+CONFIG_CPU_HAS_FPU=y
+
+#
+# Board support
+#
+# CONFIG_SH_7780_SOLUTION_ENGINE is not set
+CONFIG_SH_SDK7780=y
+# CONFIG_SH_HIGHLANDER is not set
+# CONFIG_SH_SDK7780_STANDALONE is not set
+CONFIG_SH_SDK7780_BASE=y
+
+#
+# Timer and clock configuration
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_TIMER_IRQ=28
+CONFIG_SH_PCLK_FREQ=33333333
+CONFIG_TICK_ONESHOT=y
+# CONFIG_NO_HZ is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA_API=y
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=12
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+
+#
+# Additional SuperH Device Drivers
+#
+CONFIG_HEARTBEAT=y
+# CONFIG_PUSH_SWITCH is not set
+
+#
+# Kernel features
+#
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_300 is not set
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+# CONFIG_KEXEC is not set
+# CONFIG_CRASH_DUMP is not set
+# CONFIG_PREEMPT_NONE is not set
+# CONFIG_PREEMPT_VOLUNTARY is not set
+CONFIG_PREEMPT=y
+CONFIG_PREEMPT_BKL=y
+CONFIG_GUSA=y
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x01800000
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="mem=128M console=tty0 console=ttySC0,115200 ip=bootp root=/dev/nfs nfsroot=192.168.0.1:/home/rootfs"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+# CONFIG_PCI_LEGACY is not set
+CONFIG_PCI_DEBUG=y
+CONFIG_PCCARD=y
+# CONFIG_PCMCIA_DEBUG is not set
+CONFIG_PCMCIA=y
+CONFIG_PCMCIA_LOAD_CIS=y
+CONFIG_PCMCIA_IOCTL=y
+CONFIG_CARDBUS=y
+
+#
+# PC-card bridges
+#
+CONFIG_YENTA=y
+CONFIG_YENTA_O2=y
+CONFIG_YENTA_RICOH=y
+CONFIG_YENTA_TI=y
+CONFIG_YENTA_ENE_TUNE=y
+CONFIG_YENTA_TOSHIBA=y
+# CONFIG_PD6729 is not set
+# CONFIG_I82092 is not set
+CONFIG_PCCARD_NONSTATIC=y
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_HOTPLUG_PCI_FAKE is not set
+# CONFIG_HOTPLUG_PCI_CPCI is not set
+# CONFIG_HOTPLUG_PCI_SHPC is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+# CONFIG_XFRM_USER is not set
+# CONFIG_XFRM_SUB_POLICY is not set
+# CONFIG_XFRM_MIGRATE is not set
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_MULTIPLE_TABLES is not set
+# CONFIG_IP_ROUTE_MULTIPATH is not set
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+CONFIG_IP_PNP_BOOTP=y
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_IP_MROUTE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+CONFIG_INET_TUNNEL=y
+CONFIG_INET_XFRM_MODE_TRANSPORT=y
+CONFIG_INET_XFRM_MODE_TUNNEL=y
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+CONFIG_IPV6=y
+# CONFIG_IPV6_PRIVACY is not set
+# CONFIG_IPV6_ROUTER_PREF is not set
+# CONFIG_IPV6_OPTIMISTIC_DAD is not set
+# CONFIG_INET6_AH is not set
+# CONFIG_INET6_ESP is not set
+# CONFIG_INET6_IPCOMP is not set
+# CONFIG_IPV6_MIP6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+CONFIG_INET6_XFRM_MODE_TRANSPORT=y
+CONFIG_INET6_XFRM_MODE_TUNNEL=y
+# CONFIG_INET6_XFRM_MODE_BEET is not set
+# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set
+CONFIG_IPV6_SIT=y
+# CONFIG_IPV6_TUNNEL is not set
+# CONFIG_IPV6_MULTIPLE_TABLES is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+CONFIG_NET_SCHED=y
+
+#
+# Queueing/Scheduling
+#
+# CONFIG_NET_SCH_CBQ is not set
+# CONFIG_NET_SCH_HTB is not set
+# CONFIG_NET_SCH_HFSC is not set
+# CONFIG_NET_SCH_PRIO is not set
+# CONFIG_NET_SCH_RR is not set
+# CONFIG_NET_SCH_RED is not set
+# CONFIG_NET_SCH_SFQ is not set
+# CONFIG_NET_SCH_TEQL is not set
+# CONFIG_NET_SCH_TBF is not set
+# CONFIG_NET_SCH_GRED is not set
+# CONFIG_NET_SCH_DSMARK is not set
+# CONFIG_NET_SCH_NETEM is not set
+# CONFIG_NET_SCH_INGRESS is not set
+
+#
+# Classification
+#
+# CONFIG_NET_CLS_BASIC is not set
+# CONFIG_NET_CLS_TCINDEX is not set
+# CONFIG_NET_CLS_ROUTE4 is not set
+# CONFIG_NET_CLS_FW is not set
+# CONFIG_NET_CLS_U32 is not set
+# CONFIG_NET_CLS_RSVP is not set
+# CONFIG_NET_CLS_RSVP6 is not set
+# CONFIG_NET_EMATCH is not set
+# CONFIG_NET_CLS_ACT is not set
+# CONFIG_NET_CLS_POLICE is not set
+CONFIG_NET_SCH_FIFO=y
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_DEBUG_DEVRES is not set
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+# CONFIG_MTD is not set
+CONFIG_PARPORT=y
+# CONFIG_PARPORT_PC is not set
+# CONFIG_PARPORT_GSC is not set
+# CONFIG_PARPORT_AX88796 is not set
+# CONFIG_PARPORT_1284 is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+CONFIG_IDE=y
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=y
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+# CONFIG_BLK_DEV_IDE_SATA is not set
+CONFIG_BLK_DEV_IDEDISK=y
+CONFIG_IDEDISK_MULTI_MODE=y
+# CONFIG_BLK_DEV_IDECS is not set
+# CONFIG_BLK_DEV_DELKIN is not set
+CONFIG_BLK_DEV_IDECD=y
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+# CONFIG_BLK_DEV_IDESCSI is not set
+# CONFIG_IDE_TASK_IOCTL is not set
+CONFIG_IDE_PROC_FS=y
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=y
+CONFIG_BLK_DEV_PLATFORM=y
+
+#
+# PCI IDE chipsets support
+#
+CONFIG_BLK_DEV_IDEPCI=y
+# CONFIG_IDEPCI_SHARE_IRQ is not set
+CONFIG_IDEPCI_PCIBUS_ORDER=y
+# CONFIG_BLK_DEV_OFFBOARD is not set
+CONFIG_BLK_DEV_GENERIC=y
+# CONFIG_BLK_DEV_OPTI621 is not set
+# CONFIG_BLK_DEV_AEC62XX is not set
+# CONFIG_BLK_DEV_ALI15X3 is not set
+# CONFIG_BLK_DEV_AMD74XX is not set
+# CONFIG_BLK_DEV_CMD64X is not set
+# CONFIG_BLK_DEV_TRIFLEX is not set
+# CONFIG_BLK_DEV_CY82C693 is not set
+# CONFIG_BLK_DEV_CS5520 is not set
+# CONFIG_BLK_DEV_CS5530 is not set
+# CONFIG_BLK_DEV_HPT34X is not set
+# CONFIG_BLK_DEV_HPT366 is not set
+# CONFIG_BLK_DEV_JMICRON is not set
+# CONFIG_BLK_DEV_SC1200 is not set
+# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT8213 is not set
+# CONFIG_BLK_DEV_IT821X is not set
+# CONFIG_BLK_DEV_NS87415 is not set
+# CONFIG_BLK_DEV_PDC202XX_OLD is not set
+# CONFIG_BLK_DEV_PDC202XX_NEW is not set
+# CONFIG_BLK_DEV_SVWKS is not set
+# CONFIG_BLK_DEV_SIIMAGE is not set
+# CONFIG_BLK_DEV_SLC90E66 is not set
+# CONFIG_BLK_DEV_TRM290 is not set
+# CONFIG_BLK_DEV_VIA82CXXX is not set
+# CONFIG_BLK_DEV_TC86C001 is not set
+# CONFIG_IDE_ARM is not set
+# CONFIG_BLK_DEV_IDEDMA is not set
+# CONFIG_IDE_ARCH_OBSOLETE_INIT is not set
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=y
+CONFIG_SCSI_DMA=y
+# CONFIG_SCSI_TGT is not set
+CONFIG_SCSI_NETLINK=y
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=y
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+CONFIG_BLK_DEV_SR=y
+# CONFIG_BLK_DEV_SR_VENDOR is not set
+CONFIG_CHR_DEV_SG=y
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+# CONFIG_SCSI_SCAN_ASYNC is not set
+CONFIG_SCSI_WAIT_SCAN=m
+
+#
+# SCSI Transports
+#
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_FC_ATTRS=y
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_LIBSAS is not set
+# CONFIG_SCSI_SRP_ATTRS is not set
+CONFIG_SCSI_LOWLEVEL=y
+# CONFIG_ISCSI_TCP is not set
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_AIC94XX is not set
+# CONFIG_SCSI_ARCMSR is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_HPTIOP is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_STEX is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+# CONFIG_SCSI_QLA_FC is not set
+# CONFIG_SCSI_QLA_ISCSI is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+# CONFIG_SCSI_SRP is not set
+# CONFIG_SCSI_LOWLEVEL_PCMCIA is not set
+CONFIG_ATA=y
+# CONFIG_ATA_NONSTANDARD is not set
+# CONFIG_SATA_AHCI is not set
+# CONFIG_SATA_SVW is not set
+# CONFIG_ATA_PIIX is not set
+# CONFIG_SATA_MV is not set
+# CONFIG_SATA_NV is not set
+# CONFIG_PDC_ADMA is not set
+# CONFIG_SATA_QSTOR is not set
+# CONFIG_SATA_PROMISE is not set
+# CONFIG_SATA_SX4 is not set
+# CONFIG_SATA_SIL is not set
+# CONFIG_SATA_SIL24 is not set
+# CONFIG_SATA_SIS is not set
+# CONFIG_SATA_ULI is not set
+# CONFIG_SATA_VIA is not set
+# CONFIG_SATA_VITESSE is not set
+# CONFIG_SATA_INIC162X is not set
+# CONFIG_PATA_ALI is not set
+# CONFIG_PATA_AMD is not set
+# CONFIG_PATA_ARTOP is not set
+# CONFIG_PATA_ATIIXP is not set
+# CONFIG_PATA_CMD640_PCI is not set
+# CONFIG_PATA_CMD64X is not set
+# CONFIG_PATA_CS5520 is not set
+# CONFIG_PATA_CS5530 is not set
+# CONFIG_PATA_CYPRESS is not set
+# CONFIG_PATA_EFAR is not set
+# CONFIG_ATA_GENERIC is not set
+# CONFIG_PATA_HPT366 is not set
+# CONFIG_PATA_HPT37X is not set
+# CONFIG_PATA_HPT3X2N is not set
+# CONFIG_PATA_HPT3X3 is not set
+# CONFIG_PATA_IT821X is not set
+# CONFIG_PATA_IT8213 is not set
+# CONFIG_PATA_JMICRON is not set
+# CONFIG_PATA_TRIFLEX is not set
+# CONFIG_PATA_MARVELL is not set
+# CONFIG_PATA_MPIIX is not set
+# CONFIG_PATA_OLDPIIX is not set
+# CONFIG_PATA_NETCELL is not set
+# CONFIG_PATA_NS87410 is not set
+# CONFIG_PATA_NS87415 is not set
+# CONFIG_PATA_OPTI is not set
+# CONFIG_PATA_OPTIDMA is not set
+# CONFIG_PATA_PCMCIA is not set
+# CONFIG_PATA_PDC_OLD is not set
+# CONFIG_PATA_RADISYS is not set
+# CONFIG_PATA_RZ1000 is not set
+# CONFIG_PATA_SC1200 is not set
+# CONFIG_PATA_SERVERWORKS is not set
+# CONFIG_PATA_PDC2027X is not set
+# CONFIG_PATA_SIL680 is not set
+# CONFIG_PATA_SIS is not set
+# CONFIG_PATA_VIA is not set
+# CONFIG_PATA_WINBOND is not set
+# CONFIG_PATA_PLATFORM is not set
+CONFIG_MD=y
+# CONFIG_BLK_DEV_MD is not set
+CONFIG_BLK_DEV_DM=y
+# CONFIG_DM_DEBUG is not set
+# CONFIG_DM_CRYPT is not set
+# CONFIG_DM_SNAPSHOT is not set
+# CONFIG_DM_MIRROR is not set
+# CONFIG_DM_ZERO is not set
+# CONFIG_DM_MULTIPATH is not set
+# CONFIG_DM_DELAY is not set
+# CONFIG_DM_UEVENT is not set
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_FIREWIRE is not set
+# CONFIG_IEEE1394 is not set
+# CONFIG_I2O is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_ARCNET is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_AX88796 is not set
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+CONFIG_SMC91X=y
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_NET_PCI is not set
+# CONFIG_B44 is not set
+# CONFIG_NET_POCKET is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+# CONFIG_TR is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_NET_PCMCIA is not set
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PLIP is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+CONFIG_NETCONSOLE=y
+# CONFIG_NETCONSOLE_DYNAMIC is not set
+CONFIG_NETPOLL=y
+# CONFIG_NETPOLL_TRAP is not set
+CONFIG_NET_POLL_CONTROLLER=y
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+# CONFIG_INPUT_FF_MEMLESS is not set
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+CONFIG_INPUT_MOUSEDEV_PSAUX=y
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+CONFIG_INPUT_EVDEV=y
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+CONFIG_INPUT_KEYBOARD=y
+CONFIG_KEYBOARD_ATKBD=y
+# CONFIG_KEYBOARD_SUNKBD is not set
+# CONFIG_KEYBOARD_LKKBD is not set
+# CONFIG_KEYBOARD_XTKBD is not set
+# CONFIG_KEYBOARD_NEWTON is not set
+# CONFIG_KEYBOARD_STOWAWAY is not set
+CONFIG_INPUT_MOUSE=y
+CONFIG_MOUSE_PS2=y
+CONFIG_MOUSE_PS2_ALPS=y
+CONFIG_MOUSE_PS2_LOGIPS2PP=y
+CONFIG_MOUSE_PS2_SYNAPTICS=y
+CONFIG_MOUSE_PS2_LIFEBOOK=y
+CONFIG_MOUSE_PS2_TRACKPOINT=y
+# CONFIG_MOUSE_PS2_TOUCHKIT is not set
+# CONFIG_MOUSE_SERIAL is not set
+# CONFIG_MOUSE_APPLETOUCH is not set
+# CONFIG_MOUSE_VSXXXAA is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PARKBD is not set
+# CONFIG_SERIO_PCIPS2 is not set
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_VT_HW_CONSOLE_BINDING is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_PRINTER is not set
+# CONFIG_PPDEV is not set
+# CONFIG_IPMI_HANDLER is not set
+CONFIG_HW_RANDOM=y
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# PCMCIA character devices
+#
+# CONFIG_SYNCLINK_CS is not set
+# CONFIG_CARDMAN_4000 is not set
+# CONFIG_CARDMAN_4040 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+CONFIG_DEVPORT=y
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+# CONFIG_W1 is not set
+CONFIG_POWER_SUPPLY=y
+# CONFIG_POWER_SUPPLY_DEBUG is not set
+# CONFIG_PDA_POWER is not set
+# CONFIG_BATTERY_DS2760 is not set
+# CONFIG_HWMON is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+CONFIG_SSB=y
+CONFIG_SSB_PCIHOST_POSSIBLE=y
+CONFIG_SSB_PCIHOST=y
+CONFIG_SSB_PCMCIAHOST_POSSIBLE=y
+# CONFIG_SSB_PCMCIAHOST is not set
+# CONFIG_SSB_SILENT is not set
+# CONFIG_SSB_DEBUG is not set
+CONFIG_SSB_DRIVER_PCICORE_POSSIBLE=y
+CONFIG_SSB_DRIVER_PCICORE=y
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+# CONFIG_DAB is not set
+
+#
+# Graphics support
+#
+# CONFIG_DRM is not set
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+CONFIG_FB=y
+# CONFIG_FIRMWARE_EDID is not set
+# CONFIG_FB_DDC is not set
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
+# CONFIG_FB_SYS_FILLRECT is not set
+# CONFIG_FB_SYS_COPYAREA is not set
+# CONFIG_FB_SYS_IMAGEBLIT is not set
+# CONFIG_FB_SYS_FOPS is not set
+CONFIG_FB_DEFERRED_IO=y
+# CONFIG_FB_SVGALIB is not set
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+
+#
+# Frame buffer hardware drivers
+#
+# CONFIG_FB_CIRRUS is not set
+# CONFIG_FB_PM2 is not set
+# CONFIG_FB_CYBER2000 is not set
+# CONFIG_FB_ASILIANT is not set
+# CONFIG_FB_IMSTT is not set
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_NVIDIA is not set
+# CONFIG_FB_RIVA is not set
+# CONFIG_FB_MATROX is not set
+# CONFIG_FB_RADEON is not set
+# CONFIG_FB_ATY128 is not set
+# CONFIG_FB_ATY is not set
+# CONFIG_FB_S3 is not set
+# CONFIG_FB_SAVAGE is not set
+# CONFIG_FB_SIS is not set
+# CONFIG_FB_NEOMAGIC is not set
+# CONFIG_FB_KYRO is not set
+# CONFIG_FB_3DFX is not set
+# CONFIG_FB_VOODOO1 is not set
+# CONFIG_FB_VT8623 is not set
+# CONFIG_FB_TRIDENT is not set
+# CONFIG_FB_ARK is not set
+# CONFIG_FB_PM3 is not set
+# CONFIG_FB_VIRTUAL is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+CONFIG_DISPLAY_SUPPORT=y
+
+#
+# Display hardware drivers
+#
+
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
+# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set
+# CONFIG_FONTS is not set
+CONFIG_FONT_8x8=y
+CONFIG_FONT_8x16=y
+CONFIG_LOGO=y
+CONFIG_LOGO_LINUX_MONO=y
+CONFIG_LOGO_LINUX_VGA16=y
+CONFIG_LOGO_LINUX_CLUT224=y
+CONFIG_LOGO_SUPERH_MONO=y
+CONFIG_LOGO_SUPERH_VGA16=y
+CONFIG_LOGO_SUPERH_CLUT224=y
+
+#
+# Sound
+#
+CONFIG_SOUND=y
+
+#
+# Advanced Linux Sound Architecture
+#
+# CONFIG_SND is not set
+
+#
+# Open Sound System
+#
+CONFIG_SOUND_PRIME=y
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+CONFIG_HID_SUPPORT=y
+CONFIG_HID=y
+# CONFIG_HID_DEBUG is not set
+# CONFIG_HIDRAW is not set
+
+#
+# USB Input Devices
+#
+CONFIG_USB_HID=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
+# CONFIG_HID_FF is not set
+# CONFIG_USB_HIDDEV is not set
+CONFIG_USB_SUPPORT=y
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB_ARCH_HAS_EHCI=y
+CONFIG_USB=y
+CONFIG_USB_DEBUG=y
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_DEVICE_CLASS is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_EHCI_TT_NEWSCHED is not set
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_OHCI_HCD is not set
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+# CONFIG_USB_R8A66597_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_ACM is not set
+CONFIG_USB_PRINTER=y
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=y
+# CONFIG_USB_STORAGE_DEBUG is not set
+# CONFIG_USB_STORAGE_DATAFAB is not set
+# CONFIG_USB_STORAGE_FREECOM is not set
+# CONFIG_USB_STORAGE_ISD200 is not set
+# CONFIG_USB_STORAGE_DPCM is not set
+# CONFIG_USB_STORAGE_USBAT is not set
+# CONFIG_USB_STORAGE_SDDR09 is not set
+# CONFIG_USB_STORAGE_SDDR55 is not set
+# CONFIG_USB_STORAGE_JUMPSHOT is not set
+# CONFIG_USB_STORAGE_ALAUDA is not set
+# CONFIG_USB_STORAGE_ONETOUCH is not set
+# CONFIG_USB_STORAGE_KARMA is not set
+# CONFIG_USB_LIBUSUAL is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+# CONFIG_USB_USS720 is not set
+
+#
+# USB Serial Converter support
+#
+# CONFIG_USB_SERIAL is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_ADUTUX is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_BERRY_CHARGE is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGET is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_FTDI_ELAN is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TRANCEVIBRATOR is not set
+# CONFIG_USB_IOWARRIOR is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+# CONFIG_MMC is not set
+CONFIG_NEW_LEDS=y
+CONFIG_LEDS_CLASS=y
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+# CONFIG_LEDS_TRIGGERS is not set
+# CONFIG_INFINIBAND is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_AUXDISPLAY is not set
+
+#
+# Userspace I/O
+#
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT2_FS_POSIX_ACL=y
+# CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+CONFIG_EXT3_FS_POSIX_ACL=y
+# CONFIG_EXT3_FS_SECURITY is not set
+# CONFIG_EXT4DEV_FS is not set
+CONFIG_JBD=y
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+# CONFIG_GFS2_FS is not set
+# CONFIG_OCFS2_FS is not set
+CONFIG_MINIX_FS=y
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+CONFIG_AUTOFS4_FS=y
+# CONFIG_FUSE_FS is not set
+CONFIG_GENERIC_ACL=y
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=y
+# CONFIG_JOLIET is not set
+# CONFIG_ZISOFS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=y
+CONFIG_NTFS_DEBUG=y
+CONFIG_NTFS_RW=y
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+CONFIG_HUGETLBFS=y
+CONFIG_HUGETLB_PAGE=y
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_SUNRPC_BIND34 is not set
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+CONFIG_NLS_ASCII=y
+CONFIG_NLS_ISO8859_1=y
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+CONFIG_NLS_ISO8859_15=y
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=y
+# CONFIG_DLM is not set
+# CONFIG_INSTRUMENTATION is not set
+
+#
+# Kernel hacking
+#
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+# CONFIG_ENABLE_MUST_CHECK is not set
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_UNUSED_SYMBOLS=y
+# CONFIG_DEBUG_FS is not set
+# CONFIG_HEADERS_CHECK is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_SHIRQ is not set
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHED_DEBUG is not set
+# CONFIG_SCHEDSTATS is not set
+CONFIG_TIMER_STATS=y
+# CONFIG_SLUB_DEBUG_ON is not set
+CONFIG_DEBUG_PREEMPT=y
+# CONFIG_DEBUG_RT_MUTEXES is not set
+# CONFIG_RT_MUTEX_TESTER is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_LOCK_ALLOC is not set
+# CONFIG_PROVE_LOCKING is not set
+# CONFIG_LOCK_STAT is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+CONFIG_DEBUG_BUGVERBOSE=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_DEBUG_VM is not set
+# CONFIG_DEBUG_LIST is not set
+# CONFIG_DEBUG_SG is not set
+# CONFIG_FRAME_POINTER is not set
+# CONFIG_FORCED_INLINING is not set
+# CONFIG_BOOT_PRINTK_DELAY is not set
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_FAULT_INJECTION is not set
+# CONFIG_SAMPLES is not set
+CONFIG_SH_STANDARD_BIOS=y
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_EARLY_PRINTK is not set
+# CONFIG_DEBUG_BOOTMEM is not set
+CONFIG_DEBUG_STACKOVERFLOW=y
+# CONFIG_DEBUG_STACK_USAGE is not set
+# CONFIG_4KSTACKS is not set
+# CONFIG_IRQSTACKS is not set
+# CONFIG_SH_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_ALGAPI=y
+# CONFIG_CRYPTO_MANAGER is not set
+# CONFIG_CRYPTO_HMAC is not set
+# CONFIG_CRYPTO_XCBC is not set
+# CONFIG_CRYPTO_NULL is not set
+# CONFIG_CRYPTO_MD4 is not set
+CONFIG_CRYPTO_MD5=y
+# CONFIG_CRYPTO_SHA1 is not set
+# CONFIG_CRYPTO_SHA256 is not set
+# CONFIG_CRYPTO_SHA512 is not set
+# CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
+# CONFIG_CRYPTO_GF128MUL is not set
+# CONFIG_CRYPTO_ECB is not set
+# CONFIG_CRYPTO_CBC is not set
+# CONFIG_CRYPTO_PCBC is not set
+# CONFIG_CRYPTO_LRW is not set
+# CONFIG_CRYPTO_XTS is not set
+# CONFIG_CRYPTO_CRYPTD is not set
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_FCRYPT is not set
+# CONFIG_CRYPTO_BLOWFISH is not set
+# CONFIG_CRYPTO_TWOFISH is not set
+# CONFIG_CRYPTO_SERPENT is not set
+# CONFIG_CRYPTO_AES is not set
+# CONFIG_CRYPTO_CAST5 is not set
+# CONFIG_CRYPTO_CAST6 is not set
+# CONFIG_CRYPTO_TEA is not set
+# CONFIG_CRYPTO_ARC4 is not set
+# CONFIG_CRYPTO_KHAZAD is not set
+# CONFIG_CRYPTO_ANUBIS is not set
+# CONFIG_CRYPTO_SEED is not set
+# CONFIG_CRYPTO_DEFLATE is not set
+# CONFIG_CRYPTO_MICHAEL_MIC is not set
+# CONFIG_CRYPTO_CRC32C is not set
+# CONFIG_CRYPTO_CAMELLIA is not set
+# CONFIG_CRYPTO_TEST is not set
+# CONFIG_CRYPTO_AUTHENC is not set
+CONFIG_CRYPTO_HW=y
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_PLIST=y
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_IOPORT=y
+CONFIG_HAS_DMA=y
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 1086cf111f8b..7bf2a2c823f3 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -18,6 +18,7 @@ obj-$(CONFIG_SH_SECUREEDGE5410)		+= ops-snapgear.o
 obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
 obj-$(CONFIG_SH_HIGHLANDER)		+= ops-r7780rp.o fixups-r7780rp.o
+obj-$(CONFIG_SH_SDK7780)		+= ops-sdk7780.o fixups-sdk7780.o
 obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
 obj-$(CONFIG_SH_LANDISK)		+= ops-landisk.o
 obj-$(CONFIG_SH_LBOX_RE2)		+= ops-lboxre2.o fixups-lboxre2.o
diff --git a/arch/sh/drivers/pci/fixups-sdk7780.c b/arch/sh/drivers/pci/fixups-sdk7780.c
new file mode 100644
index 000000000000..2f8863099dd1
--- /dev/null
+++ b/arch/sh/drivers/pci/fixups-sdk7780.c
@@ -0,0 +1,59 @@
+/*
+ * arch/sh/drivers/pci/fixups-sdk7780.c
+ *
+ * PCI fixups for the SDK7780SE03
+ *
+ * Copyright (C) 2003  Lineo uSolutions, Inc.
+ * Copyright (C) 2004 - 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/pci.h>
+#include "pci-sh4.h"
+#include <asm/io.h>
+
+int pci_fixup_pcic(void)
+{
+	ctrl_outl(0x00000001, SH7780_PCI_VCR2);
+
+	/* Enable all interrupts, so we know what to fix */
+	pci_write_reg(0x0000C3FF, SH7780_PCIIMR);
+	pci_write_reg(0x0000380F, SH7780_PCIAINTM);
+
+	/* Set up standard PCI config registers */
+	pci_write_reg(0xFB00, SH7780_PCISTATUS);
+	pci_write_reg(0x0047, SH7780_PCICMD);
+	pci_write_reg(0x00, SH7780_PCIPIF);
+	pci_write_reg(0x00, SH7780_PCISUB);
+	pci_write_reg(0x06, SH7780_PCIBCC);
+	pci_write_reg(0x1912, SH7780_PCISVID);
+	pci_write_reg(0x0001, SH7780_PCISID);
+
+	pci_write_reg(0x08000000, SH7780_PCIMBAR0);	/* PCI */
+	pci_write_reg(0x08000000, SH7780_PCILAR0);	/* SHwy */
+	pci_write_reg(0x07F00001, SH7780_PCILSR);	/* size 128M w/ MBAR */
+
+	pci_write_reg(0x00000000, SH7780_PCIMBAR1);
+	pci_write_reg(0x00000000, SH7780_PCILAR1);
+	pci_write_reg(0x00000000, SH7780_PCILSR1);
+
+	pci_write_reg(0xAB000801, SH7780_PCIIBAR);
+
+	/*
+	 * Set the MBR so PCI address is one-to-one with window,
+	 * meaning all calls go straight through... use ifdef to
+	 * catch erroneous assumption.
+	 */
+	pci_write_reg(0xFD000000 , SH7780_PCIMBR0);
+	pci_write_reg(0x00FC0000 , SH7780_PCIMBMR0);	/* 16M */
+
+	/* Set IOBR for window containing area specified in pci.h */
+	pci_write_reg(PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE-1), SH7780_PCIIOBR);
+	pci_write_reg((SH7780_PCI_IO_SIZE-1) & (7 << 18), SH7780_PCIIOBMR);
+
+	pci_write_reg(0xA5000C01, SH7780_PCICR);
+
+	return 0;
+}
diff --git a/arch/sh/drivers/pci/ops-sdk7780.c b/arch/sh/drivers/pci/ops-sdk7780.c
new file mode 100644
index 000000000000..66a9b4047f26
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-sdk7780.c
@@ -0,0 +1,73 @@
+/*
+ * linux/arch/sh/drivers/pci/ops-sdk7780.c
+ *
+ * Copyright (C) 2006  Nobuhiro Iwamatsu
+ *
+ * PCI initialization for the SDK7780SE03
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <asm/sdk7780.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
+
+/* IDSEL [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] */
+static char sdk7780_irq_tab[4][16] __initdata = {
+	/* INTA */
+	{ 65, 68, 67, 68, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 },
+	/* INTB */
+	{ 66, 65, -1, 65, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 },
+	/* INTC */
+	{ 67, 66, -1, 66, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 },
+	/* INTD */
+	{ 68, 67, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 },
+};
+
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
+{
+       return sdk7780_irq_tab[pin-1][slot];
+}
+
+static struct resource sdk7780_io_resource = {
+	.name	= "SH7780_IO",
+	.start	= SH7780_PCI_IO_BASE,
+	.end	= SH7780_PCI_IO_BASE + SH7780_PCI_IO_SIZE - 1,
+	.flags	= IORESOURCE_IO
+};
+
+static struct resource sdk7780_mem_resource = {
+	.name	= "SH7780_mem",
+	.start	= SH7780_PCI_MEMORY_BASE,
+	.end	= SH7780_PCI_MEMORY_BASE + SH7780_PCI_MEM_SIZE - 1,
+	.flags	= IORESOURCE_MEM
+};
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh4_pci_ops, &sdk7780_io_resource, &sdk7780_mem_resource, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+static struct sh4_pci_address_map sdk7780_pci_map = {
+	.window0	= {
+		.base	= SH7780_CS2_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+	.window1	= {
+		.base	= SH7780_CS3_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+	.flags	= SH4_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	printk(KERN_INFO "SH7780 PCI: Finished initializing PCI controller\n");
+	return sh7780_pcic_init(&sdk7780_pci_map);
+}
diff --git a/include/asm-sh/sdk7780.h b/include/asm-sh/sdk7780.h
new file mode 100644
index 000000000000..697dc865f21b
--- /dev/null
+++ b/include/asm-sh/sdk7780.h
@@ -0,0 +1,81 @@
+#ifndef __ASM_SH_RENESAS_SDK7780_H
+#define __ASM_SH_RENESAS_SDK7780_H
+
+/*
+ * linux/include/asm-sh/sdk7780.h
+ *
+ * Renesas Solutions SH7780 SDK Support
+ * Copyright (C) 2008 Nicholas Beck <nbeck@mpc-data.co.uk>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <asm/addrspace.h>
+
+/* Box specific addresses.  */
+#define SE_AREA0_WIDTH	4		/* Area0: 32bit */
+#define PA_ROM			0xa0000000	/* EPROM */
+#define PA_ROM_SIZE		0x00400000	/* EPROM size 4M byte */
+#define PA_FROM			0xa0800000	/* Flash-ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash-ROM size 4M byte */
+#define PA_EXT1			0xa4000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_SDRAM		0xa8000000	/* DDR-SDRAM(Area2/3) 128MB */
+#define PA_SDRAM_SIZE	0x08000000
+
+#define PA_EXT4			0xb0000000
+#define PA_EXT4_SIZE	0x04000000
+#define PA_EXT_USER		PA_EXT4		/* User Expansion Space */
+
+#define PA_PERIPHERAL	PA_AREA5_IO
+
+/* SRAM/Reserved */
+#define PA_RESERVED	(PA_PERIPHERAL + 0)
+/* FPGA base address */
+#define PA_FPGA		(PA_PERIPHERAL + 0x01000000)
+/* SMC LAN91C111 */
+#define PA_LAN		(PA_PERIPHERAL + 0x01800000)
+
+
+#define FPGA_SRSTR      (PA_FPGA + 0x000)	/* System reset */
+#define FPGA_IRQ0SR     (PA_FPGA + 0x010)	/* IRQ0 status */
+#define FPGA_IRQ0MR     (PA_FPGA + 0x020)	/* IRQ0 mask */
+#define FPGA_BDMR       (PA_FPGA + 0x030)	/* Board operating mode */
+#define FPGA_INTT0PRTR  (PA_FPGA + 0x040)	/* Interrupt test mode0 port */
+#define FPGA_INTT0SELR  (PA_FPGA + 0x050)	/* Int. test mode0 select */
+#define FPGA_INTT1POLR  (PA_FPGA + 0x060)	/* Int. test mode0 polarity */
+#define FPGA_NMIR       (PA_FPGA + 0x070)	/* NMI source */
+#define FPGA_NMIMR      (PA_FPGA + 0x080)	/* NMI mask */
+#define FPGA_IRQR       (PA_FPGA + 0x090)	/* IRQX source */
+#define FPGA_IRQMR      (PA_FPGA + 0x0A0)	/* IRQX mask */
+#define FPGA_SLEDR      (PA_FPGA + 0x0B0)	/* LED control */
+#define PA_LED			FPGA_SLEDR
+#define FPGA_MAPSWR     (PA_FPGA + 0x0C0)	/* Map switch */
+#define FPGA_FPVERR     (PA_FPGA + 0x0D0)	/* FPGA version */
+#define FPGA_FPDATER    (PA_FPGA + 0x0E0)	/* FPGA date */
+#define FPGA_RSE        (PA_FPGA + 0x100)	/* Reset source */
+#define FPGA_EASR       (PA_FPGA + 0x110)	/* External area select */
+#define FPGA_SPER       (PA_FPGA + 0x120)	/* Serial port enable */
+#define FPGA_IMSR       (PA_FPGA + 0x130)	/* Interrupt mode select */
+#define FPGA_PCIMR      (PA_FPGA + 0x140)	/* PCI Mode */
+#define FPGA_DIPSWMR    (PA_FPGA + 0x150)	/* DIPSW monitor */
+#define FPGA_FPODR      (PA_FPGA + 0x160)	/* Output port data */
+#define FPGA_ATAESR     (PA_FPGA + 0x170)	/* ATA extended bus status */
+#define FPGA_IRQPOLR    (PA_FPGA + 0x180)	/* IRQx polarity */
+
+
+#define SDK7780_NR_IRL			15
+/* IDE/ATA interrupt */
+#define IRQ_CFCARD				14
+/* SMC interrupt */
+#define IRQ_ETHERNET			6
+
+
+/* arch/sh/boards/renesas/sdk7780/irq.c */
+void init_sdk7780_IRQ(void);
+
+#define __IO_PREFIX		sdk7780
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_RENESAS_SDK7780_H */
-- 
cgit v1.2.3


From f93e97eaead5c50af35d73cca7301ebbfdff116c Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Thu, 24 Jan 2008 18:35:10 +0900
Subject: sh: declared coherent memory support V2

This patch adds declared coherent memory support to the sh architecture. All
functions are based on the x86 implementation. Header files are adjusted to
use the new functions instead of the former consistent_alloc() code.

This version includes the few changes what were included in the fix patch
together with modifications based on feedback from Paul.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/consistent.c      | 174 +++++++++++++++++++++++++++++++------------
 include/asm-sh/dma-mapping.h |  56 +++++---------
 2 files changed, 148 insertions(+), 82 deletions(-)

(limited to 'include')

diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c
index 65ad30031ad7..7b2131c9eeda 100644
--- a/arch/sh/mm/consistent.c
+++ b/arch/sh/mm/consistent.c
@@ -3,6 +3,8 @@
  *
  * Copyright (C) 2004 - 2007  Paul Mundt
  *
+ * Declared coherent memory functions based on arch/x86/kernel/pci-dma_32.c
+ *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
@@ -13,66 +15,146 @@
 #include <asm/addrspace.h>
 #include <asm/io.h>
 
-void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
-{
-	struct page *page, *end, *free;
-	void *ret, *vp;
-	int order;
-
-	size = PAGE_ALIGN(size);
-	order = get_order(size);
+struct dma_coherent_mem {
+	void		*virt_base;
+	u32		device_base;
+	int		size;
+	int		flags;
+	unsigned long	*bitmap;
+};
 
-	page = alloc_pages(gfp, order);
-	if (!page)
-		return NULL;
-	split_page(page, order);
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			   dma_addr_t *dma_handle, gfp_t gfp)
+{
+	void *ret;
+	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+	int order = get_order(size);
+
+	if (mem) {
+		int page = bitmap_find_free_region(mem->bitmap, mem->size,
+						     order);
+		if (page >= 0) {
+			*dma_handle = mem->device_base + (page << PAGE_SHIFT);
+			ret = mem->virt_base + (page << PAGE_SHIFT);
+			memset(ret, 0, size);
+			return ret;
+		}
+		if (mem->flags & DMA_MEMORY_EXCLUSIVE)
+			return NULL;
+	}
 
-	ret = page_address(page);
-	*handle = virt_to_phys(ret);
+	ret = (void *)__get_free_pages(gfp, order);
 
-	vp = ioremap_nocache(*handle, size);
-	if (!vp) {
-		free_pages((unsigned long)ret, order);
-		return NULL;
+	if (ret != NULL) {
+		memset(ret, 0, size);
+		/*
+		 * Pages from the page allocator may have data present in
+		 * cache. So flush the cache before using uncached memory.
+		 */
+		dma_cache_sync(NULL, ret, size, DMA_BIDIRECTIONAL);
+		*dma_handle = virt_to_phys(ret);
 	}
+	return ret;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
 
-	memset(vp, 0, size);
-
-	/*
-	 * We must flush the cache before we pass it on to the device
-	 */
-	dma_cache_sync(NULL, ret, size, DMA_BIDIRECTIONAL);
+void dma_free_coherent(struct device *dev, size_t size,
+			 void *vaddr, dma_addr_t dma_handle)
+{
+	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+	int order = get_order(size);
 
-	page = virt_to_page(ret);
-	free = page + (size >> PAGE_SHIFT);
-	end  = page + (1 << order);
+	if (mem && vaddr >= mem->virt_base && vaddr < (mem->virt_base + (mem->size << PAGE_SHIFT))) {
+		int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
 
-	while (++page < end) {
-		/* Free any unused pages */
-		if (page >= free) {
-			__free_page(page);
-		}
+		bitmap_release_region(mem->bitmap, page, order);
+	} else {
+		WARN_ON(irqs_disabled());	/* for portability */
+		BUG_ON(mem && mem->flags & DMA_MEMORY_EXCLUSIVE);
+		free_pages((unsigned long)vaddr, order);
 	}
-
-	return vp;
 }
-EXPORT_SYMBOL(consistent_alloc);
+EXPORT_SYMBOL(dma_free_coherent);
 
-void consistent_free(void *vaddr, size_t size, dma_addr_t dma_handle)
+int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+				dma_addr_t device_addr, size_t size, int flags)
 {
-	struct page *page;
-	unsigned long addr;
-
-	addr = (unsigned long)phys_to_virt((unsigned long)dma_handle);
-	page = virt_to_page(addr);
+	void __iomem *mem_base = NULL;
+	int pages = size >> PAGE_SHIFT;
+	int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
+
+	if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
+		goto out;
+	if (!size)
+		goto out;
+	if (dev->dma_mem)
+		goto out;
+
+	/* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
+
+	mem_base = ioremap_nocache(bus_addr, size);
+	if (!mem_base)
+		goto out;
+
+	dev->dma_mem = kmalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
+	if (!dev->dma_mem)
+		goto out;
+	dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
+	if (!dev->dma_mem->bitmap)
+		goto free1_out;
+
+	dev->dma_mem->virt_base = mem_base;
+	dev->dma_mem->device_base = device_addr;
+	dev->dma_mem->size = pages;
+	dev->dma_mem->flags = flags;
+
+	if (flags & DMA_MEMORY_MAP)
+		return DMA_MEMORY_MAP;
+
+	return DMA_MEMORY_IO;
+
+ free1_out:
+	kfree(dev->dma_mem);
+ out:
+	if (mem_base)
+		iounmap(mem_base);
+	return 0;
+}
+EXPORT_SYMBOL(dma_declare_coherent_memory);
 
-	free_pages(addr, get_order(size));
+void dma_release_declared_memory(struct device *dev)
+{
+	struct dma_coherent_mem *mem = dev->dma_mem;
+
+	if (!mem)
+		return;
+	dev->dma_mem = NULL;
+	iounmap(mem->virt_base);
+	kfree(mem->bitmap);
+	kfree(mem);
+}
+EXPORT_SYMBOL(dma_release_declared_memory);
 
-	iounmap(vaddr);
+void *dma_mark_declared_memory_occupied(struct device *dev,
+					dma_addr_t device_addr, size_t size)
+{
+	struct dma_coherent_mem *mem = dev->dma_mem;
+	int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	int pos, err;
+
+	if (!mem)
+		return ERR_PTR(-EINVAL);
+
+	pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
+	err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
+	if (err != 0)
+		return ERR_PTR(err);
+	return mem->virt_base + (pos << PAGE_SHIFT);
 }
-EXPORT_SYMBOL(consistent_free);
+EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
 
-void consistent_sync(void *vaddr, size_t size, int direction)
+void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
+		    enum dma_data_direction direction)
 {
 #ifdef CONFIG_CPU_SH5
 	void *p1addr = vaddr;
@@ -94,4 +176,4 @@ void consistent_sync(void *vaddr, size_t size, int direction)
 		BUG();
 	}
 }
-EXPORT_SYMBOL(consistent_sync);
+EXPORT_SYMBOL(dma_cache_sync);
diff --git a/include/asm-sh/dma-mapping.h b/include/asm-sh/dma-mapping.h
index 20ae762e5258..22cc419389fe 100644
--- a/include/asm-sh/dma-mapping.h
+++ b/include/asm-sh/dma-mapping.h
@@ -8,11 +8,6 @@
 
 extern struct bus_type pci_bus_type;
 
-/* arch/sh/mm/consistent.c */
-extern void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle);
-extern void consistent_free(void *vaddr, size_t size, dma_addr_t handle);
-extern void consistent_sync(void *vaddr, size_t size, int direction);
-
 #define dma_supported(dev, mask)	(1)
 
 static inline int dma_set_mask(struct device *dev, u64 mask)
@@ -25,44 +20,19 @@ static inline int dma_set_mask(struct device *dev, u64 mask)
 	return 0;
 }
 
-static inline void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t flag)
-{
-	if (sh_mv.mv_consistent_alloc) {
-		void *ret;
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			 dma_addr_t *dma_handle, gfp_t flag);
 
-		ret = sh_mv.mv_consistent_alloc(dev, size, dma_handle, flag);
-		if (ret != NULL)
-			return ret;
-	}
-
-	return consistent_alloc(flag, size, dma_handle);
-}
-
-static inline void dma_free_coherent(struct device *dev, size_t size,
-		       void *vaddr, dma_addr_t dma_handle)
-{
-	if (sh_mv.mv_consistent_free) {
-		int ret;
-
-		ret = sh_mv.mv_consistent_free(dev, size, vaddr, dma_handle);
-		if (ret == 0)
-			return;
-	}
+void dma_free_coherent(struct device *dev, size_t size,
+		       void *vaddr, dma_addr_t dma_handle);
 
-	consistent_free(vaddr, size, dma_handle);
-}
+void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
+		    enum dma_data_direction dir);
 
 #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
 #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
 #define dma_is_consistent(d, h) (1)
 
-static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
-				  enum dma_data_direction dir)
-{
-	consistent_sync(vaddr, size, (int)dir);
-}
-
 static inline dma_addr_t dma_map_single(struct device *dev,
 					void *ptr, size_t size,
 					enum dma_data_direction dir)
@@ -205,4 +175,18 @@ static inline int dma_mapping_error(dma_addr_t dma_addr)
 {
 	return dma_addr == 0;
 }
+
+#define ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
+
+extern int
+dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+			    dma_addr_t device_addr, size_t size, int flags);
+
+extern void
+dma_release_declared_memory(struct device *dev);
+
+extern void *
+dma_mark_declared_memory_occupied(struct device *dev,
+				  dma_addr_t device_addr, size_t size);
+
 #endif /* __ASM_SH_DMA_MAPPING_H */
-- 
cgit v1.2.3


From d4dca67bc2fd6caa4df3db28b6424841b95fde88 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Thu, 24 Jan 2008 18:45:45 +0900
Subject: sh: remove consistent alloc stuff from the machine vector

Now with the voyagergx cruft gone and the dreamcast using declared
coherent memory for pci there are no users of the consistent alloc and
free functions pointers in the machine vector.

So this little patch simply removes these function pointers from the macvec.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/machvec.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/machvec.h b/include/asm-sh/machvec.h
index 088698bacf2f..ddb18ad23303 100644
--- a/include/asm-sh/machvec.h
+++ b/include/asm-sh/machvec.h
@@ -56,9 +56,6 @@ struct sh_machine_vector {
 
 	void (*mv_heartbeat)(void);
 
-	void *(*mv_consistent_alloc)(struct device *, size_t, dma_addr_t *, gfp_t);
-	int (*mv_consistent_free)(struct device *, size_t, void *, dma_addr_t);
-
 	void __iomem *(*mv_ioport_map)(unsigned long port, unsigned int size);
 	void (*mv_ioport_unmap)(void __iomem *);
 };
-- 
cgit v1.2.3


From da2d7f4bc578651455a7353995beb87db3cd8815 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Fri, 25 Jan 2008 16:04:29 +0900
Subject: sh: update r7780rp interrupt code

This patch updates the board specific irq code for r7780rp. The new code is
very similar to the other highlander implementations, with the exception that
the r7780rp handles pci interrupts using IRL. To simplify the pci code and
use the same interrupt numbers as r7780mp and r7785rp we hook in to the
cpu specific pci vectors.

The pci interrupts and the push switch all work well with and without this
patch. CF and AX88796 are not ok though and the source of the problem is
unknown at this point. The AX88796 does for not detect it's proper mac
address (IPL gets it right) and the kernel hangs on CF access. As a workaround
this patch removes the CF and the AX88796 from the platform datain case of
r7780rp.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/r7780rp/Makefile      |  2 +-
 arch/sh/boards/renesas/r7780rp/irq-r7780rp.c | 52 +++++++++++++++++++++++++---
 arch/sh/boards/renesas/r7780rp/irq.c         | 51 ---------------------------
 arch/sh/boards/renesas/r7780rp/setup.c       | 10 +++---
 arch/sh/drivers/pci/ops-r7780rp.c            | 16 ++-------
 include/asm-sh/r7780rp.h                     | 19 ----------
 6 files changed, 57 insertions(+), 93 deletions(-)
 delete mode 100644 arch/sh/boards/renesas/r7780rp/irq.c

(limited to 'include')

diff --git a/arch/sh/boards/renesas/r7780rp/Makefile b/arch/sh/boards/renesas/r7780rp/Makefile
index dd26182fbf58..20a10080b11f 100644
--- a/arch/sh/boards/renesas/r7780rp/Makefile
+++ b/arch/sh/boards/renesas/r7780rp/Makefile
@@ -3,7 +3,7 @@
 #
 irqinit-$(CONFIG_SH_R7780MP)	:= irq-r7780mp.o
 irqinit-$(CONFIG_SH_R7785RP)	:= irq-r7785rp.o
-irqinit-$(CONFIG_SH_R7780RP)	:= irq-r7780rp.o irq.o
+irqinit-$(CONFIG_SH_R7780RP)	:= irq-r7780rp.o
 obj-y				:= setup.o $(irqinit-y)
 
 ifneq ($(CONFIG_SH_R7785RP),y)
diff --git a/arch/sh/boards/renesas/r7780rp/irq-r7780rp.c b/arch/sh/boards/renesas/r7780rp/irq-r7780rp.c
index fa4a534cade9..bd34048ed0e1 100644
--- a/arch/sh/boards/renesas/r7780rp/irq-r7780rp.c
+++ b/arch/sh/boards/renesas/r7780rp/irq-r7780rp.c
@@ -3,21 +3,65 @@
  *
  * Copyright (C) 2002  Atom Create Engineering Co., Ltd.
  * Copyright (C) 2006  Paul Mundt
+ * Copyright (C) 2008  Magnus Damm
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
 #include <linux/init.h>
+#include <linux/irq.h>
 #include <linux/io.h>
 #include <asm/r7780rp.h>
 
+enum {
+	UNUSED = 0,
+
+	/* board specific interrupt sources */
+
+	AX88796,          /* Ethernet controller */
+	PSW,              /* Push Switch */
+	CF,               /* Compact Flash */
+
+	PCI_A,
+	PCI_B,
+	PCI_C,
+	PCI_D,
+};
+
+static struct intc_vect vectors[] __initdata = {
+	INTC_IRQ(PCI_A, 65), /* dirty: overwrite cpu vectors for pci */
+	INTC_IRQ(PCI_B, 66),
+	INTC_IRQ(PCI_C, 67),
+	INTC_IRQ(PCI_D, 68),
+	INTC_IRQ(CF, IRQ_CF),
+	INTC_IRQ(PSW, IRQ_PSW),
+	INTC_IRQ(AX88796, IRQ_AX88796),
+};
+
+static struct intc_mask_reg mask_registers[] __initdata = {
+	{ 0xa5000000, 0, 16, /* IRLMSK */
+	  { PCI_A, PCI_B, PCI_C, PCI_D, CF, 0, 0, 0,
+	    0, 0, 0, 0, 0, 0, PSW, AX88796 } },
+};
+
+static unsigned char irl2irq[HL_NR_IRL] __initdata = {
+	65, 66, 67, 68,
+	IRQ_CF, 0, 0, 0,
+	0, 0, 0, 0,
+	IRQ_AX88796, IRQ_PSW
+};
+
+static DECLARE_INTC_DESC(intc_desc, "r7780rp", vectors,
+			 NULL, mask_registers, NULL, NULL);
+
 unsigned char * __init highlander_init_irq_r7780rp(void)
 {
-	int i;
-
-	for (i = 0; i < 15; i++)
-		make_r7780rp_irq(i);
+	if (ctrl_inw(0xa5000600)) {
+		printk(KERN_INFO "Using r7780rp interrupt controller.\n");
+		register_intc_controller(&intc_desc);
+		return irl2irq;
+	}
 
 	return NULL;
 }
diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c
deleted file mode 100644
index e0b8eb52f376..000000000000
--- a/arch/sh/boards/renesas/r7780rp/irq.c
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * Renesas Solutions Highlander R7780RP-1 Support.
- *
- * Copyright (C) 2002  Atom Create Engineering Co., Ltd.
- * Copyright (C) 2006  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/interrupt.h>
-#include <linux/io.h>
-#include <asm/r7780rp.h>
-
-#ifdef CONFIG_SH_R7780RP
-static int mask_pos[] = {15, 14, 13, 12, 11, 10, 9, 8, 7, 5, 6, 4, 0, 1, 2, 0};
-#elif defined(CONFIG_SH_R7780MP)
-static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0};
-#elif defined(CONFIG_SH_R7785RP)
-static int mask_pos[] = {2, 11, 2, 2, 2, 2, 9, 8, 7, 5, 10, 2, 2, 2, 2, 2};
-#endif
-
-static void enable_r7780rp_irq(unsigned int irq)
-{
-	/* Set priority in IPR back to original value */
-	ctrl_outw(ctrl_inw(IRLCNTR1) | (1 << mask_pos[irq]), IRLCNTR1);
-}
-
-static void disable_r7780rp_irq(unsigned int irq)
-{
-	/* Set the priority in IPR to 0 */
-	ctrl_outw(ctrl_inw(IRLCNTR1) & (0xffff ^ (1 << mask_pos[irq])),
-		  IRLCNTR1);
-}
-
-static struct irq_chip r7780rp_irq_chip __read_mostly = {
-	.name		= "R7780RP",
-	.mask		= disable_r7780rp_irq,
-	.unmask		= enable_r7780rp_irq,
-	.mask_ack	= disable_r7780rp_irq,
-};
-
-void make_r7780rp_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	set_irq_chip_and_handler_name(irq, &r7780rp_irq_chip,
-				      handle_level_irq, "level");
-	enable_r7780rp_irq(irq);
-}
diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c
index 0fdc0bc19145..a43b47726f54 100644
--- a/arch/sh/boards/renesas/r7780rp/setup.c
+++ b/arch/sh/boards/renesas/r7780rp/setup.c
@@ -179,9 +179,11 @@ static struct platform_device ax88796_device = {
 static struct platform_device *r7780rp_devices[] __initdata = {
 	&r8a66597_usb_host_device,
 	&m66592_usb_peripheral_device,
-	&cf_ide_device,
 	&heartbeat_device,
+#ifndef CONFIG_SH_R7780RP
+	&cf_ide_device,
 	&ax88796_device,
+#endif
 };
 
 static int __init r7780rp_devices_setup(void)
@@ -316,9 +318,9 @@ void __init highlander_init_irq(void)
 			break;
 #endif
 #ifdef CONFIG_SH_R7780RP
-		highlander_init_irq_r7780rp();
-		ucp = irl2irq;
-		break;
+		ucp = highlander_init_irq_r7780rp();
+		if (ucp)
+			break;
 #endif
 	} while (0);
 
diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c
index 48fe4032ebea..5fdadaeed6fc 100644
--- a/arch/sh/drivers/pci/ops-r7780rp.c
+++ b/arch/sh/drivers/pci/ops-r7780rp.c
@@ -17,25 +17,13 @@
 #include <asm/io.h>
 #include "pci-sh4.h"
 
-static char r7780rp_irq_tab[] __initdata = {
-	0, 1, 2, 3,
-};
-
-static char r7780mp_irq_tab[] __initdata = {
+static char irq_tab[] __initdata = {
 	65, 66, 67, 68,
 };
 
 int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
-	if (mach_is_r7780rp())
-		return r7780rp_irq_tab[slot];
-	if (mach_is_r7780mp() || mach_is_r7785rp())
-		return r7780mp_irq_tab[slot];
-
-	printk(KERN_ERR "PCI: Bad IRQ mapping "
-	       "request for slot %d, func %d\n", slot, pin-1);
-
-	return -1;
+	return irq_tab[slot];
 }
 
 static struct resource sh7780_io_resource = {
diff --git a/include/asm-sh/r7780rp.h b/include/asm-sh/r7780rp.h
index 1c2d4d18fab2..bdecea0840a0 100644
--- a/include/asm-sh/r7780rp.h
+++ b/include/asm-sh/r7780rp.h
@@ -121,21 +121,6 @@
 
 #define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
 
-#define IRQ_PCISLOT1	0		/* PCI Slot #1 IRQ */
-#define IRQ_PCISLOT2	1		/* PCI Slot #2 IRQ */
-#define IRQ_PCISLOT3	2		/* PCI Slot #3 IRQ */
-#define IRQ_PCISLOT4	3		/* PCI Slot #4 IRQ */
-#define IRQ_CFINST	5		/* CF Card Insert IRQ */
-#define IRQ_M66596	6		/* M66596 IRQ */
-#define IRQ_SDCARD	7		/* SD Card IRQ */
-#define IRQ_TUCHPANEL	8		/* Touch Panel IRQ */
-#define IRQ_SCI		9		/* SCI IRQ */
-#define IRQ_2SERIAL	10		/* Serial IRQ */
-#define	IRQ_EXTENTION	11		/* EXTn IRQ */
-#define IRQ_ONETH	12		/* On board Ethernet IRQ */
-#define IRQ_PSW		13		/* Push Switch IRQ */
-#define IRQ_ZIGBEE	14		/* Ziggbee IO IRQ */
-
 #define IVDR_CK_ON	8		/* iVDR Clock ON */
 
 #elif defined(CONFIG_SH_R7785RP)
@@ -192,9 +177,7 @@
 
 #define IRQ_AX88796		(HL_FPGA_IRQ_BASE + 0)
 #define IRQ_CF			(HL_FPGA_IRQ_BASE + 1)
-#ifndef IRQ_PSW
 #define IRQ_PSW			(HL_FPGA_IRQ_BASE + 2)
-#endif
 #define IRQ_EXT0		(HL_FPGA_IRQ_BASE + 3)
 #define IRQ_EXT1		(HL_FPGA_IRQ_BASE + 4)
 #define IRQ_EXT2		(HL_FPGA_IRQ_BASE + 5)
@@ -208,8 +191,6 @@
 #define IRQ_RTC			(HL_FPGA_IRQ_BASE + 13)
 #define IRQ_TH_ALERT		(HL_FPGA_IRQ_BASE + 14)
 
-void make_r7780rp_irq(unsigned int irq);
-
 unsigned char *highlander_init_irq_r7780mp(void);
 unsigned char *highlander_init_irq_r7780rp(void);
 unsigned char *highlander_init_irq_r7785rp(void);
-- 
cgit v1.2.3


From 6582d7b7376aa587d74b08c74457dc28abc1a9fa Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Wed, 23 Jan 2008 16:21:18 +0900
Subject: sh: add spi header and r2d platform data V3

This patch adds the header file asm/spi.h and board specific code for the
r2d board. The header file contains a structure that should be used to
point out a single spi bus. The board specific code for r2d is updated with
such a structure for the new spi_sh_sci driver. The structure contains a
chip select callback plus information about the R9701 rtc chip which is
attached to the spi bus.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/rts7751r2d/setup.c | 42 +++++++++++++++++++++++++++++++
 include/asm-sh/spi.h                      | 13 ++++++++++
 2 files changed, 55 insertions(+)
 create mode 100644 include/asm-sh/spi.h

(limited to 'include')

diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index 8528616feaa0..3452b072adde 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -16,9 +16,12 @@
 #include <linux/sm501-regs.h>
 #include <linux/pm.h>
 #include <linux/fb.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/spi_bitbang.h>
 #include <asm/machvec.h>
 #include <asm/rts7751r2d.h>
 #include <asm/io.h>
+#include <asm/spi.h>
 
 static struct resource cf_ide_resources[] = {
 	[0] = {
@@ -53,6 +56,43 @@ static struct platform_device cf_ide_device  = {
 	},
 };
 
+static struct spi_board_info spi_bus[] = {
+	{
+		.modalias	= "rtc-r9701",
+		.max_speed_hz	= 1000000,
+		.mode		= SPI_MODE_3,
+	},
+};
+
+static void r2d_chip_select(struct sh_spi_info *spi, int cs, int state)
+{
+	BUG_ON(cs != 0);  /* Single Epson RTC-9701JE attached on CS0 */
+	ctrl_outw(state == BITBANG_CS_ACTIVE, PA_RTCCE);
+}
+
+static struct sh_spi_info spi_info = {
+	.num_chipselect = 1,
+	.chip_select = r2d_chip_select,
+};
+
+static struct resource spi_sh_sci_resources[] = {
+	{
+		.start	= 0xffe00000,
+		.end	= 0xffe0001f,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+static struct platform_device spi_sh_sci_device  = {
+	.name		= "spi_sh_sci",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(spi_sh_sci_resources),
+	.resource	= spi_sh_sci_resources,
+	.dev	= {
+		.platform_data	= &spi_info,
+	},
+};
+
 static struct resource heartbeat_resources[] = {
 	[0] = {
 		.start	= PA_OUTPORT,
@@ -176,10 +216,12 @@ static struct platform_device *rts7751r2d_devices[] __initdata = {
 #endif
 	&cf_ide_device,
 	&heartbeat_device,
+	&spi_sh_sci_device,
 };
 
 static int __init rts7751r2d_devices_setup(void)
 {
+	spi_register_board_info(spi_bus, ARRAY_SIZE(spi_bus));
 	return platform_add_devices(rts7751r2d_devices,
 				    ARRAY_SIZE(rts7751r2d_devices));
 }
diff --git a/include/asm-sh/spi.h b/include/asm-sh/spi.h
new file mode 100644
index 000000000000..e96f5b0953c8
--- /dev/null
+++ b/include/asm-sh/spi.h
@@ -0,0 +1,13 @@
+#ifndef __ASM_SPI_H__
+#define __ASM_SPI_H__
+
+struct sh_spi_info;
+
+struct sh_spi_info {
+	int			 bus_num;
+	int			 num_chipselect;
+
+	void (*chip_select)(struct sh_spi_info *spi, int cs, int state);
+};
+
+#endif /* __ASM_SPI_H__ */
-- 
cgit v1.2.3


From 6da127ad0918f93ea93678dad62ce15ffed18797 Mon Sep 17 00:00:00 2001
From: Christof Schmitt <christof.schmitt@de.ibm.com>
Date: Fri, 11 Jan 2008 10:09:43 +0100
Subject: blktrace: Add blktrace ioctls to SCSI generic devices

Since the SCSI layer uses the request queues from the block layer, blktrace can
also be used to trace the requests to all SCSI devices (like SCSI tape drives),
not only disks. The only missing part is the ioctl interface to start and stop
tracing.

This patch adds the SETUP, START, STOP and TEARDOWN ioctls from blktrace to the
sg device files. With this change, blktrace can be used for SCSI devices like
for disks, e.g.: blktrace -d /dev/sg1 -o - | blkparse -i -

Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blktrace.c             | 24 ++++++++++++++----------
 block/compat_ioctl.c         |  5 ++++-
 drivers/scsi/sg.c            | 12 ++++++++++++
 include/linux/blktrace_api.h | 12 ++++++++++--
 4 files changed, 40 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/block/blktrace.c b/block/blktrace.c
index 9b4da4ae3c7d..568588cd16b2 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -235,7 +235,7 @@ static void blk_trace_cleanup(struct blk_trace *bt)
 	kfree(bt);
 }
 
-static int blk_trace_remove(struct request_queue *q)
+int blk_trace_remove(struct request_queue *q)
 {
 	struct blk_trace *bt;
 
@@ -249,6 +249,7 @@ static int blk_trace_remove(struct request_queue *q)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(blk_trace_remove);
 
 static int blk_dropped_open(struct inode *inode, struct file *filp)
 {
@@ -316,18 +317,17 @@ static struct rchan_callbacks blk_relay_callbacks = {
 /*
  * Setup everything required to start tracing
  */
-int do_blk_trace_setup(struct request_queue *q, struct block_device *bdev,
+int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
 			struct blk_user_trace_setup *buts)
 {
 	struct blk_trace *old_bt, *bt = NULL;
 	struct dentry *dir = NULL;
-	char b[BDEVNAME_SIZE];
 	int ret, i;
 
 	if (!buts->buf_size || !buts->buf_nr)
 		return -EINVAL;
 
-	strcpy(buts->name, bdevname(bdev, b));
+	strcpy(buts->name, name);
 
 	/*
 	 * some device names have larger paths - convert the slashes
@@ -352,7 +352,7 @@ int do_blk_trace_setup(struct request_queue *q, struct block_device *bdev,
 		goto err;
 
 	bt->dir = dir;
-	bt->dev = bdev->bd_dev;
+	bt->dev = dev;
 	atomic_set(&bt->dropped, 0);
 
 	ret = -EIO;
@@ -399,8 +399,8 @@ err:
 	return ret;
 }
 
-static int blk_trace_setup(struct request_queue *q, struct block_device *bdev,
-			   char __user *arg)
+int blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
+		    char __user *arg)
 {
 	struct blk_user_trace_setup buts;
 	int ret;
@@ -409,7 +409,7 @@ static int blk_trace_setup(struct request_queue *q, struct block_device *bdev,
 	if (ret)
 		return -EFAULT;
 
-	ret = do_blk_trace_setup(q, bdev, &buts);
+	ret = do_blk_trace_setup(q, name, dev, &buts);
 	if (ret)
 		return ret;
 
@@ -418,8 +418,9 @@ static int blk_trace_setup(struct request_queue *q, struct block_device *bdev,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(blk_trace_setup);
 
-static int blk_trace_startstop(struct request_queue *q, int start)
+int blk_trace_startstop(struct request_queue *q, int start)
 {
 	struct blk_trace *bt;
 	int ret;
@@ -452,6 +453,7 @@ static int blk_trace_startstop(struct request_queue *q, int start)
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(blk_trace_startstop);
 
 /**
  * blk_trace_ioctl: - handle the ioctls associated with tracing
@@ -464,6 +466,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
 {
 	struct request_queue *q;
 	int ret, start = 0;
+	char b[BDEVNAME_SIZE];
 
 	q = bdev_get_queue(bdev);
 	if (!q)
@@ -473,7 +476,8 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
 
 	switch (cmd) {
 	case BLKTRACESETUP:
-		ret = blk_trace_setup(q, bdev, arg);
+		strcpy(b, bdevname(bdev, b));
+		ret = blk_trace_setup(q, b, bdev->bd_dev, arg);
 		break;
 	case BLKTRACESTART:
 		start = 1;
diff --git a/block/compat_ioctl.c b/block/compat_ioctl.c
index cae0a852619e..b73373216b0e 100644
--- a/block/compat_ioctl.c
+++ b/block/compat_ioctl.c
@@ -545,6 +545,7 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg)
 	struct blk_user_trace_setup buts;
 	struct compat_blk_user_trace_setup cbuts;
 	struct request_queue *q;
+	char b[BDEVNAME_SIZE];
 	int ret;
 
 	q = bdev_get_queue(bdev);
@@ -554,6 +555,8 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg)
 	if (copy_from_user(&cbuts, arg, sizeof(cbuts)))
 		return -EFAULT;
 
+	strcpy(b, bdevname(bdev, b));
+
 	buts = (struct blk_user_trace_setup) {
 		.act_mask = cbuts.act_mask,
 		.buf_size = cbuts.buf_size,
@@ -565,7 +568,7 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg)
 	memcpy(&buts.name, &cbuts.name, 32);
 
 	mutex_lock(&bdev->bd_mutex);
-	ret = do_blk_trace_setup(q, bdev, &buts);
+	ret = do_blk_trace_setup(q, b, bdev->bd_dev, &buts);
 	mutex_unlock(&bdev->bd_mutex);
 	if (ret)
 		return ret;
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 17216b76efdc..aba28f335b88 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -48,6 +48,7 @@ static int sg_version_num = 30534;	/* 2 digits for each component */
 #include <linux/blkdev.h>
 #include <linux/delay.h>
 #include <linux/scatterlist.h>
+#include <linux/blktrace_api.h>
 
 #include "scsi.h"
 #include <scsi/scsi_dbg.h>
@@ -1067,6 +1068,17 @@ sg_ioctl(struct inode *inode, struct file *filp,
 	case BLKSECTGET:
 		return put_user(sdp->device->request_queue->max_sectors * 512,
 				ip);
+	case BLKTRACESETUP:
+		return blk_trace_setup(sdp->device->request_queue,
+				       sdp->disk->disk_name,
+				       sdp->device->sdev_gendev.devt,
+				       (char *)arg);
+	case BLKTRACESTART:
+		return blk_trace_startstop(sdp->device->request_queue, 1);
+	case BLKTRACESTOP:
+		return blk_trace_startstop(sdp->device->request_queue, 0);
+	case BLKTRACETEARDOWN:
+		return blk_trace_remove(sdp->device->request_queue);
 	default:
 		if (read_only)
 			return -EPERM;	/* don't know so take safe approach */
diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index 7e11d23ac36a..06dadba349ac 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -148,7 +148,7 @@ extern int blk_trace_ioctl(struct block_device *, unsigned, char __user *);
 extern void blk_trace_shutdown(struct request_queue *);
 extern void __blk_add_trace(struct blk_trace *, sector_t, int, int, u32, int, int, void *);
 extern int do_blk_trace_setup(struct request_queue *q,
-	struct block_device *bdev, struct blk_user_trace_setup *buts);
+	char *name, dev_t dev, struct blk_user_trace_setup *buts);
 
 
 /**
@@ -282,6 +282,11 @@ static inline void blk_add_trace_remap(struct request_queue *q, struct bio *bio,
 	__blk_add_trace(bt, from, bio->bi_size, bio->bi_rw, BLK_TA_REMAP, !bio_flagged(bio, BIO_UPTODATE), sizeof(r), &r);
 }
 
+extern int blk_trace_setup(request_queue_t *q, char *name, dev_t dev,
+			   char __user *arg);
+extern int blk_trace_startstop(request_queue_t *q, int start);
+extern int blk_trace_remove(request_queue_t *q);
+
 #else /* !CONFIG_BLK_DEV_IO_TRACE */
 #define blk_trace_ioctl(bdev, cmd, arg)		(-ENOTTY)
 #define blk_trace_shutdown(q)			do { } while (0)
@@ -290,7 +295,10 @@ static inline void blk_add_trace_remap(struct request_queue *q, struct bio *bio,
 #define blk_add_trace_generic(q, rq, rw, what)	do { } while (0)
 #define blk_add_trace_pdu_int(q, what, bio, pdu)	do { } while (0)
 #define blk_add_trace_remap(q, bio, dev, f, t)	do {} while (0)
-#define do_blk_trace_setup(q, bdev, buts)	(-ENOTTY)
+#define do_blk_trace_setup(q, name, dev, buts)	(-ENOTTY)
+#define blk_trace_setup(q, name, dev, arg)	(-ENOTTY)
+#define blk_trace_startstop(q, start)		(-ENOTTY)
+#define blk_trace_remove(q)			(-ENOTTY)
 #endif /* CONFIG_BLK_DEV_IO_TRACE */
 #endif /* __KERNEL__ */
 #endif
-- 
cgit v1.2.3


From 482eb689169948e9f4966fbae6be4d6bc0bfa818 Mon Sep 17 00:00:00 2001
From: Pete Wyckoff <pw@osc.edu>
Date: Tue, 1 Jan 2008 10:23:02 -0500
Subject: block: allow queue dma_alignment of zero

Let queue_dma_alignment return 0 if it was specifically set to 0.
This permits devices with no particular alignment restrictions to
use arbitrary user space buffers without copying.

Signed-off-by: Pete Wyckoff <pw@osc.edu>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 49b7a4c31a6d..c7a3ab575c24 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -836,12 +836,7 @@ static inline int bdev_hardsect_size(struct block_device *bdev)
 
 static inline int queue_dma_alignment(struct request_queue *q)
 {
-	int retval = 511;
-
-	if (q && q->dma_alignment)
-		retval = q->dma_alignment;
-
-	return retval;
+	return q ? q->dma_alignment : 511;
 }
 
 /* assumes size > 256 */
-- 
cgit v1.2.3


From 0db9299f48ebd4a860d6ad4e1d36ac50671d48e7 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 30 Nov 2007 09:16:50 +0100
Subject: SG: Move functions to lib/scatterlist.c and add sg chaining allocator
 helpers

Manually doing chained sg lists is not trivial, so add some helpers
to make sure that drivers get it right.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/scatterlist.h | 125 ++++----------------
 lib/Makefile                |   2 +-
 lib/scatterlist.c           | 281 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 307 insertions(+), 101 deletions(-)
 create mode 100644 lib/scatterlist.c

(limited to 'include')

diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index e3ff21dbac53..e9cb103417b2 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -7,6 +7,12 @@
 #include <linux/string.h>
 #include <asm/io.h>
 
+struct sg_table {
+	struct scatterlist *sgl;	/* the list */
+	unsigned int nents;		/* number of mapped entries */
+	unsigned int orig_nents;	/* original size of list */
+};
+
 /*
  * Notes on SG table design.
  *
@@ -106,71 +112,12 @@ static inline void sg_set_buf(struct scatterlist *sg, const void *buf,
 	sg_set_page(sg, virt_to_page(buf), buflen, offset_in_page(buf));
 }
 
-/**
- * sg_next - return the next scatterlist entry in a list
- * @sg:		The current sg entry
- *
- * Description:
- *   Usually the next entry will be @sg@ + 1, but if this sg element is part
- *   of a chained scatterlist, it could jump to the start of a new
- *   scatterlist array.
- *
- **/
-static inline struct scatterlist *sg_next(struct scatterlist *sg)
-{
-#ifdef CONFIG_DEBUG_SG
-	BUG_ON(sg->sg_magic != SG_MAGIC);
-#endif
-	if (sg_is_last(sg))
-		return NULL;
-
-	sg++;
-	if (unlikely(sg_is_chain(sg)))
-		sg = sg_chain_ptr(sg);
-
-	return sg;
-}
-
 /*
  * Loop over each sg element, following the pointer to a new list if necessary
  */
 #define for_each_sg(sglist, sg, nr, __i)	\
 	for (__i = 0, sg = (sglist); __i < (nr); __i++, sg = sg_next(sg))
 
-/**
- * sg_last - return the last scatterlist entry in a list
- * @sgl:	First entry in the scatterlist
- * @nents:	Number of entries in the scatterlist
- *
- * Description:
- *   Should only be used casually, it (currently) scan the entire list
- *   to get the last entry.
- *
- *   Note that the @sgl@ pointer passed in need not be the first one,
- *   the important bit is that @nents@ denotes the number of entries that
- *   exist from @sgl@.
- *
- **/
-static inline struct scatterlist *sg_last(struct scatterlist *sgl,
-					  unsigned int nents)
-{
-#ifndef ARCH_HAS_SG_CHAIN
-	struct scatterlist *ret = &sgl[nents - 1];
-#else
-	struct scatterlist *sg, *ret = NULL;
-	unsigned int i;
-
-	for_each_sg(sgl, sg, nents, i)
-		ret = sg;
-
-#endif
-#ifdef CONFIG_DEBUG_SG
-	BUG_ON(sgl[0].sg_magic != SG_MAGIC);
-	BUG_ON(!sg_is_last(ret));
-#endif
-	return ret;
-}
-
 /**
  * sg_chain - Chain two sglists together
  * @prv:	First scatterlist
@@ -222,47 +169,6 @@ static inline void sg_mark_end(struct scatterlist *sg)
 	sg->page_link &= ~0x01;
 }
 
-/**
- * sg_init_table - Initialize SG table
- * @sgl:	   The SG table
- * @nents:	   Number of entries in table
- *
- * Notes:
- *   If this is part of a chained sg table, sg_mark_end() should be
- *   used only on the last table part.
- *
- **/
-static inline void sg_init_table(struct scatterlist *sgl, unsigned int nents)
-{
-	memset(sgl, 0, sizeof(*sgl) * nents);
-#ifdef CONFIG_DEBUG_SG
-	{
-		unsigned int i;
-		for (i = 0; i < nents; i++)
-			sgl[i].sg_magic = SG_MAGIC;
-	}
-#endif
-	sg_mark_end(&sgl[nents - 1]);
-}
-
-/**
- * sg_init_one - Initialize a single entry sg list
- * @sg:		 SG entry
- * @buf:	 Virtual address for IO
- * @buflen:	 IO length
- *
- * Notes:
- *   This should not be used on a single entry that is part of a larger
- *   table. Use sg_init_table() for that.
- *
- **/
-static inline void sg_init_one(struct scatterlist *sg, const void *buf,
-			       unsigned int buflen)
-{
-	sg_init_table(sg, 1);
-	sg_set_buf(sg, buf, buflen);
-}
-
 /**
  * sg_phys - Return physical address of an sg entry
  * @sg:	     SG entry
@@ -293,4 +199,23 @@ static inline void *sg_virt(struct scatterlist *sg)
 	return page_address(sg_page(sg)) + sg->offset;
 }
 
+struct scatterlist *sg_next(struct scatterlist *);
+struct scatterlist *sg_last(struct scatterlist *s, unsigned int);
+void sg_init_table(struct scatterlist *, unsigned int);
+void sg_init_one(struct scatterlist *, const void *, unsigned int);
+
+typedef struct scatterlist *(sg_alloc_fn)(unsigned int, gfp_t);
+typedef void (sg_free_fn)(struct scatterlist *, unsigned int);
+
+void __sg_free_table(struct sg_table *, sg_free_fn *);
+void sg_free_table(struct sg_table *);
+int __sg_alloc_table(struct sg_table *, unsigned int, gfp_t, sg_alloc_fn *);
+int sg_alloc_table(struct sg_table *, unsigned int, gfp_t);
+
+/*
+ * Maximum number of entries that will be allocated in one piece, if
+ * a list larger than this is required then chaining will be utilized.
+ */
+#define SG_MAX_SINGLE_ALLOC		(PAGE_SIZE / sizeof(struct scatterlist))
+
 #endif /* _LINUX_SCATTERLIST_H */
diff --git a/lib/Makefile b/lib/Makefile
index b6793ed28d84..89841dc9d91c 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -6,7 +6,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \
 	 rbtree.o radix-tree.o dump_stack.o \
 	 idr.o int_sqrt.o extable.o prio_tree.o \
 	 sha1.o irq_regs.o reciprocal_div.o argv_split.o \
-	 proportions.o prio_heap.o
+	 proportions.o prio_heap.o scatterlist.o
 
 lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
new file mode 100644
index 000000000000..02aaa27e010e
--- /dev/null
+++ b/lib/scatterlist.c
@@ -0,0 +1,281 @@
+/*
+ * Copyright (C) 2007 Jens Axboe <jens.axboe@oracle.com>
+ *
+ * Scatterlist handling helpers.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2. See the file COPYING for more details.
+ */
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+
+/**
+ * sg_next - return the next scatterlist entry in a list
+ * @sg:		The current sg entry
+ *
+ * Description:
+ *   Usually the next entry will be @sg@ + 1, but if this sg element is part
+ *   of a chained scatterlist, it could jump to the start of a new
+ *   scatterlist array.
+ *
+ **/
+struct scatterlist *sg_next(struct scatterlist *sg)
+{
+#ifdef CONFIG_DEBUG_SG
+	BUG_ON(sg->sg_magic != SG_MAGIC);
+#endif
+	if (sg_is_last(sg))
+		return NULL;
+
+	sg++;
+	if (unlikely(sg_is_chain(sg)))
+		sg = sg_chain_ptr(sg);
+
+	return sg;
+}
+EXPORT_SYMBOL(sg_next);
+
+/**
+ * sg_last - return the last scatterlist entry in a list
+ * @sgl:	First entry in the scatterlist
+ * @nents:	Number of entries in the scatterlist
+ *
+ * Description:
+ *   Should only be used casually, it (currently) scans the entire list
+ *   to get the last entry.
+ *
+ *   Note that the @sgl@ pointer passed in need not be the first one,
+ *   the important bit is that @nents@ denotes the number of entries that
+ *   exist from @sgl@.
+ *
+ **/
+struct scatterlist *sg_last(struct scatterlist *sgl, unsigned int nents)
+{
+#ifndef ARCH_HAS_SG_CHAIN
+	struct scatterlist *ret = &sgl[nents - 1];
+#else
+	struct scatterlist *sg, *ret = NULL;
+	unsigned int i;
+
+	for_each_sg(sgl, sg, nents, i)
+		ret = sg;
+
+#endif
+#ifdef CONFIG_DEBUG_SG
+	BUG_ON(sgl[0].sg_magic != SG_MAGIC);
+	BUG_ON(!sg_is_last(ret));
+#endif
+	return ret;
+}
+EXPORT_SYMBOL(sg_last);
+
+/**
+ * sg_init_table - Initialize SG table
+ * @sgl:	   The SG table
+ * @nents:	   Number of entries in table
+ *
+ * Notes:
+ *   If this is part of a chained sg table, sg_mark_end() should be
+ *   used only on the last table part.
+ *
+ **/
+void sg_init_table(struct scatterlist *sgl, unsigned int nents)
+{
+	memset(sgl, 0, sizeof(*sgl) * nents);
+#ifdef CONFIG_DEBUG_SG
+	{
+		unsigned int i;
+		for (i = 0; i < nents; i++)
+			sgl[i].sg_magic = SG_MAGIC;
+	}
+#endif
+	sg_mark_end(&sgl[nents - 1]);
+}
+EXPORT_SYMBOL(sg_init_table);
+
+/**
+ * sg_init_one - Initialize a single entry sg list
+ * @sg:		 SG entry
+ * @buf:	 Virtual address for IO
+ * @buflen:	 IO length
+ *
+ **/
+void sg_init_one(struct scatterlist *sg, const void *buf, unsigned int buflen)
+{
+	sg_init_table(sg, 1);
+	sg_set_buf(sg, buf, buflen);
+}
+EXPORT_SYMBOL(sg_init_one);
+
+/*
+ * The default behaviour of sg_alloc_table() is to use these kmalloc/kfree
+ * helpers.
+ */
+static struct scatterlist *sg_kmalloc(unsigned int nents, gfp_t gfp_mask)
+{
+	if (nents == SG_MAX_SINGLE_ALLOC)
+		return (struct scatterlist *) __get_free_page(gfp_mask);
+	else
+		return kmalloc(nents * sizeof(struct scatterlist), gfp_mask);
+}
+
+static void sg_kfree(struct scatterlist *sg, unsigned int nents)
+{
+	if (nents == SG_MAX_SINGLE_ALLOC)
+		free_page((unsigned long) sg);
+	else
+		kfree(sg);
+}
+
+/**
+ * __sg_free_table - Free a previously mapped sg table
+ * @table:	The sg table header to use
+ * @free_fn:	Free function
+ *
+ *  Description:
+ *    Free an sg table previously allocated and setup with __sg_alloc_table().
+ *
+ **/
+void __sg_free_table(struct sg_table *table, sg_free_fn *free_fn)
+{
+	struct scatterlist *sgl, *next;
+
+	if (unlikely(!table->sgl))
+		return;
+
+	sgl = table->sgl;
+	while (table->orig_nents) {
+		unsigned int alloc_size = table->orig_nents;
+		unsigned int sg_size;
+
+		/*
+		 * If we have more than SG_MAX_SINGLE_ALLOC segments left,
+		 * then assign 'next' to the sg table after the current one.
+		 * sg_size is then one less than alloc size, since the last
+		 * element is the chain pointer.
+		 */
+		if (alloc_size > SG_MAX_SINGLE_ALLOC) {
+			next = sg_chain_ptr(&sgl[SG_MAX_SINGLE_ALLOC - 1]);
+			alloc_size = SG_MAX_SINGLE_ALLOC;
+			sg_size = alloc_size - 1;
+		} else {
+			sg_size = alloc_size;
+			next = NULL;
+		}
+
+		table->orig_nents -= sg_size;
+		free_fn(sgl, alloc_size);
+		sgl = next;
+	}
+
+	table->sgl = NULL;
+}
+EXPORT_SYMBOL(__sg_free_table);
+
+/**
+ * sg_free_table - Free a previously allocated sg table
+ * @table:	The mapped sg table header
+ *
+ **/
+void sg_free_table(struct sg_table *table)
+{
+	__sg_free_table(table, sg_kfree);
+}
+EXPORT_SYMBOL(sg_free_table);
+
+/**
+ * __sg_alloc_table - Allocate and initialize an sg table with given allocator
+ * @table:	The sg table header to use
+ * @nents:	Number of entries in sg list
+ * @gfp_mask:	GFP allocation mask
+ * @alloc_fn:	Allocator to use
+ *
+ * Notes:
+ *   If this function returns non-0 (eg failure), the caller must call
+ *   __sg_free_table() to cleanup any leftover allocations.
+ *
+ **/
+int __sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask,
+		     sg_alloc_fn *alloc_fn)
+{
+	struct scatterlist *sg, *prv;
+	unsigned int left;
+
+#ifndef ARCH_HAS_SG_CHAIN
+	BUG_ON(nents > SG_MAX_SINGLE_ALLOC);
+#endif
+
+	memset(table, 0, sizeof(*table));
+
+	left = nents;
+	prv = NULL;
+	do {
+		unsigned int sg_size, alloc_size = left;
+
+		if (alloc_size > SG_MAX_SINGLE_ALLOC) {
+			alloc_size = SG_MAX_SINGLE_ALLOC;
+			sg_size = alloc_size - 1;
+		} else
+			sg_size = alloc_size;
+
+		left -= sg_size;
+
+		sg = alloc_fn(alloc_size, gfp_mask);
+		if (unlikely(!sg))
+			return -ENOMEM;
+
+		sg_init_table(sg, alloc_size);
+		table->nents = table->orig_nents += sg_size;
+
+		/*
+		 * If this is the first mapping, assign the sg table header.
+		 * If this is not the first mapping, chain previous part.
+		 */
+		if (prv)
+			sg_chain(prv, SG_MAX_SINGLE_ALLOC, sg);
+		else
+			table->sgl = sg;
+
+		/*
+		 * If no more entries after this one, mark the end
+		 */
+		if (!left)
+			sg_mark_end(&sg[sg_size - 1]);
+
+		/*
+		 * only really needed for mempool backed sg allocations (like
+		 * SCSI), a possible improvement here would be to pass the
+		 * table pointer into the allocator and let that clear these
+		 * flags
+		 */
+		gfp_mask &= ~__GFP_WAIT;
+		gfp_mask |= __GFP_HIGH;
+		prv = sg;
+	} while (left);
+
+	return 0;
+}
+EXPORT_SYMBOL(__sg_alloc_table);
+
+/**
+ * sg_alloc_table - Allocate and initialize an sg table
+ * @table:	The sg table header to use
+ * @nents:	Number of entries in sg list
+ * @gfp_mask:	GFP allocation mask
+ *
+ *  Description:
+ *    Allocate and initialize an sg table. If @nents@ is larger than
+ *    SG_MAX_SINGLE_ALLOC a chained sg table will be setup.
+ *
+ **/
+int sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask)
+{
+	int ret;
+
+	ret = __sg_alloc_table(table, nents, gfp_mask, sg_kmalloc);
+	if (unlikely(ret))
+		__sg_free_table(table, sg_kfree);
+
+	return ret;
+}
+EXPORT_SYMBOL(sg_alloc_table);
-- 
cgit v1.2.3


From 5ed7959ede0936c55e50421a53f153b17080e876 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 15 Nov 2007 09:13:11 +0100
Subject: SG: Convert SCSI to use scatterlist helpers for sg chaining

Also change scsi_alloc_sgtable() to just return 0/failure, since it
maps to the command passed in. ->request_buffer is now no longer needed,
once drivers are adapted to use scsi_sglist() it can be killed.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 drivers/scsi/scsi_lib.c     | 138 +++++++-------------------------------------
 drivers/scsi/scsi_tgt_lib.c |   3 +-
 include/scsi/scsi_cmnd.h    |   7 +--
 3 files changed, 24 insertions(+), 124 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 4cf902efbdbf..3b5121c4c081 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -739,138 +739,41 @@ static inline unsigned int scsi_sgtable_index(unsigned short nents)
 	return index;
 }
 
-struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+static void scsi_sg_free(struct scatterlist *sgl, unsigned int nents)
 {
 	struct scsi_host_sg_pool *sgp;
-	struct scatterlist *sgl, *prev, *ret;
-	unsigned int index;
-	int this, left;
-
-	BUG_ON(!cmd->use_sg);
-
-	left = cmd->use_sg;
-	ret = prev = NULL;
-	do {
-		this = left;
-		if (this > SCSI_MAX_SG_SEGMENTS) {
-			this = SCSI_MAX_SG_SEGMENTS - 1;
-			index = SG_MEMPOOL_NR - 1;
-		} else
-			index = scsi_sgtable_index(this);
 
-		left -= this;
-
-		sgp = scsi_sg_pools + index;
+	sgp = scsi_sg_pools + scsi_sgtable_index(nents);
+	mempool_free(sgl, sgp->pool);
+}
 
-		sgl = mempool_alloc(sgp->pool, gfp_mask);
-		if (unlikely(!sgl))
-			goto enomem;
+static struct scatterlist *scsi_sg_alloc(unsigned int nents, gfp_t gfp_mask)
+{
+	struct scsi_host_sg_pool *sgp;
 
-		sg_init_table(sgl, sgp->size);
+	sgp = scsi_sg_pools + scsi_sgtable_index(nents);
+	return mempool_alloc(sgp->pool, gfp_mask);
+}
 
-		/*
-		 * first loop through, set initial index and return value
-		 */
-		if (!ret)
-			ret = sgl;
+int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+{
+	int ret;
 
-		/*
-		 * chain previous sglist, if any. we know the previous
-		 * sglist must be the biggest one, or we would not have
-		 * ended up doing another loop.
-		 */
-		if (prev)
-			sg_chain(prev, SCSI_MAX_SG_SEGMENTS, sgl);
+	BUG_ON(!cmd->use_sg);
 
-		/*
-		 * if we have nothing left, mark the last segment as
-		 * end-of-list
-		 */
-		if (!left)
-			sg_mark_end(&sgl[this - 1]);
+	ret = __sg_alloc_table(&cmd->sg_table, cmd->use_sg, gfp_mask, scsi_sg_alloc);
+	if (unlikely(ret))
+		__sg_free_table(&cmd->sg_table, scsi_sg_free);
 
-		/*
-		 * don't allow subsequent mempool allocs to sleep, it would
-		 * violate the mempool principle.
-		 */
-		gfp_mask &= ~__GFP_WAIT;
-		gfp_mask |= __GFP_HIGH;
-		prev = sgl;
-	} while (left);
-
-	/*
-	 * ->use_sg may get modified after dma mapping has potentially
-	 * shrunk the number of segments, so keep a copy of it for free.
-	 */
-	cmd->__use_sg = cmd->use_sg;
+	cmd->request_buffer = cmd->sg_table.sgl;
 	return ret;
-enomem:
-	if (ret) {
-		/*
-		 * Free entries chained off ret. Since we were trying to
-		 * allocate another sglist, we know that all entries are of
-		 * the max size.
-		 */
-		sgp = scsi_sg_pools + SG_MEMPOOL_NR - 1;
-		prev = ret;
-		ret = &ret[SCSI_MAX_SG_SEGMENTS - 1];
-
-		while ((sgl = sg_chain_ptr(ret)) != NULL) {
-			ret = &sgl[SCSI_MAX_SG_SEGMENTS - 1];
-			mempool_free(sgl, sgp->pool);
-		}
-
-		mempool_free(prev, sgp->pool);
-	}
-	return NULL;
 }
 
 EXPORT_SYMBOL(scsi_alloc_sgtable);
 
 void scsi_free_sgtable(struct scsi_cmnd *cmd)
 {
-	struct scatterlist *sgl = cmd->request_buffer;
-	struct scsi_host_sg_pool *sgp;
-
-	/*
-	 * if this is the biggest size sglist, check if we have
-	 * chained parts we need to free
-	 */
-	if (cmd->__use_sg > SCSI_MAX_SG_SEGMENTS) {
-		unsigned short this, left;
-		struct scatterlist *next;
-		unsigned int index;
-
-		left = cmd->__use_sg - (SCSI_MAX_SG_SEGMENTS - 1);
-		next = sg_chain_ptr(&sgl[SCSI_MAX_SG_SEGMENTS - 1]);
-		while (left && next) {
-			sgl = next;
-			this = left;
-			if (this > SCSI_MAX_SG_SEGMENTS) {
-				this = SCSI_MAX_SG_SEGMENTS - 1;
-				index = SG_MEMPOOL_NR - 1;
-			} else
-				index = scsi_sgtable_index(this);
-
-			left -= this;
-
-			sgp = scsi_sg_pools + index;
-
-			if (left)
-				next = sg_chain_ptr(&sgl[sgp->size - 1]);
-
-			mempool_free(sgl, sgp->pool);
-		}
-
-		/*
-		 * Restore original, will be freed below
-		 */
-		sgl = cmd->request_buffer;
-		sgp = scsi_sg_pools + SG_MEMPOOL_NR - 1;
-	} else
-		sgp = scsi_sg_pools + scsi_sgtable_index(cmd->__use_sg);
-
-	mempool_free(sgl, sgp->pool);
+	__sg_free_table(&cmd->sg_table, scsi_sg_free);
 }
 
 EXPORT_SYMBOL(scsi_free_sgtable);
@@ -1120,8 +1023,7 @@ static int scsi_init_io(struct scsi_cmnd *cmd)
 	/*
 	 * If sg table allocation fails, requeue request later.
 	 */
-	cmd->request_buffer = scsi_alloc_sgtable(cmd, GFP_ATOMIC);
-	if (unlikely(!cmd->request_buffer)) {
+	if (unlikely(scsi_alloc_sgtable(cmd, GFP_ATOMIC))) {
 		scsi_unprep_request(req);
 		return BLKPREP_DEFER;
 	}
diff --git a/drivers/scsi/scsi_tgt_lib.c b/drivers/scsi/scsi_tgt_lib.c
index 93ece8f4e5de..01e03f3f6ffa 100644
--- a/drivers/scsi/scsi_tgt_lib.c
+++ b/drivers/scsi/scsi_tgt_lib.c
@@ -359,8 +359,7 @@ static int scsi_tgt_init_cmd(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 	int count;
 
 	cmd->use_sg = rq->nr_phys_segments;
-	cmd->request_buffer = scsi_alloc_sgtable(cmd, gfp_mask);
-	if (!cmd->request_buffer)
+	if (scsi_alloc_sgtable(cmd, gfp_mask))
 		return -ENOMEM;
 
 	cmd->request_bufflen = rq->data_len;
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index abd7479ff452..a457fca66f61 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -8,7 +8,6 @@
 #include <linux/scatterlist.h>
 
 struct request;
-struct scatterlist;
 struct Scsi_Host;
 struct scsi_device;
 
@@ -68,8 +67,8 @@ struct scsi_cmnd {
 	void *request_buffer;		/* Actual requested buffer */
 
 	/* These elements define the operation we ultimately want to perform */
+	struct sg_table sg_table;
 	unsigned short use_sg;	/* Number of pieces of scatter-gather */
-	unsigned short __use_sg;
 
 	unsigned underflow;	/* Return error if less than
 				   this amount is transferred */
@@ -128,14 +127,14 @@ extern void *scsi_kmap_atomic_sg(struct scatterlist *sg, int sg_count,
 				 size_t *offset, size_t *len);
 extern void scsi_kunmap_atomic_sg(void *virt);
 
-extern struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *, gfp_t);
+extern int scsi_alloc_sgtable(struct scsi_cmnd *, gfp_t);
 extern void scsi_free_sgtable(struct scsi_cmnd *);
 
 extern int scsi_dma_map(struct scsi_cmnd *cmd);
 extern void scsi_dma_unmap(struct scsi_cmnd *cmd);
 
 #define scsi_sg_count(cmd) ((cmd)->use_sg)
-#define scsi_sglist(cmd) ((struct scatterlist *)(cmd)->request_buffer)
+#define scsi_sglist(cmd) ((cmd)->sg_table.sgl)
 #define scsi_bufflen(cmd) ((cmd)->request_bufflen)
 
 static inline void scsi_set_resid(struct scsi_cmnd *cmd, int resid)
-- 
cgit v1.2.3


From 336cdb4003200a90f4fc52a4e9ccc2baa570fffb Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:40:30 -0500
Subject: blk_end_request: add new request completion interface (take 4)

This patch adds 2 new interfaces for request completion:
  o blk_end_request()   : called without queue lock
  o __blk_end_request() : called with queue lock held

blk_end_request takes 'error' as an argument instead of 'uptodate',
which current end_that_request_* take.
The meanings of values are below and the value is used when bio is
completed.
    0 : success
  < 0 : error

Some device drivers call some generic functions below between
end_that_request_{first/chunk} and end_that_request_last().
  o add_disk_randomness()
  o blk_queue_end_tag()
  o blkdev_dequeue_request()
These are called in the blk_end_request interfaces as a part of
generic request completion.
So all device drivers become to call above functions.
To decide whether to call blkdev_dequeue_request(), blk_end_request
uses list_empty(&rq->queuelist) (blk_queued_rq() macro is added for it).
So drivers must re-initialize it using list_init() or so before calling
blk_end_request if drivers use it for its specific purpose.
(Currently, there is no driver which completes request without
 re-initializing the queuelist after used it.  So rq->queuelist
 can be used for the purpose above.)

"Normal" drivers can be converted to use blk_end_request()
in a standard way shown below.

 a) end_that_request_{chunk/first}
    spin_lock_irqsave()
    (add_disk_randomness(), blk_queue_end_tag(), blkdev_dequeue_request())
    end_that_request_last()
    spin_unlock_irqrestore()
    => blk_end_request()

 b) spin_lock_irqsave()
    end_that_request_{chunk/first}
    (add_disk_randomness(), blk_queue_end_tag(), blkdev_dequeue_request())
    end_that_request_last()
    spin_unlock_irqrestore()
    => spin_lock_irqsave()
       __blk_end_request()
       spin_unlock_irqsave()

 c) spin_lock_irqsave()
    (add_disk_randomness(), blk_queue_end_tag(), blkdev_dequeue_request())
    end_that_request_last()
    spin_unlock_irqrestore()
    => blk_end_request()   or   spin_lock_irqsave()
                                __blk_end_request()
                                spin_unlock_irqrestore()

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h |  4 +++
 2 files changed, 100 insertions(+)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 3d0422f48453..5c01911af47c 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3791,6 +3791,102 @@ void end_request(struct request *req, int uptodate)
 }
 EXPORT_SYMBOL(end_request);
 
+static void complete_request(struct request *rq, int error)
+{
+	/*
+	 * REMOVEME: This conversion is transitional and will be removed
+	 *           when old end_that_request_* are unexported.
+	 */
+	int uptodate = 1;
+	if (error)
+		uptodate = (error == -EIO) ? 0 : error;
+
+	if (blk_rq_tagged(rq))
+		blk_queue_end_tag(rq->q, rq);
+
+	if (blk_queued_rq(rq))
+		blkdev_dequeue_request(rq);
+
+	end_that_request_last(rq, uptodate);
+}
+
+/**
+ * blk_end_request - Helper function for drivers to complete the request.
+ * @rq:       the request being processed
+ * @error:    0 for success, < 0 for error
+ * @nr_bytes: number of bytes to complete
+ *
+ * Description:
+ *     Ends I/O on a number of bytes attached to @rq.
+ *     If @rq has leftover, sets it up for the next range of segments.
+ *
+ * Return:
+ *     0 - we are done with this request
+ *     1 - still buffers pending for this request
+ **/
+int blk_end_request(struct request *rq, int error, int nr_bytes)
+{
+	struct request_queue *q = rq->q;
+	unsigned long flags = 0UL;
+	/*
+	 * REMOVEME: This conversion is transitional and will be removed
+	 *           when old end_that_request_* are unexported.
+	 */
+	int uptodate = 1;
+	if (error)
+		uptodate = (error == -EIO) ? 0 : error;
+
+	if (blk_fs_request(rq) || blk_pc_request(rq)) {
+		if (__end_that_request_first(rq, uptodate, nr_bytes))
+			return 1;
+	}
+
+	add_disk_randomness(rq->rq_disk);
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	complete_request(rq, error);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(blk_end_request);
+
+/**
+ * __blk_end_request - Helper function for drivers to complete the request.
+ * @rq:       the request being processed
+ * @error:    0 for success, < 0 for error
+ * @nr_bytes: number of bytes to complete
+ *
+ * Description:
+ *     Must be called with queue lock held unlike blk_end_request().
+ *
+ * Return:
+ *     0 - we are done with this request
+ *     1 - still buffers pending for this request
+ **/
+int __blk_end_request(struct request *rq, int error, int nr_bytes)
+{
+	/*
+	 * REMOVEME: This conversion is transitional and will be removed
+	 *           when old end_that_request_* are unexported.
+	 */
+	int uptodate = 1;
+	if (error)
+		uptodate = (error == -EIO) ? 0 : error;
+
+	if (blk_fs_request(rq) || blk_pc_request(rq)) {
+		if (__end_that_request_first(rq, uptodate, nr_bytes))
+			return 1;
+	}
+
+	add_disk_randomness(rq->rq_disk);
+
+	complete_request(rq, error);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__blk_end_request);
+
 static void blk_rq_bio_prep(struct request_queue *q, struct request *rq,
 			    struct bio *bio)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 49b7a4c31a6d..3b212f02db8d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -537,6 +537,8 @@ enum {
 #define blk_fua_rq(rq)		((rq)->cmd_flags & REQ_FUA)
 #define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
 #define blk_empty_barrier(rq)	(blk_barrier_rq(rq) && blk_fs_request(rq) && !(rq)->hard_nr_sectors)
+/* rq->queuelist of dequeued request must be list_empty() */
+#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
 
 #define list_entry_rq(ptr)	list_entry((ptr), struct request, queuelist)
 
@@ -724,6 +726,8 @@ static inline void blk_run_address_space(struct address_space *mapping)
  * for parts of the original function. This prevents
  * code duplication in drivers.
  */
+extern int blk_end_request(struct request *rq, int error, int nr_bytes);
+extern int __blk_end_request(struct request *rq, int error, int nr_bytes);
 extern int end_that_request_first(struct request *, int, int);
 extern int end_that_request_chunk(struct request *, int, int);
 extern void end_that_request_last(struct request *, int);
-- 
cgit v1.2.3


From 3b11313a6c2a42425bf06e92528bda6affd58dec Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:41:17 -0500
Subject: blk_end_request: add/export functions to get request size (take 4)

This patch adds/exports functions to get the size of request in bytes.
They are useful because blk_end_request interfaces take bytes
as a completed I/O size instead of sectors.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 25 ++++++++++++++++++++++---
 include/linux/blkdev.h |  8 ++++++++
 2 files changed, 30 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 5c01911af47c..8b2b2509f60e 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3723,13 +3723,32 @@ static inline void __end_request(struct request *rq, int uptodate,
 	}
 }
 
-static unsigned int rq_byte_size(struct request *rq)
+/**
+ * blk_rq_bytes - Returns bytes left to complete in the entire request
+ **/
+unsigned int blk_rq_bytes(struct request *rq)
 {
 	if (blk_fs_request(rq))
 		return rq->hard_nr_sectors << 9;
 
 	return rq->data_len;
 }
+EXPORT_SYMBOL_GPL(blk_rq_bytes);
+
+/**
+ * blk_rq_cur_bytes - Returns bytes left to complete in the current segment
+ **/
+unsigned int blk_rq_cur_bytes(struct request *rq)
+{
+	if (blk_fs_request(rq))
+		return rq->current_nr_sectors << 9;
+
+	if (rq->bio)
+		return rq->bio->bi_size;
+
+	return rq->data_len;
+}
+EXPORT_SYMBOL_GPL(blk_rq_cur_bytes);
 
 /**
  * end_queued_request - end all I/O on a queued request
@@ -3744,7 +3763,7 @@ static unsigned int rq_byte_size(struct request *rq)
  **/
 void end_queued_request(struct request *rq, int uptodate)
 {
-	__end_request(rq, uptodate, rq_byte_size(rq), 1);
+	__end_request(rq, uptodate, blk_rq_bytes(rq), 1);
 }
 EXPORT_SYMBOL(end_queued_request);
 
@@ -3761,7 +3780,7 @@ EXPORT_SYMBOL(end_queued_request);
  **/
 void end_dequeued_request(struct request *rq, int uptodate)
 {
-	__end_request(rq, uptodate, rq_byte_size(rq), 0);
+	__end_request(rq, uptodate, blk_rq_bytes(rq), 0);
 }
 EXPORT_SYMBOL(end_dequeued_request);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 3b212f02db8d..aa2341df7932 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -736,6 +736,14 @@ extern void end_queued_request(struct request *, int);
 extern void end_dequeued_request(struct request *, int);
 extern void blk_complete_request(struct request *);
 
+/*
+ * blk_end_request() takes bytes instead of sectors as a complete size.
+ * blk_rq_bytes() returns bytes left to complete in the entire request.
+ * blk_rq_cur_bytes() returns bytes left to complete in the current segment.
+ */
+extern unsigned int blk_rq_bytes(struct request *rq);
+extern unsigned int blk_rq_cur_bytes(struct request *rq);
+
 /*
  * end_that_request_first/chunk() takes an uptodate argument. we account
  * any value <= as an io error. 0 means -EIO for compatability reasons,
-- 
cgit v1.2.3


From e19a3ab058fe91c8c54d43dc56dccf7eb386478e Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:51:02 -0500
Subject: blk_end_request: add callback feature (take 4)

This patch adds a variant of the interface, blk_end_request_callback(),
which has driver callback feature.

Drivers may need to do special works between end_that_request_first()
and end_that_request_last().
For such drivers, blk_end_request_callback() allows it to pass
a callback function which is called between end_that_request_first()
and end_that_request_last().

This interface is only for fallback of other blk_end_request interfaces.
Drivers should avoid their tricky behaviors and use other interfaces
as much as possible.

Currently, only one driver, ide-cd, needs this interface.
So this interface should/will be removed, after the driver removes
such tricky behaviors.

o ide-cd (cdrom_newpc_intr())
  In PIO mode, cdrom_newpc_intr() needs to defer end_that_request_last()
  until the device clears DRQ_STAT and raises an interrupt after
  end_that_request_first().
  So end_that_request_first() and end_that_request_last() are called
  separately in cdrom_newpc_intr().

  This means blk_end_request_callback() has to return without
  completing request even if no leftover in the request.
  To satisfy the requirement, callback function has return value
  so that drivers can tell blk_end_request_callback() to return
  without completing request.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 72 +++++++++++++++++++++++++++++++++++++++++++++-----
 include/linux/blkdev.h |  2 ++
 2 files changed, 68 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index fb951198c70e..919e4baf478b 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3825,10 +3825,14 @@ static void complete_request(struct request *rq, int error)
 }
 
 /**
- * blk_end_request - Helper function for drivers to complete the request.
- * @rq:       the request being processed
- * @error:    0 for success, < 0 for error
- * @nr_bytes: number of bytes to complete
+ * blk_end_io - Generic end_io function to complete a request.
+ * @rq:           the request being processed
+ * @error:        0 for success, < 0 for error
+ * @nr_bytes:     number of bytes to complete
+ * @drv_callback: function called between completion of bios in the request
+ *                and completion of the request.
+ *                If the callback returns non 0, this helper returns without
+ *                completion of the request.
  *
  * Description:
  *     Ends I/O on a number of bytes attached to @rq.
@@ -3836,9 +3840,10 @@ static void complete_request(struct request *rq, int error)
  *
  * Return:
  *     0 - we are done with this request
- *     1 - still buffers pending for this request
+ *     1 - this request is not freed yet, it still has pending buffers.
  **/
-int blk_end_request(struct request *rq, int error, int nr_bytes)
+static int blk_end_io(struct request *rq, int error, int nr_bytes,
+		      int (drv_callback)(struct request *))
 {
 	struct request_queue *q = rq->q;
 	unsigned long flags = 0UL;
@@ -3855,6 +3860,10 @@ int blk_end_request(struct request *rq, int error, int nr_bytes)
 			return 1;
 	}
 
+	/* Special feature for tricky drivers */
+	if (drv_callback && drv_callback(rq))
+		return 1;
+
 	add_disk_randomness(rq->rq_disk);
 
 	spin_lock_irqsave(q->queue_lock, flags);
@@ -3863,6 +3872,25 @@ int blk_end_request(struct request *rq, int error, int nr_bytes)
 
 	return 0;
 }
+
+/**
+ * blk_end_request - Helper function for drivers to complete the request.
+ * @rq:       the request being processed
+ * @error:    0 for success, < 0 for error
+ * @nr_bytes: number of bytes to complete
+ *
+ * Description:
+ *     Ends I/O on a number of bytes attached to @rq.
+ *     If @rq has leftover, sets it up for the next range of segments.
+ *
+ * Return:
+ *     0 - we are done with this request
+ *     1 - still buffers pending for this request
+ **/
+int blk_end_request(struct request *rq, int error, int nr_bytes)
+{
+	return blk_end_io(rq, error, nr_bytes, NULL);
+}
 EXPORT_SYMBOL_GPL(blk_end_request);
 
 /**
@@ -3901,6 +3929,38 @@ int __blk_end_request(struct request *rq, int error, int nr_bytes)
 }
 EXPORT_SYMBOL_GPL(__blk_end_request);
 
+/**
+ * blk_end_request_callback - Special helper function for tricky drivers
+ * @rq:           the request being processed
+ * @error:        0 for success, < 0 for error
+ * @nr_bytes:     number of bytes to complete
+ * @drv_callback: function called between completion of bios in the request
+ *                and completion of the request.
+ *                If the callback returns non 0, this helper returns without
+ *                completion of the request.
+ *
+ * Description:
+ *     Ends I/O on a number of bytes attached to @rq.
+ *     If @rq has leftover, sets it up for the next range of segments.
+ *
+ *     This special helper function is used only for existing tricky drivers.
+ *     (e.g. cdrom_newpc_intr() of ide-cd)
+ *     This interface will be removed when such drivers are rewritten.
+ *     Don't use this interface in other places anymore.
+ *
+ * Return:
+ *     0 - we are done with this request
+ *     1 - this request is not freed yet.
+ *         this request still has pending buffers or
+ *         the driver doesn't want to finish this request yet.
+ **/
+int blk_end_request_callback(struct request *rq, int error, int nr_bytes,
+			     int (drv_callback)(struct request *))
+{
+	return blk_end_io(rq, error, nr_bytes, drv_callback);
+}
+EXPORT_SYMBOL_GPL(blk_end_request_callback);
+
 static void blk_rq_bio_prep(struct request_queue *q, struct request *rq,
 			    struct bio *bio)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index aa2341df7932..63fe7542b3fa 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -734,6 +734,8 @@ extern void end_that_request_last(struct request *, int);
 extern void end_request(struct request *, int);
 extern void end_queued_request(struct request *, int);
 extern void end_dequeued_request(struct request *, int);
+extern int blk_end_request_callback(struct request *rq, int error, int nr_bytes,
+				    int (drv_callback)(struct request *));
 extern void blk_complete_request(struct request *);
 
 /*
-- 
cgit v1.2.3


From e3a04fe34a3ec81ddeddb6c73fb7299716cffbb0 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:51:46 -0500
Subject: blk_end_request: add bidi completion interface (take 4)

This patch adds a variant of the interface, blk_end_bidi_request(),
which completes a bidi request.

Bidi request must be completed as a whole, both rq and rq->next_rq
at once.  So the interface has 2 arguments for completion size.

As for ->end_io, only rq->end_io is called (rq->next_rq->end_io is not
called).  So if special completion handling is needed, the handler
must be set to rq->end_io.
And the handler must take care of freeing next_rq too, since
the interface doesn't care of it if rq->end_io is not NULL.

Cc: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 40 +++++++++++++++++++++++++++++++++++-----
 include/linux/blkdev.h |  2 ++
 2 files changed, 37 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 919e4baf478b..4889f7a8c2b7 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3821,6 +3821,9 @@ static void complete_request(struct request *rq, int error)
 	if (blk_queued_rq(rq))
 		blkdev_dequeue_request(rq);
 
+	if (blk_bidi_rq(rq) && !rq->end_io)
+		__blk_put_request(rq->next_rq->q, rq->next_rq);
+
 	end_that_request_last(rq, uptodate);
 }
 
@@ -3828,14 +3831,15 @@ static void complete_request(struct request *rq, int error)
  * blk_end_io - Generic end_io function to complete a request.
  * @rq:           the request being processed
  * @error:        0 for success, < 0 for error
- * @nr_bytes:     number of bytes to complete
+ * @nr_bytes:     number of bytes to complete @rq
+ * @bidi_bytes:   number of bytes to complete @rq->next_rq
  * @drv_callback: function called between completion of bios in the request
  *                and completion of the request.
  *                If the callback returns non 0, this helper returns without
  *                completion of the request.
  *
  * Description:
- *     Ends I/O on a number of bytes attached to @rq.
+ *     Ends I/O on a number of bytes attached to @rq and @rq->next_rq.
  *     If @rq has leftover, sets it up for the next range of segments.
  *
  * Return:
@@ -3843,7 +3847,7 @@ static void complete_request(struct request *rq, int error)
  *     1 - this request is not freed yet, it still has pending buffers.
  **/
 static int blk_end_io(struct request *rq, int error, int nr_bytes,
-		      int (drv_callback)(struct request *))
+		      int bidi_bytes, int (drv_callback)(struct request *))
 {
 	struct request_queue *q = rq->q;
 	unsigned long flags = 0UL;
@@ -3858,6 +3862,11 @@ static int blk_end_io(struct request *rq, int error, int nr_bytes,
 	if (blk_fs_request(rq) || blk_pc_request(rq)) {
 		if (__end_that_request_first(rq, uptodate, nr_bytes))
 			return 1;
+
+		/* Bidi request must be completed as a whole */
+		if (blk_bidi_rq(rq) &&
+		    __end_that_request_first(rq->next_rq, uptodate, bidi_bytes))
+			return 1;
 	}
 
 	/* Special feature for tricky drivers */
@@ -3889,7 +3898,7 @@ static int blk_end_io(struct request *rq, int error, int nr_bytes,
  **/
 int blk_end_request(struct request *rq, int error, int nr_bytes)
 {
-	return blk_end_io(rq, error, nr_bytes, NULL);
+	return blk_end_io(rq, error, nr_bytes, 0, NULL);
 }
 EXPORT_SYMBOL_GPL(blk_end_request);
 
@@ -3929,6 +3938,27 @@ int __blk_end_request(struct request *rq, int error, int nr_bytes)
 }
 EXPORT_SYMBOL_GPL(__blk_end_request);
 
+/**
+ * blk_end_bidi_request - Helper function for drivers to complete bidi request.
+ * @rq:         the bidi request being processed
+ * @error:      0 for success, < 0 for error
+ * @nr_bytes:   number of bytes to complete @rq
+ * @bidi_bytes: number of bytes to complete @rq->next_rq
+ *
+ * Description:
+ *     Ends I/O on a number of bytes attached to @rq and @rq->next_rq.
+ *
+ * Return:
+ *     0 - we are done with this request
+ *     1 - still buffers pending for this request
+ **/
+int blk_end_bidi_request(struct request *rq, int error, int nr_bytes,
+			 int bidi_bytes)
+{
+	return blk_end_io(rq, error, nr_bytes, bidi_bytes, NULL);
+}
+EXPORT_SYMBOL_GPL(blk_end_bidi_request);
+
 /**
  * blk_end_request_callback - Special helper function for tricky drivers
  * @rq:           the request being processed
@@ -3957,7 +3987,7 @@ EXPORT_SYMBOL_GPL(__blk_end_request);
 int blk_end_request_callback(struct request *rq, int error, int nr_bytes,
 			     int (drv_callback)(struct request *))
 {
-	return blk_end_io(rq, error, nr_bytes, drv_callback);
+	return blk_end_io(rq, error, nr_bytes, 0, drv_callback);
 }
 EXPORT_SYMBOL_GPL(blk_end_request_callback);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 63fe7542b3fa..029b7097f9e5 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -728,6 +728,8 @@ static inline void blk_run_address_space(struct address_space *mapping)
  */
 extern int blk_end_request(struct request *rq, int error, int nr_bytes);
 extern int __blk_end_request(struct request *rq, int error, int nr_bytes);
+extern int blk_end_bidi_request(struct request *rq, int error, int nr_bytes,
+				int bidi_bytes);
 extern int end_that_request_first(struct request *, int, int);
 extern int end_that_request_chunk(struct request *, int, int);
 extern void end_that_request_last(struct request *, int);
-- 
cgit v1.2.3


From 3bcddeac1c4c7e6fb90531b80f236b1a05dfe514 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:52:28 -0500
Subject: blk_end_request: remove/unexport end_that_request_* (take 4)

This patch removes the following functions:
  o end_that_request_first()
  o end_that_request_chunk()
and stops exporting the functions below:
  o end_that_request_last()

Cc: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 61 +++++++++++++-------------------------------------
 include/linux/blkdev.h | 13 +++++------
 2 files changed, 20 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 4889f7a8c2b7..4bd1803919ce 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3432,6 +3432,20 @@ static void blk_recalc_rq_sectors(struct request *rq, int nsect)
 	}
 }
 
+/**
+ * __end_that_request_first - end I/O on a request
+ * @req:      the request being processed
+ * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
+ * @nr_bytes: number of bytes to complete
+ *
+ * Description:
+ *     Ends I/O on a number of bytes attached to @req, and sets it up
+ *     for the next range of segments (if any) in the cluster.
+ *
+ * Return:
+ *     0 - we are done with this request, call end_that_request_last()
+ *     1 - still buffers pending for this request
+ **/
 static int __end_that_request_first(struct request *req, int uptodate,
 				    int nr_bytes)
 {
@@ -3548,49 +3562,6 @@ static int __end_that_request_first(struct request *req, int uptodate,
 	return 1;
 }
 
-/**
- * end_that_request_first - end I/O on a request
- * @req:      the request being processed
- * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
- * @nr_sectors: number of sectors to end I/O on
- *
- * Description:
- *     Ends I/O on a number of sectors attached to @req, and sets it up
- *     for the next range of segments (if any) in the cluster.
- *
- * Return:
- *     0 - we are done with this request, call end_that_request_last()
- *     1 - still buffers pending for this request
- **/
-int end_that_request_first(struct request *req, int uptodate, int nr_sectors)
-{
-	return __end_that_request_first(req, uptodate, nr_sectors << 9);
-}
-
-EXPORT_SYMBOL(end_that_request_first);
-
-/**
- * end_that_request_chunk - end I/O on a request
- * @req:      the request being processed
- * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
- * @nr_bytes: number of bytes to complete
- *
- * Description:
- *     Ends I/O on a number of bytes attached to @req, and sets it up
- *     for the next range of segments (if any). Like end_that_request_first(),
- *     but deals with bytes instead of sectors.
- *
- * Return:
- *     0 - we are done with this request, call end_that_request_last()
- *     1 - still buffers pending for this request
- **/
-int end_that_request_chunk(struct request *req, int uptodate, int nr_bytes)
-{
-	return __end_that_request_first(req, uptodate, nr_bytes);
-}
-
-EXPORT_SYMBOL(end_that_request_chunk);
-
 /*
  * splice the completion data to a local structure and hand off to
  * process_completion_queue() to complete the requests
@@ -3670,7 +3641,7 @@ EXPORT_SYMBOL(blk_complete_request);
 /*
  * queue lock must be held
  */
-void end_that_request_last(struct request *req, int uptodate)
+static void end_that_request_last(struct request *req, int uptodate)
 {
 	struct gendisk *disk = req->rq_disk;
 	int error;
@@ -3705,8 +3676,6 @@ void end_that_request_last(struct request *req, int uptodate)
 		__blk_put_request(req->q, req);
 }
 
-EXPORT_SYMBOL(end_that_request_last);
-
 static inline void __end_request(struct request *rq, int uptodate,
 				 unsigned int nr_bytes)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 029b7097f9e5..0c39ac75bed4 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -718,21 +718,18 @@ static inline void blk_run_address_space(struct address_space *mapping)
 }
 
 /*
- * end_request() and friends. Must be called with the request queue spinlock
- * acquired. All functions called within end_request() _must_be_ atomic.
+ * blk_end_request() and friends.
+ * __blk_end_request() and end_request() must be called with
+ * the request queue spinlock acquired.
  *
  * Several drivers define their own end_request and call
- * end_that_request_first() and end_that_request_last()
- * for parts of the original function. This prevents
- * code duplication in drivers.
+ * blk_end_request() for parts of the original function.
+ * This prevents code duplication in drivers.
  */
 extern int blk_end_request(struct request *rq, int error, int nr_bytes);
 extern int __blk_end_request(struct request *rq, int error, int nr_bytes);
 extern int blk_end_bidi_request(struct request *rq, int error, int nr_bytes,
 				int bidi_bytes);
-extern int end_that_request_first(struct request *, int, int);
-extern int end_that_request_chunk(struct request *, int, int);
-extern void end_that_request_last(struct request *, int);
 extern void end_request(struct request *, int);
 extern void end_queued_request(struct request *, int);
 extern void end_dequeued_request(struct request *, int);
-- 
cgit v1.2.3


From 5450d3e1d68f10be087f0855d8bad5458b50ecbe Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 11 Dec 2007 17:53:03 -0500
Subject: blk_end_request: cleanup 'uptodate' related code (take 4)

This patch converts 'uptodate' arguments of no longer exported
interfaces, end_that_request_first/last, to 'error', and removes
internal conversions for it in blk_end_request interfaces.

Also, this patch removes no longer needed end_io_error().

Cc: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c      | 56 ++++++++------------------------------------------
 include/linux/blkdev.h |  8 --------
 2 files changed, 9 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 4bd1803919ce..f5e091bc027d 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3435,7 +3435,7 @@ static void blk_recalc_rq_sectors(struct request *rq, int nsect)
 /**
  * __end_that_request_first - end I/O on a request
  * @req:      the request being processed
- * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
+ * @error:    0 for success, < 0 for error
  * @nr_bytes: number of bytes to complete
  *
  * Description:
@@ -3446,21 +3446,14 @@ static void blk_recalc_rq_sectors(struct request *rq, int nsect)
  *     0 - we are done with this request, call end_that_request_last()
  *     1 - still buffers pending for this request
  **/
-static int __end_that_request_first(struct request *req, int uptodate,
+static int __end_that_request_first(struct request *req, int error,
 				    int nr_bytes)
 {
-	int total_bytes, bio_nbytes, error, next_idx = 0;
+	int total_bytes, bio_nbytes, next_idx = 0;
 	struct bio *bio;
 
 	blk_add_trace_rq(req->q, req, BLK_TA_COMPLETE);
 
-	/*
-	 * extend uptodate bool to allow < 0 value to be direct io error
-	 */
-	error = 0;
-	if (end_io_error(uptodate))
-		error = !uptodate ? -EIO : uptodate;
-
 	/*
 	 * for a REQ_BLOCK_PC request, we want to carry any eventual
 	 * sense key with us all the way through
@@ -3468,7 +3461,7 @@ static int __end_that_request_first(struct request *req, int uptodate,
 	if (!blk_pc_request(req))
 		req->errors = 0;
 
-	if (!uptodate) {
+	if (error) {
 		if (blk_fs_request(req) && !(req->cmd_flags & REQ_QUIET))
 			printk("end_request: I/O error, dev %s, sector %llu\n",
 				req->rq_disk ? req->rq_disk->disk_name : "?",
@@ -3641,17 +3634,9 @@ EXPORT_SYMBOL(blk_complete_request);
 /*
  * queue lock must be held
  */
-static void end_that_request_last(struct request *req, int uptodate)
+static void end_that_request_last(struct request *req, int error)
 {
 	struct gendisk *disk = req->rq_disk;
-	int error;
-
-	/*
-	 * extend uptodate bool to allow < 0 value to be direct io error
-	 */
-	error = 0;
-	if (end_io_error(uptodate))
-		error = !uptodate ? -EIO : uptodate;
 
 	if (unlikely(laptop_mode) && blk_fs_request(req))
 		laptop_io_completion();
@@ -3776,14 +3761,6 @@ EXPORT_SYMBOL(end_request);
 
 static void complete_request(struct request *rq, int error)
 {
-	/*
-	 * REMOVEME: This conversion is transitional and will be removed
-	 *           when old end_that_request_* are unexported.
-	 */
-	int uptodate = 1;
-	if (error)
-		uptodate = (error == -EIO) ? 0 : error;
-
 	if (blk_rq_tagged(rq))
 		blk_queue_end_tag(rq->q, rq);
 
@@ -3793,7 +3770,7 @@ static void complete_request(struct request *rq, int error)
 	if (blk_bidi_rq(rq) && !rq->end_io)
 		__blk_put_request(rq->next_rq->q, rq->next_rq);
 
-	end_that_request_last(rq, uptodate);
+	end_that_request_last(rq, error);
 }
 
 /**
@@ -3820,21 +3797,14 @@ static int blk_end_io(struct request *rq, int error, int nr_bytes,
 {
 	struct request_queue *q = rq->q;
 	unsigned long flags = 0UL;
-	/*
-	 * REMOVEME: This conversion is transitional and will be removed
-	 *           when old end_that_request_* are unexported.
-	 */
-	int uptodate = 1;
-	if (error)
-		uptodate = (error == -EIO) ? 0 : error;
 
 	if (blk_fs_request(rq) || blk_pc_request(rq)) {
-		if (__end_that_request_first(rq, uptodate, nr_bytes))
+		if (__end_that_request_first(rq, error, nr_bytes))
 			return 1;
 
 		/* Bidi request must be completed as a whole */
 		if (blk_bidi_rq(rq) &&
-		    __end_that_request_first(rq->next_rq, uptodate, bidi_bytes))
+		    __end_that_request_first(rq->next_rq, error, bidi_bytes))
 			return 1;
 	}
 
@@ -3886,16 +3856,8 @@ EXPORT_SYMBOL_GPL(blk_end_request);
  **/
 int __blk_end_request(struct request *rq, int error, int nr_bytes)
 {
-	/*
-	 * REMOVEME: This conversion is transitional and will be removed
-	 *           when old end_that_request_* are unexported.
-	 */
-	int uptodate = 1;
-	if (error)
-		uptodate = (error == -EIO) ? 0 : error;
-
 	if (blk_fs_request(rq) || blk_pc_request(rq)) {
-		if (__end_that_request_first(rq, uptodate, nr_bytes))
+		if (__end_that_request_first(rq, error, nr_bytes))
 			return 1;
 	}
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0c39ac75bed4..0ea82d222046 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -745,14 +745,6 @@ extern void blk_complete_request(struct request *);
 extern unsigned int blk_rq_bytes(struct request *rq);
 extern unsigned int blk_rq_cur_bytes(struct request *rq);
 
-/*
- * end_that_request_first/chunk() takes an uptodate argument. we account
- * any value <= as an io error. 0 means -EIO for compatability reasons,
- * any other < 0 value is the direct error type. An uptodate value of
- * 1 indicates successful io completion
- */
-#define end_io_error(uptodate)	(unlikely((uptodate) <= 0))
-
 static inline void blkdev_dequeue_request(struct request *req)
 {
 	elv_dequeue_request(req->q, req);
-- 
cgit v1.2.3


From fd0928df98b9578be8a786ac0cb78a47a5e17a20 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 24 Jan 2008 08:52:45 +0100
Subject: ioprio: move io priority from task_struct to io_context

This is where it belongs and then it doesn't take up space for a
process that doesn't do IO.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/cfq-iosched.c       | 34 ++++++++++----------
 block/ll_rw_blk.c         | 30 ++++++++++++------
 fs/ioprio.c               | 29 ++++++++++++-----
 include/linux/blkdev.h    | 81 ++++-------------------------------------------
 include/linux/init_task.h |  1 -
 include/linux/iocontext.h | 79 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/ioprio.h    | 13 ++++----
 include/linux/sched.h     |  1 -
 kernel/fork.c             | 32 ++++++++++++++++---
 9 files changed, 178 insertions(+), 122 deletions(-)
 create mode 100644 include/linux/iocontext.h

(limited to 'include')

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 13553e015d72..533af75329e6 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -199,7 +199,7 @@ CFQ_CFQQ_FNS(sync);
 
 static void cfq_dispatch_insert(struct request_queue *, struct request *);
 static struct cfq_queue *cfq_get_queue(struct cfq_data *, int,
-				       struct task_struct *, gfp_t);
+				       struct io_context *, gfp_t);
 static struct cfq_io_context *cfq_cic_rb_lookup(struct cfq_data *,
 						struct io_context *);
 
@@ -1273,7 +1273,7 @@ cfq_alloc_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 	return cic;
 }
 
-static void cfq_init_prio_data(struct cfq_queue *cfqq)
+static void cfq_init_prio_data(struct cfq_queue *cfqq, struct io_context *ioc)
 {
 	struct task_struct *tsk = current;
 	int ioprio_class;
@@ -1281,7 +1281,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq)
 	if (!cfq_cfqq_prio_changed(cfqq))
 		return;
 
-	ioprio_class = IOPRIO_PRIO_CLASS(tsk->ioprio);
+	ioprio_class = IOPRIO_PRIO_CLASS(ioc->ioprio);
 	switch (ioprio_class) {
 		default:
 			printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class);
@@ -1293,11 +1293,11 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq)
 			cfqq->ioprio_class = IOPRIO_CLASS_BE;
 			break;
 		case IOPRIO_CLASS_RT:
-			cfqq->ioprio = task_ioprio(tsk);
+			cfqq->ioprio = task_ioprio(ioc);
 			cfqq->ioprio_class = IOPRIO_CLASS_RT;
 			break;
 		case IOPRIO_CLASS_BE:
-			cfqq->ioprio = task_ioprio(tsk);
+			cfqq->ioprio = task_ioprio(ioc);
 			cfqq->ioprio_class = IOPRIO_CLASS_BE;
 			break;
 		case IOPRIO_CLASS_IDLE:
@@ -1330,8 +1330,7 @@ static inline void changed_ioprio(struct cfq_io_context *cic)
 	cfqq = cic->cfqq[ASYNC];
 	if (cfqq) {
 		struct cfq_queue *new_cfqq;
-		new_cfqq = cfq_get_queue(cfqd, ASYNC, cic->ioc->task,
-					 GFP_ATOMIC);
+		new_cfqq = cfq_get_queue(cfqd, ASYNC, cic->ioc, GFP_ATOMIC);
 		if (new_cfqq) {
 			cic->cfqq[ASYNC] = new_cfqq;
 			cfq_put_queue(cfqq);
@@ -1363,13 +1362,13 @@ static void cfq_ioc_set_ioprio(struct io_context *ioc)
 
 static struct cfq_queue *
 cfq_find_alloc_queue(struct cfq_data *cfqd, int is_sync,
-		     struct task_struct *tsk, gfp_t gfp_mask)
+		     struct io_context *ioc, gfp_t gfp_mask)
 {
 	struct cfq_queue *cfqq, *new_cfqq = NULL;
 	struct cfq_io_context *cic;
 
 retry:
-	cic = cfq_cic_rb_lookup(cfqd, tsk->io_context);
+	cic = cfq_cic_rb_lookup(cfqd, ioc);
 	/* cic always exists here */
 	cfqq = cic_to_cfqq(cic, is_sync);
 
@@ -1412,7 +1411,7 @@ retry:
 		cfq_mark_cfqq_prio_changed(cfqq);
 		cfq_mark_cfqq_queue_new(cfqq);
 
-		cfq_init_prio_data(cfqq);
+		cfq_init_prio_data(cfqq, ioc);
 	}
 
 	if (new_cfqq)
@@ -1439,11 +1438,11 @@ cfq_async_queue_prio(struct cfq_data *cfqd, int ioprio_class, int ioprio)
 }
 
 static struct cfq_queue *
-cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct task_struct *tsk,
+cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct io_context *ioc,
 	      gfp_t gfp_mask)
 {
-	const int ioprio = task_ioprio(tsk);
-	const int ioprio_class = task_ioprio_class(tsk);
+	const int ioprio = task_ioprio(ioc);
+	const int ioprio_class = task_ioprio_class(ioc);
 	struct cfq_queue **async_cfqq = NULL;
 	struct cfq_queue *cfqq = NULL;
 
@@ -1453,7 +1452,7 @@ cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct task_struct *tsk,
 	}
 
 	if (!cfqq) {
-		cfqq = cfq_find_alloc_queue(cfqd, is_sync, tsk, gfp_mask);
+		cfqq = cfq_find_alloc_queue(cfqd, is_sync, ioc, gfp_mask);
 		if (!cfqq)
 			return NULL;
 	}
@@ -1793,7 +1792,7 @@ static void cfq_insert_request(struct request_queue *q, struct request *rq)
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct cfq_queue *cfqq = RQ_CFQQ(rq);
 
-	cfq_init_prio_data(cfqq);
+	cfq_init_prio_data(cfqq, RQ_CIC(rq)->ioc);
 
 	cfq_add_rq_rb(rq);
 
@@ -1900,7 +1899,7 @@ static int cfq_may_queue(struct request_queue *q, int rw)
 
 	cfqq = cic_to_cfqq(cic, rw & REQ_RW_SYNC);
 	if (cfqq) {
-		cfq_init_prio_data(cfqq);
+		cfq_init_prio_data(cfqq, cic->ioc);
 		cfq_prio_boost(cfqq);
 
 		return __cfq_may_queue(cfqq);
@@ -1938,7 +1937,6 @@ static int
 cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
-	struct task_struct *tsk = current;
 	struct cfq_io_context *cic;
 	const int rw = rq_data_dir(rq);
 	const int is_sync = rq_is_sync(rq);
@@ -1956,7 +1954,7 @@ cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask)
 
 	cfqq = cic_to_cfqq(cic, is_sync);
 	if (!cfqq) {
-		cfqq = cfq_get_queue(cfqd, is_sync, tsk, gfp_mask);
+		cfqq = cfq_get_queue(cfqd, is_sync, cic->ioc, gfp_mask);
 
 		if (!cfqq)
 			goto queue_fail;
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 3d0422f48453..b9bb02e845cd 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3904,6 +3904,26 @@ void exit_io_context(void)
 	put_io_context(ioc);
 }
 
+struct io_context *alloc_io_context(gfp_t gfp_flags, int node)
+{
+	struct io_context *ret;
+
+	ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node);
+	if (ret) {
+		atomic_set(&ret->refcount, 1);
+		ret->task = current;
+		ret->ioprio_changed = 0;
+		ret->ioprio = 0;
+		ret->last_waited = jiffies; /* doesn't matter... */
+		ret->nr_batch_requests = 0; /* because this is 0 */
+		ret->aic = NULL;
+		ret->cic_root.rb_node = NULL;
+		ret->ioc_data = NULL;
+	}
+
+	return ret;
+}
+
 /*
  * If the current task has no IO context then create one and initialise it.
  * Otherwise, return its existing IO context.
@@ -3921,16 +3941,8 @@ static struct io_context *current_io_context(gfp_t gfp_flags, int node)
 	if (likely(ret))
 		return ret;
 
-	ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node);
+	ret = alloc_io_context(gfp_flags, node);
 	if (ret) {
-		atomic_set(&ret->refcount, 1);
-		ret->task = current;
-		ret->ioprio_changed = 0;
-		ret->last_waited = jiffies; /* doesn't matter... */
-		ret->nr_batch_requests = 0; /* because this is 0 */
-		ret->aic = NULL;
-		ret->cic_root.rb_node = NULL;
-		ret->ioc_data = NULL;
 		/* make sure set_task_ioprio() sees the settings above */
 		smp_wmb();
 		tsk->io_context = ret;
diff --git a/fs/ioprio.c b/fs/ioprio.c
index e4e01bc7f338..a7600401ecf7 100644
--- a/fs/ioprio.c
+++ b/fs/ioprio.c
@@ -41,18 +41,29 @@ static int set_task_ioprio(struct task_struct *task, int ioprio)
 		return err;
 
 	task_lock(task);
+	do {
+		ioc = task->io_context;
+		/* see wmb() in current_io_context() */
+		smp_read_barrier_depends();
+		if (ioc)
+			break;
 
-	task->ioprio = ioprio;
-
-	ioc = task->io_context;
-	/* see wmb() in current_io_context() */
-	smp_read_barrier_depends();
+		ioc = alloc_io_context(GFP_ATOMIC, -1);
+		if (!ioc) {
+			err = -ENOMEM;
+			break;
+		}
+		task->io_context = ioc;
+		ioc->task = task;
+	} while (1);
 
-	if (ioc)
+	if (!err) {
+		ioc->ioprio = ioprio;
 		ioc->ioprio_changed = 1;
+	}
 
 	task_unlock(task);
-	return 0;
+	return err;
 }
 
 asmlinkage long sys_ioprio_set(int which, int who, int ioprio)
@@ -148,7 +159,9 @@ static int get_task_ioprio(struct task_struct *p)
 	ret = security_task_getioprio(p);
 	if (ret)
 		goto out;
-	ret = p->ioprio;
+	ret = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_NONE, IOPRIO_NORM);
+	if (p->io_context)
+		ret = p->io_context->ioprio;
 out:
 	return ret;
 }
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 49b7a4c31a6d..510a18ba1ec5 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -34,83 +34,10 @@ struct sg_io_hdr;
 #define BLKDEV_MIN_RQ	4
 #define BLKDEV_MAX_RQ	128	/* Default maximum */
 
-/*
- * This is the per-process anticipatory I/O scheduler state.
- */
-struct as_io_context {
-	spinlock_t lock;
-
-	void (*dtor)(struct as_io_context *aic); /* destructor */
-	void (*exit)(struct as_io_context *aic); /* called on task exit */
-
-	unsigned long state;
-	atomic_t nr_queued; /* queued reads & sync writes */
-	atomic_t nr_dispatched; /* number of requests gone to the drivers */
-
-	/* IO History tracking */
-	/* Thinktime */
-	unsigned long last_end_request;
-	unsigned long ttime_total;
-	unsigned long ttime_samples;
-	unsigned long ttime_mean;
-	/* Layout pattern */
-	unsigned int seek_samples;
-	sector_t last_request_pos;
-	u64 seek_total;
-	sector_t seek_mean;
-};
-
-struct cfq_queue;
-struct cfq_io_context {
-	struct rb_node rb_node;
-	void *key;
-
-	struct cfq_queue *cfqq[2];
-
-	struct io_context *ioc;
-
-	unsigned long last_end_request;
-	sector_t last_request_pos;
-
-	unsigned long ttime_total;
-	unsigned long ttime_samples;
-	unsigned long ttime_mean;
-
-	unsigned int seek_samples;
-	u64 seek_total;
-	sector_t seek_mean;
-
-	struct list_head queue_list;
-
-	void (*dtor)(struct io_context *); /* destructor */
-	void (*exit)(struct io_context *); /* called on task exit */
-};
-
-/*
- * This is the per-process I/O subsystem state.  It is refcounted and
- * kmalloc'ed. Currently all fields are modified in process io context
- * (apart from the atomic refcount), so require no locking.
- */
-struct io_context {
-	atomic_t refcount;
-	struct task_struct *task;
-
-	unsigned int ioprio_changed;
-
-	/*
-	 * For request batching
-	 */
-	unsigned long last_waited; /* Time last woken after wait for request */
-	int nr_batch_requests;     /* Number of requests left in the batch */
-
-	struct as_io_context *aic;
-	struct rb_root cic_root;
-	void *ioc_data;
-};
-
 void put_io_context(struct io_context *ioc);
 void exit_io_context(void);
 struct io_context *get_io_context(gfp_t gfp_flags, int node);
+struct io_context *alloc_io_context(gfp_t gfp_flags, int node);
 void copy_io_context(struct io_context **pdst, struct io_context **psrc);
 void swap_io_context(struct io_context **ioc1, struct io_context **ioc2);
 
@@ -894,6 +821,12 @@ static inline void exit_io_context(void)
 {
 }
 
+static inline int put_io_context(struct io_context *ioc)
+{
+	return 1;
+}
+
+
 #endif /* CONFIG_BLOCK */
 
 #endif
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 796019b22b6f..e6b3f7080679 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -137,7 +137,6 @@ extern struct group_info init_groups;
 		.time_slice	= HZ, 					\
 		.nr_cpus_allowed = NR_CPUS,				\
 	},								\
-	.ioprio		= 0,						\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
 	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
 	.ptrace_list	= LIST_HEAD_INIT(tsk.ptrace_list),		\
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
new file mode 100644
index 000000000000..186807ea62e2
--- /dev/null
+++ b/include/linux/iocontext.h
@@ -0,0 +1,79 @@
+#ifndef IOCONTEXT_H
+#define IOCONTEXT_H
+
+/*
+ * This is the per-process anticipatory I/O scheduler state.
+ */
+struct as_io_context {
+	spinlock_t lock;
+
+	void (*dtor)(struct as_io_context *aic); /* destructor */
+	void (*exit)(struct as_io_context *aic); /* called on task exit */
+
+	unsigned long state;
+	atomic_t nr_queued; /* queued reads & sync writes */
+	atomic_t nr_dispatched; /* number of requests gone to the drivers */
+
+	/* IO History tracking */
+	/* Thinktime */
+	unsigned long last_end_request;
+	unsigned long ttime_total;
+	unsigned long ttime_samples;
+	unsigned long ttime_mean;
+	/* Layout pattern */
+	unsigned int seek_samples;
+	sector_t last_request_pos;
+	u64 seek_total;
+	sector_t seek_mean;
+};
+
+struct cfq_queue;
+struct cfq_io_context {
+	struct rb_node rb_node;
+	void *key;
+
+	struct cfq_queue *cfqq[2];
+
+	struct io_context *ioc;
+
+	unsigned long last_end_request;
+	sector_t last_request_pos;
+
+	unsigned long ttime_total;
+	unsigned long ttime_samples;
+	unsigned long ttime_mean;
+
+	unsigned int seek_samples;
+	u64 seek_total;
+	sector_t seek_mean;
+
+	struct list_head queue_list;
+
+	void (*dtor)(struct io_context *); /* destructor */
+	void (*exit)(struct io_context *); /* called on task exit */
+};
+
+/*
+ * This is the per-process I/O subsystem state.  It is refcounted and
+ * kmalloc'ed. Currently all fields are modified in process io context
+ * (apart from the atomic refcount), so require no locking.
+ */
+struct io_context {
+	atomic_t refcount;
+	struct task_struct *task;
+
+	unsigned short ioprio;
+	unsigned short ioprio_changed;
+
+	/*
+	 * For request batching
+	 */
+	unsigned long last_waited; /* Time last woken after wait for request */
+	int nr_batch_requests;     /* Number of requests left in the batch */
+
+	struct as_io_context *aic;
+	struct rb_root cic_root;
+	void *ioc_data;
+};
+
+#endif
diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h
index baf29387cab4..2a3bb1bb7433 100644
--- a/include/linux/ioprio.h
+++ b/include/linux/ioprio.h
@@ -2,6 +2,7 @@
 #define IOPRIO_H
 
 #include <linux/sched.h>
+#include <linux/iocontext.h>
 
 /*
  * Gives us 8 prio classes with 13-bits of data for each class
@@ -45,18 +46,18 @@ enum {
  * the cpu scheduler nice value to an io priority
  */
 #define IOPRIO_NORM	(4)
-static inline int task_ioprio(struct task_struct *task)
+static inline int task_ioprio(struct io_context *ioc)
 {
-	if (ioprio_valid(task->ioprio))
-		return IOPRIO_PRIO_DATA(task->ioprio);
+	if (ioprio_valid(ioc->ioprio))
+		return IOPRIO_PRIO_DATA(ioc->ioprio);
 
 	return IOPRIO_NORM;
 }
 
-static inline int task_ioprio_class(struct task_struct *task)
+static inline int task_ioprio_class(struct io_context *ioc)
 {
-	if (ioprio_valid(task->ioprio))
-		return IOPRIO_PRIO_CLASS(task->ioprio);
+	if (ioprio_valid(ioc->ioprio))
+		return IOPRIO_PRIO_CLASS(ioc->ioprio);
 
 	return IOPRIO_CLASS_BE;
 }
diff --git a/include/linux/sched.h b/include/linux/sched.h
index df5b24ee80b3..80837e7d527e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -975,7 +975,6 @@ struct task_struct {
 	struct hlist_head preempt_notifiers;
 #endif
 
-	unsigned short ioprio;
 	/*
 	 * fpu_counter contains the number of consecutive context switches
 	 * that the FPU is used. If this is over a threshold, the lazy fpu
diff --git a/kernel/fork.c b/kernel/fork.c
index 39d22b3357de..2a86c9dff744 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -51,6 +51,7 @@
 #include <linux/random.h>
 #include <linux/tty.h>
 #include <linux/proc_fs.h>
+#include <linux/blkdev.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -791,6 +792,26 @@ out:
 	return error;
 }
 
+static int copy_io(struct task_struct *tsk)
+{
+#ifdef CONFIG_BLOCK
+	struct io_context *ioc = current->io_context;
+
+	if (!ioc)
+		return 0;
+
+	if (ioprio_valid(ioc->ioprio)) {
+		tsk->io_context = alloc_io_context(GFP_KERNEL, -1);
+		if (unlikely(!tsk->io_context))
+			return -ENOMEM;
+
+		tsk->io_context->task = tsk;
+		tsk->io_context->ioprio = ioc->ioprio;
+	}
+#endif
+	return 0;
+}
+
 /*
  *	Helper to unshare the files of the current task.
  *	We don't want to expose copy_files internals to
@@ -1156,15 +1177,17 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		goto bad_fork_cleanup_mm;
 	if ((retval = copy_namespaces(clone_flags, p)))
 		goto bad_fork_cleanup_keys;
+	if ((retval = copy_io(p)))
+		goto bad_fork_cleanup_namespaces;
 	retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
 	if (retval)
-		goto bad_fork_cleanup_namespaces;
+		goto bad_fork_cleanup_io;
 
 	if (pid != &init_struct_pid) {
 		retval = -ENOMEM;
 		pid = alloc_pid(task_active_pid_ns(p));
 		if (!pid)
-			goto bad_fork_cleanup_namespaces;
+			goto bad_fork_cleanup_io;
 
 		if (clone_flags & CLONE_NEWPID) {
 			retval = pid_ns_prepare_proc(task_active_pid_ns(p));
@@ -1234,9 +1257,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	/* Need tasklist lock for parent etc handling! */
 	write_lock_irq(&tasklist_lock);
 
-	/* for sys_ioprio_set(IOPRIO_WHO_PGRP) */
-	p->ioprio = current->ioprio;
-
 	/*
 	 * The task hasn't been attached yet, so its cpus_allowed mask will
 	 * not be changed, nor will its assigned CPU.
@@ -1328,6 +1348,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 bad_fork_free_pid:
 	if (pid != &init_struct_pid)
 		free_pid(pid);
+bad_fork_cleanup_io:
+	put_io_context(p->io_context);
 bad_fork_cleanup_namespaces:
 	exit_task_namespaces(p);
 bad_fork_cleanup_keys:
-- 
cgit v1.2.3


From d38ecf935fcb10264a6bc190855d9595165e6eeb Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 24 Jan 2008 08:53:35 +0100
Subject: io context sharing: preliminary support

Detach task state from ioc, instead keep track of how many processes
are accessing the ioc.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/ll_rw_blk.c         | 46 ++++++++++++++++++++++++++++++----------------
 fs/ioprio.c               |  1 -
 include/linux/blkdev.h    |  2 +-
 include/linux/iocontext.h | 22 ++++++++++++++++++----
 kernel/fork.c             |  1 -
 5 files changed, 49 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index b9bb02e845cd..d4550ecae443 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3854,12 +3854,13 @@ int __init blk_dev_init(void)
 }
 
 /*
- * IO Context helper functions
+ * IO Context helper functions. put_io_context() returns 1 if there are no
+ * more users of this io context, 0 otherwise.
  */
-void put_io_context(struct io_context *ioc)
+int put_io_context(struct io_context *ioc)
 {
 	if (ioc == NULL)
-		return;
+		return 1;
 
 	BUG_ON(atomic_read(&ioc->refcount) == 0);
 
@@ -3878,7 +3879,9 @@ void put_io_context(struct io_context *ioc)
 		rcu_read_unlock();
 
 		kmem_cache_free(iocontext_cachep, ioc);
+		return 1;
 	}
+	return 0;
 }
 EXPORT_SYMBOL(put_io_context);
 
@@ -3893,15 +3896,17 @@ void exit_io_context(void)
 	current->io_context = NULL;
 	task_unlock(current);
 
-	ioc->task = NULL;
-	if (ioc->aic && ioc->aic->exit)
-		ioc->aic->exit(ioc->aic);
-	if (ioc->cic_root.rb_node != NULL) {
-		cic = rb_entry(rb_first(&ioc->cic_root), struct cfq_io_context, rb_node);
-		cic->exit(ioc);
-	}
+	if (atomic_dec_and_test(&ioc->nr_tasks)) {
+		if (ioc->aic && ioc->aic->exit)
+			ioc->aic->exit(ioc->aic);
+		if (ioc->cic_root.rb_node != NULL) {
+			cic = rb_entry(rb_first(&ioc->cic_root),
+				struct cfq_io_context, rb_node);
+			cic->exit(ioc);
+		}
 
-	put_io_context(ioc);
+		put_io_context(ioc);
+	}
 }
 
 struct io_context *alloc_io_context(gfp_t gfp_flags, int node)
@@ -3911,7 +3916,8 @@ struct io_context *alloc_io_context(gfp_t gfp_flags, int node)
 	ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
-		ret->task = current;
+		atomic_set(&ret->nr_tasks, 1);
+		spin_lock_init(&ret->lock);
 		ret->ioprio_changed = 0;
 		ret->ioprio = 0;
 		ret->last_waited = jiffies; /* doesn't matter... */
@@ -3959,10 +3965,18 @@ static struct io_context *current_io_context(gfp_t gfp_flags, int node)
  */
 struct io_context *get_io_context(gfp_t gfp_flags, int node)
 {
-	struct io_context *ret;
-	ret = current_io_context(gfp_flags, node);
-	if (likely(ret))
-		atomic_inc(&ret->refcount);
+	struct io_context *ret = NULL;
+
+	/*
+	 * Check for unlikely race with exiting task. ioc ref count is
+	 * zero when ioc is being detached.
+	 */
+	do {
+		ret = current_io_context(gfp_flags, node);
+		if (unlikely(!ret))
+			break;
+	} while (!atomic_inc_not_zero(&ret->refcount));
+
 	return ret;
 }
 EXPORT_SYMBOL(get_io_context);
diff --git a/fs/ioprio.c b/fs/ioprio.c
index a7600401ecf7..06b5d97c5fdd 100644
--- a/fs/ioprio.c
+++ b/fs/ioprio.c
@@ -54,7 +54,6 @@ static int set_task_ioprio(struct task_struct *task, int ioprio)
 			break;
 		}
 		task->io_context = ioc;
-		ioc->task = task;
 	} while (1);
 
 	if (!err) {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 510a18ba1ec5..2483a05231c7 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -34,7 +34,7 @@ struct sg_io_hdr;
 #define BLKDEV_MIN_RQ	4
 #define BLKDEV_MAX_RQ	128	/* Default maximum */
 
-void put_io_context(struct io_context *ioc);
+int put_io_context(struct io_context *ioc);
 void exit_io_context(void);
 struct io_context *get_io_context(gfp_t gfp_flags, int node);
 struct io_context *alloc_io_context(gfp_t gfp_flags, int node);
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index 186807ea62e2..cd44d458124a 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -54,13 +54,15 @@ struct cfq_io_context {
 };
 
 /*
- * This is the per-process I/O subsystem state.  It is refcounted and
- * kmalloc'ed. Currently all fields are modified in process io context
- * (apart from the atomic refcount), so require no locking.
+ * I/O subsystem state of the associated processes.  It is refcounted
+ * and kmalloc'ed. These could be shared between processes.
  */
 struct io_context {
 	atomic_t refcount;
-	struct task_struct *task;
+	atomic_t nr_tasks;
+
+	/* all the fields below are protected by this lock */
+	spinlock_t lock;
 
 	unsigned short ioprio;
 	unsigned short ioprio_changed;
@@ -76,4 +78,16 @@ struct io_context {
 	void *ioc_data;
 };
 
+static inline struct io_context *ioc_task_link(struct io_context *ioc)
+{
+	/*
+	 * if ref count is zero, don't allow sharing (ioc is going away, it's
+	 * a race).
+	 */
+	if (ioc && atomic_inc_not_zero(&ioc->refcount))
+		return ioc;
+
+	return NULL;
+}
+
 #endif
diff --git a/kernel/fork.c b/kernel/fork.c
index 2a86c9dff744..1987c57abb08 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -805,7 +805,6 @@ static int copy_io(struct task_struct *tsk)
 		if (unlikely(!tsk->io_context))
 			return -ENOMEM;
 
-		tsk->io_context->task = tsk;
 		tsk->io_context->ioprio = ioc->ioprio;
 	}
 #endif
-- 
cgit v1.2.3


From 4ac845a2e9a816ed5a7b301f56dcc0a3d0b1ba4d Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 24 Jan 2008 08:44:49 +0100
Subject: block: cfq: make the io contect sharing lockless

The io context sharing introduced a per-ioc spinlock, that would protect
the cfq io context lookup. That is a regression from the original, since
we never needed any locking there because the ioc/cic were process private.

The cic lookup is changed from an rbtree construct to a radix tree, which
we can then use RCU to make the reader side lockless. That is the performance
critical path, modifying the radix tree is only done on process creation
(when that process first does IO, actually) and on process exit (if that
process has done IO).

As it so happens, radix trees are also much faster for this type of
lookup where the key is a pointer. It's a very sparse tree.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/cfq-iosched.c       | 267 +++++++++++++++++++++++++---------------------
 block/ll_rw_blk.c         |  49 ++++++---
 include/linux/iocontext.h |   6 +-
 3 files changed, 185 insertions(+), 137 deletions(-)

(limited to 'include')

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index dba52b62cef8..8830893e062f 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -200,7 +200,7 @@ CFQ_CFQQ_FNS(sync);
 static void cfq_dispatch_insert(struct request_queue *, struct request *);
 static struct cfq_queue *cfq_get_queue(struct cfq_data *, int,
 				       struct io_context *, gfp_t);
-static struct cfq_io_context *cfq_cic_rb_lookup(struct cfq_data *,
+static struct cfq_io_context *cfq_cic_lookup(struct cfq_data *,
 						struct io_context *);
 
 static inline struct cfq_queue *cic_to_cfqq(struct cfq_io_context *cic,
@@ -609,7 +609,7 @@ cfq_find_rq_fmerge(struct cfq_data *cfqd, struct bio *bio)
 	struct cfq_io_context *cic;
 	struct cfq_queue *cfqq;
 
-	cic = cfq_cic_rb_lookup(cfqd, tsk->io_context);
+	cic = cfq_cic_lookup(cfqd, tsk->io_context);
 	if (!cic)
 		return NULL;
 
@@ -721,7 +721,7 @@ static int cfq_allow_merge(struct request_queue *q, struct request *rq,
 	 * Lookup the cfqq that this bio will be queued with. Allow
 	 * merge only if rq is queued there.
 	 */
-	cic = cfq_cic_rb_lookup(cfqd, current->io_context);
+	cic = cfq_cic_lookup(cfqd, current->io_context);
 	if (!cic)
 		return 0;
 
@@ -1170,29 +1170,74 @@ static void cfq_put_queue(struct cfq_queue *cfqq)
 	kmem_cache_free(cfq_pool, cfqq);
 }
 
-static void cfq_free_io_context(struct io_context *ioc)
+/*
+ * Call func for each cic attached to this ioc. Returns number of cic's seen.
+ */
+#define CIC_GANG_NR	16
+static unsigned int
+call_for_each_cic(struct io_context *ioc,
+		  void (*func)(struct io_context *, struct cfq_io_context *))
 {
-	struct cfq_io_context *__cic;
-	struct rb_node *n;
-	int freed = 0;
+	struct cfq_io_context *cics[CIC_GANG_NR];
+	unsigned long index = 0;
+	unsigned int called = 0;
+	int nr;
 
-	ioc->ioc_data = NULL;
+	rcu_read_lock();
 
-	spin_lock(&ioc->lock);
+	do {
+		int i;
 
-	while ((n = rb_first(&ioc->cic_root)) != NULL) {
-		__cic = rb_entry(n, struct cfq_io_context, rb_node);
-		rb_erase(&__cic->rb_node, &ioc->cic_root);
-		kmem_cache_free(cfq_ioc_pool, __cic);
-		freed++;
-	}
+		/*
+		 * Perhaps there's a better way - this just gang lookups from
+		 * 0 to the end, restarting after each CIC_GANG_NR from the
+		 * last key + 1.
+		 */
+		nr = radix_tree_gang_lookup(&ioc->radix_root, (void **) cics,
+						index, CIC_GANG_NR);
+		if (!nr)
+			break;
+
+		called += nr;
+		index = 1 + (unsigned long) cics[nr - 1]->key;
+
+		for (i = 0; i < nr; i++)
+			func(ioc, cics[i]);
+	} while (nr == CIC_GANG_NR);
+
+	rcu_read_unlock();
+
+	return called;
+}
+
+static void cic_free_func(struct io_context *ioc, struct cfq_io_context *cic)
+{
+	unsigned long flags;
+
+	BUG_ON(!cic->dead_key);
+
+	spin_lock_irqsave(&ioc->lock, flags);
+	radix_tree_delete(&ioc->radix_root, cic->dead_key);
+	spin_unlock_irqrestore(&ioc->lock, flags);
+
+	kmem_cache_free(cfq_ioc_pool, cic);
+}
+
+static void cfq_free_io_context(struct io_context *ioc)
+{
+	int freed;
+
+	/*
+	 * ioc->refcount is zero here, so no more cic's are allowed to be
+	 * linked into this ioc. So it should be ok to iterate over the known
+	 * list, we will see all cic's since no new ones are added.
+	 */
+	freed = call_for_each_cic(ioc, cic_free_func);
 
 	elv_ioc_count_mod(ioc_count, -freed);
 
 	if (ioc_gone && !elv_ioc_count_read(ioc_count))
 		complete(ioc_gone);
-
-	spin_unlock(&ioc->lock);
 }
 
 static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
@@ -1209,7 +1254,12 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd,
 					 struct cfq_io_context *cic)
 {
 	list_del_init(&cic->queue_list);
+
+	/*
+	 * Make sure key == NULL is seen for dead queues
+	 */
 	smp_wmb();
+	cic->dead_key = (unsigned long) cic->key;
 	cic->key = NULL;
 
 	if (cic->cfqq[ASYNC]) {
@@ -1223,16 +1273,18 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd,
 	}
 }
 
-static void cfq_exit_single_io_context(struct cfq_io_context *cic)
+static void cfq_exit_single_io_context(struct io_context *ioc,
+				       struct cfq_io_context *cic)
 {
 	struct cfq_data *cfqd = cic->key;
 
 	if (cfqd) {
 		struct request_queue *q = cfqd->queue;
+		unsigned long flags;
 
-		spin_lock_irq(q->queue_lock);
+		spin_lock_irqsave(q->queue_lock, flags);
 		__cfq_exit_single_io_context(cfqd, cic);
-		spin_unlock_irq(q->queue_lock);
+		spin_unlock_irqrestore(q->queue_lock, flags);
 	}
 }
 
@@ -1242,24 +1294,8 @@ static void cfq_exit_single_io_context(struct cfq_io_context *cic)
  */
 static void cfq_exit_io_context(struct io_context *ioc)
 {
-	struct cfq_io_context *__cic;
-	struct rb_node *n;
-
-	ioc->ioc_data = NULL;
-
-	spin_lock(&ioc->lock);
-	/*
-	 * put the reference this task is holding to the various queues
-	 */
-	n = rb_first(&ioc->cic_root);
-	while (n != NULL) {
-		__cic = rb_entry(n, struct cfq_io_context, rb_node);
-
-		cfq_exit_single_io_context(__cic);
-		n = rb_next(n);
-	}
-
-	spin_unlock(&ioc->lock);
+	rcu_assign_pointer(ioc->ioc_data, NULL);
+	call_for_each_cic(ioc, cfq_exit_single_io_context);
 }
 
 static struct cfq_io_context *
@@ -1323,7 +1359,8 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq, struct io_context *ioc)
 	cfq_clear_cfqq_prio_changed(cfqq);
 }
 
-static inline void changed_ioprio(struct cfq_io_context *cic)
+static inline void changed_ioprio(struct io_context *ioc,
+				  struct cfq_io_context *cic)
 {
 	struct cfq_data *cfqd = cic->key;
 	struct cfq_queue *cfqq;
@@ -1353,22 +1390,8 @@ static inline void changed_ioprio(struct cfq_io_context *cic)
 
 static void cfq_ioc_set_ioprio(struct io_context *ioc)
 {
-	struct cfq_io_context *cic;
-	struct rb_node *n;
-
-	spin_lock(&ioc->lock);
-
+	call_for_each_cic(ioc, changed_ioprio);
 	ioc->ioprio_changed = 0;
-
-	n = rb_first(&ioc->cic_root);
-	while (n != NULL) {
-		cic = rb_entry(n, struct cfq_io_context, rb_node);
-
-		changed_ioprio(cic);
-		n = rb_next(n);
-	}
-
-	spin_unlock(&ioc->lock);
 }
 
 static struct cfq_queue *
@@ -1379,7 +1402,7 @@ cfq_find_alloc_queue(struct cfq_data *cfqd, int is_sync,
 	struct cfq_io_context *cic;
 
 retry:
-	cic = cfq_cic_rb_lookup(cfqd, ioc);
+	cic = cfq_cic_lookup(cfqd, ioc);
 	/* cic always exists here */
 	cfqq = cic_to_cfqq(cic, is_sync);
 
@@ -1480,28 +1503,42 @@ cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct io_context *ioc,
 	return cfqq;
 }
 
+static void cfq_cic_free(struct cfq_io_context *cic)
+{
+	kmem_cache_free(cfq_ioc_pool, cic);
+	elv_ioc_count_dec(ioc_count);
+
+	if (ioc_gone && !elv_ioc_count_read(ioc_count))
+		complete(ioc_gone);
+}
+
 /*
  * We drop cfq io contexts lazily, so we may find a dead one.
  */
 static void
-cfq_drop_dead_cic(struct io_context *ioc, struct cfq_io_context *cic)
+cfq_drop_dead_cic(struct cfq_data *cfqd, struct io_context *ioc,
+		  struct cfq_io_context *cic)
 {
+	unsigned long flags;
+
 	WARN_ON(!list_empty(&cic->queue_list));
 
+	spin_lock_irqsave(&ioc->lock, flags);
+
 	if (ioc->ioc_data == cic)
-		ioc->ioc_data = NULL;
+		rcu_assign_pointer(ioc->ioc_data, NULL);
 
-	rb_erase(&cic->rb_node, &ioc->cic_root);
-	kmem_cache_free(cfq_ioc_pool, cic);
-	elv_ioc_count_dec(ioc_count);
+	radix_tree_delete(&ioc->radix_root, (unsigned long) cfqd);
+	spin_unlock_irqrestore(&ioc->lock, flags);
+
+	cfq_cic_free(cic);
 }
 
 static struct cfq_io_context *
-cfq_cic_rb_lookup(struct cfq_data *cfqd, struct io_context *ioc)
+cfq_cic_lookup(struct cfq_data *cfqd, struct io_context *ioc)
 {
-	struct rb_node *n;
 	struct cfq_io_context *cic;
-	void *k, *key = cfqd;
+	void *k;
 
 	if (unlikely(!ioc))
 		return NULL;
@@ -1509,79 +1546,65 @@ cfq_cic_rb_lookup(struct cfq_data *cfqd, struct io_context *ioc)
 	/*
 	 * we maintain a last-hit cache, to avoid browsing over the tree
 	 */
-	cic = ioc->ioc_data;
+	cic = rcu_dereference(ioc->ioc_data);
 	if (cic && cic->key == cfqd)
 		return cic;
 
-	spin_lock(&ioc->lock);
-restart:
-	n = ioc->cic_root.rb_node;
-	while (n) {
-		cic = rb_entry(n, struct cfq_io_context, rb_node);
+	do {
+		rcu_read_lock();
+		cic = radix_tree_lookup(&ioc->radix_root, (unsigned long) cfqd);
+		rcu_read_unlock();
+		if (!cic)
+			break;
 		/* ->key must be copied to avoid race with cfq_exit_queue() */
 		k = cic->key;
 		if (unlikely(!k)) {
-			cfq_drop_dead_cic(ioc, cic);
-			goto restart;
+			cfq_drop_dead_cic(cfqd, ioc, cic);
+			continue;
 		}
 
-		if (key < k)
-			n = n->rb_left;
-		else if (key > k)
-			n = n->rb_right;
-		else {
-			ioc->ioc_data = cic;
-			spin_unlock(&ioc->lock);
-			return cic;
-		}
-	}
+		rcu_assign_pointer(ioc->ioc_data, cic);
+		break;
+	} while (1);
 
-	spin_unlock(&ioc->lock);
-	return NULL;
+	return cic;
 }
 
-static inline void
+/*
+ * Add cic into ioc, using cfqd as the search key. This enables us to lookup
+ * the process specific cfq io context when entered from the block layer.
+ * Also adds the cic to a per-cfqd list, used when this queue is removed.
+ */
+static inline int
 cfq_cic_link(struct cfq_data *cfqd, struct io_context *ioc,
-	     struct cfq_io_context *cic)
+	     struct cfq_io_context *cic, gfp_t gfp_mask)
 {
-	struct rb_node **p;
-	struct rb_node *parent;
-	struct cfq_io_context *__cic;
 	unsigned long flags;
-	void *k;
+	int ret;
 
-	spin_lock(&ioc->lock);
-	cic->ioc = ioc;
-	cic->key = cfqd;
+	ret = radix_tree_preload(gfp_mask);
+	if (!ret) {
+		cic->ioc = ioc;
+		cic->key = cfqd;
 
-restart:
-	parent = NULL;
-	p = &ioc->cic_root.rb_node;
-	while (*p) {
-		parent = *p;
-		__cic = rb_entry(parent, struct cfq_io_context, rb_node);
-		/* ->key must be copied to avoid race with cfq_exit_queue() */
-		k = __cic->key;
-		if (unlikely(!k)) {
-			cfq_drop_dead_cic(ioc, __cic);
-			goto restart;
-		}
+		spin_lock_irqsave(&ioc->lock, flags);
+		ret = radix_tree_insert(&ioc->radix_root,
+						(unsigned long) cfqd, cic);
+		spin_unlock_irqrestore(&ioc->lock, flags);
 
-		if (cic->key < k)
-			p = &(*p)->rb_left;
-		else if (cic->key > k)
-			p = &(*p)->rb_right;
-		else
-			BUG();
+		radix_tree_preload_end();
+
+		if (!ret) {
+			spin_lock_irqsave(cfqd->queue->queue_lock, flags);
+			list_add(&cic->queue_list, &cfqd->cic_list);
+			spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+		}
 	}
 
-	rb_link_node(&cic->rb_node, parent, p);
-	rb_insert_color(&cic->rb_node, &ioc->cic_root);
+	if (ret)
+		printk(KERN_ERR "cfq: cic link failed!\n");
 
-	spin_lock_irqsave(cfqd->queue->queue_lock, flags);
-	list_add(&cic->queue_list, &cfqd->cic_list);
-	spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
-	spin_unlock(&ioc->lock);
+	return ret;
 }
 
 /*
@@ -1601,7 +1624,7 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 	if (!ioc)
 		return NULL;
 
-	cic = cfq_cic_rb_lookup(cfqd, ioc);
+	cic = cfq_cic_lookup(cfqd, ioc);
 	if (cic)
 		goto out;
 
@@ -1609,13 +1632,17 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 	if (cic == NULL)
 		goto err;
 
-	cfq_cic_link(cfqd, ioc, cic);
+	if (cfq_cic_link(cfqd, ioc, cic, gfp_mask))
+		goto err_free;
+
 out:
 	smp_read_barrier_depends();
 	if (unlikely(ioc->ioprio_changed))
 		cfq_ioc_set_ioprio(ioc);
 
 	return cic;
+err_free:
+	cfq_cic_free(cic);
 err:
 	put_io_context(ioc);
 	return NULL;
@@ -1909,7 +1936,7 @@ static int cfq_may_queue(struct request_queue *q, int rw)
 	 * so just lookup a possibly existing queue, or return 'may queue'
 	 * if that fails
 	 */
-	cic = cfq_cic_rb_lookup(cfqd, tsk->io_context);
+	cic = cfq_cic_lookup(cfqd, tsk->io_context);
 	if (!cic)
 		return ELV_MQUEUE_MAY;
 
@@ -2174,7 +2201,7 @@ static int __init cfq_slab_setup(void)
 	if (!cfq_pool)
 		goto fail;
 
-	cfq_ioc_pool = KMEM_CACHE(cfq_io_context, 0);
+	cfq_ioc_pool = KMEM_CACHE(cfq_io_context, SLAB_DESTROY_BY_RCU);
 	if (!cfq_ioc_pool)
 		goto fail;
 
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index d4550ecae443..b901db63f6ae 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3853,6 +3853,21 @@ int __init blk_dev_init(void)
 	return 0;
 }
 
+static void cfq_dtor(struct io_context *ioc)
+{
+	struct cfq_io_context *cic[1];
+	int r;
+
+	/*
+	 * We don't have a specific key to lookup with, so use the gang
+	 * lookup to just retrieve the first item stored. The cfq exit
+	 * function will iterate the full tree, so any member will do.
+	 */
+	r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1);
+	if (r > 0)
+		cic[0]->dtor(ioc);
+}
+
 /*
  * IO Context helper functions. put_io_context() returns 1 if there are no
  * more users of this io context, 0 otherwise.
@@ -3865,18 +3880,11 @@ int put_io_context(struct io_context *ioc)
 	BUG_ON(atomic_read(&ioc->refcount) == 0);
 
 	if (atomic_dec_and_test(&ioc->refcount)) {
-		struct cfq_io_context *cic;
-
 		rcu_read_lock();
 		if (ioc->aic && ioc->aic->dtor)
 			ioc->aic->dtor(ioc->aic);
-		if (ioc->cic_root.rb_node != NULL) {
-			struct rb_node *n = rb_first(&ioc->cic_root);
-
-			cic = rb_entry(n, struct cfq_io_context, rb_node);
-			cic->dtor(ioc);
-		}
 		rcu_read_unlock();
+		cfq_dtor(ioc);
 
 		kmem_cache_free(iocontext_cachep, ioc);
 		return 1;
@@ -3885,11 +3893,26 @@ int put_io_context(struct io_context *ioc)
 }
 EXPORT_SYMBOL(put_io_context);
 
+static void cfq_exit(struct io_context *ioc)
+{
+	struct cfq_io_context *cic[1];
+	int r;
+
+	rcu_read_lock();
+	/*
+	 * See comment for cfq_dtor()
+	 */
+	r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1);
+	rcu_read_unlock();
+
+	if (r > 0)
+		cic[0]->exit(ioc);
+}
+
 /* Called by the exitting task */
 void exit_io_context(void)
 {
 	struct io_context *ioc;
-	struct cfq_io_context *cic;
 
 	task_lock(current);
 	ioc = current->io_context;
@@ -3899,11 +3922,7 @@ void exit_io_context(void)
 	if (atomic_dec_and_test(&ioc->nr_tasks)) {
 		if (ioc->aic && ioc->aic->exit)
 			ioc->aic->exit(ioc->aic);
-		if (ioc->cic_root.rb_node != NULL) {
-			cic = rb_entry(rb_first(&ioc->cic_root),
-				struct cfq_io_context, rb_node);
-			cic->exit(ioc);
-		}
+		cfq_exit(ioc);
 
 		put_io_context(ioc);
 	}
@@ -3923,7 +3942,7 @@ struct io_context *alloc_io_context(gfp_t gfp_flags, int node)
 		ret->last_waited = jiffies; /* doesn't matter... */
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
-		ret->cic_root.rb_node = NULL;
+		INIT_RADIX_TREE(&ret->radix_root, GFP_ATOMIC | __GFP_HIGH);
 		ret->ioc_data = NULL;
 	}
 
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index cd44d458124a..593b222d9dcc 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -1,6 +1,8 @@
 #ifndef IOCONTEXT_H
 #define IOCONTEXT_H
 
+#include <linux/radix-tree.h>
+
 /*
  * This is the per-process anticipatory I/O scheduler state.
  */
@@ -29,8 +31,8 @@ struct as_io_context {
 
 struct cfq_queue;
 struct cfq_io_context {
-	struct rb_node rb_node;
 	void *key;
+	unsigned long dead_key;
 
 	struct cfq_queue *cfqq[2];
 
@@ -74,7 +76,7 @@ struct io_context {
 	int nr_batch_requests;     /* Number of requests left in the batch */
 
 	struct as_io_context *aic;
-	struct rb_root cic_root;
+	struct radix_tree_root radix_root;
 	void *ioc_data;
 };
 
-- 
cgit v1.2.3


From fadad878cc0640cc9cd5569998bf54b693f7b38b Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 24 Jan 2008 08:54:47 +0100
Subject: kernel: add CLONE_IO to specifically request sharing of IO contexts

syslets (or other threads/processes that want io context sharing) can
set this to enforce sharing of io context.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/sched.h |  1 +
 kernel/fork.c         | 14 ++++++++++----
 2 files changed, 11 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 80837e7d527e..2d0546e884ea 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -27,6 +27,7 @@
 #define CLONE_NEWUSER		0x10000000	/* New user namespace */
 #define CLONE_NEWPID		0x20000000	/* New pid namespace */
 #define CLONE_NEWNET		0x40000000	/* New network namespace */
+#define CLONE_IO		0x80000000	/* Clone io context */
 
 /*
  * Scheduling policies
diff --git a/kernel/fork.c b/kernel/fork.c
index 1987c57abb08..314f5101d2b0 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -792,15 +792,21 @@ out:
 	return error;
 }
 
-static int copy_io(struct task_struct *tsk)
+static int copy_io(unsigned long clone_flags, struct task_struct *tsk)
 {
 #ifdef CONFIG_BLOCK
 	struct io_context *ioc = current->io_context;
 
 	if (!ioc)
 		return 0;
-
-	if (ioprio_valid(ioc->ioprio)) {
+	/*
+	 * Share io context with parent, if CLONE_IO is set
+	 */
+	if (clone_flags & CLONE_IO) {
+		tsk->io_context = ioc_task_link(ioc);
+		if (unlikely(!tsk->io_context))
+			return -ENOMEM;
+	} else if (ioprio_valid(ioc->ioprio)) {
 		tsk->io_context = alloc_io_context(GFP_KERNEL, -1);
 		if (unlikely(!tsk->io_context))
 			return -ENOMEM;
@@ -1176,7 +1182,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		goto bad_fork_cleanup_mm;
 	if ((retval = copy_namespaces(clone_flags, p)))
 		goto bad_fork_cleanup_keys;
-	if ((retval = copy_io(p)))
+	if ((retval = copy_io(clone_flags, p)))
 		goto bad_fork_cleanup_namespaces;
 	retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
 	if (retval)
-- 
cgit v1.2.3


From fa0ccd837e3dddb44c7db2f128a8bb7e4eabc21a Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Thu, 10 Jan 2008 11:30:36 -0600
Subject: block: implement drain buffers

These DMA drain buffer implementations in drivers are pretty horrible
to do in terms of manipulating the scatterlist.  Plus they're being
done at least in drivers/ide and drivers/ata, so we now have code
duplication.

The one use case for this, as I understand it is AHCI controllers doing
PIO mode to mmc devices but translating this to DMA at the controller
level.

So, what about adding a callback to the block layer that permits the
adding of the drain buffer for the problem devices.  The idea is that
you'd do this in slave_configure after you find one of these devices.

The beauty of doing it in the block layer is that it quietly adds the
drain buffer to the end of the sg list, so it automatically gets mapped
(and unmapped) without anything unusual having to be done to the
scatterlist in driver/scsi or drivers/ata and without any alteration to
the transfer length.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/elevator.c       | 26 +++++++++++++++++++++++++-
 block/ll_rw_blk.c      | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h |  4 ++++
 3 files changed, 78 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/block/elevator.c b/block/elevator.c
index f9736fbdab03..8cd5775acd7a 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -741,7 +741,21 @@ struct request *elv_next_request(struct request_queue *q)
 			q->boundary_rq = NULL;
 		}
 
-		if ((rq->cmd_flags & REQ_DONTPREP) || !q->prep_rq_fn)
+		if (rq->cmd_flags & REQ_DONTPREP)
+			break;
+
+		if (q->dma_drain_size && rq->data_len) {
+			/*
+			 * make sure space for the drain appears we
+			 * know we can do this because max_hw_segments
+			 * has been adjusted to be one fewer than the
+			 * device can handle
+			 */
+			rq->nr_phys_segments++;
+			rq->nr_hw_segments++;
+		}
+
+		if (!q->prep_rq_fn)
 			break;
 
 		ret = q->prep_rq_fn(q, rq);
@@ -754,6 +768,16 @@ struct request *elv_next_request(struct request_queue *q)
 			 * avoid resource deadlock.  REQ_STARTED will
 			 * prevent other fs requests from passing this one.
 			 */
+			if (q->dma_drain_size && rq->data_len &&
+			    !(rq->cmd_flags & REQ_DONTPREP)) {
+				/*
+				 * remove the space for the drain we added
+				 * so that we don't add it again
+				 */
+				--rq->nr_phys_segments;
+				--rq->nr_hw_segments;
+			}
+
 			rq = NULL;
 			break;
 		} else if (ret == BLKPREP_KILL) {
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 3d0422f48453..768987dc2697 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -725,6 +725,45 @@ void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b)
 
 EXPORT_SYMBOL(blk_queue_stack_limits);
 
+/**
+ * blk_queue_dma_drain - Set up a drain buffer for excess dma.
+ *
+ * @q:  the request queue for the device
+ * @buf:	physically contiguous buffer
+ * @size:	size of the buffer in bytes
+ *
+ * Some devices have excess DMA problems and can't simply discard (or
+ * zero fill) the unwanted piece of the transfer.  They have to have a
+ * real area of memory to transfer it into.  The use case for this is
+ * ATAPI devices in DMA mode.  If the packet command causes a transfer
+ * bigger than the transfer size some HBAs will lock up if there
+ * aren't DMA elements to contain the excess transfer.  What this API
+ * does is adjust the queue so that the buf is always appended
+ * silently to the scatterlist.
+ *
+ * Note: This routine adjusts max_hw_segments to make room for
+ * appending the drain buffer.  If you call
+ * blk_queue_max_hw_segments() or blk_queue_max_phys_segments() after
+ * calling this routine, you must set the limit to one fewer than your
+ * device can support otherwise there won't be room for the drain
+ * buffer.
+ */
+int blk_queue_dma_drain(struct request_queue *q, void *buf,
+				unsigned int size)
+{
+	if (q->max_hw_segments < 2 || q->max_phys_segments < 2)
+		return -EINVAL;
+	/* make room for appending the drain */
+	--q->max_hw_segments;
+	--q->max_phys_segments;
+	q->dma_drain_buffer = buf;
+	q->dma_drain_size = size;
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(blk_queue_dma_drain);
+
 /**
  * blk_queue_segment_boundary - set boundary rules for segment merging
  * @q:  the request queue for the device
@@ -1379,6 +1418,16 @@ new_segment:
 		bvprv = bvec;
 	} /* segments in rq */
 
+	if (q->dma_drain_size) {
+		sg->page_link &= ~0x02;
+		sg = sg_next(sg);
+		sg_set_page(sg, virt_to_page(q->dma_drain_buffer),
+			    q->dma_drain_size,
+			    ((unsigned long)q->dma_drain_buffer) &
+			    (PAGE_SIZE - 1));
+		nsegs++;
+	}
+
 	if (sg)
 		sg_mark_end(sg);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index c7a3ab575c24..e542c8fd9215 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -429,6 +429,8 @@ struct request_queue
 	unsigned int		max_segment_size;
 
 	unsigned long		seg_boundary_mask;
+	void			*dma_drain_buffer;
+	unsigned int		dma_drain_size;
 	unsigned int		dma_alignment;
 
 	struct blk_queue_tag	*queue_tags;
@@ -760,6 +762,8 @@ extern void blk_queue_max_hw_segments(struct request_queue *, unsigned short);
 extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
 extern void blk_queue_hardsect_size(struct request_queue *, unsigned short);
 extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
+extern int blk_queue_dma_drain(struct request_queue *q, void *buf,
+			       unsigned int size);
 extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
 extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
 extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
-- 
cgit v1.2.3


From 7cedb1f17fb7f4374d11501f61656ae9d3ba47e9 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Sun, 13 Jan 2008 14:15:28 -0600
Subject: SG: work with the SCSI fixed maximum allocations.

SCSI sg table allocation has a maximum size (of SCSI_MAX_SG_SEGMENTS,
currently 128) and this will cause a BUG_ON() in SCSI if something
tries an allocation over it.  This patch adds a size limit to the
chaining allocator to allow the specification of the maximum
allocation size for chaining, so we always chain in units of the
maximum SCSI allocation size.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 drivers/scsi/scsi_lib.c     |  8 +++++---
 include/linux/scatterlist.h |  5 +++--
 lib/scatterlist.c           | 41 +++++++++++++++++++++++++++--------------
 3 files changed, 35 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 3b5121c4c081..eb4911a61641 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -761,9 +761,11 @@ int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 
 	BUG_ON(!cmd->use_sg);
 
-	ret = __sg_alloc_table(&cmd->sg_table, cmd->use_sg, gfp_mask, scsi_sg_alloc);
+	ret = __sg_alloc_table(&cmd->sg_table, cmd->use_sg,
+			       SCSI_MAX_SG_SEGMENTS, gfp_mask, scsi_sg_alloc);
 	if (unlikely(ret))
-		__sg_free_table(&cmd->sg_table, scsi_sg_free);
+		__sg_free_table(&cmd->sg_table, SCSI_MAX_SG_SEGMENTS,
+				scsi_sg_free);
 
 	cmd->request_buffer = cmd->sg_table.sgl;
 	return ret;
@@ -773,7 +775,7 @@ EXPORT_SYMBOL(scsi_alloc_sgtable);
 
 void scsi_free_sgtable(struct scsi_cmnd *cmd)
 {
-	__sg_free_table(&cmd->sg_table, scsi_sg_free);
+	__sg_free_table(&cmd->sg_table, SCSI_MAX_SG_SEGMENTS, scsi_sg_free);
 }
 
 EXPORT_SYMBOL(scsi_free_sgtable);
diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index e9cb103417b2..a3d567a974e8 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -207,9 +207,10 @@ void sg_init_one(struct scatterlist *, const void *, unsigned int);
 typedef struct scatterlist *(sg_alloc_fn)(unsigned int, gfp_t);
 typedef void (sg_free_fn)(struct scatterlist *, unsigned int);
 
-void __sg_free_table(struct sg_table *, sg_free_fn *);
+void __sg_free_table(struct sg_table *, unsigned int, sg_free_fn *);
 void sg_free_table(struct sg_table *);
-int __sg_alloc_table(struct sg_table *, unsigned int, gfp_t, sg_alloc_fn *);
+int __sg_alloc_table(struct sg_table *, unsigned int, unsigned int, gfp_t,
+		     sg_alloc_fn *);
 int sg_alloc_table(struct sg_table *, unsigned int, gfp_t);
 
 /*
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index 02aaa27e010e..acca4901046c 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -130,13 +130,17 @@ static void sg_kfree(struct scatterlist *sg, unsigned int nents)
 /**
  * __sg_free_table - Free a previously mapped sg table
  * @table:	The sg table header to use
+ * @max_ents:	The maximum number of entries per single scatterlist
  * @free_fn:	Free function
  *
  *  Description:
- *    Free an sg table previously allocated and setup with __sg_alloc_table().
+ *    Free an sg table previously allocated and setup with
+ *    __sg_alloc_table().  The @max_ents value must be identical to
+ *    that previously used with __sg_alloc_table().
  *
  **/
-void __sg_free_table(struct sg_table *table, sg_free_fn *free_fn)
+void __sg_free_table(struct sg_table *table, unsigned int max_ents,
+		     sg_free_fn *free_fn)
 {
 	struct scatterlist *sgl, *next;
 
@@ -149,14 +153,14 @@ void __sg_free_table(struct sg_table *table, sg_free_fn *free_fn)
 		unsigned int sg_size;
 
 		/*
-		 * If we have more than SG_MAX_SINGLE_ALLOC segments left,
+		 * If we have more than max_ents segments left,
 		 * then assign 'next' to the sg table after the current one.
 		 * sg_size is then one less than alloc size, since the last
 		 * element is the chain pointer.
 		 */
-		if (alloc_size > SG_MAX_SINGLE_ALLOC) {
-			next = sg_chain_ptr(&sgl[SG_MAX_SINGLE_ALLOC - 1]);
-			alloc_size = SG_MAX_SINGLE_ALLOC;
+		if (alloc_size > max_ents) {
+			next = sg_chain_ptr(&sgl[max_ents - 1]);
+			alloc_size = max_ents;
 			sg_size = alloc_size - 1;
 		} else {
 			sg_size = alloc_size;
@@ -179,7 +183,7 @@ EXPORT_SYMBOL(__sg_free_table);
  **/
 void sg_free_table(struct sg_table *table)
 {
-	__sg_free_table(table, sg_kfree);
+	__sg_free_table(table, SG_MAX_SINGLE_ALLOC, sg_kfree);
 }
 EXPORT_SYMBOL(sg_free_table);
 
@@ -187,22 +191,30 @@ EXPORT_SYMBOL(sg_free_table);
  * __sg_alloc_table - Allocate and initialize an sg table with given allocator
  * @table:	The sg table header to use
  * @nents:	Number of entries in sg list
+ * @max_ents:	The maximum number of entries the allocator returns per call
  * @gfp_mask:	GFP allocation mask
  * @alloc_fn:	Allocator to use
  *
+ * Description:
+ *   This function returns a @table @nents long. The allocator is
+ *   defined to return scatterlist chunks of maximum size @max_ents.
+ *   Thus if @nents is bigger than @max_ents, the scatterlists will be
+ *   chained in units of @max_ents.
+ *
  * Notes:
  *   If this function returns non-0 (eg failure), the caller must call
  *   __sg_free_table() to cleanup any leftover allocations.
  *
  **/
-int __sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask,
+int __sg_alloc_table(struct sg_table *table, unsigned int nents,
+		     unsigned int max_ents, gfp_t gfp_mask,
 		     sg_alloc_fn *alloc_fn)
 {
 	struct scatterlist *sg, *prv;
 	unsigned int left;
 
 #ifndef ARCH_HAS_SG_CHAIN
-	BUG_ON(nents > SG_MAX_SINGLE_ALLOC);
+	BUG_ON(nents > max_ents);
 #endif
 
 	memset(table, 0, sizeof(*table));
@@ -212,8 +224,8 @@ int __sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask,
 	do {
 		unsigned int sg_size, alloc_size = left;
 
-		if (alloc_size > SG_MAX_SINGLE_ALLOC) {
-			alloc_size = SG_MAX_SINGLE_ALLOC;
+		if (alloc_size > max_ents) {
+			alloc_size = max_ents;
 			sg_size = alloc_size - 1;
 		} else
 			sg_size = alloc_size;
@@ -232,7 +244,7 @@ int __sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask,
 		 * If this is not the first mapping, chain previous part.
 		 */
 		if (prv)
-			sg_chain(prv, SG_MAX_SINGLE_ALLOC, sg);
+			sg_chain(prv, max_ents, sg);
 		else
 			table->sgl = sg;
 
@@ -272,9 +284,10 @@ int sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask)
 {
 	int ret;
 
-	ret = __sg_alloc_table(table, nents, gfp_mask, sg_kmalloc);
+	ret = __sg_alloc_table(table, nents, SG_MAX_SINGLE_ALLOC,
+			       gfp_mask, sg_kmalloc);
 	if (unlikely(ret))
-		__sg_free_table(table, sg_kfree);
+		__sg_free_table(table, SG_MAX_SINGLE_ALLOC, sg_kfree);
 
 	return ret;
 }
-- 
cgit v1.2.3


From e5c271ec3bdfaca5e8d47a9e63cfc0bf889881aa Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 22 Nov 2007 17:59:11 +0100
Subject: [ARM] 4664/1: Add basic support for HTC Magician PDA phones

This includes irda, gpio keys, pxafb, backlight, ohci and flash
(read-only).

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/Kconfig           |   6 +
 arch/arm/mach-pxa/Makefile          |   1 +
 arch/arm/mach-pxa/magician.c        | 218 ++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-pxa/magician.h | 111 ++++++++++++++++++
 4 files changed, 336 insertions(+)
 create mode 100644 arch/arm/mach-pxa/magician.c
 create mode 100644 include/asm-arm/arch-pxa/magician.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index b08f55d906c8..619f6fc2bfe1 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -112,6 +112,12 @@ config MACH_ARMCORE
 	select PXA27x
 	select IWMMXT
 
+config MACH_MAGICIAN
+	bool "Enable HTC Magician Support"
+	depends on ARCH_PXA
+	select PXA27x
+	select IWMMXT
+
 endchoice
 
 if PXA_SHARPSL
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 04f8323c8c89..fca4e819dc5d 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -22,6 +22,7 @@ obj-$(CONFIG_MACH_AKITA)	+= akita-ioexp.o
 obj-$(CONFIG_MACH_POODLE)	+= poodle.o corgi_ssp.o
 obj-$(CONFIG_MACH_TOSA)		+= tosa.o
 obj-$(CONFIG_MACH_EM_X270)	+= em-x270.o
+obj-$(CONFIG_MACH_MAGICIAN)	+= magician.o
 obj-$(CONFIG_ARCH_PXA_ESERIES)	+= eseries.o
 
 ifeq ($(CONFIG_MACH_ZYLONITE),y)
diff --git a/arch/arm/mach-pxa/magician.c b/arch/arm/mach-pxa/magician.c
new file mode 100644
index 000000000000..d98ef7ada2f8
--- /dev/null
+++ b/arch/arm/mach-pxa/magician.c
@@ -0,0 +1,218 @@
+/*
+ * Support for HTC Magician PDA phones:
+ * i-mate JAM, O2 Xda mini, Orange SPV M500, Qtek s100, Qtek s110
+ * and T-Mobile MDA Compact.
+ *
+ * Copyright (c) 2006-2007 Philipp Zabel
+ *
+ * Based on hx4700.c, spitz.c and others.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/map.h>
+#include <linux/mtd/physmap.h>
+
+#include <asm/gpio.h>
+#include <asm/hardware.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+#include <asm/arch/magician.h>
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxafb.h>
+#include <asm/arch/irda.h>
+#include <asm/arch/ohci.h>
+
+#include "generic.h"
+
+/*
+ * IRDA
+ */
+
+static void magician_irda_transceiver_mode(struct device *dev, int mode)
+{
+	gpio_set_value(GPIO83_MAGICIAN_nIR_EN, mode & IR_OFF);
+}
+
+static struct pxaficp_platform_data magician_ficp_info = {
+	.transceiver_cap  = IR_SIRMODE | IR_OFF,
+	.transceiver_mode = magician_irda_transceiver_mode,
+};
+
+/*
+ * GPIO Keys
+ */
+
+static struct gpio_keys_button magician_button_table[] = {
+	{KEY_POWER,      GPIO0_MAGICIAN_KEY_POWER,      0, "Power button"},
+	{KEY_ESC,        GPIO37_MAGICIAN_KEY_HANGUP,    0, "Hangup button"},
+	{KEY_F10,        GPIO38_MAGICIAN_KEY_CONTACTS,  0, "Contacts button"},
+	{KEY_CALENDAR,   GPIO90_MAGICIAN_KEY_CALENDAR,  0, "Calendar button"},
+	{KEY_CAMERA,     GPIO91_MAGICIAN_KEY_CAMERA,    0, "Camera button"},
+	{KEY_UP,         GPIO93_MAGICIAN_KEY_UP,        0, "Up button"},
+	{KEY_DOWN,       GPIO94_MAGICIAN_KEY_DOWN,      0, "Down button"},
+	{KEY_LEFT,       GPIO95_MAGICIAN_KEY_LEFT,      0, "Left button"},
+	{KEY_RIGHT,      GPIO96_MAGICIAN_KEY_RIGHT,     0, "Right button"},
+	{KEY_KPENTER,    GPIO97_MAGICIAN_KEY_ENTER,     0, "Action button"},
+	{KEY_RECORD,     GPIO98_MAGICIAN_KEY_RECORD,    0, "Record button"},
+	{KEY_VOLUMEUP,   GPIO100_MAGICIAN_KEY_VOL_UP,   0, "Volume up"},
+	{KEY_VOLUMEDOWN, GPIO101_MAGICIAN_KEY_VOL_DOWN, 0, "Volume down"},
+	{KEY_PHONE,      GPIO102_MAGICIAN_KEY_PHONE,    0, "Phone button"},
+	{KEY_PLAY,       GPIO99_MAGICIAN_HEADPHONE_IN,  0, "Headset button"},
+};
+
+static struct gpio_keys_platform_data gpio_keys_data = {
+	.buttons  = magician_button_table,
+	.nbuttons = ARRAY_SIZE(magician_button_table),
+};
+
+static struct platform_device gpio_keys = {
+	.name = "gpio-keys",
+	.dev  = {
+		.platform_data = &gpio_keys_data,
+	},
+	.id   = -1,
+};
+
+/*
+ * LCD - Toppoly TD028STEB1
+ */
+
+static struct pxafb_mode_info toppoly_modes[] = {
+	{
+		.pixclock     = 96153,
+		.bpp          = 16,
+		.xres         = 240,
+		.yres         = 320,
+		.hsync_len    = 11,
+		.vsync_len    = 3,
+		.left_margin  = 19,
+		.upper_margin = 2,
+		.right_margin = 10,
+		.lower_margin = 2,
+		.sync         = 0,
+	},
+};
+
+static struct pxafb_mach_info toppoly_info = {
+	.modes       = toppoly_modes,
+	.num_modes   = 1,
+	.fixed_modes = 1,
+	.lccr0       = LCCR0_Color | LCCR0_Sngl | LCCR0_Act,
+	.lccr3       = LCCR3_PixRsEdg,
+};
+
+/*
+ * Backlight
+ */
+
+static void magician_set_bl_intensity(int intensity)
+{
+	if (intensity) {
+		PWM_CTRL0 = 1;
+		PWM_PERVAL0 = 0xc8;
+		PWM_PWDUTY0 = intensity;
+		pxa_set_cken(CKEN_PWM0, 1);
+	} else {
+		pxa_set_cken(CKEN_PWM0, 0);
+	}
+}
+
+static struct generic_bl_info backlight_info = {
+	.default_intensity = 0x64,
+	.limit_mask        = 0x0b,
+	.max_intensity     = 0xc7,
+	.set_bl_intensity  = magician_set_bl_intensity,
+};
+
+static struct platform_device backlight = {
+	.name = "corgi-bl",
+	.dev  = {
+		.platform_data = &backlight_info,
+	},
+	.id   = -1,
+};
+
+
+/*
+ * USB OHCI
+ */
+
+static int magician_ohci_init(struct device *dev)
+{
+	UHCHR = (UHCHR | UHCHR_SSEP2 | UHCHR_PCPL | UHCHR_CGR) &
+	    ~(UHCHR_SSEP1 | UHCHR_SSEP3 | UHCHR_SSE);
+
+	return 0;
+}
+
+static struct pxaohci_platform_data magician_ohci_info = {
+	.port_mode    = PMM_PERPORT_MODE,
+	.init         = magician_ohci_init,
+	.power_budget = 0,
+};
+
+
+/*
+ * StrataFlash
+ */
+
+#define PXA_CS_SIZE		0x04000000
+
+static struct resource strataflash_resource = {
+	.start = PXA_CS0_PHYS,
+	.end   = PXA_CS0_PHYS + PXA_CS_SIZE - 1,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct physmap_flash_data strataflash_data = {
+	.width = 4,
+};
+
+static struct platform_device strataflash = {
+	.name          = "physmap-flash",
+	.id            = -1,
+	.num_resources = 1,
+	.resource      = &strataflash_resource,
+	.dev = {
+		.platform_data = &strataflash_data,
+	},
+};
+
+/*
+ * Platform devices
+ */
+
+static struct platform_device *devices[] __initdata = {
+	&gpio_keys,
+	&backlight,
+	&strataflash,
+};
+
+static void __init magician_init(void)
+{
+	platform_add_devices(devices, ARRAY_SIZE(devices));
+	pxa_set_ohci_info(&magician_ohci_info);
+	pxa_set_ficp_info(&magician_ficp_info);
+	set_pxa_fb_info(&toppoly_info);
+}
+
+
+MACHINE_START(MAGICIAN, "HTC Magician")
+	.phys_io = 0x40000000,
+	.io_pg_offst = (io_p2v(0x40000000) >> 18) & 0xfffc,
+	.boot_params = 0xa0000100,
+	.map_io = pxa_map_io,
+	.init_irq = pxa27x_init_irq,
+	.init_machine = magician_init,
+	.timer = &pxa_timer,
+MACHINE_END
diff --git a/include/asm-arm/arch-pxa/magician.h b/include/asm-arm/arch-pxa/magician.h
new file mode 100644
index 000000000000..337f51f06b3a
--- /dev/null
+++ b/include/asm-arm/arch-pxa/magician.h
@@ -0,0 +1,111 @@
+/*
+ * GPIO and IRQ definitions for HTC Magician PDA phones
+ *
+ * Copyright (c) 2007 Philipp Zabel
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#ifndef _MAGICIAN_H_
+#define _MAGICIAN_H_
+
+#include <asm/arch/pxa-regs.h>
+
+/*
+ * PXA GPIOs
+ */
+
+#define GPIO0_MAGICIAN_KEY_POWER		0
+#define GPIO9_MAGICIAN_UNKNOWN			9
+#define GPIO10_MAGICIAN_GSM_IRQ			10
+#define GPIO11_MAGICIAN_GSM_OUT1		11
+#define GPIO13_MAGICIAN_CPLD_IRQ		13
+#define GPIO18_MAGICIAN_UNKNOWN			18
+#define GPIO22_MAGICIAN_VIBRA_EN		22
+#define GPIO26_MAGICIAN_GSM_POWER		26
+#define GPIO27_MAGICIAN_USBC_PUEN		27
+#define GPIO30_MAGICIAN_nCHARGE_EN		30
+#define GPIO37_MAGICIAN_KEY_HANGUP		37
+#define GPIO38_MAGICIAN_KEY_CONTACTS		38
+#define GPIO40_MAGICIAN_GSM_OUT2		40
+#define GPIO48_MAGICIAN_UNKNOWN			48
+#define GPIO56_MAGICIAN_UNKNOWN			56
+#define GPIO57_MAGICIAN_CAM_RESET		57
+#define GPIO83_MAGICIAN_nIR_EN			83
+#define GPIO86_MAGICIAN_GSM_RESET		86
+#define GPIO87_MAGICIAN_GSM_SELECT		87
+#define GPIO90_MAGICIAN_KEY_CALENDAR		90
+#define GPIO91_MAGICIAN_KEY_CAMERA		91
+#define GPIO93_MAGICIAN_KEY_UP			93
+#define GPIO94_MAGICIAN_KEY_DOWN		94
+#define GPIO95_MAGICIAN_KEY_LEFT		95
+#define GPIO96_MAGICIAN_KEY_RIGHT		96
+#define GPIO97_MAGICIAN_KEY_ENTER		97
+#define GPIO98_MAGICIAN_KEY_RECORD		98
+#define GPIO99_MAGICIAN_HEADPHONE_IN		99
+#define GPIO100_MAGICIAN_KEY_VOL_UP		100
+#define GPIO101_MAGICIAN_KEY_VOL_DOWN 		101
+#define GPIO102_MAGICIAN_KEY_PHONE		102
+#define GPIO103_MAGICIAN_LED_KP			103
+#define GPIO104_MAGICIAN_LCD_POWER_1 		104
+#define GPIO105_MAGICIAN_LCD_POWER_2		105
+#define GPIO106_MAGICIAN_LCD_POWER_3		106
+#define GPIO107_MAGICIAN_DS1WM_IRQ		107
+#define GPIO108_MAGICIAN_GSM_READY		108
+#define GPIO114_MAGICIAN_UNKNOWN		114
+#define GPIO115_MAGICIAN_nPEN_IRQ		115
+#define GPIO116_MAGICIAN_nCAM_EN		116
+#define GPIO119_MAGICIAN_UNKNOWN		119
+#define GPIO120_MAGICIAN_UNKNOWN		120
+
+/*
+ * PXA GPIO alternate function mode & direction
+ */
+
+#define GPIO0_MAGICIAN_KEY_POWER_MD		(0 | GPIO_IN)
+#define GPIO9_MAGICIAN_UNKNOWN_MD		(9 | GPIO_IN)
+#define GPIO10_MAGICIAN_GSM_IRQ_MD		(10 | GPIO_IN)
+#define GPIO11_MAGICIAN_GSM_OUT1_MD		(11 | GPIO_OUT)
+#define GPIO13_MAGICIAN_CPLD_IRQ_MD		(13 | GPIO_IN)
+#define GPIO18_MAGICIAN_UNKNOWN_MD		(18 | GPIO_OUT)
+#define GPIO22_MAGICIAN_VIBRA_EN_MD		(22 | GPIO_OUT)
+#define GPIO26_MAGICIAN_GSM_POWER_MD		(26 | GPIO_OUT)
+#define GPIO27_MAGICIAN_USBC_PUEN_MD		(27 | GPIO_OUT)
+#define GPIO30_MAGICIAN_nCHARGE_EN_MD		(30 | GPIO_OUT)
+#define GPIO37_MAGICIAN_KEY_HANGUP_MD		(37 | GPIO_OUT)
+#define GPIO38_MAGICIAN_KEY_CONTACTS_MD		(38 | GPIO_OUT)
+#define GPIO40_MAGICIAN_GSM_OUT2_MD		(40 | GPIO_OUT)
+#define GPIO48_MAGICIAN_UNKNOWN_MD		(48 | GPIO_OUT)
+#define GPIO56_MAGICIAN_UNKNOWN_MD		(56 | GPIO_OUT)
+#define GPIO57_MAGICIAN_CAM_RESET_MD		(57 | GPIO_OUT)
+#define GPIO83_MAGICIAN_nIR_EN_MD		(83 | GPIO_OUT)
+#define GPIO86_MAGICIAN_GSM_RESET_MD		(86 | GPIO_OUT)
+#define GPIO87_MAGICIAN_GSM_SELECT_MD		(87 | GPIO_OUT)
+#define GPIO90_MAGICIAN_KEY_CALENDAR_MD		(90 | GPIO_OUT)
+#define GPIO91_MAGICIAN_KEY_CAMERA_MD		(91 | GPIO_OUT)
+#define GPIO93_MAGICIAN_KEY_UP_MD		(93 | GPIO_IN)
+#define GPIO94_MAGICIAN_KEY_DOWN_MD		(94 | GPIO_IN)
+#define GPIO95_MAGICIAN_KEY_LEFT_MD		(95 | GPIO_IN)
+#define GPIO96_MAGICIAN_KEY_RIGHT_MD		(96 | GPIO_IN)
+#define GPIO97_MAGICIAN_KEY_ENTER_MD		(97 | GPIO_IN)
+#define GPIO98_MAGICIAN_KEY_RECORD_MD		(98 | GPIO_IN)
+#define GPIO99_MAGICIAN_HEADPHONE_IN_MD		(99 | GPIO_IN)
+#define GPIO100_MAGICIAN_KEY_VOL_UP_MD		(100 | GPIO_IN)
+#define GPIO101_MAGICIAN_KEY_VOL_DOWN_MD 	(101 | GPIO_IN)
+#define GPIO102_MAGICIAN_KEY_PHONE_MD		(102 | GPIO_IN)
+#define GPIO103_MAGICIAN_LED_KP_MD		(103 | GPIO_OUT)
+#define GPIO104_MAGICIAN_LCD_POWER_1_MD 	(104 | GPIO_OUT)
+#define GPIO105_MAGICIAN_LCD_POWER_2_MD		(105 | GPIO_OUT)
+#define GPIO106_MAGICIAN_LCD_POWER_3_MD		(106 | GPIO_OUT)
+#define GPIO107_MAGICIAN_DS1WM_IRQ_MD		(107 | GPIO_IN)
+#define GPIO108_MAGICIAN_GSM_READY_MD		(108 | GPIO_IN)
+#define GPIO114_MAGICIAN_UNKNOWN_MD		(114 | GPIO_OUT)
+#define GPIO115_MAGICIAN_nPEN_IRQ_MD		(115 | GPIO_IN)
+#define GPIO116_MAGICIAN_nCAM_EN_MD		(116 | GPIO_OUT)
+#define GPIO119_MAGICIAN_UNKNOWN_MD		(119 | GPIO_OUT)
+#define GPIO120_MAGICIAN_UNKNOWN_MD		(120 | GPIO_OUT)
+
+#endif /* _MAGICIAN_H_ */
-- 
cgit v1.2.3


From e1d9b9532522f4a04b925e151c1790e669312c55 Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Thu, 13 Dec 2007 10:41:43 +0800
Subject: [ARM] pxa: add basic support for Littleton (PXA3xx Form Factor
 Platform)

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/Kconfig            |   5 +
 arch/arm/mach-pxa/Makefile           |   1 +
 arch/arm/mach-pxa/littleton.c        | 325 +++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-pxa/littleton.h |   6 +
 4 files changed, 337 insertions(+)
 create mode 100644 arch/arm/mach-pxa/littleton.c
 create mode 100644 include/asm-arm/arch-pxa/littleton.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index 619f6fc2bfe1..c73da5e33067 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -107,6 +107,11 @@ config MACH_ZYLONITE
 	bool "PXA3xx Development Platform"
 	select PXA3xx
 
+config MACH_LITTLETON
+	bool "PXA3xx Form Factor Platform (aka Littleton)"
+	select PXA3xx
+	select PXA_SSP
+
 config MACH_ARMCORE
 	bool "CompuLab CM-X270 modules"
 	select PXA27x
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index fca4e819dc5d..5e9aec57642d 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -30,6 +30,7 @@ ifeq ($(CONFIG_MACH_ZYLONITE),y)
   obj-$(CONFIG_CPU_PXA300)	+= zylonite_pxa300.o
   obj-$(CONFIG_CPU_PXA320)	+= zylonite_pxa320.o
 endif
+obj-$(CONFIG_MACH_LITTLETON)	+= littleton.o
 
 obj-$(CONFIG_MACH_ARMCORE)      += cm-x270.o
 
diff --git a/arch/arm/mach-pxa/littleton.c b/arch/arm/mach-pxa/littleton.c
new file mode 100644
index 000000000000..e408a3a89bba
--- /dev/null
+++ b/arch/arm/mach-pxa/littleton.c
@@ -0,0 +1,325 @@
+/*
+ *  linux/arch/arm/mach-pxa/littleton.c
+ *
+ *  Support for the Marvell Littleton Development Platform.
+ *
+ *  Author:	Jason Chagas (largely modified code)
+ *  Created:	Nov 20, 2006
+ *  Copyright:	(C) Copyright 2006 Marvell International Ltd.
+ *
+ *  2007-11-22  modified to align with latest kernel
+ *              eric miao <eric.miao@marvell.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  publishhed by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+
+#include <asm/types.h>
+#include <asm/setup.h>
+#include <asm/memory.h>
+#include <asm/mach-types.h>
+#include <asm/hardware.h>
+#include <asm/irq.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/irq.h>
+
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/mfp-pxa300.h>
+#include <asm/arch/gpio.h>
+#include <asm/arch/pxafb.h>
+#include <asm/arch/ssp.h>
+#include <asm/arch/littleton.h>
+
+#include "generic.h"
+
+#define ARRAY_AND_SIZE(x)	(x), ARRAY_SIZE(x)
+
+/* Littleton MFP configurations */
+static mfp_cfg_t littleton_mfp_cfg[] __initdata = {
+	/* LCD */
+	GPIO54_LCD_LDD_0,
+	GPIO55_LCD_LDD_1,
+	GPIO56_LCD_LDD_2,
+	GPIO57_LCD_LDD_3,
+	GPIO58_LCD_LDD_4,
+	GPIO59_LCD_LDD_5,
+	GPIO60_LCD_LDD_6,
+	GPIO61_LCD_LDD_7,
+	GPIO62_LCD_LDD_8,
+	GPIO63_LCD_LDD_9,
+	GPIO64_LCD_LDD_10,
+	GPIO65_LCD_LDD_11,
+	GPIO66_LCD_LDD_12,
+	GPIO67_LCD_LDD_13,
+	GPIO68_LCD_LDD_14,
+	GPIO69_LCD_LDD_15,
+	GPIO70_LCD_LDD_16,
+	GPIO71_LCD_LDD_17,
+	GPIO72_LCD_FCLK,
+	GPIO73_LCD_LCLK,
+	GPIO74_LCD_PCLK,
+	GPIO75_LCD_BIAS,
+
+	/* SSP2 */
+	GPIO25_SSP2_SCLK,
+	GPIO17_SSP2_FRM,
+	GPIO27_SSP2_TXD,
+
+	/* Debug Ethernet */
+	GPIO90_GPIO,
+};
+
+static struct resource smc91x_resources[] = {
+	[0] = {
+		.start	= (LITTLETON_ETH_PHYS + 0x300),
+		.end	= (LITTLETON_ETH_PHYS + 0xfffff),
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_GPIO(mfp_to_gpio(MFP_PIN_GPIO90)),
+		.end	= IRQ_GPIO(mfp_to_gpio(MFP_PIN_GPIO90)),
+		.flags	= IORESOURCE_IRQ,
+	}
+};
+
+static struct platform_device smc91x_device = {
+	.name		= "smc91x",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(smc91x_resources),
+	.resource	= smc91x_resources,
+};
+
+#if defined(CONFIG_FB_PXA) || defined(CONFIG_FB_PXA_MODULES)
+/* use bit 30, 31 as the indicator of command parameter number */
+#define CMD0(x)		((0x00000000) | ((x) << 9))
+#define CMD1(x, x1)	((0x40000000) | ((x) << 9) | 0x100 | (x1))
+#define CMD2(x, x1, x2)	((0x80000000) | ((x) << 18) | 0x20000 |\
+			 ((x1) << 9) | 0x100 | (x2))
+
+static uint32_t lcd_panel_reset[] = {
+	CMD0(0x1), /* reset */
+	CMD0(0x0), /* nop */
+	CMD0(0x0), /* nop */
+	CMD0(0x0), /* nop */
+};
+
+static uint32_t lcd_panel_on[] = {
+	CMD0(0x29),		/* Display ON */
+	CMD2(0xB8, 0xFF, 0xF9),	/* Output Control */
+	CMD0(0x11),		/* Sleep out */
+	CMD1(0xB0, 0x16),	/* Wake */
+};
+
+static uint32_t lcd_panel_off[] = {
+	CMD0(0x28),		/* Display OFF */
+	CMD2(0xB8, 0x80, 0x02),	/* Output Control */
+	CMD0(0x10),		/* Sleep in */
+	CMD1(0xB0, 0x00),	/* Deep stand by in */
+};
+
+static uint32_t lcd_vga_pass_through[] = {
+	CMD1(0xB0, 0x16),
+	CMD1(0xBC, 0x80),
+	CMD1(0xE1, 0x00),
+	CMD1(0x36, 0x50),
+	CMD1(0x3B, 0x00),
+};
+
+static uint32_t lcd_qvga_pass_through[] = {
+	CMD1(0xB0, 0x16),
+	CMD1(0xBC, 0x81),
+	CMD1(0xE1, 0x00),
+	CMD1(0x36, 0x50),
+	CMD1(0x3B, 0x22),
+};
+
+static uint32_t lcd_vga_transfer[] = {
+	CMD1(0xcf, 0x02), 	/* Blanking period control (1) */
+	CMD2(0xd0, 0x08, 0x04),	/* Blanking period control (2) */
+	CMD1(0xd1, 0x01),	/* CKV timing control on/off */
+	CMD2(0xd2, 0x14, 0x00),	/* CKV 1,2 timing control */
+	CMD2(0xd3, 0x1a, 0x0f),	/* OEV timing control */
+	CMD2(0xd4, 0x1f, 0xaf),	/* ASW timing control (1) */
+	CMD1(0xd5, 0x14),	/* ASW timing control (2) */
+	CMD0(0x21),		/* Invert for normally black display */
+	CMD0(0x29),		/* Display on */
+};
+
+static uint32_t lcd_qvga_transfer[] = {
+	CMD1(0xd6, 0x02),	/* Blanking period control (1) */
+	CMD2(0xd7, 0x08, 0x04),	/* Blanking period control (2) */
+	CMD1(0xd8, 0x01),	/* CKV timing control on/off */
+	CMD2(0xd9, 0x00, 0x08),	/* CKV 1,2 timing control */
+	CMD2(0xde, 0x05, 0x0a),	/* OEV timing control */
+	CMD2(0xdf, 0x0a, 0x19),	/* ASW timing control (1) */
+	CMD1(0xe0, 0x0a),	/* ASW timing control (2) */
+	CMD0(0x21),		/* Invert for normally black display */
+	CMD0(0x29),		/* Display on */
+};
+
+static uint32_t lcd_panel_config[] = {
+	CMD2(0xb8, 0xff, 0xf9),	/* Output control */
+	CMD0(0x11),		/* sleep out */
+	CMD1(0xba, 0x01),	/* Display mode (1) */
+	CMD1(0xbb, 0x00),	/* Display mode (2) */
+	CMD1(0x3a, 0x60),	/* Display mode 18-bit RGB */
+	CMD1(0xbf, 0x10),	/* Drive system change control */
+	CMD1(0xb1, 0x56),	/* Booster operation setup */
+	CMD1(0xb2, 0x33),	/* Booster mode setup */
+	CMD1(0xb3, 0x11),	/* Booster frequency setup */
+	CMD1(0xb4, 0x02),	/* Op amp/system clock */
+	CMD1(0xb5, 0x35),	/* VCS voltage */
+	CMD1(0xb6, 0x40),	/* VCOM voltage */
+	CMD1(0xb7, 0x03),	/* External display signal */
+	CMD1(0xbd, 0x00),	/* ASW slew rate */
+	CMD1(0xbe, 0x00),	/* Dummy data for QuadData operation */
+	CMD1(0xc0, 0x11),	/* Sleep out FR count (A) */
+	CMD1(0xc1, 0x11),	/* Sleep out FR count (B) */
+	CMD1(0xc2, 0x11),	/* Sleep out FR count (C) */
+	CMD2(0xc3, 0x20, 0x40),	/* Sleep out FR count (D) */
+	CMD2(0xc4, 0x60, 0xc0),	/* Sleep out FR count (E) */
+	CMD2(0xc5, 0x10, 0x20),	/* Sleep out FR count (F) */
+	CMD1(0xc6, 0xc0),	/* Sleep out FR count (G) */
+	CMD2(0xc7, 0x33, 0x43),	/* Gamma 1 fine tuning (1) */
+	CMD1(0xc8, 0x44),	/* Gamma 1 fine tuning (2) */
+	CMD1(0xc9, 0x33),	/* Gamma 1 inclination adjustment */
+	CMD1(0xca, 0x00),	/* Gamma 1 blue offset adjustment */
+	CMD2(0xec, 0x01, 0xf0),	/* Horizontal clock cycles */
+};
+
+static void ssp_reconfig(struct ssp_dev *dev, int nparam)
+{
+	static int last_nparam = -1;
+
+	/* check if it is necessary to re-config SSP */
+	if (nparam == last_nparam)
+		return;
+
+	ssp_disable(dev);
+	ssp_config(dev, (nparam == 2) ? 0x0010058a : 0x00100581, 0x18, 0, 0);
+
+	last_nparam = nparam;
+}
+
+static void ssp_send_cmd(uint32_t *cmd, int num)
+{
+	static int ssp_initialized;
+	static struct ssp_dev ssp2;
+
+	int i;
+
+	if (!ssp_initialized) {
+		ssp_init(&ssp2, 2, SSP_NO_IRQ);
+		ssp_initialized = 1;
+	}
+
+	clk_enable(ssp2.ssp->clk);
+	for (i = 0; i < num; i++, cmd++) {
+		ssp_reconfig(&ssp2, (*cmd >> 30) & 0x3);
+		ssp_write_word(&ssp2, *cmd & 0x3fffffff);
+
+		/* FIXME: ssp_flush() is mandatory here to work */
+		ssp_flush(&ssp2);
+	}
+	clk_disable(ssp2.ssp->clk);
+}
+
+static void littleton_lcd_power(int on, struct fb_var_screeninfo *var)
+{
+	if (on) {
+		ssp_send_cmd(ARRAY_AND_SIZE(lcd_panel_on));
+		ssp_send_cmd(ARRAY_AND_SIZE(lcd_panel_reset));
+		if (var->xres > 240) {
+			/* VGA */
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_vga_pass_through));
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_panel_config));
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_vga_transfer));
+		} else {
+			/* QVGA */
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_qvga_pass_through));
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_panel_config));
+			ssp_send_cmd(ARRAY_AND_SIZE(lcd_qvga_transfer));
+		}
+	} else
+		ssp_send_cmd(ARRAY_AND_SIZE(lcd_panel_off));
+}
+
+static struct pxafb_mode_info tpo_tdo24mtea1_modes[] = {
+	[0] = {
+		/* VGA */
+		.pixclock	= 38250,
+		.xres		= 480,
+		.yres		= 640,
+		.bpp		= 16,
+		.hsync_len	= 8,
+		.left_margin	= 8,
+		.right_margin	= 24,
+		.vsync_len	= 2,
+		.upper_margin	= 2,
+		.lower_margin	= 4,
+		.sync		= 0,
+	},
+	[1] = {
+		/* QVGA */
+		.pixclock	= 153000,
+		.xres		= 240,
+		.yres		= 320,
+		.bpp		= 16,
+		.hsync_len	= 8,
+		.left_margin	= 8,
+		.right_margin	= 88,
+		.vsync_len	= 2,
+		.upper_margin	= 2,
+		.lower_margin	= 2,
+		.sync		= 0,
+	},
+};
+
+static struct pxafb_mach_info littleton_lcd_info = {
+	.modes			= tpo_tdo24mtea1_modes,
+	.num_modes		= 2,
+	.lccr0			= LCCR0_Act,
+	.lccr3			= LCCR3_HSP | LCCR3_VSP,
+	.pxafb_lcd_power	= littleton_lcd_power,
+};
+
+static void littleton_init_lcd(void)
+{
+	set_pxa_fb_info(&littleton_lcd_info);
+}
+#else
+static inline void littleton_init_lcd(void) {};
+#endif /* CONFIG_FB_PXA || CONFIG_FB_PXA_MODULES */
+
+static void __init littleton_init(void)
+{
+	/* initialize MFP configurations */
+	pxa3xx_mfp_config(ARRAY_AND_SIZE(littleton_mfp_cfg));
+
+	/*
+	 * Note: we depend bootloader set the correct
+	 * value to MSC register for SMC91x.
+	 */
+	platform_device_register(&smc91x_device);
+
+	littleton_init_lcd();
+}
+
+MACHINE_START(LITTLETON, "Marvell Form Factor Development Platform (aka Littleton)")
+	.phys_io	= 0x40000000,
+	.boot_params	= 0xa0000100,
+	.io_pg_offst	= (io_p2v(0x40000000) >> 18) & 0xfffc,
+	.map_io		= pxa_map_io,
+	.init_irq	= pxa3xx_init_irq,
+	.timer		= &pxa_timer,
+	.init_machine	= littleton_init,
+MACHINE_END
diff --git a/include/asm-arm/arch-pxa/littleton.h b/include/asm-arm/arch-pxa/littleton.h
new file mode 100644
index 000000000000..79d209b826f4
--- /dev/null
+++ b/include/asm-arm/arch-pxa/littleton.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_ARCH_ZYLONITE_H
+#define __ASM_ARCH_ZYLONITE_H
+
+#define LITTLETON_ETH_PHYS	0x30000000
+
+#endif /* __ASM_ARCH_ZYLONITE_H */
-- 
cgit v1.2.3


From 34e31d871ee4b6a9f6c5504da7d6dcc24967844c Mon Sep 17 00:00:00 2001
From: Robert Schwebel <robert@schwebel.de>
Date: Tue, 8 Jan 2008 08:44:23 +0100
Subject: [ARM] 4744/1: pcm027: add support for phyCORE-PXA270 CPU module

This patch adds main support for the generic phyCORE-PXA270 CPU module
(aka PCM-027). Its as generic as possible to support any kind of baseboard.

Note: Neither the CPU module nor the pcm027.c implementation can work without
a baseboard support. Baseboard support can be added by the PCM-990 or any
custom variant.

V2:
 After comments by Eric Miao:
  - Currently unsupported devices moved into separate patch
  - direct call of baseboard initialisation

V3:
 After comments by Russell King
  - sort include files
  - setting RTC bit for power control removed
 - style problems fixed (discovered by checkpatch.pl)

Signed-off-by: Juergen Beisert <j.beisert@pengutronix.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/Kconfig         |   5 +
 arch/arm/mach-pxa/Makefile        |   1 +
 arch/arm/mach-pxa/pcm027.c        | 216 ++++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-pxa/irqs.h   |  10 +-
 include/asm-arm/arch-pxa/pcm027.h |  75 +++++++++++++
 5 files changed, 306 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-pxa/pcm027.c
 create mode 100644 include/asm-arm/arch-pxa/pcm027.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index c73da5e33067..958303c100b0 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -123,6 +123,11 @@ config MACH_MAGICIAN
 	select PXA27x
 	select IWMMXT
 
+config MACH_PCM027
+	bool "Phytec phyCORE-PXA270 CPU module (PCM-027)"
+	select PXA27x
+	select IWMMXT
+
 endchoice
 
 if PXA_SHARPSL
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 5e9aec57642d..61d301805c80 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_PXA_SHARP_C7xx)	+= corgi.o corgi_ssp.o corgi_lcd.o sharpsl_pm.o cor
 obj-$(CONFIG_PXA_SHARP_Cxx00)	+= spitz.o corgi_ssp.o corgi_lcd.o sharpsl_pm.o spitz_pm.o
 obj-$(CONFIG_MACH_AKITA)	+= akita-ioexp.o
 obj-$(CONFIG_MACH_POODLE)	+= poodle.o corgi_ssp.o
+obj-$(CONFIG_MACH_PCM027)	+= pcm027.o
 obj-$(CONFIG_MACH_TOSA)		+= tosa.o
 obj-$(CONFIG_MACH_EM_X270)	+= em-x270.o
 obj-$(CONFIG_MACH_MAGICIAN)	+= magician.o
diff --git a/arch/arm/mach-pxa/pcm027.c b/arch/arm/mach-pxa/pcm027.c
new file mode 100644
index 000000000000..540c3bba5f9a
--- /dev/null
+++ b/arch/arm/mach-pxa/pcm027.c
@@ -0,0 +1,216 @@
+/*
+ *  linux/arch/arm/mach-pxa/pcm027.c
+ *  Support for the Phytec phyCORE-PXA270 CPU card (aka PCM-027).
+ *
+ *  Refer
+ *   http://www.phytec.com/products/sbc/ARM-XScale/phyCORE-XScale-PXA270.html
+ *  for additional hardware info
+ *
+ *  Author:	Juergen Kilb
+ *  Created:	April 05, 2005
+ *  Copyright:	Phytec Messtechnik GmbH
+ *  e-Mail:	armlinux@phytec.de
+ *
+ *  based on Intel Mainstone Board
+ *
+ *  Copyright 2007 Juergen Beisert @ Pengutronix (j.beisert@pengutronix.de)
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/mtd/physmap.h>
+#include <linux/spi/spi.h>
+#include <linux/leds.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+#include <asm/arch/hardware.h>
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx_spi.h>
+#include <asm/arch/pcm027.h>
+#include "generic.h"
+
+/*
+ * ABSTRACT:
+ *
+ * The PXA270 processor comes with a bunch of hardware on its silicon.
+ * Not all of this hardware can be used at the same time and not all
+ * is routed to module's connectors. Also it depends on the baseboard, what
+ * kind of hardware can be used in which way.
+ * -> So this file supports the main devices on the CPU card only!
+ * Refer pcm990-baseboard.c how to extend this features to get a full
+ * blown system with many common interfaces.
+ *
+ * The PCM-027 supports the following interfaces through its connectors and
+ * will be used in pcm990-baseboard.c:
+ *
+ * - LCD support
+ * - MMC support
+ * - IDE/CF card
+ * - FFUART
+ * - BTUART
+ * - IRUART
+ * - AC97
+ * - SSP
+ * - SSP3
+ *
+ * Claimed GPIOs:
+ * GPIO0 -> IRQ input from RTC
+ * GPIO2 -> SYS_ENA*)
+ * GPIO3 -> PWR_SCL
+ * GPIO4 -> PWR_SDA
+ * GPIO5 -> PowerCap0*)
+ * GPIO6 -> PowerCap1*)
+ * GPIO7 -> PowerCap2*)
+ * GPIO8 -> PowerCap3*)
+ * GPIO15 -> /CS1
+ * GPIO20 -> /CS2
+ * GPIO21 -> /CS3
+ * GPIO33 -> /CS5 network controller select
+ * GPIO52 -> IRQ from network controller
+ * GPIO78 -> /CS2
+ * GPIO80 -> /CS4
+ * GPIO90 -> LED0
+ * GPIO91 -> LED1
+ * GPIO114 -> IRQ from CAN controller
+ * GPIO117 -> SCL
+ * GPIO118 -> SDA
+ *
+ * *) CPU internal use only
+ */
+
+/*
+ * SMC91x network controller specific stuff
+ */
+static struct resource smc91x_resources[] = {
+	[0] = {
+		.start	= PCM027_ETH_PHYS + 0x300,
+		.end	= PCM027_ETH_PHYS + PCM027_ETH_SIZE,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= PCM027_ETH_IRQ,
+		.end	= PCM027_ETH_IRQ,
+		/* note: smc91x's driver doesn't use the trigger bits yet */
+		.flags	= IORESOURCE_IRQ | PCM027_ETH_IRQ_EDGE,
+	}
+};
+
+static struct platform_device smc91x_device = {
+	.name		= "smc91x",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(smc91x_resources),
+	.resource	= smc91x_resources,
+};
+
+static struct physmap_flash_data pcm027_flash_data = {
+	.width  = 4,
+};
+
+static struct resource pcm027_flash_resource = {
+	.start          = PCM027_FLASH_PHYS,
+	.end            = PCM027_FLASH_PHYS + PCM027_FLASH_SIZE - 1 ,
+	.flags          = IORESOURCE_MEM,
+};
+
+static struct platform_device pcm027_flash = {
+	.name           = "physmap-flash",
+	.id             = 0,
+	.dev            = {
+		.platform_data  = &pcm027_flash_data,
+	},
+	.resource       = &pcm027_flash_resource,
+	.num_resources  = 1,
+};
+
+#ifdef CONFIG_LEDS_GPIO
+
+static struct gpio_led pcm027_led[] = {
+	{
+		.name = "led0:red",	/* FIXME */
+		.gpio = PCM027_LED_CPU
+	},
+	{
+		.name = "led1:green",	/* FIXME */
+		.gpio = PCM027_LED_HEARD_BEAT
+	},
+};
+
+static struct gpio_led_platform_data pcm027_led_data = {
+	.num_leds	= ARRAY_SIZE(pcm027_led),
+	.leds		= pcm027_led
+};
+
+static struct platform_device pcm027_led_dev = {
+	.name		= "leds-gpio",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &pcm027_led_data,
+	},
+};
+
+#endif /* CONFIG_LEDS_GPIO */
+
+/*
+ * declare the available device resources on this board
+ */
+static struct platform_device *devices[] __initdata = {
+	&smc91x_device,
+	&pcm027_flash,
+#ifdef CONFIG_LEDS_GPIO
+	&pcm027_led_dev
+#endif
+};
+
+/*
+ * pcm027_init - breath some life into the board
+ */
+static void __init pcm027_init(void)
+{
+	/* system bus arbiter setting
+	 * - Core_Park
+	 * - LCD_wt:DMA_wt:CORE_Wt = 2:3:4
+	 */
+	ARB_CNTRL = ARB_CORE_PARK | 0x234;
+
+	platform_add_devices(devices, ARRAY_SIZE(devices));
+
+	/* LEDs (on demand only) */
+#ifdef CONFIG_LEDS_GPIO
+	pxa_gpio_mode(PCM027_LED_CPU | GPIO_OUT);
+	pxa_gpio_mode(PCM027_LED_HEARD_BEAT | GPIO_OUT);
+#endif /* CONFIG_LEDS_GPIO */
+
+	/* at last call the baseboard to initialize itself */
+#ifdef CONFIG_MACH_PCM990_BASEBOARD
+	pcm990_baseboard_init();
+#endif
+}
+
+static void __init pcm027_map_io(void)
+{
+	pxa_map_io();
+
+	/* initialize sleep mode regs (wake-up sources, etc) */
+	PGSR0 = 0x01308000;
+	PGSR1 = 0x00CF0002;
+	PGSR2 = 0x0E294000;
+	PGSR3 = 0x0000C000;
+	PWER  = 0x40000000 | PWER_GPIO0 | PWER_GPIO1;
+	PRER  = 0x00000000;
+	PFER  = 0x00000003;
+}
+
+MACHINE_START(PCM027, "Phytec Messtechnik GmbH phyCORE-PXA270")
+	/* Maintainer: Pengutronix */
+	.boot_params	= 0xa0000100,
+	.phys_io	= 0x40000000,
+	.io_pg_offst	= (io_p2v(0x40000000) >> 18) & 0xfffc,
+	.map_io		= pcm027_map_io,
+	.init_irq	= pxa27x_init_irq,
+	.timer		= &pxa_timer,
+	.init_machine	= pcm027_init,
+MACHINE_END
diff --git a/include/asm-arm/arch-pxa/irqs.h b/include/asm-arm/arch-pxa/irqs.h
index b76ee6d1f5b4..c562b972a4a6 100644
--- a/include/asm-arm/arch-pxa/irqs.h
+++ b/include/asm-arm/arch-pxa/irqs.h
@@ -180,7 +180,8 @@
 #define NR_IRQS			(IRQ_LOCOMO_SPI_TEND + 1)
 #elif defined(CONFIG_ARCH_LUBBOCK) || \
       defined(CONFIG_MACH_LOGICPD_PXA270) || \
-      defined(CONFIG_MACH_MAINSTONE)
+      defined(CONFIG_MACH_MAINSTONE) || \
+      defined(CONFIG_MACH_PCM027)
 #define NR_IRQS			(IRQ_BOARD_END)
 #else
 #define NR_IRQS			(IRQ_BOARD_START)
@@ -227,6 +228,13 @@
 #define IRQ_LOCOMO_LT_BASE	(IRQ_BOARD_START + 2)
 #define IRQ_LOCOMO_SPI_BASE	(IRQ_BOARD_START + 3)
 
+/* phyCORE-PXA270 (PCM027) Interrupts */
+#define PCM027_IRQ(x)          (IRQ_BOARD_START + (x))
+#define PCM027_BTDET_IRQ       PCM027_IRQ(0)
+#define PCM027_FF_RI_IRQ       PCM027_IRQ(1)
+#define PCM027_MMCDET_IRQ      PCM027_IRQ(2)
+#define PCM027_PM_5V_IRQ       PCM027_IRQ(3)
+
 /* ITE8152 irqs */
 /* add IT8152 IRQs beyond BOARD_END */
 #ifdef CONFIG_PCI_HOST_ITE8152
diff --git a/include/asm-arm/arch-pxa/pcm027.h b/include/asm-arm/arch-pxa/pcm027.h
new file mode 100644
index 000000000000..7beae1472c3e
--- /dev/null
+++ b/include/asm-arm/arch-pxa/pcm027.h
@@ -0,0 +1,75 @@
+/*
+ * linux/include/asm-arm/arch-pxa/pcm027.h
+ *
+ * (c) 2003 Phytec Messtechnik GmbH <armlinux@phytec.de>
+ * (c) 2007 Juergen Beisert <j.beisert@pengutronix.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+/*
+ * Definitions of CPU card resources only
+ */
+
+/* I2C RTC */
+#define PCM027_RTC_IRQ_GPIO	0
+#define PCM027_RTC_IRQ		IRQ_GPIO(PCM027_RTC_IRQ_GPIO)
+#define PCM027_RTC_IRQ_EDGE	IRQ_TYPE_EDGE_FALLING
+#define ADR_PCM027_RTC		0x51	/* I2C address */
+
+/* I2C EEPROM */
+#define ADR_PCM027_EEPROM	0x54	/* I2C address */
+
+/* Ethernet chip (SMSC91C111) */
+#define PCM027_ETH_IRQ_GPIO	52
+#define PCM027_ETH_IRQ		IRQ_GPIO(PCM027_ETH_IRQ_GPIO)
+#define PCM027_ETH_IRQ_EDGE	IRQ_TYPE_EDGE_RISING
+#define PCM027_ETH_PHYS		PXA_CS5_PHYS
+#define PCM027_ETH_SIZE		(1*1024*1024)
+
+/* CAN controller SJA1000 (unsupported yet) */
+#define PCM027_CAN_IRQ_GPIO	114
+#define PCM027_CAN_IRQ		IRQ_GPIO(PCM027_CAN_IRQ_GPIO)
+#define PCM027_CAN_IRQ_EDGE	IRQ_TYPE_EDGE_FALLING
+#define PCM027_CAN_PHYS		0x22000000
+#define PCM027_CAN_SIZE		0x100
+
+/* SPI GPIO expander (unsupported yet) */
+#define PCM027_EGPIO_IRQ_GPIO	27
+#define PCM027_EGPIO_IRQ	IRQ_GPIO(PCM027_EGPIO_IRQ_GPIO)
+#define PCM027_EGPIO_IRQ_EDGE	IRQ_TYPE_EDGE_FALLING
+#define PCM027_EGPIO_CS		24
+/*
+ * TODO: Switch this pin from dedicated usage to GPIO if
+ * more than the MAX7301 device is connected to this SPI bus
+ */
+#define PCM027_EGPIO_CS_MODE	GPIO24_SFRM_MD
+
+/* Flash memory */
+#define PCM027_FLASH_PHYS	0x00000000
+#define PCM027_FLASH_SIZE	0x02000000
+
+/* onboard LEDs connected to GPIO */
+#define PCM027_LED_CPU		90
+#define PCM027_LED_HEARD_BEAT	91
+
+/*
+ * This CPU module needs a baseboard to work. After basic initializing
+ * its own devices, it calls baseboard's init function.
+ * TODO: Add your own basebaord init function and call it from
+ * inside pcm027_init(). This example here is for the developmen board.
+ * Refer pcm990-baseboard.c
+ */
+extern void pcm990_baseboard_init(void);
-- 
cgit v1.2.3


From 2e927b76267a57a44c018ffcc64cde5fedde1fcf Mon Sep 17 00:00:00 2001
From: Robert Schwebel <robert@schwebel.de>
Date: Tue, 8 Jan 2008 08:52:04 +0100
Subject: [ARM] 4747/1: pcm027: support for pcm990 baseboard for phyCORE-PXA270

This patch adds baseboard support for the phyCORE-PXA270 development
kit (aka PCM-990).

This example shows how to use some phyCORE-PXA270 CPU module features
on a baseboard in a standard manner. It could be used as a starting
point for custom baseboard development.

V2:
 After comments by Eric Miao:
  - IRQ chained handler fixed
  - video/graphic support moved to separate patch
  - ifdef/endif hell reduced ;-)

V3:
 After comments by Russell King
  - initialise the mmci platform data statically

V4:
 After comments by Russell King
  - wrong return value in pcm990_mci_init() fixed

Signed-off-by: Juergen Beisert <j.beisert@pengutronix.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/Kconfig                   |   9 +
 arch/arm/mach-pxa/Makefile                  |   1 +
 arch/arm/mach-pxa/pcm990-baseboard.c        | 330 ++++++++++++++++++++++++++++
 include/asm-arm/arch-pxa/pcm990_baseboard.h | 275 +++++++++++++++++++++++
 4 files changed, 615 insertions(+)
 create mode 100644 arch/arm/mach-pxa/pcm990-baseboard.c
 create mode 100644 include/asm-arm/arch-pxa/pcm990_baseboard.h

(limited to 'include')

diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index 958303c100b0..e5b7063ed818 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -130,6 +130,15 @@ config MACH_PCM027
 
 endchoice
 
+choice
+	prompt "Used baseboard"
+	depends on MACH_PCM027
+
+config MACH_PCM990_BASEBOARD
+	bool "PHYTEC PCM-990 development board"
+
+endchoice
+
 if PXA_SHARPSL
 
 choice
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 61d301805c80..1af56805b954 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -21,6 +21,7 @@ obj-$(CONFIG_PXA_SHARP_Cxx00)	+= spitz.o corgi_ssp.o corgi_lcd.o sharpsl_pm.o sp
 obj-$(CONFIG_MACH_AKITA)	+= akita-ioexp.o
 obj-$(CONFIG_MACH_POODLE)	+= poodle.o corgi_ssp.o
 obj-$(CONFIG_MACH_PCM027)	+= pcm027.o
+obj-$(CONFIG_MACH_PCM990_BASEBOARD)	+= pcm990-baseboard.o
 obj-$(CONFIG_MACH_TOSA)		+= tosa.o
 obj-$(CONFIG_MACH_EM_X270)	+= em-x270.o
 obj-$(CONFIG_MACH_MAGICIAN)	+= magician.o
diff --git a/arch/arm/mach-pxa/pcm990-baseboard.c b/arch/arm/mach-pxa/pcm990-baseboard.c
new file mode 100644
index 000000000000..3dda16a20049
--- /dev/null
+++ b/arch/arm/mach-pxa/pcm990-baseboard.c
@@ -0,0 +1,330 @@
+/*
+ *  arch/arm/mach-pxa/pcm990-baseboard.c
+ *  Support for the Phytec phyCORE-PXA270 Development Platform (PCM-990).
+ *
+ *  Refer
+ *   http://www.phytec.com/products/rdk/ARM-XScale/phyCORE-XScale-PXA270.html
+ *  for additional hardware info
+ *
+ *  Author:	Juergen Kilb
+ *  Created:	April 05, 2005
+ *  Copyright:	Phytec Messtechnik GmbH
+ *  e-Mail:	armlinux@phytec.de
+ *
+ *  based on Intel Mainstone Board
+ *
+ *  Copyright 2007 Juergen Beisert @ Pengutronix (j.beisert@pengutronix.de)
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/ide.h>
+#include <asm/mach/map.h>
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/mmc.h>
+#include <asm/arch/ohci.h>
+#include <asm/arch/pcm990_baseboard.h>
+
+/*
+ * The PCM-990 development baseboard uses PCM-027's hardeware in the
+ * following way:
+ *
+ * - LCD support is in use
+ *  - GPIO16 is output for back light on/off with PWM
+ *  - GPIO58 ... GPIO73 are outputs for display data
+ *  - GPIO74 is output output for LCDFCLK
+ *  - GPIO75 is output for LCDLCLK
+ *  - GPIO76 is output for LCDPCLK
+ *  - GPIO77 is output for LCDBIAS
+ * - MMC support is in use
+ *  - GPIO32 is output for MMCCLK
+ *  - GPIO92 is MMDAT0
+ *  - GPIO109 is MMDAT1
+ *  - GPIO110 is MMCS0
+ *  - GPIO111 is MMCS1
+ *  - GPIO112 is MMCMD
+ * - IDE/CF card is in use
+ *  - GPIO48 is output /POE
+ *  - GPIO49 is output /PWE
+ *  - GPIO50 is output /PIOR
+ *  - GPIO51 is output /PIOW
+ *  - GPIO54 is output /PCE2
+ *  - GPIO55 is output /PREG
+ *  - GPIO56 is input /PWAIT
+ *  - GPIO57 is output /PIOS16
+ *  - GPIO79 is output PSKTSEL
+ *  - GPIO85 is output /PCE1
+ * - FFUART is in use
+ *  - GPIO34 is input FFRXD
+ *  - GPIO35 is input FFCTS
+ *  - GPIO36 is input FFDCD
+ *  - GPIO37 is input FFDSR
+ *  - GPIO38 is input FFRI
+ *  - GPIO39 is output FFTXD
+ *  - GPIO40 is output FFDTR
+ *  - GPIO41 is output FFRTS
+ * - BTUART is in use
+ *  - GPIO42 is input BTRXD
+ *  - GPIO43 is output BTTXD
+ *  - GPIO44 is input BTCTS
+ *  - GPIO45 is output BTRTS
+ * - IRUART is in use
+ *  - GPIO46 is input STDRXD
+ *  - GPIO47 is output STDTXD
+ * - AC97 is in use*)
+ *  - GPIO28 is input AC97CLK
+ *  - GPIO29 is input AC97DatIn
+ *  - GPIO30 is output AC97DatO
+ *  - GPIO31 is output AC97SYNC
+ *  - GPIO113 is output AC97_RESET
+ * - SSP is in use
+ *  - GPIO23 is output SSPSCLK
+ *  - GPIO24 is output chip select to Max7301
+ *  - GPIO25 is output SSPTXD
+ *  - GPIO26 is input SSPRXD
+ *  - GPIO27 is input for Max7301 IRQ
+ *  - GPIO53 is input SSPSYSCLK
+ * - SSP3 is in use
+ *  - GPIO81 is output SSPTXD3
+ *  - GPIO82 is input SSPRXD3
+ *  - GPIO83 is output SSPSFRM
+ *  - GPIO84 is output SSPCLK3
+ *
+ * Otherwise claimed GPIOs:
+ * GPIO1 -> IRQ from user switch
+ * GPIO9 -> IRQ from power management
+ * GPIO10 -> IRQ from WML9712 AC97 controller
+ * GPIO11 -> IRQ from IDE controller
+ * GPIO12 -> IRQ from CF controller
+ * GPIO13 -> IRQ from CF controller
+ * GPIO14 -> GPIO free
+ * GPIO15 -> /CS1 selects baseboard's Control CPLD (U7, 16 bit wide data path)
+ * GPIO19 -> GPIO free
+ * GPIO20 -> /SDCS2
+ * GPIO21 -> /CS3 PC card socket select
+ * GPIO33 -> /CS5  network controller select
+ * GPIO78 -> /CS2  (16 bit wide data path)
+ * GPIO80 -> /CS4  (16 bit wide data path)
+ * GPIO86 -> GPIO free
+ * GPIO87 -> GPIO free
+ * GPIO90 -> LED0 on CPU module
+ * GPIO91 -> LED1 on CPI module
+ * GPIO117 -> SCL
+ * GPIO118 -> SDA
+ */
+
+static unsigned long pcm990_irq_enabled;
+
+static void pcm990_mask_ack_irq(unsigned int irq)
+{
+	int pcm990_irq = (irq - PCM027_IRQ(0));
+	PCM990_INTMSKENA = (pcm990_irq_enabled &= ~(1 << pcm990_irq));
+}
+
+static void pcm990_unmask_irq(unsigned int irq)
+{
+	int pcm990_irq = (irq - PCM027_IRQ(0));
+	/* the irq can be acknowledged only if deasserted, so it's done here */
+	PCM990_INTSETCLR |= 1 << pcm990_irq;
+	PCM990_INTMSKENA  = (pcm990_irq_enabled |= (1 << pcm990_irq));
+}
+
+static struct irq_chip pcm990_irq_chip = {
+	.mask_ack	= pcm990_mask_ack_irq,
+	.unmask		= pcm990_unmask_irq,
+};
+
+static void pcm990_irq_handler(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned long pending = (~PCM990_INTSETCLR) & pcm990_irq_enabled;
+
+	do {
+		GEDR(PCM990_CTRL_INT_IRQ_GPIO) =
+					GPIO_bit(PCM990_CTRL_INT_IRQ_GPIO);
+		if (likely(pending)) {
+			irq = PCM027_IRQ(0) + __ffs(pending);
+			desc = irq_desc + irq;
+			desc_handle_irq(irq, desc);
+		}
+		pending = (~PCM990_INTSETCLR) & pcm990_irq_enabled;
+	} while (pending);
+}
+
+static void __init pcm990_init_irq(void)
+{
+	int irq;
+
+	/* setup extra PCM990 irqs */
+	for (irq = PCM027_IRQ(0); irq <= PCM027_IRQ(3); irq++) {
+		set_irq_chip(irq, &pcm990_irq_chip);
+		set_irq_handler(irq, handle_level_irq);
+		set_irq_flags(irq, IRQF_VALID | IRQF_PROBE);
+	}
+
+	PCM990_INTMSKENA = 0x00;	/* disable all Interrupts */
+	PCM990_INTSETCLR = 0xFF;
+
+	set_irq_chained_handler(PCM990_CTRL_INT_IRQ, pcm990_irq_handler);
+	set_irq_type(PCM990_CTRL_INT_IRQ, PCM990_CTRL_INT_IRQ_EDGE);
+}
+
+static int pcm990_mci_init(struct device *dev, irq_handler_t mci_detect_int,
+			void *data)
+{
+	int err;
+
+	/*
+	 * enable GPIO for PXA27x MMC controller
+	 */
+	pxa_gpio_mode(GPIO32_MMCCLK_MD);
+	pxa_gpio_mode(GPIO112_MMCCMD_MD);
+	pxa_gpio_mode(GPIO92_MMCDAT0_MD);
+	pxa_gpio_mode(GPIO109_MMCDAT1_MD);
+	pxa_gpio_mode(GPIO110_MMCDAT2_MD);
+	pxa_gpio_mode(GPIO111_MMCDAT3_MD);
+
+	err = request_irq(PCM027_MMCDET_IRQ, mci_detect_int, IRQF_DISABLED,
+			     "MMC card detect", data);
+	if (err)
+		printk(KERN_ERR "pcm990_mci_init: MMC/SD: can't request MMC "
+				"card detect IRQ\n");
+
+	return err;
+}
+
+static void pcm990_mci_setpower(struct device *dev, unsigned int vdd)
+{
+	struct pxamci_platform_data *p_d = dev->platform_data;
+
+	if ((1 << vdd) & p_d->ocr_mask)
+		__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG5) =
+						PCM990_CTRL_MMC2PWR;
+	else
+		__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG5) =
+						~PCM990_CTRL_MMC2PWR;
+}
+
+static void pcm990_mci_exit(struct device *dev, void *data)
+{
+	free_irq(PCM027_MMCDET_IRQ, data);
+}
+
+#define MSECS_PER_JIFFY (1000/HZ)
+
+static struct pxamci_platform_data pcm990_mci_platform_data = {
+	.detect_delay	= 250 / MSECS_PER_JIFFY,
+	.ocr_mask	= MMC_VDD_32_33 | MMC_VDD_33_34,
+	.init 		= pcm990_mci_init,
+	.setpower 	= pcm990_mci_setpower,
+	.exit		= pcm990_mci_exit,
+};
+
+/*
+ * init OHCI hardware to work with
+ *
+ * Note: Only USB port 1 (host only) is connected
+ *
+ * GPIO88 (USBHPWR#1): overcurrent in, overcurrent when low
+ * GPIO89 (USBHPEN#1): power-on out, on when low
+ */
+static int pcm990_ohci_init(struct device *dev)
+{
+	pxa_gpio_mode(PCM990_USB_OVERCURRENT);
+	pxa_gpio_mode(PCM990_USB_PWR_EN);
+	/*
+	 * disable USB port 2 and 3
+	 * power sense is active low
+	 */
+	UHCHR = ((UHCHR) | UHCHR_PCPL | UHCHR_PSPL | UHCHR_SSEP2 |
+				UHCHR_SSEP3) & ~(UHCHR_SSEP1 | UHCHR_SSE);
+	/*
+	 * wait 10ms after Power on
+	 * overcurrent per port
+	 * power switch per port
+	 */
+	UHCRHDA = (5<<24) | (1<<11) | (1<<8);	/* FIXME: Required? */
+
+	return 0;
+}
+
+static struct pxaohci_platform_data pcm990_ohci_platform_data = {
+	.port_mode	= PMM_PERPORT_MODE,
+	.init		= pcm990_ohci_init,
+	.exit		= NULL,
+};
+
+/*
+ * AC97 support
+ * Note: The connected AC97 mixer also reports interrupts at PCM990_AC97_IRQ
+ */
+static struct resource pxa27x_ac97_resources[] = {
+	[0] = {
+		.start  = 0x40500000,
+		.end	= 0x40500000 + 0xfff,
+		.flags  = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start  = IRQ_AC97,
+		.end    = IRQ_AC97,
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static u64 pxa_ac97_dmamask = 0xffffffffUL;
+
+static struct platform_device pxa27x_device_ac97 = {
+	.name           = "pxa2xx-ac97",
+	.id             = -1,
+	.dev            = {
+		.dma_mask = &pxa_ac97_dmamask,
+		.coherent_dma_mask = 0xffffffff,
+	},
+	.num_resources  = ARRAY_SIZE(pxa27x_ac97_resources),
+	.resource       = pxa27x_ac97_resources,
+};
+
+/*
+ * enable generic access to the base board control CPLDs U6 and U7
+ */
+static struct map_desc pcm990_io_desc[] __initdata = {
+	{
+		.virtual	= PCM990_CTRL_BASE,
+		.pfn		= __phys_to_pfn(PCM990_CTRL_PHYS),
+		.length		= PCM990_CTRL_SIZE,
+		.type		= MT_DEVICE	/* CPLD */
+	}, {
+		.virtual	= PCM990_CF_PLD_BASE,
+		.pfn		= __phys_to_pfn(PCM990_CF_PLD_PHYS),
+		.length		= PCM990_CF_PLD_SIZE,
+		.type		= MT_DEVICE	/* CPLD */
+	}
+};
+
+/*
+ * system init for baseboard usage. Will be called by pcm027 init.
+ *
+ * Add platform devices present on this baseboard and init
+ * them from CPU side as far as required to use them later on
+ */
+void __init pcm990_baseboard_init(void)
+{
+	/* register CPLD access */
+	iotable_init(pcm990_io_desc, ARRAY_SIZE(pcm990_io_desc));
+
+	/* register CPLD's IRQ controller */
+	pcm990_init_irq();
+
+	platform_device_register(&pxa27x_device_ac97);
+
+	/* MMC */
+	pxa_set_mci_info(&pcm990_mci_platform_data);
+
+	/* USB host */
+	pxa_set_ohci_info(&pcm990_ohci_platform_data);
+
+	printk(KERN_INFO"PCM-990 Evaluation baseboard initialized\n");
+}
diff --git a/include/asm-arm/arch-pxa/pcm990_baseboard.h b/include/asm-arm/arch-pxa/pcm990_baseboard.h
new file mode 100644
index 000000000000..b699d0d7bdb2
--- /dev/null
+++ b/include/asm-arm/arch-pxa/pcm990_baseboard.h
@@ -0,0 +1,275 @@
+/*
+ * include/asm-arm/arch-pxa/pcm990_baseboard.h
+ *
+ * (c) 2003 Phytec Messtechnik GmbH <armlinux@phytec.de>
+ * (c) 2007 Juergen Beisert <j.beisert@pengutronix.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#include <asm/arch/pcm027.h>
+
+/*
+ * definitions relevant only when the PCM-990
+ * development base board is in use
+ */
+
+/* CPLD's interrupt controller is connected to PCM-027 GPIO 9 */
+#define PCM990_CTRL_INT_IRQ_GPIO	9
+#define PCM990_CTRL_INT_IRQ		IRQ_GPIO(PCM990_CTRL_INT_IRQ_GPIO)
+#define PCM990_CTRL_INT_IRQ_EDGE	IRQT_RISING
+#define PCM990_CTRL_PHYS		PXA_CS1_PHYS	/* 16-Bit */
+#define PCM990_CTRL_BASE		0xea000000
+#define PCM990_CTRL_SIZE		(1*1024*1024)
+
+#define PCM990_CTRL_PWR_IRQ_GPIO	14
+#define PCM990_CTRL_PWR_IRQ		IRQ_GPIO(PCM990_CTRL_PWR_IRQ_GPIO)
+#define PCM990_CTRL_PWR_IRQ_EDGE	IRQT_RISING
+
+/* visible CPLD (U7) registers */
+#define PCM990_CTRL_REG0	0x0000	/* RESET REGISTER */
+#define PCM990_CTRL_SYSRES	0x0001	/* System RESET REGISTER */
+#define PCM990_CTRL_RESOUT	0x0002	/* RESETOUT Enable REGISTER */
+#define PCM990_CTRL_RESGPIO	0x0004	/* RESETGPIO Enable REGISTER */
+
+#define PCM990_CTRL_REG1	0x0002	/* Power REGISTER */
+#define PCM990_CTRL_5VOFF	0x0001	/* Disable  5V Regulators */
+#define PCM990_CTRL_CANPWR	0x0004	/* Enable CANPWR ADUM */
+#define PCM990_CTRL_PM_5V	0x0008	/* Read 5V OK */
+
+#define PCM990_CTRL_REG2	0x0004	/* LED REGISTER */
+#define PCM990_CTRL_LEDPWR	0x0001	/* POWER LED enable */
+#define PCM990_CTRL_LEDBAS	0x0002	/* BASIS LED enable */
+#define PCM990_CTRL_LEDUSR	0x0004	/* USER LED enable */
+
+#define PCM990_CTRL_REG3	0x0006	/* LCD CTRL REGISTER 3 */
+#define PCM990_CTRL_LCDPWR	0x0001	/* RW LCD Power on */
+#define PCM990_CTRL_LCDON	0x0002	/* RW LCD Latch on */
+#define PCM990_CTRL_LCDPOS1	0x0004	/* RW POS 1 */
+#define PCM990_CTRL_LCDPOS2	0x0008	/* RW POS 2 */
+
+#define PCM990_CTRL_REG4	0x0008	/* MMC1 CTRL REGISTER 4 */
+#define PCM990_CTRL_MMC1PWR	0x0001 /* RW MMC1 Power on */
+
+#define PCM990_CTRL_REG5	0x000A	/* MMC2 CTRL REGISTER 5 */
+#define PCM990_CTRL_MMC2PWR	0x0001	/* RW MMC2 Power on */
+#define PCM990_CTRL_MMC2LED	0x0002	/* RW MMC2 LED */
+#define PCM990_CTRL_MMC2DE	0x0004	/* R MMC2 Card detect */
+#define PCM990_CTRL_MMC2WP	0x0008	/* R MMC2 Card write protect */
+
+#define PCM990_CTRL_REG6	0x000C	/* Interrupt Clear REGISTER */
+#define PCM990_CTRL_INTC0	0x0001	/* Clear Reg BT Detect */
+#define PCM990_CTRL_INTC1	0x0002	/* Clear Reg FR RI */
+#define PCM990_CTRL_INTC2	0x0004	/* Clear Reg MMC1 Detect */
+#define PCM990_CTRL_INTC3	0x0008	/* Clear Reg PM_5V off */
+
+#define PCM990_CTRL_REG7	0x000E	/* Interrupt Enable REGISTER */
+#define PCM990_CTRL_ENAINT0	0x0001	/* Enable Int BT Detect */
+#define PCM990_CTRL_ENAINT1	0x0002	/* Enable Int FR RI */
+#define PCM990_CTRL_ENAINT2	0x0004	/* Enable Int MMC1 Detect */
+#define PCM990_CTRL_ENAINT3	0x0008	/* Enable Int PM_5V off */
+
+#define PCM990_CTRL_REG8	0x0014	/* Uart REGISTER */
+#define PCM990_CTRL_FFSD	0x0001	/* BT Uart Enable */
+#define PCM990_CTRL_BTSD	0x0002	/* FF Uart Enable */
+#define PCM990_CTRL_FFRI	0x0004	/* FF Uart RI detect */
+#define PCM990_CTRL_BTRX	0x0008	/* BT Uart Rx detect */
+
+#define PCM990_CTRL_REG9	0x0010	/* AC97 Flash REGISTER */
+#define PCM990_CTRL_FLWP	0x0001	/* pC Flash Write Protect */
+#define PCM990_CTRL_FLDIS	0x0002	/* pC Flash Disable */
+#define PCM990_CTRL_AC97ENA	0x0004	/* Enable AC97 Expansion */
+
+#define PCM990_CTRL_REG10	0x0012	/* GPS-REGISTER */
+#define PCM990_CTRL_GPSPWR	0x0004	/* GPS-Modul Power on */
+#define PCM990_CTRL_GPSENA	0x0008	/* GPS-Modul Enable */
+
+#define PCM990_CTRL_REG11	0x0014	/* Accu REGISTER */
+#define PCM990_CTRL_ACENA	0x0001	/* Charge Enable */
+#define PCM990_CTRL_ACSEL	0x0002	/* Charge Akku -> DC Enable */
+#define PCM990_CTRL_ACPRES	0x0004	/* DC Present */
+#define PCM990_CTRL_ACALARM	0x0008	/* Error Akku */
+
+#define PCM990_CTRL_P2V(x)	((x) - PCM990_CTRL_PHYS + PCM990_CTRL_BASE)
+#define PCM990_CTRL_V2P(x)	((x) - PCM990_CTRL_BASE + PCM990_CTRL_PHYS)
+
+#ifndef __ASSEMBLY__
+#  define __PCM990_CTRL_REG(x) \
+		(*((volatile unsigned char *)PCM990_CTRL_P2V(x)))
+#else
+#  define __PCM990_CTRL_REG(x)	PCM990_CTRL_P2V(x)
+#endif
+
+#define PCM990_INTMSKENA __PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG7)
+#define PCM990_INTSETCLR __PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG6)
+#define PCM990_CTRL0	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG0)
+#define PCM990_CTRL1	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG1)
+#define PCM990_CTRL2	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG2)
+#define PCM990_CTRL3	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG3)
+#define PCM990_CTRL4	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG4)
+#define PCM990_CTRL5	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG5)
+#define PCM990_CTRL6	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG6)
+#define PCM990_CTRL7	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG7)
+#define PCM990_CTRL8	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG8)
+#define PCM990_CTRL9	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG9)
+#define PCM990_CTRL10	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG10)
+#define PCM990_CTRL11	__PCM990_CTRL_REG(PCM990_CTRL_PHYS + PCM990_CTRL_REG11)
+
+
+/*
+ * IDE
+ */
+#define PCM990_IDE_IRQ_GPIO	13
+#define PCM990_IDE_IRQ		IRQ_GPIO(PCM990_IDE_IRQ_GPIO)
+#define PCM990_IDE_IRQ_EDGE	IRQT_RISING
+#define PCM990_IDE_PLD_PHYS	0x20000000	/* 16 bit wide */
+#define PCM990_IDE_PLD_BASE	0xee000000
+#define PCM990_IDE_PLD_SIZE	(1*1024*1024)
+
+/* visible CPLD (U6) registers */
+#define PCM990_IDE_PLD_REG0	0x1000	/* OFFSET IDE REGISTER 0 */
+#define PCM990_IDE_PM5V		0x0004	/* R System VCC_5V */
+#define PCM990_IDE_STBY		0x0008	/* R System StandBy */
+
+#define PCM990_IDE_PLD_REG1	0x1002	/* OFFSET IDE REGISTER 1 */
+#define PCM990_IDE_IDEMODE	0x0001	/* R TrueIDE Mode */
+#define PCM990_IDE_DMAENA	0x0004	/* RW DMA Enable */
+#define PCM990_IDE_DMA1_0	0x0008	/* RW 1=DREQ1 0=DREQ0 */
+
+#define PCM990_IDE_PLD_REG2	0x1004	/* OFFSET IDE REGISTER 2 */
+#define PCM990_IDE_RESENA	0x0001	/* RW IDE Reset Bit enable */
+#define PCM990_IDE_RES		0x0002	/* RW IDE Reset Bit */
+#define PCM990_IDE_RDY		0x0008	/* RDY */
+
+#define PCM990_IDE_PLD_REG3	0x1006	/* OFFSET IDE REGISTER 3 */
+#define PCM990_IDE_IDEOE	0x0001	/* RW Latch on Databus */
+#define PCM990_IDE_IDEON	0x0002	/* RW Latch on Control Address */
+#define PCM990_IDE_IDEIN	0x0004	/* RW Latch on Interrupt usw. */
+
+#define PCM990_IDE_PLD_REG4	0x1008	/* OFFSET IDE REGISTER 4 */
+#define PCM990_IDE_PWRENA	0x0001	/* RW IDE Power enable */
+#define PCM990_IDE_5V		0x0002	/* R IDE Power 5V */
+#define PCM990_IDE_PWG		0x0008	/* R IDE Power is on */
+
+#define PCM990_IDE_PLD_P2V(x) ((x) - PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_BASE)
+#define PCM990_IDE_PLD_V2P(x) ((x) - PCM990_IDE_PLD_BASE + PCM990_IDE_PLD_PHYS)
+
+#ifndef __ASSEMBLY__
+# define  __PCM990_IDE_PLD_REG(x) \
+	(*((volatile unsigned char *)PCM990_IDE_PLD_P2V(x)))
+#else
+# define  __PCM990_IDE_PLD_REG(x)	PCM990_IDE_PLD_P2V(x)
+#endif
+
+#define PCM990_IDE0 \
+	__PCM990_IDE_PLD_REG(PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_REG0)
+#define PCM990_IDE1 \
+	__PCM990_IDE_PLD_REG(PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_REG1)
+#define PCM990_IDE2 \
+	__PCM990_IDE_PLD_REG(PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_REG2)
+#define PCM990_IDE3 \
+	__PCM990_IDE_PLD_REG(PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_REG3)
+#define PCM990_IDE4 \
+	__PCM990_IDE_PLD_REG(PCM990_IDE_PLD_PHYS + PCM990_IDE_PLD_REG4)
+
+/*
+ * Compact Flash
+ */
+#define PCM990_CF_IRQ_GPIO	11
+#define PCM990_CF_IRQ		IRQ_GPIO(PCM990_CF_IRQ_GPIO)
+#define PCM990_CF_IRQ_EDGE	IRQT_RISING
+
+#define PCM990_CF_CD_GPIO	12
+#define PCM990_CF_CD		IRQ_GPIO(PCM990_CF_CD_GPIO)
+#define PCM990_CF_CD_EDGE	IRQT_RISING
+
+#define PCM990_CF_PLD_PHYS	0x30000000	/* 16 bit wide */
+#define PCM990_CF_PLD_BASE	0xef000000
+#define PCM990_CF_PLD_SIZE	(1*1024*1024)
+#define PCM990_CF_PLD_P2V(x)	((x) - PCM990_CF_PLD_PHYS + PCM990_CF_PLD_BASE)
+#define PCM990_CF_PLD_V2P(x)	((x) - PCM990_CF_PLD_BASE + PCM990_CF_PLD_PHYS)
+
+/* visible CPLD (U6) registers */
+#define PCM990_CF_PLD_REG0	0x1000	/* OFFSET CF REGISTER 0 */
+#define PCM990_CF_REG0_LED	0x0001	/* RW LED on */
+#define PCM990_CF_REG0_BLK	0x0002	/* RW LED flash when access */
+#define PCM990_CF_REG0_PM5V	0x0004	/* R System VCC_5V enable */
+#define PCM990_CF_REG0_STBY	0x0008	/* R System StandBy */
+
+#define PCM990_CF_PLD_REG1	0x1002	/* OFFSET CF REGISTER 1 */
+#define PCM990_CF_REG1_IDEMODE	0x0001	/* RW CF card run as TrueIDE */
+#define PCM990_CF_REG1_CF0	0x0002	/* RW CF card at ADDR 0x28000000 */
+
+#define PCM990_CF_PLD_REG2	0x1004	/* OFFSET CF REGISTER 2 */
+#define PCM990_CF_REG2_RES	0x0002	/* RW CF RESET BIT */
+#define PCM990_CF_REG2_RDYENA	0x0004	/* RW Enable CF_RDY */
+#define PCM990_CF_REG2_RDY	0x0008	/* R CF_RDY auf PWAIT */
+
+#define PCM990_CF_PLD_REG3	0x1006	/* OFFSET CF REGISTER 3 */
+#define PCM990_CF_REG3_CFOE	0x0001	/* RW Latch on Databus */
+#define PCM990_CF_REG3_CFON	0x0002	/* RW Latch on Control Address */
+#define PCM990_CF_REG3_CFIN	0x0004	/* RW Latch on Interrupt usw. */
+#define PCM990_CF_REG3_CFCD	0x0008	/* RW Latch on CD1/2 VS1/2 usw */
+
+#define PCM990_CF_PLD_REG4	0x1008	/* OFFSET CF REGISTER 4 */
+#define PCM990_CF_REG4_PWRENA	0x0001	/* RW CF Power on (CD1/2 = "00") */
+#define PCM990_CF_REG4_5_3V	0x0002	/* RW 1 = 5V CF_VCC 0 = 3 V CF_VCC */
+#define PCM990_CF_REG4_3B	0x0004	/* RW 3.0V Backup from VCC (5_3V=0) */
+#define PCM990_CF_REG4_PWG	0x0008	/* R CF-Power is on */
+
+#define PCM990_CF_PLD_REG5	0x100A	/* OFFSET CF REGISTER 5 */
+#define PCM990_CF_REG5_BVD1	0x0001	/* R CF /BVD1 */
+#define PCM990_CF_REG5_BVD2	0x0002	/* R CF /BVD2 */
+#define PCM990_CF_REG5_VS1	0x0004	/* R CF /VS1 */
+#define PCM990_CF_REG5_VS2	0x0008	/* R CF /VS2 */
+
+#define PCM990_CF_PLD_REG6	0x100C	/* OFFSET CF REGISTER 6 */
+#define PCM990_CF_REG6_CD1	0x0001	/* R CF Card_Detect1 */
+#define PCM990_CF_REG6_CD2	0x0002	/* R CF Card_Detect2 */
+
+#ifndef __ASSEMBLY__
+#  define  __PCM990_CF_PLD_REG(x) \
+	(*((volatile unsigned char *)PCM990_CF_PLD_P2V(x)))
+#else
+#  define  __PCM990_CF_PLD_REG(x)	PCM990_CF_PLD_P2V(x)
+#endif
+
+#define PCM990_CF0 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG0)
+#define PCM990_CF1 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG1)
+#define PCM990_CF2 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG2)
+#define PCM990_CF3 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG3)
+#define PCM990_CF4 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG4)
+#define PCM990_CF5 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG5)
+#define PCM990_CF6 __PCM990_CF_PLD_REG(PCM990_CF_PLD_PHYS + PCM990_CF_PLD_REG6)
+
+/*
+ * Wolfson AC97 Touch
+ */
+#define PCM990_AC97_IRQ_GPIO	10
+#define PCM990_AC97_IRQ		IRQ_GPIO(PCM990_AC97_IRQ_GPIO)
+#define PCM990_AC97_IRQ_EDGE	IRQT_RISING
+
+/*
+ * MMC phyCORE
+ */
+#define PCM990_MMC0_IRQ_GPIO	9
+#define PCM990_MMC0_IRQ		IRQ_GPIO(PCM990_MMC0_IRQ_GPIO)
+#define PCM990_MMC0_IRQ_EDGE	IRQT_FALLING
+
+/*
+ * USB phyCore
+ */
+#define PCM990_USB_OVERCURRENT (88 | GPIO_ALT_FN_1_IN)
+#define PCM990_USB_PWR_EN (89 | GPIO_ALT_FN_2_OUT)
-- 
cgit v1.2.3


From ca4d6cfcee0cb2d25c0eb3b0172ecc6f223133ef Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@rpsys.net>
Date: Wed, 2 Jan 2008 01:09:54 +0100
Subject: [ARM] 4737/1: Refactor corgi_lcd to improve readability + bugfix

This patch refactors the code in corgi_lcd.c moving it to the board
specific corgi and spitz files where appropriate instead of the
existing ifdef mess which hinders readability.

Fix spitz_get_hsync_len() to call get_hsync_invperiod so pxafb can be
compiled as a module.

The confusing variables which represent the inverse horizintal sync
period are renamed to "invperiod" consistently.

An incorrect comment in corgi_ts.c is also corrected.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-pxa/corgi.c            | 183 ++++++++++++++++++++-
 arch/arm/mach-pxa/corgi_lcd.c        | 299 +----------------------------------
 arch/arm/mach-pxa/sharpsl.h          |  19 +--
 arch/arm/mach-pxa/spitz.c            | 106 ++++++++++++-
 drivers/input/touchscreen/corgi_ts.c |   8 +-
 include/asm-arm/arch-pxa/corgi.h     |   1 -
 include/asm-arm/arch-pxa/sharpsl.h   |   2 +-
 include/asm-arm/arch-pxa/spitz.h     |   2 -
 8 files changed, 292 insertions(+), 328 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c
index 2363cc64fe07..7304fdbfbb0f 100644
--- a/arch/arm/mach-pxa/corgi.c
+++ b/arch/arm/mach-pxa/corgi.c
@@ -21,6 +21,7 @@
 #include <linux/mmc/host.h>
 #include <linux/pm.h>
 #include <linux/backlight.h>
+#include <video/w100fb.h>
 
 #include <asm/setup.h>
 #include <asm/memory.h>
@@ -140,6 +141,136 @@ struct corgissp_machinfo corgi_ssp_machinfo = {
 };
 
 
+/*
+ * LCD/Framebuffer
+ */
+static void w100_lcdtg_suspend(struct w100fb_par *par)
+{
+	corgi_lcdtg_suspend();
+}
+
+static void w100_lcdtg_init(struct w100fb_par *par)
+{
+	corgi_lcdtg_hw_init(par->xres);
+}
+
+
+static struct w100_tg_info corgi_lcdtg_info = {
+	.change  = w100_lcdtg_init,
+	.suspend = w100_lcdtg_suspend,
+	.resume  = w100_lcdtg_init,
+};
+
+static struct w100_mem_info corgi_fb_mem = {
+	.ext_cntl          = 0x00040003,
+	.sdram_mode_reg    = 0x00650021,
+	.ext_timing_cntl   = 0x10002a4a,
+	.io_cntl           = 0x7ff87012,
+	.size              = 0x1fffff,
+};
+
+static struct w100_gen_regs corgi_fb_regs = {
+	.lcd_format    = 0x00000003,
+	.lcdd_cntl1    = 0x01CC0000,
+	.lcdd_cntl2    = 0x0003FFFF,
+	.genlcd_cntl1  = 0x00FFFF0D,
+	.genlcd_cntl2  = 0x003F3003,
+	.genlcd_cntl3  = 0x000102aa,
+};
+
+static struct w100_gpio_regs corgi_fb_gpio = {
+	.init_data1   = 0x000000bf,
+	.init_data2   = 0x00000000,
+	.gpio_dir1    = 0x00000000,
+	.gpio_oe1     = 0x03c0feff,
+	.gpio_dir2    = 0x00000000,
+	.gpio_oe2     = 0x00000000,
+};
+
+static struct w100_mode corgi_fb_modes[] = {
+{
+	.xres            = 480,
+	.yres            = 640,
+	.left_margin     = 0x56,
+	.right_margin    = 0x55,
+	.upper_margin    = 0x03,
+	.lower_margin    = 0x00,
+	.crtc_ss         = 0x82360056,
+	.crtc_ls         = 0xA0280000,
+	.crtc_gs         = 0x80280028,
+	.crtc_vpos_gs    = 0x02830002,
+	.crtc_rev        = 0x00400008,
+	.crtc_dclk       = 0xA0000000,
+	.crtc_gclk       = 0x8015010F,
+	.crtc_goe        = 0x80100110,
+	.crtc_ps1_active = 0x41060010,
+	.pll_freq        = 75,
+	.fast_pll_freq   = 100,
+	.sysclk_src      = CLK_SRC_PLL,
+	.sysclk_divider  = 0,
+	.pixclk_src      = CLK_SRC_PLL,
+	.pixclk_divider  = 2,
+	.pixclk_divider_rotated = 6,
+},{
+	.xres            = 240,
+	.yres            = 320,
+	.left_margin     = 0x27,
+	.right_margin    = 0x2e,
+	.upper_margin    = 0x01,
+	.lower_margin    = 0x00,
+	.crtc_ss         = 0x81170027,
+	.crtc_ls         = 0xA0140000,
+	.crtc_gs         = 0xC0140014,
+	.crtc_vpos_gs    = 0x00010141,
+	.crtc_rev        = 0x00400008,
+	.crtc_dclk       = 0xA0000000,
+	.crtc_gclk       = 0x8015010F,
+	.crtc_goe        = 0x80100110,
+	.crtc_ps1_active = 0x41060010,
+	.pll_freq        = 0,
+	.fast_pll_freq   = 0,
+	.sysclk_src      = CLK_SRC_XTAL,
+	.sysclk_divider  = 0,
+	.pixclk_src      = CLK_SRC_XTAL,
+	.pixclk_divider  = 1,
+	.pixclk_divider_rotated = 1,
+},
+
+};
+
+static struct w100fb_mach_info corgi_fb_info = {
+	.tg         = &corgi_lcdtg_info,
+	.init_mode  = INIT_MODE_ROTATED,
+	.mem        = &corgi_fb_mem,
+	.regs       = &corgi_fb_regs,
+	.modelist   = &corgi_fb_modes[0],
+	.num_modes  = 2,
+	.gpio       = &corgi_fb_gpio,
+	.xtal_freq  = 12500000,
+	.xtal_dbl   = 0,
+};
+
+static struct resource corgi_fb_resources[] = {
+	[0] = {
+		.start   = 0x08000000,
+		.end     = 0x08ffffff,
+		.flags   = IORESOURCE_MEM,
+	},
+};
+
+static struct platform_device corgifb_device = {
+	.name           = "w100fb",
+	.id             = -1,
+	.num_resources	= ARRAY_SIZE(corgi_fb_resources),
+	.resource	= corgi_fb_resources,
+	.dev            = {
+		.platform_data = &corgi_fb_info,
+		.parent = &corgissp_device.dev,
+	},
+
+};
+
+
 /*
  * Corgi Backlight Device
  */
@@ -154,6 +285,21 @@ static void corgi_bl_kick_battery(void)
 	}
 }
 
+static void corgi_bl_set_intensity(int intensity)
+{
+	if (intensity > 0x10)
+		intensity += 0x10;
+
+	/* Bits 0-4 are accessed via the SSP interface */
+	corgi_ssp_blduty_set(intensity & 0x1f);
+
+	/* Bit 5 is via SCOOP */
+	if (intensity & 0x0020)
+		set_scoop_gpio(&corgiscoop_device.dev, CORGI_SCP_BACKLIGHT_CONT);
+	else
+		reset_scoop_gpio(&corgiscoop_device.dev, CORGI_SCP_BACKLIGHT_CONT);
+}
+
 static struct generic_bl_info corgi_bl_machinfo = {
 	.name = "corgi-bl",
 	.max_intensity = 0x2f,
@@ -190,9 +336,40 @@ static struct platform_device corgiled_device = {
 	.id		= -1,
 };
 
+
 /*
  * Corgi Touch Screen Device
  */
+static unsigned long (*get_hsync_invperiod)(struct device *dev);
+
+static void inline sharpsl_wait_sync(int gpio)
+{
+	while((GPLR(gpio) & GPIO_bit(gpio)) == 0);
+	while((GPLR(gpio) & GPIO_bit(gpio)) != 0);
+}
+
+static unsigned long corgi_get_hsync_invperiod(void)
+{
+	if (!get_hsync_invperiod)
+		get_hsync_invperiod = symbol_get(w100fb_get_hsynclen);
+	if (!get_hsync_invperiod)
+		return 0;
+
+	return get_hsync_invperiod(&corgifb_device.dev);
+}
+
+static void corgi_put_hsync(void)
+{
+	if (get_hsync_invperiod)
+		symbol_put(w100fb_get_hsynclen);
+	get_hsync_invperiod = NULL;
+}
+
+static void corgi_wait_hsync(void)
+{
+	sharpsl_wait_sync(CORGI_GPIO_HSYNC);
+}
+
 static struct resource corgits_resources[] = {
 	[0] = {
 		.start		= CORGI_IRQ_GPIO_TP_INT,
@@ -202,9 +379,9 @@ static struct resource corgits_resources[] = {
 };
 
 static struct corgits_machinfo  corgi_ts_machinfo = {
-	.get_hsync_len   = corgi_get_hsync_len,
-	.put_hsync       = corgi_put_hsync,
-	.wait_hsync      = corgi_wait_hsync,
+	.get_hsync_invperiod = corgi_get_hsync_invperiod,
+	.put_hsync           = corgi_put_hsync,
+	.wait_hsync          = corgi_wait_hsync,
 };
 
 static struct platform_device corgits_device = {
diff --git a/arch/arm/mach-pxa/corgi_lcd.c b/arch/arm/mach-pxa/corgi_lcd.c
index 365b9435f748..9328df37afd1 100644
--- a/arch/arm/mach-pxa/corgi_lcd.c
+++ b/arch/arm/mach-pxa/corgi_lcd.c
@@ -173,7 +173,7 @@ static void lcdtg_set_phadadj(int mode)
 
 static int lcd_inited;
 
-static void lcdtg_hw_init(int mode)
+void corgi_lcdtg_hw_init(int mode)
 {
 	if (!lcd_inited) {
 		int comadj;
@@ -254,7 +254,7 @@ static void lcdtg_hw_init(int mode)
 	}
 }
 
-static void lcdtg_suspend(void)
+void corgi_lcdtg_suspend(void)
 {
 	/* 60Hz x 2 frame = 16.7msec x 2 = 33.4 msec */
 	mdelay(34);
@@ -288,298 +288,3 @@ static void lcdtg_suspend(void)
 	lcd_inited = 0;
 }
 
-
-/*
- * Corgi w100 Frame Buffer Device
- */
-#ifdef CONFIG_PXA_SHARP_C7xx
-
-#include <video/w100fb.h>
-
-static void w100_lcdtg_suspend(struct w100fb_par *par)
-{
-	lcdtg_suspend();
-}
-
-static void w100_lcdtg_init(struct w100fb_par *par)
-{
-	lcdtg_hw_init(par->xres);
-}
-
-
-static struct w100_tg_info corgi_lcdtg_info = {
-	.change  = w100_lcdtg_init,
-	.suspend = w100_lcdtg_suspend,
-	.resume  = w100_lcdtg_init,
-};
-
-static struct w100_mem_info corgi_fb_mem = {
-	.ext_cntl          = 0x00040003,
-	.sdram_mode_reg    = 0x00650021,
-	.ext_timing_cntl   = 0x10002a4a,
-	.io_cntl           = 0x7ff87012,
-	.size              = 0x1fffff,
-};
-
-static struct w100_gen_regs corgi_fb_regs = {
-	.lcd_format    = 0x00000003,
-	.lcdd_cntl1    = 0x01CC0000,
-	.lcdd_cntl2    = 0x0003FFFF,
-	.genlcd_cntl1  = 0x00FFFF0D,
-	.genlcd_cntl2  = 0x003F3003,
-	.genlcd_cntl3  = 0x000102aa,
-};
-
-static struct w100_gpio_regs corgi_fb_gpio = {
-	.init_data1   = 0x000000bf,
-	.init_data2   = 0x00000000,
-	.gpio_dir1    = 0x00000000,
-	.gpio_oe1     = 0x03c0feff,
-	.gpio_dir2    = 0x00000000,
-	.gpio_oe2     = 0x00000000,
-};
-
-static struct w100_mode corgi_fb_modes[] = {
-{
-	.xres            = 480,
-	.yres            = 640,
-	.left_margin     = 0x56,
-	.right_margin    = 0x55,
-	.upper_margin    = 0x03,
-	.lower_margin    = 0x00,
-	.crtc_ss         = 0x82360056,
-	.crtc_ls         = 0xA0280000,
-	.crtc_gs         = 0x80280028,
-	.crtc_vpos_gs    = 0x02830002,
-	.crtc_rev        = 0x00400008,
-	.crtc_dclk       = 0xA0000000,
-	.crtc_gclk       = 0x8015010F,
-	.crtc_goe        = 0x80100110,
-	.crtc_ps1_active = 0x41060010,
-	.pll_freq        = 75,
-	.fast_pll_freq   = 100,
-	.sysclk_src      = CLK_SRC_PLL,
-	.sysclk_divider  = 0,
-	.pixclk_src      = CLK_SRC_PLL,
-	.pixclk_divider  = 2,
-	.pixclk_divider_rotated = 6,
-},{
-	.xres            = 240,
-	.yres            = 320,
-	.left_margin     = 0x27,
-	.right_margin    = 0x2e,
-	.upper_margin    = 0x01,
-	.lower_margin    = 0x00,
-	.crtc_ss         = 0x81170027,
-	.crtc_ls         = 0xA0140000,
-	.crtc_gs         = 0xC0140014,
-	.crtc_vpos_gs    = 0x00010141,
-	.crtc_rev        = 0x00400008,
-	.crtc_dclk       = 0xA0000000,
-	.crtc_gclk       = 0x8015010F,
-	.crtc_goe        = 0x80100110,
-	.crtc_ps1_active = 0x41060010,
-	.pll_freq        = 0,
-	.fast_pll_freq   = 0,
-	.sysclk_src      = CLK_SRC_XTAL,
-	.sysclk_divider  = 0,
-	.pixclk_src      = CLK_SRC_XTAL,
-	.pixclk_divider  = 1,
-	.pixclk_divider_rotated = 1,
-},
-
-};
-
-static struct w100fb_mach_info corgi_fb_info = {
-	.tg         = &corgi_lcdtg_info,
-	.init_mode  = INIT_MODE_ROTATED,
-	.mem        = &corgi_fb_mem,
-	.regs       = &corgi_fb_regs,
-	.modelist   = &corgi_fb_modes[0],
-	.num_modes  = 2,
-	.gpio       = &corgi_fb_gpio,
-	.xtal_freq  = 12500000,
-	.xtal_dbl   = 0,
-};
-
-static struct resource corgi_fb_resources[] = {
-	[0] = {
-		.start   = 0x08000000,
-		.end     = 0x08ffffff,
-		.flags   = IORESOURCE_MEM,
-	},
-};
-
-struct platform_device corgifb_device = {
-	.name           = "w100fb",
-	.id             = -1,
-	.num_resources	= ARRAY_SIZE(corgi_fb_resources),
-	.resource	= corgi_fb_resources,
-	.dev            = {
- 		.platform_data = &corgi_fb_info,
- 		.parent = &corgissp_device.dev,
-	},
-
-};
-#endif
-
-
-/*
- * Spitz PXA Frame Buffer Device
- */
-#ifdef CONFIG_PXA_SHARP_Cxx00
-
-#include <asm/arch/pxafb.h>
-
-void spitz_lcd_power(int on, struct fb_var_screeninfo *var)
-{
-	if (on)
-		lcdtg_hw_init(var->xres);
-	else
-		lcdtg_suspend();
-}
-
-#endif
-
-
-/*
- * Corgi/Spitz Touchscreen to LCD interface
- */
-static unsigned long (*get_hsync_time)(struct device *dev);
-
-static void inline sharpsl_wait_sync(int gpio)
-{
-	while((GPLR(gpio) & GPIO_bit(gpio)) == 0);
-	while((GPLR(gpio) & GPIO_bit(gpio)) != 0);
-}
-
-#ifdef CONFIG_PXA_SHARP_C7xx
-unsigned long corgi_get_hsync_len(void)
-{
-	if (!get_hsync_time)
-		get_hsync_time = symbol_get(w100fb_get_hsynclen);
-	if (!get_hsync_time)
-		return 0;
-
-	return get_hsync_time(&corgifb_device.dev);
-}
-
-void corgi_put_hsync(void)
-{
-	if (get_hsync_time)
-		symbol_put(w100fb_get_hsynclen);
-	get_hsync_time = NULL;
-}
-
-void corgi_wait_hsync(void)
-{
-	sharpsl_wait_sync(CORGI_GPIO_HSYNC);
-}
-#endif
-
-#ifdef CONFIG_PXA_SHARP_Cxx00
-static struct device *spitz_pxafb_dev;
-
-static int is_pxafb_device(struct device * dev, void * data)
-{
-	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
-
-	return (strncmp(pdev->name, "pxa2xx-fb", 9) == 0);
-}
-
-unsigned long spitz_get_hsync_len(void)
-{
-#ifdef CONFIG_FB_PXA
-	if (!spitz_pxafb_dev) {
-		spitz_pxafb_dev = bus_find_device(&platform_bus_type, NULL, NULL, is_pxafb_device);
-		if (!spitz_pxafb_dev)
-			return 0;
-	}
-	if (!get_hsync_time)
-		get_hsync_time = symbol_get(pxafb_get_hsync_time);
-	if (!get_hsync_time)
-#endif
-		return 0;
-
-	return pxafb_get_hsync_time(spitz_pxafb_dev);
-}
-
-void spitz_put_hsync(void)
-{
-	put_device(spitz_pxafb_dev);
-	if (get_hsync_time)
-		symbol_put(pxafb_get_hsync_time);
-	spitz_pxafb_dev = NULL;
-	get_hsync_time = NULL;
-}
-
-void spitz_wait_hsync(void)
-{
-	sharpsl_wait_sync(SPITZ_GPIO_HSYNC);
-}
-#endif
-
-/*
- * Corgi/Spitz Backlight Power
- */
-#ifdef CONFIG_PXA_SHARP_C7xx
-void corgi_bl_set_intensity(int intensity)
-{
-	if (intensity > 0x10)
-		intensity += 0x10;
-
-	/* Bits 0-4 are accessed via the SSP interface */
-	corgi_ssp_blduty_set(intensity & 0x1f);
-
-	/* Bit 5 is via SCOOP */
-	if (intensity & 0x0020)
-		set_scoop_gpio(&corgiscoop_device.dev, CORGI_SCP_BACKLIGHT_CONT);
-	else
-		reset_scoop_gpio(&corgiscoop_device.dev, CORGI_SCP_BACKLIGHT_CONT);
-}
-#endif
-
-
-#if defined(CONFIG_MACH_SPITZ) || defined(CONFIG_MACH_BORZOI)
-void spitz_bl_set_intensity(int intensity)
-{
-	if (intensity > 0x10)
-		intensity += 0x10;
-
-	/* Bits 0-4 are accessed via the SSP interface */
-	corgi_ssp_blduty_set(intensity & 0x1f);
-
-	/* Bit 5 is via SCOOP */
-	if (intensity & 0x0020)
-		reset_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_CONT);
-	else
-		set_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_CONT);
-
-	if (intensity)
-		set_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_ON);
-	else
-		reset_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_ON);
-}
-#endif
-
-#ifdef CONFIG_MACH_AKITA
-void akita_bl_set_intensity(int intensity)
-{
-	if (intensity > 0x10)
-		intensity += 0x10;
-
-	/* Bits 0-4 are accessed via the SSP interface */
-	corgi_ssp_blduty_set(intensity & 0x1f);
-
-	/* Bit 5 is via IO-Expander */
-	if (intensity & 0x0020)
-		akita_reset_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_CONT);
-	else
-		akita_set_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_CONT);
-
-	if (intensity)
-		akita_set_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_ON);
-	else
-		akita_reset_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_ON);
-}
-#endif
diff --git a/arch/arm/mach-pxa/sharpsl.h b/arch/arm/mach-pxa/sharpsl.h
index da4769caaf72..047909a76651 100644
--- a/arch/arm/mach-pxa/sharpsl.h
+++ b/arch/arm/mach-pxa/sharpsl.h
@@ -26,28 +26,15 @@ void corgi_ssp_set_machinfo(struct corgissp_machinfo *machinfo);
 
 
 /*
- * SharpSL Backlight
+ * SharpSL/Corgi LCD Driver
  */
-void corgi_bl_set_intensity(int intensity);
-void spitz_bl_set_intensity(int intensity);
-void akita_bl_set_intensity(int intensity);
-
-
-/*
- * SharpSL Touchscreen Driver
- */
-unsigned long corgi_get_hsync_len(void);
-unsigned long spitz_get_hsync_len(void);
-void corgi_put_hsync(void);
-void spitz_put_hsync(void);
-void corgi_wait_hsync(void);
-void spitz_wait_hsync(void);
+void corgi_lcdtg_suspend(void);
+void corgi_lcdtg_hw_init(int mode);
 
 
 /*
  * SharpSL Battery/PM Driver
  */
-
 #define READ_GPIO_BIT(x)    (GPLR(x) & GPIO_bit(x))
 
 /* MAX1111 Channel Definitions */
diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c
index 2d78199d24af..bb4b65d352c9 100644
--- a/arch/arm/mach-pxa/spitz.c
+++ b/arch/arm/mach-pxa/spitz.c
@@ -271,6 +271,55 @@ static struct platform_device spitzled_device = {
 /*
  * Spitz Touch Screen Device
  */
+
+static unsigned long (*get_hsync_invperiod)(struct device *dev);
+
+static void inline sharpsl_wait_sync(int gpio)
+{
+	while((GPLR(gpio) & GPIO_bit(gpio)) == 0);
+	while((GPLR(gpio) & GPIO_bit(gpio)) != 0);
+}
+
+static struct device *spitz_pxafb_dev;
+
+static int is_pxafb_device(struct device * dev, void * data)
+{
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+
+	return (strncmp(pdev->name, "pxa2xx-fb", 9) == 0);
+}
+
+static unsigned long spitz_get_hsync_invperiod(void)
+{
+#ifdef CONFIG_FB_PXA
+	if (!spitz_pxafb_dev) {
+		spitz_pxafb_dev = bus_find_device(&platform_bus_type, NULL, NULL, is_pxafb_device);
+		if (!spitz_pxafb_dev)
+			return 0;
+	}
+	if (!get_hsync_invperiod)
+		get_hsync_invperiod = symbol_get(pxafb_get_hsync_time);
+	if (!get_hsync_invperiod)
+#endif
+		return 0;
+
+	return get_hsync_invperiod(spitz_pxafb_dev);
+}
+
+static void spitz_put_hsync(void)
+{
+	put_device(spitz_pxafb_dev);
+	if (get_hsync_invperiod)
+		symbol_put(pxafb_get_hsync_time);
+	spitz_pxafb_dev = NULL;
+	get_hsync_invperiod = NULL;
+}
+
+static void spitz_wait_hsync(void)
+{
+	sharpsl_wait_sync(SPITZ_GPIO_HSYNC);
+}
+
 static struct resource spitzts_resources[] = {
 	[0] = {
 		.start		= SPITZ_IRQ_GPIO_TP_INT,
@@ -280,9 +329,9 @@ static struct resource spitzts_resources[] = {
 };
 
 static struct corgits_machinfo  spitz_ts_machinfo = {
-	.get_hsync_len   = spitz_get_hsync_len,
-	.put_hsync       = spitz_put_hsync,
-	.wait_hsync      = spitz_wait_hsync,
+	.get_hsync_invperiod = spitz_get_hsync_invperiod,
+	.put_hsync           = spitz_put_hsync,
+	.wait_hsync          = spitz_wait_hsync,
 };
 
 static struct platform_device spitzts_device = {
@@ -423,6 +472,14 @@ static struct pxaficp_platform_data spitz_ficp_platform_data = {
  * Spitz PXA Framebuffer
  */
 
+static void spitz_lcd_power(int on, struct fb_var_screeninfo *var)
+{
+	if (on)
+		corgi_lcdtg_hw_init(var->xres);
+	else
+		corgi_lcdtg_suspend();
+}
+
 static struct pxafb_mode_info spitz_pxafb_modes[] = {
 {
 	.pixclock       = 19231,
@@ -520,6 +577,27 @@ static void __init common_init(void)
 	set_pxa_fb_info(&spitz_pxafb_info);
 }
 
+#if defined(CONFIG_MACH_SPITZ) || defined(CONFIG_MACH_BORZOI)
+static void spitz_bl_set_intensity(int intensity)
+{
+	if (intensity > 0x10)
+		intensity += 0x10;
+
+	/* Bits 0-4 are accessed via the SSP interface */
+	corgi_ssp_blduty_set(intensity & 0x1f);
+
+	/* Bit 5 is via SCOOP */
+	if (intensity & 0x0020)
+		reset_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_CONT);
+	else
+		set_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_CONT);
+
+	if (intensity)
+		set_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_ON);
+	else
+		reset_scoop_gpio(&spitzscoop2_device.dev, SPITZ_SCP2_BACKLIGHT_ON);
+}
+
 static void __init spitz_init(void)
 {
 	platform_scoop_config = &spitz_pcmcia_config;
@@ -530,6 +608,7 @@ static void __init spitz_init(void)
 
 	platform_device_register(&spitzscoop2_device);
 }
+#endif
 
 #ifdef CONFIG_MACH_AKITA
 /*
@@ -542,6 +621,26 @@ struct platform_device akitaioexp_device = {
 
 EXPORT_SYMBOL_GPL(akitaioexp_device);
 
+static void akita_bl_set_intensity(int intensity)
+{
+	if (intensity > 0x10)
+		intensity += 0x10;
+
+	/* Bits 0-4 are accessed via the SSP interface */
+	corgi_ssp_blduty_set(intensity & 0x1f);
+
+	/* Bit 5 is via IO-Expander */
+	if (intensity & 0x0020)
+		akita_reset_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_CONT);
+	else
+		akita_set_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_CONT);
+
+	if (intensity)
+		akita_set_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_ON);
+	else
+		akita_reset_ioexp(&akitaioexp_device.dev, AKITA_IOEXP_BACKLIGHT_ON);
+}
+
 static void __init akita_init(void)
 {
 	spitz_ficp_platform_data.transceiver_mode = akita_irda_transceiver_mode;
@@ -558,7 +657,6 @@ static void __init akita_init(void)
 }
 #endif
 
-
 static void __init fixup_spitz(struct machine_desc *desc,
 		struct tag *tags, char **cmdline, struct meminfo *mi)
 {
diff --git a/drivers/input/touchscreen/corgi_ts.c b/drivers/input/touchscreen/corgi_ts.c
index b1b2e07bf080..99d92f5c93d6 100644
--- a/drivers/input/touchscreen/corgi_ts.c
+++ b/drivers/input/touchscreen/corgi_ts.c
@@ -74,10 +74,10 @@ extern unsigned int get_clk_frequency_khz(int info);
 
 static unsigned long calc_waittime(struct corgi_ts *corgi_ts)
 {
-	unsigned long hsync_len = corgi_ts->machinfo->get_hsync_len();
+	unsigned long hsync_invperiod = corgi_ts->machinfo->get_hsync_invperiod();
 
-	if (hsync_len)
-		return get_clk_frequency_khz(0)*1000/hsync_len;
+	if (hsync_invperiod)
+		return get_clk_frequency_khz(0)*1000/hsync_invperiod;
 	else
 		return 0;
 }
@@ -114,7 +114,7 @@ static int sync_receive_data_send_cmd(struct corgi_ts *corgi_ts, int doRecive, i
 			if (timer2-timer1 > wait_time) {
 				/* too slow - timeout, try again */
 				corgi_ts->machinfo->wait_hsync();
-				/* get OSCR */
+				/* get CCNT */
 				CCNT(timer1);
 				/* Wait after HSync */
 				CCNT(timer2);
diff --git a/include/asm-arm/arch-pxa/corgi.h b/include/asm-arm/arch-pxa/corgi.h
index e554caa0d18b..bf856503baf6 100644
--- a/include/asm-arm/arch-pxa/corgi.h
+++ b/include/asm-arm/arch-pxa/corgi.h
@@ -104,7 +104,6 @@
  */
 extern struct platform_device corgiscoop_device;
 extern struct platform_device corgissp_device;
-extern struct platform_device corgifb_device;
 
 #endif /* __ASM_ARCH_CORGI_H  */
 
diff --git a/include/asm-arm/arch-pxa/sharpsl.h b/include/asm-arm/arch-pxa/sharpsl.h
index 2b0fe773213a..3b1d4a72d4d1 100644
--- a/include/asm-arm/arch-pxa/sharpsl.h
+++ b/include/asm-arm/arch-pxa/sharpsl.h
@@ -16,7 +16,7 @@ int corgi_ssp_max1111_get(unsigned long data);
  */
 
 struct corgits_machinfo {
-	unsigned long (*get_hsync_len)(void);
+	unsigned long (*get_hsync_invperiod)(void);
 	void (*put_hsync)(void);
 	void (*wait_hsync)(void);
 };
diff --git a/include/asm-arm/arch-pxa/spitz.h b/include/asm-arm/arch-pxa/spitz.h
index 4953dd324d4d..bd14365f7ed5 100644
--- a/include/asm-arm/arch-pxa/spitz.h
+++ b/include/asm-arm/arch-pxa/spitz.h
@@ -156,5 +156,3 @@ extern struct platform_device spitzscoop_device;
 extern struct platform_device spitzscoop2_device;
 extern struct platform_device spitzssp_device;
 extern struct sharpsl_charger_machinfo spitz_pm_machinfo;
-
-extern void spitz_lcd_power(int on, struct fb_var_screeninfo *var);
-- 
cgit v1.2.3


From 4e4fc05a2b6e7bd2e0facd96e0c18dceb34d9349 Mon Sep 17 00:00:00 2001
From: Daniel Mack <daniel@caiaq.org>
Date: Wed, 23 Jan 2008 14:54:50 +0100
Subject: [ARM] 4762/1: Basic support for Toradex Colibri module

This patch adds support for Toradex' PXA27x based Colibri module.
It's kept as simple as possible to only provide basic functionality.
A default config is also included.

Signed-off-by: Daniel Mack <daniel@caiaq.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/colibri_defconfig | 1481 ++++++++++++++++++++++++++++++++++++
 arch/arm/mach-pxa/Kconfig          |    4 +
 arch/arm/mach-pxa/Makefile         |    1 +
 arch/arm/mach-pxa/colibri.c        |  134 ++++
 drivers/net/dm9000.c               |    3 +-
 include/asm-arm/arch-pxa/colibri.h |   19 +
 6 files changed, 1641 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/configs/colibri_defconfig
 create mode 100644 arch/arm/mach-pxa/colibri.c
 create mode 100644 include/asm-arm/arch-pxa/colibri.h

(limited to 'include')

diff --git a/arch/arm/configs/colibri_defconfig b/arch/arm/configs/colibri_defconfig
new file mode 100644
index 000000000000..c3e3418ed4fe
--- /dev/null
+++ b/arch/arm/configs/colibri_defconfig
@@ -0,0 +1,1481 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.24-rc3
+# Mon Dec  3 13:36:09 2007
+#
+CONFIG_ARM=y
+CONFIG_SYS_SUPPORTS_APM_EMULATION=y
+CONFIG_GENERIC_GPIO=y
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+CONFIG_MMU=y
+# CONFIG_NO_IOPORT is not set
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_STACKTRACE_SUPPORT=y
+CONFIG_LOCKDEP_SUPPORT=y
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
+CONFIG_HARDIRQS_SW_RESEND=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_ZONE_DMA=y
+CONFIG_ARCH_MTD_XIP=y
+CONFIG_VECTORS_BASE=0xffff0000
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+CONFIG_SYSVIPC_SYSCTL=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_BSD_PROCESS_ACCT_V3=y
+# CONFIG_TASKSTATS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
+# CONFIG_AUDIT is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
+CONFIG_SYSFS_DEPRECATED=y
+# CONFIG_RELAY is not set
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+CONFIG_UID16=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+CONFIG_KALLSYMS_EXTRA_PASS=y
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_ANON_INODES=y
+CONFIG_EPOLL=y
+CONFIG_SIGNALFD=y
+CONFIG_EVENTFD=y
+CONFIG_SHMEM=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+CONFIG_RT_MUTEXES=y
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_MODVERSIONS=y
+CONFIG_MODULE_SRCVERSION_ALL=y
+CONFIG_KMOD=y
+CONFIG_BLOCK=y
+CONFIG_LBD=y
+# CONFIG_BLK_DEV_IO_TRACE is not set
+CONFIG_LSF=y
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+CONFIG_DEFAULT_AS=y
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="anticipatory"
+
+#
+# System Type
+#
+# CONFIG_ARCH_AAEC2000 is not set
+# CONFIG_ARCH_INTEGRATOR is not set
+# CONFIG_ARCH_REALVIEW is not set
+# CONFIG_ARCH_VERSATILE is not set
+# CONFIG_ARCH_AT91 is not set
+# CONFIG_ARCH_CLPS7500 is not set
+# CONFIG_ARCH_CLPS711X is not set
+# CONFIG_ARCH_CO285 is not set
+# CONFIG_ARCH_EBSA110 is not set
+# CONFIG_ARCH_EP93XX is not set
+# CONFIG_ARCH_FOOTBRIDGE is not set
+# CONFIG_ARCH_NETX is not set
+# CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_IMX is not set
+# CONFIG_ARCH_IOP13XX is not set
+# CONFIG_ARCH_IOP32X is not set
+# CONFIG_ARCH_IOP33X is not set
+# CONFIG_ARCH_IXP23XX is not set
+# CONFIG_ARCH_IXP2000 is not set
+# CONFIG_ARCH_IXP4XX is not set
+# CONFIG_ARCH_L7200 is not set
+# CONFIG_ARCH_KS8695 is not set
+# CONFIG_ARCH_NS9XXX is not set
+# CONFIG_ARCH_MXC is not set
+# CONFIG_ARCH_PNX4008 is not set
+CONFIG_ARCH_PXA=y
+# CONFIG_ARCH_RPC is not set
+# CONFIG_ARCH_SA1100 is not set
+# CONFIG_ARCH_S3C2410 is not set
+# CONFIG_ARCH_SHARK is not set
+# CONFIG_ARCH_LH7A40X is not set
+# CONFIG_ARCH_DAVINCI is not set
+# CONFIG_ARCH_OMAP is not set
+
+#
+# Intel PXA2xx/PXA3xx Implementations
+#
+# CONFIG_ARCH_LUBBOCK is not set
+# CONFIG_MACH_LOGICPD_PXA270 is not set
+# CONFIG_MACH_MAINSTONE is not set
+# CONFIG_ARCH_PXA_IDP is not set
+# CONFIG_PXA_SHARPSL is not set
+# CONFIG_MACH_TRIZEPS4 is not set
+# CONFIG_MACH_EM_X270 is not set
+CONFIG_MACH_COLIBRI=y
+# CONFIG_MACH_ZYLONITE is not set
+# CONFIG_MACH_ARMCORE is not set
+CONFIG_PXA27x=y
+
+#
+# Boot options
+#
+
+#
+# Power management
+#
+
+#
+# Processor Type
+#
+CONFIG_CPU_32=y
+CONFIG_CPU_XSCALE=y
+CONFIG_CPU_32v5=y
+CONFIG_CPU_ABRT_EV5T=y
+CONFIG_CPU_CACHE_VIVT=y
+CONFIG_CPU_TLB_V4WBI=y
+CONFIG_CPU_CP15=y
+CONFIG_CPU_CP15_MMU=y
+
+#
+# Processor Features
+#
+CONFIG_ARM_THUMB=y
+# CONFIG_CPU_DCACHE_DISABLE is not set
+# CONFIG_OUTER_CACHE is not set
+CONFIG_IWMMXT=y
+CONFIG_XSCALE_PMU=y
+
+#
+# Bus support
+#
+# CONFIG_PCI_SYSCALL is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+# CONFIG_PCCARD is not set
+
+#
+# Kernel Features
+#
+# CONFIG_TICK_ONESHOT is not set
+# CONFIG_NO_HZ is not set
+# CONFIG_HIGH_RES_TIMERS is not set
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
+CONFIG_PREEMPT=y
+CONFIG_HZ=100
+CONFIG_AEABI=y
+CONFIG_OABI_COMPAT=y
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4096
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_BOUNCE=y
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ALIGNMENT_TRAP=y
+
+#
+# Boot options
+#
+CONFIG_ZBOOT_ROM_TEXT=0
+CONFIG_ZBOOT_ROM_BSS=0
+CONFIG_CMDLINE=""
+# CONFIG_XIP_KERNEL is not set
+# CONFIG_KEXEC is not set
+
+#
+# Floating point emulation
+#
+
+#
+# At least one emulation must be selected
+#
+CONFIG_FPE_NWFPE=y
+# CONFIG_FPE_NWFPE_XP is not set
+# CONFIG_FPE_FASTFPE is not set
+
+#
+# Userspace binary formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+CONFIG_PM=y
+# CONFIG_PM_LEGACY is not set
+# CONFIG_PM_DEBUG is not set
+CONFIG_PM_SLEEP=y
+CONFIG_SUSPEND_UP_POSSIBLE=y
+CONFIG_SUSPEND=y
+# CONFIG_APM_EMULATION is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+CONFIG_XFRM_USER=m
+# CONFIG_XFRM_SUB_POLICY is not set
+# CONFIG_XFRM_MIGRATE is not set
+CONFIG_NET_KEY=y
+# CONFIG_NET_KEY_MIGRATE is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_XFRM_MODE_TRANSPORT=y
+CONFIG_INET_XFRM_MODE_TUNNEL=y
+CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IP_VS is not set
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETLABEL is not set
+# CONFIG_NETWORK_SECMARK is not set
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+
+#
+# Core Netfilter Configuration
+#
+# CONFIG_NETFILTER_NETLINK is not set
+# CONFIG_NF_CONNTRACK_ENABLED is not set
+# CONFIG_NF_CONNTRACK is not set
+# CONFIG_NETFILTER_XTABLES is not set
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_QUEUE=m
+# CONFIG_IP_NF_IPTABLES is not set
+# CONFIG_IP_NF_ARPTABLES is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+CONFIG_VLAN_8021Q=m
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+CONFIG_IRDA=m
+
+#
+# IrDA protocols
+#
+CONFIG_IRLAN=m
+CONFIG_IRCOMM=m
+CONFIG_IRDA_ULTRA=y
+
+#
+# IrDA options
+#
+CONFIG_IRDA_CACHE_LAST_LSAP=y
+CONFIG_IRDA_FAST_RR=y
+# CONFIG_IRDA_DEBUG is not set
+
+#
+# Infrared-port device drivers
+#
+
+#
+# SIR device drivers
+#
+CONFIG_IRTTY_SIR=m
+
+#
+# Dongle support
+#
+# CONFIG_DONGLE is not set
+# CONFIG_KINGSUN_DONGLE is not set
+# CONFIG_KSDAZZLE_DONGLE is not set
+# CONFIG_KS959_DONGLE is not set
+
+#
+# Old SIR device drivers
+#
+# CONFIG_IRPORT_SIR is not set
+
+#
+# Old Serial dongle support
+#
+
+#
+# FIR device drivers
+#
+# CONFIG_USB_IRDA is not set
+# CONFIG_SIGMATEL_FIR is not set
+# CONFIG_PXA_FICP is not set
+# CONFIG_MCS_FIR is not set
+CONFIG_BT=m
+CONFIG_BT_L2CAP=m
+CONFIG_BT_SCO=m
+CONFIG_BT_RFCOMM=m
+CONFIG_BT_RFCOMM_TTY=y
+CONFIG_BT_BNEP=m
+CONFIG_BT_BNEP_MC_FILTER=y
+CONFIG_BT_BNEP_PROTO_FILTER=y
+CONFIG_BT_HIDP=m
+
+#
+# Bluetooth device drivers
+#
+# CONFIG_BT_HCIUSB is not set
+# CONFIG_BT_HCIBTUSB is not set
+# CONFIG_BT_HCIBTSDIO is not set
+# CONFIG_BT_HCIUART is not set
+# CONFIG_BT_HCIBCM203X is not set
+# CONFIG_BT_HCIBPA10X is not set
+# CONFIG_BT_HCIBFUSB is not set
+# CONFIG_BT_HCIVHCI is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+CONFIG_CFG80211=y
+CONFIG_NL80211=y
+CONFIG_WIRELESS_EXT=y
+# CONFIG_MAC80211 is not set
+CONFIG_IEEE80211=y
+# CONFIG_IEEE80211_DEBUG is not set
+CONFIG_IEEE80211_CRYPT_WEP=y
+CONFIG_IEEE80211_CRYPT_CCMP=m
+CONFIG_IEEE80211_CRYPT_TKIP=m
+CONFIG_IEEE80211_SOFTMAC=m
+# CONFIG_IEEE80211_SOFTMAC_DEBUG is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_DEBUG_DEVRES is not set
+# CONFIG_SYS_HYPERVISOR is not set
+CONFIG_CONNECTOR=y
+CONFIG_PROC_EVENTS=y
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+# CONFIG_MTD_AFS_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+CONFIG_MTD_JEDECPROBE=y
+CONFIG_MTD_GEN_PROBE=y
+CONFIG_MTD_CFI_ADV_OPTIONS=y
+# CONFIG_MTD_CFI_NOSWAP is not set
+# CONFIG_MTD_CFI_BE_BYTE_SWAP is not set
+CONFIG_MTD_CFI_LE_BYTE_SWAP=y
+CONFIG_MTD_CFI_GEOMETRY=y
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_OTP is not set
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+# CONFIG_MTD_RAM is not set
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_XIP is not set
+
+#
+# Mapping drivers for chip access
+#
+CONFIG_MTD_COMPLEX_MAPPINGS=y
+CONFIG_MTD_PHYSMAP=y
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
+CONFIG_MTD_PHYSMAP_BANKWIDTH=2
+CONFIG_MTD_PXA2XX=y
+# CONFIG_MTD_ARM_INTEGRATOR is not set
+# CONFIG_MTD_IMPA7 is not set
+# CONFIG_MTD_SHARP_SL is not set
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+CONFIG_MTD_BLOCK2MTD=y
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+CONFIG_MTD_NAND=y
+# CONFIG_MTD_NAND_VERIFY_WRITE is not set
+# CONFIG_MTD_NAND_ECC_SMC is not set
+# CONFIG_MTD_NAND_MUSEUM_IDS is not set
+# CONFIG_MTD_NAND_H1900 is not set
+CONFIG_MTD_NAND_IDS=y
+CONFIG_MTD_NAND_DISKONCHIP=y
+CONFIG_MTD_NAND_DISKONCHIP_PROBE_ADVANCED=y
+CONFIG_MTD_NAND_DISKONCHIP_PROBE_ADDRESS=0x4000000
+CONFIG_MTD_NAND_DISKONCHIP_PROBE_HIGH=y
+CONFIG_MTD_NAND_DISKONCHIP_BBTWRITE=y
+# CONFIG_MTD_NAND_SHARPSL is not set
+# CONFIG_MTD_NAND_NANDSIM is not set
+# CONFIG_MTD_NAND_PLATFORM is not set
+# CONFIG_MTD_ALAUDA is not set
+CONFIG_MTD_ONENAND=y
+# CONFIG_MTD_ONENAND_VERIFY_WRITE is not set
+# CONFIG_MTD_ONENAND_GENERIC is not set
+# CONFIG_MTD_ONENAND_OTP is not set
+# CONFIG_MTD_ONENAND_2X_PROGRAM is not set
+# CONFIG_MTD_ONENAND_SIM is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_CRYPTOLOOP=m
+CONFIG_BLK_DEV_NBD=y
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=8
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
+CONFIG_IDE=y
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=y
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+# CONFIG_BLK_DEV_IDE_SATA is not set
+CONFIG_BLK_DEV_IDEDISK=y
+CONFIG_IDEDISK_MULTI_MODE=y
+# CONFIG_BLK_DEV_IDECD is not set
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+# CONFIG_IDE_TASK_IOCTL is not set
+CONFIG_IDE_PROC_FS=y
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=y
+# CONFIG_BLK_DEV_PLATFORM is not set
+# CONFIG_IDE_ARM is not set
+# CONFIG_BLK_DEV_IDEDMA is not set
+CONFIG_IDE_ARCH_OBSOLETE_INIT=y
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_ATA is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+CONFIG_PHYLIB=y
+
+#
+# MII PHY device drivers
+#
+# CONFIG_MARVELL_PHY is not set
+# CONFIG_DAVICOM_PHY is not set
+# CONFIG_QSEMI_PHY is not set
+# CONFIG_LXT_PHY is not set
+# CONFIG_CICADA_PHY is not set
+# CONFIG_VITESSE_PHY is not set
+# CONFIG_SMSC_PHY is not set
+# CONFIG_BROADCOM_PHY is not set
+# CONFIG_ICPLUS_PHY is not set
+# CONFIG_FIXED_PHY is not set
+# CONFIG_MDIO_BITBANG is not set
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_AX88796 is not set
+# CONFIG_SMC91X is not set
+CONFIG_DM9000=y
+# CONFIG_SMC911X is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+CONFIG_WLAN_80211=y
+# CONFIG_LIBERTAS is not set
+# CONFIG_USB_ZD1201 is not set
+CONFIG_HOSTAP=y
+CONFIG_HOSTAP_FIRMWARE=y
+CONFIG_HOSTAP_FIRMWARE_NVRAM=y
+# CONFIG_ZD1211RW is not set
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+# CONFIG_INPUT_FF_MEMLESS is not set
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+CONFIG_INPUT_MOUSEDEV_PSAUX=y
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=640
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=480
+# CONFIG_INPUT_JOYDEV is not set
+CONFIG_INPUT_EVDEV=y
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+CONFIG_INPUT_KEYBOARD=y
+CONFIG_KEYBOARD_ATKBD=m
+# CONFIG_KEYBOARD_SUNKBD is not set
+# CONFIG_KEYBOARD_LKKBD is not set
+# CONFIG_KEYBOARD_XTKBD is not set
+# CONFIG_KEYBOARD_NEWTON is not set
+# CONFIG_KEYBOARD_STOWAWAY is not set
+# CONFIG_KEYBOARD_PXA27x is not set
+# CONFIG_KEYBOARD_GPIO is not set
+CONFIG_INPUT_MOUSE=y
+# CONFIG_MOUSE_PS2 is not set
+CONFIG_MOUSE_SERIAL=m
+# CONFIG_MOUSE_APPLETOUCH is not set
+# CONFIG_MOUSE_VSXXXAA is not set
+# CONFIG_MOUSE_GPIO is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+CONFIG_INPUT_TOUCHSCREEN=y
+# CONFIG_TOUCHSCREEN_FUJITSU is not set
+# CONFIG_TOUCHSCREEN_GUNZE is not set
+# CONFIG_TOUCHSCREEN_ELO is not set
+# CONFIG_TOUCHSCREEN_MTOUCH is not set
+# CONFIG_TOUCHSCREEN_MK712 is not set
+# CONFIG_TOUCHSCREEN_PENMOUNT is not set
+# CONFIG_TOUCHSCREEN_TOUCHRIGHT is not set
+# CONFIG_TOUCHSCREEN_TOUCHWIN is not set
+CONFIG_TOUCHSCREEN_UCB1400=y
+# CONFIG_TOUCHSCREEN_USB_COMPOSITE is not set
+CONFIG_INPUT_MISC=y
+# CONFIG_INPUT_ATI_REMOTE is not set
+# CONFIG_INPUT_ATI_REMOTE2 is not set
+# CONFIG_INPUT_KEYSPAN_REMOTE is not set
+# CONFIG_INPUT_POWERMATE is not set
+# CONFIG_INPUT_YEALINK is not set
+CONFIG_INPUT_UINPUT=m
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+CONFIG_SERIO_SERPORT=y
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_VT_HW_CONSOLE_BINDING is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_PXA=y
+CONFIG_SERIAL_PXA_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+CONFIG_HW_RANDOM=y
+# CONFIG_NVRAM is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+CONFIG_I2C=y
+CONFIG_I2C_BOARDINFO=y
+CONFIG_I2C_CHARDEV=y
+
+#
+# I2C Algorithms
+#
+# CONFIG_I2C_ALGOBIT is not set
+# CONFIG_I2C_ALGOPCF is not set
+# CONFIG_I2C_ALGOPCA is not set
+
+#
+# I2C Hardware Bus support
+#
+# CONFIG_I2C_GPIO is not set
+# CONFIG_I2C_PXA is not set
+# CONFIG_I2C_OCORES is not set
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_SIMTEC is not set
+# CONFIG_I2C_TAOS_EVM is not set
+# CONFIG_I2C_STUB is not set
+# CONFIG_I2C_TINY_USB is not set
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
+# CONFIG_DS1682 is not set
+# CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_SENSORS_TSL2550 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_SENSORS_AD7418 is not set
+# CONFIG_SENSORS_ADM1021 is not set
+# CONFIG_SENSORS_ADM1025 is not set
+# CONFIG_SENSORS_ADM1026 is not set
+# CONFIG_SENSORS_ADM1029 is not set
+# CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
+# CONFIG_SENSORS_ADT7470 is not set
+# CONFIG_SENSORS_ATXP1 is not set
+# CONFIG_SENSORS_DS1621 is not set
+# CONFIG_SENSORS_F71805F is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_F75375S is not set
+# CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
+# CONFIG_SENSORS_IT87 is not set
+# CONFIG_SENSORS_LM63 is not set
+# CONFIG_SENSORS_LM75 is not set
+# CONFIG_SENSORS_LM77 is not set
+# CONFIG_SENSORS_LM78 is not set
+# CONFIG_SENSORS_LM80 is not set
+# CONFIG_SENSORS_LM83 is not set
+# CONFIG_SENSORS_LM85 is not set
+# CONFIG_SENSORS_LM87 is not set
+# CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_LM93 is not set
+# CONFIG_SENSORS_MAX1619 is not set
+# CONFIG_SENSORS_MAX6650 is not set
+# CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_PC87427 is not set
+# CONFIG_SENSORS_DME1737 is not set
+# CONFIG_SENSORS_SMSC47M1 is not set
+# CONFIG_SENSORS_SMSC47M192 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_THMC50 is not set
+# CONFIG_SENSORS_VT1211 is not set
+# CONFIG_SENSORS_W83781D is not set
+# CONFIG_SENSORS_W83791D is not set
+# CONFIG_SENSORS_W83792D is not set
+# CONFIG_SENSORS_W83793 is not set
+# CONFIG_SENSORS_W83L785TS is not set
+# CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+# CONFIG_SA1100_WATCHDOG is not set
+
+#
+# USB-based Watchdog Cards
+#
+# CONFIG_USBPCWATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+CONFIG_DAB=y
+# CONFIG_USB_DABUSB is not set
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+CONFIG_FB=y
+CONFIG_FIRMWARE_EDID=y
+# CONFIG_FB_DDC is not set
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
+# CONFIG_FB_SYS_FILLRECT is not set
+# CONFIG_FB_SYS_COPYAREA is not set
+# CONFIG_FB_SYS_IMAGEBLIT is not set
+# CONFIG_FB_SYS_FOPS is not set
+CONFIG_FB_DEFERRED_IO=y
+# CONFIG_FB_SVGALIB is not set
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+
+#
+# Frame buffer hardware drivers
+#
+# CONFIG_FB_UVESA is not set
+# CONFIG_FB_S1D13XXX is not set
+CONFIG_FB_PXA=y
+# CONFIG_FB_PXA_PARAMETERS is not set
+# CONFIG_FB_MBX is not set
+# CONFIG_FB_VIRTUAL is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=y
+CONFIG_BACKLIGHT_CLASS_DEVICE=y
+# CONFIG_BACKLIGHT_CORGI is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Console display driver support
+#
+# CONFIG_VGA_CONSOLE is not set
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+# CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY is not set
+CONFIG_FRAMEBUFFER_CONSOLE_ROTATION=y
+CONFIG_FONTS=y
+CONFIG_FONT_8x8=y
+CONFIG_FONT_8x16=y
+# CONFIG_FONT_6x11 is not set
+# CONFIG_FONT_7x14 is not set
+# CONFIG_FONT_PEARL_8x8 is not set
+# CONFIG_FONT_ACORN_8x8 is not set
+# CONFIG_FONT_MINI_4x6 is not set
+# CONFIG_FONT_SUN8x16 is not set
+# CONFIG_FONT_SUN12x22 is not set
+# CONFIG_FONT_10x18 is not set
+CONFIG_LOGO=y
+CONFIG_LOGO_LINUX_MONO=y
+CONFIG_LOGO_LINUX_VGA16=y
+CONFIG_LOGO_LINUX_CLUT224=y
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+CONFIG_AC97_BUS=y
+CONFIG_HID_SUPPORT=y
+CONFIG_HID=y
+# CONFIG_HID_DEBUG is not set
+# CONFIG_HIDRAW is not set
+
+#
+# USB Input Devices
+#
+# CONFIG_USB_HID is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+CONFIG_USB_SUPPORT=y
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_DEVICE_CLASS is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_SUSPEND is not set
+# CONFIG_USB_PERSIST is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_OHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+# CONFIG_USB_R8A66597_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# may also be needed; see USB_STORAGE Help for more information
+#
+# CONFIG_USB_LIBUSUAL is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MON is not set
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+CONFIG_USB_SERIAL=m
+# CONFIG_USB_SERIAL_GENERIC is not set
+# CONFIG_USB_SERIAL_AIRCABLE is not set
+# CONFIG_USB_SERIAL_AIRPRIME is not set
+# CONFIG_USB_SERIAL_ARK3116 is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_CH341 is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+# CONFIG_USB_SERIAL_CP2101 is not set
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+# CONFIG_USB_SERIAL_FTDI_SIO is not set
+# CONFIG_USB_SERIAL_FUNSOFT is not set
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+# CONFIG_USB_SERIAL_MOS7720 is not set
+# CONFIG_USB_SERIAL_MOS7840 is not set
+# CONFIG_USB_SERIAL_NAVMAN is not set
+# CONFIG_USB_SERIAL_PL2303 is not set
+# CONFIG_USB_SERIAL_OTI6858 is not set
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_SIERRAWIRELESS is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OPTION is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+# CONFIG_USB_SERIAL_DEBUG is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_ADUTUX is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_BERRY_CHARGE is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGET is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_FTDI_ELAN is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TRANCEVIBRATOR is not set
+# CONFIG_USB_IOWARRIOR is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+CONFIG_USB_GADGET=m
+# CONFIG_USB_GADGET_DEBUG is not set
+# CONFIG_USB_GADGET_DEBUG_FILES is not set
+# CONFIG_USB_GADGET_DEBUG_FS is not set
+CONFIG_USB_GADGET_SELECTED=y
+# CONFIG_USB_GADGET_AMD5536UDC is not set
+# CONFIG_USB_GADGET_ATMEL_USBA is not set
+# CONFIG_USB_GADGET_FSL_USB2 is not set
+# CONFIG_USB_GADGET_NET2280 is not set
+# CONFIG_USB_GADGET_PXA2XX is not set
+# CONFIG_USB_GADGET_M66592 is not set
+# CONFIG_USB_GADGET_GOKU is not set
+# CONFIG_USB_GADGET_LH7A40X is not set
+# CONFIG_USB_GADGET_OMAP is not set
+# CONFIG_USB_GADGET_S3C2410 is not set
+# CONFIG_USB_GADGET_AT91 is not set
+CONFIG_USB_GADGET_DUMMY_HCD=y
+CONFIG_USB_DUMMY_HCD=m
+CONFIG_USB_GADGET_DUALSPEED=y
+# CONFIG_USB_ZERO is not set
+# CONFIG_USB_ETH is not set
+# CONFIG_USB_GADGETFS is not set
+# CONFIG_USB_FILE_STORAGE is not set
+# CONFIG_USB_G_SERIAL is not set
+# CONFIG_USB_MIDI_GADGET is not set
+CONFIG_MMC=y
+# CONFIG_MMC_DEBUG is not set
+# CONFIG_MMC_UNSAFE_RESUME is not set
+
+#
+# MMC/SD Card Drivers
+#
+CONFIG_MMC_BLOCK=y
+CONFIG_MMC_BLOCK_BOUNCE=y
+# CONFIG_SDIO_UART is not set
+
+#
+# MMC/SD Host Controller Drivers
+#
+# CONFIG_MMC_PXA is not set
+CONFIG_NEW_LEDS=y
+# CONFIG_LEDS_CLASS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+CONFIG_LEDS_TRIGGERS=y
+CONFIG_LEDS_TRIGGER_TIMER=y
+# CONFIG_LEDS_TRIGGER_IDE_DISK is not set
+CONFIG_LEDS_TRIGGER_HEARTBEAT=y
+CONFIG_RTC_LIB=y
+CONFIG_RTC_CLASS=y
+# CONFIG_RTC_HCTOSYS is not set
+# CONFIG_RTC_DEBUG is not set
+
+#
+# RTC interfaces
+#
+CONFIG_RTC_INTF_SYSFS=y
+CONFIG_RTC_INTF_PROC=y
+CONFIG_RTC_INTF_DEV=y
+# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set
+# CONFIG_RTC_DRV_TEST is not set
+
+#
+# I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
+# CONFIG_RTC_DRV_DS1672 is not set
+# CONFIG_RTC_DRV_MAX6900 is not set
+# CONFIG_RTC_DRV_RS5C372 is not set
+# CONFIG_RTC_DRV_ISL1208 is not set
+# CONFIG_RTC_DRV_X1205 is not set
+# CONFIG_RTC_DRV_PCF8563 is not set
+CONFIG_RTC_DRV_PCF8583=m
+# CONFIG_RTC_DRV_M41T80 is not set
+
+#
+# SPI RTC drivers
+#
+
+#
+# Platform RTC drivers
+#
+# CONFIG_RTC_DRV_CMOS is not set
+# CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
+# CONFIG_RTC_DRV_DS1742 is not set
+# CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
+# CONFIG_RTC_DRV_V3020 is not set
+
+#
+# on-CPU RTC drivers
+#
+# CONFIG_RTC_DRV_SA1100 is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+# CONFIG_GFS2_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+CONFIG_AUTOFS4_FS=y
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-15"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_TMPFS_POSIX_ACL is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_CONFIGFS_FS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_ECRYPT_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=1
+CONFIG_JFFS2_FS_WRITEBUFFER=y
+CONFIG_JFFS2_FS_WBUF_VERIFY=y
+# CONFIG_JFFS2_SUMMARY is not set
+# CONFIG_JFFS2_FS_XATTR is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+CONFIG_NFS_V4=y
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+CONFIG_NFSD_V4=y
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+CONFIG_SUNRPC_GSS=y
+# CONFIG_SUNRPC_BIND34 is not set
+CONFIG_RPCSEC_GSS_KRB5=y
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-15"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+CONFIG_NLS_CODEPAGE_850=y
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+CONFIG_NLS_ASCII=y
+CONFIG_NLS_ISO8859_1=m
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+CONFIG_NLS_ISO8859_15=m
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+# CONFIG_DLM is not set
+CONFIG_INSTRUMENTATION=y
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+
+#
+# Kernel hacking
+#
+CONFIG_PRINTK_TIME=y
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_MAGIC_SYSRQ=y
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_HEADERS_CHECK is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_SHIRQ is not set
+CONFIG_DETECT_SOFTLOCKUP=y
+CONFIG_SCHED_DEBUG=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_TIMER_STATS is not set
+# CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_PREEMPT=y
+# CONFIG_DEBUG_RT_MUTEXES is not set
+# CONFIG_RT_MUTEX_TESTER is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_LOCK_ALLOC is not set
+# CONFIG_PROVE_LOCKING is not set
+# CONFIG_LOCK_STAT is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+CONFIG_DEBUG_BUGVERBOSE=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_DEBUG_VM is not set
+# CONFIG_DEBUG_LIST is not set
+# CONFIG_DEBUG_SG is not set
+CONFIG_FRAME_POINTER=y
+CONFIG_FORCED_INLINING=y
+# CONFIG_BOOT_PRINTK_DELAY is not set
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_FAULT_INJECTION is not set
+# CONFIG_SAMPLES is not set
+CONFIG_DEBUG_USER=y
+CONFIG_DEBUG_ERRORS=y
+CONFIG_DEBUG_LL=y
+# CONFIG_DEBUG_ICEDCC is not set
+
+#
+# Security options
+#
+CONFIG_KEYS=y
+CONFIG_KEYS_DEBUG_PROC_KEYS=y
+CONFIG_SECURITY=y
+# CONFIG_SECURITY_NETWORK is not set
+CONFIG_SECURITY_CAPABILITIES=y
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_SECURITY_ROOTPLUG is not set
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_ALGAPI=y
+CONFIG_CRYPTO_BLKCIPHER=y
+CONFIG_CRYPTO_MANAGER=y
+# CONFIG_CRYPTO_HMAC is not set
+# CONFIG_CRYPTO_XCBC is not set
+# CONFIG_CRYPTO_NULL is not set
+# CONFIG_CRYPTO_MD4 is not set
+CONFIG_CRYPTO_MD5=y
+CONFIG_CRYPTO_SHA1=m
+CONFIG_CRYPTO_SHA256=m
+CONFIG_CRYPTO_SHA512=m
+# CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
+# CONFIG_CRYPTO_GF128MUL is not set
+CONFIG_CRYPTO_ECB=y
+CONFIG_CRYPTO_CBC=y
+CONFIG_CRYPTO_PCBC=m
+# CONFIG_CRYPTO_LRW is not set
+# CONFIG_CRYPTO_XTS is not set
+# CONFIG_CRYPTO_CRYPTD is not set
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_FCRYPT is not set
+# CONFIG_CRYPTO_BLOWFISH is not set
+# CONFIG_CRYPTO_TWOFISH is not set
+# CONFIG_CRYPTO_SERPENT is not set
+CONFIG_CRYPTO_AES=m
+# CONFIG_CRYPTO_CAST5 is not set
+# CONFIG_CRYPTO_CAST6 is not set
+# CONFIG_CRYPTO_TEA is not set
+CONFIG_CRYPTO_ARC4=y
+# CONFIG_CRYPTO_KHAZAD is not set
+# CONFIG_CRYPTO_ANUBIS is not set
+# CONFIG_CRYPTO_SEED is not set
+CONFIG_CRYPTO_DEFLATE=m
+CONFIG_CRYPTO_MICHAEL_MIC=m
+CONFIG_CRYPTO_CRC32C=y
+# CONFIG_CRYPTO_CAMELLIA is not set
+# CONFIG_CRYPTO_TEST is not set
+# CONFIG_CRYPTO_AUTHENC is not set
+CONFIG_CRYPTO_HW=y
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+CONFIG_CRC_CCITT=y
+CONFIG_CRC16=y
+# CONFIG_CRC_ITU_T is not set
+CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
+CONFIG_LIBCRC32C=y
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
+CONFIG_REED_SOLOMON=y
+CONFIG_REED_SOLOMON_DEC16=y
+CONFIG_PLIST=y
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_IOPORT=y
+CONFIG_HAS_DMA=y
diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index e5b7063ed818..0908bea0f609 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -103,6 +103,10 @@ config MACH_EM_X270
 	bool "CompuLab EM-x270 platform"
 	select PXA27x
 
+config MACH_COLIBRI
+	bool "Toradex Colibri PX27x"
+	select PXA27x
+
 config MACH_ZYLONITE
 	bool "PXA3xx Development Platform"
 	select PXA3xx
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 1af56805b954..b5c916c0747d 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_MACH_LOGICPD_PXA270) += lpd270.o
 obj-$(CONFIG_MACH_MAINSTONE)	+= mainstone.o
 obj-$(CONFIG_ARCH_PXA_IDP)	+= idp.o
 obj-$(CONFIG_MACH_TRIZEPS4)	+= trizeps4.o
+obj-$(CONFIG_MACH_COLIBRI)	+= colibri.o
 obj-$(CONFIG_PXA_SHARP_C7xx)	+= corgi.o corgi_ssp.o corgi_lcd.o sharpsl_pm.o corgi_pm.o
 obj-$(CONFIG_PXA_SHARP_Cxx00)	+= spitz.o corgi_ssp.o corgi_lcd.o sharpsl_pm.o spitz_pm.o
 obj-$(CONFIG_MACH_AKITA)	+= akita-ioexp.o
diff --git a/arch/arm/mach-pxa/colibri.c b/arch/arm/mach-pxa/colibri.c
new file mode 100644
index 000000000000..6db54e31c397
--- /dev/null
+++ b/arch/arm/mach-pxa/colibri.c
@@ -0,0 +1,134 @@
+/*
+ *  linux/arch/arm/mach-pxa/colibri.c
+ *
+ *  Support for Toradex PXA27x based Colibri module
+ *  Daniel Mack <daniel@caiaq.de>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/sysdev.h>
+#include <linux/interrupt.h>
+#include <linux/bitops.h>
+#include <linux/ioport.h>
+#include <linux/delay.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/physmap.h>
+#include <asm/mach-types.h>
+#include <asm/hardware.h>
+#include <asm/irq.h>
+#include <asm/sizes.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/irq.h>
+#include <asm/mach/flash.h>
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/colibri.h>
+
+#include "generic.h"
+#include "devices.h"
+
+/*
+ * Flash
+ */
+static struct mtd_partition colibri_partitions[] = {
+	{
+		.name =		"Bootloader",
+		.offset =	0x00000000,
+		.size =		0x00040000,
+		.mask_flags =	MTD_WRITEABLE  /* force read-only */
+	}, {
+		.name =		"Kernel",
+		.offset =	0x00040000,
+		.size =		0x00400000,
+		.mask_flags =	0
+	}, {
+		.name =		"Rootfs",
+		.offset =	0x00440000,
+		.size =		MTDPART_SIZ_FULL,
+		.mask_flags =	0
+	}
+};
+
+static struct physmap_flash_data colibri_flash_data[] = {
+	{
+		.width		= 4,			/* bankwidth in bytes */
+		.parts		= colibri_partitions,
+		.nr_parts	= ARRAY_SIZE(colibri_partitions)
+	}
+};
+
+static struct resource flash_resource = {
+	.start	= PXA_CS0_PHYS,
+	.end	= PXA_CS0_PHYS + SZ_32M - 1,
+	.flags	= IORESOURCE_MEM,
+};
+
+static struct platform_device flash_device = {
+	.name	= "physmap-flash",
+	.id	= 0,
+	.dev 	= {
+		.platform_data = colibri_flash_data,
+	},
+	.resource = &flash_resource,
+	.num_resources = 1,
+};
+
+/*
+ * DM9000 Ethernet
+ */
+static struct resource dm9000_resources[] = {
+	[0] = {
+		.start	= COLIBRI_ETH_PHYS,
+		.end	= COLIBRI_ETH_PHYS + 3,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= COLIBRI_ETH_PHYS + 4,
+		.end	= COLIBRI_ETH_PHYS + 4 + 500,
+		.flags	= IORESOURCE_MEM,
+	},
+	[2] = {
+		.start	= COLIBRI_ETH_IRQ,
+		.end	= COLIBRI_ETH_IRQ,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device dm9000_device = {
+	.name		= "dm9000",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(dm9000_resources),
+	.resource	= dm9000_resources,
+};
+
+static struct platform_device *colibri_devices[] __initdata = {
+	&flash_device,
+	&dm9000_device,
+};
+
+static void __init colibri_init(void)
+{
+	/* DM9000 LAN */
+	pxa_gpio_mode(GPIO78_nCS_2_MD);
+	pxa_gpio_mode(GPIO_DM9000 | GPIO_IN);
+	set_irq_type(COLIBRI_ETH_IRQ, IRQT_FALLING);
+
+	platform_add_devices(colibri_devices, ARRAY_SIZE(colibri_devices));
+}
+
+MACHINE_START(COLIBRI, "Toradex Colibri PXA27x")
+	.phys_io	= 0x40000000,
+	.io_pg_offst	= (io_p2v(0x40000000) >> 18) & 0xfffc,
+	.boot_params	= COLIBRI_SDRAM_BASE + 0x100,
+	.init_machine	= colibri_init,
+	.map_io		= pxa_map_io,
+	.init_irq	= pxa27x_init_irq,
+	.timer		= &pxa_timer,
+MACHINE_END
diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c
index 3286d2a0a870..6a20a5491a96 100644
--- a/drivers/net/dm9000.c
+++ b/drivers/net/dm9000.c
@@ -66,6 +66,7 @@
 #include <linux/dm9000.h>
 #include <linux/delay.h>
 #include <linux/platform_device.h>
+#include <linux/irq.h>
 
 #include <asm/delay.h>
 #include <asm/irq.h>
@@ -113,7 +114,7 @@
 #define writesl	outsl
 #define DM9000_IRQ_FLAGS	(IRQF_SHARED | IRQF_TRIGGER_HIGH)
 #else
-#define DM9000_IRQ_FLAGS	IRQF_SHARED
+#define DM9000_IRQ_FLAGS	(IRQF_SHARED | IRQT_RISING)
 #endif
 
 /*
diff --git a/include/asm-arm/arch-pxa/colibri.h b/include/asm-arm/arch-pxa/colibri.h
new file mode 100644
index 000000000000..2ae373fb5675
--- /dev/null
+++ b/include/asm-arm/arch-pxa/colibri.h
@@ -0,0 +1,19 @@
+#ifndef _COLIBRI_H_
+#define _COLIBRI_H_
+
+/* physical memory regions */
+#define COLIBRI_FLASH_PHYS	(PXA_CS0_PHYS)  /* Flash region */
+#define COLIBRI_ETH_PHYS	(PXA_CS2_PHYS)  /* Ethernet DM9000 region */
+#define COLIBRI_SDRAM_BASE	0xa0000000      /* SDRAM region */
+
+/* virtual memory regions */
+#define COLIBRI_DISK_VIRT	0xF0000000	/* Disk On Chip region */
+
+/* size of flash */
+#define COLIBRI_FLASH_SIZE	0x02000000	/* Flash size 32 MB */
+
+/* Ethernet Controller Davicom DM9000 */
+#define GPIO_DM9000		114
+#define COLIBRI_ETH_IRQ	IRQ_GPIO(GPIO_DM9000)
+
+#endif /* _COLIBRI_H_ */
-- 
cgit v1.2.3


From 1fb4e5611a9d0b524d85f0db4402aa439ebdb331 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:29 +0100
Subject: [ARM] 4719/1: S3C2412: Update SPI register definitions for the
 S3C2412

Add S3C2412 register definitions.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/plat-s3c24xx/regs-spi.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/plat-s3c24xx/regs-spi.h b/include/asm-arm/plat-s3c24xx/regs-spi.h
index 4a499a138256..960907f17914 100644
--- a/include/asm-arm/plat-s3c24xx/regs-spi.h
+++ b/include/asm-arm/plat-s3c24xx/regs-spi.h
@@ -17,6 +17,7 @@
 
 #define S3C2410_SPCON	(0x00)
 
+#define S3C2412_SPCON_DIRC_RX	  (1<<7)
 #define S3C2410_SPCON_SMOD_DMA	  (2<<5)	/* DMA mode */
 #define S3C2410_SPCON_SMOD_INT	  (1<<5)	/* interrupt mode */
 #define S3C2410_SPCON_SMOD_POLL   (0<<5)	/* polling mode */
@@ -37,7 +38,7 @@
 #define S3C2410_SPSTA_DCOL	  (1<<2)	/* Data Collision Error */
 #define S3C2410_SPSTA_MULD	  (1<<1)	/* Multi Master Error */
 #define S3C2410_SPSTA_READY	  (1<<0)	/* Data Tx/Rx ready */
-
+#define S3C2412_SPSTA_READY_ORG	  (1<<3)
 
 #define S3C2410_SPPIN	 (0x08)
 
@@ -46,9 +47,13 @@
 #define S3C2400_SPPIN_nCS     	  (1<<1)	/* SPI Card Select */
 #define S3C2410_SPPIN_KEEP	  (1<<0)	/* Master Out keep */
 
-
 #define S3C2410_SPPRE	 (0x0C)
 #define S3C2410_SPTDAT	 (0x10)
 #define S3C2410_SPRDAT	 (0x14)
 
+#define S3C2412_TXFIFO	 (0x18)
+#define S3C2412_RXFIFO	 (0x18)
+#define S3C2412_SPFIC	 (0x24)
+
+
 #endif /* __ASM_ARCH_REGS_SPI_H */
-- 
cgit v1.2.3


From cf08167c82b38ed9f6401e68b3c91fb11496f101 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:28 +0100
Subject: [ARM] 4720/1: S3C2412: Add power configuration registers for battery
 flat behaviour.

Add the S3C2412_PWRCFG values for the action taken on detecting that
the battery is flat.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-power.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-power.h b/include/asm-arm/arch-s3c2410/regs-power.h
index f79987be55e8..13d13b7cfe98 100644
--- a/include/asm-arm/arch-s3c2410/regs-power.h
+++ b/include/asm-arm/arch-s3c2410/regs-power.h
@@ -23,7 +23,8 @@
 #define S3C2412_INFORM2		S3C24XX_PWRREG(0x78)
 #define S3C2412_INFORM3		S3C24XX_PWRREG(0x7C)
 
-#define S3C2412_PWRCFG_BATF_IGNORE		(0<<0)
+#define S3C2412_PWRCFG_BATF_IRQ			(1<<0)
+#define S3C2412_PWRCFG_BATF_IGNORE		(2<<0)
 #define S3C2412_PWRCFG_BATF_SLEEP		(3<<0)
 #define S3C2412_PWRCFG_BATF_MASK		(3<<0)
 
-- 
cgit v1.2.3


From d652074ebcedd13c7ce760c7f3327f30862852fb Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:31 +0100
Subject: [ARM] 4721/1: S3C24XX: Ensure watchdog clock is enbaled for hard
 reset

If the hard reset routine is using the watchdog, then
ensure that the clock for the watchdog has been enabled
before we try and issue a reset.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/system.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/system.h b/include/asm-arm/arch-s3c2410/system.h
index 63891786dfa0..cb9cd9fb8611 100644
--- a/include/asm-arm/arch-s3c2410/system.h
+++ b/include/asm-arm/arch-s3c2410/system.h
@@ -20,6 +20,9 @@
 #include <asm/plat-s3c/regs-watchdog.h>
 #include <asm/arch/regs-clock.h>
 
+#include <linux/clk.h>
+#include <linux/err.h>
+
 void (*s3c24xx_idle)(void);
 void (*s3c24xx_reset_hook)(void);
 
@@ -59,6 +62,8 @@ static void arch_idle(void)
 static void
 arch_reset(char mode)
 {
+	struct clk *wdtclk;
+
 	if (mode == 's') {
 		cpu_reset(0);
 	}
@@ -70,6 +75,12 @@ arch_reset(char mode)
 
 	__raw_writel(0, S3C2410_WTCON);	  /* disable watchdog, to be safe  */
 
+	wdtclk = clk_get(NULL, "watchdog");
+	if (!IS_ERR(wdtclk)) {
+		clk_enable(wdtclk);
+	} else
+		printk(KERN_WARNING "%s: warning: cannot get watchdog clock\n", __func__);
+
 	/* put initial values into count and data */
 	__raw_writel(0x100, S3C2410_WTCNT);
 	__raw_writel(0x100, S3C2410_WTDAT);
-- 
cgit v1.2.3


From ff44b49b55edee33b7ecea987be1a6413162581b Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:32 +0100
Subject: [ARM] 4722/1: S3C24XX: Improve output if watchdog reset fails

If the watchdog reset fails and we decided to take the jump
to zero approach, allow 50ms for the UARTS to drain the FIFOs
before calling into a bootloader that may flush the output.

Also reduece the waits and the timeout values as 5 seconds is
rather long.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/system.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/system.h b/include/asm-arm/arch-s3c2410/system.h
index cb9cd9fb8611..14de4e596f87 100644
--- a/include/asm-arm/arch-s3c2410/system.h
+++ b/include/asm-arm/arch-s3c2410/system.h
@@ -82,18 +82,21 @@ arch_reset(char mode)
 		printk(KERN_WARNING "%s: warning: cannot get watchdog clock\n", __func__);
 
 	/* put initial values into count and data */
-	__raw_writel(0x100, S3C2410_WTCNT);
-	__raw_writel(0x100, S3C2410_WTDAT);
+	__raw_writel(0x80, S3C2410_WTCNT);
+	__raw_writel(0x80, S3C2410_WTDAT);
 
 	/* set the watchdog to go and reset... */
 	__raw_writel(S3C2410_WTCON_ENABLE|S3C2410_WTCON_DIV16|S3C2410_WTCON_RSTEN |
 		     S3C2410_WTCON_PRESCALE(0x20), S3C2410_WTCON);
 
 	/* wait for reset to assert... */
-	mdelay(5000);
+	mdelay(500);
 
 	printk(KERN_ERR "Watchdog reset failed to assert reset\n");
 
+	/* delay to allow the serial port to show the message */
+	mdelay(50);
+
 	/* we'll take a jump through zero as a poor second */
 	cpu_reset(0);
 }
-- 
cgit v1.2.3


From b708d7bf18fd90105b8099cc558d66477dafd8ea Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:39 +0100
Subject: [ARM] 4726/1: S3C2412: IIS register definitions

The S3C2412 IIS engine differs from the previous
SoC in the range, so add a set of register definitions
in a seperate file for it.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h | 72 +++++++++++++++++++++++++
 1 file changed, 72 insertions(+)
 create mode 100644 include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h

(limited to 'include')

diff --git a/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h b/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h
new file mode 100644
index 000000000000..25d4058bcfed
--- /dev/null
+++ b/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h
@@ -0,0 +1,72 @@
+/* linux/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h
+ *
+ * Copyright 2007 Simtec Electronics <linux@simtec.co.uk>
+ *	http://armlinux.simtec.co.uk/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * S3C2412 IIS register definition
+*/
+
+#ifndef __ASM_ARCH_REGS_S3C2412_IIS_H
+#define __ASM_ARCH_REGS_S3C2412_IIS_H
+
+#define S3C2412_IISCON			(0x00)
+#define S3C2412_IISMOD			(0x04)
+#define S3C2412_IISFIC			(0x08)
+#define S3C2412_IISPSR			(0x0C)
+#define S3C2412_IISTXD			(0x10)
+#define S3C2412_IISRXD			(0x14)
+
+#define S3C2412_IISCON_LRINDEX		(1 << 11)
+#define S3C2412_IISCON_TXFIFO_EMPTY	(1 << 10)
+#define S3C2412_IISCON_RXFIFO_EMPTY	(1 << 9)
+#define S3C2412_IISCON_TXFIFO_FULL	(1 << 8)
+#define S3C2412_IISCON_RXFIFO_FULL	(1 << 7)
+#define S3C2412_IISCON_TXDMA_PAUSE	(1 << 6)
+#define S3C2412_IISCON_RXDMA_PAUSE	(1 << 5)
+#define S3C2412_IISCON_TXCH_PAUSE	(1 << 4)
+#define S3C2412_IISCON_RXCH_PAUSE	(1 << 3)
+#define S3C2412_IISCON_TXDMA_ACTIVE	(1 << 2)
+#define S3C2412_IISCON_RXDMA_ACTIVE	(1 << 1)
+#define S3C2412_IISCON_IIS_ACTIVE	(1 << 0)
+
+#define S3C2412_IISMOD_MASTER_INTERNAL	(0 << 10)
+#define S3C2412_IISMOD_MASTER_EXTERNAL	(1 << 10)
+#define S3C2412_IISMOD_SLAVE		(2 << 10)
+#define S3C2412_IISMOD_MASTER_MASK	(3 << 10)
+#define S3C2412_IISMOD_MODE_TXONLY	(0 << 8)
+#define S3C2412_IISMOD_MODE_RXONLY	(1 << 8)
+#define S3C2412_IISMOD_MODE_TXRX	(2 << 8)
+#define S3C2412_IISMOD_MODE_MASK	(3 << 8)
+#define S3C2412_IISMOD_LR_LLOW		(0 << 7)
+#define S3C2412_IISMOD_LR_RLOW		(1 << 7)
+#define S3C2412_IISMOD_SDF_IIS		(0 << 5)
+#define S3C2412_IISMOD_SDF_MSB		(0 << 5)
+#define S3C2412_IISMOD_SDF_LSB		(0 << 5)
+#define S3C2412_IISMOD_SDF_MASK		(3 << 5)
+#define S3C2412_IISMOD_RCLK_256FS	(0 << 3)
+#define S3C2412_IISMOD_RCLK_512FS	(1 << 3)
+#define S3C2412_IISMOD_RCLK_384FS	(2 << 3)
+#define S3C2412_IISMOD_RCLK_768FS	(3 << 3)
+#define S3C2412_IISMOD_RCLK_MASK 	(3 << 3)
+#define S3C2412_IISMOD_BCLK_32FS	(0 << 1)
+#define S3C2412_IISMOD_BCLK_48FS	(1 << 1)
+#define S3C2412_IISMOD_BCLK_16FS	(2 << 1)
+#define S3C2412_IISMOD_BCLK_24FS	(3 << 1)
+#define S3C2412_IISMOD_BCLK_MASK	(3 << 1)
+#define S3C2412_IISMOD_8BIT		(1 << 0)
+
+#define S3C2412_IISPSR_PSREN		(1 << 15)
+
+#define S3C2412_IISFIC_TXFLUSH		(1 << 15)
+#define S3C2412_IISFIC_RXFLUSH		(1 << 7)
+#define S3C2412_IISFIC_TXCOUNT(x)	(((x) >>  8) & 0xf)
+#define S3C2412_IISFIC_RXCOUNT(x)	(((x) >>  0) & 0xf)
+
+
+
+#endif /* __ASM_ARCH_REGS_S3C2412_IIS_H */
+
-- 
cgit v1.2.3


From 0953612a479491b57db794580b13a8e39b810724 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:37 +0100
Subject: [ARM] 4727/1: S3C2412: Remove unused GPESLPCON

S3C2412_GPESLPCON does not exist in the register
mappings, so remove it.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2412/pm.c               | 1 -
 include/asm-arm/arch-s3c2410/regs-gpio.h | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-s3c2412/pm.c b/arch/arm/mach-s3c2412/pm.c
index 8988dac388a9..974ffb130479 100644
--- a/arch/arm/mach-s3c2412/pm.c
+++ b/arch/arm/mach-s3c2412/pm.c
@@ -88,7 +88,6 @@ static struct sleep_save s3c2412_sleep[] = {
 	SAVE_ITEM(S3C2412_GPBSLPCON),
 	SAVE_ITEM(S3C2412_GPCSLPCON),
 	SAVE_ITEM(S3C2412_GPDSLPCON),
-	SAVE_ITEM(S3C2412_GPESLPCON),
 	SAVE_ITEM(S3C2412_GPFSLPCON),
 	SAVE_ITEM(S3C2412_GPGSLPCON),
 	SAVE_ITEM(S3C2412_GPHSLPCON),
diff --git a/include/asm-arm/arch-s3c2410/regs-gpio.h b/include/asm-arm/arch-s3c2410/regs-gpio.h
index b693158b2d3c..1f089fb71a80 100644
--- a/include/asm-arm/arch-s3c2410/regs-gpio.h
+++ b/include/asm-arm/arch-s3c2410/regs-gpio.h
@@ -1133,7 +1133,6 @@
 #define S3C2412_GPBSLPCON	S3C2410_GPIOREG(0x1C)
 #define S3C2412_GPCSLPCON	S3C2410_GPIOREG(0x2C)
 #define S3C2412_GPDSLPCON	S3C2410_GPIOREG(0x3C)
-#define S3C2412_GPESLPCON	S3C2410_GPIOREG(0x4C)
 #define S3C2412_GPFSLPCON	S3C2410_GPIOREG(0x5C)
 #define S3C2412_GPGSLPCON	S3C2410_GPIOREG(0x6C)
 #define S3C2412_GPHSLPCON	S3C2410_GPIOREG(0x7C)
-- 
cgit v1.2.3


From 0baada2742a42390a2ebed09e07e1fab518db884 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sun, 23 Dec 2007 03:09:33 +0100
Subject: [ARM] 4730/1: S3C2412: Ensure the PWRCFG has the right mode for RTC
 wake

Ensure that if the RTC IRQ is not selected for wake in the
base configuration, then the PWRCFG has the same value set
in it.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2412/irq.c        | 24 ++++++++++++++++++++++++
 arch/arm/plat-s3c24xx/irq.c        |  2 +-
 include/asm-arm/plat-s3c24xx/irq.h |  2 ++
 3 files changed, 27 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/arm/mach-s3c2412/irq.c b/arch/arm/mach-s3c2412/irq.c
index e9d0c769f5da..cc1917bf952a 100644
--- a/arch/arm/mach-s3c2412/irq.c
+++ b/arch/arm/mach-s3c2412/irq.c
@@ -33,6 +33,7 @@
 
 #include <asm/arch/regs-irq.h>
 #include <asm/arch/regs-gpio.h>
+#include <asm/arch/regs-power.h>
 
 #include <asm/plat-s3c24xx/cpu.h>
 #include <asm/plat-s3c24xx/irq.h>
@@ -153,6 +154,22 @@ static struct irq_chip s3c2412_irq_cfsdi = {
 	.unmask		= s3c2412_irq_cfsdi_unmask,
 };
 
+static int s3c2412_irq_rtc_wake(unsigned int irqno, unsigned int state)
+{
+	unsigned long pwrcfg;
+
+	pwrcfg = __raw_readl(S3C2412_PWRCFG);
+	if (state)
+		pwrcfg &= ~S3C2412_PWRCFG_RTC_MASKIRQ;
+	else
+		pwrcfg |= S3C2412_PWRCFG_RTC_MASKIRQ;
+	__raw_writel(pwrcfg, S3C2412_PWRCFG);
+
+	return s3c_irq_chip.set_wake(irqno, state);
+}
+
+static struct irq_chip s3c2412_irq_rtc_chip;
+
 static int s3c2412_irq_add(struct sys_device *sysdev)
 {
 	unsigned int irqno;
@@ -173,6 +190,13 @@ static int s3c2412_irq_add(struct sys_device *sysdev)
 		set_irq_flags(irqno, IRQF_VALID);
 	}
 
+	/* change RTC IRQ's set wake method */
+
+	s3c2412_irq_rtc_chip = s3c_irq_chip;
+	s3c2412_irq_rtc_chip.set_wake = s3c2412_irq_rtc_wake;
+
+	set_irq_chip(IRQ_RTC, &s3c2412_irq_rtc_chip);
+
 	return 0;
 }
 
diff --git a/arch/arm/plat-s3c24xx/irq.c b/arch/arm/plat-s3c24xx/irq.c
index 8fbc88470261..d486f5112569 100644
--- a/arch/arm/plat-s3c24xx/irq.c
+++ b/arch/arm/plat-s3c24xx/irq.c
@@ -187,7 +187,7 @@ struct irq_chip s3c_irq_level_chip = {
 	.set_wake	= s3c_irq_wake
 };
 
-static struct irq_chip s3c_irq_chip = {
+struct irq_chip s3c_irq_chip = {
 	.name		= "s3c",
 	.ack		= s3c_irq_ack,
 	.mask		= s3c_irq_mask,
diff --git a/include/asm-arm/plat-s3c24xx/irq.h b/include/asm-arm/plat-s3c24xx/irq.h
index 8af6d9579b31..45746a995343 100644
--- a/include/asm-arm/plat-s3c24xx/irq.h
+++ b/include/asm-arm/plat-s3c24xx/irq.h
@@ -15,7 +15,9 @@
 
 #define EXTINT_OFF (IRQ_EINT4 - 4)
 
+/* these are exported for arch/arm/mach-* usage */
 extern struct irq_chip s3c_irq_level_chip;
+extern struct irq_chip s3c_irq_chip;
 
 static inline void
 s3c_irqsub_mask(unsigned int irqno, unsigned int parentbit,
-- 
cgit v1.2.3


From f7275dac55008f8296cfb89a01b1e71918ac7995 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Sun, 27 Jan 2008 19:01:18 +0100
Subject: [ARM] 4775/1: s3c2410: fix compilation error if only s3c2442 cpu is
 selected

This patch fixes compilation error if only a machine with
s3c2442 cpu is selected but without s3c2440 cpu selected.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-dsc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-dsc.h b/include/asm-arm/arch-s3c2410/regs-dsc.h
index c0748511edbc..1235df70f34e 100644
--- a/include/asm-arm/arch-s3c2410/regs-dsc.h
+++ b/include/asm-arm/arch-s3c2410/regs-dsc.h
@@ -19,7 +19,7 @@
 #define S3C2412_DSC1	   S3C2410_GPIOREG(0xe0)
 #endif
 
-#if defined(CONFIG_CPU_S3C2440)
+#if defined(CONFIG_CPU_S3C244X)
 
 #define S3C2440_DSC0	   S3C2410_GPIOREG(0xc4)
 #define S3C2440_DSC1	   S3C2410_GPIOREG(0xc8)
-- 
cgit v1.2.3


From 67d729adc0e76e21c82a2c59853f25f5f784ca79 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:19 +0100
Subject: [ARM] 4779/1: S3C2412: Add s3c2412_gpio_set_sleepcfg() call

Add s3c2412_gpio_set_sleepcfg() to allow the setting of the sleep
configuration of the GPIO blocks.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2412/Makefile           |  1 +
 arch/arm/mach-s3c2412/gpio.c             | 60 ++++++++++++++++++++++++++++++++
 include/asm-arm/arch-s3c2410/hardware.h  |  5 +++
 include/asm-arm/arch-s3c2410/regs-gpio.h |  5 +++
 4 files changed, 71 insertions(+)
 create mode 100644 arch/arm/mach-s3c2412/gpio.c

(limited to 'include')

diff --git a/arch/arm/mach-s3c2412/Makefile b/arch/arm/mach-s3c2412/Makefile
index 7b782c7df7b4..267f3348301e 100644
--- a/arch/arm/mach-s3c2412/Makefile
+++ b/arch/arm/mach-s3c2412/Makefile
@@ -12,6 +12,7 @@ obj-				:=
 obj-$(CONFIG_CPU_S3C2412)	+= s3c2412.o
 obj-$(CONFIG_CPU_S3C2412)	+= irq.o
 obj-$(CONFIG_CPU_S3C2412)	+= clock.o
+obj-$(CONFIG_CPU_S3C2412)	+= gpio.o
 obj-$(CONFIG_S3C2412_DMA)	+= dma.o
 obj-$(CONFIG_S3C2412_PM)	+= pm.o sleep.o
 
diff --git a/arch/arm/mach-s3c2412/gpio.c b/arch/arm/mach-s3c2412/gpio.c
new file mode 100644
index 000000000000..8e55c3a2eab8
--- /dev/null
+++ b/arch/arm/mach-s3c2412/gpio.c
@@ -0,0 +1,60 @@
+/* linux/arch/arm/mach-s3c2412/gpio.c
+ *
+ * Copyright (c) 2007 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * http://armlinux.simtec.co.uk/.
+ *
+ * S3C2412/S3C2413 specific GPIO support
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <asm/arch/regs-gpio.h>
+
+#include <asm/hardware.h>
+
+int s3c2412_gpio_set_sleepcfg(unsigned int pin, unsigned int state)
+{
+	void __iomem *base = S3C24XX_GPIO_BASE(pin);
+	unsigned long offs = S3C2410_GPIO_OFFSET(pin);
+	unsigned long flags;
+	unsigned long slpcon;
+
+	offs *= 2;
+
+	if (pin < S3C2410_GPIO_BANKB)
+		return -EINVAL;
+
+	if (pin >= S3C2410_GPIO_BANKF &&
+	    pin <= S3C2410_GPIO_BANKG)
+		return -EINVAL;
+
+	if (pin > (S3C2410_GPIO_BANKH + 32))
+		return -EINVAL;
+
+	local_irq_save(flags);
+
+	slpcon = __raw_readl(base + 0x0C);
+
+	slpcon &= ~(3 << offs);
+	slpcon |= state << offs;
+
+	__raw_writel(slpcon, base + 0x0C);
+
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+EXPORT_SYMBOL(s3c2412_gpio_set_sleepcfg);
diff --git a/include/asm-arm/arch-s3c2410/hardware.h b/include/asm-arm/arch-s3c2410/hardware.h
index 6dadf58ff984..faeaba514ffe 100644
--- a/include/asm-arm/arch-s3c2410/hardware.h
+++ b/include/asm-arm/arch-s3c2410/hardware.h
@@ -99,6 +99,11 @@ extern int s3c2440_set_dsc(unsigned int pin, unsigned int value);
 
 #endif /* CONFIG_CPU_S3C2440 */
 
+#ifdef CONFIG_CPU_S3C2412
+
+extern int s3c2412_gpio_set_sleepcfg(unsigned int pin, unsigned int state);
+
+#endif /* CONFIG_CPU_S3C2412 */
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-arm/arch-s3c2410/regs-gpio.h b/include/asm-arm/arch-s3c2410/regs-gpio.h
index 1f089fb71a80..0ad75d716ded 100644
--- a/include/asm-arm/arch-s3c2410/regs-gpio.h
+++ b/include/asm-arm/arch-s3c2410/regs-gpio.h
@@ -1138,6 +1138,11 @@
 #define S3C2412_GPHSLPCON	S3C2410_GPIOREG(0x7C)
 
 /* definitions for each pin bit */
+#define S3C2412_GPIO_SLPCON_LOW	 ( 0x00 )
+#define S3C2412_GPIO_SLPCON_HIGH ( 0x01 )
+#define S3C2412_GPIO_SLPCON_IN   ( 0x02 )
+#define S3C2412_GPIO_SLPCON_PULL ( 0x03 )
+
 #define S3C2412_SLPCON_LOW(x)	( 0x00 << ((x) * 2))
 #define S3C2412_SLPCON_HIGH(x)	( 0x01 << ((x) * 2))
 #define S3C2412_SLPCON_IN(x)	( 0x02 << ((x) * 2))
-- 
cgit v1.2.3


From c6709e8ef5752314c22c75bc7575f9be390e215b Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:20 +0100
Subject: [ARM] 4780/1: S3C2412: Allow for seperate DMA channels for TX and RX

The current S3C24XX DMA code does not allow for an peripheral
that has one channel for RX and another for TX.

This patch adds a per-cpu dma operation to select the transmit
or receive channel, and adds support to the S3C2412 for the
seperate DMA channels for TX and RX.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2412/dma.c        | 39 ++++++++++++++++++++++++++++++++++++--
 arch/arm/plat-s3c24xx/dma.c        | 16 ++++++++++++----
 include/asm-arm/plat-s3c24xx/dma.h |  5 +++++
 3 files changed, 54 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-s3c2412/dma.c b/arch/arm/mach-s3c2412/dma.c
index c9cd51439126..1dd864993566 100644
--- a/arch/arm/mach-s3c2412/dma.c
+++ b/arch/arm/mach-s3c2412/dma.c
@@ -40,106 +40,141 @@ static struct s3c24xx_dma_map __initdata s3c2412_dma_mappings[] = {
 	[DMACH_XD0] = {
 		.name		= "xdreq0",
 		.channels	= MAP(S3C2412_DMAREQSEL_XDREQ0),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_XDREQ0),
 	},
 	[DMACH_XD1] = {
 		.name		= "xdreq1",
 		.channels	= MAP(S3C2412_DMAREQSEL_XDREQ1),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_XDREQ1),
 	},
 	[DMACH_SDI] = {
 		.name		= "sdi",
 		.channels	= MAP(S3C2412_DMAREQSEL_SDI),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_SDI),
 		.hw_addr.to	= S3C2410_PA_SDI + S3C2410_SDIDATA,
 		.hw_addr.from	= S3C2410_PA_SDI + S3C2410_SDIDATA,
 	},
 	[DMACH_SPI0] = {
 		.name		= "spi0",
 		.channels	= MAP(S3C2412_DMAREQSEL_SPI0TX),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_SPI0RX),
 		.hw_addr.to	= S3C2410_PA_SPI + S3C2410_SPTDAT,
 		.hw_addr.from	= S3C2410_PA_SPI + S3C2410_SPRDAT,
 	},
 	[DMACH_SPI1] = {
 		.name		= "spi1",
 		.channels	= MAP(S3C2412_DMAREQSEL_SPI1TX),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_SPI1RX),
 		.hw_addr.to	= S3C2410_PA_SPI + S3C2412_SPI1 + S3C2410_SPTDAT,
 		.hw_addr.from	= S3C2410_PA_SPI + S3C2412_SPI1  + S3C2410_SPRDAT,
 	},
 	[DMACH_UART0] = {
 		.name		= "uart0",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART0_0),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART0_0),
 		.hw_addr.to	= S3C2410_PA_UART0 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART0 + S3C2410_URXH,
 	},
 	[DMACH_UART1] = {
 		.name		= "uart1",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART1_0),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART1_0),
 		.hw_addr.to	= S3C2410_PA_UART1 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART1 + S3C2410_URXH,
 	},
       	[DMACH_UART2] = {
 		.name		= "uart2",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART2_0),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART2_0),
 		.hw_addr.to	= S3C2410_PA_UART2 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART2 + S3C2410_URXH,
 	},
 	[DMACH_UART0_SRC2] = {
 		.name		= "uart0",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART0_1),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART0_1),
 		.hw_addr.to	= S3C2410_PA_UART0 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART0 + S3C2410_URXH,
 	},
 	[DMACH_UART1_SRC2] = {
 		.name		= "uart1",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART1_1),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART1_1),
 		.hw_addr.to	= S3C2410_PA_UART1 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART1 + S3C2410_URXH,
 	},
       	[DMACH_UART2_SRC2] = {
 		.name		= "uart2",
 		.channels	= MAP(S3C2412_DMAREQSEL_UART2_1),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_UART2_1),
 		.hw_addr.to	= S3C2410_PA_UART2 + S3C2410_UTXH,
 		.hw_addr.from	= S3C2410_PA_UART2 + S3C2410_URXH,
 	},
 	[DMACH_TIMER] = {
 		.name		= "timer",
 		.channels	= MAP(S3C2412_DMAREQSEL_TIMER),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_TIMER),
 	},
 	[DMACH_I2S_IN] = {
 		.name		= "i2s-sdi",
 		.channels	= MAP(S3C2412_DMAREQSEL_I2SRX),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_I2SRX),
 		.hw_addr.from	= S3C2410_PA_IIS + S3C2412_IISRXD,
 	},
 	[DMACH_I2S_OUT] = {
 		.name		= "i2s-sdo",
 		.channels	= MAP(S3C2412_DMAREQSEL_I2STX),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_I2STX),
 		.hw_addr.to	= S3C2410_PA_IIS + S3C2412_IISTXD,
 	},
 	[DMACH_USB_EP1] = {
 		.name		= "usb-ep1",
 		.channels	= MAP(S3C2412_DMAREQSEL_USBEP1),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_USBEP1),
 	},
 	[DMACH_USB_EP2] = {
 		.name		= "usb-ep2",
 		.channels	= MAP(S3C2412_DMAREQSEL_USBEP2),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_USBEP2),
 	},
 	[DMACH_USB_EP3] = {
 		.name		= "usb-ep3",
 		.channels	= MAP(S3C2412_DMAREQSEL_USBEP3),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_USBEP3),
 	},
 	[DMACH_USB_EP4] = {
 		.name		= "usb-ep4",
 		.channels	= MAP(S3C2412_DMAREQSEL_USBEP4),
+		.channels_rx	= MAP(S3C2412_DMAREQSEL_USBEP4),
 	},
 };
 
+static void s3c2412_dma_direction(struct s3c2410_dma_chan *chan,
+				  struct s3c24xx_dma_map *map,
+				  enum s3c2410_dmasrc dir)
+{
+	unsigned long chsel;
+
+	if (dir == S3C2410_DMASRC_HW)
+		chsel = map->channels_rx[0];
+	else
+		chsel = map->channels[0];
+
+	chsel &= ~DMA_CH_VALID;
+	chsel |= S3C2412_DMAREQSEL_HW;
+
+	writel(chsel, chan->regs + S3C2412_DMA_DMAREQSEL);
+}
+
 static void s3c2412_dma_select(struct s3c2410_dma_chan *chan,
 			       struct s3c24xx_dma_map *map)
 {
-	writel(map->channels[0] | S3C2412_DMAREQSEL_HW,
-	       chan->regs + S3C2412_DMA_DMAREQSEL);
+	s3c2412_dma_direction(chan, map, chan->source);
 }
 
 static struct s3c24xx_dma_selection __initdata s3c2412_dma_sel = {
 	.select		= s3c2412_dma_select,
+	.direction	= s3c2412_dma_direction,
 	.dcon_mask	= 0,
 	.map		= s3c2412_dma_mappings,
 	.map_size	= ARRAY_SIZE(s3c2412_dma_mappings),
diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c
index aae1b9cbaf44..10ef3995b114 100644
--- a/arch/arm/plat-s3c24xx/dma.c
+++ b/arch/arm/plat-s3c24xx/dma.c
@@ -1184,7 +1184,7 @@ int s3c2410_dma_devconfig(int channel,
 		dma_wrreg(chan, S3C2410_DMA_DIDSTC, (0<<1) | (0<<0));
 
 		chan->addr_reg = dma_regaddr(chan, S3C2410_DMA_DIDST);
-		return 0;
+		break;
 
 	case S3C2410_DMASRC_MEM:
 		/* source is memory */
@@ -1195,11 +1195,19 @@ int s3c2410_dma_devconfig(int channel,
 		dma_wrreg(chan, S3C2410_DMA_DIDSTC, hwcfg & 3);
 
 		chan->addr_reg = dma_regaddr(chan, S3C2410_DMA_DISRC);
-		return 0;
+		break;
+
+	default:
+		printk(KERN_ERR "dma%d: invalid source type (%d)\n",
+		       channel, source);
+
+		return -EINVAL;
 	}
 
-	printk(KERN_ERR "dma%d: invalid source type (%d)\n", channel, source);
-	return -EINVAL;
+	if (dma_sel.direction != NULL)
+		(dma_sel.direction)(chan, chan->map, source);
+
+	return 0;
 }
 
 EXPORT_SYMBOL(s3c2410_dma_devconfig);
diff --git a/include/asm-arm/plat-s3c24xx/dma.h b/include/asm-arm/plat-s3c24xx/dma.h
index 2c59406435e5..c78efe316fc8 100644
--- a/include/asm-arm/plat-s3c24xx/dma.h
+++ b/include/asm-arm/plat-s3c24xx/dma.h
@@ -32,6 +32,7 @@ struct s3c24xx_dma_map {
 	struct s3c24xx_dma_addr  hw_addr;
 
 	unsigned long		 channels[S3C2410_DMA_CHANNELS];
+	unsigned long		 channels_rx[S3C2410_DMA_CHANNELS];
 };
 
 struct s3c24xx_dma_selection {
@@ -41,6 +42,10 @@ struct s3c24xx_dma_selection {
 
 	void	(*select)(struct s3c2410_dma_chan *chan,
 			  struct s3c24xx_dma_map *map);
+
+	void	(*direction)(struct s3c2410_dma_chan *chan,
+			     struct s3c24xx_dma_map *map,
+			     enum s3c2410_dmasrc dir);
 };
 
 extern int s3c24xx_dma_init_map(struct s3c24xx_dma_selection *sel);
-- 
cgit v1.2.3


From c58f7a1d36709e50398c05df9419386befef2c59 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:21 +0100
Subject: [ARM] 4781/1: S3C24XX: DMA suspend and resume support

If an DMA channel was active at suspend, then ensure that
it is correctly reconfigured when the system resumes.

Note, the previous policy was for each driver to handle their
own reconfiguration on resume. The policy has been changed to
make the individual driver's job easier.

Signed-off-by: Ben Dooks <ben-linux@flfuf.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-s3c24xx/dma.c        | 26 +++++++++++++++++++++++++-
 include/asm-arm/arch-s3c2410/dma.h |  1 +
 2 files changed, 26 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c
index 10ef3995b114..68ce89db17db 100644
--- a/arch/arm/plat-s3c24xx/dma.c
+++ b/arch/arm/plat-s3c24xx/dma.c
@@ -1173,6 +1173,7 @@ int s3c2410_dma_devconfig(int channel,
 
 	chan->source = source;
 	chan->dev_addr = devaddr;
+	chan->hw_cfg = hwcfg;
 
 	switch (source) {
 	case S3C2410_DMASRC_HW:
@@ -1235,6 +1236,10 @@ int s3c2410_dma_getposition(dmach_t channel, dma_addr_t *src, dma_addr_t *dst)
 
 EXPORT_SYMBOL(s3c2410_dma_getposition);
 
+static struct s3c2410_dma_chan *to_dma_chan(struct sys_device *dev)
+{
+	return container_of(dev, struct s3c2410_dma_chan, dev);
+}
 
 /* system device class */
 
@@ -1242,7 +1247,7 @@ EXPORT_SYMBOL(s3c2410_dma_getposition);
 
 static int s3c2410_dma_suspend(struct sys_device *dev, pm_message_t state)
 {
-	struct s3c2410_dma_chan *cp = container_of(dev, struct s3c2410_dma_chan, dev);
+	struct s3c2410_dma_chan *cp = to_dma_chan(dev);
 
 	printk(KERN_DEBUG "suspending dma channel %d\n", cp->number);
 
@@ -1264,6 +1269,24 @@ static int s3c2410_dma_suspend(struct sys_device *dev, pm_message_t state)
 
 static int s3c2410_dma_resume(struct sys_device *dev)
 {
+	struct s3c2410_dma_chan *cp = to_dma_chan(dev);
+	unsigned int no = cp->number | DMACH_LOW_LEVEL;
+
+	/* restore channel's hardware configuration */
+
+	if (!cp->in_use)
+		return 0;
+
+	printk(KERN_INFO "dma%d: restoring configuration\n", cp->number);
+
+	s3c2410_dma_config(no, cp->xfer_unit, cp->dcon);
+	s3c2410_dma_devconfig(no, cp->source, cp->hw_cfg, cp->dev_addr);
+
+	/* re-select the dma source for this channel */
+
+	if (cp->map != NULL)
+		dma_sel.select(cp, cp->map);
+
 	return 0;
 }
 
@@ -1453,6 +1476,7 @@ static struct s3c2410_dma_chan *s3c2410_dma_map_channel(int channel)
 
  found:
 	dmach = &s3c2410_chans[ch];
+	dmach->map = ch_map;
 	dma_chan_map[channel] = dmach;
 
 	/* select the channel */
diff --git a/include/asm-arm/arch-s3c2410/dma.h b/include/asm-arm/arch-s3c2410/dma.h
index c6e8d8f64938..4f291d9b7d93 100644
--- a/include/asm-arm/arch-s3c2410/dma.h
+++ b/include/asm-arm/arch-s3c2410/dma.h
@@ -214,6 +214,7 @@ struct s3c2410_dma_chan {
 	unsigned long		 dev_addr;
 	unsigned long		 load_timeout;
 	unsigned int		 flags;		/* channel flags */
+	unsigned int		 hw_cfg;	/* last hw config */
 
 	struct s3c24xx_dma_map	*map;		/* channel hw maps */
 
-- 
cgit v1.2.3


From 9f613be4a9b6d680772b0a52a43a4a94d8e131b6 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:22 +0100
Subject: [ARM] 4782/1: S3C24XX: Define FIQ_START for any FIQ users

Ensure FIQ_START is defined to allow anyone to use FIQ code on
an S3C24XX based CPU.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/irqs.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/irqs.h b/include/asm-arm/arch-s3c2410/irqs.h
index 996f65488d2d..d858b3eb5547 100644
--- a/include/asm-arm/arch-s3c2410/irqs.h
+++ b/include/asm-arm/arch-s3c2410/irqs.h
@@ -160,4 +160,7 @@
 #define NR_IRQS (IRQ_S3C2440_AC97+1)
 #endif
 
+/* Our FIQs are routable from IRQ_EINT0 to IRQ_ADCPARENT */
+#define FIQ_START		IRQ_EINT0
+
 #endif /* __ASM_ARCH_IRQ_H */
-- 
cgit v1.2.3


From bb6d9b56c1d5c1cbff39b7c337c900e2b3d8387a Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:23 +0100
Subject: [ARM] 4783/1: S3C24XX: Add s3c2410_gpio_getpull()

Add the call s3c2410_gpio_getpull() to return the
current state of the pin's pull-up.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-s3c24xx/gpio.c            | 13 +++++++++++++
 include/asm-arm/arch-s3c2410/hardware.h | 12 ++++++++++++
 2 files changed, 25 insertions(+)

(limited to 'include')

diff --git a/arch/arm/plat-s3c24xx/gpio.c b/arch/arm/plat-s3c24xx/gpio.c
index ec3a09c4d181..7cc6faec25d9 100644
--- a/arch/arm/plat-s3c24xx/gpio.c
+++ b/arch/arm/plat-s3c24xx/gpio.c
@@ -122,6 +122,19 @@ void s3c2410_gpio_pullup(unsigned int pin, unsigned int to)
 
 EXPORT_SYMBOL(s3c2410_gpio_pullup);
 
+int s3c2410_gpio_getpull(unsigned int pin)
+{
+	void __iomem *base = S3C24XX_GPIO_BASE(pin);
+	unsigned long offs = S3C2410_GPIO_OFFSET(pin);
+
+	if (pin < S3C2410_GPIO_BANKB)
+		return -EINVAL;
+
+	return (__raw_readl(base + 0x08) & (1L << offs)) ? 1 : 0;
+}
+
+EXPORT_SYMBOL(s3c2410_gpio_getpull);
+
 void s3c2410_gpio_setpin(unsigned int pin, unsigned int to)
 {
 	void __iomem *base = S3C24XX_GPIO_BASE(pin);
diff --git a/include/asm-arm/arch-s3c2410/hardware.h b/include/asm-arm/arch-s3c2410/hardware.h
index faeaba514ffe..20509e4540f6 100644
--- a/include/asm-arm/arch-s3c2410/hardware.h
+++ b/include/asm-arm/arch-s3c2410/hardware.h
@@ -87,6 +87,18 @@ extern int s3c2410_gpio_irqfilter(unsigned int pin, unsigned int on,
 
 extern void s3c2410_gpio_pullup(unsigned int pin, unsigned int to);
 
+/* s3c2410_gpio_getpull
+ *
+ * Read the state of the pull-up on a given pin
+ *
+ * return:
+ *	< 0 => error code
+ *	  0 => enabled
+ *	  1 => disabled
+*/
+
+extern int s3c2410_gpio_getpull(unsigned int pin);
+
 extern void s3c2410_gpio_setpin(unsigned int pin, unsigned int to);
 
 extern unsigned int s3c2410_gpio_getpin(unsigned int pin);
-- 
cgit v1.2.3


From e5812bf66881a7d9c67d807b09d69a023d6e4b86 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:25 +0100
Subject: [ARM] 4785/1: S3C24XX: Add _SHIFT definitions for S3C2410_BANKCON
 registers

Add definitions to allow easier decomposotion of the contents of
the S3C2410_BANKON registers

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-mem.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-mem.h b/include/asm-arm/arch-s3c2410/regs-mem.h
index e4d82341f7ba..312ff93b63c6 100644
--- a/include/asm-arm/arch-s3c2410/regs-mem.h
+++ b/include/asm-arm/arch-s3c2410/regs-mem.h
@@ -98,16 +98,19 @@
 #define S3C2410_BANKCON_Tacp3		(0x1 << 2)
 #define S3C2410_BANKCON_Tacp4		(0x2 << 2)
 #define S3C2410_BANKCON_Tacp6		(0x3 << 2)
+#define S3C2410_BANKCON_Tacp_SHIFT	(2)
 
 #define S3C2410_BANKCON_Tcah0		(0x0 << 4)
 #define S3C2410_BANKCON_Tcah1		(0x1 << 4)
 #define S3C2410_BANKCON_Tcah2		(0x2 << 4)
 #define S3C2410_BANKCON_Tcah4		(0x3 << 4)
+#define S3C2410_BANKCON_Tcah_SHIFT	(4)
 
 #define S3C2410_BANKCON_Tcoh0		(0x0 << 6)
 #define S3C2410_BANKCON_Tcoh1		(0x1 << 6)
 #define S3C2410_BANKCON_Tcoh2		(0x2 << 6)
 #define S3C2410_BANKCON_Tcoh4		(0x3 << 6)
+#define S3C2410_BANKCON_Tcoh_SHIFT	(6)
 
 #define S3C2410_BANKCON_Tacc1		(0x0 << 8)
 #define S3C2410_BANKCON_Tacc2		(0x1 << 8)
@@ -117,16 +120,19 @@
 #define S3C2410_BANKCON_Tacc8		(0x5 << 8)
 #define S3C2410_BANKCON_Tacc10		(0x6 << 8)
 #define S3C2410_BANKCON_Tacc14		(0x7 << 8)
+#define S3C2410_BANKCON_Tacc_SHIFT	(8)
 
 #define S3C2410_BANKCON_Tcos0		(0x0 << 11)
 #define S3C2410_BANKCON_Tcos1		(0x1 << 11)
 #define S3C2410_BANKCON_Tcos2		(0x2 << 11)
 #define S3C2410_BANKCON_Tcos4		(0x3 << 11)
+#define S3C2410_BANKCON_Tcos_SHIFT	(11)
 
 #define S3C2410_BANKCON_Tacs0		(0x0 << 13)
 #define S3C2410_BANKCON_Tacs1		(0x1 << 13)
 #define S3C2410_BANKCON_Tacs2		(0x2 << 13)
 #define S3C2410_BANKCON_Tacs4		(0x3 << 13)
+#define S3C2410_BANKCON_Tacs_SHIFT	(13)
 
 #define S3C2410_BANKCON_SRAM		(0x0 << 15)
 #define S3C2400_BANKCON_EDODRAM		(0x2 << 15)
-- 
cgit v1.2.3


From db9b85c527a397e4cb75956ae69acea92904f6dc Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:26 +0100
Subject: [ARM] 4786/1: S3C2412: Add SPI FIFO controll constants

Add control constants for the S3C2412 SPI unit FIFO.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/plat-s3c24xx/regs-spi.h | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/plat-s3c24xx/regs-spi.h b/include/asm-arm/plat-s3c24xx/regs-spi.h
index 960907f17914..ea565b007d04 100644
--- a/include/asm-arm/plat-s3c24xx/regs-spi.h
+++ b/include/asm-arm/plat-s3c24xx/regs-spi.h
@@ -17,7 +17,21 @@
 
 #define S3C2410_SPCON	(0x00)
 
+#define S3C2412_SPCON_RXFIFO_RB2	(0<<14)
+#define S3C2412_SPCON_RXFIFO_RB4	(1<<14)
+#define S3C2412_SPCON_RXFIFO_RB12	(2<<14)
+#define S3C2412_SPCON_RXFIFO_RB14	(3<<14)
+#define S3C2412_SPCON_TXFIFO_RB2	(0<<12)
+#define S3C2412_SPCON_TXFIFO_RB4	(1<<12)
+#define S3C2412_SPCON_TXFIFO_RB12	(2<<12)
+#define S3C2412_SPCON_TXFIFO_RB14	(3<<12)
+#define S3C2412_SPCON_RXFIFO_RESET	(1<<11) /* RxFIFO reset */
+#define S3C2412_SPCON_TXFIFO_RESET	(1<<10) /* TxFIFO reset */
+#define S3C2412_SPCON_RXFIFO_EN		(1<<9)  /* RxFIFO Enable */
+#define S3C2412_SPCON_TXFIFO_EN		(1<<8)  /* TxFIFO Enable */
+
 #define S3C2412_SPCON_DIRC_RX	  (1<<7)
+
 #define S3C2410_SPCON_SMOD_DMA	  (2<<5)	/* DMA mode */
 #define S3C2410_SPCON_SMOD_INT	  (1<<5)	/* interrupt mode */
 #define S3C2410_SPCON_SMOD_POLL   (0<<5)	/* polling mode */
@@ -35,6 +49,15 @@
 
 #define S3C2410_SPSTA	 (0x04)
 
+#define S3C2412_SPSTA_RXFIFO_AE		(1<<11)
+#define S3C2412_SPSTA_TXFIFO_AE		(1<<10)
+#define S3C2412_SPSTA_RXFIFO_ERROR	(1<<9)
+#define S3C2412_SPSTA_TXFIFO_ERROR	(1<<8)
+#define S3C2412_SPSTA_RXFIFO_FIFO	(1<<7)
+#define S3C2412_SPSTA_RXFIFO_EMPTY	(1<<6)
+#define S3C2412_SPSTA_TXFIFO_NFULL	(1<<5)
+#define S3C2412_SPSTA_TXFIFO_EMPTY	(1<<4)
+
 #define S3C2410_SPSTA_DCOL	  (1<<2)	/* Data Collision Error */
 #define S3C2410_SPSTA_MULD	  (1<<1)	/* Multi Master Error */
 #define S3C2410_SPSTA_READY	  (1<<0)	/* Data Tx/Rx ready */
-- 
cgit v1.2.3


From e95f52cd3b664eaa9c2b2b123f2a339ae0f1a92b Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:29 +0100
Subject: [ARM] 4789/1: S3C2412: Add missing CLKDIVN register values

Add S3C2412_CLKDIVN_DVSEN and S3C2412_CLKDIVN_HALFHCLK definitions to
the S3C2412_CLKDIVN set.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-clock.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-clock.h b/include/asm-arm/arch-s3c2410/regs-clock.h
index e39656b7a086..dba9df9d8713 100644
--- a/include/asm-arm/arch-s3c2410/regs-clock.h
+++ b/include/asm-arm/arch-s3c2410/regs-clock.h
@@ -138,6 +138,8 @@ s3c2410_get_pll(unsigned int pllval, unsigned int baseclk)
 #define S3C2412_CLKDIVN_PDIVN		(1<<2)
 #define S3C2412_CLKDIVN_HDIVN_MASK	(3<<0)
 #define S3C2421_CLKDIVN_ARMDIVN		(1<<3)
+#define S3C2412_CLKDIVN_DVSEN		(1<<4)
+#define S3C2412_CLKDIVN_HALFHCLK	(1<<5)
 #define S3C2412_CLKDIVN_USB48DIV	(1<<6)
 #define S3C2412_CLKDIVN_UARTDIV_MASK	(15<<8)
 #define S3C2412_CLKDIVN_UARTDIV_SHIFT	(8)
-- 
cgit v1.2.3


From 9b8c0088404778a1291191cf140bcfea082f027c Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:32 +0100
Subject: [ARM] 4792/1: S3C24XX: Remove warnings from debug-macro.S

Remove warnings left in include/asm-arm/arch-s3c2410/debug-macro.S
whilst these where being experimented with.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/debug-macro.S | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/debug-macro.S b/include/asm-arm/arch-s3c2410/debug-macro.S
index 9c8cd9abb82b..89076c322726 100644
--- a/include/asm-arm/arch-s3c2410/debug-macro.S
+++ b/include/asm-arm/arch-s3c2410/debug-macro.S
@@ -92,11 +92,9 @@
 #if defined(CONFIG_CPU_LLSERIAL_S3C2410_ONLY)
 #define fifo_full  fifo_full_s3c2410
 #define fifo_level fifo_level_s3c2410
-#warning 2410only
 #elif !defined(CONFIG_CPU_LLSERIAL_S3C2440_ONLY)
 #define fifo_full  fifo_full_s3c24xx
 #define fifo_level fifo_level_s3c24xx
-#warning generic
 #endif
 
 /* include the reset of the code which will do the work */
-- 
cgit v1.2.3


From c27cb681ac1598352569f75cb19850a12b7ef102 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 28 Jan 2008 13:01:33 +0100
Subject: [ARM] 4793/1: S3C24XX: Add IRQ->GPIO pin mapping function

Add the reverse of s3c2410_gpio_getirq to convert
a IRQ  number into a GPIO pin number.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-s3c24xx/gpio.c            | 16 ++++++++++++++++
 include/asm-arm/arch-s3c2410/hardware.h | 11 +++++++++++
 2 files changed, 27 insertions(+)

(limited to 'include')

diff --git a/arch/arm/plat-s3c24xx/gpio.c b/arch/arm/plat-s3c24xx/gpio.c
index 7cc6faec25d9..ee99dcc7f0bd 100644
--- a/arch/arm/plat-s3c24xx/gpio.c
+++ b/arch/arm/plat-s3c24xx/gpio.c
@@ -199,3 +199,19 @@ int s3c2410_gpio_getirq(unsigned int pin)
 }
 
 EXPORT_SYMBOL(s3c2410_gpio_getirq);
+
+int s3c2410_gpio_irq2pin(unsigned int irq)
+{
+	if (irq >= IRQ_EINT0 && irq <= IRQ_EINT3)
+		return S3C2410_GPF0 + (irq - IRQ_EINT0);
+
+	if (irq >= IRQ_EINT4 && irq <= IRQ_EINT7)
+		return S3C2410_GPF4 + (irq - IRQ_EINT4);
+
+	if (irq >= IRQ_EINT8 && irq <= IRQ_EINT23)
+		return S3C2410_GPG0 + (irq - IRQ_EINT8);
+
+	return -EINVAL;
+}
+
+EXPORT_SYMBOL(s3c2410_gpio_irq2pin);
diff --git a/include/asm-arm/arch-s3c2410/hardware.h b/include/asm-arm/arch-s3c2410/hardware.h
index 20509e4540f6..29592c3ebf22 100644
--- a/include/asm-arm/arch-s3c2410/hardware.h
+++ b/include/asm-arm/arch-s3c2410/hardware.h
@@ -50,6 +50,17 @@ extern unsigned int s3c2410_gpio_getcfg(unsigned int pin);
 
 extern int s3c2410_gpio_getirq(unsigned int pin);
 
+/* s3c2410_gpio_irq2pin
+ *
+ * turn the given irq number into the corresponding GPIO number
+ *
+ * returns:
+ *	< 0 = no pin
+ *	>=0 = gpio pin number
+*/
+
+extern int s3c2410_gpio_irq2pin(unsigned int irq);
+
 #ifdef CONFIG_CPU_S3C2400
 
 extern int s3c2400_gpio_getirq(unsigned int pin);
-- 
cgit v1.2.3


From 81e1a875505f2963f4d22f7e7ade39d764755f9b Mon Sep 17 00:00:00 2001
From: Michel Daenzer <michel@tungstengraphics.com>
Date: Wed, 24 Oct 2007 16:30:34 +0200
Subject: HID: Rename some code identifiers from PowerBook specific to Apple
 generic

Preserve identifiers exposed in build and run time configuration though in
order not to break existing configurations.

This is in preparation for adding support for Apple aluminum USB keyboards.

Signed-off-by: Michel Daenzer <michel@tungstengraphics.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input.c         | 75 +++++++++++++++++++++--------------------
 drivers/hid/usbhid/hid-quirks.c | 26 +++++++-------
 include/linux/hid.h             |  8 ++---
 3 files changed, 55 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index 3a18ec45a71f..85803f183fcc 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -34,10 +34,10 @@
 #include <linux/hid.h>
 #include <linux/hid-debug.h>
 
-static int hid_pb_fnmode = 1;
-module_param_named(pb_fnmode, hid_pb_fnmode, int, 0644);
+static int hid_apple_fnmode = 1;
+module_param_named(pb_fnmode, hid_apple_fnmode, int, 0644);
 MODULE_PARM_DESC(pb_fnmode,
-		"Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)");
+		"Mode of fn key on Apple keyboards (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)");
 
 #define unk	KEY_UNKNOWN
 
@@ -99,20 +99,20 @@ struct hidinput_key_translation {
 	u8 flags;
 };
 
-#define POWERBOOK_FLAG_FKEY 0x01
+#define APPLE_FLAG_FKEY 0x01
 
 static struct hidinput_key_translation powerbook_fn_keys[] = {
 	{ KEY_BACKSPACE, KEY_DELETE },
-	{ KEY_F1,       KEY_BRIGHTNESSDOWN,     POWERBOOK_FLAG_FKEY },
-	{ KEY_F2,       KEY_BRIGHTNESSUP,       POWERBOOK_FLAG_FKEY },
-	{ KEY_F3,       KEY_MUTE,               POWERBOOK_FLAG_FKEY },
-	{ KEY_F4,       KEY_VOLUMEDOWN,         POWERBOOK_FLAG_FKEY },
-	{ KEY_F5,       KEY_VOLUMEUP,           POWERBOOK_FLAG_FKEY },
-	{ KEY_F6,       KEY_NUMLOCK,            POWERBOOK_FLAG_FKEY },
-	{ KEY_F7,       KEY_SWITCHVIDEOMODE,    POWERBOOK_FLAG_FKEY },
-	{ KEY_F8,       KEY_KBDILLUMTOGGLE,     POWERBOOK_FLAG_FKEY },
-	{ KEY_F9,       KEY_KBDILLUMDOWN,       POWERBOOK_FLAG_FKEY },
-	{ KEY_F10,      KEY_KBDILLUMUP,         POWERBOOK_FLAG_FKEY },
+	{ KEY_F1,       KEY_BRIGHTNESSDOWN,     APPLE_FLAG_FKEY },
+	{ KEY_F2,       KEY_BRIGHTNESSUP,       APPLE_FLAG_FKEY },
+	{ KEY_F3,       KEY_MUTE,               APPLE_FLAG_FKEY },
+	{ KEY_F4,       KEY_VOLUMEDOWN,         APPLE_FLAG_FKEY },
+	{ KEY_F5,       KEY_VOLUMEUP,           APPLE_FLAG_FKEY },
+	{ KEY_F6,       KEY_NUMLOCK,            APPLE_FLAG_FKEY },
+	{ KEY_F7,       KEY_SWITCHVIDEOMODE,    APPLE_FLAG_FKEY },
+	{ KEY_F8,       KEY_KBDILLUMTOGGLE,     APPLE_FLAG_FKEY },
+	{ KEY_F9,       KEY_KBDILLUMDOWN,       APPLE_FLAG_FKEY },
+	{ KEY_F10,      KEY_KBDILLUMUP,         APPLE_FLAG_FKEY },
 	{ KEY_UP,       KEY_PAGEUP },
 	{ KEY_DOWN,     KEY_PAGEDOWN },
 	{ KEY_LEFT,     KEY_HOME },
@@ -143,7 +143,7 @@ static struct hidinput_key_translation powerbook_numlock_keys[] = {
 	{ }
 };
 
-static struct hidinput_key_translation powerbook_iso_keyboard[] = {
+static struct hidinput_key_translation apple_iso_keyboard[] = {
 	{ KEY_GRAVE,    KEY_102ND },
 	{ KEY_102ND,    KEY_GRAVE },
 	{ }
@@ -161,39 +161,39 @@ static struct hidinput_key_translation *find_translation(struct hidinput_key_tra
 	return NULL;
 }
 
-static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input,
+static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input,
 		struct hid_usage *usage, __s32 value)
 {
 	struct hidinput_key_translation *trans;
 
 	if (usage->code == KEY_FN) {
-		if (value) hid->quirks |=  HID_QUIRK_POWERBOOK_FN_ON;
-		else       hid->quirks &= ~HID_QUIRK_POWERBOOK_FN_ON;
+		if (value) hid->quirks |=  HID_QUIRK_APPLE_FN_ON;
+		else       hid->quirks &= ~HID_QUIRK_APPLE_FN_ON;
 
 		input_event(input, usage->type, usage->code, value);
 
 		return 1;
 	}
 
-	if (hid_pb_fnmode) {
+	if (hid_apple_fnmode) {
 		int do_translate;
 
 		trans = find_translation(powerbook_fn_keys, usage->code);
 		if (trans) {
-			if (test_bit(usage->code, hid->pb_pressed_fn))
+			if (test_bit(usage->code, hid->apple_pressed_fn))
 				do_translate = 1;
-			else if (trans->flags & POWERBOOK_FLAG_FKEY)
+			else if (trans->flags & APPLE_FLAG_FKEY)
 				do_translate =
-					(hid_pb_fnmode == 2 &&  (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) ||
-					(hid_pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON));
+					(hid_apple_fnmode == 2 &&  (hid->quirks & HID_QUIRK_APPLE_FN_ON)) ||
+					(hid_apple_fnmode == 1 && !(hid->quirks & HID_QUIRK_APPLE_FN_ON));
 			else
-				do_translate = (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON);
+				do_translate = (hid->quirks & HID_QUIRK_APPLE_FN_ON);
 
 			if (do_translate) {
 				if (value)
-					set_bit(usage->code, hid->pb_pressed_fn);
+					set_bit(usage->code, hid->apple_pressed_fn);
 				else
-					clear_bit(usage->code, hid->pb_pressed_fn);
+					clear_bit(usage->code, hid->apple_pressed_fn);
 
 				input_event(input, usage->type, trans->to, value);
 
@@ -218,8 +218,8 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input,
 		}
 	}
 
-	if (hid->quirks & HID_QUIRK_POWERBOOK_ISO_KEYBOARD) {
-		trans = find_translation(powerbook_iso_keyboard, usage->code);
+	if (hid->quirks & HID_QUIRK_APPLE_ISO_KEYBOARD) {
+		trans = find_translation(apple_iso_keyboard, usage->code);
 		if (trans) {
 			input_event(input, usage->type, trans->to, value);
 			return 1;
@@ -229,7 +229,7 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input,
 	return 0;
 }
 
-static void hidinput_pb_setup(struct input_dev *input)
+static void hidinput_apple_setup(struct input_dev *input)
 {
 	struct hidinput_key_translation *trans;
 
@@ -242,18 +242,19 @@ static void hidinput_pb_setup(struct input_dev *input)
 	for (trans = powerbook_numlock_keys; trans->from; trans++)
 		set_bit(trans->to, input->keybit);
 
-	for (trans = powerbook_iso_keyboard; trans->from; trans++)
+	for (trans = apple_iso_keyboard; trans->from; trans++)
 		set_bit(trans->to, input->keybit);
 
 }
 #else
-static inline int hidinput_pb_event(struct hid_device *hid, struct input_dev *input,
-		struct hid_usage *usage, __s32 value)
+static inline int hidinput_apple_event(struct hid_device *hid,
+				       struct input_dev *input,
+				       struct hid_usage *usage, __s32 value)
 {
 	return 0;
 }
 
-static inline void hidinput_pb_setup(struct input_dev *input)
+static inline void hidinput_apple_setup(struct input_dev *input)
 {
 }
 #endif
@@ -797,14 +798,14 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 				goto ignore;
 			break;
 
-		case HID_UP_CUSTOM: /* Reported on Logitech and Powerbook USB keyboards */
+		case HID_UP_CUSTOM: /* Reported on Logitech and Apple USB keyboards */
 
 			set_bit(EV_REP, input->evbit);
 			switch(usage->hid & HID_USAGE) {
 				case 0x003:
-					/* The fn key on Apple PowerBooks */
+					/* The fn key on Apple USB keyboards */
 					map_key_clear(KEY_FN);
-					hidinput_pb_setup(input);
+					hidinput_apple_setup(input);
 					break;
 
 				default:    goto ignore;
@@ -989,7 +990,7 @@ void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct
 		return;
 	}
 
-	if ((hid->quirks & HID_QUIRK_POWERBOOK_HAS_FN) && hidinput_pb_event(hid, input, usage, value))
+	if ((hid->quirks & HID_QUIRK_APPLE_HAS_FN) && hidinput_apple_event(hid, input, usage, value))
 		return;
 
 	if (usage->hat_min < usage->hat_max || usage->hat_dir) {
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index a2552856476b..127f24a93b09 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -540,19 +540,19 @@ static const struct hid_blacklist {
 
 	{ USB_VENDOR_ID_WISEGROUP_LTD, USB_DEVICE_ID_SMARTJOY_DUAL_PLUS, HID_QUIRK_NOGET | HID_QUIRK_MULTI_INPUT },
 
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_ANSI, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_ISO, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_ANSI, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_ISO, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_POWERBOOK_ISO_KEYBOARD},
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_JIS, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ANSI, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ISO, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_POWERBOOK_ISO_KEYBOARD},
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_JIS, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_ANSI, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_ISO, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_POWERBOOK_ISO_KEYBOARD},
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_JIS, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
-	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_APPLE_ISO_KEYBOARD},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER_JIS, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_APPLE_ISO_KEYBOARD},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_JIS, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_ANSI, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_ISO, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE | HID_QUIRK_APPLE_ISO_KEYBOARD},
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_JIS, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
+	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY, HID_QUIRK_APPLE_HAS_FN | HID_QUIRK_IGNORE_MOUSE },
 
 	{ USB_VENDOR_ID_DELL, USB_DEVICE_ID_DELL_W7658, HID_QUIRK_RESET_LEDS },
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_KBD, HID_QUIRK_RESET_LEDS },
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 6e35b92b1d2c..833f2af8aab9 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -267,10 +267,10 @@ struct hid_item {
 #define HID_QUIRK_2WHEEL_MOUSE_HACK_5		0x00000100
 #define HID_QUIRK_2WHEEL_MOUSE_HACK_ON		0x00000200
 #define HID_QUIRK_MIGHTYMOUSE			0x00000400
-#define HID_QUIRK_POWERBOOK_HAS_FN		0x00000800
-#define HID_QUIRK_POWERBOOK_FN_ON		0x00001000
+#define HID_QUIRK_APPLE_HAS_FN			0x00000800
+#define HID_QUIRK_APPLE_FN_ON			0x00001000
 #define HID_QUIRK_INVERT_HWHEEL			0x00002000
-#define HID_QUIRK_POWERBOOK_ISO_KEYBOARD        0x00004000
+#define HID_QUIRK_APPLE_ISO_KEYBOARD		0x00004000
 #define HID_QUIRK_BAD_RELATIVE_KEYS		0x00008000
 #define HID_QUIRK_SKIP_OUTPUT_REPORTS		0x00010000
 #define HID_QUIRK_IGNORE_MOUSE			0x00020000
@@ -469,7 +469,7 @@ struct hid_device {							/* device report descriptor */
 	/* handler for raw output data, used by hidraw */
 	int (*hid_output_raw_report) (struct hid_device *, __u8 *, size_t);
 #ifdef CONFIG_USB_HIDINPUT_POWERBOOK
-	unsigned long pb_pressed_fn[BITS_TO_LONGS(KEY_CNT)];
+	unsigned long apple_pressed_fn[BITS_TO_LONGS(KEY_CNT)];
 	unsigned long pb_pressed_numlock[BITS_TO_LONGS(KEY_CNT)];
 #endif
 };
-- 
cgit v1.2.3


From c80e5ffac0579499ca28444155118ffcdd9b8d7e Mon Sep 17 00:00:00 2001
From: Pavel Troller <patrol@sinus.cz>
Date: Mon, 29 Oct 2007 11:13:46 +0100
Subject: HID: Implement horizontal wheel handling for A4 Tech X5-005D

This mouse distinguishes horizontal wheel from vertical by a special "pseudo
event" GenericDesktop.00b8, with values of 0 for vertical and 8 for horizontal
wheel. Because this event is supplied by the parser too late, we need to delay
a wheel event, wait for this one and send either REL_WHEEL or REL_HWHEEL to
input depending on the event value.

Signed-off-by: Pavel Troller <patrol@sinus.cz>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input.c         | 20 +++++++++++++++++---
 drivers/hid/usbhid/hid-quirks.c |  2 ++
 include/linux/hid.h             |  3 +++
 3 files changed, 22 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index 8c4c908177fa..de270b16bde3 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -902,9 +902,10 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 			map_key(BTN_1);
 	}
 
-	if ((device->quirks & (HID_QUIRK_2WHEEL_MOUSE_HACK_7 | HID_QUIRK_2WHEEL_MOUSE_HACK_5)) &&
-		 (usage->type == EV_REL) && (usage->code == REL_WHEEL))
-			set_bit(REL_HWHEEL, bit);
+	if ((device->quirks & (HID_QUIRK_2WHEEL_MOUSE_HACK_7 | HID_QUIRK_2WHEEL_MOUSE_HACK_5 |
+			HID_QUIRK_2WHEEL_MOUSE_HACK_B8)) && (usage->type == EV_REL) &&
+			(usage->code == REL_WHEEL))
+		set_bit(REL_HWHEEL, bit);
 
 	if (((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005))
 		|| ((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007)))
@@ -1002,6 +1003,19 @@ void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct
 		return;
 	}
 
+	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
+			(usage->type == EV_REL) &&
+			(usage->code == REL_WHEEL)) {
+		hid->delayed_value = value;
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
+			(usage->hid == 0x000100b8)) {
+		input_event(input, EV_REL, value ? REL_HWHEEL : REL_WHEEL, hid->delayed_value);
+		return;
+	}
+
 	if ((hid->quirks & HID_QUIRK_INVERT_HWHEEL) && (usage->code == REL_HWHEEL)) {
 		input_event(input, usage->type, usage->code, -value);
 		return;
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 3304b0ab39f7..836b06a86173 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -19,6 +19,7 @@
 
 #define USB_VENDOR_ID_A4TECH		0x09da
 #define USB_DEVICE_ID_A4TECH_WCP32PU	0x0006
+#define USB_DEVICE_ID_A4TECH_X5_005D	0x000a
 
 #define USB_VENDOR_ID_AASHIMA		0x06d6
 #define USB_DEVICE_ID_AASHIMA_GAMEPAD	0x0025
@@ -371,6 +372,7 @@ static const struct hid_blacklist {
 } hid_blacklist[] = {
 
 	{ USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_WCP32PU, HID_QUIRK_2WHEEL_MOUSE_HACK_7 },
+	{ USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_X5_005D, HID_QUIRK_2WHEEL_MOUSE_HACK_B8 },
 	{ USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_MOUSE, HID_QUIRK_2WHEEL_MOUSE_HACK_5 },
 
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_RECEIVER, HID_QUIRK_BAD_RELATIVE_KEYS },
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 833f2af8aab9..991bbcdc1ca7 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -281,6 +281,7 @@ struct hid_item {
 #define HID_QUIRK_LOGITECH_IGNORE_DOUBLED_WHEEL	0x00400000
 #define HID_QUIRK_LOGITECH_EXPANDED_KEYMAP	0x00800000
 #define HID_QUIRK_IGNORE_HIDINPUT		0x01000000
+#define HID_QUIRK_2WHEEL_MOUSE_HACK_B8		0x02000000
 
 /*
  * Separate quirks for runtime report descriptor fixup
@@ -456,6 +457,8 @@ struct hid_device {							/* device report descriptor */
 
 	void *driver_data;
 
+	__s32 delayed_value;						/* For A4 Tech mice hwheel quirk */
+
 	/* device-specific function pointers */
 	int (*hidinput_input_event) (struct input_dev *, unsigned int, unsigned int, int);
 	int (*hid_open) (struct hid_device *);
-- 
cgit v1.2.3


From af9e0eacdc072ba28fd139b90de27023d9cb0598 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Wed, 14 Nov 2007 12:13:26 +0100
Subject: HID: add full support for Genius KB-29E

Genius KB-29E has broken report descriptor, which causes some of the
Consumer usages to appear incorrectly as Button usages. We fix it by
fixing the report descriptor before it is being parsed.

Also a few of the keys violate the HUT standard, so they need a special
handling. They currently fall into "Reserved" range as per HUT 1.12.

Reported-by: Szekeres Istvan <szekeres@iii.hu>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input.c         |  6 ++++++
 drivers/hid/usbhid/hid-quirks.c | 16 ++++++++++++++++
 include/linux/hid.h             |  1 +
 3 files changed, 23 insertions(+)

(limited to 'include')

diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index de270b16bde3..0da29cf43713 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -630,6 +630,12 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 				case 0x0f6: map_key_clear(KEY_NEXT);		break;
 				case 0x0fa: map_key_clear(KEY_BACK);		break;
 
+				/* reserved in HUT 1.12. Reported on Genius KB29E */
+				case 0x156: map_key_clear(KEY_WORDPROCESSOR);	break;
+				case 0x157: map_key_clear(KEY_SPREADSHEET);	break;
+				case 0x158: map_key_clear(KEY_PRESENTATION);	break;
+				case 0x15c: map_key_clear(KEY_STOP);		break;
+
 				case 0x182: map_key_clear(KEY_BOOKMARKS);	break;
 				case 0x183: map_key_clear(KEY_CONFIG);		break;
 				case 0x184: map_key_clear(KEY_WORDPROCESSOR);	break;
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 836b06a86173..0ab18a401b77 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -301,6 +301,9 @@
 #define USB_VENDOR_ID_MICROSOFT		0x045e
 #define USB_DEVICE_ID_SIDEWINDER_GV	0x003b
 
+#define USB_VENDOR_ID_MONTEREY		0x0566
+#define USB_DEVICE_ID_GENIUS_KB29E	0x3004
+
 #define USB_VENDOR_ID_NCR		0x0404
 #define USB_DEVICE_ID_NCR_FIRST		0x0300
 #define USB_DEVICE_ID_NCR_LAST		0x03ff
@@ -646,6 +649,8 @@ static const struct hid_rdesc_blacklist {
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_S510_RECEIVER, HID_QUIRK_RDESC_LOGITECH },
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_S510_RECEIVER_2, HID_QUIRK_RDESC_LOGITECH },
 
+	{ USB_VENDOR_ID_MONTEREY, USB_DEVICE_ID_GENIUS_KB29E, HID_QUIRK_RDESC_BUTTON_CONSUMER },
+
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER4_JIS, HID_QUIRK_RDESC_MACBOOK_JIS },
 
 	{ USB_VENDOR_ID_PETALYNX, USB_DEVICE_ID_PETALYNX_MAXTER_REMOTE, HID_QUIRK_RDESC_PETALYNX },
@@ -973,6 +978,14 @@ static void usbhid_fixup_macbook_descriptor(unsigned char *rdesc, int rsize)
 	}
 }
 
+static void usbhid_fixup_button_consumer_descriptor(unsigned char *rdesc, int rsize)
+{
+	if (rsize >= 30 && rdesc[29] == 0x05
+			&& rdesc[30] == 0x09) {
+		printk(KERN_INFO "Fixing up button/consumer in HID report descriptor\n");
+		rdesc[30] = 0x0c;
+	}
+}
 
 static void __usbhid_fixup_report_descriptor(__u32 quirks, char *rdesc, unsigned rsize)
 {
@@ -990,6 +1003,9 @@ static void __usbhid_fixup_report_descriptor(__u32 quirks, char *rdesc, unsigned
 
 	if (quirks & HID_QUIRK_RDESC_MACBOOK_JIS)
 		usbhid_fixup_macbook_descriptor(rdesc, rsize);
+
+	if (quirks & HID_QUIRK_RDESC_BUTTON_CONSUMER)
+		usbhid_fixup_button_consumer_descriptor(rdesc, rsize);
 }
 
 /**
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 991bbcdc1ca7..c67eeb516040 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -292,6 +292,7 @@ struct hid_item {
 #define HID_QUIRK_RDESC_SWAPPED_MIN_MAX		0x00000004
 #define HID_QUIRK_RDESC_PETALYNX		0x00000008
 #define HID_QUIRK_RDESC_MACBOOK_JIS		0x00000010
+#define HID_QUIRK_RDESC_BUTTON_CONSUMER		0x00000020
 
 /*
  * This is the global environment of the parser. This information is
-- 
cgit v1.2.3


From 10bd065facb2594bd508597ef464d401b212f379 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Thu, 22 Nov 2007 15:18:18 +0100
Subject: HID: refactor mapping to input subsystem for quirky devices

Currently, the handling of mapping between hid and input for devices
that don't conform to HUT 1.12 specification is very messy -- no per-device
handling, no blacklists, conditions on idVendor and idProduct placed
all over the code.

This patch moves all the device-specific input mapping to a separate
file, and introduces a blacklist-style handling for non-standard
device-specific mappings.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/Makefile           |   2 +-
 drivers/hid/hid-input-quirks.c | 326 +++++++++++++++++++++++++++++++++++++++++
 drivers/hid/hid-input.c        | 163 ++-------------------
 include/linux/hid.h            |   1 +
 4 files changed, 337 insertions(+), 155 deletions(-)
 create mode 100644 drivers/hid/hid-input-quirks.c

(limited to 'include')

diff --git a/drivers/hid/Makefile b/drivers/hid/Makefile
index 1ac5103f7c93..275dc522c738 100644
--- a/drivers/hid/Makefile
+++ b/drivers/hid/Makefile
@@ -1,7 +1,7 @@
 #
 # Makefile for the HID driver
 #
-hid-objs			:= hid-core.o hid-input.o
+hid-objs			:= hid-core.o hid-input.o hid-input-quirks.o
 
 obj-$(CONFIG_HID)		+= hid.o
 
diff --git a/drivers/hid/hid-input-quirks.c b/drivers/hid/hid-input-quirks.c
new file mode 100644
index 000000000000..e05e9ad5522e
--- /dev/null
+++ b/drivers/hid/hid-input-quirks.c
@@ -0,0 +1,326 @@
+/*
+ *  HID-input usage mapping quirks
+ *
+ *  This is used to handle HID-input mappings for devices violating
+ *  HUT 1.12 specification.
+ *
+ * Copyright (c) 2007 Jiri Kosina
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License
+ */
+
+#include <linux/input.h>
+#include <linux/hid.h>
+
+#define map_abs(c)      do { usage->code = c; usage->type = EV_ABS; bit = input->absbit; *max = ABS_MAX; } while (0)
+#define map_rel(c)      do { usage->code = c; usage->type = EV_REL; bit = input->relbit; *max = REL_MAX; } while (0)
+#define map_key(c)      do { usage->code = c; usage->type = EV_KEY; bit = input->keybit; *max = KEY_MAX; } while (0)
+#define map_led(c)      do { usage->code = c; usage->type = EV_LED; bit = input->ledbit; *max = LED_MAX; } while (0)
+
+#define map_abs_clear(c)        do { map_abs(c); clear_bit(c, bit); } while (0)
+#define map_key_clear(c)        do { map_key(c); clear_bit(c, bit); } while (0)
+
+static int quirk_belkin_wkbd(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
+		return 0;
+
+	switch (usage->hid & HID_USAGE) {
+		case 0x03a: map_key_clear(KEY_SOUND);		break;
+		case 0x03b: map_key_clear(KEY_CAMERA);		break;
+		case 0x03c: map_key_clear(KEY_DOCUMENTS);	break;
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_cherry_cymotion(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
+		return 0;
+
+	switch (usage->hid & HID_USAGE) {
+		case 0x301: map_key_clear(KEY_PROG1);		break;
+		case 0x302: map_key_clear(KEY_PROG2);		break;
+		case 0x303: map_key_clear(KEY_PROG3);		break;
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_logitech_ultrax_remote(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_LOGIVENDOR)
+		return 0;
+
+	set_bit(EV_REP, input->evbit);
+	switch(usage->hid & HID_USAGE) {
+		/* Reported on Logitech Ultra X Media Remote */
+		case 0x004: map_key_clear(KEY_AGAIN);		break;
+		case 0x00d: map_key_clear(KEY_HOME);		break;
+		case 0x024: map_key_clear(KEY_SHUFFLE);		break;
+		case 0x025: map_key_clear(KEY_TV);		break;
+		case 0x026: map_key_clear(KEY_MENU);		break;
+		case 0x031: map_key_clear(KEY_AUDIO);		break;
+		case 0x032: map_key_clear(KEY_TEXT);		break;
+		case 0x033: map_key_clear(KEY_LAST);		break;
+		case 0x047: map_key_clear(KEY_MP3);		break;
+		case 0x048: map_key_clear(KEY_DVD);		break;
+		case 0x049: map_key_clear(KEY_MEDIA);		break;
+		case 0x04a: map_key_clear(KEY_VIDEO);		break;
+		case 0x04b: map_key_clear(KEY_ANGLE);		break;
+		case 0x04c: map_key_clear(KEY_LANGUAGE);	break;
+		case 0x04d: map_key_clear(KEY_SUBTITLE);	break;
+		case 0x051: map_key_clear(KEY_RED);		break;
+		case 0x052: map_key_clear(KEY_CLOSE);		break;
+
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_chicony_tactical_pad(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
+		return 0;
+
+	set_bit(EV_REP, input->evbit);
+	switch (usage->hid & HID_USAGE) {
+		case 0xff01: map_key_clear(BTN_1);		break;
+		case 0xff02: map_key_clear(BTN_2);		break;
+		case 0xff03: map_key_clear(BTN_3);		break;
+		case 0xff04: map_key_clear(BTN_4);		break;
+		case 0xff05: map_key_clear(BTN_5);		break;
+		case 0xff06: map_key_clear(BTN_6);		break;
+		case 0xff07: map_key_clear(BTN_7);		break;
+		case 0xff08: map_key_clear(BTN_8);		break;
+		case 0xff09: map_key_clear(BTN_9);		break;
+		case 0xff0a: map_key_clear(BTN_A);		break;
+		case 0xff0b: map_key_clear(BTN_B);		break;
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_microsoft_ergonomy_kb(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
+		return 0;
+
+	switch(usage->hid & HID_USAGE) {
+		case 0xfd06: map_key_clear(KEY_CHAT);		break;
+		case 0xfd07: map_key_clear(KEY_PHONE);		break;
+		case 0xff05:
+			set_bit(EV_REP, input->evbit);
+			map_key_clear(KEY_F13);
+			set_bit(KEY_F14, input->keybit);
+			set_bit(KEY_F15, input->keybit);
+			set_bit(KEY_F16, input->keybit);
+			set_bit(KEY_F17, input->keybit);
+			set_bit(KEY_F18, input->keybit);
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_microsoft_presenter_8k(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
+		return 0;
+
+	set_bit(EV_REP, input->evbit);
+	switch(usage->hid & HID_USAGE) {
+		case 0xfd08: map_key_clear(KEY_RIGHT);		break;
+		case 0xfd09: map_key_clear(KEY_LEFT);		break;
+		case 0xfd0b: map_key_clear(KEY_PAUSE);		break;
+		case 0xfd0f: map_key_clear(KEY_F5);		break;
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_petalynx_remote(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if (((usage->hid & HID_USAGE_PAGE) != HID_UP_LOGIVENDOR) &&
+			((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER))
+		return 0;
+
+	if ((usage->hid & HID_USAGE_PAGE) == HID_UP_LOGIVENDOR)
+		switch(usage->hid & HID_USAGE) {
+			case 0x05a: map_key_clear(KEY_TEXT);		break;
+			case 0x05b: map_key_clear(KEY_RED);		break;
+			case 0x05c: map_key_clear(KEY_GREEN);		break;
+			case 0x05d: map_key_clear(KEY_YELLOW);		break;
+			case 0x05e: map_key_clear(KEY_BLUE);		break;
+			default:
+				return 0;
+		}
+
+	if ((usage->hid & HID_USAGE_PAGE) == HID_UP_CONSUMER)
+		switch(usage->hid & HID_USAGE) {
+			case 0x0f6: map_key_clear(KEY_NEXT);            break;
+			case 0x0fa: map_key_clear(KEY_BACK);            break;
+			default:
+				return 0;
+		}
+	return 1;
+}
+
+static int quirk_logitech_wireless(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
+		return 0;
+
+	switch (usage->hid & HID_USAGE) {
+		case 0x1001: map_key_clear(KEY_MESSENGER);	break;
+		case 0x1003: map_key_clear(KEY_SOUND);		break;
+		case 0x1004: map_key_clear(KEY_VIDEO);		break;
+		case 0x1005: map_key_clear(KEY_AUDIO);		break;
+		case 0x100a: map_key_clear(KEY_DOCUMENTS);	break;
+		case 0x1011: map_key_clear(KEY_PREVIOUSSONG);	break;
+		case 0x1012: map_key_clear(KEY_NEXTSONG);	break;
+		case 0x1013: map_key_clear(KEY_CAMERA);		break;
+		case 0x1014: map_key_clear(KEY_MESSENGER);	break;
+		case 0x1015: map_key_clear(KEY_RECORD);		break;
+		case 0x1016: map_key_clear(KEY_PLAYER);		break;
+		case 0x1017: map_key_clear(KEY_EJECTCD);	break;
+		case 0x1018: map_key_clear(KEY_MEDIA);		break;
+		case 0x1019: map_key_clear(KEY_PROG1);		break;
+		case 0x101a: map_key_clear(KEY_PROG2);		break;
+		case 0x101b: map_key_clear(KEY_PROG3);		break;
+		case 0x101f: map_key_clear(KEY_ZOOMIN);		break;
+		case 0x1020: map_key_clear(KEY_ZOOMOUT);	break;
+		case 0x1021: map_key_clear(KEY_ZOOMRESET);	break;
+		case 0x1023: map_key_clear(KEY_CLOSE);		break;
+		case 0x1027: map_key_clear(KEY_MENU);		break;
+		/* this one is marked as 'Rotate' */
+		case 0x1028: map_key_clear(KEY_ANGLE);		break;
+		case 0x1029: map_key_clear(KEY_SHUFFLE);	break;
+		case 0x102a: map_key_clear(KEY_BACK);		break;
+		case 0x102b: map_key_clear(KEY_CYCLEWINDOWS);	break;
+		case 0x1041: map_key_clear(KEY_BATTERY);	break;
+		case 0x1042: map_key_clear(KEY_WORDPROCESSOR);	break;
+		case 0x1043: map_key_clear(KEY_SPREADSHEET);	break;
+		case 0x1044: map_key_clear(KEY_PRESENTATION);	break;
+		case 0x1045: map_key_clear(KEY_UNDO);		break;
+		case 0x1046: map_key_clear(KEY_REDO);		break;
+		case 0x1047: map_key_clear(KEY_PRINT);		break;
+		case 0x1048: map_key_clear(KEY_SAVE);		break;
+		case 0x1049: map_key_clear(KEY_PROG1);		break;
+		case 0x104a: map_key_clear(KEY_PROG2);		break;
+		case 0x104b: map_key_clear(KEY_PROG3);		break;
+		case 0x104c: map_key_clear(KEY_PROG4);		break;
+
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+static int quirk_cherry_genius_29e(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
+		return 0;
+
+	switch (usage->hid & HID_USAGE) {
+		case 0x156: map_key_clear(KEY_WORDPROCESSOR);	break;
+		case 0x157: map_key_clear(KEY_SPREADSHEET);	break;
+		case 0x158: map_key_clear(KEY_PRESENTATION);	break;
+		case 0x15c: map_key_clear(KEY_STOP);		break;
+
+		default:
+			return 0;
+	}
+	return 1;
+}
+
+
+#define VENDOR_ID_BELKIN			0x1020
+#define DEVICE_ID_BELKIN_WIRELESS_KEYBOARD	0x0006
+
+#define VENDOR_ID_CHERRY			0x046a
+#define DEVICE_ID_CHERRY_CYMOTION		0x0023
+
+#define VENDOR_ID_CHICONY			0x04f2
+#define DEVICE_ID_CHICONY_TACTICAL_PAD		0x0418
+
+#define VENDOR_ID_LOGITECH			0x046d
+#define DEVICE_ID_LOGITECH_RECEIVER		0xc101
+#define DEVICE_ID_S510_RECEIVER			0xc50c
+#define DEVICE_ID_S510_RECEIVER_2		0xc517
+#define DEVICE_ID_MX3000_RECEIVER		0xc513
+
+#define VENDOR_ID_MICROSOFT			0x045e
+#define DEVICE_ID_MS4K				0x00db
+#define DEVICE_ID_MS6K				0x00f9
+#define DEVICE_ID_MS_PRESENTER_8K		0x0713
+
+#define VENDOR_ID_MONTEREY			0x0566
+#define DEVICE_ID_GENIUS_KB29E			0x3004
+
+#define VENDOR_ID_PETALYNX			0x18b1
+#define DEVICE_ID_PETALYNX_MAXTER_REMOTE	0x0037
+
+static const struct hid_input_blacklist {
+	__u16 idVendor;
+	__u16 idProduct;
+	int (*quirk)(struct hid_usage *, struct input_dev *, unsigned long *, int *);
+} hid_input_blacklist[] = {
+	{ VENDOR_ID_BELKIN, DEVICE_ID_BELKIN_WIRELESS_KEYBOARD, quirk_belkin_wkbd },
+
+	{ VENDOR_ID_CHERRY, DEVICE_ID_CHERRY_CYMOTION, quirk_cherry_cymotion },
+
+	{ VENDOR_ID_CHICONY, DEVICE_ID_CHICONY_TACTICAL_PAD, quirk_chicony_tactical_pad },
+
+	{ VENDOR_ID_LOGITECH, DEVICE_ID_LOGITECH_RECEIVER, quirk_logitech_ultrax_remote },
+	{ VENDOR_ID_LOGITECH, DEVICE_ID_S510_RECEIVER, quirk_logitech_wireless },
+	{ VENDOR_ID_LOGITECH, DEVICE_ID_S510_RECEIVER_2, quirk_logitech_wireless },
+	{ VENDOR_ID_LOGITECH, DEVICE_ID_MX3000_RECEIVER, quirk_logitech_wireless },
+
+	{ VENDOR_ID_MICROSOFT, DEVICE_ID_MS4K, quirk_microsoft_ergonomy_kb },
+	{ VENDOR_ID_MICROSOFT, DEVICE_ID_MS6K, quirk_microsoft_ergonomy_kb },
+	{ VENDOR_ID_MICROSOFT, DEVICE_ID_MS_PRESENTER_8K, quirk_microsoft_presenter_8k },
+
+	{ VENDOR_ID_MONTEREY, DEVICE_ID_GENIUS_KB29E, quirk_cherry_genius_29e },
+
+	{ VENDOR_ID_PETALYNX, DEVICE_ID_PETALYNX_MAXTER_REMOTE, quirk_petalynx_remote },
+	
+	{ 0, 0, 0 }
+};
+
+int hidinput_mapping_quirks(struct hid_usage *usage, 
+				   struct input_dev *input, 
+				   unsigned long *bit, int *max)
+{
+	struct hid_device *device = input_get_drvdata(input);
+	int i = 0;
+	
+	while (hid_input_blacklist[i].quirk) {
+		if (hid_input_blacklist[i].idVendor == device->vendor &&
+				hid_input_blacklist[i].idProduct == device->product)
+			return hid_input_blacklist[i].quirk(usage, input, bit, max);
+		i++;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hidinput_mapping_quirks);
+
diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index 0da29cf43713..3d448037c82f 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -367,7 +367,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 {
 	struct input_dev *input = hidinput->input;
 	struct hid_device *device = input_get_drvdata(input);
-	int max = 0, code;
+	int max = 0, code, ret;
 	unsigned long *bit = NULL;
 
 	field->hidinput = hidinput;
@@ -386,6 +386,10 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 		goto ignore;
 	}
 
+	ret = hidinput_mapping_quirks(usage, input, bit, &max);
+	if (ret)
+		goto mapped;
+
 	switch (usage->hid & HID_USAGE_PAGE) {
 
 		case HID_UP_UNDEFINED:
@@ -573,14 +577,6 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 				case 0x000: goto ignore;
 				case 0x034: map_key_clear(KEY_SLEEP);		break;
 				case 0x036: map_key_clear(BTN_MISC);		break;
-				/*
-				 * The next three are reported by Belkin wireless
-				 * keyboard (1020:0006). These values are "reserved"
-				 * in HUT 1.12.
-				 */
-				case 0x03a: map_key_clear(KEY_SOUND);           break;
-				case 0x03b: map_key_clear(KEY_CAMERA);          break;
-				case 0x03c: map_key_clear(KEY_DOCUMENTS);       break;
 
 				case 0x040: map_key_clear(KEY_MENU);		break;
 				case 0x045: map_key_clear(KEY_RADIO);		break;
@@ -626,16 +622,6 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 				case 0x0e9: map_key_clear(KEY_VOLUMEUP);	break;
 				case 0x0ea: map_key_clear(KEY_VOLUMEDOWN);	break;
 
-				/* reserved in HUT 1.12. Reported on Petalynx remote */
-				case 0x0f6: map_key_clear(KEY_NEXT);		break;
-				case 0x0fa: map_key_clear(KEY_BACK);		break;
-
-				/* reserved in HUT 1.12. Reported on Genius KB29E */
-				case 0x156: map_key_clear(KEY_WORDPROCESSOR);	break;
-				case 0x157: map_key_clear(KEY_SPREADSHEET);	break;
-				case 0x158: map_key_clear(KEY_PRESENTATION);	break;
-				case 0x15c: map_key_clear(KEY_STOP);		break;
-
 				case 0x182: map_key_clear(KEY_BOOKMARKS);	break;
 				case 0x183: map_key_clear(KEY_CONFIG);		break;
 				case 0x184: map_key_clear(KEY_WORDPROCESSOR);	break;
@@ -695,51 +681,6 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 				case 0x28b: map_key_clear(KEY_FORWARDMAIL);	break;
 				case 0x28c: map_key_clear(KEY_SEND);		break;
 
-				/* Reported on a Cherry Cymotion keyboard */
-				case 0x301: map_key_clear(KEY_PROG1);		break;
-				case 0x302: map_key_clear(KEY_PROG2);		break;
-				case 0x303: map_key_clear(KEY_PROG3);		break;
-
-				/* Reported on certain Logitech wireless keyboards */
-				case 0x1001: map_key_clear(KEY_MESSENGER);	break;
-				case 0x1003: map_key_clear(KEY_SOUND);		break;
-				case 0x1004: map_key_clear(KEY_VIDEO);		break;
-				case 0x1005: map_key_clear(KEY_AUDIO);		break;
-				case 0x100a: map_key_clear(KEY_DOCUMENTS);	break;
-				case 0x1011: map_key_clear(KEY_PREVIOUSSONG);	break;
-				case 0x1012: map_key_clear(KEY_NEXTSONG);	break;
-				case 0x1013: map_key_clear(KEY_CAMERA);		break;
-				case 0x1014: map_key_clear(KEY_MESSENGER);	break;
-				case 0x1015: map_key_clear(KEY_RECORD);		break;
-				case 0x1016: map_key_clear(KEY_PLAYER);		break;
-				case 0x1017: map_key_clear(KEY_EJECTCD);	break;
-				case 0x1018: map_key_clear(KEY_MEDIA);          break;
-				case 0x1019: map_key_clear(KEY_PROG1);		break;
-				case 0x101a: map_key_clear(KEY_PROG2);		break;
-				case 0x101b: map_key_clear(KEY_PROG3);		break;
-				case 0x101f: map_key_clear(KEY_ZOOMIN);		break;
-				case 0x1020: map_key_clear(KEY_ZOOMOUT);	break;
-				case 0x1021: map_key_clear(KEY_ZOOMRESET);	break;
-				case 0x1023: map_key_clear(KEY_CLOSE);		break;
-				case 0x1027: map_key_clear(KEY_MENU);           break;
-				/* this one is marked as 'Rotate' */
-				case 0x1028: map_key_clear(KEY_ANGLE);		break;
-				case 0x1029: map_key_clear(KEY_SHUFFLE);	break;
-				case 0x102a: map_key_clear(KEY_BACK);           break;
-				case 0x102b: map_key_clear(KEY_CYCLEWINDOWS);   break;
-				case 0x1041: map_key_clear(KEY_BATTERY);	break;
-				case 0x1042: map_key_clear(KEY_WORDPROCESSOR);	break;
-				case 0x1043: map_key_clear(KEY_SPREADSHEET);	break;
-				case 0x1044: map_key_clear(KEY_PRESENTATION);	break;
-				case 0x1045: map_key_clear(KEY_UNDO);		break;
-				case 0x1046: map_key_clear(KEY_REDO);		break;
-				case 0x1047: map_key_clear(KEY_PRINT);		break;
-				case 0x1048: map_key_clear(KEY_SAVE);		break;
-				case 0x1049: map_key_clear(KEY_PROG1);		break;
-				case 0x104a: map_key_clear(KEY_PROG2);		break;
-				case 0x104b: map_key_clear(KEY_PROG3);		break;
-				case 0x104c: map_key_clear(KEY_PROG4);		break;
-
 				default:    goto ignore;
 			}
 			break;
@@ -766,65 +707,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 
 		case HID_UP_MSVENDOR:
 
-			/* Unfortunately, there are multiple devices which
-			 * emit usages from MSVENDOR page that require different
-			 * handling. If this list grows too much in the future,
-			 * more general handling will have to be introduced here
-			 * (i.e. another blacklist).
-			 */
-
-			/* Chicony Chicony KU-0418 tactical pad */
-			if (IS_CHICONY_TACTICAL_PAD(device)) {
-				set_bit(EV_REP, input->evbit);
-				switch(usage->hid & HID_USAGE) {
-					case 0xff01: map_key_clear(BTN_1);		break;
-					case 0xff02: map_key_clear(BTN_2);		break;
-					case 0xff03: map_key_clear(BTN_3);		break;
-					case 0xff04: map_key_clear(BTN_4);		break;
-					case 0xff05: map_key_clear(BTN_5);		break;
-					case 0xff06: map_key_clear(BTN_6);		break;
-					case 0xff07: map_key_clear(BTN_7);		break;
-					case 0xff08: map_key_clear(BTN_8);		break;
-					case 0xff09: map_key_clear(BTN_9);		break;
-					case 0xff0a: map_key_clear(BTN_A);		break;
-					case 0xff0b: map_key_clear(BTN_B);		break;
-					default:    goto ignore;
-				}
-
-			/* Microsoft Natural Ergonomic Keyboard 4000 */
-			} else if (IS_MS_KB(device)) {
-				switch(usage->hid & HID_USAGE) {
-					case 0xfd06:
-						map_key_clear(KEY_CHAT);
-						break;
-					case 0xfd07:
-						map_key_clear(KEY_PHONE);
-						break;
-					case 0xff05:
-						set_bit(EV_REP, input->evbit);
-						map_key_clear(KEY_F13);
-						set_bit(KEY_F14, input->keybit);
-						set_bit(KEY_F15, input->keybit);
-						set_bit(KEY_F16, input->keybit);
-						set_bit(KEY_F17, input->keybit);
-						set_bit(KEY_F18, input->keybit);
-					default:	goto ignore;
-				}
-
-			/* Microsoft Wireless Notebook Presenter Mouse 8000 */
-			} else if (IS_MS_PRESENTER_8000(device)) {
-				set_bit(EV_REP, input->evbit);
-				switch(usage->hid & HID_USAGE) {
-					/* Useful mappings of bottom-side keys for presentations */
-					case 0xfd08: map_key_clear(KEY_RIGHT);		break;
-					case 0xfd09: map_key_clear(KEY_LEFT);		break;
-					case 0xfd0b: map_key_clear(KEY_PAUSE);		break;
-					case 0xfd0f: map_key_clear(KEY_F5);		break;
-					default:    goto ignore;
-				}
-			} else
-				goto ignore;
-			break;
+			goto ignore;
 
 		case HID_UP_CUSTOM: /* Reported on Logitech and Apple USB keyboards */
 
@@ -841,38 +724,9 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 			break;
 
 		case HID_UP_LOGIVENDOR:
-			set_bit(EV_REP, input->evbit);
-			switch(usage->hid & HID_USAGE) {
-				/* Reported on Logitech Ultra X Media Remote */
-				case 0x004: map_key_clear(KEY_AGAIN);		break;
-				case 0x00d: map_key_clear(KEY_HOME);		break;
-				case 0x024: map_key_clear(KEY_SHUFFLE);		break;
-				case 0x025: map_key_clear(KEY_TV);		break;
-				case 0x026: map_key_clear(KEY_MENU);		break;
-				case 0x031: map_key_clear(KEY_AUDIO);		break;
-				case 0x032: map_key_clear(KEY_TEXT);		break;
-				case 0x033: map_key_clear(KEY_LAST);		break;
-				case 0x047: map_key_clear(KEY_MP3);		break;
-				case 0x048: map_key_clear(KEY_DVD);		break;
-				case 0x049: map_key_clear(KEY_MEDIA);		break;
-				case 0x04a: map_key_clear(KEY_VIDEO);		break;
-				case 0x04b: map_key_clear(KEY_ANGLE);		break;
-				case 0x04c: map_key_clear(KEY_LANGUAGE);	break;
-				case 0x04d: map_key_clear(KEY_SUBTITLE);	break;
-				case 0x051: map_key_clear(KEY_RED);		break;
-				case 0x052: map_key_clear(KEY_CLOSE);		break;
-
-				/* Reported on Petalynx Maxter remote */
-				case 0x05a: map_key_clear(KEY_TEXT);		break;
-				case 0x05b: map_key_clear(KEY_RED);		break;
-				case 0x05c: map_key_clear(KEY_GREEN);		break;
-				case 0x05d: map_key_clear(KEY_YELLOW);		break;
-				case 0x05e: map_key_clear(KEY_BLUE);		break;
-
-				default:    goto ignore;
-			}
-			break;
 
+			goto ignore;
+		
 		case HID_UP_PID:
 
 			switch(usage->hid & HID_USAGE) {
@@ -899,6 +753,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 			break;
 	}
 
+mapped:
 	if (device->quirks & HID_QUIRK_MIGHTYMOUSE) {
 		if (usage->hid == HID_GD_Z)
 			map_rel(REL_HWHEEL);
diff --git a/include/linux/hid.h b/include/linux/hid.h
index c67eeb516040..cd5d562b1b7c 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -524,6 +524,7 @@ extern void hidinput_disconnect(struct hid_device *);
 int hid_set_field(struct hid_field *, unsigned, __s32);
 int hid_input_report(struct hid_device *, int type, u8 *, int, int);
 int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field);
+int hidinput_mapping_quirks(struct hid_usage *, struct input_dev *, unsigned long *, int *);
 void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt);
 void hid_output_report(struct hid_report *report, __u8 *data);
 void hid_free_device(struct hid_device *device);
-- 
cgit v1.2.3


From 87bc2aa9933afc032a93490e1642918121e7470b Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Fri, 23 Nov 2007 13:16:02 +0100
Subject: HID: separate hid-input event quirks from generic code

This patch separates also the hid-input quirks that have to be
applied at the time the event occurs, so that the generic code
handling HUT-compliant devices is not messed up by them too much.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input-quirks.c | 65 +++++++++++++++++++++++++++++++++++++++++-
 drivers/hid/hid-input.c        | 63 ++++------------------------------------
 include/linux/hid.h            |  2 ++
 3 files changed, 71 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/drivers/hid/hid-input-quirks.c b/drivers/hid/hid-input-quirks.c
index e05e9ad5522e..7f2f80b00d74 100644
--- a/drivers/hid/hid-input-quirks.c
+++ b/drivers/hid/hid-input-quirks.c
@@ -322,5 +322,68 @@ int hidinput_mapping_quirks(struct hid_usage *usage,
 	}
 	return 0;
 }
-EXPORT_SYMBOL_GPL(hidinput_mapping_quirks);
+
+#define IS_MS_KB(x) (x->vendor == 0x045e && (x->product == 0x00db || x->product == 0x00f9))
+
+void hidinput_event_quirks(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value)
+{
+	struct input_dev *input;
+	unsigned *quirks = &hid->quirks;
+
+	input = field->hidinput->input;
+
+	if (((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005))
+		|| ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) {
+		if (value) hid->quirks |=  HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
+		else       hid->quirks &= ~HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
+			(usage->type == EV_REL) &&
+			(usage->code == REL_WHEEL)) {
+		hid->delayed_value = value;
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
+			(usage->hid == 0x000100b8)) {
+		input_event(input, EV_REL, value ? REL_HWHEEL : REL_WHEEL, hid->delayed_value);
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_INVERT_HWHEEL) && (usage->code == REL_HWHEEL)) {
+		input_event(input, usage->type, usage->code, -value);
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_ON) && (usage->code == REL_WHEEL)) {
+		input_event(input, usage->type, REL_HWHEEL, value);
+		return;
+	}
+
+	if ((hid->quirks & HID_QUIRK_APPLE_HAS_FN) && hidinput_apple_event(hid, input, usage, value))
+		return;
+
+	/* Handling MS keyboards special buttons */
+	if (IS_MS_KB(hid) && usage->hid == (HID_UP_MSVENDOR | 0xff05)) {
+		int key = 0;
+		static int last_key = 0;
+		switch (value) {
+			case 0x01: key = KEY_F14; break;
+			case 0x02: key = KEY_F15; break;
+			case 0x04: key = KEY_F16; break;
+			case 0x08: key = KEY_F17; break;
+			case 0x10: key = KEY_F18; break;
+			default: break;
+		}
+		if (key) {
+			input_event(input, usage->type, key, 1);
+			last_key = key;
+		} else {
+			input_event(input, usage->type, last_key, 0);
+		}
+	}
+}
+
 
diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index 3d448037c82f..aeb018e31bfc 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -86,11 +86,6 @@ static const struct {
 #define map_abs_clear(c)	do { map_abs(c); clear_bit(c, bit); } while (0)
 #define map_key_clear(c)	do { map_key(c); clear_bit(c, bit); } while (0)
 
-/* hardware needing special handling due to colliding MSVENDOR page usages */
-#define IS_CHICONY_TACTICAL_PAD(x) (x->vendor == 0x04f2 && device->product == 0x0418)
-#define IS_MS_KB(x) (x->vendor == 0x045e && (x->product == 0x00db || x->product == 0x00f9))
-#define IS_MS_PRESENTER_8000(x) (x->vendor == 0x045e && x->product == 0x0713)
-
 #ifdef CONFIG_USB_HIDINPUT_POWERBOOK
 
 struct hidinput_key_translation {
@@ -177,7 +172,7 @@ static struct hidinput_key_translation *find_translation(struct hidinput_key_tra
 	return NULL;
 }
 
-static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input,
+int hidinput_apple_event(struct hid_device *hid, struct input_dev *input,
 		struct hid_usage *usage, __s32 value)
 {
 	struct hidinput_key_translation *trans;
@@ -269,7 +264,7 @@ static void hidinput_apple_setup(struct input_dev *input)
 
 }
 #else
-static inline int hidinput_apple_event(struct hid_device *hid,
+inline int hidinput_apple_event(struct hid_device *hid,
 				       struct input_dev *input,
 				       struct hid_usage *usage, __s32 value)
 {
@@ -386,6 +381,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 		goto ignore;
 	}
 
+	/* handle input mappings for quirky devices */
 	ret = hidinput_mapping_quirks(usage, input, bit, &max);
 	if (ret)
 		goto mapped;
@@ -857,38 +853,8 @@ void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct
 	if (!usage->type)
 		return;
 
-	if (((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005))
-		|| ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) {
-		if (value) hid->quirks |=  HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
-		else       hid->quirks &= ~HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
-		return;
-	}
-
-	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
-			(usage->type == EV_REL) &&
-			(usage->code == REL_WHEEL)) {
-		hid->delayed_value = value;
-		return;
-	}
-
-	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_B8) &&
-			(usage->hid == 0x000100b8)) {
-		input_event(input, EV_REL, value ? REL_HWHEEL : REL_WHEEL, hid->delayed_value);
-		return;
-	}
-
-	if ((hid->quirks & HID_QUIRK_INVERT_HWHEEL) && (usage->code == REL_HWHEEL)) {
-		input_event(input, usage->type, usage->code, -value);
-		return;
-	}
-
-	if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_ON) && (usage->code == REL_WHEEL)) {
-		input_event(input, usage->type, REL_HWHEEL, value);
-		return;
-	}
-
-	if ((hid->quirks & HID_QUIRK_APPLE_HAS_FN) && hidinput_apple_event(hid, input, usage, value))
-		return;
+	/* handle input events for quirky devices */
+	hidinput_event_quirks(hid, field, usage, value);
 
 	if (usage->hat_min < usage->hat_max || usage->hat_dir) {
 		int hat_dir = usage->hat_dir;
@@ -949,25 +915,6 @@ void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct
 		return;
 	}
 
-	/* Handling MS keyboards special buttons */
-	if (IS_MS_KB(hid) && usage->hid == (HID_UP_MSVENDOR | 0xff05)) {
-		int key = 0;
-		static int last_key = 0;
-		switch (value) {
-			case 0x01: key = KEY_F14; break;
-			case 0x02: key = KEY_F15; break;
-			case 0x04: key = KEY_F16; break;
-			case 0x08: key = KEY_F17; break;
-			case 0x10: key = KEY_F18; break;
-			default: break;
-		}
-		if (key) {
-			input_event(input, usage->type, key, 1);
-			last_key = key;
-		} else {
-			input_event(input, usage->type, last_key, 0);
-		}
-	}
 	/* report the usage code as scancode if the key status has changed */
 	if (usage->type == EV_KEY && !!test_bit(usage->code, input->key) != value)
 		input_event(input, EV_MSC, MSC_SCAN, usage->hid);
diff --git a/include/linux/hid.h b/include/linux/hid.h
index cd5d562b1b7c..dca5804836f3 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -525,6 +525,8 @@ int hid_set_field(struct hid_field *, unsigned, __s32);
 int hid_input_report(struct hid_device *, int type, u8 *, int, int);
 int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field);
 int hidinput_mapping_quirks(struct hid_usage *, struct input_dev *, unsigned long *, int *);
+void hidinput_event_quirks(struct hid_device *, struct hid_field *, struct hid_usage *, __s32);
+int hidinput_apple_event(struct hid_device *, struct input_dev *, struct hid_usage *, __s32);
 void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt);
 void hid_output_report(struct hid_report *report, __u8 *data);
 void hid_free_device(struct hid_device *device);
-- 
cgit v1.2.3


From 36ccaad640737899b069a9a93a82765f0e675a20 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Mon, 26 Nov 2007 13:18:00 +0100
Subject: HID: hid-input quirk for BTC 8193

BTC 8193 keyboard handles its scrollwheel in very non-standard way.
It produces two non-standard usages for scrolling up and down, in
both cases with postive value equaling to 1. We handle this by temporary
mapping, which we then catch in quirk event handler, and remap to
negative HWHEEL even in order to introduce correct behavior.

Also the button requires special mapping, as it triggers standard-violating
usage code.

Reported in kernel.org bugzilla #9385

Reported-by: Kir Kolyshkin <kir@sacred.ru>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input-quirks.c  | 34 ++++++++++++++++++++++++++++++++++
 drivers/hid/usbhid/hid-quirks.c |  5 +++++
 include/linux/hid.h             |  1 +
 3 files changed, 40 insertions(+)

(limited to 'include')

diff --git a/drivers/hid/hid-input-quirks.c b/drivers/hid/hid-input-quirks.c
index 7f2f80b00d74..e378b4a5eeac 100644
--- a/drivers/hid/hid-input-quirks.c
+++ b/drivers/hid/hid-input-quirks.c
@@ -253,6 +253,27 @@ static int quirk_cherry_genius_29e(struct hid_usage *usage, struct input_dev *in
 	return 1;
 }
 
+static int quirk_btc_8193(struct hid_usage *usage, struct input_dev *input,
+			      unsigned long *bit, int *max)
+{
+	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
+		return 0;
+
+	switch (usage->hid & HID_USAGE) {
+		case 0x230: map_key(BTN_MOUSE);			break;
+		case 0x231: map_rel(REL_WHEEL);			break;
+		/* 
+		 * this keyboard has a scrollwheel implemented in
+		 * totally broken way. We map this usage temporarily
+		 * to HWHEEL and handle it in the event quirk handler
+		 */
+		case 0x232: map_rel(REL_HWHEEL);		break;
+
+		default:
+			return 0;
+	}
+	return 1;
+}
 
 #define VENDOR_ID_BELKIN			0x1020
 #define DEVICE_ID_BELKIN_WIRELESS_KEYBOARD	0x0006
@@ -263,6 +284,9 @@ static int quirk_cherry_genius_29e(struct hid_usage *usage, struct input_dev *in
 #define VENDOR_ID_CHICONY			0x04f2
 #define DEVICE_ID_CHICONY_TACTICAL_PAD		0x0418
 
+#define VENDOR_ID_EZKEY				0x0518
+#define DEVICE_ID_BTC_8193			0x0002
+
 #define VENDOR_ID_LOGITECH			0x046d
 #define DEVICE_ID_LOGITECH_RECEIVER		0xc101
 #define DEVICE_ID_S510_RECEIVER			0xc50c
@@ -291,6 +315,8 @@ static const struct hid_input_blacklist {
 
 	{ VENDOR_ID_CHICONY, DEVICE_ID_CHICONY_TACTICAL_PAD, quirk_chicony_tactical_pad },
 
+	{ VENDOR_ID_EZKEY, DEVICE_ID_BTC_8193, quirk_btc_8193 },
+
 	{ VENDOR_ID_LOGITECH, DEVICE_ID_LOGITECH_RECEIVER, quirk_logitech_ultrax_remote },
 	{ VENDOR_ID_LOGITECH, DEVICE_ID_S510_RECEIVER, quirk_logitech_wireless },
 	{ VENDOR_ID_LOGITECH, DEVICE_ID_S510_RECEIVER_2, quirk_logitech_wireless },
@@ -323,6 +349,7 @@ int hidinput_mapping_quirks(struct hid_usage *usage,
 	return 0;
 }
 
+#define IS_BTC8193(x) (x->vendor == 0x0518 && x->product == 0x0002)
 #define IS_MS_KB(x) (x->vendor == 0x045e && (x->product == 0x00db || x->product == 0x00f9))
 
 void hidinput_event_quirks(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value)
@@ -384,6 +411,13 @@ void hidinput_event_quirks(struct hid_device *hid, struct hid_field *field, stru
 			input_event(input, usage->type, last_key, 0);
 		}
 	}
+
+	/* handle the temporary quirky mapping to HWHEEL */
+	if (hid->quirks & HID_QUIRK_HWHEEL_WHEEL_INVERT &&
+			usage->type == EV_REL && usage->code == REL_HWHEEL) {
+		input_event(input, usage->type, REL_WHEEL, -value);
+		return;
+	}
 }
 
 
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index d844757ed8de..aa470f8c0103 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -118,6 +118,9 @@
 #define USB_VENDOR_ID_ESSENTIAL_REALITY	0x0d7f
 #define USB_DEVICE_ID_ESSENTIAL_REALITY_P5 0x0100
 
+#define USB_VENDOR_ID_EZKEY 		0x0518
+#define USB_DEVICE_ID_BTC_8193		0x0002
+
 #define USB_VENDOR_ID_GAMERON		0x0810
 #define USB_DEVICE_ID_GAMERON_DUAL_PSX_ADAPTOR	0x0001
 
@@ -400,6 +403,8 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL4, HID_QUIRK_HIDDEV | HID_QUIRK_IGNORE_HIDINPUT },
 	{ USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_SIDEWINDER_GV, HID_QUIRK_HIDINPUT },
 
+	{ USB_VENDOR_ID_EZKEY, USB_DEVICE_ID_BTC_8193, HID_QUIRK_HWHEEL_WHEEL_INVERT },
+
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_01, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_10, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_20, HID_QUIRK_IGNORE },
diff --git a/include/linux/hid.h b/include/linux/hid.h
index dca5804836f3..33ec33389b37 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -282,6 +282,7 @@ struct hid_item {
 #define HID_QUIRK_LOGITECH_EXPANDED_KEYMAP	0x00800000
 #define HID_QUIRK_IGNORE_HIDINPUT		0x01000000
 #define HID_QUIRK_2WHEEL_MOUSE_HACK_B8		0x02000000
+#define HID_QUIRK_HWHEEL_WHEEL_INVERT		0x04000000
 
 /*
  * Separate quirks for runtime report descriptor fixup
-- 
cgit v1.2.3


From 628edcde87592a7ac6e72b555bb03ea265bcfbd2 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Mon, 26 Nov 2007 13:26:33 +0100
Subject: HID: proper handling of MS 4k and 6k devices

This removes ugly macros IS_* to distinguish devices that
need special handling in hid-input, and establish proper
quirks for them.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input-quirks.c  | 6 ++----
 drivers/hid/usbhid/hid-quirks.c | 5 +++++
 include/linux/hid.h             | 1 +
 3 files changed, 8 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/hid/hid-input-quirks.c b/drivers/hid/hid-input-quirks.c
index 7018be3a5a4c..fbe8b6de1a63 100644
--- a/drivers/hid/hid-input-quirks.c
+++ b/drivers/hid/hid-input-quirks.c
@@ -349,9 +349,6 @@ int hidinput_mapping_quirks(struct hid_usage *usage,
 	return 0;
 }
 
-#define IS_BTC8193(x) (x->vendor == 0x0518 && x->product == 0x0002)
-#define IS_MS_KB(x) (x->vendor == 0x045e && (x->product == 0x00db || x->product == 0x00f9))
-
 void hidinput_event_quirks(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value)
 {
 	struct input_dev *input;
@@ -392,7 +389,8 @@ void hidinput_event_quirks(struct hid_device *hid, struct hid_field *field, stru
 		return;
 
 	/* Handling MS keyboards special buttons */
-	if (IS_MS_KB(hid) && usage->hid == (HID_UP_MSVENDOR | 0xff05)) {
+	if (hid->quirks & HID_QUIRK_MICROSOFT_KEYS && 
+			usage->hid == (HID_UP_MSVENDOR | 0xff05)) {
 		int key = 0;
 		static int last_key = 0;
 		switch (value) {
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index aa470f8c0103..dcb102d87414 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -305,6 +305,8 @@
 #define USB_VENDOR_ID_MICROSOFT		0x045e
 #define USB_DEVICE_ID_SIDEWINDER_GV	0x003b
 #define USB_DEVICE_ID_WIRELESS_OPTICAL_DESKTOP_3_0 0x009d
+#define USB_DEVICE_ID_MS_NE4K		0x00db
+#define USB_DEVICE_ID_MS_LK6K		0x00f9
 
 #define USB_VENDOR_ID_MONTEREY		0x0566
 #define USB_DEVICE_ID_GENIUS_KB29E	0x3004
@@ -534,6 +536,9 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_ELITE_KBD, HID_QUIRK_LOGITECH_IGNORE_DOUBLED_WHEEL | HID_QUIRK_LOGITECH_EXPANDED_KEYMAP },
 	{ USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_CORDLESS_DESKTOP_LX500, HID_QUIRK_LOGITECH_IGNORE_DOUBLED_WHEEL | HID_QUIRK_LOGITECH_EXPANDED_KEYMAP },
 
+	{ USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_NE4K, HID_QUIRK_MICROSOFT_KEYS },
+	{ USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_LK6K, HID_QUIRK_MICROSOFT_KEYS },
+
 	{ USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MIGHTYMOUSE, HID_QUIRK_MIGHTYMOUSE | HID_QUIRK_INVERT_HWHEEL },
 
 	{ USB_VENDOR_ID_PANTHERLORD, USB_DEVICE_ID_PANTHERLORD_TWIN_USB_JOYSTICK, HID_QUIRK_MULTI_INPUT | HID_QUIRK_SKIP_OUTPUT_REPORTS },
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 33ec33389b37..24f04cd742de 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -283,6 +283,7 @@ struct hid_item {
 #define HID_QUIRK_IGNORE_HIDINPUT		0x01000000
 #define HID_QUIRK_2WHEEL_MOUSE_HACK_B8		0x02000000
 #define HID_QUIRK_HWHEEL_WHEEL_INVERT		0x04000000
+#define HID_QUIRK_MICROSOFT_KEYS		0x08000000
 
 /*
  * Separate quirks for runtime report descriptor fixup
-- 
cgit v1.2.3


From 70d215c4a7dfbddc138a2dd726d8f80f3e6d2622 Mon Sep 17 00:00:00 2001
From: Fengguang Wu <wfg@mail.ustc.edu.cn>
Date: Fri, 7 Dec 2007 16:35:14 +0800
Subject: HID: the `bit' in hidinput_mapping_quirks() is an out parameter

Fix a panic, by changing
	hidinput_mapping_quirks(,, unsigned long *bit,)
to
	hidinput_mapping_quirks(,, unsigned long **bit,)

The `bit' in this function is an out parameter.

Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-input-quirks.c | 36 ++++++++++++++++++------------------
 drivers/hid/hid-input.c        |  2 +-
 include/linux/hid.h            |  2 +-
 3 files changed, 20 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/drivers/hid/hid-input-quirks.c b/drivers/hid/hid-input-quirks.c
index fbe8b6de1a63..4bcdc9bb658e 100644
--- a/drivers/hid/hid-input-quirks.c
+++ b/drivers/hid/hid-input-quirks.c
@@ -16,16 +16,16 @@
 #include <linux/input.h>
 #include <linux/hid.h>
 
-#define map_abs(c)      do { usage->code = c; usage->type = EV_ABS; bit = input->absbit; *max = ABS_MAX; } while (0)
-#define map_rel(c)      do { usage->code = c; usage->type = EV_REL; bit = input->relbit; *max = REL_MAX; } while (0)
-#define map_key(c)      do { usage->code = c; usage->type = EV_KEY; bit = input->keybit; *max = KEY_MAX; } while (0)
-#define map_led(c)      do { usage->code = c; usage->type = EV_LED; bit = input->ledbit; *max = LED_MAX; } while (0)
+#define map_abs(c)      do { usage->code = c; usage->type = EV_ABS; *bit = input->absbit; *max = ABS_MAX; } while (0)
+#define map_rel(c)      do { usage->code = c; usage->type = EV_REL; *bit = input->relbit; *max = REL_MAX; } while (0)
+#define map_key(c)      do { usage->code = c; usage->type = EV_KEY; *bit = input->keybit; *max = KEY_MAX; } while (0)
+#define map_led(c)      do { usage->code = c; usage->type = EV_LED; *bit = input->ledbit; *max = LED_MAX; } while (0)
 
-#define map_abs_clear(c)        do { map_abs(c); clear_bit(c, bit); } while (0)
-#define map_key_clear(c)        do { map_key(c); clear_bit(c, bit); } while (0)
+#define map_abs_clear(c)        do { map_abs(c); clear_bit(c, *bit); } while (0)
+#define map_key_clear(c)        do { map_key(c); clear_bit(c, *bit); } while (0)
 
 static int quirk_belkin_wkbd(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
 		return 0;
@@ -41,7 +41,7 @@ static int quirk_belkin_wkbd(struct hid_usage *usage, struct input_dev *input,
 }
 
 static int quirk_cherry_cymotion(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
 		return 0;
@@ -57,7 +57,7 @@ static int quirk_cherry_cymotion(struct hid_usage *usage, struct input_dev *inpu
 }
 
 static int quirk_logitech_ultrax_remote(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_LOGIVENDOR)
 		return 0;
@@ -90,7 +90,7 @@ static int quirk_logitech_ultrax_remote(struct hid_usage *usage, struct input_de
 }
 
 static int quirk_chicony_tactical_pad(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
 		return 0;
@@ -115,7 +115,7 @@ static int quirk_chicony_tactical_pad(struct hid_usage *usage, struct input_dev
 }
 
 static int quirk_microsoft_ergonomy_kb(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
 		return 0;
@@ -138,7 +138,7 @@ static int quirk_microsoft_ergonomy_kb(struct hid_usage *usage, struct input_dev
 }
 
 static int quirk_microsoft_presenter_8k(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_MSVENDOR)
 		return 0;
@@ -156,7 +156,7 @@ static int quirk_microsoft_presenter_8k(struct hid_usage *usage, struct input_de
 }
 
 static int quirk_petalynx_remote(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if (((usage->hid & HID_USAGE_PAGE) != HID_UP_LOGIVENDOR) &&
 			((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER))
@@ -184,7 +184,7 @@ static int quirk_petalynx_remote(struct hid_usage *usage, struct input_dev *inpu
 }
 
 static int quirk_logitech_wireless(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
 		return 0;
@@ -236,7 +236,7 @@ static int quirk_logitech_wireless(struct hid_usage *usage, struct input_dev *in
 }
 
 static int quirk_cherry_genius_29e(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
 		return 0;
@@ -254,7 +254,7 @@ static int quirk_cherry_genius_29e(struct hid_usage *usage, struct input_dev *in
 }
 
 static int quirk_btc_8193(struct hid_usage *usage, struct input_dev *input,
-			      unsigned long *bit, int *max)
+			      unsigned long **bit, int *max)
 {
 	if ((usage->hid & HID_USAGE_PAGE) != HID_UP_CONSUMER)
 		return 0;
@@ -307,7 +307,7 @@ static int quirk_btc_8193(struct hid_usage *usage, struct input_dev *input,
 static const struct hid_input_blacklist {
 	__u16 idVendor;
 	__u16 idProduct;
-	int (*quirk)(struct hid_usage *, struct input_dev *, unsigned long *, int *);
+	int (*quirk)(struct hid_usage *, struct input_dev *, unsigned long **, int *);
 } hid_input_blacklist[] = {
 	{ VENDOR_ID_BELKIN, DEVICE_ID_BELKIN_WIRELESS_KEYBOARD, quirk_belkin_wkbd },
 
@@ -335,7 +335,7 @@ static const struct hid_input_blacklist {
 
 int hidinput_mapping_quirks(struct hid_usage *usage, 
 				   struct input_dev *input, 
-				   unsigned long *bit, int *max)
+				   unsigned long **bit, int *max)
 {
 	struct hid_device *device = input_get_drvdata(input);
 	int i = 0;
diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index aeb018e31bfc..5325d98b4328 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -382,7 +382,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
 	}
 
 	/* handle input mappings for quirky devices */
-	ret = hidinput_mapping_quirks(usage, input, bit, &max);
+	ret = hidinput_mapping_quirks(usage, input, &bit, &max);
 	if (ret)
 		goto mapped;
 
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 24f04cd742de..6a70b788ee9c 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -526,7 +526,7 @@ extern void hidinput_disconnect(struct hid_device *);
 int hid_set_field(struct hid_field *, unsigned, __s32);
 int hid_input_report(struct hid_device *, int type, u8 *, int, int);
 int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field);
-int hidinput_mapping_quirks(struct hid_usage *, struct input_dev *, unsigned long *, int *);
+int hidinput_mapping_quirks(struct hid_usage *, struct input_dev *, unsigned long **, int *);
 void hidinput_event_quirks(struct hid_device *, struct hid_field *, struct hid_usage *, __s32);
 int hidinput_apple_event(struct hid_device *, struct input_dev *, struct hid_usage *, __s32);
 void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt);
-- 
cgit v1.2.3


From fe56caa97e626cc6d6e18adbd5ccd1a9aa9a4fcf Mon Sep 17 00:00:00 2001
From: Robert Schedel <r.schedel@yahoo.de>
Date: Wed, 26 Dec 2007 00:57:40 +0100
Subject: HID: Support Samsung IR remote

Samsung USB remotes (0419:0001) are rejected by kernel 2.6.23, because the
report descriptor from the remote contains a 48 bit HID report field. HID 1.11
states: Fields may span at most 4 bytes.

This patch, based on 2.6.23, fixes this by modifying the internal report
descriptor in hid-quirks.c. Additional user space support (e.g. LIRC) is
required to fetch the information from the hiddev interface.

The burden to reconstruct the data is moved into userspace (lirc through hiddev).
There is no need to set HID_QUIRK_HIDDEV quirk, as the device has also output
applications, which trigger the creation of hiddev device automatically.

Signed-off-by: Robert Schedel <r.schedel@yahoo.de>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/usbhid/hid-quirks.c | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/hid.h             |  1 +
 2 files changed, 37 insertions(+)

(limited to 'include')

diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 68b38fdb1203..30f63e9f93a3 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -341,6 +341,9 @@
 #define USB_VENDOR_ID_SAITEK		0x06a3
 #define USB_DEVICE_ID_SAITEK_RUMBLEPAD	0xff17
 
+#define USB_VENDOR_ID_SAMSUNG		0x0419
+#define USB_DEVICE_ID_SAMSUNG_IR_REMOTE	0x0001
+
 #define USB_VENDOR_ID_SONY			0x054c
 #define USB_DEVICE_ID_SONY_PS3_CONTROLLER	0x0268
 
@@ -673,6 +676,8 @@ static const struct hid_rdesc_blacklist {
 
 	{ USB_VENDOR_ID_PETALYNX, USB_DEVICE_ID_PETALYNX_MAXTER_REMOTE, HID_QUIRK_RDESC_PETALYNX },
 
+	{ USB_VENDOR_ID_SAMSUNG, USB_DEVICE_ID_SAMSUNG_IR_REMOTE, HID_QUIRK_RDESC_SAMSUNG_REMOTE },
+
 	{ USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_BARCODE_1, HID_QUIRK_RDESC_SWAPPED_MIN_MAX },
 	{ USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_BARCODE_2, HID_QUIRK_RDESC_SWAPPED_MIN_MAX },
 
@@ -947,6 +952,33 @@ static void usbhid_fixup_logitech_descriptor(unsigned char *rdesc, int rsize)
 	}
 }
 
+/*
+ * Samsung IrDA remote controller (reports as Cypress USB Mouse).
+ *
+ * Vendor specific report #4 has a size of 48 bit,
+ * and therefore is not accepted when inspecting the descriptors.
+ * As a workaround we reinterpret the report as:
+ *   Variable type, count 6, size 8 bit, log. maximum 255
+ * The burden to reconstruct the data is moved into user space.
+ */
+static void usbhid_fixup_samsung_irda_descriptor(unsigned char *rdesc,
+						  int rsize)
+{
+	if (rsize >= 182 && rdesc[175] == 0x25
+			 && rdesc[176] == 0x40
+			 && rdesc[177] == 0x75
+			 && rdesc[178] == 0x30
+			 && rdesc[179] == 0x95
+			 && rdesc[180] == 0x01
+			 && rdesc[182] == 0x40) {
+		printk(KERN_INFO "Fixing up Samsung IrDA report descriptor\n");
+		rdesc[176] = 0xff;
+		rdesc[178] = 0x08;
+		rdesc[180] = 0x06;
+		rdesc[182] = 0x42;
+	}
+}
+
 /* Petalynx Maxter Remote has maximum for consumer page set too low */
 static void usbhid_fixup_petalynx_descriptor(unsigned char *rdesc, int rsize)
 {
@@ -1026,6 +1058,10 @@ static void __usbhid_fixup_report_descriptor(__u32 quirks, char *rdesc, unsigned
 
 	if (quirks & HID_QUIRK_RDESC_BUTTON_CONSUMER)
 		usbhid_fixup_button_consumer_descriptor(rdesc, rsize);
+
+	if (quirks & HID_QUIRK_RDESC_SAMSUNG_REMOTE)
+		usbhid_fixup_samsung_irda_descriptor(rdesc, rsize);
+
 }
 
 /**
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 6a70b788ee9c..3902690647b0 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -295,6 +295,7 @@ struct hid_item {
 #define HID_QUIRK_RDESC_PETALYNX		0x00000008
 #define HID_QUIRK_RDESC_MACBOOK_JIS		0x00000010
 #define HID_QUIRK_RDESC_BUTTON_CONSUMER		0x00000020
+#define HID_QUIRK_RDESC_SAMSUNG_REMOTE		0x00000040
 
 /*
  * This is the global environment of the parser. This information is
-- 
cgit v1.2.3


From 7998a731664ac4988b349e70669bac11e3b3a3ac Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Mon, 31 Dec 2007 14:58:38 +0100
Subject: A few corrections to include/linux/Kbuild

auxvec.h, i2c-dev.h and vt.h *should* be unifdef'ed i2o-dev.h does not need
unifdef'ing

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/linux/Kbuild | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index bd694f779346..ad99ce9f9169 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -34,7 +34,6 @@ header-y += atmsap.h
 header-y += atmsvc.h
 header-y += atm_zatm.h
 header-y += auto_fs4.h
-header-y += auxvec.h
 header-y += ax25.h
 header-y += b1lli.h
 header-y += baycom.h
@@ -73,7 +72,7 @@ header-y += gen_stats.h
 header-y += gigaset_dev.h
 header-y += hdsmart.h
 header-y += hysdn_if.h
-header-y += i2c-dev.h
+header-y += i2o-dev.h
 header-y += i8k.h
 header-y += if_arcnet.h
 header-y += if_bonding.h
@@ -158,7 +157,6 @@ header-y += veth.h
 header-y += video_decoder.h
 header-y += video_encoder.h
 header-y += videotext.h
-header-y += vt.h
 header-y += x25.h
 
 unifdef-y += acct.h
@@ -173,6 +171,7 @@ unifdef-y += atm.h
 unifdef-y += atm_tcp.h
 unifdef-y += audit.h
 unifdef-y += auto_fs.h
+unifdef-y += auxvec.h
 unifdef-y += binfmts.h
 unifdef-y += capability.h
 unifdef-y += capi.h
@@ -214,7 +213,7 @@ unifdef-y += hdreg.h
 unifdef-y += hiddev.h
 unifdef-y += hpet.h
 unifdef-y += i2c.h
-unifdef-y += i2o-dev.h
+unifdef-y += i2c-dev.h
 unifdef-y += icmp.h
 unifdef-y += icmpv6.h
 unifdef-y += if_addr.h
@@ -349,6 +348,7 @@ unifdef-y += videodev.h
 unifdef-y += virtio_config.h
 unifdef-y += virtio_blk.h
 unifdef-y += virtio_net.h
+unifdef-y += vt.h
 unifdef-y += wait.h
 unifdef-y += wanrouter.h
 unifdef-y += watchdog.h
-- 
cgit v1.2.3


From 01ba2bdc6b639764745ff678caf3fb9e5bcd745a Mon Sep 17 00:00:00 2001
From: Sam Ravnborg <sam@ravnborg.org>
Date: Sun, 20 Jan 2008 14:15:03 +0100
Subject: all archs: consolidate init and exit sections in vmlinux.lds.h

This patch consolidate all definitions of .init.text, .init.data
and .exit.text, .exit.data section definitions in
the generic vmlinux.lds.h.

This is a preparational patch - alone it does not buy
us much good.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 arch/alpha/kernel/vmlinux.lds.S     |  8 ++++----
 arch/arm/kernel/vmlinux.lds.S       | 10 +++++-----
 arch/avr32/kernel/vmlinux.lds.S     |  8 ++++----
 arch/blackfin/kernel/vmlinux.lds.S  |  8 ++++----
 arch/cris/arch-v10/vmlinux.lds.S    |  8 ++++----
 arch/cris/arch-v32/vmlinux.lds.S    |  8 ++++----
 arch/frv/kernel/vmlinux.lds.S       | 14 +++++++-------
 arch/h8300/kernel/vmlinux.lds.S     |  8 ++++----
 arch/ia64/kernel/vmlinux.lds.S      |  8 ++++----
 arch/m32r/kernel/vmlinux.lds.S      | 12 ++++++------
 arch/m68k/kernel/vmlinux-std.lds    |  8 ++++----
 arch/m68k/kernel/vmlinux-sun3.lds   |  8 ++++----
 arch/m68knommu/kernel/vmlinux.lds.S |  8 ++++----
 arch/mips/kernel/vmlinux.lds.S      |  8 ++++----
 arch/parisc/kernel/vmlinux.lds.S    |  8 ++++----
 arch/powerpc/kernel/vmlinux.lds.S   | 10 ++++++----
 arch/ppc/kernel/vmlinux.lds.S       |  8 ++++----
 arch/s390/kernel/vmlinux.lds.S      |  8 ++++----
 arch/sh/kernel/vmlinux_32.lds.S     |  8 ++++----
 arch/sh/kernel/vmlinux_64.lds.S     |  8 ++++----
 arch/sparc/kernel/vmlinux.lds.S     |  8 ++++----
 arch/sparc64/kernel/vmlinux.lds.S   |  8 ++++----
 arch/um/kernel/dyn.lds.S            |  4 ++--
 arch/um/kernel/uml.lds.S            |  4 ++--
 arch/v850/kernel/vmlinux.lds.S      | 10 +++++-----
 arch/x86/kernel/vmlinux_32.lds.S    | 14 ++++++++++----
 arch/x86/kernel/vmlinux_64.lds.S    | 19 +++++++++++++------
 arch/xtensa/kernel/vmlinux.lds.S    |  9 +++++----
 include/asm-generic/vmlinux.lds.h   |  7 +++++++
 29 files changed, 140 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/kernel/vmlinux.lds.S b/arch/alpha/kernel/vmlinux.lds.S
index 55c05b511f4c..f13249be17c5 100644
--- a/arch/alpha/kernel/vmlinux.lds.S
+++ b/arch/alpha/kernel/vmlinux.lds.S
@@ -46,11 +46,11 @@ SECTIONS
 	__init_begin = .;
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 
 	. = ALIGN(16);
@@ -136,8 +136,8 @@ SECTIONS
 
 	/* Sections to be discarded */
 	/DISCARD/ : {
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 30f732c7fdb5..4898bdcfe7dd 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -30,7 +30,7 @@ SECTIONS
 	}
 
 	.init : {			/* Init code and data		*/
-			*(.init.text)
+			INIT_TEXT
 		_einittext = .;
 		__proc_info_begin = .;
 			*(.proc.info.init)
@@ -70,15 +70,15 @@ SECTIONS
 		__per_cpu_end = .;
 #ifndef CONFIG_XIP_KERNEL
 		__init_begin = _stext;
-		*(.init.data)
+		INIT_DATA
 		. = ALIGN(4096);
 		__init_end = .;
 #endif
 	}
 
 	/DISCARD/ : {			/* Exit code and data		*/
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 #ifndef CONFIG_MMU
 		*(.fixup)
@@ -130,7 +130,7 @@ SECTIONS
 #ifdef CONFIG_XIP_KERNEL
 		. = ALIGN(4096);
 		__init_begin = .;
-		*(.init.data)
+		INIT_DATA
 		. = ALIGN(4096);
 		__init_end = .;
 #endif
diff --git a/arch/avr32/kernel/vmlinux.lds.S b/arch/avr32/kernel/vmlinux.lds.S
index 11f08e35a2eb..481cfd40c053 100644
--- a/arch/avr32/kernel/vmlinux.lds.S
+++ b/arch/avr32/kernel/vmlinux.lds.S
@@ -27,19 +27,19 @@ SECTIONS
 		__init_begin = .;
 			_sinittext = .;
 			*(.text.reset)
-			*(.init.text)
+			INIT_TEXT
 			/*
 			 * .exit.text is discarded at runtime, not
 			 * link time, to deal with references from
 			 * __bug_table
 			 */
-			*(.exit.text)
+			EXIT_TEXT
 			_einittext = .;
 		. = ALIGN(4);
 		__tagtable_begin = .;
 			*(.taglist.init)
 		__tagtable_end = .;
-			*(.init.data)
+			INIT_DATA
 		. = ALIGN(16);
 		__setup_start = .;
 			*(.init.setup)
@@ -135,7 +135,7 @@ SECTIONS
 	 * thrown away, as cleanup code is never called unless it's a module.
 	 */
 	/DISCARD/       	: {
-		*(.exit.data)
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/blackfin/kernel/vmlinux.lds.S b/arch/blackfin/kernel/vmlinux.lds.S
index 9b75bc83c71f..858722421b40 100644
--- a/arch/blackfin/kernel/vmlinux.lds.S
+++ b/arch/blackfin/kernel/vmlinux.lds.S
@@ -91,13 +91,13 @@ SECTIONS
 	{
 		. = ALIGN(PAGE_SIZE);
 		__sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		__einittext = .;
 	}
 	.init.data :
 	{
 		. = ALIGN(16);
-		*(.init.data)
+		INIT_DATA
 	}
 	.init.setup :
 	{
@@ -198,8 +198,8 @@ SECTIONS
 
 	/DISCARD/ :
 	{
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 }
diff --git a/arch/cris/arch-v10/vmlinux.lds.S b/arch/cris/arch-v10/vmlinux.lds.S
index 97a7876ed681..93c9f0ea286b 100644
--- a/arch/cris/arch-v10/vmlinux.lds.S
+++ b/arch/cris/arch-v10/vmlinux.lds.S
@@ -57,10 +57,10 @@ SECTIONS
   	__init_begin = .;
 	.init.text : { 
 		   _sinittext = .;
-		   *(.init.text)
+		   INIT_TEXT
 		   _einittext = .;
 	}
-  	.init.data : { *(.init.data) }
+	.init.data : { INIT_DATA }
   	. = ALIGN(16);
   	__setup_start = .;
   	.init.setup : { *(.init.setup) }
@@ -109,8 +109,8 @@ SECTIONS
 
 	/* Sections to be discarded */
   	/DISCARD/ : {
-        	*(.text.exit)
-        	*(.data.exit)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
         }
 
diff --git a/arch/cris/arch-v32/vmlinux.lds.S b/arch/cris/arch-v32/vmlinux.lds.S
index b076c134c0bb..fead8c59ea63 100644
--- a/arch/cris/arch-v32/vmlinux.lds.S
+++ b/arch/cris/arch-v32/vmlinux.lds.S
@@ -61,10 +61,10 @@ SECTIONS
   	__init_begin = .;
 	.init.text : {
 		   _sinittext = .;
-		   *(.init.text)
+		   INIT_TEXT
 		   _einittext = .;
 	}
-  	.init.data : { *(.init.data) }
+	.init.data : { INIT_DATA }
   	. = ALIGN(16);
   	__setup_start = .;
   	.init.setup : { *(.init.setup) }
@@ -124,8 +124,8 @@ SECTIONS
 
 	/* Sections to be discarded */
   	/DISCARD/ : {
-        	*(.text.exit)
-        	*(.data.exit)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
         }
 
diff --git a/arch/frv/kernel/vmlinux.lds.S b/arch/frv/kernel/vmlinux.lds.S
index a17a81d58bf6..f42b328b1dd0 100644
--- a/arch/frv/kernel/vmlinux.lds.S
+++ b/arch/frv/kernel/vmlinux.lds.S
@@ -28,14 +28,14 @@ SECTIONS
   .init.text : {
 	*(.text.head)
 #ifndef CONFIG_DEBUG_INFO
-	*(.init.text)
-	*(.exit.text)
-	*(.exit.data)
+	INIT_TEXT
+	EXIT_TEXT
+	EXIT_DATA
 	*(.exitcall.exit)
 #endif
   }
   _einittext = .;
-  .init.data : { *(.init.data) }
+  .init.data : { INIT_DATA }
 
   . = ALIGN(8);
   __setup_start = .;
@@ -106,8 +106,8 @@ SECTIONS
 	LOCK_TEXT
 #ifdef CONFIG_DEBUG_INFO
 	*(
-	.init.text
-	.exit.text
+	INIT_TEXT
+	EXIT_TEXT
 	.exitcall.exit
 	)
 #endif
@@ -138,7 +138,7 @@ SECTIONS
   .data : {			/* Data */
 	DATA_DATA
 	*(.data.*)
-	*(.exit.data)
+	EXIT_DATA
 	CONSTRUCTORS
 	}
 
diff --git a/arch/h8300/kernel/vmlinux.lds.S b/arch/h8300/kernel/vmlinux.lds.S
index a2e72d495551..43a87b9085b6 100644
--- a/arch/h8300/kernel/vmlinux.lds.S
+++ b/arch/h8300/kernel/vmlinux.lds.S
@@ -110,9 +110,9 @@ SECTIONS
 	. = ALIGN(0x4) ;
 	___init_begin = .;
 	__sinittext = .; 
-		*(.init.text)
+		INIT_TEXT
 	__einittext = .; 
-		*(.init.data)
+		INIT_DATA
 	. = ALIGN(0x4) ;
 	___setup_start = .;
 		*(.init.setup)
@@ -124,8 +124,8 @@ SECTIONS
 	___con_initcall_start = .;
 		*(.con_initcall.init)
 	___con_initcall_end = .;
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 #if defined(CONFIG_BLK_DEV_INITRD)
 		. = ALIGN(4);
 	___initramfs_start = .;
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index 757e419ebcf8..80622acc95de 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -27,8 +27,8 @@ SECTIONS
 {
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exit.text)
-	*(.exit.data)
+	EXIT_TEXT
+	EXIT_DATA
 	*(.exitcall.exit)
 	*(.IA_64.unwind.exit.text)
 	*(.IA_64.unwind_info.exit.text)
@@ -119,12 +119,12 @@ SECTIONS
   .init.text : AT(ADDR(.init.text) - LOAD_OFFSET)
 	{
 	  _sinittext = .;
-	  *(.init.text)
+	  INIT_TEXT
 	  _einittext = .;
 	}
 
   .init.data : AT(ADDR(.init.data) - LOAD_OFFSET)
-	{ *(.init.data) }
+	{ INIT_DATA }
 
 #ifdef CONFIG_BLK_DEV_INITRD
   .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET)
diff --git a/arch/m32r/kernel/vmlinux.lds.S b/arch/m32r/kernel/vmlinux.lds.S
index 942a8c7a4417..41b07854fcc6 100644
--- a/arch/m32r/kernel/vmlinux.lds.S
+++ b/arch/m32r/kernel/vmlinux.lds.S
@@ -76,10 +76,10 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
-  .init.data : { *(.init.data) }
+  .init.data : { INIT_DATA }
   . = ALIGN(16);
   __setup_start = .;
   .init.setup : { *(.init.setup) }
@@ -100,8 +100,8 @@ SECTIONS
   .altinstr_replacement : { *(.altinstr_replacement) }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : { *(.exit.text) }
-  .exit.data : { *(.exit.data) }
+  .exit.text : { EXIT_TEXT }
+  .exit.data : { EXIT_DATA }
 
 #ifdef CONFIG_BLK_DEV_INITRD
   . = ALIGN(4096);
@@ -124,8 +124,8 @@ SECTIONS
 
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exit.text)
-	*(.exit.data)
+	EXIT_TEXT
+	EXIT_DATA
 	*(.exitcall.exit)
 	}
 
diff --git a/arch/m68k/kernel/vmlinux-std.lds b/arch/m68k/kernel/vmlinux-std.lds
index 59fe285865ec..7537cc5e6159 100644
--- a/arch/m68k/kernel/vmlinux-std.lds
+++ b/arch/m68k/kernel/vmlinux-std.lds
@@ -45,10 +45,10 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
-  .init.data : { *(.init.data) }
+  .init.data : { INIT_DATA }
   . = ALIGN(16);
   __setup_start = .;
   .init.setup : { *(.init.setup) }
@@ -82,8 +82,8 @@ SECTIONS
 
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exit.text)
-	*(.exit.data)
+	EXIT_TEXT
+	EXIT_DATA
 	*(.exitcall.exit)
 	}
 
diff --git a/arch/m68k/kernel/vmlinux-sun3.lds b/arch/m68k/kernel/vmlinux-sun3.lds
index 4adffefb5c48..cdc313e7c299 100644
--- a/arch/m68k/kernel/vmlinux-sun3.lds
+++ b/arch/m68k/kernel/vmlinux-sun3.lds
@@ -38,10 +38,10 @@ SECTIONS
 __init_begin = .;
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
-	.init.data : { *(.init.data) }
+	.init.data : { INIT_DATA }
 	. = ALIGN(16);
 	__setup_start = .;
 	.init.setup : { *(.init.setup) }
@@ -77,8 +77,8 @@ __init_begin = .;
 
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exit.text)
-	*(.exit.data)
+	EXIT_TEXT
+	EXIT_DATA
 	*(.exitcall.exit)
 	}
 
diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S
index 07a0055602f4..b44edb08e212 100644
--- a/arch/m68knommu/kernel/vmlinux.lds.S
+++ b/arch/m68knommu/kernel/vmlinux.lds.S
@@ -143,9 +143,9 @@ SECTIONS {
 		. = ALIGN(4096);
 		__init_begin = .;
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
-		*(.init.data)
+		INIT_DATA
 		. = ALIGN(16);
 		__setup_start = .;
 		*(.init.setup)
@@ -170,8 +170,8 @@ SECTIONS {
 	} > INIT
 
 	/DISCARD/ : {
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S
index 5fc2398bdb76..b5470ceb418b 100644
--- a/arch/mips/kernel/vmlinux.lds.S
+++ b/arch/mips/kernel/vmlinux.lds.S
@@ -114,11 +114,11 @@ SECTIONS
 	__init_begin = .;
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 	. = ALIGN(16);
 	.init.setup : {
@@ -144,10 +144,10 @@ SECTIONS
 	 * references from .rodata
 	 */
 	.exit.text : {
-		*(.exit.text)
+		EXIT_TEXT
 	}
 	.exit.data : {
-		*(.exit.data)
+		EXIT_DATA
 	}
 #if defined(CONFIG_BLK_DEV_INITRD)
 	. = ALIGN(_PAGE_SIZE);
diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S
index 40d0ff9b81ab..50b4a3a25d0a 100644
--- a/arch/parisc/kernel/vmlinux.lds.S
+++ b/arch/parisc/kernel/vmlinux.lds.S
@@ -172,11 +172,11 @@ SECTIONS
 	__init_begin = .;
 	.init.text : { 
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 	. = ALIGN(16);
 	.init.setup : {
@@ -215,10 +215,10 @@ SECTIONS
 	 *  from .altinstructions and .eh_frame
 	 */
 	.exit.text : {
-		*(.exit.text)
+		EXIT_TEXT
 	}
 	.exit.data : {
-		*(.exit.data)
+		EXIT_DATA
 	}
 #ifdef CONFIG_BLK_DEV_INITRD
 	. = ALIGN(PAGE_SIZE);
diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index f66fa5d966b0..0afb9e31d2a0 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -23,7 +23,7 @@ SECTIONS
 	/* Sections to be discarded. */
 	/DISCARD/ : {
 	*(.exitcall.exit)
-	*(.exit.data)
+	EXIT_DATA
 	}
 
 	. = KERNELBASE;
@@ -76,17 +76,19 @@ SECTIONS
 
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 
 	/* .exit.text is discarded at runtime, not link time,
 	 * to deal with references from __bug_table
 	 */
-	.exit.text : { *(.exit.text) }
+	.exit.text : {
+		EXIT_TEXT
+	}
 
 	.init.data : {
-		*(.init.data);
+		INIT_DATA
 		__vtop_table_begin = .;
 		*(.vtop_fixup);
 		__vtop_table_end = .;
diff --git a/arch/ppc/kernel/vmlinux.lds.S b/arch/ppc/kernel/vmlinux.lds.S
index 98c1212674f6..52b64fcbdfc5 100644
--- a/arch/ppc/kernel/vmlinux.lds.S
+++ b/arch/ppc/kernel/vmlinux.lds.S
@@ -97,14 +97,14 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
   /* .exit.text is discarded at runtime, not link time,
      to deal with references from __bug_table */
-  .exit.text : { *(.exit.text) }
+  .exit.text : { EXIT_TEXT }
   .init.data : {
-    *(.init.data);
+    INIT_DATA
     __vtop_table_begin = .;
     *(.vtop_fixup);
     __vtop_table_end = .;
@@ -164,6 +164,6 @@ SECTIONS
   /* Sections to be discarded. */
   /DISCARD/ : {
     *(.exitcall.exit)
-    *(.exit.data)
+    EXIT_DATA
   }
 }
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index 936159199346..7d43c3cd3ef3 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -97,7 +97,7 @@ SECTIONS
 	__init_begin = .;
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	/*
@@ -105,11 +105,11 @@ SECTIONS
 	 * to deal with references from __bug_table
 	*/
 	.exit.text : {
-		*(.exit.text)
+		EXIT_TEXT
 	}
 
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 	. = ALIGN(0x100);
 	.init.setup : {
@@ -156,7 +156,7 @@ SECTIONS
 
 	/* Sections to be discarded */
 	/DISCARD/ : {
-		*(.exit.data)
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/sh/kernel/vmlinux_32.lds.S b/arch/sh/kernel/vmlinux_32.lds.S
index d549fac6d3e7..c7113786ecd4 100644
--- a/arch/sh/kernel/vmlinux_32.lds.S
+++ b/arch/sh/kernel/vmlinux_32.lds.S
@@ -84,9 +84,9 @@ SECTIONS
 	. = ALIGN(PAGE_SIZE);		/* Init code and data */
 	__init_begin = .;
 	_sinittext = .;
-	.init.text : { *(.init.text) }
+	.init.text : { INIT_TEXT }
 	_einittext = .;
-	.init.data : { *(.init.data) }
+	.init.data : { INIT_DATA }
 
 	. = ALIGN(16);
 	__setup_start = .;
@@ -122,8 +122,8 @@ SECTIONS
 	 * .exit.text is discarded at runtime, not link time, to deal with
 	 * references from __bug_table
 	 */
-	.exit.text : { *(.exit.text) }
-	.exit.data : { *(.exit.data) }
+	.exit.text : { EXIT_TEXT }
+	.exit.data : { EXIT_DATA }
 
 	. = ALIGN(PAGE_SIZE);
 	.bss : {
diff --git a/arch/sh/kernel/vmlinux_64.lds.S b/arch/sh/kernel/vmlinux_64.lds.S
index 2fd0f7401484..3f1bd6392bb3 100644
--- a/arch/sh/kernel/vmlinux_64.lds.S
+++ b/arch/sh/kernel/vmlinux_64.lds.S
@@ -96,9 +96,9 @@ SECTIONS
 	. = ALIGN(PAGE_SIZE);		/* Init code and data */
 	__init_begin = .;
 	_sinittext = .;
-	.init.text : C_PHYS(.init.text) { *(.init.text) }
+	.init.text : C_PHYS(.init.text) { INIT_TEXT }
 	_einittext = .;
-	.init.data : C_PHYS(.init.data) { *(.init.data) }
+	.init.data : C_PHYS(.init.data) { INIT_DATA }
 	. = ALIGN(L1_CACHE_BYTES);	/* Better if Cache Line aligned */
 	__setup_start = .;
 	.init.setup : C_PHYS(.init.setup) { *(.init.setup) }
@@ -134,8 +134,8 @@ SECTIONS
 	 * .exit.text is discarded at runtime, not link time, to deal with
 	 * references from __bug_table
 	 */
-	.exit.text : C_PHYS(.exit.text) { *(.exit.text) }
-	.exit.data : C_PHYS(.exit.data) { *(.exit.data) }
+	.exit.text : C_PHYS(.exit.text) { EXIT_TEXT }
+	.exit.data : C_PHYS(.exit.data) { EXIT_DATA }
 
 	. = ALIGN(PAGE_SIZE);
 	.bss : C_PHYS(.bss) {
diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S
index a8b4200f9cc3..216147d6e61f 100644
--- a/arch/sparc/kernel/vmlinux.lds.S
+++ b/arch/sparc/kernel/vmlinux.lds.S
@@ -48,12 +48,12 @@ SECTIONS
 	__init_begin = .;
 	.init.text : {
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	__init_text_end = .;
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 	. = ALIGN(16);
 	.init.setup : {
@@ -102,8 +102,8 @@ SECTIONS
 	_end = . ;
 	PROVIDE (end = .);
 	/DISCARD/ : {
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/sparc64/kernel/vmlinux.lds.S b/arch/sparc64/kernel/vmlinux.lds.S
index 9fcd503bc04a..01f809617e5e 100644
--- a/arch/sparc64/kernel/vmlinux.lds.S
+++ b/arch/sparc64/kernel/vmlinux.lds.S
@@ -56,11 +56,11 @@ SECTIONS
 	.init.text : {
 		__init_begin = .;
 		_sinittext = .;
-		*(.init.text)
+		INIT_TEXT
 		_einittext = .;
 	}
 	.init.data : {
-		*(.init.data)
+		INIT_DATA
 	}
 	. = ALIGN(16);
 	.init.setup : {
@@ -137,8 +137,8 @@ SECTIONS
 	PROVIDE (end = .);
 
 	/DISCARD/ : {
-		*(.exit.text)
-		*(.exit.data)
+		EXIT_TEXT
+		EXIT_DATA
 		*(.exitcall.exit)
 	}
 
diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S
index 3866f4960f04..26090b7f323e 100644
--- a/arch/um/kernel/dyn.lds.S
+++ b/arch/um/kernel/dyn.lds.S
@@ -17,7 +17,7 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
 
@@ -84,7 +84,7 @@ SECTIONS
 
   #include "asm/common.lds.S"
 
-  init.data : { *(.init.data) }
+  init.data : { INIT_DATA }
 
   /* Ensure the __preinit_array_start label is properly aligned.  We
      could instead move the label definition inside the section, but
diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S
index 13df191e2b41..5828c1d54505 100644
--- a/arch/um/kernel/uml.lds.S
+++ b/arch/um/kernel/uml.lds.S
@@ -23,7 +23,7 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
   . = ALIGN(4096);
@@ -48,7 +48,7 @@ SECTIONS
 
   #include "asm/common.lds.S"
 
-  init.data : { *(init.data) }
+  init.data : { INIT_DATA }
   .data    :
   {
     . = ALIGN(KERNEL_STACK_SIZE);		/* init_task */
diff --git a/arch/v850/kernel/vmlinux.lds.S b/arch/v850/kernel/vmlinux.lds.S
index 6172599b4ce2..d08cd1d27f27 100644
--- a/arch/v850/kernel/vmlinux.lds.S
+++ b/arch/v850/kernel/vmlinux.lds.S
@@ -114,7 +114,7 @@
 #define DATA_CONTENTS							      \
 		__sdata = . ;						      \
 		DATA_DATA						      \
-			*(.exit.data)	/* 2.5 convention */		      \
+			EXIT_DATA	/* 2.5 convention */		      \
 			*(.data.exit)	/* 2.4 convention */		      \
 		. = ALIGN (16) ;					      \
 		*(.data.cacheline_aligned)				      \
@@ -157,9 +157,9 @@
 		. = ALIGN (4096) ;					      \
 		__init_start = . ;					      \
 			__sinittext = .;				      \
-			*(.init.text)	/* 2.5 convention */		      \
+			INIT_TEXT	/* 2.5 convention */		      \
 			__einittext = .;				      \
-			*(.init.data)					      \
+			INIT_DATA					      \
 			*(.text.init)	/* 2.4 convention */		      \
 			*(.data.init)					      \
 		INITCALL_CONTENTS					      \
@@ -170,7 +170,7 @@
 #define ROMK_INIT_RAM_CONTENTS						      \
 		. = ALIGN (4096) ;					      \
 		__init_start = . ;					      \
-			*(.init.data)	/* 2.5 convention */		      \
+			INIT_DATA	/* 2.5 convention */		      \
 			*(.data.init)	/* 2.4 convention */		      \
 		__init_end = . ;					      \
 		. = ALIGN (4096) ;
@@ -179,7 +179,7 @@
    should go into ROM.  */	
 #define ROMK_INIT_ROM_CONTENTS						      \
 			_sinittext = .;					      \
-			*(.init.text)	/* 2.5 convention */		      \
+			INIT_TEXT	/* 2.5 convention */		      \
 			_einittext = .;					      \
 			*(.text.init)	/* 2.4 convention */		      \
 		INITCALL_CONTENTS					      \
diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index 7d72cce00529..84c913f38f98 100644
--- a/arch/x86/kernel/vmlinux_32.lds.S
+++ b/arch/x86/kernel/vmlinux_32.lds.S
@@ -131,10 +131,12 @@ SECTIONS
   .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
   	__init_begin = .;
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
-  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) }
+  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) {
+	INIT_DATA
+  }
   . = ALIGN(16);
   .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) {
   	__setup_start = .;
@@ -169,8 +171,12 @@ SECTIONS
   }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) }
-  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) }
+  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) {
+	EXIT_TEXT
+  }
+  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) {
+	EXIT_DATA
+  }
 #if defined(CONFIG_BLK_DEV_INITRD)
   . = ALIGN(4096);
   .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) {
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index ba8ea97abd21..ea5386944e67 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -155,12 +155,15 @@ SECTIONS
   __init_begin = .;
   .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT
 	_einittext = .;
   }
-  __initdata_begin = .;
-  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { *(.init.data) }
-  __initdata_end = .;
+  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) {
+	__initdata_begin = .;
+	INIT_DATA
+	__initdata_end = .;
+   }
+
   . = ALIGN(16);
   __setup_start = .;
   .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) }
@@ -187,8 +190,12 @@ SECTIONS
   }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { *(.exit.text) }
-  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { *(.exit.data) }
+  .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) {
+	EXIT_TEXT
+  }
+  .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) {
+	EXIT_DATA
+  }
 
 /* vdso blob that is mapped into user space */
   vdso_start = . ;
diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S
index ac4ed52034db..7d0f55a4982d 100644
--- a/arch/xtensa/kernel/vmlinux.lds.S
+++ b/arch/xtensa/kernel/vmlinux.lds.S
@@ -136,13 +136,13 @@ SECTIONS
   __init_begin = .;
   .init.text : {
   	_sinittext = .;
-	*(.init.literal) *(.init.text)
+	*(.init.literal) INIT_TEXT
 	_einittext = .;
   }
 
   .init.data :
   {
-    *(.init.data)
+    INIT_DATA
     . = ALIGN(0x4);
     __tagtable_begin = .;
     *(.taglist)
@@ -278,8 +278,9 @@ SECTIONS
   /* Sections to be discarded */
   /DISCARD/ :
   {
-  	*(.exit.literal .exit.text)
-  	*(.exit.data)
+	*(.exit.literal)
+	EXIT_TEXT
+	EXIT_DATA
         *(.exitcall.exit)
   }
 
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 9f584cc5c5fb..ae0166e83490 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -183,6 +183,13 @@
 		*(.kprobes.text)					\
 		VMLINUX_SYMBOL(__kprobes_text_end) = .;
 
+/* init and exit section handling */
+#define INIT_TEXT *(.init.text)
+#define INIT_DATA *(.init.data)
+#define EXIT_TEXT *(.exit.text)
+#define EXIT_DATA *(.exit.data)
+
+
 		/* DWARF debug sections.
 		Symbols in the DWARF debugging sections are relative to
 		the beginning of the section so we begin them at 0.  */
-- 
cgit v1.2.3


From f3fe866d59d707c7a2bba0b23add078e19edb3dc Mon Sep 17 00:00:00 2001
From: Sam Ravnborg <sam@ravnborg.org>
Date: Sun, 20 Jan 2008 18:54:48 +0100
Subject: compiler.h: introduce __section()

Add a new helper: __section() that makes a section definition
much shorter and more readable.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/linux/compiler.h |  5 +++++
 include/linux/init.h     | 26 +++++++++++++-------------
 2 files changed, 18 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index c68b67b86ef1..e0114a61268f 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -175,4 +175,9 @@ extern void __chk_io_ptr(const volatile void __iomem *);
 #define __cold
 #endif
 
+/* Simple shorthand for a section definition */
+#ifndef __section
+# define __section(S) __attribute__ ((__section__(#S)))
+#endif
+
 #endif /* __LINUX_COMPILER_H */
diff --git a/include/linux/init.h b/include/linux/init.h
index 5141381a7527..998076818402 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -40,10 +40,10 @@
 
 /* These are for everybody (although not all archs will actually
    discard it in modules) */
-#define __init		__attribute__ ((__section__ (".init.text"))) __cold
-#define __initdata	__attribute__ ((__section__ (".init.data")))
-#define __exitdata	__attribute__ ((__section__(".exit.data")))
-#define __exit_call	__attribute_used__ __attribute__ ((__section__ (".exitcall.exit")))
+#define __init		__section(.init.text) __cold
+#define __initdata	__section(.init.data)
+#define __exitdata	__section(.exit.data)
+#define __exit_call	__attribute_used__ __section(.exitcall.exit)
 
 /* modpost check for section mismatches during the kernel build.
  * A section mismatch happens when there are references from a
@@ -55,14 +55,14 @@
  * the init/exit section (code or data) is valid and will teach modpost
  * not to issue a warning.
  * The markers follow same syntax rules as __init / __initdata. */
-#define __init_refok     noinline __attribute__ ((__section__ (".text.init.refok")))
-#define __initdata_refok          __attribute__ ((__section__ (".data.init.refok")))
-#define __exit_refok     noinline __attribute__ ((__section__ (".exit.text.refok")))
+#define __init_refok     noinline __section(.text.init.refok)
+#define __initdata_refok          __section(.data.init.refok)
+#define __exit_refok     noinline __section(.exit.text.refok)
 
 #ifdef MODULE
-#define __exit		__attribute__ ((__section__(".exit.text"))) __cold
+#define __exit		__section(.exit.text) __cold
 #else
-#define __exit		__attribute_used__ __attribute__ ((__section__(".exit.text"))) __cold
+#define __exit		__attribute_used__ __section(.exit.text) __cold
 #endif
 
 /* For assembly routines */
@@ -142,11 +142,11 @@ void prepare_namespace(void);
 
 #define console_initcall(fn) \
 	static initcall_t __initcall_##fn \
-	__attribute_used__ __attribute__((__section__(".con_initcall.init")))=fn
+	__attribute_used__ __section(.con_initcall.init)=fn
 
 #define security_initcall(fn) \
 	static initcall_t __initcall_##fn \
-	__attribute_used__ __attribute__((__section__(".security_initcall.init"))) = fn
+	__attribute_used__ __section(.security_initcall.init) = fn
 
 struct obs_kernel_param {
 	const char *str;
@@ -164,7 +164,7 @@ struct obs_kernel_param {
 	static char __setup_str_##unique_id[] __initdata __aligned(1) = str; \
 	static struct obs_kernel_param __setup_##unique_id	\
 		__attribute_used__				\
-		__attribute__((__section__(".init.setup")))	\
+		__section(.init.setup)				\
 		__attribute__((aligned((sizeof(long)))))	\
 		= { __setup_str_##unique_id, fn, early }
 
@@ -242,7 +242,7 @@ void __init parse_early_param(void);
 #endif
 
 /* Data marked not to be saved by software suspend */
-#define __nosavedata __attribute__ ((__section__ (".data.nosave")))
+#define __nosavedata __section(.data.nosave)
 
 /* This means "can be init if no module support, otherwise module load
    may call it." */
-- 
cgit v1.2.3


From eb8f689046b857874e964463619f09df06d59fad Mon Sep 17 00:00:00 2001
From: Sam Ravnborg <sam@ravnborg.org>
Date: Sun, 20 Jan 2008 20:07:28 +0100
Subject: Use separate sections for __dev/__cpu/__mem code/data

Introducing separate sections for __dev* (HOTPLUG),
__cpu* (HOTPLUG_CPU) and __mem* (MEMORY_HOTPLUG)
allows us to do a much more reliable Section mismatch
check in modpost. We are no longer dependent on the actual
configuration of for example HOTPLUG.

This has the effect that all users see much more
Section mismatch warnings than before because they
were almost all hidden when HOTPLUG was enabled.
The advantage of this is that when building a piece
of code then it is much more likely that the Section
mismatch errors are spotted and the warnings will be
felt less random of nature.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Greg KH <greg@kroah.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Adrian Bunk <bunk@kernel.org>
---
 include/asm-generic/vmlinux.lds.h | 88 ++++++++++++++++++++++++++++++++++++---
 include/linux/init.h              | 77 +++++++++++++++++-----------------
 scripts/mod/modpost.c             | 54 +++++++++++++++++-------
 3 files changed, 159 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index ae0166e83490..e0a56fb8f813 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -9,10 +9,46 @@
 /* Align . to a 8 byte boundary equals to maximum function alignment. */
 #define ALIGN_FUNCTION()  . = ALIGN(8)
 
+/* The actual configuration determine if the init/exit sections
+ * are handled as text/data or they can be discarded (which
+ * often happens at runtime)
+ */
+#ifdef CONFIG_HOTPLUG
+#define DEV_KEEP(sec)    *(.dev##sec)
+#define DEV_DISCARD(sec)
+#else
+#define DEV_KEEP(sec)
+#define DEV_DISCARD(sec) *(.dev##sec)
+#endif
+
+#ifdef CONFIG_HOTPLUG_CPU
+#define CPU_KEEP(sec)    *(.cpu##sec)
+#define CPU_DISCARD(sec)
+#else
+#define CPU_KEEP(sec)
+#define CPU_DISCARD(sec) *(.cpu##sec)
+#endif
+
+#if defined(CONFIG_MEMORY_HOTPLUG) || defined(CONFIG_ACPI_HOTPLUG_MEMORY) \
+	|| defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
+#define MEM_KEEP(sec)    *(.mem##sec)
+#define MEM_DISCARD(sec)
+#else
+#define MEM_KEEP(sec)
+#define MEM_DISCARD(sec) *(.mem##sec)
+#endif
+
+
 /* .data section */
 #define DATA_DATA							\
 	*(.data)							\
 	*(.data.init.refok)						\
+	DEV_KEEP(init.data)						\
+	DEV_KEEP(exit.data)						\
+	CPU_KEEP(init.data)						\
+	CPU_KEEP(exit.data)						\
+	MEM_KEEP(init.data)						\
+	MEM_KEEP(exit.data)						\
 	. = ALIGN(8);							\
 	VMLINUX_SYMBOL(__start___markers) = .;				\
 	*(__markers)							\
@@ -132,6 +168,16 @@
 		*(__ksymtab_strings)					\
 	}								\
 									\
+	/* __*init sections */						\
+	__init_rodata : AT(ADDR(__init_rodata) - LOAD_OFFSET) {		\
+		DEV_KEEP(init.rodata)					\
+		DEV_KEEP(exit.rodata)					\
+		CPU_KEEP(init.rodata)					\
+		CPU_KEEP(exit.rodata)					\
+		MEM_KEEP(init.rodata)					\
+		MEM_KEEP(exit.rodata)					\
+	}								\
+									\
 	/* Built-in module parameters. */				\
 	__param : AT(ADDR(__param) - LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__start___param) = .;			\
@@ -139,7 +185,6 @@
 		VMLINUX_SYMBOL(__stop___param) = .;			\
 		VMLINUX_SYMBOL(__end_rodata) = .;			\
 	}								\
-									\
 	. = ALIGN((align));
 
 /* RODATA provided for backward compatibility.
@@ -159,7 +204,14 @@
 		ALIGN_FUNCTION();					\
 		*(.text)						\
 		*(.text.init.refok)					\
-		*(.exit.text.refok)
+		*(.exit.text.refok)					\
+	DEV_KEEP(init.text)						\
+	DEV_KEEP(exit.text)						\
+	CPU_KEEP(init.text)						\
+	CPU_KEEP(exit.text)						\
+	MEM_KEEP(init.text)						\
+	MEM_KEEP(exit.text)
+
 
 /* sched.text is aling to function alignment to secure we have same
  * address even at second ld pass when generating System.map */
@@ -184,11 +236,35 @@
 		VMLINUX_SYMBOL(__kprobes_text_end) = .;
 
 /* init and exit section handling */
-#define INIT_TEXT *(.init.text)
-#define INIT_DATA *(.init.data)
-#define EXIT_TEXT *(.exit.text)
-#define EXIT_DATA *(.exit.data)
+#define INIT_DATA							\
+	*(.init.data)							\
+	DEV_DISCARD(init.data)						\
+	DEV_DISCARD(init.rodata)					\
+	CPU_DISCARD(init.data)						\
+	CPU_DISCARD(init.rodata)					\
+	MEM_DISCARD(init.data)						\
+	MEM_DISCARD(init.rodata)
+
+#define INIT_TEXT							\
+	*(.init.text)							\
+	DEV_DISCARD(init.text)						\
+	CPU_DISCARD(init.text)						\
+	MEM_DISCARD(init.text)
+
+#define EXIT_DATA							\
+	*(.exit.data)							\
+	DEV_DISCARD(exit.data)						\
+	DEV_DISCARD(exit.rodata)					\
+	CPU_DISCARD(exit.data)						\
+	CPU_DISCARD(exit.rodata)					\
+	MEM_DISCARD(exit.data)						\
+	MEM_DISCARD(exit.rodata)
 
+#define EXIT_TEXT							\
+	*(.exit.text)							\
+	DEV_DISCARD(exit.text)						\
+	CPU_DISCARD(exit.text)						\
+	MEM_DISCARD(exit.text)
 
 		/* DWARF debug sections.
 		Symbols in the DWARF debugging sections are relative to
diff --git a/include/linux/init.h b/include/linux/init.h
index 998076818402..dcb66c76bd48 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -60,18 +60,54 @@
 #define __exit_refok     noinline __section(.exit.text.refok)
 
 #ifdef MODULE
-#define __exit		__section(.exit.text) __cold
+#define __exitused
 #else
-#define __exit		__attribute_used__ __section(.exit.text) __cold
+#define __exitused  __used
 #endif
 
+#define __exit          __section(.exit.text) __exitused __cold
+
+/* Used for HOTPLUG */
+#define __devinit        __section(.devinit.text) __cold
+#define __devinitdata    __section(.devinit.data)
+#define __devinitconst   __section(.devinit.rodata)
+#define __devexit        __section(.devexit.text) __exitused __cold
+#define __devexitdata    __section(.devexit.data)
+#define __devexitconst   __section(.devexit.rodata)
+
+/* Used for HOTPLUG_CPU */
+#define __cpuinit        __section(.cpuinit.text) __cold
+#define __cpuinitdata    __section(.cpuinit.data)
+#define __cpuinitconst   __section(.cpuinit.rodata)
+#define __cpuexit        __section(.cpuexit.text) __exitused __cold
+#define __cpuexitdata    __section(.cpuexit.data)
+#define __cpuexitconst   __section(.cpuexit.rodata)
+
+/* Used for MEMORY_HOTPLUG */
+#define __meminit        __section(.meminit.text) __cold
+#define __meminitdata    __section(.meminit.data)
+#define __meminitconst   __section(.meminit.rodata)
+#define __memexit        __section(.memexit.text) __exitused __cold
+#define __memexitdata    __section(.memexit.data)
+#define __memexitconst   __section(.memexit.rodata)
+
 /* For assembly routines */
 #define __INIT		.section	".init.text","ax"
 #define __INIT_REFOK	.section	".text.init.refok","ax"
 #define __FINIT		.previous
+
 #define __INITDATA	.section	".init.data","aw"
 #define __INITDATA_REFOK .section	".data.init.refok","aw"
 
+#define __DEVINIT        .section	".devinit.text", "ax"
+#define __DEVINITDATA    .section	".devinit.data", "aw"
+
+#define __CPUINIT        .section	".cpuinit.text", "ax"
+#define __CPUINITDATA    .section	".cpuinit.data", "aw"
+
+#define __MEMINIT        .section	".meminit.text", "ax"
+#define __MEMINITDATA    .section	".meminit.data", "aw"
+
 #ifndef __ASSEMBLY__
 /*
  * Used for initialization calls..
@@ -254,43 +290,6 @@ void __init parse_early_param(void);
 #define __initdata_or_module __initdata
 #endif /*CONFIG_MODULES*/
 
-#ifdef CONFIG_HOTPLUG
-#define __devinit
-#define __devinitdata
-#define __devexit
-#define __devexitdata
-#else
-#define __devinit __init
-#define __devinitdata __initdata
-#define __devexit __exit
-#define __devexitdata __exitdata
-#endif
-
-#ifdef CONFIG_HOTPLUG_CPU
-#define __cpuinit
-#define __cpuinitdata
-#define __cpuexit
-#define __cpuexitdata
-#else
-#define __cpuinit	__init
-#define __cpuinitdata __initdata
-#define __cpuexit __exit
-#define __cpuexitdata	__exitdata
-#endif
-
-#if defined(CONFIG_MEMORY_HOTPLUG) || defined(CONFIG_ACPI_HOTPLUG_MEMORY) \
-	|| defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
-#define __meminit
-#define __meminitdata
-#define __memexit
-#define __memexitdata
-#else
-#define __meminit	__init
-#define __meminitdata __initdata
-#define __memexit __exit
-#define __memexitdata	__exitdata
-#endif
-
 /* Functions marked as __devexit may be discarded at kernel link time, depending
    on config options.  Newer versions of binutils detect references from
    retained sections to discarded sections and flag an error.  Pointers to
diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
index 986513dcd700..730b321680cd 100644
--- a/scripts/mod/modpost.c
+++ b/scripts/mod/modpost.c
@@ -670,27 +670,41 @@ int match(const char *sym, const char * const pat[])
 static const char *section_white_list[] =
 	{ ".debug*", ".stab*", ".note*", ".got*", ".toc*", NULL };
 
-#define INIT_DATA_SECTIONS ".init.data$"
-#define EXIT_DATA_SECTIONS ".exit.data$"
+#define ALL_INIT_DATA_SECTIONS \
+	".init.data$", ".devinit.data$", ".cpuinit.data$", ".meminit.data$"
+#define ALL_EXIT_DATA_SECTIONS \
+	".exit.data$", ".devexit.data$", ".cpuexit.data$", ".memexit.data$"
 
-#define INIT_TEXT_SECTIONS ".init.text$"
-#define EXIT_TEXT_SECTIONS ".exit.text$"
+#define ALL_INIT_TEXT_SECTIONS \
+	".init.text$", ".devinit.text$", ".cpuinit.text$", ".meminit.text$"
+#define ALL_EXIT_TEXT_SECTIONS \
+	".exit.text$", ".devexit.text$", ".cpuexit.text$", ".memexit.text$"
 
-#define INIT_SECTIONS INIT_DATA_SECTIONS, INIT_TEXT_SECTIONS
-#define EXIT_SECTIONS EXIT_DATA_SECTIONS, EXIT_TEXT_SECTIONS
+#define ALL_INIT_SECTIONS ALL_INIT_DATA_SECTIONS, ALL_INIT_TEXT_SECTIONS
+#define ALL_EXIT_SECTIONS ALL_EXIT_DATA_SECTIONS, ALL_EXIT_TEXT_SECTIONS
 
 #define DATA_SECTIONS ".data$", ".data.rel$"
 #define TEXT_SECTIONS ".text$"
 
+#define INIT_SECTIONS      ".init.data$", ".init.text$"
+#define DEV_INIT_SECTIONS  ".devinit.data$", ".devinit.text$"
+#define CPU_INIT_SECTIONS  ".cpuinit.data$", ".cpuinit.text$"
+#define MEM_INIT_SECTIONS  ".meminit.data$", ".meminit.text$"
+
+#define EXIT_SECTIONS      ".exit.data$", ".exit.text$"
+#define DEV_EXIT_SECTIONS  ".devexit.data$", ".devexit.text$"
+#define CPU_EXIT_SECTIONS  ".cpuexit.data$", ".cpuexit.text$"
+#define MEM_EXIT_SECTIONS  ".memexit.data$", ".memexit.text$"
+
 /* init data sections */
-static const char *init_data_sections[] = { INIT_DATA_SECTIONS, NULL };
+static const char *init_data_sections[] = { ALL_INIT_DATA_SECTIONS, NULL };
 
 /* all init sections */
-static const char *init_sections[] = { INIT_SECTIONS, NULL };
+static const char *init_sections[] = { ALL_INIT_SECTIONS, NULL };
 
 /* All init and exit sections (code + data) */
 static const char *init_exit_sections[] =
-	{INIT_SECTIONS, EXIT_SECTIONS, NULL };
+	{ALL_INIT_SECTIONS, ALL_EXIT_SECTIONS, NULL };
 
 /* data section */
 static const char *data_sections[] = { DATA_SECTIONS, NULL };
@@ -734,22 +748,32 @@ const struct sectioncheck sectioncheck[] = {
  */
 {
 	.fromsec = { TEXT_SECTIONS, DATA_SECTIONS, NULL },
-	.tosec   = { INIT_SECTIONS, EXIT_SECTIONS, NULL }
+	.tosec   = { ALL_INIT_SECTIONS, ALL_EXIT_SECTIONS, NULL }
+},
+/* Do not reference init code/data from devinit/cpuinit/meminit code/data */
+{
+	.fromsec = { DEV_INIT_SECTIONS, CPU_INIT_SECTIONS, MEM_INIT_SECTIONS, NULL },
+	.tosec   = { INIT_SECTIONS, NULL }
+},
+/* Do not reference exit code/data from devexit/cpuexit/memexit code/data */
+{
+	.fromsec = { DEV_EXIT_SECTIONS, CPU_EXIT_SECTIONS, MEM_EXIT_SECTIONS, NULL },
+	.tosec   = { EXIT_SECTIONS, NULL }
 },
 /* Do not use exit code/data from init code */
 {
-	.fromsec = { INIT_SECTIONS, NULL },
-	.tosec   = { EXIT_SECTIONS, NULL },
+	.fromsec = { ALL_INIT_SECTIONS, NULL },
+	.tosec   = { ALL_EXIT_SECTIONS, NULL },
 },
 /* Do not use init code/data from exit code */
 {
-	.fromsec = { EXIT_SECTIONS, NULL },
-	.tosec   = { INIT_SECTIONS, NULL }
+	.fromsec = { ALL_EXIT_SECTIONS, NULL },
+	.tosec   = { ALL_INIT_SECTIONS, NULL }
 },
 /* Do not export init/exit functions or data */
 {
 	.fromsec = { "__ksymtab*", NULL },
-	.tosec   = { INIT_SECTIONS, EXIT_SECTIONS, NULL }
+	.tosec   = { ALL_INIT_SECTIONS, ALL_EXIT_SECTIONS, NULL }
 }
 };
 
-- 
cgit v1.2.3


From 3ff6eecca4e5c49a5d1dd8b58ea0e20102ce08f0 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Thu, 24 Jan 2008 22:16:20 +0100
Subject: remove __attribute_used__

Remove the deprecated __attribute_used__.

[Introduce __section in a few places to silence checkpatch /sam]

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 arch/alpha/lib/dec_and_lock.c           |  3 +--
 arch/powerpc/boot/Makefile              |  2 +-
 arch/powerpc/kernel/sysfs.c             |  2 +-
 arch/powerpc/oprofile/op_model_power4.c |  6 +++---
 arch/sparc64/kernel/unaligned.c         |  2 +-
 arch/um/include/init.h                  | 26 +++++++++++++-------------
 drivers/rapidio/rio.h                   |  4 ++--
 fs/compat_ioctl.c                       |  2 +-
 include/asm-avr32/setup.h               |  2 +-
 include/asm-ia64/gcc_intrin.h           |  2 +-
 include/asm-sh/machvec.h                |  2 +-
 include/asm-sh/thread_info.h            |  2 +-
 include/asm-x86/thread_info_32.h        |  2 +-
 include/linux/compiler-gcc3.h           |  2 --
 include/linux/compiler-gcc4.h           |  1 -
 include/linux/compiler.h                |  4 ----
 include/linux/elfnote.h                 |  2 +-
 include/linux/init.h                    | 11 +++++------
 include/linux/module.h                  |  4 ++--
 include/linux/moduleparam.h             |  4 ++--
 include/linux/pci.h                     |  2 +-
 scripts/mod/modpost.c                   |  4 ++--
 22 files changed, 41 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/lib/dec_and_lock.c b/arch/alpha/lib/dec_and_lock.c
index 6ae2500a9d9e..0f5520d2f45f 100644
--- a/arch/alpha/lib/dec_and_lock.c
+++ b/arch/alpha/lib/dec_and_lock.c
@@ -30,8 +30,7 @@ _atomic_dec_and_lock:				\n\
 	.previous				\n\
 	.end _atomic_dec_and_lock");
 
-static int __attribute_used__
-atomic_dec_and_lock_1(atomic_t *atomic, spinlock_t *lock)
+static int __used atomic_dec_and_lock_1(atomic_t *atomic, spinlock_t *lock)
 {
 	/* Slow path */
 	spin_lock(lock);
diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index 18e32719d0ed..4b1d98b8135e 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -65,7 +65,7 @@ obj-wlib := $(addsuffix .o, $(basename $(addprefix $(obj)/, $(src-wlib))))
 obj-plat := $(addsuffix .o, $(basename $(addprefix $(obj)/, $(src-plat))))
 
 quiet_cmd_copy_zlib = COPY    $@
-      cmd_copy_zlib = sed "s@__attribute_used__@@;s@<linux/\([^>]*\).*@\"\1\"@" $< > $@
+      cmd_copy_zlib = sed "s@__used@@;s@<linux/\([^>]*\).*@\"\1\"@" $< > $@
 
 quiet_cmd_copy_zlibheader = COPY    $@
       cmd_copy_zlibheader = sed "s@<linux/\([^>]*\).*@\"\1\"@" $< > $@
diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c
index 25d9a96484dd..c8127f832df0 100644
--- a/arch/powerpc/kernel/sysfs.c
+++ b/arch/powerpc/kernel/sysfs.c
@@ -158,7 +158,7 @@ static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
 	unsigned long val = run_on_cpu(cpu->sysdev.id, read_##NAME, 0); \
 	return sprintf(buf, "%lx\n", val); \
 } \
-static ssize_t __attribute_used__ \
+static ssize_t __used \
 	store_##NAME(struct sys_device *dev, const char *buf, size_t count) \
 { \
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev); \
diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c
index cddc250a6a5c..446a8bbb847b 100644
--- a/arch/powerpc/oprofile/op_model_power4.c
+++ b/arch/powerpc/oprofile/op_model_power4.c
@@ -172,15 +172,15 @@ static void power4_stop(void)
 }
 
 /* Fake functions used by canonicalize_pc */
-static void __attribute_used__ hypervisor_bucket(void)
+static void __used hypervisor_bucket(void)
 {
 }
 
-static void __attribute_used__ rtas_bucket(void)
+static void __used rtas_bucket(void)
 {
 }
 
-static void __attribute_used__ kernel_unknown_bucket(void)
+static void __used kernel_unknown_bucket(void)
 {
 }
 
diff --git a/arch/sparc64/kernel/unaligned.c b/arch/sparc64/kernel/unaligned.c
index 953be816fa25..dc7bf1b6321c 100644
--- a/arch/sparc64/kernel/unaligned.c
+++ b/arch/sparc64/kernel/unaligned.c
@@ -175,7 +175,7 @@ unsigned long compute_effective_address(struct pt_regs *regs,
 }
 
 /* This is just to make gcc think die_if_kernel does return... */
-static void __attribute_used__ unaligned_panic(char *str, struct pt_regs *regs)
+static void __used unaligned_panic(char *str, struct pt_regs *regs)
 {
 	die_if_kernel(str, regs);
 }
diff --git a/arch/um/include/init.h b/arch/um/include/init.h
index d4de7c0120ce..cebc6cae9190 100644
--- a/arch/um/include/init.h
+++ b/arch/um/include/init.h
@@ -42,15 +42,15 @@ typedef void (*exitcall_t)(void);
 
 /* These are for everybody (although not all archs will actually
    discard it in modules) */
-#define __init		__attribute__ ((__section__ (".init.text")))
-#define __initdata	__attribute__ ((__section__ (".init.data")))
-#define __exitdata	__attribute__ ((__section__(".exit.data")))
-#define __exit_call	__attribute_used__ __attribute__ ((__section__ (".exitcall.exit")))
+#define __init		__section(.init.text)
+#define __initdata	__section(.init.data)
+#define __exitdata	__section(.exit.data)
+#define __exit_call	__used __section(.exitcall.exit)
 
 #ifdef MODULE
-#define __exit		__attribute__ ((__section__(".exit.text")))
+#define __exit		__section(.exit.text)
 #else
-#define __exit		__attribute_used__ __attribute__ ((__section__(".exit.text")))
+#define __exit		__used __section(.exit.text)
 #endif
 
 #endif
@@ -103,16 +103,16 @@ extern struct uml_param __uml_setup_start, __uml_setup_end;
  * Mark functions and data as being only used at initialization
  * or exit time.
  */
-#define __uml_init_setup	__attribute_used__ __attribute__ ((__section__ (".uml.setup.init")))
-#define __uml_setup_help	__attribute_used__ __attribute__ ((__section__ (".uml.help.init")))
-#define __uml_init_call		__attribute_used__ __attribute__ ((__section__ (".uml.initcall.init")))
-#define __uml_postsetup_call	__attribute_used__ __attribute__ ((__section__ (".uml.postsetup.init")))
-#define __uml_exit_call		__attribute_used__ __attribute__ ((__section__ (".uml.exitcall.exit")))
+#define __uml_init_setup	__used __section(.uml.setup.init)
+#define __uml_setup_help	__used __section(.uml.help.init)
+#define __uml_init_call		__used __section(.uml.initcall.init)
+#define __uml_postsetup_call	__used __section(.uml.postsetup.init)
+#define __uml_exit_call		__used __section(.uml.exitcall.exit)
 
 #ifndef __KERNEL__
 
 #define __define_initcall(level,fn) \
-	static initcall_t __initcall_##fn __attribute_used__ \
+	static initcall_t __initcall_##fn __used \
 	__attribute__((__section__(".initcall" level ".init"))) = fn
 
 /* Userspace initcalls shouldn't depend on anything in the kernel, so we'll
@@ -122,7 +122,7 @@ extern struct uml_param __uml_setup_start, __uml_setup_end;
 
 #define __exitcall(fn) static exitcall_t __exitcall_##fn __exit_call = fn
 
-#define __init_call	__attribute_used__ __attribute__ ((__section__ (".initcall.init")))
+#define __init_call	__used __section(.initcall.init)
 
 #endif
 
diff --git a/drivers/rapidio/rio.h b/drivers/rapidio/rio.h
index b242cee656e7..80e3f03b5041 100644
--- a/drivers/rapidio/rio.h
+++ b/drivers/rapidio/rio.h
@@ -31,8 +31,8 @@ extern struct rio_route_ops __end_rio_route_ops[];
 
 /* Helpers internal to the RIO core code */
 #define DECLARE_RIO_ROUTE_SECTION(section, vid, did, add_hook, get_hook)  \
-        static struct rio_route_ops __rio_route_ops __attribute_used__   \
-	        __attribute__((__section__(#section))) = { vid, did, add_hook, get_hook };
+	static struct rio_route_ops __rio_route_ops __used   \
+	__section(section)= { vid, did, add_hook, get_hook };
 
 /**
  * DECLARE_RIO_ROUTE_OPS - Registers switch routing operations
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index da8cb3b3592c..ffdc022cae64 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1376,7 +1376,7 @@ static int do_atm_ioctl(unsigned int fd, unsigned int cmd32, unsigned long arg)
         return -EINVAL;
 }
 
-static __attribute_used__ int 
+static __used int
 ret_einval(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
 	return -EINVAL;
diff --git a/include/asm-avr32/setup.h b/include/asm-avr32/setup.h
index b0828d43e110..ea3070ff13a5 100644
--- a/include/asm-avr32/setup.h
+++ b/include/asm-avr32/setup.h
@@ -110,7 +110,7 @@ struct tagtable {
 	int	(*parse)(struct tag *);
 };
 
-#define __tag __attribute_used__ __attribute__((__section__(".taglist.init")))
+#define __tag __used __attribute__((__section__(".taglist.init")))
 #define __tagtable(tag, fn)						\
 	static struct tagtable __tagtable_##fn __tag = { tag, fn }
 
diff --git a/include/asm-ia64/gcc_intrin.h b/include/asm-ia64/gcc_intrin.h
index e58d3298fa10..5b6665c754c9 100644
--- a/include/asm-ia64/gcc_intrin.h
+++ b/include/asm-ia64/gcc_intrin.h
@@ -24,7 +24,7 @@
 extern void ia64_bad_param_for_setreg (void);
 extern void ia64_bad_param_for_getreg (void);
 
-register unsigned long ia64_r13 asm ("r13") __attribute_used__;
+register unsigned long ia64_r13 asm ("r13") __used;
 
 #define ia64_setreg(regnum, val)						\
 ({										\
diff --git a/include/asm-sh/machvec.h b/include/asm-sh/machvec.h
index ddb18ad23303..b2e4124070ae 100644
--- a/include/asm-sh/machvec.h
+++ b/include/asm-sh/machvec.h
@@ -65,6 +65,6 @@ extern struct sh_machine_vector sh_mv;
 #define get_system_type()	sh_mv.mv_name
 
 #define __initmv \
-	__attribute_used__ __attribute__((__section__ (".machvec.init")))
+	__used __section(.machvec.init)
 
 #endif /* _ASM_SH_MACHVEC_H */
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index c6577d3dc46d..c50e5d35fe84 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -68,7 +68,7 @@ struct thread_info {
 #define init_stack		(init_thread_union.stack)
 
 /* how to get the current stack pointer from C */
-register unsigned long current_stack_pointer asm("r15") __attribute_used__;
+register unsigned long current_stack_pointer asm("r15") __used;
 
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index ef58fd2a6eb0..a516e9192f11 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -85,7 +85,7 @@ struct thread_info {
 
 
 /* how to get the current stack pointer from C */
-register unsigned long current_stack_pointer asm("esp") __attribute_used__;
+register unsigned long current_stack_pointer asm("esp") __used;
 
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
diff --git a/include/linux/compiler-gcc3.h b/include/linux/compiler-gcc3.h
index 2d8c0f48f55e..e5eb795f78a1 100644
--- a/include/linux/compiler-gcc3.h
+++ b/include/linux/compiler-gcc3.h
@@ -7,10 +7,8 @@
 
 #if __GNUC_MINOR__ >= 3
 # define __used			__attribute__((__used__))
-# define __attribute_used__	__used				/* deprecated */
 #else
 # define __used			__attribute__((__unused__))
-# define __attribute_used__	__used				/* deprecated */
 #endif
 
 #if __GNUC_MINOR__ >= 4
diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h
index ee7ca5de970c..0ab3a3232330 100644
--- a/include/linux/compiler-gcc4.h
+++ b/include/linux/compiler-gcc4.h
@@ -15,7 +15,6 @@
 #endif
 
 #define __used			__attribute__((__used__))
-#define __attribute_used__	__used			/* deprecated */
 #define __must_check 		__attribute__((warn_unused_result))
 #define __compiler_offsetof(a,b) __builtin_offsetof(a,b)
 #define __always_inline		inline __attribute__((always_inline))
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index e0114a61268f..d0e17e1657dc 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -126,10 +126,6 @@ extern void __chk_io_ptr(const volatile void __iomem *);
  * Mark functions that are referenced only in inline assembly as __used so
  * the code is emitted even though it appears to be unreferenced.
  */
-#ifndef __attribute_used__
-# define __attribute_used__	/* deprecated */
-#endif
-
 #ifndef __used
 # define __used			/* unimplemented */
 #endif
diff --git a/include/linux/elfnote.h b/include/linux/elfnote.h
index e831759b2fb5..278e3ef05336 100644
--- a/include/linux/elfnote.h
+++ b/include/linux/elfnote.h
@@ -76,7 +76,7 @@
 		typeof(desc) _desc					\
 			     __attribute__((aligned(sizeof(Elf##size##_Word)))); \
 	} _ELFNOTE_PASTE(_note_, unique)				\
-		__attribute_used__					\
+		__used							\
 		__attribute__((section(".note." name),			\
 			       aligned(sizeof(Elf##size##_Word)),	\
 			       unused)) = {				\
diff --git a/include/linux/init.h b/include/linux/init.h
index dcb66c76bd48..dde1eaa7766b 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -43,7 +43,7 @@
 #define __init		__section(.init.text) __cold
 #define __initdata	__section(.init.data)
 #define __exitdata	__section(.exit.data)
-#define __exit_call	__attribute_used__ __section(.exitcall.exit)
+#define __exit_call	__used __section(.exitcall.exit)
 
 /* modpost check for section mismatches during the kernel build.
  * A section mismatch happens when there are references from a
@@ -144,7 +144,7 @@ void prepare_namespace(void);
  */
 
 #define __define_initcall(level,fn,id) \
-	static initcall_t __initcall_##fn##id __attribute_used__ \
+	static initcall_t __initcall_##fn##id __used \
 	__attribute__((__section__(".initcall" level ".init"))) = fn
 
 /*
@@ -178,11 +178,11 @@ void prepare_namespace(void);
 
 #define console_initcall(fn) \
 	static initcall_t __initcall_##fn \
-	__attribute_used__ __section(.con_initcall.init)=fn
+	__used __section(.con_initcall.init) = fn
 
 #define security_initcall(fn) \
 	static initcall_t __initcall_##fn \
-	__attribute_used__ __section(.security_initcall.init) = fn
+	__used __section(.security_initcall.init) = fn
 
 struct obs_kernel_param {
 	const char *str;
@@ -199,8 +199,7 @@ struct obs_kernel_param {
 #define __setup_param(str, unique_id, fn, early)			\
 	static char __setup_str_##unique_id[] __initdata __aligned(1) = str; \
 	static struct obs_kernel_param __setup_##unique_id	\
-		__attribute_used__				\
-		__section(.init.setup)				\
+		__used __section(.init.setup)			\
 		__attribute__((aligned((sizeof(long)))))	\
 		= { __setup_str_##unique_id, fn, early }
 
diff --git a/include/linux/module.h b/include/linux/module.h
index c97bdb7eb957..404838184ea5 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -178,7 +178,7 @@ void *__symbol_get_gpl(const char *symbol);
 #define __CRC_SYMBOL(sym, sec)					\
 	extern void *__crc_##sym __attribute__((weak));		\
 	static const unsigned long __kcrctab_##sym		\
-	__attribute_used__					\
+	__used							\
 	__attribute__((section("__kcrctab" sec), unused))	\
 	= (unsigned long) &__crc_##sym;
 #else
@@ -193,7 +193,7 @@ void *__symbol_get_gpl(const char *symbol);
 	__attribute__((section("__ksymtab_strings")))		\
 	= MODULE_SYMBOL_PREFIX #sym;                    	\
 	static const struct kernel_symbol __ksymtab_##sym	\
-	__attribute_used__					\
+	__used							\
 	__attribute__((section("__ksymtab" sec), unused))	\
 	= { (unsigned long)&sym, __kstrtab_##sym }
 
diff --git a/include/linux/moduleparam.h b/include/linux/moduleparam.h
index 13410b20600f..8126e55c5bdc 100644
--- a/include/linux/moduleparam.h
+++ b/include/linux/moduleparam.h
@@ -18,7 +18,7 @@
 #define __module_cat(a,b) ___module_cat(a,b)
 #define __MODULE_INFO(tag, name, info)					  \
 static const char __module_cat(name,__LINE__)[]				  \
-  __attribute_used__							  \
+  __used								  \
   __attribute__((section(".modinfo"),unused)) = __stringify(tag) "=" info
 #else  /* !MODULE */
 #define __MODULE_INFO(tag, name, info)
@@ -72,7 +72,7 @@ struct kparam_array
 	BUILD_BUG_ON_ZERO((perm) < 0 || (perm) > 0777 || ((perm) & 2));	\
 	static const char __param_str_##name[] = prefix #name;		\
 	static struct kernel_param const __param_##name			\
-	__attribute_used__						\
+	__used								\
     __attribute__ ((unused,__section__ ("__param"),aligned(sizeof(void *)))) \
 	= { __param_str_##name, perm, set, get, { arg } }
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0dd93bb62fbe..ae1006322f80 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -867,7 +867,7 @@ enum pci_fixup_pass {
 
 /* Anonymous variables would be nice... */
 #define DECLARE_PCI_FIXUP_SECTION(section, name, vendor, device, hook)	\
-	static const struct pci_fixup __pci_fixup_##name __attribute_used__ \
+	static const struct pci_fixup __pci_fixup_##name __used		\
 	__attribute__((__section__(#section))) = { vendor, device, hook };
 #define DECLARE_PCI_FIXUP_EARLY(vendor, device, hook)			\
 	DECLARE_PCI_FIXUP_SECTION(.pci_fixup_early,			\
diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
index 0a80acafd212..e75739ec9c03 100644
--- a/scripts/mod/modpost.c
+++ b/scripts/mod/modpost.c
@@ -1445,7 +1445,7 @@ static int add_versions(struct buffer *b, struct module *mod)
 
 	buf_printf(b, "\n");
 	buf_printf(b, "static const struct modversion_info ____versions[]\n");
-	buf_printf(b, "__attribute_used__\n");
+	buf_printf(b, "__used\n");
 	buf_printf(b, "__attribute__((section(\"__versions\"))) = {\n");
 
 	for (s = mod->unres; s; s = s->next) {
@@ -1476,7 +1476,7 @@ static void add_depends(struct buffer *b, struct module *mod,
 
 	buf_printf(b, "\n");
 	buf_printf(b, "static const char __module_depends[]\n");
-	buf_printf(b, "__attribute_used__\n");
+	buf_printf(b, "__used\n");
 	buf_printf(b, "__attribute__((section(\".modinfo\"))) =\n");
 	buf_printf(b, "\"depends=");
 	for (s = mod->unres; s; s = s->next) {
-- 
cgit v1.2.3


From 1a3fb6d481689d0482eacadcbe3205b49b423c11 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Thu, 24 Jan 2008 22:20:18 +0100
Subject: asm-generic/vmlix.lds.h: simplify __mem{init,exit}* dependencies

Simplify the dependencies on __mem{init,exit}* (ACPI_HOTPLUG_MEMORY requires
MEMORY_HOTPLUG).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/asm-generic/vmlinux.lds.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index e0a56fb8f813..294853053707 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -29,8 +29,7 @@
 #define CPU_DISCARD(sec) *(.cpu##sec)
 #endif
 
-#if defined(CONFIG_MEMORY_HOTPLUG) || defined(CONFIG_ACPI_HOTPLUG_MEMORY) \
-	|| defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
+#if defined(CONFIG_MEMORY_HOTPLUG)
 #define MEM_KEEP(sec)    *(.mem##sec)
 #define MEM_DISCARD(sec)
 #else
-- 
cgit v1.2.3


From 312b1485fb509c9bc32eda28ad29537896658cb8 Mon Sep 17 00:00:00 2001
From: Sam Ravnborg <sam@ravnborg.org>
Date: Mon, 28 Jan 2008 20:21:15 +0100
Subject: Introduce new section reference annotations tags: __ref, __refdata,
 __refconst

Today we have the following annotations for functions/data
referencing __init/__exit functions / data:

__init_refok     => for init functions
__initdata_refok => for init data
__exit_refok     => for exit functions

There is really no difference between the __init and __exit
versions and simplify it and to introduce a shorter annotation
the following new annotations are introduced:

__ref      => for functions (code) that
              references __*init / __*exit
__refdata  => for variables
__refconst => for const variables

Whit this annotation is it more obvious what the annotation
is for and there is no longer the arbitary division
between __init and __exit code.

The mechanishm is the same as before - a special section
is created which is made part of the usual sections
in the linker script.

We will start to see annotations like this:

-static struct pci_serial_quirk pci_serial_quirks[] = {
+static const struct pci_serial_quirk pci_serial_quirks[] __refconst = {
-----------------
-static struct notifier_block __cpuinitdata cpuid_class_cpu_notifier =
+static struct notifier_block cpuid_class_cpu_notifier __refdata =
----------------
-static int threshold_cpu_callback(struct notifier_block *nfb,
+static int __ref threshold_cpu_callback(struct notifier_block *nfb,

[The above is just random samples].

Note: No modifications were needed in modpost
to support the new sections due to the newly introduced
blacklisting.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/asm-generic/vmlinux.lds.h |  3 +++
 include/linux/init.h              | 34 +++++++++++++++++++++++++++-------
 2 files changed, 30 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 294853053707..76df771be585 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -42,6 +42,7 @@
 #define DATA_DATA							\
 	*(.data)							\
 	*(.data.init.refok)						\
+	*(.ref.data)							\
 	DEV_KEEP(init.data)						\
 	DEV_KEEP(exit.data)						\
 	CPU_KEEP(init.data)						\
@@ -169,6 +170,7 @@
 									\
 	/* __*init sections */						\
 	__init_rodata : AT(ADDR(__init_rodata) - LOAD_OFFSET) {		\
+		*(.ref.rodata)						\
 		DEV_KEEP(init.rodata)					\
 		DEV_KEEP(exit.rodata)					\
 		CPU_KEEP(init.rodata)					\
@@ -202,6 +204,7 @@
 #define TEXT_TEXT							\
 		ALIGN_FUNCTION();					\
 		*(.text)						\
+		*(.ref.text)						\
 		*(.text.init.refok)					\
 		*(.exit.text.refok)					\
 	DEV_KEEP(init.text)						\
diff --git a/include/linux/init.h b/include/linux/init.h
index dde1eaa7766b..2efbda016741 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -52,12 +52,26 @@
  * when early init has completed so all such references are potential bugs.
  * For exit sections the same issue exists.
  * The following markers are used for the cases where the reference to
- * the init/exit section (code or data) is valid and will teach modpost
- * not to issue a warning.
+ * the *init / *exit section (code or data) is valid and will teach
+ * modpost not to issue a warning.
  * The markers follow same syntax rules as __init / __initdata. */
-#define __init_refok     noinline __section(.text.init.refok)
-#define __initdata_refok          __section(.data.init.refok)
-#define __exit_refok     noinline __section(.exit.text.refok)
+#define __ref            __section(.ref.text) noinline
+#define __refdata        __section(.ref.data)
+#define __refconst       __section(.ref.rodata)
+
+/* backward compatibility note
+ *  A few places hardcode the old section names:
+ *  .text.init.refok
+ *  .data.init.refok
+ *  .exit.text.refok
+ *  They should be converted to use the defines from this file
+ */
+
+/* compatibility defines */
+#define __init_refok     __ref
+#define __initdata_refok __refdata
+#define __exit_refok     __ref
+
 
 #ifdef MODULE
 #define __exitused
@@ -93,11 +107,9 @@
 
 /* For assembly routines */
 #define __INIT		.section	".init.text","ax"
-#define __INIT_REFOK	.section	".text.init.refok","ax"
 #define __FINIT		.previous
 
 #define __INITDATA	.section	".init.data","aw"
-#define __INITDATA_REFOK .section	".data.init.refok","aw"
 
 #define __DEVINIT        .section	".devinit.text", "ax"
 #define __DEVINITDATA    .section	".devinit.data", "aw"
@@ -108,6 +120,14 @@
 #define __MEMINIT        .section	".meminit.text", "ax"
 #define __MEMINITDATA    .section	".meminit.data", "aw"
 
+/* silence warnings when references are OK */
+#define __REF            .section       ".ref.text", "ax"
+#define __REFDATA        .section       ".ref.data", "aw"
+#define __REFCONST       .section       ".ref.rodata", "aw"
+/* backward compatibility */
+#define __INIT_REFOK     .section	__REF
+#define __INITDATA_REFOK .section	__REFDATA
+
 #ifndef __ASSEMBLY__
 /*
  * Used for initialization calls..
-- 
cgit v1.2.3


From bbdfc2f70610bebb841d0874dc901c648308e43a Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 6 Nov 2007 23:29:47 -0800
Subject: [SPLICE]: Don't assume regular pages in splice_to_pipe()

Allow caller to pass in a release function, there might be
other resources that need releasing as well. Needed for
network receive.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/splice.c            | 9 ++++++++-
 include/linux/splice.h | 1 +
 2 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/fs/splice.c b/fs/splice.c
index 56b802bfbfa4..0a0b79b01d05 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -254,11 +254,16 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
 	}
 
 	while (page_nr < spd_pages)
-		page_cache_release(spd->pages[page_nr++]);
+		spd->spd_release(spd, page_nr++);
 
 	return ret;
 }
 
+static void spd_release_page(struct splice_pipe_desc *spd, unsigned int i)
+{
+	page_cache_release(spd->pages[i]);
+}
+
 static int
 __generic_file_splice_read(struct file *in, loff_t *ppos,
 			   struct pipe_inode_info *pipe, size_t len,
@@ -277,6 +282,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos,
 		.partial = partial,
 		.flags = flags,
 		.ops = &page_cache_pipe_buf_ops,
+		.spd_release = spd_release_page,
 	};
 
 	index = *ppos >> PAGE_CACHE_SHIFT;
@@ -1432,6 +1438,7 @@ static long vmsplice_to_pipe(struct file *file, const struct iovec __user *iov,
 		.partial = partial,
 		.flags = flags,
 		.ops = &user_page_pipe_buf_ops,
+		.spd_release = spd_release_page,
 	};
 
 	pipe = pipe_info(file->f_path.dentry->d_inode);
diff --git a/include/linux/splice.h b/include/linux/splice.h
index 33e447f98a54..528dcb93c2f2 100644
--- a/include/linux/splice.h
+++ b/include/linux/splice.h
@@ -53,6 +53,7 @@ struct splice_pipe_desc {
 	int nr_pages;			/* number of pages in map */
 	unsigned int flags;		/* splice flags */
 	const struct pipe_buf_operations *ops;/* ops associated with output pipe */
+	void (*spd_release)(struct splice_pipe_desc *, unsigned int);
 };
 
 typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *,
-- 
cgit v1.2.3


From 9c55e01c0cc835818475a6ce8c4d684df9949ac8 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 6 Nov 2007 23:30:13 -0800
Subject: [TCP]: Splice receive support.

Support for network splice receive.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h    |   3 +
 include/linux/skbuff.h |   6 ++
 include/net/tcp.h      |   3 +
 net/core/skbuff.c      | 246 +++++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/af_inet.c     |   1 +
 net/ipv4/tcp.c         | 129 ++++++++++++++++++++++++++
 net/socket.c           |  13 +++
 7 files changed, 401 insertions(+)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index 596131ea46f4..0235d917d5c3 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -22,6 +22,7 @@
 #include <asm/socket.h>
 
 struct poll_table_struct;
+struct pipe_inode_info;
 struct inode;
 struct net;
 
@@ -172,6 +173,8 @@ struct proto_ops {
 				      struct vm_area_struct * vma);
 	ssize_t		(*sendpage)  (struct socket *sock, struct page *page,
 				      int offset, size_t size, int flags);
+	ssize_t 	(*splice_read)(struct socket *sock,  loff_t *ppos,
+				       struct pipe_inode_info *pipe, size_t len, unsigned int flags);
 };
 
 struct net_proto_family {
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index bddd50bd6878..d39f53ef66bb 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -95,6 +95,7 @@
 
 struct net_device;
 struct scatterlist;
+struct pipe_inode_info;
 
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 struct nf_conntrack {
@@ -1559,6 +1560,11 @@ extern int	       skb_store_bits(struct sk_buff *skb, int offset,
 extern __wsum	       skb_copy_and_csum_bits(const struct sk_buff *skb,
 					      int offset, u8 *to, int len,
 					      __wsum csum);
+extern int             skb_splice_bits(struct sk_buff *skb,
+						unsigned int offset,
+						struct pipe_inode_info *pipe,
+						unsigned int len,
+						unsigned int flags);
 extern void	       skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to);
 extern void	       skb_split(struct sk_buff *skb,
 				 struct sk_buff *skb1, const u32 len);
diff --git a/include/net/tcp.h b/include/net/tcp.h
index cb5b033e0e59..d893b4480769 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -309,6 +309,9 @@ extern int			tcp_twsk_unique(struct sock *sk,
 
 extern void			tcp_twsk_destructor(struct sock *sk);
 
+extern ssize_t			tcp_splice_read(struct socket *sk, loff_t *ppos,
+					        struct pipe_inode_info *pipe, size_t len, unsigned int flags);
+
 static inline void tcp_dec_quickack_mode(struct sock *sk,
 					 const unsigned int pkts)
 {
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b6283779e93d..98420f9c4b6d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -52,6 +52,7 @@
 #endif
 #include <linux/string.h>
 #include <linux/skbuff.h>
+#include <linux/splice.h>
 #include <linux/cache.h>
 #include <linux/rtnetlink.h>
 #include <linux/init.h>
@@ -71,6 +72,40 @@
 static struct kmem_cache *skbuff_head_cache __read_mostly;
 static struct kmem_cache *skbuff_fclone_cache __read_mostly;
 
+static void sock_pipe_buf_release(struct pipe_inode_info *pipe,
+				  struct pipe_buffer *buf)
+{
+	struct sk_buff *skb = (struct sk_buff *) buf->private;
+
+	kfree_skb(skb);
+}
+
+static void sock_pipe_buf_get(struct pipe_inode_info *pipe,
+				struct pipe_buffer *buf)
+{
+	struct sk_buff *skb = (struct sk_buff *) buf->private;
+
+	skb_get(skb);
+}
+
+static int sock_pipe_buf_steal(struct pipe_inode_info *pipe,
+			       struct pipe_buffer *buf)
+{
+	return 1;
+}
+
+
+/* Pipe buffer operations for a socket. */
+static struct pipe_buf_operations sock_pipe_buf_ops = {
+	.can_merge = 0,
+	.map = generic_pipe_buf_map,
+	.unmap = generic_pipe_buf_unmap,
+	.confirm = generic_pipe_buf_confirm,
+	.release = sock_pipe_buf_release,
+	.steal = sock_pipe_buf_steal,
+	.get = sock_pipe_buf_get,
+};
+
 /*
  *	Keep out-of-line to prevent kernel bloat.
  *	__builtin_return_address is not used because it is not always
@@ -1122,6 +1157,217 @@ fault:
 	return -EFAULT;
 }
 
+/*
+ * Callback from splice_to_pipe(), if we need to release some pages
+ * at the end of the spd in case we error'ed out in filling the pipe.
+ */
+static void sock_spd_release(struct splice_pipe_desc *spd, unsigned int i)
+{
+	struct sk_buff *skb = (struct sk_buff *) spd->partial[i].private;
+
+	kfree_skb(skb);
+}
+
+/*
+ * Fill page/offset/length into spd, if it can hold more pages.
+ */
+static inline int spd_fill_page(struct splice_pipe_desc *spd, struct page *page,
+				unsigned int len, unsigned int offset,
+				struct sk_buff *skb)
+{
+	if (unlikely(spd->nr_pages == PIPE_BUFFERS))
+		return 1;
+
+	spd->pages[spd->nr_pages] = page;
+	spd->partial[spd->nr_pages].len = len;
+	spd->partial[spd->nr_pages].offset = offset;
+	spd->partial[spd->nr_pages].private = (unsigned long) skb_get(skb);
+	spd->nr_pages++;
+	return 0;
+}
+
+/*
+ * Map linear and fragment data from the skb to spd. Returns number of
+ * pages mapped.
+ */
+static int __skb_splice_bits(struct sk_buff *skb, unsigned int *offset,
+			     unsigned int *total_len,
+			     struct splice_pipe_desc *spd)
+{
+	unsigned int nr_pages = spd->nr_pages;
+	unsigned int poff, plen, len, toff, tlen;
+	int headlen, seg;
+
+	toff = *offset;
+	tlen = *total_len;
+	if (!tlen)
+		goto err;
+
+	/*
+	 * if the offset is greater than the linear part, go directly to
+	 * the fragments.
+	 */
+	headlen = skb_headlen(skb);
+	if (toff >= headlen) {
+		toff -= headlen;
+		goto map_frag;
+	}
+
+	/*
+	 * first map the linear region into the pages/partial map, skipping
+	 * any potential initial offset.
+	 */
+	len = 0;
+	while (len < headlen) {
+		void *p = skb->data + len;
+
+		poff = (unsigned long) p & (PAGE_SIZE - 1);
+		plen = min_t(unsigned int, headlen - len, PAGE_SIZE - poff);
+		len += plen;
+
+		if (toff) {
+			if (plen <= toff) {
+				toff -= plen;
+				continue;
+			}
+			plen -= toff;
+			poff += toff;
+			toff = 0;
+		}
+
+		plen = min(plen, tlen);
+		if (!plen)
+			break;
+
+		/*
+		 * just jump directly to update and return, no point
+		 * in going over fragments when the output is full.
+		 */
+		if (spd_fill_page(spd, virt_to_page(p), plen, poff, skb))
+			goto done;
+
+		tlen -= plen;
+	}
+
+	/*
+	 * then map the fragments
+	 */
+map_frag:
+	for (seg = 0; seg < skb_shinfo(skb)->nr_frags; seg++) {
+		const skb_frag_t *f = &skb_shinfo(skb)->frags[seg];
+
+		plen = f->size;
+		poff = f->page_offset;
+
+		if (toff) {
+			if (plen <= toff) {
+				toff -= plen;
+				continue;
+			}
+			plen -= toff;
+			poff += toff;
+			toff = 0;
+		}
+
+		plen = min(plen, tlen);
+		if (!plen)
+			break;
+
+		if (spd_fill_page(spd, f->page, plen, poff, skb))
+			break;
+
+		tlen -= plen;
+	}
+
+done:
+	if (spd->nr_pages - nr_pages) {
+		*offset = 0;
+		*total_len = tlen;
+		return 0;
+	}
+err:
+	return 1;
+}
+
+/*
+ * Map data from the skb to a pipe. Should handle both the linear part,
+ * the fragments, and the frag list. It does NOT handle frag lists within
+ * the frag list, if such a thing exists. We'd probably need to recurse to
+ * handle that cleanly.
+ */
+int skb_splice_bits(struct sk_buff *__skb, unsigned int offset,
+		    struct pipe_inode_info *pipe, unsigned int tlen,
+		    unsigned int flags)
+{
+	struct partial_page partial[PIPE_BUFFERS];
+	struct page *pages[PIPE_BUFFERS];
+	struct splice_pipe_desc spd = {
+		.pages = pages,
+		.partial = partial,
+		.flags = flags,
+		.ops = &sock_pipe_buf_ops,
+		.spd_release = sock_spd_release,
+	};
+	struct sk_buff *skb;
+
+	/*
+	 * I'd love to avoid the clone here, but tcp_read_sock()
+	 * ignores reference counts and unconditonally kills the sk_buff
+	 * on return from the actor.
+	 */
+	skb = skb_clone(__skb, GFP_KERNEL);
+	if (unlikely(!skb))
+		return -ENOMEM;
+
+	/*
+	 * __skb_splice_bits() only fails if the output has no room left,
+	 * so no point in going over the frag_list for the error case.
+	 */
+	if (__skb_splice_bits(skb, &offset, &tlen, &spd))
+		goto done;
+	else if (!tlen)
+		goto done;
+
+	/*
+	 * now see if we have a frag_list to map
+	 */
+	if (skb_shinfo(skb)->frag_list) {
+		struct sk_buff *list = skb_shinfo(skb)->frag_list;
+
+		for (; list && tlen; list = list->next) {
+			if (__skb_splice_bits(list, &offset, &tlen, &spd))
+				break;
+		}
+	}
+
+done:
+	/*
+	 * drop our reference to the clone, the pipe consumption will
+	 * drop the rest.
+	 */
+	kfree_skb(skb);
+
+	if (spd.nr_pages) {
+		int ret;
+
+		/*
+		 * Drop the socket lock, otherwise we have reverse
+		 * locking dependencies between sk_lock and i_mutex
+		 * here as compared to sendfile(). We enter here
+		 * with the socket lock held, and splice_to_pipe() will
+		 * grab the pipe inode lock. For sendfile() emulation,
+		 * we call into ->sendpage() with the i_mutex lock held
+		 * and networking will grab the socket lock.
+		 */
+		release_sock(__skb->sk);
+		ret = splice_to_pipe(pipe, &spd);
+		lock_sock(__skb->sk);
+		return ret;
+	}
+
+	return 0;
+}
+
 /**
  *	skb_store_bits - store bits from kernel buffer to skb
  *	@skb: destination buffer
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index d2f22e74b267..c75f20b49935 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -838,6 +838,7 @@ const struct proto_ops inet_stream_ops = {
 	.recvmsg	   = sock_common_recvmsg,
 	.mmap		   = sock_no_mmap,
 	.sendpage	   = tcp_sendpage,
+	.splice_read	   = tcp_splice_read,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_sock_common_setsockopt,
 	.compat_getsockopt = compat_sock_common_getsockopt,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8e65182f7af1..56ed40703f98 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -254,6 +254,10 @@
 #include <linux/poll.h>
 #include <linux/init.h>
 #include <linux/fs.h>
+#include <linux/skbuff.h>
+#include <linux/splice.h>
+#include <linux/net.h>
+#include <linux/socket.h>
 #include <linux/random.h>
 #include <linux/bootmem.h>
 #include <linux/cache.h>
@@ -265,6 +269,7 @@
 #include <net/xfrm.h>
 #include <net/ip.h>
 #include <net/netdma.h>
+#include <net/sock.h>
 
 #include <asm/uaccess.h>
 #include <asm/ioctls.h>
@@ -291,6 +296,15 @@ atomic_t tcp_sockets_allocated;	/* Current number of TCP sockets. */
 EXPORT_SYMBOL(tcp_memory_allocated);
 EXPORT_SYMBOL(tcp_sockets_allocated);
 
+/*
+ * TCP splice context
+ */
+struct tcp_splice_state {
+	struct pipe_inode_info *pipe;
+	size_t len;
+	unsigned int flags;
+};
+
 /*
  * Pressure flag: try to collapse.
  * Technical note: it is used by multiple contexts non atomically.
@@ -501,6 +515,120 @@ static inline void tcp_push(struct sock *sk, int flags, int mss_now,
 	}
 }
 
+int tcp_splice_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb,
+			 unsigned int offset, size_t len)
+{
+	struct tcp_splice_state *tss = rd_desc->arg.data;
+
+	return skb_splice_bits(skb, offset, tss->pipe, tss->len, tss->flags);
+}
+
+static int __tcp_splice_read(struct sock *sk, struct tcp_splice_state *tss)
+{
+	/* Store TCP splice context information in read_descriptor_t. */
+	read_descriptor_t rd_desc = {
+		.arg.data = tss,
+	};
+
+	return tcp_read_sock(sk, &rd_desc, tcp_splice_data_recv);
+}
+
+/**
+ *  tcp_splice_read - splice data from TCP socket to a pipe
+ * @sock:	socket to splice from
+ * @ppos:	position (not valid)
+ * @pipe:	pipe to splice to
+ * @len:	number of bytes to splice
+ * @flags:	splice modifier flags
+ *
+ * Description:
+ *    Will read pages from given socket and fill them into a pipe.
+ *
+ **/
+ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos,
+			struct pipe_inode_info *pipe, size_t len,
+			unsigned int flags)
+{
+	struct sock *sk = sock->sk;
+	struct tcp_splice_state tss = {
+		.pipe = pipe,
+		.len = len,
+		.flags = flags,
+	};
+	long timeo;
+	ssize_t spliced;
+	int ret;
+
+	/*
+	 * We can't seek on a socket input
+	 */
+	if (unlikely(*ppos))
+		return -ESPIPE;
+
+	ret = spliced = 0;
+
+	lock_sock(sk);
+
+	timeo = sock_rcvtimeo(sk, flags & SPLICE_F_NONBLOCK);
+	while (tss.len) {
+		ret = __tcp_splice_read(sk, &tss);
+		if (ret < 0)
+			break;
+		else if (!ret) {
+			if (spliced)
+				break;
+			if (flags & SPLICE_F_NONBLOCK) {
+				ret = -EAGAIN;
+				break;
+			}
+			if (sock_flag(sk, SOCK_DONE))
+				break;
+			if (sk->sk_err) {
+				ret = sock_error(sk);
+				break;
+			}
+			if (sk->sk_shutdown & RCV_SHUTDOWN)
+				break;
+			if (sk->sk_state == TCP_CLOSE) {
+				/*
+				 * This occurs when user tries to read
+				 * from never connected socket.
+				 */
+				if (!sock_flag(sk, SOCK_DONE))
+					ret = -ENOTCONN;
+				break;
+			}
+			if (!timeo) {
+				ret = -EAGAIN;
+				break;
+			}
+			sk_wait_data(sk, &timeo);
+			if (signal_pending(current)) {
+				ret = sock_intr_errno(timeo);
+				break;
+			}
+			continue;
+		}
+		tss.len -= ret;
+		spliced += ret;
+
+		release_sock(sk);
+		lock_sock(sk);
+
+		if (sk->sk_err || sk->sk_state == TCP_CLOSE ||
+		    (sk->sk_shutdown & RCV_SHUTDOWN) || !timeo ||
+		    signal_pending(current))
+			break;
+	}
+
+	release_sock(sk);
+
+	if (spliced)
+		return spliced;
+
+	return ret;
+}
+
 static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffset,
 			 size_t psize, int flags)
 {
@@ -2532,6 +2660,7 @@ EXPORT_SYMBOL(tcp_poll);
 EXPORT_SYMBOL(tcp_read_sock);
 EXPORT_SYMBOL(tcp_recvmsg);
 EXPORT_SYMBOL(tcp_sendmsg);
+EXPORT_SYMBOL(tcp_splice_read);
 EXPORT_SYMBOL(tcp_sendpage);
 EXPORT_SYMBOL(tcp_setsockopt);
 EXPORT_SYMBOL(tcp_shutdown);
diff --git a/net/socket.c b/net/socket.c
index 74784dfe8e5b..92fab9e1c602 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -112,6 +112,9 @@ static long compat_sock_ioctl(struct file *file,
 static int sock_fasync(int fd, struct file *filp, int on);
 static ssize_t sock_sendpage(struct file *file, struct page *page,
 			     int offset, size_t size, loff_t *ppos, int more);
+static ssize_t sock_splice_read(struct file *file, loff_t *ppos,
+			        struct pipe_inode_info *pipe, size_t len,
+				unsigned int flags);
 
 /*
  *	Socket files have a set of 'special' operations as well as the generic file ones. These don't appear
@@ -134,6 +137,7 @@ static const struct file_operations socket_file_ops = {
 	.fasync =	sock_fasync,
 	.sendpage =	sock_sendpage,
 	.splice_write = generic_splice_sendpage,
+	.splice_read =	sock_splice_read,
 };
 
 /*
@@ -691,6 +695,15 @@ static ssize_t sock_sendpage(struct file *file, struct page *page,
 	return sock->ops->sendpage(sock, page, offset, size, flags);
 }
 
+static ssize_t sock_splice_read(struct file *file, loff_t *ppos,
+			        struct pipe_inode_info *pipe, size_t len,
+				unsigned int flags)
+{
+	struct socket *sock = file->private_data;
+
+	return sock->ops->splice_read(sock, ppos, pipe, len, flags);
+}
+
 static struct sock_iocb *alloc_sock_iocb(struct kiocb *iocb,
 					 struct sock_iocb *siocb)
 {
-- 
cgit v1.2.3


From 33c732c36169d7022ad7d6eb474b0c9be43a2dc1 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Tue, 13 Nov 2007 20:30:01 -0800
Subject: [IPV4]: Add raw drops counter.

Add raw drops counter for IPv4 in /proc/net/raw .

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 11 ++++++++---
 net/core/sock.c    |  1 +
 net/ipv4/raw.c     | 17 ++++++++++-------
 3 files changed, 19 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 6e1542da33a1..f5b643714131 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -145,7 +145,8 @@ struct sock_common {
   *	@sk_forward_alloc: space allocated forward
   *	@sk_allocation: allocation mode
   *	@sk_sndbuf: size of send buffer in bytes
-  *	@sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE, %SO_OOBINLINE settings
+  *	@sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE,
+  *		   %SO_OOBINLINE settings
   *	@sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets
   *	@sk_route_caps: route capabilities (e.g. %NETIF_F_TSO)
   *	@sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4)
@@ -153,9 +154,12 @@ struct sock_common {
   *	@sk_backlog: always used with the per-socket spinlock held
   *	@sk_callback_lock: used with the callbacks in the end of this struct
   *	@sk_error_queue: rarely used
-  *	@sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt, IPV6_ADDRFORM for instance)
+  *	@sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt,
+  *			  IPV6_ADDRFORM for instance)
   *	@sk_err: last error
-  *	@sk_err_soft: errors that don't cause failure but are the cause of a persistent failure not just 'timed out'
+  *	@sk_err_soft: errors that don't cause failure but are the cause of a
+  *		      persistent failure not just 'timed out'
+  *	@sk_drops: raw drops counter
   *	@sk_ack_backlog: current listen backlog
   *	@sk_max_ack_backlog: listen backlog set in listen()
   *	@sk_priority: %SO_PRIORITY setting
@@ -239,6 +243,7 @@ struct sock {
 	rwlock_t		sk_callback_lock;
 	int			sk_err,
 				sk_err_soft;
+	atomic_t		sk_drops;
 	unsigned short		sk_ack_backlog;
 	unsigned short		sk_max_ack_backlog;
 	__u32			sk_priority;
diff --git a/net/core/sock.c b/net/core/sock.c
index c519b439b8b1..2029d095b4c5 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1611,6 +1611,7 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 	sk->sk_stamp = ktime_set(-1L, -1L);
 
 	atomic_set(&sk->sk_refcnt, 1);
+	atomic_set(&sk->sk_drops, 0);
 }
 
 void fastcall lock_sock_nested(struct sock *sk, int subclass)
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e7050f8eabeb..761056ef4932 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -241,7 +241,7 @@ static int raw_rcv_skb(struct sock * sk, struct sk_buff * skb)
 	/* Charge it to the socket. */
 
 	if (sock_queue_rcv_skb(sk, skb) < 0) {
-		/* FIXME: increment a raw drops counter here */
+		atomic_inc(&sk->sk_drops);
 		kfree_skb(skb);
 		return NET_RX_DROP;
 	}
@@ -252,6 +252,7 @@ static int raw_rcv_skb(struct sock * sk, struct sk_buff * skb)
 int raw_rcv(struct sock *sk, struct sk_buff *skb)
 {
 	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) {
+		atomic_inc(&sk->sk_drops);
 		kfree_skb(skb);
 		return NET_RX_DROP;
 	}
@@ -871,28 +872,30 @@ static __inline__ char *get_raw_sock(struct sock *sp, char *tmpbuf, int i)
 	      srcp  = inet->num;
 
 	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p",
+		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d",
 		i, src, srcp, dest, destp, sp->sk_state,
 		atomic_read(&sp->sk_wmem_alloc),
 		atomic_read(&sp->sk_rmem_alloc),
 		0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
-		atomic_read(&sp->sk_refcnt), sp);
+		atomic_read(&sp->sk_refcnt), sp, atomic_read(&sp->sk_drops));
 	return tmpbuf;
 }
 
+#define TMPSZ 128
+
 static int raw_seq_show(struct seq_file *seq, void *v)
 {
-	char tmpbuf[129];
+	char tmpbuf[TMPSZ+1];
 
 	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%-127s\n",
+		seq_printf(seq, "%-*s\n", TMPSZ-1,
 			       "  sl  local_address rem_address   st tx_queue "
 			       "rx_queue tr tm->when retrnsmt   uid  timeout "
-			       "inode");
+			       "inode  drops");
 	else {
 		struct raw_iter_state *state = raw_seq_private(seq);
 
-		seq_printf(seq, "%-127s\n",
+		seq_printf(seq, "%-*s\n", TMPSZ-1,
 			   get_raw_sock(v, tmpbuf, state->bucket));
 	}
 	return 0;
-- 
cgit v1.2.3


From b4ce92775c2e7ff9cf79cca4e0a19c8c5fd6287b Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:33:32 -0800
Subject: [IPV6]: Move nfheader_len into rt6_info

The dst member nfheader_len is only used by IPv6.  It's also currently
creating a rather ugly alignment hole in struct dst.  Therefore this patch
moves it from there into struct rt6_info.

It also reorders the fields in rt6_info to minimize holes.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h       |  1 -
 include/net/ip6_fib.h   | 11 ++++++++---
 net/ipv4/xfrm4_policy.c |  1 -
 net/ipv6/ip6_output.c   |  5 +++--
 net/ipv6/xfrm6_policy.c |  3 ++-
 5 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 2f65e894b829..69888f1502b1 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -50,7 +50,6 @@ struct dst_entry
 	unsigned long		expires;
 
 	unsigned short		header_len;	/* more space at head required */
-	unsigned short		nfheader_len;	/* more non-fragment space at head required */
 	unsigned short		trailer_len;	/* space to reserve at tail */
 
 	u32			metrics[RTAX_MAX];
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 857821360bb6..4cefcffe773e 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -99,16 +99,21 @@ struct rt6_info
 	u32				rt6i_flags;
 	u32				rt6i_metric;
 	atomic_t			rt6i_ref;
-	struct fib6_table		*rt6i_table;
 
-	struct rt6key			rt6i_dst;
-	struct rt6key			rt6i_src;
+	/* more non-fragment space at head required */
+	unsigned short			nfheader_len;
 
 	u8				rt6i_protocol;
 
+	struct fib6_table		*rt6i_table;
+
+	struct rt6key			rt6i_dst;
+
 #ifdef CONFIG_XFRM
 	u32				rt6i_flow_cache_genid;
 #endif
+
+	struct rt6key			rt6i_src;
 };
 
 static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst)
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index cc86fb110dd8..5ee3a2f9fb27 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -161,7 +161,6 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		dst_prev->flags	       |= DST_HOST;
 		dst_prev->lastuse	= jiffies;
 		dst_prev->header_len	= header_len;
-		dst_prev->nfheader_len	= 0;
 		dst_prev->trailer_len	= trailer_len;
 		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 150615758fdc..387030d2483c 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1099,7 +1099,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		sk->sk_sndmsg_page = NULL;
 		sk->sk_sndmsg_off = 0;
 		exthdrlen = rt->u.dst.header_len + (opt ? opt->opt_flen : 0) -
-			    rt->u.dst.nfheader_len;
+			    rt->nfheader_len;
 		length += exthdrlen;
 		transhdrlen += exthdrlen;
 	} else {
@@ -1114,7 +1114,8 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 
 	hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
 
-	fragheaderlen = sizeof(struct ipv6hdr) + rt->u.dst.nfheader_len + (opt ? opt->opt_nflen : 0);
+	fragheaderlen = sizeof(struct ipv6hdr) + rt->nfheader_len +
+			(opt ? opt->opt_nflen : 0);
 	maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen - sizeof(struct frag_hdr);
 
 	if (mtu <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) {
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 3cad3e8dc4da..5b02f0efd38e 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -157,7 +157,8 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		dst_prev = dst1;
 
 		if (xfrm[i]->type->flags & XFRM_TYPE_NON_FRAGMENT)
-			dst->nfheader_len += xfrm[i]->props.header_len;
+			((struct rt6_info *)dst)->nfheader_len +=
+				xfrm[i]->props.header_len;
 		header_len += xfrm[i]->props.header_len;
 		trailer_len += xfrm[i]->props.trailer_len;
 
-- 
cgit v1.2.3


From 352e512c32b634768303a43768245a0363cebbe7 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:34:06 -0800
Subject: [NET]: Eliminate duplicate copies of dst_discard

We have a number of copies of dst_discard scattered around the place
which all do the same thing, namely free a packet on the input or
output paths.

This patch deletes all of them except dst_discard and points all the
users to it.

The only non-trivial bit is decnet where it returns an error.
However, conceptually this is identical to the blackhole functions
used in IPv4 and IPv6 which do not return errors.  So they should
either all return errors or all return zero.  For now I've stuck with
the majority and picked zero as the return value.

It doesn't really matter in practice since few if any driver would
react differently depending on a zero return value or NET_RX_DROP.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h     |  1 +
 net/core/dst.c        |  3 ++-
 net/decnet/dn_route.c | 13 +------------
 net/ipv4/route.c      | 11 +++--------
 net/ipv6/exthdrs.c    | 13 ++-----------
 net/ipv6/route.c      | 21 ++++-----------------
 6 files changed, 13 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 69888f1502b1..7a0b1bde8e28 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -179,6 +179,7 @@ static inline struct dst_entry *dst_pop(struct dst_entry *dst)
 	return child;
 }
 
+extern int dst_discard(struct sk_buff *skb);
 extern void * dst_alloc(struct dst_ops * ops);
 extern void __dst_free(struct dst_entry * dst);
 extern struct dst_entry *dst_destroy(struct dst_entry * dst);
diff --git a/net/core/dst.c b/net/core/dst.c
index 03daead3592a..f538061f4b93 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -153,11 +153,12 @@ loop:
 #endif
 }
 
-static int dst_discard(struct sk_buff *skb)
+int dst_discard(struct sk_buff *skb)
 {
 	kfree_skb(skb);
 	return 0;
 }
+EXPORT_SYMBOL(dst_discard);
 
 void * dst_alloc(struct dst_ops * ops)
 {
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 23aa3556e56f..2a5bb0714c7e 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -764,17 +764,6 @@ drop:
 	return NET_RX_DROP;
 }
 
-/*
- * Drop packet. This is used for endnodes and for
- * when we should not be forwarding packets from
- * this dest.
- */
-static int dn_blackhole(struct sk_buff *skb)
-{
-	kfree_skb(skb);
-	return NET_RX_DROP;
-}
-
 /*
  * Used to catch bugs. This should never normally get
  * called.
@@ -1396,7 +1385,7 @@ make_route:
 		default:
 		case RTN_UNREACHABLE:
 		case RTN_BLACKHOLE:
-			rt->u.dst.input = dn_blackhole;
+			rt->u.dst.input = dst_discard;
 	}
 	rt->rt_flags = flags;
 	if (rt->u.dst.dev)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 49e008568ddf..137b8eb666b7 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -92,6 +92,7 @@
 #include <linux/jhash.h>
 #include <linux/rcupdate.h>
 #include <linux/times.h>
+#include <net/dst.h>
 #include <net/net_namespace.h>
 #include <net/protocol.h>
 #include <net/ip.h>
@@ -2357,12 +2358,6 @@ static struct dst_ops ipv4_dst_blackhole_ops = {
 };
 
 
-static int ipv4_blackhole_output(struct sk_buff *skb)
-{
-	kfree_skb(skb);
-	return 0;
-}
-
 static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp, struct sock *sk)
 {
 	struct rtable *ort = *rp;
@@ -2374,8 +2369,8 @@ static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp, struct sock
 
 		atomic_set(&new->__refcnt, 1);
 		new->__use = 1;
-		new->input = ipv4_blackhole_output;
-		new->output = ipv4_blackhole_output;
+		new->input = dst_discard;
+		new->output = dst_discard;
 		memcpy(new->metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32));
 
 		new->dev = ort->u.dst.dev;
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 1e89efd38a0c..cee06b1655c1 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -32,6 +32,7 @@
 #include <linux/in6.h>
 #include <linux/icmpv6.h>
 
+#include <net/dst.h>
 #include <net/sock.h>
 #include <net/snmp.h>
 
@@ -318,18 +319,8 @@ void __init ipv6_destopt_init(void)
 		printk(KERN_ERR "ipv6_destopt_init: Could not register protocol\n");
 }
 
-/********************************
-  NONE header. No data in packet.
- ********************************/
-
-static int ipv6_nodata_rcv(struct sk_buff *skb)
-{
-	kfree_skb(skb);
-	return 0;
-}
-
 static struct inet6_protocol nodata_protocol = {
-	.handler	=	ipv6_nodata_rcv,
+	.handler	=	dst_discard,
 	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 20083e0d3995..ac70e2d3b10c 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -152,7 +152,6 @@ struct rt6_info ip6_null_entry = {
 
 static int ip6_pkt_prohibit(struct sk_buff *skb);
 static int ip6_pkt_prohibit_out(struct sk_buff *skb);
-static int ip6_pkt_blk_hole(struct sk_buff *skb);
 
 struct rt6_info ip6_prohibit_entry = {
 	.u = {
@@ -181,8 +180,8 @@ struct rt6_info ip6_blk_hole_entry = {
 			.obsolete	= -1,
 			.error		= -EINVAL,
 			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
-			.input		= ip6_pkt_blk_hole,
-			.output		= ip6_pkt_blk_hole,
+			.input		= dst_discard,
+			.output		= dst_discard,
 			.ops		= &ip6_dst_ops,
 			.path		= (struct dst_entry*)&ip6_blk_hole_entry,
 		}
@@ -782,12 +781,6 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
 
 EXPORT_SYMBOL(ip6_route_output);
 
-static int ip6_blackhole_output(struct sk_buff *skb)
-{
-	kfree_skb(skb);
-	return 0;
-}
-
 int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl)
 {
 	struct rt6_info *ort = (struct rt6_info *) *dstp;
@@ -800,8 +793,8 @@ int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl
 
 		atomic_set(&new->__refcnt, 1);
 		new->__use = 1;
-		new->input = ip6_blackhole_output;
-		new->output = ip6_blackhole_output;
+		new->input = dst_discard;
+		new->output = dst_discard;
 
 		memcpy(new->metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32));
 		new->dev = ort->u.dst.dev;
@@ -1811,12 +1804,6 @@ static int ip6_pkt_prohibit_out(struct sk_buff *skb)
 	return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES);
 }
 
-static int ip6_pkt_blk_hole(struct sk_buff *skb)
-{
-	kfree_skb(skb);
-	return 0;
-}
-
 #endif
 
 /*
-- 
cgit v1.2.3


From 274b3426db25b8d63cbf25475e728ce1ee6caebd Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:34:34 -0800
Subject: [NET]: Remove unnecessary inclusion of dst.h

The file net/netevent.h only refers to struct dst_entry * so it
doesn't need to include dst.h.  I've replaced it with a forward
declaration.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netevent.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/netevent.h b/include/net/netevent.h
index e5d216241423..e82b7bab3ff3 100644
--- a/include/net/netevent.h
+++ b/include/net/netevent.h
@@ -12,7 +12,7 @@
  */
 #ifdef __KERNEL__
 
-#include <net/dst.h>
+struct dst_entry;
 
 struct netevent_redirect {
 	struct dst_entry *old;
-- 
cgit v1.2.3


From f04e7e8d7f175c05bbde3ae748bf2541da53721d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:36:51 -0800
Subject: [IPSEC]: Replace x->type->{local,remote}_addr with flags

The functions local_addr and remote_addr are more than what they're
needed for.  The same thing can be done easily with flags on the type
object.  This patch does that and simplifies the wrapper functions in
xfrm6_policy accordingly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  4 ++--
 net/ipv6/mip6.c         | 11 ++---------
 net/ipv6/xfrm6_policy.c | 20 ++++++++------------
 3 files changed, 12 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 1dd20cf17982..3434fdc7de37 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -282,6 +282,8 @@ struct xfrm_type
 	__u8			flags;
 #define XFRM_TYPE_NON_FRAGMENT	1
 #define XFRM_TYPE_REPLAY_PROT	2
+#define XFRM_TYPE_LOCAL_COADDR	4
+#define XFRM_TYPE_REMOTE_COADDR	8
 
 	int			(*init_state)(struct xfrm_state *x);
 	void			(*destructor)(struct xfrm_state *);
@@ -289,8 +291,6 @@ struct xfrm_type
 	int			(*output)(struct xfrm_state *, struct sk_buff *pskb);
 	int			(*reject)(struct xfrm_state *, struct sk_buff *, struct flowi *);
 	int			(*hdr_offset)(struct xfrm_state *, struct sk_buff *, u8 **);
-	xfrm_address_t		*(*local_addr)(struct xfrm_state *, xfrm_address_t *);
-	xfrm_address_t		*(*remote_addr)(struct xfrm_state *, xfrm_address_t *);
 	/* Estimate maximal size of result of transformation of a dgram */
 	u32			(*get_mtu)(struct xfrm_state *, int size);
 };
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index 7fd841d41019..edfd9cdd721c 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -34,11 +34,6 @@
 #include <net/xfrm.h>
 #include <net/mip6.h>
 
-static xfrm_address_t *mip6_xfrm_addr(struct xfrm_state *x, xfrm_address_t *addr)
-{
-	return x->coaddr;
-}
-
 static inline unsigned int calc_padlen(unsigned int len, unsigned int n)
 {
 	return (n - len + 16) & 0x7;
@@ -337,14 +332,13 @@ static struct xfrm_type mip6_destopt_type =
 	.description	= "MIP6DESTOPT",
 	.owner		= THIS_MODULE,
 	.proto	     	= IPPROTO_DSTOPTS,
-	.flags		= XFRM_TYPE_NON_FRAGMENT,
+	.flags		= XFRM_TYPE_NON_FRAGMENT | XFRM_TYPE_LOCAL_COADDR,
 	.init_state	= mip6_destopt_init_state,
 	.destructor	= mip6_destopt_destroy,
 	.input		= mip6_destopt_input,
 	.output		= mip6_destopt_output,
 	.reject		= mip6_destopt_reject,
 	.hdr_offset	= mip6_destopt_offset,
-	.local_addr	= mip6_xfrm_addr,
 };
 
 static int mip6_rthdr_input(struct xfrm_state *x, struct sk_buff *skb)
@@ -467,13 +461,12 @@ static struct xfrm_type mip6_rthdr_type =
 	.description	= "MIP6RT",
 	.owner		= THIS_MODULE,
 	.proto	     	= IPPROTO_ROUTING,
-	.flags		= XFRM_TYPE_NON_FRAGMENT,
+	.flags		= XFRM_TYPE_NON_FRAGMENT | XFRM_TYPE_REMOTE_COADDR,
 	.init_state	= mip6_rthdr_init_state,
 	.destructor	= mip6_rthdr_destroy,
 	.input		= mip6_rthdr_input,
 	.output		= mip6_rthdr_output,
 	.hdr_offset	= mip6_rthdr_offset,
-	.remote_addr	= mip6_xfrm_addr,
 };
 
 static int __init mip6_init(void)
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 77dc3651437e..3b38e493d151 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -87,20 +87,16 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 	return dst;
 }
 
-static inline struct in6_addr*
-__xfrm6_bundle_addr_remote(struct xfrm_state *x, struct in6_addr *addr)
+static inline xfrm_address_t *__xfrm6_bundle_addr_remote(struct xfrm_state *x)
 {
-	return (x->type->remote_addr) ?
-		(struct in6_addr*)x->type->remote_addr(x, (xfrm_address_t *)addr) :
-		(struct in6_addr*)&x->id.daddr;
+	return (x->type->flags & XFRM_TYPE_REMOTE_COADDR) ? x->coaddr :
+							    &x->id.daddr;
 }
 
-static inline struct in6_addr*
-__xfrm6_bundle_addr_local(struct xfrm_state *x, struct in6_addr *addr)
+static inline xfrm_address_t *__xfrm6_bundle_addr_local(struct xfrm_state *x)
 {
-	return (x->type->local_addr) ?
-		(struct in6_addr*)x->type->local_addr(x, (xfrm_address_t *)addr) :
-		(struct in6_addr*)&x->props.saddr;
+	return (x->type->flags & XFRM_TYPE_LOCAL_COADDR) ? x->coaddr :
+							   &x->props.saddr;
 }
 
 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
@@ -171,9 +167,9 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 				fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
 				break;
 			case AF_INET6:
-				ipv6_addr_copy(&fl_tunnel.fl6_dst, __xfrm6_bundle_addr_remote(xfrm[i], &fl->fl6_dst));
+				ipv6_addr_copy(&fl_tunnel.fl6_dst, (struct in6_addr *)__xfrm6_bundle_addr_remote(xfrm[i]));
 
-				ipv6_addr_copy(&fl_tunnel.fl6_src, __xfrm6_bundle_addr_local(xfrm[i], &fl->fl6_src));
+				ipv6_addr_copy(&fl_tunnel.fl6_src, (struct in6_addr *)__xfrm6_bundle_addr_local(xfrm[i]));
 				break;
 			default:
 				BUG_ON(1);
-- 
cgit v1.2.3


From 66cdb3ca27323a92712d289fc5edc7841d74a139 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:37:28 -0800
Subject: [IPSEC]: Move flow construction into xfrm_dst_lookup

This patch moves the flow construction from the callers of
xfrm_dst_lookup into that function.  It also changes xfrm_dst_lookup
so that it takes an xfrm state as its argument instead of explicit
addresses.

This removes any address-specific logic from the callers of
xfrm_dst_lookup which is needed to correctly support inter-family
transforms.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      | 10 ++---
 net/ipv4/xfrm4_policy.c | 80 +++++++++++++++++++---------------------
 net/ipv6/xfrm6_policy.c | 97 ++++++++++++++++++-------------------------------
 net/xfrm/xfrm_policy.c  | 25 ++++++++-----
 4 files changed, 91 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 3434fdc7de37..d427343f527b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -233,7 +233,8 @@ struct xfrm_policy_afinfo {
 	unsigned short		family;
 	struct dst_ops		*dst_ops;
 	void			(*garbage_collect)(void);
-	int			(*dst_lookup)(struct xfrm_dst **dst, struct flowi *fl);
+	struct dst_entry	*(*dst_lookup)(int tos, xfrm_address_t *saddr,
+					       xfrm_address_t *daddr);
 	int			(*get_saddr)(xfrm_address_t *saddr, xfrm_address_t *daddr);
 	struct dst_entry	*(*find_bundle)(struct flowi *fl, struct xfrm_policy *policy);
 	int			(*bundle_create)(struct xfrm_policy *policy, 
@@ -1079,7 +1080,6 @@ extern int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
 #ifdef CONFIG_XFRM
 extern int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb);
 extern int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen);
-extern int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl, unsigned short family);
 #else
 static inline int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen)
 {
@@ -1092,13 +1092,9 @@ static inline int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb)
  	kfree_skb(skb);
 	return 0;
 }
-
-static inline int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl, unsigned short family)
-{
-	return -EINVAL;
-} 
 #endif
 
+extern struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos);
 struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp);
 extern int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), void *);
 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl);
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index d903c8bdffcd..cebc84731969 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -8,7 +8,8 @@
  *
  */
 
-#include <linux/compiler.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
 #include <linux/inetdevice.h>
 #include <net/dst.h>
 #include <net/xfrm.h>
@@ -17,28 +18,44 @@
 static struct dst_ops xfrm4_dst_ops;
 static struct xfrm_policy_afinfo xfrm4_policy_afinfo;
 
-static int xfrm4_dst_lookup(struct xfrm_dst **dst, struct flowi *fl)
+static struct dst_entry *xfrm4_dst_lookup(int tos, xfrm_address_t *saddr,
+					  xfrm_address_t *daddr)
 {
-	return __ip_route_output_key((struct rtable**)dst, fl);
-}
-
-static int xfrm4_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
-{
-	struct rtable *rt;
-	struct flowi fl_tunnel = {
+	struct flowi fl = {
 		.nl_u = {
 			.ip4_u = {
+				.tos = tos,
 				.daddr = daddr->a4,
 			},
 		},
 	};
+	struct dst_entry *dst;
+	struct rtable *rt;
+	int err;
 
-	if (!xfrm4_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
-		saddr->a4 = rt->rt_src;
-		dst_release(&rt->u.dst);
-		return 0;
-	}
-	return -EHOSTUNREACH;
+	if (saddr)
+		fl.fl4_src = saddr->a4;
+
+	err = __ip_route_output_key(&rt, &fl);
+	dst = &rt->u.dst;
+	if (err)
+		dst = ERR_PTR(err);
+	return dst;
+}
+
+static int xfrm4_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
+{
+	struct dst_entry *dst;
+	struct rtable *rt;
+
+	dst = xfrm4_dst_lookup(0, NULL, daddr);
+	if (IS_ERR(dst))
+		return -EHOSTUNREACH;
+
+	rt = (struct rtable *)dst;
+	saddr->a4 = rt->rt_src;
+	dst_release(dst);
+	return 0;
 }
 
 static struct dst_entry *
@@ -73,15 +90,7 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 	struct dst_entry *dst, *dst_prev;
 	struct rtable *rt0 = (struct rtable*)(*dst_p);
 	struct rtable *rt = rt0;
-	struct flowi fl_tunnel = {
-		.nl_u = {
-			.ip4_u = {
-				.saddr = fl->fl4_src,
-				.daddr = fl->fl4_dst,
-				.tos = fl->fl4_tos
-			}
-		}
-	};
+	int tos = fl->fl4_tos;
 	int i;
 	int err;
 	int header_len = 0;
@@ -119,25 +128,12 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		trailer_len += xfrm[i]->props.trailer_len;
 
 		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
-			unsigned short encap_family = xfrm[i]->props.family;
-			switch (encap_family) {
-			case AF_INET:
-				fl_tunnel.fl4_dst = xfrm[i]->id.daddr.a4;
-				fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
-				break;
-#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
-			case AF_INET6:
-				ipv6_addr_copy(&fl_tunnel.fl6_dst, (struct in6_addr*)&xfrm[i]->id.daddr.a6);
-				ipv6_addr_copy(&fl_tunnel.fl6_src, (struct in6_addr*)&xfrm[i]->props.saddr.a6);
-				break;
-#endif
-			default:
-				BUG_ON(1);
-			}
-			err = xfrm_dst_lookup((struct xfrm_dst **)&rt,
-					      &fl_tunnel, encap_family);
-			if (err)
+			dst1 = xfrm_dst_lookup(xfrm[i], tos);
+			err = PTR_ERR(dst1);
+			if (IS_ERR(dst1))
 				goto error;
+
+			rt = (struct rtable *)dst1;
 		} else
 			dst_hold(&rt->u.dst);
 	}
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 3b38e493d151..8e78530865a6 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -11,7 +11,8 @@
  *
  */
 
-#include <linux/compiler.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
 #include <linux/netdevice.h>
 #include <net/addrconf.h>
 #include <net/dst.h>
@@ -26,35 +27,40 @@
 static struct dst_ops xfrm6_dst_ops;
 static struct xfrm_policy_afinfo xfrm6_policy_afinfo;
 
-static int xfrm6_dst_lookup(struct xfrm_dst **xdst, struct flowi *fl)
+static struct dst_entry *xfrm6_dst_lookup(int tos, xfrm_address_t *saddr,
+					  xfrm_address_t *daddr)
 {
-	struct dst_entry *dst = ip6_route_output(NULL, fl);
-	int err = dst->error;
-	if (!err)
-		*xdst = (struct xfrm_dst *) dst;
-	else
+	struct flowi fl = {};
+	struct dst_entry *dst;
+	int err;
+
+	memcpy(&fl.fl6_dst, daddr, sizeof(fl.fl6_dst));
+	if (saddr)
+		memcpy(&fl.fl6_src, saddr, sizeof(fl.fl6_src));
+
+	dst = ip6_route_output(NULL, &fl);
+
+	err = dst->error;
+	if (dst->error) {
 		dst_release(dst);
-	return err;
+		dst = ERR_PTR(err);
+	}
+
+	return dst;
 }
 
 static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
 {
-	struct rt6_info *rt;
-	struct flowi fl_tunnel = {
-		.nl_u = {
-			.ip6_u = {
-				.daddr = *(struct in6_addr *)&daddr->a6,
-			},
-		},
-	};
-
-	if (!xfrm6_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
-		ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)&daddr->a6,
-			       (struct in6_addr *)&saddr->a6);
-		dst_release(&rt->u.dst);
-		return 0;
-	}
-	return -EHOSTUNREACH;
+	struct dst_entry *dst;
+
+	dst = xfrm6_dst_lookup(0, NULL, daddr);
+	if (IS_ERR(dst))
+		return -EHOSTUNREACH;
+
+	ipv6_get_saddr(dst, (struct in6_addr *)&daddr->a6,
+		       (struct in6_addr *)&saddr->a6);
+	dst_release(dst);
+	return 0;
 }
 
 static struct dst_entry *
@@ -87,18 +93,6 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 	return dst;
 }
 
-static inline xfrm_address_t *__xfrm6_bundle_addr_remote(struct xfrm_state *x)
-{
-	return (x->type->flags & XFRM_TYPE_REMOTE_COADDR) ? x->coaddr :
-							    &x->id.daddr;
-}
-
-static inline xfrm_address_t *__xfrm6_bundle_addr_local(struct xfrm_state *x)
-{
-	return (x->type->flags & XFRM_TYPE_LOCAL_COADDR) ? x->coaddr :
-							   &x->props.saddr;
-}
-
 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
  * all the metrics... Shortly, bundle a bundle.
  */
@@ -110,14 +104,6 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 	struct dst_entry *dst, *dst_prev;
 	struct rt6_info *rt0 = (struct rt6_info*)(*dst_p);
 	struct rt6_info *rt  = rt0;
-	struct flowi fl_tunnel = {
-		.nl_u = {
-			.ip6_u = {
-				.saddr = fl->fl6_src,
-				.daddr = fl->fl6_dst,
-			}
-		}
-	};
 	int i;
 	int err;
 	int header_len = 0;
@@ -160,25 +146,12 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		trailer_len += xfrm[i]->props.trailer_len;
 
 		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
-			unsigned short encap_family = xfrm[i]->props.family;
-			switch(encap_family) {
-			case AF_INET:
-				fl_tunnel.fl4_dst = xfrm[i]->id.daddr.a4;
-				fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
-				break;
-			case AF_INET6:
-				ipv6_addr_copy(&fl_tunnel.fl6_dst, (struct in6_addr *)__xfrm6_bundle_addr_remote(xfrm[i]));
-
-				ipv6_addr_copy(&fl_tunnel.fl6_src, (struct in6_addr *)__xfrm6_bundle_addr_local(xfrm[i]));
-				break;
-			default:
-				BUG_ON(1);
-			}
-
-			err = xfrm_dst_lookup((struct xfrm_dst **) &rt,
-					      &fl_tunnel, encap_family);
-			if (err)
+			dst1 = xfrm_dst_lookup(xfrm[i], 0);
+			err = PTR_ERR(dst1);
+			if (IS_ERR(dst1))
 				goto error;
+
+			rt = (struct rt6_info *)dst1;
 		} else
 			dst_hold(&rt->u.dst);
 	}
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index df5bfa837eb3..085c19d4d1b7 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -13,6 +13,7 @@
  *
  */
 
+#include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/kmod.h>
 #include <linux/list.h>
@@ -84,21 +85,25 @@ int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl,
 	return 0;
 }
 
-int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl,
-		    unsigned short family)
+struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos)
 {
-	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
-	int err = 0;
+	xfrm_address_t *saddr = &x->props.saddr;
+	xfrm_address_t *daddr = &x->id.daddr;
+	struct xfrm_policy_afinfo *afinfo;
+	struct dst_entry *dst;
 
+	if (x->type->flags & XFRM_TYPE_LOCAL_COADDR)
+		saddr = x->coaddr;
+	if (x->type->flags & XFRM_TYPE_REMOTE_COADDR)
+		daddr = x->coaddr;
+
+	afinfo = xfrm_policy_get_afinfo(x->props.family);
 	if (unlikely(afinfo == NULL))
-		return -EAFNOSUPPORT;
+		return ERR_PTR(-EAFNOSUPPORT);
 
-	if (likely(afinfo->dst_lookup != NULL))
-		err = afinfo->dst_lookup(dst, fl);
-	else
-		err = -EINVAL;
+	dst = afinfo->dst_lookup(tos, saddr, daddr);
 	xfrm_policy_put_afinfo(afinfo);
-	return err;
+	return dst;
 }
 EXPORT_SYMBOL(xfrm_dst_lookup);
 
-- 
cgit v1.2.3


From 25ee3286dcbc830a833354bb1d15567956844813 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 11 Dec 2007 09:32:34 -0800
Subject: [IPSEC]: Merge common code into xfrm_bundle_create

Half of the code in xfrm4_bundle_create and xfrm6_bundle_create are
common.  This patch extracts that logic and puts it into
xfrm_bundle_create.  The rest of it are then accessed through afinfo.

As a result this fixes the problem with inter-family transforms where
we treat every xfrm dst in the bundle as if it belongs to the top
family.

This patch also fixes a long-standing error-path bug where we may free
the xfrm states twice.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  11 ++-
 net/ipv4/xfrm4_policy.c | 134 +++++++----------------------------
 net/ipv6/xfrm6_policy.c | 136 ++++++-----------------------------
 net/xfrm/xfrm_policy.c  | 183 +++++++++++++++++++++++++++++++++++++++++-------
 4 files changed, 211 insertions(+), 253 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d427343f527b..c9345fe3f8d2 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -227,6 +227,7 @@ struct km_event
 	u32	event;
 };
 
+struct net_device;
 struct xfrm_type;
 struct xfrm_dst;
 struct xfrm_policy_afinfo {
@@ -237,13 +238,11 @@ struct xfrm_policy_afinfo {
 					       xfrm_address_t *daddr);
 	int			(*get_saddr)(xfrm_address_t *saddr, xfrm_address_t *daddr);
 	struct dst_entry	*(*find_bundle)(struct flowi *fl, struct xfrm_policy *policy);
-	int			(*bundle_create)(struct xfrm_policy *policy, 
-						 struct xfrm_state **xfrm, 
-						 int nx,
-						 struct flowi *fl, 
-						 struct dst_entry **dst_p);
 	void			(*decode_session)(struct sk_buff *skb,
 						  struct flowi *fl);
+	int			(*get_tos)(struct flowi *fl);
+	int			(*fill_dst)(struct xfrm_dst *xdst,
+					    struct net_device *dev);
 };
 
 extern int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo);
@@ -1094,7 +1093,6 @@ static inline int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb)
 }
 #endif
 
-extern struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos);
 struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp);
 extern int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), void *);
 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl);
@@ -1113,7 +1111,6 @@ extern int xfrm_policy_flush(u8 type, struct xfrm_audit *audit_info);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
 extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst,
 			  struct flowi *fl, int family, int strict);
-extern void xfrm_init_pmtu(struct dst_entry *dst);
 
 #ifdef CONFIG_XFRM_MIGRATE
 extern int km_migrate(struct xfrm_selector *sel, u8 dir, u8 type,
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index cebc84731969..1d7524375b49 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -79,122 +79,39 @@ __xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 	return dst;
 }
 
-/* Allocate chain of dst_entry's, attach known xfrm's, calculate
- * all the metrics... Shortly, bundle a bundle.
- */
-
-static int
-__xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int nx,
-		      struct flowi *fl, struct dst_entry **dst_p)
+static int xfrm4_get_tos(struct flowi *fl)
 {
-	struct dst_entry *dst, *dst_prev;
-	struct rtable *rt0 = (struct rtable*)(*dst_p);
-	struct rtable *rt = rt0;
-	int tos = fl->fl4_tos;
-	int i;
-	int err;
-	int header_len = 0;
-	int trailer_len = 0;
-
-	dst = dst_prev = NULL;
-	dst_hold(&rt->u.dst);
-
-	for (i = 0; i < nx; i++) {
-		struct dst_entry *dst1 = dst_alloc(&xfrm4_dst_ops);
-		struct xfrm_dst *xdst;
-
-		if (unlikely(dst1 == NULL)) {
-			err = -ENOBUFS;
-			dst_release(&rt->u.dst);
-			goto error;
-		}
+	return fl->fl4_tos;
+}
 
-		if (!dst)
-			dst = dst1;
-		else {
-			dst_prev->child = dst1;
-			dst1->flags |= DST_NOHASH;
-			dst_clone(dst1);
-		}
+static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
+{
+	struct rtable *rt = (struct rtable *)xdst->route;
 
-		xdst = (struct xfrm_dst *)dst1;
-		xdst->route = &rt->u.dst;
-		xdst->genid = xfrm[i]->genid;
+	xdst->u.rt.fl = rt->fl;
 
-		dst1->next = dst_prev;
-		dst_prev = dst1;
+	xdst->u.dst.dev = dev;
+	dev_hold(dev);
 
-		header_len += xfrm[i]->props.header_len;
-		trailer_len += xfrm[i]->props.trailer_len;
+	xdst->u.rt.idev = in_dev_get(dev);
+	if (!xdst->u.rt.idev)
+		return -ENODEV;
 
-		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
-			dst1 = xfrm_dst_lookup(xfrm[i], tos);
-			err = PTR_ERR(dst1);
-			if (IS_ERR(dst1))
-				goto error;
+	xdst->u.rt.peer = rt->peer;
+	if (rt->peer)
+		atomic_inc(&rt->peer->refcnt);
 
-			rt = (struct rtable *)dst1;
-		} else
-			dst_hold(&rt->u.dst);
-	}
+	/* Sheit... I remember I did this right. Apparently,
+	 * it was magically lost, so this code needs audit */
+	xdst->u.rt.rt_flags = rt->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST |
+					      RTCF_LOCAL);
+	xdst->u.rt.rt_type = rt->rt_type;
+	xdst->u.rt.rt_src = rt->rt_src;
+	xdst->u.rt.rt_dst = rt->rt_dst;
+	xdst->u.rt.rt_gateway = rt->rt_gateway;
+	xdst->u.rt.rt_spec_dst = rt->rt_spec_dst;
 
-	dst_prev->child = &rt->u.dst;
-	dst->path = &rt->u.dst;
-
-	/* Copy neighbout for reachability confirmation */
-	dst->neighbour = neigh_clone(rt->u.dst.neighbour);
-
-	*dst_p = dst;
-	dst = dst_prev;
-
-	dst_prev = *dst_p;
-	i = 0;
-	err = -ENODEV;
-	for (; dst_prev != &rt->u.dst; dst_prev = dst_prev->child) {
-		struct xfrm_dst *x = (struct xfrm_dst*)dst_prev;
-		x->u.rt.fl = *fl;
-
-		dst_prev->xfrm = xfrm[i++];
-		dst_prev->dev = rt->u.dst.dev;
-		if (!rt->u.dst.dev)
-			goto error;
-		dev_hold(rt->u.dst.dev);
-
-		x->u.rt.idev = in_dev_get(rt->u.dst.dev);
-		if (!x->u.rt.idev)
-			goto error;
-
-		dst_prev->obsolete	= -1;
-		dst_prev->flags	       |= DST_HOST;
-		dst_prev->lastuse	= jiffies;
-		dst_prev->header_len	= header_len;
-		dst_prev->trailer_len	= trailer_len;
-		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
-
-		dst_prev->input = dst_discard;
-		dst_prev->output = dst_prev->xfrm->outer_mode->afinfo->output;
-		if (rt0->peer)
-			atomic_inc(&rt0->peer->refcnt);
-		x->u.rt.peer = rt0->peer;
-		/* Sheit... I remember I did this right. Apparently,
-		 * it was magically lost, so this code needs audit */
-		x->u.rt.rt_flags = rt0->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST|RTCF_LOCAL);
-		x->u.rt.rt_type = rt0->rt_type;
-		x->u.rt.rt_src = rt0->rt_src;
-		x->u.rt.rt_dst = rt0->rt_dst;
-		x->u.rt.rt_gateway = rt0->rt_gateway;
-		x->u.rt.rt_spec_dst = rt0->rt_spec_dst;
-		header_len -= x->u.dst.xfrm->props.header_len;
-		trailer_len -= x->u.dst.xfrm->props.trailer_len;
-	}
-
-	xfrm_init_pmtu(dst);
 	return 0;
-
-error:
-	if (dst)
-		dst_free(dst);
-	return err;
 }
 
 static void
@@ -330,8 +247,9 @@ static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
 	.dst_lookup =		xfrm4_dst_lookup,
 	.get_saddr =		xfrm4_get_saddr,
 	.find_bundle = 		__xfrm4_find_bundle,
-	.bundle_create =	__xfrm4_bundle_create,
 	.decode_session =	_decode_session4,
+	.get_tos =		xfrm4_get_tos,
+	.fill_dst =		xfrm4_fill_dst,
 };
 
 static void __init xfrm4_policy_init(void)
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 8e78530865a6..63932c5fd3c7 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -93,126 +93,33 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 	return dst;
 }
 
-/* Allocate chain of dst_entry's, attach known xfrm's, calculate
- * all the metrics... Shortly, bundle a bundle.
- */
-
-static int
-__xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int nx,
-		      struct flowi *fl, struct dst_entry **dst_p)
+static int xfrm6_get_tos(struct flowi *fl)
 {
-	struct dst_entry *dst, *dst_prev;
-	struct rt6_info *rt0 = (struct rt6_info*)(*dst_p);
-	struct rt6_info *rt  = rt0;
-	int i;
-	int err;
-	int header_len = 0;
-	int trailer_len = 0;
-
-	dst = dst_prev = NULL;
-	dst_hold(&rt->u.dst);
-
-	for (i = 0; i < nx; i++) {
-		struct dst_entry *dst1 = dst_alloc(&xfrm6_dst_ops);
-		struct xfrm_dst *xdst;
+	return 0;
+}
 
-		if (unlikely(dst1 == NULL)) {
-			err = -ENOBUFS;
-			dst_release(&rt->u.dst);
-			goto error;
-		}
+static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
+{
+	struct rt6_info *rt = (struct rt6_info*)xdst->route;
 
-		if (!dst)
-			dst = dst1;
-		else {
-			dst_prev->child = dst1;
-			dst1->flags |= DST_NOHASH;
-			dst_clone(dst1);
-		}
+	xdst->u.dst.dev = dev;
+	dev_hold(dev);
 
-		xdst = (struct xfrm_dst *)dst1;
-		xdst->route = &rt->u.dst;
-		xdst->genid = xfrm[i]->genid;
-		if (rt->rt6i_node)
-			xdst->route_cookie = rt->rt6i_node->fn_sernum;
-
-		dst1->next = dst_prev;
-		dst_prev = dst1;
-
-		if (xfrm[i]->type->flags & XFRM_TYPE_NON_FRAGMENT)
-			((struct rt6_info *)dst)->nfheader_len +=
-				xfrm[i]->props.header_len;
-		header_len += xfrm[i]->props.header_len;
-		trailer_len += xfrm[i]->props.trailer_len;
-
-		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
-			dst1 = xfrm_dst_lookup(xfrm[i], 0);
-			err = PTR_ERR(dst1);
-			if (IS_ERR(dst1))
-				goto error;
-
-			rt = (struct rt6_info *)dst1;
-		} else
-			dst_hold(&rt->u.dst);
-	}
+	xdst->u.rt6.rt6i_idev = in6_dev_get(rt->u.dst.dev);
+	if (!xdst->u.rt6.rt6i_idev)
+		return -ENODEV;
 
-	dst_prev->child = &rt->u.dst;
-	dst->path = &rt->u.dst;
-
-	/* Copy neighbour for reachability confirmation */
-	dst->neighbour = neigh_clone(rt->u.dst.neighbour);
-
-	if (rt->rt6i_node)
-		((struct xfrm_dst *)dst)->path_cookie = rt->rt6i_node->fn_sernum;
-
-	*dst_p = dst;
-	dst = dst_prev;
-
-	dst_prev = *dst_p;
-	i = 0;
-	err = -ENODEV;
-	for (; dst_prev != &rt->u.dst; dst_prev = dst_prev->child) {
-		struct xfrm_dst *x = (struct xfrm_dst*)dst_prev;
-
-		dst_prev->xfrm = xfrm[i++];
-		dst_prev->dev = rt->u.dst.dev;
-		if (!rt->u.dst.dev)
-			goto error;
-		dev_hold(rt->u.dst.dev);
-
-		x->u.rt6.rt6i_idev = in6_dev_get(rt->u.dst.dev);
-		if (!x->u.rt6.rt6i_idev)
-			goto error;
-
-		dst_prev->obsolete	= -1;
-		dst_prev->flags	       |= DST_HOST;
-		dst_prev->lastuse	= jiffies;
-		dst_prev->header_len	= header_len;
-		dst_prev->trailer_len	= trailer_len;
-		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
-
-		dst_prev->input = dst_discard;
-		dst_prev->output = dst_prev->xfrm->outer_mode->afinfo->output;
-		/* Sheit... I remember I did this right. Apparently,
-		 * it was magically lost, so this code needs audit */
-		x->u.rt6.rt6i_flags    = rt0->rt6i_flags&(RTF_ANYCAST|RTF_LOCAL);
-		x->u.rt6.rt6i_metric   = rt0->rt6i_metric;
-		x->u.rt6.rt6i_node     = rt0->rt6i_node;
-		x->u.rt6.rt6i_gateway  = rt0->rt6i_gateway;
-		memcpy(&x->u.rt6.rt6i_gateway, &rt0->rt6i_gateway, sizeof(x->u.rt6.rt6i_gateway));
-		x->u.rt6.rt6i_dst      = rt0->rt6i_dst;
-		x->u.rt6.rt6i_src      = rt0->rt6i_src;
-		header_len -= x->u.dst.xfrm->props.header_len;
-		trailer_len -= x->u.dst.xfrm->props.trailer_len;
-	}
+	/* Sheit... I remember I did this right. Apparently,
+	 * it was magically lost, so this code needs audit */
+	xdst->u.rt6.rt6i_flags = rt->rt6i_flags & (RTF_ANYCAST |
+						   RTF_LOCAL);
+	xdst->u.rt6.rt6i_metric = rt->rt6i_metric;
+	xdst->u.rt6.rt6i_node = rt->rt6i_node;
+	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
+	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
+	xdst->u.rt6.rt6i_src = rt->rt6i_src;
 
-	xfrm_init_pmtu(dst);
 	return 0;
-
-error:
-	if (dst)
-		dst_free(dst);
-	return err;
 }
 
 static inline void
@@ -355,8 +262,9 @@ static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
 	.dst_lookup =		xfrm6_dst_lookup,
 	.get_saddr = 		xfrm6_get_saddr,
 	.find_bundle =		__xfrm6_find_bundle,
-	.bundle_create =	__xfrm6_bundle_create,
 	.decode_session =	_decode_session6,
+	.get_tos =		xfrm6_get_tos,
+	.fill_dst =		xfrm6_fill_dst,
 };
 
 static void __init xfrm6_policy_init(void)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 085c19d4d1b7..b153f7482052 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -24,6 +24,7 @@
 #include <linux/netfilter.h>
 #include <linux/module.h>
 #include <linux/cache.h>
+#include <net/dst.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
 
@@ -50,6 +51,7 @@ static DEFINE_SPINLOCK(xfrm_policy_gc_lock);
 
 static struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family);
 static void xfrm_policy_put_afinfo(struct xfrm_policy_afinfo *afinfo);
+static void xfrm_init_pmtu(struct dst_entry *dst);
 
 static inline int
 __xfrm4_selector_match(struct xfrm_selector *sel, struct flowi *fl)
@@ -85,7 +87,8 @@ int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl,
 	return 0;
 }
 
-struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos)
+static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos,
+						int family)
 {
 	xfrm_address_t *saddr = &x->props.saddr;
 	xfrm_address_t *daddr = &x->id.daddr;
@@ -97,7 +100,7 @@ struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos)
 	if (x->type->flags & XFRM_TYPE_REMOTE_COADDR)
 		daddr = x->coaddr;
 
-	afinfo = xfrm_policy_get_afinfo(x->props.family);
+	afinfo = xfrm_policy_get_afinfo(family);
 	if (unlikely(afinfo == NULL))
 		return ERR_PTR(-EAFNOSUPPORT);
 
@@ -105,7 +108,6 @@ struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos)
 	xfrm_policy_put_afinfo(afinfo);
 	return dst;
 }
-EXPORT_SYMBOL(xfrm_dst_lookup);
 
 static inline unsigned long make_jiffies(long secs)
 {
@@ -1234,24 +1236,164 @@ xfrm_find_bundle(struct flowi *fl, struct xfrm_policy *policy, unsigned short fa
 	return x;
 }
 
-/* Allocate chain of dst_entry's, attach known xfrm's, calculate
- * all the metrics... Shortly, bundle a bundle.
- */
+static inline int xfrm_get_tos(struct flowi *fl, int family)
+{
+	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
+	int tos;
 
-static int
-xfrm_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int nx,
-		   struct flowi *fl, struct dst_entry **dst_p,
-		   unsigned short family)
+	if (!afinfo)
+		return -EINVAL;
+
+	tos = afinfo->get_tos(fl);
+
+	xfrm_policy_put_afinfo(afinfo);
+
+	return tos;
+}
+
+static inline struct xfrm_dst *xfrm_alloc_dst(int family)
 {
-	int err;
 	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
-	if (unlikely(afinfo == NULL))
+	struct xfrm_dst *xdst;
+
+	if (!afinfo)
+		return ERR_PTR(-EINVAL);
+
+	xdst = dst_alloc(afinfo->dst_ops) ?: ERR_PTR(-ENOBUFS);
+
+	xfrm_policy_put_afinfo(afinfo);
+
+	return xdst;
+}
+
+static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
+{
+	struct xfrm_policy_afinfo *afinfo =
+		xfrm_policy_get_afinfo(xdst->u.dst.ops->family);
+	int err;
+
+	if (!afinfo)
 		return -EINVAL;
-	err = afinfo->bundle_create(policy, xfrm, nx, fl, dst_p);
+
+	err = afinfo->fill_dst(xdst, dev);
+
 	xfrm_policy_put_afinfo(afinfo);
+
 	return err;
 }
 
+/* Allocate chain of dst_entry's, attach known xfrm's, calculate
+ * all the metrics... Shortly, bundle a bundle.
+ */
+
+static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
+					    struct xfrm_state **xfrm, int nx,
+					    struct flowi *fl,
+					    struct dst_entry *dst)
+{
+	unsigned long now = jiffies;
+	struct net_device *dev;
+	struct dst_entry *dst_prev = NULL;
+	struct dst_entry *dst0 = NULL;
+	int i = 0;
+	int err;
+	int header_len = 0;
+	int trailer_len = 0;
+	int tos;
+	int family = policy->selector.family;
+
+	tos = xfrm_get_tos(fl, family);
+	err = tos;
+	if (tos < 0)
+		goto put_states;
+
+	dst_hold(dst);
+
+	for (; i < nx; i++) {
+		struct xfrm_dst *xdst = xfrm_alloc_dst(family);
+		struct dst_entry *dst1 = &xdst->u.dst;
+
+		err = PTR_ERR(xdst);
+		if (IS_ERR(xdst)) {
+			dst_release(dst);
+			goto put_states;
+		}
+
+		if (!dst_prev)
+			dst0 = dst1;
+		else {
+			dst_prev->child = dst_clone(dst1);
+			dst1->flags |= DST_NOHASH;
+		}
+
+		xdst->route = dst;
+		memcpy(&dst1->metrics, &dst->metrics, sizeof(dst->metrics));
+
+		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
+			family = xfrm[i]->props.family;
+			dst = xfrm_dst_lookup(xfrm[i], tos, family);
+			err = PTR_ERR(dst);
+			if (IS_ERR(dst))
+				goto put_states;
+		} else
+			dst_hold(dst);
+
+		dst1->xfrm = xfrm[i];
+		xdst->genid = xfrm[i]->genid;
+
+		dst1->obsolete = -1;
+		dst1->flags |= DST_HOST;
+		dst1->lastuse = now;
+
+		dst1->input = dst_discard;
+		dst1->output = xfrm[i]->outer_mode->afinfo->output;
+
+		dst1->next = dst_prev;
+		dst_prev = dst1;
+
+		header_len += xfrm[i]->props.header_len;
+		trailer_len += xfrm[i]->props.trailer_len;
+	}
+
+	dst_prev->child = dst;
+	dst0->path = dst;
+
+	err = -ENODEV;
+	dev = dst->dev;
+	if (!dev)
+		goto free_dst;
+
+	/* Copy neighbout for reachability confirmation */
+	dst0->neighbour = neigh_clone(dst->neighbour);
+
+	xfrm_init_pmtu(dst_prev);
+
+	for (dst_prev = dst0; dst_prev != dst; dst_prev = dst_prev->child) {
+		struct xfrm_dst *xdst = (struct xfrm_dst *)dst_prev;
+
+		err = xfrm_fill_dst(xdst, dev);
+		if (err)
+			goto free_dst;
+
+		dst_prev->header_len = header_len;
+		dst_prev->trailer_len = trailer_len;
+		header_len -= xdst->u.dst.xfrm->props.header_len;
+		trailer_len -= xdst->u.dst.xfrm->props.trailer_len;
+	}
+
+out:
+	return dst0;
+
+put_states:
+	for (; i < nx; i++)
+		xfrm_state_put(xfrm[i]);
+free_dst:
+	if (dst0)
+		dst_free(dst0);
+	dst0 = ERR_PTR(err);
+	goto out;
+}
+
 static int inline
 xfrm_dst_alloc_copy(void **target, void *src, int size)
 {
@@ -1454,15 +1596,10 @@ restart:
 			return 0;
 		}
 
-		dst = dst_orig;
-		err = xfrm_bundle_create(policy, xfrm, nx, fl, &dst, family);
-
-		if (unlikely(err)) {
-			int i;
-			for (i=0; i<nx; i++)
-				xfrm_state_put(xfrm[i]);
+		dst = xfrm_bundle_create(policy, xfrm, nx, fl, dst_orig);
+		err = PTR_ERR(dst);
+		if (IS_ERR(dst))
 			goto error;
-		}
 
 		for (pi = 0; pi < npols; pi++) {
 			read_lock_bh(&pols[pi]->lock);
@@ -1886,7 +2023,7 @@ static int xfrm_flush_bundles(void)
 	return 0;
 }
 
-void xfrm_init_pmtu(struct dst_entry *dst)
+static void xfrm_init_pmtu(struct dst_entry *dst)
 {
 	do {
 		struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
@@ -1907,8 +2044,6 @@ void xfrm_init_pmtu(struct dst_entry *dst)
 	} while ((dst = dst->next));
 }
 
-EXPORT_SYMBOL(xfrm_init_pmtu);
-
 /* Check that the bundle accepts the flow and its components are
  * still valid.
  */
-- 
cgit v1.2.3


From 29bb43b4ec4e625b0659186fc8a7c8f8b7c81982 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:40:13 -0800
Subject: [INET]: Give outer DSCP directly to ip*_copy_dscp

This patch changes the prototype of ipv4_copy_dscp and ipv6_copy_dscp so
that they directly take the outer DSCP rather than the outer IP header.
This will help us to unify the code for inter-family tunnels.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_ecn.h       | 8 ++++----
 net/ipv4/xfrm4_mode_tunnel.c | 2 +-
 net/ipv6/ip6_tunnel.c        | 2 +-
 net/ipv6/xfrm6_mode_tunnel.c | 3 ++-
 4 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h
index de8399a79774..ba33db053854 100644
--- a/include/net/inet_ecn.h
+++ b/include/net/inet_ecn.h
@@ -83,9 +83,9 @@ static inline void IP_ECN_clear(struct iphdr *iph)
 	iph->tos &= ~INET_ECN_MASK;
 }
 
-static inline void ipv4_copy_dscp(struct iphdr *outer, struct iphdr *inner)
+static inline void ipv4_copy_dscp(unsigned int dscp, struct iphdr *inner)
 {
-	u32 dscp = ipv4_get_dsfield(outer) & ~INET_ECN_MASK;
+	dscp &= ~INET_ECN_MASK;
 	ipv4_change_dsfield(inner, INET_ECN_MASK, dscp);
 }
 
@@ -104,9 +104,9 @@ static inline void IP6_ECN_clear(struct ipv6hdr *iph)
 	*(__be32*)iph &= ~htonl(INET_ECN_MASK << 20);
 }
 
-static inline void ipv6_copy_dscp(struct ipv6hdr *outer, struct ipv6hdr *inner)
+static inline void ipv6_copy_dscp(unsigned int dscp, struct ipv6hdr *inner)
 {
-	u32 dscp = ipv6_get_dsfield(outer) & ~INET_ECN_MASK;
+	dscp &= ~INET_ECN_MASK;
 	ipv6_change_dsfield(inner, INET_ECN_MASK, dscp);
 }
 
diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c
index e4deecba6dd2..68a9f56ff09f 100644
--- a/net/ipv4/xfrm4_mode_tunnel.c
+++ b/net/ipv4/xfrm4_mode_tunnel.c
@@ -113,7 +113,7 @@ static int xfrm4_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 	iph = ip_hdr(skb);
 	if (iph->protocol == IPPROTO_IPIP) {
 		if (x->props.flags & XFRM_STATE_DECAP_DSCP)
-			ipv4_copy_dscp(iph, ipip_hdr(skb));
+			ipv4_copy_dscp(ipv4_get_dsfield(iph), ipip_hdr(skb));
 		if (!(x->props.flags & XFRM_STATE_NOECN))
 			ipip_ecn_decapsulate(skb);
 	}
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 5383b33db8ca..a4051afaf776 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -635,7 +635,7 @@ static void ip6ip6_dscp_ecn_decapsulate(struct ip6_tnl *t,
 					struct sk_buff *skb)
 {
 	if (t->parms.flags & IP6_TNL_F_RCV_DSCP_COPY)
-		ipv6_copy_dscp(ipv6h, ipv6_hdr(skb));
+		ipv6_copy_dscp(ipv6_get_dsfield(ipv6h), ipv6_hdr(skb));
 
 	if (INET_ECN_is_ce(ipv6_get_dsfield(ipv6h)))
 		IP6_ECN_set_ce(ipv6_hdr(skb));
diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c
index fd84e2217274..9a43ea722481 100644
--- a/net/ipv6/xfrm6_mode_tunnel.c
+++ b/net/ipv6/xfrm6_mode_tunnel.c
@@ -95,7 +95,8 @@ static int xfrm6_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 	nh = skb_network_header(skb);
 	if (nh[IP6CB(skb)->nhoff] == IPPROTO_IPV6) {
 		if (x->props.flags & XFRM_STATE_DECAP_DSCP)
-			ipv6_copy_dscp(ipv6_hdr(skb), ipipv6_hdr(skb));
+			ipv6_copy_dscp(ipv6_get_dsfield(ipv6_hdr(skb)),
+				       ipipv6_hdr(skb));
 		if (!(x->props.flags & XFRM_STATE_NOECN))
 			ipip6_ecn_decapsulate(skb);
 	} else {
-- 
cgit v1.2.3


From 36cf9acf93e8561d9faec24849e57688a81eb9c5 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:40:52 -0800
Subject: [IPSEC]: Separate inner/outer mode processing on output

With inter-family transforms the inner mode differs from the outer
mode.  Attempting to handle both sides from the same function means
that it needs to handle both IPv4 and IPv6 which creates duplication
and confusion.

This patch separates the two parts on the output path so that each
function deals with one family only.

In particular, the functions xfrm4_extract_output/xfrm6_extract_output
moves the pertinent fields from the IPv4/IPv6 IP headers into a
neutral format stored in skb->cb.  This is then used by the outer mode
output functions to write the outer IP header.  In this way the output
function no longer has to know about the inner address family.

Since the extract functions are only called by tunnel modes (the only
modes that can support inter-family transforms), I've also moved the
xfrm*_tunnel_check_size calls into them.  This allows the correct ICMP
message to be sent as opposed to now where you might call icmp_send
with an IPv6 packet and vice versa.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h           | 51 +++++++++++++++++++++++++++++++++++++++++++-
 net/ipv4/xfrm4_mode_beet.c   | 15 +++++++++++--
 net/ipv4/xfrm4_mode_tunnel.c | 37 ++++++++++----------------------
 net/ipv4/xfrm4_output.c      | 41 +++++++++++++++++++++++++----------
 net/ipv4/xfrm4_state.c       | 17 +++++++++++++++
 net/ipv6/xfrm6_mode_beet.c   | 28 ++++++++++++------------
 net/ipv6/xfrm6_mode_tunnel.c | 31 +++++++++------------------
 net/ipv6/xfrm6_output.c      | 39 +++++++++++++++++++++++++--------
 net/ipv6/xfrm6_state.c       | 18 ++++++++++++++++
 9 files changed, 193 insertions(+), 84 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index c9345fe3f8d2..138c1868be1d 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -257,6 +257,7 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 
 struct xfrm_state_afinfo {
 	unsigned int		family;
+	unsigned int		proto;
 	struct module		*owner;
 	struct xfrm_type	*type_map[IPPROTO_MAX];
 	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];
@@ -267,6 +268,8 @@ struct xfrm_state_afinfo {
 	int			(*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n);
 	int			(*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n);
 	int			(*output)(struct sk_buff *skb);
+	int			(*extract_output)(struct xfrm_state *x,
+						  struct sk_buff *skb);
 };
 
 extern int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo);
@@ -312,7 +315,18 @@ struct xfrm_mode {
 	 * header.  The value of the network header will always point
 	 * to the top IP header while skb->data will point to the payload.
 	 */
-	int (*output)(struct xfrm_state *x,struct sk_buff *skb);
+	int (*output2)(struct xfrm_state *x,struct sk_buff *skb);
+
+	/*
+	 * This is the actual output entry point.
+	 *
+	 * For transport mode and equivalent this would be identical to
+	 * output2 (which does not need to be set).  While tunnel mode
+	 * and equivalent would set this to a tunnel encapsulation function
+	 * (xfrm4_prepare_output or xfrm6_prepare_output) that would in turn
+	 * call output2.
+	 */
+	int (*output)(struct xfrm_state *x, struct sk_buff *skb);
 
 	struct xfrm_state_afinfo *afinfo;
 	struct module *owner;
@@ -454,6 +468,35 @@ struct xfrm_skb_cb {
 
 #define XFRM_SKB_CB(__skb) ((struct xfrm_skb_cb *)&((__skb)->cb[0]))
 
+/*
+ * This structure is used by the afinfo prepare_input/prepare_output functions
+ * to transmit header information to the mode input/output functions.
+ */
+struct xfrm_mode_skb_cb {
+	union {
+		struct inet_skb_parm h4;
+		struct inet6_skb_parm h6;
+	} header;
+
+	/* Copied from header for IPv4, always set to zero and DF for IPv6. */
+	__be16 id;
+	__be16 frag_off;
+
+	/* TOS for IPv4, class for IPv6. */
+	u8 tos;
+
+	/* TTL for IPv4, hop limitfor IPv6. */
+	u8 ttl;
+
+	/* Protocol for IPv4, NH for IPv6. */
+	u8 protocol;
+
+	/* Used by IPv6 only, zero for IPv4. */
+	u8 flow_lbl[3];
+};
+
+#define XFRM_MODE_SKB_CB(__skb) ((struct xfrm_mode_skb_cb *)&((__skb)->cb[0]))
+
 /* Audit Information */
 struct xfrm_audit
 {
@@ -1051,6 +1094,7 @@ extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
 extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm_output(struct sk_buff *skb);
+extern int xfrm4_extract_header(struct sk_buff *skb);
 extern int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 			   int encap_type);
 extern int xfrm4_rcv(struct sk_buff *skb);
@@ -1060,9 +1104,12 @@ static inline int xfrm4_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 	return xfrm4_rcv_encap(skb, nexthdr, spi, 0);
 }
 
+extern int xfrm4_extract_output(struct xfrm_state *x, struct sk_buff *skb);
+extern int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm4_output(struct sk_buff *skb);
 extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short family);
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler, unsigned short family);
+extern int xfrm6_extract_header(struct sk_buff *skb);
 extern int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi);
 extern int xfrm6_rcv(struct sk_buff *skb);
 extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
@@ -1072,6 +1119,8 @@ extern int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler, unsigned short
 extern __be32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr);
 extern void xfrm6_tunnel_free_spi(xfrm_address_t *saddr);
 extern __be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr);
+extern int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb);
+extern int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm6_output(struct sk_buff *skb);
 extern int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
 				 u8 **prevhdr);
diff --git a/net/ipv4/xfrm4_mode_beet.c b/net/ipv4/xfrm4_mode_beet.c
index e42e122414be..94842adce144 100644
--- a/net/ipv4/xfrm4_mode_beet.c
+++ b/net/ipv4/xfrm4_mode_beet.c
@@ -43,7 +43,17 @@ static int xfrm4_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 	ph = (struct ip_beet_phdr *)__skb_pull(skb, sizeof(*iph) - hdrlen);
 
 	top_iph = ip_hdr(skb);
-	memmove(top_iph, iph, sizeof(*iph));
+
+	top_iph->ihl = 5;
+	top_iph->version = 4;
+
+	top_iph->protocol = XFRM_MODE_SKB_CB(skb)->protocol;
+	top_iph->tos = XFRM_MODE_SKB_CB(skb)->tos;
+
+	top_iph->id = XFRM_MODE_SKB_CB(skb)->id;
+	top_iph->frag_off = XFRM_MODE_SKB_CB(skb)->frag_off;
+	top_iph->ttl = XFRM_MODE_SKB_CB(skb)->ttl;
+
 	if (unlikely(optlen)) {
 		BUG_ON(optlen < 0);
 
@@ -111,7 +121,8 @@ out:
 
 static struct xfrm_mode xfrm4_beet_mode = {
 	.input = xfrm4_beet_input,
-	.output = xfrm4_beet_output,
+	.output2 = xfrm4_beet_output,
+	.output = xfrm4_prepare_output,
 	.owner = THIS_MODULE,
 	.encap = XFRM_MODE_BEET,
 	.flags = XFRM_MODE_FLAG_TUNNEL,
diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c
index 68a9f56ff09f..cc8bbb274e37 100644
--- a/net/ipv4/xfrm4_mode_tunnel.c
+++ b/net/ipv4/xfrm4_mode_tunnel.c
@@ -36,53 +36,37 @@ static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
 static int xfrm4_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
-	struct xfrm_dst *xdst = (struct xfrm_dst*)dst;
-	struct iphdr *iph, *top_iph;
+	struct iphdr *top_iph;
 	int flags;
 
-	iph = ip_hdr(skb);
-
 	skb_set_network_header(skb, -x->props.header_len);
 	skb->mac_header = skb->network_header +
 			  offsetof(struct iphdr, protocol);
-	skb->transport_header = skb->network_header + sizeof(*iph);
+	skb->transport_header = skb->network_header + sizeof(*top_iph);
 	top_iph = ip_hdr(skb);
 
 	top_iph->ihl = 5;
 	top_iph->version = 4;
 
-	flags = x->props.flags;
+	top_iph->protocol = x->inner_mode->afinfo->proto;
 
 	/* DS disclosed */
-	if (xdst->route->ops->family == AF_INET) {
-		top_iph->protocol = IPPROTO_IPIP;
-		top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos);
-		top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ?
-			0 : (iph->frag_off & htons(IP_DF));
-	}
-#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
-	else {
-		struct ipv6hdr *ipv6h = (struct ipv6hdr*)iph;
-		top_iph->protocol = IPPROTO_IPV6;
-		top_iph->tos = INET_ECN_encapsulate(iph->tos, ipv6_get_dsfield(ipv6h));
-		top_iph->frag_off = 0;
-	}
-#endif
+	top_iph->tos = INET_ECN_encapsulate(XFRM_MODE_SKB_CB(skb)->tos,
+					    XFRM_MODE_SKB_CB(skb)->tos);
 
+	flags = x->props.flags;
 	if (flags & XFRM_STATE_NOECN)
 		IP_ECN_clear(top_iph);
 
-	if (!top_iph->frag_off)
-		__ip_select_ident(top_iph, dst->child, 0);
+	top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ?
+			    0 : XFRM_MODE_SKB_CB(skb)->frag_off;
+	ip_select_ident(top_iph, dst->child, NULL);
 
 	top_iph->ttl = dst_metric(dst->child, RTAX_HOPLIMIT);
 
 	top_iph->saddr = x->props.saddr.a4;
 	top_iph->daddr = x->id.daddr.a4;
 
-	skb->protocol = htons(ETH_P_IP);
-
-	memset(&(IPCB(skb)->opt), 0, sizeof(struct ip_options));
 	return 0;
 }
 
@@ -136,7 +120,8 @@ out:
 
 static struct xfrm_mode xfrm4_tunnel_mode = {
 	.input = xfrm4_tunnel_input,
-	.output = xfrm4_tunnel_output,
+	.output2 = xfrm4_tunnel_output,
+	.output = xfrm4_prepare_output,
 	.owner = THIS_MODULE,
 	.encap = XFRM_MODE_TUNNEL,
 	.flags = XFRM_MODE_FLAG_TUNNEL,
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index c4a7156962bd..13fd11335e28 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -8,11 +8,12 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#include <linux/compiler.h>
 #include <linux/if_ether.h>
 #include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/skbuff.h>
 #include <linux/netfilter_ipv4.h>
+#include <net/dst.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
 #include <net/icmp.h>
@@ -25,8 +26,6 @@ static int xfrm4_tunnel_check_size(struct sk_buff *skb)
 	if (IPCB(skb)->flags & IPSKB_XFRM_TUNNEL_SIZE)
 		goto out;
 
-	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
-
 	if (!(ip_hdr(skb)->frag_off & htons(IP_DF)) || skb->local_df)
 		goto out;
 
@@ -40,19 +39,39 @@ out:
 	return ret;
 }
 
+int xfrm4_extract_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	err = xfrm4_tunnel_check_size(skb);
+	if (err)
+		return err;
+
+	return xfrm4_extract_header(skb);
+}
+
+int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	err = x->inner_mode->afinfo->extract_output(x, skb);
+	if (err)
+		return err;
+
+	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
+
+	skb->protocol = htons(ETH_P_IP);
+
+	return x->outer_mode->output2(x, skb);
+}
+EXPORT_SYMBOL(xfrm4_prepare_output);
+
 static inline int xfrm4_output_one(struct sk_buff *skb)
 {
-	struct dst_entry *dst = skb->dst;
-	struct xfrm_state *x = dst->xfrm;
 	struct iphdr *iph;
 	int err;
 
-	if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
-		err = xfrm4_tunnel_check_size(skb);
-		if (err)
-			goto error_nolock;
-	}
-
 	err = xfrm_output(skb);
 	if (err)
 		goto error_nolock;
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 13d54a1c3337..e6030e74ff65 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -47,12 +47,29 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.family = AF_INET;
 }
 
+int xfrm4_extract_header(struct sk_buff *skb)
+{
+	struct iphdr *iph = ip_hdr(skb);
+
+	XFRM_MODE_SKB_CB(skb)->id = iph->id;
+	XFRM_MODE_SKB_CB(skb)->frag_off = iph->frag_off;
+	XFRM_MODE_SKB_CB(skb)->tos = iph->tos;
+	XFRM_MODE_SKB_CB(skb)->ttl = iph->ttl;
+	XFRM_MODE_SKB_CB(skb)->protocol = iph->protocol;
+	memset(XFRM_MODE_SKB_CB(skb)->flow_lbl, 0,
+	       sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl));
+
+	return 0;
+}
+
 static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
+	.proto			= IPPROTO_IPIP,
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
 	.output			= xfrm4_output,
+	.extract_output		= xfrm4_extract_output,
 };
 
 void __init xfrm4_state_init(void)
diff --git a/net/ipv6/xfrm6_mode_beet.c b/net/ipv6/xfrm6_mode_beet.c
index 2bfb4f05c14c..4988ed9c76c6 100644
--- a/net/ipv6/xfrm6_mode_beet.c
+++ b/net/ipv6/xfrm6_mode_beet.c
@@ -25,25 +25,24 @@
  */
 static int xfrm6_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct ipv6hdr *iph, *top_iph;
-	u8 *prevhdr;
-	int hdr_len;
+	struct ipv6hdr *top_iph;
 
-	iph = ipv6_hdr(skb);
-
-	hdr_len = ip6_find_1stfragopt(skb, &prevhdr);
-
-	skb_set_mac_header(skb, (prevhdr - x->props.header_len) - skb->data);
 	skb_set_network_header(skb, -x->props.header_len);
-	skb->transport_header = skb->network_header + hdr_len;
-	__skb_pull(skb, hdr_len);
-
+	skb->mac_header = skb->network_header +
+			  offsetof(struct ipv6hdr, nexthdr);
+	skb->transport_header = skb->network_header + sizeof(*top_iph);
 	top_iph = ipv6_hdr(skb);
-	memmove(top_iph, iph, hdr_len);
 
+	top_iph->version = 6;
+
+	memcpy(top_iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl,
+	       sizeof(top_iph->flow_lbl));
+	top_iph->nexthdr = XFRM_MODE_SKB_CB(skb)->protocol;
+
+	ipv6_change_dsfield(top_iph, 0, XFRM_MODE_SKB_CB(skb)->tos);
+	top_iph->hop_limit = XFRM_MODE_SKB_CB(skb)->ttl;
 	ipv6_addr_copy(&top_iph->saddr, (struct in6_addr *)&x->props.saddr);
 	ipv6_addr_copy(&top_iph->daddr, (struct in6_addr *)&x->id.daddr);
-
 	return 0;
 }
 
@@ -76,7 +75,8 @@ out:
 
 static struct xfrm_mode xfrm6_beet_mode = {
 	.input = xfrm6_beet_input,
-	.output = xfrm6_beet_output,
+	.output2 = xfrm6_beet_output,
+	.output = xfrm6_prepare_output,
 	.owner = THIS_MODULE,
 	.encap = XFRM_MODE_BEET,
 	.flags = XFRM_MODE_FLAG_TUNNEL,
diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c
index 9a43ea722481..d45ce5d44197 100644
--- a/net/ipv6/xfrm6_mode_tunnel.c
+++ b/net/ipv6/xfrm6_mode_tunnel.c
@@ -38,33 +38,22 @@ static inline void ip6ip_ecn_decapsulate(struct sk_buff *skb)
 static int xfrm6_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
-	struct xfrm_dst *xdst = (struct xfrm_dst*)dst;
-	struct ipv6hdr *iph, *top_iph;
+	struct ipv6hdr *top_iph;
 	int dsfield;
 
-	iph = ipv6_hdr(skb);
-
 	skb_set_network_header(skb, -x->props.header_len);
 	skb->mac_header = skb->network_header +
 			  offsetof(struct ipv6hdr, nexthdr);
-	skb->transport_header = skb->network_header + sizeof(*iph);
+	skb->transport_header = skb->network_header + sizeof(*top_iph);
 	top_iph = ipv6_hdr(skb);
 
 	top_iph->version = 6;
-	if (xdst->route->ops->family == AF_INET6) {
-		top_iph->priority = iph->priority;
-		top_iph->flow_lbl[0] = iph->flow_lbl[0];
-		top_iph->flow_lbl[1] = iph->flow_lbl[1];
-		top_iph->flow_lbl[2] = iph->flow_lbl[2];
-		top_iph->nexthdr = IPPROTO_IPV6;
-	} else {
-		top_iph->priority = 0;
-		top_iph->flow_lbl[0] = 0;
-		top_iph->flow_lbl[1] = 0;
-		top_iph->flow_lbl[2] = 0;
-		top_iph->nexthdr = IPPROTO_IPIP;
-	}
-	dsfield = ipv6_get_dsfield(top_iph);
+
+	memcpy(top_iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl,
+	       sizeof(top_iph->flow_lbl));
+	top_iph->nexthdr = x->inner_mode->afinfo->proto;
+
+	dsfield = XFRM_MODE_SKB_CB(skb)->tos;
 	dsfield = INET_ECN_encapsulate(dsfield, dsfield);
 	if (x->props.flags & XFRM_STATE_NOECN)
 		dsfield &= ~INET_ECN_MASK;
@@ -72,7 +61,6 @@ static int xfrm6_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 	top_iph->hop_limit = dst_metric(dst->child, RTAX_HOPLIMIT);
 	ipv6_addr_copy(&top_iph->saddr, (struct in6_addr *)&x->props.saddr);
 	ipv6_addr_copy(&top_iph->daddr, (struct in6_addr *)&x->id.daddr);
-	skb->protocol = htons(ETH_P_IPV6);
 	return 0;
 }
 
@@ -116,7 +104,8 @@ out:
 
 static struct xfrm_mode xfrm6_tunnel_mode = {
 	.input = xfrm6_tunnel_input,
-	.output = xfrm6_tunnel_output,
+	.output2 = xfrm6_tunnel_output,
+	.output = xfrm6_prepare_output,
 	.owner = THIS_MODULE,
 	.encap = XFRM_MODE_TUNNEL,
 	.flags = XFRM_MODE_FLAG_TUNNEL,
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 656976760ad4..bc2e80e3b0b1 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -10,10 +10,12 @@
  */
 
 #include <linux/if_ether.h>
-#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/skbuff.h>
 #include <linux/icmpv6.h>
 #include <linux/netfilter_ipv6.h>
+#include <net/dst.h>
 #include <net/ipv6.h>
 #include <net/xfrm.h>
 
@@ -43,19 +45,38 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb)
 	return ret;
 }
 
+int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	err = xfrm6_tunnel_check_size(skb);
+	if (err)
+		return err;
+
+	return xfrm6_extract_header(skb);
+}
+
+int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	err = x->inner_mode->afinfo->extract_output(x, skb);
+	if (err)
+		return err;
+
+	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
+
+	skb->protocol = htons(ETH_P_IPV6);
+
+	return x->outer_mode->output2(x, skb);
+}
+EXPORT_SYMBOL(xfrm6_prepare_output);
+
 static inline int xfrm6_output_one(struct sk_buff *skb)
 {
-	struct dst_entry *dst = skb->dst;
-	struct xfrm_state *x = dst->xfrm;
 	struct ipv6hdr *iph;
 	int err;
 
-	if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
-		err = xfrm6_tunnel_check_size(skb);
-		if (err)
-			goto error_nolock;
-	}
-
 	err = xfrm_output(skb);
 	if (err)
 		goto error_nolock;
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index b392bee396f1..98b05f472322 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -14,6 +14,7 @@
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
+#include <net/dsfield.h>
 #include <net/ipv6.h>
 #include <net/addrconf.h>
 
@@ -168,13 +169,30 @@ __xfrm6_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n)
 	return 0;
 }
 
+int xfrm6_extract_header(struct sk_buff *skb)
+{
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+
+	XFRM_MODE_SKB_CB(skb)->id = 0;
+	XFRM_MODE_SKB_CB(skb)->frag_off = htons(IP_DF);
+	XFRM_MODE_SKB_CB(skb)->tos = ipv6_get_dsfield(iph);
+	XFRM_MODE_SKB_CB(skb)->ttl = iph->hop_limit;
+	XFRM_MODE_SKB_CB(skb)->protocol = iph->nexthdr;
+	memcpy(XFRM_MODE_SKB_CB(skb)->flow_lbl, iph->flow_lbl,
+	       sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl));
+
+	return 0;
+}
+
 static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
+	.proto			= IPPROTO_IPV6,
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
 	.state_sort		= __xfrm6_state_sort,
 	.output			= xfrm6_output,
+	.extract_output		= xfrm6_extract_output,
 };
 
 void __init xfrm6_state_init(void)
-- 
cgit v1.2.3


From 227620e295090629fcb2c46ad3828222ab65438d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:41:28 -0800
Subject: [IPSEC]: Separate inner/outer mode processing on input

With inter-family transforms the inner mode differs from the outer
mode.  Attempting to handle both sides from the same function means
that it needs to handle both IPv4 and IPv6 which creates duplication
and confusion.

This patch separates the two parts on the input path so that each
function deals with one family only.

In particular, the functions xfrm4_extract_inut/xfrm6_extract_inut
moves the pertinent fields from the IPv4/IPv6 IP headers into a
neutral format stored in skb->cb.  This is then used by the inner mode
input functions to modify the inner IP header.  In this way the input
function no longer has to know about the outer address family.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h           | 27 ++++++++++++++++++
 net/ipv4/xfrm4_input.c       |  7 ++++-
 net/ipv4/xfrm4_mode_beet.c   | 67 ++++++++++++++++++++++++++------------------
 net/ipv4/xfrm4_mode_tunnel.c | 44 +++++++----------------------
 net/ipv4/xfrm4_state.c       |  2 ++
 net/ipv6/xfrm6_input.c       |  7 ++++-
 net/ipv6/xfrm6_mode_beet.c   | 36 ++++++++++++++++--------
 net/ipv6/xfrm6_mode_tunnel.c | 31 ++++++--------------
 net/ipv6/xfrm6_output.c      |  1 +
 net/ipv6/xfrm6_state.c       |  5 +++-
 net/xfrm/xfrm_input.c        | 13 +++++++++
 11 files changed, 141 insertions(+), 99 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 138c1868be1d..a9dbe091ae58 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -258,6 +258,7 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 struct xfrm_state_afinfo {
 	unsigned int		family;
 	unsigned int		proto;
+	unsigned int		eth_proto;
 	struct module		*owner;
 	struct xfrm_type	*type_map[IPPROTO_MAX];
 	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];
@@ -268,6 +269,8 @@ struct xfrm_state_afinfo {
 	int			(*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n);
 	int			(*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n);
 	int			(*output)(struct sk_buff *skb);
+	int			(*extract_input)(struct xfrm_state *x,
+						 struct sk_buff *skb);
 	int			(*extract_output)(struct xfrm_state *x,
 						  struct sk_buff *skb);
 };
@@ -302,6 +305,27 @@ extern int xfrm_register_type(struct xfrm_type *type, unsigned short family);
 extern int xfrm_unregister_type(struct xfrm_type *type, unsigned short family);
 
 struct xfrm_mode {
+	/*
+	 * Remove encapsulation header.
+	 *
+	 * The IP header will be moved over the top of the encapsulation
+	 * header.
+	 *
+	 * On entry, the transport header shall point to where the IP header
+	 * should be and the network header shall be set to where the IP
+	 * header currently is.  skb->data shall point to the start of the
+	 * payload.
+	 */
+	int (*input2)(struct xfrm_state *x, struct sk_buff *skb);
+
+	/*
+	 * This is the actual input entry point.
+	 *
+	 * For transport mode and equivalent this would be identical to
+	 * input2 (which does not need to be set).  While tunnel mode
+	 * and equivalent would set this to the tunnel encapsulation function
+	 * xfrm4_prepare_input that would in turn call input2.
+	 */
 	int (*input)(struct xfrm_state *x, struct sk_buff *skb);
 
 	/*
@@ -1093,8 +1117,10 @@ extern void xfrm_replay_advance(struct xfrm_state *x, __be32 seq);
 extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
 extern int xfrm_init_state(struct xfrm_state *x);
+extern int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm_output(struct sk_buff *skb);
 extern int xfrm4_extract_header(struct sk_buff *skb);
+extern int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 			   int encap_type);
 extern int xfrm4_rcv(struct sk_buff *skb);
@@ -1110,6 +1136,7 @@ extern int xfrm4_output(struct sk_buff *skb);
 extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short family);
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler, unsigned short family);
 extern int xfrm6_extract_header(struct sk_buff *skb);
+extern int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi);
 extern int xfrm6_rcv(struct sk_buff *skb);
 extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index 5e95c8a07efb..c0323d05ab69 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -16,6 +16,11 @@
 #include <net/ip.h>
 #include <net/xfrm.h>
 
+int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	return xfrm4_extract_header(skb);
+}
+
 #ifdef CONFIG_NETFILTER
 static inline int xfrm4_rcv_encap_finish(struct sk_buff *skb)
 {
@@ -91,7 +96,7 @@ int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 
 		xfrm_vec[xfrm_nr++] = x;
 
-		if (x->outer_mode->input(x, skb))
+		if (x->inner_mode->input(x, skb))
 			goto drop;
 
 		if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
diff --git a/net/ipv4/xfrm4_mode_beet.c b/net/ipv4/xfrm4_mode_beet.c
index 94842adce144..e093a7b59e18 100644
--- a/net/ipv4/xfrm4_mode_beet.c
+++ b/net/ipv4/xfrm4_mode_beet.c
@@ -17,6 +17,21 @@
 #include <net/ip.h>
 #include <net/xfrm.h>
 
+static void xfrm4_beet_make_header(struct sk_buff *skb)
+{
+	struct iphdr *iph = ip_hdr(skb);
+
+	iph->ihl = 5;
+	iph->version = 4;
+
+	iph->protocol = XFRM_MODE_SKB_CB(skb)->protocol;
+	iph->tos = XFRM_MODE_SKB_CB(skb)->tos;
+
+	iph->id = XFRM_MODE_SKB_CB(skb)->id;
+	iph->frag_off = XFRM_MODE_SKB_CB(skb)->frag_off;
+	iph->ttl = XFRM_MODE_SKB_CB(skb)->ttl;
+}
+
 /* Add encapsulation header.
  *
  * The top IP header will be constructed per draft-nikander-esp-beet-mode-06.txt.
@@ -40,20 +55,12 @@ static int xfrm4_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 			  offsetof(struct iphdr, protocol);
 	skb->transport_header = skb->network_header + sizeof(*iph);
 
+	xfrm4_beet_make_header(skb);
+
 	ph = (struct ip_beet_phdr *)__skb_pull(skb, sizeof(*iph) - hdrlen);
 
 	top_iph = ip_hdr(skb);
 
-	top_iph->ihl = 5;
-	top_iph->version = 4;
-
-	top_iph->protocol = XFRM_MODE_SKB_CB(skb)->protocol;
-	top_iph->tos = XFRM_MODE_SKB_CB(skb)->tos;
-
-	top_iph->id = XFRM_MODE_SKB_CB(skb)->id;
-	top_iph->frag_off = XFRM_MODE_SKB_CB(skb)->frag_off;
-	top_iph->ttl = XFRM_MODE_SKB_CB(skb)->ttl;
-
 	if (unlikely(optlen)) {
 		BUG_ON(optlen < 0);
 
@@ -75,43 +82,46 @@ static int xfrm4_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 
 static int xfrm4_beet_input(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct iphdr *iph = ip_hdr(skb);
-	int phlen = 0;
+	struct iphdr *iph;
 	int optlen = 0;
-	u8 ph_nexthdr = 0;
 	int err = -EINVAL;
 
-	if (unlikely(iph->protocol == IPPROTO_BEETPH)) {
+	if (unlikely(XFRM_MODE_SKB_CB(skb)->protocol == IPPROTO_BEETPH)) {
 		struct ip_beet_phdr *ph;
+		int phlen;
 
 		if (!pskb_may_pull(skb, sizeof(*ph)))
 			goto out;
-		ph = (struct ip_beet_phdr *)(ipip_hdr(skb) + 1);
+
+		ph = (struct ip_beet_phdr *)skb->data;
 
 		phlen = sizeof(*ph) + ph->padlen;
 		optlen = ph->hdrlen * 8 + (IPV4_BEET_PHMAXLEN - phlen);
 		if (optlen < 0 || optlen & 3 || optlen > 250)
 			goto out;
 
-		if (!pskb_may_pull(skb, phlen + optlen))
-			goto out;
-		skb->len -= phlen + optlen;
+		XFRM_MODE_SKB_CB(skb)->protocol = ph->nexthdr;
 
-		ph_nexthdr = ph->nexthdr;
+		if (!pskb_may_pull(skb, phlen));
+			goto out;
+		__skb_pull(skb, phlen);
 	}
 
-	skb_set_network_header(skb, phlen - sizeof(*iph));
-	memmove(skb_network_header(skb), iph, sizeof(*iph));
-	skb_set_transport_header(skb, phlen + optlen);
-	skb->data = skb_transport_header(skb);
+	skb_push(skb, sizeof(*iph));
+	skb_reset_network_header(skb);
+
+	memmove(skb->data - skb->mac_len, skb_mac_header(skb),
+		skb->mac_len);
+	skb_set_mac_header(skb, -skb->mac_len);
+
+	xfrm4_beet_make_header(skb);
 
 	iph = ip_hdr(skb);
-	iph->ihl = (sizeof(*iph) + optlen) / 4;
-	iph->tot_len = htons(skb->len + iph->ihl * 4);
+
+	iph->ihl += optlen / 4;
+	iph->tot_len = htons(skb->len);
 	iph->daddr = x->sel.daddr.a4;
 	iph->saddr = x->sel.saddr.a4;
-	if (ph_nexthdr)
-		iph->protocol = ph_nexthdr;
 	iph->check = 0;
 	iph->check = ip_fast_csum(skb_network_header(skb), iph->ihl);
 	err = 0;
@@ -120,7 +130,8 @@ out:
 }
 
 static struct xfrm_mode xfrm4_beet_mode = {
-	.input = xfrm4_beet_input,
+	.input2 = xfrm4_beet_input,
+	.input = xfrm_prepare_input,
 	.output2 = xfrm4_beet_output,
 	.output = xfrm4_prepare_output,
 	.owner = THIS_MODULE,
diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c
index cc8bbb274e37..aa335dba8ffa 100644
--- a/net/ipv4/xfrm4_mode_tunnel.c
+++ b/net/ipv4/xfrm4_mode_tunnel.c
@@ -16,19 +16,12 @@
 
 static inline void ipip_ecn_decapsulate(struct sk_buff *skb)
 {
-	struct iphdr *outer_iph = ip_hdr(skb);
 	struct iphdr *inner_iph = ipip_hdr(skb);
 
-	if (INET_ECN_is_ce(outer_iph->tos))
+	if (INET_ECN_is_ce(XFRM_MODE_SKB_CB(skb)->tos))
 		IP_ECN_set_ce(inner_iph);
 }
 
-static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
-{
-	if (INET_ECN_is_ce(iph->tos))
-		IP6_ECN_set_ce(ipv6_hdr(skb));
-}
-
 /* Add encapsulation header.
  *
  * The top IP header will be constructed per RFC 2401.
@@ -72,20 +65,11 @@ static int xfrm4_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 
 static int xfrm4_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct iphdr *iph = ip_hdr(skb);
 	const unsigned char *old_mac;
 	int err = -EINVAL;
 
-	switch (iph->protocol){
-		case IPPROTO_IPIP:
-			break;
-#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
-		case IPPROTO_IPV6:
-			break;
-#endif
-		default:
-			goto out;
-	}
+	if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPIP)
+		goto out;
 
 	if (!pskb_may_pull(skb, sizeof(struct iphdr)))
 		goto out;
@@ -94,20 +78,11 @@ static int xfrm4_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 	    (err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
 		goto out;
 
-	iph = ip_hdr(skb);
-	if (iph->protocol == IPPROTO_IPIP) {
-		if (x->props.flags & XFRM_STATE_DECAP_DSCP)
-			ipv4_copy_dscp(ipv4_get_dsfield(iph), ipip_hdr(skb));
-		if (!(x->props.flags & XFRM_STATE_NOECN))
-			ipip_ecn_decapsulate(skb);
-	}
-#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
-	else {
-		if (!(x->props.flags & XFRM_STATE_NOECN))
-			ipip6_ecn_decapsulate(iph, skb);
-		skb->protocol = htons(ETH_P_IPV6);
-	}
-#endif
+	if (x->props.flags & XFRM_STATE_DECAP_DSCP)
+		ipv4_copy_dscp(XFRM_MODE_SKB_CB(skb)->tos, ipip_hdr(skb));
+	if (!(x->props.flags & XFRM_STATE_NOECN))
+		ipip_ecn_decapsulate(skb);
+
 	old_mac = skb_mac_header(skb);
 	skb_set_mac_header(skb, -skb->mac_len);
 	memmove(skb_mac_header(skb), old_mac, skb->mac_len);
@@ -119,7 +94,8 @@ out:
 }
 
 static struct xfrm_mode xfrm4_tunnel_mode = {
-	.input = xfrm4_tunnel_input,
+	.input2 = xfrm4_tunnel_input,
+	.input = xfrm_prepare_input,
 	.output2 = xfrm4_tunnel_output,
 	.output = xfrm4_prepare_output,
 	.owner = THIS_MODULE,
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index e6030e74ff65..85f04b7b237f 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -65,10 +65,12 @@ int xfrm4_extract_header(struct sk_buff *skb)
 static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.proto			= IPPROTO_IPIP,
+	.eth_proto		= htons(ETH_P_IP),
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
 	.output			= xfrm4_output,
+	.extract_input		= xfrm4_extract_input,
 	.extract_output		= xfrm4_extract_output,
 };
 
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 515783707e86..c458d0a2e684 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -16,6 +16,11 @@
 #include <net/ipv6.h>
 #include <net/xfrm.h>
 
+int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	return xfrm6_extract_header(skb);
+}
+
 int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 {
 	int err;
@@ -68,7 +73,7 @@ int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 
 		xfrm_vec[xfrm_nr++] = x;
 
-		if (x->outer_mode->input(x, skb))
+		if (x->inner_mode->input(x, skb))
 			goto drop;
 
 		if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
diff --git a/net/ipv6/xfrm6_mode_beet.c b/net/ipv6/xfrm6_mode_beet.c
index 4988ed9c76c6..0527d11c1ae3 100644
--- a/net/ipv6/xfrm6_mode_beet.c
+++ b/net/ipv6/xfrm6_mode_beet.c
@@ -19,6 +19,20 @@
 #include <net/ipv6.h>
 #include <net/xfrm.h>
 
+static void xfrm6_beet_make_header(struct sk_buff *skb)
+{
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+
+	iph->version = 6;
+
+	memcpy(iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl,
+	       sizeof(iph->flow_lbl));
+	iph->nexthdr = XFRM_MODE_SKB_CB(skb)->protocol;
+
+	ipv6_change_dsfield(iph, 0, XFRM_MODE_SKB_CB(skb)->tos);
+	iph->hop_limit = XFRM_MODE_SKB_CB(skb)->ttl;
+}
+
 /* Add encapsulation header.
  *
  * The top IP header will be constructed per draft-nikander-esp-beet-mode-06.txt.
@@ -31,16 +45,11 @@ static int xfrm6_beet_output(struct xfrm_state *x, struct sk_buff *skb)
 	skb->mac_header = skb->network_header +
 			  offsetof(struct ipv6hdr, nexthdr);
 	skb->transport_header = skb->network_header + sizeof(*top_iph);
-	top_iph = ipv6_hdr(skb);
 
-	top_iph->version = 6;
+	xfrm6_beet_make_header(skb);
 
-	memcpy(top_iph->flow_lbl, XFRM_MODE_SKB_CB(skb)->flow_lbl,
-	       sizeof(top_iph->flow_lbl));
-	top_iph->nexthdr = XFRM_MODE_SKB_CB(skb)->protocol;
+	top_iph = ipv6_hdr(skb);
 
-	ipv6_change_dsfield(top_iph, 0, XFRM_MODE_SKB_CB(skb)->tos);
-	top_iph->hop_limit = XFRM_MODE_SKB_CB(skb)->ttl;
 	ipv6_addr_copy(&top_iph->saddr, (struct in6_addr *)&x->props.saddr);
 	ipv6_addr_copy(&top_iph->daddr, (struct in6_addr *)&x->id.daddr);
 	return 0;
@@ -51,19 +60,21 @@ static int xfrm6_beet_input(struct xfrm_state *x, struct sk_buff *skb)
 	struct ipv6hdr *ip6h;
 	const unsigned char *old_mac;
 	int size = sizeof(struct ipv6hdr);
-	int err = -EINVAL;
+	int err;
 
-	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
+	err = skb_cow_head(skb, size + skb->mac_len);
+	if (err)
 		goto out;
 
-	skb_push(skb, size);
-	memmove(skb->data, skb_network_header(skb), size);
+	__skb_push(skb, size);
 	skb_reset_network_header(skb);
 
 	old_mac = skb_mac_header(skb);
 	skb_set_mac_header(skb, -skb->mac_len);
 	memmove(skb_mac_header(skb), old_mac, skb->mac_len);
 
+	xfrm6_beet_make_header(skb);
+
 	ip6h = ipv6_hdr(skb);
 	ip6h->payload_len = htons(skb->len - size);
 	ipv6_addr_copy(&ip6h->daddr, (struct in6_addr *) &x->sel.daddr.a6);
@@ -74,7 +85,8 @@ out:
 }
 
 static struct xfrm_mode xfrm6_beet_mode = {
-	.input = xfrm6_beet_input,
+	.input2 = xfrm6_beet_input,
+	.input = xfrm_prepare_input,
 	.output2 = xfrm6_beet_output,
 	.output = xfrm6_prepare_output,
 	.owner = THIS_MODULE,
diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c
index d45ce5d44197..f7d0d6612650 100644
--- a/net/ipv6/xfrm6_mode_tunnel.c
+++ b/net/ipv6/xfrm6_mode_tunnel.c
@@ -25,12 +25,6 @@ static inline void ipip6_ecn_decapsulate(struct sk_buff *skb)
 		IP6_ECN_set_ce(inner_iph);
 }
 
-static inline void ip6ip_ecn_decapsulate(struct sk_buff *skb)
-{
-	if (INET_ECN_is_ce(ipv6_get_dsfield(ipv6_hdr(skb))))
-			IP_ECN_set_ce(ipip_hdr(skb));
-}
-
 /* Add encapsulation header.
  *
  * The top IP header will be constructed per RFC 2401.
@@ -68,10 +62,8 @@ static int xfrm6_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 {
 	int err = -EINVAL;
 	const unsigned char *old_mac;
-	const unsigned char *nh = skb_network_header(skb);
 
-	if (nh[IP6CB(skb)->nhoff] != IPPROTO_IPV6 &&
-	    nh[IP6CB(skb)->nhoff] != IPPROTO_IPIP)
+	if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPV6)
 		goto out;
 	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
 		goto out;
@@ -80,18 +72,12 @@ static int xfrm6_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 	    (err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
 		goto out;
 
-	nh = skb_network_header(skb);
-	if (nh[IP6CB(skb)->nhoff] == IPPROTO_IPV6) {
-		if (x->props.flags & XFRM_STATE_DECAP_DSCP)
-			ipv6_copy_dscp(ipv6_get_dsfield(ipv6_hdr(skb)),
-				       ipipv6_hdr(skb));
-		if (!(x->props.flags & XFRM_STATE_NOECN))
-			ipip6_ecn_decapsulate(skb);
-	} else {
-		if (!(x->props.flags & XFRM_STATE_NOECN))
-			ip6ip_ecn_decapsulate(skb);
-		skb->protocol = htons(ETH_P_IP);
-	}
+	if (x->props.flags & XFRM_STATE_DECAP_DSCP)
+		ipv6_copy_dscp(ipv6_get_dsfield(ipv6_hdr(skb)),
+			       ipipv6_hdr(skb));
+	if (!(x->props.flags & XFRM_STATE_NOECN))
+		ipip6_ecn_decapsulate(skb);
+
 	old_mac = skb_mac_header(skb);
 	skb_set_mac_header(skb, -skb->mac_len);
 	memmove(skb_mac_header(skb), old_mac, skb->mac_len);
@@ -103,7 +89,8 @@ out:
 }
 
 static struct xfrm_mode xfrm6_tunnel_mode = {
-	.input = xfrm6_tunnel_input,
+	.input2 = xfrm6_tunnel_input,
+	.input = xfrm_prepare_input,
 	.output2 = xfrm6_tunnel_output,
 	.output = xfrm6_prepare_output,
 	.owner = THIS_MODULE,
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index bc2e80e3b0b1..c45050cfe72b 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -53,6 +53,7 @@ int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 	if (err)
 		return err;
 
+	IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr);
 	return xfrm6_extract_header(skb);
 }
 
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 98b05f472322..90fef0a4726f 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -177,7 +177,8 @@ int xfrm6_extract_header(struct sk_buff *skb)
 	XFRM_MODE_SKB_CB(skb)->frag_off = htons(IP_DF);
 	XFRM_MODE_SKB_CB(skb)->tos = ipv6_get_dsfield(iph);
 	XFRM_MODE_SKB_CB(skb)->ttl = iph->hop_limit;
-	XFRM_MODE_SKB_CB(skb)->protocol = iph->nexthdr;
+	XFRM_MODE_SKB_CB(skb)->protocol =
+		skb_network_header(skb)[IP6CB(skb)->nhoff];
 	memcpy(XFRM_MODE_SKB_CB(skb)->flow_lbl, iph->flow_lbl,
 	       sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl));
 
@@ -187,11 +188,13 @@ int xfrm6_extract_header(struct sk_buff *skb)
 static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.proto			= IPPROTO_IPV6,
+	.eth_proto		= htons(ETH_P_IPV6),
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
 	.state_sort		= __xfrm6_state_sort,
 	.output			= xfrm6_output,
+	.extract_input		= xfrm6_extract_input,
 	.extract_output		= xfrm6_extract_output,
 };
 
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index cb97fda1b6df..4c803f7e74e5 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -81,6 +81,19 @@ int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, __be32 *spi, __be32 *seq)
 }
 EXPORT_SYMBOL(xfrm_parse_spi);
 
+int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	err = x->outer_mode->afinfo->extract_input(x, skb);
+	if (err)
+		return err;
+
+	skb->protocol = x->inner_mode->afinfo->eth_proto;
+	return x->inner_mode->input2(x, skb);
+}
+EXPORT_SYMBOL(xfrm_prepare_input);
+
 void __init xfrm_input_init(void)
 {
 	secpath_cachep = kmem_cache_create("secpath_cache",
-- 
cgit v1.2.3


From c439cb2e4b13cf1cb2abcd006b906315a3381323 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 11 Jan 2008 19:14:00 -0800
Subject: [IPV4]: Add ip_local_out

Most callers of the LOCAL_OUT chain will set the IP packet length and
header checksum before doing so.  They also share the same output
function dst_output.

This patch creates a new function called ip_local_out which does all
of that and converts the appropriate users over to it.

Apart from removing duplicate code, it will also help in merging the
IPsec output path once the same thing is done for IPv6.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h                |  2 ++
 include/net/ipip.h              |  5 ++---
 net/ipv4/igmp.c                 | 12 ++----------
 net/ipv4/ip_output.c            | 39 +++++++++++++++++++++++++--------------
 net/ipv4/ipvs/ip_vs_xmit.c      |  6 ++----
 net/ipv4/netfilter/ipt_REJECT.c |  8 +-------
 net/ipv4/xfrm4_output.c         |  8 +-------
 7 files changed, 35 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 50c8889b1b8d..66d51616ade8 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -102,6 +102,8 @@ extern int		ip_mc_output(struct sk_buff *skb);
 extern int		ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *));
 extern int		ip_do_nat(struct sk_buff *skb);
 extern void		ip_send_check(struct iphdr *ip);
+extern int		__ip_local_out(struct sk_buff *skb);
+extern int		ip_local_out(struct sk_buff *skb);
 extern int		ip_queue_xmit(struct sk_buff *skb, int ipfragok);
 extern void		ip_init(void);
 extern int		ip_append_data(struct sock *sk,
diff --git a/include/net/ipip.h b/include/net/ipip.h
index 7cdc914322f0..549e132bca9c 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -2,6 +2,7 @@
 #define __NET_IPIP_H 1
 
 #include <linux/if_tunnel.h>
+#include <net/ip.h>
 
 /* Keep error state on tunnel for 30 sec */
 #define IPTUNNEL_ERR_TIMEO	(30*HZ)
@@ -30,11 +31,9 @@ struct ip_tunnel
 	int pkt_len = skb->len;						\
 									\
 	skb->ip_summed = CHECKSUM_NONE;					\
-	iph->tot_len = htons(skb->len);					\
 	ip_select_ident(iph, &rt->u.dst, NULL);				\
-	ip_send_check(iph);						\
 									\
-	err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev, dst_output);\
+	err = ip_local_out(skb);					\
 	if (net_xmit_eval(err) == 0) {					\
 		stats->tx_bytes += pkt_len;				\
 		stats->tx_packets++;					\
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 701558564e96..c560a9392b1d 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -349,17 +349,12 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size)
 
 static int igmpv3_sendpack(struct sk_buff *skb)
 {
-	struct iphdr *pip = ip_hdr(skb);
 	struct igmphdr *pig = igmp_hdr(skb);
-	const int iplen = skb->tail - skb->network_header;
 	const int igmplen = skb->tail - skb->transport_header;
 
-	pip->tot_len = htons(iplen);
-	ip_send_check(pip);
 	pig->csum = ip_compute_csum(igmp_hdr(skb), igmplen);
 
-	return NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, skb->dev,
-		       dst_output);
+	return ip_local_out(skb);
 }
 
 static int grec_size(struct ip_mc_list *pmc, int type, int gdel, int sdel)
@@ -680,13 +675,11 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
 	iph->daddr    = dst;
 	iph->saddr    = rt->rt_src;
 	iph->protocol = IPPROTO_IGMP;
-	iph->tot_len  = htons(IGMP_SIZE);
 	ip_select_ident(iph, &rt->u.dst, NULL);
 	((u8*)&iph[1])[0] = IPOPT_RA;
 	((u8*)&iph[1])[1] = 4;
 	((u8*)&iph[1])[2] = 0;
 	((u8*)&iph[1])[3] = 0;
-	ip_send_check(iph);
 
 	ih = (struct igmphdr *)skb_put(skb, sizeof(struct igmphdr));
 	ih->type=type;
@@ -695,8 +688,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
 	ih->group=group;
 	ih->csum=ip_compute_csum((void *)ih, sizeof(struct igmphdr));
 
-	return NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
-		       dst_output);
+	return ip_local_out(skb);
 }
 
 static void igmp_gq_timer_expire(unsigned long data)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index bc9e57550e86..03b9b0600276 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -91,6 +91,28 @@ __inline__ void ip_send_check(struct iphdr *iph)
 	iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
 }
 
+int __ip_local_out(struct sk_buff *skb)
+{
+	struct iphdr *iph = ip_hdr(skb);
+
+	iph->tot_len = htons(skb->len);
+	ip_send_check(iph);
+	return nf_hook(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, skb->dst->dev,
+		       dst_output);
+}
+
+int ip_local_out(struct sk_buff *skb)
+{
+	int err;
+
+	err = __ip_local_out(skb);
+	if (likely(err == 1))
+		err = dst_output(skb);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ip_local_out);
+
 /* dev_loopback_xmit for use with netfilter. */
 static int ip_dev_loopback_xmit(struct sk_buff *newskb)
 {
@@ -138,20 +160,17 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
 	iph->daddr    = rt->rt_dst;
 	iph->saddr    = rt->rt_src;
 	iph->protocol = sk->sk_protocol;
-	iph->tot_len  = htons(skb->len);
 	ip_select_ident(iph, &rt->u.dst, sk);
 
 	if (opt && opt->optlen) {
 		iph->ihl += opt->optlen>>2;
 		ip_options_build(skb, opt, daddr, rt, 0);
 	}
-	ip_send_check(iph);
 
 	skb->priority = sk->sk_priority;
 
 	/* Send it out. */
-	return NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
-		       dst_output);
+	return ip_local_out(skb);
 }
 
 EXPORT_SYMBOL_GPL(ip_build_and_send_pkt);
@@ -347,7 +366,6 @@ packet_routed:
 	skb_reset_network_header(skb);
 	iph = ip_hdr(skb);
 	*((__be16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff));
-	iph->tot_len = htons(skb->len);
 	if (ip_dont_fragment(sk, &rt->u.dst) && !ipfragok)
 		iph->frag_off = htons(IP_DF);
 	else
@@ -366,13 +384,9 @@ packet_routed:
 	ip_select_ident_more(iph, &rt->u.dst, sk,
 			     (skb_shinfo(skb)->gso_segs ?: 1) - 1);
 
-	/* Add an IP checksum. */
-	ip_send_check(iph);
-
 	skb->priority = sk->sk_priority;
 
-	return NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
-		       dst_output);
+	return ip_local_out(skb);
 
 no_route:
 	IP_INC_STATS(IPSTATS_MIB_OUTNOROUTES);
@@ -1262,14 +1276,12 @@ int ip_push_pending_frames(struct sock *sk)
 		ip_options_build(skb, opt, inet->cork.addr, rt, 0);
 	}
 	iph->tos = inet->tos;
-	iph->tot_len = htons(skb->len);
 	iph->frag_off = df;
 	ip_select_ident(iph, &rt->u.dst, sk);
 	iph->ttl = ttl;
 	iph->protocol = sk->sk_protocol;
 	iph->saddr = rt->rt_src;
 	iph->daddr = rt->rt_dst;
-	ip_send_check(iph);
 
 	skb->priority = sk->sk_priority;
 	skb->dst = dst_clone(&rt->u.dst);
@@ -1279,8 +1291,7 @@ int ip_push_pending_frames(struct sock *sk)
 			skb_transport_header(skb))->type);
 
 	/* Netfilter gets whole the not fragmented skb. */
-	err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL,
-		      skb->dst->dev, dst_output);
+	err = ip_local_out(skb);
 	if (err) {
 		if (err > 0)
 			err = inet->recverr ? net_xmit_errno(err) : 0;
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index 7c074e386c17..66775ad9e328 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -16,8 +16,8 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/ip.h>
 #include <linux/tcp.h>                  /* for tcphdr */
+#include <net/ip.h>
 #include <net/tcp.h>                    /* for csum_tcpudp_magic */
 #include <net/udp.h>
 #include <net/icmp.h>                   /* for icmp_send */
@@ -406,14 +406,12 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	iph->daddr		=	rt->rt_dst;
 	iph->saddr		=	rt->rt_src;
 	iph->ttl		=	old_iph->ttl;
-	iph->tot_len		=	htons(skb->len);
 	ip_select_ident(iph, &rt->u.dst, NULL);
-	ip_send_check(iph);
 
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(skb, rt);
+	ip_local_out(skb);
 
 	LeaveFunction(10);
 
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index dcf4d21d5116..ccb2a03dcd5a 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -90,7 +90,6 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	/* Truncate to length (no data) */
 	tcph->doff = sizeof(struct tcphdr)/4;
 	skb_trim(nskb, ip_hdrlen(nskb) + sizeof(struct tcphdr));
-	niph->tot_len = htons(nskb->len);
 
 	if (tcph->ack) {
 		needs_ack = 0;
@@ -139,18 +138,13 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	/* Adjust IP TTL */
 	niph->ttl = dst_metric(nskb->dst, RTAX_HOPLIMIT);
 
-	/* Adjust IP checksum */
-	niph->check = 0;
-	niph->check = ip_fast_csum(skb_network_header(nskb), niph->ihl);
-
 	/* "Never happens" */
 	if (nskb->len > dst_mtu(nskb->dst))
 		goto free_nskb;
 
 	nf_ct_attach(nskb, oldskb);
 
-	NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, nskb, NULL, nskb->dst->dev,
-		dst_output);
+	ip_local_out(nskb);
 	return;
 
  free_nskb:
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 13fd11335e28..0ffc3d078489 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -69,17 +69,12 @@ EXPORT_SYMBOL(xfrm4_prepare_output);
 
 static inline int xfrm4_output_one(struct sk_buff *skb)
 {
-	struct iphdr *iph;
 	int err;
 
 	err = xfrm_output(skb);
 	if (err)
 		goto error_nolock;
 
-	iph = ip_hdr(skb);
-	iph->tot_len = htons(skb->len);
-	ip_send_check(iph);
-
 	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
 	err = 0;
 
@@ -97,8 +92,7 @@ static int xfrm4_output_finish2(struct sk_buff *skb)
 	while (likely((err = xfrm4_output_one(skb)) == 0)) {
 		nf_reset(skb);
 
-		err = nf_hook(PF_INET, NF_IP_LOCAL_OUT, skb, NULL,
-			      skb->dst->dev, dst_output);
+		err = __ip_local_out(skb);
 		if (unlikely(err != 1))
 			break;
 
-- 
cgit v1.2.3


From ef76bc23ef2acf20c8f7f841a542d8ab74c827c6 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 11 Jan 2008 19:15:08 -0800
Subject: [IPV6]: Add ip6_local_out

Most callers of the LOCAL_OUT chain will set the IP packet length
before doing so.  They also share the same output function dst_output.

This patch creates a new function called ip6_local_out which does all
of that and converts the appropriate users over to it.

Apart from removing duplicate code, it will also help in merging the
IPsec output path.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h               |  3 +++
 net/ipv6/ip6_output.c            | 33 +++++++++++++++++++++++++++------
 net/ipv6/ip6_tunnel.c            |  4 +---
 net/ipv6/netfilter/ip6t_REJECT.c |  4 +---
 net/ipv6/xfrm6_output.c          |  7 +------
 5 files changed, 33 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index ae328b680ff2..e90f9625cb1b 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -509,6 +509,9 @@ extern int			ip6_forward(struct sk_buff *skb);
 extern int			ip6_input(struct sk_buff *skb);
 extern int			ip6_mc_input(struct sk_buff *skb);
 
+extern int			__ip6_local_out(struct sk_buff *skb);
+extern int			ip6_local_out(struct sk_buff *skb);
+
 /*
  *	Extension header (options) processing
  */
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 387030d2483c..bd121f9ae0a7 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -29,7 +29,7 @@
  */
 
 #include <linux/errno.h>
-#include <linux/types.h>
+#include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/socket.h>
 #include <linux/net.h>
@@ -70,6 +70,31 @@ static __inline__ void ipv6_select_ident(struct sk_buff *skb, struct frag_hdr *f
 	spin_unlock_bh(&ip6_id_lock);
 }
 
+int __ip6_local_out(struct sk_buff *skb)
+{
+	int len;
+
+	len = skb->len - sizeof(struct ipv6hdr);
+	if (len > IPV6_MAXPLEN)
+		len = 0;
+	ipv6_hdr(skb)->payload_len = htons(len);
+
+	return nf_hook(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev,
+		       dst_output);
+}
+
+int ip6_local_out(struct sk_buff *skb)
+{
+	int err;
+
+	err = __ip6_local_out(skb);
+	if (likely(err == 1))
+		err = dst_output(skb);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ip6_local_out);
+
 static int ip6_output_finish(struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
@@ -1403,10 +1428,6 @@ int ip6_push_pending_frames(struct sock *sk)
 	*(__be32*)hdr = fl->fl6_flowlabel |
 		     htonl(0x60000000 | ((int)np->cork.tclass << 20));
 
-	if (skb->len <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN)
-		hdr->payload_len = htons(skb->len - sizeof(struct ipv6hdr));
-	else
-		hdr->payload_len = 0;
 	hdr->hop_limit = np->cork.hop_limit;
 	hdr->nexthdr = proto;
 	ipv6_addr_copy(&hdr->saddr, &fl->fl6_src);
@@ -1423,7 +1444,7 @@ int ip6_push_pending_frames(struct sock *sk)
 		ICMP6_INC_STATS_BH(idev, ICMP6_MIB_OUTMSGS);
 	}
 
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output);
+	err = ip6_local_out(skb);
 	if (err) {
 		if (err > 0)
 			err = np->recverr ? net_xmit_errno(err) : 0;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index a4051afaf776..29b5321e39c3 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -910,15 +910,13 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 	*(__be32*)ipv6h = fl->fl6_flowlabel | htonl(0x60000000);
 	dsfield = INET_ECN_encapsulate(0, dsfield);
 	ipv6_change_dsfield(ipv6h, ~INET_ECN_MASK, dsfield);
-	ipv6h->payload_len = htons(skb->len - sizeof(struct ipv6hdr));
 	ipv6h->hop_limit = t->parms.hop_limit;
 	ipv6h->nexthdr = proto;
 	ipv6_addr_copy(&ipv6h->saddr, &fl->fl6_src);
 	ipv6_addr_copy(&ipv6h->daddr, &fl->fl6_dst);
 	nf_reset(skb);
 	pkt_len = skb->len;
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL,
-		      skb->dst->dev, dst_output);
+	err = ip6_local_out(skb);
 
 	if (net_xmit_eval(err) == 0) {
 		stats->tx_bytes += pkt_len;
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index 1a7d2917545d..c1c663482837 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -121,7 +121,6 @@ static void send_reset(struct sk_buff *oldskb)
 	ip6h->version = 6;
 	ip6h->hop_limit = dst_metric(dst, RTAX_HOPLIMIT);
 	ip6h->nexthdr = IPPROTO_TCP;
-	ip6h->payload_len = htons(sizeof(struct tcphdr));
 	ipv6_addr_copy(&ip6h->saddr, &oip6h->daddr);
 	ipv6_addr_copy(&ip6h->daddr, &oip6h->saddr);
 
@@ -159,8 +158,7 @@ static void send_reset(struct sk_buff *oldskb)
 
 	nf_ct_attach(nskb, oldskb);
 
-	NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, nskb, NULL, nskb->dst->dev,
-		dst_output);
+	ip6_local_out(nskb);
 }
 
 static inline void
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index c45050cfe72b..0f0ff51f6dba 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -75,16 +75,12 @@ EXPORT_SYMBOL(xfrm6_prepare_output);
 
 static inline int xfrm6_output_one(struct sk_buff *skb)
 {
-	struct ipv6hdr *iph;
 	int err;
 
 	err = xfrm_output(skb);
 	if (err)
 		goto error_nolock;
 
-	iph = ipv6_hdr(skb);
-	iph->payload_len = htons(skb->len - sizeof(*iph));
-
 	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
 	err = 0;
 
@@ -102,8 +98,7 @@ static int xfrm6_output_finish2(struct sk_buff *skb)
 	while (likely((err = xfrm6_output_one(skb)) == 0)) {
 		nf_reset(skb);
 
-		err = nf_hook(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL,
-			      skb->dst->dev, dst_output);
+		err = __ip6_local_out(skb);
 		if (unlikely(err != 1))
 			break;
 
-- 
cgit v1.2.3


From 862b82c6f960cc61274d370aa78ce1112f92a83e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:43:11 -0800
Subject: [IPSEC]: Merge most of the output path

As part of the work on asynchrnous cryptographic operations, we need
to be able to resume from the spot where they occur.  As such, it
helps if we isolate them to one spot.

This patch moves most of the remaining family-specific processing into
the common output code.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h       |  1 +
 include/net/xfrm.h      |  1 +
 net/ipv4/route.c        |  1 +
 net/ipv4/xfrm4_output.c | 76 +++---------------------------------------------
 net/ipv4/xfrm4_policy.c |  1 +
 net/ipv4/xfrm4_state.c  |  2 ++
 net/ipv6/route.c        |  1 +
 net/ipv6/xfrm6_output.c | 77 +++++--------------------------------------------
 net/ipv6/xfrm6_policy.c |  1 +
 net/ipv6/xfrm6_state.c  |  2 ++
 net/xfrm/xfrm_output.c  | 70 ++++++++++++++++++++++++++++++++++++++++++--
 11 files changed, 88 insertions(+), 145 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 7a0b1bde8e28..e86b9a008ebf 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -98,6 +98,7 @@ struct dst_ops
 	struct dst_entry *	(*negative_advice)(struct dst_entry *);
 	void			(*link_failure)(struct sk_buff *);
 	void			(*update_pmtu)(struct dst_entry *dst, u32 mtu);
+	int			(*local_out)(struct sk_buff *skb);
 	int			entry_size;
 
 	atomic_t		entries;
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index a9dbe091ae58..ab9e747340b4 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -259,6 +259,7 @@ struct xfrm_state_afinfo {
 	unsigned int		family;
 	unsigned int		proto;
 	unsigned int		eth_proto;
+	unsigned int		nf_post_routing;
 	struct module		*owner;
 	struct xfrm_type	*type_map[IPPROTO_MAX];
 	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 137b8eb666b7..94ef788a2ac6 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -166,6 +166,7 @@ static struct dst_ops ipv4_dst_ops = {
 	.negative_advice =	ipv4_negative_advice,
 	.link_failure =		ipv4_link_failure,
 	.update_pmtu =		ip_rt_update_pmtu,
+	.local_out =		ip_local_out,
 	.entry_size =		sizeof(struct rtable),
 };
 
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 0ffc3d078489..2fb4efa3ff2c 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -59,7 +59,7 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 		return err;
 
 	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
-	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
+	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED;
 
 	skb->protocol = htons(ETH_P_IP);
 
@@ -67,87 +67,19 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(xfrm4_prepare_output);
 
-static inline int xfrm4_output_one(struct sk_buff *skb)
-{
-	int err;
-
-	err = xfrm_output(skb);
-	if (err)
-		goto error_nolock;
-
-	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
-	err = 0;
-
-out_exit:
-	return err;
-error_nolock:
-	kfree_skb(skb);
-	goto out_exit;
-}
-
-static int xfrm4_output_finish2(struct sk_buff *skb)
-{
-	int err;
-
-	while (likely((err = xfrm4_output_one(skb)) == 0)) {
-		nf_reset(skb);
-
-		err = __ip_local_out(skb);
-		if (unlikely(err != 1))
-			break;
-
-		if (!skb->dst->xfrm)
-			return dst_output(skb);
-
-		err = nf_hook(PF_INET, NF_IP_POST_ROUTING, skb, NULL,
-			      skb->dst->dev, xfrm4_output_finish2);
-		if (unlikely(err != 1))
-			break;
-	}
-
-	return err;
-}
-
 static int xfrm4_output_finish(struct sk_buff *skb)
 {
-	struct sk_buff *segs;
-
 #ifdef CONFIG_NETFILTER
 	if (!skb->dst->xfrm) {
 		IPCB(skb)->flags |= IPSKB_REROUTED;
 		return dst_output(skb);
 	}
-#endif
 
-	if (!skb_is_gso(skb))
-		return xfrm4_output_finish2(skb);
+	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IP);
-	segs = skb_gso_segment(skb, 0);
-	kfree_skb(skb);
-	if (unlikely(IS_ERR(segs)))
-		return PTR_ERR(segs);
-
-	do {
-		struct sk_buff *nskb = segs->next;
-		int err;
-
-		segs->next = NULL;
-		err = xfrm4_output_finish2(segs);
-
-		if (unlikely(err)) {
-			while ((segs = nskb)) {
-				nskb = segs->next;
-				segs->next = NULL;
-				kfree_skb(segs);
-			}
-			return err;
-		}
-
-		segs = nskb;
-	} while (segs);
-
-	return 0;
+	return xfrm_output(skb);
 }
 
 int xfrm4_output(struct sk_buff *skb)
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 1d7524375b49..b4948c170b3e 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -237,6 +237,7 @@ static struct dst_ops xfrm4_dst_ops = {
 	.update_pmtu =		xfrm4_update_pmtu,
 	.destroy =		xfrm4_dst_destroy,
 	.ifdown =		xfrm4_dst_ifdown,
+	.local_out =		__ip_local_out,
 	.gc_thresh =		1024,
 	.entry_size =		sizeof(struct xfrm_dst),
 };
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 85f04b7b237f..80292fbf221a 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -11,6 +11,7 @@
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
+#include <linux/netfilter_ipv4.h>
 
 static struct xfrm_state_afinfo xfrm4_state_afinfo;
 
@@ -66,6 +67,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.proto			= IPPROTO_IPIP,
 	.eth_proto		= htons(ETH_P_IP),
+	.nf_post_routing	= NF_IP_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ac70e2d3b10c..4ef2cfaa3467 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -113,6 +113,7 @@ static struct dst_ops ip6_dst_ops = {
 	.negative_advice	=	ip6_negative_advice,
 	.link_failure		=	ip6_link_failure,
 	.update_pmtu		=	ip6_rt_update_pmtu,
+	.local_out		=	ip6_local_out,
 	.entry_size		=	sizeof(struct rt6_info),
 };
 
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 0f0ff51f6dba..a0a924991c4f 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -66,6 +66,9 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 		return err;
 
 	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
+#ifdef CONFIG_NETFILTER
+	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IPV6);
 
@@ -73,80 +76,14 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(xfrm6_prepare_output);
 
-static inline int xfrm6_output_one(struct sk_buff *skb)
-{
-	int err;
-
-	err = xfrm_output(skb);
-	if (err)
-		goto error_nolock;
-
-	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
-	err = 0;
-
-out_exit:
-	return err;
-error_nolock:
-	kfree_skb(skb);
-	goto out_exit;
-}
-
-static int xfrm6_output_finish2(struct sk_buff *skb)
-{
-	int err;
-
-	while (likely((err = xfrm6_output_one(skb)) == 0)) {
-		nf_reset(skb);
-
-		err = __ip6_local_out(skb);
-		if (unlikely(err != 1))
-			break;
-
-		if (!skb->dst->xfrm)
-			return dst_output(skb);
-
-		err = nf_hook(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL,
-			      skb->dst->dev, xfrm6_output_finish2);
-		if (unlikely(err != 1))
-			break;
-	}
-
-	return err;
-}
-
 static int xfrm6_output_finish(struct sk_buff *skb)
 {
-	struct sk_buff *segs;
-
-	if (!skb_is_gso(skb))
-		return xfrm6_output_finish2(skb);
+#ifdef CONFIG_NETFILTER
+	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
+#endif
 
 	skb->protocol = htons(ETH_P_IPV6);
-	segs = skb_gso_segment(skb, 0);
-	kfree_skb(skb);
-	if (unlikely(IS_ERR(segs)))
-		return PTR_ERR(segs);
-
-	do {
-		struct sk_buff *nskb = segs->next;
-		int err;
-
-		segs->next = NULL;
-		err = xfrm6_output_finish2(segs);
-
-		if (unlikely(err)) {
-			while ((segs = nskb)) {
-				nskb = segs->next;
-				segs->next = NULL;
-				kfree_skb(segs);
-			}
-			return err;
-		}
-
-		segs = nskb;
-	} while (segs);
-
-	return 0;
+	return xfrm_output(skb);
 }
 
 int xfrm6_output(struct sk_buff *skb)
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 63932c5fd3c7..a31dd531e191 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -252,6 +252,7 @@ static struct dst_ops xfrm6_dst_ops = {
 	.update_pmtu =		xfrm6_update_pmtu,
 	.destroy =		xfrm6_dst_destroy,
 	.ifdown =		xfrm6_dst_ifdown,
+	.local_out =		__ip6_local_out,
 	.gc_thresh =		1024,
 	.entry_size =		sizeof(struct xfrm_dst),
 };
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 90fef0a4726f..bb09e85a336d 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -14,6 +14,7 @@
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
+#include <linux/netfilter_ipv6.h>
 #include <net/dsfield.h>
 #include <net/ipv6.h>
 #include <net/addrconf.h>
@@ -189,6 +190,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.proto			= IPPROTO_IPV6,
 	.eth_proto		= htons(ETH_P_IPV6),
+	.nf_post_routing	= NF_IP6_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index b1efdc8850a7..bcb3701c5cf3 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -12,6 +12,7 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/netdevice.h>
+#include <linux/netfilter.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
 #include <net/dst.h>
@@ -40,7 +41,7 @@ err:
 	return err;
 }
 
-int xfrm_output(struct sk_buff *skb)
+static int xfrm_output_one(struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
 	struct xfrm_state *x = dst->xfrm;
@@ -87,10 +88,73 @@ int xfrm_output(struct sk_buff *skb)
 
 	err = 0;
 
-error_nolock:
+out_exit:
 	return err;
 error:
 	spin_unlock_bh(&x->lock);
-	goto error_nolock;
+error_nolock:
+	kfree_skb(skb);
+	goto out_exit;
+}
+
+static int xfrm_output2(struct sk_buff *skb)
+{
+	int err;
+
+	while (likely((err = xfrm_output_one(skb)) == 0)) {
+		struct xfrm_state *x;
+
+		nf_reset(skb);
+
+		err = skb->dst->ops->local_out(skb);
+		if (unlikely(err != 1))
+			break;
+
+		x = skb->dst->xfrm;
+		if (!x)
+			return dst_output(skb);
+
+		err = nf_hook(x->inner_mode->afinfo->family,
+			      x->inner_mode->afinfo->nf_post_routing, skb,
+			      NULL, skb->dst->dev, xfrm_output2);
+		if (unlikely(err != 1))
+			break;
+	}
+
+	return err;
+}
+
+int xfrm_output(struct sk_buff *skb)
+{
+	struct sk_buff *segs;
+
+	if (!skb_is_gso(skb))
+		return xfrm_output2(skb);
+
+	segs = skb_gso_segment(skb, 0);
+	kfree_skb(skb);
+	if (unlikely(IS_ERR(segs)))
+		return PTR_ERR(segs);
+
+	do {
+		struct sk_buff *nskb = segs->next;
+		int err;
+
+		segs->next = NULL;
+		err = xfrm_output2(segs);
+
+		if (unlikely(err)) {
+			while ((segs = nskb)) {
+				nskb = segs->next;
+				segs->next = NULL;
+				kfree_skb(segs);
+			}
+			return err;
+		}
+
+		segs = nskb;
+	} while (segs);
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(xfrm_output);
-- 
cgit v1.2.3


From c6581a457e661b7070e484ad723bbf555b17aca2 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:43:43 -0800
Subject: [IPSEC]: Add async resume support on output

This patch adds support for async resumptions on output.  To do so,
the transform would return -EINPROGRESS and subsequently invoke the
function xfrm_output_resume to resume processing.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  1 +
 net/xfrm/xfrm_output.c | 57 +++++++++++++++++++++++++++++++++++---------------
 2 files changed, 41 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index ab9e747340b4..99677207a4ce 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1119,6 +1119,7 @@ extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
 extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb);
+extern int xfrm_output_resume(struct sk_buff *skb, int err);
 extern int xfrm_output(struct sk_buff *skb);
 extern int xfrm4_extract_header(struct sk_buff *skb);
 extern int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb);
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index bcb3701c5cf3..048d240c3e15 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -18,6 +18,8 @@
 #include <net/dst.h>
 #include <net/xfrm.h>
 
+static int xfrm_output2(struct sk_buff *skb);
+
 static int xfrm_state_check_space(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
@@ -41,17 +43,13 @@ err:
 	return err;
 }
 
-static int xfrm_output_one(struct sk_buff *skb)
+static int xfrm_output_one(struct sk_buff *skb, int err)
 {
 	struct dst_entry *dst = skb->dst;
 	struct xfrm_state *x = dst->xfrm;
-	int err;
 
-	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		err = skb_checksum_help(skb);
-		if (err)
-			goto error_nolock;
-	}
+	if (err <= 0)
+		goto resume;
 
 	do {
 		err = x->outer_mode->output(x, skb);
@@ -75,6 +73,8 @@ static int xfrm_output_one(struct sk_buff *skb)
 		spin_unlock_bh(&x->lock);
 
 		err = x->type->output(x, skb);
+
+resume:
 		if (err)
 			goto error_nolock;
 
@@ -97,18 +97,16 @@ error_nolock:
 	goto out_exit;
 }
 
-static int xfrm_output2(struct sk_buff *skb)
+int xfrm_output_resume(struct sk_buff *skb, int err)
 {
-	int err;
-
-	while (likely((err = xfrm_output_one(skb)) == 0)) {
+	while (likely((err = xfrm_output_one(skb, err)) == 0)) {
 		struct xfrm_state *x;
 
 		nf_reset(skb);
 
 		err = skb->dst->ops->local_out(skb);
 		if (unlikely(err != 1))
-			break;
+			goto out;
 
 		x = skb->dst->xfrm;
 		if (!x)
@@ -118,18 +116,25 @@ static int xfrm_output2(struct sk_buff *skb)
 			      x->inner_mode->afinfo->nf_post_routing, skb,
 			      NULL, skb->dst->dev, xfrm_output2);
 		if (unlikely(err != 1))
-			break;
+			goto out;
 	}
 
+	if (err == -EINPROGRESS)
+		err = 0;
+
+out:
 	return err;
 }
+EXPORT_SYMBOL_GPL(xfrm_output_resume);
 
-int xfrm_output(struct sk_buff *skb)
+static int xfrm_output2(struct sk_buff *skb)
 {
-	struct sk_buff *segs;
+	return xfrm_output_resume(skb, 1);
+}
 
-	if (!skb_is_gso(skb))
-		return xfrm_output2(skb);
+static int xfrm_output_gso(struct sk_buff *skb)
+{
+	struct sk_buff *segs;
 
 	segs = skb_gso_segment(skb, 0);
 	kfree_skb(skb);
@@ -157,4 +162,22 @@ int xfrm_output(struct sk_buff *skb)
 
 	return 0;
 }
+
+int xfrm_output(struct sk_buff *skb)
+{
+	int err;
+
+	if (skb_is_gso(skb))
+		return xfrm_output_gso(skb);
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		err = skb_checksum_help(skb);
+		if (err) {
+			kfree_skb(skb);
+			return err;
+		}
+	}
+
+	return xfrm_output2(skb);
+}
 EXPORT_SYMBOL_GPL(xfrm_output);
-- 
cgit v1.2.3


From 716062fd4c2f88a33ab409f62a1e7397ad0a7e33 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:44:23 -0800
Subject: [IPSEC]: Merge most of the input path

As part of the work on asynchronous cryptographic operations, we need
to be able to resume from the spot where they occur.  As such, it
helps if we isolate them to one spot.

This patch moves most of the remaining family-specific processing into
the common input code.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  22 +++++++++
 net/ipv4/xfrm4_input.c | 126 ++++++-------------------------------------------
 net/ipv4/xfrm4_state.c |   1 +
 net/ipv6/xfrm6_input.c | 118 +++++----------------------------------------
 net/ipv6/xfrm6_state.c |   1 +
 net/xfrm/xfrm_input.c  | 113 ++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 164 insertions(+), 217 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 99677207a4ce..329de412c0ba 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -274,6 +274,8 @@ struct xfrm_state_afinfo {
 						 struct sk_buff *skb);
 	int			(*extract_output)(struct xfrm_state *x,
 						  struct sk_buff *skb);
+	int			(*transport_finish)(struct sk_buff *skb,
+						    int async);
 };
 
 extern int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo);
@@ -522,6 +524,22 @@ struct xfrm_mode_skb_cb {
 
 #define XFRM_MODE_SKB_CB(__skb) ((struct xfrm_mode_skb_cb *)&((__skb)->cb[0]))
 
+/*
+ * This structure is used by the input processing to locate the SPI and
+ * related information.
+ */
+struct xfrm_spi_skb_cb {
+	union {
+		struct inet_skb_parm h4;
+		struct inet6_skb_parm h6;
+	} header;
+
+	unsigned int nhoff;
+	unsigned int daddroff;
+};
+
+#define XFRM_SPI_SKB_CB(__skb) ((struct xfrm_spi_skb_cb *)&((__skb)->cb[0]))
+
 /* Audit Information */
 struct xfrm_audit
 {
@@ -1119,12 +1137,15 @@ extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
 extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb);
+extern int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi,
+		      int encap_type);
 extern int xfrm_output_resume(struct sk_buff *skb, int err);
 extern int xfrm_output(struct sk_buff *skb);
 extern int xfrm4_extract_header(struct sk_buff *skb);
 extern int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 			   int encap_type);
+extern int xfrm4_transport_finish(struct sk_buff *skb, int async);
 extern int xfrm4_rcv(struct sk_buff *skb);
 
 static inline int xfrm4_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
@@ -1140,6 +1161,7 @@ extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler, unsigned short f
 extern int xfrm6_extract_header(struct sk_buff *skb);
 extern int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi);
+extern int xfrm6_transport_finish(struct sk_buff *skb, int async);
 extern int xfrm6_rcv(struct sk_buff *skb);
 extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
 			    xfrm_address_t *saddr, u8 proto);
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index c0323d05ab69..e374903dacdf 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -41,124 +41,26 @@ drop:
 int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 		    int encap_type)
 {
-	int err;
-	__be32 seq;
-	struct xfrm_state *xfrm_vec[XFRM_MAX_DEPTH];
-	struct xfrm_state *x;
-	int xfrm_nr = 0;
-	int decaps = 0;
-	unsigned int nhoff = offsetof(struct iphdr, protocol);
-
-	seq = 0;
-	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0)
-		goto drop;
-
-	do {
-		const struct iphdr *iph = ip_hdr(skb);
-
-		if (xfrm_nr == XFRM_MAX_DEPTH)
-			goto drop;
-
-		x = xfrm_state_lookup((xfrm_address_t *)&iph->daddr, spi,
-				      nexthdr, AF_INET);
-		if (x == NULL)
-			goto drop;
-
-		spin_lock(&x->lock);
-		if (unlikely(x->km.state != XFRM_STATE_VALID))
-			goto drop_unlock;
-
-		if ((x->encap ? x->encap->encap_type : 0) != encap_type)
-			goto drop_unlock;
-
-		if (x->props.replay_window && xfrm_replay_check(x, seq))
-			goto drop_unlock;
-
-		if (xfrm_state_check_expire(x))
-			goto drop_unlock;
-
-		nexthdr = x->type->input(x, skb);
-		if (nexthdr <= 0)
-			goto drop_unlock;
-
-		skb_network_header(skb)[nhoff] = nexthdr;
-
-		/* only the first xfrm gets the encap type */
-		encap_type = 0;
-
-		if (x->props.replay_window)
-			xfrm_replay_advance(x, seq);
-
-		x->curlft.bytes += skb->len;
-		x->curlft.packets++;
-
-		spin_unlock(&x->lock);
-
-		xfrm_vec[xfrm_nr++] = x;
-
-		if (x->inner_mode->input(x, skb))
-			goto drop;
-
-		if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
-			decaps = 1;
-			break;
-		}
-
-		err = xfrm_parse_spi(skb, nexthdr, &spi, &seq);
-		if (err < 0)
-			goto drop;
-	} while (!err);
-
-	/* Allocate new secpath or COW existing one. */
-
-	if (!skb->sp || atomic_read(&skb->sp->refcnt) != 1) {
-		struct sec_path *sp;
-		sp = secpath_dup(skb->sp);
-		if (!sp)
-			goto drop;
-		if (skb->sp)
-			secpath_put(skb->sp);
-		skb->sp = sp;
-	}
-	if (xfrm_nr + skb->sp->len > XFRM_MAX_DEPTH)
-		goto drop;
-
-	memcpy(skb->sp->xvec + skb->sp->len, xfrm_vec,
-	       xfrm_nr * sizeof(xfrm_vec[0]));
-	skb->sp->len += xfrm_nr;
-
-	nf_reset(skb);
+	XFRM_SPI_SKB_CB(skb)->nhoff = offsetof(struct iphdr, protocol);
+	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
+	return xfrm_input(skb, nexthdr, spi, encap_type);
+}
+EXPORT_SYMBOL(xfrm4_rcv_encap);
 
-	if (decaps) {
-		dst_release(skb->dst);
-		skb->dst = NULL;
-		netif_rx(skb);
-		return 0;
-	} else {
+int xfrm4_transport_finish(struct sk_buff *skb, int async)
+{
 #ifdef CONFIG_NETFILTER
-		__skb_push(skb, skb->data - skb_network_header(skb));
-		ip_hdr(skb)->tot_len = htons(skb->len);
-		ip_send_check(ip_hdr(skb));
+	__skb_push(skb, skb->data - skb_network_header(skb));
+	ip_hdr(skb)->tot_len = htons(skb->len);
+	ip_send_check(ip_hdr(skb));
 
-		NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
-			xfrm4_rcv_encap_finish);
-		return 0;
+	NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
+		xfrm4_rcv_encap_finish);
+	return 0;
 #else
-		return -ip_hdr(skb)->protocol;
+	return -ip_hdr(skb)->protocol;
 #endif
-	}
-
-drop_unlock:
-	spin_unlock(&x->lock);
-	xfrm_state_put(x);
-drop:
-	while (--xfrm_nr >= 0)
-		xfrm_state_put(xfrm_vec[xfrm_nr]);
-
-	kfree_skb(skb);
-	return 0;
 }
-EXPORT_SYMBOL(xfrm4_rcv_encap);
 
 /* If it's a keepalive packet, then just eat it.
  * If it's an encapsulated packet, then pass it to the
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 80292fbf221a..3b067e8b7bfe 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -74,6 +74,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.output			= xfrm4_output,
 	.extract_input		= xfrm4_extract_input,
 	.extract_output		= xfrm4_extract_output,
+	.transport_finish	= xfrm4_transport_finish,
 };
 
 void __init xfrm4_state_init(void)
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index c458d0a2e684..3b9eedf5b24a 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -23,118 +23,26 @@ int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb)
 
 int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 {
-	int err;
-	__be32 seq;
-	struct xfrm_state *xfrm_vec[XFRM_MAX_DEPTH];
-	struct xfrm_state *x;
-	int xfrm_nr = 0;
-	int decaps = 0;
-	unsigned int nhoff;
-
-	nhoff = IP6CB(skb)->nhoff;
-
-	seq = 0;
-	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0)
-		goto drop;
-
-	do {
-		struct ipv6hdr *iph = ipv6_hdr(skb);
-
-		if (xfrm_nr == XFRM_MAX_DEPTH)
-			goto drop;
-
-		x = xfrm_state_lookup((xfrm_address_t *)&iph->daddr, spi,
-				      nexthdr, AF_INET6);
-		if (x == NULL)
-			goto drop;
-		spin_lock(&x->lock);
-		if (unlikely(x->km.state != XFRM_STATE_VALID))
-			goto drop_unlock;
-
-		if (x->props.replay_window && xfrm_replay_check(x, seq))
-			goto drop_unlock;
-
-		if (xfrm_state_check_expire(x))
-			goto drop_unlock;
-
-		nexthdr = x->type->input(x, skb);
-		if (nexthdr <= 0)
-			goto drop_unlock;
-
-		skb_network_header(skb)[nhoff] = nexthdr;
-
-		if (x->props.replay_window)
-			xfrm_replay_advance(x, seq);
-
-		x->curlft.bytes += skb->len;
-		x->curlft.packets++;
-
-		spin_unlock(&x->lock);
-
-		xfrm_vec[xfrm_nr++] = x;
-
-		if (x->inner_mode->input(x, skb))
-			goto drop;
-
-		if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
-			decaps = 1;
-			break;
-		}
-
-		if ((err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) < 0)
-			goto drop;
-	} while (!err);
-
-	/* Allocate new secpath or COW existing one. */
-	if (!skb->sp || atomic_read(&skb->sp->refcnt) != 1) {
-		struct sec_path *sp;
-		sp = secpath_dup(skb->sp);
-		if (!sp)
-			goto drop;
-		if (skb->sp)
-			secpath_put(skb->sp);
-		skb->sp = sp;
-	}
-
-	if (xfrm_nr + skb->sp->len > XFRM_MAX_DEPTH)
-		goto drop;
-
-	memcpy(skb->sp->xvec + skb->sp->len, xfrm_vec,
-	       xfrm_nr * sizeof(xfrm_vec[0]));
-	skb->sp->len += xfrm_nr;
-
-	nf_reset(skb);
+	XFRM_SPI_SKB_CB(skb)->nhoff = IP6CB(skb)->nhoff;
+	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
+	return xfrm_input(skb, nexthdr, spi, 0);
+}
+EXPORT_SYMBOL(xfrm6_rcv_spi);
 
-	if (decaps) {
-		dst_release(skb->dst);
-		skb->dst = NULL;
-		netif_rx(skb);
-		return -1;
-	} else {
+int xfrm6_transport_finish(struct sk_buff *skb, int async)
+{
 #ifdef CONFIG_NETFILTER
-		ipv6_hdr(skb)->payload_len = htons(skb->len);
-		__skb_push(skb, skb->data - skb_network_header(skb));
+	ipv6_hdr(skb)->payload_len = htons(skb->len);
+	__skb_push(skb, skb->data - skb_network_header(skb));
 
-		NF_HOOK(PF_INET6, NF_IP6_PRE_ROUTING, skb, skb->dev, NULL,
-			ip6_rcv_finish);
-		return -1;
+	NF_HOOK(PF_INET6, NF_IP6_PRE_ROUTING, skb, skb->dev, NULL,
+		ip6_rcv_finish);
+	return -1;
 #else
-		return 1;
+	return 1;
 #endif
-	}
-
-drop_unlock:
-	spin_unlock(&x->lock);
-	xfrm_state_put(x);
-drop:
-	while (--xfrm_nr >= 0)
-		xfrm_state_put(xfrm_vec[xfrm_nr]);
-	kfree_skb(skb);
-	return -1;
 }
 
-EXPORT_SYMBOL(xfrm6_rcv_spi);
-
 int xfrm6_rcv(struct sk_buff *skb)
 {
 	return xfrm6_rcv_spi(skb, skb_network_header(skb)[IP6CB(skb)->nhoff],
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index bb09e85a336d..00360b514e99 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -198,6 +198,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.output			= xfrm6_output,
 	.extract_input		= xfrm6_extract_input,
 	.extract_output		= xfrm6_extract_output,
+	.transport_finish	= xfrm6_transport_finish,
 };
 
 void __init xfrm6_state_init(void)
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 4c803f7e74e5..b980095be935 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -9,6 +9,8 @@
 
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/netdevice.h>
+#include <net/dst.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
 
@@ -94,6 +96,117 @@ int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(xfrm_prepare_input);
 
+int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
+{
+	int err;
+	__be32 seq;
+	struct xfrm_state *xfrm_vec[XFRM_MAX_DEPTH];
+	struct xfrm_state *x;
+	int xfrm_nr = 0;
+	int decaps = 0;
+	unsigned int nhoff = XFRM_SPI_SKB_CB(skb)->nhoff;
+	unsigned int daddroff = XFRM_SPI_SKB_CB(skb)->daddroff;
+
+	seq = 0;
+	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0)
+		goto drop;
+
+	do {
+		if (xfrm_nr == XFRM_MAX_DEPTH)
+			goto drop;
+
+		x = xfrm_state_lookup((xfrm_address_t *)
+				      (skb_network_header(skb) + daddroff),
+				      spi, nexthdr, AF_INET);
+		if (x == NULL)
+			goto drop;
+
+		spin_lock(&x->lock);
+		if (unlikely(x->km.state != XFRM_STATE_VALID))
+			goto drop_unlock;
+
+		if ((x->encap ? x->encap->encap_type : 0) != encap_type)
+			goto drop_unlock;
+
+		if (x->props.replay_window && xfrm_replay_check(x, seq))
+			goto drop_unlock;
+
+		if (xfrm_state_check_expire(x))
+			goto drop_unlock;
+
+		nexthdr = x->type->input(x, skb);
+		if (nexthdr <= 0)
+			goto drop_unlock;
+
+		skb_network_header(skb)[nhoff] = nexthdr;
+
+		/* only the first xfrm gets the encap type */
+		encap_type = 0;
+
+		if (x->props.replay_window)
+			xfrm_replay_advance(x, seq);
+
+		x->curlft.bytes += skb->len;
+		x->curlft.packets++;
+
+		spin_unlock(&x->lock);
+
+		xfrm_vec[xfrm_nr++] = x;
+
+		if (x->inner_mode->input(x, skb))
+			goto drop;
+
+		if (x->outer_mode->flags & XFRM_MODE_FLAG_TUNNEL) {
+			decaps = 1;
+			break;
+		}
+
+		err = xfrm_parse_spi(skb, nexthdr, &spi, &seq);
+		if (err < 0)
+			goto drop;
+	} while (!err);
+
+	/* Allocate new secpath or COW existing one. */
+
+	if (!skb->sp || atomic_read(&skb->sp->refcnt) != 1) {
+		struct sec_path *sp;
+		sp = secpath_dup(skb->sp);
+		if (!sp)
+			goto drop;
+		if (skb->sp)
+			secpath_put(skb->sp);
+		skb->sp = sp;
+	}
+	if (xfrm_nr + skb->sp->len > XFRM_MAX_DEPTH)
+		goto drop;
+
+	memcpy(skb->sp->xvec + skb->sp->len, xfrm_vec,
+	       xfrm_nr * sizeof(xfrm_vec[0]));
+	skb->sp->len += xfrm_nr;
+
+	nf_reset(skb);
+
+	if (decaps) {
+		dst_release(skb->dst);
+		skb->dst = NULL;
+		netif_rx(skb);
+		return 0;
+	} else {
+		return x->inner_mode->afinfo->transport_finish(skb, 0);
+	}
+
+drop_unlock:
+	spin_unlock(&x->lock);
+	xfrm_state_put(x);
+drop:
+	while (--xfrm_nr >= 0)
+		xfrm_state_put(xfrm_vec[xfrm_nr]);
+
+	kfree_skb(skb);
+	return 0;
+}
+EXPORT_SYMBOL(xfrm_input);
+
 void __init xfrm_input_init(void)
 {
 	secpath_cachep = kmem_cache_create("secpath_cache",
-- 
cgit v1.2.3


From d26f398400311982d2433debae85746c348b7d58 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Nov 2007 21:47:08 -0800
Subject: [IPSEC]: Make x->lastused an unsigned long

Currently x->lastused is u64 which means that it cannot be
read/written atomically on all architectures.  David Miller observed
that the value stored in it is only an unsigned long which is always
atomic.

So based on his suggestion this patch changes the internal
representation from u64 to unsigned long while the user-interface
still refers to it as u64.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h       | 2 +-
 net/ipv6/xfrm6_mode_ro.c | 2 --
 net/xfrm/xfrm_user.c     | 4 ++--
 3 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 329de412c0ba..311bbd113aa7 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -183,7 +183,7 @@ struct xfrm_state
 	struct timer_list	timer;
 
 	/* Last used time */
-	u64			lastused;
+	unsigned long		lastused;
 
 	/* Reference to data common to all the instances of this
 	 * transformer. */
diff --git a/net/ipv6/xfrm6_mode_ro.c b/net/ipv6/xfrm6_mode_ro.c
index 4a01cb3c370b..63d5d493098a 100644
--- a/net/ipv6/xfrm6_mode_ro.c
+++ b/net/ipv6/xfrm6_mode_ro.c
@@ -54,9 +54,7 @@ static int xfrm6_ro_output(struct xfrm_state *x, struct sk_buff *skb)
 	__skb_pull(skb, hdr_len);
 	memmove(ipv6_hdr(skb), iph, hdr_len);
 
-	spin_lock_bh(&x->lock);
 	x->lastused = get_seconds();
-	spin_unlock_bh(&x->lock);
 
 	return 0;
 }
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index c4f6419b1769..6424e5360511 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -1986,8 +1986,8 @@ static inline size_t xfrm_sa_len(struct xfrm_state *x)
 	if (x->coaddr)
 		l += nla_total_size(sizeof(*x->coaddr));
 
-	/* Must count this as this may become non-zero behind our back. */
-	l += nla_total_size(sizeof(x->lastused));
+	/* Must count x->lastused as it may become non-zero behind our back. */
+	l += nla_total_size(sizeof(u64));
 
 	return l;
 }
-- 
cgit v1.2.3


From 60d5fcfb19d8a958fc563e52240cd05ec23f36c9 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 19 Nov 2007 18:47:58 -0800
Subject: [IPSEC]: Remove nhoff from xfrm_input

The nhoff field isn't actually necessary in xfrm_input.  For tunnel
mode transforms we now throw away the output IP header so it makes no
sense to fill in the nexthdr field.  For transport mode we can now let
the function transport_finish do the setting and it knows where the
nexthdr field is.

The only other thing that needs the nexthdr field to be set is the
header extraction code.  However, we can simply move the protocol
extraction out of the generic header extraction.

We want to minimise the amount of info we have to carry around between
transforms as this simplifies the resumption process for async crypto.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  1 -
 net/ipv4/xfrm4_input.c  | 13 +++++++------
 net/ipv4/xfrm4_output.c |  2 ++
 net/ipv4/xfrm4_state.c  |  1 -
 net/ipv6/xfrm6_input.c  |  4 +++-
 net/ipv6/xfrm6_output.c |  3 ++-
 net/ipv6/xfrm6_state.c  |  2 --
 net/xfrm/xfrm_input.c   |  5 ++---
 8 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 311bbd113aa7..cf85dc9dc420 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -534,7 +534,6 @@ struct xfrm_spi_skb_cb {
 		struct inet6_skb_parm h6;
 	} header;
 
-	unsigned int nhoff;
 	unsigned int daddroff;
 };
 
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index e374903dacdf..662d1e86cfbf 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -21,7 +21,6 @@ int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb)
 	return xfrm4_extract_header(skb);
 }
 
-#ifdef CONFIG_NETFILTER
 static inline int xfrm4_rcv_encap_finish(struct sk_buff *skb)
 {
 	if (skb->dst == NULL) {
@@ -36,12 +35,10 @@ drop:
 	kfree_skb(skb);
 	return NET_RX_DROP;
 }
-#endif
 
 int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 		    int encap_type)
 {
-	XFRM_SPI_SKB_CB(skb)->nhoff = offsetof(struct iphdr, protocol);
 	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
 	return xfrm_input(skb, nexthdr, spi, encap_type);
 }
@@ -49,16 +46,20 @@ EXPORT_SYMBOL(xfrm4_rcv_encap);
 
 int xfrm4_transport_finish(struct sk_buff *skb, int async)
 {
+	struct iphdr *iph = ip_hdr(skb);
+
+	iph->protocol = XFRM_MODE_SKB_CB(skb)->protocol;
+
 #ifdef CONFIG_NETFILTER
 	__skb_push(skb, skb->data - skb_network_header(skb));
-	ip_hdr(skb)->tot_len = htons(skb->len);
-	ip_send_check(ip_hdr(skb));
+	iph->tot_len = htons(skb->len);
+	ip_send_check(iph);
 
 	NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
 		xfrm4_rcv_encap_finish);
 	return 0;
 #else
-	return -ip_hdr(skb)->protocol;
+	return -iph->protocol;
 #endif
 }
 
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 2fb4efa3ff2c..1900200d3c0f 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -47,6 +47,8 @@ int xfrm4_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 	if (err)
 		return err;
 
+	XFRM_MODE_SKB_CB(skb)->protocol = ip_hdr(skb)->protocol;
+
 	return xfrm4_extract_header(skb);
 }
 
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 3b067e8b7bfe..d837784a2199 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -56,7 +56,6 @@ int xfrm4_extract_header(struct sk_buff *skb)
 	XFRM_MODE_SKB_CB(skb)->frag_off = iph->frag_off;
 	XFRM_MODE_SKB_CB(skb)->tos = iph->tos;
 	XFRM_MODE_SKB_CB(skb)->ttl = iph->ttl;
-	XFRM_MODE_SKB_CB(skb)->protocol = iph->protocol;
 	memset(XFRM_MODE_SKB_CB(skb)->flow_lbl, 0,
 	       sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl));
 
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 3b9eedf5b24a..5c006c845943 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -23,7 +23,6 @@ int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb)
 
 int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 {
-	XFRM_SPI_SKB_CB(skb)->nhoff = IP6CB(skb)->nhoff;
 	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
 	return xfrm_input(skb, nexthdr, spi, 0);
 }
@@ -31,6 +30,9 @@ EXPORT_SYMBOL(xfrm6_rcv_spi);
 
 int xfrm6_transport_finish(struct sk_buff *skb, int async)
 {
+	skb_network_header(skb)[IP6CB(skb)->nhoff] =
+		XFRM_MODE_SKB_CB(skb)->protocol;
+
 #ifdef CONFIG_NETFILTER
 	ipv6_hdr(skb)->payload_len = htons(skb->len);
 	__skb_push(skb, skb->data - skb_network_header(skb));
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index a0a924991c4f..318669a9cb48 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -53,7 +53,8 @@ int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 	if (err)
 		return err;
 
-	IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr);
+	XFRM_MODE_SKB_CB(skb)->protocol = ipv6_hdr(skb)->nexthdr;
+
 	return xfrm6_extract_header(skb);
 }
 
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 00360b514e99..df7e98d914fa 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -178,8 +178,6 @@ int xfrm6_extract_header(struct sk_buff *skb)
 	XFRM_MODE_SKB_CB(skb)->frag_off = htons(IP_DF);
 	XFRM_MODE_SKB_CB(skb)->tos = ipv6_get_dsfield(iph);
 	XFRM_MODE_SKB_CB(skb)->ttl = iph->hop_limit;
-	XFRM_MODE_SKB_CB(skb)->protocol =
-		skb_network_header(skb)[IP6CB(skb)->nhoff];
 	memcpy(XFRM_MODE_SKB_CB(skb)->flow_lbl, iph->flow_lbl,
 	       sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl));
 
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 5cad522e8ef6..cce9d4586045 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -102,7 +102,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	__be32 seq;
 	struct xfrm_state *x;
 	int decaps = 0;
-	unsigned int nhoff = XFRM_SPI_SKB_CB(skb)->nhoff;
 	unsigned int daddroff = XFRM_SPI_SKB_CB(skb)->daddroff;
 
 	/* Allocate new secpath or COW existing one. */
@@ -157,8 +156,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			goto drop_unlock;
 		}
 
-		skb_network_header(skb)[nhoff] = nexthdr;
-
 		/* only the first xfrm gets the encap type */
 		encap_type = 0;
 
@@ -170,6 +167,8 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 
 		spin_unlock(&x->lock);
 
+		XFRM_MODE_SKB_CB(skb)->protocol = nexthdr;
+
 		if (x->inner_mode->input(x, skb))
 			goto drop;
 
-- 
cgit v1.2.3


From 1bf06cd2e338fd6fc29169d30eaf0df982338285 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 19 Nov 2007 18:50:17 -0800
Subject: [IPSEC]: Add async resume support on input

This patch adds support for async resumptions on input.  To do so, the
transform would return -EINPROGRESS and subsequently invoke the
function xfrm_input_resume to resume processing.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  1 +
 net/ipv4/xfrm4_input.c |  3 +++
 net/ipv6/xfrm6_input.c |  3 +++
 net/xfrm/xfrm_input.c  | 38 +++++++++++++++++++++++++++++++++-----
 4 files changed, 40 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index cf85dc9dc420..37a086d71baf 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1138,6 +1138,7 @@ extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi,
 		      int encap_type);
+extern int xfrm_input_resume(struct sk_buff *skb, int nexthdr);
 extern int xfrm_output_resume(struct sk_buff *skb, int err);
 extern int xfrm_output(struct sk_buff *skb);
 extern int xfrm4_extract_header(struct sk_buff *skb);
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index 662d1e86cfbf..d5890c84a492 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -59,6 +59,9 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async)
 		xfrm4_rcv_encap_finish);
 	return 0;
 #else
+	if (async)
+		return xfrm4_rcv_encap_finish(skb);
+
 	return -iph->protocol;
 #endif
 }
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 5c006c845943..e317d0855468 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -41,6 +41,9 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async)
 		ip6_rcv_finish);
 	return -1;
 #else
+	if (async)
+		return ip6_rcv_finish(skb);
+
 	return 1;
 #endif
 }
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index cce9d4586045..96f42c1d2e8e 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -101,8 +101,17 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	int err;
 	__be32 seq;
 	struct xfrm_state *x;
+	xfrm_address_t *daddr;
 	int decaps = 0;
-	unsigned int daddroff = XFRM_SPI_SKB_CB(skb)->daddroff;
+	int async = 0;
+
+	/* A negative encap_type indicates async resumption. */
+	if (encap_type < 0) {
+		async = 1;
+		x = skb->sp->xvec[skb->sp->len - 1];
+		seq = XFRM_SKB_CB(skb)->seq;
+		goto resume;
+	}
 
 	/* Allocate new secpath or COW existing one. */
 	if (!skb->sp || atomic_read(&skb->sp->refcnt) != 1) {
@@ -116,6 +125,9 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		skb->sp = sp;
 	}
 
+	daddr = (xfrm_address_t *)(skb_network_header(skb) +
+				   XFRM_SPI_SKB_CB(skb)->daddroff);
+
 	seq = 0;
 	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0)
 		goto drop;
@@ -124,9 +136,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		if (skb->sp->len == XFRM_MAX_DEPTH)
 			goto drop;
 
-		x = xfrm_state_lookup((xfrm_address_t *)
-				      (skb_network_header(skb) + daddroff),
-				      spi, nexthdr, AF_INET);
+		x = xfrm_state_lookup(daddr, spi, nexthdr, AF_INET);
 		if (x == NULL)
 			goto drop;
 
@@ -147,8 +157,14 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 
 		spin_unlock(&x->lock);
 
+		XFRM_SKB_CB(skb)->seq = seq;
+
 		nexthdr = x->type->input(x, skb);
 
+		if (nexthdr == -EINPROGRESS)
+			return 0;
+
+resume:
 		spin_lock(&x->lock);
 		if (nexthdr <= 0) {
 			if (nexthdr == -EBADMSG)
@@ -177,6 +193,12 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			break;
 		}
 
+		/*
+		 * We need the inner address.  However, we only get here for
+		 * transport mode so the outer address is identical.
+		 */
+		daddr = &x->id.daddr;
+
 		err = xfrm_parse_spi(skb, nexthdr, &spi, &seq);
 		if (err < 0)
 			goto drop;
@@ -190,7 +212,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		netif_rx(skb);
 		return 0;
 	} else {
-		return x->inner_mode->afinfo->transport_finish(skb, 0);
+		return x->inner_mode->afinfo->transport_finish(skb, async);
 	}
 
 drop_unlock:
@@ -201,6 +223,12 @@ drop:
 }
 EXPORT_SYMBOL(xfrm_input);
 
+int xfrm_input_resume(struct sk_buff *skb, int nexthdr)
+{
+	return xfrm_input(skb, nexthdr, 0, -1);
+}
+EXPORT_SYMBOL(xfrm_input_resume);
+
 void __init xfrm_input_init(void)
 {
 	secpath_cachep = kmem_cache_create("secpath_cache",
-- 
cgit v1.2.3


From 6e23ae2a48750bda407a4a58f52a4865d7308bf5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 19 Nov 2007 18:53:30 -0800
Subject: [NETFILTER]: Introduce NF_INET_ hook values

The IPv4 and IPv6 hook values are identical, yet some code tries to figure
out the "correct" value by looking at the address family. Introduce NF_INET_*
values for both IPv4 and IPv6. The old values are kept in a #ifndef __KERNEL__
section for userspace compatibility.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                      |  9 ++++++
 include/linux/netfilter/x_tables.h             |  4 +--
 include/linux/netfilter_ipv4.h                 |  2 +-
 include/linux/netfilter_ipv4/ip_tables.h       |  8 ++---
 include/linux/netfilter_ipv6.h                 |  3 +-
 include/linux/netfilter_ipv6/ip6_tables.h      |  8 ++---
 include/net/netfilter/nf_nat.h                 |  3 +-
 net/bridge/br_netfilter.c                      | 12 +++----
 net/compat.c                                   |  6 ++--
 net/ipv4/ip_forward.c                          |  2 +-
 net/ipv4/ip_input.c                            |  4 +--
 net/ipv4/ip_output.c                           | 12 +++----
 net/ipv4/ipmr.c                                |  2 +-
 net/ipv4/ipvs/ip_vs_core.c                     | 18 +++++------
 net/ipv4/ipvs/ip_vs_xmit.c                     |  2 +-
 net/ipv4/netfilter.c                           |  8 ++---
 net/ipv4/netfilter/ip_tables.c                 | 44 +++++++++++++-------------
 net/ipv4/netfilter/ipt_MASQUERADE.c            |  4 +--
 net/ipv4/netfilter/ipt_NETMAP.c                | 13 ++++----
 net/ipv4/netfilter/ipt_REDIRECT.c              |  8 ++---
 net/ipv4/netfilter/ipt_REJECT.c                |  6 ++--
 net/ipv4/netfilter/ipt_SAME.c                  |  7 ++--
 net/ipv4/netfilter/ipt_owner.c                 |  3 +-
 net/ipv4/netfilter/iptable_filter.c            | 22 +++++++------
 net/ipv4/netfilter/iptable_mangle.c            | 40 +++++++++++------------
 net/ipv4/netfilter/iptable_raw.c               | 14 ++++----
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 18 +++++------
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  2 +-
 net/ipv4/netfilter/nf_nat_core.c               | 14 ++++----
 net/ipv4/netfilter/nf_nat_h323.c               |  8 ++---
 net/ipv4/netfilter/nf_nat_helper.c             |  4 +--
 net/ipv4/netfilter/nf_nat_pptp.c               |  4 +--
 net/ipv4/netfilter/nf_nat_rule.c               | 28 ++++++++--------
 net/ipv4/netfilter/nf_nat_sip.c                |  4 +--
 net/ipv4/netfilter/nf_nat_standalone.c         | 14 ++++----
 net/ipv4/raw.c                                 |  2 +-
 net/ipv4/xfrm4_input.c                         |  2 +-
 net/ipv4/xfrm4_output.c                        |  4 +--
 net/ipv4/xfrm4_state.c                         |  2 +-
 net/ipv6/ip6_input.c                           |  6 ++--
 net/ipv6/ip6_output.c                          | 14 ++++----
 net/ipv6/mcast.c                               |  6 ++--
 net/ipv6/ndisc.c                               |  6 ++--
 net/ipv6/netfilter.c                           |  6 ++--
 net/ipv6/netfilter/ip6_tables.c                | 26 +++++++--------
 net/ipv6/netfilter/ip6t_REJECT.c               |  6 ++--
 net/ipv6/netfilter/ip6t_eui64.c                |  4 +--
 net/ipv6/netfilter/ip6t_owner.c                |  3 +-
 net/ipv6/netfilter/ip6table_filter.c           | 22 +++++++------
 net/ipv6/netfilter/ip6table_mangle.c           | 40 +++++++++++------------
 net/ipv6/netfilter/ip6table_raw.c              | 14 ++++----
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 12 +++----
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  2 +-
 net/ipv6/raw.c                                 |  2 +-
 net/ipv6/xfrm6_input.c                         |  2 +-
 net/ipv6/xfrm6_output.c                        |  2 +-
 net/ipv6/xfrm6_state.c                         |  2 +-
 net/netfilter/nf_conntrack_netlink.c           |  8 ++---
 net/netfilter/nf_conntrack_proto_tcp.c         |  4 +--
 net/netfilter/nf_conntrack_proto_udp.c         |  4 +--
 net/netfilter/nf_conntrack_proto_udplite.c     |  3 +-
 net/netfilter/xt_CLASSIFY.c                    | 12 +++----
 net/netfilter/xt_TCPMSS.c                      | 12 +++----
 net/netfilter/xt_mac.c                         | 12 +++----
 net/netfilter/xt_physdev.c                     |  6 ++--
 net/netfilter/xt_policy.c                      |  5 ++-
 net/netfilter/xt_realm.c                       |  4 +--
 net/sched/sch_ingress.c                        |  4 +--
 security/selinux/hooks.c                       |  4 +--
 69 files changed, 321 insertions(+), 302 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 16adac688af5..25fc12260340 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -39,6 +39,15 @@
 #define NFC_ALTERED 0x8000
 #endif
 
+enum nf_inet_hooks {
+	NF_INET_PRE_ROUTING,
+	NF_INET_LOCAL_IN,
+	NF_INET_FORWARD,
+	NF_INET_LOCAL_OUT,
+	NF_INET_POST_ROUTING,
+	NF_INET_NUMHOOKS
+};
+
 #ifdef __KERNEL__
 #ifdef CONFIG_NETFILTER
 
diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 03e6ce979eaa..9657c4ee70fc 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -265,8 +265,8 @@ struct xt_table_info
 	unsigned int initial_entries;
 
 	/* Entry points and underflows */
-	unsigned int hook_entry[NF_IP_NUMHOOKS];
-	unsigned int underflow[NF_IP_NUMHOOKS];
+	unsigned int hook_entry[NF_INET_NUMHOOKS];
+	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* ipt_entry tables: one per CPU */
 	char *entries[NR_CPUS];
diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index 1a63adf5c4c1..9a10092e358c 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -36,7 +36,6 @@
 #define NFC_IP_DST_PT		0x0400
 /* Something else about the proto */
 #define NFC_IP_PROTO_UNKNOWN	0x2000
-#endif /* ! __KERNEL__ */
 
 /* IP Hooks */
 /* After promisc drops, checksum checks. */
@@ -50,6 +49,7 @@
 /* Packets about to hit the wire. */
 #define NF_IP_POST_ROUTING	4
 #define NF_IP_NUMHOOKS		5
+#endif /* ! __KERNEL__ */
 
 enum nf_ip_hook_priorities {
 	NF_IP_PRI_FIRST = INT_MIN,
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index d79ed69cbc1f..54da61603eff 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -156,10 +156,10 @@ struct ipt_getinfo
 	unsigned int valid_hooks;
 
 	/* Hook entry points: one per netfilter hook. */
-	unsigned int hook_entry[NF_IP_NUMHOOKS];
+	unsigned int hook_entry[NF_INET_NUMHOOKS];
 
 	/* Underflow points. */
-	unsigned int underflow[NF_IP_NUMHOOKS];
+	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* Number of entries */
 	unsigned int num_entries;
@@ -185,10 +185,10 @@ struct ipt_replace
 	unsigned int size;
 
 	/* Hook entry points. */
-	unsigned int hook_entry[NF_IP_NUMHOOKS];
+	unsigned int hook_entry[NF_INET_NUMHOOKS];
 
 	/* Underflow points. */
-	unsigned int underflow[NF_IP_NUMHOOKS];
+	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* Information about old entries: */
 	/* Number of counters (must be equal to current number of entries). */
diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 66ca8e3100dc..3475a65dae9b 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -40,8 +40,6 @@
 #define NFC_IP6_DST_PT           0x0400
 /* Something else about the proto */
 #define NFC_IP6_PROTO_UNKNOWN    0x2000
-#endif /* ! __KERNEL__ */
-
 
 /* IP6 Hooks */
 /* After promisc drops, checksum checks. */
@@ -55,6 +53,7 @@
 /* Packets about to hit the wire. */
 #define NF_IP6_POST_ROUTING	4
 #define NF_IP6_NUMHOOKS		5
+#endif /* ! __KERNEL__ */
 
 
 enum nf_ip6_hook_priorities {
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 7dc481ce7cba..2e98654188b3 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -216,10 +216,10 @@ struct ip6t_getinfo
 	unsigned int valid_hooks;
 
 	/* Hook entry points: one per netfilter hook. */
-	unsigned int hook_entry[NF_IP6_NUMHOOKS];
+	unsigned int hook_entry[NF_INET_NUMHOOKS];
 
 	/* Underflow points. */
-	unsigned int underflow[NF_IP6_NUMHOOKS];
+	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* Number of entries */
 	unsigned int num_entries;
@@ -245,10 +245,10 @@ struct ip6t_replace
 	unsigned int size;
 
 	/* Hook entry points. */
-	unsigned int hook_entry[NF_IP6_NUMHOOKS];
+	unsigned int hook_entry[NF_INET_NUMHOOKS];
 
 	/* Underflow points. */
-	unsigned int underflow[NF_IP6_NUMHOOKS];
+	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* Information about old entries: */
 	/* Number of counters (must be equal to current number of entries). */
diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h
index 6ae52f7c9f55..76da32292bcd 100644
--- a/include/net/netfilter/nf_nat.h
+++ b/include/net/netfilter/nf_nat.h
@@ -12,7 +12,8 @@ enum nf_nat_manip_type
 };
 
 /* SRC manip occurs POST_ROUTING or LOCAL_IN */
-#define HOOK2MANIP(hooknum) ((hooknum) != NF_IP_POST_ROUTING && (hooknum) != NF_IP_LOCAL_IN)
+#define HOOK2MANIP(hooknum) ((hooknum) != NF_INET_POST_ROUTING && \
+			     (hooknum) != NF_INET_LOCAL_IN)
 
 #define IP_NAT_RANGE_MAP_IPS 1
 #define IP_NAT_RANGE_PROTO_SPECIFIED 2
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 9f78a69d6b8b..f9ef3e58b4cb 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -511,7 +511,7 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook,
 	if (!setup_pre_routing(skb))
 		return NF_DROP;
 
-	NF_HOOK(PF_INET6, NF_IP6_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(PF_INET6, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
 		br_nf_pre_routing_finish_ipv6);
 
 	return NF_STOLEN;
@@ -584,7 +584,7 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff *skb,
 		return NF_DROP;
 	store_orig_dstaddr(skb);
 
-	NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(PF_INET, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
 		br_nf_pre_routing_finish);
 
 	return NF_STOLEN;
@@ -681,7 +681,7 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff *skb,
 	nf_bridge->mask |= BRNF_BRIDGED;
 	nf_bridge->physoutdev = skb->dev;
 
-	NF_HOOK(pf, NF_IP_FORWARD, skb, bridge_parent(in), parent,
+	NF_HOOK(pf, NF_INET_FORWARD, skb, bridge_parent(in), parent,
 		br_nf_forward_finish);
 
 	return NF_STOLEN;
@@ -832,7 +832,7 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff *skb,
 	if (nf_bridge->netoutdev)
 		realoutdev = nf_bridge->netoutdev;
 #endif
-	NF_HOOK(pf, NF_IP_POST_ROUTING, skb, NULL, realoutdev,
+	NF_HOOK(pf, NF_INET_POST_ROUTING, skb, NULL, realoutdev,
 		br_nf_dev_queue_xmit);
 
 	return NF_STOLEN;
@@ -905,12 +905,12 @@ static struct nf_hook_ops br_nf_ops[] = {
 	{ .hook = ip_sabotage_in,
 	  .owner = THIS_MODULE,
 	  .pf = PF_INET,
-	  .hooknum = NF_IP_PRE_ROUTING,
+	  .hooknum = NF_INET_PRE_ROUTING,
 	  .priority = NF_IP_PRI_FIRST, },
 	{ .hook = ip_sabotage_in,
 	  .owner = THIS_MODULE,
 	  .pf = PF_INET6,
-	  .hooknum = NF_IP6_PRE_ROUTING,
+	  .hooknum = NF_INET_PRE_ROUTING,
 	  .priority = NF_IP6_PRI_FIRST, },
 };
 
diff --git a/net/compat.c b/net/compat.c
index 377e560ab5c9..f4ef4c048652 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -325,8 +325,8 @@ struct compat_ipt_replace {
 	u32			valid_hooks;
 	u32			num_entries;
 	u32			size;
-	u32			hook_entry[NF_IP_NUMHOOKS];
-	u32			underflow[NF_IP_NUMHOOKS];
+	u32			hook_entry[NF_INET_NUMHOOKS];
+	u32			underflow[NF_INET_NUMHOOKS];
 	u32			num_counters;
 	compat_uptr_t		counters;	/* struct ipt_counters * */
 	struct ipt_entry	entries[0];
@@ -391,7 +391,7 @@ static int do_netfilter_replace(int fd, int level, int optname,
 			   origsize))
 		goto out;
 
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		if (__get_user(tmp32, &urepl->hook_entry[i]) ||
 		    __put_user(tmp32, &repl_nat->hook_entry[i]) ||
 		    __get_user(tmp32, &urepl->underflow[i]) ||
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 877da3ed52e2..0b3b328d82db 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -110,7 +110,7 @@ int ip_forward(struct sk_buff *skb)
 
 	skb->priority = rt_tos2priority(iph->tos);
 
-	return NF_HOOK(PF_INET, NF_IP_FORWARD, skb, skb->dev, rt->u.dst.dev,
+	return NF_HOOK(PF_INET, NF_INET_FORWARD, skb, skb->dev, rt->u.dst.dev,
 		       ip_forward_finish);
 
 sr_failed:
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 168c871fcd79..5b8a7603e606 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -268,7 +268,7 @@ int ip_local_deliver(struct sk_buff *skb)
 			return 0;
 	}
 
-	return NF_HOOK(PF_INET, NF_IP_LOCAL_IN, skb, skb->dev, NULL,
+	return NF_HOOK(PF_INET, NF_INET_LOCAL_IN, skb, skb->dev, NULL,
 		       ip_local_deliver_finish);
 }
 
@@ -442,7 +442,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	/* Remove any debris in the socket control block */
 	memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
 
-	return NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, dev, NULL,
+	return NF_HOOK(PF_INET, NF_INET_PRE_ROUTING, skb, dev, NULL,
 		       ip_rcv_finish);
 
 inhdr_error:
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 03b9b0600276..6dd1d9c5d52e 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -97,7 +97,7 @@ int __ip_local_out(struct sk_buff *skb)
 
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
-	return nf_hook(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, skb->dst->dev,
+	return nf_hook(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, skb->dst->dev,
 		       dst_output);
 }
 
@@ -270,8 +270,8 @@ int ip_mc_output(struct sk_buff *skb)
 		) {
 			struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 			if (newskb)
-				NF_HOOK(PF_INET, NF_IP_POST_ROUTING, newskb, NULL,
-					newskb->dev,
+				NF_HOOK(PF_INET, NF_INET_POST_ROUTING, newskb,
+					NULL, newskb->dev,
 					ip_dev_loopback_xmit);
 		}
 
@@ -286,11 +286,11 @@ int ip_mc_output(struct sk_buff *skb)
 	if (rt->rt_flags&RTCF_BROADCAST) {
 		struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 		if (newskb)
-			NF_HOOK(PF_INET, NF_IP_POST_ROUTING, newskb, NULL,
+			NF_HOOK(PF_INET, NF_INET_POST_ROUTING, newskb, NULL,
 				newskb->dev, ip_dev_loopback_xmit);
 	}
 
-	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dev,
+	return NF_HOOK_COND(PF_INET, NF_INET_POST_ROUTING, skb, NULL, skb->dev,
 			    ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
@@ -304,7 +304,7 @@ int ip_output(struct sk_buff *skb)
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
 
-	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, dev,
+	return NF_HOOK_COND(PF_INET, NF_INET_POST_ROUTING, skb, NULL, dev,
 			    ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index ba6c23cdf47b..8e5d47a60602 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1245,7 +1245,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi)
 	 * not mrouter) cannot join to more than one interface - it will
 	 * result in receiving multiple packets.
 	 */
-	NF_HOOK(PF_INET, NF_IP_FORWARD, skb, skb->dev, dev,
+	NF_HOOK(PF_INET, NF_INET_FORWARD, skb, skb->dev, dev,
 		ipmr_forward_finish);
 	return;
 
diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c
index 8fba20256f52..30e8f7571529 100644
--- a/net/ipv4/ipvs/ip_vs_core.c
+++ b/net/ipv4/ipvs/ip_vs_core.c
@@ -481,7 +481,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 
 
 /*
- *      It is hooked before NF_IP_PRI_NAT_SRC at the NF_IP_POST_ROUTING
+ *      It is hooked before NF_IP_PRI_NAT_SRC at the NF_INET_POST_ROUTING
  *      chain, and is used for VS/NAT.
  *      It detects packets for VS/NAT connections and sends the packets
  *      immediately. This can avoid that iptable_nat mangles the packets
@@ -679,7 +679,7 @@ static inline int is_tcp_reset(const struct sk_buff *skb)
 }
 
 /*
- *	It is hooked at the NF_IP_FORWARD chain, used only for VS/NAT.
+ *	It is hooked at the NF_INET_FORWARD chain, used only for VS/NAT.
  *	Check if outgoing packet belongs to the established ip_vs_conn,
  *      rewrite addresses of the packet and send it on its way...
  */
@@ -814,7 +814,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 
 	/* reassemble IP fragments */
 	if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
-		if (ip_vs_gather_frags(skb, hooknum == NF_IP_LOCAL_IN ?
+		if (ip_vs_gather_frags(skb, hooknum == NF_INET_LOCAL_IN ?
 					    IP_DEFRAG_VS_IN : IP_DEFRAG_VS_FWD))
 			return NF_STOLEN;
 	}
@@ -1003,12 +1003,12 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb,
 
 
 /*
- *	It is hooked at the NF_IP_FORWARD chain, in order to catch ICMP
+ *	It is hooked at the NF_INET_FORWARD chain, in order to catch ICMP
  *      related packets destined for 0.0.0.0/0.
  *      When fwmark-based virtual service is used, such as transparent
  *      cache cluster, TCP packets can be marked and routed to ip_vs_in,
  *      but ICMP destined for 0.0.0.0/0 cannot not be easily marked and
- *      sent to ip_vs_in_icmp. So, catch them at the NF_IP_FORWARD chain
+ *      sent to ip_vs_in_icmp. So, catch them at the NF_INET_FORWARD chain
  *      and send them to ip_vs_in_icmp.
  */
 static unsigned int
@@ -1032,7 +1032,7 @@ static struct nf_hook_ops ip_vs_in_ops = {
 	.hook		= ip_vs_in,
 	.owner		= THIS_MODULE,
 	.pf		= PF_INET,
-	.hooknum        = NF_IP_LOCAL_IN,
+	.hooknum        = NF_INET_LOCAL_IN,
 	.priority       = 100,
 };
 
@@ -1041,7 +1041,7 @@ static struct nf_hook_ops ip_vs_out_ops = {
 	.hook		= ip_vs_out,
 	.owner		= THIS_MODULE,
 	.pf		= PF_INET,
-	.hooknum        = NF_IP_FORWARD,
+	.hooknum        = NF_INET_FORWARD,
 	.priority       = 100,
 };
 
@@ -1051,7 +1051,7 @@ static struct nf_hook_ops ip_vs_forward_icmp_ops = {
 	.hook		= ip_vs_forward_icmp,
 	.owner		= THIS_MODULE,
 	.pf		= PF_INET,
-	.hooknum        = NF_IP_FORWARD,
+	.hooknum        = NF_INET_FORWARD,
 	.priority       = 99,
 };
 
@@ -1060,7 +1060,7 @@ static struct nf_hook_ops ip_vs_post_routing_ops = {
 	.hook		= ip_vs_post_routing,
 	.owner		= THIS_MODULE,
 	.pf		= PF_INET,
-	.hooknum        = NF_IP_POST_ROUTING,
+	.hooknum        = NF_INET_POST_ROUTING,
 	.priority       = NF_IP_PRI_NAT_SRC-1,
 };
 
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index 66775ad9e328..1e96bf82a0b5 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -129,7 +129,7 @@ ip_vs_dst_reset(struct ip_vs_dest *dest)
 do {							\
 	(skb)->ipvs_property = 1;			\
 	skb_forward_csum(skb);				\
-	NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, (skb), NULL,	\
+	NF_HOOK(PF_INET, NF_INET_LOCAL_OUT, (skb), NULL,	\
 		(rt)->u.dst.dev, dst_output);		\
 } while (0)
 
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 5539debf4973..d9022467e089 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -23,7 +23,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
 		addr_type = type;
 
 	/* some non-standard hacks like ipt_REJECT.c:send_reset() can cause
-	 * packets with foreign saddr to appear on the NF_IP_LOCAL_OUT hook.
+	 * packets with foreign saddr to appear on the NF_INET_LOCAL_OUT hook.
 	 */
 	if (addr_type == RTN_LOCAL) {
 		fl.nl_u.ip4_u.daddr = iph->daddr;
@@ -126,7 +126,7 @@ static void nf_ip_saveroute(const struct sk_buff *skb, struct nf_info *info)
 {
 	struct ip_rt_info *rt_info = nf_info_reroute(info);
 
-	if (info->hook == NF_IP_LOCAL_OUT) {
+	if (info->hook == NF_INET_LOCAL_OUT) {
 		const struct iphdr *iph = ip_hdr(skb);
 
 		rt_info->tos = iph->tos;
@@ -139,7 +139,7 @@ static int nf_ip_reroute(struct sk_buff *skb, const struct nf_info *info)
 {
 	const struct ip_rt_info *rt_info = nf_info_reroute(info);
 
-	if (info->hook == NF_IP_LOCAL_OUT) {
+	if (info->hook == NF_INET_LOCAL_OUT) {
 		const struct iphdr *iph = ip_hdr(skb);
 
 		if (!(iph->tos == rt_info->tos
@@ -158,7 +158,7 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 
 	switch (skb->ip_summed) {
 	case CHECKSUM_COMPLETE:
-		if (hook != NF_IP_PRE_ROUTING && hook != NF_IP_LOCAL_IN)
+		if (hook != NF_INET_PRE_ROUTING && hook != NF_INET_LOCAL_IN)
 			break;
 		if ((protocol == 0 && !csum_fold(skb->csum)) ||
 		    !csum_tcpudp_magic(iph->saddr, iph->daddr,
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index b9b189c26208..ca23c63ced37 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -220,11 +220,11 @@ unconditional(const struct ipt_ip *ip)
 #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 static const char *hooknames[] = {
-	[NF_IP_PRE_ROUTING]		= "PREROUTING",
-	[NF_IP_LOCAL_IN]		= "INPUT",
-	[NF_IP_FORWARD]			= "FORWARD",
-	[NF_IP_LOCAL_OUT]		= "OUTPUT",
-	[NF_IP_POST_ROUTING]		= "POSTROUTING",
+	[NF_INET_PRE_ROUTING]		= "PREROUTING",
+	[NF_INET_LOCAL_IN]		= "INPUT",
+	[NF_INET_FORWARD]			= "FORWARD",
+	[NF_INET_LOCAL_OUT]		= "OUTPUT",
+	[NF_INET_POST_ROUTING]		= "POSTROUTING",
 };
 
 enum nf_ip_trace_comments {
@@ -465,7 +465,7 @@ mark_source_chains(struct xt_table_info *newinfo,
 
 	/* No recursion; use packet counter to save back ptrs (reset
 	   to 0 as we leave), and comefrom to save source hook bitmask */
-	for (hook = 0; hook < NF_IP_NUMHOOKS; hook++) {
+	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
 		unsigned int pos = newinfo->hook_entry[hook];
 		struct ipt_entry *e
 			= (struct ipt_entry *)(entry0 + pos);
@@ -481,13 +481,13 @@ mark_source_chains(struct xt_table_info *newinfo,
 				= (void *)ipt_get_target(e);
 			int visited = e->comefrom & (1 << hook);
 
-			if (e->comefrom & (1 << NF_IP_NUMHOOKS)) {
+			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
 				printk("iptables: loop hook %u pos %u %08X.\n",
 				       hook, pos, e->comefrom);
 				return 0;
 			}
 			e->comefrom
-				|= ((1 << hook) | (1 << NF_IP_NUMHOOKS));
+				|= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
 
 			/* Unconditional return/END. */
 			if ((e->target_offset == sizeof(struct ipt_entry)
@@ -507,10 +507,10 @@ mark_source_chains(struct xt_table_info *newinfo,
 				/* Return: backtrack through the last
 				   big jump. */
 				do {
-					e->comefrom ^= (1<<NF_IP_NUMHOOKS);
+					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
 #ifdef DEBUG_IP_FIREWALL_USER
 					if (e->comefrom
-					    & (1 << NF_IP_NUMHOOKS)) {
+					    & (1 << NF_INET_NUMHOOKS)) {
 						duprintf("Back unset "
 							 "on hook %u "
 							 "rule %u\n",
@@ -741,7 +741,7 @@ check_entry_size_and_hooks(struct ipt_entry *e,
 	}
 
 	/* Check hooks & underflows */
-	for (h = 0; h < NF_IP_NUMHOOKS; h++) {
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
 		if ((unsigned char *)e - base == hook_entries[h])
 			newinfo->hook_entry[h] = hook_entries[h];
 		if ((unsigned char *)e - base == underflows[h])
@@ -795,7 +795,7 @@ translate_table(const char *name,
 	newinfo->number = number;
 
 	/* Init all hooks to impossible value. */
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		newinfo->hook_entry[i] = 0xFFFFFFFF;
 		newinfo->underflow[i] = 0xFFFFFFFF;
 	}
@@ -819,7 +819,7 @@ translate_table(const char *name,
 	}
 
 	/* Check hooks all assigned */
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		/* Only hooks which are valid */
 		if (!(valid_hooks & (1 << i)))
 			continue;
@@ -1107,7 +1107,7 @@ static int compat_calc_entry(struct ipt_entry *e, struct xt_table_info *info,
 	if (ret)
 		return ret;
 
-	for (i = 0; i< NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		if (info->hook_entry[i] && (e < (struct ipt_entry *)
 				(base + info->hook_entry[i])))
 			newinfo->hook_entry[i] -= off;
@@ -1130,7 +1130,7 @@ static int compat_table_info(struct xt_table_info *info,
 	memset(newinfo, 0, sizeof(struct xt_table_info));
 	newinfo->size = info->size;
 	newinfo->number = info->number;
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		newinfo->hook_entry[i] = info->hook_entry[i];
 		newinfo->underflow[i] = info->underflow[i];
 	}
@@ -1479,8 +1479,8 @@ struct compat_ipt_replace {
 	u32			valid_hooks;
 	u32			num_entries;
 	u32			size;
-	u32			hook_entry[NF_IP_NUMHOOKS];
-	u32			underflow[NF_IP_NUMHOOKS];
+	u32			hook_entry[NF_INET_NUMHOOKS];
+	u32			underflow[NF_INET_NUMHOOKS];
 	u32			num_counters;
 	compat_uptr_t		counters;	/* struct ipt_counters * */
 	struct compat_ipt_entry	entries[0];
@@ -1645,7 +1645,7 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e,
 		goto out;
 
 	/* Check hooks & underflows */
-	for (h = 0; h < NF_IP_NUMHOOKS; h++) {
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
 		if ((unsigned char *)e - base == hook_entries[h])
 			newinfo->hook_entry[h] = hook_entries[h];
 		if ((unsigned char *)e - base == underflows[h])
@@ -1700,7 +1700,7 @@ static int compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 	xt_compat_target_from_user(t, dstptr, size);
 
 	de->next_offset = e->next_offset - (origsize - *size);
-	for (h = 0; h < NF_IP_NUMHOOKS; h++) {
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
 		if ((unsigned char *)de - base < newinfo->hook_entry[h])
 			newinfo->hook_entry[h] -= origsize - *size;
 		if ((unsigned char *)de - base < newinfo->underflow[h])
@@ -1753,7 +1753,7 @@ translate_compat_table(const char *name,
 	info->number = number;
 
 	/* Init all hooks to impossible value. */
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		info->hook_entry[i] = 0xFFFFFFFF;
 		info->underflow[i] = 0xFFFFFFFF;
 	}
@@ -1778,7 +1778,7 @@ translate_compat_table(const char *name,
 	}
 
 	/* Check hooks all assigned */
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		/* Only hooks which are valid */
 		if (!(valid_hooks & (1 << i)))
 			continue;
@@ -1800,7 +1800,7 @@ translate_compat_table(const char *name,
 		goto out_unlock;
 
 	newinfo->number = number;
-	for (i = 0; i < NF_IP_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		newinfo->hook_entry[i] = info->hook_entry[i];
 		newinfo->underflow[i] = info->underflow[i];
 	}
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index 44b516e7cb79..5a18997bb3d3 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -67,7 +67,7 @@ masquerade_target(struct sk_buff *skb,
 	const struct rtable *rt;
 	__be32 newsrc;
 
-	NF_CT_ASSERT(hooknum == NF_IP_POST_ROUTING);
+	NF_CT_ASSERT(hooknum == NF_INET_POST_ROUTING);
 
 	ct = nf_ct_get(skb, &ctinfo);
 	nat = nfct_nat(ct);
@@ -172,7 +172,7 @@ static struct xt_target masquerade __read_mostly = {
 	.target		= masquerade_target,
 	.targetsize	= sizeof(struct nf_nat_multi_range_compat),
 	.table		= "nat",
-	.hooks		= 1 << NF_IP_POST_ROUTING,
+	.hooks		= 1 << NF_INET_POST_ROUTING,
 	.checkentry	= masquerade_check,
 	.me		= THIS_MODULE,
 };
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
index f8699291e33d..973bbee7ee1f 100644
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ b/net/ipv4/netfilter/ipt_NETMAP.c
@@ -56,14 +56,14 @@ target(struct sk_buff *skb,
 	const struct nf_nat_multi_range_compat *mr = targinfo;
 	struct nf_nat_range newrange;
 
-	NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING
-		     || hooknum == NF_IP_POST_ROUTING
-		     || hooknum == NF_IP_LOCAL_OUT);
+	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING
+		     || hooknum == NF_INET_POST_ROUTING
+		     || hooknum == NF_INET_LOCAL_OUT);
 	ct = nf_ct_get(skb, &ctinfo);
 
 	netmask = ~(mr->range[0].min_ip ^ mr->range[0].max_ip);
 
-	if (hooknum == NF_IP_PRE_ROUTING || hooknum == NF_IP_LOCAL_OUT)
+	if (hooknum == NF_INET_PRE_ROUTING || hooknum == NF_INET_LOCAL_OUT)
 		new_ip = ip_hdr(skb)->daddr & ~netmask;
 	else
 		new_ip = ip_hdr(skb)->saddr & ~netmask;
@@ -84,8 +84,9 @@ static struct xt_target target_module __read_mostly = {
 	.target 	= target,
 	.targetsize	= sizeof(struct nf_nat_multi_range_compat),
 	.table		= "nat",
-	.hooks		= (1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_POST_ROUTING) |
-			  (1 << NF_IP_LOCAL_OUT),
+	.hooks		= (1 << NF_INET_PRE_ROUTING) |
+			  (1 << NF_INET_POST_ROUTING) |
+			  (1 << NF_INET_LOCAL_OUT),
 	.checkentry 	= check,
 	.me 		= THIS_MODULE
 };
diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c
index f7cf7d61a2d4..4757af293ba4 100644
--- a/net/ipv4/netfilter/ipt_REDIRECT.c
+++ b/net/ipv4/netfilter/ipt_REDIRECT.c
@@ -60,14 +60,14 @@ redirect_target(struct sk_buff *skb,
 	const struct nf_nat_multi_range_compat *mr = targinfo;
 	struct nf_nat_range newrange;
 
-	NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING
-		     || hooknum == NF_IP_LOCAL_OUT);
+	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING
+		     || hooknum == NF_INET_LOCAL_OUT);
 
 	ct = nf_ct_get(skb, &ctinfo);
 	NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED));
 
 	/* Local packets: make them go to loopback */
-	if (hooknum == NF_IP_LOCAL_OUT)
+	if (hooknum == NF_INET_LOCAL_OUT)
 		newdst = htonl(0x7F000001);
 	else {
 		struct in_device *indev;
@@ -101,7 +101,7 @@ static struct xt_target redirect_reg __read_mostly = {
 	.target		= redirect_target,
 	.targetsize	= sizeof(struct nf_nat_multi_range_compat),
 	.table		= "nat",
-	.hooks		= (1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_OUT),
+	.hooks		= (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT),
 	.checkentry	= redirect_check,
 	.me		= THIS_MODULE,
 };
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index ccb2a03dcd5a..d55b262bf608 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -123,7 +123,7 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	niph->id = 0;
 
 	addr_type = RTN_UNSPEC;
-	if (hook != NF_IP_FORWARD
+	if (hook != NF_INET_FORWARD
 #ifdef CONFIG_BRIDGE_NETFILTER
 	    || (nskb->nf_bridge && nskb->nf_bridge->mask & BRNF_BRIDGED)
 #endif
@@ -234,8 +234,8 @@ static struct xt_target ipt_reject_reg __read_mostly = {
 	.target		= reject,
 	.targetsize	= sizeof(struct ipt_reject_info),
 	.table		= "filter",
-	.hooks		= (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD) |
-			  (1 << NF_IP_LOCAL_OUT),
+	.hooks		= (1 << NF_INET_LOCAL_IN) | (1 << NF_INET_FORWARD) |
+			  (1 << NF_INET_LOCAL_OUT),
 	.checkentry	= check,
 	.me		= THIS_MODULE,
 };
diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c
index 8988571436b8..f2f62b5ce9aa 100644
--- a/net/ipv4/netfilter/ipt_SAME.c
+++ b/net/ipv4/netfilter/ipt_SAME.c
@@ -119,8 +119,8 @@ same_target(struct sk_buff *skb,
 	struct nf_nat_range newrange;
 	const struct nf_conntrack_tuple *t;
 
-	NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING ||
-			hooknum == NF_IP_POST_ROUTING);
+	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING ||
+			hooknum == NF_INET_POST_ROUTING);
 	ct = nf_ct_get(skb, &ctinfo);
 
 	t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
@@ -158,7 +158,8 @@ static struct xt_target same_reg __read_mostly = {
 	.target		= same_target,
 	.targetsize	= sizeof(struct ipt_same_info),
 	.table		= "nat",
-	.hooks		= (1 << NF_IP_PRE_ROUTING | 1 << NF_IP_POST_ROUTING),
+	.hooks		= (1 << NF_INET_PRE_ROUTING) |
+			  (1 << NF_INET_POST_ROUTING),
 	.checkentry	= same_check,
 	.destroy	= same_destroy,
 	.me		= THIS_MODULE,
diff --git a/net/ipv4/netfilter/ipt_owner.c b/net/ipv4/netfilter/ipt_owner.c
index b14e77da7a33..6bc4bfea66d6 100644
--- a/net/ipv4/netfilter/ipt_owner.c
+++ b/net/ipv4/netfilter/ipt_owner.c
@@ -73,7 +73,8 @@ static struct xt_match owner_match __read_mostly = {
 	.family		= AF_INET,
 	.match		= match,
 	.matchsize	= sizeof(struct ipt_owner_info),
-	.hooks		= (1 << NF_IP_LOCAL_OUT) | (1 << NF_IP_POST_ROUTING),
+	.hooks		= (1 << NF_INET_LOCAL_OUT) |
+			  (1 << NF_INET_POST_ROUTING),
 	.checkentry	= checkentry,
 	.me		= THIS_MODULE,
 };
diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c
index ba3262c60437..06ab64e30e88 100644
--- a/net/ipv4/netfilter/iptable_filter.c
+++ b/net/ipv4/netfilter/iptable_filter.c
@@ -19,7 +19,9 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("iptables filter table");
 
-#define FILTER_VALID_HOOKS ((1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD) | (1 << NF_IP_LOCAL_OUT))
+#define FILTER_VALID_HOOKS ((1 << NF_INET_LOCAL_IN) | \
+			    (1 << NF_INET_FORWARD) | \
+			    (1 << NF_INET_LOCAL_OUT))
 
 static struct
 {
@@ -33,14 +35,14 @@ static struct
 		.num_entries = 4,
 		.size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error),
 		.hook_entry = {
-			[NF_IP_LOCAL_IN] = 0,
-			[NF_IP_FORWARD] = sizeof(struct ipt_standard),
-			[NF_IP_LOCAL_OUT] = sizeof(struct ipt_standard) * 2,
+			[NF_INET_LOCAL_IN] = 0,
+			[NF_INET_FORWARD] = sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2,
 		},
 		.underflow = {
-			[NF_IP_LOCAL_IN] = 0,
-			[NF_IP_FORWARD] = sizeof(struct ipt_standard),
-			[NF_IP_LOCAL_OUT] = sizeof(struct ipt_standard) * 2,
+			[NF_INET_LOCAL_IN] = 0,
+			[NF_INET_FORWARD] = sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2,
 		},
 	},
 	.entries = {
@@ -94,21 +96,21 @@ static struct nf_hook_ops ipt_ops[] = {
 		.hook		= ipt_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_FILTER,
 	},
 	{
 		.hook		= ipt_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_FORWARD,
+		.hooknum	= NF_INET_FORWARD,
 		.priority	= NF_IP_PRI_FILTER,
 	},
 	{
 		.hook		= ipt_local_out_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_FILTER,
 	},
 };
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index b4360a69d5ca..0335827d3e4d 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -21,11 +21,11 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("iptables mangle table");
 
-#define MANGLE_VALID_HOOKS ((1 << NF_IP_PRE_ROUTING) | \
-			    (1 << NF_IP_LOCAL_IN) | \
-			    (1 << NF_IP_FORWARD) | \
-			    (1 << NF_IP_LOCAL_OUT) | \
-			    (1 << NF_IP_POST_ROUTING))
+#define MANGLE_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | \
+			    (1 << NF_INET_LOCAL_IN) | \
+			    (1 << NF_INET_FORWARD) | \
+			    (1 << NF_INET_LOCAL_OUT) | \
+			    (1 << NF_INET_POST_ROUTING))
 
 /* Ouch - five different hooks? Maybe this should be a config option..... -- BC */
 static struct
@@ -40,18 +40,18 @@ static struct
 		.num_entries = 6,
 		.size = sizeof(struct ipt_standard) * 5 + sizeof(struct ipt_error),
 		.hook_entry = {
-			[NF_IP_PRE_ROUTING] 	= 0,
-			[NF_IP_LOCAL_IN] 	= sizeof(struct ipt_standard),
-			[NF_IP_FORWARD] 	= sizeof(struct ipt_standard) * 2,
-			[NF_IP_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 3,
-			[NF_IP_POST_ROUTING] 	= sizeof(struct ipt_standard) * 4,
+			[NF_INET_PRE_ROUTING] 	= 0,
+			[NF_INET_LOCAL_IN] 	= sizeof(struct ipt_standard),
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard) * 2,
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 3,
+			[NF_INET_POST_ROUTING] 	= sizeof(struct ipt_standard) * 4,
 		},
 		.underflow = {
-			[NF_IP_PRE_ROUTING] 	= 0,
-			[NF_IP_LOCAL_IN] 	= sizeof(struct ipt_standard),
-			[NF_IP_FORWARD] 	= sizeof(struct ipt_standard) * 2,
-			[NF_IP_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 3,
-			[NF_IP_POST_ROUTING]	= sizeof(struct ipt_standard) * 4,
+			[NF_INET_PRE_ROUTING] 	= 0,
+			[NF_INET_LOCAL_IN] 	= sizeof(struct ipt_standard),
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard) * 2,
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 3,
+			[NF_INET_POST_ROUTING]	= sizeof(struct ipt_standard) * 4,
 		},
 	},
 	.entries = {
@@ -133,35 +133,35 @@ static struct nf_hook_ops ipt_ops[] = {
 		.hook		= ipt_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
 		.hook		= ipt_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
 		.hook		= ipt_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_FORWARD,
+		.hooknum	= NF_INET_FORWARD,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
 		.hook		= ipt_local_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
 		.hook		= ipt_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 };
diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c
index f8678651250f..66be23295594 100644
--- a/net/ipv4/netfilter/iptable_raw.c
+++ b/net/ipv4/netfilter/iptable_raw.c
@@ -7,7 +7,7 @@
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <net/ip.h>
 
-#define RAW_VALID_HOOKS ((1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_OUT))
+#define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT))
 
 static struct
 {
@@ -21,12 +21,12 @@ static struct
 		.num_entries = 3,
 		.size = sizeof(struct ipt_standard) * 2 + sizeof(struct ipt_error),
 		.hook_entry = {
-			[NF_IP_PRE_ROUTING] = 0,
-			[NF_IP_LOCAL_OUT] = sizeof(struct ipt_standard)
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard)
 		},
 		.underflow = {
-			[NF_IP_PRE_ROUTING] = 0,
-			[NF_IP_LOCAL_OUT]  = sizeof(struct ipt_standard)
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_LOCAL_OUT]  = sizeof(struct ipt_standard)
 		},
 	},
 	.entries = {
@@ -78,14 +78,14 @@ static struct nf_hook_ops ipt_ops[] = {
 	{
 		.hook = ipt_hook,
 		.pf = PF_INET,
-		.hooknum = NF_IP_PRE_ROUTING,
+		.hooknum = NF_INET_PRE_ROUTING,
 		.priority = NF_IP_PRI_RAW,
 		.owner = THIS_MODULE,
 	},
 	{
 		.hook = ipt_local_hook,
 		.pf = PF_INET,
-		.hooknum = NF_IP_LOCAL_OUT,
+		.hooknum = NF_INET_LOCAL_OUT,
 		.priority = NF_IP_PRI_RAW,
 		.owner = THIS_MODULE,
 	},
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index 910dae732a0f..c91725a85789 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -150,7 +150,7 @@ static unsigned int ipv4_conntrack_defrag(unsigned int hooknum,
 	/* Gather fragments. */
 	if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
 		if (nf_ct_ipv4_gather_frags(skb,
-					    hooknum == NF_IP_PRE_ROUTING ?
+					    hooknum == NF_INET_PRE_ROUTING ?
 					    IP_DEFRAG_CONNTRACK_IN :
 					    IP_DEFRAG_CONNTRACK_OUT))
 			return NF_STOLEN;
@@ -190,56 +190,56 @@ static struct nf_hook_ops ipv4_conntrack_ops[] = {
 		.hook		= ipv4_conntrack_defrag,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_CONNTRACK_DEFRAG,
 	},
 	{
 		.hook		= ipv4_conntrack_in,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_CONNTRACK,
 	},
 	{
 		.hook           = ipv4_conntrack_defrag,
 		.owner          = THIS_MODULE,
 		.pf             = PF_INET,
-		.hooknum        = NF_IP_LOCAL_OUT,
+		.hooknum        = NF_INET_LOCAL_OUT,
 		.priority       = NF_IP_PRI_CONNTRACK_DEFRAG,
 	},
 	{
 		.hook		= ipv4_conntrack_local,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_CONNTRACK,
 	},
 	{
 		.hook		= ipv4_conntrack_help,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
 	},
 	{
 		.hook		= ipv4_conntrack_help,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
 	},
 	{
 		.hook		= ipv4_confirm,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
 	},
 	{
 		.hook		= ipv4_confirm,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
 	},
 };
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index adcbaf6d4299..0e2c448ea389 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -195,7 +195,7 @@ icmp_error(struct sk_buff *skb, unsigned int dataoff,
 	}
 
 	/* See ip_conntrack_proto_tcp.c */
-	if (nf_conntrack_checksum && hooknum == NF_IP_PRE_ROUTING &&
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
 	    nf_ip_checksum(skb, hooknum, dataoff, 0)) {
 		if (LOG_INVALID(IPPROTO_ICMP))
 			nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL,
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 86b465b176ba..d237511cf46c 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -213,9 +213,9 @@ find_best_ips_proto(struct nf_conntrack_tuple *tuple,
 	*var_ipp = htonl(minip + j % (maxip - minip + 1));
 }
 
-/* Manipulate the tuple into the range given.  For NF_IP_POST_ROUTING,
- * we change the source to map into the range.  For NF_IP_PRE_ROUTING
- * and NF_IP_LOCAL_OUT, we change the destination to map into the
+/* Manipulate the tuple into the range given.  For NF_INET_POST_ROUTING,
+ * we change the source to map into the range.  For NF_INET_PRE_ROUTING
+ * and NF_INET_LOCAL_OUT, we change the destination to map into the
  * range.  It might not be possible to get a unique tuple, but we try.
  * At worst (or if we race), we will end up with a final duplicate in
  * __ip_conntrack_confirm and drop the packet. */
@@ -293,10 +293,10 @@ nf_nat_setup_info(struct nf_conn *ct,
 		}
 	}
 
-	NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING ||
-		     hooknum == NF_IP_POST_ROUTING ||
-		     hooknum == NF_IP_LOCAL_IN ||
-		     hooknum == NF_IP_LOCAL_OUT);
+	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING ||
+		     hooknum == NF_INET_POST_ROUTING ||
+		     hooknum == NF_INET_LOCAL_IN ||
+		     hooknum == NF_INET_LOCAL_OUT);
 	BUG_ON(nf_nat_initialized(ct, maniptype));
 
 	/* What we've got will look like inverse of reply. Normally
diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c
index 93e18ef114f2..0f226df76f5c 100644
--- a/net/ipv4/netfilter/nf_nat_h323.c
+++ b/net/ipv4/netfilter/nf_nat_h323.c
@@ -391,7 +391,7 @@ static void ip_nat_q931_expect(struct nf_conn *new,
 	range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip;
 
 	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING);
+	nf_nat_setup_info(new, &range, NF_INET_POST_ROUTING);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
@@ -400,7 +400,7 @@ static void ip_nat_q931_expect(struct nf_conn *new,
 	    new->master->tuplehash[!this->dir].tuple.src.u3.ip;
 
 	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING);
+	nf_nat_setup_info(new, &range, NF_INET_PRE_ROUTING);
 }
 
 /****************************************************************************/
@@ -481,7 +481,7 @@ static void ip_nat_callforwarding_expect(struct nf_conn *new,
 	range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip;
 
 	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING);
+	nf_nat_setup_info(new, &range, NF_INET_POST_ROUTING);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
@@ -489,7 +489,7 @@ static void ip_nat_callforwarding_expect(struct nf_conn *new,
 	range.min_ip = range.max_ip = this->saved_ip;
 
 	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING);
+	nf_nat_setup_info(new, &range, NF_INET_PRE_ROUTING);
 }
 
 /****************************************************************************/
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index 8718da00ef2a..d00b8b2891fb 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -431,7 +431,7 @@ void nf_nat_follow_master(struct nf_conn *ct,
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
 	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
@@ -439,6 +439,6 @@ void nf_nat_follow_master(struct nf_conn *ct,
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.src.u3.ip;
 	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
 }
 EXPORT_SYMBOL(nf_nat_follow_master);
diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c
index 6817e7995f35..c540999f5090 100644
--- a/net/ipv4/netfilter/nf_nat_pptp.c
+++ b/net/ipv4/netfilter/nf_nat_pptp.c
@@ -94,7 +94,7 @@ static void pptp_nat_expected(struct nf_conn *ct,
 		range.min = range.max = exp->saved_proto;
 	}
 	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = IP_NAT_RANGE_MAP_IPS;
@@ -105,7 +105,7 @@ static void pptp_nat_expected(struct nf_conn *ct,
 		range.min = range.max = exp->saved_proto;
 	}
 	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
 }
 
 /* outbound packets == from PNS to PAC */
diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c
index 46b25ab5f78b..ee39ed87bb08 100644
--- a/net/ipv4/netfilter/nf_nat_rule.c
+++ b/net/ipv4/netfilter/nf_nat_rule.c
@@ -24,7 +24,9 @@
 #include <net/netfilter/nf_nat_core.h>
 #include <net/netfilter/nf_nat_rule.h>
 
-#define NAT_VALID_HOOKS ((1<<NF_IP_PRE_ROUTING) | (1<<NF_IP_POST_ROUTING) | (1<<NF_IP_LOCAL_OUT))
+#define NAT_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | \
+			 (1 << NF_INET_POST_ROUTING) | \
+			 (1 << NF_INET_LOCAL_OUT))
 
 static struct
 {
@@ -38,14 +40,14 @@ static struct
 		.num_entries = 4,
 		.size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error),
 		.hook_entry = {
-			[NF_IP_PRE_ROUTING] = 0,
-			[NF_IP_POST_ROUTING] = sizeof(struct ipt_standard),
-			[NF_IP_LOCAL_OUT] = sizeof(struct ipt_standard) * 2
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_POST_ROUTING] = sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2
 		},
 		.underflow = {
-			[NF_IP_PRE_ROUTING] = 0,
-			[NF_IP_POST_ROUTING] = sizeof(struct ipt_standard),
-			[NF_IP_LOCAL_OUT] = sizeof(struct ipt_standard) * 2
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_POST_ROUTING] = sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2
 		},
 	},
 	.entries = {
@@ -76,7 +78,7 @@ static unsigned int ipt_snat_target(struct sk_buff *skb,
 	enum ip_conntrack_info ctinfo;
 	const struct nf_nat_multi_range_compat *mr = targinfo;
 
-	NF_CT_ASSERT(hooknum == NF_IP_POST_ROUTING);
+	NF_CT_ASSERT(hooknum == NF_INET_POST_ROUTING);
 
 	ct = nf_ct_get(skb, &ctinfo);
 
@@ -118,15 +120,15 @@ static unsigned int ipt_dnat_target(struct sk_buff *skb,
 	enum ip_conntrack_info ctinfo;
 	const struct nf_nat_multi_range_compat *mr = targinfo;
 
-	NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING ||
-		     hooknum == NF_IP_LOCAL_OUT);
+	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING ||
+		     hooknum == NF_INET_LOCAL_OUT);
 
 	ct = nf_ct_get(skb, &ctinfo);
 
 	/* Connection must be valid and new. */
 	NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED));
 
-	if (hooknum == NF_IP_LOCAL_OUT &&
+	if (hooknum == NF_INET_LOCAL_OUT &&
 	    mr->range[0].flags & IP_NAT_RANGE_MAP_IPS)
 		warn_if_extra_mangle(ip_hdr(skb)->daddr,
 				     mr->range[0].min_ip);
@@ -227,7 +229,7 @@ static struct xt_target ipt_snat_reg __read_mostly = {
 	.target		= ipt_snat_target,
 	.targetsize	= sizeof(struct nf_nat_multi_range_compat),
 	.table		= "nat",
-	.hooks		= 1 << NF_IP_POST_ROUTING,
+	.hooks		= 1 << NF_INET_POST_ROUTING,
 	.checkentry	= ipt_snat_checkentry,
 	.family		= AF_INET,
 };
@@ -237,7 +239,7 @@ static struct xt_target ipt_dnat_reg __read_mostly = {
 	.target		= ipt_dnat_target,
 	.targetsize	= sizeof(struct nf_nat_multi_range_compat),
 	.table		= "nat",
-	.hooks		= (1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_OUT),
+	.hooks		= (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT),
 	.checkentry	= ipt_dnat_checkentry,
 	.family		= AF_INET,
 };
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index 8996ccb757db..b8c0720cf428 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -229,14 +229,14 @@ static void ip_nat_sdp_expect(struct nf_conn *ct,
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
 	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
 	range.min = range.max = exp->saved_proto;
 	range.min_ip = range.max_ip = exp->saved_ip;
 	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
 }
 
 /* So, this packet has hit the connection tracking matching code.
diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c
index 7db76ea9af91..84172e9dcb16 100644
--- a/net/ipv4/netfilter/nf_nat_standalone.c
+++ b/net/ipv4/netfilter/nf_nat_standalone.c
@@ -137,7 +137,7 @@ nf_nat_fn(unsigned int hooknum,
 			if (unlikely(nf_ct_is_confirmed(ct)))
 				/* NAT module was loaded late */
 				ret = alloc_null_binding_confirmed(ct, hooknum);
-			else if (hooknum == NF_IP_LOCAL_IN)
+			else if (hooknum == NF_INET_LOCAL_IN)
 				/* LOCAL_IN hook doesn't have a chain!  */
 				ret = alloc_null_binding(ct, hooknum);
 			else
@@ -279,7 +279,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_in,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_NAT_DST,
 	},
 	/* After packet filtering, change source */
@@ -287,7 +287,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_out,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
 	/* After conntrack, adjust sequence number */
@@ -295,7 +295,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_adjust,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
 	},
 	/* Before packet filtering, change destination */
@@ -303,7 +303,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_local_fn,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_NAT_DST,
 	},
 	/* After packet filtering, change source */
@@ -311,7 +311,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_fn,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
 	/* After conntrack, adjust sequence number */
@@ -319,7 +319,7 @@ static struct nf_hook_ops nf_nat_ops[] = {
 		.hook		= nf_nat_adjust,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
-		.hooknum	= NF_IP_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
 	},
 };
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 761056ef4932..b80987d2fc55 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -321,7 +321,7 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length,
 		icmp_out_count(((struct icmphdr *)
 			skb_transport_header(skb))->type);
 
-	err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
+	err = NF_HOOK(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
 		      dst_output);
 	if (err > 0)
 		err = inet->recverr ? net_xmit_errno(err) : 0;
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index d5890c84a492..0c377a66b8b5 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -55,7 +55,7 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async)
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
 
-	NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(PF_INET, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
 		xfrm4_rcv_encap_finish);
 	return 0;
 #else
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 1900200d3c0f..d5a58a818021 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -86,7 +86,7 @@ static int xfrm4_output_finish(struct sk_buff *skb)
 
 int xfrm4_output(struct sk_buff *skb)
 {
-	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev,
-			    xfrm4_output_finish,
+	return NF_HOOK_COND(PF_INET, NF_INET_POST_ROUTING, skb,
+			    NULL, skb->dst->dev, xfrm4_output_finish,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index d837784a2199..296113598944 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -66,7 +66,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.proto			= IPPROTO_IPIP,
 	.eth_proto		= htons(ETH_P_IP),
-	.nf_post_routing	= NF_IP_POST_ROUTING,
+	.nf_post_routing	= NF_INET_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index fac6f7f9dd73..79610b4bad3e 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -134,7 +134,8 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 
 	rcu_read_unlock();
 
-	return NF_HOOK(PF_INET6,NF_IP6_PRE_ROUTING, skb, dev, NULL, ip6_rcv_finish);
+	return NF_HOOK(PF_INET6, NF_INET_PRE_ROUTING, skb, dev, NULL,
+		       ip6_rcv_finish);
 err:
 	IP6_INC_STATS_BH(idev, IPSTATS_MIB_INHDRERRORS);
 drop:
@@ -229,7 +230,8 @@ discard:
 
 int ip6_input(struct sk_buff *skb)
 {
-	return NF_HOOK(PF_INET6,NF_IP6_LOCAL_IN, skb, skb->dev, NULL, ip6_input_finish);
+	return NF_HOOK(PF_INET6, NF_INET_LOCAL_IN, skb, skb->dev, NULL,
+		       ip6_input_finish);
 }
 
 int ip6_mc_input(struct sk_buff *skb)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index bd121f9ae0a7..d54da616e3af 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -79,7 +79,7 @@ int __ip6_local_out(struct sk_buff *skb)
 		len = 0;
 	ipv6_hdr(skb)->payload_len = htons(len);
 
-	return nf_hook(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev,
+	return nf_hook(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dst->dev,
 		       dst_output);
 }
 
@@ -145,8 +145,8 @@ static int ip6_output2(struct sk_buff *skb)
 			   is not supported in any case.
 			 */
 			if (newskb)
-				NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, newskb, NULL,
-					newskb->dev,
+				NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, newskb,
+					NULL, newskb->dev,
 					ip6_dev_loopback_xmit);
 
 			if (ipv6_hdr(skb)->hop_limit == 0) {
@@ -159,7 +159,8 @@ static int ip6_output2(struct sk_buff *skb)
 		IP6_INC_STATS(idev, IPSTATS_MIB_OUTMCASTPKTS);
 	}
 
-	return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb,NULL, skb->dev,ip6_output_finish);
+	return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb->dev,
+		       ip6_output_finish);
 }
 
 static inline int ip6_skb_dst_mtu(struct sk_buff *skb)
@@ -261,7 +262,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
 	if ((skb->len <= mtu) || ipfragok || skb_is_gso(skb)) {
 		IP6_INC_STATS(ip6_dst_idev(skb->dst),
 			      IPSTATS_MIB_OUTREQUESTS);
-		return NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev,
+		return NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, dst->dev,
 				dst_output);
 	}
 
@@ -525,7 +526,8 @@ int ip6_forward(struct sk_buff *skb)
 	hdr->hop_limit--;
 
 	IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS);
-	return NF_HOOK(PF_INET6,NF_IP6_FORWARD, skb, skb->dev, dst->dev, ip6_forward_finish);
+	return NF_HOOK(PF_INET6, NF_INET_FORWARD, skb, skb->dev, dst->dev,
+		       ip6_forward_finish);
 
 error:
 	IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INADDRERRORS);
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 17d7318ff7bf..82b12940c2a0 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1448,7 +1448,7 @@ static inline int mld_dev_queue_xmit2(struct sk_buff *skb)
 
 static inline int mld_dev_queue_xmit(struct sk_buff *skb)
 {
-	return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL, skb->dev,
+	return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb->dev,
 		       mld_dev_queue_xmit2);
 }
 
@@ -1469,7 +1469,7 @@ static void mld_sendpack(struct sk_buff *skb)
 	pmr->csum = csum_ipv6_magic(&pip6->saddr, &pip6->daddr, mldlen,
 		IPPROTO_ICMPV6, csum_partial(skb_transport_header(skb),
 					     mldlen, 0));
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dev,
+	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev,
 		mld_dev_queue_xmit);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS_BH(idev, ICMPV6_MLD2_REPORT);
@@ -1813,7 +1813,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 
 	idev = in6_dev_get(skb->dev);
 
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dev,
+	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev,
 		mld_dev_queue_xmit);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(idev, type);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 85947eae5bf7..b2531f80317e 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -533,7 +533,8 @@ static void __ndisc_send(struct net_device *dev,
 	idev = in6_dev_get(dst->dev);
 	IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS);
 
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output);
+	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, dst->dev,
+		      dst_output);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(idev, type);
 		ICMP6_INC_STATS(idev, ICMP6_MIB_OUTMSGS);
@@ -1538,7 +1539,8 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 	buff->dst = dst;
 	idev = in6_dev_get(dst->dev);
 	IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS);
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, buff, NULL, dst->dev, dst_output);
+	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, buff, NULL, dst->dev,
+		      dst_output);
 	if (!err) {
 		ICMP6MSGOUT_INC_STATS(idev, NDISC_REDIRECT);
 		ICMP6_INC_STATS(idev, ICMP6_MIB_OUTMSGS);
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index b1326c2bf8aa..175e19f80253 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -60,7 +60,7 @@ static void nf_ip6_saveroute(const struct sk_buff *skb, struct nf_info *info)
 {
 	struct ip6_rt_info *rt_info = nf_info_reroute(info);
 
-	if (info->hook == NF_IP6_LOCAL_OUT) {
+	if (info->hook == NF_INET_LOCAL_OUT) {
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 
 		rt_info->daddr = iph->daddr;
@@ -72,7 +72,7 @@ static int nf_ip6_reroute(struct sk_buff *skb, const struct nf_info *info)
 {
 	struct ip6_rt_info *rt_info = nf_info_reroute(info);
 
-	if (info->hook == NF_IP6_LOCAL_OUT) {
+	if (info->hook == NF_INET_LOCAL_OUT) {
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 		if (!ipv6_addr_equal(&iph->daddr, &rt_info->daddr) ||
 		    !ipv6_addr_equal(&iph->saddr, &rt_info->saddr))
@@ -89,7 +89,7 @@ __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 
 	switch (skb->ip_summed) {
 	case CHECKSUM_COMPLETE:
-		if (hook != NF_IP6_PRE_ROUTING && hook != NF_IP6_LOCAL_IN)
+		if (hook != NF_INET_PRE_ROUTING && hook != NF_INET_LOCAL_IN)
 			break;
 		if (!csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr,
 				     skb->len - dataoff, protocol,
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index acaba1537931..e1e87eff4686 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -258,11 +258,11 @@ unconditional(const struct ip6t_ip6 *ipv6)
     defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
 /* This cries for unification! */
 static const char *hooknames[] = {
-	[NF_IP6_PRE_ROUTING]		= "PREROUTING",
-	[NF_IP6_LOCAL_IN]		= "INPUT",
-	[NF_IP6_FORWARD]		= "FORWARD",
-	[NF_IP6_LOCAL_OUT]		= "OUTPUT",
-	[NF_IP6_POST_ROUTING]		= "POSTROUTING",
+	[NF_INET_PRE_ROUTING]		= "PREROUTING",
+	[NF_INET_LOCAL_IN]		= "INPUT",
+	[NF_INET_FORWARD]		= "FORWARD",
+	[NF_INET_LOCAL_OUT]		= "OUTPUT",
+	[NF_INET_POST_ROUTING]		= "POSTROUTING",
 };
 
 enum nf_ip_trace_comments {
@@ -502,7 +502,7 @@ mark_source_chains(struct xt_table_info *newinfo,
 
 	/* No recursion; use packet counter to save back ptrs (reset
 	   to 0 as we leave), and comefrom to save source hook bitmask */
-	for (hook = 0; hook < NF_IP6_NUMHOOKS; hook++) {
+	for (hook = 0; hook < NF_INET_NUMHOOKS; hook++) {
 		unsigned int pos = newinfo->hook_entry[hook];
 		struct ip6t_entry *e
 			= (struct ip6t_entry *)(entry0 + pos);
@@ -518,13 +518,13 @@ mark_source_chains(struct xt_table_info *newinfo,
 			struct ip6t_standard_target *t
 				= (void *)ip6t_get_target(e);
 
-			if (e->comefrom & (1 << NF_IP6_NUMHOOKS)) {
+			if (e->comefrom & (1 << NF_INET_NUMHOOKS)) {
 				printk("iptables: loop hook %u pos %u %08X.\n",
 				       hook, pos, e->comefrom);
 				return 0;
 			}
 			e->comefrom
-				|= ((1 << hook) | (1 << NF_IP6_NUMHOOKS));
+				|= ((1 << hook) | (1 << NF_INET_NUMHOOKS));
 
 			/* Unconditional return/END. */
 			if ((e->target_offset == sizeof(struct ip6t_entry)
@@ -544,10 +544,10 @@ mark_source_chains(struct xt_table_info *newinfo,
 				/* Return: backtrack through the last
 				   big jump. */
 				do {
-					e->comefrom ^= (1<<NF_IP6_NUMHOOKS);
+					e->comefrom ^= (1<<NF_INET_NUMHOOKS);
 #ifdef DEBUG_IP_FIREWALL_USER
 					if (e->comefrom
-					    & (1 << NF_IP6_NUMHOOKS)) {
+					    & (1 << NF_INET_NUMHOOKS)) {
 						duprintf("Back unset "
 							 "on hook %u "
 							 "rule %u\n",
@@ -746,7 +746,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e,
 	}
 
 	/* Check hooks & underflows */
-	for (h = 0; h < NF_IP6_NUMHOOKS; h++) {
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
 		if ((unsigned char *)e - base == hook_entries[h])
 			newinfo->hook_entry[h] = hook_entries[h];
 		if ((unsigned char *)e - base == underflows[h])
@@ -800,7 +800,7 @@ translate_table(const char *name,
 	newinfo->number = number;
 
 	/* Init all hooks to impossible value. */
-	for (i = 0; i < NF_IP6_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		newinfo->hook_entry[i] = 0xFFFFFFFF;
 		newinfo->underflow[i] = 0xFFFFFFFF;
 	}
@@ -824,7 +824,7 @@ translate_table(const char *name,
 	}
 
 	/* Check hooks all assigned */
-	for (i = 0; i < NF_IP6_NUMHOOKS; i++) {
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
 		/* Only hooks which are valid */
 		if (!(valid_hooks & (1 << i)))
 			continue;
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index c1c663482837..960ba1780a9c 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -164,7 +164,7 @@ static void send_reset(struct sk_buff *oldskb)
 static inline void
 send_unreach(struct sk_buff *skb_in, unsigned char code, unsigned int hooknum)
 {
-	if (hooknum == NF_IP6_LOCAL_OUT && skb_in->dev == NULL)
+	if (hooknum == NF_INET_LOCAL_OUT && skb_in->dev == NULL)
 		skb_in->dev = init_net.loopback_dev;
 
 	icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0, NULL);
@@ -243,8 +243,8 @@ static struct xt_target ip6t_reject_reg __read_mostly = {
 	.target		= reject6_target,
 	.targetsize	= sizeof(struct ip6t_reject_info),
 	.table		= "filter",
-	.hooks		= (1 << NF_IP6_LOCAL_IN) | (1 << NF_IP6_FORWARD) |
-			  (1 << NF_IP6_LOCAL_OUT),
+	.hooks		= (1 << NF_INET_LOCAL_IN) | (1 << NF_INET_FORWARD) |
+			  (1 << NF_INET_LOCAL_OUT),
 	.checkentry	= check,
 	.me		= THIS_MODULE
 };
diff --git a/net/ipv6/netfilter/ip6t_eui64.c b/net/ipv6/netfilter/ip6t_eui64.c
index 41df9a578c7a..ff71269579da 100644
--- a/net/ipv6/netfilter/ip6t_eui64.c
+++ b/net/ipv6/netfilter/ip6t_eui64.c
@@ -67,8 +67,8 @@ static struct xt_match eui64_match __read_mostly = {
 	.family		= AF_INET6,
 	.match		= match,
 	.matchsize	= sizeof(int),
-	.hooks		= (1 << NF_IP6_PRE_ROUTING) | (1 << NF_IP6_LOCAL_IN) |
-			  (1 << NF_IP6_FORWARD),
+	.hooks		= (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN) |
+			  (1 << NF_INET_FORWARD),
 	.me		= THIS_MODULE,
 };
 
diff --git a/net/ipv6/netfilter/ip6t_owner.c b/net/ipv6/netfilter/ip6t_owner.c
index 6036613aef36..1e0dc4a972cf 100644
--- a/net/ipv6/netfilter/ip6t_owner.c
+++ b/net/ipv6/netfilter/ip6t_owner.c
@@ -73,7 +73,8 @@ static struct xt_match owner_match __read_mostly = {
 	.family		= AF_INET6,
 	.match		= match,
 	.matchsize	= sizeof(struct ip6t_owner_info),
-	.hooks		= (1 << NF_IP6_LOCAL_OUT) | (1 << NF_IP6_POST_ROUTING),
+	.hooks		= (1 << NF_INET_LOCAL_OUT) |
+			  (1 << NF_INET_POST_ROUTING),
 	.checkentry	= checkentry,
 	.me		= THIS_MODULE,
 };
diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c
index 1d26b202bf30..0ae072dd6924 100644
--- a/net/ipv6/netfilter/ip6table_filter.c
+++ b/net/ipv6/netfilter/ip6table_filter.c
@@ -17,7 +17,9 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("ip6tables filter table");
 
-#define FILTER_VALID_HOOKS ((1 << NF_IP6_LOCAL_IN) | (1 << NF_IP6_FORWARD) | (1 << NF_IP6_LOCAL_OUT))
+#define FILTER_VALID_HOOKS ((1 << NF_INET_LOCAL_IN) | \
+			    (1 << NF_INET_FORWARD) | \
+			    (1 << NF_INET_LOCAL_OUT))
 
 static struct
 {
@@ -31,14 +33,14 @@ static struct
 		.num_entries = 4,
 		.size = sizeof(struct ip6t_standard) * 3 + sizeof(struct ip6t_error),
 		.hook_entry = {
-			[NF_IP6_LOCAL_IN] = 0,
-			[NF_IP6_FORWARD] = sizeof(struct ip6t_standard),
-			[NF_IP6_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2
+			[NF_INET_LOCAL_IN] = 0,
+			[NF_INET_FORWARD] = sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2
 		},
 		.underflow = {
-			[NF_IP6_LOCAL_IN] = 0,
-			[NF_IP6_FORWARD] = sizeof(struct ip6t_standard),
-			[NF_IP6_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2
+			[NF_INET_LOCAL_IN] = 0,
+			[NF_INET_FORWARD] = sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2
 		},
 	},
 	.entries = {
@@ -93,21 +95,21 @@ static struct nf_hook_ops ip6t_ops[] = {
 		.hook		= ip6t_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_FILTER,
 	},
 	{
 		.hook		= ip6t_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_FORWARD,
+		.hooknum	= NF_INET_FORWARD,
 		.priority	= NF_IP6_PRI_FILTER,
 	},
 	{
 		.hook		= ip6t_local_out_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_FILTER,
 	},
 };
diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c
index a0b6381f1e8c..8e62b2316829 100644
--- a/net/ipv6/netfilter/ip6table_mangle.c
+++ b/net/ipv6/netfilter/ip6table_mangle.c
@@ -15,11 +15,11 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("ip6tables mangle table");
 
-#define MANGLE_VALID_HOOKS ((1 << NF_IP6_PRE_ROUTING) | \
-			    (1 << NF_IP6_LOCAL_IN) | \
-			    (1 << NF_IP6_FORWARD) | \
-			    (1 << NF_IP6_LOCAL_OUT) | \
-			    (1 << NF_IP6_POST_ROUTING))
+#define MANGLE_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | \
+			    (1 << NF_INET_LOCAL_IN) | \
+			    (1 << NF_INET_FORWARD) | \
+			    (1 << NF_INET_LOCAL_OUT) | \
+			    (1 << NF_INET_POST_ROUTING))
 
 static struct
 {
@@ -33,18 +33,18 @@ static struct
 		.num_entries = 6,
 		.size = sizeof(struct ip6t_standard) * 5 + sizeof(struct ip6t_error),
 		.hook_entry = {
-			[NF_IP6_PRE_ROUTING] 	= 0,
-			[NF_IP6_LOCAL_IN]	= sizeof(struct ip6t_standard),
-			[NF_IP6_FORWARD]	= sizeof(struct ip6t_standard) * 2,
-			[NF_IP6_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 3,
-			[NF_IP6_POST_ROUTING]	= sizeof(struct ip6t_standard) * 4,
+			[NF_INET_PRE_ROUTING] 	= 0,
+			[NF_INET_LOCAL_IN]	= sizeof(struct ip6t_standard),
+			[NF_INET_FORWARD]	= sizeof(struct ip6t_standard) * 2,
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 3,
+			[NF_INET_POST_ROUTING]	= sizeof(struct ip6t_standard) * 4,
 		},
 		.underflow = {
-			[NF_IP6_PRE_ROUTING] 	= 0,
-			[NF_IP6_LOCAL_IN]	= sizeof(struct ip6t_standard),
-			[NF_IP6_FORWARD]	= sizeof(struct ip6t_standard) * 2,
-			[NF_IP6_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 3,
-			[NF_IP6_POST_ROUTING]	= sizeof(struct ip6t_standard) * 4,
+			[NF_INET_PRE_ROUTING] 	= 0,
+			[NF_INET_LOCAL_IN]	= sizeof(struct ip6t_standard),
+			[NF_INET_FORWARD]	= sizeof(struct ip6t_standard) * 2,
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 3,
+			[NF_INET_POST_ROUTING]	= sizeof(struct ip6t_standard) * 4,
 		},
 	},
 	.entries = {
@@ -125,35 +125,35 @@ static struct nf_hook_ops ip6t_ops[] = {
 		.hook		= ip6t_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP6_PRI_MANGLE,
 	},
 	{
 		.hook		= ip6t_local_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_MANGLE,
 	},
 	{
 		.hook		= ip6t_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_FORWARD,
+		.hooknum	= NF_INET_FORWARD,
 		.priority	= NF_IP6_PRI_MANGLE,
 	},
 	{
 		.hook		= ip6t_local_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_MANGLE,
 	},
 	{
 		.hook		= ip6t_route_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP6_PRI_MANGLE,
 	},
 };
diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c
index 8f7109f991e6..4fecd8de8cc2 100644
--- a/net/ipv6/netfilter/ip6table_raw.c
+++ b/net/ipv6/netfilter/ip6table_raw.c
@@ -6,7 +6,7 @@
 #include <linux/module.h>
 #include <linux/netfilter_ipv6/ip6_tables.h>
 
-#define RAW_VALID_HOOKS ((1 << NF_IP6_PRE_ROUTING) | (1 << NF_IP6_LOCAL_OUT))
+#define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT))
 
 static struct
 {
@@ -20,12 +20,12 @@ static struct
 		.num_entries = 3,
 		.size = sizeof(struct ip6t_standard) * 2 + sizeof(struct ip6t_error),
 		.hook_entry = {
-			[NF_IP6_PRE_ROUTING] = 0,
-			[NF_IP6_LOCAL_OUT] = sizeof(struct ip6t_standard)
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard)
 		},
 		.underflow = {
-			[NF_IP6_PRE_ROUTING] = 0,
-			[NF_IP6_LOCAL_OUT] = sizeof(struct ip6t_standard)
+			[NF_INET_PRE_ROUTING] = 0,
+			[NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard)
 		},
 	},
 	.entries = {
@@ -58,14 +58,14 @@ static struct nf_hook_ops ip6t_ops[] = {
 	{
 	  .hook = ip6t_hook,
 	  .pf = PF_INET6,
-	  .hooknum = NF_IP6_PRE_ROUTING,
+	  .hooknum = NF_INET_PRE_ROUTING,
 	  .priority = NF_IP6_PRI_FIRST,
 	  .owner = THIS_MODULE,
 	},
 	{
 	  .hook = ip6t_hook,
 	  .pf = PF_INET6,
-	  .hooknum = NF_IP6_LOCAL_OUT,
+	  .hooknum = NF_INET_LOCAL_OUT,
 	  .priority = NF_IP6_PRI_FIRST,
 	  .owner = THIS_MODULE,
 	},
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index ad74bab05047..50f46787fda4 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -263,42 +263,42 @@ static struct nf_hook_ops ipv6_conntrack_ops[] = {
 		.hook		= ipv6_defrag,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP6_PRI_CONNTRACK_DEFRAG,
 	},
 	{
 		.hook		= ipv6_conntrack_in,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_PRE_ROUTING,
+		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP6_PRI_CONNTRACK,
 	},
 	{
 		.hook		= ipv6_conntrack_local,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_CONNTRACK,
 	},
 	{
 		.hook		= ipv6_defrag,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_OUT,
+		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_CONNTRACK_DEFRAG,
 	},
 	{
 		.hook		= ipv6_confirm,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_POST_ROUTING,
+		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP6_PRI_LAST,
 	},
 	{
 		.hook		= ipv6_confirm,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET6,
-		.hooknum	= NF_IP6_LOCAL_IN,
+		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_LAST-1,
 	},
 };
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index fd9123f3dc04..e99384f9764d 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -192,7 +192,7 @@ icmpv6_error(struct sk_buff *skb, unsigned int dataoff,
 		return -NF_ACCEPT;
 	}
 
-	if (nf_conntrack_checksum && hooknum == NF_IP6_PRE_ROUTING &&
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
 	    nf_ip6_checksum(skb, hooknum, dataoff, IPPROTO_ICMPV6)) {
 		nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL,
 			      "nf_ct_icmpv6: ICMPv6 checksum failed\n");
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index ae314f3fea46..ad622cc11bda 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -619,7 +619,7 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
 		goto error_fault;
 
 	IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS);
-	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
+	err = NF_HOOK(PF_INET6, NF_INET_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
 		      dst_output);
 	if (err > 0)
 		err = np->recverr ? net_xmit_errno(err) : 0;
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index e317d0855468..e2c3efd2579d 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -37,7 +37,7 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async)
 	ipv6_hdr(skb)->payload_len = htons(skb->len);
 	__skb_push(skb, skb->data - skb_network_header(skb));
 
-	NF_HOOK(PF_INET6, NF_IP6_PRE_ROUTING, skb, skb->dev, NULL,
+	NF_HOOK(PF_INET6, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
 		ip6_rcv_finish);
 	return -1;
 #else
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 318669a9cb48..b34c58c65656 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -89,6 +89,6 @@ static int xfrm6_output_finish(struct sk_buff *skb)
 
 int xfrm6_output(struct sk_buff *skb)
 {
-	return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL, skb->dst->dev,
+	return NF_HOOK(PF_INET6, NF_INET_POST_ROUTING, skb, NULL, skb->dst->dev,
 		       xfrm6_output_finish);
 }
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index df7e98d914fa..29e0d25b9e1e 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -188,7 +188,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.proto			= IPPROTO_IPV6,
 	.eth_proto		= htons(ETH_P_IPV6),
-	.nf_post_routing	= NF_IP6_POST_ROUTING,
+	.nf_post_routing	= NF_INET_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 7d231243754a..a15971e9923b 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -829,18 +829,18 @@ ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[])
 						&range) < 0)
 				return -EINVAL;
 			if (nf_nat_initialized(ct,
-					       HOOK2MANIP(NF_IP_PRE_ROUTING)))
+					       HOOK2MANIP(NF_INET_PRE_ROUTING)))
 				return -EEXIST;
-			nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING);
+			nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
 		}
 		if (cda[CTA_NAT_SRC]) {
 			if (nfnetlink_parse_nat(cda[CTA_NAT_SRC], ct,
 						&range) < 0)
 				return -EINVAL;
 			if (nf_nat_initialized(ct,
-					       HOOK2MANIP(NF_IP_POST_ROUTING)))
+					       HOOK2MANIP(NF_INET_POST_ROUTING)))
 				return -EEXIST;
-			nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING);
+			nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
 		}
 #endif
 	}
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 7a3f64c1aca6..d96f18863fd2 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -783,9 +783,7 @@ static int tcp_error(struct sk_buff *skb,
 	 * because the checksum is assumed to be correct.
 	 */
 	/* FIXME: Source route IP option packets --RR */
-	if (nf_conntrack_checksum &&
-	    ((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) ||
-	     (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) &&
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
 	    nf_checksum(skb, hooknum, dataoff, IPPROTO_TCP, pf)) {
 		if (LOG_INVALID(IPPROTO_TCP))
 			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index b3e7ecb080e6..570a2e109478 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -128,9 +128,7 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff,
 	 * We skip checking packets on the outgoing path
 	 * because the checksum is assumed to be correct.
 	 * FIXME: Source route IP option packets --RR */
-	if (nf_conntrack_checksum &&
-	    ((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) ||
-	     (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) &&
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
 	    nf_checksum(skb, hooknum, dataoff, IPPROTO_UDP, pf)) {
 		if (LOG_INVALID(IPPROTO_UDP))
 			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c
index b8981dd922be..7e116d5766d1 100644
--- a/net/netfilter/nf_conntrack_proto_udplite.c
+++ b/net/netfilter/nf_conntrack_proto_udplite.c
@@ -133,8 +133,7 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff,
 
 	/* Checksum invalid? Ignore. */
 	if (nf_conntrack_checksum && !skb_csum_unnecessary(skb) &&
-	    ((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) ||
-	     (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING))) {
+	    hooknum == NF_INET_PRE_ROUTING) {
 		if (pf == PF_INET) {
 			struct iphdr *iph = ip_hdr(skb);
 
diff --git a/net/netfilter/xt_CLASSIFY.c b/net/netfilter/xt_CLASSIFY.c
index 77eeae658d42..e4f7f86d7dd5 100644
--- a/net/netfilter/xt_CLASSIFY.c
+++ b/net/netfilter/xt_CLASSIFY.c
@@ -47,9 +47,9 @@ static struct xt_target xt_classify_target[] __read_mostly = {
 		.target 	= target,
 		.targetsize	= sizeof(struct xt_classify_target_info),
 		.table		= "mangle",
-		.hooks		= (1 << NF_IP_LOCAL_OUT) |
-				  (1 << NF_IP_FORWARD) |
-				  (1 << NF_IP_POST_ROUTING),
+		.hooks		= (1 << NF_INET_LOCAL_OUT) |
+				  (1 << NF_INET_FORWARD) |
+				  (1 << NF_INET_POST_ROUTING),
 		.me 		= THIS_MODULE,
 	},
 	{
@@ -58,9 +58,9 @@ static struct xt_target xt_classify_target[] __read_mostly = {
 		.target 	= target,
 		.targetsize	= sizeof(struct xt_classify_target_info),
 		.table		= "mangle",
-		.hooks		= (1 << NF_IP6_LOCAL_OUT) |
-				  (1 << NF_IP6_FORWARD) |
-				  (1 << NF_IP6_POST_ROUTING),
+		.hooks		= (1 << NF_INET_LOCAL_OUT) |
+				  (1 << NF_INET_FORWARD) |
+				  (1 << NF_INET_POST_ROUTING),
 		.me 		= THIS_MODULE,
 	},
 };
diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c
index 8e76d1f52fbe..f183c8fa47a5 100644
--- a/net/netfilter/xt_TCPMSS.c
+++ b/net/netfilter/xt_TCPMSS.c
@@ -214,9 +214,9 @@ xt_tcpmss_checkentry4(const char *tablename,
 	const struct ipt_entry *e = entry;
 
 	if (info->mss == XT_TCPMSS_CLAMP_PMTU &&
-	    (hook_mask & ~((1 << NF_IP_FORWARD) |
-			   (1 << NF_IP_LOCAL_OUT) |
-			   (1 << NF_IP_POST_ROUTING))) != 0) {
+	    (hook_mask & ~((1 << NF_INET_FORWARD) |
+			   (1 << NF_INET_LOCAL_OUT) |
+			   (1 << NF_INET_POST_ROUTING))) != 0) {
 		printk("xt_TCPMSS: path-MTU clamping only supported in "
 		       "FORWARD, OUTPUT and POSTROUTING hooks\n");
 		return false;
@@ -239,9 +239,9 @@ xt_tcpmss_checkentry6(const char *tablename,
 	const struct ip6t_entry *e = entry;
 
 	if (info->mss == XT_TCPMSS_CLAMP_PMTU &&
-	    (hook_mask & ~((1 << NF_IP6_FORWARD) |
-			   (1 << NF_IP6_LOCAL_OUT) |
-			   (1 << NF_IP6_POST_ROUTING))) != 0) {
+	    (hook_mask & ~((1 << NF_INET_FORWARD) |
+			   (1 << NF_INET_LOCAL_OUT) |
+			   (1 << NF_INET_POST_ROUTING))) != 0) {
 		printk("xt_TCPMSS: path-MTU clamping only supported in "
 		       "FORWARD, OUTPUT and POSTROUTING hooks\n");
 		return false;
diff --git a/net/netfilter/xt_mac.c b/net/netfilter/xt_mac.c
index 00490d777a0f..6ff4479ca638 100644
--- a/net/netfilter/xt_mac.c
+++ b/net/netfilter/xt_mac.c
@@ -50,9 +50,9 @@ static struct xt_match xt_mac_match[] __read_mostly = {
 		.family		= AF_INET,
 		.match		= match,
 		.matchsize	= sizeof(struct xt_mac_info),
-		.hooks		= (1 << NF_IP_PRE_ROUTING) |
-				  (1 << NF_IP_LOCAL_IN) |
-				  (1 << NF_IP_FORWARD),
+		.hooks		= (1 << NF_INET_PRE_ROUTING) |
+				  (1 << NF_INET_LOCAL_IN) |
+				  (1 << NF_INET_FORWARD),
 		.me		= THIS_MODULE,
 	},
 	{
@@ -60,9 +60,9 @@ static struct xt_match xt_mac_match[] __read_mostly = {
 		.family		= AF_INET6,
 		.match		= match,
 		.matchsize	= sizeof(struct xt_mac_info),
-		.hooks		= (1 << NF_IP6_PRE_ROUTING) |
-				  (1 << NF_IP6_LOCAL_IN) |
-				  (1 << NF_IP6_FORWARD),
+		.hooks		= (1 << NF_INET_PRE_ROUTING) |
+				  (1 << NF_INET_LOCAL_IN) |
+				  (1 << NF_INET_FORWARD),
 		.me		= THIS_MODULE,
 	},
 };
diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c
index a4bab043a6d1..e91aee74de5e 100644
--- a/net/netfilter/xt_physdev.c
+++ b/net/netfilter/xt_physdev.c
@@ -113,12 +113,12 @@ checkentry(const char *tablename,
 	if (info->bitmask & XT_PHYSDEV_OP_OUT &&
 	    (!(info->bitmask & XT_PHYSDEV_OP_BRIDGED) ||
 	     info->invert & XT_PHYSDEV_OP_BRIDGED) &&
-	    hook_mask & ((1 << NF_IP_LOCAL_OUT) | (1 << NF_IP_FORWARD) |
-			 (1 << NF_IP_POST_ROUTING))) {
+	    hook_mask & ((1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) |
+			 (1 << NF_INET_POST_ROUTING))) {
 		printk(KERN_WARNING "physdev match: using --physdev-out in the "
 		       "OUTPUT, FORWARD and POSTROUTING chains for non-bridged "
 		       "traffic is not supported anymore.\n");
-		if (hook_mask & (1 << NF_IP_LOCAL_OUT))
+		if (hook_mask & (1 << NF_INET_LOCAL_OUT))
 			return false;
 	}
 	return true;
diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c
index 6d6d3b7fcbb5..2eaa6fd089ce 100644
--- a/net/netfilter/xt_policy.c
+++ b/net/netfilter/xt_policy.c
@@ -144,14 +144,13 @@ static bool checkentry(const char *tablename, const void *ip_void,
 				"outgoing policy selected\n");
 		return false;
 	}
-	/* hook values are equal for IPv4 and IPv6 */
-	if (hook_mask & (1 << NF_IP_PRE_ROUTING | 1 << NF_IP_LOCAL_IN)
+	if (hook_mask & (1 << NF_INET_PRE_ROUTING | 1 << NF_INET_LOCAL_IN)
 	    && info->flags & XT_POLICY_MATCH_OUT) {
 		printk(KERN_ERR "xt_policy: output policy not valid in "
 				"PRE_ROUTING and INPUT\n");
 		return false;
 	}
-	if (hook_mask & (1 << NF_IP_POST_ROUTING | 1 << NF_IP_LOCAL_OUT)
+	if (hook_mask & (1 << NF_INET_POST_ROUTING | 1 << NF_INET_LOCAL_OUT)
 	    && info->flags & XT_POLICY_MATCH_IN) {
 		printk(KERN_ERR "xt_policy: input policy not valid in "
 				"POST_ROUTING and OUTPUT\n");
diff --git a/net/netfilter/xt_realm.c b/net/netfilter/xt_realm.c
index cc3e76d77a99..91113dcbe0f5 100644
--- a/net/netfilter/xt_realm.c
+++ b/net/netfilter/xt_realm.c
@@ -41,8 +41,8 @@ static struct xt_match realm_match __read_mostly = {
 	.name		= "realm",
 	.match		= match,
 	.matchsize	= sizeof(struct xt_realm_info),
-	.hooks		= (1 << NF_IP_POST_ROUTING) | (1 << NF_IP_FORWARD) |
-			  (1 << NF_IP_LOCAL_OUT) | (1 << NF_IP_LOCAL_IN),
+	.hooks		= (1 << NF_INET_POST_ROUTING) | (1 << NF_INET_FORWARD) |
+			  (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_LOCAL_IN),
 	.family		= AF_INET,
 	.me		= THIS_MODULE
 };
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index 3f8335e6ea2e..d377deca4f20 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -235,7 +235,7 @@ static struct nf_hook_ops ing_ops = {
 	.hook           = ing_hook,
 	.owner		= THIS_MODULE,
 	.pf             = PF_INET,
-	.hooknum        = NF_IP_PRE_ROUTING,
+	.hooknum        = NF_INET_PRE_ROUTING,
 	.priority       = NF_IP_PRI_FILTER + 1,
 };
 
@@ -243,7 +243,7 @@ static struct nf_hook_ops ing6_ops = {
 	.hook           = ing_hook,
 	.owner		= THIS_MODULE,
 	.pf             = PF_INET6,
-	.hooknum        = NF_IP6_PRE_ROUTING,
+	.hooknum        = NF_INET_PRE_ROUTING,
 	.priority       = NF_IP6_PRI_FILTER + 1,
 };
 
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 0396354fff95..64d414efb404 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5281,7 +5281,7 @@ static struct nf_hook_ops selinux_ipv4_op = {
 	.hook =		selinux_ipv4_postroute_last,
 	.owner =	THIS_MODULE,
 	.pf =		PF_INET,
-	.hooknum =	NF_IP_POST_ROUTING,
+	.hooknum =	NF_INET_POST_ROUTING,
 	.priority =	NF_IP_PRI_SELINUX_LAST,
 };
 
@@ -5291,7 +5291,7 @@ static struct nf_hook_ops selinux_ipv6_op = {
 	.hook =		selinux_ipv6_postroute_last,
 	.owner =	THIS_MODULE,
 	.pf =		PF_INET6,
-	.hooknum =	NF_IP6_POST_ROUTING,
+	.hooknum =	NF_INET_POST_ROUTING,
 	.priority =	NF_IP6_PRI_SELINUX_LAST,
 };
 
-- 
cgit v1.2.3


From 294b4baf292197e13d1df1d253efa7ac84ffee3f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Wed, 14 Nov 2007 01:57:47 -0800
Subject: [IPSEC]: Kill afinfo->nf_post_routing

After changeset:

	[NETFILTER]: Introduce NF_INET_ hook values

It always evaluates to NF_INET_POST_ROUTING.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 1 -
 net/ipv4/xfrm4_state.c | 1 -
 net/ipv6/xfrm6_state.c | 1 -
 net/xfrm/xfrm_output.c | 2 +-
 4 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 37a086d71baf..ecc02e9bde32 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -259,7 +259,6 @@ struct xfrm_state_afinfo {
 	unsigned int		family;
 	unsigned int		proto;
 	unsigned int		eth_proto;
-	unsigned int		nf_post_routing;
 	struct module		*owner;
 	struct xfrm_type	*type_map[IPPROTO_MAX];
 	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 296113598944..fdeebe68a379 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -66,7 +66,6 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.proto			= IPPROTO_IPIP,
 	.eth_proto		= htons(ETH_P_IP),
-	.nf_post_routing	= NF_INET_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 29e0d25b9e1e..a7a7e8fd6a33 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -188,7 +188,6 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.proto			= IPPROTO_IPV6,
 	.eth_proto		= htons(ETH_P_IPV6),
-	.nf_post_routing	= NF_INET_POST_ROUTING,
 	.owner			= THIS_MODULE,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index 048d240c3e15..3c277a4d0e78 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -113,7 +113,7 @@ int xfrm_output_resume(struct sk_buff *skb, int err)
 			return dst_output(skb);
 
 		err = nf_hook(x->inner_mode->afinfo->family,
-			      x->inner_mode->afinfo->nf_post_routing, skb,
+			      NF_INET_POST_ROUTING, skb,
 			      NULL, skb->dst->dev, xfrm_output2);
 		if (unlikely(err != 1))
 			goto out;
-- 
cgit v1.2.3


From 2a8cc6c89039e0530a3335954253b76ed0f9339a Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 14 Nov 2007 15:56:23 +0900
Subject: [IPV6] ADDRCONF: Support RFC3484 configurable address selection
 policy table.

Policy table is implemented as an RCU linear list since we do not expect
large list nor frequent updates.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_addrlabel.h |  32 +++
 include/linux/rtnetlink.h    |   7 +
 include/net/addrconf.h       |   8 +
 net/ipv6/Makefile            |   1 +
 net/ipv6/addrconf.c          |  40 +---
 net/ipv6/addrlabel.c         | 551 +++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 608 insertions(+), 31 deletions(-)
 create mode 100644 include/linux/if_addrlabel.h
 create mode 100644 net/ipv6/addrlabel.c

(limited to 'include')

diff --git a/include/linux/if_addrlabel.h b/include/linux/if_addrlabel.h
new file mode 100644
index 000000000000..9fe79c95dd28
--- /dev/null
+++ b/include/linux/if_addrlabel.h
@@ -0,0 +1,32 @@
+/*
+ * if_addrlabel.h - netlink interface for address labels
+ *
+ * Copyright (C)2007 USAGI/WIDE Project,  All Rights Reserved.
+ *
+ * Authors:
+ *	YOSHIFUJI Hideaki @ USAGI/WIDE <yoshfuji@linux-ipv6.org>
+ */
+
+#ifndef __LINUX_IF_ADDRLABEL_H
+#define __LINUX_IF_ADDRLABEL_H
+
+struct ifaddrlblmsg
+{
+	__u8		ifal_family;		/* Address family */
+	__u8		__ifal_reserved;	/* Reserved */
+	__u8		ifal_prefixlen;		/* Prefix length */
+	__u8		ifal_flags;		/* Flags */
+	__u32		ifal_index;		/* Link index */
+	__u32		ifal_seq;		/* sequence number */
+};
+
+enum
+{
+	IFAL_ADDRESS = 1,
+	IFAL_LABEL = 2,
+	__IFAL_MAX
+};
+
+#define IFAL_MAX	(__IFAL_MAX - 1)
+
+#endif
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 4e81836191df..e20dcc89a836 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -100,6 +100,13 @@ enum {
 	RTM_NEWNDUSEROPT = 68,
 #define RTM_NEWNDUSEROPT RTM_NEWNDUSEROPT
 
+	RTM_NEWADDRLABEL = 72,
+#define RTM_NEWADDRLABEL RTM_NEWADDRLABEL
+	RTM_DELADDRLABEL,
+#define RTM_NEWADDRLABEL RTM_NEWADDRLABEL
+	RTM_GETADDRLABEL,
+#define RTM_GETADDRLABEL RTM_GETADDRLABEL
+
 	__RTM_MAX,
 #define RTM_MAX		(((__RTM_MAX + 3) & ~3) - 1)
 };
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 33b593e17441..bccc2feb99da 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -83,6 +83,14 @@ extern void			addrconf_join_solict(struct net_device *dev,
 extern void			addrconf_leave_solict(struct inet6_dev *idev,
 					struct in6_addr *addr);
 
+/*
+ *	IPv6 Address Label subsystem (addrlabel.c)
+ */
+extern int			ipv6_addr_label_init(void);
+extern void			ipv6_addr_label_rtnl_register(void);
+extern u32			ipv6_addr_label(const struct in6_addr *addr,
+						int type, int ifindex);
+
 /*
  *	multicast prototypes (mcast.c)
  */
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 87c23a73d284..5ffa98003058 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -5,6 +5,7 @@
 obj-$(CONFIG_IPV6) += ipv6.o
 
 ipv6-objs :=	af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
+		addrlabel.o \
 		route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \
 		raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \
 		exthdrs.o sysctl_net_ipv6.o datagram.o \
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index e1e591bfbdca..a70cecf8fc8d 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -874,36 +874,6 @@ static inline int ipv6_saddr_preferred(int type)
 	return 0;
 }
 
-/* static matching label */
-static inline int ipv6_addr_label(const struct in6_addr *addr, int type,
-				  int ifindex)
-{
- /*
-  * 	prefix (longest match)	label
-  * 	-----------------------------
-  * 	::1/128			0
-  * 	::/0			1
-  * 	2002::/16		2
-  * 	::/96			3
-  * 	::ffff:0:0/96		4
-  *	fc00::/7		5
-  * 	2001::/32		6
-  */
-	if (type & IPV6_ADDR_LOOPBACK)
-		return 0;
-	else if (type & IPV6_ADDR_COMPATv4)
-		return 3;
-	else if (type & IPV6_ADDR_MAPPED)
-		return 4;
-	else if (addr->s6_addr32[0] == htonl(0x20010000))
-		return 6;
-	else if (addr->s6_addr16[0] == htons(0x2002))
-		return 2;
-	else if ((addr->s6_addr[0] & 0xfe) == 0xfc)
-		return 5;
-	return 1;
-}
-
 int ipv6_dev_get_saddr(struct net_device *daddr_dev,
 		       struct in6_addr *daddr, struct in6_addr *saddr)
 {
@@ -4189,7 +4159,13 @@ EXPORT_SYMBOL(unregister_inet6addr_notifier);
 
 int __init addrconf_init(void)
 {
-	int err = 0;
+	int err;
+
+	if ((err = ipv6_addr_label_init()) < 0) {
+		printk(KERN_CRIT "IPv6 Addrconf: cannot initialize default policy table: %d.\n",
+			err);
+		return err;
+	}
 
 	/* The addrconf netdev notifier requires that loopback_dev
 	 * has it's ipv6 private information allocated and setup
@@ -4240,6 +4216,8 @@ int __init addrconf_init(void)
 	__rtnl_register(PF_INET6, RTM_GETMULTICAST, NULL, inet6_dump_ifmcaddr);
 	__rtnl_register(PF_INET6, RTM_GETANYCAST, NULL, inet6_dump_ifacaddr);
 
+	ipv6_addr_label_rtnl_register();
+
 #ifdef CONFIG_SYSCTL
 	addrconf_sysctl.sysctl_header =
 		register_sysctl_table(addrconf_sysctl.addrconf_root_dir);
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
new file mode 100644
index 000000000000..204d4d66834c
--- /dev/null
+++ b/net/ipv6/addrlabel.c
@@ -0,0 +1,551 @@
+/*
+ * IPv6 Address Label subsystem
+ * for the IPv6 "Default" Source Address Selection
+ *
+ * Copyright (C)2007 USAGI/WIDE Project
+ */
+/*
+ * Author:
+ * 	YOSHIFUJI Hideaki @ USAGI/WIDE Project <yoshfuji@linux-ipv6.org>
+ */
+
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/rcupdate.h>
+#include <linux/in6.h>
+#include <net/addrconf.h>
+#include <linux/if_addrlabel.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+
+#if 0
+#define ADDRLABEL(x...) printk(x)
+#else
+#define ADDRLABEL(x...) do { ; } while(0)
+#endif
+
+/*
+ * Policy Table
+ */
+struct ip6addrlbl_entry
+{
+	struct in6_addr prefix;
+	int prefixlen;
+	int ifindex;
+	int addrtype;
+	u32 label;
+	struct hlist_node list;
+	atomic_t refcnt;
+	struct rcu_head rcu;
+};
+
+static struct ip6addrlbl_table
+{
+	struct hlist_head head;
+	spinlock_t lock;
+	u32 seq;
+} ip6addrlbl_table;
+
+/*
+ * Default policy table (RFC3484 + extensions)
+ *
+ * prefix		addr_type	label
+ * -------------------------------------------------------------------------
+ * ::1/128		LOOPBACK	0
+ * ::/0			N/A		1
+ * 2002::/16		N/A		2
+ * ::/96		COMPATv4	3
+ * ::ffff:0:0/96	V4MAPPED	4
+ * fc00::/7		N/A		5		ULA (RFC 4193)
+ * 2001::/32		N/A		6		Teredo (RFC 4380)
+ *
+ * Note: 0xffffffff is used if we do not have any policies.
+ */
+
+#define IPV6_ADDR_LABEL_DEFAULT	0xffffffffUL
+
+static const __initdata struct ip6addrlbl_init_table
+{
+	const struct in6_addr *prefix;
+	int prefixlen;
+	u32 label;
+} ip6addrlbl_init_table[] = {
+	{	/* ::/0 */
+		.prefix = &in6addr_any,
+		.label = 1,
+	},{	/* fc00::/7 */
+		.prefix = &(struct in6_addr){{{ 0xfc }}},
+		.prefixlen = 7,
+		.label = 5,
+	},{	/* 2002::/16 */
+		.prefix = &(struct in6_addr){{{ 0x20, 0x02 }}},
+		.prefixlen = 16,
+		.label = 2,
+	},{	/* 2001::/32 */
+		.prefix = &(struct in6_addr){{{ 0x20, 0x01 }}},
+		.prefixlen = 32,
+		.label = 6,
+	},{	/* ::ffff:0:0 */
+		.prefix = &(struct in6_addr){{{ [10] = 0xff, [11] = 0xff }}},
+		.prefixlen = 96,
+		.label = 4,
+	},{	/* ::/96 */
+		.prefix = &in6addr_any,
+		.prefixlen = 96,
+		.label = 3,
+	},{	/* ::1/128 */
+		.prefix = &in6addr_loopback,
+		.prefixlen = 128,
+		.label = 0,
+	}
+};
+
+/* Object management */
+static inline void ip6addrlbl_free(struct ip6addrlbl_entry *p)
+{
+	kfree(p);
+}
+
+static inline int ip6addrlbl_hold(struct ip6addrlbl_entry *p)
+{
+	return atomic_inc_not_zero(&p->refcnt);
+}
+
+static inline void ip6addrlbl_put(struct ip6addrlbl_entry *p)
+{
+	if (atomic_dec_and_test(&p->refcnt))
+		ip6addrlbl_free(p);
+}
+
+static void ip6addrlbl_free_rcu(struct rcu_head *h)
+{
+	ip6addrlbl_free(container_of(h, struct ip6addrlbl_entry, rcu));
+}
+
+/* Find label */
+static int __ip6addrlbl_match(struct ip6addrlbl_entry *p,
+			      const struct in6_addr *addr,
+			      int addrtype, int ifindex)
+{
+	if (p->ifindex && p->ifindex != ifindex)
+		return 0;
+	if (p->addrtype && p->addrtype != addrtype)
+		return 0;
+	if (!ipv6_prefix_equal(addr, &p->prefix, p->prefixlen))
+		return 0;
+	return 1;
+}
+
+static struct ip6addrlbl_entry *__ipv6_addr_label(const struct in6_addr *addr,
+						  int type, int ifindex)
+{
+	struct hlist_node *pos;
+	struct ip6addrlbl_entry *p;
+	hlist_for_each_entry_rcu(p, pos, &ip6addrlbl_table.head, list) {
+		if (__ip6addrlbl_match(p, addr, type, ifindex))
+			return p;
+	}
+	return NULL;
+}
+
+u32 ipv6_addr_label(const struct in6_addr *addr, int type, int ifindex)
+{
+	u32 label;
+	struct ip6addrlbl_entry *p;
+
+	type &= IPV6_ADDR_MAPPED | IPV6_ADDR_COMPATv4 | IPV6_ADDR_LOOPBACK;
+
+	rcu_read_lock();
+	p = __ipv6_addr_label(addr, type, ifindex);
+	label = p ? p->label : IPV6_ADDR_LABEL_DEFAULT;
+	rcu_read_unlock();
+
+	ADDRLABEL(KERN_DEBUG "%s(addr=" NIP6_FMT ", type=%d, ifindex=%d) => %08x\n",
+			__FUNCTION__,
+			NIP6(*addr), type, ifindex,
+			label);
+
+	return label;
+}
+
+/* allocate one entry */
+struct ip6addrlbl_entry *ip6addrlbl_alloc(const struct in6_addr *prefix,
+					  int prefixlen, int ifindex,
+					  u32 label)
+{
+	struct ip6addrlbl_entry *newp;
+	int addrtype;
+
+	ADDRLABEL(KERN_DEBUG "%s(prefix=" NIP6_FMT ", prefixlen=%d, ifindex=%d, label=%u)\n",
+			__FUNCTION__,
+			NIP6(*prefix), prefixlen,
+			ifindex,
+			(unsigned int)label);
+
+	addrtype = ipv6_addr_type(prefix) & (IPV6_ADDR_MAPPED | IPV6_ADDR_COMPATv4 | IPV6_ADDR_LOOPBACK);
+
+	switch (addrtype) {
+	case IPV6_ADDR_MAPPED:
+		if (prefixlen > 96)
+			return ERR_PTR(-EINVAL);
+		if (prefixlen < 96)
+			addrtype = 0;
+		break;
+	case IPV6_ADDR_COMPATv4:
+		if (prefixlen != 96)
+			addrtype = 0;
+		break;
+	case IPV6_ADDR_LOOPBACK:
+		if (prefixlen != 128)
+			addrtype = 0;
+		break;
+	}
+
+	newp = kmalloc(sizeof(*newp), GFP_KERNEL);
+	if (!newp)
+		return ERR_PTR(-ENOMEM);
+
+	ipv6_addr_prefix(&newp->prefix, prefix, prefixlen);
+	newp->prefixlen = prefixlen;
+	newp->ifindex = ifindex;
+	newp->addrtype = addrtype;
+	newp->label = label;
+	INIT_HLIST_NODE(&newp->list);
+	atomic_set(&newp->refcnt, 1);
+	return newp;
+}
+
+/* add a label */
+int __ip6addrlbl_add(struct ip6addrlbl_entry *newp, int replace)
+{
+	int ret = 0;
+
+	ADDRLABEL(KERN_DEBUG "%s(newp=%p, replace=%d)\n",
+			__FUNCTION__,
+			newp, replace);
+
+	if (hlist_empty(&ip6addrlbl_table.head)) {
+		hlist_add_head_rcu(&newp->list, &ip6addrlbl_table.head);
+	} else {
+		struct hlist_node *pos, *n;
+		struct ip6addrlbl_entry *p = NULL;
+		hlist_for_each_entry_safe(p, pos, n,
+					  &ip6addrlbl_table.head, list) {
+			if (p->prefixlen == newp->prefixlen &&
+			    p->ifindex == newp->ifindex &&
+			    ipv6_addr_equal(&p->prefix, &newp->prefix)) {
+				if (!replace) {
+					ret = -EEXIST;
+					goto out;
+				}
+				hlist_replace_rcu(&p->list, &newp->list);
+				ip6addrlbl_put(p);
+				call_rcu(&p->rcu, ip6addrlbl_free_rcu);
+				goto out;
+			} else if ((p->prefixlen == newp->prefixlen && !p->ifindex) ||
+				   (p->prefixlen < newp->prefixlen)) {
+				hlist_add_before_rcu(&newp->list, &p->list);
+				goto out;
+			}
+		}
+		hlist_add_after_rcu(&p->list, &newp->list);
+	}
+out:
+	if (!ret)
+		ip6addrlbl_table.seq++;
+	return ret;
+}
+
+/* add a label */
+int ip6addrlbl_add(const struct in6_addr *prefix, int prefixlen,
+		       int ifindex, u32 label, int replace)
+{
+	struct ip6addrlbl_entry *newp;
+	int ret = 0;
+
+	ADDRLABEL(KERN_DEBUG "%s(prefix=" NIP6_FMT ", prefixlen=%d, ifindex=%d, label=%u, replace=%d)\n",
+			__FUNCTION__,
+			NIP6(*prefix), prefixlen,
+			ifindex,
+			(unsigned int)label,
+			replace);
+
+	newp = ip6addrlbl_alloc(prefix, prefixlen, ifindex, label);
+	if (IS_ERR(newp))
+		return PTR_ERR(newp);
+	spin_lock(&ip6addrlbl_table.lock);
+	ret = __ip6addrlbl_add(newp, replace);
+	spin_unlock(&ip6addrlbl_table.lock);
+	if (ret)
+		ip6addrlbl_free(newp);
+	return ret;
+}
+
+/* remove a label */
+int __ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
+			  int ifindex)
+{
+	struct ip6addrlbl_entry *p = NULL;
+	struct hlist_node *pos, *n;
+	int ret = -ESRCH;
+
+	ADDRLABEL(KERN_DEBUG "%s(prefix=" NIP6_FMT ", prefixlen=%d, ifindex=%d)\n",
+			__FUNCTION__,
+			NIP6(*prefix), prefixlen,
+			ifindex);
+
+	hlist_for_each_entry_safe(p, pos, n, &ip6addrlbl_table.head, list) {
+		if (p->prefixlen == prefixlen &&
+		    p->ifindex == ifindex &&
+		    ipv6_addr_equal(&p->prefix, prefix)) {
+			hlist_del_rcu(&p->list);
+			ip6addrlbl_put(p);
+			call_rcu(&p->rcu, ip6addrlbl_free_rcu);
+			ret = 0;
+			break;
+		}
+	}
+	return ret;
+}
+
+int ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
+		       int ifindex)
+{
+	struct in6_addr prefix_buf;
+	int ret;
+
+	ADDRLABEL(KERN_DEBUG "%s(prefix=" NIP6_FMT ", prefixlen=%d, ifindex=%d)\n",
+			__FUNCTION__,
+			NIP6(*prefix), prefixlen,
+			ifindex);
+
+	ipv6_addr_prefix(&prefix_buf, prefix, prefixlen);
+	spin_lock(&ip6addrlbl_table.lock);
+	ret = __ip6addrlbl_del(&prefix_buf, prefixlen, ifindex);
+	spin_unlock(&ip6addrlbl_table.lock);
+	return ret;
+}
+
+/* add default label */
+static __init int ip6addrlbl_init(void)
+{
+	int err = 0;
+	int i;
+
+	ADDRLABEL(KERN_DEBUG "%s()\n", __FUNCTION__);
+
+	for (i = 0; i < ARRAY_SIZE(ip6addrlbl_init_table); i++) {
+		int ret = ip6addrlbl_add(ip6addrlbl_init_table[i].prefix,
+					 ip6addrlbl_init_table[i].prefixlen,
+					 0,
+					 ip6addrlbl_init_table[i].label, 0);
+		/* XXX: should we free all rules when we catch an error? */
+		if (ret && (!err || err != -ENOMEM))
+			err = ret;
+	}
+	return err;
+}
+
+int __init ipv6_addr_label_init(void)
+{
+	spin_lock_init(&ip6addrlbl_table.lock);
+
+	return ip6addrlbl_init();
+}
+
+static const struct nla_policy ifal_policy[IFAL_MAX+1] = {
+	[IFAL_ADDRESS]		= { .len = sizeof(struct in6_addr), },
+	[IFAL_LABEL]		= { .len = sizeof(u32), },
+};
+
+static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh,
+			     void *arg)
+{
+	struct ifaddrlblmsg *ifal;
+	struct nlattr *tb[IFAL_MAX+1];
+	struct in6_addr *pfx;
+	u32 label;
+	int err = 0;
+
+	err = nlmsg_parse(nlh, sizeof(*ifal), tb, IFAL_MAX, ifal_policy);
+	if (err < 0)
+		return err;
+
+	ifal = nlmsg_data(nlh);
+
+	if (ifal->ifal_family != AF_INET6 ||
+	    ifal->ifal_prefixlen > 128)
+		return -EINVAL;
+
+	if (ifal->ifal_index &&
+	    !__dev_get_by_index(&init_net, ifal->ifal_index))
+		return -EINVAL;
+
+	if (!tb[IFAL_ADDRESS])
+		return -EINVAL;
+
+	pfx = nla_data(tb[IFAL_ADDRESS]);
+	if (!pfx)
+		return -EINVAL;
+
+	if (!tb[IFAL_LABEL])
+		return -EINVAL;
+	label = nla_get_u32(tb[IFAL_LABEL]);
+	if (label == IPV6_ADDR_LABEL_DEFAULT)
+		return -EINVAL;
+
+	switch(nlh->nlmsg_type) {
+	case RTM_NEWADDRLABEL:
+		err = ip6addrlbl_add(pfx, ifal->ifal_prefixlen,
+				     ifal->ifal_index, label,
+				     nlh->nlmsg_flags & NLM_F_REPLACE);
+		break;
+	case RTM_DELADDRLABEL:
+		err = ip6addrlbl_del(pfx, ifal->ifal_prefixlen,
+				     ifal->ifal_index);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+	}
+	return err;
+}
+
+static inline void ip6addrlbl_putmsg(struct nlmsghdr *nlh,
+				     int prefixlen, int ifindex, u32 lseq)
+{
+	struct ifaddrlblmsg *ifal = nlmsg_data(nlh);
+	ifal->ifal_family = AF_INET6;
+	ifal->ifal_prefixlen = prefixlen;
+	ifal->ifal_flags = 0;
+	ifal->ifal_index = ifindex;
+	ifal->ifal_seq = lseq;
+};
+
+static int ip6addrlbl_fill(struct sk_buff *skb,
+			   struct ip6addrlbl_entry *p,
+			   u32 lseq,
+			   u32 pid, u32 seq, int event,
+			   unsigned int flags)
+{
+	struct nlmsghdr *nlh = nlmsg_put(skb, pid, seq, event,
+					 sizeof(struct ifaddrlblmsg), flags);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	ip6addrlbl_putmsg(nlh, p->prefixlen, p->ifindex, lseq);
+
+	if (nla_put(skb, IFAL_ADDRESS, 16, &p->prefix) < 0 ||
+	    nla_put_u32(skb, IFAL_LABEL, p->label) < 0) {
+		nlmsg_cancel(skb, nlh);
+		return -EMSGSIZE;
+	}
+
+	return nlmsg_end(skb, nlh);
+}
+
+static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct ip6addrlbl_entry *p;
+	struct hlist_node *pos;
+	int idx = 0, s_idx = cb->args[0];
+	int err;
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(p, pos, &ip6addrlbl_table.head, list) {
+		if (idx >= s_idx) {
+			if ((err = ip6addrlbl_fill(skb, p,
+						   ip6addrlbl_table.seq,
+						   NETLINK_CB(cb->skb).pid,
+						   cb->nlh->nlmsg_seq,
+						   RTM_NEWADDRLABEL,
+						   NLM_F_MULTI)) <= 0)
+				break;
+		}
+		idx++;
+	}
+	rcu_read_unlock();
+	cb->args[0] = idx;
+	return skb->len;
+}
+
+static inline int ip6addrlbl_msgsize(void)
+{
+	return (NLMSG_ALIGN(sizeof(struct ifaddrlblmsg))
+		+ nla_total_size(16)	/* IFAL_ADDRESS */
+		+ nla_total_size(4)	/* IFAL_LABEL */
+	);
+}
+
+static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
+			  void *arg)
+{
+	struct ifaddrlblmsg *ifal;
+	struct nlattr *tb[IFAL_MAX+1];
+	struct in6_addr *addr;
+	u32 lseq;
+	int err = 0;
+	struct ip6addrlbl_entry *p;
+	struct sk_buff *skb;
+
+	err = nlmsg_parse(nlh, sizeof(*ifal), tb, IFAL_MAX, ifal_policy);
+	if (err < 0)
+		return err;
+
+	ifal = nlmsg_data(nlh);
+
+	if (ifal->ifal_family != AF_INET6 ||
+	    ifal->ifal_prefixlen != 128)
+		return -EINVAL;
+
+	if (ifal->ifal_index &&
+	    !__dev_get_by_index(&init_net, ifal->ifal_index))
+		return -EINVAL;
+
+	if (!tb[IFAL_ADDRESS])
+		return -EINVAL;
+
+	addr = nla_data(tb[IFAL_ADDRESS]);
+	if (!addr)
+		return -EINVAL;
+
+	rcu_read_lock();
+	p = __ipv6_addr_label(addr, ipv6_addr_type(addr), ifal->ifal_index);
+	if (p && ip6addrlbl_hold(p))
+		p = NULL;
+	lseq = ip6addrlbl_table.seq;
+	rcu_read_unlock();
+
+	if (!p) {
+		err = -ESRCH;
+		goto out;
+	}
+
+	if (!(skb = nlmsg_new(ip6addrlbl_msgsize(), GFP_KERNEL))) {
+		ip6addrlbl_put(p);
+		return -ENOBUFS;
+	}
+
+	err = ip6addrlbl_fill(skb, p, lseq,
+			      NETLINK_CB(in_skb).pid, nlh->nlmsg_seq,
+			      RTM_NEWADDRLABEL, 0);
+
+	ip6addrlbl_put(p);
+
+	if (err < 0) {
+		WARN_ON(err == -EMSGSIZE);
+		kfree_skb(skb);
+		goto out;
+	}
+
+	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+out:
+	return err;
+}
+
+void __init ipv6_addr_label_rtnl_register(void)
+{
+	__rtnl_register(PF_INET6, RTM_NEWADDRLABEL, ip6addrlbl_newdel, NULL);
+	__rtnl_register(PF_INET6, RTM_DELADDRLABEL, ip6addrlbl_newdel, NULL);
+	__rtnl_register(PF_INET6, RTM_GETADDRLABEL, ip6addrlbl_get, ip6addrlbl_dump);
+}
+
-- 
cgit v1.2.3


From 20fea08b5fb639c4c175b5c74a2bb346c5c5bc2e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Wed, 14 Nov 2007 01:44:41 -0800
Subject: [NET]: Move Qdisc_class_ops and Qdisc_ops in appropriate sections.

Qdisc_class_ops are const, and Qdisc_ops are mostly read.

Using "const" and "__read_mostly" qualifiers helps to reduce false
sharing.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h |  2 +-
 net/mac80211/wme.c        |  4 ++--
 net/sched/cls_api.c       |  4 ++--
 net/sched/sch_api.c       | 12 ++++++------
 net/sched/sch_atm.c       |  4 ++--
 net/sched/sch_blackhole.c |  2 +-
 net/sched/sch_cbq.c       |  4 ++--
 net/sched/sch_dsmark.c    |  4 ++--
 net/sched/sch_fifo.c      |  4 ++--
 net/sched/sch_generic.c   | 10 +++++-----
 net/sched/sch_gred.c      |  2 +-
 net/sched/sch_hfsc.c      |  4 ++--
 net/sched/sch_htb.c       |  4 ++--
 net/sched/sch_ingress.c   |  4 ++--
 net/sched/sch_netem.c     |  6 +++---
 net/sched/sch_prio.c      |  6 +++---
 net/sched/sch_red.c       |  4 ++--
 net/sched/sch_sfq.c       |  2 +-
 net/sched/sch_tbf.c       |  4 ++--
 19 files changed, 43 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 4c3b35153c37..9418ac26c986 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -86,7 +86,7 @@ struct Qdisc_class_ops
 struct Qdisc_ops
 {
 	struct Qdisc_ops	*next;
-	struct Qdisc_class_ops	*cl_ops;
+	const struct Qdisc_class_ops	*cl_ops;
 	char			id[IFNAMSIZ];
 	int			priv_size;
 
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 5b8a157975a3..8dbdededdec2 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -527,7 +527,7 @@ static struct tcf_proto ** wme_classop_find_tcf(struct Qdisc *qd,
 
 /* this qdisc is classful (i.e. has classes, some of which may have leaf qdiscs attached)
  * - these are the operations on the classes */
-static struct Qdisc_class_ops class_ops =
+static const struct Qdisc_class_ops class_ops =
 {
 	.graft = wme_classop_graft,
 	.leaf = wme_classop_leaf,
@@ -547,7 +547,7 @@ static struct Qdisc_class_ops class_ops =
 
 
 /* queueing discipline operations */
-static struct Qdisc_ops wme_qdisc_ops =
+static struct Qdisc_ops wme_qdisc_ops __read_mostly =
 {
 	.next = NULL,
 	.cl_ops = &class_ops,
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 03657976fd50..bb98045d5508 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -130,7 +130,7 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 	struct tcf_proto **back, **chain;
 	struct tcf_proto *tp;
 	struct tcf_proto_ops *tp_ops;
-	struct Qdisc_class_ops *cops;
+	const struct Qdisc_class_ops *cops;
 	unsigned long cl;
 	unsigned long fh;
 	int err;
@@ -382,7 +382,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	struct tcf_proto *tp, **chain;
 	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
 	unsigned long cl = 0;
-	struct Qdisc_class_ops *cops;
+	const struct Qdisc_class_ops *cops;
 	struct tcf_dump_args arg;
 
 	if (cb->nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*tcm)))
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 8ae137e3522b..259321be1ad8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -195,7 +195,7 @@ static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
 	struct Qdisc *leaf;
-	struct Qdisc_class_ops *cops = p->ops->cl_ops;
+	const struct Qdisc_class_ops *cops = p->ops->cl_ops;
 
 	if (cops == NULL)
 		return NULL;
@@ -373,7 +373,7 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 
 void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
 {
-	struct Qdisc_class_ops *cops;
+	const struct Qdisc_class_ops *cops;
 	unsigned long cl;
 	u32 parentid;
 
@@ -417,7 +417,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
 			*old = dev_graft_qdisc(dev, new);
 		}
 	} else {
-		struct Qdisc_class_ops *cops = parent->ops->cl_ops;
+		const struct Qdisc_class_ops *cops = parent->ops->cl_ops;
 
 		err = -EINVAL;
 
@@ -581,7 +581,7 @@ static int
 check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w)
 {
 	struct Qdisc *leaf;
-	struct Qdisc_class_ops *cops = q->ops->cl_ops;
+	const struct Qdisc_class_ops *cops = q->ops->cl_ops;
 	struct check_loop_arg *arg = (struct check_loop_arg *)w;
 
 	leaf = cops->leaf(q, cl);
@@ -924,7 +924,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 	struct rtattr **tca = arg;
 	struct net_device *dev;
 	struct Qdisc *q = NULL;
-	struct Qdisc_class_ops *cops;
+	const struct Qdisc_class_ops *cops;
 	unsigned long cl = 0;
 	unsigned long new_cl;
 	u32 pid = tcm->tcm_parent;
@@ -1039,7 +1039,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
 	struct nlmsghdr  *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
 	struct gnet_dump d;
-	struct Qdisc_class_ops *cl_ops = q->ops->cl_ops;
+	const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops;
 
 	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags);
 	tcm = NLMSG_DATA(nlh);
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index ddc4f2c54379..d870a4115d9a 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -668,7 +668,7 @@ static int atm_tc_dump(struct Qdisc *sch, struct sk_buff *skb)
 	return 0;
 }
 
-static struct Qdisc_class_ops atm_class_ops = {
+static const struct Qdisc_class_ops atm_class_ops = {
 	.graft		= atm_tc_graft,
 	.leaf		= atm_tc_leaf,
 	.get		= atm_tc_get,
@@ -683,7 +683,7 @@ static struct Qdisc_class_ops atm_class_ops = {
 	.dump_stats	= atm_tc_dump_class_stats,
 };
 
-static struct Qdisc_ops atm_qdisc_ops = {
+static struct Qdisc_ops atm_qdisc_ops __read_mostly = {
 	.cl_ops		= &atm_class_ops,
 	.id		= "atm",
 	.priv_size	= sizeof(struct atm_qdisc_data),
diff --git a/net/sched/sch_blackhole.c b/net/sched/sch_blackhole.c
index f914fc43a124..507fb488bc98 100644
--- a/net/sched/sch_blackhole.c
+++ b/net/sched/sch_blackhole.c
@@ -28,7 +28,7 @@ static struct sk_buff *blackhole_dequeue(struct Qdisc *sch)
 	return NULL;
 }
 
-static struct Qdisc_ops blackhole_qdisc_ops = {
+static struct Qdisc_ops blackhole_qdisc_ops __read_mostly = {
 	.id		= "blackhole",
 	.priv_size	= 0,
 	.enqueue	= blackhole_enqueue,
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 4de3744e65c3..bea123fc24a4 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -2045,7 +2045,7 @@ static void cbq_walk(struct Qdisc *sch, struct qdisc_walker *arg)
 	}
 }
 
-static struct Qdisc_class_ops cbq_class_ops = {
+static const struct Qdisc_class_ops cbq_class_ops = {
 	.graft		=	cbq_graft,
 	.leaf		=	cbq_leaf,
 	.qlen_notify	=	cbq_qlen_notify,
@@ -2061,7 +2061,7 @@ static struct Qdisc_class_ops cbq_class_ops = {
 	.dump_stats	=	cbq_dump_class_stats,
 };
 
-static struct Qdisc_ops cbq_qdisc_ops = {
+static struct Qdisc_ops cbq_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&cbq_class_ops,
 	.id		=	"cbq",
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 60f89199e3da..b9fe6975fbec 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -461,7 +461,7 @@ rtattr_failure:
 	return RTA_NEST_CANCEL(skb, opts);
 }
 
-static struct Qdisc_class_ops dsmark_class_ops = {
+static const struct Qdisc_class_ops dsmark_class_ops = {
 	.graft		=	dsmark_graft,
 	.leaf		=	dsmark_leaf,
 	.get		=	dsmark_get,
@@ -475,7 +475,7 @@ static struct Qdisc_class_ops dsmark_class_ops = {
 	.dump		=	dsmark_dump_class,
 };
 
-static struct Qdisc_ops dsmark_qdisc_ops = {
+static struct Qdisc_ops dsmark_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&dsmark_class_ops,
 	.id		=	"dsmark",
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index c264308f17c1..d71dbfc790c0 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -78,7 +78,7 @@ rtattr_failure:
 	return -1;
 }
 
-struct Qdisc_ops pfifo_qdisc_ops = {
+struct Qdisc_ops pfifo_qdisc_ops __read_mostly = {
 	.id		=	"pfifo",
 	.priv_size	=	sizeof(struct fifo_sched_data),
 	.enqueue	=	pfifo_enqueue,
@@ -92,7 +92,7 @@ struct Qdisc_ops pfifo_qdisc_ops = {
 	.owner		=	THIS_MODULE,
 };
 
-struct Qdisc_ops bfifo_qdisc_ops = {
+struct Qdisc_ops bfifo_qdisc_ops __read_mostly = {
 	.id		=	"bfifo",
 	.priv_size	=	sizeof(struct fifo_sched_data),
 	.enqueue	=	bfifo_enqueue,
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 84c048a54799..9be2f152455a 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -287,7 +287,7 @@ static int noop_requeue(struct sk_buff *skb, struct Qdisc* qdisc)
 	return NET_XMIT_CN;
 }
 
-struct Qdisc_ops noop_qdisc_ops = {
+struct Qdisc_ops noop_qdisc_ops __read_mostly = {
 	.id		=	"noop",
 	.priv_size	=	0,
 	.enqueue	=	noop_enqueue,
@@ -304,7 +304,7 @@ struct Qdisc noop_qdisc = {
 	.list		=	LIST_HEAD_INIT(noop_qdisc.list),
 };
 
-static struct Qdisc_ops noqueue_qdisc_ops = {
+static struct Qdisc_ops noqueue_qdisc_ops __read_mostly = {
 	.id		=	"noqueue",
 	.priv_size	=	0,
 	.enqueue	=	noop_enqueue,
@@ -406,7 +406,7 @@ static int pfifo_fast_init(struct Qdisc *qdisc, struct rtattr *opt)
 	return 0;
 }
 
-static struct Qdisc_ops pfifo_fast_ops = {
+static struct Qdisc_ops pfifo_fast_ops __read_mostly = {
 	.id		=	"pfifo_fast",
 	.priv_size	=	PFIFO_FAST_BANDS * sizeof(struct sk_buff_head),
 	.enqueue	=	pfifo_fast_enqueue,
@@ -472,7 +472,7 @@ errout:
 
 void qdisc_reset(struct Qdisc *qdisc)
 {
-	struct Qdisc_ops *ops = qdisc->ops;
+	const struct Qdisc_ops *ops = qdisc->ops;
 
 	if (ops->reset)
 		ops->reset(qdisc);
@@ -491,7 +491,7 @@ static void __qdisc_destroy(struct rcu_head *head)
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
-	struct Qdisc_ops  *ops = qdisc->ops;
+	const struct Qdisc_ops  *ops = qdisc->ops;
 
 	if (qdisc->flags & TCQ_F_BUILTIN ||
 	    !atomic_dec_and_test(&qdisc->refcnt))
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index 3cc6dda02e2e..e2bcd6682c70 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -577,7 +577,7 @@ static void gred_destroy(struct Qdisc *sch)
 	}
 }
 
-static struct Qdisc_ops gred_qdisc_ops = {
+static struct Qdisc_ops gred_qdisc_ops __read_mostly = {
 	.id		=	"gred",
 	.priv_size	=	sizeof(struct gred_sched),
 	.enqueue	=	gred_enqueue,
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index a6ad491e434b..69dc3bccf024 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1698,7 +1698,7 @@ hfsc_drop(struct Qdisc *sch)
 	return 0;
 }
 
-static struct Qdisc_class_ops hfsc_class_ops = {
+static const struct Qdisc_class_ops hfsc_class_ops = {
 	.change		= hfsc_change_class,
 	.delete		= hfsc_delete_class,
 	.graft		= hfsc_graft_class,
@@ -1714,7 +1714,7 @@ static struct Qdisc_class_ops hfsc_class_ops = {
 	.walk		= hfsc_walk
 };
 
-static struct Qdisc_ops hfsc_qdisc_ops = {
+static struct Qdisc_ops hfsc_qdisc_ops __read_mostly = {
 	.id		= "hfsc",
 	.init		= hfsc_init_qdisc,
 	.change		= hfsc_change_qdisc,
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 5e608a64935a..72beb66cec9c 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1529,7 +1529,7 @@ static void htb_walk(struct Qdisc *sch, struct qdisc_walker *arg)
 	}
 }
 
-static struct Qdisc_class_ops htb_class_ops = {
+static const struct Qdisc_class_ops htb_class_ops = {
 	.graft		=	htb_graft,
 	.leaf		=	htb_leaf,
 	.qlen_notify	=	htb_qlen_notify,
@@ -1545,7 +1545,7 @@ static struct Qdisc_class_ops htb_class_ops = {
 	.dump_stats	=	htb_dump_class_stats,
 };
 
-static struct Qdisc_ops htb_qdisc_ops = {
+static struct Qdisc_ops htb_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&htb_class_ops,
 	.id		=	"htb",
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index d377deca4f20..902d82ea764b 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -338,7 +338,7 @@ rtattr_failure:
 	return -1;
 }
 
-static struct Qdisc_class_ops ingress_class_ops = {
+static const struct Qdisc_class_ops ingress_class_ops = {
 	.graft		=	ingress_graft,
 	.leaf		=	ingress_leaf,
 	.get		=	ingress_get,
@@ -352,7 +352,7 @@ static struct Qdisc_class_ops ingress_class_ops = {
 	.dump		=	NULL,
 };
 
-static struct Qdisc_ops ingress_qdisc_ops = {
+static struct Qdisc_ops ingress_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&ingress_class_ops,
 	.id		=	"ingress",
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 9e5e87e81f00..6c344ade33c2 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -544,7 +544,7 @@ rtattr_failure:
 	return -1;
 }
 
-static struct Qdisc_ops tfifo_qdisc_ops = {
+static struct Qdisc_ops tfifo_qdisc_ops __read_mostly = {
 	.id		=	"tfifo",
 	.priv_size	=	sizeof(struct fifo_sched_data),
 	.enqueue	=	tfifo_enqueue,
@@ -705,7 +705,7 @@ static struct tcf_proto **netem_find_tcf(struct Qdisc *sch, unsigned long cl)
 	return NULL;
 }
 
-static struct Qdisc_class_ops netem_class_ops = {
+static const struct Qdisc_class_ops netem_class_ops = {
 	.graft		=	netem_graft,
 	.leaf		=	netem_leaf,
 	.get		=	netem_get,
@@ -717,7 +717,7 @@ static struct Qdisc_class_ops netem_class_ops = {
 	.dump		=	netem_dump_class,
 };
 
-static struct Qdisc_ops netem_qdisc_ops = {
+static struct Qdisc_ops netem_qdisc_ops __read_mostly = {
 	.id		=	"netem",
 	.cl_ops		=	&netem_class_ops,
 	.priv_size	=	sizeof(struct netem_sched_data),
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index de894096e442..2243aaa8d851 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -468,7 +468,7 @@ static struct tcf_proto ** prio_find_tcf(struct Qdisc *sch, unsigned long cl)
 	return &q->filter_list;
 }
 
-static struct Qdisc_class_ops prio_class_ops = {
+static const struct Qdisc_class_ops prio_class_ops = {
 	.graft		=	prio_graft,
 	.leaf		=	prio_leaf,
 	.get		=	prio_get,
@@ -483,7 +483,7 @@ static struct Qdisc_class_ops prio_class_ops = {
 	.dump_stats	=	prio_dump_class_stats,
 };
 
-static struct Qdisc_ops prio_qdisc_ops = {
+static struct Qdisc_ops prio_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&prio_class_ops,
 	.id		=	"prio",
@@ -500,7 +500,7 @@ static struct Qdisc_ops prio_qdisc_ops = {
 	.owner		=	THIS_MODULE,
 };
 
-static struct Qdisc_ops rr_qdisc_ops = {
+static struct Qdisc_ops rr_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&prio_class_ops,
 	.id		=	"rr",
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 9b95fefb70f4..f1e9647f7db7 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -359,7 +359,7 @@ static struct tcf_proto **red_find_tcf(struct Qdisc *sch, unsigned long cl)
 	return NULL;
 }
 
-static struct Qdisc_class_ops red_class_ops = {
+static const struct Qdisc_class_ops red_class_ops = {
 	.graft		=	red_graft,
 	.leaf		=	red_leaf,
 	.get		=	red_get,
@@ -371,7 +371,7 @@ static struct Qdisc_class_ops red_class_ops = {
 	.dump		=	red_dump_class,
 };
 
-static struct Qdisc_ops red_qdisc_ops = {
+static struct Qdisc_ops red_qdisc_ops __read_mostly = {
 	.id		=	"red",
 	.priv_size	=	sizeof(struct red_sched_data),
 	.cl_ops		=	&red_class_ops,
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 65293876cd66..c58fa6efc7a2 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -480,7 +480,7 @@ rtattr_failure:
 	return -1;
 }
 
-static struct Qdisc_ops sfq_qdisc_ops = {
+static struct Qdisc_ops sfq_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	NULL,
 	.id		=	"sfq",
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index b0d81098b0ee..d88fea9d6b61 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -469,7 +469,7 @@ static struct tcf_proto **tbf_find_tcf(struct Qdisc *sch, unsigned long cl)
 	return NULL;
 }
 
-static struct Qdisc_class_ops tbf_class_ops =
+static const struct Qdisc_class_ops tbf_class_ops =
 {
 	.graft		=	tbf_graft,
 	.leaf		=	tbf_leaf,
@@ -482,7 +482,7 @@ static struct Qdisc_class_ops tbf_class_ops =
 	.dump		=	tbf_dump_class,
 };
 
-static struct Qdisc_ops tbf_qdisc_ops = {
+static struct Qdisc_ops tbf_qdisc_ops __read_mostly = {
 	.next		=	NULL,
 	.cl_ops		=	&tbf_class_ops,
 	.id		=	"tbf",
-- 
cgit v1.2.3


From c0ef877b2c9f543e9fb7953bfe1a0cd3a4eae362 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 15 Nov 2007 03:03:19 -0800
Subject: [NET]: Move sock_valbool_flag to socket.c

The sock_valbool_flag() helper is used in setsockopt to
set or reset some flag on the sock. This helper is required
in the net/socket.c only, so move it there.

Besides, patch two places in sys_setsockopt() that repeat
this helper functionality manually.

Since this is not a bugfix, but a trivial cleanup, I
prepared this patch against net-2.6.25, but it also
applies (with a single offset) to the latest net-2.6.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h |  8 --------
 net/core/sock.c    | 20 +++++++++++---------
 2 files changed, 11 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index f5b643714131..9c55af8e5f81 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1396,14 +1396,6 @@ extern int net_msg_warn;
 				lock_sock(sk); \
 				}
 
-static inline void sock_valbool_flag(struct sock *sk, int bit, int valbool)
-{
-	if (valbool)
-		sock_set_flag(sk, bit);
-	else
-		sock_reset_flag(sk, bit);
-}
-
 extern __u32 sysctl_wmem_max;
 extern __u32 sysctl_rmem_max;
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 2029d095b4c5..98b243a5b323 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -419,6 +419,14 @@ out:
 	return ret;
 }
 
+static inline void sock_valbool_flag(struct sock *sk, int bit, int valbool)
+{
+	if (valbool)
+		sock_set_flag(sk, bit);
+	else
+		sock_reset_flag(sk, bit);
+}
+
 /*
  *	This is meant for all protocols to use and covers goings on
  *	at the socket level. Everything here is generic.
@@ -463,11 +471,8 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 	case SO_DEBUG:
 		if (val && !capable(CAP_NET_ADMIN)) {
 			ret = -EACCES;
-		}
-		else if (valbool)
-			sock_set_flag(sk, SOCK_DBG);
-		else
-			sock_reset_flag(sk, SOCK_DBG);
+		} else
+			sock_valbool_flag(sk, SOCK_DBG, valbool);
 		break;
 	case SO_REUSEADDR:
 		sk->sk_reuse = valbool;
@@ -477,10 +482,7 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 		ret = -ENOPROTOOPT;
 		break;
 	case SO_DONTROUTE:
-		if (valbool)
-			sock_set_flag(sk, SOCK_LOCALROUTE);
-		else
-			sock_reset_flag(sk, SOCK_LOCALROUTE);
+		sock_valbool_flag(sk, SOCK_LOCALROUTE, valbool);
 		break;
 	case SO_BROADCAST:
 		sock_valbool_flag(sk, SOCK_BROADCAST, valbool);
-- 
cgit v1.2.3


From a47e5a988a575e64c8c9bae65a1dfe3dca7cba32 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 15 Nov 2007 19:41:46 -0800
Subject: [TCP]: Convert highest_sack to sk_buff to allow direct access
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It is going to replace the sack fastpath hint quite soon... :-)

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h   |  6 ++++--
 include/net/tcp.h     | 10 ++++++++++
 net/ipv4/tcp_input.c  | 12 ++++++------
 net/ipv4/tcp_output.c | 19 ++++++++++---------
 4 files changed, 30 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index bac17c59b24e..34acee662230 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -332,8 +332,10 @@ struct tcp_sock {
 
 	struct tcp_sack_block_wire recv_sack_cache[4];
 
-	u32	highest_sack;	/* Start seq of globally highest revd SACK
-				 * (validity guaranteed only if sacked_out > 0) */
+	struct sk_buff *highest_sack;   /* highest skb with SACK received
+					 * (validity guaranteed only if
+					 * sacked_out > 0)
+					 */
 
 	/* from STCP, retrans queue hinting */
 	struct sk_buff* lost_skb_hint;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index d893b4480769..0ede804b16d4 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1312,6 +1312,16 @@ static inline int tcp_write_queue_empty(struct sock *sk)
 	return skb_queue_empty(&sk->sk_write_queue);
 }
 
+/* Start sequence of the highest skb with SACKed bit, valid only if
+ * sacked > 0 or when the caller has ensured validity by itself.
+ */
+static inline u32 tcp_highest_sack_seq(struct tcp_sock *tp)
+{
+	if (!tp->sacked_out)
+		return tp->snd_una;
+	return TCP_SKB_CB(tp->highest_sack)->seq;
+}
+
 /* /proc */
 enum tcp_seq_states {
 	TCP_SEQ_STATE_LISTENING,
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index c0e8f2b1fa7e..31294b52ad42 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1245,7 +1245,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
 	int reord = tp->packets_out;
 	int prior_fackets;
-	u32 highest_sack_end_seq = tp->lost_retrans_low;
+	u32 highest_sack_end_seq;
 	int flag = 0;
 	int found_dup_sack = 0;
 	int cached_fack_count;
@@ -1256,7 +1256,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	if (!tp->sacked_out) {
 		if (WARN_ON(tp->fackets_out))
 			tp->fackets_out = 0;
-		tp->highest_sack = tp->snd_una;
+		tp->highest_sack = tcp_write_queue_head(sk);
 	}
 	prior_fackets = tp->fackets_out;
 
@@ -1483,10 +1483,9 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 				if (fack_count > tp->fackets_out)
 					tp->fackets_out = fack_count;
 
-				if (after(TCP_SKB_CB(skb)->seq, tp->highest_sack)) {
-					tp->highest_sack = TCP_SKB_CB(skb)->seq;
-					highest_sack_end_seq = TCP_SKB_CB(skb)->end_seq;
-				}
+				if (after(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
+					tp->highest_sack = skb;
+
 			} else {
 				if (dup_sack && (sacked&TCPCB_RETRANS))
 					reord = min(fack_count, reord);
@@ -1514,6 +1513,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 			flag &= ~FLAG_ONLY_ORIG_SACKED;
 	}
 
+	highest_sack_end_seq = TCP_SKB_CB(tp->highest_sack)->end_seq;
 	if (tcp_is_fack(tp) && tp->retrans_out &&
 	    after(highest_sack_end_seq, tp->lost_retrans_low) &&
 	    icsk->icsk_ca_state == TCP_CA_Recovery)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index f4c1eef89af0..ce506af5ce07 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -657,13 +657,15 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned
  * tweak SACK fastpath hint too as it would overwrite all changes unless
  * hint is also changed.
  */
-static void tcp_adjust_fackets_out(struct tcp_sock *tp, struct sk_buff *skb,
+static void tcp_adjust_fackets_out(struct sock *sk, struct sk_buff *skb,
 				   int decr)
 {
+	struct tcp_sock *tp = tcp_sk(sk);
+
 	if (!tp->sacked_out || tcp_is_reno(tp))
 		return;
 
-	if (!before(tp->highest_sack, TCP_SKB_CB(skb)->seq))
+	if (!before(tcp_highest_sack_seq(tp), TCP_SKB_CB(skb)->seq))
 		tp->fackets_out -= decr;
 
 	/* cnt_hint is "off-by-one" compared with fackets_out (see sacktag) */
@@ -712,9 +714,8 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 	TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq;
 	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq;
 
-	if (tcp_is_sack(tp) && tp->sacked_out &&
-	    (TCP_SKB_CB(skb)->seq == tp->highest_sack))
-		tp->highest_sack = TCP_SKB_CB(buff)->seq;
+	if (tcp_is_sack(tp) && tp->sacked_out && (skb == tp->highest_sack))
+		tp->highest_sack = buff;
 
 	/* PSH and FIN should only be set in the second packet. */
 	flags = TCP_SKB_CB(skb)->flags;
@@ -772,7 +773,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 			tcp_dec_pcount_approx_int(&tp->sacked_out, diff);
 			tcp_verify_left_out(tp);
 		}
-		tcp_adjust_fackets_out(tp, skb, diff);
+		tcp_adjust_fackets_out(sk, skb, diff);
 	}
 
 	/* Link BUFF into the send queue. */
@@ -1712,7 +1713,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 		       tcp_skb_pcount(next_skb) != 1);
 
 		if (WARN_ON(tcp_is_sack(tp) && tp->sacked_out &&
-		    (TCP_SKB_CB(next_skb)->seq == tp->highest_sack)))
+		    (next_skb == tp->highest_sack)))
 			return;
 
 		/* Ok.	We will be able to collapse the packet. */
@@ -1747,7 +1748,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 		if (tcp_is_reno(tp) && tp->sacked_out)
 			tcp_dec_pcount_approx(&tp->sacked_out, next_skb);
 
-		tcp_adjust_fackets_out(tp, next_skb, tcp_skb_pcount(next_skb));
+		tcp_adjust_fackets_out(sk, next_skb, tcp_skb_pcount(next_skb));
 		tp->packets_out -= tcp_skb_pcount(next_skb);
 
 		/* changed transmit queue under us so clear hints */
@@ -2028,7 +2029,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk)
 			break;
 		tp->forward_skb_hint = skb;
 
-		if (after(TCP_SKB_CB(skb)->seq, tp->highest_sack))
+		if (after(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
 			break;
 
 		if (tcp_packets_in_flight(tp) >= tp->snd_cwnd)
-- 
cgit v1.2.3


From fd6dad616d4fe2f08d690f25ca76b0102158fb3a Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 15 Nov 2007 19:49:47 -0800
Subject: [TCP]: Earlier SACK block verification & simplify access to them
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h  |  2 +-
 net/ipv4/tcp_input.c | 85 +++++++++++++++++++++++++++++++---------------------
 2 files changed, 52 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 34acee662230..794497c7d755 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -330,7 +330,7 @@ struct tcp_sock {
 	struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */
 	struct tcp_sack_block selective_acks[4]; /* The SACKS themselves*/
 
-	struct tcp_sack_block_wire recv_sack_cache[4];
+	struct tcp_sack_block recv_sack_cache[4];
 
 	struct sk_buff *highest_sack;   /* highest skb with SACK received
 					 * (validity guaranteed only if
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a62e0f90f566..a287747e9dd6 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1340,9 +1340,11 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned char *ptr = (skb_transport_header(ack_skb) +
 			      TCP_SKB_CB(ack_skb)->sacked);
-	struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2);
+	struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2);
+	struct tcp_sack_block sp[4];
 	struct sk_buff *cached_skb;
 	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
+	int used_sacks;
 	int reord = tp->packets_out;
 	int flag = 0;
 	int found_dup_sack = 0;
@@ -1357,7 +1359,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 		tp->highest_sack = tcp_write_queue_head(sk);
 	}
 
-	found_dup_sack = tcp_check_dsack(tp, ack_skb, sp,
+	found_dup_sack = tcp_check_dsack(tp, ack_skb, sp_wire,
 					 num_sacks, prior_snd_una);
 	if (found_dup_sack)
 		flag |= FLAG_DSACKING_ACK;
@@ -1372,14 +1374,49 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	if (!tp->packets_out)
 		goto out;
 
+	used_sacks = 0;
+	first_sack_index = 0;
+	for (i = 0; i < num_sacks; i++) {
+		int dup_sack = !i && found_dup_sack;
+
+		sp[used_sacks].start_seq = ntohl(get_unaligned(&sp_wire[i].start_seq));
+		sp[used_sacks].end_seq = ntohl(get_unaligned(&sp_wire[i].end_seq));
+
+		if (!tcp_is_sackblock_valid(tp, dup_sack,
+					    sp[used_sacks].start_seq,
+					    sp[used_sacks].end_seq)) {
+			if (dup_sack) {
+				if (!tp->undo_marker)
+					NET_INC_STATS_BH(LINUX_MIB_TCPDSACKIGNOREDNOUNDO);
+				else
+					NET_INC_STATS_BH(LINUX_MIB_TCPDSACKIGNOREDOLD);
+			} else {
+				/* Don't count olds caused by ACK reordering */
+				if ((TCP_SKB_CB(ack_skb)->ack_seq != tp->snd_una) &&
+				    !after(sp[used_sacks].end_seq, tp->snd_una))
+					continue;
+				NET_INC_STATS_BH(LINUX_MIB_TCPSACKDISCARD);
+			}
+			if (i == 0)
+				first_sack_index = -1;
+			continue;
+		}
+
+		/* Ignore very old stuff early */
+		if (!after(sp[used_sacks].end_seq, prior_snd_una))
+			continue;
+
+		used_sacks++;
+	}
+
 	/* SACK fastpath:
 	 * if the only SACK change is the increase of the end_seq of
 	 * the first block then only apply that SACK block
 	 * and use retrans queue hinting otherwise slowpath */
 	force_one_sack = 1;
-	for (i = 0; i < num_sacks; i++) {
-		__be32 start_seq = sp[i].start_seq;
-		__be32 end_seq = sp[i].end_seq;
+	for (i = 0; i < used_sacks; i++) {
+		u32 start_seq = sp[i].start_seq;
+		u32 end_seq = sp[i].end_seq;
 
 		if (i == 0) {
 			if (tp->recv_sack_cache[i].start_seq != start_seq)
@@ -1398,19 +1435,17 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 		tp->recv_sack_cache[i].end_seq = 0;
 	}
 
-	first_sack_index = 0;
 	if (force_one_sack)
-		num_sacks = 1;
+		used_sacks = 1;
 	else {
 		int j;
 		tp->fastpath_skb_hint = NULL;
 
 		/* order SACK blocks to allow in order walk of the retrans queue */
-		for (i = num_sacks-1; i > 0; i--) {
+		for (i = used_sacks - 1; i > 0; i--) {
 			for (j = 0; j < i; j++){
-				if (after(ntohl(sp[j].start_seq),
-					  ntohl(sp[j+1].start_seq))){
-					struct tcp_sack_block_wire tmp;
+				if (after(sp[j].start_seq, sp[j+1].start_seq)) {
+					struct tcp_sack_block tmp;
 
 					tmp = sp[j];
 					sp[j] = sp[j+1];
@@ -1433,32 +1468,14 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 		cached_fack_count = 0;
 	}
 
-	for (i = 0; i < num_sacks; i++) {
+	for (i = 0; i < used_sacks; i++) {
 		struct sk_buff *skb;
-		__u32 start_seq = ntohl(sp->start_seq);
-		__u32 end_seq = ntohl(sp->end_seq);
+		u32 start_seq = sp[i].start_seq;
+		u32 end_seq = sp[i].end_seq;
 		int fack_count;
 		int dup_sack = (found_dup_sack && (i == first_sack_index));
 		int next_dup = (found_dup_sack && (i+1 == first_sack_index));
 
-		sp++;
-
-		if (!tcp_is_sackblock_valid(tp, dup_sack, start_seq, end_seq)) {
-			if (dup_sack) {
-				if (!tp->undo_marker)
-					NET_INC_STATS_BH(LINUX_MIB_TCPDSACKIGNOREDNOUNDO);
-				else
-					NET_INC_STATS_BH(LINUX_MIB_TCPDSACKIGNOREDOLD);
-			} else {
-				/* Don't count olds caused by ACK reordering */
-				if ((TCP_SKB_CB(ack_skb)->ack_seq != tp->snd_una) &&
-				    !after(end_seq, tp->snd_una))
-					continue;
-				NET_INC_STATS_BH(LINUX_MIB_TCPSACKDISCARD);
-			}
-			continue;
-		}
-
 		skb = cached_skb;
 		fack_count = cached_fack_count;
 
@@ -1489,8 +1506,8 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 
 			/* Due to sorting DSACK may reside within this SACK block! */
 			if (next_dup) {
-				u32 dup_start = ntohl(sp->start_seq);
-				u32 dup_end = ntohl(sp->end_seq);
+				u32 dup_start = sp[i+1].start_seq;
+				u32 dup_end = sp[i+1].end_seq;
 
 				if (before(TCP_SKB_CB(skb)->seq, dup_end)) {
 					in_sack = tcp_match_skb_to_sack(sk, skb, dup_start, dup_end);
-- 
cgit v1.2.3


From 68f8353b480e5f2e136c38a511abdbb88eaa8ce2 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 15 Nov 2007 19:50:37 -0800
Subject: [TCP]: Rewrite SACK block processing & sack_recv_cache use
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Key points of this patch are:

  - In case new SACK information is advance only type, no skb
    processing below previously discovered highest point is done
  - Optimize cases below highest point too since there's no need
    to always go up to highest point (which is very likely still
    present in that SACK), this is not entirely true though
    because I'm dropping the fastpath_skb_hint which could
    previously optimize those cases even better. Whether that's
    significant, I'm not too sure.

Currently it will provide skipping by walking. Combined with
RB-tree, all skipping would become fast too regardless of window
size (can be done incrementally later).

Previously a number of cases in TCP SACK processing fails to
take advantage of costly stored information in sack_recv_cache,
most importantly, expected events such as cumulative ACK and new
hole ACKs. Processing on such ACKs result in rather long walks
building up latencies (which easily gets nasty when window is
huge). Those latencies are often completely unnecessary
compared with the amount of _new_ information received, usually
for cumulative ACK there's no new information at all, yet TCP
walks whole queue unnecessary potentially taking a number of
costly cache misses on the way, etc.!

Since the inclusion of highest_sack, there's a lot information
that is very likely redundant (SACK fastpath hint stuff,
fackets_out, highest_sack), though there's no ultimate guarantee
that they'll remain the same whole the time (in all unearthly
scenarios). Take advantage of this knowledge here and drop
fastpath hint and use direct access to highest SACKed skb as
a replacement.

Effectively "special cased" fastpath is dropped. This change
adds some complexity to introduce better coveraged "fastpath",
though the added complexity should make TCP behave more cache
friendly.

The current ACK's SACK blocks are compared against each cached
block individially and only ranges that are new are then scanned
by the high constant walk. For other parts of write queue, even
when in previously known part of the SACK blocks, a faster skip
function is used (if necessary at all). In addition, whenever
possible, TCP fast-forwards to highest_sack skb that was made
available by an earlier patch. In typical case, no other things
but this fast-forward and mandatory markings after that occur
making the access pattern quite similar to the former fastpath
"special case".

DSACKs are special case that must always be walked.

The local to recv_sack_cache copying could be more intelligent
w.r.t DSACKs which are likely to be there only once but that
is left to a separate patch.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h   |   3 -
 include/net/tcp.h     |   1 -
 net/ipv4/tcp_input.c  | 271 +++++++++++++++++++++++++++++++-------------------
 net/ipv4/tcp_output.c |  14 +--
 4 files changed, 172 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 794497c7d755..08027f1d7f31 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -343,10 +343,7 @@ struct tcp_sock {
 	struct sk_buff *scoreboard_skb_hint;
 	struct sk_buff *retransmit_skb_hint;
 	struct sk_buff *forward_skb_hint;
-	struct sk_buff *fastpath_skb_hint;
 
-	int     fastpath_cnt_hint;	/* Lags behind by current skb's pcount
-					 * compared to respective fackets_out */
 	int     lost_cnt_hint;
 	int     retransmit_cnt_hint;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 0ede804b16d4..f0c5e7a29400 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1081,7 +1081,6 @@ static inline void tcp_clear_retrans_hints_partial(struct tcp_sock *tp)
 static inline void tcp_clear_all_retrans_hints(struct tcp_sock *tp)
 {
 	tcp_clear_retrans_hints_partial(tp);
-	tp->fastpath_skb_hint = NULL;
 }
 
 /* MD5 Signature */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a287747e9dd6..3ad6a19ad30f 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1333,6 +1333,88 @@ static int tcp_sacktag_one(struct sk_buff *skb, struct tcp_sock *tp,
 	return flag;
 }
 
+static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk,
+					struct tcp_sack_block *next_dup,
+					u32 start_seq, u32 end_seq,
+					int dup_sack_in, int *fack_count,
+					int *reord, int *flag)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	tcp_for_write_queue_from(skb, sk) {
+		int in_sack = 0;
+		int dup_sack = dup_sack_in;
+
+		if (skb == tcp_send_head(sk))
+			break;
+
+		/* queue is in-order => we can short-circuit the walk early */
+		if (!before(TCP_SKB_CB(skb)->seq, end_seq))
+			break;
+
+		if ((next_dup != NULL) &&
+		    before(TCP_SKB_CB(skb)->seq, next_dup->end_seq)) {
+			in_sack = tcp_match_skb_to_sack(sk, skb,
+							next_dup->start_seq,
+							next_dup->end_seq);
+			if (in_sack > 0)
+				dup_sack = 1;
+		}
+
+		if (in_sack <= 0)
+			in_sack = tcp_match_skb_to_sack(sk, skb, start_seq, end_seq);
+		if (unlikely(in_sack < 0))
+			break;
+
+		if (in_sack)
+			*flag |= tcp_sacktag_one(skb, tp, reord, dup_sack, *fack_count);
+
+		*fack_count += tcp_skb_pcount(skb);
+	}
+	return skb;
+}
+
+/* Avoid all extra work that is being done by sacktag while walking in
+ * a normal way
+ */
+static struct sk_buff *tcp_sacktag_skip(struct sk_buff *skb, struct sock *sk,
+					u32 skip_to_seq)
+{
+	tcp_for_write_queue_from(skb, sk) {
+		if (skb == tcp_send_head(sk))
+			break;
+
+		if (before(TCP_SKB_CB(skb)->end_seq, skip_to_seq))
+			break;
+	}
+	return skb;
+}
+
+static struct sk_buff *tcp_maybe_skipping_dsack(struct sk_buff *skb,
+						struct sock *sk,
+						struct tcp_sack_block *next_dup,
+						u32 skip_to_seq,
+						int *fack_count, int *reord,
+						int *flag)
+{
+	if (next_dup == NULL)
+		return skb;
+
+	if (before(next_dup->start_seq, skip_to_seq)) {
+		skb = tcp_sacktag_skip(skb, sk, next_dup->start_seq);
+		tcp_sacktag_walk(skb, sk, NULL,
+				 next_dup->start_seq, next_dup->end_seq,
+				 1, fack_count, reord, flag);
+	}
+
+	return skb;
+}
+
+static int tcp_sack_cache_ok(struct tcp_sock *tp, struct tcp_sack_block *cache)
+{
+	return cache < tp->recv_sack_cache + ARRAY_SIZE(tp->recv_sack_cache);
+}
+
 static int
 tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_una)
 {
@@ -1342,16 +1424,16 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 			      TCP_SKB_CB(ack_skb)->sacked);
 	struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2);
 	struct tcp_sack_block sp[4];
-	struct sk_buff *cached_skb;
+	struct tcp_sack_block *cache;
+	struct sk_buff *skb;
 	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
 	int used_sacks;
 	int reord = tp->packets_out;
 	int flag = 0;
 	int found_dup_sack = 0;
-	int cached_fack_count;
-	int i;
+	int fack_count;
+	int i, j;
 	int first_sack_index;
-	int force_one_sack;
 
 	if (!tp->sacked_out) {
 		if (WARN_ON(tp->fackets_out))
@@ -1409,132 +1491,123 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 		used_sacks++;
 	}
 
-	/* SACK fastpath:
-	 * if the only SACK change is the increase of the end_seq of
-	 * the first block then only apply that SACK block
-	 * and use retrans queue hinting otherwise slowpath */
-	force_one_sack = 1;
-	for (i = 0; i < used_sacks; i++) {
-		u32 start_seq = sp[i].start_seq;
-		u32 end_seq = sp[i].end_seq;
-
-		if (i == 0) {
-			if (tp->recv_sack_cache[i].start_seq != start_seq)
-				force_one_sack = 0;
-		} else {
-			if ((tp->recv_sack_cache[i].start_seq != start_seq) ||
-			    (tp->recv_sack_cache[i].end_seq != end_seq))
-				force_one_sack = 0;
-		}
-		tp->recv_sack_cache[i].start_seq = start_seq;
-		tp->recv_sack_cache[i].end_seq = end_seq;
-	}
-	/* Clear the rest of the cache sack blocks so they won't match mistakenly. */
-	for (; i < ARRAY_SIZE(tp->recv_sack_cache); i++) {
-		tp->recv_sack_cache[i].start_seq = 0;
-		tp->recv_sack_cache[i].end_seq = 0;
-	}
+	/* order SACK blocks to allow in order walk of the retrans queue */
+	for (i = used_sacks - 1; i > 0; i--) {
+		for (j = 0; j < i; j++){
+			if (after(sp[j].start_seq, sp[j+1].start_seq)) {
+				struct tcp_sack_block tmp;
 
-	if (force_one_sack)
-		used_sacks = 1;
-	else {
-		int j;
-		tp->fastpath_skb_hint = NULL;
-
-		/* order SACK blocks to allow in order walk of the retrans queue */
-		for (i = used_sacks - 1; i > 0; i--) {
-			for (j = 0; j < i; j++){
-				if (after(sp[j].start_seq, sp[j+1].start_seq)) {
-					struct tcp_sack_block tmp;
-
-					tmp = sp[j];
-					sp[j] = sp[j+1];
-					sp[j+1] = tmp;
-
-					/* Track where the first SACK block goes to */
-					if (j == first_sack_index)
-						first_sack_index = j+1;
-				}
+				tmp = sp[j];
+				sp[j] = sp[j+1];
+				sp[j+1] = tmp;
 
+				/* Track where the first SACK block goes to */
+				if (j == first_sack_index)
+					first_sack_index = j+1;
 			}
 		}
 	}
 
-	/* Use SACK fastpath hint if valid */
-	cached_skb = tp->fastpath_skb_hint;
-	cached_fack_count = tp->fastpath_cnt_hint;
-	if (!cached_skb) {
-		cached_skb = tcp_write_queue_head(sk);
-		cached_fack_count = 0;
+	skb = tcp_write_queue_head(sk);
+	fack_count = 0;
+	i = 0;
+
+	if (!tp->sacked_out) {
+		/* It's already past, so skip checking against it */
+		cache = tp->recv_sack_cache + ARRAY_SIZE(tp->recv_sack_cache);
+	} else {
+		cache = tp->recv_sack_cache;
+		/* Skip empty blocks in at head of the cache */
+		while (tcp_sack_cache_ok(tp, cache) && !cache->start_seq &&
+		       !cache->end_seq)
+			cache++;
 	}
 
-	for (i = 0; i < used_sacks; i++) {
-		struct sk_buff *skb;
+	while (i < used_sacks) {
 		u32 start_seq = sp[i].start_seq;
 		u32 end_seq = sp[i].end_seq;
-		int fack_count;
 		int dup_sack = (found_dup_sack && (i == first_sack_index));
-		int next_dup = (found_dup_sack && (i+1 == first_sack_index));
+		struct tcp_sack_block *next_dup = NULL;
 
-		skb = cached_skb;
-		fack_count = cached_fack_count;
+		if (found_dup_sack && ((i + 1) == first_sack_index))
+			next_dup = &sp[i + 1];
 
 		/* Event "B" in the comment above. */
 		if (after(end_seq, tp->high_seq))
 			flag |= FLAG_DATA_LOST;
 
-		tcp_for_write_queue_from(skb, sk) {
-			int in_sack = 0;
-
-			if (skb == tcp_send_head(sk))
-				break;
-
-			cached_skb = skb;
-			cached_fack_count = fack_count;
-			if (i == first_sack_index) {
-				tp->fastpath_skb_hint = skb;
-				tp->fastpath_cnt_hint = fack_count;
+		/* Skip too early cached blocks */
+		while (tcp_sack_cache_ok(tp, cache) &&
+		       !before(start_seq, cache->end_seq))
+			cache++;
+
+		/* Can skip some work by looking recv_sack_cache? */
+		if (tcp_sack_cache_ok(tp, cache) && !dup_sack &&
+		    after(end_seq, cache->start_seq)) {
+
+			/* Head todo? */
+			if (before(start_seq, cache->start_seq)) {
+				skb = tcp_sacktag_skip(skb, sk, start_seq);
+				skb = tcp_sacktag_walk(skb, sk, next_dup, start_seq,
+						       cache->start_seq, dup_sack,
+						       &fack_count, &reord, &flag);
 			}
 
-			/* The retransmission queue is always in order, so
-			 * we can short-circuit the walk early.
-			 */
-			if (!before(TCP_SKB_CB(skb)->seq, end_seq))
-				break;
-
-			dup_sack = (found_dup_sack && (i == first_sack_index));
+			/* Rest of the block already fully processed? */
+			if (!after(end_seq, cache->end_seq)) {
+				skb = tcp_maybe_skipping_dsack(skb, sk, next_dup, cache->end_seq,
+							       &fack_count, &reord, &flag);
+				goto advance_sp;
+			}
 
-			/* Due to sorting DSACK may reside within this SACK block! */
-			if (next_dup) {
-				u32 dup_start = sp[i+1].start_seq;
-				u32 dup_end = sp[i+1].end_seq;
+			/* ...tail remains todo... */
+			if (TCP_SKB_CB(tp->highest_sack)->end_seq == cache->end_seq) {
+				/* ...but better entrypoint exists! Check that DSACKs are
+				 * properly accounted while skipping here
+				 */
+				tcp_maybe_skipping_dsack(skb, sk, next_dup, cache->end_seq,
+							 &fack_count, &reord, &flag);
 
-				if (before(TCP_SKB_CB(skb)->seq, dup_end)) {
-					in_sack = tcp_match_skb_to_sack(sk, skb, dup_start, dup_end);
-					if (in_sack > 0)
-						dup_sack = 1;
-				}
+				skb = tcp_write_queue_next(sk, tp->highest_sack);
+				fack_count = tp->fackets_out;
+				cache++;
+				goto walk;
 			}
 
-			/* DSACK info lost if out-of-mem, try SACK still */
-			if (in_sack <= 0)
-				in_sack = tcp_match_skb_to_sack(sk, skb, start_seq, end_seq);
-			if (unlikely(in_sack < 0))
-				break;
-
-			if (in_sack)
-				flag |= tcp_sacktag_one(skb, tp, &reord, dup_sack, fack_count);
+			skb = tcp_sacktag_skip(skb, sk, cache->end_seq);
+			/* Check overlap against next cached too (past this one already) */
+			cache++;
+			continue;
+		}
 
-			fack_count += tcp_skb_pcount(skb);
+		if (!before(start_seq, tcp_highest_sack_seq(tp))) {
+			skb = tcp_write_queue_next(sk, tp->highest_sack);
+			fack_count = tp->fackets_out;
 		}
+		skb = tcp_sacktag_skip(skb, sk, start_seq);
+
+walk:
+		skb = tcp_sacktag_walk(skb, sk, next_dup, start_seq, end_seq,
+				       dup_sack, &fack_count, &reord, &flag);
 
+advance_sp:
 		/* SACK enhanced FRTO (RFC4138, Appendix B): Clearing correct
 		 * due to in-order walk
 		 */
 		if (after(end_seq, tp->frto_highmark))
 			flag &= ~FLAG_ONLY_ORIG_SACKED;
+
+		i++;
 	}
 
+	/* Clear the head of the cache sack blocks so we can skip it next time */
+	for (i = 0; i < ARRAY_SIZE(tp->recv_sack_cache) - used_sacks; i++) {
+		tp->recv_sack_cache[i].start_seq = 0;
+		tp->recv_sack_cache[i].end_seq = 0;
+	}
+	for (j = 0; j < used_sacks; j++)
+		tp->recv_sack_cache[i++] = sp[j];
+
 	flag |= tcp_mark_lost_retrans(sk);
 
 	tcp_verify_left_out(tp);
@@ -2821,9 +2894,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p,
 		}
 
 		tp->fackets_out -= min(pkts_acked, tp->fackets_out);
-		/* hint's skb might be NULL but we don't need to care */
-		tp->fastpath_cnt_hint -= min_t(u32, pkts_acked,
-					       tp->fastpath_cnt_hint);
+
 		if (ca_ops->pkts_acked) {
 			s32 rtt_us = -1;
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ce506af5ce07..030fc69ea217 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -653,9 +653,7 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned
 }
 
 /* When a modification to fackets out becomes necessary, we need to check
- * skb is counted to fackets_out or not. Another important thing is to
- * tweak SACK fastpath hint too as it would overwrite all changes unless
- * hint is also changed.
+ * skb is counted to fackets_out or not.
  */
 static void tcp_adjust_fackets_out(struct sock *sk, struct sk_buff *skb,
 				   int decr)
@@ -667,11 +665,6 @@ static void tcp_adjust_fackets_out(struct sock *sk, struct sk_buff *skb,
 
 	if (!before(tcp_highest_sack_seq(tp), TCP_SKB_CB(skb)->seq))
 		tp->fackets_out -= decr;
-
-	/* cnt_hint is "off-by-one" compared with fackets_out (see sacktag) */
-	if (tp->fastpath_skb_hint != NULL &&
-	    after(TCP_SKB_CB(tp->fastpath_skb_hint)->seq, TCP_SKB_CB(skb)->seq))
-		tp->fastpath_cnt_hint -= decr;
 }
 
 /* Function to create two new TCP segments.  Shrinks the given segment
@@ -1753,11 +1746,6 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 
 		/* changed transmit queue under us so clear hints */
 		tcp_clear_retrans_hints_partial(tp);
-		/* manually tune sacktag skb hint */
-		if (tp->fastpath_skb_hint == next_skb) {
-			tp->fastpath_skb_hint = skb;
-			tp->fastpath_cnt_hint -= tcp_skb_pcount(skb);
-		}
 
 		sk_stream_free_skb(sk, next_skb);
 	}
-- 
cgit v1.2.3


From cd05acfe65ed2cf2db683fa9a6adb8d35635263b Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Date: Sun, 16 Dec 2007 15:59:24 -0800
Subject: [CAN]: Allocate protocol numbers for PF_CAN

This patch adds a protocol/address family number, ARP hardware type,
ethernet packet type, and a line discipline number for the SocketCAN
implementation.

Signed-off-by: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Signed-off-by: Urs Thuermann <urs.thuermann@volkswagen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h       | 4 +++-
 include/linux/if_arp.h   | 1 +
 include/linux/if_ether.h | 1 +
 include/linux/socket.h   | 2 ++
 include/linux/tty.h      | 3 ++-
 net/core/sock.c          | 4 ++--
 6 files changed, 11 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/if.h b/include/linux/if.h
index 32bf419351f1..186070d5c544 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -50,7 +50,9 @@
 #define IFF_LOWER_UP	0x10000		/* driver signals L1 up		*/
 #define IFF_DORMANT	0x20000		/* driver signals dormant	*/
 
-#define IFF_VOLATILE	(IFF_LOOPBACK|IFF_POINTOPOINT|IFF_BROADCAST|\
+#define IFF_ECHO	0x40000		/* echo sent packets		*/
+
+#define IFF_VOLATILE	(IFF_LOOPBACK|IFF_POINTOPOINT|IFF_BROADCAST|IFF_ECHO|\
 		IFF_MASTER|IFF_SLAVE|IFF_RUNNING|IFF_LOWER_UP|IFF_DORMANT)
 
 /* Private (from user) interface flags (netdevice->priv_flags). */
diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h
index ed7b93c3083a..296e8e86e91d 100644
--- a/include/linux/if_arp.h
+++ b/include/linux/if_arp.h
@@ -52,6 +52,7 @@
 #define ARPHRD_ROSE	270
 #define ARPHRD_X25	271		/* CCITT X.25			*/
 #define ARPHRD_HWX25	272		/* Boards with X.25 in firmware	*/
+#define ARPHRD_CAN	280		/* Controller Area Network      */
 #define ARPHRD_PPP	512
 #define ARPHRD_CISCO	513		/* Cisco HDLC	 		*/
 #define ARPHRD_HDLC	ARPHRD_CISCO
diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h
index 5f9297793661..cc002cbbdc26 100644
--- a/include/linux/if_ether.h
+++ b/include/linux/if_ether.h
@@ -90,6 +90,7 @@
 #define ETH_P_WAN_PPP   0x0007          /* Dummy type for WAN PPP frames*/
 #define ETH_P_PPP_MP    0x0008          /* Dummy type for PPP MP frames */
 #define ETH_P_LOCALTALK 0x0009		/* Localtalk pseudo type 	*/
+#define ETH_P_CAN	0x000C		/* Controller Area Network      */
 #define ETH_P_PPPTALK	0x0010		/* Dummy type for Atalk over PPP*/
 #define ETH_P_TR_802_2	0x0011		/* 802.2 frames 		*/
 #define ETH_P_MOBITEX	0x0015		/* Mobitex (kaz@cafe.net)	*/
diff --git a/include/linux/socket.h b/include/linux/socket.h
index c22ef1c1afb8..eb5bdd59a644 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -185,6 +185,7 @@ struct ucred {
 #define AF_PPPOX	24	/* PPPoX sockets		*/
 #define AF_WANPIPE	25	/* Wanpipe API Sockets */
 #define AF_LLC		26	/* Linux LLC			*/
+#define AF_CAN		29	/* Controller Area Network      */
 #define AF_TIPC		30	/* TIPC sockets			*/
 #define AF_BLUETOOTH	31	/* Bluetooth sockets 		*/
 #define AF_IUCV		32	/* IUCV sockets			*/
@@ -220,6 +221,7 @@ struct ucred {
 #define PF_PPPOX	AF_PPPOX
 #define PF_WANPIPE	AF_WANPIPE
 #define PF_LLC		AF_LLC
+#define PF_CAN		AF_CAN
 #define PF_TIPC		AF_TIPC
 #define PF_BLUETOOTH	AF_BLUETOOTH
 #define PF_IUCV		AF_IUCV
diff --git a/include/linux/tty.h b/include/linux/tty.h
index defd2ab72449..402de892b3ed 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -23,7 +23,7 @@
  */
 #define NR_UNIX98_PTY_DEFAULT	4096      /* Default maximum for Unix98 ptys */
 #define NR_UNIX98_PTY_MAX	(1 << MINORBITS) /* Absolute limit */
-#define NR_LDISCS		17
+#define NR_LDISCS		18
 
 /* line disciplines */
 #define N_TTY		0
@@ -44,6 +44,7 @@
 #define N_SYNC_PPP	14	/* synchronous PPP */
 #define N_HCI		15	/* Bluetooth HCI UART */
 #define N_GIGASET_M101	16	/* Siemens Gigaset M101 serial DECT adapter */
+#define N_SLCAN		17	/* Serial / USB serial CAN Adaptors */
 
 /*
  * This character is the same as _POSIX_VDISABLE: it cannot be used as
diff --git a/net/core/sock.c b/net/core/sock.c
index 98b243a5b323..c9305a861760 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -154,7 +154,7 @@ static const char *af_family_key_strings[AF_MAX+1] = {
   "sk_lock-AF_ASH"   , "sk_lock-AF_ECONET"   , "sk_lock-AF_ATMSVC"   ,
   "sk_lock-21"       , "sk_lock-AF_SNA"      , "sk_lock-AF_IRDA"     ,
   "sk_lock-AF_PPPOX" , "sk_lock-AF_WANPIPE"  , "sk_lock-AF_LLC"      ,
-  "sk_lock-27"       , "sk_lock-28"          , "sk_lock-29"          ,
+  "sk_lock-27"       , "sk_lock-28"          , "sk_lock-AF_CAN"      ,
   "sk_lock-AF_TIPC"  , "sk_lock-AF_BLUETOOTH", "sk_lock-IUCV"        ,
   "sk_lock-AF_RXRPC" , "sk_lock-AF_MAX"
 };
@@ -168,7 +168,7 @@ static const char *af_family_slock_key_strings[AF_MAX+1] = {
   "slock-AF_ASH"   , "slock-AF_ECONET"   , "slock-AF_ATMSVC"   ,
   "slock-21"       , "slock-AF_SNA"      , "slock-AF_IRDA"     ,
   "slock-AF_PPPOX" , "slock-AF_WANPIPE"  , "slock-AF_LLC"      ,
-  "slock-27"       , "slock-28"          , "slock-29"          ,
+  "slock-27"       , "slock-28"          , "slock-AF_CAN"      ,
   "slock-AF_TIPC"  , "slock-AF_BLUETOOTH", "slock-AF_IUCV"     ,
   "slock-AF_RXRPC" , "slock-AF_MAX"
 };
-- 
cgit v1.2.3


From 0d66548a10cbbe0ef256852d63d30603f0f73f9b Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Date: Fri, 16 Nov 2007 15:52:17 -0800
Subject: [CAN]: Add PF_CAN core module

This patch adds the CAN core functionality but no protocols or drivers.
No protocol implementations are included here.  They come as separate
patches.  Protocol numbers are already in include/linux/can.h.

Signed-off-by: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Signed-off-by: Urs Thuermann <urs.thuermann@volkswagen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/can.h       | 111 ++++++
 include/linux/can/core.h  |  64 ++++
 include/linux/can/error.h |  93 +++++
 net/Kconfig               |   1 +
 net/Makefile              |   1 +
 net/can/Kconfig           |  17 +
 net/can/Makefile          |   6 +
 net/can/af_can.c          | 861 ++++++++++++++++++++++++++++++++++++++++++++++
 net/can/af_can.h          | 122 +++++++
 net/can/proc.c            | 533 ++++++++++++++++++++++++++++
 10 files changed, 1809 insertions(+)
 create mode 100644 include/linux/can.h
 create mode 100644 include/linux/can/core.h
 create mode 100644 include/linux/can/error.h
 create mode 100644 net/can/Kconfig
 create mode 100644 net/can/Makefile
 create mode 100644 net/can/af_can.c
 create mode 100644 net/can/af_can.h
 create mode 100644 net/can/proc.c

(limited to 'include')

diff --git a/include/linux/can.h b/include/linux/can.h
new file mode 100644
index 000000000000..d18333302cbd
--- /dev/null
+++ b/include/linux/can.h
@@ -0,0 +1,111 @@
+/*
+ * linux/can.h
+ *
+ * Definitions for CAN network layer (socket addr / CAN frame / CAN filter)
+ *
+ * Authors: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
+ *          Urs Thuermann   <urs.thuermann@volkswagen.de>
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef CAN_H
+#define CAN_H
+
+#include <linux/types.h>
+#include <linux/socket.h>
+
+/* controller area network (CAN) kernel definitions */
+
+/* special address description flags for the CAN_ID */
+#define CAN_EFF_FLAG 0x80000000U /* EFF/SFF is set in the MSB */
+#define CAN_RTR_FLAG 0x40000000U /* remote transmission request */
+#define CAN_ERR_FLAG 0x20000000U /* error frame */
+
+/* valid bits in CAN ID for frame formats */
+#define CAN_SFF_MASK 0x000007FFU /* standard frame format (SFF) */
+#define CAN_EFF_MASK 0x1FFFFFFFU /* extended frame format (EFF) */
+#define CAN_ERR_MASK 0x1FFFFFFFU /* omit EFF, RTR, ERR flags */
+
+/*
+ * Controller Area Network Identifier structure
+ *
+ * bit 0-28	: CAN identifier (11/29 bit)
+ * bit 29	: error frame flag (0 = data frame, 1 = error frame)
+ * bit 30	: remote transmission request flag (1 = rtr frame)
+ * bit 31	: frame format flag (0 = standard 11 bit, 1 = extended 29 bit)
+ */
+typedef __u32 canid_t;
+
+/*
+ * Controller Area Network Error Frame Mask structure
+ *
+ * bit 0-28	: error class mask (see include/linux/can/error.h)
+ * bit 29-31	: set to zero
+ */
+typedef __u32 can_err_mask_t;
+
+/**
+ * struct can_frame - basic CAN frame structure
+ * @can_id:  the CAN ID of the frame and CAN_*_FLAG flags, see above.
+ * @can_dlc: the data length field of the CAN frame
+ * @data:    the CAN frame payload.
+ */
+struct can_frame {
+	canid_t can_id;  /* 32 bit CAN_ID + EFF/RTR/ERR flags */
+	__u8    can_dlc; /* data length code: 0 .. 8 */
+	__u8    data[8] __attribute__((aligned(8)));
+};
+
+/* particular protocols of the protocol family PF_CAN */
+#define CAN_RAW		1 /* RAW sockets */
+#define CAN_BCM		2 /* Broadcast Manager */
+#define CAN_TP16	3 /* VAG Transport Protocol v1.6 */
+#define CAN_TP20	4 /* VAG Transport Protocol v2.0 */
+#define CAN_MCNET	5 /* Bosch MCNet */
+#define CAN_ISOTP	6 /* ISO 15765-2 Transport Protocol */
+#define CAN_NPROTO	7
+
+#define SOL_CAN_BASE 100
+
+/**
+ * struct sockaddr_can - the sockaddr structure for CAN sockets
+ * @can_family:  address family number AF_CAN.
+ * @can_ifindex: CAN network interface index.
+ * @can_addr:    protocol specific address information
+ */
+struct sockaddr_can {
+	sa_family_t can_family;
+	int         can_ifindex;
+	union {
+		/* transport protocol class address information (e.g. ISOTP) */
+		struct { canid_t rx_id, tx_id; } tp;
+
+		/* reserved for future CAN protocols address information */
+	} can_addr;
+};
+
+/**
+ * struct can_filter - CAN ID based filter in can_register().
+ * @can_id:   relevant bits of CAN ID which are not masked out.
+ * @can_mask: CAN mask (see description)
+ *
+ * Description:
+ * A filter matches, when
+ *
+ *          <received_can_id> & mask == can_id & mask
+ *
+ * The filter can be inverted (CAN_INV_FILTER bit set in can_id) or it can
+ * filter for error frames (CAN_ERR_FLAG bit set in mask).
+ */
+struct can_filter {
+	canid_t can_id;
+	canid_t can_mask;
+};
+
+#define CAN_INV_FILTER 0x20000000U /* to be set in can_filter.can_id */
+
+#endif /* CAN_H */
diff --git a/include/linux/can/core.h b/include/linux/can/core.h
new file mode 100644
index 000000000000..e9ca210ffa5b
--- /dev/null
+++ b/include/linux/can/core.h
@@ -0,0 +1,64 @@
+/*
+ * linux/can/core.h
+ *
+ * Protoypes and definitions for CAN protocol modules using the PF_CAN core
+ *
+ * Authors: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
+ *          Urs Thuermann   <urs.thuermann@volkswagen.de>
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef CAN_CORE_H
+#define CAN_CORE_H
+
+#include <linux/can.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+
+#define CAN_VERSION "20071116"
+
+/* increment this number each time you change some user-space interface */
+#define CAN_ABI_VERSION "8"
+
+#define CAN_VERSION_STRING "rev " CAN_VERSION " abi " CAN_ABI_VERSION
+
+#define DNAME(dev) ((dev) ? (dev)->name : "any")
+
+/**
+ * struct can_proto - CAN protocol structure
+ * @type:       type argument in socket() syscall, e.g. SOCK_DGRAM.
+ * @protocol:   protocol number in socket() syscall.
+ * @capability: capability needed to open the socket, or -1 for no restriction.
+ * @ops:        pointer to struct proto_ops for sock->ops.
+ * @prot:       pointer to struct proto structure.
+ */
+struct can_proto {
+	int              type;
+	int              protocol;
+	int              capability;
+	struct proto_ops *ops;
+	struct proto     *prot;
+};
+
+/* function prototypes for the CAN networklayer core (af_can.c) */
+
+extern int  can_proto_register(struct can_proto *cp);
+extern void can_proto_unregister(struct can_proto *cp);
+
+extern int  can_rx_register(struct net_device *dev, canid_t can_id,
+			    canid_t mask,
+			    void (*func)(struct sk_buff *, void *),
+			    void *data, char *ident);
+
+extern void can_rx_unregister(struct net_device *dev, canid_t can_id,
+			      canid_t mask,
+			      void (*func)(struct sk_buff *, void *),
+			      void *data);
+
+extern int can_send(struct sk_buff *skb, int loop);
+
+#endif /* CAN_CORE_H */
diff --git a/include/linux/can/error.h b/include/linux/can/error.h
new file mode 100644
index 000000000000..d4127fd9e681
--- /dev/null
+++ b/include/linux/can/error.h
@@ -0,0 +1,93 @@
+/*
+ * linux/can/error.h
+ *
+ * Definitions of the CAN error frame to be filtered and passed to the user.
+ *
+ * Author: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef CAN_ERROR_H
+#define CAN_ERROR_H
+
+#define CAN_ERR_DLC 8 /* dlc for error frames */
+
+/* error class (mask) in can_id */
+#define CAN_ERR_TX_TIMEOUT   0x00000001U /* TX timeout (by netdevice driver) */
+#define CAN_ERR_LOSTARB      0x00000002U /* lost arbitration    / data[0]    */
+#define CAN_ERR_CRTL         0x00000004U /* controller problems / data[1]    */
+#define CAN_ERR_PROT         0x00000008U /* protocol violations / data[2..3] */
+#define CAN_ERR_TRX          0x00000010U /* transceiver status  / data[4]    */
+#define CAN_ERR_ACK          0x00000020U /* received no ACK on transmission */
+#define CAN_ERR_BUSOFF       0x00000040U /* bus off */
+#define CAN_ERR_BUSERROR     0x00000080U /* bus error (may flood!) */
+#define CAN_ERR_RESTARTED    0x00000100U /* controller restarted */
+
+/* arbitration lost in bit ... / data[0] */
+#define CAN_ERR_LOSTARB_UNSPEC   0x00 /* unspecified */
+				      /* else bit number in bitstream */
+
+/* error status of CAN-controller / data[1] */
+#define CAN_ERR_CRTL_UNSPEC      0x00 /* unspecified */
+#define CAN_ERR_CRTL_RX_OVERFLOW 0x01 /* RX buffer overflow */
+#define CAN_ERR_CRTL_TX_OVERFLOW 0x02 /* TX buffer overflow */
+#define CAN_ERR_CRTL_RX_WARNING  0x04 /* reached warning level for RX errors */
+#define CAN_ERR_CRTL_TX_WARNING  0x08 /* reached warning level for TX errors */
+#define CAN_ERR_CRTL_RX_PASSIVE  0x10 /* reached error passive status RX */
+#define CAN_ERR_CRTL_TX_PASSIVE  0x20 /* reached error passive status TX */
+				      /* (at least one error counter exceeds */
+				      /* the protocol-defined level of 127)  */
+
+/* error in CAN protocol (type) / data[2] */
+#define CAN_ERR_PROT_UNSPEC      0x00 /* unspecified */
+#define CAN_ERR_PROT_BIT         0x01 /* single bit error */
+#define CAN_ERR_PROT_FORM        0x02 /* frame format error */
+#define CAN_ERR_PROT_STUFF       0x04 /* bit stuffing error */
+#define CAN_ERR_PROT_BIT0        0x08 /* unable to send dominant bit */
+#define CAN_ERR_PROT_BIT1        0x10 /* unable to send recessive bit */
+#define CAN_ERR_PROT_OVERLOAD    0x20 /* bus overload */
+#define CAN_ERR_PROT_ACTIVE      0x40 /* active error announcement */
+#define CAN_ERR_PROT_TX          0x80 /* error occured on transmission */
+
+/* error in CAN protocol (location) / data[3] */
+#define CAN_ERR_PROT_LOC_UNSPEC  0x00 /* unspecified */
+#define CAN_ERR_PROT_LOC_SOF     0x03 /* start of frame */
+#define CAN_ERR_PROT_LOC_ID28_21 0x02 /* ID bits 28 - 21 (SFF: 10 - 3) */
+#define CAN_ERR_PROT_LOC_ID20_18 0x06 /* ID bits 20 - 18 (SFF: 2 - 0 )*/
+#define CAN_ERR_PROT_LOC_SRTR    0x04 /* substitute RTR (SFF: RTR) */
+#define CAN_ERR_PROT_LOC_IDE     0x05 /* identifier extension */
+#define CAN_ERR_PROT_LOC_ID17_13 0x07 /* ID bits 17-13 */
+#define CAN_ERR_PROT_LOC_ID12_05 0x0F /* ID bits 12-5 */
+#define CAN_ERR_PROT_LOC_ID04_00 0x0E /* ID bits 4-0 */
+#define CAN_ERR_PROT_LOC_RTR     0x0C /* RTR */
+#define CAN_ERR_PROT_LOC_RES1    0x0D /* reserved bit 1 */
+#define CAN_ERR_PROT_LOC_RES0    0x09 /* reserved bit 0 */
+#define CAN_ERR_PROT_LOC_DLC     0x0B /* data length code */
+#define CAN_ERR_PROT_LOC_DATA    0x0A /* data section */
+#define CAN_ERR_PROT_LOC_CRC_SEQ 0x08 /* CRC sequence */
+#define CAN_ERR_PROT_LOC_CRC_DEL 0x18 /* CRC delimiter */
+#define CAN_ERR_PROT_LOC_ACK     0x19 /* ACK slot */
+#define CAN_ERR_PROT_LOC_ACK_DEL 0x1B /* ACK delimiter */
+#define CAN_ERR_PROT_LOC_EOF     0x1A /* end of frame */
+#define CAN_ERR_PROT_LOC_INTERM  0x12 /* intermission */
+
+/* error status of CAN-transceiver / data[4] */
+/*                                             CANH CANL */
+#define CAN_ERR_TRX_UNSPEC             0x00 /* 0000 0000 */
+#define CAN_ERR_TRX_CANH_NO_WIRE       0x04 /* 0000 0100 */
+#define CAN_ERR_TRX_CANH_SHORT_TO_BAT  0x05 /* 0000 0101 */
+#define CAN_ERR_TRX_CANH_SHORT_TO_VCC  0x06 /* 0000 0110 */
+#define CAN_ERR_TRX_CANH_SHORT_TO_GND  0x07 /* 0000 0111 */
+#define CAN_ERR_TRX_CANL_NO_WIRE       0x40 /* 0100 0000 */
+#define CAN_ERR_TRX_CANL_SHORT_TO_BAT  0x50 /* 0101 0000 */
+#define CAN_ERR_TRX_CANL_SHORT_TO_VCC  0x60 /* 0110 0000 */
+#define CAN_ERR_TRX_CANL_SHORT_TO_GND  0x70 /* 0111 0000 */
+#define CAN_ERR_TRX_CANL_SHORT_TO_CANH 0x80 /* 1000 0000 */
+
+/* controller specific additional information / data[5..7] */
+
+#endif /* CAN_ERROR_H */
diff --git a/net/Kconfig b/net/Kconfig
index ab4e6da5012f..58ed2f4199dc 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -218,6 +218,7 @@ endmenu
 endmenu
 
 source "net/ax25/Kconfig"
+source "net/can/Kconfig"
 source "net/irda/Kconfig"
 source "net/bluetooth/Kconfig"
 source "net/rxrpc/Kconfig"
diff --git a/net/Makefile b/net/Makefile
index bbe7d2a41486..b7a13643b549 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_LAPB)		+= lapb/
 obj-$(CONFIG_NETROM)		+= netrom/
 obj-$(CONFIG_ROSE)		+= rose/
 obj-$(CONFIG_AX25)		+= ax25/
+obj-$(CONFIG_CAN)		+= can/
 obj-$(CONFIG_IRDA)		+= irda/
 obj-$(CONFIG_BT)		+= bluetooth/
 obj-$(CONFIG_SUNRPC)		+= sunrpc/
diff --git a/net/can/Kconfig b/net/can/Kconfig
new file mode 100644
index 000000000000..8b92790747e5
--- /dev/null
+++ b/net/can/Kconfig
@@ -0,0 +1,17 @@
+#
+# Controller Area Network (CAN) network layer core configuration
+#
+
+menuconfig CAN
+	depends on NET
+	tristate "CAN bus subsystem support"
+	---help---
+	  Controller Area Network (CAN) is a slow (up to 1Mbit/s) serial
+	  communications protocol which was developed by Bosch in
+	  1991, mainly for automotive, but now widely used in marine
+	  (NMEA2000), industrial, and medical applications.
+	  More information on the CAN network protocol family PF_CAN
+	  is contained in <Documentation/networking/can.txt>.
+
+	  If you want CAN support you should say Y here and also to the
+	  specific driver for your controller(s) below.
diff --git a/net/can/Makefile b/net/can/Makefile
new file mode 100644
index 000000000000..4c7563c2ccb2
--- /dev/null
+++ b/net/can/Makefile
@@ -0,0 +1,6 @@
+#
+#  Makefile for the Linux Controller Area Network core.
+#
+
+obj-$(CONFIG_CAN)	+= can.o
+can-objs		:= af_can.o proc.o
diff --git a/net/can/af_can.c b/net/can/af_can.c
new file mode 100644
index 000000000000..5158e886630f
--- /dev/null
+++ b/net/can/af_can.c
@@ -0,0 +1,861 @@
+/*
+ * af_can.c - Protocol family CAN core module
+ *            (used by different CAN protocol modules)
+ *
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kmod.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/rcupdate.h>
+#include <linux/uaccess.h>
+#include <linux/net.h>
+#include <linux/netdevice.h>
+#include <linux/socket.h>
+#include <linux/if_ether.h>
+#include <linux/if_arp.h>
+#include <linux/skbuff.h>
+#include <linux/can.h>
+#include <linux/can/core.h>
+#include <net/net_namespace.h>
+#include <net/sock.h>
+
+#include "af_can.h"
+
+static __initdata const char banner[] = KERN_INFO
+	"can: controller area network core (" CAN_VERSION_STRING ")\n";
+
+MODULE_DESCRIPTION("Controller Area Network PF_CAN core");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Urs Thuermann <urs.thuermann@volkswagen.de>, "
+	      "Oliver Hartkopp <oliver.hartkopp@volkswagen.de>");
+
+MODULE_ALIAS_NETPROTO(PF_CAN);
+
+static int stats_timer __read_mostly = 1;
+module_param(stats_timer, int, S_IRUGO);
+MODULE_PARM_DESC(stats_timer, "enable timer for statistics (default:on)");
+
+HLIST_HEAD(can_rx_dev_list);
+static struct dev_rcv_lists can_rx_alldev_list;
+static DEFINE_SPINLOCK(can_rcvlists_lock);
+
+static struct kmem_cache *rcv_cache __read_mostly;
+
+/* table of registered CAN protocols */
+static struct can_proto *proto_tab[CAN_NPROTO] __read_mostly;
+static DEFINE_SPINLOCK(proto_tab_lock);
+
+struct timer_list can_stattimer;   /* timer for statistics update */
+struct s_stats    can_stats;       /* packet statistics */
+struct s_pstats   can_pstats;      /* receive list statistics */
+
+/*
+ * af_can socket functions
+ */
+
+static int can_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	struct sock *sk = sock->sk;
+
+	switch (cmd) {
+
+	case SIOCGSTAMP:
+		return sock_get_timestamp(sk, (struct timeval __user *)arg);
+
+	default:
+		return -ENOIOCTLCMD;
+	}
+}
+
+static void can_sock_destruct(struct sock *sk)
+{
+	skb_queue_purge(&sk->sk_receive_queue);
+}
+
+static int can_create(struct net *net, struct socket *sock, int protocol)
+{
+	struct sock *sk;
+	struct can_proto *cp;
+	char module_name[sizeof("can-proto-000")];
+	int err = 0;
+
+	sock->state = SS_UNCONNECTED;
+
+	if (protocol < 0 || protocol >= CAN_NPROTO)
+		return -EINVAL;
+
+	if (net != &init_net)
+		return -EAFNOSUPPORT;
+
+	/* try to load protocol module, when CONFIG_KMOD is defined */
+	if (!proto_tab[protocol]) {
+		sprintf(module_name, "can-proto-%d", protocol);
+		err = request_module(module_name);
+
+		/*
+		 * In case of error we only print a message but don't
+		 * return the error code immediately.  Below we will
+		 * return -EPROTONOSUPPORT
+		 */
+		if (err == -ENOSYS) {
+			if (printk_ratelimit())
+				printk(KERN_INFO "can: request_module(%s)"
+				       " not implemented.\n", module_name);
+		} else if (err) {
+			if (printk_ratelimit())
+				printk(KERN_ERR "can: request_module(%s)"
+				       " failed.\n", module_name);
+		}
+	}
+
+	spin_lock(&proto_tab_lock);
+	cp = proto_tab[protocol];
+	if (cp && !try_module_get(cp->prot->owner))
+		cp = NULL;
+	spin_unlock(&proto_tab_lock);
+
+	/* check for available protocol and correct usage */
+
+	if (!cp)
+		return -EPROTONOSUPPORT;
+
+	if (cp->type != sock->type) {
+		err = -EPROTONOSUPPORT;
+		goto errout;
+	}
+
+	if (cp->capability >= 0 && !capable(cp->capability)) {
+		err = -EPERM;
+		goto errout;
+	}
+
+	sock->ops = cp->ops;
+
+	sk = sk_alloc(net, PF_CAN, GFP_KERNEL, cp->prot);
+	if (!sk) {
+		err = -ENOMEM;
+		goto errout;
+	}
+
+	sock_init_data(sock, sk);
+	sk->sk_destruct = can_sock_destruct;
+
+	if (sk->sk_prot->init)
+		err = sk->sk_prot->init(sk);
+
+	if (err) {
+		/* release sk on errors */
+		sock_orphan(sk);
+		sock_put(sk);
+	}
+
+ errout:
+	module_put(cp->prot->owner);
+	return err;
+}
+
+/*
+ * af_can tx path
+ */
+
+/**
+ * can_send - transmit a CAN frame (optional with local loopback)
+ * @skb: pointer to socket buffer with CAN frame in data section
+ * @loop: loopback for listeners on local CAN sockets (recommended default!)
+ *
+ * Return:
+ *  0 on success
+ *  -ENETDOWN when the selected interface is down
+ *  -ENOBUFS on full driver queue (see net_xmit_errno())
+ *  -ENOMEM when local loopback failed at calling skb_clone()
+ *  -EPERM when trying to send on a non-CAN interface
+ */
+int can_send(struct sk_buff *skb, int loop)
+{
+	int err;
+
+	if (skb->dev->type != ARPHRD_CAN) {
+		kfree_skb(skb);
+		return -EPERM;
+	}
+
+	if (!(skb->dev->flags & IFF_UP)) {
+		kfree_skb(skb);
+		return -ENETDOWN;
+	}
+
+	skb->protocol = htons(ETH_P_CAN);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+
+	if (loop) {
+		/* local loopback of sent CAN frames */
+
+		/* indication for the CAN driver: do loopback */
+		skb->pkt_type = PACKET_LOOPBACK;
+
+		/*
+		 * The reference to the originating sock may be required
+		 * by the receiving socket to check whether the frame is
+		 * its own. Example: can_raw sockopt CAN_RAW_RECV_OWN_MSGS
+		 * Therefore we have to ensure that skb->sk remains the
+		 * reference to the originating sock by restoring skb->sk
+		 * after each skb_clone() or skb_orphan() usage.
+		 */
+
+		if (!(skb->dev->flags & IFF_ECHO)) {
+			/*
+			 * If the interface is not capable to do loopback
+			 * itself, we do it here.
+			 */
+			struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
+
+			if (!newskb) {
+				kfree_skb(skb);
+				return -ENOMEM;
+			}
+
+			newskb->sk = skb->sk;
+			newskb->ip_summed = CHECKSUM_UNNECESSARY;
+			newskb->pkt_type = PACKET_BROADCAST;
+			netif_rx(newskb);
+		}
+	} else {
+		/* indication for the CAN driver: no loopback required */
+		skb->pkt_type = PACKET_HOST;
+	}
+
+	/* send to netdevice */
+	err = dev_queue_xmit(skb);
+	if (err > 0)
+		err = net_xmit_errno(err);
+
+	/* update statistics */
+	can_stats.tx_frames++;
+	can_stats.tx_frames_delta++;
+
+	return err;
+}
+EXPORT_SYMBOL(can_send);
+
+/*
+ * af_can rx path
+ */
+
+static struct dev_rcv_lists *find_dev_rcv_lists(struct net_device *dev)
+{
+	struct dev_rcv_lists *d = NULL;
+	struct hlist_node *n;
+
+	/*
+	 * find receive list for this device
+	 *
+	 * The hlist_for_each_entry*() macros curse through the list
+	 * using the pointer variable n and set d to the containing
+	 * struct in each list iteration.  Therefore, after list
+	 * iteration, d is unmodified when the list is empty, and it
+	 * points to last list element, when the list is non-empty
+	 * but no match in the loop body is found.  I.e. d is *not*
+	 * NULL when no match is found.  We can, however, use the
+	 * cursor variable n to decide if a match was found.
+	 */
+
+	hlist_for_each_entry_rcu(d, n, &can_rx_dev_list, list) {
+		if (d->dev == dev)
+			break;
+	}
+
+	return n ? d : NULL;
+}
+
+static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask,
+					struct dev_rcv_lists *d)
+{
+	canid_t inv = *can_id & CAN_INV_FILTER; /* save flag before masking */
+
+	/* filter error frames */
+	if (*mask & CAN_ERR_FLAG) {
+		/* clear CAN_ERR_FLAG in list entry */
+		*mask &= CAN_ERR_MASK;
+		return &d->rx[RX_ERR];
+	}
+
+	/* ensure valid values in can_mask */
+	if (*mask & CAN_EFF_FLAG)
+		*mask &= (CAN_EFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG);
+	else
+		*mask &= (CAN_SFF_MASK | CAN_RTR_FLAG);
+
+	/* reduce condition testing at receive time */
+	*can_id &= *mask;
+
+	/* inverse can_id/can_mask filter */
+	if (inv)
+		return &d->rx[RX_INV];
+
+	/* mask == 0 => no condition testing at receive time */
+	if (!(*mask))
+		return &d->rx[RX_ALL];
+
+	/* use extra filterset for the subscription of exactly *ONE* can_id */
+	if (*can_id & CAN_EFF_FLAG) {
+		if (*mask == (CAN_EFF_MASK | CAN_EFF_FLAG)) {
+			/* RFC: a use-case for hash-tables in the future? */
+			return &d->rx[RX_EFF];
+		}
+	} else {
+		if (*mask == CAN_SFF_MASK)
+			return &d->rx_sff[*can_id];
+	}
+
+	/* default: filter via can_id/can_mask */
+	return &d->rx[RX_FIL];
+}
+
+/**
+ * can_rx_register - subscribe CAN frames from a specific interface
+ * @dev: pointer to netdevice (NULL => subcribe from 'all' CAN devices list)
+ * @can_id: CAN identifier (see description)
+ * @mask: CAN mask (see description)
+ * @func: callback function on filter match
+ * @data: returned parameter for callback function
+ * @ident: string for calling module indentification
+ *
+ * Description:
+ *  Invokes the callback function with the received sk_buff and the given
+ *  parameter 'data' on a matching receive filter. A filter matches, when
+ *
+ *          <received_can_id> & mask == can_id & mask
+ *
+ *  The filter can be inverted (CAN_INV_FILTER bit set in can_id) or it can
+ *  filter for error frames (CAN_ERR_FLAG bit set in mask).
+ *
+ * Return:
+ *  0 on success
+ *  -ENOMEM on missing cache mem to create subscription entry
+ *  -ENODEV unknown device
+ */
+int can_rx_register(struct net_device *dev, canid_t can_id, canid_t mask,
+		    void (*func)(struct sk_buff *, void *), void *data,
+		    char *ident)
+{
+	struct receiver *r;
+	struct hlist_head *rl;
+	struct dev_rcv_lists *d;
+	int err = 0;
+
+	/* insert new receiver  (dev,canid,mask) -> (func,data) */
+
+	r = kmem_cache_alloc(rcv_cache, GFP_KERNEL);
+	if (!r)
+		return -ENOMEM;
+
+	spin_lock(&can_rcvlists_lock);
+
+	d = find_dev_rcv_lists(dev);
+	if (d) {
+		rl = find_rcv_list(&can_id, &mask, d);
+
+		r->can_id  = can_id;
+		r->mask    = mask;
+		r->matches = 0;
+		r->func    = func;
+		r->data    = data;
+		r->ident   = ident;
+
+		hlist_add_head_rcu(&r->list, rl);
+		d->entries++;
+
+		can_pstats.rcv_entries++;
+		if (can_pstats.rcv_entries_max < can_pstats.rcv_entries)
+			can_pstats.rcv_entries_max = can_pstats.rcv_entries;
+	} else {
+		kmem_cache_free(rcv_cache, r);
+		err = -ENODEV;
+	}
+
+	spin_unlock(&can_rcvlists_lock);
+
+	return err;
+}
+EXPORT_SYMBOL(can_rx_register);
+
+/*
+ * can_rx_delete_device - rcu callback for dev_rcv_lists structure removal
+ */
+static void can_rx_delete_device(struct rcu_head *rp)
+{
+	struct dev_rcv_lists *d = container_of(rp, struct dev_rcv_lists, rcu);
+
+	kfree(d);
+}
+
+/*
+ * can_rx_delete_receiver - rcu callback for single receiver entry removal
+ */
+static void can_rx_delete_receiver(struct rcu_head *rp)
+{
+	struct receiver *r = container_of(rp, struct receiver, rcu);
+
+	kmem_cache_free(rcv_cache, r);
+}
+
+/**
+ * can_rx_unregister - unsubscribe CAN frames from a specific interface
+ * @dev: pointer to netdevice (NULL => unsubcribe from 'all' CAN devices list)
+ * @can_id: CAN identifier
+ * @mask: CAN mask
+ * @func: callback function on filter match
+ * @data: returned parameter for callback function
+ *
+ * Description:
+ *  Removes subscription entry depending on given (subscription) values.
+ */
+void can_rx_unregister(struct net_device *dev, canid_t can_id, canid_t mask,
+		       void (*func)(struct sk_buff *, void *), void *data)
+{
+	struct receiver *r = NULL;
+	struct hlist_head *rl;
+	struct hlist_node *next;
+	struct dev_rcv_lists *d;
+
+	spin_lock(&can_rcvlists_lock);
+
+	d = find_dev_rcv_lists(dev);
+	if (!d) {
+		printk(KERN_ERR "BUG: receive list not found for "
+		       "dev %s, id %03X, mask %03X\n",
+		       DNAME(dev), can_id, mask);
+		goto out;
+	}
+
+	rl = find_rcv_list(&can_id, &mask, d);
+
+	/*
+	 * Search the receiver list for the item to delete.  This should
+	 * exist, since no receiver may be unregistered that hasn't
+	 * been registered before.
+	 */
+
+	hlist_for_each_entry_rcu(r, next, rl, list) {
+		if (r->can_id == can_id && r->mask == mask
+		    && r->func == func && r->data == data)
+			break;
+	}
+
+	/*
+	 * Check for bugs in CAN protocol implementations:
+	 * If no matching list item was found, the list cursor variable next
+	 * will be NULL, while r will point to the last item of the list.
+	 */
+
+	if (!next) {
+		printk(KERN_ERR "BUG: receive list entry not found for "
+		       "dev %s, id %03X, mask %03X\n",
+		       DNAME(dev), can_id, mask);
+		r = NULL;
+		d = NULL;
+		goto out;
+	}
+
+	hlist_del_rcu(&r->list);
+	d->entries--;
+
+	if (can_pstats.rcv_entries > 0)
+		can_pstats.rcv_entries--;
+
+	/* remove device structure requested by NETDEV_UNREGISTER */
+	if (d->remove_on_zero_entries && !d->entries)
+		hlist_del_rcu(&d->list);
+	else
+		d = NULL;
+
+ out:
+	spin_unlock(&can_rcvlists_lock);
+
+	/* schedule the receiver item for deletion */
+	if (r)
+		call_rcu(&r->rcu, can_rx_delete_receiver);
+
+	/* schedule the device structure for deletion */
+	if (d)
+		call_rcu(&d->rcu, can_rx_delete_device);
+}
+EXPORT_SYMBOL(can_rx_unregister);
+
+static inline void deliver(struct sk_buff *skb, struct receiver *r)
+{
+	struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);
+
+	if (clone) {
+		clone->sk = skb->sk;
+		r->func(clone, r->data);
+		r->matches++;
+	}
+}
+
+static int can_rcv_filter(struct dev_rcv_lists *d, struct sk_buff *skb)
+{
+	struct receiver *r;
+	struct hlist_node *n;
+	int matches = 0;
+	struct can_frame *cf = (struct can_frame *)skb->data;
+	canid_t can_id = cf->can_id;
+
+	if (d->entries == 0)
+		return 0;
+
+	if (can_id & CAN_ERR_FLAG) {
+		/* check for error frame entries only */
+		hlist_for_each_entry_rcu(r, n, &d->rx[RX_ERR], list) {
+			if (can_id & r->mask) {
+				deliver(skb, r);
+				matches++;
+			}
+		}
+		return matches;
+	}
+
+	/* check for unfiltered entries */
+	hlist_for_each_entry_rcu(r, n, &d->rx[RX_ALL], list) {
+		deliver(skb, r);
+		matches++;
+	}
+
+	/* check for can_id/mask entries */
+	hlist_for_each_entry_rcu(r, n, &d->rx[RX_FIL], list) {
+		if ((can_id & r->mask) == r->can_id) {
+			deliver(skb, r);
+			matches++;
+		}
+	}
+
+	/* check for inverted can_id/mask entries */
+	hlist_for_each_entry_rcu(r, n, &d->rx[RX_INV], list) {
+		if ((can_id & r->mask) != r->can_id) {
+			deliver(skb, r);
+			matches++;
+		}
+	}
+
+	/* check CAN_ID specific entries */
+	if (can_id & CAN_EFF_FLAG) {
+		hlist_for_each_entry_rcu(r, n, &d->rx[RX_EFF], list) {
+			if (r->can_id == can_id) {
+				deliver(skb, r);
+				matches++;
+			}
+		}
+	} else {
+		can_id &= CAN_SFF_MASK;
+		hlist_for_each_entry_rcu(r, n, &d->rx_sff[can_id], list) {
+			deliver(skb, r);
+			matches++;
+		}
+	}
+
+	return matches;
+}
+
+static int can_rcv(struct sk_buff *skb, struct net_device *dev,
+		   struct packet_type *pt, struct net_device *orig_dev)
+{
+	struct dev_rcv_lists *d;
+	int matches;
+
+	if (dev->type != ARPHRD_CAN || dev->nd_net != &init_net) {
+		kfree_skb(skb);
+		return 0;
+	}
+
+	/* update statistics */
+	can_stats.rx_frames++;
+	can_stats.rx_frames_delta++;
+
+	rcu_read_lock();
+
+	/* deliver the packet to sockets listening on all devices */
+	matches = can_rcv_filter(&can_rx_alldev_list, skb);
+
+	/* find receive list for this device */
+	d = find_dev_rcv_lists(dev);
+	if (d)
+		matches += can_rcv_filter(d, skb);
+
+	rcu_read_unlock();
+
+	/* free the skbuff allocated by the netdevice driver */
+	kfree_skb(skb);
+
+	if (matches > 0) {
+		can_stats.matches++;
+		can_stats.matches_delta++;
+	}
+
+	return 0;
+}
+
+/*
+ * af_can protocol functions
+ */
+
+/**
+ * can_proto_register - register CAN transport protocol
+ * @cp: pointer to CAN protocol structure
+ *
+ * Return:
+ *  0 on success
+ *  -EINVAL invalid (out of range) protocol number
+ *  -EBUSY  protocol already in use
+ *  -ENOBUF if proto_register() fails
+ */
+int can_proto_register(struct can_proto *cp)
+{
+	int proto = cp->protocol;
+	int err = 0;
+
+	if (proto < 0 || proto >= CAN_NPROTO) {
+		printk(KERN_ERR "can: protocol number %d out of range\n",
+		       proto);
+		return -EINVAL;
+	}
+
+	spin_lock(&proto_tab_lock);
+	if (proto_tab[proto]) {
+		printk(KERN_ERR "can: protocol %d already registered\n",
+		       proto);
+		err = -EBUSY;
+		goto errout;
+	}
+
+	err = proto_register(cp->prot, 0);
+	if (err < 0)
+		goto errout;
+
+	proto_tab[proto] = cp;
+
+	/* use generic ioctl function if the module doesn't bring its own */
+	if (!cp->ops->ioctl)
+		cp->ops->ioctl = can_ioctl;
+
+ errout:
+	spin_unlock(&proto_tab_lock);
+
+	return err;
+}
+EXPORT_SYMBOL(can_proto_register);
+
+/**
+ * can_proto_unregister - unregister CAN transport protocol
+ * @cp: pointer to CAN protocol structure
+ */
+void can_proto_unregister(struct can_proto *cp)
+{
+	int proto = cp->protocol;
+
+	spin_lock(&proto_tab_lock);
+	if (!proto_tab[proto]) {
+		printk(KERN_ERR "BUG: can: protocol %d is not registered\n",
+		       proto);
+	}
+	proto_unregister(cp->prot);
+	proto_tab[proto] = NULL;
+	spin_unlock(&proto_tab_lock);
+}
+EXPORT_SYMBOL(can_proto_unregister);
+
+/*
+ * af_can notifier to create/remove CAN netdevice specific structs
+ */
+static int can_notifier(struct notifier_block *nb, unsigned long msg,
+			void *data)
+{
+	struct net_device *dev = (struct net_device *)data;
+	struct dev_rcv_lists *d;
+
+	if (dev->nd_net != &init_net)
+		return NOTIFY_DONE;
+
+	if (dev->type != ARPHRD_CAN)
+		return NOTIFY_DONE;
+
+	switch (msg) {
+
+	case NETDEV_REGISTER:
+
+		/*
+		 * create new dev_rcv_lists for this device
+		 *
+		 * N.B. zeroing the struct is the correct initialization
+		 * for the embedded hlist_head structs.
+		 * Another list type, e.g. list_head, would require
+		 * explicit initialization.
+		 */
+
+		d = kzalloc(sizeof(*d), GFP_KERNEL);
+		if (!d) {
+			printk(KERN_ERR
+			       "can: allocation of receive list failed\n");
+			return NOTIFY_DONE;
+		}
+		d->dev = dev;
+
+		spin_lock(&can_rcvlists_lock);
+		hlist_add_head_rcu(&d->list, &can_rx_dev_list);
+		spin_unlock(&can_rcvlists_lock);
+
+		break;
+
+	case NETDEV_UNREGISTER:
+		spin_lock(&can_rcvlists_lock);
+
+		d = find_dev_rcv_lists(dev);
+		if (d) {
+			if (d->entries) {
+				d->remove_on_zero_entries = 1;
+				d = NULL;
+			} else
+				hlist_del_rcu(&d->list);
+		} else
+			printk(KERN_ERR "can: notifier: receive list not "
+			       "found for dev %s\n", dev->name);
+
+		spin_unlock(&can_rcvlists_lock);
+
+		if (d)
+			call_rcu(&d->rcu, can_rx_delete_device);
+
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+/*
+ * af_can module init/exit functions
+ */
+
+static struct packet_type can_packet __read_mostly = {
+	.type = __constant_htons(ETH_P_CAN),
+	.dev  = NULL,
+	.func = can_rcv,
+};
+
+static struct net_proto_family can_family_ops __read_mostly = {
+	.family = PF_CAN,
+	.create = can_create,
+	.owner  = THIS_MODULE,
+};
+
+/* notifier block for netdevice event */
+static struct notifier_block can_netdev_notifier __read_mostly = {
+	.notifier_call = can_notifier,
+};
+
+static __init int can_init(void)
+{
+	printk(banner);
+
+	rcv_cache = kmem_cache_create("can_receiver", sizeof(struct receiver),
+				      0, 0, NULL);
+	if (!rcv_cache)
+		return -ENOMEM;
+
+	/*
+	 * Insert can_rx_alldev_list for reception on all devices.
+	 * This struct is zero initialized which is correct for the
+	 * embedded hlist heads, the dev pointer, and the entries counter.
+	 */
+
+	spin_lock(&can_rcvlists_lock);
+	hlist_add_head_rcu(&can_rx_alldev_list.list, &can_rx_dev_list);
+	spin_unlock(&can_rcvlists_lock);
+
+	if (stats_timer) {
+		/* the statistics are updated every second (timer triggered) */
+		setup_timer(&can_stattimer, can_stat_update, 0);
+		mod_timer(&can_stattimer, round_jiffies(jiffies + HZ));
+	} else
+		can_stattimer.function = NULL;
+
+	can_init_proc();
+
+	/* protocol register */
+	sock_register(&can_family_ops);
+	register_netdevice_notifier(&can_netdev_notifier);
+	dev_add_pack(&can_packet);
+
+	return 0;
+}
+
+static __exit void can_exit(void)
+{
+	struct dev_rcv_lists *d;
+	struct hlist_node *n, *next;
+
+	if (stats_timer)
+		del_timer(&can_stattimer);
+
+	can_remove_proc();
+
+	/* protocol unregister */
+	dev_remove_pack(&can_packet);
+	unregister_netdevice_notifier(&can_netdev_notifier);
+	sock_unregister(PF_CAN);
+
+	/* remove can_rx_dev_list */
+	spin_lock(&can_rcvlists_lock);
+	hlist_del(&can_rx_alldev_list.list);
+	hlist_for_each_entry_safe(d, n, next, &can_rx_dev_list, list) {
+		hlist_del(&d->list);
+		kfree(d);
+	}
+	spin_unlock(&can_rcvlists_lock);
+
+	kmem_cache_destroy(rcv_cache);
+}
+
+module_init(can_init);
+module_exit(can_exit);
diff --git a/net/can/af_can.h b/net/can/af_can.h
new file mode 100644
index 000000000000..18f91e37cc30
--- /dev/null
+++ b/net/can/af_can.h
@@ -0,0 +1,122 @@
+/*
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef AF_CAN_H
+#define AF_CAN_H
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/list.h>
+#include <linux/rcupdate.h>
+#include <linux/can.h>
+
+/* af_can rx dispatcher structures */
+
+struct receiver {
+	struct hlist_node list;
+	struct rcu_head rcu;
+	canid_t can_id;
+	canid_t mask;
+	unsigned long matches;
+	void (*func)(struct sk_buff *, void *);
+	void *data;
+	char *ident;
+};
+
+enum { RX_ERR, RX_ALL, RX_FIL, RX_INV, RX_EFF, RX_MAX };
+
+struct dev_rcv_lists {
+	struct hlist_node list;
+	struct rcu_head rcu;
+	struct net_device *dev;
+	struct hlist_head rx[RX_MAX];
+	struct hlist_head rx_sff[0x800];
+	int remove_on_zero_entries;
+	int entries;
+};
+
+/* statistic structures */
+
+/* can be reset e.g. by can_init_stats() */
+struct s_stats {
+	unsigned long jiffies_init;
+
+	unsigned long rx_frames;
+	unsigned long tx_frames;
+	unsigned long matches;
+
+	unsigned long total_rx_rate;
+	unsigned long total_tx_rate;
+	unsigned long total_rx_match_ratio;
+
+	unsigned long current_rx_rate;
+	unsigned long current_tx_rate;
+	unsigned long current_rx_match_ratio;
+
+	unsigned long max_rx_rate;
+	unsigned long max_tx_rate;
+	unsigned long max_rx_match_ratio;
+
+	unsigned long rx_frames_delta;
+	unsigned long tx_frames_delta;
+	unsigned long matches_delta;
+};
+
+/* persistent statistics */
+struct s_pstats {
+	unsigned long stats_reset;
+	unsigned long user_reset;
+	unsigned long rcv_entries;
+	unsigned long rcv_entries_max;
+};
+
+/* function prototypes for the CAN networklayer procfs (proc.c) */
+extern void can_init_proc(void);
+extern void can_remove_proc(void);
+extern void can_stat_update(unsigned long data);
+
+/* structures and variables from af_can.c needed in proc.c for reading */
+extern struct timer_list can_stattimer;    /* timer for statistics update */
+extern struct s_stats    can_stats;        /* packet statistics */
+extern struct s_pstats   can_pstats;       /* receive list statistics */
+extern struct hlist_head can_rx_dev_list;  /* rx dispatcher structures */
+
+#endif /* AF_CAN_H */
diff --git a/net/can/proc.c b/net/can/proc.c
new file mode 100644
index 000000000000..520fef5e5398
--- /dev/null
+++ b/net/can/proc.c
@@ -0,0 +1,533 @@
+/*
+ * proc.c - procfs support for Protocol family CAN core module
+ *
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/proc_fs.h>
+#include <linux/list.h>
+#include <linux/rcupdate.h>
+#include <linux/can/core.h>
+
+#include "af_can.h"
+
+/*
+ * proc filenames for the PF_CAN core
+ */
+
+#define CAN_PROC_VERSION     "version"
+#define CAN_PROC_STATS       "stats"
+#define CAN_PROC_RESET_STATS "reset_stats"
+#define CAN_PROC_RCVLIST_ALL "rcvlist_all"
+#define CAN_PROC_RCVLIST_FIL "rcvlist_fil"
+#define CAN_PROC_RCVLIST_INV "rcvlist_inv"
+#define CAN_PROC_RCVLIST_SFF "rcvlist_sff"
+#define CAN_PROC_RCVLIST_EFF "rcvlist_eff"
+#define CAN_PROC_RCVLIST_ERR "rcvlist_err"
+
+static struct proc_dir_entry *can_dir;
+static struct proc_dir_entry *pde_version;
+static struct proc_dir_entry *pde_stats;
+static struct proc_dir_entry *pde_reset_stats;
+static struct proc_dir_entry *pde_rcvlist_all;
+static struct proc_dir_entry *pde_rcvlist_fil;
+static struct proc_dir_entry *pde_rcvlist_inv;
+static struct proc_dir_entry *pde_rcvlist_sff;
+static struct proc_dir_entry *pde_rcvlist_eff;
+static struct proc_dir_entry *pde_rcvlist_err;
+
+static int user_reset;
+
+static const char rx_list_name[][8] = {
+	[RX_ERR] = "rx_err",
+	[RX_ALL] = "rx_all",
+	[RX_FIL] = "rx_fil",
+	[RX_INV] = "rx_inv",
+	[RX_EFF] = "rx_eff",
+};
+
+/*
+ * af_can statistics stuff
+ */
+
+static void can_init_stats(void)
+{
+	/*
+	 * This memset function is called from a timer context (when
+	 * can_stattimer is active which is the default) OR in a process
+	 * context (reading the proc_fs when can_stattimer is disabled).
+	 */
+	memset(&can_stats, 0, sizeof(can_stats));
+	can_stats.jiffies_init = jiffies;
+
+	can_pstats.stats_reset++;
+
+	if (user_reset) {
+		user_reset = 0;
+		can_pstats.user_reset++;
+	}
+}
+
+static unsigned long calc_rate(unsigned long oldjif, unsigned long newjif,
+			       unsigned long count)
+{
+	unsigned long rate;
+
+	if (oldjif == newjif)
+		return 0;
+
+	/* see can_stat_update() - this should NEVER happen! */
+	if (count > (ULONG_MAX / HZ)) {
+		printk(KERN_ERR "can: calc_rate: count exceeded! %ld\n",
+		       count);
+		return 99999999;
+	}
+
+	rate = (count * HZ) / (newjif - oldjif);
+
+	return rate;
+}
+
+void can_stat_update(unsigned long data)
+{
+	unsigned long j = jiffies; /* snapshot */
+
+	/* restart counting in timer context on user request */
+	if (user_reset)
+		can_init_stats();
+
+	/* restart counting on jiffies overflow */
+	if (j < can_stats.jiffies_init)
+		can_init_stats();
+
+	/* prevent overflow in calc_rate() */
+	if (can_stats.rx_frames > (ULONG_MAX / HZ))
+		can_init_stats();
+
+	/* prevent overflow in calc_rate() */
+	if (can_stats.tx_frames > (ULONG_MAX / HZ))
+		can_init_stats();
+
+	/* matches overflow - very improbable */
+	if (can_stats.matches > (ULONG_MAX / 100))
+		can_init_stats();
+
+	/* calc total values */
+	if (can_stats.rx_frames)
+		can_stats.total_rx_match_ratio = (can_stats.matches * 100) /
+			can_stats.rx_frames;
+
+	can_stats.total_tx_rate = calc_rate(can_stats.jiffies_init, j,
+					    can_stats.tx_frames);
+	can_stats.total_rx_rate = calc_rate(can_stats.jiffies_init, j,
+					    can_stats.rx_frames);
+
+	/* calc current values */
+	if (can_stats.rx_frames_delta)
+		can_stats.current_rx_match_ratio =
+			(can_stats.matches_delta * 100) /
+			can_stats.rx_frames_delta;
+
+	can_stats.current_tx_rate = calc_rate(0, HZ, can_stats.tx_frames_delta);
+	can_stats.current_rx_rate = calc_rate(0, HZ, can_stats.rx_frames_delta);
+
+	/* check / update maximum values */
+	if (can_stats.max_tx_rate < can_stats.current_tx_rate)
+		can_stats.max_tx_rate = can_stats.current_tx_rate;
+
+	if (can_stats.max_rx_rate < can_stats.current_rx_rate)
+		can_stats.max_rx_rate = can_stats.current_rx_rate;
+
+	if (can_stats.max_rx_match_ratio < can_stats.current_rx_match_ratio)
+		can_stats.max_rx_match_ratio = can_stats.current_rx_match_ratio;
+
+	/* clear values for 'current rate' calculation */
+	can_stats.tx_frames_delta = 0;
+	can_stats.rx_frames_delta = 0;
+	can_stats.matches_delta   = 0;
+
+	/* restart timer (one second) */
+	mod_timer(&can_stattimer, round_jiffies(jiffies + HZ));
+}
+
+/*
+ * proc read functions
+ *
+ * From known use-cases we expect about 10 entries in a receive list to be
+ * printed in the proc_fs. So PAGE_SIZE is definitely enough space here.
+ *
+ */
+
+static int can_print_rcvlist(char *page, int len, struct hlist_head *rx_list,
+			     struct net_device *dev)
+{
+	struct receiver *r;
+	struct hlist_node *n;
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(r, n, rx_list, list) {
+		char *fmt = (r->can_id & CAN_EFF_FLAG)?
+			"   %-5s  %08X  %08x  %08x  %08x  %8ld  %s\n" :
+			"   %-5s     %03X    %08x  %08lx  %08lx  %8ld  %s\n";
+
+		len += snprintf(page + len, PAGE_SIZE - len, fmt,
+				DNAME(dev), r->can_id, r->mask,
+				(unsigned long)r->func, (unsigned long)r->data,
+				r->matches, r->ident);
+
+		/* does a typical line fit into the current buffer? */
+
+		/* 100 Bytes before end of buffer */
+		if (len > PAGE_SIZE - 100) {
+			/* mark output cut off */
+			len += snprintf(page + len, PAGE_SIZE - len,
+					"   (..)\n");
+			break;
+		}
+	}
+	rcu_read_unlock();
+
+	return len;
+}
+
+static int can_print_recv_banner(char *page, int len)
+{
+	/*
+	 *                  can1.  00000000  00000000  00000000
+	 *                 .......          0  tp20
+	 */
+	len += snprintf(page + len, PAGE_SIZE - len,
+			"  device   can_id   can_mask  function"
+			"  userdata   matches  ident\n");
+
+	return len;
+}
+
+static int can_proc_read_stats(char *page, char **start, off_t off,
+			       int count, int *eof, void *data)
+{
+	int len = 0;
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+	len += snprintf(page + len, PAGE_SIZE - len,
+			" %8ld transmitted frames (TXF)\n",
+			can_stats.tx_frames);
+	len += snprintf(page + len, PAGE_SIZE - len,
+			" %8ld received frames (RXF)\n", can_stats.rx_frames);
+	len += snprintf(page + len, PAGE_SIZE - len,
+			" %8ld matched frames (RXMF)\n", can_stats.matches);
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+	if (can_stattimer.function == can_stat_update) {
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld %% total match ratio (RXMR)\n",
+				can_stats.total_rx_match_ratio);
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s total tx rate (TXR)\n",
+				can_stats.total_tx_rate);
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s total rx rate (RXR)\n",
+				can_stats.total_rx_rate);
+
+		len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld %% current match ratio (CRXMR)\n",
+				can_stats.current_rx_match_ratio);
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s current tx rate (CTXR)\n",
+				can_stats.current_tx_rate);
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s current rx rate (CRXR)\n",
+				can_stats.current_rx_rate);
+
+		len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld %% max match ratio (MRXMR)\n",
+				can_stats.max_rx_match_ratio);
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s max tx rate (MTXR)\n",
+				can_stats.max_tx_rate);
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld frames/s max rx rate (MRXR)\n",
+				can_stats.max_rx_rate);
+
+		len += snprintf(page + len, PAGE_SIZE - len, "\n");
+	}
+
+	len += snprintf(page + len, PAGE_SIZE - len,
+			" %8ld current receive list entries (CRCV)\n",
+			can_pstats.rcv_entries);
+	len += snprintf(page + len, PAGE_SIZE - len,
+			" %8ld maximum receive list entries (MRCV)\n",
+			can_pstats.rcv_entries_max);
+
+	if (can_pstats.stats_reset)
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"\n %8ld statistic resets (STR)\n",
+				can_pstats.stats_reset);
+
+	if (can_pstats.user_reset)
+		len += snprintf(page + len, PAGE_SIZE - len,
+				" %8ld user statistic resets (USTR)\n",
+				can_pstats.user_reset);
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+	*eof = 1;
+	return len;
+}
+
+static int can_proc_read_reset_stats(char *page, char **start, off_t off,
+				     int count, int *eof, void *data)
+{
+	int len = 0;
+
+	user_reset = 1;
+
+	if (can_stattimer.function == can_stat_update) {
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"Scheduled statistic reset #%ld.\n",
+				can_pstats.stats_reset + 1);
+
+	} else {
+		if (can_stats.jiffies_init != jiffies)
+			can_init_stats();
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"Performed statistic reset #%ld.\n",
+				can_pstats.stats_reset);
+	}
+
+	*eof = 1;
+	return len;
+}
+
+static int can_proc_read_version(char *page, char **start, off_t off,
+				 int count, int *eof, void *data)
+{
+	int len = 0;
+
+	len += snprintf(page + len, PAGE_SIZE - len, "%s\n",
+			CAN_VERSION_STRING);
+	*eof = 1;
+	return len;
+}
+
+static int can_proc_read_rcvlist(char *page, char **start, off_t off,
+				 int count, int *eof, void *data)
+{
+	/* double cast to prevent GCC warning */
+	int idx = (int)(long)data;
+	int len = 0;
+	struct dev_rcv_lists *d;
+	struct hlist_node *n;
+
+	len += snprintf(page + len, PAGE_SIZE - len,
+			"\nreceive list '%s':\n", rx_list_name[idx]);
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(d, n, &can_rx_dev_list, list) {
+
+		if (!hlist_empty(&d->rx[idx])) {
+			len = can_print_recv_banner(page, len);
+			len = can_print_rcvlist(page, len, &d->rx[idx], d->dev);
+		} else
+			len += snprintf(page + len, PAGE_SIZE - len,
+					"  (%s: no entry)\n", DNAME(d->dev));
+
+		/* exit on end of buffer? */
+		if (len > PAGE_SIZE - 100)
+			break;
+	}
+	rcu_read_unlock();
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+	*eof = 1;
+	return len;
+}
+
+static int can_proc_read_rcvlist_sff(char *page, char **start, off_t off,
+				     int count, int *eof, void *data)
+{
+	int len = 0;
+	struct dev_rcv_lists *d;
+	struct hlist_node *n;
+
+	/* RX_SFF */
+	len += snprintf(page + len, PAGE_SIZE - len,
+			"\nreceive list 'rx_sff':\n");
+
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(d, n, &can_rx_dev_list, list) {
+		int i, all_empty = 1;
+		/* check wether at least one list is non-empty */
+		for (i = 0; i < 0x800; i++)
+			if (!hlist_empty(&d->rx_sff[i])) {
+				all_empty = 0;
+				break;
+			}
+
+		if (!all_empty) {
+			len = can_print_recv_banner(page, len);
+			for (i = 0; i < 0x800; i++) {
+				if (!hlist_empty(&d->rx_sff[i]) &&
+				    len < PAGE_SIZE - 100)
+					len = can_print_rcvlist(page, len,
+								&d->rx_sff[i],
+								d->dev);
+			}
+		} else
+			len += snprintf(page + len, PAGE_SIZE - len,
+					"  (%s: no entry)\n", DNAME(d->dev));
+
+		/* exit on end of buffer? */
+		if (len > PAGE_SIZE - 100)
+			break;
+	}
+	rcu_read_unlock();
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+	*eof = 1;
+	return len;
+}
+
+/*
+ * proc utility functions
+ */
+
+static struct proc_dir_entry *can_create_proc_readentry(const char *name,
+							mode_t mode,
+							read_proc_t *read_proc,
+							void *data)
+{
+	if (can_dir)
+		return create_proc_read_entry(name, mode, can_dir, read_proc,
+					      data);
+	else
+		return NULL;
+}
+
+static void can_remove_proc_readentry(const char *name)
+{
+	if (can_dir)
+		remove_proc_entry(name, can_dir);
+}
+
+/*
+ * can_init_proc - create main CAN proc directory and procfs entries
+ */
+void can_init_proc(void)
+{
+	/* create /proc/net/can directory */
+	can_dir = proc_mkdir("can", init_net.proc_net);
+
+	if (!can_dir) {
+		printk(KERN_INFO "can: failed to create /proc/net/can . "
+		       "CONFIG_PROC_FS missing?\n");
+		return;
+	}
+
+	can_dir->owner = THIS_MODULE;
+
+	/* own procfs entries from the AF_CAN core */
+	pde_version     = can_create_proc_readentry(CAN_PROC_VERSION, 0644,
+					can_proc_read_version, NULL);
+	pde_stats       = can_create_proc_readentry(CAN_PROC_STATS, 0644,
+					can_proc_read_stats, NULL);
+	pde_reset_stats = can_create_proc_readentry(CAN_PROC_RESET_STATS, 0644,
+					can_proc_read_reset_stats, NULL);
+	pde_rcvlist_err = can_create_proc_readentry(CAN_PROC_RCVLIST_ERR, 0644,
+					can_proc_read_rcvlist, (void *)RX_ERR);
+	pde_rcvlist_all = can_create_proc_readentry(CAN_PROC_RCVLIST_ALL, 0644,
+					can_proc_read_rcvlist, (void *)RX_ALL);
+	pde_rcvlist_fil = can_create_proc_readentry(CAN_PROC_RCVLIST_FIL, 0644,
+					can_proc_read_rcvlist, (void *)RX_FIL);
+	pde_rcvlist_inv = can_create_proc_readentry(CAN_PROC_RCVLIST_INV, 0644,
+					can_proc_read_rcvlist, (void *)RX_INV);
+	pde_rcvlist_eff = can_create_proc_readentry(CAN_PROC_RCVLIST_EFF, 0644,
+					can_proc_read_rcvlist, (void *)RX_EFF);
+	pde_rcvlist_sff = can_create_proc_readentry(CAN_PROC_RCVLIST_SFF, 0644,
+					can_proc_read_rcvlist_sff, NULL);
+}
+
+/*
+ * can_remove_proc - remove procfs entries and main CAN proc directory
+ */
+void can_remove_proc(void)
+{
+	if (pde_version)
+		can_remove_proc_readentry(CAN_PROC_VERSION);
+
+	if (pde_stats)
+		can_remove_proc_readentry(CAN_PROC_STATS);
+
+	if (pde_reset_stats)
+		can_remove_proc_readentry(CAN_PROC_RESET_STATS);
+
+	if (pde_rcvlist_err)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_ERR);
+
+	if (pde_rcvlist_all)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_ALL);
+
+	if (pde_rcvlist_fil)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_FIL);
+
+	if (pde_rcvlist_inv)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_INV);
+
+	if (pde_rcvlist_eff)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_EFF);
+
+	if (pde_rcvlist_sff)
+		can_remove_proc_readentry(CAN_PROC_RCVLIST_SFF);
+
+	if (can_dir)
+		proc_net_remove(&init_net, "can");
+}
-- 
cgit v1.2.3


From c18ce101f2e47d97ace125033e2896895a6db3dd Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Date: Fri, 16 Nov 2007 15:53:09 -0800
Subject: [CAN]: Add raw protocol

This patch adds the CAN raw protocol.

Signed-off-by: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Signed-off-by: Urs Thuermann <urs.thuermann@volkswagen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/can/raw.h |  31 ++
 net/can/Kconfig         |  11 +
 net/can/Makefile        |   3 +
 net/can/raw.c           | 763 ++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 808 insertions(+)
 create mode 100644 include/linux/can/raw.h
 create mode 100644 net/can/raw.c

(limited to 'include')

diff --git a/include/linux/can/raw.h b/include/linux/can/raw.h
new file mode 100644
index 000000000000..b2a0f87492c5
--- /dev/null
+++ b/include/linux/can/raw.h
@@ -0,0 +1,31 @@
+/*
+ * linux/can/raw.h
+ *
+ * Definitions for raw CAN sockets
+ *
+ * Authors: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
+ *          Urs Thuermann   <urs.thuermann@volkswagen.de>
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef CAN_RAW_H
+#define CAN_RAW_H
+
+#include <linux/can.h>
+
+#define SOL_CAN_RAW (SOL_CAN_BASE + CAN_RAW)
+
+/* for socket options affecting the socket (not the global system) */
+
+enum {
+	CAN_RAW_FILTER = 1,	/* set 0 .. n can_filter(s)          */
+	CAN_RAW_ERR_FILTER,	/* set filter for error frames       */
+	CAN_RAW_LOOPBACK,	/* local loopback (default:on)       */
+	CAN_RAW_RECV_OWN_MSGS	/* receive my own msgs (default:off) */
+};
+
+#endif
diff --git a/net/can/Kconfig b/net/can/Kconfig
index 8b92790747e5..4718d1f50ab3 100644
--- a/net/can/Kconfig
+++ b/net/can/Kconfig
@@ -15,3 +15,14 @@ menuconfig CAN
 
 	  If you want CAN support you should say Y here and also to the
 	  specific driver for your controller(s) below.
+
+config CAN_RAW
+	tristate "Raw CAN Protocol (raw access with CAN-ID filtering)"
+	depends on CAN
+	default N
+	---help---
+	  The raw CAN protocol option offers access to the CAN bus via
+	  the BSD socket API. You probably want to use the raw socket in
+	  most cases where no higher level protocol is being used. The raw
+	  socket has several filter options e.g. ID masking / error frames.
+	  To receive/send raw CAN messages, use AF_CAN with protocol CAN_RAW.
diff --git a/net/can/Makefile b/net/can/Makefile
index 4c7563c2ccb2..86f1cf21fce4 100644
--- a/net/can/Makefile
+++ b/net/can/Makefile
@@ -4,3 +4,6 @@
 
 obj-$(CONFIG_CAN)	+= can.o
 can-objs		:= af_can.o proc.o
+
+obj-$(CONFIG_CAN_RAW)	+= can-raw.o
+can-raw-objs		:= raw.o
diff --git a/net/can/raw.c b/net/can/raw.c
new file mode 100644
index 000000000000..aeefd1419d00
--- /dev/null
+++ b/net/can/raw.c
@@ -0,0 +1,763 @@
+/*
+ * raw.c - Raw sockets for protocol family CAN
+ *
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/uio.h>
+#include <linux/net.h>
+#include <linux/netdevice.h>
+#include <linux/socket.h>
+#include <linux/if_arp.h>
+#include <linux/skbuff.h>
+#include <linux/can.h>
+#include <linux/can/core.h>
+#include <linux/can/raw.h>
+#include <net/sock.h>
+#include <net/net_namespace.h>
+
+#define CAN_RAW_VERSION CAN_VERSION
+static __initdata const char banner[] =
+	KERN_INFO "can: raw protocol (rev " CAN_RAW_VERSION ")\n";
+
+MODULE_DESCRIPTION("PF_CAN raw protocol");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Urs Thuermann <urs.thuermann@volkswagen.de>");
+
+#define MASK_ALL 0
+
+/*
+ * A raw socket has a list of can_filters attached to it, each receiving
+ * the CAN frames matching that filter.  If the filter list is empty,
+ * no CAN frames will be received by the socket.  The default after
+ * opening the socket, is to have one filter which receives all frames.
+ * The filter list is allocated dynamically with the exception of the
+ * list containing only one item.  This common case is optimized by
+ * storing the single filter in dfilter, to avoid using dynamic memory.
+ */
+
+struct raw_sock {
+	struct sock sk;
+	int bound;
+	int ifindex;
+	struct notifier_block notifier;
+	int loopback;
+	int recv_own_msgs;
+	int count;                 /* number of active filters */
+	struct can_filter dfilter; /* default/single filter */
+	struct can_filter *filter; /* pointer to filter(s) */
+	can_err_mask_t err_mask;
+};
+
+static inline struct raw_sock *raw_sk(const struct sock *sk)
+{
+	return (struct raw_sock *)sk;
+}
+
+static void raw_rcv(struct sk_buff *skb, void *data)
+{
+	struct sock *sk = (struct sock *)data;
+	struct raw_sock *ro = raw_sk(sk);
+	struct sockaddr_can *addr;
+	int error;
+
+	if (!ro->recv_own_msgs) {
+		/* check the received tx sock reference */
+		if (skb->sk == sk) {
+			kfree_skb(skb);
+			return;
+		}
+	}
+
+	/*
+	 *  Put the datagram to the queue so that raw_recvmsg() can
+	 *  get it from there.  We need to pass the interface index to
+	 *  raw_recvmsg().  We pass a whole struct sockaddr_can in skb->cb
+	 *  containing the interface index.
+	 */
+
+	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct sockaddr_can));
+	addr = (struct sockaddr_can *)skb->cb;
+	memset(addr, 0, sizeof(*addr));
+	addr->can_family  = AF_CAN;
+	addr->can_ifindex = skb->dev->ifindex;
+
+	error = sock_queue_rcv_skb(sk, skb);
+	if (error < 0)
+		kfree_skb(skb);
+}
+
+static int raw_enable_filters(struct net_device *dev, struct sock *sk,
+			      struct can_filter *filter,
+			      int count)
+{
+	int err = 0;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		err = can_rx_register(dev, filter[i].can_id,
+				      filter[i].can_mask,
+				      raw_rcv, sk, "raw");
+		if (err) {
+			/* clean up successfully registered filters */
+			while (--i >= 0)
+				can_rx_unregister(dev, filter[i].can_id,
+						  filter[i].can_mask,
+						  raw_rcv, sk);
+			break;
+		}
+	}
+
+	return err;
+}
+
+static int raw_enable_errfilter(struct net_device *dev, struct sock *sk,
+				can_err_mask_t err_mask)
+{
+	int err = 0;
+
+	if (err_mask)
+		err = can_rx_register(dev, 0, err_mask | CAN_ERR_FLAG,
+				      raw_rcv, sk, "raw");
+
+	return err;
+}
+
+static void raw_disable_filters(struct net_device *dev, struct sock *sk,
+			      struct can_filter *filter,
+			      int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++)
+		can_rx_unregister(dev, filter[i].can_id, filter[i].can_mask,
+				  raw_rcv, sk);
+}
+
+static inline void raw_disable_errfilter(struct net_device *dev,
+					 struct sock *sk,
+					 can_err_mask_t err_mask)
+
+{
+	if (err_mask)
+		can_rx_unregister(dev, 0, err_mask | CAN_ERR_FLAG,
+				  raw_rcv, sk);
+}
+
+static inline void raw_disable_allfilters(struct net_device *dev,
+					  struct sock *sk)
+{
+	struct raw_sock *ro = raw_sk(sk);
+
+	raw_disable_filters(dev, sk, ro->filter, ro->count);
+	raw_disable_errfilter(dev, sk, ro->err_mask);
+}
+
+static int raw_enable_allfilters(struct net_device *dev, struct sock *sk)
+{
+	struct raw_sock *ro = raw_sk(sk);
+	int err;
+
+	err = raw_enable_filters(dev, sk, ro->filter, ro->count);
+	if (!err) {
+		err = raw_enable_errfilter(dev, sk, ro->err_mask);
+		if (err)
+			raw_disable_filters(dev, sk, ro->filter, ro->count);
+	}
+
+	return err;
+}
+
+static int raw_notifier(struct notifier_block *nb,
+			unsigned long msg, void *data)
+{
+	struct net_device *dev = (struct net_device *)data;
+	struct raw_sock *ro = container_of(nb, struct raw_sock, notifier);
+	struct sock *sk = &ro->sk;
+
+	if (dev->nd_net != &init_net)
+		return NOTIFY_DONE;
+
+	if (dev->type != ARPHRD_CAN)
+		return NOTIFY_DONE;
+
+	if (ro->ifindex != dev->ifindex)
+		return NOTIFY_DONE;
+
+	switch (msg) {
+
+	case NETDEV_UNREGISTER:
+		lock_sock(sk);
+		/* remove current filters & unregister */
+		if (ro->bound)
+			raw_disable_allfilters(dev, sk);
+
+		if (ro->count > 1)
+			kfree(ro->filter);
+
+		ro->ifindex = 0;
+		ro->bound   = 0;
+		ro->count   = 0;
+		release_sock(sk);
+
+		sk->sk_err = ENODEV;
+		if (!sock_flag(sk, SOCK_DEAD))
+			sk->sk_error_report(sk);
+		break;
+
+	case NETDEV_DOWN:
+		sk->sk_err = ENETDOWN;
+		if (!sock_flag(sk, SOCK_DEAD))
+			sk->sk_error_report(sk);
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int raw_init(struct sock *sk)
+{
+	struct raw_sock *ro = raw_sk(sk);
+
+	ro->bound            = 0;
+	ro->ifindex          = 0;
+
+	/* set default filter to single entry dfilter */
+	ro->dfilter.can_id   = 0;
+	ro->dfilter.can_mask = MASK_ALL;
+	ro->filter           = &ro->dfilter;
+	ro->count            = 1;
+
+	/* set default loopback behaviour */
+	ro->loopback         = 1;
+	ro->recv_own_msgs    = 0;
+
+	/* set notifier */
+	ro->notifier.notifier_call = raw_notifier;
+
+	register_netdevice_notifier(&ro->notifier);
+
+	return 0;
+}
+
+static int raw_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+
+	unregister_netdevice_notifier(&ro->notifier);
+
+	lock_sock(sk);
+
+	/* remove current filters & unregister */
+	if (ro->bound) {
+		if (ro->ifindex) {
+			struct net_device *dev;
+
+			dev = dev_get_by_index(&init_net, ro->ifindex);
+			if (dev) {
+				raw_disable_allfilters(dev, sk);
+				dev_put(dev);
+			}
+		} else
+			raw_disable_allfilters(NULL, sk);
+	}
+
+	if (ro->count > 1)
+		kfree(ro->filter);
+
+	ro->ifindex = 0;
+	ro->bound   = 0;
+	ro->count   = 0;
+
+	release_sock(sk);
+	sock_put(sk);
+
+	return 0;
+}
+
+static int raw_bind(struct socket *sock, struct sockaddr *uaddr, int len)
+{
+	struct sockaddr_can *addr = (struct sockaddr_can *)uaddr;
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+	int ifindex;
+	int err = 0;
+	int notify_enetdown = 0;
+
+	if (len < sizeof(*addr))
+		return -EINVAL;
+
+	lock_sock(sk);
+
+	if (ro->bound && addr->can_ifindex == ro->ifindex)
+		goto out;
+
+	if (addr->can_ifindex) {
+		struct net_device *dev;
+
+		dev = dev_get_by_index(&init_net, addr->can_ifindex);
+		if (!dev) {
+			err = -ENODEV;
+			goto out;
+		}
+		if (dev->type != ARPHRD_CAN) {
+			dev_put(dev);
+			err = -ENODEV;
+			goto out;
+		}
+		if (!(dev->flags & IFF_UP))
+			notify_enetdown = 1;
+
+		ifindex = dev->ifindex;
+
+		/* filters set by default/setsockopt */
+		err = raw_enable_allfilters(dev, sk);
+		dev_put(dev);
+
+	} else {
+		ifindex = 0;
+
+		/* filters set by default/setsockopt */
+		err = raw_enable_allfilters(NULL, sk);
+	}
+
+	if (!err) {
+		if (ro->bound) {
+			/* unregister old filters */
+			if (ro->ifindex) {
+				struct net_device *dev;
+
+				dev = dev_get_by_index(&init_net, ro->ifindex);
+				if (dev) {
+					raw_disable_allfilters(dev, sk);
+					dev_put(dev);
+				}
+			} else
+				raw_disable_allfilters(NULL, sk);
+		}
+		ro->ifindex = ifindex;
+		ro->bound = 1;
+	}
+
+ out:
+	release_sock(sk);
+
+	if (notify_enetdown) {
+		sk->sk_err = ENETDOWN;
+		if (!sock_flag(sk, SOCK_DEAD))
+			sk->sk_error_report(sk);
+	}
+
+	return err;
+}
+
+static int raw_getname(struct socket *sock, struct sockaddr *uaddr,
+		       int *len, int peer)
+{
+	struct sockaddr_can *addr = (struct sockaddr_can *)uaddr;
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+
+	if (peer)
+		return -EOPNOTSUPP;
+
+	addr->can_family  = AF_CAN;
+	addr->can_ifindex = ro->ifindex;
+
+	*len = sizeof(*addr);
+
+	return 0;
+}
+
+static int raw_setsockopt(struct socket *sock, int level, int optname,
+			  char __user *optval, int optlen)
+{
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+	struct can_filter *filter = NULL;  /* dyn. alloc'ed filters */
+	struct can_filter sfilter;         /* single filter */
+	struct net_device *dev = NULL;
+	can_err_mask_t err_mask = 0;
+	int count = 0;
+	int err = 0;
+
+	if (level != SOL_CAN_RAW)
+		return -EINVAL;
+	if (optlen < 0)
+		return -EINVAL;
+
+	switch (optname) {
+
+	case CAN_RAW_FILTER:
+		if (optlen % sizeof(struct can_filter) != 0)
+			return -EINVAL;
+
+		count = optlen / sizeof(struct can_filter);
+
+		if (count > 1) {
+			/* filter does not fit into dfilter => alloc space */
+			filter = kmalloc(optlen, GFP_KERNEL);
+			if (!filter)
+				return -ENOMEM;
+
+			err = copy_from_user(filter, optval, optlen);
+			if (err) {
+				kfree(filter);
+				return err;
+			}
+		} else if (count == 1) {
+			err = copy_from_user(&sfilter, optval, optlen);
+			if (err)
+				return err;
+		}
+
+		lock_sock(sk);
+
+		if (ro->bound && ro->ifindex)
+			dev = dev_get_by_index(&init_net, ro->ifindex);
+
+		if (ro->bound) {
+			/* (try to) register the new filters */
+			if (count == 1)
+				err = raw_enable_filters(dev, sk, &sfilter, 1);
+			else
+				err = raw_enable_filters(dev, sk, filter,
+							 count);
+			if (err) {
+				if (count > 1)
+					kfree(filter);
+
+				goto out_fil;
+			}
+
+			/* remove old filter registrations */
+			raw_disable_filters(dev, sk, ro->filter, ro->count);
+		}
+
+		/* remove old filter space */
+		if (ro->count > 1)
+			kfree(ro->filter);
+
+		/* link new filters to the socket */
+		if (count == 1) {
+			/* copy filter data for single filter */
+			ro->dfilter = sfilter;
+			filter = &ro->dfilter;
+		}
+		ro->filter = filter;
+		ro->count  = count;
+
+ out_fil:
+		if (dev)
+			dev_put(dev);
+
+		release_sock(sk);
+
+		break;
+
+	case CAN_RAW_ERR_FILTER:
+		if (optlen != sizeof(err_mask))
+			return -EINVAL;
+
+		err = copy_from_user(&err_mask, optval, optlen);
+		if (err)
+			return err;
+
+		err_mask &= CAN_ERR_MASK;
+
+		lock_sock(sk);
+
+		if (ro->bound && ro->ifindex)
+			dev = dev_get_by_index(&init_net, ro->ifindex);
+
+		/* remove current error mask */
+		if (ro->bound) {
+			/* (try to) register the new err_mask */
+			err = raw_enable_errfilter(dev, sk, err_mask);
+
+			if (err)
+				goto out_err;
+
+			/* remove old err_mask registration */
+			raw_disable_errfilter(dev, sk, ro->err_mask);
+		}
+
+		/* link new err_mask to the socket */
+		ro->err_mask = err_mask;
+
+ out_err:
+		if (dev)
+			dev_put(dev);
+
+		release_sock(sk);
+
+		break;
+
+	case CAN_RAW_LOOPBACK:
+		if (optlen != sizeof(ro->loopback))
+			return -EINVAL;
+
+		err = copy_from_user(&ro->loopback, optval, optlen);
+
+		break;
+
+	case CAN_RAW_RECV_OWN_MSGS:
+		if (optlen != sizeof(ro->recv_own_msgs))
+			return -EINVAL;
+
+		err = copy_from_user(&ro->recv_own_msgs, optval, optlen);
+
+		break;
+
+	default:
+		return -ENOPROTOOPT;
+	}
+	return err;
+}
+
+static int raw_getsockopt(struct socket *sock, int level, int optname,
+			  char __user *optval, int __user *optlen)
+{
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+	int len;
+	void *val;
+	int err = 0;
+
+	if (level != SOL_CAN_RAW)
+		return -EINVAL;
+	if (get_user(len, optlen))
+		return -EFAULT;
+	if (len < 0)
+		return -EINVAL;
+
+	switch (optname) {
+
+	case CAN_RAW_FILTER:
+		lock_sock(sk);
+		if (ro->count > 0) {
+			int fsize = ro->count * sizeof(struct can_filter);
+			if (len > fsize)
+				len = fsize;
+			err = copy_to_user(optval, ro->filter, len);
+		} else
+			len = 0;
+		release_sock(sk);
+
+		if (!err)
+			err = put_user(len, optlen);
+		return err;
+
+	case CAN_RAW_ERR_FILTER:
+		if (len > sizeof(can_err_mask_t))
+			len = sizeof(can_err_mask_t);
+		val = &ro->err_mask;
+		break;
+
+	case CAN_RAW_LOOPBACK:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		val = &ro->loopback;
+		break;
+
+	case CAN_RAW_RECV_OWN_MSGS:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		val = &ro->recv_own_msgs;
+		break;
+
+	default:
+		return -ENOPROTOOPT;
+	}
+
+	if (put_user(len, optlen))
+		return -EFAULT;
+	if (copy_to_user(optval, val, len))
+		return -EFAULT;
+	return 0;
+}
+
+static int raw_sendmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, size_t size)
+{
+	struct sock *sk = sock->sk;
+	struct raw_sock *ro = raw_sk(sk);
+	struct sk_buff *skb;
+	struct net_device *dev;
+	int ifindex;
+	int err;
+
+	if (msg->msg_name) {
+		struct sockaddr_can *addr =
+			(struct sockaddr_can *)msg->msg_name;
+
+		if (addr->can_family != AF_CAN)
+			return -EINVAL;
+
+		ifindex = addr->can_ifindex;
+	} else
+		ifindex = ro->ifindex;
+
+	dev = dev_get_by_index(&init_net, ifindex);
+	if (!dev)
+		return -ENXIO;
+
+	skb = sock_alloc_send_skb(sk, size, msg->msg_flags & MSG_DONTWAIT,
+				  &err);
+	if (!skb) {
+		dev_put(dev);
+		return err;
+	}
+
+	err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size);
+	if (err < 0) {
+		kfree_skb(skb);
+		dev_put(dev);
+		return err;
+	}
+	skb->dev = dev;
+	skb->sk  = sk;
+
+	err = can_send(skb, ro->loopback);
+
+	dev_put(dev);
+
+	if (err)
+		return err;
+
+	return size;
+}
+
+static int raw_recvmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, size_t size, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct sk_buff *skb;
+	int error = 0;
+	int noblock;
+
+	noblock =  flags & MSG_DONTWAIT;
+	flags   &= ~MSG_DONTWAIT;
+
+	skb = skb_recv_datagram(sk, flags, noblock, &error);
+	if (!skb)
+		return error;
+
+	if (size < skb->len)
+		msg->msg_flags |= MSG_TRUNC;
+	else
+		size = skb->len;
+
+	error = memcpy_toiovec(msg->msg_iov, skb->data, size);
+	if (error < 0) {
+		skb_free_datagram(sk, skb);
+		return error;
+	}
+
+	sock_recv_timestamp(msg, sk, skb);
+
+	if (msg->msg_name) {
+		msg->msg_namelen = sizeof(struct sockaddr_can);
+		memcpy(msg->msg_name, skb->cb, msg->msg_namelen);
+	}
+
+	skb_free_datagram(sk, skb);
+
+	return size;
+}
+
+static struct proto_ops raw_ops __read_mostly = {
+	.family        = PF_CAN,
+	.release       = raw_release,
+	.bind          = raw_bind,
+	.connect       = sock_no_connect,
+	.socketpair    = sock_no_socketpair,
+	.accept        = sock_no_accept,
+	.getname       = raw_getname,
+	.poll          = datagram_poll,
+	.ioctl         = NULL,		/* use can_ioctl() from af_can.c */
+	.listen        = sock_no_listen,
+	.shutdown      = sock_no_shutdown,
+	.setsockopt    = raw_setsockopt,
+	.getsockopt    = raw_getsockopt,
+	.sendmsg       = raw_sendmsg,
+	.recvmsg       = raw_recvmsg,
+	.mmap          = sock_no_mmap,
+	.sendpage      = sock_no_sendpage,
+};
+
+static struct proto raw_proto __read_mostly = {
+	.name       = "CAN_RAW",
+	.owner      = THIS_MODULE,
+	.obj_size   = sizeof(struct raw_sock),
+	.init       = raw_init,
+};
+
+static struct can_proto raw_can_proto __read_mostly = {
+	.type       = SOCK_RAW,
+	.protocol   = CAN_RAW,
+	.capability = -1,
+	.ops        = &raw_ops,
+	.prot       = &raw_proto,
+};
+
+static __init int raw_module_init(void)
+{
+	int err;
+
+	printk(banner);
+
+	err = can_proto_register(&raw_can_proto);
+	if (err < 0)
+		printk(KERN_ERR "can: registration of raw protocol failed\n");
+
+	return err;
+}
+
+static __exit void raw_module_exit(void)
+{
+	can_proto_unregister(&raw_can_proto);
+}
+
+module_init(raw_module_init);
+module_exit(raw_module_exit);
-- 
cgit v1.2.3


From ffd980f976e7fd666c2e61bf8ab35107efd11828 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Date: Fri, 16 Nov 2007 15:53:52 -0800
Subject: [CAN]: Add broadcast manager (bcm) protocol

This patch adds the CAN broadcast manager (bcm) protocol.

Signed-off-by: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Signed-off-by: Urs Thuermann <urs.thuermann@volkswagen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/can/bcm.h |   65 ++
 net/can/Kconfig         |   13 +
 net/can/Makefile        |    3 +
 net/can/bcm.c           | 1561 +++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 1642 insertions(+)
 create mode 100644 include/linux/can/bcm.h
 create mode 100644 net/can/bcm.c

(limited to 'include')

diff --git a/include/linux/can/bcm.h b/include/linux/can/bcm.h
new file mode 100644
index 000000000000..7ade33a0ff03
--- /dev/null
+++ b/include/linux/can/bcm.h
@@ -0,0 +1,65 @@
+/*
+ * linux/can/bcm.h
+ *
+ * Definitions for CAN Broadcast Manager (BCM)
+ *
+ * Author: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#ifndef CAN_BCM_H
+#define CAN_BCM_H
+
+/**
+ * struct bcm_msg_head - head of messages to/from the broadcast manager
+ * @opcode:    opcode, see enum below.
+ * @flags:     special flags, see below.
+ * @count:     number of frames to send before changing interval.
+ * @ival1:     interval for the first @count frames.
+ * @ival2:     interval for the following frames.
+ * @can_id:    CAN ID of frames to be sent or received.
+ * @nframes:   number of frames appended to the message head.
+ * @frames:    array of CAN frames.
+ */
+struct bcm_msg_head {
+	int opcode;
+	int flags;
+	int count;
+	struct timeval ival1, ival2;
+	canid_t can_id;
+	int nframes;
+	struct can_frame frames[0];
+};
+
+enum {
+	TX_SETUP = 1,	/* create (cyclic) transmission task */
+	TX_DELETE,	/* remove (cyclic) transmission task */
+	TX_READ,	/* read properties of (cyclic) transmission task */
+	TX_SEND,	/* send one CAN frame */
+	RX_SETUP,	/* create RX content filter subscription */
+	RX_DELETE,	/* remove RX content filter subscription */
+	RX_READ,	/* read properties of RX content filter subscription */
+	TX_STATUS,	/* reply to TX_READ request */
+	TX_EXPIRED,	/* notification on performed transmissions (count=0) */
+	RX_STATUS,	/* reply to RX_READ request */
+	RX_TIMEOUT,	/* cyclic message is absent */
+	RX_CHANGED	/* updated CAN frame (detected content change) */
+};
+
+#define SETTIMER            0x0001
+#define STARTTIMER          0x0002
+#define TX_COUNTEVT         0x0004
+#define TX_ANNOUNCE         0x0008
+#define TX_CP_CAN_ID        0x0010
+#define RX_FILTER_ID        0x0020
+#define RX_CHECK_DLC        0x0040
+#define RX_NO_AUTOTIMER     0x0080
+#define RX_ANNOUNCE_RESUME  0x0100
+#define TX_RESET_MULTI_IDX  0x0200
+#define RX_RTR_FRAME        0x0400
+
+#endif /* CAN_BCM_H */
diff --git a/net/can/Kconfig b/net/can/Kconfig
index 4718d1f50ab3..182b96b80ebc 100644
--- a/net/can/Kconfig
+++ b/net/can/Kconfig
@@ -26,3 +26,16 @@ config CAN_RAW
 	  most cases where no higher level protocol is being used. The raw
 	  socket has several filter options e.g. ID masking / error frames.
 	  To receive/send raw CAN messages, use AF_CAN with protocol CAN_RAW.
+
+config CAN_BCM
+	tristate "Broadcast Manager CAN Protocol (with content filtering)"
+	depends on CAN
+	default N
+	---help---
+	  The Broadcast Manager offers content filtering, timeout monitoring,
+	  sending of RTR frames, and cyclic CAN messages without permanent user
+	  interaction. The BCM can be 'programmed' via the BSD socket API and
+	  informs you on demand e.g. only on content updates / timeouts.
+	  You probably want to use the bcm socket in most cases where cyclic
+	  CAN messages are used on the bus (e.g. in automotive environments).
+	  To use the Broadcast Manager, use AF_CAN with protocol CAN_BCM.
diff --git a/net/can/Makefile b/net/can/Makefile
index 86f1cf21fce4..9cd3c4b3abda 100644
--- a/net/can/Makefile
+++ b/net/can/Makefile
@@ -7,3 +7,6 @@ can-objs		:= af_can.o proc.o
 
 obj-$(CONFIG_CAN_RAW)	+= can-raw.o
 can-raw-objs		:= raw.o
+
+obj-$(CONFIG_CAN_BCM)	+= can-bcm.o
+can-bcm-objs		:= bcm.o
diff --git a/net/can/bcm.c b/net/can/bcm.c
new file mode 100644
index 000000000000..bd4282dae754
--- /dev/null
+++ b/net/can/bcm.c
@@ -0,0 +1,1561 @@
+/*
+ * bcm.c - Broadcast Manager to filter/send (cyclic) CAN content
+ *
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ * Send feedback to <socketcan-users@lists.berlios.de>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/proc_fs.h>
+#include <linux/uio.h>
+#include <linux/net.h>
+#include <linux/netdevice.h>
+#include <linux/socket.h>
+#include <linux/if_arp.h>
+#include <linux/skbuff.h>
+#include <linux/can.h>
+#include <linux/can/core.h>
+#include <linux/can/bcm.h>
+#include <net/sock.h>
+#include <net/net_namespace.h>
+
+/* use of last_frames[index].can_dlc */
+#define RX_RECV    0x40 /* received data for this element */
+#define RX_THR     0x80 /* element not been sent due to throttle feature */
+#define BCM_CAN_DLC_MASK 0x0F /* clean private flags in can_dlc by masking */
+
+/* get best masking value for can_rx_register() for a given single can_id */
+#define REGMASK(id) ((id & CAN_RTR_FLAG) | ((id & CAN_EFF_FLAG) ? \
+			(CAN_EFF_MASK | CAN_EFF_FLAG) : CAN_SFF_MASK))
+
+#define CAN_BCM_VERSION CAN_VERSION
+static __initdata const char banner[] = KERN_INFO
+	"can: broadcast manager protocol (rev " CAN_BCM_VERSION ")\n";
+
+MODULE_DESCRIPTION("PF_CAN broadcast manager protocol");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Oliver Hartkopp <oliver.hartkopp@volkswagen.de>");
+
+/* easy access to can_frame payload */
+static inline u64 GET_U64(const struct can_frame *cp)
+{
+	return *(u64 *)cp->data;
+}
+
+struct bcm_op {
+	struct list_head list;
+	int ifindex;
+	canid_t can_id;
+	int flags;
+	unsigned long j_ival1, j_ival2, j_lastmsg;
+	unsigned long frames_abs, frames_filtered;
+	struct timer_list timer, thrtimer;
+	struct timeval ival1, ival2;
+	ktime_t rx_stamp;
+	int rx_ifindex;
+	int count;
+	int nframes;
+	int currframe;
+	struct can_frame *frames;
+	struct can_frame *last_frames;
+	struct can_frame sframe;
+	struct can_frame last_sframe;
+	struct sock *sk;
+	struct net_device *rx_reg_dev;
+};
+
+static struct proc_dir_entry *proc_dir;
+
+struct bcm_sock {
+	struct sock sk;
+	int bound;
+	int ifindex;
+	struct notifier_block notifier;
+	struct list_head rx_ops;
+	struct list_head tx_ops;
+	unsigned long dropped_usr_msgs;
+	struct proc_dir_entry *bcm_proc_read;
+	char procname [9]; /* pointer printed in ASCII with \0 */
+};
+
+static inline struct bcm_sock *bcm_sk(const struct sock *sk)
+{
+	return (struct bcm_sock *)sk;
+}
+
+#define CFSIZ sizeof(struct can_frame)
+#define OPSIZ sizeof(struct bcm_op)
+#define MHSIZ sizeof(struct bcm_msg_head)
+
+/*
+ * rounded_tv2jif - calculate jiffies from timeval including optional up
+ * @tv: pointer to timeval
+ *
+ * Description:
+ * Unlike timeval_to_jiffies() provided in include/linux/jiffies.h, this
+ * function is intentionally more relaxed on precise timer ticks to get
+ * exact one jiffy for requested 1000us on a 1000HZ machine.
+ * This code is to be removed when upgrading to kernel hrtimer.
+ *
+ * Return:
+ *  calculated jiffies (max: ULONG_MAX)
+ */
+static unsigned long rounded_tv2jif(const struct timeval *tv)
+{
+	unsigned long sec  = tv->tv_sec;
+	unsigned long usec = tv->tv_usec;
+	unsigned long jif;
+
+	if (sec > ULONG_MAX / HZ)
+		return ULONG_MAX;
+
+	/* round up to get at least the requested time */
+	usec += 1000000 / HZ - 1;
+
+	jif  = usec / (1000000 / HZ);
+
+	if (sec * HZ > ULONG_MAX - jif)
+		return ULONG_MAX;
+
+	return jif + sec * HZ;
+}
+
+/*
+ * procfs functions
+ */
+static char *bcm_proc_getifname(int ifindex)
+{
+	struct net_device *dev;
+
+	if (!ifindex)
+		return "any";
+
+	/* no usage counting */
+	dev = __dev_get_by_index(&init_net, ifindex);
+	if (dev)
+		return dev->name;
+
+	return "???";
+}
+
+static int bcm_read_proc(char *page, char **start, off_t off,
+			 int count, int *eof, void *data)
+{
+	int len = 0;
+	struct sock *sk = (struct sock *)data;
+	struct bcm_sock *bo = bcm_sk(sk);
+	struct bcm_op *op;
+
+	len += snprintf(page + len, PAGE_SIZE - len, ">>> socket %p",
+			sk->sk_socket);
+	len += snprintf(page + len, PAGE_SIZE - len, " / sk %p", sk);
+	len += snprintf(page + len, PAGE_SIZE - len, " / bo %p", bo);
+	len += snprintf(page + len, PAGE_SIZE - len, " / dropped %lu",
+			bo->dropped_usr_msgs);
+	len += snprintf(page + len, PAGE_SIZE - len, " / bound %s",
+			bcm_proc_getifname(bo->ifindex));
+	len += snprintf(page + len, PAGE_SIZE - len, " <<<\n");
+
+	list_for_each_entry(op, &bo->rx_ops, list) {
+
+		unsigned long reduction;
+
+		/* print only active entries & prevent division by zero */
+		if (!op->frames_abs)
+			continue;
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"rx_op: %03X %-5s ",
+				op->can_id, bcm_proc_getifname(op->ifindex));
+		len += snprintf(page + len, PAGE_SIZE - len, "[%d]%c ",
+				op->nframes,
+				(op->flags & RX_CHECK_DLC)?'d':' ');
+		if (op->j_ival1)
+			len += snprintf(page + len, PAGE_SIZE - len,
+					"timeo=%ld ", op->j_ival1);
+
+		if (op->j_ival2)
+			len += snprintf(page + len, PAGE_SIZE - len,
+					"thr=%ld ", op->j_ival2);
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"# recv %ld (%ld) => reduction: ",
+				op->frames_filtered, op->frames_abs);
+
+		reduction = 100 - (op->frames_filtered * 100) / op->frames_abs;
+
+		len += snprintf(page + len, PAGE_SIZE - len, "%s%ld%%\n",
+				(reduction == 100)?"near ":"", reduction);
+
+		if (len > PAGE_SIZE - 200) {
+			/* mark output cut off */
+			len += snprintf(page + len, PAGE_SIZE - len, "(..)\n");
+			break;
+		}
+	}
+
+	list_for_each_entry(op, &bo->tx_ops, list) {
+
+		len += snprintf(page + len, PAGE_SIZE - len,
+				"tx_op: %03X %s [%d] ",
+				op->can_id, bcm_proc_getifname(op->ifindex),
+				op->nframes);
+		if (op->j_ival1)
+			len += snprintf(page + len, PAGE_SIZE - len, "t1=%ld ",
+					op->j_ival1);
+
+		if (op->j_ival2)
+			len += snprintf(page + len, PAGE_SIZE - len, "t2=%ld ",
+					op->j_ival2);
+
+		len += snprintf(page + len, PAGE_SIZE - len, "# sent %ld\n",
+				op->frames_abs);
+
+		if (len > PAGE_SIZE - 100) {
+			/* mark output cut off */
+			len += snprintf(page + len, PAGE_SIZE - len, "(..)\n");
+			break;
+		}
+	}
+
+	len += snprintf(page + len, PAGE_SIZE - len, "\n");
+
+	*eof = 1;
+	return len;
+}
+
+/*
+ * bcm_can_tx - send the (next) CAN frame to the appropriate CAN interface
+ *              of the given bcm tx op
+ */
+static void bcm_can_tx(struct bcm_op *op)
+{
+	struct sk_buff *skb;
+	struct net_device *dev;
+	struct can_frame *cf = &op->frames[op->currframe];
+
+	/* no target device? => exit */
+	if (!op->ifindex)
+		return;
+
+	dev = dev_get_by_index(&init_net, op->ifindex);
+	if (!dev) {
+		/* RFC: should this bcm_op remove itself here? */
+		return;
+	}
+
+	skb = alloc_skb(CFSIZ, gfp_any());
+	if (!skb)
+		goto out;
+
+	memcpy(skb_put(skb, CFSIZ), cf, CFSIZ);
+
+	/* send with loopback */
+	skb->dev = dev;
+	skb->sk = op->sk;
+	can_send(skb, 1);
+
+	/* update statistics */
+	op->currframe++;
+	op->frames_abs++;
+
+	/* reached last frame? */
+	if (op->currframe >= op->nframes)
+		op->currframe = 0;
+ out:
+	dev_put(dev);
+}
+
+/*
+ * bcm_send_to_user - send a BCM message to the userspace
+ *                    (consisting of bcm_msg_head + x CAN frames)
+ */
+static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head,
+			     struct can_frame *frames, int has_timestamp)
+{
+	struct sk_buff *skb;
+	struct can_frame *firstframe;
+	struct sockaddr_can *addr;
+	struct sock *sk = op->sk;
+	int datalen = head->nframes * CFSIZ;
+	int err;
+
+	skb = alloc_skb(sizeof(*head) + datalen, gfp_any());
+	if (!skb)
+		return;
+
+	memcpy(skb_put(skb, sizeof(*head)), head, sizeof(*head));
+
+	if (head->nframes) {
+		/* can_frames starting here */
+		firstframe = (struct can_frame *) skb_tail_pointer(skb);
+
+		memcpy(skb_put(skb, datalen), frames, datalen);
+
+		/*
+		 * the BCM uses the can_dlc-element of the can_frame
+		 * structure for internal purposes. This is only
+		 * relevant for updates that are generated by the
+		 * BCM, where nframes is 1
+		 */
+		if (head->nframes == 1)
+			firstframe->can_dlc &= BCM_CAN_DLC_MASK;
+	}
+
+	if (has_timestamp) {
+		/* restore rx timestamp */
+		skb->tstamp = op->rx_stamp;
+	}
+
+	/*
+	 *  Put the datagram to the queue so that bcm_recvmsg() can
+	 *  get it from there.  We need to pass the interface index to
+	 *  bcm_recvmsg().  We pass a whole struct sockaddr_can in skb->cb
+	 *  containing the interface index.
+	 */
+
+	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct sockaddr_can));
+	addr = (struct sockaddr_can *)skb->cb;
+	memset(addr, 0, sizeof(*addr));
+	addr->can_family  = AF_CAN;
+	addr->can_ifindex = op->rx_ifindex;
+
+	err = sock_queue_rcv_skb(sk, skb);
+	if (err < 0) {
+		struct bcm_sock *bo = bcm_sk(sk);
+
+		kfree_skb(skb);
+		/* don't care about overflows in this statistic */
+		bo->dropped_usr_msgs++;
+	}
+}
+
+/*
+ * bcm_tx_timeout_handler - performes cyclic CAN frame transmissions
+ */
+static void bcm_tx_timeout_handler(unsigned long data)
+{
+	struct bcm_op *op = (struct bcm_op *)data;
+
+	if (op->j_ival1 && (op->count > 0)) {
+
+		op->count--;
+		if (!op->count && (op->flags & TX_COUNTEVT)) {
+			struct bcm_msg_head msg_head;
+
+			/* create notification to user */
+			msg_head.opcode  = TX_EXPIRED;
+			msg_head.flags   = op->flags;
+			msg_head.count   = op->count;
+			msg_head.ival1   = op->ival1;
+			msg_head.ival2   = op->ival2;
+			msg_head.can_id  = op->can_id;
+			msg_head.nframes = 0;
+
+			bcm_send_to_user(op, &msg_head, NULL, 0);
+		}
+	}
+
+	if (op->j_ival1 && (op->count > 0)) {
+
+		/* send (next) frame */
+		bcm_can_tx(op);
+		mod_timer(&op->timer, jiffies + op->j_ival1);
+
+	} else {
+		if (op->j_ival2) {
+
+			/* send (next) frame */
+			bcm_can_tx(op);
+			mod_timer(&op->timer, jiffies + op->j_ival2);
+		}
+	}
+
+	return;
+}
+
+/*
+ * bcm_rx_changed - create a RX_CHANGED notification due to changed content
+ */
+static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data)
+{
+	struct bcm_msg_head head;
+
+	op->j_lastmsg = jiffies;
+
+	/* update statistics */
+	op->frames_filtered++;
+
+	/* prevent statistics overflow */
+	if (op->frames_filtered > ULONG_MAX/100)
+		op->frames_filtered = op->frames_abs = 0;
+
+	head.opcode  = RX_CHANGED;
+	head.flags   = op->flags;
+	head.count   = op->count;
+	head.ival1   = op->ival1;
+	head.ival2   = op->ival2;
+	head.can_id  = op->can_id;
+	head.nframes = 1;
+
+	bcm_send_to_user(op, &head, data, 1);
+}
+
+/*
+ * bcm_rx_update_and_send - process a detected relevant receive content change
+ *                          1. update the last received data
+ *                          2. send a notification to the user (if possible)
+ */
+static void bcm_rx_update_and_send(struct bcm_op *op,
+				   struct can_frame *lastdata,
+				   struct can_frame *rxdata)
+{
+	unsigned long nexttx = op->j_lastmsg + op->j_ival2;
+
+	memcpy(lastdata, rxdata, CFSIZ);
+
+	/* mark as used */
+	lastdata->can_dlc |= RX_RECV;
+
+	/* throttle bcm_rx_changed ? */
+	if ((op->thrtimer.expires) ||
+	    ((op->j_ival2) && (nexttx > jiffies))) {
+		/* we are already waiting OR we have to start waiting */
+
+		/* mark as 'throttled' */
+		lastdata->can_dlc |= RX_THR;
+
+		if (!(op->thrtimer.expires)) {
+			/* start the timer only the first time */
+			mod_timer(&op->thrtimer, nexttx);
+		}
+
+	} else {
+		/* send RX_CHANGED to the user immediately */
+		bcm_rx_changed(op, rxdata);
+	}
+}
+
+/*
+ * bcm_rx_cmp_to_index - (bit)compares the currently received data to formerly
+ *                       received data stored in op->last_frames[]
+ */
+static void bcm_rx_cmp_to_index(struct bcm_op *op, int index,
+				struct can_frame *rxdata)
+{
+	/*
+	 * no one uses the MSBs of can_dlc for comparation,
+	 * so we use it here to detect the first time of reception
+	 */
+
+	if (!(op->last_frames[index].can_dlc & RX_RECV)) {
+		/* received data for the first time => send update to user */
+		bcm_rx_update_and_send(op, &op->last_frames[index], rxdata);
+		return;
+	}
+
+	/* do a real check in can_frame data section */
+
+	if ((GET_U64(&op->frames[index]) & GET_U64(rxdata)) !=
+	    (GET_U64(&op->frames[index]) & GET_U64(&op->last_frames[index]))) {
+		bcm_rx_update_and_send(op, &op->last_frames[index], rxdata);
+		return;
+	}
+
+	if (op->flags & RX_CHECK_DLC) {
+		/* do a real check in can_frame dlc */
+		if (rxdata->can_dlc != (op->last_frames[index].can_dlc &
+					BCM_CAN_DLC_MASK)) {
+			bcm_rx_update_and_send(op, &op->last_frames[index],
+					       rxdata);
+			return;
+		}
+	}
+}
+
+/*
+ * bcm_rx_starttimer - enable timeout monitoring for CAN frame receiption
+ */
+static void bcm_rx_starttimer(struct bcm_op *op)
+{
+	if (op->flags & RX_NO_AUTOTIMER)
+		return;
+
+	if (op->j_ival1)
+		mod_timer(&op->timer, jiffies + op->j_ival1);
+}
+
+/*
+ * bcm_rx_timeout_handler - when the (cyclic) CAN frame receiption timed out
+ */
+static void bcm_rx_timeout_handler(unsigned long data)
+{
+	struct bcm_op *op = (struct bcm_op *)data;
+	struct bcm_msg_head msg_head;
+
+	msg_head.opcode  = RX_TIMEOUT;
+	msg_head.flags   = op->flags;
+	msg_head.count   = op->count;
+	msg_head.ival1   = op->ival1;
+	msg_head.ival2   = op->ival2;
+	msg_head.can_id  = op->can_id;
+	msg_head.nframes = 0;
+
+	bcm_send_to_user(op, &msg_head, NULL, 0);
+
+	/* no restart of the timer is done here! */
+
+	/* if user wants to be informed, when cyclic CAN-Messages come back */
+	if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) {
+		/* clear received can_frames to indicate 'nothing received' */
+		memset(op->last_frames, 0, op->nframes * CFSIZ);
+	}
+}
+
+/*
+ * bcm_rx_thr_handler - the time for blocked content updates is over now:
+ *                      Check for throttled data and send it to the userspace
+ */
+static void bcm_rx_thr_handler(unsigned long data)
+{
+	struct bcm_op *op = (struct bcm_op *)data;
+	int i = 0;
+
+	/* mark disabled / consumed timer */
+	op->thrtimer.expires = 0;
+
+	if (op->nframes > 1) {
+		/* for MUX filter we start at index 1 */
+		for (i = 1; i < op->nframes; i++) {
+			if ((op->last_frames) &&
+			    (op->last_frames[i].can_dlc & RX_THR)) {
+				op->last_frames[i].can_dlc &= ~RX_THR;
+				bcm_rx_changed(op, &op->last_frames[i]);
+			}
+		}
+
+	} else {
+		/* for RX_FILTER_ID and simple filter */
+		if (op->last_frames && (op->last_frames[0].can_dlc & RX_THR)) {
+			op->last_frames[0].can_dlc &= ~RX_THR;
+			bcm_rx_changed(op, &op->last_frames[0]);
+		}
+	}
+}
+
+/*
+ * bcm_rx_handler - handle a CAN frame receiption
+ */
+static void bcm_rx_handler(struct sk_buff *skb, void *data)
+{
+	struct bcm_op *op = (struct bcm_op *)data;
+	struct can_frame rxframe;
+	int i;
+
+	/* disable timeout */
+	del_timer(&op->timer);
+
+	if (skb->len == sizeof(rxframe)) {
+		memcpy(&rxframe, skb->data, sizeof(rxframe));
+		/* save rx timestamp */
+		op->rx_stamp = skb->tstamp;
+		/* save originator for recvfrom() */
+		op->rx_ifindex = skb->dev->ifindex;
+		/* update statistics */
+		op->frames_abs++;
+		kfree_skb(skb);
+
+	} else {
+		kfree_skb(skb);
+		return;
+	}
+
+	if (op->can_id != rxframe.can_id)
+		return;
+
+	if (op->flags & RX_RTR_FRAME) {
+		/* send reply for RTR-request (placed in op->frames[0]) */
+		bcm_can_tx(op);
+		return;
+	}
+
+	if (op->flags & RX_FILTER_ID) {
+		/* the easiest case */
+		bcm_rx_update_and_send(op, &op->last_frames[0], &rxframe);
+		bcm_rx_starttimer(op);
+		return;
+	}
+
+	if (op->nframes == 1) {
+		/* simple compare with index 0 */
+		bcm_rx_cmp_to_index(op, 0, &rxframe);
+		bcm_rx_starttimer(op);
+		return;
+	}
+
+	if (op->nframes > 1) {
+		/*
+		 * multiplex compare
+		 *
+		 * find the first multiplex mask that fits.
+		 * Remark: The MUX-mask is stored in index 0
+		 */
+
+		for (i = 1; i < op->nframes; i++) {
+			if ((GET_U64(&op->frames[0]) & GET_U64(&rxframe)) ==
+			    (GET_U64(&op->frames[0]) &
+			     GET_U64(&op->frames[i]))) {
+				bcm_rx_cmp_to_index(op, i, &rxframe);
+				break;
+			}
+		}
+		bcm_rx_starttimer(op);
+	}
+}
+
+/*
+ * helpers for bcm_op handling: find & delete bcm [rx|tx] op elements
+ */
+static struct bcm_op *bcm_find_op(struct list_head *ops, canid_t can_id,
+				  int ifindex)
+{
+	struct bcm_op *op;
+
+	list_for_each_entry(op, ops, list) {
+		if ((op->can_id == can_id) && (op->ifindex == ifindex))
+			return op;
+	}
+
+	return NULL;
+}
+
+static void bcm_remove_op(struct bcm_op *op)
+{
+	del_timer(&op->timer);
+	del_timer(&op->thrtimer);
+
+	if ((op->frames) && (op->frames != &op->sframe))
+		kfree(op->frames);
+
+	if ((op->last_frames) && (op->last_frames != &op->last_sframe))
+		kfree(op->last_frames);
+
+	kfree(op);
+
+	return;
+}
+
+static void bcm_rx_unreg(struct net_device *dev, struct bcm_op *op)
+{
+	if (op->rx_reg_dev == dev) {
+		can_rx_unregister(dev, op->can_id, REGMASK(op->can_id),
+				  bcm_rx_handler, op);
+
+		/* mark as removed subscription */
+		op->rx_reg_dev = NULL;
+	} else
+		printk(KERN_ERR "can-bcm: bcm_rx_unreg: registered device "
+		       "mismatch %p %p\n", op->rx_reg_dev, dev);
+}
+
+/*
+ * bcm_delete_rx_op - find and remove a rx op (returns number of removed ops)
+ */
+static int bcm_delete_rx_op(struct list_head *ops, canid_t can_id, int ifindex)
+{
+	struct bcm_op *op, *n;
+
+	list_for_each_entry_safe(op, n, ops, list) {
+		if ((op->can_id == can_id) && (op->ifindex == ifindex)) {
+
+			/*
+			 * Don't care if we're bound or not (due to netdev
+			 * problems) can_rx_unregister() is always a save
+			 * thing to do here.
+			 */
+			if (op->ifindex) {
+				/*
+				 * Only remove subscriptions that had not
+				 * been removed due to NETDEV_UNREGISTER
+				 * in bcm_notifier()
+				 */
+				if (op->rx_reg_dev) {
+					struct net_device *dev;
+
+					dev = dev_get_by_index(&init_net,
+							       op->ifindex);
+					if (dev) {
+						bcm_rx_unreg(dev, op);
+						dev_put(dev);
+					}
+				}
+			} else
+				can_rx_unregister(NULL, op->can_id,
+						  REGMASK(op->can_id),
+						  bcm_rx_handler, op);
+
+			list_del(&op->list);
+			bcm_remove_op(op);
+			return 1; /* done */
+		}
+	}
+
+	return 0; /* not found */
+}
+
+/*
+ * bcm_delete_tx_op - find and remove a tx op (returns number of removed ops)
+ */
+static int bcm_delete_tx_op(struct list_head *ops, canid_t can_id, int ifindex)
+{
+	struct bcm_op *op, *n;
+
+	list_for_each_entry_safe(op, n, ops, list) {
+		if ((op->can_id == can_id) && (op->ifindex == ifindex)) {
+			list_del(&op->list);
+			bcm_remove_op(op);
+			return 1; /* done */
+		}
+	}
+
+	return 0; /* not found */
+}
+
+/*
+ * bcm_read_op - read out a bcm_op and send it to the user (for bcm_sendmsg)
+ */
+static int bcm_read_op(struct list_head *ops, struct bcm_msg_head *msg_head,
+		       int ifindex)
+{
+	struct bcm_op *op = bcm_find_op(ops, msg_head->can_id, ifindex);
+
+	if (!op)
+		return -EINVAL;
+
+	/* put current values into msg_head */
+	msg_head->flags   = op->flags;
+	msg_head->count   = op->count;
+	msg_head->ival1   = op->ival1;
+	msg_head->ival2   = op->ival2;
+	msg_head->nframes = op->nframes;
+
+	bcm_send_to_user(op, msg_head, op->frames, 0);
+
+	return MHSIZ;
+}
+
+/*
+ * bcm_tx_setup - create or update a bcm tx op (for bcm_sendmsg)
+ */
+static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg,
+			int ifindex, struct sock *sk)
+{
+	struct bcm_sock *bo = bcm_sk(sk);
+	struct bcm_op *op;
+	int i, err;
+
+	/* we need a real device to send frames */
+	if (!ifindex)
+		return -ENODEV;
+
+	/* we need at least one can_frame */
+	if (msg_head->nframes < 1)
+		return -EINVAL;
+
+	/* check the given can_id */
+	op = bcm_find_op(&bo->tx_ops, msg_head->can_id, ifindex);
+
+	if (op) {
+		/* update existing BCM operation */
+
+		/*
+		 * Do we need more space for the can_frames than currently
+		 * allocated? -> This is a _really_ unusual use-case and
+		 * therefore (complexity / locking) it is not supported.
+		 */
+		if (msg_head->nframes > op->nframes)
+			return -E2BIG;
+
+		/* update can_frames content */
+		for (i = 0; i < msg_head->nframes; i++) {
+			err = memcpy_fromiovec((u8 *)&op->frames[i],
+					       msg->msg_iov, CFSIZ);
+			if (err < 0)
+				return err;
+
+			if (msg_head->flags & TX_CP_CAN_ID) {
+				/* copy can_id into frame */
+				op->frames[i].can_id = msg_head->can_id;
+			}
+		}
+
+	} else {
+		/* insert new BCM operation for the given can_id */
+
+		op = kzalloc(OPSIZ, GFP_KERNEL);
+		if (!op)
+			return -ENOMEM;
+
+		op->can_id    = msg_head->can_id;
+
+		/* create array for can_frames and copy the data */
+		if (msg_head->nframes > 1) {
+			op->frames = kmalloc(msg_head->nframes * CFSIZ,
+					     GFP_KERNEL);
+			if (!op->frames) {
+				kfree(op);
+				return -ENOMEM;
+			}
+		} else
+			op->frames = &op->sframe;
+
+		for (i = 0; i < msg_head->nframes; i++) {
+			err = memcpy_fromiovec((u8 *)&op->frames[i],
+					       msg->msg_iov, CFSIZ);
+			if (err < 0) {
+				if (op->frames != &op->sframe)
+					kfree(op->frames);
+				kfree(op);
+				return err;
+			}
+
+			if (msg_head->flags & TX_CP_CAN_ID) {
+				/* copy can_id into frame */
+				op->frames[i].can_id = msg_head->can_id;
+			}
+		}
+
+		/* tx_ops never compare with previous received messages */
+		op->last_frames = NULL;
+
+		/* bcm_can_tx / bcm_tx_timeout_handler needs this */
+		op->sk = sk;
+		op->ifindex = ifindex;
+
+		/* initialize uninitialized (kzalloc) structure */
+		setup_timer(&op->timer, bcm_tx_timeout_handler,
+			    (unsigned long)op);
+
+		/* currently unused in tx_ops */
+		init_timer(&op->thrtimer);
+
+		/* add this bcm_op to the list of the tx_ops */
+		list_add(&op->list, &bo->tx_ops);
+
+	} /* if ((op = bcm_find_op(&bo->tx_ops, msg_head->can_id, ifindex))) */
+
+	if (op->nframes != msg_head->nframes) {
+		op->nframes   = msg_head->nframes;
+		/* start multiple frame transmission with index 0 */
+		op->currframe = 0;
+	}
+
+	/* check flags */
+
+	op->flags = msg_head->flags;
+
+	if (op->flags & TX_RESET_MULTI_IDX) {
+		/* start multiple frame transmission with index 0 */
+		op->currframe = 0;
+	}
+
+	if (op->flags & SETTIMER) {
+		/* set timer values */
+		op->count = msg_head->count;
+		op->ival1 = msg_head->ival1;
+		op->ival2 = msg_head->ival2;
+		op->j_ival1 = rounded_tv2jif(&msg_head->ival1);
+		op->j_ival2 = rounded_tv2jif(&msg_head->ival2);
+
+		/* disable an active timer due to zero values? */
+		if (!op->j_ival1 && !op->j_ival2)
+			del_timer(&op->timer);
+	}
+
+	if ((op->flags & STARTTIMER) &&
+	    ((op->j_ival1 && op->count) || op->j_ival2)) {
+
+		/* spec: send can_frame when starting timer */
+		op->flags |= TX_ANNOUNCE;
+
+		if (op->j_ival1 && (op->count > 0)) {
+			/* op->count-- is done in bcm_tx_timeout_handler */
+			mod_timer(&op->timer, jiffies + op->j_ival1);
+		} else
+			mod_timer(&op->timer, jiffies + op->j_ival2);
+	}
+
+	if (op->flags & TX_ANNOUNCE)
+		bcm_can_tx(op);
+
+	return msg_head->nframes * CFSIZ + MHSIZ;
+}
+
+/*
+ * bcm_rx_setup - create or update a bcm rx op (for bcm_sendmsg)
+ */
+static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg,
+			int ifindex, struct sock *sk)
+{
+	struct bcm_sock *bo = bcm_sk(sk);
+	struct bcm_op *op;
+	int do_rx_register;
+	int err = 0;
+
+	if ((msg_head->flags & RX_FILTER_ID) || (!(msg_head->nframes))) {
+		/* be robust against wrong usage ... */
+		msg_head->flags |= RX_FILTER_ID;
+		/* ignore trailing garbage */
+		msg_head->nframes = 0;
+	}
+
+	if ((msg_head->flags & RX_RTR_FRAME) &&
+	    ((msg_head->nframes != 1) ||
+	     (!(msg_head->can_id & CAN_RTR_FLAG))))
+		return -EINVAL;
+
+	/* check the given can_id */
+	op = bcm_find_op(&bo->rx_ops, msg_head->can_id, ifindex);
+	if (op) {
+		/* update existing BCM operation */
+
+		/*
+		 * Do we need more space for the can_frames than currently
+		 * allocated? -> This is a _really_ unusual use-case and
+		 * therefore (complexity / locking) it is not supported.
+		 */
+		if (msg_head->nframes > op->nframes)
+			return -E2BIG;
+
+		if (msg_head->nframes) {
+			/* update can_frames content */
+			err = memcpy_fromiovec((u8 *)op->frames,
+					       msg->msg_iov,
+					       msg_head->nframes * CFSIZ);
+			if (err < 0)
+				return err;
+
+			/* clear last_frames to indicate 'nothing received' */
+			memset(op->last_frames, 0, msg_head->nframes * CFSIZ);
+		}
+
+		op->nframes = msg_head->nframes;
+
+		/* Only an update -> do not call can_rx_register() */
+		do_rx_register = 0;
+
+	} else {
+		/* insert new BCM operation for the given can_id */
+		op = kzalloc(OPSIZ, GFP_KERNEL);
+		if (!op)
+			return -ENOMEM;
+
+		op->can_id    = msg_head->can_id;
+		op->nframes   = msg_head->nframes;
+
+		if (msg_head->nframes > 1) {
+			/* create array for can_frames and copy the data */
+			op->frames = kmalloc(msg_head->nframes * CFSIZ,
+					     GFP_KERNEL);
+			if (!op->frames) {
+				kfree(op);
+				return -ENOMEM;
+			}
+
+			/* create and init array for received can_frames */
+			op->last_frames = kzalloc(msg_head->nframes * CFSIZ,
+						  GFP_KERNEL);
+			if (!op->last_frames) {
+				kfree(op->frames);
+				kfree(op);
+				return -ENOMEM;
+			}
+
+		} else {
+			op->frames = &op->sframe;
+			op->last_frames = &op->last_sframe;
+		}
+
+		if (msg_head->nframes) {
+			err = memcpy_fromiovec((u8 *)op->frames, msg->msg_iov,
+					       msg_head->nframes * CFSIZ);
+			if (err < 0) {
+				if (op->frames != &op->sframe)
+					kfree(op->frames);
+				if (op->last_frames != &op->last_sframe)
+					kfree(op->last_frames);
+				kfree(op);
+				return err;
+			}
+		}
+
+		/* bcm_can_tx / bcm_tx_timeout_handler needs this */
+		op->sk = sk;
+		op->ifindex = ifindex;
+
+		/* initialize uninitialized (kzalloc) structure */
+		setup_timer(&op->timer, bcm_rx_timeout_handler,
+			    (unsigned long)op);
+
+		/* init throttle timer for RX_CHANGED */
+		setup_timer(&op->thrtimer, bcm_rx_thr_handler,
+			    (unsigned long)op);
+
+		/* mark disabled timer */
+		op->thrtimer.expires = 0;
+
+		/* add this bcm_op to the list of the rx_ops */
+		list_add(&op->list, &bo->rx_ops);
+
+		/* call can_rx_register() */
+		do_rx_register = 1;
+
+	} /* if ((op = bcm_find_op(&bo->rx_ops, msg_head->can_id, ifindex))) */
+
+	/* check flags */
+	op->flags = msg_head->flags;
+
+	if (op->flags & RX_RTR_FRAME) {
+
+		/* no timers in RTR-mode */
+		del_timer(&op->thrtimer);
+		del_timer(&op->timer);
+
+		/*
+		 * funny feature in RX(!)_SETUP only for RTR-mode:
+		 * copy can_id into frame BUT without RTR-flag to
+		 * prevent a full-load-loopback-test ... ;-]
+		 */
+		if ((op->flags & TX_CP_CAN_ID) ||
+		    (op->frames[0].can_id == op->can_id))
+			op->frames[0].can_id = op->can_id & ~CAN_RTR_FLAG;
+
+	} else {
+		if (op->flags & SETTIMER) {
+
+			/* set timer value */
+			op->ival1 = msg_head->ival1;
+			op->ival2 = msg_head->ival2;
+			op->j_ival1 = rounded_tv2jif(&msg_head->ival1);
+			op->j_ival2 = rounded_tv2jif(&msg_head->ival2);
+
+			/* disable an active timer due to zero value? */
+			if (!op->j_ival1)
+				del_timer(&op->timer);
+
+			/* free currently blocked msgs ? */
+			if (op->thrtimer.expires) {
+				/* send blocked msgs hereafter */
+				mod_timer(&op->thrtimer, jiffies + 2);
+			}
+
+			/*
+			 * if (op->j_ival2) is zero, no (new) throttling
+			 * will happen. For details see functions
+			 * bcm_rx_update_and_send() and bcm_rx_thr_handler()
+			 */
+		}
+
+		if ((op->flags & STARTTIMER) && op->j_ival1)
+			mod_timer(&op->timer, jiffies + op->j_ival1);
+	}
+
+	/* now we can register for can_ids, if we added a new bcm_op */
+	if (do_rx_register) {
+		if (ifindex) {
+			struct net_device *dev;
+
+			dev = dev_get_by_index(&init_net, ifindex);
+			if (dev) {
+				err = can_rx_register(dev, op->can_id,
+						      REGMASK(op->can_id),
+						      bcm_rx_handler, op,
+						      "bcm");
+
+				op->rx_reg_dev = dev;
+				dev_put(dev);
+			}
+
+		} else
+			err = can_rx_register(NULL, op->can_id,
+					      REGMASK(op->can_id),
+					      bcm_rx_handler, op, "bcm");
+		if (err) {
+			/* this bcm rx op is broken -> remove it */
+			list_del(&op->list);
+			bcm_remove_op(op);
+			return err;
+		}
+	}
+
+	return msg_head->nframes * CFSIZ + MHSIZ;
+}
+
+/*
+ * bcm_tx_send - send a single CAN frame to the CAN interface (for bcm_sendmsg)
+ */
+static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk)
+{
+	struct sk_buff *skb;
+	struct net_device *dev;
+	int err;
+
+	/* we need a real device to send frames */
+	if (!ifindex)
+		return -ENODEV;
+
+	skb = alloc_skb(CFSIZ, GFP_KERNEL);
+
+	if (!skb)
+		return -ENOMEM;
+
+	err = memcpy_fromiovec(skb_put(skb, CFSIZ), msg->msg_iov, CFSIZ);
+	if (err < 0) {
+		kfree_skb(skb);
+		return err;
+	}
+
+	dev = dev_get_by_index(&init_net, ifindex);
+	if (!dev) {
+		kfree_skb(skb);
+		return -ENODEV;
+	}
+
+	skb->dev = dev;
+	skb->sk  = sk;
+	can_send(skb, 1); /* send with loopback */
+	dev_put(dev);
+
+	return CFSIZ + MHSIZ;
+}
+
+/*
+ * bcm_sendmsg - process BCM commands (opcodes) from the userspace
+ */
+static int bcm_sendmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, size_t size)
+{
+	struct sock *sk = sock->sk;
+	struct bcm_sock *bo = bcm_sk(sk);
+	int ifindex = bo->ifindex; /* default ifindex for this bcm_op */
+	struct bcm_msg_head msg_head;
+	int ret; /* read bytes or error codes as return value */
+
+	if (!bo->bound)
+		return -ENOTCONN;
+
+	/* check for alternative ifindex for this bcm_op */
+
+	if (!ifindex && msg->msg_name) {
+		/* no bound device as default => check msg_name */
+		struct sockaddr_can *addr =
+			(struct sockaddr_can *)msg->msg_name;
+
+		if (addr->can_family != AF_CAN)
+			return -EINVAL;
+
+		/* ifindex from sendto() */
+		ifindex = addr->can_ifindex;
+
+		if (ifindex) {
+			struct net_device *dev;
+
+			dev = dev_get_by_index(&init_net, ifindex);
+			if (!dev)
+				return -ENODEV;
+
+			if (dev->type != ARPHRD_CAN) {
+				dev_put(dev);
+				return -ENODEV;
+			}
+
+			dev_put(dev);
+		}
+	}
+
+	/* read message head information */
+
+	ret = memcpy_fromiovec((u8 *)&msg_head, msg->msg_iov, MHSIZ);
+	if (ret < 0)
+		return ret;
+
+	lock_sock(sk);
+
+	switch (msg_head.opcode) {
+
+	case TX_SETUP:
+		ret = bcm_tx_setup(&msg_head, msg, ifindex, sk);
+		break;
+
+	case RX_SETUP:
+		ret = bcm_rx_setup(&msg_head, msg, ifindex, sk);
+		break;
+
+	case TX_DELETE:
+		if (bcm_delete_tx_op(&bo->tx_ops, msg_head.can_id, ifindex))
+			ret = MHSIZ;
+		else
+			ret = -EINVAL;
+		break;
+
+	case RX_DELETE:
+		if (bcm_delete_rx_op(&bo->rx_ops, msg_head.can_id, ifindex))
+			ret = MHSIZ;
+		else
+			ret = -EINVAL;
+		break;
+
+	case TX_READ:
+		/* reuse msg_head for the reply to TX_READ */
+		msg_head.opcode  = TX_STATUS;
+		ret = bcm_read_op(&bo->tx_ops, &msg_head, ifindex);
+		break;
+
+	case RX_READ:
+		/* reuse msg_head for the reply to RX_READ */
+		msg_head.opcode  = RX_STATUS;
+		ret = bcm_read_op(&bo->rx_ops, &msg_head, ifindex);
+		break;
+
+	case TX_SEND:
+		/* we need at least one can_frame */
+		if (msg_head.nframes < 1)
+			ret = -EINVAL;
+		else
+			ret = bcm_tx_send(msg, ifindex, sk);
+		break;
+
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	release_sock(sk);
+
+	return ret;
+}
+
+/*
+ * notification handler for netdevice status changes
+ */
+static int bcm_notifier(struct notifier_block *nb, unsigned long msg,
+			void *data)
+{
+	struct net_device *dev = (struct net_device *)data;
+	struct bcm_sock *bo = container_of(nb, struct bcm_sock, notifier);
+	struct sock *sk = &bo->sk;
+	struct bcm_op *op;
+	int notify_enodev = 0;
+
+	if (dev->nd_net != &init_net)
+		return NOTIFY_DONE;
+
+	if (dev->type != ARPHRD_CAN)
+		return NOTIFY_DONE;
+
+	switch (msg) {
+
+	case NETDEV_UNREGISTER:
+		lock_sock(sk);
+
+		/* remove device specific receive entries */
+		list_for_each_entry(op, &bo->rx_ops, list)
+			if (op->rx_reg_dev == dev)
+				bcm_rx_unreg(dev, op);
+
+		/* remove device reference, if this is our bound device */
+		if (bo->bound && bo->ifindex == dev->ifindex) {
+			bo->bound   = 0;
+			bo->ifindex = 0;
+			notify_enodev = 1;
+		}
+
+		release_sock(sk);
+
+		if (notify_enodev) {
+			sk->sk_err = ENODEV;
+			if (!sock_flag(sk, SOCK_DEAD))
+				sk->sk_error_report(sk);
+		}
+		break;
+
+	case NETDEV_DOWN:
+		if (bo->bound && bo->ifindex == dev->ifindex) {
+			sk->sk_err = ENETDOWN;
+			if (!sock_flag(sk, SOCK_DEAD))
+				sk->sk_error_report(sk);
+		}
+	}
+
+	return NOTIFY_DONE;
+}
+
+/*
+ * initial settings for all BCM sockets to be set at socket creation time
+ */
+static int bcm_init(struct sock *sk)
+{
+	struct bcm_sock *bo = bcm_sk(sk);
+
+	bo->bound            = 0;
+	bo->ifindex          = 0;
+	bo->dropped_usr_msgs = 0;
+	bo->bcm_proc_read    = NULL;
+
+	INIT_LIST_HEAD(&bo->tx_ops);
+	INIT_LIST_HEAD(&bo->rx_ops);
+
+	/* set notifier */
+	bo->notifier.notifier_call = bcm_notifier;
+
+	register_netdevice_notifier(&bo->notifier);
+
+	return 0;
+}
+
+/*
+ * standard socket functions
+ */
+static int bcm_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct bcm_sock *bo = bcm_sk(sk);
+	struct bcm_op *op, *next;
+
+	/* remove bcm_ops, timer, rx_unregister(), etc. */
+
+	unregister_netdevice_notifier(&bo->notifier);
+
+	lock_sock(sk);
+
+	list_for_each_entry_safe(op, next, &bo->tx_ops, list)
+		bcm_remove_op(op);
+
+	list_for_each_entry_safe(op, next, &bo->rx_ops, list) {
+		/*
+		 * Don't care if we're bound or not (due to netdev problems)
+		 * can_rx_unregister() is always a save thing to do here.
+		 */
+		if (op->ifindex) {
+			/*
+			 * Only remove subscriptions that had not
+			 * been removed due to NETDEV_UNREGISTER
+			 * in bcm_notifier()
+			 */
+			if (op->rx_reg_dev) {
+				struct net_device *dev;
+
+				dev = dev_get_by_index(&init_net, op->ifindex);
+				if (dev) {
+					bcm_rx_unreg(dev, op);
+					dev_put(dev);
+				}
+			}
+		} else
+			can_rx_unregister(NULL, op->can_id,
+					  REGMASK(op->can_id),
+					  bcm_rx_handler, op);
+
+		bcm_remove_op(op);
+	}
+
+	/* remove procfs entry */
+	if (proc_dir && bo->bcm_proc_read)
+		remove_proc_entry(bo->procname, proc_dir);
+
+	/* remove device reference */
+	if (bo->bound) {
+		bo->bound   = 0;
+		bo->ifindex = 0;
+	}
+
+	release_sock(sk);
+	sock_put(sk);
+
+	return 0;
+}
+
+static int bcm_connect(struct socket *sock, struct sockaddr *uaddr, int len,
+		       int flags)
+{
+	struct sockaddr_can *addr = (struct sockaddr_can *)uaddr;
+	struct sock *sk = sock->sk;
+	struct bcm_sock *bo = bcm_sk(sk);
+
+	if (bo->bound)
+		return -EISCONN;
+
+	/* bind a device to this socket */
+	if (addr->can_ifindex) {
+		struct net_device *dev;
+
+		dev = dev_get_by_index(&init_net, addr->can_ifindex);
+		if (!dev)
+			return -ENODEV;
+
+		if (dev->type != ARPHRD_CAN) {
+			dev_put(dev);
+			return -ENODEV;
+		}
+
+		bo->ifindex = dev->ifindex;
+		dev_put(dev);
+
+	} else {
+		/* no interface reference for ifindex = 0 ('any' CAN device) */
+		bo->ifindex = 0;
+	}
+
+	bo->bound = 1;
+
+	if (proc_dir) {
+		/* unique socket address as filename */
+		sprintf(bo->procname, "%p", sock);
+		bo->bcm_proc_read = create_proc_read_entry(bo->procname, 0644,
+							   proc_dir,
+							   bcm_read_proc, sk);
+	}
+
+	return 0;
+}
+
+static int bcm_recvmsg(struct kiocb *iocb, struct socket *sock,
+		       struct msghdr *msg, size_t size, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct sk_buff *skb;
+	int error = 0;
+	int noblock;
+	int err;
+
+	noblock =  flags & MSG_DONTWAIT;
+	flags   &= ~MSG_DONTWAIT;
+	skb = skb_recv_datagram(sk, flags, noblock, &error);
+	if (!skb)
+		return error;
+
+	if (skb->len < size)
+		size = skb->len;
+
+	err = memcpy_toiovec(msg->msg_iov, skb->data, size);
+	if (err < 0) {
+		skb_free_datagram(sk, skb);
+		return err;
+	}
+
+	sock_recv_timestamp(msg, sk, skb);
+
+	if (msg->msg_name) {
+		msg->msg_namelen = sizeof(struct sockaddr_can);
+		memcpy(msg->msg_name, skb->cb, msg->msg_namelen);
+	}
+
+	skb_free_datagram(sk, skb);
+
+	return size;
+}
+
+static struct proto_ops bcm_ops __read_mostly = {
+	.family        = PF_CAN,
+	.release       = bcm_release,
+	.bind          = sock_no_bind,
+	.connect       = bcm_connect,
+	.socketpair    = sock_no_socketpair,
+	.accept        = sock_no_accept,
+	.getname       = sock_no_getname,
+	.poll          = datagram_poll,
+	.ioctl         = NULL,		/* use can_ioctl() from af_can.c */
+	.listen        = sock_no_listen,
+	.shutdown      = sock_no_shutdown,
+	.setsockopt    = sock_no_setsockopt,
+	.getsockopt    = sock_no_getsockopt,
+	.sendmsg       = bcm_sendmsg,
+	.recvmsg       = bcm_recvmsg,
+	.mmap          = sock_no_mmap,
+	.sendpage      = sock_no_sendpage,
+};
+
+static struct proto bcm_proto __read_mostly = {
+	.name       = "CAN_BCM",
+	.owner      = THIS_MODULE,
+	.obj_size   = sizeof(struct bcm_sock),
+	.init       = bcm_init,
+};
+
+static struct can_proto bcm_can_proto __read_mostly = {
+	.type       = SOCK_DGRAM,
+	.protocol   = CAN_BCM,
+	.capability = -1,
+	.ops        = &bcm_ops,
+	.prot       = &bcm_proto,
+};
+
+static int __init bcm_module_init(void)
+{
+	int err;
+
+	printk(banner);
+
+	err = can_proto_register(&bcm_can_proto);
+	if (err < 0) {
+		printk(KERN_ERR "can: registration of bcm protocol failed\n");
+		return err;
+	}
+
+	/* create /proc/net/can-bcm directory */
+	proc_dir = proc_mkdir("can-bcm", init_net.proc_net);
+
+	if (proc_dir)
+		proc_dir->owner = THIS_MODULE;
+
+	return 0;
+}
+
+static void __exit bcm_module_exit(void)
+{
+	can_proto_unregister(&bcm_can_proto);
+
+	if (proc_dir)
+		proc_net_remove(&init_net, "can-bcm");
+}
+
+module_init(bcm_module_init);
+module_exit(bcm_module_exit);
-- 
cgit v1.2.3


From 4195e31780a20e09c6e793c2d96390e05309e226 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver@hartkopp.net>
Date: Thu, 27 Dec 2007 16:50:06 -0800
Subject: [CAN]: Fix plain integer definitions in userspace header.

This patch fixes the use of plain integers instead of __u32 in a struct
that is visible from kernel space and user space.

Thanks to Sam Ravnborg for pointing out the wrong plain int usage.

Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/can/bcm.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/can/bcm.h b/include/linux/can/bcm.h
index 7ade33a0ff03..7f293273c444 100644
--- a/include/linux/can/bcm.h
+++ b/include/linux/can/bcm.h
@@ -26,12 +26,12 @@
  * @frames:    array of CAN frames.
  */
 struct bcm_msg_head {
-	int opcode;
-	int flags;
-	int count;
+	__u32 opcode;
+	__u32 flags;
+	__u32 count;
 	struct timeval ival1, ival2;
 	canid_t can_id;
-	int nframes;
+	__u32 nframes;
 	struct can_frame frames[0];
 };
 
-- 
cgit v1.2.3


From 1f98eefae8ca451d317b1602f2cedf7515b032ff Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver@hartkopp.net>
Date: Thu, 27 Dec 2007 16:51:46 -0800
Subject: [CAN]: Add missing Kbuild entries

This patch adds the missing Kbuild entries and the missing Kbuild file
in include/linux/can for the CAN subsystem.

Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/Kbuild     | 2 ++
 include/linux/can/Kbuild | 3 +++
 2 files changed, 5 insertions(+)
 create mode 100644 include/linux/can/Kbuild

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index bd694f779346..a85e87fd60d7 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -1,4 +1,5 @@
 header-y += byteorder/
+header-y += can/
 header-y += dvb/
 header-y += hdlc/
 header-y += isdn/
@@ -41,6 +42,7 @@ header-y += baycom.h
 header-y += bfs_fs.h
 header-y += blkpg.h
 header-y += bpqether.h
+header-y += can.h
 header-y += cdk.h
 header-y += chio.h
 header-y += coda_psdev.h
diff --git a/include/linux/can/Kbuild b/include/linux/can/Kbuild
new file mode 100644
index 000000000000..eff898aac02b
--- /dev/null
+++ b/include/linux/can/Kbuild
@@ -0,0 +1,3 @@
+header-y += raw.h
+header-y += bcm.h
+header-y += error.h
-- 
cgit v1.2.3


From 0953864160bdd28dfe45fd46fa462b4d2d53cb96 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@linux-foundation.org>
Date: Mon, 19 Nov 2007 19:23:29 -0800
Subject: [NETPOLL]: no need to store local_mac

The local_mac is managed by the network device, no need to keep a
spare copy and all the management problems that could cause.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/netconsole.c | 10 ++++------
 include/linux/netpoll.h  |  2 +-
 net/core/netpoll.c       |  9 +++------
 3 files changed, 8 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 5ffbb8891647..31e047dd7bb3 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -306,9 +306,11 @@ static ssize_t show_remote_ip(struct netconsole_target *nt, char *buf)
 
 static ssize_t show_local_mac(struct netconsole_target *nt, char *buf)
 {
+	struct net_device *dev = nt->np.dev;
+
 	DECLARE_MAC_BUF(mac);
 	return snprintf(buf, PAGE_SIZE, "%s\n",
-			print_mac(mac, nt->np.local_mac));
+			print_mac(mac, dev->dev_addr));
 }
 
 static ssize_t show_remote_mac(struct netconsole_target *nt, char *buf)
@@ -667,7 +669,7 @@ static int netconsole_netdev_event(struct notifier_block *this,
 	struct netconsole_target *nt;
 	struct net_device *dev = ptr;
 
-	if (!(event == NETDEV_CHANGEADDR || event == NETDEV_CHANGENAME))
+	if (!(event == NETDEV_CHANGENAME))
 		goto done;
 
 	spin_lock_irqsave(&target_list_lock, flags);
@@ -675,10 +677,6 @@ static int netconsole_netdev_event(struct notifier_block *this,
 		netconsole_target_get(nt);
 		if (nt->np.dev == dev) {
 			switch (event) {
-			case NETDEV_CHANGEADDR:
-				memcpy(nt->np.local_mac, dev->dev_addr, ETH_ALEN);
-				break;
-
 			case NETDEV_CHANGENAME:
 				strlcpy(nt->np.dev_name, dev->name, IFNAMSIZ);
 				break;
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 20250d963d72..e3d79593fb3a 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -20,7 +20,7 @@ struct netpoll {
 
 	u32 local_ip, remote_ip;
 	u16 local_port, remote_port;
- 	u8 local_mac[ETH_ALEN], remote_mac[ETH_ALEN];
+	u8 remote_mac[ETH_ALEN];
 };
 
 struct netpoll_info {
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 250868f68762..cf6acd3084a4 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -360,8 +360,8 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
 	eth = (struct ethhdr *) skb_push(skb, ETH_HLEN);
 	skb_reset_mac_header(skb);
 	skb->protocol = eth->h_proto = htons(ETH_P_IP);
-	memcpy(eth->h_source, np->local_mac, 6);
-	memcpy(eth->h_dest, np->remote_mac, 6);
+	memcpy(eth->h_source, np->dev->dev_addr, ETH_ALEN);
+	memcpy(eth->h_dest, np->remote_mac, ETH_ALEN);
 
 	skb->dev = np->dev;
 
@@ -431,7 +431,7 @@ static void arp_reply(struct sk_buff *skb)
 
 	/* Fill the device header for the ARP frame */
 	if (dev_hard_header(send_skb, skb->dev, ptype,
-			    sha, np->local_mac,
+			    sha, np->dev->dev_addr,
 			    send_skb->len) < 0) {
 		kfree_skb(send_skb);
 		return;
@@ -737,9 +737,6 @@ int netpoll_setup(struct netpoll *np)
 		}
 	}
 
-	if (is_zero_ether_addr(np->local_mac) && ndev->dev_addr)
-		memcpy(np->local_mac, ndev->dev_addr, 6);
-
 	if (!np->local_ip) {
 		rcu_read_lock();
 		in_dev = __in_dev_get_rcu(ndev);
-- 
cgit v1.2.3


From c7b6ea24b43afb5749cb704e143df19d70e23dea Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@linux-foundation.org>
Date: Mon, 19 Nov 2007 19:37:09 -0800
Subject: [NETPOLL]: Don't need rx_flags.

The rx_flags variable is redundant. Turning rx on/off is done
via setting the rx_np pointer.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netpoll.h | 7 +++----
 net/core/netpoll.c      | 4 ----
 2 files changed, 3 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index e3d79593fb3a..a0525a1f4715 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -25,7 +25,6 @@ struct netpoll {
 
 struct netpoll_info {
 	atomic_t refcnt;
-	int rx_flags;
 	spinlock_t rx_lock;
 	struct netpoll *rx_np; /* netpoll that registered an rx_hook */
 	struct sk_buff_head arp_tx; /* list of arp requests to reply to */
@@ -51,12 +50,12 @@ static inline int netpoll_rx(struct sk_buff *skb)
 	unsigned long flags;
 	int ret = 0;
 
-	if (!npinfo || (!npinfo->rx_np && !npinfo->rx_flags))
+	if (!npinfo || !npinfo->rx_np)
 		return 0;
 
 	spin_lock_irqsave(&npinfo->rx_lock, flags);
-	/* check rx_flags again with the lock held */
-	if (npinfo->rx_flags && __netpoll_rx(skb))
+	/* check rx_np again with the lock held */
+	if (npinfo->rx_np && __netpoll_rx(skb))
 		ret = 1;
 	spin_unlock_irqrestore(&npinfo->rx_lock, flags);
 
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 9e3aea0bd369..b1d5acd2fc7a 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -39,7 +39,6 @@ static struct sk_buff_head skb_pool;
 static atomic_t trapped;
 
 #define USEC_PER_POLL	50
-#define NETPOLL_RX_ENABLED  1
 
 #define MAX_SKB_SIZE \
 		(MAX_UDP_CHUNK + sizeof(struct udphdr) + \
@@ -675,7 +674,6 @@ int netpoll_setup(struct netpoll *np)
 			goto release;
 		}
 
-		npinfo->rx_flags = 0;
 		npinfo->rx_np = NULL;
 
 		spin_lock_init(&npinfo->rx_lock);
@@ -757,7 +755,6 @@ int netpoll_setup(struct netpoll *np)
 
 	if (np->rx_hook) {
 		spin_lock_irqsave(&npinfo->rx_lock, flags);
-		npinfo->rx_flags |= NETPOLL_RX_ENABLED;
 		npinfo->rx_np = np;
 		spin_unlock_irqrestore(&npinfo->rx_lock, flags);
 	}
@@ -799,7 +796,6 @@ void netpoll_cleanup(struct netpoll *np)
 			if (npinfo->rx_np == np) {
 				spin_lock_irqsave(&npinfo->rx_lock, flags);
 				npinfo->rx_np = NULL;
-				npinfo->rx_flags &= ~NETPOLL_RX_ENABLED;
 				spin_unlock_irqrestore(&npinfo->rx_lock, flags);
 			}
 
-- 
cgit v1.2.3


From c237899d1f8c5bfcfc9d6204052e0e065827ff75 Mon Sep 17 00:00:00 2001
From: Michael Wu <flamingice@sourmilk.net>
Date: Tue, 30 Oct 2007 16:50:05 -0400
Subject: ieee80211: Add IEEE80211_MAX_FRAME_LEN to linux/ieee80211.h

This patch adds IEEE80211_MAX_FRAME_LEN which is useful for drivers trying
to determine how much to allocate for their RX buffers.

It also updates the comment on IEEE80211_MAX_DATA_LEN based on revisions
in 802.11e.

IEEE80211_MAX_FRAG_THRESHOLD and IEEE80211_MAX_RTS_THRESHOLD are also
revised due to the new maximum frame size.

Signed-off-by: Michael Wu <flamingice@sourmilk.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ieee80211.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 30621c27159f..214821af3d5b 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -81,18 +81,18 @@
 
 
 /* miscellaneous IEEE 802.11 constants */
-#define IEEE80211_MAX_FRAG_THRESHOLD	2346
-#define IEEE80211_MAX_RTS_THRESHOLD	2347
+#define IEEE80211_MAX_FRAG_THRESHOLD	2352
+#define IEEE80211_MAX_RTS_THRESHOLD	2353
 #define IEEE80211_MAX_AID		2007
 #define IEEE80211_MAX_TIM_LEN		251
-#define IEEE80211_MAX_DATA_LEN		2304
 /* Maximum size for the MA-UNITDATA primitive, 802.11 standard section
    6.2.1.1.2.
 
-   The figure in section 7.1.2 suggests a body size of up to 2312
-   bytes is allowed, which is a bit confusing, I suspect this
-   represents the 2304 bytes of real data, plus a possible 8 bytes of
-   WEP IV and ICV. (this interpretation suggested by Ramiro Barreiro) */
+   802.11e clarifies the figure in section 7.1.2. The frame body is
+   up to 2304 octets long (maximum MSDU size) plus any crypt overhead. */
+#define IEEE80211_MAX_DATA_LEN		2304
+/* 30 byte 4 addr hdr, 2 byte QoS, 2304 byte MSDU, 12 byte crypt, 4 byte FCS */
+#define IEEE80211_MAX_FRAME_LEN		2352
 
 #define IEEE80211_MAX_SSID_LEN		32
 
-- 
cgit v1.2.3


From 97c53cacf00d1f5aa04adabfebcc806ca8b22b10 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 19 Nov 2007 22:26:51 -0800
Subject: [NET]: Make rtnetlink infrastructure network namespace aware (v3)

After this patch none of the netlink callback support anything
except the initial network namespace but the rtnetlink infrastructure
now handles multiple network namespaces.

Changes from v2:
- IPv6 addrlabel processing

Changes from v1:
- no need for special rtnl_unlock handling
- fixed IPv6 ndisc

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h   |  8 +++---
 include/net/net_namespace.h |  3 +++
 net/bridge/br_netlink.c     |  4 +--
 net/core/fib_rules.c        |  4 +--
 net/core/neighbour.c        |  4 +--
 net/core/rtnetlink.c        | 63 +++++++++++++++++++++++++++++++++++++--------
 net/decnet/dn_dev.c         |  4 +--
 net/decnet/dn_route.c       |  2 +-
 net/decnet/dn_table.c       |  4 +--
 net/ipv4/devinet.c          |  4 +--
 net/ipv4/fib_semantics.c    |  4 +--
 net/ipv4/ipmr.c             |  4 +--
 net/ipv4/route.c            |  2 +-
 net/ipv6/addrconf.c         | 14 +++++-----
 net/ipv6/addrlabel.c        |  2 +-
 net/ipv6/ndisc.c            |  5 ++--
 net/ipv6/route.c            |  6 ++---
 net/sched/act_api.c         |  8 +++---
 net/sched/cls_api.c         |  2 +-
 net/sched/sch_api.c         |  4 +--
 net/wireless/wext.c         |  5 +++-
 21 files changed, 102 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index e20dcc89a836..b014f6b7fe29 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -620,11 +620,11 @@ extern int __rtattr_parse_nested_compat(struct rtattr *tb[], int maxattr,
 ({	data = RTA_PAYLOAD(rta) >= len ? RTA_DATA(rta) : NULL; \
 	__rtattr_parse_nested_compat(tb, max, rta, len); })
 
-extern int rtnetlink_send(struct sk_buff *skb, u32 pid, u32 group, int echo);
-extern int rtnl_unicast(struct sk_buff *skb, u32 pid);
-extern int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group,
+extern int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, u32 group, int echo);
+extern int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid);
+extern int rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
 		       struct nlmsghdr *nlh, gfp_t flags);
-extern void rtnl_set_sk_err(u32 group, int error);
+extern void rtnl_set_sk_err(struct net *net, u32 group, int error);
 extern int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics);
 extern int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst,
 			      u32 id, u32 ts, u32 tsage, long expires,
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 5dd6d90b37eb..90802a668c23 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -10,6 +10,7 @@
 
 struct proc_dir_entry;
 struct net_device;
+struct sock;
 struct net {
 	atomic_t		count;		/* To decided when the network
 						 *  namespace should be freed.
@@ -29,6 +30,8 @@ struct net {
 	struct list_head 	dev_base_head;
 	struct hlist_head 	*dev_name_head;
 	struct hlist_head	*dev_index_head;
+
+	struct sock 		*rtnl;			/* rtnetlink socket */
 };
 
 #ifdef CONFIG_NET
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index a4ffa2b63cd5..f5d69336d97b 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -97,10 +97,10 @@ void br_ifinfo_notify(int event, struct net_bridge_port *port)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net,0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_LINK, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_LINK, err);
 }
 
 /*
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 3b20b6f0982e..0af0538343da 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -599,10 +599,10 @@ static void notify_rule_change(int event, struct fib_rule *rule,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, pid, ops->nlgroup, nlh, GFP_KERNEL);
+	err = rtnl_notify(skb, &init_net, pid, ops->nlgroup, nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(ops->nlgroup, err);
+		rtnl_set_sk_err(&init_net, ops->nlgroup, err);
 }
 
 static void attach_rules(struct list_head *rules, struct net_device *dev)
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 29f0a4d2008f..a8b72c1c7c8a 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -2467,10 +2467,10 @@ static void __neigh_notify(struct neighbour *n, int type, int flags)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_NEIGH, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_NEIGH, err);
 }
 
 #ifdef CONFIG_ARPD
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 4edc3dac4ccd..9efaf35934f4 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -60,7 +60,6 @@ struct rtnl_link
 };
 
 static DEFINE_MUTEX(rtnl_mutex);
-static struct sock *rtnl;
 
 void rtnl_lock(void)
 {
@@ -458,8 +457,9 @@ size_t rtattr_strlcpy(char *dest, const struct rtattr *rta, size_t size)
 	return ret;
 }
 
-int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
+int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned group, int echo)
 {
+	struct sock *rtnl = net->rtnl;
 	int err = 0;
 
 	NETLINK_CB(skb).dst_group = group;
@@ -471,14 +471,17 @@ int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
 	return err;
 }
 
-int rtnl_unicast(struct sk_buff *skb, u32 pid)
+int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid)
 {
+	struct sock *rtnl = net->rtnl;
+
 	return nlmsg_unicast(rtnl, skb, pid);
 }
 
-int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group,
+int rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid, u32 group,
 		struct nlmsghdr *nlh, gfp_t flags)
 {
+	struct sock *rtnl = net->rtnl;
 	int report = 0;
 
 	if (nlh)
@@ -487,8 +490,10 @@ int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group,
 	return nlmsg_notify(rtnl, skb, pid, group, report, flags);
 }
 
-void rtnl_set_sk_err(u32 group, int error)
+void rtnl_set_sk_err(struct net *net, u32 group, int error)
 {
+	struct sock *rtnl = net->rtnl;
+
 	netlink_set_err(rtnl, 0, group, error);
 }
 
@@ -1201,7 +1206,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		kfree_skb(nskb);
 		goto errout;
 	}
-	err = rtnl_unicast(nskb, NETLINK_CB(skb).pid);
+	err = rtnl_unicast(nskb, net, NETLINK_CB(skb).pid);
 errout:
 	dev_put(dev);
 
@@ -1252,10 +1257,10 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_KERNEL);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_LINK, NULL, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_LINK, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_LINK, err);
 }
 
 /* Protected by RTNL sempahore.  */
@@ -1266,6 +1271,7 @@ static int rtattr_max;
 
 static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
+	struct net *net = skb->sk->sk_net;
 	rtnl_doit_func doit;
 	int sz_idx, kind;
 	int min_len;
@@ -1294,6 +1300,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		return -EPERM;
 
 	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
+		struct sock *rtnl;
 		rtnl_dumpit_func dumpit;
 
 		dumpit = rtnl_get_dumpit(family, type);
@@ -1301,6 +1308,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 			return -EOPNOTSUPP;
 
 		__rtnl_unlock();
+		rtnl = net->rtnl;
 		err = netlink_dump_start(rtnl, skb, nlh, dumpit, NULL);
 		rtnl_lock();
 		return err;
@@ -1373,6 +1381,40 @@ static struct notifier_block rtnetlink_dev_notifier = {
 	.notifier_call	= rtnetlink_event,
 };
 
+
+static int rtnetlink_net_init(struct net *net)
+{
+	struct sock *sk;
+	sk = netlink_kernel_create(net, NETLINK_ROUTE, RTNLGRP_MAX,
+				   rtnetlink_rcv, &rtnl_mutex, THIS_MODULE);
+	if (!sk)
+		return -ENOMEM;
+
+	/* Don't hold an extra reference on the namespace */
+	put_net(sk->sk_net);
+	net->rtnl = sk;
+	return 0;
+}
+
+static void rtnetlink_net_exit(struct net *net)
+{
+	struct sock *sk = net->rtnl;
+	if (sk) {
+		/* At the last minute lie and say this is a socket for the
+		 * initial network namespace.  So the socket will be safe to
+		 * free.
+		 */
+		sk->sk_net = get_net(&init_net);
+		sock_put(sk);
+		net->rtnl = NULL;
+	}
+}
+
+static struct pernet_operations rtnetlink_net_ops = {
+	.init = rtnetlink_net_init,
+	.exit = rtnetlink_net_exit,
+};
+
 void __init rtnetlink_init(void)
 {
 	int i;
@@ -1385,10 +1427,9 @@ void __init rtnetlink_init(void)
 	if (!rta_buf)
 		panic("rtnetlink_init: cannot allocate rta_buf\n");
 
-	rtnl = netlink_kernel_create(&init_net, NETLINK_ROUTE, RTNLGRP_MAX,
-				     rtnetlink_rcv, &rtnl_mutex, THIS_MODULE);
-	if (rtnl == NULL)
+	if (register_pernet_subsys(&rtnetlink_net_ops))
 		panic("rtnetlink_init: cannot initialize rtnetlink\n");
+
 	netlink_set_nonroot(NETLINK_ROUTE, NL_NONROOT_RECV);
 	register_netdevice_notifier(&rtnetlink_dev_notifier);
 
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index 94256845a057..39c89c682047 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -793,10 +793,10 @@ static void dn_ifaddr_notify(int event, struct dn_ifaddr *ifa)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_DECnet_IFADDR, NULL, GFP_KERNEL);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_DECnet_IFADDR, NULL, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_DECnet_IFADDR, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_DECnet_IFADDR, err);
 }
 
 static int dn_nl_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 28aeba15cf12..5d742f1420d9 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -1587,7 +1587,7 @@ static int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void
 		goto out_free;
 	}
 
-	return rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+	return rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
 
 out_free:
 	kfree_skb(skb);
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index a3bdb8dd1fb2..e09d915dbd77 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -375,10 +375,10 @@ static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, u32 tb_id,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, pid, RTNLGRP_DECnet_ROUTE, nlh, GFP_KERNEL);
+	err = rtnl_notify(skb, &init_net, pid, RTNLGRP_DECnet_ROUTE, nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_DECnet_ROUTE, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_DECnet_ROUTE, err);
 }
 
 static __inline__ int dn_hash_dump_bucket(struct sk_buff *skb,
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index c0eb26a0d0bc..6e75c884e1af 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1240,10 +1240,10 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa, struct nlmsghdr *nlh,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, pid, RTNLGRP_IPV4_IFADDR, nlh, GFP_KERNEL);
+	err = rtnl_notify(skb, &init_net, pid, RTNLGRP_IPV4_IFADDR, nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV4_IFADDR, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV4_IFADDR, err);
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 1351a2617dce..33ec96001d90 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -320,11 +320,11 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, info->pid, RTNLGRP_IPV4_ROUTE,
+	err = rtnl_notify(skb, &init_net, info->pid, RTNLGRP_IPV4_ROUTE,
 			  info->nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV4_ROUTE, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV4_ROUTE, err);
 }
 
 /* Return the first fib alias matching TOS with
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 8e5d47a60602..11879283ad52 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -321,7 +321,7 @@ static void ipmr_destroy_unres(struct mfc_cache *c)
 			e->error = -ETIMEDOUT;
 			memset(&e->msg, 0, sizeof(e->msg));
 
-			rtnl_unicast(skb, NETLINK_CB(skb).pid);
+			rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid);
 		} else
 			kfree_skb(skb);
 	}
@@ -533,7 +533,7 @@ static void ipmr_cache_resolve(struct mfc_cache *uc, struct mfc_cache *c)
 				memset(&e->msg, 0, sizeof(e->msg));
 			}
 
-			rtnl_unicast(skb, NETLINK_CB(skb).pid);
+			rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid);
 		} else
 			ip_mr_forward(skb, c, 0);
 	}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 1d2839571d2f..e4b6fb4b1f4f 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2610,7 +2610,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 	if (err <= 0)
 		goto errout_free;
 
-	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
 errout:
 	return err;
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 26de8ee5095b..6c8b193474ba 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3397,7 +3397,7 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 		kfree_skb(skb);
 		goto errout_ifa;
 	}
-	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
 errout_ifa:
 	in6_ifa_put(ifa);
 errout:
@@ -3420,10 +3420,10 @@ static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV6_IFADDR, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_IFADDR, err);
 }
 
 static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
@@ -3628,10 +3628,10 @@ void inet6_ifinfo_notify(int event, struct inet6_dev *idev)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV6_IFADDR, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_IFADDR, err);
 }
 
 static inline size_t inet6_prefix_nlmsg_size(void)
@@ -3697,10 +3697,10 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, 0, RTNLGRP_IPV6_PREFIX, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_IPV6_PREFIX, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV6_PREFIX, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_PREFIX, err);
 }
 
 static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index b9b5d5707142..6f1ca607edda 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -549,7 +549,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 		goto out;
 	}
 
-	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
 out:
 	return err;
 }
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index b2531f80317e..b87f9d245e2e 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1049,7 +1049,8 @@ static void ndisc_ra_useropt(struct sk_buff *ra, struct nd_opt_hdr *opt)
 		&ipv6_hdr(ra)->saddr);
 	nlmsg_end(skb, nlh);
 
-	err = rtnl_notify(skb, 0, RTNLGRP_ND_USEROPT, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_ND_USEROPT, NULL,
+			  GFP_ATOMIC);
 	if (err < 0)
 		goto errout;
 
@@ -1059,7 +1060,7 @@ nla_put_failure:
 	nlmsg_free(skb);
 	err = -EMSGSIZE;
 errout:
-	rtnl_set_sk_err(RTNLGRP_ND_USEROPT, err);
+	rtnl_set_sk_err(&init_net, RTNLGRP_ND_USEROPT, err);
 }
 
 static void ndisc_router_discovery(struct sk_buff *skb)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 5e1c5796761b..d7ec4c9ffc4b 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2230,7 +2230,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 		goto errout;
 	}
 
-	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
 errout:
 	return err;
 }
@@ -2260,10 +2260,10 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, pid, RTNLGRP_IPV6_ROUTE, nlh, gfp_any());
+	err = rtnl_notify(skb, &init_net, pid, RTNLGRP_IPV6_ROUTE, nlh, gfp_any());
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(RTNLGRP_IPV6_ROUTE, err);
+		rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_ROUTE, err);
 }
 
 /*
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 852829139c67..81506474a4f7 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -660,7 +660,7 @@ act_get_notify(u32 pid, struct nlmsghdr *n, struct tc_action *a, int event)
 		return -EINVAL;
 	}
 
-	return rtnl_unicast(skb, pid);
+	return rtnl_unicast(skb, &init_net, pid);
 }
 
 static struct tc_action *
@@ -781,7 +781,7 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid)
 	nlh->nlmsg_flags |= NLM_F_ROOT;
 	module_put(a->ops->owner);
 	kfree(a);
-	err = rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+	err = rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
 	if (err > 0)
 		return 0;
 
@@ -844,7 +844,7 @@ tca_action_gd(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int event)
 
 		/* now do the delete */
 		tcf_action_destroy(head, 0);
-		ret = rtnetlink_send(skb, pid, RTNLGRP_TC,
+		ret = rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC,
 				     n->nlmsg_flags&NLM_F_ECHO);
 		if (ret > 0)
 			return 0;
@@ -888,7 +888,7 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event,
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	NETLINK_CB(skb).dst_group = RTNLGRP_TC;
 
-	err = rtnetlink_send(skb, pid, RTNLGRP_TC, flags&NLM_F_ECHO);
+	err = rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, flags&NLM_F_ECHO);
 	if (err > 0)
 		err = 0;
 	return err;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index fdab6a530bba..80dccac769d0 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -361,7 +361,7 @@ static int tfilter_notify(struct sk_buff *oskb, struct nlmsghdr *n,
 		return -EINVAL;
 	}
 
-	return rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+	return rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
 }
 
 struct tcf_dump_args
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index f30e3f7ad885..273c628be054 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -872,7 +872,7 @@ static int qdisc_notify(struct sk_buff *oskb, struct nlmsghdr *n,
 	}
 
 	if (skb->len)
-		return rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+		return rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
 
 err_out:
 	kfree_skb(skb);
@@ -1103,7 +1103,7 @@ static int tclass_notify(struct sk_buff *oskb, struct nlmsghdr *n,
 		return -EINVAL;
 	}
 
-	return rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
+	return rtnetlink_send(skb, &init_net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
 }
 
 struct qdisc_dump_args
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index 47e80cc2077c..db03ed5ce055 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -1137,7 +1137,7 @@ static void wireless_nlevent_process(unsigned long data)
 	struct sk_buff *skb;
 
 	while ((skb = skb_dequeue(&wireless_nlevent_queue)))
-		rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
+		rtnl_notify(skb, &init_net, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC);
 }
 
 static DECLARE_TASKLET(wireless_nlevent_tasklet, wireless_nlevent_process, 0);
@@ -1189,6 +1189,9 @@ static void rtmsg_iwinfo(struct net_device *dev, char *event, int event_len)
 	struct sk_buff *skb;
 	int err;
 
+	if (dev->nd_net != &init_net)
+		return;
+
 	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
 	if (!skb)
 		return;
-- 
cgit v1.2.3


From d12d01d6b4d197115c4d7736172b5b1ec8b1eb71 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 19 Nov 2007 22:28:35 -0800
Subject: [NET]: Make AF_PACKET handle multiple network namespaces

This is done by making packet_sklist_lock and packet_sklist per
network namespace and adding an additional filter condition on
received packets to ensure they came from the proper network
namespace.

Changes from v1:
- prohibit to call inet_dgram_ops.ioctl in other than init_net

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |   4 ++
 net/packet/af_packet.c      | 131 ++++++++++++++++++++++++++++----------------
 2 files changed, 89 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 90802a668c23..4d0d6349aa7e 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -32,6 +32,10 @@ struct net {
 	struct hlist_head	*dev_index_head;
 
 	struct sock 		*rtnl;			/* rtnetlink socket */
+
+	/* List of all packet sockets. */
+	rwlock_t		packet_sklist_lock;
+	struct hlist_head	packet_sklist;
 };
 
 #ifdef CONFIG_NET
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 8a7807dbba01..45e3cbcb2763 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -135,10 +135,6 @@ dev->hard_header == NULL (ll header is added by device, we cannot control it)
    packet classifier depends on it.
  */
 
-/* List of all packet sockets. */
-static HLIST_HEAD(packet_sklist);
-static DEFINE_RWLOCK(packet_sklist_lock);
-
 /* Private packet socket structures. */
 
 struct packet_mclist
@@ -246,9 +242,6 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev,  struct
 	struct sock *sk;
 	struct sockaddr_pkt *spkt;
 
-	if (dev->nd_net != &init_net)
-		goto out;
-
 	/*
 	 *	When we registered the protocol we saved the socket in the data
 	 *	field for just this event.
@@ -270,6 +263,9 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev,  struct
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto out;
 
+	if (dev->nd_net != sk->sk_net)
+		goto out;
+
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
 		goto oom;
 
@@ -341,7 +337,7 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock,
 	 */
 
 	saddr->spkt_device[13] = 0;
-	dev = dev_get_by_name(&init_net, saddr->spkt_device);
+	dev = dev_get_by_name(sk->sk_net, saddr->spkt_device);
 	err = -ENODEV;
 	if (dev == NULL)
 		goto out_unlock;
@@ -449,15 +445,15 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet
 	int skb_len = skb->len;
 	unsigned int snaplen, res;
 
-	if (dev->nd_net != &init_net)
-		goto drop;
-
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto drop;
 
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
+	if (dev->nd_net != sk->sk_net)
+		goto drop;
+
 	skb->dev = dev;
 
 	if (dev->header_ops) {
@@ -566,15 +562,15 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	struct sk_buff *copy_skb = NULL;
 	struct timeval tv;
 
-	if (dev->nd_net != &init_net)
-		goto drop;
-
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto drop;
 
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
+	if (dev->nd_net != sk->sk_net)
+		goto drop;
+
 	if (dev->header_ops) {
 		if (sk->sk_type != SOCK_DGRAM)
 			skb_push(skb, skb->data - skb_mac_header(skb));
@@ -732,7 +728,7 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock,
 	}
 
 
-	dev = dev_get_by_index(&init_net, ifindex);
+	dev = dev_get_by_index(sk->sk_net, ifindex);
 	err = -ENXIO;
 	if (dev == NULL)
 		goto out_unlock;
@@ -799,15 +795,17 @@ static int packet_release(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
 	struct packet_sock *po;
+	struct net *net;
 
 	if (!sk)
 		return 0;
 
+	net = sk->sk_net;
 	po = pkt_sk(sk);
 
-	write_lock_bh(&packet_sklist_lock);
+	write_lock_bh(&net->packet_sklist_lock);
 	sk_del_node_init(sk);
-	write_unlock_bh(&packet_sklist_lock);
+	write_unlock_bh(&net->packet_sklist_lock);
 
 	/*
 	 *	Unhook packet receive handler.
@@ -916,7 +914,7 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, int add
 		return -EINVAL;
 	strlcpy(name,uaddr->sa_data,sizeof(name));
 
-	dev = dev_get_by_name(&init_net, name);
+	dev = dev_get_by_name(sk->sk_net, name);
 	if (dev) {
 		err = packet_do_bind(sk, dev, pkt_sk(sk)->num);
 		dev_put(dev);
@@ -943,7 +941,7 @@ static int packet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len
 
 	if (sll->sll_ifindex) {
 		err = -ENODEV;
-		dev = dev_get_by_index(&init_net, sll->sll_ifindex);
+		dev = dev_get_by_index(sk->sk_net, sll->sll_ifindex);
 		if (dev == NULL)
 			goto out;
 	}
@@ -972,9 +970,6 @@ static int packet_create(struct net *net, struct socket *sock, int protocol)
 	__be16 proto = (__force __be16)protocol; /* weird, but documented */
 	int err;
 
-	if (net != &init_net)
-		return -EAFNOSUPPORT;
-
 	if (!capable(CAP_NET_RAW))
 		return -EPERM;
 	if (sock->type != SOCK_DGRAM && sock->type != SOCK_RAW &&
@@ -1020,9 +1015,9 @@ static int packet_create(struct net *net, struct socket *sock, int protocol)
 		po->running = 1;
 	}
 
-	write_lock_bh(&packet_sklist_lock);
-	sk_add_node(sk, &packet_sklist);
-	write_unlock_bh(&packet_sklist_lock);
+	write_lock_bh(&net->packet_sklist_lock);
+	sk_add_node(sk, &net->packet_sklist);
+	write_unlock_bh(&net->packet_sklist_lock);
 	return(0);
 out:
 	return err;
@@ -1140,7 +1135,7 @@ static int packet_getname_spkt(struct socket *sock, struct sockaddr *uaddr,
 		return -EOPNOTSUPP;
 
 	uaddr->sa_family = AF_PACKET;
-	dev = dev_get_by_index(&init_net, pkt_sk(sk)->ifindex);
+	dev = dev_get_by_index(sk->sk_net, pkt_sk(sk)->ifindex);
 	if (dev) {
 		strlcpy(uaddr->sa_data, dev->name, 15);
 		dev_put(dev);
@@ -1165,7 +1160,7 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr,
 	sll->sll_family = AF_PACKET;
 	sll->sll_ifindex = po->ifindex;
 	sll->sll_protocol = po->num;
-	dev = dev_get_by_index(&init_net, po->ifindex);
+	dev = dev_get_by_index(sk->sk_net, po->ifindex);
 	if (dev) {
 		sll->sll_hatype = dev->type;
 		sll->sll_halen = dev->addr_len;
@@ -1217,7 +1212,7 @@ static int packet_mc_add(struct sock *sk, struct packet_mreq_max *mreq)
 	rtnl_lock();
 
 	err = -ENODEV;
-	dev = __dev_get_by_index(&init_net, mreq->mr_ifindex);
+	dev = __dev_get_by_index(sk->sk_net, mreq->mr_ifindex);
 	if (!dev)
 		goto done;
 
@@ -1271,7 +1266,7 @@ static int packet_mc_drop(struct sock *sk, struct packet_mreq_max *mreq)
 			if (--ml->count == 0) {
 				struct net_device *dev;
 				*mlp = ml->next;
-				dev = dev_get_by_index(&init_net, ml->ifindex);
+				dev = dev_get_by_index(sk->sk_net, ml->ifindex);
 				if (dev) {
 					packet_dev_mc(dev, ml, -1);
 					dev_put(dev);
@@ -1299,7 +1294,7 @@ static void packet_flush_mclist(struct sock *sk)
 		struct net_device *dev;
 
 		po->mclist = ml->next;
-		if ((dev = dev_get_by_index(&init_net, ml->ifindex)) != NULL) {
+		if ((dev = dev_get_by_index(sk->sk_net, ml->ifindex)) != NULL) {
 			packet_dev_mc(dev, ml, -1);
 			dev_put(dev);
 		}
@@ -1455,12 +1450,10 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void
 	struct sock *sk;
 	struct hlist_node *node;
 	struct net_device *dev = data;
+	struct net *net = dev->nd_net;
 
-	if (dev->nd_net != &init_net)
-		return NOTIFY_DONE;
-
-	read_lock(&packet_sklist_lock);
-	sk_for_each(sk, node, &packet_sklist) {
+	read_lock(&net->packet_sklist_lock);
+	sk_for_each(sk, node, &net->packet_sklist) {
 		struct packet_sock *po = pkt_sk(sk);
 
 		switch (msg) {
@@ -1499,7 +1492,7 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void
 			break;
 		}
 	}
-	read_unlock(&packet_sklist_lock);
+	read_unlock(&net->packet_sklist_lock);
 	return NOTIFY_DONE;
 }
 
@@ -1547,6 +1540,8 @@ static int packet_ioctl(struct socket *sock, unsigned int cmd,
 		case SIOCGIFDSTADDR:
 		case SIOCSIFDSTADDR:
 		case SIOCSIFFLAGS:
+			if (sk->sk_net != &init_net)
+				return -ENOIOCTLCMD;
 			return inet_dgram_ops.ioctl(sock, cmd, arg);
 #endif
 
@@ -1862,12 +1857,12 @@ static struct notifier_block packet_netdev_notifier = {
 };
 
 #ifdef CONFIG_PROC_FS
-static inline struct sock *packet_seq_idx(loff_t off)
+static inline struct sock *packet_seq_idx(struct net *net, loff_t off)
 {
 	struct sock *s;
 	struct hlist_node *node;
 
-	sk_for_each(s, node, &packet_sklist) {
+	sk_for_each(s, node, &net->packet_sklist) {
 		if (!off--)
 			return s;
 	}
@@ -1876,21 +1871,24 @@ static inline struct sock *packet_seq_idx(loff_t off)
 
 static void *packet_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	read_lock(&packet_sklist_lock);
-	return *pos ? packet_seq_idx(*pos - 1) : SEQ_START_TOKEN;
+	struct net *net = seq->private;
+	read_lock(&net->packet_sklist_lock);
+	return *pos ? packet_seq_idx(net, *pos - 1) : SEQ_START_TOKEN;
 }
 
 static void *packet_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
+	struct net *net = seq->private;
 	++*pos;
 	return  (v == SEQ_START_TOKEN)
-		? sk_head(&packet_sklist)
+		? sk_head(&net->packet_sklist)
 		: sk_next((struct sock*)v) ;
 }
 
 static void packet_seq_stop(struct seq_file *seq, void *v)
 {
-	read_unlock(&packet_sklist_lock);
+	struct net *net = seq->private;
+	read_unlock(&net->packet_sklist_lock);
 }
 
 static int packet_seq_show(struct seq_file *seq, void *v)
@@ -1926,7 +1924,26 @@ static const struct seq_operations packet_seq_ops = {
 
 static int packet_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open(file, &packet_seq_ops);
+	struct seq_file *seq;
+	int res;
+	res = seq_open(file, &packet_seq_ops);
+	if (!res) {
+		seq = file->private_data;
+		seq->private = get_proc_net(inode);
+		if (!seq->private) {
+			seq_release(inode, file);
+			res = -ENXIO;
+		}
+	}
+	return res;
+}
+
+static int packet_seq_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq=  file->private_data;
+	struct net *net = seq->private;
+	put_net(net);
+	return seq_release(inode, file);
 }
 
 static const struct file_operations packet_seq_fops = {
@@ -1934,15 +1951,37 @@ static const struct file_operations packet_seq_fops = {
 	.open		= packet_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release,
+	.release	= packet_seq_release,
 };
 
 #endif
 
+static int packet_net_init(struct net *net)
+{
+	rwlock_init(&net->packet_sklist_lock);
+	INIT_HLIST_HEAD(&net->packet_sklist);
+
+	if (!proc_net_fops_create(net, "packet", 0, &packet_seq_fops))
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void packet_net_exit(struct net *net)
+{
+	proc_net_remove(net, "packet");
+}
+
+static struct pernet_operations packet_net_ops = {
+	.init = packet_net_init,
+	.exit = packet_net_exit,
+};
+
+
 static void __exit packet_exit(void)
 {
-	proc_net_remove(&init_net, "packet");
 	unregister_netdevice_notifier(&packet_netdev_notifier);
+	unregister_pernet_subsys(&packet_net_ops);
 	sock_unregister(PF_PACKET);
 	proto_unregister(&packet_proto);
 }
@@ -1955,8 +1994,8 @@ static int __init packet_init(void)
 		goto out;
 
 	sock_register(&packet_family_ops);
+	register_pernet_subsys(&packet_net_ops);
 	register_netdevice_notifier(&packet_netdev_notifier);
-	proc_net_fops_create(&init_net, "packet", 0, &packet_seq_fops);
 out:
 	return rc;
 }
-- 
cgit v1.2.3


From e372c41401993b45c721c4d92730e7e0a79f7c1b Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 19 Nov 2007 22:31:54 -0800
Subject: [NET]: Consolidate net namespace related proc files creation.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c       | 38 ++++++++++++++++++++++++++++++++++++++
 include/linux/seq_file.h | 13 +++++++++++++
 net/core/dev.c           | 28 +++++-----------------------
 net/core/dev_mcast.c     | 26 ++++----------------------
 net/netlink/af_netlink.c | 33 +++++++--------------------------
 net/packet/af_packet.c   | 26 ++++----------------------
 net/unix/af_unix.c       | 31 ++++++-------------------------
 net/wireless/wext.c      | 24 +++---------------------
 8 files changed, 80 insertions(+), 139 deletions(-)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 0afe21ee0607..cfc4f6c072f1 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -22,10 +22,48 @@
 #include <linux/mount.h>
 #include <linux/nsproxy.h>
 #include <net/net_namespace.h>
+#include <linux/seq_file.h>
 
 #include "internal.h"
 
 
+int seq_open_net(struct inode *ino, struct file *f,
+		 const struct seq_operations *ops, int size)
+{
+	struct net *net;
+	struct seq_net_private *p;
+
+	BUG_ON(size < sizeof(*p));
+
+	net = get_proc_net(ino);
+	if (net == NULL)
+		return -ENXIO;
+
+	p = __seq_open_private(f, ops, size);
+	if (p == NULL) {
+		put_net(net);
+		return -ENOMEM;
+	}
+	p->net = net;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(seq_open_net);
+
+int seq_release_net(struct inode *ino, struct file *f)
+{
+	struct seq_file *seq;
+	struct seq_net_private *p;
+
+	seq = f->private_data;
+	p = seq->private;
+
+	put_net(p->net);
+	seq_release_private(ino, f);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(seq_release_net);
+
+
 struct proc_dir_entry *proc_net_fops_create(struct net *net,
 	const char *name, mode_t mode, const struct file_operations *fops)
 {
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index ebbc02b325fc..648dfeb444db 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -63,5 +63,18 @@ extern struct list_head *seq_list_start_head(struct list_head *head,
 extern struct list_head *seq_list_next(void *v, struct list_head *head,
 		loff_t *ppos);
 
+struct net;
+struct seq_net_private {
+	struct net *net;
+};
+
+int seq_open_net(struct inode *, struct file *,
+		 const struct seq_operations *, int);
+int seq_release_net(struct inode *, struct file *);
+static inline struct net *seq_file_net(struct seq_file *seq)
+{
+	return ((struct seq_net_private *)seq->private)->net;
+}
+
 #endif
 #endif
diff --git a/net/core/dev.c b/net/core/dev.c
index 0879f52115eb..d0e23d8310ff 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2364,7 +2364,7 @@ static int dev_ifconf(struct net *net, char __user *arg)
  */
 void *dev_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	struct net *net = seq->private;
+	struct net *net = seq_file_net(seq);
 	loff_t off;
 	struct net_device *dev;
 
@@ -2382,7 +2382,7 @@ void *dev_seq_start(struct seq_file *seq, loff_t *pos)
 
 void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
-	struct net *net = seq->private;
+	struct net *net = seq_file_net(seq);
 	++*pos;
 	return v == SEQ_START_TOKEN ?
 		first_net_device(net) : next_net_device((struct net_device *)v);
@@ -2481,26 +2481,8 @@ static const struct seq_operations dev_seq_ops = {
 
 static int dev_seq_open(struct inode *inode, struct file *file)
 {
-	struct seq_file *seq;
-	int res;
-	res =  seq_open(file, &dev_seq_ops);
-	if (!res) {
-		seq = file->private_data;
-		seq->private = get_proc_net(inode);
-		if (!seq->private) {
-			seq_release(inode, file);
-			res = -ENXIO;
-		}
-	}
-	return res;
-}
-
-static int dev_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return seq_release(inode, file);
+	return seq_open_net(inode, file, &dev_seq_ops,
+			    sizeof(struct seq_net_private));
 }
 
 static const struct file_operations dev_seq_fops = {
@@ -2508,7 +2490,7 @@ static const struct file_operations dev_seq_fops = {
 	.open    = dev_seq_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
-	.release = dev_seq_release,
+	.release = seq_release_net,
 };
 
 static const struct seq_operations softnet_seq_ops = {
diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c
index 69fff16ece10..63f0b33d7ced 100644
--- a/net/core/dev_mcast.c
+++ b/net/core/dev_mcast.c
@@ -187,7 +187,7 @@ EXPORT_SYMBOL(dev_mc_unsync);
 #ifdef CONFIG_PROC_FS
 static void *dev_mc_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	struct net *net = seq->private;
+	struct net *net = seq_file_net(seq);
 	struct net_device *dev;
 	loff_t off = 0;
 
@@ -241,26 +241,8 @@ static const struct seq_operations dev_mc_seq_ops = {
 
 static int dev_mc_seq_open(struct inode *inode, struct file *file)
 {
-	struct seq_file *seq;
-	int res;
-	res = seq_open(file, &dev_mc_seq_ops);
-	if (!res) {
-		seq = file->private_data;
-		seq->private = get_proc_net(inode);
-		if (!seq->private) {
-			seq_release(inode, file);
-			res = -ENXIO;
-		}
-	}
-	return res;
-}
-
-static int dev_mc_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return seq_release(inode, file);
+	return seq_open_net(inode, file, &dev_mc_seq_ops,
+			    sizeof(struct seq_net_private));
 }
 
 static const struct file_operations dev_mc_seq_fops = {
@@ -268,7 +250,7 @@ static const struct file_operations dev_mc_seq_fops = {
 	.open    = dev_mc_seq_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
-	.release = dev_mc_seq_release,
+	.release = seq_release_net,
 };
 
 #endif
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index de3988ba1f46..1518244ffad9 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1681,7 +1681,7 @@ int nlmsg_notify(struct sock *sk, struct sk_buff *skb, u32 pid,
 
 #ifdef CONFIG_PROC_FS
 struct nl_seq_iter {
-	struct net *net;
+	struct seq_net_private p;
 	int link;
 	int hash_idx;
 };
@@ -1699,7 +1699,7 @@ static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos)
 
 		for (j = 0; j <= hash->mask; j++) {
 			sk_for_each(s, node, &hash->table[j]) {
-				if (iter->net != s->sk_net)
+				if (iter->p.net != s->sk_net)
 					continue;
 				if (off == pos) {
 					iter->link = i;
@@ -1734,7 +1734,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	s = v;
 	do {
 		s = sk_next(s);
-	} while (s && (iter->net != s->sk_net));
+	} while (s && (iter->p.net != s->sk_net));
 	if (s)
 		return s;
 
@@ -1746,7 +1746,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 		for (; j <= hash->mask; j++) {
 			s = sk_head(&hash->table[j]);
-			while (s && (iter->net != s->sk_net))
+			while (s && (iter->p.net != s->sk_net))
 				s = sk_next(s);
 			if (s) {
 				iter->link = i;
@@ -1802,27 +1802,8 @@ static const struct seq_operations netlink_seq_ops = {
 
 static int netlink_seq_open(struct inode *inode, struct file *file)
 {
-	struct nl_seq_iter *iter;
-
-	iter = __seq_open_private(file, &netlink_seq_ops, sizeof(*iter));
-	if (!iter)
-		return -ENOMEM;
-
-	iter->net = get_proc_net(inode);
-	if (!iter->net) {
-		seq_release_private(inode, file);
-		return -ENXIO;
-	}
-
-	return 0;
-}
-
-static int netlink_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct nl_seq_iter *iter = seq->private;
-	put_net(iter->net);
-	return seq_release_private(inode, file);
+	return seq_open_net(inode, file, &netlink_seq_ops,
+				sizeof(struct nl_seq_iter));
 }
 
 static const struct file_operations netlink_seq_fops = {
@@ -1830,7 +1811,7 @@ static const struct file_operations netlink_seq_fops = {
 	.open		= netlink_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= netlink_seq_release,
+	.release	= seq_release_net,
 };
 
 #endif
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 45e3cbcb2763..ace29f1c4c5b 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -1871,7 +1871,7 @@ static inline struct sock *packet_seq_idx(struct net *net, loff_t off)
 
 static void *packet_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	struct net *net = seq->private;
+	struct net *net = seq_file_net(seq);
 	read_lock(&net->packet_sklist_lock);
 	return *pos ? packet_seq_idx(net, *pos - 1) : SEQ_START_TOKEN;
 }
@@ -1924,26 +1924,8 @@ static const struct seq_operations packet_seq_ops = {
 
 static int packet_seq_open(struct inode *inode, struct file *file)
 {
-	struct seq_file *seq;
-	int res;
-	res = seq_open(file, &packet_seq_ops);
-	if (!res) {
-		seq = file->private_data;
-		seq->private = get_proc_net(inode);
-		if (!seq->private) {
-			seq_release(inode, file);
-			res = -ENXIO;
-		}
-	}
-	return res;
-}
-
-static int packet_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq=  file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return seq_release(inode, file);
+	return seq_open_net(inode, file, &packet_seq_ops,
+			    sizeof(struct seq_net_private));
 }
 
 static const struct file_operations packet_seq_fops = {
@@ -1951,7 +1933,7 @@ static const struct file_operations packet_seq_fops = {
 	.open		= packet_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= packet_seq_release,
+	.release	= seq_release_net,
 };
 
 #endif
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index ecad42b72ad6..eacddb21a9b4 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2018,7 +2018,7 @@ static unsigned int unix_poll(struct file * file, struct socket *sock, poll_tabl
 
 #ifdef CONFIG_PROC_FS
 struct unix_iter_state {
-	struct net *net;
+	struct seq_net_private p;
 	int i;
 };
 static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos)
@@ -2027,7 +2027,7 @@ static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos)
 	struct sock *s;
 
 	for (s = first_unix_socket(&iter->i); s; s = next_unix_socket(&iter->i, s)) {
-		if (s->sk_net != iter->net)
+		if (s->sk_net != iter->p.net)
 			continue;
 		if (off == pos)
 			return s;
@@ -2054,7 +2054,7 @@ static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 		sk = first_unix_socket(&iter->i);
 	else
 		sk = next_unix_socket(&iter->i, sk);
-	while (sk && (sk->sk_net != iter->net))
+	while (sk && (sk->sk_net != iter->p.net))
 		sk = next_unix_socket(&iter->i, sk);
 	return sk;
 }
@@ -2118,27 +2118,8 @@ static const struct seq_operations unix_seq_ops = {
 
 static int unix_seq_open(struct inode *inode, struct file *file)
 {
-	struct unix_iter_state *it;
-
-	it = __seq_open_private(file, &unix_seq_ops,
-				sizeof(struct unix_iter_state));
-	if (it == NULL)
-		return -ENOMEM;
-
-	it->net = get_proc_net(inode);
-	if (it->net == NULL) {
-		seq_release_private(inode, file);
-		return -ENXIO;
-	}
-	return 0;
-}
-
-static int unix_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct unix_iter_state *iter = seq->private;
-	put_net(iter->net);
-	return seq_release_private(inode, file);
+	return seq_open_net(inode, file, &unix_seq_ops,
+			    sizeof(struct unix_iter_state));
 }
 
 static const struct file_operations unix_seq_fops = {
@@ -2146,7 +2127,7 @@ static const struct file_operations unix_seq_fops = {
 	.open		= unix_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= unix_seq_release,
+	.release	= seq_release_net,
 };
 
 #endif
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index db03ed5ce055..1e4cf615f878 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -673,26 +673,8 @@ static const struct seq_operations wireless_seq_ops = {
 
 static int wireless_seq_open(struct inode *inode, struct file *file)
 {
-	struct seq_file *seq;
-	int res;
-	res = seq_open(file, &wireless_seq_ops);
-	if (!res) {
-		seq = file->private_data;
-		seq->private = get_proc_net(inode);
-		if (!seq->private) {
-			seq_release(inode, file);
-			res = -ENXIO;
-		}
-	}
-	return res;
-}
-
-static int wireless_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return seq_release(inode, file);
+	return seq_open_net(inode, file, &wireless_seq_ops,
+			    sizeof(struct seq_net_private));
 }
 
 static const struct file_operations wireless_seq_fops = {
@@ -700,7 +682,7 @@ static const struct file_operations wireless_seq_fops = {
 	.open    = wireless_seq_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
-	.release = wireless_seq_release,
+	.release = seq_release_net,
 };
 
 int wext_proc_init(struct net *net)
-- 
cgit v1.2.3


From 7bc54c90307b4bc3d7fb2ffd6ad8fbda0671a45e Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:35:07 -0800
Subject: [IPv4] RAW: Compact the API for the kernel

The raw sockets functions are explicitly used from
inside the kernel in two places:

1. in ip_local_deliver_finish to intercept skb-s
2. in icmp_error

For this purposes many functions and even data structures,
that are naturally internal for raw protocol, are exported.

Compact the API to two functions and hide all the other
(including hash table and rwlock) inside the net/ipv4/raw.c

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h   | 19 +++----------------
 net/ipv4/icmp.c     | 15 +--------------
 net/ipv4/ip_input.c | 16 ++++------------
 net/ipv4/raw.c      | 53 ++++++++++++++++++++++++++++++++++++++++++++++++-----
 4 files changed, 56 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index e4af59781949..7fc3c770f170 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -22,23 +22,10 @@
 
 extern struct proto raw_prot;
 
-extern void 	raw_err(struct sock *, struct sk_buff *, u32 info);
-extern int 	raw_rcv(struct sock *, struct sk_buff *);
-
-/* Note: v4 ICMP wants to get at this stuff, if you change the
- *       hashing mechanism, make sure you update icmp.c as well.
- */
-#define RAWV4_HTABLE_SIZE	MAX_INET_PROTOS
-extern struct hlist_head raw_v4_htable[RAWV4_HTABLE_SIZE];
-
-extern rwlock_t raw_v4_lock;
+void raw_icmp_error(struct sk_buff *, int, u32);
+int raw_local_deliver(struct sk_buff *, int);
 
-
-extern struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
-				    __be32 raddr, __be32 laddr,
-				    int dif);
-
-extern int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash);
+extern int 	raw_rcv(struct sock *, struct sk_buff *);
 
 #ifdef CONFIG_PROC_FS
 extern int  raw_proc_init(void);
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 82baea026484..13d74598d3e4 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -603,7 +603,6 @@ static void icmp_unreach(struct sk_buff *skb)
 	struct icmphdr *icmph;
 	int hash, protocol;
 	struct net_protocol *ipprot;
-	struct sock *raw_sk;
 	u32 info = 0;
 
 	/*
@@ -697,21 +696,9 @@ static void icmp_unreach(struct sk_buff *skb)
 	/*
 	 *	Deliver ICMP message to raw sockets. Pretty useless feature?
 	 */
+	raw_icmp_error(skb, protocol, info);
 
-	/* Note: See raw.c and net/raw.h, RAWV4_HTABLE_SIZE==MAX_INET_PROTOS */
 	hash = protocol & (MAX_INET_PROTOS - 1);
-	read_lock(&raw_v4_lock);
-	if ((raw_sk = sk_head(&raw_v4_htable[hash])) != NULL) {
-		while ((raw_sk = __raw_v4_lookup(raw_sk, protocol, iph->daddr,
-						 iph->saddr,
-						 skb->dev->ifindex)) != NULL) {
-			raw_err(raw_sk, skb, info);
-			raw_sk = sk_next(raw_sk);
-			iph = (struct iphdr *)skb->data;
-		}
-	}
-	read_unlock(&raw_v4_lock);
-
 	rcu_read_lock();
 	ipprot = rcu_dereference(inet_protos[hash]);
 	if (ipprot && ipprot->err_handler)
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 4068e178d747..65631391d479 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -204,22 +204,14 @@ static int ip_local_deliver_finish(struct sk_buff *skb)
 
 	rcu_read_lock();
 	{
-		/* Note: See raw.c and net/raw.h, RAWV4_HTABLE_SIZE==MAX_INET_PROTOS */
 		int protocol = ip_hdr(skb)->protocol;
-		int hash;
-		struct sock *raw_sk;
+		int hash, raw;
 		struct net_protocol *ipprot;
 
 	resubmit:
-		hash = protocol & (MAX_INET_PROTOS - 1);
-		raw_sk = sk_head(&raw_v4_htable[hash]);
-
-		/* If there maybe a raw socket we must check - if not we
-		 * don't care less
-		 */
-		if (raw_sk && !raw_v4_input(skb, ip_hdr(skb), hash))
-			raw_sk = NULL;
+		raw = raw_local_deliver(skb, protocol);
 
+		hash = protocol & (MAX_INET_PROTOS - 1);
 		if ((ipprot = rcu_dereference(inet_protos[hash])) != NULL) {
 			int ret;
 
@@ -237,7 +229,7 @@ static int ip_local_deliver_finish(struct sk_buff *skb)
 			}
 			IP_INC_STATS_BH(IPSTATS_MIB_INDELIVERS);
 		} else {
-			if (!raw_sk) {
+			if (!raw) {
 				if (xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
 					IP_INC_STATS_BH(IPSTATS_MIB_INUNKNOWNPROTOS);
 					icmp_send(skb, ICMP_DEST_UNREACH,
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index b80987d2fc55..8a506618b912 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -80,8 +80,10 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter_ipv4.h>
 
-struct hlist_head raw_v4_htable[RAWV4_HTABLE_SIZE];
-DEFINE_RWLOCK(raw_v4_lock);
+#define RAWV4_HTABLE_SIZE	MAX_INET_PROTOS
+
+static struct hlist_head raw_v4_htable[RAWV4_HTABLE_SIZE];
+static DEFINE_RWLOCK(raw_v4_lock);
 
 static void raw_v4_hash(struct sock *sk)
 {
@@ -102,7 +104,7 @@ static void raw_v4_unhash(struct sock *sk)
 	write_unlock_bh(&raw_v4_lock);
 }
 
-struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
+static struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
 			     __be32 raddr, __be32 laddr,
 			     int dif)
 {
@@ -150,7 +152,7 @@ static __inline__ int icmp_filter(struct sock *sk, struct sk_buff *skb)
  * RFC 1122: SHOULD pass TOS value up to the transport layer.
  * -> It does. And not only TOS, but all IP header.
  */
-int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash)
+static int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash)
 {
 	struct sock *sk;
 	struct hlist_head *head;
@@ -182,7 +184,25 @@ out:
 	return delivered;
 }
 
-void raw_err (struct sock *sk, struct sk_buff *skb, u32 info)
+int raw_local_deliver(struct sk_buff *skb, int protocol)
+{
+	int hash;
+	struct sock *raw_sk;
+
+	hash = protocol & (RAWV4_HTABLE_SIZE - 1);
+	raw_sk = sk_head(&raw_v4_htable[hash]);
+
+	/* If there maybe a raw socket we must check - if not we
+	 * don't care less
+	 */
+	if (raw_sk && !raw_v4_input(skb, ip_hdr(skb), hash))
+		raw_sk = NULL;
+
+	return raw_sk != NULL;
+
+}
+
+static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	const int type = icmp_hdr(skb)->type;
@@ -236,6 +256,29 @@ void raw_err (struct sock *sk, struct sk_buff *skb, u32 info)
 	}
 }
 
+void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info)
+{
+	int hash;
+	struct sock *raw_sk;
+	struct iphdr *iph;
+
+	hash = protocol & (RAWV4_HTABLE_SIZE - 1);
+
+	read_lock(&raw_v4_lock);
+	raw_sk = sk_head(&raw_v4_htable[hash]);
+	if (raw_sk != NULL) {
+		iph = (struct iphdr *)skb->data;
+		while ((raw_sk = __raw_v4_lookup(raw_sk, protocol, iph->daddr,
+						iph->saddr,
+						skb->dev->ifindex)) != NULL) {
+			raw_err(raw_sk, skb, info);
+			raw_sk = sk_next(raw_sk);
+			iph = (struct iphdr *)skb->data;
+		}
+	}
+	read_unlock(&raw_v4_lock);
+}
+
 static int raw_rcv_skb(struct sock * sk, struct sk_buff * skb)
 {
 	/* Charge it to the socket. */
-- 
cgit v1.2.3


From 69d6da0b0faa70249a243a14e6066c013e9294e5 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:35:57 -0800
Subject: [IPv6] RAW: Compact the API for the kernel

Same as in the previous patch for ipv4, compact the
API and hide hash table and rwlock inside the raw.c
file.

Plus fix some "bad" places from checkpatch.pl point
of view (assignments inside if()).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/rawv6.h  | 19 +++----------------
 net/ipv6/icmp.c      | 15 +--------------
 net/ipv6/ip6_input.c |  9 +++------
 net/ipv6/raw.c       | 52 ++++++++++++++++++++++++++++++++++++++++++++--------
 4 files changed, 51 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/net/rawv6.h b/include/net/rawv6.h
index a5819891d525..8a22599f26ba 100644
--- a/include/net/rawv6.h
+++ b/include/net/rawv6.h
@@ -5,26 +5,13 @@
 
 #include <net/protocol.h>
 
-#define RAWV6_HTABLE_SIZE	MAX_INET_PROTOS
-extern struct hlist_head raw_v6_htable[RAWV6_HTABLE_SIZE];
-extern rwlock_t raw_v6_lock;
-
-extern int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr);
-
-extern struct sock *__raw_v6_lookup(struct sock *sk, unsigned short num,
-				    struct in6_addr *loc_addr, struct in6_addr *rmt_addr,
-				    int dif);
+void raw6_icmp_error(struct sk_buff *, int nexthdr,
+		int type, int code, int inner_offset, __be32);
+int raw6_local_deliver(struct sk_buff *, int);
 
 extern int			rawv6_rcv(struct sock *sk,
 					  struct sk_buff *skb);
 
-
-extern void			rawv6_err(struct sock *sk,
-					  struct sk_buff *skb,
-					  struct inet6_skb_parm *opt,
-					  int type, int code, 
-					  int offset, __be32 info);
-
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
 int rawv6_mh_filter_register(int (*filter)(struct sock *sock,
 					   struct sk_buff *skb));
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index f1240688dc58..93c96cfd5ee1 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -555,9 +555,7 @@ out:
 
 static void icmpv6_notify(struct sk_buff *skb, int type, int code, __be32 info)
 {
-	struct in6_addr *saddr, *daddr;
 	struct inet6_protocol *ipprot;
-	struct sock *sk;
 	int inner_offset;
 	int hash;
 	u8 nexthdr;
@@ -579,9 +577,6 @@ static void icmpv6_notify(struct sk_buff *skb, int type, int code, __be32 info)
 	if (!pskb_may_pull(skb, inner_offset+8))
 		return;
 
-	saddr = &ipv6_hdr(skb)->saddr;
-	daddr = &ipv6_hdr(skb)->daddr;
-
 	/* BUGGG_FUTURE: we should try to parse exthdrs in this packet.
 	   Without this we will not able f.e. to make source routed
 	   pmtu discovery.
@@ -597,15 +592,7 @@ static void icmpv6_notify(struct sk_buff *skb, int type, int code, __be32 info)
 		ipprot->err_handler(skb, NULL, type, code, inner_offset, info);
 	rcu_read_unlock();
 
-	read_lock(&raw_v6_lock);
-	if ((sk = sk_head(&raw_v6_htable[hash])) != NULL) {
-		while ((sk = __raw_v6_lookup(sk, nexthdr, saddr, daddr,
-					    IP6CB(skb)->iif))) {
-			rawv6_err(sk, skb, NULL, type, code, inner_offset, info);
-			sk = sk_next(sk);
-		}
-	}
-	read_unlock(&raw_v6_lock);
+	raw6_icmp_error(skb, nexthdr, type, code, inner_offset, info);
 }
 
 /*
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 79610b4bad3e..178aebc0427a 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -153,9 +153,8 @@ out:
 static int ip6_input_finish(struct sk_buff *skb)
 {
 	struct inet6_protocol *ipprot;
-	struct sock *raw_sk;
 	unsigned int nhoff;
-	int nexthdr;
+	int nexthdr, raw;
 	u8 hash;
 	struct inet6_dev *idev;
 
@@ -171,9 +170,7 @@ resubmit:
 	nhoff = IP6CB(skb)->nhoff;
 	nexthdr = skb_network_header(skb)[nhoff];
 
-	raw_sk = sk_head(&raw_v6_htable[nexthdr & (MAX_INET_PROTOS - 1)]);
-	if (raw_sk && !ipv6_raw_deliver(skb, nexthdr))
-		raw_sk = NULL;
+	raw = raw6_local_deliver(skb, nexthdr);
 
 	hash = nexthdr & (MAX_INET_PROTOS - 1);
 	if ((ipprot = rcu_dereference(inet6_protos[hash])) != NULL) {
@@ -206,7 +203,7 @@ resubmit:
 		else if (ret == 0)
 			IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDELIVERS);
 	} else {
-		if (!raw_sk) {
+		if (!raw) {
 			if (xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
 				IP6_INC_STATS_BH(idev, IPSTATS_MIB_INUNKNOWNPROTOS);
 				icmpv6_send(skb, ICMPV6_PARAMPROB,
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index ad622cc11bda..53f01b4982c7 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -60,8 +60,10 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 
-struct hlist_head raw_v6_htable[RAWV6_HTABLE_SIZE];
-DEFINE_RWLOCK(raw_v6_lock);
+#define RAWV6_HTABLE_SIZE	MAX_INET_PROTOS
+
+static struct hlist_head raw_v6_htable[RAWV6_HTABLE_SIZE];
+static DEFINE_RWLOCK(raw_v6_lock);
 
 static void raw_v6_hash(struct sock *sk)
 {
@@ -83,10 +85,8 @@ static void raw_v6_unhash(struct sock *sk)
 }
 
 
-/* Grumble... icmp and ip_input want to get at this... */
-struct sock *__raw_v6_lookup(struct sock *sk, unsigned short num,
-			     struct in6_addr *loc_addr, struct in6_addr *rmt_addr,
-			     int dif)
+static struct sock *__raw_v6_lookup(struct sock *sk, unsigned short num,
+		struct in6_addr *loc_addr, struct in6_addr *rmt_addr, int dif)
 {
 	struct hlist_node *node;
 	int is_multicast = ipv6_addr_is_multicast(loc_addr);
@@ -167,7 +167,7 @@ EXPORT_SYMBOL(rawv6_mh_filter_unregister);
  *
  *	Caller owns SKB so we must make clones.
  */
-int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
+static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 {
 	struct in6_addr *saddr;
 	struct in6_addr *daddr;
@@ -242,6 +242,17 @@ out:
 	return delivered;
 }
 
+int raw6_local_deliver(struct sk_buff *skb, int nexthdr)
+{
+	struct sock *raw_sk;
+
+	raw_sk = sk_head(&raw_v6_htable[nexthdr & (MAX_INET_PROTOS - 1)]);
+	if (raw_sk && !ipv6_raw_deliver(skb, nexthdr))
+		raw_sk = NULL;
+
+	return raw_sk != NULL;
+}
+
 /* This cleans up af_inet6 a bit. -DaveM */
 static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
@@ -316,7 +327,7 @@ out:
 	return err;
 }
 
-void rawv6_err(struct sock *sk, struct sk_buff *skb,
+static void rawv6_err(struct sock *sk, struct sk_buff *skb,
 	       struct inet6_skb_parm *opt,
 	       int type, int code, int offset, __be32 info)
 {
@@ -350,6 +361,31 @@ void rawv6_err(struct sock *sk, struct sk_buff *skb,
 	}
 }
 
+void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
+		int type, int code, int inner_offset, __be32 info)
+{
+	struct sock *sk;
+	int hash;
+	struct in6_addr *saddr, *daddr;
+
+	hash = nexthdr & (RAWV6_HTABLE_SIZE - 1);
+
+	read_lock(&raw_v6_lock);
+	sk = sk_head(&raw_v6_htable[hash]);
+	if (sk != NULL) {
+		saddr = &ipv6_hdr(skb)->saddr;
+		daddr = &ipv6_hdr(skb)->daddr;
+
+		while ((sk = __raw_v6_lookup(sk, nexthdr, saddr, daddr,
+						IP6CB(skb)->iif))) {
+			rawv6_err(sk, skb, NULL, type, code,
+					inner_offset, info);
+			sk = sk_next(sk);
+		}
+	}
+	read_unlock(&raw_v6_lock);
+}
+
 static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
 {
 	if ((raw6_sk(sk)->checksum || sk->sk_filter) &&
-- 
cgit v1.2.3


From b673e4dfc8f29e5bfe4d342029b793e9d504f6dd Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:36:45 -0800
Subject: [RAW]: Introduce raw_hashinfo structure

The ipv4/raw.c and ipv6/raw.c contain many common code (most
of which is proc interface) which can be consolidated.

Most of the places to consolidate deal with the raw sockets
hashtable, so introduce a struct raw_hashinfo which describes
the raw sockets hash.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h |  7 +++++++
 net/ipv4/raw.c    | 50 +++++++++++++++++++++++++-------------------------
 net/ipv6/raw.c    | 49 +++++++++++++++++++++++++------------------------
 3 files changed, 57 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 7fc3c770f170..70b27c7c8167 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -27,6 +27,13 @@ int raw_local_deliver(struct sk_buff *, int);
 
 extern int 	raw_rcv(struct sock *, struct sk_buff *);
 
+#define RAW_HTABLE_SIZE	MAX_INET_PROTOS
+
+struct raw_hashinfo {
+	rwlock_t lock;
+	struct hlist_head ht[RAW_HTABLE_SIZE];
+};
+
 #ifdef CONFIG_PROC_FS
 extern int  raw_proc_init(void);
 extern void raw_proc_exit(void);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 8a506618b912..dd9f00b3ab42 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -80,28 +80,27 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter_ipv4.h>
 
-#define RAWV4_HTABLE_SIZE	MAX_INET_PROTOS
-
-static struct hlist_head raw_v4_htable[RAWV4_HTABLE_SIZE];
-static DEFINE_RWLOCK(raw_v4_lock);
+static struct raw_hashinfo raw_v4_hashinfo = {
+	.lock = __RW_LOCK_UNLOCKED(),
+};
 
 static void raw_v4_hash(struct sock *sk)
 {
-	struct hlist_head *head = &raw_v4_htable[inet_sk(sk)->num &
-						 (RAWV4_HTABLE_SIZE - 1)];
+	struct hlist_head *head = &raw_v4_hashinfo.ht[inet_sk(sk)->num &
+						 (RAW_HTABLE_SIZE - 1)];
 
-	write_lock_bh(&raw_v4_lock);
+	write_lock_bh(&raw_v4_hashinfo.lock);
 	sk_add_node(sk, head);
 	sock_prot_inc_use(sk->sk_prot);
-	write_unlock_bh(&raw_v4_lock);
+	write_unlock_bh(&raw_v4_hashinfo.lock);
 }
 
 static void raw_v4_unhash(struct sock *sk)
 {
-	write_lock_bh(&raw_v4_lock);
+	write_lock_bh(&raw_v4_hashinfo.lock);
 	if (sk_del_node_init(sk))
 		sock_prot_dec_use(sk->sk_prot);
-	write_unlock_bh(&raw_v4_lock);
+	write_unlock_bh(&raw_v4_hashinfo.lock);
 }
 
 static struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
@@ -158,8 +157,8 @@ static int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash)
 	struct hlist_head *head;
 	int delivered = 0;
 
-	read_lock(&raw_v4_lock);
-	head = &raw_v4_htable[hash];
+	read_lock(&raw_v4_hashinfo.lock);
+	head = &raw_v4_hashinfo.ht[hash];
 	if (hlist_empty(head))
 		goto out;
 	sk = __raw_v4_lookup(__sk_head(head), iph->protocol,
@@ -180,7 +179,7 @@ static int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash)
 				     skb->dev->ifindex);
 	}
 out:
-	read_unlock(&raw_v4_lock);
+	read_unlock(&raw_v4_hashinfo.lock);
 	return delivered;
 }
 
@@ -189,8 +188,8 @@ int raw_local_deliver(struct sk_buff *skb, int protocol)
 	int hash;
 	struct sock *raw_sk;
 
-	hash = protocol & (RAWV4_HTABLE_SIZE - 1);
-	raw_sk = sk_head(&raw_v4_htable[hash]);
+	hash = protocol & (RAW_HTABLE_SIZE - 1);
+	raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]);
 
 	/* If there maybe a raw socket we must check - if not we
 	 * don't care less
@@ -262,10 +261,10 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info)
 	struct sock *raw_sk;
 	struct iphdr *iph;
 
-	hash = protocol & (RAWV4_HTABLE_SIZE - 1);
+	hash = protocol & (RAW_HTABLE_SIZE - 1);
 
-	read_lock(&raw_v4_lock);
-	raw_sk = sk_head(&raw_v4_htable[hash]);
+	read_lock(&raw_v4_hashinfo.lock);
+	raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]);
 	if (raw_sk != NULL) {
 		iph = (struct iphdr *)skb->data;
 		while ((raw_sk = __raw_v4_lookup(raw_sk, protocol, iph->daddr,
@@ -276,7 +275,7 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info)
 			iph = (struct iphdr *)skb->data;
 		}
 	}
-	read_unlock(&raw_v4_lock);
+	read_unlock(&raw_v4_hashinfo.lock);
 }
 
 static int raw_rcv_skb(struct sock * sk, struct sk_buff * skb)
@@ -844,10 +843,11 @@ static struct sock *raw_get_first(struct seq_file *seq)
 	struct sock *sk;
 	struct raw_iter_state* state = raw_seq_private(seq);
 
-	for (state->bucket = 0; state->bucket < RAWV4_HTABLE_SIZE; ++state->bucket) {
+	for (state->bucket = 0; state->bucket < RAW_HTABLE_SIZE;
+			++state->bucket) {
 		struct hlist_node *node;
 
-		sk_for_each(sk, node, &raw_v4_htable[state->bucket])
+		sk_for_each(sk, node, &raw_v4_hashinfo.ht[state->bucket])
 			if (sk->sk_family == PF_INET)
 				goto found;
 	}
@@ -866,8 +866,8 @@ try_again:
 		;
 	} while (sk && sk->sk_family != PF_INET);
 
-	if (!sk && ++state->bucket < RAWV4_HTABLE_SIZE) {
-		sk = sk_head(&raw_v4_htable[state->bucket]);
+	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
+		sk = sk_head(&raw_v4_hashinfo.ht[state->bucket]);
 		goto try_again;
 	}
 	return sk;
@@ -885,7 +885,7 @@ static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos)
 
 static void *raw_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	read_lock(&raw_v4_lock);
+	read_lock(&raw_v4_hashinfo.lock);
 	return *pos ? raw_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 }
 
@@ -903,7 +903,7 @@ static void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 static void raw_seq_stop(struct seq_file *seq, void *v)
 {
-	read_unlock(&raw_v4_lock);
+	read_unlock(&raw_v4_hashinfo.lock);
 }
 
 static __inline__ char *get_raw_sock(struct sock *sp, char *tmpbuf, int i)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 53f01b4982c7..15c72a6365a6 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -54,34 +54,34 @@
 #include <net/mip6.h>
 #endif
 
+#include <net/raw.h>
 #include <net/rawv6.h>
 #include <net/xfrm.h>
 
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 
-#define RAWV6_HTABLE_SIZE	MAX_INET_PROTOS
-
-static struct hlist_head raw_v6_htable[RAWV6_HTABLE_SIZE];
-static DEFINE_RWLOCK(raw_v6_lock);
+static struct raw_hashinfo raw_v6_hashinfo = {
+	.lock = __RW_LOCK_UNLOCKED(),
+};
 
 static void raw_v6_hash(struct sock *sk)
 {
-	struct hlist_head *list = &raw_v6_htable[inet_sk(sk)->num &
-						 (RAWV6_HTABLE_SIZE - 1)];
+	struct hlist_head *list = &raw_v6_hashinfo.ht[inet_sk(sk)->num &
+						 (RAW_HTABLE_SIZE - 1)];
 
-	write_lock_bh(&raw_v6_lock);
+	write_lock_bh(&raw_v6_hashinfo.lock);
 	sk_add_node(sk, list);
 	sock_prot_inc_use(sk->sk_prot);
-	write_unlock_bh(&raw_v6_lock);
+	write_unlock_bh(&raw_v6_hashinfo.lock);
 }
 
 static void raw_v6_unhash(struct sock *sk)
 {
-	write_lock_bh(&raw_v6_lock);
+	write_lock_bh(&raw_v6_hashinfo.lock);
 	if (sk_del_node_init(sk))
 		sock_prot_dec_use(sk->sk_prot);
-	write_unlock_bh(&raw_v6_lock);
+	write_unlock_bh(&raw_v6_hashinfo.lock);
 }
 
 
@@ -180,8 +180,8 @@ static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 
 	hash = nexthdr & (MAX_INET_PROTOS - 1);
 
-	read_lock(&raw_v6_lock);
-	sk = sk_head(&raw_v6_htable[hash]);
+	read_lock(&raw_v6_hashinfo.lock);
+	sk = sk_head(&raw_v6_hashinfo.ht[hash]);
 
 	/*
 	 *	The first socket found will be delivered after
@@ -238,7 +238,7 @@ static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 				     IP6CB(skb)->iif);
 	}
 out:
-	read_unlock(&raw_v6_lock);
+	read_unlock(&raw_v6_hashinfo.lock);
 	return delivered;
 }
 
@@ -246,7 +246,7 @@ int raw6_local_deliver(struct sk_buff *skb, int nexthdr)
 {
 	struct sock *raw_sk;
 
-	raw_sk = sk_head(&raw_v6_htable[nexthdr & (MAX_INET_PROTOS - 1)]);
+	raw_sk = sk_head(&raw_v6_hashinfo.ht[nexthdr & (MAX_INET_PROTOS - 1)]);
 	if (raw_sk && !ipv6_raw_deliver(skb, nexthdr))
 		raw_sk = NULL;
 
@@ -368,10 +368,10 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
 	int hash;
 	struct in6_addr *saddr, *daddr;
 
-	hash = nexthdr & (RAWV6_HTABLE_SIZE - 1);
+	hash = nexthdr & (RAW_HTABLE_SIZE - 1);
 
-	read_lock(&raw_v6_lock);
-	sk = sk_head(&raw_v6_htable[hash]);
+	read_lock(&raw_v6_hashinfo.lock);
+	sk = sk_head(&raw_v6_hashinfo.ht[hash]);
 	if (sk != NULL) {
 		saddr = &ipv6_hdr(skb)->saddr;
 		daddr = &ipv6_hdr(skb)->daddr;
@@ -383,7 +383,7 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
 			sk = sk_next(sk);
 		}
 	}
-	read_unlock(&raw_v6_lock);
+	read_unlock(&raw_v6_hashinfo.lock);
 }
 
 static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
@@ -1221,8 +1221,9 @@ static struct sock *raw6_get_first(struct seq_file *seq)
 	struct hlist_node *node;
 	struct raw6_iter_state* state = raw6_seq_private(seq);
 
-	for (state->bucket = 0; state->bucket < RAWV6_HTABLE_SIZE; ++state->bucket)
-		sk_for_each(sk, node, &raw_v6_htable[state->bucket])
+	for (state->bucket = 0; state->bucket < RAW_HTABLE_SIZE;
+			++state->bucket)
+		sk_for_each(sk, node, &raw_v6_hashinfo.ht[state->bucket])
 			if (sk->sk_family == PF_INET6)
 				goto out;
 	sk = NULL;
@@ -1240,8 +1241,8 @@ try_again:
 		;
 	} while (sk && sk->sk_family != PF_INET6);
 
-	if (!sk && ++state->bucket < RAWV6_HTABLE_SIZE) {
-		sk = sk_head(&raw_v6_htable[state->bucket]);
+	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
+		sk = sk_head(&raw_v6_hashinfo.ht[state->bucket]);
 		goto try_again;
 	}
 	return sk;
@@ -1258,7 +1259,7 @@ static struct sock *raw6_get_idx(struct seq_file *seq, loff_t pos)
 
 static void *raw6_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	read_lock(&raw_v6_lock);
+	read_lock(&raw_v6_hashinfo.lock);
 	return *pos ? raw6_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 }
 
@@ -1276,7 +1277,7 @@ static void *raw6_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 static void raw6_seq_stop(struct seq_file *seq, void *v)
 {
-	read_unlock(&raw_v6_lock);
+	read_unlock(&raw_v6_hashinfo.lock);
 }
 
 static void raw6_sock_seq_show(struct seq_file *seq, struct sock *sp, int i)
-- 
cgit v1.2.3


From 65b4c50b47c4ac3d2b5a82e5553b8e5613fb9585 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:37:24 -0800
Subject: [RAW]: Consolidate proto->hash callback

Having the raw_hashinfo it's easy to consolidate the
raw[46]_hash functions.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h |  2 ++
 net/ipv4/raw.c    | 17 ++++++++++++-----
 net/ipv6/raw.c    |  8 +-------
 3 files changed, 15 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 70b27c7c8167..43f1e3c07380 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -39,4 +39,6 @@ extern int  raw_proc_init(void);
 extern void raw_proc_exit(void);
 #endif
 
+void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h);
+
 #endif	/* _RAW_H */
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index dd9f00b3ab42..28b95ae5d2dc 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -84,15 +84,22 @@ static struct raw_hashinfo raw_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(),
 };
 
-static void raw_v4_hash(struct sock *sk)
+void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h)
 {
-	struct hlist_head *head = &raw_v4_hashinfo.ht[inet_sk(sk)->num &
-						 (RAW_HTABLE_SIZE - 1)];
+	struct hlist_head *head;
 
-	write_lock_bh(&raw_v4_hashinfo.lock);
+	head = &h->ht[inet_sk(sk)->num & (RAW_HTABLE_SIZE - 1)];
+
+	write_lock_bh(&h->lock);
 	sk_add_node(sk, head);
 	sock_prot_inc_use(sk->sk_prot);
-	write_unlock_bh(&raw_v4_hashinfo.lock);
+	write_unlock_bh(&h->lock);
+}
+EXPORT_SYMBOL_GPL(raw_hash_sk);
+
+static void raw_v4_hash(struct sock *sk)
+{
+	raw_hash_sk(sk, &raw_v4_hashinfo);
 }
 
 static void raw_v4_unhash(struct sock *sk)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 15c72a6365a6..70db6f494538 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -67,13 +67,7 @@ static struct raw_hashinfo raw_v6_hashinfo = {
 
 static void raw_v6_hash(struct sock *sk)
 {
-	struct hlist_head *list = &raw_v6_hashinfo.ht[inet_sk(sk)->num &
-						 (RAW_HTABLE_SIZE - 1)];
-
-	write_lock_bh(&raw_v6_hashinfo.lock);
-	sk_add_node(sk, list);
-	sock_prot_inc_use(sk->sk_prot);
-	write_unlock_bh(&raw_v6_hashinfo.lock);
+	raw_hash_sk(sk, &raw_v6_hashinfo);
 }
 
 static void raw_v6_unhash(struct sock *sk)
-- 
cgit v1.2.3


From ab70768ec78c6784958bab3b58fbe3f4150006df Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:37:58 -0800
Subject: [RAW]: Consolidate proto->unhash callback

Same as the ->hash one, this is easily consolidated.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h |  1 +
 net/ipv4/raw.c    | 14 ++++++++++----
 net/ipv6/raw.c    |  5 +----
 3 files changed, 12 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 43f1e3c07380..81a1773b58ba 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -40,5 +40,6 @@ extern void raw_proc_exit(void);
 #endif
 
 void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h);
+void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h);
 
 #endif	/* _RAW_H */
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 28b95ae5d2dc..d24501a342a4 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -97,6 +97,15 @@ void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h)
 }
 EXPORT_SYMBOL_GPL(raw_hash_sk);
 
+void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h)
+{
+	write_lock_bh(&h->lock);
+	if (sk_del_node_init(sk))
+		sock_prot_dec_use(sk->sk_prot);
+	write_unlock_bh(&h->lock);
+}
+EXPORT_SYMBOL_GPL(raw_unhash_sk);
+
 static void raw_v4_hash(struct sock *sk)
 {
 	raw_hash_sk(sk, &raw_v4_hashinfo);
@@ -104,10 +113,7 @@ static void raw_v4_hash(struct sock *sk)
 
 static void raw_v4_unhash(struct sock *sk)
 {
-	write_lock_bh(&raw_v4_hashinfo.lock);
-	if (sk_del_node_init(sk))
-		sock_prot_dec_use(sk->sk_prot);
-	write_unlock_bh(&raw_v4_hashinfo.lock);
+	raw_unhash_sk(sk, &raw_v4_hashinfo);
 }
 
 static struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 70db6f494538..422d27cfbe16 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -72,10 +72,7 @@ static void raw_v6_hash(struct sock *sk)
 
 static void raw_v6_unhash(struct sock *sk)
 {
-	write_lock_bh(&raw_v6_hashinfo.lock);
-	if (sk_del_node_init(sk))
-		sock_prot_dec_use(sk->sk_prot);
-	write_unlock_bh(&raw_v6_hashinfo.lock);
+	raw_unhash_sk(sk, &raw_v6_hashinfo);
 }
 
 
-- 
cgit v1.2.3


From 42a73808ed4f30b739eb52bcbb33a02fe62ceef5 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 Nov 2007 22:38:33 -0800
Subject: [RAW]: Consolidate proc interface.

Both ipv6/raw.c and ipv4/raw.c use the seq files to walk
through the raw sockets hash and show them.

The "walking" code is rather huge, but is identical in both
cases. The difference is the hash table to walk over and
the protocol family to check (this was not in the first
virsion of the patch, which was noticed by YOSHIFUJI)

Make the ->open store the needed hash table and the family
on the allocated raw_iter_state and make the start/next/stop
callbacks work with it.

This removes most of the code.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h | 14 ++++++++++
 net/ipv4/raw.c    | 52 +++++++++++++++++++++++------------
 net/ipv6/raw.c    | 82 ++++---------------------------------------------------
 3 files changed, 53 insertions(+), 95 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 81a1773b58ba..4d1aba032bf5 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -37,6 +37,20 @@ struct raw_hashinfo {
 #ifdef CONFIG_PROC_FS
 extern int  raw_proc_init(void);
 extern void raw_proc_exit(void);
+
+struct raw_iter_state {
+	int bucket;
+	unsigned short family;
+	struct raw_hashinfo *h;
+};
+
+#define raw_seq_private(seq) ((struct raw_iter_state *)(seq)->private)
+void *raw_seq_start(struct seq_file *seq, loff_t *pos);
+void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos);
+void raw_seq_stop(struct seq_file *seq, void *v);
+int raw_seq_open(struct file *file, struct raw_hashinfo *h,
+		unsigned short family);
+
 #endif
 
 void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index d24501a342a4..2ff8214a530b 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -845,12 +845,6 @@ struct proto raw_prot = {
 };
 
 #ifdef CONFIG_PROC_FS
-struct raw_iter_state {
-	int bucket;
-};
-
-#define raw_seq_private(seq) ((struct raw_iter_state *)(seq)->private)
-
 static struct sock *raw_get_first(struct seq_file *seq)
 {
 	struct sock *sk;
@@ -860,8 +854,8 @@ static struct sock *raw_get_first(struct seq_file *seq)
 			++state->bucket) {
 		struct hlist_node *node;
 
-		sk_for_each(sk, node, &raw_v4_hashinfo.ht[state->bucket])
-			if (sk->sk_family == PF_INET)
+		sk_for_each(sk, node, &state->h->ht[state->bucket])
+			if (sk->sk_family == state->family)
 				goto found;
 	}
 	sk = NULL;
@@ -877,10 +871,10 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && sk->sk_family != PF_INET);
+	} while (sk && sk->sk_family != state->family);
 
 	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
-		sk = sk_head(&raw_v4_hashinfo.ht[state->bucket]);
+		sk = sk_head(&state->h->ht[state->bucket]);
 		goto try_again;
 	}
 	return sk;
@@ -896,13 +890,16 @@ static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos)
 	return pos ? NULL : sk;
 }
 
-static void *raw_seq_start(struct seq_file *seq, loff_t *pos)
+void *raw_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	read_lock(&raw_v4_hashinfo.lock);
+	struct raw_iter_state *state = raw_seq_private(seq);
+
+	read_lock(&state->h->lock);
 	return *pos ? raw_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 }
+EXPORT_SYMBOL_GPL(raw_seq_start);
 
-static void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	struct sock *sk;
 
@@ -913,11 +910,15 @@ static void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	++*pos;
 	return sk;
 }
+EXPORT_SYMBOL_GPL(raw_seq_next);
 
-static void raw_seq_stop(struct seq_file *seq, void *v)
+void raw_seq_stop(struct seq_file *seq, void *v)
 {
-	read_unlock(&raw_v4_hashinfo.lock);
+	struct raw_iter_state *state = raw_seq_private(seq);
+
+	read_unlock(&state->h->lock);
 }
+EXPORT_SYMBOL_GPL(raw_seq_stop);
 
 static __inline__ char *get_raw_sock(struct sock *sp, char *tmpbuf, int i)
 {
@@ -964,15 +965,30 @@ static const struct seq_operations raw_seq_ops = {
 	.show  = raw_seq_show,
 };
 
-static int raw_seq_open(struct inode *inode, struct file *file)
+int raw_seq_open(struct file *file, struct raw_hashinfo *h,
+		unsigned short family)
 {
-	return seq_open_private(file, &raw_seq_ops,
+	struct raw_iter_state *i;
+
+	i = __seq_open_private(file, &raw_seq_ops,
 			sizeof(struct raw_iter_state));
+	if (i == NULL)
+		return -ENOMEM;
+
+	i->h = h;
+	i->family = family;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(raw_seq_open);
+
+static int raw_v4_seq_open(struct inode *inode, struct file *file)
+{
+	return raw_seq_open(file, &raw_v4_hashinfo, PF_INET);
 }
 
 static const struct file_operations raw_seq_fops = {
 	.owner	 = THIS_MODULE,
-	.open	 = raw_seq_open,
+	.open	 = raw_v4_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
 	.release = seq_release_private,
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 422d27cfbe16..b34631e1b015 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1200,77 +1200,6 @@ struct proto rawv6_prot = {
 };
 
 #ifdef CONFIG_PROC_FS
-struct raw6_iter_state {
-	int bucket;
-};
-
-#define raw6_seq_private(seq) ((struct raw6_iter_state *)(seq)->private)
-
-static struct sock *raw6_get_first(struct seq_file *seq)
-{
-	struct sock *sk;
-	struct hlist_node *node;
-	struct raw6_iter_state* state = raw6_seq_private(seq);
-
-	for (state->bucket = 0; state->bucket < RAW_HTABLE_SIZE;
-			++state->bucket)
-		sk_for_each(sk, node, &raw_v6_hashinfo.ht[state->bucket])
-			if (sk->sk_family == PF_INET6)
-				goto out;
-	sk = NULL;
-out:
-	return sk;
-}
-
-static struct sock *raw6_get_next(struct seq_file *seq, struct sock *sk)
-{
-	struct raw6_iter_state* state = raw6_seq_private(seq);
-
-	do {
-		sk = sk_next(sk);
-try_again:
-		;
-	} while (sk && sk->sk_family != PF_INET6);
-
-	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
-		sk = sk_head(&raw_v6_hashinfo.ht[state->bucket]);
-		goto try_again;
-	}
-	return sk;
-}
-
-static struct sock *raw6_get_idx(struct seq_file *seq, loff_t pos)
-{
-	struct sock *sk = raw6_get_first(seq);
-	if (sk)
-		while (pos && (sk = raw6_get_next(seq, sk)) != NULL)
-			--pos;
-	return pos ? NULL : sk;
-}
-
-static void *raw6_seq_start(struct seq_file *seq, loff_t *pos)
-{
-	read_lock(&raw_v6_hashinfo.lock);
-	return *pos ? raw6_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
-}
-
-static void *raw6_seq_next(struct seq_file *seq, void *v, loff_t *pos)
-{
-	struct sock *sk;
-
-	if (v == SEQ_START_TOKEN)
-		sk = raw6_get_first(seq);
-	else
-		sk = raw6_get_next(seq, v);
-	++*pos;
-	return sk;
-}
-
-static void raw6_seq_stop(struct seq_file *seq, void *v)
-{
-	read_unlock(&raw_v6_hashinfo.lock);
-}
-
 static void raw6_sock_seq_show(struct seq_file *seq, struct sock *sp, int i)
 {
 	struct ipv6_pinfo *np = inet6_sk(sp);
@@ -1308,21 +1237,20 @@ static int raw6_seq_show(struct seq_file *seq, void *v)
 			   "st tx_queue rx_queue tr tm->when retrnsmt"
 			   "   uid  timeout inode  drops\n");
 	else
-		raw6_sock_seq_show(seq, v, raw6_seq_private(seq)->bucket);
+		raw6_sock_seq_show(seq, v, raw_seq_private(seq)->bucket);
 	return 0;
 }
 
 static const struct seq_operations raw6_seq_ops = {
-	.start =	raw6_seq_start,
-	.next =		raw6_seq_next,
-	.stop =		raw6_seq_stop,
+	.start =	raw_seq_start,
+	.next =		raw_seq_next,
+	.stop =		raw_seq_stop,
 	.show =		raw6_seq_show,
 };
 
 static int raw6_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &raw6_seq_ops,
-			sizeof(struct raw6_iter_state));
+	return raw_seq_open(file, &raw_v6_hashinfo, PF_INET6);
 }
 
 static const struct file_operations raw6_seq_fops = {
-- 
cgit v1.2.3


From dabeb344f54ab780d152714c18f1cb6b21c471a1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 9 Nov 2007 01:57:29 +0100
Subject: mac80211: provide interface iterator for drivers

Sometimes drivers need to know which interfaces are associated with
their hardware. Rather than forcing those drivers to keep track of
the interfaces that were added, this adds an iteration function to
mac80211.

As it is intended to be used from the interface add/remove callbacks,
the iteration function may currently only be called under RTNL.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h | 16 ++++++++++++++++
 net/mac80211/util.c    | 33 +++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 17b60391fcd6..1470e1b886f0 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1406,4 +1406,20 @@ void ieee80211_wake_queues(struct ieee80211_hw *hw);
  */
 void ieee80211_scan_completed(struct ieee80211_hw *hw);
 
+/**
+ * ieee80211_iterate_active_interfaces - iterate active interfaces
+ *
+ * This function iterates over the interfaces associated with a given
+ * hardware that are currently active and calls the callback for them.
+ * Must be called under RTNL.
+ *
+ * @hw: the hardware struct of which the interfaces should be iterated over
+ * @iterator: the iterator function to call
+ * @data: first argument of the iterator function
+ */
+void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
+					 void (*iterator)(void *data, u8 *mac,
+							  int if_id),
+					 void *data);
+
 #endif /* MAC80211_H */
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 5a0564e1dbd6..88f262baaa5e 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -22,6 +22,7 @@
 #include <linux/bitmap.h>
 #include <net/net_namespace.h>
 #include <net/cfg80211.h>
+#include <net/rtnetlink.h>
 
 #include "ieee80211_i.h"
 #include "ieee80211_rate.h"
@@ -484,3 +485,35 @@ void ieee80211_wake_queues(struct ieee80211_hw *hw)
 		ieee80211_wake_queue(hw, i);
 }
 EXPORT_SYMBOL(ieee80211_wake_queues);
+
+void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
+					 void (*iterator)(void *data, u8 *mac,
+							  int if_id),
+					 void *data)
+{
+	struct ieee80211_local *local = hw_to_local(hw);
+	struct ieee80211_sub_if_data *sdata;
+
+	ASSERT_RTNL();
+
+	/* we hold the RTNL here so can safely walk the list */
+	list_for_each_entry(sdata, &local->interfaces, list) {
+		switch (sdata->type) {
+		case IEEE80211_IF_TYPE_INVALID:
+		case IEEE80211_IF_TYPE_MNTR:
+		case IEEE80211_IF_TYPE_VLAN:
+			continue;
+		case IEEE80211_IF_TYPE_AP:
+		case IEEE80211_IF_TYPE_STA:
+		case IEEE80211_IF_TYPE_IBSS:
+		case IEEE80211_IF_TYPE_WDS:
+			break;
+		}
+		if (sdata->dev == local->mdev)
+			continue;
+		if (netif_running(sdata->dev))
+			iterator(data, sdata->dev->dev_addr,
+				 sdata->dev->ifindex);
+	}
+}
+EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces);
-- 
cgit v1.2.3


From 6b4e324164c683a7b4c1bd0be4d85f9a5a0f0e90 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Wed, 14 Nov 2007 19:57:38 +0200
Subject: mac80211: adding 802.11n definitions in ieee80211.h

This patch adds several structs and definitions to ieee80211.h
to support 802.11n draft specifications.
As 802.11n depends on and extends the 802.11e standard in several issues,
there are also several definitions that belong to 802.11e.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ieee80211.h | 134 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 134 insertions(+)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 214821af3d5b..3e641590d0c8 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -54,6 +54,8 @@
 #define IEEE80211_STYPE_ACTION		0x00D0
 
 /* control */
+#define IEEE80211_STYPE_BACK_REQ	0x0080
+#define IEEE80211_STYPE_BACK		0x0090
 #define IEEE80211_STYPE_PSPOLL		0x00A0
 #define IEEE80211_STYPE_RTS		0x00B0
 #define IEEE80211_STYPE_CTS		0x00C0
@@ -185,6 +187,25 @@ struct ieee80211_mgmt {
 					u8 new_chan;
 					u8 switch_count;
 				} __attribute__((packed)) chan_switch;
+				struct{
+					u8 action_code;
+					u8 dialog_token;
+					__le16 capab;
+					__le16 timeout;
+					__le16 start_seq_num;
+				} __attribute__((packed)) addba_req;
+				struct{
+					u8 action_code;
+					u8 dialog_token;
+					__le16 status;
+					__le16 capab;
+					__le16 timeout;
+				} __attribute__((packed)) addba_resp;
+				struct{
+					u8 action_code;
+					__le16 params;
+					__le16 reason_code;
+				} __attribute__((packed)) delba;
 			} u;
 		} __attribute__ ((packed)) action;
 	} u;
@@ -205,6 +226,66 @@ struct ieee80211_cts {
 	u8 ra[6];
 } __attribute__ ((packed));
 
+/**
+ * struct ieee80211_bar - HT Block Ack Request
+ *
+ * This structure refers to "HT BlockAckReq" as
+ * described in 802.11n draft section 7.2.1.7.1
+ */
+struct ieee80211_bar {
+	__le16 frame_control;
+	__le16 duration;
+	__u8 ra[6];
+	__u8 ta[6];
+	__u16 control;
+	__u16 start_seq_num;
+} __attribute__((packed));
+
+/**
+ * struct ieee80211_ht_cap - HT capabilities
+ *
+ * This structure refers to "HT capabilities element" as
+ * described in 802.11n draft section 7.3.2.52
+ */
+struct ieee80211_ht_cap {
+	__le16 cap_info;
+	u8 ampdu_params_info;
+	u8 supp_mcs_set[16];
+	__le16 extended_ht_cap_info;
+	__le32 tx_BF_cap_info;
+	u8 antenna_selection_info;
+} __attribute__ ((packed));
+
+/**
+ * struct ieee80211_ht_cap - HT additional information
+ *
+ * This structure refers to "HT information element" as
+ * described in 802.11n draft section 7.3.2.53
+ */
+struct ieee80211_ht_addt_info {
+	u8 control_chan;
+	u8 ht_param;
+	__le16 operation_mode;
+	__le16 stbc_param;
+	u8 basic_set[16];
+} __attribute__ ((packed));
+
+/* 802.11n HT capabilities masks */
+#define IEEE80211_HT_CAP_SUP_WIDTH		0x0002
+#define IEEE80211_HT_CAP_MIMO_PS		0x000C
+#define IEEE80211_HT_CAP_GRN_FLD		0x0010
+#define IEEE80211_HT_CAP_SGI_20			0x0020
+#define IEEE80211_HT_CAP_SGI_40			0x0040
+#define IEEE80211_HT_CAP_DELAY_BA		0x0400
+#define IEEE80211_HT_CAP_MAX_AMSDU		0x0800
+#define IEEE80211_HT_CAP_AMPDU_FACTOR		0x03
+#define IEEE80211_HT_CAP_AMPDU_DENSITY		0x1C
+/* 802.11n HT IE masks */
+#define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
+#define IEEE80211_HT_IE_CHA_WIDTH		0x04
+#define IEEE80211_HT_IE_HT_PROTECTION		0x0003
+#define IEEE80211_HT_IE_NON_GF_STA_PRSNT	0x0004
+#define IEEE80211_HT_IE_NON_HT_STA_PRSNT	0x0010
 
 /* Authentication algorithms */
 #define WLAN_AUTH_OPEN 0
@@ -271,6 +352,18 @@ enum ieee80211_statuscode {
 	WLAN_STATUS_UNSUPP_RSN_VERSION = 44,
 	WLAN_STATUS_INVALID_RSN_IE_CAP = 45,
 	WLAN_STATUS_CIPHER_SUITE_REJECTED = 46,
+	/* 802.11e */
+	WLAN_STATUS_UNSPECIFIED_QOS = 32,
+	WLAN_STATUS_ASSOC_DENIED_NOBANDWIDTH = 33,
+	WLAN_STATUS_ASSOC_DENIED_LOWACK = 34,
+	WLAN_STATUS_ASSOC_DENIED_UNSUPP_QOS = 35,
+	WLAN_STATUS_REQUEST_DECLINED = 37,
+	WLAN_STATUS_INVALID_QOS_PARAM = 38,
+	WLAN_STATUS_CHANGE_TSPEC = 39,
+	WLAN_STATUS_WAIT_TS_DELAY = 47,
+	WLAN_STATUS_NO_DIRECT_LINK = 48,
+	WLAN_STATUS_STA_NOT_PRESENT = 49,
+	WLAN_STATUS_STA_NOT_QSTA = 50,
 };
 
 
@@ -301,6 +394,16 @@ enum ieee80211_reasoncode {
 	WLAN_REASON_INVALID_RSN_IE_CAP = 22,
 	WLAN_REASON_IEEE8021X_FAILED = 23,
 	WLAN_REASON_CIPHER_SUITE_REJECTED = 24,
+	/* 802.11e */
+	WLAN_REASON_DISASSOC_UNSPECIFIED_QOS = 32,
+	WLAN_REASON_DISASSOC_QAP_NO_BANDWIDTH = 33,
+	WLAN_REASON_DISASSOC_LOW_ACK = 34,
+	WLAN_REASON_DISASSOC_QAP_EXCEED_TXOP = 35,
+	WLAN_REASON_QSTA_LEAVE_QBSS = 36,
+	WLAN_REASON_QSTA_NOT_USE = 37,
+	WLAN_REASON_QSTA_REQUIRE_SETUP = 38,
+	WLAN_REASON_QSTA_TIMEOUT = 39,
+	WLAN_REASON_QSTA_CIPHER_NOT_SUPP = 45,
 };
 
 
@@ -319,6 +422,15 @@ enum ieee80211_eid {
 	WLAN_EID_HP_PARAMS = 8,
 	WLAN_EID_HP_TABLE = 9,
 	WLAN_EID_REQUEST = 10,
+	/* 802.11e */
+	WLAN_EID_QBSS_LOAD = 11,
+	WLAN_EID_EDCA_PARAM_SET = 12,
+	WLAN_EID_TSPEC = 13,
+	WLAN_EID_TCLAS = 14,
+	WLAN_EID_SCHEDULE = 15,
+	WLAN_EID_TS_DELAY = 43,
+	WLAN_EID_TCLAS_PROCESSING = 44,
+	WLAN_EID_QOS_CAPA = 46,
 	/* 802.11h */
 	WLAN_EID_PWR_CONSTRAINT = 32,
 	WLAN_EID_PWR_CAPABILITY = 33,
@@ -333,6 +445,9 @@ enum ieee80211_eid {
 	/* 802.11g */
 	WLAN_EID_ERP_INFO = 42,
 	WLAN_EID_EXT_SUPP_RATES = 50,
+	/* 802.11n */
+	WLAN_EID_HT_CAPABILITY = 45,
+	WLAN_EID_HT_EXTRA_INFO = 61,
 	/* 802.11i */
 	WLAN_EID_RSN = 48,
 	WLAN_EID_WPA = 221,
@@ -341,6 +456,25 @@ enum ieee80211_eid {
 	WLAN_EID_QOS_PARAMETER = 222
 };
 
+/* Action category code */
+enum ieee80211_category {
+	WLAN_CATEGORY_SPECTRUM_MGMT = 0,
+	WLAN_CATEGORY_QOS = 1,
+	WLAN_CATEGORY_DLS = 2,
+	WLAN_CATEGORY_BACK = 3,
+	WLAN_CATEGORY_WMM = 17,
+};
+
+/* BACK action code */
+enum ieee80211_back_actioncode {
+	WLAN_ACTION_ADDBA_REQ = 0,
+	WLAN_ACTION_ADDBA_RESP = 1,
+	WLAN_ACTION_DELBA = 2,
+};
+
+/* A-MSDU 802.11n */
+#define IEEE80211_QOS_CONTROL_A_MSDU_PRESENT 0x0080
+
 /* cipher suite selectors */
 #define WLAN_CIPHER_SUITE_USE_GROUP	0x000FAC00
 #define WLAN_CIPHER_SUITE_WEP40		0x000FAC01
-- 
cgit v1.2.3


From de4d1db369785c29d68915edfee0cb70e8199f4c Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Wed, 21 Nov 2007 22:02:58 +0800
Subject: [LIB]: Introduce struct pcounter

This just generalises what was introduced by Eric Dumazet for the struct proto
inuse field in 286ab3d46058840d68e5d7d52e316c1f7e98c59f:

    [NET]: Define infrastructure to keep 'inuse' changes in an efficent SMP/NUMA way.

Please look at the comment in there to see the rationale.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pcounter.h | 102 +++++++++++++++++++++++++++++++++++++++++++++++
 lib/Makefile             |   1 +
 lib/pcounter.c           |  26 ++++++++++++
 3 files changed, 129 insertions(+)
 create mode 100644 include/linux/pcounter.h
 create mode 100644 lib/pcounter.c

(limited to 'include')

diff --git a/include/linux/pcounter.h b/include/linux/pcounter.h
new file mode 100644
index 000000000000..620aaded4e94
--- /dev/null
+++ b/include/linux/pcounter.h
@@ -0,0 +1,102 @@
+#ifndef __LINUX_PCOUNTER_H
+#define __LINUX_PCOUNTER_H
+
+struct pcounter {
+#ifdef CONFIG_SMP
+	void		(*add)(struct pcounter *self, int inc);
+	int		(*getval)(const struct pcounter *self);
+	int		*per_cpu_values;
+#else
+	int		val;
+#endif
+};
+
+/*
+ * Special macros to let pcounters use a fast version of {getvalue|add}
+ * using a static percpu variable per pcounter instead of an allocated one,
+ * saving one dereference.
+ * This might be changed if/when dynamic percpu vars become fast.
+ */
+#ifdef CONFIG_SMP
+#include <linux/cpumask.h>
+#include <linux/percpu.h>
+
+#define DEFINE_PCOUNTER(NAME)					\
+static DEFINE_PER_CPU(int, NAME##_pcounter_values);		\
+static void NAME##_pcounter_add(struct pcounter *self, int inc)	\
+{								\
+       __get_cpu_var(NAME##_pcounter_values) += inc;		\
+}								\
+								\
+static int NAME##_pcounter_getval(const struct pcounter *self)	\
+{								\
+       int res = 0, cpu;					\
+								\
+       for_each_possible_cpu(cpu)				\
+               res += per_cpu(NAME##_pcounter_values, cpu);	\
+       return res;						\
+}
+
+#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER)		\
+	MEMBER = {						\
+		.add	= NAME##_pcounter_add,			\
+		.getval = NAME##_pcounter_getval,		\
+	}
+
+extern void pcounter_def_add(struct pcounter *self, int inc);
+extern int pcounter_def_getval(const struct pcounter *self);
+
+static inline int pcounter_alloc(struct pcounter *self)
+{
+	int rc = 0;
+	if (self->add == NULL) {
+		self->per_cpu_values = alloc_percpu(int);
+		if (self->per_cpu_values != NULL) {
+			self->add    = pcounter_def_add;
+			self->getval = pcounter_def_getval;
+		} else
+			rc = 1;
+	}
+	return rc;
+}
+
+static inline void pcounter_free(struct pcounter *self)
+{
+	if (self->per_cpu_values != NULL) {
+		free_percpu(self->per_cpu_values);
+		self->per_cpu_values = NULL;
+		self->getval = NULL;
+		self->add = NULL;
+	}
+}
+
+static inline void pcounter_add(struct pcounter *self, int inc)
+{
+	self->add(self, inc);
+}
+
+static inline int pcounter_getval(const struct pcounter *self)
+{
+	return self->getval(self);
+}
+
+#else /* CONFIG_SMP */
+
+static inline void pcounter_add(struct pcounter *self, int inc)
+{
+	self->value += inc;
+}
+
+static inline int pcounter_getval(const struct pcounter *self)
+{
+	return self->val;
+}
+
+#define DEFINE_PCOUNTER(NAME)
+#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER)
+#define pcounter_alloc(self) 0
+#define pcounter_free(self)
+
+#endif /* CONFIG_SMP */
+
+#endif /* __LINUX_PCOUNTER_H */
diff --git a/lib/Makefile b/lib/Makefile
index 89841dc9d91c..543f2502b60a 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -61,6 +61,7 @@ obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
 obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o
 obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o
 obj-$(CONFIG_SMP) += percpu_counter.o
+obj-$(CONFIG_SMP) += pcounter.o
 obj-$(CONFIG_AUDIT_GENERIC) += audit.o
 
 obj-$(CONFIG_SWIOTLB) += swiotlb.o
diff --git a/lib/pcounter.c b/lib/pcounter.c
new file mode 100644
index 000000000000..93feea598251
--- /dev/null
+++ b/lib/pcounter.c
@@ -0,0 +1,26 @@
+/*
+ * Define default pcounter functions
+ * Note that often used pcounters use dedicated functions to get a speed increase.
+ * (see DEFINE_PCOUNTER/REF_PCOUNTER_MEMBER)
+ */
+
+#include <linux/module.h>
+#include <linux/pcounter.h>
+#include <linux/smp.h>
+
+void pcounter_def_add(struct pcounter *self, int inc)
+{
+	per_cpu_ptr(self->per_cpu_values, smp_processor_id())[0] += inc;
+}
+
+EXPORT_SYMBOL_GPL(pcounter_def_add);
+
+int pcounter_def_getval(const struct pcounter *self)
+{
+	int res = 0, cpu;
+	for_each_possible_cpu(cpu)
+		res += per_cpu_ptr(self->per_cpu_values, cpu)[0];
+	return res;
+}
+
+EXPORT_SYMBOL_GPL(pcounter_def_getval);
-- 
cgit v1.2.3


From ebb53d75657f86587ac8cf3e38ab0c860a8e3d4f Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Wed, 21 Nov 2007 22:08:50 +0800
Subject: [NET] proto: Use pcounters for the inuse field

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 59 +++++++---------------------------------------------
 net/core/sock.c    | 61 ++++++------------------------------------------------
 2 files changed, 14 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 9c55af8e5f81..e329d05f7995 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -47,6 +47,7 @@
 #include <linux/module.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
+#include <linux/pcounter.h>
 #include <linux/skbuff.h>	/* struct sk_buff */
 #include <linux/mm.h>
 #include <linux/security.h>
@@ -565,14 +566,9 @@ struct proto {
 	void			(*unhash)(struct sock *sk);
 	int			(*get_port)(struct sock *sk, unsigned short snum);
 
-#ifdef CONFIG_SMP
 	/* Keeping track of sockets in use */
-	void			(*inuse_add)(struct proto *prot, int inc);
-	int			(*inuse_getval)(const struct proto *prot);
-	int			*inuse_ptr;
-#else
-	int			inuse;
-#endif
+	struct pcounter		inuse;
+
 	/* Memory pressure */
 	void			(*enter_memory_pressure)(void);
 	atomic_t		*memory_allocated;	/* Current allocated memory. */
@@ -607,35 +603,8 @@ struct proto {
 #endif
 };
 
-/*
- * Special macros to let protos use a fast version of inuse{get|add}
- * using a static percpu variable per proto instead of an allocated one,
- * saving one dereference.
- * This might be changed if/when dynamic percpu vars become fast.
- */
-#ifdef CONFIG_SMP
-# define DEFINE_PROTO_INUSE(NAME)			\
-static DEFINE_PER_CPU(int, NAME##_inuse);		\
-static void NAME##_inuse_add(struct proto *prot, int inc)	\
-{							\
-	__get_cpu_var(NAME##_inuse) += inc;		\
-}							\
-							\
-static int NAME##_inuse_getval(const struct proto *prot)\
-{							\
-	int res = 0, cpu;				\
-							\
-	for_each_possible_cpu(cpu)			\
-		res += per_cpu(NAME##_inuse, cpu);	\
-	return res;					\
-}
-# define REF_PROTO_INUSE(NAME)				\
-	.inuse_add = NAME##_inuse_add,			\
-	.inuse_getval = NAME##_inuse_getval,
-#else
-# define DEFINE_PROTO_INUSE(NAME)
-# define REF_PROTO_INUSE(NAME)
-#endif
+#define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME)
+#define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
 
 extern int proto_register(struct proto *prot, int alloc_slab);
 extern void proto_unregister(struct proto *prot);
@@ -668,29 +637,17 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 /* Called with local bh disabled */
 static __inline__ void sock_prot_inc_use(struct proto *prot)
 {
-#ifdef CONFIG_SMP
-	prot->inuse_add(prot, 1);
-#else
-	prot->inuse++;
-#endif
+	pcounter_add(&prot->inuse, 1);
 }
 
 static __inline__ void sock_prot_dec_use(struct proto *prot)
 {
-#ifdef CONFIG_SMP
-	prot->inuse_add(prot, -1);
-#else
-	prot->inuse--;
-#endif
+	pcounter_add(&prot->inuse, -1);
 }
 
 static __inline__ int sock_prot_inuse(struct proto *proto)
 {
-#ifdef CONFIG_SMP
-	return proto->inuse_getval(proto);
-#else
-	return proto->inuse;
-#endif
+	return pcounter_getval(&proto->inuse);
 }
 
 /* With per-bucket locks this operation is not-atomic, so that
diff --git a/net/core/sock.c b/net/core/sock.c
index c9305a861760..eac7aa0721da 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1804,65 +1804,15 @@ EXPORT_SYMBOL(sk_common_release);
 static DEFINE_RWLOCK(proto_list_lock);
 static LIST_HEAD(proto_list);
 
-#ifdef CONFIG_SMP
-/*
- * Define default functions to keep track of inuse sockets per protocol
- * Note that often used protocols use dedicated functions to get a speed increase.
- * (see DEFINE_PROTO_INUSE/REF_PROTO_INUSE)
- */
-static void inuse_add(struct proto *prot, int inc)
-{
-	per_cpu_ptr(prot->inuse_ptr, smp_processor_id())[0] += inc;
-}
-
-static int inuse_get(const struct proto *prot)
-{
-	int res = 0, cpu;
-	for_each_possible_cpu(cpu)
-		res += per_cpu_ptr(prot->inuse_ptr, cpu)[0];
-	return res;
-}
-
-static int inuse_init(struct proto *prot)
-{
-	if (!prot->inuse_getval || !prot->inuse_add) {
-		prot->inuse_ptr = alloc_percpu(int);
-		if (prot->inuse_ptr == NULL)
-			return -ENOBUFS;
-
-		prot->inuse_getval = inuse_get;
-		prot->inuse_add = inuse_add;
-	}
-	return 0;
-}
-
-static void inuse_fini(struct proto *prot)
-{
-	if (prot->inuse_ptr != NULL) {
-		free_percpu(prot->inuse_ptr);
-		prot->inuse_ptr = NULL;
-		prot->inuse_getval = NULL;
-		prot->inuse_add = NULL;
-	}
-}
-#else
-static inline int inuse_init(struct proto *prot)
-{
-	return 0;
-}
-
-static inline void inuse_fini(struct proto *prot)
-{
-}
-#endif
-
 int proto_register(struct proto *prot, int alloc_slab)
 {
 	char *request_sock_slab_name = NULL;
 	char *timewait_sock_slab_name;
 
-	if (inuse_init(prot))
+	if (pcounter_alloc(&prot->inuse) != 0) {
+		printk(KERN_CRIT "%s: Can't alloc inuse counters!\n", prot->name);
 		goto out;
+	}
 
 	if (alloc_slab) {
 		prot->slab = kmem_cache_create(prot->name, prot->obj_size, 0,
@@ -1930,7 +1880,7 @@ out_free_sock_slab:
 	kmem_cache_destroy(prot->slab);
 	prot->slab = NULL;
 out_free_inuse:
-	inuse_fini(prot);
+	pcounter_free(&prot->inuse);
 out:
 	return -ENOBUFS;
 }
@@ -1943,7 +1893,8 @@ void proto_unregister(struct proto *prot)
 	list_del(&prot->node);
 	write_unlock(&proto_list_lock);
 
-	inuse_fini(prot);
+	pcounter_free(&prot->inuse);
+
 	if (prot->slab != NULL) {
 		kmem_cache_destroy(prot->slab);
 		prot->slab = NULL;
-- 
cgit v1.2.3


From 9b91ad2747891767c0efb4fb965c5dfed8d4f88e Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Tue, 20 Nov 2007 21:56:37 -0200
Subject: [DCCP]: Make PARTOPEN an autonomous state

This decouples PARTOPEN from TCP-specific stream-states.

It thus addresses the FIXME.

The code has been checked with regard to dependency on PARTOPEN and FIN_WAIT1
states (to which PARTOPEN previously was mapped): there is no difference, as
PARTOPEN is always referred to directly (i.e. not via the mapping to TCP
state).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 333c3ea82a5d..a0073268808f 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -229,21 +229,13 @@ struct dccp_so_feat {
 enum dccp_state {
 	DCCP_OPEN	= TCP_ESTABLISHED,
 	DCCP_REQUESTING	= TCP_SYN_SENT,
-	DCCP_PARTOPEN	= TCP_FIN_WAIT1, /* FIXME:
-					    This mapping is horrible, but TCP has
-					    no matching state for DCCP_PARTOPEN,
-					    as TCP_SYN_RECV is already used by
-					    DCCP_RESPOND, why don't stop using TCP
-					    mapping of states? OK, now we don't use
-					    sk_stream_sendmsg anymore, so doesn't
-					    seem to exist any reason for us to
-					    do the TCP mapping here */
 	DCCP_LISTEN	= TCP_LISTEN,
 	DCCP_RESPOND	= TCP_SYN_RECV,
 	DCCP_CLOSING	= TCP_CLOSING,
 	DCCP_TIME_WAIT	= TCP_TIME_WAIT,
 	DCCP_CLOSED	= TCP_CLOSE,
-	DCCP_MAX_STATES = TCP_MAX_STATES,
+	DCCP_PARTOPEN	= TCP_MAX_STATES,
+	DCCP_MAX_STATES
 };
 
 #define DCCP_STATE_MASK 0xf
@@ -252,12 +244,12 @@ enum dccp_state {
 enum {
 	DCCPF_OPEN	 = TCPF_ESTABLISHED,
 	DCCPF_REQUESTING = TCPF_SYN_SENT,
-	DCCPF_PARTOPEN	 = TCPF_FIN_WAIT1,
 	DCCPF_LISTEN	 = TCPF_LISTEN,
 	DCCPF_RESPOND	 = TCPF_SYN_RECV,
 	DCCPF_CLOSING	 = TCPF_CLOSING,
 	DCCPF_TIME_WAIT	 = TCPF_TIME_WAIT,
 	DCCPF_CLOSED	 = TCPF_CLOSE,
+	DCCPF_PARTOPEN	 = 1 << DCCP_PARTOPEN,
 };
 
 static inline struct dccp_hdr *dccp_hdr(const struct sk_buff *skb)
-- 
cgit v1.2.3


From e7d0362dd41e760f340c1b500646cc92522bd9d5 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Mon, 26 Nov 2007 23:34:54 +0800
Subject: [PCOUNTER] Fix build error without CONFIG_SMP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I keep getting this build error and couldn't find anyone fixing
it in archives. ...Maybe all net developers except me build
just SMP kernels :-).

In file included from include/net/sock.h:50,
                 from ipc/mqueue.c:35:
include/linux/pcounter.h: In function 'pcounter_add':
include/linux/pcounter.h:87: error: 'struct pcounter' has no
member named 'value'
make[1]: *** [ipc/mqueue.o] Error 1
make: *** [ipc] Error 2

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pcounter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/pcounter.h b/include/linux/pcounter.h
index 620aaded4e94..9c4760a328f3 100644
--- a/include/linux/pcounter.h
+++ b/include/linux/pcounter.h
@@ -84,7 +84,7 @@ static inline int pcounter_getval(const struct pcounter *self)
 
 static inline void pcounter_add(struct pcounter *self, int inc)
 {
-	self->value += inc;
+	self->val += inc;
 }
 
 static inline int pcounter_getval(const struct pcounter *self)
-- 
cgit v1.2.3


From 8d8ad9d7c4bfe79bc91b7fc419ecfb9dcdfe6a51 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 26 Nov 2007 20:10:50 +0800
Subject: [NET]: Name magic constants in sock_wake_async()

The sock_wake_async() performs a bit different actions
depending on "how" argument. Unfortunately this argument
ony has numerical magic values.

I propose to give names to their constants to help people
reading this function callers understand what's going on
without looking into this function all the time.

I suppose this is 2.6.25 material, but if it's not (or the
naming seems poor/bad/awful), I can rework it against the
current net-2.6 tree.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h  |  7 +++++++
 net/atm/common.c     |  2 +-
 net/core/sock.c      |  8 ++++----
 net/core/stream.c    |  2 +-
 net/dccp/input.c     |  8 ++++----
 net/dccp/output.c    |  2 +-
 net/ipv4/tcp_input.c | 12 ++++++------
 net/rxrpc/af_rxrpc.c |  2 +-
 net/sctp/socket.c    |  3 ++-
 net/socket.c         |  9 ++++-----
 net/unix/af_unix.c   |  8 ++++----
 11 files changed, 35 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index 0235d917d5c3..f95f12c5840c 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -186,6 +186,13 @@ struct net_proto_family {
 struct iovec;
 struct kvec;
 
+enum {
+	SOCK_WAKE_IO,
+	SOCK_WAKE_WAITD,
+	SOCK_WAKE_SPACE,
+	SOCK_WAKE_URG,
+};
+
 extern int	     sock_wake_async(struct socket *sk, int how, int band);
 extern int	     sock_register(const struct net_proto_family *fam);
 extern void	     sock_unregister(int family);
diff --git a/net/atm/common.c b/net/atm/common.c
index eba09a04f6bf..c865517ba449 100644
--- a/net/atm/common.c
+++ b/net/atm/common.c
@@ -113,7 +113,7 @@ static void vcc_write_space(struct sock *sk)
 		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 			wake_up_interruptible(sk->sk_sleep);
 
-		sk_wake_async(sk, 2, POLL_OUT);
+		sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
 	}
 
 	read_unlock(&sk->sk_callback_lock);
diff --git a/net/core/sock.c b/net/core/sock.c
index eac7aa0721da..118214047ed2 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1498,7 +1498,7 @@ static void sock_def_error_report(struct sock *sk)
 	read_lock(&sk->sk_callback_lock);
 	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 		wake_up_interruptible(sk->sk_sleep);
-	sk_wake_async(sk,0,POLL_ERR);
+	sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR);
 	read_unlock(&sk->sk_callback_lock);
 }
 
@@ -1507,7 +1507,7 @@ static void sock_def_readable(struct sock *sk, int len)
 	read_lock(&sk->sk_callback_lock);
 	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 		wake_up_interruptible(sk->sk_sleep);
-	sk_wake_async(sk,1,POLL_IN);
+	sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
 	read_unlock(&sk->sk_callback_lock);
 }
 
@@ -1524,7 +1524,7 @@ static void sock_def_write_space(struct sock *sk)
 
 		/* Should agree with poll, otherwise some programs break */
 		if (sock_writeable(sk))
-			sk_wake_async(sk, 2, POLL_OUT);
+			sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
 	}
 
 	read_unlock(&sk->sk_callback_lock);
@@ -1539,7 +1539,7 @@ void sk_send_sigurg(struct sock *sk)
 {
 	if (sk->sk_socket && sk->sk_socket->file)
 		if (send_sigurg(&sk->sk_socket->file->f_owner))
-			sk_wake_async(sk, 3, POLL_PRI);
+			sk_wake_async(sk, SOCK_WAKE_URG, POLL_PRI);
 }
 
 void sk_reset_timer(struct sock *sk, struct timer_list* timer,
diff --git a/net/core/stream.c b/net/core/stream.c
index b2fb846f42a4..5586879bb9b3 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -35,7 +35,7 @@ void sk_stream_write_space(struct sock *sk)
 		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 			wake_up_interruptible(sk->sk_sleep);
 		if (sock->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN))
-			sock_wake_async(sock, 2, POLL_OUT);
+			sock_wake_async(sock, SOCK_WAKE_SPACE, POLL_OUT);
 	}
 }
 
diff --git a/net/dccp/input.c b/net/dccp/input.c
index df0fb2c149a6..ef299fbd7c26 100644
--- a/net/dccp/input.c
+++ b/net/dccp/input.c
@@ -37,7 +37,7 @@ static void dccp_rcv_close(struct sock *sk, struct sk_buff *skb)
 	dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
 	dccp_fin(sk, skb);
 	dccp_set_state(sk, DCCP_CLOSED);
-	sk_wake_async(sk, 1, POLL_HUP);
+	sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
 }
 
 static void dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb)
@@ -90,7 +90,7 @@ static void dccp_rcv_reset(struct sock *sk, struct sk_buff *skb)
 	dccp_fin(sk, skb);
 
 	if (err && !sock_flag(sk, SOCK_DEAD))
-		sk_wake_async(sk, 0, POLL_ERR);
+		sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR);
 	dccp_time_wait(sk, DCCP_TIME_WAIT, 0);
 }
 
@@ -416,7 +416,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk,
 
 		if (!sock_flag(sk, SOCK_DEAD)) {
 			sk->sk_state_change(sk);
-			sk_wake_async(sk, 0, POLL_OUT);
+			sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT);
 		}
 
 		if (sk->sk_write_pending || icsk->icsk_ack.pingpong ||
@@ -624,7 +624,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		switch (old_state) {
 		case DCCP_PARTOPEN:
 			sk->sk_state_change(sk);
-			sk_wake_async(sk, 0, POLL_OUT);
+			sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT);
 			break;
 		}
 	} else if (unlikely(dh->dccph_type == DCCP_PKT_SYNC)) {
diff --git a/net/dccp/output.c b/net/dccp/output.c
index f49544618f20..33ce737ef3a8 100644
--- a/net/dccp/output.c
+++ b/net/dccp/output.c
@@ -170,7 +170,7 @@ void dccp_write_space(struct sock *sk)
 		wake_up_interruptible(sk->sk_sleep);
 	/* Should agree with poll, otherwise some programs break */
 	if (sock_writeable(sk))
-		sk_wake_async(sk, 2, POLL_OUT);
+		sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
 
 	read_unlock(&sk->sk_callback_lock);
 }
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 79996b16b942..97ea3eda206d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3595,9 +3595,9 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th)
 		/* Do not send POLL_HUP for half duplex close. */
 		if (sk->sk_shutdown == SHUTDOWN_MASK ||
 		    sk->sk_state == TCP_CLOSE)
-			sk_wake_async(sk, 1, POLL_HUP);
+			sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
 		else
-			sk_wake_async(sk, 1, POLL_IN);
+			sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
 	}
 }
 
@@ -4956,7 +4956,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 
 		if (!sock_flag(sk, SOCK_DEAD)) {
 			sk->sk_state_change(sk);
-			sk_wake_async(sk, 0, POLL_OUT);
+			sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT);
 		}
 
 		if (sk->sk_write_pending ||
@@ -5186,9 +5186,9 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 				 * are not waked up, because sk->sk_sleep ==
 				 * NULL and sk->sk_socket == NULL.
 				 */
-				if (sk->sk_socket) {
-					sk_wake_async(sk,0,POLL_OUT);
-				}
+				if (sk->sk_socket)
+					sk_wake_async(sk,
+							SOCK_WAKE_IO, POLL_OUT);
 
 				tp->snd_una = TCP_SKB_CB(skb)->ack_seq;
 				tp->snd_wnd = ntohs(th->window) <<
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index d6389450c4bf..5e82f1c0afbb 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -65,7 +65,7 @@ static void rxrpc_write_space(struct sock *sk)
 	if (rxrpc_writable(sk)) {
 		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 			wake_up_interruptible(sk->sk_sleep);
-		sk_wake_async(sk, 2, POLL_OUT);
+		sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
 	}
 	read_unlock(&sk->sk_callback_lock);
 }
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index ea9649ca0b2a..dc2f9221f092 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -6008,7 +6008,8 @@ static void __sctp_write_space(struct sctp_association *asoc)
 			 */
 			if (sock->fasync_list &&
 			    !(sk->sk_shutdown & SEND_SHUTDOWN))
-				sock_wake_async(sock, 2, POLL_OUT);
+				sock_wake_async(sock,
+						SOCK_WAKE_SPACE, POLL_OUT);
 		}
 	}
 }
diff --git a/net/socket.c b/net/socket.c
index aeeab388cc38..9ebca5c695d6 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1070,20 +1070,19 @@ int sock_wake_async(struct socket *sock, int how, int band)
 	if (!sock || !sock->fasync_list)
 		return -1;
 	switch (how) {
-	case 1:
-
+	case SOCK_WAKE_WAITD:
 		if (test_bit(SOCK_ASYNC_WAITDATA, &sock->flags))
 			break;
 		goto call_kill;
-	case 2:
+	case SOCK_WAKE_SPACE:
 		if (!test_and_clear_bit(SOCK_ASYNC_NOSPACE, &sock->flags))
 			break;
 		/* fall through */
-	case 0:
+	case SOCK_WAKE_IO:
 call_kill:
 		__kill_fasync(sock->fasync_list, SIGIO, band);
 		break;
-	case 3:
+	case SOCK_WAKE_URG:
 		__kill_fasync(sock->fasync_list, SIGURG, band);
 	}
 	return 0;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 0f1ecbf86d0f..393197afb190 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -317,7 +317,7 @@ static void unix_write_space(struct sock *sk)
 	if (unix_writable(sk)) {
 		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
 			wake_up_interruptible_sync(sk->sk_sleep);
-		sk_wake_async(sk, 2, POLL_OUT);
+		sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
 	}
 	read_unlock(&sk->sk_callback_lock);
 }
@@ -403,7 +403,7 @@ static int unix_release_sock (struct sock *sk, int embrion)
 			unix_state_unlock(skpair);
 			skpair->sk_state_change(skpair);
 			read_lock(&skpair->sk_callback_lock);
-			sk_wake_async(skpair,1,POLL_HUP);
+			sk_wake_async(skpair, SOCK_WAKE_WAITD, POLL_HUP);
 			read_unlock(&skpair->sk_callback_lock);
 		}
 		sock_put(skpair); /* It may now die */
@@ -1900,9 +1900,9 @@ static int unix_shutdown(struct socket *sock, int mode)
 			other->sk_state_change(other);
 			read_lock(&other->sk_callback_lock);
 			if (peer_mode == SHUTDOWN_MASK)
-				sk_wake_async(other,1,POLL_HUP);
+				sk_wake_async(other, SOCK_WAKE_WAITD, POLL_HUP);
 			else if (peer_mode & RCV_SHUTDOWN)
-				sk_wake_async(other,1,POLL_IN);
+				sk_wake_async(other, SOCK_WAKE_WAITD, POLL_IN);
 			read_unlock(&other->sk_callback_lock);
 		}
 		if (other)
-- 
cgit v1.2.3


From 8512430e554a84275669f78f86dce18566d5cf7a Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Mon, 26 Nov 2007 20:17:38 +0800
Subject: [TCP]: Move FRTO checks out from write queue abstraction funcs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Better place exists in update_send_head (other non-queue related
adjustments are done there as well) which is the only caller of
tcp_advance_send_head (now that the bogus call from mtu_probe is
gone).

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     | 5 -----
 net/ipv4/tcp_output.c | 4 ++++
 2 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index f0c5e7a29400..7e583261f3a0 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1236,14 +1236,9 @@ static inline struct sk_buff *tcp_send_head(struct sock *sk)
 
 static inline void tcp_advance_send_head(struct sock *sk, struct sk_buff *skb)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
-
 	sk->sk_send_head = skb->next;
 	if (sk->sk_send_head == (struct sk_buff *)&sk->sk_write_queue)
 		sk->sk_send_head = NULL;
-	/* Don't override Nagle indefinately with F-RTO */
-	if (tp->frto_counter == 2)
-		tp->frto_counter = 3;
 }
 
 static inline void tcp_check_send_head(struct sock *sk, struct sk_buff *skb_unlinked)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 030fc69ea217..546f385a405f 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -80,6 +80,10 @@ static void update_send_head(struct sock *sk, struct sk_buff *skb)
 	tcp_advance_send_head(sk, skb);
 	tp->snd_nxt = TCP_SKB_CB(skb)->end_seq;
 	tcp_packets_out_inc(sk, skb);
+
+	/* Don't override Nagle indefinately with F-RTO */
+	if (tp->frto_counter == 2)
+		tp->frto_counter = 3;
 }
 
 /* SND.NXT, if window was not shrunk.
-- 
cgit v1.2.3


From f561d0f27d6283c49359bb96048f8ac3728c812c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 29 Nov 2007 20:28:50 +1100
Subject: [NET]: Uninline the sk_stream_alloc_pskb

This function seems too big for inlining. Indeed, it saves
half-a-kilo when uninlined:

add/remove: 1/0 grow/shrink: 0/7 up/down: 195/-719 (-524)
function                                     old     new   delta
sk_stream_alloc_pskb                           -     195    +195
ip_rt_init                                   529     525      -4
__inet_lookup_listener                       284     274     -10
tcp_sendmsg                                 2583    2486     -97
tcp_sendpage                                1449    1305    -144
tso_fragment                                 417     267    -150
tcp_fragment                                1149     992    -157
__tcp_push_pending_frames                   1998    1841    -157

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 29 ++---------------------------
 net/ipv4/tcp.c     | 27 +++++++++++++++++++++++++++
 2 files changed, 29 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index e329d05f7995..7d500a825e4e 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1192,33 +1192,8 @@ static inline void sk_stream_moderate_sndbuf(struct sock *sk)
 	}
 }
 
-static inline struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
-						   int size, int mem,
-						   gfp_t gfp)
-{
-	struct sk_buff *skb;
-
-	/* The TCP header must be at least 32-bit aligned.  */
-	size = ALIGN(size, 4);
-
-	skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
-	if (skb) {
-		skb->truesize += mem;
-		if (sk_stream_wmem_schedule(sk, skb->truesize)) {
-			/*
-			 * Make sure that we have exactly size bytes
-			 * available to the caller, no more, no less.
-			 */
-			skb_reserve(skb, skb_tailroom(skb) - size);
-			return skb;
-		}
-		__kfree_skb(skb);
-	} else {
-		sk->sk_prot->enter_memory_pressure();
-		sk_stream_moderate_sndbuf(sk);
-	}
-	return NULL;
-}
+struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
+		int size, int mem, gfp_t gfp);
 
 static inline struct sk_buff *sk_stream_alloc_skb(struct sock *sk,
 						  int size,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index e055f25876df..6b35ab841db2 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -629,6 +629,33 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos,
 	return ret;
 }
 
+struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
+		int size, int mem, gfp_t gfp)
+{
+	struct sk_buff *skb;
+
+	/* The TCP header must be at least 32-bit aligned.  */
+	size = ALIGN(size, 4);
+
+	skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
+	if (skb) {
+		skb->truesize += mem;
+		if (sk_stream_wmem_schedule(sk, skb->truesize)) {
+			/*
+			 * Make sure that we have exactly size bytes
+			 * available to the caller, no more, no less.
+			 */
+			skb_reserve(skb, skb_tailroom(skb) - size);
+			return skb;
+		}
+		__kfree_skb(skb);
+	} else {
+		sk->sk_prot->enter_memory_pressure();
+		sk_stream_moderate_sndbuf(sk);
+	}
+	return NULL;
+}
+
 static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffset,
 			 size_t psize, int flags)
 {
-- 
cgit v1.2.3


From df97c708d5e6eebdd9ded1fa588eae09acf53793 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 29 Nov 2007 21:22:33 +1100
Subject: [NET]: Eliminate unused argument from sk_stream_alloc_pskb

The 3rd argument is always zero (according to grep :) Eliminate
it and merge the function with sk_stream_alloc_skb.

This saves 44 more bytes, and together with the previous patch
we have:

add/remove: 1/0 grow/shrink: 0/8 up/down: 183/-751 (-568)
function                                     old     new   delta
sk_stream_alloc_skb                            -     183    +183
ip_rt_init                                   529     525      -4
arp_ignore                                   112     107      -5
__inet_lookup_listener                       284     274     -10
tcp_sendmsg                                 2583    2481    -102
tcp_sendpage                                1449    1300    -149
tso_fragment                                 417     258    -159
tcp_fragment                                1149     988    -161
__tcp_push_pending_frames                   1998    1837    -161

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h    | 10 +---------
 net/ipv4/tcp.c        | 11 ++++-------
 net/ipv4/tcp_output.c |  2 +-
 3 files changed, 6 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 7d500a825e4e..bcbf8891dde4 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1192,15 +1192,7 @@ static inline void sk_stream_moderate_sndbuf(struct sock *sk)
 	}
 }
 
-struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
-		int size, int mem, gfp_t gfp);
-
-static inline struct sk_buff *sk_stream_alloc_skb(struct sock *sk,
-						  int size,
-						  gfp_t gfp)
-{
-	return sk_stream_alloc_pskb(sk, size, 0, gfp);
-}
+struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp);
 
 static inline struct page *sk_stream_alloc_page(struct sock *sk)
 {
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 6b35ab841db2..eb77088f3054 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -629,8 +629,7 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos,
 	return ret;
 }
 
-struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
-		int size, int mem, gfp_t gfp)
+struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp)
 {
 	struct sk_buff *skb;
 
@@ -639,7 +638,6 @@ struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
 
 	skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
 	if (skb) {
-		skb->truesize += mem;
 		if (sk_stream_wmem_schedule(sk, skb->truesize)) {
 			/*
 			 * Make sure that we have exactly size bytes
@@ -692,8 +690,7 @@ new_segment:
 			if (!sk_stream_memory_free(sk))
 				goto wait_for_sndbuf;
 
-			skb = sk_stream_alloc_pskb(sk, 0, 0,
-						   sk->sk_allocation);
+			skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation);
 			if (!skb)
 				goto wait_for_memory;
 
@@ -873,8 +870,8 @@ new_segment:
 				if (!sk_stream_memory_free(sk))
 					goto wait_for_sndbuf;
 
-				skb = sk_stream_alloc_pskb(sk, select_size(sk),
-							   0, sk->sk_allocation);
+				skb = sk_stream_alloc_skb(sk, select_size(sk),
+						sk->sk_allocation);
 				if (!skb)
 					goto wait_for_memory;
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 546f385a405f..98835afb8d8a 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1180,7 +1180,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
 	if (skb->len != skb->data_len)
 		return tcp_fragment(sk, skb, len, mss_now);
 
-	buff = sk_stream_alloc_pskb(sk, 0, 0, GFP_ATOMIC);
+	buff = sk_stream_alloc_skb(sk, 0, GFP_ATOMIC);
 	if (unlikely(buff == NULL))
 		return -ENOMEM;
 
-- 
cgit v1.2.3


From c7dc89c0ac8e7c3796bff91becf58ccdbcaf9f18 Mon Sep 17 00:00:00 2001
From: "Fred L. Templin" <fred.l.templin@boeing.com>
Date: Thu, 29 Nov 2007 22:11:40 +1100
Subject: [IPV6]: Add RFC4214 support

This patch includes support for the Intra-Site Automatic Tunnel
Addressing Protocol (ISATAP) per RFC4214. It uses the SIT
module, and is configured using extensions to the "iproute2"
utility. The diffs are specific to the Linux 2.6.24-rc2 kernel
distribution.

This version includes the diff for ./include/linux/if.h which was
missing in the v2.4 submission and is needed to make the
patch compile. The patch has been installed, compiled and
tested in a clean 2.6.24-rc2 kernel build area.

Signed-off-by: Fred L. Templin <fred.l.templin@boeing.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h        |  1 +
 include/linux/if_tunnel.h |  3 ++
 include/linux/in.h        |  8 +++++
 include/net/addrconf.h    | 19 ++++++++++++
 net/ipv6/addrconf.c       | 22 +++++++++++++-
 net/ipv6/route.c          |  2 ++
 net/ipv6/sit.c            | 77 +++++++++++++++++++++++++++++++++++++++++++++++
 7 files changed, 131 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/if.h b/include/linux/if.h
index 186070d5c544..5c9d1fa93fef 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -63,6 +63,7 @@
 #define IFF_MASTER_ALB	0x10		/* bonding master, balance-alb.	*/
 #define IFF_BONDING	0x20		/* bonding master or slave	*/
 #define IFF_SLAVE_NEEDARP 0x40		/* need ARPs for validation	*/
+#define IFF_ISATAP	0x80		/* ISATAP interface (RFC4214)	*/
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h
index 660b5010c2d9..228eb4eb3129 100644
--- a/include/linux/if_tunnel.h
+++ b/include/linux/if_tunnel.h
@@ -17,6 +17,9 @@
 #define GRE_FLAGS	__constant_htons(0x00F8)
 #define GRE_VERSION	__constant_htons(0x0007)
 
+/* i_flags values for SIT mode */
+#define	SIT_ISATAP	0x0001
+
 struct ip_tunnel_parm
 {
 	char			name[IFNAMSIZ];
diff --git a/include/linux/in.h b/include/linux/in.h
index 3975cbf52f20..a8f00cac8f79 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -253,6 +253,14 @@ struct sockaddr_in {
 #define ZERONET(x)	(((x) & htonl(0xff000000)) == htonl(0x00000000))
 #define LOCAL_MCAST(x)	(((x) & htonl(0xFFFFFF00)) == htonl(0xE0000000))
 
+/* Special-Use IPv4 Addresses (RFC3330) */
+#define PRIVATE_10(x)	(((x) & htonl(0xff000000)) == htonl(0x0A000000))
+#define LINKLOCAL_169(x) (((x) & htonl(0xffff0000)) == htonl(0xA9FE0000))
+#define PRIVATE_172(x)	(((x) & htonl(0xfff00000)) == htonl(0xAC100000))
+#define TEST_192(x)	(((x) & htonl(0xffffff00)) == htonl(0xC0000200))
+#define ANYCAST_6TO4(x)	(((x) & htonl(0xffffff00)) == htonl(0xC0586300))
+#define PRIVATE_192(x)	(((x) & htonl(0xffff0000)) == htonl(0xC0A80000))
+#define TEST_198(x)	(((x) & htonl(0xfffe0000)) == htonl(0xC6120000))
 #endif
 
 #endif	/* _LINUX_IN_H */
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index bccc2feb99da..c56827da0dee 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -17,6 +17,7 @@
 
 #define IPV6_MAX_ADDRESSES		16
 
+#include <linux/in.h>
 #include <linux/in6.h>
 
 struct prefix_info {
@@ -249,6 +250,24 @@ static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr)
 		addr->s6_addr32[3] == htonl(0x00000002));
 }
 
+static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr)
+{
+	eui[0] = (ZERONET(addr) || PRIVATE_10(addr) || LOOPBACK(addr) ||
+		  LINKLOCAL_169(addr) || PRIVATE_172(addr) || TEST_192(addr) ||
+		  ANYCAST_6TO4(addr) || PRIVATE_192(addr) || TEST_198(addr) ||
+		  MULTICAST(addr) || BADCLASS(addr)) ? 0x00 : 0x02;
+	eui[1] = 0;
+	eui[2] = 0x5E;
+	eui[3] = 0xFE;
+	memcpy (eui+4, &addr, 4);
+	return 0;
+}
+
+static inline int ipv6_addr_is_isatap(const struct in6_addr *addr)
+{
+	return ((addr->s6_addr32[2] | htonl(0x02000000)) == htonl(0x02005EFE));
+}
+
 #ifdef CONFIG_PROC_FS
 extern int if6_proc_init(void);
 extern void if6_proc_exit(void);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 6c8b193474ba..f177424c186f 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -377,6 +377,13 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 		       "%s: Disabled Privacy Extensions\n",
 		       dev->name);
 		ndev->cnf.use_tempaddr = -1;
+
+		if (dev->type == ARPHRD_SIT && (dev->priv_flags & IFF_ISATAP)) {
+			printk(KERN_INFO
+			       "%s: Disabled Multicast RS\n",
+			       dev->name);
+			ndev->cnf.rtr_solicits = 0;
+		}
 	} else {
 		in6_dev_hold(ndev);
 		ipv6_regen_rndid((unsigned long) ndev);
@@ -1409,6 +1416,9 @@ static int ipv6_generate_eui64(u8 *eui, struct net_device *dev)
 		return addrconf_ifid_arcnet(eui, dev);
 	case ARPHRD_INFINIBAND:
 		return addrconf_ifid_infiniband(eui, dev);
+	case ARPHRD_SIT:
+		if (dev->priv_flags & IFF_ISATAP)
+			return ipv6_isatap_eui64(eui, *(__be32 *)dev->dev_addr);
 	}
 	return -1;
 }
@@ -1444,7 +1454,7 @@ regen:
 	 *
 	 *  - Reserved subnet anycast (RFC 2526)
 	 *	11111101 11....11 1xxxxxxx
-	 *  - ISATAP (draft-ietf-ngtrans-isatap-13.txt) 5.1
+	 *  - ISATAP (RFC4214) 6.1
 	 *	00-00-5E-FE-xx-xx-xx-xx
 	 *  - value 0
 	 *  - XXX: already assigned to an address on the device
@@ -2175,6 +2185,16 @@ static void addrconf_sit_config(struct net_device *dev)
 		return;
 	}
 
+	if (dev->priv_flags & IFF_ISATAP) {
+		struct in6_addr addr;
+
+		ipv6_addr_set(&addr,  htonl(0xFE800000), 0, 0, 0);
+		addrconf_prefix_route(&addr, 64, dev, 0, 0);
+		if (!ipv6_generate_eui64(addr.s6_addr + 8, dev))
+			addrconf_add_linklocal(idev, &addr);
+		return;
+	}
+
 	sit_add_v4_addrs(idev);
 
 	if (dev->flags&IFF_POINTOPOINT) {
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index d7ec4c9ffc4b..e2c980dbe52d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1659,6 +1659,8 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	return rt;
 }
 
+EXPORT_SYMBOL(rt6_get_dflt_router);
+
 struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 				     struct net_device *dev,
 				     unsigned int pref)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 71433d29d884..b3b8513e9cb7 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -16,6 +16,7 @@
  *	Changes:
  * Roger Venning <r.venning@telstra.com>:	6to4 support
  * Nate Thompson <nate@thebog.net>:		6to4 support
+ * Fred L. Templin <fltemplin@acm.org>:		isatap support
  */
 
 #include <linux/module.h>
@@ -182,6 +183,9 @@ static struct ip_tunnel * ipip6_tunnel_locate(struct ip_tunnel_parm *parms, int
 	dev->init = ipip6_tunnel_init;
 	nt->parms = *parms;
 
+	if (parms->i_flags & SIT_ISATAP)
+		dev->priv_flags |= IFF_ISATAP;
+
 	if (register_netdevice(dev) < 0) {
 		free_netdev(dev);
 		goto failed;
@@ -364,6 +368,48 @@ static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
 		IP6_ECN_set_ce(ipv6_hdr(skb));
 }
 
+/* ISATAP (RFC4214) - check source address */
+static int
+isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev)
+{
+	struct neighbour *neigh;
+	struct dst_entry *dst;
+	struct rt6_info *rt;
+	struct flowi fl;
+	struct in6_addr *addr6;
+	struct in6_addr rtr;
+	struct ipv6hdr *iph6;
+	int ok = 0;
+
+	/* from onlink default router */
+	ipv6_addr_set(&rtr,  htonl(0xFE800000), 0, 0, 0);
+	ipv6_isatap_eui64(rtr.s6_addr + 8, iph->saddr);
+	if ((rt = rt6_get_dflt_router(&rtr, dev))) {
+		dst_release(&rt->u.dst);
+		return 1;
+	}
+
+	iph6 = ipv6_hdr(skb);
+	memset(&fl, 0, sizeof(fl));
+	fl.proto = iph6->nexthdr;
+	ipv6_addr_copy(&fl.fl6_dst, &iph6->saddr);
+	fl.oif = dev->ifindex;
+	security_skb_classify_flow(skb, &fl);
+
+	dst = ip6_route_output(NULL, &fl);
+	if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) {
+
+		addr6 = (struct in6_addr*)&neigh->primary_key;
+
+		/* from correct previous hop */
+		if (ipv6_addr_is_isatap(addr6) &&
+		    (addr6->s6_addr32[3] == iph->saddr))
+			ok = 1;
+	}
+	dst_release(dst);
+	return ok;
+}
+
 static int ipip6_rcv(struct sk_buff *skb)
 {
 	struct iphdr *iph;
@@ -382,6 +428,14 @@ static int ipip6_rcv(struct sk_buff *skb)
 		IPCB(skb)->flags = 0;
 		skb->protocol = htons(ETH_P_IPV6);
 		skb->pkt_type = PACKET_HOST;
+
+		if ((tunnel->dev->priv_flags & IFF_ISATAP) &&
+		    !isatap_srcok(skb, iph, tunnel->dev)) {
+			tunnel->stat.rx_errors++;
+			read_unlock(&ipip6_lock);
+			kfree_skb(skb);
+			return 0;
+		}
 		tunnel->stat.rx_packets++;
 		tunnel->stat.rx_bytes += skb->len;
 		skb->dev = tunnel->dev;
@@ -444,6 +498,29 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (skb->protocol != htons(ETH_P_IPV6))
 		goto tx_error;
 
+	/* ISATAP (RFC4214) - must come before 6to4 */
+	if (dev->priv_flags & IFF_ISATAP) {
+		struct neighbour *neigh = NULL;
+
+		if (skb->dst)
+			neigh = skb->dst->neighbour;
+
+		if (neigh == NULL) {
+			if (net_ratelimit())
+				printk(KERN_DEBUG "sit: nexthop == NULL\n");
+			goto tx_error;
+		}
+
+		addr6 = (struct in6_addr*)&neigh->primary_key;
+		addr_type = ipv6_addr_type(addr6);
+
+		if ((addr_type & IPV6_ADDR_UNICAST) &&
+		     ipv6_addr_is_isatap(addr6))
+			dst = addr6->s6_addr32[3];
+		else
+			goto tx_error;
+	}
+
 	if (!dst)
 		dst = try_6to4(&iph6->daddr);
 
-- 
cgit v1.2.3


From f11135a3442996d78dad99933bfdb90d1f6588d3 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Wed, 28 Nov 2007 11:34:53 -0200
Subject: [DCCP]: Dedicated auxiliary states to support passive-close

This adds two auxiliary states to deal with passive closes:
  * PASSIVE_CLOSE    (reached from OPEN via reception of Close)    and
  * PASSIVE_CLOSEREQ (reached from OPEN via reception of CloseReq)
as internal intermediate states.

These states are used to allow a receiver to process unread data before
acknowledging the received connection-termination-request (the Close/CloseReq).

Without such support, it will happen that passively-closed sockets enter CLOSED
state while there is still unprocessed data in the queue; leading to unexpected
and erratic API behaviour.

PASSIVE_CLOSE has been mapped into TCPF_CLOSE_WAIT, so that the code will
seamlessly work with inet_accept() (which tests for this state).

The state names are thanks to Arnaldo, who suggested this naming scheme
following an earlier revision of this patch.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h | 56 ++++++++++++++++++++++++++++++++++++----------------
 net/dccp/proto.c     | 22 +++++++++++----------
 2 files changed, 51 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index a0073268808f..8b3f9ad3cf04 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -227,29 +227,51 @@ struct dccp_so_feat {
 #include <net/tcp_states.h>
 
 enum dccp_state {
-	DCCP_OPEN	= TCP_ESTABLISHED,
-	DCCP_REQUESTING	= TCP_SYN_SENT,
-	DCCP_LISTEN	= TCP_LISTEN,
-	DCCP_RESPOND	= TCP_SYN_RECV,
-	DCCP_CLOSING	= TCP_CLOSING,
-	DCCP_TIME_WAIT	= TCP_TIME_WAIT,
-	DCCP_CLOSED	= TCP_CLOSE,
-	DCCP_PARTOPEN	= TCP_MAX_STATES,
+	DCCP_OPEN	     = TCP_ESTABLISHED,
+	DCCP_REQUESTING	     = TCP_SYN_SENT,
+	DCCP_LISTEN	     = TCP_LISTEN,
+	DCCP_RESPOND	     = TCP_SYN_RECV,
+	/*
+	 * States involved in closing a DCCP connection:
+	 * 1) ACTIVE_CLOSEREQ is entered by a server sending a CloseReq.
+	 *
+	 * 2) CLOSING can have three different meanings (RFC 4340, 8.3):
+	 *  a. Client has performed active-close, has sent a Close to the server
+	 *     from state OPEN or PARTOPEN, and is waiting for the final Reset
+	 *     (in this case, SOCK_DONE == 1).
+	 *  b. Client is asked to perform passive-close, by receiving a CloseReq
+	 *     in (PART)OPEN state. It sends a Close and waits for final Reset
+	 *     (in this case, SOCK_DONE == 0).
+	 *  c. Server performs an active-close as in (a), keeps TIMEWAIT state.
+	 *
+	 * 3) The following intermediate states are employed to give passively
+	 *    closing nodes a chance to process their unread data:
+	 *    - PASSIVE_CLOSE    (from OPEN => CLOSED) and
+	 *    - PASSIVE_CLOSEREQ (from (PART)OPEN to CLOSING; case (b) above).
+	 */
+	DCCP_ACTIVE_CLOSEREQ = TCP_FIN_WAIT1,
+	DCCP_PASSIVE_CLOSE   = TCP_CLOSE_WAIT,	/* any node receiving a Close */
+	DCCP_CLOSING	     = TCP_CLOSING,
+	DCCP_TIME_WAIT	     = TCP_TIME_WAIT,
+	DCCP_CLOSED	     = TCP_CLOSE,
+	DCCP_PARTOPEN	     = TCP_MAX_STATES,
+	DCCP_PASSIVE_CLOSEREQ,			/* clients receiving CloseReq */
 	DCCP_MAX_STATES
 };
 
-#define DCCP_STATE_MASK 0xf
+#define DCCP_STATE_MASK 0x1f
 #define DCCP_ACTION_FIN (1<<7)
 
 enum {
-	DCCPF_OPEN	 = TCPF_ESTABLISHED,
-	DCCPF_REQUESTING = TCPF_SYN_SENT,
-	DCCPF_LISTEN	 = TCPF_LISTEN,
-	DCCPF_RESPOND	 = TCPF_SYN_RECV,
-	DCCPF_CLOSING	 = TCPF_CLOSING,
-	DCCPF_TIME_WAIT	 = TCPF_TIME_WAIT,
-	DCCPF_CLOSED	 = TCPF_CLOSE,
-	DCCPF_PARTOPEN	 = 1 << DCCP_PARTOPEN,
+	DCCPF_OPEN	      = TCPF_ESTABLISHED,
+	DCCPF_REQUESTING      = TCPF_SYN_SENT,
+	DCCPF_LISTEN	      = TCPF_LISTEN,
+	DCCPF_RESPOND	      = TCPF_SYN_RECV,
+	DCCPF_ACTIVE_CLOSEREQ = TCPF_FIN_WAIT1,
+	DCCPF_CLOSING	      = TCPF_CLOSING,
+	DCCPF_TIME_WAIT	      = TCPF_TIME_WAIT,
+	DCCPF_CLOSED	      = TCPF_CLOSE,
+	DCCPF_PARTOPEN	      = (1 << DCCP_PARTOPEN),
 };
 
 static inline struct dccp_hdr *dccp_hdr(const struct sk_buff *skb)
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 73006b747678..3489d3f21f50 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -60,8 +60,7 @@ void dccp_set_state(struct sock *sk, const int state)
 {
 	const int oldstate = sk->sk_state;
 
-	dccp_pr_debug("%s(%p) %-10.10s -> %s\n",
-		      dccp_role(sk), sk,
+	dccp_pr_debug("%s(%p)  %s  -->  %s\n", dccp_role(sk), sk,
 		      dccp_state_name(oldstate), dccp_state_name(state));
 	WARN_ON(state == oldstate);
 
@@ -134,14 +133,17 @@ EXPORT_SYMBOL_GPL(dccp_packet_name);
 const char *dccp_state_name(const int state)
 {
 	static char *dccp_state_names[] = {
-	[DCCP_OPEN]	  = "OPEN",
-	[DCCP_REQUESTING] = "REQUESTING",
-	[DCCP_PARTOPEN]	  = "PARTOPEN",
-	[DCCP_LISTEN]	  = "LISTEN",
-	[DCCP_RESPOND]	  = "RESPOND",
-	[DCCP_CLOSING]	  = "CLOSING",
-	[DCCP_TIME_WAIT]  = "TIME_WAIT",
-	[DCCP_CLOSED]	  = "CLOSED",
+	[DCCP_OPEN]		= "OPEN",
+	[DCCP_REQUESTING]	= "REQUESTING",
+	[DCCP_PARTOPEN]		= "PARTOPEN",
+	[DCCP_LISTEN]		= "LISTEN",
+	[DCCP_RESPOND]		= "RESPOND",
+	[DCCP_CLOSING]		= "CLOSING",
+	[DCCP_ACTIVE_CLOSEREQ]	= "CLOSEREQ",
+	[DCCP_PASSIVE_CLOSE]	= "PASSIVE_CLOSE",
+	[DCCP_PASSIVE_CLOSEREQ]	= "PASSIVE_CLOSEREQ",
+	[DCCP_TIME_WAIT]	= "TIME_WAIT",
+	[DCCP_CLOSED]		= "CLOSED",
 	};
 
 	if (state >= DCCP_MAX_STATES)
-- 
cgit v1.2.3


From 0c869620762fea4b3acf6502d9e80840b27ec642 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Wed, 28 Nov 2007 11:59:48 -0200
Subject: [DCCP]: Integrate state transitions for passive-close

This adds the necessary state transitions for the two forms of passive-close

 * PASSIVE_CLOSE    - which is entered when a host   receives a Close;
 * PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.

Here is a detailed account of what the patch does in each state.

1) Receiving CloseReq

  The pseudo-code in 8.5 says:

     Step 13: Process CloseReq
          If P.type == CloseReq and S.state < CLOSEREQ,
              Generate Close
              S.state := CLOSING
              Set CLOSING timer.

  This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.

   * CLOSED:         silently ignore - it may be a late or duplicate CloseReq;
   * LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
   * REQUEST:        perform Step 13 directly (no need to enqueue packet);
   * OPEN/PARTOPEN:  enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.

  When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
  I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
  the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
  gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.

2) Receiving Close

  The code below from 8.5 is unconditional.

     Step 14: Process Close
          If P.type == Close,
              Generate Reset(Closed)
              Tear down connection
              Drop packet and return

  Thus we need to consider all states:
   * CLOSED:           silently ignore, since this can happen when a retransmitted or late Close arrives;
   * LISTEN:           dccp_rcv_state_process() will generate a Reset ("No Connection");
   * REQUEST:          perform Step 14 directly (no need to enqueue packet);
   * RESPOND:          dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
   * OPEN/PARTOPEN:    enter PASSIVE_CLOSE so that application has a chance to process unread data;
   * CLOSEREQ:         server performed active-close -- perform Step 14;
   * CLOSING:          simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
   * PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
   * TIMEWAIT:         packet is ignored.

   Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
   is no socket for such a connection": the socket still exists, but its state indicates it is unusable.

   Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
   sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h |  1 -
 net/dccp/input.c     | 88 +++++++++++++++++++++++++++++++++++++++++++---------
 net/dccp/proto.c     | 88 ++++++++++++++++++++++++++++++++++------------------
 3 files changed, 131 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 8b3f9ad3cf04..312b989c7edb 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -260,7 +260,6 @@ enum dccp_state {
 };
 
 #define DCCP_STATE_MASK 0x1f
-#define DCCP_ACTION_FIN (1<<7)
 
 enum {
 	DCCPF_OPEN	      = TCPF_ESTABLISHED,
diff --git a/net/dccp/input.c b/net/dccp/input.c
index ef299fbd7c26..fe4b0fbfa508 100644
--- a/net/dccp/input.c
+++ b/net/dccp/input.c
@@ -32,16 +32,56 @@ static void dccp_fin(struct sock *sk, struct sk_buff *skb)
 	sk->sk_data_ready(sk, 0);
 }
 
-static void dccp_rcv_close(struct sock *sk, struct sk_buff *skb)
+static int dccp_rcv_close(struct sock *sk, struct sk_buff *skb)
 {
-	dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
-	dccp_fin(sk, skb);
-	dccp_set_state(sk, DCCP_CLOSED);
-	sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
+	int queued = 0;
+
+	switch (sk->sk_state) {
+	/*
+	 * We ignore Close when received in one of the following states:
+	 *  - CLOSED		(may be a late or duplicate packet)
+	 *  - PASSIVE_CLOSEREQ	(the peer has sent a CloseReq earlier)
+	 *  - RESPOND		(already handled by dccp_check_req)
+	 */
+	case DCCP_CLOSING:
+		/*
+		 * Simultaneous-close: receiving a Close after sending one. This
+		 * can happen if both client and server perform active-close and
+		 * will result in an endless ping-pong of crossing and retrans-
+		 * mitted Close packets, which only terminates when one of the
+		 * nodes times out (min. 64 seconds). Quicker convergence can be
+		 * achieved when one of the nodes acts as tie-breaker.
+		 * This is ok as both ends are done with data transfer and each
+		 * end is just waiting for the other to acknowledge termination.
+		 */
+		if (dccp_sk(sk)->dccps_role != DCCP_ROLE_CLIENT)
+			break;
+		/* fall through */
+	case DCCP_REQUESTING:
+	case DCCP_ACTIVE_CLOSEREQ:
+		dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
+		dccp_done(sk);
+		break;
+	case DCCP_OPEN:
+	case DCCP_PARTOPEN:
+		/* Give waiting application a chance to read pending data */
+		queued = 1;
+		dccp_fin(sk, skb);
+		dccp_set_state(sk, DCCP_PASSIVE_CLOSE);
+		/* fall through */
+	case DCCP_PASSIVE_CLOSE:
+		/*
+		 * Retransmitted Close: we have already enqueued the first one.
+		 */
+		sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
+	}
+	return queued;
 }
 
-static void dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb)
+static int dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb)
 {
+	int queued = 0;
+
 	/*
 	 *   Step 7: Check for unexpected packet types
 	 *      If (S.is_server and P.type == CloseReq)
@@ -50,12 +90,26 @@ static void dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb)
 	 */
 	if (dccp_sk(sk)->dccps_role != DCCP_ROLE_CLIENT) {
 		dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq, DCCP_PKT_SYNC);
-		return;
+		return queued;
 	}
 
-	if (sk->sk_state != DCCP_CLOSING)
+	/* Step 13: process relevant Client states < CLOSEREQ */
+	switch (sk->sk_state) {
+	case DCCP_REQUESTING:
+		dccp_send_close(sk, 0);
 		dccp_set_state(sk, DCCP_CLOSING);
-	dccp_send_close(sk, 0);
+		break;
+	case DCCP_OPEN:
+	case DCCP_PARTOPEN:
+		/* Give waiting application a chance to read pending data */
+		queued = 1;
+		dccp_fin(sk, skb);
+		dccp_set_state(sk, DCCP_PASSIVE_CLOSEREQ);
+		/* fall through */
+	case DCCP_PASSIVE_CLOSEREQ:
+		sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
+	}
+	return queued;
 }
 
 static u8 dccp_reset_code_convert(const u8 code)
@@ -247,11 +301,13 @@ static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
 		dccp_rcv_reset(sk, skb);
 		return 0;
 	case DCCP_PKT_CLOSEREQ:
-		dccp_rcv_closereq(sk, skb);
+		if (dccp_rcv_closereq(sk, skb))
+			return 0;
 		goto discard;
 	case DCCP_PKT_CLOSE:
-		dccp_rcv_close(sk, skb);
-		return 0;
+		if (dccp_rcv_close(sk, skb))
+			return 0;
+		goto discard;
 	case DCCP_PKT_REQUEST:
 		/* Step 7
 		 *   or (S.is_server and P.type == Response)
@@ -590,11 +646,13 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		dccp_send_sync(sk, dcb->dccpd_seq, DCCP_PKT_SYNC);
 		goto discard;
 	} else if (dh->dccph_type == DCCP_PKT_CLOSEREQ) {
-		dccp_rcv_closereq(sk, skb);
+		if (dccp_rcv_closereq(sk, skb))
+			return 0;
 		goto discard;
 	} else if (dh->dccph_type == DCCP_PKT_CLOSE) {
-		dccp_rcv_close(sk, skb);
-		return 0;
+		if (dccp_rcv_close(sk, skb))
+			return 0;
+		goto discard;
 	}
 
 	switch (sk->sk_state) {
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 3489d3f21f50..60f40ec72ff3 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -71,7 +71,8 @@ void dccp_set_state(struct sock *sk, const int state)
 		break;
 
 	case DCCP_CLOSED:
-		if (oldstate == DCCP_CLOSING || oldstate == DCCP_OPEN)
+		if (oldstate == DCCP_OPEN || oldstate == DCCP_ACTIVE_CLOSEREQ ||
+		    oldstate == DCCP_CLOSING)
 			DCCP_INC_STATS(DCCP_MIB_ESTABRESETS);
 
 		sk->sk_prot->unhash(sk);
@@ -92,6 +93,24 @@ void dccp_set_state(struct sock *sk, const int state)
 
 EXPORT_SYMBOL_GPL(dccp_set_state);
 
+static void dccp_finish_passive_close(struct sock *sk)
+{
+	switch (sk->sk_state) {
+	case DCCP_PASSIVE_CLOSE:
+		/* Node (client or server) has received Close packet. */
+		dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
+		dccp_set_state(sk, DCCP_CLOSED);
+		break;
+	case DCCP_PASSIVE_CLOSEREQ:
+		/*
+		 * Client received CloseReq. We set the `active' flag so that
+		 * dccp_send_close() retransmits the Close as per RFC 4340, 8.3.
+		 */
+		dccp_send_close(sk, 1);
+		dccp_set_state(sk, DCCP_CLOSING);
+	}
+}
+
 void dccp_done(struct sock *sk)
 {
 	dccp_set_state(sk, DCCP_CLOSED);
@@ -762,19 +781,26 @@ int dccp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 		dh = dccp_hdr(skb);
 
-		if (dh->dccph_type == DCCP_PKT_DATA ||
-		    dh->dccph_type == DCCP_PKT_DATAACK)
+		switch (dh->dccph_type) {
+		case DCCP_PKT_DATA:
+		case DCCP_PKT_DATAACK:
 			goto found_ok_skb;
 
-		if (dh->dccph_type == DCCP_PKT_RESET ||
-		    dh->dccph_type == DCCP_PKT_CLOSE) {
-			dccp_pr_debug("found fin ok!\n");
+		case DCCP_PKT_CLOSE:
+		case DCCP_PKT_CLOSEREQ:
+			if (!(flags & MSG_PEEK))
+				dccp_finish_passive_close(sk);
+			/* fall through */
+		case DCCP_PKT_RESET:
+			dccp_pr_debug("found fin (%s) ok!\n",
+				      dccp_packet_name(dh->dccph_type));
 			len = 0;
 			goto found_fin_ok;
+		default:
+			dccp_pr_debug("packet_type=%s\n",
+				      dccp_packet_name(dh->dccph_type));
+			sk_eat_skb(sk, skb, 0);
 		}
-		dccp_pr_debug("packet_type=%s\n",
-			      dccp_packet_name(dh->dccph_type));
-		sk_eat_skb(sk, skb, 0);
 verify_sock_status:
 		if (sock_flag(sk, SOCK_DONE)) {
 			len = 0;
@@ -876,28 +902,30 @@ out:
 
 EXPORT_SYMBOL_GPL(inet_dccp_listen);
 
-static const unsigned char dccp_new_state[] = {
-	/* current state:   new state:      action:	*/
-	[0]		  = DCCP_CLOSED,
-	[DCCP_OPEN]	  = DCCP_CLOSING | DCCP_ACTION_FIN,
-	[DCCP_REQUESTING] = DCCP_CLOSED,
-	[DCCP_PARTOPEN]	  = DCCP_CLOSING | DCCP_ACTION_FIN,
-	[DCCP_LISTEN]	  = DCCP_CLOSED,
-	[DCCP_RESPOND]	  = DCCP_CLOSED,
-	[DCCP_CLOSING]	  = DCCP_CLOSED,
-	[DCCP_TIME_WAIT]  = DCCP_CLOSED,
-	[DCCP_CLOSED]	  = DCCP_CLOSED,
-};
-
-static int dccp_close_state(struct sock *sk)
+static void dccp_terminate_connection(struct sock *sk)
 {
-	const int next = dccp_new_state[sk->sk_state];
-	const int ns = next & DCCP_STATE_MASK;
+	u8 next_state = DCCP_CLOSED;
 
-	if (ns != sk->sk_state)
-		dccp_set_state(sk, ns);
+	switch (sk->sk_state) {
+	case DCCP_PASSIVE_CLOSE:
+	case DCCP_PASSIVE_CLOSEREQ:
+		dccp_finish_passive_close(sk);
+		break;
+	case DCCP_PARTOPEN:
+		dccp_pr_debug("Stop PARTOPEN timer (%p)\n", sk);
+		inet_csk_clear_xmit_timer(sk, ICSK_TIME_DACK);
+		/* fall through */
+	case DCCP_OPEN:
+		dccp_send_close(sk, 1);
 
-	return next & DCCP_ACTION_FIN;
+		if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER)
+			next_state = DCCP_ACTIVE_CLOSEREQ;
+		else
+			next_state = DCCP_CLOSING;
+		/* fall through */
+	default:
+		dccp_set_state(sk, next_state);
+	}
 }
 
 void dccp_close(struct sock *sk, long timeout)
@@ -940,8 +968,8 @@ void dccp_close(struct sock *sk, long timeout)
 	} else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
 		/* Check zero linger _after_ checking for unread data. */
 		sk->sk_prot->disconnect(sk, 0);
-	} else if (dccp_close_state(sk)) {
-		dccp_send_close(sk, 1);
+	} else if (sk->sk_state != DCCP_CLOSED) {
+		dccp_terminate_connection(sk);
 	}
 
 	sk_stream_wait_close(sk, timeout);
-- 
cgit v1.2.3


From a99a00cf1adef2d3dce745c93c9cc8b0a1612c50 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 30 Nov 2007 01:14:30 +1100
Subject: [NET]: Move netfilter checksum helpers to net/core/utils.c

This allows to get rid of the CONFIG_NETFILTER dependency of NET_ACT_NAT.
This patch redefines the old names to keep the noise low, the next patch
converts all users.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h | 25 ++++---------------------
 include/net/checksum.h    | 25 +++++++++++++++++++++++++
 net/core/utils.c          | 16 ++++++++++++++++
 net/netfilter/core.c      | 16 ----------------
 4 files changed, 45 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 25fc12260340..e2bf6d2ffb6c 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -298,27 +298,10 @@ extern void nf_invalidate_cache(int pf);
    Returns true or false. */
 extern int skb_make_writable(struct sk_buff *skb, unsigned int writable_len);
 
-static inline void nf_csum_replace4(__sum16 *sum, __be32 from, __be32 to)
-{
-	__be32 diff[] = { ~from, to };
-
-	*sum = csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum_unfold(*sum)));
-}
-
-static inline void nf_csum_replace2(__sum16 *sum, __be16 from, __be16 to)
-{
-	nf_csum_replace4(sum, (__force __be32)from, (__force __be32)to);
-}
-
-extern void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
-				      __be32 from, __be32 to, int pseudohdr);
-
-static inline void nf_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb,
-				      __be16 from, __be16 to, int pseudohdr)
-{
-	nf_proto_csum_replace4(sum, skb, (__force __be32)from,
-				(__force __be32)to, pseudohdr);
-}
+#define nf_csum_replace4	csum_replace4
+#define nf_csum_replace2	csum_replace2
+#define nf_proto_csum_replace4	inet_proto_csum_replace4
+#define nf_proto_csum_replace2	inet_proto_csum_replace2
 
 struct nf_afinfo {
 	unsigned short	family;
diff --git a/include/net/checksum.h b/include/net/checksum.h
index 124246172a88..07602b7fa218 100644
--- a/include/net/checksum.h
+++ b/include/net/checksum.h
@@ -93,4 +93,29 @@ static inline __wsum csum_unfold(__sum16 n)
 }
 
 #define CSUM_MANGLED_0 ((__force __sum16)0xffff)
+
+static inline void csum_replace4(__sum16 *sum, __be32 from, __be32 to)
+{
+	__be32 diff[] = { ~from, to };
+
+	*sum = csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum_unfold(*sum)));
+}
+
+static inline void csum_replace2(__sum16 *sum, __be16 from, __be16 to)
+{
+	csum_replace4(sum, (__force __be32)from, (__force __be32)to);
+}
+
+struct sk_buff;
+extern void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
+				     __be32 from, __be32 to, int pseudohdr);
+
+static inline void inet_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb,
+					    __be16 from, __be16 to,
+					    int pseudohdr)
+{
+	inet_proto_csum_replace4(sum, skb, (__force __be32)from,
+				 (__force __be32)to, pseudohdr);
+}
+
 #endif
diff --git a/net/core/utils.c b/net/core/utils.c
index 0bf17da40d52..34459c4c740c 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -293,3 +293,19 @@ out:
 }
 
 EXPORT_SYMBOL(in6_pton);
+
+void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
+			      __be32 from, __be32 to, int pseudohdr)
+{
+	__be32 diff[] = { ~from, to };
+	if (skb->ip_summed != CHECKSUM_PARTIAL) {
+		*sum = csum_fold(csum_partial(diff, sizeof(diff),
+				~csum_unfold(*sum)));
+		if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr)
+			skb->csum = ~csum_partial(diff, sizeof(diff),
+						~skb->csum);
+	} else if (pseudohdr)
+		*sum = ~csum_fold(csum_partial(diff, sizeof(diff),
+				csum_unfold(*sum)));
+}
+EXPORT_SYMBOL(inet_proto_csum_replace4);
diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index bed9ba01e8ec..631d26948314 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -217,22 +217,6 @@ int skb_make_writable(struct sk_buff *skb, unsigned int writable_len)
 }
 EXPORT_SYMBOL(skb_make_writable);
 
-void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
-			    __be32 from, __be32 to, int pseudohdr)
-{
-	__be32 diff[] = { ~from, to };
-	if (skb->ip_summed != CHECKSUM_PARTIAL) {
-		*sum = csum_fold(csum_partial(diff, sizeof(diff),
-				~csum_unfold(*sum)));
-		if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr)
-			skb->csum = ~csum_partial(diff, sizeof(diff),
-						~skb->csum);
-	} else if (pseudohdr)
-		*sum = ~csum_fold(csum_partial(diff, sizeof(diff),
-				csum_unfold(*sum)));
-}
-EXPORT_SYMBOL(nf_proto_csum_replace4);
-
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 /* This does not belong here, but locally generated errors need it if connection
    tracking in use: without this, connection may not be in hash table, and hence
-- 
cgit v1.2.3


From be0ea7d5da3d99140bde7e5cea328eb111731700 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 30 Nov 2007 01:17:11 +1100
Subject: [NETFILTER]: Convert old checksum helper names

Kill the defines again, convert to the new checksum helper names and
remove the dependency of NET_ACT_NAT on NETFILTER.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h              |  5 -----
 net/ipv4/netfilter/ipt_ECN.c           |  6 +++---
 net/ipv4/netfilter/ipt_TOS.c           |  2 +-
 net/ipv4/netfilter/ipt_TTL.c           |  4 ++--
 net/ipv4/netfilter/nf_nat_core.c       |  4 ++--
 net/ipv4/netfilter/nf_nat_helper.c     | 20 ++++++++++----------
 net/ipv4/netfilter/nf_nat_proto_icmp.c |  4 ++--
 net/ipv4/netfilter/nf_nat_proto_tcp.c  |  4 ++--
 net/ipv4/netfilter/nf_nat_proto_udp.c  |  6 +++---
 net/netfilter/xt_TCPMSS.c              | 17 +++++++++--------
 net/sched/Kconfig                      |  1 -
 net/sched/act_nat.c                    | 12 ++++++------
 12 files changed, 40 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index e2bf6d2ffb6c..f42e4362d50d 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -298,11 +298,6 @@ extern void nf_invalidate_cache(int pf);
    Returns true or false. */
 extern int skb_make_writable(struct sk_buff *skb, unsigned int writable_len);
 
-#define nf_csum_replace4	csum_replace4
-#define nf_csum_replace2	csum_replace2
-#define nf_proto_csum_replace4	inet_proto_csum_replace4
-#define nf_proto_csum_replace2	inet_proto_csum_replace2
-
 struct nf_afinfo {
 	unsigned short	family;
 	__sum16		(*checksum)(struct sk_buff *skb, unsigned int hook,
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index add110060a22..e8d5f6867970 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -38,7 +38,7 @@ set_ect_ip(struct sk_buff *skb, const struct ipt_ECN_info *einfo)
 		oldtos = iph->tos;
 		iph->tos &= ~IPT_ECN_IP_MASK;
 		iph->tos |= (einfo->ip_ect & IPT_ECN_IP_MASK);
-		nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos));
+		csum_replace2(&iph->check, htons(oldtos), htons(iph->tos));
 	}
 	return true;
 }
@@ -71,8 +71,8 @@ set_ect_tcp(struct sk_buff *skb, const struct ipt_ECN_info *einfo)
 	if (einfo->operation & IPT_ECN_OP_SET_CWR)
 		tcph->cwr = einfo->proto.tcp.cwr;
 
-	nf_proto_csum_replace2(&tcph->check, skb,
-				oldval, ((__be16 *)tcph)[6], 0);
+	inet_proto_csum_replace2(&tcph->check, skb,
+				 oldval, ((__be16 *)tcph)[6], 0);
 	return true;
 }
 
diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c
index d4573baa7f27..7b4a6cac399e 100644
--- a/net/ipv4/netfilter/ipt_TOS.c
+++ b/net/ipv4/netfilter/ipt_TOS.c
@@ -38,7 +38,7 @@ target(struct sk_buff *skb,
 		iph = ip_hdr(skb);
 		oldtos = iph->tos;
 		iph->tos = (iph->tos & IPTOS_PREC_MASK) | tosinfo->tos;
-		nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos));
+		csum_replace2(&iph->check, htons(oldtos), htons(iph->tos));
 	}
 	return XT_CONTINUE;
 }
diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c
index c620a0527666..00ddfbea2799 100644
--- a/net/ipv4/netfilter/ipt_TTL.c
+++ b/net/ipv4/netfilter/ipt_TTL.c
@@ -54,8 +54,8 @@ ipt_ttl_target(struct sk_buff *skb,
 	}
 
 	if (new_ttl != iph->ttl) {
-		nf_csum_replace2(&iph->check, htons(iph->ttl << 8),
-					      htons(new_ttl << 8));
+		csum_replace2(&iph->check, htons(iph->ttl << 8),
+					   htons(new_ttl << 8));
 		iph->ttl = new_ttl;
 	}
 
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index d237511cf46c..746c2efddcd1 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -372,10 +372,10 @@ manip_pkt(u_int16_t proto,
 	iph = (void *)skb->data + iphdroff;
 
 	if (maniptype == IP_NAT_MANIP_SRC) {
-		nf_csum_replace4(&iph->check, iph->saddr, target->src.u3.ip);
+		csum_replace4(&iph->check, iph->saddr, target->src.u3.ip);
 		iph->saddr = target->src.u3.ip;
 	} else {
-		nf_csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip);
+		csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip);
 		iph->daddr = target->dst.u3.ip;
 	}
 	return 1;
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index d00b8b2891fb..53f79a310b43 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -180,8 +180,8 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb,
 								datalen, 0));
 		}
 	} else
-		nf_proto_csum_replace2(&tcph->check, skb,
-				       htons(oldlen), htons(datalen), 1);
+		inet_proto_csum_replace2(&tcph->check, skb,
+					 htons(oldlen), htons(datalen), 1);
 
 	if (rep_len != match_len) {
 		set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
@@ -270,8 +270,8 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb,
 				udph->check = CSUM_MANGLED_0;
 		}
 	} else
-		nf_proto_csum_replace2(&udph->check, skb,
-				       htons(oldlen), htons(datalen), 1);
+		inet_proto_csum_replace2(&udph->check, skb,
+					 htons(oldlen), htons(datalen), 1);
 
 	return 1;
 }
@@ -310,10 +310,10 @@ sack_adjust(struct sk_buff *skb,
 			 ntohl(sack->start_seq), new_start_seq,
 			 ntohl(sack->end_seq), new_end_seq);
 
-		nf_proto_csum_replace4(&tcph->check, skb,
-				       sack->start_seq, new_start_seq, 0);
-		nf_proto_csum_replace4(&tcph->check, skb,
-				       sack->end_seq, new_end_seq, 0);
+		inet_proto_csum_replace4(&tcph->check, skb,
+					 sack->start_seq, new_start_seq, 0);
+		inet_proto_csum_replace4(&tcph->check, skb,
+					 sack->end_seq, new_end_seq, 0);
 		sack->start_seq = new_start_seq;
 		sack->end_seq = new_end_seq;
 		sackoff += sizeof(*sack);
@@ -397,8 +397,8 @@ nf_nat_seq_adjust(struct sk_buff *skb,
 	else
 		newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before);
 
-	nf_proto_csum_replace4(&tcph->check, skb, tcph->seq, newseq, 0);
-	nf_proto_csum_replace4(&tcph->check, skb, tcph->ack_seq, newack, 0);
+	inet_proto_csum_replace4(&tcph->check, skb, tcph->seq, newseq, 0);
+	inet_proto_csum_replace4(&tcph->check, skb, tcph->ack_seq, newack, 0);
 
 	pr_debug("Adjusting sequence number from %u->%u, ack from %u->%u\n",
 		 ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq),
diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c
index b9fc724388fc..088bb141c15d 100644
--- a/net/ipv4/netfilter/nf_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c
@@ -65,8 +65,8 @@ icmp_manip_pkt(struct sk_buff *skb,
 		return 0;
 
 	hdr = (struct icmphdr *)(skb->data + hdroff);
-	nf_proto_csum_replace2(&hdr->checksum, skb,
-			       hdr->un.echo.id, tuple->src.u.icmp.id, 0);
+	inet_proto_csum_replace2(&hdr->checksum, skb,
+				 hdr->un.echo.id, tuple->src.u.icmp.id, 0);
 	hdr->un.echo.id = tuple->src.u.icmp.id;
 	return 1;
 }
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index 6bab2e184455..633c53ff3d32 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -132,8 +132,8 @@ tcp_manip_pkt(struct sk_buff *skb,
 	if (hdrsize < sizeof(*hdr))
 		return 1;
 
-	nf_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1);
-	nf_proto_csum_replace2(&hdr->check, skb, oldport, newport, 0);
+	inet_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1);
+	inet_proto_csum_replace2(&hdr->check, skb, oldport, newport, 0);
 	return 1;
 }
 
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index cbf1a61e2908..9c6519cd15f5 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -117,9 +117,9 @@ udp_manip_pkt(struct sk_buff *skb,
 		portptr = &hdr->dest;
 	}
 	if (hdr->check || skb->ip_summed == CHECKSUM_PARTIAL) {
-		nf_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1);
-		nf_proto_csum_replace2(&hdr->check, skb, *portptr, newport,
-				       0);
+		inet_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1);
+		inet_proto_csum_replace2(&hdr->check, skb, *portptr, newport,
+					 0);
 		if (!hdr->check)
 			hdr->check = CSUM_MANGLED_0;
 	}
diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c
index f183c8fa47a5..bf6249e44068 100644
--- a/net/netfilter/xt_TCPMSS.c
+++ b/net/netfilter/xt_TCPMSS.c
@@ -95,8 +95,9 @@ tcpmss_mangle_packet(struct sk_buff *skb,
 			opt[i+2] = (newmss & 0xff00) >> 8;
 			opt[i+3] = newmss & 0x00ff;
 
-			nf_proto_csum_replace2(&tcph->check, skb,
-					       htons(oldmss), htons(newmss), 0);
+			inet_proto_csum_replace2(&tcph->check, skb,
+						 htons(oldmss), htons(newmss),
+						 0);
 			return 0;
 		}
 	}
@@ -117,19 +118,19 @@ tcpmss_mangle_packet(struct sk_buff *skb,
 	opt = (u_int8_t *)tcph + sizeof(struct tcphdr);
 	memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr));
 
-	nf_proto_csum_replace2(&tcph->check, skb,
-			       htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1);
+	inet_proto_csum_replace2(&tcph->check, skb,
+				 htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1);
 	opt[0] = TCPOPT_MSS;
 	opt[1] = TCPOLEN_MSS;
 	opt[2] = (newmss & 0xff00) >> 8;
 	opt[3] = newmss & 0x00ff;
 
-	nf_proto_csum_replace4(&tcph->check, skb, 0, *((__be32 *)opt), 0);
+	inet_proto_csum_replace4(&tcph->check, skb, 0, *((__be32 *)opt), 0);
 
 	oldval = ((__be16 *)tcph)[6];
 	tcph->doff += TCPOLEN_MSS/4;
-	nf_proto_csum_replace2(&tcph->check, skb,
-				oldval, ((__be16 *)tcph)[6], 0);
+	inet_proto_csum_replace2(&tcph->check, skb,
+				 oldval, ((__be16 *)tcph)[6], 0);
 	return TCPOLEN_MSS;
 }
 
@@ -152,7 +153,7 @@ xt_tcpmss_target4(struct sk_buff *skb,
 	if (ret > 0) {
 		iph = ip_hdr(skb);
 		newlen = htons(ntohs(iph->tot_len) + ret);
-		nf_csum_replace2(&iph->check, iph->tot_len, newlen);
+		csum_replace2(&iph->check, iph->tot_len, newlen);
 		iph->tot_len = newlen;
 	}
 	return XT_CONTINUE;
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 9c15c4888d12..f5ab54b1edd3 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -445,7 +445,6 @@ config NET_ACT_IPT
 config NET_ACT_NAT
         tristate "Stateless NAT"
         depends on NET_CLS_ACT
-        select NETFILTER
         ---help---
 	  Say Y here to do stateless NAT on IPv4 packets.  You should use
 	  netfilter for NAT unless you know what you are doing.
diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c
index c96273bcaf9c..da5c1eae4222 100644
--- a/net/sched/act_nat.c
+++ b/net/sched/act_nat.c
@@ -151,7 +151,7 @@ static int tcf_nat(struct sk_buff *skb, struct tc_action *a,
 		else
 			iph->daddr = new_addr;
 
-		nf_csum_replace4(&iph->check, addr, new_addr);
+		csum_replace4(&iph->check, addr, new_addr);
 	}
 
 	ihl = iph->ihl * 4;
@@ -169,7 +169,7 @@ static int tcf_nat(struct sk_buff *skb, struct tc_action *a,
 			goto drop;
 
 		tcph = (void *)(skb_network_header(skb) + ihl);
-		nf_proto_csum_replace4(&tcph->check, skb, addr, new_addr, 1);
+		inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr, 1);
 		break;
 	}
 	case IPPROTO_UDP:
@@ -184,8 +184,8 @@ static int tcf_nat(struct sk_buff *skb, struct tc_action *a,
 
 		udph = (void *)(skb_network_header(skb) + ihl);
 		if (udph->check || skb->ip_summed == CHECKSUM_PARTIAL) {
-			nf_proto_csum_replace4(&udph->check, skb, addr,
-					       new_addr, 1);
+			inet_proto_csum_replace4(&udph->check, skb, addr,
+						 new_addr, 1);
 			if (!udph->check)
 				udph->check = CSUM_MANGLED_0;
 		}
@@ -232,8 +232,8 @@ static int tcf_nat(struct sk_buff *skb, struct tc_action *a,
 		else
 			iph->saddr = new_addr;
 
-		nf_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
-				       1);
+		inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
+					 1);
 		break;
 	}
 	default:
-- 
cgit v1.2.3


From 29e796fd4de54b8f5bc30d897611210ece4fd0f2 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Fri, 30 Nov 2007 23:50:18 +1100
Subject: sysctl: Add register_sysctl_paths function

There are a number of modules that register a sysctl table
somewhere deeply nested in the sysctl hierarchy, such as
fs/nfs, fs/xfs, dev/cdrom, etc.

They all specify several dummy ctl_tables for the path name.
This patch implements register_sysctl_path that takes
an additional path name, and makes up dummy sysctl nodes
for each component.

This patch was originally written by Olaf Kirch and
brought to my attention and reworked some by Olaf Hering.
I have changed a few additional things so the bugs are mine.

After converting all of the easy callers Olaf Hering observed
allyesconfig ARCH=i386, the patch reduces the final binary size by 9369 bytes.

.text +897
.data -7008

   text    data     bss     dec     hex filename
   26959310        4045899 4718592 35723801        2211a19 ../vmlinux-vanilla
   26960207        4038891 4718592 35717690        221023a ../O-allyesconfig/vmlinux

So this change is both a space savings and a code simplification.

CC: Olaf Kirch <okir@suse.de>
CC: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Daniel Lezcano <dlezcano@fr.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h |  8 +++++
 kernel/sysctl.c        | 89 +++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 82 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 4f5047df8a9e..3b6e2c9fbb2e 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -1059,7 +1059,15 @@ struct ctl_table_header
 	struct completion *unregistering;
 };
 
+/* struct ctl_path describes where in the hierarchy a table is added */
+struct ctl_path {
+	const char *procname;
+	int ctl_name;
+};
+
 struct ctl_table_header *register_sysctl_table(struct ctl_table * table);
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						struct ctl_table *table);
 
 void unregister_sysctl_table(struct ctl_table_header * table);
 int sysctl_check_table(struct ctl_table *table);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 8e96558cb8f3..f580542333eb 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1561,11 +1561,12 @@ static __init int sysctl_init(void)
 core_initcall(sysctl_init);
 
 /**
- * register_sysctl_table - register a sysctl hierarchy
+ * register_sysctl_paths - register a sysctl hierarchy
+ * @path: The path to the directory the sysctl table is in.
  * @table: the top-level table structure
  *
  * Register a sysctl table hierarchy. @table should be a filled in ctl_table
- * array. An entry with a ctl_name of 0 terminates the table. 
+ * array. A completely 0 filled entry terminates the table.
  *
  * The members of the &struct ctl_table structure are used as follows:
  *
@@ -1628,25 +1629,76 @@ core_initcall(sysctl_init);
  * This routine returns %NULL on a failure to register, and a pointer
  * to the table header on success.
  */
-struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						struct ctl_table *table)
 {
-	struct ctl_table_header *tmp;
-	tmp = kmalloc(sizeof(struct ctl_table_header), GFP_KERNEL);
-	if (!tmp)
+	struct ctl_table_header *header;
+	struct ctl_table *new, **prevp;
+	unsigned int n, npath;
+
+	/* Count the path components */
+	for (npath = 0; path[npath].ctl_name || path[npath].procname; ++npath)
+		;
+
+	/*
+	 * For each path component, allocate a 2-element ctl_table array.
+	 * The first array element will be filled with the sysctl entry
+	 * for this, the second will be the sentinel (ctl_name == 0).
+	 *
+	 * We allocate everything in one go so that we don't have to
+	 * worry about freeing additional memory in unregister_sysctl_table.
+	 */
+	header = kzalloc(sizeof(struct ctl_table_header) +
+			 (2 * npath * sizeof(struct ctl_table)), GFP_KERNEL);
+	if (!header)
 		return NULL;
-	tmp->ctl_table = table;
-	INIT_LIST_HEAD(&tmp->ctl_entry);
-	tmp->used = 0;
-	tmp->unregistering = NULL;
-	sysctl_set_parent(NULL, table);
-	if (sysctl_check_table(tmp->ctl_table)) {
-		kfree(tmp);
+
+	new = (struct ctl_table *) (header + 1);
+
+	/* Now connect the dots */
+	prevp = &header->ctl_table;
+	for (n = 0; n < npath; ++n, ++path) {
+		/* Copy the procname */
+		new->procname = path->procname;
+		new->ctl_name = path->ctl_name;
+		new->mode     = 0555;
+
+		*prevp = new;
+		prevp = &new->child;
+
+		new += 2;
+	}
+	*prevp = table;
+
+	INIT_LIST_HEAD(&header->ctl_entry);
+	header->used = 0;
+	header->unregistering = NULL;
+	sysctl_set_parent(NULL, header->ctl_table);
+	if (sysctl_check_table(header->ctl_table)) {
+		kfree(header);
 		return NULL;
 	}
 	spin_lock(&sysctl_lock);
-	list_add_tail(&tmp->ctl_entry, &root_table_header.ctl_entry);
+	list_add_tail(&header->ctl_entry, &root_table_header.ctl_entry);
 	spin_unlock(&sysctl_lock);
-	return tmp;
+
+	return header;
+}
+
+/**
+ * register_sysctl_table - register a sysctl table hierarchy
+ * @table: the top-level table structure
+ *
+ * Register a sysctl table hierarchy. @table should be a filled in ctl_table
+ * array. A completely 0 filled entry terminates the table.
+ *
+ * See register_sysctl_paths for more details.
+ */
+struct ctl_table_header *register_sysctl_table(struct ctl_table *table)
+{
+	static const struct ctl_path null_path[] = { {} };
+
+	return register_sysctl_paths(null_path, table);
 }
 
 /**
@@ -1675,6 +1727,12 @@ struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
 	return NULL;
 }
 
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						    struct ctl_table *table)
+{
+	return NULL;
+}
+
 void unregister_sysctl_table(struct ctl_table_header * table)
 {
 }
@@ -2733,6 +2791,7 @@ EXPORT_SYMBOL(proc_dostring);
 EXPORT_SYMBOL(proc_doulongvec_minmax);
 EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax);
 EXPORT_SYMBOL(register_sysctl_table);
+EXPORT_SYMBOL(register_sysctl_paths);
 EXPORT_SYMBOL(sysctl_intvec);
 EXPORT_SYMBOL(sysctl_jiffies);
 EXPORT_SYMBOL(sysctl_ms_jiffies);
-- 
cgit v1.2.3


From 23eb06de7d2d333a0f7ebba2da663e00c9c9483e Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Fri, 30 Nov 2007 23:52:10 +1100
Subject: sysctl: Remember the ctl_table we passed to register_sysctl_paths

By doing this we allow users of register_sysctl_paths that build
and dynamically allocate their ctl_table to be simpler.  This allows
them to just remember the ctl_table_header returned from
register_sysctl_paths from which they can now find the
ctl_table array they need to free.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Daniel Lezcano <dlezcano@fr.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h | 1 +
 kernel/sysctl.c        | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 3b6e2c9fbb2e..77de3bfd8744 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -1057,6 +1057,7 @@ struct ctl_table_header
 	struct list_head ctl_entry;
 	int used;
 	struct completion *unregistering;
+	struct ctl_table *ctl_table_arg;
 };
 
 /* struct ctl_path describes where in the hierarchy a table is added */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index f580542333eb..89b7d95ecf51 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1669,6 +1669,7 @@ struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 		new += 2;
 	}
 	*prevp = table;
+	header->ctl_table_arg = table;
 
 	INIT_LIST_HEAD(&header->ctl_entry);
 	header->used = 0;
-- 
cgit v1.2.3


From e51b6ba077791f2f8c876022b37419be7a2ceec3 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Fri, 30 Nov 2007 23:54:00 +1100
Subject: sysctl: Infrastructure for per namespace sysctls

This patch implements the basic infrastructure for per namespace sysctls.

A list of lists of sysctl headers is added, allowing each namespace to have
it's own list of sysctl headers.

Each list of sysctl headers has a lookup function to find the first
sysctl header in the list, allowing the lists to have a per namespace
instance.

register_sysct_root is added to tell sysctl.c about additional
lists of sysctl_headers.  As all of the users are expected to be in
kernel no unregister function is provided.

sysctl_head_next is updated to walk through the list of lists.

__register_sysctl_paths is added to add a new sysctl table on
a non-default sysctl list.

The only intrusive part of this patch is propagating the information
to decided which list of sysctls to use for sysctl_check_table.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Daniel Lezcano <dlezcano@fr.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h | 17 ++++++++-
 kernel/sysctl.c        | 93 ++++++++++++++++++++++++++++++++++++++++++++------
 kernel/sysctl_check.c  | 25 ++++++++------
 3 files changed, 112 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 77de3bfd8744..89faebfe48b8 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -945,7 +945,10 @@ enum
 
 /* For the /proc/sys support */
 struct ctl_table;
+struct nsproxy;
 extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
+extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
+						struct ctl_table_header *prev);
 extern void sysctl_head_finish(struct ctl_table_header *prev);
 extern int sysctl_perm(struct ctl_table *table, int op);
 
@@ -1049,6 +1052,13 @@ struct ctl_table
 	void *extra2;
 };
 
+struct ctl_table_root {
+	struct list_head root_list;
+	struct list_head header_list;
+	struct list_head *(*lookup)(struct ctl_table_root *root,
+					   struct nsproxy *namespaces);
+};
+
 /* struct ctl_table_header is used to maintain dynamic lists of
    struct ctl_table trees. */
 struct ctl_table_header
@@ -1058,6 +1068,7 @@ struct ctl_table_header
 	int used;
 	struct completion *unregistering;
 	struct ctl_table *ctl_table_arg;
+	struct ctl_table_root *root;
 };
 
 /* struct ctl_path describes where in the hierarchy a table is added */
@@ -1066,12 +1077,16 @@ struct ctl_path {
 	int ctl_name;
 };
 
+void register_sysctl_root(struct ctl_table_root *root);
+struct ctl_table_header *__register_sysctl_paths(
+	struct ctl_table_root *root, struct nsproxy *namespaces,
+	const struct ctl_path *path, struct ctl_table *table);
 struct ctl_table_header *register_sysctl_table(struct ctl_table * table);
 struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 						struct ctl_table *table);
 
 void unregister_sysctl_table(struct ctl_table_header * table);
-int sysctl_check_table(struct ctl_table *table);
+int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table);
 
 #else /* __KERNEL__ */
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 89b7d95ecf51..45e76f209dcb 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -157,8 +157,16 @@ static int proc_dointvec_taint(struct ctl_table *table, int write, struct file *
 #endif
 
 static struct ctl_table root_table[];
-static struct ctl_table_header root_table_header =
-	{ root_table, LIST_HEAD_INIT(root_table_header.ctl_entry) };
+static struct ctl_table_root sysctl_table_root;
+static struct ctl_table_header root_table_header = {
+	.ctl_table = root_table,
+	.ctl_entry = LIST_HEAD_INIT(sysctl_table_root.header_list),
+	.root = &sysctl_table_root,
+};
+static struct ctl_table_root sysctl_table_root = {
+	.root_list = LIST_HEAD_INIT(sysctl_table_root.root_list),
+	.header_list = LIST_HEAD_INIT(root_table_header.ctl_entry),
+};
 
 static struct ctl_table kern_table[];
 static struct ctl_table vm_table[];
@@ -1371,12 +1379,27 @@ void sysctl_head_finish(struct ctl_table_header *head)
 	spin_unlock(&sysctl_lock);
 }
 
-struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
+static struct list_head *
+lookup_header_list(struct ctl_table_root *root, struct nsproxy *namespaces)
 {
+	struct list_head *header_list;
+	header_list = &root->header_list;
+	if (root->lookup)
+		header_list = root->lookup(root, namespaces);
+	return header_list;
+}
+
+struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
+					    struct ctl_table_header *prev)
+{
+	struct ctl_table_root *root;
+	struct list_head *header_list;
 	struct ctl_table_header *head;
 	struct list_head *tmp;
+
 	spin_lock(&sysctl_lock);
 	if (prev) {
+		head = prev;
 		tmp = &prev->ctl_entry;
 		unuse_table(prev);
 		goto next;
@@ -1390,14 +1413,38 @@ struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
 		spin_unlock(&sysctl_lock);
 		return head;
 	next:
+		root = head->root;
 		tmp = tmp->next;
-		if (tmp == &root_table_header.ctl_entry)
-			break;
+		header_list = lookup_header_list(root, namespaces);
+		if (tmp != header_list)
+			continue;
+
+		do {
+			root = list_entry(root->root_list.next,
+					struct ctl_table_root, root_list);
+			if (root == &sysctl_table_root)
+				goto out;
+			header_list = lookup_header_list(root, namespaces);
+		} while (list_empty(header_list));
+		tmp = header_list->next;
 	}
+out:
 	spin_unlock(&sysctl_lock);
 	return NULL;
 }
 
+struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
+{
+	return __sysctl_head_next(current->nsproxy, prev);
+}
+
+void register_sysctl_root(struct ctl_table_root *root)
+{
+	spin_lock(&sysctl_lock);
+	list_add_tail(&root->root_list, &sysctl_table_root.root_list);
+	spin_unlock(&sysctl_lock);
+}
+
 #ifdef CONFIG_SYSCTL_SYSCALL
 int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp,
 	       void __user *newval, size_t newlen)
@@ -1554,14 +1601,16 @@ static __init int sysctl_init(void)
 {
 	int err;
 	sysctl_set_parent(NULL, root_table);
-	err = sysctl_check_table(root_table);
+	err = sysctl_check_table(current->nsproxy, root_table);
 	return 0;
 }
 
 core_initcall(sysctl_init);
 
 /**
- * register_sysctl_paths - register a sysctl hierarchy
+ * __register_sysctl_paths - register a sysctl hierarchy
+ * @root: List of sysctl headers to register on
+ * @namespaces: Data to compute which lists of sysctl entries are visible
  * @path: The path to the directory the sysctl table is in.
  * @table: the top-level table structure
  *
@@ -1629,9 +1678,12 @@ core_initcall(sysctl_init);
  * This routine returns %NULL on a failure to register, and a pointer
  * to the table header on success.
  */
-struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
-						struct ctl_table *table)
+struct ctl_table_header *__register_sysctl_paths(
+	struct ctl_table_root *root,
+	struct nsproxy *namespaces,
+	const struct ctl_path *path, struct ctl_table *table)
 {
+	struct list_head *header_list;
 	struct ctl_table_header *header;
 	struct ctl_table *new, **prevp;
 	unsigned int n, npath;
@@ -1674,18 +1726,37 @@ struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 	INIT_LIST_HEAD(&header->ctl_entry);
 	header->used = 0;
 	header->unregistering = NULL;
+	header->root = root;
 	sysctl_set_parent(NULL, header->ctl_table);
-	if (sysctl_check_table(header->ctl_table)) {
+	if (sysctl_check_table(namespaces, header->ctl_table)) {
 		kfree(header);
 		return NULL;
 	}
 	spin_lock(&sysctl_lock);
-	list_add_tail(&header->ctl_entry, &root_table_header.ctl_entry);
+	header_list = lookup_header_list(root, namespaces);
+	list_add_tail(&header->ctl_entry, header_list);
 	spin_unlock(&sysctl_lock);
 
 	return header;
 }
 
+/**
+ * register_sysctl_table_path - register a sysctl table hierarchy
+ * @path: The path to the directory the sysctl table is in.
+ * @table: the top-level table structure
+ *
+ * Register a sysctl table hierarchy. @table should be a filled in ctl_table
+ * array. A completely 0 filled entry terminates the table.
+ *
+ * See __register_sysctl_paths for more details.
+ */
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						struct ctl_table *table)
+{
+	return __register_sysctl_paths(&sysctl_table_root, current->nsproxy,
+					path, table);
+}
+
 /**
  * register_sysctl_table - register a sysctl table hierarchy
  * @table: the top-level table structure
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index d8a5558a47b4..c3206fa50048 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -1342,7 +1342,8 @@ static void sysctl_repair_table(struct ctl_table *table)
 	}
 }
 
-static struct ctl_table *sysctl_check_lookup(struct ctl_table *table)
+static struct ctl_table *sysctl_check_lookup(struct nsproxy *namespaces,
+						struct ctl_table *table)
 {
 	struct ctl_table_header *head;
 	struct ctl_table *ref, *test;
@@ -1350,8 +1351,8 @@ static struct ctl_table *sysctl_check_lookup(struct ctl_table *table)
 
 	depth = sysctl_depth(table);
 
-	for (head = sysctl_head_next(NULL); head;
-	     head = sysctl_head_next(head)) {
+	for (head = __sysctl_head_next(namespaces, NULL); head;
+	     head = __sysctl_head_next(namespaces, head)) {
 		cur_depth = depth;
 		ref = head->ctl_table;
 repeat:
@@ -1396,13 +1397,14 @@ static void set_fail(const char **fail, struct ctl_table *table, const char *str
 	*fail = str;
 }
 
-static int sysctl_check_dir(struct ctl_table *table)
+static int sysctl_check_dir(struct nsproxy *namespaces,
+				struct ctl_table *table)
 {
 	struct ctl_table *ref;
 	int error;
 
 	error = 0;
-	ref = sysctl_check_lookup(table);
+	ref = sysctl_check_lookup(namespaces, table);
 	if (ref) {
 		int match = 0;
 		if ((!table->procname && !ref->procname) ||
@@ -1427,11 +1429,12 @@ static int sysctl_check_dir(struct ctl_table *table)
 	return error;
 }
 
-static void sysctl_check_leaf(struct ctl_table *table, const char **fail)
+static void sysctl_check_leaf(struct nsproxy *namespaces,
+				struct ctl_table *table, const char **fail)
 {
 	struct ctl_table *ref;
 
-	ref = sysctl_check_lookup(table);
+	ref = sysctl_check_lookup(namespaces, table);
 	if (ref && (ref != table))
 		set_fail(fail, table, "Sysctl already exists");
 }
@@ -1455,7 +1458,7 @@ static void sysctl_check_bin_path(struct ctl_table *table, const char **fail)
 	}
 }
 
-int sysctl_check_table(struct ctl_table *table)
+int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table)
 {
 	int error = 0;
 	for (; table->ctl_name || table->procname; table++) {
@@ -1485,7 +1488,7 @@ int sysctl_check_table(struct ctl_table *table)
 				set_fail(&fail, table, "Directory with extra1");
 			if (table->extra2)
 				set_fail(&fail, table, "Directory with extra2");
-			if (sysctl_check_dir(table))
+			if (sysctl_check_dir(namespaces, table))
 				set_fail(&fail, table, "Inconsistent directory names");
 		} else {
 			if ((table->strategy == sysctl_data) ||
@@ -1534,7 +1537,7 @@ int sysctl_check_table(struct ctl_table *table)
 			if (!table->procname && table->proc_handler)
 				set_fail(&fail, table, "proc_handler without procname");
 #endif
-			sysctl_check_leaf(table, &fail);
+			sysctl_check_leaf(namespaces, table, &fail);
 		}
 		sysctl_check_bin_path(table, &fail);
 		if (fail) {
@@ -1542,7 +1545,7 @@ int sysctl_check_table(struct ctl_table *table)
 			error = -EINVAL;
 		}
 		if (table->child)
-			error |= sysctl_check_table(table->child);
+			error |= sysctl_check_table(namespaces, table->child);
 	}
 	return error;
 }
-- 
cgit v1.2.3


From 95bdfccb2bf4ea21c0065772c6a2c75cbaf6ad0d Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Fri, 30 Nov 2007 23:55:42 +1100
Subject: [NET]: Implement the per network namespace sysctl infrastructure

The user interface is: register_net_sysctl_table and
unregister_net_sysctl_table.  Very much like the current
interface except there is a network namespace parameter.

With this any sysctl registered with register_net_sysctl_table
will only show up to tasks in the same network namespace.

All other sysctls continue to be globally visible.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Daniel Lezcano <dlezcano@fr.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  9 +++++++
 net/sysctl_net.c            | 57 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 4d0d6349aa7e..235214c4c231 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -25,6 +25,8 @@ struct net {
 	struct proc_dir_entry 	*proc_net_stat;
 	struct proc_dir_entry 	*proc_net_root;
 
+	struct list_head	sysctl_table_headers;
+
 	struct net_device       *loopback_dev;          /* The loopback */
 
 	struct list_head 	dev_base_head;
@@ -144,4 +146,11 @@ extern void unregister_pernet_subsys(struct pernet_operations *);
 extern int register_pernet_device(struct pernet_operations *);
 extern void unregister_pernet_device(struct pernet_operations *);
 
+struct ctl_path;
+struct ctl_table;
+struct ctl_table_header;
+extern struct ctl_table_header *register_net_sysctl_table(struct net *net,
+	const struct ctl_path *path, struct ctl_table *table);
+extern void unregister_net_sysctl_table(struct ctl_table_header *header);
+
 #endif /* __NET_NET_NAMESPACE_H */
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index cd4eafbab1b8..c50c793aa7f0 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -14,6 +14,7 @@
 
 #include <linux/mm.h>
 #include <linux/sysctl.h>
+#include <linux/nsproxy.h>
 
 #include <net/sock.h>
 
@@ -54,3 +55,59 @@ struct ctl_table net_table[] = {
 #endif
 	{ 0 },
 };
+
+static struct list_head *
+net_ctl_header_lookup(struct ctl_table_root *root, struct nsproxy *namespaces)
+{
+	return &namespaces->net_ns->sysctl_table_headers;
+}
+
+static struct ctl_table_root net_sysctl_root = {
+	.lookup = net_ctl_header_lookup,
+};
+
+static int sysctl_net_init(struct net *net)
+{
+	INIT_LIST_HEAD(&net->sysctl_table_headers);
+	return 0;
+}
+
+static void sysctl_net_exit(struct net *net)
+{
+	WARN_ON(!list_empty(&net->sysctl_table_headers));
+	return;
+}
+
+static struct pernet_operations sysctl_pernet_ops = {
+	.init = sysctl_net_init,
+	.exit = sysctl_net_exit,
+};
+
+static __init int sysctl_init(void)
+{
+	int ret;
+	ret = register_pernet_subsys(&sysctl_pernet_ops);
+	if (ret)
+		goto out;
+	register_sysctl_root(&net_sysctl_root);
+out:
+	return ret;
+}
+subsys_initcall(sysctl_init);
+
+struct ctl_table_header *register_net_sysctl_table(struct net *net,
+	const struct ctl_path *path, struct ctl_table *table)
+{
+	struct nsproxy namespaces;
+	namespaces = *current->nsproxy;
+	namespaces.net_ns = net;
+	return __register_sysctl_paths(&net_sysctl_root,
+					&namespaces, path, table);
+}
+EXPORT_SYMBOL_GPL(register_net_sysctl_table);
+
+void unregister_net_sysctl_table(struct ctl_table_header *header)
+{
+	return unregister_sysctl_table(header);
+}
+EXPORT_SYMBOL_GPL(unregister_net_sysctl_table);
-- 
cgit v1.2.3


From 97577e38284f48ca773392e2d401af2f6c95dc08 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 1 Dec 2007 23:40:40 +1100
Subject: [UNIX]: Extend unix_sysctl_(un)register prototypes

Add the struct net * argument to both of them to use in
the future. Also make the register one return an error code.

It is useless right now, but will make the future patches
much simpler.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/af_unix.h      | 8 ++++----
 net/unix/af_unix.c         | 4 ++--
 net/unix/sysctl_net_unix.c | 5 +++--
 3 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index a1c805d7f488..e0fba269d092 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -60,11 +60,11 @@ struct unix_sock {
 
 #ifdef CONFIG_SYSCTL
 extern int sysctl_unix_max_dgram_qlen;
-extern void unix_sysctl_register(void);
-extern void unix_sysctl_unregister(void);
+extern int unix_sysctl_register(struct net *net);
+extern void unix_sysctl_unregister(struct net *net);
 #else
-static inline void unix_sysctl_register(void) {}
-static inline void unix_sysctl_unregister(void) {}
+static inline int unix_sysctl_register(struct net *net) { return 0; }
+static inline void unix_sysctl_unregister(struct net *net) {}
 #endif
 #endif
 #endif
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 393197afb190..a0aa6d3c2c08 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2175,7 +2175,7 @@ static int __init af_unix_init(void)
 
 	sock_register(&unix_family_ops);
 	register_pernet_subsys(&unix_net_ops);
-	unix_sysctl_register();
+	unix_sysctl_register(&init_net);
 out:
 	return rc;
 }
@@ -2183,7 +2183,7 @@ out:
 static void __exit af_unix_exit(void)
 {
 	sock_unregister(PF_UNIX);
-	unix_sysctl_unregister();
+	unix_sysctl_unregister(&init_net);
 	proto_unregister(&unix_proto);
 	unregister_pernet_subsys(&unix_net_ops);
 }
diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c
index eb0bd57ebada..b2e0407e1b7b 100644
--- a/net/unix/sysctl_net_unix.c
+++ b/net/unix/sysctl_net_unix.c
@@ -48,12 +48,13 @@ static ctl_table unix_root_table[] = {
 
 static struct ctl_table_header * unix_sysctl_header;
 
-void unix_sysctl_register(void)
+int unix_sysctl_register(struct net *net)
 {
 	unix_sysctl_header = register_sysctl_table(unix_root_table);
+	return unix_sysctl_header == NULL ? -ENOMEM : 0;
 }
 
-void unix_sysctl_unregister(void)
+void unix_sysctl_unregister(struct net *net)
 {
 	unregister_sysctl_table(unix_sysctl_header);
 }
-- 
cgit v1.2.3


From d392e49756a99ba51577d265d6db83e3f01c6ac8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 1 Dec 2007 23:44:15 +1100
Subject: [UNIX]: Move the sysctl_unix_max_dgram_qlen

This will make all the sub-namespaces always use the
default value (10) and leave the tuning via sysctl
to the init namespace only.

Per-namespace tuning is coming.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 3 +++
 net/unix/af_unix.c          | 6 +++---
 net/unix/sysctl_net_unix.c  | 2 +-
 3 files changed, 7 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 235214c4c231..b0cf07519b81 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -38,6 +38,9 @@ struct net {
 	/* List of all packet sockets. */
 	rwlock_t		packet_sklist_lock;
 	struct hlist_head	packet_sklist;
+
+	/* unix sockets */
+	int			sysctl_unix_max_dgram_qlen;
 };
 
 #ifdef CONFIG_NET
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index a0aa6d3c2c08..73620d61762b 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -117,8 +117,6 @@
 #include <net/checksum.h>
 #include <linux/security.h>
 
-int sysctl_unix_max_dgram_qlen __read_mostly = 10;
-
 static struct hlist_head unix_socket_table[UNIX_HASH_SIZE + 1];
 static DEFINE_SPINLOCK(unix_table_lock);
 static atomic_t unix_nr_socks = ATOMIC_INIT(0);
@@ -594,7 +592,7 @@ static struct sock * unix_create1(struct net *net, struct socket *sock)
 				&af_unix_sk_receive_queue_lock_key);
 
 	sk->sk_write_space	= unix_write_space;
-	sk->sk_max_ack_backlog	= sysctl_unix_max_dgram_qlen;
+	sk->sk_max_ack_backlog	= net->sysctl_unix_max_dgram_qlen;
 	sk->sk_destruct		= unix_sock_destructor;
 	u	  = unix_sk(sk);
 	u->dentry = NULL;
@@ -2140,6 +2138,8 @@ static int unix_net_init(struct net *net)
 {
 	int error = -ENOMEM;
 
+	net->sysctl_unix_max_dgram_qlen = 10;
+
 #ifdef CONFIG_PROC_FS
 	if (!proc_net_fops_create(net, "unix", 0, &unix_seq_fops))
 		goto out;
diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c
index b2e0407e1b7b..c46cec04d23f 100644
--- a/net/unix/sysctl_net_unix.c
+++ b/net/unix/sysctl_net_unix.c
@@ -18,7 +18,7 @@ static ctl_table unix_table[] = {
 	{
 		.ctl_name	= NET_UNIX_MAX_DGRAM_QLEN,
 		.procname	= "max_dgram_qlen",
-		.data		= &sysctl_unix_max_dgram_qlen,
+		.data		= &init_net.sysctl_unix_max_dgram_qlen,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
-- 
cgit v1.2.3


From 1597fbc0faf88c42edb3fe42e92e53b83c3f120e Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 1 Dec 2007 23:51:01 +1100
Subject: [UNIX]: Make the unix sysctl tables per-namespace

This is the core.

 * add the ctl_table_header on the struct net;
 * make the unix_sysctl_register and _unregister clone the table;
 * moves calls to them into per-net init and exit callbacks;
 * move the .data pointer in the proper place.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  3 +++
 net/unix/af_unix.c          |  9 ++++++---
 net/unix/sysctl_net_unix.c  | 27 ++++++++++++++++++++++-----
 3 files changed, 31 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index b0cf07519b81..f97b2a4469ae 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -11,6 +11,8 @@
 struct proc_dir_entry;
 struct net_device;
 struct sock;
+struct ctl_table_header;
+
 struct net {
 	atomic_t		count;		/* To decided when the network
 						 *  namespace should be freed.
@@ -41,6 +43,7 @@ struct net {
 
 	/* unix sockets */
 	int			sysctl_unix_max_dgram_qlen;
+	struct ctl_table_header	*unix_ctl;
 };
 
 #ifdef CONFIG_NET
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 73620d61762b..b8a2189fb5c0 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2139,10 +2139,14 @@ static int unix_net_init(struct net *net)
 	int error = -ENOMEM;
 
 	net->sysctl_unix_max_dgram_qlen = 10;
+	if (unix_sysctl_register(net))
+		goto out;
 
 #ifdef CONFIG_PROC_FS
-	if (!proc_net_fops_create(net, "unix", 0, &unix_seq_fops))
+	if (!proc_net_fops_create(net, "unix", 0, &unix_seq_fops)) {
+		unix_sysctl_unregister(net);
 		goto out;
+	}
 #endif
 	error = 0;
 out:
@@ -2151,6 +2155,7 @@ out:
 
 static void unix_net_exit(struct net *net)
 {
+	unix_sysctl_unregister(net);
 	proc_net_remove(net, "unix");
 }
 
@@ -2175,7 +2180,6 @@ static int __init af_unix_init(void)
 
 	sock_register(&unix_family_ops);
 	register_pernet_subsys(&unix_net_ops);
-	unix_sysctl_register(&init_net);
 out:
 	return rc;
 }
@@ -2183,7 +2187,6 @@ out:
 static void __exit af_unix_exit(void)
 {
 	sock_unregister(PF_UNIX);
-	unix_sysctl_unregister(&init_net);
 	proto_unregister(&unix_proto);
 	unregister_pernet_subsys(&unix_net_ops);
 }
diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c
index 43dd356730fd..553ef6a487d0 100644
--- a/net/unix/sysctl_net_unix.c
+++ b/net/unix/sysctl_net_unix.c
@@ -32,16 +32,33 @@ static struct ctl_path unix_path[] = {
 	{ },
 };
 
-static struct ctl_table_header * unix_sysctl_header;
-
 int unix_sysctl_register(struct net *net)
 {
-	unix_sysctl_header = register_sysctl_paths(unix_path, unix_table);
-	return unix_sysctl_header == NULL ? -ENOMEM : 0;
+	struct ctl_table *table;
+
+	table = kmemdup(unix_table, sizeof(unix_table), GFP_KERNEL);
+	if (table == NULL)
+		goto err_alloc;
+
+	table[0].data = &net->sysctl_unix_max_dgram_qlen;
+	net->unix_ctl = register_net_sysctl_table(net, unix_path, table);
+	if (net->unix_ctl == NULL)
+		goto err_reg;
+
+	return 0;
+
+err_reg:
+	kfree(table);
+err_alloc:
+	return -ENOMEM;
 }
 
 void unix_sysctl_unregister(struct net *net)
 {
-	unregister_sysctl_table(unix_sysctl_header);
+	struct ctl_table *table;
+
+	table = net->unix_ctl->ctl_table_arg;
+	unregister_sysctl_table(net->unix_ctl);
+	kfree(table);
 }
 
-- 
cgit v1.2.3


From e38bad4766a110b61fa6038f10be16ced8c6cc38 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 28 Nov 2007 10:55:32 +0100
Subject: mac80211: make ieee80211_iterate_active_interfaces not need rtnl

Interface iteration in mac80211 can be done without holding any
locks because I converted it to RCU. Initially, I thought this
wouldn't be needed for ieee80211_iterate_active_interfaces but
it's turning out that multi-BSS AP support can be much simpler
in a driver if ieee80211_iterate_active_interfaces can be called
without holding locks. This converts it to use RCU, it adds a
requirement that the callback it invokes cannot sleep.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h | 3 +--
 net/mac80211/util.c    | 7 ++++---
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 1470e1b886f0..4cc5b1c8a4f1 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1411,10 +1411,9 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw);
  *
  * This function iterates over the interfaces associated with a given
  * hardware that are currently active and calls the callback for them.
- * Must be called under RTNL.
  *
  * @hw: the hardware struct of which the interfaces should be iterated over
- * @iterator: the iterator function to call
+ * @iterator: the iterator function to call, cannot sleep
  * @data: first argument of the iterator function
  */
 void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 5227c16f113e..7b278e9aa1a4 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -482,10 +482,9 @@ void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata;
 
-	ASSERT_RTNL();
+	rcu_read_lock();
 
-	/* we hold the RTNL here so can safely walk the list */
-	list_for_each_entry(sdata, &local->interfaces, list) {
+	list_for_each_entry_rcu(sdata, &local->interfaces, list) {
 		switch (sdata->type) {
 		case IEEE80211_IF_TYPE_INVALID:
 		case IEEE80211_IF_TYPE_MNTR:
@@ -503,5 +502,7 @@ void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
 			iterator(data, sdata->dev->dev_addr,
 				 sdata->dev->ifindex);
 	}
+
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces);
-- 
cgit v1.2.3


From 10816d40f2e9500057cb46d7608a362a1d10bb9b Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 26 Nov 2007 16:14:30 +0200
Subject: mac80211: adding 802.11n HT framework definitions

New structures:
 - ieee80211_ht_info: describing STA's HT capabilities
 - ieee80211_ht_bss_info: describing BSS's HT characteristics
Changed structures:
 - ieee80211_hw_mode: now also holds PHY HT capabilities for each HW mode
 - ieee80211_conf: ht_conf holds current self HT configuration
                   ht_bss_conf holds current BSS HT configuration
 - flag IEEE80211_CONF_SUPPORT_HT_MODE added to indicate if HT use is
   desired
 - sta_info: now also holds Peer's HT capabilities

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h  | 56 +++++++++++++++++++++++++++++++++++++++++++------
 net/mac80211/sta_info.h |  3 +++
 2 files changed, 53 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 4cc5b1c8a4f1..79e96455438a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -138,18 +138,55 @@ enum ieee80211_phymode {
 	NUM_IEEE80211_MODES
 };
 
+/**
+ * struct ieee80211_ht_info - describing STA's HT capabilities
+ *
+ * This structure describes most essential parameters needed
+ * to describe 802.11n HT capabilities for an STA.
+ *
+ * @ht_supported: is HT supported by STA, 0: no, 1: yes
+ * @cap: HT capabilities map as described in 802.11n spec
+ * @ampdu_factor: Maximum A-MPDU length factor
+ * @ampdu_density: Minimum A-MPDU spacing
+ * @supp_mcs_set: Supported MCS set as described in 802.11n spec
+ */
+struct ieee80211_ht_info {
+	u8 ht_supported;
+	u16 cap; /* use IEEE80211_HT_CAP_ */
+	u8 ampdu_factor;
+	u8 ampdu_density;
+	u8 supp_mcs_set[16];
+};
+
+/**
+ * struct ieee80211_ht_bss_info - describing BSS's HT characteristics
+ *
+ * This structure describes most essential parameters needed
+ * to describe 802.11n HT characteristics in a BSS
+ *
+ * @primary_channel: channel number of primery channel
+ * @bss_cap: 802.11n's general BSS capabilities (e.g. channel width)
+ * @bss_op_mode: 802.11n's BSS operation modes (e.g. HT protection)
+ */
+struct ieee80211_ht_bss_info {
+	u8 primary_channel;
+	u8 bss_cap;  /* use IEEE80211_HT_IE_CHA_ */
+	u8 bss_op_mode; /* use IEEE80211_HT_IE_ */
+};
+
 /**
  * struct ieee80211_hw_mode - PHY mode definition
  *
  * This structure describes the capabilities supported by the device
  * in a single PHY mode.
  *
+ * @list: internal
+ * @channels: pointer to array of supported channels
+ * @rates: pointer to array of supported bitrates
  * @mode: the PHY mode for this definition
  * @num_channels: number of supported channels
- * @channels: pointer to array of supported channels
  * @num_rates: number of supported bitrates
- * @rates: pointer to array of supported bitrates
- * @list: internal
+ * @ht_info: PHY's 802.11n HT abilities for this mode
  */
 struct ieee80211_hw_mode {
 	struct list_head list;
@@ -158,6 +195,7 @@ struct ieee80211_hw_mode {
 	enum ieee80211_phymode mode;
 	int num_channels;
 	int num_rates;
+	struct ieee80211_ht_info ht_info;
 };
 
 /**
@@ -406,11 +444,12 @@ struct ieee80211_tx_status {
  *
  * @IEEE80211_CONF_SHORT_SLOT_TIME: use 802.11g short slot time
  * @IEEE80211_CONF_RADIOTAP: add radiotap header at receive time (if supported)
- *
+ * @IEEE80211_CONF_SUPPORT_HT_MODE: use 802.11n HT capabilities (if supported)
  */
 enum ieee80211_conf_flags {
-	IEEE80211_CONF_SHORT_SLOT_TIME	= 1<<0,
-	IEEE80211_CONF_RADIOTAP		= 1<<1,
+	IEEE80211_CONF_SHORT_SLOT_TIME	= (1<<0),
+	IEEE80211_CONF_RADIOTAP		= (1<<1),
+	IEEE80211_CONF_SUPPORT_HT_MODE	= (1<<2),
 };
 
 /**
@@ -434,6 +473,8 @@ enum ieee80211_conf_flags {
  * @antenna_sel_tx: transmit antenna selection, 0: default/diversity,
  *	1/2: antenna 0/1
  * @antenna_sel_rx: receive antenna selection, like @antenna_sel_tx
+ * @ht_conf: describes current self configuration of 802.11n HT capabilies
+ * @ht_bss_conf: describes current BSS configuration of 802.11n HT parameters
  */
 struct ieee80211_conf {
 	int channel;			/* IEEE 802.11 channel number */
@@ -452,6 +493,9 @@ struct ieee80211_conf {
 	u8 antenna_max;
 	u8 antenna_sel_tx;
 	u8 antenna_sel_rx;
+
+	struct ieee80211_ht_info ht_conf;
+	struct ieee80211_ht_bss_info ht_bss_conf;
 };
 
 /**
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 8f7ebe41c024..e1a4ac1d48d7 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -99,6 +99,9 @@ struct sta_info {
 
 	u16 listen_interval;
 
+	struct ieee80211_ht_info ht_info; /* 802.11n HT capabilities
+					     of this STA */
+
 #ifdef CONFIG_MAC80211_DEBUGFS
 	struct sta_info_debugfsdentries {
 		struct dentry *dir;
-- 
cgit v1.2.3


From d3c990fb26b78f60614885d9ecaf7b7686b7b098 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 26 Nov 2007 16:14:34 +0200
Subject: mac80211: adding 802.11n configuration flows

This patch configures the 802.11n mode of operation
internally in ieee80211_conf structure and in the low-level
driver as well (through op conf_ht).
It does not include AP configuration flows.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h       |  3 +++
 net/mac80211/ieee80211.c     | 51 ++++++++++++++++++++++++++++++++++++++++++++
 net/mac80211/ieee80211_i.h   |  3 +++
 net/mac80211/ieee80211_sta.c | 32 +++++++++++++++++++++++++++
 4 files changed, 89 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 79e96455438a..0d67b331ee72 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1041,6 +1041,8 @@ enum ieee80211_erp_change_flags {
  * @tx_last_beacon: Determine whether the last IBSS beacon was sent by us.
  *	This is needed only for IBSS mode and the result of this function is
  *	used to determine whether to reply to Probe Requests.
+ *
+ * @conf_ht: Configures low level driver with 802.11n HT data. Must be atomic.
  */
 struct ieee80211_ops {
 	int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb,
@@ -1086,6 +1088,7 @@ struct ieee80211_ops {
 			     struct sk_buff *skb,
 			     struct ieee80211_tx_control *control);
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
+	int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
 };
 
 /**
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 4f8b6653e364..d6a97a68a62e 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -34,6 +34,8 @@
 #include "debugfs.h"
 #include "debugfs_netdev.h"
 
+#define SUPP_MCS_SET_LEN 16
+
 /*
  * For seeing transmitted packets on monitor interfaces
  * we have a radiotap header too.
@@ -563,6 +565,55 @@ int ieee80211_hw_config(struct ieee80211_local *local)
 	return ret;
 }
 
+/**
+ * ieee80211_hw_config_ht should be used only after legacy configuration
+ * has been determined, as ht configuration depends upon the hardware's
+ * HT abilities for a _specific_ band.
+ */
+int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
+			   struct ieee80211_ht_info *req_ht_cap,
+			   struct ieee80211_ht_bss_info *req_bss_cap)
+{
+	struct ieee80211_conf *conf = &local->hw.conf;
+	struct ieee80211_hw_mode *mode = conf->mode;
+	int i;
+
+	/* HT is not supported */
+	if (!mode->ht_info.ht_supported) {
+		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
+		return -EOPNOTSUPP;
+	}
+
+	/* disable HT */
+	if (!enable_ht) {
+		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
+	} else {
+		conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
+		conf->ht_conf.cap = req_ht_cap->cap & mode->ht_info.cap;
+		conf->ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
+		conf->ht_conf.cap |=
+			mode->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+		conf->ht_bss_conf.primary_channel =
+			req_bss_cap->primary_channel;
+		conf->ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
+		conf->ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
+		for (i = 0; i < SUPP_MCS_SET_LEN; i++)
+			conf->ht_conf.supp_mcs_set[i] =
+				mode->ht_info.supp_mcs_set[i] &
+				  req_ht_cap->supp_mcs_set[i];
+
+		/* In STA mode, this gives us indication
+		 * to the AP's mode of operation */
+		conf->ht_conf.ht_supported = 1;
+		conf->ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
+		conf->ht_conf.ampdu_density = req_ht_cap->ampdu_density;
+	}
+
+	local->ops->conf_ht(local_to_hw(local), &local->hw.conf);
+
+	return 0;
+}
+
 void ieee80211_erp_info_change_notify(struct net_device *dev, u8 changes)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 66b6cf3a62ac..2dcb9425fe8f 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -708,6 +708,9 @@ int ieee80211_if_update_wds(struct net_device *dev, u8 *remote_addr);
 void ieee80211_if_setup(struct net_device *dev);
 struct ieee80211_rate *ieee80211_get_rate(struct ieee80211_local *local,
 					  int phymode, int hwrate);
+int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
+			   struct ieee80211_ht_info *req_ht_cap,
+			   struct ieee80211_ht_bss_info *req_bss_cap);
 
 /* ieee80211_ioctl.c */
 extern const struct iw_handler_def ieee80211_iw_handler_def;
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 1d553d78b227..87830c04a1ca 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -1437,6 +1437,19 @@ static void ieee80211_rx_mgmt_assoc_resp(struct net_device *dev,
 	}
 	sta->supp_rates = rates;
 
+	if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param &&
+	    local->ops->conf_ht) {
+		struct ieee80211_ht_bss_info bss_info;
+
+		ieee80211_ht_cap_ie_to_ht_info(
+				(struct ieee80211_ht_cap *)
+				elems.ht_cap_elem, &sta->ht_info);
+		ieee80211_ht_addt_info_ie_to_ht_bss_info(
+				(struct ieee80211_ht_addt_info *)
+				elems.ht_info_elem, &bss_info);
+		ieee80211_hw_config_ht(local, 1, &sta->ht_info, &bss_info);
+	}
+
 	rate_control_rate_init(sta, local);
 
 	if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) {
@@ -1859,6 +1872,8 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 	struct ieee80211_if_sta *ifsta;
 	size_t baselen;
 	struct ieee802_11_elems elems;
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_conf *conf = &local->hw.conf;
 
 	ieee80211_rx_bss_info(dev, mgmt, len, rx_status, 1);
 
@@ -1881,6 +1896,23 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 	if (elems.erp_info && elems.erp_info_len >= 1)
 		ieee80211_handle_erp_ie(dev, elems.erp_info[0]);
 
+	if (elems.ht_cap_elem && elems.ht_info_elem &&
+	    elems.wmm_param && local->ops->conf_ht &&
+	    conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) {
+		struct ieee80211_ht_bss_info bss_info;
+
+		ieee80211_ht_addt_info_ie_to_ht_bss_info(
+				(struct ieee80211_ht_addt_info *)
+				elems.ht_info_elem, &bss_info);
+		/* check if AP changed bss inforamation */
+		if ((conf->ht_bss_conf.primary_channel !=
+		     bss_info.primary_channel) ||
+		    (conf->ht_bss_conf.bss_cap != bss_info.bss_cap) ||
+		    (conf->ht_bss_conf.bss_op_mode != bss_info.bss_op_mode))
+			ieee80211_hw_config_ht(local, 1, &conf->ht_conf,
+						&bss_info);
+	}
+
 	if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) {
 		ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param,
 					 elems.wmm_param_len);
-- 
cgit v1.2.3


From c3a05c6050a339c92e49fae0ba77dbba0d41fd99 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Sun, 2 Dec 2007 00:47:59 +0200
Subject: [TCP]: Cong.ctrl modules: remove unused good_ack from cong_avoid
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h        | 4 ++--
 net/ipv4/tcp_bic.c       | 3 +--
 net/ipv4/tcp_cong.c      | 2 +-
 net/ipv4/tcp_cubic.c     | 3 +--
 net/ipv4/tcp_highspeed.c | 3 +--
 net/ipv4/tcp_htcp.c      | 3 +--
 net/ipv4/tcp_hybla.c     | 5 ++---
 net/ipv4/tcp_illinois.c  | 3 +--
 net/ipv4/tcp_input.c     | 9 ++++-----
 net/ipv4/tcp_lp.c        | 4 ++--
 net/ipv4/tcp_scalable.c  | 3 +--
 net/ipv4/tcp_vegas.c     | 7 +++----
 net/ipv4/tcp_veno.c      | 7 +++----
 net/ipv4/tcp_yeah.c      | 3 +--
 14 files changed, 24 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7e583261f3a0..cdd0050b97a4 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -652,7 +652,7 @@ struct tcp_congestion_ops {
 	/* lower bound for congestion window (optional) */
 	u32 (*min_cwnd)(const struct sock *sk);
 	/* do new cwnd calculation (required) */
-	void (*cong_avoid)(struct sock *sk, u32 ack, u32 in_flight, int good_ack);
+	void (*cong_avoid)(struct sock *sk, u32 ack, u32 in_flight);
 	/* call before changing ca_state (optional) */
 	void (*set_state)(struct sock *sk, u8 new_state);
 	/* call when cwnd event occurs (optional) */
@@ -683,7 +683,7 @@ extern void tcp_slow_start(struct tcp_sock *tp);
 
 extern struct tcp_congestion_ops tcp_init_congestion_ops;
 extern u32 tcp_reno_ssthresh(struct sock *sk);
-extern void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight, int flag);
+extern void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight);
 extern u32 tcp_reno_min_cwnd(const struct sock *sk);
 extern struct tcp_congestion_ops tcp_reno;
 
diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c
index 5dba0fc8f579..5212ed9b0c98 100644
--- a/net/ipv4/tcp_bic.c
+++ b/net/ipv4/tcp_bic.c
@@ -136,8 +136,7 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
 		ca->cnt = 1;
 }
 
-static void bictcp_cong_avoid(struct sock *sk, u32 ack,
-			      u32 in_flight, int data_acked)
+static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bictcp *ca = inet_csk_ca(sk);
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 55fca1820c34..4451750b478c 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -324,7 +324,7 @@ EXPORT_SYMBOL_GPL(tcp_slow_start);
 /* This is Jacobson's slow start and congestion avoidance.
  * SIGCOMM '88, p. 328.
  */
-void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight, int flag)
+void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 80bd084a9f91..3aa0b23c1ea0 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -246,8 +246,7 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
 		ca->cnt = 1;
 }
 
-static void bictcp_cong_avoid(struct sock *sk, u32 ack,
-			      u32 in_flight, int data_acked)
+static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bictcp *ca = inet_csk_ca(sk);
diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c
index 14a073d8b60f..8b6caaf75bb9 100644
--- a/net/ipv4/tcp_highspeed.c
+++ b/net/ipv4/tcp_highspeed.c
@@ -109,8 +109,7 @@ static void hstcp_init(struct sock *sk)
 	tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128);
 }
 
-static void hstcp_cong_avoid(struct sock *sk, u32 adk,
-			     u32 in_flight, int data_acked)
+static void hstcp_cong_avoid(struct sock *sk, u32 adk, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct hstcp *ca = inet_csk_ca(sk);
diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c
index 5215691f2760..af99776146ff 100644
--- a/net/ipv4/tcp_htcp.c
+++ b/net/ipv4/tcp_htcp.c
@@ -225,8 +225,7 @@ static u32 htcp_recalc_ssthresh(struct sock *sk)
 	return max((tp->snd_cwnd * ca->beta) >> 7, 2U);
 }
 
-static void htcp_cong_avoid(struct sock *sk, u32 ack,
-			    u32 in_flight, int data_acked)
+static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct htcp *ca = inet_csk_ca(sk);
diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp_hybla.c
index b3e55cf56171..44618b675916 100644
--- a/net/ipv4/tcp_hybla.c
+++ b/net/ipv4/tcp_hybla.c
@@ -85,8 +85,7 @@ static inline u32 hybla_fraction(u32 odds)
  *     o Give cwnd a new value based on the model proposed
  *     o remember increments <1
  */
-static void hybla_cong_avoid(struct sock *sk, u32 ack,
-			    u32 in_flight, int flag)
+static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct hybla *ca = inet_csk_ca(sk);
@@ -103,7 +102,7 @@ static void hybla_cong_avoid(struct sock *sk, u32 ack,
 		return;
 
 	if (!ca->hybla_en)
-		return tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+		return tcp_reno_cong_avoid(sk, ack, in_flight);
 
 	if (ca->rho == 0)
 		hybla_recalc_param(sk);
diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c
index 5aa5f5496d6d..1eba160b72dc 100644
--- a/net/ipv4/tcp_illinois.c
+++ b/net/ipv4/tcp_illinois.c
@@ -256,8 +256,7 @@ static void tcp_illinois_state(struct sock *sk, u8 new_state)
 /*
  * Increase window in response to successful acknowledgment.
  */
-static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack,
-				    u32 in_flight, int flag)
+static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct illinois *ca = inet_csk_ca(sk);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index d313dea361a6..cb441188870c 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2710,11 +2710,10 @@ static inline void tcp_ack_update_rtt(struct sock *sk, const int flag,
 		tcp_ack_no_tstamp(sk, seq_rtt, flag);
 }
 
-static void tcp_cong_avoid(struct sock *sk, u32 ack,
-			   u32 in_flight, int good)
+static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	const struct inet_connection_sock *icsk = inet_csk(sk);
-	icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight, good);
+	icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight);
 	tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp;
 }
 
@@ -3238,11 +3237,11 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 		/* Advance CWND, if state allows this. */
 		if ((flag & FLAG_DATA_ACKED) && !frto_cwnd &&
 		    tcp_may_raise_cwnd(sk, flag))
-			tcp_cong_avoid(sk, ack, prior_in_flight, 0);
+			tcp_cong_avoid(sk, ack, prior_in_flight);
 		tcp_fastretrans_alert(sk, prior_packets - tp->packets_out, flag);
 	} else {
 		if ((flag & FLAG_DATA_ACKED) && !frto_cwnd)
-			tcp_cong_avoid(sk, ack, prior_in_flight, 1);
+			tcp_cong_avoid(sk, ack, prior_in_flight);
 	}
 
 	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP))
diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c
index e7f5ef92cbd8..ce3c41ff50b2 100644
--- a/net/ipv4/tcp_lp.c
+++ b/net/ipv4/tcp_lp.c
@@ -115,12 +115,12 @@ static void tcp_lp_init(struct sock *sk)
  * Will only call newReno CA when away from inference.
  * From TCP-LP's paper, this will be handled in additive increasement.
  */
-static void tcp_lp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight, int flag)
+static void tcp_lp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct lp *lp = inet_csk_ca(sk);
 
 	if (!(lp->flag & LP_WITHIN_INF))
-		tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+		tcp_reno_cong_avoid(sk, ack, in_flight);
 }
 
 /**
diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c
index be27a33a1c68..2747ec7bfb63 100644
--- a/net/ipv4/tcp_scalable.c
+++ b/net/ipv4/tcp_scalable.c
@@ -15,8 +15,7 @@
 #define TCP_SCALABLE_AI_CNT	50U
 #define TCP_SCALABLE_MD_SCALE	3
 
-static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack,
-				    u32 in_flight, int flag)
+static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c
index 007304e99842..be24d6ee34bd 100644
--- a/net/ipv4/tcp_vegas.c
+++ b/net/ipv4/tcp_vegas.c
@@ -162,14 +162,13 @@ void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event)
 }
 EXPORT_SYMBOL_GPL(tcp_vegas_cwnd_event);
 
-static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack,
-				 u32 in_flight, int flag)
+static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct vegas *vegas = inet_csk_ca(sk);
 
 	if (!vegas->doing_vegas_now)
-		return tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+		return tcp_reno_cong_avoid(sk, ack, in_flight);
 
 	/* The key players are v_beg_snd_una and v_beg_snd_nxt.
 	 *
@@ -228,7 +227,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack,
 			/* We don't have enough RTT samples to do the Vegas
 			 * calculation, so we'll behave like Reno.
 			 */
-			tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+			tcp_reno_cong_avoid(sk, ack, in_flight);
 		} else {
 			u32 rtt, target_cwnd, diff;
 
diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c
index 8fb2aee0b1a4..d16689e98516 100644
--- a/net/ipv4/tcp_veno.c
+++ b/net/ipv4/tcp_veno.c
@@ -114,14 +114,13 @@ static void tcp_veno_cwnd_event(struct sock *sk, enum tcp_ca_event event)
 		tcp_veno_init(sk);
 }
 
-static void tcp_veno_cong_avoid(struct sock *sk, u32 ack,
-				u32 in_flight, int flag)
+static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct veno *veno = inet_csk_ca(sk);
 
 	if (!veno->doing_veno_now)
-		return tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+		return tcp_reno_cong_avoid(sk, ack, in_flight);
 
 	/* limited by applications */
 	if (!tcp_is_cwnd_limited(sk, in_flight))
@@ -132,7 +131,7 @@ static void tcp_veno_cong_avoid(struct sock *sk, u32 ack,
 		/* We don't have enough rtt samples to do the Veno
 		 * calculation, so we'll behave like Reno.
 		 */
-		tcp_reno_cong_avoid(sk, ack, in_flight, flag);
+		tcp_reno_cong_avoid(sk, ack, in_flight);
 	} else {
 		u32 rtt, target_cwnd;
 
diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c
index c107fba7430e..e03b10183a8b 100644
--- a/net/ipv4/tcp_yeah.c
+++ b/net/ipv4/tcp_yeah.c
@@ -69,8 +69,7 @@ static void tcp_yeah_pkts_acked(struct sock *sk, u32 pkts_acked, s32 rtt_us)
 	tcp_vegas_pkts_acked(sk, pkts_acked, rtt_us);
 }
 
-static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack,
-				u32 in_flight, int flag)
+static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct yeah *yeah = inet_csk_ca(sk);
-- 
cgit v1.2.3


From 234b68607006f3721679e900809ccb99e8bfb10c Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Sun, 2 Dec 2007 00:48:02 +0200
Subject: [TCP]: Add tcp_for_write_queue_from_safe and use it in mtu_probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     | 5 +++++
 net/ipv4/tcp_output.c | 8 ++++----
 2 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index cdd0050b97a4..6e392babda4a 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1229,6 +1229,11 @@ static inline struct sk_buff *tcp_write_queue_next(struct sock *sk, struct sk_bu
 		for (; (skb != (struct sk_buff *)&(sk)->sk_write_queue);\
 		     skb = skb->next)
 
+#define tcp_for_write_queue_from_safe(skb, tmp, sk)			\
+		for (tmp = skb->next;					\
+		     (skb != (struct sk_buff *)&(sk)->sk_write_queue);	\
+		     skb = tmp, tmp = skb->next)
+
 static inline struct sk_buff *tcp_send_head(struct sock *sk)
 {
 	return sk->sk_send_head;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b41176f380d7..7d8583a15d02 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1350,9 +1350,7 @@ static int tcp_mtu_probe(struct sock *sk)
 	tcp_insert_write_queue_before(nskb, skb, sk);
 
 	len = 0;
-	while (len < probe_size) {
-		next = tcp_write_queue_next(sk, skb);
-
+	tcp_for_write_queue_from_safe(skb, next, sk) {
 		copy = min_t(int, skb->len, probe_size - len);
 		if (nskb->ip_summed)
 			skb_copy_bits(skb, 0, skb_put(nskb, copy), copy);
@@ -1381,7 +1379,9 @@ static int tcp_mtu_probe(struct sock *sk)
 		}
 
 		len += copy;
-		skb = next;
+
+		if (len >= probe_size)
+			break;
 	}
 	tcp_init_tso_segs(sk, nskb, nskb->len);
 
-- 
cgit v1.2.3


From 6859d49475d4f32abe640372117e4b687906e6b6 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Sun, 2 Dec 2007 00:48:06 +0200
Subject: [TCP]: Abstract tp->highest_sack accessing & point to next skb
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pointing to the next skb is necessary to avoid referencing
already SACKed skbs which will soon be on a separate list.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     | 35 ++++++++++++++++++++++++++++++++++-
 net/ipv4/tcp_input.c  | 27 +++++++++++++++------------
 net/ipv4/tcp_output.c | 11 +++--------
 3 files changed, 52 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 6e392babda4a..5ec1cacca8a1 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1267,8 +1267,12 @@ static inline void tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb
 	__tcp_add_write_queue_tail(sk, skb);
 
 	/* Queue it, remembering where we must start sending. */
-	if (sk->sk_send_head == NULL)
+	if (sk->sk_send_head == NULL) {
 		sk->sk_send_head = skb;
+
+		if (tcp_sk(sk)->highest_sack == NULL)
+			tcp_sk(sk)->highest_sack = skb;
+	}
 }
 
 static inline void __tcp_add_write_queue_head(struct sock *sk, struct sk_buff *skb)
@@ -1318,9 +1322,38 @@ static inline u32 tcp_highest_sack_seq(struct tcp_sock *tp)
 {
 	if (!tp->sacked_out)
 		return tp->snd_una;
+
+	if (tp->highest_sack == NULL)
+		return tp->snd_nxt;
+
 	return TCP_SKB_CB(tp->highest_sack)->seq;
 }
 
+static inline void tcp_advance_highest_sack(struct sock *sk, struct sk_buff *skb)
+{
+	tcp_sk(sk)->highest_sack = tcp_skb_is_last(sk, skb) ? NULL :
+						tcp_write_queue_next(sk, skb);
+}
+
+static inline struct sk_buff *tcp_highest_sack(struct sock *sk)
+{
+	return tcp_sk(sk)->highest_sack;
+}
+
+static inline void tcp_highest_sack_reset(struct sock *sk)
+{
+	tcp_sk(sk)->highest_sack = tcp_write_queue_head(sk);
+}
+
+/* Called when old skb is about to be deleted (to be combined with new skb) */
+static inline void tcp_highest_sack_combine(struct sock *sk,
+					    struct sk_buff *old,
+					    struct sk_buff *new)
+{
+	if (tcp_sk(sk)->sacked_out && (old == tcp_sk(sk)->highest_sack))
+		tcp_sk(sk)->highest_sack = new;
+}
+
 /* /proc */
 enum tcp_seq_states {
 	TCP_SEQ_STATE_LISTENING,
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 263c536def5c..bc2d5f70966e 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1125,7 +1125,7 @@ static void tcp_mark_lost_retrans(struct sock *sk)
 	struct sk_buff *skb;
 	int cnt = 0;
 	u32 new_low_seq = tp->snd_nxt;
-	u32 received_upto = TCP_SKB_CB(tp->highest_sack)->end_seq;
+	u32 received_upto = tcp_highest_sack_seq(tp);
 
 	if (!tcp_is_fack(tp) || !tp->retrans_out ||
 	    !after(received_upto, tp->lost_retrans_low) ||
@@ -1236,9 +1236,10 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb,
 	return in_sack;
 }
 
-static int tcp_sacktag_one(struct sk_buff *skb, struct tcp_sock *tp,
+static int tcp_sacktag_one(struct sk_buff *skb, struct sock *sk,
 			   int *reord, int dup_sack, int fack_count)
 {
+	struct tcp_sock *tp = tcp_sk(sk);
 	u8 sacked = TCP_SKB_CB(skb)->sacked;
 	int flag = 0;
 
@@ -1307,8 +1308,8 @@ static int tcp_sacktag_one(struct sk_buff *skb, struct tcp_sock *tp,
 		if (fack_count > tp->fackets_out)
 			tp->fackets_out = fack_count;
 
-		if (after(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
-			tp->highest_sack = skb;
+		if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
+			tcp_advance_highest_sack(sk, skb);
 	}
 
 	/* D-SACK. We can detect redundant retransmission in S|R and plain R
@@ -1330,8 +1331,6 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk,
 					int dup_sack_in, int *fack_count,
 					int *reord, int *flag)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
-
 	tcp_for_write_queue_from(skb, sk) {
 		int in_sack = 0;
 		int dup_sack = dup_sack_in;
@@ -1358,7 +1357,7 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk,
 			break;
 
 		if (in_sack)
-			*flag |= tcp_sacktag_one(skb, tp, reord, dup_sack, *fack_count);
+			*flag |= tcp_sacktag_one(skb, sk, reord, dup_sack, *fack_count);
 
 		*fack_count += tcp_skb_pcount(skb);
 	}
@@ -1429,7 +1428,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	if (!tp->sacked_out) {
 		if (WARN_ON(tp->fackets_out))
 			tp->fackets_out = 0;
-		tp->highest_sack = tcp_write_queue_head(sk);
+		tcp_highest_sack_reset(sk);
 	}
 
 	found_dup_sack = tcp_check_dsack(tp, ack_skb, sp_wire,
@@ -1552,9 +1551,11 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 						       &fack_count, &reord, &flag);
 
 			/* ...tail remains todo... */
-			if (TCP_SKB_CB(tp->highest_sack)->end_seq == cache->end_seq) {
+			if (tcp_highest_sack_seq(tp) == cache->end_seq) {
 				/* ...but better entrypoint exists! */
-				skb = tcp_write_queue_next(sk, tp->highest_sack);
+				skb = tcp_highest_sack(sk);
+				if (skb == NULL)
+					break;
 				fack_count = tp->fackets_out;
 				cache++;
 				goto walk;
@@ -1566,8 +1567,10 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 			continue;
 		}
 
-		if (tp->sacked_out && !before(start_seq, tcp_highest_sack_seq(tp))) {
-			skb = tcp_write_queue_next(sk, tp->highest_sack);
+		if (!before(start_seq, tcp_highest_sack_seq(tp))) {
+			skb = tcp_highest_sack(sk);
+			if (skb == NULL)
+				break;
 			fack_count = tp->fackets_out;
 		}
 		skb = tcp_sacktag_skip(skb, sk, start_seq);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 7d8583a15d02..9a985b55e7d8 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -667,7 +667,7 @@ static void tcp_adjust_fackets_out(struct sock *sk, struct sk_buff *skb,
 	if (!tp->sacked_out || tcp_is_reno(tp))
 		return;
 
-	if (!before(tcp_highest_sack_seq(tp), TCP_SKB_CB(skb)->seq))
+	if (after(tcp_highest_sack_seq(tp), TCP_SKB_CB(skb)->seq))
 		tp->fackets_out -= decr;
 }
 
@@ -711,9 +711,6 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 	TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq;
 	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq;
 
-	if (tcp_is_sack(tp) && tp->sacked_out && (skb == tp->highest_sack))
-		tp->highest_sack = buff;
-
 	/* PSH and FIN should only be set in the second packet. */
 	flags = TCP_SKB_CB(skb)->flags;
 	TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
@@ -1707,9 +1704,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 		BUG_ON(tcp_skb_pcount(skb) != 1 ||
 		       tcp_skb_pcount(next_skb) != 1);
 
-		if (WARN_ON(tcp_is_sack(tp) && tp->sacked_out &&
-		    (next_skb == tp->highest_sack)))
-			return;
+		tcp_highest_sack_combine(sk, next_skb, skb);
 
 		/* Ok.	We will be able to collapse the packet. */
 		tcp_unlink_write_queue(next_skb, sk);
@@ -2019,7 +2014,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk)
 			break;
 		tp->forward_skb_hint = skb;
 
-		if (after(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
+		if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp)))
 			break;
 
 		if (tcp_packets_in_flight(tp) >= tp->snd_cwnd)
-- 
cgit v1.2.3


From 2fcb45b6b87914f072314e5b5d9c196f45984683 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 3 Dec 2007 22:54:12 -0800
Subject: [IPSEC]: Use the correct family for input state lookup

When merging the input paths of IPsec I accidentally left a hard-coded
AF_INET for the state lookup call.  This broke IPv6 obviously.  This
patch fixes by getting the input callers to specify the family through
skb->cb.

Credit goes to Kazunori Miyazawa for diagnosing this and providing an
initial patch.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 1 +
 net/ipv4/xfrm4_input.c | 1 +
 net/ipv6/xfrm6_input.c | 1 +
 net/xfrm/xfrm_input.c  | 5 ++++-
 4 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index ecc02e9bde32..182f10b29d88 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -534,6 +534,7 @@ struct xfrm_spi_skb_cb {
 	} header;
 
 	unsigned int daddroff;
+	unsigned int family;
 };
 
 #define XFRM_SPI_SKB_CB(__skb) ((struct xfrm_spi_skb_cb *)&((__skb)->cb[0]))
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index 0c377a66b8b5..33f990d56c91 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -39,6 +39,7 @@ drop:
 int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 		    int encap_type)
 {
+	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
 	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
 	return xfrm_input(skb, nexthdr, spi, encap_type);
 }
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index e2c3efd2579d..74f3aacebb5e 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -23,6 +23,7 @@ int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb)
 
 int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi)
 {
+	XFRM_SPI_SKB_CB(skb)->family = AF_INET6;
 	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
 	return xfrm_input(skb, nexthdr, spi, 0);
 }
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 96f42c1d2e8e..8b2b1b59133e 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -102,6 +102,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	__be32 seq;
 	struct xfrm_state *x;
 	xfrm_address_t *daddr;
+	unsigned int family;
 	int decaps = 0;
 	int async = 0;
 
@@ -127,6 +128,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 
 	daddr = (xfrm_address_t *)(skb_network_header(skb) +
 				   XFRM_SPI_SKB_CB(skb)->daddroff);
+	family = XFRM_SPI_SKB_CB(skb)->family;
 
 	seq = 0;
 	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0)
@@ -136,7 +138,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		if (skb->sp->len == XFRM_MAX_DEPTH)
 			goto drop;
 
-		x = xfrm_state_lookup(daddr, spi, nexthdr, AF_INET);
+		x = xfrm_state_lookup(daddr, spi, nexthdr, family);
 		if (x == NULL)
 			goto drop;
 
@@ -198,6 +200,7 @@ resume:
 		 * transport mode so the outer address is identical.
 		 */
 		daddr = &x->id.daddr;
+		family = x->outer_mode->afinfo->family;
 
 		err = xfrm_parse_spi(skb, nexthdr, &spi, &seq);
 		if (err < 0)
-- 
cgit v1.2.3


From 0eeb8ffcfeaa0d909ce39147f7b8fdd6cef1aacd Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 4 Dec 2007 01:15:45 -0800
Subject: [NET]: netns compilation speedup

This patch speedups compilation when net_namespace.h is changed.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_cls.h | 3 ++-
 include/net/sock.h    | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index f285de69c615..2eaf20480024 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -2,7 +2,6 @@
 #define __NET_PKT_CLS_H
 
 #include <linux/pkt_cls.h>
-#include <net/net_namespace.h>
 #include <net/sch_generic.h>
 #include <net/act_api.h>
 
@@ -336,6 +335,8 @@ static inline int tcf_valid_offset(const struct sk_buff *skb,
 }
 
 #ifdef CONFIG_NET_CLS_IND
+#include <net/net_namespace.h>
+
 static inline int
 tcf_change_indev(struct tcf_proto *tp, char *indev, struct rtattr *indev_tlv)
 {
diff --git a/include/net/sock.h b/include/net/sock.h
index bcbf8891dde4..800e73a62d8f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -57,7 +57,6 @@
 #include <asm/atomic.h>
 #include <net/dst.h>
 #include <net/checksum.h>
-#include <net/net_namespace.h>
 
 /*
  * This structure really needs to be cleaned up.
@@ -95,6 +94,7 @@ typedef struct {
 
 struct sock;
 struct proto;
+struct net;
 
 /**
  *	struct sock_common - minimal network layer representation of sockets
-- 
cgit v1.2.3


From 338e8a79262c3709e314fbcc7ca193323e534934 Mon Sep 17 00:00:00 2001
From: Sven Schnelle <svens@bitebene.org>
Date: Tue, 4 Dec 2007 23:21:50 -0800
Subject: [NETFILTER]: x_tables: add TCPOPTSTRIP target

Signed-off-by: Sven Schnelle <svens@bitebene.org>
Signed-off-by: Jan Engelhardt <jengelh@gmx.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_TCPOPTSTRIP.h |  13 +++
 net/netfilter/Kconfig                    |   8 ++
 net/netfilter/Makefile                   |   1 +
 net/netfilter/xt_TCPOPTSTRIP.c           | 147 +++++++++++++++++++++++++++++++
 4 files changed, 169 insertions(+)
 create mode 100644 include/linux/netfilter/xt_TCPOPTSTRIP.h
 create mode 100644 net/netfilter/xt_TCPOPTSTRIP.c

(limited to 'include')

diff --git a/include/linux/netfilter/xt_TCPOPTSTRIP.h b/include/linux/netfilter/xt_TCPOPTSTRIP.h
new file mode 100644
index 000000000000..2db543214ff5
--- /dev/null
+++ b/include/linux/netfilter/xt_TCPOPTSTRIP.h
@@ -0,0 +1,13 @@
+#ifndef _XT_TCPOPTSTRIP_H
+#define _XT_TCPOPTSTRIP_H
+
+#define tcpoptstrip_set_bit(bmap, idx) \
+	(bmap[(idx) >> 5] |= 1U << (idx & 31))
+#define tcpoptstrip_test_bit(bmap, idx) \
+	(((1U << (idx & 31)) & bmap[(idx) >> 5]) != 0)
+
+struct xt_tcpoptstrip_target_info {
+	u_int32_t strip_bmap[8];
+};
+
+#endif /* _XT_TCPOPTSTRIP_H */
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 21a9fcc03796..693f861a03b4 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -411,6 +411,14 @@ config NETFILTER_XT_TARGET_TCPMSS
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config NETFILTER_XT_TARGET_TCPOPTSTRIP
+	tristate '"TCPOPTSTRIP" target support (EXPERIMENTAL)'
+	depends on EXPERIMENTAL && NETFILTER_XTABLES
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	help
+	  This option adds a "TCPOPTSTRIP" target, which allows you to strip
+	  TCP options from TCP packets.
+
 config NETFILTER_XT_MATCH_COMMENT
 	tristate  '"comment" match support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index ad0e36ebea3d..7763dea17be4 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP) += xt_TCPOPTSTRIP.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_TRACE) += xt_TRACE.o
 
 # matches
diff --git a/net/netfilter/xt_TCPOPTSTRIP.c b/net/netfilter/xt_TCPOPTSTRIP.c
new file mode 100644
index 000000000000..43d6ac224f5e
--- /dev/null
+++ b/net/netfilter/xt_TCPOPTSTRIP.c
@@ -0,0 +1,147 @@
+/*
+ * A module for stripping a specific TCP option from TCP packets.
+ *
+ * Copyright (C) 2007 Sven Schnelle <svens@bitebene.org>
+ * Copyright © CC Computer Consultants GmbH, 2007
+ * Contact: Jan Engelhardt <jengelh@computergmbh.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
+#include <net/ipv6.h>
+#include <net/tcp.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_TCPOPTSTRIP.h>
+
+static inline unsigned int optlen(const u_int8_t *opt, unsigned int offset)
+{
+	/* Beware zero-length options: make finite progress */
+	if (opt[offset] <= TCPOPT_NOP || opt[offset+1] == 0)
+		return 1;
+	else
+		return opt[offset+1];
+}
+
+static unsigned int
+tcpoptstrip_mangle_packet(struct sk_buff *skb,
+			  const struct xt_tcpoptstrip_target_info *info,
+			  unsigned int tcphoff, unsigned int minlen)
+{
+	unsigned int optl, i, j;
+	struct tcphdr *tcph;
+	u_int16_t n, o;
+	u_int8_t *opt;
+
+	if (!skb_make_writable(skb, skb->len))
+		return NF_DROP;
+
+	tcph = (struct tcphdr *)(skb_network_header(skb) + tcphoff);
+	opt  = (u_int8_t *)tcph;
+
+	/*
+	 * Walk through all TCP options - if we find some option to remove,
+	 * set all octets to %TCPOPT_NOP and adjust checksum.
+	 */
+	for (i = sizeof(struct tcphdr); i < tcp_hdrlen(skb); i += optl) {
+		optl = optlen(opt, i);
+
+		if (i + optl > tcp_hdrlen(skb))
+			break;
+
+		if (!tcpoptstrip_test_bit(info->strip_bmap, opt[i]))
+			continue;
+
+		for (j = 0; j < optl; ++j) {
+			o = opt[i+j];
+			n = TCPOPT_NOP;
+			if ((i + j) % 2 == 0) {
+				o <<= 8;
+				n <<= 8;
+			}
+			inet_proto_csum_replace2(&tcph->check, skb, htons(o),
+						 htons(n), 0);
+		}
+		memset(opt + i, TCPOPT_NOP, optl);
+	}
+
+	return XT_CONTINUE;
+}
+
+static unsigned int
+tcpoptstrip_tg4(struct sk_buff *skb, const struct net_device *in,
+		const struct net_device *out, unsigned int hooknum,
+		const struct xt_target *target, const void *targinfo)
+{
+	return tcpoptstrip_mangle_packet(skb, targinfo, ip_hdrlen(skb),
+	       sizeof(struct iphdr) + sizeof(struct tcphdr));
+}
+
+#if defined(CONFIG_IP6_NF_MANGLE) || defined(CONFIG_IP6_NF_MANGLE_MODULE)
+static unsigned int
+tcpoptstrip_tg6(struct sk_buff *skb, const struct net_device *in,
+		const struct net_device *out, unsigned int hooknum,
+		const struct xt_target *target, const void *targinfo)
+{
+	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
+	unsigned int tcphoff;
+	u_int8_t nexthdr;
+
+	nexthdr = ipv6h->nexthdr;
+	tcphoff = ipv6_skip_exthdr(skb, sizeof(*ipv6h), &nexthdr);
+	if (tcphoff < 0)
+		return NF_DROP;
+
+	return tcpoptstrip_mangle_packet(skb, targinfo, tcphoff,
+	       sizeof(*ipv6h) + sizeof(struct tcphdr));
+}
+#endif
+
+static struct xt_target tcpoptstrip_tg_reg[] __read_mostly = {
+	{
+		.name       = "TCPOPTSTRIP",
+		.family     = AF_INET,
+		.table      = "mangle",
+		.proto      = IPPROTO_TCP,
+		.target     = tcpoptstrip_tg4,
+		.targetsize = sizeof(struct xt_tcpoptstrip_target_info),
+		.me         = THIS_MODULE,
+	},
+#if defined(CONFIG_IP6_NF_MANGLE) || defined(CONFIG_IP6_NF_MANGLE_MODULE)
+	{
+		.name       = "TCPOPTSTRIP",
+		.family     = AF_INET6,
+		.table      = "mangle",
+		.proto      = IPPROTO_TCP,
+		.target     = tcpoptstrip_tg6,
+		.targetsize = sizeof(struct xt_tcpoptstrip_target_info),
+		.me         = THIS_MODULE,
+	},
+#endif
+};
+
+static int __init tcpoptstrip_tg_init(void)
+{
+	return xt_register_targets(tcpoptstrip_tg_reg,
+				   ARRAY_SIZE(tcpoptstrip_tg_reg));
+}
+
+static void __exit tcpoptstrip_tg_exit(void)
+{
+	xt_unregister_targets(tcpoptstrip_tg_reg,
+			      ARRAY_SIZE(tcpoptstrip_tg_reg));
+}
+
+module_init(tcpoptstrip_tg_init);
+module_exit(tcpoptstrip_tg_exit);
+MODULE_AUTHOR("Sven Schnelle <svens@bitebene.org>, Jan Engelhardt <jengelh@computergmbh.de>");
+MODULE_DESCRIPTION("netfilter \"TCPOPTSTRIP\" target module");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_TCPOPTSTRIP");
+MODULE_ALIAS("ip6t_TCPOPTSTRIP");
-- 
cgit v1.2.3


From 259d4e41f3ec25f22169daece42729f597b89f9a Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 4 Dec 2007 23:24:56 -0800
Subject: [NETFILTER]: x_tables: struct xt_table_info diet

Instead of using a big array of NR_CPUS entries, we can compute the size
needed at runtime, using nr_cpu_ids

This should save some ram (especially on David's machines where NR_CPUS=4096 :
32 KB can be saved per table, and 64KB for dynamically allocated ones (because
of slab/slub alignements) )

In particular, the 'bootstrap' tables are not any more static (in data
section) but on stack as their size is now very small.

This also should reduce the size used on stack in compat functions
(get_info() declares an automatic variable, that could be bigger than kernel
stack size for big NR_CPUS)

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h |  5 ++++-
 net/ipv4/netfilter/arp_tables.c    |  5 ++---
 net/ipv4/netfilter/ip_tables.c     | 24 +++++++++---------------
 net/ipv6/netfilter/ip6_tables.c    |  5 ++---
 net/netfilter/x_tables.c           |  2 +-
 5 files changed, 18 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 9657c4ee70fc..e305f2d0d4d1 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -269,9 +269,12 @@ struct xt_table_info
 	unsigned int underflow[NF_INET_NUMHOOKS];
 
 	/* ipt_entry tables: one per CPU */
-	char *entries[NR_CPUS];
+	/* Note : this field MUST be the last one, see XT_TABLE_INFO_SZ */
+	char *entries[1];
 };
 
+#define XT_TABLE_INFO_SZ (offsetof(struct xt_table_info, entries) \
+			  + nr_cpu_ids * sizeof(char *))
 extern int xt_register_target(struct xt_target *target);
 extern void xt_unregister_target(struct xt_target *target);
 extern int xt_register_targets(struct xt_target *target, unsigned int n);
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 2909c92ecd99..a21722d5c9fa 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -811,8 +811,7 @@ static int do_replace(void __user *user, unsigned int len)
 		return -ENOPROTOOPT;
 
 	/* overflow check */
-	if (tmp.size >= (INT_MAX - sizeof(struct xt_table_info)) / NR_CPUS -
-			SMP_CACHE_BYTES)
+	if (tmp.size >= INT_MAX / num_possible_cpus())
 		return -ENOMEM;
 	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
 		return -ENOMEM;
@@ -1090,7 +1089,7 @@ int arpt_register_table(struct arpt_table *table,
 {
 	int ret;
 	struct xt_table_info *newinfo;
-	static struct xt_table_info bootstrap
+	struct xt_table_info bootstrap
 		= { 0, 0, 0, { 0 }, { 0 }, { } };
 	void *loc_cpu_entry;
 
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index ca23c63ced37..87d369244bd9 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1090,7 +1090,8 @@ compat_calc_match(struct ipt_entry_match *m, int * size)
 	return 0;
 }
 
-static int compat_calc_entry(struct ipt_entry *e, struct xt_table_info *info,
+static int compat_calc_entry(struct ipt_entry *e,
+		const struct xt_table_info *info,
 		void *base, struct xt_table_info *newinfo)
 {
 	struct ipt_entry_target *t;
@@ -1118,22 +1119,17 @@ static int compat_calc_entry(struct ipt_entry *e, struct xt_table_info *info,
 	return 0;
 }
 
-static int compat_table_info(struct xt_table_info *info,
+static int compat_table_info(const struct xt_table_info *info,
 		struct xt_table_info *newinfo)
 {
 	void *loc_cpu_entry;
-	int i;
 
 	if (!newinfo || !info)
 		return -EINVAL;
 
-	memset(newinfo, 0, sizeof(struct xt_table_info));
-	newinfo->size = info->size;
-	newinfo->number = info->number;
-	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
-		newinfo->hook_entry[i] = info->hook_entry[i];
-		newinfo->underflow[i] = info->underflow[i];
-	}
+	/* we dont care about newinfo->entries[] */
+	memcpy(newinfo, info, offsetof(struct xt_table_info, entries));
+	newinfo->initial_entries = 0;
 	loc_cpu_entry = info->entries[raw_smp_processor_id()];
 	return IPT_ENTRY_ITERATE(loc_cpu_entry, info->size,
 			compat_calc_entry, info, loc_cpu_entry, newinfo);
@@ -1327,8 +1323,7 @@ do_replace(void __user *user, unsigned int len)
 		return -ENOPROTOOPT;
 
 	/* overflow check */
-	if (tmp.size >= (INT_MAX - sizeof(struct xt_table_info)) / NR_CPUS -
-			SMP_CACHE_BYTES)
+	if (tmp.size >= INT_MAX / num_possible_cpus())
 		return -ENOMEM;
 	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
 		return -ENOMEM;
@@ -1868,8 +1863,7 @@ compat_do_replace(void __user *user, unsigned int len)
 		return -ENOPROTOOPT;
 
 	/* overflow check */
-	if (tmp.size >= (INT_MAX - sizeof(struct xt_table_info)) / NR_CPUS -
-			SMP_CACHE_BYTES)
+	if (tmp.size >= INT_MAX / num_possible_cpus())
 		return -ENOMEM;
 	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
 		return -ENOMEM;
@@ -2126,7 +2120,7 @@ int ipt_register_table(struct xt_table *table, const struct ipt_replace *repl)
 {
 	int ret;
 	struct xt_table_info *newinfo;
-	static struct xt_table_info bootstrap
+	struct xt_table_info bootstrap
 		= { 0, 0, 0, { 0 }, { 0 }, { } };
 	void *loc_cpu_entry;
 
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index e1e87eff4686..e60c1b4b1ec8 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -1042,8 +1042,7 @@ do_replace(void __user *user, unsigned int len)
 		return -EFAULT;
 
 	/* overflow check */
-	if (tmp.size >= (INT_MAX - sizeof(struct xt_table_info)) / NR_CPUS -
-			SMP_CACHE_BYTES)
+	if (tmp.size >= INT_MAX / num_possible_cpus())
 		return -ENOMEM;
 	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
 		return -ENOMEM;
@@ -1339,7 +1338,7 @@ int ip6t_register_table(struct xt_table *table,
 {
 	int ret;
 	struct xt_table_info *newinfo;
-	static struct xt_table_info bootstrap
+	struct xt_table_info bootstrap
 		= { 0, 0, 0, { 0 }, { 0 }, { } };
 	void *loc_cpu_entry;
 
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index b6160e41eb1c..07bb465d9518 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -499,7 +499,7 @@ struct xt_table_info *xt_alloc_table_info(unsigned int size)
 	if ((SMP_ALIGN(size) >> PAGE_SHIFT) + 2 > num_physpages)
 		return NULL;
 
-	newinfo = kzalloc(sizeof(struct xt_table_info), GFP_KERNEL);
+	newinfo = kzalloc(XT_TABLE_INFO_SZ, GFP_KERNEL);
 	if (!newinfo)
 		return NULL;
 
-- 
cgit v1.2.3


From 0265ab44bacc1a1e0e3f5873d8ca2d5a29e33db2 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Tue, 4 Dec 2007 23:27:38 -0800
Subject: [NETFILTER]: merge ipt_owner/ip6t_owner in xt_owner

xt_owner merges ipt_owner and ip6t_owner, and adds a flag to match
on socket (non-)existence.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild     |   1 +
 include/linux/netfilter/xt_owner.h |  16 +++
 net/ipv4/netfilter/Kconfig         |   9 --
 net/ipv4/netfilter/Makefile        |   1 -
 net/ipv4/netfilter/ipt_owner.c     |  87 ---------------
 net/ipv6/netfilter/Kconfig         |   9 --
 net/ipv6/netfilter/Makefile        |   1 -
 net/ipv6/netfilter/ip6t_owner.c    |  87 ---------------
 net/netfilter/Kconfig              |   8 ++
 net/netfilter/Makefile             |   1 +
 net/netfilter/xt_owner.c           | 211 +++++++++++++++++++++++++++++++++++++
 11 files changed, 237 insertions(+), 194 deletions(-)
 create mode 100644 include/linux/netfilter/xt_owner.h
 delete mode 100644 net/ipv4/netfilter/ipt_owner.c
 delete mode 100644 net/ipv6/netfilter/ip6t_owner.c
 create mode 100644 net/netfilter/xt_owner.c

(limited to 'include')

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index b87e83a5e070..1e6900278318 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -26,6 +26,7 @@ header-y += xt_limit.h
 header-y += xt_mac.h
 header-y += xt_mark.h
 header-y += xt_multiport.h
+header-y += xt_owner.h
 header-y += xt_pkttype.h
 header-y += xt_policy.h
 header-y += xt_realm.h
diff --git a/include/linux/netfilter/xt_owner.h b/include/linux/netfilter/xt_owner.h
new file mode 100644
index 000000000000..eacd34efebd5
--- /dev/null
+++ b/include/linux/netfilter/xt_owner.h
@@ -0,0 +1,16 @@
+#ifndef _XT_OWNER_MATCH_H
+#define _XT_OWNER_MATCH_H
+
+enum {
+	XT_OWNER_UID    = 1 << 0,
+	XT_OWNER_GID    = 1 << 1,
+	XT_OWNER_SOCKET = 1 << 2,
+};
+
+struct xt_owner_match_info {
+	u_int32_t uid;
+	u_int32_t gid;
+	u_int8_t match, invert;
+};
+
+#endif /* _XT_OWNER_MATCH_H */
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 9aca9c55687c..6c563d908c78 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -111,15 +111,6 @@ config IP_NF_MATCH_TTL
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-config IP_NF_MATCH_OWNER
-	tristate "Owner match support"
-	depends on IP_NF_IPTABLES
-	help
-	  Packet owner matching allows you to match locally-generated packets
-	  based on who created them: the user, group, process or session.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 config IP_NF_MATCH_ADDRTYPE
 	tristate  'address type match support'
 	depends on IP_NF_IPTABLES
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index 7456833d6ade..42199e93b86c 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -45,7 +45,6 @@ obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
 obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o
 obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o
 obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o
-obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o
 obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o
 obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o
 obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o
diff --git a/net/ipv4/netfilter/ipt_owner.c b/net/ipv4/netfilter/ipt_owner.c
deleted file mode 100644
index 4f1aa897d4b0..000000000000
--- a/net/ipv4/netfilter/ipt_owner.c
+++ /dev/null
@@ -1,87 +0,0 @@
-/* Kernel module to match various things tied to sockets associated with
-   locally generated outgoing packets. */
-
-/* (C) 2000 Marc Boucher <marc@mbsi.ca>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/module.h>
-#include <linux/skbuff.h>
-#include <linux/file.h>
-#include <linux/rcupdate.h>
-#include <net/sock.h>
-
-#include <linux/netfilter_ipv4/ipt_owner.h>
-#include <linux/netfilter/x_tables.h>
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>");
-MODULE_DESCRIPTION("iptables owner match");
-
-static bool
-owner_mt(const struct sk_buff *skb, const struct net_device *in,
-         const struct net_device *out, const struct xt_match *match,
-         const void *matchinfo, int offset, unsigned int protoff,
-         bool *hotdrop)
-{
-	const struct ipt_owner_info *info = matchinfo;
-
-	if (!skb->sk || !skb->sk->sk_socket || !skb->sk->sk_socket->file)
-		return false;
-
-	if(info->match & IPT_OWNER_UID) {
-		if ((skb->sk->sk_socket->file->f_uid != info->uid) ^
-		    !!(info->invert & IPT_OWNER_UID))
-			return false;
-	}
-
-	if(info->match & IPT_OWNER_GID) {
-		if ((skb->sk->sk_socket->file->f_gid != info->gid) ^
-		    !!(info->invert & IPT_OWNER_GID))
-			return false;
-	}
-
-	return true;
-}
-
-static bool
-owner_mt_check(const char *tablename, const void *ip,
-               const struct xt_match *match, void *matchinfo,
-               unsigned int hook_mask)
-{
-	const struct ipt_owner_info *info = matchinfo;
-
-	if (info->match & (IPT_OWNER_PID|IPT_OWNER_SID|IPT_OWNER_COMM)) {
-		printk("ipt_owner: pid, sid and command matching "
-		       "not supported anymore\n");
-		return false;
-	}
-	return true;
-}
-
-static struct xt_match owner_mt_reg __read_mostly = {
-	.name		= "owner",
-	.family		= AF_INET,
-	.match		= owner_mt,
-	.matchsize	= sizeof(struct ipt_owner_info),
-	.hooks		= (1 << NF_INET_LOCAL_OUT) |
-			  (1 << NF_INET_POST_ROUTING),
-	.checkentry	= owner_mt_check,
-	.me		= THIS_MODULE,
-};
-
-static int __init owner_mt_init(void)
-{
-	return xt_register_match(&owner_mt_reg);
-}
-
-static void __exit owner_mt_exit(void)
-{
-	xt_unregister_match(&owner_mt_reg);
-}
-
-module_init(owner_mt_init);
-module_exit(owner_mt_exit);
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 838b8ddee8c0..30d48529d988 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -89,15 +89,6 @@ config IP6_NF_MATCH_HL
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-config IP6_NF_MATCH_OWNER
-	tristate "Owner match support"
-	depends on IP6_NF_IPTABLES
-	help
-	  Packet owner matching allows you to match locally-generated packets
-	  based on who created them: the user, group, process or session.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 config IP6_NF_MATCH_IPV6HEADER
 	tristate "IPv6 Extension Headers Match"
 	depends on IP6_NF_IPTABLES
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index e789ec44d23b..fbf2c14ed887 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -23,7 +23,6 @@ obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o
 obj-$(CONFIG_IP6_NF_MATCH_IPV6HEADER) += ip6t_ipv6header.o
 obj-$(CONFIG_IP6_NF_MATCH_MH) += ip6t_mh.o
 obj-$(CONFIG_IP6_NF_MATCH_OPTS) += ip6t_hbh.o
-obj-$(CONFIG_IP6_NF_MATCH_OWNER) += ip6t_owner.o
 obj-$(CONFIG_IP6_NF_MATCH_RT) += ip6t_rt.o
 
 # targets
diff --git a/net/ipv6/netfilter/ip6t_owner.c b/net/ipv6/netfilter/ip6t_owner.c
deleted file mode 100644
index 6a52ed985166..000000000000
--- a/net/ipv6/netfilter/ip6t_owner.c
+++ /dev/null
@@ -1,87 +0,0 @@
-/* Kernel module to match various things tied to sockets associated with
-   locally generated outgoing packets. */
-
-/* (C) 2000-2001 Marc Boucher <marc@mbsi.ca>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/module.h>
-#include <linux/skbuff.h>
-#include <linux/file.h>
-#include <linux/rcupdate.h>
-#include <net/sock.h>
-
-#include <linux/netfilter_ipv6/ip6t_owner.h>
-#include <linux/netfilter_ipv6/ip6_tables.h>
-#include <linux/netfilter/x_tables.h>
-
-MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>");
-MODULE_DESCRIPTION("IP6 tables owner matching module");
-MODULE_LICENSE("GPL");
-
-
-static bool
-owner_mt6(const struct sk_buff *skb, const struct net_device *in,
-          const struct net_device *out, const struct xt_match *match,
-          const void *matchinfo, int offset, unsigned int protoff,
-          bool *hotdrop)
-{
-	const struct ip6t_owner_info *info = matchinfo;
-
-	if (!skb->sk || !skb->sk->sk_socket || !skb->sk->sk_socket->file)
-		return false;
-
-	if (info->match & IP6T_OWNER_UID)
-		if ((skb->sk->sk_socket->file->f_uid != info->uid) ^
-		    !!(info->invert & IP6T_OWNER_UID))
-			return false;
-
-	if (info->match & IP6T_OWNER_GID)
-		if ((skb->sk->sk_socket->file->f_gid != info->gid) ^
-		    !!(info->invert & IP6T_OWNER_GID))
-			return false;
-
-	return true;
-}
-
-static bool
-owner_mt6_check(const char *tablename, const void *ip,
-                const struct xt_match *match, void *matchinfo,
-                unsigned int hook_mask)
-{
-	const struct ip6t_owner_info *info = matchinfo;
-
-	if (info->match & (IP6T_OWNER_PID | IP6T_OWNER_SID)) {
-		printk("ipt_owner: pid and sid matching "
-		       "not supported anymore\n");
-		return false;
-	}
-	return true;
-}
-
-static struct xt_match owner_mt6_reg __read_mostly = {
-	.name		= "owner",
-	.family		= AF_INET6,
-	.match		= owner_mt6,
-	.matchsize	= sizeof(struct ip6t_owner_info),
-	.hooks		= (1 << NF_INET_LOCAL_OUT) |
-			  (1 << NF_INET_POST_ROUTING),
-	.checkentry	= owner_mt6_check,
-	.me		= THIS_MODULE,
-};
-
-static int __init owner_mt6_init(void)
-{
-	return xt_register_match(&owner_mt6_reg);
-}
-
-static void __exit owner_mt6_exit(void)
-{
-	xt_unregister_match(&owner_mt6_reg);
-}
-
-module_init(owner_mt6_init);
-module_exit(owner_mt6_exit);
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 693f861a03b4..4bc0552b75fd 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -554,6 +554,14 @@ config NETFILTER_XT_MATCH_MARK
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config NETFILTER_XT_MATCH_OWNER
+	tristate '"owner" match support'
+	depends on NETFILTER_XTABLES
+	---help---
+	Socket owner matching allows you to match locally-generated packets
+	based on who created the socket: the user or group. It is also
+	possible to check whether a socket actually exists.
+
 config NETFILTER_XT_MATCH_POLICY
 	tristate 'IPsec "policy" match support'
 	depends on NETFILTER_XTABLES && XFRM
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 7763dea17be4..28f59a35aeef 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -67,6 +67,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_LIMIT) += xt_limit.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_MAC) += xt_mac.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_MARK) += xt_mark.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_MULTIPORT) += xt_multiport.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_OWNER) += xt_owner.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_PKTTYPE) += xt_pkttype.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_POLICY) += xt_policy.o
diff --git a/net/netfilter/xt_owner.c b/net/netfilter/xt_owner.c
new file mode 100644
index 000000000000..4222fa2c1b14
--- /dev/null
+++ b/net/netfilter/xt_owner.c
@@ -0,0 +1,211 @@
+/*
+ * Kernel module to match various things tied to sockets associated with
+ * locally generated outgoing packets.
+ *
+ * (C) 2000 Marc Boucher <marc@mbsi.ca>
+ *
+ * Copyright © CC Computer Consultants GmbH, 2007
+ * Contact: <jengelh@computergmbh.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/file.h>
+#include <net/sock.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_owner.h>
+#include <linux/netfilter_ipv4/ipt_owner.h>
+#include <linux/netfilter_ipv6/ip6t_owner.h>
+
+static bool
+owner_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+            const struct net_device *out, const struct xt_match *match,
+            const void *matchinfo, int offset, unsigned int protoff,
+            bool *hotdrop)
+{
+	const struct ipt_owner_info *info = matchinfo;
+	const struct file *filp;
+
+	if (skb->sk == NULL || skb->sk->sk_socket == NULL)
+		return false;
+
+	filp = skb->sk->sk_socket->file;
+	if (filp == NULL)
+		return false;
+
+	if (info->match & IPT_OWNER_UID)
+		if ((filp->f_uid != info->uid) ^
+		    !!(info->invert & IPT_OWNER_UID))
+			return false;
+
+	if (info->match & IPT_OWNER_GID)
+		if ((filp->f_gid != info->gid) ^
+		    !!(info->invert & IPT_OWNER_GID))
+			return false;
+
+	return true;
+}
+
+static bool
+owner_mt6_v0(const struct sk_buff *skb, const struct net_device *in,
+             const struct net_device *out, const struct xt_match *match,
+             const void *matchinfo, int offset, unsigned int protoff,
+             bool *hotdrop)
+{
+	const struct ip6t_owner_info *info = matchinfo;
+	const struct file *filp;
+
+	if (skb->sk == NULL || skb->sk->sk_socket == NULL)
+		return false;
+
+	filp = skb->sk->sk_socket->file;
+	if (filp == NULL)
+		return false;
+
+	if (info->match & IP6T_OWNER_UID)
+		if ((filp->f_uid != info->uid) ^
+		    !!(info->invert & IP6T_OWNER_UID))
+			return false;
+
+	if (info->match & IP6T_OWNER_GID)
+		if ((filp->f_gid != info->gid) ^
+		    !!(info->invert & IP6T_OWNER_GID))
+			return false;
+
+	return true;
+}
+
+static bool
+owner_mt(const struct sk_buff *skb, const struct net_device *in,
+         const struct net_device *out, const struct xt_match *match,
+         const void *matchinfo, int offset, unsigned int protoff,
+         bool *hotdrop)
+{
+	const struct xt_owner_match_info *info = matchinfo;
+	const struct file *filp;
+
+	if (skb->sk == NULL || skb->sk->sk_socket == NULL)
+		return (info->match ^ info->invert) == 0;
+	else if (info->match & info->invert & XT_OWNER_SOCKET)
+		/*
+		 * Socket exists but user wanted ! --socket-exists.
+		 * (Single ampersands intended.)
+		 */
+		return false;
+
+	filp = skb->sk->sk_socket->file;
+	if (filp == NULL)
+		return ((info->match ^ info->invert) &
+		       (XT_OWNER_UID | XT_OWNER_GID)) == 0;
+
+	if (info->match & XT_OWNER_UID)
+		if ((filp->f_uid != info->uid) ^
+		    !!(info->invert & XT_OWNER_UID))
+			return false;
+
+	if (info->match & XT_OWNER_GID)
+		if ((filp->f_gid != info->gid) ^
+		    !!(info->invert & XT_OWNER_GID))
+			return false;
+
+	return true;
+}
+
+static bool
+owner_mt_check_v0(const char *tablename, const void *ip,
+                  const struct xt_match *match, void *matchinfo,
+                  unsigned int hook_mask)
+{
+	const struct ipt_owner_info *info = matchinfo;
+
+	if (info->match & (IPT_OWNER_PID | IPT_OWNER_SID | IPT_OWNER_COMM)) {
+		printk(KERN_WARNING KBUILD_MODNAME
+		       ": PID, SID and command matching is not "
+		       "supported anymore\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool
+owner_mt6_check_v0(const char *tablename, const void *ip,
+                   const struct xt_match *match, void *matchinfo,
+                   unsigned int hook_mask)
+{
+	const struct ip6t_owner_info *info = matchinfo;
+
+	if (info->match & (IP6T_OWNER_PID | IP6T_OWNER_SID)) {
+		printk(KERN_WARNING KBUILD_MODNAME
+		       ": PID and SID matching is not supported anymore\n");
+		return false;
+	}
+
+	return true;
+}
+
+static struct xt_match owner_mt_reg[] __read_mostly = {
+	{
+		.name       = "owner",
+		.revision   = 0,
+		.family     = AF_INET,
+		.match      = owner_mt_v0,
+		.matchsize  = sizeof(struct ipt_owner_info),
+		.checkentry = owner_mt_check_v0,
+		.hooks      = (1 << NF_INET_LOCAL_OUT) |
+		              (1 << NF_INET_POST_ROUTING),
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "owner",
+		.revision   = 0,
+		.family     = AF_INET6,
+		.match      = owner_mt6_v0,
+		.matchsize  = sizeof(struct ip6t_owner_info),
+		.checkentry = owner_mt6_check_v0,
+		.hooks      = (1 << NF_INET_LOCAL_OUT) |
+		              (1 << NF_INET_POST_ROUTING),
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "owner",
+		.revision   = 1,
+		.family     = AF_INET,
+		.match      = owner_mt,
+		.matchsize  = sizeof(struct xt_owner_match_info),
+		.hooks      = (1 << NF_INET_LOCAL_OUT) |
+		              (1 << NF_INET_POST_ROUTING),
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "owner",
+		.revision   = 1,
+		.family     = AF_INET6,
+		.match      = owner_mt,
+		.matchsize  = sizeof(struct xt_owner_match_info),
+		.hooks      = (1 << NF_INET_LOCAL_OUT) |
+		              (1 << NF_INET_POST_ROUTING),
+		.me         = THIS_MODULE,
+	},
+};
+
+static int __init owner_mt_init(void)
+{
+	return xt_register_matches(owner_mt_reg, ARRAY_SIZE(owner_mt_reg));
+}
+
+static void __exit owner_mt_exit(void)
+{
+	xt_unregister_matches(owner_mt_reg, ARRAY_SIZE(owner_mt_reg));
+}
+
+module_init(owner_mt_init);
+module_exit(owner_mt_exit);
+MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>");
+MODULE_DESCRIPTION("netfilter \"owner\" match module");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_owner");
+MODULE_ALIAS("ip6t_owner");
-- 
cgit v1.2.3


From 0553811612a6178365f3b062c30234913b218a96 Mon Sep 17 00:00:00 2001
From: Laszlo Attila Toth <panther@balabit.hu>
Date: Tue, 4 Dec 2007 23:28:46 -0800
Subject: [IPV4]: Add inet_dev_addr_type()

Address type search can be limited to an interface by
inet_dev_addr_type function.

Signed-off-by: Laszlo Attila Toth <panther@balabit.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h     |  1 +
 net/ipv4/fib_frontend.c | 21 +++++++++++++++++++--
 2 files changed, 20 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 59b0b19205a2..76b08c1affc0 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -118,6 +118,7 @@ extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mt
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
 extern unsigned		inet_addr_type(__be32 addr);
+extern unsigned		inet_dev_addr_type(const struct net_device *dev, __be32 addr);
 extern void		ip_rt_multicast_event(struct in_device *);
 extern int		ip_rt_ioctl(unsigned int cmd, void __user *arg);
 extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index e02aba5fa136..d90b42f3630c 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -155,7 +155,12 @@ out:
 	return dev;
 }
 
-unsigned inet_addr_type(__be32 addr)
+/*
+ * Find address type as if only "dev" was present in the system. If
+ * on_dev is NULL then all interfaces are taken into consideration.
+ */
+static inline unsigned __inet_dev_addr_type(const struct net_device *dev,
+					    __be32 addr)
 {
 	struct flowi		fl = { .nl_u = { .ip4_u = { .daddr = addr } } };
 	struct fib_result	res;
@@ -175,13 +180,24 @@ unsigned inet_addr_type(__be32 addr)
 	if (local_table) {
 		ret = RTN_UNICAST;
 		if (!local_table->tb_lookup(local_table, &fl, &res)) {
-			ret = res.type;
+			if (!dev || dev == res.fi->fib_dev)
+				ret = res.type;
 			fib_res_put(&res);
 		}
 	}
 	return ret;
 }
 
+unsigned int inet_addr_type(__be32 addr)
+{
+	return __inet_dev_addr_type(NULL, addr);
+}
+
+unsigned int inet_dev_addr_type(const struct net_device *dev, __be32 addr)
+{
+       return __inet_dev_addr_type(dev, addr);
+}
+
 /* Given (packet source, input interface) and optional (dst, oif, tos):
    - (main) check, that source is valid i.e. not broadcast or our local
      address.
@@ -940,4 +956,5 @@ void __init ip_fib_init(void)
 }
 
 EXPORT_SYMBOL(inet_addr_type);
+EXPORT_SYMBOL(inet_dev_addr_type);
 EXPORT_SYMBOL(ip_dev_find);
-- 
cgit v1.2.3


From e2cf5ecbea861ff05105bbd40f4f0d7823d9e213 Mon Sep 17 00:00:00 2001
From: Laszlo Attila Toth <panther@balabit.hu>
Date: Tue, 4 Dec 2007 23:30:18 -0800
Subject: [NETFILTER]: ipt_addrtype: limit address type checking to an
 interface

Addrtype match has a new revision (1), which lets address type checking
limited to the interface the current packet belongs to. Either incoming
or outgoing interface can be used depending on the current hook. In the
FORWARD hook two maches should be used if both interfaces have to be checked.
The new structure is ipt_addrtype_info_v1.

Revision 0 lets older userspace programs use the match as earlier.
ipt_addrtype_info is used.

Signed-off-by: Laszlo Attila Toth <panther@balabit.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ipt_addrtype.h |  14 ++++
 net/ipv4/netfilter/ipt_addrtype.c           | 104 +++++++++++++++++++++++-----
 2 files changed, 102 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ipt_addrtype.h b/include/linux/netfilter_ipv4/ipt_addrtype.h
index 166ed01a8122..446de6aef983 100644
--- a/include/linux/netfilter_ipv4/ipt_addrtype.h
+++ b/include/linux/netfilter_ipv4/ipt_addrtype.h
@@ -1,6 +1,20 @@
 #ifndef _IPT_ADDRTYPE_H
 #define _IPT_ADDRTYPE_H
 
+enum {
+	IPT_ADDRTYPE_INVERT_SOURCE	= 0x0001,
+	IPT_ADDRTYPE_INVERT_DEST	= 0x0002,
+	IPT_ADDRTYPE_LIMIT_IFACE_IN	= 0x0004,
+	IPT_ADDRTYPE_LIMIT_IFACE_OUT	= 0x0008,
+};
+
+struct ipt_addrtype_info_v1 {
+	u_int16_t	source;		/* source-type mask */
+	u_int16_t	dest;		/* dest-type mask */
+	u_int32_t	flags;
+};
+
+/* revision 0 */
 struct ipt_addrtype_info {
 	u_int16_t	source;		/* source-type mask */
 	u_int16_t	dest;		/* dest-type mask */
diff --git a/net/ipv4/netfilter/ipt_addrtype.c b/net/ipv4/netfilter/ipt_addrtype.c
index b75421c5e087..14394c6a3c29 100644
--- a/net/ipv4/netfilter/ipt_addrtype.c
+++ b/net/ipv4/netfilter/ipt_addrtype.c
@@ -2,6 +2,7 @@
  *  iptables module to match inet_addr_type() of an ip.
  *
  *  Copyright (c) 2004 Patrick McHardy <kaber@trash.net>
+ *  (C) 2007 Laszlo Attila Toth <panther@balabit.hu>
  *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License version 2 as
@@ -22,45 +23,116 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
 MODULE_DESCRIPTION("iptables addrtype match");
 
-static inline bool match_type(__be32 addr, u_int16_t mask)
+static inline bool match_type(const struct net_device *dev, __be32 addr,
+			      u_int16_t mask)
 {
-	return !!(mask & (1 << inet_addr_type(addr)));
+	return !!(mask & (1 << inet_dev_addr_type(dev, addr)));
 }
 
 static bool
-addrtype_mt(const struct sk_buff *skb, const struct net_device *in,
-            const struct net_device *out, const struct xt_match *match,
-            const void *matchinfo, int offset, unsigned int protoff,
-            bool *hotdrop)
+addrtype_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+	       const struct net_device *out, const struct xt_match *match,
+	       const void *matchinfo, int offset, unsigned int protoff,
+	       bool *hotdrop)
 {
 	const struct ipt_addrtype_info *info = matchinfo;
 	const struct iphdr *iph = ip_hdr(skb);
 	bool ret = true;
 
 	if (info->source)
-		ret &= match_type(iph->saddr, info->source)^info->invert_source;
+		ret &= match_type(NULL, iph->saddr, info->source) ^
+		       info->invert_source;
 	if (info->dest)
-		ret &= match_type(iph->daddr, info->dest)^info->invert_dest;
+		ret &= match_type(NULL, iph->daddr, info->dest) ^
+		       info->invert_dest;
 
 	return ret;
 }
 
-static struct xt_match addrtype_mt_reg __read_mostly = {
-	.name		= "addrtype",
-	.family		= AF_INET,
-	.match		= addrtype_mt,
-	.matchsize	= sizeof(struct ipt_addrtype_info),
-	.me		= THIS_MODULE
+static bool
+addrtype_mt_v1(const struct sk_buff *skb, const struct net_device *in,
+	       const struct net_device *out, const struct xt_match *match,
+	       const void *matchinfo, int offset, unsigned int protoff,
+	       bool *hotdrop)
+{
+	const struct ipt_addrtype_info_v1 *info = matchinfo;
+	const struct iphdr *iph = ip_hdr(skb);
+	const struct net_device *dev = NULL;
+	bool ret = true;
+
+	if (info->flags & IPT_ADDRTYPE_LIMIT_IFACE_IN)
+		dev = in;
+	else if (info->flags & IPT_ADDRTYPE_LIMIT_IFACE_OUT)
+		dev = out;
+
+	if (info->source)
+		ret &= match_type(dev, iph->saddr, info->source) ^
+		       (info->flags & IPT_ADDRTYPE_INVERT_SOURCE);
+	if (ret && info->dest)
+		ret &= match_type(dev, iph->daddr, info->dest) ^
+		       (info->flags & IPT_ADDRTYPE_INVERT_DEST);
+	return ret;
+}
+
+static bool
+addrtype_mt_checkentry_v1(const char *tablename, const void *ip_void,
+			  const struct xt_match *match, void *matchinfo,
+			  unsigned int hook_mask)
+{
+	struct ipt_addrtype_info_v1 *info = matchinfo;
+
+	if (info->flags & IPT_ADDRTYPE_LIMIT_IFACE_IN &&
+	    info->flags & IPT_ADDRTYPE_LIMIT_IFACE_OUT) {
+		printk(KERN_ERR "ipt_addrtype: both incoming and outgoing "
+				"interface limitation cannot be selected\n");
+		return false;
+	}
+
+	if (hook_mask & (1 << NF_INET_PRE_ROUTING | 1 << NF_INET_LOCAL_IN) &&
+	    info->flags & IPT_ADDRTYPE_LIMIT_IFACE_OUT) {
+		printk(KERN_ERR "ipt_addrtype: output interface limitation "
+				"not valid in PRE_ROUTING and INPUT\n");
+		return false;
+	}
+
+	if (hook_mask & (1 << NF_INET_POST_ROUTING | 1 << NF_INET_LOCAL_OUT) &&
+	    info->flags & IPT_ADDRTYPE_LIMIT_IFACE_IN) {
+		printk(KERN_ERR "ipt_addrtype: input interface limitation "
+				"not valid in POST_ROUTING and OUTPUT\n");
+		return false;
+	}
+
+	return true;
+}
+
+static struct xt_match addrtype_mt_reg[] __read_mostly = {
+	{
+		.name		= "addrtype",
+		.family		= AF_INET,
+		.match		= addrtype_mt_v0,
+		.matchsize	= sizeof(struct ipt_addrtype_info),
+		.me		= THIS_MODULE
+	},
+	{
+		.name		= "addrtype",
+		.family		= AF_INET,
+		.revision	= 1,
+		.match		= addrtype_mt_v1,
+		.checkentry	= addrtype_mt_checkentry_v1,
+		.matchsize	= sizeof(struct ipt_addrtype_info_v1),
+		.me		= THIS_MODULE
+	}
 };
 
 static int __init addrtype_mt_init(void)
 {
-	return xt_register_match(&addrtype_mt_reg);
+	return xt_register_matches(addrtype_mt_reg,
+				   ARRAY_SIZE(addrtype_mt_reg));
 }
 
 static void __exit addrtype_mt_exit(void)
 {
-	xt_unregister_match(&addrtype_mt_reg);
+	xt_unregister_matches(addrtype_mt_reg, ARRAY_SIZE(addrtype_mt_reg));
 }
 
 module_init(addrtype_mt_init);
-- 
cgit v1.2.3


From 3c3f486603438130b93b33acd92db0f994ec3e55 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Tue, 4 Dec 2007 23:37:29 -0800
Subject: [NET]: Constify include/net/dsfield.h

Constify include/net/dsfield.h

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dsfield.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/dsfield.h b/include/net/dsfield.h
index eb65bf2e2502..8a8d4e06900d 100644
--- a/include/net/dsfield.h
+++ b/include/net/dsfield.h
@@ -12,15 +12,15 @@
 #include <asm/byteorder.h>
 
 
-static inline __u8 ipv4_get_dsfield(struct iphdr *iph)
+static inline __u8 ipv4_get_dsfield(const struct iphdr *iph)
 {
 	return iph->tos;
 }
 
 
-static inline __u8 ipv6_get_dsfield(struct ipv6hdr *ipv6h)
+static inline __u8 ipv6_get_dsfield(const struct ipv6hdr *ipv6h)
 {
-	return ntohs(*(__be16 *) ipv6h) >> 4;
+	return ntohs(*(const __be16 *)ipv6h) >> 4;
 }
 
 
-- 
cgit v1.2.3


From f1095ab51d4297d4a84b64a65c71054183a73486 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Tue, 4 Dec 2007 23:38:30 -0800
Subject: [NETFILTER]: IPv6 capable xt_tos v1 match

Extends the xt_dscp match by xt_tos v1 to add support for selectively
matching any bit in the IPv4 TOS and IPv6 Priority fields. (ipt_tos
and xt_dscp only accepted a limited range of possible values.)

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_dscp.h |  6 ++++++
 net/netfilter/xt_dscp.c           | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 38 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_dscp.h b/include/linux/netfilter/xt_dscp.h
index 1da61e6acaf7..f49bc1a648dc 100644
--- a/include/linux/netfilter/xt_dscp.h
+++ b/include/linux/netfilter/xt_dscp.h
@@ -20,4 +20,10 @@ struct xt_dscp_info {
 	u_int8_t invert;
 };
 
+struct xt_tos_match_info {
+	u_int8_t tos_mask;
+	u_int8_t tos_value;
+	u_int8_t invert;
+};
+
 #endif /* _XT_DSCP_H */
diff --git a/net/netfilter/xt_dscp.c b/net/netfilter/xt_dscp.c
index 75b0df990d4f..834e43720315 100644
--- a/net/netfilter/xt_dscp.c
+++ b/net/netfilter/xt_dscp.c
@@ -23,6 +23,7 @@ MODULE_LICENSE("GPL");
 MODULE_ALIAS("ipt_dscp");
 MODULE_ALIAS("ip6t_dscp");
 MODULE_ALIAS("ipt_tos");
+MODULE_ALIAS("ip6t_tos");
 
 static bool
 dscp_mt(const struct sk_buff *skb, const struct net_device *in,
@@ -72,6 +73,21 @@ static bool tos_mt_v0(const struct sk_buff *skb, const struct net_device *in,
 	return (ip_hdr(skb)->tos == info->tos) ^ info->invert;
 }
 
+static bool tos_mt(const struct sk_buff *skb, const struct net_device *in,
+                   const struct net_device *out, const struct xt_match *match,
+                   const void *matchinfo, int offset, unsigned int protoff,
+                   bool *hotdrop)
+{
+	const struct xt_tos_match_info *info = matchinfo;
+
+	if (match->family == AF_INET)
+		return ((ip_hdr(skb)->tos & info->tos_mask) ==
+		       info->tos_value) ^ !!info->invert;
+	else
+		return ((ipv6_get_dsfield(ipv6_hdr(skb)) & info->tos_mask) ==
+		       info->tos_value) ^ !!info->invert;
+}
+
 static struct xt_match dscp_mt_reg[] __read_mostly = {
 	{
 		.name		= "dscp",
@@ -97,6 +113,22 @@ static struct xt_match dscp_mt_reg[] __read_mostly = {
 		.matchsize	= sizeof(struct ipt_tos_info),
 		.me		= THIS_MODULE,
 	},
+	{
+		.name		= "tos",
+		.revision	= 1,
+		.family		= AF_INET,
+		.match		= tos_mt,
+		.matchsize	= sizeof(struct xt_tos_match_info),
+		.me		= THIS_MODULE,
+	},
+	{
+		.name		= "tos",
+		.revision	= 1,
+		.family		= AF_INET6,
+		.match		= tos_mt,
+		.matchsize	= sizeof(struct xt_tos_match_info),
+		.me		= THIS_MODULE,
+	},
 };
 
 static int __init dscp_mt_init(void)
-- 
cgit v1.2.3


From 5c350e5a380333c64da8580fa134a2fd8e71fea4 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Tue, 4 Dec 2007 23:39:09 -0800
Subject: [NETFILTER]: IPv6 capable xt_TOS v1 target

Extends the xt_DSCP target by xt_TOS v1 to add support for selectively
setting and flipping any bit in the IPv4 TOS and IPv6 Priority fields.
(ipt_TOS and xt_DSCP only accepted a limited range of possible
values.)

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_DSCP.h |  5 ++++
 net/netfilter/Kconfig             |  2 +-
 net/netfilter/xt_DSCP.c           | 63 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 69 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_DSCP.h b/include/linux/netfilter/xt_DSCP.h
index 3c7c963997bd..14da1968e2c6 100644
--- a/include/linux/netfilter/xt_DSCP.h
+++ b/include/linux/netfilter/xt_DSCP.h
@@ -17,4 +17,9 @@ struct xt_DSCP_info {
 	u_int8_t dscp;
 };
 
+struct xt_tos_target_info {
+	u_int8_t tos_value;
+	u_int8_t tos_mask;
+};
+
 #endif /* _XT_DSCP_TARGET_H */
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 9c82d4cc86bf..7bde6315999a 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -304,7 +304,7 @@ config NETFILTER_XT_TARGET_DSCP
 
 	  It also adds the "TOS" target, which allows you to create rules in
 	  the "mangle" table which alter the Type Of Service field of an IPv4
-	  packet prior to routing.
+	  or the Priority field of an IPv6 packet, prior to routing.
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
diff --git a/net/netfilter/xt_DSCP.c b/net/netfilter/xt_DSCP.c
index 40a4f1d71916..fd7500ecadfd 100644
--- a/net/netfilter/xt_DSCP.c
+++ b/net/netfilter/xt_DSCP.c
@@ -26,6 +26,7 @@ MODULE_LICENSE("GPL");
 MODULE_ALIAS("ipt_DSCP");
 MODULE_ALIAS("ip6t_DSCP");
 MODULE_ALIAS("ipt_TOS");
+MODULE_ALIAS("ip6t_TOS");
 
 static unsigned int
 dscp_tg(struct sk_buff *skb, const struct net_device *in,
@@ -117,6 +118,50 @@ tos_tg_check_v0(const char *tablename, const void *e_void,
 	return true;
 }
 
+static unsigned int
+tos_tg(struct sk_buff *skb, const struct net_device *in,
+       const struct net_device *out, unsigned int hooknum,
+       const struct xt_target *target, const void *targinfo)
+{
+	const struct xt_tos_target_info *info = targinfo;
+	struct iphdr *iph = ip_hdr(skb);
+	u_int8_t orig, nv;
+
+	orig = ipv4_get_dsfield(iph);
+	nv   = (orig & info->tos_mask) ^ info->tos_value;
+
+	if (orig != nv) {
+		if (!skb_make_writable(skb, sizeof(struct iphdr)))
+			return NF_DROP;
+		iph = ip_hdr(skb);
+		ipv4_change_dsfield(iph, ~0, nv);
+	}
+
+	return XT_CONTINUE;
+}
+
+static unsigned int
+tos_tg6(struct sk_buff *skb, const struct net_device *in,
+        const struct net_device *out, unsigned int hooknum,
+        const struct xt_target *target, const void *targinfo)
+{
+	const struct xt_tos_target_info *info = targinfo;
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+	u_int8_t orig, nv;
+
+	orig = ipv6_get_dsfield(iph);
+	nv   = (orig & info->tos_mask) ^ info->tos_value;
+
+	if (orig != nv) {
+		if (!skb_make_writable(skb, sizeof(struct iphdr)))
+			return NF_DROP;
+		iph = ipv6_hdr(skb);
+		ipv6_change_dsfield(iph, ~0, nv);
+	}
+
+	return XT_CONTINUE;
+}
+
 static struct xt_target dscp_tg_reg[] __read_mostly = {
 	{
 		.name		= "DSCP",
@@ -146,6 +191,24 @@ static struct xt_target dscp_tg_reg[] __read_mostly = {
 		.checkentry	= tos_tg_check_v0,
 		.me		= THIS_MODULE,
 	},
+	{
+		.name		= "TOS",
+		.revision	= 1,
+		.family		= AF_INET,
+		.table		= "mangle",
+		.target		= tos_tg,
+		.targetsize	= sizeof(struct xt_tos_target_info),
+		.me		= THIS_MODULE,
+	},
+	{
+		.name		= "TOS",
+		.revision	= 1,
+		.family		= AF_INET6,
+		.table		= "mangle",
+		.target		= tos_tg6,
+		.targetsize	= sizeof(struct xt_tos_target_info),
+		.me		= THIS_MODULE,
+	},
 };
 
 static int __init dscp_tg_init(void)
-- 
cgit v1.2.3


From 5859034d7eb8793d3d78d3af515c4175e7b9d03a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 4 Dec 2007 23:40:05 -0800
Subject: [NETFILTER]: x_tables: add RATEEST target

Add new rate estimator target (using gen_estimator). In combination with
the rateest match (next patch) this can be used for load-based multipath
routing.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild       |   1 +
 include/linux/netfilter/xt_RATEEST.h |  11 ++
 include/net/netfilter/xt_rateest.h   |  17 +++
 net/netfilter/Kconfig                |  10 ++
 net/netfilter/Makefile               |   1 +
 net/netfilter/xt_RATEEST.c           | 204 +++++++++++++++++++++++++++++++++++
 6 files changed, 244 insertions(+)
 create mode 100644 include/linux/netfilter/xt_RATEEST.h
 create mode 100644 include/net/netfilter/xt_rateest.h
 create mode 100644 net/netfilter/xt_RATEEST.c

(limited to 'include')

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index 1e6900278318..707a15854430 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -10,6 +10,7 @@ header-y += xt_DSCP.h
 header-y += xt_MARK.h
 header-y += xt_NFLOG.h
 header-y += xt_NFQUEUE.h
+header-y += xt_RATEEST.h
 header-y += xt_SECMARK.h
 header-y += xt_TCPMSS.h
 header-y += xt_comment.h
diff --git a/include/linux/netfilter/xt_RATEEST.h b/include/linux/netfilter/xt_RATEEST.h
new file mode 100644
index 000000000000..670f2e49d4fb
--- /dev/null
+++ b/include/linux/netfilter/xt_RATEEST.h
@@ -0,0 +1,11 @@
+#ifndef _XT_RATEEST_TARGET_H
+#define _XT_RATEEST_TARGET_H
+
+struct xt_rateest_target_info {
+	char			name[IFNAMSIZ];
+	int8_t			interval;
+	u_int8_t		ewma_log;
+	struct xt_rateest	*est __attribute__((aligned(8)));
+};
+
+#endif /* _XT_RATEEST_TARGET_H */
diff --git a/include/net/netfilter/xt_rateest.h b/include/net/netfilter/xt_rateest.h
new file mode 100644
index 000000000000..65d594dffbff
--- /dev/null
+++ b/include/net/netfilter/xt_rateest.h
@@ -0,0 +1,17 @@
+#ifndef _XT_RATEEST_H
+#define _XT_RATEEST_H
+
+struct xt_rateest {
+	struct hlist_node		list;
+	char				name[IFNAMSIZ];
+	unsigned int			refcnt;
+	spinlock_t			lock;
+	struct gnet_estimator		params;
+	struct gnet_stats_rate_est	rstats;
+	struct gnet_stats_basic		bstats;
+};
+
+extern struct xt_rateest *xt_rateest_lookup(const char *name);
+extern void xt_rateest_put(struct xt_rateest *est);
+
+#endif /* _XT_RATEEST_H */
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 7bde6315999a..22d1f10e88b0 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -357,6 +357,16 @@ config NETFILTER_XT_TARGET_NOTRACK
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+config NETFILTER_XT_TARGET_RATEEST
+	tristate '"RATEEST" target support'
+	depends on NETFILTER_XTABLES
+	help
+	  This option adds a `RATEEST' target, which allows to measure
+	  rates similar to TC estimators. The `rateest' match can be
+	  used to match on the measured rates.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config NETFILTER_XT_TARGET_TRACE
 	tristate  '"TRACE" target support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 28f59a35aeef..413afaad361f 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_RATEEST) += xt_RATEEST.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP) += xt_TCPOPTSTRIP.o
diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c
new file mode 100644
index 000000000000..c0088839fde2
--- /dev/null
+++ b/net/netfilter/xt_RATEEST.c
@@ -0,0 +1,204 @@
+/*
+ * (C) 2007 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/gen_stats.h>
+#include <linux/jhash.h>
+#include <linux/rtnetlink.h>
+#include <linux/random.h>
+#include <net/gen_stats.h>
+
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_RATEEST.h>
+#include <net/netfilter/xt_rateest.h>
+
+static DEFINE_MUTEX(xt_rateest_mutex);
+
+#define RATEEST_HSIZE	16
+static struct hlist_head rateest_hash[RATEEST_HSIZE] __read_mostly;
+static unsigned int jhash_rnd __read_mostly;
+
+static unsigned int xt_rateest_hash(const char *name)
+{
+	return jhash(name, FIELD_SIZEOF(struct xt_rateest, name), jhash_rnd) &
+	       (RATEEST_HSIZE - 1);
+}
+
+static void xt_rateest_hash_insert(struct xt_rateest *est)
+{
+	unsigned int h;
+
+	h = xt_rateest_hash(est->name);
+	hlist_add_head(&est->list, &rateest_hash[h]);
+}
+
+struct xt_rateest *xt_rateest_lookup(const char *name)
+{
+	struct xt_rateest *est;
+	struct hlist_node *n;
+	unsigned int h;
+
+	h = xt_rateest_hash(name);
+	mutex_lock(&xt_rateest_mutex);
+	hlist_for_each_entry(est, n, &rateest_hash[h], list) {
+		if (strcmp(est->name, name) == 0) {
+			est->refcnt++;
+			mutex_unlock(&xt_rateest_mutex);
+			return est;
+		}
+	}
+	mutex_unlock(&xt_rateest_mutex);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(xt_rateest_lookup);
+
+void xt_rateest_put(struct xt_rateest *est)
+{
+	mutex_lock(&xt_rateest_mutex);
+	if (--est->refcnt == 0) {
+		hlist_del(&est->list);
+		gen_kill_estimator(&est->bstats, &est->rstats);
+		kfree(est);
+	}
+	mutex_unlock(&xt_rateest_mutex);
+}
+EXPORT_SYMBOL_GPL(xt_rateest_put);
+
+static unsigned int
+xt_rateest_tg(struct sk_buff *skb,
+	      const struct net_device *in,
+	      const struct net_device *out,
+	      unsigned int hooknum,
+	      const struct xt_target *target,
+	      const void *targinfo)
+{
+	const struct xt_rateest_target_info *info = targinfo;
+	struct gnet_stats_basic *stats = &info->est->bstats;
+
+	spin_lock_bh(&info->est->lock);
+	stats->bytes += skb->len;
+	stats->packets++;
+	spin_unlock_bh(&info->est->lock);
+
+	return XT_CONTINUE;
+}
+
+static bool
+xt_rateest_tg_checkentry(const char *tablename,
+			 const void *entry,
+			 const struct xt_target *target,
+			 void *targinfo,
+			 unsigned int hook_mask)
+{
+	struct xt_rateest_target_info *info = (void *)targinfo;
+	struct xt_rateest *est;
+	struct {
+		struct rtattr		opt;
+		struct gnet_estimator	est;
+	} cfg;
+
+	est = xt_rateest_lookup(info->name);
+	if (est) {
+		/*
+		 * If estimator parameters are specified, they must match the
+		 * existing estimator.
+		 */
+		if ((!info->interval && !info->ewma_log) ||
+		    (info->interval != est->params.interval ||
+		     info->ewma_log != est->params.ewma_log)) {
+			xt_rateest_put(est);
+			return false;
+		}
+		info->est = est;
+		return true;
+	}
+
+	est = kzalloc(sizeof(*est), GFP_KERNEL);
+	if (!est)
+		goto err1;
+
+	strlcpy(est->name, info->name, sizeof(est->name));
+	spin_lock_init(&est->lock);
+	est->refcnt		= 1;
+	est->params.interval	= info->interval;
+	est->params.ewma_log	= info->ewma_log;
+
+	cfg.opt.rta_len		= RTA_LENGTH(sizeof(cfg.est));
+	cfg.opt.rta_type	= TCA_STATS_RATE_EST;
+	cfg.est.interval	= info->interval;
+	cfg.est.ewma_log	= info->ewma_log;
+
+	if (gen_new_estimator(&est->bstats, &est->rstats, &est->lock,
+			      &cfg.opt) < 0)
+		goto err2;
+
+	info->est = est;
+	xt_rateest_hash_insert(est);
+
+	return true;
+
+err2:
+	kfree(est);
+err1:
+	return false;
+}
+
+static void xt_rateest_tg_destroy(const struct xt_target *target,
+				  void *targinfo)
+{
+	struct xt_rateest_target_info *info = targinfo;
+
+	xt_rateest_put(info->est);
+}
+
+static struct xt_target xt_rateest_target[] __read_mostly = {
+	{
+		.family		= AF_INET,
+		.name		= "RATEEST",
+		.target		= xt_rateest_tg,
+		.checkentry	= xt_rateest_tg_checkentry,
+		.destroy	= xt_rateest_tg_destroy,
+		.targetsize	= sizeof(struct xt_rateest_target_info),
+		.me		= THIS_MODULE,
+	},
+	{
+		.family		= AF_INET6,
+		.name		= "RATEEST",
+		.target		= xt_rateest_tg,
+		.checkentry	= xt_rateest_tg_checkentry,
+		.destroy	= xt_rateest_tg_destroy,
+		.targetsize	= sizeof(struct xt_rateest_target_info),
+		.me		= THIS_MODULE,
+	},
+};
+
+static int __init xt_rateest_tg_init(void)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(rateest_hash); i++)
+		INIT_HLIST_HEAD(&rateest_hash[i]);
+
+	get_random_bytes(&jhash_rnd, sizeof(jhash_rnd));
+	return xt_register_targets(xt_rateest_target,
+				   ARRAY_SIZE(xt_rateest_target));
+}
+
+static void __exit xt_rateest_tg_fini(void)
+{
+	xt_unregister_targets(xt_rateest_target, ARRAY_SIZE(xt_rateest_target));
+}
+
+
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("xtables rate estimator");
+MODULE_ALIAS("ipt_RATEEST");
+MODULE_ALIAS("ip6t_RATEEST");
+module_init(xt_rateest_tg_init);
+module_exit(xt_rateest_tg_fini);
-- 
cgit v1.2.3


From 50c164a81f1c0dfad056f99e5685537fdd0f07dd Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 4 Dec 2007 13:02:19 +0100
Subject: [NETFILTER]: x_tables: add rateest match

Add rate estimator match. The rate estimator match can match on
estimated rates by the RATEEST target. It supports matching on
absolute bps/pps values, comparing two rate estimators and matching
on the difference between two rate estimators.

This is what I use to route outgoing data connections from a FTP
server over two lines based on the  available bandwidth:

# estimate outgoing rates
iptables -t mangle -A POSTROUTING -o eth0 -j RATEEST --rateest-name eth0 \
                                                     --rateest-interval 250ms \
                                                     --rateest-ewma 0.5s
iptables -t mangle -A POSTROUTING -o ppp0 -j RATEEST --rateest-name ppp0 \
                                                     --rateest-interval 250ms \
                                                     --rateest-ewma 0.5s

# mark based on available bandwidth
iptables -t mangle -A BALANCE -m state --state NEW \
                              -m helper --helper ftp \
                              -m rateest --rateest-delta \
                                         --rateest1 eth0 \
                                         --rateest-bps1 2.5mbit \
                                         --rateest-gt \
                                         --rateest2 ppp0 \
                                         --rateest-bps2 2mbit \
                              -j CONNMARK --set-mark 0x1

iptables -t mangle -A BALANCE -m state --state NEW \
                              -m helper --helper ftp \
                              -m rateest --rateest-delta \
                                         --rateest1 ppp0 \
                                         --rateest-bps1 2mbit \
                                         --rateest-gt \
                                         --rateest2 eth0 \
                                         --rateest-bps2 2.5mbit \
                              -j CONNMARK --set-mark 0x2

iptables -t mangle -A BALANCE -j CONNMARK --restore-mark

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild       |   1 +
 include/linux/netfilter/xt_rateest.h |  33 +++++++
 net/netfilter/Kconfig                |  10 ++
 net/netfilter/Makefile               |   1 +
 net/netfilter/xt_rateest.c           | 178 +++++++++++++++++++++++++++++++++++
 5 files changed, 223 insertions(+)
 create mode 100644 include/linux/netfilter/xt_rateest.h
 create mode 100644 net/netfilter/xt_rateest.c

(limited to 'include')

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index 707a15854430..ac9e6429f747 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -30,6 +30,7 @@ header-y += xt_multiport.h
 header-y += xt_owner.h
 header-y += xt_pkttype.h
 header-y += xt_policy.h
+header-y += xt_rateest.h
 header-y += xt_realm.h
 header-y += xt_sctp.h
 header-y += xt_state.h
diff --git a/include/linux/netfilter/xt_rateest.h b/include/linux/netfilter/xt_rateest.h
new file mode 100644
index 000000000000..51948e15aea2
--- /dev/null
+++ b/include/linux/netfilter/xt_rateest.h
@@ -0,0 +1,33 @@
+#ifndef _XT_RATEEST_MATCH_H
+#define _XT_RATEEST_MATCH_H
+
+enum xt_rateest_match_flags {
+	XT_RATEEST_MATCH_INVERT	= 1<<0,
+	XT_RATEEST_MATCH_ABS	= 1<<1,
+	XT_RATEEST_MATCH_REL	= 1<<2,
+	XT_RATEEST_MATCH_DELTA	= 1<<3,
+	XT_RATEEST_MATCH_BPS	= 1<<4,
+	XT_RATEEST_MATCH_PPS	= 1<<5,
+};
+
+enum xt_rateest_match_mode {
+	XT_RATEEST_MATCH_NONE,
+	XT_RATEEST_MATCH_EQ,
+	XT_RATEEST_MATCH_LT,
+	XT_RATEEST_MATCH_GT,
+};
+
+struct xt_rateest_match_info {
+	char			name1[IFNAMSIZ];
+	char			name2[IFNAMSIZ];
+	u_int16_t		flags;
+	u_int16_t		mode;
+	u_int32_t		bps1;
+	u_int32_t		pps1;
+	u_int32_t		bps2;
+	u_int32_t		pps2;
+	struct xt_rateest	*est1 __attribute__((aligned(8)));
+	struct xt_rateest	*est2 __attribute__((aligned(8)));
+};
+
+#endif /* _XT_RATEEST_MATCH_H */
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 22d1f10e88b0..4182393186ae 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -631,6 +631,16 @@ config NETFILTER_XT_MATCH_QUOTA
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+config NETFILTER_XT_MATCH_RATEEST
+	tristate '"rateest" match support'
+	depends on NETFILTER_XTABLES
+	select NETFILTER_XT_TARGET_RATEEST
+	help
+	  This option adds a `rateest' match, which allows to match on the
+	  rate estimated by the RATEEST target.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config NETFILTER_XT_MATCH_REALM
 	tristate  '"realm" match support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 413afaad361f..3b9ea8fb3a07 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -73,6 +73,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_PKTTYPE) += xt_pkttype.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_POLICY) += xt_policy.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_QUOTA) += xt_quota.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_RATEEST) += xt_rateest.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_REALM) += xt_realm.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_SCTP) += xt_sctp.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_STATE) += xt_state.o
diff --git a/net/netfilter/xt_rateest.c b/net/netfilter/xt_rateest.c
new file mode 100644
index 000000000000..fdb86a515146
--- /dev/null
+++ b/net/netfilter/xt_rateest.c
@@ -0,0 +1,178 @@
+/*
+ * (C) 2007 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/gen_stats.h>
+
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_rateest.h>
+#include <net/netfilter/xt_rateest.h>
+
+
+static bool xt_rateest_mt(const struct sk_buff *skb,
+			  const struct net_device *in,
+			  const struct net_device *out,
+			  const struct xt_match *match,
+			  const void *matchinfo,
+			  int offset,
+			  unsigned int protoff,
+			  bool *hotdrop)
+{
+	const struct xt_rateest_match_info *info = matchinfo;
+	struct gnet_stats_rate_est *r;
+	u_int32_t bps1, bps2, pps1, pps2;
+	bool ret = true;
+
+	spin_lock_bh(&info->est1->lock);
+	r = &info->est1->rstats;
+	if (info->flags & XT_RATEEST_MATCH_DELTA) {
+		bps1 = info->bps1 >= r->bps ? info->bps1 - r->bps : 0;
+		pps1 = info->pps1 >= r->pps ? info->pps1 - r->pps : 0;
+	} else {
+		bps1 = r->bps;
+		pps1 = r->pps;
+	}
+	spin_unlock_bh(&info->est1->lock);
+
+	if (info->flags & XT_RATEEST_MATCH_ABS) {
+		bps2 = info->bps2;
+		pps2 = info->pps2;
+	} else {
+		spin_lock_bh(&info->est2->lock);
+		r = &info->est2->rstats;
+		if (info->flags & XT_RATEEST_MATCH_DELTA) {
+			bps2 = info->bps2 >= r->bps ? info->bps2 - r->bps : 0;
+			pps2 = info->pps2 >= r->pps ? info->pps2 - r->pps : 0;
+		} else {
+			bps2 = r->bps;
+			pps2 = r->pps;
+		}
+		spin_unlock_bh(&info->est2->lock);
+	}
+
+	switch (info->mode) {
+	case XT_RATEEST_MATCH_LT:
+		if (info->flags & XT_RATEEST_MATCH_BPS)
+			ret &= bps1 < bps2;
+		if (info->flags & XT_RATEEST_MATCH_PPS)
+			ret &= pps1 < pps2;
+		break;
+	case XT_RATEEST_MATCH_GT:
+		if (info->flags & XT_RATEEST_MATCH_BPS)
+			ret &= bps1 > bps2;
+		if (info->flags & XT_RATEEST_MATCH_PPS)
+			ret &= pps1 > pps2;
+		break;
+	case XT_RATEEST_MATCH_EQ:
+		if (info->flags & XT_RATEEST_MATCH_BPS)
+			ret &= bps1 == bps2;
+		if (info->flags & XT_RATEEST_MATCH_PPS)
+			ret &= pps2 == pps2;
+		break;
+	}
+
+	ret ^= info->flags & XT_RATEEST_MATCH_INVERT ? true : false;
+	return ret;
+}
+
+static bool xt_rateest_mt_checkentry(const char *tablename,
+				     const void *ip,
+				     const struct xt_match *match,
+				     void *matchinfo,
+				     unsigned int hook_mask)
+{
+	struct xt_rateest_match_info *info = (void *)matchinfo;
+	struct xt_rateest *est1, *est2;
+
+	if (hweight32(info->flags & (XT_RATEEST_MATCH_ABS |
+				     XT_RATEEST_MATCH_REL)) != 1)
+		goto err1;
+
+	if (!(info->flags & (XT_RATEEST_MATCH_BPS | XT_RATEEST_MATCH_PPS)))
+		goto err1;
+
+	switch (info->mode) {
+	case XT_RATEEST_MATCH_EQ:
+	case XT_RATEEST_MATCH_LT:
+	case XT_RATEEST_MATCH_GT:
+		break;
+	default:
+		goto err1;
+	}
+
+	est1 = xt_rateest_lookup(info->name1);
+	if (!est1)
+		goto err1;
+
+	if (info->flags & XT_RATEEST_MATCH_REL) {
+		est2 = xt_rateest_lookup(info->name2);
+		if (!est2)
+			goto err2;
+	} else
+		est2 = NULL;
+
+
+	info->est1 = est1;
+	info->est2 = est2;
+	return true;
+
+err2:
+	xt_rateest_put(est1);
+err1:
+	return false;
+}
+
+static void xt_rateest_mt_destroy(const struct xt_match *match,
+				  void *matchinfo)
+{
+	struct xt_rateest_match_info *info = (void *)matchinfo;
+
+	xt_rateest_put(info->est1);
+	if (info->est2)
+		xt_rateest_put(info->est2);
+}
+
+static struct xt_match xt_rateest_match[] __read_mostly = {
+	{
+		.family		= AF_INET,
+		.name		= "rateest",
+		.match		= xt_rateest_mt,
+		.checkentry	= xt_rateest_mt_checkentry,
+		.destroy	= xt_rateest_mt_destroy,
+		.matchsize	= sizeof(struct xt_rateest_match_info),
+		.me		= THIS_MODULE,
+	},
+	{
+		.family		= AF_INET6,
+		.name		= "rateest",
+		.match		= xt_rateest_mt,
+		.checkentry	= xt_rateest_mt_checkentry,
+		.destroy	= xt_rateest_mt_destroy,
+		.matchsize	= sizeof(struct xt_rateest_match_info),
+		.me		= THIS_MODULE,
+	},
+};
+
+static int __init xt_rateest_mt_init(void)
+{
+	return xt_register_matches(xt_rateest_match,
+				   ARRAY_SIZE(xt_rateest_match));
+}
+
+static void __exit xt_rateest_mt_fini(void)
+{
+	xt_unregister_matches(xt_rateest_match, ARRAY_SIZE(xt_rateest_match));
+}
+
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("xtables rate estimator match");
+MODULE_ALIAS("ipt_rateest");
+MODULE_ALIAS("ip6t_rateest");
+module_init(xt_rateest_mt_init);
+module_exit(xt_rateest_mt_fini);
-- 
cgit v1.2.3


From 1841a4c7ae106b7a3e2521db55f4d8bb8a0988d5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 01:22:05 -0800
Subject: [NETFILTER]: nf_ct_h323: remove ipv6 module dependency

nf_conntrack_h323 needs ip6_route_output for the call forwarding filter.
Add a ->route function to nf_afinfo and use that to avoid pulling in the
ipv6 module.

Fix the #ifdef for the IPv6 code while I'm at it - the IPv6 support is
only needed when IPv6 conntrack is enabled.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h              |  2 ++
 net/ipv4/netfilter.c                   |  6 ++++++
 net/ipv6/netfilter.c                   |  7 +++++++
 net/netfilter/nf_conntrack_h323_main.c | 19 ++++++++++++-------
 4 files changed, 27 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index f42e4362d50d..9bfc7d4f5868 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -298,10 +298,12 @@ extern void nf_invalidate_cache(int pf);
    Returns true or false. */
 extern int skb_make_writable(struct sk_buff *skb, unsigned int writable_len);
 
+struct flowi;
 struct nf_afinfo {
 	unsigned short	family;
 	__sum16		(*checksum)(struct sk_buff *skb, unsigned int hook,
 				    unsigned int dataoff, u_int8_t protocol);
+	int		(*route)(struct dst_entry **dst, struct flowi *fl);
 	void		(*saveroute)(const struct sk_buff *skb,
 				     struct nf_info *info);
 	int		(*reroute)(struct sk_buff *skb,
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index d9022467e089..599d448ef57e 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -182,9 +182,15 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 
 EXPORT_SYMBOL(nf_ip_checksum);
 
+static int nf_ip_route(struct dst_entry **dst, struct flowi *fl)
+{
+	return ip_route_output_key((struct rtable **)dst, fl);
+}
+
 static struct nf_afinfo nf_ip_afinfo = {
 	.family		= AF_INET,
 	.checksum	= nf_ip_checksum,
+	.route		= nf_ip_route,
 	.saveroute	= nf_ip_saveroute,
 	.reroute	= nf_ip_reroute,
 	.route_key_size	= sizeof(struct ip_rt_info),
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 175e19f80253..281f732e3c97 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -81,6 +81,12 @@ static int nf_ip6_reroute(struct sk_buff *skb, const struct nf_info *info)
 	return 0;
 }
 
+static int nf_ip6_route(struct dst_entry **dst, struct flowi *fl)
+{
+	*dst = ip6_route_output(NULL, fl);
+	return (*dst)->error;
+}
+
 __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 			     unsigned int dataoff, u_int8_t protocol)
 {
@@ -118,6 +124,7 @@ EXPORT_SYMBOL(nf_ip6_checksum);
 static struct nf_afinfo nf_ip6_afinfo = {
 	.family		= AF_INET6,
 	.checksum	= nf_ip6_checksum,
+	.route		= nf_ip6_route,
 	.saveroute	= nf_ip6_saveroute,
 	.reroute	= nf_ip6_reroute,
 	.route_key_size	= sizeof(struct ip6_rt_info),
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index f23fd9598e19..c550257b5462 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -708,9 +708,15 @@ static int callforward_do_filter(union nf_conntrack_address *src,
 				 union nf_conntrack_address *dst,
 				 int family)
 {
+	struct nf_afinfo *afinfo;
 	struct flowi fl1, fl2;
 	int ret = 0;
 
+	/* rcu_read_lock()ed by nf_hook_slow() */
+	afinfo = nf_get_afinfo(family);
+	if (!afinfo)
+		return 0;
+
 	memset(&fl1, 0, sizeof(fl1));
 	memset(&fl2, 0, sizeof(fl2));
 
@@ -720,8 +726,8 @@ static int callforward_do_filter(union nf_conntrack_address *src,
 
 		fl1.fl4_dst = src->ip;
 		fl2.fl4_dst = dst->ip;
-		if (ip_route_output_key(&rt1, &fl1) == 0) {
-			if (ip_route_output_key(&rt2, &fl2) == 0) {
+		if (!afinfo->route((struct dst_entry **)&rt1, &fl1)) {
+			if (!afinfo->route((struct dst_entry **)&rt2, &fl2)) {
 				if (rt1->rt_gateway == rt2->rt_gateway &&
 				    rt1->u.dst.dev  == rt2->u.dst.dev)
 					ret = 1;
@@ -731,16 +737,15 @@ static int callforward_do_filter(union nf_conntrack_address *src,
 		}
 		break;
 	}
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+#if defined(CONFIG_NF_CONNTRACK_IPV6) || \
+    defined(CONFIG_NF_CONNTRACK_IPV6_MODULE)
 	case AF_INET6: {
 		struct rt6_info *rt1, *rt2;
 
 		memcpy(&fl1.fl6_dst, src, sizeof(fl1.fl6_dst));
 		memcpy(&fl2.fl6_dst, dst, sizeof(fl2.fl6_dst));
-		rt1 = (struct rt6_info *)ip6_route_output(NULL, &fl1);
-		if (rt1) {
-			rt2 = (struct rt6_info *)ip6_route_output(NULL, &fl2);
-			if (rt2) {
+		if (!afinfo->route((struct dst_entry **)&rt1, &fl1)) {
+			if (!afinfo->route((struct dst_entry **)&rt2, &fl2)) {
 				if (!memcmp(&rt1->rt6i_gateway, &rt2->rt6i_gateway,
 					    sizeof(rt1->rt6i_gateway)) &&
 				    rt1->u.dst.dev == rt2->u.dst.dev)
-- 
cgit v1.2.3


From e3ac5298159c5286cef86f0865d4fa6a606bd391 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 01:23:57 -0800
Subject: [NETFILTER]: nf_queue: make queue_handler const

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h       |  8 ++++----
 net/ipv4/netfilter/ip_queue.c   |  2 +-
 net/ipv6/netfilter/ip6_queue.c  |  2 +-
 net/netfilter/nf_queue.c        | 12 ++++++------
 net/netfilter/nfnetlink_queue.c |  2 +-
 5 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 9bfc7d4f5868..c2c3fafa5fd0 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -281,11 +281,11 @@ struct nf_queue_handler {
 	void *data;
 	char *name;
 };
-extern int nf_register_queue_handler(int pf, 
-                                     struct nf_queue_handler *qh);
+extern int nf_register_queue_handler(int pf,
+				     const struct nf_queue_handler *qh);
 extern int nf_unregister_queue_handler(int pf,
-				       struct nf_queue_handler *qh);
-extern void nf_unregister_queue_handlers(struct nf_queue_handler *qh);
+				       const struct nf_queue_handler *qh);
+extern void nf_unregister_queue_handlers(const struct nf_queue_handler *qh);
 extern void nf_reinject(struct sk_buff *skb,
 			struct nf_info *info,
 			unsigned int verdict);
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 14d64a383db1..062ff196f2c5 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -645,7 +645,7 @@ static const struct file_operations ip_queue_proc_fops = {
 	.owner		= THIS_MODULE,
 };
 
-static struct nf_queue_handler nfqh = {
+static const struct nf_queue_handler nfqh = {
 	.name	= "ip_queue",
 	.outfn	= &ipq_enqueue_packet,
 };
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index e273605eef85..d6e971bd9fe1 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -634,7 +634,7 @@ static const struct file_operations ip6_queue_proc_fops = {
 	.owner		= THIS_MODULE,
 };
 
-static struct nf_queue_handler nfqh = {
+static const struct nf_queue_handler nfqh = {
 	.name	= "ip6_queue",
 	.outfn	= &ipq_enqueue_packet,
 };
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 0bea88c30e51..dd18126a1a6d 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -15,13 +15,13 @@
  * long term mutex.  The handler must provide an an outfn() to accept packets
  * for queueing and must reinject all packets it receives, no matter what.
  */
-static struct nf_queue_handler *queue_handler[NPROTO];
+static const struct nf_queue_handler *queue_handler[NPROTO];
 
 static DEFINE_MUTEX(queue_handler_mutex);
 
 /* return EBUSY when somebody else is registered, return EEXIST if the
  * same handler is registered, return 0 in case of success. */
-int nf_register_queue_handler(int pf, struct nf_queue_handler *qh)
+int nf_register_queue_handler(int pf, const struct nf_queue_handler *qh)
 {
 	int ret;
 
@@ -44,7 +44,7 @@ int nf_register_queue_handler(int pf, struct nf_queue_handler *qh)
 EXPORT_SYMBOL(nf_register_queue_handler);
 
 /* The caller must flush their queue before this */
-int nf_unregister_queue_handler(int pf, struct nf_queue_handler *qh)
+int nf_unregister_queue_handler(int pf, const struct nf_queue_handler *qh)
 {
 	if (pf >= NPROTO)
 		return -EINVAL;
@@ -64,7 +64,7 @@ int nf_unregister_queue_handler(int pf, struct nf_queue_handler *qh)
 }
 EXPORT_SYMBOL(nf_unregister_queue_handler);
 
-void nf_unregister_queue_handlers(struct nf_queue_handler *qh)
+void nf_unregister_queue_handlers(const struct nf_queue_handler *qh)
 {
 	int pf;
 
@@ -98,7 +98,7 @@ static int __nf_queue(struct sk_buff *skb,
 	struct net_device *physoutdev = NULL;
 #endif
 	struct nf_afinfo *afinfo;
-	struct nf_queue_handler *qh;
+	const struct nf_queue_handler *qh;
 
 	/* QUEUE == DROP if noone is waiting, to be safe. */
 	rcu_read_lock();
@@ -313,7 +313,7 @@ static int seq_show(struct seq_file *s, void *v)
 {
 	int ret;
 	loff_t *pos = v;
-	struct nf_queue_handler *qh;
+	const struct nf_queue_handler *qh;
 
 	rcu_read_lock();
 	qh = rcu_dereference(queue_handler[*pos]);
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 3ceeffcf6f9d..b75091c8ae5e 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -849,7 +849,7 @@ static const struct nla_policy nfqa_cfg_policy[NFQA_CFG_MAX+1] = {
 	[NFQA_CFG_PARAMS]	= { .len = sizeof(struct nfqnl_msg_config_params) },
 };
 
-static struct nf_queue_handler nfqh = {
+static const struct nf_queue_handler nfqh = {
 	.name 	= "nf_queue",
 	.outfn	= &nfqnl_enqueue_packet,
 };
-- 
cgit v1.2.3


From f9d8928f8340ab8e76f1da4799cb19a6ff58b83d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 01:24:30 -0800
Subject: [NETFILTER]: nf_queue: remove unused data pointer

Remove the data pointer from struct nf_queue_handler. It has never been used
and is useless for the only handler that really matters, nfnetlink_queue,
since the handler is shared between all instances.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h       | 3 +--
 net/ipv4/netfilter/ip_queue.c   | 2 +-
 net/ipv6/netfilter/ip6_queue.c  | 2 +-
 net/netfilter/nf_queue.c        | 2 +-
 net/netfilter/nfnetlink_queue.c | 2 +-
 5 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index c2c3fafa5fd0..1ba60112ab83 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -277,8 +277,7 @@ int compat_nf_getsockopt(struct sock *sk, int pf, int optval,
 /* Packet queuing */
 struct nf_queue_handler {
 	int (*outfn)(struct sk_buff *skb, struct nf_info *info,
-		     unsigned int queuenum, void *data);
-	void *data;
+		     unsigned int queuenum);
 	char *name;
 };
 extern int nf_register_queue_handler(int pf,
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 062ff196f2c5..08e7f8b4e951 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -272,7 +272,7 @@ nlmsg_failure:
 
 static int
 ipq_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		   unsigned int queuenum, void *data)
+		   unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index d6e971bd9fe1..5a9ca0d4fb2f 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -269,7 +269,7 @@ nlmsg_failure:
 
 static int
 ipq_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		   unsigned int queuenum, void *data)
+		   unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index dd18126a1a6d..c098ccbbbcee 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -153,7 +153,7 @@ static int __nf_queue(struct sk_buff *skb,
 	}
 #endif
 	afinfo->saveroute(skb, info);
-	status = qh->outfn(skb, info, queuenum, qh->data);
+	status = qh->outfn(skb, info, queuenum);
 
 	rcu_read_unlock();
 
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index b75091c8ae5e..94ec1c263d03 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -534,7 +534,7 @@ nla_put_failure:
 
 static int
 nfqnl_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		     unsigned int queuenum, void *data)
+		     unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
-- 
cgit v1.2.3


From c01cd429fc118c5db92475c5f08b307718aa4efc Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 01:24:48 -0800
Subject: [NETFILTER]: nf_queue: move queueing related functions/struct to
 seperate header

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h        | 32 ++------------------------------
 include/net/netfilter/nf_queue.h | 32 ++++++++++++++++++++++++++++++++
 net/ipv4/netfilter.c             |  1 +
 net/ipv4/netfilter/ip_queue.c    |  1 +
 net/ipv6/netfilter.c             |  1 +
 net/ipv6/netfilter/ip6_queue.c   |  1 +
 net/netfilter/nf_queue.c         |  1 +
 net/netfilter/nfnetlink_queue.c  |  1 +
 8 files changed, 40 insertions(+), 30 deletions(-)
 create mode 100644 include/net/netfilter/nf_queue.h

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 1ba60112ab83..5fe4ef401cc8 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -101,19 +101,6 @@ struct nf_sockopt_ops
 	struct module *owner;
 };
 
-/* Each queued (to userspace) skbuff has one of these. */
-struct nf_info
-{
-	/* The ops struct which sent us to userspace. */
-	struct nf_hook_ops *elem;
-	
-	/* If we're sent to userspace, this keeps housekeeping info */
-	int pf;
-	unsigned int hook;
-	struct net_device *indev, *outdev;
-	int (*okfn)(struct sk_buff *);
-};
-                                                                                
 /* Function to register/unregister hook points. */
 int nf_register_hook(struct nf_hook_ops *reg);
 void nf_unregister_hook(struct nf_hook_ops *reg);
@@ -274,21 +261,6 @@ int compat_nf_setsockopt(struct sock *sk, int pf, int optval,
 int compat_nf_getsockopt(struct sock *sk, int pf, int optval,
 		char __user *opt, int *len);
 
-/* Packet queuing */
-struct nf_queue_handler {
-	int (*outfn)(struct sk_buff *skb, struct nf_info *info,
-		     unsigned int queuenum);
-	char *name;
-};
-extern int nf_register_queue_handler(int pf,
-				     const struct nf_queue_handler *qh);
-extern int nf_unregister_queue_handler(int pf,
-				       const struct nf_queue_handler *qh);
-extern void nf_unregister_queue_handlers(const struct nf_queue_handler *qh);
-extern void nf_reinject(struct sk_buff *skb,
-			struct nf_info *info,
-			unsigned int verdict);
-
 /* FIXME: Before cache is ever used, this must be implemented for real. */
 extern void nf_invalidate_cache(int pf);
 
@@ -298,6 +270,8 @@ extern void nf_invalidate_cache(int pf);
 extern int skb_make_writable(struct sk_buff *skb, unsigned int writable_len);
 
 struct flowi;
+struct nf_info;
+
 struct nf_afinfo {
 	unsigned short	family;
 	__sum16		(*checksum)(struct sk_buff *skb, unsigned int hook,
@@ -334,8 +308,6 @@ nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff,
 extern int nf_register_afinfo(struct nf_afinfo *afinfo);
 extern void nf_unregister_afinfo(struct nf_afinfo *afinfo);
 
-#define nf_info_reroute(x) ((void *)x + sizeof(struct nf_info))
-
 #include <net/flow.h>
 extern void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
 
diff --git a/include/net/netfilter/nf_queue.h b/include/net/netfilter/nf_queue.h
new file mode 100644
index 000000000000..8c6b382fd865
--- /dev/null
+++ b/include/net/netfilter/nf_queue.h
@@ -0,0 +1,32 @@
+#ifndef _NF_QUEUE_H
+#define _NF_QUEUE_H
+
+/* Each queued (to userspace) skbuff has one of these. */
+struct nf_info {
+	struct nf_hook_ops	*elem;
+	int			pf;
+	unsigned int		hook;
+	struct net_device	*indev;
+	struct net_device	*outdev;
+	int			(*okfn)(struct sk_buff *);
+};
+
+#define nf_info_reroute(x) ((void *)x + sizeof(struct nf_info))
+
+/* Packet queuing */
+struct nf_queue_handler {
+	int			(*outfn)(struct sk_buff *skb,
+					 struct nf_info *info,
+					 unsigned int queuenum);
+	char			*name;
+};
+
+extern int nf_register_queue_handler(int pf,
+				     const struct nf_queue_handler *qh);
+extern int nf_unregister_queue_handler(int pf,
+				       const struct nf_queue_handler *qh);
+extern void nf_unregister_queue_handlers(const struct nf_queue_handler *qh);
+extern void nf_reinject(struct sk_buff *skb, struct nf_info *info,
+			unsigned int verdict);
+
+#endif /* _NF_QUEUE_H */
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 599d448ef57e..f7166084a5ab 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -7,6 +7,7 @@
 #include <net/route.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
+#include <net/netfilter/nf_queue.h>
 
 /* route_me_harder function, used by iptable_nat, iptable_mangle + ip_queue */
 int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 08e7f8b4e951..2966fbddce8b 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -28,6 +28,7 @@
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/route.h>
+#include <net/netfilter/nf_queue.h>
 
 #define IPQ_QMAX_DEFAULT 1024
 #define IPQ_PROC_FS_NAME "ip_queue"
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 281f732e3c97..55ea9c6ec744 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -8,6 +8,7 @@
 #include <net/ip6_route.h>
 #include <net/xfrm.h>
 #include <net/ip6_checksum.h>
+#include <net/netfilter/nf_queue.h>
 
 int ip6_route_me_harder(struct sk_buff *skb)
 {
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index 5a9ca0d4fb2f..7ff9915750a3 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -29,6 +29,7 @@
 #include <net/sock.h>
 #include <net/ipv6.h>
 #include <net/ip6_route.h>
+#include <net/netfilter/nf_queue.h>
 #include <linux/netfilter_ipv4/ip_queue.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv6/ip6_tables.h>
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index c098ccbbbcee..bd71f433b85e 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -7,6 +7,7 @@
 #include <linux/seq_file.h>
 #include <linux/rcupdate.h>
 #include <net/protocol.h>
+#include <net/netfilter/nf_queue.h>
 
 #include "nf_internals.h"
 
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 94ec1c263d03..3a09f021065a 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -27,6 +27,7 @@
 #include <linux/netfilter/nfnetlink_queue.h>
 #include <linux/list.h>
 #include <net/sock.h>
+#include <net/netfilter/nf_queue.h>
 
 #include <asm/atomic.h>
 
-- 
cgit v1.2.3


From 02f014d88831f73b895c1fe09badb66c88e932d3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 01:26:33 -0800
Subject: [NETFILTER]: nf_queue: move list_head/skb/id to struct nf_info

Move common fields for queue management to struct nf_info and rename it
to struct nf_queue_entry. The avoids one allocation/free per packet and
simplifies the code a bit.

Alternatively we could add some private room at the tail, but since
all current users use identical structs this seems easier.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h        |  6 ++--
 include/net/netfilter/nf_queue.h | 14 ++++----
 net/ipv4/netfilter.c             | 14 ++++----
 net/ipv4/netfilter/ip_queue.c    | 68 +++++++++++++-----------------------
 net/ipv6/netfilter.c             | 14 ++++----
 net/ipv6/netfilter/ip6_queue.c   | 67 +++++++++++++-----------------------
 net/netfilter/nf_queue.c         | 65 +++++++++++++++++++----------------
 net/netfilter/nfnetlink_queue.c  | 74 +++++++++++++---------------------------
 8 files changed, 132 insertions(+), 190 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 5fe4ef401cc8..f25eec595807 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -270,7 +270,7 @@ extern void nf_invalidate_cache(int pf);
 extern int skb_make_writable(struct sk_buff *skb, unsigned int writable_len);
 
 struct flowi;
-struct nf_info;
+struct nf_queue_entry;
 
 struct nf_afinfo {
 	unsigned short	family;
@@ -278,9 +278,9 @@ struct nf_afinfo {
 				    unsigned int dataoff, u_int8_t protocol);
 	int		(*route)(struct dst_entry **dst, struct flowi *fl);
 	void		(*saveroute)(const struct sk_buff *skb,
-				     struct nf_info *info);
+				     struct nf_queue_entry *entry);
 	int		(*reroute)(struct sk_buff *skb,
-				   const struct nf_info *info);
+				   const struct nf_queue_entry *entry);
 	int		route_key_size;
 };
 
diff --git a/include/net/netfilter/nf_queue.h b/include/net/netfilter/nf_queue.h
index 8c6b382fd865..d030044e9235 100644
--- a/include/net/netfilter/nf_queue.h
+++ b/include/net/netfilter/nf_queue.h
@@ -2,7 +2,11 @@
 #define _NF_QUEUE_H
 
 /* Each queued (to userspace) skbuff has one of these. */
-struct nf_info {
+struct nf_queue_entry {
+	struct list_head	list;
+	struct sk_buff		*skb;
+	unsigned int		id;
+
 	struct nf_hook_ops	*elem;
 	int			pf;
 	unsigned int		hook;
@@ -11,12 +15,11 @@ struct nf_info {
 	int			(*okfn)(struct sk_buff *);
 };
 
-#define nf_info_reroute(x) ((void *)x + sizeof(struct nf_info))
+#define nf_queue_entry_reroute(x) ((void *)x + sizeof(struct nf_queue_entry))
 
 /* Packet queuing */
 struct nf_queue_handler {
-	int			(*outfn)(struct sk_buff *skb,
-					 struct nf_info *info,
+	int			(*outfn)(struct nf_queue_entry *entry,
 					 unsigned int queuenum);
 	char			*name;
 };
@@ -26,7 +29,6 @@ extern int nf_register_queue_handler(int pf,
 extern int nf_unregister_queue_handler(int pf,
 				       const struct nf_queue_handler *qh);
 extern void nf_unregister_queue_handlers(const struct nf_queue_handler *qh);
-extern void nf_reinject(struct sk_buff *skb, struct nf_info *info,
-			unsigned int verdict);
+extern void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict);
 
 #endif /* _NF_QUEUE_H */
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index f7166084a5ab..7bf5e4a199f0 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -123,11 +123,12 @@ struct ip_rt_info {
 	u_int8_t tos;
 };
 
-static void nf_ip_saveroute(const struct sk_buff *skb, struct nf_info *info)
+static void nf_ip_saveroute(const struct sk_buff *skb,
+			    struct nf_queue_entry *entry)
 {
-	struct ip_rt_info *rt_info = nf_info_reroute(info);
+	struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
 
-	if (info->hook == NF_INET_LOCAL_OUT) {
+	if (entry->hook == NF_INET_LOCAL_OUT) {
 		const struct iphdr *iph = ip_hdr(skb);
 
 		rt_info->tos = iph->tos;
@@ -136,11 +137,12 @@ static void nf_ip_saveroute(const struct sk_buff *skb, struct nf_info *info)
 	}
 }
 
-static int nf_ip_reroute(struct sk_buff *skb, const struct nf_info *info)
+static int nf_ip_reroute(struct sk_buff *skb,
+			 const struct nf_queue_entry *entry)
 {
-	const struct ip_rt_info *rt_info = nf_info_reroute(info);
+	const struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
 
-	if (info->hook == NF_INET_LOCAL_OUT) {
+	if (entry->hook == NF_INET_LOCAL_OUT) {
 		const struct iphdr *iph = ip_hdr(skb);
 
 		if (!(iph->tos == rt_info->tos
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index df2957c5bcb4..f1affd2344a9 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -35,13 +35,7 @@
 #define NET_IPQ_QMAX 2088
 #define NET_IPQ_QMAX_NAME "ip_queue_maxlen"
 
-struct ipq_queue_entry {
-	struct list_head list;
-	struct nf_info *info;
-	struct sk_buff *skb;
-};
-
-typedef int (*ipq_cmpfn)(struct ipq_queue_entry *, unsigned long);
+typedef int (*ipq_cmpfn)(struct nf_queue_entry *, unsigned long);
 
 static unsigned char copy_mode __read_mostly = IPQ_COPY_NONE;
 static unsigned int queue_maxlen __read_mostly = IPQ_QMAX_DEFAULT;
@@ -56,7 +50,7 @@ static LIST_HEAD(queue_list);
 static DEFINE_MUTEX(ipqnl_mutex);
 
 static void
-ipq_issue_verdict(struct ipq_queue_entry *entry, int verdict)
+ipq_issue_verdict(struct nf_queue_entry *entry, int verdict)
 {
 	/* TCP input path (and probably other bits) assume to be called
 	 * from softirq context, not from syscall, like ipq_issue_verdict is
@@ -64,14 +58,12 @@ ipq_issue_verdict(struct ipq_queue_entry *entry, int verdict)
 	 * softirq, e.g.  We therefore emulate this by local_bh_disable() */
 
 	local_bh_disable();
-	nf_reinject(entry->skb, entry->info, verdict);
+	nf_reinject(entry, verdict);
 	local_bh_enable();
-
-	kfree(entry);
 }
 
 static inline void
-__ipq_enqueue_entry(struct ipq_queue_entry *entry)
+__ipq_enqueue_entry(struct nf_queue_entry *entry)
 {
        list_add_tail(&entry->list, &queue_list);
        queue_total++;
@@ -114,10 +106,10 @@ __ipq_reset(void)
 	__ipq_flush(NULL, 0);
 }
 
-static struct ipq_queue_entry *
+static struct nf_queue_entry *
 ipq_find_dequeue_entry(unsigned long id)
 {
-	struct ipq_queue_entry *entry = NULL, *i;
+	struct nf_queue_entry *entry = NULL, *i;
 
 	write_lock_bh(&queue_lock);
 
@@ -140,7 +132,7 @@ ipq_find_dequeue_entry(unsigned long id)
 static void
 __ipq_flush(ipq_cmpfn cmpfn, unsigned long data)
 {
-	struct ipq_queue_entry *entry, *next;
+	struct nf_queue_entry *entry, *next;
 
 	list_for_each_entry_safe(entry, next, &queue_list, list) {
 		if (!cmpfn || cmpfn(entry, data)) {
@@ -160,7 +152,7 @@ ipq_flush(ipq_cmpfn cmpfn, unsigned long data)
 }
 
 static struct sk_buff *
-ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
+ipq_build_packet_message(struct nf_queue_entry *entry, int *errp)
 {
 	sk_buff_data_t old_tail;
 	size_t size = 0;
@@ -217,20 +209,20 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 	pmsg->timestamp_sec   = tv.tv_sec;
 	pmsg->timestamp_usec  = tv.tv_usec;
 	pmsg->mark            = entry->skb->mark;
-	pmsg->hook            = entry->info->hook;
+	pmsg->hook            = entry->hook;
 	pmsg->hw_protocol     = entry->skb->protocol;
 
-	if (entry->info->indev)
-		strcpy(pmsg->indev_name, entry->info->indev->name);
+	if (entry->indev)
+		strcpy(pmsg->indev_name, entry->indev->name);
 	else
 		pmsg->indev_name[0] = '\0';
 
-	if (entry->info->outdev)
-		strcpy(pmsg->outdev_name, entry->info->outdev->name);
+	if (entry->outdev)
+		strcpy(pmsg->outdev_name, entry->outdev->name);
 	else
 		pmsg->outdev_name[0] = '\0';
 
-	if (entry->info->indev && entry->skb->dev) {
+	if (entry->indev && entry->skb->dev) {
 		pmsg->hw_type = entry->skb->dev->type;
 		pmsg->hw_addrlen = dev_parse_header(entry->skb,
 						    pmsg->hw_addr);
@@ -252,28 +244,17 @@ nlmsg_failure:
 }
 
 static int
-ipq_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		   unsigned int queuenum)
+ipq_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
-	struct ipq_queue_entry *entry;
 
 	if (copy_mode == IPQ_COPY_NONE)
 		return -EAGAIN;
 
-	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
-	if (entry == NULL) {
-		printk(KERN_ERR "ip_queue: OOM in ipq_enqueue_packet()\n");
-		return -ENOMEM;
-	}
-
-	entry->info = info;
-	entry->skb = skb;
-
 	nskb = ipq_build_packet_message(entry, &status);
 	if (nskb == NULL)
-		goto err_out_free;
+		return status;
 
 	write_lock_bh(&queue_lock);
 
@@ -307,14 +288,11 @@ err_out_free_nskb:
 
 err_out_unlock:
 	write_unlock_bh(&queue_lock);
-
-err_out_free:
-	kfree(entry);
 	return status;
 }
 
 static int
-ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
+ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct nf_queue_entry *e)
 {
 	int diff;
 	int err;
@@ -352,7 +330,7 @@ ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
 static int
 ipq_set_verdict(struct ipq_verdict_msg *vmsg, unsigned int len)
 {
-	struct ipq_queue_entry *entry;
+	struct nf_queue_entry *entry;
 
 	if (vmsg->value > NF_MAX_VERDICT)
 		return -EINVAL;
@@ -412,13 +390,13 @@ ipq_receive_peer(struct ipq_peer_msg *pmsg,
 }
 
 static int
-dev_cmp(struct ipq_queue_entry *entry, unsigned long ifindex)
+dev_cmp(struct nf_queue_entry *entry, unsigned long ifindex)
 {
-	if (entry->info->indev)
-		if (entry->info->indev->ifindex == ifindex)
+	if (entry->indev)
+		if (entry->indev->ifindex == ifindex)
 			return 1;
-	if (entry->info->outdev)
-		if (entry->info->outdev->ifindex == ifindex)
+	if (entry->outdev)
+		if (entry->outdev->ifindex == ifindex)
 			return 1;
 #ifdef CONFIG_BRIDGE_NETFILTER
 	if (entry->skb->nf_bridge) {
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 55ea9c6ec744..945e6ae19569 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -57,11 +57,12 @@ struct ip6_rt_info {
 	struct in6_addr saddr;
 };
 
-static void nf_ip6_saveroute(const struct sk_buff *skb, struct nf_info *info)
+static void nf_ip6_saveroute(const struct sk_buff *skb,
+			     struct nf_queue_entry *entry)
 {
-	struct ip6_rt_info *rt_info = nf_info_reroute(info);
+	struct ip6_rt_info *rt_info = nf_queue_entry_reroute(entry);
 
-	if (info->hook == NF_INET_LOCAL_OUT) {
+	if (entry->hook == NF_INET_LOCAL_OUT) {
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 
 		rt_info->daddr = iph->daddr;
@@ -69,11 +70,12 @@ static void nf_ip6_saveroute(const struct sk_buff *skb, struct nf_info *info)
 	}
 }
 
-static int nf_ip6_reroute(struct sk_buff *skb, const struct nf_info *info)
+static int nf_ip6_reroute(struct sk_buff *skb,
+			  const struct nf_queue_entry *entry)
 {
-	struct ip6_rt_info *rt_info = nf_info_reroute(info);
+	struct ip6_rt_info *rt_info = nf_queue_entry_reroute(entry);
 
-	if (info->hook == NF_INET_LOCAL_OUT) {
+	if (entry->hook == NF_INET_LOCAL_OUT) {
 		struct ipv6hdr *iph = ipv6_hdr(skb);
 		if (!ipv6_addr_equal(&iph->daddr, &rt_info->daddr) ||
 		    !ipv6_addr_equal(&iph->saddr, &rt_info->saddr))
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index 9c50cb19b39b..9014adae4fb1 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -39,13 +39,7 @@
 #define NET_IPQ_QMAX 2088
 #define NET_IPQ_QMAX_NAME "ip6_queue_maxlen"
 
-struct ipq_queue_entry {
-	struct list_head list;
-	struct nf_info *info;
-	struct sk_buff *skb;
-};
-
-typedef int (*ipq_cmpfn)(struct ipq_queue_entry *, unsigned long);
+typedef int (*ipq_cmpfn)(struct nf_queue_entry *, unsigned long);
 
 static unsigned char copy_mode __read_mostly = IPQ_COPY_NONE;
 static unsigned int queue_maxlen __read_mostly = IPQ_QMAX_DEFAULT;
@@ -60,16 +54,15 @@ static LIST_HEAD(queue_list);
 static DEFINE_MUTEX(ipqnl_mutex);
 
 static void
-ipq_issue_verdict(struct ipq_queue_entry *entry, int verdict)
+ipq_issue_verdict(struct nf_queue_entry *entry, int verdict)
 {
 	local_bh_disable();
-	nf_reinject(entry->skb, entry->info, verdict);
+	nf_reinject(entry, verdict);
 	local_bh_enable();
-	kfree(entry);
 }
 
 static inline void
-__ipq_enqueue_entry(struct ipq_queue_entry *entry)
+__ipq_enqueue_entry(struct nf_queue_entry *entry)
 {
        list_add_tail(&entry->list, &queue_list);
        queue_total++;
@@ -112,10 +105,10 @@ __ipq_reset(void)
 	__ipq_flush(NULL, 0);
 }
 
-static struct ipq_queue_entry *
+static struct nf_queue_entry *
 ipq_find_dequeue_entry(unsigned long id)
 {
-	struct ipq_queue_entry *entry = NULL, *i;
+	struct nf_queue_entry *entry = NULL, *i;
 
 	write_lock_bh(&queue_lock);
 
@@ -138,7 +131,7 @@ ipq_find_dequeue_entry(unsigned long id)
 static void
 __ipq_flush(ipq_cmpfn cmpfn, unsigned long data)
 {
-	struct ipq_queue_entry *entry, *next;
+	struct nf_queue_entry *entry, *next;
 
 	list_for_each_entry_safe(entry, next, &queue_list, list) {
 		if (!cmpfn || cmpfn(entry, data)) {
@@ -158,7 +151,7 @@ ipq_flush(ipq_cmpfn cmpfn, unsigned long data)
 }
 
 static struct sk_buff *
-ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
+ipq_build_packet_message(struct nf_queue_entry *entry, int *errp)
 {
 	sk_buff_data_t old_tail;
 	size_t size = 0;
@@ -215,20 +208,20 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 	pmsg->timestamp_sec   = tv.tv_sec;
 	pmsg->timestamp_usec  = tv.tv_usec;
 	pmsg->mark            = entry->skb->mark;
-	pmsg->hook            = entry->info->hook;
+	pmsg->hook            = entry->hook;
 	pmsg->hw_protocol     = entry->skb->protocol;
 
-	if (entry->info->indev)
-		strcpy(pmsg->indev_name, entry->info->indev->name);
+	if (entry->indev)
+		strcpy(pmsg->indev_name, entry->indev->name);
 	else
 		pmsg->indev_name[0] = '\0';
 
-	if (entry->info->outdev)
-		strcpy(pmsg->outdev_name, entry->info->outdev->name);
+	if (entry->outdev)
+		strcpy(pmsg->outdev_name, entry->outdev->name);
 	else
 		pmsg->outdev_name[0] = '\0';
 
-	if (entry->info->indev && entry->skb->dev) {
+	if (entry->indev && entry->skb->dev) {
 		pmsg->hw_type = entry->skb->dev->type;
 		pmsg->hw_addrlen = dev_parse_header(entry->skb, pmsg->hw_addr);
 	}
@@ -249,28 +242,17 @@ nlmsg_failure:
 }
 
 static int
-ipq_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		   unsigned int queuenum)
+ipq_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
-	struct ipq_queue_entry *entry;
 
 	if (copy_mode == IPQ_COPY_NONE)
 		return -EAGAIN;
 
-	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
-	if (entry == NULL) {
-		printk(KERN_ERR "ip6_queue: OOM in ipq_enqueue_packet()\n");
-		return -ENOMEM;
-	}
-
-	entry->info = info;
-	entry->skb = skb;
-
 	nskb = ipq_build_packet_message(entry, &status);
 	if (nskb == NULL)
-		goto err_out_free;
+		return status;
 
 	write_lock_bh(&queue_lock);
 
@@ -304,14 +286,11 @@ err_out_free_nskb:
 
 err_out_unlock:
 	write_unlock_bh(&queue_lock);
-
-err_out_free:
-	kfree(entry);
 	return status;
 }
 
 static int
-ipq_mangle_ipv6(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
+ipq_mangle_ipv6(ipq_verdict_msg_t *v, struct nf_queue_entry *e)
 {
 	int diff;
 	int err;
@@ -349,7 +328,7 @@ ipq_mangle_ipv6(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
 static int
 ipq_set_verdict(struct ipq_verdict_msg *vmsg, unsigned int len)
 {
-	struct ipq_queue_entry *entry;
+	struct nf_queue_entry *entry;
 
 	if (vmsg->value > NF_MAX_VERDICT)
 		return -EINVAL;
@@ -409,14 +388,14 @@ ipq_receive_peer(struct ipq_peer_msg *pmsg,
 }
 
 static int
-dev_cmp(struct ipq_queue_entry *entry, unsigned long ifindex)
+dev_cmp(struct nf_queue_entry *entry, unsigned long ifindex)
 {
-	if (entry->info->indev)
-		if (entry->info->indev->ifindex == ifindex)
+	if (entry->indev)
+		if (entry->indev->ifindex == ifindex)
 			return 1;
 
-	if (entry->info->outdev)
-		if (entry->info->outdev->ifindex == ifindex)
+	if (entry->outdev)
+		if (entry->outdev->ifindex == ifindex)
 			return 1;
 #ifdef CONFIG_BRIDGE_NETFILTER
 	if (entry->skb->nf_bridge) {
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index bd71f433b85e..d9d3dc4ce1a3 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -93,7 +93,7 @@ static int __nf_queue(struct sk_buff *skb,
 		      unsigned int queuenum)
 {
 	int status;
-	struct nf_info *info;
+	struct nf_queue_entry *entry;
 #ifdef CONFIG_BRIDGE_NETFILTER
 	struct net_device *physindev = NULL;
 	struct net_device *physoutdev = NULL;
@@ -118,8 +118,8 @@ static int __nf_queue(struct sk_buff *skb,
 		return 1;
 	}
 
-	info = kmalloc(sizeof(*info) + afinfo->route_key_size, GFP_ATOMIC);
-	if (!info) {
+	entry = kmalloc(sizeof(*entry) + afinfo->route_key_size, GFP_ATOMIC);
+	if (!entry) {
 		if (net_ratelimit())
 			printk(KERN_ERR "OOM queueing packet %p\n",
 			       skb);
@@ -128,13 +128,20 @@ static int __nf_queue(struct sk_buff *skb,
 		return 1;
 	}
 
-	*info = (struct nf_info) {
-		(struct nf_hook_ops *)elem, pf, hook, indev, outdev, okfn };
+	*entry = (struct nf_queue_entry) {
+		.skb	= skb,
+		.elem	= list_entry(elem, struct nf_hook_ops, list),
+		.pf	= pf,
+		.hook	= hook,
+		.indev	= indev,
+		.outdev	= outdev,
+		.okfn	= okfn,
+	};
 
 	/* If it's going away, ignore hook. */
-	if (!try_module_get(info->elem->owner)) {
+	if (!try_module_get(entry->elem->owner)) {
 		rcu_read_unlock();
-		kfree(info);
+		kfree(entry);
 		return 0;
 	}
 
@@ -153,8 +160,8 @@ static int __nf_queue(struct sk_buff *skb,
 			dev_hold(physoutdev);
 	}
 #endif
-	afinfo->saveroute(skb, info);
-	status = qh->outfn(skb, info, queuenum);
+	afinfo->saveroute(skb, entry);
+	status = qh->outfn(entry, queuenum);
 
 	rcu_read_unlock();
 
@@ -170,8 +177,8 @@ static int __nf_queue(struct sk_buff *skb,
 		if (physoutdev)
 			dev_put(physoutdev);
 #endif
-		module_put(info->elem->owner);
-		kfree(info);
+		module_put(entry->elem->owner);
+		kfree(entry);
 		kfree_skb(skb);
 
 		return 1;
@@ -220,19 +227,19 @@ int nf_queue(struct sk_buff *skb,
 	return 1;
 }
 
-void nf_reinject(struct sk_buff *skb, struct nf_info *info,
-		 unsigned int verdict)
+void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict)
 {
-	struct list_head *elem = &info->elem->list;
+	struct sk_buff *skb = entry->skb;
+	struct list_head *elem = &entry->elem->list;
 	struct nf_afinfo *afinfo;
 
 	rcu_read_lock();
 
 	/* Release those devices we held, or Alexey will kill me. */
-	if (info->indev)
-		dev_put(info->indev);
-	if (info->outdev)
-		dev_put(info->outdev);
+	if (entry->indev)
+		dev_put(entry->indev);
+	if (entry->outdev)
+		dev_put(entry->outdev);
 #ifdef CONFIG_BRIDGE_NETFILTER
 	if (skb->nf_bridge) {
 		if (skb->nf_bridge->physindev)
@@ -243,7 +250,7 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
 #endif
 
 	/* Drop reference to owner of hook which queued us. */
-	module_put(info->elem->owner);
+	module_put(entry->elem->owner);
 
 	/* Continue traversal iff userspace said ok... */
 	if (verdict == NF_REPEAT) {
@@ -252,28 +259,28 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
 	}
 
 	if (verdict == NF_ACCEPT) {
-		afinfo = nf_get_afinfo(info->pf);
-		if (!afinfo || afinfo->reroute(skb, info) < 0)
+		afinfo = nf_get_afinfo(entry->pf);
+		if (!afinfo || afinfo->reroute(skb, entry) < 0)
 			verdict = NF_DROP;
 	}
 
 	if (verdict == NF_ACCEPT) {
 	next_hook:
-		verdict = nf_iterate(&nf_hooks[info->pf][info->hook],
-				     skb, info->hook,
-				     info->indev, info->outdev, &elem,
-				     info->okfn, INT_MIN);
+		verdict = nf_iterate(&nf_hooks[entry->pf][entry->hook],
+				     skb, entry->hook,
+				     entry->indev, entry->outdev, &elem,
+				     entry->okfn, INT_MIN);
 	}
 
 	switch (verdict & NF_VERDICT_MASK) {
 	case NF_ACCEPT:
 	case NF_STOP:
-		info->okfn(skb);
+		entry->okfn(skb);
 	case NF_STOLEN:
 		break;
 	case NF_QUEUE:
-		if (!__nf_queue(skb, elem, info->pf, info->hook,
-				info->indev, info->outdev, info->okfn,
+		if (!__nf_queue(skb, elem, entry->pf, entry->hook,
+				entry->indev, entry->outdev, entry->okfn,
 				verdict >> NF_VERDICT_BITS))
 			goto next_hook;
 		break;
@@ -281,7 +288,7 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
 		kfree_skb(skb);
 	}
 	rcu_read_unlock();
-	kfree(info);
+	kfree(entry);
 	return;
 }
 EXPORT_SYMBOL(nf_reinject);
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index cb901cf75776..a4937649d006 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -45,13 +45,6 @@
 #define QDEBUG(x, ...)
 #endif
 
-struct nfqnl_queue_entry {
-	struct list_head list;
-	struct nf_info *info;
-	struct sk_buff *skb;
-	unsigned int id;
-};
-
 struct nfqnl_instance {
 	struct hlist_node hlist;		/* global list of queues */
 	atomic_t use;
@@ -73,7 +66,7 @@ struct nfqnl_instance {
 	struct list_head queue_list;		/* packets in queue */
 };
 
-typedef int (*nfqnl_cmpfn)(struct nfqnl_queue_entry *, unsigned long);
+typedef int (*nfqnl_cmpfn)(struct nf_queue_entry *, unsigned long);
 
 static DEFINE_RWLOCK(instances_lock);
 
@@ -212,7 +205,7 @@ instance_destroy(struct nfqnl_instance *inst)
 
 
 static void
-issue_verdict(struct nfqnl_queue_entry *entry, int verdict)
+issue_verdict(struct nf_queue_entry *entry, int verdict)
 {
 	QDEBUG("entering for entry %p, verdict %u\n", entry, verdict);
 
@@ -222,15 +215,12 @@ issue_verdict(struct nfqnl_queue_entry *entry, int verdict)
 	 * softirq, e.g.  We therefore emulate this by local_bh_disable() */
 
 	local_bh_disable();
-	nf_reinject(entry->skb, entry->info, verdict);
+	nf_reinject(entry, verdict);
 	local_bh_enable();
-
-	kfree(entry);
 }
 
 static inline void
-__enqueue_entry(struct nfqnl_instance *queue,
-		      struct nfqnl_queue_entry *entry)
+__enqueue_entry(struct nfqnl_instance *queue, struct nf_queue_entry *entry)
 {
        list_add_tail(&entry->list, &queue->queue_list);
        queue->queue_total++;
@@ -265,10 +255,10 @@ __nfqnl_set_mode(struct nfqnl_instance *queue,
 	return status;
 }
 
-static struct nfqnl_queue_entry *
+static struct nf_queue_entry *
 find_dequeue_entry(struct nfqnl_instance *queue, unsigned int id)
 {
-	struct nfqnl_queue_entry *entry = NULL, *i;
+	struct nf_queue_entry *entry = NULL, *i;
 
 	spin_lock_bh(&queue->lock);
 
@@ -292,7 +282,7 @@ find_dequeue_entry(struct nfqnl_instance *queue, unsigned int id)
 static void
 nfqnl_flush(struct nfqnl_instance *queue, nfqnl_cmpfn cmpfn, unsigned long data)
 {
-	struct nfqnl_queue_entry *entry, *next;
+	struct nf_queue_entry *entry, *next;
 
 	spin_lock_bh(&queue->lock);
 	list_for_each_entry_safe(entry, next, &queue->queue_list, list) {
@@ -307,7 +297,7 @@ nfqnl_flush(struct nfqnl_instance *queue, nfqnl_cmpfn cmpfn, unsigned long data)
 
 static struct sk_buff *
 nfqnl_build_packet_message(struct nfqnl_instance *queue,
-			   struct nfqnl_queue_entry *entry, int *errp)
+			   struct nf_queue_entry *entry, int *errp)
 {
 	sk_buff_data_t old_tail;
 	size_t size;
@@ -316,7 +306,6 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 	struct nfqnl_msg_packet_hdr pmsg;
 	struct nlmsghdr *nlh;
 	struct nfgenmsg *nfmsg;
-	struct nf_info *entinf = entry->info;
 	struct sk_buff *entskb = entry->skb;
 	struct net_device *indev;
 	struct net_device *outdev;
@@ -336,7 +325,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 		+ nla_total_size(sizeof(struct nfqnl_msg_packet_hw))
 		+ nla_total_size(sizeof(struct nfqnl_msg_packet_timestamp));
 
-	outdev = entinf->outdev;
+	outdev = entry->outdev;
 
 	spin_lock_bh(&queue->lock);
 
@@ -379,23 +368,23 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 			NFNL_SUBSYS_QUEUE << 8 | NFQNL_MSG_PACKET,
 			sizeof(struct nfgenmsg));
 	nfmsg = NLMSG_DATA(nlh);
-	nfmsg->nfgen_family = entinf->pf;
+	nfmsg->nfgen_family = entry->pf;
 	nfmsg->version = NFNETLINK_V0;
 	nfmsg->res_id = htons(queue->queue_num);
 
 	pmsg.packet_id 		= htonl(entry->id);
 	pmsg.hw_protocol	= entskb->protocol;
-	pmsg.hook		= entinf->hook;
+	pmsg.hook		= entry->hook;
 
 	NLA_PUT(skb, NFQA_PACKET_HDR, sizeof(pmsg), &pmsg);
 
-	indev = entinf->indev;
+	indev = entry->indev;
 	if (indev) {
 		tmp_uint = htonl(indev->ifindex);
 #ifndef CONFIG_BRIDGE_NETFILTER
 		NLA_PUT(skb, NFQA_IFINDEX_INDEV, sizeof(tmp_uint), &tmp_uint);
 #else
-		if (entinf->pf == PF_BRIDGE) {
+		if (entry->pf == PF_BRIDGE) {
 			/* Case 1: indev is physical input device, we need to
 			 * look for bridge group (when called from
 			 * netfilter_bridge) */
@@ -425,7 +414,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 #ifndef CONFIG_BRIDGE_NETFILTER
 		NLA_PUT(skb, NFQA_IFINDEX_OUTDEV, sizeof(tmp_uint), &tmp_uint);
 #else
-		if (entinf->pf == PF_BRIDGE) {
+		if (entry->pf == PF_BRIDGE) {
 			/* Case 1: outdev is physical output device, we need to
 			 * look for bridge group (when called from
 			 * netfilter_bridge) */
@@ -504,13 +493,11 @@ nla_put_failure:
 }
 
 static int
-nfqnl_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
-		     unsigned int queuenum)
+nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 {
 	int status = -EINVAL;
 	struct sk_buff *nskb;
 	struct nfqnl_instance *queue;
-	struct nfqnl_queue_entry *entry;
 
 	QDEBUG("entered\n");
 
@@ -526,22 +513,11 @@ nfqnl_enqueue_packet(struct sk_buff *skb, struct nf_info *info,
 		goto err_out_put;
 	}
 
-	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
-	if (entry == NULL) {
-		if (net_ratelimit())
-			printk(KERN_ERR
-				"nf_queue: OOM in nfqnl_enqueue_packet()\n");
-		status = -ENOMEM;
-		goto err_out_put;
-	}
-
-	entry->info = info;
-	entry->skb = skb;
 	entry->id = atomic_inc_return(&queue->id_sequence);
 
 	nskb = nfqnl_build_packet_message(queue, entry, &status);
 	if (nskb == NULL)
-		goto err_out_free;
+		goto err_out_put;
 
 	spin_lock_bh(&queue->lock);
 
@@ -577,15 +553,13 @@ err_out_free_nskb:
 err_out_unlock:
 	spin_unlock_bh(&queue->lock);
 
-err_out_free:
-	kfree(entry);
 err_out_put:
 	instance_put(queue);
 	return status;
 }
 
 static int
-nfqnl_mangle(void *data, int data_len, struct nfqnl_queue_entry *e)
+nfqnl_mangle(void *data, int data_len, struct nf_queue_entry *e)
 {
 	int diff;
 	int err;
@@ -630,15 +604,13 @@ nfqnl_set_mode(struct nfqnl_instance *queue,
 }
 
 static int
-dev_cmp(struct nfqnl_queue_entry *entry, unsigned long ifindex)
+dev_cmp(struct nf_queue_entry *entry, unsigned long ifindex)
 {
-	struct nf_info *entinf = entry->info;
-
-	if (entinf->indev)
-		if (entinf->indev->ifindex == ifindex)
+	if (entry->indev)
+		if (entry->indev->ifindex == ifindex)
 			return 1;
-	if (entinf->outdev)
-		if (entinf->outdev->ifindex == ifindex)
+	if (entry->outdev)
+		if (entry->outdev->ifindex == ifindex)
 			return 1;
 #ifdef CONFIG_BRIDGE_NETFILTER
 	if (entry->skb->nf_bridge) {
@@ -748,7 +720,7 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb,
 	struct nfqnl_msg_verdict_hdr *vhdr;
 	struct nfqnl_instance *queue;
 	unsigned int verdict;
-	struct nfqnl_queue_entry *entry;
+	struct nf_queue_entry *entry;
 	int err;
 
 	queue = instance_lookup_get(queue_num);
-- 
cgit v1.2.3


From 33eb9cfc700ae9ce621d47d6ca6d6b4ad7cd97f3 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 5 Dec 2007 01:37:34 -0800
Subject: [NET]: Isolate the net/core/ sysctl table

Using ctl paths we can put all the stuff, related to net/core/
sysctl table, into one file and remove all the references on it.

As a good side effect this hides the "core_table" name from
the global scope :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h         |  4 ----
 net/core/sysctl_net_core.c | 19 ++++++++++++++++++-
 net/sysctl_net.c           |  6 ------
 3 files changed, 18 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 800e73a62d8f..4ce37ce8c411 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1325,10 +1325,6 @@ extern __u32 sysctl_rmem_max;
 
 extern void sk_init(void);
 
-#ifdef CONFIG_SYSCTL
-extern struct ctl_table core_table[];
-#endif
-
 extern int sysctl_optmem_max;
 
 extern __u32 sysctl_wmem_default;
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 277c8faaf732..e322713e590a 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -10,10 +10,11 @@
 #include <linux/module.h>
 #include <linux/socket.h>
 #include <linux/netdevice.h>
+#include <linux/init.h>
 #include <net/sock.h>
 #include <net/xfrm.h>
 
-ctl_table core_table[] = {
+static struct ctl_table net_core_table[] = {
 #ifdef CONFIG_NET
 	{
 		.ctl_name	= NET_CORE_WMEM_MAX,
@@ -149,3 +150,19 @@ ctl_table core_table[] = {
 	},
 	{ .ctl_name = 0 }
 };
+
+static __initdata struct ctl_path net_core_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "core", .ctl_name = NET_CORE, },
+	{ },
+};
+
+static __init int sysctl_core_init(void)
+{
+	struct ctl_table_header *hdr;
+
+	hdr = register_sysctl_paths(net_core_path, net_core_table);
+	return hdr == NULL ? -ENOMEM : 0;
+}
+
+__initcall(sysctl_core_init);
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index c50c793aa7f0..747fc55b28f1 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -31,12 +31,6 @@
 #endif
 
 struct ctl_table net_table[] = {
-	{
-		.ctl_name	= NET_CORE,
-		.procname	= "core",
-		.mode		= 0555,
-		.child		= core_table,
-	},
 #ifdef CONFIG_INET
 	{
 		.ctl_name	= NET_IPV4,
-- 
cgit v1.2.3


From 3e37c3f9978839d91188e4ca0cc662a7245f28e4 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 5 Dec 2007 01:41:26 -0800
Subject: [IPV4]: Use ctl paths to register net/ipv4/ table

This is the same as I did for the net/core/ table in the
second patch in his series: use the paths and isolate the
whole table in the .c file.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h           |  2 --
 net/ipv4/sysctl_net_ipv4.c | 19 ++++++++++++++++++-
 net/sysctl_net.c           |  8 --------
 3 files changed, 18 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 66d51616ade8..17d1189723fc 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -395,6 +395,4 @@ int ipv4_doint_and_flush_strategy(ctl_table *table, int __user *name, int nlen,
 extern int ip_misc_proc_init(void);
 #endif
 
-extern struct ctl_table ipv4_table[];
-
 #endif	/* _IP_H */
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 3546424765f5..bfd0dec6238d 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -13,6 +13,7 @@
 #include <linux/igmp.h>
 #include <linux/inetdevice.h>
 #include <linux/seqlock.h>
+#include <linux/init.h>
 #include <net/snmp.h>
 #include <net/icmp.h>
 #include <net/ip.h>
@@ -247,7 +248,7 @@ static int strategy_allowed_congestion_control(ctl_table *table, int __user *nam
 
 }
 
-ctl_table ipv4_table[] = {
+static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_TCP_TIMESTAMPS,
 		.procname	= "tcp_timestamps",
@@ -878,3 +879,19 @@ ctl_table ipv4_table[] = {
 	},
 	{ .ctl_name = 0 }
 };
+
+static __initdata struct ctl_path net_ipv4_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
+	{ },
+};
+
+static __init int sysctl_ipv4_init(void)
+{
+	struct ctl_table_header *hdr;
+
+	hdr = register_sysctl_paths(net_ipv4_path, ipv4_table);
+	return hdr == NULL ? -ENOMEM : 0;
+}
+
+__initcall(sysctl_ipv4_init);
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index 747fc55b28f1..a4f0ed8d0e79 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -31,14 +31,6 @@
 #endif
 
 struct ctl_table net_table[] = {
-#ifdef CONFIG_INET
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= ipv4_table
-	},
-#endif
 #ifdef CONFIG_TR
 	{
 		.ctl_name	= NET_TR,
-- 
cgit v1.2.3


From 36f0bebd9865dc7e327777fca34b75e65cbfd1a6 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 24 Jan 2008 17:04:49 -0800
Subject: [TR]: Use ctl paths to register net/token-ring/ table

The same thing for token-ring - use ctl paths and get
rid of external references on the tr_table.

Unfortunately, I couldn't split this patch into cleanup and
use-the-paths parts.

As a lame excuse I can say, that the cleanup is just moving
the tr_table from one file to another - closet to a single
variable, that this ctl table tunes. Since the source  file
becomes empty after the move, I remove it.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_tr.h    |  3 ---
 net/802/Makefile         |  3 +--
 net/802/sysctl_net_802.c | 33 ---------------------------------
 net/802/tr.c             | 25 ++++++++++++++++++++++++-
 net/sysctl_net.c         |  8 --------
 5 files changed, 25 insertions(+), 47 deletions(-)
 delete mode 100644 net/802/sysctl_net_802.c

(limited to 'include')

diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h
index 046e9d95ba9a..5bcec8b2c5e2 100644
--- a/include/linux/if_tr.h
+++ b/include/linux/if_tr.h
@@ -49,9 +49,6 @@ static inline struct trh_hdr *tr_hdr(const struct sk_buff *skb)
 {
 	return (struct trh_hdr *)skb_mac_header(skb);
 }
-#ifdef CONFIG_SYSCTL
-extern struct ctl_table tr_table[];
-#endif
 #endif
 
 /* This is an Token-Ring LLC structure */
diff --git a/net/802/Makefile b/net/802/Makefile
index 977704a54f68..68569ffddea1 100644
--- a/net/802/Makefile
+++ b/net/802/Makefile
@@ -3,9 +3,8 @@
 #
 
 # Check the p8022 selections against net/core/Makefile.
-obj-$(CONFIG_SYSCTL)	+= sysctl_net_802.o
 obj-$(CONFIG_LLC)	+= p8022.o psnap.o
-obj-$(CONFIG_TR)	+= p8022.o psnap.o tr.o sysctl_net_802.o
+obj-$(CONFIG_TR)	+= p8022.o psnap.o tr.o
 obj-$(CONFIG_NET_FC)	+=                 fc.o
 obj-$(CONFIG_FDDI)	+=                 fddi.o
 obj-$(CONFIG_HIPPI)	+=                 hippi.o
diff --git a/net/802/sysctl_net_802.c b/net/802/sysctl_net_802.c
deleted file mode 100644
index ead56037398b..000000000000
--- a/net/802/sysctl_net_802.c
+++ /dev/null
@@ -1,33 +0,0 @@
-/* -*- linux-c -*-
- *		sysctl_net_802.c: sysctl interface to net 802 subsystem.
- *
- *		Begun April 1, 1996, Mike Shaver.
- *		Added /proc/sys/net/802 directory entry (empty =) ). [MS]
- *
- *		This program is free software; you can redistribute it and/or
- *		modify it under the terms of the GNU General Public License
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- */
-
-#include <linux/mm.h>
-#include <linux/if_tr.h>
-#include <linux/sysctl.h>
-
-#ifdef CONFIG_TR
-extern int sysctl_tr_rif_timeout;
-#endif
-
-struct ctl_table tr_table[] = {
-#ifdef CONFIG_TR
-	{
-		.ctl_name	= NET_TR_RIF_TIMEOUT,
-		.procname	= "rif_timeout",
-		.data		= &sysctl_tr_rif_timeout,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-#endif /* CONFIG_TR */
-	{ 0 },
-};
diff --git a/net/802/tr.c b/net/802/tr.c
index 151855dd4599..3f16b1720554 100644
--- a/net/802/tr.c
+++ b/net/802/tr.c
@@ -35,6 +35,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/init.h>
+#include <linux/sysctl.h>
 #include <net/arp.h>
 #include <net/net_namespace.h>
 
@@ -634,6 +635,26 @@ struct net_device *alloc_trdev(int sizeof_priv)
 	return alloc_netdev(sizeof_priv, "tr%d", tr_setup);
 }
 
+#ifdef CONFIG_SYSCTL
+static struct ctl_table tr_table[] = {
+	{
+		.ctl_name	= NET_TR_RIF_TIMEOUT,
+		.procname	= "rif_timeout",
+		.data		= &sysctl_tr_rif_timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{ 0 },
+};
+
+static __initdata struct ctl_path tr_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "token-ring", .ctl_name = NET_TR, },
+	{ }
+};
+#endif
+
 /*
  *	Called during bootup.  We don't actually have to initialise
  *	too much for this.
@@ -644,7 +665,9 @@ static int __init rif_init(void)
 	rif_timer.expires  = jiffies + sysctl_tr_rif_timeout;
 	setup_timer(&rif_timer, rif_check_expire, 0);
 	add_timer(&rif_timer);
-
+#ifdef CONFIG_SYSCTL
+	register_sysctl_paths(tr_path, tr_table);
+#endif
 	proc_net_fops_create(&init_net, "tr_rif", S_IRUGO, &rif_seq_fops);
 	return 0;
 }
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index a4f0ed8d0e79..16ad14b5d572 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -31,14 +31,6 @@
 #endif
 
 struct ctl_table net_table[] = {
-#ifdef CONFIG_TR
-	{
-		.ctl_name	= NET_TR,
-		.procname	= "token-ring",
-		.mode		= 0555,
-		.child		= tr_table,
-	},
-#endif
 	{ 0 },
 };
 
-- 
cgit v1.2.3


From 08913681e484f3f0db949dd0809012e089846216 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 5 Dec 2007 01:42:49 -0800
Subject: [NET]: Remove the empty net_table

I have removed all the entries from this table (core_table,
ipv4_table and tr_table), so now we can safely drop it.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h | 1 -
 kernel/sysctl.c     | 8 --------
 net/sysctl_net.c    | 4 ----
 3 files changed, 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index f95f12c5840c..c414d90e647b 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -337,7 +337,6 @@ static const struct proto_ops name##_ops = {			\
 
 #ifdef CONFIG_SYSCTL
 #include <linux/sysctl.h>
-extern ctl_table net_table[];
 extern int net_msg_cost;
 extern int net_msg_burst;
 #endif
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 45e76f209dcb..4bc8e48434a7 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -200,14 +200,6 @@ static struct ctl_table root_table[] = {
 		.mode		= 0555,
 		.child		= vm_table,
 	},
-#ifdef CONFIG_NET
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= net_table,
-	},
-#endif
 	{
 		.ctl_name	= CTL_FS,
 		.procname	= "fs",
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index 16ad14b5d572..665e856675a4 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -30,10 +30,6 @@
 #include <linux/if_tr.h>
 #endif
 
-struct ctl_table net_table[] = {
-	{ 0 },
-};
-
 static struct list_head *
 net_ctl_header_lookup(struct ctl_table_root *root, struct nsproxy *namespaces)
 {
-- 
cgit v1.2.3


From 68dd299bc84dede6aef32e6f4777a676314f5d21 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 5 Dec 2007 01:44:58 -0800
Subject: [INET]: Merge sys.net.ipv4.ip_forward and
 sys.net.ipv4.conf.all.forwarding

AFAIS these two entries should do the same thing - change the
forwarding state on ipv4_devconf and on all the devices.

I propose to merge the handlers together using ctl paths.

The inet_forward_change() is static after this and I move
it higher to be closer to other "propagation" helpers and
to avoid diff making patches based on { and } matching :)
i.e. - make them easier to read.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h |  1 -
 net/ipv4/devinet.c         | 66 ++++++++++++++++++++++++++++++----------------
 net/ipv4/sysctl_net_ipv4.c | 65 ---------------------------------------------
 3 files changed, 44 insertions(+), 88 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index d83fee2dc643..dd093ea4c489 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -135,7 +135,6 @@ extern struct in_device	*inetdev_by_index(int);
 extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
 extern __be32		inet_confirm_addr(const struct net_device *dev, __be32 dst, __be32 local, int scope);
 extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, __be32 mask);
-extern void		inet_forward_change(void);
 
 static __inline__ int inet_ifa_match(__be32 addr, struct in_ifaddr *ifa)
 {
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 9e2747aab252..d1dc0150647d 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1263,6 +1263,28 @@ static void devinet_copy_dflt_conf(int i)
 	read_unlock(&dev_base_lock);
 }
 
+static void inet_forward_change(void)
+{
+	struct net_device *dev;
+	int on = IPV4_DEVCONF_ALL(FORWARDING);
+
+	IPV4_DEVCONF_ALL(ACCEPT_REDIRECTS) = !on;
+	IPV4_DEVCONF_DFLT(FORWARDING) = on;
+
+	read_lock(&dev_base_lock);
+	for_each_netdev(&init_net, dev) {
+		struct in_device *in_dev;
+		rcu_read_lock();
+		in_dev = __in_dev_get_rcu(dev);
+		if (in_dev)
+			IN_DEV_CONF_SET(in_dev, FORWARDING, on);
+		rcu_read_unlock();
+	}
+	read_unlock(&dev_base_lock);
+
+	rt_cache_flush(0);
+}
+
 static int devinet_conf_proc(ctl_table *ctl, int write,
 			     struct file* filp, void __user *buffer,
 			     size_t *lenp, loff_t *ppos)
@@ -1332,28 +1354,6 @@ static int devinet_conf_sysctl(ctl_table *table, int __user *name, int nlen,
 	return 1;
 }
 
-void inet_forward_change(void)
-{
-	struct net_device *dev;
-	int on = IPV4_DEVCONF_ALL(FORWARDING);
-
-	IPV4_DEVCONF_ALL(ACCEPT_REDIRECTS) = !on;
-	IPV4_DEVCONF_DFLT(FORWARDING) = on;
-
-	read_lock(&dev_base_lock);
-	for_each_netdev(&init_net, dev) {
-		struct in_device *in_dev;
-		rcu_read_lock();
-		in_dev = __in_dev_get_rcu(dev);
-		if (in_dev)
-			IN_DEV_CONF_SET(in_dev, FORWARDING, on);
-		rcu_read_unlock();
-	}
-	read_unlock(&dev_base_lock);
-
-	rt_cache_flush(0);
-}
-
 static int devinet_sysctl_forward(ctl_table *ctl, int write,
 				  struct file* filp, void __user *buffer,
 				  size_t *lenp, loff_t *ppos)
@@ -1536,6 +1536,27 @@ static void devinet_sysctl_unregister(struct ipv4_devconf *p)
 }
 #endif
 
+static struct ctl_table ctl_forward_entry[] = {
+	{
+		.ctl_name	= NET_IPV4_FORWARD,
+		.procname	= "ip_forward",
+		.data		= &ipv4_devconf.data[
+					NET_IPV4_CONF_FORWARDING - 1],
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= devinet_sysctl_forward,
+		.strategy	= devinet_conf_sysctl,
+		.extra1		= &ipv4_devconf,
+	},
+	{ },
+};
+
+static __initdata struct ctl_path net_ipv4_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
+	{ },
+};
+
 void __init devinet_init(void)
 {
 	register_gifconf(PF_INET, inet_gifconf);
@@ -1549,6 +1570,7 @@ void __init devinet_init(void)
 			&ipv4_devconf);
 	__devinet_sysctl_register("default", NET_PROTO_CONF_DEFAULT,
 			&ipv4_devconf_dflt);
+	register_sysctl_paths(net_ipv4_path, ctl_forward_entry);
 #endif
 }
 
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index bfd0dec6238d..844f26fab06f 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -27,62 +27,6 @@ static int tcp_retr1_max = 255;
 static int ip_local_port_range_min[] = { 1, 1 };
 static int ip_local_port_range_max[] = { 65535, 65535 };
 
-static
-int ipv4_sysctl_forward(ctl_table *ctl, int write, struct file * filp,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	int val = IPV4_DEVCONF_ALL(FORWARDING);
-	int ret;
-
-	ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
-
-	if (write && IPV4_DEVCONF_ALL(FORWARDING) != val)
-		inet_forward_change();
-
-	return ret;
-}
-
-static int ipv4_sysctl_forward_strategy(ctl_table *table,
-			 int __user *name, int nlen,
-			 void __user *oldval, size_t __user *oldlenp,
-			 void __user *newval, size_t newlen)
-{
-	int *valp = table->data;
-	int new;
-
-	if (!newval || !newlen)
-		return 0;
-
-	if (newlen != sizeof(int))
-		return -EINVAL;
-
-	if (get_user(new, (int __user *)newval))
-		return -EFAULT;
-
-	if (new == *valp)
-		return 0;
-
-	if (oldval && oldlenp) {
-		size_t len;
-
-		if (get_user(len, oldlenp))
-			return -EFAULT;
-
-		if (len) {
-			if (len > table->maxlen)
-				len = table->maxlen;
-			if (copy_to_user(oldval, valp, len))
-				return -EFAULT;
-			if (put_user(len, oldlenp))
-				return -EFAULT;
-		}
-	}
-
-	*valp = new;
-	inet_forward_change();
-	return 1;
-}
-
 extern seqlock_t sysctl_port_range_lock;
 extern int sysctl_local_port_range[2];
 
@@ -281,15 +225,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV4_FORWARD,
-		.procname	= "ip_forward",
-		.data		= &IPV4_DEVCONF_ALL(FORWARDING),
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &ipv4_sysctl_forward,
-		.strategy	= &ipv4_sysctl_forward_strategy
-	},
 	{
 		.ctl_name	= NET_IPV4_DEFAULT_TTL,
 		.procname	= "ip_default_ttl",
-- 
cgit v1.2.3


From 27ab2568649d5ba6c5a20212079b7c4f6da4ca0d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 5 Dec 2007 01:51:58 -0800
Subject: [UDP]: Avoid repeated counting of checksum errors due to peeking

Currently it is possible for two processes to peek on the same socket
and end up incrementing the error counter twice for the same packet.

This patch fixes it by making skb_kill_datagram return whether it
succeeded in unlinking the packet and only incrementing the counter
if it did.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 2 +-
 net/core/datagram.c    | 9 ++++++++-
 net/ipv4/udp.c         | 5 ++---
 net/ipv6/udp.c         | 4 ++--
 4 files changed, 13 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index d39f53ef66bb..17b3f70fbbc3 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1549,7 +1549,7 @@ extern int	       skb_copy_and_csum_datagram_iovec(struct sk_buff *skb,
 							int hlen,
 							struct iovec *iov);
 extern void	       skb_free_datagram(struct sock *sk, struct sk_buff *skb);
-extern void	       skb_kill_datagram(struct sock *sk, struct sk_buff *skb,
+extern int	       skb_kill_datagram(struct sock *sk, struct sk_buff *skb,
 					 unsigned int flags);
 extern __wsum	       skb_checksum(const struct sk_buff *skb, int offset,
 				    int len, __wsum csum);
diff --git a/net/core/datagram.c b/net/core/datagram.c
index 029b93e246b4..fbd6c76436d0 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -217,20 +217,27 @@ void skb_free_datagram(struct sock *sk, struct sk_buff *skb)
  *	This function currently only disables BH when acquiring the
  *	sk_receive_queue lock.  Therefore it must not be used in a
  *	context where that lock is acquired in an IRQ context.
+ *
+ *	It returns 0 if the packet was removed by us.
  */
 
-void skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags)
+int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags)
 {
+	int err = 0;
+
 	if (flags & MSG_PEEK) {
+		err = -ENOENT;
 		spin_lock_bh(&sk->sk_receive_queue.lock);
 		if (skb == skb_peek(&sk->sk_receive_queue)) {
 			__skb_unlink(skb, &sk->sk_receive_queue);
 			atomic_dec(&skb->users);
+			err = 0;
 		}
 		spin_unlock_bh(&sk->sk_receive_queue.lock);
 	}
 
 	kfree_skb(skb);
+	return err;
 }
 
 EXPORT_SYMBOL(skb_kill_datagram);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index d0283b7fcec5..f50de5d5218d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -899,9 +899,8 @@ out:
 	return err;
 
 csum_copy_err:
-	UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
-
-	skb_kill_datagram(sk, skb, flags);
+	if (!skb_kill_datagram(sk, skb, flags))
+		UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
 
 	if (noblock)
 		return -EAGAIN;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 77ab31b99232..87bccec9882a 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -207,8 +207,8 @@ out:
 	return err;
 
 csum_copy_err:
-	UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
-	skb_kill_datagram(sk, skb, flags);
+	if (!skb_kill_datagram(sk, skb, flags))
+		UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
 
 	if (flags & MSG_DONTWAIT)
 		return -EAGAIN;
-- 
cgit v1.2.3


From 1781f7f5804e52ee2d35328b129602146a8d8254 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 11 Dec 2007 11:30:32 -0800
Subject: [UDP]: Restore missing inDatagrams increments

The previous move of the the UDP inDatagrams counter caused the
counting of encapsulated packets, SUNRPC data (as opposed to call)
packets and RXRPC packets to go missing.

This patch restores all of these.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h    |  9 ---------
 include/net/udp.h     | 25 +++++++++++++++++++++++++
 include/net/udplite.h |  3 ---
 net/ipv4/udp.c        |  6 +++++-
 net/ipv6/proc.c       |  1 +
 net/ipv6/udp.c        |  2 ++
 net/rxrpc/ar-input.c  |  4 ++++
 net/sunrpc/xprtsock.c |  6 +++++-
 8 files changed, 42 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e90f9625cb1b..a84f3f697a34 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -164,15 +164,6 @@ DECLARE_SNMP_STAT(struct icmpv6msg_mib, icmpv6msg_statistics);
 #define ICMP6MSGIN_INC_STATS_USER(idev, field) \
 	_DEVINC(icmpv6msg, _USER, idev, field)
 
-DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6);
-DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
-#define UDP6_INC_STATS_BH(field, is_udplite) 			      do  {  \
-	if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field);         \
-	else		SNMP_INC_STATS_BH(udp_stats_in6, field);    } while(0)
-#define UDP6_INC_STATS_USER(field, is_udplite)			       do {    \
-	if (is_udplite) SNMP_INC_STATS_USER(udplite_stats_in6, field);         \
-	else		SNMP_INC_STATS_USER(udp_stats_in6, field);    } while(0)
-
 struct ip6_ra_chain
 {
 	struct ip6_ra_chain	*next;
diff --git a/include/net/udp.h b/include/net/udp.h
index 98755ebaf163..98cb09ca3a27 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -139,6 +139,12 @@ extern int 	udp_lib_setsockopt(struct sock *sk, int level, int optname,
 				   int (*push_pending_frames)(struct sock *));
 
 DECLARE_SNMP_STAT(struct udp_mib, udp_statistics);
+DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6);
+
+/* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */
+DECLARE_SNMP_STAT(struct udp_mib, udplite_statistics);
+DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
+
 /*
  * 	SNMP statistics for UDP and UDP-Lite
  */
@@ -149,6 +155,25 @@ DECLARE_SNMP_STAT(struct udp_mib, udp_statistics);
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field);         \
 	else		SNMP_INC_STATS_BH(udp_statistics, field);    }  while(0)
 
+#define UDP6_INC_STATS_BH(field, is_udplite) 			      do  {  \
+	if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field);         \
+	else		SNMP_INC_STATS_BH(udp_stats_in6, field);    } while(0)
+#define UDP6_INC_STATS_USER(field, is_udplite)			       do {    \
+	if (is_udplite) SNMP_INC_STATS_USER(udplite_stats_in6, field);         \
+	else		SNMP_INC_STATS_USER(udp_stats_in6, field);    } while(0)
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+#define UDPX_INC_STATS_BH(sk, field) \
+	do { \
+		if ((sk)->sk_family == AF_INET) \
+			UDP_INC_STATS_BH(field, 0); \
+		else \
+			UDP6_INC_STATS_BH(field, 0); \
+	} while (0);
+#else
+#define UDPX_INC_STATS_BH(sk, field) UDP_INC_STATS_BH(field, 0)
+#endif
+
 /* /proc */
 struct udp_seq_afinfo {
 	struct module		*owner;
diff --git a/include/net/udplite.h b/include/net/udplite.h
index 635b0eafca95..b76b2e377af4 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -13,9 +13,6 @@
 extern struct proto 		udplite_prot;
 extern struct hlist_head 	udplite_hash[UDP_HTABLE_SIZE];
 
-/* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */
-DECLARE_SNMP_STAT(struct udp_mib, udplite_statistics);
-
 /*
  *	Checksum computation is all in software, hence simpler getfrag.
  */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index f50de5d5218d..78cfcb4a1b3f 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -110,6 +110,7 @@
  */
 
 DEFINE_SNMP_STAT(struct udp_mib, udp_statistics) __read_mostly;
+EXPORT_SYMBOL(udp_statistics);
 
 struct hlist_head udp_hash[UDP_HTABLE_SIZE];
 DEFINE_RWLOCK(udp_hash_lock);
@@ -969,8 +970,11 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 			int ret;
 
 			ret = (*up->encap_rcv)(sk, skb);
-			if (ret <= 0)
+			if (ret <= 0) {
+				UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS,
+						 is_udplite);
 				return -ret;
+			}
 		}
 
 		/* FALLTHROUGH -- it's a UDP Packet */
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 44937616057e..41e9980b3e0e 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -27,6 +27,7 @@
 #include <net/ip.h>
 #include <net/sock.h>
 #include <net/tcp.h>
+#include <net/udp.h>
 #include <net/transp_v6.h>
 #include <net/ipv6.h>
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 87bccec9882a..36bdcd2e1b52 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -34,6 +34,7 @@
 #include <linux/ipv6.h>
 #include <linux/icmpv6.h>
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/skbuff.h>
 #include <asm/uaccess.h>
 
@@ -51,6 +52,7 @@
 #include "udp_impl.h"
 
 DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly;
+EXPORT_SYMBOL(udp_stats_in6);
 
 static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
 {
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index 91b5bbb003e2..f446d9b9925f 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -20,6 +20,7 @@
 #include <net/sock.h>
 #include <net/af_rxrpc.h>
 #include <net/ip.h>
+#include <net/udp.h>
 #include "ar-internal.h"
 
 unsigned long rxrpc_ack_timeout = 1;
@@ -707,10 +708,13 @@ void rxrpc_data_ready(struct sock *sk, int count)
 	if (skb_checksum_complete(skb)) {
 		rxrpc_free_skb(skb);
 		rxrpc_put_local(local);
+		UDP_INC_STATS_BH(UDP_MIB_INERRORS, 0);
 		_leave(" [CSUM failed]");
 		return;
 	}
 
+	UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, 0);
+
 	/* the socket buffer we have is owned by UDP, with UDP's data all over
 	 * it, but we really want our own */
 	skb_orphan(skb);
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 2f630a512ab7..6fa52f44de0f 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -838,8 +838,12 @@ static void xs_udp_data_ready(struct sock *sk, int len)
 		copied = repsize;
 
 	/* Suck it into the iovec, verify checksum if not done by hw. */
-	if (csum_partial_copy_to_xdr(&rovr->rq_private_buf, skb))
+	if (csum_partial_copy_to_xdr(&rovr->rq_private_buf, skb)) {
+		UDPX_INC_STATS_BH(sk, UDP_MIB_INERRORS);
 		goto out_unlock;
+	}
+
+	UDPX_INC_STATS_BH(sk, UDP_MIB_INDATAGRAMS);
 
 	/* Something worked... */
 	dst_confirm(skb->dst);
-- 
cgit v1.2.3


From a59322be07c964e916d15be3df473fb7ba20c41e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 5 Dec 2007 01:53:40 -0800
Subject: [UDP]: Only increment counter on first peek/recv

The previous move of the the UDP inDatagrams counter caused each
peek of the same packet to be counted separately.  This may be
undesirable.

This patch fixes this by adding a bit to sk_buff to record whether
this packet has already been seen through skb_recv_datagram.  We
then only increment the counter when the packet is seen for the
first time.

The only dodgy part is the fact that skb_recv_datagram doesn't have
a good way of returning this new bit of information.  So I've added
a new function __skb_recv_datagram that does return this and made
skb_recv_datagram a wrapper around it.

The plan is to eventually replace all uses of skb_recv_datagram with
this new function at which time it can be renamed its proper name.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h |  3 +++
 net/core/datagram.c    | 43 +++++++++++++++++++++++++++----------------
 net/ipv4/udp.c         |  7 +++++--
 net/ipv6/udp.c         |  7 +++++--
 4 files changed, 40 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 17b3f70fbbc3..c618fbf7d173 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -288,6 +288,7 @@ struct sk_buff {
 	__u8			pkt_type:3,
 				fclone:2,
 				ipvs_property:1,
+				peeked:1,
 				nf_trace:1;
 	__be16			protocol;
 
@@ -1538,6 +1539,8 @@ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
 		     skb = skb->prev)
 
 
+extern struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
+					   int *peeked, int *err);
 extern struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
 					 int noblock, int *err);
 extern unsigned int    datagram_poll(struct file *file, struct socket *sock,
diff --git a/net/core/datagram.c b/net/core/datagram.c
index fbd6c76436d0..2d733131d7ce 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -115,10 +115,10 @@ out_noerr:
 }
 
 /**
- *	skb_recv_datagram - Receive a datagram skbuff
+ *	__skb_recv_datagram - Receive a datagram skbuff
  *	@sk: socket
  *	@flags: MSG_ flags
- *	@noblock: blocking operation?
+ *	@peeked: returns non-zero if this packet has been seen before
  *	@err: error code returned
  *
  *	Get a datagram skbuff, understands the peeking, nonblocking wakeups
@@ -143,8 +143,8 @@ out_noerr:
  *	quite explicitly by POSIX 1003.1g, don't change them without having
  *	the standard around please.
  */
-struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
-				  int noblock, int *err)
+struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
+				    int *peeked, int *err)
 {
 	struct sk_buff *skb;
 	long timeo;
@@ -156,7 +156,7 @@ struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
 	if (error)
 		goto no_packet;
 
-	timeo = sock_rcvtimeo(sk, noblock);
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 
 	do {
 		/* Again only user level code calls this function, so nothing
@@ -165,18 +165,19 @@ struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
 		 * Look at current nfs client by the way...
 		 * However, this function was corrent in any case. 8)
 		 */
-		if (flags & MSG_PEEK) {
-			unsigned long cpu_flags;
-
-			spin_lock_irqsave(&sk->sk_receive_queue.lock,
-					  cpu_flags);
-			skb = skb_peek(&sk->sk_receive_queue);
-			if (skb)
+		unsigned long cpu_flags;
+
+		spin_lock_irqsave(&sk->sk_receive_queue.lock, cpu_flags);
+		skb = skb_peek(&sk->sk_receive_queue);
+		if (skb) {
+			*peeked = skb->peeked;
+			if (flags & MSG_PEEK) {
+				skb->peeked = 1;
 				atomic_inc(&skb->users);
-			spin_unlock_irqrestore(&sk->sk_receive_queue.lock,
-					       cpu_flags);
-		} else
-			skb = skb_dequeue(&sk->sk_receive_queue);
+			} else
+				__skb_unlink(skb, &sk->sk_receive_queue);
+		}
+		spin_unlock_irqrestore(&sk->sk_receive_queue.lock, cpu_flags);
 
 		if (skb)
 			return skb;
@@ -194,6 +195,16 @@ no_packet:
 	*err = error;
 	return NULL;
 }
+EXPORT_SYMBOL(__skb_recv_datagram);
+
+struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
+				  int noblock, int *err)
+{
+	int peeked;
+
+	return __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+				   &peeked, err);
+}
 
 void skb_free_datagram(struct sock *sk, struct sk_buff *skb)
 {
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 78cfcb4a1b3f..9ed6393c65d9 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -827,6 +827,7 @@ int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
 	struct sk_buff *skb;
 	unsigned int ulen, copied;
+	int peeked;
 	int err;
 	int is_udplite = IS_UDPLITE(sk);
 
@@ -840,7 +841,8 @@ int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		return ip_recv_error(sk, msg, len);
 
 try_again:
-	skb = skb_recv_datagram(sk, flags, noblock, &err);
+	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+				  &peeked, &err);
 	if (!skb)
 		goto out;
 
@@ -875,7 +877,8 @@ try_again:
 	if (err)
 		goto out_free;
 
-	UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+	if (!peeked)
+		UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
 
 	sock_recv_timestamp(msg, sk, skb);
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 36bdcd2e1b52..fa640765385e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -123,6 +123,7 @@ int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 	struct inet_sock *inet = inet_sk(sk);
 	struct sk_buff *skb;
 	unsigned int ulen, copied;
+	int peeked;
 	int err;
 	int is_udplite = IS_UDPLITE(sk);
 
@@ -133,7 +134,8 @@ int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 		return ipv6_recv_error(sk, msg, len);
 
 try_again:
-	skb = skb_recv_datagram(sk, flags, noblock, &err);
+	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+				  &peeked, &err);
 	if (!skb)
 		goto out;
 
@@ -166,7 +168,8 @@ try_again:
 	if (err)
 		goto out_free;
 
-	UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+	if (!peeked)
+		UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
 
 	sock_recv_timestamp(msg, sk, skb);
 
-- 
cgit v1.2.3


From f4d900a2cae94256f56be7769734100c7054bf00 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 5 Dec 2007 03:31:53 -0800
Subject: [NETLINK]: Mark attribute construction exception unlikely

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h | 2 +-
 include/net/netlink.h   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index d5bfaba595c7..2aee0f510876 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -245,7 +245,7 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len, int flags)
 }
 
 #define NLMSG_NEW(skb, pid, seq, type, len, flags) \
-({	if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) \
+({	if (unlikely(skb_tailroom(skb) < (int)NLMSG_SPACE(len))) \
 		goto nlmsg_failure; \
 	__nlmsg_put(skb, pid, seq, type, len, flags); })
 
diff --git a/include/net/netlink.h b/include/net/netlink.h
index 9298218c07f9..db4b935b6c7e 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -862,7 +862,7 @@ static inline int nla_put_msecs(struct sk_buff *skb, int attrtype,
 
 #define NLA_PUT(skb, attrtype, attrlen, data) \
 	do { \
-		if (nla_put(skb, attrtype, attrlen, data) < 0) \
+		if (unlikely(nla_put(skb, attrtype, attrlen, data) < 0)) \
 			goto nla_put_failure; \
 	} while(0)
 
-- 
cgit v1.2.3


From d63bddbe90c4fd924b2155ca92a879393d856170 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Dec 2007 00:40:34 -0800
Subject: [IPV6]: Make fib6_init to return an error code.

If there is an error in the initialization function, nothing is
followed up to the caller. So I add a return value to be set for the
init function.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h |  2 +-
 net/ipv6/ip6_fib.c    | 14 +++++++++++---
 2 files changed, 12 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 4cefcffe773e..5d39ce92afcf 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -224,7 +224,7 @@ extern void			fib6_run_gc(unsigned long dummy);
 
 extern void			fib6_gc_cleanup(void);
 
-extern void			fib6_init(void);
+extern int			fib6_init(void);
 
 extern void			fib6_rules_init(void);
 extern void			fib6_rules_cleanup(void);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 31b60a02512a..c100b44f2b87 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1473,16 +1473,24 @@ void fib6_run_gc(unsigned long dummy)
 	spin_unlock_bh(&fib6_gc_lock);
 }
 
-void __init fib6_init(void)
+int __init fib6_init(void)
 {
+	int ret;
 	fib6_node_kmem = kmem_cache_create("fib6_nodes",
 					   sizeof(struct fib6_node),
 					   0, SLAB_HWCACHE_ALIGN|SLAB_PANIC,
 					   NULL);
-
 	fib6_tables_init();
 
-	__rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
+	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
+	if (ret)
+		goto out_kmem_cache_create;
+out:
+	return ret;
+
+out_kmem_cache_create:
+	kmem_cache_destroy(fib6_node_kmem);
+	goto out;
 }
 
 void fib6_gc_cleanup(void)
-- 
cgit v1.2.3


From 0013cabab30ec55830ce63d34c0bdd887eb87644 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Dec 2007 00:42:11 -0800
Subject: [IPV6]: Make xfrm6_init to return an error code.

The xfrm initialization function does not return any error code, so if
there is an error, the caller can not be advise of that.  This patch
checks the return code of the different called functions in order to
return a successful or failed initialization.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  4 ++--
 net/ipv6/xfrm6_policy.c | 22 +++++++++++++++++-----
 net/ipv6/xfrm6_state.c  |  4 ++--
 3 files changed, 21 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 182f10b29d88..18260921eeb5 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1066,11 +1066,11 @@ struct xfrm6_tunnel {
 
 extern void xfrm_init(void);
 extern void xfrm4_init(void);
-extern void xfrm6_init(void);
+extern int xfrm6_init(void);
 extern void xfrm6_fini(void);
 extern void xfrm_state_init(void);
 extern void xfrm4_state_init(void);
-extern void xfrm6_state_init(void);
+extern int xfrm6_state_init(void);
 extern void xfrm6_state_fini(void);
 
 extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *);
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 4d54951cea08..181cf91538f1 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -269,9 +269,9 @@ static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
 	.fill_dst =		xfrm6_fill_dst,
 };
 
-static void __init xfrm6_policy_init(void)
+static int __init xfrm6_policy_init(void)
 {
-	xfrm_policy_register_afinfo(&xfrm6_policy_afinfo);
+	return xfrm_policy_register_afinfo(&xfrm6_policy_afinfo);
 }
 
 static void xfrm6_policy_fini(void)
@@ -279,10 +279,22 @@ static void xfrm6_policy_fini(void)
 	xfrm_policy_unregister_afinfo(&xfrm6_policy_afinfo);
 }
 
-void __init xfrm6_init(void)
+int __init xfrm6_init(void)
 {
-	xfrm6_policy_init();
-	xfrm6_state_init();
+	int ret;
+
+	ret = xfrm6_policy_init();
+	if (ret)
+		goto out;
+
+	ret = xfrm6_state_init();
+	if (ret)
+		goto out_policy;
+out:
+	return ret;
+out_policy:
+	xfrm6_policy_fini();
+	goto out;
 }
 
 void xfrm6_fini(void)
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index a7a7e8fd6a33..dc817e035e23 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -198,9 +198,9 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.transport_finish	= xfrm6_transport_finish,
 };
 
-void __init xfrm6_state_init(void)
+int __init xfrm6_state_init(void)
 {
-	xfrm_state_register_afinfo(&xfrm6_state_afinfo);
+	return xfrm_state_register_afinfo(&xfrm6_state_afinfo);
 }
 
 void xfrm6_state_fini(void)
-- 
cgit v1.2.3


From 9eb87f3f7e0686a256c5bb4f886dede0171245f2 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Dec 2007 00:42:52 -0800
Subject: [IPV6]: Make fib6_rules_init to return an error code.

When the fib_rules initialization finished, no return code is provided
so there is no way to know, for the caller, if the initialization has
been successful or has failed. This patch fix that.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h |  1 +
 include/net/ip6_fib.h   |  2 +-
 net/core/fib_rules.c    |  5 +++--
 net/ipv6/fib6_rules.c   | 19 ++++++++++++++++---
 4 files changed, 21 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 41a301e38643..2364db1a47e6 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -103,6 +103,7 @@ static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla)
 
 extern int			fib_rules_register(struct fib_rules_ops *);
 extern int			fib_rules_unregister(struct fib_rules_ops *);
+extern void                     fib_rules_cleanup_ops(struct fib_rules_ops *);
 
 extern int			fib_rules_lookup(struct fib_rules_ops *,
 						 struct flowi *, int flags,
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 5d39ce92afcf..eaa315868792 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -226,7 +226,7 @@ extern void			fib6_gc_cleanup(void);
 
 extern int			fib6_init(void);
 
-extern void			fib6_rules_init(void);
+extern int			fib6_rules_init(void);
 extern void			fib6_rules_cleanup(void);
 
 #endif
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 0af0538343da..fcbf41c0a5d4 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -102,7 +102,7 @@ errout:
 
 EXPORT_SYMBOL_GPL(fib_rules_register);
 
-static void cleanup_ops(struct fib_rules_ops *ops)
+void fib_rules_cleanup_ops(struct fib_rules_ops *ops)
 {
 	struct fib_rule *rule, *tmp;
 
@@ -111,6 +111,7 @@ static void cleanup_ops(struct fib_rules_ops *ops)
 		fib_rule_put(rule);
 	}
 }
+EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops);
 
 int fib_rules_unregister(struct fib_rules_ops *ops)
 {
@@ -121,7 +122,7 @@ int fib_rules_unregister(struct fib_rules_ops *ops)
 	list_for_each_entry(o, &rules_ops, list) {
 		if (o == ops) {
 			list_del_rcu(&o->list);
-			cleanup_ops(ops);
+			fib_rules_cleanup_ops(ops);
 			goto out;
 		}
 	}
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 428c6b0e26d8..9ce2e0a6748a 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -265,10 +265,23 @@ static int __init fib6_default_rules_init(void)
 	return 0;
 }
 
-void __init fib6_rules_init(void)
+int __init fib6_rules_init(void)
 {
-	BUG_ON(fib6_default_rules_init());
-	fib_rules_register(&fib6_rules_ops);
+	int ret;
+
+	ret = fib6_default_rules_init();
+	if (ret)
+		goto out;
+
+	ret = fib_rules_register(&fib6_rules_ops);
+	if (ret)
+		goto out_default_rules_init;
+out:
+	return ret;
+
+out_default_rules_init:
+	fib_rules_cleanup_ops(&fib6_rules_ops);
+	goto out;
 }
 
 void fib6_rules_cleanup(void)
-- 
cgit v1.2.3


From 433d49c3bb14b8a2351fe97df8359e4ad0de4c7c Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Dec 2007 00:43:48 -0800
Subject: [IPV6]: Make ip6_route_init to return an error code.

The route initialization function does not return any value to notify
if the initialization is successful or not. This patch checks all
calls made for the initilization in order to return a value for the
caller.

Unfortunately, proc_net_fops_create will return a NULL pointer if
CONFIG_PROC_FS is off, so we can not check the return code without an
ifdef CONFIG_PROC_FS block in the ip6_route_init function.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  2 +-
 net/ipv6/route.c        | 64 ++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 54 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 5456fdd6d047..1a0698b1e86c 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -50,7 +50,7 @@ extern void			ip6_route_input(struct sk_buff *skb);
 extern struct dst_entry *	ip6_route_output(struct sock *sk,
 						 struct flowi *fl);
 
-extern void			ip6_route_init(void);
+extern int			ip6_route_init(void);
 extern void			ip6_route_cleanup(void);
 
 extern int			ipv6_route_ioctl(unsigned int cmd, void __user *arg);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 452111fa4c52..d7754abf9216 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2468,26 +2468,70 @@ ctl_table ipv6_route_table[] = {
 
 #endif
 
-void __init ip6_route_init(void)
+int __init ip6_route_init(void)
 {
+	int ret;
+
 	ip6_dst_ops.kmem_cachep =
 		kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0,
 				  SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
 	ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep;
 
-	fib6_init();
-	proc_net_fops_create(&init_net, "ipv6_route", 0, &ipv6_route_proc_fops);
-	proc_net_fops_create(&init_net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops);
+	ret = fib6_init();
+	if (ret)
+		goto out_kmem_cache;
+
+#ifdef CONFIG_PROC_FS
+	ret = -ENOMEM;
+	if (!proc_net_fops_create(&init_net, "ipv6_route",
+				  0, &ipv6_route_proc_fops))
+		goto out_fib6_init;
+
+	if (!proc_net_fops_create(&init_net, "rt6_stats",
+				  S_IRUGO, &rt6_stats_seq_fops))
+		goto out_proc_ipv6_route;
+#endif
+
 #ifdef CONFIG_XFRM
-	xfrm6_init();
+	ret = xfrm6_init();
+	if (ret)
+		goto out_proc_rt6_stats;
 #endif
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	fib6_rules_init();
+	ret = fib6_rules_init();
+	if (ret)
+		goto xfrm6_init;
 #endif
+	ret = -ENOBUFS;
+	if (__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL) ||
+	    __rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL) ||
+	    __rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL))
+		goto fib6_rules_init;
 
-	__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL);
-	__rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL);
-	__rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL);
+	ret = 0;
+out:
+	return ret;
+
+fib6_rules_init:
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	fib6_rules_cleanup();
+xfrm6_init:
+#endif
+#ifdef CONFIG_XFRM
+	xfrm6_fini();
+out_proc_rt6_stats:
+#endif
+#ifdef CONFIG_PROC_FS
+	proc_net_remove(&init_net, "rt6_stats");
+out_proc_ipv6_route:
+	proc_net_remove(&init_net, "ipv6_route");
+out_fib6_init:
+#endif
+	rt6_ifdown(NULL);
+	fib6_gc_cleanup();
+out_kmem_cache:
+	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
+	goto out;
 }
 
 void ip6_route_cleanup(void)
@@ -2495,10 +2539,8 @@ void ip6_route_cleanup(void)
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	fib6_rules_cleanup();
 #endif
-#ifdef CONFIG_PROC_FS
 	proc_net_remove(&init_net, "ipv6_route");
 	proc_net_remove(&init_net, "rt6_stats");
-#endif
 #ifdef CONFIG_XFRM
 	xfrm6_fini();
 #endif
-- 
cgit v1.2.3


From bb803175865dd030420eebbc87298414c7725019 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Fri, 7 Dec 2007 00:46:11 -0800
Subject: [IPV4]: Remove ip_fib_local_table and ip_fib_main_table defines.

There are only 2 users and it doesn't hurt to call fib_get_table
instead, and it makes it easier to make the fib network namespace
aware.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h | 3 ---
 net/ipv4/fib_hash.c  | 5 +++--
 2 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index ed514bfb61ba..690fb4dd7505 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -191,9 +191,6 @@ extern void __init fib4_rules_init(void);
 extern u32 fib_rules_tclass(struct fib_result *res);
 #endif
 
-#define ip_fib_local_table fib_get_table(RT_TABLE_LOCAL)
-#define ip_fib_main_table fib_get_table(RT_TABLE_MAIN)
-
 extern int fib_lookup(struct flowi *flp, struct fib_result *res);
 
 extern struct fib_table *fib_new_table(u32 id);
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index eb1aa8eead09..638185eb315f 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -811,7 +811,8 @@ struct fib_iter_state {
 static struct fib_alias *fib_get_first(struct seq_file *seq)
 {
 	struct fib_iter_state *iter = seq->private;
-	struct fn_hash *table = (struct fn_hash *) ip_fib_main_table->tb_data;
+	struct fib_table *main_table = fib_get_table(RT_TABLE_MAIN);
+	struct fn_hash *table = (struct fn_hash *)main_table->tb_data;
 
 	iter->bucket    = 0;
 	iter->hash_head = NULL;
@@ -950,7 +951,7 @@ static void *fib_seq_start(struct seq_file *seq, loff_t *pos)
 	void *v = NULL;
 
 	read_lock(&fib_hash_lock);
-	if (ip_fib_main_table)
+	if (fib_get_table(RT_TABLE_MAIN))
 		v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 	return v;
 }
-- 
cgit v1.2.3


From cbbb90e68cd073b8d63b491166066e347902b7e9 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 7 Dec 2007 23:56:57 -0800
Subject: [SNMP]: Remove unused devconf macros.

The SNMP_INC_STATS_OFFSET_BH is used only by ICMP6_INC_STATS_OFFSET_BH.
The ICMP6_INC_STATS_OFFSET_BH is unused.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h | 8 --------
 include/net/snmp.h | 2 --
 2 files changed, 10 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index a84f3f697a34..38df94b5fb26 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -143,14 +143,6 @@ DECLARE_SNMP_STAT(struct icmpv6msg_mib, icmpv6msg_statistics);
 #define ICMP6_INC_STATS_BH(idev, field)	_DEVINC(icmpv6, _BH, idev, field)
 #define ICMP6_INC_STATS_USER(idev, field) _DEVINC(icmpv6, _USER, idev, field)
 
-#define ICMP6_INC_STATS_OFFSET_BH(idev, field, offset)	({			\
-	struct inet6_dev *_idev = idev;						\
-	__typeof__(offset) _offset = (offset);					\
-	if (likely(_idev != NULL))						\
-		SNMP_INC_STATS_OFFSET_BH(_idev->stats.icmpv6, field, _offset);	\
-	SNMP_INC_STATS_OFFSET_BH(icmpv6_statistics, field, _offset);    	\
-})
-
 #define ICMP6MSGOUT_INC_STATS(idev, field) \
 	_DEVINC(icmpv6msg, , idev, field +256)
 #define ICMP6MSGOUT_INC_STATS_BH(idev, field) \
diff --git a/include/net/snmp.h b/include/net/snmp.h
index ea206bff0dc4..9c5793db637f 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -134,8 +134,6 @@ struct linux_mib {
 
 #define SNMP_INC_STATS_BH(mib, field) 	\
 	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field]++)
-#define SNMP_INC_STATS_OFFSET_BH(mib, field, offset)	\
-	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field + (offset)]++)
 #define SNMP_INC_STATS_USER(mib, field) \
 	(per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field]++)
 #define SNMP_INC_STATS(mib, field) 	\
-- 
cgit v1.2.3


From 024626e36d75fc8c6e32d50d4c68bfc3b8df5fdf Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 8 Dec 2007 00:09:24 -0800
Subject: [NET] sysctl: make the sys.net.core sysctls per-namespace

Making them per-namespace is required for the following
two reasons:

 First, some ctl values have a per-namespace meaning.
 Second, making them writable from the sub-namespace
 is an isolation hole.

So I introduce the pernet operations to create these
tables. For init_net I use the existing statically
declared tables, for sub-namespace they are duplicated
and the write bits are removed from the mode.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  3 +++
 net/core/sysctl_net_core.c  | 50 ++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 48 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index f97b2a4469ae..d5936115d972 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -37,6 +37,9 @@ struct net {
 
 	struct sock 		*rtnl;			/* rtnetlink socket */
 
+	/* core sysctls */
+	struct ctl_table_header	*sysctl_core_hdr;
+
 	/* List of all packet sockets. */
 	rwlock_t		packet_sklist_lock;
 	struct hlist_head	packet_sklist;
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index e322713e590a..57a7eadb8551 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -151,18 +151,58 @@ static struct ctl_table net_core_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static __initdata struct ctl_path net_core_path[] = {
+static __net_initdata struct ctl_path net_core_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "core", .ctl_name = NET_CORE, },
 	{ },
 };
 
-static __init int sysctl_core_init(void)
+static __net_init int sysctl_core_net_init(struct net *net)
 {
-	struct ctl_table_header *hdr;
+	struct ctl_table *tbl, *tmp;
+
+	tbl = net_core_table;
+	if (net != &init_net) {
+		tbl = kmemdup(tbl, sizeof(net_core_table), GFP_KERNEL);
+		if (tbl == NULL)
+			goto err_dup;
+
+		for (tmp = tbl; tmp->procname; tmp++)
+			tmp->mode &= ~0222;
+	}
+
+	net->sysctl_core_hdr = register_net_sysctl_table(net,
+			net_core_path, tbl);
+	if (net->sysctl_core_hdr == NULL)
+		goto err_reg;
 
-	hdr = register_sysctl_paths(net_core_path, net_core_table);
-	return hdr == NULL ? -ENOMEM : 0;
+	return 0;
+
+err_reg:
+	if (tbl != net_core_table)
+		kfree(tbl);
+err_dup:
+	return -ENOMEM;
+}
+
+static __net_exit void sysctl_core_net_exit(struct net *net)
+{
+	struct ctl_table *tbl;
+
+	tbl = net->sysctl_core_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->sysctl_core_hdr);
+	BUG_ON(tbl == net_core_table);
+	kfree(tbl);
+}
+
+static __net_initdata struct pernet_operations sysctl_core_ops = {
+	.init = sysctl_core_net_init,
+	.exit = sysctl_core_net_exit,
+};
+
+static __init int sysctl_core_init(void)
+{
+	return register_pernet_subsys(&sysctl_core_ops);
 }
 
 __initcall(sysctl_core_init);
-- 
cgit v1.2.3


From b8e1f9b5c37e77cc8f978a58859b35fe5edd5542 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 8 Dec 2007 00:12:33 -0800
Subject: [NET] sysctl: make sysctl_somaxconn per-namespace

Just move the variable on the struct net and adjust
its usage.

Others sysctls from sys.net.core table are more
difficult to virtualize (i.e. make them per-namespace),
but I'll look at them as well a bit later.

Signed-off-by: Pavel Emelyanov <xemul@oenvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/socket.h      | 1 -
 include/net/net_namespace.h | 1 +
 net/core/sysctl_net_core.c  | 4 +++-
 net/socket.c                | 8 ++++----
 4 files changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/socket.h b/include/linux/socket.h
index eb5bdd59a644..bd2b30a74e76 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -24,7 +24,6 @@ struct __kernel_sockaddr_storage {
 #include <linux/types.h>		/* pid_t			*/
 #include <linux/compiler.h>		/* __user			*/
 
-extern int sysctl_somaxconn;
 #ifdef CONFIG_PROC_FS
 struct seq_file;
 extern void socket_seq_show(struct seq_file *seq);
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index d5936115d972..b62e31fca474 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -39,6 +39,7 @@ struct net {
 
 	/* core sysctls */
 	struct ctl_table_header	*sysctl_core_hdr;
+	int			sysctl_somaxconn;
 
 	/* List of all packet sockets. */
 	rwlock_t		packet_sklist_lock;
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index dc4cf7dda9d1..130338f83ae5 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -127,7 +127,7 @@ static struct ctl_table net_core_table[] = {
 	{
 		.ctl_name	= NET_CORE_SOMAXCONN,
 		.procname	= "somaxconn",
-		.data		= &sysctl_somaxconn,
+		.data		= &init_net.sysctl_somaxconn,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -161,6 +161,8 @@ static __net_init int sysctl_core_net_init(struct net *net)
 {
 	struct ctl_table *tbl, *tmp;
 
+	net->sysctl_somaxconn = SOMAXCONN;
+
 	tbl = net_core_table;
 	if (net != &init_net) {
 		tbl = kmemdup(tbl, sizeof(net_core_table), GFP_KERNEL);
diff --git a/net/socket.c b/net/socket.c
index 9ebca5c695d6..7651de008502 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1365,17 +1365,17 @@ asmlinkage long sys_bind(int fd, struct sockaddr __user *umyaddr, int addrlen)
  *	ready for listening.
  */
 
-int sysctl_somaxconn __read_mostly = SOMAXCONN;
-
 asmlinkage long sys_listen(int fd, int backlog)
 {
 	struct socket *sock;
 	int err, fput_needed;
+	int somaxconn;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (sock) {
-		if ((unsigned)backlog > sysctl_somaxconn)
-			backlog = sysctl_somaxconn;
+		somaxconn = sock->sk->sk_net->sysctl_somaxconn;
+		if ((unsigned)backlog > somaxconn)
+			backlog = somaxconn;
 
 		err = security_socket_listen(sock, backlog);
 		if (!err)
-- 
cgit v1.2.3


From c35b7e72cd48bc7163b6900fb3689fa54b572bba Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Sat, 8 Dec 2007 00:14:11 -0800
Subject: [IPV6]: remove ifdef in route6 for xfrm6

The following patch create the usual static inline functions to disable
the xfrm6_init and xfrm6_fini function when XFRM is off.
That's allow to remove some ifdef and make the code a little more clear.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h | 16 +++++++++++++---
 net/ipv6/route.c   |  7 +------
 2 files changed, 14 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 18260921eeb5..503d0d294a4c 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -842,7 +842,6 @@ xfrm_state_addr_cmp(struct xfrm_tmpl *tmpl, struct xfrm_state *x, unsigned short
 }
 
 #ifdef CONFIG_XFRM
-
 extern int __xfrm_policy_check(struct sock *, int dir, struct sk_buff *skb, unsigned short family);
 
 static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, unsigned short family)
@@ -1066,12 +1065,23 @@ struct xfrm6_tunnel {
 
 extern void xfrm_init(void);
 extern void xfrm4_init(void);
-extern int xfrm6_init(void);
-extern void xfrm6_fini(void);
 extern void xfrm_state_init(void);
 extern void xfrm4_state_init(void);
+#ifdef CONFIG_XFRM
+extern int xfrm6_init(void);
+extern void xfrm6_fini(void);
 extern int xfrm6_state_init(void);
 extern void xfrm6_state_fini(void);
+#else
+static inline int xfrm6_init(void)
+{
+	return 0;
+}
+static inline void xfrm6_fini(void)
+{
+	;
+}
+#endif
 
 extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *);
 extern struct xfrm_state *xfrm_state_alloc(void);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index dbdae143ef5f..c4e890abb2d4 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2522,11 +2522,10 @@ int __init ip6_route_init(void)
 	if (ret)
 		goto out_fib6_init;
 
-#ifdef CONFIG_XFRM
 	ret = xfrm6_init();
 	if (ret)
 		goto out_proc_init;
-#endif
+
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	ret = fib6_rules_init();
 	if (ret)
@@ -2547,9 +2546,7 @@ fib6_rules_init:
 	fib6_rules_cleanup();
 xfrm6_init:
 #endif
-#ifdef CONFIG_XFRM
 	xfrm6_fini();
-#endif
 out_proc_init:
 	ipv6_route_proc_fini(&init_net);
 out_fib6_init:
@@ -2566,9 +2563,7 @@ void ip6_route_cleanup(void)
 	fib6_rules_cleanup();
 #endif
 	ipv6_route_proc_fini(&init_net);
-#ifdef CONFIG_XFRM
 	xfrm6_fini();
-#endif
 	rt6_ifdown(NULL);
 	fib6_gc_cleanup();
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
-- 
cgit v1.2.3


From 7e5449c21562f1554d2c355db1ec9d3e4f434288 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Sat, 8 Dec 2007 00:14:54 -0800
Subject: [IPV6]: route6 remove ifdef for fib_rules

The patch defines the usual static inline functions when the code is
disabled for fib6_rules. That's allow to remove some ifdef in route.c
file and make the code a little more clear.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h | 12 +++++++++++-
 net/ipv6/route.c      |  7 +------
 2 files changed, 12 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index eaa315868792..14830edc2ac0 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -226,8 +226,18 @@ extern void			fib6_gc_cleanup(void);
 
 extern int			fib6_init(void);
 
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
 extern int			fib6_rules_init(void);
 extern void			fib6_rules_cleanup(void);
-
+#else
+static inline int               fib6_rules_init(void)
+{
+	return 0;
+}
+static inline void              fib6_rules_cleanup(void)
+{
+	return ;
+}
+#endif
 #endif
 #endif
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index c4e890abb2d4..11ef456d67c5 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2526,11 +2526,10 @@ int __init ip6_route_init(void)
 	if (ret)
 		goto out_proc_init;
 
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	ret = fib6_rules_init();
 	if (ret)
 		goto xfrm6_init;
-#endif
+
 	ret = -ENOBUFS;
 	if (__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL) ||
 	    __rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL) ||
@@ -2542,10 +2541,8 @@ out:
 	return ret;
 
 fib6_rules_init:
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	fib6_rules_cleanup();
 xfrm6_init:
-#endif
 	xfrm6_fini();
 out_proc_init:
 	ipv6_route_proc_fini(&init_net);
@@ -2559,9 +2556,7 @@ out_kmem_cache:
 
 void ip6_route_cleanup(void)
 {
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	fib6_rules_cleanup();
-#endif
 	ipv6_route_proc_fini(&init_net);
 	xfrm6_fini();
 	rt6_ifdown(NULL);
-- 
cgit v1.2.3


From 971b893e79db0f7dccfcea15dbdebca3ca64a84d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 8 Dec 2007 00:32:23 -0800
Subject: [IPV4]: last default route is a fib table property

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h |  1 +
 net/ipv4/fib_hash.c  | 16 ++++++++--------
 net/ipv4/fib_trie.c  | 18 +++++++++---------
 3 files changed, 18 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 690fb4dd7505..d70b9b49f8b1 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -141,6 +141,7 @@ struct fib_table {
 	struct hlist_node tb_hlist;
 	u32		tb_id;
 	unsigned	tb_stamp;
+	int		tb_default;
 	int		(*tb_lookup)(struct fib_table *tb, const struct flowi *flp, struct fib_result *res);
 	int		(*tb_insert)(struct fib_table *, struct fib_config *);
 	int		(*tb_delete)(struct fib_table *, struct fib_config *);
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 86087d45c64e..9d5404158476 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -272,8 +272,6 @@ out:
 	return err;
 }
 
-static int fn_hash_last_dflt=-1;
-
 static void
 fn_hash_select_default(struct fib_table *tb, const struct flowi *flp, struct fib_result *res)
 {
@@ -314,9 +312,9 @@ fn_hash_select_default(struct fib_table *tb, const struct flowi *flp, struct fib
 				if (next_fi != res->fi)
 					break;
 			} else if (!fib_detect_death(fi, order, &last_resort,
-						     &last_idx, fn_hash_last_dflt)) {
+						&last_idx, tb->tb_default)) {
 				fib_result_assign(res, fi);
-				fn_hash_last_dflt = order;
+				tb->tb_default = order;
 				goto out;
 			}
 			fi = next_fi;
@@ -325,19 +323,20 @@ fn_hash_select_default(struct fib_table *tb, const struct flowi *flp, struct fib
 	}
 
 	if (order <= 0 || fi == NULL) {
-		fn_hash_last_dflt = -1;
+		tb->tb_default = -1;
 		goto out;
 	}
 
-	if (!fib_detect_death(fi, order, &last_resort, &last_idx, fn_hash_last_dflt)) {
+	if (!fib_detect_death(fi, order, &last_resort, &last_idx,
+				tb->tb_default)) {
 		fib_result_assign(res, fi);
-		fn_hash_last_dflt = order;
+		tb->tb_default = order;
 		goto out;
 	}
 
 	if (last_idx >= 0)
 		fib_result_assign(res, last_resort);
-	fn_hash_last_dflt = last_idx;
+	tb->tb_default = last_idx;
 out:
 	read_unlock(&fib_hash_lock);
 }
@@ -773,6 +772,7 @@ struct fib_table * __init fib_hash_init(u32 id)
 		return NULL;
 
 	tb->tb_id = id;
+	tb->tb_default = -1;
 	tb->tb_lookup = fn_hash_lookup;
 	tb->tb_insert = fn_hash_insert;
 	tb->tb_delete = fn_hash_delete;
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index c7c5c6c802df..9126eea71670 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1782,8 +1782,6 @@ static int fn_trie_flush(struct fib_table *tb)
 	return found;
 }
 
-static int trie_last_dflt = -1;
-
 static void
 fn_trie_select_default(struct fib_table *tb, const struct flowi *flp, struct fib_result *res)
 {
@@ -1830,28 +1828,29 @@ fn_trie_select_default(struct fib_table *tb, const struct flowi *flp, struct fib
 			if (next_fi != res->fi)
 				break;
 		} else if (!fib_detect_death(fi, order, &last_resort,
-					     &last_idx, trie_last_dflt)) {
+					     &last_idx, tb->tb_default)) {
 			fib_result_assign(res, fi);
-			trie_last_dflt = order;
+			tb->tb_default = order;
 			goto out;
 		}
 		fi = next_fi;
 		order++;
 	}
 	if (order <= 0 || fi == NULL) {
-		trie_last_dflt = -1;
+		tb->tb_default = -1;
 		goto out;
 	}
 
-	if (!fib_detect_death(fi, order, &last_resort, &last_idx, trie_last_dflt)) {
+	if (!fib_detect_death(fi, order, &last_resort, &last_idx,
+				tb->tb_default)) {
 		fib_result_assign(res, fi);
-		trie_last_dflt = order;
+		tb->tb_default = order;
 		goto out;
 	}
 	if (last_idx >= 0)
 		fib_result_assign(res, last_resort);
-	trie_last_dflt = last_idx;
- out:;
+	tb->tb_default = last_idx;
+out:
 	rcu_read_unlock();
 }
 
@@ -1978,6 +1977,7 @@ struct fib_table * __init fib_hash_init(u32 id)
 		return NULL;
 
 	tb->tb_id = id;
+	tb->tb_default = -1;
 	tb->tb_lookup = fn_trie_lookup;
 	tb->tb_insert = fn_trie_insert;
 	tb->tb_delete = fn_trie_delete;
-- 
cgit v1.2.3


From c69bce20dda7f79160856a338298d65a284ba303 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 23 Jan 2008 22:31:45 -0800
Subject: [NET]: Remove unused "mibalign" argument for snmp_mib_init().

With fixes from Arnaldo Carvalho de Melo.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h    |  2 +-
 net/ipv4/af_inet.c  | 23 ++++++++---------------
 net/ipv6/addrconf.c |  9 +++------
 net/ipv6/af_inet6.c | 17 ++++++++---------
 4 files changed, 20 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 17d1189723fc..571bcf2a1cff 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -171,7 +171,7 @@ DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_ADD_STATS_USER(field, adnd)	SNMP_ADD_STATS_USER(net_statistics, field, adnd)
 
 extern unsigned long snmp_fold_field(void *mib[], int offt);
-extern int snmp_mib_init(void *ptr[2], size_t mibsize, size_t mibalign);
+extern int snmp_mib_init(void *ptr[2], size_t mibsize);
 extern void snmp_mib_free(void *ptr[2]);
 
 extern void inet_get_local_port_range(int *low, int *high);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 0e4b6eba2340..7f8b27ff94ff 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1242,7 +1242,7 @@ unsigned long snmp_fold_field(void *mib[], int offt)
 }
 EXPORT_SYMBOL_GPL(snmp_fold_field);
 
-int snmp_mib_init(void *ptr[2], size_t mibsize, size_t mibalign)
+int snmp_mib_init(void *ptr[2], size_t mibsize)
 {
 	BUG_ON(ptr == NULL);
 	ptr[0] = __alloc_percpu(mibsize);
@@ -1296,32 +1296,25 @@ static struct net_protocol icmp_protocol = {
 static int __init init_ipv4_mibs(void)
 {
 	if (snmp_mib_init((void **)net_statistics,
-			  sizeof(struct linux_mib),
-			  __alignof__(struct linux_mib)) < 0)
+			  sizeof(struct linux_mib)) < 0)
 		goto err_net_mib;
 	if (snmp_mib_init((void **)ip_statistics,
-			  sizeof(struct ipstats_mib),
-			  __alignof__(struct ipstats_mib)) < 0)
+			  sizeof(struct ipstats_mib)) < 0)
 		goto err_ip_mib;
 	if (snmp_mib_init((void **)icmp_statistics,
-			  sizeof(struct icmp_mib),
-			  __alignof__(struct icmp_mib)) < 0)
+			  sizeof(struct icmp_mib)) < 0)
 		goto err_icmp_mib;
 	if (snmp_mib_init((void **)icmpmsg_statistics,
-			  sizeof(struct icmpmsg_mib),
-			  __alignof__(struct icmpmsg_mib)) < 0)
+			  sizeof(struct icmpmsg_mib)) < 0)
 		goto err_icmpmsg_mib;
 	if (snmp_mib_init((void **)tcp_statistics,
-			  sizeof(struct tcp_mib),
-			  __alignof__(struct tcp_mib)) < 0)
+			  sizeof(struct tcp_mib)) < 0)
 		goto err_tcp_mib;
 	if (snmp_mib_init((void **)udp_statistics,
-			  sizeof(struct udp_mib),
-			  __alignof__(struct udp_mib)) < 0)
+			  sizeof(struct udp_mib)) < 0)
 		goto err_udp_mib;
 	if (snmp_mib_init((void **)udplite_statistics,
-			  sizeof(struct udp_mib),
-			  __alignof__(struct udp_mib)) < 0)
+			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
 
 	tcp_mib_init();
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 95cf3aa41c98..ba6f7925c178 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -256,16 +256,13 @@ static void addrconf_mod_timer(struct inet6_ifaddr *ifp,
 static int snmp6_alloc_dev(struct inet6_dev *idev)
 {
 	if (snmp_mib_init((void **)idev->stats.ipv6,
-			  sizeof(struct ipstats_mib),
-			  __alignof__(struct ipstats_mib)) < 0)
+			  sizeof(struct ipstats_mib)) < 0)
 		goto err_ip;
 	if (snmp_mib_init((void **)idev->stats.icmpv6,
-			  sizeof(struct icmpv6_mib),
-			  __alignof__(struct icmpv6_mib)) < 0)
+			  sizeof(struct icmpv6_mib)) < 0)
 		goto err_icmp;
 	if (snmp_mib_init((void **)idev->stats.icmpv6msg,
-			  sizeof(struct icmpv6msg_mib),
-			  __alignof__(struct icmpv6msg_mib)) < 0)
+			  sizeof(struct icmpv6msg_mib)) < 0)
 		goto err_icmpmsg;
 
 	return 0;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 5ab8ba7a5863..90d2f723fc46 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -714,20 +714,19 @@ EXPORT_SYMBOL_GPL(ipv6_opt_accepted);
 
 static int __init init_ipv6_mibs(void)
 {
-	if (snmp_mib_init((void **)ipv6_statistics, sizeof (struct ipstats_mib),
-			  __alignof__(struct ipstats_mib)) < 0)
+	if (snmp_mib_init((void **)ipv6_statistics,
+			  sizeof(struct ipstats_mib)) < 0)
 		goto err_ip_mib;
-	if (snmp_mib_init((void **)icmpv6_statistics, sizeof (struct icmpv6_mib),
-			  __alignof__(struct icmpv6_mib)) < 0)
+	if (snmp_mib_init((void **)icmpv6_statistics,
+			  sizeof(struct icmpv6_mib)) < 0)
 		goto err_icmp_mib;
 	if (snmp_mib_init((void **)icmpv6msg_statistics,
-	    sizeof (struct icmpv6msg_mib), __alignof__(struct icmpv6_mib)) < 0)
+			  sizeof(struct icmpv6msg_mib)) < 0)
 		goto err_icmpmsg_mib;
-	if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib),
-			  __alignof__(struct udp_mib)) < 0)
+	if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0)
 		goto err_udp_mib;
-	if (snmp_mib_init((void **)udplite_stats_in6, sizeof (struct udp_mib),
-			  __alignof__(struct udp_mib)) < 0)
+	if (snmp_mib_init((void **)udplite_stats_in6,
+			  sizeof (struct udp_mib)) < 0)
 		goto err_udplite_mib;
 	return 0;
 
-- 
cgit v1.2.3


From 005011211f559113686938c2c252b8ee1ab855b5 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 11 Dec 2007 01:53:43 -0800
Subject: [IPSEC]: Add xfrm_input_state helper

This patch adds the xfrm_input_state helper function which returns the
current xfrm state being processed on the input path given an sk_buff.
This is currently only used by xfrm_input but will be used by ESP upon
asynchronous resumption.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    | 5 +++++
 net/xfrm/xfrm_input.c | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 503d0d294a4c..fe881b6e2bd0 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1317,4 +1317,9 @@ static inline void xfrm_states_delete(struct xfrm_state **states, int n)
 }
 #endif
 
+static inline struct xfrm_state *xfrm_input_state(struct sk_buff *skb)
+{
+	return skb->sp->xvec[skb->sp->len - 1];
+}
+
 #endif	/* _NET_XFRM_H */
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 8b2b1b59133e..8624cbdb2a1e 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -109,7 +109,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	/* A negative encap_type indicates async resumption. */
 	if (encap_type < 0) {
 		async = 1;
-		x = skb->sp->xvec[skb->sp->len - 1];
+		x = xfrm_input_state(skb);
 		seq = XFRM_SKB_CB(skb)->seq;
 		goto resume;
 	}
-- 
cgit v1.2.3


From 01ecfe9ba63aa0f23bbdb15371916ba4d5382330 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 11 Dec 2007 02:16:47 -0800
Subject: [IPV4]: Cleanup IN_DEV_MFORWARD macro

This is essentially IN_DEV_ANDCONF with proper arguments.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index dd093ea4c489..962a062b44ff 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -78,9 +78,7 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev)
 	(max(IPV4_DEVCONF_ALL(attr), IN_DEV_CONF_GET((in_dev), attr)))
 
 #define IN_DEV_FORWARD(in_dev)		IN_DEV_CONF_GET((in_dev), FORWARDING)
-#define IN_DEV_MFORWARD(in_dev)		(IPV4_DEVCONF_ALL(MC_FORWARDING) && \
-					 IPV4_DEVCONF((in_dev)->cnf, \
-						      MC_FORWARDING))
+#define IN_DEV_MFORWARD(in_dev)		IN_DEV_ANDCONF((in_dev), MC_FORWARDING)
 #define IN_DEV_RPFILTER(in_dev)		IN_DEV_ANDCONF((in_dev), RP_FILTER)
 #define IN_DEV_SOURCE_ROUTE(in_dev)	IN_DEV_ANDCONF((in_dev), \
 						       ACCEPT_SOURCE_ROUTE)
-- 
cgit v1.2.3


From 0a3e78ac2c555441f5bc00588070058533bc8d6b Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 11 Dec 2007 02:23:18 -0800
Subject: [IPV6]: make flowlabel to return an error

This patch makes the flowlab subsystem to return an error code and makes
some cleanup with procfs ifdefs.
The af_inet6 will use the flowlabel init return code to check the initialization
was correct.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h       |  2 +-
 net/ipv6/af_inet6.c      |  5 ++++-
 net/ipv6/ip6_flowlabel.c | 30 +++++++++++++++++++++++-------
 3 files changed, 28 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 38df94b5fb26..4d9106580950 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -219,7 +219,7 @@ extern struct ipv6_txoptions	*fl6_merge_options(struct ipv6_txoptions * opt_spac
 						   struct ipv6_txoptions * fopt);
 extern void			fl6_free_socklist(struct sock *sk);
 extern int			ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen);
-extern void			ip6_flowlabel_init(void);
+extern int			ip6_flowlabel_init(void);
 extern void			ip6_flowlabel_cleanup(void);
 
 static inline void fl6_sock_release(struct ip6_flowlabel *fl)
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 90d2f723fc46..614f3d905dd1 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -851,7 +851,9 @@ static int __init inet6_init(void)
 	err = ip6_route_init();
 	if (err)
 		goto ip6_route_fail;
-	ip6_flowlabel_init();
+	err = ip6_flowlabel_init();
+	if (err)
+		goto ip6_flowlabel_fail;
 	err = addrconf_init();
 	if (err)
 		goto addrconf_fail;
@@ -874,6 +876,7 @@ out:
 
 addrconf_fail:
 	ip6_flowlabel_cleanup();
+ip6_flowlabel_fail:
 	ip6_route_cleanup();
 ip6_route_fail:
 #ifdef CONFIG_PROC_FS
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index b12cc22e7745..d0babea89819 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -692,20 +692,36 @@ static const struct file_operations ip6fl_seq_fops = {
 	.llseek		=	seq_lseek,
 	.release	=	seq_release_private,
 };
-#endif
 
+static int ip6_flowlabel_proc_init(struct net *net)
+{
+	if (!proc_net_fops_create(net, "ip6_flowlabel", S_IRUGO, &ip6fl_seq_fops))
+		return -ENOMEM;
+	return 0;
+}
 
-void ip6_flowlabel_init(void)
+static void ip6_flowlabel_proc_fini(struct net *net)
 {
-#ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "ip6_flowlabel", S_IRUGO, &ip6fl_seq_fops);
+	proc_net_remove(net, "ip6_flowlabel");
+}
+#else
+static inline int ip6_flowlabel_proc_init(struct net *net)
+{
+	return 0;
+}
+static inline void ip6_flowlabel_proc_fini(struct net *net)
+{
+	return ;
+}
 #endif
+
+int ip6_flowlabel_init(void)
+{
+	return ip6_flowlabel_proc_init(&init_net);
 }
 
 void ip6_flowlabel_cleanup(void)
 {
 	del_timer(&ip6_fl_gc_timer);
-#ifdef CONFIG_PROC_FS
-	proc_net_remove(&init_net, "ip6_flowlabel");
-#endif
+	ip6_flowlabel_proc_fini(&init_net);
 }
-- 
cgit v1.2.3


From 248b238dc960a42aa235057ba0a51a98ae2b0f0d Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 11 Dec 2007 02:23:54 -0800
Subject: [IPV6]: make extended headers to return an error at initialization

This patch factorize the code for the differents init functions for rthdr,
nodata, destopt in a single function exthdrs_init.
This function returns an error so the af_inet6 module can check correctly
the initialization.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/transp_v6.h |  5 ++--
 net/ipv6/af_inet6.c     | 10 +++++---
 net/ipv6/exthdrs.c      | 64 ++++++++++++++++++++++++++++++-------------------
 3 files changed, 48 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 409da3a9a455..610b1bb775c9 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -17,10 +17,9 @@ extern struct proto tcpv6_prot;
 struct flowi;
 
 /* extention headers */
-extern void				ipv6_rthdr_init(void);
+extern int				ipv6_exthdrs_init(void);
+extern void				ipv6_exthdrs_exit(void);
 extern void				ipv6_frag_init(void);
-extern void				ipv6_nodata_init(void);
-extern void				ipv6_destopt_init(void);
 
 /* transport protocols */
 extern void				rawv6_init(void);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 614f3d905dd1..442c298c1d7c 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -859,10 +859,11 @@ static int __init inet6_init(void)
 		goto addrconf_fail;
 
 	/* Init v6 extension headers. */
-	ipv6_rthdr_init();
+	err = ipv6_exthdrs_init();
+	if (err)
+		goto ipv6_exthdrs_fail;
+
 	ipv6_frag_init();
-	ipv6_nodata_init();
-	ipv6_destopt_init();
 
 	/* Init v6 transport protocols. */
 	udpv6_init();
@@ -874,6 +875,8 @@ static int __init inet6_init(void)
 out:
 	return err;
 
+ipv6_exthdrs_fail:
+	addrconf_cleanup();
 addrconf_fail:
 	ip6_flowlabel_cleanup();
 ip6_flowlabel_fail:
@@ -932,6 +935,7 @@ static void __exit inet6_exit(void)
 	/* Cleanup code parts. */
 	ipv6_packet_cleanup();
 
+	ipv6_exthdrs_exit();
 	addrconf_cleanup();
 	ip6_flowlabel_cleanup();
 	ip6_route_cleanup();
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index cee06b1655c1..2df34ed276f1 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -308,28 +308,6 @@ static int ipv6_destopt_rcv(struct sk_buff *skb)
 	return -1;
 }
 
-static struct inet6_protocol destopt_protocol = {
-	.handler	=	ipv6_destopt_rcv,
-	.flags		=	INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR,
-};
-
-void __init ipv6_destopt_init(void)
-{
-	if (inet6_add_protocol(&destopt_protocol, IPPROTO_DSTOPTS) < 0)
-		printk(KERN_ERR "ipv6_destopt_init: Could not register protocol\n");
-}
-
-static struct inet6_protocol nodata_protocol = {
-	.handler	=	dst_discard,
-	.flags		=	INET6_PROTO_NOPOLICY,
-};
-
-void __init ipv6_nodata_init(void)
-{
-	if (inet6_add_protocol(&nodata_protocol, IPPROTO_NONE) < 0)
-		printk(KERN_ERR "ipv6_nodata_init: Could not register protocol\n");
-}
-
 /********************************
   Routing header.
  ********************************/
@@ -527,12 +505,48 @@ static struct inet6_protocol rthdr_protocol = {
 	.flags		=	INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR,
 };
 
-void __init ipv6_rthdr_init(void)
+static struct inet6_protocol destopt_protocol = {
+	.handler	=	ipv6_destopt_rcv,
+	.flags		=	INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR,
+};
+
+static struct inet6_protocol nodata_protocol = {
+	.handler	=	dst_discard,
+	.flags		=	INET6_PROTO_NOPOLICY,
+};
+
+int __init ipv6_exthdrs_init(void)
 {
-	if (inet6_add_protocol(&rthdr_protocol, IPPROTO_ROUTING) < 0)
-		printk(KERN_ERR "ipv6_rthdr_init: Could not register protocol\n");
+	int ret;
+
+	ret = inet6_add_protocol(&rthdr_protocol, IPPROTO_ROUTING);
+	if (ret)
+		goto out;
+
+	ret = inet6_add_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+	if (ret)
+		goto out_rthdr;
+
+	ret = inet6_add_protocol(&nodata_protocol, IPPROTO_NONE);
+	if (ret)
+		goto out_destopt;
+
+out:
+	return ret;
+out_rthdr:
+	inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
+out_destopt:
+	inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+	goto out;
 };
 
+void ipv6_exthdrs_exit(void)
+{
+	inet6_del_protocol(&nodata_protocol, IPPROTO_NONE);
+	inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+	inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
+}
+
 /**********************************
   Hop-by-hop options.
  **********************************/
-- 
cgit v1.2.3


From 853cbbaaa4ccdf221be5ab6afe967aa9998546b7 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 11 Dec 2007 02:24:29 -0800
Subject: [IPV6]: make frag to return an error at initialization

This patch makes the frag_init to return an error code, so the af_inet6
module can handle the error.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/transp_v6.h |  3 ++-
 net/ipv6/af_inet6.c     |  8 ++++++--
 net/ipv6/reassembly.c   | 16 +++++++++++++---
 3 files changed, 21 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 610b1bb775c9..aa9a4a6b99df 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -19,7 +19,8 @@ struct flowi;
 /* extention headers */
 extern int				ipv6_exthdrs_init(void);
 extern void				ipv6_exthdrs_exit(void);
-extern void				ipv6_frag_init(void);
+extern int				ipv6_frag_init(void);
+extern void				ipv6_frag_exit(void);
 
 /* transport protocols */
 extern void				rawv6_init(void);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 442c298c1d7c..a75c4bc9281a 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -863,7 +863,9 @@ static int __init inet6_init(void)
 	if (err)
 		goto ipv6_exthdrs_fail;
 
-	ipv6_frag_init();
+	err = ipv6_frag_init();
+	if (err)
+		goto ipv6_frag_fail;
 
 	/* Init v6 transport protocols. */
 	udpv6_init();
@@ -875,6 +877,8 @@ static int __init inet6_init(void)
 out:
 	return err;
 
+ipv6_frag_fail:
+	ipv6_exthdrs_exit();
 ipv6_exthdrs_fail:
 	addrconf_cleanup();
 addrconf_fail:
@@ -934,7 +938,7 @@ static void __exit inet6_exit(void)
 
 	/* Cleanup code parts. */
 	ipv6_packet_cleanup();
-
+	ipv6_frag_exit();
 	ipv6_exthdrs_exit();
 	addrconf_cleanup();
 	ip6_flowlabel_cleanup();
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 76c88a93b9b5..bf4173daecbb 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -632,11 +632,13 @@ static struct inet6_protocol frag_protocol =
 	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
-void __init ipv6_frag_init(void)
+int __init ipv6_frag_init(void)
 {
-	if (inet6_add_protocol(&frag_protocol, IPPROTO_FRAGMENT) < 0)
-		printk(KERN_ERR "ipv6_frag_init: Could not register protocol\n");
+	int ret;
 
+	ret = inet6_add_protocol(&frag_protocol, IPPROTO_FRAGMENT);
+	if (ret)
+		goto out;
 	ip6_frags.ctl = &ip6_frags_ctl;
 	ip6_frags.hashfn = ip6_hashfn;
 	ip6_frags.constructor = ip6_frag_init;
@@ -646,4 +648,12 @@ void __init ipv6_frag_init(void)
 	ip6_frags.match = ip6_frag_match;
 	ip6_frags.frag_expire = ip6_frag_expire;
 	inet_frags_init(&ip6_frags);
+out:
+	return ret;
+}
+
+void ipv6_frag_exit(void)
+{
+	inet_frags_fini(&ip6_frags);
+	inet6_del_protocol(&frag_protocol, IPPROTO_FRAGMENT);
 }
-- 
cgit v1.2.3


From 87c3efbfdd1f98af14a1f60ff19f73d9a8d8da98 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 11 Dec 2007 02:25:01 -0800
Subject: [IPV6]: make inet6_register_protosw to return an error code

This patch makes the inet6_register_protosw to return an error code.
The different protocols can be aware the registration was successful or
not and can pass the error to the initial caller, af_inet6.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/protocol.h |  2 +-
 net/ipv6/af_inet6.c    | 11 +++++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/protocol.h b/include/net/protocol.h
index 1166ffb4b3ec..ad8c584233a6 100644
--- a/include/net/protocol.h
+++ b/include/net/protocol.h
@@ -102,7 +102,7 @@ extern void	inet_unregister_protosw(struct inet_protosw *p);
 #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
 extern int	inet6_add_protocol(struct inet6_protocol *prot, unsigned char num);
 extern int	inet6_del_protocol(struct inet6_protocol *prot, unsigned char num);
-extern void	inet6_register_protosw(struct inet_protosw *p);
+extern int	inet6_register_protosw(struct inet_protosw *p);
 extern void	inet6_unregister_protosw(struct inet_protosw *p);
 #endif
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index a75c4bc9281a..53b06de696bd 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -565,21 +565,23 @@ static struct inet_protosw rawv6_protosw = {
 	.flags		= INET_PROTOSW_REUSE,
 };
 
-void
-inet6_register_protosw(struct inet_protosw *p)
+int inet6_register_protosw(struct inet_protosw *p)
 {
 	struct list_head *lh;
 	struct inet_protosw *answer;
-	int protocol = p->protocol;
 	struct list_head *last_perm;
+	int protocol = p->protocol;
+	int ret;
 
 	spin_lock_bh(&inetsw6_lock);
 
+	ret = -EINVAL;
 	if (p->type >= SOCK_MAX)
 		goto out_illegal;
 
 	/* If we are trying to override a permanent protocol, bail. */
 	answer = NULL;
+	ret = -EPERM;
 	last_perm = &inetsw6[p->type];
 	list_for_each(lh, &inetsw6[p->type]) {
 		answer = list_entry(lh, struct inet_protosw, list);
@@ -603,9 +605,10 @@ inet6_register_protosw(struct inet_protosw *p)
 	 * system automatically returns to the old behavior.
 	 */
 	list_add_rcu(&p->list, last_perm);
+	ret = 0;
 out:
 	spin_unlock_bh(&inetsw6_lock);
-	return;
+	return ret;
 
 out_permanent:
 	printk(KERN_ERR "Attempt to override permanent protocol %d.\n",
-- 
cgit v1.2.3


From 7f4e4868f3ce0e946f116c28fa4fe033be5e4ba9 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 11 Dec 2007 02:25:35 -0800
Subject: [IPV6]: make the protocol initialization to return an error code

This patchset makes the different protocols to return an error code, so
the af_inet6 module can check the initialization was correct or not.

The raw6 was taken into account to be consistent with the rest of the
protocols, but the registration is at the same place.
Because the raw6 has its own init function, the proto and the ops structure
can be moved inside the raw6.c file.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h       |  2 +-
 include/net/transp_v6.h  | 12 +++++---
 net/ipv6/af_inet6.c      | 77 +++++++++++++++++++++---------------------------
 net/ipv6/ipv6_sockglue.c |  3 +-
 net/ipv6/raw.c           | 52 ++++++++++++++++++++++++++++++++
 net/ipv6/tcp_ipv6.c      | 36 +++++++++++++++++-----
 net/ipv6/udp.c           | 26 +++++++++++++---
 net/ipv6/udplite.c       | 25 +++++++++++++---
 8 files changed, 168 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 4d9106580950..f2adedff927f 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -545,7 +545,7 @@ extern int			compat_ipv6_getsockopt(struct sock *sk,
 						char __user *optval,
 						int __user *optlen);
 
-extern void			ipv6_packet_init(void);
+extern int			ipv6_packet_init(void);
 
 extern void			ipv6_packet_cleanup(void);
 
diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index aa9a4a6b99df..27394e0447d8 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -23,10 +23,14 @@ extern int				ipv6_frag_init(void);
 extern void				ipv6_frag_exit(void);
 
 /* transport protocols */
-extern void				rawv6_init(void);
-extern void				udpv6_init(void);
-extern void 				udplitev6_init(void);
-extern void				tcpv6_init(void);
+extern int				rawv6_init(void);
+extern void				rawv6_exit(void);
+extern int				udpv6_init(void);
+extern void				udpv6_exit(void);
+extern int 				udplitev6_init(void);
+extern void 				udplitev6_exit(void);
+extern int				tcpv6_init(void);
+extern void				tcpv6_exit(void);
 
 extern int				udpv6_connect(struct sock *sk,
 						      struct sockaddr *uaddr,
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 53b06de696bd..34c20533ba5d 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -529,42 +529,6 @@ static struct net_proto_family inet6_family_ops = {
 	.owner	= THIS_MODULE,
 };
 
-/* Same as inet6_dgram_ops, sans udp_poll.  */
-static const struct proto_ops inet6_sockraw_ops = {
-	.family		   = PF_INET6,
-	.owner		   = THIS_MODULE,
-	.release	   = inet6_release,
-	.bind		   = inet6_bind,
-	.connect	   = inet_dgram_connect,	/* ok		*/
-	.socketpair	   = sock_no_socketpair,	/* a do nothing	*/
-	.accept		   = sock_no_accept,		/* a do nothing	*/
-	.getname	   = inet6_getname,
-	.poll		   = datagram_poll,		/* ok		*/
-	.ioctl		   = inet6_ioctl,		/* must change  */
-	.listen		   = sock_no_listen,		/* ok		*/
-	.shutdown	   = inet_shutdown,		/* ok		*/
-	.setsockopt	   = sock_common_setsockopt,	/* ok		*/
-	.getsockopt	   = sock_common_getsockopt,	/* ok		*/
-	.sendmsg	   = inet_sendmsg,		/* ok		*/
-	.recvmsg	   = sock_common_recvmsg,	/* ok		*/
-	.mmap		   = sock_no_mmap,
-	.sendpage	   = sock_no_sendpage,
-#ifdef CONFIG_COMPAT
-	.compat_setsockopt = compat_sock_common_setsockopt,
-	.compat_getsockopt = compat_sock_common_getsockopt,
-#endif
-};
-
-static struct inet_protosw rawv6_protosw = {
-	.type		= SOCK_RAW,
-	.protocol	= IPPROTO_IP,	/* wild card */
-	.prot		= &rawv6_prot,
-	.ops		= &inet6_sockraw_ops,
-	.capability	= CAP_NET_RAW,
-	.no_check	= UDP_CSUM_DEFAULT,
-	.flags		= INET_PROTOSW_REUSE,
-};
-
 int inet6_register_protosw(struct inet_protosw *p)
 {
 	struct list_head *lh;
@@ -771,7 +735,6 @@ static int __init inet6_init(void)
 	__this_module.can_unload = &ipv6_unload;
 #endif
 #endif
-
 	err = proto_register(&tcpv6_prot, 1);
 	if (err)
 		goto out;
@@ -796,14 +759,16 @@ static int __init inet6_init(void)
 	/* We MUST register RAW sockets before we create the ICMP6,
 	 * IGMP6, or NDISC control sockets.
 	 */
-	inet6_register_protosw(&rawv6_protosw);
+	err = rawv6_init();
+	if (err)
+		goto out_unregister_raw_proto;
 
 	/* Register the family here so that the init calls below will
 	 * be able to create sockets. (?? is this dangerous ??)
 	 */
 	err = sock_register(&inet6_family_ops);
 	if (err)
-		goto out_unregister_raw_proto;
+		goto out_sock_register_fail;
 
 	/* Initialise ipv6 mibs */
 	err = init_ipv6_mibs();
@@ -871,15 +836,32 @@ static int __init inet6_init(void)
 		goto ipv6_frag_fail;
 
 	/* Init v6 transport protocols. */
-	udpv6_init();
-	udplitev6_init();
-	tcpv6_init();
+	err = udpv6_init();
+	if (err)
+		goto udpv6_fail;
 
-	ipv6_packet_init();
-	err = 0;
+	err = udplitev6_init();
+	if (err)
+		goto udplitev6_fail;
+
+	err = tcpv6_init();
+	if (err)
+		goto tcpv6_fail;
+
+	err = ipv6_packet_init();
+	if (err)
+		goto ipv6_packet_fail;
 out:
 	return err;
 
+ipv6_packet_fail:
+	tcpv6_exit();
+tcpv6_fail:
+	udplitev6_exit();
+udplitev6_fail:
+	udpv6_exit();
+udpv6_fail:
+	ipv6_frag_exit();
 ipv6_frag_fail:
 	ipv6_exthdrs_exit();
 ipv6_exthdrs_fail:
@@ -920,6 +902,8 @@ icmp_fail:
 out_unregister_sock:
 	sock_unregister(PF_INET6);
 	rtnl_unregister_all(PF_INET6);
+out_sock_register_fail:
+	rawv6_exit();
 out_unregister_raw_proto:
 	proto_unregister(&rawv6_prot);
 out_unregister_udplite_proto:
@@ -939,6 +923,10 @@ static void __exit inet6_exit(void)
 	/* Disallow any further netlink messages */
 	rtnl_unregister_all(PF_INET6);
 
+	udpv6_exit();
+	udplitev6_exit();
+	tcpv6_exit();
+
 	/* Cleanup code parts. */
 	ipv6_packet_cleanup();
 	ipv6_frag_exit();
@@ -961,6 +949,7 @@ static void __exit inet6_exit(void)
 	igmp6_cleanup();
 	ndisc_cleanup();
 	icmpv6_cleanup();
+	rawv6_exit();
 #ifdef CONFIG_SYSCTL
 	ipv6_sysctl_unregister();
 #endif
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 8c5f80fd03ad..20fece4ad3d8 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -1128,9 +1128,10 @@ int compat_ipv6_getsockopt(struct sock *sk, int level, int optname,
 EXPORT_SYMBOL(compat_ipv6_getsockopt);
 #endif
 
-void __init ipv6_packet_init(void)
+int __init ipv6_packet_init(void)
 {
 	dev_add_pack(&ipv6_packet_type);
+	return 0;
 }
 
 void ipv6_packet_cleanup(void)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index b34631e1b015..850b83e430bc 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1273,3 +1273,55 @@ void raw6_proc_exit(void)
 	proc_net_remove(&init_net, "raw6");
 }
 #endif	/* CONFIG_PROC_FS */
+
+/* Same as inet6_dgram_ops, sans udp_poll.  */
+static const struct proto_ops inet6_sockraw_ops = {
+	.family		   = PF_INET6,
+	.owner		   = THIS_MODULE,
+	.release	   = inet6_release,
+	.bind		   = inet6_bind,
+	.connect	   = inet_dgram_connect,	/* ok		*/
+	.socketpair	   = sock_no_socketpair,	/* a do nothing	*/
+	.accept		   = sock_no_accept,		/* a do nothing	*/
+	.getname	   = inet6_getname,
+	.poll		   = datagram_poll,		/* ok		*/
+	.ioctl		   = inet6_ioctl,		/* must change  */
+	.listen		   = sock_no_listen,		/* ok		*/
+	.shutdown	   = inet_shutdown,		/* ok		*/
+	.setsockopt	   = sock_common_setsockopt,	/* ok		*/
+	.getsockopt	   = sock_common_getsockopt,	/* ok		*/
+	.sendmsg	   = inet_sendmsg,		/* ok		*/
+	.recvmsg	   = sock_common_recvmsg,	/* ok		*/
+	.mmap		   = sock_no_mmap,
+	.sendpage	   = sock_no_sendpage,
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_sock_common_setsockopt,
+	.compat_getsockopt = compat_sock_common_getsockopt,
+#endif
+};
+
+static struct inet_protosw rawv6_protosw = {
+	.type		= SOCK_RAW,
+	.protocol	= IPPROTO_IP,	/* wild card */
+	.prot		= &rawv6_prot,
+	.ops		= &inet6_sockraw_ops,
+	.capability	= CAP_NET_RAW,
+	.no_check	= UDP_CSUM_DEFAULT,
+	.flags		= INET_PROTOSW_REUSE,
+};
+
+int __init rawv6_init(void)
+{
+	int ret;
+
+	ret = inet6_register_protosw(&rawv6_protosw);
+	if (ret)
+		goto out;
+out:
+	return ret;
+}
+
+void __exit rawv6_exit(void)
+{
+	inet6_unregister_protosw(&rawv6_protosw);
+}
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 93980c3b83e6..9544beb6d1ca 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2166,14 +2166,36 @@ static struct inet_protosw tcpv6_protosw = {
 				INET_PROTOSW_ICSK,
 };
 
-void __init tcpv6_init(void)
+int __init tcpv6_init(void)
 {
+	int ret;
+
+	ret = inet6_add_protocol(&tcpv6_protocol, IPPROTO_TCP);
+	if (ret)
+		goto out;
+
 	/* register inet6 protocol */
-	if (inet6_add_protocol(&tcpv6_protocol, IPPROTO_TCP) < 0)
-		printk(KERN_ERR "tcpv6_init: Could not register protocol\n");
-	inet6_register_protosw(&tcpv6_protosw);
+	ret = inet6_register_protosw(&tcpv6_protosw);
+	if (ret)
+		goto out_tcpv6_protocol;
+
+	ret = inet_csk_ctl_sock_create(&tcp6_socket, PF_INET6,
+				       SOCK_RAW, IPPROTO_TCP);
+	if (ret)
+		goto out_tcpv6_protosw;
+out:
+	return ret;
 
-	if (inet_csk_ctl_sock_create(&tcp6_socket, PF_INET6, SOCK_RAW,
-				     IPPROTO_TCP) < 0)
-		panic("Failed to create the TCPv6 control socket.\n");
+out_tcpv6_protocol:
+	inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP);
+out_tcpv6_protosw:
+	inet6_unregister_protosw(&tcpv6_protosw);
+	goto out;
+}
+
+void __exit tcpv6_exit(void)
+{
+	sock_release(tcp6_socket);
+	inet6_unregister_protosw(&tcpv6_protosw);
+	inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP);
 }
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index fa640765385e..1e3bd39f54ec 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1016,9 +1016,27 @@ static struct inet_protosw udpv6_protosw = {
 };
 
 
-void __init udpv6_init(void)
+int __init udpv6_init(void)
 {
-	if (inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP) < 0)
-		printk(KERN_ERR "udpv6_init: Could not register protocol\n");
-	inet6_register_protosw(&udpv6_protosw);
+	int ret;
+
+	ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP);
+	if (ret)
+		goto out;
+
+	ret = inet6_register_protosw(&udpv6_protosw);
+	if (ret)
+		goto out_udpv6_protocol;
+out:
+	return ret;
+
+out_udpv6_protocol:
+	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
+	goto out;
+}
+
+void __exit udpv6_exit(void)
+{
+	inet6_unregister_protosw(&udpv6_protosw);
+	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
 }
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 5a0379f71415..f20b376689fb 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -77,12 +77,29 @@ static struct inet_protosw udplite6_protosw = {
 	.flags		= INET_PROTOSW_PERMANENT,
 };
 
-void __init udplitev6_init(void)
+int __init udplitev6_init(void)
 {
-	if (inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE) < 0)
-		printk(KERN_ERR "%s: Could not register.\n", __FUNCTION__);
+	int ret;
 
-	inet6_register_protosw(&udplite6_protosw);
+	ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
+	if (ret)
+		goto out;
+
+	ret = inet6_register_protosw(&udplite6_protosw);
+	if (ret)
+		goto out_udplitev6_protocol;
+out:
+	return ret;
+
+out_udplitev6_protocol:
+	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
+	goto out;
+}
+
+void __exit udplitev6_exit(void)
+{
+	inet6_unregister_protosw(&udplite6_protosw);
+	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
 }
 
 #ifdef CONFIG_PROC_FS
-- 
cgit v1.2.3


From 27147c9e6e8316af91dea487aa2c14b18dcff4cf Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 11 Dec 2007 04:18:41 -0800
Subject: [AF_UNIX]: Remove unused declaration of sysctl_unix_max_dgram_qlen.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/af_unix.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index e0fba269d092..2dfa96b0575e 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -59,7 +59,6 @@ struct unix_sock {
 #define unix_sk(__sk) ((struct unix_sock *)__sk)
 
 #ifdef CONFIG_SYSCTL
-extern int sysctl_unix_max_dgram_qlen;
 extern int unix_sysctl_register(struct net *net);
 extern void unix_sysctl_unregister(struct net *net);
 #else
-- 
cgit v1.2.3


From a0a53c8ba95451feef6c1975016f0a1eb3044ad4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 11 Dec 2007 04:19:17 -0800
Subject: [NETNS]: struct net content re-work (v3)

Recently David Miller and Herbert Xu pointed out that struct net becomes
overbloated and un-maintainable. There are two solutions:
- provide a pointer to a network subsystem definition from struct net.
  This costs an additional dereferrence
- place sub-system definition into the structure itself. This will speedup
  run-time access at the cost of recompilation time

The second approach looks better for us. Other sub-systems will follow.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  6 +++---
 include/net/netns/unix.h    | 13 +++++++++++++
 net/unix/af_unix.c          |  4 ++--
 net/unix/sysctl_net_unix.c  | 12 ++++++------
 4 files changed, 24 insertions(+), 11 deletions(-)
 create mode 100644 include/net/netns/unix.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index b62e31fca474..d943fd4eaba5 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -8,6 +8,8 @@
 #include <linux/workqueue.h>
 #include <linux/list.h>
 
+#include <net/netns/unix.h>
+
 struct proc_dir_entry;
 struct net_device;
 struct sock;
@@ -45,9 +47,7 @@ struct net {
 	rwlock_t		packet_sklist_lock;
 	struct hlist_head	packet_sklist;
 
-	/* unix sockets */
-	int			sysctl_unix_max_dgram_qlen;
-	struct ctl_table_header	*unix_ctl;
+	struct netns_unix	unx;
 };
 
 #ifdef CONFIG_NET
diff --git a/include/net/netns/unix.h b/include/net/netns/unix.h
new file mode 100644
index 000000000000..284649d4dfb4
--- /dev/null
+++ b/include/net/netns/unix.h
@@ -0,0 +1,13 @@
+/*
+ * Unix network namespace
+ */
+#ifndef __NETNS_UNIX_H__
+#define __NETNS_UNIX_H__
+
+struct ctl_table_header;
+struct netns_unix {
+	int			sysctl_max_dgram_qlen;
+	struct ctl_table_header	*ctl;
+};
+
+#endif /* __NETNS_UNIX_H__ */
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index b8a2189fb5c0..63a9239571a4 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -592,7 +592,7 @@ static struct sock * unix_create1(struct net *net, struct socket *sock)
 				&af_unix_sk_receive_queue_lock_key);
 
 	sk->sk_write_space	= unix_write_space;
-	sk->sk_max_ack_backlog	= net->sysctl_unix_max_dgram_qlen;
+	sk->sk_max_ack_backlog	= net->unx.sysctl_max_dgram_qlen;
 	sk->sk_destruct		= unix_sock_destructor;
 	u	  = unix_sk(sk);
 	u->dentry = NULL;
@@ -2138,7 +2138,7 @@ static int unix_net_init(struct net *net)
 {
 	int error = -ENOMEM;
 
-	net->sysctl_unix_max_dgram_qlen = 10;
+	net->unx.sysctl_max_dgram_qlen = 10;
 	if (unix_sysctl_register(net))
 		goto out;
 
diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c
index 553ef6a487d0..77513d7e35f2 100644
--- a/net/unix/sysctl_net_unix.c
+++ b/net/unix/sysctl_net_unix.c
@@ -18,7 +18,7 @@ static ctl_table unix_table[] = {
 	{
 		.ctl_name	= NET_UNIX_MAX_DGRAM_QLEN,
 		.procname	= "max_dgram_qlen",
-		.data		= &init_net.sysctl_unix_max_dgram_qlen,
+		.data		= &init_net.unx.sysctl_max_dgram_qlen,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -40,9 +40,9 @@ int unix_sysctl_register(struct net *net)
 	if (table == NULL)
 		goto err_alloc;
 
-	table[0].data = &net->sysctl_unix_max_dgram_qlen;
-	net->unix_ctl = register_net_sysctl_table(net, unix_path, table);
-	if (net->unix_ctl == NULL)
+	table[0].data = &net->unx.sysctl_max_dgram_qlen;
+	net->unx.ctl = register_net_sysctl_table(net, unix_path, table);
+	if (net->unx.ctl == NULL)
 		goto err_reg;
 
 	return 0;
@@ -57,8 +57,8 @@ void unix_sysctl_unregister(struct net *net)
 {
 	struct ctl_table *table;
 
-	table = net->unix_ctl->ctl_table_arg;
-	unregister_sysctl_table(net->unix_ctl);
+	table = net->unx.ctl->ctl_table_arg;
+	unregister_sysctl_table(net->unx.ctl);
 	kfree(table);
 }
 
-- 
cgit v1.2.3


From 2aaef4e47fef8a6c0bc7fc5d9d3eea4af290e04c Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 11 Dec 2007 04:19:54 -0800
Subject: [NETNS]: separate af_packet netns data

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  6 ++----
 include/net/netns/packet.h  | 15 +++++++++++++++
 net/packet/af_packet.c      | 28 ++++++++++++++--------------
 3 files changed, 31 insertions(+), 18 deletions(-)
 create mode 100644 include/net/netns/packet.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index d943fd4eaba5..18da0af6192f 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -9,6 +9,7 @@
 #include <linux/list.h>
 
 #include <net/netns/unix.h>
+#include <net/netns/packet.h>
 
 struct proc_dir_entry;
 struct net_device;
@@ -43,10 +44,7 @@ struct net {
 	struct ctl_table_header	*sysctl_core_hdr;
 	int			sysctl_somaxconn;
 
-	/* List of all packet sockets. */
-	rwlock_t		packet_sklist_lock;
-	struct hlist_head	packet_sklist;
-
+	struct netns_packet	packet;
 	struct netns_unix	unx;
 };
 
diff --git a/include/net/netns/packet.h b/include/net/netns/packet.h
new file mode 100644
index 000000000000..637daf698884
--- /dev/null
+++ b/include/net/netns/packet.h
@@ -0,0 +1,15 @@
+/*
+ * Packet network namespace
+ */
+#ifndef __NETNS_PACKET_H__
+#define __NETNS_PACKET_H__
+
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+struct netns_packet {
+	rwlock_t		sklist_lock;
+	struct hlist_head	sklist;
+};
+
+#endif /* __NETNS_PACKET_H__ */
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ace29f1c4c5b..485af5691d64 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -803,9 +803,9 @@ static int packet_release(struct socket *sock)
 	net = sk->sk_net;
 	po = pkt_sk(sk);
 
-	write_lock_bh(&net->packet_sklist_lock);
+	write_lock_bh(&net->packet.sklist_lock);
 	sk_del_node_init(sk);
-	write_unlock_bh(&net->packet_sklist_lock);
+	write_unlock_bh(&net->packet.sklist_lock);
 
 	/*
 	 *	Unhook packet receive handler.
@@ -1015,9 +1015,9 @@ static int packet_create(struct net *net, struct socket *sock, int protocol)
 		po->running = 1;
 	}
 
-	write_lock_bh(&net->packet_sklist_lock);
-	sk_add_node(sk, &net->packet_sklist);
-	write_unlock_bh(&net->packet_sklist_lock);
+	write_lock_bh(&net->packet.sklist_lock);
+	sk_add_node(sk, &net->packet.sklist);
+	write_unlock_bh(&net->packet.sklist_lock);
 	return(0);
 out:
 	return err;
@@ -1452,8 +1452,8 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void
 	struct net_device *dev = data;
 	struct net *net = dev->nd_net;
 
-	read_lock(&net->packet_sklist_lock);
-	sk_for_each(sk, node, &net->packet_sklist) {
+	read_lock(&net->packet.sklist_lock);
+	sk_for_each(sk, node, &net->packet.sklist) {
 		struct packet_sock *po = pkt_sk(sk);
 
 		switch (msg) {
@@ -1492,7 +1492,7 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void
 			break;
 		}
 	}
-	read_unlock(&net->packet_sklist_lock);
+	read_unlock(&net->packet.sklist_lock);
 	return NOTIFY_DONE;
 }
 
@@ -1862,7 +1862,7 @@ static inline struct sock *packet_seq_idx(struct net *net, loff_t off)
 	struct sock *s;
 	struct hlist_node *node;
 
-	sk_for_each(s, node, &net->packet_sklist) {
+	sk_for_each(s, node, &net->packet.sklist) {
 		if (!off--)
 			return s;
 	}
@@ -1872,7 +1872,7 @@ static inline struct sock *packet_seq_idx(struct net *net, loff_t off)
 static void *packet_seq_start(struct seq_file *seq, loff_t *pos)
 {
 	struct net *net = seq_file_net(seq);
-	read_lock(&net->packet_sklist_lock);
+	read_lock(&net->packet.sklist_lock);
 	return *pos ? packet_seq_idx(net, *pos - 1) : SEQ_START_TOKEN;
 }
 
@@ -1881,14 +1881,14 @@ static void *packet_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	struct net *net = seq->private;
 	++*pos;
 	return  (v == SEQ_START_TOKEN)
-		? sk_head(&net->packet_sklist)
+		? sk_head(&net->packet.sklist)
 		: sk_next((struct sock*)v) ;
 }
 
 static void packet_seq_stop(struct seq_file *seq, void *v)
 {
 	struct net *net = seq->private;
-	read_unlock(&net->packet_sklist_lock);
+	read_unlock(&net->packet.sklist_lock);
 }
 
 static int packet_seq_show(struct seq_file *seq, void *v)
@@ -1940,8 +1940,8 @@ static const struct file_operations packet_seq_fops = {
 
 static int packet_net_init(struct net *net)
 {
-	rwlock_init(&net->packet_sklist_lock);
-	INIT_HLIST_HEAD(&net->packet_sklist);
+	rwlock_init(&net->packet.sklist_lock);
+	INIT_HLIST_HEAD(&net->packet.sklist);
 
 	if (!proc_net_fops_create(net, "packet", 0, &packet_seq_fops))
 		return -ENOMEM;
-- 
cgit v1.2.3


From 815f4e57e9fc67456624ecde0515a901368c78d2 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 12 Dec 2007 10:36:59 -0800
Subject: [IPSEC]: Make xfrm_lookup flags argument a bit-field

This patch introduces an enum for bits in the flags argument of xfrm_lookup.
This is so that we can cram more information into it later.

Since all current users use just the values 0 and 1, XFRM_LOOKUP_WAIT has
been added with the value 1 << 0 to represent the current meaning of flags.

The test in __xfrm_lookup has been changed accordingly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h      | 5 +++++
 net/xfrm/xfrm_policy.c | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index e86b9a008ebf..aaa2dbb50179 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -265,6 +265,11 @@ static inline struct dst_entry *dst_check(struct dst_entry *dst, u32 cookie)
 
 extern void		dst_init(void);
 
+/* Flags for xfrm_lookup flags argument. */
+enum {
+	XFRM_LOOKUP_WAIT = 1 << 0,
+};
+
 struct flowi;
 #ifndef CONFIG_XFRM
 static inline int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 95dc581861e8..3d516d57b5b2 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1565,7 +1565,7 @@ restart:
 				xfrm_pol_put(policy);
 				return -EREMOTE;
 			}
-			if (err == -EAGAIN && flags) {
+			if (err == -EAGAIN && (flags & XFRM_LOOKUP_WAIT)) {
 				DECLARE_WAITQUEUE(wait, current);
 
 				add_wait_queue(&km_waitq, &wait);
-- 
cgit v1.2.3


From d5422efe680fc55010c6ddca2370ca9548a96355 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 12 Dec 2007 10:44:16 -0800
Subject: [IPSEC]: Added xfrm_decode_session_reverse and
 xfrmX_policy_check_reverse

RFC 4301 requires us to relookup ICMP traffic that does not match any
policies using the reverse of its payload.  This patch adds the functions
xfrm_decode_session_reverse and xfrmX_policy_check_reverse so we can get
the reverse flow to perform such a lookup.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h    |  1 +
 include/net/xfrm.h      | 63 +++++++++++++++++++++++++++++++++++++++++++++----
 net/ipv4/xfrm4_policy.c | 10 ++++----
 net/ipv6/xfrm6_policy.c | 10 ++++----
 net/xfrm/xfrm_policy.c  | 17 ++++++++-----
 5 files changed, 80 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index b58adc52448d..c0e41e02234f 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -114,6 +114,7 @@ enum
 	XFRM_POLICY_IN	= 0,
 	XFRM_POLICY_OUT	= 1,
 	XFRM_POLICY_FWD	= 2,
+	XFRM_POLICY_MASK = 3,
 	XFRM_POLICY_MAX	= 3
 };
 
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index fe881b6e2bd0..d6dae5ae7abe 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -239,7 +239,8 @@ struct xfrm_policy_afinfo {
 	int			(*get_saddr)(xfrm_address_t *saddr, xfrm_address_t *daddr);
 	struct dst_entry	*(*find_bundle)(struct flowi *fl, struct xfrm_policy *policy);
 	void			(*decode_session)(struct sk_buff *skb,
-						  struct flowi *fl);
+						  struct flowi *fl,
+						  int reverse);
 	int			(*get_tos)(struct flowi *fl);
 	int			(*fill_dst)(struct xfrm_dst *xdst,
 					    struct net_device *dev);
@@ -844,14 +845,23 @@ xfrm_state_addr_cmp(struct xfrm_tmpl *tmpl, struct xfrm_state *x, unsigned short
 #ifdef CONFIG_XFRM
 extern int __xfrm_policy_check(struct sock *, int dir, struct sk_buff *skb, unsigned short family);
 
-static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, unsigned short family)
+static inline int __xfrm_policy_check2(struct sock *sk, int dir,
+				       struct sk_buff *skb,
+				       unsigned int family, int reverse)
 {
+	int ndir = dir | (reverse ? XFRM_POLICY_MASK + 1 : 0);
+
 	if (sk && sk->sk_policy[XFRM_POLICY_IN])
-		return __xfrm_policy_check(sk, dir, skb, family);
+		return __xfrm_policy_check(sk, ndir, skb, family);
 
 	return	(!xfrm_policy_count[dir] && !skb->sp) ||
 		(skb->dst->flags & DST_NOPOLICY) ||
-		__xfrm_policy_check(sk, dir, skb, family);
+		__xfrm_policy_check(sk, ndir, skb, family);
+}
+
+static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, unsigned short family)
+{
+	return __xfrm_policy_check2(sk, dir, skb, family, 0);
 }
 
 static inline int xfrm4_policy_check(struct sock *sk, int dir, struct sk_buff *skb)
@@ -864,7 +874,34 @@ static inline int xfrm6_policy_check(struct sock *sk, int dir, struct sk_buff *s
 	return xfrm_policy_check(sk, dir, skb, AF_INET6);
 }
 
-extern int xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family);
+static inline int xfrm4_policy_check_reverse(struct sock *sk, int dir,
+					     struct sk_buff *skb)
+{
+	return __xfrm_policy_check2(sk, dir, skb, AF_INET, 1);
+}
+
+static inline int xfrm6_policy_check_reverse(struct sock *sk, int dir,
+					     struct sk_buff *skb)
+{
+	return __xfrm_policy_check2(sk, dir, skb, AF_INET6, 1);
+}
+
+extern int __xfrm_decode_session(struct sk_buff *skb, struct flowi *fl,
+				 unsigned int family, int reverse);
+
+static inline int xfrm_decode_session(struct sk_buff *skb, struct flowi *fl,
+				      unsigned int family)
+{
+	return __xfrm_decode_session(skb, fl, family, 0);
+}
+
+static inline int xfrm_decode_session_reverse(struct sk_buff *skb,
+					      struct flowi *fl,
+					      unsigned int family)
+{
+	return __xfrm_decode_session(skb, fl, family, 1);
+}
+
 extern int __xfrm_route_forward(struct sk_buff *skb, unsigned short family);
 
 static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family)
@@ -925,6 +962,22 @@ static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *sk
 {
 	return 1;
 }
+static inline int xfrm_decode_session_reverse(struct sk_buff *skb,
+					      struct flowi *fl,
+					      unsigned int family)
+{
+	return -ENOSYS;
+}
+static inline int xfrm4_policy_check_reverse(struct sock *sk, int dir,
+					     struct sk_buff *skb)
+{
+	return 1;
+}
+static inline int xfrm6_policy_check_reverse(struct sock *sk, int dir,
+					     struct sk_buff *skb)
+{
+	return 1;
+}
 #endif
 
 static __inline__
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 10b72d185bb9..5ccae3a463c2 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -115,7 +115,7 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 }
 
 static void
-_decode_session4(struct sk_buff *skb, struct flowi *fl)
+_decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 {
 	struct iphdr *iph = ip_hdr(skb);
 	u8 *xprth = skb_network_header(skb) + iph->ihl * 4;
@@ -131,8 +131,8 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
 				__be16 *ports = (__be16 *)xprth;
 
-				fl->fl_ip_sport = ports[0];
-				fl->fl_ip_dport = ports[1];
+				fl->fl_ip_sport = ports[!!reverse];
+				fl->fl_ip_dport = ports[!reverse];
 			}
 			break;
 
@@ -174,8 +174,8 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 		}
 	}
 	fl->proto = iph->protocol;
-	fl->fl4_dst = iph->daddr;
-	fl->fl4_src = iph->saddr;
+	fl->fl4_dst = reverse ? iph->saddr : iph->daddr;
+	fl->fl4_src = reverse ? iph->daddr : iph->saddr;
 	fl->fl4_tos = iph->tos;
 }
 
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 181cf91538f1..d26b7dc3f33b 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -123,7 +123,7 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 }
 
 static inline void
-_decode_session6(struct sk_buff *skb, struct flowi *fl)
+_decode_session6(struct sk_buff *skb, struct flowi *fl, int reverse)
 {
 	u16 offset = skb_network_header_len(skb);
 	struct ipv6hdr *hdr = ipv6_hdr(skb);
@@ -132,8 +132,8 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl)
 	u8 nexthdr = nh[IP6CB(skb)->nhoff];
 
 	memset(fl, 0, sizeof(struct flowi));
-	ipv6_addr_copy(&fl->fl6_dst, &hdr->daddr);
-	ipv6_addr_copy(&fl->fl6_src, &hdr->saddr);
+	ipv6_addr_copy(&fl->fl6_dst, reverse ? &hdr->saddr : &hdr->daddr);
+	ipv6_addr_copy(&fl->fl6_src, reverse ? &hdr->daddr : &hdr->saddr);
 
 	while (pskb_may_pull(skb, nh + offset + 1 - skb->data)) {
 		nh = skb_network_header(skb);
@@ -156,8 +156,8 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl)
 			if (pskb_may_pull(skb, nh + offset + 4 - skb->data)) {
 				__be16 *ports = (__be16 *)exthdr;
 
-				fl->fl_ip_sport = ports[0];
-				fl->fl_ip_dport = ports[1];
+				fl->fl_ip_sport = ports[!!reverse];
+				fl->fl_ip_dport = ports[!reverse];
 			}
 			fl->proto = nexthdr;
 			return;
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 3d516d57b5b2..2e10d46c0e8c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1732,8 +1732,8 @@ xfrm_policy_ok(struct xfrm_tmpl *tmpl, struct sec_path *sp, int start,
 	return start;
 }
 
-int
-xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family)
+int __xfrm_decode_session(struct sk_buff *skb, struct flowi *fl,
+			  unsigned int family, int reverse)
 {
 	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
 	int err;
@@ -1741,12 +1741,12 @@ xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family
 	if (unlikely(afinfo == NULL))
 		return -EAFNOSUPPORT;
 
-	afinfo->decode_session(skb, fl);
+	afinfo->decode_session(skb, fl, reverse);
 	err = security_xfrm_decode_session(skb, &fl->secid);
 	xfrm_policy_put_afinfo(afinfo);
 	return err;
 }
-EXPORT_SYMBOL(xfrm_decode_session);
+EXPORT_SYMBOL(__xfrm_decode_session);
 
 static inline int secpath_has_nontransport(struct sec_path *sp, int k, int *idxp)
 {
@@ -1768,11 +1768,16 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 	int npols = 0;
 	int xfrm_nr;
 	int pi;
+	int reverse;
 	struct flowi fl;
-	u8 fl_dir = policy_to_flow_dir(dir);
+	u8 fl_dir;
 	int xerr_idx = -1;
 
-	if (xfrm_decode_session(skb, &fl, family) < 0)
+	reverse = dir & ~XFRM_POLICY_MASK;
+	dir &= XFRM_POLICY_MASK;
+	fl_dir = policy_to_flow_dir(dir);
+
+	if (__xfrm_decode_session(skb, &fl, family, reverse) < 0)
 		return 0;
 	nf_nat_decode_session(skb, &fl, family);
 
-- 
cgit v1.2.3


From 8b7817f3a959ed99d7443afc12f78a7e1fcc2063 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 12 Dec 2007 10:44:43 -0800
Subject: [IPSEC]: Add ICMP host relookup support

RFC 4301 requires us to relookup ICMP traffic that does not match any
policies using the reverse of its payload.  This patch implements this
for ICMP traffic that originates from or terminates on localhost.

This is activated on outbound with the new policy flag XFRM_POLICY_ICMP,
and on inbound by the new state flag XFRM_STATE_ICMP.

On inbound the policy check is now performed by the ICMP protocol so
that it can repeat the policy check where necessary.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h   |  3 ++
 include/net/dst.h      |  1 +
 net/ipv4/af_inet.c     |  1 +
 net/ipv4/icmp.c        | 82 ++++++++++++++++++++++++++++++++++++++++++++++++--
 net/ipv6/icmp.c        | 60 +++++++++++++++++++++++++++++++++---
 net/xfrm/xfrm_policy.c | 17 +++++++++--
 6 files changed, 154 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index c0e41e02234f..1131eabfaa2a 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -329,6 +329,7 @@ struct xfrm_usersa_info {
 #define XFRM_STATE_DECAP_DSCP	2
 #define XFRM_STATE_NOPMTUDISC	4
 #define XFRM_STATE_WILDRECV	8
+#define XFRM_STATE_ICMP		16
 };
 
 struct xfrm_usersa_id {
@@ -363,6 +364,8 @@ struct xfrm_userpolicy_info {
 #define XFRM_POLICY_BLOCK	1
 	__u8				flags;
 #define XFRM_POLICY_LOCALOK	1	/* Allow user to override global policy */
+	/* Automatically expand selector to include matching ICMP payloads. */
+#define XFRM_POLICY_ICMP	2
 	__u8				share;
 };
 
diff --git a/include/net/dst.h b/include/net/dst.h
index aaa2dbb50179..31468c9aa877 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -268,6 +268,7 @@ extern void		dst_init(void);
 /* Flags for xfrm_lookup flags argument. */
 enum {
 	XFRM_LOOKUP_WAIT = 1 << 0,
+	XFRM_LOOKUP_ICMP = 1 << 1,
 };
 
 struct flowi;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 7f8b27ff94ff..5089a369e99c 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1291,6 +1291,7 @@ static struct net_protocol udp_protocol = {
 
 static struct net_protocol icmp_protocol = {
 	.handler =	icmp_rcv,
+	.no_policy =	1,
 };
 
 static int __init init_ipv4_mibs(void)
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 13d74598d3e4..3c41a6f7e6ec 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -92,6 +92,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <net/checksum.h>
+#include <net/xfrm.h>
 
 /*
  *	Build xmit assembly blocks
@@ -563,11 +564,71 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 				}
 			}
 		};
+		int err;
+		struct rtable *rt2;
+
 		security_skb_classify_flow(skb_in, &fl);
-		if (ip_route_output_key(&rt, &fl))
+		if (__ip_route_output_key(&rt, &fl))
+			goto out_unlock;
+
+		/* No need to clone since we're just using its address. */
+		rt2 = rt;
+
+		err = xfrm_lookup((struct dst_entry **)&rt, &fl, NULL, 0);
+		switch (err) {
+		case 0:
+			if (rt != rt2)
+				goto route_done;
+			break;
+		case -EPERM:
+			rt = NULL;
+			break;
+		default:
+			goto out_unlock;
+		}
+
+		if (xfrm_decode_session_reverse(skb_in, &fl, AF_INET))
+			goto out_unlock;
+
+		if (inet_addr_type(fl.fl4_src) == RTN_LOCAL)
+			err = __ip_route_output_key(&rt2, &fl);
+		else {
+			struct flowi fl2 = {};
+			struct dst_entry *odst;
+
+			fl2.fl4_dst = fl.fl4_src;
+			if (ip_route_output_key(&rt2, &fl2))
+				goto out_unlock;
+
+			/* Ugh! */
+			odst = skb_in->dst;
+			err = ip_route_input(skb_in, fl.fl4_dst, fl.fl4_src,
+					     RT_TOS(tos), rt2->u.dst.dev);
+
+			dst_release(&rt2->u.dst);
+			rt2 = (struct rtable *)skb_in->dst;
+			skb_in->dst = odst;
+		}
+
+		if (err)
+			goto out_unlock;
+
+		err = xfrm_lookup((struct dst_entry **)&rt2, &fl, NULL,
+				  XFRM_LOOKUP_ICMP);
+		if (err == -ENOENT) {
+			if (!rt)
+				goto out_unlock;
+			goto route_done;
+		}
+
+		dst_release(&rt->u.dst);
+		rt = rt2;
+
+		if (err)
 			goto out_unlock;
 	}
 
+route_done:
 	if (!icmpv4_xrlim_allow(rt, type, code))
 		goto ende;
 
@@ -916,6 +977,22 @@ int icmp_rcv(struct sk_buff *skb)
 	struct icmphdr *icmph;
 	struct rtable *rt = (struct rtable *)skb->dst;
 
+	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb) &&
+	    skb->sp->xvec[skb->sp->len - 1]->props.flags & XFRM_STATE_ICMP) {
+		int nh;
+
+		if (!pskb_may_pull(skb, sizeof(*icmph) + sizeof(struct iphdr)))
+			goto drop;
+
+		nh = skb_network_offset(skb);
+		skb_set_network_header(skb, sizeof(*icmph));
+
+		if (!xfrm4_policy_check_reverse(NULL, XFRM_POLICY_IN, skb))
+			goto drop;
+
+		skb_set_network_header(skb, nh);
+	}
+
 	ICMP_INC_STATS_BH(ICMP_MIB_INMSGS);
 
 	switch (skb->ip_summed) {
@@ -929,8 +1006,7 @@ int icmp_rcv(struct sk_buff *skb)
 			goto error;
 	}
 
-	if (!pskb_pull(skb, sizeof(struct icmphdr)))
-		goto error;
+	__skb_pull(skb, sizeof(*icmph));
 
 	icmph = icmp_hdr(skb);
 
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 93c96cfd5ee1..c0bea7bfaa8a 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -63,6 +63,7 @@
 #include <net/ip6_route.h>
 #include <net/addrconf.h>
 #include <net/icmp.h>
+#include <net/xfrm.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -86,7 +87,7 @@ static int icmpv6_rcv(struct sk_buff *skb);
 
 static struct inet6_protocol icmpv6_protocol = {
 	.handler	=	icmpv6_rcv,
-	.flags		=	INET6_PROTO_FINAL,
+	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
 };
 
 static __inline__ int icmpv6_xmit_lock(void)
@@ -310,8 +311,10 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 	struct ipv6_pinfo *np;
 	struct in6_addr *saddr = NULL;
 	struct dst_entry *dst;
+	struct dst_entry *dst2;
 	struct icmp6hdr tmp_hdr;
 	struct flowi fl;
+	struct flowi fl2;
 	struct icmpv6_msg msg;
 	int iif = 0;
 	int addr_type = 0;
@@ -418,9 +421,42 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 		goto out_dst_release;
 	}
 
-	if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0)
+	/* No need to clone since we're just using its address. */
+	dst2 = dst;
+
+	err = xfrm_lookup(&dst, &fl, sk, 0);
+	switch (err) {
+	case 0:
+		if (dst != dst2)
+			goto route_done;
+		break;
+	case -EPERM:
+		dst = NULL;
+		break;
+	default:
+		goto out;
+	}
+
+	if (xfrm_decode_session_reverse(skb, &fl2, AF_INET6))
+		goto out;
+
+	if (ip6_dst_lookup(sk, &dst2, &fl))
 		goto out;
 
+	err = xfrm_lookup(&dst2, &fl, sk, XFRM_LOOKUP_ICMP);
+	if (err == -ENOENT) {
+		if (!dst)
+			goto out;
+		goto route_done;
+	}
+
+	dst_release(dst);
+	dst = dst2;
+
+	if (err)
+		goto out;
+
+route_done:
 	if (ipv6_addr_is_multicast(&fl.fl6_dst))
 		hlimit = np->mcast_hops;
 	else
@@ -608,6 +644,22 @@ static int icmpv6_rcv(struct sk_buff *skb)
 	struct icmp6hdr *hdr;
 	int type;
 
+	if (xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb) &&
+	    skb->sp->xvec[skb->sp->len - 1]->props.flags & XFRM_STATE_ICMP) {
+		int nh;
+
+		if (!pskb_may_pull(skb, sizeof(*hdr) + sizeof(*orig_hdr)))
+			goto drop_no_count;
+
+		nh = skb_network_offset(skb);
+		skb_set_network_header(skb, sizeof(*hdr));
+
+		if (!xfrm6_policy_check_reverse(NULL, XFRM_POLICY_IN, skb))
+			goto drop_no_count;
+
+		skb_set_network_header(skb, nh);
+	}
+
 	ICMP6_INC_STATS_BH(idev, ICMP6_MIB_INMSGS);
 
 	saddr = &ipv6_hdr(skb)->saddr;
@@ -630,8 +682,7 @@ static int icmpv6_rcv(struct sk_buff *skb)
 		}
 	}
 
-	if (!pskb_pull(skb, sizeof(struct icmp6hdr)))
-		goto discard_it;
+	__skb_pull(skb, sizeof(*hdr));
 
 	hdr = icmp6_hdr(skb);
 
@@ -717,6 +768,7 @@ static int icmpv6_rcv(struct sk_buff *skb)
 
 discard_it:
 	ICMP6_INC_STATS_BH(idev, ICMP6_MIB_INERRORS);
+drop_no_count:
 	kfree_skb(skb);
 	return 0;
 }
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 2e10d46c0e8c..a83b5e1349ed 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1469,11 +1469,13 @@ restart:
 			goto dropdst;
 	}
 
+	err = -ENOENT;
+
 	if (!policy) {
 		/* To accelerate a bit...  */
 		if ((dst_orig->flags & DST_NOXFRM) ||
 		    !xfrm_policy_count[XFRM_POLICY_OUT])
-			return 0;
+			goto nopol;
 
 		policy = flow_cache_lookup(fl, dst_orig->ops->family,
 					   dir, xfrm_policy_lookup);
@@ -1483,14 +1485,18 @@ restart:
 	}
 
 	if (!policy)
-		return 0;
+		goto nopol;
 
 	family = dst_orig->ops->family;
-	policy->curlft.use_time = get_seconds();
 	pols[0] = policy;
 	npols ++;
 	xfrm_nr += pols[0]->xfrm_nr;
 
+	if ((flags & XFRM_LOOKUP_ICMP) && !(policy->flags & XFRM_POLICY_ICMP))
+		goto error;
+
+	policy->curlft.use_time = get_seconds();
+
 	switch (policy->action) {
 	default:
 	case XFRM_POLICY_BLOCK:
@@ -1649,6 +1655,11 @@ dropdst:
 	dst_release(dst_orig);
 	*dst_p = NULL;
 	return err;
+
+nopol:
+	if (flags & XFRM_LOOKUP_ICMP)
+		goto dropdst;
+	return 0;
 }
 EXPORT_SYMBOL(__xfrm_lookup);
 
-- 
cgit v1.2.3


From 41380930d2cbdc0abf7513a675864258b7ac973d Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 12 Dec 2007 10:46:51 -0800
Subject: [NET]: Remove FASTCALL macro

X86_32 was the last user of the FASTCALL macro, now that it
uses regparm(3) by default, this macro expands to nothing.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ns83820.c          | 5 +----
 include/net/bluetooth/rfcomm.h | 4 ++--
 include/net/sock.h             | 4 ++--
 3 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ns83820.c b/drivers/net/ns83820.c
index ea71f6d82661..972acc3d1276 100644
--- a/drivers/net/ns83820.c
+++ b/drivers/net/ns83820.c
@@ -611,7 +611,7 @@ static inline int rx_refill(struct net_device *ndev, gfp_t gfp)
 	return i ? 0 : -ENOMEM;
 }
 
-static void FASTCALL(rx_refill_atomic(struct net_device *ndev));
+static void rx_refill_atomic(struct net_device *ndev));
 static void fastcall rx_refill_atomic(struct net_device *ndev)
 {
 	rx_refill(ndev, GFP_ATOMIC);
@@ -633,7 +633,6 @@ static inline void clear_rx_desc(struct ns83820 *dev, unsigned i)
 	build_rx_desc(dev, dev->rx_info.descs + (DESC_SIZE * i), 0, 0, CMDSTS_OWN, 0);
 }
 
-static void FASTCALL(phy_intr(struct net_device *ndev));
 static void fastcall phy_intr(struct net_device *ndev)
 {
 	struct ns83820 *dev = PRIV(ndev);
@@ -832,7 +831,6 @@ static void ns83820_cleanup_rx(struct ns83820 *dev)
 	}
 }
 
-static void FASTCALL(ns83820_rx_kick(struct net_device *ndev));
 static void fastcall ns83820_rx_kick(struct net_device *ndev)
 {
 	struct ns83820 *dev = PRIV(ndev);
@@ -854,7 +852,6 @@ static void fastcall ns83820_rx_kick(struct net_device *ndev)
 /* rx_irq
  *
  */
-static void FASTCALL(rx_irq(struct net_device *ndev));
 static void fastcall rx_irq(struct net_device *ndev)
 {
 	struct ns83820 *dev = PRIV(ndev);
diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h
index 25aa575db807..98ec7a320689 100644
--- a/include/net/bluetooth/rfcomm.h
+++ b/include/net/bluetooth/rfcomm.h
@@ -252,8 +252,8 @@ static inline void rfcomm_dlc_put(struct rfcomm_dlc *d)
 		rfcomm_dlc_free(d);
 }
 
-extern void FASTCALL(__rfcomm_dlc_throttle(struct rfcomm_dlc *d));
-extern void FASTCALL(__rfcomm_dlc_unthrottle(struct rfcomm_dlc *d));
+extern void __rfcomm_dlc_throttle(struct rfcomm_dlc *d);
+extern void __rfcomm_dlc_unthrottle(struct rfcomm_dlc *d);
 
 static inline void rfcomm_dlc_throttle(struct rfcomm_dlc *d)
 {
diff --git a/include/net/sock.h b/include/net/sock.h
index 4ce37ce8c411..416bc994adad 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -774,14 +774,14 @@ do {									\
 	lockdep_init_map(&(sk)->sk_lock.dep_map, (name), (key), 0);	\
 } while (0)
 
-extern void FASTCALL(lock_sock_nested(struct sock *sk, int subclass));
+extern void lock_sock_nested(struct sock *sk, int subclass);
 
 static inline void lock_sock(struct sock *sk)
 {
 	lock_sock_nested(sk, 0);
 }
 
-extern void FASTCALL(release_sock(struct sock *sk));
+extern void release_sock(struct sock *sk);
 
 /* BH context may only use the following locking interface. */
 #define bh_lock_sock(__sk)	spin_lock(&((__sk)->sk_lock.slock))
-- 
cgit v1.2.3


From b8599d20708fa3bde1e414689f3474560c2d990b Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Thu, 13 Dec 2007 12:25:01 -0200
Subject: [DCCP]: Support for server holding timewait state

This adds a socket option and signalling support for the case where the server
holds timewait state on closing the connection, as described in RFC 4340, 8.3.

Since holding timewait state at the server is the non-usual case, it is enabled
via a socket option. Documentation for this socket option has been added.

The setsockopt statement has been made resilient against different possible cases
of expressing boolean `true' values using a suggestion by Ian McDonald.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/dccp.txt |  6 ++++++
 include/linux/dccp.h              |  3 +++
 net/dccp/output.c                 |  6 ++++--
 net/dccp/proto.c                  | 13 ++++++++++++-
 4 files changed, 25 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt
index d76905a5a087..39131a3c78f8 100644
--- a/Documentation/networking/dccp.txt
+++ b/Documentation/networking/dccp.txt
@@ -57,6 +57,12 @@ can be set before calling bind().
 DCCP_SOCKOPT_GET_CUR_MPS is read-only and retrieves the current maximum packet
 size (application payload size) in bytes, see RFC 4340, section 14.
 
+DCCP_SOCKOPT_SERVER_TIMEWAIT enables the server (listening socket) to hold
+timewait state when closing the connection (RFC 4340, 8.3). The usual case is
+that the closing server sends a CloseReq, whereupon the client holds timewait
+state. When this boolean socket option is on, the server sends a Close instead
+and will enter TIMEWAIT. This option must be set after accept() returns.
+
 DCCP_SOCKOPT_SEND_CSCOV and DCCP_SOCKOPT_RECV_CSCOV are used for setting the
 partial checksum coverage (RFC 4340, sec. 9.2). The default is that checksums
 always cover the entire packet and that only fully covered application data is
diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 312b989c7edb..c676021603f5 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -205,6 +205,7 @@ struct dccp_so_feat {
 #define DCCP_SOCKOPT_CHANGE_L		3
 #define DCCP_SOCKOPT_CHANGE_R		4
 #define DCCP_SOCKOPT_GET_CUR_MPS	5
+#define DCCP_SOCKOPT_SERVER_TIMEWAIT	6
 #define DCCP_SOCKOPT_SEND_CSCOV		10
 #define DCCP_SOCKOPT_RECV_CSCOV		11
 #define DCCP_SOCKOPT_CCID_RX_INFO	128
@@ -492,6 +493,7 @@ struct dccp_ackvec;
  * @dccps_role - role of this sock, one of %dccp_role
  * @dccps_hc_rx_insert_options - receiver wants to add options when acking
  * @dccps_hc_tx_insert_options - sender wants to add options when sending
+ * @dccps_server_timewait - server holds timewait state on close (RFC 4340, 8.3)
  * @dccps_xmit_timer - timer for when CCID is not ready to send
  * @dccps_syn_rtt - RTT sample from Request/Response exchange (in usecs)
  */
@@ -528,6 +530,7 @@ struct dccp_sock {
 	enum dccp_role			dccps_role:2;
 	__u8				dccps_hc_rx_insert_options:1;
 	__u8				dccps_hc_tx_insert_options:1;
+	__u8				dccps_server_timewait:1;
 	struct timer_list		dccps_xmit_timer;
 };
 
diff --git a/net/dccp/output.c b/net/dccp/output.c
index e97584aa4898..b2e17910930d 100644
--- a/net/dccp/output.c
+++ b/net/dccp/output.c
@@ -567,8 +567,10 @@ void dccp_send_close(struct sock *sk, const int active)
 
 	/* Reserve space for headers and prepare control bits. */
 	skb_reserve(skb, sk->sk_prot->max_header);
-	DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ?
-					DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ;
+	if (dp->dccps_role == DCCP_ROLE_SERVER && !dp->dccps_server_timewait)
+		DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSEREQ;
+	else
+		DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSE;
 
 	if (active) {
 		dccp_write_xmit(sk, 1);
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 8a73c8f98d76..cc87c500bfb8 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -551,6 +551,12 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname,
 						     (struct dccp_so_feat __user *)
 						     optval);
 		break;
+	case DCCP_SOCKOPT_SERVER_TIMEWAIT:
+		if (dp->dccps_role != DCCP_ROLE_SERVER)
+			err = -EOPNOTSUPP;
+		else
+			dp->dccps_server_timewait = (val != 0);
+		break;
 	case DCCP_SOCKOPT_SEND_CSCOV:	/* sender side, RFC 4340, sec. 9.2 */
 		if (val < 0 || val > 15)
 			err = -EINVAL;
@@ -653,6 +659,10 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname,
 		val = dp->dccps_mss_cache;
 		len = sizeof(val);
 		break;
+	case DCCP_SOCKOPT_SERVER_TIMEWAIT:
+		val = dp->dccps_server_timewait;
+		len = sizeof(val);
+		break;
 	case DCCP_SOCKOPT_SEND_CSCOV:
 		val = dp->dccps_pcslen;
 		len = sizeof(val);
@@ -918,7 +928,8 @@ static void dccp_terminate_connection(struct sock *sk)
 	case DCCP_OPEN:
 		dccp_send_close(sk, 1);
 
-		if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER)
+		if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER &&
+		    !dccp_sk(sk)->dccps_server_timewait)
 			next_state = DCCP_ACTIVE_CLOSEREQ;
 		else
 			next_state = DCCP_CLOSING;
-- 
cgit v1.2.3


From 8b819412481494fb6861c08d360b75fabcbbfbbf Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Thu, 13 Dec 2007 12:29:24 -0200
Subject: [DCCP]: Allow to parse options on Request Sockets

The option parsing code currently only parses on full sk's. This causes a problem for
options sent during the initial handshake (in particular timestamps and feature-negotiation
options). Therefore, this patch extends the option parsing code with an additional argument
for request_socks: if it is non-NULL, options are parsed on the request socket, otherwise
the normal path (parsing on the sk) is used.

Subsequent patches, which implement feature negotiation during connection setup, make use
of this facility.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h |  5 +++--
 net/dccp/input.c     |  6 +++---
 net/dccp/ipv4.c      |  8 ++++----
 net/dccp/ipv6.c      |  8 ++++----
 net/dccp/options.c   | 34 +++++++++++++++++++++++-----------
 5 files changed, 37 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index c676021603f5..7214031461d3 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -407,8 +407,6 @@ struct dccp_opt_pend {
 
 extern void dccp_minisock_init(struct dccp_minisock *dmsk);
 
-extern int dccp_parse_options(struct sock *sk, struct sk_buff *skb);
-
 struct dccp_request_sock {
 	struct inet_request_sock dreq_inet_rsk;
 	__u64			 dreq_iss;
@@ -423,6 +421,9 @@ static inline struct dccp_request_sock *dccp_rsk(const struct request_sock *req)
 
 extern struct inet_timewait_death_row dccp_death_row;
 
+extern int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
+			      struct sk_buff *skb);
+
 struct dccp_options_received {
 	u32	dccpor_ndp; /* only 24 bits */
 	u32	dccpor_timestamp;
diff --git a/net/dccp/input.c b/net/dccp/input.c
index dacd4fd3c63c..08392ed86c25 100644
--- a/net/dccp/input.c
+++ b/net/dccp/input.c
@@ -369,7 +369,7 @@ int dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
 	if (dccp_check_seqno(sk, skb))
 		goto discard;
 
-	if (dccp_parse_options(sk, skb))
+	if (dccp_parse_options(sk, NULL, skb))
 		goto discard;
 
 	if (DCCP_SKB_CB(skb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ)
@@ -427,7 +427,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk,
 			goto out_invalid_packet;
 		}
 
-		if (dccp_parse_options(sk, skb))
+		if (dccp_parse_options(sk, NULL, skb))
 			goto out_invalid_packet;
 
 		/* Obtain usec RTT sample from SYN exchange (used by CCID 3) */
@@ -609,7 +609,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		/*
 		 * Step 8: Process options and mark acknowledgeable
 		 */
-		if (dccp_parse_options(sk, skb))
+		if (dccp_parse_options(sk, NULL, skb))
 			goto discard;
 
 		if (dcb->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ)
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index db17b83e8d3e..02fc91ce2505 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -600,11 +600,12 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (req == NULL)
 		goto drop;
 
-	if (dccp_parse_options(sk, skb))
-		goto drop_and_free;
-
 	dccp_reqsk_init(req, skb);
 
+	dreq = dccp_rsk(req);
+	if (dccp_parse_options(sk, dreq, skb))
+		goto drop_and_free;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
 
@@ -621,7 +622,6 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	 * In fact we defer setting S.GSR, S.SWL, S.SWH to
 	 * dccp_create_openreq_child.
 	 */
-	dreq = dccp_rsk(req);
 	dreq->dreq_isr	   = dcb->dccpd_seq;
 	dreq->dreq_iss	   = dccp_v4_init_sequence(skb);
 	dreq->dreq_service = service;
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index a08e2cb11915..f42b75ce7f5c 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -415,11 +415,12 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (req == NULL)
 		goto drop;
 
-	if (dccp_parse_options(sk, skb))
-		goto drop_and_free;
-
 	dccp_reqsk_init(req, skb);
 
+	dreq = dccp_rsk(req);
+	if (dccp_parse_options(sk, dreq, skb))
+		goto drop_and_free;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
 
@@ -449,7 +450,6 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	 *   In fact we defer setting S.GSR, S.SWL, S.SWH to
 	 *   dccp_create_openreq_child.
 	 */
-	dreq = dccp_rsk(req);
 	dreq->dreq_isr	   = dcb->dccpd_seq;
 	dreq->dreq_iss	   = dccp_v6_init_sequence(skb);
 	dreq->dreq_service = service;
diff --git a/net/dccp/options.c b/net/dccp/options.c
index 523250b45ea5..f496d4dc7efc 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -46,7 +46,13 @@ static u32 dccp_decode_value_var(const unsigned char *bf, const u8 len)
 	return value;
 }
 
-int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
+/**
+ * dccp_parse_options  -  Parse DCCP options present in @skb
+ * @sk: client|server|listening dccp socket (when @dreq != NULL)
+ * @dreq: request socket to use during connection setup, or NULL
+ */
+int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
+		       struct sk_buff *skb)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
 	const struct dccp_hdr *dh = dccp_hdr(skb);
@@ -92,6 +98,20 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
 				goto out_invalid_option;
 		}
 
+		/*
+		 * CCID-Specific Options (from RFC 4340, sec. 10.3):
+		 *
+		 * Option numbers 128 through 191 are for options sent from the
+		 * HC-Sender to the HC-Receiver; option numbers 192 through 255
+		 * are for options sent from the HC-Receiver to	the HC-Sender.
+		 *
+		 * CCID-specific options are ignored during connection setup, as
+		 * negotiation may still be in progress (see RFC 4340, 10.3).
+		 *
+		 */
+		if (dreq != NULL && opt >= 128)
+			goto ignore_option;
+
 		switch (opt) {
 		case DCCPO_PADDING:
 			break;
@@ -150,6 +170,7 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
 			opt_val = get_unaligned((__be32 *)value);
 			opt_recv->dccpor_timestamp = ntohl(opt_val);
 
+			/* FIXME: if dreq != NULL, don't store this on listening socket */
 			dp->dccps_timestamp_echo = opt_recv->dccpor_timestamp;
 			dp->dccps_timestamp_time = ktime_get_real();
 
@@ -213,15 +234,6 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
 			dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n",
 				      dccp_role(sk), elapsed_time);
 			break;
-			/*
-			 * From RFC 4340, sec. 10.3:
-			 *
-			 *	Option numbers 128 through 191 are for
-			 *	options sent from the HC-Sender to the
-			 *	HC-Receiver; option numbers 192 through 255
-			 *	are for options sent from the HC-Receiver to
-			 *	the HC-Sender.
-			 */
 		case 128 ... 191: {
 			const u16 idx = value - options;
 
@@ -245,7 +257,7 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
 				  "implemented, ignoring", sk, opt, len);
 			break;
 		}
-
+ignore_option:
 		if (opt != DCCPO_MANDATORY)
 			mandatory = 0;
 	}
-- 
cgit v1.2.3


From b4d4f7c70fd3361c6c889752e08ea9be304cf5f4 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Thu, 13 Dec 2007 12:37:19 -0200
Subject: [DCCP]: Handle timestamps on Request/Response exchange separately

In DCCP, timestamps can occur on packets anytime, CCID3 uses a timestamp(/echo) on the Request/Response
exchange. This patch addresses the following situation:
	* timestamps are recorded on the listening socket;
	* Responses are sent from dccp_request_sockets;
	* suppose two connections reach the listening socket with very small time in between:
	* the first timestamp value gets overwritten by the second connection request.

This is not really good, so this patch separates timestamps into
 * those which are received by the server during the initial handshake (on dccp_request_sock);
 * those which are received by the client or the client after connection establishment.

As before, a timestamp of 0 is regarded as indicating that no (meaningful) timestamp has been
received (in addition, a warning message is printed if hosts send 0-valued timestamps).

The timestamp-echoing now works as follows:
 * when a timestamp is present on the initial Request, it is placed into dreq, due to the
   call to dccp_parse_options in dccp_v{4,6}_conn_request;
 * when a timestamp is present on the Ack leading from RESPOND => OPEN, it is copied over
   from the request_sock into the child cocket in dccp_create_openreq_child;
 * timestamps received on an (established) dccp_sock are treated as before.

Since Elapsed Time is measured in hundredths of milliseconds (13.2), the new dccp_timestamp()
function is used, as it is expected that the time between receiving the timestamp and
sending the timestamp echo will be very small against the wrap-around time. As a byproduct,
this allows smaller timestamping-time fields.

Furthermore, inserting the Timestamp Echo option has been taken out of the block starting with
'!dccp_packet_without_ack()', since Timestamp Echo can be carried on any packet (5.8 and 13.3).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h | 16 +++++++++++++--
 net/dccp/minisocks.c | 21 ++++++++++++--------
 net/dccp/options.c   | 56 +++++++++++++++++++++++++++++++++-------------------
 3 files changed, 63 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 7214031461d3..484e45c7c89a 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -407,11 +407,23 @@ struct dccp_opt_pend {
 
 extern void dccp_minisock_init(struct dccp_minisock *dmsk);
 
+/**
+ * struct dccp_request_sock  -  represent DCCP-specific connection request
+ * @dreq_inet_rsk: structure inherited from
+ * @dreq_iss: initial sequence number sent on the Response (RFC 4340, 7.1)
+ * @dreq_isr: initial sequence number received on the Request
+ * @dreq_service: service code present on the Request (there is just one)
+ * The following two fields are analogous to the ones in dccp_sock:
+ * @dreq_timestamp_echo: last received timestamp to echo (13.1)
+ * @dreq_timestamp_echo: the time of receiving the last @dreq_timestamp_echo
+ */
 struct dccp_request_sock {
 	struct inet_request_sock dreq_inet_rsk;
 	__u64			 dreq_iss;
 	__u64			 dreq_isr;
 	__be32			 dreq_service;
+	__u32			 dreq_timestamp_echo;
+	__u32			 dreq_timestamp_time;
 };
 
 static inline struct dccp_request_sock *dccp_rsk(const struct request_sock *req)
@@ -477,8 +489,8 @@ struct dccp_ackvec;
  * @dccps_gar - greatest valid ack number received on a non-Sync; initialized to %dccps_iss
  * @dccps_service - first (passive sock) or unique (active sock) service code
  * @dccps_service_list - second .. last service code on passive socket
- * @dccps_timestamp_time - time of latest TIMESTAMP option
  * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option
+ * @dccps_timestamp_time - time of receiving latest @dccps_timestamp_echo
  * @dccps_l_ack_ratio - feature-local Ack Ratio
  * @dccps_r_ack_ratio - feature-remote Ack Ratio
  * @dccps_pcslen - sender   partial checksum coverage (via sockopt)
@@ -514,8 +526,8 @@ struct dccp_sock {
 	__u64				dccps_gar;
 	__be32				dccps_service;
 	struct dccp_service_list	*dccps_service_list;
-	ktime_t				dccps_timestamp_time;
 	__u32				dccps_timestamp_echo;
+	__u32				dccps_timestamp_time;
 	__u16				dccps_l_ack_ratio;
 	__u16				dccps_r_ack_ratio;
 	__u16				dccps_pcslen;
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index b1d5da61f6af..027d1814e1ab 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -117,11 +117,13 @@ struct sock *dccp_create_openreq_child(struct sock *sk,
 		struct dccp_sock *newdp = dccp_sk(newsk);
 		struct dccp_minisock *newdmsk = dccp_msk(newsk);
 
-		newdp->dccps_role	   = DCCP_ROLE_SERVER;
-		newdp->dccps_hc_rx_ackvec  = NULL;
-		newdp->dccps_service_list  = NULL;
-		newdp->dccps_service	   = dreq->dreq_service;
-		newicsk->icsk_rto	   = DCCP_TIMEOUT_INIT;
+		newdp->dccps_role	    = DCCP_ROLE_SERVER;
+		newdp->dccps_hc_rx_ackvec   = NULL;
+		newdp->dccps_service_list   = NULL;
+		newdp->dccps_service	    = dreq->dreq_service;
+		newdp->dccps_timestamp_echo = dreq->dreq_timestamp_echo;
+		newdp->dccps_timestamp_time = dreq->dreq_timestamp_time;
+		newicsk->icsk_rto	    = DCCP_TIMEOUT_INIT;
 
 		if (dccp_feat_clone(sk, newsk))
 			goto out_free;
@@ -303,9 +305,12 @@ EXPORT_SYMBOL_GPL(dccp_reqsk_send_ack);
 
 void dccp_reqsk_init(struct request_sock *req, struct sk_buff *skb)
 {
-	inet_rsk(req)->rmt_port = dccp_hdr(skb)->dccph_sport;
-	inet_rsk(req)->acked	= 0;
-	req->rcv_wnd		= sysctl_dccp_feat_sequence_window;
+	struct dccp_request_sock *dreq = dccp_rsk(req);
+
+	inet_rsk(req)->rmt_port	  = dccp_hdr(skb)->dccph_sport;
+	inet_rsk(req)->acked	  = 0;
+	req->rcv_wnd		  = sysctl_dccp_feat_sequence_window;
+	dreq->dreq_timestamp_echo = 0;
 }
 
 EXPORT_SYMBOL_GPL(dccp_reqsk_init);
diff --git a/net/dccp/options.c b/net/dccp/options.c
index f496d4dc7efc..0c996d8c79a3 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -166,16 +166,27 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 		case DCCPO_TIMESTAMP:
 			if (len != 4)
 				goto out_invalid_option;
-
+			/*
+			 * RFC 4340 13.1: "The precise time corresponding to
+			 * Timestamp Value zero is not specified". We use
+			 * zero to indicate absence of a meaningful timestamp.
+			 */
 			opt_val = get_unaligned((__be32 *)value);
-			opt_recv->dccpor_timestamp = ntohl(opt_val);
-
-			/* FIXME: if dreq != NULL, don't store this on listening socket */
-			dp->dccps_timestamp_echo = opt_recv->dccpor_timestamp;
-			dp->dccps_timestamp_time = ktime_get_real();
+			if (unlikely(opt_val == 0)) {
+				DCCP_WARN("Timestamp with zero value\n");
+				break;
+			}
 
+			if (dreq != NULL) {
+				dreq->dreq_timestamp_echo = ntohl(opt_val);
+				dreq->dreq_timestamp_time = dccp_timestamp();
+			} else {
+				opt_recv->dccpor_timestamp =
+					dp->dccps_timestamp_echo = ntohl(opt_val);
+				dp->dccps_timestamp_time = dccp_timestamp();
+			}
 			dccp_pr_debug("%s rx opt: TIMESTAMP=%u, ackno=%llu\n",
-				      dccp_role(sk), opt_recv->dccpor_timestamp,
+				      dccp_role(sk), ntohl(opt_val),
 				      (unsigned long long)
 				      DCCP_SKB_CB(skb)->dccpd_ack_seq);
 			break;
@@ -393,16 +404,24 @@ int dccp_insert_option_timestamp(struct sock *sk, struct sk_buff *skb)
 
 EXPORT_SYMBOL_GPL(dccp_insert_option_timestamp);
 
-static int dccp_insert_option_timestamp_echo(struct sock *sk,
+static int dccp_insert_option_timestamp_echo(struct dccp_sock *dp,
+					     struct dccp_request_sock *dreq,
 					     struct sk_buff *skb)
 {
-	struct dccp_sock *dp = dccp_sk(sk);
 	__be32 tstamp_echo;
-	int len, elapsed_time_len;
 	unsigned char *to;
-	const suseconds_t delta = ktime_us_delta(ktime_get_real(),
-						 dp->dccps_timestamp_time);
-	u32 elapsed_time = delta / 10;
+	u32 elapsed_time, elapsed_time_len, len;
+
+	if (dreq != NULL) {
+		elapsed_time = dccp_timestamp() - dreq->dreq_timestamp_time;
+		tstamp_echo  = htonl(dreq->dreq_timestamp_echo);
+		dreq->dreq_timestamp_echo = 0;
+	} else {
+		elapsed_time = dccp_timestamp() - dp->dccps_timestamp_time;
+		tstamp_echo  = htonl(dp->dccps_timestamp_echo);
+		dp->dccps_timestamp_echo = 0;
+	}
+
 	elapsed_time_len = dccp_elapsed_time_len(elapsed_time);
 	len = 6 + elapsed_time_len;
 
@@ -415,7 +434,6 @@ static int dccp_insert_option_timestamp_echo(struct sock *sk,
 	*to++ = DCCPO_TIMESTAMP_ECHO;
 	*to++ = len;
 
-	tstamp_echo = htonl(dp->dccps_timestamp_echo);
 	memcpy(to, &tstamp_echo, 4);
 	to += 4;
 
@@ -427,8 +445,6 @@ static int dccp_insert_option_timestamp_echo(struct sock *sk,
 		memcpy(to, &var32, 4);
 	}
 
-	dp->dccps_timestamp_echo = 0;
-	dp->dccps_timestamp_time = ktime_set(0, 0);
 	return 0;
 }
 
@@ -537,10 +553,6 @@ int dccp_insert_options(struct sock *sk, struct sk_buff *skb)
 		    dccp_ackvec_pending(dp->dccps_hc_rx_ackvec) &&
 		    dccp_insert_option_ackvec(sk, skb))
 			return -1;
-
-		if (dp->dccps_timestamp_echo != 0 &&
-		    dccp_insert_option_timestamp_echo(sk, skb))
-			return -1;
 	}
 
 	if (dp->dccps_hc_rx_insert_options) {
@@ -564,6 +576,10 @@ int dccp_insert_options(struct sock *sk, struct sk_buff *skb)
 	    dccp_insert_option_timestamp(sk, skb))
 		return -1;
 
+	if (dp->dccps_timestamp_echo != 0 &&
+	    dccp_insert_option_timestamp_echo(dp, NULL, skb))
+		return -1;
+
 	/* XXX: insert other options when appropriate */
 
 	if (DCCP_SKB_CB(skb)->dccpd_opt_len != 0) {
-- 
cgit v1.2.3


From 8afd351c772d1815347d5c46716b099fde00a579 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 16 Dec 2007 13:29:36 -0800
Subject: [NETNS]: Add the netns_ipv4 struct

The ipv4 will store its parameters inside this structure.
This one is empty now, but it will be eventually filled.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 2 ++
 include/net/netns/ipv4.h    | 9 +++++++++
 2 files changed, 11 insertions(+)
 create mode 100644 include/net/netns/ipv4.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 18da0af6192f..d04ddf2f80da 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -10,6 +10,7 @@
 
 #include <net/netns/unix.h>
 #include <net/netns/packet.h>
+#include <net/netns/ipv4.h>
 
 struct proc_dir_entry;
 struct net_device;
@@ -46,6 +47,7 @@ struct net {
 
 	struct netns_packet	packet;
 	struct netns_unix	unx;
+	struct netns_ipv4	ipv4;
 };
 
 #ifdef CONFIG_NET
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
new file mode 100644
index 000000000000..ce830d57c35c
--- /dev/null
+++ b/include/net/netns/ipv4.h
@@ -0,0 +1,9 @@
+/*
+ * ipv4 in net namespaces
+ */
+
+#ifndef __NETNS_IPV4_H__
+#define __NETNS_IPV4_H__
+struct netns_ipv4 {
+};
+#endif
-- 
cgit v1.2.3


From 32e569b7277f13c4b27bb29c761189963e49ce7a Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 16 Dec 2007 13:30:39 -0800
Subject: [IPV4]: Pass the net pointer to the arp_req_set_proxy()

This one will need to set the IPV4_DEVCONF_ALL(PROXY_ARP), but
there's no ways to get the net right in place, so we have to
pull one from the inet_ioctl's struct sock.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/arp.h  |  2 +-
 net/ipv4/af_inet.c |  2 +-
 net/ipv4/arp.c     | 28 ++++++++++++++++------------
 3 files changed, 18 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/arp.h b/include/net/arp.h
index f02664568600..36482bf9ad36 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -11,7 +11,7 @@ extern struct neigh_table arp_tbl;
 
 extern void	arp_init(void);
 extern int	arp_find(unsigned char *haddr, struct sk_buff *skb);
-extern int	arp_ioctl(unsigned int cmd, void __user *arg);
+extern int	arp_ioctl(struct net *net, unsigned int cmd, void __user *arg);
 extern void     arp_send(int type, int ptype, __be32 dest_ip,
 			 struct net_device *dev, __be32 src_ip,
 			 unsigned char *dest_hw, unsigned char *src_hw, unsigned char *th);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 5089a369e99c..03633b7b9b4a 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -798,7 +798,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		case SIOCDARP:
 		case SIOCGARP:
 		case SIOCSARP:
-			err = arp_ioctl(cmd, (void __user *)arg);
+			err = arp_ioctl(sk->sk_net, cmd, (void __user *)arg);
 			break;
 		case SIOCGIFADDR:
 		case SIOCSIFADDR:
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 477f3e5fe8d1..5daf504ba3b3 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -952,7 +952,7 @@ out_of_mem:
  *	Set (create) an ARP cache entry.
  */
 
-static int arp_req_set_proxy(struct net_device *dev, int on)
+static int arp_req_set_proxy(struct net *net, struct net_device *dev, int on)
 {
 	if (dev == NULL) {
 		IPV4_DEVCONF_ALL(PROXY_ARP) = on;
@@ -965,7 +965,8 @@ static int arp_req_set_proxy(struct net_device *dev, int on)
 	return -ENXIO;
 }
 
-static int arp_req_set_public(struct arpreq *r, struct net_device *dev)
+static int arp_req_set_public(struct net *net, struct arpreq *r,
+		struct net_device *dev)
 {
 	__be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
 	__be32 mask = ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr;
@@ -984,17 +985,18 @@ static int arp_req_set_public(struct arpreq *r, struct net_device *dev)
 		return 0;
 	}
 
-	return arp_req_set_proxy(dev, 1);
+	return arp_req_set_proxy(net, dev, 1);
 }
 
-static int arp_req_set(struct arpreq *r, struct net_device * dev)
+static int arp_req_set(struct net *net, struct arpreq *r,
+		struct net_device * dev)
 {
 	__be32 ip;
 	struct neighbour *neigh;
 	int err;
 
 	if (r->arp_flags & ATF_PUBL)
-		return arp_req_set_public(r, dev);
+		return arp_req_set_public(net, r, dev);
 
 	ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
 	if (r->arp_flags & ATF_PERM)
@@ -1080,7 +1082,8 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev)
 	return err;
 }
 
-static int arp_req_delete_public(struct arpreq *r, struct net_device *dev)
+static int arp_req_delete_public(struct net *net, struct arpreq *r,
+		struct net_device *dev)
 {
 	__be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr;
 	__be32 mask = ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr;
@@ -1091,17 +1094,18 @@ static int arp_req_delete_public(struct arpreq *r, struct net_device *dev)
 	if (mask)
 		return -EINVAL;
 
-	return arp_req_set_proxy(dev, 0);
+	return arp_req_set_proxy(net, dev, 0);
 }
 
-static int arp_req_delete(struct arpreq *r, struct net_device * dev)
+static int arp_req_delete(struct net *net, struct arpreq *r,
+		struct net_device * dev)
 {
 	int err;
 	__be32 ip;
 	struct neighbour *neigh;
 
 	if (r->arp_flags & ATF_PUBL)
-		return arp_req_delete_public(r, dev);
+		return arp_req_delete_public(net, r, dev);
 
 	ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
 	if (dev == NULL) {
@@ -1131,7 +1135,7 @@ static int arp_req_delete(struct arpreq *r, struct net_device * dev)
  *	Handle an ARP layer I/O control request.
  */
 
-int arp_ioctl(unsigned int cmd, void __user *arg)
+int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg)
 {
 	int err;
 	struct arpreq r;
@@ -1179,10 +1183,10 @@ int arp_ioctl(unsigned int cmd, void __user *arg)
 
 	switch (cmd) {
 	case SIOCDARP:
-		err = arp_req_delete(&r, dev);
+		err = arp_req_delete(net, &r, dev);
 		break;
 	case SIOCSARP:
-		err = arp_req_set(&r, dev);
+		err = arp_req_set(net, &r, dev);
 		break;
 	case SIOCGARP:
 		err = arp_req_get(&r, dev);
-- 
cgit v1.2.3


From 752d14dc6aa9d0fc8f3b25e5052596fb549e5157 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 16 Dec 2007 13:31:47 -0800
Subject: [IPV4]: Move the devinet pointers on the struct net

This is the core.

Add all and default pointers on the netns_ipv4 and register
a new pernet subsys to initialize them.

Also add the ctl_table_header to register the
net.ipv4.ip_forward ctl.

I don't allocate additional memory for init_net, but use
global devinets.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |   6 +++
 net/ipv4/devinet.c       | 105 +++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 102 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index ce830d57c35c..e06d7cfd6d64 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -4,6 +4,12 @@
 
 #ifndef __NETNS_IPV4_H__
 #define __NETNS_IPV4_H__
+struct ctl_table_header;
+struct ipv4_devconf;
+
 struct netns_ipv4 {
+	struct ctl_table_header	*forw_hdr;
+	struct ipv4_devconf	*devconf_all;
+	struct ipv4_devconf	*devconf_dflt;
 };
 #endif
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 8b2a44c4f82a..a2d48173828a 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -62,6 +62,7 @@
 #include <net/route.h>
 #include <net/ip_fib.h>
 #include <net/rtnetlink.h>
+#include <net/net_namespace.h>
 
 struct ipv4_devconf ipv4_devconf = {
 	.data = {
@@ -1497,7 +1498,7 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name,
 	devinet_ctl_path[DEVINET_CTL_PATH_DEV].procname = t->dev_name;
 	devinet_ctl_path[DEVINET_CTL_PATH_DEV].ctl_name = ctl_name;
 
-	t->sysctl_header = register_sysctl_paths(devinet_ctl_path,
+	t->sysctl_header = register_net_sysctl_table(net, devinet_ctl_path,
 			t->devinet_vars);
 	if (!t->sysctl_header)
 		goto free_procname;
@@ -1557,27 +1558,113 @@ static struct ctl_table ctl_forward_entry[] = {
 	{ },
 };
 
-static __initdata struct ctl_path net_ipv4_path[] = {
+static __net_initdata struct ctl_path net_ipv4_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
 	{ },
 };
 
+static __net_init int devinet_init_net(struct net *net)
+{
+	int err;
+	struct ctl_table *tbl;
+	struct ipv4_devconf *all, *dflt;
+	struct ctl_table_header *forw_hdr;
+
+	err = -ENOMEM;
+	all = &ipv4_devconf;
+	dflt = &ipv4_devconf_dflt;
+	tbl = ctl_forward_entry;
+
+	if (net != &init_net) {
+		all = kmemdup(all, sizeof(ipv4_devconf), GFP_KERNEL);
+		if (all == NULL)
+			goto err_alloc_all;
+
+		dflt = kmemdup(dflt, sizeof(ipv4_devconf_dflt), GFP_KERNEL);
+		if (dflt == NULL)
+			goto err_alloc_dflt;
+
+		tbl = kmemdup(tbl, sizeof(ctl_forward_entry), GFP_KERNEL);
+		if (tbl == NULL)
+			goto err_alloc_ctl;
+
+		tbl[0].data = &all->data[NET_IPV4_CONF_FORWARDING - 1];
+		tbl[0].extra1 = all;
+		tbl[0].extra2 = net;
+	}
+
+#ifdef CONFIG_SYSCTL
+	err = __devinet_sysctl_register(net, "all",
+			NET_PROTO_CONF_ALL, all);
+	if (err < 0)
+		goto err_reg_all;
+
+	err = __devinet_sysctl_register(net, "default",
+			NET_PROTO_CONF_DEFAULT, dflt);
+	if (err < 0)
+		goto err_reg_dflt;
+
+	err = -ENOMEM;
+	forw_hdr = register_net_sysctl_table(net, net_ipv4_path, tbl);
+	if (forw_hdr == NULL)
+		goto err_reg_ctl;
+#endif
+
+	net->ipv4.forw_hdr = forw_hdr;
+	net->ipv4.devconf_all = all;
+	net->ipv4.devconf_dflt = dflt;
+	return 0;
+
+#ifdef CONFIG_SYSCTL
+err_reg_ctl:
+	__devinet_sysctl_unregister(dflt);
+err_reg_dflt:
+	__devinet_sysctl_unregister(all);
+err_reg_all:
+	if (tbl != ctl_forward_entry)
+		kfree(tbl);
+#endif
+err_alloc_ctl:
+	if (dflt != &ipv4_devconf_dflt)
+		kfree(dflt);
+err_alloc_dflt:
+	if (all != &ipv4_devconf)
+		kfree(all);
+err_alloc_all:
+	return err;
+}
+
+static __net_exit void devinet_exit_net(struct net *net)
+{
+	struct ctl_table *tbl;
+
+	tbl = net->ipv4.forw_hdr->ctl_table_arg;
+#ifdef CONFIG_SYSCTL
+	unregister_net_sysctl_table(net->ipv4.forw_hdr);
+	__devinet_sysctl_unregister(net->ipv4.devconf_dflt);
+	__devinet_sysctl_unregister(net->ipv4.devconf_all);
+#endif
+	kfree(tbl);
+	kfree(net->ipv4.devconf_dflt);
+	kfree(net->ipv4.devconf_all);
+}
+
+static __net_initdata struct pernet_operations devinet_ops = {
+	.init = devinet_init_net,
+	.exit = devinet_exit_net,
+};
+
 void __init devinet_init(void)
 {
+	register_pernet_subsys(&devinet_ops);
+
 	register_gifconf(PF_INET, inet_gifconf);
 	register_netdevice_notifier(&ip_netdev_notifier);
 
 	rtnl_register(PF_INET, RTM_NEWADDR, inet_rtm_newaddr, NULL);
 	rtnl_register(PF_INET, RTM_DELADDR, inet_rtm_deladdr, NULL);
 	rtnl_register(PF_INET, RTM_GETADDR, NULL, inet_dump_ifaddr);
-#ifdef CONFIG_SYSCTL
-	__devinet_sysctl_register(&init_net, "all", NET_PROTO_CONF_ALL,
-			&ipv4_devconf);
-	__devinet_sysctl_register(&init_net, "default", NET_PROTO_CONF_DEFAULT,
-			&ipv4_devconf_dflt);
-	register_sysctl_paths(net_ipv4_path, ctl_forward_entry);
-#endif
 }
 
 EXPORT_SYMBOL(in_dev_finish_destroy);
-- 
cgit v1.2.3


From 586f12115264b767ea6a48ce081ca25a39c1e3dd Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 16 Dec 2007 13:32:48 -0800
Subject: [IPV4]: Switch users of ipv4_devconf(_all) to use the pernet one

These are scattered over the code, but almost all the
"critical" places already have the proper struct net
at hand except for snmp proc showing function and routing
rtnl handler.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h | 12 ++++++++----
 net/ipv4/arp.c             |  4 ++--
 net/ipv4/devinet.c         |  6 +++---
 net/ipv4/igmp.c            |  4 ++--
 net/ipv4/ipmr.c            |  4 ++--
 net/ipv4/proc.c            |  3 ++-
 net/ipv4/route.c           |  2 +-
 7 files changed, 20 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 962a062b44ff..b3c5081de02d 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -44,7 +44,8 @@ struct in_device
 };
 
 #define IPV4_DEVCONF(cnf, attr) ((cnf).data[NET_IPV4_CONF_ ## attr - 1])
-#define IPV4_DEVCONF_ALL(attr) IPV4_DEVCONF(ipv4_devconf, attr)
+#define IPV4_DEVCONF_ALL(net, attr) \
+	IPV4_DEVCONF((*(net)->ipv4.devconf_all), attr)
 
 static inline int ipv4_devconf_get(struct in_device *in_dev, int index)
 {
@@ -71,11 +72,14 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev)
 	ipv4_devconf_set((in_dev), NET_IPV4_CONF_ ## attr, (val))
 
 #define IN_DEV_ANDCONF(in_dev, attr) \
-	(IPV4_DEVCONF_ALL(attr) && IN_DEV_CONF_GET((in_dev), attr))
+	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) && \
+	 IN_DEV_CONF_GET((in_dev), attr))
 #define IN_DEV_ORCONF(in_dev, attr) \
-	(IPV4_DEVCONF_ALL(attr) || IN_DEV_CONF_GET((in_dev), attr))
+	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) || \
+	 IN_DEV_CONF_GET((in_dev), attr))
 #define IN_DEV_MAXCONF(in_dev, attr) \
-	(max(IPV4_DEVCONF_ALL(attr), IN_DEV_CONF_GET((in_dev), attr)))
+	(max(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr), \
+	     IN_DEV_CONF_GET((in_dev), attr)))
 
 #define IN_DEV_FORWARD(in_dev)		IN_DEV_CONF_GET((in_dev), FORWARDING)
 #define IN_DEV_MFORWARD(in_dev)		IN_DEV_ANDCONF((in_dev), MC_FORWARDING)
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 5daf504ba3b3..1102fb3d8018 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -860,7 +860,7 @@ static int arp_process(struct sk_buff *skb)
 
 	n = __neigh_lookup(&arp_tbl, &sip, dev, 0);
 
-	if (IPV4_DEVCONF_ALL(ARP_ACCEPT)) {
+	if (IPV4_DEVCONF_ALL(dev->nd_net, ARP_ACCEPT)) {
 		/* Unsolicited ARP is not accepted by default.
 		   It is possible, that this option should be enabled for some
 		   devices (strip is candidate)
@@ -955,7 +955,7 @@ out_of_mem:
 static int arp_req_set_proxy(struct net *net, struct net_device *dev, int on)
 {
 	if (dev == NULL) {
-		IPV4_DEVCONF_ALL(PROXY_ARP) = on;
+		IPV4_DEVCONF_ALL(net, PROXY_ARP) = on;
 		return 0;
 	}
 	if (__in_dev_get_rtnl(dev)) {
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 4c01c55c381d..1f21f4a2df86 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1258,9 +1258,9 @@ static void devinet_copy_dflt_conf(struct net *net, int i)
 static void inet_forward_change(struct net *net)
 {
 	struct net_device *dev;
-	int on = IPV4_DEVCONF_ALL(FORWARDING);
+	int on = IPV4_DEVCONF_ALL(net, FORWARDING);
 
-	IPV4_DEVCONF_ALL(ACCEPT_REDIRECTS) = !on;
+	IPV4_DEVCONF_ALL(net, ACCEPT_REDIRECTS) = !on;
 	IPV4_DEVCONF_DFLT(net, FORWARDING) = on;
 
 	read_lock(&dev_base_lock);
@@ -1360,7 +1360,7 @@ static int devinet_sysctl_forward(ctl_table *ctl, int write,
 	if (write && *valp != val) {
 		struct net *net = ctl->extra2;
 
-		if (valp == &IPV4_DEVCONF_ALL(FORWARDING))
+		if (valp == &IPV4_DEVCONF_ALL(net, FORWARDING))
 			inet_forward_change(net);
 		else if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING))
 			rt_cache_flush(0);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index c560a9392b1d..d3d5906e1b3d 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -130,12 +130,12 @@
  */
 
 #define IGMP_V1_SEEN(in_dev) \
-	(IPV4_DEVCONF_ALL(FORCE_IGMP_VERSION) == 1 || \
+	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 1 || \
 	 IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 1 || \
 	 ((in_dev)->mr_v1_seen && \
 	  time_before(jiffies, (in_dev)->mr_v1_seen)))
 #define IGMP_V2_SEEN(in_dev) \
-	(IPV4_DEVCONF_ALL(FORCE_IGMP_VERSION) == 2 || \
+	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 2 || \
 	 IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 2 || \
 	 ((in_dev)->mr_v2_seen && \
 	  time_before(jiffies, (in_dev)->mr_v2_seen)))
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 11879283ad52..9947f523862e 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -849,7 +849,7 @@ static void mrtsock_destruct(struct sock *sk)
 {
 	rtnl_lock();
 	if (sk == mroute_socket) {
-		IPV4_DEVCONF_ALL(MC_FORWARDING)--;
+		IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)--;
 
 		write_lock_bh(&mrt_lock);
 		mroute_socket=NULL;
@@ -898,7 +898,7 @@ int ip_mroute_setsockopt(struct sock *sk,int optname,char __user *optval,int opt
 			mroute_socket=sk;
 			write_unlock_bh(&mrt_lock);
 
-			IPV4_DEVCONF_ALL(MC_FORWARDING)++;
+			IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)++;
 		}
 		rtnl_unlock();
 		return ret;
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index ce34b281803f..41734db677be 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -309,7 +309,8 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 		seq_printf(seq, " %s", snmp4_ipstats_list[i].name);
 
 	seq_printf(seq, "\nIp: %d %d",
-		   IPV4_DEVCONF_ALL(FORWARDING) ? 1 : 2, sysctl_ip_default_ttl);
+		   IPV4_DEVCONF_ALL(&init_net, FORWARDING) ? 1 : 2,
+		   sysctl_ip_default_ttl);
 
 	for (i = 0; snmp4_ipstats_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 1b70ffd12615..36c7add8de84 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2619,7 +2619,7 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 		__be32 dst = rt->rt_dst;
 
 		if (MULTICAST(dst) && !LOCAL_MCAST(dst) &&
-		    IPV4_DEVCONF_ALL(MC_FORWARDING)) {
+		    IPV4_DEVCONF_ALL(&init_net, MC_FORWARDING)) {
 			int err = ipmr_get_route(skb, r, nowait);
 			if (err <= 0) {
 				if (!nowait) {
-- 
cgit v1.2.3


From 2658fa803111dae1353602e7f586de8e537803e2 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 16 Dec 2007 13:42:49 -0800
Subject: [IPV4]: Create ipv4_is_<type>(__be32 addr) functions

Change IPV4 specific macros LOOPBACK MULTICAST LOCAL_MCAST BADCLASS
and ZERONET macros to inline functions ipv4_is_<type>(__be32 addr)

Adds type safety and arguably some readability.

Changes since last submission:

Removed ipv4_addr_octets function
Used hex constants
Converted recently added rfc3330 macros

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 87 ++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 74 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index a8f00cac8f79..f8d607379017 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -246,21 +246,82 @@ struct sockaddr_in {
 #include <asm/byteorder.h> 
 
 #ifdef __KERNEL__
-/* Some random defines to make it easier in the kernel.. */
-#define LOOPBACK(x)	(((x) & htonl(0xff000000)) == htonl(0x7f000000))
-#define MULTICAST(x)	(((x) & htonl(0xf0000000)) == htonl(0xe0000000))
-#define BADCLASS(x)	(((x) & htonl(0xf0000000)) == htonl(0xf0000000))
-#define ZERONET(x)	(((x) & htonl(0xff000000)) == htonl(0x00000000))
-#define LOCAL_MCAST(x)	(((x) & htonl(0xFFFFFF00)) == htonl(0xE0000000))
+
+static inline bool ipv4_is_loopback(__be32 addr)
+{
+	return (addr & htonl(0xff000000)) == htonl(0x7f000000);
+}
+
+static inline bool ipv4_is_multicast(__be32 addr)
+{
+	return (addr & htonl(0xf0000000)) == htonl(0xe0000000);
+}
+
+static inline bool ipv4_is_local_multicast(__be32 addr)
+{
+	return (addr & htonl(0xffffff00)) == htonl(0xe0000000);
+}
+
+static inline bool ipv4_is_badclass(__be32 addr)
+{
+	return (addr & htonl(0xf0000000)) == htonl(0xf0000000);
+}
+
+static inline bool ipv4_is_zeronet(__be32 addr)
+{
+	return (addr & htonl(0xff000000)) == htonl(0x00000000);
+}
+
+#define LOOPBACK(x)		ipv4_is_loopback(x)
+#define MULTICAST(x)		ipv4_is_multicast(x)
+#define BADCLASS(x)		ipv4_is_badclass(x)
+#define ZERONET(x)		ipv4_is_zeronet(x)
+#define LOCAL_MCAST(x)		ipv4_is_local_multicast(x)
 
 /* Special-Use IPv4 Addresses (RFC3330) */
-#define PRIVATE_10(x)	(((x) & htonl(0xff000000)) == htonl(0x0A000000))
-#define LINKLOCAL_169(x) (((x) & htonl(0xffff0000)) == htonl(0xA9FE0000))
-#define PRIVATE_172(x)	(((x) & htonl(0xfff00000)) == htonl(0xAC100000))
-#define TEST_192(x)	(((x) & htonl(0xffffff00)) == htonl(0xC0000200))
-#define ANYCAST_6TO4(x)	(((x) & htonl(0xffffff00)) == htonl(0xC0586300))
-#define PRIVATE_192(x)	(((x) & htonl(0xffff0000)) == htonl(0xC0A80000))
-#define TEST_198(x)	(((x) & htonl(0xfffe0000)) == htonl(0xC6120000))
+
+static inline bool ipv4_is_private_10(__be32 addr)
+{
+	return (addr & htonl(0xff000000)) == htonl(0x0a000000);
+}
+
+static inline bool ipv4_is_private_172(__be32 addr)
+{
+	return (addr & htonl(0xfff00000)) == htonl(0xac100000);
+}
+
+static inline bool ipv4_is_private_192(__be32 addr)
+{
+	return (addr & htonl(0xffff0000)) == htonl(0xc0a80000);
+}
+
+static inline bool ipv4_is_linklocal_169(__be32 addr)
+{
+	return (addr & htonl(0xffff0000)) == htonl(0xa9fe0000);
+}
+
+static inline bool ipv4_is_anycast_6to4(__be32 addr)
+{
+	return (addr & htonl(0xffffff00)) == htonl(0xc0586300);
+}
+
+static inline bool ipv4_is_test_192(__be32 addr)
+{
+	return (addr & htonl(0xffffff00)) == htonl(0xc0000200);
+}
+
+static inline bool ipv4_is_test_198(__be32 addr)
+{
+	return (addr & htonl(0xfffe0000)) == htonl(0xc6120000);
+}
 #endif
 
+#define PRIVATE_10(x)		ipv4_is_private_10(x)
+#define LINKLOCAL_169(x)	ipv4_is_linklocal_169(x)
+#define PRIVATE_172(x)		ipv4_is_private_172(x)
+#define TEST_192(x)		ipv4_is_test_192(x)
+#define ANYCAST_6TO4(x)		ipv4_is_anycast_6to4(x)
+#define PRIVATE_192(x)		ipv4_is_private_192(x)
+#define TEST_198(x)		ipv4_is_test_198(x)
+
 #endif	/* _LINUX_IN_H */
-- 
cgit v1.2.3


From 3db8cda362dced00caf19865ffda3fa1028c59bc Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 16 Dec 2007 13:43:24 -0800
Subject: [IPV4] include/net: Use ipv4_is_<type>

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index c56827da0dee..1c3a5602990e 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -252,10 +252,12 @@ static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr)
 
 static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr)
 {
-	eui[0] = (ZERONET(addr) || PRIVATE_10(addr) || LOOPBACK(addr) ||
-		  LINKLOCAL_169(addr) || PRIVATE_172(addr) || TEST_192(addr) ||
-		  ANYCAST_6TO4(addr) || PRIVATE_192(addr) || TEST_198(addr) ||
-		  MULTICAST(addr) || BADCLASS(addr)) ? 0x00 : 0x02;
+	eui[0] = (ipv4_is_zeronet(addr) || ipv4_is_private_10(addr) ||
+		  ipv4_is_loopback(addr) || ipv4_is_linklocal_169(addr) ||
+		  ipv4_is_private_172(addr) || ipv4_is_test_192(addr) ||
+		  ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) ||
+		  ipv4_is_test_198(addr) || ipv4_is_multicast(addr) ||
+		  ipv4_is_badclass(addr)) ? 0x00 : 0x02;
 	eui[1] = 0;
 	eui[2] = 0x5E;
 	eui[3] = 0xFE;
-- 
cgit v1.2.3


From b5cb2bbc4c6cb489941be881e8adfe136ee45b8e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 16 Dec 2007 13:46:59 -0800
Subject: [IPV4] sctp: Use ipv4_is_<type>

Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/constants.h | 36 ++++++------------------------------
 net/sctp/protocol.c          | 12 +++++++-----
 2 files changed, 13 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h
index 05f22a6afbcd..fefcba67bd1e 100644
--- a/include/net/sctp/constants.h
+++ b/include/net/sctp/constants.h
@@ -365,36 +365,12 @@ typedef enum {
  * Also, RFC 8.4, non-unicast addresses are not considered valid SCTP
  * addresses.
  */
-#define IS_IPV4_UNUSABLE_ADDRESS(a) \
-	((htonl(INADDR_BROADCAST) == *a) || \
-	(MULTICAST(*a)) || \
-	(((unsigned char *)(a))[0] == 0) || \
-	((((unsigned char *)(a))[0] == 198) && \
-	(((unsigned char *)(a))[1] == 18) && \
-	(((unsigned char *)(a))[2] == 0)) || \
-	((((unsigned char *)(a))[0] == 192) && \
-	(((unsigned char *)(a))[1] == 88) && \
-	(((unsigned char *)(a))[2] == 99)))
-
-/* IPv4 Link-local addresses: 169.254.0.0/16.  */
-#define IS_IPV4_LINK_ADDRESS(a) \
-	((((unsigned char *)(a))[0] == 169) && \
-	(((unsigned char *)(a))[1] == 254))
-
-/* RFC 1918 "Address Allocation for Private Internets" defines the IPv4
- * private address space as the following:
- *
- * 10.0.0.0 - 10.255.255.255 (10/8 prefix)
- * 172.16.0.0.0 - 172.31.255.255 (172.16/12 prefix)
- * 192.168.0.0 - 192.168.255.255 (192.168/16 prefix)
- */
-#define IS_IPV4_PRIVATE_ADDRESS(a) \
-	((((unsigned char *)(a))[0] == 10) || \
-	((((unsigned char *)(a))[0] == 172) && \
-	(((unsigned char *)(a))[1] >= 16) && \
-	(((unsigned char *)(a))[1] < 32)) || \
-	((((unsigned char *)(a))[0] == 192) && \
-	(((unsigned char *)(a))[1] == 168)))
+#define IS_IPV4_UNUSABLE_ADDRESS(a)	    \
+	((htonl(INADDR_BROADCAST) == a) ||  \
+	 ipv4_is_multicast(a) ||	    \
+	 ipv4_is_zeronet(a) ||		    \
+	 ipv4_is_test_198(a) ||		    \
+	 ipv4_is_anycast_6to4(a))
 
 /* Flags used for the bind address copy functions.  */
 #define SCTP_ADDR6_ALLOWED	0x00000001	/* IPv6 address is allowed by
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index d50f610d1b02..dc22d7108494 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -359,7 +359,7 @@ static int sctp_v4_addr_valid(union sctp_addr *addr,
 			      const struct sk_buff *skb)
 {
 	/* Is this a non-unicast address or a unusable SCTP address? */
-	if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr))
+	if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr))
 		return 0;
 
 	/* Is this a broadcast address? */
@@ -408,13 +408,15 @@ static sctp_scope_t sctp_v4_scope(union sctp_addr *addr)
 	 */
 
 	/* Check for unusable SCTP addresses. */
-	if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr)) {
+	if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) {
 		retval =  SCTP_SCOPE_UNUSABLE;
-	} else if (LOOPBACK(addr->v4.sin_addr.s_addr)) {
+	} else if (ipv4_is_loopback(addr->v4.sin_addr.s_addr)) {
 		retval = SCTP_SCOPE_LOOPBACK;
-	} else if (IS_IPV4_LINK_ADDRESS(&addr->v4.sin_addr.s_addr)) {
+	} else if (ipv4_is_linklocal_169(addr->v4.sin_addr.s_addr)) {
 		retval = SCTP_SCOPE_LINK;
-	} else if (IS_IPV4_PRIVATE_ADDRESS(&addr->v4.sin_addr.s_addr)) {
+	} else if (ipv4_is_private_10(addr->v4.sin_addr.s_addr) ||
+		   ipv4_is_private_172(addr->v4.sin_addr.s_addr) ||
+		   ipv4_is_private_192(addr->v4.sin_addr.s_addr)) {
 		retval = SCTP_SCOPE_PRIVATE;
 	} else {
 		retval = SCTP_SCOPE_GLOBAL;
-- 
cgit v1.2.3


From 2d4d29802ff76de5af6123ef26c24ab512181223 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 16 Dec 2007 13:48:11 -0800
Subject: [IPV4]: Remove unused IPV4TYPE macros

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 14 --------------
 1 file changed, 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index f8d607379017..27d8a5ae9f75 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -272,12 +272,6 @@ static inline bool ipv4_is_zeronet(__be32 addr)
 	return (addr & htonl(0xff000000)) == htonl(0x00000000);
 }
 
-#define LOOPBACK(x)		ipv4_is_loopback(x)
-#define MULTICAST(x)		ipv4_is_multicast(x)
-#define BADCLASS(x)		ipv4_is_badclass(x)
-#define ZERONET(x)		ipv4_is_zeronet(x)
-#define LOCAL_MCAST(x)		ipv4_is_local_multicast(x)
-
 /* Special-Use IPv4 Addresses (RFC3330) */
 
 static inline bool ipv4_is_private_10(__be32 addr)
@@ -316,12 +310,4 @@ static inline bool ipv4_is_test_198(__be32 addr)
 }
 #endif
 
-#define PRIVATE_10(x)		ipv4_is_private_10(x)
-#define LINKLOCAL_169(x)	ipv4_is_linklocal_169(x)
-#define PRIVATE_172(x)		ipv4_is_private_172(x)
-#define TEST_192(x)		ipv4_is_test_192(x)
-#define ANYCAST_6TO4(x)		ipv4_is_anycast_6to4(x)
-#define PRIVATE_192(x)		ipv4_is_private_192(x)
-#define TEST_198(x)		ipv4_is_test_198(x)
-
 #endif	/* _LINUX_IN_H */
-- 
cgit v1.2.3


From 9ad0977fe10bd5d052a6db7738afe017367c2e32 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Sun, 16 Dec 2007 14:06:41 -0800
Subject: [SCTP]: Use crc32c library for checksum calculations.

The crc32c library used an identical table and algorithm
as SCTP.  Switch to using the library instead of carrying
our own table.  Using crypto layer proved to have too
much overhead compared to using the library directly.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/checksum.h |  78 ++++++++++++++++
 include/net/sctp/sctp.h     |   7 --
 net/sctp/Kconfig            |   1 +
 net/sctp/Makefile           |   2 +-
 net/sctp/crc32c.c           | 222 --------------------------------------------
 net/sctp/input.c            |   1 +
 net/sctp/output.c           |   1 +
 7 files changed, 82 insertions(+), 230 deletions(-)
 create mode 100644 include/net/sctp/checksum.h
 delete mode 100644 net/sctp/crc32c.c

(limited to 'include')

diff --git a/include/net/sctp/checksum.h b/include/net/sctp/checksum.h
new file mode 100644
index 000000000000..ba75c67cb992
--- /dev/null
+++ b/include/net/sctp/checksum.h
@@ -0,0 +1,78 @@
+/* SCTP kernel reference Implementation
+ * Copyright (c) 1999-2001 Motorola, Inc.
+ * Copyright (c) 2001-2003 International Business Machines, Corp.
+ *
+ * This file is part of the SCTP kernel reference Implementation
+ *
+ * SCTP Checksum functions
+ *
+ * The SCTP reference implementation is free software;
+ * you can redistribute it and/or modify it under the terms of
+ * the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * The SCTP reference implementation is distributed in the hope that it
+ * will be useful, but WITHOUT ANY WARRANTY; without even the implied
+ *                 ************************
+ * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ * See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with GNU CC; see the file COPYING.  If not, write to
+ * the Free Software Foundation, 59 Temple Place - Suite 330,
+ * Boston, MA 02111-1307, USA.
+ *
+ * Please send any bug reports or fixes you make to the
+ * email address(es):
+ *    lksctp developers <lksctp-developers@lists.sourceforge.net>
+ *
+ * Or submit a bug report through the following website:
+ *    http://www.sf.net/projects/lksctp
+ *
+ * Written or modified by:
+ *    Dinakaran Joseph
+ *    Jon Grimm <jgrimm@us.ibm.com>
+ *    Sridhar Samudrala <sri@us.ibm.com>
+ *
+ * Rewritten to use libcrc32c by:
+ *    Vlad Yasevich <vladislav.yasevich@hp.com>
+ *
+ * Any bugs reported given to us we will try to fix... any fixes shared will
+ * be incorporated into the next SCTP release.
+ */
+
+#include <linux/types.h>
+#include <net/sctp/sctp.h>
+#include <linux/crc32c.h>
+
+static inline __u32 sctp_start_cksum(__u8 *buffer, __u16 length)
+{
+	__u32 crc = ~(__u32) 0;
+	__u8  zero[sizeof(__u32)] = {0};
+
+	/* Optimize this routine to be SCTP specific, knowing how
+	 * to skip the checksum field of the SCTP header.
+	 */
+
+	/* Calculate CRC up to the checksum. */
+	crc = crc32c(crc, buffer, sizeof(struct sctphdr) - sizeof(__u32));
+
+	/* Skip checksum field of the header. */
+	crc = crc32c(crc, zero, sizeof(__u32));
+
+	/* Calculate the rest of the CRC. */
+	crc = crc32c(crc, &buffer[sizeof(struct sctphdr)],
+			    length - sizeof(struct sctphdr));
+	return crc;
+}
+
+static inline __u32 sctp_update_cksum(__u8 *buffer, __u16 length, __u32 crc32)
+{
+	return crc32c(crc32, buffer, length);
+}
+
+static inline __u32 sctp_end_cksum(__u32 crc32)
+{
+	return ntohl(~crc32);
+}
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 34318a33a94c..1b81ede7c2bc 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -149,13 +149,6 @@ int sctp_primitive_SEND(struct sctp_association *, void *arg);
 int sctp_primitive_REQUESTHEARTBEAT(struct sctp_association *, void *arg);
 int sctp_primitive_ASCONF(struct sctp_association *, void *arg);
 
-/*
- * sctp/crc32c.c
- */
-__u32 sctp_start_cksum(__u8 *ptr, __u16 count);
-__u32 sctp_update_cksum(__u8 *ptr, __u16 count, __u32 cksum);
-__u32 sctp_end_cksum(__u32 cksum);
-
 /*
  * sctp/input.c
  */
diff --git a/net/sctp/Kconfig b/net/sctp/Kconfig
index 5390bc792159..0b79f869c4ea 100644
--- a/net/sctp/Kconfig
+++ b/net/sctp/Kconfig
@@ -10,6 +10,7 @@ menuconfig IP_SCTP
 	select CRYPTO_HMAC
 	select CRYPTO_SHA1
 	select CRYPTO_MD5 if SCTP_HMAC_MD5
+	select LIBCRC32C
 	---help---
 	  Stream Control Transmission Protocol
 
diff --git a/net/sctp/Makefile b/net/sctp/Makefile
index 1da7204d9b42..f5356b9d5ee3 100644
--- a/net/sctp/Makefile
+++ b/net/sctp/Makefile
@@ -9,7 +9,7 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \
 	  transport.o chunk.o sm_make_chunk.o ulpevent.o \
 	  inqueue.o outqueue.o ulpqueue.o command.o \
 	  tsnmap.o bind_addr.o socket.o primitive.o \
-	  output.o input.o debug.o ssnmap.o proc.o crc32c.o \
+	  output.o input.o debug.o ssnmap.o proc.o \
 	  auth.o
 
 sctp-$(CONFIG_SCTP_DBG_OBJCNT) += objcnt.o
diff --git a/net/sctp/crc32c.c b/net/sctp/crc32c.c
deleted file mode 100644
index 181edabdb8ca..000000000000
--- a/net/sctp/crc32c.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/* SCTP kernel reference Implementation
- * Copyright (c) 1999-2001 Motorola, Inc.
- * Copyright (c) 2001-2003 International Business Machines, Corp.
- *
- * This file is part of the SCTP kernel reference Implementation
- *
- * SCTP Checksum functions
- *
- * The SCTP reference implementation is free software;
- * you can redistribute it and/or modify it under the terms of
- * the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * The SCTP reference implementation is distributed in the hope that it
- * will be useful, but WITHOUT ANY WARRANTY; without even the implied
- *                 ************************
- * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
- * See the GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with GNU CC; see the file COPYING.  If not, write to
- * the Free Software Foundation, 59 Temple Place - Suite 330,
- * Boston, MA 02111-1307, USA.
- *
- * Please send any bug reports or fixes you make to the
- * email address(es):
- *    lksctp developers <lksctp-developers@lists.sourceforge.net>
- *
- * Or submit a bug report through the following website:
- *    http://www.sf.net/projects/lksctp
- *
- * Written or modified by:
- *    Dinakaran Joseph
- *    Jon Grimm <jgrimm@us.ibm.com>
- *    Sridhar Samudrala <sri@us.ibm.com>
- *
- * Any bugs reported given to us we will try to fix... any fixes shared will
- * be incorporated into the next SCTP release.
- */
-
-/* The following code has been taken directly from
- * draft-ietf-tsvwg-sctpcsum-03.txt
- *
- * The code has now been modified specifically for SCTP knowledge.
- */
-
-#include <linux/types.h>
-#include <net/sctp/sctp.h>
-
-#define CRC32C_POLY 0x1EDC6F41
-#define CRC32C(c,d) (c=(c>>8)^crc_c[(c^(d))&0xFF])
-/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
-/* Copyright 2001, D. Otis.  Use this program, code or tables    */
-/* extracted from it, as desired without restriction.            */
-/*                                                               */
-/* 32 Bit Reflected CRC table generation for SCTP.               */
-/* To accommodate serial byte data being shifted out least       */
-/* significant bit first, the table's 32 bit words are reflected */
-/* which flips both byte and bit MS and LS positions.  The CRC   */
-/* is calculated MS bits first from the perspective of the serial*/
-/* stream.  The x^32 term is implied and the x^0 term may also   */
-/* be shown as +1.  The polynomial code used is 0x1EDC6F41.      */
-/* Castagnoli93                                                  */
-/* x^32+x^28+x^27+x^26+x^25+x^23+x^22+x^20+x^19+x^18+x^14+x^13+  */
-/* x^11+x^10+x^9+x^8+x^6+x^0                                     */
-/* Guy Castagnoli Stefan Braeuer and Martin Herrman              */
-/* "Optimization of Cyclic Redundancy-Check Codes                */
-/* with 24 and 32 Parity Bits",                                  */
-/* IEEE Transactions on Communications, Vol.41, No.6, June 1993  */
-/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
-static const __u32 crc_c[256] = {
-	0x00000000, 0xF26B8303, 0xE13B70F7, 0x1350F3F4,
-	0xC79A971F, 0x35F1141C, 0x26A1E7E8, 0xD4CA64EB,
-	0x8AD958CF, 0x78B2DBCC, 0x6BE22838, 0x9989AB3B,
-	0x4D43CFD0, 0xBF284CD3, 0xAC78BF27, 0x5E133C24,
-	0x105EC76F, 0xE235446C, 0xF165B798, 0x030E349B,
-	0xD7C45070, 0x25AFD373, 0x36FF2087, 0xC494A384,
-	0x9A879FA0, 0x68EC1CA3, 0x7BBCEF57, 0x89D76C54,
-	0x5D1D08BF, 0xAF768BBC, 0xBC267848, 0x4E4DFB4B,
-	0x20BD8EDE, 0xD2D60DDD, 0xC186FE29, 0x33ED7D2A,
-	0xE72719C1, 0x154C9AC2, 0x061C6936, 0xF477EA35,
-	0xAA64D611, 0x580F5512, 0x4B5FA6E6, 0xB93425E5,
-	0x6DFE410E, 0x9F95C20D, 0x8CC531F9, 0x7EAEB2FA,
-	0x30E349B1, 0xC288CAB2, 0xD1D83946, 0x23B3BA45,
-	0xF779DEAE, 0x05125DAD, 0x1642AE59, 0xE4292D5A,
-	0xBA3A117E, 0x4851927D, 0x5B016189, 0xA96AE28A,
-	0x7DA08661, 0x8FCB0562, 0x9C9BF696, 0x6EF07595,
-	0x417B1DBC, 0xB3109EBF, 0xA0406D4B, 0x522BEE48,
-	0x86E18AA3, 0x748A09A0, 0x67DAFA54, 0x95B17957,
-	0xCBA24573, 0x39C9C670, 0x2A993584, 0xD8F2B687,
-	0x0C38D26C, 0xFE53516F, 0xED03A29B, 0x1F682198,
-	0x5125DAD3, 0xA34E59D0, 0xB01EAA24, 0x42752927,
-	0x96BF4DCC, 0x64D4CECF, 0x77843D3B, 0x85EFBE38,
-	0xDBFC821C, 0x2997011F, 0x3AC7F2EB, 0xC8AC71E8,
-	0x1C661503, 0xEE0D9600, 0xFD5D65F4, 0x0F36E6F7,
-	0x61C69362, 0x93AD1061, 0x80FDE395, 0x72966096,
-	0xA65C047D, 0x5437877E, 0x4767748A, 0xB50CF789,
-	0xEB1FCBAD, 0x197448AE, 0x0A24BB5A, 0xF84F3859,
-	0x2C855CB2, 0xDEEEDFB1, 0xCDBE2C45, 0x3FD5AF46,
-	0x7198540D, 0x83F3D70E, 0x90A324FA, 0x62C8A7F9,
-	0xB602C312, 0x44694011, 0x5739B3E5, 0xA55230E6,
-	0xFB410CC2, 0x092A8FC1, 0x1A7A7C35, 0xE811FF36,
-	0x3CDB9BDD, 0xCEB018DE, 0xDDE0EB2A, 0x2F8B6829,
-	0x82F63B78, 0x709DB87B, 0x63CD4B8F, 0x91A6C88C,
-	0x456CAC67, 0xB7072F64, 0xA457DC90, 0x563C5F93,
-	0x082F63B7, 0xFA44E0B4, 0xE9141340, 0x1B7F9043,
-	0xCFB5F4A8, 0x3DDE77AB, 0x2E8E845F, 0xDCE5075C,
-	0x92A8FC17, 0x60C37F14, 0x73938CE0, 0x81F80FE3,
-	0x55326B08, 0xA759E80B, 0xB4091BFF, 0x466298FC,
-	0x1871A4D8, 0xEA1A27DB, 0xF94AD42F, 0x0B21572C,
-	0xDFEB33C7, 0x2D80B0C4, 0x3ED04330, 0xCCBBC033,
-	0xA24BB5A6, 0x502036A5, 0x4370C551, 0xB11B4652,
-	0x65D122B9, 0x97BAA1BA, 0x84EA524E, 0x7681D14D,
-	0x2892ED69, 0xDAF96E6A, 0xC9A99D9E, 0x3BC21E9D,
-	0xEF087A76, 0x1D63F975, 0x0E330A81, 0xFC588982,
-	0xB21572C9, 0x407EF1CA, 0x532E023E, 0xA145813D,
-	0x758FE5D6, 0x87E466D5, 0x94B49521, 0x66DF1622,
-	0x38CC2A06, 0xCAA7A905, 0xD9F75AF1, 0x2B9CD9F2,
-	0xFF56BD19, 0x0D3D3E1A, 0x1E6DCDEE, 0xEC064EED,
-	0xC38D26C4, 0x31E6A5C7, 0x22B65633, 0xD0DDD530,
-	0x0417B1DB, 0xF67C32D8, 0xE52CC12C, 0x1747422F,
-	0x49547E0B, 0xBB3FFD08, 0xA86F0EFC, 0x5A048DFF,
-	0x8ECEE914, 0x7CA56A17, 0x6FF599E3, 0x9D9E1AE0,
-	0xD3D3E1AB, 0x21B862A8, 0x32E8915C, 0xC083125F,
-	0x144976B4, 0xE622F5B7, 0xF5720643, 0x07198540,
-	0x590AB964, 0xAB613A67, 0xB831C993, 0x4A5A4A90,
-	0x9E902E7B, 0x6CFBAD78, 0x7FAB5E8C, 0x8DC0DD8F,
-	0xE330A81A, 0x115B2B19, 0x020BD8ED, 0xF0605BEE,
-	0x24AA3F05, 0xD6C1BC06, 0xC5914FF2, 0x37FACCF1,
-	0x69E9F0D5, 0x9B8273D6, 0x88D28022, 0x7AB90321,
-	0xAE7367CA, 0x5C18E4C9, 0x4F48173D, 0xBD23943E,
-	0xF36E6F75, 0x0105EC76, 0x12551F82, 0xE03E9C81,
-	0x34F4F86A, 0xC69F7B69, 0xD5CF889D, 0x27A40B9E,
-	0x79B737BA, 0x8BDCB4B9, 0x988C474D, 0x6AE7C44E,
-	0xBE2DA0A5, 0x4C4623A6, 0x5F16D052, 0xAD7D5351,
-};
-
-__u32 sctp_start_cksum(__u8 *buffer, __u16 length)
-{
-	__u32 crc32 = ~(__u32) 0;
-	__u32 i;
-
-	/* Optimize this routine to be SCTP specific, knowing how
-	 * to skip the checksum field of the SCTP header.
-	 */
-
-	/* Calculate CRC up to the checksum. */
-	for (i = 0; i < (sizeof(struct sctphdr) - sizeof(__u32)); i++)
-		CRC32C(crc32, buffer[i]);
-
-	/* Skip checksum field of the header. */
-	for (i = 0; i < sizeof(__u32); i++)
-		CRC32C(crc32, 0);
-
-	/* Calculate the rest of the CRC. */
-	for (i = sizeof(struct sctphdr); i < length ; i++)
-		CRC32C(crc32, buffer[i]);
-
-	return crc32;
-}
-
-__u32 sctp_update_cksum(__u8 *buffer, __u16 length, __u32 crc32)
-{
-	__u32 i;
-
-	for (i = 0; i < length ; i++)
-		CRC32C(crc32, buffer[i]);
-
-	return crc32;
-}
-
-#if 0
-__u32 sctp_update_copy_cksum(__u8 *to, __u8 *from, __u16 length, __u32 crc32)
-{
-	__u32 i;
-	__u32 *_to = (__u32 *)to;
-	__u32 *_from = (__u32 *)from;
-
-	for (i = 0; i < (length/4); i++) {
-		_to[i] = _from[i];
-		CRC32C(crc32, from[i*4]);
-		CRC32C(crc32, from[i*4+1]);
-		CRC32C(crc32, from[i*4+2]);
-		CRC32C(crc32, from[i*4+3]);
-	}
-
-	return crc32;
-}
-#endif  /*  0  */
-
-__u32 sctp_end_cksum(__u32 crc32)
-{
-	__u32 result;
-	__u8 byte0, byte1, byte2, byte3;
-
-	result = ~crc32;
-
-	/*  result  now holds the negated polynomial remainder;
-	 *  since the table and algorithm is "reflected" [williams95].
-	 *  That is,  result has the same value as if we mapped the message
-	 *  to a polyomial, computed the host-bit-order polynomial
-	 *  remainder, performed final negation, then did an end-for-end
-	 *  bit-reversal.
-	 *  Note that a 32-bit bit-reversal is identical to four inplace
-	 *  8-bit reversals followed by an end-for-end byteswap.
-	 *  In other words, the bytes of each bit are in the right order,
-	 *  but the bytes have been byteswapped.  So we now do an explicit
-	 *  byteswap.  On a little-endian machine, this byteswap and
-	 *  the final ntohl cancel out and could be elided.
-	 */
-	byte0 = result & 0xff;
-	byte1 = (result>>8) & 0xff;
-	byte2 = (result>>16) & 0xff;
-	byte3 = (result>>24) & 0xff;
-
-	crc32 = ((byte0 << 24) |
-		 (byte1 << 16) |
-		 (byte2 << 8)  |
-		 byte3);
-	return crc32;
-}
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 91ae463b079b..b08c7cb5c9d6 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -60,6 +60,7 @@
 #include <net/xfrm.h>
 #include <net/sctp/sctp.h>
 #include <net/sctp/sm.h>
+#include <net/sctp/checksum.h>
 
 /* Forward declarations for internal helpers. */
 static int sctp_rcv_ootb(struct sk_buff *);
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 847639d542c0..5e811b91f21c 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -60,6 +60,7 @@
 
 #include <net/sctp/sctp.h>
 #include <net/sctp/sm.h>
+#include <net/sctp/checksum.h>
 
 /* Forward declarations for private helpers. */
 static sctp_xmit_t sctp_packet_append_data(struct sctp_packet *packet,
-- 
cgit v1.2.3


From c49e5ea322c2fb43f430abb3c4a49eae1394287e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 11 Dec 2007 21:33:42 +0100
Subject: mac80211: conditionally include timestamp in radiotap information

This makes mac80211 include the low-level MAC timestamp
in the radiotap header if the driver indicated (by a new
RX flag) that the timestamp is valid.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h |  3 +++
 net/mac80211/rx.c      | 48 ++++++++++++++++++++++++++++++++----------------
 2 files changed, 35 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 0d67b331ee72..3bd970f29f29 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -350,6 +350,8 @@ struct ieee80211_tx_control {
  *	the frame.
  * @RX_FLAG_FAILED_PLCP_CRC: Set this flag if the PCLP check failed on
  *	the frame.
+ * @RX_FLAG_TSFT: The timestamp passed in the RX status (@mactime field)
+ *	is valid.
  */
 enum mac80211_rx_flags {
 	RX_FLAG_MMIC_ERROR	= 1<<0,
@@ -359,6 +361,7 @@ enum mac80211_rx_flags {
 	RX_FLAG_IV_STRIPPED	= 1<<4,
 	RX_FLAG_FAILED_FCS_CRC	= 1<<5,
 	RX_FLAG_FAILED_PLCP_CRC = 1<<6,
+	RX_FLAG_TSFT		= 1<<7,
 };
 
 /**
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index c542fc1c983d..9cd59ecbcd67 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -79,8 +79,9 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_rate *rate;
 	int needed_headroom = 0;
-	struct ieee80211_rtap_hdr {
-		struct ieee80211_radiotap_header hdr;
+	struct ieee80211_radiotap_header *rthdr;
+	__le64 *rttsft = NULL;
+	struct ieee80211_rtap_fixed_data {
 		u8 flags;
 		u8 rate;
 		__le16 chan_freq;
@@ -88,7 +89,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 		u8 antsignal;
 		u8 padding_for_rxflags;
 		__le16 rx_flags;
-	} __attribute__ ((packed)) *rthdr;
+	} __attribute__ ((packed)) *rtfixed;
 	struct sk_buff *skb, *skb2;
 	struct net_device *prev_dev = NULL;
 	int present_fcs_len = 0;
@@ -105,7 +106,8 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 	if (status->flag & RX_FLAG_RADIOTAP)
 		rtap_len = ieee80211_get_radiotap_len(origskb->data);
 	else
-		needed_headroom = sizeof(*rthdr);
+		/* room for radiotap header, always present fields and TSFT */
+		needed_headroom = sizeof(*rthdr) + sizeof(*rtfixed) + 8;
 
 	if (local->hw.flags & IEEE80211_HW_RX_INCLUDES_FCS)
 		present_fcs_len = FCS_LEN;
@@ -133,7 +135,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 		 * them allocate enough headroom to start with.
 		 */
 		if (skb_headroom(skb) < needed_headroom &&
-		    pskb_expand_head(skb, sizeof(*rthdr), 0, GFP_ATOMIC)) {
+		    pskb_expand_head(skb, needed_headroom, 0, GFP_ATOMIC)) {
 			dev_kfree_skb(skb);
 			return NULL;
 		}
@@ -152,42 +154,56 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 
 	/* if necessary, prepend radiotap information */
 	if (!(status->flag & RX_FLAG_RADIOTAP)) {
+		rtfixed = (void *) skb_push(skb, sizeof(*rtfixed));
+		rtap_len = sizeof(*rthdr) + sizeof(*rtfixed);
+		if (status->flag & RX_FLAG_TSFT) {
+			rttsft = (void *) skb_push(skb, sizeof(*rttsft));
+			rtap_len += 8;
+		}
 		rthdr = (void *) skb_push(skb, sizeof(*rthdr));
 		memset(rthdr, 0, sizeof(*rthdr));
-		rthdr->hdr.it_len = cpu_to_le16(sizeof(*rthdr));
-		rthdr->hdr.it_present =
+		memset(rtfixed, 0, sizeof(*rtfixed));
+		rthdr->it_present =
 			cpu_to_le32((1 << IEEE80211_RADIOTAP_FLAGS) |
 				    (1 << IEEE80211_RADIOTAP_RATE) |
 				    (1 << IEEE80211_RADIOTAP_CHANNEL) |
 				    (1 << IEEE80211_RADIOTAP_DB_ANTSIGNAL) |
 				    (1 << IEEE80211_RADIOTAP_RX_FLAGS));
-		rthdr->flags = local->hw.flags & IEEE80211_HW_RX_INCLUDES_FCS ?
-			       IEEE80211_RADIOTAP_F_FCS : 0;
+		rtfixed->flags = 0;
+		if (local->hw.flags & IEEE80211_HW_RX_INCLUDES_FCS)
+			rtfixed->flags |= IEEE80211_RADIOTAP_F_FCS;
+
+		if (rttsft) {
+			*rttsft = cpu_to_le64(status->mactime);
+			rthdr->it_present |=
+				cpu_to_le32(1 << IEEE80211_RADIOTAP_TSFT);
+		}
 
 		/* FIXME: when radiotap gets a 'bad PLCP' flag use it here */
-		rthdr->rx_flags = 0;
+		rtfixed->rx_flags = 0;
 		if (status->flag &
 		    (RX_FLAG_FAILED_FCS_CRC | RX_FLAG_FAILED_PLCP_CRC))
-			rthdr->rx_flags |=
+			rtfixed->rx_flags |=
 				cpu_to_le16(IEEE80211_RADIOTAP_F_RX_BADFCS);
 
 		rate = ieee80211_get_rate(local, status->phymode,
 					  status->rate);
 		if (rate)
-			rthdr->rate = rate->rate / 5;
+			rtfixed->rate = rate->rate / 5;
 
-		rthdr->chan_freq = cpu_to_le16(status->freq);
+		rtfixed->chan_freq = cpu_to_le16(status->freq);
 
 		if (status->phymode == MODE_IEEE80211A)
-			rthdr->chan_flags =
+			rtfixed->chan_flags =
 				cpu_to_le16(IEEE80211_CHAN_OFDM |
 					    IEEE80211_CHAN_5GHZ);
 		else
-			rthdr->chan_flags =
+			rtfixed->chan_flags =
 				cpu_to_le16(IEEE80211_CHAN_DYN |
 					    IEEE80211_CHAN_2GHZ);
 
-		rthdr->antsignal = status->ssi;
+		rtfixed->antsignal = status->ssi;
+		rthdr->it_len = cpu_to_le16(rtap_len);
 	}
 
 	skb_set_mac_header(skb, 0);
-- 
cgit v1.2.3


From 374fdfbc67837c1f4369eedb0f371ce3e6cce832 Mon Sep 17 00:00:00 2001
From: Dan Williams <dcbw@redhat.com>
Date: Wed, 12 Dec 2007 10:25:07 -0500
Subject: introduce WEXT scan capabilities

Introduce scan capabilities to WEXT so that userspace can do intelligent
things with scan behavior such as handling hidden SSIDs more gracefully.
If the driver reports a specific scan capability, the driver must
respect the options specified in the iw_scan_req structure when handling
the SIOCSIWSCAN call, unless it's mode or state does not allow it to do
so, in which case it must return an error.

This version switches to Dave Kilroy's suggestion of claiming unused
padding space for the scan_capa field.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireless/hostap/hostap_ioctl.c |  3 +++
 drivers/net/wireless/ipw2200.c             |  2 ++
 include/linux/wireless.h                   | 13 +++++++++++++
 net/mac80211/ieee80211_ioctl.c             |  2 ++
 4 files changed, 20 insertions(+)

(limited to 'include')

diff --git a/drivers/net/wireless/hostap/hostap_ioctl.c b/drivers/net/wireless/hostap/hostap_ioctl.c
index d8f5efcfcab9..3a57d48cc360 100644
--- a/drivers/net/wireless/hostap/hostap_ioctl.c
+++ b/drivers/net/wireless/hostap/hostap_ioctl.c
@@ -1089,6 +1089,9 @@ static int prism2_ioctl_giwrange(struct net_device *dev,
 	range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 |
 		IW_ENC_CAPA_CIPHER_TKIP | IW_ENC_CAPA_CIPHER_CCMP;
 
+	if (local->sta_fw_ver >= PRISM2_FW_VER(1,3,1))
+		range->scan_capa = IW_SCAN_CAPA_ESSID;
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c
index 003f73f89efa..be31304dfad5 100644
--- a/drivers/net/wireless/ipw2200.c
+++ b/drivers/net/wireless/ipw2200.c
@@ -8912,6 +8912,8 @@ static int ipw_wx_get_range(struct net_device *dev,
 	range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 |
 		IW_ENC_CAPA_CIPHER_TKIP | IW_ENC_CAPA_CIPHER_CCMP;
 
+	range->scan_capa = IW_SCAN_CAPA_ESSID | IW_SCAN_CAPA_TYPE;
+
 	IPW_DEBUG_WX("GET Range\n");
 	return 0;
 }
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 0987aa7a6cf5..74e84caa1e20 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -541,6 +541,16 @@
 /* Maximum size of returned data */
 #define IW_SCAN_MAX_DATA	4096	/* In bytes */
 
+/* Scan capability flags - in (struct iw_range *)->scan_capa */
+#define IW_SCAN_CAPA_NONE		0x00
+#define IW_SCAN_CAPA_ESSID		0x01
+#define IW_SCAN_CAPA_BSSID		0x02
+#define IW_SCAN_CAPA_CHANNEL	0x04
+#define IW_SCAN_CAPA_MODE		0x08
+#define IW_SCAN_CAPA_RATE		0x10
+#define IW_SCAN_CAPA_TYPE		0x20
+#define IW_SCAN_CAPA_TIME		0x40
+
 /* Max number of char in custom event - use multiple of them if needed */
 #define IW_CUSTOM_MAX		256	/* In bytes */
 
@@ -963,6 +973,9 @@ struct	iw_range
 	__u16		old_num_channels;
 	__u8		old_num_frequency;
 
+	/* Scan capabilities */
+	__u8		scan_capa; 	/* IW_SCAN_CAPA_* bit field */
+
 	/* Wireless event capability bitmasks */
 	__u32		event_capa[6];
 
diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c
index 161f3bdec41b..04ddce76135b 100644
--- a/net/mac80211/ieee80211_ioctl.c
+++ b/net/mac80211/ieee80211_ioctl.c
@@ -218,6 +218,8 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev,
 	IW_EVENT_CAPA_SET(range->event_capa, SIOCGIWAP);
 	IW_EVENT_CAPA_SET(range->event_capa, SIOCGIWSCAN);
 
+	range->scan_capa |= IW_SCAN_CAPA_ESSID;
+
 	return 0;
 }
 
-- 
cgit v1.2.3


From 8956695131b8a7878891667469899d667eb5892b Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:46:40 -0800
Subject: [NETFILTER]: x_tables: make xt_compat_match_from_user usable in
 iterator macros

Make xt_compat_match_from_user return an int to make it usable in the
*tables iterator macros and kill a now unnecessary wrapper function.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h |  4 ++--
 net/ipv4/netfilter/ip_tables.c     | 13 +------------
 net/netfilter/x_tables.c           |  5 +++--
 3 files changed, 6 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index e305f2d0d4d1..616e6f4ede59 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -382,8 +382,8 @@ extern void xt_compat_lock(int af);
 extern void xt_compat_unlock(int af);
 
 extern int xt_compat_match_offset(struct xt_match *match);
-extern void xt_compat_match_from_user(struct xt_entry_match *m,
-				      void **dstptr, int *size);
+extern int xt_compat_match_from_user(struct xt_entry_match *m,
+				     void **dstptr, int *size);
 extern int xt_compat_match_to_user(struct xt_entry_match *m,
 				   void __user **dstptr, int *size);
 
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 7d24262331a4..4586af397ef4 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1654,16 +1654,6 @@ release_matches:
 	return ret;
 }
 
-static inline int
-compat_copy_match_from_user(struct ipt_entry_match *m,
-			    void **dstptr, compat_uint_t *size,
-			    const char *name, const struct ipt_ip *ip,
-			    unsigned int hookmask)
-{
-	xt_compat_match_from_user(m, dstptr, size);
-	return 0;
-}
-
 static int
 compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 			    unsigned int *size, const char *name,
@@ -1681,8 +1671,7 @@ compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 	memcpy(de, e, sizeof(struct ipt_entry));
 
 	*dstptr += sizeof(struct compat_ipt_entry);
-	ret = IPT_MATCH_ITERATE(e, compat_copy_match_from_user, dstptr, size,
-				name, &de->ip, de->comefrom);
+	ret = IPT_MATCH_ITERATE(e, xt_compat_match_from_user, dstptr, size);
 	if (ret)
 		return ret;
 	de->target_offset = e->target_offset - (origsize - *size);
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index 07bb465d9518..b95284ee4fd4 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -342,8 +342,8 @@ int xt_compat_match_offset(struct xt_match *match)
 }
 EXPORT_SYMBOL_GPL(xt_compat_match_offset);
 
-void xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr,
-			       int *size)
+int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr,
+			      int *size)
 {
 	struct xt_match *match = m->u.kernel.match;
 	struct compat_xt_entry_match *cm = (struct compat_xt_entry_match *)m;
@@ -365,6 +365,7 @@ void xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr,
 
 	*size += off;
 	*dstptr += msize;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(xt_compat_match_from_user);
 
-- 
cgit v1.2.3


From 89c002d66aafab93814b38d8dae43fa50aec390a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:46:59 -0800
Subject: [NETFILTER]: {ip,ip6,arp}_tables: consolidate iterator macros

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h        | 43 ++++++++++++++++++++++++
 include/linux/netfilter_arp/arp_tables.h  | 17 ++--------
 include/linux/netfilter_ipv4/ip_tables.h  | 55 ++++---------------------------
 include/linux/netfilter_ipv6/ip6_tables.h | 36 +++-----------------
 4 files changed, 55 insertions(+), 96 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 616e6f4ede59..8ab754e14ec0 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -126,6 +126,49 @@ struct xt_counters_info
 
 #define XT_INV_PROTO		0x40	/* Invert the sense of PROTO. */
 
+/* fn returns 0 to continue iteration */
+#define XT_MATCH_ITERATE(type, e, fn, args...)			\
+({								\
+	unsigned int __i;					\
+	int __ret = 0;						\
+	struct xt_entry_match *__m;				\
+								\
+	for (__i = sizeof(type);				\
+	     __i < (e)->target_offset;				\
+	     __i += __m->u.match_size) {			\
+		__m = (void *)e + __i;				\
+								\
+		__ret = fn(__m , ## args);			\
+		if (__ret != 0)					\
+			break;					\
+	}							\
+	__ret;							\
+})
+
+/* fn returns 0 to continue iteration */
+#define XT_ENTRY_ITERATE_CONTINUE(type, entries, size, n, fn, args...) \
+({								\
+	unsigned int __i, __n;					\
+	int __ret = 0;						\
+	type *__entry;						\
+								\
+	for (__i = 0, __n = 0; __i < (size);			\
+	     __i += __entry->next_offset, __n++) { 		\
+		__entry = (void *)(entries) + __i;		\
+		if (__n < n)					\
+			continue;				\
+								\
+		__ret = fn(__entry , ## args);			\
+		if (__ret != 0)					\
+			break;					\
+	}							\
+	__ret;							\
+})
+
+/* fn returns 0 to continue iteration */
+#define XT_ENTRY_ITERATE(type, entries, size, fn, args...) \
+	XT_ENTRY_ITERATE_CONTINUE(type, entries, size, 0, fn, args)
+
 #ifdef __KERNEL__
 
 #include <linux/netdevice.h>
diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 2fc73fa8e37f..e44811b9be6c 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -217,21 +217,8 @@ static __inline__ struct arpt_entry_target *arpt_get_target(struct arpt_entry *e
 }
 
 /* fn returns 0 to continue iteration */
-#define ARPT_ENTRY_ITERATE(entries, size, fn, args...)		\
-({								\
-	unsigned int __i;					\
-	int __ret = 0;						\
-	struct arpt_entry *__entry;				\
-								\
-	for (__i = 0; __i < (size); __i += __entry->next_offset) { \
-		__entry = (void *)(entries) + __i;		\
-								\
-		__ret = fn(__entry , ## args);			\
-		if (__ret != 0)					\
-			break;					\
-	}							\
-	__ret;							\
-})
+#define ARPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct arpt_entry, entries, size, fn, ## args)
 
 /*
  *	Main firewall chains definitions and global var's definitions.
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index 54da61603eff..1e0cfca7f354 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -229,60 +229,17 @@ ipt_get_target(struct ipt_entry *e)
 }
 
 /* fn returns 0 to continue iteration */
-#define IPT_MATCH_ITERATE(e, fn, args...)	\
-({						\
-	unsigned int __i;			\
-	int __ret = 0;				\
-	struct ipt_entry_match *__match;	\
-						\
-	for (__i = sizeof(struct ipt_entry);	\
-	     __i < (e)->target_offset;		\
-	     __i += __match->u.match_size) {	\
-		__match = (void *)(e) + __i;	\
-						\
-		__ret = fn(__match , ## args);	\
-		if (__ret != 0)			\
-			break;			\
-	}					\
-	__ret;					\
-})
+#define IPT_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct ipt_entry, e, fn, ## args)
 
 /* fn returns 0 to continue iteration */
-#define IPT_ENTRY_ITERATE(entries, size, fn, args...)		\
-({								\
-	unsigned int __i;					\
-	int __ret = 0;						\
-	struct ipt_entry *__entry;				\
-								\
-	for (__i = 0; __i < (size); __i += __entry->next_offset) { \
-		__entry = (void *)(entries) + __i;		\
-								\
-		__ret = fn(__entry , ## args);			\
-		if (__ret != 0)					\
-			break;					\
-	}							\
-	__ret;							\
-})
+#define IPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args)
 
 /* fn returns 0 to continue iteration */
 #define IPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \
-({								\
-	unsigned int __i, __n;					\
-	int __ret = 0;						\
-	struct ipt_entry *__entry;				\
-								\
-	for (__i = 0, __n = 0; __i < (size);			\
-	     __i += __entry->next_offset, __n++) { 		\
-		__entry = (void *)(entries) + __i;		\
-		if (__n < n)					\
-			continue;				\
-								\
-		__ret = fn(__entry , ## args);			\
-		if (__ret != 0)					\
-			break;					\
-	}							\
-	__ret;							\
-})
+	XT_ENTRY_ITERATE_CONTINUE(struct ipt_entry, entries, size, n, fn, \
+				  ## args)
 
 /*
  *	Main firewall chains definitions and global var's definitions.
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 2e98654188b3..8257b52015f3 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -289,40 +289,12 @@ ip6t_get_target(struct ip6t_entry *e)
 }
 
 /* fn returns 0 to continue iteration */
-#define IP6T_MATCH_ITERATE(e, fn, args...)	\
-({						\
-	unsigned int __i;			\
-	int __ret = 0;				\
-	struct ip6t_entry_match *__m;		\
-						\
-	for (__i = sizeof(struct ip6t_entry);	\
-	     __i < (e)->target_offset;		\
-	     __i += __m->u.match_size) {	\
-		__m = (void *)(e) + __i;	\
-						\
-		__ret = fn(__m , ## args);	\
-		if (__ret != 0)			\
-			break;			\
-	}					\
-	__ret;					\
-})
+#define IP6T_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct ip6t_entry, e, fn, ## args)
 
 /* fn returns 0 to continue iteration */
-#define IP6T_ENTRY_ITERATE(entries, size, fn, args...)		\
-({								\
-	unsigned int __i;					\
-	int __ret = 0;						\
-	struct ip6t_entry *__e;					\
-								\
-	for (__i = 0; __i < (size); __i += __e->next_offset) {	\
-		__e = (void *)(entries) + __i;			\
-								\
-		__ret = fn(__e , ## args);			\
-		if (__ret != 0)					\
-			break;					\
-	}							\
-	__ret;							\
-})
+#define IP6T_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct ip6t_entry, entries, size, fn, ## args)
 
 /*
  *	Main firewall chains definitions and global var's definitions.
-- 
cgit v1.2.3


From 73cd598df46a73d6f02063f2520df115a9b88aa5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:47:32 -0800
Subject: [NETFILTER]: ip_tables: fix compat types

Use compat types and compat iterators when dealing with compat entries for
clarity. This doesn't actually make a difference for ip_tables, but is
needed for ip6_tables and arp_tables.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_tables.h | 25 ++++++++++++----
 net/ipv4/netfilter/ip_tables.c           | 51 +++++++++++++++++---------------
 2 files changed, 47 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index 1e0cfca7f354..45fcad91e67b 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -236,11 +236,6 @@ ipt_get_target(struct ipt_entry *e)
 #define IPT_ENTRY_ITERATE(entries, size, fn, args...) \
 	XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args)
 
-/* fn returns 0 to continue iteration */
-#define IPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \
-	XT_ENTRY_ITERATE_CONTINUE(struct ipt_entry, entries, size, n, fn, \
-				  ## args)
-
 /*
  *	Main firewall chains definitions and global var's definitions.
  */
@@ -316,8 +311,28 @@ struct compat_ipt_entry
 	unsigned char elems[0];
 };
 
+/* Helper functions */
+static inline struct ipt_entry_target *
+compat_ipt_get_target(struct compat_ipt_entry *e)
+{
+	return (void *)e + e->target_offset;
+}
+
 #define COMPAT_IPT_ALIGN(s) 	COMPAT_XT_ALIGN(s)
 
+/* fn returns 0 to continue iteration */
+#define COMPAT_IPT_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct compat_ipt_entry, e, fn, ## args)
+
+/* fn returns 0 to continue iteration */
+#define COMPAT_IPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct compat_ipt_entry, entries, size, fn, ## args)
+
+/* fn returns 0 to continue iteration */
+#define COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \
+	XT_ENTRY_ITERATE_CONTINUE(struct compat_ipt_entry, entries, size, n, \
+				  fn, ## args)
+
 #endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _IPTABLES_H */
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index cc896fe2fd92..d8caa1ed4879 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1559,7 +1559,7 @@ compat_release_match(struct ipt_entry_match *m, unsigned int *i)
 }
 
 static inline int
-compat_release_entry(struct ipt_entry *e, unsigned int *i)
+compat_release_entry(struct compat_ipt_entry *e, unsigned int *i)
 {
 	struct ipt_entry_target *t;
 
@@ -1567,14 +1567,14 @@ compat_release_entry(struct ipt_entry *e, unsigned int *i)
 		return 1;
 
 	/* Cleanup all matches */
-	IPT_MATCH_ITERATE(e, compat_release_match, NULL);
-	t = ipt_get_target(e);
+	COMPAT_IPT_MATCH_ITERATE(e, compat_release_match, NULL);
+	t = compat_ipt_get_target(e);
 	module_put(t->u.kernel.target->me);
 	return 0;
 }
 
 static inline int
-check_compat_entry_size_and_hooks(struct ipt_entry *e,
+check_compat_entry_size_and_hooks(struct compat_ipt_entry *e,
 				  struct xt_table_info *newinfo,
 				  unsigned int *size,
 				  unsigned char *base,
@@ -1603,19 +1603,20 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e,
 		return -EINVAL;
 	}
 
-	ret = check_entry(e, name);
+	/* For purposes of check_entry casting the compat entry is fine */
+	ret = check_entry((struct ipt_entry *)e, name);
 	if (ret)
 		return ret;
 
 	off = sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry);
 	entry_offset = (void *)e - (void *)base;
 	j = 0;
-	ret = IPT_MATCH_ITERATE(e, compat_find_calc_match, name, &e->ip,
-				e->comefrom, &off, &j);
+	ret = COMPAT_IPT_MATCH_ITERATE(e, compat_find_calc_match, name,
+				       &e->ip, e->comefrom, &off, &j);
 	if (ret != 0)
 		goto release_matches;
 
-	t = ipt_get_target(e);
+	t = compat_ipt_get_target(e);
 	target = try_then_request_module(xt_find_target(AF_INET,
 							t->u.user.name,
 							t->u.user.revision),
@@ -1643,7 +1644,7 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e,
 	}
 
 	/* Clear counters and comefrom */
-	e->counters = ((struct ipt_counters) { 0, 0 });
+	memset(&e->counters, 0, sizeof(e->counters));
 	e->comefrom = 0;
 
 	(*i)++;
@@ -1657,7 +1658,7 @@ release_matches:
 }
 
 static int
-compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
+compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr,
 			    unsigned int *size, const char *name,
 			    struct xt_table_info *newinfo, unsigned char *base)
 {
@@ -1671,15 +1672,17 @@ compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 	origsize = *size;
 	de = (struct ipt_entry *)*dstptr;
 	memcpy(de, e, sizeof(struct ipt_entry));
+	memcpy(&de->counters, &e->counters, sizeof(e->counters));
 
-	*dstptr += sizeof(struct compat_ipt_entry);
+	*dstptr += sizeof(struct ipt_entry);
 	*size += sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry);
 
-	ret = IPT_MATCH_ITERATE(e, xt_compat_match_from_user, dstptr, size);
+	ret = COMPAT_IPT_MATCH_ITERATE(e, xt_compat_match_from_user,
+				       dstptr, size);
 	if (ret)
 		return ret;
 	de->target_offset = e->target_offset - (origsize - *size);
-	t = ipt_get_target(e);
+	t = compat_ipt_get_target(e);
 	target = t->u.kernel.target;
 	xt_compat_target_from_user(t, dstptr, size);
 
@@ -1746,11 +1749,11 @@ translate_compat_table(const char *name,
 	j = 0;
 	xt_compat_lock(AF_INET);
 	/* Walk through entries, checking offsets. */
-	ret = IPT_ENTRY_ITERATE(entry0, total_size,
-				check_compat_entry_size_and_hooks,
-				info, &size, entry0,
-				entry0 + total_size,
-				hook_entries, underflows, &j, name);
+	ret = COMPAT_IPT_ENTRY_ITERATE(entry0, total_size,
+				       check_compat_entry_size_and_hooks,
+				       info, &size, entry0,
+				       entry0 + total_size,
+				       hook_entries, underflows, &j, name);
 	if (ret != 0)
 		goto out_unlock;
 
@@ -1791,9 +1794,9 @@ translate_compat_table(const char *name,
 	entry1 = newinfo->entries[raw_smp_processor_id()];
 	pos = entry1;
 	size = total_size;
-	ret = IPT_ENTRY_ITERATE(entry0, total_size,
-				compat_copy_entry_from_user, &pos, &size,
-				name, newinfo, entry1);
+	ret = COMPAT_IPT_ENTRY_ITERATE(entry0, total_size,
+				       compat_copy_entry_from_user, &pos, &size,
+				       name, newinfo, entry1);
 	compat_flush_offsets();
 	xt_compat_unlock(AF_INET);
 	if (ret)
@@ -1808,8 +1811,8 @@ translate_compat_table(const char *name,
 				name, &i);
 	if (ret) {
 		j -= i;
-		IPT_ENTRY_ITERATE_CONTINUE(entry1, newinfo->size, i,
-					   compat_release_entry, &j);
+		COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i,
+						  compat_release_entry, &j);
 		IPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i);
 		xt_free_table_info(newinfo);
 		return ret;
@@ -1828,7 +1831,7 @@ translate_compat_table(const char *name,
 free_newinfo:
 	xt_free_table_info(newinfo);
 out:
-	IPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
+	COMPAT_IPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
 	return ret;
 out_unlock:
 	compat_flush_offsets();
-- 
cgit v1.2.3


From b386d9f5960a9afce7f077edf2095fccfbb1a8e6 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:47:48 -0800
Subject: [NETFILTER]: ip_tables: move compat offset calculation to x_tables

Its needed by ip6_tables and arp_tables as well.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h |  4 +++
 net/ipv4/netfilter/ip_tables.c     | 67 +++++---------------------------------
 net/netfilter/x_tables.c           | 58 +++++++++++++++++++++++++++++++++
 3 files changed, 70 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 8ab754e14ec0..b99ede51318a 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -424,6 +424,10 @@ struct compat_xt_counters_info
 extern void xt_compat_lock(int af);
 extern void xt_compat_unlock(int af);
 
+extern int xt_compat_add_offset(int af, unsigned int offset, short delta);
+extern void xt_compat_flush_offsets(int af);
+extern short xt_compat_calc_jump(int af, unsigned int offset);
+
 extern int xt_compat_match_offset(struct xt_match *match);
 extern int xt_compat_match_from_user(struct xt_entry_match *m,
 				     void **dstptr, int *size);
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index d8caa1ed4879..07be12cc3fe3 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1014,63 +1014,12 @@ copy_entries_to_user(unsigned int total_size,
 }
 
 #ifdef CONFIG_COMPAT
-struct compat_delta {
-	struct compat_delta *next;
-	unsigned int offset;
-	short delta;
-};
-
-static struct compat_delta *compat_offsets;
-
-static int compat_add_offset(unsigned int offset, short delta)
-{
-	struct compat_delta *tmp;
-
-	tmp = kmalloc(sizeof(struct compat_delta), GFP_KERNEL);
-	if (!tmp)
-		return -ENOMEM;
-	tmp->offset = offset;
-	tmp->delta = delta;
-	if (compat_offsets) {
-		tmp->next = compat_offsets->next;
-		compat_offsets->next = tmp;
-	} else {
-		compat_offsets = tmp;
-		tmp->next = NULL;
-	}
-	return 0;
-}
-
-static void compat_flush_offsets(void)
-{
-	struct compat_delta *tmp, *next;
-
-	if (compat_offsets) {
-		for (tmp = compat_offsets; tmp; tmp = next) {
-			next = tmp->next;
-			kfree(tmp);
-		}
-		compat_offsets = NULL;
-	}
-}
-
-static short compat_calc_jump(unsigned int offset)
-{
-	struct compat_delta *tmp;
-	short delta;
-
-	for (tmp = compat_offsets, delta = 0; tmp; tmp = tmp->next)
-		if (tmp->offset < offset)
-			delta += tmp->delta;
-	return delta;
-}
-
 static void compat_standard_from_user(void *dst, void *src)
 {
 	int v = *(compat_int_t *)src;
 
 	if (v > 0)
-		v += compat_calc_jump(v);
+		v += xt_compat_calc_jump(AF_INET, v);
 	memcpy(dst, &v, sizeof(v));
 }
 
@@ -1079,7 +1028,7 @@ static int compat_standard_to_user(void __user *dst, void *src)
 	compat_int_t cv = *(int *)src;
 
 	if (cv > 0)
-		cv -= compat_calc_jump(cv);
+		cv -= xt_compat_calc_jump(AF_INET, cv);
 	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
 }
 
@@ -1104,7 +1053,7 @@ static int compat_calc_entry(struct ipt_entry *e,
 	t = ipt_get_target(e);
 	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
-	ret = compat_add_offset(entry_offset, off);
+	ret = xt_compat_add_offset(AF_INET, entry_offset, off);
 	if (ret)
 		return ret;
 
@@ -1167,7 +1116,7 @@ static int get_info(void __user *user, int *len, int compat)
 		if (compat) {
 			struct xt_table_info tmp;
 			ret = compat_table_info(private, &tmp);
-			compat_flush_offsets();
+			xt_compat_flush_offsets(AF_INET);
 			private = &tmp;
 		}
 #endif
@@ -1631,7 +1580,7 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e,
 
 	off += xt_compat_target_offset(target);
 	*size += off;
-	ret = compat_add_offset(entry_offset, off);
+	ret = xt_compat_add_offset(AF_INET, entry_offset, off);
 	if (ret)
 		goto out;
 
@@ -1797,7 +1746,7 @@ translate_compat_table(const char *name,
 	ret = COMPAT_IPT_ENTRY_ITERATE(entry0, total_size,
 				       compat_copy_entry_from_user, &pos, &size,
 				       name, newinfo, entry1);
-	compat_flush_offsets();
+	xt_compat_flush_offsets(AF_INET);
 	xt_compat_unlock(AF_INET);
 	if (ret)
 		goto free_newinfo;
@@ -1834,7 +1783,7 @@ out:
 	COMPAT_IPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
 	return ret;
 out_unlock:
-	compat_flush_offsets();
+	xt_compat_flush_offsets(AF_INET);
 	xt_compat_unlock(AF_INET);
 	goto out;
 }
@@ -1997,7 +1946,7 @@ compat_get_entries(struct compat_ipt_get_entries __user *uptr, int *len)
 				 get.size);
 			ret = -EINVAL;
 		}
-		compat_flush_offsets();
+		xt_compat_flush_offsets(AF_INET);
 		module_put(t->me);
 		xt_table_unlock(t);
 	} else
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index b95284ee4fd4..8d4fca96a4a7 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -34,12 +34,21 @@ MODULE_DESCRIPTION("[ip,ip6,arp]_tables backend module");
 
 #define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1))
 
+struct compat_delta {
+	struct compat_delta *next;
+	unsigned int offset;
+	short delta;
+};
+
 struct xt_af {
 	struct mutex mutex;
 	struct list_head match;
 	struct list_head target;
 	struct list_head tables;
+#ifdef CONFIG_COMPAT
 	struct mutex compat_mutex;
+	struct compat_delta *compat_offsets;
+#endif
 };
 
 static struct xt_af *xt;
@@ -335,6 +344,54 @@ int xt_check_match(const struct xt_match *match, unsigned short family,
 EXPORT_SYMBOL_GPL(xt_check_match);
 
 #ifdef CONFIG_COMPAT
+int xt_compat_add_offset(int af, unsigned int offset, short delta)
+{
+	struct compat_delta *tmp;
+
+	tmp = kmalloc(sizeof(struct compat_delta), GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	tmp->offset = offset;
+	tmp->delta = delta;
+
+	if (xt[af].compat_offsets) {
+		tmp->next = xt[af].compat_offsets->next;
+		xt[af].compat_offsets->next = tmp;
+	} else {
+		xt[af].compat_offsets = tmp;
+		tmp->next = NULL;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xt_compat_add_offset);
+
+void xt_compat_flush_offsets(int af)
+{
+	struct compat_delta *tmp, *next;
+
+	if (xt[af].compat_offsets) {
+		for (tmp = xt[af].compat_offsets; tmp; tmp = next) {
+			next = tmp->next;
+			kfree(tmp);
+		}
+		xt[af].compat_offsets = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(xt_compat_flush_offsets);
+
+short xt_compat_calc_jump(int af, unsigned int offset)
+{
+	struct compat_delta *tmp;
+	short delta;
+
+	for (tmp = xt[af].compat_offsets, delta = 0; tmp; tmp = tmp->next)
+		if (tmp->offset < offset)
+			delta += tmp->delta;
+	return delta;
+}
+EXPORT_SYMBOL_GPL(xt_compat_calc_jump);
+
 int xt_compat_match_offset(struct xt_match *match)
 {
 	u_int16_t csize = match->compatsize ? : match->matchsize;
@@ -873,6 +930,7 @@ static int __init xt_init(void)
 		mutex_init(&xt[i].mutex);
 #ifdef CONFIG_COMPAT
 		mutex_init(&xt[i].compat_mutex);
+		xt[i].compat_offsets = NULL;
 #endif
 		INIT_LIST_HEAD(&xt[i].target);
 		INIT_LIST_HEAD(&xt[i].match);
-- 
cgit v1.2.3


From 3bc3fe5eed5e866c0871db6d745f3bf58af004ef Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:50:37 -0800
Subject: [NETFILTER]: ip6_tables: add compat support

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6/ip6_tables.h |  35 ++
 net/compat.c                              | 106 ----
 net/ipv6/netfilter/ip6_tables.c           | 823 ++++++++++++++++++++++++++++--
 3 files changed, 802 insertions(+), 162 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 8257b52015f3..c1124826bf29 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -326,5 +326,40 @@ extern int ip6_masked_addrcmp(const struct in6_addr *addr1,
 
 #define IP6T_ALIGN(s) (((s) + (__alignof__(struct ip6t_entry)-1)) & ~(__alignof__(struct ip6t_entry)-1))
 
+#ifdef CONFIG_COMPAT
+#include <net/compat.h>
+
+struct compat_ip6t_entry
+{
+	struct ip6t_ip6 ipv6;
+	compat_uint_t nfcache;
+	u_int16_t target_offset;
+	u_int16_t next_offset;
+	compat_uint_t comefrom;
+	struct compat_xt_counters counters;
+	unsigned char elems[0];
+};
+
+static inline struct ip6t_entry_target *
+compat_ip6t_get_target(struct compat_ip6t_entry *e)
+{
+	return (void *)e + e->target_offset;
+}
+
+#define COMPAT_IP6T_ALIGN(s)	COMPAT_XT_ALIGN(s)
+
+/* fn returns 0 to continue iteration */
+#define COMPAT_IP6T_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct compat_ip6t_entry, e, fn, ## args)
+
+/* fn returns 0 to continue iteration */
+#define COMPAT_IP6T_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct compat_ip6t_entry, entries, size, fn, ## args)
+
+#define COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \
+	XT_ENTRY_ITERATE_CONTINUE(struct compat_ip6t_entry, entries, size, n, \
+				  fn, ## args)
+
+#endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _IP6_TABLES_H */
diff --git a/net/compat.c b/net/compat.c
index f4ef4c048652..80013fb69a61 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -20,7 +20,6 @@
 #include <linux/syscalls.h>
 #include <linux/filter.h>
 #include <linux/compat.h>
-#include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/security.h>
 
 #include <net/scm.h>
@@ -316,107 +315,6 @@ void scm_detach_fds_compat(struct msghdr *kmsg, struct scm_cookie *scm)
 	__scm_destroy(scm);
 }
 
-/*
- * For now, we assume that the compatibility and native version
- * of struct ipt_entry are the same - sfr.  FIXME
- */
-struct compat_ipt_replace {
-	char			name[IPT_TABLE_MAXNAMELEN];
-	u32			valid_hooks;
-	u32			num_entries;
-	u32			size;
-	u32			hook_entry[NF_INET_NUMHOOKS];
-	u32			underflow[NF_INET_NUMHOOKS];
-	u32			num_counters;
-	compat_uptr_t		counters;	/* struct ipt_counters * */
-	struct ipt_entry	entries[0];
-};
-
-static int do_netfilter_replace(int fd, int level, int optname,
-				char __user *optval, int optlen)
-{
-	struct compat_ipt_replace __user *urepl;
-	struct ipt_replace __user *repl_nat;
-	char name[IPT_TABLE_MAXNAMELEN];
-	u32 origsize, tmp32, num_counters;
-	unsigned int repl_nat_size;
-	int ret;
-	int i;
-	compat_uptr_t ucntrs;
-
-	urepl = (struct compat_ipt_replace __user *)optval;
-	if (get_user(origsize, &urepl->size))
-		return -EFAULT;
-
-	/* Hack: Causes ipchains to give correct error msg --RR */
-	if (optlen != sizeof(*urepl) + origsize)
-		return -ENOPROTOOPT;
-
-	/* XXX Assumes that size of ipt_entry is the same both in
-	 *     native and compat environments.
-	 */
-	repl_nat_size = sizeof(*repl_nat) + origsize;
-	repl_nat = compat_alloc_user_space(repl_nat_size);
-
-	ret = -EFAULT;
-	if (put_user(origsize, &repl_nat->size))
-		goto out;
-
-	if (!access_ok(VERIFY_READ, urepl, optlen) ||
-	    !access_ok(VERIFY_WRITE, repl_nat, optlen))
-		goto out;
-
-	if (__copy_from_user(name, urepl->name, sizeof(urepl->name)) ||
-	    __copy_to_user(repl_nat->name, name, sizeof(repl_nat->name)))
-		goto out;
-
-	if (__get_user(tmp32, &urepl->valid_hooks) ||
-	    __put_user(tmp32, &repl_nat->valid_hooks))
-		goto out;
-
-	if (__get_user(tmp32, &urepl->num_entries) ||
-	    __put_user(tmp32, &repl_nat->num_entries))
-		goto out;
-
-	if (__get_user(num_counters, &urepl->num_counters) ||
-	    __put_user(num_counters, &repl_nat->num_counters))
-		goto out;
-
-	if (__get_user(ucntrs, &urepl->counters) ||
-	    __put_user(compat_ptr(ucntrs), &repl_nat->counters))
-		goto out;
-
-	if (__copy_in_user(&repl_nat->entries[0],
-			   &urepl->entries[0],
-			   origsize))
-		goto out;
-
-	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
-		if (__get_user(tmp32, &urepl->hook_entry[i]) ||
-		    __put_user(tmp32, &repl_nat->hook_entry[i]) ||
-		    __get_user(tmp32, &urepl->underflow[i]) ||
-		    __put_user(tmp32, &repl_nat->underflow[i]))
-			goto out;
-	}
-
-	/*
-	 * Since struct ipt_counters just contains two u_int64_t members
-	 * we can just do the access_ok check here and pass the (converted)
-	 * pointer into the standard syscall.  We hope that the pointer is
-	 * not misaligned ...
-	 */
-	if (!access_ok(VERIFY_WRITE, compat_ptr(ucntrs),
-		       num_counters * sizeof(struct ipt_counters)))
-		goto out;
-
-
-	ret = sys_setsockopt(fd, level, optname,
-			     (char __user *)repl_nat, repl_nat_size);
-
-out:
-	return ret;
-}
-
 /*
  * A struct sock_filter is architecture independent.
  */
@@ -485,10 +383,6 @@ asmlinkage long compat_sys_setsockopt(int fd, int level, int optname,
 	int err;
 	struct socket *sock;
 
-	if (level == SOL_IPV6 && optname == IPT_SO_SET_REPLACE)
-		return do_netfilter_replace(fd, level, optname,
-					    optval, optlen);
-
 	if (optlen < 0)
 		return -EINVAL;
 
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 6fcc0d5bc27f..db0dc96be55c 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -19,9 +19,11 @@
 #include <linux/poison.h>
 #include <linux/icmpv6.h>
 #include <net/ipv6.h>
+#include <net/compat.h>
 #include <asm/uaccess.h>
 #include <linux/mutex.h>
 #include <linux/proc_fs.h>
+#include <linux/err.h>
 #include <linux/cpumask.h>
 
 #include <linux/netfilter_ipv6/ip6_tables.h>
@@ -1037,7 +1039,80 @@ copy_entries_to_user(unsigned int total_size,
 	return ret;
 }
 
-static int get_info(void __user *user, int *len)
+#ifdef CONFIG_COMPAT
+static void compat_standard_from_user(void *dst, void *src)
+{
+	int v = *(compat_int_t *)src;
+
+	if (v > 0)
+		v += xt_compat_calc_jump(AF_INET6, v);
+	memcpy(dst, &v, sizeof(v));
+}
+
+static int compat_standard_to_user(void __user *dst, void *src)
+{
+	compat_int_t cv = *(int *)src;
+
+	if (cv > 0)
+		cv -= xt_compat_calc_jump(AF_INET6, cv);
+	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
+}
+
+static inline int
+compat_calc_match(struct ip6t_entry_match *m, int *size)
+{
+	*size += xt_compat_match_offset(m->u.kernel.match);
+	return 0;
+}
+
+static int compat_calc_entry(struct ip6t_entry *e,
+			     const struct xt_table_info *info,
+			     void *base, struct xt_table_info *newinfo)
+{
+	struct ip6t_entry_target *t;
+	unsigned int entry_offset;
+	int off, i, ret;
+
+	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
+	entry_offset = (void *)e - base;
+	IP6T_MATCH_ITERATE(e, compat_calc_match, &off);
+	t = ip6t_get_target(e);
+	off += xt_compat_target_offset(t->u.kernel.target);
+	newinfo->size -= off;
+	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
+		if (info->hook_entry[i] &&
+		    (e < (struct ip6t_entry *)(base + info->hook_entry[i])))
+			newinfo->hook_entry[i] -= off;
+		if (info->underflow[i] &&
+		    (e < (struct ip6t_entry *)(base + info->underflow[i])))
+			newinfo->underflow[i] -= off;
+	}
+	return 0;
+}
+
+static int compat_table_info(const struct xt_table_info *info,
+			     struct xt_table_info *newinfo)
+{
+	void *loc_cpu_entry;
+
+	if (!newinfo || !info)
+		return -EINVAL;
+
+	/* we dont care about newinfo->entries[] */
+	memcpy(newinfo, info, offsetof(struct xt_table_info, entries));
+	newinfo->initial_entries = 0;
+	loc_cpu_entry = info->entries[raw_smp_processor_id()];
+	return IP6T_ENTRY_ITERATE(loc_cpu_entry, info->size,
+				  compat_calc_entry, info, loc_cpu_entry,
+				  newinfo);
+}
+#endif
+
+static int get_info(void __user *user, int *len, int compat)
 {
 	char name[IP6T_TABLE_MAXNAMELEN];
 	struct xt_table *t;
@@ -1053,13 +1128,24 @@ static int get_info(void __user *user, int *len)
 		return -EFAULT;
 
 	name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
-
+#ifdef CONFIG_COMPAT
+	if (compat)
+		xt_compat_lock(AF_INET6);
+#endif
 	t = try_then_request_module(xt_find_table_lock(AF_INET6, name),
 				    "ip6table_%s", name);
 	if (t && !IS_ERR(t)) {
 		struct ip6t_getinfo info;
 		struct xt_table_info *private = t->private;
 
+#ifdef CONFIG_COMPAT
+		if (compat) {
+			struct xt_table_info tmp;
+			ret = compat_table_info(private, &tmp);
+			xt_compat_flush_offsets(AF_INET6);
+			private = &tmp;
+		}
+#endif
 		info.valid_hooks = t->valid_hooks;
 		memcpy(info.hook_entry, private->hook_entry,
 		       sizeof(info.hook_entry));
@@ -1078,6 +1164,10 @@ static int get_info(void __user *user, int *len)
 		module_put(t->me);
 	} else
 		ret = t ? PTR_ERR(t) : -ENOENT;
+#ifdef CONFIG_COMPAT
+	if (compat)
+		xt_compat_unlock(AF_INET6);
+#endif
 	return ret;
 }
 
@@ -1121,65 +1211,40 @@ get_entries(struct ip6t_get_entries __user *uptr, int *len)
 }
 
 static int
-do_replace(void __user *user, unsigned int len)
+__do_replace(const char *name, unsigned int valid_hooks,
+	     struct xt_table_info *newinfo, unsigned int num_counters,
+	     void __user *counters_ptr)
 {
 	int ret;
-	struct ip6t_replace tmp;
 	struct xt_table *t;
-	struct xt_table_info *newinfo, *oldinfo;
+	struct xt_table_info *oldinfo;
 	struct xt_counters *counters;
-	void *loc_cpu_entry, *loc_cpu_old_entry;
-
-	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
-		return -EFAULT;
-
-	/* overflow check */
-	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
-		return -ENOMEM;
-
-	newinfo = xt_alloc_table_info(tmp.size);
-	if (!newinfo)
-		return -ENOMEM;
-
-	/* choose the copy that is on our node/cpu */
-	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
-	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
-			   tmp.size) != 0) {
-		ret = -EFAULT;
-		goto free_newinfo;
-	}
+	void *loc_cpu_old_entry;
 
-	counters = vmalloc_node(tmp.num_counters * sizeof(struct xt_counters),
+	ret = 0;
+	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
 				numa_node_id());
 	if (!counters) {
 		ret = -ENOMEM;
-		goto free_newinfo;
+		goto out;
 	}
 
-	ret = translate_table(tmp.name, tmp.valid_hooks,
-			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
-			      tmp.hook_entry, tmp.underflow);
-	if (ret != 0)
-		goto free_newinfo_counters;
-
-	duprintf("ip_tables: Translated table\n");
-
-	t = try_then_request_module(xt_find_table_lock(AF_INET6, tmp.name),
-				    "ip6table_%s", tmp.name);
+	t = try_then_request_module(xt_find_table_lock(AF_INET6, name),
+				    "ip6table_%s", name);
 	if (!t || IS_ERR(t)) {
 		ret = t ? PTR_ERR(t) : -ENOENT;
 		goto free_newinfo_counters_untrans;
 	}
 
 	/* You lied! */
-	if (tmp.valid_hooks != t->valid_hooks) {
+	if (valid_hooks != t->valid_hooks) {
 		duprintf("Valid hook crap: %08X vs %08X\n",
-			 tmp.valid_hooks, t->valid_hooks);
+			 valid_hooks, t->valid_hooks);
 		ret = -EINVAL;
 		goto put_module;
 	}
 
-	oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret);
+	oldinfo = xt_replace_table(t, num_counters, newinfo, &ret);
 	if (!oldinfo)
 		goto put_module;
 
@@ -1197,10 +1262,11 @@ do_replace(void __user *user, unsigned int len)
 	get_counters(oldinfo, counters);
 	/* Decrease module usage counts and free resource */
 	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
-	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL);
+	IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,
+			   NULL);
 	xt_free_table_info(oldinfo);
-	if (copy_to_user(tmp.counters, counters,
-			 sizeof(struct xt_counters) * tmp.num_counters) != 0)
+	if (copy_to_user(counters_ptr, counters,
+			 sizeof(struct xt_counters) * num_counters) != 0)
 		ret = -EFAULT;
 	vfree(counters);
 	xt_table_unlock(t);
@@ -1210,9 +1276,54 @@ do_replace(void __user *user, unsigned int len)
 	module_put(t->me);
 	xt_table_unlock(t);
  free_newinfo_counters_untrans:
-	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry,NULL);
- free_newinfo_counters:
 	vfree(counters);
+ out:
+	return ret;
+}
+
+static int
+do_replace(void __user *user, unsigned int len)
+{
+	int ret;
+	struct ip6t_replace tmp;
+	struct xt_table_info *newinfo;
+	void *loc_cpu_entry;
+
+	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+		return -EFAULT;
+
+	/* overflow check */
+	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
+		return -ENOMEM;
+
+	newinfo = xt_alloc_table_info(tmp.size);
+	if (!newinfo)
+		return -ENOMEM;
+
+	/* choose the copy that is on our node/cpu */
+	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
+	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
+			   tmp.size) != 0) {
+		ret = -EFAULT;
+		goto free_newinfo;
+	}
+
+	ret = translate_table(tmp.name, tmp.valid_hooks,
+			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
+			      tmp.hook_entry, tmp.underflow);
+	if (ret != 0)
+		goto free_newinfo;
+
+	duprintf("ip_tables: Translated table\n");
+
+	ret = __do_replace(tmp.name, tmp.valid_hooks, newinfo,
+			   tmp.num_counters, tmp.counters);
+	if (ret)
+		goto free_newinfo_untrans;
+	return 0;
+
+ free_newinfo_untrans:
+	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
  free_newinfo:
 	xt_free_table_info(newinfo);
 	return ret;
@@ -1241,31 +1352,59 @@ add_counter_to_entry(struct ip6t_entry *e,
 }
 
 static int
-do_add_counters(void __user *user, unsigned int len)
+do_add_counters(void __user *user, unsigned int len, int compat)
 {
 	unsigned int i;
-	struct xt_counters_info tmp, *paddc;
+	struct xt_counters_info tmp;
+	struct xt_counters *paddc;
+	unsigned int num_counters;
+	char *name;
+	int size;
+	void *ptmp;
 	struct xt_table_info *private;
 	struct xt_table *t;
 	int ret = 0;
 	void *loc_cpu_entry;
+#ifdef CONFIG_COMPAT
+	struct compat_xt_counters_info compat_tmp;
 
-	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+	if (compat) {
+		ptmp = &compat_tmp;
+		size = sizeof(struct compat_xt_counters_info);
+	} else
+#endif
+	{
+		ptmp = &tmp;
+		size = sizeof(struct xt_counters_info);
+	}
+
+	if (copy_from_user(ptmp, user, size) != 0)
 		return -EFAULT;
 
-	if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters))
+#ifdef CONFIG_COMPAT
+	if (compat) {
+		num_counters = compat_tmp.num_counters;
+		name = compat_tmp.name;
+	} else
+#endif
+	{
+		num_counters = tmp.num_counters;
+		name = tmp.name;
+	}
+
+	if (len != size + num_counters * sizeof(struct xt_counters))
 		return -EINVAL;
 
-	paddc = vmalloc_node(len, numa_node_id());
+	paddc = vmalloc_node(len - size, numa_node_id());
 	if (!paddc)
 		return -ENOMEM;
 
-	if (copy_from_user(paddc, user, len) != 0) {
+	if (copy_from_user(paddc, user + size, len - size) != 0) {
 		ret = -EFAULT;
 		goto free;
 	}
 
-	t = xt_find_table_lock(AF_INET6, tmp.name);
+	t = xt_find_table_lock(AF_INET6, name);
 	if (!t || IS_ERR(t)) {
 		ret = t ? PTR_ERR(t) : -ENOENT;
 		goto free;
@@ -1273,7 +1412,7 @@ do_add_counters(void __user *user, unsigned int len)
 
 	write_lock_bh(&t->lock);
 	private = t->private;
-	if (private->number != tmp.num_counters) {
+	if (private->number != num_counters) {
 		ret = -EINVAL;
 		goto unlock_up_free;
 	}
@@ -1284,7 +1423,7 @@ do_add_counters(void __user *user, unsigned int len)
 	IP6T_ENTRY_ITERATE(loc_cpu_entry,
 			  private->size,
 			  add_counter_to_entry,
-			  paddc->counters,
+			  paddc,
 			  &i);
  unlock_up_free:
 	write_unlock_bh(&t->lock);
@@ -1296,6 +1435,567 @@ do_add_counters(void __user *user, unsigned int len)
 	return ret;
 }
 
+#ifdef CONFIG_COMPAT
+struct compat_ip6t_replace {
+	char			name[IP6T_TABLE_MAXNAMELEN];
+	u32			valid_hooks;
+	u32			num_entries;
+	u32			size;
+	u32			hook_entry[NF_INET_NUMHOOKS];
+	u32			underflow[NF_INET_NUMHOOKS];
+	u32			num_counters;
+	compat_uptr_t		counters;	/* struct ip6t_counters * */
+	struct compat_ip6t_entry entries[0];
+};
+
+static int
+compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr,
+			  compat_uint_t *size, struct xt_counters *counters,
+			  unsigned int *i)
+{
+	struct ip6t_entry_target *t;
+	struct compat_ip6t_entry __user *ce;
+	u_int16_t target_offset, next_offset;
+	compat_uint_t origsize;
+	int ret;
+
+	ret = -EFAULT;
+	origsize = *size;
+	ce = (struct compat_ip6t_entry __user *)*dstptr;
+	if (copy_to_user(ce, e, sizeof(struct ip6t_entry)))
+		goto out;
+
+	if (copy_to_user(&ce->counters, &counters[*i], sizeof(counters[*i])))
+		goto out;
+
+	*dstptr += sizeof(struct compat_ip6t_entry);
+	*size -= sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
+
+	ret = IP6T_MATCH_ITERATE(e, xt_compat_match_to_user, dstptr, size);
+	target_offset = e->target_offset - (origsize - *size);
+	if (ret)
+		goto out;
+	t = ip6t_get_target(e);
+	ret = xt_compat_target_to_user(t, dstptr, size);
+	if (ret)
+		goto out;
+	ret = -EFAULT;
+	next_offset = e->next_offset - (origsize - *size);
+	if (put_user(target_offset, &ce->target_offset))
+		goto out;
+	if (put_user(next_offset, &ce->next_offset))
+		goto out;
+
+	(*i)++;
+	return 0;
+out:
+	return ret;
+}
+
+static inline int
+compat_find_calc_match(struct ip6t_entry_match *m,
+		       const char *name,
+		       const struct ip6t_ip6 *ipv6,
+		       unsigned int hookmask,
+		       int *size, int *i)
+{
+	struct xt_match *match;
+
+	match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name,
+						      m->u.user.revision),
+					"ip6t_%s", m->u.user.name);
+	if (IS_ERR(match) || !match) {
+		duprintf("compat_check_calc_match: `%s' not found\n",
+			 m->u.user.name);
+		return match ? PTR_ERR(match) : -ENOENT;
+	}
+	m->u.kernel.match = match;
+	*size += xt_compat_match_offset(match);
+
+	(*i)++;
+	return 0;
+}
+
+static inline int
+compat_release_match(struct ip6t_entry_match *m, unsigned int *i)
+{
+	if (i && (*i)-- == 0)
+		return 1;
+
+	module_put(m->u.kernel.match->me);
+	return 0;
+}
+
+static inline int
+compat_release_entry(struct compat_ip6t_entry *e, unsigned int *i)
+{
+	struct ip6t_entry_target *t;
+
+	if (i && (*i)-- == 0)
+		return 1;
+
+	/* Cleanup all matches */
+	COMPAT_IP6T_MATCH_ITERATE(e, compat_release_match, NULL);
+	t = compat_ip6t_get_target(e);
+	module_put(t->u.kernel.target->me);
+	return 0;
+}
+
+static inline int
+check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e,
+				  struct xt_table_info *newinfo,
+				  unsigned int *size,
+				  unsigned char *base,
+				  unsigned char *limit,
+				  unsigned int *hook_entries,
+				  unsigned int *underflows,
+				  unsigned int *i,
+				  const char *name)
+{
+	struct ip6t_entry_target *t;
+	struct xt_target *target;
+	unsigned int entry_offset;
+	int ret, off, h, j;
+
+	duprintf("check_compat_entry_size_and_hooks %p\n", e);
+	if ((unsigned long)e % __alignof__(struct compat_ip6t_entry) != 0
+	    || (unsigned char *)e + sizeof(struct compat_ip6t_entry) >= limit) {
+		duprintf("Bad offset %p, limit = %p\n", e, limit);
+		return -EINVAL;
+	}
+
+	if (e->next_offset < sizeof(struct compat_ip6t_entry) +
+			     sizeof(struct compat_xt_entry_target)) {
+		duprintf("checking: element %p size %u\n",
+			 e, e->next_offset);
+		return -EINVAL;
+	}
+
+	/* For purposes of check_entry casting the compat entry is fine */
+	ret = check_entry((struct ip6t_entry *)e, name);
+	if (ret)
+		return ret;
+
+	off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
+	entry_offset = (void *)e - (void *)base;
+	j = 0;
+	ret = COMPAT_IP6T_MATCH_ITERATE(e, compat_find_calc_match, name,
+					&e->ipv6, e->comefrom, &off, &j);
+	if (ret != 0)
+		goto release_matches;
+
+	t = compat_ip6t_get_target(e);
+	target = try_then_request_module(xt_find_target(AF_INET6,
+							t->u.user.name,
+							t->u.user.revision),
+					 "ip6t_%s", t->u.user.name);
+	if (IS_ERR(target) || !target) {
+		duprintf("check_compat_entry_size_and_hooks: `%s' not found\n",
+			 t->u.user.name);
+		ret = target ? PTR_ERR(target) : -ENOENT;
+		goto release_matches;
+	}
+	t->u.kernel.target = target;
+
+	off += xt_compat_target_offset(target);
+	*size += off;
+	ret = xt_compat_add_offset(AF_INET6, entry_offset, off);
+	if (ret)
+		goto out;
+
+	/* Check hooks & underflows */
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
+		if ((unsigned char *)e - base == hook_entries[h])
+			newinfo->hook_entry[h] = hook_entries[h];
+		if ((unsigned char *)e - base == underflows[h])
+			newinfo->underflow[h] = underflows[h];
+	}
+
+	/* Clear counters and comefrom */
+	memset(&e->counters, 0, sizeof(e->counters));
+	e->comefrom = 0;
+
+	(*i)++;
+	return 0;
+
+out:
+	module_put(t->u.kernel.target->me);
+release_matches:
+	IP6T_MATCH_ITERATE(e, compat_release_match, &j);
+	return ret;
+}
+
+static int
+compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr,
+			    unsigned int *size, const char *name,
+			    struct xt_table_info *newinfo, unsigned char *base)
+{
+	struct ip6t_entry_target *t;
+	struct xt_target *target;
+	struct ip6t_entry *de;
+	unsigned int origsize;
+	int ret, h;
+
+	ret = 0;
+	origsize = *size;
+	de = (struct ip6t_entry *)*dstptr;
+	memcpy(de, e, sizeof(struct ip6t_entry));
+	memcpy(&de->counters, &e->counters, sizeof(e->counters));
+
+	*dstptr += sizeof(struct ip6t_entry);
+	*size += sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry);
+
+	ret = COMPAT_IP6T_MATCH_ITERATE(e, xt_compat_match_from_user,
+					dstptr, size);
+	if (ret)
+		return ret;
+	de->target_offset = e->target_offset - (origsize - *size);
+	t = compat_ip6t_get_target(e);
+	target = t->u.kernel.target;
+	xt_compat_target_from_user(t, dstptr, size);
+
+	de->next_offset = e->next_offset - (origsize - *size);
+	for (h = 0; h < NF_INET_NUMHOOKS; h++) {
+		if ((unsigned char *)de - base < newinfo->hook_entry[h])
+			newinfo->hook_entry[h] -= origsize - *size;
+		if ((unsigned char *)de - base < newinfo->underflow[h])
+			newinfo->underflow[h] -= origsize - *size;
+	}
+	return ret;
+}
+
+static inline int compat_check_entry(struct ip6t_entry *e, const char *name,
+				     unsigned int *i)
+{
+	int j, ret;
+
+	j = 0;
+	ret = IP6T_MATCH_ITERATE(e, check_match, name, &e->ipv6,
+				 e->comefrom, &j);
+	if (ret)
+		goto cleanup_matches;
+
+	ret = check_target(e, name);
+	if (ret)
+		goto cleanup_matches;
+
+	(*i)++;
+	return 0;
+
+ cleanup_matches:
+	IP6T_MATCH_ITERATE(e, cleanup_match, &j);
+	return ret;
+}
+
+static int
+translate_compat_table(const char *name,
+		       unsigned int valid_hooks,
+		       struct xt_table_info **pinfo,
+		       void **pentry0,
+		       unsigned int total_size,
+		       unsigned int number,
+		       unsigned int *hook_entries,
+		       unsigned int *underflows)
+{
+	unsigned int i, j;
+	struct xt_table_info *newinfo, *info;
+	void *pos, *entry0, *entry1;
+	unsigned int size;
+	int ret;
+
+	info = *pinfo;
+	entry0 = *pentry0;
+	size = total_size;
+	info->number = number;
+
+	/* Init all hooks to impossible value. */
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
+		info->hook_entry[i] = 0xFFFFFFFF;
+		info->underflow[i] = 0xFFFFFFFF;
+	}
+
+	duprintf("translate_compat_table: size %u\n", info->size);
+	j = 0;
+	xt_compat_lock(AF_INET6);
+	/* Walk through entries, checking offsets. */
+	ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size,
+					check_compat_entry_size_and_hooks,
+					info, &size, entry0,
+					entry0 + total_size,
+					hook_entries, underflows, &j, name);
+	if (ret != 0)
+		goto out_unlock;
+
+	ret = -EINVAL;
+	if (j != number) {
+		duprintf("translate_compat_table: %u not %u entries\n",
+			 j, number);
+		goto out_unlock;
+	}
+
+	/* Check hooks all assigned */
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
+		/* Only hooks which are valid */
+		if (!(valid_hooks & (1 << i)))
+			continue;
+		if (info->hook_entry[i] == 0xFFFFFFFF) {
+			duprintf("Invalid hook entry %u %u\n",
+				 i, hook_entries[i]);
+			goto out_unlock;
+		}
+		if (info->underflow[i] == 0xFFFFFFFF) {
+			duprintf("Invalid underflow %u %u\n",
+				 i, underflows[i]);
+			goto out_unlock;
+		}
+	}
+
+	ret = -ENOMEM;
+	newinfo = xt_alloc_table_info(size);
+	if (!newinfo)
+		goto out_unlock;
+
+	newinfo->number = number;
+	for (i = 0; i < NF_INET_NUMHOOKS; i++) {
+		newinfo->hook_entry[i] = info->hook_entry[i];
+		newinfo->underflow[i] = info->underflow[i];
+	}
+	entry1 = newinfo->entries[raw_smp_processor_id()];
+	pos = entry1;
+	size = total_size;
+	ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size,
+					compat_copy_entry_from_user,
+					&pos, &size, name, newinfo, entry1);
+	xt_compat_flush_offsets(AF_INET6);
+	xt_compat_unlock(AF_INET6);
+	if (ret)
+		goto free_newinfo;
+
+	ret = -ELOOP;
+	if (!mark_source_chains(newinfo, valid_hooks, entry1))
+		goto free_newinfo;
+
+	i = 0;
+	ret = IP6T_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry,
+				 name, &i);
+	if (ret) {
+		j -= i;
+		COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i,
+						   compat_release_entry, &j);
+		IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i);
+		xt_free_table_info(newinfo);
+		return ret;
+	}
+
+	/* And one copy for every other CPU */
+	for_each_possible_cpu(i)
+		if (newinfo->entries[i] && newinfo->entries[i] != entry1)
+			memcpy(newinfo->entries[i], entry1, newinfo->size);
+
+	*pinfo = newinfo;
+	*pentry0 = entry1;
+	xt_free_table_info(info);
+	return 0;
+
+free_newinfo:
+	xt_free_table_info(newinfo);
+out:
+	COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
+	return ret;
+out_unlock:
+	xt_compat_flush_offsets(AF_INET6);
+	xt_compat_unlock(AF_INET6);
+	goto out;
+}
+
+static int
+compat_do_replace(void __user *user, unsigned int len)
+{
+	int ret;
+	struct compat_ip6t_replace tmp;
+	struct xt_table_info *newinfo;
+	void *loc_cpu_entry;
+
+	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+		return -EFAULT;
+
+	/* overflow check */
+	if (tmp.size >= INT_MAX / num_possible_cpus())
+		return -ENOMEM;
+	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
+		return -ENOMEM;
+
+	newinfo = xt_alloc_table_info(tmp.size);
+	if (!newinfo)
+		return -ENOMEM;
+
+	/* choose the copy that is our node/cpu */
+	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
+	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
+			   tmp.size) != 0) {
+		ret = -EFAULT;
+		goto free_newinfo;
+	}
+
+	ret = translate_compat_table(tmp.name, tmp.valid_hooks,
+				     &newinfo, &loc_cpu_entry, tmp.size,
+				     tmp.num_entries, tmp.hook_entry,
+				     tmp.underflow);
+	if (ret != 0)
+		goto free_newinfo;
+
+	duprintf("compat_do_replace: Translated table\n");
+
+	ret = __do_replace(tmp.name, tmp.valid_hooks, newinfo,
+			   tmp.num_counters, compat_ptr(tmp.counters));
+	if (ret)
+		goto free_newinfo_untrans;
+	return 0;
+
+ free_newinfo_untrans:
+	IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
+ free_newinfo:
+	xt_free_table_info(newinfo);
+	return ret;
+}
+
+static int
+compat_do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user,
+		       unsigned int len)
+{
+	int ret;
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	switch (cmd) {
+	case IP6T_SO_SET_REPLACE:
+		ret = compat_do_replace(user, len);
+		break;
+
+	case IP6T_SO_SET_ADD_COUNTERS:
+		ret = do_add_counters(user, len, 1);
+		break;
+
+	default:
+		duprintf("do_ip6t_set_ctl:  unknown request %i\n", cmd);
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+struct compat_ip6t_get_entries {
+	char name[IP6T_TABLE_MAXNAMELEN];
+	compat_uint_t size;
+	struct compat_ip6t_entry entrytable[0];
+};
+
+static int
+compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table,
+			    void __user *userptr)
+{
+	struct xt_counters *counters;
+	struct xt_table_info *private = table->private;
+	void __user *pos;
+	unsigned int size;
+	int ret = 0;
+	void *loc_cpu_entry;
+	unsigned int i = 0;
+
+	counters = alloc_counters(table);
+	if (IS_ERR(counters))
+		return PTR_ERR(counters);
+
+	/* choose the copy that is on our node/cpu, ...
+	 * This choice is lazy (because current thread is
+	 * allowed to migrate to another cpu)
+	 */
+	loc_cpu_entry = private->entries[raw_smp_processor_id()];
+	pos = userptr;
+	size = total_size;
+	ret = IP6T_ENTRY_ITERATE(loc_cpu_entry, total_size,
+				 compat_copy_entry_to_user,
+				 &pos, &size, counters, &i);
+
+	vfree(counters);
+	return ret;
+}
+
+static int
+compat_get_entries(struct compat_ip6t_get_entries __user *uptr, int *len)
+{
+	int ret;
+	struct compat_ip6t_get_entries get;
+	struct xt_table *t;
+
+	if (*len < sizeof(get)) {
+		duprintf("compat_get_entries: %u < %u\n",
+			 *len, (unsigned int)sizeof(get));
+		return -EINVAL;
+	}
+
+	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
+		return -EFAULT;
+
+	if (*len != sizeof(struct compat_ip6t_get_entries) + get.size) {
+		duprintf("compat_get_entries: %u != %u\n", *len,
+			 (unsigned int)(sizeof(struct compat_ip6t_get_entries) +
+			 get.size));
+		return -EINVAL;
+	}
+
+	xt_compat_lock(AF_INET6);
+	t = xt_find_table_lock(AF_INET6, get.name);
+	if (t && !IS_ERR(t)) {
+		struct xt_table_info *private = t->private;
+		struct xt_table_info info;
+		duprintf("t->private->number = %u\n",
+			 private->number);
+		ret = compat_table_info(private, &info);
+		if (!ret && get.size == info.size) {
+			ret = compat_copy_entries_to_user(private->size,
+							  t, uptr->entrytable);
+		} else if (!ret) {
+			duprintf("compat_get_entries: I've got %u not %u!\n",
+				 private->size,
+				 get.size);
+			ret = -EINVAL;
+		}
+		xt_compat_flush_offsets(AF_INET6);
+		module_put(t->me);
+		xt_table_unlock(t);
+	} else
+		ret = t ? PTR_ERR(t) : -ENOENT;
+
+	xt_compat_unlock(AF_INET6);
+	return ret;
+}
+
+static int do_ip6t_get_ctl(struct sock *, int, void __user *, int *);
+
+static int
+compat_do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
+{
+	int ret;
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	switch (cmd) {
+	case IP6T_SO_GET_INFO:
+		ret = get_info(user, len, 1);
+		break;
+	case IP6T_SO_GET_ENTRIES:
+		ret = compat_get_entries(user, len);
+		break;
+	default:
+		ret = do_ip6t_get_ctl(sk, cmd, user, len);
+	}
+	return ret;
+}
+#endif
+
 static int
 do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
 {
@@ -1310,7 +2010,7 @@ do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
 		break;
 
 	case IP6T_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(user, len);
+		ret = do_add_counters(user, len, 0);
 		break;
 
 	default:
@@ -1331,7 +2031,7 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	switch (cmd) {
 	case IP6T_SO_GET_INFO:
-		ret = get_info(user, len);
+		ret = get_info(user, len, 0);
 		break;
 
 	case IP6T_SO_GET_ENTRIES:
@@ -1483,6 +2183,11 @@ static struct xt_target ip6t_standard_target __read_mostly = {
 	.name		= IP6T_STANDARD_TARGET,
 	.targetsize	= sizeof(int),
 	.family		= AF_INET6,
+#ifdef CONFIG_COMPAT
+	.compatsize	= sizeof(compat_int_t),
+	.compat_from_user = compat_standard_from_user,
+	.compat_to_user	= compat_standard_to_user,
+#endif
 };
 
 static struct xt_target ip6t_error_target __read_mostly = {
@@ -1497,9 +2202,15 @@ static struct nf_sockopt_ops ip6t_sockopts = {
 	.set_optmin	= IP6T_BASE_CTL,
 	.set_optmax	= IP6T_SO_SET_MAX+1,
 	.set		= do_ip6t_set_ctl,
+#ifdef CONFIG_COMPAT
+	.compat_set	= compat_do_ip6t_set_ctl,
+#endif
 	.get_optmin	= IP6T_BASE_CTL,
 	.get_optmax	= IP6T_SO_GET_MAX+1,
 	.get		= do_ip6t_get_ctl,
+#ifdef CONFIG_COMPAT
+	.compat_get	= compat_do_ip6t_get_ctl,
+#endif
 	.owner		= THIS_MODULE,
 };
 
-- 
cgit v1.2.3


From 06e1374a7ed45f1788353a2944a20133adc55649 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:53:40 -0800
Subject: [NETFILTER]: ip6_tables: use XT_ALIGN

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6/ip6_tables.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index c1124826bf29..110801d699ee 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -324,7 +324,7 @@ extern int ip6_masked_addrcmp(const struct in6_addr *addr1,
 			      const struct in6_addr *mask,
 			      const struct in6_addr *addr2);
 
-#define IP6T_ALIGN(s) (((s) + (__alignof__(struct ip6t_entry)-1)) & ~(__alignof__(struct ip6t_entry)-1))
+#define IP6T_ALIGN(s) XT_ALIGN(s)
 
 #ifdef CONFIG_COMPAT
 #include <net/compat.h>
-- 
cgit v1.2.3


From 0495cf957bfacbca123cb4c4e1c4cf0e265f522e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 21:55:34 -0800
Subject: [NETFILTER]: arp_tables: use XT_ALIGN

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_arp/arp_tables.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index e44811b9be6c..7ade26b025a4 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -280,6 +280,6 @@ extern unsigned int arpt_do_table(struct sk_buff *skb,
 				  const struct net_device *out,
 				  struct arpt_table *table);
 
-#define ARPT_ALIGN(s) (((s) + (__alignof__(struct arpt_entry)-1)) & ~(__alignof__(struct arpt_entry)-1))
+#define ARPT_ALIGN(s) XT_ALIGN(s)
 #endif /*__KERNEL__*/
 #endif /* _ARPTABLES_H */
-- 
cgit v1.2.3


From d6a2ba07c31b0497fc82a8c175400ea8747da2ef Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:26:54 -0800
Subject: [NETFILTER]: arp_tables: add compat support

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_arp/arp_tables.h |  31 ++
 net/ipv4/netfilter/arp_tables.c          | 748 ++++++++++++++++++++++++++++---
 2 files changed, 721 insertions(+), 58 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 7ade26b025a4..53dd4df27aa1 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -281,5 +281,36 @@ extern unsigned int arpt_do_table(struct sk_buff *skb,
 				  struct arpt_table *table);
 
 #define ARPT_ALIGN(s) XT_ALIGN(s)
+
+#ifdef CONFIG_COMPAT
+#include <net/compat.h>
+
+struct compat_arpt_entry
+{
+	struct arpt_arp arp;
+	u_int16_t target_offset;
+	u_int16_t next_offset;
+	compat_uint_t comefrom;
+	struct compat_xt_counters counters;
+	unsigned char elems[0];
+};
+
+static inline struct arpt_entry_target *
+compat_arpt_get_target(struct compat_arpt_entry *e)
+{
+	return (void *)e + e->target_offset;
+}
+
+#define COMPAT_ARPT_ALIGN(s)	COMPAT_XT_ALIGN(s)
+
+/* fn returns 0 to continue iteration */
+#define COMPAT_ARPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct compat_arpt_entry, entries, size, fn, ## args)
+
+#define COMPAT_ARPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \
+	XT_ENTRY_ITERATE_CONTINUE(struct compat_arpt_entry, entries, size, n, \
+				  fn, ## args)
+
+#endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _ARPTABLES_H */
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 029df76b9707..ad2da6db0e26 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -19,9 +19,10 @@
 #include <linux/proc_fs.h>
 #include <linux/module.h>
 #include <linux/init.h>
-
-#include <asm/uaccess.h>
 #include <linux/mutex.h>
+#include <linux/err.h>
+#include <net/compat.h>
+#include <asm/uaccess.h>
 
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_arp/arp_tables.h>
@@ -782,7 +783,73 @@ static int copy_entries_to_user(unsigned int total_size,
 	return ret;
 }
 
-static int get_info(void __user *user, int *len)
+#ifdef CONFIG_COMPAT
+static void compat_standard_from_user(void *dst, void *src)
+{
+	int v = *(compat_int_t *)src;
+
+	if (v > 0)
+		v += xt_compat_calc_jump(NF_ARP, v);
+	memcpy(dst, &v, sizeof(v));
+}
+
+static int compat_standard_to_user(void __user *dst, void *src)
+{
+	compat_int_t cv = *(int *)src;
+
+	if (cv > 0)
+		cv -= xt_compat_calc_jump(NF_ARP, cv);
+	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
+}
+
+static int compat_calc_entry(struct arpt_entry *e,
+			     const struct xt_table_info *info,
+			     void *base, struct xt_table_info *newinfo)
+{
+	struct arpt_entry_target *t;
+	unsigned int entry_offset;
+	int off, i, ret;
+
+	off = sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
+	entry_offset = (void *)e - base;
+
+	t = arpt_get_target(e);
+	off += xt_compat_target_offset(t->u.kernel.target);
+	newinfo->size -= off;
+	ret = xt_compat_add_offset(NF_ARP, entry_offset, off);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
+		if (info->hook_entry[i] &&
+		    (e < (struct arpt_entry *)(base + info->hook_entry[i])))
+			newinfo->hook_entry[i] -= off;
+		if (info->underflow[i] &&
+		    (e < (struct arpt_entry *)(base + info->underflow[i])))
+			newinfo->underflow[i] -= off;
+	}
+	return 0;
+}
+
+static int compat_table_info(const struct xt_table_info *info,
+			     struct xt_table_info *newinfo)
+{
+	void *loc_cpu_entry;
+
+	if (!newinfo || !info)
+		return -EINVAL;
+
+	/* we dont care about newinfo->entries[] */
+	memcpy(newinfo, info, offsetof(struct xt_table_info, entries));
+	newinfo->initial_entries = 0;
+	loc_cpu_entry = info->entries[raw_smp_processor_id()];
+	return ARPT_ENTRY_ITERATE(loc_cpu_entry, info->size,
+				  compat_calc_entry, info, loc_cpu_entry,
+				  newinfo);
+}
+#endif
+
+static int get_info(void __user *user, int *len, int compat)
 {
 	char name[ARPT_TABLE_MAXNAMELEN];
 	struct arpt_table *t;
@@ -798,13 +865,24 @@ static int get_info(void __user *user, int *len)
 		return -EFAULT;
 
 	name[ARPT_TABLE_MAXNAMELEN-1] = '\0';
-
+#ifdef CONFIG_COMPAT
+	if (compat)
+		xt_compat_lock(NF_ARP);
+#endif
 	t = try_then_request_module(xt_find_table_lock(NF_ARP, name),
 				    "arptable_%s", name);
 	if (t && !IS_ERR(t)) {
 		struct arpt_getinfo info;
 		struct xt_table_info *private = t->private;
 
+#ifdef CONFIG_COMPAT
+		if (compat) {
+			struct xt_table_info tmp;
+			ret = compat_table_info(private, &tmp);
+			xt_compat_flush_offsets(NF_ARP);
+			private = &tmp;
+		}
+#endif
 		info.valid_hooks = t->valid_hooks;
 		memcpy(info.hook_entry, private->hook_entry,
 		       sizeof(info.hook_entry));
@@ -822,6 +900,10 @@ static int get_info(void __user *user, int *len)
 		module_put(t->me);
 	} else
 		ret = t ? PTR_ERR(t) : -ENOENT;
+#ifdef CONFIG_COMPAT
+	if (compat)
+		xt_compat_unlock(NF_ARP);
+#endif
 	return ret;
 }
 
@@ -864,65 +946,41 @@ static int get_entries(struct arpt_get_entries __user *uptr, int *len)
 	return ret;
 }
 
-static int do_replace(void __user *user, unsigned int len)
+static int __do_replace(const char *name, unsigned int valid_hooks,
+			struct xt_table_info *newinfo,
+			unsigned int num_counters,
+			void __user *counters_ptr)
 {
 	int ret;
-	struct arpt_replace tmp;
 	struct arpt_table *t;
-	struct xt_table_info *newinfo, *oldinfo;
+	struct xt_table_info *oldinfo;
 	struct xt_counters *counters;
-	void *loc_cpu_entry, *loc_cpu_old_entry;
-
-	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
-		return -EFAULT;
-
-	/* overflow check */
-	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
-		return -ENOMEM;
-
-	newinfo = xt_alloc_table_info(tmp.size);
-	if (!newinfo)
-		return -ENOMEM;
+	void *loc_cpu_old_entry;
 
-	/* choose the copy that is on our node/cpu */
-	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
-	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
-			   tmp.size) != 0) {
-		ret = -EFAULT;
-		goto free_newinfo;
-	}
-
-	counters = vmalloc_node(tmp.num_counters * sizeof(struct xt_counters),
+	ret = 0;
+	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
 				numa_node_id());
 	if (!counters) {
 		ret = -ENOMEM;
-		goto free_newinfo;
+		goto out;
 	}
 
-	ret = translate_table(tmp.name, tmp.valid_hooks,
-			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
-			      tmp.hook_entry, tmp.underflow);
-	if (ret != 0)
-		goto free_newinfo_counters;
-
-	duprintf("arp_tables: Translated table\n");
-
-	t = try_then_request_module(xt_find_table_lock(NF_ARP, tmp.name),
-				    "arptable_%s", tmp.name);
+	t = try_then_request_module(xt_find_table_lock(NF_ARP, name),
+				    "arptable_%s", name);
 	if (!t || IS_ERR(t)) {
 		ret = t ? PTR_ERR(t) : -ENOENT;
 		goto free_newinfo_counters_untrans;
 	}
 
 	/* You lied! */
-	if (tmp.valid_hooks != t->valid_hooks) {
+	if (valid_hooks != t->valid_hooks) {
 		duprintf("Valid hook crap: %08X vs %08X\n",
-			 tmp.valid_hooks, t->valid_hooks);
+			 valid_hooks, t->valid_hooks);
 		ret = -EINVAL;
 		goto put_module;
 	}
 
-	oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret);
+	oldinfo = xt_replace_table(t, num_counters, newinfo, &ret);
 	if (!oldinfo)
 		goto put_module;
 
@@ -940,11 +998,12 @@ static int do_replace(void __user *user, unsigned int len)
 	get_counters(oldinfo, counters);
 	/* Decrease module usage counts and free resource */
 	loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()];
-	ARPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL);
+	ARPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,
+			   NULL);
 
 	xt_free_table_info(oldinfo);
-	if (copy_to_user(tmp.counters, counters,
-			 sizeof(struct xt_counters) * tmp.num_counters) != 0)
+	if (copy_to_user(counters_ptr, counters,
+			 sizeof(struct xt_counters) * num_counters) != 0)
 		ret = -EFAULT;
 	vfree(counters);
 	xt_table_unlock(t);
@@ -954,9 +1013,53 @@ static int do_replace(void __user *user, unsigned int len)
 	module_put(t->me);
 	xt_table_unlock(t);
  free_newinfo_counters_untrans:
-	ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
- free_newinfo_counters:
 	vfree(counters);
+ out:
+	return ret;
+}
+
+static int do_replace(void __user *user, unsigned int len)
+{
+	int ret;
+	struct arpt_replace tmp;
+	struct xt_table_info *newinfo;
+	void *loc_cpu_entry;
+
+	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+		return -EFAULT;
+
+	/* overflow check */
+	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
+		return -ENOMEM;
+
+	newinfo = xt_alloc_table_info(tmp.size);
+	if (!newinfo)
+		return -ENOMEM;
+
+	/* choose the copy that is on our node/cpu */
+	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
+	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp),
+			   tmp.size) != 0) {
+		ret = -EFAULT;
+		goto free_newinfo;
+	}
+
+	ret = translate_table(tmp.name, tmp.valid_hooks,
+			      newinfo, loc_cpu_entry, tmp.size, tmp.num_entries,
+			      tmp.hook_entry, tmp.underflow);
+	if (ret != 0)
+		goto free_newinfo;
+
+	duprintf("arp_tables: Translated table\n");
+
+	ret = __do_replace(tmp.name, tmp.valid_hooks, newinfo,
+			   tmp.num_counters, tmp.counters);
+	if (ret)
+		goto free_newinfo_untrans;
+	return 0;
+
+ free_newinfo_untrans:
+	ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
  free_newinfo:
 	xt_free_table_info(newinfo);
 	return ret;
@@ -976,31 +1079,59 @@ static inline int add_counter_to_entry(struct arpt_entry *e,
 	return 0;
 }
 
-static int do_add_counters(void __user *user, unsigned int len)
+static int do_add_counters(void __user *user, unsigned int len, int compat)
 {
 	unsigned int i;
-	struct xt_counters_info tmp, *paddc;
+	struct xt_counters_info tmp;
+	struct xt_counters *paddc;
+	unsigned int num_counters;
+	char *name;
+	int size;
+	void *ptmp;
 	struct arpt_table *t;
 	struct xt_table_info *private;
 	int ret = 0;
 	void *loc_cpu_entry;
+#ifdef CONFIG_COMPAT
+	struct compat_xt_counters_info compat_tmp;
 
-	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+	if (compat) {
+		ptmp = &compat_tmp;
+		size = sizeof(struct compat_xt_counters_info);
+	} else
+#endif
+	{
+		ptmp = &tmp;
+		size = sizeof(struct xt_counters_info);
+	}
+
+	if (copy_from_user(ptmp, user, size) != 0)
 		return -EFAULT;
 
-	if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters))
+#ifdef CONFIG_COMPAT
+	if (compat) {
+		num_counters = compat_tmp.num_counters;
+		name = compat_tmp.name;
+	} else
+#endif
+	{
+		num_counters = tmp.num_counters;
+		name = tmp.name;
+	}
+
+	if (len != size + num_counters * sizeof(struct xt_counters))
 		return -EINVAL;
 
-	paddc = vmalloc_node(len, numa_node_id());
+	paddc = vmalloc_node(len - size, numa_node_id());
 	if (!paddc)
 		return -ENOMEM;
 
-	if (copy_from_user(paddc, user, len) != 0) {
+	if (copy_from_user(paddc, user + size, len - size) != 0) {
 		ret = -EFAULT;
 		goto free;
 	}
 
-	t = xt_find_table_lock(NF_ARP, tmp.name);
+	t = xt_find_table_lock(NF_ARP, name);
 	if (!t || IS_ERR(t)) {
 		ret = t ? PTR_ERR(t) : -ENOENT;
 		goto free;
@@ -1008,7 +1139,7 @@ static int do_add_counters(void __user *user, unsigned int len)
 
 	write_lock_bh(&t->lock);
 	private = t->private;
-	if (private->number != tmp.num_counters) {
+	if (private->number != num_counters) {
 		ret = -EINVAL;
 		goto unlock_up_free;
 	}
@@ -1019,7 +1150,7 @@ static int do_add_counters(void __user *user, unsigned int len)
 	ARPT_ENTRY_ITERATE(loc_cpu_entry,
 			   private->size,
 			   add_counter_to_entry,
-			   paddc->counters,
+			   paddc,
 			   &i);
  unlock_up_free:
 	write_unlock_bh(&t->lock);
@@ -1031,6 +1162,496 @@ static int do_add_counters(void __user *user, unsigned int len)
 	return ret;
 }
 
+#ifdef CONFIG_COMPAT
+static inline int
+compat_release_entry(struct compat_arpt_entry *e, unsigned int *i)
+{
+	struct arpt_entry_target *t;
+
+	if (i && (*i)-- == 0)
+		return 1;
+
+	t = compat_arpt_get_target(e);
+	module_put(t->u.kernel.target->me);
+	return 0;
+}
+
+static inline int
+check_compat_entry_size_and_hooks(struct compat_arpt_entry *e,
+				  struct xt_table_info *newinfo,
+				  unsigned int *size,
+				  unsigned char *base,
+				  unsigned char *limit,
+				  unsigned int *hook_entries,
+				  unsigned int *underflows,
+				  unsigned int *i,
+				  const char *name)
+{
+	struct arpt_entry_target *t;
+	struct xt_target *target;
+	unsigned int entry_offset;
+	int ret, off, h;
+
+	duprintf("check_compat_entry_size_and_hooks %p\n", e);
+	if ((unsigned long)e % __alignof__(struct compat_arpt_entry) != 0
+	    || (unsigned char *)e + sizeof(struct compat_arpt_entry) >= limit) {
+		duprintf("Bad offset %p, limit = %p\n", e, limit);
+		return -EINVAL;
+	}
+
+	if (e->next_offset < sizeof(struct compat_arpt_entry) +
+			     sizeof(struct compat_xt_entry_target)) {
+		duprintf("checking: element %p size %u\n",
+			 e, e->next_offset);
+		return -EINVAL;
+	}
+
+	/* For purposes of check_entry casting the compat entry is fine */
+	ret = check_entry((struct arpt_entry *)e, name);
+	if (ret)
+		return ret;
+
+	off = sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
+	entry_offset = (void *)e - (void *)base;
+
+	t = compat_arpt_get_target(e);
+	target = try_then_request_module(xt_find_target(NF_ARP,
+							t->u.user.name,
+							t->u.user.revision),
+					 "arpt_%s", t->u.user.name);
+	if (IS_ERR(target) || !target) {
+		duprintf("check_compat_entry_size_and_hooks: `%s' not found\n",
+			 t->u.user.name);
+		ret = target ? PTR_ERR(target) : -ENOENT;
+		goto out;
+	}
+	t->u.kernel.target = target;
+
+	off += xt_compat_target_offset(target);
+	*size += off;
+	ret = xt_compat_add_offset(NF_ARP, entry_offset, off);
+	if (ret)
+		goto release_target;
+
+	/* Check hooks & underflows */
+	for (h = 0; h < NF_ARP_NUMHOOKS; h++) {
+		if ((unsigned char *)e - base == hook_entries[h])
+			newinfo->hook_entry[h] = hook_entries[h];
+		if ((unsigned char *)e - base == underflows[h])
+			newinfo->underflow[h] = underflows[h];
+	}
+
+	/* Clear counters and comefrom */
+	memset(&e->counters, 0, sizeof(e->counters));
+	e->comefrom = 0;
+
+	(*i)++;
+	return 0;
+
+release_target:
+	module_put(t->u.kernel.target->me);
+out:
+	return ret;
+}
+
+static int
+compat_copy_entry_from_user(struct compat_arpt_entry *e, void **dstptr,
+			    unsigned int *size, const char *name,
+			    struct xt_table_info *newinfo, unsigned char *base)
+{
+	struct arpt_entry_target *t;
+	struct xt_target *target;
+	struct arpt_entry *de;
+	unsigned int origsize;
+	int ret, h;
+
+	ret = 0;
+	origsize = *size;
+	de = (struct arpt_entry *)*dstptr;
+	memcpy(de, e, sizeof(struct arpt_entry));
+	memcpy(&de->counters, &e->counters, sizeof(e->counters));
+
+	*dstptr += sizeof(struct arpt_entry);
+	*size += sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
+
+	de->target_offset = e->target_offset - (origsize - *size);
+	t = compat_arpt_get_target(e);
+	target = t->u.kernel.target;
+	xt_compat_target_from_user(t, dstptr, size);
+
+	de->next_offset = e->next_offset - (origsize - *size);
+	for (h = 0; h < NF_ARP_NUMHOOKS; h++) {
+		if ((unsigned char *)de - base < newinfo->hook_entry[h])
+			newinfo->hook_entry[h] -= origsize - *size;
+		if ((unsigned char *)de - base < newinfo->underflow[h])
+			newinfo->underflow[h] -= origsize - *size;
+	}
+	return ret;
+}
+
+static inline int compat_check_entry(struct arpt_entry *e, const char *name,
+				     unsigned int *i)
+{
+	int ret;
+
+	ret = check_target(e, name);
+	if (ret)
+		return ret;
+
+	(*i)++;
+	return 0;
+}
+
+static int translate_compat_table(const char *name,
+				  unsigned int valid_hooks,
+				  struct xt_table_info **pinfo,
+				  void **pentry0,
+				  unsigned int total_size,
+				  unsigned int number,
+				  unsigned int *hook_entries,
+				  unsigned int *underflows)
+{
+	unsigned int i, j;
+	struct xt_table_info *newinfo, *info;
+	void *pos, *entry0, *entry1;
+	unsigned int size;
+	int ret;
+
+	info = *pinfo;
+	entry0 = *pentry0;
+	size = total_size;
+	info->number = number;
+
+	/* Init all hooks to impossible value. */
+	for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
+		info->hook_entry[i] = 0xFFFFFFFF;
+		info->underflow[i] = 0xFFFFFFFF;
+	}
+
+	duprintf("translate_compat_table: size %u\n", info->size);
+	j = 0;
+	xt_compat_lock(NF_ARP);
+	/* Walk through entries, checking offsets. */
+	ret = COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size,
+					check_compat_entry_size_and_hooks,
+					info, &size, entry0,
+					entry0 + total_size,
+					hook_entries, underflows, &j, name);
+	if (ret != 0)
+		goto out_unlock;
+
+	ret = -EINVAL;
+	if (j != number) {
+		duprintf("translate_compat_table: %u not %u entries\n",
+			 j, number);
+		goto out_unlock;
+	}
+
+	/* Check hooks all assigned */
+	for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
+		/* Only hooks which are valid */
+		if (!(valid_hooks & (1 << i)))
+			continue;
+		if (info->hook_entry[i] == 0xFFFFFFFF) {
+			duprintf("Invalid hook entry %u %u\n",
+				 i, hook_entries[i]);
+			goto out_unlock;
+		}
+		if (info->underflow[i] == 0xFFFFFFFF) {
+			duprintf("Invalid underflow %u %u\n",
+				 i, underflows[i]);
+			goto out_unlock;
+		}
+	}
+
+	ret = -ENOMEM;
+	newinfo = xt_alloc_table_info(size);
+	if (!newinfo)
+		goto out_unlock;
+
+	newinfo->number = number;
+	for (i = 0; i < NF_ARP_NUMHOOKS; i++) {
+		newinfo->hook_entry[i] = info->hook_entry[i];
+		newinfo->underflow[i] = info->underflow[i];
+	}
+	entry1 = newinfo->entries[raw_smp_processor_id()];
+	pos = entry1;
+	size = total_size;
+	ret = COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size,
+					compat_copy_entry_from_user,
+					&pos, &size, name, newinfo, entry1);
+	xt_compat_flush_offsets(NF_ARP);
+	xt_compat_unlock(NF_ARP);
+	if (ret)
+		goto free_newinfo;
+
+	ret = -ELOOP;
+	if (!mark_source_chains(newinfo, valid_hooks, entry1))
+		goto free_newinfo;
+
+	i = 0;
+	ret = ARPT_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry,
+				 name, &i);
+	if (ret) {
+		j -= i;
+		COMPAT_ARPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i,
+						   compat_release_entry, &j);
+		ARPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i);
+		xt_free_table_info(newinfo);
+		return ret;
+	}
+
+	/* And one copy for every other CPU */
+	for_each_possible_cpu(i)
+		if (newinfo->entries[i] && newinfo->entries[i] != entry1)
+			memcpy(newinfo->entries[i], entry1, newinfo->size);
+
+	*pinfo = newinfo;
+	*pentry0 = entry1;
+	xt_free_table_info(info);
+	return 0;
+
+free_newinfo:
+	xt_free_table_info(newinfo);
+out:
+	COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j);
+	return ret;
+out_unlock:
+	xt_compat_flush_offsets(NF_ARP);
+	xt_compat_unlock(NF_ARP);
+	goto out;
+}
+
+struct compat_arpt_replace {
+	char				name[ARPT_TABLE_MAXNAMELEN];
+	u32				valid_hooks;
+	u32				num_entries;
+	u32				size;
+	u32				hook_entry[NF_ARP_NUMHOOKS];
+	u32				underflow[NF_ARP_NUMHOOKS];
+	u32				num_counters;
+	compat_uptr_t			counters;
+	struct compat_arpt_entry	entries[0];
+};
+
+static int compat_do_replace(void __user *user, unsigned int len)
+{
+	int ret;
+	struct compat_arpt_replace tmp;
+	struct xt_table_info *newinfo;
+	void *loc_cpu_entry;
+
+	if (copy_from_user(&tmp, user, sizeof(tmp)) != 0)
+		return -EFAULT;
+
+	/* overflow check */
+	if (tmp.size >= INT_MAX / num_possible_cpus())
+		return -ENOMEM;
+	if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters))
+		return -ENOMEM;
+
+	newinfo = xt_alloc_table_info(tmp.size);
+	if (!newinfo)
+		return -ENOMEM;
+
+	/* choose the copy that is on our node/cpu */
+	loc_cpu_entry = newinfo->entries[raw_smp_processor_id()];
+	if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) {
+		ret = -EFAULT;
+		goto free_newinfo;
+	}
+
+	ret = translate_compat_table(tmp.name, tmp.valid_hooks,
+				     &newinfo, &loc_cpu_entry, tmp.size,
+				     tmp.num_entries, tmp.hook_entry,
+				     tmp.underflow);
+	if (ret != 0)
+		goto free_newinfo;
+
+	duprintf("compat_do_replace: Translated table\n");
+
+	ret = __do_replace(tmp.name, tmp.valid_hooks, newinfo,
+			   tmp.num_counters, compat_ptr(tmp.counters));
+	if (ret)
+		goto free_newinfo_untrans;
+	return 0;
+
+ free_newinfo_untrans:
+	ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL);
+ free_newinfo:
+	xt_free_table_info(newinfo);
+	return ret;
+}
+
+static int compat_do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user,
+				  unsigned int len)
+{
+	int ret;
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	switch (cmd) {
+	case ARPT_SO_SET_REPLACE:
+		ret = compat_do_replace(user, len);
+		break;
+
+	case ARPT_SO_SET_ADD_COUNTERS:
+		ret = do_add_counters(user, len, 1);
+		break;
+
+	default:
+		duprintf("do_arpt_set_ctl:  unknown request %i\n", cmd);
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static int compat_copy_entry_to_user(struct arpt_entry *e, void __user **dstptr,
+				     compat_uint_t *size,
+				     struct xt_counters *counters,
+				     unsigned int *i)
+{
+	struct arpt_entry_target *t;
+	struct compat_arpt_entry __user *ce;
+	u_int16_t target_offset, next_offset;
+	compat_uint_t origsize;
+	int ret;
+
+	ret = -EFAULT;
+	origsize = *size;
+	ce = (struct compat_arpt_entry __user *)*dstptr;
+	if (copy_to_user(ce, e, sizeof(struct arpt_entry)))
+		goto out;
+
+	if (copy_to_user(&ce->counters, &counters[*i], sizeof(counters[*i])))
+		goto out;
+
+	*dstptr += sizeof(struct compat_arpt_entry);
+	*size -= sizeof(struct arpt_entry) - sizeof(struct compat_arpt_entry);
+
+	target_offset = e->target_offset - (origsize - *size);
+
+	t = arpt_get_target(e);
+	ret = xt_compat_target_to_user(t, dstptr, size);
+	if (ret)
+		goto out;
+	ret = -EFAULT;
+	next_offset = e->next_offset - (origsize - *size);
+	if (put_user(target_offset, &ce->target_offset))
+		goto out;
+	if (put_user(next_offset, &ce->next_offset))
+		goto out;
+
+	(*i)++;
+	return 0;
+out:
+	return ret;
+}
+
+static int compat_copy_entries_to_user(unsigned int total_size,
+				       struct arpt_table *table,
+				       void __user *userptr)
+{
+	struct xt_counters *counters;
+	struct xt_table_info *private = table->private;
+	void __user *pos;
+	unsigned int size;
+	int ret = 0;
+	void *loc_cpu_entry;
+	unsigned int i = 0;
+
+	counters = alloc_counters(table);
+	if (IS_ERR(counters))
+		return PTR_ERR(counters);
+
+	/* choose the copy on our node/cpu */
+	loc_cpu_entry = private->entries[raw_smp_processor_id()];
+	pos = userptr;
+	size = total_size;
+	ret = ARPT_ENTRY_ITERATE(loc_cpu_entry, total_size,
+				 compat_copy_entry_to_user,
+				 &pos, &size, counters, &i);
+	vfree(counters);
+	return ret;
+}
+
+struct compat_arpt_get_entries {
+	char name[ARPT_TABLE_MAXNAMELEN];
+	compat_uint_t size;
+	struct compat_arpt_entry entrytable[0];
+};
+
+static int compat_get_entries(struct compat_arpt_get_entries __user *uptr,
+			      int *len)
+{
+	int ret;
+	struct compat_arpt_get_entries get;
+	struct arpt_table *t;
+
+	if (*len < sizeof(get)) {
+		duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get));
+		return -EINVAL;
+	}
+	if (copy_from_user(&get, uptr, sizeof(get)) != 0)
+		return -EFAULT;
+	if (*len != sizeof(struct compat_arpt_get_entries) + get.size) {
+		duprintf("compat_get_entries: %u != %zu\n",
+			 *len, sizeof(get) + get.size);
+		return -EINVAL;
+	}
+
+	xt_compat_lock(NF_ARP);
+	t = xt_find_table_lock(NF_ARP, get.name);
+	if (t && !IS_ERR(t)) {
+		struct xt_table_info *private = t->private;
+		struct xt_table_info info;
+
+		duprintf("t->private->number = %u\n", private->number);
+		ret = compat_table_info(private, &info);
+		if (!ret && get.size == info.size) {
+			ret = compat_copy_entries_to_user(private->size,
+							  t, uptr->entrytable);
+		} else if (!ret) {
+			duprintf("compat_get_entries: I've got %u not %u!\n",
+				 private->size, get.size);
+			ret = -EINVAL;
+		}
+		xt_compat_flush_offsets(NF_ARP);
+		module_put(t->me);
+		xt_table_unlock(t);
+	} else
+		ret = t ? PTR_ERR(t) : -ENOENT;
+
+	xt_compat_unlock(NF_ARP);
+	return ret;
+}
+
+static int do_arpt_get_ctl(struct sock *, int, void __user *, int *);
+
+static int compat_do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user,
+				  int *len)
+{
+	int ret;
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	switch (cmd) {
+	case ARPT_SO_GET_INFO:
+		ret = get_info(user, len, 1);
+		break;
+	case ARPT_SO_GET_ENTRIES:
+		ret = compat_get_entries(user, len);
+		break;
+	default:
+		ret = do_arpt_get_ctl(sk, cmd, user, len);
+	}
+	return ret;
+}
+#endif
+
 static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
 {
 	int ret;
@@ -1044,7 +1665,7 @@ static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned
 		break;
 
 	case ARPT_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(user, len);
+		ret = do_add_counters(user, len, 0);
 		break;
 
 	default:
@@ -1064,7 +1685,7 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len
 
 	switch (cmd) {
 	case ARPT_SO_GET_INFO:
-		ret = get_info(user, len);
+		ret = get_info(user, len, 0);
 		break;
 
 	case ARPT_SO_GET_ENTRIES:
@@ -1156,6 +1777,11 @@ static struct arpt_target arpt_standard_target __read_mostly = {
 	.name		= ARPT_STANDARD_TARGET,
 	.targetsize	= sizeof(int),
 	.family		= NF_ARP,
+#ifdef CONFIG_COMPAT
+	.compatsize	= sizeof(compat_int_t),
+	.compat_from_user = compat_standard_from_user,
+	.compat_to_user	= compat_standard_to_user,
+#endif
 };
 
 static struct arpt_target arpt_error_target __read_mostly = {
@@ -1170,9 +1796,15 @@ static struct nf_sockopt_ops arpt_sockopts = {
 	.set_optmin	= ARPT_BASE_CTL,
 	.set_optmax	= ARPT_SO_SET_MAX+1,
 	.set		= do_arpt_set_ctl,
+#ifdef CONFIG_COMPAT
+	.compat_set	= compat_do_arpt_set_ctl,
+#endif
 	.get_optmin	= ARPT_BASE_CTL,
 	.get_optmax	= ARPT_SO_GET_MAX+1,
 	.get		= do_arpt_get_ctl,
+#ifdef CONFIG_COMPAT
+	.compat_get	= compat_do_arpt_get_ctl,
+#endif
 	.owner		= THIS_MODULE,
 };
 
-- 
cgit v1.2.3


From 13eae15a244bb29beaa47bf86a24fd29ca7f8a4c Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 17 Dec 2007 22:28:00 -0800
Subject: [NETFILTER]: ctnetlink: add support for NAT sequence adjustments

The combination of NAT and helpers may produce TCP sequence adjustments.
In failover setups, this information needs to be replicated in order to
achieve a successful recovery of mangled, related connections. This patch is
particularly useful for conntrackd, see:

http://people.netfilter.org/pablo/conntrack-tools/

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_common.h |   4 +
 include/linux/netfilter/nfnetlink_conntrack.h |  10 +++
 net/ipv4/netfilter/nf_nat_helper.c            |   3 +
 net/netfilter/nf_conntrack_netlink.c          | 124 +++++++++++++++++++++++++-
 4 files changed, 140 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index 9e0dae07861e..19747e8f71cf 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -129,6 +129,10 @@ enum ip_conntrack_events
 	/* Mark is set */
 	IPCT_MARK_BIT = 12,
 	IPCT_MARK = (1 << IPCT_MARK_BIT),
+
+	/* NAT sequence adjustment */
+	IPCT_NATSEQADJ_BIT = 13,
+	IPCT_NATSEQADJ = (1 << IPCT_NATSEQADJ_BIT),
 };
 
 enum ip_conntrack_expect_events {
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 4affa3fe78e0..c19d976b1b75 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -37,6 +37,8 @@ enum ctattr_type {
 	CTA_ID,
 	CTA_NAT_DST,
 	CTA_TUPLE_MASTER,
+	CTA_NAT_SEQ_ADJ_ORIG,
+	CTA_NAT_SEQ_ADJ_REPLY,
 	__CTA_MAX
 };
 #define CTA_MAX (__CTA_MAX - 1)
@@ -119,6 +121,14 @@ enum ctattr_protonat {
 };
 #define CTA_PROTONAT_MAX (__CTA_PROTONAT_MAX - 1)
 
+enum ctattr_natseq {
+	CTA_NAT_SEQ_CORRECTION_POS,
+	CTA_NAT_SEQ_OFFSET_BEFORE,
+	CTA_NAT_SEQ_OFFSET_AFTER,
+	__CTA_NAT_SEQ_MAX
+};
+#define CTA_NAT_SEQ_MAX (__CTA_NAT_SEQ_MAX - 1)
+
 enum ctattr_expect {
 	CTA_EXPECT_UNSPEC,
 	CTA_EXPECT_MASTER,
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index 53f79a310b43..d24f3d947394 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -20,6 +20,7 @@
 #include <linux/netfilter_ipv4.h>
 #include <net/netfilter/nf_conntrack.h>
 #include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_ecache.h>
 #include <net/netfilter/nf_conntrack_expect.h>
 #include <net/netfilter/nf_nat.h>
 #include <net/netfilter/nf_nat_protocol.h>
@@ -191,6 +192,8 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb,
 		/* Tell TCP window tracking about seq change */
 		nf_conntrack_tcp_update(skb, ip_hdrlen(skb),
 					ct, CTINFO2DIR(ctinfo));
+
+		nf_conntrack_event_cache(IPCT_NATSEQADJ, skb);
 	}
 	return 1;
 }
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index a15971e9923b..d7da167ef1b8 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -254,6 +254,55 @@ nla_put_failure:
 #define ctnetlink_dump_mark(a, b) (0)
 #endif
 
+#ifdef CONFIG_NF_NAT_NEEDED
+static inline int
+dump_nat_seq_adj(struct sk_buff *skb, const struct nf_nat_seq *natseq, int type)
+{
+	__be32 tmp;
+	struct nlattr *nest_parms;
+
+	nest_parms = nla_nest_start(skb, type | NLA_F_NESTED);
+	if (!nest_parms)
+		goto nla_put_failure;
+
+	tmp = htonl(natseq->correction_pos);
+	NLA_PUT(skb, CTA_NAT_SEQ_CORRECTION_POS, sizeof(tmp), &tmp);
+	tmp = htonl(natseq->offset_before);
+	NLA_PUT(skb, CTA_NAT_SEQ_OFFSET_BEFORE, sizeof(tmp), &tmp);
+	tmp = htonl(natseq->offset_after);
+	NLA_PUT(skb, CTA_NAT_SEQ_OFFSET_AFTER, sizeof(tmp), &tmp);
+
+	nla_nest_end(skb, nest_parms);
+
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+
+static inline int
+ctnetlink_dump_nat_seq_adj(struct sk_buff *skb, const struct nf_conn *ct)
+{
+	struct nf_nat_seq *natseq;
+	struct nf_conn_nat *nat = nfct_nat(ct);
+
+	if (!(ct->status & IPS_SEQ_ADJUST) || !nat)
+		return 0;
+
+	natseq = &nat->seq[IP_CT_DIR_ORIGINAL];
+	if (dump_nat_seq_adj(skb, natseq, CTA_NAT_SEQ_ADJ_ORIG) == -1)
+		return -1;
+
+	natseq = &nat->seq[IP_CT_DIR_REPLY];
+	if (dump_nat_seq_adj(skb, natseq, CTA_NAT_SEQ_ADJ_REPLY) == -1)
+		return -1;
+
+	return 0;
+}
+#else
+#define ctnetlink_dump_nat_seq_adj(a, b) (0)
+#endif
+
 static inline int
 ctnetlink_dump_id(struct sk_buff *skb, const struct nf_conn *ct)
 {
@@ -321,7 +370,8 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
 	    ctnetlink_dump_helpinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_mark(skb, ct) < 0 ||
 	    ctnetlink_dump_id(skb, ct) < 0 ||
-	    ctnetlink_dump_use(skb, ct) < 0)
+	    ctnetlink_dump_use(skb, ct) < 0 ||
+	    ctnetlink_dump_nat_seq_adj(skb, ct) < 0)
 		goto nla_put_failure;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
@@ -424,6 +474,10 @@ static int ctnetlink_conntrack_event(struct notifier_block *this,
 		    (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 ||
 		     ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0))
 			goto nla_put_failure;
+
+		if (events & IPCT_NATSEQADJ &&
+		    ctnetlink_dump_nat_seq_adj(skb, ct) < 0)
+			goto nla_put_failure;
 	}
 
 	nlh->nlmsg_len = skb->tail - b;
@@ -935,6 +989,66 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, struct nlattr *cda[])
 	return err;
 }
 
+#ifdef CONFIG_NF_NAT_NEEDED
+static inline int
+change_nat_seq_adj(struct nf_nat_seq *natseq, struct nlattr *attr)
+{
+	struct nlattr *cda[CTA_NAT_SEQ_MAX+1];
+
+	nla_parse_nested(cda, CTA_NAT_SEQ_MAX, attr, NULL);
+
+	if (!cda[CTA_NAT_SEQ_CORRECTION_POS])
+		return -EINVAL;
+
+	natseq->correction_pos =
+		ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_CORRECTION_POS]));
+
+	if (!cda[CTA_NAT_SEQ_OFFSET_BEFORE])
+		return -EINVAL;
+
+	natseq->offset_before =
+		ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_OFFSET_BEFORE]));
+
+	if (!cda[CTA_NAT_SEQ_OFFSET_AFTER])
+		return -EINVAL;
+
+	natseq->offset_after =
+		ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_OFFSET_AFTER]));
+
+	return 0;
+}
+
+static int
+ctnetlink_change_nat_seq_adj(struct nf_conn *ct, struct nlattr *cda[])
+{
+	int ret = 0;
+	struct nf_conn_nat *nat = nfct_nat(ct);
+
+	if (!nat)
+		return 0;
+
+	if (cda[CTA_NAT_SEQ_ADJ_ORIG]) {
+		ret = change_nat_seq_adj(&nat->seq[IP_CT_DIR_ORIGINAL],
+					 cda[CTA_NAT_SEQ_ADJ_ORIG]);
+		if (ret < 0)
+			return ret;
+
+		ct->status |= IPS_SEQ_ADJUST;
+	}
+
+	if (cda[CTA_NAT_SEQ_ADJ_REPLY]) {
+		ret = change_nat_seq_adj(&nat->seq[IP_CT_DIR_REPLY],
+					 cda[CTA_NAT_SEQ_ADJ_REPLY]);
+		if (ret < 0)
+			return ret;
+
+		ct->status |= IPS_SEQ_ADJUST;
+	}
+
+	return 0;
+}
+#endif
+
 static int
 ctnetlink_change_conntrack(struct nf_conn *ct, struct nlattr *cda[])
 {
@@ -969,6 +1083,14 @@ ctnetlink_change_conntrack(struct nf_conn *ct, struct nlattr *cda[])
 		ct->mark = ntohl(*(__be32 *)nla_data(cda[CTA_MARK]));
 #endif
 
+#ifdef CONFIG_NF_NAT_NEEDED
+	if (cda[CTA_NAT_SEQ_ADJ_ORIG] || cda[CTA_NAT_SEQ_ADJ_REPLY]) {
+		err = ctnetlink_change_nat_seq_adj(ct, cda);
+		if (err < 0)
+			return err;
+	}
+#endif
+
 	return 0;
 }
 
-- 
cgit v1.2.3


From 37fccd8577d38e249dde71512fb38d2f6a4d9d3c Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 17 Dec 2007 22:28:41 -0800
Subject: [NETFILTER]: ctnetlink: add support for secmark

This patch adds support for James Morris' connsecmark.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_common.h |  4 ++++
 include/linux/netfilter/nfnetlink_conntrack.h |  1 +
 net/netfilter/nf_conntrack_netlink.c          | 22 ++++++++++++++++++++++
 net/netfilter/xt_CONNSECMARK.c                |  5 ++++-
 4 files changed, 31 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index 19747e8f71cf..bad1eb760f61 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -133,6 +133,10 @@ enum ip_conntrack_events
 	/* NAT sequence adjustment */
 	IPCT_NATSEQADJ_BIT = 13,
 	IPCT_NATSEQADJ = (1 << IPCT_NATSEQADJ_BIT),
+
+	/* Secmark is set */
+	IPCT_SECMARK_BIT = 14,
+	IPCT_SECMARK = (1 << IPCT_SECMARK_BIT),
 };
 
 enum ip_conntrack_expect_events {
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index c19d976b1b75..e3e1533aba2d 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -39,6 +39,7 @@ enum ctattr_type {
 	CTA_TUPLE_MASTER,
 	CTA_NAT_SEQ_ADJ_ORIG,
 	CTA_NAT_SEQ_ADJ_REPLY,
+	CTA_SECMARK,
 	__CTA_MAX
 };
 #define CTA_MAX (__CTA_MAX - 1)
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 94027c84be52..d4eedc68cc76 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -254,6 +254,22 @@ nla_put_failure:
 #define ctnetlink_dump_mark(a, b) (0)
 #endif
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static inline int
+ctnetlink_dump_secmark(struct sk_buff *skb, const struct nf_conn *ct)
+{
+	__be32 mark = htonl(ct->secmark);
+
+	NLA_PUT(skb, CTA_SECMARK, sizeof(u_int32_t), &mark);
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+#else
+#define ctnetlink_dump_secmark(a, b) (0)
+#endif
+
 #define master_tuple(ct) &(ct->master->tuplehash[IP_CT_DIR_ORIGINAL].tuple)
 
 static inline int
@@ -392,6 +408,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
 	    ctnetlink_dump_protoinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_helpinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_mark(skb, ct) < 0 ||
+	    ctnetlink_dump_secmark(skb, ct) < 0 ||
 	    ctnetlink_dump_id(skb, ct) < 0 ||
 	    ctnetlink_dump_use(skb, ct) < 0 ||
 	    ctnetlink_dump_master(skb, ct) < 0 ||
@@ -493,6 +510,11 @@ static int ctnetlink_conntrack_event(struct notifier_block *this,
 		    && ctnetlink_dump_mark(skb, ct) < 0)
 			goto nla_put_failure;
 #endif
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+		if ((events & IPCT_SECMARK || ct->secmark)
+		    && ctnetlink_dump_secmark(skb, ct) < 0)
+			goto nla_put_failure;
+#endif
 
 		if (events & IPCT_COUNTER_FILLING &&
 		    (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 ||
diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c
index 2c265e87f396..2333f7e29bc9 100644
--- a/net/netfilter/xt_CONNSECMARK.c
+++ b/net/netfilter/xt_CONNSECMARK.c
@@ -20,6 +20,7 @@
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_CONNSECMARK.h>
 #include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_ecache.h>
 
 #define PFX "CONNSECMARK: "
 
@@ -40,8 +41,10 @@ static void secmark_save(const struct sk_buff *skb)
 		enum ip_conntrack_info ctinfo;
 
 		ct = nf_ct_get(skb, &ctinfo);
-		if (ct && !ct->secmark)
+		if (ct && !ct->secmark) {
 			ct->secmark = skb->secmark;
+			nf_conntrack_event_cache(IPCT_SECMARK, skb);
+		}
 	}
 }
 
-- 
cgit v1.2.3


From 838965ba22066c7fcdbacfc543c387d0eb76c14c Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:29:26 -0800
Subject: [NETLINK]: Add NLA_PUT_BE16/nla_get_be16()

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index db4b935b6c7e..a52d03601754 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -881,6 +881,9 @@ static inline int nla_put_msecs(struct sk_buff *skb, int attrtype,
 #define NLA_PUT_LE16(skb, attrtype, value) \
 	NLA_PUT_TYPE(skb, __le16, attrtype, value)
 
+#define NLA_PUT_BE16(skb, attrtype, value) \
+	NLA_PUT_TYPE(skb, __be16, attrtype, value)
+
 #define NLA_PUT_U32(skb, attrtype, value) \
 	NLA_PUT_TYPE(skb, u32, attrtype, value)
 
@@ -926,6 +929,15 @@ static inline u16 nla_get_u16(struct nlattr *nla)
 	return *(u16 *) nla_data(nla);
 }
 
+/**
+ * nla_get_be16 - return payload of __be16 attribute
+ * @nla: __be16 netlink attribute
+ */
+static inline __be16 nla_get_be16(struct nlattr *nla)
+{
+	return *(__be16 *) nla_data(nla);
+}
+
 /**
  * nla_get_le16 - return payload of __le16 attribute
  * @nla: __le16 netlink attribute
-- 
cgit v1.2.3


From 2b628a0866860d44652362aafe403e5b5895583d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:37:36 -0800
Subject: [NETFILTER]: nf_nat: mark NAT protocols const

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_nat_protocol.h   | 18 +++++++++---------
 net/ipv4/netfilter/nf_nat_core.c          | 20 ++++++++++----------
 net/ipv4/netfilter/nf_nat_proto_gre.c     |  2 +-
 net/ipv4/netfilter/nf_nat_proto_icmp.c    |  2 +-
 net/ipv4/netfilter/nf_nat_proto_tcp.c     |  2 +-
 net/ipv4/netfilter/nf_nat_proto_udp.c     |  2 +-
 net/ipv4/netfilter/nf_nat_proto_unknown.c |  2 +-
 net/netfilter/nf_conntrack_netlink.c      |  2 +-
 8 files changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h
index 04578bfe23e1..4aa0edbb5b96 100644
--- a/include/net/netfilter/nf_nat_protocol.h
+++ b/include/net/netfilter/nf_nat_protocol.h
@@ -46,21 +46,21 @@ struct nf_nat_protocol
 };
 
 /* Protocol registration. */
-extern int nf_nat_protocol_register(struct nf_nat_protocol *proto);
-extern void nf_nat_protocol_unregister(struct nf_nat_protocol *proto);
+extern int nf_nat_protocol_register(const struct nf_nat_protocol *proto);
+extern void nf_nat_protocol_unregister(const struct nf_nat_protocol *proto);
 
-extern struct nf_nat_protocol *nf_nat_proto_find_get(u_int8_t protocol);
-extern void nf_nat_proto_put(struct nf_nat_protocol *proto);
+extern const struct nf_nat_protocol *nf_nat_proto_find_get(u_int8_t protocol);
+extern void nf_nat_proto_put(const struct nf_nat_protocol *proto);
 
 /* Built-in protocols. */
-extern struct nf_nat_protocol nf_nat_protocol_tcp;
-extern struct nf_nat_protocol nf_nat_protocol_udp;
-extern struct nf_nat_protocol nf_nat_protocol_icmp;
-extern struct nf_nat_protocol nf_nat_unknown_protocol;
+extern const struct nf_nat_protocol nf_nat_protocol_tcp;
+extern const struct nf_nat_protocol nf_nat_protocol_udp;
+extern const struct nf_nat_protocol nf_nat_protocol_icmp;
+extern const struct nf_nat_protocol nf_nat_unknown_protocol;
 
 extern int init_protocols(void) __init;
 extern void cleanup_protocols(void);
-extern struct nf_nat_protocol *find_nat_proto(u_int16_t protonum);
+extern const struct nf_nat_protocol *find_nat_proto(u_int16_t protonum);
 
 extern int nf_nat_port_range_to_nlattr(struct sk_buff *skb,
 				       const struct nf_nat_range *range);
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 4ee67e9f42ba..a772445228ac 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -42,18 +42,18 @@ static int nf_nat_vmalloced;
 static struct hlist_head *bysource;
 
 #define MAX_IP_NAT_PROTO 256
-static struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO];
+static const struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO];
 
-static inline struct nf_nat_protocol *
+static inline const struct nf_nat_protocol *
 __nf_nat_proto_find(u_int8_t protonum)
 {
 	return rcu_dereference(nf_nat_protos[protonum]);
 }
 
-struct nf_nat_protocol *
+const struct nf_nat_protocol *
 nf_nat_proto_find_get(u_int8_t protonum)
 {
-	struct nf_nat_protocol *p;
+	const struct nf_nat_protocol *p;
 
 	rcu_read_lock();
 	p = __nf_nat_proto_find(protonum);
@@ -66,7 +66,7 @@ nf_nat_proto_find_get(u_int8_t protonum)
 EXPORT_SYMBOL_GPL(nf_nat_proto_find_get);
 
 void
-nf_nat_proto_put(struct nf_nat_protocol *p)
+nf_nat_proto_put(const struct nf_nat_protocol *p)
 {
 	module_put(p->me);
 }
@@ -105,7 +105,7 @@ static int
 in_range(const struct nf_conntrack_tuple *tuple,
 	 const struct nf_nat_range *range)
 {
-	struct nf_nat_protocol *proto;
+	const struct nf_nat_protocol *proto;
 	int ret = 0;
 
 	/* If we are supposed to map IPs, then we must be in the
@@ -226,7 +226,7 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 struct nf_conn *ct,
 		 enum nf_nat_manip_type maniptype)
 {
-	struct nf_nat_protocol *proto;
+	const struct nf_nat_protocol *proto;
 
 	/* 1) If this srcip/proto/src-proto-part is currently mapped,
 	   and that same mapping gives a unique tuple within the given
@@ -355,7 +355,7 @@ manip_pkt(u_int16_t proto,
 	  enum nf_nat_manip_type maniptype)
 {
 	struct iphdr *iph;
-	struct nf_nat_protocol *p;
+	const struct nf_nat_protocol *p;
 
 	if (!skb_make_writable(skb, iphdroff + sizeof(*iph)))
 		return 0;
@@ -515,7 +515,7 @@ int nf_nat_icmp_reply_translation(struct nf_conn *ct,
 EXPORT_SYMBOL_GPL(nf_nat_icmp_reply_translation);
 
 /* Protocol registration. */
-int nf_nat_protocol_register(struct nf_nat_protocol *proto)
+int nf_nat_protocol_register(const struct nf_nat_protocol *proto)
 {
 	int ret = 0;
 
@@ -532,7 +532,7 @@ int nf_nat_protocol_register(struct nf_nat_protocol *proto)
 EXPORT_SYMBOL(nf_nat_protocol_register);
 
 /* Noone stores the protocol anywhere; simply delete it. */
-void nf_nat_protocol_unregister(struct nf_nat_protocol *proto)
+void nf_nat_protocol_unregister(const struct nf_nat_protocol *proto)
 {
 	write_lock_bh(&nf_nat_lock);
 	rcu_assign_pointer(nf_nat_protos[proto->protonum],
diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c
index 945e0ae78ffa..9fa272e73113 100644
--- a/net/ipv4/netfilter/nf_nat_proto_gre.c
+++ b/net/ipv4/netfilter/nf_nat_proto_gre.c
@@ -135,7 +135,7 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 	return 1;
 }
 
-static struct nf_nat_protocol gre __read_mostly = {
+static const struct nf_nat_protocol gre = {
 	.name			= "GRE",
 	.protonum		= IPPROTO_GRE,
 	.me			= THIS_MODULE,
diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c
index 088bb141c15d..a0e44c953cb6 100644
--- a/net/ipv4/netfilter/nf_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c
@@ -71,7 +71,7 @@ icmp_manip_pkt(struct sk_buff *skb,
 	return 1;
 }
 
-struct nf_nat_protocol nf_nat_protocol_icmp = {
+const struct nf_nat_protocol nf_nat_protocol_icmp = {
 	.name			= "ICMP",
 	.protonum		= IPPROTO_ICMP,
 	.me			= THIS_MODULE,
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index 633c53ff3d32..da23e9fbe679 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -137,7 +137,7 @@ tcp_manip_pkt(struct sk_buff *skb,
 	return 1;
 }
 
-struct nf_nat_protocol nf_nat_protocol_tcp = {
+const struct nf_nat_protocol nf_nat_protocol_tcp = {
 	.name			= "TCP",
 	.protonum		= IPPROTO_TCP,
 	.me			= THIS_MODULE,
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index 9c6519cd15f5..10df4db078af 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -127,7 +127,7 @@ udp_manip_pkt(struct sk_buff *skb,
 	return 1;
 }
 
-struct nf_nat_protocol nf_nat_protocol_udp = {
+const struct nf_nat_protocol nf_nat_protocol_udp = {
 	.name			= "UDP",
 	.protonum		= IPPROTO_UDP,
 	.me			= THIS_MODULE,
diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c
index cfd2742e9706..a26efeb073cb 100644
--- a/net/ipv4/netfilter/nf_nat_proto_unknown.c
+++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c
@@ -45,7 +45,7 @@ unknown_manip_pkt(struct sk_buff *skb,
 	return 1;
 }
 
-struct nf_nat_protocol nf_nat_unknown_protocol = {
+const struct nf_nat_protocol nf_nat_unknown_protocol = {
 	.name			= "unknown",
 	/* .me isn't set: getting a ref to this cannot fail. */
 	.manip_pkt		= unknown_manip_pkt,
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 75012585efe0..7851065ef206 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -695,7 +695,7 @@ static int nfnetlink_parse_nat_proto(struct nlattr *attr,
 				     struct nf_nat_range *range)
 {
 	struct nlattr *tb[CTA_PROTONAT_MAX+1];
-	struct nf_nat_protocol *npt;
+	const struct nf_nat_protocol *npt;
 	int err;
 
 	err = nla_parse_nested(tb, CTA_PROTONAT_MAX, attr, protonat_nla_policy);
-- 
cgit v1.2.3


From cc01dcbd26865addfe9eb5431f1f9dbc511515ba Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:38:20 -0800
Subject: [NETFILTER]: nf_nat: pass manip type instead of hook to
 nf_nat_setup_info

nf_nat_setup_info gets the hook number and translates that to the
manip type to perform. This is a relict from the time when one
manip per hook could exist, the exact hook number doesn't matter
anymore, its converted to the manip type. Most callers already
know what kind of NAT they want to perform, so pass the maniptype
in directly.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_nat.h       |  2 +-
 net/ipv4/netfilter/ipt_MASQUERADE.c  |  2 +-
 net/ipv4/netfilter/ipt_NETMAP.c      |  2 +-
 net/ipv4/netfilter/ipt_REDIRECT.c    |  2 +-
 net/ipv4/netfilter/nf_nat_core.c     |  9 +++------
 net/ipv4/netfilter/nf_nat_h323.c     | 16 ++++------------
 net/ipv4/netfilter/nf_nat_helper.c   |  6 ++----
 net/ipv4/netfilter/nf_nat_pptp.c     |  6 ++----
 net/ipv4/netfilter/nf_nat_rule.c     |  8 ++++----
 net/ipv4/netfilter/nf_nat_sip.c      |  6 ++----
 net/netfilter/nf_conntrack_netlink.c | 10 ++++------
 11 files changed, 25 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h
index 76da32292bcd..9dc1039ff78b 100644
--- a/include/net/netfilter/nf_nat.h
+++ b/include/net/netfilter/nf_nat.h
@@ -80,7 +80,7 @@ struct nf_conn_nat
 /* Set up the info structure to map into this range. */
 extern unsigned int nf_nat_setup_info(struct nf_conn *ct,
 				      const struct nf_nat_range *range,
-				      unsigned int hooknum);
+				      enum nf_nat_manip_type maniptype);
 
 /* Is this tuple already taken? (not by us)*/
 extern int nf_nat_used_tuple(const struct nf_conntrack_tuple *tuple,
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index f54150356cee..1cbff7b3084f 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -95,7 +95,7 @@ masquerade_tg(struct sk_buff *skb, const struct net_device *in,
 		  mr->range[0].min, mr->range[0].max });
 
 	/* Hand modified range to generic setup. */
-	return nf_nat_setup_info(ct, &newrange, hooknum);
+	return nf_nat_setup_info(ct, &newrange, IP_NAT_MANIP_SRC);
 }
 
 static int
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
index 8b8263e63579..5b71ef4d848a 100644
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ b/net/ipv4/netfilter/ipt_NETMAP.c
@@ -70,7 +70,7 @@ netmap_tg(struct sk_buff *skb, const struct net_device *in,
 		  mr->range[0].min, mr->range[0].max });
 
 	/* Hand modified range to generic setup. */
-	return nf_nat_setup_info(ct, &newrange, hooknum);
+	return nf_nat_setup_info(ct, &newrange, HOOK2MANIP(hooknum));
 }
 
 static struct xt_target netmap_tg_reg __read_mostly = {
diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c
index 74ce7e1e9d7f..3d9ec5c34c52 100644
--- a/net/ipv4/netfilter/ipt_REDIRECT.c
+++ b/net/ipv4/netfilter/ipt_REDIRECT.c
@@ -87,7 +87,7 @@ redirect_tg(struct sk_buff *skb, const struct net_device *in,
 		  mr->range[0].min, mr->range[0].max });
 
 	/* Hand modified range to generic setup. */
-	return nf_nat_setup_info(ct, &newrange, hooknum);
+	return nf_nat_setup_info(ct, &newrange, IP_NAT_MANIP_DST);
 }
 
 static struct xt_target redirect_tg_reg __read_mostly = {
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 7cc179cfc7c1..aec157d0ad93 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -277,12 +277,11 @@ out:
 unsigned int
 nf_nat_setup_info(struct nf_conn *ct,
 		  const struct nf_nat_range *range,
-		  unsigned int hooknum)
+		  enum nf_nat_manip_type maniptype)
 {
 	struct nf_conntrack_tuple curr_tuple, new_tuple;
 	struct nf_conn_nat *nat;
 	int have_to_hash = !(ct->status & IPS_NAT_DONE_MASK);
-	enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum);
 
 	/* nat helper or nfctnetlink also setup binding */
 	nat = nfct_nat(ct);
@@ -294,10 +293,8 @@ nf_nat_setup_info(struct nf_conn *ct,
 		}
 	}
 
-	NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING ||
-		     hooknum == NF_INET_POST_ROUTING ||
-		     hooknum == NF_INET_LOCAL_IN ||
-		     hooknum == NF_INET_LOCAL_OUT);
+	NF_CT_ASSERT(maniptype == IP_NAT_MANIP_SRC ||
+		     maniptype == IP_NAT_MANIP_DST);
 	BUG_ON(nf_nat_initialized(ct, maniptype));
 
 	/* What we've got will look like inverse of reply. Normally
diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c
index 0f226df76f5c..2e4bdee92c4a 100644
--- a/net/ipv4/netfilter/nf_nat_h323.c
+++ b/net/ipv4/netfilter/nf_nat_h323.c
@@ -389,18 +389,14 @@ static void ip_nat_q931_expect(struct nf_conn *new,
 	/* Change src to where master sends to */
 	range.flags = IP_NAT_RANGE_MAP_IPS;
 	range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip;
-
-	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(new, &range, NF_INET_POST_ROUTING);
+	nf_nat_setup_info(new, &range, IP_NAT_MANIP_SRC);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
 	range.min = range.max = this->saved_proto;
 	range.min_ip = range.max_ip =
 	    new->master->tuplehash[!this->dir].tuple.src.u3.ip;
-
-	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(new, &range, NF_INET_PRE_ROUTING);
+	nf_nat_setup_info(new, &range, IP_NAT_MANIP_DST);
 }
 
 /****************************************************************************/
@@ -479,17 +475,13 @@ static void ip_nat_callforwarding_expect(struct nf_conn *new,
 	/* Change src to where master sends to */
 	range.flags = IP_NAT_RANGE_MAP_IPS;
 	range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip;
-
-	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(new, &range, NF_INET_POST_ROUTING);
+	nf_nat_setup_info(new, &range, IP_NAT_MANIP_SRC);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
 	range.min = range.max = this->saved_proto;
 	range.min_ip = range.max_ip = this->saved_ip;
-
-	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(new, &range, NF_INET_PRE_ROUTING);
+	nf_nat_setup_info(new, &range, IP_NAT_MANIP_DST);
 }
 
 /****************************************************************************/
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index d24f3d947394..4c0232842e75 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -433,15 +433,13 @@ void nf_nat_follow_master(struct nf_conn *ct,
 	range.flags = IP_NAT_RANGE_MAP_IPS;
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
-	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
 	range.min = range.max = exp->saved_proto;
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.src.u3.ip;
-	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
 }
 EXPORT_SYMBOL(nf_nat_follow_master);
diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c
index c540999f5090..e63b944a2ebb 100644
--- a/net/ipv4/netfilter/nf_nat_pptp.c
+++ b/net/ipv4/netfilter/nf_nat_pptp.c
@@ -93,8 +93,7 @@ static void pptp_nat_expected(struct nf_conn *ct,
 		range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED;
 		range.min = range.max = exp->saved_proto;
 	}
-	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = IP_NAT_RANGE_MAP_IPS;
@@ -104,8 +103,7 @@ static void pptp_nat_expected(struct nf_conn *ct,
 		range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED;
 		range.min = range.max = exp->saved_proto;
 	}
-	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
 }
 
 /* outbound packets == from PNS to PAC */
diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c
index ee39ed87bb08..4391aec56abd 100644
--- a/net/ipv4/netfilter/nf_nat_rule.c
+++ b/net/ipv4/netfilter/nf_nat_rule.c
@@ -87,7 +87,7 @@ static unsigned int ipt_snat_target(struct sk_buff *skb,
 			    ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY));
 	NF_CT_ASSERT(out);
 
-	return nf_nat_setup_info(ct, &mr->range[0], hooknum);
+	return nf_nat_setup_info(ct, &mr->range[0], IP_NAT_MANIP_SRC);
 }
 
 /* Before 2.6.11 we did implicit source NAT if required. Warn about change. */
@@ -133,7 +133,7 @@ static unsigned int ipt_dnat_target(struct sk_buff *skb,
 		warn_if_extra_mangle(ip_hdr(skb)->daddr,
 				     mr->range[0].min_ip);
 
-	return nf_nat_setup_info(ct, &mr->range[0], hooknum);
+	return nf_nat_setup_info(ct, &mr->range[0], IP_NAT_MANIP_DST);
 }
 
 static bool ipt_snat_checkentry(const char *tablename,
@@ -184,7 +184,7 @@ alloc_null_binding(struct nf_conn *ct, unsigned int hooknum)
 
 	pr_debug("Allocating NULL binding for %p (%u.%u.%u.%u)\n",
 		 ct, NIPQUAD(ip));
-	return nf_nat_setup_info(ct, &range, hooknum);
+	return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum));
 }
 
 unsigned int
@@ -203,7 +203,7 @@ alloc_null_binding_confirmed(struct nf_conn *ct, unsigned int hooknum)
 
 	pr_debug("Allocating NULL binding for confirmed %p (%u.%u.%u.%u)\n",
 		 ct, NIPQUAD(ip));
-	return nf_nat_setup_info(ct, &range, hooknum);
+	return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum));
 }
 
 int nf_nat_rule_find(struct sk_buff *skb,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index b8c0720cf428..606a170bf4ca 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -228,15 +228,13 @@ static void ip_nat_sdp_expect(struct nf_conn *ct,
 	range.flags = IP_NAT_RANGE_MAP_IPS;
 	range.min_ip = range.max_ip
 		= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
-	/* hook doesn't matter, but it has to do source manip */
-	nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
 
 	/* For DST manip, map port here to where it's expected. */
 	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
 	range.min = range.max = exp->saved_proto;
 	range.min_ip = range.max_ip = exp->saved_ip;
-	/* hook doesn't matter, but it has to do destination manip */
-	nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
 }
 
 /* So, this packet has hit the connection tracking matching code.
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 7851065ef206..3a065f43ddd8 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -918,19 +918,17 @@ ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[])
 			if (nfnetlink_parse_nat(cda[CTA_NAT_DST], ct,
 						&range) < 0)
 				return -EINVAL;
-			if (nf_nat_initialized(ct,
-					       HOOK2MANIP(NF_INET_PRE_ROUTING)))
+			if (nf_nat_initialized(ct, IP_NAT_MANIP_DST))
 				return -EEXIST;
-			nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING);
+			nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
 		}
 		if (cda[CTA_NAT_SRC]) {
 			if (nfnetlink_parse_nat(cda[CTA_NAT_SRC], ct,
 						&range) < 0)
 				return -EINVAL;
-			if (nf_nat_initialized(ct,
-					       HOOK2MANIP(NF_INET_POST_ROUTING)))
+			if (nf_nat_initialized(ct, IP_NAT_MANIP_SRC))
 				return -EEXIST;
-			nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING);
+			nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
 		}
 #endif
 	}
-- 
cgit v1.2.3


From f01ffbd6e7d001ccf9168b33507958a51ce0ffcf Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:38:49 -0800
Subject: [NETFILTER]: nf_log: move logging stuff to seperate header

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                      | 55 ------------------------
 include/net/netfilter/nf_log.h                 | 59 ++++++++++++++++++++++++++
 net/bridge/netfilter/ebt_log.c                 |  1 +
 net/bridge/netfilter/ebt_ulog.c                |  1 +
 net/ipv4/netfilter/ip_tables.c                 |  1 +
 net/ipv4/netfilter/ipt_LOG.c                   |  1 +
 net/ipv4/netfilter/ipt_ULOG.c                  |  1 +
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  1 +
 net/ipv6/netfilter/ip6_tables.c                |  1 +
 net/ipv6/netfilter/ip6t_LOG.c                  |  1 +
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  1 +
 net/netfilter/nf_conntrack_proto_tcp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udp.c         |  1 +
 net/netfilter/nf_conntrack_proto_udplite.c     |  1 +
 net/netfilter/nf_log.c                         |  1 +
 net/netfilter/nfnetlink_log.c                  |  1 +
 net/netfilter/xt_NFLOG.c                       |  1 +
 17 files changed, 74 insertions(+), 55 deletions(-)
 create mode 100644 include/net/netfilter/nf_log.h

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index f25eec595807..368b7ed1f1bf 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -124,61 +124,6 @@ extern struct ctl_table nf_net_ipv4_netfilter_sysctl_path[];
 
 extern struct list_head nf_hooks[NPROTO][NF_MAX_HOOKS];
 
-/* those NF_LOG_* defines and struct nf_loginfo are legacy definitios that will
- * disappear once iptables is replaced with pkttables.  Please DO NOT use them
- * for any new code! */
-#define NF_LOG_TCPSEQ		0x01	/* Log TCP sequence numbers */
-#define NF_LOG_TCPOPT		0x02	/* Log TCP options */
-#define NF_LOG_IPOPT		0x04	/* Log IP options */
-#define NF_LOG_UID		0x08	/* Log UID owning local socket */
-#define NF_LOG_MASK		0x0f
-
-#define NF_LOG_TYPE_LOG		0x01
-#define NF_LOG_TYPE_ULOG	0x02
-
-struct nf_loginfo {
-	u_int8_t type;
-	union {
-		struct {
-			u_int32_t copy_len;
-			u_int16_t group;
-			u_int16_t qthreshold;
-		} ulog;
-		struct {
-			u_int8_t level;
-			u_int8_t logflags;
-		} log;
-	} u;
-};
-
-typedef void nf_logfn(unsigned int pf,
-		      unsigned int hooknum,
-		      const struct sk_buff *skb,
-		      const struct net_device *in,
-		      const struct net_device *out,
-		      const struct nf_loginfo *li,
-		      const char *prefix);
-
-struct nf_logger {
-	struct module	*me;
-	nf_logfn 	*logfn;
-	char		*name;
-};
-
-/* Function to register/unregister log function. */
-int nf_log_register(int pf, struct nf_logger *logger);
-void nf_log_unregister(struct nf_logger *logger);
-void nf_log_unregister_pf(int pf);
-
-/* Calls the registered backend logging function */
-void nf_log_packet(int pf,
-		   unsigned int hooknum,
-		   const struct sk_buff *skb,
-		   const struct net_device *in,
-		   const struct net_device *out,
-		   struct nf_loginfo *li,
-		   const char *fmt, ...);
-
 int nf_hook_slow(int pf, unsigned int hook, struct sk_buff *skb,
 		 struct net_device *indev, struct net_device *outdev,
 		 int (*okfn)(struct sk_buff *), int thresh);
diff --git a/include/net/netfilter/nf_log.h b/include/net/netfilter/nf_log.h
new file mode 100644
index 000000000000..f0426e59f229
--- /dev/null
+++ b/include/net/netfilter/nf_log.h
@@ -0,0 +1,59 @@
+#ifndef _NF_LOG_H
+#define _NF_LOG_H
+
+/* those NF_LOG_* defines and struct nf_loginfo are legacy definitios that will
+ * disappear once iptables is replaced with pkttables.  Please DO NOT use them
+ * for any new code! */
+#define NF_LOG_TCPSEQ		0x01	/* Log TCP sequence numbers */
+#define NF_LOG_TCPOPT		0x02	/* Log TCP options */
+#define NF_LOG_IPOPT		0x04	/* Log IP options */
+#define NF_LOG_UID		0x08	/* Log UID owning local socket */
+#define NF_LOG_MASK		0x0f
+
+#define NF_LOG_TYPE_LOG		0x01
+#define NF_LOG_TYPE_ULOG	0x02
+
+struct nf_loginfo {
+	u_int8_t type;
+	union {
+		struct {
+			u_int32_t copy_len;
+			u_int16_t group;
+			u_int16_t qthreshold;
+		} ulog;
+		struct {
+			u_int8_t level;
+			u_int8_t logflags;
+		} log;
+	} u;
+};
+
+typedef void nf_logfn(unsigned int pf,
+		      unsigned int hooknum,
+		      const struct sk_buff *skb,
+		      const struct net_device *in,
+		      const struct net_device *out,
+		      const struct nf_loginfo *li,
+		      const char *prefix);
+
+struct nf_logger {
+	struct module	*me;
+	nf_logfn 	*logfn;
+	char		*name;
+};
+
+/* Function to register/unregister log function. */
+int nf_log_register(int pf, struct nf_logger *logger);
+void nf_log_unregister(struct nf_logger *logger);
+void nf_log_unregister_pf(int pf);
+
+/* Calls the registered backend logging function */
+void nf_log_packet(int pf,
+		   unsigned int hooknum,
+		   const struct sk_buff *skb,
+		   const struct net_device *in,
+		   const struct net_device *out,
+		   struct nf_loginfo *li,
+		   const char *fmt, ...);
+
+#endif /* _NF_LOG_H */
diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index 457815fb5584..fcb3b54dc191 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -17,6 +17,7 @@
 #include <linux/in.h>
 #include <linux/if_arp.h>
 #include <linux/spinlock.h>
+#include <net/netfilter/nf_log.h>
 
 static DEFINE_SPINLOCK(ebt_log_lock);
 
diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c
index e7cfd30bac75..1b9ca07f44ff 100644
--- a/net/bridge/netfilter/ebt_ulog.c
+++ b/net/bridge/netfilter/ebt_ulog.c
@@ -38,6 +38,7 @@
 #include <linux/netdevice.h>
 #include <linux/netfilter_bridge/ebtables.h>
 #include <linux/netfilter_bridge/ebt_ulog.h>
+#include <net/netfilter/nf_log.h>
 #include <net/sock.h>
 #include "../br_private.h"
 
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 439b2925765d..271f6a5d3d4c 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -26,6 +26,7 @@
 
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
+#include <net/netfilter/nf_log.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c
index f8c613a6eb04..4b346e59bf2d 100644
--- a/net/ipv4/netfilter/ipt_LOG.c
+++ b/net/ipv4/netfilter/ipt_LOG.c
@@ -22,6 +22,7 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ipt_LOG.h>
+#include <net/netfilter/nf_log.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index 4139042a63a1..1d8e146345e4 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -43,6 +43,7 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ipt_ULOG.h>
+#include <net/netfilter/nf_log.h>
 #include <net/sock.h>
 #include <linux/bitops.h>
 #include <asm/unaligned.h>
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 3e2e5cdda9de..cd0d6690627e 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -18,6 +18,7 @@
 #include <net/netfilter/nf_conntrack_tuple.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_core.h>
+#include <net/netfilter/nf_log.h>
 
 static unsigned long nf_ct_icmp_timeout __read_mostly = 30*HZ;
 
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index d910d56d22da..bb50d0e66734 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -28,6 +28,7 @@
 
 #include <linux/netfilter_ipv6/ip6_tables.h>
 #include <linux/netfilter/x_tables.h>
+#include <net/netfilter/nf_log.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index 19523242991d..e6a2b1e9469c 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -23,6 +23,7 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv6/ip6_tables.h>
+#include <net/netfilter/nf_log.h>
 
 MODULE_AUTHOR("Jan Rekorajski <baggins@pld.org.pl>");
 MODULE_DESCRIPTION("IP6 tables LOG target module");
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index 44689d444410..02d60dfbab80 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -24,6 +24,7 @@
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/ipv6/nf_conntrack_icmpv6.h>
+#include <net/netfilter/nf_log.h>
 
 static unsigned long nf_ct_icmpv6_timeout __read_mostly = 30*HZ;
 
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 600b476d225c..1d496b912bd0 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -24,6 +24,7 @@
 #include <net/netfilter/nf_conntrack.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_ecache.h>
+#include <net/netfilter/nf_log.h>
 
 /* Protects conntrack->proto.tcp */
 static DEFINE_RWLOCK(tcp_lock);
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index 570a2e109478..7ac607319562 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -21,6 +21,7 @@
 #include <linux/netfilter_ipv6.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_ecache.h>
+#include <net/netfilter/nf_log.h>
 
 static unsigned int nf_ct_udp_timeout __read_mostly = 30*HZ;
 static unsigned int nf_ct_udp_timeout_stream __read_mostly = 180*HZ;
diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c
index 7e116d5766d1..6518bcd17d65 100644
--- a/net/netfilter/nf_conntrack_proto_udplite.c
+++ b/net/netfilter/nf_conntrack_proto_udplite.c
@@ -22,6 +22,7 @@
 #include <linux/netfilter_ipv6.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_ecache.h>
+#include <net/netfilter/nf_log.h>
 
 static unsigned int nf_ct_udplite_timeout __read_mostly = 30*HZ;
 static unsigned int nf_ct_udplite_timeout_stream __read_mostly = 180*HZ;
diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index d67c4fbf6031..fad97d69481c 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -6,6 +6,7 @@
 #include <linux/netfilter.h>
 #include <linux/seq_file.h>
 #include <net/protocol.h>
+#include <net/netfilter/nf_log.h>
 
 #include "nf_internals.h"
 
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 2c7bd2eb0294..959a0cb131f1 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -29,6 +29,7 @@
 #include <linux/jhash.h>
 #include <linux/random.h>
 #include <net/sock.h>
+#include <net/netfilter/nf_log.h>
 
 #include <asm/atomic.h>
 
diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c
index 83af124e88cd..866facfa4f43 100644
--- a/net/netfilter/xt_NFLOG.c
+++ b/net/netfilter/xt_NFLOG.c
@@ -12,6 +12,7 @@
 
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_NFLOG.h>
+#include <net/netfilter/nf_log.h>
 
 MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
 MODULE_DESCRIPTION("x_tables NFLOG target");
-- 
cgit v1.2.3


From 7b2f9631e789c3e7d59201c21f09a24cd6ce3b1a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:39:08 -0800
Subject: [NETFILTER]: nf_log: constify struct nf_logger and nf_log_packet
 loginfo arg

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_log.h  |  6 +++---
 net/bridge/netfilter/ebt_log.c  |  2 +-
 net/bridge/netfilter/ebt_ulog.c |  2 +-
 net/ipv4/netfilter/ipt_LOG.c    |  2 +-
 net/ipv6/netfilter/ip6t_LOG.c   |  2 +-
 net/netfilter/nf_log.c          | 10 +++++-----
 net/netfilter/nfnetlink_log.c   |  2 +-
 7 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_log.h b/include/net/netfilter/nf_log.h
index f0426e59f229..037e82403f91 100644
--- a/include/net/netfilter/nf_log.h
+++ b/include/net/netfilter/nf_log.h
@@ -43,8 +43,8 @@ struct nf_logger {
 };
 
 /* Function to register/unregister log function. */
-int nf_log_register(int pf, struct nf_logger *logger);
-void nf_log_unregister(struct nf_logger *logger);
+int nf_log_register(int pf, const struct nf_logger *logger);
+void nf_log_unregister(const struct nf_logger *logger);
 void nf_log_unregister_pf(int pf);
 
 /* Calls the registered backend logging function */
@@ -53,7 +53,7 @@ void nf_log_packet(int pf,
 		   const struct sk_buff *skb,
 		   const struct net_device *in,
 		   const struct net_device *out,
-		   struct nf_loginfo *li,
+		   const struct nf_loginfo *li,
 		   const char *fmt, ...);
 
 #endif /* _NF_LOG_H */
diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index fcb3b54dc191..3be9e9898553 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -183,7 +183,7 @@ static struct ebt_watcher log =
 	.me		= THIS_MODULE,
 };
 
-static struct nf_logger ebt_log_logger = {
+static const struct nf_logger ebt_log_logger = {
 	.name 		= "ebt_log",
 	.logfn		= &ebt_log_packet,
 	.me		= THIS_MODULE,
diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c
index 1b9ca07f44ff..b73ba28bcbe8 100644
--- a/net/bridge/netfilter/ebt_ulog.c
+++ b/net/bridge/netfilter/ebt_ulog.c
@@ -279,7 +279,7 @@ static struct ebt_watcher ulog = {
 	.me		= THIS_MODULE,
 };
 
-static struct nf_logger ebt_ulog_logger = {
+static const struct nf_logger ebt_ulog_logger = {
 	.name		= EBT_ULOG_WATCHER,
 	.logfn		= &ebt_log_packet,
 	.me		= THIS_MODULE,
diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c
index 4b346e59bf2d..5acdddfa6add 100644
--- a/net/ipv4/netfilter/ipt_LOG.c
+++ b/net/ipv4/netfilter/ipt_LOG.c
@@ -465,7 +465,7 @@ static struct xt_target log_tg_reg __read_mostly = {
 	.me		= THIS_MODULE,
 };
 
-static struct nf_logger ipt_log_logger ={
+static const struct nf_logger ipt_log_logger ={
 	.name		= "ipt_LOG",
 	.logfn		= &ipt_log_packet,
 	.me		= THIS_MODULE,
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index e6a2b1e9469c..474c2b12621a 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -478,7 +478,7 @@ static struct xt_target log_tg6_reg __read_mostly = {
 	.me 		= THIS_MODULE,
 };
 
-static struct nf_logger ip6t_logger = {
+static const struct nf_logger ip6t_logger = {
 	.name		= "ip6t_LOG",
 	.logfn		= &ip6t_log_packet,
 	.me		= THIS_MODULE,
diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index fad97d69481c..ed9116dc78b5 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -15,12 +15,12 @@
 
 #define NF_LOG_PREFIXLEN		128
 
-static struct nf_logger *nf_loggers[NPROTO];
+static const struct nf_logger *nf_loggers[NPROTO] __read_mostly;
 static DEFINE_MUTEX(nf_log_mutex);
 
 /* return EBUSY if somebody else is registered, EEXIST if the same logger
  * is registred, 0 on success. */
-int nf_log_register(int pf, struct nf_logger *logger)
+int nf_log_register(int pf, const struct nf_logger *logger)
 {
 	int ret;
 
@@ -58,7 +58,7 @@ void nf_log_unregister_pf(int pf)
 }
 EXPORT_SYMBOL(nf_log_unregister_pf);
 
-void nf_log_unregister(struct nf_logger *logger)
+void nf_log_unregister(const struct nf_logger *logger)
 {
 	int i;
 
@@ -78,12 +78,12 @@ void nf_log_packet(int pf,
 		   const struct sk_buff *skb,
 		   const struct net_device *in,
 		   const struct net_device *out,
-		   struct nf_loginfo *loginfo,
+		   const struct nf_loginfo *loginfo,
 		   const char *fmt, ...)
 {
 	va_list args;
 	char prefix[NF_LOG_PREFIXLEN];
-	struct nf_logger *logger;
+	const struct nf_logger *logger;
 
 	rcu_read_lock();
 	logger = rcu_dereference(nf_loggers[pf]);
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 959a0cb131f1..02e63577e156 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -726,7 +726,7 @@ nfulnl_recv_unsupp(struct sock *ctnl, struct sk_buff *skb,
 	return -ENOTSUPP;
 }
 
-static struct nf_logger nfulnl_logger = {
+static const struct nf_logger nfulnl_logger = {
 	.name	= "nfnetlink_log",
 	.logfn	= &nfulnl_log_packet,
 	.me	= THIS_MODULE,
-- 
cgit v1.2.3


From 76aa1ce139f649e432272f6ad75204b763ef13bd Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:41:52 -0800
Subject: [NETFILTER]: nfnetlink_log: include GID in netlink message

Similar to Maciej Soltysiak's ipt_LOG patch, include GID in addition
to UID in netlink message.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nfnetlink_log.h | 1 +
 net/netfilter/nfnetlink_log.c           | 3 +++
 2 files changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h
index 5966afa026e9..a85721332924 100644
--- a/include/linux/netfilter/nfnetlink_log.h
+++ b/include/linux/netfilter/nfnetlink_log.h
@@ -47,6 +47,7 @@ enum nfulnl_attr_type {
 	NFULA_UID,			/* user id of socket */
 	NFULA_SEQ,			/* instance-local sequence number */
 	NFULA_SEQ_GLOBAL,		/* global sequence number */
+	NFULA_GID,			/* group id of socket */
 
 	__NFULA_MAX
 };
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 950b1f0713dd..5013cb97ce2b 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -467,9 +467,11 @@ __build_packet_message(struct nfulnl_instance *inst,
 		read_lock_bh(&skb->sk->sk_callback_lock);
 		if (skb->sk->sk_socket && skb->sk->sk_socket->file) {
 			__be32 uid = htonl(skb->sk->sk_socket->file->f_uid);
+			__be32 gid = htons(skb->sk->sk_socket->file->f_gid);
 			/* need to unlock here since NLA_PUT may goto */
 			read_unlock_bh(&skb->sk->sk_callback_lock);
 			NLA_PUT_BE32(inst->skb, NFULA_UID, uid);
+			NLA_PUT_BE32(inst->skb, NFULA_GID, gid);
 		} else
 			read_unlock_bh(&skb->sk->sk_callback_lock);
 	}
@@ -564,6 +566,7 @@ nfulnl_log_packet(unsigned int pf,
 #endif
 		+ nla_total_size(sizeof(u_int32_t))	/* mark */
 		+ nla_total_size(sizeof(u_int32_t))	/* uid */
+		+ nla_total_size(sizeof(u_int32_t))	/* gid */
 		+ nla_total_size(plen)			/* prefix */
 		+ nla_total_size(sizeof(struct nfulnl_msg_packet_hw))
 		+ nla_total_size(sizeof(struct nfulnl_msg_packet_timestamp));
-- 
cgit v1.2.3


From 90a9ba8dd90bcffe279d3272545bccba6dcc8d7c Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:42:09 -0800
Subject: [NETFILTER]: Kill function prototype for non-existing function

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 368b7ed1f1bf..bd4a2dd54233 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -206,9 +206,6 @@ int compat_nf_setsockopt(struct sock *sk, int pf, int optval,
 int compat_nf_getsockopt(struct sock *sk, int pf, int optval,
 		char __user *opt, int *len);
 
-/* FIXME: Before cache is ever used, this must be implemented for real. */
-extern void nf_invalidate_cache(int pf);
-
 /* Call this before modifying an existing packet: ensures it is
    modifiable and linear to the point you care about (writable_len).
    Returns true or false. */
-- 
cgit v1.2.3


From 1e796fda00f06bac584f0e4ad8750ab9430d79d3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:42:27 -0800
Subject: [NETFILTER]: constify nf_afinfo

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h              | 10 +++++-----
 net/ipv4/netfilter.c                   |  2 +-
 net/ipv6/netfilter.c                   |  2 +-
 net/netfilter/core.c                   |  6 +++---
 net/netfilter/nf_conntrack_h323_main.c |  2 +-
 net/netfilter/nf_queue.c               |  4 ++--
 6 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index bd4a2dd54233..0947424d01d6 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -226,8 +226,8 @@ struct nf_afinfo {
 	int		route_key_size;
 };
 
-extern struct nf_afinfo *nf_afinfo[];
-static inline struct nf_afinfo *nf_get_afinfo(unsigned short family)
+extern const struct nf_afinfo *nf_afinfo[NPROTO];
+static inline const struct nf_afinfo *nf_get_afinfo(unsigned short family)
 {
 	return rcu_dereference(nf_afinfo[family]);
 }
@@ -236,7 +236,7 @@ static inline __sum16
 nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff,
 	    u_int8_t protocol, unsigned short family)
 {
-	struct nf_afinfo *afinfo;
+	const struct nf_afinfo *afinfo;
 	__sum16 csum = 0;
 
 	rcu_read_lock();
@@ -247,8 +247,8 @@ nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff,
 	return csum;
 }
 
-extern int nf_register_afinfo(struct nf_afinfo *afinfo);
-extern void nf_unregister_afinfo(struct nf_afinfo *afinfo);
+extern int nf_register_afinfo(const struct nf_afinfo *afinfo);
+extern void nf_unregister_afinfo(const struct nf_afinfo *afinfo);
 
 #include <net/flow.h>
 extern void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 7bf5e4a199f0..4011f8f987c0 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -190,7 +190,7 @@ static int nf_ip_route(struct dst_entry **dst, struct flowi *fl)
 	return ip_route_output_key((struct rtable **)dst, fl);
 }
 
-static struct nf_afinfo nf_ip_afinfo = {
+static const struct nf_afinfo nf_ip_afinfo = {
 	.family		= AF_INET,
 	.checksum	= nf_ip_checksum,
 	.route		= nf_ip_route,
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 945e6ae19569..2e06724dc348 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -124,7 +124,7 @@ __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 
 EXPORT_SYMBOL(nf_ip6_checksum);
 
-static struct nf_afinfo nf_ip6_afinfo = {
+static const struct nf_afinfo nf_ip6_afinfo = {
 	.family		= AF_INET6,
 	.checksum	= nf_ip6_checksum,
 	.route		= nf_ip6_route,
diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 95e18635ce7a..e0263445484a 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -26,10 +26,10 @@
 
 static DEFINE_MUTEX(afinfo_mutex);
 
-struct nf_afinfo *nf_afinfo[NPROTO] __read_mostly;
+const struct nf_afinfo *nf_afinfo[NPROTO] __read_mostly;
 EXPORT_SYMBOL(nf_afinfo);
 
-int nf_register_afinfo(struct nf_afinfo *afinfo)
+int nf_register_afinfo(const struct nf_afinfo *afinfo)
 {
 	int err;
 
@@ -42,7 +42,7 @@ int nf_register_afinfo(struct nf_afinfo *afinfo)
 }
 EXPORT_SYMBOL_GPL(nf_register_afinfo);
 
-void nf_unregister_afinfo(struct nf_afinfo *afinfo)
+void nf_unregister_afinfo(const struct nf_afinfo *afinfo)
 {
 	mutex_lock(&afinfo_mutex);
 	rcu_assign_pointer(nf_afinfo[afinfo->family], NULL);
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index c550257b5462..b636ca60a770 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -708,7 +708,7 @@ static int callforward_do_filter(union nf_conntrack_address *src,
 				 union nf_conntrack_address *dst,
 				 int family)
 {
-	struct nf_afinfo *afinfo;
+	const struct nf_afinfo *afinfo;
 	struct flowi fl1, fl2;
 	int ret = 0;
 
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 77965114b091..bfc2928c1912 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -119,7 +119,7 @@ static int __nf_queue(struct sk_buff *skb,
 	struct net_device *physindev;
 	struct net_device *physoutdev;
 #endif
-	struct nf_afinfo *afinfo;
+	const struct nf_afinfo *afinfo;
 	const struct nf_queue_handler *qh;
 
 	/* QUEUE == DROP if noone is waiting, to be safe. */
@@ -233,7 +233,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict)
 {
 	struct sk_buff *skb = entry->skb;
 	struct list_head *elem = &entry->elem->list;
-	struct nf_afinfo *afinfo;
+	const struct nf_afinfo *afinfo;
 
 	rcu_read_lock();
 
-- 
cgit v1.2.3


From 051578ccbcdad3b24b621dfb652194e36759e8d5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 17 Dec 2007 22:42:51 -0800
Subject: [NETFILTER]: nf_nat: properly use RCU for ip_nat_decode_session

We need to use rcu_assign_pointer/rcu_dereference to avoid races.
Also remove an obsolete CONFIG_IP_NAT_NEEDED ifdef.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h              | 11 ++++++++---
 net/ipv4/netfilter/nf_nat_standalone.c |  6 +++---
 2 files changed, 11 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 0947424d01d6..1a8487325a4f 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -256,11 +256,16 @@ extern void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
 static inline void
 nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, int family)
 {
-#if defined(CONFIG_IP_NF_NAT_NEEDED) || defined(CONFIG_NF_NAT_NEEDED)
+#ifdef CONFIG_NF_NAT_NEEDED
 	void (*decodefn)(struct sk_buff *, struct flowi *);
 
-	if (family == AF_INET && (decodefn = ip_nat_decode_session) != NULL)
-		decodefn(skb, fl);
+	if (family == AF_INET) {
+		rcu_read_lock();
+		decodefn = rcu_dereference(ip_nat_decode_session);
+		if (decodefn)
+			decodefn(skb, fl);
+		rcu_read_unlock();
+	}
 #endif
 }
 
diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c
index a2b02f01cc5c..99b2c788d5a8 100644
--- a/net/ipv4/netfilter/nf_nat_standalone.c
+++ b/net/ipv4/netfilter/nf_nat_standalone.c
@@ -332,7 +332,7 @@ static int __init nf_nat_standalone_init(void)
 
 #ifdef CONFIG_XFRM
 	BUG_ON(ip_nat_decode_session != NULL);
-	ip_nat_decode_session = nat_decode_session;
+	rcu_assign_pointer(ip_nat_decode_session, nat_decode_session);
 #endif
 	ret = nf_nat_rule_init();
 	if (ret < 0) {
@@ -350,7 +350,7 @@ static int __init nf_nat_standalone_init(void)
 	nf_nat_rule_cleanup();
  cleanup_decode_session:
 #ifdef CONFIG_XFRM
-	ip_nat_decode_session = NULL;
+	rcu_assign_pointer(ip_nat_decode_session, NULL);
 	synchronize_net();
 #endif
 	return ret;
@@ -361,7 +361,7 @@ static void __exit nf_nat_standalone_fini(void)
 	nf_unregister_hooks(nf_nat_ops, ARRAY_SIZE(nf_nat_ops));
 	nf_nat_rule_cleanup();
 #ifdef CONFIG_XFRM
-	ip_nat_decode_session = NULL;
+	rcu_assign_pointer(ip_nat_decode_session, NULL);
 	synchronize_net();
 #endif
 	/* Conntrack caches are unregistered in nf_conntrack_cleanup */
-- 
cgit v1.2.3


From 643a2c15a407faf08101a20e1a3461160711899d Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 17 Dec 2007 22:43:50 -0800
Subject: [NETFILTER]: Introduce nf_inet_address

A few netfilter modules provide their own union of IPv4 and IPv6
address storage. Will unify that in this patch series.

(1/4): Rename union nf_conntrack_address to union nf_inet_addr and
move it to x_tables.h.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                   |  6 +++++
 include/linux/netfilter/nf_conntrack_h323.h |  6 ++---
 include/net/netfilter/nf_conntrack_expect.h |  4 ++--
 include/net/netfilter/nf_conntrack_tuple.h  | 17 +++++----------
 net/ipv4/netfilter/nf_nat_h323.c            | 10 ++++-----
 net/netfilter/nf_conntrack_expect.c         |  4 ++--
 net/netfilter/nf_conntrack_ftp.c            |  2 +-
 net/netfilter/nf_conntrack_h323_main.c      | 34 ++++++++++++++---------------
 net/netfilter/nf_conntrack_sip.c            |  8 +++----
 net/netfilter/xt_connlimit.c                | 20 ++++++++---------
 10 files changed, 55 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 1a8487325a4f..d190d560de61 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -48,6 +48,12 @@ enum nf_inet_hooks {
 	NF_INET_NUMHOOKS
 };
 
+union nf_inet_addr {
+	u_int32_t	all[4];
+	__be32		ip;
+	__be32		ip6[4];
+};
+
 #ifdef __KERNEL__
 #ifdef CONFIG_NETFILTER
 
diff --git a/include/linux/netfilter/nf_conntrack_h323.h b/include/linux/netfilter/nf_conntrack_h323.h
index aabd24ac7631..26f9226ea72b 100644
--- a/include/linux/netfilter/nf_conntrack_h323.h
+++ b/include/linux/netfilter/nf_conntrack_h323.h
@@ -31,7 +31,7 @@ struct nf_conn;
 
 extern int get_h225_addr(struct nf_conn *ct, unsigned char *data,
 			 TransportAddress *taddr,
-			 union nf_conntrack_address *addr, __be16 *port);
+			 union nf_inet_addr *addr, __be16 *port);
 extern void nf_conntrack_h245_expect(struct nf_conn *new,
 				     struct nf_conntrack_expect *this);
 extern void nf_conntrack_q931_expect(struct nf_conn *new,
@@ -39,12 +39,12 @@ extern void nf_conntrack_q931_expect(struct nf_conn *new,
 extern int (*set_h245_addr_hook) (struct sk_buff *skb,
 				  unsigned char **data, int dataoff,
 				  H245_TransportAddress *taddr,
-				  union nf_conntrack_address *addr,
+				  union nf_inet_addr *addr,
 				  __be16 port);
 extern int (*set_h225_addr_hook) (struct sk_buff *skb,
 				  unsigned char **data, int dataoff,
 				  TransportAddress *taddr,
-				  union nf_conntrack_address *addr,
+				  union nf_inet_addr *addr,
 				  __be16 port);
 extern int (*set_sig_addr_hook) (struct sk_buff *skb,
 				 struct nf_conn *ct,
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index b47c04f12dbe..6c3fd254c28e 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -73,8 +73,8 @@ void nf_ct_unexpect_related(struct nf_conntrack_expect *exp);
    nf_ct_expect_related.  You will have to call put afterwards. */
 struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me);
 void nf_ct_expect_init(struct nf_conntrack_expect *, int,
-		       union nf_conntrack_address *,
-		       union nf_conntrack_address *,
+		       union nf_inet_addr *,
+		       union nf_inet_addr *,
 		       u_int8_t, __be16 *, __be16 *);
 void nf_ct_expect_put(struct nf_conntrack_expect *exp);
 int nf_ct_expect_related(struct nf_conntrack_expect *expect);
diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index c48e390f4b0f..45cb17cdcfd0 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -10,6 +10,7 @@
 #ifndef _NF_CONNTRACK_TUPLE_H
 #define _NF_CONNTRACK_TUPLE_H
 
+#include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/nf_conntrack_tuple_common.h>
 
 /* A `tuple' is a structure containing the information to uniquely
@@ -20,15 +21,7 @@
   "non-manipulatable" lines, for the benefit of the NAT code.
 */
 
-#define NF_CT_TUPLE_L3SIZE	4
-
-/* The l3 protocol-specific manipulable parts of the tuple: always in
-   network order! */
-union nf_conntrack_address {
-	u_int32_t all[NF_CT_TUPLE_L3SIZE];
-	__be32 ip;
-	__be32 ip6[4];
-};
+#define NF_CT_TUPLE_L3SIZE	ARRAY_SIZE(((union nf_inet_addr *)NULL)->all)
 
 /* The protocol-specific manipulable parts of the tuple: always in
    network order! */
@@ -57,7 +50,7 @@ union nf_conntrack_man_proto
 /* The manipulable part of the tuple. */
 struct nf_conntrack_man
 {
-	union nf_conntrack_address u3;
+	union nf_inet_addr u3;
 	union nf_conntrack_man_proto u;
 	/* Layer 3 protocol */
 	u_int16_t l3num;
@@ -70,7 +63,7 @@ struct nf_conntrack_tuple
 
 	/* These are the parts of the tuple which are fixed. */
 	struct {
-		union nf_conntrack_address u3;
+		union nf_inet_addr u3;
 		union {
 			/* Add other protocols here. */
 			__be16 all;
@@ -103,7 +96,7 @@ struct nf_conntrack_tuple
 struct nf_conntrack_tuple_mask
 {
 	struct {
-		union nf_conntrack_address u3;
+		union nf_inet_addr u3;
 		union nf_conntrack_man_proto u;
 	} src;
 };
diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c
index 2e4bdee92c4a..a121989fdad7 100644
--- a/net/ipv4/netfilter/nf_nat_h323.c
+++ b/net/ipv4/netfilter/nf_nat_h323.c
@@ -76,7 +76,7 @@ static int set_addr(struct sk_buff *skb,
 static int set_h225_addr(struct sk_buff *skb,
 			 unsigned char **data, int dataoff,
 			 TransportAddress *taddr,
-			 union nf_conntrack_address *addr, __be16 port)
+			 union nf_inet_addr *addr, __be16 port)
 {
 	return set_addr(skb, data, dataoff, taddr->ipAddress.ip,
 			addr->ip, port);
@@ -86,7 +86,7 @@ static int set_h225_addr(struct sk_buff *skb,
 static int set_h245_addr(struct sk_buff *skb,
 			 unsigned char **data, int dataoff,
 			 H245_TransportAddress *taddr,
-			 union nf_conntrack_address *addr, __be16 port)
+			 union nf_inet_addr *addr, __be16 port)
 {
 	return set_addr(skb, data, dataoff,
 			taddr->unicastAddress.iPAddress.network,
@@ -103,7 +103,7 @@ static int set_sig_addr(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int i;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 
 	for (i = 0; i < count; i++) {
 		if (get_h225_addr(ct, *data, &taddr[i], &addr, &port)) {
@@ -155,7 +155,7 @@ static int set_ras_addr(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int i;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 
 	for (i = 0; i < count; i++) {
 		if (get_h225_addr(ct, *data, &taddr[i], &addr, &port) &&
@@ -408,7 +408,7 @@ static int nat_q931(struct sk_buff *skb, struct nf_conn *ct,
 	struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info;
 	int dir = CTINFO2DIR(ctinfo);
 	u_int16_t nated_port = ntohs(port);
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 
 	/* Set expectations for NAT */
 	exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port;
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 175c8d1a1992..0efbf343eac8 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -226,8 +226,8 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me)
 EXPORT_SYMBOL_GPL(nf_ct_expect_alloc);
 
 void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family,
-		       union nf_conntrack_address *saddr,
-		       union nf_conntrack_address *daddr,
+		       union nf_inet_addr *saddr,
+		       union nf_inet_addr *daddr,
 		       u_int8_t proto, __be16 *src, __be16 *dst)
 {
 	int len;
diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c
index 6df259067f7e..6770baf2e845 100644
--- a/net/netfilter/nf_conntrack_ftp.c
+++ b/net/netfilter/nf_conntrack_ftp.c
@@ -358,7 +358,7 @@ static int help(struct sk_buff *skb,
 	unsigned int matchlen, matchoff;
 	struct nf_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info;
 	struct nf_conntrack_expect *exp;
-	union nf_conntrack_address *daddr;
+	union nf_inet_addr *daddr;
 	struct nf_conntrack_man cmd = {};
 	unsigned int i;
 	int found = 0, ends_in_nl;
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index b636ca60a770..872c1aa3124c 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -50,12 +50,12 @@ MODULE_PARM_DESC(callforward_filter, "only create call forwarding expectations "
 int (*set_h245_addr_hook) (struct sk_buff *skb,
 			   unsigned char **data, int dataoff,
 			   H245_TransportAddress *taddr,
-			   union nf_conntrack_address *addr, __be16 port)
+			   union nf_inet_addr *addr, __be16 port)
 			   __read_mostly;
 int (*set_h225_addr_hook) (struct sk_buff *skb,
 			   unsigned char **data, int dataoff,
 			   TransportAddress *taddr,
-			   union nf_conntrack_address *addr, __be16 port)
+			   union nf_inet_addr *addr, __be16 port)
 			   __read_mostly;
 int (*set_sig_addr_hook) (struct sk_buff *skb,
 			  struct nf_conn *ct,
@@ -214,7 +214,7 @@ static int get_tpkt_data(struct sk_buff *skb, unsigned int protoff,
 /****************************************************************************/
 static int get_h245_addr(struct nf_conn *ct, unsigned char *data,
 			 H245_TransportAddress *taddr,
-			 union nf_conntrack_address *addr, __be16 *port)
+			 union nf_inet_addr *addr, __be16 *port)
 {
 	unsigned char *p;
 	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
@@ -257,7 +257,7 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 	int ret = 0;
 	__be16 port;
 	__be16 rtp_port, rtcp_port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *rtp_exp;
 	struct nf_conntrack_expect *rtcp_exp;
 	typeof(nat_rtp_rtcp_hook) nat_rtp_rtcp;
@@ -330,7 +330,7 @@ static int expect_t120(struct sk_buff *skb,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 	typeof(nat_t120_hook) nat_t120;
 
@@ -623,7 +623,7 @@ static struct nf_conntrack_helper nf_conntrack_helper_h245 __read_mostly = {
 /****************************************************************************/
 int get_h225_addr(struct nf_conn *ct, unsigned char *data,
 		  TransportAddress *taddr,
-		  union nf_conntrack_address *addr, __be16 *port)
+		  union nf_inet_addr *addr, __be16 *port)
 {
 	unsigned char *p;
 	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
@@ -662,7 +662,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 	typeof(nat_h245_hook) nat_h245;
 
@@ -704,8 +704,8 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct,
 
 /* If the calling party is on the same side of the forward-to party,
  * we don't need to track the second call */
-static int callforward_do_filter(union nf_conntrack_address *src,
-				 union nf_conntrack_address *dst,
+static int callforward_do_filter(union nf_inet_addr *src,
+				 union nf_inet_addr *dst,
 				 int family)
 {
 	const struct nf_afinfo *afinfo;
@@ -772,7 +772,7 @@ static int expect_callforwarding(struct sk_buff *skb,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 	typeof(nat_callforwarding_hook) nat_callforwarding;
 
@@ -828,7 +828,7 @@ static int process_setup(struct sk_buff *skb, struct nf_conn *ct,
 	int ret;
 	int i;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	typeof(set_h225_addr_hook) set_h225_addr;
 
 	pr_debug("nf_ct_q931: Setup\n");
@@ -1200,7 +1200,7 @@ static unsigned char *get_udp_data(struct sk_buff *skb, unsigned int protoff,
 
 /****************************************************************************/
 static struct nf_conntrack_expect *find_expect(struct nf_conn *ct,
-					       union nf_conntrack_address *addr,
+					       union nf_inet_addr *addr,
 					       __be16 port)
 {
 	struct nf_conntrack_expect *exp;
@@ -1242,7 +1242,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct,
 	int ret = 0;
 	int i;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 	typeof(nat_q931_hook) nat_q931;
 
@@ -1311,7 +1311,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 
 	pr_debug("nf_ct_ras: GCF\n");
@@ -1471,7 +1471,7 @@ static int process_arq(struct sk_buff *skb, struct nf_conn *ct,
 	struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info;
 	int dir = CTINFO2DIR(ctinfo);
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	typeof(set_h225_addr_hook) set_h225_addr;
 
 	pr_debug("nf_ct_ras: ARQ\n");
@@ -1513,7 +1513,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 	typeof(set_sig_addr_hook) set_sig_addr;
 
@@ -1576,7 +1576,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	__be16 port;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	struct nf_conntrack_expect *exp;
 
 	pr_debug("nf_ct_ras: LCF\n");
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 515abffc4a09..47d8947cf263 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -247,7 +247,7 @@ static int skp_digits_len(struct nf_conn *ct, const char *dptr,
 }
 
 static int parse_addr(struct nf_conn *ct, const char *cp, const char **endp,
-		      union nf_conntrack_address *addr, const char *limit)
+		      union nf_inet_addr *addr, const char *limit)
 {
 	const char *end;
 	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
@@ -275,7 +275,7 @@ static int parse_addr(struct nf_conn *ct, const char *cp, const char **endp,
 static int epaddr_len(struct nf_conn *ct, const char *dptr,
 		      const char *limit, int *shift)
 {
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	const char *aux = dptr;
 
 	if (!parse_addr(ct, dptr, &dptr, &addr, limit)) {
@@ -366,7 +366,7 @@ EXPORT_SYMBOL_GPL(ct_sip_get_info);
 static int set_expected_rtp(struct sk_buff *skb,
 			    struct nf_conn *ct,
 			    enum ip_conntrack_info ctinfo,
-			    union nf_conntrack_address *addr,
+			    union nf_inet_addr *addr,
 			    __be16 port,
 			    const char *dptr)
 {
@@ -403,7 +403,7 @@ static int sip_help(struct sk_buff *skb,
 		    enum ip_conntrack_info ctinfo)
 {
 	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
-	union nf_conntrack_address addr;
+	union nf_inet_addr addr;
 	unsigned int dataoff, datalen;
 	const char *dptr;
 	int ret = NF_ACCEPT;
diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c
index 26d12b00a9c1..b7a684607c73 100644
--- a/net/netfilter/xt_connlimit.c
+++ b/net/netfilter/xt_connlimit.c
@@ -53,10 +53,10 @@ static inline unsigned int connlimit_iphash(__be32 addr)
 }
 
 static inline unsigned int
-connlimit_iphash6(const union nf_conntrack_address *addr,
-                  const union nf_conntrack_address *mask)
+connlimit_iphash6(const union nf_inet_addr *addr,
+                  const union nf_inet_addr *mask)
 {
-	union nf_conntrack_address res;
+	union nf_inet_addr res;
 	unsigned int i;
 
 	if (unlikely(!connlimit_rnd_inited)) {
@@ -81,14 +81,14 @@ static inline bool already_closed(const struct nf_conn *conn)
 }
 
 static inline unsigned int
-same_source_net(const union nf_conntrack_address *addr,
-		const union nf_conntrack_address *mask,
-		const union nf_conntrack_address *u3, unsigned int family)
+same_source_net(const union nf_inet_addr *addr,
+		const union nf_inet_addr *mask,
+		const union nf_inet_addr *u3, unsigned int family)
 {
 	if (family == AF_INET) {
 		return (addr->ip & mask->ip) == (u3->ip & mask->ip);
 	} else {
-		union nf_conntrack_address lh, rh;
+		union nf_inet_addr lh, rh;
 		unsigned int i;
 
 		for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i) {
@@ -102,8 +102,8 @@ same_source_net(const union nf_conntrack_address *addr,
 
 static int count_them(struct xt_connlimit_data *data,
 		      const struct nf_conntrack_tuple *tuple,
-		      const union nf_conntrack_address *addr,
-		      const union nf_conntrack_address *mask,
+		      const union nf_inet_addr *addr,
+		      const union nf_inet_addr *mask,
 		      const struct xt_match *match)
 {
 	struct nf_conntrack_tuple_hash *found;
@@ -185,7 +185,7 @@ connlimit_mt(const struct sk_buff *skb, const struct net_device *in,
              bool *hotdrop)
 {
 	const struct xt_connlimit_info *info = matchinfo;
-	union nf_conntrack_address addr, mask;
+	union nf_inet_addr addr, mask;
 	struct nf_conntrack_tuple tuple;
 	const struct nf_conntrack_tuple *tuple_ptr = &tuple;
 	enum ip_conntrack_info ctinfo;
-- 
cgit v1.2.3


From 22c2d8bca212a655c120fd6617328ffa3480afad Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 17 Dec 2007 22:44:47 -0800
Subject: [NETFILTER]: xt_connlimit: use the new union nf_inet_addr

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_connlimit.h | 9 +++++++--
 net/netfilter/xt_connlimit.c           | 7 +++----
 2 files changed, 10 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_connlimit.h b/include/linux/netfilter/xt_connlimit.h
index 37e933c9987d..315d2dce9da6 100644
--- a/include/linux/netfilter/xt_connlimit.h
+++ b/include/linux/netfilter/xt_connlimit.h
@@ -5,8 +5,13 @@ struct xt_connlimit_data;
 
 struct xt_connlimit_info {
 	union {
-		__be32 v4_mask;
-		__be32 v6_mask[4];
+		union nf_inet_addr mask;
+#ifndef __KERNEL__
+		union {
+			__be32 v4_mask;
+			__be32 v6_mask[4];
+		};
+#endif
 	};
 	unsigned int limit, inverse;
 
diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c
index b7a684607c73..6a9e2a357186 100644
--- a/net/netfilter/xt_connlimit.c
+++ b/net/netfilter/xt_connlimit.c
@@ -185,7 +185,7 @@ connlimit_mt(const struct sk_buff *skb, const struct net_device *in,
              bool *hotdrop)
 {
 	const struct xt_connlimit_info *info = matchinfo;
-	union nf_inet_addr addr, mask;
+	union nf_inet_addr addr;
 	struct nf_conntrack_tuple tuple;
 	const struct nf_conntrack_tuple *tuple_ptr = &tuple;
 	enum ip_conntrack_info ctinfo;
@@ -202,15 +202,14 @@ connlimit_mt(const struct sk_buff *skb, const struct net_device *in,
 	if (match->family == AF_INET6) {
 		const struct ipv6hdr *iph = ipv6_hdr(skb);
 		memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr));
-		memcpy(&mask.ip6, info->v6_mask, sizeof(info->v6_mask));
 	} else {
 		const struct iphdr *iph = ip_hdr(skb);
 		addr.ip = iph->saddr;
-		mask.ip = info->v4_mask;
 	}
 
 	spin_lock_bh(&info->data->lock);
-	connections = count_them(info->data, tuple_ptr, &addr, &mask, match);
+	connections = count_them(info->data, tuple_ptr, &addr,
+	                         &info->mask, match);
 	spin_unlock_bh(&info->data->lock);
 
 	if (connections < 0) {
-- 
cgit v1.2.3


From d647b36a69bf0a630ebf981bde3c0651e2779e5e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 20 Dec 2007 04:13:21 -0800
Subject: [SNMP]: Fix SNMP counters with PREEMPT

The SNMP macros use raw_smp_processor_id() in process context
which is illegal because the process may be preempted and then
migrated to another CPU.

This patch makes it use get_cpu/put_cpu to disable preemption.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/snmp.h | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/snmp.h b/include/net/snmp.h
index 9c5793db637f..fbb66663a42c 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -23,6 +23,7 @@
 
 #include <linux/cache.h>
 #include <linux/snmp.h>
+#include <linux/smp.h>
 
 /*
  * Mibs are stored in array of unsigned long.
@@ -135,14 +136,26 @@ struct linux_mib {
 #define SNMP_INC_STATS_BH(mib, field) 	\
 	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field]++)
 #define SNMP_INC_STATS_USER(mib, field) \
-	(per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field]++)
+	do { \
+		per_cpu_ptr(mib[1], get_cpu())->mibs[field]++; \
+		put_cpu(); \
+	} while (0)
 #define SNMP_INC_STATS(mib, field) 	\
-	(per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]++)
+	do { \
+		per_cpu_ptr(mib[!in_softirq()], get_cpu())->mibs[field]++; \
+		put_cpu(); \
+	} while (0)
 #define SNMP_DEC_STATS(mib, field) 	\
-	(per_cpu_ptr(mib[!in_softirq()], raw_smp_processor_id())->mibs[field]--)
+	do { \
+		per_cpu_ptr(mib[!in_softirq()], get_cpu())->mibs[field]--; \
+		put_cpu(); \
+	} while (0)
 #define SNMP_ADD_STATS_BH(mib, field, addend) 	\
 	(per_cpu_ptr(mib[0], raw_smp_processor_id())->mibs[field] += addend)
 #define SNMP_ADD_STATS_USER(mib, field, addend) 	\
-	(per_cpu_ptr(mib[1], raw_smp_processor_id())->mibs[field] += addend)
+	do { \
+		per_cpu_ptr(mib[1], get_cpu())->mibs[field] += addend; \
+		put_cpu(); \
+	} while (0)
 
 #endif
-- 
cgit v1.2.3


From 42e30bf3463cd37d73839376662cb79b4d5c416c Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 20 Dec 2007 14:08:56 -0800
Subject: [SCTP]: Handle the wildcard ADD-IP Address parameter

The Address Parameter in the parameter list of the ASCONF chunk
may be a wildcard address.  In this case special processing
is required.  For the 'add' case, the source IP of the packet is
added.  In the 'del' case, all addresses except the source IP
of packet are removed. In the "mark primary" case, the source
address is marked as primary.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  2 ++
 net/sctp/associola.c       | 17 +++++++++++++++++
 net/sctp/sm_make_chunk.c   | 40 ++++++++++++++++++++++++++++++++++++----
 3 files changed, 55 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index bb965742b64e..96621db22466 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1938,6 +1938,8 @@ void sctp_assoc_rwnd_increase(struct sctp_association *, unsigned);
 void sctp_assoc_rwnd_decrease(struct sctp_association *, unsigned);
 void sctp_assoc_set_primary(struct sctp_association *,
 			    struct sctp_transport *);
+void sctp_assoc_del_nonprimary_peers(struct sctp_association *,
+				    struct sctp_transport *);
 int sctp_assoc_set_bind_addr_from_ep(struct sctp_association *,
 				     gfp_t);
 int sctp_assoc_set_bind_addr_from_cookie(struct sctp_association *,
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 33ae9b01131e..61bebb9b96e6 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -730,6 +730,23 @@ struct sctp_transport *sctp_assoc_lookup_paddr(
 	return NULL;
 }
 
+/* Remove all transports except a give one */
+void sctp_assoc_del_nonprimary_peers(struct sctp_association *asoc,
+				     struct sctp_transport *primary)
+{
+	struct sctp_transport	*temp;
+	struct sctp_transport	*t;
+
+	list_for_each_entry_safe(t, temp, &asoc->peer.transport_addr_list,
+				 transports) {
+		/* if the current transport is not the primary one, delete it */
+		if (t != primary)
+			sctp_assoc_rm_peer(asoc, t);
+	}
+
+	return;
+}
+
 /* Engage in transport control operations.
  * Mark the transport up or down and send a notification to the user.
  * Select and update the new active and retran paths.
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 3cc629d3c9ff..64790b53323b 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2727,7 +2727,6 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc,
 	struct sctp_transport *peer;
 	struct sctp_af *af;
 	union sctp_addr	addr;
-	struct list_head *pos;
 	union sctp_addr_param *addr_param;
 
 	addr_param = (union sctp_addr_param *)
@@ -2738,8 +2737,24 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc,
 		return SCTP_ERROR_INV_PARAM;
 
 	af->from_addr_param(&addr, addr_param, htons(asoc->peer.port), 0);
+
+	/* ADDIP 4.2.1  This parameter MUST NOT contain a broadcast
+	 * or multicast address.
+	 * (note: wildcard is permitted and requires special handling so
+	 *  make sure we check for that)
+	 */
+	if (!af->is_any(&addr) && !af->addr_valid(&addr, NULL, asconf->skb))
+		return SCTP_ERROR_INV_PARAM;
+
 	switch (asconf_param->param_hdr.type) {
 	case SCTP_PARAM_ADD_IP:
+		/* Section 4.2.1:
+		 * If the address 0.0.0.0 or ::0 is provided, the source
+		 * address of the packet MUST be added.
+		 */
+		if (af->is_any(&addr))
+			memcpy(&addr, &asconf->source, sizeof(addr));
+
 		/* ADDIP 4.3 D9) If an endpoint receives an ADD IP address
 		 * request and does not have the local resources to add this
 		 * new address to the association, it MUST return an Error
@@ -2761,8 +2776,7 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc,
 		 * MUST send an Error Cause TLV with the error cause set to the
 		 * new error code 'Request to Delete Last Remaining IP Address'.
 		 */
-		pos = asoc->peer.transport_addr_list.next;
-		if (pos->next == &asoc->peer.transport_addr_list)
+		if (asoc->peer.transport_count == 1)
 			return SCTP_ERROR_DEL_LAST_IP;
 
 		/* ADDIP 4.3 D8) If a request is received to delete an IP
@@ -2775,9 +2789,27 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc,
 		if (sctp_cmp_addr_exact(sctp_source(asconf), &addr))
 			return SCTP_ERROR_DEL_SRC_IP;
 
-		sctp_assoc_del_peer(asoc, &addr);
+		/* Section 4.2.2
+		 * If the address 0.0.0.0 or ::0 is provided, all
+		 * addresses of the peer except	the source address of the
+		 * packet MUST be deleted.
+		 */
+		if (af->is_any(&addr)) {
+			sctp_assoc_set_primary(asoc, asconf->transport);
+			sctp_assoc_del_nonprimary_peers(asoc,
+							asconf->transport);
+		} else
+			sctp_assoc_del_peer(asoc, &addr);
 		break;
 	case SCTP_PARAM_SET_PRIMARY:
+		/* ADDIP Section 4.2.4
+		 * If the address 0.0.0.0 or ::0 is provided, the receiver
+		 * MAY mark the source address of the packet as its
+		 * primary.
+		 */
+		if (af->is_any(&addr))
+			memcpy(&addr.v4, sctp_source(asconf), sizeof(addr));
+
 		peer = sctp_assoc_lookup_paddr(asoc, &addr);
 		if (!peer)
 			return SCTP_ERROR_INV_PARAM;
-- 
cgit v1.2.3


From d6de3097592b7ae7f8e233a4dafb088e2aa8170f Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 20 Dec 2007 14:10:00 -0800
Subject: [SCTP]: Add the handling of "Set Primary IP Address" parameter to
 INIT

The ADD-IP "Set Primary IP Address" parameter is allowed in the
INIT/INIT-ACK exchange.  Allow processing of this parameter during
the INIT/INIT-ACK.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  1 +
 net/sctp/sm_make_chunk.c   | 27 +++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 96621db22466..8a0808eab811 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -451,6 +451,7 @@ union sctp_params {
 	struct sctp_random_param *random;
 	struct sctp_chunks_param *chunks;
 	struct sctp_hmac_algo_param *hmac_algo;
+	struct sctp_addip_param *addip;
 };
 
 /* RFC 2960.  Section 3.3.5 Heartbeat.
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 64790b53323b..8138bbd93210 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1969,6 +1969,11 @@ static sctp_ierror_t sctp_verify_param(const struct sctp_association *asoc,
 	case SCTP_PARAM_SUPPORTED_EXT:
 		break;
 
+	case SCTP_PARAM_SET_PRIMARY:
+		if (sctp_addip_enable)
+			break;
+		goto fallthrough;
+
 	case SCTP_PARAM_HOST_NAME_ADDRESS:
 		/* Tell the peer, we won't support this param.  */
 		sctp_process_hn_param(asoc, param, chunk, err_chunk);
@@ -2286,6 +2291,8 @@ static int sctp_process_param(struct sctp_association *asoc,
 	sctp_scope_t scope;
 	time_t stale;
 	struct sctp_af *af;
+	union sctp_addr_param *addr_param;
+	struct sctp_transport *t;
 
 	/* We maintain all INIT parameters in network byte order all the
 	 * time.  This allows us to not worry about whether the parameters
@@ -2376,6 +2383,26 @@ static int sctp_process_param(struct sctp_association *asoc,
 		asoc->peer.adaptation_ind = param.aind->adaptation_ind;
 		break;
 
+	case SCTP_PARAM_SET_PRIMARY:
+		addr_param = param.v + sizeof(sctp_addip_param_t);
+
+		af = sctp_get_af_specific(param_type2af(param.p->type));
+		af->from_addr_param(&addr, addr_param,
+				    htons(asoc->peer.port), 0);
+
+		/* if the address is invalid, we can't process it.
+		 * XXX: see spec for what to do.
+		 */
+		if (!af->addr_valid(&addr, NULL, NULL))
+			break;
+
+		t = sctp_assoc_lookup_paddr(asoc, &addr);
+		if (!t)
+			break;
+
+		sctp_assoc_set_primary(asoc, t);
+		break;
+
 	case SCTP_PARAM_SUPPORTED_EXT:
 		sctp_process_ext_param(asoc, param);
 		break;
-- 
cgit v1.2.3


From a08de64d074b36a56ee3bb985cd171281db78e96 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 20 Dec 2007 14:11:47 -0800
Subject: [SCTP]: Update ASCONF processing to conform to spec.

The processing of the ASCONF chunks has changed a lot in the
spec.  New items are:
    1. A list of ASCONF-ACK chunks is now cached
    2. The source of the packet is used in response.
    3. New handling for unexpect ASCONF chunks.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h | 24 +++++++++--------
 net/sctp/associola.c       | 58 +++++++++++++++++++++++++++++++++++++++--
 net/sctp/outqueue.c        | 29 +++++++++++++++++++--
 net/sctp/sm_make_chunk.c   | 12 ++++-----
 net/sctp/sm_statefuns.c    | 64 ++++++++++++++++++++++++++++++----------------
 5 files changed, 143 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 8a0808eab811..fa87873fee76 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -744,6 +744,7 @@ struct sctp_chunk {
 	__u8 tsn_missing_report; /* Data chunk missing counter. */
 	__u8 data_accepted; 	/* At least 1 chunk in this packet accepted */
 	__u8 auth;		/* IN: was auth'ed | OUT: needs auth */
+	__u8 has_asconf;	/* IN: have seen an asconf before */
 };
 
 void sctp_chunk_hold(struct sctp_chunk *);
@@ -1785,20 +1786,16 @@ struct sctp_association {
 	 */
 	struct sctp_chunk *addip_last_asconf;
 
-	/* ADDIP Section 4.2 Upon reception of an ASCONF Chunk.
+	/* ADDIP Section 5.2 Upon reception of an ASCONF Chunk.
 	 *
-	 * IMPLEMENTATION NOTE: As an optimization a receiver may wish
-	 * to save the last ASCONF-ACK for some predetermined period
-	 * of time and instead of re-processing the ASCONF (with the
-	 * same serial number) it may just re-transmit the
-	 * ASCONF-ACK. It may wish to use the arrival of a new serial
-	 * number to discard the previously saved ASCONF-ACK or any
-	 * other means it may choose to expire the saved ASCONF-ACK.
+	 * This is needed to implement itmes E1 - E4 of the updated
+	 * spec.  Here is the justification:
 	 *
-	 * [This is our saved ASCONF-ACK.  We invalidate it when a new
-	 * ASCONF serial number arrives.]
+	 * Since the peer may bundle multiple ASCONF chunks toward us,
+	 * we now need the ability to cache multiple ACKs.  The section
+	 * describes in detail how they are cached and cleaned up.
 	 */
-	struct sctp_chunk *addip_last_asconf_ack;
+	struct list_head asconf_ack_list;
 
 	/* These ASCONF chunks are waiting to be sent.
 	 *
@@ -1947,6 +1944,11 @@ int sctp_assoc_set_bind_addr_from_cookie(struct sctp_association *,
 					 struct sctp_cookie*,
 					 gfp_t gfp);
 int sctp_assoc_set_id(struct sctp_association *, gfp_t);
+void sctp_assoc_clean_asconf_ack_cache(const struct sctp_association *asoc);
+struct sctp_chunk *sctp_assoc_lookup_asconf_ack(
+					const struct sctp_association *asoc,
+					__be32 serial);
+
 
 int sctp_cmp_addr_exact(const union sctp_addr *ss1,
 			const union sctp_addr *ss2);
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 61bebb9b96e6..a016e78061f4 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -61,6 +61,7 @@
 
 /* Forward declarations for internal functions. */
 static void sctp_assoc_bh_rcv(struct work_struct *work);
+static void sctp_assoc_free_asconf_acks(struct sctp_association *asoc);
 
 
 /* 1st Level Abstractions. */
@@ -242,6 +243,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
 	asoc->addip_serial = asoc->c.initial_tsn;
 
 	INIT_LIST_HEAD(&asoc->addip_chunk_list);
+	INIT_LIST_HEAD(&asoc->asconf_ack_list);
 
 	/* Make an empty list of remote transport addresses.  */
 	INIT_LIST_HEAD(&asoc->peer.transport_addr_list);
@@ -431,8 +433,7 @@ void sctp_association_free(struct sctp_association *asoc)
 	asoc->peer.transport_count = 0;
 
 	/* Free any cached ASCONF_ACK chunk. */
-	if (asoc->addip_last_asconf_ack)
-		sctp_chunk_free(asoc->addip_last_asconf_ack);
+	sctp_assoc_free_asconf_acks(asoc);
 
 	/* Free any cached ASCONF chunk. */
 	if (asoc->addip_last_asconf)
@@ -1485,3 +1486,56 @@ retry:
 	asoc->assoc_id = (sctp_assoc_t) assoc_id;
 	return error;
 }
+
+/* Free asconf_ack cache */
+static void sctp_assoc_free_asconf_acks(struct sctp_association *asoc)
+{
+	struct sctp_chunk *ack;
+	struct sctp_chunk *tmp;
+
+	list_for_each_entry_safe(ack, tmp, &asoc->asconf_ack_list,
+				transmitted_list) {
+		list_del_init(&ack->transmitted_list);
+		sctp_chunk_free(ack);
+	}
+}
+
+/* Clean up the ASCONF_ACK queue */
+void sctp_assoc_clean_asconf_ack_cache(const struct sctp_association *asoc)
+{
+	struct sctp_chunk *ack;
+	struct sctp_chunk *tmp;
+
+	/* We can remove all the entries from the queue upto
+	 * the "Peer-Sequence-Number".
+	 */
+	list_for_each_entry_safe(ack, tmp, &asoc->asconf_ack_list,
+				transmitted_list) {
+		if (ack->subh.addip_hdr->serial ==
+				htonl(asoc->peer.addip_serial))
+			break;
+
+		list_del_init(&ack->transmitted_list);
+		sctp_chunk_free(ack);
+	}
+}
+
+/* Find the ASCONF_ACK whose serial number matches ASCONF */
+struct sctp_chunk *sctp_assoc_lookup_asconf_ack(
+					const struct sctp_association *asoc,
+					__be32 serial)
+{
+	struct sctp_chunk *ack = NULL;
+
+	/* Walk through the list of cached ASCONF-ACKs and find the
+	 * ack chunk whose serial number matches that of the request.
+	 */
+	list_for_each_entry(ack, &asoc->asconf_ack_list, transmitted_list) {
+		if (ack->subh.addip_hdr->serial == serial) {
+			sctp_chunk_hold(ack);
+			break;
+		}
+	}
+
+	return ack;
+}
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index fa76f235169b..a42af865c2ef 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -716,7 +716,29 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 		new_transport = chunk->transport;
 
 		if (!new_transport) {
-			new_transport = asoc->peer.active_path;
+			/*
+			 * If we have a prior transport pointer, see if
+			 * the destination address of the chunk
+			 * matches the destination address of the
+			 * current transport.  If not a match, then
+			 * try to look up the transport with a given
+			 * destination address.  We do this because
+			 * after processing ASCONFs, we may have new
+			 * transports created.
+			 */
+			if (transport &&
+			    sctp_cmp_addr_exact(&chunk->dest,
+						&transport->ipaddr))
+					new_transport = transport;
+			else
+				new_transport = sctp_assoc_lookup_paddr(asoc,
+								&chunk->dest);
+
+			/* if we still don't have a new transport, then
+			 * use the current active path.
+			 */
+			if (!new_transport)
+				new_transport = asoc->peer.active_path;
 		} else if ((new_transport->state == SCTP_INACTIVE) ||
 			   (new_transport->state == SCTP_UNCONFIRMED)) {
 			/* If the chunk is Heartbeat or Heartbeat Ack,
@@ -729,9 +751,12 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 			 * address of the IP datagram containing the
 			 * HEARTBEAT chunk to which this ack is responding.
 			 * ...
+			 *
+			 * ASCONF_ACKs also must be sent to the source.
 			 */
 			if (chunk->chunk_hdr->type != SCTP_CID_HEARTBEAT &&
-			    chunk->chunk_hdr->type != SCTP_CID_HEARTBEAT_ACK)
+			    chunk->chunk_hdr->type != SCTP_CID_HEARTBEAT_ACK &&
+			    chunk->chunk_hdr->type != SCTP_CID_ASCONF_ACK)
 				new_transport = asoc->peer.active_path;
 		}
 
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 8138bbd93210..7fd6a6b68614 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1275,6 +1275,9 @@ nodata:
 /* Release the memory occupied by a chunk.  */
 static void sctp_chunk_destroy(struct sctp_chunk *chunk)
 {
+	BUG_ON(!list_empty(&chunk->list));
+	list_del_init(&chunk->transmitted_list);
+
 	/* Free the chunk skb data and the SCTP_chunk stub itself. */
 	dev_kfree_skb(chunk->skb);
 
@@ -1285,9 +1288,6 @@ static void sctp_chunk_destroy(struct sctp_chunk *chunk)
 /* Possibly, free the chunk.  */
 void sctp_chunk_free(struct sctp_chunk *chunk)
 {
-	BUG_ON(!list_empty(&chunk->list));
-	list_del_init(&chunk->transmitted_list);
-
 	/* Release our reference on the message tracker. */
 	if (chunk->msg)
 		sctp_datamsg_put(chunk->msg);
@@ -2980,11 +2980,9 @@ done:
 	 * after freeing the reference to old asconf ack if any.
 	 */
 	if (asconf_ack) {
-		if (asoc->addip_last_asconf_ack)
-			sctp_chunk_free(asoc->addip_last_asconf_ack);
-
 		sctp_chunk_hold(asconf_ack);
-		asoc->addip_last_asconf_ack = asconf_ack;
+		list_add_tail(&asconf_ack->transmitted_list,
+			      &asoc->asconf_ack_list);
 	}
 
 	return asconf_ack;
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index b6aaa7e97d82..a1be9d93f1a8 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -3402,48 +3402,68 @@ sctp_disposition_t sctp_sf_do_asconf(const struct sctp_endpoint *ep,
 
 	/* Verify the ASCONF chunk before processing it. */
 	if (!sctp_verify_asconf(asoc,
-	    (sctp_paramhdr_t *)((void *)addr_param + length),
-	    (void *)chunk->chunk_end,
-	    &err_param))
+			    (sctp_paramhdr_t *)((void *)addr_param + length),
+			    (void *)chunk->chunk_end,
+			    &err_param))
 		return sctp_sf_violation_paramlen(ep, asoc, type,
-			   (void *)&err_param, commands);
+						  (void *)&err_param, commands);
 
-	/* ADDIP 4.2 C1) Compare the value of the serial number to the value
+	/* ADDIP 5.2 E1) Compare the value of the serial number to the value
 	 * the endpoint stored in a new association variable
 	 * 'Peer-Serial-Number'.
 	 */
 	if (serial == asoc->peer.addip_serial + 1) {
-		/* ADDIP 4.2 C2) If the value found in the serial number is
-		 * equal to the ('Peer-Serial-Number' + 1), the endpoint MUST
-		 * do V1-V5.
+		/* If this is the first instance of ASCONF in the packet,
+		 * we can clean our old ASCONF-ACKs.
+		 */
+		if (!chunk->has_asconf)
+			sctp_assoc_clean_asconf_ack_cache(asoc);
+
+		/* ADDIP 5.2 E4) When the Sequence Number matches the next one
+		 * expected, process the ASCONF as described below and after
+		 * processing the ASCONF Chunk, append an ASCONF-ACK Chunk to
+		 * the response packet and cache a copy of it (in the event it
+		 * later needs to be retransmitted).
+		 *
+		 * Essentially, do V1-V5.
 		 */
 		asconf_ack = sctp_process_asconf((struct sctp_association *)
 						 asoc, chunk);
 		if (!asconf_ack)
 			return SCTP_DISPOSITION_NOMEM;
-	} else if (serial == asoc->peer.addip_serial) {
-		/* ADDIP 4.2 C3) If the value found in the serial number is
-		 * equal to the value stored in the 'Peer-Serial-Number'
-		 * IMPLEMENTATION NOTE: As an optimization a receiver may wish
-		 * to save the last ASCONF-ACK for some predetermined period of
-		 * time and instead of re-processing the ASCONF (with the same
-		 * serial number) it may just re-transmit the ASCONF-ACK.
+	} else if (serial < asoc->peer.addip_serial + 1) {
+		/* ADDIP 5.2 E2)
+		 * If the value found in the Sequence Number is less than the
+		 * ('Peer- Sequence-Number' + 1), simply skip to the next
+		 * ASCONF, and include in the outbound response packet
+		 * any previously cached ASCONF-ACK response that was
+		 * sent and saved that matches the Sequence Number of the
+		 * ASCONF.  Note: It is possible that no cached ASCONF-ACK
+		 * Chunk exists.  This will occur when an older ASCONF
+		 * arrives out of order.  In such a case, the receiver
+		 * should skip the ASCONF Chunk and not include ASCONF-ACK
+		 * Chunk for that chunk.
 		 */
-		if (asoc->addip_last_asconf_ack)
-			asconf_ack = asoc->addip_last_asconf_ack;
-		else
+		asconf_ack = sctp_assoc_lookup_asconf_ack(asoc, hdr->serial);
+		if (!asconf_ack)
 			return SCTP_DISPOSITION_DISCARD;
 	} else {
-		/* ADDIP 4.2 C4) Otherwise, the ASCONF Chunk is discarded since
+		/* ADDIP 5.2 E5) Otherwise, the ASCONF Chunk is discarded since
 		 * it must be either a stale packet or from an attacker.
 		 */
 		return SCTP_DISPOSITION_DISCARD;
 	}
 
-	/* ADDIP 4.2 C5) In both cases C2 and C3 the ASCONF-ACK MUST be sent
-	 * back to the source address contained in the IP header of the ASCONF
-	 * being responded to.
+	/* ADDIP 5.2 E6)  The destination address of the SCTP packet
+	 * containing the ASCONF-ACK Chunks MUST be the source address of
+	 * the SCTP packet that held the ASCONF Chunks.
+	 *
+	 * To do this properly, we'll set the destination address of the chunk
+	 * and at the transmit time, will try look up the transport to use.
+	 * Since ASCONFs may be bundled, the correct transport may not be
+	 * created untill we process the entire packet, thus this workaround.
 	 */
+	asconf_ack->dest = chunk->source;
 	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(asconf_ack));
 
 	return SCTP_DISPOSITION_CONSUME;
-- 
cgit v1.2.3


From f57d96b2e92d209ab3991bba9a44e0d6ef7614a8 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 20 Dec 2007 14:12:24 -0800
Subject: [SCTP]: Change use_as_src into a full address state

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h | 10 ++++++++--
 net/sctp/bind_addr.c       |  9 +++++----
 net/sctp/ipv6.c            |  2 +-
 net/sctp/protocol.c        |  8 ++++----
 net/sctp/sm_make_chunk.c   |  6 +++---
 net/sctp/socket.c          |  8 ++++----
 6 files changed, 25 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index fa87873fee76..2528f8a837d9 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -760,12 +760,18 @@ void sctp_init_addrs(struct sctp_chunk *, union sctp_addr *,
 		     union sctp_addr *);
 const union sctp_addr *sctp_source(const struct sctp_chunk *chunk);
 
+enum {
+	SCTP_ADDR_NEW,		/* new address added to assoc/ep */
+	SCTP_ADDR_SRC,		/* address can be used as source */
+	SCTP_ADDR_DEL,		/* address about to be deleted */
+};
+
 /* This is a structure for holding either an IPv6 or an IPv4 address.  */
 struct sctp_sockaddr_entry {
 	struct list_head list;
 	struct rcu_head	rcu;
 	union sctp_addr a;
-	__u8 use_as_src;
+	__u8 state;
 	__u8 valid;
 };
 
@@ -1190,7 +1196,7 @@ int sctp_bind_addr_dup(struct sctp_bind_addr *dest,
 			const struct sctp_bind_addr *src,
 			gfp_t gfp);
 int sctp_add_bind_addr(struct sctp_bind_addr *, union sctp_addr *,
-		       __u8 use_as_src, gfp_t gfp);
+		       __u8 addr_state, gfp_t gfp);
 int sctp_del_bind_addr(struct sctp_bind_addr *, union sctp_addr *);
 int sctp_bind_addr_match(struct sctp_bind_addr *, const union sctp_addr *,
 			 struct sctp_sock *);
diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c
index 6a7d01091f0c..432661174789 100644
--- a/net/sctp/bind_addr.c
+++ b/net/sctp/bind_addr.c
@@ -171,7 +171,7 @@ void sctp_bind_addr_free(struct sctp_bind_addr *bp)
 
 /* Add an address to the bind address list in the SCTP_bind_addr structure. */
 int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new,
-		       __u8 use_as_src, gfp_t gfp)
+		       __u8 addr_state, gfp_t gfp)
 {
 	struct sctp_sockaddr_entry *addr;
 
@@ -188,7 +188,7 @@ int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new,
 	if (!addr->a.v4.sin_port)
 		addr->a.v4.sin_port = htons(bp->port);
 
-	addr->use_as_src = use_as_src;
+	addr->state = addr_state;
 	addr->valid = 1;
 
 	INIT_LIST_HEAD(&addr->list);
@@ -312,7 +312,7 @@ int sctp_raw_to_bind_addrs(struct sctp_bind_addr *bp, __u8 *raw_addr_list,
 		}
 
 		af->from_addr_param(&addr, rawaddr, htons(port), 0);
-		retval = sctp_add_bind_addr(bp, &addr, 1, gfp);
+		retval = sctp_add_bind_addr(bp, &addr, SCTP_ADDR_SRC, gfp);
 		if (retval) {
 			/* Can't finish building the list, clean up. */
 			sctp_bind_addr_clean(bp);
@@ -411,7 +411,8 @@ static int sctp_copy_one_addr(struct sctp_bind_addr *dest,
 		    (((AF_INET6 == addr->sa.sa_family) &&
 		      (flags & SCTP_ADDR6_ALLOWED) &&
 		      (flags & SCTP_ADDR6_PEERSUPP))))
-			error = sctp_add_bind_addr(dest, addr, 1, gfp);
+			error = sctp_add_bind_addr(dest, addr, SCTP_ADDR_SRC,
+						    gfp);
 	}
 
 	return error;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 7f31ff638bc6..bd04aed673cb 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -330,7 +330,7 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc,
 	list_for_each_entry_rcu(laddr, &bp->address_list, list) {
 		if (!laddr->valid)
 			continue;
-		if ((laddr->use_as_src) &&
+		if ((laddr->state == SCTP_ADDR_SRC) &&
 		    (laddr->a.sa.sa_family == AF_INET6) &&
 		    (scope <= sctp_scope(&laddr->a))) {
 			bmatchlen = sctp_v6_addr_match_len(daddr, &laddr->a);
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index dc22d7108494..e466e00b9a9f 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -229,8 +229,8 @@ int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope,
 			    (((AF_INET6 == addr->a.sa.sa_family) &&
 			      (copy_flags & SCTP_ADDR6_ALLOWED) &&
 			      (copy_flags & SCTP_ADDR6_PEERSUPP)))) {
-				error = sctp_add_bind_addr(bp, &addr->a, 1,
-						    GFP_ATOMIC);
+				error = sctp_add_bind_addr(bp, &addr->a,
+						    SCTP_ADDR_SRC, GFP_ATOMIC);
 				if (error)
 					goto end_copy;
 			}
@@ -472,7 +472,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc,
 		 */
 		rcu_read_lock();
 		list_for_each_entry_rcu(laddr, &bp->address_list, list) {
-			if (!laddr->valid || !laddr->use_as_src)
+			if (!laddr->valid || (laddr->state != SCTP_ADDR_SRC))
 				continue;
 			sctp_v4_dst_saddr(&dst_saddr, dst, htons(bp->port));
 			if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a))
@@ -494,7 +494,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc,
 	list_for_each_entry_rcu(laddr, &bp->address_list, list) {
 		if (!laddr->valid)
 			continue;
-		if ((laddr->use_as_src) &&
+		if ((laddr->state == SCTP_ADDR_SRC) &&
 		    (AF_INET == laddr->a.sa.sa_family)) {
 			fl.fl4_src = laddr->a.v4.sin_addr.s_addr;
 			if (!ip_route_output_key(&rt, &fl)) {
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 7fd6a6b68614..46f54188f00a 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1692,8 +1692,8 @@ no_hmac:
 
 	/* Also, add the destination address. */
 	if (list_empty(&retval->base.bind_addr.address_list)) {
-		sctp_add_bind_addr(&retval->base.bind_addr, &chunk->dest, 1,
-				GFP_ATOMIC);
+		sctp_add_bind_addr(&retval->base.bind_addr, &chunk->dest,
+				SCTP_ADDR_SRC, GFP_ATOMIC);
 	}
 
 	retval->next_tsn = retval->c.initial_tsn;
@@ -3016,7 +3016,7 @@ static int sctp_asconf_param_success(struct sctp_association *asoc,
 		local_bh_disable();
 		list_for_each_entry(saddr, &bp->address_list, list) {
 			if (sctp_cmp_addr_exact(&saddr->a, &addr))
-				saddr->use_as_src = 1;
+				saddr->state = SCTP_ADDR_SRC;
 		}
 		local_bh_enable();
 		break;
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index dc2f9221f092..7a8650f01d08 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -390,7 +390,7 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len)
 	/* Add the address to the bind address list.
 	 * Use GFP_ATOMIC since BHs will be disabled.
 	 */
-	ret = sctp_add_bind_addr(bp, addr, 1, GFP_ATOMIC);
+	ret = sctp_add_bind_addr(bp, addr, SCTP_ADDR_SRC, GFP_ATOMIC);
 
 	/* Copy back into socket for getsockname() use. */
 	if (!ret) {
@@ -585,8 +585,8 @@ static int sctp_send_asconf_add_ip(struct sock		*sk,
 			addr = (union sctp_addr *)addr_buf;
 			af = sctp_get_af_specific(addr->v4.sin_family);
 			memcpy(&saveaddr, addr, af->sockaddr_len);
-			retval = sctp_add_bind_addr(bp, &saveaddr, 0,
-						    GFP_ATOMIC);
+			retval = sctp_add_bind_addr(bp, &saveaddr,
+						    SCTP_ADDR_NEW, GFP_ATOMIC);
 			addr_buf += af->sockaddr_len;
 		}
 	}
@@ -777,7 +777,7 @@ static int sctp_send_asconf_del_ip(struct sock		*sk,
 			af = sctp_get_af_specific(laddr->v4.sin_family);
 			list_for_each_entry(saddr, &bp->address_list, list) {
 				if (sctp_cmp_addr_exact(&saddr->a, laddr))
-					saddr->use_as_src = 0;
+					saddr->state = SCTP_ADDR_DEL;
 			}
 			addr_buf += af->sockaddr_len;
 		}
-- 
cgit v1.2.3


From 75205f478331cc64ce729ea72d3c8c1837fb59cb Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 20 Dec 2007 14:12:59 -0800
Subject: [SCTP]: Implement ADD-IP special case processing for ABORT chunk

ADD-IP spec has a special case for processing ABORTs:
    F4) ... One special consideration is that ABORT
        Chunks arriving destined to the IP address being deleted MUST be
        ignored (see Section 5.3.1 for further details).

Check if the address we received on is in the DEL state, and if
so, ignore the ABORT.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  2 ++
 net/sctp/bind_addr.c       | 26 +++++++++++++++++++++++
 net/sctp/sm_statefuns.c    | 52 ++++++++++++++++++++++++++++++++++++++++++----
 3 files changed, 76 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 2528f8a837d9..4d591bfce452 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1200,6 +1200,8 @@ int sctp_add_bind_addr(struct sctp_bind_addr *, union sctp_addr *,
 int sctp_del_bind_addr(struct sctp_bind_addr *, union sctp_addr *);
 int sctp_bind_addr_match(struct sctp_bind_addr *, const union sctp_addr *,
 			 struct sctp_sock *);
+int sctp_bind_addr_state(const struct sctp_bind_addr *bp,
+			 const union sctp_addr *addr);
 union sctp_addr *sctp_find_unmatch_addr(struct sctp_bind_addr	*bp,
 					const union sctp_addr	*addrs,
 					int			addrcnt,
diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c
index 432661174789..13fbfb449a55 100644
--- a/net/sctp/bind_addr.c
+++ b/net/sctp/bind_addr.c
@@ -353,6 +353,32 @@ int sctp_bind_addr_match(struct sctp_bind_addr *bp,
 	return match;
 }
 
+/* Get the state of the entry in the bind_addr_list */
+int sctp_bind_addr_state(const struct sctp_bind_addr *bp,
+			 const union sctp_addr *addr)
+{
+	struct sctp_sockaddr_entry *laddr;
+	struct sctp_af *af;
+	int state = -1;
+
+	af = sctp_get_af_specific(addr->sa.sa_family);
+	if (unlikely(!af))
+		return state;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(laddr, &bp->address_list, list) {
+		if (!laddr->valid)
+			continue;
+		if (af->cmp_addr(&laddr->a, addr)) {
+			state = laddr->state;
+			break;
+		}
+	}
+	rcu_read_unlock();
+
+	return state;
+}
+
 /* Find the first address in the bind address list that is not present in
  * the addrs packed array.
  */
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index a1be9d93f1a8..0c9f37eb7d8d 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -143,6 +143,12 @@ static sctp_ierror_t sctp_sf_authenticate(const struct sctp_endpoint *ep,
 				    const sctp_subtype_t type,
 				    struct sctp_chunk *chunk);
 
+static sctp_disposition_t __sctp_sf_do_9_1_abort(const struct sctp_endpoint *ep,
+					const struct sctp_association *asoc,
+					const sctp_subtype_t type,
+					void *arg,
+					sctp_cmd_seq_t *commands);
+
 /* Small helper function that checks if the chunk length
  * is of the appropriate length.  The 'required_length' argument
  * is set to be the size of a specific chunk we are testing.
@@ -2073,11 +2079,20 @@ sctp_disposition_t sctp_sf_shutdown_pending_abort(
 	if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t)))
 		return sctp_sf_pdiscard(ep, asoc, type, arg, commands);
 
+	/* ADD-IP: Special case for ABORT chunks
+	 * F4)  One special consideration is that ABORT Chunks arriving
+	 * destined to the IP address being deleted MUST be
+	 * ignored (see Section 5.3.1 for further details).
+	 */
+	if (SCTP_ADDR_DEL ==
+		    sctp_bind_addr_state(&asoc->base.bind_addr, &chunk->dest))
+		return sctp_sf_discard_chunk(ep, asoc, type, arg, commands);
+
 	/* Stop the T5-shutdown guard timer.  */
 	sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP,
 			SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD));
 
-	return sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands);
+	return __sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands);
 }
 
 /*
@@ -2109,6 +2124,15 @@ sctp_disposition_t sctp_sf_shutdown_sent_abort(const struct sctp_endpoint *ep,
 	if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t)))
 		return sctp_sf_pdiscard(ep, asoc, type, arg, commands);
 
+	/* ADD-IP: Special case for ABORT chunks
+	 * F4)  One special consideration is that ABORT Chunks arriving
+	 * destined to the IP address being deleted MUST be
+	 * ignored (see Section 5.3.1 for further details).
+	 */
+	if (SCTP_ADDR_DEL ==
+		    sctp_bind_addr_state(&asoc->base.bind_addr, &chunk->dest))
+		return sctp_sf_discard_chunk(ep, asoc, type, arg, commands);
+
 	/* Stop the T2-shutdown timer. */
 	sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP,
 			SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN));
@@ -2117,7 +2141,7 @@ sctp_disposition_t sctp_sf_shutdown_sent_abort(const struct sctp_endpoint *ep,
 	sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP,
 			SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD));
 
-	return sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands);
+	return __sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands);
 }
 
 /*
@@ -2344,8 +2368,6 @@ sctp_disposition_t sctp_sf_do_9_1_abort(const struct sctp_endpoint *ep,
 					sctp_cmd_seq_t *commands)
 {
 	struct sctp_chunk *chunk = arg;
-	unsigned len;
-	__be16 error = SCTP_ERROR_NO_ERROR;
 
 	if (!sctp_vtag_verify_either(chunk, asoc))
 		return sctp_sf_pdiscard(ep, asoc, type, arg, commands);
@@ -2363,6 +2385,28 @@ sctp_disposition_t sctp_sf_do_9_1_abort(const struct sctp_endpoint *ep,
 	if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t)))
 		return sctp_sf_pdiscard(ep, asoc, type, arg, commands);
 
+	/* ADD-IP: Special case for ABORT chunks
+	 * F4)  One special consideration is that ABORT Chunks arriving
+	 * destined to the IP address being deleted MUST be
+	 * ignored (see Section 5.3.1 for further details).
+	 */
+	if (SCTP_ADDR_DEL ==
+		    sctp_bind_addr_state(&asoc->base.bind_addr, &chunk->dest))
+		return sctp_sf_discard_chunk(ep, asoc, type, arg, commands);
+
+	return __sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands);
+}
+
+static sctp_disposition_t __sctp_sf_do_9_1_abort(const struct sctp_endpoint *ep,
+					const struct sctp_association *asoc,
+					const sctp_subtype_t type,
+					void *arg,
+					sctp_cmd_seq_t *commands)
+{
+	struct sctp_chunk *chunk = arg;
+	unsigned len;
+	__be16 error = SCTP_ERROR_NO_ERROR;
+
 	/* See if we have an error cause code in the chunk.  */
 	len = ntohs(chunk->chunk_hdr->length);
 	if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr))
-- 
cgit v1.2.3


From 152da81deb9a4870aeac352336184b2b14d4b2ba Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 20 Dec 2007 15:31:33 -0800
Subject: [INET]: Uninline the __inet_hash function.

This one is used in quite many places in the networking code and
seems to big to be inline.

After the patch net/ipv4/build-in.o loses ~650 bytes:
add/remove: 2/0 grow/shrink: 0/5 up/down: 461/-1114 (-653)
function                                     old     new   delta
__inet_hash_nolisten                           -     282    +282
__inet_hash                                    -     179    +179
tcp_sacktag_write_queue                     2255    2254      -1
__inet_lookup_listener                       284     274     -10
tcp_v4_syn_recv_sock                         755     493    -262
tcp_v4_hash                                  389      35    -354
inet_hash_connect                           1086     599    -487

This version addresses the issue pointed by Eric, that
while being inline this function was optimized by gcc
in respect to the 'listen_possible' argument.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 29 +++------------------------
 net/dccp/ipv4.c               |  2 +-
 net/ipv4/inet_hashtables.c    | 46 +++++++++++++++++++++++++++++++++++++++++--
 net/ipv4/tcp_ipv4.c           |  2 +-
 4 files changed, 49 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 37f6cb112127..982d487751f9 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -264,37 +264,14 @@ static inline void inet_listen_unlock(struct inet_hashinfo *hashinfo)
 		wake_up(&hashinfo->lhash_wait);
 }
 
-static inline void __inet_hash(struct inet_hashinfo *hashinfo,
-			       struct sock *sk, const int listen_possible)
-{
-	struct hlist_head *list;
-	rwlock_t *lock;
-
-	BUG_TRAP(sk_unhashed(sk));
-	if (listen_possible && sk->sk_state == TCP_LISTEN) {
-		list = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)];
-		lock = &hashinfo->lhash_lock;
-		inet_listen_wlock(hashinfo);
-	} else {
-		struct inet_ehash_bucket *head;
-		sk->sk_hash = inet_sk_ehashfn(sk);
-		head = inet_ehash_bucket(hashinfo, sk->sk_hash);
-		list = &head->chain;
-		lock = inet_ehash_lockp(hashinfo, sk->sk_hash);
-		write_lock(lock);
-	}
-	__sk_add_node(sk, list);
-	sock_prot_inc_use(sk->sk_prot);
-	write_unlock(lock);
-	if (listen_possible && sk->sk_state == TCP_LISTEN)
-		wake_up(&hashinfo->lhash_wait);
-}
+extern void __inet_hash(struct inet_hashinfo *hashinfo, struct sock *sk);
+extern void __inet_hash_nolisten(struct inet_hashinfo *hinfo, struct sock *sk);
 
 static inline void inet_hash(struct inet_hashinfo *hashinfo, struct sock *sk)
 {
 	if (sk->sk_state != TCP_CLOSE) {
 		local_bh_disable();
-		__inet_hash(hashinfo, sk, 1);
+		__inet_hash(hashinfo, sk);
 		local_bh_enable();
 	}
 }
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 02fc91ce2505..f450df2fc863 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -408,7 +408,7 @@ struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb,
 
 	dccp_sync_mss(newsk, dst_mtu(dst));
 
-	__inet_hash(&dccp_hashinfo, newsk, 0);
+	__inet_hash_nolisten(&dccp_hashinfo, newsk);
 	__inet_inherit_port(&dccp_hashinfo, sk, newsk);
 
 	return newsk;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 67704da04fc4..8dfd5a691e5f 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -267,6 +267,48 @@ static inline u32 inet_sk_port_offset(const struct sock *sk)
 					  inet->dport);
 }
 
+void __inet_hash_nolisten(struct inet_hashinfo *hashinfo, struct sock *sk)
+{
+	struct hlist_head *list;
+	rwlock_t *lock;
+	struct inet_ehash_bucket *head;
+
+	BUG_TRAP(sk_unhashed(sk));
+
+	sk->sk_hash = inet_sk_ehashfn(sk);
+	head = inet_ehash_bucket(hashinfo, sk->sk_hash);
+	list = &head->chain;
+	lock = inet_ehash_lockp(hashinfo, sk->sk_hash);
+
+	write_lock(lock);
+	__sk_add_node(sk, list);
+	sock_prot_inc_use(sk->sk_prot);
+	write_unlock(lock);
+}
+EXPORT_SYMBOL_GPL(__inet_hash_nolisten);
+
+void __inet_hash(struct inet_hashinfo *hashinfo, struct sock *sk)
+{
+	struct hlist_head *list;
+	rwlock_t *lock;
+
+	if (sk->sk_state != TCP_LISTEN) {
+		__inet_hash_nolisten(hashinfo, sk);
+		return;
+	}
+
+	BUG_TRAP(sk_unhashed(sk));
+	list = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)];
+	lock = &hashinfo->lhash_lock;
+
+	inet_listen_wlock(hashinfo);
+	__sk_add_node(sk, list);
+	sock_prot_inc_use(sk->sk_prot);
+	write_unlock(lock);
+	wake_up(&hashinfo->lhash_wait);
+}
+EXPORT_SYMBOL_GPL(__inet_hash);
+
 /*
  * Bind a port for a connect operation and hash it.
  */
@@ -334,7 +376,7 @@ ok:
 		inet_bind_hash(sk, tb, port);
 		if (sk_unhashed(sk)) {
 			inet_sk(sk)->sport = htons(port);
-			__inet_hash(hinfo, sk, 0);
+			__inet_hash_nolisten(hinfo, sk);
 		}
 		spin_unlock(&head->lock);
 
@@ -351,7 +393,7 @@ ok:
 	tb  = inet_csk(sk)->icsk_bind_hash;
 	spin_lock_bh(&head->lock);
 	if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
-		__inet_hash(hinfo, sk, 0);
+		__inet_hash_nolisten(hinfo, sk);
 		spin_unlock_bh(&head->lock);
 		return 0;
 	} else {
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 652c32368ccc..fc9bdd8b7dcf 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1478,7 +1478,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 	}
 #endif
 
-	__inet_hash(&tcp_hashinfo, newsk, 0);
+	__inet_hash_nolisten(&tcp_hashinfo, newsk);
 	__inet_inherit_port(&tcp_hashinfo, sk, newsk);
 
 	return newsk;
-- 
cgit v1.2.3


From 77a5ba55dab7b4ece12f37c717022819e3f77b44 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 20 Dec 2007 15:32:17 -0800
Subject: [INET]: Uninline the __inet_lookup_established function.

This is -700 bytes from the net/ipv4/built-in.o

add/remove: 1/0 grow/shrink: 1/3 up/down: 340/-1040 (-700)
function                                     old     new   delta
__inet_lookup_established                      -     339    +339
tcp_sacktag_write_queue                     2254    2255      +1
tcp_v4_err                                  1304     973    -331
tcp_v4_rcv                                  2089    1744    -345
tcp_v4_do_rcv                                826     462    -364

Exporting is for dccp module (used via e.g. inet_lookup).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 40 +++-------------------------------------
 net/ipv4/inet_hashtables.c    | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 982d487751f9..65ddb25d62e8 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -374,43 +374,9 @@ typedef __u64 __bitwise __addrpair;
  *
  * Local BH must be disabled here.
  */
-static inline struct sock *
-	__inet_lookup_established(struct inet_hashinfo *hashinfo,
-				  const __be32 saddr, const __be16 sport,
-				  const __be32 daddr, const u16 hnum,
-				  const int dif)
-{
-	INET_ADDR_COOKIE(acookie, saddr, daddr)
-	const __portpair ports = INET_COMBINED_PORTS(sport, hnum);
-	struct sock *sk;
-	const struct hlist_node *node;
-	/* Optimize here for direct hit, only listening connections can
-	 * have wildcards anyways.
-	 */
-	unsigned int hash = inet_ehashfn(daddr, hnum, saddr, sport);
-	struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash);
-	rwlock_t *lock = inet_ehash_lockp(hashinfo, hash);
-
-	prefetch(head->chain.first);
-	read_lock(lock);
-	sk_for_each(sk, node, &head->chain) {
-		if (INET_MATCH(sk, hash, acookie, saddr, daddr, ports, dif))
-			goto hit; /* You sunk my battleship! */
-	}
-
-	/* Must check for a TIME_WAIT'er before going to listener hash. */
-	sk_for_each(sk, node, &head->twchain) {
-		if (INET_TW_MATCH(sk, hash, acookie, saddr, daddr, ports, dif))
-			goto hit;
-	}
-	sk = NULL;
-out:
-	read_unlock(lock);
-	return sk;
-hit:
-	sock_hold(sk);
-	goto out;
-}
+extern struct sock * __inet_lookup_established(struct inet_hashinfo *hashinfo,
+		const __be32 saddr, const __be16 sport,
+		const __be32 daddr, const u16 hnum, const int dif);
 
 static inline struct sock *
 	inet_lookup_established(struct inet_hashinfo *hashinfo,
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 8dfd5a691e5f..2e5814a84360 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -190,6 +190,44 @@ sherry_cache:
 }
 EXPORT_SYMBOL_GPL(__inet_lookup_listener);
 
+struct sock * __inet_lookup_established(struct inet_hashinfo *hashinfo,
+				  const __be32 saddr, const __be16 sport,
+				  const __be32 daddr, const u16 hnum,
+				  const int dif)
+{
+	INET_ADDR_COOKIE(acookie, saddr, daddr)
+	const __portpair ports = INET_COMBINED_PORTS(sport, hnum);
+	struct sock *sk;
+	const struct hlist_node *node;
+	/* Optimize here for direct hit, only listening connections can
+	 * have wildcards anyways.
+	 */
+	unsigned int hash = inet_ehashfn(daddr, hnum, saddr, sport);
+	struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash);
+	rwlock_t *lock = inet_ehash_lockp(hashinfo, hash);
+
+	prefetch(head->chain.first);
+	read_lock(lock);
+	sk_for_each(sk, node, &head->chain) {
+		if (INET_MATCH(sk, hash, acookie, saddr, daddr, ports, dif))
+			goto hit; /* You sunk my battleship! */
+	}
+
+	/* Must check for a TIME_WAIT'er before going to listener hash. */
+	sk_for_each(sk, node, &head->twchain) {
+		if (INET_TW_MATCH(sk, hash, acookie, saddr, daddr, ports, dif))
+			goto hit;
+	}
+	sk = NULL;
+out:
+	read_unlock(lock);
+	return sk;
+hit:
+	sock_hold(sk);
+	goto out;
+}
+EXPORT_SYMBOL_GPL(__inet_lookup_established);
+
 /* called with local bh disabled */
 static int __inet_check_established(struct inet_timewait_death_row *death_row,
 				    struct sock *sk, __u16 lport,
-- 
cgit v1.2.3


From 7054fb9376e111d0edc06efcedbac6930a6caf76 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 20 Dec 2007 15:32:54 -0800
Subject: [INET]: Uninline the inet_twsk_put function.

This one is not that big, but is widely used: saves 1200 bytes
from net/ipv4/built-in.o

add/remove: 1/0 grow/shrink: 1/12 up/down: 97/-1300 (-1203)
function                                     old     new   delta
inet_twsk_put                                  -      87     +87
__inet_lookup_listener                       274     284     +10
tcp_sacktag_write_queue                     2255    2254      -1
tcp_time_wait                                482     411     -71
__inet_check_established                     796     722     -74
tcp_v4_err                                   973     898     -75
__inet_twsk_kill                             230     154     -76
inet_twsk_deschedule                         180     103     -77
tcp_v4_do_rcv                                462     384     -78
inet_hash_connect                            686     607     -79
inet_twdr_do_twkill_work                     236     150     -86
inet_twdr_twcal_tick                         395     307     -88
tcp_v4_rcv                                  1744    1480    -264
tcp_timewait_state_process                   975     644    -331

Export it for ipv6 module.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_timewait_sock.h | 14 +-------------
 net/ipv4/inet_timewait_sock.c    | 15 +++++++++++++++
 2 files changed, 16 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index abaff0597270..67e925065aae 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -193,19 +193,7 @@ static inline __be32 inet_rcv_saddr(const struct sock *sk)
 		inet_sk(sk)->rcv_saddr : inet_twsk(sk)->tw_rcv_saddr;
 }
 
-static inline void inet_twsk_put(struct inet_timewait_sock *tw)
-{
-	if (atomic_dec_and_test(&tw->tw_refcnt)) {
-		struct module *owner = tw->tw_prot->owner;
-		twsk_destructor((struct sock *)tw);
-#ifdef SOCK_REFCNT_DEBUG
-		printk(KERN_DEBUG "%s timewait_sock %p released\n",
-		       tw->tw_prot->name, tw);
-#endif
-		kmem_cache_free(tw->tw_prot->twsk_prot->twsk_slab, tw);
-		module_put(owner);
-	}
-}
+extern void inet_twsk_put(struct inet_timewait_sock *tw);
 
 extern struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
 						  const int state);
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index d43e787031a4..1b7db4208969 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -48,6 +48,21 @@ static void __inet_twsk_kill(struct inet_timewait_sock *tw,
 	inet_twsk_put(tw);
 }
 
+void inet_twsk_put(struct inet_timewait_sock *tw)
+{
+	if (atomic_dec_and_test(&tw->tw_refcnt)) {
+		struct module *owner = tw->tw_prot->owner;
+		twsk_destructor((struct sock *)tw);
+#ifdef SOCK_REFCNT_DEBUG
+		printk(KERN_DEBUG "%s timewait_sock %p released\n",
+		       tw->tw_prot->name, tw);
+#endif
+		kmem_cache_free(tw->tw_prot->twsk_prot->twsk_slab, tw);
+		module_put(owner);
+	}
+}
+EXPORT_SYMBOL_GPL(inet_twsk_put);
+
 /*
  * Enter the time wait state. This is called with locally disabled BH.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
-- 
cgit v1.2.3


From a1b051405bc16222d92c73b0c26d65b333a154ee Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Thu, 20 Dec 2007 20:41:12 -0800
Subject: [XFRM] IPv6: Fix dst/routing check at transformation.

IPv6 specific thing is wrongly removed from transformation at net-2.6.25.
This patch recovers it with current design.

o Update "path" of xfrm_dst since IPv6 transformation should
  care about routing changes. It is required by MIPv6 and
  off-link destined IPsec.
o Rename nfheader_len which is for non-fragment transformation used by
  MIPv6 to rt6i_nfheader_len as IPv6 name space.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  2 +-
 include/net/xfrm.h      |  3 +++
 net/ipv4/xfrm4_policy.c |  7 +++++++
 net/ipv6/ip6_output.c   |  4 ++--
 net/ipv6/xfrm6_policy.c | 17 +++++++++++++++++
 net/xfrm/xfrm_policy.c  | 21 +++++++++++++++++++++
 6 files changed, 51 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 14830edc2ac0..d8d85b13364d 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -101,7 +101,7 @@ struct rt6_info
 	atomic_t			rt6i_ref;
 
 	/* more non-fragment space at head required */
-	unsigned short			nfheader_len;
+	unsigned short			rt6i_nfheader_len;
 
 	u8				rt6i_protocol;
 
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d6dae5ae7abe..eea1c327c93e 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -242,6 +242,9 @@ struct xfrm_policy_afinfo {
 						  struct flowi *fl,
 						  int reverse);
 	int			(*get_tos)(struct flowi *fl);
+	int			(*init_path)(struct xfrm_dst *path,
+					     struct dst_entry *dst,
+					     int nfheader_len);
 	int			(*fill_dst)(struct xfrm_dst *xdst,
 					    struct net_device *dev);
 };
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 5ccae3a463c2..656345f75e0d 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -84,6 +84,12 @@ static int xfrm4_get_tos(struct flowi *fl)
 	return fl->fl4_tos;
 }
 
+static int xfrm4_init_path(struct xfrm_dst *path, struct dst_entry *dst,
+			   int nfheader_len)
+{
+	return 0;
+}
+
 static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 {
 	struct rtable *rt = (struct rtable *)xdst->route;
@@ -251,6 +257,7 @@ static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
 	.find_bundle = 		__xfrm4_find_bundle,
 	.decode_session =	_decode_session4,
 	.get_tos =		xfrm4_get_tos,
+	.init_path =		xfrm4_init_path,
 	.fill_dst =		xfrm4_fill_dst,
 };
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index d54da616e3af..4686646058d3 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1126,7 +1126,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		sk->sk_sndmsg_page = NULL;
 		sk->sk_sndmsg_off = 0;
 		exthdrlen = rt->u.dst.header_len + (opt ? opt->opt_flen : 0) -
-			    rt->nfheader_len;
+			    rt->rt6i_nfheader_len;
 		length += exthdrlen;
 		transhdrlen += exthdrlen;
 	} else {
@@ -1141,7 +1141,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 
 	hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
 
-	fragheaderlen = sizeof(struct ipv6hdr) + rt->nfheader_len +
+	fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +
 			(opt ? opt->opt_nflen : 0);
 	maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen - sizeof(struct frag_hdr);
 
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index d26b7dc3f33b..cf373b46a1ba 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -98,6 +98,20 @@ static int xfrm6_get_tos(struct flowi *fl)
 	return 0;
 }
 
+static int xfrm6_init_path(struct xfrm_dst *path, struct dst_entry *dst,
+			   int nfheader_len)
+{
+	if (dst->ops->family == AF_INET6) {
+		struct rt6_info *rt = (struct rt6_info*)dst;
+		if (rt->rt6i_node)
+			path->path_cookie = rt->rt6i_node->fn_sernum;
+	}
+
+	path->u.rt6.rt6i_nfheader_len = nfheader_len;
+
+	return 0;
+}
+
 static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 {
 	struct rt6_info *rt = (struct rt6_info*)xdst->route;
@@ -115,6 +129,8 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 						   RTF_LOCAL);
 	xdst->u.rt6.rt6i_metric = rt->rt6i_metric;
 	xdst->u.rt6.rt6i_node = rt->rt6i_node;
+	if (rt->rt6i_node)
+		xdst->route_cookie = rt->rt6i_node->fn_sernum;
 	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
 	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
 	xdst->u.rt6.rt6i_src = rt->rt6i_src;
@@ -266,6 +282,7 @@ static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
 	.find_bundle =		__xfrm6_find_bundle,
 	.decode_session =	_decode_session6,
 	.get_tos =		xfrm6_get_tos,
+	.init_path =		xfrm6_init_path,
 	.fill_dst =		xfrm6_fill_dst,
 };
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 8023a3c0dad5..521cb6e12561 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1266,6 +1266,23 @@ static inline struct xfrm_dst *xfrm_alloc_dst(int family)
 	return xdst;
 }
 
+static inline int xfrm_init_path(struct xfrm_dst *path, struct dst_entry *dst,
+				 int nfheader_len)
+{
+	struct xfrm_policy_afinfo *afinfo =
+		xfrm_policy_get_afinfo(dst->ops->family);
+	int err;
+
+	if (!afinfo)
+		return -EINVAL;
+
+	err = afinfo->init_path(path, dst, nfheader_len);
+
+	xfrm_policy_put_afinfo(afinfo);
+
+	return err;
+}
+
 static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
 {
 	struct xfrm_policy_afinfo *afinfo =
@@ -1298,6 +1315,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 	int i = 0;
 	int err;
 	int header_len = 0;
+	int nfheader_len = 0;
 	int trailer_len = 0;
 	int tos;
 	int family = policy->selector.family;
@@ -1352,6 +1370,8 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 		dst_prev = dst1;
 
 		header_len += xfrm[i]->props.header_len;
+		if (xfrm[i]->type->flags & XFRM_TYPE_NON_FRAGMENT)
+			nfheader_len += xfrm[i]->props.header_len;
 		trailer_len += xfrm[i]->props.trailer_len;
 	}
 
@@ -1366,6 +1386,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 	/* Copy neighbout for reachability confirmation */
 	dst0->neighbour = neigh_clone(dst->neighbour);
 
+	xfrm_init_path((struct xfrm_dst *)dst0, dst, nfheader_len);
 	xfrm_init_pmtu(dst_prev);
 
 	for (dst_prev = dst0; dst_prev != dst; dst_prev = dst_prev->child) {
-- 
cgit v1.2.3


From 558f82ef6e0d25e87f7468c07b6db1fbbf95a855 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Thu, 20 Dec 2007 20:42:57 -0800
Subject: [XFRM]: Define packet dropping statistics.

This statistics is shown factor dropped by transformation
at /proc/net/xfrm_stat for developer.
It is a counter designed from current transformation source code
and defined as linux private MIB.

See Documentation/networking/xfrm_proc.txt for the detail.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/xfrm_proc.txt | 71 +++++++++++++++++++++++++
 include/linux/snmp.h                   | 31 +++++++++++
 include/net/snmp.h                     |  5 ++
 include/net/xfrm.h                     | 18 +++++++
 net/xfrm/Makefile                      |  1 +
 net/xfrm/xfrm_policy.c                 | 24 +++++++++
 net/xfrm/xfrm_proc.c                   | 96 ++++++++++++++++++++++++++++++++++
 7 files changed, 246 insertions(+)
 create mode 100644 Documentation/networking/xfrm_proc.txt
 create mode 100644 net/xfrm/xfrm_proc.c

(limited to 'include')

diff --git a/Documentation/networking/xfrm_proc.txt b/Documentation/networking/xfrm_proc.txt
new file mode 100644
index 000000000000..ec9045b5c34a
--- /dev/null
+++ b/Documentation/networking/xfrm_proc.txt
@@ -0,0 +1,71 @@
+XFRM proc - /proc/net/xfrm_* files
+==================================
+Masahide NAKAMURA <nakam@linux-ipv6.org>
+
+
+Transformation Statistics
+-------------------------
+xfrm_proc is a statistics shown factor dropped by transformation
+for developer.
+It is a counter designed from current transformation source code
+and defined like linux private MIB.
+
+Inbound statistics
+~~~~~~~~~~~~~~~~~~
+XfrmInError:
+	All errors which is not matched others
+XfrmInBufferError:
+	No buffer is left
+XfrmInHdrError:
+	Header error
+XfrmInNoStates:
+	No state is found
+	i.e. Either inbound SPI, address, or IPsec protocol at SA is wrong
+XfrmInStateProtoError:
+	Transformation protocol specific error
+	e.g. SA key is wrong
+XfrmInStateModeError:
+	Transformation mode specific error
+XfrmInSeqOutOfWindow:
+	Sequence out of window
+XfrmInStateExpired:
+	State is expired
+XfrmInStateMismatch:
+	State has mismatch option
+	e.g. UDP encapsulation type is mismatch
+XfrmInStateInvalid:
+	State is invalid
+XfrmInTmplMismatch:
+	No matching template for states
+	e.g. Inbound SAs are correct but SP rule is wrong
+XfrmInNoPols:
+	No policy is found for states
+	e.g. Inbound SAs are correct but no SP is found
+XfrmInPolBlock:
+	Policy discards
+XfrmInPolError:
+	Policy error
+
+Outbound errors
+~~~~~~~~~~~~~~~
+XfrmOutError:
+	All errors which is not matched others
+XfrmOutBundleGenError:
+	Bundle generation error
+XfrmOutBundleCheckError:
+	Bundle check error
+XfrmOutNoStates:
+	No state is found
+XfrmOutStateProtoError:
+	Transformation protocol specific error
+XfrmOutStateModeError:
+	Transformation mode specific error
+	e.g. Outer header space is not enough
+XfrmOutStateExpired:
+	State is expired
+XfrmOutPolBlock:
+	Policy discards
+XfrmOutPolDead:
+	Policy is dead
+XfrmOutPolError:
+	Policy error
diff --git a/include/linux/snmp.h b/include/linux/snmp.h
index 89f0c2b5f405..86d3effb2836 100644
--- a/include/linux/snmp.h
+++ b/include/linux/snmp.h
@@ -217,4 +217,35 @@ enum
 	__LINUX_MIB_MAX
 };
 
+/* linux Xfrm mib definitions */
+enum
+{
+	LINUX_MIB_XFRMNUM = 0,
+	LINUX_MIB_XFRMINERROR,			/* XfrmInError */
+	LINUX_MIB_XFRMINBUFFERERROR,		/* XfrmInBufferError */
+	LINUX_MIB_XFRMINHDRERROR,		/* XfrmInHdrError */
+	LINUX_MIB_XFRMINNOSTATES,		/* XfrmInNoStates */
+	LINUX_MIB_XFRMINSTATEPROTOERROR,	/* XfrmInStateProtoError */
+	LINUX_MIB_XFRMINSTATEMODEERROR,		/* XfrmInStateModeError */
+	LINUX_MIB_XFRMINSEQOUTOFWINDOW,		/* XfrmInSeqOutOfWindow */
+	LINUX_MIB_XFRMINSTATEEXPIRED,		/* XfrmInStateExpired */
+	LINUX_MIB_XFRMINSTATEMISMATCH,		/* XfrmInStateMismatch */
+	LINUX_MIB_XFRMINSTATEINVALID,		/* XfrmInStateInvalid */
+	LINUX_MIB_XFRMINTMPLMISMATCH,		/* XfrmInTmplMismatch */
+	LINUX_MIB_XFRMINNOPOLS,			/* XfrmInNoPols */
+	LINUX_MIB_XFRMINPOLBLOCK,		/* XfrmInPolBlock */
+	LINUX_MIB_XFRMINPOLERROR,		/* XfrmInPolError */
+	LINUX_MIB_XFRMOUTERROR,			/* XfrmOutError */
+	LINUX_MIB_XFRMOUTBUNDLEGENERROR,	/* XfrmOutBundleGenError */
+	LINUX_MIB_XFRMOUTBUNDLECHECKERROR,	/* XfrmOutBundleCheckError */
+	LINUX_MIB_XFRMOUTNOSTATES,		/* XfrmOutNoStates */
+	LINUX_MIB_XFRMOUTSTATEPROTOERROR,	/* XfrmOutStateProtoError */
+	LINUX_MIB_XFRMOUTSTATEMODEERROR,	/* XfrmOutStateModeError */
+	LINUX_MIB_XFRMOUTSTATEEXPIRED,		/* XfrmOutStateExpired */
+	LINUX_MIB_XFRMOUTPOLBLOCK,		/* XfrmOutPolBlock */
+	LINUX_MIB_XFRMOUTPOLDEAD,		/* XfrmOutPolDead */
+	LINUX_MIB_XFRMOUTPOLERROR,		/* XfrmOutPolError */
+	__LINUX_MIB_XFRMMAX
+};
+
 #endif	/* _LINUX_SNMP_H */
diff --git a/include/net/snmp.h b/include/net/snmp.h
index fbb66663a42c..ce2f48507510 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -118,6 +118,11 @@ struct linux_mib {
 	unsigned long	mibs[LINUX_MIB_MAX];
 };
 
+/* Linux Xfrm */
+#define LINUX_MIB_XFRMMAX	__LINUX_MIB_XFRMMAX
+struct linux_xfrm_mib {
+	unsigned long	mibs[LINUX_MIB_XFRMMAX];
+};
 
 /* 
  * FIXME: On x86 and some other CPUs the split into user and softirq parts
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index eea1c327c93e..a79702bcdcd0 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -19,6 +19,9 @@
 #include <net/route.h>
 #include <net/ipv6.h>
 #include <net/ip6_fib.h>
+#ifdef CONFIG_XFRM_STATISTICS
+#include <net/snmp.h>
+#endif
 
 #define XFRM_PROTO_ESP		50
 #define XFRM_PROTO_AH		51
@@ -34,6 +37,17 @@
 #define MODULE_ALIAS_XFRM_TYPE(family, proto) \
 	MODULE_ALIAS("xfrm-type-" __stringify(family) "-" __stringify(proto))
 
+#ifdef CONFIG_XFRM_STATISTICS
+DECLARE_SNMP_STAT(struct linux_xfrm_mib, xfrm_statistics);
+#define XFRM_INC_STATS(field)		SNMP_INC_STATS(xfrm_statistics, field)
+#define XFRM_INC_STATS_BH(field)	SNMP_INC_STATS_BH(xfrm_statistics, field)
+#define XFRM_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(xfrm_statistics, field)
+#else
+#define XFRM_INC_STATS(field)
+#define XFRM_INC_STATS_BH(field)
+#define XFRM_INC_STATS_USER(field)
+#endif
+
 extern struct sock *xfrm_nl;
 extern u32 sysctl_xfrm_aevent_etime;
 extern u32 sysctl_xfrm_aevent_rseqth;
@@ -1139,6 +1153,10 @@ static inline void xfrm6_fini(void)
 }
 #endif
 
+#ifdef CONFIG_XFRM_STATISTICS
+extern int xfrm_proc_init(void);
+#endif
+
 extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *);
 extern struct xfrm_state *xfrm_state_alloc(void);
 extern struct xfrm_state *xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, 
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index 45744a3d3a51..332cfb0ff566 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -4,5 +4,6 @@
 
 obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \
 		      xfrm_input.o xfrm_output.o xfrm_algo.o
+obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 521cb6e12561..32ddb7b12e7f 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -27,11 +27,19 @@
 #include <net/dst.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
+#ifdef CONFIG_XFRM_STATISTICS
+#include <net/snmp.h>
+#endif
 
 #include "xfrm_hash.h"
 
 int sysctl_xfrm_larval_drop __read_mostly;
 
+#ifdef CONFIG_XFRM_STATISTICS
+DEFINE_SNMP_STAT(struct linux_xfrm_mib, xfrm_statistics) __read_mostly;
+EXPORT_SYMBOL(xfrm_statistics);
+#endif
+
 DEFINE_MUTEX(xfrm_cfg_mutex);
 EXPORT_SYMBOL(xfrm_cfg_mutex);
 
@@ -2258,6 +2266,16 @@ static struct notifier_block xfrm_dev_notifier = {
 	0
 };
 
+#ifdef CONFIG_XFRM_STATISTICS
+static int __init xfrm_statistics_init(void)
+{
+	if (snmp_mib_init((void **)xfrm_statistics,
+			  sizeof(struct linux_xfrm_mib)) < 0)
+		return -ENOMEM;
+	return 0;
+}
+#endif
+
 static void __init xfrm_policy_init(void)
 {
 	unsigned int hmask, sz;
@@ -2294,9 +2312,15 @@ static void __init xfrm_policy_init(void)
 
 void __init xfrm_init(void)
 {
+#ifdef CONFIG_XFRM_STATISTICS
+	xfrm_statistics_init();
+#endif
 	xfrm_state_init();
 	xfrm_policy_init();
 	xfrm_input_init();
+#ifdef CONFIG_XFRM_STATISTICS
+	xfrm_proc_init();
+#endif
 }
 
 #ifdef CONFIG_AUDITSYSCALL
diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c
new file mode 100644
index 000000000000..31d035415ecd
--- /dev/null
+++ b/net/xfrm/xfrm_proc.c
@@ -0,0 +1,96 @@
+/*
+ * xfrm_proc.c
+ *
+ * Copyright (C)2006-2007 USAGI/WIDE Project
+ *
+ * Authors:	Masahide NAKAMURA <nakam@linux-ipv6.org>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <net/snmp.h>
+#include <net/xfrm.h>
+
+static struct snmp_mib xfrm_mib_list[] = {
+	SNMP_MIB_ITEM("XfrmInError", LINUX_MIB_XFRMINERROR),
+	SNMP_MIB_ITEM("XfrmInBufferError", LINUX_MIB_XFRMINBUFFERERROR),
+	SNMP_MIB_ITEM("XfrmInHdrError", LINUX_MIB_XFRMINHDRERROR),
+	SNMP_MIB_ITEM("XfrmInNoStates", LINUX_MIB_XFRMINNOSTATES),
+	SNMP_MIB_ITEM("XfrmInStateProtoError", LINUX_MIB_XFRMINSTATEPROTOERROR),
+	SNMP_MIB_ITEM("XfrmInStateModeError", LINUX_MIB_XFRMINSTATEMODEERROR),
+	SNMP_MIB_ITEM("XfrmInSeqOutOfWindow", LINUX_MIB_XFRMINSEQOUTOFWINDOW),
+	SNMP_MIB_ITEM("XfrmInStateExpired", LINUX_MIB_XFRMINSTATEEXPIRED),
+	SNMP_MIB_ITEM("XfrmInStateMismatch", LINUX_MIB_XFRMINSTATEMISMATCH),
+	SNMP_MIB_ITEM("XfrmInStateInvalid", LINUX_MIB_XFRMINSTATEINVALID),
+	SNMP_MIB_ITEM("XfrmInTmplMismatch", LINUX_MIB_XFRMINTMPLMISMATCH),
+	SNMP_MIB_ITEM("XfrmInNoPols", LINUX_MIB_XFRMINNOPOLS),
+	SNMP_MIB_ITEM("XfrmInPolBlock", LINUX_MIB_XFRMINPOLBLOCK),
+	SNMP_MIB_ITEM("XfrmInPolError", LINUX_MIB_XFRMINPOLERROR),
+	SNMP_MIB_ITEM("XfrmOutError", LINUX_MIB_XFRMOUTERROR),
+	SNMP_MIB_ITEM("XfrmOutBundleGenError", LINUX_MIB_XFRMOUTBUNDLEGENERROR),
+	SNMP_MIB_ITEM("XfrmOutBundleCheckError", LINUX_MIB_XFRMOUTBUNDLECHECKERROR),
+	SNMP_MIB_ITEM("XfrmOutNoStates", LINUX_MIB_XFRMOUTNOSTATES),
+	SNMP_MIB_ITEM("XfrmOutStateProtoError", LINUX_MIB_XFRMOUTSTATEPROTOERROR),
+	SNMP_MIB_ITEM("XfrmOutStateModeError", LINUX_MIB_XFRMOUTSTATEMODEERROR),
+	SNMP_MIB_ITEM("XfrmOutStateExpired", LINUX_MIB_XFRMOUTSTATEEXPIRED),
+	SNMP_MIB_ITEM("XfrmOutPolBlock", LINUX_MIB_XFRMOUTPOLBLOCK),
+	SNMP_MIB_ITEM("XfrmOutPolDead", LINUX_MIB_XFRMOUTPOLDEAD),
+	SNMP_MIB_ITEM("XfrmOutPolError", LINUX_MIB_XFRMOUTPOLERROR),
+	SNMP_MIB_SENTINEL
+};
+
+static unsigned long
+fold_field(void *mib[], int offt)
+{
+        unsigned long res = 0;
+        int i;
+
+        for_each_possible_cpu(i) {
+                res += *(((unsigned long *)per_cpu_ptr(mib[0], i)) + offt);
+                res += *(((unsigned long *)per_cpu_ptr(mib[1], i)) + offt);
+        }
+        return res;
+}
+
+static int xfrm_statistics_seq_show(struct seq_file *seq, void *v)
+{
+	int i;
+	for (i=0; xfrm_mib_list[i].name; i++)
+		seq_printf(seq, "%-24s\t%lu\n", xfrm_mib_list[i].name,
+			   fold_field((void **)xfrm_statistics,
+				      xfrm_mib_list[i].entry));
+	return 0;
+}
+
+static int xfrm_statistics_seq_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, xfrm_statistics_seq_show, NULL);
+}
+
+static struct file_operations xfrm_statistics_seq_fops = {
+	.owner	 = THIS_MODULE,
+	.open	 = xfrm_statistics_seq_open,
+	.read	 = seq_read,
+	.llseek	 = seq_lseek,
+	.release = single_release,
+};
+
+int __init xfrm_proc_init(void)
+{
+	int rc = 0;
+
+	if (!proc_net_fops_create(&init_net, "xfrm_stat", S_IRUGO,
+				  &xfrm_statistics_seq_fops))
+		goto stat_fail;
+
+ out:
+	return rc;
+
+ stat_fail:
+	rc = -ENOMEM;
+	goto out;
+}
-- 
cgit v1.2.3


From 68277accb3a5f004344f4346498640601b8b7016 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Thu, 20 Dec 2007 20:49:33 -0800
Subject: [XFRM]: Assorted IPsec fixups

This patch fixes a number of small but potentially troublesome things in the
XFRM/IPsec code:

 * Use the 'audit_enabled' variable already in include/linux/audit.h
   Removed the need for extern declarations local to each XFRM audit fuction

 * Convert 'sid' to 'secid' everywhere we can
   The 'sid' name is specific to SELinux, 'secid' is the common naming
   convention used by the kernel when refering to tokenized LSM labels,
   unfortunately we have to leave 'ctx_sid' in 'struct xfrm_sec_ctx' otherwise
   we risk breaking userspace

 * Convert address display to use standard NIP* macros
   Similar to what was recently done with the SPD audit code, this also also
   includes the removal of some unnecessary memcpy() calls

 * Move common code to xfrm_audit_common_stateinfo()
   Code consolidation from the "less is more" book on software development

 * Proper spacing around commas in function arguments
   Minor style tweak since I was already touching the code

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 14 ++++++-------
 net/xfrm/xfrm_policy.c | 15 +++++++-------
 net/xfrm/xfrm_state.c  | 53 +++++++++++++++++++++-----------------------------
 3 files changed, 36 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index a79702bcdcd0..f333c95c4189 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -565,7 +565,7 @@ struct xfrm_audit
 };
 
 #ifdef CONFIG_AUDITSYSCALL
-static inline struct audit_buffer *xfrm_audit_start(u32 auid, u32 sid)
+static inline struct audit_buffer *xfrm_audit_start(u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf = NULL;
 	char *secctx;
@@ -578,8 +578,8 @@ static inline struct audit_buffer *xfrm_audit_start(u32 auid, u32 sid)
 
 	audit_log_format(audit_buf, "auid=%u", auid);
 
-	if (sid != 0 &&
-	    security_secid_to_secctx(sid, &secctx, &secctx_len) == 0) {
+	if (secid != 0 &&
+	    security_secid_to_secctx(secid, &secctx, &secctx_len) == 0) {
 		audit_log_format(audit_buf, " subj=%s", secctx);
 		security_release_secctx(secctx, secctx_len);
 	} else
@@ -588,13 +588,13 @@ static inline struct audit_buffer *xfrm_audit_start(u32 auid, u32 sid)
 }
 
 extern void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
-				  u32 auid, u32 sid);
+				  u32 auid, u32 secid);
 extern void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
-				  u32 auid, u32 sid);
+				  u32 auid, u32 secid);
 extern void xfrm_audit_state_add(struct xfrm_state *x, int result,
-				 u32 auid, u32 sid);
+				 u32 auid, u32 secid);
 extern void xfrm_audit_state_delete(struct xfrm_state *x, int result,
-				    u32 auid, u32 sid);
+				    u32 auid, u32 secid);
 #else
 #define xfrm_audit_policy_add(x, r, a, s)	do { ; } while (0)
 #define xfrm_audit_policy_delete(x, r, a, s)	do { ; } while (0)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 74807a7d3d69..abc3e39b115b 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -24,6 +24,7 @@
 #include <linux/netfilter.h>
 #include <linux/module.h>
 #include <linux/cache.h>
+#include <linux/audit.h>
 #include <net/dst.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
@@ -2401,15 +2402,14 @@ static inline void xfrm_audit_common_policyinfo(struct xfrm_policy *xp,
 	}
 }
 
-void
-xfrm_audit_policy_add(struct xfrm_policy *xp, int result, u32 auid, u32 sid)
+void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
+			   u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf;
-	extern int audit_enabled;
 
 	if (audit_enabled == 0)
 		return;
-	audit_buf = xfrm_audit_start(auid, sid);
+	audit_buf = xfrm_audit_start(auid, secid);
 	if (audit_buf == NULL)
 		return;
 	audit_log_format(audit_buf, " op=SPD-add res=%u", result);
@@ -2418,15 +2418,14 @@ xfrm_audit_policy_add(struct xfrm_policy *xp, int result, u32 auid, u32 sid)
 }
 EXPORT_SYMBOL_GPL(xfrm_audit_policy_add);
 
-void
-xfrm_audit_policy_delete(struct xfrm_policy *xp, int result, u32 auid, u32 sid)
+void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
+			      u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf;
-	extern int audit_enabled;
 
 	if (audit_enabled == 0)
 		return;
-	audit_buf = xfrm_audit_start(auid, sid);
+	audit_buf = xfrm_audit_start(auid, secid);
 	if (audit_buf == NULL)
 		return;
 	audit_log_format(audit_buf, " op=SPD-delete res=%u", result);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index f7c0951c9fd9..9e57378c51df 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -19,6 +19,7 @@
 #include <linux/ipsec.h>
 #include <linux/module.h>
 #include <linux/cache.h>
+#include <linux/audit.h>
 #include <asm/uaccess.h>
 
 #include "xfrm_hash.h"
@@ -1998,69 +1999,59 @@ void __init xfrm_state_init(void)
 static inline void xfrm_audit_common_stateinfo(struct xfrm_state *x,
 					       struct audit_buffer *audit_buf)
 {
-	if (x->security)
+	struct xfrm_sec_ctx *ctx = x->security;
+	u32 spi = ntohl(x->id.spi);
+
+	if (ctx)
 		audit_log_format(audit_buf, " sec_alg=%u sec_doi=%u sec_obj=%s",
-				 x->security->ctx_alg, x->security->ctx_doi,
-				 x->security->ctx_str);
+				 ctx->ctx_alg, ctx->ctx_doi, ctx->ctx_str);
 
 	switch(x->props.family) {
 	case AF_INET:
-		audit_log_format(audit_buf, " src=%u.%u.%u.%u dst=%u.%u.%u.%u",
+		audit_log_format(audit_buf,
+				 " src=" NIPQUAD_FMT " dst=" NIPQUAD_FMT,
 				 NIPQUAD(x->props.saddr.a4),
 				 NIPQUAD(x->id.daddr.a4));
 		break;
 	case AF_INET6:
-		{
-			struct in6_addr saddr6, daddr6;
-
-			memcpy(&saddr6, x->props.saddr.a6,
-				sizeof(struct in6_addr));
-			memcpy(&daddr6, x->id.daddr.a6,
-				sizeof(struct in6_addr));
-			audit_log_format(audit_buf,
-					 " src=" NIP6_FMT " dst=" NIP6_FMT,
-					 NIP6(saddr6), NIP6(daddr6));
-		}
+		audit_log_format(audit_buf,
+				 " src=" NIP6_FMT " dst=" NIP6_FMT,
+				 NIP6(*(struct in6_addr *)x->props.saddr.a6),
+				 NIP6(*(struct in6_addr *)x->id.daddr.a6));
 		break;
 	}
+
+	audit_log_format(audit_buf, " spi=%u(0x%x)", spi, spi);
 }
 
-void
-xfrm_audit_state_add(struct xfrm_state *x, int result, u32 auid, u32 sid)
+void xfrm_audit_state_add(struct xfrm_state *x, int result,
+			  u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf;
-	u32 spi;
-	extern int audit_enabled;
 
 	if (audit_enabled == 0)
 		return;
-	audit_buf = xfrm_audit_start(auid, sid);
+	audit_buf = xfrm_audit_start(auid, secid);
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SAD-add res=%u",result);
+	audit_log_format(audit_buf, " op=SAD-add res=%u", result);
 	xfrm_audit_common_stateinfo(x, audit_buf);
-	spi = ntohl(x->id.spi);
-	audit_log_format(audit_buf, " spi=%u(0x%x)", spi, spi);
 	audit_log_end(audit_buf);
 }
 EXPORT_SYMBOL_GPL(xfrm_audit_state_add);
 
-void
-xfrm_audit_state_delete(struct xfrm_state *x, int result, u32 auid, u32 sid)
+void xfrm_audit_state_delete(struct xfrm_state *x, int result,
+			     u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf;
-	u32 spi;
-	extern int audit_enabled;
 
 	if (audit_enabled == 0)
 		return;
-	audit_buf = xfrm_audit_start(auid, sid);
+	audit_buf = xfrm_audit_start(auid, secid);
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SAD-delete res=%u",result);
+	audit_log_format(audit_buf, " op=SAD-delete res=%u", result);
 	xfrm_audit_common_stateinfo(x, audit_buf);
-	spi = ntohl(x->id.spi);
-	audit_log_format(audit_buf, " spi=%u(0x%x)", spi, spi);
 	audit_log_end(audit_buf);
 }
 EXPORT_SYMBOL_GPL(xfrm_audit_state_delete);
-- 
cgit v1.2.3


From 678f5f7117d5780d3a51b201c9f44b7bf90f6a76 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 01:31:23 +0100
Subject: mac80211: clean up eapol handling in TX path

The previous patch left only one user of the ieee80211_is_eapol()
function and that user can be eliminated easily by introducing
a new "frame is EAPOL" flag to handle the frame specially (we
already have this information) instead of doing the (expensive)
ieee80211_is_eapol() all the time.

Also, allow unencrypted frames to be sent when they are injected.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h     |  1 +
 net/mac80211/ieee80211.c   |  2 ++
 net/mac80211/ieee80211_i.h |  2 +-
 net/mac80211/tx.c          |  7 ++++++-
 net/mac80211/util.c        | 17 -----------------
 5 files changed, 10 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3bd970f29f29..2606ca282c5e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -307,6 +307,7 @@ struct ieee80211_tx_control {
 						  * using the through
 						  * set_retry_limit configured
 						  * long retry value */
+#define IEEE80211_TXCTL_EAPOL_FRAME	(1<<11) /* internal to mac80211 */
 	u32 flags;			       /* tx control flags defined
 						* above */
 	u8 key_idx;		/* keyidx from hw->set_key(), undefined if
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index c0dbf77547f7..9c14e3d303c5 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -732,6 +732,8 @@ static void ieee80211_remove_tx_extra(struct ieee80211_local *local,
 		pkt_data->flags |= IEEE80211_TXPD_DO_NOT_ENCRYPT;
 	if (control->flags & IEEE80211_TXCTL_REQUEUE)
 		pkt_data->flags |= IEEE80211_TXPD_REQUEUE;
+	if (control->flags & IEEE80211_TXCTL_EAPOL_FRAME)
+		pkt_data->flags |= IEEE80211_TXPD_EAPOL_FRAME;
 	pkt_data->queue = control->queue;
 
 	hdrlen = ieee80211_get_hdrlen_from_skb(skb);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 214109b8d95a..baf53c047127 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -164,6 +164,7 @@ struct ieee80211_txrx_data {
 #define IEEE80211_TXPD_REQ_TX_STATUS	BIT(0)
 #define IEEE80211_TXPD_DO_NOT_ENCRYPT	BIT(1)
 #define IEEE80211_TXPD_REQUEUE		BIT(2)
+#define IEEE80211_TXPD_EAPOL_FRAME	BIT(3)
 /* Stored in sk_buff->cb */
 struct ieee80211_tx_packet_data {
 	int ifindex;
@@ -798,7 +799,6 @@ extern void *mac80211_wiphy_privid; /* for wiphy privid */
 extern const unsigned char rfc1042_header[6];
 extern const unsigned char bridge_tunnel_header[6];
 u8 *ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len);
-int ieee80211_is_eapol(const struct sk_buff *skb, int hdrlen);
 int ieee80211_frame_duration(struct ieee80211_local *local, size_t len,
 			     int rate, int erp, int short_preamble);
 void mac80211_ev_michael_mic_failure(struct net_device *dev, int keyidx,
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 6dbd91842881..e177a8dc23b9 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -437,7 +437,8 @@ ieee80211_tx_h_select_key(struct ieee80211_txrx_data *tx)
 	else if ((key = rcu_dereference(tx->sdata->default_key)))
 		tx->key = key;
 	else if (tx->sdata->drop_unencrypted &&
-		 !ieee80211_is_eapol(tx->skb, ieee80211_get_hdrlen(fc))) {
+		 !(tx->u.tx.control->flags & IEEE80211_TXCTL_EAPOL_FRAME) &&
+		 !(tx->flags & IEEE80211_TXRXD_TX_INJECTED)) {
 		I802_DEBUG_INC(tx->local->tx_handlers_drop_unencrypted);
 		return TXRX_DROP;
 	} else {
@@ -1241,6 +1242,8 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 		control.flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
 	if (pkt_data->flags & IEEE80211_TXPD_REQUEUE)
 		control.flags |= IEEE80211_TXCTL_REQUEUE;
+	if (pkt_data->flags & IEEE80211_TXPD_EAPOL_FRAME)
+		control.flags |= IEEE80211_TXCTL_EAPOL_FRAME;
 	control.queue = pkt_data->queue;
 
 	ret = ieee80211_tx(odev, skb, &control);
@@ -1514,6 +1517,8 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
 	memset(pkt_data, 0, sizeof(struct ieee80211_tx_packet_data));
 	pkt_data->ifindex = dev->ifindex;
+	if (ethertype == ETH_P_PAE)
+		pkt_data->flags |= IEEE80211_TXPD_EAPOL_FRAME;
 
 	skb->dev = local->mdev;
 	dev->stats.tx_packets++;
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index fb7fd896cd0d..2b02b2b9d645 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -40,10 +40,6 @@ const unsigned char rfc1042_header[] =
 const unsigned char bridge_tunnel_header[] =
 	{ 0xaa, 0xaa, 0x03, 0x00, 0x00, 0xf8 };
 
-/* No encapsulation header if EtherType < 0x600 (=length) */
-static const unsigned char eapol_header[] =
-	{ 0xaa, 0xaa, 0x03, 0x00, 0x00, 0x00, 0x88, 0x8e };
-
 
 static int rate_list_match(const int *rate_list, int rate)
 {
@@ -223,19 +219,6 @@ int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
 }
 EXPORT_SYMBOL(ieee80211_get_hdrlen_from_skb);
 
-int ieee80211_is_eapol(const struct sk_buff *skb, int hdrlen)
-{
-	if (unlikely(skb->len < 10))
-		return 0;
-
-	if (unlikely(skb->len >= hdrlen + sizeof(eapol_header) &&
-		     memcmp(skb->data + hdrlen, eapol_header,
-			    sizeof(eapol_header)) == 0))
-		return 1;
-
-	return 0;
-}
-
 void ieee80211_tx_set_iswep(struct ieee80211_txrx_data *tx)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) tx->skb->data;
-- 
cgit v1.2.3


From 7d54d0ddd66678ada6635159dac1eb82ccbe34b5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 01:31:25 +0100
Subject: mac80211: allow easier multicast/broadcast buffering in hardware

There are various decisions influencing the decision whether to buffer
a frame for after the next DTIM beacon. The "do we have stations in PS
mode" condition cannot be tested by the driver so mac80211 has to do
that. To ease driver writing for hardware that can buffer frames until
after the next DTIM beacon, introduce a new txctl flag telling the
driver to buffer a specific frame.

While at it, restructure and comment the code for multicast buffering
and remove spurious "inline" directives.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h |  2 ++
 net/mac80211/tx.c      | 32 +++++++++++++++++++++++---------
 2 files changed, 25 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 2606ca282c5e..5b9e7a262448 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -308,6 +308,8 @@ struct ieee80211_tx_control {
 						  * set_retry_limit configured
 						  * long retry value */
 #define IEEE80211_TXCTL_EAPOL_FRAME	(1<<11) /* internal to mac80211 */
+#define IEEE80211_TXCTL_SEND_AFTER_DTIM	(1<<12) /* send this frame after DTIM
+						 * beacon */
 	u32 flags;			       /* tx control flags defined
 						* above */
 	u8 key_idx;		/* keyidx from hw->set_key(), undefined if
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index e177a8dc23b9..f7aff2e97eea 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -322,16 +322,27 @@ static void purge_old_ps_buffers(struct ieee80211_local *local)
 	       wiphy_name(local->hw.wiphy), purged);
 }
 
-static inline ieee80211_txrx_result
+static ieee80211_txrx_result
 ieee80211_tx_h_multicast_ps_buf(struct ieee80211_txrx_data *tx)
 {
-	/* broadcast/multicast frame */
-	/* If any of the associated stations is in power save mode,
-	 * the frame is buffered to be sent after DTIM beacon frame */
-	if ((tx->local->hw.flags & IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING) &&
-	    tx->sdata->type != IEEE80211_IF_TYPE_WDS &&
-	    tx->sdata->bss && atomic_read(&tx->sdata->bss->num_sta_ps) &&
-	    !(tx->fc & IEEE80211_FCTL_ORDER)) {
+	/*
+	 * broadcast/multicast frame
+	 *
+	 * If any of the associated stations is in power save mode,
+	 * the frame is buffered to be sent after DTIM beacon frame.
+	 * This is done either by the hardware or us.
+	 */
+
+	/* not AP/IBSS or ordered frame */
+	if (!tx->sdata->bss || (tx->fc & IEEE80211_FCTL_ORDER))
+		return TXRX_CONTINUE;
+
+	/* no stations in PS mode */
+	if (!atomic_read(&tx->sdata->bss->num_sta_ps))
+		return TXRX_CONTINUE;
+
+	/* buffered in mac80211 */
+	if (tx->local->hw.flags & IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING) {
 		if (tx->local->total_ps_buffered >= TOTAL_MAX_TX_BUFFER)
 			purge_old_ps_buffers(tx->local);
 		if (skb_queue_len(&tx->sdata->bss->ps_bc_buf) >=
@@ -348,10 +359,13 @@ ieee80211_tx_h_multicast_ps_buf(struct ieee80211_txrx_data *tx)
 		return TXRX_QUEUED;
 	}
 
+	/* buffered in hardware */
+	tx->u.tx.control->flags |= IEEE80211_TXCTL_SEND_AFTER_DTIM;
+
 	return TXRX_CONTINUE;
 }
 
-static inline ieee80211_txrx_result
+static ieee80211_txrx_result
 ieee80211_tx_h_unicast_ps_buf(struct ieee80211_txrx_data *tx)
 {
 	struct sta_info *sta = tx->sta;
-- 
cgit v1.2.3


From 41ade00f21a72d30911c6351a93823a491fffa39 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 02:03:29 +0100
Subject: cfg80211/nl80211: introduce key handling

This introduces key handling to cfg80211/nl80211. Default
and group keys can be added, changed and removed; sequence
counters for each key can be retrieved.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/nl80211.h |  34 ++++++
 include/net/cfg80211.h  |  44 ++++++++
 net/wireless/core.c     |   3 +
 net/wireless/nl80211.c  | 289 ++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 370 insertions(+)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 538ee1dd3d0a..8dc807d9c29a 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -37,6 +37,16 @@
  *	userspace to request deletion of a virtual interface, then requires
  *	attribute %NL80211_ATTR_IFINDEX.
  *
+ * @NL80211_CMD_GET_KEY: Get sequence counter information for a key specified
+ *	by %NL80211_ATTR_KEY_IDX and/or %NL80211_ATTR_MAC.
+ * @NL80211_CMD_SET_KEY: Set key attributes %NL80211_ATTR_KEY_DEFAULT or
+ *	%NL80211_ATTR_KEY_THRESHOLD.
+ * @NL80211_CMD_NEW_KEY: add a key with given %NL80211_ATTR_KEY_DATA,
+ *	%NL80211_ATTR_KEY_IDX, %NL80211_ATTR_MAC and %NL80211_ATTR_KEY_CIPHER
+ *	attributes.
+ * @NL80211_CMD_DEL_KEY: delete a key identified by %NL80211_ATTR_KEY_IDX
+ *	or %NL80211_ATTR_MAC.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -54,6 +64,11 @@ enum nl80211_commands {
 	NL80211_CMD_NEW_INTERFACE,
 	NL80211_CMD_DEL_INTERFACE,
 
+	NL80211_CMD_GET_KEY,
+	NL80211_CMD_SET_KEY,
+	NL80211_CMD_NEW_KEY,
+	NL80211_CMD_DEL_KEY,
+
 	/* add commands here */
 
 	/* used to define NL80211_CMD_MAX below */
@@ -75,6 +90,17 @@ enum nl80211_commands {
  * @NL80211_ATTR_IFNAME: network interface name
  * @NL80211_ATTR_IFTYPE: type of virtual interface, see &enum nl80211_iftype
  *
+ * @NL80211_ATTR_MAC: MAC address (various uses)
+ *
+ * @NL80211_ATTR_KEY_DATA: (temporal) key data; for TKIP this consists of
+ *	16 bytes encryption key followed by 8 bytes each for TX and RX MIC
+ *	keys
+ * @NL80211_ATTR_KEY_IDX: key ID (u8, 0-3)
+ * @NL80211_ATTR_KEY_CIPHER: key cipher suite (u32, as defined by IEEE 802.11
+ *	section 7.3.2.25.1, e.g. 0x000FAC04)
+ * @NL80211_ATTR_KEY_SEQ: transmit key sequence number (IV/PN) for TKIP and
+ *	CCMP keys, each six bytes in little endian
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -89,6 +115,14 @@ enum nl80211_attrs {
 	NL80211_ATTR_IFNAME,
 	NL80211_ATTR_IFTYPE,
 
+	NL80211_ATTR_MAC,
+
+	NL80211_ATTR_KEY_DATA,
+	NL80211_ATTR_KEY_IDX,
+	NL80211_ATTR_KEY_CIPHER,
+	NL80211_ATTR_KEY_SEQ,
+	NL80211_ATTR_KEY_DEFAULT,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index d30960e1755c..3db7dfa53b6f 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -49,6 +49,26 @@ extern int ieee80211_radiotap_iterator_next(
    struct ieee80211_radiotap_iterator *iterator);
 
 
+ /**
+ * struct key_params - key information
+ *
+ * Information about a key
+ *
+ * @key: key material
+ * @key_len: length of key material
+ * @cipher: cipher suite selector
+ * @seq: sequence counter (IV/PN) for TKIP and CCMP keys, only used
+ *	with the get_key() callback, must be in little endian,
+ *	length given by @seq_len.
+ */
+struct key_params {
+	u8 *key;
+	u8 *seq;
+	int key_len;
+	int seq_len;
+	u32 cipher;
+};
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -71,6 +91,18 @@ struct wiphy;
  *
  * @change_virtual_intf: change type of virtual interface
  *
+ * @add_key: add a key with the given parameters. @mac_addr will be %NULL
+ *	when adding a group key.
+ *
+ * @get_key: get information about the key with the given parameters.
+ *	@mac_addr will be %NULL when requesting information for a group
+ *	key. All pointers given to the @callback function need not be valid
+ *	after it returns.
+ *
+ * @del_key: remove a key given the @mac_addr (%NULL for a group key)
+ *	and @key_index
+ *
+ * @set_default_key: set the default key on an interface
  */
 struct cfg80211_ops {
 	int	(*add_virtual_intf)(struct wiphy *wiphy, char *name,
@@ -78,6 +110,18 @@ struct cfg80211_ops {
 	int	(*del_virtual_intf)(struct wiphy *wiphy, int ifindex);
 	int	(*change_virtual_intf)(struct wiphy *wiphy, int ifindex,
 				       enum nl80211_iftype type);
+
+	int	(*add_key)(struct wiphy *wiphy, struct net_device *netdev,
+			   u8 key_index, u8 *mac_addr,
+			   struct key_params *params);
+	int	(*get_key)(struct wiphy *wiphy, struct net_device *netdev,
+			   u8 key_index, u8 *mac_addr, void *cookie,
+			   void (*callback)(void *cookie, struct key_params*));
+	int	(*del_key)(struct wiphy *wiphy, struct net_device *netdev,
+			   u8 key_index, u8 *mac_addr);
+	int	(*set_default_key)(struct wiphy *wiphy,
+				   struct net_device *netdev,
+				   u8 key_index);
 };
 
 #endif /* __NET_CFG80211_H */
diff --git a/net/wireless/core.c b/net/wireless/core.c
index febc33bc9c09..cfc5fc5f9e75 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -184,6 +184,9 @@ struct wiphy *wiphy_new(struct cfg80211_ops *ops, int sizeof_priv)
 	struct cfg80211_registered_device *drv;
 	int alloc_size;
 
+	WARN_ON(!ops->add_key && ops->del_key);
+	WARN_ON(ops->add_key && !ops->del_key);
+
 	alloc_size = sizeof(*drv) + sizeof_priv;
 
 	drv = kzalloc(alloc_size, GFP_KERNEL);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 48b0d453e4e1..090936388528 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -61,6 +61,14 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = {
 	[NL80211_ATTR_IFTYPE] = { .type = NLA_U32 },
 	[NL80211_ATTR_IFINDEX] = { .type = NLA_U32 },
 	[NL80211_ATTR_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ-1 },
+
+	[NL80211_ATTR_MAC] = { .type = NLA_BINARY, .len = ETH_ALEN },
+
+	[NL80211_ATTR_KEY_DATA] = { .type = NLA_BINARY,
+				    .len = WLAN_MAX_KEY_LEN },
+	[NL80211_ATTR_KEY_IDX] = { .type = NLA_U8 },
+	[NL80211_ATTR_KEY_CIPHER] = { .type = NLA_U32 },
+	[NL80211_ATTR_KEY_DEFAULT] = { .type = NLA_FLAG },
 };
 
 /* message building helper */
@@ -335,6 +343,263 @@ static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info)
 	return err;
 }
 
+struct get_key_cookie {
+	struct sk_buff *msg;
+	int error;
+};
+
+static void get_key_callback(void *c, struct key_params *params)
+{
+	struct get_key_cookie *cookie = c;
+
+	if (params->key)
+		NLA_PUT(cookie->msg, NL80211_ATTR_KEY_DATA,
+			params->key_len, params->key);
+
+	if (params->seq)
+		NLA_PUT(cookie->msg, NL80211_ATTR_KEY_SEQ,
+			params->seq_len, params->seq);
+
+	if (params->cipher)
+		NLA_PUT_U32(cookie->msg, NL80211_ATTR_KEY_CIPHER,
+			    params->cipher);
+
+	return;
+ nla_put_failure:
+	cookie->error = 1;
+}
+
+static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 key_idx = 0;
+	u8 *mac_addr = NULL;
+	struct get_key_cookie cookie = {
+		.error = 0,
+	};
+	void *hdr;
+	struct sk_buff *msg;
+
+	if (info->attrs[NL80211_ATTR_KEY_IDX])
+		key_idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
+
+	if (key_idx > 3)
+		return -EINVAL;
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->get_key) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!msg) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0,
+			     NL80211_CMD_NEW_KEY);
+
+	if (IS_ERR(hdr)) {
+		err = PTR_ERR(hdr);
+		goto out;
+	}
+
+	cookie.msg = msg;
+
+	NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex);
+	NLA_PUT_U8(msg, NL80211_ATTR_KEY_IDX, key_idx);
+	if (mac_addr)
+		NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr);
+
+	rtnl_lock();
+	err = drv->ops->get_key(&drv->wiphy, dev, key_idx, mac_addr,
+				&cookie, get_key_callback);
+	rtnl_unlock();
+
+	if (err)
+		goto out;
+
+	if (cookie.error)
+		goto nla_put_failure;
+
+	genlmsg_end(msg, hdr);
+	err = genlmsg_unicast(msg, info->snd_pid);
+	goto out;
+
+ nla_put_failure:
+	err = -ENOBUFS;
+	nlmsg_free(msg);
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_set_key(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 key_idx;
+
+	if (!info->attrs[NL80211_ATTR_KEY_IDX])
+		return -EINVAL;
+
+	key_idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
+
+	if (key_idx > 3)
+		return -EINVAL;
+
+	/* currently only support setting default key */
+	if (!info->attrs[NL80211_ATTR_KEY_DEFAULT])
+		return -EINVAL;
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->set_default_key) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->set_default_key(&drv->wiphy, dev, key_idx);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct key_params params;
+	u8 key_idx = 0;
+	u8 *mac_addr = NULL;
+
+	memset(&params, 0, sizeof(params));
+
+	if (!info->attrs[NL80211_ATTR_KEY_CIPHER])
+		return -EINVAL;
+
+	if (info->attrs[NL80211_ATTR_KEY_DATA]) {
+		params.key = nla_data(info->attrs[NL80211_ATTR_KEY_DATA]);
+		params.key_len = nla_len(info->attrs[NL80211_ATTR_KEY_DATA]);
+	}
+
+	if (info->attrs[NL80211_ATTR_KEY_IDX])
+		key_idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
+
+	params.cipher = nla_get_u32(info->attrs[NL80211_ATTR_KEY_CIPHER]);
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	if (key_idx > 3)
+		return -EINVAL;
+
+	/*
+	 * Disallow pairwise keys with non-zero index unless it's WEP
+	 * (because current deployments use pairwise WEP keys with
+	 * non-zero indizes but 802.11i clearly specifies to use zero)
+	 */
+	if (mac_addr && key_idx &&
+	    params.cipher != WLAN_CIPHER_SUITE_WEP40 &&
+	    params.cipher != WLAN_CIPHER_SUITE_WEP104)
+		return -EINVAL;
+
+	/* TODO: add definitions for the lengths to linux/ieee80211.h */
+	switch (params.cipher) {
+	case WLAN_CIPHER_SUITE_WEP40:
+		if (params.key_len != 5)
+			return -EINVAL;
+		break;
+	case WLAN_CIPHER_SUITE_TKIP:
+		if (params.key_len != 32)
+			return -EINVAL;
+		break;
+	case WLAN_CIPHER_SUITE_CCMP:
+		if (params.key_len != 16)
+			return -EINVAL;
+		break;
+	case WLAN_CIPHER_SUITE_WEP104:
+		if (params.key_len != 13)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->add_key) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->add_key(&drv->wiphy, dev, key_idx, mac_addr, &params);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 key_idx = 0;
+	u8 *mac_addr = NULL;
+
+	if (info->attrs[NL80211_ATTR_KEY_IDX])
+		key_idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
+
+	if (key_idx > 3)
+		return -EINVAL;
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->del_key) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->del_key(&drv->wiphy, dev, key_idx, mac_addr);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
 static struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_WIPHY,
@@ -374,6 +639,30 @@ static struct genl_ops nl80211_ops[] = {
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
+	{
+		.cmd = NL80211_CMD_GET_KEY,
+		.doit = nl80211_get_key,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_SET_KEY,
+		.doit = nl80211_set_key,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_NEW_KEY,
+		.doit = nl80211_new_key,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_DEL_KEY,
+		.doit = nl80211_del_key,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
 };
 
 /* multicast groups */
-- 
cgit v1.2.3


From 62da92fb75c346b503bca765fd1337e08771c9fe Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 02:03:31 +0100
Subject: mac80211: support getting key sequence counters via cfg80211

This implements cfg80211's get_key() to allow retrieving the sequence
counter for a TKIP or CCMP key from userspace. It also cleans up and
documents the associated low-level driver interface.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h | 14 +++------
 net/mac80211/cfg.c     | 85 +++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 89 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5b9e7a262448..a762a75ae62d 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -647,9 +647,6 @@ struct ieee80211_key_conf {
 	u8 key[0];
 };
 
-#define IEEE80211_SEQ_COUNTER_RX	0
-#define IEEE80211_SEQ_COUNTER_TX	1
-
 /**
  * enum set_key_cmd - key command
  *
@@ -996,9 +993,9 @@ enum ieee80211_erp_change_flags {
  *
  * @get_stats: return low-level statistics
  *
- * @get_sequence_counter: For devices that have internal sequence counters this
- *	callback allows mac80211 to access the current value of a counter.
- *	This callback seems not well-defined, tell us if you need it.
+ * @get_tkip_seq: If your device implements TKIP encryption in hardware this
+ *	callback should be provided to read the TKIP transmit IVs (both IV32
+ *	and IV16) for the given key from hardware.
  *
  * @set_rts_threshold: Configuration of RTS threshold (if device needs it)
  *
@@ -1073,9 +1070,8 @@ struct ieee80211_ops {
 	int (*hw_scan)(struct ieee80211_hw *hw, u8 *ssid, size_t len);
 	int (*get_stats)(struct ieee80211_hw *hw,
 			 struct ieee80211_low_level_stats *stats);
-	int (*get_sequence_counter)(struct ieee80211_hw *hw,
-				    u8* addr, u8 keyidx, u8 txrx,
-				    u32* iv32, u16* iv16);
+	void (*get_tkip_seq)(struct ieee80211_hw *hw, u8 hw_key_idx,
+			     u32 *iv32, u16 *iv16);
 	int (*set_rts_threshold)(struct ieee80211_hw *hw, u32 value);
 	int (*set_frag_threshold)(struct ieee80211_hw *hw, u32 value);
 	int (*set_retry_limit)(struct ieee80211_hw *hw,
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index d49f7b58b05c..4c1ce353c662 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1,7 +1,7 @@
 /*
  * mac80211 configuration hooks for cfg80211
  *
- * Copyright 2006	Johannes Berg <johannes@sipsolutions.net>
+ * Copyright 2006, 2007	Johannes Berg <johannes@sipsolutions.net>
  *
  * This file is GPLv2 as found in COPYING.
  */
@@ -175,6 +175,88 @@ static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev,
 	return 0;
 }
 
+static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
+			     u8 key_idx, u8 *mac_addr, void *cookie,
+			     void (*callback)(void *cookie,
+					      struct key_params *params))
+{
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct sta_info *sta = NULL;
+	u8 seq[6] = {0};
+	struct key_params params;
+	struct ieee80211_key *key;
+	u32 iv32;
+	u16 iv16;
+	int err = -ENOENT;
+
+	if (mac_addr) {
+		sta = sta_info_get(sdata->local, mac_addr);
+		if (!sta)
+			goto out;
+
+		key = sta->key;
+	} else
+		key = sdata->keys[key_idx];
+
+	if (!key)
+		goto out;
+
+	memset(&params, 0, sizeof(params));
+
+	switch (key->conf.alg) {
+	case ALG_TKIP:
+		params.cipher = WLAN_CIPHER_SUITE_TKIP;
+
+		iv32 = key->u.tkip.iv32;
+		iv16 = key->u.tkip.iv16;
+
+		if (key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE &&
+		    sdata->local->ops->get_tkip_seq)
+			sdata->local->ops->get_tkip_seq(
+				local_to_hw(sdata->local),
+				key->conf.hw_key_idx,
+				&iv32, &iv16);
+
+		seq[0] = iv16 & 0xff;
+		seq[1] = (iv16 >> 8) & 0xff;
+		seq[2] = iv32 & 0xff;
+		seq[3] = (iv32 >> 8) & 0xff;
+		seq[4] = (iv32 >> 16) & 0xff;
+		seq[5] = (iv32 >> 24) & 0xff;
+		params.seq = seq;
+		params.seq_len = 6;
+		break;
+	case ALG_CCMP:
+		params.cipher = WLAN_CIPHER_SUITE_CCMP;
+		seq[0] = key->u.ccmp.tx_pn[5];
+		seq[1] = key->u.ccmp.tx_pn[4];
+		seq[2] = key->u.ccmp.tx_pn[3];
+		seq[3] = key->u.ccmp.tx_pn[2];
+		seq[4] = key->u.ccmp.tx_pn[1];
+		seq[5] = key->u.ccmp.tx_pn[0];
+		params.seq = seq;
+		params.seq_len = 6;
+		break;
+	case ALG_WEP:
+		if (key->conf.keylen == 5)
+			params.cipher = WLAN_CIPHER_SUITE_WEP40;
+		else
+			params.cipher = WLAN_CIPHER_SUITE_WEP104;
+		break;
+	}
+
+	params.key = key->conf.key;
+	params.key_len = key->conf.keylen;
+
+	callback(cookie, &params);
+	err = 0;
+
+ out:
+	if (sta)
+		sta_info_put(sta);
+	return err;
+}
+
 static int ieee80211_config_default_key(struct wiphy *wiphy,
 					struct net_device *dev,
 					u8 key_idx)
@@ -193,5 +275,6 @@ struct cfg80211_ops mac80211_config_ops = {
 	.change_virtual_intf = ieee80211_change_iface,
 	.add_key = ieee80211_add_key,
 	.del_key = ieee80211_del_key,
+	.get_key = ieee80211_get_key,
 	.set_default_key = ieee80211_config_default_key,
 };
-- 
cgit v1.2.3


From ed1b6cc7f80f831e192704b05b9917f9cc37be15 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 02:03:32 +0100
Subject: cfg80211/nl80211: add beacon settings

This adds the necessary API to cfg80211/nl80211 to allow
changing beaconing settings.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/nl80211.h |  24 +++++++++
 include/net/cfg80211.h  |  33 ++++++++++++
 net/wireless/nl80211.c  | 133 ++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 190 insertions(+)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 8dc807d9c29a..f1e455a8b4de 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -47,6 +47,15 @@
  * @NL80211_CMD_DEL_KEY: delete a key identified by %NL80211_ATTR_KEY_IDX
  *	or %NL80211_ATTR_MAC.
  *
+ * @NL80211_CMD_GET_BEACON: retrieve beacon information (returned in a
+ *	%NL80222_CMD_NEW_BEACON message)
+ * @NL80211_CMD_SET_BEACON: set the beacon on an access point interface
+ *	using the %NL80211_ATTR_BEACON_INTERVAL, %NL80211_ATTR_DTIM_PERIOD,
+ *	%NL80211_BEACON_HEAD and %NL80211_BEACON_TAIL attributes.
+ * @NL80211_CMD_NEW_BEACON: add a new beacon to an access point interface,
+ *	parameters are like for %NL80211_CMD_SET_BEACON.
+ * @NL80211_CMD_DEL_BEACON: remove the beacon, stop sending it
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -69,6 +78,11 @@ enum nl80211_commands {
 	NL80211_CMD_NEW_KEY,
 	NL80211_CMD_DEL_KEY,
 
+	NL80211_CMD_GET_BEACON,
+	NL80211_CMD_SET_BEACON,
+	NL80211_CMD_NEW_BEACON,
+	NL80211_CMD_DEL_BEACON,
+
 	/* add commands here */
 
 	/* used to define NL80211_CMD_MAX below */
@@ -101,6 +115,11 @@ enum nl80211_commands {
  * @NL80211_ATTR_KEY_SEQ: transmit key sequence number (IV/PN) for TKIP and
  *	CCMP keys, each six bytes in little endian
  *
+ * @NL80211_ATTR_BEACON_INTERVAL: beacon interval in TU
+ * @NL80211_ATTR_DTIM_PERIOD: DTIM period for beaconing
+ * @NL80211_ATTR_BEACON_HEAD: portion of the beacon before the TIM IE
+ * @NL80211_ATTR_BEACON_TAIL: portion of the beacon after the TIM IE
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -123,6 +142,11 @@ enum nl80211_attrs {
 	NL80211_ATTR_KEY_SEQ,
 	NL80211_ATTR_KEY_DEFAULT,
 
+	NL80211_ATTR_BEACON_INTERVAL,
+	NL80211_ATTR_DTIM_PERIOD,
+	NL80211_ATTR_BEACON_HEAD,
+	NL80211_ATTR_BEACON_TAIL,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 3db7dfa53b6f..fc94852e967b 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -69,6 +69,26 @@ struct key_params {
 	u32 cipher;
 };
 
+/**
+ * struct beacon_parameters - beacon parameters
+ *
+ * Used to configure the beacon for an interface.
+ *
+ * @head: head portion of beacon (before TIM IE)
+ *     or %NULL if not changed
+ * @tail: tail portion of beacon (after TIM IE)
+ *     or %NULL if not changed
+ * @interval: beacon interval or zero if not changed
+ * @dtim_period: DTIM period or zero if not changed
+ * @head_len: length of @head
+ * @tail_len: length of @tail
+ */
+struct beacon_parameters {
+	u8 *head, *tail;
+	int interval, dtim_period;
+	int head_len, tail_len;
+};
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -103,6 +123,13 @@ struct wiphy;
  *	and @key_index
  *
  * @set_default_key: set the default key on an interface
+ *
+ * @add_beacon: Add a beacon with given parameters, @head, @interval
+ *	and @dtim_period will be valid, @tail is optional.
+ * @set_beacon: Change the beacon parameters for an access point mode
+ *	interface. This should reject the call when no beacon has been
+ *	configured.
+ * @del_beacon: Remove beacon configuration and stop sending the beacon.
  */
 struct cfg80211_ops {
 	int	(*add_virtual_intf)(struct wiphy *wiphy, char *name,
@@ -122,6 +149,12 @@ struct cfg80211_ops {
 	int	(*set_default_key)(struct wiphy *wiphy,
 				   struct net_device *netdev,
 				   u8 key_index);
+
+	int	(*add_beacon)(struct wiphy *wiphy, struct net_device *dev,
+			      struct beacon_parameters *info);
+	int	(*set_beacon)(struct wiphy *wiphy, struct net_device *dev,
+			      struct beacon_parameters *info);
+	int	(*del_beacon)(struct wiphy *wiphy, struct net_device *dev);
 };
 
 #endif /* __NET_CFG80211_H */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 090936388528..306ae019ea81 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -69,6 +69,13 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = {
 	[NL80211_ATTR_KEY_IDX] = { .type = NLA_U8 },
 	[NL80211_ATTR_KEY_CIPHER] = { .type = NLA_U32 },
 	[NL80211_ATTR_KEY_DEFAULT] = { .type = NLA_FLAG },
+
+	[NL80211_ATTR_BEACON_INTERVAL] = { .type = NLA_U32 },
+	[NL80211_ATTR_DTIM_PERIOD] = { .type = NLA_U32 },
+	[NL80211_ATTR_BEACON_HEAD] = { .type = NLA_BINARY,
+				       .len = IEEE80211_MAX_DATA_LEN },
+	[NL80211_ATTR_BEACON_TAIL] = { .type = NLA_BINARY,
+				       .len = IEEE80211_MAX_DATA_LEN },
 };
 
 /* message building helper */
@@ -600,6 +607,114 @@ static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info)
 	return err;
 }
 
+static int nl80211_addset_beacon(struct sk_buff *skb, struct genl_info *info)
+{
+        int (*call)(struct wiphy *wiphy, struct net_device *dev,
+		    struct beacon_parameters *info);
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct beacon_parameters params;
+	int haveinfo = 0;
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	switch (info->genlhdr->cmd) {
+	case NL80211_CMD_NEW_BEACON:
+		/* these are required for NEW_BEACON */
+		if (!info->attrs[NL80211_ATTR_BEACON_INTERVAL] ||
+		    !info->attrs[NL80211_ATTR_DTIM_PERIOD] ||
+		    !info->attrs[NL80211_ATTR_BEACON_HEAD]) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		call = drv->ops->add_beacon;
+		break;
+	case NL80211_CMD_SET_BEACON:
+		call = drv->ops->set_beacon;
+		break;
+	default:
+		WARN_ON(1);
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	if (!call) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	memset(&params, 0, sizeof(params));
+
+	if (info->attrs[NL80211_ATTR_BEACON_INTERVAL]) {
+		params.interval =
+		    nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]);
+		haveinfo = 1;
+	}
+
+	if (info->attrs[NL80211_ATTR_DTIM_PERIOD]) {
+		params.dtim_period =
+		    nla_get_u32(info->attrs[NL80211_ATTR_DTIM_PERIOD]);
+		haveinfo = 1;
+	}
+
+	if (info->attrs[NL80211_ATTR_BEACON_HEAD]) {
+		params.head = nla_data(info->attrs[NL80211_ATTR_BEACON_HEAD]);
+		params.head_len =
+		    nla_len(info->attrs[NL80211_ATTR_BEACON_HEAD]);
+		haveinfo = 1;
+	}
+
+	if (info->attrs[NL80211_ATTR_BEACON_TAIL]) {
+		params.tail = nla_data(info->attrs[NL80211_ATTR_BEACON_TAIL]);
+		params.tail_len =
+		    nla_len(info->attrs[NL80211_ATTR_BEACON_TAIL]);
+		haveinfo = 1;
+	}
+
+	if (!haveinfo) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = call(&drv->wiphy, dev, &params);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_del_beacon(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->del_beacon) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->del_beacon(&drv->wiphy, dev);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
 static struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_WIPHY,
@@ -663,6 +778,24 @@ static struct genl_ops nl80211_ops[] = {
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
+	{
+		.cmd = NL80211_CMD_SET_BEACON,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+		.doit = nl80211_addset_beacon,
+	},
+	{
+		.cmd = NL80211_CMD_NEW_BEACON,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+		.doit = nl80211_addset_beacon,
+	},
+	{
+		.cmd = NL80211_CMD_DEL_BEACON,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+		.doit = nl80211_del_beacon,
+	},
 };
 
 /* multicast groups */
-- 
cgit v1.2.3


From 5727ef1b2e797a1922f5bc239b6afb2b4cfb80bc Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 02:03:34 +0100
Subject: cfg80211/nl80211: station handling

This patch adds station handling to cfg80211/nl80211.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/nl80211.h |  68 ++++++++++++++
 include/net/cfg80211.h  |  55 ++++++++++++
 net/wireless/nl80211.c  | 235 ++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 358 insertions(+)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index f1e455a8b4de..85e2d7d1f9e3 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -6,6 +6,18 @@
  * Copyright 2006, 2007 Johannes Berg <johannes@sipsolutions.net>
  */
 
+/**
+ * DOC: Station handling
+ *
+ * Stations are added per interface, but a special case exists with VLAN
+ * interfaces. When a station is bound to an AP interface, it may be moved
+ * into a VLAN identified by a VLAN interface index (%NL80211_ATTR_STA_VLAN).
+ * The station is still assumed to belong to the AP interface it was added
+ * to.
+ *
+ * TODO: need more info?
+ */
+
 /**
  * enum nl80211_commands - supported nl80211 commands
  *
@@ -56,6 +68,16 @@
  *	parameters are like for %NL80211_CMD_SET_BEACON.
  * @NL80211_CMD_DEL_BEACON: remove the beacon, stop sending it
  *
+ * @NL80211_CMD_GET_STATION: Get station attributes for station identified by
+ *	%NL80211_ATTR_MAC on the interface identified by %NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_SET_STATION: Set station attributes for station identified by
+ *	%NL80211_ATTR_MAC on the interface identified by %NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_NEW_STATION: Add a station with given attributes to the
+ *	the interface identified by %NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_DEL_STATION: Remove a station identified by %NL80211_ATTR_MAC
+ *	or, if no MAC address given, all stations, on the interface identified
+ *	by %NL80211_ATTR_IFINDEX.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -83,6 +105,11 @@ enum nl80211_commands {
 	NL80211_CMD_NEW_BEACON,
 	NL80211_CMD_DEL_BEACON,
 
+	NL80211_CMD_GET_STATION,
+	NL80211_CMD_SET_STATION,
+	NL80211_CMD_NEW_STATION,
+	NL80211_CMD_DEL_STATION,
+
 	/* add commands here */
 
 	/* used to define NL80211_CMD_MAX below */
@@ -120,6 +147,17 @@ enum nl80211_commands {
  * @NL80211_ATTR_BEACON_HEAD: portion of the beacon before the TIM IE
  * @NL80211_ATTR_BEACON_TAIL: portion of the beacon after the TIM IE
  *
+ * @NL80211_ATTR_STA_AID: Association ID for the station (u16)
+ * @NL80211_ATTR_STA_FLAGS: flags, nested element with NLA_FLAG attributes of
+ *	&enum nl80211_sta_flags.
+ * @NL80211_ATTR_STA_LISTEN_INTERVAL: listen interval as defined by
+ *	IEEE 802.11 7.3.1.6 (u16).
+ * @NL80211_ATTR_STA_SUPPORTED_RATES: supported rates, array of supported
+ *	rates as defined by IEEE 802.11 7.3.2.2 but without the length
+ *	restriction (at most %NL80211_MAX_SUPP_RATES).
+ * @NL80211_ATTR_STA_VLAN: interface index of VLAN interface to move station
+ *	to, or the AP interface the station was originally added to to.
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -147,12 +185,20 @@ enum nl80211_attrs {
 	NL80211_ATTR_BEACON_HEAD,
 	NL80211_ATTR_BEACON_TAIL,
 
+	NL80211_ATTR_STA_AID,
+	NL80211_ATTR_STA_FLAGS,
+	NL80211_ATTR_STA_LISTEN_INTERVAL,
+	NL80211_ATTR_STA_SUPPORTED_RATES,
+	NL80211_ATTR_STA_VLAN,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
 
+#define NL80211_MAX_SUPP_RATES	32
+
 /**
  * enum nl80211_iftype - (virtual) interface types
  *
@@ -184,4 +230,26 @@ enum nl80211_iftype {
 	NL80211_IFTYPE_MAX = __NL80211_IFTYPE_AFTER_LAST - 1
 };
 
+/**
+ * enum nl80211_sta_flags - station flags
+ *
+ * Station flags. When a station is added to an AP interface, it is
+ * assumed to be already associated (and hence authenticated.)
+ *
+ * @NL80211_STA_FLAG_AUTHORIZED: station is authorized (802.1X)
+ * @NL80211_STA_FLAG_SHORT_PREAMBLE: station is capable of receiving frames
+ *	with short barker preamble
+ * @NL80211_STA_FLAG_WME: station is WME/QoS capable
+ */
+enum nl80211_sta_flags {
+	__NL80211_STA_FLAG_INVALID,
+	NL80211_STA_FLAG_AUTHORIZED,
+	NL80211_STA_FLAG_SHORT_PREAMBLE,
+	NL80211_STA_FLAG_WME,
+
+	/* keep last */
+	__NL80211_STA_FLAG_AFTER_LAST,
+	NL80211_STA_FLAG_MAX = __NL80211_STA_FLAG_AFTER_LAST - 1
+};
+
 #endif /* __LINUX_NL80211_H */
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index fc94852e967b..df650935e268 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -89,6 +89,47 @@ struct beacon_parameters {
 	int head_len, tail_len;
 };
 
+/**
+ * enum station_flags - station flags
+ *
+ * Station capability flags. Note that these must be the bits
+ * according to the nl80211 flags.
+ *
+ * @STATION_FLAG_CHANGED: station flags were changed
+ * @STATION_FLAG_AUTHORIZED: station is authorized to send frames (802.1X)
+ * @STATION_FLAG_SHORT_PREAMBLE: station is capable of receiving frames
+ *	with short preambles
+ * @STATION_FLAG_WME: station is WME/QoS capable
+ */
+enum station_flags {
+	STATION_FLAG_CHANGED		= 1<<0,
+	STATION_FLAG_AUTHORIZED		= 1<<NL80211_STA_FLAG_AUTHORIZED,
+	STATION_FLAG_SHORT_PREAMBLE	= 1<<NL80211_STA_FLAG_SHORT_PREAMBLE,
+	STATION_FLAG_WME		= 1<<NL80211_STA_FLAG_WME,
+};
+
+/**
+ * struct station_parameters - station parameters
+ *
+ * Used to change and create a new station.
+ *
+ * @vlan: vlan interface station should belong to
+ * @supported_rates: supported rates in IEEE 802.11 format
+ *	(or NULL for no change)
+ * @supported_rates_len: number of supported rates
+ * @station_flags: station flags (see &enum station_flags)
+ * @listen_interval: listen interval or -1 for no change
+ * @aid: AID or zero for no change
+ */
+struct station_parameters {
+	u8 *supported_rates;
+	struct net_device *vlan;
+	u32 station_flags;
+	int listen_interval;
+	u16 aid;
+	u8 supported_rates_len;
+};
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -130,6 +171,12 @@ struct wiphy;
  *	interface. This should reject the call when no beacon has been
  *	configured.
  * @del_beacon: Remove beacon configuration and stop sending the beacon.
+ *
+ * @add_station: Add a new station.
+ *
+ * @del_station: Remove a station; @mac may be NULL to remove all stations.
+ *
+ * @change_station: Modify a given station.
  */
 struct cfg80211_ops {
 	int	(*add_virtual_intf)(struct wiphy *wiphy, char *name,
@@ -155,6 +202,14 @@ struct cfg80211_ops {
 	int	(*set_beacon)(struct wiphy *wiphy, struct net_device *dev,
 			      struct beacon_parameters *info);
 	int	(*del_beacon)(struct wiphy *wiphy, struct net_device *dev);
+
+
+	int	(*add_station)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *mac, struct station_parameters *params);
+	int	(*del_station)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *mac);
+	int	(*change_station)(struct wiphy *wiphy, struct net_device *dev,
+				  u8 *mac, struct station_parameters *params);
 };
 
 #endif /* __NET_CFG80211_H */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 306ae019ea81..431835126e88 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -76,6 +76,12 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = {
 				       .len = IEEE80211_MAX_DATA_LEN },
 	[NL80211_ATTR_BEACON_TAIL] = { .type = NLA_BINARY,
 				       .len = IEEE80211_MAX_DATA_LEN },
+	[NL80211_ATTR_STA_AID] = { .type = NLA_U16 },
+	[NL80211_ATTR_STA_FLAGS] = { .type = NLA_NESTED },
+	[NL80211_ATTR_STA_LISTEN_INTERVAL] = { .type = NLA_U16 },
+	[NL80211_ATTR_STA_SUPPORTED_RATES] = { .type = NLA_BINARY,
+					       .len = NL80211_MAX_SUPP_RATES },
+	[NL80211_ATTR_STA_VLAN] = { .type = NLA_U32 },
 };
 
 /* message building helper */
@@ -715,6 +721,210 @@ static int nl80211_del_beacon(struct sk_buff *skb, struct genl_info *info)
 	return err;
 }
 
+static const struct nla_policy sta_flags_policy[NL80211_STA_FLAG_MAX + 1] = {
+	[NL80211_STA_FLAG_AUTHORIZED] = { .type = NLA_FLAG },
+	[NL80211_STA_FLAG_SHORT_PREAMBLE] = { .type = NLA_FLAG },
+	[NL80211_STA_FLAG_WME] = { .type = NLA_FLAG },
+};
+
+static int parse_station_flags(struct nlattr *nla, u32 *staflags)
+{
+	struct nlattr *flags[NL80211_STA_FLAG_MAX + 1];
+	int flag;
+
+	*staflags = 0;
+
+	if (!nla)
+		return 0;
+
+	if (nla_parse_nested(flags, NL80211_STA_FLAG_MAX,
+			     nla, sta_flags_policy))
+		return -EINVAL;
+
+	*staflags = STATION_FLAG_CHANGED;
+
+	for (flag = 1; flag <= NL80211_STA_FLAG_MAX; flag++)
+		if (flags[flag])
+			*staflags |= (1<<flag);
+
+	return 0;
+}
+
+static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info)
+{
+	return -EOPNOTSUPP;
+}
+
+/*
+ * Get vlan interface making sure it is on the right wiphy.
+ */
+static int get_vlan(struct nlattr *vlanattr,
+		    struct cfg80211_registered_device *rdev,
+		    struct net_device **vlan)
+{
+	*vlan = NULL;
+
+	if (vlanattr) {
+		*vlan = dev_get_by_index(&init_net, nla_get_u32(vlanattr));
+		if (!*vlan)
+			return -ENODEV;
+		if (!(*vlan)->ieee80211_ptr)
+			return -EINVAL;
+		if ((*vlan)->ieee80211_ptr->wiphy != &rdev->wiphy)
+			return -EINVAL;
+	}
+	return 0;
+}
+
+static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct station_parameters params;
+	u8 *mac_addr = NULL;
+
+	memset(&params, 0, sizeof(params));
+
+	params.listen_interval = -1;
+
+	if (info->attrs[NL80211_ATTR_STA_AID])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	if (info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]) {
+		params.supported_rates =
+			nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
+		params.supported_rates_len =
+			nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
+	}
+
+	if (info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL])
+		params.listen_interval =
+		    nla_get_u16(info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL]);
+
+	if (parse_station_flags(info->attrs[NL80211_ATTR_STA_FLAGS],
+				&params.station_flags))
+		return -EINVAL;
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
+	if (err)
+		goto out;
+
+	if (!drv->ops->change_station) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->change_station(&drv->wiphy, dev, mac_addr, &params);
+	rtnl_unlock();
+
+ out:
+	if (params.vlan)
+		dev_put(params.vlan);
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct station_parameters params;
+	u8 *mac_addr = NULL;
+
+	memset(&params, 0, sizeof(params));
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_STA_AID])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES])
+		return -EINVAL;
+
+	mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+	params.supported_rates =
+		nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
+	params.supported_rates_len =
+		nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
+	params.listen_interval =
+		nla_get_u16(info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL]);
+	params.listen_interval = nla_get_u16(info->attrs[NL80211_ATTR_STA_AID]);
+
+	if (parse_station_flags(info->attrs[NL80211_ATTR_STA_FLAGS],
+				&params.station_flags))
+		return -EINVAL;
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
+	if (err)
+		goto out;
+
+	if (!drv->ops->add_station) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->add_station(&drv->wiphy, dev, mac_addr, &params);
+	rtnl_unlock();
+
+ out:
+	if (params.vlan)
+		dev_put(params.vlan);
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 *mac_addr = NULL;
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->del_station) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->del_station(&drv->wiphy, dev, mac_addr);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
 static struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_WIPHY,
@@ -796,6 +1006,31 @@ static struct genl_ops nl80211_ops[] = {
 		.flags = GENL_ADMIN_PERM,
 		.doit = nl80211_del_beacon,
 	},
+	{
+		.cmd = NL80211_CMD_GET_STATION,
+		.doit = nl80211_get_station,
+		/* TODO: implement dumpit */
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_SET_STATION,
+		.doit = nl80211_set_station,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_NEW_STATION,
+		.doit = nl80211_new_station,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_DEL_STATION,
+		.doit = nl80211_del_station,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
 };
 
 /* multicast groups */
-- 
cgit v1.2.3


From fd5b74dcb88cfc109d6576b22deaef6f47f82c12 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 02:03:36 +0100
Subject: cfg80211/nl80211: implement station attribute retrieval

After a station is added to the kernel's structures, userspace
has to be able to retrieve statistics about that station, especially
whether the station was idle and how much bytes were transferred
to and from it. This adds the necessary code to nl80211.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/nl80211.h | 28 +++++++++++++++++
 include/net/cfg80211.h  | 35 +++++++++++++++++++++
 net/wireless/nl80211.c  | 82 ++++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 144 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 85e2d7d1f9e3..9fecf902419c 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -157,6 +157,9 @@ enum nl80211_commands {
  *	restriction (at most %NL80211_MAX_SUPP_RATES).
  * @NL80211_ATTR_STA_VLAN: interface index of VLAN interface to move station
  *	to, or the AP interface the station was originally added to to.
+ * @NL80211_ATTR_STA_STATS: statistics for a station, part of station info
+ *	given for %NL80211_CMD_GET_STATION, nested attribute containing
+ *	info as possible, see &enum nl80211_sta_stats.
  *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
@@ -190,6 +193,7 @@ enum nl80211_attrs {
 	NL80211_ATTR_STA_LISTEN_INTERVAL,
 	NL80211_ATTR_STA_SUPPORTED_RATES,
 	NL80211_ATTR_STA_VLAN,
+	NL80211_ATTR_STA_STATS,
 
 	/* add attributes here, update the policy in nl80211.c */
 
@@ -252,4 +256,28 @@ enum nl80211_sta_flags {
 	NL80211_STA_FLAG_MAX = __NL80211_STA_FLAG_AFTER_LAST - 1
 };
 
+/**
+ * enum nl80211_sta_stats - station statistics
+ *
+ * These attribute types are used with %NL80211_ATTR_STA_STATS
+ * when getting information about a station.
+ *
+ * @__NL80211_STA_STAT_INVALID: attribute number 0 is reserved
+ * @NL80211_STA_STAT_INACTIVE_TIME: time since last activity (u32, msecs)
+ * @NL80211_STA_STAT_RX_BYTES: total received bytes (u32, from this station)
+ * @NL80211_STA_STAT_TX_BYTES: total transmitted bytes (u32, to this station)
+ * @__NL80211_STA_STAT_AFTER_LAST: internal
+ * @NL80211_STA_STAT_MAX: highest possible station stats attribute
+ */
+enum nl80211_sta_stats {
+	__NL80211_STA_STAT_INVALID,
+	NL80211_STA_STAT_INACTIVE_TIME,
+	NL80211_STA_STAT_RX_BYTES,
+	NL80211_STA_STAT_TX_BYTES,
+
+	/* keep last */
+	__NL80211_STA_STAT_AFTER_LAST,
+	NL80211_STA_STAT_MAX = __NL80211_STA_STAT_AFTER_LAST - 1
+};
+
 #endif /* __LINUX_NL80211_H */
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index df650935e268..bcc480b8892a 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -130,6 +130,39 @@ struct station_parameters {
 	u8 supported_rates_len;
 };
 
+/**
+ * enum station_stats_flags - station statistics flags
+ *
+ * Used by the driver to indicate which info in &struct station_stats
+ * it has filled in during get_station().
+ *
+ * @STATION_STAT_INACTIVE_TIME: @inactive_time filled
+ * @STATION_STAT_RX_BYTES: @rx_bytes filled
+ * @STATION_STAT_TX_BYTES: @tx_bytes filled
+ */
+enum station_stats_flags {
+	STATION_STAT_INACTIVE_TIME	= 1<<0,
+	STATION_STAT_RX_BYTES		= 1<<1,
+	STATION_STAT_TX_BYTES		= 1<<2,
+};
+
+/**
+ * struct station_stats - station statistics
+ *
+ * Station information filled by driver for get_station().
+ *
+ * @filled: bitflag of flags from &enum station_stats_flags
+ * @inactive_time: time since last station activity (tx/rx) in milliseconds
+ * @rx_bytes: bytes received from this station
+ * @tx_bytes: bytes transmitted to this station
+ */
+struct station_stats {
+	u32 filled;
+	u32 inactive_time;
+	u32 rx_bytes;
+	u32 tx_bytes;
+};
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -210,6 +243,8 @@ struct cfg80211_ops {
 			       u8 *mac);
 	int	(*change_station)(struct wiphy *wiphy, struct net_device *dev,
 				  u8 *mac, struct station_parameters *params);
+	int	(*get_station)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *mac, struct station_stats *stats);
 };
 
 #endif /* __NET_CFG80211_H */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 431835126e88..e3a214f63f91 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -750,9 +750,89 @@ static int parse_station_flags(struct nlattr *nla, u32 *staflags)
 	return 0;
 }
 
+static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq,
+				int flags, struct net_device *dev,
+				u8 *mac_addr, struct station_stats *stats)
+{
+	void *hdr;
+	struct nlattr *statsattr;
+
+	hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_STATION);
+	if (!hdr)
+		return -1;
+
+	NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex);
+	NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr);
+
+	statsattr = nla_nest_start(msg, NL80211_ATTR_STA_STATS);
+	if (!statsattr)
+		goto nla_put_failure;
+	if (stats->filled & STATION_STAT_INACTIVE_TIME)
+		NLA_PUT_U32(msg, NL80211_STA_STAT_INACTIVE_TIME,
+			    stats->inactive_time);
+	if (stats->filled & STATION_STAT_RX_BYTES)
+		NLA_PUT_U32(msg, NL80211_STA_STAT_RX_BYTES,
+			    stats->rx_bytes);
+	if (stats->filled & STATION_STAT_TX_BYTES)
+		NLA_PUT_U32(msg, NL80211_STA_STAT_TX_BYTES,
+			    stats->tx_bytes);
+
+	nla_nest_end(msg, statsattr);
+
+	return genlmsg_end(msg, hdr);
+
+ nla_put_failure:
+	return genlmsg_cancel(msg, hdr);
+}
+
+
 static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info)
 {
-	return -EOPNOTSUPP;
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct station_stats stats;
+	struct sk_buff *msg;
+	u8 *mac_addr = NULL;
+
+	memset(&stats, 0, sizeof(stats));
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->get_station) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->get_station(&drv->wiphy, dev, mac_addr, &stats);
+	rtnl_unlock();
+
+	msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!msg)
+		goto out;
+
+	if (nl80211_send_station(msg, info->snd_pid, info->snd_seq, 0,
+				 dev, mac_addr, &stats) < 0)
+		goto out_free;
+
+	err = genlmsg_unicast(msg, info->snd_pid);
+	goto out;
+
+ out_free:
+	nlmsg_free(msg);
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
 }
 
 /*
-- 
cgit v1.2.3


From 9cb5734e5b9b26097c7fa28a9c6426a204cc15e3 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 12 Jan 2008 02:16:03 -0800
Subject: [TCP]: Convert several length variable to unsigned.

Several length variables cannot be negative, so convert int to
unsigned int.  This also allows us to do sane shift operations
on those variables.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 6 ++++--
 net/ipv4/tcp_ipv4.c | 6 +++---
 net/ipv6/tcp_ipv6.c | 8 ++++----
 3 files changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 5ec1cacca8a1..13ebe11a0af7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1155,7 +1155,8 @@ extern int			tcp_v4_calc_md5_hash(char *md5_hash,
 						     struct dst_entry *dst,
 						     struct request_sock *req,
 						     struct tcphdr *th,
-						     int protocol, int tcplen);
+						     int protocol,
+						     unsigned int tcplen);
 extern struct tcp_md5sig_key	*tcp_v4_md5_lookup(struct sock *sk,
 						   struct sock *addr_sk);
 
@@ -1404,7 +1405,8 @@ struct tcp_sock_af_ops {
 						  struct dst_entry *dst,
 						  struct request_sock *req,
 						  struct tcphdr *th,
-						  int protocol, int len);
+						  int protocol,
+						  unsigned int len);
 	int			(*md5_add) (struct sock *sk,
 					    struct sock *addr_sk,
 					    u8 *newkey,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index fc9bdd8b7dcf..9aea88b8d4fc 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -99,7 +99,7 @@ static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk,
 static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   __be32 saddr, __be32 daddr,
 				   struct tcphdr *th, int protocol,
-				   int tcplen);
+				   unsigned int tcplen);
 #endif
 
 struct inet_hashinfo __cacheline_aligned tcp_hashinfo = {
@@ -1020,7 +1020,7 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, char __user *optval,
 static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   __be32 saddr, __be32 daddr,
 				   struct tcphdr *th, int protocol,
-				   int tcplen)
+				   unsigned int tcplen)
 {
 	struct scatterlist sg[4];
 	__u16 data_len;
@@ -1113,7 +1113,7 @@ int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 			 struct dst_entry *dst,
 			 struct request_sock *req,
 			 struct tcphdr *th, int protocol,
-			 int tcplen)
+			 unsigned int tcplen)
 {
 	__be32 saddr, daddr;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 0268e118f0b4..00c08399837d 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -733,7 +733,7 @@ static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   struct in6_addr *saddr,
 				   struct in6_addr *daddr,
 				   struct tcphdr *th, int protocol,
-				   int tcplen)
+				   unsigned int tcplen)
 {
 	struct scatterlist sg[4];
 	__u16 data_len;
@@ -818,7 +818,7 @@ static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				struct dst_entry *dst,
 				struct request_sock *req,
 				struct tcphdr *th, int protocol,
-				int tcplen)
+				unsigned int tcplen)
 {
 	struct in6_addr *saddr, *daddr;
 
@@ -985,7 +985,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 	struct tcphdr *th = tcp_hdr(skb), *t1;
 	struct sk_buff *buff;
 	struct flowi fl;
-	int tot_len = sizeof(*th);
+	unsigned int tot_len = sizeof(*th);
 #ifdef CONFIG_TCP_MD5SIG
 	struct tcp_md5sig_key *key;
 #endif
@@ -1085,7 +1085,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
 	struct tcphdr *th = tcp_hdr(skb), *t1;
 	struct sk_buff *buff;
 	struct flowi fl;
-	int tot_len = sizeof(struct tcphdr);
+	unsigned int tot_len = sizeof(struct tcphdr);
 	__be32 *topt;
 #ifdef CONFIG_TCP_MD5SIG
 	struct tcp_md5sig_key *key;
-- 
cgit v1.2.3


From 8df09ea3b8ccfe0c94844102d33fa46f57c08d9e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Fri, 21 Dec 2007 03:07:41 -0800
Subject: [SOCK] Avoid integer divides where not necessary in
 include/net/sock.h

Because sk_wmem_queued, sk_sndbuf are signed, a divide per two
may force compiler to use an integer divide.

We can instead use a right shift.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 416bc994adad..e178b49eefbb 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -445,7 +445,7 @@ static inline int sk_acceptq_is_full(struct sock *sk)
  */
 static inline int sk_stream_min_wspace(struct sock *sk)
 {
-	return sk->sk_wmem_queued / 2;
+	return sk->sk_wmem_queued >> 1;
 }
 
 static inline int sk_stream_wspace(struct sock *sk)
@@ -1187,7 +1187,7 @@ static inline void sk_wake_async(struct sock *sk, int how, int band)
 static inline void sk_stream_moderate_sndbuf(struct sock *sk)
 {
 	if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK)) {
-		sk->sk_sndbuf = min(sk->sk_sndbuf, sk->sk_wmem_queued / 2);
+		sk->sk_sndbuf = min(sk->sk_sndbuf, sk->sk_wmem_queued >> 1);
 		sk->sk_sndbuf = max(sk->sk_sndbuf, SOCK_MIN_SNDBUF);
 	}
 }
@@ -1211,7 +1211,7 @@ static inline struct page *sk_stream_alloc_page(struct sock *sk)
  */
 static inline int sock_writeable(const struct sock *sk) 
 {
-	return atomic_read(&sk->sk_wmem_alloc) < (sk->sk_sndbuf / 2);
+	return atomic_read(&sk->sk_wmem_alloc) < (sk->sk_sndbuf >> 1);
 }
 
 static inline gfp_t gfp_any(void)
-- 
cgit v1.2.3


From afeb14b49098ba7a51c96e083a4105a0301f94c4 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Fri, 21 Dec 2007 14:58:11 -0800
Subject: [XFRM]: RFC4303 compliant auditing

This patch adds a number of new IPsec audit events to meet the auditing
requirements of RFC4303.  This includes audit hooks for the following events:

 * Could not find a valid SA [sections 2.1, 3.4.2]
   . xfrm_audit_state_notfound()
   . xfrm_audit_state_notfound_simple()

 * Sequence number overflow [section 3.3.3]
   . xfrm_audit_state_replay_overflow()

 * Replayed packet [section 3.4.3]
   . xfrm_audit_state_replay()

 * Integrity check failure [sections 3.4.4.1, 3.4.4.2]
   . xfrm_audit_state_icvfail()

While RFC4304 deals only with ESP most of the changes in this patch apply to
IPsec in general, i.e. both AH and ESP.  The one case, integrity check
failure, where ESP specific code had to be modified the same was done to the
AH code for the sake of consistency.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  33 ++++++++---
 net/ipv4/ah4.c         |   4 +-
 net/ipv4/esp4.c        |   1 +
 net/ipv6/ah6.c         |   2 +-
 net/ipv6/esp6.c        |   1 +
 net/ipv6/xfrm6_input.c |   1 +
 net/xfrm/xfrm_input.c  |   3 +-
 net/xfrm/xfrm_output.c |   2 +
 net/xfrm/xfrm_policy.c |  14 ++---
 net/xfrm/xfrm_state.c  | 153 +++++++++++++++++++++++++++++++++++++++++++------
 10 files changed, 180 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index f333c95c4189..5d5580ac0101 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -565,26 +565,33 @@ struct xfrm_audit
 };
 
 #ifdef CONFIG_AUDITSYSCALL
-static inline struct audit_buffer *xfrm_audit_start(u32 auid, u32 secid)
+static inline struct audit_buffer *xfrm_audit_start(const char *op)
 {
 	struct audit_buffer *audit_buf = NULL;
-	char *secctx;
-	u32 secctx_len;
 
+	if (audit_enabled == 0)
+		return NULL;
 	audit_buf = audit_log_start(current->audit_context, GFP_ATOMIC,
-			      AUDIT_MAC_IPSEC_EVENT);
+				    AUDIT_MAC_IPSEC_EVENT);
 	if (audit_buf == NULL)
 		return NULL;
+	audit_log_format(audit_buf, "op=%s", op);
+	return audit_buf;
+}
 
-	audit_log_format(audit_buf, "auid=%u", auid);
+static inline void xfrm_audit_helper_usrinfo(u32 auid, u32 secid,
+					     struct audit_buffer *audit_buf)
+{
+	char *secctx;
+	u32 secctx_len;
 
+	audit_log_format(audit_buf, " auid=%u", auid);
 	if (secid != 0 &&
 	    security_secid_to_secctx(secid, &secctx, &secctx_len) == 0) {
 		audit_log_format(audit_buf, " subj=%s", secctx);
 		security_release_secctx(secctx, secctx_len);
 	} else
 		audit_log_task_context(audit_buf);
-	return audit_buf;
 }
 
 extern void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
@@ -595,11 +602,22 @@ extern void xfrm_audit_state_add(struct xfrm_state *x, int result,
 				 u32 auid, u32 secid);
 extern void xfrm_audit_state_delete(struct xfrm_state *x, int result,
 				    u32 auid, u32 secid);
+extern void xfrm_audit_state_replay_overflow(struct xfrm_state *x,
+					     struct sk_buff *skb);
+extern void xfrm_audit_state_notfound_simple(struct sk_buff *skb, u16 family);
+extern void xfrm_audit_state_notfound(struct sk_buff *skb, u16 family,
+				      __be32 net_spi, __be32 net_seq);
+extern void xfrm_audit_state_icvfail(struct xfrm_state *x,
+				     struct sk_buff *skb, u8 proto);
 #else
 #define xfrm_audit_policy_add(x, r, a, s)	do { ; } while (0)
 #define xfrm_audit_policy_delete(x, r, a, s)	do { ; } while (0)
 #define xfrm_audit_state_add(x, r, a, s)	do { ; } while (0)
 #define xfrm_audit_state_delete(x, r, a, s)	do { ; } while (0)
+#define xfrm_audit_state_replay_overflow(x, s)	do { ; } while (0)
+#define xfrm_audit_state_notfound_simple(s, f)	do { ; } while (0)
+#define xfrm_audit_state_notfound(s, f, sp, sq)	do { ; } while (0)
+#define xfrm_audit_state_icvfail(x, s, p)	do { ; } while (0)
 #endif /* CONFIG_AUDITSYSCALL */
 
 static inline void xfrm_pol_hold(struct xfrm_policy *policy)
@@ -1214,7 +1232,8 @@ extern int xfrm_state_delete(struct xfrm_state *x);
 extern int xfrm_state_flush(u8 proto, struct xfrm_audit *audit_info);
 extern void xfrm_sad_getinfo(struct xfrmk_sadinfo *si);
 extern void xfrm_spd_getinfo(struct xfrmk_spdinfo *si);
-extern int xfrm_replay_check(struct xfrm_state *x, __be32 seq);
+extern int xfrm_replay_check(struct xfrm_state *x,
+			     struct sk_buff *skb, __be32 seq);
 extern void xfrm_replay_advance(struct xfrm_state *x, __be32 seq);
 extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index d76803a3dcae..ec8de0aa20ec 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -179,8 +179,10 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb)
 		err = ah_mac_digest(ahp, skb, ah->auth_data);
 		if (err)
 			goto unlock;
-		if (memcmp(ahp->work_icv, auth_data, ahp->icv_trunc_len))
+		if (memcmp(ahp->work_icv, auth_data, ahp->icv_trunc_len)) {
+			xfrm_audit_state_icvfail(x, skb, IPPROTO_AH);
 			err = -EBADMSG;
+		}
 	}
 unlock:
 	spin_unlock(&x->lock);
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 28ea5c77ca23..b334c7619c08 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -191,6 +191,7 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 			BUG();
 
 		if (unlikely(memcmp(esp->auth.work_icv, sum, alen))) {
+			xfrm_audit_state_icvfail(x, skb, IPPROTO_ESP);
 			err = -EBADMSG;
 			goto unlock;
 		}
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 1b51d1eedbde..2d32772c87c3 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -381,7 +381,7 @@ static int ah6_input(struct xfrm_state *x, struct sk_buff *skb)
 		if (err)
 			goto unlock;
 		if (memcmp(ahp->work_icv, auth_data, ahp->icv_trunc_len)) {
-			LIMIT_NETDEBUG(KERN_WARNING "ipsec ah authentication error\n");
+			xfrm_audit_state_icvfail(x, skb, IPPROTO_AH);
 			err = -EBADMSG;
 		}
 	}
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 5bd5292ad9fa..e10f10bfe2c9 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -186,6 +186,7 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
 			BUG();
 
 		if (unlikely(memcmp(esp->auth.work_icv, sum, alen))) {
+			xfrm_audit_state_icvfail(x, skb, IPPROTO_ESP);
 			ret = -EBADMSG;
 			goto unlock;
 		}
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 6644fc6d5427..063ce6ed1bd0 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -152,6 +152,7 @@ int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
 
 	if (!x) {
 		XFRM_INC_STATS(LINUX_MIB_XFRMINNOSTATES);
+		xfrm_audit_state_notfound_simple(skb, AF_INET6);
 		goto drop;
 	}
 
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 493243fc5fe5..1b250f33ad5b 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -147,6 +147,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		x = xfrm_state_lookup(daddr, spi, nexthdr, family);
 		if (x == NULL) {
 			XFRM_INC_STATS(LINUX_MIB_XFRMINNOSTATES);
+			xfrm_audit_state_notfound(skb, family, spi, seq);
 			goto drop;
 		}
 
@@ -163,7 +164,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			goto drop_unlock;
 		}
 
-		if (x->props.replay_window && xfrm_replay_check(x, seq)) {
+		if (x->props.replay_window && xfrm_replay_check(x, skb, seq)) {
 			XFRM_INC_STATS(LINUX_MIB_XFRMINSEQOUTOFWINDOW);
 			goto drop_unlock;
 		}
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index 867484a046af..09514449fe8a 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -57,6 +57,8 @@ static int xfrm_output_one(struct sk_buff *skb, int err)
 
 		if (x->type->flags & XFRM_TYPE_REPLAY_PROT) {
 			XFRM_SKB_CB(skb)->seq = ++x->replay.oseq;
+			if (unlikely(x->replay.oseq == 0))
+				xfrm_audit_state_replay_overflow(x, skb);
 			if (xfrm_aevent_is_on())
 				xfrm_replay_notify(x, XFRM_REPLAY_UPDATE);
 		}
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index abc3e39b115b..280f8ded975c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -2407,12 +2407,11 @@ void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
 {
 	struct audit_buffer *audit_buf;
 
-	if (audit_enabled == 0)
-		return;
-	audit_buf = xfrm_audit_start(auid, secid);
+	audit_buf = xfrm_audit_start("SPD-add");
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SPD-add res=%u", result);
+	xfrm_audit_helper_usrinfo(auid, secid, audit_buf);
+	audit_log_format(audit_buf, " res=%u", result);
 	xfrm_audit_common_policyinfo(xp, audit_buf);
 	audit_log_end(audit_buf);
 }
@@ -2423,12 +2422,11 @@ void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
 {
 	struct audit_buffer *audit_buf;
 
-	if (audit_enabled == 0)
-		return;
-	audit_buf = xfrm_audit_start(auid, secid);
+	audit_buf = xfrm_audit_start("SPD-delete");
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SPD-delete res=%u", result);
+	xfrm_audit_helper_usrinfo(auid, secid, audit_buf);
+	audit_log_format(audit_buf, " res=%u", result);
 	xfrm_audit_common_policyinfo(xp, audit_buf);
 	audit_log_end(audit_buf);
 }
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 9e57378c51df..6bf876c866df 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -61,6 +61,13 @@ static unsigned int xfrm_state_genid;
 static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family);
 static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo);
 
+#ifdef CONFIG_AUDITSYSCALL
+static void xfrm_audit_state_replay(struct xfrm_state *x,
+				    struct sk_buff *skb, __be32 net_seq);
+#else
+#define xfrm_audit_state_replay(x, s, sq)	do { ; } while (0)
+#endif /* CONFIG_AUDITSYSCALL */
+
 static inline unsigned int xfrm_dst_hash(xfrm_address_t *daddr,
 					 xfrm_address_t *saddr,
 					 u32 reqid,
@@ -1609,13 +1616,14 @@ static void xfrm_replay_timer_handler(unsigned long data)
 	spin_unlock(&x->lock);
 }
 
-int xfrm_replay_check(struct xfrm_state *x, __be32 net_seq)
+int xfrm_replay_check(struct xfrm_state *x,
+		      struct sk_buff *skb, __be32 net_seq)
 {
 	u32 diff;
 	u32 seq = ntohl(net_seq);
 
 	if (unlikely(seq == 0))
-		return -EINVAL;
+		goto err;
 
 	if (likely(seq > x->replay.seq))
 		return 0;
@@ -1624,14 +1632,18 @@ int xfrm_replay_check(struct xfrm_state *x, __be32 net_seq)
 	if (diff >= min_t(unsigned int, x->props.replay_window,
 			  sizeof(x->replay.bitmap) * 8)) {
 		x->stats.replay_window++;
-		return -EINVAL;
+		goto err;
 	}
 
 	if (x->replay.bitmap & (1U << diff)) {
 		x->stats.replay++;
-		return -EINVAL;
+		goto err;
 	}
 	return 0;
+
+err:
+	xfrm_audit_state_replay(x, skb, net_seq);
+	return -EINVAL;
 }
 EXPORT_SYMBOL(xfrm_replay_check);
 
@@ -1996,8 +2008,8 @@ void __init xfrm_state_init(void)
 }
 
 #ifdef CONFIG_AUDITSYSCALL
-static inline void xfrm_audit_common_stateinfo(struct xfrm_state *x,
-					       struct audit_buffer *audit_buf)
+static inline void xfrm_audit_helper_sainfo(struct xfrm_state *x,
+					    struct audit_buffer *audit_buf)
 {
 	struct xfrm_sec_ctx *ctx = x->security;
 	u32 spi = ntohl(x->id.spi);
@@ -2024,18 +2036,45 @@ static inline void xfrm_audit_common_stateinfo(struct xfrm_state *x,
 	audit_log_format(audit_buf, " spi=%u(0x%x)", spi, spi);
 }
 
+static inline void xfrm_audit_helper_pktinfo(struct sk_buff *skb, u16 family,
+					     struct audit_buffer *audit_buf)
+{
+	struct iphdr *iph4;
+	struct ipv6hdr *iph6;
+
+	switch (family) {
+	case AF_INET:
+		iph4 = ip_hdr(skb);
+		audit_log_format(audit_buf,
+				 " src=" NIPQUAD_FMT " dst=" NIPQUAD_FMT,
+				 NIPQUAD(iph4->saddr),
+				 NIPQUAD(iph4->daddr));
+		break;
+	case AF_INET6:
+		iph6 = ipv6_hdr(skb);
+		audit_log_format(audit_buf,
+				 " src=" NIP6_FMT " dst=" NIP6_FMT
+				 " flowlbl=0x%x%x%x",
+				 NIP6(iph6->saddr),
+				 NIP6(iph6->daddr),
+				 iph6->flow_lbl[0] & 0x0f,
+				 iph6->flow_lbl[1],
+				 iph6->flow_lbl[2]);
+		break;
+	}
+}
+
 void xfrm_audit_state_add(struct xfrm_state *x, int result,
 			  u32 auid, u32 secid)
 {
 	struct audit_buffer *audit_buf;
 
-	if (audit_enabled == 0)
-		return;
-	audit_buf = xfrm_audit_start(auid, secid);
+	audit_buf = xfrm_audit_start("SAD-add");
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SAD-add res=%u", result);
-	xfrm_audit_common_stateinfo(x, audit_buf);
+	xfrm_audit_helper_usrinfo(auid, secid, audit_buf);
+	xfrm_audit_helper_sainfo(x, audit_buf);
+	audit_log_format(audit_buf, " res=%u", result);
 	audit_log_end(audit_buf);
 }
 EXPORT_SYMBOL_GPL(xfrm_audit_state_add);
@@ -2045,14 +2084,96 @@ void xfrm_audit_state_delete(struct xfrm_state *x, int result,
 {
 	struct audit_buffer *audit_buf;
 
-	if (audit_enabled == 0)
-		return;
-	audit_buf = xfrm_audit_start(auid, secid);
+	audit_buf = xfrm_audit_start("SAD-delete");
 	if (audit_buf == NULL)
 		return;
-	audit_log_format(audit_buf, " op=SAD-delete res=%u", result);
-	xfrm_audit_common_stateinfo(x, audit_buf);
+	xfrm_audit_helper_usrinfo(auid, secid, audit_buf);
+	xfrm_audit_helper_sainfo(x, audit_buf);
+	audit_log_format(audit_buf, " res=%u", result);
 	audit_log_end(audit_buf);
 }
 EXPORT_SYMBOL_GPL(xfrm_audit_state_delete);
+
+void xfrm_audit_state_replay_overflow(struct xfrm_state *x,
+				      struct sk_buff *skb)
+{
+	struct audit_buffer *audit_buf;
+	u32 spi;
+
+	audit_buf = xfrm_audit_start("SA-replay-overflow");
+	if (audit_buf == NULL)
+		return;
+	xfrm_audit_helper_pktinfo(skb, x->props.family, audit_buf);
+	/* don't record the sequence number because it's inherent in this kind
+	 * of audit message */
+	spi = ntohl(x->id.spi);
+	audit_log_format(audit_buf, " spi=%u(0x%x)", spi, spi);
+	audit_log_end(audit_buf);
+}
+EXPORT_SYMBOL_GPL(xfrm_audit_state_replay_overflow);
+
+static void xfrm_audit_state_replay(struct xfrm_state *x,
+			     struct sk_buff *skb, __be32 net_seq)
+{
+	struct audit_buffer *audit_buf;
+	u32 spi;
+
+	audit_buf = xfrm_audit_start("SA-replayed-pkt");
+	if (audit_buf == NULL)
+		return;
+	xfrm_audit_helper_pktinfo(skb, x->props.family, audit_buf);
+	spi = ntohl(x->id.spi);
+	audit_log_format(audit_buf, " spi=%u(0x%x) seqno=%u",
+			 spi, spi, ntohl(net_seq));
+	audit_log_end(audit_buf);
+}
+
+void xfrm_audit_state_notfound_simple(struct sk_buff *skb, u16 family)
+{
+	struct audit_buffer *audit_buf;
+
+	audit_buf = xfrm_audit_start("SA-notfound");
+	if (audit_buf == NULL)
+		return;
+	xfrm_audit_helper_pktinfo(skb, family, audit_buf);
+	audit_log_end(audit_buf);
+}
+EXPORT_SYMBOL_GPL(xfrm_audit_state_notfound_simple);
+
+void xfrm_audit_state_notfound(struct sk_buff *skb, u16 family,
+			       __be32 net_spi, __be32 net_seq)
+{
+	struct audit_buffer *audit_buf;
+	u32 spi;
+
+	audit_buf = xfrm_audit_start("SA-notfound");
+	if (audit_buf == NULL)
+		return;
+	xfrm_audit_helper_pktinfo(skb, family, audit_buf);
+	spi = ntohl(net_spi);
+	audit_log_format(audit_buf, " spi=%u(0x%x) seqno=%u",
+			 spi, spi, ntohl(net_seq));
+	audit_log_end(audit_buf);
+}
+EXPORT_SYMBOL_GPL(xfrm_audit_state_notfound);
+
+void xfrm_audit_state_icvfail(struct xfrm_state *x,
+			      struct sk_buff *skb, u8 proto)
+{
+	struct audit_buffer *audit_buf;
+	__be32 net_spi;
+	__be32 net_seq;
+
+	audit_buf = xfrm_audit_start("SA-icv-failure");
+	if (audit_buf == NULL)
+		return;
+	xfrm_audit_helper_pktinfo(skb, x->props.family, audit_buf);
+	if (xfrm_parse_spi(skb, proto, &net_spi, &net_seq) == 0) {
+		u32 spi = ntohl(net_spi);
+		audit_log_format(audit_buf, " spi=%u(0x%x) seqno=%u",
+				 spi, spi, ntohl(net_seq));
+	}
+	audit_log_end(audit_buf);
+}
+EXPORT_SYMBOL_GPL(xfrm_audit_state_icvfail);
 #endif /* CONFIG_AUDITSYSCALL */
-- 
cgit v1.2.3


From 426b5303eb435d98b9bee37a807be386bc2b3320 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Thu, 24 Jan 2008 00:13:18 -0800
Subject: [NETNS]: Modify the neighbour table code so it handles multiple
 network namespaces

I'm actually surprised at how much was involved.  At first glance it
appears that the neighbour table data structures are already split by
network device so all that should be needed is to modify the user
interface commands to filter the set of neighbours by the network
namespace of their devices.

However a couple things turned up while I was reading through the
code.  The proxy neighbour table allows entries with no network
device, and the neighbour parms are per network device (except for the
defaults) so they now need a per network namespace default.

So I updated the two structures (which surprised me) with their very
own network namespace parameter.  Updated the relevant lookup and
destroy routines with a network namespace parameter and modified the
code that interacts with users to filter out neighbour table entries
for devices of other namespaces.

I'm a little concerned that we can modify and display the global table
configuration and from all network namespaces.  But this appears good
enough for now.

I keep thinking modifying the neighbour table to have per network
namespace instances of each table type would should be cleaner.  The
hash table is already dynamically sized so there are it is not a
limiter.  The default parameter would be straight forward to take care
of.  However when I look at the how the network table is built and
used I still find some assumptions that there is only a single
neighbour table for each type of table in the kernel.  The netlink
operations, neigh_seq_start, the non-core network users that call
neigh_lookup.  So while it might be doable it would require more
refactoring than my current approach of just doing a little extra
filtering in the code.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h |  10 +++--
 net/atm/clip.c          |  15 ++++++-
 net/core/neighbour.c    | 116 ++++++++++++++++++++++++++++++------------------
 net/decnet/dn_neigh.c   |   6 +--
 net/decnet/dn_route.c   |   2 +-
 net/ipv4/arp.c          |  12 ++---
 net/ipv6/ip6_output.c   |   2 +-
 net/ipv6/ndisc.c        |   4 +-
 8 files changed, 106 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index a4f26187fc1a..11590f23d99d 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -34,6 +34,7 @@ struct neighbour;
 
 struct neigh_parms
 {
+	struct net *net;
 	struct net_device *dev;
 	struct neigh_parms *next;
 	int	(*neigh_setup)(struct neighbour *);
@@ -126,7 +127,8 @@ struct neigh_ops
 struct pneigh_entry
 {
 	struct pneigh_entry	*next;
-	struct net_device		*dev;
+	struct net		*net;
+	struct net_device	*dev;
 	u8			flags;
 	u8			key[0];
 };
@@ -187,6 +189,7 @@ extern struct neighbour *	neigh_lookup(struct neigh_table *tbl,
 					     const void *pkey,
 					     struct net_device *dev);
 extern struct neighbour *	neigh_lookup_nodev(struct neigh_table *tbl,
+						   struct net *net,
 						   const void *pkey);
 extern struct neighbour *	neigh_create(struct neigh_table *tbl,
 					     const void *pkey,
@@ -211,8 +214,8 @@ extern unsigned long		neigh_rand_reach_time(unsigned long base);
 
 extern void			pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
 					       struct sk_buff *skb);
-extern struct pneigh_entry	*pneigh_lookup(struct neigh_table *tbl, const void *key, struct net_device *dev, int creat);
-extern int			pneigh_delete(struct neigh_table *tbl, const void *key, struct net_device *dev);
+extern struct pneigh_entry	*pneigh_lookup(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev, int creat);
+extern int			pneigh_delete(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev);
 
 extern void neigh_app_ns(struct neighbour *n);
 extern void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie);
@@ -220,6 +223,7 @@ extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct n
 extern void pneigh_for_each(struct neigh_table *tbl, void (*cb)(struct pneigh_entry *));
 
 struct neigh_seq_state {
+	struct net *net;
 	struct neigh_table *tbl;
 	void *(*neigh_sub_iter)(struct neigh_seq_state *state,
 				struct neighbour *n, loff_t *pos);
diff --git a/net/atm/clip.c b/net/atm/clip.c
index 741742f00797..47fbdc0c5f72 100644
--- a/net/atm/clip.c
+++ b/net/atm/clip.c
@@ -949,6 +949,11 @@ static int arp_seq_open(struct inode *inode, struct file *file)
 
 	seq = file->private_data;
 	seq->private = state;
+	state->ns.net = get_proc_net(inode);
+	if (!state->ns.net) {
+		seq_release_private(inode, file);
+		rc = -ENXIO;
+	}
 out:
 	return rc;
 
@@ -957,11 +962,19 @@ out_kfree:
 	goto out;
 }
 
+static int arp_seq_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct clip_seq_state *state = seq->private;
+	put_net(state->ns.net);
+	return seq_release_private(inode, file);
+}
+
 static const struct file_operations arp_seq_fops = {
 	.open		= arp_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release_private,
+	.release	= arp_seq_release,
 	.owner		= THIS_MODULE
 };
 #endif
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 9a283fcde9a6..bd899d557737 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -375,7 +375,8 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 	return n;
 }
 
-struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, const void *pkey)
+struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
+				     const void *pkey)
 {
 	struct neighbour *n;
 	int key_len = tbl->key_len;
@@ -385,7 +386,8 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, const void *pkey)
 
 	read_lock_bh(&tbl->lock);
 	for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) {
-		if (!memcmp(n->primary_key, pkey, key_len)) {
+		if (!memcmp(n->primary_key, pkey, key_len) &&
+		    (net == n->dev->nd_net)) {
 			neigh_hold(n);
 			NEIGH_CACHE_STAT_INC(tbl, hits);
 			break;
@@ -463,7 +465,8 @@ out_neigh_release:
 	goto out;
 }
 
-struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
+struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
+				    struct net *net, const void *pkey,
 				    struct net_device *dev, int creat)
 {
 	struct pneigh_entry *n;
@@ -479,6 +482,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
 
 	for (n = tbl->phash_buckets[hash_val]; n; n = n->next) {
 		if (!memcmp(n->key, pkey, key_len) &&
+		    (n->net == net) &&
 		    (n->dev == dev || !n->dev)) {
 			read_unlock_bh(&tbl->lock);
 			goto out;
@@ -495,6 +499,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
 	if (!n)
 		goto out;
 
+	n->net = hold_net(net);
 	memcpy(n->key, pkey, key_len);
 	n->dev = dev;
 	if (dev)
@@ -517,7 +522,7 @@ out:
 }
 
 
-int pneigh_delete(struct neigh_table *tbl, const void *pkey,
+int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey,
 		  struct net_device *dev)
 {
 	struct pneigh_entry *n, **np;
@@ -532,13 +537,15 @@ int pneigh_delete(struct neigh_table *tbl, const void *pkey,
 	write_lock_bh(&tbl->lock);
 	for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL;
 	     np = &n->next) {
-		if (!memcmp(n->key, pkey, key_len) && n->dev == dev) {
+		if (!memcmp(n->key, pkey, key_len) && n->dev == dev &&
+		    (n->net == net)) {
 			*np = n->next;
 			write_unlock_bh(&tbl->lock);
 			if (tbl->pdestructor)
 				tbl->pdestructor(n);
 			if (n->dev)
 				dev_put(n->dev);
+			release_net(n->net);
 			kfree(n);
 			return 0;
 		}
@@ -561,6 +568,7 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev)
 					tbl->pdestructor(n);
 				if (n->dev)
 					dev_put(n->dev);
+				release_net(n->net);
 				kfree(n);
 				continue;
 			}
@@ -1261,12 +1269,37 @@ void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
 	spin_unlock(&tbl->proxy_queue.lock);
 }
 
+static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
+						      struct net *net, int ifindex)
+{
+	struct neigh_parms *p;
+
+	for (p = &tbl->parms; p; p = p->next) {
+		if (p->net != net)
+			continue;
+		if ((p->dev && p->dev->ifindex == ifindex) ||
+		    (!p->dev && !ifindex))
+			return p;
+	}
+
+	return NULL;
+}
 
 struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
 				      struct neigh_table *tbl)
 {
-	struct neigh_parms *p = kmemdup(&tbl->parms, sizeof(*p), GFP_KERNEL);
+	struct neigh_parms *p, *ref;
+	struct net *net;
+
+	net = &init_net;
+	if (dev)
+		net = dev->nd_net;
+
+	ref = lookup_neigh_params(tbl, net, 0);
+	if (!ref)
+		return NULL;
 
+	p = kmemdup(ref, sizeof(*p), GFP_KERNEL);
 	if (p) {
 		p->tbl		  = tbl;
 		atomic_set(&p->refcnt, 1);
@@ -1282,6 +1315,7 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
 			dev_hold(dev);
 			p->dev = dev;
 		}
+		p->net = hold_net(net);
 		p->sysctl_table = NULL;
 		write_lock_bh(&tbl->lock);
 		p->next		= tbl->parms.next;
@@ -1323,6 +1357,7 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms)
 
 void neigh_parms_destroy(struct neigh_parms *parms)
 {
+	release_net(parms->net);
 	kfree(parms);
 }
 
@@ -1333,6 +1368,7 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl)
 	unsigned long now = jiffies;
 	unsigned long phsize;
 
+	tbl->parms.net = &init_net;
 	atomic_set(&tbl->parms.refcnt, 1);
 	INIT_RCU_HEAD(&tbl->parms.rcu_head);
 	tbl->parms.reachable_time =
@@ -1446,9 +1482,6 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	struct net_device *dev = NULL;
 	int err = -EINVAL;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	if (nlmsg_len(nlh) < sizeof(*ndm))
 		goto out;
 
@@ -1477,7 +1510,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 			goto out_dev_put;
 
 		if (ndm->ndm_flags & NTF_PROXY) {
-			err = pneigh_delete(tbl, nla_data(dst_attr), dev);
+			err = pneigh_delete(tbl, net, nla_data(dst_attr), dev);
 			goto out_dev_put;
 		}
 
@@ -1515,9 +1548,6 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	struct net_device *dev = NULL;
 	int err;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL);
 	if (err < 0)
 		goto out;
@@ -1557,7 +1587,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 			struct pneigh_entry *pn;
 
 			err = -ENOBUFS;
-			pn = pneigh_lookup(tbl, dst, dev, 1);
+			pn = pneigh_lookup(tbl, net, dst, dev, 1);
 			if (pn) {
 				pn->flags = ndm->ndm_flags;
 				err = 0;
@@ -1752,19 +1782,6 @@ errout:
 	return -EMSGSIZE;
 }
 
-static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
-						      int ifindex)
-{
-	struct neigh_parms *p;
-
-	for (p = &tbl->parms; p; p = p->next)
-		if ((p->dev && p->dev->ifindex == ifindex) ||
-		    (!p->dev && !ifindex))
-			return p;
-
-	return NULL;
-}
-
 static const struct nla_policy nl_neightbl_policy[NDTA_MAX+1] = {
 	[NDTA_NAME]		= { .type = NLA_STRING },
 	[NDTA_THRESH1]		= { .type = NLA_U32 },
@@ -1798,9 +1815,6 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	struct nlattr *tb[NDTA_MAX+1];
 	int err;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	err = nlmsg_parse(nlh, sizeof(*ndtmsg), tb, NDTA_MAX,
 			  nl_neightbl_policy);
 	if (err < 0)
@@ -1845,7 +1859,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		if (tbp[NDTPA_IFINDEX])
 			ifindex = nla_get_u32(tbp[NDTPA_IFINDEX]);
 
-		p = lookup_neigh_params(tbl, ifindex);
+		p = lookup_neigh_params(tbl, net, ifindex);
 		if (p == NULL) {
 			err = -ENOENT;
 			goto errout_tbl_lock;
@@ -1926,9 +1940,6 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 	int neigh_skip = cb->args[1];
 	struct neigh_table *tbl;
 
-	if (net != &init_net)
-		return 0;
-
 	family = ((struct rtgenmsg *) nlmsg_data(cb->nlh))->rtgen_family;
 
 	read_lock(&neigh_tbl_lock);
@@ -1943,8 +1954,11 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 				       NLM_F_MULTI) <= 0)
 			break;
 
-		for (nidx = 0, p = tbl->parms.next; p; p = p->next, nidx++) {
-			if (nidx < neigh_skip)
+		for (nidx = 0, p = tbl->parms.next; p; p = p->next) {
+			if (net != p->net)
+				continue;
+
+			if (nidx++ < neigh_skip)
 				continue;
 
 			if (neightbl_fill_param_info(skb, tbl, p,
@@ -2020,6 +2034,7 @@ static void neigh_update_notify(struct neighbour *neigh)
 static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
 			    struct netlink_callback *cb)
 {
+	struct net * net = skb->sk->sk_net;
 	struct neighbour *n;
 	int rc, h, s_h = cb->args[1];
 	int idx, s_idx = idx = cb->args[2];
@@ -2030,8 +2045,12 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
 			continue;
 		if (h > s_h)
 			s_idx = 0;
-		for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next, idx++) {
-			if (idx < s_idx)
+		for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) {
+			int lidx;
+			if (n->dev->nd_net != net)
+				continue;
+			lidx = idx++;
+			if (lidx < s_idx)
 				continue;
 			if (neigh_fill_info(skb, n, NETLINK_CB(cb->skb).pid,
 					    cb->nlh->nlmsg_seq,
@@ -2053,13 +2072,9 @@ out:
 
 static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
 	struct neigh_table *tbl;
 	int t, family, s_t;
 
-	if (net != &init_net)
-		return 0;
-
 	read_lock(&neigh_tbl_lock);
 	family = ((struct rtgenmsg *) nlmsg_data(cb->nlh))->rtgen_family;
 	s_t = cb->args[0];
@@ -2127,6 +2142,7 @@ EXPORT_SYMBOL(__neigh_for_each_release);
 static struct neighbour *neigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
+	struct net *net = state->net;
 	struct neigh_table *tbl = state->tbl;
 	struct neighbour *n = NULL;
 	int bucket = state->bucket;
@@ -2136,6 +2152,8 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 		n = tbl->hash_buckets[bucket];
 
 		while (n) {
+			if (n->dev->nd_net != net)
+				goto next;
 			if (state->neigh_sub_iter) {
 				loff_t fakep = 0;
 				void *v;
@@ -2165,6 +2183,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 					loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
+	struct net *net = state->net;
 	struct neigh_table *tbl = state->tbl;
 
 	if (state->neigh_sub_iter) {
@@ -2176,6 +2195,8 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 
 	while (1) {
 		while (n) {
+			if (n->dev->nd_net != net)
+				goto next;
 			if (state->neigh_sub_iter) {
 				void *v = state->neigh_sub_iter(state, n, pos);
 				if (v)
@@ -2222,6 +2243,7 @@ static struct neighbour *neigh_get_idx(struct seq_file *seq, loff_t *pos)
 static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
+	struct net * net = state->net;
 	struct neigh_table *tbl = state->tbl;
 	struct pneigh_entry *pn = NULL;
 	int bucket = state->bucket;
@@ -2229,6 +2251,8 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 	state->flags |= NEIGH_SEQ_IS_PNEIGH;
 	for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) {
 		pn = tbl->phash_buckets[bucket];
+		while (pn && (pn->net != net))
+			pn = pn->next;
 		if (pn)
 			break;
 	}
@@ -2242,6 +2266,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 					    loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
+	struct net * net = state->net;
 	struct neigh_table *tbl = state->tbl;
 
 	pn = pn->next;
@@ -2249,6 +2274,8 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 		if (++state->bucket > PNEIGH_HASHMASK)
 			break;
 		pn = tbl->phash_buckets[state->bucket];
+		while (pn && (pn->net != net))
+			pn = pn->next;
 		if (pn)
 			break;
 	}
@@ -2450,6 +2477,7 @@ static inline size_t neigh_nlmsg_size(void)
 
 static void __neigh_notify(struct neighbour *n, int type, int flags)
 {
+	struct net *net = n->dev->nd_net;
 	struct sk_buff *skb;
 	int err = -ENOBUFS;
 
@@ -2464,10 +2492,10 @@ static void __neigh_notify(struct neighbour *n, int type, int flags)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, &init_net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
+	err = rtnl_notify(skb, net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(&init_net, RTNLGRP_NEIGH, err);
+		rtnl_set_sk_err(net, RTNLGRP_NEIGH, err);
 }
 
 #ifdef CONFIG_ARPD
diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c
index e851b143cca3..1ca13b17974d 100644
--- a/net/decnet/dn_neigh.c
+++ b/net/decnet/dn_neigh.c
@@ -580,8 +580,8 @@ static const struct seq_operations dn_neigh_seq_ops = {
 
 static int dn_neigh_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &dn_neigh_seq_ops,
-			sizeof(struct neigh_seq_state));
+	return seq_open_net(inode, file, &dn_neigh_seq_ops,
+			    sizeof(struct neigh_seq_state));
 }
 
 static const struct file_operations dn_neigh_seq_fops = {
@@ -589,7 +589,7 @@ static const struct file_operations dn_neigh_seq_fops = {
 	.open		= dn_neigh_seq_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release_private,
+	.release	= seq_release_net,
 };
 
 #endif
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 1ae5efcdbd5f..938ba7da217c 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -984,7 +984,7 @@ source_ok:
 		 * here
 		 */
 		if (!try_hard) {
-			neigh = neigh_lookup_nodev(&dn_neigh_table, &fl.fld_dst);
+			neigh = neigh_lookup_nodev(&dn_neigh_table, &init_net, &fl.fld_dst);
 			if (neigh) {
 				if ((oldflp->oif &&
 				    (neigh->dev->ifindex != oldflp->oif)) ||
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index fdf12d1c350e..9eb6d3ab2977 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -837,7 +837,7 @@ static int arp_process(struct sk_buff *skb)
 		} else if (IN_DEV_FORWARD(in_dev)) {
 			if ((rt->rt_flags&RTCF_DNAT) ||
 			    (addr_type == RTN_UNICAST  && rt->u.dst.dev != dev &&
-			     (arp_fwd_proxy(in_dev, rt) || pneigh_lookup(&arp_tbl, &tip, dev, 0)))) {
+			     (arp_fwd_proxy(in_dev, rt) || pneigh_lookup(&arp_tbl, &init_net, &tip, dev, 0)))) {
 				n = neigh_event_ns(&arp_tbl, sha, &sip, dev);
 				if (n)
 					neigh_release(n);
@@ -980,7 +980,7 @@ static int arp_req_set_public(struct net *net, struct arpreq *r,
 			return -ENODEV;
 	}
 	if (mask) {
-		if (pneigh_lookup(&arp_tbl, &ip, dev, 1) == NULL)
+		if (pneigh_lookup(&arp_tbl, &init_net, &ip, dev, 1) == NULL)
 			return -ENOBUFS;
 		return 0;
 	}
@@ -1089,7 +1089,7 @@ static int arp_req_delete_public(struct net *net, struct arpreq *r,
 	__be32 mask = ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr;
 
 	if (mask == htonl(0xFFFFFFFF))
-		return pneigh_delete(&arp_tbl, &ip, dev);
+		return pneigh_delete(&arp_tbl, &init_net, &ip, dev);
 
 	if (mask)
 		return -EINVAL;
@@ -1375,8 +1375,8 @@ static const struct seq_operations arp_seq_ops = {
 
 static int arp_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &arp_seq_ops,
-			sizeof(struct neigh_seq_state));
+	return seq_open_net(inode, file, &arp_seq_ops,
+			    sizeof(struct neigh_seq_state));
 }
 
 static const struct file_operations arp_seq_fops = {
@@ -1384,7 +1384,7 @@ static const struct file_operations arp_seq_fops = {
 	.open           = arp_seq_open,
 	.read           = seq_read,
 	.llseek         = seq_lseek,
-	.release	= seq_release_private,
+	.release	= seq_release_net,
 };
 
 static int __init arp_proc_init(void)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 4686646058d3..ba7c8aaf2782 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -449,7 +449,7 @@ int ip6_forward(struct sk_buff *skb)
 
 	/* XXX: idev->cnf.proxy_ndp? */
 	if (ipv6_devconf.proxy_ndp &&
-	    pneigh_lookup(&nd_tbl, &hdr->daddr, skb->dev, 0)) {
+	    pneigh_lookup(&nd_tbl, &init_net, &hdr->daddr, skb->dev, 0)) {
 		int proxied = ip6_forward_proxy_check(skb);
 		if (proxied > 0)
 			return ip6_input(skb);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index b87f9d245e2e..b66a1f81bd83 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -789,7 +789,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 		if (ipv6_chk_acast_addr(dev, &msg->target) ||
 		    (idev->cnf.forwarding &&
 		     (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) &&
-		     (pneigh = pneigh_lookup(&nd_tbl,
+		     (pneigh = pneigh_lookup(&nd_tbl, &init_net,
 					     &msg->target, dev, 0)) != NULL)) {
 			if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) &&
 			    skb->pkt_type != PACKET_HOST &&
@@ -930,7 +930,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 		 */
 		if (lladdr && !memcmp(lladdr, dev->dev_addr, dev->addr_len) &&
 		    ipv6_devconf.forwarding && ipv6_devconf.proxy_ndp &&
-		    pneigh_lookup(&nd_tbl, &msg->target, dev, 0)) {
+		    pneigh_lookup(&nd_tbl, &init_net, &msg->target, dev, 0)) {
 			/* XXX: idev->cnf.prixy_ndp */
 			goto out;
 		}
-- 
cgit v1.2.3


From 21371f768bf7127ee45bfaadd17899df6a439e8f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Mon, 24 Dec 2007 20:57:56 -0800
Subject: [SOCK] Avoid divides in sk_stream_pages() and
 __sk_stream_mem_reclaim()

sk_forward_alloc being signed, we should take care of divides by
SK_STREAM_MEM_QUANTUM we do in sk_stream_pages() and
__sk_stream_mem_reclaim()

This patchs introduces SK_STREAM_MEM_QUANTUM_SHIFT, defined
as ilog2(SK_STREAM_MEM_QUANTUM), to be able to use right
shifts instead of plain divides.

This should help compiler to choose right shifts instead of
expensive divides (as seen with CONFIG_CC_OPTIMIZE_FOR_SIZE=y on x86)

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 3 ++-
 net/core/stream.c  | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index e178b49eefbb..d27ba6fdd039 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -716,10 +716,11 @@ extern void __sk_stream_mem_reclaim(struct sock *sk);
 extern int sk_stream_mem_schedule(struct sock *sk, int size, int kind);
 
 #define SK_STREAM_MEM_QUANTUM ((int)PAGE_SIZE)
+#define SK_STREAM_MEM_QUANTUM_SHIFT ilog2(SK_STREAM_MEM_QUANTUM)
 
 static inline int sk_stream_pages(int amt)
 {
-	return DIV_ROUND_UP(amt, SK_STREAM_MEM_QUANTUM);
+	return (amt + SK_STREAM_MEM_QUANTUM - 1) >> SK_STREAM_MEM_QUANTUM_SHIFT;
 }
 
 static inline void sk_stream_mem_reclaim(struct sock *sk)
diff --git a/net/core/stream.c b/net/core/stream.c
index 5586879bb9b3..bf188ffdbdbe 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -196,7 +196,7 @@ EXPORT_SYMBOL(sk_stream_error);
 
 void __sk_stream_mem_reclaim(struct sock *sk)
 {
-	atomic_sub(sk->sk_forward_alloc / SK_STREAM_MEM_QUANTUM,
+	atomic_sub(sk->sk_forward_alloc >> SK_STREAM_MEM_QUANTUM_SHIFT,
 		   sk->sk_prot->memory_allocated);
 	sk->sk_forward_alloc &= SK_STREAM_MEM_QUANTUM - 1;
 	if (*sk->sk_prot->memory_pressure &&
-- 
cgit v1.2.3


From 7ffc49a6ee92b7138c2ee28073a8e10e58335d62 Mon Sep 17 00:00:00 2001
From: Michael Chan <mchan@broadcom.com>
Date: Mon, 24 Dec 2007 21:28:09 -0800
Subject: [ETH]: Combine format_addr() with print_mac().

print_mac() used many most net drivers and format_addr() used by
net-sysfs.c are very similar and they can be intergrated.

format_addr() is also identically redefined in the qla4xxx iscsi
driver.

Export a new function sysfs_format_mac() to be used by net-sysfs,
qla4xxx and others in the future.  Both print_mac() and
sysfs_format_mac() call _format_mac_addr() to do the formatting.

Changed print_mac() to use unsigned char * to be consistent with
net_device struct's dev_addr.  Added buffer length overrun checking
as suggested by Joe Perches.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/scsi/qla4xxx/ql4_os.c | 14 +-------------
 include/linux/if_ether.h      |  8 +++++---
 net/core/net-sysfs.c          | 15 ++-------------
 net/ethernet/eth.c            | 30 +++++++++++++++++++++++++++---
 4 files changed, 35 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index f55b9f7d9396..d3f86646cb08 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -173,18 +173,6 @@ static void qla4xxx_conn_stop(struct iscsi_cls_conn *conn, int flag)
 		printk(KERN_ERR "iscsi: invalid stop flag %d\n", flag);
 }
 
-static ssize_t format_addr(char *buf, const unsigned char *addr, int len)
-{
-	int i;
-	char *cp = buf;
-
-	for (i = 0; i < len; i++)
-		cp += sprintf(cp, "%02x%c", addr[i],
-			      i == (len - 1) ? '\n' : ':');
-	return cp - buf;
-}
-
-
 static int qla4xxx_host_get_param(struct Scsi_Host *shost,
 				  enum iscsi_host_param param, char *buf)
 {
@@ -193,7 +181,7 @@ static int qla4xxx_host_get_param(struct Scsi_Host *shost,
 
 	switch (param) {
 	case ISCSI_HOST_PARAM_HWADDRESS:
-		len = format_addr(buf, ha->my_mac, MAC_ADDR_LEN);
+		len = sysfs_format_mac(buf, ha->my_mac, MAC_ADDR_LEN);
 		break;
 	case ISCSI_HOST_PARAM_IPADDRESS:
 		len = sprintf(buf, "%d.%d.%d.%d\n", ha->ip_address[0],
diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h
index cc002cbbdc26..7a1e011b8a2c 100644
--- a/include/linux/if_ether.h
+++ b/include/linux/if_ether.h
@@ -124,12 +124,14 @@ int eth_header_parse(const struct sk_buff *skb, unsigned char *haddr);
 extern struct ctl_table ether_table[];
 #endif
 
+extern ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len);
+
 /*
  *	Display a 6 byte device address (MAC) in a readable format.
  */
-#define MAC_FMT "%02x:%02x:%02x:%02x:%02x:%02x"
-extern char *print_mac(char *buf, const u8 *addr);
-#define DECLARE_MAC_BUF(var) char var[18] __maybe_unused
+extern char *print_mac(char *buf, const unsigned char *addr);
+#define MAC_BUF_SIZE	18
+#define DECLARE_MAC_BUF(var) char var[MAC_BUF_SIZE] __maybe_unused
 
 #endif
 
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index e41f4b9d2e7e..7635d3f72723 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -95,17 +95,6 @@ NETDEVICE_SHOW(type, fmt_dec);
 NETDEVICE_SHOW(link_mode, fmt_dec);
 
 /* use same locking rules as GIFHWADDR ioctl's */
-static ssize_t format_addr(char *buf, const unsigned char *addr, int len)
-{
-	int i;
-	char *cp = buf;
-
-	for (i = 0; i < len; i++)
-		cp += sprintf(cp, "%02x%c", addr[i],
-			      i == (len - 1) ? '\n' : ':');
-	return cp - buf;
-}
-
 static ssize_t show_address(struct device *dev, struct device_attribute *attr,
 			    char *buf)
 {
@@ -114,7 +103,7 @@ static ssize_t show_address(struct device *dev, struct device_attribute *attr,
 
 	read_lock(&dev_base_lock);
 	if (dev_isalive(net))
-	    ret = format_addr(buf, net->dev_addr, net->addr_len);
+		ret = sysfs_format_mac(buf, net->dev_addr, net->addr_len);
 	read_unlock(&dev_base_lock);
 	return ret;
 }
@@ -124,7 +113,7 @@ static ssize_t show_broadcast(struct device *dev,
 {
 	struct net_device *net = to_net_dev(dev);
 	if (dev_isalive(net))
-		return format_addr(buf, net->broadcast, net->addr_len);
+		return sysfs_format_mac(buf, net->broadcast, net->addr_len);
 	return -EINVAL;
 }
 
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 6b2e454ae313..a7b417523e9b 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -359,10 +359,34 @@ struct net_device *alloc_etherdev_mq(int sizeof_priv, unsigned int queue_count)
 }
 EXPORT_SYMBOL(alloc_etherdev_mq);
 
-char *print_mac(char *buf, const u8 *addr)
+static size_t _format_mac_addr(char *buf, int buflen,
+				const unsigned char *addr, int len)
 {
-	sprintf(buf, MAC_FMT,
-		addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
+	int i;
+	char *cp = buf;
+
+	for (i = 0; i < len; i++) {
+		cp += scnprintf(cp, buflen - (cp - buf), "%02x", addr[i]);
+		if (i == len - 1)
+			break;
+		cp += strlcpy(cp, ":", buflen - (cp - buf));
+	}
+	return cp - buf;
+}
+
+ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len)
+{
+	size_t l;
+
+	l = _format_mac_addr(buf, PAGE_SIZE, addr, len);
+	l += strlcpy(buf + l, "\n", PAGE_SIZE - l);
+	return ((ssize_t) l);
+}
+EXPORT_SYMBOL(sysfs_format_mac);
+
+char *print_mac(char *buf, const unsigned char *addr)
+{
+	_format_mac_addr(buf, MAC_BUF_SIZE, addr, ETH_ALEN);
 	return buf;
 }
 EXPORT_SYMBOL(print_mac);
-- 
cgit v1.2.3


From ef39592f786b6d56d9faf988a3f18786eeb050b3 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Sun, 30 Dec 2007 23:16:06 -0800
Subject: [ATM]: Convert struct class_device to struct device

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
---
 include/linux/atmdev.h |  4 +--
 net/atm/atm_sysfs.c    | 66 +++++++++++++++++++++++++++-----------------------
 2 files changed, 38 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/linux/atmdev.h b/include/linux/atmdev.h
index 2096e5c72827..a3d07c29d16c 100644
--- a/include/linux/atmdev.h
+++ b/include/linux/atmdev.h
@@ -359,7 +359,7 @@ struct atm_dev {
 	struct proc_dir_entry *proc_entry; /* proc entry */
 	char *proc_name;		/* proc entry name */
 #endif
-	struct class_device class_dev;	/* sysfs class device */
+	struct device class_dev;	/* sysfs device */
 	struct list_head dev_list;	/* linkage */
 };
 
@@ -461,7 +461,7 @@ static inline void atm_dev_put(struct atm_dev *dev)
 		BUG_ON(!test_bit(ATM_DF_REMOVED, &dev->flags));
 		if (dev->ops->dev_close)
 			dev->ops->dev_close(dev);
-		class_device_put(&dev->class_dev);
+		put_device(&dev->class_dev);
 	}
 }
 
diff --git a/net/atm/atm_sysfs.c b/net/atm/atm_sysfs.c
index 9ef07eda2c43..1b88311f2130 100644
--- a/net/atm/atm_sysfs.c
+++ b/net/atm/atm_sysfs.c
@@ -9,13 +9,15 @@
 
 #define to_atm_dev(cldev) container_of(cldev, struct atm_dev, class_dev)
 
-static ssize_t show_type(struct class_device *cdev, char *buf)
+static ssize_t show_type(struct device *cdev,
+			 struct device_attribute *attr, char *buf)
 {
 	struct atm_dev *adev = to_atm_dev(cdev);
 	return sprintf(buf, "%s\n", adev->type);
 }
 
-static ssize_t show_address(struct class_device *cdev, char *buf)
+static ssize_t show_address(struct device *cdev,
+			    struct device_attribute *attr, char *buf)
 {
 	char *pos = buf;
 	struct atm_dev *adev = to_atm_dev(cdev);
@@ -28,7 +30,8 @@ static ssize_t show_address(struct class_device *cdev, char *buf)
 	return pos - buf;
 }
 
-static ssize_t show_atmaddress(struct class_device *cdev, char *buf)
+static ssize_t show_atmaddress(struct device *cdev,
+			       struct device_attribute *attr, char *buf)
 {
 	unsigned long flags;
 	char *pos = buf;
@@ -54,7 +57,8 @@ static ssize_t show_atmaddress(struct class_device *cdev, char *buf)
 	return pos - buf;
 }
 
-static ssize_t show_carrier(struct class_device *cdev, char *buf)
+static ssize_t show_carrier(struct device *cdev,
+			    struct device_attribute *attr, char *buf)
 {
 	char *pos = buf;
 	struct atm_dev *adev = to_atm_dev(cdev);
@@ -65,7 +69,8 @@ static ssize_t show_carrier(struct class_device *cdev, char *buf)
 	return pos - buf;
 }
 
-static ssize_t show_link_rate(struct class_device *cdev, char *buf)
+static ssize_t show_link_rate(struct device *cdev,
+			      struct device_attribute *attr, char *buf)
 {
 	char *pos = buf;
 	struct atm_dev *adev = to_atm_dev(cdev);
@@ -90,22 +95,23 @@ static ssize_t show_link_rate(struct class_device *cdev, char *buf)
 	return pos - buf;
 }
 
-static CLASS_DEVICE_ATTR(address, S_IRUGO, show_address, NULL);
-static CLASS_DEVICE_ATTR(atmaddress, S_IRUGO, show_atmaddress, NULL);
-static CLASS_DEVICE_ATTR(carrier, S_IRUGO, show_carrier, NULL);
-static CLASS_DEVICE_ATTR(type, S_IRUGO, show_type, NULL);
-static CLASS_DEVICE_ATTR(link_rate, S_IRUGO, show_link_rate, NULL);
-
-static struct class_device_attribute *atm_attrs[] = {
-	&class_device_attr_atmaddress,
-	&class_device_attr_address,
-	&class_device_attr_carrier,
-	&class_device_attr_type,
-	&class_device_attr_link_rate,
+static DEVICE_ATTR(address, S_IRUGO, show_address, NULL);
+static DEVICE_ATTR(atmaddress, S_IRUGO, show_atmaddress, NULL);
+static DEVICE_ATTR(carrier, S_IRUGO, show_carrier, NULL);
+static DEVICE_ATTR(type, S_IRUGO, show_type, NULL);
+static DEVICE_ATTR(link_rate, S_IRUGO, show_link_rate, NULL);
+
+static struct device_attribute *atm_attrs[] = {
+	&dev_attr_atmaddress,
+	&dev_attr_address,
+	&dev_attr_carrier,
+	&dev_attr_type,
+	&dev_attr_link_rate,
 	NULL
 };
 
-static int atm_uevent(struct class_device *cdev, struct kobj_uevent_env *env)
+
+static int atm_uevent(struct device *cdev, struct kobj_uevent_env *env)
 {
 	struct atm_dev *adev;
 
@@ -122,7 +128,7 @@ static int atm_uevent(struct class_device *cdev, struct kobj_uevent_env *env)
 	return 0;
 }
 
-static void atm_release(struct class_device *cdev)
+static void atm_release(struct device *cdev)
 {
 	struct atm_dev *adev = to_atm_dev(cdev);
 
@@ -131,25 +137,25 @@ static void atm_release(struct class_device *cdev)
 
 static struct class atm_class = {
 	.name		= "atm",
-	.release	= atm_release,
-	.uevent		= atm_uevent,
+	.dev_release	= atm_release,
+	.dev_uevent		= atm_uevent,
 };
 
 int atm_register_sysfs(struct atm_dev *adev)
 {
-	struct class_device *cdev = &adev->class_dev;
+	struct device *cdev = &adev->class_dev;
 	int i, j, err;
 
 	cdev->class = &atm_class;
-	class_set_devdata(cdev, adev);
+	dev_set_drvdata(cdev, adev);
 
-	snprintf(cdev->class_id, BUS_ID_SIZE, "%s%d", adev->type, adev->number);
-	err = class_device_register(cdev);
+	snprintf(cdev->bus_id, BUS_ID_SIZE, "%s%d", adev->type, adev->number);
+	err = device_register(cdev);
 	if (err < 0)
 		return err;
 
 	for (i = 0; atm_attrs[i]; i++) {
-		err = class_device_create_file(cdev, atm_attrs[i]);
+		err = device_create_file(cdev, atm_attrs[i]);
 		if (err)
 			goto err_out;
 	}
@@ -158,16 +164,16 @@ int atm_register_sysfs(struct atm_dev *adev)
 
 err_out:
 	for (j = 0; j < i; j++)
-		class_device_remove_file(cdev, atm_attrs[j]);
-	class_device_del(cdev);
+		device_remove_file(cdev, atm_attrs[j]);
+	device_del(cdev);
 	return err;
 }
 
 void atm_unregister_sysfs(struct atm_dev *adev)
 {
-	struct class_device *cdev = &adev->class_dev;
+	struct device *cdev = &adev->class_dev;
 
-	class_device_del(cdev);
+	device_del(cdev);
 }
 
 int __init atm_sysfs_init(void)
-- 
cgit v1.2.3


From 097b19a9987204b898299260ee3ebff4cf716800 Mon Sep 17 00:00:00 2001
From: Eric Kinzie <ekinzie@cmf.nrl.navy.mil>
Date: Sun, 30 Dec 2007 23:17:53 -0800
Subject: [ATM]: [br2864] routed support

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atmbr2684.h |  18 +++++-
 net/atm/br2684.c          | 148 ++++++++++++++++++++++++++++++++++++----------
 2 files changed, 133 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/linux/atmbr2684.h b/include/linux/atmbr2684.h
index 969fb6c9e1cc..ccdab6c216cd 100644
--- a/include/linux/atmbr2684.h
+++ b/include/linux/atmbr2684.h
@@ -14,6 +14,9 @@
 #define BR2684_MEDIA_FDDI	(3)
 #define BR2684_MEDIA_802_6	(4)	/* 802.6 */
 
+					/* used only at device creation:  */
+#define BR2684_FLAG_ROUTED	(1<<16) /* payload is routed, not bridged */
+
 /*
  * Is there FCS inbound on this VC?  This currently isn't supported.
  */
@@ -35,6 +38,14 @@
 #define BR2684_ENCAPS_LLC	(1)
 #define BR2684_ENCAPS_AUTODETECT (2)	/* Unsuported */
 
+/*
+ * Is this VC bridged or routed?
+ */
+
+#define BR2684_PAYLOAD_ROUTED   (0)
+#define BR2684_PAYLOAD_BRIDGED  (1)
+
+
 /*
  * This is for the ATM_NEWBACKENDIF call - these are like socket families:
  * the first element of the structure is the backend number and the rest
@@ -42,7 +53,7 @@
  */
 struct atm_newif_br2684 {
 	atm_backend_t	backend_num;	/* ATM_BACKEND_BR2684 */
-	int		media;		/* BR2684_MEDIA_* */
+	int		media;		/* BR2684_MEDIA_*, flags in upper bits */
 	char		ifname[IFNAMSIZ];
 	int		mtu;
 };
@@ -95,6 +106,11 @@ struct br2684_filter_set {
 	struct br2684_filter filter;
 };
 
+enum br2684_payload {
+        p_routed = BR2684_PAYLOAD_ROUTED,
+        p_bridged = BR2684_PAYLOAD_BRIDGED,
+};
+
 #define BR2684_SETFILT	_IOW( 'a', ATMIOC_BACKEND + 0, \
 				struct br2684_filter_set)
 
diff --git a/net/atm/br2684.c b/net/atm/br2684.c
index ba6428f204f9..d9bb2a16b7cb 100644
--- a/net/atm/br2684.c
+++ b/net/atm/br2684.c
@@ -1,7 +1,8 @@
 /*
-Experimental ethernet netdevice using ATM AAL5 as underlying carrier
-(RFC1483 obsoleted by RFC2684) for Linux 2.4
-Author: Marcell GAL, 2000, XDSL Ltd, Hungary
+Ethernet netdevice using ATM AAL5 as underlying carrier
+(RFC1483 obsoleted by RFC2684) for Linux
+Authors: Marcell GAL, 2000, XDSL Ltd, Hungary
+         Eric Kinzie, 2006-2007, US Naval Research Laboratory
 */
 
 #include <linux/module.h>
@@ -39,9 +40,27 @@ static void skb_debug(const struct sk_buff *skb)
 #define skb_debug(skb)	do {} while (0)
 #endif
 
+#define BR2684_ETHERTYPE_LEN	2
+#define BR2684_PAD_LEN		2
+
+#define LLC		0xaa, 0xaa, 0x03
+#define SNAP_BRIDGED	0x00, 0x80, 0xc2
+#define SNAP_ROUTED	0x00, 0x00, 0x00
+#define PID_ETHERNET	0x00, 0x07
+#define ETHERTYPE_IPV4	0x08, 0x00
+#define ETHERTYPE_IPV6	0x86, 0xdd
+#define PAD_BRIDGED	0x00, 0x00
+
+static unsigned char ethertype_ipv4[] =
+	{ ETHERTYPE_IPV4 };
+static unsigned char ethertype_ipv6[] =
+	{ ETHERTYPE_IPV6 };
 static unsigned char llc_oui_pid_pad[] =
-    { 0xAA, 0xAA, 0x03, 0x00, 0x80, 0xC2, 0x00, 0x07, 0x00, 0x00 };
-#define PADLEN	(2)
+	{ LLC, SNAP_BRIDGED, PID_ETHERNET, PAD_BRIDGED };
+static unsigned char llc_oui_ipv4[] =
+	{ LLC, SNAP_ROUTED, ETHERTYPE_IPV4 };
+static unsigned char llc_oui_ipv6[] =
+	{ LLC, SNAP_ROUTED, ETHERTYPE_IPV6 };
 
 enum br2684_encaps {
 	e_vc  = BR2684_ENCAPS_VC,
@@ -69,6 +88,7 @@ struct br2684_dev {
 	struct list_head brvccs; /* one device <=> one vcc (before xmas) */
 	struct net_device_stats stats;
 	int mac_was_set;
+	enum br2684_payload payload;
 };
 
 /*
@@ -136,6 +156,7 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
 {
 	struct atm_vcc *atmvcc;
 	int minheadroom = (brvcc->encaps == e_llc) ? 10 : 2;
+
 	if (skb_headroom(skb) < minheadroom) {
 		struct sk_buff *skb2 = skb_realloc_headroom(skb, minheadroom);
 		brvcc->copies_needed++;
@@ -146,11 +167,32 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
 		}
 		skb = skb2;
 	}
-	skb_push(skb, minheadroom);
-	if (brvcc->encaps == e_llc)
-		skb_copy_to_linear_data(skb, llc_oui_pid_pad, 10);
-	else
-		memset(skb->data, 0, 2);
+
+	if (brvcc->encaps == e_llc) {
+		if (brdev->payload == p_bridged) {
+			skb_push(skb, sizeof(llc_oui_pid_pad));
+			skb_copy_to_linear_data(skb, llc_oui_pid_pad, sizeof(llc_oui_pid_pad));
+		} else if (brdev->payload == p_routed) {
+			unsigned short prot = ntohs(skb->protocol);
+
+			skb_push(skb, sizeof(llc_oui_ipv4));
+			switch (prot) {
+				case ETH_P_IP:
+					skb_copy_to_linear_data(skb, llc_oui_ipv4, sizeof(llc_oui_ipv4));
+					break;
+				case ETH_P_IPV6:
+					skb_copy_to_linear_data(skb, llc_oui_ipv6, sizeof(llc_oui_ipv6));
+					break;
+				default:
+					dev_kfree_skb(skb);
+					return 0;
+			}
+		}
+	} else {
+		skb_push(skb, 2);
+		if (brdev->payload == p_bridged)
+			memset(skb->data, 0, 2);
+	}
 	skb_debug(skb);
 
 	ATM_SKB(skb)->vcc = atmvcc = brvcc->atmvcc;
@@ -299,7 +341,6 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 	struct br2684_vcc *brvcc = BR2684_VCC(atmvcc);
 	struct net_device *net_dev = brvcc->device;
 	struct br2684_dev *brdev = BRPRIV(net_dev);
-	int plen = sizeof(llc_oui_pid_pad) + ETH_HLEN;
 
 	pr_debug("br2684_push\n");
 
@@ -320,35 +361,50 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 	atm_return(atmvcc, skb->truesize);
 	pr_debug("skb from brdev %p\n", brdev);
 	if (brvcc->encaps == e_llc) {
+
+		if (skb->len > 7 && skb->data[7] == 0x01)
+			__skb_trim(skb, skb->len - 4);
+
+		/* accept packets that have "ipv[46]" in the snap header */
+		if ((skb->len >= (sizeof(llc_oui_ipv4)))
+		    && (memcmp(skb->data, llc_oui_ipv4, sizeof(llc_oui_ipv4) - BR2684_ETHERTYPE_LEN) == 0)) {
+			if (memcmp(skb->data + 6, ethertype_ipv6, sizeof(ethertype_ipv6)) == 0)
+				skb->protocol = __constant_htons(ETH_P_IPV6);
+			else if (memcmp(skb->data + 6, ethertype_ipv4, sizeof(ethertype_ipv4)) == 0)
+				skb->protocol = __constant_htons(ETH_P_IP);
+			else {
+				brdev->stats.rx_errors++;
+				dev_kfree_skb(skb);
+				return;
+			}
+			skb_pull(skb, sizeof(llc_oui_ipv4));
+			skb_reset_network_header(skb);
+			skb->pkt_type = PACKET_HOST;
+
 		/* let us waste some time for checking the encapsulation.
 		   Note, that only 7 char is checked so frames with a valid FCS
 		   are also accepted (but FCS is not checked of course) */
-		if (memcmp(skb->data, llc_oui_pid_pad, 7)) {
+		} else if ((skb->len >= sizeof(llc_oui_pid_pad)) &&
+		           (memcmp(skb->data, llc_oui_pid_pad, 7) == 0)) {
+			skb_pull(skb, sizeof(llc_oui_pid_pad));
+			skb->protocol = eth_type_trans(skb, net_dev);
+		} else {
 			brdev->stats.rx_errors++;
 			dev_kfree_skb(skb);
 			return;
 		}
 
-		/* Strip FCS if present */
-		if (skb->len > 7 && skb->data[7] == 0x01)
-			__skb_trim(skb, skb->len - 4);
 	} else {
-		plen = PADLEN + ETH_HLEN;	/* pad, dstmac,srcmac, ethtype */
 		/* first 2 chars should be 0 */
 		if (*((u16 *) (skb->data)) != 0) {
 			brdev->stats.rx_errors++;
 			dev_kfree_skb(skb);
 			return;
 		}
-	}
-	if (skb->len < plen) {
-		brdev->stats.rx_errors++;
-		dev_kfree_skb(skb);	/* dev_ not needed? */
-		return;
+		skb_pull(skb, BR2684_PAD_LEN + ETH_HLEN); /* pad, dstmac, srcmac, ethtype */
+		skb->protocol = eth_type_trans(skb, net_dev);
 	}
 
-	skb_pull(skb, plen - ETH_HLEN);
-	skb->protocol = eth_type_trans(skb, net_dev);
 #ifdef CONFIG_ATM_BR2684_IPFILTER
 	if (unlikely(packet_fails_filter(skb->protocol, brvcc, skb))) {
 		brdev->stats.rx_dropped++;
@@ -482,25 +538,52 @@ static void br2684_setup(struct net_device *netdev)
 	INIT_LIST_HEAD(&brdev->brvccs);
 }
 
+static void br2684_setup_routed(struct net_device *netdev)
+{
+	struct br2684_dev *brdev = BRPRIV(netdev);
+	brdev->net_dev = netdev;
+
+	netdev->hard_header_len = 0;
+	my_eth_mac_addr = netdev->set_mac_address;
+	netdev->set_mac_address = br2684_mac_addr;
+	netdev->hard_start_xmit = br2684_start_xmit;
+	netdev->get_stats = br2684_get_stats;
+	netdev->addr_len = 0;
+	netdev->mtu = 1500;
+	netdev->type = ARPHRD_PPP;
+	netdev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
+	netdev->tx_queue_len = 100;
+	INIT_LIST_HEAD(&brdev->brvccs);
+}
+
 static int br2684_create(void __user *arg)
 {
 	int err;
 	struct net_device *netdev;
 	struct br2684_dev *brdev;
 	struct atm_newif_br2684 ni;
+	enum br2684_payload payload;
 
 	pr_debug("br2684_create\n");
 
 	if (copy_from_user(&ni, arg, sizeof ni)) {
 		return -EFAULT;
 	}
+
+	if (ni.media & BR2684_FLAG_ROUTED)
+		payload = p_routed;
+	else
+		payload = p_bridged;
+	ni.media &= 0xffff; /* strip flags */
+
 	if (ni.media != BR2684_MEDIA_ETHERNET || ni.mtu != 1500) {
 		return -EINVAL;
 	}
 
 	netdev = alloc_netdev(sizeof(struct br2684_dev),
 			      ni.ifname[0] ? ni.ifname : "nas%d",
-			      br2684_setup);
+			      (payload == p_routed) ?
+			       br2684_setup_routed : br2684_setup);
 	if (!netdev)
 		return -ENOMEM;
 
@@ -516,6 +599,7 @@ static int br2684_create(void __user *arg)
 	}
 
 	write_lock_irq(&devs_lock);
+	brdev->payload = payload;
 	brdev->number = list_empty(&br2684_devs) ? 1 :
 	    BRPRIV(list_entry_brdev(br2684_devs.prev))->number + 1;
 	list_add_tail(&brdev->br2684_devs, &br2684_devs);
@@ -601,14 +685,14 @@ static int br2684_seq_show(struct seq_file *seq, void *v)
 		   brdev->mac_was_set ? "set" : "auto");
 
 	list_for_each_entry(brvcc, &brdev->brvccs, brvccs) {
-		seq_printf(seq, "  vcc %d.%d.%d: encaps=%s"
-				    ", failed copies %u/%u"
-				    "\n", brvcc->atmvcc->dev->number,
-				    brvcc->atmvcc->vpi, brvcc->atmvcc->vci,
-				    (brvcc->encaps == e_llc) ? "LLC" : "VC"
-				    , brvcc->copies_failed
-				    , brvcc->copies_needed
-				    );
+		seq_printf(seq, "  vcc %d.%d.%d: encaps=%s payload=%s"
+			        ", failed copies %u/%u"
+			        "\n", brvcc->atmvcc->dev->number,
+				      brvcc->atmvcc->vpi, brvcc->atmvcc->vci,
+				      (brvcc->encaps == e_llc) ? "LLC" : "VC",
+				      (brdev->payload == p_bridged) ? "bridged" : "routed",
+				      brvcc->copies_failed,
+				      brvcc->copies_needed);
 #ifdef CONFIG_ATM_BR2684_IPFILTER
 #define b1(var, byte)	((u8 *) &brvcc->filter.var)[byte]
 #define bs(var)		b1(var, 0), b1(var, 1), b1(var, 2), b1(var, 3)
-- 
cgit v1.2.3


From fb64c735a52f396aa261844b851cd820a80dee46 Mon Sep 17 00:00:00 2001
From: Chas Williams <chas@cmf.nrl.navy.mil>
Date: Sun, 30 Dec 2007 23:18:29 -0800
Subject: [ATM]: [br2864] whitespace cleanup

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atmbr2684.h |  43 +++++----
 net/atm/br2684.c          | 222 ++++++++++++++++++++++++----------------------
 2 files changed, 135 insertions(+), 130 deletions(-)

(limited to 'include')

diff --git a/include/linux/atmbr2684.h b/include/linux/atmbr2684.h
index ccdab6c216cd..52bf72affbba 100644
--- a/include/linux/atmbr2684.h
+++ b/include/linux/atmbr2684.h
@@ -15,7 +15,7 @@
 #define BR2684_MEDIA_802_6	(4)	/* 802.6 */
 
 					/* used only at device creation:  */
-#define BR2684_FLAG_ROUTED	(1<<16) /* payload is routed, not bridged */
+#define BR2684_FLAG_ROUTED	(1<<16)	/* payload is routed, not bridged */
 
 /*
  * Is there FCS inbound on this VC?  This currently isn't supported.
@@ -45,17 +45,16 @@
 #define BR2684_PAYLOAD_ROUTED   (0)
 #define BR2684_PAYLOAD_BRIDGED  (1)
 
-
 /*
  * This is for the ATM_NEWBACKENDIF call - these are like socket families:
  * the first element of the structure is the backend number and the rest
  * is per-backend specific
  */
 struct atm_newif_br2684 {
-	atm_backend_t	backend_num;	/* ATM_BACKEND_BR2684 */
-	int		media;		/* BR2684_MEDIA_*, flags in upper bits */
-	char		ifname[IFNAMSIZ];
-	int		mtu;
+	atm_backend_t backend_num;	/* ATM_BACKEND_BR2684 */
+	int media;		/* BR2684_MEDIA_*, flags in upper bits */
+	char ifname[IFNAMSIZ];
+	int mtu;
 };
 
 /*
@@ -66,10 +65,10 @@ struct atm_newif_br2684 {
 #define BR2684_FIND_BYNUM	(1)
 #define BR2684_FIND_BYIFNAME	(2)
 struct br2684_if_spec {
-	int method;			/* BR2684_FIND_* */
+	int method;		/* BR2684_FIND_* */
 	union {
-		char		ifname[IFNAMSIZ];
-		int		devnum;
+		char ifname[IFNAMSIZ];
+		int devnum;
 	} spec;
 };
 
@@ -79,16 +78,16 @@ struct br2684_if_spec {
  * is per-backend specific
  */
 struct atm_backend_br2684 {
-	atm_backend_t	backend_num;	/* ATM_BACKEND_BR2684 */
+	atm_backend_t backend_num;	/* ATM_BACKEND_BR2684 */
 	struct br2684_if_spec ifspec;
-	int	fcs_in;		/* BR2684_FCSIN_* */
-	int	fcs_out;	/* BR2684_FCSOUT_* */
-	int	fcs_auto;	/* 1: fcs_{in,out} disabled if no FCS rx'ed */
-	int	encaps;		/* BR2684_ENCAPS_* */
-	int	has_vpiid;	/* 1: use vpn_id - Unsupported */
-	__u8	vpn_id[7];
-	int	send_padding;	/* unsupported */
-	int	min_size;	/* we will pad smaller packets than this */
+	int fcs_in;		/* BR2684_FCSIN_* */
+	int fcs_out;		/* BR2684_FCSOUT_* */
+	int fcs_auto;		/* 1: fcs_{in,out} disabled if no FCS rx'ed */
+	int encaps;		/* BR2684_ENCAPS_* */
+	int has_vpiid;		/* 1: use vpn_id - Unsupported */
+	__u8 vpn_id[7];
+	int send_padding;	/* unsupported */
+	int min_size;		/* we will pad smaller packets than this */
 };
 
 /*
@@ -97,8 +96,8 @@ struct atm_backend_br2684 {
  * efficient per-if in/out filters, this support will be removed
  */
 struct br2684_filter {
-	__be32	prefix;		/* network byte order */
-	__be32	netmask;	/* 0 = disable filter */
+	__be32 prefix;		/* network byte order */
+	__be32 netmask;		/* 0 = disable filter */
 };
 
 struct br2684_filter_set {
@@ -107,8 +106,8 @@ struct br2684_filter_set {
 };
 
 enum br2684_payload {
-        p_routed = BR2684_PAYLOAD_ROUTED,
-        p_bridged = BR2684_PAYLOAD_BRIDGED,
+	p_routed = BR2684_PAYLOAD_ROUTED,
+	p_bridged = BR2684_PAYLOAD_BRIDGED,
 };
 
 #define BR2684_SETFILT	_IOW( 'a', ATMIOC_BACKEND + 0, \
diff --git a/net/atm/br2684.c b/net/atm/br2684.c
index d9bb2a16b7cb..5c8a0dcb1045 100644
--- a/net/atm/br2684.c
+++ b/net/atm/br2684.c
@@ -1,9 +1,10 @@
 /*
-Ethernet netdevice using ATM AAL5 as underlying carrier
-(RFC1483 obsoleted by RFC2684) for Linux
-Authors: Marcell GAL, 2000, XDSL Ltd, Hungary
-         Eric Kinzie, 2006-2007, US Naval Research Laboratory
-*/
+ * Ethernet netdevice using ATM AAL5 as underlying carrier
+ * (RFC1483 obsoleted by RFC2684) for Linux
+ *
+ * Authors: Marcell GAL, 2000, XDSL Ltd, Hungary
+ *          Eric Kinzie, 2006-2007, US Naval Research Laboratory
+ */
 
 #include <linux/module.h>
 #include <linux/init.h>
@@ -51,28 +52,24 @@ static void skb_debug(const struct sk_buff *skb)
 #define ETHERTYPE_IPV6	0x86, 0xdd
 #define PAD_BRIDGED	0x00, 0x00
 
-static unsigned char ethertype_ipv4[] =
-	{ ETHERTYPE_IPV4 };
-static unsigned char ethertype_ipv6[] =
-	{ ETHERTYPE_IPV6 };
+static unsigned char ethertype_ipv4[] = { ETHERTYPE_IPV4 };
+static unsigned char ethertype_ipv6[] = { ETHERTYPE_IPV6 };
 static unsigned char llc_oui_pid_pad[] =
-	{ LLC, SNAP_BRIDGED, PID_ETHERNET, PAD_BRIDGED };
-static unsigned char llc_oui_ipv4[] =
-	{ LLC, SNAP_ROUTED, ETHERTYPE_IPV4 };
-static unsigned char llc_oui_ipv6[] =
-	{ LLC, SNAP_ROUTED, ETHERTYPE_IPV6 };
+			{ LLC, SNAP_BRIDGED, PID_ETHERNET, PAD_BRIDGED };
+static unsigned char llc_oui_ipv4[] = { LLC, SNAP_ROUTED, ETHERTYPE_IPV4 };
+static unsigned char llc_oui_ipv6[] = { LLC, SNAP_ROUTED, ETHERTYPE_IPV6 };
 
 enum br2684_encaps {
-	e_vc  = BR2684_ENCAPS_VC,
+	e_vc = BR2684_ENCAPS_VC,
 	e_llc = BR2684_ENCAPS_LLC,
 };
 
 struct br2684_vcc {
-	struct atm_vcc  *atmvcc;
+	struct atm_vcc *atmvcc;
 	struct net_device *device;
-	/* keep old push,pop functions for chaining */
-	void (*old_push)(struct atm_vcc *vcc,struct sk_buff *skb);
-	/* void (*old_pop)(struct atm_vcc *vcc,struct sk_buff *skb); */
+	/* keep old push, pop functions for chaining */
+	void (*old_push) (struct atm_vcc * vcc, struct sk_buff * skb);
+	/* void (*old_pop)(struct atm_vcc *vcc, struct sk_buff *skb); */
 	enum br2684_encaps encaps;
 	struct list_head brvccs;
 #ifdef CONFIG_ATM_BR2684_IPFILTER
@@ -85,7 +82,7 @@ struct br2684_dev {
 	struct net_device *net_dev;
 	struct list_head br2684_devs;
 	int number;
-	struct list_head brvccs; /* one device <=> one vcc (before xmas) */
+	struct list_head brvccs;	/* one device <=> one vcc (before xmas) */
 	struct net_device_stats stats;
 	int mac_was_set;
 	enum br2684_payload payload;
@@ -104,7 +101,7 @@ static LIST_HEAD(br2684_devs);
 
 static inline struct br2684_dev *BRPRIV(const struct net_device *net_dev)
 {
-	return (struct br2684_dev *) net_dev->priv;
+	return (struct br2684_dev *)net_dev->priv;
 }
 
 static inline struct net_device *list_entry_brdev(const struct list_head *le)
@@ -114,7 +111,7 @@ static inline struct net_device *list_entry_brdev(const struct list_head *le)
 
 static inline struct br2684_vcc *BR2684_VCC(const struct atm_vcc *atmvcc)
 {
-	return (struct br2684_vcc *) (atmvcc->user_back);
+	return (struct br2684_vcc *)(atmvcc->user_back);
 }
 
 static inline struct br2684_vcc *list_entry_brvcc(const struct list_head *le)
@@ -152,7 +149,7 @@ static struct net_device *br2684_find_dev(const struct br2684_if_spec *s)
  * otherwise false
  */
 static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
-	struct br2684_vcc *brvcc)
+			   struct br2684_vcc *brvcc)
 {
 	struct atm_vcc *atmvcc;
 	int minheadroom = (brvcc->encaps == e_llc) ? 10 : 2;
@@ -171,21 +168,24 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
 	if (brvcc->encaps == e_llc) {
 		if (brdev->payload == p_bridged) {
 			skb_push(skb, sizeof(llc_oui_pid_pad));
-			skb_copy_to_linear_data(skb, llc_oui_pid_pad, sizeof(llc_oui_pid_pad));
+			skb_copy_to_linear_data(skb, llc_oui_pid_pad,
+						sizeof(llc_oui_pid_pad));
 		} else if (brdev->payload == p_routed) {
 			unsigned short prot = ntohs(skb->protocol);
 
 			skb_push(skb, sizeof(llc_oui_ipv4));
 			switch (prot) {
-				case ETH_P_IP:
-					skb_copy_to_linear_data(skb, llc_oui_ipv4, sizeof(llc_oui_ipv4));
-					break;
-				case ETH_P_IPV6:
-					skb_copy_to_linear_data(skb, llc_oui_ipv6, sizeof(llc_oui_ipv6));
-					break;
-				default:
-					dev_kfree_skb(skb);
-					return 0;
+			case ETH_P_IP:
+				skb_copy_to_linear_data(skb, llc_oui_ipv4,
+							sizeof(llc_oui_ipv4));
+				break;
+			case ETH_P_IPV6:
+				skb_copy_to_linear_data(skb, llc_oui_ipv6,
+							sizeof(llc_oui_ipv6));
+				break;
+			default:
+				dev_kfree_skb(skb);
+				return 0;
 			}
 		}
 	} else {
@@ -198,13 +198,14 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
 	ATM_SKB(skb)->vcc = atmvcc = brvcc->atmvcc;
 	pr_debug("atm_skb(%p)->vcc(%p)->dev(%p)\n", skb, atmvcc, atmvcc->dev);
 	if (!atm_may_send(atmvcc, skb->truesize)) {
-		/* we free this here for now, because we cannot know in a higher
-			layer whether the skb point it supplied wasn't freed yet.
-			now, it always is.
-		*/
+		/*
+		 * We free this here for now, because we cannot know in a higher
+		 * layer whether the skb pointer it supplied wasn't freed yet.
+		 * Now, it always is.
+		 */
 		dev_kfree_skb(skb);
 		return 0;
-		}
+	}
 	atomic_add(skb->truesize, &sk_atm(atmvcc)->sk_wmem_alloc);
 	ATM_SKB(skb)->atm_options = atmvcc->atm_options;
 	brdev->stats.tx_packets++;
@@ -214,10 +215,9 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
 }
 
 static inline struct br2684_vcc *pick_outgoing_vcc(struct sk_buff *skb,
-	struct br2684_dev *brdev)
+						   struct br2684_dev *brdev)
 {
-	return list_empty(&brdev->brvccs) ? NULL :
-	    list_entry_brvcc(brdev->brvccs.next); /* 1 vcc/dev right now */
+	return list_empty(&brdev->brvccs) ? NULL : list_entry_brvcc(brdev->brvccs.next);	/* 1 vcc/dev right now */
 }
 
 static int br2684_start_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -241,11 +241,10 @@ static int br2684_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		/*
 		 * We should probably use netif_*_queue() here, but that
 		 * involves added complication.  We need to walk before
-		 * we can run
+		 * we can run.
+		 *
+		 * Don't free here! this pointer might be no longer valid!
 		 */
-		/* don't free here! this pointer might be no longer valid!
-		dev_kfree_skb(skb);
-		*/
 		brdev->stats.tx_errors++;
 		brdev->stats.tx_fifo_errors++;
 	}
@@ -259,12 +258,11 @@ static struct net_device_stats *br2684_get_stats(struct net_device *dev)
 	return &BRPRIV(dev)->stats;
 }
 
-
 /*
  * We remember when the MAC gets set, so we don't override it later with
  * the ESI of the ATM card of the first VC
  */
-static int (*my_eth_mac_addr)(struct net_device *, void *);
+static int (*my_eth_mac_addr) (struct net_device *, void *);
 static int br2684_mac_addr(struct net_device *dev, void *p)
 {
 	int err = my_eth_mac_addr(dev, p);
@@ -275,7 +273,7 @@ static int br2684_mac_addr(struct net_device *dev, void *p)
 
 #ifdef CONFIG_ATM_BR2684_IPFILTER
 /* this IOCTL is experimental. */
-static int br2684_setfilt(struct atm_vcc *atmvcc, void __user *arg)
+static int br2684_setfilt(struct atm_vcc *atmvcc, void __user * arg)
 {
 	struct br2684_vcc *brvcc;
 	struct br2684_filter_set fs;
@@ -285,13 +283,12 @@ static int br2684_setfilt(struct atm_vcc *atmvcc, void __user *arg)
 	if (fs.ifspec.method != BR2684_FIND_BYNOTHING) {
 		/*
 		 * This is really a per-vcc thing, but we can also search
-		 * by device
+		 * by device.
 		 */
 		struct br2684_dev *brdev;
 		read_lock(&devs_lock);
 		brdev = BRPRIV(br2684_find_dev(&fs.ifspec));
-		if (brdev == NULL || list_empty(&brdev->brvccs) ||
-		    brdev->brvccs.next != brdev->brvccs.prev)  /* >1 VCC */
+		if (brdev == NULL || list_empty(&brdev->brvccs) || brdev->brvccs.next != brdev->brvccs.prev)	/* >1 VCC */
 			brvcc = NULL;
 		else
 			brvcc = list_entry_brvcc(brdev->brvccs.next);
@@ -309,15 +306,16 @@ static inline int
 packet_fails_filter(__be16 type, struct br2684_vcc *brvcc, struct sk_buff *skb)
 {
 	if (brvcc->filter.netmask == 0)
-		return 0;			/* no filter in place */
+		return 0;	/* no filter in place */
 	if (type == htons(ETH_P_IP) &&
-	    (((struct iphdr *) (skb->data))->daddr & brvcc->filter.
+	    (((struct iphdr *)(skb->data))->daddr & brvcc->filter.
 	     netmask) == brvcc->filter.prefix)
 		return 0;
 	if (type == htons(ETH_P_ARP))
 		return 0;
-	/* TODO: we should probably filter ARPs too.. don't want to have
-	 *   them returning values that don't make sense, or is that ok?
+	/*
+	 * TODO: we should probably filter ARPs too.. don't want to have
+	 * them returning values that don't make sense, or is that ok?
 	 */
 	return 1;		/* drop */
 }
@@ -367,10 +365,17 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 
 		/* accept packets that have "ipv[46]" in the snap header */
 		if ((skb->len >= (sizeof(llc_oui_ipv4)))
-		    && (memcmp(skb->data, llc_oui_ipv4, sizeof(llc_oui_ipv4) - BR2684_ETHERTYPE_LEN) == 0)) {
-			if (memcmp(skb->data + 6, ethertype_ipv6, sizeof(ethertype_ipv6)) == 0)
+		    &&
+		    (memcmp
+		     (skb->data, llc_oui_ipv4,
+		      sizeof(llc_oui_ipv4) - BR2684_ETHERTYPE_LEN) == 0)) {
+			if (memcmp
+			    (skb->data + 6, ethertype_ipv6,
+			     sizeof(ethertype_ipv6)) == 0)
 				skb->protocol = __constant_htons(ETH_P_IPV6);
-			else if (memcmp(skb->data + 6, ethertype_ipv4, sizeof(ethertype_ipv4)) == 0)
+			else if (memcmp
+				 (skb->data + 6, ethertype_ipv4,
+				  sizeof(ethertype_ipv4)) == 0)
 				skb->protocol = __constant_htons(ETH_P_IP);
 			else {
 				brdev->stats.rx_errors++;
@@ -380,12 +385,13 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 			skb_pull(skb, sizeof(llc_oui_ipv4));
 			skb_reset_network_header(skb);
 			skb->pkt_type = PACKET_HOST;
-
-		/* let us waste some time for checking the encapsulation.
-		   Note, that only 7 char is checked so frames with a valid FCS
-		   are also accepted (but FCS is not checked of course) */
+			/*
+			 * Let us waste some time for checking the encapsulation.
+			 * Note, that only 7 char is checked so frames with a valid FCS
+			 * are also accepted (but FCS is not checked of course).
+			 */
 		} else if ((skb->len >= sizeof(llc_oui_pid_pad)) &&
-		           (memcmp(skb->data, llc_oui_pid_pad, 7) == 0)) {
+			   (memcmp(skb->data, llc_oui_pid_pad, 7) == 0)) {
 			skb_pull(skb, sizeof(llc_oui_pid_pad));
 			skb->protocol = eth_type_trans(skb, net_dev);
 		} else {
@@ -401,7 +407,7 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 			dev_kfree_skb(skb);
 			return;
 		}
-		skb_pull(skb, BR2684_PAD_LEN + ETH_HLEN); /* pad, dstmac, srcmac, ethtype */
+		skb_pull(skb, BR2684_PAD_LEN + ETH_HLEN);	/* pad, dstmac, srcmac, ethtype */
 		skb->protocol = eth_type_trans(skb, net_dev);
 	}
 
@@ -428,11 +434,12 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
 	netif_rx(skb);
 }
 
-static int br2684_regvcc(struct atm_vcc *atmvcc, void __user *arg)
+/*
+ * Assign a vcc to a dev
+ * Note: we do not have explicit unassign, but look at _push()
+ */
+static int br2684_regvcc(struct atm_vcc *atmvcc, void __user * arg)
 {
-/* assign a vcc to a dev
-Note: we do not have explicit unassign, but look at _push()
-*/
 	int err;
 	struct br2684_vcc *brvcc;
 	struct sk_buff *skb;
@@ -451,7 +458,7 @@ Note: we do not have explicit unassign, but look at _push()
 	net_dev = br2684_find_dev(&be.ifspec);
 	if (net_dev == NULL) {
 		printk(KERN_ERR
-		    "br2684: tried to attach to non-existant device\n");
+		       "br2684: tried to attach to non-existant device\n");
 		err = -ENXIO;
 		goto error;
 	}
@@ -467,13 +474,15 @@ Note: we do not have explicit unassign, but look at _push()
 	}
 	if (be.fcs_in != BR2684_FCSIN_NO || be.fcs_out != BR2684_FCSOUT_NO ||
 	    be.fcs_auto || be.has_vpiid || be.send_padding || (be.encaps !=
-	    BR2684_ENCAPS_VC && be.encaps != BR2684_ENCAPS_LLC) ||
-	    be.min_size != 0) {
+							       BR2684_ENCAPS_VC
+							       && be.encaps !=
+							       BR2684_ENCAPS_LLC)
+	    || be.min_size != 0) {
 		err = -EINVAL;
 		goto error;
 	}
-	pr_debug("br2684_regvcc vcc=%p, encaps=%d, brvcc=%p\n", atmvcc, be.encaps,
-		brvcc);
+	pr_debug("br2684_regvcc vcc=%p, encaps=%d, brvcc=%p\n", atmvcc,
+		 be.encaps, brvcc);
 	if (list_empty(&brdev->brvccs) && !brdev->mac_was_set) {
 		unsigned char *esi = atmvcc->dev->esi;
 		if (esi[0] | esi[1] | esi[2] | esi[3] | esi[4] | esi[5])
@@ -486,7 +495,7 @@ Note: we do not have explicit unassign, but look at _push()
 	brvcc->device = net_dev;
 	brvcc->atmvcc = atmvcc;
 	atmvcc->user_back = brvcc;
-	brvcc->encaps = (enum br2684_encaps) be.encaps;
+	brvcc->encaps = (enum br2684_encaps)be.encaps;
 	brvcc->old_push = atmvcc->push;
 	barrier();
 	atmvcc->push = br2684_push;
@@ -517,7 +526,7 @@ Note: we do not have explicit unassign, but look at _push()
 	}
 	__module_get(THIS_MODULE);
 	return 0;
-    error:
+      error:
 	write_unlock_irq(&devs_lock);
 	kfree(brvcc);
 	return err;
@@ -556,7 +565,7 @@ static void br2684_setup_routed(struct net_device *netdev)
 	INIT_LIST_HEAD(&brdev->brvccs);
 }
 
-static int br2684_create(void __user *arg)
+static int br2684_create(void __user * arg)
 {
 	int err;
 	struct net_device *netdev;
@@ -574,7 +583,7 @@ static int br2684_create(void __user *arg)
 		payload = p_routed;
 	else
 		payload = p_bridged;
-	ni.media &= 0xffff; /* strip flags */
+	ni.media &= 0xffff;	/* strip flags */
 
 	if (ni.media != BR2684_MEDIA_ETHERNET || ni.mtu != 1500) {
 		return -EINVAL;
@@ -583,7 +592,7 @@ static int br2684_create(void __user *arg)
 	netdev = alloc_netdev(sizeof(struct br2684_dev),
 			      ni.ifname[0] ? ni.ifname : "nas%d",
 			      (payload == p_routed) ?
-			       br2684_setup_routed : br2684_setup);
+			      br2684_setup_routed : br2684_setup);
 	if (!netdev)
 		return -ENOMEM;
 
@@ -612,16 +621,16 @@ static int br2684_create(void __user *arg)
  * -ENOIOCTLCMD for any unrecognized ioctl
  */
 static int br2684_ioctl(struct socket *sock, unsigned int cmd,
-	unsigned long arg)
+			unsigned long arg)
 {
 	struct atm_vcc *atmvcc = ATM_SD(sock);
 	void __user *argp = (void __user *)arg;
+	atm_backend_t b;
 
 	int err;
-	switch(cmd) {
+	switch (cmd) {
 	case ATM_SETBACKEND:
-	case ATM_NEWBACKENDIF: {
-		atm_backend_t b;
+	case ATM_NEWBACKENDIF:
 		err = get_user(b, (atm_backend_t __user *) argp);
 		if (err)
 			return -EFAULT;
@@ -633,7 +642,6 @@ static int br2684_ioctl(struct socket *sock, unsigned int cmd,
 			return br2684_regvcc(atmvcc, argp);
 		else
 			return br2684_create(argp);
-		}
 #ifdef CONFIG_ATM_BR2684_IPFILTER
 	case BR2684_SETFILT:
 		if (atmvcc->push != br2684_push)
@@ -641,6 +649,7 @@ static int br2684_ioctl(struct socket *sock, unsigned int cmd,
 		if (!capable(CAP_NET_ADMIN))
 			return -EPERM;
 		err = br2684_setfilt(atmvcc, argp);
+
 		return err;
 #endif /* CONFIG_ATM_BR2684_IPFILTER */
 	}
@@ -648,19 +657,18 @@ static int br2684_ioctl(struct socket *sock, unsigned int cmd,
 }
 
 static struct atm_ioctl br2684_ioctl_ops = {
-	.owner	= THIS_MODULE,
-	.ioctl	= br2684_ioctl,
+	.owner = THIS_MODULE,
+	.ioctl = br2684_ioctl,
 };
 
-
 #ifdef CONFIG_PROC_FS
-static void *br2684_seq_start(struct seq_file *seq, loff_t *pos)
+static void *br2684_seq_start(struct seq_file *seq, loff_t * pos)
 {
 	read_lock(&devs_lock);
 	return seq_list_start(&br2684_devs, *pos);
 }
 
-static void *br2684_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+static void *br2684_seq_next(struct seq_file *seq, void *v, loff_t * pos)
 {
 	return seq_list_next(v, &br2684_devs, pos);
 }
@@ -673,7 +681,7 @@ static void br2684_seq_stop(struct seq_file *seq, void *v)
 static int br2684_seq_show(struct seq_file *seq, void *v)
 {
 	const struct br2684_dev *brdev = list_entry(v, struct br2684_dev,
-			br2684_devs);
+						    br2684_devs);
 	const struct net_device *net_dev = brdev->net_dev;
 	const struct br2684_vcc *brvcc;
 	DECLARE_MAC_BUF(mac);
@@ -686,20 +694,18 @@ static int br2684_seq_show(struct seq_file *seq, void *v)
 
 	list_for_each_entry(brvcc, &brdev->brvccs, brvccs) {
 		seq_printf(seq, "  vcc %d.%d.%d: encaps=%s payload=%s"
-			        ", failed copies %u/%u"
-			        "\n", brvcc->atmvcc->dev->number,
-				      brvcc->atmvcc->vpi, brvcc->atmvcc->vci,
-				      (brvcc->encaps == e_llc) ? "LLC" : "VC",
-				      (brdev->payload == p_bridged) ? "bridged" : "routed",
-				      brvcc->copies_failed,
-				      brvcc->copies_needed);
+			   ", failed copies %u/%u"
+			   "\n", brvcc->atmvcc->dev->number,
+			   brvcc->atmvcc->vpi, brvcc->atmvcc->vci,
+			   (brvcc->encaps == e_llc) ? "LLC" : "VC",
+			   (brdev->payload == p_bridged) ? "bridged" : "routed",
+			   brvcc->copies_failed, brvcc->copies_needed);
 #ifdef CONFIG_ATM_BR2684_IPFILTER
 #define b1(var, byte)	((u8 *) &brvcc->filter.var)[byte]
 #define bs(var)		b1(var, 0), b1(var, 1), b1(var, 2), b1(var, 3)
-			if (brvcc->filter.netmask != 0)
-				seq_printf(seq, "    filter=%d.%d.%d.%d/"
-						"%d.%d.%d.%d\n",
-						bs(prefix), bs(netmask));
+		if (brvcc->filter.netmask != 0)
+			seq_printf(seq, "    filter=%d.%d.%d.%d/"
+				   "%d.%d.%d.%d\n", bs(prefix), bs(netmask));
 #undef bs
 #undef b1
 #endif /* CONFIG_ATM_BR2684_IPFILTER */
@@ -709,9 +715,9 @@ static int br2684_seq_show(struct seq_file *seq, void *v)
 
 static const struct seq_operations br2684_seq_ops = {
 	.start = br2684_seq_start,
-	.next  = br2684_seq_next,
-	.stop  = br2684_seq_stop,
-	.show  = br2684_seq_show,
+	.next = br2684_seq_next,
+	.stop = br2684_seq_stop,
+	.show = br2684_seq_show,
 };
 
 static int br2684_proc_open(struct inode *inode, struct file *file)
@@ -720,15 +726,15 @@ static int br2684_proc_open(struct inode *inode, struct file *file)
 }
 
 static const struct file_operations br2684_proc_ops = {
-	.owner   = THIS_MODULE,
-	.open    = br2684_proc_open,
-	.read    = seq_read,
-	.llseek  = seq_lseek,
+	.owner = THIS_MODULE,
+	.open = br2684_proc_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
 	.release = seq_release,
 };
 
 extern struct proc_dir_entry *atm_proc_root;	/* from proc.c */
-#endif
+#endif /* CONFIG_PROC_FS */
 
 static int __init br2684_init(void)
 {
-- 
cgit v1.2.3


From 04ce99c4839a86f4ef476f811cced8d1f11999e4 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Sun, 30 Dec 2007 23:23:02 -0800
Subject: [IPV4]: Remove unused define in include/net/arp.h (HAVE_ARP_CREATE)

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/arp.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/arp.h b/include/net/arp.h
index 36482bf9ad36..752eb47b2678 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -5,7 +5,6 @@
 #include <linux/if_arp.h>
 #include <net/neighbour.h>
 
-#define HAVE_ARP_CREATE
 
 extern struct neigh_table arp_tbl;
 
-- 
cgit v1.2.3


From f624357959001c9156ee7a475283fc6041f78e0e Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Sun, 30 Dec 2007 23:25:31 -0800
Subject: [NEIGH]: Remove unused method from include/net/neighbour.h

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 11590f23d99d..a9dda2908211 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -292,10 +292,6 @@ static inline int neigh_is_connected(struct neighbour *neigh)
 	return neigh->nud_state&NUD_CONNECTED;
 }
 
-static inline int neigh_is_valid(struct neighbour *neigh)
-{
-	return neigh->nud_state&NUD_VALID;
-}
 
 static inline int neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
 {
-- 
cgit v1.2.3


From 3ab224be6d69de912ee21302745ea45a99274dbc Mon Sep 17 00:00:00 2001
From: Hideo Aoki <haoki@redhat.com>
Date: Mon, 31 Dec 2007 00:11:19 -0800
Subject: [NET] CORE: Introducing new memory accounting interface.

This patch introduces new memory accounting functions for each network
protocol. Most of them are renamed from memory accounting functions
for stream protocols. At the same time, some stream memory accounting
functions are removed since other functions do same thing.

Renaming:
	sk_stream_free_skb()		->	sk_wmem_free_skb()
	__sk_stream_mem_reclaim()	->	__sk_mem_reclaim()
	sk_stream_mem_reclaim()		->	sk_mem_reclaim()
	sk_stream_mem_schedule 		->    	__sk_mem_schedule()
	sk_stream_pages()      		->	sk_mem_pages()
	sk_stream_rmem_schedule()	->	sk_rmem_schedule()
	sk_stream_wmem_schedule()	->	sk_wmem_schedule()
	sk_charge_skb()			->	sk_mem_charge()

Removeing
	sk_stream_rfree():	consolidates into sock_rfree()
	sk_stream_set_owner_r(): consolidates into skb_set_owner_r()
	sk_stream_mem_schedule()

The following functions are added.
    	sk_has_account(): check if the protocol supports accounting
	sk_mem_uncharge(): do the opposite of sk_mem_charge()

In addition, to achieve consolidation, updating sk_wmem_queued is
removed from sk_mem_charge().

Next, to consolidate memory accounting functions, this patch adds
memory accounting calls to network core functions. Moreover, present
memory accounting call is renamed to new accounting call.

Finally we replace present memory accounting calls with new interface
in TCP and SCTP.

Signed-off-by: Takahiro Yasui <tyasui@redhat.com>
Signed-off-by: Hideo Aoki <haoki@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h |   3 +-
 include/net/sock.h      |  98 ++++++++++++++++++++++++++-------------------
 include/net/tcp.h       |   4 +-
 net/core/datagram.c     |   2 +
 net/core/sock.c         | 104 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/core/stream.c       |  84 +-------------------------------------
 net/ipv4/tcp.c          |  23 ++++++-----
 net/ipv4/tcp_input.c    |  26 ++++++------
 net/ipv4/tcp_output.c   |  26 +++++++-----
 net/ipv4/tcp_timer.c    |   8 ++--
 net/sctp/protocol.c     |   2 +-
 net/sctp/sm_statefuns.c |   2 +-
 net/sctp/socket.c       |  11 ++---
 net/sctp/ulpevent.c     |   2 +-
 net/sctp/ulpqueue.c     |   2 +-
 15 files changed, 222 insertions(+), 175 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 1b81ede7c2bc..4977b0a81535 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -463,8 +463,7 @@ static inline void sctp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 	skb->destructor = sctp_sock_rfree;
 	atomic_add(event->rmem_len, &sk->sk_rmem_alloc);
 	/*
-	 * This mimics the behavior of
-	 * sk_stream_set_owner_r
+	 * This mimics the behavior of skb_set_owner_r
 	 */
 	sk->sk_forward_alloc -= event->rmem_len;
 }
diff --git a/include/net/sock.h b/include/net/sock.h
index d27ba6fdd039..3d938f6c6725 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -460,25 +460,6 @@ static inline int sk_stream_memory_free(struct sock *sk)
 	return sk->sk_wmem_queued < sk->sk_sndbuf;
 }
 
-extern void sk_stream_rfree(struct sk_buff *skb);
-
-static inline void sk_stream_set_owner_r(struct sk_buff *skb, struct sock *sk)
-{
-	skb->sk = sk;
-	skb->destructor = sk_stream_rfree;
-	atomic_add(skb->truesize, &sk->sk_rmem_alloc);
-	sk->sk_forward_alloc -= skb->truesize;
-}
-
-static inline void sk_stream_free_skb(struct sock *sk, struct sk_buff *skb)
-{
-	skb_truesize_check(skb);
-	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
-	sk->sk_wmem_queued   -= skb->truesize;
-	sk->sk_forward_alloc += skb->truesize;
-	__kfree_skb(skb);
-}
-
 /* The per-socket spinlock must be held here. */
 static inline void sk_add_backlog(struct sock *sk, struct sk_buff *skb)
 {
@@ -576,7 +557,7 @@ struct proto {
 	/*
 	 * Pressure flag: try to collapse.
 	 * Technical note: it is used by multiple contexts non atomically.
-	 * All the sk_stream_mem_schedule() is of this nature: accounting
+	 * All the __sk_mem_schedule() is of this nature: accounting
 	 * is strict, actions are advisory and have some latency.
 	 */
 	int			*memory_pressure;
@@ -712,33 +693,73 @@ static inline struct inode *SOCK_INODE(struct socket *socket)
 	return &container_of(socket, struct socket_alloc, socket)->vfs_inode;
 }
 
-extern void __sk_stream_mem_reclaim(struct sock *sk);
-extern int sk_stream_mem_schedule(struct sock *sk, int size, int kind);
+/*
+ * Functions for memory accounting
+ */
+extern int __sk_mem_schedule(struct sock *sk, int size, int kind);
+extern void __sk_mem_reclaim(struct sock *sk);
 
-#define SK_STREAM_MEM_QUANTUM ((int)PAGE_SIZE)
-#define SK_STREAM_MEM_QUANTUM_SHIFT ilog2(SK_STREAM_MEM_QUANTUM)
+#define SK_MEM_QUANTUM ((int)PAGE_SIZE)
+#define SK_MEM_QUANTUM_SHIFT ilog2(SK_MEM_QUANTUM)
+#define SK_MEM_SEND	0
+#define SK_MEM_RECV	1
 
-static inline int sk_stream_pages(int amt)
+static inline int sk_mem_pages(int amt)
 {
-	return (amt + SK_STREAM_MEM_QUANTUM - 1) >> SK_STREAM_MEM_QUANTUM_SHIFT;
+	return (amt + SK_MEM_QUANTUM - 1) >> SK_MEM_QUANTUM_SHIFT;
 }
 
-static inline void sk_stream_mem_reclaim(struct sock *sk)
+static inline int sk_has_account(struct sock *sk)
 {
-	if (sk->sk_forward_alloc >= SK_STREAM_MEM_QUANTUM)
-		__sk_stream_mem_reclaim(sk);
+	/* return true if protocol supports memory accounting */
+	return !!sk->sk_prot->memory_allocated;
 }
 
-static inline int sk_stream_rmem_schedule(struct sock *sk, struct sk_buff *skb)
+static inline int sk_wmem_schedule(struct sock *sk, int size)
 {
-	return (int)skb->truesize <= sk->sk_forward_alloc ||
-		sk_stream_mem_schedule(sk, skb->truesize, 1);
+	if (!sk_has_account(sk))
+		return 1;
+	return size <= sk->sk_forward_alloc ||
+		__sk_mem_schedule(sk, size, SK_MEM_SEND);
 }
 
-static inline int sk_stream_wmem_schedule(struct sock *sk, int size)
+static inline int sk_rmem_schedule(struct sock *sk, int size)
 {
+	if (!sk_has_account(sk))
+		return 1;
 	return size <= sk->sk_forward_alloc ||
-	       sk_stream_mem_schedule(sk, size, 0);
+		__sk_mem_schedule(sk, size, SK_MEM_RECV);
+}
+
+static inline void sk_mem_reclaim(struct sock *sk)
+{
+	if (!sk_has_account(sk))
+		return;
+	if (sk->sk_forward_alloc >= SK_MEM_QUANTUM)
+		__sk_mem_reclaim(sk);
+}
+
+static inline void sk_mem_charge(struct sock *sk, int size)
+{
+	if (!sk_has_account(sk))
+		return;
+	sk->sk_forward_alloc -= size;
+}
+
+static inline void sk_mem_uncharge(struct sock *sk, int size)
+{
+	if (!sk_has_account(sk))
+		return;
+	sk->sk_forward_alloc += size;
+}
+
+static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb)
+{
+	skb_truesize_check(skb);
+	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
+	sk->sk_wmem_queued -= skb->truesize;
+	sk_mem_uncharge(sk, skb->truesize);
+	__kfree_skb(skb);
 }
 
 /* Used by processes to "lock" a socket state, so that
@@ -1076,12 +1097,6 @@ static inline int sk_can_gso(const struct sock *sk)
 
 extern void sk_setup_caps(struct sock *sk, struct dst_entry *dst);
 
-static inline void sk_charge_skb(struct sock *sk, struct sk_buff *skb)
-{
-	sk->sk_wmem_queued   += skb->truesize;
-	sk->sk_forward_alloc -= skb->truesize;
-}
-
 static inline int skb_copy_to_page(struct sock *sk, char __user *from,
 				   struct sk_buff *skb, struct page *page,
 				   int off, int copy)
@@ -1101,7 +1116,7 @@ static inline int skb_copy_to_page(struct sock *sk, char __user *from,
 	skb->data_len	     += copy;
 	skb->truesize	     += copy;
 	sk->sk_wmem_queued   += copy;
-	sk->sk_forward_alloc -= copy;
+	sk_mem_charge(sk, copy);
 	return 0;
 }
 
@@ -1127,6 +1142,7 @@ static inline void skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 	skb->sk = sk;
 	skb->destructor = sock_rfree;
 	atomic_add(skb->truesize, &sk->sk_rmem_alloc);
+	sk_mem_charge(sk, skb->truesize);
 }
 
 extern void sk_reset_timer(struct sock *sk, struct timer_list* timer,
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 13ebe11a0af7..76286e80205a 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1196,8 +1196,8 @@ static inline void tcp_write_queue_purge(struct sock *sk)
 	struct sk_buff *skb;
 
 	while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL)
-		sk_stream_free_skb(sk, skb);
-	sk_stream_mem_reclaim(sk);
+		sk_wmem_free_skb(sk, skb);
+	sk_mem_reclaim(sk);
 }
 
 static inline struct sk_buff *tcp_write_queue_head(struct sock *sk)
diff --git a/net/core/datagram.c b/net/core/datagram.c
index 2d733131d7ce..8a28fc93b724 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -209,6 +209,7 @@ struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
 void skb_free_datagram(struct sock *sk, struct sk_buff *skb)
 {
 	kfree_skb(skb);
+	sk_mem_reclaim(sk);
 }
 
 /**
@@ -248,6 +249,7 @@ int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags)
 	}
 
 	kfree_skb(skb);
+	sk_mem_reclaim(sk);
 	return err;
 }
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 118214047ed2..8c184c4a3811 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -282,6 +282,11 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (err)
 		goto out;
 
+	if (!sk_rmem_schedule(sk, skb->truesize)) {
+		err = -ENOBUFS;
+		goto out;
+	}
+
 	skb->dev = NULL;
 	skb_set_owner_r(skb, sk);
 
@@ -1107,7 +1112,9 @@ void sock_rfree(struct sk_buff *skb)
 {
 	struct sock *sk = skb->sk;
 
+	skb_truesize_check(skb);
 	atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
+	sk_mem_uncharge(skb->sk, skb->truesize);
 }
 
 
@@ -1384,6 +1391,103 @@ int sk_wait_data(struct sock *sk, long *timeo)
 
 EXPORT_SYMBOL(sk_wait_data);
 
+/**
+ *	__sk_mem_schedule - increase sk_forward_alloc and memory_allocated
+ *	@sk: socket
+ *	@size: memory size to allocate
+ *	@kind: allocation type
+ *
+ *	If kind is SK_MEM_SEND, it means wmem allocation. Otherwise it means
+ *	rmem allocation. This function assumes that protocols which have
+ *	memory_pressure use sk_wmem_queued as write buffer accounting.
+ */
+int __sk_mem_schedule(struct sock *sk, int size, int kind)
+{
+	struct proto *prot = sk->sk_prot;
+	int amt = sk_mem_pages(size);
+	int allocated;
+
+	sk->sk_forward_alloc += amt * SK_MEM_QUANTUM;
+	allocated = atomic_add_return(amt, prot->memory_allocated);
+
+	/* Under limit. */
+	if (allocated <= prot->sysctl_mem[0]) {
+		if (prot->memory_pressure && *prot->memory_pressure)
+			*prot->memory_pressure = 0;
+		return 1;
+	}
+
+	/* Under pressure. */
+	if (allocated > prot->sysctl_mem[1])
+		if (prot->enter_memory_pressure)
+			prot->enter_memory_pressure();
+
+	/* Over hard limit. */
+	if (allocated > prot->sysctl_mem[2])
+		goto suppress_allocation;
+
+	/* guarantee minimum buffer size under pressure */
+	if (kind == SK_MEM_RECV) {
+		if (atomic_read(&sk->sk_rmem_alloc) < prot->sysctl_rmem[0])
+			return 1;
+	} else { /* SK_MEM_SEND */
+		if (sk->sk_type == SOCK_STREAM) {
+			if (sk->sk_wmem_queued < prot->sysctl_wmem[0])
+				return 1;
+		} else if (atomic_read(&sk->sk_wmem_alloc) <
+			   prot->sysctl_wmem[0])
+				return 1;
+	}
+
+	if (prot->memory_pressure) {
+		if (!*prot->memory_pressure ||
+		    prot->sysctl_mem[2] > atomic_read(prot->sockets_allocated) *
+		    sk_mem_pages(sk->sk_wmem_queued +
+				 atomic_read(&sk->sk_rmem_alloc) +
+				 sk->sk_forward_alloc))
+			return 1;
+	}
+
+suppress_allocation:
+
+	if (kind == SK_MEM_SEND && sk->sk_type == SOCK_STREAM) {
+		sk_stream_moderate_sndbuf(sk);
+
+		/* Fail only if socket is _under_ its sndbuf.
+		 * In this case we cannot block, so that we have to fail.
+		 */
+		if (sk->sk_wmem_queued + size >= sk->sk_sndbuf)
+			return 1;
+	}
+
+	/* Alas. Undo changes. */
+	sk->sk_forward_alloc -= amt * SK_MEM_QUANTUM;
+	atomic_sub(amt, prot->memory_allocated);
+	return 0;
+}
+
+EXPORT_SYMBOL(__sk_mem_schedule);
+
+/**
+ *	__sk_reclaim - reclaim memory_allocated
+ *	@sk: socket
+ */
+void __sk_mem_reclaim(struct sock *sk)
+{
+	struct proto *prot = sk->sk_prot;
+
+	atomic_sub(sk->sk_forward_alloc / SK_MEM_QUANTUM,
+		   prot->memory_allocated);
+	sk->sk_forward_alloc &= SK_MEM_QUANTUM - 1;
+
+	if (prot->memory_pressure && *prot->memory_pressure &&
+	    (atomic_read(prot->memory_allocated) < prot->sysctl_mem[0]))
+		*prot->memory_pressure = 0;
+}
+
+EXPORT_SYMBOL(__sk_mem_reclaim);
+
+
 /*
  * Set of default routines for initialising struct proto_ops when
  * the protocol does not support a particular function. In certain
diff --git a/net/core/stream.c b/net/core/stream.c
index bf188ffdbdbe..4a0ad152c9c4 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -172,17 +172,6 @@ do_interrupted:
 
 EXPORT_SYMBOL(sk_stream_wait_memory);
 
-void sk_stream_rfree(struct sk_buff *skb)
-{
-	struct sock *sk = skb->sk;
-
-	skb_truesize_check(skb);
-	atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
-	sk->sk_forward_alloc += skb->truesize;
-}
-
-EXPORT_SYMBOL(sk_stream_rfree);
-
 int sk_stream_error(struct sock *sk, int flags, int err)
 {
 	if (err == -EPIPE)
@@ -194,77 +183,6 @@ int sk_stream_error(struct sock *sk, int flags, int err)
 
 EXPORT_SYMBOL(sk_stream_error);
 
-void __sk_stream_mem_reclaim(struct sock *sk)
-{
-	atomic_sub(sk->sk_forward_alloc >> SK_STREAM_MEM_QUANTUM_SHIFT,
-		   sk->sk_prot->memory_allocated);
-	sk->sk_forward_alloc &= SK_STREAM_MEM_QUANTUM - 1;
-	if (*sk->sk_prot->memory_pressure &&
-	    (atomic_read(sk->sk_prot->memory_allocated) <
-	     sk->sk_prot->sysctl_mem[0]))
-		*sk->sk_prot->memory_pressure = 0;
-}
-
-EXPORT_SYMBOL(__sk_stream_mem_reclaim);
-
-int sk_stream_mem_schedule(struct sock *sk, int size, int kind)
-{
-	int amt = sk_stream_pages(size);
-	struct proto *prot = sk->sk_prot;
-
-	sk->sk_forward_alloc += amt * SK_STREAM_MEM_QUANTUM;
-	atomic_add(amt, prot->memory_allocated);
-
-	/* Under limit. */
-	if (atomic_read(prot->memory_allocated) < prot->sysctl_mem[0]) {
-		if (*prot->memory_pressure)
-			*prot->memory_pressure = 0;
-		return 1;
-	}
-
-	/* Over hard limit. */
-	if (atomic_read(prot->memory_allocated) > prot->sysctl_mem[2]) {
-		prot->enter_memory_pressure();
-		goto suppress_allocation;
-	}
-
-	/* Under pressure. */
-	if (atomic_read(prot->memory_allocated) > prot->sysctl_mem[1])
-		prot->enter_memory_pressure();
-
-	if (kind) {
-		if (atomic_read(&sk->sk_rmem_alloc) < prot->sysctl_rmem[0])
-			return 1;
-	} else if (sk->sk_wmem_queued < prot->sysctl_wmem[0])
-		return 1;
-
-	if (!*prot->memory_pressure ||
-	    prot->sysctl_mem[2] > atomic_read(prot->sockets_allocated) *
-				sk_stream_pages(sk->sk_wmem_queued +
-						atomic_read(&sk->sk_rmem_alloc) +
-						sk->sk_forward_alloc))
-		return 1;
-
-suppress_allocation:
-
-	if (!kind) {
-		sk_stream_moderate_sndbuf(sk);
-
-		/* Fail only if socket is _under_ its sndbuf.
-		 * In this case we cannot block, so that we have to fail.
-		 */
-		if (sk->sk_wmem_queued + size >= sk->sk_sndbuf)
-			return 1;
-	}
-
-	/* Alas. Undo changes. */
-	sk->sk_forward_alloc -= amt * SK_STREAM_MEM_QUANTUM;
-	atomic_sub(amt, prot->memory_allocated);
-	return 0;
-}
-
-EXPORT_SYMBOL(sk_stream_mem_schedule);
-
 void sk_stream_kill_queues(struct sock *sk)
 {
 	/* First the read buffer. */
@@ -277,7 +195,7 @@ void sk_stream_kill_queues(struct sock *sk)
 	BUG_TRAP(skb_queue_empty(&sk->sk_write_queue));
 
 	/* Account for returned memory. */
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 	BUG_TRAP(!sk->sk_wmem_queued);
 	BUG_TRAP(!sk->sk_forward_alloc);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index fdaf965a6794..2cbfa6df7976 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -308,7 +308,7 @@ struct tcp_splice_state {
 /*
  * Pressure flag: try to collapse.
  * Technical note: it is used by multiple contexts non atomically.
- * All the sk_stream_mem_schedule() is of this nature: accounting
+ * All the __sk_mem_schedule() is of this nature: accounting
  * is strict, actions are advisory and have some latency.
  */
 int tcp_memory_pressure __read_mostly;
@@ -485,7 +485,8 @@ static inline void skb_entail(struct sock *sk, struct sk_buff *skb)
 	tcb->sacked  = 0;
 	skb_header_release(skb);
 	tcp_add_write_queue_tail(sk, skb);
-	sk_charge_skb(sk, skb);
+	sk->sk_wmem_queued += skb->truesize;
+	sk_mem_charge(sk, skb->truesize);
 	if (tp->nonagle & TCP_NAGLE_PUSH)
 		tp->nonagle &= ~TCP_NAGLE_PUSH;
 }
@@ -638,7 +639,7 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp)
 
 	skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
 	if (skb) {
-		if (sk_stream_wmem_schedule(sk, skb->truesize)) {
+		if (sk_wmem_schedule(sk, skb->truesize)) {
 			/*
 			 * Make sure that we have exactly size bytes
 			 * available to the caller, no more, no less.
@@ -707,7 +708,7 @@ new_segment:
 			tcp_mark_push(tp, skb);
 			goto new_segment;
 		}
-		if (!sk_stream_wmem_schedule(sk, copy))
+		if (!sk_wmem_schedule(sk, copy))
 			goto wait_for_memory;
 
 		if (can_coalesce) {
@@ -721,7 +722,7 @@ new_segment:
 		skb->data_len += copy;
 		skb->truesize += copy;
 		sk->sk_wmem_queued += copy;
-		sk->sk_forward_alloc -= copy;
+		sk_mem_charge(sk, copy);
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		tp->write_seq += copy;
 		TCP_SKB_CB(skb)->end_seq += copy;
@@ -928,7 +929,7 @@ new_segment:
 				if (copy > PAGE_SIZE - off)
 					copy = PAGE_SIZE - off;
 
-				if (!sk_stream_wmem_schedule(sk, copy))
+				if (!sk_wmem_schedule(sk, copy))
 					goto wait_for_memory;
 
 				if (!page) {
@@ -1019,7 +1020,7 @@ do_fault:
 		 * reset, where we can be unlinking the send_head.
 		 */
 		tcp_check_send_head(sk, skb);
-		sk_stream_free_skb(sk, skb);
+		sk_wmem_free_skb(sk, skb);
 	}
 
 do_error:
@@ -1738,7 +1739,7 @@ void tcp_close(struct sock *sk, long timeout)
 		__kfree_skb(skb);
 	}
 
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 	/* As outlined in RFC 2525, section 2.17, we send a RST here because
 	 * data was lost. To witness the awful effects of the old behavior of
@@ -1841,7 +1842,7 @@ adjudge_to_death:
 		}
 	}
 	if (sk->sk_state != TCP_CLOSE) {
-		sk_stream_mem_reclaim(sk);
+		sk_mem_reclaim(sk);
 		if (tcp_too_many_orphans(sk,
 				atomic_read(sk->sk_prot->orphan_count))) {
 			if (net_ratelimit())
@@ -2658,11 +2659,11 @@ void __init tcp_init(void)
 	limit = ((unsigned long)sysctl_tcp_mem[1]) << (PAGE_SHIFT - 7);
 	max_share = min(4UL*1024*1024, limit);
 
-	sysctl_tcp_wmem[0] = SK_STREAM_MEM_QUANTUM;
+	sysctl_tcp_wmem[0] = SK_MEM_QUANTUM;
 	sysctl_tcp_wmem[1] = 16*1024;
 	sysctl_tcp_wmem[2] = max(64*1024, max_share);
 
-	sysctl_tcp_rmem[0] = SK_STREAM_MEM_QUANTUM;
+	sysctl_tcp_rmem[0] = SK_MEM_QUANTUM;
 	sysctl_tcp_rmem[1] = 87380;
 	sysctl_tcp_rmem[2] = max(87380, max_share);
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index efea9873208e..722c9cbb91e3 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -591,7 +591,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb)
 			 * restart window, so that we send ACKs quickly.
 			 */
 			tcp_incr_quickack(sk);
-			sk_stream_mem_reclaim(sk);
+			sk_mem_reclaim(sk);
 		}
 	}
 	icsk->icsk_ack.lrcvtime = now;
@@ -2851,7 +2851,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets)
 			break;
 
 		tcp_unlink_write_queue(skb, sk);
-		sk_stream_free_skb(sk, skb);
+		sk_wmem_free_skb(sk, skb);
 		tcp_clear_all_retrans_hints(tp);
 	}
 
@@ -3567,7 +3567,7 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th)
 	__skb_queue_purge(&tp->out_of_order_queue);
 	if (tcp_is_sack(tp))
 		tcp_sack_reset(&tp->rx_opt);
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
 		sk->sk_state_change(sk);
@@ -3850,12 +3850,12 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 queue_and_out:
 			if (eaten < 0 &&
 			    (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
-			     !sk_stream_rmem_schedule(sk, skb))) {
+			     !sk_rmem_schedule(sk, skb->truesize))) {
 				if (tcp_prune_queue(sk) < 0 ||
-				    !sk_stream_rmem_schedule(sk, skb))
+				    !sk_rmem_schedule(sk, skb->truesize))
 					goto drop;
 			}
-			sk_stream_set_owner_r(skb, sk);
+			skb_set_owner_r(skb, sk);
 			__skb_queue_tail(&sk->sk_receive_queue, skb);
 		}
 		tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
@@ -3924,9 +3924,9 @@ drop:
 	TCP_ECN_check_ce(tp, skb);
 
 	if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
-	    !sk_stream_rmem_schedule(sk, skb)) {
+	    !sk_rmem_schedule(sk, skb->truesize)) {
 		if (tcp_prune_queue(sk) < 0 ||
-		    !sk_stream_rmem_schedule(sk, skb))
+		    !sk_rmem_schedule(sk, skb->truesize))
 			goto drop;
 	}
 
@@ -3937,7 +3937,7 @@ drop:
 	SOCK_DEBUG(sk, "out of order segment: rcv_next %X seq %X - %X\n",
 		   tp->rcv_nxt, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);
 
-	sk_stream_set_owner_r(skb, sk);
+	skb_set_owner_r(skb, sk);
 
 	if (!skb_peek(&tp->out_of_order_queue)) {
 		/* Initial out of order segment, build 1 SACK. */
@@ -4079,7 +4079,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list,
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
 		TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
 		__skb_insert(nskb, skb->prev, skb, list);
-		sk_stream_set_owner_r(nskb, sk);
+		skb_set_owner_r(nskb, sk);
 
 		/* Copy data, releasing collapsed skbs. */
 		while (copy > 0) {
@@ -4177,7 +4177,7 @@ static int tcp_prune_queue(struct sock *sk)
 		     sk->sk_receive_queue.next,
 		     (struct sk_buff*)&sk->sk_receive_queue,
 		     tp->copied_seq, tp->rcv_nxt);
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 	if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf)
 		return 0;
@@ -4197,7 +4197,7 @@ static int tcp_prune_queue(struct sock *sk)
 		 */
 		if (tcp_is_sack(tp))
 			tcp_sack_reset(&tp->rx_opt);
-		sk_stream_mem_reclaim(sk);
+		sk_mem_reclaim(sk);
 	}
 
 	if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf)
@@ -4699,7 +4699,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 				/* Bulk data transfer: receiver */
 				__skb_pull(skb,tcp_header_len);
 				__skb_queue_tail(&sk->sk_receive_queue, skb);
-				sk_stream_set_owner_r(skb, sk);
+				skb_set_owner_r(skb, sk);
 				tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
 			}
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 9058e0a25107..7a4834a2ae84 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -637,7 +637,8 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
 	tp->write_seq = TCP_SKB_CB(skb)->end_seq;
 	skb_header_release(skb);
 	tcp_add_write_queue_tail(sk, skb);
-	sk_charge_skb(sk, skb);
+	sk->sk_wmem_queued += skb->truesize;
+	sk_mem_charge(sk, skb->truesize);
 }
 
 static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
@@ -701,7 +702,8 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 	if (buff == NULL)
 		return -ENOMEM; /* We'll just try again later. */
 
-	sk_charge_skb(sk, buff);
+	sk->sk_wmem_queued += buff->truesize;
+	sk_mem_charge(sk, buff->truesize);
 	nlen = skb->len - len - nsize;
 	buff->truesize += nlen;
 	skb->truesize -= nlen;
@@ -825,7 +827,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
 
 	skb->truesize	     -= len;
 	sk->sk_wmem_queued   -= len;
-	sk->sk_forward_alloc += len;
+	sk_mem_uncharge(sk, len);
 	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
 
 	/* Any change of skb->len requires recalculation of tso
@@ -1197,7 +1199,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
 	if (unlikely(buff == NULL))
 		return -ENOMEM;
 
-	sk_charge_skb(sk, buff);
+	sk->sk_wmem_queued += buff->truesize;
+	sk_mem_charge(sk, buff->truesize);
 	buff->truesize += nlen;
 	skb->truesize -= nlen;
 
@@ -1350,7 +1353,8 @@ static int tcp_mtu_probe(struct sock *sk)
 	/* We're allowed to probe.  Build it now. */
 	if ((nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC)) == NULL)
 		return -1;
-	sk_charge_skb(sk, nskb);
+	sk->sk_wmem_queued += nskb->truesize;
+	sk_mem_charge(sk, nskb->truesize);
 
 	skb = tcp_send_head(sk);
 
@@ -1377,7 +1381,7 @@ static int tcp_mtu_probe(struct sock *sk)
 			 * Throw it away. */
 			TCP_SKB_CB(nskb)->flags |= TCP_SKB_CB(skb)->flags;
 			tcp_unlink_write_queue(skb, sk);
-			sk_stream_free_skb(sk, skb);
+			sk_wmem_free_skb(sk, skb);
 		} else {
 			TCP_SKB_CB(nskb)->flags |= TCP_SKB_CB(skb)->flags &
 						   ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
@@ -1744,7 +1748,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 		/* changed transmit queue under us so clear hints */
 		tcp_clear_retrans_hints_partial(tp);
 
-		sk_stream_free_skb(sk, next_skb);
+		sk_wmem_free_skb(sk, next_skb);
 	}
 }
 
@@ -2139,8 +2143,9 @@ int tcp_send_synack(struct sock *sk)
 			tcp_unlink_write_queue(skb, sk);
 			skb_header_release(nskb);
 			__tcp_add_write_queue_head(sk, nskb);
-			sk_stream_free_skb(sk, skb);
-			sk_charge_skb(sk, nskb);
+			sk_wmem_free_skb(sk, skb);
+			sk->sk_wmem_queued += nskb->truesize;
+			sk_mem_charge(sk, nskb->truesize);
 			skb = nskb;
 		}
 
@@ -2343,7 +2348,8 @@ int tcp_connect(struct sock *sk)
 	tp->retrans_stamp = TCP_SKB_CB(buff)->when;
 	skb_header_release(buff);
 	__tcp_add_write_queue_tail(sk, buff);
-	sk_charge_skb(sk, buff);
+	sk->sk_wmem_queued += buff->truesize;
+	sk_mem_charge(sk, buff->truesize);
 	tp->packets_out += tcp_skb_pcount(buff);
 	tcp_transmit_skb(sk, buff, 1, GFP_KERNEL);
 
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index ea85bc00c61f..17931be6d584 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -186,7 +186,7 @@ static void tcp_delack_timer(unsigned long data)
 		goto out_unlock;
 	}
 
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 	if (sk->sk_state == TCP_CLOSE || !(icsk->icsk_ack.pending & ICSK_ACK_TIMER))
 		goto out;
@@ -226,7 +226,7 @@ static void tcp_delack_timer(unsigned long data)
 
 out:
 	if (tcp_memory_pressure)
-		sk_stream_mem_reclaim(sk);
+		sk_mem_reclaim(sk);
 out_unlock:
 	bh_unlock_sock(sk);
 	sock_put(sk);
@@ -420,7 +420,7 @@ static void tcp_write_timer(unsigned long data)
 	TCP_CHECK_TIMER(sk);
 
 out:
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 out_unlock:
 	bh_unlock_sock(sk);
 	sock_put(sk);
@@ -514,7 +514,7 @@ static void tcp_keepalive_timer (unsigned long data)
 	}
 
 	TCP_CHECK_TIMER(sk);
-	sk_stream_mem_reclaim(sk);
+	sk_mem_reclaim(sk);
 
 resched:
 	inet_csk_reset_keepalive_timer (sk, elapsed);
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index e466e00b9a9f..b92196495027 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -1109,7 +1109,7 @@ SCTP_STATIC __init int sctp_init(void)
 	sysctl_sctp_rmem[1] = (1500 *(sizeof(struct sk_buff) + 1));
 	sysctl_sctp_rmem[2] = max(sysctl_sctp_rmem[1], max_share);
 
-	sysctl_sctp_wmem[0] = SK_STREAM_MEM_QUANTUM;
+	sysctl_sctp_wmem[0] = SK_MEM_QUANTUM;
 	sysctl_sctp_wmem[1] = 16*1024;
 	sysctl_sctp_wmem[2] = max(64*1024, max_share);
 
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 511d8c9a171a..b1267519183b 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -5844,7 +5844,7 @@ static int sctp_eat_data(const struct sctp_association *asoc,
 	/*
 	 * Also try to renege to limit our memory usage in the event that
 	 * we are under memory pressure
-	 * If we can't renege, don't worry about it, the sk_stream_rmem_schedule
+	 * If we can't renege, don't worry about it, the sk_rmem_schedule
 	 * in sctp_ulpevent_make_rcvmsg will drop the frame if we grow our
 	 * memory usage too much
 	 */
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 7a8650f01d08..710df67a6785 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -174,7 +174,8 @@ static inline void sctp_set_owner_w(struct sctp_chunk *chunk)
 				sizeof(struct sctp_chunk);
 
 	atomic_add(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc);
-	sk_charge_skb(sk, chunk->skb);
+	sk->sk_wmem_queued += chunk->skb->truesize;
+	sk_mem_charge(sk, chunk->skb->truesize);
 }
 
 /* Verify that this is a valid address. */
@@ -6035,10 +6036,10 @@ static void sctp_wfree(struct sk_buff *skb)
 	atomic_sub(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc);
 
 	/*
-	 * This undoes what is done via sk_charge_skb
+	 * This undoes what is done via sctp_set_owner_w and sk_mem_charge
 	 */
 	sk->sk_wmem_queued   -= skb->truesize;
-	sk->sk_forward_alloc += skb->truesize;
+	sk_mem_uncharge(sk, skb->truesize);
 
 	sock_wfree(skb);
 	__sctp_write_space(asoc);
@@ -6059,9 +6060,9 @@ void sctp_sock_rfree(struct sk_buff *skb)
 	atomic_sub(event->rmem_len, &sk->sk_rmem_alloc);
 
 	/*
-	 * Mimic the behavior of sk_stream_rfree
+	 * Mimic the behavior of sock_rfree
 	 */
-	sk->sk_forward_alloc += event->rmem_len;
+	sk_mem_uncharge(sk, event->rmem_len);
 }
 
 
diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c
index 307314356e16..047c27df98f4 100644
--- a/net/sctp/ulpevent.c
+++ b/net/sctp/ulpevent.c
@@ -700,7 +700,7 @@ struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc,
 	if (rx_count >= asoc->base.sk->sk_rcvbuf) {
 
 		if ((asoc->base.sk->sk_userlocks & SOCK_RCVBUF_LOCK) ||
-		   (!sk_stream_rmem_schedule(asoc->base.sk, chunk->skb)))
+		    (!sk_rmem_schedule(asoc->base.sk, chunk->skb->truesize)))
 			goto fail;
 	}
 
diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c
index 1733fa29a501..c25caefa3bcb 100644
--- a/net/sctp/ulpqueue.c
+++ b/net/sctp/ulpqueue.c
@@ -1046,7 +1046,7 @@ void sctp_ulpq_renege(struct sctp_ulpq *ulpq, struct sctp_chunk *chunk,
 		sctp_ulpq_partial_delivery(ulpq, chunk, gfp);
 	}
 
-	sk_stream_mem_reclaim(asoc->base.sk);
+	sk_mem_reclaim(asoc->base.sk);
 	return;
 }
 
-- 
cgit v1.2.3


From 95766fff6b9a78d11fc2d3812dd035381690b55d Mon Sep 17 00:00:00 2001
From: Hideo Aoki <haoki@redhat.com>
Date: Mon, 31 Dec 2007 00:29:24 -0800
Subject: [UDP]: Add memory accounting.

Signed-off-by: Takahiro Yasui <tyasui@redhat.com>
Signed-off-by: Hideo Aoki <haoki@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ip-sysctl.txt | 27 ++++++++++++++++
 include/net/udp.h                      |  9 ++++++
 net/ipv4/af_inet.c                     |  5 +++
 net/ipv4/proc.c                        |  3 +-
 net/ipv4/sysctl_net_ipv4.c             | 31 ++++++++++++++++++
 net/ipv4/udp.c                         | 57 ++++++++++++++++++++++++++++++++--
 net/ipv6/udp.c                         | 32 ++++++++++++++++---
 7 files changed, 157 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 6f7872ba1def..17a6e46fbd43 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -446,6 +446,33 @@ tcp_dma_copybreak - INTEGER
 	and CONFIG_NET_DMA is enabled.
 	Default: 4096
 
+UDP variables:
+
+udp_mem - vector of 3 INTEGERs: min, pressure, max
+	Number of pages allowed for queueing by all UDP sockets.
+
+	min: Below this number of pages UDP is not bothered about its
+	memory appetite. When amount of memory allocated by UDP exceeds
+	this number, UDP starts to moderate memory usage.
+
+	pressure: This value was introduced to follow format of tcp_mem.
+
+	max: Number of pages allowed for queueing by all UDP sockets.
+
+	Default is calculated at boot time from amount of available memory.
+
+udp_rmem_min - INTEGER
+	Minimal size of receive buffer used by UDP sockets in moderation.
+	Each UDP socket is able to use the size for receiving data, even if
+	total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
+	Default: 4096
+
+udp_wmem_min - INTEGER
+	Minimal size of send buffer used by UDP sockets in moderation.
+	Each UDP socket is able to use the size for sending data, even if
+	total pages of UDP sockets exceed udp_mem pressure. The unit is byte.
+	Default: 4096
+
 CIPSOv4 Variables:
 
 cipso_cache_enable - BOOLEAN
diff --git a/include/net/udp.h b/include/net/udp.h
index 98cb09ca3a27..93796beac8ff 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -65,6 +65,13 @@ extern rwlock_t udp_hash_lock;
 
 extern struct proto udp_prot;
 
+extern atomic_t udp_memory_allocated;
+
+/* sysctl variables for udp */
+extern int sysctl_udp_mem[3];
+extern int sysctl_udp_rmem_min;
+extern int sysctl_udp_wmem_min;
+
 struct sk_buff;
 
 /*
@@ -198,4 +205,6 @@ extern void udp_proc_unregister(struct udp_seq_afinfo *afinfo);
 extern int  udp4_proc_init(void);
 extern void udp4_proc_exit(void);
 #endif
+
+extern void udp_init(void);
 #endif	/* _UDP_H */
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 03633b7b9b4a..0e12cf646071 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -139,6 +139,8 @@ void inet_sock_destruct(struct sock *sk)
 	__skb_queue_purge(&sk->sk_receive_queue);
 	__skb_queue_purge(&sk->sk_error_queue);
 
+	sk_mem_reclaim(sk);
+
 	if (sk->sk_type == SOCK_STREAM && sk->sk_state != TCP_CLOSE) {
 		printk("Attempt to release TCP socket in state %d %p\n",
 		       sk->sk_state, sk);
@@ -1417,6 +1419,9 @@ static int __init inet_init(void)
 	/* Setup TCP slab cache for open requests. */
 	tcp_init();
 
+	/* Setup UDP memory threshold */
+	udp_init();
+
 	/* Add UDP-Lite (RFC 3828) */
 	udplite4_register();
 
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 41734db677be..53bc010beefd 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -56,7 +56,8 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 		   sock_prot_inuse(&tcp_prot), atomic_read(&tcp_orphan_count),
 		   tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated),
 		   atomic_read(&tcp_memory_allocated));
-	seq_printf(seq, "UDP: inuse %d\n", sock_prot_inuse(&udp_prot));
+	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse(&udp_prot),
+		   atomic_read(&udp_memory_allocated));
 	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse(&udplite_prot));
 	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 844f26fab06f..a5a9f8e3bb25 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -19,6 +19,7 @@
 #include <net/ip.h>
 #include <net/route.h>
 #include <net/tcp.h>
+#include <net/udp.h>
 #include <net/cipso_ipv4.h>
 #include <net/inet_frag.h>
 
@@ -812,6 +813,36 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "udp_mem",
+		.data		= &sysctl_udp_mem,
+		.maxlen		= sizeof(sysctl_udp_mem),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "udp_rmem_min",
+		.data		= &sysctl_udp_rmem_min,
+		.maxlen		= sizeof(sysctl_udp_rmem_min),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "udp_wmem_min",
+		.data		= &sysctl_udp_wmem_min,
+		.maxlen		= sizeof(sysctl_udp_wmem_min),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero
+	},
 	{ .ctl_name = 0 }
 };
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 1ce6b60b7f93..353284360751 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -82,6 +82,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/ioctls.h>
+#include <linux/bootmem.h>
 #include <linux/types.h>
 #include <linux/fcntl.h>
 #include <linux/module.h>
@@ -118,6 +119,17 @@ EXPORT_SYMBOL(udp_stats_in6);
 struct hlist_head udp_hash[UDP_HTABLE_SIZE];
 DEFINE_RWLOCK(udp_hash_lock);
 
+int sysctl_udp_mem[3] __read_mostly;
+int sysctl_udp_rmem_min __read_mostly;
+int sysctl_udp_wmem_min __read_mostly;
+
+EXPORT_SYMBOL(sysctl_udp_mem);
+EXPORT_SYMBOL(sysctl_udp_rmem_min);
+EXPORT_SYMBOL(sysctl_udp_wmem_min);
+
+atomic_t udp_memory_allocated;
+EXPORT_SYMBOL(udp_memory_allocated);
+
 static inline int __udp_lib_lport_inuse(__u16 num,
 					const struct hlist_head udptable[])
 {
@@ -901,13 +913,17 @@ try_again:
 		err = ulen;
 
 out_free:
+	lock_sock(sk);
 	skb_free_datagram(sk, skb);
+	release_sock(sk);
 out:
 	return err;
 
 csum_copy_err:
+	lock_sock(sk);
 	if (!skb_kill_datagram(sk, skb, flags))
 		UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+	release_sock(sk);
 
 	if (noblock)
 		return -EAGAIN;
@@ -1072,7 +1088,15 @@ static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
 				skb1 = skb_clone(skb, GFP_ATOMIC);
 
 			if (skb1) {
-				int ret = udp_queue_rcv_skb(sk, skb1);
+				int ret = 0;
+
+				bh_lock_sock_nested(sk);
+				if (!sock_owned_by_user(sk))
+					ret = udp_queue_rcv_skb(sk, skb1);
+				else
+					sk_add_backlog(sk, skb1);
+				bh_unlock_sock(sk);
+
 				if (ret > 0)
 					/* we should probably re-process instead
 					 * of dropping packets here. */
@@ -1165,7 +1189,13 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 			       inet_iif(skb), udptable);
 
 	if (sk != NULL) {
-		int ret = udp_queue_rcv_skb(sk, skb);
+		int ret = 0;
+		bh_lock_sock_nested(sk);
+		if (!sock_owned_by_user(sk))
+			ret = udp_queue_rcv_skb(sk, skb);
+		else
+			sk_add_backlog(sk, skb);
+		bh_unlock_sock(sk);
 		sock_put(sk);
 
 		/* a return value > 0 means to resubmit the input, but
@@ -1460,6 +1490,10 @@ struct proto udp_prot = {
 	.hash		   = udp_lib_hash,
 	.unhash		   = udp_lib_unhash,
 	.get_port	   = udp_v4_get_port,
+	.memory_allocated  = &udp_memory_allocated,
+	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem	   = &sysctl_udp_wmem_min,
+	.sysctl_rmem	   = &sysctl_udp_rmem_min,
 	.obj_size	   = sizeof(struct udp_sock),
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udp_setsockopt,
@@ -1655,6 +1689,25 @@ void udp4_proc_exit(void)
 }
 #endif /* CONFIG_PROC_FS */
 
+void __init udp_init(void)
+{
+	unsigned long limit;
+
+	/* Set the pressure threshold up by the same strategy of TCP. It is a
+	 * fraction of global memory that is up to 1/2 at 256 MB, decreasing
+	 * toward zero with the amount of memory, with a floor of 128 pages.
+	 */
+	limit = min(nr_all_pages, 1UL<<(28-PAGE_SHIFT)) >> (20-PAGE_SHIFT);
+	limit = (limit * (nr_all_pages >> (20-PAGE_SHIFT))) >> (PAGE_SHIFT-11);
+	limit = max(limit, 128UL);
+	sysctl_udp_mem[0] = limit / 4 * 3;
+	sysctl_udp_mem[1] = limit;
+	sysctl_udp_mem[2] = sysctl_udp_mem[0] * 2;
+
+	sysctl_udp_rmem_min = SK_MEM_QUANTUM;
+	sysctl_udp_wmem_min = SK_MEM_QUANTUM;
+}
+
 EXPORT_SYMBOL(udp_disconnect);
 EXPORT_SYMBOL(udp_hash);
 EXPORT_SYMBOL(udp_hash_lock);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index c9a97b405511..bf58acab2064 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -204,13 +204,17 @@ try_again:
 		err = ulen;
 
 out_free:
+	lock_sock(sk);
 	skb_free_datagram(sk, skb);
+	release_sock(sk);
 out:
 	return err;
 
 csum_copy_err:
+	lock_sock(sk);
 	if (!skb_kill_datagram(sk, skb, flags))
 		UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+	release_sock(sk);
 
 	if (flags & MSG_DONTWAIT)
 		return -EAGAIN;
@@ -366,10 +370,21 @@ static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr,
 	while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr,
 					uh->source, saddr, dif))) {
 		struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC);
-		if (buff)
-			udpv6_queue_rcv_skb(sk2, buff);
+		if (buff) {
+			bh_lock_sock_nested(sk2);
+			if (!sock_owned_by_user(sk2))
+				udpv6_queue_rcv_skb(sk2, buff);
+			else
+				sk_add_backlog(sk2, buff);
+			bh_unlock_sock(sk2);
+		}
 	}
-	udpv6_queue_rcv_skb(sk, skb);
+	bh_lock_sock_nested(sk);
+	if (!sock_owned_by_user(sk))
+		udpv6_queue_rcv_skb(sk, skb);
+	else
+		sk_add_backlog(sk, skb);
+	bh_unlock_sock(sk);
 out:
 	read_unlock(&udp_hash_lock);
 	return 0;
@@ -482,7 +497,12 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 
 	/* deliver */
 
-	udpv6_queue_rcv_skb(sk, skb);
+	bh_lock_sock_nested(sk);
+	if (!sock_owned_by_user(sk))
+		udpv6_queue_rcv_skb(sk, skb);
+	else
+		sk_add_backlog(sk, skb);
+	bh_unlock_sock(sk);
 	sock_put(sk);
 	return 0;
 
@@ -994,6 +1014,10 @@ struct proto udpv6_prot = {
 	.hash		   = udp_lib_hash,
 	.unhash		   = udp_lib_unhash,
 	.get_port	   = udp_v6_get_port,
+	.memory_allocated  = &udp_memory_allocated,
+	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem	   = &sysctl_udp_wmem_min,
+	.sysctl_rmem	   = &sysctl_udp_rmem_min,
 	.obj_size	   = sizeof(struct udp6_sock),
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udpv6_setsockopt,
-- 
cgit v1.2.3


From 61f1ab41b8ede8e2a26c349a4e3372100545c5ec Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Mon, 31 Dec 2007 04:22:09 -0800
Subject: [IPV4]: Remove unused multipath cached routing defintion in
 net/flow.h

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/flow.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/flow.h b/include/net/flow.h
index af59fa5cc1f8..ad16e0076c89 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -48,7 +48,6 @@ struct flowi {
 
 	__u8	proto;
 	__u8	flags;
-#define FLOWI_FLAG_MULTIPATHOLDROUTE 0x01
 	union {
 		struct {
 			__be16	sport;
-- 
cgit v1.2.3


From 90840defabbd819180c7528e12d550776848f833 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Mon, 31 Dec 2007 04:48:41 -0800
Subject: [TCP]: Introduce tcp_wnd_end() to reduce line lengths
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     |  6 ++++++
 net/ipv4/tcp_input.c  |  3 +--
 net/ipv4/tcp_output.c | 20 ++++++++++----------
 3 files changed, 17 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 76286e80205a..6a732d4919fe 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -785,6 +785,12 @@ static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp)
 	return 3;
 }
 
+/* Returns end sequence number of the receiver's advertised window */
+static inline u32 tcp_wnd_end(const struct tcp_sock *tp)
+{
+	return tp->snd_una + tp->snd_wnd;
+}
+
 /* RFC2861 Check whether we are limited by application or congestion window
  * This is the inverse of cwnd check in tcp_tso_should_defer
  */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 41f4b862319d..366f63ae4d5c 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2921,8 +2921,7 @@ static void tcp_ack_probe(struct sock *sk)
 
 	/* Was it a usable window open? */
 
-	if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq,
-		   tp->snd_una + tp->snd_wnd)) {
+	if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq, tcp_wnd_end(tp))) {
 		icsk->icsk_backoff = 0;
 		inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0);
 		/* Socket must be waked up by subsequent tcp_data_snd_check().
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 1ca638b83168..821fae271428 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -89,10 +89,10 @@ static inline __u32 tcp_acceptable_seq(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
-	if (!before(tp->snd_una+tp->snd_wnd, tp->snd_nxt))
+	if (!before(tcp_wnd_end(tp), tp->snd_nxt))
 		return tp->snd_nxt;
 	else
-		return tp->snd_una+tp->snd_wnd;
+		return tcp_wnd_end(tp);
 }
 
 /* Calculate mss to advertise in SYN segment.
@@ -1024,7 +1024,7 @@ static unsigned int tcp_mss_split_point(struct sock *sk, struct sk_buff *skb,
 	struct tcp_sock *tp = tcp_sk(sk);
 	u32 needed, window, cwnd_len;
 
-	window = (tp->snd_una + tp->snd_wnd - TCP_SKB_CB(skb)->seq);
+	window = tcp_wnd_end(tp) - TCP_SKB_CB(skb)->seq;
 	cwnd_len = mss_now * cwnd;
 
 	if (likely(cwnd_len <= window && skb != tcp_write_queue_tail(sk)))
@@ -1134,7 +1134,7 @@ static inline int tcp_snd_wnd_test(struct tcp_sock *tp, struct sk_buff *skb, uns
 	if (skb->len > cur_mss)
 		end_seq = TCP_SKB_CB(skb)->seq + cur_mss;
 
-	return !after(end_seq, tp->snd_una + tp->snd_wnd);
+	return !after(end_seq, tcp_wnd_end(tp));
 }
 
 /* This checks if the data bearing packet SKB (usually tcp_send_head(sk))
@@ -1251,7 +1251,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb)
 	BUG_ON(tcp_skb_pcount(skb) <= 1 ||
 	       (tp->snd_cwnd <= in_flight));
 
-	send_win = (tp->snd_una + tp->snd_wnd) - TCP_SKB_CB(skb)->seq;
+	send_win = tcp_wnd_end(tp) - TCP_SKB_CB(skb)->seq;
 
 	/* From in_flight test above, we know that cwnd > in_flight.  */
 	cong_win = (tp->snd_cwnd - in_flight) * tp->mss_cache;
@@ -1332,7 +1332,7 @@ static int tcp_mtu_probe(struct sock *sk)
 
 	if (tp->snd_wnd < size_needed)
 		return -1;
-	if (after(tp->snd_nxt + size_needed, tp->snd_una + tp->snd_wnd))
+	if (after(tp->snd_nxt + size_needed, tcp_wnd_end(tp)))
 		return 0;
 
 	/* Do we need to wait to drain cwnd? With none in flight, don't stall */
@@ -1687,7 +1687,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 			return;
 
 		/* Next skb is out of window. */
-		if (after(TCP_SKB_CB(next_skb)->end_seq, tp->snd_una+tp->snd_wnd))
+		if (after(TCP_SKB_CB(next_skb)->end_seq, tcp_wnd_end(tp)))
 			return;
 
 		/* Punt if not enough space exists in the first SKB for
@@ -1831,7 +1831,7 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 	 * case, when window is shrunk to zero. In this case
 	 * our retransmit serves as a zero window probe.
 	 */
-	if (!before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd)
+	if (!before(TCP_SKB_CB(skb)->seq, tcp_wnd_end(tp))
 	    && TCP_SKB_CB(skb)->seq != tp->snd_una)
 		return -EAGAIN;
 
@@ -2497,10 +2497,10 @@ int tcp_write_wakeup(struct sock *sk)
 		struct sk_buff *skb;
 
 		if ((skb = tcp_send_head(sk)) != NULL &&
-		    before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd)) {
+		    before(TCP_SKB_CB(skb)->seq, tcp_wnd_end(tp))) {
 			int err;
 			unsigned int mss = tcp_current_mss(sk, 0);
-			unsigned int seg_size = tp->snd_una+tp->snd_wnd-TCP_SKB_CB(skb)->seq;
+			unsigned int seg_size = tcp_wnd_end(tp) - TCP_SKB_CB(skb)->seq;
 
 			if (before(tp->pushed_seq, TCP_SKB_CB(skb)->end_seq))
 				tp->pushed_seq = TCP_SKB_CB(skb)->end_seq;
-- 
cgit v1.2.3


From 4828e7f49a402930e8b3e72de695c8d37e0f98ee Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Mon, 31 Dec 2007 04:50:19 -0800
Subject: [TCP]: Remove TCPCB_URG & TCPCB_AT_TAIL as unnecessary
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The snd_up check should be enough. I suspect this has been
there to provide a minor optimization in clean_rtx_queue which
used to have a small if (!->sacked) block which could skip
snd_up check among the other work.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     | 4 ----
 net/ipv4/tcp.c        | 1 -
 net/ipv4/tcp_input.c  | 3 +--
 net/ipv4/tcp_output.c | 7 +++----
 4 files changed, 4 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 6a732d4919fe..48081ada92a7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -578,10 +578,6 @@ struct tcp_skb_cb {
 #define TCPCB_EVER_RETRANS	0x80	/* Ever retransmitted frame	*/
 #define TCPCB_RETRANS		(TCPCB_SACKED_RETRANS|TCPCB_EVER_RETRANS)
 
-#define TCPCB_URG		0x20	/* Urgent pointer advanced here	*/
-
-#define TCPCB_AT_TAIL		(TCPCB_URG)
-
 	__u16		urg_ptr;	/* Valid w/URG flags is set.	*/
 	__u32		ack_seq;	/* Sequence number ACK'd	*/
 };
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 2cbfa6df7976..34085e3a4096 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -497,7 +497,6 @@ static inline void tcp_mark_urg(struct tcp_sock *tp, int flags,
 	if (flags & MSG_OOB) {
 		tp->urg_mode = 1;
 		tp->snd_up = tp->write_seq;
-		TCP_SKB_CB(skb)->sacked |= TCPCB_URG;
 	}
 }
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 7bac1fac065f..1e7fd8113663 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2821,8 +2821,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets)
 		if (sacked & TCPCB_LOST)
 			tp->lost_out -= acked_pcount;
 
-		if (unlikely((sacked & TCPCB_URG) && tp->urg_mode &&
-			     !before(end_seq, tp->snd_up)))
+		if (unlikely(tp->urg_mode && !before(end_seq, tp->snd_up)))
 			tp->urg_mode = 0;
 
 		tp->packets_out -= acked_pcount;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 821fae271428..cd21528665f3 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -711,7 +711,6 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 	TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
 	TCP_SKB_CB(buff)->flags = flags;
 	TCP_SKB_CB(buff)->sacked = TCP_SKB_CB(skb)->sacked;
-	TCP_SKB_CB(skb)->sacked &= ~TCPCB_AT_TAIL;
 
 	if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_PARTIAL) {
 		/* Copy and checksum data tail into the new buffer. */
@@ -1726,7 +1725,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 		/* All done, get rid of second SKB and account for it so
 		 * packet counting does not break.
 		 */
-		TCP_SKB_CB(skb)->sacked |= TCP_SKB_CB(next_skb)->sacked&(TCPCB_EVER_RETRANS|TCPCB_AT_TAIL);
+		TCP_SKB_CB(skb)->sacked |= TCP_SKB_CB(next_skb)->sacked & TCPCB_EVER_RETRANS;
 		if (TCP_SKB_CB(next_skb)->sacked&TCPCB_SACKED_RETRANS)
 			tp->retrans_out -= tcp_skb_pcount(next_skb);
 		if (TCP_SKB_CB(next_skb)->sacked&TCPCB_LOST)
@@ -2475,7 +2474,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent)
 	skb_reserve(skb, MAX_TCP_HEADER);
 	skb->csum = 0;
 	TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK;
-	TCP_SKB_CB(skb)->sacked = urgent;
+	TCP_SKB_CB(skb)->sacked = 0;
 	skb_shinfo(skb)->gso_segs = 1;
 	skb_shinfo(skb)->gso_size = 0;
 	skb_shinfo(skb)->gso_type = 0;
@@ -2527,7 +2526,7 @@ int tcp_write_wakeup(struct sock *sk)
 		} else {
 			if (tp->urg_mode &&
 			    between(tp->snd_up, tp->snd_una+1, tp->snd_una+0xFFFF))
-				tcp_xmit_probe_skb(sk, TCPCB_URG);
+				tcp_xmit_probe_skb(sk, 1);
 			return tcp_xmit_probe_skb(sk, 0);
 		}
 	}
-- 
cgit v1.2.3


From 9ef32d0d1f64cad414697f34bda1b269f632f0cd Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 1 Jan 2008 19:07:50 -0800
Subject: [IPSEC]: Kill duplicate xfrm_policy_flush prototype

For five years we had two xfrm_policy_flush prototypes and every time that
function's signature changed people have been diligently updating both of
them without noticing :)

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 5d5580ac0101..a419a4372e21 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1310,7 +1310,6 @@ extern int xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi);
 struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto,
 				  xfrm_address_t *daddr, xfrm_address_t *saddr,
 				  int create, unsigned short family);
-extern int xfrm_policy_flush(u8 type, struct xfrm_audit *audit_info);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
 extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst,
 			  struct flowi *fl, int family, int strict);
-- 
cgit v1.2.3


From f1862b0ae2294f6970f695abf02392d025e02dbe Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sun, 27 Jan 2008 23:04:43 -0800
Subject: [SHAPER]: The scheduled shaper removal.

This patch contains the scheduled removal of the shaper driver.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/feature-removal-schedule.txt |   9 -
 Documentation/networking/00-INDEX          |   2 -
 Documentation/networking/shaper.txt        |  48 ---
 drivers/net/Kconfig                        |  17 -
 drivers/net/Makefile                       |   1 -
 drivers/net/shaper.c                       | 603 -----------------------------
 include/linux/Kbuild                       |   1 -
 include/linux/if_shaper.h                  |  51 ---
 8 files changed, 732 deletions(-)
 delete mode 100644 Documentation/networking/shaper.txt
 delete mode 100644 drivers/net/shaper.c
 delete mode 100644 include/linux/if_shaper.h

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 2c3ae6c02a47..cc02e67239ec 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -280,15 +280,6 @@ Who:	Thomas Gleixner <tglx@linutronix.de>
 
 ---------------------------
 
-What:	shaper network driver
-When:	January 2008
-Files:	drivers/net/shaper.c, include/linux/if_shaper.h
-Why:	This driver has been marked obsolete for many years.
-	It was only designed to work on lower speed links and has design
-	flaws that lead to machine crashes. The qdisc infrastructure in
-	2.4 or later kernels, provides richer features and is more robust.
-Who:	Stephen Hemminger <shemminger@linux-foundation.org>
-
 ---------------------------
 
 What:	i2c-i810, i2c-prosavage and i2c-savage4
diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX
index b4eefadb9adf..02e56d447a8f 100644
--- a/Documentation/networking/00-INDEX
+++ b/Documentation/networking/00-INDEX
@@ -84,8 +84,6 @@ policy-routing.txt
 	- IP policy-based routing
 ray_cs.txt
 	- Raylink Wireless LAN card driver info.
-shaper.txt
-	- info on the module that can shape/limit transmitted traffic.
 sk98lin.txt
 	- Marvell Yukon Chipset / SysKonnect SK-98xx compliant Gigabit
 	  Ethernet Adapter family driver info
diff --git a/Documentation/networking/shaper.txt b/Documentation/networking/shaper.txt
deleted file mode 100644
index 6c4ebb66a906..000000000000
--- a/Documentation/networking/shaper.txt
+++ /dev/null
@@ -1,48 +0,0 @@
-Traffic Shaper For Linux
-
-This is the current BETA release of the traffic shaper for Linux. It works
-within the following limits:
-
-o	Minimum shaping speed is currently about 9600 baud (it can only
-shape down to 1 byte per clock tick)
-
-o	Maximum is about 256K, it will go above this but get a bit blocky.
-
-o	If you ifconfig the master device that a shaper is attached to down
-then your machine will follow.
-
-o	The shaper must be a module.
-
-
-Setup:
-
-	A shaper device is configured using the shapeconfig program.
-Typically you will do something like this
-
-shapecfg attach shaper0 eth1
-shapecfg speed shaper0 64000
-ifconfig shaper0 myhost netmask 255.255.255.240 broadcast 1.2.3.4.255 up
-route add -net some.network netmask a.b.c.d dev shaper0
-
-The shaper should have the same IP address as the device it is attached to
-for normal use.
-
-Gotchas:
-
-	The shaper shapes transmitted traffic. It's rather impossible to
-shape received traffic except at the end (or a router) transmitting it.
-
-	Gated/routed/rwhod/mrouted all see the shaper as an additional device
-and will treat it as such unless patched. Note that for mrouted you can run
-mrouted tunnels via a traffic shaper to control bandwidth usage.
-
-	The shaper is device/route based. This makes it very easy to use
-with any setup BUT less flexible. You may need to use iproute2 to set up
-multiple route tables to get the flexibility.
-
-	There is no "borrowing" or "sharing" scheme. This is a simple
-traffic limiter. We implement Van Jacobson and Sally Floyd's CBQ
-architecture into Linux 2.2. This is the preferred solution. Shaper is
-for simple or back compatible setups.
-
-Alan
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index a6728661c416..777bb81d9ada 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -3015,23 +3015,6 @@ config NET_FC
 	  adaptor below. You also should have said Y to "SCSI support" and
 	  "SCSI generic support".
 
-config SHAPER
-	tristate "Traffic Shaper (OBSOLETE)"
-	depends on EXPERIMENTAL
-	---help---
-	  The traffic shaper is a virtual network device that allows you to
-	  limit the rate of outgoing data flow over some other network device.
-	  The traffic that you want to slow down can then be routed through
-	  these virtual devices. See
-	  <file:Documentation/networking/shaper.txt> for more information.
-
-	  An alternative to this traffic shaper are traffic schedulers which
-	  you'll get if you say Y to "QoS and/or fair queuing" in
-	  "Networking options".
-
-	  To compile this driver as a module, choose M here: the module
-	  will be called shaper.  If unsure, say N.
-
 config NETCONSOLE
 	tristate "Network console logging support (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 17a2ffd73d7d..a2f2662c244b 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -93,7 +93,6 @@ obj-$(CONFIG_NET_SB1000) += sb1000.o
 obj-$(CONFIG_MAC8390) += mac8390.o
 obj-$(CONFIG_APNE) += apne.o 8390.o
 obj-$(CONFIG_PCMCIA_PCNET) += 8390.o
-obj-$(CONFIG_SHAPER) += shaper.o
 obj-$(CONFIG_HP100) += hp100.o
 obj-$(CONFIG_SMC9194) += smc9194.o
 obj-$(CONFIG_FEC) += fec.o
diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c
deleted file mode 100644
index 228f650250f6..000000000000
--- a/drivers/net/shaper.c
+++ /dev/null
@@ -1,603 +0,0 @@
-/*
- *			Simple traffic shaper for Linux NET3.
- *
- *	(c) Copyright 1996 Alan Cox <alan@redhat.com>, All Rights Reserved.
- *				http://www.redhat.com
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- *	Neither Alan Cox nor CymruNet Ltd. admit liability nor provide
- *	warranty for any of this software. This material is provided
- *	"AS-IS" and at no charge.
- *
- *
- *	Algorithm:
- *
- *	Queue Frame:
- *		Compute time length of frame at regulated speed
- *		Add frame to queue at appropriate point
- *		Adjust time length computation for followup frames
- *		Any frame that falls outside of its boundaries is freed
- *
- *	We work to the following constants
- *
- *		SHAPER_QLEN	Maximum queued frames
- *		SHAPER_LATENCY	Bounding latency on a frame. Leaving this latency
- *				window drops the frame. This stops us queueing
- *				frames for a long time and confusing a remote
- *				host.
- *		SHAPER_MAXSLIP	Maximum time a priority frame may jump forward.
- *				That bounds the penalty we will inflict on low
- *				priority traffic.
- *		SHAPER_BURST	Time range we call "now" in order to reduce
- *				system load. The more we make this the burstier
- *				the behaviour, the better local performance you
- *				get through packet clustering on routers and the
- *				worse the remote end gets to judge rtts.
- *
- *	This is designed to handle lower speed links ( < 200K/second or so). We
- *	run off a 100-150Hz base clock typically. This gives us a resolution at
- *	200Kbit/second of about 2Kbit or 256 bytes. Above that our timer
- *	resolution may start to cause much more burstiness in the traffic. We
- *	could avoid a lot of that by calling kick_shaper() at the end of the
- *	tied device transmissions. If you run above about 100K second you
- *	may need to tune the supposed speed rate for the right values.
- *
- *	BUGS:
- *		Downing the interface under the shaper before the shaper
- *		will render your machine defunct. Don't for now shape over
- *		PPP or SLIP therefore!
- *		This will be fixed in BETA4
- *
- * Update History :
- *
- *              bh_atomic() SMP races fixes and rewritten the locking code to
- *              be SMP safe and irq-mask friendly.
- *              NOTE: we can't use start_bh_atomic() in kick_shaper()
- *              because it's going to be recalled from an irq handler,
- *              and synchronize_bh() is a nono if called from irq context.
- *						1999  Andrea Arcangeli
- *
- *              Device statistics (tx_pakets, tx_bytes,
- *              tx_drops: queue_over_time and collisions: max_queue_exceded)
- *                               1999/06/18 Jordi Murgo <savage@apostols.org>
- *
- *		Use skb->cb for private data.
- *				 2000/03 Andi Kleen
- */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/fcntl.h>
-#include <linux/mm.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/if_arp.h>
-#include <linux/init.h>
-#include <linux/if_shaper.h>
-#include <linux/jiffies.h>
-
-#include <net/dst.h>
-#include <net/arp.h>
-#include <net/net_namespace.h>
-
-struct shaper_cb {
-	unsigned long	shapeclock;		/* Time it should go out */
-	unsigned long	shapestamp;		/* Stamp for shaper    */
-	__u32		shapelatency;		/* Latency on frame */
-	__u32		shapelen;		/* Frame length in clocks */
-	__u16		shapepend;		/* Pending */
-};
-#define SHAPERCB(skb) ((struct shaper_cb *) ((skb)->cb))
-
-static int sh_debug;		/* Debug flag */
-
-#define SHAPER_BANNER	"CymruNet Traffic Shaper BETA 0.04 for Linux 2.1\n"
-
-static void shaper_kick(struct shaper *sh);
-
-/*
- *	Compute clocks on a buffer
- */
-
-static int shaper_clocks(struct shaper *shaper, struct sk_buff *skb)
-{
- 	int t=skb->len/shaper->bytespertick;
- 	return t;
-}
-
-/*
- *	Set the speed of a shaper. We compute this in bytes per tick since
- *	thats how the machine wants to run. Quoted input is in bits per second
- *	as is traditional (note not BAUD). We assume 8 bit bytes.
- */
-
-static void shaper_setspeed(struct shaper *shaper, int bitspersec)
-{
-	shaper->bitspersec=bitspersec;
-	shaper->bytespertick=(bitspersec/HZ)/8;
-	if(!shaper->bytespertick)
-		shaper->bytespertick++;
-}
-
-/*
- *	Throw a frame at a shaper.
- */
-
-
-static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct shaper *shaper = dev->priv;
- 	struct sk_buff *ptr;
-
-	spin_lock(&shaper->lock);
- 	ptr=shaper->sendq.prev;
-
- 	/*
- 	 *	Set up our packet details
- 	 */
-
- 	SHAPERCB(skb)->shapelatency=0;
- 	SHAPERCB(skb)->shapeclock=shaper->recovery;
- 	if(time_before(SHAPERCB(skb)->shapeclock, jiffies))
- 		SHAPERCB(skb)->shapeclock=jiffies;
- 	skb->priority=0;	/* short term bug fix */
- 	SHAPERCB(skb)->shapestamp=jiffies;
-
- 	/*
- 	 *	Time slots for this packet.
- 	 */
-
- 	SHAPERCB(skb)->shapelen= shaper_clocks(shaper,skb);
-
-	{
-		struct sk_buff *tmp;
-		/*
-		 *	Up our shape clock by the time pending on the queue
-		 *	(Should keep this in the shaper as a variable..)
-		 */
-		for(tmp=skb_peek(&shaper->sendq); tmp!=NULL &&
-			tmp!=(struct sk_buff *)&shaper->sendq; tmp=tmp->next)
-			SHAPERCB(skb)->shapeclock+=SHAPERCB(tmp)->shapelen;
-		/*
-		 *	Queue over time. Spill packet.
-		 */
-		if(time_after(SHAPERCB(skb)->shapeclock,jiffies + SHAPER_LATENCY)) {
-			dev_kfree_skb(skb);
-			dev->stats.tx_dropped++;
-		} else
-			skb_queue_tail(&shaper->sendq, skb);
-	}
-
-	if(sh_debug)
- 		printk("Frame queued.\n");
- 	if(skb_queue_len(&shaper->sendq)>SHAPER_QLEN)
- 	{
- 		ptr=skb_dequeue(&shaper->sendq);
-                dev_kfree_skb(ptr);
-                dev->stats.collisions++;
- 	}
-	shaper_kick(shaper);
-	spin_unlock(&shaper->lock);
- 	return 0;
-}
-
-/*
- *	Transmit from a shaper
- */
-
-static void shaper_queue_xmit(struct shaper *shaper, struct sk_buff *skb)
-{
-	struct sk_buff *newskb=skb_clone(skb, GFP_ATOMIC);
-	if(sh_debug)
-		printk("Kick frame on %p\n",newskb);
-	if(newskb)
-	{
-		newskb->dev=shaper->dev;
-		newskb->priority=2;
-		if(sh_debug)
-			printk("Kick new frame to %s, %d\n",
-				shaper->dev->name,newskb->priority);
-		dev_queue_xmit(newskb);
-
-                shaper->dev->stats.tx_bytes += skb->len;
-		shaper->dev->stats.tx_packets++;
-
-                if(sh_debug)
-			printk("Kicked new frame out.\n");
-		dev_kfree_skb(skb);
-	}
-}
-
-/*
- *	Timer handler for shaping clock
- */
-
-static void shaper_timer(unsigned long data)
-{
-	struct shaper *shaper = (struct shaper *)data;
-
-	spin_lock(&shaper->lock);
-	shaper_kick(shaper);
-	spin_unlock(&shaper->lock);
-}
-
-/*
- *	Kick a shaper queue and try and do something sensible with the
- *	queue.
- */
-
-static void shaper_kick(struct shaper *shaper)
-{
-	struct sk_buff *skb;
-
-	/*
-	 *	Walk the list (may be empty)
-	 */
-
-	while((skb=skb_peek(&shaper->sendq))!=NULL)
-	{
-		/*
-		 *	Each packet due to go out by now (within an error
-		 *	of SHAPER_BURST) gets kicked onto the link
-		 */
-
-		if(sh_debug)
-			printk("Clock = %ld, jiffies = %ld\n", SHAPERCB(skb)->shapeclock, jiffies);
-		if(time_before_eq(SHAPERCB(skb)->shapeclock, jiffies + SHAPER_BURST))
-		{
-			/*
-			 *	Pull the frame and get interrupts back on.
-			 */
-
-			skb_unlink(skb, &shaper->sendq);
-			if (shaper->recovery <
-			    SHAPERCB(skb)->shapeclock + SHAPERCB(skb)->shapelen)
-				shaper->recovery = SHAPERCB(skb)->shapeclock + SHAPERCB(skb)->shapelen;
-			/*
-			 *	Pass on to the physical target device via
-			 *	our low level packet thrower.
-			 */
-
-			SHAPERCB(skb)->shapepend=0;
-			shaper_queue_xmit(shaper, skb);	/* Fire */
-		}
-		else
-			break;
-	}
-
-	/*
-	 *	Next kick.
-	 */
-
-	if(skb!=NULL)
-		mod_timer(&shaper->timer, SHAPERCB(skb)->shapeclock);
-}
-
-
-/*
- *	Bring the interface up. We just disallow this until a
- *	bind.
- */
-
-static int shaper_open(struct net_device *dev)
-{
-	struct shaper *shaper=dev->priv;
-
-	/*
-	 *	Can't open until attached.
-	 *	Also can't open until speed is set, or we'll get
-	 *	a division by zero.
-	 */
-
-	if(shaper->dev==NULL)
-		return -ENODEV;
-	if(shaper->bitspersec==0)
-		return -EINVAL;
-	return 0;
-}
-
-/*
- *	Closing a shaper flushes the queues.
- */
-
-static int shaper_close(struct net_device *dev)
-{
-	struct shaper *shaper=dev->priv;
-	struct sk_buff *skb;
-
-	while ((skb = skb_dequeue(&shaper->sendq)) != NULL)
-		dev_kfree_skb(skb);
-
-	spin_lock_bh(&shaper->lock);
-	shaper_kick(shaper);
-	spin_unlock_bh(&shaper->lock);
-
-	del_timer_sync(&shaper->timer);
-	return 0;
-}
-
-/*
- *	Revectored calls. We alter the parameters and call the functions
- *	for our attached device. This enables us to bandwidth allocate after
- *	ARP and other resolutions and not before.
- */
-
-static int shaper_header(struct sk_buff *skb, struct net_device *dev,
-			 unsigned short type,
-			 const void *daddr, const void *saddr, unsigned len)
-{
-	struct shaper *sh=dev->priv;
-	int v;
-	if(sh_debug)
-		printk("Shaper header\n");
-	skb->dev = sh->dev;
-	v = dev_hard_header(skb, sh->dev, type, daddr, saddr, len);
-	skb->dev = dev;
-	return v;
-}
-
-static int shaper_rebuild_header(struct sk_buff *skb)
-{
-	struct shaper *sh=skb->dev->priv;
-	struct net_device *dev=skb->dev;
-	int v;
-	if(sh_debug)
-		printk("Shaper rebuild header\n");
-	skb->dev=sh->dev;
-	v = sh->dev->header_ops->rebuild(skb);
-	skb->dev=dev;
-	return v;
-}
-
-#if 0
-static int shaper_cache(struct neighbour *neigh, struct hh_cache *hh)
-{
-	struct shaper *sh=neigh->dev->priv;
-	struct net_device *tmp;
-	int ret;
-	if(sh_debug)
-		printk("Shaper header cache bind\n");
-	tmp=neigh->dev;
-	neigh->dev=sh->dev;
-	ret=sh->hard_header_cache(neigh,hh);
-	neigh->dev=tmp;
-	return ret;
-}
-
-static void shaper_cache_update(struct hh_cache *hh, struct net_device *dev,
-	unsigned char *haddr)
-{
-	struct shaper *sh=dev->priv;
-	if(sh_debug)
-		printk("Shaper cache update\n");
-	sh->header_cache_update(hh, sh->dev, haddr);
-}
-#endif
-
-#ifdef CONFIG_INET
-
-static int shaper_neigh_setup(struct neighbour *n)
-{
-#ifdef CONFIG_INET
-	if (n->nud_state == NUD_NONE) {
-		n->ops = &arp_broken_ops;
-		n->output = n->ops->output;
-	}
-#endif
-	return 0;
-}
-
-static int shaper_neigh_setup_dev(struct net_device *dev, struct neigh_parms *p)
-{
-#ifdef CONFIG_INET
-	if (p->tbl->family == AF_INET) {
-		p->neigh_setup = shaper_neigh_setup;
-		p->ucast_probes = 0;
-		p->mcast_probes = 0;
-	}
-#endif
-	return 0;
-}
-
-#else /* !(CONFIG_INET) */
-
-static int shaper_neigh_setup_dev(struct net_device *dev, struct neigh_parms *p)
-{
-	return 0;
-}
-
-#endif
-
-static const struct header_ops shaper_ops = {
-	.create	 = shaper_header,
-	.rebuild = shaper_rebuild_header,
-};
-
-static int shaper_attach(struct net_device *shdev, struct shaper *sh, struct net_device *dev)
-{
-	sh->dev = dev;
-	sh->get_stats=dev->get_stats;
-
-	shdev->neigh_setup = shaper_neigh_setup_dev;
-	shdev->hard_header_len=dev->hard_header_len;
-	shdev->type=dev->type;
-	shdev->addr_len=dev->addr_len;
-	shdev->mtu=dev->mtu;
-	sh->bitspersec=0;
-	return 0;
-}
-
-static int shaper_ioctl(struct net_device *dev,  struct ifreq *ifr, int cmd)
-{
-	struct shaperconf *ss= (struct shaperconf *)&ifr->ifr_ifru;
-	struct shaper *sh=dev->priv;
-
-	if(ss->ss_cmd == SHAPER_SET_DEV || ss->ss_cmd == SHAPER_SET_SPEED)
-	{
-		if(!capable(CAP_NET_ADMIN))
-			return -EPERM;
-	}
-
-	switch(ss->ss_cmd)
-	{
-		case SHAPER_SET_DEV:
-		{
-			struct net_device *them=__dev_get_by_name(&init_net, ss->ss_name);
-			if(them==NULL)
-				return -ENODEV;
-			if(sh->dev)
-				return -EBUSY;
-			return shaper_attach(dev,dev->priv, them);
-		}
-		case SHAPER_GET_DEV:
-			if(sh->dev==NULL)
-				return -ENODEV;
-			strcpy(ss->ss_name, sh->dev->name);
-			return 0;
-		case SHAPER_SET_SPEED:
-			shaper_setspeed(sh,ss->ss_speed);
-			return 0;
-		case SHAPER_GET_SPEED:
-			ss->ss_speed=sh->bitspersec;
-			return 0;
-		default:
-			return -EINVAL;
-	}
-}
-
-static void shaper_init_priv(struct net_device *dev)
-{
-	struct shaper *sh = dev->priv;
-
-	skb_queue_head_init(&sh->sendq);
-	init_timer(&sh->timer);
-	sh->timer.function=shaper_timer;
-	sh->timer.data=(unsigned long)sh;
-	spin_lock_init(&sh->lock);
-}
-
-/*
- *	Add a shaper device to the system
- */
-
-static void __init shaper_setup(struct net_device *dev)
-{
-	/*
-	 *	Set up the shaper.
-	 */
-
-	shaper_init_priv(dev);
-
-	dev->open		= shaper_open;
-	dev->stop		= shaper_close;
-	dev->hard_start_xmit 	= shaper_start_xmit;
-	dev->set_multicast_list = NULL;
-
-	/*
-	 *	Intialise the packet queues
-	 */
-
-	/*
-	 *	Handlers for when we attach to a device.
-	 */
-
-	dev->neigh_setup	= shaper_neigh_setup_dev;
-	dev->do_ioctl		= shaper_ioctl;
-	dev->hard_header_len	= 0;
-	dev->type		= ARPHRD_ETHER;	/* initially */
-	dev->set_mac_address	= NULL;
-	dev->mtu		= 1500;
-	dev->addr_len		= 0;
-	dev->tx_queue_len	= 10;
-	dev->flags		= 0;
-}
-
-static int shapers = 1;
-#ifdef MODULE
-
-module_param(shapers, int, 0);
-MODULE_PARM_DESC(shapers, "Traffic shaper: maximum number of shapers");
-
-#else /* MODULE */
-
-static int __init set_num_shapers(char *str)
-{
-	shapers = simple_strtol(str, NULL, 0);
-	return 1;
-}
-
-__setup("shapers=", set_num_shapers);
-
-#endif /* MODULE */
-
-static struct net_device **devs;
-
-static unsigned int shapers_registered = 0;
-
-static int __init shaper_init(void)
-{
-	int i;
-	size_t alloc_size;
-	struct net_device *dev;
-	char name[IFNAMSIZ];
-
-	if (shapers < 1)
-		return -ENODEV;
-
-	alloc_size = sizeof(*dev) * shapers;
-	devs = kzalloc(alloc_size, GFP_KERNEL);
-	if (!devs)
-		return -ENOMEM;
-
-	for (i = 0; i < shapers; i++) {
-
-		snprintf(name, IFNAMSIZ, "shaper%d", i);
-		dev = alloc_netdev(sizeof(struct shaper), name,
-				   shaper_setup);
-		if (!dev)
-			break;
-
-		if (register_netdev(dev)) {
-			free_netdev(dev);
-			break;
-		}
-
-		devs[i] = dev;
-		shapers_registered++;
-	}
-
-	if (!shapers_registered) {
-		kfree(devs);
-		devs = NULL;
-	}
-
-	return (shapers_registered ? 0 : -ENODEV);
-}
-
-static void __exit shaper_exit (void)
-{
-	int i;
-
-	for (i = 0; i < shapers_registered; i++) {
-		if (devs[i]) {
-			unregister_netdev(devs[i]);
-			free_netdev(devs[i]);
-		}
-	}
-
-	kfree(devs);
-	devs = NULL;
-}
-
-module_init(shaper_init);
-module_exit(shaper_exit);
-MODULE_LICENSE("GPL");
-
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index a85e87fd60d7..bc33a5c87d64 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -231,7 +231,6 @@ unifdef-y += if_ltalk.h
 unifdef-y += if_link.h
 unifdef-y += if_pppol2tp.h
 unifdef-y += if_pppox.h
-unifdef-y += if_shaper.h
 unifdef-y += if_tr.h
 unifdef-y += if_tun.h
 unifdef-y += if_vlan.h
diff --git a/include/linux/if_shaper.h b/include/linux/if_shaper.h
deleted file mode 100644
index 3b1b7ba19825..000000000000
--- a/include/linux/if_shaper.h
+++ /dev/null
@@ -1,51 +0,0 @@
-#ifndef __LINUX_SHAPER_H
-#define __LINUX_SHAPER_H
-
-#ifdef __KERNEL__
-
-#define SHAPER_QLEN	10
-/*
- *	This is a bit speed dependent (read it shouldn't be a constant!)
- *
- *	5 is about right for 28.8 upwards. Below that double for every
- *	halving of speed or so. - ie about 20 for 9600 baud.
- */
-#define SHAPER_LATENCY	(5*HZ)
-#define SHAPER_MAXSLIP	2
-#define SHAPER_BURST	(HZ/50)		/* Good for >128K then */
-
-struct shaper
-{
-	struct sk_buff_head sendq;
-	__u32 bytespertick;
-	__u32 bitspersec;
-	__u32 shapelatency;
-	__u32 shapeclock;
-	unsigned long recovery;	/* Time we can next clock a packet out on
-				   an empty queue */
-	spinlock_t lock;
-	struct net_device *dev;
-	struct net_device_stats* (*get_stats)(struct net_device *dev);
-	struct timer_list timer;
-};
-
-#endif
-
-#define SHAPER_SET_DEV		0x0001
-#define SHAPER_SET_SPEED	0x0002
-#define SHAPER_GET_DEV		0x0003
-#define SHAPER_GET_SPEED	0x0004
-
-struct shaperconf
-{
-	__u16	ss_cmd;
-	union
-	{
-		char 	ssu_name[14];
-		__u32	ssu_speed;
-	} ss_u;
-#define ss_speed ss_u.ssu_speed
-#define ss_name ss_u.ssu_name
-};
-
-#endif
-- 
cgit v1.2.3


From b798232fcc1c608433e1927df1d49b13b066b698 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Tue, 1 Jan 2008 21:13:09 -0800
Subject: [IPV4]: Remove three declarations of unimplemented methods and
 correct a typo in include/net/ip.h

These three declarations in include/net/ip.h are not implemented
anywhere:

ip_mc_dropsocket(), ip_mc_dropdevice() and ip_net_unreachable().

Also, correct a comment to be "Functions provided by ip_fragment.c"
(instead of by ip_fragment.o) in consistency with the other comments
in this header.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 571bcf2a1cff..cbeb97d38860 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -82,8 +82,6 @@ struct packet_type;
 struct rtable;
 struct sockaddr;
 
-extern void		ip_mc_dropsocket(struct sock *);
-extern void		ip_mc_dropdevice(struct net_device *dev);
 extern int		igmp_mc_proc_init(void);
 
 /*
@@ -321,7 +319,7 @@ static __inline__ void inet_reset_saddr(struct sock *sk)
 extern int	ip_call_ra_chain(struct sk_buff *skb);
 
 /*
- *	Functions provided by ip_fragment.o
+ *	Functions provided by ip_fragment.c
  */
 
 enum ip_defrag_users
@@ -344,7 +342,6 @@ int ip_frag_nqueues(void);
  */
  
 extern int ip_forward(struct sk_buff *skb);
-extern int ip_net_unreachable(struct sk_buff *skb);
  
 /*
  *	Functions provided by ip_options.c
-- 
cgit v1.2.3


From b950dfcf504842a0f2bb142a33ef938f1c00d300 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Tue, 1 Jan 2008 21:17:19 -0800
Subject: [IPVS]: Remove declaration of unimplemented method and remove unused
 definition from include/net/ip_vs.h

In include/net/ip_vs.h:
- The ip_vs_secure_tcp_set() method is not implemented anywhere.
- IP_VS_APP_TYPE_FTP is an unused definition.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_vs.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 8a7d59be8a0d..3de6d1e6afa7 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -676,7 +676,6 @@ extern const char *ip_vs_proto_name(unsigned proto);
 extern void ip_vs_init_hash_table(struct list_head *table, int rows);
 #define IP_VS_INIT_HASH_TABLE(t) ip_vs_init_hash_table(t, sizeof(t)/sizeof(t[0]))
 
-#define IP_VS_APP_TYPE_UNSPEC	0
 #define IP_VS_APP_TYPE_FTP	1
 
 /*
@@ -735,7 +734,6 @@ extern const char * ip_vs_state_name(__u16 proto, int state);
 
 extern void ip_vs_tcp_conn_listen(struct ip_vs_conn *cp);
 extern int ip_vs_check_template(struct ip_vs_conn *ct);
-extern void ip_vs_secure_tcp_set(int on);
 extern void ip_vs_random_dropentry(void);
 extern int ip_vs_conn_init(void);
 extern void ip_vs_conn_cleanup(void);
-- 
cgit v1.2.3


From 571e7682026fd0e25833d103a3eeb74be29bf199 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Thu, 3 Jan 2008 20:41:28 -0800
Subject: [LIB] pcounter : unline too big functions

Before pushing pcounter to Linus tree, I would like to make some adjustments.

Goal is to reduce kernel text size, by unlining too big functions.

When a pcounter is bound to a statically defined per_cpu variable,
we define two small helpers functions. (No more folding function
using the fat for_each_possible_cpu(cpu) ... )

static DEFINE_PER_CPU(int, NAME##_pcounter_values);
static void NAME##_pcounter_add(struct pcounter *self, int val)
{
       __get_cpu_var(NAME##_pcounter_values) += val;
}
static int NAME##_pcounter_getval(const struct pcounter *self, int cpu)
{
       return per_cpu(NAME##_pcounter_values, cpu);
}

Fast path is therefore unchanged, while folding/alloc/free is now unlined.

This saves 228 bytes on i386

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pcounter.h | 80 ++++++++++++++++--------------------------------
 lib/pcounter.c           | 42 ++++++++++++++++++++++---
 2 files changed, 63 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/include/linux/pcounter.h b/include/linux/pcounter.h
index 9c4760a328f3..a82d9f2628ca 100644
--- a/include/linux/pcounter.h
+++ b/include/linux/pcounter.h
@@ -1,41 +1,39 @@
 #ifndef __LINUX_PCOUNTER_H
 #define __LINUX_PCOUNTER_H
-
+/*
+ * Using a dynamic percpu 'int' variable has a cost :
+ * 1) Extra dereference
+ * Current per_cpu_ptr() implementation uses an array per 'percpu variable'.
+ * 2) memory cost of NR_CPUS*(32+sizeof(void *)) instead of num_possible_cpus()*4
+ *
+ * This pcounter implementation is an abstraction to be able to use
+ * either a static or a dynamic per cpu variable.
+ * One dynamic per cpu variable gets a fast & cheap implementation, we can
+ * change pcounter implementation too.
+ */
 struct pcounter {
 #ifdef CONFIG_SMP
 	void		(*add)(struct pcounter *self, int inc);
-	int		(*getval)(const struct pcounter *self);
+	int		(*getval)(const struct pcounter *self, int cpu);
 	int		*per_cpu_values;
 #else
 	int		val;
 #endif
 };
 
-/*
- * Special macros to let pcounters use a fast version of {getvalue|add}
- * using a static percpu variable per pcounter instead of an allocated one,
- * saving one dereference.
- * This might be changed if/when dynamic percpu vars become fast.
- */
 #ifdef CONFIG_SMP
-#include <linux/cpumask.h>
 #include <linux/percpu.h>
 
-#define DEFINE_PCOUNTER(NAME)					\
-static DEFINE_PER_CPU(int, NAME##_pcounter_values);		\
-static void NAME##_pcounter_add(struct pcounter *self, int inc)	\
-{								\
-       __get_cpu_var(NAME##_pcounter_values) += inc;		\
-}								\
-								\
-static int NAME##_pcounter_getval(const struct pcounter *self)	\
-{								\
-       int res = 0, cpu;					\
-								\
-       for_each_possible_cpu(cpu)				\
-               res += per_cpu(NAME##_pcounter_values, cpu);	\
-       return res;						\
-}
+#define DEFINE_PCOUNTER(NAME)						\
+static DEFINE_PER_CPU(int, NAME##_pcounter_values);			\
+static void NAME##_pcounter_add(struct pcounter *self, int val)		\
+{									\
+       __get_cpu_var(NAME##_pcounter_values) += val;			\
+}									\
+static int NAME##_pcounter_getval(const struct pcounter *self, int cpu)	\
+{									\
+	return per_cpu(NAME##_pcounter_values, cpu);			\
+}									\
 
 #define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER)		\
 	MEMBER = {						\
@@ -43,42 +41,16 @@ static int NAME##_pcounter_getval(const struct pcounter *self)	\
 		.getval = NAME##_pcounter_getval,		\
 	}
 
-extern void pcounter_def_add(struct pcounter *self, int inc);
-extern int pcounter_def_getval(const struct pcounter *self);
-
-static inline int pcounter_alloc(struct pcounter *self)
-{
-	int rc = 0;
-	if (self->add == NULL) {
-		self->per_cpu_values = alloc_percpu(int);
-		if (self->per_cpu_values != NULL) {
-			self->add    = pcounter_def_add;
-			self->getval = pcounter_def_getval;
-		} else
-			rc = 1;
-	}
-	return rc;
-}
-
-static inline void pcounter_free(struct pcounter *self)
-{
-	if (self->per_cpu_values != NULL) {
-		free_percpu(self->per_cpu_values);
-		self->per_cpu_values = NULL;
-		self->getval = NULL;
-		self->add = NULL;
-	}
-}
 
 static inline void pcounter_add(struct pcounter *self, int inc)
 {
 	self->add(self, inc);
 }
 
-static inline int pcounter_getval(const struct pcounter *self)
-{
-	return self->getval(self);
-}
+extern int pcounter_getval(const struct pcounter *self);
+extern int pcounter_alloc(struct pcounter *self);
+extern void pcounter_free(struct pcounter *self);
+
 
 #else /* CONFIG_SMP */
 
diff --git a/lib/pcounter.c b/lib/pcounter.c
index 93feea598251..9b56807da93b 100644
--- a/lib/pcounter.c
+++ b/lib/pcounter.c
@@ -7,20 +7,52 @@
 #include <linux/module.h>
 #include <linux/pcounter.h>
 #include <linux/smp.h>
+#include <linux/cpumask.h>
 
-void pcounter_def_add(struct pcounter *self, int inc)
+static void pcounter_dyn_add(struct pcounter *self, int inc)
 {
 	per_cpu_ptr(self->per_cpu_values, smp_processor_id())[0] += inc;
 }
 
-EXPORT_SYMBOL_GPL(pcounter_def_add);
+static int pcounter_dyn_getval(const struct pcounter *self, int cpu)
+{
+	return per_cpu_ptr(self->per_cpu_values, cpu)[0];
+}
 
-int pcounter_def_getval(const struct pcounter *self)
+int pcounter_getval(const struct pcounter *self)
 {
 	int res = 0, cpu;
+
 	for_each_possible_cpu(cpu)
-		res += per_cpu_ptr(self->per_cpu_values, cpu)[0];
+		res += self->getval(self, cpu);
+
 	return res;
 }
+EXPORT_SYMBOL_GPL(pcounter_getval);
+
+int pcounter_alloc(struct pcounter *self)
+{
+	int rc = 0;
+	if (self->add == NULL) {
+		self->per_cpu_values = alloc_percpu(int);
+		if (self->per_cpu_values != NULL) {
+			self->add    = pcounter_dyn_add;
+			self->getval = pcounter_dyn_getval;
+		} else
+			rc = 1;
+	}
+	return rc;
+}
+EXPORT_SYMBOL_GPL(pcounter_alloc);
+
+void pcounter_free(struct pcounter *self)
+{
+	if (self->per_cpu_values != NULL) {
+		free_percpu(self->per_cpu_values);
+		self->per_cpu_values = NULL;
+		self->getval = NULL;
+		self->add = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(pcounter_free);
 
-EXPORT_SYMBOL_GPL(pcounter_def_getval);
-- 
cgit v1.2.3


From 65f7651788e18fadb2fbb7276af935d7871e1803 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Thu, 3 Jan 2008 20:46:48 -0800
Subject: [NET]: prot_inuse cleanups and optimizations

1) Cleanups (all functions are prefixed by sock_prot_inuse)

sock_prot_inc_use(prot) -> sock_prot_inuse_add(prot,-1)
sock_prot_dec_use(prot) -> sock_prot_inuse_add(prot,-1)
sock_prot_inuse()       -> sock_prot_inuse_get()

New functions :

sock_prot_inuse_init() and sock_prot_inuse_free() to abstract pcounter use.

2) if CONFIG_PROC_FS=n, we can zap 'inuse' member from "struct proto",
since nobody wants to read the inuse value.

This saves 1372 bytes on i386/SMP and some cpu cycles.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h |  2 +-
 include/net/sock.h            | 40 ++++++++++++++++++++++++++++++----------
 include/net/udp.h             |  2 +-
 net/core/sock.c               |  6 +++---
 net/ipv4/inet_hashtables.c    |  6 +++---
 net/ipv4/inet_timewait_sock.c |  2 +-
 net/ipv4/proc.c               |  9 +++++----
 net/ipv4/raw.c                |  4 ++--
 net/ipv4/udp.c                |  2 +-
 net/ipv6/inet6_hashtables.c   |  4 ++--
 net/ipv6/ipv6_sockglue.c      |  8 ++++----
 net/ipv6/proc.c               |  8 ++++----
 12 files changed, 57 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 65ddb25d62e8..761bdc01425d 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -293,7 +293,7 @@ static inline void inet_unhash(struct inet_hashinfo *hashinfo, struct sock *sk)
 	}
 
 	if (__sk_del_node_init(sk))
-		sock_prot_dec_use(sk->sk_prot);
+		sock_prot_inuse_add(sk->sk_prot, -1);
 	write_unlock_bh(lock);
 out:
 	if (sk->sk_state == TCP_LISTEN)
diff --git a/include/net/sock.h b/include/net/sock.h
index 3d938f6c6725..786fae858e77 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -548,7 +548,9 @@ struct proto {
 	int			(*get_port)(struct sock *sk, unsigned short snum);
 
 	/* Keeping track of sockets in use */
+#ifdef CONFIG_PROC_FS
 	struct pcounter		inuse;
+#endif
 
 	/* Memory pressure */
 	void			(*enter_memory_pressure)(void);
@@ -584,9 +586,6 @@ struct proto {
 #endif
 };
 
-#define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME)
-#define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
-
 extern int proto_register(struct proto *prot, int alloc_slab);
 extern void proto_unregister(struct proto *prot);
 
@@ -615,21 +614,42 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 #define sk_refcnt_debug_release(sk) do { } while (0)
 #endif /* SOCK_REFCNT_DEBUG */
 
+
+#ifdef CONFIG_PROC_FS
+# define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME)
+# define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
 /* Called with local bh disabled */
-static __inline__ void sock_prot_inc_use(struct proto *prot)
+static inline void sock_prot_inuse_add(struct proto *prot, int inc)
 {
-	pcounter_add(&prot->inuse, 1);
+	pcounter_add(&prot->inuse, inc);
 }
-
-static __inline__ void sock_prot_dec_use(struct proto *prot)
+static inline int sock_prot_inuse_init(struct proto *proto)
 {
-	pcounter_add(&prot->inuse, -1);
+	return pcounter_alloc(&proto->inuse);
 }
-
-static __inline__ int sock_prot_inuse(struct proto *proto)
+static inline int sock_prot_inuse_get(struct proto *proto)
 {
 	return pcounter_getval(&proto->inuse);
 }
+static inline void sock_prot_inuse_free(struct proto *proto)
+{
+	pcounter_free(&proto->inuse);
+}
+#else
+# define DEFINE_PROTO_INUSE(NAME)
+# define REF_PROTO_INUSE(NAME)
+static void inline sock_prot_inuse_add(struct proto *prot, int inc)
+{
+}
+static int inline sock_prot_inuse_init(struct proto *proto)
+{
+	return 0;
+}
+static void inline sock_prot_inuse_free(struct proto *proto)
+{
+}
+#endif
+
 
 /* With per-bucket locks this operation is not-atomic, so that
  * this version is not worse.
diff --git a/include/net/udp.h b/include/net/udp.h
index 93796beac8ff..c6669c0a74c7 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -115,7 +115,7 @@ static inline void udp_lib_unhash(struct sock *sk)
 	write_lock_bh(&udp_hash_lock);
 	if (sk_del_node_init(sk)) {
 		inet_sk(sk)->num = 0;
-		sock_prot_dec_use(sk->sk_prot);
+		sock_prot_inuse_add(sk->sk_prot, -1);
 	}
 	write_unlock_bh(&udp_hash_lock);
 }
diff --git a/net/core/sock.c b/net/core/sock.c
index 3d7757ee2fc8..1c4b1cd16d65 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1913,7 +1913,7 @@ int proto_register(struct proto *prot, int alloc_slab)
 	char *request_sock_slab_name = NULL;
 	char *timewait_sock_slab_name;
 
-	if (pcounter_alloc(&prot->inuse) != 0) {
+	if (sock_prot_inuse_init(prot) != 0) {
 		printk(KERN_CRIT "%s: Can't alloc inuse counters!\n", prot->name);
 		goto out;
 	}
@@ -1984,7 +1984,7 @@ out_free_sock_slab:
 	kmem_cache_destroy(prot->slab);
 	prot->slab = NULL;
 out_free_inuse:
-	pcounter_free(&prot->inuse);
+	sock_prot_inuse_free(prot);
 out:
 	return -ENOBUFS;
 }
@@ -1997,7 +1997,7 @@ void proto_unregister(struct proto *prot)
 	list_del(&prot->node);
 	write_unlock(&proto_list_lock);
 
-	pcounter_free(&prot->inuse);
+	sock_prot_inuse_free(prot);
 
 	if (prot->slab != NULL) {
 		kmem_cache_destroy(prot->slab);
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 88a059e04e30..619c63c6948a 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -278,7 +278,7 @@ unique:
 	sk->sk_hash = hash;
 	BUG_TRAP(sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock(lock);
 
 	if (twp) {
@@ -321,7 +321,7 @@ void __inet_hash_nolisten(struct inet_hashinfo *hashinfo, struct sock *sk)
 
 	write_lock(lock);
 	__sk_add_node(sk, list);
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock(lock);
 }
 EXPORT_SYMBOL_GPL(__inet_hash_nolisten);
@@ -342,7 +342,7 @@ void __inet_hash(struct inet_hashinfo *hashinfo, struct sock *sk)
 
 	inet_listen_wlock(hashinfo);
 	__sk_add_node(sk, list);
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock(lock);
 	wake_up(&hashinfo->lhash_wait);
 }
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 1b7db4208969..876169f3a528 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -91,7 +91,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 
 	/* Step 2: Remove SK from established hash. */
 	if (__sk_del_node_init(sk))
-		sock_prot_dec_use(sk->sk_prot);
+		sock_prot_inuse_add(sk->sk_prot, -1);
 
 	/* Step 3: Hash TW into TIMEWAIT chain. */
 	inet_twsk_add_node(tw, &ehead->twchain);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 53bc010beefd..cb3787fbeb90 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -53,13 +53,14 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 {
 	socket_seq_show(seq);
 	seq_printf(seq, "TCP: inuse %d orphan %d tw %d alloc %d mem %d\n",
-		   sock_prot_inuse(&tcp_prot), atomic_read(&tcp_orphan_count),
+		   sock_prot_inuse_get(&tcp_prot),
+		   atomic_read(&tcp_orphan_count),
 		   tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated),
 		   atomic_read(&tcp_memory_allocated));
-	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse(&udp_prot),
+	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot),
 		   atomic_read(&udp_memory_allocated));
-	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse(&udplite_prot));
-	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse(&raw_prot));
+	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
+	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
 			ip_frag_nqueues(), ip_frag_mem());
 	return 0;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 5aec5a5e5f16..e811034b1b01 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -92,7 +92,7 @@ void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h)
 
 	write_lock_bh(&h->lock);
 	sk_add_node(sk, head);
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock_bh(&h->lock);
 }
 EXPORT_SYMBOL_GPL(raw_hash_sk);
@@ -101,7 +101,7 @@ void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h)
 {
 	write_lock_bh(&h->lock);
 	if (sk_del_node_init(sk))
-		sock_prot_dec_use(sk->sk_prot);
+		sock_prot_inuse_add(sk->sk_prot, -1);
 	write_unlock_bh(&h->lock);
 }
 EXPORT_SYMBOL_GPL(raw_unhash_sk);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 02fcccd0486e..cb2411cb87da 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -230,7 +230,7 @@ gotit:
 	if (sk_unhashed(sk)) {
 		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
 		sk_add_node(sk, head);
-		sock_prot_inc_use(sk->sk_prot);
+		sock_prot_inuse_add(sk->sk_prot, 1);
 	}
 	error = 0;
 fail:
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 0765d8bd380f..a66a7d8e2811 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -43,7 +43,7 @@ void __inet6_hash(struct inet_hashinfo *hashinfo,
 	}
 
 	__sk_add_node(sk, list);
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock(lock);
 }
 EXPORT_SYMBOL(__inet6_hash);
@@ -216,7 +216,7 @@ unique:
 	BUG_TRAP(sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
 	sk->sk_hash = hash;
-	sock_prot_inc_use(sk->sk_prot);
+	sock_prot_inuse_add(sk->sk_prot, 1);
 	write_unlock(lock);
 
 	if (twp != NULL) {
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 20fece4ad3d8..bf2a686aa13d 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -268,8 +268,8 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 				struct inet_connection_sock *icsk = inet_csk(sk);
 
 				local_bh_disable();
-				sock_prot_dec_use(sk->sk_prot);
-				sock_prot_inc_use(&tcp_prot);
+				sock_prot_inuse_add(sk->sk_prot, -1);
+				sock_prot_inuse_add(&tcp_prot, 1);
 				local_bh_enable();
 				sk->sk_prot = &tcp_prot;
 				icsk->icsk_af_ops = &ipv4_specific;
@@ -282,8 +282,8 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 				if (sk->sk_protocol == IPPROTO_UDPLITE)
 					prot = &udplite_prot;
 				local_bh_disable();
-				sock_prot_dec_use(sk->sk_prot);
-				sock_prot_inc_use(prot);
+				sock_prot_inuse_add(sk->sk_prot, -1);
+				sock_prot_inuse_add(prot, 1);
 				local_bh_enable();
 				sk->sk_prot = prot;
 				sk->sk_socket->ops = &inet_dgram_ops;
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 41e9980b3e0e..571d95a21c15 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -36,13 +36,13 @@ static struct proc_dir_entry *proc_net_devsnmp6;
 static int sockstat6_seq_show(struct seq_file *seq, void *v)
 {
 	seq_printf(seq, "TCP6: inuse %d\n",
-		       sock_prot_inuse(&tcpv6_prot));
+		       sock_prot_inuse_get(&tcpv6_prot));
 	seq_printf(seq, "UDP6: inuse %d\n",
-		       sock_prot_inuse(&udpv6_prot));
+		       sock_prot_inuse_get(&udpv6_prot));
 	seq_printf(seq, "UDPLITE6: inuse %d\n",
-			sock_prot_inuse(&udplitev6_prot));
+			sock_prot_inuse_get(&udplitev6_prot));
 	seq_printf(seq, "RAW6: inuse %d\n",
-		       sock_prot_inuse(&rawv6_prot));
+		       sock_prot_inuse_get(&rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
 		       ip6_frag_nqueues(), ip6_frag_mem());
 	return 0;
-- 
cgit v1.2.3


From 96a899655e2c3ba53f083eda69706ee4eb05271f Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 4 Jan 2008 01:59:20 -0800
Subject: [CONNECTOR]: Cleanup struct cn_queue_dev

Struct member netlink_groups is never used, and I don't see how it can
be useful.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/connector/cn_queue.c | 1 -
 include/linux/connector.h    | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include')

diff --git a/drivers/connector/cn_queue.c b/drivers/connector/cn_queue.c
index 12ceed54ab18..dbda08c7b3db 100644
--- a/drivers/connector/cn_queue.c
+++ b/drivers/connector/cn_queue.c
@@ -146,7 +146,6 @@ struct cn_queue_dev *cn_queue_alloc_dev(char *name, struct sock *nls)
 	spin_lock_init(&dev->queue_lock);
 
 	dev->nls = nls;
-	dev->netlink_groups = 0;
 
 	dev->cn_queue = create_workqueue(dev->name);
 	if (!dev->cn_queue) {
diff --git a/include/linux/connector.h b/include/linux/connector.h
index 13fc4541bf23..7e18311d655e 100644
--- a/include/linux/connector.h
+++ b/include/linux/connector.h
@@ -112,7 +112,6 @@ struct cn_queue_dev {
 	struct list_head queue_list;
 	spinlock_t queue_lock;
 
-	int netlink_groups;
 	struct sock *nls;
 };
 
-- 
cgit v1.2.3


From 6e32814bc89e7103e2b75b841155faf51f60a8f1 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 4 Jan 2008 01:59:42 -0800
Subject: [CONNECTOR]: Cleanup struct cn_callback_entry

- 'cb' is a fake struct member. In a previous patch struct cn_callback
was renamed to cn_callback_id, so 'cb' should have been deleted at that
time.

- 'nls' isn't used and is redundant, we can retrieve this data through
cn_callback_entry.pdev->nls.

- 'seq' and 'group' should be u32, as they are declared to be u32 in
other places.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/connector/cn_queue.c | 1 -
 include/linux/connector.h    | 4 +---
 2 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/connector/cn_queue.c b/drivers/connector/cn_queue.c
index dbda08c7b3db..5732ca3259f9 100644
--- a/drivers/connector/cn_queue.c
+++ b/drivers/connector/cn_queue.c
@@ -104,7 +104,6 @@ int cn_queue_add_callback(struct cn_queue_dev *dev, char *name, struct cb_id *id
 		return -EINVAL;
 	}
 
-	cbq->nls = dev->nls;
 	cbq->seq = 0;
 	cbq->group = cbq->id.id.idx;
 
diff --git a/include/linux/connector.h b/include/linux/connector.h
index 7e18311d655e..da6dd957f908 100644
--- a/include/linux/connector.h
+++ b/include/linux/connector.h
@@ -132,15 +132,13 @@ struct cn_callback_data {
 
 struct cn_callback_entry {
 	struct list_head callback_entry;
-	struct cn_callback *cb;
 	struct work_struct work;
 	struct cn_queue_dev *pdev;
 
 	struct cn_callback_id id;
 	struct cn_callback_data data;
 
-	int seq, group;
-	struct sock *nls;
+	u32 seq, group;
 };
 
 struct cn_ctl_entry {
-- 
cgit v1.2.3


From 2a75de0c1de2dde9ef41aeb45a21048681421b8a Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Sat, 5 Jan 2008 23:08:49 -0800
Subject: [NETNS]: Should build with CONFIG_SYSCTL=n

Previous NETNS patches broke CONFIG_SYSCTL=n case

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  2 ++
 net/ipv4/devinet.c       | 17 ++++++++++-------
 2 files changed, 12 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index e06d7cfd6d64..61a28ff9440a 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -8,7 +8,9 @@ struct ctl_table_header;
 struct ipv4_devconf;
 
 struct netns_ipv4 {
+#ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*forw_hdr;
+#endif
 	struct ipv4_devconf	*devconf_all;
 	struct ipv4_devconf	*devconf_dflt;
 };
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 44cb252d2f61..03db15b10309 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1542,7 +1542,6 @@ static void devinet_sysctl_unregister(struct in_device *idev)
 	__devinet_sysctl_unregister(&idev->cnf);
 	neigh_sysctl_unregister(idev->arp_parms);
 }
-#endif
 
 static struct ctl_table ctl_forward_entry[] = {
 	{
@@ -1565,18 +1564,20 @@ static __net_initdata struct ctl_path net_ipv4_path[] = {
 	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
 	{ },
 };
+#endif
 
 static __net_init int devinet_init_net(struct net *net)
 {
 	int err;
-	struct ctl_table *tbl;
 	struct ipv4_devconf *all, *dflt;
+#ifdef CONFIG_SYSCTL
+	struct ctl_table *tbl = ctl_forward_entry;
 	struct ctl_table_header *forw_hdr;
+#endif
 
 	err = -ENOMEM;
 	all = &ipv4_devconf;
 	dflt = &ipv4_devconf_dflt;
-	tbl = ctl_forward_entry;
 
 	if (net != &init_net) {
 		all = kmemdup(all, sizeof(ipv4_devconf), GFP_KERNEL);
@@ -1587,6 +1588,7 @@ static __net_init int devinet_init_net(struct net *net)
 		if (dflt == NULL)
 			goto err_alloc_dflt;
 
+#ifdef CONFIG_SYSCTL
 		tbl = kmemdup(tbl, sizeof(ctl_forward_entry), GFP_KERNEL);
 		if (tbl == NULL)
 			goto err_alloc_ctl;
@@ -1594,6 +1596,7 @@ static __net_init int devinet_init_net(struct net *net)
 		tbl[0].data = &all->data[NET_IPV4_CONF_FORWARDING - 1];
 		tbl[0].extra1 = all;
 		tbl[0].extra2 = net;
+#endif
 	}
 
 #ifdef CONFIG_SYSCTL
@@ -1611,9 +1614,9 @@ static __net_init int devinet_init_net(struct net *net)
 	forw_hdr = register_net_sysctl_table(net, net_ipv4_path, tbl);
 	if (forw_hdr == NULL)
 		goto err_reg_ctl;
+	net->ipv4.forw_hdr = forw_hdr;
 #endif
 
-	net->ipv4.forw_hdr = forw_hdr;
 	net->ipv4.devconf_all = all;
 	net->ipv4.devconf_dflt = dflt;
 	return 0;
@@ -1626,8 +1629,8 @@ err_reg_dflt:
 err_reg_all:
 	if (tbl != ctl_forward_entry)
 		kfree(tbl);
-#endif
 err_alloc_ctl:
+#endif
 	if (dflt != &ipv4_devconf_dflt)
 		kfree(dflt);
 err_alloc_dflt:
@@ -1639,15 +1642,15 @@ err_alloc_all:
 
 static __net_exit void devinet_exit_net(struct net *net)
 {
+#ifdef CONFIG_SYSCTL
 	struct ctl_table *tbl;
 
 	tbl = net->ipv4.forw_hdr->ctl_table_arg;
-#ifdef CONFIG_SYSCTL
 	unregister_net_sysctl_table(net->ipv4.forw_hdr);
 	__devinet_sysctl_unregister(net->ipv4.devconf_dflt);
 	__devinet_sysctl_unregister(net->ipv4.devconf_all);
-#endif
 	kfree(tbl);
+#endif
 	kfree(net->ipv4.devconf_dflt);
 	kfree(net->ipv4.devconf_all);
 }
-- 
cgit v1.2.3


From 64c31b3f76482bb64459e786f9eca3bd0164d153 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Mon, 7 Jan 2008 22:34:29 -0800
Subject: [XFRM] xfrm_policy_destroy: Rename and relative fixes.

Since __xfrm_policy_destroy is used to destory the resources
allocated by xfrm_policy_alloc. So using the name
__xfrm_policy_destroy is not correspond with xfrm_policy_alloc.
Rename it to xfrm_policy_destroy.

And along with some instances that call xfrm_policy_alloc
but not using xfrm_policy_destroy to destroy the resource,
fix them.

Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 4 ++--
 net/key/af_key.c       | 6 ++----
 net/xfrm/xfrm_policy.c | 4 ++--
 net/xfrm/xfrm_user.c   | 2 +-
 4 files changed, 7 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index a419a4372e21..5ebb9ba479b1 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -626,12 +626,12 @@ static inline void xfrm_pol_hold(struct xfrm_policy *policy)
 		atomic_inc(&policy->refcnt);
 }
 
-extern void __xfrm_policy_destroy(struct xfrm_policy *policy);
+extern void xfrm_policy_destroy(struct xfrm_policy *policy);
 
 static inline void xfrm_pol_put(struct xfrm_policy *policy)
 {
 	if (atomic_dec_and_test(&policy->refcnt))
-		__xfrm_policy_destroy(policy);
+		xfrm_policy_destroy(policy);
 }
 
 #ifdef CONFIG_XFRM_SUB_POLICY
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 76dcd882f87b..16b72b5570c3 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -2291,8 +2291,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h
 	return 0;
 
 out:
-	security_xfrm_policy_free(xp);
-	kfree(xp);
+	xfrm_policy_destroy(xp);
 	return err;
 }
 
@@ -3236,8 +3235,7 @@ static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt,
 	return xp;
 
 out:
-	security_xfrm_policy_free(xp);
-	kfree(xp);
+	xfrm_policy_destroy(xp);
 	return NULL;
 }
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index d83227baaa09..534b29eb46fe 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -221,7 +221,7 @@ EXPORT_SYMBOL(xfrm_policy_alloc);
 
 /* Destroy xfrm_policy: descendant resources must be released to this moment. */
 
-void __xfrm_policy_destroy(struct xfrm_policy *policy)
+void xfrm_policy_destroy(struct xfrm_policy *policy)
 {
 	BUG_ON(!policy->dead);
 
@@ -233,7 +233,7 @@ void __xfrm_policy_destroy(struct xfrm_policy *policy)
 	security_xfrm_policy_free(policy);
 	kfree(policy);
 }
-EXPORT_SYMBOL(__xfrm_policy_destroy);
+EXPORT_SYMBOL(xfrm_policy_destroy);
 
 static void xfrm_policy_gc_kill(struct xfrm_policy *policy)
 {
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 6424e5360511..35fc16ae50ac 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -1043,7 +1043,7 @@ static struct xfrm_policy *xfrm_policy_construct(struct xfrm_userpolicy_info *p,
 	return xp;
  error:
 	*errp = err;
-	kfree(xp);
+	xfrm_policy_destroy(xp);
 	return NULL;
 }
 
-- 
cgit v1.2.3


From 3becd578c5c9aafde6f562bb1dfe20d420bce6e8 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Mon, 7 Jan 2008 22:36:54 -0800
Subject: [NET]: Remove unused member of dst_entry

The info placeholder member of dst_entry seems to be unused in the
network stack.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 31468c9aa877..e03ea0c04e01 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -80,7 +80,6 @@ struct dst_entry
 		struct rt6_info   *rt6_next;
 		struct dn_route  *dn_next;
 	};
-	char			info[0];
 };
 
 
-- 
cgit v1.2.3


From 1b7d03acbfe9d6f1ecf169e6494c5eca29fa0ed3 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Tue, 25 Dec 2007 17:00:31 +0200
Subject: mac80211: A-MPDU Rx add low level driver API

This patch adds the API to perform A-MPDU actions between mac80211 and low
level driver.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index a762a75ae62d..3784a61a8489 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -918,6 +918,18 @@ enum ieee80211_erp_change_flags {
 	IEEE80211_ERP_CHANGE_PREAMBLE	= 1<<1,
 };
 
+/**
+ * enum ieee80211_ampdu_mlme_action - A-MPDU actions
+ *
+ * These flags are used with the ampdu_action() callback in
+ * &struct ieee80211_ops to indicate which action is needed.
+ * @IEEE80211_AMPDU_RX_START: start Rx aggregation
+ * @IEEE80211_AMPDU_RX_STOP: stop Rx aggregation
+ */
+enum ieee80211_ampdu_mlme_action {
+	IEEE80211_AMPDU_RX_START,
+	IEEE80211_AMPDU_RX_STOP,
+};
 
 /**
  * struct ieee80211_ops - callbacks from mac80211 to the driver
@@ -1046,6 +1058,12 @@ enum ieee80211_erp_change_flags {
  *	used to determine whether to reply to Probe Requests.
  *
  * @conf_ht: Configures low level driver with 802.11n HT data. Must be atomic.
+ *
+ * @ampdu_action: Perform a certain A-MPDU action
+ * 	The RA/TID combination determines the destination and TID we want
+ * 	the ampdu action to be performed for. The action is defined through
+ * 	ieee80211_ampdu_mlme_action. Starting sequence number (@ssn)
+ * 	is the first frame we expect to perform the action on.
  */
 struct ieee80211_ops {
 	int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb,
@@ -1091,6 +1109,9 @@ struct ieee80211_ops {
 			     struct ieee80211_tx_control *control);
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
 	int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
+	int (*ampdu_action)(struct ieee80211_hw *hw,
+			    enum ieee80211_ampdu_mlme_action action,
+			    const u8 *ra, u16 tid, u16 ssn);
 };
 
 /**
-- 
cgit v1.2.3


From 07db218396650933abff3c5c1ad1e2a6e0cfedeb Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Tue, 25 Dec 2007 17:00:33 +0200
Subject: mac80211: A-MPDU Rx adding basic functionality

This patch adds the basic needed abilities and functions for A-MPDU Rx session
changed functions:
 - ieee80211_sta_process_addba_request - Rx A-MPDU initialization enabled
 - ieee80211_stop - stops all A-MPDU Rx in case interface goes down
added functions:
 - ieee80211_send_delba - used for sending out Del BA in A-MPDU sessions
 - ieee80211_sta_stop_rx_BA_session - stopping Rx A-MPDU session
 - sta_rx_agg_session_timer_expired - stops A-MPDU Rx use if load is too
low

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ieee80211.h    |   7 ++
 net/mac80211/ieee80211.c     |   9 ++
 net/mac80211/ieee80211_i.h   |   3 +
 net/mac80211/ieee80211_sta.c | 221 ++++++++++++++++++++++++++++++++++++++++---
 net/mac80211/sta_info.c      |   3 +
 5 files changed, 232 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 3e641590d0c8..4d5a4c9dcba7 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -472,6 +472,13 @@ enum ieee80211_back_actioncode {
 	WLAN_ACTION_DELBA = 2,
 };
 
+/* BACK (block-ack) parties */
+enum ieee80211_back_parties {
+	WLAN_BACK_RECIPIENT = 0,
+	WLAN_BACK_INITIATOR = 1,
+	WLAN_BACK_TIMER = 2,
+};
+
 /* A-MSDU 802.11n */
 #define IEEE80211_QOS_CONTROL_A_MSDU_PRESENT 0x0080
 
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 9c14e3d303c5..2011c726f2b1 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -292,9 +292,18 @@ static int ieee80211_stop(struct net_device *dev)
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_if_init_conf conf;
+	struct sta_info *sta;
+	int i;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
+	list_for_each_entry(sta, &local->sta_list, list) {
+		for (i = 0; i <  STA_TID_NUM; i++)
+			ieee80211_sta_stop_rx_ba_session(sta->dev, sta->addr,
+						i, WLAN_BACK_RECIPIENT,
+						WLAN_REASON_QSTA_LEAVE_QBSS);
+	}
+
 	netif_stop_queue(dev);
 
 	/*
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index baf53c047127..740d69d5bbd3 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -767,6 +767,9 @@ int ieee80211_ht_cap_ie_to_ht_info(struct ieee80211_ht_cap *ht_cap_ie,
 int ieee80211_ht_addt_info_ie_to_ht_bss_info(
 			struct ieee80211_ht_addt_info *ht_add_info_ie,
 			struct ieee80211_ht_bss_info *bss_info);
+void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *da,
+				u16 tid, u16 initiator, u16 reason);
+void sta_rx_agg_session_timer_expired(unsigned long data);
 /* ieee80211_iface.c */
 int ieee80211_if_add(struct net_device *dev, const char *name,
 		     struct net_device **new_dev, int type);
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 5b8f484c1673..d5a7683fab3a 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -64,6 +64,11 @@
 #define IEEE80211_ADDBA_PARAM_TID_MASK 0x003C
 #define IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK 0xFFA0
 
+/* next values represent the buffer size for A-MPDU frame.
+ * According to IEEE802.11n spec size varies from 8K to 64K (in powers of 2) */
+#define IEEE80211_MIN_AMPDU_BUF 0x8
+#define IEEE80211_MAX_AMPDU_BUF 0x40
+
 static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst,
 				     u8 *ssid, size_t ssid_len);
 static struct ieee80211_sta_bss *
@@ -1005,7 +1010,8 @@ static void ieee80211_send_addba_resp(struct net_device *dev, u8 *da, u16 tid,
 	struct ieee80211_mgmt *mgmt;
 	u16 capab;
 
-	skb = dev_alloc_skb(sizeof(*mgmt) + local->hw.extra_tx_headroom);
+	skb = dev_alloc_skb(sizeof(*mgmt) + local->hw.extra_tx_headroom + 1 +
+					sizeof(mgmt->u.action.u.addba_resp));
 	if (!skb) {
 		printk(KERN_DEBUG "%s: failed to allocate buffer "
 		       "for addba resp frame\n", dev->name);
@@ -1047,9 +1053,14 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev,
 						size_t len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_hw *hw = &local->hw;
+	struct ieee80211_conf *conf = &hw->conf;
 	struct sta_info *sta;
-	u16 capab, tid, timeout, ba_policy, buf_size, status;
+	struct tid_ampdu_rx *tid_agg_rx;
+	u16 capab, tid, timeout, ba_policy, buf_size, start_seq_num, status;
 	u8 dialog_token;
+	int ret = -EOPNOTSUPP;
+	DECLARE_MAC_BUF(mac);
 
 	sta = sta_info_get(local, mgmt->sa);
 	if (!sta)
@@ -1058,28 +1069,216 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev,
 	/* extract session parameters from addba request frame */
 	dialog_token = mgmt->u.action.u.addba_req.dialog_token;
 	timeout = le16_to_cpu(mgmt->u.action.u.addba_req.timeout);
+	start_seq_num =
+		le16_to_cpu(mgmt->u.action.u.addba_req.start_seq_num) >> 4;
 
 	capab = le16_to_cpu(mgmt->u.action.u.addba_req.capab);
 	ba_policy = (capab & IEEE80211_ADDBA_PARAM_POLICY_MASK) >> 1;
 	tid = (capab & IEEE80211_ADDBA_PARAM_TID_MASK) >> 2;
 	buf_size = (capab & IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK) >> 6;
 
-	/* TODO - currently aggregation is declined (A-MPDU add BA request
-	* acceptance is not obligatory by 802.11n draft), but here is
-	* the entry point for dealing with it */
-#ifdef MAC80211_HT_DEBUG
-	if (net_ratelimit())
-		printk(KERN_DEBUG "Add Block Ack request arrived,"
-				   " currently denying it\n");
-#endif /* MAC80211_HT_DEBUG */
-
 	status = WLAN_STATUS_REQUEST_DECLINED;
 
+	/* sanity check for incoming parameters:
+	 * check if configuration can support the BA policy
+	 * and if buffer size does not exceeds max value */
+	if (((ba_policy != 1)
+		&& (!(conf->ht_conf.cap & IEEE80211_HT_CAP_DELAY_BA)))
+		|| (buf_size > IEEE80211_MAX_AMPDU_BUF)) {
+		status = WLAN_STATUS_INVALID_QOS_PARAM;
+#ifdef CONFIG_MAC80211_HT_DEBUG
+		if (net_ratelimit())
+			printk(KERN_DEBUG "Block Ack Req with bad params from "
+				"%s on tid %u. policy %d, buffer size %d\n",
+				print_mac(mac, mgmt->sa), tid, ba_policy,
+				buf_size);
+#endif /* CONFIG_MAC80211_HT_DEBUG */
+		goto end_no_lock;
+	}
+	/* determine default buffer size */
+	if (buf_size == 0) {
+		struct ieee80211_hw_mode *mode = conf->mode;
+		buf_size = IEEE80211_MIN_AMPDU_BUF;
+		buf_size = buf_size << mode->ht_info.ampdu_factor;
+	}
+
+	tid_agg_rx = &sta->ampdu_mlme.tid_rx[tid];
+
+	/* examine state machine */
+	spin_lock_bh(&sta->ampdu_mlme.ampdu_rx);
+
+	if (tid_agg_rx->state != HT_AGG_STATE_IDLE) {
+#ifdef CONFIG_MAC80211_HT_DEBUG
+		if (net_ratelimit())
+			printk(KERN_DEBUG "unexpected Block Ack Req from "
+				"%s on tid %u\n",
+				print_mac(mac, mgmt->sa), tid);
+#endif /* CONFIG_MAC80211_HT_DEBUG */
+		goto end;
+	}
+
+	/* prepare reordering buffer */
+	tid_agg_rx->reorder_buf =
+		kmalloc(buf_size * sizeof(struct sk_buf *), GFP_ATOMIC);
+	if ((!tid_agg_rx->reorder_buf) && net_ratelimit()) {
+		printk(KERN_ERR "can not allocate reordering buffer "
+						"to tid %d\n", tid);
+		goto end;
+	}
+	memset(tid_agg_rx->reorder_buf, 0,
+		buf_size * sizeof(struct sk_buf *));
+
+	if (local->ops->ampdu_action)
+		ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_START,
+					       sta->addr, tid, start_seq_num);
+#ifdef CONFIG_MAC80211_HT_DEBUG
+	printk(KERN_DEBUG "Rx A-MPDU on tid %d result %d", tid, ret);
+#endif /* CONFIG_MAC80211_HT_DEBUG */
+
+	if (ret) {
+		kfree(tid_agg_rx->reorder_buf);
+		goto end;
+	}
+
+	/* change state and send addba resp */
+	tid_agg_rx->state = HT_AGG_STATE_OPERATIONAL;
+	tid_agg_rx->dialog_token = dialog_token;
+	tid_agg_rx->ssn = start_seq_num;
+	tid_agg_rx->head_seq_num = start_seq_num;
+	tid_agg_rx->buf_size = buf_size;
+	tid_agg_rx->timeout = timeout;
+	tid_agg_rx->stored_mpdu_num = 0;
+	status = WLAN_STATUS_SUCCESS;
+end:
+	spin_unlock_bh(&sta->ampdu_mlme.ampdu_rx);
+
+end_no_lock:
 	ieee80211_send_addba_resp(sta->dev, sta->addr, tid, dialog_token,
 				status, 1, buf_size, timeout);
 	sta_info_put(sta);
 }
 
+void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
+				u16 initiator, u16 reason_code)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_if_sta *ifsta = &sdata->u.sta;
+	struct sk_buff *skb;
+	struct ieee80211_mgmt *mgmt;
+	u16 params;
+
+	skb = dev_alloc_skb(sizeof(*mgmt) + local->hw.extra_tx_headroom + 1 +
+					sizeof(mgmt->u.action.u.delba));
+
+	if (!skb) {
+		printk(KERN_ERR "%s: failed to allocate buffer "
+					"for delba frame\n", dev->name);
+		return;
+	}
+
+	skb_reserve(skb, local->hw.extra_tx_headroom);
+	mgmt = (struct ieee80211_mgmt *) skb_put(skb, 24);
+	memset(mgmt, 0, 24);
+	memcpy(mgmt->da, da, ETH_ALEN);
+	memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN);
+	if (sdata->type == IEEE80211_IF_TYPE_AP)
+		memcpy(mgmt->bssid, dev->dev_addr, ETH_ALEN);
+	else
+		memcpy(mgmt->bssid, ifsta->bssid, ETH_ALEN);
+	mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
+					IEEE80211_STYPE_ACTION);
+
+	skb_put(skb, 1 + sizeof(mgmt->u.action.u.delba));
+
+	mgmt->u.action.category = WLAN_CATEGORY_BACK;
+	mgmt->u.action.u.delba.action_code = WLAN_ACTION_DELBA;
+	params = (u16)(initiator << 11); 	/* bit 11 initiator */
+	params |= (u16)(tid << 12); 		/* bit 15:12 TID number */
+
+	mgmt->u.action.u.delba.params = cpu_to_le16(params);
+	mgmt->u.action.u.delba.reason_code = cpu_to_le16(reason_code);
+
+	ieee80211_sta_tx(dev, skb, 0);
+}
+
+void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *ra, u16 tid,
+					u16 initiator, u16 reason)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_hw *hw = &local->hw;
+	struct sta_info *sta;
+	int ret;
+
+	sta = sta_info_get(local, ra);
+	if (!sta)
+		return;
+
+	/* check if TID is in operational state */
+	spin_lock_bh(&sta->ampdu_mlme.ampdu_rx);
+	if (sta->ampdu_mlme.tid_rx[tid].state
+				!= HT_AGG_STATE_OPERATIONAL) {
+		spin_unlock_bh(&sta->ampdu_mlme.ampdu_rx);
+		if (net_ratelimit())
+			printk(KERN_DEBUG "rx BA session requested to stop on "
+				"inactive tid %d\n", tid);
+		sta_info_put(sta);
+		return;
+	}
+	sta->ampdu_mlme.tid_rx[tid].state =
+		HT_AGG_STATE_REQ_STOP_BA_MSK |
+		(initiator << HT_AGG_STATE_INITIATOR_SHIFT);
+		spin_unlock_bh(&sta->ampdu_mlme.ampdu_rx);
+
+	/* stop HW Rx aggregation. ampdu_action existence
+	 * already verified in session init so we add the BUG_ON */
+	BUG_ON(!local->ops->ampdu_action);
+
+	ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_STOP,
+					ra, tid, EINVAL);
+	if (ret)
+		printk(KERN_DEBUG "HW problem - can not stop rx "
+				"aggergation for tid %d\n", tid);
+
+	/* shutdown timer has not expired */
+	if (initiator != WLAN_BACK_TIMER)
+		del_timer_sync(&sta->ampdu_mlme.tid_rx[tid].
+					session_timer);
+
+	/* check if this is a self generated aggregation halt */
+	if (initiator == WLAN_BACK_RECIPIENT || initiator == WLAN_BACK_TIMER)
+		ieee80211_send_delba(dev, ra, tid, 0, reason);
+
+	/* free the reordering buffer */
+	kfree(sta->ampdu_mlme.tid_rx[tid].reorder_buf);
+
+	sta->ampdu_mlme.tid_rx[tid].state = HT_AGG_STATE_IDLE;
+	sta_info_put(sta);
+}
+
+/*
+ * After receiving Block Ack Request (BAR) we activated a
+ * timer after each frame arrives from the originator.
+ * if this timer expires ieee80211_sta_stop_rx_ba_session will be executed.
+ */
+void sta_rx_agg_session_timer_expired(unsigned long data)
+{
+	/* not an elegant detour, but there is no choice as the timer passes
+	 * only one argument, and verious sta_info are needed here, so init
+	 * flow in sta_info_add gives the TID as data, while the timer_to_id
+	 * array gives the sta through container_of */
+	u8 *ptid = (u8 *)data;
+	u8 *timer_to_id = ptid - *ptid;
+	struct sta_info *sta = container_of(timer_to_id, struct sta_info,
+					 timer_to_tid[0]);
+
+	printk(KERN_DEBUG "rx session timer expired on tid %d\n", (u16)*ptid);
+	ieee80211_sta_stop_rx_ba_session(sta->dev, sta->addr, (u16)*ptid,
+					 WLAN_BACK_TIMER,
+					 WLAN_REASON_QSTA_TIMEOUT);
+}
+
+
 static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 				   struct ieee80211_if_sta *ifsta,
 				   struct ieee80211_mgmt *mgmt,
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index ffe8a49d8927..60ca07804056 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -104,6 +104,7 @@ static void sta_info_release(struct kref *kref)
 	struct sta_info *sta = container_of(kref, struct sta_info, kref);
 	struct ieee80211_local *local = sta->local;
 	struct sk_buff *skb;
+	int i;
 
 	/* free sta structure; it has already been removed from
 	 * hash table etc. external structures. Make sure that all
@@ -116,6 +117,8 @@ static void sta_info_release(struct kref *kref)
 	while ((skb = skb_dequeue(&sta->tx_filtered)) != NULL) {
 		dev_kfree_skb_any(skb);
 	}
+	for (i = 0; i <  STA_TID_NUM; i++)
+		del_timer_sync(&sta->ampdu_mlme.tid_rx[i].session_timer);
 	rate_control_free_sta(sta->rate_ctrl, sta->rate_ctrl_priv);
 	rate_control_put(sta->rate_ctrl);
 	kfree(sta);
-- 
cgit v1.2.3


From cdcb006fbe7a74b5f7827f5c5c27e11399a2fab7 Mon Sep 17 00:00:00 2001
From: Ivo van Doorn <ivdoorn@gmail.com>
Date: Mon, 7 Jan 2008 19:45:24 +0100
Subject: mac80211: Add radio led trigger

Some devices have a seperate LED which indicates if the radio is
enabled or not. This adds a LED trigger to mac80211 where drivers
can hook into when they are interested in radio status changes.

v2: Check hw.conf.radio_enabled when calling start().

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h         | 29 +++++++++++++++++++++++++++++
 net/mac80211/ieee80211.c       |  3 +++
 net/mac80211/ieee80211_i.h     |  5 +++--
 net/mac80211/ieee80211_ioctl.c |  2 ++
 net/mac80211/ieee80211_led.c   | 35 +++++++++++++++++++++++++++++++++++
 net/mac80211/ieee80211_led.h   |  6 ++++++
 6 files changed, 78 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3784a61a8489..be2a383b4776 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1143,6 +1143,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw);
 extern char *__ieee80211_get_tx_led_name(struct ieee80211_hw *hw);
 extern char *__ieee80211_get_rx_led_name(struct ieee80211_hw *hw);
 extern char *__ieee80211_get_assoc_led_name(struct ieee80211_hw *hw);
+extern char *__ieee80211_get_radio_led_name(struct ieee80211_hw *hw);
 #endif
 /**
  * ieee80211_get_tx_led_name - get name of TX LED
@@ -1182,6 +1183,16 @@ static inline char *ieee80211_get_rx_led_name(struct ieee80211_hw *hw)
 #endif
 }
 
+/**
+ * ieee80211_get_assoc_led_name - get name of association LED
+ *
+ * mac80211 creates a association LED trigger for each wireless hardware
+ * that can be used to drive LEDs if your driver registers a LED device.
+ * This function returns the name (or %NULL if not configured for LEDs)
+ * of the trigger so you can automatically link the LED device.
+ *
+ * @hw: the hardware to get the LED trigger name for
+ */
 static inline char *ieee80211_get_assoc_led_name(struct ieee80211_hw *hw)
 {
 #ifdef CONFIG_MAC80211_LEDS
@@ -1191,6 +1202,24 @@ static inline char *ieee80211_get_assoc_led_name(struct ieee80211_hw *hw)
 #endif
 }
 
+/**
+ * ieee80211_get_radio_led_name - get name of radio LED
+ *
+ * mac80211 creates a radio change LED trigger for each wireless hardware
+ * that can be used to drive LEDs if your driver registers a LED device.
+ * This function returns the name (or %NULL if not configured for LEDs)
+ * of the trigger so you can automatically link the LED device.
+ *
+ * @hw: the hardware to get the LED trigger name for
+ */
+static inline char *ieee80211_get_radio_led_name(struct ieee80211_hw *hw)
+{
+#ifdef CONFIG_MAC80211_LEDS
+	return __ieee80211_get_radio_led_name(hw);
+#else
+	return NULL;
+#endif
+}
 
 /* Register a new hardware PHYMODE capability to the stack. */
 int ieee80211_register_hwmode(struct ieee80211_hw *hw,
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 8ba69ae676ca..4807e5215a79 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -219,6 +219,7 @@ static int ieee80211_open(struct net_device *dev)
 		if (res)
 			return res;
 		ieee80211_hw_config(local);
+		ieee80211_led_radio(local, local->hw.conf.radio_enabled);
 	}
 
 	switch (sdata->type) {
@@ -392,6 +393,8 @@ static int ieee80211_stop(struct net_device *dev)
 		if (local->ops->stop)
 			local->ops->stop(local_to_hw(local));
 
+		ieee80211_led_radio(local, 0);
+
 		tasklet_disable(&local->tx_pending_tasklet);
 		tasklet_disable(&local->tasklet);
 	}
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 662e69ef7eed..b898b316bda1 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -500,8 +500,9 @@ struct ieee80211_local {
 
 #ifdef CONFIG_MAC80211_LEDS
 	int tx_led_counter, rx_led_counter;
-	struct led_trigger *tx_led, *rx_led, *assoc_led;
-	char tx_led_name[32], rx_led_name[32], assoc_led_name[32];
+	struct led_trigger *tx_led, *rx_led, *assoc_led, *radio_led;
+	char tx_led_name[32], rx_led_name[32],
+	     assoc_led_name[32], radio_led_name[32];
 #endif
 
 	u32 channel_use;
diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c
index 04ddce76135b..2604e21c05a8 100644
--- a/net/mac80211/ieee80211_ioctl.c
+++ b/net/mac80211/ieee80211_ioctl.c
@@ -21,6 +21,7 @@
 
 #include <net/mac80211.h>
 #include "ieee80211_i.h"
+#include "ieee80211_led.h"
 #include "ieee80211_rate.h"
 #include "wpa.h"
 #include "aes_ccm.h"
@@ -652,6 +653,7 @@ static int ieee80211_ioctl_siwtxpower(struct net_device *dev,
 	if (local->hw.conf.radio_enabled != !(data->txpower.disabled)) {
 		local->hw.conf.radio_enabled = !(data->txpower.disabled);
 		need_reconfig = 1;
+		ieee80211_led_radio(local, local->hw.conf.radio_enabled);
 	}
 
 	if (need_reconfig) {
diff --git a/net/mac80211/ieee80211_led.c b/net/mac80211/ieee80211_led.c
index 4cf89af9d100..f401484ab6d7 100644
--- a/net/mac80211/ieee80211_led.c
+++ b/net/mac80211/ieee80211_led.c
@@ -43,6 +43,16 @@ void ieee80211_led_assoc(struct ieee80211_local *local, bool associated)
 		led_trigger_event(local->assoc_led, LED_OFF);
 }
 
+void ieee80211_led_radio(struct ieee80211_local *local, bool enabled)
+{
+	if (unlikely(!local->radio_led))
+		return;
+	if (enabled)
+		led_trigger_event(local->radio_led, LED_FULL);
+	else
+		led_trigger_event(local->radio_led, LED_OFF);
+}
+
 void ieee80211_led_init(struct ieee80211_local *local)
 {
 	local->rx_led = kzalloc(sizeof(struct led_trigger), GFP_KERNEL);
@@ -77,10 +87,25 @@ void ieee80211_led_init(struct ieee80211_local *local)
 			local->assoc_led = NULL;
 		}
 	}
+
+	local->radio_led = kzalloc(sizeof(struct led_trigger), GFP_KERNEL);
+	if (local->radio_led) {
+		snprintf(local->radio_led_name, sizeof(local->radio_led_name),
+			 "%sradio", wiphy_name(local->hw.wiphy));
+		local->radio_led->name = local->radio_led_name;
+		if (led_trigger_register(local->radio_led)) {
+			kfree(local->radio_led);
+			local->radio_led = NULL;
+		}
+	}
 }
 
 void ieee80211_led_exit(struct ieee80211_local *local)
 {
+	if (local->radio_led) {
+		led_trigger_unregister(local->radio_led);
+		kfree(local->radio_led);
+	}
 	if (local->assoc_led) {
 		led_trigger_unregister(local->assoc_led);
 		kfree(local->assoc_led);
@@ -95,6 +120,16 @@ void ieee80211_led_exit(struct ieee80211_local *local)
 	}
 }
 
+char *__ieee80211_get_radio_led_name(struct ieee80211_hw *hw)
+{
+	struct ieee80211_local *local = hw_to_local(hw);
+
+	if (local->radio_led)
+		return local->radio_led_name;
+	return NULL;
+}
+EXPORT_SYMBOL(__ieee80211_get_radio_led_name);
+
 char *__ieee80211_get_assoc_led_name(struct ieee80211_hw *hw)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
diff --git a/net/mac80211/ieee80211_led.h b/net/mac80211/ieee80211_led.h
index 0feb22619835..77b1e1ba6039 100644
--- a/net/mac80211/ieee80211_led.h
+++ b/net/mac80211/ieee80211_led.h
@@ -16,6 +16,8 @@ extern void ieee80211_led_rx(struct ieee80211_local *local);
 extern void ieee80211_led_tx(struct ieee80211_local *local, int q);
 extern void ieee80211_led_assoc(struct ieee80211_local *local,
 				bool associated);
+extern void ieee80211_led_radio(struct ieee80211_local *local,
+				bool enabled);
 extern void ieee80211_led_init(struct ieee80211_local *local);
 extern void ieee80211_led_exit(struct ieee80211_local *local);
 #else
@@ -29,6 +31,10 @@ static inline void ieee80211_led_assoc(struct ieee80211_local *local,
 				       bool associated)
 {
 }
+static inline void ieee80211_led_radio(struct ieee80211_local *local,
+				       bool enabled)
+{
+}
 static inline void ieee80211_led_init(struct ieee80211_local *local)
 {
 }
-- 
cgit v1.2.3


From f0b5a0dcf125ce43855961ef4f965a91112bea23 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 8 Jan 2008 23:54:43 -0800
Subject: [VLAN]: Avoid expensive divides

We can avoid divides (as seen with CONFIG_CC_OPTIMIZE_FOR_SIZE=y on
x86) changing vlan_group_get_device()/vlan_group_set_device() id
parameter from signed to unsigned.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 976d4b1067d1..4562105fdb2b 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -81,14 +81,16 @@ struct vlan_group {
 	struct rcu_head		rcu;
 };
 
-static inline struct net_device *vlan_group_get_device(struct vlan_group *vg, int vlan_id)
+static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
+						       unsigned int vlan_id)
 {
 	struct net_device **array;
 	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
 	return array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN];
 }
 
-static inline void vlan_group_set_device(struct vlan_group *vg, int vlan_id,
+static inline void vlan_group_set_device(struct vlan_group *vg,
+					 unsigned int vlan_id,
 					 struct net_device *dev)
 {
 	struct net_device **array;
-- 
cgit v1.2.3


From 90754f8ec011c91382ece2849a43b8f8aa8a0105 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 12 Jan 2008 02:33:50 -0800
Subject: [IPVS]: Switch to using ctl_paths.

The feature of ipvs ctls is that the net/ipv4/vs path
is common for core ipvs ctls and for two schedulers,
so I make it exported and re-use it in modules.

Two other .c files required linux/sysctl.h to make the
extern declaration of this path compile well.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_vs.h         |  3 +++
 net/ipv4/ipvs/ip_vs_ctl.c   | 35 +++++++----------------------------
 net/ipv4/ipvs/ip_vs_est.c   |  1 +
 net/ipv4/ipvs/ip_vs_lblc.c  | 31 +------------------------------
 net/ipv4/ipvs/ip_vs_lblcr.c | 31 +------------------------------
 net/ipv4/ipvs/ip_vs_sched.c |  1 +
 6 files changed, 14 insertions(+), 88 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 3de6d1e6afa7..56f3c94ae620 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -9,6 +9,8 @@
 #include <asm/types.h>		/* For __uXX types */
 #include <linux/types.h>	/* For __beXX types in userland */
 
+#include <linux/sysctl.h>	/* For ctl_path */
+
 #define IP_VS_VERSION_CODE	0x010201
 #define NVERSION(version)			\
 	(version >> 16) & 0xFF,			\
@@ -854,6 +856,7 @@ extern int sysctl_ip_vs_expire_quiescent_template;
 extern int sysctl_ip_vs_sync_threshold[2];
 extern int sysctl_ip_vs_nat_icmp_send;
 extern struct ip_vs_stats ip_vs_stats;
+extern struct ctl_path net_vs_ctl_path[];
 
 extern struct ip_vs_service *
 ip_vs_service_get(__u32 fwmark, __u16 protocol, __be32 vaddr, __be16 vport);
diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
index 693d92490c11..9fecfe7d1164 100644
--- a/net/ipv4/ipvs/ip_vs_ctl.c
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -1591,34 +1591,13 @@ static struct ctl_table vs_vars[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table vs_table[] = {
-	{
-		.procname	= "vs",
-		.mode		= 0555,
-		.child		= vs_vars
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table ipvs_ipv4_table[] = {
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= vs_table,
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table vs_root_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= ipvs_ipv4_table,
-	},
-	{ .ctl_name = 0 }
+struct ctl_path net_vs_ctl_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
+	{ .procname = "vs", },
+	{ }
 };
+EXPORT_SYMBOL_GPL(net_vs_ctl_path);
 
 static struct ctl_table_header * sysctl_header;
 
@@ -2345,7 +2324,7 @@ int ip_vs_control_init(void)
 	proc_net_fops_create(&init_net, "ip_vs", 0, &ip_vs_info_fops);
 	proc_net_fops_create(&init_net, "ip_vs_stats",0, &ip_vs_stats_fops);
 
-	sysctl_header = register_sysctl_table(vs_root_table);
+	sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars);
 
 	/* Initialize ip_vs_svc_table, ip_vs_svc_fwm_table, ip_vs_rtable */
 	for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++)  {
diff --git a/net/ipv4/ipvs/ip_vs_est.c b/net/ipv4/ipvs/ip_vs_est.c
index efdd74e4fa28..dfa0d713c801 100644
--- a/net/ipv4/ipvs/ip_vs_est.c
+++ b/net/ipv4/ipvs/ip_vs_est.c
@@ -18,6 +18,7 @@
 #include <linux/slab.h>
 #include <linux/types.h>
 #include <linux/interrupt.h>
+#include <linux/sysctl.h>
 
 #include <net/ip_vs.h>
 
diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
index bf8c04a57548..3888642706ad 100644
--- a/net/ipv4/ipvs/ip_vs_lblc.c
+++ b/net/ipv4/ipvs/ip_vs_lblc.c
@@ -123,35 +123,6 @@ static ctl_table vs_vars_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table vs_table[] = {
-	{
-		.procname	= "vs",
-		.mode		= 0555,
-		.child		= vs_vars_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table ipvs_ipv4_table[] = {
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= vs_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table lblc_root_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= ipvs_ipv4_table
-	},
-	{ .ctl_name = 0 }
-};
-
 static struct ctl_table_header * sysctl_header;
 
 /*
@@ -582,7 +553,7 @@ static int __init ip_vs_lblc_init(void)
 	int ret;
 
 	INIT_LIST_HEAD(&ip_vs_lblc_scheduler.n_list);
-	sysctl_header = register_sysctl_table(lblc_root_table);
+	sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table);
 	ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler);
 	if (ret)
 		unregister_sysctl_table(sysctl_header);
diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
index f50da6411378..daa260eb21cf 100644
--- a/net/ipv4/ipvs/ip_vs_lblcr.c
+++ b/net/ipv4/ipvs/ip_vs_lblcr.c
@@ -311,35 +311,6 @@ static ctl_table vs_vars_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table vs_table[] = {
-	{
-		.procname	= "vs",
-		.mode		= 0555,
-		.child		= vs_vars_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table ipvs_ipv4_table[] = {
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= vs_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table lblcr_root_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= ipvs_ipv4_table
-	},
-	{ .ctl_name = 0 }
-};
-
 static struct ctl_table_header * sysctl_header;
 
 /*
@@ -771,7 +742,7 @@ static int __init ip_vs_lblcr_init(void)
 	int ret;
 
 	INIT_LIST_HEAD(&ip_vs_lblcr_scheduler.n_list);
-	sysctl_header = register_sysctl_table(lblcr_root_table);
+	sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table);
 	ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
 	if (ret)
 		unregister_sysctl_table(sysctl_header);
diff --git a/net/ipv4/ipvs/ip_vs_sched.c b/net/ipv4/ipvs/ip_vs_sched.c
index 432235861908..121a32b1b756 100644
--- a/net/ipv4/ipvs/ip_vs_sched.c
+++ b/net/ipv4/ipvs/ip_vs_sched.c
@@ -24,6 +24,7 @@
 #include <linux/interrupt.h>
 #include <asm/string.h>
 #include <linux/kmod.h>
+#include <linux/sysctl.h>
 
 #include <net/ip_vs.h>
 
-- 
cgit v1.2.3


From 3d7cc2ba628dcc6b55a2bafc7eaf35019fdcc201 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 9 Jan 2008 00:33:11 -0800
Subject: [NETFILTER]: Switch to using ctl_paths in nf_queue and conntrack
 modules

This includes the most simple cases for netfilter.

The first part is tne queue modules for ipv4 and ipv6,
on which the net/ipv4/ and net/ipv6/ paths are reused
from the appropriate ipv4 and ipv6 code.

The conntrack module is also patched, but this hunk is
very small and simple.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h                        |  2 ++
 include/net/ipv6.h                      |  2 ++
 net/ipv4/netfilter/ip_queue.c           | 23 ++---------------------
 net/ipv4/sysctl_net_ipv4.c              |  5 +++--
 net/ipv6/netfilter/ip6_queue.c          | 22 +---------------------
 net/ipv6/sysctl_net_ipv6.c              |  6 ++++--
 net/netfilter/nf_conntrack_standalone.c | 15 ++++++---------
 7 files changed, 20 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index cbeb97d38860..6850a80a6886 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -177,6 +177,8 @@ extern void inet_get_local_port_range(int *low, int *high);
 extern int sysctl_ip_default_ttl;
 extern int sysctl_ip_nonlocal_bind;
 
+extern struct ctl_path net_ipv4_ctl_path[];
+
 /* From ip_fragment.c */
 struct inet_frags_ctl;
 extern struct inet_frags_ctl ip4_frags_ctl;
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index f2adedff927f..e371f322017d 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -112,6 +112,8 @@ struct frag_hdr {
 extern int sysctl_ipv6_bindv6only;
 extern int sysctl_mld_max_msf;
 
+extern struct ctl_path net_ipv6_ctl_path[];
+
 #define _DEVINC(statname, modifier, idev, field)			\
 ({									\
 	struct inet6_dev *_idev = (idev);				\
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 68b12ce8ba55..7361315f20c6 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -29,6 +29,7 @@
 #include <net/sock.h>
 #include <net/route.h>
 #include <net/netfilter/nf_queue.h>
+#include <net/ip.h>
 
 #define IPQ_QMAX_DEFAULT 1024
 #define IPQ_PROC_FS_NAME "ip_queue"
@@ -525,26 +526,6 @@ static ctl_table ipq_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table ipq_dir_table[] = {
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= ipq_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table ipq_root_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= ipq_dir_table
-	},
-	{ .ctl_name = 0 }
-};
-
 static int ip_queue_show(struct seq_file *m, void *v)
 {
 	read_lock_bh(&queue_lock);
@@ -610,7 +591,7 @@ static int __init ip_queue_init(void)
 	}
 
 	register_netdevice_notifier(&ipq_dev_notifier);
-	ipq_sysctl_header = register_sysctl_table(ipq_root_table);
+	ipq_sysctl_header = register_sysctl_paths(net_ipv4_ctl_path, ipq_table);
 
 	status = nf_register_queue_handler(PF_INET, &nfqh);
 	if (status < 0) {
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index a5a9f8e3bb25..45536a91266a 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -846,17 +846,18 @@ static struct ctl_table ipv4_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static __initdata struct ctl_path net_ipv4_path[] = {
+struct ctl_path net_ipv4_ctl_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
 	{ },
 };
+EXPORT_SYMBOL_GPL(net_ipv4_ctl_path);
 
 static __init int sysctl_ipv4_init(void)
 {
 	struct ctl_table_header *hdr;
 
-	hdr = register_sysctl_paths(net_ipv4_path, ipv4_table);
+	hdr = register_sysctl_paths(net_ipv4_ctl_path, ipv4_table);
 	return hdr == NULL ? -ENOMEM : 0;
 }
 
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index e5b0059582f5..a20db0bb5a1f 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -529,26 +529,6 @@ static ctl_table ipq_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table ipq_dir_table[] = {
-	{
-		.ctl_name	= NET_IPV6,
-		.procname	= "ipv6",
-		.mode		= 0555,
-		.child		= ipq_table
-	},
-	{ .ctl_name = 0 }
-};
-
-static ctl_table ipq_root_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= ipq_dir_table
-	},
-	{ .ctl_name = 0 }
-};
-
 static int ip6_queue_show(struct seq_file *m, void *v)
 {
 	read_lock_bh(&queue_lock);
@@ -614,7 +594,7 @@ static int __init ip6_queue_init(void)
 	}
 
 	register_netdevice_notifier(&ipq_dev_notifier);
-	ipq_sysctl_header = register_sysctl_table(ipq_root_table);
+	ipq_sysctl_header = register_sysctl_paths(net_ipv6_ctl_path, ipq_table);
 
 	status = nf_register_queue_handler(PF_INET6, &nfqh);
 	if (status < 0) {
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 0b5bec3cb797..4ad8d9d3cb7a 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -82,17 +82,19 @@ static ctl_table ipv6_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static struct ctl_path ipv6_ctl_path[] = {
+struct ctl_path net_ipv6_ctl_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "ipv6", .ctl_name = NET_IPV6, },
 	{ },
 };
+EXPORT_SYMBOL_GPL(net_ipv6_ctl_path);
 
 static struct ctl_table_header *ipv6_sysctl_header;
 
 void ipv6_sysctl_register(void)
 {
-	ipv6_sysctl_header = register_sysctl_paths(ipv6_ctl_path, ipv6_table);
+	ipv6_sysctl_header = register_sysctl_paths(net_ipv6_ctl_path,
+			ipv6_table);
 }
 
 void ipv6_sysctl_unregister(void)
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 9efdd37fc195..2ad49331302b 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -383,15 +383,11 @@ static ctl_table nf_ct_netfilter_table[] = {
 	{ .ctl_name = 0 }
 };
 
-static ctl_table nf_ct_net_table[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= nf_ct_netfilter_table,
-	},
-	{ .ctl_name = 0 }
+struct ctl_path nf_ct_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ }
 };
+
 EXPORT_SYMBOL_GPL(nf_ct_log_invalid);
 #endif /* CONFIG_SYSCTL */
 
@@ -418,7 +414,8 @@ static int __init nf_conntrack_standalone_init(void)
 	proc_stat->owner = THIS_MODULE;
 #endif
 #ifdef CONFIG_SYSCTL
-	nf_ct_sysctl_header = register_sysctl_table(nf_ct_net_table);
+	nf_ct_sysctl_header = register_sysctl_paths(nf_ct_path,
+			nf_ct_netfilter_table);
 	if (nf_ct_sysctl_header == NULL) {
 		printk("nf_conntrack: can't register to sysctl.\n");
 		ret = -ENOMEM;
-- 
cgit v1.2.3


From b3fd3ffe39d830e7c24ef63b7f28703b485da2e3 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 9 Jan 2008 00:34:02 -0800
Subject: [NETFILTER]: Use the ctl paths instead of hand-made analogue

The conntracks subsystem has a similar infrastructure
to maintain ctl_paths, but since we already have it
on the generic level, I think it's OK to switch to
using it.

So, basically, this patch just replaces the ctl_table-s
with ctl_path-s, nf_register_sysctl_table with
register_sysctl_paths() and removes no longer needed code.

After this the net/netfilter/nf_sysctl.c file contains
the paths only.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                    |   8 +-
 include/net/netfilter/nf_conntrack_l3proto.h |   2 +-
 net/netfilter/nf_conntrack_proto.c           |   7 +-
 net/netfilter/nf_sysctl.c                    | 127 ++-------------------------
 4 files changed, 16 insertions(+), 128 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index d190d560de61..c41f6438095d 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -120,12 +120,8 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg);
 
 #ifdef CONFIG_SYSCTL
 /* Sysctl registration */
-struct ctl_table_header *nf_register_sysctl_table(struct ctl_table *path,
-						  struct ctl_table *table);
-void nf_unregister_sysctl_table(struct ctl_table_header *header,
-				struct ctl_table *table);
-extern struct ctl_table nf_net_netfilter_sysctl_path[];
-extern struct ctl_table nf_net_ipv4_netfilter_sysctl_path[];
+extern struct ctl_path nf_net_netfilter_sysctl_path[];
+extern struct ctl_path nf_net_ipv4_netfilter_sysctl_path[];
 #endif /* CONFIG_SYSCTL */
 
 extern struct list_head nf_hooks[NPROTO][NF_MAX_HOOKS];
diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h
index 15888fc7b72d..875c6d41eaab 100644
--- a/include/net/netfilter/nf_conntrack_l3proto.h
+++ b/include/net/netfilter/nf_conntrack_l3proto.h
@@ -73,7 +73,7 @@ struct nf_conntrack_l3proto
 
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*ctl_table_header;
-	struct ctl_table	*ctl_table_path;
+	struct ctl_path		*ctl_table_path;
 	struct ctl_table	*ctl_table;
 #endif /* CONFIG_SYSCTL */
 
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c
index 6d947068c58f..8595b5946acf 100644
--- a/net/netfilter/nf_conntrack_proto.c
+++ b/net/netfilter/nf_conntrack_proto.c
@@ -36,11 +36,11 @@ static DEFINE_MUTEX(nf_ct_proto_mutex);
 
 #ifdef CONFIG_SYSCTL
 static int
-nf_ct_register_sysctl(struct ctl_table_header **header, struct ctl_table *path,
+nf_ct_register_sysctl(struct ctl_table_header **header, struct ctl_path *path,
 		      struct ctl_table *table, unsigned int *users)
 {
 	if (*header == NULL) {
-		*header = nf_register_sysctl_table(path, table);
+		*header = register_sysctl_paths(path, table);
 		if (*header == NULL)
 			return -ENOMEM;
 	}
@@ -55,7 +55,8 @@ nf_ct_unregister_sysctl(struct ctl_table_header **header,
 {
 	if (users != NULL && --*users > 0)
 		return;
-	nf_unregister_sysctl_table(*header, table);
+
+	unregister_sysctl_table(*header);
 	*header = NULL;
 }
 #endif
diff --git a/net/netfilter/nf_sysctl.c b/net/netfilter/nf_sysctl.c
index ee34589e48a4..d9fcc893301d 100644
--- a/net/netfilter/nf_sysctl.c
+++ b/net/netfilter/nf_sysctl.c
@@ -7,128 +7,19 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 
-static void
-path_free(struct ctl_table *path, struct ctl_table *table)
-{
-	struct ctl_table *t, *next;
-
-	for (t = path; t != NULL && t != table; t = next) {
-		next = t->child;
-		kfree(t);
-	}
-}
-
-static struct ctl_table *
-path_dup(struct ctl_table *path, struct ctl_table *table)
-{
-	struct ctl_table *t, *last = NULL, *tmp;
-
-	for (t = path; t != NULL; t = t->child) {
-		/* twice the size since path elements are terminated by an
-		 * empty element */
-		tmp = kmemdup(t, 2 * sizeof(*t), GFP_KERNEL);
-		if (tmp == NULL) {
-			if (last != NULL)
-				path_free(path, table);
-			return NULL;
-		}
-
-		if (last != NULL)
-			last->child = tmp;
-		else
-			path = tmp;
-		last = tmp;
-	}
-
-	if (last != NULL)
-		last->child = table;
-	else
-		path = table;
-
-	return path;
-}
-
-struct ctl_table_header *
-nf_register_sysctl_table(struct ctl_table *path, struct ctl_table *table)
-{
-	struct ctl_table_header *header;
-
-	path = path_dup(path, table);
-	if (path == NULL)
-		return NULL;
-	header = register_sysctl_table(path);
-	if (header == NULL)
-		path_free(path, table);
-	return header;
-}
-EXPORT_SYMBOL_GPL(nf_register_sysctl_table);
-
-void
-nf_unregister_sysctl_table(struct ctl_table_header *header,
-			   struct ctl_table *table)
-{
-	struct ctl_table *path = header->ctl_table;
-
-	unregister_sysctl_table(header);
-	path_free(path, table);
-}
-EXPORT_SYMBOL_GPL(nf_unregister_sysctl_table);
-
 /* net/netfilter */
-static struct ctl_table nf_net_netfilter_table[] = {
-	{
-		.ctl_name	= NET_NETFILTER,
-		.procname	= "netfilter",
-		.mode		= 0555,
-	},
-	{
-		.ctl_name	= 0
-	}
-};
-struct ctl_table nf_net_netfilter_sysctl_path[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= nf_net_netfilter_table,
-	},
-	{
-		.ctl_name	= 0
-	}
+struct ctl_path nf_net_netfilter_sysctl_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "netfilter", .ctl_name = NET_NETFILTER, },
+	{ }
 };
 EXPORT_SYMBOL_GPL(nf_net_netfilter_sysctl_path);
 
 /* net/ipv4/netfilter */
-static struct ctl_table nf_net_ipv4_netfilter_table[] = {
-	{
-		.ctl_name	= NET_IPV4_NETFILTER,
-		.procname	= "netfilter",
-		.mode		= 0555,
-	},
-	{
-		.ctl_name	= 0
-	}
-};
-static struct ctl_table nf_net_ipv4_table[] = {
-	{
-		.ctl_name	= NET_IPV4,
-		.procname	= "ipv4",
-		.mode		= 0555,
-		.child		= nf_net_ipv4_netfilter_table,
-	},
-	{
-		.ctl_name	= 0
-	}
-};
-struct ctl_table nf_net_ipv4_netfilter_sysctl_path[] = {
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= nf_net_ipv4_table,
-	},
-	{
-		.ctl_name	= 0
-	}
+struct ctl_path nf_net_ipv4_netfilter_sysctl_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
+	{ .procname = "netfilter", .ctl_name = NET_IPV4_NETFILTER, },
+	{ }
 };
 EXPORT_SYMBOL_GPL(nf_net_ipv4_netfilter_sysctl_path);
-- 
cgit v1.2.3


From 291480c09a9452a3d8852a9bfeb5ba2cbcfe662c Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 02:47:55 -0800
Subject: [NETNS][IPV6]: Make ipv6_sysctl_register to return a value.

This patch makes the function ipv6_sysctl_register to return a
value. The af_inet6 init function is now able to handle an error and
catch it from the initialization of the sysctl.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h         | 2 +-
 net/ipv6/af_inet6.c        | 5 ++++-
 net/ipv6/sysctl_net_ipv6.c | 9 +++++++--
 3 files changed, 12 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e371f322017d..3e086f8bb449 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -622,7 +622,7 @@ static inline int snmp6_unregister_dev(struct inet6_dev *idev)
 extern ctl_table ipv6_route_table[];
 extern ctl_table ipv6_icmp_table[];
 
-extern void ipv6_sysctl_register(void);
+extern int ipv6_sysctl_register(void);
 extern void ipv6_sysctl_unregister(void);
 #endif
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 34c20533ba5d..a2842400a09e 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -783,7 +783,9 @@ static int __init inet6_init(void)
 	 */
 
 #ifdef CONFIG_SYSCTL
-	ipv6_sysctl_register();
+	err = ipv6_sysctl_register();
+	if (err)
+		goto sysctl_fail;
 #endif
 	err = icmpv6_init(&inet6_family_ops);
 	if (err)
@@ -897,6 +899,7 @@ ndisc_fail:
 icmp_fail:
 #ifdef CONFIG_SYSCTL
 	ipv6_sysctl_unregister();
+sysctl_fail:
 #endif
 	cleanup_ipv6_mibs();
 out_unregister_sock:
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 4ad8d9d3cb7a..f713fbf34c23 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -91,10 +91,15 @@ EXPORT_SYMBOL_GPL(net_ipv6_ctl_path);
 
 static struct ctl_table_header *ipv6_sysctl_header;
 
-void ipv6_sysctl_register(void)
+int ipv6_sysctl_register(void)
 {
 	ipv6_sysctl_header = register_sysctl_paths(net_ipv6_ctl_path,
-			ipv6_table);
+						   ipv6_table);
+	if (!ipv6_sysctl_header)
+		return -ENOMEM;
+
+	return 0;
+
 }
 
 void ipv6_sysctl_unregister(void)
-- 
cgit v1.2.3


From b0f159db7c95babf1e321334eb33d42db94f2854 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 02:49:06 -0800
Subject: [NETNS][IPV6]: Add ipv6 structure for netns.

Like the ipv4 part, this patch adds an ipv6 structure in the net
structure to aggregate the different resources to make ipv6 per
namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  4 ++++
 include/net/netns/ipv6.h    | 10 ++++++++++
 2 files changed, 14 insertions(+)
 create mode 100644 include/net/netns/ipv6.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index d04ddf2f80da..b0a5f904d252 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -11,6 +11,7 @@
 #include <net/netns/unix.h>
 #include <net/netns/packet.h>
 #include <net/netns/ipv4.h>
+#include <net/netns/ipv6.h>
 
 struct proc_dir_entry;
 struct net_device;
@@ -48,6 +49,9 @@ struct net {
 	struct netns_packet	packet;
 	struct netns_unix	unx;
 	struct netns_ipv4	ipv4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netns_ipv6	ipv6;
+#endif
 };
 
 #ifdef CONFIG_NET
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
new file mode 100644
index 000000000000..6f82046e3013
--- /dev/null
+++ b/include/net/netns/ipv6.h
@@ -0,0 +1,10 @@
+/*
+ * ipv6 in net namespaces
+ */
+
+#ifndef __NETNS_IPV6_H__
+#define __NETNS_IPV6_H__
+
+struct netns_ipv6 {
+};
+#endif
-- 
cgit v1.2.3


From 760f2d0186225f06d46e07232d65219c5055cad3 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 02:53:43 -0800
Subject: [NETNS][IPV6]: Make multiple instance of sysctl tables.

Each network namespace wants its own set of sysctl value, eg. we
should not be able from a namespace to set a sysctl value for another
namespace , especially for the initial network namespace.

This patch duplicates the sysctl table when we register a new network
namespace for ipv6. The duplicated table are postfixed with the
"template" word to notify the developper the table is cloned.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h         |  4 +--
 include/net/netns/ipv6.h   |  9 +++++++
 net/ipv6/icmp.c            | 12 ++++++++-
 net/ipv6/route.c           | 11 +++++++-
 net/ipv6/sysctl_net_ipv6.c | 67 +++++++++++++++++++++++++++++++++++++++-------
 5 files changed, 89 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 3e086f8bb449..5519035491e7 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -619,8 +619,8 @@ static inline int snmp6_unregister_dev(struct inet6_dev *idev)
 #endif
 
 #ifdef CONFIG_SYSCTL
-extern ctl_table ipv6_route_table[];
-extern ctl_table ipv6_icmp_table[];
+extern ctl_table ipv6_route_table_template[];
+extern ctl_table ipv6_icmp_table_template[];
 
 extern int ipv6_sysctl_register(void);
 extern void ipv6_sysctl_unregister(void);
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 6f82046e3013..58fbf30d9fdb 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -5,6 +5,15 @@
 #ifndef __NETNS_IPV6_H__
 #define __NETNS_IPV6_H__
 
+struct ctl_table_header;
+
+struct netns_sysctl_ipv6 {
+#ifdef CONFIG_SYSCTL
+	struct ctl_table_header *table;
+#endif
+};
+
 struct netns_ipv6 {
+	struct netns_sysctl_ipv6 sysctl;
 };
 #endif
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index c3bbd8687307..dfe3b37c43e9 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -907,7 +907,7 @@ int icmpv6_err_convert(int type, int code, int *err)
 EXPORT_SYMBOL(icmpv6_err_convert);
 
 #ifdef CONFIG_SYSCTL
-ctl_table ipv6_icmp_table[] = {
+ctl_table ipv6_icmp_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_ICMP_RATELIMIT,
 		.procname	= "ratelimit",
@@ -918,5 +918,15 @@ ctl_table ipv6_icmp_table[] = {
 	},
 	{ .ctl_name = 0 },
 };
+
+struct ctl_table *ipv6_icmp_sysctl_init(struct net *net)
+{
+	struct ctl_table *table;
+
+	table = kmemdup(ipv6_icmp_table_template,
+			sizeof(ipv6_icmp_table_template),
+			GFP_KERNEL);
+	return table;
+}
 #endif
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b80ef5784207..0c7382f4fb85 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2409,7 +2409,7 @@ int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp,
 		return -EINVAL;
 }
 
-ctl_table ipv6_route_table[] = {
+ctl_table ipv6_route_table_template[] = {
 	{
 		.procname	=	"flush",
 		.data		=	&flush_delay,
@@ -2499,6 +2499,15 @@ ctl_table ipv6_route_table[] = {
 	{ .ctl_name = 0 }
 };
 
+struct ctl_table *ipv6_route_sysctl_init(struct net *net)
+{
+	struct ctl_table *table;
+
+	table = kmemdup(ipv6_route_table_template,
+			sizeof(ipv6_route_table_template),
+			GFP_KERNEL);
+	return table;
+}
 #endif
 
 int __init ip6_route_init(void)
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 7329decf1f9d..7970f3366f87 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -14,20 +14,23 @@
 #include <net/addrconf.h>
 #include <net/inet_frag.h>
 
-static ctl_table ipv6_table[] = {
+extern struct ctl_table *ipv6_route_sysctl_init(struct net *net);
+extern struct ctl_table *ipv6_icmp_sysctl_init(struct net *net);
+
+static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_ROUTE,
 		.procname	= "route",
 		.maxlen		= 0,
 		.mode		= 0555,
-		.child		= ipv6_route_table
+		.child		= ipv6_route_table_template
 	},
 	{
 		.ctl_name	= NET_IPV6_ICMP,
 		.procname	= "icmp",
 		.maxlen		= 0,
 		.mode		= 0555,
-		.child		= ipv6_icmp_table
+		.child		= ipv6_icmp_table_template
 	},
 	{
 		.ctl_name	= NET_IPV6_BINDV6ONLY,
@@ -89,22 +92,66 @@ struct ctl_path net_ipv6_ctl_path[] = {
 };
 EXPORT_SYMBOL_GPL(net_ipv6_ctl_path);
 
-static struct ctl_table_header *ipv6_sysctl_header;
-
 static int ipv6_sysctl_net_init(struct net *net)
 {
-	ipv6_sysctl_header = register_net_sysctl_table(net, net_ipv6_ctl_path,
-						       ipv6_table);
-	if (!ipv6_sysctl_header)
+	struct ctl_table *ipv6_table;
+	struct ctl_table *ipv6_route_table;
+	struct ctl_table *ipv6_icmp_table;
+	int err;
+
+	err = -ENOMEM;
+	ipv6_table = kmemdup(ipv6_table_template, sizeof(ipv6_table_template),
+			     GFP_KERNEL);
+	if (!ipv6_table)
+		goto out;
+
+	ipv6_route_table = ipv6_route_sysctl_init(net);
+	if (!ipv6_route_table)
+		goto out_ipv6_table;
+
+	ipv6_icmp_table = ipv6_icmp_sysctl_init(net);
+	if (!ipv6_icmp_table)
+		goto out_ipv6_route_table;
+
+	ipv6_table[0].child = ipv6_route_table;
+	ipv6_table[1].child = ipv6_icmp_table;
+
+	net->ipv6.sysctl.table = register_net_sysctl_table(net, net_ipv6_ctl_path,
+							   ipv6_table);
+	if (!net->ipv6.sysctl.table)
 		return -ENOMEM;
 
-	return 0;
+	if (!net->ipv6.sysctl.table)
+		goto out_ipv6_icmp_table;
+
+	err = 0;
+out:
+	return err;
 
+out_ipv6_icmp_table:
+	kfree(ipv6_icmp_table);
+out_ipv6_route_table:
+	kfree(ipv6_route_table);
+out_ipv6_table:
+	kfree(ipv6_table);
+	goto out;
 }
 
 static void ipv6_sysctl_net_exit(struct net *net)
 {
-	unregister_net_sysctl_table(ipv6_sysctl_header);
+	struct ctl_table *ipv6_table;
+	struct ctl_table *ipv6_route_table;
+	struct ctl_table *ipv6_icmp_table;
+
+	ipv6_table = net->ipv6.sysctl.table->ctl_table_arg;
+	ipv6_route_table = ipv6_table[0].child;
+	ipv6_icmp_table = ipv6_table[1].child;
+
+	unregister_net_sysctl_table(net->ipv6.sysctl.table);
+
+	kfree(ipv6_table);
+	kfree(ipv6_route_table);
+	kfree(ipv6_icmp_table);
 }
 
 static struct pernet_operations ipv6_sysctl_net_ops = {
-- 
cgit v1.2.3


From 99bc9c4e45e7e783cf0b0a25cc03a103c038f254 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 02:54:53 -0800
Subject: [NETNS][IPV6]: Make bindv6only sysctl per namespace.

This patch moves the bindv6only sysctl to the network namespace
structure. Until the ipv6 protocol is not per namespace, the sysctl
variable is always from the initial network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h         | 1 -
 include/net/netns/ipv6.h   | 1 +
 net/ipv6/af_inet6.c        | 5 ++---
 net/ipv6/sysctl_net_ipv6.c | 4 +++-
 4 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5519035491e7..d03a4076e227 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -109,7 +109,6 @@ struct frag_hdr {
 #include <net/sock.h>
 
 /* sysctls */
-extern int sysctl_ipv6_bindv6only;
 extern int sysctl_mld_max_msf;
 
 extern struct ctl_path net_ipv6_ctl_path[];
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 58fbf30d9fdb..42b9b412fb87 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -11,6 +11,7 @@ struct netns_sysctl_ipv6 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header *table;
 #endif
+	int bindv6only;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 72b898fe2dde..70662bf8ab98 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -66,8 +66,6 @@ MODULE_AUTHOR("Cast of dozens");
 MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
 MODULE_LICENSE("GPL");
 
-int sysctl_ipv6_bindv6only __read_mostly;
-
 /* The inetsw6 table contains everything that inet6_create needs to
  * build a new socket.
  */
@@ -193,7 +191,7 @@ lookup_protocol:
 	np->mcast_hops	= -1;
 	np->mc_loop	= 1;
 	np->pmtudisc	= IPV6_PMTUDISC_WANT;
-	np->ipv6only	= sysctl_ipv6_bindv6only;
+	np->ipv6only	= init_net.ipv6.sysctl.bindv6only;
 
 	/* Init the ipv4 part of the socket since we can have sockets
 	 * using v6 API for ipv4.
@@ -721,6 +719,7 @@ static void cleanup_ipv6_mibs(void)
 
 static int inet6_net_init(struct net *net)
 {
+	net->ipv6.sysctl.bindv6only = 0;
 	return 0;
 }
 
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 7970f3366f87..13be97a928cb 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -35,7 +35,7 @@ static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_BINDV6ONLY,
 		.procname	= "bindv6only",
-		.data		= &sysctl_ipv6_bindv6only,
+		.data		= &init_net.ipv6.sysctl.bindv6only,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -116,6 +116,8 @@ static int ipv6_sysctl_net_init(struct net *net)
 	ipv6_table[0].child = ipv6_route_table;
 	ipv6_table[1].child = ipv6_icmp_table;
 
+	ipv6_table[2].data = &net->ipv6.sysctl.bindv6only;
+
 	net->ipv6.sysctl.table = register_net_sysctl_table(net, net_ipv6_ctl_path,
 							   ipv6_table);
 	if (!net->ipv6.sysctl.table)
-- 
cgit v1.2.3


From e71e0349eb32bc438fa80d8990c6f3592967d111 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 02:56:03 -0800
Subject: [NETNS][IPV6]: Make ip6_frags per namespace.

The ip6_frags is moved to the network namespace structure.  Because
there can be multiple instances of the network namespaces, and the
ip6_frags is no longer a global static variable, a helper function has
been added to facilitate the initialization of the variables.

Until the ipv6 protocol is not per namespace, the variables are
accessed relatively from the initial network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h         |  3 ---
 include/net/netns/ipv6.h   |  3 +++
 net/ipv6/af_inet6.c        |  8 ++++++++
 net/ipv6/reassembly.c      | 16 +++++++---------
 net/ipv6/sysctl_net_ipv6.c | 12 ++++++++----
 5 files changed, 26 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index d03a4076e227..c8e8cb241090 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -572,9 +572,6 @@ extern int inet6_hash_connect(struct inet_timewait_death_row *death_row,
 /*
  * reassembly.c
  */
-struct inet_frags_ctl;
-extern struct inet_frags_ctl ip6_frags_ctl;
-
 extern const struct proto_ops inet6_stream_ops;
 extern const struct proto_ops inet6_dgram_ops;
 
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 42b9b412fb87..ea4a71ac23d4 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -2,6 +2,8 @@
  * ipv6 in net namespaces
  */
 
+#include <net/inet_frag.h>
+
 #ifndef __NETNS_IPV6_H__
 #define __NETNS_IPV6_H__
 
@@ -11,6 +13,7 @@ struct netns_sysctl_ipv6 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header *table;
 #endif
+	struct inet_frags_ctl frags;
 	int bindv6only;
 };
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 70662bf8ab98..c4a1882fa80f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -72,6 +72,8 @@ MODULE_LICENSE("GPL");
 static struct list_head inetsw6[SOCK_MAX];
 static DEFINE_SPINLOCK(inetsw6_lock);
 
+void ipv6_frag_sysctl_init(struct net *net);
+
 static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk)
 {
 	const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo);
@@ -720,6 +722,12 @@ static void cleanup_ipv6_mibs(void)
 static int inet6_net_init(struct net *net)
 {
 	net->ipv6.sysctl.bindv6only = 0;
+	net->ipv6.sysctl.frags.high_thresh = 256 * 1024;
+	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
+	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
+	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
+	ipv6_frag_sysctl_init(net);
+
 	return 0;
 }
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index bf4173daecbb..5cd0bc693a5f 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -82,13 +82,6 @@ struct frag_queue
 	__u16			nhoffset;
 };
 
-struct inet_frags_ctl ip6_frags_ctl __read_mostly = {
-	.high_thresh 	 = 256 * 1024,
-	.low_thresh	 = 192 * 1024,
-	.timeout	 = IPV6_FRAG_TIMEOUT,
-	.secret_interval = 10 * 60 * HZ,
-};
-
 static struct inet_frags ip6_frags;
 
 int ip6_frag_nqueues(void)
@@ -605,7 +598,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 		return 1;
 	}
 
-	if (atomic_read(&ip6_frags.mem) > ip6_frags_ctl.high_thresh)
+	if (atomic_read(&ip6_frags.mem) > init_net.ipv6.sysctl.frags.high_thresh)
 		ip6_evictor(ip6_dst_idev(skb->dst));
 
 	if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr,
@@ -632,6 +625,11 @@ static struct inet6_protocol frag_protocol =
 	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
+void ipv6_frag_sysctl_init(struct net *net)
+{
+	ip6_frags.ctl = &net->ipv6.sysctl.frags;
+}
+
 int __init ipv6_frag_init(void)
 {
 	int ret;
@@ -639,7 +637,7 @@ int __init ipv6_frag_init(void)
 	ret = inet6_add_protocol(&frag_protocol, IPPROTO_FRAGMENT);
 	if (ret)
 		goto out;
-	ip6_frags.ctl = &ip6_frags_ctl;
+
 	ip6_frags.hashfn = ip6_hashfn;
 	ip6_frags.constructor = ip6_frag_init;
 	ip6_frags.destructor = NULL;
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 13be97a928cb..ae3cfd1b8e0e 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -43,7 +43,7 @@ static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_HIGH_THRESH,
 		.procname	= "ip6frag_high_thresh",
-		.data		= &ip6_frags_ctl.high_thresh,
+		.data		= &init_net.ipv6.sysctl.frags.high_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -51,7 +51,7 @@ static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_LOW_THRESH,
 		.procname	= "ip6frag_low_thresh",
-		.data		= &ip6_frags_ctl.low_thresh,
+		.data		= &init_net.ipv6.sysctl.frags.low_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -59,7 +59,7 @@ static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_TIME,
 		.procname	= "ip6frag_time",
-		.data		= &ip6_frags_ctl.timeout,
+		.data		= &init_net.ipv6.sysctl.frags.timeout,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -68,7 +68,7 @@ static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_SECRET_INTERVAL,
 		.procname	= "ip6frag_secret_interval",
-		.data		= &ip6_frags_ctl.secret_interval,
+		.data		= &init_net.ipv6.sysctl.frags.secret_interval,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -117,6 +117,10 @@ static int ipv6_sysctl_net_init(struct net *net)
 	ipv6_table[1].child = ipv6_icmp_table;
 
 	ipv6_table[2].data = &net->ipv6.sysctl.bindv6only;
+	ipv6_table[3].data = &net->ipv6.sysctl.frags.high_thresh;
+	ipv6_table[4].data = &net->ipv6.sysctl.frags.low_thresh;
+	ipv6_table[5].data = &net->ipv6.sysctl.frags.timeout;
+	ipv6_table[6].data = &net->ipv6.sysctl.frags.secret_interval;
 
 	net->ipv6.sysctl.table = register_net_sysctl_table(net, net_ipv6_ctl_path,
 							   ipv6_table);
-- 
cgit v1.2.3


From 4990509f19e8f1e000a83a88fc46328f73b8a88a Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 03:01:01 -0800
Subject: [NETNS][IPV6]: Make sysctls route per namespace.

All the sysctl concerning the routes are moved to the network
namespace structure. A helper function is called to initialize the
variables.

Because the ipv6 protocol is not yet per namespace, the variables are
accessed relatively from the network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h    |  2 --
 include/net/netns/ipv6.h   |  8 ++++++++
 net/ipv6/af_inet6.c        |  8 ++++++++
 net/ipv6/ip6_fib.c         | 14 ++++++++-----
 net/ipv6/route.c           | 49 +++++++++++++++++++---------------------------
 net/ipv6/sysctl_net_ipv6.c | 11 +++++++++++
 6 files changed, 56 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 1a0698b1e86c..faac0eee1ef3 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -43,8 +43,6 @@ extern struct rt6_info	ip6_prohibit_entry;
 extern struct rt6_info	ip6_blk_hole_entry;
 #endif
 
-extern int ip6_rt_gc_interval;
-
 extern void			ip6_route_input(struct sk_buff *skb);
 
 extern struct dst_entry *	ip6_route_output(struct sock *sk,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index ea4a71ac23d4..050d08809ef2 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -15,6 +15,14 @@ struct netns_sysctl_ipv6 {
 #endif
 	struct inet_frags_ctl frags;
 	int bindv6only;
+	int flush_delay;
+	int ip6_rt_max_size;
+	int ip6_rt_gc_min_interval;
+	int ip6_rt_gc_timeout;
+	int ip6_rt_gc_interval;
+	int ip6_rt_gc_elasticity;
+	int ip6_rt_mtu_expires;
+	int ip6_rt_min_advmss;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index c4a1882fa80f..3aea84a1822a 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -726,6 +726,14 @@ static int inet6_net_init(struct net *net)
 	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
 	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
 	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
+	net->ipv6.sysctl.flush_delay = 0;
+	net->ipv6.sysctl.ip6_rt_max_size = 4096;
+	net->ipv6.sysctl.ip6_rt_gc_min_interval = HZ / 2;
+	net->ipv6.sysctl.ip6_rt_gc_timeout = 60*HZ;
+	net->ipv6.sysctl.ip6_rt_gc_interval = 30*HZ;
+	net->ipv6.sysctl.ip6_rt_gc_elasticity = 9;
+	net->ipv6.sysctl.ip6_rt_mtu_expires = 10*60*HZ;
+	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
 	ipv6_frag_sysctl_init(net);
 
 	return 0;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 7165a5e90f45..0e83164aa3e6 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -681,13 +681,15 @@ static __inline__ void fib6_start_gc(struct rt6_info *rt)
 {
 	if (ip6_fib_timer.expires == 0 &&
 	    (rt->rt6i_flags & (RTF_EXPIRES|RTF_CACHE)))
-		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
+		mod_timer(&ip6_fib_timer, jiffies +
+			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
 }
 
 void fib6_force_start_gc(void)
 {
 	if (ip6_fib_timer.expires == 0)
-		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
+		mod_timer(&ip6_fib_timer, jiffies +
+			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
 }
 
 /*
@@ -1447,7 +1449,8 @@ void fib6_run_gc(unsigned long dummy)
 {
 	if (dummy != ~0UL) {
 		spin_lock_bh(&fib6_gc_lock);
-		gc_args.timeout = dummy ? (int)dummy : ip6_rt_gc_interval;
+		gc_args.timeout = dummy ? (int)dummy :
+			init_net.ipv6.sysctl.ip6_rt_gc_interval;
 	} else {
 		local_bh_disable();
 		if (!spin_trylock(&fib6_gc_lock)) {
@@ -1455,7 +1458,7 @@ void fib6_run_gc(unsigned long dummy)
 			local_bh_enable();
 			return;
 		}
-		gc_args.timeout = ip6_rt_gc_interval;
+		gc_args.timeout = init_net.ipv6.sysctl.ip6_rt_gc_interval;
 	}
 	gc_args.more = 0;
 
@@ -1463,7 +1466,8 @@ void fib6_run_gc(unsigned long dummy)
 	fib6_clean_all(fib6_age, 0, NULL);
 
 	if (gc_args.more)
-		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
+		mod_timer(&ip6_fib_timer, jiffies +
+			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
 	else {
 		del_timer(&ip6_fib_timer);
 		ip6_fib_timer.expires = 0;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 0c7382f4fb85..d2b3cf695aff 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -73,14 +73,6 @@
 
 #define CLONE_OFFLINK_ROUTE 0
 
-static int ip6_rt_max_size = 4096;
-static int ip6_rt_gc_min_interval = HZ / 2;
-static int ip6_rt_gc_timeout = 60*HZ;
-int ip6_rt_gc_interval = 30*HZ;
-static int ip6_rt_gc_elasticity = 9;
-static int ip6_rt_mtu_expires = 10*60*HZ;
-static int ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
-
 static struct rt6_info * ip6_rt_copy(struct rt6_info *ort);
 static struct dst_entry	*ip6_dst_check(struct dst_entry *dst, u32 cookie);
 static struct dst_entry *ip6_negative_advice(struct dst_entry *);
@@ -894,8 +886,8 @@ static inline unsigned int ipv6_advmss(unsigned int mtu)
 {
 	mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr);
 
-	if (mtu < ip6_rt_min_advmss)
-		mtu = ip6_rt_min_advmss;
+	if (mtu < init_net.ipv6.sysctl.ip6_rt_min_advmss)
+		mtu = init_net.ipv6.sysctl.ip6_rt_min_advmss;
 
 	/*
 	 * Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and
@@ -995,19 +987,19 @@ static int ip6_dst_gc(void)
 	static unsigned long last_gc;
 	unsigned long now = jiffies;
 
-	if (time_after(last_gc + ip6_rt_gc_min_interval, now) &&
-	    atomic_read(&ip6_dst_ops.entries) <= ip6_rt_max_size)
+	if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) &&
+	    atomic_read(&ip6_dst_ops.entries) <= init_net.ipv6.sysctl.ip6_rt_max_size)
 		goto out;
 
 	expire++;
 	fib6_run_gc(expire);
 	last_gc = now;
 	if (atomic_read(&ip6_dst_ops.entries) < ip6_dst_ops.gc_thresh)
-		expire = ip6_rt_gc_timeout>>1;
+		expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1;
 
 out:
-	expire -= expire>>ip6_rt_gc_elasticity;
-	return (atomic_read(&ip6_dst_ops.entries) > ip6_rt_max_size);
+	expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity;
+	return (atomic_read(&ip6_dst_ops.entries) > init_net.ipv6.sysctl.ip6_rt_max_size);
 }
 
 /* Clean host part of a prefix. Not necessary in radix tree,
@@ -1513,7 +1505,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		rt->u.dst.metrics[RTAX_MTU-1] = pmtu;
 		if (allfrag)
 			rt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
-		dst_set_expires(&rt->u.dst, ip6_rt_mtu_expires);
+		dst_set_expires(&rt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires);
 		rt->rt6i_flags |= RTF_MODIFIED|RTF_EXPIRES;
 		goto out;
 	}
@@ -1539,7 +1531,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		 * which is 10 mins. After 10 mins the decreased pmtu is expired
 		 * and detecting PMTU increase will be automatically happened.
 		 */
-		dst_set_expires(&nrt->u.dst, ip6_rt_mtu_expires);
+		dst_set_expires(&nrt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires);
 		nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES;
 
 		ip6_ins_rt(nrt);
@@ -2395,15 +2387,14 @@ static inline void ipv6_route_proc_fini(struct net *net)
 
 #ifdef CONFIG_SYSCTL
 
-static int flush_delay;
-
 static
 int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp,
 			      void __user *buffer, size_t *lenp, loff_t *ppos)
 {
+	int delay = init_net.ipv6.sysctl.flush_delay;
 	if (write) {
 		proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
-		fib6_run_gc(flush_delay <= 0 ? ~0UL : (unsigned long)flush_delay);
+		fib6_run_gc(delay <= 0 ? ~0UL : (unsigned long)delay);
 		return 0;
 	} else
 		return -EINVAL;
@@ -2412,7 +2403,7 @@ int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp,
 ctl_table ipv6_route_table_template[] = {
 	{
 		.procname	=	"flush",
-		.data		=	&flush_delay,
+		.data		=	&init_net.ipv6.sysctl.flush_delay,
 		.maxlen		=	sizeof(int),
 		.mode		=	0200,
 		.proc_handler	=	&ipv6_sysctl_rtcache_flush
@@ -2428,7 +2419,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_MAX_SIZE,
 		.procname	=	"max_size",
-		.data		=	&ip6_rt_max_size,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_max_size,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec,
@@ -2436,7 +2427,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_GC_MIN_INTERVAL,
 		.procname	=	"gc_min_interval",
-		.data		=	&ip6_rt_gc_min_interval,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2445,7 +2436,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_GC_TIMEOUT,
 		.procname	=	"gc_timeout",
-		.data		=	&ip6_rt_gc_timeout,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_gc_timeout,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2454,7 +2445,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_GC_INTERVAL,
 		.procname	=	"gc_interval",
-		.data		=	&ip6_rt_gc_interval,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_gc_interval,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2463,7 +2454,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_GC_ELASTICITY,
 		.procname	=	"gc_elasticity",
-		.data		=	&ip6_rt_gc_elasticity,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_gc_elasticity,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2472,7 +2463,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_MTU_EXPIRES,
 		.procname	=	"mtu_expires",
-		.data		=	&ip6_rt_mtu_expires,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_mtu_expires,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2481,7 +2472,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_MIN_ADVMSS,
 		.procname	=	"min_adv_mss",
-		.data		=	&ip6_rt_min_advmss,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_min_advmss,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_jiffies,
@@ -2490,7 +2481,7 @@ ctl_table ipv6_route_table_template[] = {
 	{
 		.ctl_name	=	NET_IPV6_ROUTE_GC_MIN_INTERVAL_MS,
 		.procname	=	"gc_min_interval_ms",
-		.data		=	&ip6_rt_gc_min_interval,
+		.data		=	&init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
 		.maxlen		=	sizeof(int),
 		.mode		=	0644,
 		.proc_handler	=	&proc_dointvec_ms_jiffies,
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index d223159638d3..b4ba422f2714 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -113,7 +113,18 @@ static int ipv6_sysctl_net_init(struct net *net)
 	if (!ipv6_icmp_table)
 		goto out_ipv6_route_table;
 
+	ipv6_route_table[0].data = &net->ipv6.sysctl.flush_delay;
+	/* ipv6_route_table[1].data will be handled when we have
+	   routes per namespace */
+	ipv6_route_table[2].data = &net->ipv6.sysctl.ip6_rt_max_size;
+	ipv6_route_table[3].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
+	ipv6_route_table[4].data = &net->ipv6.sysctl.ip6_rt_gc_timeout;
+	ipv6_route_table[5].data = &net->ipv6.sysctl.ip6_rt_gc_interval;
+	ipv6_route_table[6].data = &net->ipv6.sysctl.ip6_rt_gc_elasticity;
+	ipv6_route_table[7].data = &net->ipv6.sysctl.ip6_rt_mtu_expires;
+	ipv6_route_table[8].data = &net->ipv6.sysctl.ip6_rt_min_advmss;
 	ipv6_table[0].child = ipv6_route_table;
+
 	ipv6_table[1].child = ipv6_icmp_table;
 
 	ipv6_table[2].data = &net->ipv6.sysctl.bindv6only;
-- 
cgit v1.2.3


From 41a76906b3225997036efd88cbaae69d60b1e947 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 03:02:40 -0800
Subject: [NETNS][IPV6]: Make icmpv6_time sysctl per namespace.

This patch moves the icmpv6_time sysctl to the network namespace
structure.

Because the ipv6 protocol is not yet per namespace, the variable is
accessed relatively to the initial network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h   | 1 +
 net/ipv6/af_inet6.c        | 1 +
 net/ipv6/icmp.c            | 6 ++----
 net/ipv6/sysctl_net_ipv6.c | 1 +
 4 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 050d08809ef2..10733a6f1bd4 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -23,6 +23,7 @@ struct netns_sysctl_ipv6 {
 	int ip6_rt_gc_elasticity;
 	int ip6_rt_mtu_expires;
 	int ip6_rt_min_advmss;
+	int icmpv6_time;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3aea84a1822a..218b8b3050a1 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -734,6 +734,7 @@ static int inet6_net_init(struct net *net)
 	net->ipv6.sysctl.ip6_rt_gc_elasticity = 9;
 	net->ipv6.sysctl.ip6_rt_mtu_expires = 10*60*HZ;
 	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
+	net->ipv6.sysctl.icmpv6_time = 1*HZ;
 	ipv6_frag_sysctl_init(net);
 
 	return 0;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index dfe3b37c43e9..5395afe55ca5 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -154,8 +154,6 @@ static int is_ineligible(struct sk_buff *skb)
 	return 0;
 }
 
-static int sysctl_icmpv6_time __read_mostly = 1*HZ;
-
 /*
  * Check the ICMP output rate limit
  */
@@ -186,7 +184,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type,
 		res = 1;
 	} else {
 		struct rt6_info *rt = (struct rt6_info *)dst;
-		int tmo = sysctl_icmpv6_time;
+		int tmo = init_net.ipv6.sysctl.icmpv6_time;
 
 		/* Give more bandwidth to wider prefixes. */
 		if (rt->rt6i_dst.plen < 128)
@@ -911,7 +909,7 @@ ctl_table ipv6_icmp_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_ICMP_RATELIMIT,
 		.procname	= "ratelimit",
-		.data		= &sysctl_icmpv6_time,
+		.data		= &init_net.ipv6.sysctl.icmpv6_time,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index b4ba422f2714..5e0af4d46324 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -125,6 +125,7 @@ static int ipv6_sysctl_net_init(struct net *net)
 	ipv6_route_table[8].data = &net->ipv6.sysctl.ip6_rt_min_advmss;
 	ipv6_table[0].child = ipv6_route_table;
 
+	ipv6_icmp_table[0].data = &net->ipv6.sysctl.icmpv6_time;
 	ipv6_table[1].child = ipv6_icmp_table;
 
 	ipv6_table[2].data = &net->ipv6.sysctl.bindv6only;
-- 
cgit v1.2.3


From f8c26b8d589867aed8251db2935f8aa03aa68717 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:17:29 -0800
Subject: [NETNS]: Add netns parameter to fib_rules_(un)register.

The patch extends the different fib rules API in order to pass the
network namespace pointer. That will allow to access the different
tables from a namespace relative object. As usual, the pointer to the
init_net variable is passed as parameter so we don't break the
network.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 4 ++--
 net/core/fib_rules.c    | 4 ++--
 net/decnet/dn_rules.c   | 4 ++--
 net/ipv4/fib_rules.c    | 2 +-
 net/ipv6/fib6_rules.c   | 4 ++--
 5 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 2364db1a47e6..f7351b0ad552 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -101,8 +101,8 @@ static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla)
 	return frh->table;
 }
 
-extern int			fib_rules_register(struct fib_rules_ops *);
-extern int			fib_rules_unregister(struct fib_rules_ops *);
+extern int fib_rules_register(struct net *, struct fib_rules_ops *);
+extern int fib_rules_unregister(struct net *, struct fib_rules_ops *);
 extern void                     fib_rules_cleanup_ops(struct fib_rules_ops *);
 
 extern int			fib_rules_lookup(struct fib_rules_ops *,
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index fcbf41c0a5d4..ada9c8155e3c 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -74,7 +74,7 @@ static void flush_route_cache(struct fib_rules_ops *ops)
 		ops->flush_cache();
 }
 
-int fib_rules_register(struct fib_rules_ops *ops)
+int fib_rules_register(struct net *net, struct fib_rules_ops *ops)
 {
 	int err = -EEXIST;
 	struct fib_rules_ops *o;
@@ -113,7 +113,7 @@ void fib_rules_cleanup_ops(struct fib_rules_ops *ops)
 }
 EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops);
 
-int fib_rules_unregister(struct fib_rules_ops *ops)
+int fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
 {
 	int err = 0;
 	struct fib_rules_ops *o;
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index ffebea04cc99..0b5e2b97e15a 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -255,12 +255,12 @@ void __init dn_fib_rules_init(void)
 {
 	BUG_ON(fib_default_rule_add(&dn_fib_rules_ops, 0x7fff,
 			            RT_TABLE_MAIN, 0));
-	fib_rules_register(&dn_fib_rules_ops);
+	fib_rules_register(&init_net, &dn_fib_rules_ops);
 }
 
 void __exit dn_fib_rules_cleanup(void)
 {
-	fib_rules_unregister(&dn_fib_rules_ops);
+	fib_rules_unregister(&init_net, &dn_fib_rules_ops);
 }
 
 
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index a0ada3a8d8dd..eac3f717ecab 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -314,5 +314,5 @@ static int __init fib_default_rules_init(void)
 void __init fib4_rules_init(void)
 {
 	BUG_ON(fib_default_rules_init());
-	fib_rules_register(&fib4_rules_ops);
+	fib_rules_register(&init_net, &fib4_rules_ops);
 }
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 9ce2e0a6748a..e4d7e5a08b40 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -273,7 +273,7 @@ int __init fib6_rules_init(void)
 	if (ret)
 		goto out;
 
-	ret = fib_rules_register(&fib6_rules_ops);
+	ret = fib_rules_register(&init_net, &fib6_rules_ops);
 	if (ret)
 		goto out_default_rules_init;
 out:
@@ -286,5 +286,5 @@ out_default_rules_init:
 
 void fib6_rules_cleanup(void)
 {
-	fib_rules_unregister(&fib6_rules_ops);
+	fib_rules_unregister(&init_net, &fib6_rules_ops);
 }
-- 
cgit v1.2.3


From 868d13ac811746e28e4c806f2b1bd8575796f9af Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:18:25 -0800
Subject: [NETNS]: Pass fib_rules_ops into default_pref method.

fib_rules_ops contains operations and the list of configured rules. ops will
become per/namespace soon, so we need them to be known in the default_pref
callback.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 2 +-
 net/core/fib_rules.c    | 2 +-
 net/decnet/dn_rules.c   | 2 +-
 net/ipv4/fib_rules.c    | 2 +-
 net/ipv6/fib6_rules.c   | 2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index f7351b0ad552..e9a074cf59ed 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -56,7 +56,7 @@ struct fib_rules_ops
 	int			(*fill)(struct fib_rule *, struct sk_buff *,
 					struct nlmsghdr *,
 					struct fib_rule_hdr *);
-	u32			(*default_pref)(void);
+	u32			(*default_pref)(struct fib_rules_ops *ops);
 	size_t			(*nlmsg_payload)(struct fib_rule *);
 
 	/* Called after modifications to the rules set, must flush
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index ada9c8155e3c..e12e9f583949 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -285,7 +285,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	rule->table = frh_get_table(frh, tb);
 
 	if (!rule->pref && ops->default_pref)
-		rule->pref = ops->default_pref();
+		rule->pref = ops->default_pref(ops);
 
 	err = -EINVAL;
 	if (tb[FRA_GOTO]) {
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 0b5e2b97e15a..c1fae23d2263 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -212,7 +212,7 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-static u32 dn_fib_rule_default_pref(void)
+static u32 dn_fib_rule_default_pref(struct fib_rules_ops *ops)
 {
 	struct list_head *pos;
 	struct fib_rule *rule;
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index eac3f717ecab..afe669dd1bd5 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -245,7 +245,7 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-static u32 fib4_rule_default_pref(void)
+static u32 fib4_rule_default_pref(struct fib_rules_ops *ops)
 {
 	struct list_head *pos;
 	struct fib_rule *rule;
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index e4d7e5a08b40..76437a1fcab0 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -223,7 +223,7 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-static u32 fib6_rule_default_pref(void)
+static u32 fib6_rule_default_pref(struct fib_rules_ops *ops)
 {
 	return 0x3FFF;
 }
-- 
cgit v1.2.3


From 5fd30ee7c48bf7f9cd16ab44c8a09fa4a57cc21d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:20:28 -0800
Subject: [NETNS]: Namespacing in the generic fib rules code.

Move static rules_ops & rules_mod_lock to the struct net, register the
pernet subsys to init them and enjoy the fact that the core rules
infrastructure works in the namespace.

Real IPv4 fib rules virtualization requires fib tables support in the
namespace and will be done seriously later in the patchset.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  4 ++
 net/core/fib_rules.c        | 91 +++++++++++++++++++++++++++------------------
 2 files changed, 58 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index b0a5f904d252..b8c1d60ba9e4 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -40,6 +40,10 @@ struct net {
 	struct hlist_head 	*dev_name_head;
 	struct hlist_head	*dev_index_head;
 
+	/* core fib_rules */
+	struct list_head	rules_ops;
+	spinlock_t		rules_mod_lock;
+
 	struct sock 		*rtnl;			/* rtnetlink socket */
 
 	/* core sysctls */
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index e12e9f583949..c5f78fed6885 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -15,9 +15,6 @@
 #include <net/sock.h>
 #include <net/fib_rules.h>
 
-static LIST_HEAD(rules_ops);
-static DEFINE_SPINLOCK(rules_mod_lock);
-
 int fib_default_rule_add(struct fib_rules_ops *ops,
 			 u32 pref, u32 table, u32 flags)
 {
@@ -40,16 +37,17 @@ int fib_default_rule_add(struct fib_rules_ops *ops,
 }
 EXPORT_SYMBOL(fib_default_rule_add);
 
-static void notify_rule_change(int event, struct fib_rule *rule,
+static void notify_rule_change(struct net *net, int event,
+			       struct fib_rule *rule,
 			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
 			       u32 pid);
 
-static struct fib_rules_ops *lookup_rules_ops(int family)
+static struct fib_rules_ops *lookup_rules_ops(struct net *net, int family)
 {
 	struct fib_rules_ops *ops;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(ops, &rules_ops, list) {
+	list_for_each_entry_rcu(ops, &net->rules_ops, list) {
 		if (ops->family == family) {
 			if (!try_module_get(ops->owner))
 				ops = NULL;
@@ -87,15 +85,16 @@ int fib_rules_register(struct net *net, struct fib_rules_ops *ops)
 	    ops->action == NULL)
 		return -EINVAL;
 
-	spin_lock(&rules_mod_lock);
-	list_for_each_entry(o, &rules_ops, list)
+	spin_lock(&net->rules_mod_lock);
+	list_for_each_entry(o, &net->rules_ops, list)
 		if (ops->family == o->family)
 			goto errout;
 
-	list_add_tail_rcu(&ops->list, &rules_ops);
+	hold_net(net);
+	list_add_tail_rcu(&ops->list, &net->rules_ops);
 	err = 0;
 errout:
-	spin_unlock(&rules_mod_lock);
+	spin_unlock(&net->rules_mod_lock);
 
 	return err;
 }
@@ -118,8 +117,8 @@ int fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
 	int err = 0;
 	struct fib_rules_ops *o;
 
-	spin_lock(&rules_mod_lock);
-	list_for_each_entry(o, &rules_ops, list) {
+	spin_lock(&net->rules_mod_lock);
+	list_for_each_entry(o, &net->rules_ops, list) {
 		if (o == ops) {
 			list_del_rcu(&o->list);
 			fib_rules_cleanup_ops(ops);
@@ -129,9 +128,11 @@ int fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
 
 	err = -ENOENT;
 out:
-	spin_unlock(&rules_mod_lock);
+	spin_unlock(&net->rules_mod_lock);
 
 	synchronize_rcu();
+	if (!err)
+		release_net(net);
 
 	return err;
 }
@@ -229,13 +230,10 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	struct nlattr *tb[FRA_MAX+1];
 	int err = -EINVAL, unresolved = 0;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
 		goto errout;
 
-	ops = lookup_rules_ops(frh->family);
+	ops = lookup_rules_ops(net, frh->family);
 	if (ops == NULL) {
 		err = EAFNOSUPPORT;
 		goto errout;
@@ -348,7 +346,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	else
 		list_add_rcu(&rule->list, &ops->rules_list);
 
-	notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
+	notify_rule_change(net, RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
 	flush_route_cache(ops);
 	rules_ops_put(ops);
 	return 0;
@@ -369,13 +367,10 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	struct nlattr *tb[FRA_MAX+1];
 	int err = -EINVAL;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
 		goto errout;
 
-	ops = lookup_rules_ops(frh->family);
+	ops = lookup_rules_ops(net, frh->family);
 	if (ops == NULL) {
 		err = EAFNOSUPPORT;
 		goto errout;
@@ -441,7 +436,7 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		}
 
 		synchronize_rcu();
-		notify_rule_change(RTM_DELRULE, rule, ops, nlh,
+		notify_rule_change(net, RTM_DELRULE, rule, ops, nlh,
 				   NETLINK_CB(skb).pid);
 		fib_rule_put(rule);
 		flush_route_cache(ops);
@@ -551,13 +546,10 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb)
 	struct fib_rules_ops *ops;
 	int idx = 0, family;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	family = rtnl_msg_family(cb->nlh);
 	if (family != AF_UNSPEC) {
 		/* Protocol specific dump request */
-		ops = lookup_rules_ops(family);
+		ops = lookup_rules_ops(net, family);
 		if (ops == NULL)
 			return -EAFNOSUPPORT;
 
@@ -565,7 +557,7 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb)
 	}
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(ops, &rules_ops, list) {
+	list_for_each_entry_rcu(ops, &net->rules_ops, list) {
 		if (idx < cb->args[0] || !try_module_get(ops->owner))
 			goto skip;
 
@@ -582,7 +574,7 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
-static void notify_rule_change(int event, struct fib_rule *rule,
+static void notify_rule_change(struct net *net, int event, struct fib_rule *rule,
 			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
 			       u32 pid)
 {
@@ -600,10 +592,10 @@ static void notify_rule_change(int event, struct fib_rule *rule,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, &init_net, pid, ops->nlgroup, nlh, GFP_KERNEL);
+	err = rtnl_notify(skb, net, pid, ops->nlgroup, nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(&init_net, ops->nlgroup, err);
+		rtnl_set_sk_err(net, ops->nlgroup, err);
 }
 
 static void attach_rules(struct list_head *rules, struct net_device *dev)
@@ -631,22 +623,20 @@ static int fib_rules_event(struct notifier_block *this, unsigned long event,
 			    void *ptr)
 {
 	struct net_device *dev = ptr;
+	struct net *net = dev->nd_net;
 	struct fib_rules_ops *ops;
 
-	if (dev->nd_net != &init_net)
-		return NOTIFY_DONE;
-
 	ASSERT_RTNL();
 	rcu_read_lock();
 
 	switch (event) {
 	case NETDEV_REGISTER:
-		list_for_each_entry(ops, &rules_ops, list)
+		list_for_each_entry(ops, &net->rules_ops, list)
 			attach_rules(&ops->rules_list, dev);
 		break;
 
 	case NETDEV_UNREGISTER:
-		list_for_each_entry(ops, &rules_ops, list)
+		list_for_each_entry(ops, &net->rules_ops, list)
 			detach_rules(&ops->rules_list, dev);
 		break;
 	}
@@ -660,13 +650,40 @@ static struct notifier_block fib_rules_notifier = {
 	.notifier_call = fib_rules_event,
 };
 
+static int fib_rules_net_init(struct net *net)
+{
+	INIT_LIST_HEAD(&net->rules_ops);
+	spin_lock_init(&net->rules_mod_lock);
+	return 0;
+}
+
+static struct pernet_operations fib_rules_net_ops = {
+	.init = fib_rules_net_init,
+};
+
 static int __init fib_rules_init(void)
 {
+	int err;
 	rtnl_register(PF_UNSPEC, RTM_NEWRULE, fib_nl_newrule, NULL);
 	rtnl_register(PF_UNSPEC, RTM_DELRULE, fib_nl_delrule, NULL);
 	rtnl_register(PF_UNSPEC, RTM_GETRULE, NULL, fib_nl_dumprule);
 
-	return register_netdevice_notifier(&fib_rules_notifier);
+	err = register_netdevice_notifier(&fib_rules_notifier);
+	if (err < 0)
+		goto fail;
+
+	err = register_pernet_subsys(&fib_rules_net_ops);
+	if (err < 0)
+		goto fail_unregister;
+	return 0;
+
+fail_unregister:
+	unregister_netdevice_notifier(&fib_rules_notifier);
+fail:
+	rtnl_unregister(PF_UNSPEC, RTM_NEWRULE);
+	rtnl_unregister(PF_UNSPEC, RTM_DELRULE);
+	rtnl_unregister(PF_UNSPEC, RTM_GETRULE);
+	return err;
 }
 
 subsys_initcall(fib_rules_init);
-- 
cgit v1.2.3


From 61a0265344786a548e8a0b26cb668e78a71f9602 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:21:09 -0800
Subject: [NETNS]: Add namespace to API for routing /proc entries creation.

This adds netns parameter to fib_proc_init/exit and replaces __init
specifier with __net_init. After this, we will not yet have these proc
files show info from the specific namespace - this will be done when
these tables become namespaced.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h |  4 ++--
 net/ipv4/af_inet.c   |  4 ++--
 net/ipv4/fib_hash.c  |  8 ++++----
 net/ipv4/fib_trie.c  | 21 +++++++++++----------
 4 files changed, 19 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index d70b9b49f8b1..f74cbb21af45 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -253,8 +253,8 @@ static inline void fib_res_put(struct fib_result *res)
 }
 
 #ifdef CONFIG_PROC_FS
-extern int  fib_proc_init(void);
-extern void fib_proc_exit(void);
+extern int __net_init  fib_proc_init(struct net *net);
+extern void __net_exit fib_proc_exit(struct net *net);
 #endif
 
 #endif  /* _NET_FIB_H */
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 0e12cf646071..d5b8cb1dad67 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1475,14 +1475,14 @@ static int __init ipv4_proc_init(void)
 		goto out_tcp;
 	if (udp4_proc_init())
 		goto out_udp;
-	if (fib_proc_init())
+	if (fib_proc_init(&init_net))
 		goto out_fib;
 	if (ip_misc_proc_init())
 		goto out_misc;
 out:
 	return rc;
 out_misc:
-	fib_proc_exit();
+	fib_proc_exit(&init_net);
 out_fib:
 	udp4_proc_exit();
 out_udp:
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index ee1ffdb3044f..8de21bc4947a 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -1041,15 +1041,15 @@ static const struct file_operations fib_seq_fops = {
 	.release	= seq_release_private,
 };
 
-int __init fib_proc_init(void)
+int __net_init fib_proc_init(struct net *net)
 {
-	if (!proc_net_fops_create(&init_net, "route", S_IRUGO, &fib_seq_fops))
+	if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_seq_fops))
 		return -ENOMEM;
 	return 0;
 }
 
-void __init fib_proc_exit(void)
+void __net_exit fib_proc_exit(struct net *net)
 {
-	proc_net_remove(&init_net, "route");
+	proc_net_remove(net, "route");
 }
 #endif /* CONFIG_PROC_FS */
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index a842204df7ba..4bbfaeedf210 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2508,32 +2508,33 @@ static const struct file_operations fib_route_fops = {
 	.release = seq_release_private,
 };
 
-int __init fib_proc_init(void)
+int __net_init fib_proc_init(struct net *net)
 {
-	if (!proc_net_fops_create(&init_net, "fib_trie", S_IRUGO, &fib_trie_fops))
+	if (!proc_net_fops_create(net, "fib_trie", S_IRUGO, &fib_trie_fops))
 		goto out1;
 
-	if (!proc_net_fops_create(&init_net, "fib_triestat", S_IRUGO, &fib_triestat_fops))
+	if (!proc_net_fops_create(net, "fib_triestat", S_IRUGO,
+				  &fib_triestat_fops))
 		goto out2;
 
-	if (!proc_net_fops_create(&init_net, "route", S_IRUGO, &fib_route_fops))
+	if (!proc_net_fops_create(net, "route", S_IRUGO, &fib_route_fops))
 		goto out3;
 
 	return 0;
 
 out3:
-	proc_net_remove(&init_net, "fib_triestat");
+	proc_net_remove(net, "fib_triestat");
 out2:
-	proc_net_remove(&init_net, "fib_trie");
+	proc_net_remove(net, "fib_trie");
 out1:
 	return -ENOMEM;
 }
 
-void __init fib_proc_exit(void)
+void __net_exit fib_proc_exit(struct net *net)
 {
-	proc_net_remove(&init_net, "fib_trie");
-	proc_net_remove(&init_net, "fib_triestat");
-	proc_net_remove(&init_net, "route");
+	proc_net_remove(net, "fib_trie");
+	proc_net_remove(net, "fib_triestat");
+	proc_net_remove(net, "route");
 }
 
 #endif /* CONFIG_PROC_FS */
-- 
cgit v1.2.3


From dbb50165b512f6c9b7aae10af73ae5b6d811f4d0 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:21:49 -0800
Subject: [IPV4]: Check fib4_rules_init failure.

This adds error paths into both versions of fib4_rules_init
(with/without CONFIG_IP_MULTIPLE_TABLES) and returns error code to the
caller.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  2 +-
 net/ipv4/fib_frontend.c | 18 +++++++++++++++---
 net/ipv4/fib_rules.c    | 14 ++++++++++++--
 3 files changed, 28 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index f74cbb21af45..cbff18d748dd 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -186,7 +186,7 @@ static inline void fib_select_default(const struct flowi *flp, struct fib_result
 }
 
 #else /* CONFIG_IP_MULTIPLE_TABLES */
-extern void __init fib4_rules_init(void);
+extern int __init fib4_rules_init(void);
 
 #ifdef CONFIG_NET_CLS_ROUTE
 extern u32 fib_rules_tclass(struct fib_result *res);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index ac6238a3b0fd..1bb956b3be28 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -59,12 +59,24 @@ struct fib_table *ip_fib_main_table;
 #define FIB_TABLE_HASHSZ 1
 static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
-static void __init fib4_rules_init(void)
+static int __init fib4_rules_init(void)
 {
 	ip_fib_local_table = fib_hash_init(RT_TABLE_LOCAL);
-	hlist_add_head_rcu(&ip_fib_local_table->tb_hlist, &fib_table_hash[0]);
+	if (ip_fib_local_table == NULL)
+		return -ENOMEM;
+
 	ip_fib_main_table  = fib_hash_init(RT_TABLE_MAIN);
+	if (ip_fib_main_table == NULL)
+		goto fail;
+
+	hlist_add_head_rcu(&ip_fib_local_table->tb_hlist, &fib_table_hash[0]);
 	hlist_add_head_rcu(&ip_fib_main_table->tb_hlist, &fib_table_hash[0]);
+	return 0;
+
+fail:
+	kfree(ip_fib_local_table);
+	ip_fib_local_table = NULL;
+	return -ENOMEM;
 }
 #else
 
@@ -944,7 +956,7 @@ void __init ip_fib_init(void)
 	for (i = 0; i < FIB_TABLE_HASHSZ; i++)
 		INIT_HLIST_HEAD(&fib_table_hash[i]);
 
-	fib4_rules_init();
+	BUG_ON(fib4_rules_init());
 
 	register_netdevice_notifier(&fib_netdev_notifier);
 	register_inetaddr_notifier(&fib_inetaddr_notifier);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index afe669dd1bd5..0751734ecf41 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -311,8 +311,18 @@ static int __init fib_default_rules_init(void)
 	return 0;
 }
 
-void __init fib4_rules_init(void)
+int __init fib4_rules_init()
 {
-	BUG_ON(fib_default_rules_init());
+	int err;
+
 	fib_rules_register(&init_net, &fib4_rules_ops);
+	err = fib_default_rules_init();
+	if (err < 0)
+		goto fail;
+	return 0;
+
+fail:
+	/* also cleans all rules already added */
+	fib_rules_unregister(&init_net, &fib4_rules_ops);
+	return err;
 }
-- 
cgit v1.2.3


From 7b1a74fdbb9ec38a9780620fae25519fde4b21ee Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:22:17 -0800
Subject: [NETNS]: Refactor fib initialization so it can handle multiple
 namespaces.

This patch makes the fib to be initialized as a subsystem for the
network namespaces. The code does not handle several namespaces yet,
so in case of a creation of a network namespace, the
creation/initialization will not occur.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  3 +-
 net/ipv4/af_inet.c      |  4 ---
 net/ipv4/fib_frontend.c | 88 ++++++++++++++++++++++++++++++++++++++++++++-----
 net/ipv4/fib_hash.c     |  6 +---
 net/ipv4/fib_rules.c    | 11 +++++--
 net/ipv4/fib_trie.c     |  6 +---
 6 files changed, 92 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index cbff18d748dd..338d3ed10c31 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -186,7 +186,8 @@ static inline void fib_select_default(const struct flowi *flp, struct fib_result
 }
 
 #else /* CONFIG_IP_MULTIPLE_TABLES */
-extern int __init fib4_rules_init(void);
+extern int __net_init fib4_rules_init(struct net *net);
+extern void __net_exit fib4_rules_exit(struct net *net);
 
 #ifdef CONFIG_NET_CLS_ROUTE
 extern u32 fib_rules_tclass(struct fib_result *res);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index d5b8cb1dad67..40ecaae7be32 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1475,15 +1475,11 @@ static int __init ipv4_proc_init(void)
 		goto out_tcp;
 	if (udp4_proc_init())
 		goto out_udp;
-	if (fib_proc_init(&init_net))
-		goto out_fib;
 	if (ip_misc_proc_init())
 		goto out_misc;
 out:
 	return rc;
 out_misc:
-	fib_proc_exit(&init_net);
-out_fib:
 	udp4_proc_exit();
 out_udp:
 	tcp4_proc_exit();
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 1bb956b3be28..0484cae02736 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -59,7 +59,7 @@ struct fib_table *ip_fib_main_table;
 #define FIB_TABLE_HASHSZ 1
 static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
-static int __init fib4_rules_init(void)
+static int __net_init fib4_rules_init(struct net *net)
 {
 	ip_fib_local_table = fib_hash_init(RT_TABLE_LOCAL);
 	if (ip_fib_local_table == NULL)
@@ -863,10 +863,18 @@ static void nl_fib_input(struct sk_buff *skb)
 	netlink_unicast(fibnl, skb, pid, MSG_DONTWAIT);
 }
 
-static void nl_fib_lookup_init(void)
+static int nl_fib_lookup_init(struct net *net)
 {
-	fibnl = netlink_kernel_create(&init_net, NETLINK_FIB_LOOKUP, 0,
+	fibnl = netlink_kernel_create(net, NETLINK_FIB_LOOKUP, 0,
 				      nl_fib_input, NULL, THIS_MODULE);
+	if (fibnl == NULL)
+		return -EAFNOSUPPORT;
+	return 0;
+}
+
+static void nl_fib_lookup_exit(struct net *net)
+{
+	sock_put(fibnl);
 }
 
 static void fib_disable_ip(struct net_device *dev, int force)
@@ -949,22 +957,86 @@ static struct notifier_block fib_netdev_notifier = {
 	.notifier_call =fib_netdev_event,
 };
 
-void __init ip_fib_init(void)
+static int __net_init ip_fib_net_init(struct net *net)
 {
 	unsigned int i;
 
 	for (i = 0; i < FIB_TABLE_HASHSZ; i++)
 		INIT_HLIST_HEAD(&fib_table_hash[i]);
 
-	BUG_ON(fib4_rules_init());
+	return fib4_rules_init(net);
+}
 
-	register_netdevice_notifier(&fib_netdev_notifier);
-	register_inetaddr_notifier(&fib_inetaddr_notifier);
-	nl_fib_lookup_init();
+static void __net_exit ip_fib_net_exit(struct net *net)
+{
+	unsigned int i;
+
+#ifdef CONFIG_IP_MULTIPLE_TABLES
+	fib4_rules_exit(net);
+#endif
+
+	for (i = 0; i < FIB_TABLE_HASHSZ; i++) {
+		struct fib_table *tb;
+		struct hlist_head *head;
+		struct hlist_node *node, *tmp;
 
+		head = &fib_table_hash[i];
+		hlist_for_each_entry_safe(tb, node, tmp, head, tb_hlist) {
+			hlist_del(node);
+			tb->tb_flush(tb);
+			kfree(tb);
+		}
+	}
+}
+
+static int __net_init fib_net_init(struct net *net)
+{
+	int error;
+
+	error = 0;
+	if (net != &init_net)
+		goto out;
+
+	error = ip_fib_net_init(net);
+	if (error < 0)
+		goto out;
+	error = nl_fib_lookup_init(net);
+	if (error < 0)
+		goto out_nlfl;
+	error = fib_proc_init(net);
+	if (error < 0)
+		goto out_proc;
+out:
+	return error;
+
+out_proc:
+	nl_fib_lookup_exit(net);
+out_nlfl:
+	ip_fib_net_exit(net);
+	goto out;
+}
+
+static void __net_exit fib_net_exit(struct net *net)
+{
+	fib_proc_exit(net);
+	nl_fib_lookup_exit(net);
+	ip_fib_net_exit(net);
+}
+
+static struct pernet_operations fib_net_ops = {
+	.init = fib_net_init,
+	.exit = fib_net_exit,
+};
+
+void __init ip_fib_init(void)
+{
 	rtnl_register(PF_INET, RTM_NEWROUTE, inet_rtm_newroute, NULL);
 	rtnl_register(PF_INET, RTM_DELROUTE, inet_rtm_delroute, NULL);
 	rtnl_register(PF_INET, RTM_GETROUTE, NULL, inet_dump_fib);
+
+	register_pernet_subsys(&fib_net_ops);
+	register_netdevice_notifier(&fib_netdev_notifier);
+	register_inetaddr_notifier(&fib_inetaddr_notifier);
 }
 
 EXPORT_SYMBOL(inet_addr_type);
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 8de21bc4947a..694a072d8090 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -746,11 +746,7 @@ static int fn_hash_dump(struct fib_table *tb, struct sk_buff *skb, struct netlin
 	return skb->len;
 }
 
-#ifdef CONFIG_IP_MULTIPLE_TABLES
-struct fib_table * fib_hash_init(u32 id)
-#else
-struct fib_table * __init fib_hash_init(u32 id)
-#endif
+struct fib_table *fib_hash_init(u32 id)
 {
 	struct fib_table *tb;
 
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 0751734ecf41..1aae61c8ea6a 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -311,11 +311,11 @@ static int __init fib_default_rules_init(void)
 	return 0;
 }
 
-int __init fib4_rules_init()
+int __net_init fib4_rules_init(struct net *net)
 {
 	int err;
 
-	fib_rules_register(&init_net, &fib4_rules_ops);
+	fib_rules_register(net, &fib4_rules_ops);
 	err = fib_default_rules_init();
 	if (err < 0)
 		goto fail;
@@ -323,6 +323,11 @@ int __init fib4_rules_init()
 
 fail:
 	/* also cleans all rules already added */
-	fib_rules_unregister(&init_net, &fib4_rules_ops);
+	fib_rules_unregister(net, &fib4_rules_ops);
 	return err;
 }
+
+void __net_exit fib4_rules_exit(struct net *net)
+{
+	fib_rules_unregister(net, &fib4_rules_ops);
+}
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4bbfaeedf210..92b687e6a472 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1956,11 +1956,7 @@ out:
 
 /* Fix more generic FIB names for init later */
 
-#ifdef CONFIG_IP_MULTIPLE_TABLES
-struct fib_table * fib_hash_init(u32 id)
-#else
-struct fib_table * __init fib_hash_init(u32 id)
-#endif
+struct fib_table *fib_hash_init(u32 id)
 {
 	struct fib_table *tb;
 	struct trie *t;
-- 
cgit v1.2.3


From 93456b6d7753def8760b423ac6b986eb9d5a4a95 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:23:38 -0800
Subject: [IPV4]: Unify access to the routing tables.

Replace the direct pointers to local and main tables with
calls to fib_get_table() with appropriate argument.

This doesn't introduce additional dereferences, but makes the access to fib
tables uniform in any (CONFIG_IP_MULTIPLE_TABLES) case.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 39 ++++++++++++++++++++++++++++-----------
 net/ipv4/fib_frontend.c | 29 ++++++++++++-----------------
 2 files changed, 40 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 338d3ed10c31..249af662e897 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -120,16 +120,22 @@ struct fib_result_nl {
 	int             err;      
 };
 
+extern struct hlist_head fib_table_hash[];
+
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 
 #define FIB_RES_NH(res)		((res).fi->fib_nh[(res).nh_sel])
 #define FIB_RES_RESET(res)	((res).nh_sel = 0)
 
+#define FIB_TABLE_HASHSZ 2
+
 #else /* CONFIG_IP_ROUTE_MULTIPATH */
 
 #define FIB_RES_NH(res)		((res).fi->fib_nh[0])
 #define FIB_RES_RESET(res)
 
+#define FIB_TABLE_HASHSZ 256
+
 #endif /* CONFIG_IP_ROUTE_MULTIPATH */
 
 #define FIB_RES_PREFSRC(res)		((res).fi->fib_prefsrc ? : __fib_res_prefsrc(&res))
@@ -156,14 +162,17 @@ struct fib_table {
 
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 
-extern struct fib_table *ip_fib_local_table;
-extern struct fib_table *ip_fib_main_table;
+#define TABLE_LOCAL_INDEX	0
+#define TABLE_MAIN_INDEX	1
 
 static inline struct fib_table *fib_get_table(u32 id)
 {
-	if (id != RT_TABLE_LOCAL)
-		return ip_fib_main_table;
-	return ip_fib_local_table;
+	struct hlist_head *ptr;
+
+	ptr = id == RT_TABLE_LOCAL ?
+		&fib_table_hash[TABLE_LOCAL_INDEX] :
+		&fib_table_hash[TABLE_MAIN_INDEX];
+	return hlist_entry(ptr->first, struct fib_table, tb_hlist);
 }
 
 static inline struct fib_table *fib_new_table(u32 id)
@@ -173,16 +182,24 @@ static inline struct fib_table *fib_new_table(u32 id)
 
 static inline int fib_lookup(const struct flowi *flp, struct fib_result *res)
 {
-	if (ip_fib_local_table->tb_lookup(ip_fib_local_table, flp, res) &&
-	    ip_fib_main_table->tb_lookup(ip_fib_main_table, flp, res))
-		return -ENETUNREACH;
-	return 0;
+	struct fib_table *table;
+
+	table = fib_get_table(RT_TABLE_LOCAL);
+	if (!table->tb_lookup(table, flp, res))
+		return 0;
+
+	table = fib_get_table(RT_TABLE_MAIN);
+	if (!table->tb_lookup(table, flp, res))
+		return 0;
+	return -ENETUNREACH;
 }
 
-static inline void fib_select_default(const struct flowi *flp, struct fib_result *res)
+static inline void fib_select_default(const struct flowi *flp,
+				      struct fib_result *res)
 {
+	struct fib_table *table = fib_get_table(RT_TABLE_MAIN);
 	if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
-		ip_fib_main_table->tb_select_default(ip_fib_main_table, flp, res);
+		table->tb_select_default(table, flp, res);
 }
 
 #else /* CONFIG_IP_MULTIPLE_TABLES */
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 0484cae02736..9ff1e6669ef2 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -50,39 +50,34 @@
 #define FFprint(a...) printk(KERN_DEBUG a)
 
 static struct sock *fibnl;
+struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 
-struct fib_table *ip_fib_local_table;
-struct fib_table *ip_fib_main_table;
-
-#define FIB_TABLE_HASHSZ 1
-static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
-
 static int __net_init fib4_rules_init(struct net *net)
 {
-	ip_fib_local_table = fib_hash_init(RT_TABLE_LOCAL);
-	if (ip_fib_local_table == NULL)
+	struct fib_table *local_table, *main_table;
+
+	local_table = fib_hash_init(RT_TABLE_LOCAL);
+	if (local_table == NULL)
 		return -ENOMEM;
 
-	ip_fib_main_table  = fib_hash_init(RT_TABLE_MAIN);
-	if (ip_fib_main_table == NULL)
+	main_table  = fib_hash_init(RT_TABLE_MAIN);
+	if (main_table == NULL)
 		goto fail;
 
-	hlist_add_head_rcu(&ip_fib_local_table->tb_hlist, &fib_table_hash[0]);
-	hlist_add_head_rcu(&ip_fib_main_table->tb_hlist, &fib_table_hash[0]);
+	hlist_add_head_rcu(&local_table->tb_hlist,
+				&fib_table_hash[TABLE_LOCAL_INDEX]);
+	hlist_add_head_rcu(&main_table->tb_hlist,
+				&fib_table_hash[TABLE_MAIN_INDEX]);
 	return 0;
 
 fail:
-	kfree(ip_fib_local_table);
-	ip_fib_local_table = NULL;
+	kfree(local_table);
 	return -ENOMEM;
 }
 #else
 
-#define FIB_TABLE_HASHSZ 256
-static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
-
 struct fib_table *fib_new_table(u32 id)
 {
 	struct fib_table *tb;
-- 
cgit v1.2.3


From 8ad4942cd5bdad4143f7aa1d1bd4f7b2526c19c5 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:24:11 -0800
Subject: [NETNS]: Add netns parameter to fib_get_table/fib_new_table.

This patch extends the fib_get_table and the fib_new_table functions
with the network namespace pointer. That will allow to access the
table relatively from the network namespace.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 16 ++++++++--------
 net/ipv4/fib_frontend.c | 24 ++++++++++++------------
 net/ipv4/fib_hash.c     |  4 ++--
 net/ipv4/fib_rules.c    | 12 ++++++------
 net/ipv4/fib_trie.c     |  8 ++++----
 5 files changed, 32 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 249af662e897..dfb95d732aa5 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -165,7 +165,7 @@ struct fib_table {
 #define TABLE_LOCAL_INDEX	0
 #define TABLE_MAIN_INDEX	1
 
-static inline struct fib_table *fib_get_table(u32 id)
+static inline struct fib_table *fib_get_table(struct net *net, u32 id)
 {
 	struct hlist_head *ptr;
 
@@ -175,20 +175,20 @@ static inline struct fib_table *fib_get_table(u32 id)
 	return hlist_entry(ptr->first, struct fib_table, tb_hlist);
 }
 
-static inline struct fib_table *fib_new_table(u32 id)
+static inline struct fib_table *fib_new_table(struct net *net, u32 id)
 {
-	return fib_get_table(id);
+	return fib_get_table(net, id);
 }
 
 static inline int fib_lookup(const struct flowi *flp, struct fib_result *res)
 {
 	struct fib_table *table;
 
-	table = fib_get_table(RT_TABLE_LOCAL);
+	table = fib_get_table(&init_net, RT_TABLE_LOCAL);
 	if (!table->tb_lookup(table, flp, res))
 		return 0;
 
-	table = fib_get_table(RT_TABLE_MAIN);
+	table = fib_get_table(&init_net, RT_TABLE_MAIN);
 	if (!table->tb_lookup(table, flp, res))
 		return 0;
 	return -ENETUNREACH;
@@ -197,7 +197,7 @@ static inline int fib_lookup(const struct flowi *flp, struct fib_result *res)
 static inline void fib_select_default(const struct flowi *flp,
 				      struct fib_result *res)
 {
-	struct fib_table *table = fib_get_table(RT_TABLE_MAIN);
+	struct fib_table *table = fib_get_table(&init_net, RT_TABLE_MAIN);
 	if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
 		table->tb_select_default(table, flp, res);
 }
@@ -212,8 +212,8 @@ extern u32 fib_rules_tclass(struct fib_result *res);
 
 extern int fib_lookup(struct flowi *flp, struct fib_result *res);
 
-extern struct fib_table *fib_new_table(u32 id);
-extern struct fib_table *fib_get_table(u32 id);
+extern struct fib_table *fib_new_table(struct net *net, u32 id);
+extern struct fib_table *fib_get_table(struct net *net, u32 id);
 extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
 
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 9ff1e6669ef2..7718823711e3 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -78,14 +78,14 @@ fail:
 }
 #else
 
-struct fib_table *fib_new_table(u32 id)
+struct fib_table *fib_new_table(struct net *net, u32 id)
 {
 	struct fib_table *tb;
 	unsigned int h;
 
 	if (id == 0)
 		id = RT_TABLE_MAIN;
-	tb = fib_get_table(id);
+	tb = fib_get_table(net, id);
 	if (tb)
 		return tb;
 	tb = fib_hash_init(id);
@@ -96,7 +96,7 @@ struct fib_table *fib_new_table(u32 id)
 	return tb;
 }
 
-struct fib_table *fib_get_table(u32 id)
+struct fib_table *fib_get_table(struct net *net, u32 id)
 {
 	struct fib_table *tb;
 	struct hlist_node *node;
@@ -148,7 +148,7 @@ struct net_device * ip_dev_find(__be32 addr)
 	res.r = NULL;
 #endif
 
-	local_table = fib_get_table(RT_TABLE_LOCAL);
+	local_table = fib_get_table(&init_net, RT_TABLE_LOCAL);
 	if (!local_table || local_table->tb_lookup(local_table, &fl, &res))
 		return NULL;
 	if (res.type != RTN_LOCAL)
@@ -183,7 +183,7 @@ static inline unsigned __inet_dev_addr_type(const struct net_device *dev,
 	res.r = NULL;
 #endif
 
-	local_table = fib_get_table(RT_TABLE_LOCAL);
+	local_table = fib_get_table(&init_net, RT_TABLE_LOCAL);
 	if (local_table) {
 		ret = RTN_UNICAST;
 		if (!local_table->tb_lookup(local_table, &fl, &res)) {
@@ -453,13 +453,13 @@ int ip_rt_ioctl(unsigned int cmd, void __user *arg)
 			struct fib_table *tb;
 
 			if (cmd == SIOCDELRT) {
-				tb = fib_get_table(cfg.fc_table);
+				tb = fib_get_table(&init_net, cfg.fc_table);
 				if (tb)
 					err = tb->tb_delete(tb, &cfg);
 				else
 					err = -ESRCH;
 			} else {
-				tb = fib_new_table(cfg.fc_table);
+				tb = fib_new_table(&init_net, cfg.fc_table);
 				if (tb)
 					err = tb->tb_insert(tb, &cfg);
 				else
@@ -573,7 +573,7 @@ static int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *ar
 	if (err < 0)
 		goto errout;
 
-	tb = fib_get_table(cfg.fc_table);
+	tb = fib_get_table(net, cfg.fc_table);
 	if (tb == NULL) {
 		err = -ESRCH;
 		goto errout;
@@ -598,7 +598,7 @@ static int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *ar
 	if (err < 0)
 		goto errout;
 
-	tb = fib_new_table(cfg.fc_table);
+	tb = fib_new_table(&init_net, cfg.fc_table);
 	if (tb == NULL) {
 		err = -ENOBUFS;
 		goto errout;
@@ -671,9 +671,9 @@ static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifad
 	};
 
 	if (type == RTN_UNICAST)
-		tb = fib_new_table(RT_TABLE_MAIN);
+		tb = fib_new_table(&init_net, RT_TABLE_MAIN);
 	else
-		tb = fib_new_table(RT_TABLE_LOCAL);
+		tb = fib_new_table(&init_net, RT_TABLE_LOCAL);
 
 	if (tb == NULL)
 		return;
@@ -848,7 +848,7 @@ static void nl_fib_input(struct sk_buff *skb)
 	nlh = nlmsg_hdr(skb);
 
 	frn = (struct fib_result_nl *) NLMSG_DATA(nlh);
-	tb = fib_get_table(frn->tb_id_in);
+	tb = fib_get_table(&init_net, frn->tb_id_in);
 
 	nl_fib_lookup(frn, tb);
 
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 694a072d8090..d20ae2e28ab5 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -796,7 +796,7 @@ struct fib_iter_state {
 static struct fib_alias *fib_get_first(struct seq_file *seq)
 {
 	struct fib_iter_state *iter = seq->private;
-	struct fib_table *main_table = fib_get_table(RT_TABLE_MAIN);
+	struct fib_table *main_table = fib_get_table(&init_net, RT_TABLE_MAIN);
 	struct fn_hash *table = (struct fn_hash *)main_table->tb_data;
 
 	iter->bucket    = 0;
@@ -937,7 +937,7 @@ static void *fib_seq_start(struct seq_file *seq, loff_t *pos)
 	void *v = NULL;
 
 	read_lock(&fib_hash_lock);
-	if (fib_get_table(RT_TABLE_MAIN))
+	if (fib_get_table(&init_net, RT_TABLE_MAIN))
 		v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 	return v;
 }
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 1aae61c8ea6a..49819fe7e4c3 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -93,7 +93,7 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto errout;
 	}
 
-	if ((tbl = fib_get_table(rule->table)) == NULL)
+	if ((tbl = fib_get_table(&init_net, rule->table)) == NULL)
 		goto errout;
 
 	err = tbl->tb_lookup(tbl, flp, (struct fib_result *) arg->result);
@@ -109,7 +109,7 @@ void fib_select_default(const struct flowi *flp, struct fib_result *res)
 	if (res->r && res->r->action == FR_ACT_TO_TBL &&
 	    FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK) {
 		struct fib_table *tb;
-		if ((tb = fib_get_table(res->r->table)) != NULL)
+		if ((tb = fib_get_table(&init_net, res->r->table)) != NULL)
 			tb->tb_select_default(tb, flp, res);
 	}
 }
@@ -130,13 +130,13 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 	return 1;
 }
 
-static struct fib_table *fib_empty_table(void)
+static struct fib_table *fib_empty_table(struct net *net)
 {
 	u32 id;
 
 	for (id = 1; id <= RT_TABLE_MAX; id++)
-		if (fib_get_table(id) == NULL)
-			return fib_new_table(id);
+		if (fib_get_table(net, id) == NULL)
+			return fib_new_table(net, id);
 	return NULL;
 }
 
@@ -159,7 +159,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		if (rule->action == FR_ACT_TO_TBL) {
 			struct fib_table *table;
 
-			table = fib_empty_table();
+			table = fib_empty_table(&init_net);
 			if (table == NULL) {
 				err = -ENOBUFS;
 				goto errout;
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 92b687e6a472..fc0624e6a649 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2167,12 +2167,12 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v)
 	struct fib_table *tb;
 
 	trie_local = NULL;
-	tb = fib_get_table(RT_TABLE_LOCAL);
+	tb = fib_get_table(&init_net, RT_TABLE_LOCAL);
 	if (tb)
 		trie_local = (struct trie *) tb->tb_data;
 
 	trie_main = NULL;
-	tb = fib_get_table(RT_TABLE_MAIN);
+	tb = fib_get_table(&init_net, RT_TABLE_MAIN);
 	if (tb)
 		trie_main = (struct trie *) tb->tb_data;
 
@@ -2239,12 +2239,12 @@ static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos)
 	struct fib_table *tb;
 
 	if (!iter->trie_local) {
-		tb = fib_get_table(RT_TABLE_LOCAL);
+		tb = fib_get_table(&init_net, RT_TABLE_LOCAL);
 		if (tb)
 			iter->trie_local = (struct trie *) tb->tb_data;
 	}
 	if (!iter->trie_main) {
-		tb = fib_get_table(RT_TABLE_MAIN);
+		tb = fib_get_table(&init_net, RT_TABLE_MAIN);
 		if (tb)
 			iter->trie_main = (struct trie *) tb->tb_data;
 	}
-- 
cgit v1.2.3


From 6b175b26c1048d331508940ad3516ead1998084f Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Thu, 10 Jan 2008 03:25:28 -0800
Subject: [NETNS]: Add netns parameter to inet_(dev_)add_type.

The patch extends the inet_addr_type and inet_dev_addr_type with the
network namespace pointer. That allows to access the different tables
relatively to the network namespace.

The modification of the signature function is reported in all the
callers of the inet_addr_type using the pointer to the well known
init_net.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_main.c      |  2 +-
 include/net/route.h               |  4 ++--
 net/atm/clip.c                    |  2 +-
 net/ipv4/af_inet.c                |  2 +-
 net/ipv4/arp.c                    | 12 ++++++------
 net/ipv4/fib_frontend.c           | 18 ++++++++++--------
 net/ipv4/fib_semantics.c          |  4 ++--
 net/ipv4/icmp.c                   |  4 ++--
 net/ipv4/ip_options.c             |  4 ++--
 net/ipv4/ipvs/ip_vs_core.c        |  2 +-
 net/ipv4/ipvs/ip_vs_ctl.c         |  4 ++--
 net/ipv4/netfilter.c              |  2 +-
 net/ipv4/netfilter/ipt_addrtype.c |  2 +-
 net/ipv4/raw.c                    |  2 +-
 net/ipv4/route.c                  |  2 +-
 net/ipv6/af_inet6.c               |  2 +-
 net/sctp/protocol.c               |  2 +-
 17 files changed, 36 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c
index ff999ff0b627..f1866a283850 100644
--- a/drivers/s390/net/qeth_main.c
+++ b/drivers/s390/net/qeth_main.c
@@ -8340,7 +8340,7 @@ qeth_arp_constructor(struct neighbour *neigh)
 	neigh->parms = neigh_parms_clone(parms);
 	rcu_read_unlock();
 
-	neigh->type = inet_addr_type(*(__be32 *) neigh->primary_key);
+	neigh->type = inet_addr_type(&init_net, *(__be32 *) neigh->primary_key);
 	neigh->nud_state = NUD_NOARP;
 	neigh->ops = arp_direct_ops;
 	neigh->output = neigh->ops->queue_xmit;
diff --git a/include/net/route.h b/include/net/route.h
index 76b08c1affc0..b77700059252 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -117,8 +117,8 @@ extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, stru
 extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
-extern unsigned		inet_addr_type(__be32 addr);
-extern unsigned		inet_dev_addr_type(const struct net_device *dev, __be32 addr);
+extern unsigned		inet_addr_type(struct net *net, __be32 addr);
+extern unsigned		inet_dev_addr_type(struct net *net, const struct net_device *dev, __be32 addr);
 extern void		ip_rt_multicast_event(struct in_device *);
 extern int		ip_rt_ioctl(unsigned int cmd, void __user *arg);
 extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
diff --git a/net/atm/clip.c b/net/atm/clip.c
index 47fbdc0c5f72..df7d218a6827 100644
--- a/net/atm/clip.c
+++ b/net/atm/clip.c
@@ -285,7 +285,7 @@ static int clip_constructor(struct neighbour *neigh)
 	struct neigh_parms *parms;
 
 	pr_debug("clip_constructor (neigh %p, entry %p)\n", neigh, entry);
-	neigh->type = inet_addr_type(entry->ip);
+	neigh->type = inet_addr_type(&init_net, entry->ip);
 	if (neigh->type != RTN_UNICAST)
 		return -EINVAL;
 
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 40ecaae7be32..d76f8d8d5eba 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -446,7 +446,7 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	if (addr_len < sizeof(struct sockaddr_in))
 		goto out;
 
-	chk_addr_ret = inet_addr_type(addr->sin_addr.s_addr);
+	chk_addr_ret = inet_addr_type(&init_net, addr->sin_addr.s_addr);
 
 	/* Not specified by any standard per-se, however it breaks too
 	 * many applications when removed.  It is unfortunate since
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 46edf1c32a40..3f0730ec0a21 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -235,7 +235,7 @@ static int arp_constructor(struct neighbour *neigh)
 	struct in_device *in_dev;
 	struct neigh_parms *parms;
 
-	neigh->type = inet_addr_type(addr);
+	neigh->type = inet_addr_type(&init_net, addr);
 
 	rcu_read_lock();
 	in_dev = __in_dev_get_rcu(dev);
@@ -341,14 +341,14 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 	switch (IN_DEV_ARP_ANNOUNCE(in_dev)) {
 	default:
 	case 0:		/* By default announce any local IP */
-		if (skb && inet_addr_type(ip_hdr(skb)->saddr) == RTN_LOCAL)
+		if (skb && inet_addr_type(&init_net, ip_hdr(skb)->saddr) == RTN_LOCAL)
 			saddr = ip_hdr(skb)->saddr;
 		break;
 	case 1:		/* Restrict announcements of saddr in same subnet */
 		if (!skb)
 			break;
 		saddr = ip_hdr(skb)->saddr;
-		if (inet_addr_type(saddr) == RTN_LOCAL) {
+		if (inet_addr_type(&init_net, saddr) == RTN_LOCAL) {
 			/* saddr should be known to target */
 			if (inet_addr_onlink(in_dev, target, saddr))
 				break;
@@ -479,7 +479,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb)
 
 	paddr = ((struct rtable*)skb->dst)->rt_gateway;
 
-	if (arp_set_predefined(inet_addr_type(paddr), haddr, paddr, dev))
+	if (arp_set_predefined(inet_addr_type(&init_net, paddr), haddr, paddr, dev))
 		return 0;
 
 	n = __neigh_lookup(&arp_tbl, &paddr, dev, 1);
@@ -806,7 +806,7 @@ static int arp_process(struct sk_buff *skb)
 	/* Special case: IPv4 duplicate address detection packet (RFC2131) */
 	if (sip == 0) {
 		if (arp->ar_op == htons(ARPOP_REQUEST) &&
-		    inet_addr_type(tip) == RTN_LOCAL &&
+		    inet_addr_type(&init_net, tip) == RTN_LOCAL &&
 		    !arp_ignore(in_dev,dev,sip,tip))
 			arp_send(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip, sha,
 				 dev->dev_addr, sha);
@@ -866,7 +866,7 @@ static int arp_process(struct sk_buff *skb)
 		 */
 		if (n == NULL &&
 		    arp->ar_op == htons(ARPOP_REPLY) &&
-		    inet_addr_type(sip) == RTN_UNICAST)
+		    inet_addr_type(&init_net, sip) == RTN_UNICAST)
 			n = __neigh_lookup(&arp_tbl, &sip, dev, 1);
 	}
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 7718823711e3..d1a45cb6f6b0 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -166,7 +166,8 @@ out:
  * Find address type as if only "dev" was present in the system. If
  * on_dev is NULL then all interfaces are taken into consideration.
  */
-static inline unsigned __inet_dev_addr_type(const struct net_device *dev,
+static inline unsigned __inet_dev_addr_type(struct net *net,
+					    const struct net_device *dev,
 					    __be32 addr)
 {
 	struct flowi		fl = { .nl_u = { .ip4_u = { .daddr = addr } } };
@@ -183,7 +184,7 @@ static inline unsigned __inet_dev_addr_type(const struct net_device *dev,
 	res.r = NULL;
 #endif
 
-	local_table = fib_get_table(&init_net, RT_TABLE_LOCAL);
+	local_table = fib_get_table(net, RT_TABLE_LOCAL);
 	if (local_table) {
 		ret = RTN_UNICAST;
 		if (!local_table->tb_lookup(local_table, &fl, &res)) {
@@ -195,14 +196,15 @@ static inline unsigned __inet_dev_addr_type(const struct net_device *dev,
 	return ret;
 }
 
-unsigned int inet_addr_type(__be32 addr)
+unsigned int inet_addr_type(struct net *net, __be32 addr)
 {
-	return __inet_dev_addr_type(NULL, addr);
+	return __inet_dev_addr_type(net, NULL, addr);
 }
 
-unsigned int inet_dev_addr_type(const struct net_device *dev, __be32 addr)
+unsigned int inet_dev_addr_type(struct net *net, const struct net_device *dev,
+				__be32 addr)
 {
-       return __inet_dev_addr_type(dev, addr);
+       return __inet_dev_addr_type(net, dev, addr);
 }
 
 /* Given (packet source, input interface) and optional (dst, oif, tos):
@@ -391,7 +393,7 @@ static int rtentry_to_fib_config(int cmd, struct rtentry *rt,
 	if (rt->rt_gateway.sa_family == AF_INET && addr) {
 		cfg->fc_gw = addr;
 		if (rt->rt_flags & RTF_GATEWAY &&
-		    inet_addr_type(addr) == RTN_UNICAST)
+		    inet_addr_type(&init_net, addr) == RTN_UNICAST)
 			cfg->fc_scope = RT_SCOPE_UNIVERSE;
 	}
 
@@ -782,7 +784,7 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
 		fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim);
 
 		/* Check, that this local address finally disappeared. */
-		if (inet_addr_type(ifa->ifa_local) != RTN_LOCAL) {
+		if (inet_addr_type(&init_net, ifa->ifa_local) != RTN_LOCAL) {
 			/* And the last, but not the least thing.
 			   We must flush stray FIB entries.
 
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index bbd4a247b195..c1263e23876b 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -531,7 +531,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi,
 
 			if (cfg->fc_scope >= RT_SCOPE_LINK)
 				return -EINVAL;
-			if (inet_addr_type(nh->nh_gw) != RTN_UNICAST)
+			if (inet_addr_type(&init_net, nh->nh_gw) != RTN_UNICAST)
 				return -EINVAL;
 			if ((dev = __dev_get_by_index(&init_net, nh->nh_oif)) == NULL)
 				return -ENODEV;
@@ -809,7 +809,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg)
 	if (fi->fib_prefsrc) {
 		if (cfg->fc_type != RTN_LOCAL || !cfg->fc_dst ||
 		    fi->fib_prefsrc != cfg->fc_dst)
-			if (inet_addr_type(fi->fib_prefsrc) != RTN_LOCAL)
+			if (inet_addr_type(&init_net, fi->fib_prefsrc) != RTN_LOCAL)
 				goto err_inval;
 	}
 
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 1a6024978e28..e57f1673bf6b 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -590,7 +590,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 		if (xfrm_decode_session_reverse(skb_in, &fl, AF_INET))
 			goto out_unlock;
 
-		if (inet_addr_type(fl.fl4_src) == RTN_LOCAL)
+		if (inet_addr_type(&init_net, fl.fl4_src) == RTN_LOCAL)
 			err = __ip_route_output_key(&rt2, &fl);
 		else {
 			struct flowi fl2 = {};
@@ -733,7 +733,7 @@ static void icmp_unreach(struct sk_buff *skb)
 	 */
 
 	if (!sysctl_icmp_ignore_bogus_error_responses &&
-	    inet_addr_type(iph->daddr) == RTN_BROADCAST) {
+	    inet_addr_type(&init_net, iph->daddr) == RTN_BROADCAST) {
 		if (net_ratelimit())
 			printk(KERN_WARNING "%u.%u.%u.%u sent an invalid ICMP "
 					    "type %u, code %u "
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 2f14745a9e1f..4d315158fd3c 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -151,7 +151,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 						__be32 addr;
 
 						memcpy(&addr, sptr+soffset-1, 4);
-						if (inet_addr_type(addr) != RTN_LOCAL) {
+						if (inet_addr_type(&init_net, addr) != RTN_LOCAL) {
 							dopt->ts_needtime = 1;
 							soffset += 8;
 						}
@@ -400,7 +400,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 					{
 						__be32 addr;
 						memcpy(&addr, &optptr[optptr[2]-1], 4);
-						if (inet_addr_type(addr) == RTN_UNICAST)
+						if (inet_addr_type(&init_net, addr) == RTN_UNICAST)
 							break;
 						if (skb)
 							timeptr = (__be32*)&optptr[optptr[2]+3];
diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c
index 041f5120808a..963981a9d501 100644
--- a/net/ipv4/ipvs/ip_vs_core.c
+++ b/net/ipv4/ipvs/ip_vs_core.c
@@ -423,7 +423,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 	   and the destination is RTN_UNICAST (and not local), then create
 	   a cache_bypass connection entry */
 	if (sysctl_ip_vs_cache_bypass && svc->fwmark
-	    && (inet_addr_type(iph->daddr) == RTN_UNICAST)) {
+	    && (inet_addr_type(&init_net, iph->daddr) == RTN_UNICAST)) {
 		int ret, cs;
 		struct ip_vs_conn *cp;
 
diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
index 9fecfe7d1164..94c5767c8e01 100644
--- a/net/ipv4/ipvs/ip_vs_ctl.c
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -704,7 +704,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc,
 	conn_flags = udest->conn_flags | IP_VS_CONN_F_INACTIVE;
 
 	/* check if local node and update the flags */
-	if (inet_addr_type(udest->addr) == RTN_LOCAL) {
+	if (inet_addr_type(&init_net, udest->addr) == RTN_LOCAL) {
 		conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK)
 			| IP_VS_CONN_F_LOCALNODE;
 	}
@@ -756,7 +756,7 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest,
 
 	EnterFunction(2);
 
-	atype = inet_addr_type(udest->addr);
+	atype = inet_addr_type(&init_net, udest->addr);
 	if (atype != RTN_LOCAL && atype != RTN_UNICAST)
 		return -EINVAL;
 
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 4011f8f987c0..0ed843ed420a 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -19,7 +19,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
 	unsigned int hh_len;
 	unsigned int type;
 
-	type = inet_addr_type(iph->saddr);
+	type = inet_addr_type(&init_net, iph->saddr);
 	if (addr_type == RTN_UNSPEC)
 		addr_type = type;
 
diff --git a/net/ipv4/netfilter/ipt_addrtype.c b/net/ipv4/netfilter/ipt_addrtype.c
index 14394c6a3c29..8763902944b7 100644
--- a/net/ipv4/netfilter/ipt_addrtype.c
+++ b/net/ipv4/netfilter/ipt_addrtype.c
@@ -26,7 +26,7 @@ MODULE_DESCRIPTION("iptables addrtype match");
 static inline bool match_type(const struct net_device *dev, __be32 addr,
 			      u_int16_t mask)
 {
-	return !!(mask & (1 << inet_dev_addr_type(dev, addr)));
+	return !!(mask & (1 << inet_dev_addr_type(&init_net, dev, addr)));
 }
 
 static bool
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e811034b1b01..747911a8241c 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -620,7 +620,7 @@ static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 
 	if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in))
 		goto out;
-	chk_addr_ret = inet_addr_type(addr->sin_addr.s_addr);
+	chk_addr_ret = inet_addr_type(&init_net, addr->sin_addr.s_addr);
 	ret = -EADDRNOTAVAIL;
 	if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
 	    chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 78c4ce424a68..1e59c0d4b11e 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1164,7 +1164,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 		if (IN_DEV_SEC_REDIRECTS(in_dev) && ip_fib_check_default(new_gw, dev))
 			goto reject_redirect;
 	} else {
-		if (inet_addr_type(new_gw) != RTN_UNICAST)
+		if (inet_addr_type(&init_net, new_gw) != RTN_UNICAST)
 			goto reject_redirect;
 	}
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 218b8b3050a1..ac8772dd968f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -280,7 +280,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	/* Check if the address belongs to the host. */
 	if (addr_type == IPV6_ADDR_MAPPED) {
 		v4addr = addr->sin6_addr.s6_addr32[3];
-		if (inet_addr_type(v4addr) != RTN_LOCAL) {
+		if (inet_addr_type(&init_net, v4addr) != RTN_LOCAL) {
 			err = -EADDRNOTAVAIL;
 			goto out;
 		}
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index b92196495027..3f7def2936b4 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -372,7 +372,7 @@ static int sctp_v4_addr_valid(union sctp_addr *addr,
 /* Should this be available for binding?   */
 static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp)
 {
-	int ret = inet_addr_type(addr->v4.sin_addr.s_addr);
+	int ret = inet_addr_type(&init_net, addr->v4.sin_addr.s_addr);
 
 
 	if (addr->v4.sin_addr.s_addr != INADDR_ANY &&
-- 
cgit v1.2.3


From 4d1169c1e781e5853317c6b75620d678b2c4854e Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:26:13 -0800
Subject: [NETNS]: Add netns to nl_info structure.

nl_info is used to track the end-user destination of routing change
notification. This is a natural object to hold a namespace on. Place
it there and utilize the context in the appropriate places.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h    |  1 +
 net/ipv4/fib_frontend.c  |  5 +++++
 net/ipv4/fib_semantics.c | 16 ++++++++++------
 net/ipv6/ip6_fib.c       |  4 +++-
 net/ipv6/route.c         |  8 ++++++--
 5 files changed, 25 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index a52d03601754..a5e6d0ef51dd 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -217,6 +217,7 @@ struct nla_policy {
  */
 struct nl_info {
 	struct nlmsghdr		*nlh;
+	struct net		*nl_net;
 	u32			pid;
 };
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index d1a45cb6f6b0..90aa05f47f8e 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -310,6 +310,7 @@ static int rtentry_to_fib_config(int cmd, struct rtentry *rt,
 	int plen;
 
 	memset(cfg, 0, sizeof(*cfg));
+	cfg->fc_nlinfo.nl_net = &init_net;
 
 	if (rt->rt_dst.sa_family != AF_INET)
 		return -EAFNOSUPPORT;
@@ -516,6 +517,7 @@ static int rtm_to_fib_config(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 	cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid;
 	cfg->fc_nlinfo.nlh = nlh;
+	cfg->fc_nlinfo.nl_net = &init_net;
 
 	if (cfg->fc_type > RTN_MAX) {
 		err = -EINVAL;
@@ -670,6 +672,9 @@ static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifad
 		.fc_prefsrc = ifa->ifa_local,
 		.fc_oif = ifa->ifa_dev->dev->ifindex,
 		.fc_nlflags = NLM_F_CREATE | NLM_F_APPEND,
+		.fc_nlinfo = {
+			.nl_net = &init_net,
+		},
 	};
 
 	if (type == RTN_UNICAST)
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index c1263e23876b..0de6102020e2 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -320,11 +320,11 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, &init_net, info->pid, RTNLGRP_IPV4_ROUTE,
+	err = rtnl_notify(skb, info->nl_net, info->pid, RTNLGRP_IPV4_ROUTE,
 			  info->nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(&init_net, RTNLGRP_IPV4_ROUTE, err);
+		rtnl_set_sk_err(info->nl_net, RTNLGRP_IPV4_ROUTE, err);
 }
 
 /* Return the first fib alias matching TOS with
@@ -531,9 +531,11 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi,
 
 			if (cfg->fc_scope >= RT_SCOPE_LINK)
 				return -EINVAL;
-			if (inet_addr_type(&init_net, nh->nh_gw) != RTN_UNICAST)
+			if (inet_addr_type(cfg->fc_nlinfo.nl_net,
+					   nh->nh_gw) != RTN_UNICAST)
 				return -EINVAL;
-			if ((dev = __dev_get_by_index(&init_net, nh->nh_oif)) == NULL)
+			if ((dev = __dev_get_by_index(cfg->fc_nlinfo.nl_net,
+						      nh->nh_oif)) == NULL)
 				return -ENODEV;
 			if (!(dev->flags&IFF_UP))
 				return -ENETDOWN;
@@ -795,7 +797,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg)
 		if (nhs != 1 || nh->nh_gw)
 			goto err_inval;
 		nh->nh_scope = RT_SCOPE_NOWHERE;
-		nh->nh_dev = dev_get_by_index(&init_net, fi->fib_nh->nh_oif);
+		nh->nh_dev = dev_get_by_index(cfg->fc_nlinfo.nl_net,
+					      fi->fib_nh->nh_oif);
 		err = -ENODEV;
 		if (nh->nh_dev == NULL)
 			goto failure;
@@ -809,7 +812,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg)
 	if (fi->fib_prefsrc) {
 		if (cfg->fc_type != RTN_LOCAL || !cfg->fc_dst ||
 		    fi->fib_prefsrc != cfg->fc_dst)
-			if (inet_addr_type(&init_net, fi->fib_prefsrc) != RTN_LOCAL)
+			if (inet_addr_type(cfg->fc_nlinfo.nl_net,
+					   fi->fib_prefsrc) != RTN_LOCAL)
 				goto err_inval;
 	}
 
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 0e83164aa3e6..f93407cf6515 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1314,7 +1314,9 @@ static int fib6_walk(struct fib6_walker_t *w)
 
 static int fib6_clean_node(struct fib6_walker_t *w)
 {
-	struct nl_info info = {};
+	struct nl_info info = {
+		.nl_net = &init_net,
+	};
 	int res;
 	struct rt6_info *rt;
 	struct fib6_cleaner_t *c = container_of(w, struct fib6_cleaner_t, w);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index d2b3cf695aff..48c8d7cb9020 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -601,7 +601,9 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info)
 
 int ip6_ins_rt(struct rt6_info *rt)
 {
-	struct nl_info info = {};
+	struct nl_info info = {
+		.nl_net = &init_net,
+	};
 	return __ip6_ins_rt(rt, &info);
 }
 
@@ -1259,7 +1261,9 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 
 int ip6_del_rt(struct rt6_info *rt)
 {
-	struct nl_info info = {};
+	struct nl_info info = {
+		.nl_net = &init_net,
+	};
 	return __ip6_del_rt(rt, &info);
 }
 
-- 
cgit v1.2.3


From e4e4971c5f8b70daccdd401132a81b723dc8337e Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:27:51 -0800
Subject: [NETNS]: Namespacing IPv4 fib rules.

The final trick for rules: place fib4_rules_ops into struct net and
modify initialization path for this.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  5 +++++
 net/ipv4/fib_rules.c     | 44 ++++++++++++++++++++++++--------------------
 2 files changed, 29 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 61a28ff9440a..a7bd5d83e43e 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -4,8 +4,10 @@
 
 #ifndef __NETNS_IPV4_H__
 #define __NETNS_IPV4_H__
+
 struct ctl_table_header;
 struct ipv4_devconf;
+struct fib_rules_ops;
 
 struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
@@ -13,5 +15,8 @@ struct netns_ipv4 {
 #endif
 	struct ipv4_devconf	*devconf_all;
 	struct ipv4_devconf	*devconf_dflt;
+#ifdef CONFIG_IP_MULTIPLE_TABLES
+	struct fib_rules_ops	*rules_ops;
+#endif
 };
 #endif
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 49819fe7e4c3..72232ab4ecb1 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -32,8 +32,6 @@
 #include <net/ip_fib.h>
 #include <net/fib_rules.h>
 
-static struct fib_rules_ops fib4_rules_ops;
-
 struct fib4_rule
 {
 	struct fib_rule		common;
@@ -63,7 +61,7 @@ int fib_lookup(struct flowi *flp, struct fib_result *res)
 	};
 	int err;
 
-	err = fib_rules_lookup(&fib4_rules_ops, flp, 0, &arg);
+	err = fib_rules_lookup(init_net.ipv4.rules_ops, flp, 0, &arg);
 	res->r = arg.rule;
 
 	return err;
@@ -149,6 +147,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 			       struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
 			       struct nlattr **tb)
 {
+	struct net *net = skb->sk->sk_net;
 	int err = -EINVAL;
 	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
@@ -159,7 +158,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		if (rule->action == FR_ACT_TO_TBL) {
 			struct fib_table *table;
 
-			table = fib_empty_table(&init_net);
+			table = fib_empty_table(net);
 			if (table == NULL) {
 				err = -ENOBUFS;
 				goto errout;
@@ -250,9 +249,9 @@ static u32 fib4_rule_default_pref(struct fib_rules_ops *ops)
 	struct list_head *pos;
 	struct fib_rule *rule;
 
-	if (!list_empty(&fib4_rules_ops.rules_list)) {
-		pos = fib4_rules_ops.rules_list.next;
-		if (pos->next != &fib4_rules_ops.rules_list) {
+	if (!list_empty(&ops->rules_list)) {
+		pos = ops->rules_list.next;
+		if (pos->next != &ops->rules_list) {
 			rule = list_entry(pos->next, struct fib_rule, list);
 			if (rule->pref)
 				return rule->pref - 1;
@@ -274,7 +273,7 @@ static void fib4_rule_flush_cache(void)
 	rt_cache_flush(-1);
 }
 
-static struct fib_rules_ops fib4_rules_ops = {
+static struct fib_rules_ops fib4_rules_ops_template = {
 	.family		= AF_INET,
 	.rule_size	= sizeof(struct fib4_rule),
 	.addr_size	= sizeof(u32),
@@ -288,24 +287,20 @@ static struct fib_rules_ops fib4_rules_ops = {
 	.flush_cache	= fib4_rule_flush_cache,
 	.nlgroup	= RTNLGRP_IPV4_RULE,
 	.policy		= fib4_rule_policy,
-	.rules_list	= LIST_HEAD_INIT(fib4_rules_ops.rules_list),
 	.owner		= THIS_MODULE,
 };
 
-static int __init fib_default_rules_init(void)
+static int fib_default_rules_init(struct fib_rules_ops *ops)
 {
 	int err;
 
-	err = fib_default_rule_add(&fib4_rules_ops, 0,
-				   RT_TABLE_LOCAL, FIB_RULE_PERMANENT);
+	err = fib_default_rule_add(ops, 0, RT_TABLE_LOCAL, FIB_RULE_PERMANENT);
 	if (err < 0)
 		return err;
-	err = fib_default_rule_add(&fib4_rules_ops, 0x7FFE,
-				   RT_TABLE_MAIN, 0);
+	err = fib_default_rule_add(ops, 0x7FFE, RT_TABLE_MAIN, 0);
 	if (err < 0)
 		return err;
-	err = fib_default_rule_add(&fib4_rules_ops, 0x7FFF,
-				   RT_TABLE_DEFAULT, 0);
+	err = fib_default_rule_add(ops, 0x7FFF, RT_TABLE_DEFAULT, 0);
 	if (err < 0)
 		return err;
 	return 0;
@@ -314,20 +309,29 @@ static int __init fib_default_rules_init(void)
 int __net_init fib4_rules_init(struct net *net)
 {
 	int err;
+	struct fib_rules_ops *ops;
+
+	ops = kmemdup(&fib4_rules_ops_template, sizeof(*ops), GFP_KERNEL);
+	if (ops == NULL)
+		return -ENOMEM;
+	INIT_LIST_HEAD(&ops->rules_list);
+	fib_rules_register(net, ops);
 
-	fib_rules_register(net, &fib4_rules_ops);
-	err = fib_default_rules_init();
+	err = fib_default_rules_init(ops);
 	if (err < 0)
 		goto fail;
+	net->ipv4.rules_ops = ops;
 	return 0;
 
 fail:
 	/* also cleans all rules already added */
-	fib_rules_unregister(net, &fib4_rules_ops);
+	fib_rules_unregister(net, ops);
+	kfree(ops);
 	return err;
 }
 
 void __net_exit fib4_rules_exit(struct net *net)
 {
-	fib_rules_unregister(net, &fib4_rules_ops);
+	fib_rules_unregister(net, net->ipv4.rules_ops);
+	kfree(net->ipv4.rules_ops);
 }
-- 
cgit v1.2.3


From e4aef8aea31e6fc61b33a57120968a6e9824d138 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:28:24 -0800
Subject: [NETNS]: Place fib tables into netns.

The preparatory work has been done. All we need is to substitute
fib_table_hash with net->ipv4.fib_table_hash. Netns context is
available when required.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     |  6 ++----
 include/net/netns/ipv4.h |  2 ++
 net/ipv4/fib_frontend.c  | 36 ++++++++++++++++++++++++------------
 3 files changed, 28 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index dfb95d732aa5..025b225181e1 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -120,8 +120,6 @@ struct fib_result_nl {
 	int             err;      
 };
 
-extern struct hlist_head fib_table_hash[];
-
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 
 #define FIB_RES_NH(res)		((res).fi->fib_nh[(res).nh_sel])
@@ -170,8 +168,8 @@ static inline struct fib_table *fib_get_table(struct net *net, u32 id)
 	struct hlist_head *ptr;
 
 	ptr = id == RT_TABLE_LOCAL ?
-		&fib_table_hash[TABLE_LOCAL_INDEX] :
-		&fib_table_hash[TABLE_MAIN_INDEX];
+		&net->ipv4.fib_table_hash[TABLE_LOCAL_INDEX] :
+		&net->ipv4.fib_table_hash[TABLE_MAIN_INDEX];
 	return hlist_entry(ptr->first, struct fib_table, tb_hlist);
 }
 
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index a7bd5d83e43e..2dd31937d022 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -8,6 +8,7 @@
 struct ctl_table_header;
 struct ipv4_devconf;
 struct fib_rules_ops;
+struct hlist_head;
 
 struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
@@ -18,5 +19,6 @@ struct netns_ipv4 {
 #ifdef CONFIG_IP_MULTIPLE_TABLES
 	struct fib_rules_ops	*rules_ops;
 #endif
+	struct hlist_head	*fib_table_hash;
 };
 #endif
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 90aa05f47f8e..7fe54a3c168b 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -50,7 +50,6 @@
 #define FFprint(a...) printk(KERN_DEBUG a)
 
 static struct sock *fibnl;
-struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 
@@ -67,9 +66,9 @@ static int __net_init fib4_rules_init(struct net *net)
 		goto fail;
 
 	hlist_add_head_rcu(&local_table->tb_hlist,
-				&fib_table_hash[TABLE_LOCAL_INDEX]);
+				&net->ipv4.fib_table_hash[TABLE_LOCAL_INDEX]);
 	hlist_add_head_rcu(&main_table->tb_hlist,
-				&fib_table_hash[TABLE_MAIN_INDEX]);
+				&net->ipv4.fib_table_hash[TABLE_MAIN_INDEX]);
 	return 0;
 
 fail:
@@ -92,7 +91,7 @@ struct fib_table *fib_new_table(struct net *net, u32 id)
 	if (!tb)
 		return NULL;
 	h = id & (FIB_TABLE_HASHSZ - 1);
-	hlist_add_head_rcu(&tb->tb_hlist, &fib_table_hash[h]);
+	hlist_add_head_rcu(&tb->tb_hlist, &net->ipv4.fib_table_hash[h]);
 	return tb;
 }
 
@@ -100,13 +99,16 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
 {
 	struct fib_table *tb;
 	struct hlist_node *node;
+	struct hlist_head *head;
 	unsigned int h;
 
 	if (id == 0)
 		id = RT_TABLE_MAIN;
 	h = id & (FIB_TABLE_HASHSZ - 1);
+
 	rcu_read_lock();
-	hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb_hlist) {
+	head = &net->ipv4.fib_table_hash[h];
+	hlist_for_each_entry_rcu(tb, node, head, tb_hlist) {
 		if (tb->tb_id == id) {
 			rcu_read_unlock();
 			return tb;
@@ -117,15 +119,17 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
 }
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
-static void fib_flush(void)
+static void fib_flush(struct net *net)
 {
 	int flushed = 0;
 	struct fib_table *tb;
 	struct hlist_node *node;
+	struct hlist_head *head;
 	unsigned int h;
 
 	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
-		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb_hlist)
+		head = &net->ipv4.fib_table_hash[h];
+		hlist_for_each_entry(tb, node, head, tb_hlist)
 			flushed += tb->tb_flush(tb);
 	}
 
@@ -620,6 +624,7 @@ static int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 	unsigned int e = 0, s_e;
 	struct fib_table *tb;
 	struct hlist_node *node;
+	struct hlist_head *head;
 	int dumped = 0;
 
 	if (net != &init_net)
@@ -634,7 +639,8 @@ static int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 
 	for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) {
 		e = 0;
-		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb_hlist) {
+		head = &net->ipv4.fib_table_hash[h];
+		hlist_for_each_entry(tb, node, head, tb_hlist) {
 			if (e < s_e)
 				goto next;
 			if (dumped)
@@ -797,7 +803,7 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
 			   for stray nexthop entries, then ignite fib_flush.
 			*/
 			if (fib_sync_down(ifa->ifa_local, NULL, 0))
-				fib_flush();
+				fib_flush(&init_net);
 		}
 	}
 #undef LOCAL_OK
@@ -882,7 +888,7 @@ static void nl_fib_lookup_exit(struct net *net)
 static void fib_disable_ip(struct net_device *dev, int force)
 {
 	if (fib_sync_down(0, dev, force))
-		fib_flush();
+		fib_flush(&init_net);
 	rt_cache_flush(0);
 	arp_ifdown(dev);
 }
@@ -963,8 +969,13 @@ static int __net_init ip_fib_net_init(struct net *net)
 {
 	unsigned int i;
 
+	net->ipv4.fib_table_hash = kzalloc(
+			sizeof(struct hlist_head)*FIB_TABLE_HASHSZ, GFP_KERNEL);
+	if (net->ipv4.fib_table_hash == NULL)
+		return -ENOMEM;
+
 	for (i = 0; i < FIB_TABLE_HASHSZ; i++)
-		INIT_HLIST_HEAD(&fib_table_hash[i]);
+		INIT_HLIST_HEAD(&net->ipv4.fib_table_hash[i]);
 
 	return fib4_rules_init(net);
 }
@@ -982,13 +993,14 @@ static void __net_exit ip_fib_net_exit(struct net *net)
 		struct hlist_head *head;
 		struct hlist_node *node, *tmp;
 
-		head = &fib_table_hash[i];
+		head = &net->ipv4.fib_table_hash[i];
 		hlist_for_each_entry_safe(tb, node, tmp, head, tb_hlist) {
 			hlist_del(node);
 			tb->tb_flush(tb);
 			kfree(tb);
 		}
 	}
+	kfree(net->ipv4.fib_table_hash);
 }
 
 static int __net_init fib_net_init(struct net *net)
-- 
cgit v1.2.3


From 6bd48fcf73019219495f7599028296c65b749bb4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:28:55 -0800
Subject: [NETNS]: Provide correct namespace for fibnl netlink socket.

This patch makes the netlink socket to be per namespace. That allows
to have each namespace its own socket for routing queries.

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  2 ++
 net/ipv4/fib_frontend.c  | 24 ++++++++++++++++--------
 2 files changed, 18 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 2dd31937d022..3872aa7e294b 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -9,6 +9,7 @@ struct ctl_table_header;
 struct ipv4_devconf;
 struct fib_rules_ops;
 struct hlist_head;
+struct sock;
 
 struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
@@ -20,5 +21,6 @@ struct netns_ipv4 {
 	struct fib_rules_ops	*rules_ops;
 #endif
 	struct hlist_head	*fib_table_hash;
+	struct sock		*fibnl;
 };
 #endif
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 7fe54a3c168b..a5e81677a2b5 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -49,8 +49,6 @@
 
 #define FFprint(a...) printk(KERN_DEBUG a)
 
-static struct sock *fibnl;
-
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 
 static int __net_init fib4_rules_init(struct net *net)
@@ -845,11 +843,13 @@ static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb )
 
 static void nl_fib_input(struct sk_buff *skb)
 {
+	struct net *net;
 	struct fib_result_nl *frn;
 	struct nlmsghdr *nlh;
 	struct fib_table *tb;
 	u32 pid;
 
+	net = skb->sk->sk_net;
 	nlh = nlmsg_hdr(skb);
 	if (skb->len < NLMSG_SPACE(0) || skb->len < nlh->nlmsg_len ||
 	    nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*frn)))
@@ -861,28 +861,36 @@ static void nl_fib_input(struct sk_buff *skb)
 	nlh = nlmsg_hdr(skb);
 
 	frn = (struct fib_result_nl *) NLMSG_DATA(nlh);
-	tb = fib_get_table(&init_net, frn->tb_id_in);
+	tb = fib_get_table(net, frn->tb_id_in);
 
 	nl_fib_lookup(frn, tb);
 
 	pid = NETLINK_CB(skb).pid;       /* pid of sending process */
 	NETLINK_CB(skb).pid = 0;         /* from kernel */
 	NETLINK_CB(skb).dst_group = 0;  /* unicast */
-	netlink_unicast(fibnl, skb, pid, MSG_DONTWAIT);
+	netlink_unicast(net->ipv4.fibnl, skb, pid, MSG_DONTWAIT);
 }
 
 static int nl_fib_lookup_init(struct net *net)
 {
-	fibnl = netlink_kernel_create(net, NETLINK_FIB_LOOKUP, 0,
-				      nl_fib_input, NULL, THIS_MODULE);
-	if (fibnl == NULL)
+	struct sock *sk;
+	sk = netlink_kernel_create(net, NETLINK_FIB_LOOKUP, 0,
+				   nl_fib_input, NULL, THIS_MODULE);
+	if (sk == NULL)
 		return -EAFNOSUPPORT;
+	/* Don't hold an extra reference on the namespace */
+	put_net(sk->sk_net);
+	net->ipv4.fibnl = sk;
 	return 0;
 }
 
 static void nl_fib_lookup_exit(struct net *net)
 {
-	sock_put(fibnl);
+	/* At the last minute lie and say this is a socket for the
+	 * initial network namespace. So the socket will  be safe to free.
+	 */
+	net->ipv4.fibnl->sk_net = get_net(&init_net);
+	sock_put(net->ipv4.fibnl);
 }
 
 static void fib_disable_ip(struct net_device *dev, int force)
-- 
cgit v1.2.3


From 1bad118a330d494b23663fce94d4e9d9d5065fa7 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:29:53 -0800
Subject: [NETNS]: Pass namespace through ip_rt_ioctl.

... up to rtentry_to_fib_config

Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h     | 2 +-
 net/ipv4/af_inet.c      | 2 +-
 net/ipv4/fib_frontend.c | 8 ++++----
 net/ipv4/ipconfig.c     | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index b77700059252..5847e6fabe8e 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -120,7 +120,7 @@ extern void		ip_rt_send_redirect(struct sk_buff *skb);
 extern unsigned		inet_addr_type(struct net *net, __be32 addr);
 extern unsigned		inet_dev_addr_type(struct net *net, const struct net_device *dev, __be32 addr);
 extern void		ip_rt_multicast_event(struct in_device *);
-extern int		ip_rt_ioctl(unsigned int cmd, void __user *arg);
+extern int		ip_rt_ioctl(struct net *, unsigned int cmd, void __user *arg);
 extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
 extern int		ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb);
 
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index d76f8d8d5eba..bcf8c8a4a3ad 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -795,7 +795,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		case SIOCADDRT:
 		case SIOCDELRT:
 		case SIOCRTMSG:
-			err = ip_rt_ioctl(cmd, (void __user *)arg);
+			err = ip_rt_ioctl(sk->sk_net, cmd, (void __user *)arg);
 			break;
 		case SIOCDARP:
 		case SIOCGARP:
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 8b44e3195242..15909a9fa9b2 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -437,7 +437,7 @@ static int rtentry_to_fib_config(struct net *net, int cmd, struct rtentry *rt,
  *	Handle IP routing ioctl calls. These are used to manipulate the routing tables
  */
 
-int ip_rt_ioctl(unsigned int cmd, void __user *arg)
+int ip_rt_ioctl(struct net *net, unsigned int cmd, void __user *arg)
 {
 	struct fib_config cfg;
 	struct rtentry rt;
@@ -453,18 +453,18 @@ int ip_rt_ioctl(unsigned int cmd, void __user *arg)
 			return -EFAULT;
 
 		rtnl_lock();
-		err = rtentry_to_fib_config(&init_net, cmd, &rt, &cfg);
+		err = rtentry_to_fib_config(net, cmd, &rt, &cfg);
 		if (err == 0) {
 			struct fib_table *tb;
 
 			if (cmd == SIOCDELRT) {
-				tb = fib_get_table(&init_net, cfg.fc_table);
+				tb = fib_get_table(net, cfg.fc_table);
 				if (tb)
 					err = tb->tb_delete(tb, &cfg);
 				else
 					err = -ESRCH;
 			} else {
-				tb = fib_new_table(&init_net, cfg.fc_table);
+				tb = fib_new_table(net, cfg.fc_table);
 				if (tb)
 					err = tb->tb_insert(tb, &cfg);
 				else
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 4bc3a57c183d..a52b5853aaa8 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -302,7 +302,7 @@ static int __init ic_route_ioctl(unsigned int cmd, struct rtentry *arg)
 
 	mm_segment_t oldfs = get_fs();
 	set_fs(get_ds());
-	res = ip_rt_ioctl(cmd, (void __user *) arg);
+	res = ip_rt_ioctl(&init_net, cmd, (void __user *) arg);
 	set_fs(oldfs);
 	return res;
 }
-- 
cgit v1.2.3


From e5d69b9f4a6ce17f0d09595da45e37b870fee5ae Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:51:41 -0800
Subject: [ATM]: Oops reading net/atm/arp

cat /proc/net/atm/arp causes the NULL pointer dereference in the
get_proc_net+0xc/0x3a. This happens as proc_get_net believes that the
parent proc dir entry contains struct net.

Fix this assumption for "net/atm" case.

The problem is introduced by the commit c0097b07abf5f92ab135d024dd41bd2aada1512f
from Eric W. Biederman/Daniel Lezcano.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c      | 17 +++++++++++++----
 include/linux/proc_fs.h |  2 ++
 net/atm/proc.c          |  4 ++--
 3 files changed, 17 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index cfc4f6c072f1..4823c9677fac 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -96,6 +96,17 @@ static struct proc_dir_entry *proc_net_shadow(struct task_struct *task,
 	return task->nsproxy->net_ns->proc_net;
 }
 
+struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name,
+		struct proc_dir_entry *parent)
+{
+	struct proc_dir_entry *pde;
+	pde = proc_mkdir_mode(name, S_IRUGO | S_IXUGO, parent);
+	if (pde != NULL)
+		pde->data = net;
+	return pde;
+}
+EXPORT_SYMBOL_GPL(proc_net_mkdir);
+
 static __net_init int proc_net_ns_init(struct net *net)
 {
 	struct proc_dir_entry *root, *netd, *net_statd;
@@ -107,18 +118,16 @@ static __net_init int proc_net_ns_init(struct net *net)
 		goto out;
 
 	err = -EEXIST;
-	netd = proc_mkdir("net", root);
+	netd = proc_net_mkdir(net, "net", root);
 	if (!netd)
 		goto free_root;
 
 	err = -EEXIST;
-	net_statd = proc_mkdir("stat", netd);
+	net_statd = proc_net_mkdir(net, "stat", netd);
 	if (!net_statd)
 		goto free_net;
 
 	root->data = net;
-	netd->data = net;
-	net_statd->data = net;
 
 	net->proc_net_root = root;
 	net->proc_net = netd;
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index a5316829215b..8f92546b403d 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -201,6 +201,8 @@ static inline struct proc_dir_entry *create_proc_info_entry(const char *name,
 extern struct proc_dir_entry *proc_net_fops_create(struct net *net,
 	const char *name, mode_t mode, const struct file_operations *fops);
 extern void proc_net_remove(struct net *net, const char *name);
+extern struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name,
+	struct proc_dir_entry *parent);
 
 #else
 
diff --git a/net/atm/proc.c b/net/atm/proc.c
index 5d9d5ffba145..565e75e62ca4 100644
--- a/net/atm/proc.c
+++ b/net/atm/proc.c
@@ -476,7 +476,7 @@ static void atm_proc_dirs_remove(void)
 		if (e->dirent)
 			remove_proc_entry(e->name, atm_proc_root);
 	}
-	remove_proc_entry("atm", init_net.proc_net);
+	proc_net_remove(&init_net, "atm");
 }
 
 int __init atm_proc_init(void)
@@ -484,7 +484,7 @@ int __init atm_proc_init(void)
 	static struct atm_proc_entry *e;
 	int ret;
 
-	atm_proc_root = proc_mkdir("atm", init_net.proc_net);
+	atm_proc_root = proc_net_mkdir(&init_net, "atm", init_net.proc_net);
 	if (!atm_proc_root)
 		goto err_out;
 	for (e = atm_proc_ents; e->name; e++) {
-- 
cgit v1.2.3


From 4250846146c04ac6f17bf92619ddfef6db2cf34f Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 10 Jan 2008 03:53:12 -0800
Subject: [NEIGH]: Make /proc/net/arp opening consistent with seq_net_open
 semantics

seq_open_net requires that first field of the seq->private data to be
struct seq_net_private. In reality this is a single pointer to a
struct net for now. The patch makes code consistent.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 2 +-
 net/core/neighbour.c    | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index a9dda2908211..09f9fc60a779 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -223,7 +223,7 @@ extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct n
 extern void pneigh_for_each(struct neigh_table *tbl, void (*cb)(struct pneigh_entry *));
 
 struct neigh_seq_state {
-	struct net *net;
+	struct seq_net_private p;
 	struct neigh_table *tbl;
 	void *(*neigh_sub_iter)(struct neigh_seq_state *state,
 				struct neighbour *n, loff_t *pos);
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 802493327a87..19c0dd12fd58 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -2142,7 +2142,7 @@ EXPORT_SYMBOL(__neigh_for_each_release);
 static struct neighbour *neigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net *net = state->net;
+	struct net *net = state->p.net;
 	struct neigh_table *tbl = state->tbl;
 	struct neighbour *n = NULL;
 	int bucket = state->bucket;
@@ -2183,7 +2183,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 					loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net *net = state->net;
+	struct net *net = state->p.net;
 	struct neigh_table *tbl = state->tbl;
 
 	if (state->neigh_sub_iter) {
@@ -2243,7 +2243,7 @@ static struct neighbour *neigh_get_idx(struct seq_file *seq, loff_t *pos)
 static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net * net = state->net;
+	struct net * net = state->p.net;
 	struct neigh_table *tbl = state->tbl;
 	struct pneigh_entry *pn = NULL;
 	int bucket = state->bucket;
@@ -2266,7 +2266,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 					    loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net * net = state->net;
+	struct net * net = state->p.net;
 	struct neigh_table *tbl = state->tbl;
 
 	pn = pn->next;
-- 
cgit v1.2.3


From e0da5a480cafc7ca228d6b5a05dbd77344a6bd29 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 10 Jan 2008 17:42:55 -0800
Subject: [NETNS]: Create ipv6 devconf-s for namespaces

This is the core. Declare and register the pernet subsys for
addrconf. The init callback the will create the devconf-s.

The init_net will reuse the existing statically declared confs,
so that accessing them from inside the ipv6 code will still
work.

The register_pernet_subsys() is moved above the ipv6_add_dev()
call for loopback, because this function will need the
net->devconf_dflt pointer to be already set.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  2 ++
 net/ipv6/addrconf.c      | 82 +++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 72 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 10733a6f1bd4..06b4dc034bbd 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -28,5 +28,7 @@ struct netns_sysctl_ipv6 {
 
 struct netns_ipv6 {
 	struct netns_sysctl_ipv6 sysctl;
+	struct ipv6_devconf	*devconf_all;
+	struct ipv6_devconf	*devconf_dflt;
 };
 #endif
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index bde50c686722..3ad081e9366b 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4135,6 +4135,70 @@ static void addrconf_sysctl_unregister(struct inet6_dev *idev)
 
 #endif
 
+static int addrconf_init_net(struct net *net)
+{
+	int err;
+	struct ipv6_devconf *all, *dflt;
+
+	err = -ENOMEM;
+	all = &ipv6_devconf;
+	dflt = &ipv6_devconf_dflt;
+
+	if (net != &init_net) {
+		all = kmemdup(all, sizeof(ipv6_devconf), GFP_KERNEL);
+		if (all == NULL)
+			goto err_alloc_all;
+
+		dflt = kmemdup(dflt, sizeof(ipv6_devconf_dflt), GFP_KERNEL);
+		if (dflt == NULL)
+			goto err_alloc_dflt;
+	}
+
+	net->ipv6.devconf_all = all;
+	net->ipv6.devconf_dflt = dflt;
+
+#ifdef CONFIG_SYSCTL
+	err = __addrconf_sysctl_register(net, "all", NET_PROTO_CONF_ALL,
+			NULL, all);
+	if (err < 0)
+		goto err_reg_all;
+
+	err = __addrconf_sysctl_register(net, "default", NET_PROTO_CONF_DEFAULT,
+			NULL, dflt);
+	if (err < 0)
+		goto err_reg_dflt;
+#endif
+	return 0;
+
+#ifdef CONFIG_SYSCTL
+err_reg_dflt:
+	__addrconf_sysctl_unregister(all);
+err_reg_all:
+	kfree(dflt);
+#endif
+err_alloc_dflt:
+	kfree(all);
+err_alloc_all:
+	return err;
+}
+
+static void addrconf_exit_net(struct net *net)
+{
+#ifdef CONFIG_SYSCTL
+	__addrconf_sysctl_unregister(net->ipv6.devconf_dflt);
+	__addrconf_sysctl_unregister(net->ipv6.devconf_all);
+#endif
+	if (net != &init_net) {
+		kfree(net->ipv6.devconf_dflt);
+		kfree(net->ipv6.devconf_all);
+	}
+}
+
+static struct pernet_operations addrconf_ops = {
+	.init = addrconf_init_net,
+	.exit = addrconf_exit_net,
+};
+
 /*
  *      Device notifier
  */
@@ -4167,6 +4231,8 @@ int __init addrconf_init(void)
 		return err;
 	}
 
+	register_pernet_subsys(&addrconf_ops);
+
 	/* The addrconf netdev notifier requires that loopback_dev
 	 * has it's ipv6 private information allocated and setup
 	 * before it can bring up and give link-local addresses
@@ -4190,7 +4256,7 @@ int __init addrconf_init(void)
 		err = -ENOMEM;
 	rtnl_unlock();
 	if (err)
-		return err;
+		goto errlo;
 
 	ip6_null_entry.u.dst.dev = init_net.loopback_dev;
 	ip6_null_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev);
@@ -4218,16 +4284,11 @@ int __init addrconf_init(void)
 
 	ipv6_addr_label_rtnl_register();
 
-#ifdef CONFIG_SYSCTL
-	__addrconf_sysctl_register(&init_net, "all", NET_PROTO_CONF_ALL,
-			NULL, &ipv6_devconf);
-	__addrconf_sysctl_register(&init_net, "default", NET_PROTO_CONF_DEFAULT,
-			NULL, &ipv6_devconf_dflt);
-#endif
-
 	return 0;
 errout:
 	unregister_netdevice_notifier(&ipv6_dev_notf);
+errlo:
+	unregister_pernet_subsys(&addrconf_ops);
 
 	return err;
 }
@@ -4240,10 +4301,7 @@ void addrconf_cleanup(void)
 
 	unregister_netdevice_notifier(&ipv6_dev_notf);
 
-#ifdef CONFIG_SYSCTL
-	__addrconf_sysctl_unregister(&ipv6_devconf_dflt);
-	__addrconf_sysctl_unregister(&ipv6_devconf);
-#endif
+	unregister_pernet_subsys(&addrconf_ops);
 
 	rtnl_lock();
 
-- 
cgit v1.2.3


From 9993e7d313e80bdc005d09c7def91903e0068f07 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 10 Jan 2008 21:56:38 -0800
Subject: [TCP]: Do not purge sk_forward_alloc entirely in tcp_delack_timer().

Otherwise we beat heavily on the global tcp_memory atomics
when all of the sockets in the system are slowly sending
perioding packet clumps.

Noticed and suggested by Eric Dumazet.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h   | 8 ++++++++
 net/ipv4/tcp_timer.c | 2 +-
 2 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 786fae858e77..902324488d0f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -759,6 +759,14 @@ static inline void sk_mem_reclaim(struct sock *sk)
 		__sk_mem_reclaim(sk);
 }
 
+static inline void sk_mem_reclaim_partial(struct sock *sk)
+{
+	if (!sk_has_account(sk))
+		return;
+	if (sk->sk_forward_alloc > SK_MEM_QUANTUM)
+		__sk_mem_reclaim(sk);
+}
+
 static inline void sk_mem_charge(struct sock *sk, int size)
 {
 	if (!sk_has_account(sk))
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 17931be6d584..803d758a2b12 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -186,7 +186,7 @@ static void tcp_delack_timer(unsigned long data)
 		goto out_unlock;
 	}
 
-	sk_mem_reclaim(sk);
+	sk_mem_reclaim_partial(sk);
 
 	if (sk->sk_state == TCP_CLOSE || !(icsk->icsk_ack.pending & ICSK_ACK_TIMER))
 		goto out;
-- 
cgit v1.2.3


From 39971554d3469b48154f34257ad4a35081b7ee40 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 10 Jan 2008 22:37:16 -0800
Subject: [NEIGH]: Add a comment describing what a NUD stands for.

When I studied the neighbor code I puzzled over what the NUD can mean
for quite a long time.

Finally I asked Alexey and he said that this was smth like "neighbor
unreachability detection".

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 09f9fc60a779..a0d42a54cfb2 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -26,6 +26,10 @@
 #include <linux/sysctl.h>
 #include <net/rtnetlink.h>
 
+/*
+ * NUD stands for "neighbor unreachability detection"
+ */
+
 #define NUD_IN_TIMER	(NUD_INCOMPLETE|NUD_REACHABLE|NUD_DELAY|NUD_PROBE)
 #define NUD_VALID	(NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE|NUD_PROBE|NUD_STALE|NUD_DELAY)
 #define NUD_CONNECTED	(NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE)
-- 
cgit v1.2.3


From bfeade087005278fc8cafe230b7658a4f40c5acb Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 22:43:18 -0800
Subject: [NETNS][IPV6]: inet6_addr - check ipv6 address per namespace

When a new address is added, we must check if the new address does not
already exists.  This patch makes this check to be aware of a network
namespace, so the check will look if the address already exists for
the specified network namespace. While the addresses are browsed, the
addresses which do not belong to the namespace are discarded.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 4 +++-
 net/ipv6/addrconf.c    | 6 ++++--
 net/ipv6/af_inet6.c    | 3 ++-
 net/ipv6/anycast.c     | 2 +-
 net/ipv6/datagram.c    | 3 ++-
 net/ipv6/icmp.c        | 2 +-
 net/ipv6/ip6_tunnel.c  | 8 ++++----
 net/ipv6/ndisc.c       | 2 +-
 net/ipv6/raw.c         | 3 ++-
 net/sctp/ipv6.c        | 5 +++--
 10 files changed, 23 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 1c3a5602990e..d1697b587a1a 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -59,9 +59,11 @@ extern int			addrconf_add_ifaddr(void __user *arg);
 extern int			addrconf_del_ifaddr(void __user *arg);
 extern int			addrconf_set_dstaddr(void __user *arg);
 
-extern int			ipv6_chk_addr(struct in6_addr *addr,
+extern int			ipv6_chk_addr(struct net *net,
+					      struct in6_addr *addr,
 					      struct net_device *dev,
 					      int strict);
+
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
 extern int			ipv6_chk_home_addr(struct in6_addr *addr);
 #endif
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index d7b440343e97..f35c3df410df 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1206,13 +1206,16 @@ static int ipv6_count_addresses(struct inet6_dev *idev)
 	return cnt;
 }
 
-int ipv6_chk_addr(struct in6_addr *addr, struct net_device *dev, int strict)
+int ipv6_chk_addr(struct net *net, struct in6_addr *addr,
+		  struct net_device *dev, int strict)
 {
 	struct inet6_ifaddr * ifp;
 	u8 hash = ipv6_addr_hash(addr);
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
+		if (ifp->idev->dev->nd_net != net)
+			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr) &&
 		    !(ifp->flags&IFA_F_TENTATIVE)) {
 			if (dev == NULL || ifp->idev->dev == dev ||
@@ -1223,7 +1226,6 @@ int ipv6_chk_addr(struct in6_addr *addr, struct net_device *dev, int strict)
 	read_unlock_bh(&addrconf_hash_lock);
 	return ifp != NULL;
 }
-
 EXPORT_SYMBOL(ipv6_chk_addr);
 
 static
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index ac8772dd968f..3150c4be3c0c 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -314,7 +314,8 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 			 */
 			v4addr = LOOPBACK4_IPV6;
 			if (!(addr_type & IPV6_ADDR_MULTICAST))	{
-				if (!ipv6_chk_addr(&addr->sin6_addr, dev, 0)) {
+				if (!ipv6_chk_addr(&init_net, &addr->sin6_addr,
+						   dev, 0)) {
 					if (dev)
 						dev_put(dev);
 					err = -EADDRNOTAVAIL;
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 5c4190060e75..9c7f83fbc3a1 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -89,7 +89,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 		return -EPERM;
 	if (ipv6_addr_is_multicast(addr))
 		return -EINVAL;
-	if (ipv6_chk_addr(addr, NULL, 0))
+	if (ipv6_chk_addr(&init_net, addr, NULL, 0))
 		return -EINVAL;
 
 	pac = sock_kmalloc(sk, sizeof(struct ipv6_ac_socklist), GFP_KERNEL);
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index f49a06aa97de..94fa6ae77cfe 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -549,7 +549,8 @@ int datagram_send_ctl(struct msghdr *msg, struct flowi *fl,
 						return -ENODEV;
 				}
 			}
-			if (!ipv6_chk_addr(&src_info->ipi6_addr, dev, 0)) {
+			if (!ipv6_chk_addr(&init_net, &src_info->ipi6_addr,
+					   dev, 0)) {
 				if (dev)
 					dev_put(dev);
 				err = -EINVAL;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 5395afe55ca5..cbb5b9cf84ad 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -332,7 +332,7 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 	 */
 	addr_type = ipv6_addr_type(&hdr->daddr);
 
-	if (ipv6_chk_addr(&hdr->daddr, skb->dev, 0))
+	if (ipv6_chk_addr(&init_net, &hdr->daddr, skb->dev, 0))
 		saddr = &hdr->daddr;
 
 	/*
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 29b5321e39c3..425c9ae8b315 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -653,8 +653,8 @@ static inline int ip6_tnl_rcv_ctl(struct ip6_tnl *t)
 			ldev = dev_get_by_index(&init_net, p->link);
 
 		if ((ipv6_addr_is_multicast(&p->laddr) ||
-		     likely(ipv6_chk_addr(&p->laddr, ldev, 0))) &&
-		    likely(!ipv6_chk_addr(&p->raddr, NULL, 0)))
+		     likely(ipv6_chk_addr(&init_net, &p->laddr, ldev, 0))) &&
+		    likely(!ipv6_chk_addr(&init_net, &p->raddr, NULL, 0)))
 			ret = 1;
 
 		if (ldev)
@@ -788,12 +788,12 @@ static inline int ip6_tnl_xmit_ctl(struct ip6_tnl *t)
 		if (p->link)
 			ldev = dev_get_by_index(&init_net, p->link);
 
-		if (unlikely(!ipv6_chk_addr(&p->laddr, ldev, 0)))
+		if (unlikely(!ipv6_chk_addr(&init_net, &p->laddr, ldev, 0)))
 			printk(KERN_WARNING
 			       "%s xmit: Local address not yet configured!\n",
 			       p->name);
 		else if (!ipv6_addr_is_multicast(&p->raddr) &&
-			 unlikely(ipv6_chk_addr(&p->raddr, NULL, 0)))
+			 unlikely(ipv6_chk_addr(&init_net, &p->raddr, NULL, 0)))
 			printk(KERN_WARNING
 			       "%s xmit: Routing loop! "
 			       "Remote address found on this node!\n",
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index b66a1f81bd83..e217d3ff00f3 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -653,7 +653,7 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb)
 	struct in6_addr *target = (struct in6_addr *)&neigh->primary_key;
 	int probes = atomic_read(&neigh->probes);
 
-	if (skb && ipv6_chk_addr(&ipv6_hdr(skb)->saddr, dev, 1))
+	if (skb && ipv6_chk_addr(&init_net, &ipv6_hdr(skb)->saddr, dev, 1))
 		saddr = &ipv6_hdr(skb)->saddr;
 
 	if ((probes -= neigh->parms->ucast_probes) < 0) {
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 45a580e843dc..cb0b110a2ac8 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -298,7 +298,8 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 		v4addr = LOOPBACK4_IPV6;
 		if (!(addr_type & IPV6_ADDR_MULTICAST))	{
 			err = -EADDRNOTAVAIL;
-			if (!ipv6_chk_addr(&addr->sin6_addr, dev, 0)) {
+			if (!ipv6_chk_addr(&init_net, &addr->sin6_addr,
+					   dev, 0)) {
 				if (dev)
 					dev_put(dev);
 				goto out;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index bd04aed673cb..74f106a7a7e9 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -556,7 +556,7 @@ static int sctp_v6_available(union sctp_addr *addr, struct sctp_sock *sp)
 	if (!(type & IPV6_ADDR_UNICAST))
 		return 0;
 
-	return ipv6_chk_addr(in6, NULL, 0);
+	return ipv6_chk_addr(&init_net, in6, NULL, 0);
 }
 
 /* This function checks if the address is a valid address to be used for
@@ -858,7 +858,8 @@ static int sctp_inet6_bind_verify(struct sctp_sock *opt, union sctp_addr *addr)
 			dev = dev_get_by_index(&init_net, addr->v6.sin6_scope_id);
 			if (!dev)
 				return 0;
-			if (!ipv6_chk_addr(&addr->v6.sin6_addr, dev, 0)) {
+			if (!ipv6_chk_addr(&init_net, &addr->v6.sin6_addr,
+					   dev, 0)) {
 				dev_put(dev);
 				return 0;
 			}
-- 
cgit v1.2.3


From 1cab3da6be6c7659f62d0d297b389cc0e48b2178 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 22:44:09 -0800
Subject: [NETNS][IPV6]: inet6_addr - ipv6_get_ifaddr namespace aware

The inet6_addr_lst is browsed taking into account the network
namespace specified as parameter. If an address does not belong
to the specified namespace, it is ignored.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h |  8 +++++---
 net/ipv6/addrconf.c    | 11 +++++++----
 net/ipv6/ip6_output.c  |  3 ++-
 net/ipv6/ndisc.c       |  9 +++++----
 4 files changed, 19 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index d1697b587a1a..77f630f12806 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -67,9 +67,11 @@ extern int			ipv6_chk_addr(struct net *net,
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
 extern int			ipv6_chk_home_addr(struct in6_addr *addr);
 #endif
-extern struct inet6_ifaddr *	ipv6_get_ifaddr(struct in6_addr *addr,
-						struct net_device *dev,
-						int strict);
+extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
+						 struct in6_addr *addr,
+						 struct net_device *dev,
+						 int strict);
+
 extern int			ipv6_get_saddr(struct dst_entry *dst, 
 					       struct in6_addr *daddr,
 					       struct in6_addr *saddr);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 41cc31ee297e..c4df6cdff650 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1247,13 +1247,16 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr,
 	return ifp != NULL;
 }
 
-struct inet6_ifaddr * ipv6_get_ifaddr(struct in6_addr *addr, struct net_device *dev, int strict)
+struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr,
+				     struct net_device *dev, int strict)
 {
 	struct inet6_ifaddr * ifp;
 	u8 hash = ipv6_addr_hash(addr);
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
+		if (ifp->idev->dev->nd_net != net)
+			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr)) {
 			if (dev == NULL || ifp->idev->dev == dev ||
 			    !(ifp->scope&(IFA_LINK|IFA_HOST) || strict)) {
@@ -1739,7 +1742,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 
 ok:
 
-		ifp = ipv6_get_ifaddr(&addr, dev, 1);
+		ifp = ipv6_get_ifaddr(&init_net, &addr, dev, 1);
 
 		if (ifp == NULL && valid_lft) {
 			int max_addresses = in6_dev->cnf.max_addresses;
@@ -3135,7 +3138,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	/* We ignore other flags so far. */
 	ifa_flags = ifm->ifa_flags & (IFA_F_NODAD | IFA_F_HOMEADDRESS);
 
-	ifa = ipv6_get_ifaddr(pfx, dev, 1);
+	ifa = ipv6_get_ifaddr(net, pfx, dev, 1);
 	if (ifa == NULL) {
 		/*
 		 * It would be best to check for !NLM_F_CREATE here but
@@ -3442,7 +3445,7 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 	if (ifm->ifa_index)
 		dev = __dev_get_by_index(&init_net, ifm->ifa_index);
 
-	if ((ifa = ipv6_get_ifaddr(addr, dev, 1)) == NULL) {
+	if ((ifa = ipv6_get_ifaddr(net, addr, dev, 1)) == NULL) {
 		err = -EADDRNOTAVAIL;
 		goto errout;
 	}
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index ba7c8aaf2782..15c4f6cee3e6 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -936,7 +936,8 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 			struct flowi fl_gw;
 			int redirect;
 
-			ifp = ipv6_get_ifaddr(&fl->fl6_src, (*dst)->dev, 1);
+			ifp = ipv6_get_ifaddr(&init_net, &fl->fl6_src,
+					      (*dst)->dev, 1);
 
 			redirect = (ifp && ifp->flags & IFA_F_OPTIMISTIC);
 			if (ifp)
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index e217d3ff00f3..bdfc4ea61941 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -556,7 +556,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
 	};
 
 	/* for anycast or proxy, solicited_addr != src_addr */
-	ifp = ipv6_get_ifaddr(solicited_addr, dev, 1);
+	ifp = ipv6_get_ifaddr(&init_net, solicited_addr, dev, 1);
 	if (ifp) {
 		src_addr = solicited_addr;
 		if (ifp->flags & IFA_F_OPTIMISTIC)
@@ -616,7 +616,8 @@ void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr,
 	 * suppress the inclusion of the sllao.
 	 */
 	if (send_sllao) {
-		struct inet6_ifaddr *ifp = ipv6_get_ifaddr(saddr, dev, 1);
+		struct inet6_ifaddr *ifp = ipv6_get_ifaddr(&init_net, saddr,
+							   dev, 1);
 		if (ifp) {
 			if (ifp->flags & IFA_F_OPTIMISTIC)  {
 				send_sllao = 0;
@@ -741,7 +742,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 
 	inc = ipv6_addr_is_multicast(daddr);
 
-	if ((ifp = ipv6_get_ifaddr(&msg->target, dev, 1)) != NULL) {
+	if ((ifp = ipv6_get_ifaddr(&init_net, &msg->target, dev, 1)) != NULL) {
 
 		if (ifp->flags & (IFA_F_TENTATIVE|IFA_F_OPTIMISTIC)) {
 			if (dad) {
@@ -899,7 +900,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 			return;
 		}
 	}
-	if ((ifp = ipv6_get_ifaddr(&msg->target, dev, 1))) {
+	if ((ifp = ipv6_get_ifaddr(&init_net, &msg->target, dev, 1))) {
 		if (ifp->flags & IFA_F_TENTATIVE) {
 			addrconf_dad_failure(ifp);
 			return;
-- 
cgit v1.2.3


From 389f661224cdbdf178553fb09a52dc6c8bf86890 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Thu, 10 Jan 2008 22:44:40 -0800
Subject: [NETNS][IPV6]: inet6_addr - make ipv6_chk_home_addr namespace aware

Looks if the address is belonging to the network namespace, otherwise
discard the address for the check.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 3 ++-
 net/ipv6/addrconf.c    | 4 +++-
 net/ipv6/exthdrs.c     | 2 +-
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 77f630f12806..8b1509bfc695 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -65,7 +65,8 @@ extern int			ipv6_chk_addr(struct net *net,
 					      int strict);
 
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
-extern int			ipv6_chk_home_addr(struct in6_addr *addr);
+extern int			ipv6_chk_home_addr(struct net *net,
+						   struct in6_addr *addr);
 #endif
 extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
 						 struct in6_addr *addr,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index c4df6cdff650..803caf1a3890 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2883,13 +2883,15 @@ void if6_proc_exit(void)
 
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
 /* Check if address is a home address configured on any interface. */
-int ipv6_chk_home_addr(struct in6_addr *addr)
+int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr)
 {
 	int ret = 0;
 	struct inet6_ifaddr * ifp;
 	u8 hash = ipv6_addr_hash(addr);
 	read_lock_bh(&addrconf_hash_lock);
 	for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) {
+		if (ifp->idev->dev->nd_net != net)
+			continue;
 		if (ipv6_addr_cmp(&ifp->addr, addr) == 0 &&
 		    (ifp->flags & IFA_F_HOMEADDRESS)) {
 			ret = 1;
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 2df34ed276f1..3cd1c993d52b 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -445,7 +445,7 @@ looped_back:
 			kfree_skb(skb);
 			return -1;
 		}
-		if (!ipv6_chk_home_addr(addr)) {
+		if (!ipv6_chk_home_addr(&init_net, addr)) {
 			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
 					 IPSTATS_MIB_INADDRERRORS);
 			kfree_skb(skb);
-- 
cgit v1.2.3


From 490d5046930276aae50dd16942649bfc626056f7 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Sat, 12 Jan 2008 03:17:20 -0800
Subject: [TCP]: Uninline tcp_set_state
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

net/ipv4/tcp.c:
  tcp_close_state | -226
  tcp_done        | -145
  tcp_close       | -564
  tcp_disconnect  | -141
 4 functions changed, 1076 bytes removed, diff: -1076

net/ipv4/tcp_input.c:
  tcp_fin               |  -86
  tcp_rcv_state_process | -164
 2 functions changed, 250 bytes removed, diff: -250

net/ipv4/tcp_ipv4.c:
  tcp_v4_connect | -209
 1 function changed, 209 bytes removed, diff: -209

net/ipv4/arp.c:
  arp_ignore |   +5
 1 function changed, 5 bytes added, diff: +5

net/ipv6/tcp_ipv6.c:
  tcp_v6_connect | -158
 1 function changed, 158 bytes removed, diff: -158

net/sunrpc/xprtsock.c:
  xs_sendpages |   -2
 1 function changed, 2 bytes removed, diff: -2

net/dccp/ccids/ccid3.c:
  ccid3_update_send_interval |   +7
 1 function changed, 7 bytes added, diff: +7

net/ipv4/tcp.c:
  tcp_set_state | +238
 1 function changed, 238 bytes added, diff: +238

built-in.o:
 12 functions changed, 250 bytes added, 1695 bytes removed, diff: -1445

I've no explanation why some unrelated changes seem to occur
consistently as well (arp_ignore, ccid3_update_send_interval;
I checked the arp_ignore asm and it seems to be due to some
reordered of operation order causing some extra opcodes to be
generated). Still, the benefits are pretty obvious from the
codiff's results.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 35 +----------------------------------
 net/ipv4/tcp.c    | 35 +++++++++++++++++++++++++++++++++++
 2 files changed, 36 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 48081ada92a7..306580ce1e53 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -926,40 +926,7 @@ static const char *statename[]={
 	"Close Wait","Last ACK","Listen","Closing"
 };
 #endif
-
-static inline void tcp_set_state(struct sock *sk, int state)
-{
-	int oldstate = sk->sk_state;
-
-	switch (state) {
-	case TCP_ESTABLISHED:
-		if (oldstate != TCP_ESTABLISHED)
-			TCP_INC_STATS(TCP_MIB_CURRESTAB);
-		break;
-
-	case TCP_CLOSE:
-		if (oldstate == TCP_CLOSE_WAIT || oldstate == TCP_ESTABLISHED)
-			TCP_INC_STATS(TCP_MIB_ESTABRESETS);
-
-		sk->sk_prot->unhash(sk);
-		if (inet_csk(sk)->icsk_bind_hash &&
-		    !(sk->sk_userlocks & SOCK_BINDPORT_LOCK))
-			inet_put_port(&tcp_hashinfo, sk);
-		/* fall through */
-	default:
-		if (oldstate==TCP_ESTABLISHED)
-			TCP_DEC_STATS(TCP_MIB_CURRESTAB);
-	}
-
-	/* Change state AFTER socket is unhashed to avoid closed
-	 * socket sitting in hash tables.
-	 */
-	sk->sk_state = state;
-
-#ifdef STATE_TRACE
-	SOCK_DEBUG(sk, "TCP sk=%p, State %s -> %s\n",sk, statename[oldstate],statename[state]);
-#endif	
-}
+extern void tcp_set_state(struct sock *sk, int state);
 
 extern void tcp_done(struct sock *sk);
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 34085e3a4096..a0d373bd9065 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1652,6 +1652,41 @@ recv_urg:
 	goto out;
 }
 
+void tcp_set_state(struct sock *sk, int state)
+{
+	int oldstate = sk->sk_state;
+
+	switch (state) {
+	case TCP_ESTABLISHED:
+		if (oldstate != TCP_ESTABLISHED)
+			TCP_INC_STATS(TCP_MIB_CURRESTAB);
+		break;
+
+	case TCP_CLOSE:
+		if (oldstate == TCP_CLOSE_WAIT || oldstate == TCP_ESTABLISHED)
+			TCP_INC_STATS(TCP_MIB_ESTABRESETS);
+
+		sk->sk_prot->unhash(sk);
+		if (inet_csk(sk)->icsk_bind_hash &&
+		    !(sk->sk_userlocks & SOCK_BINDPORT_LOCK))
+			inet_put_port(&tcp_hashinfo, sk);
+		/* fall through */
+	default:
+		if (oldstate==TCP_ESTABLISHED)
+			TCP_DEC_STATS(TCP_MIB_CURRESTAB);
+	}
+
+	/* Change state AFTER socket is unhashed to avoid closed
+	 * socket sitting in hash tables.
+	 */
+	sk->sk_state = state;
+
+#ifdef STATE_TRACE
+	SOCK_DEBUG(sk, "TCP sk=%p, State %s -> %s\n",sk, statename[oldstate],statename[state]);
+#endif
+}
+EXPORT_SYMBOL_GPL(tcp_set_state);
+
 /*
  *	State processing on a close. This implements the state shift for
  *	sending our FIN frame. Note that we only send a FIN for some
-- 
cgit v1.2.3


From cea14e0ed6082bd3dff19203d01340de6e7ca067 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Sat, 12 Jan 2008 03:19:12 -0800
Subject: [TCP]: Uninline tcp_is_cwnd_limited
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

net/ipv4/tcp_cong.c:
  tcp_reno_cong_avoid |  -65
 1 function changed, 65 bytes removed, diff: -65

net/ipv4/arp.c:
  arp_ignore |   -5
 1 function changed, 5 bytes removed, diff: -5

net/ipv4/tcp_bic.c:
  bictcp_cong_avoid |  -57
 1 function changed, 57 bytes removed, diff: -57

net/ipv4/tcp_cubic.c:
  bictcp_cong_avoid |  -61
 1 function changed, 61 bytes removed, diff: -61

net/ipv4/tcp_highspeed.c:
  hstcp_cong_avoid |  -63
 1 function changed, 63 bytes removed, diff: -63

net/ipv4/tcp_hybla.c:
  hybla_cong_avoid |  -85
 1 function changed, 85 bytes removed, diff: -85

net/ipv4/tcp_htcp.c:
  htcp_cong_avoid |  -57
 1 function changed, 57 bytes removed, diff: -57

net/ipv4/tcp_veno.c:
  tcp_veno_cong_avoid |  -52
 1 function changed, 52 bytes removed, diff: -52

net/ipv4/tcp_scalable.c:
  tcp_scalable_cong_avoid |  -61
 1 function changed, 61 bytes removed, diff: -61

net/ipv4/tcp_yeah.c:
  tcp_yeah_cong_avoid |  -75
 1 function changed, 75 bytes removed, diff: -75

net/ipv4/tcp_illinois.c:
  tcp_illinois_cong_avoid |  -54
 1 function changed, 54 bytes removed, diff: -54

net/dccp/ccids/ccid3.c:
  ccid3_update_send_interval |   -7
  ccid3_hc_tx_packet_recv    |   +7
 2 functions changed, 7 bytes added, 7 bytes removed, diff: +0

net/ipv4/tcp_cong.c:
  tcp_is_cwnd_limited |  +88
 1 function changed, 88 bytes added, diff: +88

built-in.o:
 14 functions changed, 95 bytes added, 642 bytes removed, diff: -547

...Again some gcc artifacts visible as well.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 22 +---------------------
 net/ipv4/tcp_cong.c | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 306580ce1e53..7de4ea3a04d9 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -786,27 +786,7 @@ static inline u32 tcp_wnd_end(const struct tcp_sock *tp)
 {
 	return tp->snd_una + tp->snd_wnd;
 }
-
-/* RFC2861 Check whether we are limited by application or congestion window
- * This is the inverse of cwnd check in tcp_tso_should_defer
- */
-static inline int tcp_is_cwnd_limited(const struct sock *sk, u32 in_flight)
-{
-	const struct tcp_sock *tp = tcp_sk(sk);
-	u32 left;
-
-	if (in_flight >= tp->snd_cwnd)
-		return 1;
-
-	if (!sk_can_gso(sk))
-		return 0;
-
-	left = tp->snd_cwnd - in_flight;
-	if (sysctl_tcp_tso_win_divisor)
-		return left * sysctl_tcp_tso_win_divisor < tp->snd_cwnd;
-	else
-		return left <= tcp_max_burst(tp);
-}
+extern int tcp_is_cwnd_limited(const struct sock *sk, u32 in_flight);
 
 static inline void tcp_minshall_update(struct tcp_sock *tp, unsigned int mss,
 				       const struct sk_buff *skb)
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 4451750b478c..3a6be23d222f 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -274,6 +274,27 @@ int tcp_set_congestion_control(struct sock *sk, const char *name)
 	return err;
 }
 
+/* RFC2861 Check whether we are limited by application or congestion window
+ * This is the inverse of cwnd check in tcp_tso_should_defer
+ */
+int tcp_is_cwnd_limited(const struct sock *sk, u32 in_flight)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	u32 left;
+
+	if (in_flight >= tp->snd_cwnd)
+		return 1;
+
+	if (!sk_can_gso(sk))
+		return 0;
+
+	left = tp->snd_cwnd - in_flight;
+	if (sysctl_tcp_tso_win_divisor)
+		return left * sysctl_tcp_tso_win_divisor < tp->snd_cwnd;
+	else
+		return left <= tcp_max_burst(tp);
+}
+EXPORT_SYMBOL_GPL(tcp_is_cwnd_limited);
 
 /*
  * Slow start is used when congestion window is less than slow start
-- 
cgit v1.2.3


From ba749ae98d5aa9d2ce9a7facde0deed454f92230 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Sat, 12 Jan 2008 21:30:23 -0800
Subject: [XFRM]: alg_key_len should be unsigned to avoid integer divides

alg_key_len is currently defined as 'signed int'. This unfortunatly
leads to integer divides in several paths.

Converting it to unsigned is safe and saves 208 bytes of text on i386.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 1131eabfaa2a..f8507eed0b79 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -92,7 +92,7 @@ struct xfrm_replay_state
 
 struct xfrm_algo {
 	char	alg_name[64];
-	int	alg_key_len;    /* in bits */
+	unsigned int	alg_key_len;    /* in bits */
 	char	alg_key[0];
 };
 
-- 
cgit v1.2.3


From 3f4afb6443aaa1d69b2d8f0461c8191e40d54c3c Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sat, 12 Jan 2008 21:31:29 -0800
Subject: [XFRM]: Fix struct xfrm_algo code formatting.

Realign struct members.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index f8507eed0b79..9b5b00c4ef9d 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -91,9 +91,9 @@ struct xfrm_replay_state
 };
 
 struct xfrm_algo {
-	char	alg_name[64];
+	char		alg_name[64];
 	unsigned int	alg_key_len;    /* in bits */
-	char	alg_key[0];
+	char		alg_key[0];
 };
 
 struct xfrm_stats {
-- 
cgit v1.2.3


From f51d599fbecf4881a36466f0355da6b0b346ea70 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jan 2008 05:35:57 -0800
Subject: [NETNS][RAW]: Make /proc/net/raw(6) show per-namespace socket list.

Pull the struct net pointer up to the showing functions
to filter the sockets depending on their namespaces.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h |  3 ++-
 net/ipv4/raw.c    | 20 ++++++++++++--------
 net/ipv6/raw.c    |  4 ++--
 3 files changed, 16 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 4d1aba032bf5..cca81d8b2d8b 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -39,6 +39,7 @@ extern int  raw_proc_init(void);
 extern void raw_proc_exit(void);
 
 struct raw_iter_state {
+	struct seq_net_private p;
 	int bucket;
 	unsigned short family;
 	struct raw_hashinfo *h;
@@ -48,7 +49,7 @@ struct raw_iter_state {
 void *raw_seq_start(struct seq_file *seq, loff_t *pos);
 void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos);
 void raw_seq_stop(struct seq_file *seq, void *v);
-int raw_seq_open(struct file *file, struct raw_hashinfo *h,
+int raw_seq_open(struct inode *ino, struct file *file, struct raw_hashinfo *h,
 		unsigned short family);
 
 #endif
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index a490a9d54712..4e95372a78e7 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -860,7 +860,8 @@ static struct sock *raw_get_first(struct seq_file *seq)
 		struct hlist_node *node;
 
 		sk_for_each(sk, node, &state->h->ht[state->bucket])
-			if (sk->sk_family == state->family)
+			if (sk->sk_net == state->p.net &&
+					sk->sk_family == state->family)
 				goto found;
 	}
 	sk = NULL;
@@ -876,7 +877,8 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && sk->sk_family != state->family);
+	} while (sk && sk->sk_net != state->p.net &&
+			sk->sk_family != state->family);
 
 	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
 		sk = sk_head(&state->h->ht[state->bucket]);
@@ -970,16 +972,18 @@ static const struct seq_operations raw_seq_ops = {
 	.show  = raw_seq_show,
 };
 
-int raw_seq_open(struct file *file, struct raw_hashinfo *h,
+int raw_seq_open(struct inode *ino, struct file *file, struct raw_hashinfo *h,
 		unsigned short family)
 {
+	int err;
 	struct raw_iter_state *i;
 
-	i = __seq_open_private(file, &raw_seq_ops,
+	err = seq_open_net(ino, file, &raw_seq_ops,
 			sizeof(struct raw_iter_state));
-	if (i == NULL)
-		return -ENOMEM;
+	if (err < 0)
+		return err;
 
+	i = raw_seq_private((struct seq_file *)file->private_data);
 	i->h = h;
 	i->family = family;
 	return 0;
@@ -988,7 +992,7 @@ EXPORT_SYMBOL_GPL(raw_seq_open);
 
 static int raw_v4_seq_open(struct inode *inode, struct file *file)
 {
-	return raw_seq_open(file, &raw_v4_hashinfo, PF_INET);
+	return raw_seq_open(inode, file, &raw_v4_hashinfo, PF_INET);
 }
 
 static const struct file_operations raw_seq_fops = {
@@ -996,7 +1000,7 @@ static const struct file_operations raw_seq_fops = {
 	.open	 = raw_v4_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = seq_release_private,
+	.release = seq_release_net,
 };
 
 int __init raw_proc_init(void)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 6f20086064b2..026fa910cb70 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1259,7 +1259,7 @@ static const struct seq_operations raw6_seq_ops = {
 
 static int raw6_seq_open(struct inode *inode, struct file *file)
 {
-	return raw_seq_open(file, &raw_v6_hashinfo, PF_INET6);
+	return raw_seq_open(inode, file, &raw_v6_hashinfo, PF_INET6);
 }
 
 static const struct file_operations raw6_seq_fops = {
@@ -1267,7 +1267,7 @@ static const struct file_operations raw6_seq_fops = {
 	.open =		raw6_seq_open,
 	.read =		seq_read,
 	.llseek =	seq_lseek,
-	.release =	seq_release_private,
+	.release =	seq_release_net,
 };
 
 int __init raw6_proc_init(void)
-- 
cgit v1.2.3


From 72132c1b6c7a7605cdfde86a9f58b6ca2b2195a6 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 14 Jan 2008 22:59:30 -0800
Subject: [IPV4]: fib_rules_unregister is essentially void.

fib_rules_unregister is called only after successful register and the
return code is never checked.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h |  2 +-
 net/core/fib_rules.c    | 21 ++++-----------------
 2 files changed, 5 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index e9a074cf59ed..4f472507129e 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -102,7 +102,7 @@ static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla)
 }
 
 extern int fib_rules_register(struct net *, struct fib_rules_ops *);
-extern int fib_rules_unregister(struct net *, struct fib_rules_ops *);
+extern void fib_rules_unregister(struct net *, struct fib_rules_ops *);
 extern void                     fib_rules_cleanup_ops(struct fib_rules_ops *);
 
 extern int			fib_rules_lookup(struct fib_rules_ops *,
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index c5f78fed6885..541728aa2bae 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -112,29 +112,16 @@ void fib_rules_cleanup_ops(struct fib_rules_ops *ops)
 }
 EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops);
 
-int fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
+void fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
 {
-	int err = 0;
-	struct fib_rules_ops *o;
 
 	spin_lock(&net->rules_mod_lock);
-	list_for_each_entry(o, &net->rules_ops, list) {
-		if (o == ops) {
-			list_del_rcu(&o->list);
-			fib_rules_cleanup_ops(ops);
-			goto out;
-		}
-	}
-
-	err = -ENOENT;
-out:
+	list_del_rcu(&ops->list);
+	fib_rules_cleanup_ops(ops);
 	spin_unlock(&net->rules_mod_lock);
 
 	synchronize_rcu();
-	if (!err)
-		release_net(net);
-
-	return err;
+	release_net(net);
 }
 
 EXPORT_SYMBOL_GPL(fib_rules_unregister);
-- 
cgit v1.2.3


From 06f0511df1b3b32fc8e0840514d4b207150f1fa7 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 24 Jan 2008 00:30:58 -0800
Subject: [ARP]: neigh_parms_put(destroy) are essentially local to
 core/neighbour.c.

Make them static.

[ Moved the inline before, instead of after, call sites. -DaveM ]

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 7 -------
 net/core/neighbour.c    | 9 ++++++++-
 2 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index a0d42a54cfb2..ebbfb509822e 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -213,7 +213,6 @@ extern struct neighbour 	*neigh_event_ns(struct neigh_table *tbl,
 
 extern struct neigh_parms	*neigh_parms_alloc(struct net_device *dev, struct neigh_table *tbl);
 extern void			neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms);
-extern void			neigh_parms_destroy(struct neigh_parms *parms);
 extern unsigned long		neigh_rand_reach_time(unsigned long base);
 
 extern void			pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
@@ -254,12 +253,6 @@ static inline void __neigh_parms_put(struct neigh_parms *parms)
 	atomic_dec(&parms->refcnt);
 }
 
-static inline void neigh_parms_put(struct neigh_parms *parms)
-{
-	if (atomic_dec_and_test(&parms->refcnt))
-		neigh_parms_destroy(parms);
-}
-
 static inline struct neigh_parms *neigh_parms_clone(struct neigh_parms *parms)
 {
 	atomic_inc(&parms->refcnt);
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index a8be5ef13337..a16cf1ec5e5e 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -577,6 +577,13 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev)
 	return -ENOENT;
 }
 
+static void neigh_parms_destroy(struct neigh_parms *parms);
+
+static inline void neigh_parms_put(struct neigh_parms *parms)
+{
+	if (atomic_dec_and_test(&parms->refcnt))
+		neigh_parms_destroy(parms);
+}
 
 /*
  *	neighbour must already be out of the table;
@@ -1350,7 +1357,7 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms)
 	NEIGH_PRINTK1("neigh_parms_release: not found\n");
 }
 
-void neigh_parms_destroy(struct neigh_parms *parms)
+static void neigh_parms_destroy(struct neigh_parms *parms)
 {
 	release_net(parms->net);
 	kfree(parms);
-- 
cgit v1.2.3


From 9bd85e32644d4d3744117b0a196ad4382f8acf35 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 14 Jan 2008 23:05:55 -0800
Subject: [IPV4]: Remove extra argument from arp_ignore.

arp_ignore has two arguments: dev & in_dev. dev is used for
inet_confirm_addr calling only.

inet_confirm_addr, in turn, either gets in_dev from the device passed
or iterates over all network devices if the device passed is NULL. It
seems logical to directly pass in_dev into inet_confirm_addr.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h |  2 +-
 net/ipv4/arp.c             | 11 +++++------
 net/ipv4/devinet.c         | 17 ++++++-----------
 3 files changed, 12 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index b3c5081de02d..45f373107530 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -135,7 +135,7 @@ extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
 extern struct in_device	*inetdev_by_index(int);
 extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
-extern __be32		inet_confirm_addr(const struct net_device *dev, __be32 dst, __be32 local, int scope);
+extern __be32		inet_confirm_addr(struct in_device *in_dev, __be32 dst, __be32 local, int scope);
 extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, __be32 mask);
 
 static __inline__ int inet_ifa_match(__be32 addr, struct in_ifaddr *ifa)
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 357e8987146b..6f0827b2b154 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -382,8 +382,7 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 		read_unlock_bh(&neigh->lock);
 }
 
-static int arp_ignore(struct in_device *in_dev, struct net_device *dev,
-		      __be32 sip, __be32 tip)
+static int arp_ignore(struct in_device *in_dev, __be32 sip, __be32 tip)
 {
 	int scope;
 
@@ -403,7 +402,7 @@ static int arp_ignore(struct in_device *in_dev, struct net_device *dev,
 	case 3:	/* Do not reply for scope host addresses */
 		sip = 0;
 		scope = RT_SCOPE_LINK;
-		dev = NULL;
+		in_dev = NULL;
 		break;
 	case 4:	/* Reserved */
 	case 5:
@@ -415,7 +414,7 @@ static int arp_ignore(struct in_device *in_dev, struct net_device *dev,
 	default:
 		return 0;
 	}
-	return !inet_confirm_addr(dev, sip, tip, scope);
+	return !inet_confirm_addr(in_dev, sip, tip, scope);
 }
 
 static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev)
@@ -807,7 +806,7 @@ static int arp_process(struct sk_buff *skb)
 	if (sip == 0) {
 		if (arp->ar_op == htons(ARPOP_REQUEST) &&
 		    inet_addr_type(&init_net, tip) == RTN_LOCAL &&
-		    !arp_ignore(in_dev,dev,sip,tip))
+		    !arp_ignore(in_dev, sip, tip))
 			arp_send(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip, sha,
 				 dev->dev_addr, sha);
 		goto out;
@@ -825,7 +824,7 @@ static int arp_process(struct sk_buff *skb)
 				int dont_send = 0;
 
 				if (!dont_send)
-					dont_send |= arp_ignore(in_dev,dev,sip,tip);
+					dont_send |= arp_ignore(in_dev,sip,tip);
 				if (!dont_send && IN_DEV_ARPFILTER(in_dev))
 					dont_send |= arp_filter(sip,tip,dev);
 				if (!dont_send)
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 03db15b10309..dc1665a2b075 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -968,24 +968,19 @@ static __be32 confirm_addr_indev(struct in_device *in_dev, __be32 dst,
 
 /*
  * Confirm that local IP address exists using wildcards:
- * - dev: only on this interface, 0=any interface
+ * - in_dev: only on this interface, 0=any interface
  * - dst: only in the same subnet as dst, 0=any dst
  * - local: address, 0=autoselect the local address
  * - scope: maximum allowed scope value for the local address
  */
-__be32 inet_confirm_addr(const struct net_device *dev, __be32 dst, __be32 local, int scope)
+__be32 inet_confirm_addr(struct in_device *in_dev,
+			 __be32 dst, __be32 local, int scope)
 {
 	__be32 addr = 0;
-	struct in_device *in_dev;
-
-	if (dev) {
-		rcu_read_lock();
-		if ((in_dev = __in_dev_get_rcu(dev)))
-			addr = confirm_addr_indev(in_dev, dst, local, scope);
-		rcu_read_unlock();
+	struct net_device *dev;
 
-		return addr;
-	}
+	if (in_dev != NULL)
+		return confirm_addr_indev(in_dev, dst, local, scope);
 
 	read_lock(&dev_base_lock);
 	rcu_read_lock();
-- 
cgit v1.2.3


From 7f9b80529b8a2ad8b3273b15fb444a0e34b760a9 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <stephen.hemminger@vyatta.com>
Date: Mon, 14 Jan 2008 23:14:20 -0800
Subject: [IPV4]: fib hash|trie initialization

Initialization of the slab cache's should be done when IP is
initialized to make sure of available memory, and that code can be
marked __init.

Signed-off-by: Stephen Hemminger <stephen.hemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  5 +++--
 net/ipv4/fib_frontend.c |  9 ++++++---
 net/ipv4/fib_hash.c     | 24 +++++++++++-------------
 net/ipv4/fib_trie.c     | 16 ++++++++--------
 4 files changed, 28 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 025b225181e1..f580542f08ed 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -231,8 +231,9 @@ extern int fib_sync_down(__be32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
 extern __be32  __fib_res_prefsrc(struct fib_result *res);
 
-/* Exported by fib_hash.c */
-extern struct fib_table *fib_hash_init(u32 id);
+/* Exported by fib_{hash|trie}.c */
+extern void fib_hash_init(void);
+extern struct fib_table *fib_hash_table(u32 id);
 
 static inline void fib_combine_itag(u32 *itag, struct fib_result *res)
 {
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 02b5ff73357f..8ddcd3f91a17 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -53,11 +53,11 @@ static int __net_init fib4_rules_init(struct net *net)
 {
 	struct fib_table *local_table, *main_table;
 
-	local_table = fib_hash_init(RT_TABLE_LOCAL);
+	local_table = fib_hash_table(RT_TABLE_LOCAL);
 	if (local_table == NULL)
 		return -ENOMEM;
 
-	main_table  = fib_hash_init(RT_TABLE_MAIN);
+	main_table  = fib_hash_table(RT_TABLE_MAIN);
 	if (main_table == NULL)
 		goto fail;
 
@@ -83,7 +83,8 @@ struct fib_table *fib_new_table(struct net *net, u32 id)
 	tb = fib_get_table(net, id);
 	if (tb)
 		return tb;
-	tb = fib_hash_init(id);
+
+	tb = fib_hash_table(id);
 	if (!tb)
 		return NULL;
 	h = id & (FIB_TABLE_HASHSZ - 1);
@@ -1042,6 +1043,8 @@ void __init ip_fib_init(void)
 	register_pernet_subsys(&fib_net_ops);
 	register_netdevice_notifier(&fib_netdev_notifier);
 	register_inetaddr_notifier(&fib_inetaddr_notifier);
+
+	fib_hash_init();
 }
 
 EXPORT_SYMBOL(inet_addr_type);
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 258214f57fba..499522f3b305 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -747,21 +747,19 @@ static int fn_hash_dump(struct fib_table *tb, struct sk_buff *skb, struct netlin
 	return skb->len;
 }
 
-struct fib_table *fib_hash_init(u32 id)
+void __init fib_hash_init(void)
 {
-	struct fib_table *tb;
+	fn_hash_kmem = kmem_cache_create("ip_fib_hash", sizeof(struct fib_node),
+					 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+
+	fn_alias_kmem = kmem_cache_create("ip_fib_alias", sizeof(struct fib_alias),
+					  0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+
+}
 
-	if (fn_hash_kmem == NULL)
-		fn_hash_kmem = kmem_cache_create("ip_fib_hash",
-						 sizeof(struct fib_node),
-						 0, SLAB_HWCACHE_ALIGN,
-						 NULL);
-
-	if (fn_alias_kmem == NULL)
-		fn_alias_kmem = kmem_cache_create("ip_fib_alias",
-						  sizeof(struct fib_alias),
-						  0, SLAB_HWCACHE_ALIGN,
-						  NULL);
+struct fib_table *fib_hash_table(u32 id)
+{
+	struct fib_table *tb;
 
 	tb = kmalloc(sizeof(struct fib_table) + sizeof(struct fn_hash),
 		     GFP_KERNEL);
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index a15cb0d2e113..fbc80d15827b 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1926,19 +1926,19 @@ out:
 	return -1;
 }
 
-/* Fix more generic FIB names for init later */
+void __init fib_hash_init(void)
+{
+	fn_alias_kmem = kmem_cache_create("ip_fib_alias", sizeof(struct fib_alias),
+					  0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+}
 
-struct fib_table *fib_hash_init(u32 id)
+
+/* Fix more generic FIB names for init later */
+struct fib_table *fib_hash_table(u32 id)
 {
 	struct fib_table *tb;
 	struct trie *t;
 
-	if (fn_alias_kmem == NULL)
-		fn_alias_kmem = kmem_cache_create("ip_fib_alias",
-						  sizeof(struct fib_alias),
-						  0, SLAB_HWCACHE_ALIGN,
-						  NULL);
-
 	tb = kmalloc(sizeof(struct fib_table) + sizeof(struct trie),
 		     GFP_KERNEL);
 	if (tb == NULL)
-- 
cgit v1.2.3


From 8b6f3f62fea7b85fce8f7d12aabba7b191bf60d2 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:33:14 -0800
Subject: [NETFILTER]: Annotate start of kernel fields in NF headers

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_RATEEST.h         | 2 ++
 include/linux/netfilter/xt_connlimit.h       | 2 +-
 include/linux/netfilter/xt_hashlimit.h       | 2 +-
 include/linux/netfilter/xt_quota.h           | 2 ++
 include/linux/netfilter/xt_rateest.h         | 2 ++
 include/linux/netfilter/xt_statistic.h       | 1 +
 include/linux/netfilter/xt_string.h          | 2 ++
 include/linux/netfilter_ipv4/ipt_CLUSTERIP.h | 1 +
 8 files changed, 12 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_RATEEST.h b/include/linux/netfilter/xt_RATEEST.h
index 670f2e49d4fb..f79e3133cbea 100644
--- a/include/linux/netfilter/xt_RATEEST.h
+++ b/include/linux/netfilter/xt_RATEEST.h
@@ -5,6 +5,8 @@ struct xt_rateest_target_info {
 	char			name[IFNAMSIZ];
 	int8_t			interval;
 	u_int8_t		ewma_log;
+
+	/* Used internally by the kernel */
 	struct xt_rateest	*est __attribute__((aligned(8)));
 };
 
diff --git a/include/linux/netfilter/xt_connlimit.h b/include/linux/netfilter/xt_connlimit.h
index 315d2dce9da6..7e3284bcbd2b 100644
--- a/include/linux/netfilter/xt_connlimit.h
+++ b/include/linux/netfilter/xt_connlimit.h
@@ -15,7 +15,7 @@ struct xt_connlimit_info {
 	};
 	unsigned int limit, inverse;
 
-	/* this needs to be at the end */
+	/* Used internally by the kernel */
 	struct xt_connlimit_data *data __attribute__((aligned(8)));
 };
 
diff --git a/include/linux/netfilter/xt_hashlimit.h b/include/linux/netfilter/xt_hashlimit.h
index b4556b8edbfd..c19972e4564d 100644
--- a/include/linux/netfilter/xt_hashlimit.h
+++ b/include/linux/netfilter/xt_hashlimit.h
@@ -29,9 +29,9 @@ struct hashlimit_cfg {
 struct xt_hashlimit_info {
 	char name [IFNAMSIZ];		/* name */
 	struct hashlimit_cfg cfg;
-	struct xt_hashlimit_htable *hinfo;
 
 	/* Used internally by the kernel */
+	struct xt_hashlimit_htable *hinfo;
 	union {
 		void *ptr;
 		struct xt_hashlimit_info *master;
diff --git a/include/linux/netfilter/xt_quota.h b/include/linux/netfilter/xt_quota.h
index acd7fd77bbee..4c8368d781e5 100644
--- a/include/linux/netfilter/xt_quota.h
+++ b/include/linux/netfilter/xt_quota.h
@@ -9,6 +9,8 @@ enum xt_quota_flags {
 struct xt_quota_info {
 	u_int32_t		flags;
 	u_int32_t		pad;
+
+	/* Used internally by the kernel */
 	aligned_u64		quota;
 	struct xt_quota_info	*master;
 };
diff --git a/include/linux/netfilter/xt_rateest.h b/include/linux/netfilter/xt_rateest.h
index 51948e15aea2..2010cb74250f 100644
--- a/include/linux/netfilter/xt_rateest.h
+++ b/include/linux/netfilter/xt_rateest.h
@@ -26,6 +26,8 @@ struct xt_rateest_match_info {
 	u_int32_t		pps1;
 	u_int32_t		bps2;
 	u_int32_t		pps2;
+
+	/* Used internally by the kernel */
 	struct xt_rateest	*est1 __attribute__((aligned(8)));
 	struct xt_rateest	*est2 __attribute__((aligned(8)));
 };
diff --git a/include/linux/netfilter/xt_statistic.h b/include/linux/netfilter/xt_statistic.h
index c344e9916e23..3d38bc975048 100644
--- a/include/linux/netfilter/xt_statistic.h
+++ b/include/linux/netfilter/xt_statistic.h
@@ -23,6 +23,7 @@ struct xt_statistic_info {
 		struct {
 			u_int32_t	every;
 			u_int32_t	packet;
+			/* Used internally by the kernel */
 			u_int32_t	count;
 		} nth;
 	} u;
diff --git a/include/linux/netfilter/xt_string.h b/include/linux/netfilter/xt_string.h
index 3b3419f2637d..bb21dd1aee2d 100644
--- a/include/linux/netfilter/xt_string.h
+++ b/include/linux/netfilter/xt_string.h
@@ -12,6 +12,8 @@ struct xt_string_info
 	char 	  pattern[XT_STRING_MAX_PATTERN_SIZE];
 	u_int8_t  patlen;
 	u_int8_t  invert;
+
+	/* Used internally by the kernel */
 	struct ts_config __attribute__((aligned(8))) *config;
 };
 
diff --git a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
index daf50be22c9d..e5a3687c8a72 100644
--- a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
+++ b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h
@@ -27,6 +27,7 @@ struct ipt_clusterip_tgt_info {
 	u_int32_t hash_mode;
 	u_int32_t hash_initval;
 
+	/* Used internally by the kernel */
 	struct clusterip_config *config;
 };
 
-- 
cgit v1.2.3


From 0dc8c76029f4675c2345eefd947f123e64de1aae Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:38:34 -0800
Subject: [NETFILTER]: xt_CONNMARK target, revision 1

Introduces the xt_CONNMARK target revision 1. It uses fixed types, and
also uses the more expressive XOR logic. Futhermore, it allows to
selectively pick bits from both the ctmark and the nfmark in the SAVE
and RESTORE operations.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_CONNMARK.h |   5 ++
 net/netfilter/xt_CONNMARK.c           | 117 ++++++++++++++++++++++++++++------
 2 files changed, 102 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_CONNMARK.h b/include/linux/netfilter/xt_CONNMARK.h
index 9f744689fffc..4e58ba43c289 100644
--- a/include/linux/netfilter/xt_CONNMARK.h
+++ b/include/linux/netfilter/xt_CONNMARK.h
@@ -22,4 +22,9 @@ struct xt_connmark_target_info {
 	u_int8_t mode;
 };
 
+struct xt_connmark_tginfo1 {
+	u_int32_t ctmark, ctmask, nfmask;
+	u_int8_t mode;
+};
+
 #endif /*_XT_CONNMARK_H_target*/
diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c
index ec2eb341d2e6..b9bd77233460 100644
--- a/net/netfilter/xt_CONNMARK.c
+++ b/net/netfilter/xt_CONNMARK.c
@@ -1,8 +1,10 @@
-/* This kernel module is used to modify the connection mark values, or
- * to optionally restore the skb nfmark from the connection mark
+/*
+ *	xt_CONNMARK - Netfilter module to modify the connection mark values
  *
- * Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
- * by Henrik Nordstrom <hno@marasystems.com>
+ *	Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
+ *	by Henrik Nordstrom <hno@marasystems.com>
+ *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
+ *	Jan Engelhardt <jengelh@computergmbh.de>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -34,9 +36,9 @@ MODULE_ALIAS("ip6t_CONNMARK");
 #include <net/netfilter/nf_conntrack_ecache.h>
 
 static unsigned int
-connmark_tg(struct sk_buff *skb, const struct net_device *in,
-            const struct net_device *out, unsigned int hooknum,
-            const struct xt_target *target, const void *targinfo)
+connmark_tg_v0(struct sk_buff *skb, const struct net_device *in,
+               const struct net_device *out, unsigned int hooknum,
+               const struct xt_target *target, const void *targinfo)
 {
 	const struct xt_connmark_target_info *markinfo = targinfo;
 	struct nf_conn *ct;
@@ -74,10 +76,50 @@ connmark_tg(struct sk_buff *skb, const struct net_device *in,
 	return XT_CONTINUE;
 }
 
+static unsigned int
+connmark_tg(struct sk_buff *skb, const struct net_device *in,
+            const struct net_device *out, unsigned int hooknum,
+            const struct xt_target *target, const void *targinfo)
+{
+	const struct xt_connmark_tginfo1 *info = targinfo;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+	u_int32_t newmark;
+
+	ct = nf_ct_get(skb, &ctinfo);
+	if (ct == NULL)
+		return XT_CONTINUE;
+
+	switch (info->mode) {
+	case XT_CONNMARK_SET:
+		newmark = (ct->mark & ~info->ctmask) ^ info->ctmark;
+		if (ct->mark != newmark) {
+			ct->mark = newmark;
+			nf_conntrack_event_cache(IPCT_MARK, skb);
+		}
+		break;
+	case XT_CONNMARK_SAVE:
+		newmark = (ct->mark & ~info->ctmask) ^
+		          (skb->mark & info->nfmask);
+		if (ct->mark != newmark) {
+			ct->mark = newmark;
+			nf_conntrack_event_cache(IPCT_MARK, skb);
+		}
+		break;
+	case XT_CONNMARK_RESTORE:
+		newmark = (skb->mark & ~info->nfmask) ^
+		          (ct->mark & info->ctmask);
+		skb->mark = newmark;
+		break;
+	}
+
+	return XT_CONTINUE;
+}
+
 static bool
-connmark_tg_check(const char *tablename, const void *entry,
-                  const struct xt_target *target, void *targinfo,
-                  unsigned int hook_mask)
+connmark_tg_check_v0(const char *tablename, const void *entry,
+                     const struct xt_target *target, void *targinfo,
+                     unsigned int hook_mask)
 {
 	const struct xt_connmark_target_info *matchinfo = targinfo;
 
@@ -101,6 +143,19 @@ connmark_tg_check(const char *tablename, const void *entry,
 	return true;
 }
 
+static bool
+connmark_tg_check(const char *tablename, const void *entry,
+                  const struct xt_target *target, void *targinfo,
+                  unsigned int hook_mask)
+{
+	if (nf_ct_l3proto_try_module_get(target->family) < 0) {
+		printk(KERN_WARNING "cannot load conntrack support for "
+		       "proto=%u\n", target->family);
+		return false;
+	}
+	return true;
+}
+
 static void
 connmark_tg_destroy(const struct xt_target *target, void *targinfo)
 {
@@ -115,7 +170,7 @@ struct compat_xt_connmark_target_info {
 	u_int16_t	__pad2;
 };
 
-static void connmark_tg_compat_from_user(void *dst, void *src)
+static void connmark_tg_compat_from_user_v0(void *dst, void *src)
 {
 	const struct compat_xt_connmark_target_info *cm = src;
 	struct xt_connmark_target_info m = {
@@ -126,7 +181,7 @@ static void connmark_tg_compat_from_user(void *dst, void *src)
 	memcpy(dst, &m, sizeof(m));
 }
 
-static int connmark_tg_compat_to_user(void __user *dst, void *src)
+static int connmark_tg_compat_to_user_v0(void __user *dst, void *src)
 {
 	const struct xt_connmark_target_info *m = src;
 	struct compat_xt_connmark_target_info cm = {
@@ -141,32 +196,54 @@ static int connmark_tg_compat_to_user(void __user *dst, void *src)
 static struct xt_target connmark_tg_reg[] __read_mostly = {
 	{
 		.name		= "CONNMARK",
+		.revision	= 0,
 		.family		= AF_INET,
-		.checkentry	= connmark_tg_check,
+		.checkentry	= connmark_tg_check_v0,
 		.destroy	= connmark_tg_destroy,
-		.target		= connmark_tg,
+		.target		= connmark_tg_v0,
 		.targetsize	= sizeof(struct xt_connmark_target_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_connmark_target_info),
-		.compat_from_user = connmark_tg_compat_from_user,
-		.compat_to_user	= connmark_tg_compat_to_user,
+		.compat_from_user = connmark_tg_compat_from_user_v0,
+		.compat_to_user	= connmark_tg_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE
 	},
 	{
 		.name		= "CONNMARK",
+		.revision	= 0,
 		.family		= AF_INET6,
-		.checkentry	= connmark_tg_check,
+		.checkentry	= connmark_tg_check_v0,
 		.destroy	= connmark_tg_destroy,
-		.target		= connmark_tg,
+		.target		= connmark_tg_v0,
 		.targetsize	= sizeof(struct xt_connmark_target_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_connmark_target_info),
-		.compat_from_user = connmark_tg_compat_from_user,
-		.compat_to_user	= connmark_tg_compat_to_user,
+		.compat_from_user = connmark_tg_compat_from_user_v0,
+		.compat_to_user	= connmark_tg_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE
 	},
+	{
+		.name           = "CONNMARK",
+		.revision       = 1,
+		.family         = AF_INET,
+		.checkentry     = connmark_tg_check,
+		.target         = connmark_tg,
+		.targetsize     = sizeof(struct xt_connmark_tginfo1),
+		.destroy        = connmark_tg_destroy,
+		.me             = THIS_MODULE,
+	},
+	{
+		.name           = "CONNMARK",
+		.revision       = 1,
+		.family         = AF_INET6,
+		.checkentry     = connmark_tg_check,
+		.target         = connmark_tg,
+		.targetsize     = sizeof(struct xt_connmark_tginfo1),
+		.destroy        = connmark_tg_destroy,
+		.me             = THIS_MODULE,
+	},
 };
 
 static int __init connmark_tg_init(void)
-- 
cgit v1.2.3


From e0a812aea5cbf2085f7645bf2bfd9cba91c8a672 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:38:52 -0800
Subject: [NETFILTER]: xt_MARK target, revision 2

Introduces the xt_MARK target revision 2. It uses fixed types, and
also uses the more expressive XOR logic.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_MARK.h |  4 +++
 net/netfilter/xt_MARK.c           | 74 +++++++++++++++++++++++++++------------
 2 files changed, 56 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_MARK.h b/include/linux/netfilter/xt_MARK.h
index b021e93ee5d6..778b278fd9f2 100644
--- a/include/linux/netfilter/xt_MARK.h
+++ b/include/linux/netfilter/xt_MARK.h
@@ -18,4 +18,8 @@ struct xt_mark_target_info_v1 {
 	u_int8_t mode;
 };
 
+struct xt_mark_tginfo2 {
+	u_int32_t mark, mask;
+};
+
 #endif /*_XT_MARK_H_target */
diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c
index 57c6d55e33d3..1c3fb75adfd1 100644
--- a/net/netfilter/xt_MARK.c
+++ b/net/netfilter/xt_MARK.c
@@ -1,10 +1,13 @@
-/* This is a module which is used for setting the NFMARK field of an skb. */
-
-/* (C) 1999-2001 Marc Boucher <marc@mbsi.ca>
+/*
+ *	xt_MARK - Netfilter module to modify the NFMARK field of an skb
+ *
+ *	(C) 1999-2001 Marc Boucher <marc@mbsi.ca>
+ *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
+ *	Jan Engelhardt <jengelh@computergmbh.de>
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ *	This program is free software; you can redistribute it and/or modify
+ *	it under the terms of the GNU General Public License version 2 as
+ *	published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
@@ -33,9 +36,9 @@ mark_tg_v0(struct sk_buff *skb, const struct net_device *in,
 }
 
 static unsigned int
-mark_tg(struct sk_buff *skb, const struct net_device *in,
-        const struct net_device *out, unsigned int hooknum,
-        const struct xt_target *target, const void *targinfo)
+mark_tg_v1(struct sk_buff *skb, const struct net_device *in,
+           const struct net_device *out, unsigned int hooknum,
+           const struct xt_target *target, const void *targinfo)
 {
 	const struct xt_mark_target_info_v1 *markinfo = targinfo;
 	int mark = 0;
@@ -58,6 +61,17 @@ mark_tg(struct sk_buff *skb, const struct net_device *in,
 	return XT_CONTINUE;
 }
 
+static unsigned int
+mark_tg(struct sk_buff *skb, const struct net_device *in,
+        const struct net_device *out, unsigned int hooknum,
+        const struct xt_target *target, const void *targinfo)
+{
+	const struct xt_mark_tginfo2 *info = targinfo;
+
+	skb->mark = (skb->mark & ~info->mask) ^ info->mark;
+	return XT_CONTINUE;
+}
+
 static bool
 mark_tg_check_v0(const char *tablename, const void *entry,
                  const struct xt_target *target, void *targinfo,
@@ -73,9 +87,9 @@ mark_tg_check_v0(const char *tablename, const void *entry,
 }
 
 static bool
-mark_tg_check(const char *tablename, const void *entry,
-              const struct xt_target *target, void *targinfo,
-              unsigned int hook_mask)
+mark_tg_check_v1(const char *tablename, const void *entry,
+                 const struct xt_target *target, void *targinfo,
+                 unsigned int hook_mask)
 {
 	const struct xt_mark_target_info_v1 *markinfo = targinfo;
 
@@ -98,7 +112,7 @@ struct compat_xt_mark_target_info {
 	compat_ulong_t	mark;
 };
 
-static void mark_tg_compat_from_user(void *dst, void *src)
+static void mark_tg_compat_from_user_v0(void *dst, void *src)
 {
 	const struct compat_xt_mark_target_info *cm = src;
 	struct xt_mark_target_info m = {
@@ -107,7 +121,7 @@ static void mark_tg_compat_from_user(void *dst, void *src)
 	memcpy(dst, &m, sizeof(m));
 }
 
-static int mark_tg_compat_to_user(void __user *dst, void *src)
+static int mark_tg_compat_to_user_v0(void __user *dst, void *src)
 {
 	const struct xt_mark_target_info *m = src;
 	struct compat_xt_mark_target_info cm = {
@@ -154,8 +168,8 @@ static struct xt_target mark_tg_reg[] __read_mostly = {
 		.targetsize	= sizeof(struct xt_mark_target_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_target_info),
-		.compat_from_user = mark_tg_compat_from_user,
-		.compat_to_user	= mark_tg_compat_to_user,
+		.compat_from_user = mark_tg_compat_from_user_v0,
+		.compat_to_user	= mark_tg_compat_to_user_v0,
 #endif
 		.table		= "mangle",
 		.me		= THIS_MODULE,
@@ -164,8 +178,8 @@ static struct xt_target mark_tg_reg[] __read_mostly = {
 		.name		= "MARK",
 		.family		= AF_INET,
 		.revision	= 1,
-		.checkentry	= mark_tg_check,
-		.target		= mark_tg,
+		.checkentry	= mark_tg_check_v1,
+		.target		= mark_tg_v1,
 		.targetsize	= sizeof(struct xt_mark_target_info_v1),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_target_info_v1),
@@ -184,8 +198,8 @@ static struct xt_target mark_tg_reg[] __read_mostly = {
 		.targetsize	= sizeof(struct xt_mark_target_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_target_info),
-		.compat_from_user = mark_tg_compat_from_user,
-		.compat_to_user	= mark_tg_compat_to_user,
+		.compat_from_user = mark_tg_compat_from_user_v0,
+		.compat_to_user	= mark_tg_compat_to_user_v0,
 #endif
 		.table		= "mangle",
 		.me		= THIS_MODULE,
@@ -194,8 +208,8 @@ static struct xt_target mark_tg_reg[] __read_mostly = {
 		.name		= "MARK",
 		.family		= AF_INET6,
 		.revision	= 1,
-		.checkentry	= mark_tg_check,
-		.target		= mark_tg,
+		.checkentry	= mark_tg_check_v1,
+		.target		= mark_tg_v1,
 		.targetsize	= sizeof(struct xt_mark_target_info_v1),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_target_info_v1),
@@ -205,6 +219,22 @@ static struct xt_target mark_tg_reg[] __read_mostly = {
 		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
+	{
+		.name           = "MARK",
+		.revision       = 2,
+		.family         = AF_INET,
+		.target         = mark_tg,
+		.targetsize     = sizeof(struct xt_mark_tginfo2),
+		.me             = THIS_MODULE,
+	},
+	{
+		.name           = "MARK",
+		.revision       = 2,
+		.family         = AF_INET6,
+		.target         = mark_tg,
+		.targetsize     = sizeof(struct xt_mark_tginfo2),
+		.me             = THIS_MODULE,
+	},
 };
 
 static int __init mark_tg_init(void)
-- 
cgit v1.2.3


From 96e3227265852ffad332f911887c9cf26c85e40a Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:39:13 -0800
Subject: [NETFILTER]: xt_connmark match, revision 1

Introduces the xt_connmark match revision 1. It uses fixed types,
eventually obsoleting revision 0 some day (uses nonfixed types).
(Unfixed types like "unsigned long" do not play well with mixed
user-/kernelspace "bitness", e.g. 32/64, as is common on SPARC64,
and need extra compat code.)

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_connmark.h |  5 ++
 net/netfilter/xt_connmark.c           | 88 ++++++++++++++++++++++++++++-------
 2 files changed, 76 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_connmark.h b/include/linux/netfilter/xt_connmark.h
index c592f6ae0883..359ef86918dc 100644
--- a/include/linux/netfilter/xt_connmark.h
+++ b/include/linux/netfilter/xt_connmark.h
@@ -15,4 +15,9 @@ struct xt_connmark_info {
 	u_int8_t invert;
 };
 
+struct xt_connmark_mtinfo1 {
+	u_int32_t mark, mask;
+	u_int8_t invert;
+};
+
 #endif /*_XT_CONNMARK_H*/
diff --git a/net/netfilter/xt_connmark.c b/net/netfilter/xt_connmark.c
index 8ad875bc1589..55c62350b1f2 100644
--- a/net/netfilter/xt_connmark.c
+++ b/net/netfilter/xt_connmark.c
@@ -1,8 +1,10 @@
-/* This kernel module matches connection mark values set by the
- * CONNMARK target
+/*
+ *	xt_connmark - Netfilter module to match connection mark values
  *
- * Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
- * by Henrik Nordstrom <hno@marasystems.com>
+ *	Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
+ *	by Henrik Nordstrom <hno@marasystems.com>
+ *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
+ *	Jan Engelhardt <jengelh@computergmbh.de>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -36,6 +38,23 @@ connmark_mt(const struct sk_buff *skb, const struct net_device *in,
             const struct net_device *out, const struct xt_match *match,
             const void *matchinfo, int offset, unsigned int protoff,
             bool *hotdrop)
+{
+	const struct xt_connmark_mtinfo1 *info = matchinfo;
+	enum ip_conntrack_info ctinfo;
+	const struct nf_conn *ct;
+
+	ct = nf_ct_get(skb, &ctinfo);
+	if (ct == NULL)
+		return false;
+
+	return ((ct->mark & info->mask) == info->mark) ^ info->invert;
+}
+
+static bool
+connmark_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+               const struct net_device *out, const struct xt_match *match,
+               const void *matchinfo, int offset, unsigned int protoff,
+               bool *hotdrop)
 {
 	const struct xt_connmark_info *info = matchinfo;
 	const struct nf_conn *ct;
@@ -49,9 +68,9 @@ connmark_mt(const struct sk_buff *skb, const struct net_device *in,
 }
 
 static bool
-connmark_mt_check(const char *tablename, const void *ip,
-                  const struct xt_match *match, void *matchinfo,
-                  unsigned int hook_mask)
+connmark_mt_check_v0(const char *tablename, const void *ip,
+                     const struct xt_match *match, void *matchinfo,
+                     unsigned int hook_mask)
 {
 	const struct xt_connmark_info *cm = matchinfo;
 
@@ -67,6 +86,19 @@ connmark_mt_check(const char *tablename, const void *ip,
 	return true;
 }
 
+static bool
+connmark_mt_check(const char *tablename, const void *ip,
+                  const struct xt_match *match, void *matchinfo,
+                  unsigned int hook_mask)
+{
+	if (nf_ct_l3proto_try_module_get(match->family) < 0) {
+		printk(KERN_WARNING "cannot load conntrack support for "
+		       "proto=%u\n", match->family);
+		return false;
+	}
+	return true;
+}
+
 static void
 connmark_mt_destroy(const struct xt_match *match, void *matchinfo)
 {
@@ -81,7 +113,7 @@ struct compat_xt_connmark_info {
 	u_int16_t	__pad2;
 };
 
-static void connmark_mt_compat_from_user(void *dst, void *src)
+static void connmark_mt_compat_from_user_v0(void *dst, void *src)
 {
 	const struct compat_xt_connmark_info *cm = src;
 	struct xt_connmark_info m = {
@@ -92,7 +124,7 @@ static void connmark_mt_compat_from_user(void *dst, void *src)
 	memcpy(dst, &m, sizeof(m));
 }
 
-static int connmark_mt_compat_to_user(void __user *dst, void *src)
+static int connmark_mt_compat_to_user_v0(void __user *dst, void *src)
 {
 	const struct xt_connmark_info *m = src;
 	struct compat_xt_connmark_info cm = {
@@ -107,32 +139,54 @@ static int connmark_mt_compat_to_user(void __user *dst, void *src)
 static struct xt_match connmark_mt_reg[] __read_mostly = {
 	{
 		.name		= "connmark",
+		.revision	= 0,
 		.family		= AF_INET,
-		.checkentry	= connmark_mt_check,
-		.match		= connmark_mt,
+		.checkentry	= connmark_mt_check_v0,
+		.match		= connmark_mt_v0,
 		.destroy	= connmark_mt_destroy,
 		.matchsize	= sizeof(struct xt_connmark_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_connmark_info),
-		.compat_from_user = connmark_mt_compat_from_user,
-		.compat_to_user	= connmark_mt_compat_to_user,
+		.compat_from_user = connmark_mt_compat_from_user_v0,
+		.compat_to_user	= connmark_mt_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE
 	},
 	{
 		.name		= "connmark",
+		.revision	= 0,
 		.family		= AF_INET6,
-		.checkentry	= connmark_mt_check,
-		.match		= connmark_mt,
+		.checkentry	= connmark_mt_check_v0,
+		.match		= connmark_mt_v0,
 		.destroy	= connmark_mt_destroy,
 		.matchsize	= sizeof(struct xt_connmark_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_connmark_info),
-		.compat_from_user = connmark_mt_compat_from_user,
-		.compat_to_user	= connmark_mt_compat_to_user,
+		.compat_from_user = connmark_mt_compat_from_user_v0,
+		.compat_to_user	= connmark_mt_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE
 	},
+	{
+		.name           = "connmark",
+		.revision       = 1,
+		.family         = AF_INET,
+		.checkentry     = connmark_mt_check,
+		.match          = connmark_mt,
+		.matchsize      = sizeof(struct xt_connmark_mtinfo1),
+		.destroy        = connmark_mt_destroy,
+		.me             = THIS_MODULE,
+	},
+	{
+		.name           = "connmark",
+		.revision       = 1,
+		.family         = AF_INET6,
+		.checkentry     = connmark_mt_check,
+		.match          = connmark_mt,
+		.matchsize      = sizeof(struct xt_connmark_mtinfo1),
+		.destroy        = connmark_mt_destroy,
+		.me             = THIS_MODULE,
+	},
 };
 
 static int __init connmark_mt_init(void)
-- 
cgit v1.2.3


From 2e3075a2c4364c0e8726ac2a0f3b1708da781bac Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:40:34 -0800
Subject: [NETFILTER]: Extend nf_inet_addr with in{,6}_addr

Extend union nf_inet_addr with struct in_addr and in6_addr. Useful
because a lot of in-kernel IPv4 and IPv6 functions use
in_addr/in6_addr.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index c41f6438095d..d74e79bacd2d 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -7,6 +7,8 @@
 #include <linux/skbuff.h>
 #include <linux/net.h>
 #include <linux/if.h>
+#include <linux/in.h>
+#include <linux/in6.h>
 #include <linux/wait.h>
 #include <linux/list.h>
 #endif
@@ -52,6 +54,8 @@ union nf_inet_addr {
 	u_int32_t	all[4];
 	__be32		ip;
 	__be32		ip6[4];
+	struct in_addr	in;
+	struct in6_addr	in6;
 };
 
 #ifdef __KERNEL__
-- 
cgit v1.2.3


From 64eb12f9972d45f3b9b0f0a33a966e311c3d5275 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:40:53 -0800
Subject: [NETFILTER]: xt_conntrack match, revision 1

Introduces the xt_conntrack match revision 1. It uses fixed types, the
new nf_inet_addr and comes with IPv6 support, thereby completely
superseding xt_state.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_conntrack.h |  16 ++-
 net/netfilter/xt_conntrack.c           | 207 +++++++++++++++++++++++++++++----
 2 files changed, 197 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_conntrack.h b/include/linux/netfilter/xt_conntrack.h
index 70b6f718cf4c..d2492a3329be 100644
--- a/include/linux/netfilter/xt_conntrack.h
+++ b/include/linux/netfilter/xt_conntrack.h
@@ -6,7 +6,9 @@
 #define _XT_CONNTRACK_H
 
 #include <linux/netfilter/nf_conntrack_tuple_common.h>
-#include <linux/in.h>
+#ifdef __KERNEL__
+#	include <linux/in.h>
+#endif
 
 #define XT_CONNTRACK_STATE_BIT(ctinfo) (1 << ((ctinfo)%IP_CT_IS_REPLY+1))
 #define XT_CONNTRACK_STATE_INVALID (1 << 0)
@@ -60,4 +62,16 @@ struct xt_conntrack_info
 	/* Inverse flags */
 	u_int8_t invflags;
 };
+
+struct xt_conntrack_mtinfo1 {
+	union nf_inet_addr origsrc_addr, origsrc_mask;
+	union nf_inet_addr origdst_addr, origdst_mask;
+	union nf_inet_addr replsrc_addr, replsrc_mask;
+	union nf_inet_addr repldst_addr, repldst_mask;
+	u_int32_t expires_min, expires_max;
+	u_int16_t l4proto;
+	u_int8_t state_mask, status_mask;
+	u_int8_t match_flags, invert_flags;
+};
+
 #endif /*_XT_CONNTRACK_H*/
diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c
index 3f8bfbaa9b19..dc9e73786864 100644
--- a/net/netfilter/xt_conntrack.c
+++ b/net/netfilter/xt_conntrack.c
@@ -1,15 +1,19 @@
-/* Kernel module to match connection tracking information.
- * Superset of Rusty's minimalistic state match.
+/*
+ *	xt_conntrack - Netfilter module to match connection tracking
+ *	information. (Superset of Rusty's minimalistic state match.)
  *
- * (C) 2001  Marc Boucher (marc@mbsi.ca).
+ *	(C) 2001  Marc Boucher (marc@mbsi.ca).
+ *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
+ *	Jan Engelhardt <jengelh@computergmbh.de>
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ *	This program is free software; you can redistribute it and/or modify
+ *	it under the terms of the GNU General Public License version 2 as
+ *	published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
 #include <linux/skbuff.h>
+#include <net/ipv6.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_conntrack.h>
 #include <net/netfilter/nf_conntrack.h>
@@ -18,12 +22,13 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>");
 MODULE_DESCRIPTION("iptables connection tracking match module");
 MODULE_ALIAS("ipt_conntrack");
+MODULE_ALIAS("ip6t_conntrack");
 
 static bool
-conntrack_mt(const struct sk_buff *skb, const struct net_device *in,
-             const struct net_device *out, const struct xt_match *match,
-             const void *matchinfo, int offset, unsigned int protoff,
-             bool *hotdrop)
+conntrack_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+                const struct net_device *out, const struct xt_match *match,
+                const void *matchinfo, int offset, unsigned int protoff,
+                bool *hotdrop)
 {
 	const struct xt_conntrack_info *sinfo = matchinfo;
 	const struct nf_conn *ct;
@@ -111,6 +116,134 @@ conntrack_mt(const struct sk_buff *skb, const struct net_device *in,
 #undef FWINV
 }
 
+static bool
+conntrack_addrcmp(const union nf_inet_addr *kaddr,
+                  const union nf_inet_addr *uaddr,
+                  const union nf_inet_addr *umask, unsigned int l3proto)
+{
+	if (l3proto == AF_INET)
+		return (kaddr->ip & umask->ip) == uaddr->ip;
+	else if (l3proto == AF_INET6)
+		return ipv6_masked_addr_cmp(&kaddr->in6, &umask->in6,
+		       &uaddr->in6) == 0;
+	else
+		return false;
+}
+
+static inline bool
+conntrack_mt_origsrc(const struct nf_conn *ct,
+                     const struct xt_conntrack_mtinfo1 *info,
+                     unsigned int family)
+{
+	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3,
+	       &info->origsrc_addr, &info->origsrc_mask, family);
+}
+
+static inline bool
+conntrack_mt_origdst(const struct nf_conn *ct,
+                     const struct xt_conntrack_mtinfo1 *info,
+                     unsigned int family)
+{
+	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3,
+	       &info->origdst_addr, &info->origdst_mask, family);
+}
+
+static inline bool
+conntrack_mt_replsrc(const struct nf_conn *ct,
+                     const struct xt_conntrack_mtinfo1 *info,
+                     unsigned int family)
+{
+	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3,
+	       &info->replsrc_addr, &info->replsrc_mask, family);
+}
+
+static inline bool
+conntrack_mt_repldst(const struct nf_conn *ct,
+                     const struct xt_conntrack_mtinfo1 *info,
+                     unsigned int family)
+{
+	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3,
+	       &info->repldst_addr, &info->repldst_mask, family);
+}
+
+static bool
+conntrack_mt(const struct sk_buff *skb, const struct net_device *in,
+             const struct net_device *out, const struct xt_match *match,
+             const void *matchinfo, int offset, unsigned int protoff,
+             bool *hotdrop)
+{
+	const struct xt_conntrack_mtinfo1 *info = matchinfo;
+	enum ip_conntrack_info ctinfo;
+	const struct nf_conn *ct;
+	unsigned int statebit;
+
+	ct = nf_ct_get(skb, &ctinfo);
+
+	if (ct == &nf_conntrack_untracked)
+		statebit = XT_CONNTRACK_STATE_UNTRACKED;
+	else if (ct != NULL)
+		statebit = XT_CONNTRACK_STATE_BIT(ctinfo);
+	else
+		statebit = XT_CONNTRACK_STATE_INVALID;
+
+	if (info->match_flags & XT_CONNTRACK_STATE) {
+		if (ct != NULL) {
+			if (test_bit(IPS_SRC_NAT_BIT, &ct->status))
+				statebit |= XT_CONNTRACK_STATE_SNAT;
+			if (test_bit(IPS_DST_NAT_BIT, &ct->status))
+				statebit |= XT_CONNTRACK_STATE_DNAT;
+		}
+		if ((info->state_mask & statebit) ^
+		    !(info->invert_flags & XT_CONNTRACK_STATE))
+			return false;
+	}
+
+	if (ct == NULL)
+		return info->match_flags & XT_CONNTRACK_STATE;
+
+	if ((info->match_flags & XT_CONNTRACK_PROTO) &&
+	    ((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum ==
+	    info->l4proto) ^ !(info->invert_flags & XT_CONNTRACK_PROTO)))
+		return false;
+
+	if (info->match_flags & XT_CONNTRACK_ORIGSRC)
+		if (conntrack_mt_origsrc(ct, info, match->family) ^
+		    !(info->invert_flags & XT_CONNTRACK_ORIGSRC))
+			return false;
+
+	if (info->match_flags & XT_CONNTRACK_ORIGDST)
+		if (conntrack_mt_origdst(ct, info, match->family) ^
+		    !(info->invert_flags & XT_CONNTRACK_ORIGDST))
+			return false;
+
+	if (info->match_flags & XT_CONNTRACK_REPLSRC)
+		if (conntrack_mt_replsrc(ct, info, match->family) ^
+		    !(info->invert_flags & XT_CONNTRACK_REPLSRC))
+			return false;
+
+	if (info->match_flags & XT_CONNTRACK_REPLDST)
+		if (conntrack_mt_repldst(ct, info, match->family) ^
+		    !(info->invert_flags & XT_CONNTRACK_REPLDST))
+			return false;
+
+	if ((info->match_flags & XT_CONNTRACK_STATUS) &&
+	    (!!(info->status_mask & ct->status) ^
+	    !(info->invert_flags & XT_CONNTRACK_STATUS)))
+		return false;
+
+	if (info->match_flags & XT_CONNTRACK_EXPIRES) {
+		unsigned long expires = 0;
+
+		if (timer_pending(&ct->timeout))
+			expires = (ct->timeout.expires - jiffies) / HZ;
+		if ((expires >= info->expires_min &&
+		    expires <= info->expires_max) ^
+		    !(info->invert_flags & XT_CONNTRACK_EXPIRES))
+			return false;
+	}
+	return true;
+}
+
 static bool
 conntrack_mt_check(const char *tablename, const void *ip,
                    const struct xt_match *match, void *matchinfo,
@@ -144,7 +277,7 @@ struct compat_xt_conntrack_info
 	u_int8_t			invflags;
 };
 
-static void conntrack_mt_compat_from_user(void *dst, void *src)
+static void conntrack_mt_compat_from_user_v0(void *dst, void *src)
 {
 	const struct compat_xt_conntrack_info *cm = src;
 	struct xt_conntrack_info m = {
@@ -161,7 +294,7 @@ static void conntrack_mt_compat_from_user(void *dst, void *src)
 	memcpy(dst, &m, sizeof(m));
 }
 
-static int conntrack_mt_compat_to_user(void __user *dst, void *src)
+static int conntrack_mt_compat_to_user_v0(void __user *dst, void *src)
 {
 	const struct xt_conntrack_info *m = src;
 	struct compat_xt_conntrack_info cm = {
@@ -179,29 +312,53 @@ static int conntrack_mt_compat_to_user(void __user *dst, void *src)
 }
 #endif
 
-static struct xt_match conntrack_mt_reg __read_mostly = {
-	.name		= "conntrack",
-	.match		= conntrack_mt,
-	.checkentry	= conntrack_mt_check,
-	.destroy	= conntrack_mt_destroy,
-	.matchsize	= sizeof(struct xt_conntrack_info),
+static struct xt_match conntrack_mt_reg[] __read_mostly = {
+	{
+		.name       = "conntrack",
+		.revision   = 0,
+		.family     = AF_INET,
+		.match      = conntrack_mt_v0,
+		.checkentry = conntrack_mt_check,
+		.destroy    = conntrack_mt_destroy,
+		.matchsize  = sizeof(struct xt_conntrack_info),
+		.me         = THIS_MODULE,
 #ifdef CONFIG_COMPAT
-	.compatsize	= sizeof(struct compat_xt_conntrack_info),
-	.compat_from_user = conntrack_mt_compat_from_user,
-	.compat_to_user	= conntrack_mt_compat_to_user,
+		.compatsize       = sizeof(struct compat_xt_conntrack_info),
+		.compat_from_user = conntrack_mt_compat_from_user_v0,
+		.compat_to_user   = conntrack_mt_compat_to_user_v0,
 #endif
-	.family		= AF_INET,
-	.me		= THIS_MODULE,
+	},
+	{
+		.name       = "conntrack",
+		.revision   = 1,
+		.family     = AF_INET,
+		.matchsize  = sizeof(struct xt_conntrack_mtinfo1),
+		.match      = conntrack_mt,
+		.checkentry = conntrack_mt_check,
+		.destroy    = conntrack_mt_destroy,
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "conntrack",
+		.revision   = 1,
+		.family     = AF_INET6,
+		.matchsize  = sizeof(struct xt_conntrack_mtinfo1),
+		.match      = conntrack_mt,
+		.checkentry = conntrack_mt_check,
+		.destroy    = conntrack_mt_destroy,
+		.me         = THIS_MODULE,
+	},
 };
 
 static int __init conntrack_mt_init(void)
 {
-	return xt_register_match(&conntrack_mt_reg);
+	return xt_register_matches(conntrack_mt_reg,
+	       ARRAY_SIZE(conntrack_mt_reg));
 }
 
 static void __exit conntrack_mt_exit(void)
 {
-	xt_unregister_match(&conntrack_mt_reg);
+	xt_unregister_matches(conntrack_mt_reg, ARRAY_SIZE(conntrack_mt_reg));
 }
 
 module_init(conntrack_mt_init);
-- 
cgit v1.2.3


From 17b0d7ef658583842da75eebf8001dc617f0b52e Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:41:11 -0800
Subject: [NETFILTER]: xt_mark match, revision 1

Introduces the xt_mark match revision 1. It uses fixed types,
eventually obsoleting revision 0 some day (uses nonfixed types).

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_mark.h |  5 +++
 net/netfilter/xt_mark.c           | 72 ++++++++++++++++++++++++++++-----------
 2 files changed, 57 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_mark.h b/include/linux/netfilter/xt_mark.h
index 802dd4842caf..fae74bc3f34e 100644
--- a/include/linux/netfilter/xt_mark.h
+++ b/include/linux/netfilter/xt_mark.h
@@ -6,4 +6,9 @@ struct xt_mark_info {
     u_int8_t invert;
 };
 
+struct xt_mark_mtinfo1 {
+	u_int32_t mark, mask;
+	u_int8_t invert;
+};
+
 #endif /*_XT_MARK_H*/
diff --git a/net/netfilter/xt_mark.c b/net/netfilter/xt_mark.c
index ce8735e97627..5cc8cc57c722 100644
--- a/net/netfilter/xt_mark.c
+++ b/net/netfilter/xt_mark.c
@@ -1,10 +1,13 @@
-/* Kernel module to match NFMARK values. */
-
-/* (C) 1999-2001 Marc Boucher <marc@mbsi.ca>
+/*
+ *	xt_mark - Netfilter module to match NFMARK value
+ *
+ *	(C) 1999-2001 Marc Boucher <marc@mbsi.ca>
+ *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
+ *	Jan Engelhardt <jengelh@computergmbh.de>
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
+ *	This program is free software; you can redistribute it and/or modify
+ *	it under the terms of the GNU General Public License version 2 as
+ *	published by the Free Software Foundation.
  */
 
 #include <linux/module.h>
@@ -19,20 +22,31 @@ MODULE_DESCRIPTION("iptables mark matching module");
 MODULE_ALIAS("ipt_mark");
 MODULE_ALIAS("ip6t_mark");
 
+static bool
+mark_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+           const struct net_device *out, const struct xt_match *match,
+           const void *matchinfo, int offset, unsigned int protoff,
+           bool *hotdrop)
+{
+	const struct xt_mark_info *info = matchinfo;
+
+	return ((skb->mark & info->mask) == info->mark) ^ info->invert;
+}
+
 static bool
 mark_mt(const struct sk_buff *skb, const struct net_device *in,
         const struct net_device *out, const struct xt_match *match,
         const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop)
 {
-	const struct xt_mark_info *info = matchinfo;
+	const struct xt_mark_mtinfo1 *info = matchinfo;
 
 	return ((skb->mark & info->mask) == info->mark) ^ info->invert;
 }
 
 static bool
-mark_mt_check(const char *tablename, const void *entry,
-              const struct xt_match *match, void *matchinfo,
-              unsigned int hook_mask)
+mark_mt_check_v0(const char *tablename, const void *entry,
+                 const struct xt_match *match, void *matchinfo,
+                 unsigned int hook_mask)
 {
 	const struct xt_mark_info *minfo = matchinfo;
 
@@ -51,7 +65,7 @@ struct compat_xt_mark_info {
 	u_int16_t	__pad2;
 };
 
-static void mark_mt_compat_from_user(void *dst, void *src)
+static void mark_mt_compat_from_user_v0(void *dst, void *src)
 {
 	const struct compat_xt_mark_info *cm = src;
 	struct xt_mark_info m = {
@@ -62,7 +76,7 @@ static void mark_mt_compat_from_user(void *dst, void *src)
 	memcpy(dst, &m, sizeof(m));
 }
 
-static int mark_mt_compat_to_user(void __user *dst, void *src)
+static int mark_mt_compat_to_user_v0(void __user *dst, void *src)
 {
 	const struct xt_mark_info *m = src;
 	struct compat_xt_mark_info cm = {
@@ -77,30 +91,48 @@ static int mark_mt_compat_to_user(void __user *dst, void *src)
 static struct xt_match mark_mt_reg[] __read_mostly = {
 	{
 		.name		= "mark",
+		.revision	= 0,
 		.family		= AF_INET,
-		.checkentry	= mark_mt_check,
-		.match		= mark_mt,
+		.checkentry	= mark_mt_check_v0,
+		.match		= mark_mt_v0,
 		.matchsize	= sizeof(struct xt_mark_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_info),
-		.compat_from_user = mark_mt_compat_from_user,
-		.compat_to_user	= mark_mt_compat_to_user,
+		.compat_from_user = mark_mt_compat_from_user_v0,
+		.compat_to_user	= mark_mt_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE,
 	},
 	{
 		.name		= "mark",
+		.revision	= 0,
 		.family		= AF_INET6,
-		.checkentry	= mark_mt_check,
-		.match		= mark_mt,
+		.checkentry	= mark_mt_check_v0,
+		.match		= mark_mt_v0,
 		.matchsize	= sizeof(struct xt_mark_info),
 #ifdef CONFIG_COMPAT
 		.compatsize	= sizeof(struct compat_xt_mark_info),
-		.compat_from_user = mark_mt_compat_from_user,
-		.compat_to_user	= mark_mt_compat_to_user,
+		.compat_from_user = mark_mt_compat_from_user_v0,
+		.compat_to_user	= mark_mt_compat_to_user_v0,
 #endif
 		.me		= THIS_MODULE,
 	},
+	{
+		.name           = "mark",
+		.revision       = 1,
+		.family         = AF_INET,
+		.match          = mark_mt,
+		.matchsize      = sizeof(struct xt_mark_mtinfo1),
+		.me             = THIS_MODULE,
+	},
+	{
+		.name           = "mark",
+		.revision       = 1,
+		.family         = AF_INET6,
+		.match          = mark_mt,
+		.matchsize      = sizeof(struct xt_mark_mtinfo1),
+		.me             = THIS_MODULE,
+	},
 };
 
 static int __init mark_mt_init(void)
-- 
cgit v1.2.3


From 917b6fbd6e8dd952c64d1d7468897160467d2cc0 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:42:06 -0800
Subject: [NETFILTER]: xt_policy: use the new union nf_inet_addr

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_policy.h | 23 +++++++++++++++++++----
 net/netfilter/xt_policy.c           | 15 ++++++++-------
 2 files changed, 27 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_policy.h b/include/linux/netfilter/xt_policy.h
index 45654d359a68..053d8cc65464 100644
--- a/include/linux/netfilter/xt_policy.h
+++ b/include/linux/netfilter/xt_policy.h
@@ -27,18 +27,33 @@ struct xt_policy_spec
 			reqid:1;
 };
 
+#ifndef __KERNEL__
 union xt_policy_addr
 {
 	struct in_addr	a4;
 	struct in6_addr	a6;
 };
+#endif
 
 struct xt_policy_elem
 {
-	union xt_policy_addr	saddr;
-	union xt_policy_addr	smask;
-	union xt_policy_addr	daddr;
-	union xt_policy_addr	dmask;
+	union {
+#ifdef __KERNEL__
+		struct {
+			union nf_inet_addr saddr;
+			union nf_inet_addr smask;
+			union nf_inet_addr daddr;
+			union nf_inet_addr dmask;
+		};
+#else
+		struct {
+			union xt_policy_addr saddr;
+			union xt_policy_addr smask;
+			union xt_policy_addr daddr;
+			union xt_policy_addr dmask;
+		};
+#endif
+	};
 	__be32			spi;
 	u_int32_t		reqid;
 	u_int8_t		proto;
diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c
index 45731ca15b7d..47c2e4328a1b 100644
--- a/net/netfilter/xt_policy.c
+++ b/net/netfilter/xt_policy.c
@@ -13,6 +13,7 @@
 #include <linux/init.h>
 #include <net/xfrm.h>
 
+#include <linux/netfilter.h>
 #include <linux/netfilter/xt_policy.h>
 #include <linux/netfilter/x_tables.h>
 
@@ -21,14 +22,14 @@ MODULE_DESCRIPTION("Xtables IPsec policy matching module");
 MODULE_LICENSE("GPL");
 
 static inline bool
-xt_addr_cmp(const union xt_policy_addr *a1, const union xt_policy_addr *m,
-	    const union xt_policy_addr *a2, unsigned short family)
+xt_addr_cmp(const union nf_inet_addr *a1, const union nf_inet_addr *m,
+	    const union nf_inet_addr *a2, unsigned short family)
 {
 	switch (family) {
 	case AF_INET:
-		return !((a1->a4.s_addr ^ a2->a4.s_addr) & m->a4.s_addr);
+		return ((a1->ip ^ a2->ip) & m->ip) == 0;
 	case AF_INET6:
-		return !ipv6_masked_addr_cmp(&a1->a6, &m->a6, &a2->a6);
+		return ipv6_masked_addr_cmp(&a1->in6, &m->in6, &a2->in6) == 0;
 	}
 	return false;
 }
@@ -38,12 +39,12 @@ match_xfrm_state(const struct xfrm_state *x, const struct xt_policy_elem *e,
 		 unsigned short family)
 {
 #define MATCH_ADDR(x,y,z)	(!e->match.x ||			       \
-				 (xt_addr_cmp(&e->x, &e->y, (z), family) \
+				 (xt_addr_cmp(&e->x, &e->y, (const union nf_inet_addr *)(z), family) \
 				  ^ e->invert.x))
 #define MATCH(x,y)		(!e->match.x || ((e->x == (y)) ^ e->invert.x))
 
-	return MATCH_ADDR(saddr, smask, (union xt_policy_addr *)&x->props.saddr) &&
-	       MATCH_ADDR(daddr, dmask, (union xt_policy_addr *)&x->id.daddr) &&
+	return MATCH_ADDR(saddr, smask, &x->props.saddr) &&
+	       MATCH_ADDR(daddr, dmask, &x->id.daddr) &&
 	       MATCH(proto, x->id.proto) &&
 	       MATCH(mode, x->props.mode) &&
 	       MATCH(spi, x->id.spi) &&
-- 
cgit v1.2.3


From f72e25a897c7edda03a0e1f767925d98772684da Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Jan 2008 23:42:47 -0800
Subject: [NETFILTER]: Rename ipt_iprange to xt_iprange

This patch moves ipt_iprange to xt_iprange, in preparation for adding
IPv6 support to xt_iprange.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild             |  1 +
 include/linux/netfilter/xt_iprange.h       | 17 +++++++
 include/linux/netfilter_ipv4/ipt_iprange.h |  6 +--
 net/ipv4/netfilter/Kconfig                 | 10 ----
 net/ipv4/netfilter/Makefile                |  1 -
 net/ipv4/netfilter/ipt_iprange.c           | 77 ------------------------------
 net/netfilter/Kconfig                      | 11 +++++
 net/netfilter/Makefile                     |  1 +
 net/netfilter/xt_iprange.c                 | 76 +++++++++++++++++++++++++++++
 9 files changed, 107 insertions(+), 93 deletions(-)
 create mode 100644 include/linux/netfilter/xt_iprange.h
 delete mode 100644 net/ipv4/netfilter/ipt_iprange.c
 create mode 100644 net/netfilter/xt_iprange.c

(limited to 'include')

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index ac9e6429f747..91fef0cae42f 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -21,6 +21,7 @@ header-y += xt_dccp.h
 header-y += xt_dscp.h
 header-y += xt_esp.h
 header-y += xt_hashlimit.h
+header-y += xt_iprange.h
 header-y += xt_helper.h
 header-y += xt_length.h
 header-y += xt_limit.h
diff --git a/include/linux/netfilter/xt_iprange.h b/include/linux/netfilter/xt_iprange.h
new file mode 100644
index 000000000000..a4299c7d3680
--- /dev/null
+++ b/include/linux/netfilter/xt_iprange.h
@@ -0,0 +1,17 @@
+#ifndef _LINUX_NETFILTER_XT_IPRANGE_H
+#define _LINUX_NETFILTER_XT_IPRANGE_H 1
+
+enum {
+	IPRANGE_SRC     = 1 << 0,	/* match source IP address */
+	IPRANGE_DST     = 1 << 1,	/* match destination IP address */
+	IPRANGE_SRC_INV = 1 << 4,	/* negate the condition */
+	IPRANGE_DST_INV = 1 << 5,	/* -"- */
+};
+
+struct xt_iprange_mtinfo {
+	union nf_inet_addr src_min, src_max;
+	union nf_inet_addr dst_min, dst_max;
+	u_int8_t flags;
+};
+
+#endif /* _LINUX_NETFILTER_XT_IPRANGE_H */
diff --git a/include/linux/netfilter_ipv4/ipt_iprange.h b/include/linux/netfilter_ipv4/ipt_iprange.h
index a92fefc3c7ec..5f1aebde4d2f 100644
--- a/include/linux/netfilter_ipv4/ipt_iprange.h
+++ b/include/linux/netfilter_ipv4/ipt_iprange.h
@@ -2,11 +2,7 @@
 #define _IPT_IPRANGE_H
 
 #include <linux/types.h>
-
-#define IPRANGE_SRC		0x01	/* Match source IP address */
-#define IPRANGE_DST		0x02	/* Match destination IP address */
-#define IPRANGE_SRC_INV		0x10	/* Negate the condition */
-#define IPRANGE_DST_INV		0x20	/* Negate the condition */
+#include <linux/netfilter/xt_iprange.h>
 
 struct ipt_iprange {
 	/* Inclusive: network order. */
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 10ca307b8499..9a077cb24798 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -57,16 +57,6 @@ config IP_NF_IPTABLES
 	  To compile it as a module, choose M here.  If unsure, say N.
 
 # The matches.
-config IP_NF_MATCH_IPRANGE
-	tristate '"iprange" match support'
-	depends on IP_NF_IPTABLES
-	depends on NETFILTER_ADVANCED
-	help
-	  This option makes possible to match IP addresses against IP address
-	  ranges.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 config IP_NF_MATCH_RECENT
 	tristate '"recent" match support'
 	depends on IP_NF_IPTABLES
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index fd7d4a5b436c..0c7dc78a62e9 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -44,7 +44,6 @@ obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o
 obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
 obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o
 obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o
-obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o
 obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o
 obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o
 
diff --git a/net/ipv4/netfilter/ipt_iprange.c b/net/ipv4/netfilter/ipt_iprange.c
deleted file mode 100644
index 9a2aba816c9b..000000000000
--- a/net/ipv4/netfilter/ipt_iprange.c
+++ /dev/null
@@ -1,77 +0,0 @@
-/*
- * iptables module to match IP address ranges
- *
- * (C) 2003 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/module.h>
-#include <linux/skbuff.h>
-#include <linux/ip.h>
-#include <linux/netfilter/x_tables.h>
-#include <linux/netfilter_ipv4/ipt_iprange.h>
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
-MODULE_DESCRIPTION("Xtables: arbitrary IPv4 range matching");
-
-static bool
-iprange_mt(const struct sk_buff *skb, const struct net_device *in,
-           const struct net_device *out, const struct xt_match *match,
-           const void *matchinfo, int offset, unsigned int protoff,
-           bool *hotdrop)
-{
-	const struct ipt_iprange_info *info = matchinfo;
-	const struct iphdr *iph = ip_hdr(skb);
-
-	if (info->flags & IPRANGE_SRC) {
-		if ((ntohl(iph->saddr) < ntohl(info->src.min_ip)
-			  || ntohl(iph->saddr) > ntohl(info->src.max_ip))
-			 ^ !!(info->flags & IPRANGE_SRC_INV)) {
-			pr_debug("src IP %u.%u.%u.%u NOT in range %s"
-				 "%u.%u.%u.%u-%u.%u.%u.%u\n",
-				 NIPQUAD(iph->saddr),
-				 info->flags & IPRANGE_SRC_INV ? "(INV) " : "",
-				 NIPQUAD(info->src.min_ip),
-				 NIPQUAD(info->src.max_ip));
-			return false;
-		}
-	}
-	if (info->flags & IPRANGE_DST) {
-		if ((ntohl(iph->daddr) < ntohl(info->dst.min_ip)
-			  || ntohl(iph->daddr) > ntohl(info->dst.max_ip))
-			 ^ !!(info->flags & IPRANGE_DST_INV)) {
-			pr_debug("dst IP %u.%u.%u.%u NOT in range %s"
-				 "%u.%u.%u.%u-%u.%u.%u.%u\n",
-				 NIPQUAD(iph->daddr),
-				 info->flags & IPRANGE_DST_INV ? "(INV) " : "",
-				 NIPQUAD(info->dst.min_ip),
-				 NIPQUAD(info->dst.max_ip));
-			return false;
-		}
-	}
-	return true;
-}
-
-static struct xt_match iprange_mt_reg __read_mostly = {
-	.name		= "iprange",
-	.family		= AF_INET,
-	.match		= iprange_mt,
-	.matchsize	= sizeof(struct ipt_iprange_info),
-	.me		= THIS_MODULE
-};
-
-static int __init iprange_mt_init(void)
-{
-	return xt_register_match(&iprange_mt_reg);
-}
-
-static void __exit iprange_mt_exit(void)
-{
-	xt_unregister_match(&iprange_mt_reg);
-}
-
-module_init(iprange_mt_init);
-module_exit(iprange_mt_exit);
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 79d71437e310..daf5b881064d 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -567,6 +567,17 @@ config NETFILTER_XT_MATCH_HELPER
 
 	  To compile it as a module, choose M here.  If unsure, say Y.
 
+config NETFILTER_XT_MATCH_IPRANGE
+	tristate '"iprange" address range match support'
+	depends on NETFILTER_XTABLES
+	depends on NETFILTER_ADVANCED
+	---help---
+	This option adds a "iprange" match, which allows you to match based on
+	an IP address range. (Normal iptables only matches on single addresses
+	with an optional mask.)
+
+	If unsure, say M.
+
 config NETFILTER_XT_MATCH_LENGTH
 	tristate '"length" match support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 3b9ea8fb3a07..c910caee0d4f 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -63,6 +63,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_DSCP) += xt_dscp.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_ESP) += xt_esp.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_HELPER) += xt_helper.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_IPRANGE) += xt_iprange.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_LENGTH) += xt_length.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_LIMIT) += xt_limit.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_MAC) += xt_mac.o
diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c
new file mode 100644
index 000000000000..c57a6cf8a081
--- /dev/null
+++ b/net/netfilter/xt_iprange.c
@@ -0,0 +1,76 @@
+/*
+ *	xt_iprange - Netfilter module to match IP address ranges
+ *
+ *	(C) 2003 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
+ *
+ *	This program is free software; you can redistribute it and/or modify
+ *	it under the terms of the GNU General Public License version 2 as
+ *	published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter_ipv4/ipt_iprange.h>
+
+static bool
+iprange_mt_v0(const struct sk_buff *skb, const struct net_device *in,
+              const struct net_device *out, const struct xt_match *match,
+              const void *matchinfo, int offset, unsigned int protoff,
+              bool *hotdrop)
+{
+	const struct ipt_iprange_info *info = matchinfo;
+	const struct iphdr *iph = ip_hdr(skb);
+
+	if (info->flags & IPRANGE_SRC) {
+		if ((ntohl(iph->saddr) < ntohl(info->src.min_ip)
+			  || ntohl(iph->saddr) > ntohl(info->src.max_ip))
+			 ^ !!(info->flags & IPRANGE_SRC_INV)) {
+			pr_debug("src IP %u.%u.%u.%u NOT in range %s"
+				 "%u.%u.%u.%u-%u.%u.%u.%u\n",
+				 NIPQUAD(iph->saddr),
+				 info->flags & IPRANGE_SRC_INV ? "(INV) " : "",
+				 NIPQUAD(info->src.min_ip),
+				 NIPQUAD(info->src.max_ip));
+			return false;
+		}
+	}
+	if (info->flags & IPRANGE_DST) {
+		if ((ntohl(iph->daddr) < ntohl(info->dst.min_ip)
+			  || ntohl(iph->daddr) > ntohl(info->dst.max_ip))
+			 ^ !!(info->flags & IPRANGE_DST_INV)) {
+			pr_debug("dst IP %u.%u.%u.%u NOT in range %s"
+				 "%u.%u.%u.%u-%u.%u.%u.%u\n",
+				 NIPQUAD(iph->daddr),
+				 info->flags & IPRANGE_DST_INV ? "(INV) " : "",
+				 NIPQUAD(info->dst.min_ip),
+				 NIPQUAD(info->dst.max_ip));
+			return false;
+		}
+	}
+	return true;
+}
+
+static struct xt_match iprange_mt_reg __read_mostly = {
+	.name		= "iprange",
+	.family		= AF_INET,
+	.match		= iprange_mt_v0,
+	.matchsize	= sizeof(struct ipt_iprange_info),
+	.me		= THIS_MODULE
+};
+
+static int __init iprange_mt_init(void)
+{
+	return xt_register_match(&iprange_mt_reg);
+}
+
+static void __exit iprange_mt_exit(void)
+{
+	xt_unregister_match(&iprange_mt_reg);
+}
+
+module_init(iprange_mt_init);
+module_exit(iprange_mt_exit);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
+MODULE_DESCRIPTION("Xtables: arbitrary IPv4 range matching");
-- 
cgit v1.2.3


From e37b386c95fff34eb0eebeaf257c4f5a8b69b81f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jan 2008 23:47:44 -0800
Subject: [NETFILTER]: nf_conntrack_sctp: remove unused ttag field from
 conntrack data

Spotted by Pablo Neira Ayuso <pablo@netfilter.org>.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sctp.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h
index 5cf2c115cce4..768f78c4ac53 100644
--- a/include/linux/netfilter/nf_conntrack_sctp.h
+++ b/include/linux/netfilter/nf_conntrack_sctp.h
@@ -21,7 +21,6 @@ struct ip_ct_sctp
 	enum sctp_conntrack state;
 
 	__be32 vtag[IP_CT_DIR_MAX];
-	u_int32_t ttag[IP_CT_DIR_MAX];
 };
 
 #endif /* _NF_CONNTRACK_SCTP_H */
-- 
cgit v1.2.3


From b334aadc3c5cd4dae2a44f3dac09b3ef718ccde1 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jan 2008 23:48:57 -0800
Subject: [NETFILTER]: nf_conntrack: clean up a few header files

- Remove declarations of non-existing variables and functions
- Move helper init/cleanup function declarations to nf_conntrack_helper.h
- Remove unneeded __nf_conntrack_attach declaration and make it static

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h        |  7 -------
 include/net/netfilter/nf_conntrack_core.h   | 12 ------------
 include/net/netfilter/nf_conntrack_helper.h |  4 ++++
 net/netfilter/nf_conntrack_core.c           |  5 ++---
 4 files changed, 6 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 4ac5ab187c2a..857d89951790 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -223,8 +223,6 @@ extern void nf_conntrack_tcp_update(struct sk_buff *skb,
 /* Fake conntrack entry for untracked connections */
 extern struct nf_conn nf_conntrack_untracked;
 
-extern int nf_ct_no_defrag;
-
 /* Iterate over all conntracks: if iter returns true, it's deleted. */
 extern void
 nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data);
@@ -264,10 +262,5 @@ do {							\
 	local_bh_enable();				\
 } while (0)
 
-extern int
-nf_conntrack_register_cache(u_int32_t features, const char *name, size_t size);
-extern void
-nf_conntrack_unregister_cache(u_int32_t features);
-
 #endif /* __KERNEL__ */
 #endif /* _NF_CONNTRACK_H */
diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h
index a532e7b5ed6a..7ad0828f05cf 100644
--- a/include/net/netfilter/nf_conntrack_core.h
+++ b/include/net/netfilter/nf_conntrack_core.h
@@ -30,16 +30,6 @@ extern void nf_conntrack_cleanup(void);
 extern int nf_conntrack_proto_init(void);
 extern void nf_conntrack_proto_fini(void);
 
-extern int nf_conntrack_helper_init(void);
-extern void nf_conntrack_helper_fini(void);
-
-struct nf_conntrack_l3proto;
-extern struct nf_conntrack_l3proto *nf_ct_find_l3proto(u_int16_t pf);
-/* Like above, but you already have conntrack read lock. */
-extern struct nf_conntrack_l3proto *__nf_ct_find_l3proto(u_int16_t l3proto);
-
-struct nf_conntrack_l4proto;
-
 extern int
 nf_ct_get_tuple(const struct sk_buff *skb,
 		unsigned int nhoff,
@@ -76,8 +66,6 @@ static inline int nf_conntrack_confirm(struct sk_buff *skb)
 	return ret;
 }
 
-extern void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb);
-
 int
 print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple,
 	    struct nf_conntrack_l3proto *l3proto,
diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h
index d7b2d5483a71..2f3af00643cf 100644
--- a/include/net/netfilter/nf_conntrack_helper.h
+++ b/include/net/netfilter/nf_conntrack_helper.h
@@ -58,4 +58,8 @@ static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct)
 {
 	return nf_ct_ext_find(ct, NF_CT_EXT_HELPER);
 }
+
+extern int nf_conntrack_helper_init(void);
+extern void nf_conntrack_helper_fini(void);
+
 #endif /*_NF_CONNTRACK_HELPER_H*/
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 25564073ee23..078fff0335ad 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -861,7 +861,7 @@ EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_to_tuple);
 #endif
 
 /* Used by ipt_REJECT and ip6t_REJECT. */
-void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb)
+static void nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb)
 {
 	struct nf_conn *ct;
 	enum ip_conntrack_info ctinfo;
@@ -878,7 +878,6 @@ void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb)
 	nskb->nfctinfo = ctinfo;
 	nf_conntrack_get(nskb->nfct);
 }
-EXPORT_SYMBOL_GPL(__nf_conntrack_attach);
 
 static inline int
 do_iter(const struct nf_conntrack_tuple_hash *i,
@@ -1122,7 +1121,7 @@ int __init nf_conntrack_init(void)
 		goto out_fini_expect;
 
 	/* For use by REJECT target */
-	rcu_assign_pointer(ip_ct_attach, __nf_conntrack_attach);
+	rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach);
 	rcu_assign_pointer(nf_ct_destroy, destroy_conntrack);
 
 	/* Set up fake conntrack:
-- 
cgit v1.2.3


From c56cc9c07b049acc9d2ca97be0b540978c0c80bf Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jan 2008 23:49:17 -0800
Subject: [NETFILTER]: nf_conntrack: remove print_conntrack function from
 l3protos

Its unused and unlikely to ever be used.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack_l3proto.h          | 3 ---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c        | 7 -------
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c | 3 ---
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c        | 7 -------
 net/netfilter/nf_conntrack_l3proto_generic.c          | 7 -------
 net/netfilter/nf_conntrack_standalone.c               | 3 ---
 6 files changed, 30 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h
index 875c6d41eaab..d5526bcce147 100644
--- a/include/net/netfilter/nf_conntrack_l3proto.h
+++ b/include/net/netfilter/nf_conntrack_l3proto.h
@@ -42,9 +42,6 @@ struct nf_conntrack_l3proto
 	int (*print_tuple)(struct seq_file *s,
 			   const struct nf_conntrack_tuple *);
 
-	/* Print out the private part of the conntrack. */
-	int (*print_conntrack)(struct seq_file *s, const struct nf_conn *);
-
 	/* Returns verdict for packet, or -1 for invalid. */
 	int (*packet)(struct nf_conn *conntrack,
 		      const struct sk_buff *skb,
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index 78e64951137a..ac3d61d8026e 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -56,12 +56,6 @@ static int ipv4_print_tuple(struct seq_file *s,
 			  NIPQUAD(tuple->dst.u3.ip));
 }
 
-static int ipv4_print_conntrack(struct seq_file *s,
-				const struct nf_conn *conntrack)
-{
-	return 0;
-}
-
 /* Returns new sk_buff, or NULL */
 static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user)
 {
@@ -403,7 +397,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
 	.pkt_to_tuple	 = ipv4_pkt_to_tuple,
 	.invert_tuple	 = ipv4_invert_tuple,
 	.print_tuple	 = ipv4_print_tuple,
-	.print_conntrack = ipv4_print_conntrack,
 	.get_l4proto	 = ipv4_get_l4proto,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
index 741f3dfaa5a1..acde9952cadf 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
@@ -121,9 +121,6 @@ static int ct_seq_show(struct seq_file *s, void *v)
 		      ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0)
 		return -ENOSPC;
 
-	if (l3proto->print_conntrack(s, ct))
-		return -ENOSPC;
-
 	if (l4proto->print_conntrack(s, ct))
 		return -ENOSPC;
 
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index 97a553036dde..cf42f5cfc338 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -60,12 +60,6 @@ static int ipv6_print_tuple(struct seq_file *s,
 			  NIP6(*((struct in6_addr *)tuple->dst.u3.ip6)));
 }
 
-static int ipv6_print_conntrack(struct seq_file *s,
-				const struct nf_conn *conntrack)
-{
-	return 0;
-}
-
 /*
  * Based on ipv6_skip_exthdr() in net/ipv6/exthdr.c
  *
@@ -376,7 +370,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 __read_mostly = {
 	.pkt_to_tuple		= ipv6_pkt_to_tuple,
 	.invert_tuple		= ipv6_invert_tuple,
 	.print_tuple		= ipv6_print_tuple,
-	.print_conntrack	= ipv6_print_conntrack,
 	.get_l4proto		= ipv6_get_l4proto,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 	.tuple_to_nlattr	= ipv6_tuple_to_nlattr,
diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c
index 991c52c9a28b..8e914e5ffea8 100644
--- a/net/netfilter/nf_conntrack_l3proto_generic.c
+++ b/net/netfilter/nf_conntrack_l3proto_generic.c
@@ -55,12 +55,6 @@ static int generic_print_tuple(struct seq_file *s,
 	return 0;
 }
 
-static int generic_print_conntrack(struct seq_file *s,
-				const struct nf_conn *conntrack)
-{
-	return 0;
-}
-
 static int generic_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
 			       unsigned int *dataoff, u_int8_t *protonum)
 {
@@ -75,7 +69,6 @@ struct nf_conntrack_l3proto nf_conntrack_l3proto_generic __read_mostly = {
 	.pkt_to_tuple	 = generic_pkt_to_tuple,
 	.invert_tuple	 = generic_invert_tuple,
 	.print_tuple	 = generic_print_tuple,
-	.print_conntrack = generic_print_conntrack,
 	.get_l4proto	 = generic_get_l4proto,
 };
 EXPORT_SYMBOL_GPL(nf_conntrack_l3proto_generic);
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 2ad49331302b..d54ca78502c4 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -142,9 +142,6 @@ static int ct_seq_show(struct seq_file *s, void *v)
 		       ? (long)(conntrack->timeout.expires - jiffies)/HZ : 0) != 0)
 		return -ENOSPC;
 
-	if (l3proto->print_conntrack(s, conntrack))
-		return -ENOSPC;
-
 	if (l4proto->print_conntrack(s, conntrack))
 		return -ENOSPC;
 
-- 
cgit v1.2.3


From 569d36452ee26c08523cc9f658901c5188640853 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 18 Jan 2008 03:56:57 -0800
Subject: [NETNS][DST] dst: pass the dst_ops as parameter to the gc functions

The garbage collection function receive the dst_ops structure as
parameter. This is useful for the next incoming patchset because it
will need the dst_ops (there will be several instances) and the
network namespace pointer (contained in the dst_ops).

The protocols which do not take care of the namespaces will not be
impacted by this change (expect for the function signature), they do
just ignore the parameter.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h       | 2 +-
 net/core/dst.c          | 2 +-
 net/decnet/dn_route.c   | 4 ++--
 net/ipv4/route.c        | 6 +++---
 net/ipv4/xfrm4_policy.c | 2 +-
 net/ipv6/route.c        | 4 ++--
 net/ipv6/xfrm6_policy.c | 2 +-
 7 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index e03ea0c04e01..e24a41644c00 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -89,7 +89,7 @@ struct dst_ops
 	__be16			protocol;
 	unsigned		gc_thresh;
 
-	int			(*gc)(void);
+	int			(*gc)(struct dst_ops *ops);
 	struct dst_entry *	(*check)(struct dst_entry *, __u32 cookie);
 	void			(*destroy)(struct dst_entry *);
 	void			(*ifdown)(struct dst_entry *,
diff --git a/net/core/dst.c b/net/core/dst.c
index 7eceebaabaaa..7deef483c79f 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -165,7 +165,7 @@ void * dst_alloc(struct dst_ops * ops)
 	struct dst_entry * dst;
 
 	if (ops->gc && atomic_read(&ops->entries) > ops->gc_thresh) {
-		if (ops->gc())
+		if (ops->gc(ops))
 			return NULL;
 	}
 	dst = kmem_cache_zalloc(ops->kmem_cachep, GFP_ATOMIC);
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 938ba7da217c..31be29b8b5a3 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -107,7 +107,7 @@ static const int dn_rt_mtu_expires = 10 * 60 * HZ;
 
 static unsigned long dn_rt_deadline;
 
-static int dn_dst_gc(void);
+static int dn_dst_gc(struct dst_ops *ops);
 static struct dst_entry *dn_dst_check(struct dst_entry *, __u32);
 static struct dst_entry *dn_dst_negative_advice(struct dst_entry *);
 static void dn_dst_link_failure(struct sk_buff *);
@@ -185,7 +185,7 @@ static void dn_dst_check_expire(unsigned long dummy)
 	mod_timer(&dn_route_timer, now + decnet_dst_gc_interval * HZ);
 }
 
-static int dn_dst_gc(void)
+static int dn_dst_gc(struct dst_ops *ops)
 {
 	struct dn_route *rt, **rtp;
 	int i;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 1e59c0d4b11e..fc0145385e80 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -154,7 +154,7 @@ static void		 ipv4_dst_ifdown(struct dst_entry *dst,
 static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst);
 static void		 ipv4_link_failure(struct sk_buff *skb);
 static void		 ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu);
-static int rt_garbage_collect(void);
+static int rt_garbage_collect(struct dst_ops *ops);
 
 
 static struct dst_ops ipv4_dst_ops = {
@@ -820,7 +820,7 @@ static void rt_secret_rebuild(unsigned long dummy)
    and when load increases it reduces to limit cache size.
  */
 
-static int rt_garbage_collect(void)
+static int rt_garbage_collect(struct dst_ops *ops)
 {
 	static unsigned long expire = RT_GC_TIMEOUT;
 	static unsigned long last_gc;
@@ -1035,7 +1035,7 @@ restart:
 				int saved_int = ip_rt_gc_min_interval;
 				ip_rt_gc_elasticity	= 1;
 				ip_rt_gc_min_interval	= 0;
-				rt_garbage_collect();
+				rt_garbage_collect(&ipv4_dst_ops);
 				ip_rt_gc_min_interval	= saved_int;
 				ip_rt_gc_elasticity	= saved_elasticity;
 				goto restart;
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 656345f75e0d..f04516c880fb 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -185,7 +185,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 	fl->fl4_tos = iph->tos;
 }
 
-static inline int xfrm4_garbage_collect(void)
+static inline int xfrm4_garbage_collect(struct dst_ops *ops)
 {
 	xfrm4_policy_afinfo.garbage_collect();
 	return (atomic_read(&xfrm4_dst_ops.entries) > xfrm4_dst_ops.gc_thresh*2);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index a429900d16aa..4004c5f0b8d7 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -79,7 +79,7 @@ static struct dst_entry *ip6_negative_advice(struct dst_entry *);
 static void		ip6_dst_destroy(struct dst_entry *);
 static void		ip6_dst_ifdown(struct dst_entry *,
 				       struct net_device *dev, int how);
-static int		 ip6_dst_gc(void);
+static int		 ip6_dst_gc(struct dst_ops *ops);
 
 static int		ip6_pkt_discard(struct sk_buff *skb);
 static int		ip6_pkt_discard_out(struct sk_buff *skb);
@@ -983,7 +983,7 @@ int ndisc_dst_gc(int *more)
 	return freed;
 }
 
-static int ip6_dst_gc(void)
+static int ip6_dst_gc(struct dst_ops *ops)
 {
 	static unsigned expire = 30*HZ;
 	static unsigned long last_gc;
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index cf373b46a1ba..c25a6b527fc4 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -212,7 +212,7 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl, int reverse)
 	}
 }
 
-static inline int xfrm6_garbage_collect(void)
+static inline int xfrm6_garbage_collect(struct dst_ops *ops)
 {
 	xfrm6_policy_afinfo.garbage_collect();
 	return (atomic_read(&xfrm6_dst_ops.entries) > xfrm6_dst_ops.gc_thresh*2);
-- 
cgit v1.2.3


From d4fa26ff44e31c2636a985e3092e2cd55d8045de Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 18 Jan 2008 03:58:07 -0800
Subject: [NETNS][DST]: Add the network namespace pointer in dst_ops

The network namespace pointer can be stored into the dst_ops structure.
This is usefull when there are multiple instances of the dst_ops for a
protocol. When there are no several instances, this field will be never
used in the protocol. So there is no impact for the protocols which do
implement the network namespaces.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index e24a41644c00..c45dcc31b3bb 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -102,6 +102,7 @@ struct dst_ops
 
 	atomic_t		entries;
 	struct kmem_cache 		*kmem_cachep;
+	struct net              *dst_net;
 };
 
 #ifdef __KERNEL__
-- 
cgit v1.2.3


From a2fbb9ea235467b0be6db3cec0132b6c83c0b9fb Mon Sep 17 00:00:00 2001
From: Eliezer Tamir <eliezert@broadcom.com>
Date: Thu, 15 Nov 2007 20:09:02 +0200
Subject: add bnx2x driver for BCM57710

Signed-off-by: Eliezer Tamir <eliezert@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig             |    9 +
 drivers/net/Makefile            |    1 +
 drivers/net/bnx2x.c             | 9065 +++++++++++++++++++++++++++++++++++++++
 drivers/net/bnx2x.h             | 1071 +++++
 drivers/net/bnx2x_fw_defs.h     |  198 +
 drivers/net/bnx2x_hsi.h         | 2176 ++++++++++
 drivers/net/bnx2x_init.h        |  564 +++
 drivers/net/bnx2x_init_values.h | 6368 +++++++++++++++++++++++++++
 drivers/net/bnx2x_reg.h         | 4394 +++++++++++++++++++
 include/linux/pci_ids.h         |    1 +
 10 files changed, 23847 insertions(+)
 create mode 100644 drivers/net/bnx2x.c
 create mode 100644 drivers/net/bnx2x.h
 create mode 100644 drivers/net/bnx2x_fw_defs.h
 create mode 100644 drivers/net/bnx2x_hsi.h
 create mode 100644 drivers/net/bnx2x_init.h
 create mode 100644 drivers/net/bnx2x_init_values.h
 create mode 100644 drivers/net/bnx2x_reg.h

(limited to 'include')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 7ae9024b5833..b034410b7ab6 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2597,6 +2597,15 @@ config TEHUTI
 	help
 	  Tehuti Networks 10G Ethernet NIC
 
+config BNX2X
+	tristate "Broadcom NetXtremeII 10Gb support"
+	depends on PCI
+	help
+	  This driver supports Broadcom NetXtremeII 10 gigabit Ethernet cards.
+	  To compile this driver as a module, choose M here: the module
+	  will be called bnx2x.  This is recommended.
+
+
 endif # NETDEV_10000
 
 source "drivers/net/tokenring/Kconfig"
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 5dd2d5eb1918..5e36f203ce5c 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -65,6 +65,7 @@ obj-$(CONFIG_STNIC) += stnic.o 8390.o
 obj-$(CONFIG_FEALNX) += fealnx.o
 obj-$(CONFIG_TIGON3) += tg3.o
 obj-$(CONFIG_BNX2) += bnx2.o
+obj-$(CONFIG_BNX2X) += bnx2x.o
 spidernet-y += spider_net.o spider_net_ethtool.o
 obj-$(CONFIG_SPIDER_NET) += spidernet.o sungem_phy.o
 obj-$(CONFIG_GELIC_NET) += ps3_gelic.o
diff --git a/drivers/net/bnx2x.c b/drivers/net/bnx2x.c
new file mode 100644
index 000000000000..e8c5754798ea
--- /dev/null
+++ b/drivers/net/bnx2x.c
@@ -0,0 +1,9065 @@
+/* bnx2x.c: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * Written by: Eliezer Tamir <eliezert@broadcom.com>
+ * Based on code from Michael Chan's bnx2 driver
+ * UDP CSUM errata workaround by Arik Gendelman
+ * Slowpath rework by Vladislav Zolotarov
+ * Statistics and Link managment by Yitchak Gertner
+ *
+ */
+
+/* define this to make the driver freeze on error
+ * to allow getting debug info
+ * (you will need to reboot afterwords)
+ */
+/*#define BNX2X_STOP_ON_ERROR*/
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/kernel.h>
+#include <linux/device.h>  /* for dev_info() */
+#include <linux/timer.h>
+#include <linux/errno.h>
+#include <linux/ioport.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/skbuff.h>
+#include <linux/dma-mapping.h>
+#include <linux/bitops.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+#include <asm/byteorder.h>
+#include <linux/time.h>
+#include <linux/ethtool.h>
+#include <linux/mii.h>
+#ifdef NETIF_F_HW_VLAN_TX
+	#include <linux/if_vlan.h>
+	#define BCM_VLAN 1
+#endif
+#include <net/ip.h>
+#include <net/tcp.h>
+#include <net/checksum.h>
+#include <linux/workqueue.h>
+#include <linux/crc32.h>
+#include <linux/prefetch.h>
+#include <linux/zlib.h>
+#include <linux/version.h>
+#include <linux/io.h>
+
+#include "bnx2x_reg.h"
+#include "bnx2x_fw_defs.h"
+#include "bnx2x_hsi.h"
+#include "bnx2x.h"
+#include "bnx2x_init.h"
+
+#define DRV_MODULE_VERSION      "0.40.15"
+#define DRV_MODULE_RELDATE      "$DateTime: 2007/11/15 07:28:37 $"
+#define BNX2X_BC_VER    	0x040009
+
+/* Time in jiffies before concluding the transmitter is hung. */
+#define TX_TIMEOUT      	(5*HZ)
+
+static const char version[] __devinitdata =
+	"Broadcom NetXtreme II 577xx 10Gigabit Ethernet Driver "
+	DRV_MODULE_NAME " " DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
+
+MODULE_AUTHOR("Eliezer Tamir <eliezert@broadcom.com>");
+MODULE_DESCRIPTION("Broadcom NetXtreme II BCM57710 Driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_MODULE_VERSION);
+MODULE_INFO(cvs_version, "$Revision: #356 $");
+
+static int use_inta;
+static int poll;
+static int onefunc;
+static int nomcp;
+static int debug;
+static int use_multi;
+
+module_param(use_inta, int, 0);
+module_param(poll, int, 0);
+module_param(onefunc, int, 0);
+module_param(debug, int, 0);
+MODULE_PARM_DESC(use_inta, "use INT#A instead of MSI-X");
+MODULE_PARM_DESC(poll, "use polling (for debug)");
+MODULE_PARM_DESC(onefunc, "enable only first function");
+MODULE_PARM_DESC(nomcp, "ignore managment CPU (Implies onefunc)");
+MODULE_PARM_DESC(debug, "defualt debug msglevel");
+
+#ifdef BNX2X_MULTI
+module_param(use_multi, int, 0);
+MODULE_PARM_DESC(use_multi, "use per-CPU queues");
+#endif
+
+enum bnx2x_board_type {
+	BCM57710 = 0,
+};
+
+/* indexed by board_t, above */
+static const struct {
+	char *name;
+} board_info[] __devinitdata = {
+	{ "Broadcom NetXtreme II BCM57710 XGb" }
+};
+
+static const struct pci_device_id bnx2x_pci_tbl[] = {
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57710,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57710 },
+	{ 0 }
+};
+
+MODULE_DEVICE_TABLE(pci, bnx2x_pci_tbl);
+
+/****************************************************************************
+* General service functions
+****************************************************************************/
+
+/* used only at init
+ * locking is done by mcp
+ */
+static void bnx2x_reg_wr_ind(struct bnx2x *bp, u32 addr, u32 val)
+{
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS, addr);
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_DATA, val);
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS,
+			       PCICFG_VENDOR_ID_OFFSET);
+}
+
+#ifdef BNX2X_IND_RD
+static u32 bnx2x_reg_rd_ind(struct bnx2x *bp, u32 addr)
+{
+	u32 val;
+
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS, addr);
+	pci_read_config_dword(bp->pdev, PCICFG_GRC_DATA, &val);
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS,
+			       PCICFG_VENDOR_ID_OFFSET);
+
+	return val;
+}
+#endif
+
+static const u32 dmae_reg_go_c[] = {
+	DMAE_REG_GO_C0, DMAE_REG_GO_C1, DMAE_REG_GO_C2, DMAE_REG_GO_C3,
+	DMAE_REG_GO_C4, DMAE_REG_GO_C5, DMAE_REG_GO_C6, DMAE_REG_GO_C7,
+	DMAE_REG_GO_C8, DMAE_REG_GO_C9, DMAE_REG_GO_C10, DMAE_REG_GO_C11,
+	DMAE_REG_GO_C12, DMAE_REG_GO_C13, DMAE_REG_GO_C14, DMAE_REG_GO_C15
+};
+
+/* copy command into DMAE command memory and set DMAE command go */
+static void bnx2x_post_dmae(struct bnx2x *bp, struct dmae_command *dmae,
+			    int idx)
+{
+	u32 cmd_offset;
+	int i;
+
+	cmd_offset = (DMAE_REG_CMD_MEM + sizeof(struct dmae_command) * idx);
+	for (i = 0; i < (sizeof(struct dmae_command)/4); i++) {
+		REG_WR(bp, cmd_offset + i*4, *(((u32 *)dmae) + i));
+
+/*      	DP(NETIF_MSG_DMAE, "DMAE cmd[%d].%d (0x%08x) : 0x%08x\n",
+		   idx, i, cmd_offset + i*4, *(((u32 *)dmae) + i)); */
+	}
+	REG_WR(bp, dmae_reg_go_c[idx], 1);
+}
+
+static void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr,
+			     u32 dst_addr, u32 len32)
+{
+	struct dmae_command *dmae = &bp->dmae;
+	int port = bp->port;
+	u32 *wb_comp = bnx2x_sp(bp, wb_comp);
+	int timeout = 200;
+
+	memset(dmae, 0, sizeof(struct dmae_command));
+
+	dmae->opcode = (DMAE_CMD_SRC_PCI | DMAE_CMD_DST_GRC |
+			DMAE_CMD_C_DST_PCI | DMAE_CMD_C_ENABLE |
+			DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+			DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+			DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+			(port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+	dmae->src_addr_lo = U64_LO(dma_addr);
+	dmae->src_addr_hi = U64_HI(dma_addr);
+	dmae->dst_addr_lo = dst_addr >> 2;
+	dmae->dst_addr_hi = 0;
+	dmae->len = len32;
+	dmae->comp_addr_lo = U64_LO(bnx2x_sp_mapping(bp, wb_comp));
+	dmae->comp_addr_hi = U64_HI(bnx2x_sp_mapping(bp, wb_comp));
+	dmae->comp_val = BNX2X_WB_COMP_VAL;
+
+/*
+	DP(NETIF_MSG_DMAE, "dmae: opcode 0x%08x\n"
+	   DP_LEVEL "src_addr  [%x:%08x]  len [%d *4]  "
+		    "dst_addr [%x:%08x (%08x)]\n"
+	   DP_LEVEL "comp_addr [%x:%08x]  comp_val 0x%08x\n",
+	   dmae->opcode, dmae->src_addr_hi, dmae->src_addr_lo,
+	   dmae->len, dmae->dst_addr_hi, dmae->dst_addr_lo, dst_addr,
+	   dmae->comp_addr_hi, dmae->comp_addr_lo, dmae->comp_val);
+*/
+/*
+	DP(NETIF_MSG_DMAE, "data [0x%08x 0x%08x 0x%08x 0x%08x]\n",
+	   bp->slowpath->wb_data[0], bp->slowpath->wb_data[1],
+	   bp->slowpath->wb_data[2], bp->slowpath->wb_data[3]);
+*/
+
+	*wb_comp = 0;
+
+	bnx2x_post_dmae(bp, dmae, port * 8);
+
+	udelay(5);
+	/* adjust timeout for emulation/FPGA */
+	if (CHIP_REV_IS_SLOW(bp))
+		timeout *= 100;
+	while (*wb_comp != BNX2X_WB_COMP_VAL) {
+/*      	DP(NETIF_MSG_DMAE, "wb_comp 0x%08x\n", *wb_comp); */
+		udelay(5);
+		if (!timeout) {
+			BNX2X_ERR("dmae timeout!\n");
+			break;
+		}
+		timeout--;
+	}
+}
+
+#ifdef BNX2X_DMAE_RD
+static void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
+{
+	struct dmae_command *dmae = &bp->dmae;
+	int port = bp->port;
+	u32 *wb_comp = bnx2x_sp(bp, wb_comp);
+	int timeout = 200;
+
+	memset(bnx2x_sp(bp, wb_data[0]), 0, sizeof(u32) * 4);
+	memset(dmae, 0, sizeof(struct dmae_command));
+
+	dmae->opcode = (DMAE_CMD_SRC_GRC | DMAE_CMD_DST_PCI |
+			DMAE_CMD_C_DST_PCI | DMAE_CMD_C_ENABLE |
+			DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+			DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+			DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+			(port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+	dmae->src_addr_lo = src_addr >> 2;
+	dmae->src_addr_hi = 0;
+	dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, wb_data));
+	dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, wb_data));
+	dmae->len = len32;
+	dmae->comp_addr_lo = U64_LO(bnx2x_sp_mapping(bp, wb_comp));
+	dmae->comp_addr_hi = U64_HI(bnx2x_sp_mapping(bp, wb_comp));
+	dmae->comp_val = BNX2X_WB_COMP_VAL;
+
+/*
+	DP(NETIF_MSG_DMAE, "dmae: opcode 0x%08x\n"
+	   DP_LEVEL "src_addr  [%x:%08x]  len [%d *4]  "
+		    "dst_addr [%x:%08x (%08x)]\n"
+	   DP_LEVEL "comp_addr [%x:%08x]  comp_val 0x%08x\n",
+	   dmae->opcode, dmae->src_addr_hi, dmae->src_addr_lo,
+	   dmae->len, dmae->dst_addr_hi, dmae->dst_addr_lo, src_addr,
+	   dmae->comp_addr_hi, dmae->comp_addr_lo, dmae->comp_val);
+*/
+
+	*wb_comp = 0;
+
+	bnx2x_post_dmae(bp, dmae, port * 8);
+
+	udelay(5);
+	while (*wb_comp != BNX2X_WB_COMP_VAL) {
+		udelay(5);
+		if (!timeout) {
+			BNX2X_ERR("dmae timeout!\n");
+			break;
+		}
+		timeout--;
+	}
+/*
+	DP(NETIF_MSG_DMAE, "data [0x%08x 0x%08x 0x%08x 0x%08x]\n",
+	   bp->slowpath->wb_data[0], bp->slowpath->wb_data[1],
+	   bp->slowpath->wb_data[2], bp->slowpath->wb_data[3]);
+*/
+}
+#endif
+
+static int bnx2x_mc_assert(struct bnx2x *bp)
+{
+	int i, j;
+	int rc = 0;
+	char last_idx;
+	const char storm[] = {"XTCU"};
+	const u32 intmem_base[] = {
+		BAR_XSTRORM_INTMEM,
+		BAR_TSTRORM_INTMEM,
+		BAR_CSTRORM_INTMEM,
+		BAR_USTRORM_INTMEM
+	};
+
+	/* Go through all instances of all SEMIs */
+	for (i = 0; i < 4; i++) {
+		last_idx = REG_RD8(bp, XSTORM_ASSERT_LIST_INDEX_OFFSET +
+				   intmem_base[i]);
+		BNX2X_ERR("DATA %cSTORM_ASSERT_LIST_INDEX 0x%x\n",
+			  storm[i], last_idx);
+
+		/* print the asserts */
+		for (j = 0; j < STROM_ASSERT_ARRAY_SIZE; j++) {
+			u32 row0, row1, row2, row3;
+
+			row0 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) +
+				      intmem_base[i]);
+			row1 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 4 +
+				      intmem_base[i]);
+			row2 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 8 +
+				      intmem_base[i]);
+			row3 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 12 +
+				      intmem_base[i]);
+
+			if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+				BNX2X_ERR("DATA %cSTORM_ASSERT_INDEX 0x%x ="
+					  " 0x%08x 0x%08x 0x%08x 0x%08x\n",
+					  storm[i], j, row3, row2, row1, row0);
+				rc++;
+			} else {
+				break;
+			}
+		}
+	}
+	return rc;
+}
+static void bnx2x_fw_dump(struct bnx2x *bp)
+{
+	u32 mark, offset;
+	u32 data[9];
+	int word;
+
+	mark = REG_RD(bp, MCP_REG_MCPR_SCRATCH + 0xf104);
+	printk(KERN_ERR PFX "begin fw dump (mark 0x%x)\n", mark);
+
+	for (offset = mark - 0x08000000; offset <= 0xF900; offset += 0x8*4) {
+		for (word = 0; word < 8; word++)
+			data[word] = htonl(REG_RD(bp, MCP_REG_MCPR_SCRATCH +
+						  offset + 4*word));
+		data[8] = 0x0;
+		printk(KERN_ERR PFX "%s", (char *)data);
+	}
+	for (offset = 0xF108; offset <= mark - 0x08000000; offset += 0x8*4) {
+		for (word = 0; word < 8; word++)
+			data[word] = htonl(REG_RD(bp, MCP_REG_MCPR_SCRATCH +
+						  offset + 4*word));
+		data[8] = 0x0;
+		printk(KERN_ERR PFX "%s", (char *)data);
+	}
+	printk("\n" KERN_ERR PFX "end of fw dump\n");
+}
+
+static void bnx2x_panic_dump(struct bnx2x *bp)
+{
+	int i;
+	u16 j, start, end;
+
+	BNX2X_ERR("begin crash dump -----------------\n");
+
+	for_each_queue(bp, i) {
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+		struct eth_tx_db_data *hw_prods = fp->hw_tx_prods;
+
+		BNX2X_ERR("queue[%d]: tx_pkt_prod(%x)  tx_pkt_cons(%x)"
+			  "  tx_bd_prod(%x)  tx_bd_cons(%x)  *tx_cons_sb(%x)"
+			  "  *rx_cons_sb(%x)  rx_comp_prod(%x)"
+			  "  rx_comp_cons(%x)  fp_c_idx(%x)  fp_u_idx(%x)"
+			  "  bd data(%x,%x)\n",
+			  i, fp->tx_pkt_prod, fp->tx_pkt_cons, fp->tx_bd_prod,
+			  fp->tx_bd_cons, *fp->tx_cons_sb, *fp->rx_cons_sb,
+			  fp->rx_comp_prod, fp->rx_comp_cons, fp->fp_c_idx,
+			  fp->fp_u_idx, hw_prods->packets_prod,
+			  hw_prods->bds_prod);
+
+		start = TX_BD(le16_to_cpu(*fp->tx_cons_sb) - 10);
+		end = TX_BD(le16_to_cpu(*fp->tx_cons_sb) + 245);
+		for (j = start; j < end; j++) {
+			struct sw_tx_bd *sw_bd = &fp->tx_buf_ring[j];
+
+			BNX2X_ERR("packet[%x]=[%p,%x]\n", j,
+				  sw_bd->skb, sw_bd->first_bd);
+		}
+
+		start = TX_BD(fp->tx_bd_cons - 10);
+		end = TX_BD(fp->tx_bd_cons + 254);
+		for (j = start; j < end; j++) {
+			u32 *tx_bd = (u32 *)&fp->tx_desc_ring[j];
+
+			BNX2X_ERR("tx_bd[%x]=[%x:%x:%x:%x]\n",
+				  j, tx_bd[0], tx_bd[1], tx_bd[2], tx_bd[3]);
+		}
+
+		start = RX_BD(le16_to_cpu(*fp->rx_cons_sb) - 10);
+		end = RX_BD(le16_to_cpu(*fp->rx_cons_sb) + 503);
+		for (j = start; j < end; j++) {
+			u32 *rx_bd = (u32 *)&fp->rx_desc_ring[j];
+			struct sw_rx_bd *sw_bd = &fp->rx_buf_ring[j];
+
+			BNX2X_ERR("rx_bd[%x]=[%x:%x]  sw_bd=[%p]\n",
+				  j, rx_bd[0], rx_bd[1], sw_bd->skb);
+		}
+
+		start = RCQ_BD(fp->rx_comp_cons - 10);
+		end = RCQ_BD(fp->rx_comp_cons + 503);
+		for (j = start; j < end; j++) {
+			u32 *cqe = (u32 *)&fp->rx_comp_ring[j];
+
+			BNX2X_ERR("cqe[%x]=[%x:%x:%x:%x]\n",
+				  j, cqe[0], cqe[1], cqe[2], cqe[3]);
+		}
+	}
+
+	BNX2X_ERR("def_c_idx(%u)  def_u_idx(%u)  def_t_idx(%u)"
+		  "  def_x_idx(%u)  def_att_idx(%u)  attn_state(%u)"
+		  "  spq_prod_idx(%u)\n",
+		  bp->def_c_idx, bp->def_u_idx, bp->def_t_idx, bp->def_x_idx,
+		  bp->def_att_idx, bp->attn_state, bp->spq_prod_idx);
+
+
+	bnx2x_mc_assert(bp);
+	BNX2X_ERR("end crash dump -----------------\n");
+
+	bp->stats_state = STATS_STATE_DISABLE;
+	DP(BNX2X_MSG_STATS, "stats_state - DISABLE\n");
+}
+
+static void bnx2x_enable_int(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
+	u32 val = REG_RD(bp, addr);
+	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
+
+	if (msix) {
+		val &= ~HC_CONFIG_0_REG_SINGLE_ISR_EN_0;
+		val |= (HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 |
+			HC_CONFIG_0_REG_ATTN_BIT_EN_0);
+	} else {
+		val |= (HC_CONFIG_0_REG_SINGLE_ISR_EN_0 |
+			HC_CONFIG_0_REG_INT_LINE_EN_0 |
+			HC_CONFIG_0_REG_ATTN_BIT_EN_0);
+		val &= ~HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0;
+	}
+
+	DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)  msi %d\n",
+	   val, port, addr, msix);
+
+	REG_WR(bp, addr, val);
+}
+
+static void bnx2x_disable_int(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
+	u32 val = REG_RD(bp, addr);
+
+	val &= ~(HC_CONFIG_0_REG_SINGLE_ISR_EN_0 |
+		 HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 |
+		 HC_CONFIG_0_REG_INT_LINE_EN_0 |
+		 HC_CONFIG_0_REG_ATTN_BIT_EN_0);
+
+	DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)\n",
+	   val, port, addr);
+
+	REG_WR(bp, addr, val);
+	if (REG_RD(bp, addr) != val)
+		BNX2X_ERR("BUG! proper val not read from IGU!\n");
+}
+
+static void bnx2x_disable_int_sync(struct bnx2x *bp)
+{
+
+	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
+	int i;
+
+	atomic_inc(&bp->intr_sem);
+	/* prevent the HW from sending interrupts*/
+	bnx2x_disable_int(bp);
+
+	/* make sure all ISRs are done */
+	if (msix) {
+		for_each_queue(bp, i)
+			synchronize_irq(bp->msix_table[i].vector);
+
+		/* one more for the Slow Path IRQ */
+		synchronize_irq(bp->msix_table[i].vector);
+	} else
+		synchronize_irq(bp->pdev->irq);
+
+	/* make sure sp_task is not running */
+	cancel_work_sync(&bp->sp_task);
+
+}
+
+/* fast path code */
+
+/*
+ * general service functions
+ */
+
+static inline void bnx2x_ack_sb(struct bnx2x *bp, u8 id,
+				u8 storm, u16 index, u8 op, u8 update)
+{
+	u32 igu_addr = (IGU_ADDR_INT_ACK + IGU_PORT_BASE * bp->port) * 8;
+	struct igu_ack_register igu_ack;
+
+	igu_ack.status_block_index = index;
+	igu_ack.sb_id_and_flags =
+			((id << IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT) |
+			 (storm << IGU_ACK_REGISTER_STORM_ID_SHIFT) |
+			 (update << IGU_ACK_REGISTER_UPDATE_INDEX_SHIFT) |
+			 (op << IGU_ACK_REGISTER_INTERRUPT_MODE_SHIFT));
+
+/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
+	   (*(u32 *)&igu_ack), BAR_IGU_INTMEM + igu_addr); */
+	REG_WR(bp, BAR_IGU_INTMEM + igu_addr, (*(u32 *)&igu_ack));
+}
+
+static inline u16 bnx2x_update_fpsb_idx(struct bnx2x_fastpath *fp)
+{
+	struct host_status_block *fpsb = fp->status_blk;
+	u16 rc = 0;
+
+	barrier(); /* status block is written to by the chip */
+	if (fp->fp_c_idx != fpsb->c_status_block.status_block_index) {
+		fp->fp_c_idx = fpsb->c_status_block.status_block_index;
+		rc |= 1;
+	}
+	if (fp->fp_u_idx != fpsb->u_status_block.status_block_index) {
+		fp->fp_u_idx = fpsb->u_status_block.status_block_index;
+		rc |= 2;
+	}
+	return rc;
+}
+
+static inline int bnx2x_has_work(struct bnx2x_fastpath *fp)
+{
+	u16 rx_cons_sb = le16_to_cpu(*fp->rx_cons_sb);
+
+	if ((rx_cons_sb & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
+		rx_cons_sb++;
+
+	if ((rx_cons_sb != fp->rx_comp_cons) ||
+	    (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons))
+		return 1;
+
+	return 0;
+}
+
+static u16 bnx2x_ack_int(struct bnx2x *bp)
+{
+	u32 igu_addr = (IGU_ADDR_SIMD_MASK + IGU_PORT_BASE * bp->port) * 8;
+	u32 result = REG_RD(bp, BAR_IGU_INTMEM + igu_addr);
+
+/*      DP(NETIF_MSG_INTR, "read 0x%08x from IGU addr 0x%x\n",
+	   result, BAR_IGU_INTMEM + igu_addr); */
+
+#ifdef IGU_DEBUG
+#warning IGU_DEBUG active
+	if (result == 0) {
+		BNX2X_ERR("read %x from IGU\n", result);
+		REG_WR(bp, TM_REG_TIMER_SOFT_RST, 0);
+	}
+#endif
+	return result;
+}
+
+
+/*
+ * fast path service functions
+ */
+
+/* free skb in the packet ring at pos idx
+ * return idx of last bd freed
+ */
+static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
+			     u16 idx)
+{
+	struct sw_tx_bd *tx_buf = &fp->tx_buf_ring[idx];
+	struct eth_tx_bd *tx_bd;
+	struct sk_buff *skb = tx_buf->skb;
+	u16 bd_idx = tx_buf->first_bd;
+	int nbd;
+
+	DP(BNX2X_MSG_OFF, "pkt_idx %d  buff @(%p)->skb %p\n",
+	   idx, tx_buf, skb);
+
+	/* unmap first bd */
+	DP(BNX2X_MSG_OFF, "free bd_idx %d\n", bd_idx);
+	tx_bd = &fp->tx_desc_ring[bd_idx];
+	pci_unmap_single(bp->pdev, BD_UNMAP_ADDR(tx_bd),
+			 BD_UNMAP_LEN(tx_bd), PCI_DMA_TODEVICE);
+
+	nbd = le16_to_cpu(tx_bd->nbd) - 1;
+#ifdef BNX2X_STOP_ON_ERROR
+	if (nbd > (MAX_SKB_FRAGS + 2)) {
+		BNX2X_ERR("bad nbd!\n");
+		bnx2x_panic();
+	}
+#endif
+
+	/* Skip a parse bd and the TSO split header bd
+	   since they have no mapping */
+	if (nbd)
+		bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+
+	if (tx_bd->bd_flags.as_bitfield & (ETH_TX_BD_FLAGS_IP_CSUM |
+					   ETH_TX_BD_FLAGS_TCP_CSUM |
+					   ETH_TX_BD_FLAGS_SW_LSO)) {
+		if (--nbd)
+			bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+		tx_bd = &fp->tx_desc_ring[bd_idx];
+		/* is this a TSO split header bd? */
+		if (tx_bd->bd_flags.as_bitfield & ETH_TX_BD_FLAGS_SW_LSO) {
+			if (--nbd)
+				bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+		}
+	}
+
+	/* now free frags */
+	while (nbd > 0) {
+
+		DP(BNX2X_MSG_OFF, "free frag bd_idx %d\n", bd_idx);
+		tx_bd = &fp->tx_desc_ring[bd_idx];
+		pci_unmap_page(bp->pdev, BD_UNMAP_ADDR(tx_bd),
+			       BD_UNMAP_LEN(tx_bd), PCI_DMA_TODEVICE);
+		if (--nbd)
+			bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+	}
+
+	/* release skb */
+	BUG_TRAP(skb);
+	dev_kfree_skb(skb);
+	tx_buf->first_bd = 0;
+	tx_buf->skb = NULL;
+
+	return bd_idx;
+}
+
+static inline u32 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
+{
+	u16 used;
+	u32 prod;
+	u32 cons;
+
+	/* Tell compiler that prod and cons can change */
+	barrier();
+	prod = fp->tx_bd_prod;
+	cons = fp->tx_bd_cons;
+
+	used = (NUM_TX_BD - NUM_TX_RINGS + prod - cons +
+		(cons / TX_DESC_CNT) - (prod / TX_DESC_CNT));
+
+	if (prod >= cons) {
+		/* used = prod - cons - prod/size + cons/size */
+		used -= NUM_TX_BD - NUM_TX_RINGS;
+	}
+
+	BUG_TRAP(used <= fp->bp->tx_ring_size);
+	BUG_TRAP((fp->bp->tx_ring_size - used) <= MAX_TX_AVAIL);
+
+	return (fp->bp->tx_ring_size - used);
+}
+
+static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
+{
+	struct bnx2x *bp = fp->bp;
+	u16 hw_cons, sw_cons, bd_cons = fp->tx_bd_cons;
+	int done = 0;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return;
+#endif
+
+	hw_cons = le16_to_cpu(*fp->tx_cons_sb);
+	sw_cons = fp->tx_pkt_cons;
+
+	while (sw_cons != hw_cons) {
+		u16 pkt_cons;
+
+		pkt_cons = TX_BD(sw_cons);
+
+		/* prefetch(bp->tx_buf_ring[pkt_cons].skb); */
+
+		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %d\n",
+		   hw_cons, sw_cons, pkt_cons);
+
+/*      	if (NEXT_TX_IDX(sw_cons) != hw_cons) {
+			rmb();
+			prefetch(fp->tx_buf_ring[NEXT_TX_IDX(sw_cons)].skb);
+		}
+*/
+		bd_cons = bnx2x_free_tx_pkt(bp, fp, pkt_cons);
+		sw_cons++;
+		done++;
+
+		if (done == work)
+			break;
+	}
+
+	fp->tx_pkt_cons = sw_cons;
+	fp->tx_bd_cons = bd_cons;
+
+	/* Need to make the tx_cons update visible to start_xmit()
+	 * before checking for netif_queue_stopped().  Without the
+	 * memory barrier, there is a small possibility that start_xmit()
+	 * will miss it and cause the queue to be stopped forever.
+	 */
+	smp_mb();
+
+	/* TBD need a thresh? */
+	if (unlikely(netif_queue_stopped(bp->dev))) {
+
+		netif_tx_lock(bp->dev);
+
+		if (netif_queue_stopped(bp->dev) &&
+		    (bnx2x_tx_avail(fp) >= MAX_SKB_FRAGS + 3))
+			netif_wake_queue(bp->dev);
+
+		netif_tx_unlock(bp->dev);
+
+	}
+}
+
+static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
+			   union eth_rx_cqe *rr_cqe)
+{
+	struct bnx2x *bp = fp->bp;
+	int cid = SW_CID(rr_cqe->ramrod_cqe.conn_and_cmd_data);
+	int command = CQE_CMD(rr_cqe->ramrod_cqe.conn_and_cmd_data);
+
+	DP(NETIF_MSG_RX_STATUS,
+	   "fp %d  cid %d  got ramrod #%d  state is %x  type is %d\n",
+	   fp->index, cid, command, bp->state, rr_cqe->ramrod_cqe.type);
+
+	bp->spq_left++;
+
+	if (fp->index) {
+		switch (command | fp->state) {
+		case (RAMROD_CMD_ID_ETH_CLIENT_SETUP |
+						BNX2X_FP_STATE_OPENING):
+			DP(NETIF_MSG_IFUP, "got MULTI[%d] setup ramrod\n",
+			   cid);
+			fp->state = BNX2X_FP_STATE_OPEN;
+			break;
+
+		case (RAMROD_CMD_ID_ETH_HALT | BNX2X_FP_STATE_HALTING):
+			DP(NETIF_MSG_IFDOWN, "got MULTI[%d] halt ramrod\n",
+			   cid);
+			fp->state = BNX2X_FP_STATE_HALTED;
+			break;
+
+		default:
+			BNX2X_ERR("unexpected MC reply(%d)  state is %x\n",
+				  command, fp->state);
+		}
+		mb(); /* force bnx2x_wait_ramrod to see the change */
+		return;
+	}
+	switch (command | bp->state) {
+	case (RAMROD_CMD_ID_ETH_PORT_SETUP | BNX2X_STATE_OPENING_WAIT4_PORT):
+		DP(NETIF_MSG_IFUP, "got setup ramrod\n");
+		bp->state = BNX2X_STATE_OPEN;
+		break;
+
+	case (RAMROD_CMD_ID_ETH_HALT | BNX2X_STATE_CLOSING_WAIT4_HALT):
+		DP(NETIF_MSG_IFDOWN, "got halt ramrod\n");
+		bp->state = BNX2X_STATE_CLOSING_WAIT4_DELETE;
+		fp->state = BNX2X_FP_STATE_HALTED;
+		break;
+
+	case (RAMROD_CMD_ID_ETH_PORT_DEL | BNX2X_STATE_CLOSING_WAIT4_DELETE):
+		DP(NETIF_MSG_IFDOWN, "got delete ramrod\n");
+		bp->state = BNX2X_STATE_CLOSING_WAIT4_UNLOAD;
+		break;
+
+	case (RAMROD_CMD_ID_ETH_CFC_DEL | BNX2X_STATE_CLOSING_WAIT4_HALT):
+		DP(NETIF_MSG_IFDOWN, "got delete ramrod for MULTI[%d]\n", cid);
+		bnx2x_fp(bp, cid, state) = BNX2X_FP_STATE_DELETED;
+		break;
+
+	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_OPEN):
+		DP(NETIF_MSG_IFUP, "got set mac ramrod\n");
+		break;
+
+	default:
+		BNX2X_ERR("unexpected ramrod (%d)  state is %x\n",
+			  command, bp->state);
+	}
+
+	mb(); /* force bnx2x_wait_ramrod to see the change */
+}
+
+static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
+				     struct bnx2x_fastpath *fp, u16 index)
+{
+	struct sk_buff *skb;
+	struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[index];
+	struct eth_rx_bd *rx_bd = &fp->rx_desc_ring[index];
+	dma_addr_t mapping;
+
+	skb = netdev_alloc_skb(bp->dev, bp->rx_buf_size);
+	if (unlikely(skb == NULL))
+		return -ENOMEM;
+
+	mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size,
+				 PCI_DMA_FROMDEVICE);
+	if (unlikely(dma_mapping_error(mapping))) {
+
+		dev_kfree_skb(skb);
+		return -ENOMEM;
+	}
+
+	rx_buf->skb = skb;
+	pci_unmap_addr_set(rx_buf, mapping, mapping);
+
+	rx_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
+	rx_bd->addr_lo = cpu_to_le32(U64_LO(mapping));
+
+	return 0;
+}
+
+/* note that we are not allocating a new skb,
+ * we are just moving one from cons to prod
+ * we are not creating a new mapping,
+ * so there is no need to check for dma_mapping_error().
+ */
+static void bnx2x_reuse_rx_skb(struct bnx2x_fastpath *fp,
+			       struct sk_buff *skb, u16 cons, u16 prod)
+{
+	struct bnx2x *bp = fp->bp;
+	struct sw_rx_bd *cons_rx_buf = &fp->rx_buf_ring[cons];
+	struct sw_rx_bd *prod_rx_buf = &fp->rx_buf_ring[prod];
+	struct eth_rx_bd *cons_bd = &fp->rx_desc_ring[cons];
+	struct eth_rx_bd *prod_bd = &fp->rx_desc_ring[prod];
+
+	pci_dma_sync_single_for_device(bp->pdev,
+				       pci_unmap_addr(cons_rx_buf, mapping),
+				       bp->rx_offset + RX_COPY_THRESH,
+				       PCI_DMA_FROMDEVICE);
+
+	prod_rx_buf->skb = cons_rx_buf->skb;
+	pci_unmap_addr_set(prod_rx_buf, mapping,
+			   pci_unmap_addr(cons_rx_buf, mapping));
+	*prod_bd = *cons_bd;
+}
+
+static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
+{
+	struct bnx2x *bp = fp->bp;
+	u16 bd_cons, bd_prod, comp_ring_cons;
+	u16 hw_comp_cons, sw_comp_cons, sw_comp_prod;
+	int rx_pkt = 0;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return 0;
+#endif
+
+	hw_comp_cons = le16_to_cpu(*fp->rx_cons_sb);
+	if ((hw_comp_cons & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
+		hw_comp_cons++;
+
+	bd_cons = fp->rx_bd_cons;
+	bd_prod = fp->rx_bd_prod;
+	sw_comp_cons = fp->rx_comp_cons;
+	sw_comp_prod = fp->rx_comp_prod;
+
+	/* Memory barrier necessary as speculative reads of the rx
+	 * buffer can be ahead of the index in the status block
+	 */
+	rmb();
+
+	DP(NETIF_MSG_RX_STATUS,
+	   "queue[%d]:  hw_comp_cons %u  sw_comp_cons %u\n",
+	   fp->index, hw_comp_cons, sw_comp_cons);
+
+	while (sw_comp_cons != hw_comp_cons) {
+		unsigned int len, pad;
+		struct sw_rx_bd *rx_buf;
+		struct sk_buff *skb;
+		union eth_rx_cqe *cqe;
+
+		comp_ring_cons = RCQ_BD(sw_comp_cons);
+		bd_prod = RX_BD(bd_prod);
+		bd_cons = RX_BD(bd_cons);
+
+		cqe = &fp->rx_comp_ring[comp_ring_cons];
+
+		DP(NETIF_MSG_RX_STATUS, "hw_comp_cons %u  sw_comp_cons %u"
+		   "  comp_ring (%u)  bd_ring (%u,%u)\n",
+		   hw_comp_cons, sw_comp_cons,
+		   comp_ring_cons, bd_prod, bd_cons);
+		DP(NETIF_MSG_RX_STATUS, "CQE type %x  err %x  status %x"
+		   "  queue %x  vlan %x  len %x\n",
+		   cqe->fast_path_cqe.type,
+		   cqe->fast_path_cqe.error_type_flags,
+		   cqe->fast_path_cqe.status_flags,
+		   cqe->fast_path_cqe.rss_hash_result,
+		   cqe->fast_path_cqe.vlan_tag, cqe->fast_path_cqe.pkt_len);
+
+		/* is this a slowpath msg? */
+		if (unlikely(cqe->fast_path_cqe.type)) {
+			bnx2x_sp_event(fp, cqe);
+			goto next_cqe;
+
+		/* this is an rx packet */
+		} else {
+			rx_buf = &fp->rx_buf_ring[bd_cons];
+			skb = rx_buf->skb;
+
+			len = le16_to_cpu(cqe->fast_path_cqe.pkt_len);
+			pad = cqe->fast_path_cqe.placement_offset;
+
+			pci_dma_sync_single_for_device(bp->pdev,
+					pci_unmap_addr(rx_buf, mapping),
+						       pad + RX_COPY_THRESH,
+						       PCI_DMA_FROMDEVICE);
+			prefetch(skb);
+			prefetch(((char *)(skb)) + 128);
+
+			/* is this an error packet? */
+			if (unlikely(cqe->fast_path_cqe.error_type_flags &
+							ETH_RX_ERROR_FALGS)) {
+			/* do we sometimes forward error packets anyway? */
+				DP(NETIF_MSG_RX_ERR,
+				   "ERROR flags(%u) Rx packet(%u)\n",
+				   cqe->fast_path_cqe.error_type_flags,
+				   sw_comp_cons);
+				/* TBD make sure MC counts this as a drop */
+				goto reuse_rx;
+			}
+
+			/* Since we don't have a jumbo ring
+			 * copy small packets if mtu > 1500
+			 */
+			if ((bp->dev->mtu > ETH_MAX_PACKET_SIZE) &&
+			    (len <= RX_COPY_THRESH)) {
+				struct sk_buff *new_skb;
+
+				new_skb = netdev_alloc_skb(bp->dev,
+							   len + pad);
+				if (new_skb == NULL) {
+					DP(NETIF_MSG_RX_ERR,
+					   "ERROR packet dropped "
+					   "because of alloc failure\n");
+					/* TBD count this as a drop? */
+					goto reuse_rx;
+				}
+
+				/* aligned copy */
+				skb_copy_from_linear_data_offset(skb, pad,
+						    new_skb->data + pad, len);
+				skb_reserve(new_skb, pad);
+				skb_put(new_skb, len);
+
+				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
+
+				skb = new_skb;
+
+			} else if (bnx2x_alloc_rx_skb(bp, fp, bd_prod) == 0) {
+				pci_unmap_single(bp->pdev,
+					pci_unmap_addr(rx_buf, mapping),
+						 bp->rx_buf_use_size,
+						 PCI_DMA_FROMDEVICE);
+				skb_reserve(skb, pad);
+				skb_put(skb, len);
+
+			} else {
+				DP(NETIF_MSG_RX_ERR,
+				   "ERROR packet dropped because "
+				   "of alloc failure\n");
+reuse_rx:
+				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
+				goto next_rx;
+			}
+
+			skb->protocol = eth_type_trans(skb, bp->dev);
+
+			skb->ip_summed = CHECKSUM_NONE;
+			if (bp->rx_csum && BNX2X_RX_SUM_OK(cqe))
+				skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+			/* TBD do we pass bad csum packets in promisc */
+		}
+
+#ifdef BCM_VLAN
+		if ((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags)
+				& PARSING_FLAGS_NUMBER_OF_NESTED_VLANS)
+		    && (bp->vlgrp != NULL))
+			vlan_hwaccel_receive_skb(skb, bp->vlgrp,
+				le16_to_cpu(cqe->fast_path_cqe.vlan_tag));
+		else
+#endif
+		netif_receive_skb(skb);
+
+		bp->dev->last_rx = jiffies;
+
+next_rx:
+		rx_buf->skb = NULL;
+
+		bd_cons = NEXT_RX_IDX(bd_cons);
+		bd_prod = NEXT_RX_IDX(bd_prod);
+next_cqe:
+		sw_comp_prod = NEXT_RCQ_IDX(sw_comp_prod);
+		sw_comp_cons = NEXT_RCQ_IDX(sw_comp_cons);
+		rx_pkt++;
+
+		if ((rx_pkt == budget))
+			break;
+	} /* while */
+
+	fp->rx_bd_cons = bd_cons;
+	fp->rx_bd_prod = bd_prod;
+	fp->rx_comp_cons = sw_comp_cons;
+	fp->rx_comp_prod = sw_comp_prod;
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_RCQ_PROD_OFFSET(bp->port, fp->index), sw_comp_prod);
+
+	mmiowb(); /* keep prod updates ordered */
+
+	fp->rx_pkt += rx_pkt;
+	fp->rx_calls++;
+
+	return rx_pkt;
+}
+
+static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
+{
+	struct bnx2x_fastpath *fp = fp_cookie;
+	struct bnx2x *bp = fp->bp;
+	struct net_device *dev = bp->dev;
+	int index = fp->index;
+
+	DP(NETIF_MSG_INTR, "got an msix interrupt on [%d]\n", index);
+	bnx2x_ack_sb(bp, index, USTORM_ID, 0, IGU_INT_DISABLE, 0);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return IRQ_HANDLED;
+#endif
+
+	prefetch(fp->rx_cons_sb);
+	prefetch(fp->tx_cons_sb);
+	prefetch(&fp->status_blk->c_status_block.status_block_index);
+	prefetch(&fp->status_blk->u_status_block.status_block_index);
+
+	netif_rx_schedule(dev, &bnx2x_fp(bp, index, napi));
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
+{
+	struct net_device *dev = dev_instance;
+	struct bnx2x *bp = netdev_priv(dev);
+	u16 status = bnx2x_ack_int(bp);
+
+	if (unlikely(status == 0)) {
+		DP(NETIF_MSG_INTR, "not our interrupt!\n");
+		return IRQ_NONE;
+	}
+
+	DP(NETIF_MSG_INTR, "got an interrupt status is %u\n", status);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return IRQ_HANDLED;
+#endif
+
+	/* Return here if interrupt is shared and is disabled */
+	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
+		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
+		return IRQ_HANDLED;
+	}
+
+	if (status & 0x2) {
+		struct bnx2x_fastpath *fp = &bp->fp[0];
+
+		prefetch(fp->rx_cons_sb);
+		prefetch(fp->tx_cons_sb);
+		prefetch(&fp->status_blk->c_status_block.status_block_index);
+		prefetch(&fp->status_blk->u_status_block.status_block_index);
+
+		netif_rx_schedule(dev, &bnx2x_fp(bp, 0, napi));
+
+		status &= ~0x2;
+		if (!status)
+			return IRQ_HANDLED;
+	}
+
+	if (unlikely(status & 0x1)) {
+
+		schedule_work(&bp->sp_task);
+
+		status &= ~0x1;
+		if (!status)
+			return IRQ_HANDLED;
+	}
+
+	DP(NETIF_MSG_INTR, "got an unknown interrupt! (status is %u)\n",
+	   status);
+
+	return IRQ_HANDLED;
+}
+
+/* end of fast path */
+
+/* PHY/MAC */
+
+/*
+ * General service functions
+ */
+
+static void bnx2x_leds_set(struct bnx2x *bp, unsigned int speed)
+{
+	int port = bp->port;
+
+	NIG_WR(NIG_REG_LED_MODE_P0 + port*4,
+	       ((bp->hw_config & SHARED_HW_CFG_LED_MODE_MASK) >>
+		SHARED_HW_CFG_LED_MODE_SHIFT));
+	NIG_WR(NIG_REG_LED_CONTROL_OVERRIDE_TRAFFIC_P0 + port*4, 0);
+
+	/* Set blinking rate to ~15.9Hz */
+	NIG_WR(NIG_REG_LED_CONTROL_BLINK_RATE_P0 + port*4,
+	       LED_BLINK_RATE_VAL);
+	NIG_WR(NIG_REG_LED_CONTROL_BLINK_RATE_ENA_P0 + port*4, 1);
+
+	/* On Ax chip versions for speeds less than 10G
+	   LED scheme is different */
+	if ((CHIP_REV(bp) == CHIP_REV_Ax) && (speed < SPEED_10000)) {
+		NIG_WR(NIG_REG_LED_CONTROL_OVERRIDE_TRAFFIC_P0 + port*4, 1);
+		NIG_WR(NIG_REG_LED_CONTROL_TRAFFIC_P0 + port*4, 0);
+		NIG_WR(NIG_REG_LED_CONTROL_BLINK_TRAFFIC_P0 + port*4, 1);
+	}
+}
+
+static void bnx2x_leds_unset(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	NIG_WR(NIG_REG_LED_10G_P0 + port*4, 0);
+	NIG_WR(NIG_REG_LED_MODE_P0 + port*4, SHARED_HW_CFG_LED_MAC1);
+}
+
+static u32 bnx2x_bits_en(struct bnx2x *bp, u32 reg, u32 bits)
+{
+	u32 val = REG_RD(bp, reg);
+
+	val |= bits;
+	REG_WR(bp, reg, val);
+	return val;
+}
+
+static u32 bnx2x_bits_dis(struct bnx2x *bp, u32 reg, u32 bits)
+{
+	u32 val = REG_RD(bp, reg);
+
+	val &= ~bits;
+	REG_WR(bp, reg, val);
+	return val;
+}
+
+static int bnx2x_mdio22_write(struct bnx2x *bp, u32 reg, u32 val)
+{
+	int rc;
+	u32 tmp, i;
+	int port = bp->port;
+	u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
+
+/*      DP(NETIF_MSG_HW, "phy_addr 0x%x  reg 0x%x  val 0x%08x\n",
+	   bp->phy_addr, reg, val); */
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		tmp &= ~EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+		REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		udelay(40);
+	}
+
+	tmp = ((bp->phy_addr << 21) | (reg << 16) |
+	       (val & EMAC_MDIO_COMM_DATA) |
+	       EMAC_MDIO_COMM_COMMAND_WRITE_22 |
+	       EMAC_MDIO_COMM_START_BUSY);
+	EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, tmp);
+
+	for (i = 0; i < 50; i++) {
+		udelay(10);
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+		if (!(tmp & EMAC_MDIO_COMM_START_BUSY)) {
+			udelay(5);
+			break;
+		}
+	}
+
+	if (tmp & EMAC_MDIO_COMM_START_BUSY) {
+		BNX2X_ERR("write phy register failed\n");
+
+		rc = -EBUSY;
+	} else {
+		rc = 0;
+	}
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		tmp |= EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+	}
+
+	return rc;
+}
+
+static int bnx2x_mdio22_read(struct bnx2x *bp, u32 reg, u32 *ret_val)
+{
+	int port = bp->port;
+	u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
+	u32 val, i;
+	int rc;
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		val &= ~EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+		REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		udelay(40);
+	}
+
+	val = ((bp->phy_addr << 21) | (reg << 16) |
+	       EMAC_MDIO_COMM_COMMAND_READ_22 |
+	       EMAC_MDIO_COMM_START_BUSY);
+	EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, val);
+
+	for (i = 0; i < 50; i++) {
+		udelay(10);
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+		if (!(val & EMAC_MDIO_COMM_START_BUSY)) {
+			val &= EMAC_MDIO_COMM_DATA;
+			break;
+		}
+	}
+
+	if (val & EMAC_MDIO_COMM_START_BUSY) {
+		BNX2X_ERR("read phy register failed\n");
+
+		*ret_val = 0x0;
+		rc = -EBUSY;
+	} else {
+		*ret_val = val;
+		rc = 0;
+	}
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		val |= EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+	}
+
+/*      DP(NETIF_MSG_HW, "phy_addr 0x%x  reg 0x%x  ret_val 0x%08x\n",
+	   bp->phy_addr, reg, *ret_val); */
+
+	return rc;
+}
+
+static int bnx2x_mdio45_write(struct bnx2x *bp, u32 reg, u32 addr, u32 val)
+{
+	int rc = 0;
+	u32 tmp, i;
+	int port = bp->port;
+	u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		tmp &= ~EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+		REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		udelay(40);
+	}
+
+	/* set clause 45 mode */
+	tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+	tmp |= EMAC_MDIO_MODE_CLAUSE_45;
+	EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+
+	/* address */
+	tmp = ((bp->phy_addr << 21) | (reg << 16) | addr |
+	       EMAC_MDIO_COMM_COMMAND_ADDRESS |
+	       EMAC_MDIO_COMM_START_BUSY);
+	EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, tmp);
+
+	for (i = 0; i < 50; i++) {
+		udelay(10);
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+		if (!(tmp & EMAC_MDIO_COMM_START_BUSY)) {
+			udelay(5);
+			break;
+		}
+	}
+
+	if (tmp & EMAC_MDIO_COMM_START_BUSY) {
+		BNX2X_ERR("write phy register failed\n");
+
+		rc = -EBUSY;
+	} else {
+		/* data */
+		tmp = ((bp->phy_addr << 21) | (reg << 16) | val |
+		       EMAC_MDIO_COMM_COMMAND_WRITE_45 |
+		       EMAC_MDIO_COMM_START_BUSY);
+		EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, tmp);
+
+		for (i = 0; i < 50; i++) {
+			udelay(10);
+
+			tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+			if (!(tmp & EMAC_MDIO_COMM_START_BUSY)) {
+				udelay(5);
+				break;
+			}
+		}
+
+		if (tmp & EMAC_MDIO_COMM_START_BUSY) {
+			BNX2X_ERR("write phy register failed\n");
+
+			rc = -EBUSY;
+		}
+	}
+
+	/* unset clause 45 mode */
+	tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+	tmp &= ~EMAC_MDIO_MODE_CLAUSE_45;
+	EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		tmp = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		tmp |= EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, tmp);
+	}
+
+	return rc;
+}
+
+static int bnx2x_mdio45_read(struct bnx2x *bp, u32 reg, u32 addr,
+			     u32 *ret_val)
+{
+	int port = bp->port;
+	u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
+	u32 val, i;
+	int rc = 0;
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		val &= ~EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+		REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		udelay(40);
+	}
+
+	/* set clause 45 mode */
+	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+	val |= EMAC_MDIO_MODE_CLAUSE_45;
+	EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+
+	/* address */
+	val = ((bp->phy_addr << 21) | (reg << 16) | addr |
+	       EMAC_MDIO_COMM_COMMAND_ADDRESS |
+	       EMAC_MDIO_COMM_START_BUSY);
+	EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, val);
+
+	for (i = 0; i < 50; i++) {
+		udelay(10);
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+		if (!(val & EMAC_MDIO_COMM_START_BUSY)) {
+			udelay(5);
+			break;
+		}
+	}
+
+	if (val & EMAC_MDIO_COMM_START_BUSY) {
+		BNX2X_ERR("read phy register failed\n");
+
+		*ret_val = 0;
+		rc = -EBUSY;
+	} else {
+		/* data */
+		val = ((bp->phy_addr << 21) | (reg << 16) |
+		       EMAC_MDIO_COMM_COMMAND_READ_45 |
+		       EMAC_MDIO_COMM_START_BUSY);
+		EMAC_WR(EMAC_REG_EMAC_MDIO_COMM, val);
+
+		for (i = 0; i < 50; i++) {
+			udelay(10);
+
+			val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_COMM);
+			if (!(val & EMAC_MDIO_COMM_START_BUSY)) {
+				val &= EMAC_MDIO_COMM_DATA;
+				break;
+			}
+		}
+
+		if (val & EMAC_MDIO_COMM_START_BUSY) {
+			BNX2X_ERR("read phy register failed\n");
+
+			val = 0;
+			rc = -EBUSY;
+		}
+
+		*ret_val = val;
+	}
+
+	/* unset clause 45 mode */
+	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+	val &= ~EMAC_MDIO_MODE_CLAUSE_45;
+	EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+
+	if (bp->phy_flags & PHY_INT_MODE_AUTO_POLLING_FLAG) {
+
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MDIO_MODE);
+		val |= EMAC_MDIO_MODE_AUTO_POLL;
+		EMAC_WR(EMAC_REG_EMAC_MDIO_MODE, val);
+	}
+
+	return rc;
+}
+
+static int bnx2x_mdio45_vwrite(struct bnx2x *bp, u32 reg, u32 addr, u32 val)
+{
+	int i;
+	u32 rd_val;
+
+	might_sleep();
+	for (i = 0; i < 10; i++) {
+		bnx2x_mdio45_write(bp, reg, addr, val);
+		msleep(5);
+		bnx2x_mdio45_read(bp, reg, addr, &rd_val);
+		/* if the read value is not the same as the value we wrote,
+		   we should write it again */
+		if (rd_val == val)
+			return 0;
+	}
+	BNX2X_ERR("MDIO write in CL45 failed\n");
+	return -EBUSY;
+}
+
+/*
+ * link managment
+ */
+
+static void bnx2x_flow_ctrl_resolve(struct bnx2x *bp, u32 gp_status)
+{
+	u32 ld_pause;   /* local driver */
+	u32 lp_pause;   /* link partner */
+	u32 pause_result;
+
+	bp->flow_ctrl = 0;
+
+	/* reolve from gp_status in case of AN complete and not sgmii */
+	if ((bp->req_autoneg & AUTONEG_FLOW_CTRL) &&
+	    (gp_status & MDIO_AN_CL73_OR_37_COMPLETE) &&
+	    (!(bp->phy_flags & PHY_SGMII_FLAG)) &&
+	    (XGXS_EXT_PHY_TYPE(bp) == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT)) {
+
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+		bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_AUTO_NEG_ADV,
+				  &ld_pause);
+		bnx2x_mdio22_read(bp,
+			MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1,
+				  &lp_pause);
+		pause_result = (ld_pause &
+				MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_MASK)>>5;
+		pause_result |= (lp_pause &
+				 MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_MASK)>>7;
+		DP(NETIF_MSG_LINK, "pause_result 0x%x\n", pause_result);
+
+		switch (pause_result) { 		/* ASYM P ASYM P */
+		case 0xb:       			/*   1  0   1  1 */
+			bp->flow_ctrl = FLOW_CTRL_TX;
+			break;
+
+		case 0xe:       			/*   1  1   1  0 */
+			bp->flow_ctrl = FLOW_CTRL_RX;
+			break;
+
+		case 0x5:       			/*   0  1   0  1 */
+		case 0x7:       			/*   0  1   1  1 */
+		case 0xd:       			/*   1  1   0  1 */
+		case 0xf:       			/*   1  1   1  1 */
+			bp->flow_ctrl = FLOW_CTRL_BOTH;
+			break;
+
+		default:
+			break;
+		}
+
+	} else { /* forced mode */
+		switch (bp->req_flow_ctrl) {
+		case FLOW_CTRL_AUTO:
+			if (bp->dev->mtu <= 4500)
+				bp->flow_ctrl = FLOW_CTRL_BOTH;
+			else
+				bp->flow_ctrl = FLOW_CTRL_TX;
+			break;
+
+		case FLOW_CTRL_TX:
+		case FLOW_CTRL_RX:
+		case FLOW_CTRL_BOTH:
+			bp->flow_ctrl = bp->req_flow_ctrl;
+			break;
+
+		case FLOW_CTRL_NONE:
+		default:
+			break;
+		}
+	}
+	DP(NETIF_MSG_LINK, "flow_ctrl 0x%x\n", bp->flow_ctrl);
+}
+
+static void bnx2x_link_settings_status(struct bnx2x *bp, u32 gp_status)
+{
+	bp->link_status = 0;
+
+	if (gp_status & MDIO_GP_STATUS_TOP_AN_STATUS1_LINK_STATUS) {
+		DP(NETIF_MSG_LINK, "link up\n");
+
+		bp->link_up = 1;
+		bp->link_status |= LINK_STATUS_LINK_UP;
+
+		if (gp_status & MDIO_GP_STATUS_TOP_AN_STATUS1_DUPLEX_STATUS)
+			bp->duplex = DUPLEX_FULL;
+		else
+			bp->duplex = DUPLEX_HALF;
+
+		bnx2x_flow_ctrl_resolve(bp, gp_status);
+
+		switch (gp_status & GP_STATUS_SPEED_MASK) {
+		case GP_STATUS_10M:
+			bp->line_speed = SPEED_10;
+			if (bp->duplex == DUPLEX_FULL)
+				bp->link_status |= LINK_10TFD;
+			else
+				bp->link_status |= LINK_10THD;
+			break;
+
+		case GP_STATUS_100M:
+			bp->line_speed = SPEED_100;
+			if (bp->duplex == DUPLEX_FULL)
+				bp->link_status |= LINK_100TXFD;
+			else
+				bp->link_status |= LINK_100TXHD;
+			break;
+
+		case GP_STATUS_1G:
+		case GP_STATUS_1G_KX:
+			bp->line_speed = SPEED_1000;
+			if (bp->duplex == DUPLEX_FULL)
+				bp->link_status |= LINK_1000TFD;
+			else
+				bp->link_status |= LINK_1000THD;
+			break;
+
+		case GP_STATUS_2_5G:
+			bp->line_speed = SPEED_2500;
+			if (bp->duplex == DUPLEX_FULL)
+				bp->link_status |= LINK_2500TFD;
+			else
+				bp->link_status |= LINK_2500THD;
+			break;
+
+		case GP_STATUS_5G:
+		case GP_STATUS_6G:
+			BNX2X_ERR("link speed unsupported  gp_status 0x%x\n",
+				  gp_status);
+			break;
+
+		case GP_STATUS_10G_KX4:
+		case GP_STATUS_10G_HIG:
+		case GP_STATUS_10G_CX4:
+			bp->line_speed = SPEED_10000;
+			bp->link_status |= LINK_10GTFD;
+			break;
+
+		case GP_STATUS_12G_HIG:
+			bp->line_speed = SPEED_12000;
+			bp->link_status |= LINK_12GTFD;
+			break;
+
+		case GP_STATUS_12_5G:
+			bp->line_speed = SPEED_12500;
+			bp->link_status |= LINK_12_5GTFD;
+			break;
+
+		case GP_STATUS_13G:
+			bp->line_speed = SPEED_13000;
+			bp->link_status |= LINK_13GTFD;
+			break;
+
+		case GP_STATUS_15G:
+			bp->line_speed = SPEED_15000;
+			bp->link_status |= LINK_15GTFD;
+			break;
+
+		case GP_STATUS_16G:
+			bp->line_speed = SPEED_16000;
+			bp->link_status |= LINK_16GTFD;
+			break;
+
+		default:
+			BNX2X_ERR("link speed unsupported  gp_status 0x%x\n",
+				  gp_status);
+			break;
+		}
+
+		bp->link_status |= LINK_STATUS_SERDES_LINK;
+
+		if (bp->req_autoneg & AUTONEG_SPEED) {
+			bp->link_status |= LINK_STATUS_AUTO_NEGOTIATE_ENABLED;
+
+			if (gp_status & MDIO_AN_CL73_OR_37_COMPLETE)
+				bp->link_status |=
+					LINK_STATUS_AUTO_NEGOTIATE_COMPLETE;
+
+			if (bp->autoneg & AUTONEG_PARALLEL)
+				bp->link_status |=
+					LINK_STATUS_PARALLEL_DETECTION_USED;
+		}
+
+		if (bp->flow_ctrl & FLOW_CTRL_TX)
+		       bp->link_status |= LINK_STATUS_TX_FLOW_CONTROL_ENABLED;
+
+		if (bp->flow_ctrl & FLOW_CTRL_RX)
+		       bp->link_status |= LINK_STATUS_RX_FLOW_CONTROL_ENABLED;
+
+	} else { /* link_down */
+		DP(NETIF_MSG_LINK, "link down\n");
+
+		bp->link_up = 0;
+
+		bp->line_speed = 0;
+		bp->duplex = DUPLEX_FULL;
+		bp->flow_ctrl = 0;
+	}
+
+	DP(NETIF_MSG_LINK, "gp_status 0x%x  link_up %d\n"
+	   DP_LEVEL "  line_speed %d  duplex %d  flow_ctrl 0x%x"
+		    "  link_status 0x%x\n",
+	   gp_status, bp->link_up, bp->line_speed, bp->duplex, bp->flow_ctrl,
+	   bp->link_status);
+}
+
+static void bnx2x_link_int_ack(struct bnx2x *bp, int is_10g)
+{
+	int port = bp->port;
+
+	/* first reset all status
+	 * we asume only one line will be change at a time */
+	bnx2x_bits_dis(bp, NIG_REG_STATUS_INTERRUPT_PORT0 + port*4,
+		       (NIG_XGXS0_LINK_STATUS |
+			NIG_SERDES0_LINK_STATUS |
+			NIG_STATUS_INTERRUPT_XGXS0_LINK10G));
+	if (bp->link_up) {
+		if (is_10g) {
+			/* Disable the 10G link interrupt
+			 * by writing 1 to the status register
+			 */
+			DP(NETIF_MSG_LINK, "10G XGXS link up\n");
+			bnx2x_bits_en(bp,
+				      NIG_REG_STATUS_INTERRUPT_PORT0 + port*4,
+				      NIG_STATUS_INTERRUPT_XGXS0_LINK10G);
+
+		} else if (bp->phy_flags & PHY_XGXS_FLAG) {
+			/* Disable the link interrupt
+			 * by writing 1 to the relevant lane
+			 * in the status register
+			 */
+			DP(NETIF_MSG_LINK, "1G XGXS link up\n");
+			bnx2x_bits_en(bp,
+				      NIG_REG_STATUS_INTERRUPT_PORT0 + port*4,
+				      ((1 << bp->ser_lane) <<
+				       NIG_XGXS0_LINK_STATUS_SIZE));
+
+		} else { /* SerDes */
+			DP(NETIF_MSG_LINK, "SerDes link up\n");
+			/* Disable the link interrupt
+			 * by writing 1 to the status register
+			 */
+			bnx2x_bits_en(bp,
+				      NIG_REG_STATUS_INTERRUPT_PORT0 + port*4,
+				      NIG_SERDES0_LINK_STATUS);
+		}
+
+	} else { /* link_down */
+	}
+}
+
+static int bnx2x_ext_phy_is_link_up(struct bnx2x *bp)
+{
+	u32 ext_phy_type;
+	u32 ext_phy_addr;
+	u32 local_phy;
+	u32 val = 0;
+	u32 rx_sd, pcs_status;
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		local_phy = bp->phy_addr;
+		ext_phy_addr = ((bp->ext_phy_config &
+				 PORT_HW_CFG_XGXS_EXT_PHY_ADDR_MASK) >>
+				PORT_HW_CFG_XGXS_EXT_PHY_ADDR_SHIFT);
+		bp->phy_addr = (u8)ext_phy_addr;
+
+		ext_phy_type = XGXS_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "XGXS Direct\n");
+			val = 1;
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
+			DP(NETIF_MSG_LINK, "XGXS 8705\n");
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_WIS_DEVAD,
+					  EXT_PHY_OPT_LASI_STATUS, &val);
+			DP(NETIF_MSG_LINK, "8705 LASI status is %d\n", val);
+
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_WIS_DEVAD,
+					  EXT_PHY_OPT_LASI_STATUS, &val);
+			DP(NETIF_MSG_LINK, "8705 LASI status is %d\n", val);
+
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					  EXT_PHY_OPT_PMD_RX_SD, &rx_sd);
+			val = (rx_sd & 0x1);
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
+			DP(NETIF_MSG_LINK, "XGXS 8706\n");
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					  EXT_PHY_OPT_LASI_STATUS, &val);
+			DP(NETIF_MSG_LINK, "8706 LASI status is %d\n", val);
+
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					  EXT_PHY_OPT_LASI_STATUS, &val);
+			DP(NETIF_MSG_LINK, "8706 LASI status is %d\n", val);
+
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					  EXT_PHY_OPT_PMD_RX_SD, &rx_sd);
+			bnx2x_mdio45_read(bp, EXT_PHY_OPT_PCS_DEVAD,
+					 EXT_PHY_OPT_PCS_STATUS, &pcs_status);
+			DP(NETIF_MSG_LINK, "8706 rx_sd 0x%x"
+			   "  pcs_status 0x%x\n", rx_sd, pcs_status);
+			/* link is up if both bit 0 of pmd_rx and
+			 * bit 0 of pcs_status are set
+			 */
+			val = (rx_sd & pcs_status);
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD XGXS ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			val = 0;
+			break;
+		}
+		bp->phy_addr = local_phy;
+
+	} else { /* SerDes */
+		ext_phy_type = SERDES_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "SerDes Direct\n");
+			val = 1;
+			break;
+
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
+			DP(NETIF_MSG_LINK, "SerDes 5482\n");
+			val = 1;
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD SerDes ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			val = 0;
+			break;
+		}
+	}
+
+	return val;
+}
+
+static void bnx2x_bmac_enable(struct bnx2x *bp, int is_lb)
+{
+	int port = bp->port;
+	u32 bmac_addr = port ? NIG_REG_INGRESS_BMAC1_MEM :
+			       NIG_REG_INGRESS_BMAC0_MEM;
+	u32 wb_write[2];
+	u32 val;
+
+	DP(NETIF_MSG_LINK, "enableing BigMAC\n");
+	/* reset and unreset the BigMac */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+	       (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
+	msleep(5);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
+	       (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
+
+	/* enable access for bmac registers */
+	NIG_WR(NIG_REG_BMAC0_REGS_OUT_EN + port*4, 0x1);
+
+	/* XGXS control */
+	wb_write[0] = 0x3c;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_BMAC_XGXS_CONTROL,
+		    wb_write, 2);
+
+	/* tx MAC SA */
+	wb_write[0] = ((bp->dev->dev_addr[2] << 24) |
+		       (bp->dev->dev_addr[3] << 16) |
+		       (bp->dev->dev_addr[4] << 8) |
+			bp->dev->dev_addr[5]);
+	wb_write[1] = ((bp->dev->dev_addr[0] << 8) |
+			bp->dev->dev_addr[1]);
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_TX_SOURCE_ADDR,
+		    wb_write, 2);
+
+	/* tx control */
+	val = 0xc0;
+	if (bp->flow_ctrl & FLOW_CTRL_TX)
+		val |= 0x800000;
+	wb_write[0] = val;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_TX_CONTROL, wb_write, 2);
+
+	/* set tx mtu */
+	wb_write[0] = ETH_MAX_JUMBO_PACKET_SIZE + ETH_OVREHEAD; /* -CRC */
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_TX_MAX_SIZE, wb_write, 2);
+
+	/* mac control */
+	val = 0x3;
+	if (is_lb) {
+		val |= 0x4;
+		DP(NETIF_MSG_LINK, "enable bmac loopback\n");
+	}
+	wb_write[0] = val;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_BMAC_CONTROL,
+		    wb_write, 2);
+
+	/* rx control set to don't strip crc */
+	val = 0x14;
+	if (bp->flow_ctrl & FLOW_CTRL_RX)
+		val |= 0x20;
+	wb_write[0] = val;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_RX_CONTROL, wb_write, 2);
+
+	/* set rx mtu */
+	wb_write[0] = ETH_MAX_JUMBO_PACKET_SIZE + ETH_OVREHEAD;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_RX_MAX_SIZE, wb_write, 2);
+
+	/* set cnt max size */
+	wb_write[0] = ETH_MAX_JUMBO_PACKET_SIZE + ETH_OVREHEAD; /* -VLAN */
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_CNT_MAX_SIZE,
+		    wb_write, 2);
+
+	/* configure safc */
+	wb_write[0] = 0x1000200;
+	wb_write[1] = 0;
+	REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_RX_LLFC_MSG_FLDS,
+		    wb_write, 2);
+
+	/* fix for emulation */
+	if (CHIP_REV(bp) == CHIP_REV_EMUL) {
+		wb_write[0] = 0xf000;
+		wb_write[1] = 0;
+		REG_WR_DMAE(bp,
+			    bmac_addr + BIGMAC_REGISTER_TX_PAUSE_THRESHOLD,
+			    wb_write, 2);
+	}
+
+	/* reset old bmac stats */
+	memset(&bp->old_bmac, 0, sizeof(struct bmac_stats));
+
+	NIG_WR(NIG_REG_XCM0_OUT_EN + port*4, 0x0);
+
+	/* select XGXS */
+	NIG_WR(NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 0x1);
+	NIG_WR(NIG_REG_XGXS_LANE_SEL_P0 + port*4, 0x0);
+
+	/* disable the NIG in/out to the emac */
+	NIG_WR(NIG_REG_EMAC0_IN_EN + port*4, 0x0);
+	NIG_WR(NIG_REG_EMAC0_PAUSE_OUT_EN + port*4, 0x0);
+	NIG_WR(NIG_REG_EGRESS_EMAC0_OUT_EN + port*4, 0x0);
+
+	/* enable the NIG in/out to the bmac */
+	NIG_WR(NIG_REG_EGRESS_EMAC0_PORT + port*4, 0x0);
+
+	NIG_WR(NIG_REG_BMAC0_IN_EN + port*4, 0x1);
+	val = 0;
+	if (bp->flow_ctrl & FLOW_CTRL_TX)
+		val = 1;
+	NIG_WR(NIG_REG_BMAC0_PAUSE_OUT_EN + port*4, val);
+	NIG_WR(NIG_REG_BMAC0_OUT_EN + port*4, 0x1);
+
+	bp->phy_flags |= PHY_BMAC_FLAG;
+
+	bp->stats_state = STATS_STATE_ENABLE;
+}
+
+static void bnx2x_emac_enable(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 emac_base = port ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
+	u32 val;
+	int timeout;
+
+	DP(NETIF_MSG_LINK, "enableing EMAC\n");
+	/* reset and unreset the emac core */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+	       (MISC_REGISTERS_RESET_REG_2_RST_EMAC0_HARD_CORE << port));
+	msleep(5);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
+	       (MISC_REGISTERS_RESET_REG_2_RST_EMAC0_HARD_CORE << port));
+
+	/* enable emac and not bmac */
+	NIG_WR(NIG_REG_EGRESS_EMAC0_PORT + port*4, 1);
+
+	/* for paladium */
+	if (CHIP_REV(bp) == CHIP_REV_EMUL) {
+		/* Use lane 1 (of lanes 0-3) */
+		NIG_WR(NIG_REG_XGXS_LANE_SEL_P0 + port*4, 1);
+		NIG_WR(NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 1);
+	}
+	/* for fpga */
+	else if (CHIP_REV(bp) == CHIP_REV_FPGA) {
+		/* Use lane 1 (of lanes 0-3) */
+		NIG_WR(NIG_REG_XGXS_LANE_SEL_P0 + port*4, 1);
+		NIG_WR(NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 0);
+	}
+	/* ASIC */
+	else {
+		if (bp->phy_flags & PHY_XGXS_FLAG) {
+			DP(NETIF_MSG_LINK, "XGXS\n");
+			/* select the master lanes (out of 0-3) */
+			NIG_WR(NIG_REG_XGXS_LANE_SEL_P0 + port*4,
+			       bp->ser_lane);
+			/* select XGXS */
+			NIG_WR(NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 1);
+
+		} else { /* SerDes */
+			DP(NETIF_MSG_LINK, "SerDes\n");
+			/* select SerDes */
+			NIG_WR(NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 0);
+		}
+	}
+
+	/* enable emac */
+	NIG_WR(NIG_REG_NIG_EMAC0_EN + port*4, 1);
+
+	/* init emac - use read-modify-write */
+	/* self clear reset */
+	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MODE);
+	EMAC_WR(EMAC_REG_EMAC_MODE, (val | EMAC_MODE_RESET));
+
+	timeout = 200;
+	while (val & EMAC_MODE_RESET) {
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MODE);
+		DP(NETIF_MSG_LINK, "EMAC reset reg is %u\n", val);
+		if (!timeout) {
+			BNX2X_ERR("EMAC timeout!\n");
+			break;
+		}
+		timeout--;
+	}
+
+	/* reset tx part */
+	EMAC_WR(EMAC_REG_EMAC_TX_MODE, EMAC_TX_MODE_RESET);
+
+	timeout = 200;
+	while (val & EMAC_TX_MODE_RESET) {
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_TX_MODE);
+		DP(NETIF_MSG_LINK, "EMAC reset reg is %u\n", val);
+		if (!timeout) {
+			BNX2X_ERR("EMAC timeout!\n");
+			break;
+		}
+		timeout--;
+	}
+
+	if (CHIP_REV_IS_SLOW(bp)) {
+		/* config GMII mode */
+		val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MODE);
+		EMAC_WR(EMAC_REG_EMAC_MODE, (val | EMAC_MODE_PORT_GMII));
+
+	} else { /* ASIC */
+		/* pause enable/disable */
+		bnx2x_bits_dis(bp, emac_base + EMAC_REG_EMAC_RX_MODE,
+			       EMAC_RX_MODE_FLOW_EN);
+		if (bp->flow_ctrl & FLOW_CTRL_RX)
+			bnx2x_bits_en(bp, emac_base + EMAC_REG_EMAC_RX_MODE,
+				      EMAC_RX_MODE_FLOW_EN);
+
+		bnx2x_bits_dis(bp, emac_base + EMAC_REG_EMAC_TX_MODE,
+			       EMAC_TX_MODE_EXT_PAUSE_EN);
+		if (bp->flow_ctrl & FLOW_CTRL_TX)
+			bnx2x_bits_en(bp, emac_base + EMAC_REG_EMAC_TX_MODE,
+				      EMAC_TX_MODE_EXT_PAUSE_EN);
+	}
+
+	/* KEEP_VLAN_TAG, promiscous */
+	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_RX_MODE);
+	val |= EMAC_RX_MODE_KEEP_VLAN_TAG | EMAC_RX_MODE_PROMISCUOUS;
+	EMAC_WR(EMAC_REG_EMAC_RX_MODE, val);
+
+	/* identify magic packets */
+	val = REG_RD(bp, emac_base + EMAC_REG_EMAC_MODE);
+	EMAC_WR(EMAC_REG_EMAC_MODE, (val | EMAC_MODE_MPKT));
+
+	/* enable emac for jumbo packets */
+	EMAC_WR(EMAC_REG_EMAC_RX_MTU_SIZE,
+		(EMAC_RX_MTU_SIZE_JUMBO_ENA |
+		 (ETH_MAX_JUMBO_PACKET_SIZE + ETH_OVREHEAD))); /* -VLAN */
+
+	/* strip CRC */
+	NIG_WR(NIG_REG_NIG_INGRESS_EMAC0_NO_CRC + port*4, 0x1);
+
+	val = ((bp->dev->dev_addr[0] << 8) |
+		bp->dev->dev_addr[1]);
+	EMAC_WR(EMAC_REG_EMAC_MAC_MATCH, val);
+
+	val = ((bp->dev->dev_addr[2] << 24) |
+	       (bp->dev->dev_addr[3] << 16) |
+	       (bp->dev->dev_addr[4] << 8) |
+		bp->dev->dev_addr[5]);
+	EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + 4, val);
+
+	/* disable the NIG in/out to the bmac */
+	NIG_WR(NIG_REG_BMAC0_IN_EN + port*4, 0x0);
+	NIG_WR(NIG_REG_BMAC0_PAUSE_OUT_EN + port*4, 0x0);
+	NIG_WR(NIG_REG_BMAC0_OUT_EN + port*4, 0x0);
+
+	/* enable the NIG in/out to the emac */
+	NIG_WR(NIG_REG_EMAC0_IN_EN + port*4, 0x1);
+	val = 0;
+	if (bp->flow_ctrl & FLOW_CTRL_TX)
+		val = 1;
+	NIG_WR(NIG_REG_EMAC0_PAUSE_OUT_EN + port*4, val);
+	NIG_WR(NIG_REG_EGRESS_EMAC0_OUT_EN + port*4, 0x1);
+
+	if (CHIP_REV(bp) == CHIP_REV_FPGA) {
+		/* take the BigMac out of reset */
+		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
+		       (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
+
+		/* enable access for bmac registers */
+		NIG_WR(NIG_REG_BMAC0_REGS_OUT_EN + port*4, 0x1);
+	}
+
+	bp->phy_flags |= PHY_EMAC_FLAG;
+
+	bp->stats_state = STATS_STATE_ENABLE;
+}
+
+static void bnx2x_emac_program(struct bnx2x *bp)
+{
+	u16 mode = 0;
+	int port = bp->port;
+
+	DP(NETIF_MSG_LINK, "setting link speed & duplex\n");
+	bnx2x_bits_dis(bp, GRCBASE_EMAC0 + port*0x400 + EMAC_REG_EMAC_MODE,
+		       (EMAC_MODE_25G_MODE |
+			EMAC_MODE_PORT_MII_10M |
+			EMAC_MODE_HALF_DUPLEX));
+	switch (bp->line_speed) {
+	case SPEED_10:
+		mode |= EMAC_MODE_PORT_MII_10M;
+		break;
+
+	case SPEED_100:
+		mode |= EMAC_MODE_PORT_MII;
+		break;
+
+	case SPEED_1000:
+		mode |= EMAC_MODE_PORT_GMII;
+		break;
+
+	case SPEED_2500:
+		mode |= (EMAC_MODE_25G_MODE | EMAC_MODE_PORT_GMII);
+		break;
+
+	default:
+		/* 10G not valid for EMAC */
+		BNX2X_ERR("Invalid line_speed 0x%x\n", bp->line_speed);
+		break;
+	}
+
+	if (bp->duplex == DUPLEX_HALF)
+		mode |= EMAC_MODE_HALF_DUPLEX;
+	bnx2x_bits_en(bp, GRCBASE_EMAC0 + port*0x400 + EMAC_REG_EMAC_MODE,
+		      mode);
+
+	bnx2x_leds_set(bp, bp->line_speed);
+}
+
+static void bnx2x_set_sgmii_tx_driver(struct bnx2x *bp)
+{
+	u32 lp_up2;
+	u32 tx_driver;
+
+	/* read precomp */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_OVER_1G);
+	bnx2x_mdio22_read(bp, MDIO_OVER_1G_LP_UP2, &lp_up2);
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_TX0);
+	bnx2x_mdio22_read(bp, MDIO_TX0_TX_DRIVER, &tx_driver);
+
+	/* bits [10:7] at lp_up2, positioned at [15:12] */
+	lp_up2 = (((lp_up2 & MDIO_OVER_1G_LP_UP2_PREEMPHASIS_MASK) >>
+		   MDIO_OVER_1G_LP_UP2_PREEMPHASIS_SHIFT) <<
+		  MDIO_TX0_TX_DRIVER_PREEMPHASIS_SHIFT);
+
+	if ((lp_up2 != 0) &&
+	    (lp_up2 != (tx_driver & MDIO_TX0_TX_DRIVER_PREEMPHASIS_MASK))) {
+		/* replace tx_driver bits [15:12] */
+		tx_driver &= ~MDIO_TX0_TX_DRIVER_PREEMPHASIS_MASK;
+		tx_driver |= lp_up2;
+		bnx2x_mdio22_write(bp, MDIO_TX0_TX_DRIVER, tx_driver);
+	}
+}
+
+static void bnx2x_pbf_update(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 init_crd, crd;
+	u32 count = 1000;
+	u32 pause = 0;
+
+
+	/* disable port */
+	REG_WR(bp, PBF_REG_DISABLE_NEW_TASK_PROC_P0 + port*4, 0x1);
+
+	/* wait for init credit */
+	init_crd = REG_RD(bp, PBF_REG_P0_INIT_CRD + port*4);
+	crd = REG_RD(bp, PBF_REG_P0_CREDIT + port*8);
+	DP(NETIF_MSG_LINK, "init_crd 0x%x  crd 0x%x\n", init_crd, crd);
+
+	while ((init_crd != crd) && count) {
+		msleep(5);
+
+		crd = REG_RD(bp, PBF_REG_P0_CREDIT + port*8);
+		count--;
+	}
+	crd = REG_RD(bp, PBF_REG_P0_CREDIT + port*8);
+	if (init_crd != crd)
+		BNX2X_ERR("BUG! init_crd 0x%x != crd 0x%x\n", init_crd, crd);
+
+	if (bp->flow_ctrl & FLOW_CTRL_RX)
+		pause = 1;
+	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + port*4, pause);
+	if (pause) {
+		/* update threshold */
+		REG_WR(bp, PBF_REG_P0_ARB_THRSH + port*4, 0);
+		/* update init credit */
+		init_crd = 778; 	/* (800-18-4) */
+
+	} else {
+		u32 thresh = (ETH_MAX_JUMBO_PACKET_SIZE + ETH_OVREHEAD)/16;
+
+		/* update threshold */
+		REG_WR(bp, PBF_REG_P0_ARB_THRSH + port*4, thresh);
+		/* update init credit */
+		switch (bp->line_speed) {
+		case SPEED_10:
+		case SPEED_100:
+		case SPEED_1000:
+			init_crd = thresh + 55 - 22;
+			break;
+
+		case SPEED_2500:
+			init_crd = thresh + 138 - 22;
+			break;
+
+		case SPEED_10000:
+			init_crd = thresh + 553 - 22;
+			break;
+
+		default:
+			BNX2X_ERR("Invalid line_speed 0x%x\n",
+				  bp->line_speed);
+			break;
+		}
+	}
+	REG_WR(bp, PBF_REG_P0_INIT_CRD + port*4, init_crd);
+	DP(NETIF_MSG_LINK, "PBF updated to speed %d credit %d\n",
+	   bp->line_speed, init_crd);
+
+	/* probe the credit changes */
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 0x1);
+	msleep(5);
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 0x0);
+
+	/* enable port */
+	REG_WR(bp, PBF_REG_DISABLE_NEW_TASK_PROC_P0 + port*4, 0x0);
+}
+
+static void bnx2x_update_mng(struct bnx2x *bp)
+{
+	if (!nomcp)
+		SHMEM_WR(bp, drv_fw_mb[bp->port].link_status,
+			 bp->link_status);
+}
+
+static void bnx2x_link_report(struct bnx2x *bp)
+{
+	if (bp->link_up) {
+		netif_carrier_on(bp->dev);
+		printk(KERN_INFO PFX "%s NIC Link is Up, ", bp->dev->name);
+
+		printk("%d Mbps ", bp->line_speed);
+
+		if (bp->duplex == DUPLEX_FULL)
+			printk("full duplex");
+		else
+			printk("half duplex");
+
+		if (bp->flow_ctrl) {
+			if (bp->flow_ctrl & FLOW_CTRL_RX) {
+				printk(", receive ");
+				if (bp->flow_ctrl & FLOW_CTRL_TX)
+					printk("& transmit ");
+			} else {
+				printk(", transmit ");
+			}
+			printk("flow control ON");
+		}
+		printk("\n");
+
+	} else { /* link_down */
+		netif_carrier_off(bp->dev);
+		printk(KERN_INFO PFX "%s NIC Link is Down\n", bp->dev->name);
+	}
+}
+
+static void bnx2x_link_up(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	/* PBF - link up */
+	bnx2x_pbf_update(bp);
+
+	/* disable drain */
+	NIG_WR(NIG_REG_EGRESS_DRAIN0_MODE + port*4, 0);
+
+	/* update shared memory */
+	bnx2x_update_mng(bp);
+
+	/* indicate link up */
+	bnx2x_link_report(bp);
+}
+
+static void bnx2x_link_down(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	/* notify stats */
+	if (bp->stats_state != STATS_STATE_DISABLE) {
+		bp->stats_state = STATS_STATE_STOP;
+		DP(BNX2X_MSG_STATS, "stats_state - STOP\n");
+	}
+
+	/* indicate link down */
+	bp->phy_flags &= ~(PHY_BMAC_FLAG | PHY_EMAC_FLAG);
+
+	/* reset BigMac */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+	       (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
+
+	/* ignore drain flag interrupt */
+	/* activate nig drain */
+	NIG_WR(NIG_REG_EGRESS_DRAIN0_MODE + port*4, 1);
+
+	/* update shared memory */
+	bnx2x_update_mng(bp);
+
+	/* indicate link down */
+	bnx2x_link_report(bp);
+}
+
+static void bnx2x_init_mac_stats(struct bnx2x *bp);
+
+/* This function is called upon link interrupt */
+static void bnx2x_link_update(struct bnx2x *bp)
+{
+	u32 gp_status;
+	int port = bp->port;
+	int i;
+	int link_10g;
+
+	DP(NETIF_MSG_LINK, "port %x, is xgxs %x, stat_mask 0x%x,"
+	   " int_mask 0x%x, saved_mask 0x%x, MI_INT %x, SERDES_LINK %x,"
+	   " 10G %x, XGXS_LINK %x\n", port, (bp->phy_flags & PHY_XGXS_FLAG),
+	   REG_RD(bp, NIG_REG_STATUS_INTERRUPT_PORT0 + port*4),
+	   REG_RD(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4), bp->nig_mask,
+	   REG_RD(bp, NIG_REG_EMAC0_STATUS_MISC_MI_INT + port*0x18),
+	   REG_RD(bp, NIG_REG_SERDES0_STATUS_LINK_STATUS + port*0x3c),
+	   REG_RD(bp, NIG_REG_XGXS0_STATUS_LINK10G + port*0x68),
+	   REG_RD(bp, NIG_REG_XGXS0_STATUS_LINK_STATUS + port*0x68)
+	);
+
+	might_sleep();
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_GP_STATUS);
+	/* avoid fast toggling */
+	for (i = 0 ; i < 10 ; i++) {
+		msleep(10);
+		bnx2x_mdio22_read(bp, MDIO_GP_STATUS_TOP_AN_STATUS1,
+				  &gp_status);
+	}
+
+	bnx2x_link_settings_status(bp, gp_status);
+
+	/* anything 10 and over uses the bmac */
+	link_10g = ((bp->line_speed >= SPEED_10000) &&
+		    (bp->line_speed <= SPEED_16000));
+
+	bnx2x_link_int_ack(bp, link_10g);
+
+	/* link is up only if both local phy and external phy are up */
+	if (bp->link_up && bnx2x_ext_phy_is_link_up(bp)) {
+		if (link_10g) {
+			bnx2x_bmac_enable(bp, 0);
+			bnx2x_leds_set(bp, SPEED_10000);
+
+		} else {
+			bnx2x_emac_enable(bp);
+			bnx2x_emac_program(bp);
+
+			/* AN complete? */
+			if (gp_status & MDIO_AN_CL73_OR_37_COMPLETE) {
+				if (!(bp->phy_flags & PHY_SGMII_FLAG))
+					bnx2x_set_sgmii_tx_driver(bp);
+			}
+		}
+		bnx2x_link_up(bp);
+
+	} else { /* link down */
+		bnx2x_leds_unset(bp);
+		bnx2x_link_down(bp);
+	}
+
+	bnx2x_init_mac_stats(bp);
+}
+
+/*
+ * Init service functions
+ */
+
+static void bnx2x_set_aer_mmd(struct bnx2x *bp)
+{
+	u16 offset = (bp->phy_flags & PHY_XGXS_FLAG) ?
+					(bp->phy_addr + bp->ser_lane) : 0;
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_AER_BLOCK);
+	bnx2x_mdio22_write(bp, MDIO_AER_BLOCK_AER_REG, 0x3800 + offset);
+}
+
+static void bnx2x_set_master_ln(struct bnx2x *bp)
+{
+	u32 new_master_ln;
+
+	/* set the master_ln for AN */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_XGXS_BLOCK2);
+	bnx2x_mdio22_read(bp, MDIO_XGXS_BLOCK2_TEST_MODE_LANE,
+			  &new_master_ln);
+	bnx2x_mdio22_write(bp, MDIO_XGXS_BLOCK2_TEST_MODE_LANE,
+			   (new_master_ln | bp->ser_lane));
+}
+
+static void bnx2x_reset_unicore(struct bnx2x *bp)
+{
+	u32 mii_control;
+	int i;
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+	bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL, &mii_control);
+	/* reset the unicore */
+	bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+			   (mii_control | MDIO_COMBO_IEEO_MII_CONTROL_RESET));
+
+	/* wait for the reset to self clear */
+	for (i = 0; i < MDIO_ACCESS_TIMEOUT; i++) {
+		udelay(5);
+
+		/* the reset erased the previous bank value */
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+		bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				  &mii_control);
+
+		if (!(mii_control & MDIO_COMBO_IEEO_MII_CONTROL_RESET)) {
+			udelay(5);
+			return;
+		}
+	}
+
+	BNX2X_ERR("BUG! unicore is still in reset!\n");
+}
+
+static void bnx2x_set_swap_lanes(struct bnx2x *bp)
+{
+	/* Each two bits represents a lane number:
+	   No swap is 0123 => 0x1b no need to enable the swap */
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_XGXS_BLOCK2);
+	if (bp->rx_lane_swap != 0x1b) {
+		bnx2x_mdio22_write(bp, MDIO_XGXS_BLOCK2_RX_LN_SWAP,
+				   (bp->rx_lane_swap |
+				    MDIO_XGXS_BLOCK2_RX_LN_SWAP_ENABLE |
+				   MDIO_XGXS_BLOCK2_RX_LN_SWAP_FORCE_ENABLE));
+	} else {
+		bnx2x_mdio22_write(bp, MDIO_XGXS_BLOCK2_RX_LN_SWAP, 0);
+	}
+
+	if (bp->tx_lane_swap != 0x1b) {
+		bnx2x_mdio22_write(bp, MDIO_XGXS_BLOCK2_TX_LN_SWAP,
+				   (bp->tx_lane_swap |
+				    MDIO_XGXS_BLOCK2_TX_LN_SWAP_ENABLE));
+	} else {
+		bnx2x_mdio22_write(bp, MDIO_XGXS_BLOCK2_TX_LN_SWAP, 0);
+	}
+}
+
+static void bnx2x_set_parallel_detection(struct bnx2x *bp)
+{
+	u32 control2;
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_SERDES_DIGITAL);
+	bnx2x_mdio22_read(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL2,
+			  &control2);
+
+	if (bp->autoneg & AUTONEG_PARALLEL) {
+		control2 |= MDIO_SERDES_DIGITAL_A_1000X_CONTROL2_PRL_DT_EN;
+	} else {
+		control2 &= ~MDIO_SERDES_DIGITAL_A_1000X_CONTROL2_PRL_DT_EN;
+	}
+	bnx2x_mdio22_write(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL2,
+			   control2);
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		DP(NETIF_MSG_LINK, "XGXS\n");
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_10G_PARALLEL_DETECT);
+
+		bnx2x_mdio22_write(bp,
+				   MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_LINK,
+			       MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_LINK_CNT);
+
+		bnx2x_mdio22_read(bp,
+				 MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL,
+				  &control2);
+
+		if (bp->autoneg & AUTONEG_PARALLEL) {
+			control2 |=
+		    MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL_PARDET10G_EN;
+		} else {
+			control2 &=
+		   ~MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL_PARDET10G_EN;
+		}
+		bnx2x_mdio22_write(bp,
+				 MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL,
+				   control2);
+	}
+}
+
+static void bnx2x_set_autoneg(struct bnx2x *bp)
+{
+	u32 reg_val;
+
+	/* CL37 Autoneg */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+	bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL, &reg_val);
+	if ((bp->req_autoneg & AUTONEG_SPEED) &&
+	    (bp->autoneg & AUTONEG_CL37)) {
+		/* CL37 Autoneg Enabled */
+		reg_val |= MDIO_COMBO_IEEO_MII_CONTROL_AN_EN;
+	} else {
+		/* CL37 Autoneg Disabled */
+		reg_val &= ~(MDIO_COMBO_IEEO_MII_CONTROL_AN_EN |
+			     MDIO_COMBO_IEEO_MII_CONTROL_RESTART_AN);
+	}
+	bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL, reg_val);
+
+	/* Enable/Disable Autodetection */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_SERDES_DIGITAL);
+	bnx2x_mdio22_read(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL1, &reg_val);
+	reg_val &= ~MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_SIGNAL_DETECT_EN;
+
+	if ((bp->req_autoneg & AUTONEG_SPEED) &&
+	    (bp->autoneg & AUTONEG_SGMII_FIBER_AUTODET)) {
+		reg_val |= MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_AUTODET;
+	} else {
+		reg_val &= ~MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_AUTODET;
+	}
+	bnx2x_mdio22_write(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL1, reg_val);
+
+	/* Enable TetonII and BAM autoneg */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_BAM_NEXT_PAGE);
+	bnx2x_mdio22_read(bp, MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL,
+			  &reg_val);
+	if ((bp->req_autoneg & AUTONEG_SPEED) &&
+	    (bp->autoneg & AUTONEG_CL37) && (bp->autoneg & AUTONEG_BAM)) {
+		/* Enable BAM aneg Mode and TetonII aneg Mode */
+		reg_val |= (MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_BAM_MODE |
+			    MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_TETON_AN);
+	} else {
+		/* TetonII and BAM Autoneg Disabled */
+		reg_val &= ~(MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_BAM_MODE |
+			     MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_TETON_AN);
+	}
+	bnx2x_mdio22_write(bp, MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL,
+			   reg_val);
+
+	/* Enable Clause 73 Aneg */
+	if ((bp->req_autoneg & AUTONEG_SPEED) &&
+	    (bp->autoneg & AUTONEG_CL73)) {
+		/* Enable BAM Station Manager */
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_CL73_USERB0);
+		bnx2x_mdio22_write(bp, MDIO_CL73_USERB0_CL73_BAM_CTRL1,
+				   (MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_EN |
+			MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_STATION_MNGR_EN |
+			MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_NP_AFTER_BP_EN));
+
+		/* Merge CL73 and CL37 aneg resolution */
+		bnx2x_mdio22_read(bp, MDIO_CL73_USERB0_CL73_BAM_CTRL3,
+				  &reg_val);
+		bnx2x_mdio22_write(bp, MDIO_CL73_USERB0_CL73_BAM_CTRL3,
+				   (reg_val |
+			MDIO_CL73_USERB0_CL73_BAM_CTRL3_USE_CL73_HCD_MR));
+
+		/* Set the CL73 AN speed */
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_CL73_IEEEB1);
+		bnx2x_mdio22_read(bp, MDIO_CL73_IEEEB1_AN_ADV2, &reg_val);
+		/* In the SerDes we support only the 1G.
+		   In the XGXS we support the 10G KX4
+		   but we currently do not support the KR */
+		if (bp->phy_flags & PHY_XGXS_FLAG) {
+			DP(NETIF_MSG_LINK, "XGXS\n");
+			/* 10G KX4 */
+			reg_val |= MDIO_CL73_IEEEB1_AN_ADV2_ADVR_10G_KX4;
+		} else {
+			DP(NETIF_MSG_LINK, "SerDes\n");
+			/* 1000M KX */
+			reg_val |= MDIO_CL73_IEEEB1_AN_ADV2_ADVR_1000M_KX;
+		}
+		bnx2x_mdio22_write(bp, MDIO_CL73_IEEEB1_AN_ADV2, reg_val);
+
+		/* CL73 Autoneg Enabled */
+		reg_val = MDIO_CL73_IEEEB0_CL73_AN_CONTROL_AN_EN;
+	} else {
+		/* CL73 Autoneg Disabled */
+		reg_val = 0;
+	}
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_CL73_IEEEB0);
+	bnx2x_mdio22_write(bp, MDIO_CL73_IEEEB0_CL73_AN_CONTROL, reg_val);
+}
+
+/* program SerDes, forced speed */
+static void bnx2x_program_serdes(struct bnx2x *bp)
+{
+	u32 reg_val;
+
+	/* program duplex, disable autoneg */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+	bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL, &reg_val);
+	reg_val &= ~(MDIO_COMBO_IEEO_MII_CONTROL_FULL_DUPLEX |
+		     MDIO_COMBO_IEEO_MII_CONTROL_AN_EN);
+	if (bp->req_duplex == DUPLEX_FULL)
+		reg_val |= MDIO_COMBO_IEEO_MII_CONTROL_FULL_DUPLEX;
+	bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL, reg_val);
+
+	/* program speed
+	   - needed only if the speed is greater than 1G (2.5G or 10G) */
+	if (bp->req_line_speed > SPEED_1000) {
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_SERDES_DIGITAL);
+		bnx2x_mdio22_read(bp, MDIO_SERDES_DIGITAL_MISC1, &reg_val);
+		/* clearing the speed value before setting the right speed */
+		reg_val &= ~MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_MASK;
+		reg_val |= (MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_156_25M |
+			    MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_SEL);
+		if (bp->req_line_speed == SPEED_10000)
+			reg_val |=
+				MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_10G_CX4;
+		bnx2x_mdio22_write(bp, MDIO_SERDES_DIGITAL_MISC1, reg_val);
+	}
+}
+
+static void bnx2x_set_brcm_cl37_advertisment(struct bnx2x *bp)
+{
+	u32 val = 0;
+
+	/* configure the 48 bits for BAM AN */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_OVER_1G);
+
+	/* set extended capabilities */
+	if (bp->advertising & ADVERTISED_2500baseT_Full)
+		val |= MDIO_OVER_1G_UP1_2_5G;
+	if (bp->advertising & ADVERTISED_10000baseT_Full)
+		val |= MDIO_OVER_1G_UP1_10G;
+	bnx2x_mdio22_write(bp, MDIO_OVER_1G_UP1, val);
+
+	bnx2x_mdio22_write(bp, MDIO_OVER_1G_UP3, 0);
+}
+
+static void bnx2x_set_ieee_aneg_advertisment(struct bnx2x *bp)
+{
+	u32 an_adv;
+
+	/* for AN, we are always publishing full duplex */
+	an_adv = MDIO_COMBO_IEEE0_AUTO_NEG_ADV_FULL_DUPLEX;
+
+	/* set pause */
+	switch (bp->pause_mode) {
+	case PAUSE_SYMMETRIC:
+		an_adv |= MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_SYMMETRIC;
+		break;
+	case PAUSE_ASYMMETRIC:
+		an_adv |= MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_ASYMMETRIC;
+		break;
+	case PAUSE_BOTH:
+		an_adv |= MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_BOTH;
+		break;
+	case PAUSE_NONE:
+		an_adv |= MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_NONE;
+		break;
+	}
+
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+	bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_AUTO_NEG_ADV, an_adv);
+}
+
+static void bnx2x_restart_autoneg(struct bnx2x *bp)
+{
+	if (bp->autoneg & AUTONEG_CL73) {
+		/* enable and restart clause 73 aneg */
+		u32 an_ctrl;
+
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_CL73_IEEEB0);
+		bnx2x_mdio22_read(bp, MDIO_CL73_IEEEB0_CL73_AN_CONTROL,
+				  &an_ctrl);
+		bnx2x_mdio22_write(bp, MDIO_CL73_IEEEB0_CL73_AN_CONTROL,
+				   (an_ctrl |
+				    MDIO_CL73_IEEEB0_CL73_AN_CONTROL_AN_EN |
+				MDIO_CL73_IEEEB0_CL73_AN_CONTROL_RESTART_AN));
+
+	} else {
+		/* Enable and restart BAM/CL37 aneg */
+		u32 mii_control;
+
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+		bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				  &mii_control);
+		bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				   (mii_control |
+				    MDIO_COMBO_IEEO_MII_CONTROL_AN_EN |
+				    MDIO_COMBO_IEEO_MII_CONTROL_RESTART_AN));
+	}
+}
+
+static void bnx2x_initialize_sgmii_process(struct bnx2x *bp)
+{
+	u32 control1;
+
+	/* in SGMII mode, the unicore is always slave */
+	MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_SERDES_DIGITAL);
+	bnx2x_mdio22_read(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL1,
+			  &control1);
+	control1 |= MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_INVERT_SIGNAL_DETECT;
+	/* set sgmii mode (and not fiber) */
+	control1 &= ~(MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_FIBER_MODE |
+		      MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_AUTODET |
+		      MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_MSTR_MODE);
+	bnx2x_mdio22_write(bp, MDIO_SERDES_DIGITAL_A_1000X_CONTROL1,
+			   control1);
+
+	/* if forced speed */
+	if (!(bp->req_autoneg & AUTONEG_SPEED)) {
+		/* set speed, disable autoneg */
+		u32 mii_control;
+
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+		bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				  &mii_control);
+		mii_control &= ~(MDIO_COMBO_IEEO_MII_CONTROL_AN_EN |
+			       MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_MASK |
+				 MDIO_COMBO_IEEO_MII_CONTROL_FULL_DUPLEX);
+
+		switch (bp->req_line_speed) {
+		case SPEED_100:
+			mii_control |=
+				MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_100;
+			break;
+		case SPEED_1000:
+			mii_control |=
+				MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_1000;
+			break;
+		case SPEED_10:
+			/* there is nothing to set for 10M */
+			break;
+		default:
+			/* invalid speed for SGMII */
+			DP(NETIF_MSG_LINK, "Invalid req_line_speed 0x%x\n",
+			   bp->req_line_speed);
+			break;
+		}
+
+		/* setting the full duplex */
+		if (bp->req_duplex == DUPLEX_FULL)
+			mii_control |=
+				MDIO_COMBO_IEEO_MII_CONTROL_FULL_DUPLEX;
+		bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				   mii_control);
+
+	} else { /* AN mode */
+		/* enable and restart AN */
+		bnx2x_restart_autoneg(bp);
+	}
+}
+
+static void bnx2x_link_int_enable(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	/* setting the status to report on link up
+	   for either XGXS or SerDes */
+	bnx2x_bits_dis(bp, NIG_REG_STATUS_INTERRUPT_PORT0 + port*4,
+		       (NIG_XGXS0_LINK_STATUS |
+			NIG_STATUS_INTERRUPT_XGXS0_LINK10G |
+			NIG_SERDES0_LINK_STATUS));
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		/* TBD -
+		 * in force mode (not AN) we can enable just the relevant
+		 * interrupt
+		 * Even in AN we might enable only one according to the AN
+		 * speed mask
+		 */
+		bnx2x_bits_en(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+			      (NIG_MASK_XGXS0_LINK_STATUS |
+			       NIG_MASK_XGXS0_LINK10G));
+		DP(NETIF_MSG_LINK, "enable XGXS interrupt\n");
+
+	} else { /* SerDes */
+		bnx2x_bits_en(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+			      NIG_MASK_SERDES0_LINK_STATUS);
+		DP(NETIF_MSG_LINK, "enable SerDes interrupt\n");
+	}
+}
+
+static void bnx2x_ext_phy_init(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 ext_phy_type;
+	u32 ext_phy_addr;
+	u32 local_phy;
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		local_phy = bp->phy_addr;
+		ext_phy_addr = ((bp->ext_phy_config &
+				 PORT_HW_CFG_XGXS_EXT_PHY_ADDR_MASK) >>
+				PORT_HW_CFG_XGXS_EXT_PHY_ADDR_SHIFT);
+
+		ext_phy_type = XGXS_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "XGXS Direct\n");
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
+			DP(NETIF_MSG_LINK, "XGXS 8705\n");
+			bnx2x_bits_en(bp,
+				      NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+				      NIG_MASK_MI_INT);
+			DP(NETIF_MSG_LINK, "enabled extenal phy int\n");
+
+			bp->phy_addr = ext_phy_type;
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					    EXT_PHY_OPT_PMD_MISC_CNTL,
+					    0x8288);
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					    EXT_PHY_OPT_PHY_IDENTIFIER,
+					    0x7fbf);
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					    EXT_PHY_OPT_CMU_PLL_BYPASS,
+					    0x0100);
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_WIS_DEVAD,
+					    EXT_PHY_OPT_LASI_CNTL, 0x1);
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
+			DP(NETIF_MSG_LINK, "XGXS 8706\n");
+			bnx2x_bits_en(bp,
+				      NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+				      NIG_MASK_MI_INT);
+			DP(NETIF_MSG_LINK, "enabled extenal phy int\n");
+
+			bp->phy_addr = ext_phy_type;
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					    EXT_PHY_OPT_PMD_DIGITAL_CNT,
+					    0x400);
+			bnx2x_mdio45_vwrite(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					    EXT_PHY_OPT_LASI_CNTL, 0x1);
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD XGXS ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			break;
+		}
+		bp->phy_addr = local_phy;
+
+	} else { /* SerDes */
+/*      	ext_phy_addr = ((bp->ext_phy_config &
+				 PORT_HW_CFG_SERDES_EXT_PHY_ADDR_MASK) >>
+				PORT_HW_CFG_SERDES_EXT_PHY_ADDR_SHIFT);
+*/
+		ext_phy_type = SERDES_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "SerDes Direct\n");
+			break;
+
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
+			DP(NETIF_MSG_LINK, "SerDes 5482\n");
+			bnx2x_bits_en(bp,
+				      NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+				      NIG_MASK_MI_INT);
+			DP(NETIF_MSG_LINK, "enabled extenal phy int\n");
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD SerDes ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			break;
+		}
+	}
+}
+
+static void bnx2x_ext_phy_reset(struct bnx2x *bp)
+{
+	u32 ext_phy_type;
+	u32 ext_phy_addr;
+	u32 local_phy;
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		ext_phy_type = XGXS_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "XGXS Direct\n");
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
+			DP(NETIF_MSG_LINK, "XGXS 8705/6\n");
+			local_phy = bp->phy_addr;
+			ext_phy_addr = ((bp->ext_phy_config &
+					PORT_HW_CFG_XGXS_EXT_PHY_ADDR_MASK) >>
+					PORT_HW_CFG_XGXS_EXT_PHY_ADDR_SHIFT);
+			bp->phy_addr = (u8)ext_phy_addr;
+			bnx2x_mdio45_write(bp, EXT_PHY_OPT_PMA_PMD_DEVAD,
+					   EXT_PHY_OPT_CNTL, 0xa040);
+			bp->phy_addr = local_phy;
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD XGXS ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			break;
+		}
+
+	} else { /* SerDes */
+		ext_phy_type = SERDES_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_DIRECT:
+			DP(NETIF_MSG_LINK, "SerDes Direct\n");
+			break;
+
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
+			DP(NETIF_MSG_LINK, "SerDes 5482\n");
+			break;
+
+		default:
+			DP(NETIF_MSG_LINK, "BAD SerDes ext_phy_config 0x%x\n",
+			   bp->ext_phy_config);
+			break;
+		}
+	}
+}
+
+static void bnx2x_link_initialize(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	/* disable attentions */
+	bnx2x_bits_dis(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+		       (NIG_MASK_XGXS0_LINK_STATUS |
+			NIG_MASK_XGXS0_LINK10G |
+			NIG_MASK_SERDES0_LINK_STATUS |
+			NIG_MASK_MI_INT));
+
+	bnx2x_ext_phy_reset(bp);
+
+	bnx2x_set_aer_mmd(bp);
+
+	if (bp->phy_flags & PHY_XGXS_FLAG)
+		bnx2x_set_master_ln(bp);
+
+	/* reset the SerDes and wait for reset bit return low */
+	bnx2x_reset_unicore(bp);
+
+	bnx2x_set_aer_mmd(bp);
+
+	/* setting the masterLn_def again after the reset */
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		bnx2x_set_master_ln(bp);
+		bnx2x_set_swap_lanes(bp);
+	}
+
+	/* Set Parallel Detect */
+	if (bp->req_autoneg & AUTONEG_SPEED)
+		bnx2x_set_parallel_detection(bp);
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		if (bp->req_line_speed &&
+		    bp->req_line_speed < SPEED_1000) {
+			bp->phy_flags |= PHY_SGMII_FLAG;
+		} else {
+			bp->phy_flags &= ~PHY_SGMII_FLAG;
+		}
+	}
+
+	if (!(bp->phy_flags & PHY_SGMII_FLAG)) {
+		u16 bank, rx_eq;
+
+		rx_eq = ((bp->serdes_config &
+			  PORT_HW_CFG_SERDES_RX_DRV_EQUALIZER_MASK) >>
+			 PORT_HW_CFG_SERDES_RX_DRV_EQUALIZER_SHIFT);
+
+		DP(NETIF_MSG_LINK, "setting rx eq to %d\n", rx_eq);
+		for (bank = MDIO_REG_BANK_RX0; bank <= MDIO_REG_BANK_RX_ALL;
+			    bank += (MDIO_REG_BANK_RX1 - MDIO_REG_BANK_RX0)) {
+			MDIO_SET_REG_BANK(bp, bank);
+			bnx2x_mdio22_write(bp, MDIO_RX0_RX_EQ_BOOST,
+					   ((rx_eq &
+				MDIO_RX0_RX_EQ_BOOST_EQUALIZER_CTRL_MASK) |
+				MDIO_RX0_RX_EQ_BOOST_OFFSET_CTRL));
+		}
+
+		/* forced speed requested? */
+		if (!(bp->req_autoneg & AUTONEG_SPEED)) {
+			DP(NETIF_MSG_LINK, "not SGMII, no AN\n");
+
+			/* disable autoneg */
+			bnx2x_set_autoneg(bp);
+
+			/* program speed and duplex */
+			bnx2x_program_serdes(bp);
+
+		} else { /* AN_mode */
+			DP(NETIF_MSG_LINK, "not SGMII, AN\n");
+
+			/* AN enabled */
+			bnx2x_set_brcm_cl37_advertisment(bp);
+
+			/* program duplex & pause advertisment (for aneg) */
+			bnx2x_set_ieee_aneg_advertisment(bp);
+
+			/* enable autoneg */
+			bnx2x_set_autoneg(bp);
+
+			/* enalbe and restart AN */
+			bnx2x_restart_autoneg(bp);
+		}
+
+	} else { /* SGMII mode */
+		DP(NETIF_MSG_LINK, "SGMII\n");
+
+		bnx2x_initialize_sgmii_process(bp);
+	}
+
+	/* enable the interrupt */
+	bnx2x_link_int_enable(bp);
+
+	/* init ext phy and enable link state int */
+	bnx2x_ext_phy_init(bp);
+}
+
+static void bnx2x_phy_deassert(struct bnx2x *bp)
+{
+	int port = bp->port;
+	u32 val;
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		DP(NETIF_MSG_LINK, "XGXS\n");
+		val = XGXS_RESET_BITS;
+
+	} else { /* SerDes */
+		DP(NETIF_MSG_LINK, "SerDes\n");
+		val = SERDES_RESET_BITS;
+	}
+
+	val = val << (port*16);
+
+	/* reset and unreset the SerDes/XGXS */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_3_CLEAR, val);
+	msleep(5);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_3_SET, val);
+}
+
+static int bnx2x_phy_init(struct bnx2x *bp)
+{
+	DP(NETIF_MSG_LINK, "started\n");
+	if (CHIP_REV(bp) == CHIP_REV_FPGA) {
+		bp->phy_flags |= PHY_EMAC_FLAG;
+		bp->link_up = 1;
+		bp->line_speed = SPEED_10000;
+		bp->duplex = DUPLEX_FULL;
+		NIG_WR(NIG_REG_EGRESS_DRAIN0_MODE + bp->port*4, 0);
+		bnx2x_emac_enable(bp);
+		bnx2x_link_report(bp);
+		return 0;
+
+	} else if (CHIP_REV(bp) == CHIP_REV_EMUL) {
+		bp->phy_flags |= PHY_BMAC_FLAG;
+		bp->link_up = 1;
+		bp->line_speed = SPEED_10000;
+		bp->duplex = DUPLEX_FULL;
+		NIG_WR(NIG_REG_EGRESS_DRAIN0_MODE + bp->port*4, 0);
+		bnx2x_bmac_enable(bp, 0);
+		bnx2x_link_report(bp);
+		return 0;
+
+	} else {
+		bnx2x_phy_deassert(bp);
+		bnx2x_link_initialize(bp);
+	}
+
+	return 0;
+}
+
+static void bnx2x_link_reset(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	/* disable attentions */
+	bnx2x_bits_dis(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+		       (NIG_MASK_XGXS0_LINK_STATUS |
+			NIG_MASK_XGXS0_LINK10G |
+			NIG_MASK_SERDES0_LINK_STATUS |
+			NIG_MASK_MI_INT));
+
+	bnx2x_ext_phy_reset(bp);
+
+	/* reset the SerDes/XGXS */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_3_CLEAR,
+	       (0x1ff << (port*16)));
+
+	/* reset EMAC / BMAC and disable NIG interfaces */
+	NIG_WR(NIG_REG_BMAC0_IN_EN + port*4, 0);
+	NIG_WR(NIG_REG_BMAC0_OUT_EN + port*4, 0);
+
+	NIG_WR(NIG_REG_NIG_EMAC0_EN + port*4, 0);
+	NIG_WR(NIG_REG_EMAC0_IN_EN + port*4, 0);
+	NIG_WR(NIG_REG_EGRESS_EMAC0_OUT_EN + port*4, 0);
+
+	NIG_WR(NIG_REG_EGRESS_DRAIN0_MODE + port*4, 1);
+}
+
+#ifdef BNX2X_XGXS_LB
+static void bnx2x_set_xgxs_loopback(struct bnx2x *bp, int is_10g)
+{
+	int port = bp->port;
+
+	if (is_10g) {
+		u32 md_devad;
+
+		DP(NETIF_MSG_LINK, "XGXS 10G loopback enable\n");
+
+		/* change the uni_phy_addr in the nig */
+		REG_RD(bp, (NIG_REG_XGXS0_CTRL_MD_DEVAD + port*0x18),
+		       &md_devad);
+		NIG_WR(NIG_REG_XGXS0_CTRL_MD_DEVAD + port*0x18, 0x5);
+
+		/* change the aer mmd */
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_AER_BLOCK);
+		bnx2x_mdio22_write(bp, MDIO_AER_BLOCK_AER_REG, 0x2800);
+
+		/* config combo IEEE0 control reg for loopback */
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_CL73_IEEEB0);
+		bnx2x_mdio22_write(bp, MDIO_CL73_IEEEB0_CL73_AN_CONTROL,
+				   0x6041);
+
+		/* set aer mmd back */
+		bnx2x_set_aer_mmd(bp);
+
+		/* and md_devad */
+		NIG_WR(NIG_REG_XGXS0_CTRL_MD_DEVAD + port*0x18, md_devad);
+
+	} else {
+		u32 mii_control;
+
+		DP(NETIF_MSG_LINK, "XGXS 1G loopback enable\n");
+
+		MDIO_SET_REG_BANK(bp, MDIO_REG_BANK_COMBO_IEEE0);
+		bnx2x_mdio22_read(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				  &mii_control);
+		bnx2x_mdio22_write(bp, MDIO_COMBO_IEEE0_MII_CONTROL,
+				   (mii_control |
+				    MDIO_COMBO_IEEO_MII_CONTROL_LOOPBACK));
+	}
+}
+#endif
+
+/* end of PHY/MAC */
+
+/* slow path */
+
+/*
+ * General service functions
+ */
+
+/* the slow path queue is odd since completions arrive on the fastpath ring */
+static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
+			 u32 data_hi, u32 data_lo, int common)
+{
+	int port = bp->port;
+
+	DP(NETIF_MSG_TIMER,
+	   "spe (%x:%x)  command %x  hw_cid %x  data (%x:%x)  left %x\n",
+	   (u32)U64_HI(bp->spq_mapping), (u32)(U64_LO(bp->spq_mapping) +
+	   (void *)bp->spq_prod_bd - (void *)bp->spq), command,
+	   HW_CID(bp, cid), data_hi, data_lo, bp->spq_left);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return -EIO;
+#endif
+
+	spin_lock(&bp->spq_lock);
+
+	if (!bp->spq_left) {
+		BNX2X_ERR("BUG! SPQ ring full!\n");
+		spin_unlock(&bp->spq_lock);
+		bnx2x_panic();
+		return -EBUSY;
+	}
+	/* CID needs port number to be encoded int it */
+	bp->spq_prod_bd->hdr.conn_and_cmd_data =
+			cpu_to_le32(((command << SPE_HDR_CMD_ID_SHIFT) |
+				     HW_CID(bp, cid)));
+	bp->spq_prod_bd->hdr.type = cpu_to_le16(ETH_CONNECTION_TYPE);
+	if (common)
+		bp->spq_prod_bd->hdr.type |=
+			cpu_to_le16((1 << SPE_HDR_COMMON_RAMROD_SHIFT));
+
+	bp->spq_prod_bd->data.mac_config_addr.hi = cpu_to_le32(data_hi);
+	bp->spq_prod_bd->data.mac_config_addr.lo = cpu_to_le32(data_lo);
+
+	bp->spq_left--;
+
+	if (bp->spq_prod_bd == bp->spq_last_bd) {
+		bp->spq_prod_bd = bp->spq;
+		bp->spq_prod_idx = 0;
+		DP(NETIF_MSG_TIMER, "end of spq\n");
+
+	} else {
+		bp->spq_prod_bd++;
+		bp->spq_prod_idx++;
+	}
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(port),
+	       bp->spq_prod_idx);
+
+	spin_unlock(&bp->spq_lock);
+	return 0;
+}
+
+/* acquire split MCP access lock register */
+static int bnx2x_lock_alr(struct bnx2x *bp)
+{
+	int rc = 0;
+	u32 i, j, val;
+
+	might_sleep();
+	i = 100;
+	for (j = 0; j < i*10; j++) {
+		val = (1UL << 31);
+		REG_WR(bp, GRCBASE_MCP + 0x9c, val);
+		val = REG_RD(bp, GRCBASE_MCP + 0x9c);
+		if (val & (1L << 31))
+			break;
+
+		msleep(5);
+	}
+
+	if (!(val & (1L << 31))) {
+		BNX2X_ERR("Cannot acquire nvram interface\n");
+
+		rc = -EBUSY;
+	}
+
+	return rc;
+}
+
+/* Release split MCP access lock register */
+static void bnx2x_unlock_alr(struct bnx2x *bp)
+{
+	u32 val = 0;
+
+	REG_WR(bp, GRCBASE_MCP + 0x9c, val);
+}
+
+static inline u16 bnx2x_update_dsb_idx(struct bnx2x *bp)
+{
+	struct host_def_status_block *def_sb = bp->def_status_blk;
+	u16 rc = 0;
+
+	barrier(); /* status block is written to by the chip */
+
+	if (bp->def_att_idx != def_sb->atten_status_block.attn_bits_index) {
+		bp->def_att_idx = def_sb->atten_status_block.attn_bits_index;
+		rc |= 1;
+	}
+	if (bp->def_c_idx != def_sb->c_def_status_block.status_block_index) {
+		bp->def_c_idx = def_sb->c_def_status_block.status_block_index;
+		rc |= 2;
+	}
+	if (bp->def_u_idx != def_sb->u_def_status_block.status_block_index) {
+		bp->def_u_idx = def_sb->u_def_status_block.status_block_index;
+		rc |= 4;
+	}
+	if (bp->def_x_idx != def_sb->x_def_status_block.status_block_index) {
+		bp->def_x_idx = def_sb->x_def_status_block.status_block_index;
+		rc |= 8;
+	}
+	if (bp->def_t_idx != def_sb->t_def_status_block.status_block_index) {
+		bp->def_t_idx = def_sb->t_def_status_block.status_block_index;
+		rc |= 16;
+	}
+	return rc;
+}
+
+/*
+ * slow path service functions
+ */
+
+static void bnx2x_attn_int_asserted(struct bnx2x *bp, u32 asserted)
+{
+	int port = bp->port;
+	u32 igu_addr = (IGU_ADDR_ATTN_BITS_SET + IGU_PORT_BASE * port) * 8;
+	u32 aeu_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
+			      MISC_REG_AEU_MASK_ATTN_FUNC_0;
+	u32 nig_mask_addr = port ? NIG_REG_MASK_INTERRUPT_PORT1 :
+				   NIG_REG_MASK_INTERRUPT_PORT0;
+
+	if (~bp->aeu_mask & (asserted & 0xff))
+		BNX2X_ERR("IGU ERROR\n");
+	if (bp->attn_state & asserted)
+		BNX2X_ERR("IGU ERROR\n");
+
+	DP(NETIF_MSG_HW, "aeu_mask %x  newly asserted %x\n",
+	   bp->aeu_mask, asserted);
+	bp->aeu_mask &= ~(asserted & 0xff);
+	DP(NETIF_MSG_HW, "after masking: aeu_mask %x\n", bp->aeu_mask);
+
+	REG_WR(bp, aeu_addr, bp->aeu_mask);
+
+	bp->attn_state |= asserted;
+
+	if (asserted & ATTN_HARD_WIRED_MASK) {
+		if (asserted & ATTN_NIG_FOR_FUNC) {
+			u32 nig_status_port;
+			u32 nig_int_addr = port ?
+					NIG_REG_STATUS_INTERRUPT_PORT1 :
+					NIG_REG_STATUS_INTERRUPT_PORT0;
+
+			bp->nig_mask = REG_RD(bp, nig_mask_addr);
+			REG_WR(bp, nig_mask_addr, 0);
+
+			nig_status_port = REG_RD(bp, nig_int_addr);
+			bnx2x_link_update(bp);
+
+			/* handle unicore attn? */
+		}
+		if (asserted & ATTN_SW_TIMER_4_FUNC)
+			DP(NETIF_MSG_HW, "ATTN_SW_TIMER_4_FUNC!\n");
+
+		if (asserted & GPIO_2_FUNC)
+			DP(NETIF_MSG_HW, "GPIO_2_FUNC!\n");
+
+		if (asserted & GPIO_3_FUNC)
+			DP(NETIF_MSG_HW, "GPIO_3_FUNC!\n");
+
+		if (asserted & GPIO_4_FUNC)
+			DP(NETIF_MSG_HW, "GPIO_4_FUNC!\n");
+
+		if (port == 0) {
+			if (asserted & ATTN_GENERAL_ATTN_1) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_1!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_1, 0x0);
+			}
+			if (asserted & ATTN_GENERAL_ATTN_2) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_2!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_2, 0x0);
+			}
+			if (asserted & ATTN_GENERAL_ATTN_3) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_3!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_3, 0x0);
+			}
+		} else {
+			if (asserted & ATTN_GENERAL_ATTN_4) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_4!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_4, 0x0);
+			}
+			if (asserted & ATTN_GENERAL_ATTN_5) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_5!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_5, 0x0);
+			}
+			if (asserted & ATTN_GENERAL_ATTN_6) {
+				DP(NETIF_MSG_HW, "ATTN_GENERAL_ATTN_6!\n");
+				REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_6, 0x0);
+			}
+		}
+
+	} /* if hardwired */
+
+	DP(NETIF_MSG_HW, "about to mask 0x%08x at IGU addr 0x%x\n",
+	   asserted, BAR_IGU_INTMEM + igu_addr);
+	REG_WR(bp, BAR_IGU_INTMEM + igu_addr, asserted);
+
+	/* now set back the mask */
+	if (asserted & ATTN_NIG_FOR_FUNC)
+		REG_WR(bp, nig_mask_addr, bp->nig_mask);
+}
+
+static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
+{
+	int port = bp->port;
+	int index;
+	struct attn_route attn;
+	struct attn_route group_mask;
+	u32 reg_addr;
+	u32 val;
+
+	/* need to take HW lock because MCP or other port might also
+	   try to handle this event */
+	bnx2x_lock_alr(bp);
+
+	attn.sig[0] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_1_FUNC_0 + port*4);
+	attn.sig[1] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_2_FUNC_0 + port*4);
+	attn.sig[2] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_3_FUNC_0 + port*4);
+	attn.sig[3] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_4_FUNC_0 + port*4);
+	DP(NETIF_MSG_HW, "attn %llx\n", (unsigned long long)attn.sig[0]);
+
+	for (index = 0; index < MAX_DYNAMIC_ATTN_GRPS; index++) {
+		if (deasserted & (1 << index)) {
+			group_mask = bp->attn_group[index];
+
+			DP(NETIF_MSG_HW, "group[%d]: %llx\n", index,
+			   (unsigned long long)group_mask.sig[0]);
+
+			if (attn.sig[3] & group_mask.sig[3] &
+			    EVEREST_GEN_ATTN_IN_USE_MASK) {
+
+				if (attn.sig[3] & BNX2X_MC_ASSERT_BITS) {
+
+					BNX2X_ERR("MC assert!\n");
+					bnx2x_panic();
+
+				} else if (attn.sig[3] & BNX2X_MCP_ASSERT) {
+
+					BNX2X_ERR("MCP assert!\n");
+					REG_WR(bp,
+					     MISC_REG_AEU_GENERAL_ATTN_11, 0);
+					bnx2x_mc_assert(bp);
+
+				} else {
+					BNX2X_ERR("UNKOWEN HW ASSERT!\n");
+				}
+			}
+
+			if (attn.sig[1] & group_mask.sig[1] &
+			    BNX2X_DOORQ_ASSERT) {
+
+				val = REG_RD(bp, DORQ_REG_DORQ_INT_STS_CLR);
+				BNX2X_ERR("DB hw attention 0x%x\n", val);
+				/* DORQ discard attention */
+				if (val & 0x2)
+					BNX2X_ERR("FATAL error from DORQ\n");
+			}
+
+			if (attn.sig[2] & group_mask.sig[2] &
+			    AEU_INPUTS_ATTN_BITS_CFC_HW_INTERRUPT) {
+
+				val = REG_RD(bp, CFC_REG_CFC_INT_STS_CLR);
+				BNX2X_ERR("CFC hw attention 0x%x\n", val);
+				/* CFC error attention */
+				if (val & 0x2)
+					BNX2X_ERR("FATAL error from CFC\n");
+			}
+
+			if (attn.sig[2] & group_mask.sig[2] &
+			    AEU_INPUTS_ATTN_BITS_PXP_HW_INTERRUPT) {
+
+				val = REG_RD(bp, PXP_REG_PXP_INT_STS_CLR_0);
+				BNX2X_ERR("PXP hw attention 0x%x\n", val);
+				/* RQ_USDMDP_FIFO_OVERFLOW */
+				if (val & 0x18000)
+					BNX2X_ERR("FATAL error from PXP\n");
+			}
+
+			if (attn.sig[3] & group_mask.sig[3] &
+			    EVEREST_LATCHED_ATTN_IN_USE_MASK) {
+
+				REG_WR(bp, MISC_REG_AEU_CLR_LATCH_SIGNAL,
+				       0x7ff);
+				DP(NETIF_MSG_HW, "got latched bits 0x%x\n",
+				   attn.sig[3]);
+			}
+
+			if ((attn.sig[0] & group_mask.sig[0] &
+						HW_INTERRUT_ASSERT_SET_0) ||
+			    (attn.sig[1] & group_mask.sig[1] &
+						HW_INTERRUT_ASSERT_SET_1) ||
+			    (attn.sig[2] & group_mask.sig[2] &
+						HW_INTERRUT_ASSERT_SET_2))
+				BNX2X_ERR("FATAL HW block attention\n");
+
+			if ((attn.sig[0] & group_mask.sig[0] &
+						HW_PRTY_ASSERT_SET_0) ||
+			    (attn.sig[1] & group_mask.sig[1] &
+						HW_PRTY_ASSERT_SET_1) ||
+			    (attn.sig[2] & group_mask.sig[2] &
+						HW_PRTY_ASSERT_SET_2))
+				BNX2X_ERR("FATAL HW block parity atention\n");
+		}
+	}
+
+	bnx2x_unlock_alr(bp);
+
+	reg_addr = (IGU_ADDR_ATTN_BITS_CLR + IGU_PORT_BASE * port) * 8;
+
+	val = ~deasserted;
+/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
+	   val, BAR_IGU_INTMEM + reg_addr); */
+	REG_WR(bp, BAR_IGU_INTMEM + reg_addr, val);
+
+	if (bp->aeu_mask & (deasserted & 0xff))
+		BNX2X_ERR("IGU BUG\n");
+	if (~bp->attn_state & deasserted)
+		BNX2X_ERR("IGU BUG\n");
+
+	reg_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
+			  MISC_REG_AEU_MASK_ATTN_FUNC_0;
+
+	DP(NETIF_MSG_HW, "aeu_mask %x\n", bp->aeu_mask);
+	bp->aeu_mask |= (deasserted & 0xff);
+
+	DP(NETIF_MSG_HW, "new mask %x\n", bp->aeu_mask);
+	REG_WR(bp, reg_addr, bp->aeu_mask);
+
+	DP(NETIF_MSG_HW, "attn_state %x\n", bp->attn_state);
+	bp->attn_state &= ~deasserted;
+	DP(NETIF_MSG_HW, "new state %x\n", bp->attn_state);
+}
+
+static void bnx2x_attn_int(struct bnx2x *bp)
+{
+	/* read local copy of bits */
+	u32 attn_bits = bp->def_status_blk->atten_status_block.attn_bits;
+	u32 attn_ack = bp->def_status_blk->atten_status_block.attn_bits_ack;
+	u32 attn_state = bp->attn_state;
+
+	/* look for changed bits */
+	u32 asserted   =  attn_bits & ~attn_ack & ~attn_state;
+	u32 deasserted = ~attn_bits &  attn_ack &  attn_state;
+
+	DP(NETIF_MSG_HW,
+	   "attn_bits %x  attn_ack %x  asserted %x  deasserted %x\n",
+	   attn_bits, attn_ack, asserted, deasserted);
+
+	if (~(attn_bits ^ attn_ack) & (attn_bits ^ attn_state))
+		BNX2X_ERR("bad attention state\n");
+
+	/* handle bits that were raised */
+	if (asserted)
+		bnx2x_attn_int_asserted(bp, asserted);
+
+	if (deasserted)
+		bnx2x_attn_int_deasserted(bp, deasserted);
+}
+
+static void bnx2x_sp_task(struct work_struct *work)
+{
+	struct bnx2x *bp = container_of(work, struct bnx2x, sp_task);
+	u16 status;
+
+	/* Return here if interrupt is disabled */
+	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
+		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
+		return;
+	}
+
+	status = bnx2x_update_dsb_idx(bp);
+	if (status == 0)
+		BNX2X_ERR("spurious slowpath interrupt!\n");
+
+	DP(NETIF_MSG_INTR, "got a slowpath interrupt (updated %x)\n", status);
+
+	if (status & 0x1) {
+		/* HW attentions */
+		bnx2x_attn_int(bp);
+	}
+
+	/* CStorm events: query_stats, cfc delete ramrods */
+	if (status & 0x2)
+		bp->stat_pending = 0;
+
+	bnx2x_ack_sb(bp, DEF_SB_ID, ATTENTION_ID, bp->def_att_idx,
+		     IGU_INT_NOP, 1);
+	bnx2x_ack_sb(bp, DEF_SB_ID, USTORM_ID, le16_to_cpu(bp->def_u_idx),
+		     IGU_INT_NOP, 1);
+	bnx2x_ack_sb(bp, DEF_SB_ID, CSTORM_ID, le16_to_cpu(bp->def_c_idx),
+		     IGU_INT_NOP, 1);
+	bnx2x_ack_sb(bp, DEF_SB_ID, XSTORM_ID, le16_to_cpu(bp->def_x_idx),
+		     IGU_INT_NOP, 1);
+	bnx2x_ack_sb(bp, DEF_SB_ID, TSTORM_ID, le16_to_cpu(bp->def_t_idx),
+		     IGU_INT_ENABLE, 1);
+}
+
+static irqreturn_t bnx2x_msix_sp_int(int irq, void *dev_instance)
+{
+	struct net_device *dev = dev_instance;
+	struct bnx2x *bp = netdev_priv(dev);
+
+	/* Return here if interrupt is disabled */
+	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
+		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
+		return IRQ_HANDLED;
+	}
+
+	bnx2x_ack_sb(bp, 16, XSTORM_ID, 0, IGU_INT_DISABLE, 0);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return IRQ_HANDLED;
+#endif
+
+	schedule_work(&bp->sp_task);
+
+	return IRQ_HANDLED;
+}
+
+/* end of slow path */
+
+/* Statistics */
+
+/****************************************************************************
+* Macros
+****************************************************************************/
+
+#define UPDATE_STAT(s, t) \
+	do { \
+		estats->t += new->s - old->s; \
+		old->s = new->s; \
+	} while (0)
+
+/* sum[hi:lo] += add[hi:lo] */
+#define ADD_64(s_hi, a_hi, s_lo, a_lo) \
+	do { \
+		s_lo += a_lo; \
+		s_hi += a_hi + (s_lo < a_lo) ? 1 : 0; \
+	} while (0)
+
+/* difference = minuend - subtrahend */
+#define DIFF_64(d_hi, m_hi, s_hi, d_lo, m_lo, s_lo) \
+	do { \
+		if (m_lo < s_lo) {      /* underflow */ \
+			d_hi = m_hi - s_hi; \
+			if (d_hi > 0) { /* we can 'loan' 1 */ \
+				d_hi--; \
+				d_lo = m_lo + (UINT_MAX - s_lo) + 1; \
+			} else {	/* m_hi <= s_hi */ \
+				d_hi = 0; \
+				d_lo = 0; \
+			} \
+		} else {		/* m_lo >= s_lo */ \
+			if (m_hi < s_hi) { \
+			    d_hi = 0; \
+			    d_lo = 0; \
+			} else {	/* m_hi >= s_hi */ \
+			    d_hi = m_hi - s_hi; \
+			    d_lo = m_lo - s_lo; \
+			} \
+		} \
+	} while (0)
+
+/* minuend -= subtrahend */
+#define SUB_64(m_hi, s_hi, m_lo, s_lo) \
+	do { \
+		DIFF_64(m_hi, m_hi, s_hi, m_lo, m_lo, s_lo); \
+	} while (0)
+
+#define UPDATE_STAT64(s_hi, t_hi, s_lo, t_lo) \
+	do { \
+		DIFF_64(diff.hi, new->s_hi, old->s_hi, \
+			diff.lo, new->s_lo, old->s_lo); \
+		old->s_hi = new->s_hi; \
+		old->s_lo = new->s_lo; \
+		ADD_64(estats->t_hi, diff.hi, \
+		       estats->t_lo, diff.lo); \
+	} while (0)
+
+/* sum[hi:lo] += add */
+#define ADD_EXTEND_64(s_hi, s_lo, a) \
+	do { \
+		s_lo += a; \
+		s_hi += (s_lo < a) ? 1 : 0; \
+	} while (0)
+
+#define UPDATE_EXTEND_STAT(s, t_hi, t_lo) \
+	do { \
+		ADD_EXTEND_64(estats->t_hi, estats->t_lo, new->s); \
+	} while (0)
+
+#define UPDATE_EXTEND_TSTAT(s, t_hi, t_lo) \
+	do { \
+		diff = le32_to_cpu(tclient->s) - old_tclient->s; \
+		old_tclient->s = le32_to_cpu(tclient->s); \
+		ADD_EXTEND_64(estats->t_hi, estats->t_lo, diff); \
+	} while (0)
+
+/*
+ * General service functions
+ */
+
+static inline long bnx2x_hilo(u32 *hiref)
+{
+	u32 lo = *(hiref + 1);
+#if (BITS_PER_LONG == 64)
+	u32 hi = *hiref;
+
+	return HILO_U64(hi, lo);
+#else
+	return lo;
+#endif
+}
+
+/*
+ * Init service functions
+ */
+
+static void bnx2x_init_mac_stats(struct bnx2x *bp)
+{
+	struct dmae_command *dmae;
+	int port = bp->port;
+	int loader_idx = port * 8;
+	u32 opcode;
+	u32 mac_addr;
+
+	bp->executer_idx = 0;
+	if (bp->fw_mb) {
+		/* MCP */
+		opcode = (DMAE_CMD_SRC_PCI | DMAE_CMD_DST_GRC |
+			  DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+			  DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+			  DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+			  (port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+
+		if (bp->link_up)
+			opcode |= (DMAE_CMD_C_DST_GRC | DMAE_CMD_C_ENABLE);
+
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = U64_LO(bnx2x_sp_mapping(bp, eth_stats) +
+					   sizeof(u32));
+		dmae->src_addr_hi = U64_HI(bnx2x_sp_mapping(bp, eth_stats) +
+					   sizeof(u32));
+		dmae->dst_addr_lo = bp->fw_mb >> 2;
+		dmae->dst_addr_hi = 0;
+		dmae->len = (offsetof(struct bnx2x_eth_stats, mac_stx_end) -
+			     sizeof(u32)) >> 2;
+		if (bp->link_up) {
+			dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+			dmae->comp_addr_hi = 0;
+			dmae->comp_val = 1;
+		} else {
+			dmae->comp_addr_lo = 0;
+			dmae->comp_addr_hi = 0;
+			dmae->comp_val = 0;
+		}
+	}
+
+	if (!bp->link_up) {
+		/* no need to collect statistics in link down */
+		return;
+	}
+
+	opcode = (DMAE_CMD_SRC_GRC | DMAE_CMD_DST_PCI |
+		  DMAE_CMD_C_DST_GRC | DMAE_CMD_C_ENABLE |
+		  DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+		  DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+		  DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+		  (port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+
+	if (bp->phy_flags & PHY_BMAC_FLAG) {
+
+		mac_addr = (port ? NIG_REG_INGRESS_BMAC1_MEM :
+				   NIG_REG_INGRESS_BMAC0_MEM);
+
+		/* BIGMAC_REGISTER_TX_STAT_GTPKT ..
+		   BIGMAC_REGISTER_TX_STAT_GTBYT */
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = (mac_addr +
+				     BIGMAC_REGISTER_TX_STAT_GTPKT) >> 2;
+		dmae->src_addr_hi = 0;
+		dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, mac_stats));
+		dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, mac_stats));
+		dmae->len = (8 + BIGMAC_REGISTER_TX_STAT_GTBYT -
+			     BIGMAC_REGISTER_TX_STAT_GTPKT) >> 2;
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+
+		/* BIGMAC_REGISTER_RX_STAT_GR64 ..
+		   BIGMAC_REGISTER_RX_STAT_GRIPJ */
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = (mac_addr +
+				     BIGMAC_REGISTER_RX_STAT_GR64) >> 2;
+		dmae->src_addr_hi = 0;
+		dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, mac_stats) +
+					offsetof(struct bmac_stats, rx_gr64));
+		dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, mac_stats) +
+					offsetof(struct bmac_stats, rx_gr64));
+		dmae->len = (8 + BIGMAC_REGISTER_RX_STAT_GRIPJ -
+			     BIGMAC_REGISTER_RX_STAT_GR64) >> 2;
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+
+	} else if (bp->phy_flags & PHY_EMAC_FLAG) {
+
+		mac_addr = (port ? GRCBASE_EMAC1 : GRCBASE_EMAC0);
+
+		/* EMAC_REG_EMAC_RX_STAT_AC (EMAC_REG_EMAC_RX_STAT_AC_COUNT)*/
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = (mac_addr +
+				     EMAC_REG_EMAC_RX_STAT_AC) >> 2;
+		dmae->src_addr_hi = 0;
+		dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, mac_stats));
+		dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, mac_stats));
+		dmae->len = EMAC_REG_EMAC_RX_STAT_AC_COUNT;
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+
+		/* EMAC_REG_EMAC_RX_STAT_AC_28 */
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = (mac_addr +
+				     EMAC_REG_EMAC_RX_STAT_AC_28) >> 2;
+		dmae->src_addr_hi = 0;
+		dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, mac_stats) +
+					   offsetof(struct emac_stats,
+						    rx_falsecarriererrors));
+		dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, mac_stats) +
+					   offsetof(struct emac_stats,
+						    rx_falsecarriererrors));
+		dmae->len = 1;
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+
+		/* EMAC_REG_EMAC_TX_STAT_AC (EMAC_REG_EMAC_TX_STAT_AC_COUNT)*/
+		dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+		dmae->opcode = opcode;
+		dmae->src_addr_lo = (mac_addr +
+				     EMAC_REG_EMAC_TX_STAT_AC) >> 2;
+		dmae->src_addr_hi = 0;
+		dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, mac_stats) +
+					   offsetof(struct emac_stats,
+						    tx_ifhcoutoctets));
+		dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, mac_stats) +
+					   offsetof(struct emac_stats,
+						    tx_ifhcoutoctets));
+		dmae->len = EMAC_REG_EMAC_TX_STAT_AC_COUNT;
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+	}
+
+	/* NIG */
+	dmae = bnx2x_sp(bp, dmae[bp->executer_idx++]);
+	dmae->opcode = (DMAE_CMD_SRC_GRC | DMAE_CMD_DST_PCI |
+			DMAE_CMD_C_DST_PCI | DMAE_CMD_C_ENABLE |
+			DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+			DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+			DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+			(port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+	dmae->src_addr_lo = (port ? NIG_REG_STAT1_BRB_DISCARD :
+				    NIG_REG_STAT0_BRB_DISCARD) >> 2;
+	dmae->src_addr_hi = 0;
+	dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, nig));
+	dmae->dst_addr_hi = U64_HI(bnx2x_sp_mapping(bp, nig));
+	dmae->len = (sizeof(struct nig_stats) - 2*sizeof(u32)) >> 2;
+	dmae->comp_addr_lo = U64_LO(bnx2x_sp_mapping(bp, nig) +
+				    offsetof(struct nig_stats, done));
+	dmae->comp_addr_hi = U64_HI(bnx2x_sp_mapping(bp, nig) +
+				    offsetof(struct nig_stats, done));
+	dmae->comp_val = 0xffffffff;
+}
+
+static void bnx2x_init_stats(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	bp->stats_state = STATS_STATE_DISABLE;
+	bp->executer_idx = 0;
+
+	bp->old_brb_discard = REG_RD(bp,
+				     NIG_REG_STAT0_BRB_DISCARD + port*0x38);
+
+	memset(&bp->old_bmac, 0, sizeof(struct bmac_stats));
+	memset(&bp->old_tclient, 0, sizeof(struct tstorm_per_client_stats));
+	memset(&bp->dev->stats, 0, sizeof(struct net_device_stats));
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_STATS_FLAGS_OFFSET(port), 1);
+	REG_WR(bp, BAR_XSTRORM_INTMEM +
+	       XSTORM_STATS_FLAGS_OFFSET(port) + 4, 0);
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_STATS_FLAGS_OFFSET(port), 1);
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_STATS_FLAGS_OFFSET(port) + 4, 0);
+
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_STATS_FLAGS_OFFSET(port), 0);
+	REG_WR(bp, BAR_CSTRORM_INTMEM +
+	       CSTORM_STATS_FLAGS_OFFSET(port) + 4, 0);
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM +
+	       XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port),
+	       U64_LO(bnx2x_sp_mapping(bp, fw_stats)));
+	REG_WR(bp, BAR_XSTRORM_INTMEM +
+	       XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port) + 4,
+	       U64_HI(bnx2x_sp_mapping(bp, fw_stats)));
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port),
+	       U64_LO(bnx2x_sp_mapping(bp, fw_stats)));
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port) + 4,
+	       U64_HI(bnx2x_sp_mapping(bp, fw_stats)));
+}
+
+static void bnx2x_stop_stats(struct bnx2x *bp)
+{
+	might_sleep();
+	if (bp->stats_state != STATS_STATE_DISABLE) {
+		int timeout = 10;
+
+		bp->stats_state = STATS_STATE_STOP;
+		DP(BNX2X_MSG_STATS, "stats_state - STOP\n");
+
+		while (bp->stats_state != STATS_STATE_DISABLE) {
+			if (!timeout) {
+				BNX2X_ERR("timeout wating for stats stop\n");
+				break;
+			}
+			timeout--;
+			msleep(100);
+		}
+	}
+	DP(BNX2X_MSG_STATS, "stats_state - DISABLE\n");
+}
+
+/*
+ * Statistics service functions
+ */
+
+static void bnx2x_update_bmac_stats(struct bnx2x *bp)
+{
+	struct regp diff;
+	struct regp sum;
+	struct bmac_stats *new = bnx2x_sp(bp, mac_stats.bmac);
+	struct bmac_stats *old = &bp->old_bmac;
+	struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
+
+	sum.hi = 0;
+	sum.lo = 0;
+
+	UPDATE_STAT64(tx_gtbyt.hi, total_bytes_transmitted_hi,
+		      tx_gtbyt.lo, total_bytes_transmitted_lo);
+
+	UPDATE_STAT64(tx_gtmca.hi, total_multicast_packets_transmitted_hi,
+		      tx_gtmca.lo, total_multicast_packets_transmitted_lo);
+	ADD_64(sum.hi, diff.hi, sum.lo, diff.lo);
+
+	UPDATE_STAT64(tx_gtgca.hi, total_broadcast_packets_transmitted_hi,
+		      tx_gtgca.lo, total_broadcast_packets_transmitted_lo);
+	ADD_64(sum.hi, diff.hi, sum.lo, diff.lo);
+
+	UPDATE_STAT64(tx_gtpkt.hi, total_unicast_packets_transmitted_hi,
+		      tx_gtpkt.lo, total_unicast_packets_transmitted_lo);
+	SUB_64(estats->total_unicast_packets_transmitted_hi, sum.hi,
+	       estats->total_unicast_packets_transmitted_lo, sum.lo);
+
+	UPDATE_STAT(tx_gtxpf.lo, pause_xoff_frames_transmitted);
+	UPDATE_STAT(tx_gt64.lo, frames_transmitted_64_bytes);
+	UPDATE_STAT(tx_gt127.lo, frames_transmitted_65_127_bytes);
+	UPDATE_STAT(tx_gt255.lo, frames_transmitted_128_255_bytes);
+	UPDATE_STAT(tx_gt511.lo, frames_transmitted_256_511_bytes);
+	UPDATE_STAT(tx_gt1023.lo, frames_transmitted_512_1023_bytes);
+	UPDATE_STAT(tx_gt1518.lo, frames_transmitted_1024_1522_bytes);
+	UPDATE_STAT(tx_gt2047.lo, frames_transmitted_1523_9022_bytes);
+	UPDATE_STAT(tx_gt4095.lo, frames_transmitted_1523_9022_bytes);
+	UPDATE_STAT(tx_gt9216.lo, frames_transmitted_1523_9022_bytes);
+	UPDATE_STAT(tx_gt16383.lo, frames_transmitted_1523_9022_bytes);
+
+	UPDATE_STAT(rx_grfcs.lo, crc_receive_errors);
+	UPDATE_STAT(rx_grund.lo, runt_packets_received);
+	UPDATE_STAT(rx_grovr.lo, stat_Dot3statsFramesTooLong);
+	UPDATE_STAT(rx_grxpf.lo, pause_xoff_frames_received);
+	UPDATE_STAT(rx_grxcf.lo, control_frames_received);
+	/* UPDATE_STAT(rx_grxpf.lo, control_frames_received); */
+	UPDATE_STAT(rx_grfrg.lo, error_runt_packets_received);
+	UPDATE_STAT(rx_grjbr.lo, error_jabber_packets_received);
+
+	UPDATE_STAT64(rx_grerb.hi, stat_IfHCInBadOctets_hi,
+		      rx_grerb.lo, stat_IfHCInBadOctets_lo);
+	UPDATE_STAT64(tx_gtufl.hi, stat_IfHCOutBadOctets_hi,
+		      tx_gtufl.lo, stat_IfHCOutBadOctets_lo);
+	UPDATE_STAT(tx_gterr.lo, stat_Dot3statsInternalMacTransmitErrors);
+	/* UPDATE_STAT(rx_grxpf.lo, stat_XoffStateEntered); */
+	estats->stat_XoffStateEntered = estats->pause_xoff_frames_received;
+}
+
+static void bnx2x_update_emac_stats(struct bnx2x *bp)
+{
+	struct emac_stats *new = bnx2x_sp(bp, mac_stats.emac);
+	struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
+
+	UPDATE_EXTEND_STAT(tx_ifhcoutoctets, total_bytes_transmitted_hi,
+					     total_bytes_transmitted_lo);
+	UPDATE_EXTEND_STAT(tx_ifhcoutucastpkts,
+					total_unicast_packets_transmitted_hi,
+					total_unicast_packets_transmitted_lo);
+	UPDATE_EXTEND_STAT(tx_ifhcoutmulticastpkts,
+				      total_multicast_packets_transmitted_hi,
+				      total_multicast_packets_transmitted_lo);
+	UPDATE_EXTEND_STAT(tx_ifhcoutbroadcastpkts,
+				      total_broadcast_packets_transmitted_hi,
+				      total_broadcast_packets_transmitted_lo);
+
+	estats->pause_xon_frames_transmitted += new->tx_outxonsent;
+	estats->pause_xoff_frames_transmitted += new->tx_outxoffsent;
+	estats->single_collision_transmit_frames +=
+				new->tx_dot3statssinglecollisionframes;
+	estats->multiple_collision_transmit_frames +=
+				new->tx_dot3statsmultiplecollisionframes;
+	estats->late_collision_frames += new->tx_dot3statslatecollisions;
+	estats->excessive_collision_frames +=
+				new->tx_dot3statsexcessivecollisions;
+	estats->frames_transmitted_64_bytes += new->tx_etherstatspkts64octets;
+	estats->frames_transmitted_65_127_bytes +=
+				new->tx_etherstatspkts65octetsto127octets;
+	estats->frames_transmitted_128_255_bytes +=
+				new->tx_etherstatspkts128octetsto255octets;
+	estats->frames_transmitted_256_511_bytes +=
+				new->tx_etherstatspkts256octetsto511octets;
+	estats->frames_transmitted_512_1023_bytes +=
+				new->tx_etherstatspkts512octetsto1023octets;
+	estats->frames_transmitted_1024_1522_bytes +=
+				new->tx_etherstatspkts1024octetsto1522octet;
+	estats->frames_transmitted_1523_9022_bytes +=
+				new->tx_etherstatspktsover1522octets;
+
+	estats->crc_receive_errors += new->rx_dot3statsfcserrors;
+	estats->alignment_errors += new->rx_dot3statsalignmenterrors;
+	estats->false_carrier_detections += new->rx_falsecarriererrors;
+	estats->runt_packets_received += new->rx_etherstatsundersizepkts;
+	estats->stat_Dot3statsFramesTooLong += new->rx_dot3statsframestoolong;
+	estats->pause_xon_frames_received += new->rx_xonpauseframesreceived;
+	estats->pause_xoff_frames_received += new->rx_xoffpauseframesreceived;
+	estats->control_frames_received += new->rx_maccontrolframesreceived;
+	estats->error_runt_packets_received += new->rx_etherstatsfragments;
+	estats->error_jabber_packets_received += new->rx_etherstatsjabbers;
+
+	UPDATE_EXTEND_STAT(rx_ifhcinbadoctets, stat_IfHCInBadOctets_hi,
+					       stat_IfHCInBadOctets_lo);
+	UPDATE_EXTEND_STAT(tx_ifhcoutbadoctets, stat_IfHCOutBadOctets_hi,
+						stat_IfHCOutBadOctets_lo);
+	estats->stat_Dot3statsInternalMacTransmitErrors +=
+				new->tx_dot3statsinternalmactransmiterrors;
+	estats->stat_Dot3StatsCarrierSenseErrors +=
+				new->rx_dot3statscarriersenseerrors;
+	estats->stat_Dot3StatsDeferredTransmissions +=
+				new->tx_dot3statsdeferredtransmissions;
+	estats->stat_FlowControlDone += new->tx_flowcontroldone;
+	estats->stat_XoffStateEntered += new->rx_xoffstateentered;
+}
+
+static int bnx2x_update_storm_stats(struct bnx2x *bp)
+{
+	struct eth_stats_query *stats = bnx2x_sp(bp, fw_stats);
+	struct tstorm_common_stats *tstats = &stats->tstorm_common;
+	struct tstorm_per_client_stats *tclient =
+						&tstats->client_statistics[0];
+	struct tstorm_per_client_stats *old_tclient = &bp->old_tclient;
+	struct xstorm_common_stats *xstats = &stats->xstorm_common;
+	struct nig_stats *nstats = bnx2x_sp(bp, nig);
+	struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
+	u32 diff;
+
+	/* are DMAE stats valid? */
+	if (nstats->done != 0xffffffff) {
+		DP(BNX2X_MSG_STATS, "stats not updated by dmae\n");
+		return -1;
+	}
+
+	/* are storm stats valid? */
+	if (tstats->done.hi != 0xffffffff) {
+		DP(BNX2X_MSG_STATS, "stats not updated by tstorm\n");
+		return -2;
+	}
+	if (xstats->done.hi != 0xffffffff) {
+		DP(BNX2X_MSG_STATS, "stats not updated by xstorm\n");
+		return -3;
+	}
+
+	estats->total_bytes_received_hi =
+	estats->valid_bytes_received_hi =
+				le32_to_cpu(tclient->total_rcv_bytes.hi);
+	estats->total_bytes_received_lo =
+	estats->valid_bytes_received_lo =
+				le32_to_cpu(tclient->total_rcv_bytes.lo);
+	ADD_64(estats->total_bytes_received_hi,
+	       le32_to_cpu(tclient->rcv_error_bytes.hi),
+	       estats->total_bytes_received_lo,
+	       le32_to_cpu(tclient->rcv_error_bytes.lo));
+
+	UPDATE_EXTEND_TSTAT(rcv_unicast_pkts,
+					total_unicast_packets_received_hi,
+					total_unicast_packets_received_lo);
+	UPDATE_EXTEND_TSTAT(rcv_multicast_pkts,
+					total_multicast_packets_received_hi,
+					total_multicast_packets_received_lo);
+	UPDATE_EXTEND_TSTAT(rcv_broadcast_pkts,
+					total_broadcast_packets_received_hi,
+					total_broadcast_packets_received_lo);
+
+	estats->frames_received_64_bytes = MAC_STX_NA;
+	estats->frames_received_65_127_bytes = MAC_STX_NA;
+	estats->frames_received_128_255_bytes = MAC_STX_NA;
+	estats->frames_received_256_511_bytes = MAC_STX_NA;
+	estats->frames_received_512_1023_bytes = MAC_STX_NA;
+	estats->frames_received_1024_1522_bytes = MAC_STX_NA;
+	estats->frames_received_1523_9022_bytes = MAC_STX_NA;
+
+	estats->x_total_sent_bytes_hi =
+				le32_to_cpu(xstats->total_sent_bytes.hi);
+	estats->x_total_sent_bytes_lo =
+				le32_to_cpu(xstats->total_sent_bytes.lo);
+	estats->x_total_sent_pkts = le32_to_cpu(xstats->total_sent_pkts);
+
+	estats->t_rcv_unicast_bytes_hi =
+				le32_to_cpu(tclient->rcv_unicast_bytes.hi);
+	estats->t_rcv_unicast_bytes_lo =
+				le32_to_cpu(tclient->rcv_unicast_bytes.lo);
+	estats->t_rcv_broadcast_bytes_hi =
+				le32_to_cpu(tclient->rcv_broadcast_bytes.hi);
+	estats->t_rcv_broadcast_bytes_lo =
+				le32_to_cpu(tclient->rcv_broadcast_bytes.lo);
+	estats->t_rcv_multicast_bytes_hi =
+				le32_to_cpu(tclient->rcv_multicast_bytes.hi);
+	estats->t_rcv_multicast_bytes_lo =
+				le32_to_cpu(tclient->rcv_multicast_bytes.lo);
+	estats->t_total_rcv_pkt = le32_to_cpu(tclient->total_rcv_pkts);
+
+	estats->checksum_discard = le32_to_cpu(tclient->checksum_discard);
+	estats->packets_too_big_discard =
+				le32_to_cpu(tclient->packets_too_big_discard);
+	estats->jabber_packets_received = estats->packets_too_big_discard +
+					  estats->stat_Dot3statsFramesTooLong;
+	estats->no_buff_discard = le32_to_cpu(tclient->no_buff_discard);
+	estats->ttl0_discard = le32_to_cpu(tclient->ttl0_discard);
+	estats->mac_discard = le32_to_cpu(tclient->mac_discard);
+	estats->mac_filter_discard = le32_to_cpu(tstats->mac_filter_discard);
+	estats->xxoverflow_discard = le32_to_cpu(tstats->xxoverflow_discard);
+	estats->brb_truncate_discard =
+				le32_to_cpu(tstats->brb_truncate_discard);
+
+	estats->brb_discard += nstats->brb_discard - bp->old_brb_discard;
+	bp->old_brb_discard = nstats->brb_discard;
+
+	estats->brb_packet = nstats->brb_packet;
+	estats->brb_truncate = nstats->brb_truncate;
+	estats->flow_ctrl_discard = nstats->flow_ctrl_discard;
+	estats->flow_ctrl_octets = nstats->flow_ctrl_octets;
+	estats->flow_ctrl_packet = nstats->flow_ctrl_packet;
+	estats->mng_discard = nstats->mng_discard;
+	estats->mng_octet_inp = nstats->mng_octet_inp;
+	estats->mng_octet_out = nstats->mng_octet_out;
+	estats->mng_packet_inp = nstats->mng_packet_inp;
+	estats->mng_packet_out = nstats->mng_packet_out;
+	estats->pbf_octets = nstats->pbf_octets;
+	estats->pbf_packet = nstats->pbf_packet;
+	estats->safc_inp = nstats->safc_inp;
+
+	xstats->done.hi = 0;
+	tstats->done.hi = 0;
+	nstats->done = 0;
+
+	return 0;
+}
+
+static void bnx2x_update_net_stats(struct bnx2x *bp)
+{
+	struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
+	struct net_device_stats *nstats = &bp->dev->stats;
+
+	nstats->rx_packets =
+		bnx2x_hilo(&estats->total_unicast_packets_received_hi) +
+		bnx2x_hilo(&estats->total_multicast_packets_received_hi) +
+		bnx2x_hilo(&estats->total_broadcast_packets_received_hi);
+
+	nstats->tx_packets =
+		bnx2x_hilo(&estats->total_unicast_packets_transmitted_hi) +
+		bnx2x_hilo(&estats->total_multicast_packets_transmitted_hi) +
+		bnx2x_hilo(&estats->total_broadcast_packets_transmitted_hi);
+
+	nstats->rx_bytes = bnx2x_hilo(&estats->total_bytes_received_hi);
+
+	nstats->tx_bytes =
+		bnx2x_hilo(&estats->total_bytes_transmitted_hi);
+
+	nstats->rx_dropped = estats->checksum_discard +
+				   estats->mac_discard;
+	nstats->tx_dropped = 0;
+
+	nstats->multicast =
+		bnx2x_hilo(&estats->total_multicast_packets_transmitted_hi);
+
+	nstats->collisions =
+		estats->single_collision_transmit_frames +
+		estats->multiple_collision_transmit_frames +
+		estats->late_collision_frames +
+		estats->excessive_collision_frames;
+
+	nstats->rx_length_errors = estats->runt_packets_received +
+				   estats->jabber_packets_received;
+	nstats->rx_over_errors = estats->no_buff_discard;
+	nstats->rx_crc_errors = estats->crc_receive_errors;
+	nstats->rx_frame_errors = estats->alignment_errors;
+	nstats->rx_fifo_errors = estats->brb_discard +
+				       estats->brb_truncate_discard;
+	nstats->rx_missed_errors = estats->xxoverflow_discard;
+
+	nstats->rx_errors = nstats->rx_length_errors +
+			    nstats->rx_over_errors +
+			    nstats->rx_crc_errors +
+			    nstats->rx_frame_errors +
+			    nstats->rx_fifo_errors;
+
+	nstats->tx_aborted_errors = estats->late_collision_frames +
+					  estats->excessive_collision_frames;
+	nstats->tx_carrier_errors = estats->false_carrier_detections;
+	nstats->tx_fifo_errors = 0;
+	nstats->tx_heartbeat_errors = 0;
+	nstats->tx_window_errors = 0;
+
+	nstats->tx_errors = nstats->tx_aborted_errors +
+			    nstats->tx_carrier_errors;
+
+	estats->mac_stx_start = ++estats->mac_stx_end;
+}
+
+static void bnx2x_update_stats(struct bnx2x *bp)
+{
+	int i;
+
+	if (!bnx2x_update_storm_stats(bp)) {
+
+		if (bp->phy_flags & PHY_BMAC_FLAG) {
+			bnx2x_update_bmac_stats(bp);
+
+		} else if (bp->phy_flags & PHY_EMAC_FLAG) {
+			bnx2x_update_emac_stats(bp);
+
+		} else { /* unreached */
+			BNX2X_ERR("no MAC active\n");
+			return;
+		}
+
+		bnx2x_update_net_stats(bp);
+	}
+
+	if (bp->msglevel & NETIF_MSG_TIMER) {
+		struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
+		struct net_device_stats *nstats = &bp->dev->stats;
+
+		printk(KERN_DEBUG "%s:\n", bp->dev->name);
+		printk(KERN_DEBUG "  tx avail (%4x)  tx hc idx (%x)"
+				  "  tx pkt (%lx)\n",
+		       bnx2x_tx_avail(bp->fp),
+		       *bp->fp->tx_cons_sb, nstats->tx_packets);
+		printk(KERN_DEBUG "  rx usage (%4x)  rx hc idx (%x)"
+				  "  rx pkt (%lx)\n",
+		       (u16)(*bp->fp->rx_cons_sb - bp->fp->rx_comp_cons),
+		       *bp->fp->rx_cons_sb, nstats->rx_packets);
+		printk(KERN_DEBUG "  %s (Xoff events %u)  brb drops %u\n",
+		       netif_queue_stopped(bp->dev)? "Xoff" : "Xon",
+		       estats->driver_xoff, estats->brb_discard);
+		printk(KERN_DEBUG "tstats: checksum_discard %u  "
+			"packets_too_big_discard %u  no_buff_discard %u  "
+			"mac_discard %u  mac_filter_discard %u  "
+			"xxovrflow_discard %u  brb_truncate_discard %u  "
+			"ttl0_discard %u\n",
+		       estats->checksum_discard,
+		       estats->packets_too_big_discard,
+		       estats->no_buff_discard, estats->mac_discard,
+		       estats->mac_filter_discard, estats->xxoverflow_discard,
+		       estats->brb_truncate_discard, estats->ttl0_discard);
+
+		for_each_queue(bp, i) {
+			printk(KERN_DEBUG "[%d]: %lu\t%lu\t%lu\n", i,
+			       bnx2x_fp(bp, i, tx_pkt),
+			       bnx2x_fp(bp, i, rx_pkt),
+			       bnx2x_fp(bp, i, rx_calls));
+		}
+	}
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(BNX2X_MSG_STATS, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return;
+#endif
+
+	/* loader */
+	if (bp->executer_idx) {
+		struct dmae_command *dmae = &bp->dmae;
+		int port = bp->port;
+		int loader_idx = port * 8;
+
+		memset(dmae, 0, sizeof(struct dmae_command));
+
+		dmae->opcode = (DMAE_CMD_SRC_PCI | DMAE_CMD_DST_GRC |
+				DMAE_CMD_C_DST_GRC | DMAE_CMD_C_ENABLE |
+				DMAE_CMD_DST_RESET |
+#ifdef __BIG_ENDIAN
+				DMAE_CMD_ENDIANITY_B_DW_SWAP |
+#else
+				DMAE_CMD_ENDIANITY_DW_SWAP |
+#endif
+				(port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+		dmae->src_addr_lo = U64_LO(bnx2x_sp_mapping(bp, dmae[0]));
+		dmae->src_addr_hi = U64_HI(bnx2x_sp_mapping(bp, dmae[0]));
+		dmae->dst_addr_lo = (DMAE_REG_CMD_MEM +
+				     sizeof(struct dmae_command) *
+				     (loader_idx + 1)) >> 2;
+		dmae->dst_addr_hi = 0;
+		dmae->len = sizeof(struct dmae_command) >> 2;
+		dmae->len--;    /* !!! for A0/1 only */
+		dmae->comp_addr_lo = dmae_reg_go_c[loader_idx + 1] >> 2;
+		dmae->comp_addr_hi = 0;
+		dmae->comp_val = 1;
+
+		bnx2x_post_dmae(bp, dmae, loader_idx);
+	}
+
+	if (bp->stats_state != STATS_STATE_ENABLE) {
+		bp->stats_state = STATS_STATE_DISABLE;
+		return;
+	}
+
+	if (bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_STAT_QUERY, 0, 0, 0, 0) == 0) {
+		/* stats ramrod has it's own slot on the spe */
+		bp->spq_left++;
+		bp->stat_pending = 1;
+	}
+}
+
+static void bnx2x_timer(unsigned long data)
+{
+	struct bnx2x *bp = (struct bnx2x *) data;
+
+	if (!netif_running(bp->dev))
+		return;
+
+	if (atomic_read(&bp->intr_sem) != 0)
+		goto bnx2x_restart_timer;
+
+	if (poll) {
+		struct bnx2x_fastpath *fp = &bp->fp[0];
+		int rc;
+
+		bnx2x_tx_int(fp, 1000);
+		rc = bnx2x_rx_int(fp, 1000);
+	}
+
+	if (!nomcp && (bp->bc_ver >= 0x040003)) {
+		int port = bp->port;
+		u32 drv_pulse;
+		u32 mcp_pulse;
+
+		++bp->fw_drv_pulse_wr_seq;
+		bp->fw_drv_pulse_wr_seq &= DRV_PULSE_SEQ_MASK;
+		/* TBD - add SYSTEM_TIME */
+		drv_pulse = bp->fw_drv_pulse_wr_seq;
+		SHMEM_WR(bp, drv_fw_mb[port].drv_pulse_mb, drv_pulse);
+
+		mcp_pulse = (SHMEM_RD(bp, drv_fw_mb[port].mcp_pulse_mb) &
+			     MCP_PULSE_SEQ_MASK);
+		/* The delta between driver pulse and mcp response
+		 * should be 1 (before mcp response) or 0 (after mcp response)
+		 */
+		if ((drv_pulse != mcp_pulse) &&
+		    (drv_pulse != ((mcp_pulse + 1) & MCP_PULSE_SEQ_MASK))) {
+			/* someone lost a heartbeat... */
+			BNX2X_ERR("drv_pulse (0x%x) != mcp_pulse (0x%x)\n",
+				  drv_pulse, mcp_pulse);
+		}
+	}
+
+	if (bp->stats_state == STATS_STATE_DISABLE)
+		goto bnx2x_restart_timer;
+
+	bnx2x_update_stats(bp);
+
+bnx2x_restart_timer:
+	mod_timer(&bp->timer, jiffies + bp->current_interval);
+}
+
+/* end of Statistics */
+
+/* nic init */
+
+/*
+ * nic init service functions
+ */
+
+static void bnx2x_init_sb(struct bnx2x *bp, struct host_status_block *sb,
+			  dma_addr_t mapping, int id)
+{
+	int port = bp->port;
+	u64 section;
+	int index;
+
+	/* USTORM */
+	section = ((u64)mapping) + offsetof(struct host_status_block,
+					    u_status_block);
+	sb->u_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_USTRORM_INTMEM +
+	       USTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	REG_WR(bp, BAR_USTRORM_INTMEM +
+	       ((USTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       U64_HI(section));
+
+	for (index = 0; index < HC_USTORM_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_USTRORM_INTMEM +
+			 USTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+
+	/* CSTORM */
+	section = ((u64)mapping) + offsetof(struct host_status_block,
+					    c_status_block);
+	sb->c_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_CSTRORM_INTMEM +
+	       CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	REG_WR(bp, BAR_CSTRORM_INTMEM +
+	       ((CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       U64_HI(section));
+
+	for (index = 0; index < HC_CSTORM_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_CSTRORM_INTMEM +
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+
+	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+}
+
+static void bnx2x_init_def_sb(struct bnx2x *bp,
+			      struct host_def_status_block *def_sb,
+			      dma_addr_t mapping, int id)
+{
+	int port = bp->port;
+	int index, val, reg_offset;
+	u64 section;
+
+	/* ATTN */
+	section = ((u64)mapping) + offsetof(struct host_def_status_block,
+					    atten_status_block);
+	def_sb->atten_status_block.status_block_id = id;
+
+	reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
+			     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
+
+	for (index = 0; index < 3; index++) {
+		bp->attn_group[index].sig[0] = REG_RD(bp,
+						     reg_offset + 0x10*index);
+		bp->attn_group[index].sig[1] = REG_RD(bp,
+					       reg_offset + 0x4 + 0x10*index);
+		bp->attn_group[index].sig[2] = REG_RD(bp,
+					       reg_offset + 0x8 + 0x10*index);
+		bp->attn_group[index].sig[3] = REG_RD(bp,
+					       reg_offset + 0xc + 0x10*index);
+	}
+
+	bp->aeu_mask = REG_RD(bp, (port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
+					  MISC_REG_AEU_MASK_ATTN_FUNC_0));
+
+	reg_offset = (port ? HC_REG_ATTN_MSG1_ADDR_L :
+			     HC_REG_ATTN_MSG0_ADDR_L);
+
+	REG_WR(bp, reg_offset, U64_LO(section));
+	REG_WR(bp, reg_offset + 4, U64_HI(section));
+
+	reg_offset = (port ? HC_REG_ATTN_NUM_P1 : HC_REG_ATTN_NUM_P0);
+
+	val = REG_RD(bp, reg_offset);
+	val |= id;
+	REG_WR(bp, reg_offset, val);
+
+	/* USTORM */
+	section = ((u64)mapping) + offsetof(struct host_def_status_block,
+					    u_def_status_block);
+	def_sb->u_def_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_USTRORM_INTMEM +
+	       USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	REG_WR(bp, BAR_USTRORM_INTMEM +
+	       ((USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       U64_HI(section));
+	REG_WR(bp, BAR_USTRORM_INTMEM + USTORM_HC_BTR_OFFSET(port),
+	       BNX2X_BTR);
+
+	for (index = 0; index < HC_USTORM_DEF_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_USTRORM_INTMEM +
+			 USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+
+	/* CSTORM */
+	section = ((u64)mapping) + offsetof(struct host_def_status_block,
+					    c_def_status_block);
+	def_sb->c_def_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_CSTRORM_INTMEM +
+	       CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	REG_WR(bp, BAR_CSTRORM_INTMEM +
+	       ((CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       U64_HI(section));
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_HC_BTR_OFFSET(port),
+	       BNX2X_BTR);
+
+	for (index = 0; index < HC_CSTORM_DEF_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_CSTRORM_INTMEM +
+			 CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+
+	/* TSTORM */
+	section = ((u64)mapping) + offsetof(struct host_def_status_block,
+					    t_def_status_block);
+	def_sb->t_def_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       ((TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       U64_HI(section));
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_HC_BTR_OFFSET(port),
+	       BNX2X_BTR);
+
+	for (index = 0; index < HC_TSTORM_DEF_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_TSTRORM_INTMEM +
+			 TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+
+	/* XSTORM */
+	section = ((u64)mapping) + offsetof(struct host_def_status_block,
+					    x_def_status_block);
+	def_sb->x_def_status_block.status_block_id = id;
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM +
+	       XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	REG_WR(bp, BAR_XSTRORM_INTMEM +
+	       ((XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       U64_HI(section));
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_HC_BTR_OFFSET(port),
+	       BNX2X_BTR);
+
+	for (index = 0; index < HC_XSTORM_DEF_SB_NUM_INDICES; index++)
+		REG_WR16(bp, BAR_XSTRORM_INTMEM +
+			 XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+
+	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+}
+
+static void bnx2x_update_coalesce(struct bnx2x *bp)
+{
+	int port = bp->port;
+	int i;
+
+	for_each_queue(bp, i) {
+
+		/* HC_INDEX_U_ETH_RX_CQ_CONS */
+		REG_WR8(bp, BAR_USTRORM_INTMEM +
+			USTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+						   HC_INDEX_U_ETH_RX_CQ_CONS),
+			bp->rx_ticks_int/12);
+		REG_WR16(bp, BAR_USTRORM_INTMEM +
+			 USTORM_SB_HC_DISABLE_OFFSET(port, i,
+						   HC_INDEX_U_ETH_RX_CQ_CONS),
+			 bp->rx_ticks_int ? 0 : 1);
+
+		/* HC_INDEX_C_ETH_TX_CQ_CONS */
+		REG_WR8(bp, BAR_CSTRORM_INTMEM +
+			CSTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+						   HC_INDEX_C_ETH_TX_CQ_CONS),
+			bp->tx_ticks_int/12);
+		REG_WR16(bp, BAR_CSTRORM_INTMEM +
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, i,
+						   HC_INDEX_C_ETH_TX_CQ_CONS),
+			 bp->tx_ticks_int ? 0 : 1);
+	}
+}
+
+static void bnx2x_init_rx_rings(struct bnx2x *bp)
+{
+	u16 ring_prod;
+	int i, j;
+	int port = bp->port;
+
+	bp->rx_buf_use_size = bp->dev->mtu;
+
+	bp->rx_buf_use_size += bp->rx_offset + ETH_OVREHEAD;
+	bp->rx_buf_size = bp->rx_buf_use_size + 64;
+
+	for_each_queue(bp, j) {
+		struct bnx2x_fastpath *fp = &bp->fp[j];
+
+		fp->rx_bd_cons = 0;
+		fp->rx_cons_sb = BNX2X_RX_SB_INDEX;
+
+		for (i = 1; i <= NUM_RX_RINGS; i++) {
+			struct eth_rx_bd *rx_bd;
+
+			rx_bd = &fp->rx_desc_ring[RX_DESC_CNT * i - 2];
+			rx_bd->addr_hi =
+				cpu_to_le32(U64_HI(fp->rx_desc_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+			rx_bd->addr_lo =
+				cpu_to_le32(U64_LO(fp->rx_desc_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+
+		}
+
+		for (i = 1; i <= NUM_RCQ_RINGS; i++) {
+			struct eth_rx_cqe_next_page *nextpg;
+
+			nextpg = (struct eth_rx_cqe_next_page *)
+				&fp->rx_comp_ring[RCQ_DESC_CNT * i - 1];
+			nextpg->addr_hi =
+				cpu_to_le32(U64_HI(fp->rx_comp_mapping +
+					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+			nextpg->addr_lo =
+				cpu_to_le32(U64_LO(fp->rx_comp_mapping +
+					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+		}
+
+		/* rx completion queue */
+		fp->rx_comp_cons = ring_prod = 0;
+
+		for (i = 0; i < bp->rx_ring_size; i++) {
+			if (bnx2x_alloc_rx_skb(bp, fp, ring_prod) < 0) {
+				BNX2X_ERR("was only able to allocate "
+					  "%d rx skbs\n", i);
+				break;
+			}
+			ring_prod = NEXT_RX_IDX(ring_prod);
+			BUG_TRAP(ring_prod > i);
+		}
+
+		fp->rx_bd_prod = fp->rx_comp_prod = ring_prod;
+		fp->rx_pkt = fp->rx_calls = 0;
+
+		/* Warning! this will genrate an interrupt (to the TSTORM) */
+		/* must only be done when chip is initialized */
+		REG_WR(bp, BAR_TSTRORM_INTMEM +
+		       TSTORM_RCQ_PROD_OFFSET(port, j), ring_prod);
+		if (j != 0)
+			continue;
+
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port),
+		       U64_LO(fp->rx_comp_mapping));
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port) + 4,
+		       U64_HI(fp->rx_comp_mapping));
+	}
+}
+
+static void bnx2x_init_tx_ring(struct bnx2x *bp)
+{
+	int i, j;
+
+	for_each_queue(bp, j) {
+		struct bnx2x_fastpath *fp = &bp->fp[j];
+
+		for (i = 1; i <= NUM_TX_RINGS; i++) {
+			struct eth_tx_bd *tx_bd =
+				&fp->tx_desc_ring[TX_DESC_CNT * i - 1];
+
+			tx_bd->addr_hi =
+				cpu_to_le32(U64_HI(fp->tx_desc_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+			tx_bd->addr_lo =
+				cpu_to_le32(U64_LO(fp->tx_desc_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+		}
+
+		fp->tx_pkt_prod = 0;
+		fp->tx_pkt_cons = 0;
+		fp->tx_bd_prod = 0;
+		fp->tx_bd_cons = 0;
+		fp->tx_cons_sb = BNX2X_TX_SB_INDEX;
+		fp->tx_pkt = 0;
+	}
+}
+
+static void bnx2x_init_sp_ring(struct bnx2x *bp)
+{
+	int port = bp->port;
+
+	spin_lock_init(&bp->spq_lock);
+
+	bp->spq_left = MAX_SPQ_PENDING;
+	bp->spq_prod_idx = 0;
+	bp->dsb_sp_prod_idx = 0;
+	bp->dsb_sp_prod = BNX2X_SP_DSB_INDEX;
+	bp->spq_prod_bd = bp->spq;
+	bp->spq_last_bd = bp->spq_prod_bd + MAX_SP_DESC_CNT;
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port),
+	       U64_LO(bp->spq_mapping));
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port) + 4,
+	       U64_HI(bp->spq_mapping));
+
+	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PROD_OFFSET(port),
+	       bp->spq_prod_idx);
+}
+
+static void bnx2x_init_context(struct bnx2x *bp)
+{
+	int i;
+
+	for_each_queue(bp, i) {
+		struct eth_context *context = bnx2x_sp(bp, context[i].eth);
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+
+		context->xstorm_st_context.tx_bd_page_base_hi =
+						U64_HI(fp->tx_desc_mapping);
+		context->xstorm_st_context.tx_bd_page_base_lo =
+						U64_LO(fp->tx_desc_mapping);
+		context->xstorm_st_context.db_data_addr_hi =
+						U64_HI(fp->tx_prods_mapping);
+		context->xstorm_st_context.db_data_addr_lo =
+						U64_LO(fp->tx_prods_mapping);
+
+		context->ustorm_st_context.rx_bd_page_base_hi =
+						U64_HI(fp->rx_desc_mapping);
+		context->ustorm_st_context.rx_bd_page_base_lo =
+						U64_LO(fp->rx_desc_mapping);
+		context->ustorm_st_context.status_block_id = i;
+		context->ustorm_st_context.sb_index_number =
+						HC_INDEX_U_ETH_RX_CQ_CONS;
+		context->ustorm_st_context.rcq_base_address_hi =
+						U64_HI(fp->rx_comp_mapping);
+		context->ustorm_st_context.rcq_base_address_lo =
+						U64_LO(fp->rx_comp_mapping);
+		context->ustorm_st_context.flags =
+				USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT;
+		context->ustorm_st_context.mc_alignment_size = 64;
+		context->ustorm_st_context.num_rss = bp->num_queues;
+
+		context->cstorm_st_context.sb_index_number =
+						HC_INDEX_C_ETH_TX_CQ_CONS;
+		context->cstorm_st_context.status_block_id = i;
+
+		context->xstorm_ag_context.cdu_reserved =
+			CDU_RSRVD_VALUE_TYPE_A(HW_CID(bp, i),
+					       CDU_REGION_NUMBER_XCM_AG,
+					       ETH_CONNECTION_TYPE);
+		context->ustorm_ag_context.cdu_usage =
+			CDU_RSRVD_VALUE_TYPE_A(HW_CID(bp, i),
+					       CDU_REGION_NUMBER_UCM_AG,
+					       ETH_CONNECTION_TYPE);
+	}
+}
+
+static void bnx2x_init_ind_table(struct bnx2x *bp)
+{
+	int port = bp->port;
+	int i;
+
+	if (!is_multi(bp))
+		return;
+
+	for (i = 0; i < TSTORM_INDIRECTION_TABLE_SIZE; i++)
+		REG_WR8(bp, TSTORM_INDIRECTION_TABLE_OFFSET(port) + i,
+			i % bp->num_queues);
+
+	REG_WR(bp, PRS_REG_A_PRSU_20, 0xf);
+}
+
+static void bnx2x_set_storm_rx_mode(struct bnx2x *bp)
+{
+	int mode = bp->rx_mode;
+	int port = bp->port;
+	struct tstorm_eth_mac_filter_config tstorm_mac_filter = {0};
+	int i;
+
+	DP(NETIF_MSG_RX_STATUS, "rx mode is %d\n", mode);
+
+	switch (mode) {
+	case BNX2X_RX_MODE_NONE: /* no Rx */
+		tstorm_mac_filter.ucast_drop_all = 1;
+		tstorm_mac_filter.mcast_drop_all = 1;
+		tstorm_mac_filter.bcast_drop_all = 1;
+		break;
+	case BNX2X_RX_MODE_NORMAL:
+		tstorm_mac_filter.bcast_accept_all = 1;
+		break;
+	case BNX2X_RX_MODE_ALLMULTI:
+		tstorm_mac_filter.mcast_accept_all = 1;
+		tstorm_mac_filter.bcast_accept_all = 1;
+		break;
+	case BNX2X_RX_MODE_PROMISC:
+		tstorm_mac_filter.ucast_accept_all = 1;
+		tstorm_mac_filter.mcast_accept_all = 1;
+		tstorm_mac_filter.bcast_accept_all = 1;
+		break;
+	default:
+		BNX2X_ERR("bad rx mode (%d)\n", mode);
+	}
+
+	for (i = 0; i < sizeof(struct tstorm_eth_mac_filter_config)/4; i++) {
+		REG_WR(bp, BAR_TSTRORM_INTMEM +
+		       TSTORM_MAC_FILTER_CONFIG_OFFSET(port) + i * 4,
+		       ((u32 *)&tstorm_mac_filter)[i]);
+
+/*      	DP(NETIF_MSG_IFUP, "tstorm_mac_filter[%d]: 0x%08x\n", i,
+		   ((u32 *)&tstorm_mac_filter)[i]); */
+	}
+}
+
+static void bnx2x_set_client_config(struct bnx2x *bp, int client_id)
+{
+#ifdef BCM_VLAN
+	int mode = bp->rx_mode;
+#endif
+	int port = bp->port;
+	struct tstorm_eth_client_config tstorm_client = {0};
+
+	tstorm_client.mtu = bp->dev->mtu;
+	tstorm_client.statistics_counter_id = 0;
+	tstorm_client.config_flags =
+		TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE;
+#ifdef BCM_VLAN
+	if (mode && bp->vlgrp) {
+		tstorm_client.config_flags |=
+				TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE;
+		DP(NETIF_MSG_IFUP, "vlan removal enabled\n");
+	}
+#endif
+	tstorm_client.drop_flags = (TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR |
+				    TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR |
+				    TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR |
+				    TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR);
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_CLIENT_CONFIG_OFFSET(port, client_id),
+	       ((u32 *)&tstorm_client)[0]);
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_CLIENT_CONFIG_OFFSET(port, client_id) + 4,
+	       ((u32 *)&tstorm_client)[1]);
+
+/*      DP(NETIF_MSG_IFUP, "tstorm_client: 0x%08x 0x%08x\n",
+	   ((u32 *)&tstorm_client)[0], ((u32 *)&tstorm_client)[1]); */
+}
+
+static void bnx2x_init_internal(struct bnx2x *bp)
+{
+	int port = bp->port;
+	struct tstorm_eth_function_common_config tstorm_config = {0};
+	struct stats_indication_flags stats_flags = {0};
+	int i;
+
+	if (is_multi(bp)) {
+		tstorm_config.config_flags = MULTI_FLAGS;
+		tstorm_config.rss_result_mask = MULTI_MASK;
+	}
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM +
+	       TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(port),
+	       (*(u32 *)&tstorm_config));
+
+/*      DP(NETIF_MSG_IFUP, "tstorm_config: 0x%08x\n",
+	   (*(u32 *)&tstorm_config)); */
+
+	bp->rx_mode = BNX2X_RX_MODE_NONE; /* no rx untill link is up */
+	bnx2x_set_storm_rx_mode(bp);
+
+	for_each_queue(bp, i)
+		bnx2x_set_client_config(bp, i);
+
+
+	stats_flags.collect_eth = cpu_to_le32(1);
+
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_STATS_FLAGS_OFFSET(port),
+	       ((u32 *)&stats_flags)[0]);
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_STATS_FLAGS_OFFSET(port) + 4,
+	       ((u32 *)&stats_flags)[1]);
+
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_STATS_FLAGS_OFFSET(port),
+	       ((u32 *)&stats_flags)[0]);
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_STATS_FLAGS_OFFSET(port) + 4,
+	       ((u32 *)&stats_flags)[1]);
+
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_STATS_FLAGS_OFFSET(port),
+	       ((u32 *)&stats_flags)[0]);
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_STATS_FLAGS_OFFSET(port) + 4,
+	       ((u32 *)&stats_flags)[1]);
+
+/*      DP(NETIF_MSG_IFUP, "stats_flags: 0x%08x 0x%08x\n",
+	   ((u32 *)&stats_flags)[0], ((u32 *)&stats_flags)[1]); */
+}
+
+static void bnx2x_nic_init(struct bnx2x *bp)
+{
+	int i;
+
+	for_each_queue(bp, i) {
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+
+		fp->state = BNX2X_FP_STATE_CLOSED;
+		DP(NETIF_MSG_IFUP, "bnx2x_init_sb(%p,%p,%d);\n",
+		   bp, fp->status_blk, i);
+		fp->index = i;
+		bnx2x_init_sb(bp, fp->status_blk, fp->status_blk_mapping, i);
+	}
+
+	bnx2x_init_def_sb(bp, bp->def_status_blk,
+			  bp->def_status_blk_mapping, 0x10);
+	bnx2x_update_coalesce(bp);
+	bnx2x_init_rx_rings(bp);
+	bnx2x_init_tx_ring(bp);
+	bnx2x_init_sp_ring(bp);
+	bnx2x_init_context(bp);
+	bnx2x_init_internal(bp);
+	bnx2x_init_stats(bp);
+	bnx2x_init_ind_table(bp);
+	bnx2x_enable_int(bp);
+
+}
+
+/* end of nic init */
+
+/*
+ * gzip service functions
+ */
+
+static int bnx2x_gunzip_init(struct bnx2x *bp)
+{
+	bp->gunzip_buf = pci_alloc_consistent(bp->pdev, FW_BUF_SIZE,
+					      &bp->gunzip_mapping);
+	if (bp->gunzip_buf  == NULL)
+		goto gunzip_nomem1;
+
+	bp->strm = kmalloc(sizeof(*bp->strm), GFP_KERNEL);
+	if (bp->strm  == NULL)
+		goto gunzip_nomem2;
+
+	bp->strm->workspace = kmalloc(zlib_inflate_workspacesize(),
+				      GFP_KERNEL);
+	if (bp->strm->workspace == NULL)
+		goto gunzip_nomem3;
+
+	return 0;
+
+gunzip_nomem3:
+	kfree(bp->strm);
+	bp->strm = NULL;
+
+gunzip_nomem2:
+	pci_free_consistent(bp->pdev, FW_BUF_SIZE, bp->gunzip_buf,
+			    bp->gunzip_mapping);
+	bp->gunzip_buf = NULL;
+
+gunzip_nomem1:
+	printk(KERN_ERR PFX "%s: Cannot allocate firmware buffer for"
+	       " uncompression\n", bp->dev->name);
+	return -ENOMEM;
+}
+
+static void bnx2x_gunzip_end(struct bnx2x *bp)
+{
+	kfree(bp->strm->workspace);
+
+	kfree(bp->strm);
+	bp->strm = NULL;
+
+	if (bp->gunzip_buf) {
+		pci_free_consistent(bp->pdev, FW_BUF_SIZE, bp->gunzip_buf,
+				    bp->gunzip_mapping);
+		bp->gunzip_buf = NULL;
+	}
+}
+
+static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len)
+{
+	int n, rc;
+
+	/* check gzip header */
+	if ((zbuf[0] != 0x1f) || (zbuf[1] != 0x8b) || (zbuf[2] != Z_DEFLATED))
+		return -EINVAL;
+
+	n = 10;
+
+#define FNAME   			0x8
+
+	if (zbuf[3] & FNAME)
+		while ((zbuf[n++] != 0) && (n < len));
+
+	bp->strm->next_in = zbuf + n;
+	bp->strm->avail_in = len - n;
+	bp->strm->next_out = bp->gunzip_buf;
+	bp->strm->avail_out = FW_BUF_SIZE;
+
+	rc = zlib_inflateInit2(bp->strm, -MAX_WBITS);
+	if (rc != Z_OK)
+		return rc;
+
+	rc = zlib_inflate(bp->strm, Z_FINISH);
+	if ((rc != Z_OK) && (rc != Z_STREAM_END))
+		printk(KERN_ERR PFX "%s: Firmware decompression error: %s\n",
+		       bp->dev->name, bp->strm->msg);
+
+	bp->gunzip_outlen = (FW_BUF_SIZE - bp->strm->avail_out);
+	if (bp->gunzip_outlen & 0x3)
+		printk(KERN_ERR PFX "%s: Firmware decompression error:"
+				    " gunzip_outlen (%d) not aligned\n",
+		       bp->dev->name, bp->gunzip_outlen);
+	bp->gunzip_outlen >>= 2;
+
+	zlib_inflateEnd(bp->strm);
+
+	if (rc == Z_STREAM_END)
+		return 0;
+
+	return rc;
+}
+
+/* nic load/unload */
+
+/*
+ * general service functions
+ */
+
+/* send a NIG loopback debug packet */
+static void bnx2x_lb_pckt(struct bnx2x *bp)
+{
+#ifdef USE_DMAE
+	u32 wb_write[3];
+#endif
+
+	/* Ethernet source and destination addresses */
+#ifdef USE_DMAE
+	wb_write[0] = 0x55555555;
+	wb_write[1] = 0x55555555;
+	wb_write[2] = 0x20;     	/* SOP */
+	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
+#else
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x55555555);
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
+	/* SOP */
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x20);
+#endif
+
+	/* NON-IP protocol */
+#ifdef USE_DMAE
+	wb_write[0] = 0x09000000;
+	wb_write[1] = 0x55555555;
+	wb_write[2] = 0x10;     	/* EOP, eop_bvalid = 0 */
+	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
+#else
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x09000000);
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
+	/* EOP, eop_bvalid = 0 */
+	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x10);
+#endif
+}
+
+/* some of the internal memories
+ * are not directly readable from the driver
+ * to test them we send debug packets
+ */
+static int bnx2x_int_mem_test(struct bnx2x *bp)
+{
+	int factor;
+	int count, i;
+	u32 val = 0;
+
+	switch (CHIP_REV(bp)) {
+	case CHIP_REV_EMUL:
+		factor = 200;
+		break;
+	case CHIP_REV_FPGA:
+		factor = 120;
+		break;
+	default:
+		factor = 1;
+		break;
+	}
+
+	DP(NETIF_MSG_HW, "start part1\n");
+
+	/* Disable inputs of parser neighbor blocks */
+	REG_WR(bp, TSDM_REG_ENABLE_IN1, 0x0);
+	REG_WR(bp, TCM_REG_PRS_IFEN, 0x0);
+	REG_WR(bp, CFC_REG_DEBUG0, 0x1);
+	NIG_WR(NIG_REG_PRS_REQ_IN_EN, 0x0);
+
+	/*  Write 0 to parser credits for CFC search request */
+	REG_WR(bp, PRS_REG_CFC_SEARCH_INITIAL_CREDIT, 0x0);
+
+	/* send Ethernet packet */
+	bnx2x_lb_pckt(bp);
+
+	/* TODO do i reset NIG statistic? */
+	/* Wait until NIG register shows 1 packet of size 0x10 */
+	count = 1000 * factor;
+	while (count) {
+#ifdef BNX2X_DMAE_RD
+		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
+		val = *bnx2x_sp(bp, wb_data[0]);
+#else
+		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
+		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
+#endif
+		if (val == 0x10)
+			break;
+
+		msleep(10);
+		count--;
+	}
+	if (val != 0x10) {
+		BNX2X_ERR("NIG timeout  val = 0x%x\n", val);
+		return -1;
+	}
+
+	/* Wait until PRS register shows 1 packet */
+	count = 1000 * factor;
+	while (count) {
+		val = REG_RD(bp, PRS_REG_NUM_OF_PACKETS);
+
+		if (val == 1)
+			break;
+
+		msleep(10);
+		count--;
+	}
+	if (val != 0x1) {
+		BNX2X_ERR("PRS timeout val = 0x%x\n", val);
+		return -2;
+	}
+
+	/* Reset and init BRB, PRS */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x3);
+	msleep(50);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x3);
+	msleep(50);
+	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+
+	DP(NETIF_MSG_HW, "part2\n");
+
+	/* Disable inputs of parser neighbor blocks */
+	REG_WR(bp, TSDM_REG_ENABLE_IN1, 0x0);
+	REG_WR(bp, TCM_REG_PRS_IFEN, 0x0);
+	REG_WR(bp, CFC_REG_DEBUG0, 0x1);
+	NIG_WR(NIG_REG_PRS_REQ_IN_EN, 0x0);
+
+	/* Write 0 to parser credits for CFC search request */
+	REG_WR(bp, PRS_REG_CFC_SEARCH_INITIAL_CREDIT, 0x0);
+
+	/* send 10 Ethernet packets */
+	for (i = 0; i < 10; i++)
+		bnx2x_lb_pckt(bp);
+
+	/* Wait until NIG register shows 10 + 1
+	   packets of size 11*0x10 = 0xb0 */
+	count = 1000 * factor;
+	while (count) {
+#ifdef BNX2X_DMAE_RD
+		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
+		val = *bnx2x_sp(bp, wb_data[0]);
+#else
+		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
+		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
+#endif
+		if (val == 0xb0)
+			break;
+
+		msleep(10);
+		count--;
+	}
+	if (val != 0xb0) {
+		BNX2X_ERR("NIG timeout  val = 0x%x\n", val);
+		return -3;
+	}
+
+	/* Wait until PRS register shows 2 packets */
+	val = REG_RD(bp, PRS_REG_NUM_OF_PACKETS);
+	if (val != 2)
+		BNX2X_ERR("PRS timeout  val = 0x%x\n", val);
+
+	/* Write 1 to parser credits for CFC search request */
+	REG_WR(bp, PRS_REG_CFC_SEARCH_INITIAL_CREDIT, 0x1);
+
+	/* Wait until PRS register shows 3 packets */
+	msleep(10 * factor);
+	/* Wait until NIG register shows 1 packet of size 0x10 */
+	val = REG_RD(bp, PRS_REG_NUM_OF_PACKETS);
+	if (val != 3)
+		BNX2X_ERR("PRS timeout  val = 0x%x\n", val);
+
+	/* clear NIG EOP FIFO */
+	for (i = 0; i < 11; i++)
+		REG_RD(bp, NIG_REG_INGRESS_EOP_LB_FIFO);
+	val = REG_RD(bp, NIG_REG_INGRESS_EOP_LB_EMPTY);
+	if (val != 1) {
+		BNX2X_ERR("clear of NIG failed\n");
+		return -4;
+	}
+
+	/* Reset and init BRB, PRS, NIG */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x03);
+	msleep(50);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x03);
+	msleep(50);
+	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+#ifndef BCM_ISCSI
+	/* set NIC mode */
+	REG_WR(bp, PRS_REG_NIC_MODE, 1);
+#endif
+
+	/* Enable inputs of parser neighbor blocks */
+	REG_WR(bp, TSDM_REG_ENABLE_IN1, 0x7fffffff);
+	REG_WR(bp, TCM_REG_PRS_IFEN, 0x1);
+	REG_WR(bp, CFC_REG_DEBUG0, 0x0);
+	NIG_WR(NIG_REG_PRS_REQ_IN_EN, 0x1);
+
+	DP(NETIF_MSG_HW, "done\n");
+
+	return 0; /* OK */
+}
+
+static void enable_blocks_attention(struct bnx2x *bp)
+{
+	REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
+	REG_WR(bp, PXP_REG_PXP_INT_MASK_1, 0);
+	REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
+	REG_WR(bp, CFC_REG_CFC_INT_MASK, 0);
+	REG_WR(bp, QM_REG_QM_INT_MASK, 0);
+	REG_WR(bp, TM_REG_TM_INT_MASK, 0);
+	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_0, 0);
+	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_1, 0);
+	REG_WR(bp, XCM_REG_XCM_INT_MASK, 0);
+/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_0, 0); */
+/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_1, 0); */
+	REG_WR(bp, USDM_REG_USDM_INT_MASK_0, 0);
+	REG_WR(bp, USDM_REG_USDM_INT_MASK_1, 0);
+	REG_WR(bp, UCM_REG_UCM_INT_MASK, 0);
+/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_0, 0); */
+/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_1, 0); */
+	REG_WR(bp, GRCBASE_UPB + PB_REG_PB_INT_MASK, 0);
+	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_0, 0);
+	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_1, 0);
+	REG_WR(bp, CCM_REG_CCM_INT_MASK, 0);
+/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_0, 0); */
+/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_1, 0); */
+	REG_WR(bp, PXP2_REG_PXP2_INT_MASK, 0x480000);
+	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_0, 0);
+	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_1, 0);
+	REG_WR(bp, TCM_REG_TCM_INT_MASK, 0);
+/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_0, 0); */
+/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_1, 0); */
+	REG_WR(bp, CDU_REG_CDU_INT_MASK, 0);
+	REG_WR(bp, DMAE_REG_DMAE_INT_MASK, 0);
+/*      REG_WR(bp, MISC_REG_MISC_INT_MASK, 0); */
+	REG_WR(bp, PBF_REG_PBF_INT_MASK, 0X18); 	/* bit 3,4 masked */
+}
+
+static int bnx2x_function_init(struct bnx2x *bp, int mode)
+{
+	int func = bp->port;
+	int port = func ? PORT1 : PORT0;
+	u32 val, i;
+#ifdef USE_DMAE
+	u32 wb_write[2];
+#endif
+
+	DP(BNX2X_MSG_MCP, "function is %d  mode is %x\n", func, mode);
+	if ((func != 0) && (func != 1)) {
+		BNX2X_ERR("BAD function number (%d)\n", func);
+		return -ENODEV;
+	}
+
+	bnx2x_gunzip_init(bp);
+
+	if (mode & 0x1) {       /* init common */
+		DP(BNX2X_MSG_MCP, "starting common init  func %d  mode %x\n",
+		   func, mode);
+		REG_WR(bp, MISC_REG_RESET_REG_1, 0xffffffff);
+		REG_WR(bp, MISC_REG_RESET_REG_2, 0xfffc);
+		bnx2x_init_block(bp, MISC_COMMON_START, MISC_COMMON_END);
+
+		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x100);
+		msleep(30);
+		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x0);
+
+		bnx2x_init_block(bp, PXP_COMMON_START, PXP_COMMON_END);
+		bnx2x_init_block(bp, PXP2_COMMON_START, PXP2_COMMON_END);
+
+		bnx2x_init_pxp(bp);
+
+		if (CHIP_REV(bp) == CHIP_REV_Ax) {
+			/* enable HW interrupt from PXP on USDM
+			   overflow bit 16 on INT_MASK_0 */
+			REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
+		}
+
+#ifdef __BIG_ENDIAN
+		REG_WR(bp, PXP2_REG_RQ_QM_ENDIAN_M, 1);
+		REG_WR(bp, PXP2_REG_RQ_TM_ENDIAN_M, 1);
+		REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
+		REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
+		REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
+		REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 1);
+
+/*      	REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
+		REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
+		REG_WR(bp, PXP2_REG_RD_TM_SWAP_MODE, 1);
+		REG_WR(bp, PXP2_REG_RD_SRC_SWAP_MODE, 1);
+		REG_WR(bp, PXP2_REG_RD_CDURD_SWAP_MODE, 1);
+#endif
+
+#ifndef BCM_ISCSI
+		/* set NIC mode */
+		REG_WR(bp, PRS_REG_NIC_MODE, 1);
+#endif
+
+		REG_WR(bp, PXP2_REG_RQ_CDU_P_SIZE, 5);
+#ifdef BCM_ISCSI
+		REG_WR(bp, PXP2_REG_RQ_TM_P_SIZE, 5);
+		REG_WR(bp, PXP2_REG_RQ_QM_P_SIZE, 5);
+		REG_WR(bp, PXP2_REG_RQ_SRC_P_SIZE, 5);
+#endif
+
+		bnx2x_init_block(bp, DMAE_COMMON_START, DMAE_COMMON_END);
+
+		/* let the HW do it's magic ... */
+		msleep(100);
+		/* finish PXP init
+		   (can be moved up if we want to use the DMAE) */
+		val = REG_RD(bp, PXP2_REG_RQ_CFG_DONE);
+		if (val != 1) {
+			BNX2X_ERR("PXP2 CFG failed\n");
+			return -EBUSY;
+		}
+
+		val = REG_RD(bp, PXP2_REG_RD_INIT_DONE);
+		if (val != 1) {
+			BNX2X_ERR("PXP2 RD_INIT failed\n");
+			return -EBUSY;
+		}
+
+		REG_WR(bp, PXP2_REG_RQ_DISABLE_INPUTS, 0);
+		REG_WR(bp, PXP2_REG_RD_DISABLE_INPUTS, 0);
+
+		bnx2x_init_fill(bp, TSEM_REG_PRAM, 0, 8);
+
+		bnx2x_init_block(bp, TCM_COMMON_START, TCM_COMMON_END);
+		bnx2x_init_block(bp, UCM_COMMON_START, UCM_COMMON_END);
+		bnx2x_init_block(bp, CCM_COMMON_START, CCM_COMMON_END);
+		bnx2x_init_block(bp, XCM_COMMON_START, XCM_COMMON_END);
+
+#ifdef BNX2X_DMAE_RD
+		bnx2x_read_dmae(bp, XSEM_REG_PASSIVE_BUFFER, 3);
+		bnx2x_read_dmae(bp, CSEM_REG_PASSIVE_BUFFER, 3);
+		bnx2x_read_dmae(bp, TSEM_REG_PASSIVE_BUFFER, 3);
+		bnx2x_read_dmae(bp, USEM_REG_PASSIVE_BUFFER, 3);
+#else
+		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER);
+		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 4);
+		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 8);
+		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER);
+		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 4);
+		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 8);
+		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER);
+		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 4);
+		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 8);
+		REG_RD(bp, USEM_REG_PASSIVE_BUFFER);
+		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 4);
+		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 8);
+#endif
+		bnx2x_init_block(bp, QM_COMMON_START, QM_COMMON_END);
+		/* softrest pulse */
+		REG_WR(bp, QM_REG_SOFT_RESET, 1);
+		REG_WR(bp, QM_REG_SOFT_RESET, 0);
+
+#ifdef BCM_ISCSI
+		bnx2x_init_block(bp, TIMERS_COMMON_START, TIMERS_COMMON_END);
+#endif
+		bnx2x_init_block(bp, DQ_COMMON_START, DQ_COMMON_END);
+		REG_WR(bp, DORQ_REG_DPM_CID_OFST, BCM_PAGE_BITS);
+		if (CHIP_REV(bp) == CHIP_REV_Ax) {
+			/* enable hw interrupt from doorbell Q */
+			REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
+		}
+
+		bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+
+		if (CHIP_REV_IS_SLOW(bp)) {
+			/* fix for emulation and FPGA for no pause */
+			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_0, 513);
+			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_1, 513);
+			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_0, 0);
+			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_1, 0);
+		}
+
+		bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+
+		bnx2x_init_block(bp, TSDM_COMMON_START, TSDM_COMMON_END);
+		bnx2x_init_block(bp, CSDM_COMMON_START, CSDM_COMMON_END);
+		bnx2x_init_block(bp, USDM_COMMON_START, USDM_COMMON_END);
+		bnx2x_init_block(bp, XSDM_COMMON_START, XSDM_COMMON_END);
+
+		bnx2x_init_fill(bp, TSTORM_INTMEM_ADDR, 0, STORM_INTMEM_SIZE);
+		bnx2x_init_fill(bp, CSTORM_INTMEM_ADDR, 0, STORM_INTMEM_SIZE);
+		bnx2x_init_fill(bp, XSTORM_INTMEM_ADDR, 0, STORM_INTMEM_SIZE);
+		bnx2x_init_fill(bp, USTORM_INTMEM_ADDR, 0, STORM_INTMEM_SIZE);
+
+		bnx2x_init_block(bp, TSEM_COMMON_START, TSEM_COMMON_END);
+		bnx2x_init_block(bp, USEM_COMMON_START, USEM_COMMON_END);
+		bnx2x_init_block(bp, CSEM_COMMON_START, CSEM_COMMON_END);
+		bnx2x_init_block(bp, XSEM_COMMON_START, XSEM_COMMON_END);
+
+		/* sync semi rtc */
+		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+		       0x80000000);
+		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
+		       0x80000000);
+
+		bnx2x_init_block(bp, UPB_COMMON_START, UPB_COMMON_END);
+		bnx2x_init_block(bp, XPB_COMMON_START, XPB_COMMON_END);
+		bnx2x_init_block(bp, PBF_COMMON_START, PBF_COMMON_END);
+
+		REG_WR(bp, SRC_REG_SOFT_RST, 1);
+		for (i = SRC_REG_KEYRSS0_0; i <= SRC_REG_KEYRSS1_9; i += 4) {
+			REG_WR(bp, i, 0xc0cac01a);
+			/* TODO: repleace with something meaningfull */
+		}
+		/* SRCH COMMON comes here */
+		REG_WR(bp, SRC_REG_SOFT_RST, 0);
+
+		if (sizeof(union cdu_context) != 1024) {
+			/* we currently assume that a context is 1024 bytes */
+			printk(KERN_ALERT PFX "please adjust the size of"
+			       " cdu_context(%ld)\n",
+			       (long)sizeof(union cdu_context));
+		}
+		val = (4 << 24) + (0 << 12) + 1024;
+		REG_WR(bp, CDU_REG_CDU_GLOBAL_PARAMS, val);
+		bnx2x_init_block(bp, CDU_COMMON_START, CDU_COMMON_END);
+
+		bnx2x_init_block(bp, CFC_COMMON_START, CFC_COMMON_END);
+		REG_WR(bp, CFC_REG_INIT_REG, 0x7FF);
+
+		bnx2x_init_block(bp, HC_COMMON_START, HC_COMMON_END);
+		bnx2x_init_block(bp, MISC_AEU_COMMON_START,
+				 MISC_AEU_COMMON_END);
+		/* RXPCS COMMON comes here */
+		/* EMAC0 COMMON comes here */
+		/* EMAC1 COMMON comes here */
+		/* DBU COMMON comes here */
+		/* DBG COMMON comes here */
+		bnx2x_init_block(bp, NIG_COMMON_START, NIG_COMMON_END);
+
+		if (CHIP_REV_IS_SLOW(bp))
+			msleep(200);
+
+		/* finish CFC init */
+		val = REG_RD(bp, CFC_REG_LL_INIT_DONE);
+		if (val != 1) {
+			BNX2X_ERR("CFC LL_INIT failed\n");
+			return -EBUSY;
+		}
+
+		val = REG_RD(bp, CFC_REG_AC_INIT_DONE);
+		if (val != 1) {
+			BNX2X_ERR("CFC AC_INIT failed\n");
+			return -EBUSY;
+		}
+
+		val = REG_RD(bp, CFC_REG_CAM_INIT_DONE);
+		if (val != 1) {
+			BNX2X_ERR("CFC CAM_INIT failed\n");
+			return -EBUSY;
+		}
+
+		REG_WR(bp, CFC_REG_DEBUG0, 0);
+
+		/* read NIG statistic
+		   to see if this is our first up since powerup */
+#ifdef BNX2X_DMAE_RD
+		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
+		val = *bnx2x_sp(bp, wb_data[0]);
+#else
+		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
+		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
+#endif
+		/* do internal memory self test */
+		if ((val == 0) && bnx2x_int_mem_test(bp)) {
+			BNX2X_ERR("internal mem selftest failed\n");
+			return -EBUSY;
+		}
+
+		/* clear PXP2 attentions */
+		REG_RD(bp, PXP2_REG_PXP2_INT_STS_CLR);
+
+		enable_blocks_attention(bp);
+		/* enable_blocks_parity(bp); */
+
+	} /* end of common init */
+
+	/* per port init */
+
+	/* the phys address is shifted right 12 bits and has an added
+	   1=valid bit added to the 53rd bit
+	   then since this is a wide register(TM)
+	   we split it into two 32 bit writes
+	 */
+#define RQ_ONCHIP_AT_PORT_SIZE  384
+#define ONCHIP_ADDR1(x)   ((u32)(((u64)x >> 12) & 0xFFFFFFFF))
+#define ONCHIP_ADDR2(x)   ((u32)((1 << 20) | ((u64)x >> 44)))
+#define PXP_ONE_ILT(x)    ((x << 10) | x)
+
+	DP(BNX2X_MSG_MCP, "starting per-function init port is %x\n", func);
+
+	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + func*4, 0);
+
+	/* Port PXP comes here */
+	/* Port PXP2 comes here */
+
+	/* Offset is
+	 * Port0  0
+	 * Port1  384 */
+	i = func * RQ_ONCHIP_AT_PORT_SIZE;
+#ifdef USE_DMAE
+	wb_write[0] = ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context));
+	wb_write[1] = ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context));
+	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
+#else
+	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8,
+		   ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context)));
+	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8 + 4,
+		   ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context)));
+#endif
+	REG_WR(bp, PXP2_REG_PSWRQ_CDU0_L2P + func*4, PXP_ONE_ILT(i));
+
+#ifdef BCM_ISCSI
+	/* Port0  1
+	 * Port1  385 */
+	i++;
+	wb_write[0] = ONCHIP_ADDR1(bp->timers_mapping);
+	wb_write[1] = ONCHIP_ADDR2(bp->timers_mapping);
+	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
+	REG_WR(bp, PXP2_REG_PSWRQ_TM0_L2P + func*4, PXP_ONE_ILT(i));
+
+	/* Port0  2
+	 * Port1  386 */
+	i++;
+	wb_write[0] = ONCHIP_ADDR1(bp->qm_mapping);
+	wb_write[1] = ONCHIP_ADDR2(bp->qm_mapping);
+	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
+	REG_WR(bp, PXP2_REG_PSWRQ_QM0_L2P + func*4, PXP_ONE_ILT(i));
+
+	/* Port0  3
+	 * Port1  387 */
+	i++;
+	wb_write[0] = ONCHIP_ADDR1(bp->t1_mapping);
+	wb_write[1] = ONCHIP_ADDR2(bp->t1_mapping);
+	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
+	REG_WR(bp, PXP2_REG_PSWRQ_SRC0_L2P + func*4, PXP_ONE_ILT(i));
+#endif
+
+	/* Port TCM comes here */
+	/* Port UCM comes here */
+	/* Port CCM comes here */
+	bnx2x_init_block(bp, func ? XCM_PORT1_START : XCM_PORT0_START,
+			     func ? XCM_PORT1_END : XCM_PORT0_END);
+
+#ifdef USE_DMAE
+	wb_write[0] = 0;
+	wb_write[1] = 0;
+#endif
+	for (i = 0; i < 32; i++) {
+		REG_WR(bp, QM_REG_BASEADDR + (func*32 + i)*4, 1024 * 4 * i);
+#ifdef USE_DMAE
+		REG_WR_DMAE(bp, QM_REG_PTRTBL + (func*32 + i)*8, wb_write, 2);
+#else
+		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8, 0);
+		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8 + 4, 0);
+#endif
+	}
+	REG_WR(bp, QM_REG_CONNNUM_0 + func*4, 1024/16 - 1);
+
+	/* Port QM comes here */
+
+#ifdef BCM_ISCSI
+	REG_WR(bp, TM_REG_LIN0_SCAN_TIME + func*4, 1024/64*20);
+	REG_WR(bp, TM_REG_LIN0_MAX_ACTIVE_CID + func*4, 31);
+
+	bnx2x_init_block(bp, func ? TIMERS_PORT1_START : TIMERS_PORT0_START,
+			     func ? TIMERS_PORT1_END : TIMERS_PORT0_END);
+#endif
+	/* Port DQ comes here */
+	/* Port BRB1 comes here */
+	bnx2x_init_block(bp, func ? PRS_PORT1_START : PRS_PORT0_START,
+			     func ? PRS_PORT1_END : PRS_PORT0_END);
+	/* Port TSDM comes here */
+	/* Port CSDM comes here */
+	/* Port USDM comes here */
+	/* Port XSDM comes here */
+	bnx2x_init_block(bp, func ? TSEM_PORT1_START : TSEM_PORT0_START,
+			     func ? TSEM_PORT1_END : TSEM_PORT0_END);
+	bnx2x_init_block(bp, func ? USEM_PORT1_START : USEM_PORT0_START,
+			     func ? USEM_PORT1_END : USEM_PORT0_END);
+	bnx2x_init_block(bp, func ? CSEM_PORT1_START : CSEM_PORT0_START,
+			     func ? CSEM_PORT1_END : CSEM_PORT0_END);
+	bnx2x_init_block(bp, func ? XSEM_PORT1_START : XSEM_PORT0_START,
+			     func ? XSEM_PORT1_END : XSEM_PORT0_END);
+	/* Port UPB comes here */
+	/* Port XSDM comes here */
+	bnx2x_init_block(bp, func ? PBF_PORT1_START : PBF_PORT0_START,
+			     func ? PBF_PORT1_END : PBF_PORT0_END);
+
+	/* configure PBF to work without PAUSE mtu 9000 */
+	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + func*4, 0);
+
+	/* update threshold */
+	REG_WR(bp, PBF_REG_P0_ARB_THRSH + func*4, (9040/16));
+	/* update init credit */
+	REG_WR(bp, PBF_REG_P0_INIT_CRD + func*4, (9040/16) + 553 - 22);
+
+	/* probe changes */
+	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 1);
+	msleep(5);
+	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 0);
+
+#ifdef BCM_ISCSI
+	/* tell the searcher where the T2 table is */
+	REG_WR(bp, SRC_REG_COUNTFREE0 + func*4, 16*1024/64);
+
+	wb_write[0] = U64_LO(bp->t2_mapping);
+	wb_write[1] = U64_HI(bp->t2_mapping);
+	REG_WR_DMAE(bp, SRC_REG_FIRSTFREE0 + func*4, wb_write, 2);
+	wb_write[0] = U64_LO((u64)bp->t2_mapping + 16*1024 - 64);
+	wb_write[1] = U64_HI((u64)bp->t2_mapping + 16*1024 - 64);
+	REG_WR_DMAE(bp, SRC_REG_LASTFREE0 + func*4, wb_write, 2);
+
+	REG_WR(bp, SRC_REG_NUMBER_HASH_BITS0 + func*4, 10);
+	/* Port SRCH comes here */
+#endif
+	/* Port CDU comes here */
+	/* Port CFC comes here */
+	bnx2x_init_block(bp, func ? HC_PORT1_START : HC_PORT0_START,
+			     func ? HC_PORT1_END : HC_PORT0_END);
+	bnx2x_init_block(bp, func ? MISC_AEU_PORT1_START :
+				    MISC_AEU_PORT0_START,
+			     func ? MISC_AEU_PORT1_END : MISC_AEU_PORT0_END);
+	/* Port PXPCS comes here */
+	/* Port EMAC0 comes here */
+	/* Port EMAC1 comes here */
+	/* Port DBU comes here */
+	/* Port DBG comes here */
+	bnx2x_init_block(bp, func ? NIG_PORT1_START : NIG_PORT0_START,
+			     func ? NIG_PORT1_END : NIG_PORT0_END);
+	REG_WR(bp, NIG_REG_XGXS_SERDES0_MODE_SEL + func*4, 1);
+	/* Port MCP comes here */
+	/* Port DMAE comes here */
+
+	bnx2x_link_reset(bp);
+
+	/* Reset pciex errors for debug */
+	REG_WR(bp, 0x2114, 0xffffffff);
+	REG_WR(bp, 0x2120, 0xffffffff);
+	REG_WR(bp, 0x2814, 0xffffffff);
+
+	/* !!! move to init_values.h */
+	REG_WR(bp, XSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+	REG_WR(bp, USDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+	REG_WR(bp, CSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+	REG_WR(bp, TSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+
+	REG_WR(bp, DBG_REG_PCI_REQ_CREDIT, 0x1);
+	REG_WR(bp, TM_REG_PCIARB_CRDCNT_VAL, 0x1);
+	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x264);
+	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x0);
+
+	bnx2x_gunzip_end(bp);
+
+	if (!nomcp) {
+		port = bp->port;
+
+		bp->fw_drv_pulse_wr_seq =
+				(SHMEM_RD(bp, drv_fw_mb[port].drv_pulse_mb) &
+				 DRV_PULSE_SEQ_MASK);
+		bp->fw_mb = SHMEM_RD(bp, drv_fw_mb[port].fw_mb_param);
+		DP(BNX2X_MSG_MCP, "drv_pulse 0x%x  fw_mb 0x%x\n",
+		   bp->fw_drv_pulse_wr_seq, bp->fw_mb);
+	} else {
+		bp->fw_mb = 0;
+	}
+
+	return 0;
+}
+
+
+/* send the MCP a request, block untill there is a reply */
+static u32 bnx2x_fw_command(struct bnx2x *bp, u32 command)
+{
+	u32 rc = 0;
+	u32 seq = ++bp->fw_seq;
+	int port = bp->port;
+
+	SHMEM_WR(bp, drv_fw_mb[port].drv_mb_header, command|seq);
+	DP(BNX2X_MSG_MCP, "wrote command (%x) to FW MB\n", command|seq);
+
+	/* let the FW do it's magic ... */
+	msleep(100); /* TBD */
+
+	if (CHIP_REV_IS_SLOW(bp))
+		msleep(900);
+
+	rc = SHMEM_RD(bp, drv_fw_mb[port].fw_mb_header);
+
+	DP(BNX2X_MSG_MCP, "read (%x) seq is (%x) from FW MB\n", rc, seq);
+
+	/* is this a reply to our command? */
+	if (seq == (rc & FW_MSG_SEQ_NUMBER_MASK)) {
+		rc &= FW_MSG_CODE_MASK;
+	} else {
+		/* FW BUG! */
+		BNX2X_ERR("FW failed to respond!\n");
+		bnx2x_fw_dump(bp);
+		rc = 0;
+	}
+	return rc;
+}
+
+static void bnx2x_free_mem(struct bnx2x *bp)
+{
+
+#define BNX2X_PCI_FREE(x, y, size) \
+	do { \
+		if (x) { \
+			pci_free_consistent(bp->pdev, size, x, y); \
+			x = NULL; \
+			y = 0; \
+		} \
+	} while (0)
+
+#define BNX2X_FREE(x) \
+	do { \
+		if (x) { \
+			vfree(x); \
+			x = NULL; \
+		} \
+	} while (0)
+
+	int i;
+
+	/* fastpath */
+	for_each_queue(bp, i) {
+
+		/* Status blocks */
+		BNX2X_PCI_FREE(bnx2x_fp(bp, i, status_blk),
+			       bnx2x_fp(bp, i, status_blk_mapping),
+			       sizeof(struct host_status_block) +
+			       sizeof(struct eth_tx_db_data));
+
+		/* fast path rings: tx_buf tx_desc rx_buf rx_desc rx_comp */
+		BNX2X_FREE(bnx2x_fp(bp, i, tx_buf_ring));
+		BNX2X_PCI_FREE(bnx2x_fp(bp, i, tx_desc_ring),
+			       bnx2x_fp(bp, i, tx_desc_mapping),
+			       sizeof(struct eth_tx_bd) * NUM_TX_BD);
+
+		BNX2X_FREE(bnx2x_fp(bp, i, rx_buf_ring));
+		BNX2X_PCI_FREE(bnx2x_fp(bp, i, rx_desc_ring),
+			       bnx2x_fp(bp, i, rx_desc_mapping),
+			       sizeof(struct eth_rx_bd) * NUM_RX_BD);
+
+		BNX2X_PCI_FREE(bnx2x_fp(bp, i, rx_comp_ring),
+			       bnx2x_fp(bp, i, rx_comp_mapping),
+			       sizeof(struct eth_fast_path_rx_cqe) *
+			       NUM_RCQ_BD);
+	}
+
+	BNX2X_FREE(bp->fp);
+
+	/* end of fastpath */
+
+	BNX2X_PCI_FREE(bp->def_status_blk, bp->def_status_blk_mapping,
+		       (sizeof(struct host_def_status_block)));
+
+	BNX2X_PCI_FREE(bp->slowpath, bp->slowpath_mapping,
+		       (sizeof(struct bnx2x_slowpath)));
+
+#ifdef BCM_ISCSI
+	BNX2X_PCI_FREE(bp->t1, bp->t1_mapping, 64*1024);
+	BNX2X_PCI_FREE(bp->t2, bp->t2_mapping, 16*1024);
+	BNX2X_PCI_FREE(bp->timers, bp->timers_mapping, 8*1024);
+	BNX2X_PCI_FREE(bp->qm, bp->qm_mapping, 128*1024);
+#endif
+	BNX2X_PCI_FREE(bp->spq, bp->spq_mapping, PAGE_SIZE);
+
+#undef BNX2X_PCI_FREE
+#undef BNX2X_KFREE
+}
+
+static int bnx2x_alloc_mem(struct bnx2x *bp)
+{
+
+#define BNX2X_PCI_ALLOC(x, y, size) \
+	do { \
+		x = pci_alloc_consistent(bp->pdev, size, y); \
+		if (x == NULL) \
+			goto alloc_mem_err; \
+		memset(x, 0, size); \
+	} while (0)
+
+#define BNX2X_ALLOC(x, size) \
+	do { \
+		x = vmalloc(size); \
+		if (x == NULL) \
+			goto alloc_mem_err; \
+		memset(x, 0, size); \
+	} while (0)
+
+	int i;
+
+	/* fastpath */
+	BNX2X_ALLOC(bp->fp, sizeof(struct bnx2x_fastpath) * bp->num_queues);
+
+	for_each_queue(bp, i) {
+		bnx2x_fp(bp, i, bp) = bp;
+
+		/* Status blocks */
+		BNX2X_PCI_ALLOC(bnx2x_fp(bp, i, status_blk),
+				&bnx2x_fp(bp, i, status_blk_mapping),
+				sizeof(struct host_status_block) +
+				sizeof(struct eth_tx_db_data));
+
+		bnx2x_fp(bp, i, hw_tx_prods) =
+				(void *)(bnx2x_fp(bp, i, status_blk) + 1);
+
+		bnx2x_fp(bp, i, tx_prods_mapping) =
+				bnx2x_fp(bp, i, status_blk_mapping) +
+				sizeof(struct host_status_block);
+
+		/* fast path rings: tx_buf tx_desc rx_buf rx_desc rx_comp */
+		BNX2X_ALLOC(bnx2x_fp(bp, i, tx_buf_ring),
+				sizeof(struct sw_tx_bd) * NUM_TX_BD);
+		BNX2X_PCI_ALLOC(bnx2x_fp(bp, i, tx_desc_ring),
+				&bnx2x_fp(bp, i, tx_desc_mapping),
+				sizeof(struct eth_tx_bd) * NUM_TX_BD);
+
+		BNX2X_ALLOC(bnx2x_fp(bp, i, rx_buf_ring),
+				sizeof(struct sw_rx_bd) * NUM_RX_BD);
+		BNX2X_PCI_ALLOC(bnx2x_fp(bp, i, rx_desc_ring),
+				&bnx2x_fp(bp, i, rx_desc_mapping),
+				sizeof(struct eth_rx_bd) * NUM_RX_BD);
+
+		BNX2X_PCI_ALLOC(bnx2x_fp(bp, i, rx_comp_ring),
+				&bnx2x_fp(bp, i, rx_comp_mapping),
+				sizeof(struct eth_fast_path_rx_cqe) *
+				NUM_RCQ_BD);
+
+	}
+	/* end of fastpath */
+
+	BNX2X_PCI_ALLOC(bp->def_status_blk, &bp->def_status_blk_mapping,
+			sizeof(struct host_def_status_block));
+
+	BNX2X_PCI_ALLOC(bp->slowpath, &bp->slowpath_mapping,
+			sizeof(struct bnx2x_slowpath));
+
+#ifdef BCM_ISCSI
+	BNX2X_PCI_ALLOC(bp->t1, &bp->t1_mapping, 64*1024);
+
+	/* Initialize T1 */
+	for (i = 0; i < 64*1024; i += 64) {
+		*(u64 *)((char *)bp->t1 + i + 56) = 0x0UL;
+		*(u64 *)((char *)bp->t1 + i + 3) = 0x0UL;
+	}
+
+	/* allocate searcher T2 table
+	   we allocate 1/4 of alloc num for T2
+	  (which is not entered into the ILT) */
+	BNX2X_PCI_ALLOC(bp->t2, &bp->t2_mapping, 16*1024);
+
+	/* Initialize T2 */
+	for (i = 0; i < 16*1024; i += 64)
+		* (u64 *)((char *)bp->t2 + i + 56) = bp->t2_mapping + i + 64;
+
+	/* now sixup the last line in the block to point to the next block */
+	*(u64 *)((char *)bp->t2 + 1024*16-8) = bp->t2_mapping;
+
+	/* Timer block array (MAX_CONN*8) phys uncached for now 1024 conns */
+	BNX2X_PCI_ALLOC(bp->timers, &bp->timers_mapping, 8*1024);
+
+	/* QM queues (128*MAX_CONN) */
+	BNX2X_PCI_ALLOC(bp->qm, &bp->qm_mapping, 128*1024);
+#endif
+
+	/* Slow path ring */
+	BNX2X_PCI_ALLOC(bp->spq, &bp->spq_mapping, BCM_PAGE_SIZE);
+
+	return 0;
+
+alloc_mem_err:
+	bnx2x_free_mem(bp);
+	return -ENOMEM;
+
+#undef BNX2X_PCI_ALLOC
+#undef BNX2X_ALLOC
+}
+
+static void bnx2x_free_tx_skbs(struct bnx2x *bp)
+{
+	int i;
+
+	for_each_queue(bp, i) {
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+
+		u16 bd_cons = fp->tx_bd_cons;
+		u16 sw_prod = fp->tx_pkt_prod;
+		u16 sw_cons = fp->tx_pkt_cons;
+
+		BUG_TRAP(fp->tx_buf_ring != NULL);
+
+		while (sw_cons != sw_prod) {
+			bd_cons = bnx2x_free_tx_pkt(bp, fp, TX_BD(sw_cons));
+			sw_cons++;
+		}
+	}
+}
+
+static void bnx2x_free_rx_skbs(struct bnx2x *bp)
+{
+	int i, j;
+
+	for_each_queue(bp, j) {
+		struct bnx2x_fastpath *fp = &bp->fp[j];
+
+		BUG_TRAP(fp->rx_buf_ring != NULL);
+
+		for (i = 0; i < NUM_RX_BD; i++) {
+			struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[i];
+			struct sk_buff *skb = rx_buf->skb;
+
+			if (skb == NULL)
+				continue;
+
+			pci_unmap_single(bp->pdev,
+					 pci_unmap_addr(rx_buf, mapping),
+					 bp->rx_buf_use_size,
+					 PCI_DMA_FROMDEVICE);
+
+			rx_buf->skb = NULL;
+			dev_kfree_skb(skb);
+		}
+	}
+}
+
+static void bnx2x_free_skbs(struct bnx2x *bp)
+{
+	bnx2x_free_tx_skbs(bp);
+	bnx2x_free_rx_skbs(bp);
+}
+
+static void bnx2x_free_msix_irqs(struct bnx2x *bp)
+{
+	int i;
+
+	free_irq(bp->msix_table[0].vector, bp->dev);
+	DP(NETIF_MSG_IFDOWN, "rleased sp irq (%d)\n",
+	   bp->msix_table[0].vector);
+
+	for_each_queue(bp, i) {
+		DP(NETIF_MSG_IFDOWN, "about to rlease fp #%d->%d irq  "
+		   "state(%x)\n", i, bp->msix_table[i + 1].vector,
+		   bnx2x_fp(bp, i, state));
+
+		if (bnx2x_fp(bp, i, state) != BNX2X_FP_STATE_CLOSED) {
+
+			free_irq(bp->msix_table[i + 1].vector, &bp->fp[i]);
+			bnx2x_fp(bp, i, state) = BNX2X_FP_STATE_CLOSED;
+
+		} else
+			DP(NETIF_MSG_IFDOWN, "irq not freed\n");
+
+	}
+
+}
+
+static void bnx2x_free_irq(struct bnx2x *bp)
+{
+
+	if (bp->flags & USING_MSIX_FLAG) {
+
+		bnx2x_free_msix_irqs(bp);
+		pci_disable_msix(bp->pdev);
+
+		bp->flags &= ~USING_MSIX_FLAG;
+
+	} else
+		free_irq(bp->pdev->irq, bp->dev);
+}
+
+static int bnx2x_enable_msix(struct bnx2x *bp)
+{
+
+	int i;
+
+	bp->msix_table[0].entry = 0;
+	for_each_queue(bp, i)
+		bp->msix_table[i + 1].entry = i + 1;
+
+	if (pci_enable_msix(bp->pdev, &bp->msix_table[0],
+				     bp->num_queues + 1)){
+		BNX2X_ERR("failed to enable msix\n");
+		return -1;
+
+	}
+
+	bp->flags |= USING_MSIX_FLAG;
+
+	return 0;
+
+}
+
+
+static int bnx2x_req_msix_irqs(struct bnx2x *bp)
+{
+
+
+	int i, rc;
+
+	DP(NETIF_MSG_IFUP, "about to request sp irq\n");
+
+	rc = request_irq(bp->msix_table[0].vector, bnx2x_msix_sp_int, 0,
+			 bp->dev->name, bp->dev);
+
+	if (rc) {
+		BNX2X_ERR("request sp irq failed\n");
+		return -EBUSY;
+	}
+
+	for_each_queue(bp, i) {
+		rc = request_irq(bp->msix_table[i + 1].vector,
+				 bnx2x_msix_fp_int, 0,
+				 bp->dev->name, &bp->fp[i]);
+
+		if (rc) {
+			BNX2X_ERR("request fp #%d irq failed\n", i);
+			bnx2x_free_msix_irqs(bp);
+			return -EBUSY;
+		}
+
+		bnx2x_fp(bp, i, state) = BNX2X_FP_STATE_IRQ;
+
+	}
+
+	return 0;
+
+}
+
+static int bnx2x_req_irq(struct bnx2x *bp)
+{
+
+	int rc = request_irq(bp->pdev->irq, bnx2x_interrupt,
+			     IRQF_SHARED, bp->dev->name, bp->dev);
+	if (!rc)
+		bnx2x_fp(bp, 0, state) = BNX2X_FP_STATE_IRQ;
+
+	return rc;
+
+}
+
+/*
+ * Init service functions
+ */
+
+static void bnx2x_set_mac_addr(struct bnx2x *bp)
+{
+	struct mac_configuration_cmd *config = bnx2x_sp(bp, mac_config);
+
+	/* CAM allocation
+	 * unicasts 0-31:port0 32-63:port1
+	 * multicast 64-127:port0 128-191:port1
+	 */
+	config->hdr.length_6b = 2;
+	config->hdr.offset = bp->port ? 31 : 0;
+	config->hdr.reserved0 = 0;
+	config->hdr.reserved1 = 0;
+
+	/* primary MAC */
+	config->config_table[0].cam_entry.msb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[0]);
+	config->config_table[0].cam_entry.middle_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[2]);
+	config->config_table[0].cam_entry.lsb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[4]);
+	config->config_table[0].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[0].target_table_entry.flags = 0;
+	config->config_table[0].target_table_entry.client_id = 0;
+	config->config_table[0].target_table_entry.vlan_id = 0;
+
+	DP(NETIF_MSG_IFUP, "setting MAC (%04x:%04x:%04x)\n",
+	   config->config_table[0].cam_entry.msb_mac_addr,
+	   config->config_table[0].cam_entry.middle_mac_addr,
+	   config->config_table[0].cam_entry.lsb_mac_addr);
+
+	/* broadcast */
+	config->config_table[1].cam_entry.msb_mac_addr = 0xffff;
+	config->config_table[1].cam_entry.middle_mac_addr = 0xffff;
+	config->config_table[1].cam_entry.lsb_mac_addr = 0xffff;
+	config->config_table[1].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[1].target_table_entry.flags =
+				TSTORM_CAM_TARGET_TABLE_ENTRY_BROADCAST;
+	config->config_table[1].target_table_entry.client_id = 0;
+	config->config_table[1].target_table_entry.vlan_id = 0;
+
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+		      U64_HI(bnx2x_sp_mapping(bp, mac_config)),
+		      U64_LO(bnx2x_sp_mapping(bp, mac_config)), 0);
+}
+
+static int bnx2x_wait_ramrod(struct bnx2x *bp, int state, int idx,
+			     int *state_p, int poll)
+{
+	/* can take a while if any port is running */
+	int timeout = 500;
+
+	/* DP("waiting for state to become %d on IDX [%d]\n",
+	state, sb_idx); */
+
+	might_sleep();
+
+	while (timeout) {
+
+		if (poll) {
+			bnx2x_rx_int(bp->fp, 10);
+			/* If index is different from 0
+			 * The reply for some commands will
+			 * be on the none default queue
+			 */
+			if (idx)
+				bnx2x_rx_int(&bp->fp[idx], 10);
+		}
+
+		mb(); /* state is changed by bnx2x_sp_event()*/
+
+		if (*state_p != state)
+			return 0;
+
+		timeout--;
+		msleep(1);
+
+	}
+
+
+	/* timeout! */
+	BNX2X_ERR("timeout waiting for ramrod %d on %d\n", state, idx);
+	return -EBUSY;
+
+}
+
+static int bnx2x_setup_leading(struct bnx2x *bp)
+{
+
+	/* reset IGU staae */
+	bnx2x_ack_sb(bp, DEF_SB_ID, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+
+	/* SETUP ramrod */
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_PORT_SETUP, 0, 0, 0, 0);
+
+	return bnx2x_wait_ramrod(bp, BNX2X_STATE_OPEN, 0, &(bp->state), 0);
+
+}
+
+static int bnx2x_setup_multi(struct bnx2x *bp, int index)
+{
+
+	/* reset IGU state */
+	bnx2x_ack_sb(bp, index, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+
+	bp->fp[index].state = BNX2X_FP_STATE_OPENING;
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_CLIENT_SETUP, index, 0, index, 0);
+
+	/* Wait for completion */
+	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_OPEN, index,
+				 &(bp->fp[index].state), 1);
+
+}
+
+
+static int bnx2x_poll(struct napi_struct *napi, int budget);
+static void bnx2x_set_rx_mode(struct net_device *dev);
+
+static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
+{
+	int rc;
+	int i = 0;
+
+	bp->state = BNX2X_STATE_OPENING_WAIT4_LOAD;
+
+	/* Send LOAD_REQUEST command to MCP.
+	   Returns the type of LOAD command: if it is the
+	   first port to be initialized common blocks should be
+	   initialized, otherwise - not.
+	*/
+	if (!nomcp) {
+		rc = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_REQ);
+		if (rc == FW_MSG_CODE_DRV_LOAD_REFUSED) {
+			return -EBUSY; /* other port in diagnostic mode */
+		}
+	} else {
+		rc = FW_MSG_CODE_DRV_LOAD_COMMON;
+	}
+
+	DP(NETIF_MSG_IFUP, "set number of queues to %d\n", bp->num_queues);
+
+	/* if we can't use msix we only need one fp,
+	 * so try to enable msix with the requested number of fp's
+	 * and fallback to inta with one fp
+	 */
+	if (req_irq) {
+
+		if (use_inta) {
+			bp->num_queues = 1;
+		} else {
+			if (use_multi > 1 && use_multi <= 16)
+				/* user requested number */
+				bp->num_queues = use_multi;
+			else if (use_multi == 1)
+				bp->num_queues = num_online_cpus();
+			else
+				bp->num_queues = 1;
+
+			if (bnx2x_enable_msix(bp)) {
+				/* faild to enable msix */
+				bp->num_queues = 1;
+				if (use_multi)
+					BNX2X_ERR("Muti requested but failed"
+						  " to enable MSI-X\n");
+			}
+		}
+	}
+
+	if (bnx2x_alloc_mem(bp))
+		return -ENOMEM;
+
+	if (req_irq) {
+		if (bp->flags & USING_MSIX_FLAG) {
+			if (bnx2x_req_msix_irqs(bp)) {
+				pci_disable_msix(bp->pdev);
+				goto out_error;
+			}
+
+		} else {
+			if (bnx2x_req_irq(bp)) {
+				BNX2X_ERR("IRQ request failed, aborting\n");
+				goto out_error;
+			}
+		}
+	}
+
+	for_each_queue(bp, i)
+		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
+			       bnx2x_poll, 128);
+
+
+	/* Initialize HW */
+	if (bnx2x_function_init(bp, (rc == FW_MSG_CODE_DRV_LOAD_COMMON))) {
+		BNX2X_ERR("HW init failed, aborting\n");
+		goto out_error;
+	}
+
+
+	atomic_set(&bp->intr_sem, 0);
+
+	/* Reenable SP tasklet */
+	/*if (bp->sp_task_en) { 	       */
+	/*        tasklet_enable(&bp->sp_task);*/
+	/*} else {      		       */
+	/*        bp->sp_task_en = 1;          */
+	/*}     			       */
+
+	/* Setup NIC internals and enable interrupts */
+	bnx2x_nic_init(bp);
+
+	/* Send LOAD_DONE command to MCP */
+	if (!nomcp) {
+		rc = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE);
+		DP(NETIF_MSG_IFUP, "rc = 0x%x\n", rc);
+		if (!rc) {
+			BNX2X_ERR("MCP response failure, unloading\n");
+			goto int_disable;
+		}
+	}
+
+	bp->state = BNX2X_STATE_OPENING_WAIT4_PORT;
+
+	/* Enable Rx interrupt handling before sending the ramrod
+	   as it's completed on Rx FP queue */
+	for_each_queue(bp, i)
+		napi_enable(&bnx2x_fp(bp, i, napi));
+
+	if (bnx2x_setup_leading(bp))
+		goto stop_netif;
+
+	for_each_nondefault_queue(bp, i)
+		if (bnx2x_setup_multi(bp, i))
+			goto stop_netif;
+
+	bnx2x_set_mac_addr(bp);
+
+	bnx2x_phy_init(bp);
+
+	/* Start fast path */
+	if (req_irq) { /* IRQ is only requested from bnx2x_open */
+		netif_start_queue(bp->dev);
+		if (bp->flags & USING_MSIX_FLAG)
+			printk(KERN_INFO PFX "%s: using MSI-X\n",
+			       bp->dev->name);
+
+	/* Otherwise Tx queue should be only reenabled */
+	} else if (netif_running(bp->dev)) {
+		netif_wake_queue(bp->dev);
+		bnx2x_set_rx_mode(bp->dev);
+	}
+
+	/* start the timer */
+	mod_timer(&bp->timer, jiffies + bp->current_interval);
+
+	return 0;
+
+stop_netif:
+	for_each_queue(bp, i)
+		napi_disable(&bnx2x_fp(bp, i, napi));
+
+int_disable:
+	bnx2x_disable_int_sync(bp);
+
+	bnx2x_free_skbs(bp);
+	bnx2x_free_irq(bp);
+
+out_error:
+	bnx2x_free_mem(bp);
+
+	/* TBD we really need to reset the chip
+	   if we want to recover from this */
+	return rc;
+}
+
+static void bnx2x_netif_stop(struct bnx2x *bp)
+{
+	int i;
+
+	bp->rx_mode = BNX2X_RX_MODE_NONE;
+	bnx2x_set_storm_rx_mode(bp);
+
+	bnx2x_disable_int_sync(bp);
+	bnx2x_link_reset(bp);
+
+	for_each_queue(bp, i)
+		napi_disable(&bnx2x_fp(bp, i, napi));
+
+	if (netif_running(bp->dev)) {
+		netif_tx_disable(bp->dev);
+		bp->dev->trans_start = jiffies; /* prevent tx timeout */
+	}
+}
+
+static void bnx2x_reset_chip(struct bnx2x *bp, u32 reset_code)
+{
+	int port = bp->port;
+#ifdef USE_DMAE
+	u32 wb_write[2];
+#endif
+	int base, i;
+
+	DP(NETIF_MSG_IFDOWN, "reset called with code %x\n", reset_code);
+
+	/* Do not rcv packets to BRB */
+	REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK + port*4, 0x0);
+	/* Do not direct rcv packets that are not for MCP to the BRB */
+	REG_WR(bp, (port ? NIG_REG_LLH1_BRB1_NOT_MCP :
+			   NIG_REG_LLH0_BRB1_NOT_MCP), 0x0);
+
+	/* Configure IGU and AEU */
+	REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
+	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4, 0);
+
+	/* TODO: Close Doorbell port? */
+
+	/* Clear ILT */
+#ifdef USE_DMAE
+	wb_write[0] = 0;
+	wb_write[1] = 0;
+#endif
+	base = port * RQ_ONCHIP_AT_PORT_SIZE;
+	for (i = base; i < base + RQ_ONCHIP_AT_PORT_SIZE; i++) {
+#ifdef USE_DMAE
+		REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
+#else
+		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT, 0);
+		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + 4, 0);
+#endif
+	}
+
+	if (reset_code == FW_MSG_CODE_DRV_UNLOAD_COMMON) {
+		/* reset_common */
+		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+		       0xd3ffff7f);
+		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+		       0x1403);
+	}
+}
+
+static int bnx2x_stop_multi(struct bnx2x *bp, int index)
+{
+
+	int rc;
+
+	/* halt the connnection */
+	bp->fp[index].state = BNX2X_FP_STATE_HALTING;
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, index, 0, 0, 0);
+
+
+	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, index,
+				       &(bp->fp[index].state), 1);
+	if (rc) /* timout */
+		return rc;
+
+	/* delete cfc entry */
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_CFC_DEL, index, 0, 0, 1);
+
+	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_DELETED, index,
+				 &(bp->fp[index].state), 1);
+
+}
+
+
+static void bnx2x_stop_leading(struct bnx2x *bp)
+{
+
+	/* if the other port is hadling traffic,
+	   this can take a lot of time */
+	int timeout = 500;
+
+	might_sleep();
+
+	/* Send HALT ramrod */
+	bp->fp[0].state = BNX2X_FP_STATE_HALTING;
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, 0, 0, 0, 0);
+
+	if (bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, 0,
+			       &(bp->fp[0].state), 1))
+		return;
+
+	bp->dsb_sp_prod_idx = *bp->dsb_sp_prod;
+
+	/* Send CFC_DELETE ramrod */
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_PORT_DEL, 0, 0, 0, 1);
+
+	/*
+	   Wait for completion.
+	   we are going to reset the chip anyway
+	   so there is not much to do if this times out
+	 */
+	while (bp->dsb_sp_prod_idx == *bp->dsb_sp_prod && timeout) {
+			timeout--;
+			msleep(1);
+	}
+
+}
+
+static int bnx2x_nic_unload(struct bnx2x *bp, int fre_irq)
+{
+	u32 reset_code = 0;
+	int rc;
+	int i;
+
+	bp->state = BNX2X_STATE_CLOSING_WAIT4_HALT;
+
+	/* Calling flush_scheduled_work() may deadlock because
+	 * linkwatch_event() may be on the workqueue and it will try to get
+	 * the rtnl_lock which we are holding.
+	 */
+
+	while (bp->in_reset_task)
+		msleep(1);
+
+	/* Delete the timer: do it before disabling interrupts, as it
+	   may be stil STAT_QUERY ramrod pending after stopping the timer */
+	del_timer_sync(&bp->timer);
+
+	/* Wait until stat ramrod returns and all SP tasks complete */
+	while (bp->stat_pending && (bp->spq_left != MAX_SPQ_PENDING))
+		msleep(1);
+
+	/* Stop fast path, disable MAC, disable interrupts, disable napi */
+	bnx2x_netif_stop(bp);
+
+	if (bp->flags & NO_WOL_FLAG)
+		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_MCP;
+	else if (bp->wol) {
+		u32 emac_base = bp->port ? GRCBASE_EMAC0 : GRCBASE_EMAC1;
+		u8 *mac_addr = bp->dev->dev_addr;
+		u32 val = (EMAC_MODE_MPKT | EMAC_MODE_MPKT_RCVD |
+			   EMAC_MODE_ACPI_RCVD);
+
+		EMAC_WR(EMAC_REG_EMAC_MODE, val);
+
+		val = (mac_addr[0] << 8) | mac_addr[1];
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH, val);
+
+		val = (mac_addr[2] << 24) | (mac_addr[3] << 16) |
+		      (mac_addr[4] << 8) | mac_addr[5];
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + 4, val);
+
+		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_EN;
+	} else
+		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
+
+	for_each_nondefault_queue(bp, i)
+		if (bnx2x_stop_multi(bp, i))
+			goto error;
+
+
+	bnx2x_stop_leading(bp);
+
+error:
+	if (!nomcp)
+		rc = bnx2x_fw_command(bp, reset_code);
+	else
+		rc = FW_MSG_CODE_DRV_UNLOAD_COMMON;
+
+	/* Release IRQs */
+	if (fre_irq)
+		bnx2x_free_irq(bp);
+
+	/* Reset the chip */
+	bnx2x_reset_chip(bp, rc);
+
+	/* Report UNLOAD_DONE to MCP */
+	if (!nomcp)
+		bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE);
+
+	/* Free SKBs and driver internals */
+	bnx2x_free_skbs(bp);
+	bnx2x_free_mem(bp);
+
+	bp->state = BNX2X_STATE_CLOSED;
+	/* Set link down */
+	bp->link_up = 0;
+	netif_carrier_off(bp->dev);
+
+	return 0;
+}
+
+/* end of nic load/unload */
+
+/* ethtool_ops */
+
+/*
+ * Init service functions
+ */
+
+static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
+{
+	int port = bp->port;
+	u32 ext_phy_type;
+
+	bp->phy_flags = 0;
+
+	switch (switch_cfg) {
+	case SWITCH_CFG_1G:
+		BNX2X_DEV_INFO("switch_cfg 0x%x (1G)\n", switch_cfg);
+
+		ext_phy_type = SERDES_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_DIRECT:
+			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
+				       ext_phy_type);
+
+			bp->supported |= (SUPPORTED_10baseT_Half |
+					  SUPPORTED_10baseT_Full |
+					  SUPPORTED_100baseT_Half |
+					  SUPPORTED_100baseT_Full |
+					  SUPPORTED_1000baseT_Full |
+					  SUPPORTED_2500baseT_Full |
+					  SUPPORTED_TP | SUPPORTED_FIBRE |
+					  SUPPORTED_Autoneg |
+					  SUPPORTED_Pause |
+					  SUPPORTED_Asym_Pause);
+			break;
+
+		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
+			BNX2X_DEV_INFO("ext_phy_type 0x%x (5482)\n",
+				       ext_phy_type);
+
+			bp->phy_flags |= PHY_SGMII_FLAG;
+
+			bp->supported |= (/* SUPPORTED_10baseT_Half |
+					     SUPPORTED_10baseT_Full |
+					     SUPPORTED_100baseT_Half |
+					     SUPPORTED_100baseT_Full |*/
+					  SUPPORTED_1000baseT_Full |
+					  SUPPORTED_TP | SUPPORTED_FIBRE |
+					  SUPPORTED_Autoneg |
+					  SUPPORTED_Pause |
+					  SUPPORTED_Asym_Pause);
+			break;
+
+		default:
+			BNX2X_ERR("NVRAM config error. "
+				  "BAD SerDes ext_phy_config 0x%x\n",
+				  bp->ext_phy_config);
+			return;
+		}
+
+		bp->phy_addr = REG_RD(bp, NIG_REG_SERDES0_CTRL_PHY_ADDR +
+				      port*0x10);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+		break;
+
+	case SWITCH_CFG_10G:
+		BNX2X_DEV_INFO("switch_cfg 0x%x (10G)\n", switch_cfg);
+
+		bp->phy_flags |= PHY_XGXS_FLAG;
+
+		ext_phy_type = XGXS_EXT_PHY_TYPE(bp);
+		switch (ext_phy_type) {
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT:
+			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
+				       ext_phy_type);
+
+			bp->supported |= (SUPPORTED_10baseT_Half |
+					  SUPPORTED_10baseT_Full |
+					  SUPPORTED_100baseT_Half |
+					  SUPPORTED_100baseT_Full |
+					  SUPPORTED_1000baseT_Full |
+					  SUPPORTED_2500baseT_Full |
+					  SUPPORTED_10000baseT_Full |
+					  SUPPORTED_TP | SUPPORTED_FIBRE |
+					  SUPPORTED_Autoneg |
+					  SUPPORTED_Pause |
+					  SUPPORTED_Asym_Pause);
+			break;
+
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
+		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
+			BNX2X_DEV_INFO("ext_phy_type 0x%x (8705/6)\n",
+				       ext_phy_type);
+
+			bp->supported |= (SUPPORTED_10000baseT_Full |
+					  SUPPORTED_FIBRE |
+					  SUPPORTED_Pause |
+					  SUPPORTED_Asym_Pause);
+			break;
+
+		default:
+			BNX2X_ERR("NVRAM config error. "
+				  "BAD XGXS ext_phy_config 0x%x\n",
+				  bp->ext_phy_config);
+			return;
+		}
+
+		bp->phy_addr = REG_RD(bp, NIG_REG_XGXS0_CTRL_PHY_ADDR +
+				      port*0x18);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+
+		bp->ser_lane = ((bp->lane_config &
+				 PORT_HW_CFG_LANE_SWAP_CFG_MASTER_MASK) >>
+				PORT_HW_CFG_LANE_SWAP_CFG_MASTER_SHIFT);
+		bp->rx_lane_swap = ((bp->lane_config &
+				     PORT_HW_CFG_LANE_SWAP_CFG_RX_MASK) >>
+				    PORT_HW_CFG_LANE_SWAP_CFG_RX_SHIFT);
+		bp->tx_lane_swap = ((bp->lane_config &
+				     PORT_HW_CFG_LANE_SWAP_CFG_TX_MASK) >>
+				    PORT_HW_CFG_LANE_SWAP_CFG_TX_SHIFT);
+		BNX2X_DEV_INFO("rx_lane_swap 0x%x  tx_lane_swap 0x%x\n",
+			       bp->rx_lane_swap, bp->tx_lane_swap);
+		break;
+
+	default:
+		BNX2X_ERR("BAD switch_cfg link_config 0x%x\n",
+			  bp->link_config);
+		return;
+	}
+
+	/* mask what we support according to speed_cap_mask */
+	if (!(bp->speed_cap_mask &
+	      PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF))
+		bp->supported &= ~SUPPORTED_10baseT_Half;
+
+	if (!(bp->speed_cap_mask &
+	      PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL))
+		bp->supported &= ~SUPPORTED_10baseT_Full;
+
+	if (!(bp->speed_cap_mask &
+	      PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF))
+		bp->supported &= ~SUPPORTED_100baseT_Half;
+
+	if (!(bp->speed_cap_mask &
+	      PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL))
+		bp->supported &= ~SUPPORTED_100baseT_Full;
+
+	if (!(bp->speed_cap_mask & PORT_HW_CFG_SPEED_CAPABILITY_D0_1G))
+		bp->supported &= ~(SUPPORTED_1000baseT_Half |
+				   SUPPORTED_1000baseT_Full);
+
+	if (!(bp->speed_cap_mask & PORT_HW_CFG_SPEED_CAPABILITY_D0_2_5G))
+		bp->supported &= ~SUPPORTED_2500baseT_Full;
+
+	if (!(bp->speed_cap_mask & PORT_HW_CFG_SPEED_CAPABILITY_D0_10G))
+		bp->supported &= ~SUPPORTED_10000baseT_Full;
+
+	BNX2X_DEV_INFO("supported 0x%x\n", bp->supported);
+}
+
+static void bnx2x_link_settings_requested(struct bnx2x *bp)
+{
+	bp->req_autoneg = 0;
+	bp->req_duplex = DUPLEX_FULL;
+
+	switch (bp->link_config & PORT_FEATURE_LINK_SPEED_MASK) {
+	case PORT_FEATURE_LINK_SPEED_AUTO:
+		if (bp->supported & SUPPORTED_Autoneg) {
+			bp->req_autoneg |= AUTONEG_SPEED;
+			bp->req_line_speed = 0;
+			bp->advertising = bp->supported;
+		} else {
+			u32 ext_phy_type;
+
+			ext_phy_type = XGXS_EXT_PHY_TYPE(bp);
+			if ((ext_phy_type ==
+				PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705) ||
+			    (ext_phy_type ==
+				PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706)) {
+				/* force 10G, no AN */
+				bp->req_line_speed = SPEED_10000;
+				bp->advertising =
+						(ADVERTISED_10000baseT_Full |
+						 ADVERTISED_FIBRE);
+				break;
+			}
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  Autoneg not supported\n",
+				  bp->link_config);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_10M_FULL:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL) {
+			bp->req_line_speed = SPEED_10;
+			bp->advertising = (ADVERTISED_10baseT_Full |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_10M_HALF:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF) {
+			bp->req_line_speed = SPEED_10;
+			bp->req_duplex = DUPLEX_HALF;
+			bp->advertising = (ADVERTISED_10baseT_Half |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_100M_FULL:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL) {
+			bp->req_line_speed = SPEED_100;
+			bp->advertising = (ADVERTISED_100baseT_Full |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_100M_HALF:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF) {
+			bp->req_line_speed = SPEED_100;
+			bp->req_duplex = DUPLEX_HALF;
+			bp->advertising = (ADVERTISED_100baseT_Half |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_1G:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_1G) {
+			bp->req_line_speed = SPEED_1000;
+			bp->advertising = (ADVERTISED_1000baseT_Full |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_2_5G:
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_2_5G) {
+			bp->req_line_speed = SPEED_2500;
+			bp->advertising = (ADVERTISED_2500baseT_Full |
+					   ADVERTISED_TP);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	case PORT_FEATURE_LINK_SPEED_10G_CX4:
+	case PORT_FEATURE_LINK_SPEED_10G_KX4:
+	case PORT_FEATURE_LINK_SPEED_10G_KR:
+		if (!(bp->phy_flags & PHY_XGXS_FLAG)) {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  phy_flags 0x%x\n",
+				  bp->link_config, bp->phy_flags);
+			return;
+		}
+		if (bp->speed_cap_mask &
+		    PORT_HW_CFG_SPEED_CAPABILITY_D0_10G) {
+			bp->req_line_speed = SPEED_10000;
+			bp->advertising = (ADVERTISED_10000baseT_Full |
+					   ADVERTISED_FIBRE);
+		} else {
+			BNX2X_ERR("NVRAM config error. "
+				  "Invalid link_config 0x%x"
+				  "  speed_cap_mask 0x%x\n",
+				  bp->link_config, bp->speed_cap_mask);
+			return;
+		}
+		break;
+
+	default:
+		BNX2X_ERR("NVRAM config error. "
+			  "BAD link speed link_config 0x%x\n",
+			  bp->link_config);
+		bp->req_autoneg |= AUTONEG_SPEED;
+		bp->req_line_speed = 0;
+		bp->advertising = bp->supported;
+		break;
+	}
+	BNX2X_DEV_INFO("req_line_speed %d  req_duplex %d\n",
+		       bp->req_line_speed, bp->req_duplex);
+
+	bp->req_flow_ctrl = (bp->link_config &
+			     PORT_FEATURE_FLOW_CONTROL_MASK);
+	/* Please refer to Table 28B-3 of the 802.3ab-1999 spec */
+	switch (bp->req_flow_ctrl) {
+	case FLOW_CTRL_AUTO:
+		bp->req_autoneg |= AUTONEG_FLOW_CTRL;
+		if (bp->dev->mtu <= 4500) {
+			bp->pause_mode = PAUSE_BOTH;
+			bp->advertising |= (ADVERTISED_Pause |
+					    ADVERTISED_Asym_Pause);
+		} else {
+			bp->pause_mode = PAUSE_ASYMMETRIC;
+			bp->advertising |= ADVERTISED_Asym_Pause;
+		}
+		break;
+
+	case FLOW_CTRL_TX:
+		bp->pause_mode = PAUSE_ASYMMETRIC;
+		bp->advertising |= ADVERTISED_Asym_Pause;
+		break;
+
+	case FLOW_CTRL_RX:
+	case FLOW_CTRL_BOTH:
+		bp->pause_mode = PAUSE_BOTH;
+		bp->advertising |= (ADVERTISED_Pause |
+				    ADVERTISED_Asym_Pause);
+		break;
+
+	case FLOW_CTRL_NONE:
+	default:
+		bp->pause_mode = PAUSE_NONE;
+		bp->advertising &= ~(ADVERTISED_Pause |
+				     ADVERTISED_Asym_Pause);
+		break;
+	}
+	BNX2X_DEV_INFO("req_autoneg 0x%x  req_flow_ctrl 0x%x\n"
+	     KERN_INFO "  pause_mode %d  advertising 0x%x\n",
+		       bp->req_autoneg, bp->req_flow_ctrl,
+		       bp->pause_mode, bp->advertising);
+}
+
+static void bnx2x_get_hwinfo(struct bnx2x *bp)
+{
+	u32 val, val2, val3, val4, id;
+	int port = bp->port;
+	u32 switch_cfg;
+
+	bp->shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
+	BNX2X_DEV_INFO("shmem offset is %x\n", bp->shmem_base);
+
+	/* Get the chip revision id and number. */
+	/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
+	val = REG_RD(bp, MISC_REG_CHIP_NUM);
+	id = ((val & 0xffff) << 16);
+	val = REG_RD(bp, MISC_REG_CHIP_REV);
+	id |= ((val & 0xf) << 12);
+	val = REG_RD(bp, MISC_REG_CHIP_METAL);
+	id |= ((val & 0xff) << 4);
+	REG_RD(bp, MISC_REG_BOND_ID);
+	id |= (val & 0xf);
+	bp->chip_id = id;
+	BNX2X_DEV_INFO("chip ID is %x\n", id);
+
+	if (!bp->shmem_base || (bp->shmem_base != 0xAF900)) {
+		BNX2X_DEV_INFO("MCP not active\n");
+		nomcp = 1;
+		goto set_mac;
+	}
+
+	val = SHMEM_RD(bp, validity_map[port]);
+	if ((val & (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		!= (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		BNX2X_ERR("MCP validity signature bad\n");
+
+	bp->fw_seq = (SHMEM_RD(bp, drv_fw_mb[port].drv_mb_header) &
+		      DRV_MSG_SEQ_NUMBER_MASK);
+
+	bp->hw_config = SHMEM_RD(bp, dev_info.shared_hw_config.config);
+
+	bp->serdes_config =
+		SHMEM_RD(bp, dev_info.port_hw_config[bp->port].serdes_config);
+	bp->lane_config =
+		SHMEM_RD(bp, dev_info.port_hw_config[port].lane_config);
+	bp->ext_phy_config =
+		SHMEM_RD(bp,
+			 dev_info.port_hw_config[port].external_phy_config);
+	bp->speed_cap_mask =
+		SHMEM_RD(bp,
+			 dev_info.port_hw_config[port].speed_capability_mask);
+
+	bp->link_config =
+		SHMEM_RD(bp, dev_info.port_feature_config[port].link_config);
+
+	BNX2X_DEV_INFO("hw_config (%08x)  serdes_config (%08x)\n"
+	     KERN_INFO "  lane_config (%08x)  ext_phy_config (%08x)\n"
+	     KERN_INFO "  speed_cap_mask (%08x)  link_config (%08x)"
+		       "  fw_seq (%08x)\n",
+		       bp->hw_config, bp->serdes_config, bp->lane_config,
+		       bp->ext_phy_config, bp->speed_cap_mask,
+		       bp->link_config, bp->fw_seq);
+
+	switch_cfg = (bp->link_config & PORT_FEATURE_CONNECTED_SWITCH_MASK);
+	bnx2x_link_settings_supported(bp, switch_cfg);
+
+	bp->autoneg = (bp->hw_config & SHARED_HW_CFG_AN_ENABLE_MASK);
+	/* for now disable cl73 */
+	bp->autoneg &= ~SHARED_HW_CFG_AN_ENABLE_CL73;
+	BNX2X_DEV_INFO("autoneg 0x%x\n", bp->autoneg);
+
+	bnx2x_link_settings_requested(bp);
+
+	val2 = SHMEM_RD(bp, dev_info.port_hw_config[port].mac_upper);
+	val = SHMEM_RD(bp, dev_info.port_hw_config[port].mac_lower);
+	bp->dev->dev_addr[0] = (u8)(val2 >> 8 & 0xff);
+	bp->dev->dev_addr[1] = (u8)(val2 & 0xff);
+	bp->dev->dev_addr[2] = (u8)(val >> 24 & 0xff);
+	bp->dev->dev_addr[3] = (u8)(val >> 16 & 0xff);
+	bp->dev->dev_addr[4] = (u8)(val >> 8  & 0xff);
+	bp->dev->dev_addr[5] = (u8)(val & 0xff);
+
+	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, 6);
+
+
+	val = SHMEM_RD(bp, dev_info.shared_hw_config.part_num);
+	val2 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[4]);
+	val3 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[8]);
+	val4 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[12]);
+
+	printk(KERN_INFO PFX "part number %X-%X-%X-%X\n",
+	       val, val2, val3, val4);
+
+	/* bc ver */
+	if (!nomcp) {
+		bp->bc_ver = val = ((SHMEM_RD(bp, dev_info.bc_rev)) >> 8);
+		BNX2X_DEV_INFO("bc_ver %X\n", val);
+		if (val < BNX2X_BC_VER) {
+			/* for now only warn
+			 * later we might need to enforce this */
+			BNX2X_ERR("This driver needs bc_ver %X but found %X,"
+				  " please upgrade BC\n", BNX2X_BC_VER, val);
+		}
+	} else {
+		bp->bc_ver = 0;
+	}
+
+	val = REG_RD(bp, MCP_REG_MCPR_NVM_CFG4);
+	bp->flash_size = (NVRAM_1MB_SIZE << (val & MCPR_NVM_CFG4_FLASH_SIZE));
+	BNX2X_DEV_INFO("flash_size 0x%x (%d)\n",
+		       bp->flash_size, bp->flash_size);
+
+	return;
+
+set_mac: /* only supposed to happen on emulation/FPGA */
+	BNX2X_ERR("warning constant MAC workaround active\n");
+	bp->dev->dev_addr[0] = 0;
+	bp->dev->dev_addr[1] = 0x50;
+	bp->dev->dev_addr[2] = 0xc2;
+	bp->dev->dev_addr[3] = 0x2c;
+	bp->dev->dev_addr[4] = 0x71;
+	bp->dev->dev_addr[5] = port ? 0x0d : 0x0e;
+
+	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, 6);
+
+}
+
+/*
+ * ethtool service functions
+ */
+
+/* All ethtool functions called with rtnl_lock */
+
+static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	cmd->supported = bp->supported;
+	cmd->advertising = bp->advertising;
+
+	if (netif_carrier_ok(dev)) {
+		cmd->speed = bp->line_speed;
+		cmd->duplex = bp->duplex;
+	} else {
+		cmd->speed = bp->req_line_speed;
+		cmd->duplex = bp->req_duplex;
+	}
+
+	if (bp->phy_flags & PHY_XGXS_FLAG) {
+		cmd->port = PORT_FIBRE;
+	} else {
+		cmd->port = PORT_TP;
+	}
+
+	cmd->phy_address = bp->phy_addr;
+	cmd->transceiver = XCVR_INTERNAL;
+
+	if (bp->req_autoneg & AUTONEG_SPEED) {
+		cmd->autoneg = AUTONEG_ENABLE;
+	} else {
+		cmd->autoneg = AUTONEG_DISABLE;
+	}
+
+	cmd->maxtxpkt = 0;
+	cmd->maxrxpkt = 0;
+
+	DP(NETIF_MSG_LINK, "ethtool_cmd: cmd %d\n"
+	   DP_LEVEL "  supported 0x%x  advertising 0x%x  speed %d\n"
+	   DP_LEVEL "  duplex %d  port %d  phy_address %d  transceiver %d\n"
+	   DP_LEVEL "  autoneg %d  maxtxpkt %d  maxrxpkt %d\n",
+	   cmd->cmd, cmd->supported, cmd->advertising, cmd->speed,
+	   cmd->duplex, cmd->port, cmd->phy_address, cmd->transceiver,
+	   cmd->autoneg, cmd->maxtxpkt, cmd->maxrxpkt);
+
+	return 0;
+}
+
+static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	u32 advertising;
+
+	DP(NETIF_MSG_LINK, "ethtool_cmd: cmd %d\n"
+	   DP_LEVEL "  supported 0x%x  advertising 0x%x  speed %d\n"
+	   DP_LEVEL "  duplex %d  port %d  phy_address %d  transceiver %d\n"
+	   DP_LEVEL "  autoneg %d  maxtxpkt %d  maxrxpkt %d\n",
+	   cmd->cmd, cmd->supported, cmd->advertising, cmd->speed,
+	   cmd->duplex, cmd->port, cmd->phy_address, cmd->transceiver,
+	   cmd->autoneg, cmd->maxtxpkt, cmd->maxrxpkt);
+
+	switch (cmd->port) {
+	case PORT_TP:
+		if (!(bp->supported & SUPPORTED_TP))
+			return -EINVAL;
+
+		if (bp->phy_flags & PHY_XGXS_FLAG) {
+			bnx2x_link_reset(bp);
+			bnx2x_link_settings_supported(bp, SWITCH_CFG_1G);
+			bnx2x_phy_deassert(bp);
+		}
+		break;
+
+	case PORT_FIBRE:
+		if (!(bp->supported & SUPPORTED_FIBRE))
+			return -EINVAL;
+
+		if (!(bp->phy_flags & PHY_XGXS_FLAG)) {
+			bnx2x_link_reset(bp);
+			bnx2x_link_settings_supported(bp, SWITCH_CFG_10G);
+			bnx2x_phy_deassert(bp);
+		}
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (cmd->autoneg == AUTONEG_ENABLE) {
+		if (!(bp->supported & SUPPORTED_Autoneg))
+			return -EINVAL;
+
+		/* advertise the requested speed and duplex if supported */
+		cmd->advertising &= bp->supported;
+
+		bp->req_autoneg |= AUTONEG_SPEED;
+		bp->req_line_speed = 0;
+		bp->req_duplex = DUPLEX_FULL;
+		bp->advertising |= (ADVERTISED_Autoneg | cmd->advertising);
+
+	} else { /* forced speed */
+		/* advertise the requested speed and duplex if supported */
+		switch (cmd->speed) {
+		case SPEED_10:
+			if (cmd->duplex == DUPLEX_FULL) {
+				if (!(bp->supported & SUPPORTED_10baseT_Full))
+					return -EINVAL;
+
+				advertising = (ADVERTISED_10baseT_Full |
+					       ADVERTISED_TP);
+			} else {
+				if (!(bp->supported & SUPPORTED_10baseT_Half))
+					return -EINVAL;
+
+				advertising = (ADVERTISED_10baseT_Half |
+					       ADVERTISED_TP);
+			}
+			break;
+
+		case SPEED_100:
+			if (cmd->duplex == DUPLEX_FULL) {
+				if (!(bp->supported &
+						SUPPORTED_100baseT_Full))
+					return -EINVAL;
+
+				advertising = (ADVERTISED_100baseT_Full |
+					       ADVERTISED_TP);
+			} else {
+				if (!(bp->supported &
+						SUPPORTED_100baseT_Half))
+					return -EINVAL;
+
+				advertising = (ADVERTISED_100baseT_Half |
+					       ADVERTISED_TP);
+			}
+			break;
+
+		case SPEED_1000:
+			if (cmd->duplex != DUPLEX_FULL)
+				return -EINVAL;
+
+			if (!(bp->supported & SUPPORTED_1000baseT_Full))
+				return -EINVAL;
+
+			advertising = (ADVERTISED_1000baseT_Full |
+				       ADVERTISED_TP);
+			break;
+
+		case SPEED_2500:
+			if (cmd->duplex != DUPLEX_FULL)
+				return -EINVAL;
+
+			if (!(bp->supported & SUPPORTED_2500baseT_Full))
+				return -EINVAL;
+
+			advertising = (ADVERTISED_2500baseT_Full |
+				       ADVERTISED_TP);
+			break;
+
+		case SPEED_10000:
+			if (cmd->duplex != DUPLEX_FULL)
+				return -EINVAL;
+
+			if (!(bp->supported & SUPPORTED_10000baseT_Full))
+				return -EINVAL;
+
+			advertising = (ADVERTISED_10000baseT_Full |
+				       ADVERTISED_FIBRE);
+			break;
+
+		default:
+			return -EINVAL;
+		}
+
+		bp->req_autoneg &= ~AUTONEG_SPEED;
+		bp->req_line_speed = cmd->speed;
+		bp->req_duplex = cmd->duplex;
+		bp->advertising = advertising;
+	}
+
+	DP(NETIF_MSG_LINK, "req_autoneg 0x%x  req_line_speed %d\n"
+	   DP_LEVEL "  req_duplex %d  advertising 0x%x\n",
+	   bp->req_autoneg, bp->req_line_speed, bp->req_duplex,
+	   bp->advertising);
+
+	bnx2x_stop_stats(bp);
+	bnx2x_link_initialize(bp);
+
+	return 0;
+}
+
+static void bnx2x_get_drvinfo(struct net_device *dev,
+			      struct ethtool_drvinfo *info)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	strcpy(info->driver, DRV_MODULE_NAME);
+	strcpy(info->version, DRV_MODULE_VERSION);
+	snprintf(info->fw_version, 32, "%d.%d.%d:%d (BC VER %x)",
+		 BCM_5710_FW_MAJOR_VERSION, BCM_5710_FW_MINOR_VERSION,
+		 BCM_5710_FW_REVISION_VERSION, BCM_5710_FW_COMPILE_FLAGS,
+		 bp->bc_ver);
+	strcpy(info->bus_info, pci_name(bp->pdev));
+	info->n_stats = BNX2X_NUM_STATS;
+	info->testinfo_len = BNX2X_NUM_TESTS;
+	info->eedump_len = bp->flash_size;
+	info->regdump_len = 0;
+}
+
+static void bnx2x_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if (bp->flags & NO_WOL_FLAG) {
+		wol->supported = 0;
+		wol->wolopts = 0;
+	} else {
+		wol->supported = WAKE_MAGIC;
+		if (bp->wol)
+			wol->wolopts = WAKE_MAGIC;
+		else
+			wol->wolopts = 0;
+	}
+	memset(&wol->sopass, 0, sizeof(wol->sopass));
+}
+
+static int bnx2x_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if (wol->wolopts & ~WAKE_MAGIC)
+		return -EINVAL;
+
+	if (wol->wolopts & WAKE_MAGIC) {
+		if (bp->flags & NO_WOL_FLAG)
+			return -EINVAL;
+
+		bp->wol = 1;
+	} else {
+		bp->wol = 0;
+	}
+	return 0;
+}
+
+static u32 bnx2x_get_msglevel(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	return bp->msglevel;
+}
+
+static void bnx2x_set_msglevel(struct net_device *dev, u32 level)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if (capable(CAP_NET_ADMIN))
+		bp->msglevel = level;
+}
+
+static int bnx2x_nway_reset(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_PROBE, "state is %x, returning\n", bp->state);
+		return -EAGAIN;
+	}
+
+	bnx2x_stop_stats(bp);
+	bnx2x_link_initialize(bp);
+
+	return 0;
+}
+
+static int bnx2x_get_eeprom_len(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	return bp->flash_size;
+}
+
+static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
+{
+	int port = bp->port;
+	int count, i;
+	u32 val = 0;
+
+	/* adjust timeout for emulation/FPGA */
+	count = NVRAM_TIMEOUT_COUNT;
+	if (CHIP_REV_IS_SLOW(bp))
+		count *= 100;
+
+	/* request access to nvram interface */
+	REG_WR(bp, MCP_REG_MCPR_NVM_SW_ARB,
+	       (MCPR_NVM_SW_ARB_ARB_REQ_SET1 << port));
+
+	for (i = 0; i < count*10; i++) {
+		val = REG_RD(bp, MCP_REG_MCPR_NVM_SW_ARB);
+		if (val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port))
+			break;
+
+		udelay(5);
+	}
+
+	if (!(val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port))) {
+		DP(NETIF_MSG_NVM, "cannot get access to nvram interface\n");
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int bnx2x_release_nvram_lock(struct bnx2x *bp)
+{
+	int port = bp->port;
+	int count, i;
+	u32 val = 0;
+
+	/* adjust timeout for emulation/FPGA */
+	count = NVRAM_TIMEOUT_COUNT;
+	if (CHIP_REV_IS_SLOW(bp))
+		count *= 100;
+
+	/* relinquish nvram interface */
+	REG_WR(bp, MCP_REG_MCPR_NVM_SW_ARB,
+	       (MCPR_NVM_SW_ARB_ARB_REQ_CLR1 << port));
+
+	for (i = 0; i < count*10; i++) {
+		val = REG_RD(bp, MCP_REG_MCPR_NVM_SW_ARB);
+		if (!(val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port)))
+			break;
+
+		udelay(5);
+	}
+
+	if (val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port)) {
+		DP(NETIF_MSG_NVM, "cannot free access to nvram interface\n");
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static void bnx2x_enable_nvram_access(struct bnx2x *bp)
+{
+	u32 val;
+
+	val = REG_RD(bp, MCP_REG_MCPR_NVM_ACCESS_ENABLE);
+
+	/* enable both bits, even on read */
+	REG_WR(bp, MCP_REG_MCPR_NVM_ACCESS_ENABLE,
+	       (val | MCPR_NVM_ACCESS_ENABLE_EN |
+		      MCPR_NVM_ACCESS_ENABLE_WR_EN));
+}
+
+static void bnx2x_disable_nvram_access(struct bnx2x *bp)
+{
+	u32 val;
+
+	val = REG_RD(bp, MCP_REG_MCPR_NVM_ACCESS_ENABLE);
+
+	/* disable both bits, even after read */
+	REG_WR(bp, MCP_REG_MCPR_NVM_ACCESS_ENABLE,
+	       (val & ~(MCPR_NVM_ACCESS_ENABLE_EN |
+			MCPR_NVM_ACCESS_ENABLE_WR_EN)));
+}
+
+static int bnx2x_nvram_read_dword(struct bnx2x *bp, u32 offset, u32 *ret_val,
+				  u32 cmd_flags)
+{
+	int rc;
+	int count, i;
+	u32 val;
+
+	/* build the command word */
+	cmd_flags |= MCPR_NVM_COMMAND_DOIT;
+
+	/* need to clear DONE bit separately */
+	REG_WR(bp, MCP_REG_MCPR_NVM_COMMAND, MCPR_NVM_COMMAND_DONE);
+
+	/* address of the NVRAM to read from */
+	REG_WR(bp, MCP_REG_MCPR_NVM_ADDR,
+	       (offset & MCPR_NVM_ADDR_NVM_ADDR_VALUE));
+
+	/* issue a read command */
+	REG_WR(bp, MCP_REG_MCPR_NVM_COMMAND, cmd_flags);
+
+	/* adjust timeout for emulation/FPGA */
+	count = NVRAM_TIMEOUT_COUNT;
+	if (CHIP_REV_IS_SLOW(bp))
+		count *= 100;
+
+	/* wait for completion */
+	*ret_val = 0;
+	rc = -EBUSY;
+	for (i = 0; i < count; i++) {
+		udelay(5);
+		val = REG_RD(bp, MCP_REG_MCPR_NVM_COMMAND);
+
+		if (val & MCPR_NVM_COMMAND_DONE) {
+			val = REG_RD(bp, MCP_REG_MCPR_NVM_READ);
+			DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
+			/* we read nvram data in cpu order
+			 * but ethtool sees it as an array of bytes
+			 * converting to big-endian will do the work */
+			val = cpu_to_be32(val);
+			*ret_val = val;
+			rc = 0;
+			break;
+		}
+	}
+
+	return rc;
+}
+
+static int bnx2x_nvram_read(struct bnx2x *bp, u32 offset, u8 *ret_buf,
+			    int buf_size)
+{
+	int rc;
+	u32 cmd_flags;
+	u32 val;
+
+	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
+		DP(NETIF_MSG_NVM,
+		   "Invalid paramter: offset 0x%x  buf_size 0x%x\n",
+		   offset, buf_size);
+		return -EINVAL;
+	}
+
+	if (offset + buf_size > bp->flash_size) {
+		DP(NETIF_MSG_NVM, "Invalid paramter: offset (0x%x) +"
+				  " buf_size (0x%x) > flash_size (0x%x)\n",
+		   offset, buf_size, bp->flash_size);
+		return -EINVAL;
+	}
+
+	/* request access to nvram interface */
+	rc = bnx2x_acquire_nvram_lock(bp);
+	if (rc)
+		return rc;
+
+	/* enable access to nvram interface */
+	bnx2x_enable_nvram_access(bp);
+
+	/* read the first word(s) */
+	cmd_flags = MCPR_NVM_COMMAND_FIRST;
+	while ((buf_size > sizeof(u32)) && (rc == 0)) {
+		rc = bnx2x_nvram_read_dword(bp, offset, &val, cmd_flags);
+		memcpy(ret_buf, &val, 4);
+
+		/* advance to the next dword */
+		offset += sizeof(u32);
+		ret_buf += sizeof(u32);
+		buf_size -= sizeof(u32);
+		cmd_flags = 0;
+	}
+
+	if (rc == 0) {
+		cmd_flags |= MCPR_NVM_COMMAND_LAST;
+		rc = bnx2x_nvram_read_dword(bp, offset, &val, cmd_flags);
+		memcpy(ret_buf, &val, 4);
+	}
+
+	/* disable access to nvram interface */
+	bnx2x_disable_nvram_access(bp);
+	bnx2x_release_nvram_lock(bp);
+
+	return rc;
+}
+
+static int bnx2x_get_eeprom(struct net_device *dev,
+			    struct ethtool_eeprom *eeprom, u8 *eebuf)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	int rc;
+
+	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
+	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
+	   eeprom->len, eeprom->len);
+
+	/* parameters already validated in ethtool_get_eeprom */
+
+	rc = bnx2x_nvram_read(bp, eeprom->offset, eebuf, eeprom->len);
+
+	return rc;
+}
+
+static int bnx2x_nvram_write_dword(struct bnx2x *bp, u32 offset, u32 val,
+				   u32 cmd_flags)
+{
+	int rc;
+	int count, i;
+
+	/* build the command word */
+	cmd_flags |= MCPR_NVM_COMMAND_DOIT | MCPR_NVM_COMMAND_WR;
+
+	/* need to clear DONE bit separately */
+	REG_WR(bp, MCP_REG_MCPR_NVM_COMMAND, MCPR_NVM_COMMAND_DONE);
+
+	/* write the data */
+	REG_WR(bp, MCP_REG_MCPR_NVM_WRITE, val);
+
+	/* address of the NVRAM to write to */
+	REG_WR(bp, MCP_REG_MCPR_NVM_ADDR,
+	       (offset & MCPR_NVM_ADDR_NVM_ADDR_VALUE));
+
+	/* issue the write command */
+	REG_WR(bp, MCP_REG_MCPR_NVM_COMMAND, cmd_flags);
+
+	/* adjust timeout for emulation/FPGA */
+	count = NVRAM_TIMEOUT_COUNT;
+	if (CHIP_REV_IS_SLOW(bp))
+		count *= 100;
+
+	/* wait for completion */
+	rc = -EBUSY;
+	for (i = 0; i < count; i++) {
+		udelay(5);
+		val = REG_RD(bp, MCP_REG_MCPR_NVM_COMMAND);
+		if (val & MCPR_NVM_COMMAND_DONE) {
+			rc = 0;
+			break;
+		}
+	}
+
+	return rc;
+}
+
+#define BYTE_OFFSET(offset)     	(8 * (offset & 0x03))
+
+static int bnx2x_nvram_write1(struct bnx2x *bp, u32 offset, u8 *data_buf,
+			      int buf_size)
+{
+	int rc;
+	u32 cmd_flags;
+	u32 align_offset;
+	u32 val;
+
+	if (offset + buf_size > bp->flash_size) {
+		DP(NETIF_MSG_NVM, "Invalid paramter: offset (0x%x) +"
+				  " buf_size (0x%x) > flash_size (0x%x)\n",
+		   offset, buf_size, bp->flash_size);
+		return -EINVAL;
+	}
+
+	/* request access to nvram interface */
+	rc = bnx2x_acquire_nvram_lock(bp);
+	if (rc)
+		return rc;
+
+	/* enable access to nvram interface */
+	bnx2x_enable_nvram_access(bp);
+
+	cmd_flags = (MCPR_NVM_COMMAND_FIRST | MCPR_NVM_COMMAND_LAST);
+	align_offset = (offset & ~0x03);
+	rc = bnx2x_nvram_read_dword(bp, align_offset, &val, cmd_flags);
+
+	if (rc == 0) {
+		val &= ~(0xff << BYTE_OFFSET(offset));
+		val |= (*data_buf << BYTE_OFFSET(offset));
+
+		/* nvram data is returned as an array of bytes
+		 * convert it back to cpu order */
+		val = be32_to_cpu(val);
+
+		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
+
+		rc = bnx2x_nvram_write_dword(bp, align_offset, val,
+					     cmd_flags);
+	}
+
+	/* disable access to nvram interface */
+	bnx2x_disable_nvram_access(bp);
+	bnx2x_release_nvram_lock(bp);
+
+	return rc;
+}
+
+static int bnx2x_nvram_write(struct bnx2x *bp, u32 offset, u8 *data_buf,
+			     int buf_size)
+{
+	int rc;
+	u32 cmd_flags;
+	u32 val;
+	u32 written_so_far;
+
+	if (buf_size == 1) {    /* ethtool */
+		return bnx2x_nvram_write1(bp, offset, data_buf, buf_size);
+	}
+
+	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
+		DP(NETIF_MSG_NVM,
+		   "Invalid paramter: offset 0x%x  buf_size 0x%x\n",
+		   offset, buf_size);
+		return -EINVAL;
+	}
+
+	if (offset + buf_size > bp->flash_size) {
+		DP(NETIF_MSG_NVM, "Invalid paramter: offset (0x%x) +"
+				  " buf_size (0x%x) > flash_size (0x%x)\n",
+		   offset, buf_size, bp->flash_size);
+		return -EINVAL;
+	}
+
+	/* request access to nvram interface */
+	rc = bnx2x_acquire_nvram_lock(bp);
+	if (rc)
+		return rc;
+
+	/* enable access to nvram interface */
+	bnx2x_enable_nvram_access(bp);
+
+	written_so_far = 0;
+	cmd_flags = MCPR_NVM_COMMAND_FIRST;
+	while ((written_so_far < buf_size) && (rc == 0)) {
+		if (written_so_far == (buf_size - sizeof(u32)))
+			cmd_flags |= MCPR_NVM_COMMAND_LAST;
+		else if (((offset + 4) % NVRAM_PAGE_SIZE) == 0)
+			cmd_flags |= MCPR_NVM_COMMAND_LAST;
+		else if ((offset % NVRAM_PAGE_SIZE) == 0)
+			cmd_flags |= MCPR_NVM_COMMAND_FIRST;
+
+		memcpy(&val, data_buf, 4);
+		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
+
+		rc = bnx2x_nvram_write_dword(bp, offset, val, cmd_flags);
+
+		/* advance to the next dword */
+		offset += sizeof(u32);
+		data_buf += sizeof(u32);
+		written_so_far += sizeof(u32);
+		cmd_flags = 0;
+	}
+
+	/* disable access to nvram interface */
+	bnx2x_disable_nvram_access(bp);
+	bnx2x_release_nvram_lock(bp);
+
+	return rc;
+}
+
+static int bnx2x_set_eeprom(struct net_device *dev,
+			    struct ethtool_eeprom *eeprom, u8 *eebuf)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	int rc;
+
+	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
+	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
+	   eeprom->len, eeprom->len);
+
+	/* parameters already validated in ethtool_set_eeprom */
+
+	rc = bnx2x_nvram_write(bp, eeprom->offset, eebuf, eeprom->len);
+
+	return rc;
+}
+
+static int bnx2x_get_coalesce(struct net_device *dev,
+			      struct ethtool_coalesce *coal)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	memset(coal, 0, sizeof(struct ethtool_coalesce));
+
+	coal->rx_coalesce_usecs = bp->rx_ticks;
+	coal->tx_coalesce_usecs = bp->tx_ticks;
+	coal->stats_block_coalesce_usecs = bp->stats_ticks;
+
+	return 0;
+}
+
+static int bnx2x_set_coalesce(struct net_device *dev,
+			      struct ethtool_coalesce *coal)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	bp->rx_ticks = (u16) coal->rx_coalesce_usecs;
+	if (bp->rx_ticks > 3000)
+		bp->rx_ticks = 3000;
+
+	bp->tx_ticks = (u16) coal->tx_coalesce_usecs;
+	if (bp->tx_ticks > 0x3000)
+		bp->tx_ticks = 0x3000;
+
+	bp->stats_ticks = coal->stats_block_coalesce_usecs;
+	if (bp->stats_ticks > 0xffff00)
+		bp->stats_ticks = 0xffff00;
+	bp->stats_ticks &= 0xffff00;
+
+	if (netif_running(bp->dev))
+		bnx2x_update_coalesce(bp);
+
+	return 0;
+}
+
+static void bnx2x_get_ringparam(struct net_device *dev,
+				struct ethtool_ringparam *ering)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	ering->rx_max_pending = MAX_RX_AVAIL;
+	ering->rx_mini_max_pending = 0;
+	ering->rx_jumbo_max_pending = 0;
+
+	ering->rx_pending = bp->rx_ring_size;
+	ering->rx_mini_pending = 0;
+	ering->rx_jumbo_pending = 0;
+
+	ering->tx_max_pending = MAX_TX_AVAIL;
+	ering->tx_pending = bp->tx_ring_size;
+}
+
+static int bnx2x_set_ringparam(struct net_device *dev,
+			       struct ethtool_ringparam *ering)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if ((ering->rx_pending > MAX_RX_AVAIL) ||
+	    (ering->tx_pending > MAX_TX_AVAIL) ||
+	    (ering->tx_pending <= MAX_SKB_FRAGS + 4))
+		return -EINVAL;
+
+	bp->rx_ring_size = ering->rx_pending;
+	bp->tx_ring_size = ering->tx_pending;
+
+	if (netif_running(bp->dev)) {
+		bnx2x_nic_unload(bp, 0);
+		bnx2x_nic_load(bp, 0);
+	}
+
+	return 0;
+}
+
+static void bnx2x_get_pauseparam(struct net_device *dev,
+				 struct ethtool_pauseparam *epause)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	epause->autoneg =
+		((bp->req_autoneg & AUTONEG_FLOW_CTRL) == AUTONEG_FLOW_CTRL);
+	epause->rx_pause = ((bp->flow_ctrl & FLOW_CTRL_RX) == FLOW_CTRL_RX);
+	epause->tx_pause = ((bp->flow_ctrl & FLOW_CTRL_TX) == FLOW_CTRL_TX);
+
+	DP(NETIF_MSG_LINK, "ethtool_pauseparam: cmd %d\n"
+	   DP_LEVEL "  autoneg %d  rx_pause %d  tx_pause %d\n",
+	   epause->cmd, epause->autoneg, epause->rx_pause, epause->tx_pause);
+}
+
+static int bnx2x_set_pauseparam(struct net_device *dev,
+				struct ethtool_pauseparam *epause)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	DP(NETIF_MSG_LINK, "ethtool_pauseparam: cmd %d\n"
+	   DP_LEVEL "  autoneg %d  rx_pause %d  tx_pause %d\n",
+	   epause->cmd, epause->autoneg, epause->rx_pause, epause->tx_pause);
+
+	bp->req_flow_ctrl = FLOW_CTRL_AUTO;
+	if (epause->autoneg) {
+		bp->req_autoneg |= AUTONEG_FLOW_CTRL;
+		if (bp->dev->mtu <= 4500) {
+			bp->pause_mode = PAUSE_BOTH;
+			bp->advertising |= (ADVERTISED_Pause |
+					    ADVERTISED_Asym_Pause);
+		} else {
+			bp->pause_mode = PAUSE_ASYMMETRIC;
+			bp->advertising |= ADVERTISED_Asym_Pause;
+		}
+
+	} else {
+		bp->req_autoneg &= ~AUTONEG_FLOW_CTRL;
+
+		if (epause->rx_pause)
+			bp->req_flow_ctrl |= FLOW_CTRL_RX;
+		if (epause->tx_pause)
+			bp->req_flow_ctrl |= FLOW_CTRL_TX;
+
+		switch (bp->req_flow_ctrl) {
+		case FLOW_CTRL_AUTO:
+			bp->req_flow_ctrl = FLOW_CTRL_NONE;
+			bp->pause_mode = PAUSE_NONE;
+			bp->advertising &= ~(ADVERTISED_Pause |
+					     ADVERTISED_Asym_Pause);
+			break;
+
+		case FLOW_CTRL_TX:
+			bp->pause_mode = PAUSE_ASYMMETRIC;
+			bp->advertising |= ADVERTISED_Asym_Pause;
+			break;
+
+		case FLOW_CTRL_RX:
+		case FLOW_CTRL_BOTH:
+			bp->pause_mode = PAUSE_BOTH;
+			bp->advertising |= (ADVERTISED_Pause |
+					    ADVERTISED_Asym_Pause);
+			break;
+		}
+	}
+
+	DP(NETIF_MSG_LINK, "req_autoneg 0x%x  req_flow_ctrl 0x%x\n"
+	   DP_LEVEL "  pause_mode %d  advertising 0x%x\n",
+	   bp->req_autoneg, bp->req_flow_ctrl, bp->pause_mode,
+	   bp->advertising);
+
+	bnx2x_stop_stats(bp);
+	bnx2x_link_initialize(bp);
+
+	return 0;
+}
+
+static u32 bnx2x_get_rx_csum(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	return bp->rx_csum;
+}
+
+static int bnx2x_set_rx_csum(struct net_device *dev, u32 data)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	bp->rx_csum = data;
+	return 0;
+}
+
+static int bnx2x_set_tso(struct net_device *dev, u32 data)
+{
+	if (data)
+		dev->features |= (NETIF_F_TSO | NETIF_F_TSO_ECN);
+	else
+		dev->features &= ~(NETIF_F_TSO | NETIF_F_TSO_ECN);
+	return 0;
+}
+
+static struct {
+	char string[ETH_GSTRING_LEN];
+} bnx2x_tests_str_arr[BNX2X_NUM_TESTS] = {
+	{ "MC Errors  (online)" }
+};
+
+static int bnx2x_self_test_count(struct net_device *dev)
+{
+	return BNX2X_NUM_TESTS;
+}
+
+static void bnx2x_self_test(struct net_device *dev,
+			    struct ethtool_test *etest, u64 *buf)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	int stats_state;
+
+	memset(buf, 0, sizeof(u64) * BNX2X_NUM_TESTS);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_PROBE, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+	stats_state = bp->stats_state;
+	bnx2x_stop_stats(bp);
+
+	if (bnx2x_mc_assert(bp) != 0) {
+		buf[0] = 1;
+		etest->flags |= ETH_TEST_FL_FAILED;
+	}
+
+#ifdef BNX2X_EXTRA_DEBUG
+	bnx2x_panic_dump(bp);
+#endif
+	bp->stats_state = stats_state;
+}
+
+static struct {
+	char string[ETH_GSTRING_LEN];
+} bnx2x_stats_str_arr[BNX2X_NUM_STATS] = {
+	{ "rx_bytes"},  			 /*  0 */
+	{ "rx_error_bytes"},    		 /*  1 */
+	{ "tx_bytes"},  			 /*  2 */
+	{ "tx_error_bytes"},    		 /*  3 */
+	{ "rx_ucast_packets"},  		 /*  4 */
+	{ "rx_mcast_packets"},  		 /*  5 */
+	{ "rx_bcast_packets"},  		 /*  6 */
+	{ "tx_ucast_packets"},  		 /*  7 */
+	{ "tx_mcast_packets"},  		 /*  8 */
+	{ "tx_bcast_packets"},  		 /*  9 */
+	{ "tx_mac_errors"},     		 /* 10 */
+	{ "tx_carrier_errors"}, 		 /* 11 */
+	{ "rx_crc_errors"},     		 /* 12 */
+	{ "rx_align_errors"},   		 /* 13 */
+	{ "tx_single_collisions"},      	 /* 14 */
+	{ "tx_multi_collisions"},       	 /* 15 */
+	{ "tx_deferred"},       		 /* 16 */
+	{ "tx_excess_collisions"},      	 /* 17 */
+	{ "tx_late_collisions"},		 /* 18 */
+	{ "tx_total_collisions"},       	 /* 19 */
+	{ "rx_fragments"},      		 /* 20 */
+	{ "rx_jabbers"},			 /* 21 */
+	{ "rx_undersize_packets"},      	 /* 22 */
+	{ "rx_oversize_packets"},       	 /* 23 */
+	{ "rx_xon_frames"},     		 /* 24 */
+	{ "rx_xoff_frames"},    		 /* 25 */
+	{ "tx_xon_frames"},     		 /* 26 */
+	{ "tx_xoff_frames"},    		 /* 27 */
+	{ "rx_mac_ctrl_frames"},		 /* 28 */
+	{ "rx_filtered_packets"},       	 /* 29 */
+	{ "rx_discards"},       		 /* 30 */
+};
+
+#define STATS_OFFSET32(offset_name) \
+	(offsetof(struct bnx2x_eth_stats, offset_name) / 4)
+
+static unsigned long bnx2x_stats_offset_arr[BNX2X_NUM_STATS] = {
+	STATS_OFFSET32(total_bytes_received_hi),		     /*  0 */
+	STATS_OFFSET32(stat_IfHCInBadOctets_hi),		     /*  1 */
+	STATS_OFFSET32(total_bytes_transmitted_hi),     	     /*  2 */
+	STATS_OFFSET32(stat_IfHCOutBadOctets_hi),       	     /*  3 */
+	STATS_OFFSET32(total_unicast_packets_received_hi),           /*  4 */
+	STATS_OFFSET32(total_multicast_packets_received_hi),         /*  5 */
+	STATS_OFFSET32(total_broadcast_packets_received_hi),         /*  6 */
+	STATS_OFFSET32(total_unicast_packets_transmitted_hi),        /*  7 */
+	STATS_OFFSET32(total_multicast_packets_transmitted_hi),      /*  8 */
+	STATS_OFFSET32(total_broadcast_packets_transmitted_hi),      /*  9 */
+	STATS_OFFSET32(stat_Dot3statsInternalMacTransmitErrors),     /* 10 */
+	STATS_OFFSET32(stat_Dot3StatsCarrierSenseErrors),            /* 11 */
+	STATS_OFFSET32(crc_receive_errors),     		     /* 12 */
+	STATS_OFFSET32(alignment_errors),       		     /* 13 */
+	STATS_OFFSET32(single_collision_transmit_frames),            /* 14 */
+	STATS_OFFSET32(multiple_collision_transmit_frames),          /* 15 */
+	STATS_OFFSET32(stat_Dot3StatsDeferredTransmissions),         /* 16 */
+	STATS_OFFSET32(excessive_collision_frames),     	     /* 17 */
+	STATS_OFFSET32(late_collision_frames),  		     /* 18 */
+	STATS_OFFSET32(number_of_bugs_found_in_stats_spec),          /* 19 */
+	STATS_OFFSET32(runt_packets_received),  		     /* 20 */
+	STATS_OFFSET32(jabber_packets_received),		     /* 21 */
+	STATS_OFFSET32(error_runt_packets_received),    	     /* 22 */
+	STATS_OFFSET32(error_jabber_packets_received),  	     /* 23 */
+	STATS_OFFSET32(pause_xon_frames_received),      	     /* 24 */
+	STATS_OFFSET32(pause_xoff_frames_received),     	     /* 25 */
+	STATS_OFFSET32(pause_xon_frames_transmitted),   	     /* 26 */
+	STATS_OFFSET32(pause_xoff_frames_transmitted),  	     /* 27 */
+	STATS_OFFSET32(control_frames_received),		     /* 28 */
+	STATS_OFFSET32(mac_filter_discard),     		     /* 29 */
+	STATS_OFFSET32(no_buff_discard),			     /* 30 */
+};
+
+static u8 bnx2x_stats_len_arr[BNX2X_NUM_STATS] = {
+	8, 0, 8, 0, 8, 8, 8, 8, 8, 8,
+	4, 0, 4, 4, 4, 4, 4, 4, 4, 4,
+	4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
+	4,
+};
+
+static void bnx2x_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
+{
+	switch (stringset) {
+	case ETH_SS_STATS:
+		memcpy(buf, bnx2x_stats_str_arr, sizeof(bnx2x_stats_str_arr));
+		break;
+
+	case ETH_SS_TEST:
+		memcpy(buf, bnx2x_tests_str_arr, sizeof(bnx2x_tests_str_arr));
+		break;
+	}
+}
+
+static int bnx2x_get_stats_count(struct net_device *dev)
+{
+	return BNX2X_NUM_STATS;
+}
+
+static void bnx2x_get_ethtool_stats(struct net_device *dev,
+				    struct ethtool_stats *stats, u64 *buf)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	u32 *hw_stats = (u32 *)bnx2x_sp_check(bp, eth_stats);
+	int i;
+
+	for (i = 0; i < BNX2X_NUM_STATS; i++) {
+		if (bnx2x_stats_len_arr[i] == 0) {
+			/* skip this counter */
+			buf[i] = 0;
+			continue;
+		}
+		if (!hw_stats) {
+			buf[i] = 0;
+			continue;
+		}
+		if (bnx2x_stats_len_arr[i] == 4) {
+			/* 4-byte counter */
+		       buf[i] = (u64) *(hw_stats + bnx2x_stats_offset_arr[i]);
+			continue;
+		}
+		/* 8-byte counter */
+		buf[i] = HILO_U64(*(hw_stats + bnx2x_stats_offset_arr[i]),
+				 *(hw_stats + bnx2x_stats_offset_arr[i] + 1));
+	}
+}
+
+static int bnx2x_phys_id(struct net_device *dev, u32 data)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	int i;
+
+	if (data == 0)
+		data = 2;
+
+	for (i = 0; i < (data * 2); i++) {
+		if ((i % 2) == 0) {
+			bnx2x_leds_set(bp, SPEED_1000);
+		} else {
+			bnx2x_leds_unset(bp);
+		}
+		msleep_interruptible(500);
+		if (signal_pending(current))
+			break;
+	}
+
+	if (bp->link_up)
+		bnx2x_leds_set(bp, bp->line_speed);
+
+	return 0;
+}
+
+static struct ethtool_ops bnx2x_ethtool_ops = {
+	.get_settings   	= bnx2x_get_settings,
+	.set_settings   	= bnx2x_set_settings,
+	.get_drvinfo    	= bnx2x_get_drvinfo,
+	.get_wol		= bnx2x_get_wol,
+	.set_wol		= bnx2x_set_wol,
+	.get_msglevel   	= bnx2x_get_msglevel,
+	.set_msglevel   	= bnx2x_set_msglevel,
+	.nway_reset     	= bnx2x_nway_reset,
+	.get_link       	= ethtool_op_get_link,
+	.get_eeprom_len 	= bnx2x_get_eeprom_len,
+	.get_eeprom     	= bnx2x_get_eeprom,
+	.set_eeprom     	= bnx2x_set_eeprom,
+	.get_coalesce   	= bnx2x_get_coalesce,
+	.set_coalesce   	= bnx2x_set_coalesce,
+	.get_ringparam  	= bnx2x_get_ringparam,
+	.set_ringparam  	= bnx2x_set_ringparam,
+	.get_pauseparam 	= bnx2x_get_pauseparam,
+	.set_pauseparam 	= bnx2x_set_pauseparam,
+	.get_rx_csum    	= bnx2x_get_rx_csum,
+	.set_rx_csum    	= bnx2x_set_rx_csum,
+	.get_tx_csum    	= ethtool_op_get_tx_csum,
+	.set_tx_csum    	= ethtool_op_set_tx_csum,
+	.get_sg 		= ethtool_op_get_sg,
+	.set_sg 		= ethtool_op_set_sg,
+	.get_tso		= ethtool_op_get_tso,
+	.set_tso		= bnx2x_set_tso,
+	.self_test_count	= bnx2x_self_test_count,
+	.self_test      	= bnx2x_self_test,
+	.get_strings    	= bnx2x_get_strings,
+	.phys_id		= bnx2x_phys_id,
+	.get_stats_count	= bnx2x_get_stats_count,
+	.get_ethtool_stats      = bnx2x_get_ethtool_stats
+};
+
+/* end of ethtool_ops */
+
+/****************************************************************************
+* General service functions
+****************************************************************************/
+
+static int bnx2x_set_power_state(struct bnx2x *bp, pci_power_t state)
+{
+	u16 pmcsr;
+
+	pci_read_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, &pmcsr);
+
+	switch (state) {
+	case PCI_D0:
+		pci_write_config_word(bp->pdev,
+				      bp->pm_cap + PCI_PM_CTRL,
+				      ((pmcsr & ~PCI_PM_CTRL_STATE_MASK) |
+				       PCI_PM_CTRL_PME_STATUS));
+
+		if (pmcsr & PCI_PM_CTRL_STATE_MASK)
+		/* delay required during transition out of D3hot */
+			msleep(20);
+		break;
+
+	case PCI_D3hot:
+		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
+		pmcsr |= 3;
+
+		if (bp->wol)
+			pmcsr |= PCI_PM_CTRL_PME_ENABLE;
+
+		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
+				      pmcsr);
+
+		/* No more memory access after this point until
+		* device is brought back to D0.
+		*/
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * net_device service functions
+ */
+
+/* Called with rtnl_lock from vlan functions and also netif_tx_lock
+ * from set_multicast.
+ */
+static void bnx2x_set_rx_mode(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	u32 rx_mode = BNX2X_RX_MODE_NORMAL;
+
+	DP(NETIF_MSG_IFUP, "called dev->flags = %x\n", dev->flags);
+
+	if (dev->flags & IFF_PROMISC)
+		rx_mode = BNX2X_RX_MODE_PROMISC;
+
+	else if ((dev->flags & IFF_ALLMULTI) ||
+		 (dev->mc_count > BNX2X_MAX_MULTICAST))
+		rx_mode = BNX2X_RX_MODE_ALLMULTI;
+
+	else { /* some multicasts */
+		int i, old, offset;
+		struct dev_mc_list *mclist;
+		struct mac_configuration_cmd *config =
+						bnx2x_sp(bp, mcast_config);
+
+		for (i = 0, mclist = dev->mc_list;
+		     mclist && (i < dev->mc_count);
+		     i++, mclist = mclist->next) {
+
+			config->config_table[i].cam_entry.msb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[0]);
+			config->config_table[i].cam_entry.middle_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[2]);
+			config->config_table[i].cam_entry.lsb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[4]);
+			config->config_table[i].cam_entry.flags =
+							cpu_to_le16(bp->port);
+			config->config_table[i].target_table_entry.flags = 0;
+			config->config_table[i].target_table_entry.
+								client_id = 0;
+			config->config_table[i].target_table_entry.
+								vlan_id = 0;
+
+			DP(NETIF_MSG_IFUP,
+			   "setting MCAST[%d] (%04x:%04x:%04x)\n",
+			   i, config->config_table[i].cam_entry.msb_mac_addr,
+			   config->config_table[i].cam_entry.middle_mac_addr,
+			   config->config_table[i].cam_entry.lsb_mac_addr);
+		}
+		old = config->hdr.length_6b;
+		if (old > i) {
+			for (; i < old; i++) {
+				if (CAM_IS_INVALID(config->config_table[i])) {
+					i--; /* already invalidated */
+					break;
+				}
+				/* invalidate */
+				CAM_INVALIDATE(config->config_table[i]);
+			}
+		}
+
+		if (CHIP_REV_IS_SLOW(bp))
+			offset = BNX2X_MAX_EMUL_MULTI*(1 + bp->port);
+		else
+			offset = BNX2X_MAX_MULTICAST*(1 + bp->port);
+
+		config->hdr.length_6b = i;
+		config->hdr.offset = offset;
+		config->hdr.reserved0 = 0;
+		config->hdr.reserved1 = 0;
+
+		bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+			      U64_HI(bnx2x_sp_mapping(bp, mcast_config)),
+			      U64_LO(bnx2x_sp_mapping(bp, mcast_config)), 0);
+	}
+
+	bp->rx_mode = rx_mode;
+	bnx2x_set_storm_rx_mode(bp);
+}
+
+static int bnx2x_poll(struct napi_struct *napi, int budget)
+{
+	struct bnx2x_fastpath *fp = container_of(napi, struct bnx2x_fastpath,
+						 napi);
+	struct bnx2x *bp = fp->bp;
+	int work_done = 0;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		goto out_panic;
+#endif
+
+	prefetch(fp->tx_buf_ring[TX_BD(fp->tx_pkt_cons)].skb);
+	prefetch(fp->rx_buf_ring[RX_BD(fp->rx_bd_cons)].skb);
+	prefetch((char *)(fp->rx_buf_ring[RX_BD(fp->rx_bd_cons)].skb) + 256);
+
+	bnx2x_update_fpsb_idx(fp);
+
+	if (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons)
+		bnx2x_tx_int(fp, budget);
+
+
+	if (le16_to_cpu(*fp->rx_cons_sb) != fp->rx_comp_cons)
+		work_done = bnx2x_rx_int(fp, budget);
+
+
+	rmb(); /* bnx2x_has_work() reads the status block */
+
+	/* must not complete if we consumed full budget */
+	if ((work_done < budget) && !bnx2x_has_work(fp)) {
+
+#ifdef BNX2X_STOP_ON_ERROR
+out_panic:
+#endif
+		netif_rx_complete(bp->dev, napi);
+
+		bnx2x_ack_sb(bp, fp->index, USTORM_ID,
+			     le16_to_cpu(fp->fp_u_idx), IGU_INT_NOP, 1);
+		bnx2x_ack_sb(bp, fp->index, CSTORM_ID,
+			     le16_to_cpu(fp->fp_c_idx), IGU_INT_ENABLE, 1);
+	}
+
+	return work_done;
+}
+
+/* Called with netif_tx_lock.
+ * bnx2x_tx_int() runs without netif_tx_lock unless it needs to call
+ * netif_wake_queue().
+ */
+static int bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	struct bnx2x_fastpath *fp;
+	struct sw_tx_bd *tx_buf;
+	struct eth_tx_bd *tx_bd;
+	struct eth_tx_parse_bd *pbd = NULL;
+	u16 pkt_prod, bd_prod;
+	int nbd, fp_index = 0;
+	dma_addr_t mapping;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return NETDEV_TX_BUSY;
+#endif
+
+	fp_index = smp_processor_id() % (bp->num_queues);
+
+	fp = &bp->fp[fp_index];
+	if (unlikely(bnx2x_tx_avail(bp->fp) <
+					(skb_shinfo(skb)->nr_frags + 3))) {
+		bp->slowpath->eth_stats.driver_xoff++,
+		netif_stop_queue(dev);
+		BNX2X_ERR("BUG! Tx ring full when queue awake!\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	/*
+	This is a bit ugly. First we use one BD which we mark as start,
+	then for TSO or xsum we have a parsing info BD,
+	and only then we have the rest of the TSO bds.
+	(don't forget to mark the last one as last,
+	and to unmap only AFTER you write to the BD ...)
+	I would like to thank DovH for this mess.
+	*/
+
+	pkt_prod = fp->tx_pkt_prod++;
+	bd_prod = fp->tx_bd_prod;
+	bd_prod = TX_BD(bd_prod);
+
+	/* get a tx_buff and first bd */
+	tx_buf = &fp->tx_buf_ring[TX_BD(pkt_prod)];
+	tx_bd = &fp->tx_desc_ring[bd_prod];
+
+	tx_bd->bd_flags.as_bitfield = ETH_TX_BD_FLAGS_START_BD;
+	tx_bd->general_data = (UNICAST_ADDRESS <<
+			       ETH_TX_BD_ETH_ADDR_TYPE_SHIFT);
+	tx_bd->general_data |= 1; /* header nbd */
+
+	/* remeber the first bd of the packet */
+	tx_buf->first_bd = bd_prod;
+
+	DP(NETIF_MSG_TX_QUEUED,
+	   "sending pkt %u @%p  next_idx %u  bd %u @%p\n",
+	   pkt_prod, tx_buf, fp->tx_pkt_prod, bd_prod, tx_bd);
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		struct iphdr *iph = ip_hdr(skb);
+		u8 len;
+
+		tx_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_IP_CSUM;
+
+		/* turn on parsing and get a bd */
+		bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
+		pbd = (void *)&fp->tx_desc_ring[bd_prod];
+		len = ((u8 *)iph - (u8 *)skb->data) / 2;
+
+		/* for now NS flag is not used in Linux */
+		pbd->global_data = (len |
+				    ((skb->protocol == ETH_P_8021Q) <<
+				     ETH_TX_PARSE_BD_LLC_SNAP_EN_SHIFT));
+		pbd->ip_hlen = ip_hdrlen(skb) / 2;
+		pbd->total_hlen = cpu_to_le16(len + pbd->ip_hlen);
+		if (iph->protocol == IPPROTO_TCP) {
+			struct tcphdr *th = tcp_hdr(skb);
+
+			tx_bd->bd_flags.as_bitfield |=
+						ETH_TX_BD_FLAGS_TCP_CSUM;
+			pbd->tcp_flags = htonl(tcp_flag_word(skb)) & 0xFFFF;
+			pbd->total_hlen += cpu_to_le16(tcp_hdrlen(skb) / 2);
+			pbd->tcp_pseudo_csum = swab16(th->check);
+
+		} else if (iph->protocol == IPPROTO_UDP) {
+			struct udphdr *uh = udp_hdr(skb);
+
+			tx_bd->bd_flags.as_bitfield |=
+						ETH_TX_BD_FLAGS_TCP_CSUM;
+			pbd->total_hlen += cpu_to_le16(4);
+			pbd->global_data |= ETH_TX_PARSE_BD_CS_ANY_FLG;
+			pbd->cs_offset = 5; /* 10 >> 1 */
+			pbd->tcp_pseudo_csum = 0;
+			/* HW bug: we need to subtract 10 bytes before the
+			 * UDP header from the csum
+			 */
+			uh->check = (u16) ~csum_fold(csum_sub(uh->check,
+				csum_partial(((u8 *)(uh)-10), 10, 0)));
+		}
+	}
+
+	if ((bp->vlgrp != NULL) && vlan_tx_tag_present(skb)) {
+		tx_bd->vlan = cpu_to_le16(vlan_tx_tag_get(skb));
+		tx_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_VLAN_TAG;
+	} else {
+		tx_bd->vlan = cpu_to_le16(pkt_prod);
+	}
+
+	mapping = pci_map_single(bp->pdev, skb->data,
+				 skb->len, PCI_DMA_TODEVICE);
+
+	tx_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
+	tx_bd->addr_lo = cpu_to_le32(U64_LO(mapping));
+	nbd = skb_shinfo(skb)->nr_frags + ((pbd == NULL)? 1 : 2);
+	tx_bd->nbd = cpu_to_le16(nbd);
+	tx_bd->nbytes = cpu_to_le16(skb_headlen(skb));
+
+	DP(NETIF_MSG_TX_QUEUED, "first bd @%p  addr (%x:%x)  nbd %d"
+	   "  nbytes %d  flags %x  vlan %u\n",
+	   tx_bd, tx_bd->addr_hi, tx_bd->addr_lo, tx_bd->nbd,
+	   tx_bd->nbytes, tx_bd->bd_flags.as_bitfield, tx_bd->vlan);
+
+	if (skb_shinfo(skb)->gso_size &&
+	    (skb->len > (bp->dev->mtu + ETH_HLEN))) {
+		int hlen = 2 * le32_to_cpu(pbd->total_hlen);
+
+		DP(NETIF_MSG_TX_QUEUED,
+		   "TSO packet len %d  hlen %d  total len %d  tso size %d\n",
+		   skb->len, hlen, skb_headlen(skb),
+		   skb_shinfo(skb)->gso_size);
+
+		tx_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_SW_LSO;
+
+		if (tx_bd->nbytes > cpu_to_le16(hlen)) {
+			/* we split the first bd into headers and data bds
+			 * to ease the pain of our fellow micocode engineers
+			 * we use one mapping for both bds
+			 * So far this has only been observed to happen
+			 * in Other Operating Systems(TM)
+			 */
+
+			/* first fix first bd */
+			nbd++;
+			tx_bd->nbd = cpu_to_le16(nbd);
+			tx_bd->nbytes = cpu_to_le16(hlen);
+
+			/* we only print this as an error
+			 * because we don't think this will ever happen.
+			 */
+			BNX2X_ERR("TSO split header size is %d (%x:%x)"
+				  "  nbd %d\n", tx_bd->nbytes, tx_bd->addr_hi,
+				  tx_bd->addr_lo, tx_bd->nbd);
+
+			/* now get a new data bd
+			 * (after the pbd) and fill it */
+			bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
+			tx_bd = &fp->tx_desc_ring[bd_prod];
+
+			tx_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
+			tx_bd->addr_lo = cpu_to_le32(U64_LO(mapping) + hlen);
+			tx_bd->nbytes = cpu_to_le16(skb_headlen(skb) - hlen);
+			tx_bd->vlan = cpu_to_le16(pkt_prod);
+			/* this marks the bd
+			 * as one that has no individual mapping
+			 * the FW ignors this flag in a bd not maked start
+			 */
+			tx_bd->bd_flags.as_bitfield = ETH_TX_BD_FLAGS_SW_LSO;
+			DP(NETIF_MSG_TX_QUEUED,
+			   "TSO split data size is %d (%x:%x)\n",
+			   tx_bd->nbytes, tx_bd->addr_hi, tx_bd->addr_lo);
+		}
+
+		if (!pbd) {
+			/* supposed to be unreached
+			 * (and therefore not handled properly...)
+			 */
+			BNX2X_ERR("LSO with no PBD\n");
+			BUG();
+		}
+
+		pbd->lso_mss = cpu_to_le16(skb_shinfo(skb)->gso_size);
+		pbd->tcp_send_seq = swab32(tcp_hdr(skb)->seq);
+		pbd->ip_id = swab16(ip_hdr(skb)->id);
+		pbd->tcp_pseudo_csum =
+				swab16(~csum_tcpudp_magic(ip_hdr(skb)->saddr,
+							  ip_hdr(skb)->daddr,
+							  0, IPPROTO_TCP, 0));
+		pbd->global_data |= ETH_TX_PARSE_BD_PSEUDO_CS_WITHOUT_LEN;
+	}
+
+	{
+		int i;
+
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+			bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
+			tx_bd = &fp->tx_desc_ring[bd_prod];
+
+			mapping = pci_map_page(bp->pdev, frag->page,
+					       frag->page_offset,
+					       frag->size, PCI_DMA_TODEVICE);
+
+			tx_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
+			tx_bd->addr_lo = cpu_to_le32(U64_LO(mapping));
+			tx_bd->nbytes = cpu_to_le16(frag->size);
+			tx_bd->vlan = cpu_to_le16(pkt_prod);
+			tx_bd->bd_flags.as_bitfield = 0;
+			DP(NETIF_MSG_TX_QUEUED, "frag %d  bd @%p"
+			   "  addr (%x:%x)  nbytes %d  flags %x\n",
+			   i, tx_bd, tx_bd->addr_hi, tx_bd->addr_lo,
+			   tx_bd->nbytes, tx_bd->bd_flags.as_bitfield);
+		} /* for */
+	}
+
+	/* now at last mark the bd as the last bd */
+	tx_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_END_BD;
+
+	DP(NETIF_MSG_TX_QUEUED, "last bd @%p  flags %x\n",
+	   tx_bd, tx_bd->bd_flags.as_bitfield);
+
+	tx_buf->skb = skb;
+
+	bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
+
+	/* now send a tx doorbell, counting the next bd
+	 * if the packet contains or ends with it
+	 */
+	if (TX_BD_POFF(bd_prod) < nbd)
+		nbd++;
+
+	if (pbd)
+		DP(NETIF_MSG_TX_QUEUED,
+		   "PBD @%p  ip_data %x  ip_hlen %u  ip_id %u  lso_mss %u"
+		   "  tcp_flags %x  xsum %x  seq %u  hlen %u\n",
+		   pbd, pbd->global_data, pbd->ip_hlen, pbd->ip_id,
+		   pbd->lso_mss, pbd->tcp_flags, pbd->tcp_pseudo_csum,
+		   pbd->tcp_send_seq, pbd->total_hlen);
+
+	DP(NETIF_MSG_TX_QUEUED, "doorbell: nbd %u  bd %d\n", nbd, bd_prod);
+
+	fp->hw_tx_prods->bds_prod += cpu_to_le16(nbd);
+	mb(); /* FW restriction: must not reorder writing nbd and packets */
+	fp->hw_tx_prods->packets_prod += cpu_to_le32(1);
+	DOORBELL(bp, fp_index, 0);
+
+	mmiowb();
+
+	fp->tx_bd_prod = bd_prod;
+	dev->trans_start = jiffies;
+
+	if (unlikely(bnx2x_tx_avail(fp) < MAX_SKB_FRAGS + 3)) {
+		netif_stop_queue(dev);
+		bp->slowpath->eth_stats.driver_xoff++;
+		if (bnx2x_tx_avail(fp) >= MAX_SKB_FRAGS + 3)
+			netif_wake_queue(dev);
+	}
+	fp->tx_pkt++;
+
+	return NETDEV_TX_OK;
+}
+
+static struct net_device_stats *bnx2x_get_stats(struct net_device *dev)
+{
+	return &dev->stats;
+}
+
+/* Called with rtnl_lock */
+static int bnx2x_open(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	bnx2x_set_power_state(bp, PCI_D0);
+
+	return bnx2x_nic_load(bp, 1);
+}
+
+/* Called with rtnl_lock */
+static int bnx2x_close(struct net_device *dev)
+{
+	int rc;
+	struct bnx2x *bp = netdev_priv(dev);
+
+	/* Unload the driver, release IRQs */
+	rc = bnx2x_nic_unload(bp, 1);
+	if (rc) {
+		BNX2X_ERR("bnx2x_nic_unload failed: %d\n", rc);
+		return rc;
+	}
+	bnx2x_set_power_state(bp, PCI_D3hot);
+
+	return 0;
+}
+
+/* Called with rtnl_lock */
+static int bnx2x_change_mac_addr(struct net_device *dev, void *p)
+{
+	struct sockaddr *addr = p;
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EINVAL;
+
+	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	if (netif_running(dev))
+		bnx2x_set_mac_addr(bp);
+
+	return 0;
+}
+
+/* Called with rtnl_lock */
+static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct mii_ioctl_data *data = if_mii(ifr);
+	struct bnx2x *bp = netdev_priv(dev);
+	int err;
+
+	switch (cmd) {
+	case SIOCGMIIPHY:
+		data->phy_id = bp->phy_addr;
+
+		/* fallthru */
+	case SIOCGMIIREG: {
+		u32 mii_regval;
+
+		spin_lock_bh(&bp->phy_lock);
+		if (bp->state == BNX2X_STATE_OPEN) {
+			err = bnx2x_mdio22_read(bp, data->reg_num & 0x1f,
+						&mii_regval);
+
+			data->val_out = mii_regval;
+		} else {
+			err = -EAGAIN;
+		}
+		spin_unlock_bh(&bp->phy_lock);
+		return err;
+	}
+
+	case SIOCSMIIREG:
+		if (!capable(CAP_NET_ADMIN))
+			return -EPERM;
+
+		spin_lock_bh(&bp->phy_lock);
+		if (bp->state == BNX2X_STATE_OPEN) {
+			err = bnx2x_mdio22_write(bp, data->reg_num & 0x1f,
+						 data->val_in);
+		} else {
+			err = -EAGAIN;
+		}
+		spin_unlock_bh(&bp->phy_lock);
+		return err;
+
+	default:
+		/* do nothing */
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+/* Called with rtnl_lock */
+static int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	if ((new_mtu > ETH_MAX_JUMBO_PACKET_SIZE) ||
+	    ((new_mtu + ETH_HLEN) < ETH_MIN_PACKET_SIZE))
+		return -EINVAL;
+
+	/* This does not race with packet allocation
+	 * because the actuall alloc size is
+	 * only updated as part of load
+	 */
+	dev->mtu = new_mtu;
+
+	if (netif_running(dev)) {
+		bnx2x_nic_unload(bp, 0);
+		bnx2x_nic_load(bp, 0);
+	}
+	return 0;
+}
+
+static void bnx2x_tx_timeout(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (!bp->panic)
+		bnx2x_panic();
+#endif
+	/* This allows the netif to be shutdown gracefully before resetting */
+	schedule_work(&bp->reset_task);
+}
+
+#ifdef BCM_VLAN
+/* Called with rtnl_lock */
+static void bnx2x_vlan_rx_register(struct net_device *dev,
+				   struct vlan_group *vlgrp)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	bp->vlgrp = vlgrp;
+	if (netif_running(dev))
+		bnx2x_set_rx_mode(dev);
+}
+#endif
+
+#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
+static void poll_bnx2x(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	disable_irq(bp->pdev->irq);
+	bnx2x_interrupt(bp->pdev->irq, dev);
+	enable_irq(bp->pdev->irq);
+}
+#endif
+
+static void bnx2x_reset_task(struct work_struct *work)
+{
+	struct bnx2x *bp = container_of(work, struct bnx2x, reset_task);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	BNX2X_ERR("reset task called but STOP_ON_ERROR defined"
+		  " so reset not done to allow debug dump,\n"
+	 KERN_ERR " you will need to reboot when done\n");
+	return;
+#endif
+
+	if (!netif_running(bp->dev))
+		return;
+
+	bp->in_reset_task = 1;
+
+	bnx2x_netif_stop(bp);
+
+	bnx2x_nic_unload(bp, 0);
+	bnx2x_nic_load(bp, 0);
+
+	bp->in_reset_task = 0;
+}
+
+static int __devinit bnx2x_init_board(struct pci_dev *pdev,
+				      struct net_device *dev)
+{
+	struct bnx2x *bp;
+	int rc;
+
+	SET_NETDEV_DEV(dev, &pdev->dev);
+	bp = netdev_priv(dev);
+
+	bp->flags = 0;
+	bp->port = PCI_FUNC(pdev->devfn);
+
+	rc = pci_enable_device(pdev);
+	if (rc) {
+		printk(KERN_ERR PFX "Cannot enable PCI device, aborting\n");
+		goto err_out;
+	}
+
+	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
+		printk(KERN_ERR PFX "Cannot find PCI device base address,"
+		       " aborting\n");
+		rc = -ENODEV;
+		goto err_out_disable;
+	}
+
+	if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
+		printk(KERN_ERR PFX "Cannot find second PCI device"
+		       " base address, aborting\n");
+		rc = -ENODEV;
+		goto err_out_disable;
+	}
+
+	rc = pci_request_regions(pdev, DRV_MODULE_NAME);
+	if (rc) {
+		printk(KERN_ERR PFX "Cannot obtain PCI resources,"
+		       " aborting\n");
+		goto err_out_disable;
+	}
+
+	pci_set_master(pdev);
+
+	bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
+	if (bp->pm_cap == 0) {
+		printk(KERN_ERR PFX "Cannot find power management"
+		       " capability, aborting\n");
+		rc = -EIO;
+		goto err_out_release;
+	}
+
+	bp->pcie_cap = pci_find_capability(pdev, PCI_CAP_ID_EXP);
+	if (bp->pcie_cap == 0) {
+		printk(KERN_ERR PFX "Cannot find PCI Express capability,"
+		       " aborting\n");
+		rc = -EIO;
+		goto err_out_release;
+	}
+
+	if (pci_set_dma_mask(pdev, DMA_64BIT_MASK) == 0) {
+		bp->flags |= USING_DAC_FLAG;
+		if (pci_set_consistent_dma_mask(pdev, DMA_64BIT_MASK) != 0) {
+			printk(KERN_ERR PFX "pci_set_consistent_dma_mask"
+			       " failed, aborting\n");
+			rc = -EIO;
+			goto err_out_release;
+		}
+
+	} else if (pci_set_dma_mask(pdev, DMA_32BIT_MASK) != 0) {
+		printk(KERN_ERR PFX "System does not support DMA,"
+		       " aborting\n");
+		rc = -EIO;
+		goto err_out_release;
+	}
+
+	bp->dev = dev;
+	bp->pdev = pdev;
+
+	spin_lock_init(&bp->phy_lock);
+
+	bp->in_reset_task = 0;
+
+	INIT_WORK(&bp->reset_task, bnx2x_reset_task);
+	INIT_WORK(&bp->sp_task, bnx2x_sp_task);
+
+	dev->base_addr = dev->mem_start = pci_resource_start(pdev, 0);
+	dev->mem_end = pci_resource_end(pdev, 0);
+
+	dev->irq = pdev->irq;
+
+	bp->regview = ioremap_nocache(dev->base_addr,
+				      pci_resource_len(pdev, 0));
+	if (!bp->regview) {
+		printk(KERN_ERR PFX "Cannot map register space, aborting\n");
+		rc = -ENOMEM;
+		goto err_out_release;
+	}
+
+	bp->doorbells = ioremap_nocache(pci_resource_start(pdev , 2),
+					pci_resource_len(pdev, 2));
+	if (!bp->doorbells) {
+		printk(KERN_ERR PFX "Cannot map doorbell space, aborting\n");
+		rc = -ENOMEM;
+		goto err_out_unmap;
+	}
+
+	bnx2x_set_power_state(bp, PCI_D0);
+
+	bnx2x_get_hwinfo(bp);
+
+	if (CHIP_REV(bp) == CHIP_REV_FPGA) {
+		printk(KERN_ERR PFX "FPGA detacted. MCP disabled,"
+		       " will only init first device\n");
+		onefunc = 1;
+		nomcp = 1;
+	}
+
+	if (nomcp) {
+		printk(KERN_ERR PFX "MCP disabled, will only"
+		       " init first device\n");
+		onefunc = 1;
+	}
+
+	if (onefunc && bp->port) {
+		printk(KERN_ERR PFX "Second device disabled, exiting\n");
+		rc = -ENODEV;
+		goto err_out_unmap;
+	}
+
+	bp->tx_ring_size = MAX_TX_AVAIL;
+	bp->rx_ring_size = MAX_RX_AVAIL;
+
+	bp->rx_csum = 1;
+
+	bp->rx_offset = 0;
+
+	bp->tx_quick_cons_trip_int = 0xff;
+	bp->tx_quick_cons_trip = 0xff;
+	bp->tx_ticks_int = 50;
+	bp->tx_ticks = 50;
+
+	bp->rx_quick_cons_trip_int = 0xff;
+	bp->rx_quick_cons_trip = 0xff;
+	bp->rx_ticks_int = 25;
+	bp->rx_ticks = 25;
+
+	bp->stats_ticks = 1000000 & 0xffff00;
+
+	bp->timer_interval = HZ;
+	bp->current_interval = (poll ? poll : HZ);
+
+	init_timer(&bp->timer);
+	bp->timer.expires = jiffies + bp->current_interval;
+	bp->timer.data = (unsigned long) bp;
+	bp->timer.function = bnx2x_timer;
+
+	return 0;
+
+err_out_unmap:
+	if (bp->regview) {
+		iounmap(bp->regview);
+		bp->regview = NULL;
+	}
+
+	if (bp->doorbells) {
+		iounmap(bp->doorbells);
+		bp->doorbells = NULL;
+	}
+
+err_out_release:
+	pci_release_regions(pdev);
+
+err_out_disable:
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+
+err_out:
+	return rc;
+}
+
+static int __devinit bnx2x_init_one(struct pci_dev *pdev,
+				    const struct pci_device_id *ent)
+{
+	static int version_printed;
+	struct net_device *dev = NULL;
+	struct bnx2x *bp;
+	int rc, i;
+	int port = PCI_FUNC(pdev->devfn);
+
+	if (version_printed++ == 0)
+		printk(KERN_INFO "%s", version);
+
+	/* dev zeroed in init_etherdev */
+	dev = alloc_etherdev(sizeof(*bp));
+	if (!dev)
+		return -ENOMEM;
+
+	netif_carrier_off(dev);
+
+	bp = netdev_priv(dev);
+	bp->msglevel = debug;
+
+	if (port && onefunc) {
+		printk(KERN_ERR PFX "second function disabled. exiting\n");
+		return 0;
+	}
+
+	rc = bnx2x_init_board(pdev, dev);
+	if (rc < 0) {
+		free_netdev(dev);
+		return rc;
+	}
+
+	dev->hard_start_xmit = bnx2x_start_xmit;
+	dev->watchdog_timeo = TX_TIMEOUT;
+
+	dev->get_stats = bnx2x_get_stats;
+	dev->ethtool_ops = &bnx2x_ethtool_ops;
+	dev->open = bnx2x_open;
+	dev->stop = bnx2x_close;
+	dev->set_multicast_list = bnx2x_set_rx_mode;
+	dev->set_mac_address = bnx2x_change_mac_addr;
+	dev->do_ioctl = bnx2x_ioctl;
+	dev->change_mtu = bnx2x_change_mtu;
+	dev->tx_timeout = bnx2x_tx_timeout;
+#ifdef BCM_VLAN
+	dev->vlan_rx_register = bnx2x_vlan_rx_register;
+#endif
+#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
+	dev->poll_controller = poll_bnx2x;
+#endif
+	dev->features |= NETIF_F_SG;
+	if (bp->flags & USING_DAC_FLAG)
+		dev->features |= NETIF_F_HIGHDMA;
+	dev->features |= NETIF_F_IP_CSUM;
+#ifdef BCM_VLAN
+	dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
+#endif
+	dev->features |= NETIF_F_TSO | NETIF_F_TSO_ECN;
+
+	rc = register_netdev(dev);
+	if (rc) {
+		printk(KERN_ERR PFX "Cannot register net device\n");
+		if (bp->regview)
+			iounmap(bp->regview);
+		if (bp->doorbells)
+			iounmap(bp->doorbells);
+		pci_release_regions(pdev);
+		pci_disable_device(pdev);
+		pci_set_drvdata(pdev, NULL);
+		free_netdev(dev);
+		return rc;
+	}
+
+	pci_set_drvdata(pdev, dev);
+
+	bp->name = board_info[ent->driver_data].name;
+	printk(KERN_INFO "%s: %s (%c%d) PCI%s %s %dMHz "
+	       "found at mem %lx, IRQ %d, ",
+	       dev->name, bp->name,
+	       ((CHIP_ID(bp) & 0xf000) >> 12) + 'A',
+	       ((CHIP_ID(bp) & 0x0ff0) >> 4),
+	       ((bp->flags & PCIX_FLAG) ? "-X" : ""),
+	       ((bp->flags & PCI_32BIT_FLAG) ? "32-bit" : "64-bit"),
+	       bp->bus_speed_mhz,
+	       dev->base_addr,
+	       bp->pdev->irq);
+
+	printk("node addr ");
+	for (i = 0; i < 6; i++)
+		printk("%2.2x", dev->dev_addr[i]);
+	printk("\n");
+
+	return 0;
+}
+
+static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+	struct bnx2x *bp = netdev_priv(dev);
+
+	flush_scheduled_work();
+	/*tasklet_kill(&bp->sp_task);*/
+	unregister_netdev(dev);
+
+	if (bp->regview)
+		iounmap(bp->regview);
+
+	if (bp->doorbells)
+		iounmap(bp->doorbells);
+
+	free_netdev(dev);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+}
+
+static int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+	struct bnx2x *bp = netdev_priv(dev);
+	int rc;
+
+	if (!netif_running(dev))
+		return 0;
+
+	rc = bnx2x_nic_unload(bp, 0);
+	if (!rc)
+		return rc;
+
+	netif_device_detach(dev);
+	pci_save_state(pdev);
+
+	bnx2x_set_power_state(bp, pci_choose_state(pdev, state));
+	return 0;
+}
+
+static int bnx2x_resume(struct pci_dev *pdev)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+	struct bnx2x *bp = netdev_priv(dev);
+	int rc;
+
+	if (!netif_running(dev))
+		return 0;
+
+	pci_restore_state(pdev);
+
+	bnx2x_set_power_state(bp, PCI_D0);
+	netif_device_attach(dev);
+
+	rc = bnx2x_nic_load(bp, 0);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static struct pci_driver bnx2x_pci_driver = {
+	.name       = DRV_MODULE_NAME,
+	.id_table   = bnx2x_pci_tbl,
+	.probe      = bnx2x_init_one,
+	.remove     = __devexit_p(bnx2x_remove_one),
+	.suspend    = bnx2x_suspend,
+	.resume     = bnx2x_resume,
+};
+
+static int __init bnx2x_init(void)
+{
+	return pci_register_driver(&bnx2x_pci_driver);
+}
+
+static void __exit bnx2x_cleanup(void)
+{
+	pci_unregister_driver(&bnx2x_pci_driver);
+}
+
+module_init(bnx2x_init);
+module_exit(bnx2x_cleanup);
+
diff --git a/drivers/net/bnx2x.h b/drivers/net/bnx2x.h
new file mode 100644
index 000000000000..4f7ae6f77452
--- /dev/null
+++ b/drivers/net/bnx2x.h
@@ -0,0 +1,1071 @@
+/* bnx2x.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * Written by: Eliezer Tamir <eliezert@broadcom.com>
+ * Based on code from Michael Chan's bnx2 driver
+ */
+
+#ifndef BNX2X_H
+#define BNX2X_H
+
+/* error/debug prints */
+
+#define DRV_MODULE_NAME 	"bnx2x"
+#define PFX DRV_MODULE_NAME     ": "
+
+/* for messages that are currently off */
+#define BNX2X_MSG_OFF   		0
+#define BNX2X_MSG_MCP   		0x10000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_STATS 		0x20000 /* was: NETIF_MSG_TIMER */
+#define NETIF_MSG_NVM   		0x40000 /* was: NETIF_MSG_HW */
+#define NETIF_MSG_DMAE  		0x80000 /* was: NETIF_MSG_HW */
+
+#define DP_LEVEL			KERN_NOTICE     /* was: KERN_DEBUG */
+
+/* regular debug print */
+#define DP(__mask, __fmt, __args...) do { \
+	if (bp->msglevel & (__mask)) \
+		printk(DP_LEVEL "[%s:%d(%s)]" __fmt, __FUNCTION__, \
+		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+	} while (0)
+
+/* for errors (never masked) */
+#define BNX2X_ERR(__fmt, __args...) do { \
+	printk(KERN_ERR "[%s:%d(%s)]" __fmt, __FUNCTION__, \
+		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+	} while (0)
+
+/* before we have a dev->name use dev_info() */
+#define BNX2X_DEV_INFO(__fmt, __args...) do { \
+	if (bp->msglevel & NETIF_MSG_PROBE) \
+		dev_info(&bp->pdev->dev, __fmt, ##__args); \
+	} while (0)
+
+
+#ifdef BNX2X_STOP_ON_ERROR
+#define bnx2x_panic() do { \
+		bp->panic = 1; \
+		BNX2X_ERR("driver assert\n"); \
+		bnx2x_disable_int(bp); \
+		bnx2x_panic_dump(bp); \
+	} while (0)
+#else
+#define bnx2x_panic() do { \
+		BNX2X_ERR("driver assert\n"); \
+		bnx2x_panic_dump(bp); \
+	} while (0)
+#endif
+
+
+#define U64_LO(x)       		(((u64)x) & 0xffffffff)
+#define U64_HI(x)       		(((u64)x) >> 32)
+#define HILO_U64(hi, lo)		(((u64)hi << 32) + lo)
+
+
+#define REG_ADDR(bp, offset)    	(bp->regview + offset)
+
+#define REG_RD(bp, offset)      	readl(REG_ADDR(bp, offset))
+#define REG_RD8(bp, offset)     	readb(REG_ADDR(bp, offset))
+#define REG_RD64(bp, offset)    	readq(REG_ADDR(bp, offset))
+
+#define REG_WR(bp, offset, val) 	writel((u32)val, REG_ADDR(bp, offset))
+#define REG_WR8(bp, offset, val)	writeb((u8)val, REG_ADDR(bp, offset))
+#define REG_WR16(bp, offset, val)       writew((u16)val, REG_ADDR(bp, offset))
+#define REG_WR32(bp, offset, val)       REG_WR(bp, offset, val)
+
+#define REG_RD_IND(bp, offset)  	bnx2x_reg_rd_ind(bp, offset)
+#define REG_WR_IND(bp, offset, val)     bnx2x_reg_wr_ind(bp, offset, val)
+
+#define REG_WR_DMAE(bp, offset, val, len32) \
+	do { \
+		memcpy(bnx2x_sp(bp, wb_data[0]), val, len32 * 4); \
+		bnx2x_write_dmae(bp, bnx2x_sp_mapping(bp, wb_data), \
+				 offset, len32); \
+	} while (0)
+
+#define SHMEM_RD(bp, type) \
+	REG_RD(bp, bp->shmem_base + offsetof(struct shmem_region, type))
+#define SHMEM_WR(bp, type, val) \
+	REG_WR(bp, bp->shmem_base + offsetof(struct shmem_region, type), val)
+
+#define NIG_WR(reg, val)	REG_WR(bp, reg, val)
+#define EMAC_WR(reg, val)       REG_WR(bp, emac_base + reg, val)
+#define BMAC_WR(reg, val)       REG_WR(bp, GRCBASE_NIG + bmac_addr + reg, val)
+
+
+#define for_each_queue(bp, var) for (var = 0; var < bp->num_queues; var++)
+
+#define for_each_nondefault_queue(bp, var) \
+				for (var = 1; var < bp->num_queues; var++)
+#define is_multi(bp)    	(bp->num_queues > 1)
+
+
+struct regp {
+	u32 lo;
+	u32 hi;
+};
+
+struct bmac_stats {
+	struct regp tx_gtpkt;
+	struct regp tx_gtxpf;
+	struct regp tx_gtfcs;
+	struct regp tx_gtmca;
+	struct regp tx_gtgca;
+	struct regp tx_gtfrg;
+	struct regp tx_gtovr;
+	struct regp tx_gt64;
+	struct regp tx_gt127;
+	struct regp tx_gt255;   /* 10 */
+	struct regp tx_gt511;
+	struct regp tx_gt1023;
+	struct regp tx_gt1518;
+	struct regp tx_gt2047;
+	struct regp tx_gt4095;
+	struct regp tx_gt9216;
+	struct regp tx_gt16383;
+	struct regp tx_gtmax;
+	struct regp tx_gtufl;
+	struct regp tx_gterr;   /* 20 */
+	struct regp tx_gtbyt;
+
+	struct regp rx_gr64;
+	struct regp rx_gr127;
+	struct regp rx_gr255;
+	struct regp rx_gr511;
+	struct regp rx_gr1023;
+	struct regp rx_gr1518;
+	struct regp rx_gr2047;
+	struct regp rx_gr4095;
+	struct regp rx_gr9216;  /* 30 */
+	struct regp rx_gr16383;
+	struct regp rx_grmax;
+	struct regp rx_grpkt;
+	struct regp rx_grfcs;
+	struct regp rx_grmca;
+	struct regp rx_grbca;
+	struct regp rx_grxcf;
+	struct regp rx_grxpf;
+	struct regp rx_grxuo;
+	struct regp rx_grjbr;   /* 40 */
+	struct regp rx_grovr;
+	struct regp rx_grflr;
+	struct regp rx_grmeg;
+	struct regp rx_grmeb;
+	struct regp rx_grbyt;
+	struct regp rx_grund;
+	struct regp rx_grfrg;
+	struct regp rx_grerb;
+	struct regp rx_grfre;
+	struct regp rx_gripj;   /* 50 */
+};
+
+struct emac_stats {
+	u32 rx_ifhcinoctets     		   ;
+	u32 rx_ifhcinbadoctets  		   ;
+	u32 rx_etherstatsfragments      	   ;
+	u32 rx_ifhcinucastpkts  		   ;
+	u32 rx_ifhcinmulticastpkts      	   ;
+	u32 rx_ifhcinbroadcastpkts      	   ;
+	u32 rx_dot3statsfcserrors       	   ;
+	u32 rx_dot3statsalignmenterrors 	   ;
+	u32 rx_dot3statscarriersenseerrors         ;
+	u32 rx_xonpauseframesreceived   	   ;    /* 10 */
+	u32 rx_xoffpauseframesreceived  	   ;
+	u32 rx_maccontrolframesreceived 	   ;
+	u32 rx_xoffstateentered 		   ;
+	u32 rx_dot3statsframestoolong   	   ;
+	u32 rx_etherstatsjabbers		   ;
+	u32 rx_etherstatsundersizepkts  	   ;
+	u32 rx_etherstatspkts64octets   	   ;
+	u32 rx_etherstatspkts65octetsto127octets   ;
+	u32 rx_etherstatspkts128octetsto255octets  ;
+	u32 rx_etherstatspkts256octetsto511octets  ;    /* 20 */
+	u32 rx_etherstatspkts512octetsto1023octets ;
+	u32 rx_etherstatspkts1024octetsto1522octets;
+	u32 rx_etherstatspktsover1522octets        ;
+
+	u32 rx_falsecarriererrors       	   ;
+
+	u32 tx_ifhcoutoctets    		   ;
+	u32 tx_ifhcoutbadoctets 		   ;
+	u32 tx_etherstatscollisions     	   ;
+	u32 tx_outxonsent       		   ;
+	u32 tx_outxoffsent      		   ;
+	u32 tx_flowcontroldone  		   ;    /* 30 */
+	u32 tx_dot3statssinglecollisionframes      ;
+	u32 tx_dot3statsmultiplecollisionframes    ;
+	u32 tx_dot3statsdeferredtransmissions      ;
+	u32 tx_dot3statsexcessivecollisions        ;
+	u32 tx_dot3statslatecollisions  	   ;
+	u32 tx_ifhcoutucastpkts 		   ;
+	u32 tx_ifhcoutmulticastpkts     	   ;
+	u32 tx_ifhcoutbroadcastpkts     	   ;
+	u32 tx_etherstatspkts64octets   	   ;
+	u32 tx_etherstatspkts65octetsto127octets   ;    /* 40 */
+	u32 tx_etherstatspkts128octetsto255octets  ;
+	u32 tx_etherstatspkts256octetsto511octets  ;
+	u32 tx_etherstatspkts512octetsto1023octets ;
+	u32 tx_etherstatspkts1024octetsto1522octet ;
+	u32 tx_etherstatspktsover1522octets        ;
+	u32 tx_dot3statsinternalmactransmiterrors  ;    /* 46 */
+};
+
+union mac_stats {
+	struct emac_stats emac;
+	struct bmac_stats bmac;
+};
+
+struct nig_stats {
+	u32 brb_discard;
+	u32 brb_packet;
+	u32 brb_truncate;
+	u32 flow_ctrl_discard;
+	u32 flow_ctrl_octets;
+	u32 flow_ctrl_packet;
+	u32 mng_discard;
+	u32 mng_octet_inp;
+	u32 mng_octet_out;
+	u32 mng_packet_inp;
+	u32 mng_packet_out;
+	u32 pbf_octets;
+	u32 pbf_packet;
+	u32 safc_inp;
+	u32 done;
+	u32 pad;
+};
+
+struct bnx2x_eth_stats {
+	u32 pad;	/* to make long counters u64 aligned */
+	u32 mac_stx_start;
+	u32 total_bytes_received_hi;
+	u32 total_bytes_received_lo;
+	u32 total_bytes_transmitted_hi;
+	u32 total_bytes_transmitted_lo;
+	u32 total_unicast_packets_received_hi;
+	u32 total_unicast_packets_received_lo;
+	u32 total_multicast_packets_received_hi;
+	u32 total_multicast_packets_received_lo;
+	u32 total_broadcast_packets_received_hi;
+	u32 total_broadcast_packets_received_lo;
+	u32 total_unicast_packets_transmitted_hi;
+	u32 total_unicast_packets_transmitted_lo;
+	u32 total_multicast_packets_transmitted_hi;
+	u32 total_multicast_packets_transmitted_lo;
+	u32 total_broadcast_packets_transmitted_hi;
+	u32 total_broadcast_packets_transmitted_lo;
+	u32 crc_receive_errors;
+	u32 alignment_errors;
+	u32 false_carrier_detections;
+	u32 runt_packets_received;
+	u32 jabber_packets_received;
+	u32 pause_xon_frames_received;
+	u32 pause_xoff_frames_received;
+	u32 pause_xon_frames_transmitted;
+	u32 pause_xoff_frames_transmitted;
+	u32 single_collision_transmit_frames;
+	u32 multiple_collision_transmit_frames;
+	u32 late_collision_frames;
+	u32 excessive_collision_frames;
+	u32 control_frames_received;
+	u32 frames_received_64_bytes;
+	u32 frames_received_65_127_bytes;
+	u32 frames_received_128_255_bytes;
+	u32 frames_received_256_511_bytes;
+	u32 frames_received_512_1023_bytes;
+	u32 frames_received_1024_1522_bytes;
+	u32 frames_received_1523_9022_bytes;
+	u32 frames_transmitted_64_bytes;
+	u32 frames_transmitted_65_127_bytes;
+	u32 frames_transmitted_128_255_bytes;
+	u32 frames_transmitted_256_511_bytes;
+	u32 frames_transmitted_512_1023_bytes;
+	u32 frames_transmitted_1024_1522_bytes;
+	u32 frames_transmitted_1523_9022_bytes;
+	u32 valid_bytes_received_hi;
+	u32 valid_bytes_received_lo;
+	u32 error_runt_packets_received;
+	u32 error_jabber_packets_received;
+	u32 mac_stx_end;
+
+	u32 pad2;
+	u32 stat_IfHCInBadOctets_hi;
+	u32 stat_IfHCInBadOctets_lo;
+	u32 stat_IfHCOutBadOctets_hi;
+	u32 stat_IfHCOutBadOctets_lo;
+	u32 stat_Dot3statsFramesTooLong;
+	u32 stat_Dot3statsInternalMacTransmitErrors;
+	u32 stat_Dot3StatsCarrierSenseErrors;
+	u32 stat_Dot3StatsDeferredTransmissions;
+	u32 stat_FlowControlDone;
+	u32 stat_XoffStateEntered;
+
+	u32 x_total_sent_bytes_hi;
+	u32 x_total_sent_bytes_lo;
+	u32 x_total_sent_pkts;
+
+	u32 t_rcv_unicast_bytes_hi;
+	u32 t_rcv_unicast_bytes_lo;
+	u32 t_rcv_broadcast_bytes_hi;
+	u32 t_rcv_broadcast_bytes_lo;
+	u32 t_rcv_multicast_bytes_hi;
+	u32 t_rcv_multicast_bytes_lo;
+	u32 t_total_rcv_pkt;
+
+	u32 checksum_discard;
+	u32 packets_too_big_discard;
+	u32 no_buff_discard;
+	u32 ttl0_discard;
+	u32 mac_discard;
+	u32 mac_filter_discard;
+	u32 xxoverflow_discard;
+	u32 brb_truncate_discard;
+
+	u32 brb_discard;
+	u32 brb_packet;
+	u32 brb_truncate;
+	u32 flow_ctrl_discard;
+	u32 flow_ctrl_octets;
+	u32 flow_ctrl_packet;
+	u32 mng_discard;
+	u32 mng_octet_inp;
+	u32 mng_octet_out;
+	u32 mng_packet_inp;
+	u32 mng_packet_out;
+	u32 pbf_octets;
+	u32 pbf_packet;
+	u32 safc_inp;
+	u32 driver_xoff;
+	u32 number_of_bugs_found_in_stats_spec; /* just kidding */
+};
+
+#define MAC_STX_NA      		0xffffffff
+
+#ifdef BNX2X_MULTI
+#define MAX_CONTEXT     		16
+#else
+#define MAX_CONTEXT     		1
+#endif
+
+union cdu_context {
+	struct eth_context eth;
+	char pad[1024];
+};
+
+#define MAX_DMAE_C      		5
+
+/* DMA memory not used in fastpath */
+struct bnx2x_slowpath {
+	union cdu_context       	context[MAX_CONTEXT];
+	struct eth_stats_query  	fw_stats;
+	struct mac_configuration_cmd    mac_config;
+	struct mac_configuration_cmd    mcast_config;
+
+	/* used by dmae command executer */
+	struct dmae_command     	dmae[MAX_DMAE_C];
+
+	union mac_stats 		mac_stats;
+	struct nig_stats		nig;
+	struct bnx2x_eth_stats  	eth_stats;
+
+	u32     			wb_comp;
+#define BNX2X_WB_COMP_VAL       	0xe0d0d0ae
+	u32     			wb_data[4];
+};
+
+#define bnx2x_sp(bp, var)       	(&bp->slowpath->var)
+#define bnx2x_sp_check(bp, var) ((bp->slowpath) ? (&bp->slowpath->var) : NULL)
+#define bnx2x_sp_mapping(bp, var) \
+		(bp->slowpath_mapping + offsetof(struct bnx2x_slowpath, var))
+
+
+struct sw_rx_bd {
+	struct sk_buff  *skb;
+	DECLARE_PCI_UNMAP_ADDR(mapping)
+};
+
+struct sw_tx_bd {
+	struct sk_buff  *skb;
+	u16     	first_bd;
+};
+
+struct bnx2x_fastpath {
+
+	struct napi_struct      napi;
+
+	struct host_status_block *status_blk;
+	dma_addr_t      	status_blk_mapping;
+
+	struct eth_tx_db_data   *hw_tx_prods;
+	dma_addr_t      	tx_prods_mapping;
+
+	struct sw_tx_bd 	*tx_buf_ring;
+
+	struct eth_tx_bd	*tx_desc_ring;
+	dma_addr_t      	tx_desc_mapping;
+
+	struct sw_rx_bd 	*rx_buf_ring;
+
+	struct eth_rx_bd	*rx_desc_ring;
+	dma_addr_t      	rx_desc_mapping;
+
+	union eth_rx_cqe	*rx_comp_ring;
+	dma_addr_t      	rx_comp_mapping;
+
+	int     		state;
+#define BNX2X_FP_STATE_CLOSED   	0
+#define BNX2X_FP_STATE_IRQ      	0x80000
+#define BNX2X_FP_STATE_OPENING  	0x90000
+#define BNX2X_FP_STATE_OPEN     	0xa0000
+#define BNX2X_FP_STATE_HALTING  	0xb0000
+#define BNX2X_FP_STATE_HALTED   	0xc0000
+#define BNX2X_FP_STATE_DELETED  	0xd0000
+#define BNX2X_FP_STATE_CLOSE_IRQ	0xe0000
+
+	int     		index;
+
+	u16     		tx_pkt_prod;
+	u16     		tx_pkt_cons;
+	u16     		tx_bd_prod;
+	u16     		tx_bd_cons;
+	u16     		*tx_cons_sb;
+
+	u16     		fp_c_idx;
+	u16     		fp_u_idx;
+
+	u16     		rx_bd_prod;
+	u16     		rx_bd_cons;
+	u16     		rx_comp_prod;
+	u16     		rx_comp_cons;
+	u16     		*rx_cons_sb;
+
+	unsigned long   	tx_pkt,
+				rx_pkt,
+				rx_calls;
+
+	struct bnx2x    	*bp; /* parent */
+};
+
+#define bnx2x_fp(bp, nr, var)   	(bp->fp[nr].var)
+
+
+/* attn group wiring */
+#define MAX_DYNAMIC_ATTN_GRPS   	8
+
+struct attn_route {
+	u32     sig[4];
+};
+
+struct bnx2x {
+	/* Fields used in the tx and intr/napi performance paths
+	 * are grouped together in the beginning of the structure
+	 */
+	struct bnx2x_fastpath   *fp;
+	void __iomem    	*regview;
+	void __iomem    	*doorbells;
+
+	struct net_device       *dev;
+	struct pci_dev  	*pdev;
+
+	atomic_t		intr_sem;
+	struct msix_entry       msix_table[MAX_CONTEXT+1];
+
+	int     		tx_ring_size;
+
+#ifdef BCM_VLAN
+	struct vlan_group       *vlgrp;
+#endif
+
+	u32     		rx_csum;
+	u32     		rx_offset;
+	u32     		rx_buf_use_size;	/* useable size */
+	u32     		rx_buf_size;    	/* with alignment */
+#define ETH_OVREHEAD    		(ETH_HLEN + 8)  /* 8 for CRC + VLAN */
+#define ETH_MIN_PACKET_SIZE     	60
+#define ETH_MAX_PACKET_SIZE     	1500
+#define ETH_MAX_JUMBO_PACKET_SIZE       9600
+
+	struct host_def_status_block *def_status_blk;
+#define DEF_SB_ID       	16
+	u16     		def_c_idx;
+	u16     		def_u_idx;
+	u16     		def_t_idx;
+	u16     		def_x_idx;
+	u16     		def_att_idx;
+	u32     		attn_state;
+	struct attn_route       attn_group[MAX_DYNAMIC_ATTN_GRPS];
+	u32     		aeu_mask;
+	u32     		nig_mask;
+
+	/* slow path ring */
+	struct eth_spe  	*spq;
+	dma_addr_t      	spq_mapping;
+	u16     		spq_prod_idx;
+	u16     		dsb_sp_prod_idx;
+	struct eth_spe  	*spq_prod_bd;
+	struct eth_spe  	*spq_last_bd;
+	u16     		*dsb_sp_prod;
+	u16     		spq_left; /* serialize spq */
+	spinlock_t      	spq_lock;
+
+	/* Flag for marking that there is either
+	 * STAT_QUERY or CFC DELETE ramrod pending
+	 */
+	u8      		stat_pending;
+
+	/* End of fileds used in the performance code paths */
+
+	int     		panic;
+	int     		msglevel;
+
+	u32     		flags;
+#define PCIX_FLAG       		1
+#define PCI_32BIT_FLAG  		2
+#define ONE_TDMA_FLAG   		4       /* no longer used */
+#define NO_WOL_FLAG     		8
+#define USING_DAC_FLAG  		0x10
+#define USING_MSIX_FLAG 		0x20
+#define ASF_ENABLE_FLAG 		0x40
+
+	int     		port;
+
+	int     		pm_cap;
+	int     		pcie_cap;
+
+	/* Used to synchronize phy accesses */
+	spinlock_t      	phy_lock;
+
+	struct work_struct      reset_task;
+	u16     		in_reset_task;
+
+	struct work_struct      sp_task;
+
+	struct timer_list       timer;
+	int     		timer_interval;
+	int     		current_interval;
+
+	u32     		shmem_base;
+
+	u32     		chip_id;
+/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
+#define CHIP_ID(bp)     		(((bp)->chip_id) & 0xfffffff0)
+
+#define CHIP_NUM(bp)    		(((bp)->chip_id) & 0xffff0000)
+#define CHIP_NUM_5710   		0x57100000
+
+#define CHIP_REV(bp)    		(((bp)->chip_id) & 0x0000f000)
+#define CHIP_REV_Ax     		0x00000000
+#define CHIP_REV_Bx     		0x00001000
+#define CHIP_REV_Cx     		0x00002000
+#define CHIP_REV_EMUL   		0x0000e000
+#define CHIP_REV_FPGA   		0x0000f000
+#define CHIP_REV_IS_SLOW(bp)    	((CHIP_REV(bp) == CHIP_REV_EMUL) || \
+					 (CHIP_REV(bp) == CHIP_REV_FPGA))
+
+#define CHIP_METAL(bp)  		(((bp)->chip_id) & 0x00000ff0)
+#define CHIP_BOND_ID(bp)		(((bp)->chip_id) & 0x0000000f)
+
+	u16     		fw_seq;
+	u16     		fw_drv_pulse_wr_seq;
+	u32     		fw_mb;
+
+	u32     		hw_config;
+	u32     		serdes_config;
+	u32     		lane_config;
+	u32     		ext_phy_config;
+#define XGXS_EXT_PHY_TYPE(bp)   	(bp->ext_phy_config & \
+					 PORT_HW_CFG_XGXS_EXT_PHY_TYPE_MASK)
+#define SERDES_EXT_PHY_TYPE(bp) 	(bp->ext_phy_config & \
+					 PORT_HW_CFG_SERDES_EXT_PHY_TYPE_MASK)
+
+	u32     		speed_cap_mask;
+	u32     		link_config;
+#define SWITCH_CFG_1G   		PORT_FEATURE_CON_SWITCH_1G_SWITCH
+#define SWITCH_CFG_10G  		PORT_FEATURE_CON_SWITCH_10G_SWITCH
+#define SWITCH_CFG_AUTO_DETECT  	PORT_FEATURE_CON_SWITCH_AUTO_DETECT
+#define SWITCH_CFG_ONE_TIME_DETECT      \
+				PORT_FEATURE_CON_SWITCH_ONE_TIME_DETECT
+
+	u8      		ser_lane;
+	u8      		rx_lane_swap;
+	u8      		tx_lane_swap;
+
+	u8      		link_up;
+
+	u32     		supported;
+/* link settings - missing defines */
+#define SUPPORTED_2500baseT_Full	(1 << 15)
+#define SUPPORTED_CX4   		(1 << 16)
+
+	u32     		phy_flags;
+/*#define PHY_SERDES_FLAG       		0x1*/
+#define PHY_BMAC_FLAG   		0x2
+#define PHY_EMAC_FLAG   		0x4
+#define PHY_XGXS_FLAG   		0x8
+#define PHY_SGMII_FLAG  		0x10
+#define PHY_INT_MODE_MASK_FLAG  	0x300
+#define PHY_INT_MODE_AUTO_POLLING_FLAG  0x100
+#define PHY_INT_MODE_LINK_READY_FLAG    0x200
+
+	u32     		phy_addr;
+	u32     		phy_id;
+
+	u32     		autoneg;
+#define AUTONEG_CL37    		SHARED_HW_CFG_AN_ENABLE_CL37
+#define AUTONEG_CL73    		SHARED_HW_CFG_AN_ENABLE_CL73
+#define AUTONEG_BAM     		SHARED_HW_CFG_AN_ENABLE_BAM
+#define AUTONEG_PARALLEL		\
+				SHARED_HW_CFG_AN_ENABLE_PARALLEL_DETECTION
+#define AUTONEG_SGMII_FIBER_AUTODET     \
+				SHARED_HW_CFG_AN_EN_SGMII_FIBER_AUTO_DETECT
+#define AUTONEG_REMOTE_PHY      	SHARED_HW_CFG_AN_ENABLE_REMOTE_PHY
+
+	u32     		req_autoneg;
+#define AUTONEG_SPEED   		0x1
+#define AUTONEG_FLOW_CTRL       	0x2
+
+	u32     		req_line_speed;
+/* link settings - missing defines */
+#define SPEED_12000     		12000
+#define SPEED_12500     		12500
+#define SPEED_13000     		13000
+#define SPEED_15000     		15000
+#define SPEED_16000     		16000
+
+	u32     		req_duplex;
+	u32     		req_flow_ctrl;
+#define FLOW_CTRL_AUTO  		PORT_FEATURE_FLOW_CONTROL_AUTO
+#define FLOW_CTRL_TX    		PORT_FEATURE_FLOW_CONTROL_TX
+#define FLOW_CTRL_RX    		PORT_FEATURE_FLOW_CONTROL_RX
+#define FLOW_CTRL_BOTH  		PORT_FEATURE_FLOW_CONTROL_BOTH
+#define FLOW_CTRL_NONE  		PORT_FEATURE_FLOW_CONTROL_NONE
+
+	u32     		pause_mode;
+#define PAUSE_NONE      		0
+#define PAUSE_SYMMETRIC 		1
+#define PAUSE_ASYMMETRIC		2
+#define PAUSE_BOTH      		3
+
+	u32     		advertising;
+/* link settings - missing defines */
+#define ADVERTISED_2500baseT_Full       (1 << 15)
+#define ADVERTISED_CX4  		(1 << 16)
+
+	u32     		link_status;
+	u32     		line_speed;
+	u32     		duplex;
+	u32     		flow_ctrl;
+
+	u32     		bc_ver;
+
+	int     		flash_size;
+#define NVRAM_1MB_SIZE  		0x20000 /* 1M bit in bytes */
+#define NVRAM_TIMEOUT_COUNT     	30000
+#define NVRAM_PAGE_SIZE 		256
+
+	int     		rx_ring_size;
+
+	u16     		tx_quick_cons_trip_int;
+	u16     		tx_quick_cons_trip;
+	u16     		tx_ticks_int;
+	u16     		tx_ticks;
+
+	u16     		rx_quick_cons_trip_int;
+	u16     		rx_quick_cons_trip;
+	u16     		rx_ticks_int;
+	u16     		rx_ticks;
+
+	u32     		stats_ticks;
+
+	int     		state;
+#define BNX2X_STATE_CLOSED      	0x0
+#define BNX2X_STATE_OPENING_WAIT4_LOAD  0x1000
+#define BNX2X_STATE_OPENING_WAIT4_PORT  0x2000
+#define BNX2X_STATE_OPEN		0x3000
+#define BNX2X_STATE_CLOSING_WAIT4_HALT  0x4000
+#define BNX2X_STATE_CLOSING_WAIT4_DELETE 0x5000
+#define BNX2X_STATE_CLOSING_WAIT4_UNLOAD 0x6000
+#define BNX2X_STATE_ERROR       	0xF000
+
+	int     		num_queues;
+
+	u32     		rx_mode;
+#define BNX2X_RX_MODE_NONE      	0
+#define BNX2X_RX_MODE_NORMAL    	1
+#define BNX2X_RX_MODE_ALLMULTI  	2
+#define BNX2X_RX_MODE_PROMISC   	3
+#define BNX2X_MAX_MULTICAST     	64
+#define BNX2X_MAX_EMUL_MULTI    	16
+
+	dma_addr_t      	def_status_blk_mapping;
+
+	struct bnx2x_slowpath   *slowpath;
+	dma_addr_t      	slowpath_mapping;
+
+#ifdef BCM_ISCSI
+	void    		*t1;
+	dma_addr_t      	t1_mapping;
+	void    		*t2;
+	dma_addr_t      	t2_mapping;
+	void    		*timers;
+	dma_addr_t      	timers_mapping;
+	void    		*qm;
+	dma_addr_t      	qm_mapping;
+#endif
+
+	char    		*name;
+	u16     		bus_speed_mhz;
+	u8      		wol;
+	u8      		pad;
+
+	/* used to synchronize stats collecting */
+	int     		stats_state;
+#define STATS_STATE_DISABLE     	0
+#define STATS_STATE_ENABLE      	1
+#define STATS_STATE_STOP		2 /* stop stats on next iteration */
+
+	/* used by dmae command loader */
+	struct dmae_command     dmae;
+	int     		executer_idx;
+
+	u32     		old_brb_discard;
+	struct bmac_stats       old_bmac;
+	struct tstorm_per_client_stats old_tclient;
+	struct z_stream_s       *strm;
+	void    		*gunzip_buf;
+	dma_addr_t      	gunzip_mapping;
+	int     		gunzip_outlen;
+#define FW_BUF_SIZE     		0x8000
+
+};
+
+
+/* DMAE command defines */
+#define DMAE_CMD_SRC_PCI		0
+#define DMAE_CMD_SRC_GRC		DMAE_COMMAND_SRC
+
+#define DMAE_CMD_DST_PCI		(1 << DMAE_COMMAND_DST_SHIFT)
+#define DMAE_CMD_DST_GRC		(2 << DMAE_COMMAND_DST_SHIFT)
+
+#define DMAE_CMD_C_DST_PCI      	0
+#define DMAE_CMD_C_DST_GRC      	(1 << DMAE_COMMAND_C_DST_SHIFT)
+
+#define DMAE_CMD_C_ENABLE       	DMAE_COMMAND_C_TYPE_ENABLE
+
+#define DMAE_CMD_ENDIANITY_NO_SWAP      (0 << DMAE_COMMAND_ENDIANITY_SHIFT)
+#define DMAE_CMD_ENDIANITY_B_SWAP       (1 << DMAE_COMMAND_ENDIANITY_SHIFT)
+#define DMAE_CMD_ENDIANITY_DW_SWAP      (2 << DMAE_COMMAND_ENDIANITY_SHIFT)
+#define DMAE_CMD_ENDIANITY_B_DW_SWAP    (3 << DMAE_COMMAND_ENDIANITY_SHIFT)
+
+#define DMAE_CMD_PORT_0 		0
+#define DMAE_CMD_PORT_1 		DMAE_COMMAND_PORT
+
+#define DMAE_CMD_SRC_RESET      	DMAE_COMMAND_SRC_RESET
+#define DMAE_CMD_DST_RESET      	DMAE_COMMAND_DST_RESET
+
+#define DMAE_LEN32_MAX  		0x400
+
+
+/* MC hsi */
+#define RX_COPY_THRESH  		92
+#define BCM_PAGE_BITS   		12
+#define BCM_PAGE_SIZE   		(1 << BCM_PAGE_BITS)
+
+#define NUM_TX_RINGS    		16
+#define TX_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_tx_bd))
+#define MAX_TX_DESC_CNT 		(TX_DESC_CNT - 1)
+#define NUM_TX_BD       		(TX_DESC_CNT * NUM_TX_RINGS)
+#define MAX_TX_BD       		(NUM_TX_BD - 1)
+#define MAX_TX_AVAIL    		(MAX_TX_DESC_CNT * NUM_TX_RINGS - 2)
+#define NEXT_TX_IDX(x)  	((((x) & MAX_TX_DESC_CNT) == \
+				 (MAX_TX_DESC_CNT - 1)) ? (x) + 2 : (x) + 1)
+#define TX_BD(x)			((x) & MAX_TX_BD)
+#define TX_BD_POFF(x)   		((x) & MAX_TX_DESC_CNT)
+
+/* The RX BD ring is special, each bd is 8 bytes but the last one is 16 */
+#define NUM_RX_RINGS    		8
+#define RX_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_rx_bd))
+#define MAX_RX_DESC_CNT 		(RX_DESC_CNT - 2)
+#define RX_DESC_MASK    		(RX_DESC_CNT - 1)
+#define NUM_RX_BD       		(RX_DESC_CNT * NUM_RX_RINGS)
+#define MAX_RX_BD       		(NUM_RX_BD - 1)
+#define MAX_RX_AVAIL    		(MAX_RX_DESC_CNT * NUM_RX_RINGS - 2)
+#define NEXT_RX_IDX(x)  	((((x) & RX_DESC_MASK) == \
+				 (MAX_RX_DESC_CNT - 1)) ? (x) + 3 : (x) + 1)
+#define RX_BD(x)			((x) & MAX_RX_BD)
+
+#define NUM_RCQ_RINGS   		(NUM_RX_RINGS * 2)
+#define RCQ_DESC_CNT    	(BCM_PAGE_SIZE / sizeof(union eth_rx_cqe))
+#define MAX_RCQ_DESC_CNT		(RCQ_DESC_CNT - 1)
+#define NUM_RCQ_BD      		(RCQ_DESC_CNT * NUM_RCQ_RINGS)
+#define MAX_RCQ_BD      		(NUM_RCQ_BD - 1)
+#define MAX_RCQ_AVAIL   		(MAX_RCQ_DESC_CNT * NUM_RCQ_RINGS - 2)
+#define NEXT_RCQ_IDX(x) 	((((x) & MAX_RCQ_DESC_CNT) == \
+				 (MAX_RCQ_DESC_CNT - 1)) ? (x) + 2 : (x) + 1)
+#define RCQ_BD(x)       		((x) & MAX_RCQ_BD)
+
+
+/* used on a CID received from the HW */
+#define SW_CID(x)       		(le32_to_cpu(x) & \
+					 (COMMON_RAMROD_ETH_RX_CQE_CID >> 1))
+#define CQE_CMD(x)      		(le32_to_cpu(x) >> \
+					COMMON_RAMROD_ETH_RX_CQE_CMD_ID_SHIFT)
+
+#define BD_UNMAP_ADDR(bd)       	HILO_U64(le32_to_cpu((bd)->addr_hi), \
+						 le32_to_cpu((bd)->addr_lo))
+#define BD_UNMAP_LEN(bd)		(le16_to_cpu((bd)->nbytes))
+
+
+#define STROM_ASSERT_ARRAY_SIZE 	50
+
+
+#define MDIO_INDIRECT_REG_ADDR  	0x1f
+#define MDIO_SET_REG_BANK(bp, reg_bank) \
+		bnx2x_mdio22_write(bp, MDIO_INDIRECT_REG_ADDR, reg_bank)
+
+#define MDIO_ACCESS_TIMEOUT     	1000
+
+
+/* must be used on a CID before placing it on a HW ring */
+#define HW_CID(bp, x)   		(x | (bp->port << 23))
+
+#define SP_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_spe))
+#define MAX_SP_DESC_CNT 		(SP_DESC_CNT - 1)
+
+#define ATTN_NIG_FOR_FUNC       	(1L << 8)
+#define ATTN_SW_TIMER_4_FUNC    	(1L << 9)
+#define GPIO_2_FUNC     		(1L << 10)
+#define GPIO_3_FUNC     		(1L << 11)
+#define GPIO_4_FUNC     		(1L << 12)
+#define ATTN_GENERAL_ATTN_1     	(1L << 13)
+#define ATTN_GENERAL_ATTN_2     	(1L << 14)
+#define ATTN_GENERAL_ATTN_3     	(1L << 15)
+#define ATTN_GENERAL_ATTN_4     	(1L << 13)
+#define ATTN_GENERAL_ATTN_5     	(1L << 14)
+#define ATTN_GENERAL_ATTN_6     	(1L << 15)
+
+#define ATTN_HARD_WIRED_MASK    	0xff00
+#define ATTENTION_ID    		4
+
+
+#define BNX2X_BTR       		3
+#define MAX_SPQ_PENDING 		8
+
+
+#define BNX2X_NUM_STATS 		31
+#define BNX2X_NUM_TESTS 		2
+
+
+#define DPM_TRIGER_TYPE 		0x40
+#define DOORBELL(bp, cid, val) \
+	do { \
+		writel((u32)val, (bp)->doorbells + (BCM_PAGE_SIZE * cid) + \
+		       DPM_TRIGER_TYPE); \
+	} while (0)
+
+
+/* stuff added to make the code fit 80Col */
+
+#define TPA_TYPE_START  		ETH_FAST_PATH_RX_CQE_START_FLG
+#define TPA_TYPE_END    		ETH_FAST_PATH_RX_CQE_END_FLG
+#define TPA_TYPE(cqe)   	(cqe->fast_path_cqe.error_type_flags & \
+				 (TPA_TYPE_START | TPA_TYPE_END))
+#define BNX2X_RX_SUM_OK(cqe) \
+			(!(cqe->fast_path_cqe.status_flags & \
+			 (ETH_FAST_PATH_RX_CQE_IP_XSUM_NO_VALIDATION_FLG | \
+			  ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG)))
+
+#define BNX2X_RX_SUM_FIX(cqe) \
+			((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) & \
+			  PARSING_FLAGS_OVER_ETHERNET_PROTOCOL) == \
+			 (1 << PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT))
+
+
+#define MDIO_AN_CL73_OR_37_COMPLETE \
+		(MDIO_GP_STATUS_TOP_AN_STATUS1_CL73_AUTONEG_COMPLETE | \
+		 MDIO_GP_STATUS_TOP_AN_STATUS1_CL37_AUTONEG_COMPLETE)
+
+#define GP_STATUS_PAUSE_RSOLUTION_TXSIDE \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_PAUSE_RSOLUTION_TXSIDE
+#define GP_STATUS_PAUSE_RSOLUTION_RXSIDE \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_PAUSE_RSOLUTION_RXSIDE
+#define GP_STATUS_SPEED_MASK \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_MASK
+#define GP_STATUS_10M   MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10M
+#define GP_STATUS_100M  MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_100M
+#define GP_STATUS_1G    MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_1G
+#define GP_STATUS_2_5G  MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_2_5G
+#define GP_STATUS_5G    MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_5G
+#define GP_STATUS_6G    MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_6G
+#define GP_STATUS_10G_HIG \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_HIG
+#define GP_STATUS_10G_CX4 \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_CX4
+#define GP_STATUS_12G_HIG \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_12G_HIG
+#define GP_STATUS_12_5G MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_12_5G
+#define GP_STATUS_13G   MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_13G
+#define GP_STATUS_15G   MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_15G
+#define GP_STATUS_16G   MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_16G
+#define GP_STATUS_1G_KX MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_1G_KX
+#define GP_STATUS_10G_KX4 \
+			MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_KX4
+
+#define LINK_10THD      		LINK_STATUS_SPEED_AND_DUPLEX_10THD
+#define LINK_10TFD      		LINK_STATUS_SPEED_AND_DUPLEX_10TFD
+#define LINK_100TXHD    		LINK_STATUS_SPEED_AND_DUPLEX_100TXHD
+#define LINK_100T4      		LINK_STATUS_SPEED_AND_DUPLEX_100T4
+#define LINK_100TXFD    		LINK_STATUS_SPEED_AND_DUPLEX_100TXFD
+#define LINK_1000THD    		LINK_STATUS_SPEED_AND_DUPLEX_1000THD
+#define LINK_1000TFD    		LINK_STATUS_SPEED_AND_DUPLEX_1000TFD
+#define LINK_1000XFD    		LINK_STATUS_SPEED_AND_DUPLEX_1000XFD
+#define LINK_2500THD    		LINK_STATUS_SPEED_AND_DUPLEX_2500THD
+#define LINK_2500TFD    		LINK_STATUS_SPEED_AND_DUPLEX_2500TFD
+#define LINK_2500XFD    		LINK_STATUS_SPEED_AND_DUPLEX_2500XFD
+#define LINK_10GTFD     		LINK_STATUS_SPEED_AND_DUPLEX_10GTFD
+#define LINK_10GXFD     		LINK_STATUS_SPEED_AND_DUPLEX_10GXFD
+#define LINK_12GTFD     		LINK_STATUS_SPEED_AND_DUPLEX_12GTFD
+#define LINK_12GXFD     		LINK_STATUS_SPEED_AND_DUPLEX_12GXFD
+#define LINK_12_5GTFD   		LINK_STATUS_SPEED_AND_DUPLEX_12_5GTFD
+#define LINK_12_5GXFD   		LINK_STATUS_SPEED_AND_DUPLEX_12_5GXFD
+#define LINK_13GTFD     		LINK_STATUS_SPEED_AND_DUPLEX_13GTFD
+#define LINK_13GXFD     		LINK_STATUS_SPEED_AND_DUPLEX_13GXFD
+#define LINK_15GTFD     		LINK_STATUS_SPEED_AND_DUPLEX_15GTFD
+#define LINK_15GXFD     		LINK_STATUS_SPEED_AND_DUPLEX_15GXFD
+#define LINK_16GTFD     		LINK_STATUS_SPEED_AND_DUPLEX_16GTFD
+#define LINK_16GXFD     		LINK_STATUS_SPEED_AND_DUPLEX_16GXFD
+
+#define NIG_STATUS_INTERRUPT_XGXS0_LINK10G \
+		NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK10G
+#define NIG_XGXS0_LINK_STATUS \
+		NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK_STATUS
+#define NIG_XGXS0_LINK_STATUS_SIZE \
+		NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK_STATUS_SIZE
+#define NIG_SERDES0_LINK_STATUS \
+		NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_SERDES0_LINK_STATUS
+#define NIG_MASK_MI_INT \
+		NIG_MASK_INTERRUPT_PORT0_REG_MASK_EMAC0_MISC_MI_INT
+#define NIG_MASK_XGXS0_LINK10G \
+		NIG_MASK_INTERRUPT_PORT0_REG_MASK_XGXS0_LINK10G
+#define NIG_MASK_XGXS0_LINK_STATUS \
+		NIG_MASK_INTERRUPT_PORT0_REG_MASK_XGXS0_LINK_STATUS
+#define NIG_MASK_SERDES0_LINK_STATUS \
+		NIG_MASK_INTERRUPT_PORT0_REG_MASK_SERDES0_LINK_STATUS
+
+#define XGXS_RESET_BITS \
+	(MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_RSTB_HW |   \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_IDDQ |      \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_PWRDWN |    \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_PWRDWN_SD | \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_TXD_FIFO_RSTB)
+
+#define SERDES_RESET_BITS \
+	(MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_RSTB_HW | \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_IDDQ |    \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_PWRDWN |  \
+	 MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_PWRDWN_SD)
+
+
+#define BNX2X_MC_ASSERT_BITS \
+	(GENERAL_ATTEN_OFFSET(TSTORM_FATAL_ASSERT_ATTENTION_BIT) | \
+	 GENERAL_ATTEN_OFFSET(USTORM_FATAL_ASSERT_ATTENTION_BIT) | \
+	 GENERAL_ATTEN_OFFSET(CSTORM_FATAL_ASSERT_ATTENTION_BIT) | \
+	 GENERAL_ATTEN_OFFSET(XSTORM_FATAL_ASSERT_ATTENTION_BIT))
+
+#define BNX2X_MCP_ASSERT \
+	GENERAL_ATTEN_OFFSET(MCP_FATAL_ASSERT_ATTENTION_BIT)
+
+#define BNX2X_DOORQ_ASSERT \
+	AEU_INPUTS_ATTN_BITS_DOORBELLQ_HW_INTERRUPT
+
+#define HW_INTERRUT_ASSERT_SET_0 \
+				(AEU_INPUTS_ATTN_BITS_TSDM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_TCM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_TSEMI_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_PBF_HW_INTERRUPT)
+#define HW_PRTY_ASSERT_SET_0    (AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_PARSER_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_TSDM_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_SEARCHER_PARITY_ERROR |\
+				 AEU_INPUTS_ATTN_BITS_TSEMI_PARITY_ERROR)
+#define HW_INTERRUT_ASSERT_SET_1 \
+				(AEU_INPUTS_ATTN_BITS_QM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_TIMERS_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_XSDM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_XCM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_XSEMI_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_USDM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_UCM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_USEMI_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_UPB_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_CSDM_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_CCM_HW_INTERRUPT)
+#define HW_PRTY_ASSERT_SET_1    (AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR |\
+				 AEU_INPUTS_ATTN_BITS_QM_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_XSDM_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_XSEMI_PARITY_ERROR | \
+				AEU_INPUTS_ATTN_BITS_DOORBELLQ_PARITY_ERROR |\
+			    AEU_INPUTS_ATTN_BITS_VAUX_PCI_CORE_PARITY_ERROR |\
+				 AEU_INPUTS_ATTN_BITS_DEBUG_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_USDM_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_USEMI_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_UPB_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_CSDM_PARITY_ERROR)
+#define HW_INTERRUT_ASSERT_SET_2 \
+				(AEU_INPUTS_ATTN_BITS_CSEMI_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_CDU_HW_INTERRUPT | \
+				 AEU_INPUTS_ATTN_BITS_DMAE_HW_INTERRUPT | \
+			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_HW_INTERRUPT |\
+				 AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT)
+#define HW_PRTY_ASSERT_SET_2    (AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_PXP_PARITY_ERROR | \
+			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_PARITY_ERROR |\
+				 AEU_INPUTS_ATTN_BITS_CFC_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_CDU_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_IGU_PARITY_ERROR | \
+				 AEU_INPUTS_ATTN_BITS_MISC_PARITY_ERROR)
+
+
+#define ETH_RX_ERROR_FALGS      (ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG | \
+				 ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG | \
+				 ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG)
+
+
+#define MULTI_FLAGS \
+	(TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY     | \
+	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \
+	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY     | \
+	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \
+	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE)
+
+#define MULTI_MASK      0x7f
+
+
+#define U_SB_ETH_RX_CQ_INDEX    	HC_INDEX_U_ETH_RX_CQ_CONS
+#define C_SB_ETH_TX_CQ_INDEX    	HC_INDEX_C_ETH_TX_CQ_CONS
+#define C_DEF_SB_SP_INDEX       	HC_INDEX_DEF_C_ETH_SLOW_PATH
+
+#define BNX2X_RX_SB_INDEX \
+	&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_CQ_INDEX]
+
+#define BNX2X_TX_SB_INDEX \
+	&fp->status_blk->c_status_block.index_values[C_SB_ETH_TX_CQ_INDEX]
+
+#define BNX2X_SP_DSB_INDEX \
+&bp->def_status_blk->c_def_status_block.index_values[C_DEF_SB_SP_INDEX]
+
+
+#define CAM_IS_INVALID(x) \
+(x.target_table_entry.flags == TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE)
+
+#define CAM_INVALIDATE(x) \
+x.target_table_entry.flags = TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE
+
+
+/* MISC_REG_RESET_REG - this is here for the hsi to work don't touch */
+
+#endif /* bnx2x.h */
diff --git a/drivers/net/bnx2x_fw_defs.h b/drivers/net/bnx2x_fw_defs.h
new file mode 100644
index 000000000000..62a6eb81025a
--- /dev/null
+++ b/drivers/net/bnx2x_fw_defs.h
@@ -0,0 +1,198 @@
+/* bnx2x_fw_defs.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+
+#define CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
+	(0x1922 + (port * 0x40) + (index * 0x4))
+#define CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
+	(0x1900 + (port * 0x40))
+#define CSTORM_HC_BTR_OFFSET(port)\
+	(0x1984 + (port * 0xc0))
+#define CSTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
+	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
+#define CSTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
+	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
+#define CSTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
+#define CSTORM_STATS_FLAGS_OFFSET(port) 		(0x5108 + (port * 0x8))
+#define TSTORM_CLIENT_CONFIG_OFFSET(port, client_id)\
+	(0x1510 + (port * 0x240) + (client_id * 0x20))
+#define TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
+	(0x138a + (port * 0x28) + (index * 0x4))
+#define TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
+	(0x1370 + (port * 0x28))
+#define TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
+	(0x4b70 + (port * 0x8))
+#define TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(function)\
+	(0x1418 + (function * 0x30))
+#define TSTORM_HC_BTR_OFFSET(port)\
+	(0x13c4 + (port * 0x18))
+#define TSTORM_INDIRECTION_TABLE_OFFSET(port)\
+	(0x22c8 + (port * 0x80))
+#define TSTORM_INDIRECTION_TABLE_SIZE			0x80
+#define TSTORM_MAC_FILTER_CONFIG_OFFSET(port)\
+	(0x1420 + (port * 0x30))
+#define TSTORM_RCQ_PROD_OFFSET(port, client_id)\
+	(0x1508 + (port * 0x240) + (client_id * 0x20))
+#define TSTORM_STATS_FLAGS_OFFSET(port) 		(0x4b90 + (port * 0x8))
+#define USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
+	(0x191a + (port * 0x28) + (index * 0x4))
+#define USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
+	(0x1900 + (port * 0x28))
+#define USTORM_HC_BTR_OFFSET(port)\
+	(0x1954 + (port * 0xb8))
+#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port)\
+	(0x5408 + (port * 0x8))
+#define USTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
+	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
+#define USTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
+	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
+#define USTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
+#define XSTORM_ASSERT_LIST_INDEX_OFFSET 		0x1000
+#define XSTORM_ASSERT_LIST_OFFSET(idx)			(0x1020 + (idx * 0x10))
+#define XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
+	(0x141a + (port * 0x28) + (index * 0x4))
+#define XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
+	(0x1400 + (port * 0x28))
+#define XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
+	(0x5408 + (port * 0x8))
+#define XSTORM_HC_BTR_OFFSET(port)\
+	(0x1454 + (port * 0x18))
+#define XSTORM_SPQ_PAGE_BASE_OFFSET(port)\
+	(0x5328 + (port * 0x18))
+#define XSTORM_SPQ_PROD_OFFSET(port)\
+	(0x5330 + (port * 0x18))
+#define XSTORM_STATS_FLAGS_OFFSET(port) 		(0x53f8 + (port * 0x8))
+#define COMMON_ASM_INVALID_ASSERT_OPCODE 0x0
+
+/**
+* This file defines HSI constatnts for the ETH flow
+*/
+
+/* hash types */
+#define DEFAULT_HASH_TYPE			0
+#define IPV4_HASH_TYPE				1
+#define TCP_IPV4_HASH_TYPE			2
+#define IPV6_HASH_TYPE				3
+#define TCP_IPV6_HASH_TYPE			4
+
+/* values of command IDs in the ramrod message */
+#define RAMROD_CMD_ID_ETH_PORT_SETUP			(80)
+#define RAMROD_CMD_ID_ETH_CLIENT_SETUP			(85)
+#define RAMROD_CMD_ID_ETH_STAT_QUERY			(90)
+#define RAMROD_CMD_ID_ETH_UPDATE			(100)
+#define RAMROD_CMD_ID_ETH_HALT				(105)
+#define RAMROD_CMD_ID_ETH_SET_MAC			(110)
+#define RAMROD_CMD_ID_ETH_CFC_DEL			(115)
+#define RAMROD_CMD_ID_ETH_PORT_DEL			(120)
+#define RAMROD_CMD_ID_ETH_FORWARD_SETUP 		(125)
+
+
+/* command values for set mac command */
+#define T_ETH_MAC_COMMAND_SET				0
+#define T_ETH_MAC_COMMAND_INVALIDATE			1
+
+#define T_ETH_INDIRECTION_TABLE_SIZE			128
+
+/* Maximal L2 clients supported */
+#define ETH_MAX_RX_CLIENTS				(18)
+
+/**
+* This file defines HSI constatnts common to all microcode flows
+*/
+
+/* Connection types */
+#define ETH_CONNECTION_TYPE			0
+
+#define PROTOCOL_STATE_BIT_OFFSET		6
+
+#define ETH_STATE	(ETH_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+
+/* microcode fixed page page size 4K (chains and ring segments) */
+#define MC_PAGE_SIZE						(4096)
+
+/* Host coalescing constants */
+
+/* IGU constants */
+#define IGU_PORT_BASE				0x0400
+
+#define IGU_ADDR_MSIX				0x0000
+#define IGU_ADDR_INT_ACK			0x0200
+#define IGU_ADDR_PROD_UPD			0x0201
+#define IGU_ADDR_ATTN_BITS_UPD			0x0202
+#define IGU_ADDR_ATTN_BITS_SET			0x0203
+#define IGU_ADDR_ATTN_BITS_CLR			0x0204
+#define IGU_ADDR_COALESCE_NOW			0x0205
+#define IGU_ADDR_SIMD_MASK			0x0206
+#define IGU_ADDR_SIMD_NOMASK			0x0207
+#define IGU_ADDR_MSI_CTL			0x0210
+#define IGU_ADDR_MSI_ADDR_LO			0x0211
+#define IGU_ADDR_MSI_ADDR_HI			0x0212
+#define IGU_ADDR_MSI_DATA			0x0213
+
+#define IGU_INT_ENABLE				0
+#define IGU_INT_DISABLE 			1
+#define IGU_INT_NOP				2
+#define IGU_INT_NOP2				3
+
+/* index numbers */
+#define HC_USTORM_DEF_SB_NUM_INDICES		4
+#define HC_CSTORM_DEF_SB_NUM_INDICES		8
+#define HC_XSTORM_DEF_SB_NUM_INDICES		4
+#define HC_TSTORM_DEF_SB_NUM_INDICES		4
+#define HC_USTORM_SB_NUM_INDICES		4
+#define HC_CSTORM_SB_NUM_INDICES		4
+
+/* index values - which counterto update */
+
+#define HC_INDEX_U_ETH_RX_CQ_CONS		1
+
+#define HC_INDEX_C_ETH_TX_CQ_CONS		1
+
+#define HC_INDEX_DEF_X_SPQ_CONS 		0
+
+#define HC_INDEX_DEF_C_ETH_FW_TX_CQ_CONS	2
+#define HC_INDEX_DEF_C_ETH_SLOW_PATH		3
+
+/* used by the driver to get the SB offset */
+#define USTORM_ID			0
+#define CSTORM_ID			1
+#define XSTORM_ID			2
+#define TSTORM_ID			3
+#define ATTENTION_ID			4
+
+/* max number of slow path commands per port */
+#define MAX_RAMRODS_PER_PORT		(8)
+
+/* values for RX ETH CQE type field */
+#define RX_ETH_CQE_TYPE_ETH_FASTPATH	(0)
+#define RX_ETH_CQE_TYPE_ETH_RAMROD		(1)
+
+/* MAC address list size */
+#define T_MAC_ADDRESS_LIST_SIZE 	(96)
+
+#define XSTORM_IP_ID_ROLL_HALF 0x8000
+#define XSTORM_IP_ID_ROLL_ALL 0
+
+#define FW_LOG_LIST_SIZE	(50)
+
+#define NUM_OF_PROTOCOLS		4
+#define MAX_COS_NUMBER			16
+#define MAX_T_STAT_COUNTER_ID	18
+
+#define T_FAIR							1
+#define FAIR_MEM						2
+#define RS_PERIODIC_TIMEOUT_IN_SDM_TICS 25
+
+#define UNKNOWN_ADDRESS 	0
+#define UNICAST_ADDRESS 	1
+#define MULTICAST_ADDRESS	2
+#define BROADCAST_ADDRESS	3
+
diff --git a/drivers/net/bnx2x_hsi.h b/drivers/net/bnx2x_hsi.h
new file mode 100644
index 000000000000..6fd959c34d1f
--- /dev/null
+++ b/drivers/net/bnx2x_hsi.h
@@ -0,0 +1,2176 @@
+/* bnx2x_hsi.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+
+#define FUNC_0				0
+#define FUNC_1				1
+#define FUNC_MAX			2
+
+
+/* This value (in milliseconds) determines the frequency of the driver
+ * issuing the PULSE message code.  The firmware monitors this periodic
+ * pulse to determine when to switch to an OS-absent mode. */
+#define DRV_PULSE_PERIOD_MS		250
+
+/* This value (in milliseconds) determines how long the driver should
+ * wait for an acknowledgement from the firmware before timing out.  Once
+ * the firmware has timed out, the driver will assume there is no firmware
+ * running and there won't be any firmware-driver synchronization during a
+ * driver reset. */
+#define FW_ACK_TIME_OUT_MS		5000
+
+#define FW_ACK_POLL_TIME_MS		1
+
+#define FW_ACK_NUM_OF_POLL	(FW_ACK_TIME_OUT_MS/FW_ACK_POLL_TIME_MS)
+
+/* LED Blink rate that will achieve ~15.9Hz */
+#define LED_BLINK_RATE_VAL		480
+
+/****************************************************************************
+ * Driver <-> FW Mailbox						    *
+ ****************************************************************************/
+struct drv_fw_mb {
+	u32 drv_mb_header;
+#define DRV_MSG_CODE_MASK			0xffff0000
+#define DRV_MSG_CODE_LOAD_REQ			0x10000000
+#define DRV_MSG_CODE_LOAD_DONE			0x11000000
+#define DRV_MSG_CODE_UNLOAD_REQ_WOL_EN		0x20000000
+#define DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS 	0x20010000
+#define DRV_MSG_CODE_UNLOAD_REQ_WOL_MCP 	0x20020000
+#define DRV_MSG_CODE_UNLOAD_DONE		0x21000000
+#define DRV_MSG_CODE_DIAG_ENTER_REQ		0x50000000
+#define DRV_MSG_CODE_DIAG_EXIT_REQ		0x60000000
+#define DRV_MSG_CODE_VALIDATE_KEY		0x70000000
+#define DRV_MSG_CODE_GET_CURR_KEY		0x80000000
+#define DRV_MSG_CODE_GET_UPGRADE_KEY		0x81000000
+#define DRV_MSG_CODE_GET_MANUF_KEY		0x82000000
+#define DRV_MSG_CODE_LOAD_L2B_PRAM		0x90000000
+
+#define DRV_MSG_SEQ_NUMBER_MASK 		0x0000ffff
+
+	u32 drv_mb_param;
+
+	u32 fw_mb_header;
+#define FW_MSG_CODE_MASK			0xffff0000
+#define FW_MSG_CODE_DRV_LOAD_COMMON		0x11000000
+#define FW_MSG_CODE_DRV_LOAD_PORT		0x12000000
+#define FW_MSG_CODE_DRV_LOAD_REFUSED		0x13000000
+#define FW_MSG_CODE_DRV_LOAD_DONE		0x14000000
+#define FW_MSG_CODE_DRV_UNLOAD_COMMON		0x21000000
+#define FW_MSG_CODE_DRV_UNLOAD_PORT		0x22000000
+#define FW_MSG_CODE_DRV_UNLOAD_DONE		0x23000000
+#define FW_MSG_CODE_DIAG_ENTER_DONE		0x50000000
+#define FW_MSG_CODE_DIAG_REFUSE 		0x51000000
+#define FW_MSG_CODE_VALIDATE_KEY_SUCCESS	0x70000000
+#define FW_MSG_CODE_VALIDATE_KEY_FAILURE	0x71000000
+#define FW_MSG_CODE_GET_KEY_DONE		0x80000000
+#define FW_MSG_CODE_NO_KEY			0x8f000000
+#define FW_MSG_CODE_LIC_INFO_NOT_READY		0x8f800000
+#define FW_MSG_CODE_L2B_PRAM_LOADED		0x90000000
+#define FW_MSG_CODE_L2B_PRAM_T_LOAD_FAILURE	0x91000000
+#define FW_MSG_CODE_L2B_PRAM_C_LOAD_FAILURE	0x92000000
+#define FW_MSG_CODE_L2B_PRAM_X_LOAD_FAILURE	0x93000000
+#define FW_MSG_CODE_L2B_PRAM_U_LOAD_FAILURE	0x94000000
+
+#define FW_MSG_SEQ_NUMBER_MASK			0x0000ffff
+
+	u32 fw_mb_param;
+
+	u32 link_status;
+	/* Driver should update this field on any link change event */
+
+#define LINK_STATUS_LINK_FLAG_MASK		0x00000001
+#define LINK_STATUS_LINK_UP			0x00000001
+#define LINK_STATUS_SPEED_AND_DUPLEX_MASK	0x0000001E
+#define LINK_STATUS_SPEED_AND_DUPLEX_AN_NOT_COMPLETE	(0<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_10THD		(1<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_10TFD		(2<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_100TXHD		(3<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_100T4		(4<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_100TXFD		(5<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_1000THD		(6<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_1000TFD		(7<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_1000XFD		(7<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_2500THD		(8<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_2500TFD		(9<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_2500XFD		(9<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_10GTFD		(10<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_10GXFD		(10<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_12GTFD		(11<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_12GXFD		(11<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_12_5GTFD		(12<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_12_5GXFD		(12<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_13GTFD		(13<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_13GXFD		(13<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_15GTFD		(14<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_15GXFD		(14<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_16GTFD		(15<<1)
+#define LINK_STATUS_SPEED_AND_DUPLEX_16GXFD		(15<<1)
+
+#define LINK_STATUS_AUTO_NEGOTIATE_FLAG_MASK		0x00000020
+#define LINK_STATUS_AUTO_NEGOTIATE_ENABLED		0x00000020
+
+#define LINK_STATUS_AUTO_NEGOTIATE_COMPLETE		0x00000040
+#define LINK_STATUS_PARALLEL_DETECTION_FLAG_MASK	0x00000080
+#define LINK_STATUS_PARALLEL_DETECTION_USED		0x00000080
+
+#define LINK_STATUS_LINK_PARTNER_1000TFD_CAPABLE	0x00000200
+#define LINK_STATUS_LINK_PARTNER_1000THD_CAPABLE	0x00000400
+#define LINK_STATUS_LINK_PARTNER_100T4_CAPABLE		0x00000800
+#define LINK_STATUS_LINK_PARTNER_100TXFD_CAPABLE	0x00001000
+#define LINK_STATUS_LINK_PARTNER_100TXHD_CAPABLE	0x00002000
+#define LINK_STATUS_LINK_PARTNER_10TFD_CAPABLE		0x00004000
+#define LINK_STATUS_LINK_PARTNER_10THD_CAPABLE		0x00008000
+
+#define LINK_STATUS_TX_FLOW_CONTROL_FLAG_MASK		0x00010000
+#define LINK_STATUS_TX_FLOW_CONTROL_ENABLED		0x00010000
+
+#define LINK_STATUS_RX_FLOW_CONTROL_FLAG_MASK		0x00020000
+#define LINK_STATUS_RX_FLOW_CONTROL_ENABLED		0x00020000
+
+#define LINK_STATUS_LINK_PARTNER_FLOW_CONTROL_MASK	0x000C0000
+#define LINK_STATUS_LINK_PARTNER_NOT_PAUSE_CAPABLE	(0<<18)
+#define LINK_STATUS_LINK_PARTNER_SYMMETRIC_PAUSE	(1<<18)
+#define LINK_STATUS_LINK_PARTNER_ASYMMETRIC_PAUSE	(2<<18)
+#define LINK_STATUS_LINK_PARTNER_BOTH_PAUSE		(3<<18)
+
+#define LINK_STATUS_SERDES_LINK 			0x00100000
+
+#define LINK_STATUS_LINK_PARTNER_2500XFD_CAPABLE	0x00200000
+#define LINK_STATUS_LINK_PARTNER_2500XHD_CAPABLE	0x00400000
+#define LINK_STATUS_LINK_PARTNER_10GXFD_CAPABLE 	0x00800000
+#define LINK_STATUS_LINK_PARTNER_12GXFD_CAPABLE 	0x01000000
+#define LINK_STATUS_LINK_PARTNER_12_5GXFD_CAPABLE	0x02000000
+#define LINK_STATUS_LINK_PARTNER_13GXFD_CAPABLE 	0x04000000
+#define LINK_STATUS_LINK_PARTNER_15GXFD_CAPABLE 	0x08000000
+#define LINK_STATUS_LINK_PARTNER_16GXFD_CAPABLE 	0x10000000
+
+	u32 drv_pulse_mb;
+#define DRV_PULSE_SEQ_MASK				0x00007fff
+#define DRV_PULSE_SYSTEM_TIME_MASK			0xffff0000
+	/* The system time is in the format of
+	 * (year-2001)*12*32 + month*32 + day. */
+#define DRV_PULSE_ALWAYS_ALIVE				0x00008000
+	/* Indicate to the firmware not to go into the
+	 * OS-absent when it is not getting driver pulse.
+	 * This is used for debugging as well for PXE(MBA). */
+
+	u32 mcp_pulse_mb;
+#define MCP_PULSE_SEQ_MASK				0x00007fff
+#define MCP_PULSE_ALWAYS_ALIVE				0x00008000
+	/* Indicates to the driver not to assert due to lack
+	 * of MCP response */
+#define MCP_EVENT_MASK					0xffff0000
+#define MCP_EVENT_OTHER_DRIVER_RESET_REQ		0x00010000
+
+};
+
+
+/****************************************************************************
+ * Shared HW configuration						    *
+ ****************************************************************************/
+struct shared_hw_cfg {					 /* NVRAM Offset */
+	/* Up to 16 bytes of NULL-terminated string */
+	u8  part_num[16];					/* 0x104 */
+
+	u32 config;						/* 0x114 */
+#define SHARED_HW_CFG_MDIO_VOLTAGE_MASK 	    0x00000001
+#define SHARED_HW_CFG_MDIO_VOLTAGE_SHIFT	    0
+#define SHARED_HW_CFG_MDIO_VOLTAGE_1_2V 	    0x00000000
+#define SHARED_HW_CFG_MDIO_VOLTAGE_2_5V 	    0x00000001
+#define SHARED_HW_CFG_MCP_RST_ON_CORE_RST_EN	    0x00000002
+
+#define SHARED_HW_CFG_PORT_SWAP 		    0x00000004
+
+#define SHARED_HW_CFG_BEACON_WOL_EN		    0x00000008
+
+#define SHARED_HW_CFG_MFW_SELECT_MASK		    0x00000700
+#define SHARED_HW_CFG_MFW_SELECT_SHIFT		    8
+	/* Whatever MFW found in NVM
+	   (if multiple found, priority order is: NC-SI, UMP, IPMI) */
+#define SHARED_HW_CFG_MFW_SELECT_DEFAULT	    0x00000000
+#define SHARED_HW_CFG_MFW_SELECT_NC_SI		    0x00000100
+#define SHARED_HW_CFG_MFW_SELECT_UMP		    0x00000200
+#define SHARED_HW_CFG_MFW_SELECT_IPMI		    0x00000300
+	/* Use SPIO4 as an arbiter between: 0-NC_SI, 1-IPMI
+	  (can only be used when an add-in board, not BMC, pulls-down SPIO4) */
+#define SHARED_HW_CFG_MFW_SELECT_SPIO4_NC_SI_IPMI   0x00000400
+	/* Use SPIO4 as an arbiter between: 0-UMP, 1-IPMI
+	  (can only be used when an add-in board, not BMC, pulls-down SPIO4) */
+#define SHARED_HW_CFG_MFW_SELECT_SPIO4_UMP_IPMI     0x00000500
+	/* Use SPIO4 as an arbiter between: 0-NC-SI, 1-UMP
+	  (can only be used when an add-in board, not BMC, pulls-down SPIO4) */
+#define SHARED_HW_CFG_MFW_SELECT_SPIO4_NC_SI_UMP    0x00000600
+
+#define SHARED_HW_CFG_LED_MODE_MASK		    0x000f0000
+#define SHARED_HW_CFG_LED_MODE_SHIFT		    16
+#define SHARED_HW_CFG_LED_MAC1			    0x00000000
+#define SHARED_HW_CFG_LED_PHY1			    0x00010000
+#define SHARED_HW_CFG_LED_PHY2			    0x00020000
+#define SHARED_HW_CFG_LED_PHY3			    0x00030000
+#define SHARED_HW_CFG_LED_MAC2			    0x00040000
+#define SHARED_HW_CFG_LED_PHY4			    0x00050000
+#define SHARED_HW_CFG_LED_PHY5			    0x00060000
+#define SHARED_HW_CFG_LED_PHY6			    0x00070000
+#define SHARED_HW_CFG_LED_MAC3			    0x00080000
+#define SHARED_HW_CFG_LED_PHY7			    0x00090000
+#define SHARED_HW_CFG_LED_PHY9			    0x000a0000
+#define SHARED_HW_CFG_LED_PHY11 		    0x000b0000
+#define SHARED_HW_CFG_LED_MAC4			    0x000c0000
+#define SHARED_HW_CFG_LED_PHY8			    0x000d0000
+
+#define SHARED_HW_CFG_AN_ENABLE_MASK		    0x3f000000
+#define SHARED_HW_CFG_AN_ENABLE_SHIFT		    24
+#define SHARED_HW_CFG_AN_ENABLE_CL37		    0x01000000
+#define SHARED_HW_CFG_AN_ENABLE_CL73		    0x02000000
+#define SHARED_HW_CFG_AN_ENABLE_BAM		    0x04000000
+#define SHARED_HW_CFG_AN_ENABLE_PARALLEL_DETECTION  0x08000000
+#define SHARED_HW_CFG_AN_EN_SGMII_FIBER_AUTO_DETECT 0x10000000
+#define SHARED_HW_CFG_AN_ENABLE_REMOTE_PHY	    0x20000000
+
+	u32 config2;						/* 0x118 */
+	/* one time auto detect grace period (in sec) */
+#define SHARED_HW_CFG_GRACE_PERIOD_MASK 	    0x000000ff
+#define SHARED_HW_CFG_GRACE_PERIOD_SHIFT	    0
+
+#define SHARED_HW_CFG_PCIE_GEN2_ENABLED 	    0x00000100
+
+	/* The default value for the core clock is 250MHz and it is
+	   achieved by setting the clock change to 4 */
+#define SHARED_HW_CFG_CLOCK_CHANGE_MASK 	    0x00000e00
+#define SHARED_HW_CFG_CLOCK_CHANGE_SHIFT	    9
+
+#define SHARED_HW_CFG_SMBUS_TIMING_100KHZ	    0x00000000
+#define SHARED_HW_CFG_SMBUS_TIMING_400KHZ	    0x00001000
+
+#define SHARED_HW_CFG_HIDE_FUNC1		    0x00002000
+
+	u32 power_dissipated;					/* 0x11c */
+#define SHARED_HW_CFG_POWER_DIS_CMN_MASK	    0xff000000
+#define SHARED_HW_CFG_POWER_DIS_CMN_SHIFT	    24
+
+#define SHARED_HW_CFG_POWER_MGNT_SCALE_MASK	    0x00ff0000
+#define SHARED_HW_CFG_POWER_MGNT_SCALE_SHIFT	    16
+#define SHARED_HW_CFG_POWER_MGNT_UNKNOWN_SCALE	    0x00000000
+#define SHARED_HW_CFG_POWER_MGNT_DOT_1_WATT	    0x00010000
+#define SHARED_HW_CFG_POWER_MGNT_DOT_01_WATT	    0x00020000
+#define SHARED_HW_CFG_POWER_MGNT_DOT_001_WATT	    0x00030000
+
+	u32 ump_nc_si_config;					/* 0x120 */
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_MASK	    0x00000003
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_SHIFT	    0
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_MAC	    0x00000000
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_PHY	    0x00000001
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_MII	    0x00000000
+#define SHARED_HW_CFG_UMP_NC_SI_MII_MODE_RMII	    0x00000002
+
+#define SHARED_HW_CFG_UMP_NC_SI_NUM_DEVS_MASK	    0x00000f00
+#define SHARED_HW_CFG_UMP_NC_SI_NUM_DEVS_SHIFT	    8
+
+#define SHARED_HW_CFG_UMP_NC_SI_EXT_PHY_TYPE_MASK   0x00ff0000
+#define SHARED_HW_CFG_UMP_NC_SI_EXT_PHY_TYPE_SHIFT  16
+#define SHARED_HW_CFG_UMP_NC_SI_EXT_PHY_TYPE_NONE   0x00000000
+#define SHARED_HW_CFG_UMP_NC_SI_EXT_PHY_TYPE_BCM5221 0x00010000
+
+	u32 board;						/* 0x124 */
+#define SHARED_HW_CFG_BOARD_TYPE_MASK		    0x0000ffff
+#define SHARED_HW_CFG_BOARD_TYPE_SHIFT		    0
+#define SHARED_HW_CFG_BOARD_TYPE_NONE		    0x00000000
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1000     0x00000001
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1001     0x00000002
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1002G    0x00000003
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1004G    0x00000004
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1007G    0x00000005
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1015G    0x00000006
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1020G    0x00000007
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1003G    0x00000008
+
+#define SHARED_HW_CFG_BOARD_VER_MASK		    0xffff0000
+#define SHARED_HW_CFG_BOARD_VER_SHIFT		    16
+#define SHARED_HW_CFG_BOARD_MAJOR_VER_MASK	    0xf0000000
+#define SHARED_HW_CFG_BOARD_MAJOR_VER_SHIFT	    28
+#define SHARED_HW_CFG_BOARD_MINOR_VER_MASK	    0x0f000000
+#define SHARED_HW_CFG_BOARD_MINOR_VER_SHIFT	    24
+#define SHARED_HW_CFG_BOARD_REV_MASK		    0x00ff0000
+#define SHARED_HW_CFG_BOARD_REV_SHIFT		    16
+
+	u32 reserved;						/* 0x128 */
+
+};
+
+/****************************************************************************
+ * Port HW configuration						    *
+ ****************************************************************************/
+struct port_hw_cfg {	/* function 0: 0x12c-0x2bb, function 1: 0x2bc-0x44b */
+
+	/* Fields below are port specific (in anticipation of dual port
+	   devices */
+	u32 pci_id;
+#define PORT_HW_CFG_PCI_VENDOR_ID_MASK		    0xffff0000
+#define PORT_HW_CFG_PCI_DEVICE_ID_MASK		    0x0000ffff
+
+	u32 pci_sub_id;
+#define PORT_HW_CFG_PCI_SUBSYS_DEVICE_ID_MASK	    0xffff0000
+#define PORT_HW_CFG_PCI_SUBSYS_VENDOR_ID_MASK	    0x0000ffff
+
+	u32 power_dissipated;
+#define PORT_HW_CFG_POWER_DIS_D3_MASK		    0xff000000
+#define PORT_HW_CFG_POWER_DIS_D3_SHIFT		    24
+#define PORT_HW_CFG_POWER_DIS_D2_MASK		    0x00ff0000
+#define PORT_HW_CFG_POWER_DIS_D2_SHIFT		    16
+#define PORT_HW_CFG_POWER_DIS_D1_MASK		    0x0000ff00
+#define PORT_HW_CFG_POWER_DIS_D1_SHIFT		    8
+#define PORT_HW_CFG_POWER_DIS_D0_MASK		    0x000000ff
+#define PORT_HW_CFG_POWER_DIS_D0_SHIFT		    0
+
+	u32 power_consumed;
+#define PORT_HW_CFG_POWER_CONS_D3_MASK		    0xff000000
+#define PORT_HW_CFG_POWER_CONS_D3_SHIFT 	    24
+#define PORT_HW_CFG_POWER_CONS_D2_MASK		    0x00ff0000
+#define PORT_HW_CFG_POWER_CONS_D2_SHIFT 	    16
+#define PORT_HW_CFG_POWER_CONS_D1_MASK		    0x0000ff00
+#define PORT_HW_CFG_POWER_CONS_D1_SHIFT 	    8
+#define PORT_HW_CFG_POWER_CONS_D0_MASK		    0x000000ff
+#define PORT_HW_CFG_POWER_CONS_D0_SHIFT 	    0
+
+	u32 mac_upper;
+#define PORT_HW_CFG_UPPERMAC_MASK		    0x0000ffff
+#define PORT_HW_CFG_UPPERMAC_SHIFT		    0
+	u32 mac_lower;
+
+	u32 iscsi_mac_upper;  /* Upper 16 bits are always zeroes */
+	u32 iscsi_mac_lower;
+
+	u32 rdma_mac_upper;   /* Upper 16 bits are always zeroes */
+	u32 rdma_mac_lower;
+
+	u32 serdes_config;
+	/* for external PHY, or forced mode or during AN */
+#define PORT_HW_CFG_SERDES_TX_DRV_PRE_EMPHASIS_MASK 0xffff0000
+#define PORT_HW_CFG_SERDES_TX_DRV_PRE_EMPHASIS_SHIFT  16
+
+#define PORT_HW_CFG_SERDES_RX_DRV_EQUALIZER_MASK    0x0000ffff
+#define PORT_HW_CFG_SERDES_RX_DRV_EQUALIZER_SHIFT   0
+
+	u16 serdes_tx_driver_pre_emphasis[16];
+	u16 serdes_rx_driver_equalizer[16];
+
+	u32 xgxs_config_lane0;
+	u32 xgxs_config_lane1;
+	u32 xgxs_config_lane2;
+	u32 xgxs_config_lane3;
+	/* for external PHY, or forced mode or during AN */
+#define PORT_HW_CFG_XGXS_TX_DRV_PRE_EMPHASIS_MASK   0xffff0000
+#define PORT_HW_CFG_XGXS_TX_DRV_PRE_EMPHASIS_SHIFT  16
+
+#define PORT_HW_CFG_XGXS_RX_DRV_EQUALIZER_MASK	    0x0000ffff
+#define PORT_HW_CFG_XGXS_RX_DRV_EQUALIZER_SHIFT     0
+
+	u16 xgxs_tx_driver_pre_emphasis_lane0[16];
+	u16 xgxs_tx_driver_pre_emphasis_lane1[16];
+	u16 xgxs_tx_driver_pre_emphasis_lane2[16];
+	u16 xgxs_tx_driver_pre_emphasis_lane3[16];
+
+	u16 xgxs_rx_driver_equalizer_lane0[16];
+	u16 xgxs_rx_driver_equalizer_lane1[16];
+	u16 xgxs_rx_driver_equalizer_lane2[16];
+	u16 xgxs_rx_driver_equalizer_lane3[16];
+
+	u32 lane_config;
+#define PORT_HW_CFG_LANE_SWAP_CFG_MASK		    0x0000ffff
+#define PORT_HW_CFG_LANE_SWAP_CFG_SHIFT 	    0
+#define PORT_HW_CFG_LANE_SWAP_CFG_TX_MASK	    0x000000ff
+#define PORT_HW_CFG_LANE_SWAP_CFG_TX_SHIFT	    0
+#define PORT_HW_CFG_LANE_SWAP_CFG_RX_MASK	    0x0000ff00
+#define PORT_HW_CFG_LANE_SWAP_CFG_RX_SHIFT	    8
+#define PORT_HW_CFG_LANE_SWAP_CFG_MASTER_MASK	    0x0000c000
+#define PORT_HW_CFG_LANE_SWAP_CFG_MASTER_SHIFT	    14
+	/* AN and forced */
+#define PORT_HW_CFG_LANE_SWAP_CFG_01230123	    0x00001b1b
+	/* forced only */
+#define PORT_HW_CFG_LANE_SWAP_CFG_01233210	    0x00001be4
+	/* forced only */
+#define PORT_HW_CFG_LANE_SWAP_CFG_31203120	    0x0000d8d8
+	/* forced only */
+#define PORT_HW_CFG_LANE_SWAP_CFG_32103210	    0x0000e4e4
+
+	u32 external_phy_config;
+#define PORT_HW_CFG_SERDES_EXT_PHY_TYPE_MASK	    0xff000000
+#define PORT_HW_CFG_SERDES_EXT_PHY_TYPE_SHIFT	    24
+#define PORT_HW_CFG_SERDES_EXT_PHY_TYPE_DIRECT	    0x00000000
+#define PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482     0x01000000
+#define PORT_HW_CFG_SERDES_EXT_PHY_TYPE_NOT_CONN    0xff000000
+
+#define PORT_HW_CFG_SERDES_EXT_PHY_ADDR_MASK	    0x00ff0000
+#define PORT_HW_CFG_SERDES_EXT_PHY_ADDR_SHIFT	    16
+
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_MASK	    0x0000ff00
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_SHIFT	    8
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT	    0x00000000
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8071	    0x00000100
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8072	    0x00000200
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073	    0x00000300
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705	    0x00000400
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706	    0x00000500
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8276	    0x00000600
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8481	    0x00000700
+#define PORT_HW_CFG_XGXS_EXT_PHY_TYPE_NOT_CONN	    0x0000ff00
+
+#define PORT_HW_CFG_XGXS_EXT_PHY_ADDR_MASK	    0x000000ff
+#define PORT_HW_CFG_XGXS_EXT_PHY_ADDR_SHIFT	    0
+
+	u32 speed_capability_mask;
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_MASK	    0xffff0000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_SHIFT	    16
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL    0x00010000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF    0x00020000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF   0x00040000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL   0x00080000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_1G	    0x00100000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_2_5G	    0x00200000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_10G	    0x00400000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_12G	    0x00800000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_12_5G	    0x01000000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_13G	    0x02000000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_15G	    0x04000000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_16G	    0x08000000
+#define PORT_HW_CFG_SPEED_CAPABILITY_D0_RESERVED    0xf0000000
+
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_MASK	    0x0000ffff
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_SHIFT	    0
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_10M_FULL    0x00000001
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_10M_HALF    0x00000002
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_100M_HALF   0x00000004
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_100M_FULL   0x00000008
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_1G	    0x00000010
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_2_5G	    0x00000020
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_10G	    0x00000040
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_12G	    0x00000080
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_12_5G	    0x00000100
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_13G	    0x00000200
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_15G	    0x00000400
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_16G	    0x00000800
+#define PORT_HW_CFG_SPEED_CAPABILITY_D3_RESERVED    0x0000f000
+
+	u32 reserved[2];
+
+};
+
+/****************************************************************************
+ * Shared Feature configuration 					    *
+ ****************************************************************************/
+struct shared_feat_cfg {				 /* NVRAM Offset */
+	u32 bmc_common; 					/* 0x450 */
+#define SHARED_FEATURE_BMC_ECHO_MODE_EN 	    0x00000001
+
+};
+
+
+/****************************************************************************
+ * Port Feature configuration						    *
+ ****************************************************************************/
+struct port_feat_cfg {	/* function 0: 0x454-0x4c7, function 1: 0x4c8-0x53b */
+	u32 config;
+#define PORT_FEATURE_BAR1_SIZE_MASK		    0x0000000f
+#define PORT_FEATURE_BAR1_SIZE_SHIFT		    0
+#define PORT_FEATURE_BAR1_SIZE_DISABLED 	    0x00000000
+#define PORT_FEATURE_BAR1_SIZE_64K		    0x00000001
+#define PORT_FEATURE_BAR1_SIZE_128K		    0x00000002
+#define PORT_FEATURE_BAR1_SIZE_256K		    0x00000003
+#define PORT_FEATURE_BAR1_SIZE_512K		    0x00000004
+#define PORT_FEATURE_BAR1_SIZE_1M		    0x00000005
+#define PORT_FEATURE_BAR1_SIZE_2M		    0x00000006
+#define PORT_FEATURE_BAR1_SIZE_4M		    0x00000007
+#define PORT_FEATURE_BAR1_SIZE_8M		    0x00000008
+#define PORT_FEATURE_BAR1_SIZE_16M		    0x00000009
+#define PORT_FEATURE_BAR1_SIZE_32M		    0x0000000a
+#define PORT_FEATURE_BAR1_SIZE_64M		    0x0000000b
+#define PORT_FEATURE_BAR1_SIZE_128M		    0x0000000c
+#define PORT_FEATURE_BAR1_SIZE_256M		    0x0000000d
+#define PORT_FEATURE_BAR1_SIZE_512M		    0x0000000e
+#define PORT_FEATURE_BAR1_SIZE_1G		    0x0000000f
+#define PORT_FEATURE_BAR2_SIZE_MASK		    0x000000f0
+#define PORT_FEATURE_BAR2_SIZE_SHIFT		    4
+#define PORT_FEATURE_BAR2_SIZE_DISABLED 	    0x00000000
+#define PORT_FEATURE_BAR2_SIZE_64K		    0x00000010
+#define PORT_FEATURE_BAR2_SIZE_128K		    0x00000020
+#define PORT_FEATURE_BAR2_SIZE_256K		    0x00000030
+#define PORT_FEATURE_BAR2_SIZE_512K		    0x00000040
+#define PORT_FEATURE_BAR2_SIZE_1M		    0x00000050
+#define PORT_FEATURE_BAR2_SIZE_2M		    0x00000060
+#define PORT_FEATURE_BAR2_SIZE_4M		    0x00000070
+#define PORT_FEATURE_BAR2_SIZE_8M		    0x00000080
+#define PORT_FEATURE_BAR2_SIZE_16M		    0x00000090
+#define PORT_FEATURE_BAR2_SIZE_32M		    0x000000a0
+#define PORT_FEATURE_BAR2_SIZE_64M		    0x000000b0
+#define PORT_FEATURE_BAR2_SIZE_128M		    0x000000c0
+#define PORT_FEATURE_BAR2_SIZE_256M		    0x000000d0
+#define PORT_FEATURE_BAR2_SIZE_512M		    0x000000e0
+#define PORT_FEATURE_BAR2_SIZE_1G		    0x000000f0
+#define PORT_FEATURE_EN_SIZE_MASK		    0x07000000
+#define PORT_FEATURE_EN_SIZE_SHIFT		    24
+#define PORT_FEATURE_WOL_ENABLED		    0x01000000
+#define PORT_FEATURE_MBA_ENABLED		    0x02000000
+#define PORT_FEATURE_MFW_ENABLED		    0x04000000
+
+	u32 wol_config;
+	/* Default is used when driver sets to "auto" mode */
+#define PORT_FEATURE_WOL_DEFAULT_MASK		    0x00000003
+#define PORT_FEATURE_WOL_DEFAULT_SHIFT		    0
+#define PORT_FEATURE_WOL_DEFAULT_DISABLE	    0x00000000
+#define PORT_FEATURE_WOL_DEFAULT_MAGIC		    0x00000001
+#define PORT_FEATURE_WOL_DEFAULT_ACPI		    0x00000002
+#define PORT_FEATURE_WOL_DEFAULT_MAGIC_AND_ACPI     0x00000003
+#define PORT_FEATURE_WOL_RES_PAUSE_CAP		    0x00000004
+#define PORT_FEATURE_WOL_RES_ASYM_PAUSE_CAP	    0x00000008
+#define PORT_FEATURE_WOL_ACPI_UPON_MGMT 	    0x00000010
+
+	u32 mba_config;
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_MASK	    0x00000003
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_SHIFT	    0
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_PXE	    0x00000000
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_RPL	    0x00000001
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_BOOTP	    0x00000002
+#define PORT_FEATURE_MBA_BOOT_AGENT_TYPE_ISCSIB     0x00000003
+#define PORT_FEATURE_MBA_RES_PAUSE_CAP		    0x00000100
+#define PORT_FEATURE_MBA_RES_ASYM_PAUSE_CAP	    0x00000200
+#define PORT_FEATURE_MBA_SETUP_PROMPT_ENABLE	    0x00000400
+#define PORT_FEATURE_MBA_HOTKEY_CTRL_S		    0x00000000
+#define PORT_FEATURE_MBA_HOTKEY_CTRL_B		    0x00000800
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_MASK	    0x000ff000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_SHIFT	    12
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_DISABLED	    0x00000000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_2K	    0x00001000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_4K	    0x00002000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_8K	    0x00003000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_16K	    0x00004000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_32K	    0x00005000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_64K	    0x00006000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_128K	    0x00007000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_256K	    0x00008000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_512K	    0x00009000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_1M	    0x0000a000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_2M	    0x0000b000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_4M	    0x0000c000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_8M	    0x0000d000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_16M	    0x0000e000
+#define PORT_FEATURE_MBA_EXP_ROM_SIZE_32M	    0x0000f000
+#define PORT_FEATURE_MBA_MSG_TIMEOUT_MASK	    0x00f00000
+#define PORT_FEATURE_MBA_MSG_TIMEOUT_SHIFT	    20
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_MASK	    0x03000000
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_SHIFT	    24
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_AUTO	    0x00000000
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_BBS	    0x01000000
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_INT18H	    0x02000000
+#define PORT_FEATURE_MBA_BIOS_BOOTSTRAP_INT19H	    0x03000000
+#define PORT_FEATURE_MBA_LINK_SPEED_MASK	    0x3c000000
+#define PORT_FEATURE_MBA_LINK_SPEED_SHIFT	    26
+#define PORT_FEATURE_MBA_LINK_SPEED_AUTO	    0x00000000
+#define PORT_FEATURE_MBA_LINK_SPEED_10HD	    0x04000000
+#define PORT_FEATURE_MBA_LINK_SPEED_10FD	    0x08000000
+#define PORT_FEATURE_MBA_LINK_SPEED_100HD	    0x0c000000
+#define PORT_FEATURE_MBA_LINK_SPEED_100FD	    0x10000000
+#define PORT_FEATURE_MBA_LINK_SPEED_1GBPS	    0x14000000
+#define PORT_FEATURE_MBA_LINK_SPEED_2_5GBPS	    0x18000000
+#define PORT_FEATURE_MBA_LINK_SPEED_10GBPS_CX4	    0x1c000000
+#define PORT_FEATURE_MBA_LINK_SPEED_10GBPS_KX4	    0x20000000
+#define PORT_FEATURE_MBA_LINK_SPEED_10GBPS_KR	    0x24000000
+#define PORT_FEATURE_MBA_LINK_SPEED_12GBPS	    0x28000000
+#define PORT_FEATURE_MBA_LINK_SPEED_12_5GBPS	    0x2c000000
+#define PORT_FEATURE_MBA_LINK_SPEED_13GBPS	    0x30000000
+#define PORT_FEATURE_MBA_LINK_SPEED_15GBPS	    0x34000000
+#define PORT_FEATURE_MBA_LINK_SPEED_16GBPS	    0x38000000
+
+	u32 bmc_config;
+#define PORT_FEATURE_BMC_LINK_OVERRIDE_DEFAULT	    0x00000000
+#define PORT_FEATURE_BMC_LINK_OVERRIDE_EN	    0x00000001
+
+	u32 mba_vlan_cfg;
+#define PORT_FEATURE_MBA_VLAN_TAG_MASK		    0x0000ffff
+#define PORT_FEATURE_MBA_VLAN_TAG_SHIFT 	    0
+#define PORT_FEATURE_MBA_VLAN_EN		    0x00010000
+
+	u32 resource_cfg;
+#define PORT_FEATURE_RESOURCE_CFG_VALID 	    0x00000001
+#define PORT_FEATURE_RESOURCE_CFG_DIAG		    0x00000002
+#define PORT_FEATURE_RESOURCE_CFG_L2		    0x00000004
+#define PORT_FEATURE_RESOURCE_CFG_ISCSI 	    0x00000008
+#define PORT_FEATURE_RESOURCE_CFG_RDMA		    0x00000010
+
+	u32 smbus_config;
+	/* Obsolete */
+#define PORT_FEATURE_SMBUS_EN			    0x00000001
+#define PORT_FEATURE_SMBUS_ADDR_MASK		    0x000000fe
+#define PORT_FEATURE_SMBUS_ADDR_SHIFT		    1
+
+	u32 iscsib_boot_cfg;
+#define PORT_FEATURE_ISCSIB_SKIP_TARGET_BOOT	    0x00000001
+
+	u32 link_config;    /* Used as HW defaults for the driver */
+#define PORT_FEATURE_CONNECTED_SWITCH_MASK	    0x03000000
+#define PORT_FEATURE_CONNECTED_SWITCH_SHIFT	    24
+	/* (forced) low speed switch (< 10G) */
+#define PORT_FEATURE_CON_SWITCH_1G_SWITCH	    0x00000000
+	/* (forced) high speed switch (>= 10G) */
+#define PORT_FEATURE_CON_SWITCH_10G_SWITCH	    0x01000000
+#define PORT_FEATURE_CON_SWITCH_AUTO_DETECT	    0x02000000
+#define PORT_FEATURE_CON_SWITCH_ONE_TIME_DETECT     0x03000000
+
+#define PORT_FEATURE_LINK_SPEED_MASK		    0x000f0000
+#define PORT_FEATURE_LINK_SPEED_SHIFT		    16
+#define PORT_FEATURE_LINK_SPEED_AUTO		    0x00000000
+#define PORT_FEATURE_LINK_SPEED_10M_FULL	    0x00010000
+#define PORT_FEATURE_LINK_SPEED_10M_HALF	    0x00020000
+#define PORT_FEATURE_LINK_SPEED_100M_HALF	    0x00030000
+#define PORT_FEATURE_LINK_SPEED_100M_FULL	    0x00040000
+#define PORT_FEATURE_LINK_SPEED_1G		    0x00050000
+#define PORT_FEATURE_LINK_SPEED_2_5G		    0x00060000
+#define PORT_FEATURE_LINK_SPEED_10G_CX4 	    0x00070000
+#define PORT_FEATURE_LINK_SPEED_10G_KX4 	    0x00080000
+#define PORT_FEATURE_LINK_SPEED_10G_KR		    0x00090000
+#define PORT_FEATURE_LINK_SPEED_12G		    0x000a0000
+#define PORT_FEATURE_LINK_SPEED_12_5G		    0x000b0000
+#define PORT_FEATURE_LINK_SPEED_13G		    0x000c0000
+#define PORT_FEATURE_LINK_SPEED_15G		    0x000d0000
+#define PORT_FEATURE_LINK_SPEED_16G		    0x000e0000
+
+#define PORT_FEATURE_FLOW_CONTROL_MASK		    0x00000700
+#define PORT_FEATURE_FLOW_CONTROL_SHIFT 	    8
+#define PORT_FEATURE_FLOW_CONTROL_AUTO		    0x00000000
+#define PORT_FEATURE_FLOW_CONTROL_TX		    0x00000100
+#define PORT_FEATURE_FLOW_CONTROL_RX		    0x00000200
+#define PORT_FEATURE_FLOW_CONTROL_BOTH		    0x00000300
+#define PORT_FEATURE_FLOW_CONTROL_NONE		    0x00000400
+
+	/* The default for MCP link configuration,
+	   uses the same defines as link_config */
+	u32 mfw_wol_link_cfg;
+
+	u32 reserved[19];
+
+};
+
+
+/****************************************************************************
+ * Device Information							    *
+ ****************************************************************************/
+struct dev_info {						    /* size */
+
+	u32    bc_rev; /* 8 bits each: major, minor, build */	       /* 4 */
+
+	struct shared_hw_cfg	 shared_hw_config;		      /* 40 */
+
+	struct port_hw_cfg	 port_hw_config[FUNC_MAX];     /* 400*2=800 */
+
+	struct shared_feat_cfg	 shared_feature_config; 	       /* 4 */
+
+	struct port_feat_cfg	 port_feature_config[FUNC_MAX];/* 116*2=232 */
+
+};
+
+
+/****************************************************************************
+ * Management firmware state						    *
+ ****************************************************************************/
+/* Allocate 320 bytes for management firmware: still not known exactly
+ * how much IMD needs. */
+#define MGMTFW_STATE_WORD_SIZE				    80
+
+struct mgmtfw_state {
+	u32 opaque[MGMTFW_STATE_WORD_SIZE];
+};
+
+
+/****************************************************************************
+ * Shared Memory Region 						    *
+ ****************************************************************************/
+struct shmem_region {			       /*   SharedMem Offset (size) */
+	u32		    validity_map[FUNC_MAX];    /* 0x0 (4 * 2 = 0x8) */
+#define SHR_MEM_VALIDITY_PCI_CFG		    0x00000001
+#define SHR_MEM_VALIDITY_MB			    0x00000002
+#define SHR_MEM_VALIDITY_DEV_INFO		    0x00000004
+	/* One licensing bit should be set */
+#define SHR_MEM_VALIDITY_LIC_KEY_IN_EFFECT_MASK     0x00000038
+#define SHR_MEM_VALIDITY_LIC_MANUF_KEY_IN_EFFECT    0x00000008
+#define SHR_MEM_VALIDITY_LIC_UPGRADE_KEY_IN_EFFECT  0x00000010
+#define SHR_MEM_VALIDITY_LIC_NO_KEY_IN_EFFECT	    0x00000020
+
+	struct drv_fw_mb    drv_fw_mb[FUNC_MAX];     /* 0x8 (28 * 2 = 0x38) */
+
+	struct dev_info     dev_info;			    /* 0x40 (0x438) */
+
+#ifdef _LICENSE_H
+	license_key_t	    drv_lic_key[FUNC_MAX]; /* 0x478 (52 * 2 = 0x68) */
+#else /* Linux! */
+	u8		    reserved[52*FUNC_MAX];
+#endif
+
+	/* FW information (for internal FW use) */
+	u32		    fw_info_fio_offset; 	   /* 0x4e0 (0x4)   */
+	struct mgmtfw_state mgmtfw_state;		   /* 0x4e4 (0x140) */
+
+};							   /* 0x624 */
+
+
+#define BCM_5710_FW_MAJOR_VERSION			4
+#define BCM_5710_FW_MINOR_VERSION			0
+#define BCM_5710_FW_REVISION_VERSION			14
+#define BCM_5710_FW_COMPILE_FLAGS			1
+
+
+/*
+ * attention bits
+ */
+struct atten_def_status_block {
+	u32 attn_bits;
+	u32 attn_bits_ack;
+#if defined(__BIG_ENDIAN)
+	u16 attn_bits_index;
+	u8 reserved0;
+	u8 status_block_id;
+#elif defined(__LITTLE_ENDIAN)
+	u8 status_block_id;
+	u8 reserved0;
+	u16 attn_bits_index;
+#endif
+	u32 reserved1;
+};
+
+
+/*
+ * common data for all protocols
+ */
+struct doorbell_hdr {
+	u8 header;
+#define DOORBELL_HDR_RX (0x1<<0)
+#define DOORBELL_HDR_RX_SHIFT 0
+#define DOORBELL_HDR_DB_TYPE (0x1<<1)
+#define DOORBELL_HDR_DB_TYPE_SHIFT 1
+#define DOORBELL_HDR_DPM_SIZE (0x3<<2)
+#define DOORBELL_HDR_DPM_SIZE_SHIFT 2
+#define DOORBELL_HDR_CONN_TYPE (0xF<<4)
+#define DOORBELL_HDR_CONN_TYPE_SHIFT 4
+};
+
+/*
+ * doorbell message send to the chip
+ */
+struct doorbell {
+#if defined(__BIG_ENDIAN)
+	u16 zero_fill2;
+	u8 zero_fill1;
+	struct doorbell_hdr header;
+#elif defined(__LITTLE_ENDIAN)
+	struct doorbell_hdr header;
+	u8 zero_fill1;
+	u16 zero_fill2;
+#endif
+};
+
+
+/*
+ * IGU driver acknowlegement register
+ */
+struct igu_ack_register {
+#if defined(__BIG_ENDIAN)
+	u16 sb_id_and_flags;
+#define IGU_ACK_REGISTER_STATUS_BLOCK_ID (0x1F<<0)
+#define IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT 0
+#define IGU_ACK_REGISTER_STORM_ID (0x7<<5)
+#define IGU_ACK_REGISTER_STORM_ID_SHIFT 5
+#define IGU_ACK_REGISTER_UPDATE_INDEX (0x1<<8)
+#define IGU_ACK_REGISTER_UPDATE_INDEX_SHIFT 8
+#define IGU_ACK_REGISTER_INTERRUPT_MODE (0x3<<9)
+#define IGU_ACK_REGISTER_INTERRUPT_MODE_SHIFT 9
+#define IGU_ACK_REGISTER_RESERVED (0x1F<<11)
+#define IGU_ACK_REGISTER_RESERVED_SHIFT 11
+	u16 status_block_index;
+#elif defined(__LITTLE_ENDIAN)
+	u16 status_block_index;
+	u16 sb_id_and_flags;
+#define IGU_ACK_REGISTER_STATUS_BLOCK_ID (0x1F<<0)
+#define IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT 0
+#define IGU_ACK_REGISTER_STORM_ID (0x7<<5)
+#define IGU_ACK_REGISTER_STORM_ID_SHIFT 5
+#define IGU_ACK_REGISTER_UPDATE_INDEX (0x1<<8)
+#define IGU_ACK_REGISTER_UPDATE_INDEX_SHIFT 8
+#define IGU_ACK_REGISTER_INTERRUPT_MODE (0x3<<9)
+#define IGU_ACK_REGISTER_INTERRUPT_MODE_SHIFT 9
+#define IGU_ACK_REGISTER_RESERVED (0x1F<<11)
+#define IGU_ACK_REGISTER_RESERVED_SHIFT 11
+#endif
+};
+
+
+/*
+ * Parser parsing flags field
+ */
+struct parsing_flags {
+	u16 flags;
+#define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE (0x1<<0)
+#define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE_SHIFT 0
+#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS (0x3<<1)
+#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS_SHIFT 1
+#define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL (0x3<<3)
+#define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT 3
+#define PARSING_FLAGS_IP_OPTIONS (0x1<<5)
+#define PARSING_FLAGS_IP_OPTIONS_SHIFT 5
+#define PARSING_FLAGS_FRAGMENTATION_STATUS (0x1<<6)
+#define PARSING_FLAGS_FRAGMENTATION_STATUS_SHIFT 6
+#define PARSING_FLAGS_OVER_IP_PROTOCOL (0x3<<7)
+#define PARSING_FLAGS_OVER_IP_PROTOCOL_SHIFT 7
+#define PARSING_FLAGS_PURE_ACK_INDICATION (0x1<<9)
+#define PARSING_FLAGS_PURE_ACK_INDICATION_SHIFT 9
+#define PARSING_FLAGS_TCP_OPTIONS_EXIST (0x1<<10)
+#define PARSING_FLAGS_TCP_OPTIONS_EXIST_SHIFT 10
+#define PARSING_FLAGS_TIME_STAMP_EXIST_FLAG (0x1<<11)
+#define PARSING_FLAGS_TIME_STAMP_EXIST_FLAG_SHIFT 11
+#define PARSING_FLAGS_CONNECTION_MATCH (0x1<<12)
+#define PARSING_FLAGS_CONNECTION_MATCH_SHIFT 12
+#define PARSING_FLAGS_LLC_SNAP (0x1<<13)
+#define PARSING_FLAGS_LLC_SNAP_SHIFT 13
+#define PARSING_FLAGS_RESERVED0 (0x3<<14)
+#define PARSING_FLAGS_RESERVED0_SHIFT 14
+};
+
+
+/*
+ * dmae command structure
+ */
+struct dmae_command {
+	u32 opcode;
+#define DMAE_COMMAND_SRC (0x1<<0)
+#define DMAE_COMMAND_SRC_SHIFT 0
+#define DMAE_COMMAND_DST (0x3<<1)
+#define DMAE_COMMAND_DST_SHIFT 1
+#define DMAE_COMMAND_C_DST (0x1<<3)
+#define DMAE_COMMAND_C_DST_SHIFT 3
+#define DMAE_COMMAND_C_TYPE_ENABLE (0x1<<4)
+#define DMAE_COMMAND_C_TYPE_ENABLE_SHIFT 4
+#define DMAE_COMMAND_C_TYPE_CRC_ENABLE (0x1<<5)
+#define DMAE_COMMAND_C_TYPE_CRC_ENABLE_SHIFT 5
+#define DMAE_COMMAND_C_TYPE_CRC_OFFSET (0x7<<6)
+#define DMAE_COMMAND_C_TYPE_CRC_OFFSET_SHIFT 6
+#define DMAE_COMMAND_ENDIANITY (0x3<<9)
+#define DMAE_COMMAND_ENDIANITY_SHIFT 9
+#define DMAE_COMMAND_PORT (0x1<<11)
+#define DMAE_COMMAND_PORT_SHIFT 11
+#define DMAE_COMMAND_CRC_RESET (0x1<<12)
+#define DMAE_COMMAND_CRC_RESET_SHIFT 12
+#define DMAE_COMMAND_SRC_RESET (0x1<<13)
+#define DMAE_COMMAND_SRC_RESET_SHIFT 13
+#define DMAE_COMMAND_DST_RESET (0x1<<14)
+#define DMAE_COMMAND_DST_RESET_SHIFT 14
+#define DMAE_COMMAND_RESERVED0 (0x1FFFF<<15)
+#define DMAE_COMMAND_RESERVED0_SHIFT 15
+	u32 src_addr_lo;
+	u32 src_addr_hi;
+	u32 dst_addr_lo;
+	u32 dst_addr_hi;
+#if defined(__BIG_ENDIAN)
+	u16 reserved1;
+	u16 len;
+#elif defined(__LITTLE_ENDIAN)
+	u16 len;
+	u16 reserved1;
+#endif
+	u32 comp_addr_lo;
+	u32 comp_addr_hi;
+	u32 comp_val;
+	u32 crc32;
+	u32 crc32_c;
+#if defined(__BIG_ENDIAN)
+	u16 crc16_c;
+	u16 crc16;
+#elif defined(__LITTLE_ENDIAN)
+	u16 crc16;
+	u16 crc16_c;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 reserved2;
+	u16 crc_t10;
+#elif defined(__LITTLE_ENDIAN)
+	u16 crc_t10;
+	u16 reserved2;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 xsum8;
+	u16 xsum16;
+#elif defined(__LITTLE_ENDIAN)
+	u16 xsum16;
+	u16 xsum8;
+#endif
+};
+
+
+struct double_regpair {
+	u32 regpair0_lo;
+	u32 regpair0_hi;
+	u32 regpair1_lo;
+	u32 regpair1_hi;
+};
+
+
+/*
+ * The eth Rx Buffer Descriptor
+ */
+struct eth_rx_bd {
+	u32 addr_lo;
+	u32 addr_hi;
+};
+
+/*
+ * The eth storm context of Ustorm
+ */
+struct ustorm_eth_st_context {
+#if defined(__BIG_ENDIAN)
+	u8 sb_index_number;
+	u8 status_block_id;
+	u8 __local_rx_bd_cons;
+	u8 __local_rx_bd_prod;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __local_rx_bd_prod;
+	u8 __local_rx_bd_cons;
+	u8 status_block_id;
+	u8 sb_index_number;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 rcq_cons;
+	u16 rx_bd_cons;
+#elif defined(__LITTLE_ENDIAN)
+	u16 rx_bd_cons;
+	u16 rcq_cons;
+#endif
+	u32 rx_bd_page_base_lo;
+	u32 rx_bd_page_base_hi;
+	u32 rcq_base_address_lo;
+	u32 rcq_base_address_hi;
+#if defined(__BIG_ENDIAN)
+	u16 __num_of_returned_cqes;
+	u8 num_rss;
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
+#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
+#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
+#elif defined(__LITTLE_ENDIAN)
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
+#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
+#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
+	u8 num_rss;
+	u16 __num_of_returned_cqes;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 mc_alignment_size;
+	u16 agg_threshold;
+#elif defined(__LITTLE_ENDIAN)
+	u16 agg_threshold;
+	u16 mc_alignment_size;
+#endif
+	struct eth_rx_bd __local_bd_ring[16];
+};
+
+/*
+ * The eth storm context of Tstorm
+ */
+struct tstorm_eth_st_context {
+	u32 __reserved0[28];
+};
+
+/*
+ * The eth aggregative context section of Xstorm
+ */
+struct xstorm_eth_extra_ag_context_section {
+#if defined(__BIG_ENDIAN)
+	u8 __tcp_agg_vars1;
+	u8 __reserved50;
+	u16 __mss;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __mss;
+	u8 __reserved50;
+	u8 __tcp_agg_vars1;
+#endif
+	u32 __snd_nxt;
+	u32 __tx_wnd;
+	u32 __snd_una;
+	u32 __reserved53;
+#if defined(__BIG_ENDIAN)
+	u8 __agg_val8_th;
+	u8 __agg_val8;
+	u16 __tcp_agg_vars2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __tcp_agg_vars2;
+	u8 __agg_val8;
+	u8 __agg_val8_th;
+#endif
+	u32 __reserved58;
+	u32 __reserved59;
+	u32 __reserved60;
+	u32 __reserved61;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val7_th;
+	u16 __agg_val7;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val7;
+	u16 __agg_val7_th;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 __tcp_agg_vars5;
+	u8 __tcp_agg_vars4;
+	u8 __tcp_agg_vars3;
+	u8 __reserved62;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __reserved62;
+	u8 __tcp_agg_vars3;
+	u8 __tcp_agg_vars4;
+	u8 __tcp_agg_vars5;
+#endif
+	u32 __tcp_agg_vars6;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_misc6;
+	u16 __tcp_agg_vars7;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __tcp_agg_vars7;
+	u16 __agg_misc6;
+#endif
+	u32 __agg_val10;
+	u32 __agg_val10_th;
+#if defined(__BIG_ENDIAN)
+	u16 __reserved3;
+	u8 __reserved2;
+	u8 __agg_misc7;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __agg_misc7;
+	u8 __reserved2;
+	u16 __reserved3;
+#endif
+};
+
+/*
+ * The eth aggregative context of Xstorm
+ */
+struct xstorm_eth_ag_context {
+#if defined(__BIG_ENDIAN)
+	u16 __bd_prod;
+	u8 __agg_vars1;
+	u8 __state;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __state;
+	u8 __agg_vars1;
+	u16 __bd_prod;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 cdu_reserved;
+	u8 __agg_vars4;
+	u8 __agg_vars3;
+	u8 __agg_vars2;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __agg_vars2;
+	u8 __agg_vars3;
+	u8 __agg_vars4;
+	u8 cdu_reserved;
+#endif
+	u32 __more_packets_to_send;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_vars5;
+	u16 __agg_val4_th;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val4_th;
+	u16 __agg_vars5;
+#endif
+	struct xstorm_eth_extra_ag_context_section __extra_section;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_vars7;
+	u8 __agg_val3_th;
+	u8 __agg_vars6;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __agg_vars6;
+	u8 __agg_val3_th;
+	u16 __agg_vars7;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val11_th;
+	u16 __agg_val11;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val11;
+	u16 __agg_val11_th;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 __reserved1;
+	u8 __agg_val6_th;
+	u16 __agg_val9;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val9;
+	u8 __agg_val6_th;
+	u8 __reserved1;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val2_th;
+	u16 __agg_val2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val2;
+	u16 __agg_val2_th;
+#endif
+	u32 __agg_vars8;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_misc0;
+	u16 __agg_val4;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val4;
+	u16 __agg_misc0;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 __agg_val3;
+	u8 __agg_val6;
+	u8 __agg_val5_th;
+	u8 __agg_val5;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __agg_val5;
+	u8 __agg_val5_th;
+	u8 __agg_val6;
+	u8 __agg_val3;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __agg_misc1;
+	u16 __bd_ind_max_val;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __bd_ind_max_val;
+	u16 __agg_misc1;
+#endif
+	u32 __reserved57;
+	u32 __agg_misc4;
+	u32 __agg_misc5;
+};
+
+/*
+ * The eth aggregative context section of Tstorm
+ */
+struct tstorm_eth_extra_ag_context_section {
+	u32 __agg_val1;
+#if defined(__BIG_ENDIAN)
+	u8 __tcp_agg_vars2;
+	u8 __agg_val3;
+	u16 __agg_val2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val2;
+	u8 __agg_val3;
+	u8 __tcp_agg_vars2;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val5;
+	u8 __agg_val6;
+	u8 __tcp_agg_vars3;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __tcp_agg_vars3;
+	u8 __agg_val6;
+	u16 __agg_val5;
+#endif
+	u32 __reserved63;
+	u32 __reserved64;
+	u32 __reserved65;
+	u32 __reserved66;
+	u32 __reserved67;
+	u32 __tcp_agg_vars1;
+	u32 __reserved61;
+	u32 __reserved62;
+	u32 __reserved2;
+};
+
+/*
+ * The eth aggregative context of Tstorm
+ */
+struct tstorm_eth_ag_context {
+#if defined(__BIG_ENDIAN)
+	u16 __reserved54;
+	u8 __agg_vars1;
+	u8 __state;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __state;
+	u8 __agg_vars1;
+	u16 __reserved54;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val4;
+	u16 __agg_vars2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_vars2;
+	u16 __agg_val4;
+#endif
+	struct tstorm_eth_extra_ag_context_section __extra_section;
+};
+
+/*
+ * The eth aggregative context of Cstorm
+ */
+struct cstorm_eth_ag_context {
+	u32 __agg_vars1;
+#if defined(__BIG_ENDIAN)
+	u8 __aux1_th;
+	u8 __aux1_val;
+	u16 __agg_vars2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_vars2;
+	u8 __aux1_val;
+	u8 __aux1_th;
+#endif
+	u32 __num_of_treated_packet;
+	u32 __last_packet_treated;
+#if defined(__BIG_ENDIAN)
+	u16 __reserved58;
+	u16 __reserved57;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __reserved57;
+	u16 __reserved58;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 __reserved62;
+	u8 __reserved61;
+	u8 __reserved60;
+	u8 __reserved59;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __reserved59;
+	u8 __reserved60;
+	u8 __reserved61;
+	u8 __reserved62;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __reserved64;
+	u16 __reserved63;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __reserved63;
+	u16 __reserved64;
+#endif
+	u32 __reserved65;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_vars3;
+	u16 __rq_inv_cnt;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __rq_inv_cnt;
+	u16 __agg_vars3;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __packet_index_th;
+	u16 __packet_index;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __packet_index;
+	u16 __packet_index_th;
+#endif
+};
+
+/*
+ * The eth aggregative context of Ustorm
+ */
+struct ustorm_eth_ag_context {
+#if defined(__BIG_ENDIAN)
+	u8 __aux_counter_flags;
+	u8 __agg_vars2;
+	u8 __agg_vars1;
+	u8 __state;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __state;
+	u8 __agg_vars1;
+	u8 __agg_vars2;
+	u8 __aux_counter_flags;
+#endif
+#if defined(__BIG_ENDIAN)
+	u8 cdu_usage;
+	u8 __agg_misc2;
+	u16 __agg_misc1;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_misc1;
+	u8 __agg_misc2;
+	u8 cdu_usage;
+#endif
+	u32 __agg_misc4;
+#if defined(__BIG_ENDIAN)
+	u8 __agg_val3_th;
+	u8 __agg_val3;
+	u16 __agg_misc3;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_misc3;
+	u8 __agg_val3;
+	u8 __agg_val3_th;
+#endif
+	u32 __agg_val1;
+	u32 __agg_misc4_th;
+#if defined(__BIG_ENDIAN)
+	u16 __agg_val2_th;
+	u16 __agg_val2;
+#elif defined(__LITTLE_ENDIAN)
+	u16 __agg_val2;
+	u16 __agg_val2_th;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 __reserved2;
+	u8 __decision_rules;
+	u8 __decision_rule_enable_bits;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __decision_rule_enable_bits;
+	u8 __decision_rules;
+	u16 __reserved2;
+#endif
+};
+
+/*
+ * Timers connection context
+ */
+struct timers_block_context {
+	u32 __reserved_0;
+	u32 __reserved_1;
+	u32 __reserved_2;
+	u32 __reserved_flags;
+};
+
+/*
+ * structure for easy accessability to assembler
+ */
+struct eth_tx_bd_flags {
+	u8 as_bitfield;
+#define ETH_TX_BD_FLAGS_VLAN_TAG (0x1<<0)
+#define ETH_TX_BD_FLAGS_VLAN_TAG_SHIFT 0
+#define ETH_TX_BD_FLAGS_IP_CSUM (0x1<<1)
+#define ETH_TX_BD_FLAGS_IP_CSUM_SHIFT 1
+#define ETH_TX_BD_FLAGS_TCP_CSUM (0x1<<2)
+#define ETH_TX_BD_FLAGS_TCP_CSUM_SHIFT 2
+#define ETH_TX_BD_FLAGS_END_BD (0x1<<3)
+#define ETH_TX_BD_FLAGS_END_BD_SHIFT 3
+#define ETH_TX_BD_FLAGS_START_BD (0x1<<4)
+#define ETH_TX_BD_FLAGS_START_BD_SHIFT 4
+#define ETH_TX_BD_FLAGS_HDR_POOL (0x1<<5)
+#define ETH_TX_BD_FLAGS_HDR_POOL_SHIFT 5
+#define ETH_TX_BD_FLAGS_SW_LSO (0x1<<6)
+#define ETH_TX_BD_FLAGS_SW_LSO_SHIFT 6
+#define ETH_TX_BD_FLAGS_IPV6 (0x1<<7)
+#define ETH_TX_BD_FLAGS_IPV6_SHIFT 7
+};
+
+/*
+ * The eth Tx Buffer Descriptor
+ */
+struct eth_tx_bd {
+	u32 addr_lo;
+	u32 addr_hi;
+	u16 nbd;
+	u16 nbytes;
+	u16 vlan;
+	struct eth_tx_bd_flags bd_flags;
+	u8 general_data;
+#define ETH_TX_BD_HDR_NBDS (0x3F<<0)
+#define ETH_TX_BD_HDR_NBDS_SHIFT 0
+#define ETH_TX_BD_ETH_ADDR_TYPE (0x3<<6)
+#define ETH_TX_BD_ETH_ADDR_TYPE_SHIFT 6
+};
+
+/*
+ * Tx parsing BD structure for ETH,Relevant in START
+ */
+struct eth_tx_parse_bd {
+	u8 global_data;
+#define ETH_TX_PARSE_BD_IP_HDR_START_OFFSET (0xF<<0)
+#define ETH_TX_PARSE_BD_IP_HDR_START_OFFSET_SHIFT 0
+#define ETH_TX_PARSE_BD_CS_ANY_FLG (0x1<<4)
+#define ETH_TX_PARSE_BD_CS_ANY_FLG_SHIFT 4
+#define ETH_TX_PARSE_BD_PSEUDO_CS_WITHOUT_LEN (0x1<<5)
+#define ETH_TX_PARSE_BD_PSEUDO_CS_WITHOUT_LEN_SHIFT 5
+#define ETH_TX_PARSE_BD_LLC_SNAP_EN (0x1<<6)
+#define ETH_TX_PARSE_BD_LLC_SNAP_EN_SHIFT 6
+#define ETH_TX_PARSE_BD_NS_FLG (0x1<<7)
+#define ETH_TX_PARSE_BD_NS_FLG_SHIFT 7
+	u8 tcp_flags;
+#define ETH_TX_PARSE_BD_FIN_FLG (0x1<<0)
+#define ETH_TX_PARSE_BD_FIN_FLG_SHIFT 0
+#define ETH_TX_PARSE_BD_SYN_FLG (0x1<<1)
+#define ETH_TX_PARSE_BD_SYN_FLG_SHIFT 1
+#define ETH_TX_PARSE_BD_RST_FLG (0x1<<2)
+#define ETH_TX_PARSE_BD_RST_FLG_SHIFT 2
+#define ETH_TX_PARSE_BD_PSH_FLG (0x1<<3)
+#define ETH_TX_PARSE_BD_PSH_FLG_SHIFT 3
+#define ETH_TX_PARSE_BD_ACK_FLG (0x1<<4)
+#define ETH_TX_PARSE_BD_ACK_FLG_SHIFT 4
+#define ETH_TX_PARSE_BD_URG_FLG (0x1<<5)
+#define ETH_TX_PARSE_BD_URG_FLG_SHIFT 5
+#define ETH_TX_PARSE_BD_ECE_FLG (0x1<<6)
+#define ETH_TX_PARSE_BD_ECE_FLG_SHIFT 6
+#define ETH_TX_PARSE_BD_CWR_FLG (0x1<<7)
+#define ETH_TX_PARSE_BD_CWR_FLG_SHIFT 7
+	u8 ip_hlen;
+	s8 cs_offset;
+	u16 total_hlen;
+	u16 lso_mss;
+	u16 tcp_pseudo_csum;
+	u16 ip_id;
+	u32 tcp_send_seq;
+};
+
+/*
+ * The last BD in the BD memory will hold a pointer to the next BD memory
+ */
+struct eth_tx_next_bd {
+	u32 addr_lo;
+	u32 addr_hi;
+	u8 reserved[8];
+};
+
+/*
+ * union for 3 Bd types
+ */
+union eth_tx_bd_types {
+	struct eth_tx_bd reg_bd;
+	struct eth_tx_parse_bd parse_bd;
+	struct eth_tx_next_bd next_bd;
+};
+
+/*
+ * The eth storm context of Xstorm
+ */
+struct xstorm_eth_st_context {
+	u32 tx_bd_page_base_lo;
+	u32 tx_bd_page_base_hi;
+#if defined(__BIG_ENDIAN)
+	u16 tx_bd_cons;
+	u8 __reserved0;
+	u8 __local_tx_bd_prod;
+#elif defined(__LITTLE_ENDIAN)
+	u8 __local_tx_bd_prod;
+	u8 __reserved0;
+	u16 tx_bd_cons;
+#endif
+	u32 db_data_addr_lo;
+	u32 db_data_addr_hi;
+	u32 __pkt_cons;
+	u32 __gso_next;
+	u32 is_eth_conn_1b;
+	union eth_tx_bd_types __bds[13];
+};
+
+/*
+ * The eth storm context of Cstorm
+ */
+struct cstorm_eth_st_context {
+#if defined(__BIG_ENDIAN)
+	u16 __reserved0;
+	u8 sb_index_number;
+	u8 status_block_id;
+#elif defined(__LITTLE_ENDIAN)
+	u8 status_block_id;
+	u8 sb_index_number;
+	u16 __reserved0;
+#endif
+	u32 __reserved1[3];
+};
+
+/*
+ * Ethernet connection context
+ */
+struct eth_context {
+	struct ustorm_eth_st_context ustorm_st_context;
+	struct tstorm_eth_st_context tstorm_st_context;
+	struct xstorm_eth_ag_context xstorm_ag_context;
+	struct tstorm_eth_ag_context tstorm_ag_context;
+	struct cstorm_eth_ag_context cstorm_ag_context;
+	struct ustorm_eth_ag_context ustorm_ag_context;
+	struct timers_block_context timers_context;
+	struct xstorm_eth_st_context xstorm_st_context;
+	struct cstorm_eth_st_context cstorm_st_context;
+};
+
+
+/*
+ * ethernet doorbell
+ */
+struct eth_tx_doorbell {
+#if defined(__BIG_ENDIAN)
+	u16 npackets;
+	u8 params;
+#define ETH_TX_DOORBELL_NUM_BDS (0x3F<<0)
+#define ETH_TX_DOORBELL_NUM_BDS_SHIFT 0
+#define ETH_TX_DOORBELL_RESERVED_TX_FIN_FLAG (0x1<<6)
+#define ETH_TX_DOORBELL_RESERVED_TX_FIN_FLAG_SHIFT 6
+#define ETH_TX_DOORBELL_SPARE (0x1<<7)
+#define ETH_TX_DOORBELL_SPARE_SHIFT 7
+	struct doorbell_hdr hdr;
+#elif defined(__LITTLE_ENDIAN)
+	struct doorbell_hdr hdr;
+	u8 params;
+#define ETH_TX_DOORBELL_NUM_BDS (0x3F<<0)
+#define ETH_TX_DOORBELL_NUM_BDS_SHIFT 0
+#define ETH_TX_DOORBELL_RESERVED_TX_FIN_FLAG (0x1<<6)
+#define ETH_TX_DOORBELL_RESERVED_TX_FIN_FLAG_SHIFT 6
+#define ETH_TX_DOORBELL_SPARE (0x1<<7)
+#define ETH_TX_DOORBELL_SPARE_SHIFT 7
+	u16 npackets;
+#endif
+};
+
+
+/*
+ * ustorm status block
+ */
+struct ustorm_def_status_block {
+	u16 index_values[HC_USTORM_DEF_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * cstorm status block
+ */
+struct cstorm_def_status_block {
+	u16 index_values[HC_CSTORM_DEF_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * xstorm status block
+ */
+struct xstorm_def_status_block {
+	u16 index_values[HC_XSTORM_DEF_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * tstorm status block
+ */
+struct tstorm_def_status_block {
+	u16 index_values[HC_TSTORM_DEF_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * host status block
+ */
+struct host_def_status_block {
+	struct atten_def_status_block atten_status_block;
+	struct ustorm_def_status_block u_def_status_block;
+	struct cstorm_def_status_block c_def_status_block;
+	struct xstorm_def_status_block x_def_status_block;
+	struct tstorm_def_status_block t_def_status_block;
+};
+
+
+/*
+ * ustorm status block
+ */
+struct ustorm_status_block {
+	u16 index_values[HC_USTORM_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * cstorm status block
+ */
+struct cstorm_status_block {
+	u16 index_values[HC_CSTORM_SB_NUM_INDICES];
+	u16 status_block_index;
+	u8 reserved0;
+	u8 status_block_id;
+	u32 __flags;
+};
+
+/*
+ * host status block
+ */
+struct host_status_block {
+	struct ustorm_status_block u_status_block;
+	struct cstorm_status_block c_status_block;
+};
+
+
+/*
+ * The data for RSS setup ramrod
+ */
+struct eth_client_setup_ramrod_data {
+	u32 client_id_5b;
+	u8 is_rdma_1b;
+	u8 reserved0;
+	u16 reserved1;
+};
+
+
+/*
+ * L2 dynamic host coalescing init parameters
+ */
+struct eth_dynamic_hc_config {
+	u32 threshold[3];
+	u8 hc_timeout[4];
+};
+
+
+/*
+ * regular eth FP CQE parameters struct
+ */
+struct eth_fast_path_rx_cqe {
+	u8 type;
+	u8 error_type_flags;
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG (0x1<<0)
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG_SHIFT 0
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG (0x1<<1)
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG_SHIFT 1
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG (0x1<<2)
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG_SHIFT 2
+#define ETH_FAST_PATH_RX_CQE_START_FLG (0x1<<3)
+#define ETH_FAST_PATH_RX_CQE_START_FLG_SHIFT 3
+#define ETH_FAST_PATH_RX_CQE_END_FLG (0x1<<4)
+#define ETH_FAST_PATH_RX_CQE_END_FLG_SHIFT 4
+#define ETH_FAST_PATH_RX_CQE_RESERVED0 (0x7<<5)
+#define ETH_FAST_PATH_RX_CQE_RESERVED0_SHIFT 5
+	u8 status_flags;
+#define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE (0x7<<0)
+#define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE_SHIFT 0
+#define ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG (0x1<<3)
+#define ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG_SHIFT 3
+#define ETH_FAST_PATH_RX_CQE_BROADCAST_FLG (0x1<<4)
+#define ETH_FAST_PATH_RX_CQE_BROADCAST_FLG_SHIFT 4
+#define ETH_FAST_PATH_RX_CQE_MAC_MATCH_FLG (0x1<<5)
+#define ETH_FAST_PATH_RX_CQE_MAC_MATCH_FLG_SHIFT 5
+#define ETH_FAST_PATH_RX_CQE_IP_XSUM_NO_VALIDATION_FLG (0x1<<6)
+#define ETH_FAST_PATH_RX_CQE_IP_XSUM_NO_VALIDATION_FLG_SHIFT 6
+#define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG (0x1<<7)
+#define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG_SHIFT 7
+	u8 placement_offset;
+	u32 rss_hash_result;
+	u16 vlan_tag;
+	u16 pkt_len;
+	u16 queue_index;
+	struct parsing_flags pars_flags;
+};
+
+
+/*
+ * The data for RSS setup ramrod
+ */
+struct eth_halt_ramrod_data {
+	u32 client_id_5b;
+	u32 reserved0;
+};
+
+
+/*
+ * Place holder for ramrods protocol specific data
+ */
+struct ramrod_data {
+	u32 data_lo;
+	u32 data_hi;
+};
+
+/*
+ * union for ramrod data for ethernet protocol (CQE) (force size of 16 bits)
+ */
+union eth_ramrod_data {
+	struct ramrod_data general;
+};
+
+
+/*
+ * Rx Last BD in page (in ETH)
+ */
+struct eth_rx_bd_next_page {
+	u32 addr_lo;
+	u32 addr_hi;
+	u8 reserved[8];
+};
+
+
+/*
+ * Eth Rx Cqe structure- general structure for ramrods
+ */
+struct common_ramrod_eth_rx_cqe {
+	u8 type;
+	u8 conn_type_3b;
+	u16 reserved;
+	u32 conn_and_cmd_data;
+#define COMMON_RAMROD_ETH_RX_CQE_CID (0xFFFFFF<<0)
+#define COMMON_RAMROD_ETH_RX_CQE_CID_SHIFT 0
+#define COMMON_RAMROD_ETH_RX_CQE_CMD_ID (0xFF<<24)
+#define COMMON_RAMROD_ETH_RX_CQE_CMD_ID_SHIFT 24
+	struct ramrod_data protocol_data;
+};
+
+/*
+ * Rx Last CQE in page (in ETH)
+ */
+struct eth_rx_cqe_next_page {
+	u32 addr_lo;
+	u32 addr_hi;
+	u32 reserved0;
+	u32 reserved1;
+};
+
+/*
+ * union for all eth rx cqe types (fix their sizes)
+ */
+union eth_rx_cqe {
+	struct eth_fast_path_rx_cqe fast_path_cqe;
+	struct common_ramrod_eth_rx_cqe ramrod_cqe;
+	struct eth_rx_cqe_next_page next_page_cqe;
+};
+
+
+/*
+ * common data for all protocols
+ */
+struct spe_hdr {
+	u32 conn_and_cmd_data;
+#define SPE_HDR_CID (0xFFFFFF<<0)
+#define SPE_HDR_CID_SHIFT 0
+#define SPE_HDR_CMD_ID (0xFF<<24)
+#define SPE_HDR_CMD_ID_SHIFT 24
+	u16 type;
+#define SPE_HDR_CONN_TYPE (0xFF<<0)
+#define SPE_HDR_CONN_TYPE_SHIFT 0
+#define SPE_HDR_COMMON_RAMROD (0xFF<<8)
+#define SPE_HDR_COMMON_RAMROD_SHIFT 8
+	u16 reserved;
+};
+
+struct regpair {
+	u32 lo;
+	u32 hi;
+};
+
+/*
+ * ethernet slow path element
+ */
+union eth_specific_data {
+	u8 protocol_data[8];
+	struct regpair mac_config_addr;
+	struct eth_client_setup_ramrod_data client_setup_ramrod_data;
+	struct eth_halt_ramrod_data halt_ramrod_data;
+	struct regpair leading_cqe_addr;
+	struct regpair update_data_addr;
+};
+
+/*
+ * ethernet slow path element
+ */
+struct eth_spe {
+	struct spe_hdr hdr;
+	union eth_specific_data data;
+};
+
+
+/*
+ * doorbell data in host memory
+ */
+struct eth_tx_db_data {
+	u32 packets_prod;
+	u16 bds_prod;
+	u16 reserved;
+};
+
+
+/*
+ * Common configuration parameters per port in Tstorm
+ */
+struct tstorm_eth_function_common_config {
+	u32 config_flags;
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY (0x1<<0)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY_SHIFT 0
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY (0x1<<1)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY_SHIFT 1
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY (0x1<<2)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY_SHIFT 2
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY (0x1<<3)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY_SHIFT 3
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE (0x1<<4)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE_SHIFT 4
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE (0x1<<5)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE_SHIFT 5
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x3FFFFFF<<6)
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 6
+#if defined(__BIG_ENDIAN)
+	u16 __secondary_vlan_id;
+	u8 leading_client_id;
+	u8 rss_result_mask;
+#elif defined(__LITTLE_ENDIAN)
+	u8 rss_result_mask;
+	u8 leading_client_id;
+	u16 __secondary_vlan_id;
+#endif
+};
+
+/*
+ * parameters for eth update ramrod
+ */
+struct eth_update_ramrod_data {
+	struct tstorm_eth_function_common_config func_config;
+	u8 indirectionTable[128];
+};
+
+
+/*
+ * MAC filtering configuration command header
+ */
+struct mac_configuration_hdr {
+	u8 length_6b;
+	u8 offset;
+	u16 reserved0;
+	u32 reserved1;
+};
+
+/*
+ * MAC address in list for ramrod
+ */
+struct tstorm_cam_entry {
+	u16 lsb_mac_addr;
+	u16 middle_mac_addr;
+	u16 msb_mac_addr;
+	u16 flags;
+#define TSTORM_CAM_ENTRY_PORT_ID (0x1<<0)
+#define TSTORM_CAM_ENTRY_PORT_ID_SHIFT 0
+#define TSTORM_CAM_ENTRY_RSRVVAL0 (0x7<<1)
+#define TSTORM_CAM_ENTRY_RSRVVAL0_SHIFT 1
+#define TSTORM_CAM_ENTRY_RESERVED0 (0xFFF<<4)
+#define TSTORM_CAM_ENTRY_RESERVED0_SHIFT 4
+};
+
+/*
+ * MAC filtering: CAM target table entry
+ */
+struct tstorm_cam_target_table_entry {
+	u8 flags;
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_BROADCAST (0x1<<0)
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_BROADCAST_SHIFT 0
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_OVERRIDE_VLAN_REMOVAL (0x1<<1)
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_OVERRIDE_VLAN_REMOVAL_SHIFT 1
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE (0x1<<2)
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE_SHIFT 2
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_RDMA_MAC (0x1<<3)
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_RDMA_MAC_SHIFT 3
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_RESERVED0 (0xF<<4)
+#define TSTORM_CAM_TARGET_TABLE_ENTRY_RESERVED0_SHIFT 4
+	u8 client_id;
+	u16 vlan_id;
+};
+
+/*
+ * MAC address in list for ramrod
+ */
+struct mac_configuration_entry {
+	struct tstorm_cam_entry cam_entry;
+	struct tstorm_cam_target_table_entry target_table_entry;
+};
+
+/*
+ * MAC filtering configuration command
+ */
+struct mac_configuration_cmd {
+	struct mac_configuration_hdr hdr;
+	struct mac_configuration_entry config_table[64];
+};
+
+
+/*
+ * Configuration parameters per client in Tstorm
+ */
+struct tstorm_eth_client_config {
+#if defined(__BIG_ENDIAN)
+	u16 statistics_counter_id;
+	u16 mtu;
+#elif defined(__LITTLE_ENDIAN)
+	u16 mtu;
+	u16 statistics_counter_id;
+#endif
+#if defined(__BIG_ENDIAN)
+	u16 drop_flags;
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR (0x1<<0)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+	u16 config_flags;
+#define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
+#define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
+#define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
+#define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+#elif defined(__LITTLE_ENDIAN)
+	u16 config_flags;
+#define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
+#define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
+#define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
+#define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+	u16 drop_flags;
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR (0x1<<0)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+#endif
+};
+
+
+/*
+ * MAC filtering configuration parameters per port in Tstorm
+ */
+struct tstorm_eth_mac_filter_config {
+	u32 ucast_drop_all;
+	u32 ucast_accept_all;
+	u32 mcast_drop_all;
+	u32 mcast_accept_all;
+	u32 bcast_drop_all;
+	u32 bcast_accept_all;
+	u32 strict_vlan;
+	u32 __secondary_vlan_clients;
+};
+
+
+struct rate_shaping_per_protocol {
+#if defined(__BIG_ENDIAN)
+	u16 reserved0;
+	u16 protocol_rate;
+#elif defined(__LITTLE_ENDIAN)
+	u16 protocol_rate;
+	u16 reserved0;
+#endif
+	u32 protocol_quota;
+	s32 current_credit;
+	u32 reserved;
+};
+
+struct rate_shaping_vars {
+	struct rate_shaping_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
+	u32 pause_mask;
+	u32 periodic_stop;
+	u32 rs_periodic_timeout;
+	u32 rs_threshold;
+	u32 last_periodic_time;
+	u32 reserved;
+};
+
+struct fairness_per_protocol {
+	u32 credit_delta;
+	s32 fair_credit;
+#if defined(__BIG_ENDIAN)
+	u16 reserved0;
+	u8 state;
+	u8 weight;
+#elif defined(__LITTLE_ENDIAN)
+	u8 weight;
+	u8 state;
+	u16 reserved0;
+#endif
+	u32 reserved1;
+};
+
+struct fairness_vars {
+	struct fairness_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
+	u32 upper_bound;
+	u32 port_rate;
+	u32 pause_mask;
+	u32 fair_threshold;
+};
+
+struct safc_struct {
+	u32 cur_pause_mask;
+	u32 expire_time;
+#if defined(__BIG_ENDIAN)
+	u16 reserved0;
+	u8 cur_cos_types;
+	u8 safc_timeout_usec;
+#elif defined(__LITTLE_ENDIAN)
+	u8 safc_timeout_usec;
+	u8 cur_cos_types;
+	u16 reserved0;
+#endif
+	u32 reserved1;
+};
+
+struct demo_struct {
+	u8 con_number[NUM_OF_PROTOCOLS];
+#if defined(__BIG_ENDIAN)
+	u8 reserved1;
+	u8 fairness_enable;
+	u8 rate_shaping_enable;
+	u8 cmng_enable;
+#elif defined(__LITTLE_ENDIAN)
+	u8 cmng_enable;
+	u8 rate_shaping_enable;
+	u8 fairness_enable;
+	u8 reserved1;
+#endif
+};
+
+struct cmng_struct {
+	struct rate_shaping_vars rs_vars;
+	struct fairness_vars fair_vars;
+	struct safc_struct safc_vars;
+	struct demo_struct demo_vars;
+};
+
+
+struct cos_to_protocol {
+	u8 mask[MAX_COS_NUMBER];
+};
+
+
+/*
+ * Common statistics collected by the Xstorm (per port)
+ */
+struct xstorm_common_stats {
+	struct regpair total_sent_bytes;
+	u32 total_sent_pkts;
+	u32 unicast_pkts_sent;
+	struct regpair unicast_bytes_sent;
+	struct regpair multicast_bytes_sent;
+	u32 multicast_pkts_sent;
+	u32 broadcast_pkts_sent;
+	struct regpair broadcast_bytes_sent;
+	struct regpair done;
+};
+
+/*
+ * Protocol-common statistics collected by the Tstorm (per client)
+ */
+struct tstorm_per_client_stats {
+	struct regpair total_rcv_bytes;
+	struct regpair rcv_unicast_bytes;
+	struct regpair rcv_broadcast_bytes;
+	struct regpair rcv_multicast_bytes;
+	struct regpair rcv_error_bytes;
+	u32 checksum_discard;
+	u32 packets_too_big_discard;
+	u32 total_rcv_pkts;
+	u32 rcv_unicast_pkts;
+	u32 rcv_broadcast_pkts;
+	u32 rcv_multicast_pkts;
+	u32 no_buff_discard;
+	u32 ttl0_discard;
+	u32 mac_discard;
+	u32 reserved;
+};
+
+/*
+ * Protocol-common statistics collected by the Tstorm (per port)
+ */
+struct tstorm_common_stats {
+	struct tstorm_per_client_stats client_statistics[MAX_T_STAT_COUNTER_ID];
+	u32 mac_filter_discard;
+	u32 xxoverflow_discard;
+	u32 brb_truncate_discard;
+	u32 reserved;
+	struct regpair done;
+};
+
+/*
+ * Eth statistics query sturcture for the eth_stats_quesry ramrod
+ */
+struct eth_stats_query {
+	struct xstorm_common_stats xstorm_common;
+	struct tstorm_common_stats tstorm_common;
+};
+
+
+/*
+ * FW version stored in the Xstorm RAM
+ */
+struct fw_version {
+#if defined(__BIG_ENDIAN)
+	u16 patch;
+	u8 primary;
+	u8 client;
+#elif defined(__LITTLE_ENDIAN)
+	u8 client;
+	u8 primary;
+	u16 patch;
+#endif
+	u32 flags;
+#define FW_VERSION_OPTIMIZED (0x1<<0)
+#define FW_VERSION_OPTIMIZED_SHIFT 0
+#define FW_VERSION_BIG_ENDIEN (0x1<<1)
+#define FW_VERSION_BIG_ENDIEN_SHIFT 1
+#define __FW_VERSION_RESERVED (0x3FFFFFFF<<2)
+#define __FW_VERSION_RESERVED_SHIFT 2
+};
+
+
+/*
+ * FW version stored in first line of pram
+ */
+struct pram_fw_version {
+#if defined(__BIG_ENDIAN)
+	u16 patch;
+	u8 primary;
+	u8 client;
+#elif defined(__LITTLE_ENDIAN)
+	u8 client;
+	u8 primary;
+	u16 patch;
+#endif
+	u8 flags;
+#define PRAM_FW_VERSION_OPTIMIZED (0x1<<0)
+#define PRAM_FW_VERSION_OPTIMIZED_SHIFT 0
+#define PRAM_FW_VERSION_STORM_ID (0x3<<1)
+#define PRAM_FW_VERSION_STORM_ID_SHIFT 1
+#define PRAM_FW_VERSION_BIG_ENDIEN (0x1<<3)
+#define PRAM_FW_VERSION_BIG_ENDIEN_SHIFT 3
+#define __PRAM_FW_VERSION_RESERVED0 (0xF<<4)
+#define __PRAM_FW_VERSION_RESERVED0_SHIFT 4
+};
+
+
+/*
+ * The send queue element
+ */
+struct slow_path_element {
+	struct spe_hdr hdr;
+	u8 protocol_data[8];
+};
+
+
+/*
+ * eth/toe flags that indicate if to query
+ */
+struct stats_indication_flags {
+	u32 collect_eth;
+	u32 collect_toe;
+};
+
+
diff --git a/drivers/net/bnx2x_init.h b/drivers/net/bnx2x_init.h
new file mode 100644
index 000000000000..04f93bff2ef4
--- /dev/null
+++ b/drivers/net/bnx2x_init.h
@@ -0,0 +1,564 @@
+/* bnx2x_init.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * Written by: Eliezer Tamir <eliezert@broadcom.com>
+ */
+
+#ifndef BNX2X_INIT_H
+#define BNX2X_INIT_H
+
+#define COMMON				0x1
+#define PORT0				0x2
+#define PORT1				0x4
+
+#define INIT_EMULATION			0x1
+#define INIT_FPGA			0x2
+#define INIT_ASIC			0x4
+#define INIT_HARDWARE			0x7
+
+#define STORM_INTMEM_SIZE		(0x5800 / 4)
+#define TSTORM_INTMEM_ADDR		0x1a0000
+#define CSTORM_INTMEM_ADDR		0x220000
+#define XSTORM_INTMEM_ADDR		0x2a0000
+#define USTORM_INTMEM_ADDR		0x320000
+
+
+/* Init operation types and structures */
+
+#define OP_RD			0x1 /* read single register */
+#define OP_WR			0x2 /* write single register */
+#define OP_IW			0x3 /* write single register using mailbox */
+#define OP_SW			0x4 /* copy a string to the device */
+#define OP_SI			0x5 /* copy a string using mailbox */
+#define OP_ZR			0x6 /* clear memory */
+#define OP_ZP			0x7 /* unzip then copy with DMAE */
+#define OP_WB			0x8 /* copy a string using DMAE */
+
+struct raw_op {
+	u32 op		:8;
+	u32 offset	:24;
+	u32 raw_data;
+};
+
+struct op_read {
+	u32 op		:8;
+	u32 offset	:24;
+	u32 pad;
+};
+
+struct op_write {
+	u32 op		:8;
+	u32 offset	:24;
+	u32 val;
+};
+
+struct op_string_write {
+	u32 op		:8;
+	u32 offset	:24;
+#ifdef __LITTLE_ENDIAN
+	u16 data_off;
+	u16 data_len;
+#else /* __BIG_ENDIAN */
+	u16 data_len;
+	u16 data_off;
+#endif
+};
+
+struct op_zero {
+	u32 op		:8;
+	u32 offset	:24;
+	u32 len;
+};
+
+union init_op {
+	struct op_read		read;
+	struct op_write		write;
+	struct op_string_write	str_wr;
+	struct op_zero		zero;
+	struct raw_op		raw;
+};
+
+#include "bnx2x_init_values.h"
+
+static void bnx2x_reg_wr_ind(struct bnx2x *bp, u32 addr, u32 val);
+
+static void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr,
+			     u32 dst_addr, u32 len32);
+
+static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len);
+
+static void bnx2x_init_str_wr(struct bnx2x *bp, u32 addr, const u32 *data,
+			      u32 len)
+{
+	int i;
+
+	for (i = 0; i < len; i++) {
+		REG_WR(bp, addr + i*4, data[i]);
+		if (!(i % 10000)) {
+			touch_softlockup_watchdog();
+			cpu_relax();
+		}
+	}
+}
+
+#define INIT_MEM_WR(reg, data, reg_off, len) \
+	bnx2x_init_str_wr(bp, reg + reg_off*4, data, len)
+
+static void bnx2x_init_ind_wr(struct bnx2x *bp, u32 addr, const u32 *data,
+			      u16 len)
+{
+	int i;
+
+	for (i = 0; i < len; i++) {
+		REG_WR_IND(bp, addr + i*4, data[i]);
+		if (!(i % 10000)) {
+			touch_softlockup_watchdog();
+			cpu_relax();
+		}
+	}
+}
+
+static void bnx2x_init_wr_wb(struct bnx2x *bp, u32 addr, const u32 *data,
+			     u32 len, int gunzip)
+{
+	int offset = 0;
+
+	if (gunzip) {
+		int rc;
+#ifdef __BIG_ENDIAN
+		int i, size;
+		u32 *temp;
+
+		temp = kmalloc(len, GFP_KERNEL);
+		size = (len / 4) + ((len % 4) ? 1 : 0);
+		for (i = 0; i < size; i++)
+			temp[i] = swab32(data[i]);
+		data = temp;
+#endif
+		rc = bnx2x_gunzip(bp, (u8 *)data, len);
+		if (rc) {
+			DP(NETIF_MSG_HW, "gunzip failed ! rc %d\n", rc);
+			return;
+		}
+		len = bp->gunzip_outlen;
+#ifdef __BIG_ENDIAN
+		kfree(temp);
+		for (i = 0; i < len; i++)
+			 ((u32 *)bp->gunzip_buf)[i] =
+					swab32(((u32 *)bp->gunzip_buf)[i]);
+#endif
+	} else {
+		if ((len * 4) > FW_BUF_SIZE) {
+			BNX2X_ERR("LARGE DMAE OPERATION ! len 0x%x\n", len*4);
+			return;
+		}
+		memcpy(bp->gunzip_buf, data, len * 4);
+	}
+
+	while (len > DMAE_LEN32_MAX) {
+		bnx2x_write_dmae(bp, bp->gunzip_mapping + offset,
+				 addr + offset, DMAE_LEN32_MAX);
+		offset += DMAE_LEN32_MAX * 4;
+		len -= DMAE_LEN32_MAX;
+	}
+	bnx2x_write_dmae(bp, bp->gunzip_mapping + offset, addr + offset, len);
+}
+
+#define INIT_MEM_WB(reg, data, reg_off, len) \
+	bnx2x_init_wr_wb(bp, reg + reg_off*4, data, len, 0)
+
+#define INIT_GUNZIP_DMAE(reg, data, reg_off, len) \
+	bnx2x_init_wr_wb(bp, reg + reg_off*4, data, len, 1)
+
+static void bnx2x_init_fill(struct bnx2x *bp, u32 addr, int fill, u32 len)
+{
+	int offset = 0;
+
+	if ((len * 4) > FW_BUF_SIZE) {
+		BNX2X_ERR("LARGE DMAE OPERATION ! len 0x%x\n", len * 4);
+		return;
+	}
+	memset(bp->gunzip_buf, fill, len * 4);
+
+	while (len > DMAE_LEN32_MAX) {
+		bnx2x_write_dmae(bp, bp->gunzip_mapping + offset,
+				 addr + offset, DMAE_LEN32_MAX);
+		offset += DMAE_LEN32_MAX * 4;
+		len -= DMAE_LEN32_MAX;
+	}
+	bnx2x_write_dmae(bp, bp->gunzip_mapping + offset, addr + offset, len);
+}
+
+static void bnx2x_init_block(struct bnx2x *bp, u32 op_start, u32 op_end)
+{
+	int i;
+	union init_op *op;
+	u32 op_type, addr, len;
+	const u32 *data;
+
+	for (i = op_start; i < op_end; i++) {
+
+		op = (union init_op *)&(init_ops[i]);
+
+		op_type = op->str_wr.op;
+		addr = op->str_wr.offset;
+		len = op->str_wr.data_len;
+		data = init_data + op->str_wr.data_off;
+
+		switch (op_type) {
+		case OP_RD:
+			REG_RD(bp, addr);
+			break;
+		case OP_WR:
+			REG_WR(bp, addr, op->write.val);
+			break;
+		case OP_SW:
+			bnx2x_init_str_wr(bp, addr, data, len);
+			break;
+		case OP_WB:
+			bnx2x_init_wr_wb(bp, addr, data, len, 0);
+			break;
+		case OP_SI:
+			bnx2x_init_ind_wr(bp, addr, data, len);
+			break;
+		case OP_ZR:
+			bnx2x_init_fill(bp, addr, 0, op->zero.len);
+			break;
+		case OP_ZP:
+			bnx2x_init_wr_wb(bp, addr, data, len, 1);
+			break;
+		default:
+			BNX2X_ERR("BAD init operation!\n");
+		}
+	}
+}
+
+
+/****************************************************************************
+* PXP
+****************************************************************************/
+/*
+ * This code configures the PCI read/write arbiter
+ * which implements a wighted round robin
+ * between the virtual queues in the chip.
+ *
+ * The values were derived for each PCI max payload and max request size.
+ * since max payload and max request size are only known at run time,
+ * this is done as a separate init stage.
+ */
+
+#define NUM_WR_Q			13
+#define NUM_RD_Q			29
+#define MAX_RD_ORD			3
+#define MAX_WR_ORD			2
+
+/* configuration for one arbiter queue */
+struct arb_line {
+	int l;
+	int add;
+	int ubound;
+};
+
+/* derived configuration for each read queue for each max request size */
+static const struct arb_line read_arb_data[NUM_RD_Q][MAX_RD_ORD + 1] = {
+	{{8 , 64 , 25}, {16 , 64 , 25}, {32 , 64 , 25}, {64 , 64 , 41} },
+	{{4 , 8 , 4},   {4 , 8 , 4},    {4 , 8 , 4},    {4 , 8 , 4} },
+	{{4 , 3 , 3},   {4 , 3 , 3},    {4 , 3 , 3},    {4 , 3 , 3} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {16 , 3 , 11},  {16 , 3 , 11} },
+	{{8 , 64 , 25}, {16 , 64 , 25}, {32 , 64 , 25}, {64 , 64 , 41} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {64 , 3 , 41} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {64 , 3 , 41} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {64 , 3 , 41} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {64 , 3 , 41} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 3 , 6},   {16 , 3 , 11},  {32 , 3 , 21},  {32 , 3 , 21} },
+	{{8 , 64 , 25}, {16 , 64 , 41}, {32 , 64 , 81}, {64 , 64 , 120} }
+};
+
+/* derived configuration for each write queue for each max request size */
+static const struct arb_line write_arb_data[NUM_WR_Q][MAX_WR_ORD + 1] = {
+	{{4 , 6 , 3},   {4 , 6 , 3},    {4 , 6 , 3} },
+	{{4 , 2 , 3},   {4 , 2 , 3},    {4 , 2 , 3} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {16 , 2 , 11} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {32 , 2 , 21} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {32 , 2 , 21} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {32 , 2 , 21} },
+	{{8 , 64 , 25}, {16 , 64 , 25}, {32 , 64 , 25} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {16 , 2 , 11} },
+	{{8 , 2 , 6},   {16 , 2 , 11},  {16 , 2 , 11} },
+	{{8 , 9 , 6},   {16 , 9 , 11},  {32 , 9 , 21} },
+	{{8 , 47 , 19}, {16 , 47 , 19}, {32 , 47 , 21} },
+	{{8 , 9 , 6},   {16 , 9 , 11},  {16 , 9 , 11} },
+	{{8 , 64 , 25}, {16 , 64 , 41}, {32 , 64 , 81} }
+};
+
+/* register adresses for read queues */
+static const struct arb_line read_arb_addr[NUM_RD_Q-1] = {
+	{PXP2_REG_RQ_BW_RD_L0, PXP2_REG_RQ_BW_RD_ADD0,
+		PXP2_REG_RQ_BW_RD_UBOUND0},
+	{PXP2_REG_PSWRQ_BW_L1, PXP2_REG_PSWRQ_BW_ADD1,
+		PXP2_REG_PSWRQ_BW_UB1},
+	{PXP2_REG_PSWRQ_BW_L2, PXP2_REG_PSWRQ_BW_ADD2,
+		PXP2_REG_PSWRQ_BW_UB2},
+	{PXP2_REG_PSWRQ_BW_L3, PXP2_REG_PSWRQ_BW_ADD3,
+		PXP2_REG_PSWRQ_BW_UB3},
+	{PXP2_REG_RQ_BW_RD_L4, PXP2_REG_RQ_BW_RD_ADD4,
+		PXP2_REG_RQ_BW_RD_UBOUND4},
+	{PXP2_REG_RQ_BW_RD_L5, PXP2_REG_RQ_BW_RD_ADD5,
+		PXP2_REG_RQ_BW_RD_UBOUND5},
+	{PXP2_REG_PSWRQ_BW_L6, PXP2_REG_PSWRQ_BW_ADD6,
+		PXP2_REG_PSWRQ_BW_UB6},
+	{PXP2_REG_PSWRQ_BW_L7, PXP2_REG_PSWRQ_BW_ADD7,
+		PXP2_REG_PSWRQ_BW_UB7},
+	{PXP2_REG_PSWRQ_BW_L8, PXP2_REG_PSWRQ_BW_ADD8,
+		PXP2_REG_PSWRQ_BW_UB8},
+	{PXP2_REG_PSWRQ_BW_L9, PXP2_REG_PSWRQ_BW_ADD9,
+		PXP2_REG_PSWRQ_BW_UB9},
+	{PXP2_REG_PSWRQ_BW_L10, PXP2_REG_PSWRQ_BW_ADD10,
+		PXP2_REG_PSWRQ_BW_UB10},
+	{PXP2_REG_PSWRQ_BW_L11, PXP2_REG_PSWRQ_BW_ADD11,
+		PXP2_REG_PSWRQ_BW_UB11},
+	{PXP2_REG_RQ_BW_RD_L12, PXP2_REG_RQ_BW_RD_ADD12,
+		PXP2_REG_RQ_BW_RD_UBOUND12},
+	{PXP2_REG_RQ_BW_RD_L13, PXP2_REG_RQ_BW_RD_ADD13,
+		PXP2_REG_RQ_BW_RD_UBOUND13},
+	{PXP2_REG_RQ_BW_RD_L14, PXP2_REG_RQ_BW_RD_ADD14,
+		PXP2_REG_RQ_BW_RD_UBOUND14},
+	{PXP2_REG_RQ_BW_RD_L15, PXP2_REG_RQ_BW_RD_ADD15,
+		PXP2_REG_RQ_BW_RD_UBOUND15},
+	{PXP2_REG_RQ_BW_RD_L16, PXP2_REG_RQ_BW_RD_ADD16,
+		PXP2_REG_RQ_BW_RD_UBOUND16},
+	{PXP2_REG_RQ_BW_RD_L17, PXP2_REG_RQ_BW_RD_ADD17,
+		PXP2_REG_RQ_BW_RD_UBOUND17},
+	{PXP2_REG_RQ_BW_RD_L18, PXP2_REG_RQ_BW_RD_ADD18,
+		PXP2_REG_RQ_BW_RD_UBOUND18},
+	{PXP2_REG_RQ_BW_RD_L19, PXP2_REG_RQ_BW_RD_ADD19,
+		PXP2_REG_RQ_BW_RD_UBOUND19},
+	{PXP2_REG_RQ_BW_RD_L20, PXP2_REG_RQ_BW_RD_ADD20,
+		PXP2_REG_RQ_BW_RD_UBOUND20},
+	{PXP2_REG_RQ_BW_RD_L22, PXP2_REG_RQ_BW_RD_ADD22,
+		PXP2_REG_RQ_BW_RD_UBOUND22},
+	{PXP2_REG_RQ_BW_RD_L23, PXP2_REG_RQ_BW_RD_ADD23,
+		PXP2_REG_RQ_BW_RD_UBOUND23},
+	{PXP2_REG_RQ_BW_RD_L24, PXP2_REG_RQ_BW_RD_ADD24,
+		PXP2_REG_RQ_BW_RD_UBOUND24},
+	{PXP2_REG_RQ_BW_RD_L25, PXP2_REG_RQ_BW_RD_ADD25,
+		PXP2_REG_RQ_BW_RD_UBOUND25},
+	{PXP2_REG_RQ_BW_RD_L26, PXP2_REG_RQ_BW_RD_ADD26,
+		PXP2_REG_RQ_BW_RD_UBOUND26},
+	{PXP2_REG_RQ_BW_RD_L27, PXP2_REG_RQ_BW_RD_ADD27,
+		PXP2_REG_RQ_BW_RD_UBOUND27},
+	{PXP2_REG_PSWRQ_BW_L28, PXP2_REG_PSWRQ_BW_ADD28,
+		PXP2_REG_PSWRQ_BW_UB28}
+};
+
+/* register adresses for wrtie queues */
+static const struct arb_line write_arb_addr[NUM_WR_Q-1] = {
+	{PXP2_REG_PSWRQ_BW_L1, PXP2_REG_PSWRQ_BW_ADD1,
+		PXP2_REG_PSWRQ_BW_UB1},
+	{PXP2_REG_PSWRQ_BW_L2, PXP2_REG_PSWRQ_BW_ADD2,
+		PXP2_REG_PSWRQ_BW_UB2},
+	{PXP2_REG_PSWRQ_BW_L3, PXP2_REG_PSWRQ_BW_ADD3,
+		PXP2_REG_PSWRQ_BW_UB3},
+	{PXP2_REG_PSWRQ_BW_L6, PXP2_REG_PSWRQ_BW_ADD6,
+		PXP2_REG_PSWRQ_BW_UB6},
+	{PXP2_REG_PSWRQ_BW_L7, PXP2_REG_PSWRQ_BW_ADD7,
+		PXP2_REG_PSWRQ_BW_UB7},
+	{PXP2_REG_PSWRQ_BW_L8, PXP2_REG_PSWRQ_BW_ADD8,
+		PXP2_REG_PSWRQ_BW_UB8},
+	{PXP2_REG_PSWRQ_BW_L9, PXP2_REG_PSWRQ_BW_ADD9,
+		PXP2_REG_PSWRQ_BW_UB9},
+	{PXP2_REG_PSWRQ_BW_L10, PXP2_REG_PSWRQ_BW_ADD10,
+		PXP2_REG_PSWRQ_BW_UB10},
+	{PXP2_REG_PSWRQ_BW_L11, PXP2_REG_PSWRQ_BW_ADD11,
+		PXP2_REG_PSWRQ_BW_UB11},
+	{PXP2_REG_PSWRQ_BW_L28, PXP2_REG_PSWRQ_BW_ADD28,
+		PXP2_REG_PSWRQ_BW_UB28},
+	{PXP2_REG_RQ_BW_WR_L29, PXP2_REG_RQ_BW_WR_ADD29,
+		PXP2_REG_RQ_BW_WR_UBOUND29},
+	{PXP2_REG_RQ_BW_WR_L30, PXP2_REG_RQ_BW_WR_ADD30,
+		PXP2_REG_RQ_BW_WR_UBOUND30}
+};
+
+static void bnx2x_init_pxp(struct bnx2x *bp)
+{
+	int r_order, w_order;
+	u32 val, i;
+
+	pci_read_config_word(bp->pdev,
+			     bp->pcie_cap + PCI_EXP_DEVCTL, (u16 *)&val);
+	DP(NETIF_MSG_HW, "read 0x%x from devctl\n", val);
+	w_order = ((val & PCI_EXP_DEVCTL_PAYLOAD) >> 5);
+	r_order = ((val & PCI_EXP_DEVCTL_READRQ) >> 12);
+
+	if (r_order > MAX_RD_ORD) {
+		DP(NETIF_MSG_HW, "read order of %d  order adjusted to %d\n",
+		   r_order, MAX_RD_ORD);
+		r_order = MAX_RD_ORD;
+	}
+	if (w_order > MAX_WR_ORD) {
+		DP(NETIF_MSG_HW, "write order of %d  order adjusted to %d\n",
+		   w_order, MAX_WR_ORD);
+		w_order = MAX_WR_ORD;
+	}
+	DP(NETIF_MSG_HW, "read order %d  write order %d\n", r_order, w_order);
+
+	for (i = 0; i < NUM_RD_Q-1; i++) {
+		REG_WR(bp, read_arb_addr[i].l, read_arb_data[i][r_order].l);
+		REG_WR(bp, read_arb_addr[i].add,
+		       read_arb_data[i][r_order].add);
+		REG_WR(bp, read_arb_addr[i].ubound,
+		       read_arb_data[i][r_order].ubound);
+	}
+
+	for (i = 0; i < NUM_WR_Q-1; i++) {
+		if ((write_arb_addr[i].l == PXP2_REG_RQ_BW_WR_L29) ||
+		    (write_arb_addr[i].l == PXP2_REG_RQ_BW_WR_L30)) {
+
+			REG_WR(bp, write_arb_addr[i].l,
+			       write_arb_data[i][w_order].l);
+
+			REG_WR(bp, write_arb_addr[i].add,
+			       write_arb_data[i][w_order].add);
+
+			REG_WR(bp, write_arb_addr[i].ubound,
+			       write_arb_data[i][w_order].ubound);
+		} else {
+
+			val = REG_RD(bp, write_arb_addr[i].l);
+			REG_WR(bp, write_arb_addr[i].l,
+			       val | (write_arb_data[i][w_order].l << 10));
+
+			val = REG_RD(bp, write_arb_addr[i].add);
+			REG_WR(bp, write_arb_addr[i].add,
+			       val | (write_arb_data[i][w_order].add << 10));
+
+			val = REG_RD(bp, write_arb_addr[i].ubound);
+			REG_WR(bp, write_arb_addr[i].ubound,
+			       val | (write_arb_data[i][w_order].ubound << 7));
+		}
+	}
+
+	val =  write_arb_data[NUM_WR_Q-1][w_order].add;
+	val += write_arb_data[NUM_WR_Q-1][w_order].ubound << 10;
+	val += write_arb_data[NUM_WR_Q-1][w_order].l << 17;
+	REG_WR(bp, PXP2_REG_PSWRQ_BW_RD, val);
+
+	val =  read_arb_data[NUM_RD_Q-1][r_order].add;
+	val += read_arb_data[NUM_RD_Q-1][r_order].ubound << 10;
+	val += read_arb_data[NUM_RD_Q-1][r_order].l << 17;
+	REG_WR(bp, PXP2_REG_PSWRQ_BW_WR, val);
+
+	REG_WR(bp, PXP2_REG_RQ_WR_MBS0, w_order);
+	REG_WR(bp, PXP2_REG_RQ_WR_MBS0 + 8, w_order);
+	REG_WR(bp, PXP2_REG_RQ_RD_MBS0, r_order);
+	REG_WR(bp, PXP2_REG_RQ_RD_MBS0 + 8, r_order);
+
+	REG_WR(bp, PXP2_REG_WR_DMAE_TH, (128 << w_order)/16);
+}
+
+
+/****************************************************************************
+* CDU
+****************************************************************************/
+
+#define CDU_REGION_NUMBER_XCM_AG	2
+#define CDU_REGION_NUMBER_UCM_AG	4
+
+/**
+ * String-to-compress [31:8] = CID (all 24 bits)
+ * String-to-compress [7:4] = Region
+ * String-to-compress [3:0] = Type
+ */
+#define CDU_VALID_DATA(_cid, _region, _type) \
+		(((_cid) << 8) | (((_region) & 0xf) << 4) | (((_type) & 0xf)))
+#define CDU_CRC8(_cid, _region, _type) \
+			calc_crc8(CDU_VALID_DATA(_cid, _region, _type), 0xff)
+#define CDU_RSRVD_VALUE_TYPE_A(_cid, _region, _type) \
+			(0x80 | (CDU_CRC8(_cid, _region, _type) & 0x7f))
+#define CDU_RSRVD_VALUE_TYPE_B(_crc, _type) \
+	(0x80 | ((_type) & 0xf << 3) | (CDU_CRC8(_cid, _region, _type) & 0x7))
+#define CDU_RSRVD_INVALIDATE_CONTEXT_VALUE(_val)	((_val) & ~0x80)
+
+/*****************************************************************************
+ * Description:
+ *         Calculates crc 8 on a word value: polynomial 0-1-2-8
+ *         Code was translated from Verilog.
+ ****************************************************************************/
+static u8 calc_crc8(u32 data, u8 crc)
+{
+	u8 D[32];
+	u8 NewCRC[8];
+	u8 C[8];
+	u8 crc_res;
+	u8 i;
+
+	/* split the data into 31 bits */
+	for (i = 0; i < 32; i++) {
+		D[i] = data & 1;
+		data = data >> 1;
+	}
+
+	/* split the crc into 8 bits */
+	for (i = 0; i < 8; i++) {
+		C[i] = crc & 1;
+		crc = crc >> 1;
+	}
+
+	NewCRC[0] = D[31] ^ D[30] ^ D[28] ^ D[23] ^ D[21] ^ D[19] ^ D[18] ^
+		D[16] ^ D[14] ^ D[12] ^ D[8] ^ D[7] ^ D[6] ^ D[0] ^ C[4] ^
+		C[6] ^ C[7];
+	NewCRC[1] = D[30] ^ D[29] ^ D[28] ^ D[24] ^ D[23] ^ D[22] ^ D[21] ^
+		D[20] ^ D[18] ^ D[17] ^ D[16] ^ D[15] ^ D[14] ^ D[13] ^
+		D[12] ^ D[9] ^ D[6] ^ D[1] ^ D[0] ^ C[0] ^ C[4] ^ C[5] ^ C[6];
+	NewCRC[2] = D[29] ^ D[28] ^ D[25] ^ D[24] ^ D[22] ^ D[17] ^ D[15] ^
+		D[13] ^ D[12] ^ D[10] ^ D[8] ^ D[6] ^ D[2] ^ D[1] ^ D[0] ^
+		C[0] ^ C[1] ^ C[4] ^ C[5];
+	NewCRC[3] = D[30] ^ D[29] ^ D[26] ^ D[25] ^ D[23] ^ D[18] ^ D[16] ^
+		D[14] ^ D[13] ^ D[11] ^ D[9] ^ D[7] ^ D[3] ^ D[2] ^ D[1] ^
+		C[1] ^ C[2] ^ C[5] ^ C[6];
+	NewCRC[4] = D[31] ^ D[30] ^ D[27] ^ D[26] ^ D[24] ^ D[19] ^ D[17] ^
+		D[15] ^ D[14] ^ D[12] ^ D[10] ^ D[8] ^ D[4] ^ D[3] ^ D[2] ^
+		C[0] ^ C[2] ^ C[3] ^ C[6] ^ C[7];
+	NewCRC[5] = D[31] ^ D[28] ^ D[27] ^ D[25] ^ D[20] ^ D[18] ^ D[16] ^
+		D[15] ^ D[13] ^ D[11] ^ D[9] ^ D[5] ^ D[4] ^ D[3] ^ C[1] ^
+		C[3] ^ C[4] ^ C[7];
+	NewCRC[6] = D[29] ^ D[28] ^ D[26] ^ D[21] ^ D[19] ^ D[17] ^ D[16] ^
+		D[14] ^ D[12] ^ D[10] ^ D[6] ^ D[5] ^ D[4] ^ C[2] ^ C[4] ^
+		C[5];
+	NewCRC[7] = D[30] ^ D[29] ^ D[27] ^ D[22] ^ D[20] ^ D[18] ^ D[17] ^
+		D[15] ^ D[13] ^ D[11] ^ D[7] ^ D[6] ^ D[5] ^ C[3] ^ C[5] ^
+		C[6];
+
+	crc_res = 0;
+	for (i = 0; i < 8; i++)
+		crc_res |= (NewCRC[i] << i);
+
+	return crc_res;
+}
+
+
+#endif /* BNX2X_INIT_H */
+
diff --git a/drivers/net/bnx2x_init_values.h b/drivers/net/bnx2x_init_values.h
new file mode 100644
index 000000000000..bef0a9b19d68
--- /dev/null
+++ b/drivers/net/bnx2x_init_values.h
@@ -0,0 +1,6368 @@
+#ifndef __BNX2X_INIT_VALUES_H__
+#define __BNX2X_INIT_VALUES_H__
+
+/* This array contains the list of operations needed to initialize the chip.
+ *
+ * For each block in the chip there are three init stages:
+ * common - HW used by both ports,
+ * port1 and port2 - initialization for a specific Ethernet port.
+ * When a port is opened or closed, the management CPU tells the driver
+ * whether to init/disable common HW in addition to the port HW.
+ * This way the first port going up will first initializes the common HW,
+ * and the last port going down also resets the common HW
+ *
+ * For each init stage/block there is a list of actions needed in a format:
+ * {operation, register, data}
+ * where:
+ * OP_WR - write a value to the chip.
+ * OP_RD - read a register (usually a clear on read register).
+ * OP_SW - string write, write a section of consecutive addresses to the chip.
+ * OP_SI - copy a string using indirect writes.
+ * OP_ZR - clear a range of memory.
+ * OP_ZP - unzip and copy using DMAE.
+ * OP_WB - string copy using DMAE.
+ *
+ * The #defines mark the stages.
+ *
+ */
+
+static const struct raw_op init_ops[] = {
+#define PRS_COMMON_START        0
+	{OP_WR, PRS_REG_INC_VALUE, 0xf},
+	{OP_WR, PRS_REG_EVENT_ID_1, 0x45},
+	{OP_WR, PRS_REG_EVENT_ID_2, 0x84},
+	{OP_WR, PRS_REG_EVENT_ID_3, 0x6},
+	{OP_WR, PRS_REG_NO_MATCH_EVENT_ID, 0x4},
+	{OP_WR, PRS_REG_CM_HDR_TYPE_0, 0x0},
+	{OP_WR, PRS_REG_CM_HDR_TYPE_1, 0x12170000},
+	{OP_WR, PRS_REG_CM_HDR_TYPE_2, 0x22170000},
+	{OP_WR, PRS_REG_CM_HDR_TYPE_3, 0x32170000},
+	{OP_ZR, PRS_REG_CM_HDR_TYPE_4, 0x5},
+	{OP_WR, PRS_REG_CM_HDR_LOOPBACK_TYPE_1, 0x12150000},
+	{OP_WR, PRS_REG_CM_HDR_LOOPBACK_TYPE_2, 0x22150000},
+	{OP_WR, PRS_REG_CM_HDR_LOOPBACK_TYPE_3, 0x32150000},
+	{OP_ZR, PRS_REG_CM_HDR_LOOPBACK_TYPE_4, 0x4},
+	{OP_WR, PRS_REG_CM_NO_MATCH_HDR, 0x2100000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_0, 0x100000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_1, 0x10100000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_2, 0x20100000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_3, 0x30100000},
+	{OP_ZR, PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_4, 0x4},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_0, 0x100000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_1, 0x12140000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_2, 0x22140000},
+	{OP_WR, PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_3, 0x32140000},
+	{OP_ZR, PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_4, 0x4},
+	{OP_RD, PRS_REG_NUM_OF_PACKETS, 0x0},
+	{OP_RD, PRS_REG_NUM_OF_CFC_FLUSH_MESSAGES, 0x0},
+	{OP_RD, PRS_REG_NUM_OF_TRANSPARENT_FLUSH_MESSAGES, 0x0},
+	{OP_RD, PRS_REG_NUM_OF_DEAD_CYCLES, 0x0},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_0, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_1, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_2, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_3, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_4, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_5, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_6, 0xff},
+	{OP_WR, PRS_REG_FLUSH_REGIONS_TYPE_7, 0xff},
+	{OP_WR, PRS_REG_PURE_REGIONS, 0x3e},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_0, 0x0},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_1, 0x3f},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_2, 0x3f},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_3, 0x3f},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_4, 0x0},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_5, 0x3f},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_6, 0x3f},
+	{OP_WR, PRS_REG_PACKET_REGIONS_TYPE_7, 0x3f},
+#define PRS_COMMON_END          46
+#define PRS_PORT0_START         46
+	{OP_WR, PRS_REG_CID_PORT_0, 0x0},
+#define PRS_PORT0_END           47
+#define PRS_PORT1_START         47
+	{OP_WR, PRS_REG_CID_PORT_1, 0x800000},
+#define PRS_PORT1_END           48
+#define TSDM_COMMON_START       48
+	{OP_WR, TSDM_REG_CFC_RSP_START_ADDR, 0x411},
+	{OP_WR, TSDM_REG_CMP_COUNTER_START_ADDR, 0x400},
+	{OP_WR, TSDM_REG_Q_COUNTER_START_ADDR, 0x404},
+	{OP_WR, TSDM_REG_PCK_END_MSG_START_ADDR, 0x419},
+	{OP_WR, TSDM_REG_CMP_COUNTER_MAX0, 0xffff},
+	{OP_WR, TSDM_REG_CMP_COUNTER_MAX1, 0xffff},
+	{OP_WR, TSDM_REG_CMP_COUNTER_MAX2, 0xffff},
+	{OP_WR, TSDM_REG_CMP_COUNTER_MAX3, 0xffff},
+	{OP_ZR, TSDM_REG_AGG_INT_EVENT_0, 0x80},
+	{OP_WR, TSDM_REG_ENABLE_IN1, 0x7ffffff},
+	{OP_WR, TSDM_REG_ENABLE_IN2, 0x3f},
+	{OP_WR, TSDM_REG_ENABLE_OUT1, 0x7ffffff},
+	{OP_WR, TSDM_REG_ENABLE_OUT2, 0xf},
+	{OP_RD, TSDM_REG_NUM_OF_Q0_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q1_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q3_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q4_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q5_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q6_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q7_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q8_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q9_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q10_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_Q11_CMD, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_PKT_END_MSG, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_PXP_ASYNC_REQ, 0x0},
+	{OP_RD, TSDM_REG_NUM_OF_ACK_AFTER_PLACE, 0x0},
+	{OP_WR, TSDM_REG_TIMER_TICK, 0x3e8},
+#define TSDM_COMMON_END         76
+#define TCM_COMMON_START        76
+	{OP_WR, TCM_REG_XX_MAX_LL_SZ, 0x20},
+	{OP_WR, TCM_REG_XX_OVFL_EVNT_ID, 0x32},
+	{OP_WR, TCM_REG_TQM_TCM_HDR_P, 0x2150020},
+	{OP_WR, TCM_REG_TQM_TCM_HDR_S, 0x2150020},
+	{OP_WR, TCM_REG_TM_TCM_HDR, 0x30},
+	{OP_WR, TCM_REG_ERR_TCM_HDR, 0x8100000},
+	{OP_WR, TCM_REG_ERR_EVNT_ID, 0x33},
+	{OP_WR, TCM_REG_EXPR_EVNT_ID, 0x30},
+	{OP_WR, TCM_REG_STOP_EVNT_ID, 0x31},
+	{OP_WR, TCM_REG_PRS_WEIGHT, 0x4},
+	{OP_WR, TCM_REG_PBF_WEIGHT, 0x5},
+	{OP_WR, TCM_REG_CP_WEIGHT, 0x0},
+	{OP_WR, TCM_REG_TSDM_WEIGHT, 0x4},
+	{OP_WR, TCM_REG_TCM_TQM_USE_Q, 0x1},
+	{OP_WR, TCM_REG_GR_ARB_TYPE, 0x1},
+	{OP_WR, TCM_REG_GR_LD0_PR, 0x1},
+	{OP_WR, TCM_REG_GR_LD1_PR, 0x2},
+	{OP_WR, TCM_REG_CFC_INIT_CRD, 0x1},
+	{OP_WR, TCM_REG_FIC0_INIT_CRD, 0x40},
+	{OP_WR, TCM_REG_FIC1_INIT_CRD, 0x40},
+	{OP_WR, TCM_REG_TQM_INIT_CRD, 0x20},
+	{OP_WR, TCM_REG_XX_INIT_CRD, 0x13},
+	{OP_WR, TCM_REG_XX_MSG_NUM, 0x20},
+	{OP_ZR, TCM_REG_XX_TABLE, 0xa},
+	{OP_SW, TCM_REG_XX_DESCR_TABLE, 0x200000},
+	{OP_WR, TCM_REG_N_SM_CTX_LD_0, 0x7},
+	{OP_WR, TCM_REG_N_SM_CTX_LD_1, 0x7},
+	{OP_WR, TCM_REG_N_SM_CTX_LD_2, 0x8},
+	{OP_WR, TCM_REG_N_SM_CTX_LD_3, 0x8},
+	{OP_ZR, TCM_REG_N_SM_CTX_LD_4, 0x4},
+	{OP_WR, TCM_REG_TCM_REG0_SZ, 0x6},
+	{OP_WR, TCM_REG_PHYS_QNUM0_0, 0xd},
+	{OP_WR, TCM_REG_PHYS_QNUM0_1, 0x2d},
+	{OP_ZR, TCM_REG_PHYS_QNUM1_0, 0x6},
+	{OP_WR, TCM_REG_TCM_STORM0_IFEN, 0x1},
+	{OP_WR, TCM_REG_TCM_STORM1_IFEN, 0x1},
+	{OP_WR, TCM_REG_TCM_TQM_IFEN, 0x1},
+	{OP_WR, TCM_REG_STORM_TCM_IFEN, 0x1},
+	{OP_WR, TCM_REG_TQM_TCM_IFEN, 0x1},
+	{OP_WR, TCM_REG_TSDM_IFEN, 0x1},
+	{OP_WR, TCM_REG_TM_TCM_IFEN, 0x1},
+	{OP_WR, TCM_REG_PRS_IFEN, 0x1},
+	{OP_WR, TCM_REG_PBF_IFEN, 0x1},
+	{OP_WR, TCM_REG_USEM_IFEN, 0x1},
+	{OP_WR, TCM_REG_CSEM_IFEN, 0x1},
+	{OP_WR, TCM_REG_CDU_AG_WR_IFEN, 0x1},
+	{OP_WR, TCM_REG_CDU_AG_RD_IFEN, 0x1},
+	{OP_WR, TCM_REG_CDU_SM_WR_IFEN, 0x1},
+	{OP_WR, TCM_REG_CDU_SM_RD_IFEN, 0x1},
+	{OP_WR, TCM_REG_TCM_CFC_IFEN, 0x1},
+#define TCM_COMMON_END          126
+#define BRB1_COMMON_START       126
+	{OP_SW, BRB1_REG_LL_RAM, 0x2000020},
+	{OP_WR, BRB1_REG_SOFT_RESET, 0x1},
+	{OP_RD, BRB1_REG_NUM_OF_PAUSE_CYCLES_0, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_PAUSE_CYCLES_1, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_PAUSE_CYCLES_2, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_PAUSE_CYCLES_3, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_FULL_CYCLES_0, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_FULL_CYCLES_1, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_FULL_CYCLES_2, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_FULL_CYCLES_3, 0x0},
+	{OP_RD, BRB1_REG_NUM_OF_FULL_CYCLES_4, 0x0},
+	{OP_SW, BRB1_REG_FREE_LIST_PRS_CRDT, 0x30220},
+	{OP_WR, BRB1_REG_SOFT_RESET, 0x0},
+#define BRB1_COMMON_END         139
+#define TSEM_COMMON_START       139
+	{OP_RD, TSEM_REG_MSG_NUM_FIC0, 0x0},
+	{OP_RD, TSEM_REG_MSG_NUM_FIC1, 0x0},
+	{OP_RD, TSEM_REG_MSG_NUM_FOC0, 0x0},
+	{OP_RD, TSEM_REG_MSG_NUM_FOC1, 0x0},
+	{OP_RD, TSEM_REG_MSG_NUM_FOC2, 0x0},
+	{OP_RD, TSEM_REG_MSG_NUM_FOC3, 0x0},
+	{OP_WR, TSEM_REG_ARB_ELEMENT0, 0x1},
+	{OP_WR, TSEM_REG_ARB_ELEMENT1, 0x2},
+	{OP_WR, TSEM_REG_ARB_ELEMENT2, 0x3},
+	{OP_WR, TSEM_REG_ARB_ELEMENT3, 0x0},
+	{OP_WR, TSEM_REG_ARB_ELEMENT4, 0x4},
+	{OP_WR, TSEM_REG_ARB_CYCLE_SIZE, 0x1},
+	{OP_WR, TSEM_REG_TS_0_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_1_AS, 0x1},
+	{OP_WR, TSEM_REG_TS_2_AS, 0x4},
+	{OP_WR, TSEM_REG_TS_3_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_4_AS, 0x1},
+	{OP_WR, TSEM_REG_TS_5_AS, 0x3},
+	{OP_WR, TSEM_REG_TS_6_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_7_AS, 0x1},
+	{OP_WR, TSEM_REG_TS_8_AS, 0x4},
+	{OP_WR, TSEM_REG_TS_9_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_10_AS, 0x1},
+	{OP_WR, TSEM_REG_TS_11_AS, 0x3},
+	{OP_WR, TSEM_REG_TS_12_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_13_AS, 0x1},
+	{OP_WR, TSEM_REG_TS_14_AS, 0x4},
+	{OP_WR, TSEM_REG_TS_15_AS, 0x0},
+	{OP_WR, TSEM_REG_TS_16_AS, 0x4},
+	{OP_WR, TSEM_REG_TS_17_AS, 0x3},
+	{OP_ZR, TSEM_REG_TS_18_AS, 0x2},
+	{OP_WR, TSEM_REG_ENABLE_IN, 0x3fff},
+	{OP_WR, TSEM_REG_ENABLE_OUT, 0x3ff},
+	{OP_WR, TSEM_REG_FIC0_DISABLE, 0x0},
+	{OP_WR, TSEM_REG_FIC1_DISABLE, 0x0},
+	{OP_WR, TSEM_REG_PAS_DISABLE, 0x0},
+	{OP_WR, TSEM_REG_THREADS_LIST, 0xff},
+	{OP_ZR, TSEM_REG_PASSIVE_BUFFER, 0x400},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x18bc0, 0x1},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x18000, 0x34},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x18040, 0x18},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x18080, 0xc},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x180c0, 0x20},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x18300, 0x7a120},
+	{OP_WR, TSEM_REG_FAST_MEMORY + 0x183c0, 0x1f4},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x2000, 0x1b3},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x2000 + 0x6cc, 0x10223},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1020, 0xc8},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1000, 0x2},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x800, 0x2},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x808, 0x2},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x810, 0x4},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1fa0, 0x4},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x4cf0, 0x80224},
+	{OP_ZP, TSEM_REG_INT_TABLE, 0x8c022c},
+	{OP_ZP, TSEM_REG_PRAM, 0x3395024f},
+	{OP_ZP, TSEM_REG_PRAM + 0x8000, 0x2c760f35},
+	{OP_ZP, TSEM_REG_PRAM + 0x10000, 0x5e1a53},
+	{OP_ZP, TSEM_REG_PRAM + 0x18000, 0x5e1a6b},
+	{OP_ZP, TSEM_REG_PRAM + 0x20000, 0x5e1a83},
+	{OP_ZP, TSEM_REG_PRAM + 0x28000, 0x5e1a9b},
+	{OP_ZP, TSEM_REG_PRAM + 0x30000, 0x5e1ab3},
+	{OP_ZP, TSEM_REG_PRAM + 0x38000, 0x5e1acb},
+#define TSEM_COMMON_END         202
+#define TSEM_PORT0_START        202
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x4000, 0x16c},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x4000 + 0x5b0, 0x21ae3},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1370, 0xa},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x13c0, 0x6},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1418, 0xc},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1478, 0x12},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1508, 0x90},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x800, 0x2},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x820, 0x10},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x820 + 0x40, 0x21ae5},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x2908, 0xa},
+#define TSEM_PORT0_END          213
+#define TSEM_PORT1_START        213
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x45b8, 0x16c},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x45b8 + 0x5b0, 0x21ae7},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1398, 0xa},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x13d8, 0x6},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1448, 0xc},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x14c0, 0x12},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x1748, 0x90},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x808, 0x2},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x868, 0x10},
+	{OP_SW, TSEM_REG_FAST_MEMORY + 0x868 + 0x40, 0x21ae9},
+	{OP_ZR, TSEM_REG_FAST_MEMORY + 0x2930, 0xa},
+#define TSEM_PORT1_END          224
+#define MISC_COMMON_START       224
+	{OP_WR, MISC_REG_GRC_TIMEOUT_EN, 0x1},
+	{OP_WR, MISC_REG_PLL_STORM_CTRL_1, 0x71d2911},
+	{OP_WR, MISC_REG_PLL_STORM_CTRL_2, 0x0},
+	{OP_WR, MISC_REG_PLL_STORM_CTRL_3, 0x9c0424},
+	{OP_WR, MISC_REG_PLL_STORM_CTRL_4, 0x0},
+	{OP_WR, MISC_REG_LCPLL_CTRL_1, 0x209},
+#define MISC_COMMON_END         230
+#define NIG_COMMON_START        230
+	{OP_WR, NIG_REG_PBF_LB_IN_EN, 0x1},
+	{OP_WR, NIG_REG_PRS_REQ_IN_EN, 0x1},
+	{OP_WR, NIG_REG_EGRESS_DEBUG_IN_EN, 0x1},
+	{OP_WR, NIG_REG_BRB_LB_OUT_EN, 0x1},
+	{OP_WR, NIG_REG_PRS_EOP_OUT_EN, 0x1},
+#define NIG_COMMON_END          235
+#define NIG_PORT0_START         235
+	{OP_WR, NIG_REG_LLH0_CM_HEADER, 0x300000},
+	{OP_WR, NIG_REG_LLH0_EVENT_ID, 0x26},
+	{OP_WR, NIG_REG_LLH0_ERROR_MASK, 0x0},
+	{OP_WR, NIG_REG_LLH0_XCM_MASK, 0x4},
+	{OP_WR, NIG_REG_LLH0_BRB1_NOT_MCP, 0x1},
+	{OP_WR, NIG_REG_STATUS_INTERRUPT_PORT0, 0x0},
+	{OP_WR, NIG_REG_LLH0_XCM_INIT_CREDIT, 0x30},
+	{OP_WR, NIG_REG_BRB0_PAUSE_IN_EN, 0x1},
+	{OP_WR, NIG_REG_EGRESS_PBF0_IN_EN, 0x1},
+	{OP_WR, NIG_REG_BRB0_OUT_EN, 0x1},
+	{OP_WR, NIG_REG_XCM0_OUT_EN, 0x1},
+#define NIG_PORT0_END           246
+#define NIG_PORT1_START         246
+	{OP_WR, NIG_REG_LLH1_CM_HEADER, 0x300000},
+	{OP_WR, NIG_REG_LLH1_EVENT_ID, 0x26},
+	{OP_WR, NIG_REG_LLH1_ERROR_MASK, 0x0},
+	{OP_WR, NIG_REG_LLH1_XCM_MASK, 0x4},
+	{OP_WR, NIG_REG_LLH1_BRB1_NOT_MCP, 0x1},
+	{OP_WR, NIG_REG_STATUS_INTERRUPT_PORT1, 0x0},
+	{OP_WR, NIG_REG_LLH1_XCM_INIT_CREDIT, 0x30},
+	{OP_WR, NIG_REG_BRB1_PAUSE_IN_EN, 0x1},
+	{OP_WR, NIG_REG_EGRESS_PBF1_IN_EN, 0x1},
+	{OP_WR, NIG_REG_BRB1_OUT_EN, 0x1},
+	{OP_WR, NIG_REG_XCM1_OUT_EN, 0x1},
+#define NIG_PORT1_END           257
+#define UPB_COMMON_START        257
+	{OP_WR, GRCBASE_UPB + PB_REG_CONTROL, 0x20},
+#define UPB_COMMON_END          258
+#define CSDM_COMMON_START       258
+	{OP_WR, CSDM_REG_CFC_RSP_START_ADDR, 0xa11},
+	{OP_WR, CSDM_REG_CMP_COUNTER_START_ADDR, 0xa00},
+	{OP_WR, CSDM_REG_Q_COUNTER_START_ADDR, 0xa04},
+	{OP_WR, CSDM_REG_CMP_COUNTER_MAX0, 0xffff},
+	{OP_WR, CSDM_REG_CMP_COUNTER_MAX1, 0xffff},
+	{OP_WR, CSDM_REG_CMP_COUNTER_MAX2, 0xffff},
+	{OP_WR, CSDM_REG_CMP_COUNTER_MAX3, 0xffff},
+	{OP_ZR, CSDM_REG_AGG_INT_EVENT_0, 0x80},
+	{OP_WR, CSDM_REG_ENABLE_IN1, 0x7ffffff},
+	{OP_WR, CSDM_REG_ENABLE_IN2, 0x3f},
+	{OP_WR, CSDM_REG_ENABLE_OUT1, 0x7ffffff},
+	{OP_WR, CSDM_REG_ENABLE_OUT2, 0xf},
+	{OP_RD, CSDM_REG_NUM_OF_Q0_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q1_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q3_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q4_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q5_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q6_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q7_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q8_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q9_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q10_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_Q11_CMD, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_PKT_END_MSG, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_PXP_ASYNC_REQ, 0x0},
+	{OP_RD, CSDM_REG_NUM_OF_ACK_AFTER_PLACE, 0x0},
+	{OP_WR, CSDM_REG_TIMER_TICK, 0x3e8},
+#define CSDM_COMMON_END         285
+#define USDM_COMMON_START       285
+	{OP_WR, USDM_REG_CFC_RSP_START_ADDR, 0xa11},
+	{OP_WR, USDM_REG_CMP_COUNTER_START_ADDR, 0xa00},
+	{OP_WR, USDM_REG_Q_COUNTER_START_ADDR, 0xa04},
+	{OP_WR, USDM_REG_PCK_END_MSG_START_ADDR, 0xa21},
+	{OP_WR, USDM_REG_CMP_COUNTER_MAX0, 0xffff},
+	{OP_WR, USDM_REG_CMP_COUNTER_MAX1, 0xffff},
+	{OP_WR, USDM_REG_CMP_COUNTER_MAX2, 0xffff},
+	{OP_WR, USDM_REG_CMP_COUNTER_MAX3, 0xffff},
+	{OP_WR, USDM_REG_AGG_INT_EVENT_0, 0x46},
+	{OP_ZR, USDM_REG_AGG_INT_EVENT_1, 0x5f},
+	{OP_WR, USDM_REG_AGG_INT_MODE_0, 0x1},
+	{OP_ZR, USDM_REG_AGG_INT_MODE_1, 0x1f},
+	{OP_WR, USDM_REG_ENABLE_IN1, 0x7ffffff},
+	{OP_WR, USDM_REG_ENABLE_IN2, 0x3f},
+	{OP_WR, USDM_REG_ENABLE_OUT1, 0x7ffffff},
+	{OP_WR, USDM_REG_ENABLE_OUT2, 0xf},
+	{OP_RD, USDM_REG_NUM_OF_Q0_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q1_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q2_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q3_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q4_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q5_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q6_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q7_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q8_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q9_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q10_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_Q11_CMD, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_PKT_END_MSG, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_PXP_ASYNC_REQ, 0x0},
+	{OP_RD, USDM_REG_NUM_OF_ACK_AFTER_PLACE, 0x0},
+	{OP_WR, USDM_REG_TIMER_TICK, 0x3e8},
+#define USDM_COMMON_END         317
+#define CCM_COMMON_START        317
+	{OP_WR, CCM_REG_XX_OVFL_EVNT_ID, 0x32},
+	{OP_WR, CCM_REG_CQM_CCM_HDR_P, 0x2150020},
+	{OP_WR, CCM_REG_CQM_CCM_HDR_S, 0x2150020},
+	{OP_WR, CCM_REG_ERR_CCM_HDR, 0x8100000},
+	{OP_WR, CCM_REG_ERR_EVNT_ID, 0x33},
+	{OP_WR, CCM_REG_TSEM_WEIGHT, 0x0},
+	{OP_WR, CCM_REG_XSEM_WEIGHT, 0x4},
+	{OP_WR, CCM_REG_USEM_WEIGHT, 0x4},
+	{OP_ZR, CCM_REG_PBF_WEIGHT, 0x2},
+	{OP_WR, CCM_REG_CQM_P_WEIGHT, 0x2},
+	{OP_WR, CCM_REG_CCM_CQM_USE_Q, 0x1},
+	{OP_WR, CCM_REG_CNT_AUX1_Q, 0x2},
+	{OP_WR, CCM_REG_CNT_AUX2_Q, 0x2},
+	{OP_WR, CCM_REG_INV_DONE_Q, 0x1},
+	{OP_WR, CCM_REG_GR_ARB_TYPE, 0x1},
+	{OP_WR, CCM_REG_GR_LD0_PR, 0x1},
+	{OP_WR, CCM_REG_GR_LD1_PR, 0x2},
+	{OP_WR, CCM_REG_CFC_INIT_CRD, 0x1},
+	{OP_WR, CCM_REG_CQM_INIT_CRD, 0x20},
+	{OP_WR, CCM_REG_FIC0_INIT_CRD, 0x40},
+	{OP_WR, CCM_REG_FIC1_INIT_CRD, 0x40},
+	{OP_WR, CCM_REG_XX_INIT_CRD, 0x3},
+	{OP_WR, CCM_REG_XX_MSG_NUM, 0x18},
+	{OP_ZR, CCM_REG_XX_TABLE, 0x12},
+	{OP_SW, CCM_REG_XX_DESCR_TABLE, 0x241aeb},
+	{OP_WR, CCM_REG_N_SM_CTX_LD_0, 0x1},
+	{OP_WR, CCM_REG_N_SM_CTX_LD_1, 0x2},
+	{OP_WR, CCM_REG_N_SM_CTX_LD_2, 0x8},
+	{OP_WR, CCM_REG_N_SM_CTX_LD_3, 0x8},
+	{OP_ZR, CCM_REG_N_SM_CTX_LD_4, 0x4},
+	{OP_WR, CCM_REG_CCM_REG0_SZ, 0x4},
+	{OP_WR, CCM_REG_QOS_PHYS_QNUM0_0, 0x9},
+	{OP_WR, CCM_REG_QOS_PHYS_QNUM0_1, 0x29},
+	{OP_WR, CCM_REG_QOS_PHYS_QNUM1_0, 0xa},
+	{OP_WR, CCM_REG_QOS_PHYS_QNUM1_1, 0x2a},
+	{OP_ZR, CCM_REG_QOS_PHYS_QNUM2_0, 0x4},
+	{OP_WR, CCM_REG_PHYS_QNUM1_0, 0xc},
+	{OP_WR, CCM_REG_PHYS_QNUM1_1, 0x2c},
+	{OP_WR, CCM_REG_PHYS_QNUM2_0, 0xb},
+	{OP_WR, CCM_REG_PHYS_QNUM2_1, 0x2b},
+	{OP_ZR, CCM_REG_PHYS_QNUM3_0, 0x2},
+	{OP_WR, CCM_REG_CCM_STORM0_IFEN, 0x1},
+	{OP_WR, CCM_REG_CCM_STORM1_IFEN, 0x1},
+	{OP_WR, CCM_REG_CCM_CQM_IFEN, 0x1},
+	{OP_WR, CCM_REG_STORM_CCM_IFEN, 0x1},
+	{OP_WR, CCM_REG_CQM_CCM_IFEN, 0x1},
+	{OP_WR, CCM_REG_CSDM_IFEN, 0x1},
+	{OP_WR, CCM_REG_TSEM_IFEN, 0x1},
+	{OP_WR, CCM_REG_XSEM_IFEN, 0x1},
+	{OP_WR, CCM_REG_USEM_IFEN, 0x1},
+	{OP_WR, CCM_REG_PBF_IFEN, 0x1},
+	{OP_WR, CCM_REG_CDU_AG_WR_IFEN, 0x1},
+	{OP_WR, CCM_REG_CDU_AG_RD_IFEN, 0x1},
+	{OP_WR, CCM_REG_CDU_SM_WR_IFEN, 0x1},
+	{OP_WR, CCM_REG_CDU_SM_RD_IFEN, 0x1},
+	{OP_WR, CCM_REG_CCM_CFC_IFEN, 0x1},
+#define CCM_COMMON_END          373
+#define UCM_COMMON_START        373
+	{OP_WR, UCM_REG_XX_OVFL_EVNT_ID, 0x32},
+	{OP_WR, UCM_REG_UQM_UCM_HDR_P, 0x2150020},
+	{OP_WR, UCM_REG_UQM_UCM_HDR_S, 0x2150020},
+	{OP_WR, UCM_REG_TM_UCM_HDR, 0x30},
+	{OP_WR, UCM_REG_ERR_UCM_HDR, 0x8100000},
+	{OP_WR, UCM_REG_ERR_EVNT_ID, 0x33},
+	{OP_WR, UCM_REG_EXPR_EVNT_ID, 0x30},
+	{OP_WR, UCM_REG_STOP_EVNT_ID, 0x31},
+	{OP_WR, UCM_REG_TSEM_WEIGHT, 0x3},
+	{OP_WR, UCM_REG_CSEM_WEIGHT, 0x0},
+	{OP_WR, UCM_REG_CP_WEIGHT, 0x0},
+	{OP_WR, UCM_REG_UQM_P_WEIGHT, 0x6},
+	{OP_WR, UCM_REG_UCM_UQM_USE_Q, 0x1},
+	{OP_WR, UCM_REG_INV_CFLG_Q, 0x1},
+	{OP_WR, UCM_REG_GR_ARB_TYPE, 0x1},
+	{OP_WR, UCM_REG_GR_LD0_PR, 0x1},
+	{OP_WR, UCM_REG_GR_LD1_PR, 0x2},
+	{OP_WR, UCM_REG_CFC_INIT_CRD, 0x1},
+	{OP_WR, UCM_REG_FIC0_INIT_CRD, 0x40},
+	{OP_WR, UCM_REG_FIC1_INIT_CRD, 0x40},
+	{OP_WR, UCM_REG_TM_INIT_CRD, 0x4},
+	{OP_WR, UCM_REG_UQM_INIT_CRD, 0x20},
+	{OP_WR, UCM_REG_XX_INIT_CRD, 0xc},
+	{OP_WR, UCM_REG_XX_MSG_NUM, 0x20},
+	{OP_ZR, UCM_REG_XX_TABLE, 0x12},
+	{OP_SW, UCM_REG_XX_DESCR_TABLE, 0x201b0f},
+	{OP_WR, UCM_REG_N_SM_CTX_LD_0, 0xa},
+	{OP_WR, UCM_REG_N_SM_CTX_LD_1, 0x7},
+	{OP_WR, UCM_REG_N_SM_CTX_LD_2, 0xf},
+	{OP_WR, UCM_REG_N_SM_CTX_LD_3, 0x10},
+	{OP_ZR, UCM_REG_N_SM_CTX_LD_4, 0x4},
+	{OP_WR, UCM_REG_UCM_REG0_SZ, 0x3},
+	{OP_WR, UCM_REG_PHYS_QNUM0_0, 0xf},
+	{OP_WR, UCM_REG_PHYS_QNUM0_1, 0x2f},
+	{OP_WR, UCM_REG_PHYS_QNUM1_0, 0xe},
+	{OP_WR, UCM_REG_PHYS_QNUM1_1, 0x2e},
+	{OP_WR, UCM_REG_UCM_STORM0_IFEN, 0x1},
+	{OP_WR, UCM_REG_UCM_STORM1_IFEN, 0x1},
+	{OP_WR, UCM_REG_UCM_UQM_IFEN, 0x1},
+	{OP_WR, UCM_REG_STORM_UCM_IFEN, 0x1},
+	{OP_WR, UCM_REG_UQM_UCM_IFEN, 0x1},
+	{OP_WR, UCM_REG_USDM_IFEN, 0x1},
+	{OP_WR, UCM_REG_TM_UCM_IFEN, 0x1},
+	{OP_WR, UCM_REG_UCM_TM_IFEN, 0x1},
+	{OP_WR, UCM_REG_TSEM_IFEN, 0x1},
+	{OP_WR, UCM_REG_CSEM_IFEN, 0x1},
+	{OP_WR, UCM_REG_XSEM_IFEN, 0x1},
+	{OP_WR, UCM_REG_DORQ_IFEN, 0x1},
+	{OP_WR, UCM_REG_CDU_AG_WR_IFEN, 0x1},
+	{OP_WR, UCM_REG_CDU_AG_RD_IFEN, 0x1},
+	{OP_WR, UCM_REG_CDU_SM_WR_IFEN, 0x1},
+	{OP_WR, UCM_REG_CDU_SM_RD_IFEN, 0x1},
+	{OP_WR, UCM_REG_UCM_CFC_IFEN, 0x1},
+#define UCM_COMMON_END          426
+#define USEM_COMMON_START       426
+	{OP_RD, USEM_REG_MSG_NUM_FIC0, 0x0},
+	{OP_RD, USEM_REG_MSG_NUM_FIC1, 0x0},
+	{OP_RD, USEM_REG_MSG_NUM_FOC0, 0x0},
+	{OP_RD, USEM_REG_MSG_NUM_FOC1, 0x0},
+	{OP_RD, USEM_REG_MSG_NUM_FOC2, 0x0},
+	{OP_RD, USEM_REG_MSG_NUM_FOC3, 0x0},
+	{OP_WR, USEM_REG_ARB_ELEMENT0, 0x1},
+	{OP_WR, USEM_REG_ARB_ELEMENT1, 0x2},
+	{OP_WR, USEM_REG_ARB_ELEMENT2, 0x3},
+	{OP_WR, USEM_REG_ARB_ELEMENT3, 0x0},
+	{OP_WR, USEM_REG_ARB_ELEMENT4, 0x4},
+	{OP_WR, USEM_REG_ARB_CYCLE_SIZE, 0x1},
+	{OP_WR, USEM_REG_TS_0_AS, 0x0},
+	{OP_WR, USEM_REG_TS_1_AS, 0x1},
+	{OP_WR, USEM_REG_TS_2_AS, 0x4},
+	{OP_WR, USEM_REG_TS_3_AS, 0x0},
+	{OP_WR, USEM_REG_TS_4_AS, 0x1},
+	{OP_WR, USEM_REG_TS_5_AS, 0x3},
+	{OP_WR, USEM_REG_TS_6_AS, 0x0},
+	{OP_WR, USEM_REG_TS_7_AS, 0x1},
+	{OP_WR, USEM_REG_TS_8_AS, 0x4},
+	{OP_WR, USEM_REG_TS_9_AS, 0x0},
+	{OP_WR, USEM_REG_TS_10_AS, 0x1},
+	{OP_WR, USEM_REG_TS_11_AS, 0x3},
+	{OP_WR, USEM_REG_TS_12_AS, 0x0},
+	{OP_WR, USEM_REG_TS_13_AS, 0x1},
+	{OP_WR, USEM_REG_TS_14_AS, 0x4},
+	{OP_WR, USEM_REG_TS_15_AS, 0x0},
+	{OP_WR, USEM_REG_TS_16_AS, 0x4},
+	{OP_WR, USEM_REG_TS_17_AS, 0x3},
+	{OP_ZR, USEM_REG_TS_18_AS, 0x2},
+	{OP_WR, USEM_REG_ENABLE_IN, 0x3fff},
+	{OP_WR, USEM_REG_ENABLE_OUT, 0x3ff},
+	{OP_WR, USEM_REG_FIC0_DISABLE, 0x0},
+	{OP_WR, USEM_REG_FIC1_DISABLE, 0x0},
+	{OP_WR, USEM_REG_PAS_DISABLE, 0x0},
+	{OP_WR, USEM_REG_THREADS_LIST, 0xffff},
+	{OP_ZR, USEM_REG_PASSIVE_BUFFER, 0x800},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18bc0, 0x1},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18000, 0x1a},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18040, 0x4e},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18080, 0x10},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x180c0, 0x20},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18300, 0x7a120},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x183c0, 0x1f4},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x18380, 0x1dcd6500},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x5000, 0x102},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1020, 0xc8},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1000, 0x2},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1e20, 0x40},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3000, 0x400},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x2400, 0x2},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x2408, 0x2},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x2410, 0x6},
+	{OP_SW, USEM_REG_FAST_MEMORY + 0x2410 + 0x18, 0x21b2f},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4b68, 0x2},
+	{OP_SW, USEM_REG_FAST_MEMORY + 0x4b68 + 0x8, 0x21b31},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4b10, 0x2},
+	{OP_SW, USEM_REG_FAST_MEMORY + 0x2c30, 0x21b33},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x10800, 0x1000000},
+	{OP_SW, USEM_REG_FAST_MEMORY + 0x10c00, 0x101b35},
+	{OP_WR, USEM_REG_FAST_MEMORY + 0x10800, 0x0},
+	{OP_SW, USEM_REG_FAST_MEMORY + 0x10c40, 0x101b45},
+	{OP_ZP, USEM_REG_INT_TABLE, 0xb41b55},
+	{OP_ZP, USEM_REG_PRAM, 0x32d01b82},
+	{OP_ZP, USEM_REG_PRAM + 0x8000, 0x32172836},
+	{OP_ZP, USEM_REG_PRAM + 0x10000, 0x1a7a34bc},
+	{OP_ZP, USEM_REG_PRAM + 0x18000, 0x5f3b5b},
+	{OP_ZP, USEM_REG_PRAM + 0x20000, 0x5f3b73},
+	{OP_ZP, USEM_REG_PRAM + 0x28000, 0x5f3b8b},
+	{OP_ZP, USEM_REG_PRAM + 0x30000, 0x5f3ba3},
+	{OP_ZP, USEM_REG_PRAM + 0x38000, 0x5f3bbb},
+#define USEM_COMMON_END         498
+#define USEM_PORT0_START        498
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1400, 0xa0},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1900, 0xa},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1950, 0x2e},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1d00, 0x24},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3000, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3100, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3200, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3300, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3400, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3500, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3600, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3700, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3800, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3900, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3a00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3b00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3c00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3d00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3e00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3f00, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x2400, 0x2},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4b78, 0x52},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4e08, 0xc},
+#define USEM_PORT0_END          521
+#define USEM_PORT1_START        521
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1680, 0xa0},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1928, 0xa},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1a08, 0x2e},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x1d90, 0x24},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3080, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3180, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3280, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3380, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3480, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3580, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3680, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3780, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3880, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3980, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3a80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3b80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3c80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3d80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3e80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x3f80, 0x20},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x2408, 0x2},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4cc0, 0x52},
+	{OP_ZR, USEM_REG_FAST_MEMORY + 0x4e38, 0xc},
+#define USEM_PORT1_END          544
+#define CSEM_COMMON_START       544
+	{OP_RD, CSEM_REG_MSG_NUM_FIC0, 0x0},
+	{OP_RD, CSEM_REG_MSG_NUM_FIC1, 0x0},
+	{OP_RD, CSEM_REG_MSG_NUM_FOC0, 0x0},
+	{OP_RD, CSEM_REG_MSG_NUM_FOC1, 0x0},
+	{OP_RD, CSEM_REG_MSG_NUM_FOC2, 0x0},
+	{OP_RD, CSEM_REG_MSG_NUM_FOC3, 0x0},
+	{OP_WR, CSEM_REG_ARB_ELEMENT0, 0x1},
+	{OP_WR, CSEM_REG_ARB_ELEMENT1, 0x2},
+	{OP_WR, CSEM_REG_ARB_ELEMENT2, 0x3},
+	{OP_WR, CSEM_REG_ARB_ELEMENT3, 0x0},
+	{OP_WR, CSEM_REG_ARB_ELEMENT4, 0x4},
+	{OP_WR, CSEM_REG_ARB_CYCLE_SIZE, 0x1},
+	{OP_WR, CSEM_REG_TS_0_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_1_AS, 0x1},
+	{OP_WR, CSEM_REG_TS_2_AS, 0x4},
+	{OP_WR, CSEM_REG_TS_3_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_4_AS, 0x1},
+	{OP_WR, CSEM_REG_TS_5_AS, 0x3},
+	{OP_WR, CSEM_REG_TS_6_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_7_AS, 0x1},
+	{OP_WR, CSEM_REG_TS_8_AS, 0x4},
+	{OP_WR, CSEM_REG_TS_9_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_10_AS, 0x1},
+	{OP_WR, CSEM_REG_TS_11_AS, 0x3},
+	{OP_WR, CSEM_REG_TS_12_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_13_AS, 0x1},
+	{OP_WR, CSEM_REG_TS_14_AS, 0x4},
+	{OP_WR, CSEM_REG_TS_15_AS, 0x0},
+	{OP_WR, CSEM_REG_TS_16_AS, 0x4},
+	{OP_WR, CSEM_REG_TS_17_AS, 0x3},
+	{OP_ZR, CSEM_REG_TS_18_AS, 0x2},
+	{OP_WR, CSEM_REG_ENABLE_IN, 0x3fff},
+	{OP_WR, CSEM_REG_ENABLE_OUT, 0x3ff},
+	{OP_WR, CSEM_REG_FIC0_DISABLE, 0x0},
+	{OP_WR, CSEM_REG_FIC1_DISABLE, 0x0},
+	{OP_WR, CSEM_REG_PAS_DISABLE, 0x0},
+	{OP_WR, CSEM_REG_THREADS_LIST, 0xffff},
+	{OP_ZR, CSEM_REG_PASSIVE_BUFFER, 0x800},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x18bc0, 0x1},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x18000, 0x10},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x18040, 0x12},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x18080, 0x30},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x180c0, 0xe},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x183c0, 0x1f4},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x5000, 0x42},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1020, 0xc8},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1000, 0x2},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x2000, 0xc0},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x3070, 0x80},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x4280, 0x4},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x25c0, 0x240},
+	{OP_SW, CSEM_REG_FAST_MEMORY + 0x25c0 + 0x900, 0x83bd3},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x10800, 0x13fffff},
+	{OP_SW, CSEM_REG_FAST_MEMORY + 0x10c00, 0x103bdb},
+	{OP_WR, CSEM_REG_FAST_MEMORY + 0x10800, 0x0},
+	{OP_SW, CSEM_REG_FAST_MEMORY + 0x10c40, 0x103beb},
+	{OP_ZP, CSEM_REG_INT_TABLE, 0x5f3bfb},
+	{OP_ZP, CSEM_REG_PRAM, 0x32423c13},
+	{OP_ZP, CSEM_REG_PRAM + 0x8000, 0xf2148a4},
+	{OP_ZP, CSEM_REG_PRAM + 0x10000, 0x5f4c6d},
+	{OP_ZP, CSEM_REG_PRAM + 0x18000, 0x5f4c85},
+	{OP_ZP, CSEM_REG_PRAM + 0x20000, 0x5f4c9d},
+	{OP_ZP, CSEM_REG_PRAM + 0x28000, 0x5f4cb5},
+	{OP_ZP, CSEM_REG_PRAM + 0x30000, 0x5f4ccd},
+	{OP_ZP, CSEM_REG_PRAM + 0x38000, 0x5f4ce5},
+#define CSEM_COMMON_END         609
+#define CSEM_PORT0_START        609
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1400, 0xa0},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1900, 0x10},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1980, 0x30},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x2300, 0x2},
+	{OP_SW, CSEM_REG_FAST_MEMORY + 0x2300 + 0x8, 0x24cfd},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x3040, 0x6},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x2410, 0x30},
+#define CSEM_PORT0_END          616
+#define CSEM_PORT1_START        616
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1680, 0xa0},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1940, 0x10},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x1a40, 0x30},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x2310, 0x2},
+	{OP_SW, CSEM_REG_FAST_MEMORY + 0x2310 + 0x8, 0x24cff},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x3058, 0x6},
+	{OP_ZR, CSEM_REG_FAST_MEMORY + 0x24d0, 0x30},
+#define CSEM_PORT1_END          623
+#define XPB_COMMON_START        623
+	{OP_WR, GRCBASE_XPB + PB_REG_CONTROL, 0x20},
+#define XPB_COMMON_END          624
+#define DQ_COMMON_START         624
+	{OP_WR, DORQ_REG_MODE_ACT, 0x2},
+	{OP_WR, DORQ_REG_NORM_CID_OFST, 0x3},
+	{OP_WR, DORQ_REG_OUTST_REQ, 0x4},
+	{OP_WR, DORQ_REG_DPM_CID_ADDR, 0x8},
+	{OP_WR, DORQ_REG_RSP_INIT_CRD, 0x2},
+	{OP_WR, DORQ_REG_NORM_CMHEAD_TX, 0x90},
+	{OP_WR, DORQ_REG_CMHEAD_RX, 0x90},
+	{OP_WR, DORQ_REG_SHRT_CMHEAD, 0x800090},
+	{OP_WR, DORQ_REG_ERR_CMHEAD, 0x8140000},
+	{OP_WR, DORQ_REG_AGG_CMD0, 0x8a},
+	{OP_WR, DORQ_REG_AGG_CMD1, 0x80},
+	{OP_WR, DORQ_REG_AGG_CMD2, 0x90},
+	{OP_WR, DORQ_REG_AGG_CMD3, 0x80},
+	{OP_WR, DORQ_REG_SHRT_ACT_CNT, 0x6},
+	{OP_WR, DORQ_REG_DQ_FIFO_FULL_TH, 0x7d0},
+	{OP_WR, DORQ_REG_DQ_FIFO_AFULL_TH, 0x76c},
+	{OP_WR, DORQ_REG_REGN, 0x7c1004},
+	{OP_WR, DORQ_REG_IF_EN, 0xf},
+#define DQ_COMMON_END           642
+#define TIMERS_COMMON_START     642
+	{OP_ZR, TM_REG_CLIN_PRIOR0_CLIENT, 0x2},
+	{OP_WR, TM_REG_LIN_SETCLR_FIFO_ALFULL_THR, 0x1c},
+	{OP_WR, TM_REG_CFC_AC_CRDCNT_VAL, 0x1},
+	{OP_WR, TM_REG_CFC_CLD_CRDCNT_VAL, 0x1},
+	{OP_WR, TM_REG_CLOUT_CRDCNT0_VAL, 0x1},
+	{OP_WR, TM_REG_CLOUT_CRDCNT1_VAL, 0x1},
+	{OP_WR, TM_REG_CLOUT_CRDCNT2_VAL, 0x1},
+	{OP_WR, TM_REG_EXP_CRDCNT_VAL, 0x1},
+	{OP_WR, TM_REG_PCIARB_CRDCNT_VAL, 0x2},
+	{OP_WR, TM_REG_TIMER_TICK_SIZE, 0x3d090},
+	{OP_WR, TM_REG_CL0_CONT_REGION, 0x8},
+	{OP_WR, TM_REG_CL1_CONT_REGION, 0xc},
+	{OP_WR, TM_REG_CL2_CONT_REGION, 0x10},
+	{OP_WR, TM_REG_TM_CONTEXT_REGION, 0x20},
+	{OP_WR, TM_REG_EN_TIMERS, 0x1},
+	{OP_WR, TM_REG_EN_REAL_TIME_CNT, 0x1},
+	{OP_WR, TM_REG_EN_CL0_INPUT, 0x1},
+	{OP_WR, TM_REG_EN_CL1_INPUT, 0x1},
+	{OP_WR, TM_REG_EN_CL2_INPUT, 0x1},
+#define TIMERS_COMMON_END       661
+#define TIMERS_PORT0_START      661
+	{OP_ZR, TM_REG_LIN0_PHY_ADDR, 0x2},
+#define TIMERS_PORT0_END        662
+#define TIMERS_PORT1_START      662
+	{OP_ZR, TM_REG_LIN1_PHY_ADDR, 0x2},
+#define TIMERS_PORT1_END        663
+#define XSDM_COMMON_START       663
+	{OP_WR, XSDM_REG_CFC_RSP_START_ADDR, 0xa14},
+	{OP_WR, XSDM_REG_CMP_COUNTER_START_ADDR, 0xa00},
+	{OP_WR, XSDM_REG_Q_COUNTER_START_ADDR, 0xa04},
+	{OP_WR, XSDM_REG_CMP_COUNTER_MAX0, 0xffff},
+	{OP_WR, XSDM_REG_CMP_COUNTER_MAX1, 0xffff},
+	{OP_WR, XSDM_REG_CMP_COUNTER_MAX2, 0xffff},
+	{OP_WR, XSDM_REG_CMP_COUNTER_MAX3, 0xffff},
+	{OP_WR, XSDM_REG_AGG_INT_EVENT_0, 0x20},
+	{OP_WR, XSDM_REG_AGG_INT_EVENT_1, 0x20},
+	{OP_ZR, XSDM_REG_AGG_INT_EVENT_2, 0x5e},
+	{OP_WR, XSDM_REG_AGG_INT_MODE_0, 0x1},
+	{OP_ZR, XSDM_REG_AGG_INT_MODE_1, 0x1f},
+	{OP_WR, XSDM_REG_ENABLE_IN1, 0x7ffffff},
+	{OP_WR, XSDM_REG_ENABLE_IN2, 0x3f},
+	{OP_WR, XSDM_REG_ENABLE_OUT1, 0x7ffffff},
+	{OP_WR, XSDM_REG_ENABLE_OUT2, 0xf},
+	{OP_RD, XSDM_REG_NUM_OF_Q0_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q1_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q3_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q4_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q5_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q6_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q7_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q8_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q9_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q10_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_Q11_CMD, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_PKT_END_MSG, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_PXP_ASYNC_REQ, 0x0},
+	{OP_RD, XSDM_REG_NUM_OF_ACK_AFTER_PLACE, 0x0},
+	{OP_WR, XSDM_REG_TIMER_TICK, 0x3e8},
+#define XSDM_COMMON_END         694
+#define QM_COMMON_START         694
+	{OP_WR, QM_REG_ACTCTRINITVAL_0, 0x6},
+	{OP_WR, QM_REG_ACTCTRINITVAL_1, 0x5},
+	{OP_WR, QM_REG_ACTCTRINITVAL_2, 0xa},
+	{OP_WR, QM_REG_ACTCTRINITVAL_3, 0x5},
+	{OP_WR, QM_REG_PCIREQAT, 0x2},
+	{OP_WR, QM_REG_CMINITCRD_0, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_1, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_2, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_3, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_4, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_5, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_6, 0x4},
+	{OP_WR, QM_REG_CMINITCRD_7, 0x4},
+	{OP_WR, QM_REG_OUTLDREQ, 0x4},
+	{OP_WR, QM_REG_CTXREG_0, 0x7c},
+	{OP_WR, QM_REG_CTXREG_1, 0x3d},
+	{OP_WR, QM_REG_CTXREG_2, 0x3f},
+	{OP_WR, QM_REG_CTXREG_3, 0x9c},
+	{OP_WR, QM_REG_ENSEC, 0x7},
+	{OP_ZR, QM_REG_QVOQIDX_0, 0x5},
+	{OP_WR, QM_REG_WRRWEIGHTS_0, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_5, 0x0},
+	{OP_WR, QM_REG_QVOQIDX_6, 0x4},
+	{OP_WR, QM_REG_QVOQIDX_7, 0x4},
+	{OP_WR, QM_REG_QVOQIDX_8, 0x2},
+	{OP_WR, QM_REG_WRRWEIGHTS_1, 0x8012004},
+	{OP_WR, QM_REG_QVOQIDX_9, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_10, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_11, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_12, 0x5},
+	{OP_WR, QM_REG_WRRWEIGHTS_2, 0x20081001},
+	{OP_WR, QM_REG_QVOQIDX_13, 0x8},
+	{OP_WR, QM_REG_QVOQIDX_14, 0x6},
+	{OP_WR, QM_REG_QVOQIDX_15, 0x7},
+	{OP_WR, QM_REG_QVOQIDX_16, 0x0},
+	{OP_WR, QM_REG_WRRWEIGHTS_3, 0x1010120},
+	{OP_ZR, QM_REG_QVOQIDX_17, 0x4},
+	{OP_WR, QM_REG_WRRWEIGHTS_4, 0x1010101},
+	{OP_ZR, QM_REG_QVOQIDX_21, 0x4},
+	{OP_WR, QM_REG_WRRWEIGHTS_5, 0x1010101},
+	{OP_ZR, QM_REG_QVOQIDX_25, 0x4},
+	{OP_WR, QM_REG_WRRWEIGHTS_6, 0x1010101},
+	{OP_ZR, QM_REG_QVOQIDX_29, 0x3},
+	{OP_WR, QM_REG_QVOQIDX_32, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_7, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_33, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_34, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_35, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_36, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_8, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_37, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_38, 0x4},
+	{OP_WR, QM_REG_QVOQIDX_39, 0x4},
+	{OP_WR, QM_REG_QVOQIDX_40, 0x2},
+	{OP_WR, QM_REG_WRRWEIGHTS_9, 0x8012004},
+	{OP_WR, QM_REG_QVOQIDX_41, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_42, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_43, 0x5},
+	{OP_WR, QM_REG_QVOQIDX_44, 0x5},
+	{OP_WR, QM_REG_WRRWEIGHTS_10, 0x20081001},
+	{OP_WR, QM_REG_QVOQIDX_45, 0x8},
+	{OP_WR, QM_REG_QVOQIDX_46, 0x6},
+	{OP_WR, QM_REG_QVOQIDX_47, 0x7},
+	{OP_WR, QM_REG_QVOQIDX_48, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_11, 0x1010120},
+	{OP_WR, QM_REG_QVOQIDX_49, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_50, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_51, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_52, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_12, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_53, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_54, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_55, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_56, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_13, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_57, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_58, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_59, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_60, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_14, 0x1010101},
+	{OP_WR, QM_REG_QVOQIDX_61, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_62, 0x1},
+	{OP_WR, QM_REG_QVOQIDX_63, 0x1},
+	{OP_WR, QM_REG_WRRWEIGHTS_15, 0x1010101},
+	{OP_WR, QM_REG_VOQQMASK_0_LSB, 0xffff003f},
+	{OP_ZR, QM_REG_VOQQMASK_0_MSB, 0x2},
+	{OP_WR, QM_REG_VOQQMASK_1_MSB, 0xffff003f},
+	{OP_WR, QM_REG_VOQQMASK_2_LSB, 0x100},
+	{OP_WR, QM_REG_VOQQMASK_2_MSB, 0x100},
+	{OP_ZR, QM_REG_VOQQMASK_3_LSB, 0x2},
+	{OP_WR, QM_REG_VOQQMASK_4_LSB, 0xc0},
+	{OP_WR, QM_REG_VOQQMASK_4_MSB, 0xc0},
+	{OP_WR, QM_REG_VOQQMASK_5_LSB, 0x1e00},
+	{OP_WR, QM_REG_VOQQMASK_5_MSB, 0x1e00},
+	{OP_WR, QM_REG_VOQQMASK_6_LSB, 0x4000},
+	{OP_WR, QM_REG_VOQQMASK_6_MSB, 0x4000},
+	{OP_WR, QM_REG_VOQQMASK_7_LSB, 0x8000},
+	{OP_WR, QM_REG_VOQQMASK_7_MSB, 0x8000},
+	{OP_WR, QM_REG_VOQQMASK_8_LSB, 0x2000},
+	{OP_WR, QM_REG_VOQQMASK_8_MSB, 0x2000},
+	{OP_ZR, QM_REG_VOQQMASK_9_LSB, 0x7},
+	{OP_WR, QM_REG_VOQPORT_1, 0x1},
+	{OP_ZR, QM_REG_VOQPORT_2, 0xa},
+	{OP_WR, QM_REG_CMINTVOQMASK_0, 0xc08},
+	{OP_WR, QM_REG_CMINTVOQMASK_1, 0x40},
+	{OP_WR, QM_REG_CMINTVOQMASK_2, 0x100},
+	{OP_WR, QM_REG_CMINTVOQMASK_3, 0x20},
+	{OP_WR, QM_REG_CMINTVOQMASK_4, 0x17},
+	{OP_WR, QM_REG_CMINTVOQMASK_5, 0x80},
+	{OP_WR, QM_REG_CMINTVOQMASK_6, 0x200},
+	{OP_WR, QM_REG_CMINTVOQMASK_7, 0x0},
+	{OP_WR, QM_REG_HWAEMPTYMASK_LSB, 0xffff01ff},
+	{OP_WR, QM_REG_HWAEMPTYMASK_MSB, 0xffff01ff},
+	{OP_WR, QM_REG_ENBYPVOQMASK, 0x13},
+	{OP_WR, QM_REG_VOQCREDITAFULLTHR, 0x13f},
+	{OP_WR, QM_REG_VOQINITCREDIT_0, 0x140},
+	{OP_WR, QM_REG_VOQINITCREDIT_1, 0x140},
+	{OP_ZR, QM_REG_VOQINITCREDIT_2, 0x2},
+	{OP_WR, QM_REG_VOQINITCREDIT_4, 0xc0},
+	{OP_ZR, QM_REG_VOQINITCREDIT_5, 0x7},
+	{OP_WR, QM_REG_TASKCRDCOST_0, 0x48},
+	{OP_WR, QM_REG_TASKCRDCOST_1, 0x48},
+	{OP_ZR, QM_REG_TASKCRDCOST_2, 0x2},
+	{OP_WR, QM_REG_TASKCRDCOST_4, 0x48},
+	{OP_ZR, QM_REG_TASKCRDCOST_5, 0x7},
+	{OP_WR, QM_REG_BYTECRDINITVAL, 0x8000},
+	{OP_WR, QM_REG_BYTECRDCOST, 0x25e4},
+	{OP_WR, QM_REG_BYTECREDITAFULLTHR, 0x7fff},
+	{OP_WR, QM_REG_ENBYTECRD_LSB, 0x7},
+	{OP_WR, QM_REG_ENBYTECRD_MSB, 0x7},
+	{OP_WR, QM_REG_BYTECRDPORT_LSB, 0x0},
+	{OP_WR, QM_REG_BYTECRDPORT_MSB, 0xffffffff},
+	{OP_WR, QM_REG_FUNCNUMSEL_LSB, 0x0},
+	{OP_WR, QM_REG_FUNCNUMSEL_MSB, 0xffffffff},
+	{OP_WR, QM_REG_CMINTEN, 0xff},
+#define QM_COMMON_END           829
+#define PBF_COMMON_START        829
+	{OP_WR, PBF_REG_INIT, 0x1},
+	{OP_WR, PBF_REG_INIT_P4, 0x1},
+	{OP_WR, PBF_REG_MAC_LB_ENABLE, 0x1},
+	{OP_WR, PBF_REG_IF_ENABLE_REG, 0x7fff},
+	{OP_WR, PBF_REG_INIT_P4, 0x0},
+	{OP_WR, PBF_REG_INIT, 0x0},
+	{OP_WR, PBF_REG_DISABLE_NEW_TASK_PROC_P4, 0x0},
+#define PBF_COMMON_END          836
+#define PBF_PORT0_START         836
+	{OP_WR, PBF_REG_INIT_P0, 0x1},
+	{OP_WR, PBF_REG_MAC_IF0_ENABLE, 0x1},
+	{OP_WR, PBF_REG_INIT_P0, 0x0},
+	{OP_WR, PBF_REG_DISABLE_NEW_TASK_PROC_P0, 0x0},
+#define PBF_PORT0_END           840
+#define PBF_PORT1_START         840
+	{OP_WR, PBF_REG_INIT_P1, 0x1},
+	{OP_WR, PBF_REG_MAC_IF1_ENABLE, 0x1},
+	{OP_WR, PBF_REG_INIT_P1, 0x0},
+	{OP_WR, PBF_REG_DISABLE_NEW_TASK_PROC_P1, 0x0},
+#define PBF_PORT1_END           844
+#define XCM_COMMON_START        844
+	{OP_WR, XCM_REG_XX_OVFL_EVNT_ID, 0x32},
+	{OP_WR, XCM_REG_XQM_XCM_HDR_P, 0x3150020},
+	{OP_WR, XCM_REG_XQM_XCM_HDR_S, 0x3150020},
+	{OP_WR, XCM_REG_TM_XCM_HDR, 0x1000030},
+	{OP_WR, XCM_REG_ERR_XCM_HDR, 0x8100000},
+	{OP_WR, XCM_REG_ERR_EVNT_ID, 0x33},
+	{OP_WR, XCM_REG_EXPR_EVNT_ID, 0x30},
+	{OP_WR, XCM_REG_STOP_EVNT_ID, 0x31},
+	{OP_WR, XCM_REG_STORM_WEIGHT, 0x2},
+	{OP_WR, XCM_REG_TSEM_WEIGHT, 0x5},
+	{OP_WR, XCM_REG_CSEM_WEIGHT, 0x2},
+	{OP_WR, XCM_REG_USEM_WEIGHT, 0x2},
+	{OP_WR, XCM_REG_PBF_WEIGHT, 0x7},
+	{OP_WR, XCM_REG_NIG1_WEIGHT, 0x1},
+	{OP_WR, XCM_REG_CP_WEIGHT, 0x0},
+	{OP_WR, XCM_REG_XSDM_WEIGHT, 0x5},
+	{OP_WR, XCM_REG_XQM_P_WEIGHT, 0x3},
+	{OP_WR, XCM_REG_XCM_XQM_USE_Q, 0x1},
+	{OP_WR, XCM_REG_XQM_BYP_ACT_UPD, 0x6},
+	{OP_WR, XCM_REG_UNA_GT_NXT_Q, 0x0},
+	{OP_WR, XCM_REG_AUX1_Q, 0x2},
+	{OP_WR, XCM_REG_AUX_CNT_FLG_Q_19, 0x1},
+	{OP_WR, XCM_REG_GR_ARB_TYPE, 0x1},
+	{OP_WR, XCM_REG_GR_LD0_PR, 0x1},
+	{OP_WR, XCM_REG_GR_LD1_PR, 0x2},
+	{OP_WR, XCM_REG_CFC_INIT_CRD, 0x1},
+	{OP_WR, XCM_REG_FIC0_INIT_CRD, 0x40},
+	{OP_WR, XCM_REG_FIC1_INIT_CRD, 0x40},
+	{OP_WR, XCM_REG_TM_INIT_CRD, 0x4},
+	{OP_WR, XCM_REG_XQM_INIT_CRD, 0x20},
+	{OP_WR, XCM_REG_XX_INIT_CRD, 0x2},
+	{OP_WR, XCM_REG_XX_MSG_NUM, 0x1f},
+	{OP_ZR, XCM_REG_XX_TABLE, 0x12},
+	{OP_SW, XCM_REG_XX_DESCR_TABLE, 0x1f4d01},
+	{OP_WR, XCM_REG_N_SM_CTX_LD_0, 0xf},
+	{OP_WR, XCM_REG_N_SM_CTX_LD_1, 0x7},
+	{OP_WR, XCM_REG_N_SM_CTX_LD_2, 0xb},
+	{OP_WR, XCM_REG_N_SM_CTX_LD_3, 0xe},
+	{OP_ZR, XCM_REG_N_SM_CTX_LD_4, 0x4},
+	{OP_WR, XCM_REG_XCM_REG0_SZ, 0x4},
+	{OP_WR, XCM_REG_XCM_STORM0_IFEN, 0x1},
+	{OP_WR, XCM_REG_XCM_STORM1_IFEN, 0x1},
+	{OP_WR, XCM_REG_XCM_XQM_IFEN, 0x1},
+	{OP_WR, XCM_REG_STORM_XCM_IFEN, 0x1},
+	{OP_WR, XCM_REG_XQM_XCM_IFEN, 0x1},
+	{OP_WR, XCM_REG_XSDM_IFEN, 0x1},
+	{OP_WR, XCM_REG_TM_XCM_IFEN, 0x1},
+	{OP_WR, XCM_REG_XCM_TM_IFEN, 0x1},
+	{OP_WR, XCM_REG_TSEM_IFEN, 0x1},
+	{OP_WR, XCM_REG_CSEM_IFEN, 0x1},
+	{OP_WR, XCM_REG_USEM_IFEN, 0x1},
+	{OP_WR, XCM_REG_DORQ_IFEN, 0x1},
+	{OP_WR, XCM_REG_PBF_IFEN, 0x1},
+	{OP_WR, XCM_REG_NIG0_IFEN, 0x1},
+	{OP_WR, XCM_REG_NIG1_IFEN, 0x1},
+	{OP_WR, XCM_REG_CDU_AG_WR_IFEN, 0x1},
+	{OP_WR, XCM_REG_CDU_AG_RD_IFEN, 0x1},
+	{OP_WR, XCM_REG_CDU_SM_WR_IFEN, 0x1},
+	{OP_WR, XCM_REG_CDU_SM_RD_IFEN, 0x1},
+	{OP_WR, XCM_REG_XCM_CFC_IFEN, 0x1},
+#define XCM_COMMON_END          904
+#define XCM_PORT0_START         904
+	{OP_WR, XCM_REG_GLB_DEL_ACK_TMR_VAL_0, 0xc8},
+	{OP_WR, XCM_REG_GLB_DEL_ACK_MAX_CNT_0, 0x2},
+	{OP_WR, XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD00, 0x0},
+	{OP_WR, XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD10, 0x0},
+	{OP_WR, XCM_REG_WU_DA_CNT_CMD00, 0x2},
+	{OP_WR, XCM_REG_WU_DA_CNT_CMD10, 0x2},
+	{OP_WR, XCM_REG_WU_DA_CNT_UPD_VAL00, 0xff},
+	{OP_WR, XCM_REG_WU_DA_CNT_UPD_VAL10, 0xff},
+#define XCM_PORT0_END           912
+#define XCM_PORT1_START         912
+	{OP_WR, XCM_REG_GLB_DEL_ACK_TMR_VAL_1, 0xc8},
+	{OP_WR, XCM_REG_GLB_DEL_ACK_MAX_CNT_1, 0x2},
+	{OP_WR, XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD01, 0x0},
+	{OP_WR, XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD11, 0x0},
+	{OP_WR, XCM_REG_WU_DA_CNT_CMD01, 0x2},
+	{OP_WR, XCM_REG_WU_DA_CNT_CMD11, 0x2},
+	{OP_WR, XCM_REG_WU_DA_CNT_UPD_VAL01, 0xff},
+	{OP_WR, XCM_REG_WU_DA_CNT_UPD_VAL11, 0xff},
+#define XCM_PORT1_END           920
+#define XSEM_COMMON_START       920
+	{OP_RD, XSEM_REG_MSG_NUM_FIC0, 0x0},
+	{OP_RD, XSEM_REG_MSG_NUM_FIC1, 0x0},
+	{OP_RD, XSEM_REG_MSG_NUM_FOC0, 0x0},
+	{OP_RD, XSEM_REG_MSG_NUM_FOC1, 0x0},
+	{OP_RD, XSEM_REG_MSG_NUM_FOC2, 0x0},
+	{OP_RD, XSEM_REG_MSG_NUM_FOC3, 0x0},
+	{OP_WR, XSEM_REG_ARB_ELEMENT0, 0x1},
+	{OP_WR, XSEM_REG_ARB_ELEMENT1, 0x2},
+	{OP_WR, XSEM_REG_ARB_ELEMENT2, 0x3},
+	{OP_WR, XSEM_REG_ARB_ELEMENT3, 0x0},
+	{OP_WR, XSEM_REG_ARB_ELEMENT4, 0x4},
+	{OP_WR, XSEM_REG_ARB_CYCLE_SIZE, 0x1},
+	{OP_WR, XSEM_REG_TS_0_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_1_AS, 0x1},
+	{OP_WR, XSEM_REG_TS_2_AS, 0x4},
+	{OP_WR, XSEM_REG_TS_3_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_4_AS, 0x1},
+	{OP_WR, XSEM_REG_TS_5_AS, 0x3},
+	{OP_WR, XSEM_REG_TS_6_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_7_AS, 0x1},
+	{OP_WR, XSEM_REG_TS_8_AS, 0x4},
+	{OP_WR, XSEM_REG_TS_9_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_10_AS, 0x1},
+	{OP_WR, XSEM_REG_TS_11_AS, 0x3},
+	{OP_WR, XSEM_REG_TS_12_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_13_AS, 0x1},
+	{OP_WR, XSEM_REG_TS_14_AS, 0x4},
+	{OP_WR, XSEM_REG_TS_15_AS, 0x0},
+	{OP_WR, XSEM_REG_TS_16_AS, 0x4},
+	{OP_WR, XSEM_REG_TS_17_AS, 0x3},
+	{OP_ZR, XSEM_REG_TS_18_AS, 0x2},
+	{OP_WR, XSEM_REG_ENABLE_IN, 0x3fff},
+	{OP_WR, XSEM_REG_ENABLE_OUT, 0x3ff},
+	{OP_WR, XSEM_REG_FIC0_DISABLE, 0x0},
+	{OP_WR, XSEM_REG_FIC1_DISABLE, 0x0},
+	{OP_WR, XSEM_REG_PAS_DISABLE, 0x0},
+	{OP_WR, XSEM_REG_THREADS_LIST, 0xffff},
+	{OP_ZR, XSEM_REG_PASSIVE_BUFFER, 0x800},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18bc0, 0x1},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18000, 0x0},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18040, 0x18},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18080, 0xc},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x180c0, 0x66},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18300, 0x7a120},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x183c0, 0x1f4},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18340, 0x1f4},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x18380, 0x1dcd6500},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x55d8, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x5000, 0x48},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1020, 0xc8},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1000, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x5128, 0x92},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x5378, 0x0},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x5380, 0x24d20},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x5428, 0x44d22},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1518, 0x1},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1830, 0x0},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1838, 0x0},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x1820, 0x24d26},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4ac0, 0x2},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x4ad8, 0x24d28},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4b08, 0x4},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x1f50, 0x24d2a},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x10800, 0x0},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x10c00, 0x104d2c},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x10800, 0x1000000},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x10c40, 0x84d3c},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x10800, 0x2000000},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x10c60, 0x84d44},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x10800, 0x3000000},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x10c80, 0x84d4c},
+	{OP_ZP, XSEM_REG_INT_TABLE, 0x814d54},
+	{OP_ZP, XSEM_REG_PRAM, 0x35774d75},
+	{OP_ZP, XSEM_REG_PRAM + 0x8000, 0x36525ad3},
+	{OP_ZP, XSEM_REG_PRAM + 0x10000, 0x27266868},
+	{OP_ZP, XSEM_REG_PRAM + 0x18000, 0x5e7232},
+	{OP_ZP, XSEM_REG_PRAM + 0x20000, 0x5e724a},
+	{OP_ZP, XSEM_REG_PRAM + 0x28000, 0x5e7262},
+	{OP_ZP, XSEM_REG_PRAM + 0x30000, 0x5e727a},
+	{OP_ZP, XSEM_REG_PRAM + 0x38000, 0x5e7292},
+#define XSEM_COMMON_END         1000
+#define XSEM_PORT0_START        1000
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1400, 0xa},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1450, 0x6},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x5388, 0xc},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x5388 + 0x30, 0x272aa},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x55e0, 0x772ac},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x5600, 0x7},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1500, 0x0},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1508, 0x1},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3020, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3030, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3000, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3010, 0x2},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x3040, 0x0},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3048, 0xc},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x3048 + 0x30, 0x272b3},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x30b8, 0x1},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x4ac8, 0x272b5},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4b18, 0x42},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4d28, 0x4},
+#define XSEM_PORT0_END          1019
+#define XSEM_PORT1_START        1019
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1428, 0xa},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x1468, 0x6},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x53c0, 0xc},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x53c0 + 0x30, 0x272b7},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x5620, 0x772b9},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x5640, 0x7},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x1504, 0x0},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x150c, 0x1},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3028, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3038, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3008, 0x2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3018, 0x2},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x3044, 0x0},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x3080, 0xc},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x3080 + 0x30, 0x272c0},
+	{OP_WR, XSEM_REG_FAST_MEMORY + 0x30bc, 0x1},
+	{OP_SW, XSEM_REG_FAST_MEMORY + 0x4ad0, 0x272c2},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4c20, 0x42},
+	{OP_ZR, XSEM_REG_FAST_MEMORY + 0x4d38, 0x4},
+#define XSEM_PORT1_END          1038
+#define CDU_COMMON_START        1038
+	{OP_WR, CDU_REG_CDU_CONTROL0, 0x1},
+	{OP_WR, CDU_REG_CDU_CHK_MASK0, 0x3d000},
+	{OP_WR, CDU_REG_CDU_CHK_MASK1, 0x3d},
+	{OP_WB, CDU_REG_L1TT, 0x20072c4},
+	{OP_WB, CDU_REG_MATT, 0x2074c4},
+	{OP_ZR, CDU_REG_MATT + 0x80, 0x20},
+#define CDU_COMMON_END          1044
+#define DMAE_COMMON_START       1044
+	{OP_WR, DMAE_REG_CRC16C_INIT, 0x0},
+	{OP_WR, DMAE_REG_CRC16T10_INIT, 0x1},
+	{OP_WR, DMAE_REG_PXP_REQ_INIT_CRD, 0x2},
+	{OP_WR, DMAE_REG_PCI_IFEN, 0x1},
+	{OP_WR, DMAE_REG_GRC_IFEN, 0x1},
+#define DMAE_COMMON_END         1049
+#define PXP_COMMON_START        1049
+	{OP_SI, PXP_REG_HST_INBOUND_INT + 0x400, 0x574e4},
+	{OP_SI, PXP_REG_HST_INBOUND_INT + 0x420, 0x574e9},
+	{OP_SI, PXP_REG_HST_INBOUND_INT, 0x574ee},
+#define PXP_COMMON_END          1052
+#define CFC_COMMON_START        1052
+	{OP_WR, CFC_REG_CONTROL0, 0x10},
+	{OP_WR, CFC_REG_DISABLE_ON_ERROR, 0x3fff},
+	{OP_WR, CFC_REG_LCREQ_WEIGHTS, 0x84924a},
+#define CFC_COMMON_END          1055
+#define HC_COMMON_START         1055
+	{OP_ZR, HC_REG_USTORM_ADDR_FOR_COALESCE, 0x4},
+#define HC_COMMON_END           1056
+#define HC_PORT0_START          1056
+	{OP_WR, HC_REG_CONFIG_0, 0x1080},
+	{OP_ZR, HC_REG_UC_RAM_ADDR_0, 0x2},
+	{OP_WR, HC_REG_ATTN_NUM_P0, 0x10},
+	{OP_WR, HC_REG_LEADING_EDGE_0, 0xffff},
+	{OP_WR, HC_REG_TRAILING_EDGE_0, 0xffff},
+	{OP_WR, HC_REG_AGG_INT_0, 0x0},
+	{OP_WR, HC_REG_ATTN_IDX, 0x0},
+	{OP_ZR, HC_REG_ATTN_BIT, 0x2},
+	{OP_WR, HC_REG_VQID_0, 0x2b5},
+	{OP_WR, HC_REG_PCI_CONFIG_0, 0x0},
+	{OP_ZR, HC_REG_P0_PROD_CONS, 0x4a},
+	{OP_ZR, HC_REG_PBA_COMMAND, 0x2},
+	{OP_WR, HC_REG_INT_MASK, 0x1ffff},
+	{OP_WR, HC_REG_CONFIG_0, 0x1a82},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS, 0x24},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x120, 0x4a},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x370, 0x4a},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x5c0, 0x4a},
+#define HC_PORT0_END            1074
+#define HC_PORT1_START          1074
+	{OP_WR, HC_REG_CONFIG_1, 0x1080},
+	{OP_ZR, HC_REG_UC_RAM_ADDR_1, 0x2},
+	{OP_WR, HC_REG_ATTN_NUM_P1, 0x10},
+	{OP_WR, HC_REG_LEADING_EDGE_1, 0xffff},
+	{OP_WR, HC_REG_TRAILING_EDGE_1, 0xffff},
+	{OP_WR, HC_REG_AGG_INT_1, 0x0},
+	{OP_WR, HC_REG_ATTN_IDX + 0x4, 0x0},
+	{OP_ZR, HC_REG_ATTN_BIT + 0x8, 0x2},
+	{OP_WR, HC_REG_VQID_1, 0x2b5},
+	{OP_WR, HC_REG_PCI_CONFIG_1, 0x0},
+	{OP_ZR, HC_REG_P1_PROD_CONS, 0x4a},
+	{OP_ZR, HC_REG_PBA_COMMAND + 0x8, 0x2},
+	{OP_WR, HC_REG_INT_MASK + 0x4, 0x1ffff},
+	{OP_WR, HC_REG_CONFIG_1, 0x1a82},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x90, 0x24},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x248, 0x4a},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x498, 0x4a},
+	{OP_ZR, HC_REG_STATISTIC_COUNTERS + 0x6e8, 0x4a},
+#define HC_PORT1_END            1092
+#define PXP2_COMMON_START       1092
+	{OP_WR, PXP2_REG_PGL_CONTROL0, 0xe38324},
+	{OP_WR, PXP2_REG_PGL_CONTROL1, 0x3c10},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_0, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_1, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_2, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_3, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_4, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_5, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_6, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_TSDM_7, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_1, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_2, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_3, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_4, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_5, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_6, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_7, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_2, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_3, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_4, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_5, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_6, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_7, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_0, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_1, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_2, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_3, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_4, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_5, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_6, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_CSDM_7, 0xffffffff},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_0, 0xffff5330},
+	{OP_WR, PXP2_REG_PGL_INT_XSDM_1, 0xffff5348},
+	{OP_WR, PXP2_REG_PGL_INT_USDM_0, 0xf0003000},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ6, 0x8},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ9, 0x8},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ10, 0x8},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ11, 0x2},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ17, 0x4},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ18, 0x5},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ19, 0x4},
+	{OP_WR, PXP2_REG_RD_MAX_BLKS_VQ22, 0x0},
+	{OP_WR, PXP2_REG_RD_START_INIT, 0x1},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD0, 0x40},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD1, 0x1808},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD2, 0x803},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD3, 0x803},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD4, 0x40},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD5, 0x3},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD6, 0x803},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD7, 0x803},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD8, 0x803},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD9, 0x10003},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD10, 0x803},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD11, 0x803},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD12, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD13, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD14, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD15, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD16, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD17, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD18, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD19, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD20, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD22, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD23, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD24, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD25, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD26, 0x3},
+	{OP_WR, PXP2_REG_RQ_BW_RD_ADD27, 0x3},
+	{OP_WR, PXP2_REG_PSWRQ_BW_ADD28, 0x2403},
+	{OP_WR, PXP2_REG_RQ_BW_WR_ADD29, 0x2f},
+	{OP_WR, PXP2_REG_RQ_BW_WR_ADD30, 0x9},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND0, 0x19},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB1, 0x184},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB2, 0x183},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB3, 0x306},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND4, 0x19},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND5, 0x6},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB6, 0x306},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB7, 0x306},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB8, 0x306},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB9, 0xc86},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB10, 0x306},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB11, 0x306},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND12, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND13, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND14, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND15, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND16, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND17, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND18, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND19, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND20, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND22, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND23, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND24, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND25, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND26, 0x6},
+	{OP_WR, PXP2_REG_RQ_BW_RD_UBOUND27, 0x6},
+	{OP_WR, PXP2_REG_PSWRQ_BW_UB28, 0x306},
+	{OP_WR, PXP2_REG_RQ_BW_WR_UBOUND29, 0x13},
+	{OP_WR, PXP2_REG_RQ_BW_WR_UBOUND30, 0x6},
+	{OP_WR, PXP2_REG_PSWRQ_BW_L1, 0x1004},
+	{OP_WR, PXP2_REG_PSWRQ_BW_L2, 0x1004},
+	{OP_WR, PXP2_REG_PSWRQ_BW_RD, 0x106440},
+	{OP_WR, PXP2_REG_PSWRQ_BW_WR, 0x106440},
+	{OP_WR, PXP2_REG_RQ_RBC_DONE, 0x1},
+#define PXP2_COMMON_END         1200
+#define MISC_AEU_COMMON_START   1200
+	{OP_ZR, MISC_REG_AEU_GENERAL_ATTN_0, 0x16},
+#define MISC_AEU_COMMON_END     1201
+#define MISC_AEU_PORT0_START    1201
+	{OP_WR, MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0, 0xbf5c0000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_FUNC_0_OUT_0, 0xfff51fef},
+	{OP_WR, MISC_REG_AEU_ENABLE3_FUNC_0_OUT_0, 0xffff},
+	{OP_WR, MISC_REG_AEU_ENABLE4_FUNC_0_OUT_0, 0x500003e0},
+	{OP_WR, MISC_REG_AEU_ENABLE1_FUNC_0_OUT_1, 0x0},
+	{OP_WR, MISC_REG_AEU_ENABLE2_FUNC_0_OUT_1, 0xa000},
+	{OP_ZR, MISC_REG_AEU_ENABLE3_FUNC_0_OUT_1, 0x5},
+	{OP_WR, MISC_REG_AEU_ENABLE4_FUNC_0_OUT_2, 0xfe00000},
+	{OP_ZR, MISC_REG_AEU_ENABLE1_FUNC_0_OUT_3, 0x14},
+	{OP_WR, MISC_REG_AEU_ENABLE1_NIG_0, 0x55540000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_NIG_0, 0x55555555},
+	{OP_WR, MISC_REG_AEU_ENABLE3_NIG_0, 0x5555},
+	{OP_WR, MISC_REG_AEU_ENABLE4_NIG_0, 0x0},
+	{OP_WR, MISC_REG_AEU_ENABLE1_PXP_0, 0x55540000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_PXP_0, 0x55555555},
+	{OP_WR, MISC_REG_AEU_ENABLE3_PXP_0, 0x5555},
+	{OP_WR, MISC_REG_AEU_ENABLE4_PXP_0, 0x0},
+	{OP_WR, MISC_REG_AEU_INVERTER_1_FUNC_0, 0x0},
+	{OP_ZR, MISC_REG_AEU_INVERTER_2_FUNC_0, 0x3},
+	{OP_WR, MISC_REG_AEU_MASK_ATTN_FUNC_0, 0x7},
+#define MISC_AEU_PORT0_END      1221
+#define MISC_AEU_PORT1_START    1221
+	{OP_WR, MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0, 0xbf5c0000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_FUNC_1_OUT_0, 0xfff51fef},
+	{OP_WR, MISC_REG_AEU_ENABLE3_FUNC_1_OUT_0, 0xffff},
+	{OP_WR, MISC_REG_AEU_ENABLE4_FUNC_1_OUT_0, 0x500003e0},
+	{OP_WR, MISC_REG_AEU_ENABLE1_FUNC_1_OUT_1, 0x0},
+	{OP_WR, MISC_REG_AEU_ENABLE2_FUNC_1_OUT_1, 0xa000},
+	{OP_ZR, MISC_REG_AEU_ENABLE3_FUNC_1_OUT_1, 0x5},
+	{OP_WR, MISC_REG_AEU_ENABLE4_FUNC_1_OUT_2, 0xfe00000},
+	{OP_ZR, MISC_REG_AEU_ENABLE1_FUNC_1_OUT_3, 0x14},
+	{OP_WR, MISC_REG_AEU_ENABLE1_NIG_1, 0x55540000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_NIG_1, 0x55555555},
+	{OP_WR, MISC_REG_AEU_ENABLE3_NIG_1, 0x5555},
+	{OP_WR, MISC_REG_AEU_ENABLE4_NIG_1, 0x0},
+	{OP_WR, MISC_REG_AEU_ENABLE1_PXP_1, 0x55540000},
+	{OP_WR, MISC_REG_AEU_ENABLE2_PXP_1, 0x55555555},
+	{OP_WR, MISC_REG_AEU_ENABLE3_PXP_1, 0x5555},
+	{OP_WR, MISC_REG_AEU_ENABLE4_PXP_1, 0x0},
+	{OP_WR, MISC_REG_AEU_INVERTER_1_FUNC_1, 0x0},
+	{OP_ZR, MISC_REG_AEU_INVERTER_2_FUNC_1, 0x3},
+	{OP_WR, MISC_REG_AEU_MASK_ATTN_FUNC_1, 0x7}
+#define MISC_AEU_PORT1_END      1241
+};
+
+static const u32 init_data[] = {
+	0x00010000, 0x000204c0, 0x00030980, 0x00040e40, 0x00051300, 0x000617c0,
+	0x00071c80, 0x00082140, 0x00092600, 0x000a2ac0, 0x000b2f80, 0x000c3440,
+	0x000d3900, 0x000e3dc0, 0x000f4280, 0x00104740, 0x00114c00, 0x001250c0,
+	0x00135580, 0x00145a40, 0x00155f00, 0x001663c0, 0x00176880, 0x00186d40,
+	0x00197200, 0x001a76c0, 0x001b7b80, 0x001c8040, 0x001d8500, 0x001e89c0,
+	0x001f8e80, 0x00209340, 0x00002000, 0x00004000, 0x00006000, 0x00008000,
+	0x0000a000, 0x0000c000, 0x0000e000, 0x00010000, 0x00012000, 0x00014000,
+	0x00016000, 0x00018000, 0x0001a000, 0x0001c000, 0x0001e000, 0x00020000,
+	0x00022000, 0x00024000, 0x00026000, 0x00028000, 0x0002a000, 0x0002c000,
+	0x0002e000, 0x00030000, 0x00032000, 0x00034000, 0x00036000, 0x00038000,
+	0x0003a000, 0x0003c000, 0x0003e000, 0x00040000, 0x00042000, 0x00044000,
+	0x00046000, 0x00048000, 0x0004a000, 0x0004c000, 0x0004e000, 0x00050000,
+	0x00052000, 0x00054000, 0x00056000, 0x00058000, 0x0005a000, 0x0005c000,
+	0x0005e000, 0x00060000, 0x00062000, 0x00064000, 0x00066000, 0x00068000,
+	0x0006a000, 0x0006c000, 0x0006e000, 0x00070000, 0x00072000, 0x00074000,
+	0x00076000, 0x00078000, 0x0007a000, 0x0007c000, 0x0007e000, 0x00080000,
+	0x00082000, 0x00084000, 0x00086000, 0x00088000, 0x0008a000, 0x0008c000,
+	0x0008e000, 0x00090000, 0x00092000, 0x00094000, 0x00096000, 0x00098000,
+	0x0009a000, 0x0009c000, 0x0009e000, 0x000a0000, 0x000a2000, 0x000a4000,
+	0x000a6000, 0x000a8000, 0x000aa000, 0x000ac000, 0x000ae000, 0x000b0000,
+	0x000b2000, 0x000b4000, 0x000b6000, 0x000b8000, 0x000ba000, 0x000bc000,
+	0x000be000, 0x000c0000, 0x000c2000, 0x000c4000, 0x000c6000, 0x000c8000,
+	0x000ca000, 0x000cc000, 0x000ce000, 0x000d0000, 0x000d2000, 0x000d4000,
+	0x000d6000, 0x000d8000, 0x000da000, 0x000dc000, 0x000de000, 0x000e0000,
+	0x000e2000, 0x000e4000, 0x000e6000, 0x000e8000, 0x000ea000, 0x000ec000,
+	0x000ee000, 0x000f0000, 0x000f2000, 0x000f4000, 0x000f6000, 0x000f8000,
+	0x000fa000, 0x000fc000, 0x000fe000, 0x00100000, 0x00102000, 0x00104000,
+	0x00106000, 0x00108000, 0x0010a000, 0x0010c000, 0x0010e000, 0x00110000,
+	0x00112000, 0x00114000, 0x00116000, 0x00118000, 0x0011a000, 0x0011c000,
+	0x0011e000, 0x00120000, 0x00122000, 0x00124000, 0x00126000, 0x00128000,
+	0x0012a000, 0x0012c000, 0x0012e000, 0x00130000, 0x00132000, 0x00134000,
+	0x00136000, 0x00138000, 0x0013a000, 0x0013c000, 0x0013e000, 0x00140000,
+	0x00142000, 0x00144000, 0x00146000, 0x00148000, 0x0014a000, 0x0014c000,
+	0x0014e000, 0x00150000, 0x00152000, 0x00154000, 0x00156000, 0x00158000,
+	0x0015a000, 0x0015c000, 0x0015e000, 0x00160000, 0x00162000, 0x00164000,
+	0x00166000, 0x00168000, 0x0016a000, 0x0016c000, 0x0016e000, 0x00170000,
+	0x00172000, 0x00174000, 0x00176000, 0x00178000, 0x0017a000, 0x0017c000,
+	0x0017e000, 0x00180000, 0x00182000, 0x00184000, 0x00186000, 0x00188000,
+	0x0018a000, 0x0018c000, 0x0018e000, 0x00190000, 0x00192000, 0x00194000,
+	0x00196000, 0x00198000, 0x0019a000, 0x0019c000, 0x0019e000, 0x001a0000,
+	0x001a2000, 0x001a4000, 0x001a6000, 0x001a8000, 0x001aa000, 0x001ac000,
+	0x001ae000, 0x001b0000, 0x001b2000, 0x001b4000, 0x001b6000, 0x001b8000,
+	0x001ba000, 0x001bc000, 0x001be000, 0x001c0000, 0x001c2000, 0x001c4000,
+	0x001c6000, 0x001c8000, 0x001ca000, 0x001cc000, 0x001ce000, 0x001d0000,
+	0x001d2000, 0x001d4000, 0x001d6000, 0x001d8000, 0x001da000, 0x001dc000,
+	0x001de000, 0x001e0000, 0x001e2000, 0x001e4000, 0x001e6000, 0x001e8000,
+	0x001ea000, 0x001ec000, 0x001ee000, 0x001f0000, 0x001f2000, 0x001f4000,
+	0x001f6000, 0x001f8000, 0x001fa000, 0x001fc000, 0x001fe000, 0x00200000,
+	0x00202000, 0x00204000, 0x00206000, 0x00208000, 0x0020a000, 0x0020c000,
+	0x0020e000, 0x00210000, 0x00212000, 0x00214000, 0x00216000, 0x00218000,
+	0x0021a000, 0x0021c000, 0x0021e000, 0x00220000, 0x00222000, 0x00224000,
+	0x00226000, 0x00228000, 0x0022a000, 0x0022c000, 0x0022e000, 0x00230000,
+	0x00232000, 0x00234000, 0x00236000, 0x00238000, 0x0023a000, 0x0023c000,
+	0x0023e000, 0x00240000, 0x00242000, 0x00244000, 0x00246000, 0x00248000,
+	0x0024a000, 0x0024c000, 0x0024e000, 0x00250000, 0x00252000, 0x00254000,
+	0x00256000, 0x00258000, 0x0025a000, 0x0025c000, 0x0025e000, 0x00260000,
+	0x00262000, 0x00264000, 0x00266000, 0x00268000, 0x0026a000, 0x0026c000,
+	0x0026e000, 0x00270000, 0x00272000, 0x00274000, 0x00276000, 0x00278000,
+	0x0027a000, 0x0027c000, 0x0027e000, 0x00280000, 0x00282000, 0x00284000,
+	0x00286000, 0x00288000, 0x0028a000, 0x0028c000, 0x0028e000, 0x00290000,
+	0x00292000, 0x00294000, 0x00296000, 0x00298000, 0x0029a000, 0x0029c000,
+	0x0029e000, 0x002a0000, 0x002a2000, 0x002a4000, 0x002a6000, 0x002a8000,
+	0x002aa000, 0x002ac000, 0x002ae000, 0x002b0000, 0x002b2000, 0x002b4000,
+	0x002b6000, 0x002b8000, 0x002ba000, 0x002bc000, 0x002be000, 0x002c0000,
+	0x002c2000, 0x002c4000, 0x002c6000, 0x002c8000, 0x002ca000, 0x002cc000,
+	0x002ce000, 0x002d0000, 0x002d2000, 0x002d4000, 0x002d6000, 0x002d8000,
+	0x002da000, 0x002dc000, 0x002de000, 0x002e0000, 0x002e2000, 0x002e4000,
+	0x002e6000, 0x002e8000, 0x002ea000, 0x002ec000, 0x002ee000, 0x002f0000,
+	0x002f2000, 0x002f4000, 0x002f6000, 0x002f8000, 0x002fa000, 0x002fc000,
+	0x002fe000, 0x00300000, 0x00302000, 0x00304000, 0x00306000, 0x00308000,
+	0x0030a000, 0x0030c000, 0x0030e000, 0x00310000, 0x00312000, 0x00314000,
+	0x00316000, 0x00318000, 0x0031a000, 0x0031c000, 0x0031e000, 0x00320000,
+	0x00322000, 0x00324000, 0x00326000, 0x00328000, 0x0032a000, 0x0032c000,
+	0x0032e000, 0x00330000, 0x00332000, 0x00334000, 0x00336000, 0x00338000,
+	0x0033a000, 0x0033c000, 0x0033e000, 0x00340000, 0x00342000, 0x00344000,
+	0x00346000, 0x00348000, 0x0034a000, 0x0034c000, 0x0034e000, 0x00350000,
+	0x00352000, 0x00354000, 0x00356000, 0x00358000, 0x0035a000, 0x0035c000,
+	0x0035e000, 0x00360000, 0x00362000, 0x00364000, 0x00366000, 0x00368000,
+	0x0036a000, 0x0036c000, 0x0036e000, 0x00370000, 0x00372000, 0x00374000,
+	0x00376000, 0x00378000, 0x0037a000, 0x0037c000, 0x0037e000, 0x00380000,
+	0x00382000, 0x00384000, 0x00386000, 0x00388000, 0x0038a000, 0x0038c000,
+	0x0038e000, 0x00390000, 0x00392000, 0x00394000, 0x00396000, 0x00398000,
+	0x0039a000, 0x0039c000, 0x0039e000, 0x003a0000, 0x003a2000, 0x003a4000,
+	0x003a6000, 0x003a8000, 0x003aa000, 0x003ac000, 0x003ae000, 0x003b0000,
+	0x003b2000, 0x003b4000, 0x003b6000, 0x003b8000, 0x003ba000, 0x003bc000,
+	0x003be000, 0x003c0000, 0x003c2000, 0x003c4000, 0x003c6000, 0x003c8000,
+	0x003ca000, 0x003cc000, 0x003ce000, 0x003d0000, 0x003d2000, 0x003d4000,
+	0x003d6000, 0x003d8000, 0x003da000, 0x003dc000, 0x003de000, 0x003e0000,
+	0x003e2000, 0x003e4000, 0x003e6000, 0x003e8000, 0x003ea000, 0x003ec000,
+	0x003ee000, 0x003f0000, 0x003f2000, 0x003f4000, 0x003f6000, 0x003f8000,
+	0x003fa000, 0x003fc000, 0x003fe000, 0x003fe001, 0x00000000, 0x000001ff,
+	0x00000200, 0x00000001, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0x00088b1f, 0x00000000,
+	0x51fbff00, 0x03f0c0cf, 0x3130ef8a, 0x22b1c430, 0x3b0143f8, 0x02ecdd01,
+	0xdc406ec4, 0x19b7c404, 0x23dfd348, 0xf1476080, 0x03343031, 0x032f3731,
+	0x423f2483, 0x4d5011fc, 0x02ef9025, 0xa40cdb15, 0x77280475, 0xf2c060fb,
+	0x77629812, 0x056c1144, 0x58c8f22c, 0x4dde4d11, 0x44af950c, 0xe340ff40,
+	0xfca8b235, 0x6d081948, 0x8b5f150b, 0x95051f26, 0xd0849577, 0xe76964eb,
+	0x00607a36, 0x2726b9d6, 0x00000400, 0x00088b1f, 0x00000000, 0x7dedff00,
+	0xd554780b, 0x333ef0b5, 0x64ccce67, 0x093c991e, 0x20f264af, 0xf09c0682,
+	0x93a8a808, 0x07be3040, 0x0e22a5e4, 0x27902018, 0xf5e8bd48, 0x620c19bf,
+	0x2f06d6b4, 0x93a45a2a, 0xb6968a80, 0x6c1a06c1, 0x822203b4, 0x6b06f5bf,
+	0x368b6d7b, 0x2062a28a, 0xa5ebd8b9, 0xaffadaf7, 0x99def6b5, 0x91332673,
+	0xfebffdaa, 0x5fa7f7df, 0xf7b3ecdd, 0xf5ed7bd9, 0xb3ef6b5e, 0xa66e6547,
+	0x97d8ce5d, 0x9be507f8, 0x232c630a, 0xa1bbd65a, 0xed58cc9c, 0x9ef8731e,
+	0xec66c65c, 0x4f2e44b1, 0x12ab7a87, 0xf4dd42b6, 0x4fda9d92, 0x7af5e56f,
+	0x9743f773, 0xb9fb3b40, 0x05053d99, 0x589bb1eb, 0x6c276309, 0xf2f5ff8c,
+	0xaf3b72fa, 0x5feeb6d6, 0x557fa0cc, 0xe1d995a7, 0x661d13fd, 0x3cd7d63f,
+	0xc01984a5, 0x3eefbb50, 0xbf8c046d, 0xdbb4ac22, 0x0a7f50bd, 0xcafb421e,
+	0xfb18730e, 0x33bbb9f7, 0x4ec64e03, 0x5798da36, 0x937ef843, 0xd8c453d9,
+	0x59eef0aa, 0xaadfa023, 0x04cf8a5d, 0xaadaacf3, 0x8c9f2e44, 0x19b095cf,
+	0xe9dea886, 0x1cb1de60, 0xcd192f86, 0xf358eb4b, 0xe30bcc24, 0x0b45b532,
+	0x4dbe70b8, 0xc515d79a, 0x0f46c9cf, 0xb5eb23cd, 0xf03cc2cf, 0x144fdd5e,
+	0xceb12e1f, 0x30ed82c4, 0xf67de9ff, 0xb89ddb85, 0xa15af5be, 0x258ebf4b,
+	0xab1d717b, 0x2cdaadc2, 0xaad5c227, 0x8e8a2f2d, 0xcd33bd57, 0xfc96d708,
+	0x7b5d4161, 0x91b2796c, 0xb4616f31, 0x7f318abe, 0x0fe113bb, 0x47c7b36b,
+	0x29641f9f, 0x9deacf44, 0x45b5e666, 0x442c67c7, 0x17cccdcf, 0x2eb2bc41,
+	0xb74f4f97, 0xdd231e33, 0x7788a4d6, 0x7df3c013, 0x024d8741, 0xf843df4f,
+	0x7bf64ca0, 0xfeb0abd6, 0xa3cc99e4, 0x26fef10c, 0x1ed85b0b, 0x900bbd67,
+	0x1630a619, 0xb7822664, 0xc26f058e, 0x50d4cfb2, 0x5fc3c005, 0xeb002b24,
+	0xefe14fbd, 0xd4bccf5f, 0x9ad1beff, 0xe9bae91f, 0xe6ed92ca, 0x7496b15c,
+	0xfa7f2fac, 0xb5321801, 0xbf10cfc2, 0x88ade22a, 0x43321e16, 0xca576bbb,
+	0x7abc07c4, 0xc72d95fc, 0x4d93dcf9, 0xa678fa06, 0xa9ea1927, 0xf0635333,
+	0xb89cf4eb, 0x4e01d440, 0x827fa9ab, 0x6958cf9a, 0xedf88db6, 0xe48d6c8e,
+	0x38cb8ee6, 0x3b64775c, 0x7fa821c3, 0x08b85f17, 0x42f05aea, 0xc07c1c4f,
+	0x859626cc, 0xa6c4d065, 0x466f6e0d, 0x941f023c, 0xf8517ce5, 0xa6f5941e,
+	0x2814c2fe, 0x21a52b57, 0xc446cbc4, 0x330e9423, 0x3b75c06b, 0xd4f08cac,
+	0x7b64a63c, 0xfba78748, 0xb94f0173, 0xb7ef71d1, 0x1f316434, 0xca840f63,
+	0xc070ea43, 0xf7102e6f, 0x3cb78462, 0xf7802a12, 0x42c8ef73, 0x9034da7c,
+	0x1afcfd03, 0xf3445fcc, 0x1f1e20b7, 0x9d8c7415, 0xcd3856df, 0xaf3dbf30,
+	0x5dbf30ca, 0x2781f983, 0x2b5d089f, 0x8e3c07e6, 0xec07ec60, 0x96df9a5a,
+	0x6fe68eb7, 0x619558d6, 0xf981a4fe, 0xd3ef38c3, 0x2e6fe609, 0xfeb8d8bc,
+	0xf5c655bc, 0xcffd7c6b, 0xf989e685, 0x3ffd6893, 0xfaf8f362, 0xebe2eef3,
+	0xfab8c2b7, 0xf803ddf9, 0xefbc476f, 0xdb7f804d, 0xeb8dd379, 0xae31afcf,
+	0x7fe7cadf, 0x988b7421, 0xfff349df, 0xd7c2dd82, 0xcd377f9f, 0xf836b56f,
+	0xcd64d03c, 0x23086a49, 0x7fe17b5f, 0x802ca0c3, 0x5942a679, 0xbc18ca94,
+	0x47961dff, 0x2923b878, 0xfff61e78, 0xcdf8093c, 0x2c0bd519, 0xb94151bc,
+	0x5d3c13af, 0xf6896bb9, 0xb2a5783d, 0x064beb93, 0xc00c74fa, 0xb3f3ba77,
+	0xf000ffcf, 0xf628ee56, 0x8f24bd99, 0x265bdf0c, 0xe66f5296, 0x902c60f8,
+	0xfa85db3d, 0x673d9029, 0x59f9353c, 0x4645e826, 0xe3e20e30, 0x13962d65,
+	0x5af93a3d, 0x5f58c5b1, 0x25d63619, 0x24c8a5dc, 0xd8ca8650, 0xf79806d8,
+	0x0623e804, 0xd07df27a, 0x647e5847, 0xdda2b761, 0x15f400f8, 0xb572f4d3,
+	0x4272e89e, 0xb13ff8e5, 0xf8f241c5, 0x1ad5a6f9, 0x1c7847cb, 0x7cdd6480,
+	0x1156f621, 0x58be73ac, 0x04b9e127, 0xcf5f15f5, 0x6bdaaefc, 0xdc02c4c0,
+	0x4ef78669, 0xd416225b, 0xf0b0b75b, 0xfe3059bd, 0xb6ee0f6d, 0xf8ff4904,
+	0xae489a47, 0xc81348d9, 0x968582f5, 0xef747bf7, 0x64d8ec2d, 0x8de50919,
+	0x9bf3e341, 0xd3f58cab, 0x84c5b096, 0xc2a57976, 0x5bfc615a, 0x72ed8c1a,
+	0x54b13f9e, 0xdf31674e, 0xf0c5a07c, 0x06575c54, 0xe1e82fd1, 0x3ebb00eb,
+	0x87da246b, 0x53df14db, 0xfb05bf50, 0x1e3a444d, 0xe2f9f0d6, 0x07be2965,
+	0x997860d8, 0xdf40930a, 0x78dd8577, 0x743cb557, 0xfe183291, 0x7e1c979e,
+	0xebc184d3, 0x56fb8588, 0xdc3a21e6, 0x7cf8ceba, 0x7d762849, 0x3bea0f9c,
+	0xd03ed34b, 0xbf6daf3d, 0x1d03ed32, 0x9cef54bf, 0x0cafa86d, 0xbfe868df,
+	0x4312cb62, 0x9596b2fb, 0x9adbf686, 0x4bea1b57, 0xfa1a7742, 0xbd6ebadf,
+	0x8696fda1, 0x37ed0dfb, 0xd4326d57, 0x6e3e0c6f, 0x6160bfe8, 0x795da1ab,
+	0xfa0e6569, 0x305af537, 0xfde03867, 0xacacefd4, 0xb1f2894d, 0x1b8ff3e4,
+	0xd93ca8b3, 0x3d72a5e8, 0xbfc82bca, 0xeb2f1f69, 0xa0e496db, 0xffbe4b9c,
+	0x8d90d2c8, 0xdfcb1272, 0xcb18f2b1, 0x6837c8c7, 0xf3d91287, 0x5005851a,
+	0x6e14fbee, 0x77f3e48f, 0xec65fe84, 0x1ab7921e, 0xcd63cb8d, 0x50cbc3f3,
+	0x48b46a5e, 0xf1338361, 0xa15dacb8, 0x46d63075, 0x830cace3, 0x9ae81854,
+	0x77b3806f, 0xafe699bf, 0x22e3e743, 0x2581f7b4, 0x791fce0a, 0xf186fb39,
+	0x297f8f08, 0x48333bd5, 0x5636f62f, 0x22a07da4, 0x7e5402fe, 0xca90b8dc,
+	0x2a418d13, 0xa2ac683f, 0x06c6fdf2, 0xd71a7b2a, 0x636ef951, 0x8d63ca88,
+	0xbefe54cd, 0xb7ca85b1, 0xf950b71b, 0x9530c6db, 0x54fd1ba7, 0xb5a9f00e,
+	0x43fd10bf, 0x432b07f6, 0x0ebd9717, 0xcdc816fe, 0x7737e919, 0xe11afaf2,
+	0x4bc22737, 0xe213dd2c, 0x434c858f, 0x89292dd1, 0xc4c923d3, 0xf9c8182a,
+	0xfaf6e303, 0x8abaf296, 0xe008032a, 0x0397fbd3, 0x860d22e3, 0xde3d357d,
+	0xf683bb41, 0xd93365ef, 0x99f9163f, 0x1e9706ef, 0xd423401f, 0x8474bf37,
+	0x35fd029f, 0x7e72f14a, 0x9cbc0af9, 0x8dddbc39, 0x964d747a, 0xa4c1f3c9,
+	0x6dabebc4, 0x7538f5cf, 0x1a77d4f0, 0x945a67eb, 0x7a0fee0a, 0x478ee793,
+	0x3e78f07e, 0x65ba4028, 0x59c72951, 0x3e79a593, 0x617ec348, 0x95db0f5a,
+	0xf105fc42, 0xb6fbf508, 0x4e3448e1, 0x760e8e14, 0x1f27de1c, 0xff713f3b,
+	0xfea17c84, 0x9a3f4349, 0x473e5975, 0xff856abb, 0x1401897f, 0xc72ea953,
+	0x87376fad, 0xf3e217ac, 0xe0f9865d, 0xf58caf3d, 0x8a1bccbe, 0x427654ff,
+	0xa4a807f2, 0xacde22a3, 0x18f769de, 0xa18a75f5, 0xdc39df5e, 0xf8dfd063,
+	0x3657900f, 0x5ed15153, 0xe8b608d5, 0x0acd9d53, 0xf90bb7c0, 0xaf52e806,
+	0xb6b57ef0, 0x8f082d1e, 0xcd3474ce, 0x3d8bc4bf, 0xb1bd685b, 0x3c6c9df0,
+	0xc4d555ec, 0xcf9b57f0, 0xe38811dc, 0xf0ae7f97, 0xc4c6a538, 0x0b665ffd,
+	0x584e51b9, 0x873dc856, 0x07399bf8, 0x0b7143f0, 0xcbaba3cc, 0xe9afc071,
+	0x7acf678f, 0x8dafdc4d, 0x526ad79d, 0x757f09ce, 0xc2ce8ebb, 0xb1e2c775,
+	0xb43ff3ae, 0xcc8dc520, 0xdf894780, 0x6ac04a5d, 0xed57f182, 0xf9434f7c,
+	0x2a12d8fa, 0xc4dce7fc, 0xbf0c19f8, 0x2384eb33, 0x7b35ceba, 0xad5fe45c,
+	0xede224d9, 0x79c6eb10, 0x13134e97, 0x74bd017f, 0x62e58070, 0x26dfa826,
+	0x5dee326a, 0xfe4d51da, 0xa42c87d5, 0x89f53fa1, 0xfd04a5b9, 0xaded5583,
+	0x3ce01f9a, 0x88154cc5, 0x4dec53af, 0xbd1d24f7, 0xd11a4c8a, 0xa366b6e9,
+	0xa6fe00df, 0xa6fe0ea1, 0xcd9cbea1, 0x7638c4c9, 0xb80b66f9, 0x434eb4be,
+	0x879328fb, 0x3582b0e8, 0x7afb446c, 0xfcd263bc, 0x3f0e904b, 0xf104b27f,
+	0x479a14bb, 0x8f1e22a6, 0xd6ff1e12, 0xfbe257f9, 0x1bcf0713, 0x7c98dbe5,
+	0xed43c080, 0x8e1fc54e, 0x991737c3, 0xfe4abf38, 0x4ff080da, 0x2dcdfa89,
+	0x4d6bf531, 0x6b724a8a, 0xe3a46666, 0x642d8f29, 0x5f76a64a, 0x7a12f004,
+	0x026beade, 0x12a3fafe, 0xbb226d98, 0x74c0991e, 0x04a8fefd, 0xf2af79e9,
+	0x013472fa, 0xc04c8d4c, 0x06d80b3c, 0xd04d3be2, 0x60f08ad7, 0x1aa59cbc,
+	0x728f59d6, 0x9dd8e30d, 0xb7df0c1d, 0xf1da637b, 0xc637681f, 0x8e1bb232,
+	0x2776c6b1, 0x87f219b0, 0xbe7c67cb, 0x180fc842, 0x4c3be222, 0xebdfa17b,
+	0x662339b6, 0x34d94bf0, 0xce2077b5, 0xc878c3c8, 0xfe91813d, 0x645e52f3,
+	0xaff787ad, 0xb5847913, 0x4b0d94ef, 0xa97fc21d, 0x84b61b3f, 0xb57574d3,
+	0xa97e435f, 0x12c3bb3f, 0xf40df49e, 0x989617a0, 0x279b9519, 0xca236094,
+	0xd49bc4ad, 0x3c3d517f, 0xcb97a3ca, 0xfef431ad, 0x03a470da, 0xec70753d,
+	0x482d8252, 0xbe3858f7, 0x8359f8f6, 0xadfadfc6, 0x68305f8e, 0xf0f19dfc,
+	0x631c7a78, 0x0337a4dd, 0xfee80cc9, 0xa8f3dd9f, 0x8ff7444c, 0x85233e41,
+	0x58f84fe8, 0x4b79e344, 0xb8f8cbac, 0x59e5ebaa, 0x81575718, 0xfe05eb7f,
+	0x485ac95d, 0x294c448f, 0xb335cfc2, 0x55de0e88, 0xfea6bf5d, 0xff783aab,
+	0xfd4ed66a, 0x3119bf31, 0xe1927bac, 0x7def293d, 0x9cd49ec2, 0x1d11612d,
+	0x790b763f, 0xe087bf00, 0x106bcf93, 0xb26bcc3e, 0xb6a79a6e, 0xcf30cd3d,
+	0xf6735f80, 0xf9d11662, 0x376ab53d, 0xd9ed77f1, 0x019e68cb, 0x6afe067b,
+	0xfc6a1b44, 0xa3b80691, 0xc0334fe3, 0x3a7f1d1b, 0xcfc72bcd, 0xca72039a,
+	0x2be513f5, 0xc293cb42, 0xb7a7804d, 0x91f7c1ac, 0x9b5cb25f, 0x76415f73,
+	0x04f1fa5a, 0x744fc7d4, 0x1d34df6c, 0xea09fae8, 0x9b975c39, 0x739eb9b9,
+	0x92174c86, 0x7c853afe, 0x18bfa030, 0x43f1afe8, 0xb7e8c7b4, 0x8c0af851,
+	0x28f59a3f, 0x3746dfa8, 0x8e50f518, 0xef84dd33, 0x3e33ace5, 0x019ea2f7,
+	0x521b95bd, 0x9f5ce263, 0x6fcfc709, 0x9f4f3e36, 0xc73aba51, 0xd3516e07,
+	0x2798a533, 0xba505f4a, 0xb187a47b, 0x7957d40b, 0xe2d299fd, 0x79ff50e1,
+	0x1532bad3, 0x7ff4d5fc, 0x86ce2d2b, 0x91b2d3ba, 0xca15abd7, 0xb7cc592b,
+	0x89be8594, 0xcac0081e, 0x6294728d, 0x0a9cfc06, 0xccbf9b1c, 0x5fb47a8b,
+	0x9e8478e3, 0xd3d19f80, 0x39467e38, 0xfc46ed8f, 0xd67a98e4, 0xc3f973a3,
+	0x052bff17, 0xe62f4643, 0x1e0f013d, 0xf483c1b6, 0xbed34781, 0x8ebfc21c,
+	0x5f2533f5, 0xf305b14c, 0x4b938e10, 0x167f7ec7, 0xeaa27d3c, 0x5b0dc500,
+	0x71f9fe44, 0xf46eb710, 0x326f55bb, 0x3364e3f2, 0x0965d7cf, 0x378ceebf,
+	0xfd487937, 0xa195959e, 0x53eae0ea, 0x9fc8ddfb, 0x5f1c5f7d, 0xd6237e8c,
+	0x13d08653, 0x32a359f5, 0xa3139254, 0x749e667e, 0xc14aa3e2, 0x5b378847,
+	0xb6cf3466, 0x7e510942, 0xc0b5fa44, 0xbf5c1e01, 0x7f31aa12, 0x8edfc179,
+	0xe0d6e3a7, 0x775866d5, 0x83c85985, 0xbdb0b98b, 0xf08f2ab7, 0xf6836e96,
+	0x6b3d688c, 0xe809cee9, 0x0398a4e7, 0xc37be2d7, 0xc6cd9f97, 0x43d98e7a,
+	0xb4a2dbf8, 0x00f8470c, 0xc3cfb48f, 0x569d82f6, 0xdcc93168, 0xcf26f64f,
+	0x69ce2219, 0x1f4acc6b, 0x55cf2fca, 0xde718b83, 0xf7bebdfc, 0xf1fc619b,
+	0xf70f9a95, 0x9bafc65b, 0xccf88e99, 0xc03d7132, 0x72f390ee, 0xf82f5f3d,
+	0xf8aacfae, 0x77ded0fd, 0x8435a7bb, 0xda0b33ed, 0xb519afd1, 0xfdc3ebce,
+	0x42873e80, 0x7a35f7ef, 0xe7282f29, 0x1c95dd1d, 0x9e49bbb1, 0xf3c8373f,
+	0xd07d633a, 0x3b7a5f7c, 0x7fbc707b, 0x9c42f8f6, 0x1d0f949e, 0x67d9e82d,
+	0x725dfbe6, 0xb42cd1be, 0x7391fd1f, 0xcc5fef9d, 0x7fda74ae, 0xba037410,
+	0x925e9084, 0xfbdf402e, 0x121b5f10, 0x78bef7d3, 0x6b6e5df4, 0x5db946c8,
+	0x659f6815, 0xe625e781, 0xafd883e0, 0x21f76166, 0x50decd0b, 0xe927fc88,
+	0x70d0d4ce, 0x740353d4, 0xfc21497f, 0x7c717667, 0xb7361fd6, 0xcb1e7ee2,
+	0x40b7ae0a, 0x42ca99fb, 0xa22627ca, 0xfe75f8f3, 0x017cbd74, 0xfce2b7dd,
+	0x9a8ebbe3, 0xef493e10, 0x49de44c7, 0xffecadca, 0xbdfa598c, 0x0f76be62,
+	0xfc55f7cd, 0x166eb457, 0x7c780b8d, 0x80dfcf56, 0xff9c76c7, 0x1fe166c7,
+	0x3b63c60f, 0x3c073ff6, 0x3c1739ee, 0x939de87e, 0xcfe30799, 0x7f093b32,
+	0x87329f2b, 0xbea4b0ad, 0xf52fd2d9, 0x333c335f, 0x5e5627aa, 0xd71fb0d4,
+	0x8549f2b9, 0x6bb2acf8, 0xc26c35c5, 0xf96378e0, 0xe3e910c2, 0xd903b8d8,
+	0x5ab2f912, 0x7f13178e, 0xff07b354, 0x734e4cc3, 0xf54ef498, 0xbffb634e,
+	0x9b30f945, 0x24ce04f2, 0xf1b1b79e, 0xf1d0b2f1, 0xe248be47, 0x8b26717c,
+	0x619e1c91, 0xd0f7c429, 0xbb608bee, 0x33fca2e6, 0xdc6291db, 0x314cac9f,
+	0x6e8e4fec, 0x4ff21930, 0xc9b1f3b1, 0x5fec4cca, 0x6730e42a, 0x6fb0c99d,
+	0xee321b97, 0xa737e5eb, 0x5a63ea1a, 0x9ffa1846, 0xd0cf25b7, 0xbe7c2e3e,
+	0x9d82fda1, 0xf1f50d13, 0xfa1b17ed, 0x6659d85f, 0xae4717a8, 0x749ff432,
+	0x3ea18e78, 0x4326fba5, 0x2f6fa9ff, 0xd45fb435, 0x7ed0cab3, 0x0c6bc7c9,
+	0xeffb4bf5, 0x2c8ff432, 0x00ac38a9, 0xff025efd, 0xe086f314, 0x67c8a857,
+	0x99acfe07, 0xfc0e4f90, 0xa8c4d0b9, 0x2bc867dd, 0xa673f81e, 0xf81f9f1f,
+	0xfb1f8973, 0xe7cb3295, 0x5952e2f1, 0xdb19b0ce, 0x0be05553, 0xc0d96fb4,
+	0x569a9817, 0x7203f28d, 0x39454235, 0x0fc3a7fb, 0x469d0215, 0x46482cac,
+	0x34f20d3b, 0xbd373d42, 0x27ef8394, 0x962da792, 0x290f1b96, 0xac7ded83,
+	0x38e590be, 0x5bade655, 0xbad37de8, 0x65a7616c, 0xb230f17d, 0x5f9f9233,
+	0x7a1c5333, 0xe3d70366, 0x86a43667, 0xdf9efef9, 0x0fe75c39, 0x24e8399e,
+	0x9efdcefe, 0xaf01db93, 0x892cebd3, 0x7f2901da, 0x323f22f9, 0x7945582a,
+	0xe3fe418f, 0xb8f59ea2, 0x1ff573ce, 0x99c6fd63, 0xff021e92, 0x05d1d709,
+	0xf1f68972, 0x27da1483, 0x7161e35a, 0x36b6eff7, 0x16770f36, 0x4b2e7e0c,
+	0xdce1cd65, 0xb40f30fb, 0xe159f14f, 0x1fd86dfd, 0xbc587bec, 0xb5ea26d7,
+	0xbd454d21, 0xebe22b56, 0x521e71d8, 0x7e466f4b, 0xdd96a1d8, 0xc5d14eac,
+	0xdc5a0dd3, 0xd077c7bc, 0x0473c089, 0x0d80efaa, 0xcf5fefe6, 0xff306d7c,
+	0x0ca78b1a, 0xf4ee6beb, 0x9af21d12, 0x7ae1667b, 0x5dc13a73, 0x73fbd9d7,
+	0xc056a780, 0xd4f08b53, 0xa714fda9, 0xb0a8bff5, 0x879c5cb6, 0x5f66bb8f,
+	0x487e432e, 0xd51ce704, 0x72057c42, 0x738e3770, 0x4ed09aa7, 0xf28e7ddf,
+	0x638d5eb0, 0xf2ec065f, 0xf15bed52, 0x137072c7, 0x9eb44a78, 0xe196d0ff,
+	0x9d39b728, 0xc0a89fa5, 0x60e039b7, 0x9f22a726, 0xd1bef632, 0xebef623f,
+	0xf344d0fc, 0x7689557e, 0xf1157def, 0x0631f7ce, 0xc7231dc1, 0xe46f75de,
+	0xebe9a417, 0x4c17e7af, 0x10950012, 0x306c14bc, 0xe5e2db88, 0x95f6738b,
+	0x2427561e, 0xbdefe786, 0x377881ce, 0x4bfd5e46, 0x7c8f5ec7, 0xb311fb77,
+	0xb2845562, 0xc84178c1, 0xe677b2eb, 0xf0e8a664, 0xe5b1b5b3, 0xaf27556a,
+	0xf7f012a3, 0x9ff72a9b, 0x5c3635b6, 0xe380595d, 0x3d7a4c7a, 0xca13dd90,
+	0xd27988dd, 0x8c2f870e, 0x9327acfa, 0x702b58f0, 0xbe0a6c90, 0xf4bcfc89,
+	0xd7ca6ec2, 0xe97c88e6, 0x297e7168, 0x2ae3be76, 0x97f847cf, 0x8c2958da,
+	0x16cda7fb, 0x34b277f2, 0x08e28798, 0x5a5b9d07, 0x0a6c0636, 0x93db2fb4,
+	0x0c53f5c6, 0xd0c6e15c, 0xa20a1bd5, 0xc71c02f3, 0x3dbc1e51, 0x7f3859c1,
+	0x4451c22e, 0xe11276f2, 0xe84d48fb, 0x473c9dcf, 0xd21beb79, 0x43c6c4fe,
+	0xf0e2fd09, 0xa7e485fd, 0x7e4dffe6, 0xa1c3fa31, 0x0c3fa8a2, 0xd058b2eb,
+	0xeb8c594e, 0x8b55d9d2, 0x95d6dce5, 0xaeda1964, 0x92416e3e, 0x32407389,
+	0x9ae1230f, 0x57f2e169, 0xe618d1ef, 0x4e51ead3, 0x54d9f0e3, 0xbb88628b,
+	0x3138adbf, 0xede6553f, 0x7887bda4, 0xe5c2e311, 0x33a3e24e, 0x0bfbd6de,
+	0x877c7327, 0xc8f7c086, 0x04322375, 0x0f887d5e, 0x16f1c789, 0xbc069ea3,
+	0xb5e2e2fa, 0x326af006, 0x57ca75e0, 0x95fc95e0, 0xa98f9daf, 0x9cefee34,
+	0xa5dff816, 0xde703304, 0x5bbc3753, 0xbab0f982, 0x086b9f79, 0x679ed7d7,
+	0xe7e9b32e, 0x981ef991, 0x9cf383c7, 0x9aafe1be, 0xe825a7d0, 0x39ed3667,
+	0x35cf192f, 0x93f38ad4, 0xfd5af6a7, 0x3e5f8e8a, 0x1e70f2fd, 0x1b4fa5ea,
+	0xf2c7ccaf, 0x5bcf2a54, 0xf53bda9e, 0xa9d5f210, 0xb12d1c9f, 0x9ea738c0,
+	0xd2a25a3f, 0x60d85b3e, 0xef2f6a7f, 0x203b412e, 0x7c03daf3, 0xdfaf8b67,
+	0xfcfe0b56, 0x4aafebda, 0xc8d5e788, 0xef1ca707, 0xebfae096, 0x0f09675c,
+	0x7cb9db0f, 0x6da1923c, 0xd32332c4, 0x17fa10f4, 0x4fc1a45a, 0x5b8788db,
+	0x75265822, 0x7971fbb7, 0x23bbb7ce, 0xfa7bba58, 0xbd7618fd, 0xe201fe05,
+	0x67b1f3d5, 0xbe1fd63f, 0x33e504f2, 0x1ab017d9, 0x80916380, 0x3b59f7e3,
+	0x3a04a91e, 0xf37d2efc, 0x7e86e423, 0xf685eb01, 0xa0dff19b, 0x1e818e87,
+	0xdc6c80f2, 0x7cfb72a4, 0xb512a553, 0x191ff853, 0xe9187947, 0x8a3545d3,
+	0xeea5aec1, 0xd3f94199, 0x6429af3e, 0x61a47581, 0xbff81a56, 0x66ceeb4f,
+	0xc5cec0df, 0x3883bf96, 0x9d9d7de5, 0x0952acfa, 0x67df30a9, 0x7fac8f42,
+	0xd3f1127d, 0x178a65e6, 0xfefb18f1, 0xcb5eeb63, 0x727a099f, 0xf3879dd8,
+	0x6efc8267, 0x90027916, 0xa7cf0537, 0x80f22c3d, 0x877b866b, 0x9ae3ede9,
+	0x8252f323, 0x46defaba, 0x89d6dfbe, 0x7ca15ee7, 0xed7b9c41, 0xfaa2cc6a,
+	0x2a5e8d4b, 0xe9638d23, 0x434b66f9, 0x7bdf9e0a, 0x3342c2cd, 0x77bf97c1,
+	0xe51bff78, 0x1ddb42e0, 0x699b2fbf, 0x33d6f187, 0xd6f8cf3a, 0x13ddefbd,
+	0x1df00e9a, 0x1929a4f8, 0xd2a740ed, 0xa61c478e, 0x23fdbe2a, 0xcb5d4f51,
+	0x75a54f48, 0x306e742a, 0x3611cbad, 0xb7d3fce9, 0x765edc2c, 0x72de785b,
+	0xb442da2a, 0x7fa5b32f, 0xef413fd7, 0x0fea0cc7, 0x7b9e1f81, 0xa98f9d8e,
+	0x4ecf94fd, 0x7c30def4, 0xbfb9fc6f, 0xad4c77d7, 0x2a89f3a0, 0x66a77fb7,
+	0x903eb667, 0xcfebfbcb, 0x79e1f497, 0xfa7238ce, 0xbdef3117, 0xc7f74e16,
+	0x64768d72, 0x8fdb9597, 0x1df316ff, 0xa392dc68, 0x9651bf3f, 0xffe8dfc0,
+	0x335ff2a6, 0xe488ff68, 0x68f596be, 0x3f72bcfc, 0x11fc49e5, 0x3619af8e,
+	0x577d470e, 0xca04b2b1, 0x97ae4e19, 0xe50eb0d2, 0xeb9c550c, 0x689cfc66,
+	0x2b56455a, 0xe28e5f9e, 0x5a73af81, 0x1c16bf4f, 0x7dc3dfcf, 0x7c0fb4df,
+	0x29ebf20a, 0x7fee39a4, 0xfd11b6cc, 0xaa825b67, 0xbc52f486, 0x9d22578f,
+	0x3a238cf6, 0xec958fd9, 0x5c3a464f, 0xde4cf2cb, 0x2dec1f28, 0x9c627a05,
+	0xb6f14fb4, 0xa73ee8b1, 0xe340bf9c, 0x23e595f8, 0xc877e04a, 0xea5a05e7,
+	0x784fdc0c, 0x82581ec4, 0x7e113247, 0x678c165b, 0x478cde0f, 0x5debedd2,
+	0xf4e0154d, 0x3b40b5ed, 0xba360bd4, 0xfe00369b, 0x4704a6f6, 0x6b9796e2,
+	0x30e9a3bd, 0xb970a7b6, 0x68e8d828, 0xd1c8b942, 0xe61f3c3c, 0x5f22e7f4,
+	0x503dd40d, 0xdbabaf1f, 0xaf2aea62, 0xbed575cb, 0xc843e7fc, 0x5c5d75a7,
+	0xd6f006be, 0xfea0ac59, 0x7e135f3c, 0xf4398392, 0x547fddfb, 0xbe341382,
+	0xedc85d8f, 0x4a07dc98, 0x81f62ed0, 0x828b58b2, 0x1e04767a, 0xff47cb60,
+	0x6cca9bba, 0xdca54fd6, 0x59dbc8df, 0xea04607c, 0x76605353, 0x35e7d239,
+	0x18f7970b, 0x72b1efbc, 0x2d6bdd3e, 0xf985659d, 0x63d01e1e, 0x46ba5f7c,
+	0x81b669e8, 0xc15eedeb, 0x9b369fbb, 0xc9c01a42, 0x79e83876, 0x553f388d,
+	0x4c81cef3, 0xd4f8d7dc, 0xc9e22acf, 0x7b085cc5, 0x5b5f8f11, 0xb946822e,
+	0xef907fa8, 0xcc7c819b, 0x21d57c45, 0xd937505c, 0x847ce9ba, 0x87f69ee5,
+	0xff3718cc, 0x73c1e2d8, 0xbeb9d58e, 0xa2fe78fd, 0x0fae5f7a, 0xc3bde555,
+	0x9d577c79, 0xd1399ed2, 0xbacbe519, 0x21d94dce, 0xfb27f7c0, 0x95f2477f,
+	0x61ec878f, 0xa38edafc, 0xd79e105a, 0xa1a7bc75, 0x4524fd61, 0xda3e81fa,
+	0xb877bdb7, 0x9c12f92f, 0x3c6657ef, 0xa7803859, 0x14ee96d1, 0xc3c979fd,
+	0xa0fde7a0, 0x3d035fcf, 0xd021e58f, 0xf3d07eff, 0xd7f8be47, 0xaf3fd44c,
+	0xccd7c25a, 0xcd9e71c1, 0xe15e60f7, 0xc1f84f53, 0xb4ff5089, 0x3d6f8c60,
+	0x6b3e57d4, 0x7fbc600e, 0x96716d44, 0x6083ff40, 0x9dfde847, 0xc2f4eaff,
+	0xdb0583f3, 0xf471bd33, 0xd36795dd, 0x60981f99, 0x7cef997b, 0x36078f94,
+	0x976c9bfa, 0xbeb9f802, 0xea18ac55, 0x0d93795a, 0xa99365e4, 0x1f62d1fd,
+	0xc73867d5, 0xce597f81, 0xf2b187e5, 0x278e52a6, 0x71b4dfaa, 0xdfdc0cd3,
+	0x8a7bbb5b, 0xc89a7c82, 0x7d9d56de, 0xc825ce8e, 0x6b263e7a, 0xd4aaef80,
+	0xaa59e510, 0xce84fece, 0x271acc33, 0x61fe68f3, 0x1f2dc68c, 0x8edd7886,
+	0x13177c7b, 0xbdf843c9, 0x68b2f6c2, 0xabc1f51d, 0xd0be02a9, 0x41334d8d,
+	0x0cb11fbe, 0xdcf1fbb4, 0xe12b9aa0, 0x523c2167, 0xfb26aafc, 0x099a6e2f,
+	0xcd76ea41, 0x36867e38, 0x02afdfd0, 0x249abe1f, 0xfc682beb, 0x6293fd41,
+	0xf1c997fa, 0xf3c2e3eb, 0xb75a3d88, 0x2ff57f41, 0x59fbe8d6, 0x3b68ff88,
+	0xc727978d, 0x56f0475f, 0xf6f3370f, 0xbf65e850, 0x4dcfa51a, 0xeb4dfdda,
+	0xdaaf6f01, 0x97a8a57d, 0xa465affa, 0x7cff916f, 0x49b87e3a, 0xb3aad9e8,
+	0x39fd3157, 0x58b49b92, 0x90c75cba, 0xde3ee76e, 0xa2f186f9, 0x7a604ce3,
+	0x7e047ffa, 0x6fb7584d, 0xedd762bf, 0x03ce90b3, 0xc510b3f3, 0xf35b55f3,
+	0x07a75859, 0x5f1c3ffe, 0x027f0e17, 0x5ffe6638, 0x61ac6386, 0x451c28be,
+	0xfefef806, 0x80259470, 0x02b7d663, 0x959dac47, 0x67617ee5, 0x0d5452fb,
+	0xbdb63bf7, 0x8ffda564, 0x96c85374, 0xceba7687, 0x5b7a7644, 0xbf1f8b6a,
+	0xc6b6d8bf, 0x1ee1e74b, 0x722a04b1, 0xdd0a7fc1, 0xabfda269, 0x6ebe796d,
+	0xf6dadfdc, 0x26be792f, 0x8b3d349c, 0x071662ec, 0x3c5b6d81, 0x087e2aff,
+	0xa3efef8d, 0x3e3b44e9, 0xf646c532, 0xe59cdb72, 0x9f4fff94, 0xda1c7673,
+	0x0a7e9d49, 0x66531d91, 0xf3353f93, 0x3c26d6be, 0x149191ef, 0x2ffd1c78,
+	0x74dad425, 0x76bc7878, 0x58667fae, 0xbe6066db, 0xcf595b68, 0x73938c38,
+	0x803e702e, 0xfcf5abaa, 0xff311311, 0xbc1adc73, 0xbb780ad6, 0x1bf22aca,
+	0xf2852c29, 0xc9051562, 0x6dd2c2a8, 0x5a253ed0, 0xa7caf87f, 0x3191cf29,
+	0xefc860af, 0xaf814b18, 0xa67a3abc, 0xf641ffa9, 0x461afeb9, 0x8bf47179,
+	0x172feff5, 0x027d7d7e, 0xb4e81dff, 0xad1ce11c, 0xd4c21cf4, 0xd1e75d61,
+	0xd45f18d9, 0x139955ae, 0xe46d496c, 0x8bffdc77, 0xfd6207f1, 0x22de595a,
+	0x7c44acfc, 0xf3958cf1, 0x25ff80e5, 0xe40fadfc, 0xb3f0a37e, 0x1ea2ed2a,
+	0x9e506e73, 0xfe13ec17, 0xfc62df4f, 0x37d61268, 0xd769978c, 0xee7ce904,
+	0xe5deb172, 0xd1334a5b, 0x7fdca9bf, 0xd963ad23, 0xf7e4d9e6, 0x84bcdf2f,
+	0x784a7bce, 0x9bfc632a, 0x51265acf, 0x1d112bbe, 0xfd221fa4, 0x273f5899,
+	0xde157bd1, 0x41cfa256, 0x58edafcf, 0xe67cc36e, 0xf67fa8c4, 0x7367c42a,
+	0xf1a6dca0, 0xb31e09ef, 0x1b6d38f3, 0xb6005f7e, 0x399650ff, 0xb957d715,
+	0xdf092329, 0x716b623f, 0xf91e5ca2, 0x61d7d7e2, 0xf093f43f, 0x9529ccbc,
+	0x9eb2a5fb, 0xb72b9e08, 0x55e93f70, 0xe1accf31, 0xeff3661f, 0x1879c4ab,
+	0xe97ca696, 0xec7efc48, 0xa69d9879, 0xaf903bb2, 0x9d7810ef, 0x873e0b62,
+	0x155793f5, 0x5ea79f12, 0xc7cdd079, 0x79d79e37, 0xd3a5c44c, 0xee4dc3fb,
+	0xbfdf3a74, 0x71e5e1db, 0x23ac1946, 0x653e4fed, 0x7f1e8ea9, 0xaf8978b1,
+	0xc528fe31, 0x47d7a438, 0x702ce0f1, 0x0579f82f, 0x7e48df23, 0xa6f3eafb,
+	0x2dff066d, 0x69f4de7a, 0x8b4edbf2, 0xacf5e5b8, 0x724be6c8, 0x51ef6c31,
+	0xdebd4e31, 0xbf416e99, 0x16bcc030, 0xf5ae4b8c, 0x16eb9cc5, 0x11aedae7,
+	0x7dc209fa, 0xfa0daefa, 0xaf15f619, 0xcfac669f, 0x37eb1bf6, 0x7effc454,
+	0xee374d8b, 0xdf169fdb, 0xf0299e75, 0xb10ca726, 0x002d82de, 0x3bbd6eb4,
+	0xae127f54, 0x9bcb4777, 0x26fe7d6e, 0xb0584f36, 0x4f212345, 0x900eee34,
+	0xbca43c0f, 0x9d688580, 0x99dda94f, 0xc0a0b8fa, 0xb579ce22, 0x12f33aff,
+	0x36c391a0, 0x9fea266c, 0xe7eb955e, 0x4afefbbe, 0x4b18ebf1, 0xcefc0de4,
+	0x168bcc78, 0x2853ee2f, 0xff285a2f, 0x4edb3f51, 0xa5bf03df, 0xf7a1935c,
+	0x3900c7c7, 0x5d33ae10, 0x5f3c8663, 0x9bacf73a, 0xb79ec5fa, 0xe7cdfbff,
+	0xcec73c24, 0xe23e5ec7, 0x16007bc6, 0xf9d1366c, 0x5c8b85ab, 0xa4889df4,
+	0x6338ec6f, 0xe171a8b6, 0x84ff844e, 0xb8f4b774, 0x74e6ff71, 0x4af299bb,
+	0x6de3edb1, 0x5f8b37ca, 0xbaf31a17, 0x43fd9ea7, 0x57d60461, 0x375cf4ab,
+	0xd8e2972e, 0x638c36f1, 0x5f719d37, 0x3fbe7a49, 0xaefba033, 0x86fdb7a9,
+	0xe3cf20a4, 0x4903be18, 0x2f128fc4, 0xea70f089, 0x7e16edf0, 0xd164327e,
+	0x28c6d7e7, 0xef9439fe, 0xa7b2a4db, 0xd1a77ce2, 0x7be71364, 0xc44c9ed5,
+	0xfda5dd57, 0x2f2ab714, 0x41eee87e, 0x7ef160bd, 0x23e0e29f, 0xe97b0710,
+	0x3c775efb, 0xfc1bfdfb, 0xc4ec10ee, 0x8fb71b02, 0x12f3edb1, 0x17f8a39c,
+	0xecdf8741, 0x9bd1e4ff, 0xd013e087, 0x6f30733c, 0xd551ef35, 0x3aaef983,
+	0xdbc90f4b, 0x3718affc, 0x76fc7f11, 0xa0f29dbf, 0xe9da160b, 0xf9d888fe,
+	0x8f3dab4e, 0x9df30cde, 0xff707806, 0x00a8f4ff, 0x66f4f1e1, 0xf6fb83d5,
+	0x983ae7f3, 0x54ffd74d, 0x08f9f1c7, 0x0cce99df, 0xba701f3a, 0x49df3e2b,
+	0xfb655d3c, 0x45c57986, 0xf209fcc6, 0x8eadca1f, 0x3c57feff, 0xc70abbfe,
+	0xd7e1c29e, 0xd53a8b9f, 0x8f1c9f91, 0x7c1f72ae, 0x1cf18b4c, 0x2ee80955,
+	0x775eb71e, 0xb180acb0, 0x0c8eb261, 0xa075f30e, 0x7071e39e, 0xfb24f104,
+	0xdd7cd43f, 0x8577dc4c, 0x03f11fba, 0x9d5bf9ef, 0x7762bfd4, 0xe8cfc211,
+	0x059ab9bb, 0x3b01e64e, 0x74a399f4, 0x54f00db3, 0x43407dd2, 0x6475857f,
+	0xc46bfad0, 0x6ef0b1fb, 0x68fd8bb3, 0xe4ca87bb, 0xc23ffdb7, 0xba3a4a7b,
+	0x49f8bc93, 0xfc6a4756, 0x5d9a0073, 0x60039c31, 0xf1e50f6a, 0x3cf06632,
+	0x473eff6b, 0x87894db8, 0x3bfcf59f, 0x2abb18af, 0x8d46b7e5, 0x11ad7d90,
+	0xe6abf5f6, 0x5bdf10ef, 0x9ac492be, 0x355df46d, 0xbfe013e0, 0xb41717fe,
+	0x7af9403d, 0x1cac9dfe, 0xf31164a7, 0x47fae9b7, 0xe33cf26c, 0x833d7cc3,
+	0xbd70e7de, 0xfb033a02, 0x4cf6a667, 0x01bff250, 0x7e316e1f, 0xeb6b5267,
+	0xf43d4429, 0x15fc1284, 0xedfeadd2, 0x77ac4c9e, 0xfc9279bc, 0x67c573ae,
+	0x7f3205ee, 0xeed0d114, 0x3a5dddac, 0xb57f286c, 0x9d40e306, 0x7de28c7e,
+	0xf187c516, 0x5abcc3c6, 0x2c11d9cd, 0xdd143694, 0x126859b2, 0x9fdf33c7,
+	0xdf8099e8, 0x6e3ac1ef, 0x2ff5039d, 0x3fccf462, 0x69e9fc2a, 0x574cffdc,
+	0x730d41d9, 0xdd376a6c, 0x214bf0bf, 0xf214bf2f, 0x33dff1d2, 0x74fa46ae,
+	0xdcf269bc, 0x97e3ab93, 0xc1b21624, 0x7b5bd77b, 0x8efe827e, 0xebf6cbac,
+	0x869d2fc1, 0x72e36afc, 0x66b5e523, 0x7d397c51, 0xfe26e6d2, 0xc476f61c,
+	0xc7ce798d, 0xff23e951, 0x7f271fb7, 0xccd3a3a1, 0x61cd93f0, 0xa0e6828b,
+	0xc1ff8a33, 0x2a3d70e3, 0x96a21de5, 0xde425c0b, 0xe57bdf91, 0x951e59f7,
+	0xc2b5c7fc, 0x3c65ec7f, 0xfb11a87e, 0xf02a7fac, 0x1afe39d3, 0x5c600b3b,
+	0x3f8f27f7, 0x78fc8539, 0xdf2d9fa3, 0xbdf67e93, 0xa55e39eb, 0x38a362de,
+	0xf266647f, 0xaf296bfd, 0xf687ab2c, 0x207f24b1, 0x03d3c4ff, 0xd47a0a97,
+	0xe819359c, 0x46a7a51a, 0xdfcf94f4, 0xd2127b07, 0xfa5cc39d, 0x377d0171,
+	0x51f31b73, 0x66b6e33e, 0xe4e90b07, 0x3e51e3b6, 0xc63dc76f, 0x798967f9,
+	0x3e38925b, 0xf877c43b, 0x91dfaa1d, 0x37d5145a, 0xae075f80, 0xabcbfd4f,
+	0x79498f14, 0x95dce390, 0x7ee1f7c4, 0xadf357f1, 0x2bb67db2, 0x76a5deba,
+	0x2f24b7d3, 0x27de437e, 0x24999f8a, 0xc5e58947, 0x92f215f3, 0x4d05f84e,
+	0xf97b46ea, 0x9f1ff09c, 0x68aa6c3b, 0xdb7fc38b, 0xe73d7c6d, 0xec01a867,
+	0xdb844177, 0x32fd44b3, 0x15f9c553, 0x3c589781, 0x3fc7e7a4, 0xd056b2ff,
+	0x7f8ca6a5, 0x22f78ccf, 0xca0c3e60, 0xd78c1f6c, 0xd10ce65b, 0xef22cbf3,
+	0xcd493ee2, 0xca2f56f3, 0xf6f3ef6f, 0xf49bcc2a, 0x6aad6f3e, 0xc7a60e73,
+	0x1e8d6fe6, 0xebfd6f7f, 0xf42bf49f, 0xbecd3b63, 0xf3c1de31, 0x845dd927,
+	0x93bd019f, 0x68d83fc0, 0x3b3c7273, 0x792d9e38, 0x5ba78f5c, 0xb1ff788d,
+	0x33748790, 0xd7a0b644, 0xfe7905a6, 0x90bb39e9, 0x5b87900f, 0x4eec2f90,
+	0x2c7efe03, 0xafa969fc, 0xc6575f1e, 0x7ee1a78e, 0x6ff6886c, 0xc91f7e3b,
+	0x4b7a79c6, 0xb56fdfe3, 0x6309f05b, 0xea6eff46, 0xfb42af4c, 0x678aede8,
+	0x9ff68b0c, 0xe7bfceda, 0x061ddb35, 0xf02cdbbf, 0x7638e521, 0x3cc40f9e,
+	0x90cfa81d, 0xc31ece2e, 0x03e4bb3e, 0x7ae4d7a9, 0xfd911da4, 0x67f82f7e,
+	0x8ccfe0ba, 0xfc1f3e26, 0xca9ac2b3, 0xdeff01ed, 0x7841eb9d, 0x9da2b21e,
+	0x78f34160, 0xaddd67a0, 0x16f7d18c, 0x33fc1f62, 0xbb2df047, 0x07e81d6e,
+	0x7e16fe92, 0xd738875f, 0xd407fd28, 0x7a9d134f, 0x0e1c43b8, 0x55f5c2b3,
+	0x70927e8c, 0x7cfc3ac0, 0x138d171c, 0x8edc6fb5, 0xb58f42a9, 0x6322ef8a,
+	0xfe7fe6e5, 0x569b911f, 0xe9025e71, 0xbcdf4f10, 0xc96eec25, 0x0562fedb,
+	0xc04c6bf8, 0x624fc3e9, 0x1ce7ba4a, 0xb0bd40b0, 0x37cb414b, 0x82855720,
+	0x45bc70df, 0x18f8a3e3, 0xf032e49d, 0x3e2a3fdf, 0x0366df0a, 0xa3a2dff0,
+	0x2e2494e7, 0xf32a19c7, 0xc3fedf13, 0x80381fc2, 0x00d4efdf, 0xdf34505d,
+	0x969bf01a, 0xbf4763dc, 0x43e66de7, 0xfb74e9bf, 0x8c78d312, 0xbf70b75f,
+	0x6e2f4065, 0xbd07ebe6, 0x6d7a8ed6, 0x477edbcc, 0x2167d411, 0x5bce30fd,
+	0xd379f8c4, 0xe3185776, 0x198f2925, 0xe0667b73, 0xa39d473e, 0xf75dbd61,
+	0x277fa07c, 0xca8dcff4, 0x35ca0d75, 0x82b53bcc, 0x2b4eb3b7, 0x8b160fdd,
+	0x47bec62f, 0xeff226f3, 0x7ae24c6e, 0xd234f50d, 0x6ec21ca9, 0x4afdc03b,
+	0x798cc86e, 0xe4c657d4, 0x957dc44e, 0xa2c14943, 0xa67ef035, 0x29983738,
+	0x26669bce, 0x126b72f1, 0x0e99cbd7, 0x5c1d065e, 0x79ba67bf, 0xef79d8ff,
+	0x54defd12, 0xed7f064c, 0x7c3c8204, 0x38af9f30, 0x327e88cb, 0xed42bf4f,
+	0x8002cdff, 0x9997456f, 0x4d3b07f2, 0xea24fac1, 0xfb1d12a7, 0x0f8b3d4c,
+	0x1ef10123, 0xf3128ba5, 0xb167d493, 0xa3ea34fe, 0xfcf20d9b, 0x3be3825d,
+	0x9190afb8, 0xfafcbfd8, 0x027e7d7c, 0xb7c26b7f, 0xfa496fe2, 0x8c80bccb,
+	0x307f25bf, 0x6bad20f6, 0xbfce4cea, 0x6177afa0, 0xbafa0afc, 0xebc91a68,
+	0x6bbc780f, 0x719178e3, 0x4f66667c, 0xc0b7c61d, 0xbc9ef487, 0xe4f729f7,
+	0xe41e749b, 0xf537248d, 0xb7f0e5db, 0x5bc9c526, 0x6869a88d, 0x8487fba2,
+	0x42f795dd, 0xafdf6a8d, 0x19be3aa8, 0xfee1c555, 0x60c59269, 0x7d443a17,
+	0xd3971277, 0xa5115551, 0xd88dfa00, 0x015143ba, 0xfced119e, 0x0688f77c,
+	0x0c9ff746, 0xf0e61bcc, 0x95870d8e, 0x70a28adb, 0x17669b37, 0xe69e0079,
+	0x09af874c, 0xe0ce699e, 0xcd77080b, 0x039d08b1, 0x5bac7a9d, 0xd32978c5,
+	0xf205ffe9, 0x6bd7849b, 0xbcc08cf6, 0x8718e3bd, 0x9fc065b3, 0x79e6c9ef,
+	0x3c0aefc9, 0xc9fb09b7, 0x4efdd30e, 0x336f6e4e, 0xa6db19ef, 0xfd26673f,
+	0xccd7e858, 0x72e3ee28, 0x614f799e, 0xb8ba03db, 0xb03f3cb1, 0x0f979f03,
+	0xbf11371b, 0x307fcfb7, 0xc73ee2a2, 0x154d5b0b, 0xf6fc38e1, 0xa75394b6,
+	0x2eb8d7ff, 0xc957ff09, 0x4967ec87, 0x1def9a1a, 0xb7060f82, 0xa5ed443f,
+	0x84dfa64f, 0xd06cd6f0, 0x48f16e18, 0x033a67bc, 0x239f1027, 0xc4155bda,
+	0x9f8e63c9, 0xae0e7828, 0xce798aba, 0xaef6d203, 0x7bc1cb6b, 0x7cee6a97,
+	0x79ef072e, 0x38c0f092, 0x98d6ef84, 0x0fd28b73, 0x67e11ff9, 0x67033b94,
+	0xcebafc70, 0xb7bfb44d, 0x5fbf0772, 0xcfbae127, 0x9eb02f11, 0xd5bfc8f5,
+	0x5478bfc0, 0x8dfb0147, 0x6859acf4, 0x6a26efa3, 0x63fe7176, 0x18f0b41b,
+	0x8eefb7ac, 0xe47da098, 0xa61f7edd, 0xed77abef, 0xf0e99656, 0x4c26b235,
+	0xb9e77bfb, 0x97227ee4, 0x4bed11f7, 0xb38f886a, 0x63e43e1c, 0x71d92fcc,
+	0xfdcecf3f, 0x1e52deaa, 0x425af49f, 0xafd0bd63, 0x87c93737, 0xbabd3bdf,
+	0xdefa0afb, 0x35bcf5d0, 0x7ca3b8a2, 0x6f7ac0f3, 0x40f9c229, 0xdf20f339,
+	0x835fde80, 0x2726f2f5, 0x1a9fbe43, 0x77a9c743, 0x3bf6be73, 0xc31ef2dc,
+	0x63d7c889, 0x81aedc1d, 0xeebf205d, 0xf3c38590, 0xdd74e33c, 0x75c8149a,
+	0x87f7e064, 0x54f18611, 0x86bb454a, 0x387608f9, 0xbd5bf095, 0x37d683b5,
+	0x7bf71772, 0x92b577aa, 0x82fd4a3c, 0x7b7bc0f8, 0x457be2a7, 0x47040fc4,
+	0x943b93e9, 0x927fd0e3, 0xbcf3ccc3, 0xc7ec775b, 0xf019887b, 0xd02fc60b,
+	0xce4affbe, 0xe96e2a3f, 0xce3dc80d, 0x59e7c453, 0x097b93a6, 0xbff91c77,
+	0x1e77e118, 0xe5c83eeb, 0x6fefc23f, 0xe921fc12, 0xf5c1c330, 0x21c7a48f,
+	0xe300e80d, 0xa24bca1c, 0xa23ca471, 0xfa11ccfc, 0xc68ce4c9, 0x090e1249,
+	0xd9472ee7, 0xb8298e15, 0x775a6eee, 0xe09556ed, 0x0502d72e, 0x3b7c75ab,
+	0xfbf9c1cf, 0xb57782c0, 0xf9be781e, 0x9f0947e7, 0x3f0a2db5, 0x5472dac7,
+	0x3e7a879b, 0xe319acf4, 0xff19ab9d, 0x51cf747b, 0x7757cfd2, 0x76d7bc70,
+	0x736f56ec, 0xf728f946, 0x74c581e4, 0x95bb751f, 0xe9923fa1, 0x788cde0c,
+	0x024fc418, 0x25cf520f, 0xf4871392, 0xf91a4253, 0xe421e029, 0xf520dc62,
+	0xdf1156e4, 0xfe7f2730, 0x515d624f, 0x41ac10fd, 0xf927947c, 0xdf50d80e,
+	0x3ce8544d, 0xe933b487, 0xf51e3833, 0x37bdc4f0, 0xd287e64f, 0xb25f8c98,
+	0x63bee8e7, 0xe2b7d018, 0x5d04bbfd, 0x3f8479e1, 0x37687967, 0x74b1dc25,
+	0x6bf081fb, 0x1351f0e2, 0x476b1fc2, 0xfac05f24, 0xf73db19a, 0x41ec89a4,
+	0x86d6c86d, 0x98af26fb, 0x75df90e6, 0xf09b8160, 0xdf90b357, 0x67f9bc63,
+	0x84d4af5c, 0x777f749c, 0x256e704d, 0xdba543ed, 0xc50fb87d, 0x79b50782,
+	0xf774c1ae, 0xce9fbf3e, 0xcedd5c31, 0x8f4060e5, 0x4bfa73b6, 0x1fd72efd,
+	0xcfc7375d, 0xec858d2c, 0x6dbc51a9, 0x7d08f24d, 0xdfe5cc10, 0x3ccd7fa3,
+	0x7b1666f7, 0x88b3d29f, 0xcf5b5a8f, 0x2faadc21, 0xd35fbb79, 0xf2c727ef,
+	0xd7baf557, 0xac2ed10a, 0xa54c3aca, 0xb7a0ff3e, 0x7d057b57, 0x8be5149e,
+	0xa7985f48, 0xc93b727a, 0x0d07f54d, 0x0ace6f24, 0xc79c67ae, 0xd6307a8c,
+	0xb4772a5f, 0xdefdc2cf, 0x9e2a1ff2, 0x5ed89626, 0xde7850ac, 0x2fada3f3,
+	0xc19cf1bb, 0xe26bbdc0, 0x0e1bf03c, 0x3bbdffce, 0x97c4e101, 0xfe50da0d,
+	0xc78d57e2, 0x74296f22, 0x7dc0808f, 0x1719aeef, 0xd2e7887d, 0x7e532677,
+	0x375dd3e7, 0xc021f9dc, 0x72e3bbdd, 0x4de943b2, 0xfe9fe201, 0xf70fbb0c,
+	0xa97e20df, 0xebee8050, 0xe3765fdb, 0x3f27e4fe, 0xda2270ba, 0x9c050fb7,
+	0x5e667f27, 0xe7b97887, 0x865e9f99, 0x2561d5ed, 0xef6cf50d, 0x15fe1fce,
+	0x8ed570f5, 0x9a9f8a7a, 0x0d4bf47e, 0xe14d39f7, 0x677cfe56, 0x7dfb07b2,
+	0x65d37db2, 0xf6a6ef88, 0xb93f6a26, 0x07ea95bd, 0x4be9f99e, 0x33f68dff,
+	0xbf61f6e0, 0x4fdb3f93, 0xaa09a974, 0x04f0defc, 0xeee8fd7c, 0xa19f5cfd,
+	0xb07287bf, 0xc919eb90, 0x3f7f92d7, 0x8bd6e679, 0xc6bbfaf4, 0x60a5181c,
+	0x1f5661ff, 0x7d07d79e, 0x15fbfa4e, 0xc4ed533b, 0xce9cadf7, 0xaf74d5e6,
+	0xd73e7943, 0xdc61883b, 0x72febaf7, 0x91477d07, 0x4afb10e1, 0x293f1176,
+	0xbcc3ed0a, 0x6d05e972, 0x217dba4d, 0x9457bf49, 0x1756a3d9, 0x8b379f45,
+	0xb8939b79, 0xf514f93d, 0xd539c430, 0xac1efc65, 0x5f6c7f4f, 0xd6b5fe48,
+	0xd76bf743, 0x34be7114, 0x9a3d393b, 0xb7e57e45, 0x39630ee0, 0xcf02e41e,
+	0x8eeb40d5, 0xfadf292f, 0x1a6fd913, 0xbed357c2, 0x784752da, 0x0e7f11f9,
+	0x8ee7c658, 0x354b06a2, 0x1b592f3c, 0xe91c5e3f, 0x3dfd3274, 0x663a2376,
+	0xde92f28f, 0xe8c6fab1, 0x5511d2fc, 0x27dc26c1, 0x906d6ab8, 0xe17ea9ee,
+	0x5ee937e3, 0x2ac667e3, 0x30df8893, 0x2f76847f, 0x6a79a4e4, 0x70078f31,
+	0x7114ef9e, 0x17a987a3, 0x32aff1eb, 0x75fb49d5, 0x6a97a4fd, 0xdc5d859b,
+	0x5b19e221, 0x6bde367f, 0x2f39ef0a, 0xc09bedd6, 0xd5274e7b, 0x8fe7b3fb,
+	0x5a143b47, 0xabc87648, 0xa527bc11, 0x09a161fe, 0xa1e2e8b7, 0xf51738d8,
+	0xca77e61b, 0xe77ea59d, 0xe3296894, 0xdd3c58df, 0x3877da25, 0xea277efd,
+	0x16e79dc7, 0x7a09d687, 0x2cf4aa3f, 0xed0e3e47, 0x2f907bb3, 0x29eb2eeb,
+	0x3bb1ebd4, 0x0b4e9e04, 0x41dfc13e, 0x64cdb8a0, 0xb5df110a, 0xa4dad05d,
+	0xe77bb60c, 0x44757f09, 0xba41e2a7, 0xc73ad88c, 0x9bf03f40, 0x7f686bd6,
+	0xe4252ac6, 0x7eae76cf, 0x11e61c6d, 0xaf586e81, 0x495d98eb, 0x833b03f8,
+	0xe75d6cfe, 0x73840eec, 0x3eb5fa64, 0x3f0797dc, 0xe77d1664, 0xa0a6c351,
+	0xda3a3efc, 0xad246fbf, 0x1e786143, 0x384b51f3, 0xa3f7ed0b, 0x787d0ae3,
+	0x582ea63e, 0x0af24359, 0x2f7d2d7e, 0xfe40dc15, 0x26f79d6b, 0xc6a7ed1f,
+	0x853ef9ca, 0x1be054fc, 0xab8b1dfa, 0x7c2cbbf1, 0x7920ec3f, 0x2dccebaf,
+	0x9eaef845, 0x9e60f3e9, 0x8fc494ed, 0xf4dd2d2c, 0xf7a7183b, 0x27485958,
+	0x03eeaf97, 0xf80266bf, 0xafc17b35, 0x3fdf8099, 0x1571d9fa, 0x0d7a71c6,
+	0x5deab779, 0xcdef74b9, 0x7f8ed129, 0x2d7ddd5e, 0xe21195b5, 0xcfe11acb,
+	0xc5fd5ba6, 0x26e8ebb7, 0xfa5894fd, 0xf6e4ee3f, 0xe727bce8, 0x00463d12,
+	0x239f103d, 0xa26e9ab6, 0x863485f5, 0x8aaabbeb, 0xbe01ea45, 0x229ab6ab,
+	0xfa884f97, 0x3de3b131, 0x4f7114e9, 0xf518a05a, 0xa702f95b, 0xca87af4f,
+	0xbb26fc93, 0x2c628fce, 0x92e31930, 0x845d338f, 0xaf4e1df7, 0x7e0c87f4,
+	0xd3b8e0cf, 0xe1f8c3ee, 0x04d0969a, 0xaf0e87dc, 0xf2885a4f, 0x49d47a08,
+	0x781eebfb, 0xb6d6e31f, 0xb8f3b64b, 0xf9d689b5, 0xc247a77d, 0xe301b937,
+	0x5ac0fcdc, 0xbe5c7f6e, 0xe5d6fb78, 0x2ee7e94e, 0xbef3c62e, 0x3e9d7375,
+	0xff36af8c, 0xe38a5942, 0x746c85d7, 0x705d6e39, 0xf2579bed, 0x5fdf8038,
+	0xbaa2fc23, 0x347d1bf4, 0x7a7e20ff, 0x4e8ddb8f, 0x4ebaaf4e, 0x1c642f6e,
+	0x5e217a8b, 0x29ca5ede, 0x6b8c63ea, 0x39497f24, 0x3c79573d, 0xfc90096d,
+	0x2375869c, 0xe1f00e3f, 0x630f9c0b, 0xff8934e3, 0x8bc2f8d8, 0xef7506c2,
+	0x8f7d0776, 0xe69af52f, 0x3d451c0c, 0x6476d754, 0xd0abb123, 0xdc2f7543,
+	0x3f94385f, 0x78f95365, 0x3d2b9036, 0x1359e0df, 0x432a44d7, 0xc0df3a4e,
+	0xc6df5fb8, 0x1d8e8fef, 0x4f42b7ef, 0xbf4a7c49, 0xcf82f319, 0xc29ff943,
+	0xfa227d7e, 0xcf29d5fd, 0xfc0d8457, 0xdf0535e7, 0x44f17dc3, 0xf7be56fa,
+	0x7226f34d, 0x63efa97c, 0x99f325e9, 0x45d98eb8, 0xecb5fb89, 0x0c48f117,
+	0x3e0b14ff, 0x80e7037e, 0x07724f75, 0xbea0b917, 0xd3f99928, 0x7e8bdf00,
+	0x29f7e569, 0xc42fed8c, 0x655aaa78, 0x5797cc21, 0xf584b4ef, 0xa1f736e7,
+	0xf5ef040f, 0x33216cb9, 0x3bb95df0, 0x0671c1e8, 0xcbbd87ee, 0xc1e1e74e,
+	0x7282994d, 0x481ed23b, 0xf7e06575, 0x4f743d89, 0x1e0e63ec, 0x311ff44f,
+	0xc5cbcf87, 0xe87bbe7b, 0xacce84f7, 0xea7ca7cd, 0x61fc2f3b, 0xede01fcf,
+	0x1a3e7b53, 0xd493cfad, 0x55364cd9, 0xad3c6f97, 0x2b67dfc4, 0x167d7176,
+	0xf3f65bcd, 0x438be5bc, 0x63eb41e7, 0xf3d1fbf2, 0xfdf9b1b3, 0x6fa878e8,
+	0xfbf263eb, 0x6f5c5d07, 0xf9e7df47, 0xf14c77fd, 0xc7bd6fbf, 0x8bafdc65,
+	0xc5cfbf6d, 0x8f23abf9, 0xe3ff45cb, 0xe4fd184a, 0x115d6be5, 0x396b8fbc,
+	0x63bf463c, 0x311427b9, 0xf9ae51b9, 0xcdfc323e, 0xbe5f81b5, 0xe807d844,
+	0xbe474ea7, 0xff68790c, 0xd0f42831, 0xff57ce92, 0x72f4f18c, 0xc3a6d995,
+	0xbe6d4b38, 0x13fbf843, 0xba673a1f, 0x7e56ed51, 0x744c86e7, 0x6a2d86e6,
+	0xf455bca0, 0xf374f57c, 0xfabe72df, 0x57302bb0, 0x7fd21bf2, 0x5f5cedd1,
+	0x9756fec8, 0xe467373c, 0xd01377b7, 0x665dc739, 0x3ababe46, 0xc698b029,
+	0x4194fdbd, 0x7a569caf, 0xf201fe9e, 0x7bfe6ae3, 0xd7efb147, 0x8afaf98a,
+	0xc53ee9bb, 0xc97fd5be, 0xebaebf26, 0xff51b7d5, 0x817bcad5, 0x7b862c3b,
+	0xfc932b81, 0xf7efc4d5, 0xfba14f4d, 0x73c186ce, 0x1ab82782, 0xf6fe41e1,
+	0xdbf9eba7, 0x2ce78c9b, 0xba01b5ea, 0x921ef7d0, 0xcb946c2e, 0x23ae8f48,
+	0x36a58f44, 0x701bdd3f, 0xf0403f9b, 0x4c8e4a2d, 0x5ff018f1, 0xf7dd37f4,
+	0xf6e42958, 0xfdfe2d1d, 0x3e09f7e2, 0x8dfb283c, 0x4efafce0, 0x1e7aab36,
+	0xff234ccb, 0x2ce42c3e, 0x71cbde2c, 0xbc421ffb, 0x18795e80, 0x8e7400f1,
+	0xcaf5e8a4, 0xe7ba7ee4, 0x374a6fa2, 0xae7d1f7d, 0x80ac66f8, 0x8e784bd6,
+	0xd1385e73, 0xf0b56f7a, 0xb5df0075, 0x47af48b9, 0x87bfebc6, 0x5b80b9ff,
+	0x493be5ca, 0x142cbc46, 0xe091fbeb, 0x87974a93, 0xfd7c1be7, 0x3bce38d9,
+	0xb8f1f1a9, 0x768f8a28, 0x7fb583c2, 0x6fd01a71, 0x6d35d83e, 0x12a7e122,
+	0x938a0f9f, 0x49ff228a, 0x5c6b6dfb, 0x0bc5ed0c, 0xfdc9be72, 0x963f8f3c,
+	0x885a0bcf, 0x57d96eff, 0xc2bd6de0, 0x3145e77f, 0x2b6844c3, 0xdf2f32c7,
+	0x7fce980f, 0x203ecaaf, 0x195c8797, 0xcec4fc51, 0xc09b0c7b, 0xc4bb779e,
+	0xe63e40f7, 0x1f16f7c3, 0x8b75ccfd, 0xf43df0d7, 0xef4e385b, 0x1fc8479d,
+	0x0a45a63b, 0xda719c5b, 0xc7b6c8b5, 0x6b3b6894, 0xadc17ee0, 0xbbe99acf,
+	0x9ea66398, 0x2d7f7047, 0xfc99a6f4, 0x3dde3fa8, 0xfa2bbe83, 0xffa30ef4,
+	0x45d0527a, 0x20309f7e, 0x830b783f, 0xc5eff079, 0xcb9ce6dc, 0x3ee629e5,
+	0x14164af6, 0xfb13fbbf, 0xe103427d, 0x893df079, 0xe73c87eb, 0x7ce6e995,
+	0xf38269f3, 0xebfb8a19, 0x3ee327da, 0xc4bed45d, 0xf1f1d3ee, 0x474e6edc,
+	0xf6fea769, 0x304aaef9, 0xe77175df, 0xba7cf393, 0xb3bafe3c, 0xd58edcdc,
+	0x461becc7, 0xb3e795c7, 0xa49fa85a, 0x2cd9fd77, 0xd2eb2e7a, 0x37be8a8f,
+	0xfd8bc68a, 0x13e6472e, 0xa7beff25, 0x9efd1b30, 0xf224db11, 0x78099e86,
+	0x687bd0bf, 0x06df412f, 0xf52f60af, 0xfe4b8800, 0x9aa17879, 0xd8efe517,
+	0x9d5f0075, 0xd51effd4, 0x4fb8b704, 0x217642ef, 0xf821177d, 0xe59df179,
+	0xf25aee74, 0x854f546f, 0x71adbc78, 0xa7319e3f, 0x18b3ff41, 0x13a4b0fb,
+	0xb204c1ec, 0xfc0e74cb, 0x12378bf8, 0x5f32571f, 0x17d1e926, 0xe93f74ad,
+	0x120bb5d7, 0x4b421b9e, 0x8adefc2c, 0x6bc7ba74, 0x86ff988e, 0x981fee2e,
+	0xfbfe7e6c, 0xf5e67e55, 0x03dff264, 0xca05fb21, 0xa5fa855d, 0x00cc8eb2,
+	0x3162e9f5, 0xf00fbe0a, 0xb64a2a5f, 0xf2859834, 0xb7ffef26, 0x961670b3,
+	0xbcf9174d, 0x04ae5e58, 0xa7070e54, 0x30d62cf3, 0x6c9bfb0a, 0xff9c663d,
+	0x58da7991, 0x4fa01fbc, 0xdf89f249, 0x74fba3fb, 0xf3f303ad, 0x9dd64bba,
+	0x0eefc32e, 0x7dc5bbd4, 0xa366360f, 0x4673ebdf, 0x057da13e, 0x616cda51,
+	0x0701ddfa, 0xcdba448b, 0x8c00fbf8, 0x614f9bbe, 0x3dfc112d, 0xff17be8c,
+	0x1e1c518b, 0x76fc0fbd, 0xef2194f7, 0xc7b6370b, 0x2a9e17dc, 0x7e634ae0,
+	0xad221fba, 0x635f78c2, 0x42ad7386, 0x3016e13f, 0xc444cf3e, 0x0b61ff43,
+	0x6f0bef63, 0xe1fc7051, 0x1b73e341, 0x5f7a463b, 0xbeb20489, 0x93de8b4c,
+	0xdf8920b4, 0x22eff44b, 0x115b9c8e, 0x6f015728, 0xafd88bf3, 0xe8de50fc,
+	0xe1177e09, 0x4370f1bf, 0x17d40d6d, 0xe3c6cebe, 0xc9c530f7, 0x185dea9e,
+	0xefcadd7f, 0x3efe2d6b, 0xec361d67, 0xf0c78a3c, 0xc9b51707, 0x2b2ace78,
+	0xb17d718c, 0xd48fb443, 0x7f18bda8, 0x3fdf1100, 0x95f74bcf, 0x74f38bd4,
+	0x9a7987f8, 0xc333f3d6, 0x67d30768, 0x7ac46ff4, 0xee8cfdff, 0x19f79059,
+	0x408ec76d, 0xbf4afa9f, 0x87da0325, 0xe23e64cf, 0x03bed149, 0xebcf9dff,
+	0xdf889797, 0x5d47c08d, 0x2733ca9b, 0xc6fa9b13, 0x5cfcc1b0, 0x18bb35df,
+	0x73f954f3, 0x297f13a7, 0xba0a2f82, 0x7c51ff14, 0x3f68c22f, 0x972bbeb5,
+	0x1c17dfce, 0x719e51f6, 0xd4e1a981, 0x1bdf5c1e, 0x13effe3c, 0xc84cc6fb,
+	0x7918d4f3, 0xafca24b9, 0xe72c5ed4, 0x7accfb97, 0x3ce046be, 0xfc44a0f0,
+	0xc7bd12c4, 0x1662ece5, 0x6671c3fd, 0xbd25c744, 0xbdb45332, 0x3fdf0b72,
+	0xf646aa9b, 0x225e1ca3, 0x4e5162ca, 0xe2330166, 0x7f73c4ab, 0x5177ce13,
+	0xf21a68e0, 0xaf9b74fd, 0x9f46af19, 0xb86150bb, 0x4e9cc4af, 0xffb17ebe,
+	0x3d8af52d, 0xfb67e799, 0xddbab3dc, 0xdf997dba, 0x30df85ba, 0xb98d8d96,
+	0x24f7ff67, 0x3bffa517, 0xdb18dc91, 0x4066e463, 0x7d3f313f, 0x7dfb18b7,
+	0xff3f3ce1, 0xc05e6a6e, 0x7e3e7cf8, 0xe71f97d7, 0x15d0ec77, 0x0b8e8ccb,
+	0x78416472, 0x6ec77f62, 0x2ab7f48e, 0xa2d6fe8c, 0x8522fa7f, 0xa7b0c0ec,
+	0xc1ccbf6b, 0xe25ce67c, 0xb12ebacb, 0x43fc0f3f, 0xfd89e3df, 0xe34efaf9,
+	0xd77d7cfe, 0x74337b92, 0x9d75f3fb, 0xcf3fb55f, 0x5bbfeeed, 0xd30f74c2,
+	0xd604afbf, 0x165e7982, 0x7107c2fb, 0x9462783f, 0x3cc5aeee, 0xe35073af,
+	0x3107f4fb, 0x4bcba37f, 0xfcf070cc, 0x340b8ce8, 0xe5573e7a, 0x2129f2fb,
+	0x0b92f927, 0x665defc4, 0x7e41d998, 0x1fe7847e, 0xeb3e22c2, 0xb40d7e16,
+	0x47fca06f, 0xa465dc8c, 0x6547ca31, 0xf2963cdc, 0x8abb0b3a, 0x0d3e6327,
+	0x7fe533f4, 0x7a5eddbb, 0x98b0617f, 0x395723f2, 0x8af3a46f, 0x6aad5df5,
+	0x7adf4ea5, 0xf5f78d93, 0x9a5df9c4, 0xfa29e482, 0xe019a5df, 0x489e7a27,
+	0xa7dfcb0f, 0x9bbe62ac, 0x7de2e597, 0x9d774001, 0x3194b79d, 0x53fb4f7e,
+	0x5d2b9d0a, 0x76483ce5, 0x630ed578, 0xd45b943f, 0x6bafd57b, 0xcd537f74,
+	0x467c41ae, 0x5de52d72, 0xa1b517de, 0x3ee0afd4, 0xe611df4f, 0x77fa436b,
+	0xc05b8cc4, 0xe04f98ed, 0xa5e858b3, 0x3666f349, 0x24e53ee1, 0x565f9713,
+	0x7fc91be9, 0xe7917377, 0xc7551d84, 0xfe50cad3, 0x63eb9f40, 0x067dfcb3,
+	0xc4531fd4, 0x717ab1f8, 0x2e03f08a, 0x137d6153, 0x0a7d76a3, 0xa683f5e3,
+	0xbddd332e, 0x3c9ce02f, 0x9fdc4720, 0xd24a7a90, 0xc6e5ef03, 0x92cb5df3,
+	0x63c87e85, 0x184f4879, 0xc157f89f, 0x8d2a73cc, 0xbca1521f, 0xf407cc44,
+	0x84df210e, 0x9dfa364c, 0x034a8d8f, 0xfa1b79e9, 0xa03e508d, 0xfd7f4904,
+	0x46642abc, 0x1ac45ebc, 0x1726b86e, 0x03b758f5, 0xb07c05e9, 0x4d1fcf52,
+	0x5bbd922c, 0xb509f031, 0xe8c4a6f7, 0x3d12a61d, 0xc238304c, 0xa5ad7669,
+	0xc42bf47d, 0x6e53a273, 0xc9d51fb4, 0x3fb4618f, 0x87c01fea, 0xf52fddb1,
+	0x93ea3309, 0x8f05f65c, 0xf43feed3, 0xfb0f8c1d, 0x8c5fc22c, 0xa1dd4ba1,
+	0x641f8b93, 0xbe49fa51, 0x983598ba, 0xca1f00b3, 0xf75e792f, 0x2a5ba462,
+	0xfc8c51fa, 0x892c69a7, 0xe64d8be4, 0x57c907f9, 0xa2255e92, 0x4fa484f6,
+	0x7841cb0e, 0x53b4ed2f, 0xa462e9de, 0x58bdd013, 0x4aed2f8c, 0x6af4e4c4,
+	0xb0174879, 0x7ec1e61b, 0x97aa665e, 0xe4c933e8, 0x614cfa27, 0xdba434be,
+	0xd7ddf6cf, 0xe8eb7084, 0xe3192db0, 0x80e80545, 0x3ed8357c, 0xe463db18,
+	0x94bbe259, 0x2c9fc979, 0x9c394665, 0x670d04e1, 0x8dcafba2, 0xa0b363f7,
+	0x7b94f08a, 0x6f8fdd06, 0x41b1d232, 0xc8ae7845, 0x444bf632, 0xa7a41ccf,
+	0x89690fb4, 0x6ae529e9, 0xa9d39db7, 0xfa84e81e, 0xd2afd266, 0x34813867,
+	0x25ccade7, 0xc53bb0fb, 0xfaf584db, 0xc5f4851a, 0xcbdf6ee4, 0x106fcf58,
+	0x337eb9ed, 0xafd683ee, 0xde53ede5, 0x7ac0cc0a, 0x5ede3ac1, 0xc9b04f7c,
+	0xaddfe385, 0x2fd8bdd0, 0xcbc7727c, 0x47be1ed8, 0xcfca5fbc, 0xe06fd6e1,
+	0xf9847fef, 0x3cf10bde, 0x7c0ccb44, 0x9739ae3e, 0xf0b00eff, 0xfc172e67,
+	0xe3695bb2, 0xc83c20f7, 0xc8715dc7, 0xce28c496, 0x65cafee1, 0x6f979d62,
+	0x3f4cebe5, 0x1785557e, 0x37b2a9f8, 0x2d95d740, 0x1109c164, 0x3372ef58,
+	0xf1f9f204, 0x10bc1be4, 0x0d288f28, 0x500aa744, 0x72fbf49d, 0xc0b67cd3,
+	0xc7ef9e78, 0x92fa79ef, 0xf329bc79, 0x35ffb8e5, 0xfb4f675b, 0x469e7992,
+	0x654ccbf6, 0x09ef3f7c, 0x91a3f917, 0x7c8298c7, 0xfec0b88f, 0x57faf289,
+	0xbd9037e4, 0xe64722ba, 0xa5d47bf3, 0x97f15f30, 0x178acd9f, 0xcc9e7f07,
+	0x62c7e4ed, 0xaf3e4f1f, 0xbdfdf30a, 0xf4cbc8dd, 0x30e9ad7d, 0xe19853d5,
+	0xb9ddab1e, 0xfe7aaebc, 0x28f35f7a, 0x06b35dfc, 0x36ca8435, 0xcc7da15b,
+	0x679fe0eb, 0xffe14627, 0x2830d93f, 0x00800092, 0x00000000, 0x00088b1f,
+	0x00000000, 0x7dedff00, 0x45947c09, 0xf37f78b2, 0x093215cd, 0x87213b93,
+	0x98884013, 0x861c2184, 0x4109264b, 0xe8098414, 0x720d7282, 0xeb22dc85,
+	0x97f75763, 0xd9110441, 0x73d6f8dd, 0x0160763d, 0x18896151, 0xc3824830,
+	0x12a20882, 0x75040411, 0x0844ae22, 0xf1e20c49, 0xabaf2e1e, 0xbe667bba,
+	0xfc38666f, 0xddbf7ffb, 0xdb2e23f7, 0xaaefafa9, 0xeaeaeaea, 0x084c8eaa,
+	0x908238b9, 0xadc4b45b, 0x9680a1cf, 0xc8401bfe, 0xd5fa25dc, 0x3f02242b,
+	0x213c6376, 0x33fe1277, 0x192d7aec, 0xf01dc844, 0x289085bb, 0x2afda4b3,
+	0x9fdefe83, 0xd328bff4, 0xf3bfcf72, 0xc84d94a3, 0x3a558caf, 0xd50a1dd2,
+	0x8459ece8, 0x9c9b359c, 0x7c84be9a, 0xcce2392e, 0x7d690903, 0x965cff76,
+	0x64beceef, 0x47e696be, 0xb048d4d0, 0xefde62df, 0x13d2e27c, 0x977cdfda,
+	0x918f0bee, 0xfd22ed0d, 0xa43a6a57, 0xae9a1a27, 0x232f7e57, 0x41e93d1e,
+	0xf2ad7948, 0xe271257b, 0xe57acaf7, 0x91d99277, 0x845efd06, 0xf69f8a1f,
+	0x5907cb67, 0xcfff6932, 0xe6147fbc, 0xb9346c57, 0x97129a65, 0xc193d5ae,
+	0x1e7ce1eb, 0x4e157f34, 0xc8fba793, 0x64246f17, 0x6cc89752, 0xbb9095d2,
+	0xe67e8ecc, 0xf99c4238, 0x146529e6, 0xe1e9cebf, 0x5c5025d6, 0x4c396536,
+	0x6308fdb4, 0x8bce9b96, 0x4cc588d8, 0x79f12df1, 0x9f0c0d4a, 0x3c52471f,
+	0x4832f8c3, 0xe699e006, 0x14d1f53b, 0x8448e3ee, 0x93881bf1, 0x88c23e00,
+	0x4252112b, 0xbf1846ac, 0x42475e1f, 0x2179adff, 0x57ccaef1, 0x1f027de1,
+	0xd36244cf, 0x47137f41, 0xa775f12b, 0xfbd22169, 0xfbf2bb4e, 0xcb1388af,
+	0xc2ac9a4f, 0x24c9b12f, 0xe0aed7de, 0xca1cbb93, 0x9e041372, 0x3cf1ab47,
+	0xc515fccc, 0x01309d2f, 0x54d24c7c, 0x58c9e3fd, 0xb30d6f0a, 0x1cf6c5ae,
+	0xb852178f, 0xbac12eb3, 0x4b44c002, 0xfdb409d7, 0x926c404a, 0x3d22ae8f,
+	0xab189d58, 0xe707e9ed, 0x93761991, 0xee389e0f, 0x1499a0d8, 0xcfe5d22e,
+	0x06913c03, 0x29837ffa, 0xd210c53f, 0xb2f80994, 0x18262574, 0x52d47107,
+	0xde92d3b8, 0xd814da35, 0x132b488f, 0xfdb4e313, 0xdb4bf8fd, 0x3e066911,
+	0xb0cf20b8, 0x2a383267, 0x53f552f3, 0xc131b4e2, 0x5acf37fa, 0x91787809,
+	0xa38e81e1, 0xadf943de, 0x2b5f7d73, 0x2a7210e5, 0x942468ce, 0xa67467de,
+	0x1f5f52e5, 0xea2ee63e, 0x4a95ed86, 0xd0b6efae, 0xc764836f, 0x52cf3023,
+	0xf8cf5b8d, 0xa13a6aa9, 0x6f3f96ed, 0xf748adb0, 0xf8d43fcd, 0x2d23f008,
+	0xc93bffa2, 0x1d7ad2d0, 0xeaeb1752, 0x0bdefc74, 0xddca2ce8, 0x2b17451f,
+	0x9185ef5d, 0xdd60278e, 0xe8b8c3b7, 0xa309697a, 0xaf8e86eb, 0x01932245,
+	0x74e2549f, 0x7e02ca2d, 0x04eba3bc, 0x951297ca, 0x81d1c953, 0x351856fd,
+	0x5e1fb764, 0x757d78db, 0x096655b9, 0x9989f7c7, 0x4dc7ff60, 0xdf30d5b6,
+	0xae293d16, 0xe96c78a5, 0x9a48d3c2, 0xa01bfad2, 0xcf2840fc, 0x70e0edea,
+	0xabe5868b, 0xbee517db, 0xdd9236ac, 0x545fac74, 0xf29bb213, 0x4369e597,
+	0x4bb68b95, 0x147fd2a6, 0xa2dbed2f, 0x453e0156, 0xbce67dbd, 0xdd1c604f,
+	0x802df64d, 0xcd35d98f, 0xfb5f877a, 0xe01f30ed, 0x579b1d13, 0x454b187b,
+	0x45ccfb7f, 0xef1e209f, 0xbcdd20f2, 0xdb0160f0, 0xe1f6fdb1, 0x1fd21e4c,
+	0x72c3ac0d, 0xc5cb0ebe, 0xae53ab7a, 0x04a64937, 0x63f9468d, 0x5aa0b941,
+	0xc67a957f, 0x17ef844f, 0x4ebac4d3, 0xfb42d089, 0x64535953, 0x1fbec35a,
+	0xa64248a5, 0x64b2c91b, 0x21a33f9a, 0x7175e501, 0x1997b3a4, 0x0cd23ce9,
+	0x0ce489fb, 0x5f862f97, 0xe4b4bf8a, 0xd93e312b, 0xa9ed6fc7, 0xf6cf384d,
+	0x81a63f55, 0x786f2ebc, 0xa875e507, 0x191f3979, 0x48f38516, 0xd6e0fb60,
+	0xd9391b6a, 0x059787d3, 0x428f3666, 0x0f40acb9, 0x313664f2, 0x7ea7e81b,
+	0xc3d3fba4, 0x57b95c28, 0x6f285c47, 0x93901a6c, 0x9c63c397, 0x7e2c435d,
+	0x72612df2, 0xf0c96217, 0x688e49c1, 0xdc5c9c05, 0xb53e514d, 0xff22bbaa,
+	0x147d45ba, 0x1ccee9f9, 0xf963dc05, 0xebf94510, 0x7015f26b, 0x536e67ef,
+	0x35c1bf94, 0xc2770141, 0x9e3f202f, 0xca4e5d97, 0xbfae37a7, 0xde7f515b,
+	0xbf4859be, 0x895ebe3a, 0x4994d6d3, 0xd05b30f1, 0x80686e9f, 0x670e52c7,
+	0x68a3ded5, 0xb725b89f, 0xe1cfcb44, 0x69bb31fc, 0xf94258ee, 0x897d6fa3,
+	0x67221fa2, 0xd28e30ca, 0xa18912ed, 0x46a73eac, 0x7bcc4fb9, 0x37170946,
+	0x992cbe3c, 0xf407ea04, 0xdb4b4327, 0xac5d2129, 0xf5a8dfa0, 0x11519e84,
+	0xdd0fc8b1, 0x57c63748, 0x79303e6a, 0x074861f1, 0xf4f5e83e, 0x3e3e01b0,
+	0x19ebe2f0, 0xb9d5f109, 0x7b8e07b7, 0xee3a36de, 0x18fc40f4, 0x10fc9512,
+	0x43f25166, 0x1f928678, 0x7e4aac22, 0xe4ab9ae8, 0xa3be3507, 0x20fe4a6c,
+	0x105ffa1e, 0x568cc77c, 0xbca94df2, 0xefe00f72, 0x5f9e7872, 0x1fe90abf,
+	0xf0a987a0, 0x1af01e3a, 0xff6bc3ee, 0x2625b6dd, 0x61df660b, 0xe478041c,
+	0x197620f8, 0x046c52e5, 0x3e561dea, 0xcca7a14e, 0x107ae288, 0xd69fa76f,
+	0xf7d613bd, 0xa049c932, 0x967fd09e, 0xdcd3bae4, 0xc639df14, 0xd520a02a,
+	0xa2b3ec6f, 0xd32c1ca8, 0xdf21b3bb, 0xefc1ef4f, 0x417bd2ad, 0xa7411ef5,
+	0x7281c439, 0xa61f0025, 0x2f50f427, 0x3f40ff80, 0xff8411ed, 0xed077ea8,
+	0x5179f256, 0x7234b93a, 0x06a4be20, 0x0d911dc9, 0x72106cd3, 0x3fb171d6,
+	0x157ca366, 0x4a0da2f8, 0xe84772fe, 0x817da28f, 0x448f211f, 0x255206bb,
+	0xab7f0227, 0x50c913e4, 0xa9371de2, 0x223ff841, 0x35e80bd8, 0x416f9bc4,
+	0x0899037e, 0xbc4de94a, 0xd30f723f, 0x06a48d4b, 0xb242b2e9, 0x7f69972e,
+	0x4be975f2, 0x7f4ba508, 0x64ffd533, 0xa69b42d8, 0xd18d77c0, 0x773ec329,
+	0xfe60b91d, 0x3f5a06ea, 0xc5f95d61, 0xbff2d098, 0x5b8d8dfc, 0x281f549d,
+	0x2464f4d9, 0x5afaf4a1, 0xa56f135b, 0x2ed6cbf5, 0xb4d1b48d, 0x7d44eb45,
+	0x3c976bea, 0x7f6415da, 0xed8a3f7e, 0x264274df, 0x7c153f90, 0x795fa72a,
+	0xd7ac1c6c, 0xccead6ef, 0xb9522d32, 0xd036e0de, 0x375151f5, 0x46c92c7f,
+	0x8fc199b8, 0x54f8037c, 0x5e4a1b9f, 0x82de327a, 0x5933a3fb, 0x3ba90893,
+	0x62b4d099, 0x81463fcd, 0x3a49567e, 0xd4b8c196, 0x1fffa0f5, 0x595afee8,
+	0x77d4cf2b, 0xe11e720d, 0xb5d05aeb, 0x6fb1eeb2, 0xcdf848e4, 0x557e07a3,
+	0xbd062f97, 0x3b67ea8e, 0x89bb3c84, 0x45eb04dc, 0xe2b883e0, 0xdf93ef50,
+	0x99bb2178, 0xf66badc2, 0x787e4a47, 0xc421e422, 0xaced674f, 0x0969bcbf,
+	0xaa36874d, 0xfe5e90d7, 0xcf585cf5, 0x9ee12f52, 0x5a581a4a, 0x0e4773a6,
+	0xb654aca5, 0x687a874a, 0xaebfc6d9, 0x33c533fb, 0x5eb06639, 0xdd2fdb4f,
+	0x43c7f6f8, 0xa6bfbe40, 0xa6cb1d8b, 0x7de29261, 0xffa4a974, 0xe8f8e516,
+	0xdbfaf250, 0xfb164b7a, 0x8fb460fe, 0xedafa51e, 0x40ac93d4, 0xeed4147c,
+	0x414c7bd4, 0x3cdb52e8, 0x1e02648a, 0xdd03d27e, 0x9121debf, 0xf820f484,
+	0xa476f55c, 0xe909e383, 0xa5e4e9c8, 0x5cfabd5b, 0x412bf4c9, 0x2bbe7527,
+	0x0afa8cca, 0xca7a87b4, 0xf6ff1b1b, 0xf14fc526, 0xcc2b0627, 0xf76ff8bf,
+	0x4e3cc245, 0x5f18ab69, 0xde34b0bf, 0xd62dfb46, 0x98cde339, 0x2fc5efb2,
+	0x6df91afe, 0x8ebf87de, 0xd93fa827, 0x38b93492, 0xdf9824cf, 0xc4efe79c,
+	0x0e25cdf9, 0x17e2bbe4, 0x9fe83b64, 0xf464b5e7, 0x45d6416b, 0x175b3441,
+	0x15d356a7, 0x4a77e74d, 0xf38044cf, 0x8b5aeca0, 0x517bd287, 0x28dafc18,
+	0x8402ffd6, 0x9d191bcf, 0xf86b45ba, 0xdf4ba2ba, 0x3cd75f9f, 0x39f404d2,
+	0x004b6a9d, 0x6892af3d, 0xd21f1c9d, 0xfcaeda45, 0xa3026161, 0x8568b10f,
+	0xa8f2a47c, 0xf943be00, 0x2c2d9d26, 0x5c8f515b, 0xbff5f74c, 0xc74ff790,
+	0xd0f405dd, 0xf5c1f153, 0xa668caf2, 0x269fd327, 0x7e5fc7e2, 0xf7ad9beb,
+	0xf5b28f35, 0xabbbd62d, 0x7fe43468, 0x1fad887b, 0x839c97a8, 0x602772e3,
+	0x243bcefe, 0xe2060794, 0xd7aeb60e, 0x56baf8e8, 0xa13e0be6, 0xb9c2d6e3,
+	0x842fcbc7, 0xf14e6f9d, 0x27dcc8f3, 0x7dbf3d68, 0x8bfa12be, 0xac4f6c0c,
+	0xfd7c79db, 0xf7cd1cdf, 0xc9febe5d, 0xcf4171f3, 0x5b73eabf, 0xeeb5ce99,
+	0xe95eb0cf, 0x605a74fa, 0x30eceabd, 0x761b33ef, 0x5d377c7d, 0xc56a6727,
+	0x4fca553a, 0xde2d3af5, 0xa7f236fc, 0xa99e9d7a, 0xe83e3b7d, 0xeded7d3a,
+	0xb0664f78, 0x2bbffa75, 0xe6fe053f, 0xa7e52b46, 0xf0a18790, 0x728796a8,
+	0xa9f105b4, 0xe7f48796, 0xc8141910, 0x04ef827f, 0x5abc95bf, 0x1dc81d7e,
+	0xbe0a5f2f, 0xe0a5f2f3, 0x957cf53b, 0x5be753f8, 0xf9472ce1, 0xf0edbec4,
+	0xcf4a64a8, 0x6ac906b2, 0x3dbd60b0, 0xd93c5300, 0x35231b70, 0xfd63927b,
+	0xae0a9761, 0xa9b79555, 0xbcaabb60, 0xa9570543, 0x5c70bbca, 0xf4bbf565,
+	0x5b81e504, 0x10f75e51, 0x3d804671, 0xad7651a1, 0xa76ca8a6, 0xeb465e63,
+	0x47486fb7, 0x12697b7d, 0xc474dbb3, 0xfc8197e9, 0x4477376a, 0x75fc5346,
+	0xb373f184, 0xca3223bd, 0xdcd32ecf, 0xf5fca320, 0xdae8c8b8, 0xaa1af620,
+	0x3cfd2249, 0x62fe5424, 0x4d90bf66, 0xef408b69, 0xa0ad91af, 0x3c5ecd0b,
+	0xcef41229, 0xe31e86f6, 0x73efd327, 0x43baa1f6, 0x903db9da, 0x49b4039e,
+	0x38b3fe94, 0xe710b63f, 0x7ce8c353, 0xa2943566, 0xa57b3e3e, 0x4d735e92,
+	0x9ccf510b, 0x1efd04fb, 0xe12fdf44, 0xaf3f14ed, 0x2b212fcc, 0x1d289f91,
+	0x297760f5, 0x77cca359, 0x2ecffb70, 0x48fde7c1, 0x87b9c53f, 0x3942d5cb,
+	0xb064fc73, 0x0995ec57, 0x836749f0, 0xdcef788c, 0x9e08b920, 0xa83d986b,
+	0x3ca07482, 0x8569f813, 0x2d0fa0f5, 0x7cb75a6f, 0x5a01e83a, 0xf1eb6ca4,
+	0xda68e7ec, 0xaf58f5ba, 0x71bf00ed, 0x0938ed24, 0x08ed3fb2, 0x64dda1b3,
+	0x00cb8447, 0x3f7c6ff4, 0xa3b401ed, 0xce30ac18, 0x67df1df6, 0x3fb4df03,
+	0x79062864, 0xecca9ed5, 0xbf1a4b23, 0xe57266cf, 0x5b57a461, 0x37e99eb4,
+	0x99fdff0d, 0x06fae704, 0xf7d82b3a, 0xf9e38477, 0x30cdf2ad, 0xbff6c117,
+	0xfd30e41f, 0xd0a9f372, 0x61f20713, 0x19bc9904, 0x704517c6, 0x07db953e,
+	0xed3175ff, 0xa76ebcbf, 0x2fcd167e, 0x6ca7eb78, 0x37db0fd8, 0x4da67d33,
+	0x5dd2b847, 0xad1e68f2, 0x61c616d7, 0x056eefe7, 0x160db77d, 0x9e423747,
+	0xab70d98e, 0x4f0c14d3, 0x5e21a53f, 0xd13f10ae, 0xfbe47a31, 0x48fc7d7f,
+	0x2bf28236, 0x609465c2, 0x767e3fde, 0x0ec1e41c, 0x98a9f203, 0xc5da99ef,
+	0x6e5c7fa8, 0xfe62fdde, 0x4ae2f85d, 0xafb92de7, 0x7ff6de64, 0xe21f7a71,
+	0x5d7aa8fb, 0xe379f204, 0x8c6f08fd, 0xaf78e2b0, 0xfeabdf90, 0x337dd1af,
+	0xe5e47ba5, 0x7fc7fde7, 0xa68779f1, 0x9ce9befb, 0xc3b066ee, 0x13f5be55,
+	0xe56aef8e, 0x537fa6dc, 0xba7a17be, 0x3890befa, 0xf0a5a225, 0x744f0e70,
+	0xd2e79e4a, 0xc388f77d, 0xd2113a79, 0xdd475048, 0xcad38f27, 0x2a4b24c3,
+	0xde132828, 0x7a071dec, 0x8aa932cb, 0x4737eca8, 0x6476d3b7, 0x940c474d,
+	0x2d0e242b, 0x217217bb, 0x3c91df1e, 0x15f9d768, 0xb4852db1, 0x9a8ed482,
+	0x3bdd6902, 0x97b42547, 0x923f2184, 0x9e3df57a, 0x0d11f516, 0x1de8019d,
+	0xce2b00a5, 0xbe9f5a77, 0x8bc20722, 0xa293d8e9, 0x313f7f2f, 0xcfccd63a,
+	0x58bd1d7d, 0xc46fd33e, 0xfbe8727d, 0x66f8e55c, 0x4ba76699, 0xf36f41dd,
+	0x8c15c71b, 0x291e0093, 0x3ca6de9e, 0x7bfaa463, 0xd1d1dc12, 0x618fecf1,
+	0x72ef2df4, 0x141b63e2, 0xe694d01b, 0xa9563f7f, 0xec193627, 0xae0d5b24,
+	0xad4fd0b5, 0x1b4ddef2, 0xa1dbcecb, 0xef13d0f1, 0x421e0453, 0x3efaabf1,
+	0x1be80665, 0x667c6103, 0x74c6f8c5, 0x29631bb1, 0xf3d47a9c, 0xd5f1a495,
+	0xd75b942f, 0x60fa073d, 0x2b7fd33d, 0xf57bb386, 0xda326d8a, 0x0c1ada0f,
+	0xb18e892e, 0xde177eb8, 0x1b56b4c7, 0x9eb7dc09, 0xab48eba6, 0x55fe04bd,
+	0x371c0df3, 0xcd67075b, 0xea86a746, 0xee5896d5, 0xb24d908f, 0x3fd31e61,
+	0xcd31173c, 0x9ede7682, 0x5e76feb1, 0xab37fdef, 0xedcf9017, 0xedef54c0,
+	0xcce2f20a, 0x65ef04fd, 0x6d23b689, 0x1cbf9642, 0x23a3ce50, 0xa9e3d186,
+	0xb4167b68, 0xf294dcd7, 0x2857b414, 0x06e9ec37, 0xcec859f8, 0xc40c2359,
+	0x64e94619, 0x56b5ef90, 0xde95c9f9, 0x9cf196bf, 0xfbc2e9fd, 0xfa6567a2,
+	0x692dedda, 0x741eecec, 0xa5bf67f3, 0x988ef4ed, 0xd3daf78a, 0x93bcec0c,
+	0xbbb87f7e, 0x6f5f5d07, 0xf81189e9, 0xeabdd74b, 0x4e70ade0, 0xabad928a,
+	0x32b17f69, 0x217c0919, 0xce1b1bc6, 0xe7be23e6, 0x0246b0bd, 0x8fbe22f9,
+	0x7fd1d258, 0xb1f54d01, 0xf8c53f8c, 0xf4741446, 0xf0f932af, 0xe7e466de,
+	0xd716b319, 0xb4c3b093, 0xec0979c1, 0xeb0e7024, 0x3db7c84b, 0x5e1f786d,
+	0x20ec4611, 0xdb35bef7, 0x7ce146ba, 0xb66f782c, 0xb7acacd0, 0x04fcd3f0,
+	0xa0571f28, 0xe91787bc, 0xc76658ac, 0x4341fb29, 0xe2767e9c, 0x0f5bad32,
+	0x1fb281f6, 0x77e9c4f4, 0x9d05fc0f, 0x5cefd3b7, 0x4f4b4f7e, 0x22cc27b4,
+	0xcc27b456, 0x9c4e1e5a, 0x3aa2e77e, 0xbd350e3e, 0x56fa71cf, 0x96896944,
+	0x2242abce, 0xdb954bf6, 0x9f764ffe, 0x23ff1eaf, 0x70e36eb0, 0x779fb87e,
+	0xab5c1a34, 0x9dcb03d3, 0xb8f6618b, 0x3dc51bf4, 0xc5cf07bf, 0x25e5b70b,
+	0x8aed2090, 0x1b07e2b4, 0xa5e1f971, 0x7983d682, 0x9230fc56, 0x6f07fab8,
+	0x66f982cf, 0x81cede05, 0x5fe81a7c, 0x4df1163f, 0xb091c44f, 0x97a95e7e,
+	0xf13d8347, 0xd19abd95, 0x678c83ab, 0x9f679f03, 0xf5f22d38, 0x5b626e8d,
+	0xa29e8415, 0xdb347c5c, 0x459f0117, 0xa8788159, 0x074cf7cb, 0x2469b87e,
+	0xefeb05b1, 0xc42f9c2f, 0x19a2ebb8, 0x18762b7e, 0xdb405ed0, 0x011455ed,
+	0x71733ddf, 0x74959b3d, 0x8d863cfb, 0x4f36eea3, 0x76d3f41a, 0x083ff72b,
+	0xe173df41, 0xe018f68e, 0x30b449c3, 0xda4187e5, 0xb7ecc92a, 0xbe94ae3e,
+	0xd6ac3f15, 0x102bd3f9, 0xae74ebef, 0xc576befe, 0x22f3643d, 0xe6def519,
+	0x95fefc83, 0x62eee3db, 0x499a4cac, 0x97b02ed5, 0x68495734, 0x41dbbbc7,
+	0xdeba22c8, 0x14b87aef, 0x97d34ded, 0x8234bebe, 0xca907f7f, 0x0912bb8d,
+	0x77a699b9, 0x09da56cf, 0xb70596a6, 0x3f7839a7, 0xfa0748d3, 0xa85f4cf9,
+	0xfb702fb2, 0xe541e383, 0x17e02ff7, 0xc60fd72a, 0x0e43bb72, 0xbd7901c9,
+	0x3d39e960, 0xf8a109a8, 0xc3ca85ba, 0x8b905a9e, 0xab5b4c46, 0x157e98cd,
+	0xf604fb6b, 0x8eb1f3a5, 0xb6ee414f, 0x52e4c87f, 0x4c73cae7, 0x418f2f9e,
+	0x2ef2c73b, 0xd78dff6c, 0xd3ea3b58, 0x3e3efe41, 0x8fc51c71, 0xeb5a2fe1,
+	0x9b878a12, 0xc0d3353c, 0x35e84525, 0x905aac8f, 0x127e8b11, 0xefc041d6,
+	0x32deafb1, 0x7133ec1a, 0x00db86af, 0x5cf22b7f, 0x62b18f78, 0xe7f41771,
+	0x0fcdd232, 0x0d8f7ae4, 0x3fb81a79, 0xb16ea1bd, 0xd76d1ba6, 0x0dd3b0dc,
+	0x35cb0efe, 0x6b3da0df, 0xf984cdf9, 0x287ef17f, 0x6919fb3e, 0x4c13f13b,
+	0x171d2f78, 0x50d44aee, 0xbd46ee38, 0x17ea0975, 0xe0a95f4c, 0x2c976834,
+	0x7e0a1c7f, 0x8264fb64, 0x16b13e7e, 0x087daf94, 0x0b92bfb3, 0x0f945f14,
+	0x8fc40c60, 0x2dab5aef, 0x92fc9f68, 0xd0fb29db, 0xce8ea1f3, 0x4f8a2be7,
+	0x6a1ecb80, 0xebe3051a, 0xabac48d8, 0xb7d89169, 0x912df292, 0xea337780,
+	0x13aba7bb, 0x16e31531, 0x6bd947fb, 0x3f41a36b, 0x9999229e, 0x43d7095f,
+	0x10121f05, 0x4c9e1456, 0x8bbf9c98, 0xa8539feb, 0xe3a30dd8, 0xd35887bd,
+	0x46dddea0, 0xf8a1a912, 0x3b50529f, 0x5299f710, 0xe7ec145d, 0xe30542a0,
+	0x8bfa5087, 0x72ea2758, 0x81af41a7, 0xc98db8f5, 0x11758a83, 0x6bbbb44e,
+	0xecffb03a, 0x7b1ebe38, 0xd22e3f31, 0xfd60d893, 0xbde8ec54, 0xa062e80a,
+	0xa5e303fc, 0x9b98ca72, 0xd19cfa83, 0x7f3fe748, 0x8a5b987b, 0xd9a7a3f4,
+	0xfa034fd6, 0x1c9484fc, 0x4a3a97f4, 0xd80447e5, 0xb8693c71, 0xeb47419e,
+	0x7d5cc306, 0x54ebdc1b, 0xfb30dc50, 0xee0ccbce, 0x2d6761c6, 0x7978d383,
+	0x777066e2, 0xdbddff51, 0xc743fec0, 0x621fcbcb, 0x1f011f70, 0x45b3e566,
+	0xce0c63f9, 0x1927f911, 0x7fb4483c, 0x1843c18a, 0xd3c9fc7c, 0x0f4139b7,
+	0x8ffae289, 0xaee783d5, 0x6b7b378a, 0xde96fdfc, 0xf3726389, 0x772b1874,
+	0x98cc9d3a, 0x08cbf400, 0x5e55fe5e, 0x02c5e85e, 0x493ea27f, 0x266f5e0c,
+	0x80bf37af, 0x64e129eb, 0x5bf31879, 0xba524675, 0xbde86eee, 0xc617ca1a,
+	0xd03a7345, 0xdaf80676, 0xed11f411, 0x244b7e89, 0xe9b497a6, 0xbbd89d98,
+	0xfdb1c66f, 0x30f3f684, 0xd293d3f6, 0x909bda15, 0x81f80fd2, 0xf39681f2,
+	0x80f81681, 0x7c6c5590, 0xc028ddb4, 0x3fa3b9b9, 0xbeedbff8, 0xfba7cb40,
+	0xcbe6adc6, 0xcddf3807, 0x81cec25f, 0x703d511d, 0xce1118bf, 0x403244f4,
+	0x380f8f39, 0x8b89de1e, 0xd2f4f362, 0x7ec24125, 0xa63c7096, 0xc84d7a78,
+	0xff79faba, 0x97e78f9e, 0xbd1fb79e, 0xd7e4326c, 0xcd9ab0df, 0xd60bd187,
+	0x85ea3447, 0x3bda2b8f, 0x3f5558bd, 0x98702e97, 0xac0f746f, 0x311848ff,
+	0xdab277e2, 0x9b3d2f5b, 0xc7a0abe9, 0x27ca2f96, 0x74ff6127, 0x85f78cde,
+	0x952a5c65, 0xe3d26f2f, 0x64ad6bfd, 0xef3cbf43, 0xe00ec73c, 0x79f334f3,
+	0x5d3cd99a, 0x87257e0a, 0x88eed23a, 0x2f63e408, 0x65edbdfe, 0x14f6f0e6,
+	0x7b582372, 0xd1422720, 0xd34bfdc5, 0x85538c45, 0x559e2f4f, 0x84f813e3,
+	0x32f58df5, 0xc6297c95, 0xf5a87af9, 0xfb33fe9d, 0xcaf563cb, 0x08e9437f,
+	0xc945a7fa, 0xe975e5d7, 0x7d327b9a, 0xae8036f5, 0xf3f3e717, 0xfdc71939,
+	0x0a485286, 0x7cce0fec, 0xfec37562, 0xfb80d9b9, 0xfd92e7fe, 0x505c5c54,
+	0xaefc20ff, 0x47b6173e, 0x1bec09e9, 0x7b40fb6a, 0x06c9b59c, 0x0967e5f7,
+	0x550edffa, 0x6fc033d9, 0x57c9b373, 0x295fb7cf, 0x4cf111e1, 0xc3b40cbf,
+	0xe68a7dd0, 0xb868300f, 0x018dd73c, 0x5c203d79, 0xf0d57705, 0xe0c8c4fd,
+	0x293b0e04, 0x2034f14c, 0x3187e459, 0x3f513af8, 0x51bf958c, 0x5f70e794,
+	0xa208fe55, 0xfdb28784, 0xb0a9e624, 0x20ff7031, 0x78d8a0f4, 0x2e43ce0b,
+	0x4be010d0, 0x80d53803, 0xc196f5be, 0xa08d7a05, 0xce94a853, 0x3dd82ab3,
+	0xf0bc6a7e, 0x977e6033, 0xbb40f2e8, 0x047615ca, 0x4761568f, 0x3a348f68,
+	0xc768aee5, 0xadf605eb, 0xf747788c, 0x32bf5a78, 0x3a7a75bc, 0xff0a1f85,
+	0x256cfd0c, 0x258fd31b, 0xa54c998b, 0x575a95e3, 0x23df41d8, 0x16a35625,
+	0x4aedc4e9, 0xe94bf770, 0xcbc77db1, 0x6eb3240f, 0xc18e30ec, 0xee54ae5c,
+	0xe3bbf1f8, 0x369fffb0, 0x7dfae159, 0xf7b3c370, 0xb838be81, 0x4c8fff61,
+	0x5fa1304f, 0x60f718ac, 0xdfa29bf0, 0x380deb73, 0x370dc1fe, 0x43f281cc,
+	0x01e86a3f, 0x84fc377c, 0x9988097b, 0x9237c6f6, 0xb91f281d, 0x25392c9f,
+	0xd5c19c41, 0x7c5e45ea, 0x7e8f43fc, 0x231f141e, 0xcc2e3e19, 0xc5b4fc05,
+	0x41e65c1d, 0x8b0a62a3, 0xe9f5e067, 0xb9fd03b9, 0x2b17493b, 0x23db5dd8,
+	0x3b6a3b58, 0x3a22d1e4, 0x787dfa3f, 0x9f811d07, 0x07e23fde, 0x3e4c1523,
+	0x29448f38, 0x926302e0, 0x39a17ca2, 0x4b7f915d, 0xfc8a0de2, 0x14fd2d5b,
+	0xc18c98f0, 0x5bb7f28a, 0xf8f014f3, 0xe5155bdc, 0x0a456c8b, 0x56bb9fb8,
+	0xdeffbe51, 0x21fe657a, 0x721d0322, 0x41775a8e, 0x1b4bf03a, 0xb507359a,
+	0x7482ed54, 0x1d0cb7f0, 0x90ec3b6a, 0xc1776a1f, 0x15876177, 0xce30f0fa,
+	0x3b2ec22f, 0xf4b93649, 0x89afedc1, 0x959e9a5a, 0x649f283c, 0xafed1d8d,
+	0x078e61c9, 0x473d29e2, 0x2814aa91, 0xb0ec0237, 0x33850894, 0xfc01fece,
+	0x32a73aad, 0x2051dc7b, 0xbbbbc80c, 0x788ed022, 0xf1b55754, 0xce43ea38,
+	0xc3a39054, 0xe62b263b, 0x8707e23d, 0x6479de62, 0xf74e8abe, 0xf342e02a,
+	0x6fe5146f, 0xc8ac5c49, 0xa8f2d5bf, 0x2b8f4e8a, 0xcfaddbe0, 0x4ab4e8aa,
+	0x1597a745, 0x1bbf82f9, 0x076c7db8, 0x3a68b5d8, 0x0054d3cc, 0xa5b9b874,
+	0xdb91e903, 0x74005354, 0x16b4721a, 0x6dc035e9, 0x3bf1002f, 0x2ed56f68,
+	0x5cf4dd48, 0x4efef503, 0xee9035cf, 0xb023e7a6, 0xcea9ed8f, 0xb56f74c0,
+	0x5bbfbf15, 0x7be98b9d, 0x3f4c36d5, 0x698d1ea8, 0xd31db553, 0x2c5aeada,
+	0x9ebab9bf, 0xf862d7d9, 0x0bf1473e, 0x38368c76, 0x9cbca253, 0x0db2f28a,
+	0x5f2850ce, 0x177066ef, 0x0c29930e, 0xebe5648e, 0xf2829a9f, 0x8cf83f74,
+	0x30bf81c4, 0xd824cf4b, 0xdd3409ec, 0x8186e70b, 0x5a89b8be, 0xe4a7688d,
+	0xb1eb9031, 0x525fbc31, 0x4296c632, 0x5ec5d2c8, 0xc18431f8, 0x2307eb28,
+	0x1bc1f8b3, 0xe21b6b61, 0xc61edfab, 0x97c6f7f2, 0x4173818c, 0xf330fae9,
+	0x29747e97, 0xc0528daf, 0xe649f595, 0x88e8fd2f, 0x7654c97e, 0xe5091492,
+	0x36c3d09f, 0x20438e48, 0x1e8dca7f, 0x1aa627f4, 0x6c2bfa30, 0x8044b16c,
+	0x85e0ea5d, 0xcf8011de, 0x672edc83, 0x3374fd23, 0xeda79a69, 0xff7d83a7,
+	0x124f4aa8, 0xef2941e9, 0x56b76853, 0xf413bd63, 0x3a6de724, 0x413ffc02,
+	0x51007662, 0xa604ddee, 0x6edfb4af, 0x21e0f3c0, 0xc60e6d8d, 0x9f024811,
+	0x149f718b, 0x11c2f7f3, 0x7489df22, 0xf9128f6b, 0xe5538e1c, 0x1d60a927,
+	0x5f19df2b, 0x05fe748d, 0x904ff9f1, 0xca7c6a0b, 0x97a66609, 0xa0a96db9,
+	0x50f808c7, 0xe41cad91, 0xdb3ca1c7, 0x83ef30f5, 0x17930376, 0xf5d708f6,
+	0x0db33768, 0xfecfd3bc, 0xc3a09d55, 0x1f00777a, 0xddc4fdc3, 0x3515ca32,
+	0x604a5ffe, 0x15f22afa, 0x0aefbb9c, 0x21cd73c7, 0x1cde290d, 0x387b5e84,
+	0x0974ce1d, 0x277df76e, 0xf9fbed81, 0x6e37cb13, 0x85cf7b81, 0xbb74d861,
+	0x93ea37bd, 0xf31255ed, 0x3aa96903, 0x560df2cb, 0x33f4128e, 0xd1bf029f,
+	0xf70d8dfc, 0xcb4c3f9d, 0xdc30b67f, 0xb9967b8a, 0x6eefe20f, 0x8c5a785e,
+	0x69bff511, 0x61f35213, 0xfdc29e1f, 0x6847a79a, 0x59bdd1e3, 0xd337a51b,
+	0xfa9b6676, 0x2418a32d, 0xf48b6373, 0x4e2e1451, 0xa41977b9, 0xd71f3d25,
+	0x595dd5c3, 0x40cb6be6, 0x22a72f0d, 0x5bca2063, 0x74d5fbcf, 0xc916335c,
+	0xe51e7616, 0xf4adb8fb, 0x045de2aa, 0xf1e21fbf, 0x78091447, 0xf1cbdede,
+	0xe5edea2d, 0xe0482af8, 0x245922f7, 0xb8aca2eb, 0x0125b8f7, 0xbda63dee,
+	0xbdc42bdf, 0xeb9f3288, 0x59eb1fbf, 0x75eba5bf, 0xb9104f2e, 0x9d812fbc,
+	0x89e90f9b, 0x9d55ee0d, 0x277db377, 0x87b9f27b, 0x67f9d206, 0xe8359837,
+	0x7ca33457, 0xb3a1d00b, 0x4751d39e, 0x4dfd61ea, 0xecfcce74, 0x96261230,
+	0x5e1fc7f8, 0x36b67f6e, 0xf7fc0b99, 0x7b1e01e7, 0x9cfeff81, 0x005351db,
+	0x8fd8b0fb, 0xbba83dc5, 0x03ef98ba, 0x77bf2855, 0x78422f57, 0xd313b54f,
+	0xf9cbd5bd, 0xf983503e, 0xc33f55ef, 0xdf1701f7, 0xf7e3d607, 0x98c9ea86,
+	0x6076a8ee, 0xbd8ba63a, 0x1df4e402, 0x47c60746, 0x81acfdbe, 0x675e7dc0,
+	0x47e30183, 0xf1788ae8, 0x538858b9, 0x58579832, 0xd495e302, 0x1f978afb,
+	0xb3fb7e54, 0x0b7ed8bb, 0x99e378fd, 0x2ebd3f40, 0x241d3f34, 0xdf390e5c,
+	0xd0743945, 0x07d6e42c, 0x7808b60c, 0xa2ad833f, 0xf72807fc, 0xbe78da0e,
+	0xed833f73, 0x601c9f3c, 0xefc043b0, 0x944ab831, 0x73dac03f, 0x063df80a,
+	0x4df288d7, 0x7831760c, 0xa21fd3ed, 0xd223cc5c, 0x056f586e, 0x2091f4c6,
+	0x79cd0b62, 0xdd9d61bb, 0x24733892, 0x826333ac, 0x56fcf905, 0x00dde2cb,
+	0xa558c98f, 0xcdba1305, 0xa861efd4, 0xa45b1f97, 0x53237a85, 0xea187dee,
+	0xb8f9ea8d, 0xe8debfd6, 0x061ed7cc, 0x9575ac78, 0x236bf416, 0xbe81c5c8,
+	0xf22c0476, 0xe2213ec5, 0xd836faf0, 0xaeca3f71, 0x247ee3b7, 0x477f9cf0,
+	0x9c3352d1, 0xb325ea03, 0x43b8e304, 0x404f06f2, 0x05c5630c, 0x1e291aba,
+	0xe13c12ea, 0x2b3ce98b, 0xf0821e01, 0xd97f8b63, 0x109dd5fd, 0xc2383f64,
+	0x2be807e5, 0x23ffb445, 0x3f835d74, 0x4bc70442, 0x88ccc365, 0x755e7899,
+	0x189e8044, 0x0490f5c3, 0x2dfe6948, 0xbfcb1d3c, 0xef6c941f, 0x30f8bc82,
+	0x03b79f81, 0x1861cde5, 0x73279507, 0xe40238d9, 0xcb9f328d, 0xdf62239b,
+	0xb2e2c01a, 0x8b0d49cb, 0x7c4cbee7, 0xb8dc405f, 0x84a71e2a, 0xa5b1eed4,
+	0x6f178b07, 0x2e2f1a0a, 0xfd031bfc, 0xb0b9f96d, 0x93cf930b, 0xc3d17641,
+	0xb9686760, 0xab971dfb, 0xcca0f0a1, 0xe0e285ff, 0xf90e2389, 0xbc3c5423,
+	0xf9c08be0, 0xb6f3b0d5, 0xea9fe196, 0x81c4c69f, 0x8823bf20, 0x6794ab1c,
+	0xcf212f87, 0x97f1f2d4, 0x6693bb92, 0xfdc1a794, 0x98b91043, 0xcb54e30a,
+	0xc93e6745, 0x7505fb80, 0x0a68f42a, 0xd3de213b, 0xed0c23b9, 0x8e31f514,
+	0x0fd0448c, 0x83be3bf3, 0xeb122d78, 0x371c3c41, 0xf6846e39, 0xbff33d25,
+	0x2dcdf110, 0x753f7d5e, 0x7923c62c, 0xfb0a7bf5, 0x9efdf7e9, 0xf20b1d49,
+	0xdec9bbad, 0x1921da1f, 0xc85d1fcc, 0xbef1508f, 0x3ed91343, 0xff92981f,
+	0x8e5838b7, 0xae7a369f, 0xf4e74dda, 0x62bf86ce, 0x57d9ccdc, 0x56c1c79a,
+	0x265dba5c, 0xb68f8a46, 0x0dc4110d, 0xe52d1be7, 0xd4969a3f, 0x2b8c36c9,
+	0x4d07bad2, 0x5bd7f08c, 0xe974ff98, 0xcadc08a6, 0x5c1b364b, 0x92fd6963,
+	0xd1b327ad, 0x0ea1f5af, 0xbe730fbc, 0x22d75922, 0x13d93cfa, 0x1b594fc1,
+	0x353e4edc, 0xad87e991, 0x8327c2db, 0xbf5a56bd, 0x25deed13, 0x707493b5,
+	0x4dca2a0f, 0x3e21d44c, 0xa7b43968, 0xb8965f3b, 0xe41eeb1e, 0xfc2caf7f,
+	0x665f5bb7, 0x7a92a976, 0xc60c2b9d, 0x36b5feb7, 0x3929c788, 0x4dfb406a,
+	0xf46ffae5, 0x4d82c740, 0x1f01231b, 0xd932f595, 0x550fbcad, 0x97ac41c6,
+	0xe3f006dd, 0x7f472baf, 0x23370a6e, 0x1e3c1b5c, 0xfafe72e8, 0xb71a9ba5,
+	0x4b0a293a, 0x12297f5f, 0x374e29e2, 0xbc048dad, 0x192f76d3, 0xf7101487,
+	0x25e8fb50, 0x134e1ee3, 0x134b03c6, 0xeb237155, 0xcb71b863, 0x3fdfc83d,
+	0x274787c9, 0xf3c6f3a3, 0x5e5c422f, 0x2f33e6eb, 0xfecb78c2, 0xf045ee79,
+	0xc23b26cf, 0xad0f60cc, 0xc630e57c, 0x7f7a8935, 0x627a6449, 0xe12fcbdb,
+	0x31db6a3d, 0x35b7afa6, 0x6dbe429e, 0x78c7ed7b, 0xd85af388, 0x0efe8858,
+	0x2414b70b, 0x02e72090, 0xbadf0291, 0x4d1e20ae, 0x755daf4d, 0xfc8fd0bb,
+	0xe837a52c, 0x9638dea1, 0x171672da, 0x2bf1e164, 0xfdc7821b, 0xd4abc405,
+	0x1c3f1c4d, 0x3a75e2ec, 0x415c85ab, 0x4a21cadc, 0x4057bec7, 0x728f0dbf,
+	0xf4e0fe7f, 0x50bf0b3b, 0x2a26353a, 0xd7ce56e3, 0x9fcbf9cd, 0x262eaf21,
+	0xfb8adc64, 0xe226f00a, 0xda24570a, 0x6c3c82f6, 0xc62afcdd, 0xc6dd0109,
+	0x6db689d2, 0xf1069f3c, 0x8eba2d98, 0xc1efdce9, 0x25acf70c, 0xfd70478d,
+	0x10f96db4, 0x8f08f6e3, 0xde236bef, 0x258f161f, 0x89ae79f1, 0x1c2ef160,
+	0x030f10ff, 0x3bf4d0f7, 0x7c617b8f, 0x4261e22c, 0x4c5cc1c5, 0x2df6ceec,
+	0xbe58ef1e, 0x2b8f38e6, 0xca4bd1f1, 0x5f353a05, 0xe048be32, 0xb8780cc1,
+	0x2bae3fc9, 0x3791bc78, 0x0bd38ffe, 0x5cf107e8, 0x0a2db8de, 0x3a58430f,
+	0x83c88242, 0x1e2c55b6, 0xe3ff7d4c, 0x3843cfa3, 0xd152ffaf, 0xf433cc59,
+	0xd3ad9239, 0x6aff8dcb, 0xd12673f1, 0x127388a2, 0x10bd6a78, 0x26a78e5c,
+	0x67eafdb1, 0x00b6b7eb, 0x092536fd, 0x66badbf4, 0xe62b7e85, 0x05754adb,
+	0xafadef90, 0x960c77b0, 0x3942eed5, 0x141796ae, 0xf760b62f, 0x98937d0d,
+	0x67a8aa9c, 0xe6eb44fb, 0xb6899708, 0x6df7fad3, 0x1b15cfc8, 0x8fb0f59a,
+	0xbaff59ef, 0xb317f99e, 0x84fb03f5, 0xadfc6fcb, 0x8d99db7f, 0x09fa1b72,
+	0xbce1d742, 0xe086ca9f, 0xf86e54fd, 0xa94fd146, 0xfccf56b3, 0xdfdd2e8d,
+	0x126dad69, 0x49ba77b4, 0x7269a607, 0x6df996ba, 0x24b7fba9, 0x4abdf8e9,
+	0xa8f7d70d, 0xc83f47bc, 0x9c271e61, 0x51c3735b, 0xc5e6b63e, 0x35a45da3,
+	0x6965af55, 0x7fff842f, 0x935bfbdb, 0xbfac8d72, 0x371f1851, 0xd6e8c62e,
+	0x8073f5d0, 0xaf513fbf, 0x00fff677, 0x8707cfee, 0x78e3dfa1, 0x4dbefcc9,
+	0x6827bc0a, 0xdf3537dd, 0x5f781d92, 0x91912ad0, 0x65279512, 0xf0a59acd,
+	0x0fef6aed, 0x477d16ca, 0xa19d1de0, 0x78102e8b, 0xefeae7df, 0xb3a29a13,
+	0x70b7602d, 0xfa6d812c, 0x829ad6a6, 0x597216ae, 0x0edcef3f, 0x67b012d7,
+	0xdec0b88c, 0xba4a18b7, 0x4daae158, 0xfccfddca, 0xbeaa79be, 0x31ae9877,
+	0x83ddf4b9, 0xe7909fbc, 0xdc82481e, 0x867b8837, 0x8ac730e4, 0xf71081a4,
+	0xc01beecf, 0xe4f915f3, 0x3fa6c7b7, 0x3081e780, 0xe71573f4, 0x27f58d99,
+	0x9fac60e3, 0xb93dc537, 0x34af7e9d, 0x79f8877f, 0x8f9f8932, 0xf28ac4ac,
+	0xf50f7e4b, 0xfb25cff8, 0xfb8a49b3, 0x08b9ef58, 0x63efabbf, 0x8cee7f85,
+	0xec276597, 0x0dbf85a3, 0xc608781b, 0x03961751, 0xc31fd8e3, 0x980641fd,
+	0xf1401f6f, 0xf5cb7673, 0x5b3fe039, 0x9eefb5e4, 0x88bcbd47, 0x338963db,
+	0xafd876f3, 0xd7bd848a, 0xdcf961e9, 0xd9febbde, 0x3ac3b7de, 0xa1c1fe14,
+	0xa2c56de4, 0x421f9137, 0xda12cbe6, 0xa6e4c939, 0xbd3f22b4, 0x072e4df9,
+	0x397efdc4, 0xa78b0bf7, 0xf68dc8d3, 0xbc5791f9, 0xba46e611, 0xcec4d97e,
+	0xe976dbc5, 0x09af1d99, 0xa78f21cf, 0x7c02de3a, 0x81da2c55, 0x762626f8,
+	0x6c96396e, 0x1d31fee6, 0x0eae5099, 0x6eb027ce, 0x7da184b6, 0xd1eb4136,
+	0x2ab8e95b, 0x8bf4128e, 0x33c7c638, 0xd40be9a8, 0x70e6bff7, 0x37e67ad7,
+	0xd7c3c6a1, 0xf1aa1149, 0xc60c37bf, 0x2231f02f, 0xfdc2d49c, 0x958f9d30,
+	0x98aac7cf, 0x814fd78f, 0x6ff74bd6, 0xebf0a2d7, 0x5b2af312, 0xdffc49d4,
+	0x1c5d6ad3, 0xe311df7c, 0x97dbdf56, 0x840f7698, 0x2ae4a37d, 0x02a7fafd,
+	0x3e01c46f, 0x871eaa59, 0x7ee32d3a, 0x17efea1d, 0xf67dc492, 0x7fae3b2f,
+	0x69bd7bb7, 0xf9153b55, 0xbbed1633, 0x713f7ba2, 0x75bff71f, 0xdeea17ff,
+	0xf66ec1ef, 0x8e946ecc, 0x29fcdfe3, 0x9df108f3, 0x1105eddb, 0xbe51c718,
+	0xc8156135, 0x7e470fe5, 0xc2f60f14, 0xf71231fd, 0x1cb4bde7, 0xf7bb52b9,
+	0x717bd567, 0x393f625c, 0x5846ed51, 0x3a9f076d, 0xa7bc31e2, 0xdb2dfc1e,
+	0xb05e8637, 0xf2c9b91d, 0xc231edf3, 0x32ec230e, 0xd4cf3b6a, 0xefe4666e,
+	0xec29bfe9, 0x9417fcca, 0xac472e7c, 0xffdc83d1, 0x17ee7cb9, 0xd65d3347,
+	0xe2877b40, 0xce89bda1, 0xb0475ee6, 0xc23de51d, 0x7b8e2e98, 0x73134e80,
+	0x88863f99, 0x5b942a5b, 0x7a049c9f, 0xfb8c934e, 0x7ddfc2d7, 0x2ce7b7f4,
+	0x34a8f481, 0x131ca277, 0x5eeb1774, 0x86a543f0, 0x14e7b1f9, 0x097fd058,
+	0x57d811e4, 0x0d34746b, 0x299ee742, 0x3c600bf9, 0x0e3afded, 0x8abf024f,
+	0x669dff2d, 0x09aebd41, 0xb39779cf, 0xbffd80df, 0xd73b105b, 0xe8393cb2,
+	0x298977fb, 0x6aceff4c, 0xbbfbb0d6, 0x7aeb92f9, 0x347973b0, 0x39acaa88,
+	0xe0192ee9, 0x559f963a, 0x2892e51e, 0x8c38359c, 0x851bdfe5, 0x88d3df71,
+	0xf3fb0747, 0xa60af00c, 0xc1624ce9, 0xf1b11674, 0xba3ae5f7, 0x9cfeb44b,
+	0x83f32306, 0x309e9e25, 0xfa10ec9f, 0xcc6eceef, 0x3bb7d3be, 0xf37ddf93,
+	0x19fc7295, 0x726eff9e, 0xef9849fe, 0x67b66ee9, 0xdd1bed21, 0x704fe2b4,
+	0x21a2dcde, 0x1a9df7a0, 0x9bb41bfc, 0x24df7a8a, 0x8dd8bfe7, 0x3124dbd6,
+	0x23674aff, 0x1d22a78f, 0xb6f6fef1, 0xd60b93d7, 0x02af7b40, 0xae3c815f,
+	0x29be462d, 0x935c3332, 0x457cec91, 0xca5da5eb, 0xf857d20e, 0x5c87c2f7,
+	0x86e117ec, 0xfdba4f3f, 0xbeef3a61, 0x073dfa1a, 0x02d8e5c6, 0xf74beae3,
+	0xff8c1373, 0x40a71e05, 0xdee0d2aa, 0x068bde6c, 0xc79bec0d, 0xbf03bca8,
+	0xa7e0e3eb, 0xb94fd34d, 0xdce7f63f, 0xfc19bb1f, 0x927a69ba, 0xe7482dca,
+	0x842560fc, 0x64c8e7f2, 0x95f2235b, 0x883bf82e, 0x21e726ef, 0xb4ebb31e,
+	0xd7678fb8, 0xfae7227b, 0x377f65d1, 0xb59a1f16, 0x5848cbe5, 0xef666d4f,
+	0xed126cbc, 0x73b71c6a, 0x0736efee, 0xb26ef5e7, 0xe7cc24ed, 0x01c764f5,
+	0xa3feb76c, 0x753f04ff, 0x7fde3eda, 0x4f1d4bb4, 0xf102950d, 0xa37f9dea,
+	0xe39fcd9e, 0xd3545379, 0x69bfa80a, 0x2cc0774b, 0x2c7ee2c9, 0x0527e804,
+	0xe7b33fc3, 0x0c74378d, 0xf8d17380, 0xff50728b, 0x3fef3131, 0xf38b3a0a,
+	0x286ba70e, 0xcf9287ba, 0x80cc78de, 0x442e1377, 0xe22f5c82, 0xbafa9376,
+	0x6fd0079e, 0xe04caf7f, 0x27ca23dd, 0x498efe0a, 0x48807a66, 0xf8f51ab3,
+	0xc6a6cf5e, 0x36ce7aaf, 0xce7a1d21, 0xe4e5e5a1, 0x507a15eb, 0xff1fe0d6,
+	0x2c58a7fc, 0x92412f0f, 0x42aac8bb, 0x629af7b8, 0x4c36f9dc, 0xb9b99a61,
+	0x66bdee10, 0x207b4a24, 0x1f49699d, 0xe25df705, 0x232b38dc, 0xecb00f6f,
+	0xca116b58, 0xc4583ce1, 0x814bfb7e, 0x1eee1c31, 0x39a170bc, 0x92df0146,
+	0x7f28ac98, 0x2287a5ab, 0x8dc64c7f, 0x9faf7c8a, 0xefbfb5bc, 0x47c0777b,
+	0xc3b07747, 0x9ca079d3, 0xb4656b21, 0x396d66db, 0x148ddb74, 0x4eeeb71e,
+	0x9b036c2b, 0xcdb8fe78, 0x4fd82e05, 0xb00a5de1, 0x3ee4b463, 0xa09fd4fa,
+	0x5f7b2d31, 0x74611e63, 0xebfc6b28, 0x799b6028, 0x3da1e6c3, 0x8c9de036,
+	0xd5dfc135, 0x52c1f610, 0x65e9ecc2, 0xbbf6d41b, 0x550ecc95, 0xe54faca7,
+	0x3c827fd1, 0x0ef5bbea, 0x6e97f833, 0x3763fc1a, 0x64d6e790, 0x284bbc65,
+	0xbff2f17e, 0x1fe03964, 0x17e6a1e7, 0xe9120761, 0x6e2c83b8, 0xc5e56ae1,
+	0xf85e403f, 0xe57917c5, 0x31fb9e98, 0xc41278f1, 0x3f1f0f47, 0x628e4b4f,
+	0x5aaf67e5, 0xd0b8ef0a, 0x13c4277c, 0x4b9db291, 0x85efaec1, 0xf32b3617,
+	0x2996941f, 0x95d387e0, 0x65bccaf0, 0xad43f5fe, 0x66fa8abb, 0xefca83bd,
+	0x39d8278a, 0x639f7024, 0x992ee8eb, 0x6fdb0554, 0x4a31fe44, 0x7853f40b,
+	0xc70bed83, 0xfef47aba, 0xc96527fe, 0xbb3b4f4c, 0x7ddff1ed, 0x6de27607,
+	0xce46fd04, 0x5cfb3146, 0x8d29dd76, 0x177ae406, 0xb36caf86, 0xec7e0347,
+	0x4ee1c5ed, 0x9cac83f0, 0x24fe811f, 0x866ce17d, 0xbfedc9e7, 0x7843d1b7,
+	0xbf09efe5, 0x03ff0f57, 0xd13f379c, 0x46fb1468, 0xff9589bb, 0x6cede48f,
+	0xeb67c8df, 0x83d03a63, 0x974261cb, 0xe252e566, 0xeeb0cb95, 0xbf003e60,
+	0xf78edcf6, 0xd37b009f, 0x5a3eb63e, 0x29e3fb89, 0x4cd8bfb8, 0x95a3ebfe,
+	0x7e77b2d4, 0x427dec2d, 0x7218bf71, 0xd1aeecbd, 0xbf8dfbc3, 0x2279baf4,
+	0x843bdf99, 0xc7071f5e, 0xdca93e50, 0x3b1afcc6, 0x4aefae1b, 0x804bd6fc,
+	0x068cf69e, 0x6dc05af9, 0x3bfdfc0d, 0xd8a17c55, 0x17b58837, 0xf78b13fa,
+	0x3ae80f3e, 0x541fa026, 0xc6fd16a3, 0x092ce816, 0x6fea9751, 0xfc5886b9,
+	0x68dbebfa, 0xe62df940, 0x73666ed1, 0x3e78dbed, 0x8171de1a, 0xc3ea3174,
+	0x87fd8a35, 0x58ba73ee, 0xec455810, 0x6370ec5e, 0xe1d03e42, 0x199c44b0,
+	0x309a9839, 0x35fb9bce, 0x27fc244d, 0xffeb4331, 0x6fd0b6ae, 0xf3bc6cef,
+	0x3e87d07a, 0xcee7bbae, 0x24fbf401, 0xe29b9778, 0xfb7740f7, 0x81df9834,
+	0x73f4255f, 0xfa91fc2a, 0x77f7acfc, 0xf307e50b, 0x65dd80c6, 0xfe83312a,
+	0xfa665e3c, 0xad8b7916, 0x2a37212b, 0xe868a6a5, 0x09e4f5ff, 0xbce99287,
+	0x01fd1613, 0x3df569f5, 0x5bea95f3, 0x26b0f285, 0xebede81b, 0x6e7f1625,
+	0x31577c65, 0x9393ccec, 0x7d508e40, 0xf28ff41e, 0x1ef89dfa, 0xe6dcfa81,
+	0xe7d5abd5, 0xf38dabef, 0xedcbd00f, 0x7071c96d, 0x7877388e, 0xcfd40912,
+	0xcd56f27d, 0x021f5477, 0xdc36b479, 0xa6aef8c5, 0x6fce94ad, 0xb364dac1,
+	0xf6e78a13, 0xa2d6f162, 0xdba7f0fc, 0xcbf3437e, 0x83370a5e, 0x9d30bdda,
+	0x818de138, 0xa37aab85, 0xdeafbbbd, 0xd006eaf2, 0xa1d65eaf, 0xe06bf45d,
+	0xbd7f0a1d, 0xd81706f9, 0x2c69e783, 0x8cf7847f, 0xf9193ac9, 0x060bd4ed,
+	0x1fe75fdc, 0x2bfb81e9, 0xa50b23ac, 0x4254a9cf, 0xfcdeb4cc, 0x5db193d7,
+	0xaf7625f9, 0x02537f3b, 0xfa0aa1c6, 0x8610bcb6, 0xd9c591f1, 0xf4dcbdb5,
+	0x47449edb, 0xbc1c4a54, 0x975ae7ff, 0xf054f144, 0xc01f4fd2, 0x537f36be,
+	0x47cbd21c, 0x61fbe0e7, 0xb60f4367, 0x84be39ce, 0xef05fedc, 0x0653120f,
+	0x87895f82, 0xeece77bf, 0x7db8720f, 0xd977cd4e, 0x7e32ffbc, 0xd840c3ce,
+	0x67cff058, 0xdca24cf1, 0x628d14a8, 0x4f942d3f, 0x3e076c54, 0xf9e037ad,
+	0xb82c35b2, 0x0a486a9d, 0x8c2b57db, 0x49ec2adf, 0x507e8b15, 0x322b13d8,
+	0xe7bbde09, 0xbff4b12c, 0x71f7a697, 0xd9c4a7fc, 0x4a7cc387, 0xfadf1f6a,
+	0xf9f1e21e, 0x7e813ffb, 0x3bd89af7, 0x7c31ef6f, 0x6f02ff0b, 0xbe0ec67b,
+	0xf429570e, 0x476db6cb, 0x7f781db2, 0x382bfc77, 0x3c6d9f0f, 0x25a654f2,
+	0xc7cc7ec8, 0x658cec2b, 0x60fede30, 0x4b8db487, 0xb473b3f1, 0x5efc1d7f,
+	0xeeaff44b, 0x53646cf7, 0xb6dbbc36, 0x07698200, 0x46cff2fa, 0x3ec36537,
+	0x4437fc2b, 0x14cccedc, 0x4a947fee, 0xfbc44352, 0x0b8949cc, 0xdf83ef1a,
+	0xe7de3aaf, 0x19f78b3c, 0x7de333fe, 0x0c9f393a, 0xef1e0988, 0x5fcf04f2,
+	0x43d96a42, 0x43ee5c72, 0x077f0dd8, 0x67bc5658, 0x8899e5c9, 0xc0da18f3,
+	0xb943ffec, 0x6364ab1f, 0x88432e71, 0x7aa7b59b, 0xe5640597, 0x89119054,
+	0x113ae237, 0xc606fdef, 0x7be13425, 0x0e9337cd, 0xacc46d4c, 0x1ba29485,
+	0x8ec491d3, 0x855228e9, 0x835268e9, 0x16a46de9, 0x09a4b1d3, 0x8dc7d253,
+	0xff78c7ef, 0xf3adc67e, 0x0cffb034, 0x6f5c1498, 0xbfaf837b, 0x5243e04b,
+	0x9578afbc, 0xd84518f8, 0xf121c5fb, 0x7e819292, 0xea2f0258, 0x9c31bfbf,
+	0x119db078, 0x80f651b0, 0xef0594f3, 0x0d880a4f, 0xc42bc2e1, 0xa1c7ae47,
+	0x2532f665, 0xe0e9025e, 0x1bd4854c, 0x19ee4607, 0x2b038ca7, 0x5e718719,
+	0x5cec39fd, 0xe26ff2c0, 0xb1f79de5, 0x2824507c, 0x82ca45fc, 0x5c71fe71,
+	0xef190f72, 0xb187261d, 0x6ef17ab8, 0x4230e4c9, 0xb5bf809a, 0xcf9438ce,
+	0xde297a8b, 0x392530f3, 0x355ca00e, 0x142d23ba, 0x16670b7f, 0xb1c6ffa2,
+	0xdb8dcb38, 0x7e3c5f25, 0xd75b0a4b, 0xc4aaa547, 0x8f1e743e, 0x77dfa7e4,
+	0xbfafba7f, 0x04e7e82b, 0x3f755bac, 0x3c3ac46f, 0xf6313fd0, 0x2cfcf3c2,
+	0xde6cafbe, 0xaadb343c, 0xaccd7e81, 0xdd67dfdf, 0x32cd3f77, 0x424edaa7,
+	0x821496f9, 0x5da44deb, 0xf02dece2, 0x893bc45d, 0xbf33edef, 0xbe26eadf,
+	0x531faa38, 0x5ca1671c, 0xd693c6a5, 0x571d0e5f, 0x5c46c4b6, 0x655be47a,
+	0x2e3a993c, 0x3df7b68b, 0xee52abcc, 0xd3fc0b98, 0xfe91f516, 0xcaefea2c,
+	0x1dc6acb8, 0x687114fb, 0xe3781d62, 0x1c593ad0, 0x438850da, 0xbff1f48b,
+	0x7c134388, 0x10c3d62c, 0x72c34388, 0xf281cc37, 0x813a3fc3, 0x4e236871,
+	0xda1c7878, 0x9a1c6511, 0x159ffbed, 0xb1de2687, 0xbd85b058, 0xdb80bfa7,
+	0xfd060077, 0x5f7894e2, 0xc4579f86, 0xdf7ca4b9, 0xdf9eecdb, 0x041bff49,
+	0xa7ec907e, 0xe859f9f8, 0x0fe0122f, 0x0cb5c279, 0x0fe01dc6, 0x09aff2a9,
+	0x568277fd, 0xe9fbf506, 0x2bcfca93, 0x5c219f7c, 0xcbc7997f, 0xaec0fcc0,
+	0x8f51a0b5, 0x1782d1fa, 0xfe82b3ee, 0x64d520b4, 0xf2b77945, 0x514be43e,
+	0xfccf3c9e, 0x9b7ac473, 0xdea09cbf, 0xac4f3de5, 0xeff0d833, 0x404bac23,
+	0x3e7bcafd, 0x0814f6e7, 0xebfb84d3, 0xdd1d514d, 0x536ddee1, 0xf4158663,
+	0x5b220fbb, 0x71b14256, 0x5e9f0ce1, 0x0e98797d, 0x9603aff5, 0x74888298,
+	0xf3b13adf, 0xf99e4275, 0x3af98335, 0x7bb136b3, 0xf6a39c85, 0xdc0b4882,
+	0x84fbb02b, 0xefc12d41, 0x4ae7c7a9, 0xc08a7b8a, 0xcb16c9f7, 0x7fb8255d,
+	0x6a10f760, 0xbfb8bd3b, 0x2985b22e, 0xdd423fee, 0x0ef38b67, 0x1e89efdd,
+	0x118d7bf6, 0xc48337fd, 0xf7233dd3, 0xf663bdf0, 0xf3c0e48e, 0xa6c61dd0,
+	0x4fabe034, 0x53731c63, 0xcb6e20e3, 0xce39416a, 0xf17ceae3, 0xbca06f40,
+	0x073ef114, 0x20e32da5, 0x54deea1e, 0xa3d40547, 0x067b30d3, 0x9947def9,
+	0xc5f27b99, 0xa7682dd7, 0xb7f54caf, 0x3bbff3d3, 0x1e2f5ea2, 0x59e3058d,
+	0x5095bf9c, 0x8d0338af, 0x15f3c60c, 0x8a674ff2, 0xdbf98729, 0x4adefeaa,
+	0xdfe5576c, 0xf167a95b, 0xad89a50d, 0x59b3af4c, 0xfa6e2cad, 0x959dfde7,
+	0x9553ddcb, 0xf428b626, 0xe621b255, 0xbea6ec8a, 0xfdae5df9, 0x5a7f6c36,
+	0x7e0a87d5, 0x98e8b198, 0x3a66cf82, 0xb5f9959b, 0xaf7838d2, 0x1567d28c,
+	0x35b3c7d0, 0x0b14d469, 0xf01801f8, 0xf7cbbf5d, 0x77f11a7b, 0xcf09baa2,
+	0x9f5e628b, 0xd8451f55, 0xc651ed53, 0xa979f067, 0x704f8bfd, 0xdc1399bf,
+	0xf704e66f, 0xfdc1399b, 0xbf704e5a, 0x7c87e116, 0x4c8aa242, 0xe47f5074,
+	0x828f8137, 0xc23723f8, 0xc8fe90f7, 0x0a71c7e4, 0xb7e7647f, 0xe47f0b08,
+	0x91e7a95d, 0xec9bbb8f, 0xc43b436f, 0x9ffbd1e6, 0x23f943bb, 0xeeca9fbf,
+	0xa5503e93, 0x22f3d29f, 0xb4f1a9fa, 0x0bd95e50, 0x37ac59df, 0xcb879010,
+	0xf1a6d924, 0xb0f18fbf, 0x16ce5377, 0x1dfc28a5, 0x8b7f962a, 0x0f7e17f8,
+	0xd0196f39, 0xdf7ca04f, 0xd7405ec1, 0xe71d53c7, 0x1dfd9952, 0xab2e87df,
+	0x5fbfb033, 0xbf21cfc7, 0xebf139e0, 0x718181d4, 0x77e6a114, 0x53d34e76,
+	0xe55e7499, 0xef153b91, 0x8af18b22, 0xfa718979, 0x5f8cb9da, 0xc50ef2c9,
+	0xf6db6b3a, 0x63e02f48, 0x9c167ef1, 0xa904275b, 0xe8c69f1f, 0x5287cf42,
+	0xa4bfed1a, 0x7d0f9d8a, 0xffb9adf4, 0x3ec7a013, 0x509dca13, 0x2cf3d5f5,
+	0xf10fb1df, 0xd2173ea1, 0x0b4d7679, 0x3b9fcf32, 0x6bf5abbf, 0x4e217437,
+	0x715097e0, 0x9b250751, 0x2d77c605, 0x4b5e96bd, 0xd2d7a5af, 0xf4b5e96b,
+	0xbd2d7a5a, 0xaf4b5e96, 0x6bd2d7a5, 0xf4e5ffe9, 0xfffd007f, 0x8000c102,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x30001131, 0xafb00408,
+	0x521cae88, 0x11447fea, 0x992c9a42, 0x326ebaf3, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0x7f6db6db, 0x98a102fc, 0x80005382,
+	0x00008000, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0x00002000, 0x000040c0, 0x00006180,
+	0x00008240, 0x0000a300, 0x0000c3c0, 0x0000e480, 0x00010540, 0x00012600,
+	0x000146c0, 0x00016780, 0x00018840, 0x0001a900, 0x0001c9c0, 0x0001ea80,
+	0x00020b40, 0x00022c00, 0x00024cc0, 0x00026d80, 0x00028e40, 0x0002af00,
+	0x0002cfc0, 0x0002f080, 0x00031140, 0x00033200, 0x000352c0, 0x00037380,
+	0x00039440, 0x0003b500, 0x0003d5c0, 0x0003f680, 0x00041740, 0x00043800,
+	0x000458c0, 0x00047980, 0x00049a40, 0x00008000, 0x00010300, 0x00018600,
+	0x00020900, 0x00028c00, 0x00030f00, 0x00039200, 0x00041500, 0x00049800,
+	0x00051b00, 0x00059e00, 0x00062100, 0x0006a400, 0x00072700, 0x0007aa00,
+	0x00082d00, 0x0008b000, 0x00093300, 0x0009b600, 0x000a3900, 0x000abc00,
+	0x000b3f00, 0x000bc200, 0x000c4500, 0x000cc800, 0x000d4b00, 0x000dce00,
+	0x000e5100, 0x000ed400, 0x000f5700, 0x000fda00, 0x00105d00, 0x00000028,
+	0x00000000, 0x00100000, 0x00000000, 0x00000000, 0xffffffff, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x00088b1f, 0x00000000, 0x51fbff00, 0x03f0c0cf, 0x65e21f09,
+	0x63e62860, 0x88237860, 0xcc2b4e2a, 0xfe9942ce, 0x0c0cccf3, 0x32f88117,
+	0xe2055f10, 0xe9a48cd3, 0xb045e2b7, 0x30327377, 0x7df90358, 0x9b8b5a40,
+	0xc8014181, 0xb3e201b6, 0x204bfe40, 0xadc40afe, 0xdc0c0c3c, 0x6a0c0c5c,
+	0xc4042c40, 0xcdf8bcb6, 0xff2023b7, 0xaf951b9f, 0x17ca83cd, 0x3fafc6e6,
+	0x7cbf0789, 0x6c790106, 0xf928b3f8, 0x4620e1f1, 0x2d43749f, 0xca86aeac,
+	0x6065522f, 0xe7c40df8, 0x681ae2a1, 0x10aac5f2, 0x03329cfa, 0x7e1ab243,
+	0xc80853b3, 0x000c060f, 0x4022bae9, 0x00000400, 0x00088b1f, 0x00000000,
+	0x7dedff00, 0xd554780b, 0x733ef0b5, 0x27bcce66, 0x20212793, 0xf0841e4c,
+	0x04242074, 0x11093a8c, 0x5076c403, 0xc2ab16fe, 0x25786784, 0x5ae5a911,
+	0xc0133bff, 0x51b91688, 0x7e2da5a8, 0x68bc104e, 0x01226f69, 0x903a4483,
+	0xbd08a5c0, 0x168b6ad1, 0xe088786d, 0x7e929205, 0xcfe956de, 0xe7dad6bf,
+	0x9939cccc, 0x77e8f881, 0xbf41ffbf, 0xdecfb3ba, 0xd7b5ef67, 0xcfb5ef5e,
+	0x30733d1e, 0xd7632776, 0x73941ff1, 0x4158c645, 0xf81d0ca4, 0x614b3ce2,
+	0xc18b72ec, 0x6c64cc65, 0xe8431eca, 0x68afa84e, 0xd7588214, 0x3633301d,
+	0x5b2bb181, 0xec46bd79, 0xb41b78dc, 0xb645d7b3, 0xd3b60a03, 0x8c8563a6,
+	0xfe31379d, 0xb1d76f4f, 0x9916c634, 0xe3457579, 0x7e3839c1, 0x9991ab55,
+	0x2df3fde1, 0x2beb07a2, 0x26410877, 0x27befb40, 0x7f180ca5, 0x32685071,
+	0xe3bea0bb, 0x6765c959, 0x2f81cd6c, 0x0abce0ae, 0x4dfa9f5c, 0x7c2125cc,
+	0x8d75a5ef, 0xb0696c66, 0x8c066a79, 0xe09774b3, 0x4b3cc1d6, 0x3ba10ab7,
+	0x2173e027, 0xcbb8c228, 0x2b995a76, 0x9d70c38f, 0x8eb4bcd1, 0xbcc02f95,
+	0x83670899, 0xe647b5e4, 0xa9f3f053, 0x72b981bf, 0xaf15f523, 0x5ef03cc0,
+	0x3704aff5, 0xc75ab12e, 0xff304d7c, 0xd9e67d95, 0x7d70d92f, 0x3dae5275,
+	0x7b951d7a, 0x834b1d71, 0x1b54d28d, 0x2d28d5c2, 0x6f191d1a, 0xcc4dced9,
+	0x6b828a65, 0x5f00f64b, 0xfcb6bb54, 0xf054cfd4, 0xd3cc652c, 0x83f84364,
+	0x9131ccd2, 0x22422de7, 0xbe38a963, 0xb19f1aa8, 0x99173d00, 0x6778843b,
+	0xf477766d, 0xaa755dbe, 0xa2656df7, 0x781c6efc, 0x87de45fe, 0x2a4efd67,
+	0xa9d3f937, 0x45dfb1fc, 0x9fc4fe70, 0xfd4fe547, 0x73fe7a6e, 0x2f95117f,
+	0xbe543df8, 0xb2a72fe8, 0x7ea5efd1, 0x5367f92f, 0xa3efc3b9, 0xafeaffe7,
+	0x7f15f2a2, 0x85ff3d2d, 0xbf95357f, 0xf9e807f4, 0x53d7f9bf, 0x04fb8cd9,
+	0x16ff6ee5, 0x83f8f72a, 0xff75efd4, 0xf9f72a4e, 0x33f9e89b, 0x13283f88,
+	0x3d3ace3c, 0x19424fec, 0xb0790273, 0xceb7af64, 0x59e4f500, 0xd7c03ebc,
+	0x7e012750, 0x3a802c06, 0xdaa0dfec, 0xb4233ace, 0xdbc55a0f, 0x6b9c0687,
+	0x743ed04c, 0x6ecf6f1d, 0x4331aef0, 0xde66f67b, 0xb0d83c3e, 0xc3ed02c6,
+	0x51f6f3b7, 0xac6b9d4d, 0x85aa3ed0, 0x721adfb7, 0xfd41b5ae, 0xcf5e0ed6,
+	0xb5aef4ef, 0x5dff3d43, 0xd5d89f5e, 0xef01d6b0, 0xdfc73c4f, 0xd9bde2d7,
+	0x60cf9ce0, 0xee0736af, 0x4207e8f2, 0xc23e725d, 0x1cbd8cdc, 0x77d516f8,
+	0xded4dc1b, 0x47aef81a, 0x92ea093f, 0x058fda9b, 0xc7bed47c, 0x53f6a5e0,
+	0xbed42581, 0xfb52f247, 0x6a4ac095, 0x4b50dd7f, 0x87eeaced, 0x54bafed4,
+	0x7549ed4b, 0xcfbea8eb, 0x8c196d00, 0xe33dd4e5, 0xcdfd0009, 0x511b7c11,
+	0x6aeb747e, 0x12ca18ab, 0xf91d287d, 0xcd8149e9, 0x5d80fe46, 0xf27bba23,
+	0x7c83c916, 0xc29bfcd7, 0x994cc497, 0xbeb052d6, 0x05a7e5e8, 0xd827e71a,
+	0x71825baf, 0xce3e6fb2, 0xaa4563b8, 0xab1dc671, 0x1209c652, 0x8dfe963c,
+	0xb26c7cd3, 0x4b639e1a, 0xa15e7195, 0x5bfd1c71, 0x8ab7b8d7, 0xb5bcf0d5,
+	0x35e7195c, 0x7fa9271a, 0x1b7eecf0, 0xc64423f0, 0xe036fdd9, 0xd9c69327,
+	0xf5a4e34f, 0x77fc9e0f, 0xe4e3548a, 0x8ca553bf, 0xce7841d3, 0xbb38dfe8,
+	0x86ac99df, 0xcaa59de7, 0x9c682738, 0x575bfd21, 0xd58aeffe, 0xb96efcf0,
+	0x69efce32, 0x8dfeac9c, 0x357de7b3, 0x3f79ecfd, 0xa27f3f4c, 0xb7fb7271,
+	0xd40f82ae, 0x41f053f4, 0x102e7e98, 0x6ff6479e, 0xa81f3d9c, 0x07cf67e9,
+	0x33f9fa61, 0x7fb633c1, 0xa3f82aeb, 0xfe0a7e9a, 0xcf9fa618, 0xfdf19e09,
+	0xd3f5e783, 0x7ebcfc6a, 0x511f8c3a, 0xfb0a7840, 0x33c4cf07, 0x3c4cfc6a,
+	0x8d8fc613, 0xbfdc99c6, 0xa33f5e71, 0x67ebcfc6, 0x1549f8c2, 0xf4775d70,
+	0xcf135d6f, 0xf133f1aa, 0x433f186c, 0x19529e08, 0x8ce3061e, 0xbece3f89,
+	0xecfc6a8b, 0x33f1c8bb, 0x3a27d3be, 0xce90dda0, 0xe4ce08dc, 0x0080de0b,
+	0xa814faed, 0x30f4b0bb, 0x3d3bc81f, 0xf8d1ddea, 0x84405772, 0xc4aefb76,
+	0xa2bb4ff1, 0x58e96ff5, 0x8263975c, 0x3877654e, 0x7aaa2d8b, 0x4a925952,
+	0x6454a73f, 0xb369eaa8, 0xf4f554b2, 0xdeaa4707, 0x5e3058cf, 0x7cbc1f55,
+	0x643eaab2, 0x7deaa955, 0x55d3e3d7, 0xeeeb59ed, 0x3673d555, 0xcf554f7c,
+	0x554f3cdd, 0x54badbcf, 0xedc8de35, 0xd1f5552b, 0xd5531ebb, 0x51acb6c7,
+	0x4f6dddd5, 0x3be3eaab, 0x3f8d539e, 0x54cff8e1, 0xb777c2f5, 0xa745eaa9,
+	0x3fbd555e, 0x1a6bdcf9, 0x3ce6c6fb, 0x8bae4a3f, 0xc95acf68, 0x910e33ce,
+	0x192d5ec8, 0xa3faa262, 0x7f546c43, 0x70df0ef4, 0x8fe3ec27, 0x6bd005f0,
+	0x5bbe4a8f, 0x2fec319d, 0x4b1d39d8, 0x63a7d2c7, 0xd7db7d45, 0x80ce801b,
+	0x5e874a2e, 0x650d3dcb, 0x3727ae8d, 0xe8dfca11, 0x2e916a4a, 0x4474151f,
+	0xd591567f, 0x8d58dd22, 0xfd70e88c, 0x0297c2b0, 0x54f87451, 0xcaf39726,
+	0x6a194fb8, 0x3eef51d3, 0x0066f756, 0x9993f2f5, 0x664cbc60, 0x9bc287ec,
+	0xbc0e40a1, 0x3a68f99f, 0x09268bda, 0xf9e8d0ec, 0x347f9825, 0xf9c01fcd,
+	0x6fcce3eb, 0xcd522dca, 0x52aace6f, 0x60966fcd, 0x4f63277e, 0xe7961dd7,
+	0xffde9137, 0xe6987091, 0xa6ad6537, 0xac4bd7f9, 0x25bf354a, 0xf2037f3c,
+	0x7f38f3c1, 0xe7f58c41, 0x3faf564e, 0xfd7aa96b, 0xcfff5f12, 0xf3c84eea,
+	0x3ffd685b, 0xf5f04e17, 0xd7c63d67, 0xa371846f, 0xf18477e7, 0x5ff9c41f,
+	0x69bf338e, 0xfd7ab178, 0xebd5cbd9, 0x5ff9f237, 0x9e4f7bad, 0xfff346df,
+	0xaf8f7842, 0x9a71fb3f, 0x75ac68df, 0x38c9a13b, 0x441b52d4, 0x2fe600b9,
+	0x050c801b, 0xc6444cf3, 0xde0c6248, 0xd164285f, 0x016a3c3e, 0x9ffe83ba,
+	0xd0de8059, 0x72ed977e, 0xd017ac16, 0x6996af0e, 0x4aaec17a, 0x657819e9,
+	0x57f72b41, 0xc333f818, 0xad2bbe00, 0x11fd9fbd, 0x7db05de0, 0xa025c91c,
+	0x2def8674, 0x23529794, 0xe6307c72, 0xe413321d, 0x877bd329, 0x9a863bdc,
+	0x2fd7683c, 0x1f630466, 0x60d85e3a, 0xc515df38, 0xa8b78b6f, 0xea824beb,
+	0x097702ba, 0x81940912, 0x0b579652, 0x7f079ef3, 0x7ddd8180, 0x2c6dec06,
+	0xb9b0313f, 0x0e7e785e, 0x452605fc, 0xed1d55f3, 0xabbd79cb, 0x73d61d3e,
+	0xc3704abe, 0x0d81ec88, 0x3a196395, 0x7bcf0172, 0x77ef0039, 0x852ff298,
+	0x7273e9d6, 0x003617f8, 0x7ff1477f, 0xb0bbc88d, 0x0ee08517, 0x7fbfe790,
+	0xba054573, 0x83d7d0bc, 0x5e577def, 0xc92de2f7, 0xde98f7aa, 0xa9cdef44,
+	0x1ea20c4c, 0xc36ceef8, 0x3b552eb9, 0xf1c03f30, 0xdd0fec91, 0xf1eebacf,
+	0xdfb99e82, 0x75d7eea7, 0x8dd1eac0, 0xbb4b7285, 0x3dbec181, 0x8719a550,
+	0x1f7f6879, 0xa878e192, 0x973edd56, 0x5fd42539, 0x9fab5773, 0xe3b40657,
+	0xcf00f800, 0xf81c793c, 0xe6f844e7, 0xd7f36dcb, 0x7daa021b, 0x0a947a02,
+	0xa22cf6e5, 0xbd0f5a02, 0x3fb7dd9f, 0x434041d0, 0xfb588107, 0xf9f7fffe,
+	0xae7d8dff, 0x3e492d69, 0x8ad70517, 0xe7c969a6, 0xdb2dfaa2, 0x5f555339,
+	0xd55fbc12, 0x4a96f17f, 0x86c2fb55, 0xf9f6aa25, 0xd5561feb, 0xa4ff032b,
+	0x75773fea, 0xa1fdaa9d, 0xed54a7da, 0xab3d540f, 0xfbefdfaa, 0x77ffaaa9,
+	0xdaaa3767, 0x864b0385, 0x46a8dc62, 0x881d0df2, 0x8df7aef1, 0xa1109b6c,
+	0x2fae855b, 0xfca9d3f8, 0xb2a2efd4, 0xc830e50e, 0x5e3edd75, 0x1cab6cd6,
+	0xa15bf28a, 0xb50c4ff3, 0x69fe5043, 0x672b55e9, 0x6325d0fb, 0x88bed083,
+	0x2e596717, 0xf3ca2f30, 0x19db816e, 0xb213f3ca, 0x0d6c675e, 0x8972d7f1,
+	0xd7f2dff3, 0x3f229606, 0xb942972c, 0xe3f8b6b5, 0xd68126b2, 0x1a36d8c1,
+	0x10704967, 0xfbd17918, 0xf382ad9a, 0xe69141d2, 0x6e1f46af, 0xdec1da12,
+	0x3f9c3a0d, 0xbee43ed4, 0xe78f0831, 0xc1c01d19, 0xdf49535a, 0x2797f25d,
+	0xfe23d39f, 0xf05e54bc, 0xbcf2a6cf, 0x39e547df, 0x6795157f, 0x7654b5fc,
+	0xf95357f9, 0xe5403f91, 0x2a7aff29, 0x5037f03f, 0x85bfd279, 0x83fbdfca,
+	0x7bf15e54, 0xf4e854a8, 0xb57d96ff, 0xf2bdb8e8, 0xffafb8df, 0x16afbd1e,
+	0x23c3939d, 0x0d06ebef, 0x55794abc, 0x61797027, 0xf31676e4, 0x788e2958,
+	0xc3078955, 0x94571448, 0xf0ddc1e3, 0xd3f62777, 0x1831e537, 0xcfe8eaef,
+	0x7b8e3173, 0xb191a4f5, 0xd3ebf8cc, 0x3c5c5fee, 0x46ee80b0, 0x3858b7f8,
+	0x72aa015e, 0xdcec1e36, 0xf3dd7b11, 0x92cf5d78, 0x7726fdf0, 0x3c355c74,
+	0xc072e22e, 0xd16af1cd, 0xdacf7b71, 0x847c722e, 0xb3efbb76, 0xc6b5e847,
+	0x718b93d8, 0x2347f662, 0x36c63d43, 0x055871cb, 0x575e015f, 0x3a7ff5f0,
+	0xfda04e3e, 0x1f1f105b, 0x47e0a5dd, 0x32bff0c4, 0x574aafc5, 0xae90439d,
+	0x7bf4cab6, 0x027865b5, 0xe7191d1b, 0xd38b9d6e, 0x7ef04df5, 0x0e4be8f5,
+	0xa657f3b4, 0xf2e79ceb, 0x456fe3e3, 0x16d3a653, 0xaab4a76d, 0x841b57eb,
+	0x70371e72, 0x831acc25, 0x3ce9d1f2, 0x2c728397, 0x9c654a9d, 0xf392e995,
+	0x0bf6325c, 0x2576f872, 0xea7e5f12, 0x6fbf5137, 0xe3788e1a, 0x6168e044,
+	0x247de167, 0xf1d9ebd1, 0x3ba5c7fb, 0x8c245ffd, 0xc336b0c7, 0x985f6867,
+	0xd1d7ffe0, 0xaa7a63ee, 0x34d6e32b, 0xade30189, 0x6a977cf8, 0xa5cf783e,
+	0x67ae94e4, 0xf5be88dd, 0x7f42bcec, 0xa0e7ca20, 0x1079489d, 0xbf6d53b3,
+	0x35eaaa24, 0x2a9d03ad, 0x00ff067b, 0x04cac739, 0x72c1d23d, 0xf5ba2c99,
+	0xc002f5b6, 0x2e9d0937, 0xf78147a4, 0x92ea615b, 0xafa42e7c, 0xcdb9dfb5,
+	0x681b3d85, 0x60f85abd, 0xf5ea66ff, 0x265fe1f1, 0x1bd47d5e, 0x22d7fe83,
+	0xd01ea3fd, 0x999f3f71, 0x43e40e02, 0xdc701e96, 0x6cc3fc4b, 0x9ae0fc61,
+	0x5513cc19, 0xff01c657, 0x5f6d47ad, 0xa4bbbe72, 0x76bbfee3, 0xbcf2cdde,
+	0xf55fc337, 0xe2a7b7ae, 0xac78b5ba, 0xe4b3fd6e, 0x5314fca3, 0xefc151e1,
+	0x59bbd124, 0xee97f984, 0x7f212d3c, 0x7204305d, 0xb903787f, 0x5b7e17df,
+	0x646b79fa, 0xd8f72b9d, 0xf95fbec9, 0x0bd79cac, 0x779c4eb0, 0xf13c915e,
+	0x0515e835, 0xa67c6807, 0xd3d88c2b, 0x5e6d7e4a, 0x27f232cb, 0x0901663e,
+	0xc51792e3, 0x9fb44286, 0x4a353697, 0x32f3801e, 0xbe2395d3, 0x1c77b32e,
+	0xb5fb51f9, 0x9f08944f, 0x899f6760, 0x6cec1386, 0x6cec1d55, 0xbb117d55,
+	0x51f30d37, 0xc01b368b, 0xaa75a7c5, 0x9d115fda, 0x5662ee34, 0xc9768f5b,
+	0xeffc96d9, 0xf0fb86dc, 0x88850bfa, 0x3cd079df, 0xf8f10928, 0xcbf1e136,
+	0xdf04bfce, 0xde77df1f, 0xf26add08, 0x561e328e, 0xc13c14ec, 0x44737418,
+	0x7d0ad796, 0x5f8641fd, 0xc59c0f89, 0xdea4417a, 0xa17fa9ad, 0xfe657fa9,
+	0x4bba014d, 0x24f7ec8b, 0xe91677d1, 0x38f906ad, 0x5fb7c221, 0x3ef4fc79,
+	0xa37f77e3, 0xf3560fe3, 0xc5ff8c0d, 0x4e7e3a37, 0xfa8df81f, 0x77e3be82,
+	0x59631d9b, 0xc4a9d902, 0x3a71815f, 0x1fe9bbf2, 0xd2136bda, 0xc0a69fc9,
+	0x39ddd638, 0xd9a38c41, 0x14e99dcd, 0x54bf07d7, 0x6024f30f, 0x0edf907f,
+	0xf9f016e6, 0x2cd37ed3, 0xf8ca8794, 0xe55eb776, 0xe914267f, 0x7b28fe35,
+	0x61f00737, 0x38a6a793, 0xfcedd41b, 0x5b6b666f, 0xe30b7c4c, 0x04e8fca8,
+	0x2ffe45d6, 0x9e7aa78c, 0xa1d59a38, 0xa8f44a1a, 0x6d154f9f, 0x9c4b6f14,
+	0x9c4e28c6, 0xfad7ae5c, 0xd2c315f7, 0x4740e0a2, 0x32c9ea1b, 0xcfa6e01c,
+	0x67fde014, 0xf08cf4cb, 0xc5ff8012, 0xbc593380, 0x01f9c216, 0x39df155e,
+	0x65f540f2, 0xe7aabc11, 0x4d19e700, 0xe50f207c, 0xb8b31246, 0x53801e71,
+	0xdbd1eed5, 0x0e1e2853, 0xa3e517f2, 0xc989d4cd, 0xaaef9c7f, 0xf4ebaf0a,
+	0x190b23e9, 0xfc385085, 0x942fa601, 0x83ef8bf7, 0xd313bfaa, 0x8f04d9d4,
+	0xd8361c7a, 0x6f8f2d19, 0x8309fd95, 0xe9133bc1, 0xad6870aa, 0xa717f388,
+	0xaec872bb, 0x3a214b68, 0xca5d5acc, 0x685d42ce, 0x9cf78460, 0x39ef1c0d,
+	0xf71e3a3d, 0x0e09675c, 0x268cb183, 0x9c833016, 0x301cad02, 0x8259c78c,
+	0x5ae49fb2, 0xcd3cc3a3, 0x7829a07a, 0x54a7adaf, 0xa73610b7, 0xfc609b09,
+	0x3e3de96f, 0x67c1c63d, 0xe375d2fa, 0x9ae161f0, 0xa5a0c18e, 0xb8211b19,
+	0x05080b41, 0xc2de8e87, 0xdf184a43, 0x790f4185, 0xd72ed46c, 0xdaed0859,
+	0x2faf0b06, 0xae9417d7, 0x988ff787, 0x5d71cbbc, 0x891ce4cc, 0x4a162b79,
+	0xb7c6f5a2, 0x0724024f, 0xb6f2b815, 0x4cb54d77, 0x2814fe40, 0x1ecc3fcf,
+	0x24255bb6, 0xa5361dae, 0x6ed0921e, 0x2dbcb6f5, 0xa2bb7fa1, 0xd76a6de9,
+	0xa1ca1d61, 0x07535dc9, 0xd4f88981, 0xf4d2b2ba, 0x4b5e3011, 0xd60eba65,
+	0xb2badcf5, 0x4f75ea43, 0x1c51d35b, 0x775eb759, 0x77dd66b7, 0xcfdd70df,
+	0x5d2eefe9, 0x75ffdc5f, 0x8b3fbec2, 0xa25199f8, 0xeefe510b, 0xf0282da4,
+	0xec99bb78, 0xc69c7941, 0x1287f638, 0x7e718437, 0xe850eb8a, 0xa14f7947,
+	0x9174789b, 0x79239ccd, 0x1f4d503e, 0x6d9ef4c7, 0x7a4bd708, 0x4054894f,
+	0x787b25eb, 0x2ee7cf86, 0x2dd08fcb, 0xd1f2f9b3, 0xe61832be, 0x57e702fd,
+	0xb2519ae0, 0x7802cc52, 0x21ec6177, 0xd49eb235, 0x2e7301d5, 0x03d533ac,
+	0x7be177d8, 0xcf7c66be, 0x8427bb31, 0x32262647, 0x4620be5f, 0xaf995ee8,
+	0xdf1e4dfe, 0x05d4afaf, 0xed5c90d7, 0x08bc7012, 0xe869743d, 0x43d387a1,
+	0x87a269eb, 0x9cd3b6ea, 0x4aeb5a1e, 0xdbaa9e91, 0x85d1a704, 0x8719eaeb,
+	0xebeffcfb, 0x61ea97e6, 0x7b5a879f, 0x1d0107ea, 0x69655818, 0x7af09985,
+	0xb3ebe118, 0xca7c323f, 0x6fede2ba, 0x1facd1c3, 0x707d468d, 0xe61ee75c,
+	0xe94de08a, 0xd12f4977, 0x7c906c3e, 0x295e90ef, 0xcb823aea, 0x3be8ed7f,
+	0x1c5bb40e, 0x883defc7, 0xc1663bef, 0xf322b90e, 0x535f1c36, 0x1a67e78a,
+	0x64381d60, 0xa009eb02, 0x1f1f13dd, 0xf8c25f9c, 0xf84d3968, 0x9e1d61d1,
+	0xa7e72abf, 0x5ec8cf80, 0x7ee56a19, 0x1e7ac7ee, 0x5e9bfb3f, 0xef6e4b94,
+	0x2abf1ea6, 0x5c4fd0b9, 0xa9fea3a7, 0x5e8e8e58, 0xf7ea3aff, 0x055e5437,
+	0xfb287c11, 0x4813d0a9, 0x5fe62b4a, 0x577c5996, 0x6493f4d5, 0x4f1a8acd,
+	0x9fa36f79, 0x7f8017b9, 0xde63b137, 0x81ff2823, 0xf543fe7e, 0x198318de,
+	0x47f4a6f1, 0x3fb58cfa, 0x14331ef0, 0xd05c79f0, 0xa563b9bf, 0x7f617688,
+	0xf650b5f7, 0xacad0b8f, 0xbf7835f6, 0xac4a0130, 0x01f5cb27, 0x2f107d72,
+	0x75b94fb3, 0x946f3c02, 0x6e9effb8, 0x7a92e390, 0x3bcc48e7, 0x3db81ba5,
+	0x923e9c58, 0xb7c2bee4, 0x2484be88, 0x58021b32, 0xe964f566, 0x3e14df70,
+	0x42e8f7d6, 0x594a5076, 0x0dce977a, 0x61f98fcf, 0x9ff037dd, 0xbd108fa6,
+	0xd1d1ecb1, 0xfcbf5d06, 0x5276b898, 0x6c3b1f57, 0xec9521b4, 0x42f78da3,
+	0xaf447abd, 0xc6b5fb84, 0x8ddb7d5a, 0xe169efce, 0x13ed763f, 0x577bfa07,
+	0x8f2c6afe, 0x5078f715, 0xff51dbc6, 0xb14f7a6c, 0x30ab68a3, 0x3ec0cfbe,
+	0xd70bf0e2, 0x7d92548d, 0x3c890774, 0x3e80c7dc, 0x55a7f509, 0x777e73a0,
+	0xebb3fa46, 0x8477e7c0, 0xb1b05696, 0xb72bf950, 0x3ff04c6f, 0xe98faff4,
+	0xed1e747d, 0x9a79e318, 0x3b6d6b42, 0xd33199d6, 0xf985c3bd, 0xa29e753e,
+	0x2ca1a130, 0x42489509, 0xc533cd26, 0x0e4f69a4, 0xb2fe51e8, 0xe71e4cec,
+	0x945f3062, 0x44f40e9f, 0x7c3658f9, 0x3395d924, 0x65dd2af2, 0x48bdfbf2,
+	0x731b7eac, 0xfae44426, 0xeefdd734, 0xd1dd9532, 0xfd3946e0, 0xe245dfd6,
+	0x7f1e2d6b, 0x7d7ce065, 0x9775efd4, 0x63c5f4e5, 0x67584ba9, 0xa7c5f577,
+	0xfc5a5f16, 0xc95fc3c6, 0x52597729, 0xf1693c0c, 0x46ff0c72, 0x06368f3c,
+	0x9de20203, 0x4b98e9e3, 0x78188160, 0x29d2ec52, 0xa4683176, 0x9dd90f76,
+	0x68f0c977, 0x4f17d6f1, 0xd2bb8a9e, 0x411ff022, 0x4e91e87c, 0x884aefdc,
+	0xdb0304ff, 0xdd0ceb81, 0x6ab7f949, 0x434c1fb9, 0xf93bb579, 0x8375e150,
+	0xf087707c, 0x386301bd, 0x6d55fd7d, 0xb83dde7f, 0x14a0dd72, 0xfb9803c4,
+	0xd4671f1c, 0xf1b8c374, 0x7da4cd61, 0xdba8fc07, 0x587e3993, 0xcbbf6f32,
+	0xd0d769a9, 0x0abd768d, 0x3e1bfcdd, 0x3f17d498, 0x30ecdd0a, 0x067276c8,
+	0x1e788c1b, 0x02426fbb, 0x78da2dc5, 0x4f4b5ede, 0xc37a4f13, 0x47f4c8e3,
+	0x5f3025f4, 0xfb33b9f8, 0x569bc86b, 0xf25166d0, 0x7be8b866, 0x51bc9623,
+	0x7976fff6, 0x81676e9f, 0x4b03a283, 0x74c25ffd, 0xff591993, 0xda275c21,
+	0x20d883e7, 0x8318e75c, 0x7969cb8b, 0xcb75e19d, 0xe58a09e7, 0xb64d8bf9,
+	0x7acb8e10, 0xf8c297f4, 0x5cf8a2cf, 0xc08a4a40, 0xb3fb9061, 0xdb8cb521,
+	0x79d20f5d, 0xbb0bc01c, 0xf1c08d6e, 0xa7a08307, 0xc70c9578, 0x80a69de5,
+	0x57ac2b78, 0xf511fdf4, 0xf8e125d6, 0x3e9f5a58, 0x4a7a8b94, 0xdadbdedb,
+	0xd0ed8d51, 0x93f5116f, 0x71865fe1, 0x9ea7a007, 0x6dc18d37, 0x8a48e231,
+	0x4a6c093d, 0x83e15cf2, 0x6505513b, 0x92dfa5f7, 0x1c7af1d6, 0xa5b74b5f,
+	0x1da159e0, 0xbbd7a5aa, 0x7c1ccf58, 0x4553a5ae, 0x7a597eaf, 0x30a9ec50,
+	0x74d4845d, 0xd003d0a1, 0xfa42ffc7, 0xf1f09ee8, 0x9efe68f3, 0x169bd098,
+	0xbd810f4b, 0xeca293a7, 0xb3f678e8, 0x48273762, 0xc7da333e, 0x595d797a,
+	0x9e1d2e79, 0xf25dba8f, 0x62b9f058, 0x0fccdc4f, 0x80e01d60, 0x91cc7d02,
+	0x83a4ddf1, 0xc0dbe006, 0xf288214a, 0x63d621e6, 0xaa1ef9c0, 0xc9fca835,
+	0x8f9ee639, 0xe6fd087e, 0x3d270dd4, 0xbf28c3e4, 0x38f3ef48, 0xffcf3d44,
+	0x5c9ebcb3, 0xaee6bff4, 0xd1fc6836, 0x39e3a224, 0x23ecfe4e, 0x09303be8,
+	0x4d4aedc7, 0x76f94d1a, 0xfced7f55, 0xe3b43aca, 0x91abdf6a, 0xee78b2e3,
+	0xf7851dca, 0x83b71251, 0x131ef0c2, 0x6f80bc39, 0x6aebac1d, 0xd3e088db,
+	0x2954e831, 0x8e5d267d, 0xef52393e, 0x3b38d37a, 0x955f6318, 0x3f7c1479,
+	0x7c153ac8, 0x0e8b1e7e, 0xffdc7d8b, 0x910e6dd6, 0x2f1dacbd, 0xec8ca7a7,
+	0x81269cb0, 0x4de5deec, 0x2af9ba25, 0xa3fb9bfb, 0xa6efec7d, 0xf2de3483,
+	0xfedc8396, 0x3b23aa6e, 0xf70b797e, 0xa43ec047, 0xc80463f3, 0x15cbb268,
+	0x7983d33f, 0xb79815a2, 0xbd21ede4, 0xf3d5cf0b, 0x78d235fe, 0xf21ca9f1,
+	0xc0ff7913, 0xa7e8b5c4, 0xa85afd86, 0xbab5fb47, 0x9576c2ba, 0x6c281b5f,
+	0x5b5f91bf, 0x05385f53, 0x8fe05afd, 0xcdc15245, 0xe4b663f9, 0x797fa1ba,
+	0x41ade4c4, 0x02e2533e, 0x6abffdda, 0xfd202572, 0xbdfb253d, 0x23cf57d3,
+	0xfc3cde7b, 0xa37990da, 0x7df2de51, 0x76e24c7f, 0x7be88a4b, 0x3eebc70c,
+	0xcf172d74, 0x7fb4d4cb, 0xbcf911ba, 0x1ec3ce4d, 0x5d9157f7, 0x13dfeca3,
+	0xa777b712, 0xa077688f, 0xc14f7ffd, 0xcfe81cef, 0xbcf3d65c, 0x4922a5cf,
+	0x18f66f65, 0xd7027aee, 0x1e8094ad, 0xf16ee511, 0x82938e84, 0x8ffcbe23,
+	0x808f77c8, 0x7b5d9df2, 0x5bd70d3b, 0x81d872d3, 0x7e84bef8, 0x5bc9c610,
+	0x8f30d3da, 0x77f9a2b8, 0x9e7bf40e, 0x7b5db897, 0x40e9df59, 0x1ef96ebf,
+	0x5fc9dbb0, 0x003da50a, 0x32d4ba78, 0x3e69f0bf, 0xbf5a298d, 0x85f57fe1,
+	0x76edbb11, 0xeb282053, 0x284e253b, 0xefa1875c, 0x663d116c, 0xedfb5f47,
+	0x4dacd1f2, 0xcc0fb815, 0x0f861ee0, 0xec83acf5, 0x09962539, 0x21cf0a92,
+	0xbf2b901d, 0xefbde96e, 0xe7162236, 0xda0251ba, 0x0c14be3d, 0x8db69592,
+	0x9188375b, 0x73d950be, 0x12ef7b5a, 0x1b77dd73, 0x7aa3b9ca, 0xd6b37686,
+	0xfdd163fe, 0x9f7fe387, 0xd6bcbc25, 0xfdc56d92, 0x59d320da, 0x54ef4b4d,
+	0x166a97e4, 0x6f7e0efa, 0x3a22dd55, 0x7246ec43, 0xf85bdd77, 0x9b7753fd,
+	0xd8f485c1, 0x34d876dd, 0x22adeb16, 0x4e6a99c5, 0x3a428f14, 0x2866d2ba,
+	0xbcd7bc47, 0xf903b1f9, 0x57cfd15b, 0xde7a44da, 0x9e9676dd, 0x695ee32b,
+	0x7a851e4d, 0xfdc759a5, 0x9f238ebc, 0x7fdf915b, 0x326aed7d, 0x7eb297f9,
+	0xaaa7618d, 0xddad4eec, 0x6e19e53a, 0xb5494e47, 0xec579c93, 0xf8fa3ecb,
+	0x67675b14, 0x6b1f425f, 0x4919f0f4, 0x616ef3b6, 0xd8af31e3, 0x19792d7a,
+	0x9b7765f2, 0xe286b9d5, 0xe88edc1e, 0xb9f4f8ee, 0xb8fcf819, 0xee9c7148,
+	0xf1d1226b, 0x86cb03a1, 0xf1a2d976, 0xf5fb4f5c, 0x5f3fc0d9, 0x3ffd0fd9,
+	0x07716c3b, 0xf77a7d70, 0xd63daf28, 0x78e504be, 0x97432841, 0x4320658e,
+	0xec35f457, 0xe27d93c4, 0x7f91d7f7, 0xa67a9857, 0xddcd18a2, 0xfe425a7a,
+	0x057920df, 0x2687e923, 0x8075f22b, 0x73c20e95, 0x561655cc, 0x53d4f48a,
+	0xa0db9cc7, 0x9d51cf75, 0x30fb04ce, 0x1ba64761, 0xf955edb7, 0xa692f35c,
+	0x4a332906, 0xbf912c5e, 0x0860bb3b, 0x7a8a490c, 0xcb91a4fe, 0x97870bab,
+	0xeef248be, 0x161b237d, 0x362dbf43, 0x5bb57d72, 0x85d8cc0a, 0xe8e8cefd,
+	0x238ec1fb, 0xf3c11c6f, 0xba4e4b10, 0x28bea1c6, 0x6ffa7231, 0x3d6afe3d,
+	0x3d690b9e, 0x76f478c5, 0x83a8a57b, 0x1733f779, 0xabbd3f8f, 0x1c23ff23,
+	0x649e63b7, 0x461de75c, 0x1ff6672e, 0xc74795d5, 0xe7cbf605, 0x1a356e1c,
+	0x4661fc68, 0x8ed7ca91, 0x23cf2696, 0x19e5f3be, 0x75ceef91, 0x915c3bd5,
+	0xef629578, 0xb5f77a4c, 0xcfe06319, 0x4e749749, 0xbbd76edc, 0xaf5c44db,
+	0x896302eb, 0x07fb5f42, 0xf88f1bc6, 0x7fdbc657, 0x70078f89, 0x54be421c,
+	0xf41a0ff0, 0xbea1a347, 0x849f53bf, 0xf82813df, 0xcca1d657, 0x9e626aef,
+	0x8eb11b46, 0xc41bb5e7, 0x8f52314d, 0x54ef91f4, 0x7449cef8, 0x29e486f9,
+	0x1810f4e5, 0x870bf582, 0xec57642d, 0x4b11f596, 0xf88a1caf, 0xbe69f653,
+	0x0513eb91, 0xcff3789e, 0xca54e191, 0xafd74b2f, 0xa2015c1a, 0xc5023d9b,
+	0x90815c19, 0xfb053b29, 0xe739d365, 0x37d8de48, 0xbd87f707, 0xd6ffa6ab,
+	0x8ace9d2e, 0x214df1f2, 0xfde0d291, 0x74c34fb4, 0xfa226fa6, 0x88c93d6b,
+	0x7cee978f, 0x58cdc798, 0x728e9068, 0xc84181c8, 0x9c3fc8b7, 0x0503437f,
+	0x35b52728, 0xf8441dae, 0xaf90390d, 0xbe027822, 0x41b6b2c1, 0xb25d7a42,
+	0xe58b978a, 0x6c9647ef, 0x2e7f11f1, 0x15395ada, 0x98d5a7ed, 0x28bd40f4,
+	0x59d32bd3, 0x4ebcf032, 0x85df6caf, 0xdd3897e1, 0xd4d96c5e, 0x7d02f38e,
+	0x63e233fd, 0xcfb3eb72, 0xbd0ab823, 0xa06edc5e, 0xde90aa71, 0xad74f8a3,
+	0xde7bafbc, 0x693ef4e5, 0x7bf4a1bb, 0x0149eeb0, 0xec38c4e6, 0x443bedfe,
+	0x91fbdf94, 0x9f39339e, 0x70be9ea3, 0x63eb69f8, 0xeb4ecfa8, 0x777bc618,
+	0x39aaac4b, 0x1a3a2f8a, 0xeff1575f, 0x33759dd2, 0x95dcae28, 0xa66ba024,
+	0xf782adf0, 0x1a7b92dd, 0x34b7fff1, 0xdfc65ffc, 0x81ff87df, 0xfe649b7e,
+	0xfc0e54df, 0xb7128f0f, 0xce380a27, 0x79f6296e, 0xfd8d57a1, 0x5710b73b,
+	0x23ef1eae, 0x72355f57, 0x0d91dbff, 0x73df57f3, 0x30effc95, 0x6cdd327c,
+	0x26e8532a, 0x7b71f053, 0xf8839753, 0x3d22bc02, 0xf289bd93, 0xb82b24b6,
+	0x8f45884d, 0x967c946b, 0x72236f8a, 0x5f5fe14d, 0x5853c21b, 0xed16bb56,
+	0x008ae4e4, 0xc145e9d1, 0xf4093437, 0xe7e445da, 0x67e17218, 0x333e5e32,
+	0x1be754f8, 0x32c1c2fb, 0xc0abae90, 0x69f3a2e3, 0xee27055b, 0xdaeed28f,
+	0xf782fdca, 0x9e573e65, 0xf915ffbf, 0x53f78c3e, 0xeface27d, 0x933eb91b,
+	0xdcceb04c, 0x4c67d720, 0x60be2528, 0xc819e9ba, 0x8f404b6f, 0x8f4a2f7d,
+	0x1034345d, 0x38782abd, 0x67fbd32b, 0xcfdd7c89, 0xc41e70c9, 0x5be12998,
+	0x7a0947ce, 0x5af9c0bd, 0x8bbf3814, 0xc3c5fe38, 0xdd11a793, 0xee9b13df,
+	0x246aed97, 0x772c4cfa, 0xb3b9438e, 0x2736ef49, 0x8f258e8e, 0xec765483,
+	0x49f909f8, 0xf7c8c748, 0x0aa0f8f1, 0xa0c580f9, 0x7fd020ff, 0x7429fed4,
+	0x1d717400, 0xd73ca5f1, 0x379285d2, 0x5bb2f053, 0x669b7793, 0x601a78e2,
+	0x25def2e2, 0xfa78a9cb, 0xf3ef2be5, 0x2bf87dbb, 0xe6a054cd, 0x456fdc2f,
+	0x13717c4e, 0xe2679b3e, 0xfa644fed, 0x780ede4e, 0xe4a0627a, 0xeb720b5a,
+	0xe766d809, 0xa63e48f8, 0x595f9e49, 0xa09348c7, 0xf368dbca, 0x6f830629,
+	0x9f0a2984, 0x8d73f798, 0x334bd08e, 0xd81cfce3, 0x43cf946e, 0x5f765538,
+	0xc6912981, 0x40c1ea77, 0x967fd405, 0x0361d959, 0x0ceeeeb8, 0xe1718c54,
+	0x5981db77, 0xbae264f7, 0x853e829c, 0x53ace5d6, 0xdf20ce8e, 0x7c50b71d,
+	0x3ddb3f24, 0x974cfc8a, 0xf7c71f3b, 0xeca4fbc8, 0x6b75bc67, 0xfdbc5cf5,
+	0x21e3279c, 0xcad76e55, 0x2e26418f, 0xf9657a37, 0xf83fe5a1, 0x526f9317,
+	0x3b242fb7, 0x6d8c9d90, 0x326fde07, 0xff5179f0, 0x00d4bf40, 0x23ddfeff,
+	0xdcec27ef, 0xc343c41c, 0xa39d2376, 0x31d7f850, 0xf18e9bcb, 0xe48a4713,
+	0xaeb8c176, 0x7e1a6c59, 0xe8f2efbf, 0xd651bfc8, 0x26e1ff5f, 0x7ea26fba,
+	0x7da19187, 0x57ef209f, 0x0c162cc2, 0xbaf9f109, 0xb1e80763, 0xd3fe2894,
+	0xee89eb07, 0x5103bf80, 0xd93437fc, 0x2d806e97, 0xdbfac1fe, 0xc0b74cca,
+	0x8ccc993c, 0xa82aea16, 0x347c006f, 0x96047924, 0xb675f548, 0x0fe3f181,
+	0x2e4d303f, 0xbf04f3c3, 0x6fbe7a32, 0xc438a54a, 0x1c2cb00f, 0x28b22b50,
+	0x1d99bdbe, 0xb1f955f3, 0xb0738e46, 0x69469036, 0x9ce2f8aa, 0x2f8a2e9d,
+	0x507c8d3e, 0x79fe719d, 0x8da6ae1c, 0x078beb9f, 0x678c8a47, 0x17c4e790,
+	0x115ade29, 0xcf857ffd, 0xc925379e, 0x3011ebfb, 0x1dd0aefe, 0x93da29f2,
+	0xe31d7e4a, 0x2f8a762e, 0xb0ee8907, 0xc23e309e, 0x742a1b7f, 0x61e2b1e5,
+	0xafb64cbb, 0xf23487a7, 0xb17db589, 0x4bd031f7, 0x48b7f835, 0x29976af2,
+	0x54d81eca, 0xc78d2671, 0x439d19cd, 0xe39e447e, 0xffee35c5, 0xe1f28e18,
+	0x947efcb7, 0x80294671, 0xfdc0aae7, 0xe4d33975, 0xda9d90f1, 0xfc1d9084,
+	0xd1cf9add, 0xf37f0173, 0x2f951d73, 0x3cb7b3fe, 0x56d9ed13, 0x032c1b9d,
+	0x5c93ebcc, 0x41650f70, 0x9373d013, 0x2a9b9f85, 0x9079b769, 0x1305bbfe,
+	0x13e99b9d, 0x19ea2f8f, 0xd7da0bb2, 0x411f01b6, 0xb184bbb5, 0xef57844e,
+	0x0ad0f3a6, 0xd89ec77e, 0x63df4794, 0x37dc1be9, 0xcbc55b8a, 0xb6af16df,
+	0x894adc50, 0xf4f71593, 0xb5649d10, 0x61f8ea65, 0x5e8f3e29, 0xd792af5e,
+	0xbef1926d, 0x226be8fb, 0xb2c2b1ea, 0xef68b586, 0xef7d1867, 0x2e4ddd23,
+	0x1339db56, 0x3b76da84, 0x4634d7d2, 0x1fc2827e, 0x20ec5dc9, 0x4f12a986,
+	0xcfd07976, 0x73c17c9a, 0x04a9fd10, 0x77f8132f, 0x24f3e192, 0x58ff70e7,
+	0x07b8478a, 0x13fc29c6, 0xe328bc23, 0x2a5a3eec, 0xd2919da0, 0x46b82659,
+	0x2defa5f7, 0xe04e77e8, 0x5f701ee7, 0xe8687f15, 0x6196462f, 0x2fe9487c,
+	0xbd5cb99a, 0x429be7cd, 0xafa34fa1, 0x349de717, 0xa98ed83b, 0x81558997,
+	0x84f10c76, 0xd184b3d7, 0x02ba8378, 0x27818eb4, 0x9658257d, 0xf7048948,
+	0x3a210be4, 0x96a0312f, 0x5b9231d1, 0xfb25df3a, 0x4eda96e1, 0xf771f701,
+	0xef0443d1, 0x5e0ecee0, 0x9af57cca, 0xb814ffac, 0xa27f644a, 0xd5c38e44,
+	0x8e9f9e5f, 0xecd181fe, 0xa338c12b, 0x12fa8eb2, 0xbf9e3a5f, 0x3ce782dc,
+	0xe73fc426, 0x14966691, 0x8cc1373f, 0xaeecee7e, 0x5a7b4823, 0xf8ed39bf,
+	0xb48fda19, 0xdd1ddcd5, 0x80a57c72, 0xc1e5b5f4, 0x22b37ed8, 0x71c8b62f,
+	0x14f3a8c4, 0x7ad1cd83, 0x9ac932b9, 0xc28f2515, 0x9e722672, 0x9e7cb762,
+	0x6f7aa6c2, 0xa46613cc, 0x27834ea7, 0xdf2b9f07, 0xdf2abce1, 0x71f8a01b,
+	0x961ebe0a, 0x9e68d3f7, 0x1086e749, 0xe3a04e4f, 0xbfb1cd57, 0xd6685987,
+	0xd3977ac9, 0xf4e4eb0b, 0xb688eb82, 0xb999a633, 0xe3cdbd5e, 0x74e7a29b,
+	0x9c788452, 0x476c1454, 0x1e589dba, 0xbffa4768, 0x5274c985, 0x4b44e9ce,
+	0x939bf7ef, 0xa1e93a64, 0xe0260a7b, 0xe43145f7, 0xdb998563, 0xa59d2977,
+	0x0fc03f44, 0x857ce177, 0x03456a83, 0x339cd0a5, 0xf0f8ef50, 0x9e7822f2,
+	0x4769e45f, 0xbc7a737a, 0xa8db8e22, 0xc9ba7277, 0xb8c49ef4, 0xe4aa7d77,
+	0x38d02b5e, 0xe0c736df, 0xa7e416ab, 0x4a7e11d1, 0x5a47e391, 0xc3229cba,
+	0x1cf193e5, 0x43df1e7f, 0xa74a4f38, 0x33f783ce, 0x4c25b9e0, 0x055c4e46,
+	0x96f912fa, 0xd2f9d275, 0x09672aa4, 0xe5a0ddf2, 0xf9f7016e, 0x016fbd25,
+	0x3c76b3ee, 0x350c9f45, 0xbffe8879, 0x6268deb2, 0xf2ff8d26, 0xb5db99b0,
+	0x7e512b1b, 0x8499f2bf, 0xf0a5d5f2, 0xd6f1d1e5, 0xf3745179, 0xf83a5f74,
+	0xa07458a7, 0x3a5a7f95, 0xb50bc888, 0xadbc03fc, 0xedc3fbe5, 0x55ce94dc,
+	0x6a1b9722, 0x7f6f1779, 0x0d0dc8fa, 0x3de98f1e, 0x097ae0cf, 0xf39105e9,
+	0x74f4f1db, 0xae1ff9a3, 0x31cd1ba7, 0xd4ceb4f4, 0x9f3c1bf3, 0x744da2bb,
+	0x5f225c61, 0xf6ba8f3e, 0x4f3a24f1, 0xe09362f0, 0xd1439746, 0x502244d3,
+	0x6127b2bf, 0x4a15bb2c, 0x6ae78078, 0xb2250302, 0x08c759cf, 0x24d633ca,
+	0x1bac70ca, 0xdbb7fa41, 0x51e600a9, 0x256063eb, 0x91d3fa01, 0x684b181d,
+	0xf310783a, 0x988b6b2c, 0xb77000cf, 0x80f3f1b7, 0x485bd62f, 0x7e3eef9f,
+	0x400f37f5, 0x1d71e73d, 0x0e047f1d, 0x4ab673a6, 0x03b145de, 0xe8590e3a,
+	0xb32f703c, 0x7754d1b8, 0xf8790d8b, 0x21fbec02, 0xe8d983c4, 0xe2281fcb,
+	0x77aec439, 0xf1eca38a, 0xccbfd6de, 0xe48ae439, 0x7d0e9a97, 0xdf6d185e,
+	0x14a9f929, 0x3996f7d1, 0x963fdfca, 0x77cb9677, 0x055fb9cd, 0x9c5e309e,
+	0x5da27f1c, 0x2375f04f, 0x48bf687c, 0x10fb861e, 0xe3d648f8, 0xfb94de32,
+	0x0adbba0f, 0xbe80fa4e, 0x98393c5e, 0x7b5be1c7, 0xef7768cd, 0x214ef932,
+	0xf4b7899d, 0xf6a9f9d0, 0xe811deb6, 0xf9fc834f, 0x1ef79f08, 0x9d149b3f,
+	0xeb12feff, 0x3f5a24fb, 0x729f5bc4, 0x1f4b5e3e, 0xda167dda, 0x184dfe0e,
+	0x50c07c4b, 0xb086054e, 0xf567fa20, 0x17adc663, 0x31672371, 0x341dc517,
+	0x999fb479, 0x8e65ffbc, 0x4c679fe2, 0xde21bfde, 0x4efc1e71, 0x05f9d603,
+	0x989ec7d6, 0xdb478f9f, 0xf53ac2be, 0x18fb9896, 0x215afda2, 0xe60567af,
+	0x537cf0fd, 0x3852e18b, 0xef231ddf, 0x8091fb07, 0xdd4a05e3, 0x9fb87d27,
+	0x6bdcd81f, 0x4efa1f11, 0x0091da27, 0xb3b6227f, 0x97ba4aa2, 0x50f060c7,
+	0x3ad5f47e, 0x5bbd7e7e, 0x1e31b8b0, 0x6491dcd6, 0x33f87941, 0xfb1d2f93,
+	0x9d5dd48f, 0x09fa7183, 0x3cdd5e9d, 0x41be53d2, 0xe753b216, 0x8e09eaf7,
+	0xdfda315a, 0xcfeb12fe, 0xe7dcc4b1, 0xc790fb3e, 0xa1f7c5d7, 0xf462e7df,
+	0x3e9c7bfc, 0x45d651f9, 0xfb83f6f1, 0x3cc5c93e, 0x4f81f552, 0x7ead3fc8,
+	0x5b9b48c3, 0x0488c6f6, 0x252be3a4, 0x4b89edb1, 0x1eb75afe, 0x967199ec,
+	0xcb9b922c, 0x23dd2759, 0x61e9a2b0, 0x5e4a3eb7, 0x71d3f62a, 0x8e056743,
+	0x6440bc7f, 0xbd774075, 0x580389ee, 0x94ac390c, 0x097b45bf, 0xfd1f7f60,
+	0x9f9bd4d2, 0x9de6d283, 0x338fb4fb, 0x42a11c53, 0xb3cc8763, 0x88f73d70,
+	0x8fbc325c, 0xbcbce005, 0xe0e74ed7, 0xcf7f8f2e, 0x5c4b1f81, 0x7e5df96e,
+	0xb0ffcf00, 0xbf3047ce, 0x8d8bf816, 0x718aa8fc, 0xd9d7e5ef, 0x3b76fc23,
+	0xdbb0418f, 0x70e13f5f, 0xe68aff41, 0x144094f6, 0xa1e7b4fe, 0xc7d7911e,
+	0x6e745df4, 0x98bfed84, 0x563ef786, 0x829c3bec, 0xec37fdeb, 0xaedfc318,
+	0xf44d5d58, 0x9f22d533, 0x1960caa3, 0x8bfa0f28, 0x0573ed07, 0x5dc7c70f,
+	0xf77a821c, 0xd23b1c4a, 0x75ebf72f, 0xf6801a4f, 0xca14854d, 0x55bf1589,
+	0xbc2d2bfc, 0x3b373b37, 0x58f7e243, 0xb6af9f6a, 0xebdf3d70, 0xf000b336,
+	0x068d23e9, 0xc4eeaefa, 0xcacc3fb5, 0x64f8fd87, 0x06b9c1ab, 0x85eaadf8,
+	0x309bc1f6, 0x48ce395e, 0x5c4f97f1, 0xcba41ca1, 0xf3f02fc5, 0x7d6c9fb8,
+	0xc8f30d24, 0xb5f409ff, 0xdc58e01c, 0x55e7a068, 0xe7a7df8d, 0x5bba6517,
+	0xe52a42fc, 0x3c744f97, 0xaceb869d, 0xdf93fbfa, 0x966ffe41, 0x41c6a39c,
+	0xa0977cb9, 0x74b7773a, 0x575559bb, 0xbacad2f2, 0x8f391099, 0x7e7802e7,
+	0xe79af969, 0x89f57780, 0xc2d2fe5b, 0xdd6f8bed, 0x2eaeeb07, 0xaefb25e8,
+	0x8c8588fd, 0x44db5fa8, 0x71a1dfd2, 0xadd6f5ff, 0x77f29f50, 0xd31f9bbd,
+	0xb239274d, 0xece5c4c8, 0x70fa7d94, 0xff719cbe, 0x615776dd, 0x08797cfc,
+	0xdbd200e4, 0xe18a9f94, 0x79f479bc, 0x5cde6d54, 0xf36b8fc9, 0x029c78e6,
+	0xf2cdf1e5, 0x1c3f0ae6, 0x49cf84b3, 0x1fb7dd12, 0x45f279e1, 0x2f6e45c7,
+	0xfc4e44dd, 0xe3d6e116, 0x107b1efe, 0x6427dde3, 0xf6b89c6c, 0xafbc3ac9,
+	0x012293e1, 0x93ec2033, 0x58f2c66b, 0x2287707c, 0x659cf02f, 0xef5f645e,
+	0xd7092f71, 0x5dbe0d91, 0x1ceab0f4, 0xd5e5ff68, 0x49f0f98d, 0x419eff0b,
+	0xfcb4cf2b, 0x67928ff2, 0x0f0ee315, 0x7ca9d809, 0xcd4bcb19, 0xe568ae5c,
+	0x14a7581b, 0xe4e55a87, 0xa5c0a2f4, 0x5fc57785, 0x3f3058b3, 0xf5177e1a,
+	0x94efa15b, 0xf8acbe50, 0x260b2e6b, 0x50fea37f, 0x3f8635b2, 0xf095fa04,
+	0x07d42815, 0xfd408fd1, 0xdb46fd8f, 0x0025c95c, 0x96673ad2, 0x2772e515,
+	0x416b7fe8, 0x6a313e24, 0xe50e8357, 0x376fd845, 0xc577fda8, 0x72b2c95f,
+	0x795a31ba, 0xddfc98fd, 0x5df3bc4f, 0x3c4af8ca, 0xf184bea4, 0xb91273ef,
+	0x839ca6fd, 0x3967eb5c, 0x933e3df8, 0xd26bc3bf, 0xc8f0eff1, 0xfcf5c4a7,
+	0x83fdc8d8, 0x5c5574bc, 0x29bf09f0, 0x792def07, 0xf7a4efd2, 0x53a7f29f,
+	0x8bbf23f9, 0x79fc67ca, 0x135efc39, 0xc7c8fe46, 0xd18d2c87, 0xe49b04e7,
+	0xb093cdbc, 0xd062d7de, 0xe55623e1, 0xf497a03c, 0xd8d2c2db, 0x567bfdf2,
+	0xe4a5e13d, 0xa087fbf9, 0xc1f1f1dd, 0x097dd1ef, 0x83e26b1f, 0x6e6f250b,
+	0x13f37d2c, 0xc9379f99, 0x32c7e5fa, 0xfd900d7f, 0x9e729047, 0x4f78d1e6,
+	0xff773f30, 0xc31dfcac, 0xa9abfd4f, 0xf3cad3bc, 0x8ff76923, 0xf0073b7f,
+	0x3f036e75, 0xa4ce749f, 0x6cf3bf23, 0x85bcadad, 0xfd8ef0e0, 0xde76511f,
+	0xf1dc329d, 0xfd6355d6, 0x057e726a, 0xdea072eb, 0x876fd30a, 0x5c3e1e3a,
+	0xfa432df4, 0x695d66eb, 0xe846afd8, 0x2315ddcc, 0x79209b8f, 0xff84ac6b,
+	0x5f28219e, 0xed02fd95, 0xfefa2a31, 0xe72b228c, 0x7ce38fcb, 0xa65ab6ff,
+	0xcf25dbdb, 0xfa9e3e6a, 0x195fc37e, 0xf2b43f3d, 0xb4ffe678, 0x8c02f9ca,
+	0xce4894df, 0xdb9d7c70, 0xd635dedb, 0x78997118, 0x8e16bf7e, 0x9e9c6c5b,
+	0xa500f1b4, 0xcf01e254, 0x65044da0, 0x7049d5a6, 0x1803e36b, 0x07158d27,
+	0x8f2b4ff8, 0x3c7ba61f, 0x9f15d5a2, 0xcef8272f, 0x6c9b7dd0, 0x7bf3a336,
+	0x0e9b8a03, 0x60643c62, 0xd90eb059, 0xcb513ea5, 0xc63301a7, 0xba7cb249,
+	0xa4dbcc34, 0xa33f0275, 0x6a9781bd, 0xd69d22a5, 0x4f5d21d6, 0xf168e9f0,
+	0x412c3a13, 0x54e67141, 0x660266e7, 0x0bfeef39, 0xda82f3c2, 0x1db718b5,
+	0x797fbcb4, 0x9f7936f7, 0x84d9bc95, 0xc3c206ed, 0xa63fed84, 0xed84d6fc,
+	0x2885eb77, 0x35693efe, 0x878bff50, 0xff250e97, 0xaa39ce38, 0x26dc794e,
+	0xfc417bc4, 0xadc710f3, 0x8879f8df, 0xb8a8d6e3, 0xab8af54a, 0x5a77c724,
+	0x39f8feb9, 0x9f0169e6, 0xe4a05f3f, 0xfe291b69, 0x9a6a5bb6, 0x40887986,
+	0x43ce8eb5, 0x0bdcf386, 0xf08f79f8, 0x86ef98fb, 0xe2af3fe7, 0xc181eef9,
+	0xf6df2967, 0x03e79999, 0x6461db3f, 0xf25d2798, 0xefe51376, 0xb1d72cb6,
+	0xc364f758, 0x9cf9064c, 0xefcf018e, 0x7ca3fee0, 0x081df653, 0x08fdc24b,
+	0xf9b5c3e7, 0x8bae193f, 0xe9723e72, 0x4869e6ce, 0x93cb1927, 0x563fbc04,
+	0xeaebeb81, 0xbf506f91, 0xe5f7975e, 0x6f55fa66, 0xb55fa18b, 0xe0d3b57a,
+	0x6f52da31, 0x641b5fae, 0x2ab5eafa, 0x78e406e7, 0x6247e637, 0xd633cbec,
+	0x34ce713e, 0x738857db, 0x3149adfa, 0xf7ca9ce3, 0x6c490d9e, 0xfc18ccee,
+	0x37cc98f0, 0x743dcdba, 0x4da5230e, 0x4a972988, 0x17c94924, 0xc545b08d,
+	0x35f2b5c7, 0xd8dd7e06, 0xd0aec47f, 0x3e35ec21, 0x668ac731, 0x9c73e000,
+	0xf2719f0a, 0x9533bf0c, 0xe4a27d90, 0xd49a2e15, 0x6c77dbf6, 0x6dcf3f4a,
+	0xc4b78fb8, 0x6c53db91, 0x297f0282, 0xc5594bf6, 0x2572f154, 0x1b529e4e,
+	0xaffbf010, 0x3f230f14, 0xafb26862, 0x7afd00ce, 0x36f73126, 0xe83c27a8,
+	0x9779412c, 0x883ee554, 0xcbfadda1, 0x0fcaf7f2, 0x395efa33, 0x65f5ef9c,
+	0x7883bd91, 0x4f0efe8f, 0xfaca0fd1, 0x6bdd6f40, 0xb2cd70e3, 0x7d6f40dd,
+	0xcf317c83, 0x58be3fd6, 0xf769708c, 0x81ef1b2d, 0x7fd98d4a, 0xb0feac45,
+	0xd9fdd5cf, 0xbc720de7, 0xa4f9bfbf, 0xf5f0bd46, 0xf17de5e2, 0xbc2c44f2,
+	0xe5d0a86c, 0x009479e1, 0xe963b8b3, 0x7fb62d9d, 0x8b7edc78, 0x9edc462d,
+	0xfebf24dd, 0x49773fe2, 0xe2560f18, 0x507b61cf, 0x9e575793, 0xf909b757,
+	0xb49597e8, 0xb7fb235f, 0x64e7e0fb, 0x8fcdf6ff, 0x025779f3, 0xdc7f2279,
+	0xbf6f9f25, 0xf7fb95ba, 0xa4fd7322, 0x39a8e018, 0x6841606a, 0x606f8abf,
+	0x882116fb, 0x3d5f0bf7, 0x8bea853a, 0x192c09f7, 0xe9e02409, 0x160cefc4,
+	0x425e3e30, 0x8f754e91, 0x7d47a275, 0x7972784f, 0xea9eaa92, 0x3e13df55,
+	0xaa967660, 0xa46fa07d, 0xd5507daa, 0x96fd5578, 0xe13df55a, 0x457f4143,
+	0xd30333d7, 0xd7b3fd55, 0x0faaabdf, 0x13df506b, 0xcbadf8ae, 0xdf0cf5e4,
+	0x2f7d4f32, 0x61084877, 0xbca163be, 0x38e40bdd, 0x52e088c5, 0xc4cb38d5,
+	0xbbf0470d, 0x57de9b8a, 0x7448dad4, 0x68afbdaa, 0x7d754ed9, 0x46fb89ec,
+	0x5e0cf3f1, 0xc675373a, 0xdc5f8ea6, 0xc3035f74, 0x1abed6bf, 0x6afbd5d3,
+	0x886e5976, 0x7a9e6b37, 0x6f7c999f, 0x766f168c, 0xbfff7ab7, 0xed19bc69,
+	0x0c193c6a, 0xe31d8b3e, 0x44abad80, 0x5cb2defa, 0x4bee4e7d, 0xed93ddc4,
+	0xcaeee231, 0xcf376abd, 0x0dffb4a7, 0xd04664f3, 0x173c0634, 0x8536e455,
+	0x48e627f3, 0xaf2b5d9c, 0x75e41294, 0x2af29d8e, 0x5ad3bc91, 0xff21df86,
+	0x2b708bde, 0xf7588979, 0x05ce6880, 0xbe71c68e, 0x4d0afdea, 0xd240e7f8,
+	0xae76f983, 0x51ec0b07, 0x8739507a, 0x78c1e8a2, 0x897928e3, 0x4870a578,
+	0x0f44b8a4, 0xc86257e6, 0xbfdb6ff1, 0x7be93ea6, 0xf6dbf38e, 0xe0bf2882,
+	0x5047bdfb, 0xefde039e, 0xedd9f5c5, 0x888527a0, 0xd589fc9e, 0xf3d3fc41,
+	0x3c38311b, 0x48be7233, 0xe1689fc9, 0x0f7cd62d, 0x87b9a6d9, 0xb73c7eb4,
+	0xb9dae63b, 0xdf2c3d5e, 0xd7235143, 0x7a76346b, 0x8fee7acd, 0x6d15bde2,
+	0x8d5ff651, 0xe7fd4edd, 0x106a4be8, 0xf70b46ee, 0x8105eaab, 0xa67d78f4,
+	0x6e3ef340, 0x71f6a52a, 0x5c6a92f6, 0xacb930f7, 0x2a5e58c1, 0xed443bdd,
+	0x5efb1892, 0x6e765582, 0x7896cf2f, 0xddfd2fff, 0x0ee60ad4, 0x9bbe44a5,
+	0xd3be7162, 0x3d83c6d7, 0x140ece79, 0x3bf55f8b, 0x9a1fc42a, 0x3f1d0ade,
+	0x43e70f63, 0xfaf9fbe8, 0x66e7aecc, 0x109ebee0, 0xdb1e78dd, 0x1d602a75,
+	0x57c7829b, 0xa02256f1, 0x8d6cad93, 0x038bea19, 0x18f78a2e, 0x8fb25fc0,
+	0xcbddf08f, 0xc7c24cf6, 0xd3219398, 0x789f6858, 0xe3107bca, 0x58b3b9c3,
+	0x06ed8a12, 0x7f0530d6, 0x55b0291f, 0x5ab31bce, 0x19abfdfa, 0xe8a52b25,
+	0x5ba94073, 0xe2f3a61f, 0xf87579ca, 0x28fc4167, 0x9e47dff7, 0x4338f3f3,
+	0xe32f654d, 0x98e5297e, 0xc578ec60, 0xccbbf96c, 0x5ddef587, 0x5a63dfb8,
+	0xbddee8ba, 0x5b1c8ac5, 0x9defa3a0, 0xe1f793ef, 0x0f3cadbe, 0xbf5daecc,
+	0xce46162e, 0x0cf8c1fd, 0x2cf173f9, 0xb3bbdd3b, 0xbe932db8, 0x815e668b,
+	0x6d351be2, 0xbf497b6f, 0x44e0984c, 0x01f3d37f, 0x4b7ef013, 0x039701c4,
+	0x952efbea, 0x95efd37b, 0xa77ca7d7, 0x8372f20a, 0x78bbf015, 0x81a1dd6b,
+	0x2f7bde30, 0xa18fba31, 0xf240acf0, 0x37168981, 0xcb85edda, 0x05cfba04,
+	0xbdd1f40f, 0x6d862314, 0x7e3b8e8b, 0xb78bcb2a, 0xe5a078fe, 0x9c297404,
+	0x2a48f952, 0xfda768ec, 0x5dfc54c2, 0x95fdc2a8, 0xa338648f, 0x7b2a9bea,
+	0xde11bb07, 0x80cb530b, 0xe760fe4e, 0x85d74cf0, 0xd9bbfcf4, 0x543fca4e,
+	0xd20f24d8, 0x9d2d8569, 0x80ae37ee, 0xf400bdad, 0xdfa3fd35, 0x9f2e51eb,
+	0x05f7cbb2, 0x5c5cb2e4, 0x77c3b267, 0x2accf89e, 0xaf10339f, 0x0fb68d8d,
+	0xe539f4a3, 0xacdbba47, 0xf9586994, 0x3fe53666, 0xfbb4f795, 0xdf9f9cfa,
+	0xf0ec9873, 0x57fdc79c, 0xfade81fa, 0x93ddeb1c, 0x1db97f3f, 0xfc52bdc6,
+	0x17082197, 0x677e8fe0, 0x4ebfb941, 0x296dc719, 0x3f2762b9, 0x3f717986,
+	0xf9e3936e, 0xbb6d9d93, 0xebb7dd71, 0x0e516b42, 0xa9fef9be, 0x28671c64,
+	0x3eb8e078, 0xf5d573f8, 0x4570bee4, 0xbf7abee4, 0xf9f0a70b, 0xef2bbf0b,
+	0x8bc79b9b, 0x500df7bb, 0x0eed8bc5, 0xbfb86fbe, 0xc0937fd4, 0xcfd9f1ef,
+	0x937c564b, 0x9055fef7, 0x05824f7f, 0xe577e844, 0xc00fde4d, 0x66a87f21,
+	0x13a270ff, 0xb3cb839f, 0x51e7acc9, 0xb6e4e75c, 0x87efa77f, 0xf2792b27,
+	0xb93cea0b, 0xe2a59f0f, 0xf43d23b5, 0xb5f7701d, 0x63796fde, 0x43a95bf0,
+	0xfbf0fbc9, 0x837ec3f0, 0xf87e1f7c, 0xec5342bc, 0x7197f0fb, 0x89f120ca,
+	0xc52f0fdf, 0xe12fe03a, 0xe907f9ef, 0xc166c1fc, 0xb04c798d, 0xcfef5483,
+	0x87ded3c8, 0xf17fefcf, 0xe3df93c9, 0xc91f33ee, 0x1e194adf, 0x5dff1b5f,
+	0xcf3c0e82, 0xdeea7803, 0x51377fc0, 0x777fcf7b, 0x05b7e739, 0x4ba085eb,
+	0x553bfba0, 0xd6650bed, 0x6776085e, 0x3cec9328, 0x3dc477cd, 0x5c5f6c5e,
+	0x041befcc, 0xded34fbb, 0x754fe0a0, 0xf961ec4f, 0xfe1671f7, 0xf237d658,
+	0x64f32fbd, 0x590f31f8, 0xaf591692, 0xf9147fa2, 0xe16050b9, 0x1abee9f7,
+	0xed463ef9, 0xbb95e7fa, 0x36f3cf09, 0x761c6e08, 0xd067aff1, 0x078053f7,
+	0x2f995e95, 0x837e73af, 0x6ff3e740, 0xb2a98f76, 0xeeb845ee, 0xaef6becc,
+	0x50587ef6, 0x5f842e6f, 0xc6b1f990, 0x49afe1c8, 0x885f7466, 0xba25ec90,
+	0x125dc780, 0xf83b0f94, 0x0de10b39, 0xe627c923, 0xf2937b05, 0xe44e6574,
+	0x85058143, 0xcffaa3eb, 0x30fbe1af, 0xf842e7c2, 0x12c98efb, 0xc74cdbf9,
+	0xca3b6dfd, 0xd223d01d, 0x4e10dac9, 0xf2b126b7, 0x67bc0731, 0x20f06aea,
+	0x342e58cb, 0x2cdcb085, 0x3e7bbbc7, 0xaffc1134, 0x846cbe0b, 0x2a18a27f,
+	0xe79af927, 0xee93341b, 0xcaf9e3af, 0x167f328d, 0xaefa06b1, 0xa26b3260,
+	0x527598b2, 0xa1eb0779, 0xac5f83d2, 0xe03da364, 0x798cb2a6, 0x6aca9ca9,
+	0x0d672ca9, 0x712b7e54, 0xc002625e, 0x51359d35, 0x39e402bd, 0x32890bdd,
+	0xf74de319, 0xeecec89a, 0xac1cd37e, 0x0e6af800, 0x38bcb918, 0x577d2560,
+	0x08466073, 0x1fddf9e3, 0x7bf85cd9, 0xa7b176a0, 0xbf48e29a, 0xf7f93ab6,
+	0xbdd52e34, 0x2bb935dc, 0x78e3f77b, 0x7c7f421f, 0xbaad9e90, 0xafe27ba7,
+	0x7a8f907f, 0xe1ade260, 0xe81cf27b, 0xcb6d5fcf, 0xcafe3f26, 0x9446dd40,
+	0x5219d5d3, 0x7ab7e23d, 0x95ee8598, 0x05d3cee9, 0x415b2824, 0xe9fde3fe,
+	0x05efe06f, 0xff26d940, 0xf901c05e, 0xde10f57e, 0xdab60eaf, 0x951e7f21,
+	0x95377e1d, 0x5445fc47, 0xa87bf51e, 0x397f31df, 0x5efdc795, 0x9d1eaf2a,
+	0x77c7df4f, 0x839e9fa7, 0xc87427bf, 0x19a4f951, 0xb28fe1f4, 0x44fc2e7c,
+	0xfdf4eb08, 0x45aec0c8, 0xd508fdf8, 0xd7d67f41, 0xd19dd2f9, 0x555352ed,
+	0x92ea1db8, 0x7008fdfc, 0x629e93d2, 0xea4527a8, 0x118dfdf8, 0xbc199ce3,
+	0x57f3065e, 0xfb5220bf, 0x303f7811, 0xc72bd618, 0x2e094ca3, 0x03461f2e,
+	0x74e18cbd, 0x7c65df83, 0xea99d5d8, 0x69762310, 0x4e5c13e5, 0xedc0be54,
+	0xec67cace, 0x9eeef7f2, 0xef0d2132, 0x89cb511d, 0xf94c45f9, 0xbded13fd,
+	0xea2c45a5, 0x83b9983e, 0xe60f2475, 0xd3f756ae, 0x4ce807ca, 0x95aad3bf,
+	0x643a1f0b, 0x8d9733fb, 0x97faefc1, 0x1d93b1cf, 0xa76ec976, 0x763f7a73,
+	0x0164a475, 0x44fbd5bc, 0x0ab413ca, 0xb5b07640, 0xeab8de7f, 0xced0136e,
+	0xd7a7159d, 0xbeb1b8a6, 0xaefc49fa, 0x4ff352e2, 0x233aa4a7, 0xf4ec59f5,
+	0xdda007c6, 0xddf56eb9, 0x2ffac0f7, 0xb8c0bd23, 0xfc0d5b97, 0x12939b1e,
+	0xded32efd, 0x0b21af4c, 0x763f7bea, 0xa6ff3c33, 0x99bbf5e0, 0xbddbfbfc,
+	0x2b41be99, 0xa7deab47, 0x1afd61df, 0x6bf133ef, 0x00dc2fbe, 0x8a7bed7e,
+	0x7efbdb5c, 0xa77f0dee, 0x6dead976, 0xcea27243, 0x37d32c15, 0x658279d1,
+	0x921e5fab, 0xa5cdf603, 0xe88d96e2, 0x7835ef77, 0xf942ef12, 0x71e06cd5,
+	0xec904366, 0x89af4f80, 0x02466f34, 0x0609fbd7, 0x7c249f6e, 0x3d446a68,
+	0x1382f603, 0x6bdcafbf, 0xa439f0e6, 0xc112572e, 0xc2c3ebc9, 0x86f1f1ef,
+	0xf7b97025, 0xb25eb5e9, 0xd76e7ba7, 0x8082efd3, 0xc4faa07c, 0x6884523b,
+	0x226f67bb, 0x8af3de7f, 0xcc52fd48, 0xf1e8c49e, 0x2f1debb5, 0xdcd7b7cf,
+	0xc53f1aa3, 0xd653dfc0, 0xe36f9da3, 0x9de51324, 0x496e289e, 0xb2ebfcc3,
+	0x207bfe12, 0x8abc3245, 0xa3f1357d, 0xe1a3dba9, 0xeeff289b, 0x26e4ceb0,
+	0x557079fa, 0x7dfa38e2, 0xe297a742, 0xe71e73fb, 0x38f269f5, 0xe08e4520,
+	0xdcb9e772, 0xf20cd153, 0x28657c39, 0x79e417b7, 0x8afd608f, 0xf510a7a1,
+	0x24aab886, 0x33de0030, 0x9f495e3c, 0xd5d719f7, 0x6f34b005, 0xe1036468,
+	0xcdcdc2f9, 0xa13df70c, 0x8a176cf9, 0xefe69499, 0xaf6d7c53, 0xd813a641,
+	0xfe51998f, 0xa96736d7, 0xd7e60137, 0x391db939, 0x4245ac7d, 0x467a607b,
+	0xf5f9d906, 0xcc1cc1a5, 0x308cf70f, 0x3a627c53, 0x9f80258c, 0xc5886c72,
+	0xd412fe83, 0xcf58893d, 0xa7fd4c98, 0x86591d8f, 0xc658f87a, 0xb78c74f7,
+	0x9e2bc7af, 0x5fc0658e, 0xc21fda0e, 0x8969cf57, 0x963eafbe, 0xf92a74b1,
+	0xbd382315, 0x6c78cb1f, 0xd22bf62a, 0xc2cbee05, 0xf289a176, 0x70278fae,
+	0x0c5d3f3d, 0x09efc71d, 0xcef819ef, 0x566ff471, 0xc77ec826, 0x622b9018,
+	0x5c981fc4, 0x9f8edcd4, 0xde7b15cb, 0xeaea829e, 0xb73ff28e, 0x26f1e963,
+	0x2f7aa196, 0x03ae3d9d, 0xccdf32cb, 0xe8731032, 0xf88e3caf, 0x70f55a9e,
+	0x01ffd607, 0x88e3cafe, 0x55c992e7, 0x9f37fbf4, 0xa3e5dfdd, 0x7024e93c,
+	0x7dc463d4, 0x4b09f622, 0x781af7e2, 0x485f5092, 0xdf7db86b, 0x7b244fb3,
+	0x39751582, 0xea6e5c74, 0x33aafe90, 0x9e90d2bb, 0x38f207eb, 0x976715cb,
+	0xfdf2ea83, 0xf322ff02, 0x57624541, 0xc561d21a, 0x9e0261be, 0xe1e2bd3b,
+	0x56a04a76, 0x3061e7e4, 0xff2694cb, 0xd51678ee, 0x93fceec9, 0xfcf5ebc0,
+	0xf3d676ee, 0x23ed237d, 0x2f10cf8c, 0xe30f0f3d, 0xeabe0170, 0xb92f5ef1,
+	0xb983c33f, 0x52abf748, 0x87f21ee9, 0xfc24a72c, 0xe3302469, 0xae3e50d2,
+	0xe276ca68, 0xef06cdf3, 0x15f71a4c, 0x96e389c1, 0x3c966cef, 0x2a4fd236,
+	0x90f41e3b, 0x7ba7eed6, 0x3e32e687, 0xd126dcdf, 0x9aba0663, 0x03fcd2db,
+	0xeecd2e81, 0x8ff7c832, 0x4d6bdea9, 0xfa9f74e0, 0x4fba66cb, 0x87ce1976,
+	0xee81ab65, 0xab365d95, 0xf1443245, 0x3dd6b2cb, 0x577cafd4, 0xcbc5e533,
+	0x2f64ecf3, 0xeb0d634f, 0xdf2d5f28, 0x53e3e457, 0xc9add3b0, 0x65003fcf,
+	0xbe399fc5, 0x3efc53d8, 0x0f83f9b5, 0xbb2f91d9, 0xfc53960a, 0x942143e7,
+	0xf8bd001a, 0x0ee76650, 0xc92e1ff4, 0x0fd68ef6, 0x0b1dfad1, 0xa2a7335c,
+	0xdfd7d149, 0x38e8b660, 0xb03f168a, 0xdef11b1d, 0xa24d45a3, 0xd131ef13,
+	0xfe0a23fa, 0x953f78b4, 0xc7db2290, 0x3f706320, 0x20eca69b, 0xfa71933f,
+	0xd6017b20, 0x987dc971, 0xd39ab3ee, 0x6279efc4, 0x8a14a13d, 0xb3f06b03,
+	0x8cda1f41, 0xd60549f7, 0x02fee819, 0x5c02fe1f, 0x8dfdac52, 0xec4a27ba,
+	0xeb2d3f5b, 0x5bcf3ecb, 0x7a775fb9, 0x1b9f28bd, 0xdbcfb751, 0xb3247f97,
+	0x26fbe31f, 0xeed27df0, 0xcfd002c7, 0xdf8217e4, 0xafa97287, 0xdfa5e85f,
+	0xfed9b883, 0xdffff828, 0xc7a90a29, 0x00008000, 0x00088b1f, 0x00000000,
+	0x7dedff00, 0xc554780b, 0x3d9cf0d9, 0xcd8dcd7b, 0x09c246fd, 0xb8094404,
+	0x9fb1dc24, 0x4a34021b, 0x414045d0, 0x2dc8d812, 0x088d9242, 0x59b6b696,
+	0x5a4062e4, 0x7da5aac1, 0x2c142ea8, 0x11a0d05a, 0x86ec5d43, 0xba8b4508,
+	0x8ad45cb1, 0x14178026, 0xb16d0042, 0xfbdfad1f, 0xbb2733be, 0x6a2364e7,
+	0xffefefd5, 0x27a3cbff, 0x9cccce73, 0x997ef799, 0x6318c399, 0xb17fc39f,
+	0x50dff876, 0x261d8ac6, 0xd8c21b27, 0x4fab569c, 0x8a6c61c9, 0xef74676b,
+	0xa79cc624, 0x18564c0d, 0xedfd2e6b, 0xd543262f, 0x8ad79b24, 0xcb7693f7,
+	0xcd942f0e, 0x3b58eef1, 0xdaf4b7b4, 0xd5f6c468, 0x512c490f, 0x6724ac62,
+	0x618b126f, 0x9b0e576c, 0x7783cae5, 0xd0daefe1, 0x950ed135, 0x6cdb1992,
+	0xfb622577, 0x1b32dee7, 0x1ec60f58, 0x87f5e78c, 0xe3db99bd, 0xfd5098b6,
+	0x687f5841, 0xd5b23ca8, 0x467f58c0, 0xe8c79c3f, 0xb318a30f, 0xebdfca86,
+	0xaf94d048, 0xa9a198bd, 0x68fac85f, 0x0759179e, 0xb38f9e68, 0xdfca6817,
+	0xa9ad1b4f, 0xa4529d7f, 0x3fe84f29, 0xa27f5341, 0xbca6b263, 0xeb8ac8d6,
+	0x63675e61, 0x8f4b5e8c, 0xd8463cd0, 0xb787040d, 0x478702d3, 0x683b584b,
+	0x8576c572, 0xa98d5957, 0x7dec35a3, 0x9c38da0f, 0x819c5d58, 0x4eec630d,
+	0xffa899f5, 0x58df0143, 0x7be0d599, 0xdd46a303, 0xb5bc046f, 0x160d941f,
+	0x42f32fc0, 0xcec614bb, 0x1a17768b, 0xbe207a0b, 0x7f7e01d8, 0xdfdf8d91,
+	0xded1f025, 0xc335e0df, 0x816b5bb8, 0xde85fa26, 0x660c56e3, 0xdd7e1843,
+	0x8259b28d, 0x730370f2, 0x17dd7be3, 0x1059cccd, 0x5163071a, 0xde76bdfc,
+	0xe64e3abf, 0x8defe68c, 0xaedff7e7, 0xec62e245, 0x5d2b5a9d, 0x82cf0e7f,
+	0xc0633e38, 0x691fa0cc, 0x34f8cc74, 0x06f6b7a0, 0xfa016ec9, 0x366c6cac,
+	0xed17f8e3, 0x316549cc, 0x5ea7e15d, 0x31a6d78f, 0x5aabfa05, 0xd52ab2dc,
+	0x1f6ef401, 0x682bf752, 0xe303555f, 0xb1a91200, 0x965bab0f, 0xbb62d8cb,
+	0x6716f442, 0x7f43f981, 0xbff4feef, 0xf0073ccf, 0xd4b3fe3b, 0xfd07e47c,
+	0xffb559f3, 0x17e8f4fc, 0x27f77f3c, 0x83f63f7f, 0xd3ff6a2f, 0x65fbdecf,
+	0xc6eef3d8, 0x932ebb3f, 0x30746129, 0xace1cccc, 0xe90cb7af, 0x3b3ffa0a,
+	0x358f1fea, 0xb2497f43, 0xe01d997b, 0x27cd7edc, 0xc51e0cd9, 0xcc34bf0e,
+	0x37e2131d, 0x095ffb7d, 0xb1bc037e, 0xfb338018, 0x15b7cd81, 0x0ddf06e9,
+	0x924b63e5, 0x5e906b7d, 0x669ac15d, 0x95b1f718, 0xe06b1c7d, 0x0ec7e53d,
+	0x86f7338e, 0x2f3cb1f2, 0x24cbfbc3, 0x2cfb8307, 0x3a446acd, 0x5703899d,
+	0x2b2ef868, 0xd31674e0, 0xf9d02dd2, 0x7c61adfb, 0xb6a96b33, 0x96d5ee5c,
+	0x61fce387, 0xd899cf1e, 0xbc30fab4, 0xeffcc4fb, 0x047c2f89, 0xce3b8be5,
+	0x5376e54e, 0xfac85dc7, 0xbfb4f58c, 0x81fa2d1f, 0x32008e39, 0xb7b2f1c5,
+	0x9fcf34ac, 0xf89183e1, 0xf2266f3f, 0x4f82dbe3, 0xd6c4c4cf, 0x6707c049,
+	0x6f070e14, 0xd2e1cc8d, 0xed056013, 0x6e2777ab, 0xdfc0b822, 0xb3ee5451,
+	0xa4c7cb19, 0xde17b240, 0xfb07265b, 0x28cffbe2, 0x1d630fad, 0x4668e2ef,
+	0x53b491ed, 0x10fa7c04, 0x30f3148c, 0xe7801f01, 0x69d946cc, 0xf1f0441b,
+	0x29a3e0ea, 0xa8d0fe8f, 0x43af7f29, 0xa2f6be5b, 0xb4c85cb6, 0xdaac8bed,
+	0x8e02cbf2, 0xfbdaece3, 0x96d34fdf, 0x1e3a55df, 0x1d8bf2e2, 0xf3c30dca,
+	0x259521cc, 0xcfe00e2c, 0x179c66ec, 0xb864f78c, 0xc11e2b1c, 0x3adb78e1,
+	0x21cbe7c9, 0x5f9e6afc, 0xecab1cc7, 0xdaf74879, 0x418bf0fb, 0xe82b9a7a,
+	0xed3aee67, 0x19dfebf3, 0xbc019e35, 0xe325d84e, 0x207f78fb, 0x4e78881b,
+	0x419f738c, 0xc744f03e, 0xc37d420d, 0x46b9ede4, 0xabd9ff78, 0xe3990e6c,
+	0x3b21cb81, 0x3938ff1f, 0xe7c011c6, 0x058768fe, 0xd3da2d6e, 0x8fe51527,
+	0x37cf5eda, 0x52dc800f, 0x9c7be1fa, 0x99af4867, 0x25d20559, 0x28fa021b,
+	0xe4c2c81d, 0xf6878f8c, 0xceb71dcf, 0x3be04772, 0xef955ce0, 0x77c8259c,
+	0x35f4e63f, 0x1ec8cda6, 0x352c71c7, 0x9c20b26d, 0x8fc427eb, 0xb999fa03,
+	0x5b9df38c, 0x3064ac0a, 0x647b99bf, 0x25cce782, 0x44498f92, 0xf435e72f,
+	0xbea0f022, 0x5e71f00f, 0x329a5fc7, 0xfdd12850, 0xecc62683, 0x678ebef0,
+	0x1466df25, 0x648d53d9, 0xbb6d542f, 0x337047c2, 0xef1117b2, 0x2a8f816e,
+	0xa81e2323, 0x0658fe04, 0x4bd4e7f5, 0x178fae34, 0xa8067c5b, 0x261db55f,
+	0xae3fcf08, 0x2824f931, 0x3328dc7f, 0xbeb10fce, 0x7db550be, 0x7c2f2f3b,
+	0xf82d96fc, 0xff3c5dbc, 0x2db8fb78, 0xa2b5bef8, 0x9062ef7f, 0x6ddf504b,
+	0x3e6b5664, 0x0259b75c, 0x63b7a076, 0x4757f651, 0xe8ed3d21, 0x3e0cf0f3,
+	0xc618c524, 0xab59c74f, 0x85d0e109, 0x9f03b69f, 0x2be575c5, 0xa0f1d237,
+	0x50b9f4ae, 0x839039aa, 0x9267e372, 0x95d2ab63, 0xa974f54e, 0xc17cedd2,
+	0xd9b3d01f, 0x1bca1035, 0x1d38f5f4, 0x1fd635ac, 0xb9e715b9, 0x3bfea642,
+	0xb71009e7, 0xdc2d4902, 0x28c8f7d1, 0xdabe5e78, 0x32efe884, 0x33f3aecf,
+	0x6ce6fcd1, 0x673274e7, 0xa453e4a9, 0xcfd5f3ae, 0x98bccbb3, 0xa63d956b,
+	0xcebafceb, 0x218fd383, 0xf1104876, 0xa2d1be75, 0x9dd1fda0, 0xab5d3b7e,
+	0xff411b64, 0x675adbb6, 0x238c0732, 0x2575bb7e, 0x907688c8, 0x179fa8c1,
+	0xec5ddfac, 0x2d39be05, 0x4a6f7971, 0xf95874e6, 0x5926ea74, 0x6953f50a,
+	0x4073f40e, 0x15e3bbbd, 0x81a56382, 0xc1fa09eb, 0x4e0f4e38, 0xc606bcd8,
+	0x856de601, 0x32679fa6, 0xe361c937, 0x8604b4ab, 0x3b41fa69, 0x7c9fbc26,
+	0x1c19cfde, 0x9fa85e4f, 0xf1da5daa, 0x9a839954, 0xfb0954f1, 0x0ce7ed48,
+	0x9fea078e, 0xf784f1c1, 0x3c769cf3, 0x66a09655, 0x769d553c, 0x9f27e895,
+	0xae55d3f7, 0x73f634f8, 0xb64e7d55, 0x2b4f0c31, 0xfb1631af, 0x0f4dd822,
+	0xbff973c4, 0x7ecd9b70, 0x557eac32, 0xeca6bc7d, 0x9fb74c69, 0x859fa30c,
+	0x3bc807f8, 0x032418b6, 0x314ad3c8, 0xfbb291c4, 0xa3d47881, 0x011faf49,
+	0x5b559a0b, 0xa748cc3b, 0xfcfa5d8c, 0x865861f3, 0xcf1c653f, 0x38450394,
+	0x1d93469f, 0xf00cfd59, 0x71e017bf, 0xf5f061c6, 0x7fddf986, 0x119b83e9,
+	0x4dbb313e, 0x7940f709, 0xab45e3ad, 0xdb5fe302, 0xe5f62a68, 0x67dddea1,
+	0x5f261d9d, 0xfdb93ad7, 0xeb019ccd, 0xfba181e4, 0x8f5f8331, 0x0cf17b43,
+	0x5e2316b3, 0xd6031ad7, 0x587061c1, 0x3eaa2a0b, 0xc5765c02, 0x5fd744cb,
+	0xb54b2e1b, 0x01e15170, 0xf085d9f0, 0x535b58f1, 0xffbb4ed0, 0x6871e39a,
+	0x360d15a7, 0x9ca7a44a, 0x9721e912, 0xabaff39f, 0x0d7f2644, 0x08525626,
+	0xf441bbf0, 0xfa0ad677, 0xdbf06ae9, 0xd885091c, 0xbc71c6ee, 0xd2f978dd,
+	0x9e50c9a1, 0xca993dfc, 0x4d99eeaf, 0xe2600d72, 0x27bc0008, 0xbf258d4f,
+	0x3863fa0e, 0x9fa8a9be, 0xe126d481, 0xa937682d, 0xe005f258, 0x308e377d,
+	0xf3e70f1d, 0x796d4e49, 0x0cb6b172, 0x335f91fd, 0xafbda5d5, 0xe28759da,
+	0x07de9363, 0xb1458c17, 0xfed1cf8c, 0x6632fa11, 0x1f6b5ef0, 0x11d9d718,
+	0xebcd0766, 0xeca9dfd2, 0xb6c0c6bd, 0xa7337b62, 0xb28d9ffb, 0x84e796fa,
+	0x443dfbd3, 0x030e4db7, 0x37fb6133, 0x9bd43e2c, 0xbb33db7f, 0x77a08b17,
+	0xd198ed43, 0x35e34690, 0xdcfeb832, 0xfd8c51d4, 0x067bfcff, 0x8ffa0cb6,
+	0xfb76651b, 0x27bfb422, 0x4231f85e, 0xef072dde, 0x2f2dfd0f, 0x256c728b,
+	0x3f2148bb, 0x58b7f54d, 0xab2a1f5a, 0x340c967e, 0x34fa4419, 0x9e3d1076,
+	0x81e9f7f7, 0xe68d8ce3, 0xa07f3c78, 0x2f6e1ed9, 0x047fa234, 0xd01fc676,
+	0xa36b823f, 0xb5fd7fb0, 0x00bf2a76, 0x035e0e7e, 0x4ff90ab5, 0xd3fb6b7b,
+	0x075ac5f5, 0x8ed7f75d, 0x5a81eba0, 0xb97bd527, 0x07ae98b6, 0x75745d6b,
+	0xf9f025c7, 0xec5db918, 0x920afe79, 0xdbafe073, 0x88e28612, 0x3e20066b,
+	0x793e3bcf, 0x0af0b7f2, 0xafd3a8cf, 0xacf5c116, 0x735771d4, 0x3ff9fc98,
+	0x3f074fcd, 0x7fc71728, 0xef5d7f39, 0xb73d542f, 0x70179763, 0x87a9e0b4,
+	0xfbcc18e2, 0x0a5d7194, 0xadcae1ca, 0x709ce32b, 0x0e3bb305, 0x9f29e1f5,
+	0x3e891c5c, 0xfa54134a, 0x2848c670, 0xabe718e7, 0xefb89e8f, 0xbfabe406,
+	0x51d9356d, 0x7ec0785f, 0xae7e80b1, 0xf89db86e, 0xbc6dca12, 0x1939ef7c,
+	0xfe5a8dcb, 0x87bbf059, 0xa2589452, 0xb74e4cbb, 0xdba44c81, 0x8def2dea,
+	0xb0f23a47, 0xfad3d20e, 0x8bfef876, 0xbd4f7a00, 0xbd7e8e5c, 0xffcab9fe,
+	0x9eae411e, 0xc9cefb86, 0xab57f871, 0xafaef119, 0x0e34815f, 0x41313c7c,
+	0x3670f1f0, 0x156ded1c, 0x4c97cc1e, 0xf73f2cfc, 0xa86d419b, 0x39fcb7bf,
+	0xcaebca4e, 0x67ebe7af, 0xa507fa3f, 0x176cfe7b, 0x6bd7af74, 0x200c234a,
+	0x7e7e8a15, 0xe20534ad, 0xb9fda346, 0xacf244c9, 0x1f6fc772, 0xd3da35fb,
+	0x7d1c5a6f, 0xf69bb415, 0xe505191d, 0x494d3b85, 0x311a7c25, 0x2f084a52,
+	0x696ff81e, 0x0b0f087e, 0x5667d99e, 0x77fbf206, 0xb1f08427, 0x99936770,
+	0x8f0aec0d, 0x7326faa2, 0xbc044c6b, 0x478db7d4, 0xfb9f96be, 0x7ef119a7,
+	0xa58905ea, 0xe78044e6, 0x6e717da6, 0xa3c22701, 0xde2753c0, 0x803df574,
+	0x1553a417, 0x57e7fa4f, 0xc67a7027, 0xa23c94fe, 0x00ff27e7, 0xcb78a9f0,
+	0x57f9c0e2, 0x1c67793c, 0x6303e3ce, 0xdfeba70d, 0xc2714cac, 0xde4c2b7b,
+	0xf95d3b14, 0x96478afd, 0x6fcdbd10, 0xc8c29259, 0x1ba670ee, 0xd33cd046,
+	0x37737e71, 0xcdf9a54e, 0xbba26e63, 0x2ec8df8a, 0x353e2bb4, 0x119de2b2,
+	0xc175e2f8, 0x066c1f17, 0xfbf3046d, 0xef73c840, 0xcb8fb2eb, 0xca183bcd,
+	0xdcaa25c9, 0xfca88b64, 0xe9e395a9, 0xe08304f1, 0x70f2e02b, 0x6f72dd7a,
+	0x51f3f110, 0xfd153396, 0xbf030f47, 0x7d21f797, 0x2dcafd0e, 0x7e3952e3,
+	0xed4cbdb8, 0xc2290ec0, 0x683f58fb, 0xac3b7f22, 0x273cfb1d, 0xe2c5fef4,
+	0x28ee30fd, 0x07f3e409, 0xd36edc29, 0x095fefcf, 0xbf028e3c, 0xafa30b20,
+	0xfce6fe30, 0xe73565be, 0x7ddf956f, 0xf9f18a93, 0xff388727, 0x7cccbb60,
+	0x9520571a, 0x914d379e, 0x58581e48, 0x6f5f1220, 0xd6be0931, 0x04e9573e,
+	0x38c46dc6, 0x60bd2746, 0x7f0409ff, 0x3de787b2, 0xe4c2ed08, 0x17df07a1,
+	0x953fb5d7, 0xfe0be76f, 0x3ff9057f, 0x6cffc43a, 0xeefbe723, 0xf097adf5,
+	0x3df26957, 0x57e46bf6, 0x82af4fe0, 0xe1726afc, 0xdede3fbc, 0xf456e47c,
+	0x3e55eb1b, 0x3d4fcad5, 0xbd3c569f, 0xcf53e88f, 0xefd71fa7, 0xa7c8894b,
+	0xb4a5f73b, 0xd3e245e6, 0xaa7e56ef, 0x2754fbf0, 0xd879553f, 0xf2f51c1d,
+	0x250481f0, 0xdf843ca2, 0x2bac3621, 0xa774a9fd, 0xdf82dbd2, 0xb942f557,
+	0x2a9749d3, 0xa5d275dd, 0xf9fa774a, 0x7fa7e16a, 0x158047fe, 0xb90df9e2,
+	0xf7da1863, 0x45b8c153, 0x03054dbc, 0x6b1f34a3, 0x4d5f11a5, 0xfef3e311,
+	0xb7184983, 0x79c21241, 0xf4bff54d, 0x0df7be19, 0xdfa2f313, 0x2dcf767b,
+	0xab57da0a, 0x990ead92, 0xe68f84e4, 0x5eec8efd, 0xb00f7189, 0x8e51aaa3,
+	0xdb47f74c, 0xddb96ed1, 0x32709eea, 0x7f448411, 0x8dcbe489, 0x07b2656b,
+	0x2e3c79ad, 0xf2611ae1, 0x9efa5fa1, 0x2be345c1, 0xfd5e1829, 0xf73d1a23,
+	0xb93dd1cf, 0x2157c606, 0x8c3f87c6, 0xc2ceddd7, 0x0f5faa7e, 0x46bbce5d,
+	0xa1c5440a, 0xe3b12c3f, 0xddb87061, 0xefe32da7, 0xfd4252bf, 0x777eae5e,
+	0x5f4df47c, 0x3aba05df, 0x1b787e0e, 0xba7c8d5e, 0xcb3cf4e1, 0xa9e8e1cf,
+	0x031fc894, 0x615c9bbc, 0x44bad57e, 0xbddef72e, 0xcce280bb, 0xec0ff785,
+	0x2e6de5c1, 0x423cd7c5, 0x2af79e7e, 0xa8af503d, 0x03f9eef6, 0xe8efe445,
+	0xe822d3c7, 0xdec22ffc, 0x5a563b3d, 0x9ddf0327, 0x3f641d65, 0xd8dac779,
+	0x6f7e08d6, 0x7ea7624d, 0xd430f260, 0x86afea4b, 0x4f13b0f8, 0x329e276b,
+	0x7f0aea0e, 0xe5571dfc, 0xc3effca0, 0xe109f915, 0x1d6b62d9, 0xcf308aef,
+	0x4adc047b, 0x46667ef6, 0x3f937639, 0xb8f084c7, 0xb1acac69, 0x28947f63,
+	0x2b0144e4, 0xec9571c0, 0x246e2ebd, 0xec80e3e2, 0x3d5c01b0, 0xf76673fe,
+	0x11c3cbc0, 0xf5fb47fb, 0x0f31c984, 0x67e278e1, 0x3e05e636, 0x45c44578,
+	0xf51391e0, 0x44e5741a, 0x225f7dfd, 0x9bbc078f, 0x4eb82768, 0x2534292e,
+	0x1af37bc3, 0x6f3183ec, 0xbfea2191, 0xf3d0a89b, 0xad4ed14c, 0xebc95197,
+	0xd5ed99b5, 0x23dfe691, 0x6a5f53b9, 0xdefb35f0, 0x0fe8cd3b, 0xabe16fff,
+	0xc0d8fef0, 0x8d5f8c18, 0x746e4895, 0xfc4663a5, 0x63fc1e70, 0xfbcf13f4,
+	0x7cf2bc79, 0x02c38f09, 0x735db93c, 0x8d764493, 0xf87ae74a, 0xabe0273e,
+	0x28a8ffbe, 0x9db913fb, 0x8557c2a4, 0x6abc7eab, 0xd757907d, 0x1bfbcb86,
+	0x293f071b, 0x2fc23e61, 0x55d00dd9, 0x8cd5b1b5, 0x5c69e77c, 0x667f426f,
+	0x65fbc2ba, 0x29a0d746, 0x6ef37461, 0xdbde91a2, 0x267a3f9c, 0xb585c7d2,
+	0x77f38616, 0x871739ce, 0x8315f37a, 0xc8d1b5e3, 0xe4caaf7f, 0xc707778f,
+	0xafdad6bf, 0xef0c6b1f, 0x39c68737, 0x84d0b2b9, 0x654669ea, 0x8c352993,
+	0x4c4dde8e, 0xfdeae7a0, 0xf2854a8c, 0x1a3ef062, 0x16e7f787, 0xd5e49bca,
+	0xc344c707, 0xf13519f3, 0xc02afe9c, 0x8e72a6f0, 0xbea68f26, 0x5347cf47,
+	0xe535a24a, 0x3dad45d9, 0x96252e11, 0x044762c0, 0xa926f4fa, 0xf535f9f4,
+	0x37e81382, 0x5ffedd02, 0xab1617a7, 0x9355a17a, 0xa92ebecf, 0x8d485e8b,
+	0x2d1617a4, 0x0f115253, 0xd165be6a, 0x2252e34b, 0xb5c385e9, 0x1472df3c,
+	0xcf5e19e4, 0x70bd014e, 0x985e9875, 0xd13a2362, 0xba482bb7, 0xf1505e8c,
+	0x99cb1df4, 0x2217a8c3, 0x24f8f5f0, 0xd9b85ead, 0xe17a4e5f, 0x533229e6,
+	0x497df3c2, 0x56eefec2, 0x4ca6142f, 0x11c9b2a7, 0x106e811d, 0x3a17189e,
+	0xf8eef22a, 0x3f0fd41e, 0xbc13b24e, 0x3ba27af4, 0x39799e39, 0x04bffe39,
+	0x2f8e555f, 0xbd912a9a, 0x18ee95d7, 0xe9a2efd1, 0x16ade512, 0x7c72e1ed,
+	0xf1c81951, 0xae45d61c, 0x0eae50ba, 0xbcab9709, 0xeb9bb57d, 0xce634f01,
+	0xf9e14b2f, 0x0e4cebad, 0x35bfd42b, 0x85876724, 0x271fd9cb, 0x2b672375,
+	0xe0bd9cb8, 0x45f1e82f, 0x4263cb71, 0xcafc82be, 0x8ba04475, 0x9ad1a569,
+	0x06270375, 0x19ef29ff, 0x64578fe4, 0x27bc878a, 0xc1c1fdf4, 0xf6821a7f,
+	0x0f1c61ff, 0x49b7efe0, 0x9f80ef5c, 0xe058fd71, 0xdc39a2fb, 0x2edaae5e,
+	0x1baafc13, 0x6f57bf38, 0x3d0e2893, 0x3333ff3e, 0xf0f6ede5, 0x971a86b8,
+	0xf0bff156, 0x9955d695, 0xc4937cf8, 0xfc0acce3, 0xcd87cb0b, 0xbf242c73,
+	0xad06fe64, 0xe96250ff, 0x4aed8fd8, 0x3639277e, 0xec8bfc91, 0x7e89556f,
+	0x3ed36e16, 0x16ac72af, 0xe48057fe, 0xf6bd5576, 0x5fc15ef9, 0xde40e573,
+	0x7bcf9833, 0xd244764d, 0x4ebd6a67, 0x9bfc5478, 0x780168f0, 0xf0eaf90c,
+	0x44dfddb8, 0x0baf9379, 0x5ffcfd07, 0x3e46e554, 0xf6fc821b, 0x7249c19e,
+	0xcf32fbc3, 0xdf59ce3b, 0xfe3c3537, 0x8ba0bafc, 0x55ee0c57, 0x9aaf58e9,
+	0x6fe73fe7, 0xca1a8704, 0x306cf5a5, 0x37f36efa, 0xa76bf568, 0x48506f3e,
+	0xecd6975f, 0x43cb6e94, 0x5abcf3b8, 0xf8fd6f85, 0xb65e780c, 0xd437f414,
+	0x716f9d9f, 0xc5a8fc63, 0x517241d6, 0x05cfcfa5, 0xe053fcbf, 0xc74644e3,
+	0x8d3bd258, 0x2ef89d92, 0x087ae360, 0x728362f0, 0x8fbfc8f3, 0x5bb07817,
+	0xde392f88, 0xe427803c, 0x517ef1db, 0xddf6e7ae, 0xee39ff0c, 0x4c9bfbff,
+	0x91a2aed8, 0xa0236baf, 0x8c3be986, 0x84a61447, 0x1b4f73e1, 0x92adefa7,
+	0xe0f86e83, 0xe7c197df, 0xdaa8f065, 0x9f165f23, 0x4f3e5c2a, 0xc0827e12,
+	0x582d8b6b, 0xfc798052, 0x3363923d, 0x53c7f656, 0xbae4e15c, 0xd8df843e,
+	0x3d578e64, 0x80dfa3dd, 0xf2d4a6eb, 0xfa201c24, 0x7636c5f4, 0x6f6f7d42,
+	0xcb855f6e, 0xa3be351d, 0xbe6abf5e, 0x1482c3cc, 0x66f352af, 0x83f76523,
+	0x7257a889, 0xa44f5a25, 0x4660cfe6, 0x40dfd7c4, 0xbe9313ff, 0x66967fa0,
+	0xecedebe7, 0xe13b7dea, 0x6e5136e9, 0xb4761ace, 0xa17be014, 0xbae0764e,
+	0xca0c5bb4, 0x9e506b91, 0xe43131ac, 0x4d070b8f, 0xa3df2e4e, 0xbf949f49,
+	0xfa84c272, 0x82731cb0, 0x09ffc798, 0xc4271bd7, 0x3d02de78, 0x2898c4ec,
+	0x093a1fab, 0xd1c53fde, 0x9cf30e34, 0xacb48753, 0xe4ee38c4, 0x2391fb22,
+	0x32d5cf88, 0xaf1e7c43, 0xeb3f222b, 0xe2243bce, 0x6f0037af, 0x94ea7ed8,
+	0xe047239c, 0xc97565bd, 0x75295314, 0x0d3cdf26, 0x6507a5ca, 0x66f50287,
+	0x831b18df, 0xfd09d7f1, 0x830ee665, 0xae0b194f, 0x6957f8c4, 0x71531dcd,
+	0xe7a015dc, 0xe15d1f30, 0x8a97196f, 0xb5128de5, 0xe6dd78f7, 0xe09d2054,
+	0x21e67386, 0x5079fef0, 0xd9eec10f, 0xb679ebc8, 0x754c0e48, 0x27aff3cd,
+	0x8b6f3879, 0x598970c4, 0x2b58ec10, 0x8567d7ef, 0xa11b837e, 0x63e7943d,
+	0x4b77ec7b, 0x90d3cdfc, 0xd49c618f, 0xfdb2c47e, 0x41e72f28, 0x82cc783b,
+	0xfea0ee5b, 0xb128df6e, 0x62afe834, 0x47cfce44, 0x4d077f60, 0xb47e7c0e,
+	0x671457e9, 0x4bef099f, 0xcb67fe87, 0x9d68e3ad, 0x3b258fd7, 0xb92abbad,
+	0x5677f099, 0x9b4b19f0, 0x47baf64a, 0x6f3daa62, 0x3edb8f0c, 0xe89bdc96,
+	0x65d2de78, 0xad5e70ab, 0x031ce4de, 0x5a39d5d1, 0xe0127897, 0xf43b06ed,
+	0x83c79984, 0xcc27bd9b, 0xb37261ad, 0xf10fb939, 0x42c69ce5, 0x3aea2f92,
+	0x6e32f70a, 0xb4adb467, 0x36987ad1, 0xef1b3d93, 0xd0b4af37, 0x5c1df6fd,
+	0xb567b4bf, 0xdc7e83be, 0xcc4d1b07, 0xbe5c630b, 0x88c3cfb1, 0x9bb60fc7,
+	0xdfe4de88, 0xcd8e4898, 0x5be3c1df, 0xbf791bcd, 0x72e36c1f, 0x5337e8ad,
+	0x9ebca3c7, 0x42c9f20f, 0xafcad3f3, 0x1e02ca79, 0x89b75d47, 0x9557cbf4,
+	0x53e45d53, 0x5f1455cf, 0x7aef4eec, 0x77aa8c58, 0x5ebe31f2, 0x7335955e,
+	0x196fd203, 0x03068c4f, 0x087d3fbb, 0x2c6569f5, 0x46d53bc3, 0xa712c3fb,
+	0x96c7ca07, 0x8a7e77fa, 0x6a515ff4, 0xde1e2073, 0xbf23f847, 0x78a44dac,
+	0x395f3cea, 0xab3e79c1, 0xdfcf3821, 0x356fa5a0, 0x2ddd386e, 0x9eb86733,
+	0xce6688aa, 0x7f477ab0, 0xef0a7402, 0xf928d599, 0xdb96126c, 0x42192ff6,
+	0x6b8d3b65, 0x6bfbd0a9, 0x8f8e8963, 0x4adcf0fe, 0x7799d3a4, 0x6eed0c4b,
+	0x42af75e6, 0x41bd57bd, 0xd6bd7052, 0xfb79bbfc, 0xc71ed0f9, 0x9c57f47d,
+	0x15ae809e, 0x8deafba4, 0xfd53f7eb, 0xfd82922d, 0x4e54dd1f, 0x57854276,
+	0x4b7cf466, 0xa001f91b, 0xd30f2897, 0x905395ec, 0x5857abee, 0x57d798ec,
+	0xc7a547e9, 0x661f9136, 0xc9a0c756, 0xb2516ed0, 0xe50e9112, 0xdea75269,
+	0x78f17a44, 0xedb57d39, 0x25903930, 0x7c8f4f2e, 0x6f4a825d, 0x90dbd232,
+	0x6f35bd10, 0xfd23322c, 0xebe3e07d, 0x5e908a11, 0x571f1e91, 0x262c2ae3,
+	0xe9523edc, 0x07e80ab1, 0xe8e5f4e5, 0xd3fa3bb1, 0x02f81dd3, 0xcee47eb4,
+	0x771a3695, 0x37ddaad1, 0xc47ce52f, 0xab459c70, 0x8179487e, 0xff1e4679,
+	0xeb88e156, 0xd24f30da, 0x2e0f1c65, 0x997fa733, 0xf1c6e8f1, 0x10d8e48b,
+	0x96e803fd, 0x01bac59b, 0xe5039443, 0x5698c4d3, 0x31e304a3, 0x51187525,
+	0xe14f0f3f, 0xc23979fa, 0x3325eb0a, 0x67281f5a, 0xacee081a, 0xa3d7c297,
+	0x876476e6, 0xa67b63c7, 0xb06efea8, 0x189fde27, 0x5851e50d, 0xff404db6,
+	0x5745a26e, 0x60a061ee, 0xbc516313, 0x5eca2a9c, 0x4c9d0328, 0x479f54cc,
+	0xd12f72f3, 0xebcd1bde, 0xc79f500f, 0xead5f2f0, 0x7f5e18f3, 0x99e7a334,
+	0x7013b129, 0x14cf573d, 0x237c66bb, 0xfd0d4951, 0x62d957fa, 0x6eb489a8,
+	0x9a8a6fa7, 0xaf146787, 0xf4fff976, 0x5cb5e606, 0xf90098d6, 0xe9d34e07,
+	0x867b30d6, 0x13ca7589, 0xa4b23d04, 0xae0bb238, 0x2f01bfc7, 0x26818c13,
+	0x06b06f6c, 0xb0573f08, 0xfbf7f615, 0xf65222e3, 0xdd05e7e0, 0x56a86bfb,
+	0x6ff30fde, 0x4b632725, 0xc524ae81, 0xe590f94e, 0x46db994b, 0xed4b94eb,
+	0xcc49bf68, 0xaf9cf869, 0xfbe17657, 0xefa8f1d5, 0x802e32ef, 0x7c2131b4,
+	0x1fa91b5c, 0x07675c75, 0xb798bc5b, 0xfd618ef4, 0x74e0c26b, 0xf08ae916,
+	0xe22264b4, 0x098daced, 0xac728bbc, 0x8447fcfa, 0xeaf1b3d7, 0xd64e5173,
+	0x95d74f3d, 0xa4febc4d, 0x78272b94, 0x64b3bde1, 0x6814a4e7, 0xd0c73163,
+	0x28baeaba, 0xbee5debe, 0xed17b764, 0x5a73a023, 0x32cde497, 0x76317f93,
+	0x5f0ce488, 0xd043315f, 0xb114b1f5, 0xaeb2d94e, 0x0ded05a6, 0x4196c3f0,
+	0x0ba5cf7d, 0x4c4f8c66, 0x9199b7f4, 0x401baaee, 0xcb15e26e, 0x68dd5798,
+	0xf75a4676, 0x56452a76, 0xf59a17e3, 0xfab2fbc0, 0xb9424cad, 0xafcb873c,
+	0xc7e3bb45, 0x39517961, 0xbd6ff7b7, 0x0438e766, 0x69ab9941, 0xb1675e25,
+	0x8a687dd6, 0x74517ea3, 0x4edc25fa, 0xd449ab2e, 0x03ad6794, 0x75c20e6c,
+	0xfef5da99, 0xa7b9e44c, 0xe30fd0b3, 0xde04d38d, 0xa5ada7a7, 0xb774f7bc,
+	0x44e73d76, 0x026332ff, 0x2246bb8e, 0xb8e6093d, 0xf066ed82, 0x6c35c219,
+	0x7dc0385d, 0x67d0a35c, 0x97ce5c95, 0xb4a97f0a, 0xdd17af82, 0xc2666597,
+	0x7f71f2df, 0x84d8fd6a, 0xa3f71421, 0x7f1efef5, 0xcfeb098b, 0x137e0258,
+	0x4ec97bf7, 0xda0ea598, 0x86e60ceb, 0xe3c61b8d, 0xe755fa45, 0x3ee50d35,
+	0xf5edc3ae, 0x10ed09be, 0x5d668f98, 0x6d0a8ce9, 0x5b46acb9, 0xf733afd6,
+	0x3b3908a5, 0x7e400d80, 0x19229adb, 0xf944ff89, 0x5791d674, 0xefae0a03,
+	0x194b6d74, 0x0747eb4a, 0x295ca1a1, 0xf9dca9bb, 0xe6a81cdc, 0xc78c07fb,
+	0x25cf895b, 0x26e5dfe2, 0xe310bda7, 0x0fdc5529, 0x35cf819e, 0x7d7806eb,
+	0xcb5dcb2a, 0xfee14d7f, 0x6b9b72da, 0xc12dc531, 0xc13cc54f, 0x98d3a358,
+	0x307c4790, 0x7b3c922d, 0xabffd05b, 0x8c15b40c, 0x25121fab, 0xe4dfa5e0,
+	0xcf8c3e1e, 0xb963fbf5, 0xfd53e317, 0x0aef000a, 0x779e05ca, 0xaef2209d,
+	0xb33e8037, 0x5c9fb5c1, 0xc7f48d3f, 0x176e66d5, 0x2aac7c8b, 0x6b8f9c4f,
+	0x7a13b1f3, 0xe50cf81e, 0xd61b0dce, 0xdb99ff92, 0x1ebf99c7, 0x8d9f77c6,
+	0xddbeffbf, 0xcea98fdc, 0x35f24fdb, 0xf4f229d5, 0x36ce4269, 0xe7559c82,
+	0x454df80f, 0x64fe87b9, 0x37bd1e49, 0x87bd73bd, 0x14a4e493, 0x1cf86318,
+	0x78aa57e4, 0x0a599def, 0x41652bc1, 0xcef384bd, 0x6865de62, 0x6819ff37,
+	0xd7abef0a, 0x533ebdaf, 0x56ebf491, 0x541d2067, 0xf58ab5a5, 0xdfcf5faa,
+	0xe7fe82fc, 0xf7d41ca2, 0xebe7d30f, 0xec4ced04, 0xf144b847, 0xde2e3540,
+	0xad917e9d, 0x30dc8d06, 0xf234a3d9, 0xad7f4265, 0xd0e899bc, 0x471b99ca,
+	0x9f3cd1f4, 0x79a01ce4, 0x40b8b93e, 0x1aea9e53, 0xcb7fa9ad, 0xd94d22b4,
+	0xa6bd7692, 0x49b94dbe, 0x7fee8e53, 0xac7ea6ab, 0x73cd36e3, 0x4ecf4bcb,
+	0x662e0093, 0x0dd5b7ab, 0xde03a870, 0xd2109118, 0xace2031b, 0xaa287eb4,
+	0x6c7e46c0, 0x004b60dd, 0x832ea4e9, 0xbcb0d3ef, 0x002d24b4, 0x6c7b58fd,
+	0xf5c216b7, 0x0c6f92f7, 0x62ac73c6, 0x7755f61d, 0xf95e8d33, 0x2be70665,
+	0xaa657af5, 0x5c1a687b, 0x9929b6f3, 0xf068048c, 0x358fc42a, 0x6bbb718a,
+	0x7437df17, 0x91cfea35, 0x64967ea1, 0xc54a0e15, 0xaf3b7776, 0x91b3bf23,
+	0xbd42fbd9, 0x91bf62cd, 0xda999777, 0x1ab635e5, 0xfdfc619c, 0x0f2e5487,
+	0xb98d721e, 0x32f8898e, 0x0deac3ea, 0xf7e8cdc7, 0x689bfb57, 0xd720e1dd,
+	0x16f7da71, 0x9af42fba, 0xd70a23b1, 0x1f553d3d, 0xc2e23d79, 0x6bcd3354,
+	0x65e390bf, 0x7f61644b, 0x2aa4f8f3, 0x7676cb97, 0x1b8c6404, 0x93e36954,
+	0xf20d397a, 0xded76359, 0x7140223a, 0x11d3900d, 0x0f5545fd, 0x3cd6dc80,
+	0x3af6859f, 0x7943275f, 0x9ae39023, 0x4553d3e7, 0xcb006b3c, 0x6fc275c3,
+	0xe7110cb8, 0x1e0cdbf7, 0x5dc37ebf, 0x087f7044, 0xc21216e3, 0x0e2f0f53,
+	0x69c1cbc7, 0x15bf9a87, 0x7bd7d2f8, 0x6be3832a, 0x3a723747, 0xceedca90,
+	0x1f3052cd, 0x15c7a885, 0x033af445, 0x2014a5d7, 0x0ab6fe1e, 0x67b2a8fd,
+	0xf4fa8625, 0xd12f4daf, 0xae33263a, 0xa438c24f, 0x6954bf58, 0x7fa19652,
+	0x40e0f400, 0x31baa6fb, 0xaaafe340, 0xfbc2682f, 0xce3c8d55, 0xf1e069e2,
+	0x4ad66acb, 0x6ee6dd7e, 0x2a6f75c6, 0x5207efa7, 0xc7bf1a15, 0xbd44ef0d,
+	0xf9f160de, 0x2b7fc424, 0x76ca57c5, 0x941d6c4b, 0x4eb6af37, 0x3d2f648b,
+	0xe2fde02f, 0xf5c3336a, 0x1ed095e6, 0x6b3fb1e7, 0x3be319bf, 0xca253b65,
+	0x5bea7867, 0xd6be62a6, 0xd0c4d8b3, 0x77c7508e, 0x7e482b26, 0x84739be4,
+	0xc787391f, 0xcc5cde2d, 0xee790fed, 0x9d57147b, 0xc8fc255e, 0x0e281739,
+	0xf08566f4, 0xb8f2e723, 0xd43bf4c7, 0x8e968ebc, 0xd3d53a55, 0xf56e9e9f,
+	0xe955faf4, 0xa76a4773, 0xf508319d, 0x1c7be03a, 0x1fdef1fa, 0x4c27ca32,
+	0x426cd4bb, 0x2af7bf3f, 0x281ee3bb, 0x2b7d940f, 0x9ace5e31, 0x95c1f3f9,
+	0x5f2e249b, 0x3e715b94, 0x86667599, 0x020239fb, 0x53c4014f, 0x43a94e32,
+	0x357e468a, 0x1ca37b06, 0x0175c1ca, 0xc1a7c700, 0xf949dfcf, 0x065f2e25,
+	0x69b6973e, 0xd6bd184d, 0xcaa2b908, 0xe66601bd, 0x017189de, 0xc6067d63,
+	0xf8987717, 0xd16c963e, 0x1d3edf5f, 0xb5c127e3, 0x65ffbc4d, 0x7fcb5571,
+	0x94fe10c9, 0x75c11673, 0x5da3c8a2, 0x8a5034bf, 0x744ec8f3, 0x6f315307,
+	0x12a47d99, 0xef5d5af5, 0x1fd0f397, 0xabc3d751, 0x91e6f471, 0x13bc6c1f,
+	0x32bb6be6, 0xdb69f08a, 0xf48edeb3, 0x243beb6b, 0x57de5235, 0x2a2e7abd,
+	0x53fc0dcc, 0x5d37b17d, 0xbda4f2d5, 0x6dfd68a7, 0x78dcffb5, 0x698b8fe2,
+	0xf76d5f74, 0x81ee3127, 0xf32cff62, 0x065e9127, 0x911c6cb9, 0xc2a7aa1f,
+	0x68e75d41, 0xf085bbeb, 0x3c1de775, 0x7ad8c1bf, 0x654667ea, 0xfd0e548b,
+	0x9a7262dc, 0x47de0062, 0x709b4c7a, 0xd5d71732, 0xc1c1b8f3, 0xa5c52b5a,
+	0x53b5ad5e, 0x469b5839, 0x156bd7e5, 0xec915eba, 0x34e104fa, 0xbe9ac5f5,
+	0xace6b708, 0xa7985d87, 0x44d07b3c, 0xaecc60f0, 0xc17de7d9, 0xea3c532f,
+	0x5f5f0342, 0xe67f8ea2, 0x4ee60cc5, 0x8a6371e4, 0x0bfd2c57, 0xc54739c9,
+	0xf7e8f1b9, 0x5b6cae0c, 0xd2cf6585, 0xee9d694f, 0x282fa03b, 0xe3c8d1ef,
+	0x2d8d55a7, 0xf4b9ceb4, 0xb3f50262, 0x1e31d0c3, 0x82fbccf9, 0xde3beb44,
+	0x4e673c69, 0xd288fae5, 0x7c451b3a, 0x3c5a4879, 0x1fe3c1c1, 0xe5f8e06c,
+	0x8245d579, 0xfaae9a1e, 0xc3e21c47, 0xf5a154ba, 0x6407d522, 0x19acb38a,
+	0x8f6711d3, 0x6e317720, 0xde93eb8d, 0xf494eb12, 0xbd99a7cf, 0xc07d717d,
+	0xfde44334, 0x50b48ecf, 0x75d71f7e, 0xc7f53297, 0x3f83d7f7, 0xca88eb43,
+	0x8ca745fc, 0x6ef9ad78, 0x196acff2, 0xfdf0c1fd, 0x320fe806, 0x34bdfbcd,
+	0xae564b7e, 0x85d6ef26, 0x98335bfe, 0xaeaa7802, 0x20173003, 0xd7455f9d,
+	0x2af7ddf6, 0xa8d637c8, 0x2f51878e, 0xf2067eb0, 0x5abcaa27, 0xe84b273b,
+	0xee0cacad, 0x89a7173f, 0x2fc2b2fd, 0x4267ff02, 0xe27ff71a, 0x5f8d32f6,
+	0x7e4fbfa4, 0xc7ec5591, 0x7978001e, 0xafb8c23a, 0xc2e318cc, 0x1d6cd976,
+	0xfa0337d1, 0xcfd2ba46, 0x1f1a4673, 0xd3cfca97, 0x28f39f91, 0xacfc4ece,
+	0x464877ef, 0xcfb40ce1, 0x5fb37756, 0xefc12aeb, 0xc1181b55, 0x8121b3cb,
+	0x6d8674e0, 0x7a018d70, 0xd8e3033c, 0xd678f40c, 0x878e8ae5, 0xaedd67f6,
+	0x109884e8, 0x886f57ff, 0x73ab76a2, 0x872c4a6f, 0xf7a08d72, 0x2b22c6f6,
+	0x06e679c2, 0xfbea77ce, 0x4f4c09de, 0x6a1a3e44, 0xde6330ea, 0x2dbfa7ae,
+	0xa1bbed0e, 0x222727bb, 0x4e3775ff, 0x779f3a77, 0xb445d2d5, 0xb87f155e,
+	0xd0a1627e, 0x3a7b9e63, 0xd57ca11e, 0xdbf44652, 0xd5afebb4, 0xfeedbfc8,
+	0x9e6bca68, 0x7a4d13d9, 0x3cd3ecf0, 0xd4c679af, 0xaa60eb4a, 0xf476cb77,
+	0x7c2f3d07, 0x9ff230fc, 0xa1459767, 0xdeffbd78, 0xb6aeb873, 0xad1d7fd2,
+	0xfedca8c3, 0xd541f2dd, 0xfec8bbd2, 0xd71fcb51, 0x37285d5a, 0x2e5c8ddb,
+	0x867c6c2c, 0x9785d9ec, 0xfcc20c2e, 0xff3d99ef, 0x98e50c3d, 0x86178fe7,
+	0xe7ec3ed1, 0xe7c30c2f, 0x6ba2e79e, 0xef25d922, 0x8e38f066, 0x0e731faa,
+	0x9ef13519, 0x73fe8209, 0xc62bac56, 0xb6864d73, 0xaf84714c, 0xd7da1b1b,
+	0x8617fb40, 0x389b1e1e, 0x8addac37, 0x392fdf20, 0xf46c65a4, 0x747a309c,
+	0xb8d49866, 0xd38ad7d1, 0xa11fb130, 0x34132e33, 0xd5843ed0, 0x6f3f2341,
+	0xa1f242a0, 0x8a7600ff, 0x4bd91dbe, 0xe44f6f67, 0x5d1832e7, 0x1d220613,
+	0xf697dedc, 0x75af3387, 0x3d986e5f, 0xa0683d34, 0xfcd3ddf5, 0xefd90096,
+	0xefad2341, 0x99a4ed56, 0xcd883ee7, 0x5dc42291, 0x0e3ebffa, 0xfe3eb6e5,
+	0x6ec6b3d2, 0x2fef0898, 0x08ecc15d, 0x063c7d3f, 0xfb9db17e, 0xf8d75992,
+	0xd07c128c, 0x5c95ed8a, 0xa7c71d83, 0xb78899da, 0x3efc0886, 0xb3f39d2d,
+	0x822b47ca, 0x8ac7c206, 0x2e6b18e0, 0xe789275c, 0x7136a00d, 0xff0ae8df,
+	0x1f18ade4, 0x96e97158, 0x9ef081e8, 0x626e8715, 0xdbbf20f7, 0x62fb58c7,
+	0xb7df4bbb, 0xd10bcd4e, 0x4a7dfe89, 0x2127d73a, 0x7eb2207b, 0xfbfc178b,
+	0xd8b9e95d, 0x3d6cfff4, 0x52757e07, 0xabf250fa, 0x07b8f067, 0xfaf5abf7,
+	0x6abb9541, 0x7e04e3bf, 0x53ddcabb, 0xff80bf64, 0x5c77724b, 0x7af542ba,
+	0x922527fa, 0x4a687c5f, 0xb5d312a2, 0x87463eff, 0x6baf2121, 0xe69daaff,
+	0x59f73af1, 0xf2717fd1, 0x8a6796fc, 0x90a493e4, 0x4971e75d, 0xa9649f7c,
+	0xd4f587c9, 0xf0a70571, 0xa754e9b8, 0xaa86f289, 0x87daa7fd, 0x3c1dcf9d,
+	0x8bdaa8af, 0x74185daa, 0x94ec38f1, 0x15f9fc21, 0x106beec9, 0xf588ef5e,
+	0xf247432d, 0xf6b98cfb, 0xcb747fa1, 0x7e2312cd, 0x4aa2f617, 0x527b7a9f,
+	0x1ddf6f5d, 0xba0a371e, 0x5b2bf954, 0x9eaa17c7, 0xba7a11aa, 0x2274f51a,
+	0x34cfe9ea, 0x5fbeb776, 0xf4aed3d0, 0x08c76c64, 0x1b0e43be, 0xeefc915d,
+	0x375fa213, 0x3dae0fe2, 0xe2162f83, 0x7058b378, 0xdd8bfa19, 0xb0974fe9,
+	0x5e7bba63, 0x5f9fc693, 0x08d78f8e, 0x7c577ffc, 0x9c3feabc, 0xbeed3b8f,
+	0xba23ce6e, 0x7ef0a332, 0xb928dba4, 0x7c885826, 0xc51367f7, 0x4af5f5c9,
+	0xff063e85, 0xfbbf4355, 0xf406ccb2, 0xdc153c77, 0x578bafff, 0xe4bf235e,
+	0xd5d69925, 0x0eff042c, 0x1705b3ed, 0x44ad74ed, 0x33ae183b, 0xfe50b5b6,
+	0x76854660, 0x831875fc, 0xe477f6c5, 0x32fc7ad0, 0x44e29636, 0x3b78fe65,
+	0x4baf23ee, 0x86f5a7ae, 0xb2b2adde, 0xedfebc35, 0x8f2b7d6a, 0xf74d2d9b,
+	0x6b79e72d, 0xab2fbe8d, 0x5af3dbd1, 0xc45a3ffd, 0x03b3ba7d, 0x71b26f9e,
+	0x26407279, 0x78c6aefe, 0x998cd5d1, 0x4bd5f5d5, 0xdaa39c79, 0x23f68cbf,
+	0x1fe78957, 0x4714831a, 0x5417fca2, 0x9f5fe53d, 0x9cc87c82, 0xd54f7cda,
+	0x2724c396, 0x849f79f5, 0x3c8f2e06, 0xf3c8b295, 0x7d68930c, 0xa7d8a363,
+	0x223728aa, 0xcc5c61dc, 0x863ee0a7, 0xce3fe413, 0xf325fdd8, 0x17f84e3c,
+	0x7fb3cf31, 0xf972a65f, 0x63754f3c, 0x84fd098b, 0xde7be5cb, 0xc38cef49,
+	0x97b09c50, 0x7b5fc791, 0xc63b25f9, 0xf5897a97, 0x4757c37e, 0xc77ff5f1,
+	0xdf2224e6, 0x9f6978ab, 0xfed9e1c4, 0xd88527e6, 0xa331d86e, 0x3e11d71d,
+	0xbfe2e6ff, 0xad32f264, 0x6c7f18d3, 0x25fa3471, 0x4516c7ed, 0x879f3c23,
+	0xb9e3e22b, 0xe3118c37, 0x78a1e589, 0x7bd205f2, 0xb2788b77, 0x6c5ef411,
+	0x6fa32e28, 0xd72153f7, 0x8adf30ea, 0xe79bb974, 0xddd2ebfd, 0x57f953d7,
+	0x13ff69bd, 0xbcc279c3, 0xc63a019e, 0x9800cfb8, 0x9f95e62c, 0x302b628c,
+	0xec9753cf, 0xe6f8997e, 0xf9050657, 0x7e53da06, 0x2153bbc8, 0x55a1883d,
+	0x020cc3cc, 0x7da563dd, 0x8f944979, 0x2fe04ab9, 0xf452fc1d, 0x5334610f,
+	0x403c8f30, 0x450663e4, 0x7ff62331, 0xe6bdc99f, 0x34ffcc4a, 0x88cb1d19,
+	0x93891de8, 0x85e6bb62, 0x77cf2724, 0x3521f9ab, 0x7e5dddef, 0x0cf7c248,
+	0xf945f4e2, 0x3094e620, 0x81e6ae1f, 0x4a2ecdd8, 0x511f2d4e, 0x7967e743,
+	0x9d37d8b2, 0x8b4a29ff, 0xa5719f90, 0x6bef6131, 0xfa6b81f3, 0xfa744d3f,
+	0x6b80f355, 0x71e9bfd2, 0xd5efdc2b, 0xfc487b8d, 0xa3f214f9, 0x3a77f9fe,
+	0xffb560b7, 0xb3bde902, 0x0e71161d, 0x938f2b45, 0x2f144dfb, 0x58fe0b38,
+	0xdd3e9872, 0xddfbc69e, 0xfa222feb, 0x8d8e086f, 0xc487fac6, 0xa687d5fd,
+	0xde7b464c, 0xfa3a341b, 0x7fb848d2, 0x0e3410d5, 0x1be4fb45, 0x7a694bc4,
+	0xa359f984, 0xf3d23eb8, 0xbfe09c51, 0xe70e0cb8, 0x5fb937a2, 0xf3cc59e5,
+	0x1d75761f, 0x73cf0d70, 0x914acfef, 0x6a77f8d4, 0x6ec8fca4, 0xecbd2ebe,
+	0x2df68976, 0xc8fcd97f, 0x7fb6c68c, 0x8fa23f2b, 0x219ddf3b, 0x790322f2,
+	0xdd23b9dc, 0xa84bf34e, 0xfb885c19, 0x7ef9d49e, 0x901f7083, 0x43e6edf9,
+	0xedf9efaf, 0x352fdff6, 0x837cdd02, 0xcb7d97fd, 0x96fc2ffd, 0xf52fbffb,
+	0xdb7f85db, 0xcbffdcb7, 0x8fff72df, 0xff07d244, 0xf5ebe6af, 0xfe7d78f9,
+	0xb7979f5e, 0x8bd734bc, 0xd695eecf, 0x6ae00476, 0x4d35db8d, 0x8c687902,
+	0xbd4562df, 0xe7923259, 0xb58f56af, 0xa14936fa, 0xaf0abe3c, 0x3fdc5991,
+	0xf39de7b3, 0x6ce356e2, 0x226c7067, 0x3246bbc6, 0xdacf6ff2, 0x198b1e78,
+	0xd798e9ed, 0x37e99a82, 0x67521d81, 0x7d079c8b, 0xf7f0a7af, 0x4b1bd1ba,
+	0xc8af0ab2, 0xfed5a64c, 0x59f648ce, 0xdb2f88ad, 0x17c5191b, 0xbf495199,
+	0xa4abde89, 0xe3d3fda3, 0x2e3f7e45, 0x689b9cc0, 0x4dce4a5c, 0x928f6e74,
+	0x407be383, 0xae1367e4, 0xf18397ef, 0x03b004e4, 0x1eb78f31, 0xa7e479f3,
+	0xfa9ea9da, 0xdc99ddbe, 0xc4c07a4f, 0xef305ce8, 0x11dd7c95, 0xbd9156b3,
+	0x0ee1e6a6, 0xcb27029a, 0x8dc3ca25, 0x62bff1c1, 0xce6de408, 0x449c4411,
+	0x3f296dfb, 0xdfc8eaf8, 0x7bd377f5, 0x86fecd14, 0x6187f466, 0x29ee771c,
+	0xb19dfa20, 0x6fed2477, 0x091ae4f0, 0x46fc5dce, 0x43780711, 0x38d431c4,
+	0x7a8f9a80, 0xde3fc45b, 0x3f4ab8c1, 0xa1584f3c, 0xc34caceb, 0x5a9e77bf,
+	0x159cfe57, 0x203d3f49, 0xf78fb8f1, 0x29f71e15, 0x84fa3b30, 0xccac7faf,
+	0x0f2c78e4, 0x1a83bcf0, 0xcc4279c7, 0x7287e0a7, 0x687602c0, 0x6681aac1,
+	0xab05ebd4, 0x3af0a2b2, 0xa433e9c8, 0xcfa7bc00, 0xf15069d9, 0xee8e8917,
+	0x89f114ca, 0x8f941d03, 0x9797467b, 0xd7c99ddf, 0x7165e41b, 0x64ceeefc,
+	0xafb3be54, 0x8b9a3971, 0xcbb1ae3d, 0x6541f291, 0x38e2f7e0, 0xb8f8ebcb,
+	0x05977855, 0x4f7c2294, 0x558ebedb, 0xc8afcd78, 0x79b8a229, 0xb455b9e5,
+	0x85d2e9bf, 0x1c52d7fd, 0xffa0accc, 0x1b55b330, 0xc5909b4f, 0xa0cfe7e3,
+	0xbb708dc1, 0x6ffe48ce, 0xa24675c1, 0x1b32849f, 0x2609a7f9, 0x7960c726,
+	0xf283b712, 0xb08a938b, 0x93c4a6cc, 0x407f97fd, 0x179b9f89, 0x5c256f8a,
+	0xe11938b7, 0xa8bcb974, 0x22bb271a, 0xa44bf9d9, 0x179aafdf, 0x88783aad,
+	0xf3aa7bd6, 0xe6a2e152, 0x8bcbbb37, 0x3c12ebaa, 0xd6689fc5, 0x1d8c7851,
+	0x58ae31e3, 0xaccda7b0, 0xcb4b37ee, 0x0b4de509, 0x7ca39f76, 0xf51f9aa9,
+	0x3b239cd4, 0xb199da1a, 0x3fa0017d, 0x4364e41d, 0xee9bcf2d, 0x1a7bfb12,
+	0xae78fe61, 0xfe601ff6, 0xfcc3f578, 0xb50bfef1, 0xa81ae9fd, 0x6f75d075,
+	0x7aba08ed, 0xcf28e7fc, 0x36a642fa, 0x8f99d75f, 0x084d5d93, 0x1d7cb0af,
+	0x3af9f595, 0x63c78a39, 0xee303f70, 0xb41cf21f, 0xf42efc92, 0x7f42f797,
+	0xee2d6fe0, 0x6aee385e, 0xd18597cc, 0x61d3bb45, 0xdb6c73c3, 0xe61b0e5d,
+	0x4e9a0ac9, 0x6a847ed1, 0x15ce9063, 0xa6073ce8, 0x775f1ce1, 0x21b42feb,
+	0xd79b2f7c, 0x79fffc6a, 0xd7932fad, 0x32f905ba, 0xf23f2439, 0x26e7065d,
+	0xda701ebc, 0x87c986e6, 0x179e2ed5, 0x9bdb9d59, 0x5741eff8, 0x6139f8ef,
+	0xe628adfd, 0xeb848a71, 0xf7e16d72, 0x79e2c89f, 0xb98904ac, 0xe4b1c922,
+	0xbcb8df9e, 0xef648e4e, 0xf402bfb7, 0xc3d03578, 0x5abdf9ee, 0x71b4ae89,
+	0x73c5d5b6, 0xdbb9e17a, 0xac2c37de, 0x69c38f2f, 0x974c6ce2, 0xde2eeb4f,
+	0x40d8a336, 0x7bc2e2e7, 0xeb5f51b3, 0xd8cc62e3, 0xb9f5cd91, 0xe4739c53,
+	0x582364cd, 0x5bdabc97, 0x173877cc, 0x362f4fea, 0xf3509e90, 0x46c5f96b,
+	0x99f40b56, 0xa3e4d4bf, 0xb1b15abc, 0xff0b9e8d, 0x3f3d8570, 0x3bc9e513,
+	0x8c76cd07, 0x60e3e9a3, 0x752de28e, 0xa3bda1bd, 0x6d7da252, 0x373fd65d,
+	0xf8c38d36, 0xaeb660d9, 0x1bd42c6e, 0xe3b06b9c, 0xe5cbe57a, 0x3d85f2e1,
+	0x219f8fa3, 0xa1fda15d, 0x2f5fa38f, 0xb0bd7ef0, 0xbf43ee97, 0x8fd43932,
+	0x2f4f4eda, 0xbe94f68f, 0xbf7f2e30, 0x71456fe8, 0xc45ceafe, 0xa1818967,
+	0x3c5158de, 0xd61b9ac6, 0x234f595f, 0x7986af5f, 0xbcf0a4bd, 0xe9dfcf1e,
+	0x3caa79f3, 0x8cfa682b, 0x083efff6, 0x95f51972, 0xffc78d27, 0x3c285e0b,
+	0x34227280, 0x1f9fae4e, 0x37bf534c, 0x7c6788e0, 0xd14fb45e, 0xb8a247bb,
+	0x27c97e5b, 0xcca1a37b, 0x7fb4f15c, 0x773073de, 0xef939e39, 0xc50e3129,
+	0x971e3afd, 0xc22799d4, 0xd7e4d07c, 0xf5c7ae2a, 0x075305fc, 0xa67e20b7,
+	0x9d689a96, 0xa6d7e4de, 0xaae539d1, 0xba982b9e, 0x7c289ee9, 0xf3a25fde,
+	0x56bf261a, 0x4f9f063c, 0x88c01ed8, 0xb3889b38, 0x3c193907, 0x7c88d278,
+	0xfbb24572, 0x0537c827, 0xa89e4493, 0xa9e3eb94, 0xc62649fe, 0x7e8d46ed,
+	0x1bface6c, 0xf3d479c5, 0x7d711204, 0x2f485e42, 0x7d01ec15, 0x57d21b15,
+	0x7a9eaeff, 0xe09da7b4, 0x2be83579, 0x24c24f1c, 0x62ef4f9e, 0x77fc08bf,
+	0x93ba7fa7, 0xc467a8bb, 0x529983f9, 0xad7d575a, 0x38673c60, 0x9d12e5da,
+	0xfd1dd683, 0xcfec26bd, 0xb6ec6ef8, 0x76fd04d7, 0xf1a9aebb, 0xfd13c9bb,
+	0x787fe7b9, 0x57f8489e, 0xf548be6a, 0x7b09fac7, 0x215ebd57, 0x7b7fbbfc,
+	0xd12a013f, 0x1209fb88, 0x9827ef22, 0xbda12f59, 0xd827eea4, 0x3b856667,
+	0xab3d7093, 0x73e44fd8, 0xefb41d91, 0x7aad6fd7, 0x7bce893c, 0xa0cc6dc0,
+	0x9f485af7, 0x43fd8157, 0x63ff92f6, 0xe7ad3f1e, 0x1aabb1eb, 0xd496fbb5,
+	0x5f616ceb, 0x50f9858f, 0x77ac459e, 0x91ce2f1d, 0xef21c508, 0x21c78632,
+	0x59304a3d, 0xe3eb475d, 0x5c4d1779, 0x6438bbb2, 0x75a0eb35, 0x9215d5b3,
+	0xb7c3f503, 0x91ec971d, 0xddaf5da5, 0x83cb9327, 0x42496e69, 0x706ad5fb,
+	0xf8bad255, 0x6c7ef817, 0x66fa7c3d, 0x413c7971, 0x331e31e4, 0x6c787a73,
+	0x5bfb4494, 0xe21c1f70, 0xf817b1f9, 0xf2c7973e, 0x35173ef9, 0x5feea16f,
+	0xb79432b9, 0xea6ffda0, 0x1f7517be, 0x780fba8b, 0x00a886d2, 0x99f707da,
+	0x1db8d3ea, 0x3c3a25eb, 0xf3d193dc, 0x6f3f0b28, 0xf8ce3f5d, 0x28c3c2ac,
+	0xf37f0f87, 0xac88d176, 0x25a67108, 0xaffc2fc2, 0x98fb2eac, 0x30bffb54,
+	0xfec6a86b, 0xfedeb2a7, 0x16f2a6ff, 0x6f9f0141, 0x32abde70, 0xe6f8ddbf,
+	0xfc72df67, 0x7b247cfa, 0x4b88b976, 0xd9e7e9e9, 0x68bf552c, 0xbac34bbf,
+	0xfae43fec, 0x8e524ebb, 0x2f6bb9d2, 0x542573f1, 0xcb175a39, 0xe4215ffb,
+	0x547d7aa7, 0x914ceb5c, 0x884f5fe3, 0xe764b3e7, 0x0193a472, 0xc8f1b4db,
+	0x78daf581, 0x098dd5e4, 0x40e0f29a, 0xc17ea686, 0xe79ade81, 0x69578343,
+	0x237f0f9e, 0xd91e535f, 0x7f534a3a, 0xb46387f4, 0x68755ae7, 0x5ed791e3,
+	0x2fb749bc, 0x2bf047d2, 0x6f0dca04, 0xad9d6457, 0xa15997a9, 0x8d56579d,
+	0x96bd5e76, 0x2fcee826, 0x78e6fde1, 0x53ebf3b5, 0x025f9da7, 0xbef2d3e6,
+	0xd4cd3e7e, 0xecf7a153, 0x61b1f7f5, 0x738a07bb, 0xa99f686d, 0xc1db99fb,
+	0x6411786e, 0xffb69fdf, 0x9cd7d696, 0x06eda279, 0x79e6cf3a, 0x0379e199,
+	0x5963dbed, 0xa3e3de80, 0x34162873, 0x661ef4c8, 0x1c9b0c0c, 0x9d1afe76,
+	0xe9f28f4c, 0x911258a9, 0xf8b69e0b, 0x82fda04a, 0xe106275d, 0x7bf692fe,
+	0xfbcb3e78, 0x31f14484, 0x841d6de6, 0x60584bdf, 0xa89dfc8c, 0xbcb8b2d7,
+	0x9a3bbd3b, 0xdb49ebd6, 0xbf52669a, 0x3098b7f5, 0xdbd6a69f, 0xa849fc2f,
+	0x3d3df7c7, 0x5e77d12e, 0x71e17c70, 0x27bbf54d, 0xfae71fa7, 0x89a3416f,
+	0xb059efce, 0x9e3b676d, 0x0de0bd57, 0xe65b9d0b, 0xf48e76d9, 0x3b53d16a,
+	0x469ece08, 0xd5a8bd99, 0x8273da57, 0x5db8f1b7, 0xe9154375, 0xff932841,
+	0xf91f5d55, 0x9dcbecbc, 0xa9f0bb67, 0x00bcbb3d, 0xfccf85c2, 0x1ee30e2e,
+	0xc6b7051f, 0xedbedf91, 0xeee9cf8b, 0xfd702e72, 0x173c2fea, 0x03fc23f8,
+	0x2cc5cbe4, 0xdbcab9dd, 0xa9f7c512, 0xf7c2c302, 0xe8e087f9, 0xf1de78e5,
+	0x6e7823e9, 0x3fa7e9fd, 0x5bc7047e, 0x8f0baff9, 0xadf498fc, 0xc3728ecd,
+	0x3a61c4f3, 0x976d5bf7, 0x40e5cd90, 0xa3bfe3fb, 0xdfdbd2f3, 0xb7b038b4,
+	0x641f3df6, 0xefb77d23, 0xe3052cfc, 0x8cf78b5c, 0xe2f078a3, 0xf0be8b67,
+	0xb7486b7b, 0xe3e17ebe, 0x3e7375f2, 0xcd5f16b1, 0x38d9cd0d, 0x6f8033ce,
+	0x9e75f38c, 0xc55f5cf3, 0xc456cdcf, 0x23dbcef9, 0xe5df8b9f, 0xc986e73c,
+	0xed0f7e37, 0x2e7e069e, 0x75cf65cc, 0xbc78043e, 0x6fe04bfa, 0xbe716afb,
+	0x9dfc6449, 0x4f003fc1, 0xb9ed6c8e, 0xcc369fce, 0xf8d7f47d, 0xdbb121ac,
+	0x5f039d73, 0xb9ea6e02, 0x72ccfffb, 0x3e46ce8e, 0x74f7fa7c, 0xade2368e,
+	0x3a73c144, 0x7f46cd3e, 0x075343ee, 0xb1d72b9d, 0xefc762c7, 0x73d82d37,
+	0xfda7f894, 0x64b71e59, 0x76fbfce2, 0xe7e3f9d6, 0x8a988b95, 0x3f2c4b67,
+	0x1fc05a0f, 0xe2568bcf, 0x1738bee8, 0x7ab1d39a, 0xe777745e, 0xeeca2f44,
+	0xf89cf5cd, 0x96d5401e, 0x9fb99abf, 0x3497c21a, 0x7fe9f682, 0x2adbf9e1,
+	0xbfc2ec1c, 0x96dbc4eb, 0x7771e7c8, 0xf02f8f9e, 0x73c3f885, 0x35cf3a3c,
+	0x5435f287, 0x08b28e96, 0x137da0f2, 0xbdbd2f3f, 0xeea9fd1b, 0xd6caaa0f,
+	0x6798a9f1, 0xf2a1fc77, 0xffc6e5e9, 0x3333e155, 0x8bd0d15b, 0x97d3a70a,
+	0x75f1cb22, 0xb6bfef82, 0x62ece3fd, 0xf48fdcff, 0x66d0d558, 0x44d9fbe6,
+	0xf7cc547e, 0xf4dfd834, 0x6427a4fd, 0xd557e3b2, 0xd71c9337, 0x6eea8bcb,
+	0xb39b6df2, 0x7153bac5, 0xa7e9c14e, 0xefb87d63, 0xdddef4ea, 0xf6effd7c,
+	0xfa17b336, 0x19edbfdb, 0x721b8f33, 0xe13d41fe, 0xe9ff1b0b, 0x91069b27,
+	0xdd66eedc, 0x8f78bd77, 0x3df91fc7, 0x34e7ed9e, 0xade859ef, 0x72efff27,
+	0xe8c8fbef, 0xfbfb05bc, 0x3bf7c828, 0x4ffb847e, 0x17ffbc23, 0x176126d9,
+	0x22eb13ae, 0x3f03de41, 0x9ee02bbf, 0x3cf5cbd6, 0x9d1999c2, 0xe3be1083,
+	0x2725917e, 0xaf979f01, 0x7fa8ec95, 0x4c18af67, 0x9ccf9274, 0x89652744,
+	0xfa937bfc, 0xe6cf1cf7, 0x1ab6fc7a, 0xc3fa4313, 0x280cded3, 0xb624eb7f,
+	0x3197fb86, 0xe7425313, 0x6ead56d1, 0x0abc2389, 0xef3d533f, 0x3e66a2b6,
+	0x39fd07c9, 0xdeac9538, 0x2d6fe78c, 0x2dcfcb58, 0xa38b9fd4, 0xcf5dab97,
+	0x85fb5a89, 0x956df672, 0x81cc1cef, 0x07e7377c, 0xd7e456b1, 0x4a68f0ae,
+	0xb3ca05be, 0x07a4c80f, 0x2eb0b854, 0xebcbfddf, 0x7f3cf881, 0x973f2fa0,
+	0xb152515c, 0x3c1e9cde, 0xdd19282e, 0x597ef1c3, 0x19b87ba2, 0xb19beae5,
+	0xb596c7b8, 0x7c50345d, 0x36595980, 0x59d9938e, 0x537db876, 0x60a7b26e,
+	0x16cbc4de, 0x58fc51a5, 0xeb0addff, 0x1eff20da, 0x157bf683, 0x5f9c8965,
+	0xeff3763a, 0x1e46b1ef, 0xeb32b6bf, 0x9c778c32, 0xe0a5d652, 0x95ae5411,
+	0xc53dfe26, 0x4714f1ed, 0x91456db8, 0xba5fcf19, 0xf69e21ad, 0x93af30b5,
+	0xe2a8cfbe, 0xc1456db1, 0xf7e834fe, 0x871cfab6, 0x269fc5ac, 0xaea0bc1e,
+	0xc9dbc54d, 0x64ede0cb, 0xbe01bc24, 0x7a13f734, 0xeec5f169, 0xe94cf6e6,
+	0x62fafb1d, 0x0d7ceefe, 0x77631eff, 0x340e92f2, 0x35c5094a, 0xe33a338b,
+	0xef4829b5, 0x250ed617, 0x55f1e9bc, 0xc3d9affc, 0x7b337f71, 0x05ce962b,
+	0x66c3eb2e, 0x6ff184dd, 0xd953f389, 0xa316262f, 0x8bd01ade, 0xb56bfdf3,
+	0x82d9fe3c, 0x787ecff1, 0xca9dba6e, 0xf3fc4587, 0xe13fc628, 0x3fc626fa,
+	0x7c527ae1, 0x33e5433f, 0xd7a20020, 0x7f38bdc1, 0x8925ee0d, 0x34e2ddf8,
+	0xff8a4bb4, 0x73f38b25, 0x61637d7e, 0xce878d0c, 0xa3fef9cb, 0x7bd1bebf,
+	0x3a37630d, 0x69b4f1b5, 0xb4f1b453, 0x663915f2, 0x736fd7f4, 0xa3d4d68d,
+	0x0f5abc88, 0x925e98ba, 0x337e79a8, 0xbfcc34be, 0xf0a5ee2f, 0x1f7147f8,
+	0x4b5bbd4d, 0x2a5bdfb5, 0xb3c668bf, 0x62dbd985, 0x26b3309f, 0x87bfc219,
+	0xf1ca4402, 0x3e6de762, 0x416e3a31, 0x60e4fc7a, 0x5e7284b2, 0xaf150f8f,
+	0x080037b3, 0x1b39dfa0, 0xbd97940c, 0x55e3b66b, 0xaeeccf3c, 0x7bf7f8a7,
+	0x467bae4e, 0x39805df3, 0xbf84994f, 0xa1fb2763, 0xf232c568, 0x607dca3b,
+	0xcb88022f, 0xa23ef45f, 0x467e3196, 0x0f91594b, 0x80bec90e, 0xe2727844,
+	0x4fbd83f7, 0x7e47eff0, 0x0ed86208, 0x2b7677fa, 0x1287c71f, 0x53e5815b,
+	0xfdfbb7b6, 0x6675e600, 0x0679bae2, 0x307c8dc4, 0xd3e2e499, 0x2cfa4332,
+	0xe744c3bd, 0x587ede39, 0x31f4efa6, 0xc5533f29, 0xd9cbe62b, 0x1dfe1613,
+	0x04a00e9b, 0x71f8b7a8, 0x8f9f7d0b, 0xcfcb0328, 0x11adf40e, 0xf2f48874,
+	0xc6733c61, 0xfe7960d3, 0x6496a3e0, 0x82119f5a, 0xce106f9b, 0xfae09bd7,
+	0xa3a02f51, 0x06f9e68f, 0x3bbafc91, 0x0f47f899, 0x2e6693df, 0xef48bc9f,
+	0xea2e4852, 0x26d996ed, 0xc7df7d07, 0xa3a486cb, 0x96b93a5c, 0xf40f477f,
+	0x3f5f1c4a, 0xabd134f3, 0x46939efa, 0xef8e3c65, 0xe90e5d64, 0x0ad6dbaf,
+	0x06f4137f, 0xfae3efaf, 0x46ecbd6a, 0x275bc3bb, 0xaf9db9b9, 0x4db78fa5,
+	0x688b9ede, 0xd68db1fd, 0x6809dcaf, 0x9dca3cc5, 0x2ec9533b, 0x5bec2998,
+	0x2ddbc696, 0x3b3a73b6, 0x0f4fa07f, 0xea7ad7dc, 0xea01e010, 0x75f3a7b4,
+	0xacc6a705, 0x3747861b, 0x1dec369d, 0xffb803f6, 0xfbef44f1, 0xc8dbc046,
+	0x5ee6f76c, 0xac329f18, 0x4473c6cd, 0xfc4fface, 0xf50fbf1c, 0x83930ed1,
+	0x97e974d8, 0x0c4ed3b3, 0xeb835d8d, 0xf2eb67c2, 0x3f042704, 0x5bf8293c,
+	0x8b6fcfd5, 0xa7036cf2, 0x2482f63f, 0x34c46bbe, 0x9e217872, 0xe4872a43,
+	0x2b872aa5, 0xc7fadd85, 0x3e6edc5e, 0x60a5d6c5, 0x4fa005bd, 0x3b3cfef1,
+	0xb72a4bb7, 0x7ee7eff1, 0x4c9c7c84, 0xf5c31a7d, 0xe4167cf2, 0xea014e7a,
+	0x77dd0949, 0xc0388b45, 0x409fbe3a, 0xb9e7f72f, 0xe06ffb7d, 0xa3003bfd,
+	0x7cffe45d, 0x8f1b1ae6, 0x3c6deb71, 0xff97ac9e, 0xb9b7e84c, 0x3d7a7de2,
+	0xbbe85730, 0x377d3ab9, 0xce8e71cf, 0x0edca91d, 0xb2b945ef, 0xc2efab4e,
+	0x9a44a9e6, 0x872f3ce7, 0x873f3cf7, 0x769d7bcb, 0x08fda30d, 0x3d59dc2a,
+	0x77b4d3fc, 0x791c61c5, 0xc33283fe, 0x78aa1cef, 0x823e66a8, 0xb02cc7f7,
+	0x8f18ed0a, 0x062e31bc, 0xa1f0ba70, 0x2dc7ecbf, 0xa06b944a, 0x7efef3f3,
+	0x9427b3bf, 0xa4b2f37b, 0x6f7efdfc, 0xd570aecf, 0x829bf1c3, 0x6ef9131b,
+	0x9863076a, 0x377f85bb, 0xc91eac3a, 0x0901fffb, 0x00d0a8f5, 0x0000d0a8,
+	0x00088b1f, 0x00000000, 0x3bedff00, 0xe5557469, 0x73dcfbb5, 0xe1dc8487,
+	0xc2040464, 0xc06e49b9, 0x612f4865, 0x1213d41e, 0xde0d4b22, 0x00c10580,
+	0x86120137, 0x7c07504c, 0x0040e3e2, 0xaa1a4583, 0xf50af8a5, 0x22d28342,
+	0xb04a0834, 0xc141170c, 0xd6de8ba7, 0xa44f7d6a, 0x40932861, 0xd2bb6a52,
+	0xf7b79457, 0x3b939df7, 0xed83a404, 0x58b593af, 0x9ef37efb, 0x77f7bdbf,
+	0x52015000, 0xaaa8e601, 0xd08eceb5, 0x092bd00c, 0x72c06e60, 0xd80e35b6,
+	0x37fc0ddf, 0xbb746b7f, 0x01e5e696, 0xf7473754, 0xa41c5fe3, 0xbfe6cc01,
+	0xff5616a1, 0xd845cc41, 0x766f3d12, 0xe890eadc, 0x556679a4, 0x086e17eb,
+	0xa1e6ca0c, 0x840cfada, 0x9f12a24d, 0x6e0f351b, 0xca7a01b8, 0x7773948e,
+	0xde0bc361, 0xc5c2221b, 0x002300c9, 0x5c78174a, 0x439c1cfe, 0x2186f77f,
+	0xb09e7468, 0x39cd23c0, 0x098a6584, 0x68c79cf0, 0xb1c6994f, 0x477f788d,
+	0x617c67e2, 0xc1864ef6, 0x15483f08, 0x3a6f3c54, 0xeba236e1, 0x71eb15be,
+	0xdf8841e7, 0x11f8137a, 0xcbbf5b9e, 0xfe3e47e9, 0x695dc5fe, 0x88772b04,
+	0x66084410, 0xdcf015bf, 0x2f65f582, 0x56fb833f, 0xdaf78d6c, 0x88e2bbb0,
+	0x3c7008bf, 0x252b7c0f, 0xe3d7971c, 0xe070c341, 0x71dd8445, 0x8428f23d,
+	0xdfd982fd, 0x513de68f, 0xd927efe1, 0xd187faa3, 0xf4cc4fbe, 0xee3f7cff,
+	0x04b6c4fb, 0x9248caaf, 0x44e1a682, 0x84041489, 0x7b7f0bd5, 0x04fc8cd2,
+	0x9e8f83f1, 0xf2b8014a, 0xb6c408ba, 0x817f24ec, 0xcfdc4a9c, 0x703d05eb,
+	0x5d29b89e, 0x22f6c4ca, 0xeff9ecec, 0x8d4ce7e7, 0x4fc4e59f, 0xd599e914,
+	0xf3a50e1f, 0xbe70374f, 0x4d99face, 0xf397a7e3, 0xe7e73573, 0x339f8d4c,
+	0x7e35788f, 0x4f892bca, 0x9f8d6af2, 0xeace1c0d, 0x3a7537c2, 0x4fc4f5e1,
+	0xa7155e6c, 0xa60f885c, 0x173f909a, 0x4f508a76, 0x12a80b6d, 0x4485c3da,
+	0xdf511250, 0x3f8d04ff, 0x7e610a0e, 0xe50ecdce, 0xd29e9106, 0xc49d47b5,
+	0x453bb9d7, 0x3f2c704e, 0x73eee941, 0xc81f9e44, 0x24d53c25, 0x087942df,
+	0x408aa5c0, 0xb7fbe12e, 0x825a7df2, 0x79e3e522, 0xde2dfbed, 0x36efc8cd,
+	0x85c01587, 0x8384f676, 0x227265c1, 0x3a81679b, 0x505f3c00, 0xf8845814,
+	0xdd3e50db, 0xb2159e90, 0x5905c6cc, 0x87ec5282, 0x7e13200d, 0x24288282,
+	0x1d2fed75, 0x3e462283, 0x814c3bb5, 0x67ff48cd, 0x3c80d0d6, 0xffaf187b,
+	0xe913b192, 0x03c835f8, 0xd252bce8, 0xccafc235, 0xfc8cd815, 0x6c4c0879,
+	0x0cc087df, 0x982708a5, 0xd8713541, 0xba224f60, 0x3fa85990, 0xf1bf35e3,
+	0x7410e1ed, 0xef82af21, 0x3fae7081, 0x73a55d30, 0xc14d0726, 0x1e0fe87e,
+	0x090f0732, 0xb2714d08, 0x7c73c245, 0x913c84e3, 0x660cb0e4, 0xf15decf8,
+	0x21f846db, 0x1326a5bd, 0x8129ebc7, 0x0a28f1d1, 0xdd32f89d, 0x8444470c,
+	0x4439e28f, 0x5904b4f7, 0x0816f365, 0xa30ccdde, 0x54df8614, 0x2881bf0d,
+	0x4bc87387, 0x0df76ddc, 0x3547d3d4, 0x069f707c, 0xf1cf5b27, 0x1edf9588,
+	0xad004f8d, 0xcd6be256, 0x227fef63, 0x71359d20, 0x49f920f6, 0x6e5cb168,
+	0x39f9afed, 0x48afe1e4, 0xc05f384e, 0xf4f90083, 0x04db06a8, 0xb87b3bb4,
+	0x1e0d21ff, 0xa5ace8d7, 0xde0bca36, 0xdc5b0212, 0x03e73fb7, 0xd3ad8939,
+	0x3b0347f7, 0x1026e34d, 0xed6bfc80, 0xb2562dfe, 0xbf5a79bf, 0xb22187ee,
+	0x5dacde2d, 0x169f912a, 0x41ba6fc0, 0x8b3edad0, 0x13dc2958, 0x7afb25ef,
+	0xa1ecdad4, 0x5e76277b, 0x5874df2c, 0xef53805c, 0x3e75f9a0, 0x03c03465,
+	0x0bdbd3df, 0xcbda0a7c, 0xe39de1a4, 0x8e164a8d, 0x8e7f177f, 0x7828e864,
+	0x452fc133, 0xd7253ede, 0xfce39685, 0x8a728f0a, 0xdcd39ffc, 0xa0fecc0a,
+	0x58bbf191, 0x28061324, 0x28a37d1c, 0xb858f911, 0x6c9e605b, 0xe636e8ea,
+	0x50f13a77, 0xd478619f, 0x76f413ec, 0x71bbc148, 0x36ceeeea, 0xb070a3c3,
+	0xf27c6a2b, 0x9f1a8868, 0x1be91d78, 0x6a71b5a9, 0xd194f488, 0x4a8d9ff1,
+	0x235b6c6f, 0x42dae3e6, 0x4ba74f3a, 0x65ffef0e, 0x835f19e0, 0x9d75d199,
+	0xcc5075f9, 0xba6101be, 0x91df665c, 0x5dff223f, 0xe2a48229, 0xfc5f1aba,
+	0x6117f26a, 0x3d181a7a, 0x8f9d78a1, 0xf1a37cd1, 0x06bfa442, 0x8d55f1f4,
+	0x67cdf20a, 0xd2af9cea, 0x3f3ab1fc, 0x337f7d70, 0x7d6c4973, 0x4686a6d2,
+	0x9fa61ed5, 0xbe18b3b5, 0x7c21ee49, 0x295fd339, 0xc42f48ff, 0x4ef906b3,
+	0x5ca74557, 0x5f8c75d5, 0x6d1131a1, 0xa62337c2, 0xff022ddd, 0x52fc11de,
+	0x59f949d7, 0x7cd2f811, 0x8b4a67c4, 0x127fbd10, 0xb4ddcf1f, 0x1d1d51ae,
+	0x90fc036e, 0x5e8227f1, 0x9d299fc6, 0x9be75ef2, 0x6ac7eb85, 0xfee019fc,
+	0x45e88fa1, 0x7d3972fa, 0xd2fd2881, 0x65bcaef5, 0x14b48aed, 0x2004b38a,
+	0x68f2c37a, 0xe7fc91f2, 0x9bff3a4b, 0x586f8442, 0xa13dbc1e, 0xa9bf3f59,
+	0x64abf491, 0x6f2f4eb7, 0xf70bf587, 0xbff216fb, 0xea7c7a21, 0x53f5e8f2,
+	0x667f73a0, 0xf53a5357, 0xfa43cf17, 0xbcfc69e6, 0x4853537f, 0xbf76fae5,
+	0x67ac353b, 0xad12b365, 0xed85bf74, 0x6be31b71, 0x93b8de27, 0xbc0fe380,
+	0x07c7cd1f, 0xb27b79ea, 0xc1af9b7b, 0x4d7ed91e, 0x7764e3e7, 0x7d72bd94,
+	0xc4fd1df9, 0x4024f37b, 0x161fbd94, 0x4487dba3, 0x437baf9d, 0xa750e58f,
+	0x1f5efe7e, 0xfc0fb7dd, 0x719f1a30, 0xc7bdbd36, 0x6fe91857, 0x5f1937d4,
+	0xf9f4fc74, 0xd7a25f45, 0xddaaf3c3, 0x729afa6f, 0x320cdee8, 0xc49a56ff,
+	0x71c75c21, 0xb56a47bc, 0xbf68a938, 0xbcd00be8, 0xc2cf416e, 0x873d309c,
+	0xff99169e, 0x8085eb56, 0x196cfd07, 0x04ba56c8, 0xa5628267, 0xcb90f5fb,
+	0x9c15e8b5, 0x7169f87f, 0x3df70051, 0xc59928b9, 0xadb81f9b, 0x5d87ff8c,
+	0x6366d37d, 0x0cc250fb, 0x0ef38cab, 0xc1963b3d, 0x7fe93bb7, 0x337a47d7,
+	0xc9a5ce2b, 0x97e65df9, 0xf6f0250e, 0xbedbf72a, 0xd5af28a5, 0xe7ed996e,
+	0x5ad2924f, 0x39e13f50, 0x09916c0b, 0x098fef4f, 0x6de7b2bf, 0xfc239f03,
+	0x4695de96, 0x41ff2ef2, 0x9c4f5149, 0xe2c2be57, 0xab5e7015, 0x7c274a49,
+	0x658dcffd, 0xb96b30fa, 0x18fbd506, 0x4fd5f83c, 0xcf749dea, 0x1c58146e,
+	0xcd17f773, 0xbb62e7ef, 0x4bd321b2, 0x30e81ea8, 0x9ce78481, 0x202fdf5e,
+	0x899dcaa2, 0x3c777baf, 0xbf340f9f, 0x7c7cbaf2, 0x2f9a60fc, 0xfe70cb4a,
+	0x7017ec39, 0x712adcfc, 0x397840b3, 0x2404dd1d, 0xc097dce9, 0x2b73f0ae,
+	0xb865816c, 0x7fbb69cf, 0x673a7afa, 0xd29605bd, 0x93d7413a, 0x4e5f198f,
+	0x3f5cba4b, 0xa1cfcc6e, 0x2dbd3679, 0xca9f2195, 0x157a9d4c, 0x2aefbc04,
+	0xb5fb7912, 0xb6f91f4d, 0x2f73fae0, 0xaaf163f5, 0x276ca19c, 0xa19d3d7d,
+	0x3ffb20ec, 0x462ed5f8, 0x3c4bfb5f, 0x51d03b6e, 0x111e569e, 0x315c5427,
+	0x68387d33, 0x7c231f95, 0x23eb760f, 0x6a2f2ca8, 0xf8a12e4e, 0xe28cf511,
+	0x45ebd46d, 0x1f126ebb, 0x2645ba36, 0x4ddf43af, 0xea472d1d, 0xf9867ab7,
+	0xde64e289, 0x6d6eea3e, 0x9cafd154, 0x33822db5, 0xfaf5e159, 0xe8aef9fe,
+	0x39983938, 0x81d128e1, 0xca30c679, 0xeb7cf453, 0x0733f533, 0xf533f1c9,
+	0xe3630653, 0x43a74ad4, 0xde7fe725, 0x1c9330e1, 0x9c979a4e, 0x62b938a3,
+	0xb9f985a3, 0x8ba98d8a, 0x414e29db, 0x5740f17d, 0x2c32927a, 0xda5b9e8c,
+	0x2beedaa5, 0x07191dec, 0x5dad7fb4, 0xf36a6eb6, 0x6d6fd935, 0x5b129597,
+	0x2d1b4503, 0xcaabb23e, 0xa0330e21, 0xf1b31cbf, 0xefa42d9d, 0xf3e20b95,
+	0xe578886c, 0x53931b46, 0x4aa1c3ab, 0x3a77ee38, 0xaf0889cc, 0x6e8c4dba,
+	0x60cc8657, 0xb8e60881, 0x5397063c, 0xf10d711e, 0xc5f6c649, 0xa3ac2f45,
+	0x0412e4b4, 0x3d430d26, 0xfd70a6bd, 0x0c5bb6bc, 0x66047e50, 0xf98c5cc5,
+	0x7ee01cd3, 0x5dbdfaa2, 0x39fc7cd6, 0x41f2aea7, 0xf9f965c8, 0xfcfc98e9,
+	0x8e1e7474, 0xfe51a61b, 0x30aaffcd, 0xc43b24de, 0x5e6a3d7c, 0xd51e7e44,
+	0xf2d0c96f, 0xa3c83ceb, 0x9d5aee38, 0xdd829b25, 0x13da3b4a, 0x41b059d2,
+	0x2e56f860, 0xc4360312, 0xe79920b7, 0xa0976747, 0x5e8caddc, 0x2576c255,
+	0x37c4b1e5, 0xf8710210, 0x0b0dbef3, 0x1767e4cc, 0x4ca57e6b, 0xdbe91f00,
+	0xfa6f7899, 0x36c78a02, 0x56be33c6, 0xc7ee78fe, 0xd6bae12f, 0xa229c5a5,
+	0x5e87d61c, 0xfee03e9a, 0xde149710, 0x67d2fce7, 0x0bdf49d2, 0x4857ee5e,
+	0x22657871, 0x26e298ec, 0x974cf539, 0x9579e1ed, 0xe6ce9ecb, 0x8083e2f3,
+	0x2ad970f1, 0x6f01d191, 0x521e02d8, 0x4520fec8, 0xf2d66df2, 0x1cd6acf7,
+	0xd34a5bfe, 0xfc991ee2, 0x9c5faf37, 0xced171c6, 0xc676a7f8, 0x066ff8bd,
+	0xf993d5ed, 0x57d13f09, 0x4c66126a, 0x5bb4fb60, 0x328b8ec9, 0xb7701e78,
+	0xbdd2cb06, 0xea8f09d3, 0x32ef3635, 0x1c47df37, 0xb181e1aa, 0x5429b54d,
+	0xf3c223f1, 0x0b779b4d, 0xbc796dc9, 0xb924ef12, 0x0e7af282, 0xad98c7ab,
+	0xa72fb449, 0x76e64ab3, 0x989ffd8c, 0xc1dabfb1, 0x227560fd, 0x15cedbf2,
+	0xb7fb84c4, 0xbe4cafd1, 0xf0dccf5f, 0x85f3cefa, 0xb99df8f0, 0x49e4afc4,
+	0xdb1e108f, 0xc8af6645, 0x29c3358c, 0x0e2ed96e, 0xf1a64e7a, 0x57e445c3,
+	0xc81cefe7, 0x1db2b6e2, 0x617bb21f, 0x56e726be, 0xc7296f2d, 0x33844ef6,
+	0xf8bdb832, 0x666ba845, 0xe8bf9e6d, 0xabc7d6f2, 0x9c985957, 0x937fd1aa,
+	0xfa5ea8e3, 0xb56fb65b, 0x79469423, 0x961bf556, 0xcb6fe4a1, 0x97cbfc35,
+	0x09fd19f6, 0xc5b94fea, 0x543dd125, 0xa95b16a5, 0xb028d55d, 0x6a5d7876,
+	0xe7e3e93a, 0x4938f7cb, 0x07c4ce4f, 0xb4df743d, 0x52f7882f, 0xc0da887e,
+	0x97cf93ee, 0x873f367b, 0xc85259ed, 0xacf8e021, 0xc89332c7, 0x82949f2f,
+	0xd93f1a56, 0x5b5136ec, 0xd4d1a491, 0x62dfcc56, 0x55d1cfed, 0x43fa6b35,
+	0xe046a5fe, 0xed7aacf2, 0xcf0335b0, 0xffc9a95b, 0xb30ff6ca, 0xa756cfc9,
+	0x397db287, 0x14d9de4c, 0x04cfc2ff, 0xbcd6df76, 0xba60ebc6, 0xde486bbc,
+	0x34f35f68, 0xdeecd739, 0x7de924f3, 0xaf3bc90d, 0xdea0beab, 0xaffd611e,
+	0xbe022a6f, 0xc5e908fe, 0x53cc7899, 0x05b97b79, 0xbcacc7db, 0xbfe6acfd,
+	0xf790bfb0, 0x4f3c39ab, 0xac6b79e3, 0xbe4adfbe, 0x263f244d, 0xa7f31b0f,
+	0xadbd1a4d, 0x9270deab, 0x3f6caefc, 0xc1c8eefc, 0x57e61784, 0x5ef44df2,
+	0xd12d47e3, 0xb3f864ef, 0xe896a1b8, 0xefe98675, 0xdecd73d4, 0xa7d3816a,
+	0xd66b5bd3, 0xb5c7d7e8, 0x3a345bf5, 0xd7e340ab, 0xbedc7eee, 0x7b227b34,
+	0xa749fca5, 0xd7c49fcf, 0xbef9faeb, 0x4d6edf46, 0x766a414f, 0xbe487f12,
+	0x7f5067e4, 0xcd50eb14, 0x2b94e06f, 0xfcb1373b, 0x8e979751, 0xfbe1a7ff,
+	0x0a4c4942, 0x04d5b1cb, 0x5ded9a73, 0x1a23c5ef, 0xe8ad4f1f, 0xfc82de9e,
+	0x06f31cfb, 0xbea36e96, 0xd3ce239a, 0xaf3f46f5, 0xfdc95b60, 0x86600eac,
+	0x9fafed20, 0x304d527b, 0x81273ed3, 0xbf8e87db, 0xf505976e, 0xe81feed3,
+	0x1c5779a4, 0x37443999, 0x0cd3ff3f, 0x5457ad89, 0x4af3f307, 0x7010108b,
+	0xebc094e8, 0xec56a782, 0x84e79671, 0x327ffdf5, 0x62ff4e84, 0xa78aeb4e,
+	0xfa117fae, 0x782cf0cc, 0xd4fdf276, 0x7a16f4fe, 0xd74df3a9, 0xca9c584f,
+	0x3d5893e2, 0xca7dedad, 0xf6c5e74a, 0x19caf1eb, 0xfec97bca, 0x10faf2f7,
+	0x5923872c, 0xeeb0f711, 0x0ecff1d7, 0x84f7c343, 0x65e72f7e, 0x2efa3066,
+	0x1f35b341, 0x85d578fd, 0xa6539150, 0xf5644fb8, 0x08bfbba0, 0xb3d42e8a,
+	0xbceeddaa, 0xeaad7bc8, 0xa7a611f2, 0x756210d5, 0x1f55b2cb, 0x7d230f16,
+	0x45c629fc, 0x8c5757e4, 0x3d5c52b5, 0xa1cccb77, 0x0dd3e724, 0xb8badefc,
+	0x142bc87c, 0x786e9f17, 0x0e666d21, 0x1a282bad, 0x6b65fd89, 0x9e393207,
+	0x53150f89, 0x1171fdbb, 0x6a358ea8, 0x53c590cb, 0xd1ca3db1, 0x39e89137,
+	0x8e1624d9, 0x2cfb908c, 0xa54f1690, 0x92eccc78, 0x6a4e0ec8, 0xc2deabfe,
+	0x933bd2e5, 0x277279eb, 0xa0bafee4, 0x44db3ebe, 0x7e84db7c, 0xc5aee913,
+	0xd507a018, 0xff1e054d, 0x5c42b1cc, 0x1fd63ccf, 0x4dfb13d7, 0x09aa2709,
+	0x13c7b6ff, 0x4ddbaa24, 0x883781b3, 0xfb0769ed, 0xa6e851e4, 0xc6287168,
+	0x9d5cecab, 0xbdf9fa4f, 0xc51eb932, 0xf5e20bee, 0xc76fa4a1, 0x19d63ab4,
+	0xcb8bef0d, 0xee893a7f, 0x3d5b8ba5, 0xf9f74449, 0x140df1fd, 0x70b79cc7,
+	0x7fefa57f, 0xd88f77bb, 0x77d88ef7, 0x0c65800d, 0x189901af, 0x6740253f,
+	0xe3eba4fc, 0x70baeaed, 0xcdd78d22, 0xfaf53db5, 0x5ecd1ffa, 0xd64de257,
+	0x12f2f0e9, 0xce365864, 0x9e083aa3, 0xe5f0985f, 0xd9c510ae, 0x86c9368d,
+	0xbd259ef4, 0x86fab6c6, 0x1481eac8, 0x9f9e9313, 0x8050637d, 0x138ab7f2,
+	0x2cd15dbc, 0x7884a804, 0x22b3ed64, 0x718d9fcd, 0x2221775e, 0xed717d6e,
+	0x6bff5224, 0x9ecaff5e, 0x56daff38, 0xe037bd81, 0x5e263db0, 0xfca7b77d,
+	0xf5578bef, 0xe754bffc, 0xa42ef6bb, 0x57cf5bf3, 0x0c5ea20e, 0xedf30bcf,
+	0x2c3992ea, 0xc4bcbd5e, 0x709b60cf, 0x5628cd6a, 0xbab176e7, 0x5e5a2fef,
+	0xdc917b10, 0x4ebc0f77, 0xbb48e779, 0xbdbb224b, 0x401164ba, 0x63abad36,
+	0x5ebb9750, 0xf77bb9df, 0x6aeeb621, 0xa37da0cb, 0xbb01dd70, 0xcee7f98b,
+	0xf79892f7, 0xbc58ef91, 0x7dd913dd, 0x7451dfc1, 0xd772ebf9, 0xded4abe9,
+	0x2ffc7445, 0xdee5c53d, 0xd33aded9, 0xfa1ef449, 0x11bef251, 0x9704ef24,
+	0xb9da7144, 0x5c5fdcf5, 0xf2bd7388, 0xc2fbca20, 0x80698986, 0xa1f6727e,
+	0x7f512787, 0xa2417ec5, 0xbfbbdcf8, 0xd3f949c4, 0x4eadcdef, 0xbf7ef3ca,
+	0x0abf7f5e, 0x469fe488, 0x3489838e, 0x0cd0647f, 0xe233ee32, 0x2337ae41,
+	0x0a039b2e, 0xcfd83bc5, 0xbbd716ea, 0xfe05bab1, 0xfde09725, 0xaf2f7b39,
+	0xd072e873, 0x1a51adf3, 0x9333079d, 0xfeb4ff38, 0xffbce182, 0xe35d86fc,
+	0x4d8bbf69, 0xbef08916, 0xfb6164da, 0xc93cc3cf, 0x5b33ef44, 0xed421e79,
+	0x37d93778, 0x35e637cf, 0x7aa37cf3, 0xbab14581, 0x4167eaba, 0xe7ca3cc4,
+	0x73a64fb0, 0xa7fa6ae7, 0x853cedeb, 0xef0f37d3, 0x5637a235, 0x99a1138f,
+	0x4edddbbd, 0x3d56ff9f, 0x37e31dde, 0xa982584e, 0xef9f8993, 0x77565e89,
+	0xe87e85b9, 0x403f257c, 0xc3ebac7a, 0x0517d43c, 0x079a9287, 0xea3e93e6,
+	0x9bfba1fa, 0xa5e35bf9, 0x2dfe51b3, 0xdb5daa31, 0x1b3a53d2, 0x7a69e719,
+	0x99823908, 0xe9d2abe8, 0x535d6b8b, 0xf8a6b082, 0xc421150f, 0xafe88473,
+	0x9152a4f0, 0xc05e8bef, 0xc69ed964, 0x642091de, 0x061ef601, 0x6a17ff4c,
+	0xa2b1b075, 0xe519bfbd, 0xf3742a58, 0x96d93e52, 0x2e4d50ef, 0xf9327796,
+	0x04253e96, 0xc44ffb00, 0x9ea9733b, 0x0e420330, 0xbb3fb637, 0x464d7643,
+	0xf5ebd5f9, 0x2d63b56a, 0xe65767d5, 0xfffb5a8b, 0x3aa8944a, 0x8afdffc8,
+	0x530f20ea, 0xffb5943d, 0xf6f5425a, 0x71b477b6, 0xe3425eb8, 0xab402b6d,
+	0xdbc57d8b, 0x85c7d03f, 0xdb5fb409, 0xe45c7d2a, 0xfdf5e7ed, 0x98daf6b5,
+	0xe331a5f1, 0x1ff6f12f, 0x504e227e, 0x88f5c273, 0x9ef7b527, 0x7d5a65d2,
+	0x8336f4c2, 0x93df2cf6, 0x1d17cf54, 0xe7941d67, 0x27947914, 0x7bf0dde9,
+	0xcd7ffd7d, 0x44af3844, 0x9f78ad50, 0xd9026fc1, 0xda75e7c4, 0xe1ffdd7c,
+	0xc9bac83c, 0x6f3439bc, 0xcb3cd448, 0xda7de6cd, 0x376ebf75, 0x89e2979b,
+	0xe77f4fde, 0x9c56e29d, 0x278f5f34, 0x94afef62, 0x8c471e2c, 0x699254ca,
+	0xb40b239e, 0xe9bc42b8, 0x16615a3b, 0xbb4a038f, 0x99b677f1, 0x32711d43,
+	0x37e431e2, 0xf9077afe, 0xbe3c0db3, 0x98efb57a, 0xf41fb43e, 0xc757be8e,
+	0xcc43ef21, 0x3cb3b5af, 0x86fb8ff2, 0xd1bfc8b8, 0xebaf24db, 0xea9f7da5,
+	0x759dc5ee, 0x67dfbea6, 0xacb74def, 0x409df543, 0x5f71a875, 0x66f3bd71,
+	0xc93efcf5, 0x01b0da75, 0xb48fc8dd, 0x25effba4, 0xbb0da63c, 0xf9da88bb,
+	0xaef662b6, 0xefafa3dd, 0x6b3fea1e, 0xa1089e52, 0x6346fb3e, 0x7f611b20,
+	0xa2130a13, 0x983f0fda, 0x3d844a0c, 0x57b87a4e, 0xd291fb54, 0x53f9a8cc,
+	0x3515dd67, 0x0d0b6c5e, 0xf69c7966, 0x78ab7cce, 0xc61df3a2, 0x9f3debc1,
+	0xc7d3fbed, 0x6a1fa28f, 0x5d59126b, 0x2dfeee77, 0x70828f1f, 0xec46805c,
+	0x74bdfb22, 0x93ca044c, 0x5757eb63, 0xb46eb420, 0x80f135f0, 0xe5c58474,
+	0x8d1fa8ff, 0xc6a34bc8, 0xfb4567a7, 0x1f756bd3, 0x23180ba5, 0xac7568d5,
+	0x594f7b2b, 0xbbff5e65, 0xeafd0371, 0xde90b47d, 0xfe903b3b, 0x2d7fa841,
+	0x38adaca7, 0xe64b23de, 0x05ef4779, 0xe5007966, 0x4acc0b11, 0xa1171e84,
+	0xbea8b81d, 0xde3f71d2, 0x83474878, 0x74d7d6c2, 0xb8bb5898, 0x5465b4d7,
+	0xc336aa6f, 0xc754dd80, 0x6e2c1d90, 0x7df9fe2a, 0x67e580e3, 0x63f74282,
+	0xe7c2a355, 0x4c1ebf8b, 0x1db1b0f5, 0xc53ed8e3, 0x50131c22, 0xe0281f37,
+	0x31d5d17e, 0x76d7f581, 0x6e0f9b00, 0x54ace2ff, 0x7144aefe, 0x29aefe6f,
+	0x72888985, 0x86795bb2, 0x2297ebf4, 0xfbbb222c, 0x86f7bde5, 0x5f7a1494,
+	0x32283418, 0xc59358ff, 0x6a400ddf, 0x5df74ebf, 0x990481ed, 0xc5eac8e4,
+	0xb7ed0f14, 0xec7c77b1, 0xda90f567, 0x94cdb603, 0xdb3ad5e7, 0x0afec9b3,
+	0x8907c60e, 0xe307438e, 0xcdbf2ccd, 0x73a1aff4, 0x2bf60e0b, 0xd9537ca4,
+	0xf3033367, 0x98f7ca63, 0xca589e3d, 0xd25d3ae4, 0xea37fb04, 0xbfbf6c63,
+	0x4cabf381, 0x872ebbe3, 0x7eef8d72, 0x1a659f8d, 0xdf1067df, 0xc5f2ee91,
+	0x587b6fc8, 0x29a2c52e, 0xb260fd7b, 0xca5e3e8f, 0x15bf743f, 0x8eaaf7cb,
+	0xeaf7c638, 0x7c9a1e8e, 0xa1c945bb, 0xb58323bc, 0xdb7dba07, 0x7ae2ce89,
+	0x5fee2f94, 0x287a0f6e, 0x17519a3f, 0x87f7a5ef, 0x45dd8b6b, 0xab85d7c4,
+	0xfee6e1b0, 0x8ab5ee8b, 0x74add4df, 0x4a65c844, 0xb8c82e47, 0x2c8ec4df,
+	0xebed5f06, 0x2e724bdf, 0xcecf18eb, 0x66c8ef29, 0xa95c62e7, 0x23fef8b0,
+	0xfbffbe3d, 0x20b9c7be, 0x21811b30, 0x75dfca8a, 0x26ef3602, 0xcbfdfc21,
+	0x54238c4c, 0x1696673c, 0x5b650e7f, 0x6c9b7f22, 0xf8424381, 0x72ff18fb,
+	0x9da86a91, 0x3be5d7f6, 0x678c7c99, 0x26e5dd70, 0xd543930b, 0x950f5fa1,
+	0x9092fc70, 0x4eec9ddf, 0xc47ff24d, 0x810995ee, 0xcf6c161d, 0x73977e7f,
+	0x8fd61c84, 0x43d07366, 0xf919a67e, 0xe6b5f548, 0x00b8e2d5, 0x942667af,
+	0x9eb5b80c, 0x83b8ebcc, 0xb79febf1, 0x286b4dd3, 0x66cb869f, 0xed879796,
+	0xb78ef375, 0xb8127f48, 0x6ecadcfe, 0x58912cf7, 0xbf6b167c, 0x959eddba,
+	0xf9227ffd, 0xcfadbbde, 0xfee11fe0, 0x584a5363, 0x99ef1cbd, 0x87583ca8,
+	0x7f31eec9, 0xb3b7f9a1, 0xcac7bb6f, 0xcf6b12f8, 0x6db37d7a, 0xb2019dff,
+	0xf68093f7, 0x2fb47a4f, 0xd9a2de41, 0xdd07f33b, 0xb3af815c, 0x339951f7,
+	0xddef14ed, 0x9e1ddec4, 0xdece6fd4, 0xf23939bf, 0xf85201bc, 0x1c846afb,
+	0xa43d50d7, 0xc9852db8, 0x2e9b9751, 0xba7e5d47, 0xfd963ebc, 0x637d1177,
+	0xd775fdb6, 0xd55762ff, 0x53073deb, 0x793b767c, 0xfee1e3dc, 0x8f285ff6,
+	0xf7373c7b, 0xfe7af7bf, 0x7e8bff09, 0xe26ec072, 0x5bc3a64f, 0xa3f9872f,
+	0xb741e3cb, 0x2433d8bf, 0x68241fcc, 0x6f4fd530, 0x9ee98fc7, 0x99fefc6f,
+	0x67fa0b7e, 0xfff433fc, 0x1292f9c3, 0x78b12f9d, 0x49e50466, 0x5527963e,
+	0xfc83bd53, 0x4fe818df, 0x15bd59f1, 0x9aa43e58, 0xee4979ba, 0x42d97625,
+	0xc5259771, 0x3afd216b, 0xbea211ea, 0x48f56bef, 0x2496087d, 0xbd0ddfe8,
+	0x128cf2cd, 0x43cb4997, 0x30f6f30e, 0x09bdfc7d, 0x0c94d794, 0x18e1c6d2,
+	0x6c089f1f, 0x472df9fc, 0xcdb25e59, 0xfc20aca3, 0x4c48f05e, 0x74b7173e,
+	0xfde7e3ef, 0x5747c40d, 0x04bbffb2, 0x5676ad6b, 0xda136b4f, 0xc3b91773,
+	0xdfb57ff3, 0x8817f271, 0xaca3c877, 0x502f79a1, 0x12352231, 0x16ad79f2,
+	0x84e6f748, 0x5601cbfb, 0xae58bdd0, 0x6044fd1e, 0x1c3c1df1, 0x1fbe9437,
+	0x959f7604, 0x35b1dfeb, 0x06c77f44, 0xb3e3175e, 0xfea11dfe, 0xb09de09d,
+	0xd7bed25c, 0xf92370db, 0x866ce4eb, 0xafdf1173, 0x2e9059b5, 0x22efd757,
+	0xff477d85, 0xabd6fb42, 0xf2cf1ad0, 0xd7efe8d9, 0x4f3013aa, 0x7b287e45,
+	0xca3b411c, 0x04f6e171, 0xab986756, 0xa47b8fd7, 0x31f44bb0, 0x3d03d389,
+	0xfd7dd8d2, 0xe74ba6b8, 0x2cff4097, 0x6fcdc533, 0x1e23355f, 0xeff1f66f,
+	0x99543925, 0x285f483e, 0xdea45d53, 0x95d0f749, 0x4a97a21d, 0x9d5a5530,
+	0xb5529a1e, 0xbc188afa, 0xb98fc717, 0xf54cdcef, 0xfe1ec3bd, 0x0dd9714e,
+	0x5d519dec, 0x10b6936a, 0xa70fd72e, 0x71ed8d3f, 0xe979f719, 0xcef2eefd,
+	0xa366b367, 0xc05ad01c, 0x00fae2c9, 0x36c1675b, 0xd2b91f7d, 0x96ba42fd,
+	0x4220fbbf, 0xbb9f3a1e, 0xa28f1ff0, 0x0e29bdfa, 0x09dede98, 0xc78a241b,
+	0xa58c3975, 0x838a53c7, 0xc8efe0de, 0x6ce8050b, 0x3e725c53, 0x5dfbe78e,
+	0xb0ef8beb, 0x52cd017c, 0x6cdbaa8d, 0xfcf37998, 0xa5602522, 0xbae9bfad,
+	0xf68f25e3, 0x06103e4e, 0x791deff2, 0x9b52a35d, 0x952c5633, 0x6a4f44cd,
+	0xbfe4979c, 0x77666370, 0x7c3043f2, 0x83ce441d, 0x2f4cc90d, 0x1845ba56,
+	0xb7367a8e, 0xfba4c905, 0x160ee5a8, 0x92d83df7, 0xe313a0c4, 0xf8bfea4e,
+	0x7983fbca, 0xa7cb871d, 0x63eed5f8, 0x6afb1c58, 0xf563063f, 0x4fb9fcd6,
+	0xd84b77a1, 0x123ab023, 0x75ca12f9, 0xc7563d75, 0x6eac8378, 0xeec096f7,
+	0xbf9b293d, 0x75495463, 0xd20e1dec, 0xc22f2c7d, 0x2ec94b25, 0x6cddf45d,
+	0x8faeff12, 0xf7dacce5, 0x3e4d4eb1, 0x7df9d009, 0x881bef6c, 0xccb67277,
+	0x3d208bbf, 0x264949c1, 0x5cfc9bb7, 0x393fb66a, 0xe6bde29b, 0x9f54459d,
+	0x4aea0b9a, 0x13d94770, 0x6ae848dc, 0x77b497a7, 0x2b0f5d31, 0x852173b5,
+	0xc6855feb, 0x3e8d4ad6, 0x2b5418cb, 0x4772cfa4, 0x377ae6f6, 0x80fe8691,
+	0x1f0adf7d, 0x40fe3dfc, 0x3f7407bf, 0x2cf535fd, 0xc7d2cfc6, 0x7ce3ddb9,
+	0x0ffd175e, 0x511f5021, 0x6faceac7, 0x77b0233d, 0x326ebd6d, 0xbd0a0ccd,
+	0x095422ef, 0xd2f7d296, 0x8598e104, 0x62e38b9d, 0x8cc3ab12, 0x63ebfc62,
+	0xd39d5952, 0x9830ebce, 0xfd43dc27, 0xbf66ead7, 0x7920f8cb, 0xc51bf7d9,
+	0x6c4f5e36, 0xdf644f7f, 0xee8add1f, 0x555e5af1, 0x6f62e07b, 0x5f71dbb2,
+	0xf7e27e89, 0x11e38b71, 0x67c753ca, 0xf197f7f3, 0x9828a6d9, 0xe98bfb2e,
+	0xea3b3fdc, 0x5f909ba6, 0xc167be7f, 0x8771df2b, 0xae887ae9, 0x40a7ba2c,
+	0x4d3e90b6, 0x371de504, 0xf06df56e, 0x5064f6b3, 0xf7496c19, 0xa54eca4f,
+	0xac0310ef, 0x7dc47c7d, 0x8b125eac, 0x655e2c7f, 0x2c5efcd9, 0xc14536cb,
+	0xcfdfb271, 0x5e1852f1, 0xd1365fb8, 0x24f5e3a5, 0xdfea4ee9, 0x88555f25,
+	0x7ccaef90, 0x02d7df91, 0xf2c3afbf, 0x25496b1d, 0xbd108ef9, 0x0c17ab4c,
+	0xe0f48479, 0xa5eadd9a, 0xbe5d3b77, 0x9bde8108, 0x638e72c3, 0x88ba52fa,
+	0x6c91cbeb, 0x6a0baf37, 0x580e4772, 0xf942ae28, 0x9a7c4d3f, 0x75e44237,
+	0x3fe50938, 0x4addc980, 0x82995e36, 0x2ae8cf72, 0xde2d5adf, 0x66e2d1ad,
+	0x66f51fbd, 0xeae9f105, 0x89fd61de, 0x4607b53f, 0x7a9fc4c8, 0xd4dbcf13,
+	0xd0ead6ff, 0xa38bc99a, 0x47ef62ee, 0x297af660, 0x45be2aff, 0x9355b29e,
+	0x961ef1cb, 0x5fb60171, 0x4076bbf4, 0xa6809fd3, 0x9c70ff57, 0xe2a21fd5,
+	0xffa211ac, 0x754fe265, 0xf32b55dc, 0xfe7bda63, 0x19b37a56, 0xd11fd612,
+	0xb26ff0ea, 0xf5cbe247, 0x55f9f506, 0x44e7ff5a, 0xa64ce7d4, 0xef28e7cf,
+	0x75e56e3c, 0x156456e7, 0xff107ffd, 0xccbbfccb, 0x3e9a77f0, 0xbc512b06,
+	0x8042e704, 0xbad0f9c1, 0x9fce1072, 0x469658d8, 0xfc9854f8, 0xa2ad3eec,
+	0x3be9bc2f, 0xec2f681b, 0x76fcfc81, 0x7a77d2c3, 0x5e538a20, 0x5c285b74,
+	0xb041b49b, 0x97be40f4, 0x41582d34, 0xb78fb46d, 0xb70f4e48, 0x0ee77a6c,
+	0xe39d17ec, 0x8f1c7d98, 0x1cac5391, 0x45fc6b0e, 0xd39b787b, 0x70ba30a5,
+	0xa1ed4dfd, 0xe12bdd06, 0x01ffffa5, 0xd5b93efd, 0xefd023ff, 0xe3781b15,
+	0x7a6ec8e2, 0x3fbfcc83, 0xf3e7f68f, 0xfa27df80, 0x923afa7e, 0xf397dd20,
+	0xf4979ba1, 0xfb8592ea, 0x807d4bcd, 0x0125c1d9, 0xf0bcec2f, 0xfee906fd,
+	0xe199791e, 0xf4d6a37b, 0x39f92e5a, 0xc77e57bc, 0xe699e2fc, 0xe4dd0395,
+	0x64e903ff, 0xd99f96af, 0x735ff5bc, 0x9510ccbf, 0x80ceb4d3, 0x01a03406,
+	0x0340680d, 0x0680d01a, 0x0d01a034, 0x1a034068, 0x340680d0, 0x680d01a0,
+	0xd01a0340, 0xa0340680, 0x40680d01, 0x80d01a03, 0x01a03406, 0x0340680d,
+	0x0680d01a, 0x0d01a034, 0x1a034068, 0x340680d0, 0x680d01a0, 0xd01a0340,
+	0xa0340680, 0x40680d01, 0x80d01a03, 0x01a03406, 0x0340680d, 0x055ff01a,
+	0x328d1fff, 0x800060f6, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00,
+	0x30001131, 0xee300408, 0xd80ea5ea, 0xabdef271, 0x964d2104, 0x5dbbcce4,
+	0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0xee017e3f, 0x0014ab55, 0x000014ab, 0x00088b1f, 0x00000000, 0xc5edff00,
+	0x30001131, 0xee300408, 0xd80ea5ea, 0xabdef271, 0x964d2104, 0x5dbbcce4,
+	0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0xee017e3f, 0x0014ab55, 0x000014ab, 0x00088b1f, 0x00000000, 0xc5edff00,
+	0x30001131, 0xee300408, 0xd80ea5ea, 0xabdef271, 0x964d2104, 0x5dbbcce4,
+	0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0xee017e3f, 0x0014ab55, 0x000014ab, 0x00088b1f, 0x00000000, 0xc5edff00,
+	0x30001131, 0xee300408, 0xd80ea5ea, 0xabdef271, 0x964d2104, 0x5dbbcce4,
+	0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0xee017e3f, 0x0014ab55, 0x000014ab, 0x00088b1f, 0x00000000, 0xc5edff00,
+	0x30001131, 0xee300408, 0xd80ea5ea, 0xabdef271, 0x964d2104, 0x5dbbcce4,
+	0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6,
+	0xee017e3f, 0x0014ab55, 0x000014ab, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x00088b1f, 0x00000000, 0x62f3ff00, 0x51f86063, 0x408cc10f,
+	0x7f120cb6, 0x66476028, 0x48107d08, 0xf3e2061f, 0x2fe9a48c, 0xb9b04160,
+	0x40afec80, 0xa8597833, 0x88a1bee7, 0xcfd2738f, 0x81ae792e, 0x66322ff7,
+	0xe86067e6, 0x6ff047e4, 0xb3caa3f2, 0x3dd7d3f0, 0xb000c6b4, 0x00eeff4a,
+	0x0000eeff, 0x00088b1f, 0x00000000, 0x7dd5ff00, 0xc554780b, 0x3d9cf0d9,
+	0x3764dd97, 0x2485cd9b, 0x200d8410, 0x125c40a2, 0x126e20ee, 0xc3116088,
+	0x65e28145, 0xb201ae41, 0xdb3f6911, 0x5cbb7ffa, 0xc6d6a444, 0xa0b45b4b,
+	0x06a2828b, 0x82482459, 0xa5c8ba1b, 0x5835b5b4, 0x0da978aa, 0x24c4dc88,
+	0x97f4b004, 0x3bef3fca, 0xce7bbb33, 0xf4bc4c6e, 0x7d6f9ffb, 0x7339867c,
+	0xef3bccce, 0x6779de7d, 0x313b10a2, 0x0ae42775, 0x64246efc, 0x19084c99,
+	0xbbc32d14, 0x32413cbf, 0x46bcf909, 0x08f39a75, 0x86a1d929, 0xc5f5a46f,
+	0xd400a41b, 0x0893bb4d, 0xd3484819, 0xcaf7563a, 0xa08e8f0d, 0x4ca764ed,
+	0x8482d136, 0x4592266d, 0x610b09c8, 0xd60a673f, 0xeab126e7, 0xee7b0de2,
+	0xd7e7fe82, 0xfd12499a, 0xec8d44fe, 0x4d92fa86, 0xb4488052, 0xad21eebd,
+	0x5e7fed0b, 0x39260a40, 0x63d37d69, 0x9085339a, 0x6cbfbbe5, 0x5c057182,
+	0xe2167c67, 0xbfbe00d4, 0x23f63565, 0xadb03ca4, 0xa7ed0bb4, 0x99725abc,
+	0xebe538e8, 0x038df0a4, 0x9014b9e1, 0x370bbf69, 0x0fb3895b, 0xc10ae183,
+	0x4b1e7171, 0x6971d3d6, 0xd2a9447f, 0xe3a252ca, 0xf19cfc09, 0x4471c5ad,
+	0x744bf17d, 0xc55dfa1c, 0xa60b92ab, 0x9e226158, 0xa9bf1d20, 0xf3da692e,
+	0x32df9836, 0xec4a77ad, 0x0ebe663c, 0x6c0f28f3, 0x006d4ad0, 0x6e6d06d7,
+	0x0b7f68bf, 0x0f56ff37, 0xd99edad7, 0x6376989e, 0x5147c679, 0x2f5a3bda,
+	0xb41dc427, 0x6d01fc01, 0xf3e8ff62, 0x908d295a, 0x4cbfd04e, 0xff68969c,
+	0xb9f1f884, 0xc0615c4c, 0xb8a5093b, 0xead6e962, 0x496dd3d0, 0x0db6ff1a,
+	0xe9c97e78, 0x81cf98f0, 0x1cbe13e5, 0x73e57f2c, 0xf1bf9c22, 0x29f2c1f5,
+	0xff9f0b9f, 0xcb1437d6, 0x96373ef5, 0x62c6facf, 0x8657c1b9, 0x9bef3def,
+	0x9f26e586, 0xe8bf9f07, 0x4be58f9b, 0xfe7c4abe, 0x2c7eef8a, 0xf8fcf8b7,
+	0x356fab7c, 0x557cdb96, 0xf4e76e58, 0xe00be1da, 0x9b7d3b7b, 0x05f3acb1,
+	0xbe1bf9f1, 0x017f2c5a, 0xaa65a478, 0xa14cb1db, 0x4d1d4a74, 0x484d941c,
+	0x32d95946, 0xa633d695, 0xa7b67ab0, 0xf146996a, 0xd69b3d94, 0x2b731d29,
+	0x6999961b, 0xd652ee7b, 0x58efddde, 0xddeda16e, 0x9ef6b257, 0x93cb6555,
+	0x27cf7b68, 0x8135fb59, 0xb4c9e5aa, 0xac8d9afd, 0x61b06fbd, 0xf7b695b9,
+	0xd7ed61ad, 0x6c2b1d87, 0x3efd7eb4, 0x286c2f56, 0xeb42915b, 0x7d598785,
+	0x0ad56348, 0xc87efd3b, 0x88fdf671, 0x4ceca096, 0x55db1fc0, 0xd2843dd7,
+	0x52fed7ee, 0xd68c32b7, 0xc47dd735, 0xff66c845, 0x33a56ead, 0x865a87c5,
+	0xd3fcbbed, 0xc32b5d58, 0x33fcb7f6, 0x6a9dfdf1, 0xec7fb625, 0x9df6c72f,
+	0xb7b6255a, 0xf6c3eff8, 0xdb0ab53a, 0x601ecb4d, 0xdb0aad75, 0x883d9733,
+	0x52a1bfef, 0xd2138760, 0x6ea3d97b, 0x966f853f, 0x7d02a9e4, 0x9caa6cd2,
+	0x28b7404f, 0xaf901ce1, 0x0d322487, 0xc4a7e5e4, 0xa1e6fa89, 0xc83734ab,
+	0x8343f6e1, 0xc539079f, 0xe7d4265f, 0x0f26b0be, 0xfb0a79fb, 0x3d3f68d1,
+	0xf0a7efdb, 0x7ebaa19d, 0x2f99df0a, 0xf80e79fa, 0x63b939be, 0x677f6cfd,
+	0x779e1eb8, 0xaf3f45ca, 0x8ef63f60, 0xaff0abcd, 0xfcf0f523, 0xa7e89175,
+	0xde95e706, 0xbc767831, 0x1793f14f, 0xfbc767ed, 0xf6123f14, 0x64fd8f53,
+	0x419e0c75, 0xf5d50c1f, 0xf983e833, 0xd8039fa2, 0xf58e974f, 0xe183e3b3,
+	0x283e787a, 0x8dbcfd17, 0xd8eb74fd, 0x387d06bc, 0x87cf0f52, 0x473f448b,
+	0x1d1e9fb0, 0x51e767eb, 0x1e767e3d, 0x28e7e08d, 0x63bbd3f6, 0x4c721af3,
+	0xc7219f8f, 0x8339f822, 0x63aebf74, 0xa63cecfd, 0x63cecfc7, 0x479cfc11,
+	0x363bf278, 0xe89f21af, 0x93e433f1, 0x632e7e08, 0x831d053f, 0x1eb4eea7,
+	0x23a7753f, 0x982551f8, 0xc18ee0d7, 0x1e8cec33, 0x44cec33f, 0x009763f0,
+	0xeb1de19e, 0x1e8ceea7, 0x2267753f, 0x7846c9f8, 0x5e6c7546, 0xe3d33ec3,
+	0x1167d867, 0x9e1138fc, 0x08cdcae2, 0xf4fda10f, 0x379fbb6f, 0xcff5e9ce,
+	0xfe98e71b, 0x8fc47819, 0x21cd57fa, 0x6467002e, 0xa25f994b, 0xd6932575,
+	0x89ba509d, 0x055fda87, 0xf83f70ec, 0xd32fd1b8, 0xb4ea94f6, 0x58ce3582,
+	0x02ec5c7b, 0x4d03ec78, 0x18ef7b3a, 0xa7abac99, 0xd9d74e8f, 0x5df1cceb,
+	0x29acf574, 0x9cf5743d, 0x7dd3ae3b, 0x817665df, 0xd175deae, 0xdbbd5d70,
+	0xf7dd62d2, 0xd66e07ce, 0x3958f7b5, 0xf5ef5749, 0xf5750fc8, 0xd2ce4fde,
+	0x2bacfbd5, 0xdd77f5d7, 0x7aba25c6, 0xea9feabf, 0xcb35f9ea, 0x68577575,
+	0xb05eae8d, 0xff5d71ef, 0x5a7adf03, 0xf87c1f57, 0xe87d5d39, 0xbeeb2f47,
+	0x35fc7e1f, 0xd9e47d5d, 0xc7ff065d, 0x97d138e6, 0x77f065d7, 0x9a07e8ad,
+	0xcfe869bb, 0x60b37516, 0x6cddc7fd, 0x28f1ff58, 0x4a0e35cf, 0x07c39fd7,
+	0x58fdeed4, 0x48ef5cf3, 0x25297f60, 0x4a07244d, 0x8d0c898f, 0x52ad5f6f,
+	0x527ca3be, 0xef72fddc, 0x9af4b4a3, 0xd1a77a5a, 0xa2a6eff2, 0x3f17c0a5,
+	0x97e84c95, 0x09526559, 0x1335647c, 0x57d5cbe4, 0x1fde7e0f, 0xe70fdfc3,
+	0xfd14be31, 0xe656ac3e, 0xeffe1d80, 0x83f3bf65, 0x4f287e1d, 0x93ec3b43,
+	0xf7f6177e, 0xfde7dfa2, 0x49fc0738, 0xd4fd468b, 0x3a35f1d8, 0x7f1c3f7e,
+	0xa6bf8c25, 0xbd2df18d, 0x4fc6ea87, 0x375f31ea, 0xc746927e, 0xe8f21077,
+	0xf1f3dfb2, 0x50bafe29, 0x3dfa50bf, 0x71e96f8e, 0x5abf8e3f, 0xe375f323,
+	0x7fc64727, 0x69fde412, 0x82507f18, 0xae1ef7f9, 0x728f7f9f, 0x3635fcfd,
+	0xbd9667ff, 0xc64fc7cd, 0xbd2b3ff9, 0xe3dfe6cd, 0x66fe6ca7, 0xf8ec6fda,
+	0x37fe08f6, 0xb72ff8c2, 0x9ae5be31, 0xf7f9fa91, 0xbf9fa45c, 0xeaff8d99,
+	0xf8f8f7b2, 0xab7f1c36, 0x7f9b1ef4, 0x7c7007cf, 0xe5d2b9b3, 0xeee64da0,
+	0xd00195c9, 0x263dd413, 0x746020d9, 0x2820219c, 0xefd084e9, 0x903d4817,
+	0xa64e3f0e, 0xffdf477c, 0x1bf29922, 0x6fdcf7f8, 0x12558127, 0x1dd74c19,
+	0x5767a79c, 0x417ed4cf, 0xcb5d993f, 0xae8433d6, 0xf0a6ae77, 0xff9ed535,
+	0xef0a43e2, 0x50038dec, 0x274e514f, 0x3e1cddf0, 0xf0ccaf15, 0xb7594841,
+	0x12bf5489, 0xa4dbaca5, 0x06e4d8fe, 0x411e7fbf, 0x78f8e318, 0xa27a954e,
+	0xe6fe4631, 0x17d7d5ad, 0x0175f404, 0x09301177, 0xa4c5fd2d, 0xf71d254f,
+	0x10275d18, 0x447f97ea, 0x594270fd, 0xdc9513f6, 0xf2f1465d, 0x404f5d31,
+	0x27ae91bd, 0x4cae9da0, 0x14c72e91, 0xb63bf382, 0x7a001ada, 0x0193f17c,
+	0xbb87d81a, 0x1bf2a7ef, 0x727edf2b, 0x5fc28738, 0xd339e7c6, 0x076fa19f,
+	0x4c3c53bc, 0x0b2bf3e4, 0x974f1e2d, 0xd59ee4c4, 0x459954b9, 0xf24a6bdf,
+	0x54ffd0a8, 0xe9b356e2, 0x9d2e7778, 0x5f107aa8, 0xca135cee, 0x5135c939,
+	0x56e89f39, 0xf4c6358f, 0x8066a93f, 0x93227b4a, 0xdf9feac7, 0x37afa656,
+	0x44d56ea9, 0x5134d3e9, 0xdda81b22, 0xdf4d3a99, 0x2339c62e, 0xc5fce3a5,
+	0x6e746578, 0x9e61d39d, 0xeffa63f4, 0x3cd83a1f, 0x34bc3ae8, 0x3f73d617,
+	0x689b5c90, 0x57bc2a7d, 0xa83d7400, 0xb769f3c1, 0x90d6b9a4, 0x92897878,
+	0xe11cfd83, 0xca23bceb, 0xcd0c5fa6, 0x83ff878b, 0x29b756dd, 0x461e7e3f,
+	0xa093a7b8, 0xba76f60a, 0xc67cff47, 0x18fce37c, 0x74e09c5f, 0x407c063e,
+	0x07c093cb, 0x39c7cf14, 0x2bf5441f, 0xc01383e3, 0x4eb09907, 0xa7307c66,
+	0x6ffd312a, 0x479ff4e7, 0x29676afc, 0x9ca9f74a, 0xba73b7ee, 0xe7ab5bcf,
+	0x3cf9b88f, 0x1cc1e1b9, 0x3cf98267, 0xa62547b2, 0x657979f8, 0x2ff91f06,
+	0x8e89b029, 0xbea5677b, 0xfd0376fd, 0x777e9c3c, 0x7165e846, 0xbf3e5197,
+	0x011927eb, 0x7c7133d0, 0x31f5e9c1, 0xb5b4d7a7, 0x4cefed8f, 0xe46be1e2,
+	0xf6b13af4, 0xfb44d579, 0x112f9c6b, 0xf08c6ff0, 0x9bbbe11a, 0xdfa03438,
+	0x9febf7dd, 0xa627f4cf, 0x28b6b79f, 0xf3cc78e3, 0x1971c1c5, 0x68e463c7,
+	0xba89e6e1, 0x74c082fa, 0xd6f3fbdd, 0xcfb5d4ce, 0x6ba05aa9, 0xbdf567bf,
+	0xff4cfaba, 0xdfef744f, 0x5d32ff7d, 0x0f959dfb, 0xcc67daeb, 0x9f574c7f,
+	0xf74a79ee, 0x1b69d4fe, 0x95b7ed74, 0x9f6ba4bd, 0xae9b763c, 0xa75dd13e,
+	0xbdda5f7b, 0x9fc43ba0, 0xd9e79abc, 0xbd9b886c, 0x49cf266f, 0xe84c7af1,
+	0x273ee3bb, 0xebe13f3e, 0x9f29e583, 0xf19f9f0b, 0x1972c50d, 0x176a71e8,
+	0xf4ac754c, 0x3bb33e3e, 0x07217fe8, 0x6eaea89f, 0x7a8d3fa0, 0xc4e7a8cf,
+	0x46bd46fb, 0xa7e0b97f, 0x59cf15b4, 0xb0329124, 0xf312cd0b, 0x43f3c457,
+	0xb2123edc, 0x1a45cb1c, 0xaf45d393, 0x39062e75, 0x7b85a45a, 0xeedada57,
+	0x9c250497, 0x9eaa9ccf, 0xade7383a, 0xef806bed, 0x05abe439, 0xf3dd4281,
+	0xe8479b85, 0x47bb3bbe, 0x402af846, 0xff8187b6, 0x04deb65d, 0xb7bc31f6,
+	0x00fee4db, 0x1319fbaf, 0x09db5bc0, 0xdf59eaed, 0x0ae38cd8, 0xcb0cb9e0,
+	0x5869be53, 0xb079f09e, 0x8f9bee3c, 0x255f31e5, 0xfbbe8d96, 0xe7d8fcb1,
+	0xdf23f2c7, 0xf03f2c6a, 0xc4796155, 0x77cb16b7, 0x0f2c017d, 0xf96336fb,
+	0x65882f8e, 0xcb16af83, 0x4b1b9f26, 0xfc312f21, 0x0c73bdd2, 0x277d08bf,
+	0xce29e1f4, 0x5f80672f, 0x73c5f112, 0xa938be7a, 0x185f2256, 0x0e91a1f5,
+	0x6df77d46, 0xf4dd21f9, 0x9443f0fd, 0x7a06b9de, 0xde8f6cf7, 0x397d221f,
+	0x0e4dd3bd, 0xf7a70f06, 0x45780623, 0x041281a9, 0x7e6665bd, 0x595b711e,
+	0xabc453dc, 0x7b8b3249, 0x337578f2, 0xbc0377fd, 0x10fb04a7, 0xbf6b5fda,
+	0xe21bdbff, 0xcc47adcc, 0x3db5e484, 0xdafd233f, 0x9bf103c4, 0xc9ff0ebe,
+	0x14af0cc9, 0xda84b1aa, 0xbed73587, 0xeac59aee, 0x423b010a, 0xb8d8ae82,
+	0xc05fa46f, 0x78fb7665, 0x7f31374d, 0x6657bad3, 0x776f08f8, 0xf2894e4d,
+	0xd2dcd7ab, 0x95f833bb, 0xd8099a1f, 0xe9b01233, 0xf5a7e5f1, 0xb57c51c1,
+	0xf4f68f35, 0xfc2fffde, 0xda8fa7b4, 0xa48743d3, 0x735fc7af, 0x24f75d31,
+	0x008e2f58, 0x6df41c3e, 0x1ba1b0f8, 0xf81061f0, 0x5adff69d, 0xe92a4d73,
+	0x66e7ed17, 0x13af9393, 0xde4e9ebe, 0xd37e0854, 0x1334537e, 0xcbae9eb9,
+	0xf24dcf65, 0x5dcff020, 0x6f4a7e8d, 0x6bab6eff, 0x49d61385, 0x7612fe67,
+	0x75fa17c2, 0xe7c1eb73, 0x59aeb0ed, 0x1c726392, 0x3e9b6c2f, 0x0e79838b,
+	0x547fac5b, 0x56b7afab, 0xdc596349, 0xa53a99c4, 0x8769e83f, 0xd02f78e3,
+	0x8a58399b, 0x68e4967e, 0xf019e38e, 0x79c9805e, 0x81ea1a75, 0xbe0b3e33,
+	0xd1c37df7, 0x38513f56, 0xf427c86a, 0x2b89107b, 0xbf7edb3d, 0x2e56be4d,
+	0x228f1068, 0x580bf521, 0xffd1252f, 0x7e80f4af, 0x7e8bac15, 0x6fd941bd,
+	0x579e1ebe, 0xdfa3c6eb, 0x1b1ec539, 0xfcb6d77c, 0xe084e428, 0x3f374a5f,
+	0x2fb8a7f8, 0xfbd0e494, 0xbaba696d, 0x43b3b6b7, 0x3fc1c7e7, 0x2d1cd1c0,
+	0x8b47c029, 0x2b355b38, 0x8512d5b6, 0x322c066f, 0xf6259f81, 0xa34b62de,
+	0x968fa1e6, 0x7f4148ec, 0x7c63c978, 0x1a4fcd9f, 0x4359fbe0, 0x29970779,
+	0x51fd3154, 0xb44c470b, 0x89af0f7e, 0x3559aefd, 0xff5fd10b, 0xdd71f894,
+	0x1aae5a77, 0x796a7e0c, 0xa6e68370, 0x989cfb74, 0xffbe82c6, 0x2f63bc9c,
+	0x57b7918c, 0x7be70d64, 0xffbd6acf, 0xbab1d74a, 0xeac75d3a, 0xa4c973ea,
+	0x9a17d82c, 0x974aa4fb, 0xd5aeb8d4, 0x7656ffb5, 0x91e77ce0, 0x876055ca,
+	0xbc29a93f, 0x71e0047d, 0xb0c1fae7, 0x9b459c4e, 0x29fea973, 0x03487cdf,
+	0xe3a6f939, 0x0bef802f, 0xf315c7e2, 0xe407db8a, 0x59bc5878, 0xd8120772,
+	0xf76e5abd, 0x3bfa5e84, 0x1eb31fc0, 0x23e7cd3f, 0x3f9adfea, 0x65c32932,
+	0x4339e945, 0xc1921cf6, 0xa49f008e, 0x5be2f946, 0xd7139ff7, 0xd7bfdfff,
+	0x1d12bbfe, 0x4ffed37f, 0xdefc5e83, 0x0715effa, 0xf044c5ff, 0xb22c3a0b,
+	0x17c01c34, 0xbe096e1d, 0x14dcb853, 0x7c60714d, 0x725ab9aa, 0xc4fbe999,
+	0x2c8e9c4d, 0x3d601fa6, 0x6a5afcd9, 0xe1bd413e, 0x1f3a6e20, 0x552e7f2b,
+	0xd123d9ef, 0x3b4ae175, 0xfe51a771, 0x9953e954, 0x692ad1ca, 0xe5a24580,
+	0x7f739ae2, 0x9bf68379, 0x88099214, 0xa1853374, 0xdd97f701, 0xac776069,
+	0x03b411e4, 0x257ec7b0, 0xd1a70a23, 0x0578103e, 0x8d8397ec, 0x61b6463b,
+	0x9d75a6fc, 0x9955718c, 0x919bf75c, 0x0f782f54, 0xeda2f952, 0xb405f0ad,
+	0x5d7095af, 0x04b07adf, 0x24eaf3eb, 0x6beb4192, 0x60f9ec93, 0x26dcd2af,
+	0x63373c5d, 0x4e7e3eac, 0x5806eefc, 0xf3e6cfff, 0x1c00c405, 0x53bd790b,
+	0x7de1ba59, 0x5c4c5ffa, 0xee67b66c, 0x93f14278, 0x771f4d3d, 0x723f285f,
+	0xed3ae200, 0x06ed6dca, 0x9463900d, 0xf59b85db, 0x21b78175, 0x09e64a4a,
+	0x6236c9b0, 0xadb5bc03, 0x67c1dbc7, 0x8cdfef63, 0x7c71d81c, 0xce64f11a,
+	0xc935e3d1, 0xb5abc7a9, 0x066f018f, 0xf4b8a6bc, 0x3a66a2f8, 0xa9af0890,
+	0xfae87b43, 0x7af1bef9, 0x6fb3e017, 0x57f49f14, 0xa262c966, 0x56767007,
+	0x46ffc8c2, 0x85b5c979, 0x55f7f825, 0xf8658199, 0xf5052b95, 0x5f870e40,
+	0xb3becd9a, 0x60c8bc82, 0x2f285d55, 0x3b046910, 0x391f256e, 0xf2268e80,
+	0xb7d2b909, 0x37de3a6d, 0x59d1df71, 0xc075abf4, 0x1f23d7cf, 0xe9fc5d20,
+	0x0f22b024, 0x00e5d1e4, 0x59f416bb, 0xbae9b39c, 0x711f55f1, 0xcdfea6ce,
+	0x96073e6b, 0xf7b0168f, 0x0091fb9b, 0xe1831bac, 0x884e590b, 0xeba1cf67,
+	0x1fffaa34, 0x80265ad5, 0x9517fc3f, 0x55eecfad, 0xab772c35, 0x47ffae26,
+	0x79f201ff, 0x00ffa99a, 0xd0e6a5f1, 0x50d5cb26, 0xbf3195af, 0x75569bc0,
+	0x14df8c2d, 0xb35c1952, 0xf64f182d, 0xc768abd6, 0xc48f669b, 0x9f311a7c,
+	0x163c90a6, 0xa1cdd9e2, 0xc4f7ec4f, 0xdaa69dd8, 0xbb05ae27, 0xb12994f7,
+	0x628cfbff, 0xc87e6efa, 0xfa3d38ba, 0xf509d28c, 0xe83da3e7, 0xb6262a3c,
+	0x9becbf22, 0x77ce337b, 0x9a724847, 0xf2a5dc29, 0x0fe83f1d, 0x537f86fb,
+	0xee24de98, 0x58b333f1, 0x5f563ce8, 0xabd4888c, 0x7ec32afe, 0xc3c45d25,
+	0xcac0e2be, 0x8c87e8b9, 0x0292490e, 0xc108ebfa, 0x0fc8467e, 0x7f32c094,
+	0xe3a28350, 0x0921c1c7, 0x91b836f1, 0x90d37e60, 0x61f6fa23, 0xae218a8d,
+	0x9f819016, 0x3fd29c46, 0x6d7c13af, 0xd7c05927, 0x25965e4f, 0x5b9a4ff0,
+	0x1d396b88, 0x977679bf, 0xf9a78022, 0xd046ec02, 0x363cb2f2, 0x8ff7fad1,
+	0x9e87f30a, 0xc6fed8d2, 0x816d7353, 0x6fbe9465, 0xb8c72dce, 0xf13f17c4,
+	0xbfb44f74, 0x5287a315, 0xdc83684d, 0x75cfb0a9, 0x1295c4df, 0xd552031a,
+	0xfe252cb9, 0x82bfc17d, 0x771f059f, 0x09ec9b9e, 0xb2671824, 0xe294fcca,
+	0x8d247db9, 0x5bb4d7c2, 0x4da57422, 0x747b969a, 0xe9630ef1, 0x0173cb27,
+	0x79ed6f1e, 0x2594ec0d, 0xfa0fd894, 0x3ca3b14c, 0x2ff72d34, 0x1fb5bc83,
+	0x3bf46153, 0x27dfaebf, 0x48723e31, 0x2cefb271, 0x9e813355, 0xe42c732d,
+	0xd6379639, 0xf00a5279, 0x55b8c97b, 0xcba7de14, 0x9f37684c, 0xbf085d52,
+	0x7961317f, 0xa2ff7c71, 0x61b204eb, 0x5cfc8c4d, 0x95248a54, 0x0ea92bf6,
+	0xe3c33fec, 0xd77e0092, 0x8769ffbf, 0x4be690fd, 0x1763a466, 0xdf10f71e,
+	0xfad95575, 0xfeea8371, 0xd0f58152, 0x0b944cfc, 0x9ceec797, 0xf5f2ed4d,
+	0x09ef14b5, 0xffe146fc, 0x98e87ba5, 0x9b749e14, 0xb892f909, 0xd5074edd,
+	0x7e225e77, 0x8883947e, 0xa225845c, 0x51ea8e0c, 0x53852429, 0x681c3ea8,
+	0x5a6585b1, 0xf3a7cfa6, 0x6231fd86, 0x2835fd61, 0x40fb2367, 0x7ca1aeb8,
+	0x4d821839, 0xaef587f4, 0xafd0bfd8, 0x84af2f42, 0x67ed36b8, 0x604a63e5,
+	0x9d67ed05, 0x357498d2, 0xcedcf32d, 0xba412062, 0x5003f4e2, 0x2e27cd57,
+	0xf6b1210a, 0xf1169f5c, 0x0ed0a847, 0x78c64af8, 0xd38e9180, 0xf4f2f822,
+	0xf3fce952, 0x09bcb60f, 0x892b8b6e, 0x6fdae78d, 0xca593b41, 0x4fb0eaed,
+	0x17d0014d, 0xee237c24, 0x112d799b, 0x17d3a0fd, 0xc4686ec3, 0xef61ae7c,
+	0x54a87833, 0x41fde1b9, 0x7d723c11, 0xa136fdc2, 0x8284863f, 0x2eaeb085,
+	0xb5ee1508, 0x3e0bf1d1, 0xa71b7262, 0x15d20e3f, 0xaf6f8f5b, 0x307c8c37,
+	0x90e54c20, 0x1d1fa5d3, 0x23a0f0bf, 0x52f68752, 0x7019ea95, 0xc99eb916,
+	0x1b30389c, 0x7f38bdf7, 0xe42af119, 0x05e728d9, 0x9af6bbf7, 0x0c885c19,
+	0x8a4438e3, 0x09ece1c2, 0x392de87a, 0x1157f415, 0xeb8503c5, 0x696f2e73,
+	0x60e30217, 0x6e864ffe, 0x65df224a, 0x7e99b25d, 0xc62f631d, 0x20d99ec9,
+	0xd85ed477, 0x7b150f01, 0x8f4c682e, 0x11de9185, 0x0172a7be, 0x10b909df,
+	0x6fe46efc, 0xe82b0f21, 0x189ea657, 0x4760249f, 0xce082965, 0xeb61d52e,
+	0xe157801c, 0x1a7f1337, 0x3a9663c6, 0x78eb38d8, 0x24d2ca5e, 0xd7fa7677,
+	0x25ef5fe8, 0xc4c40913, 0x8b4abc39, 0x7bd17a06, 0x8c812349, 0x8d9f80ef,
+	0xad59f871, 0xcfc78a76, 0xebf3c86c, 0xfe02bff4, 0x87fe1180, 0x5ff8519c,
+	0x421456b7, 0xe70efe9d, 0xef9bd2c4, 0x15f90a13, 0xb6b43af2, 0x9cb52f72,
+	0x2719f271, 0x0c80a197, 0x4f978a78, 0xccf41932, 0x23f33088, 0x817a728c,
+	0xf6f9a3cf, 0xefc0de96, 0x0ea7b2f9, 0xf7def7e0, 0x2fe83745, 0xc63249df,
+	0x6bf32053, 0x2bd81765, 0xb3a190d2, 0xda76f107, 0xf5c4e8cf, 0x9f9f057a,
+	0xd4f94b54, 0x9e05fb0e, 0x0affd07d, 0x41ff5b07, 0xc7f41e40, 0x7bf3c545,
+	0x852b0b90, 0xa7e2192d, 0x1879c27d, 0x45c832a7, 0x3560725b, 0xce9079f0,
+	0x9f8cec17, 0xa5ab99da, 0x5879e1b6, 0x52f0634d, 0xa23f07e9, 0x06c260eb,
+	0x8fb820ab, 0xe59f35da, 0x602d74af, 0x729fd6f9, 0x50c27e7d, 0x3eca545e,
+	0x0aa97986, 0xbe0b9bf1, 0x2a00dc3d, 0xaa4e5fc4, 0x1f008fee, 0x9c6eb196,
+	0xb883ae47, 0x72276a24, 0xa2e4a095, 0x722fa470, 0x9fc6bda2, 0xbfb49fb0,
+	0x22cef1dc, 0xd52e77a0, 0xcaf71089, 0x91247f05, 0xc22aa85c, 0x3ef9adfd,
+	0x5f07e710, 0x407f7ce1, 0x18cdf3a3, 0xbcb1a2df, 0xd4ce2a85, 0xdbf0092e,
+	0x4e3b7294, 0xc4a2e9b6, 0x8634fcfc, 0x08bae367, 0x52f30608, 0x8d41fd78,
+	0xb8e94928, 0xbec3f16e, 0xc164dfb2, 0xb8958b75, 0xbae7a082, 0xbe1e2c68,
+	0x86ad724a, 0x16952ffa, 0x78f9e40b, 0xfefd6cad, 0x2e49c8e8, 0xbf62be85,
+	0xd61f35a1, 0x3bcf949e, 0xa6fdf469, 0x57bdad91, 0xba5a0ce2, 0xcb737e31,
+	0x27c12f30, 0xb56e8215, 0x496e8eaa, 0xa1888f10, 0xad148adb, 0x4dc5f80f,
+	0xd60bb252, 0x240b4d29, 0xe164a706, 0xff9053f3, 0x1e127122, 0xf2d42857,
+	0x6f423f10, 0x86acf9d3, 0xfb0dc052, 0xf92fd0fa, 0x89b0359d, 0x8f1bfabe,
+	0x81c433a5, 0xed8e67b3, 0xaf67710c, 0x491ba5bd, 0xf3a60710, 0xcbf3001b,
+	0x40d32b56, 0x79df65cf, 0x30f40edc, 0xd7f62dc2, 0x4bfb07f0, 0x586c656f,
+	0x7125fe83, 0xe7cf7ab3, 0x05094eb0, 0x60e39978, 0x2a49cff0, 0xb7d58acb,
+	0x132cfa4f, 0xef5a2515, 0xec4bc914, 0xffe7cba3, 0x89daefb5, 0xbd7be9a3,
+	0x59372635, 0x9fe3f5a4, 0xd6e671d1, 0x3f105d58, 0x8cf65478, 0x773fb803,
+	0x70975f3b, 0x01295d1d, 0x959f8b1c, 0xf079d3c8, 0x43fb611d, 0x1f943f1c,
+	0xd98ecf3a, 0x2ef3a037, 0xd3c506d3, 0xb52559b5, 0x7212e710, 0x2ca7c999,
+	0x4ba082f4, 0xea9bd026, 0xe5465205, 0x8dce2634, 0xdc6166af, 0xe8527f0a,
+	0xbf02fff7, 0x31437ed1, 0x1ba09dca, 0x48f4c3e9, 0x4c75c812, 0xfeb9aabc,
+	0x14fcabbd, 0x9fffbf8e, 0xb68429f3, 0x496943ff, 0xf53e3901, 0xdc535fc0,
+	0x00f60f90, 0x977acf5b, 0xeebce2a4, 0x8ff77ee2, 0xb46a9e38, 0xd79872b2,
+	0x637fbedd, 0x6fa2379c, 0x8efb67ef, 0x8ffaa80b, 0x4b6a2fdd, 0x1479c32f,
+	0xfd7793a2, 0x8562ecbe, 0x7e30eb9d, 0xeb6349bf, 0x1bd505e6, 0xc0275cb4,
+	0x806ff9e3, 0x5f5f14e7, 0xf161aa9c, 0xacdf011e, 0xf81a01ea, 0x9908ffdc,
+	0x1fed9849, 0xb95f4ca9, 0xb7c5126d, 0x08d1e387, 0x179e01f7, 0x3257e739,
+	0xe933c1fa, 0xb953f758, 0x601684f7, 0xff08c71c, 0x52bcba9c, 0xf3987fc8,
+	0x29eb84b0, 0xaa01ff78, 0x3ff73d3f, 0x68e80e74, 0x9f319f9c, 0x171f18c4,
+	0x43c8b37e, 0x983c4d9b, 0x27171297, 0xfb1cfd0a, 0x9e8bc637, 0x797af8d5,
+	0xfb021930, 0xbef96725, 0x621b9c3a, 0xcb04ebfb, 0x14925072, 0x1e968b9f,
+	0xb926052a, 0xc61b72a3, 0x33f8c5eb, 0xf833f8f8, 0xaae15438, 0xe633c7c0,
+	0xf36217ab, 0xdc68b6dc, 0xc7421e6f, 0x89e3a393, 0x39731671, 0x7a0be314,
+	0xe72abe0a, 0x7f515d74, 0xd3235bb2, 0x0c498703, 0x9ff88c7d, 0xca2b3bf7,
+	0xa1cf0447, 0xfd039c5f, 0x51736738, 0xec271769, 0xc28e545f, 0xcfe269fc,
+	0xe16af961, 0x1bb7581c, 0x36a4def1, 0xd8257376, 0xfee2689f, 0xb9717b5c,
+	0x03c89a1d, 0x94f2c2a8, 0x67e07552, 0x191b87c0, 0xb9de2c1f, 0xb79075e5,
+	0xb2bfadf2, 0x569dbc83, 0xbc60b2aa, 0xb41e9b45, 0x29c57ec3, 0xbd076fc5,
+	0xe4c03a78, 0x8aefc8ce, 0x305bec59, 0xe0f6157e, 0x9f3fcbcc, 0xc5f9fc00,
+	0x7a01d526, 0x1cd9bbde, 0x547c5336, 0xd0fc30d4, 0x7f5651fa, 0xade2f108,
+	0xfe5dc3d5, 0x9cb2afe2, 0xe7969f6c, 0x9d7185f3, 0xd3f1b15f, 0x2f0f9052,
+	0x1d7dc169, 0xfe3077e3, 0x97dc74a5, 0xc6a5a99b, 0x11be0bb7, 0x8d6f05eb,
+	0x3c469be0, 0x7f7c665f, 0x172fd73e, 0xfe72bf05, 0x0120f811, 0x9de457a6,
+	0xd28ff3eb, 0x7f8dd9fa, 0x62fd4bb2, 0xbfcb09f5, 0xf3de0d68, 0xdb672eec,
+	0xe71360fe, 0x0177e8c3, 0xcb59cefc, 0xac4230f3, 0xed86a45c, 0xb4591710,
+	0xecf5c541, 0xbfcf2da2, 0xbdd834f0, 0xabaecdf7, 0xb70bff69, 0x38777fec,
+	0xa6dc2fad, 0xd3678e66, 0x763149b0, 0x5a7c0bda, 0xf943a510, 0xb39afdf6,
+	0x40fe7b3f, 0x0b1e947a, 0x48d1edb7, 0x947c78ff, 0x8c68f704, 0x09740dff,
+	0x7d852d1e, 0xbfe7796e, 0xe9fba035, 0xbc8112dd, 0x6e7c38cc, 0xd2fb8fd8,
+	0x23a42780, 0x6e8453a7, 0x94bdf786, 0x39e7682e, 0xefed8c9d, 0x3fe40e69,
+	0x5dce0fe8, 0x757e7e51, 0xfccfc517, 0x982ecc0f, 0xef57fcff, 0x4e3cc3b3,
+	0xaf8c952a, 0x05983fd7, 0xeabe6476, 0xc96072cf, 0xe67fcf9e, 0xf36fc847,
+	0x8b28fd0e, 0x99d2fd30, 0x9dfcd1c5, 0xd6737e61, 0x9bf386dd, 0x77c83c4b,
+	0x65cbd7f3, 0xa8babf10, 0x12dbb190, 0xb94fc5c8, 0xce7c9c5c, 0x27f8fb8c,
+	0xb8531ec1, 0xccc8effc, 0xa2f721cf, 0x9f11fd7d, 0x58fcb50b, 0x18e3c8bf,
+	0x1e22cd13, 0x2c72b54c, 0xa87c6ebf, 0xe673e801, 0xc436772a, 0xf144bd01,
+	0x57487c73, 0xa1f2bf68, 0x3f5cd931, 0x7f189539, 0xc39e04b7, 0x3baa0dfb,
+	0x1df75f29, 0x1d1792b9, 0x9f1f297f, 0xfb0c9dc2, 0x12b327c6, 0x305f87c7,
+	0xd611624b, 0xa303f3a0, 0x7de4cdf2, 0xbe4cc3e3, 0x2607bc85, 0x0bf6858e,
+	0xc0fc9987, 0x53c7e077, 0xd8d9c2b8, 0x848a67be, 0xc3da80fa, 0x51bd7244,
+	0x6fad72e5, 0xe1427c17, 0x897fc056, 0xfb610bf0, 0x7e7927bc, 0xe927dcfb,
+	0xc1763177, 0xe3d9e30b, 0x3b7213fb, 0x6617f84f, 0x7593fcbd, 0x5bc6129d,
+	0x22c47e8c, 0xd3c2f035, 0x3e54af20, 0xda2edf41, 0xbcc196a2, 0xfbdeab3f,
+	0x9ca90fee, 0x7214167c, 0x469fd7aa, 0xdaff9d39, 0xa34fe47e, 0xdc167e9c,
+	0x4e50959f, 0x73c7edb7, 0xfa72f1b7, 0x053fab3f, 0xf01bd6fe, 0x5b61f427,
+	0xc3ea3478, 0x4fc410e1, 0xb7090fa0, 0x13fe46c3, 0xadf8277c, 0xddc595fc,
+	0xf844ef41, 0x845df052, 0xc5df052f, 0xf0ead1e5, 0x083f3717, 0x589f01a2,
+	0x75828ed6, 0xf6c5c586, 0xcb39ae13, 0x819693e2, 0x67d6fd6c, 0x43f582ad,
+	0xb020c726, 0xec4613fb, 0xa427ecfc, 0xbef9fe6d, 0x963e9dcb, 0x65075fa3,
+	0x895f042f, 0x1090e3a5, 0x16fbd383, 0x7d40f409, 0xf76396e0, 0x930b640b,
+	0xbd1f9df7, 0x2c780856, 0xca97ee6a, 0xae1babf8, 0xb5fc445b, 0x054cefd5,
+	0xea1ada7d, 0xad4f7989, 0x4494e507, 0xe25fa5e7, 0xfb857df4, 0x491bcd5b,
+	0xadf94c95, 0xdfa0b499, 0xc168dfe0, 0xbe38b67c, 0xe67f3330, 0xe803e2bb,
+	0x2978f80f, 0x5134b1a9, 0xf9f09a3c, 0x5ff3784e, 0x0baf1f68, 0xbf81306d,
+	0x9346dd7e, 0x24be4fce, 0x1480baf8, 0x784f5f00, 0x20652933, 0x2ecaf3ee,
+	0x0dee5738, 0x093cd39d, 0xf38c95f2, 0xdfc1fe87, 0xe0716fea, 0xc1a86fb8,
+	0xdbdcf07b, 0xeb0216e7, 0x6aad3a85, 0x3b382261, 0xeb383125, 0x7cba338e,
+	0x6a29785f, 0x618b85b6, 0x61237a7f, 0x3a38f671, 0xde0eb613, 0xaa7f704c,
+	0x7ab32435, 0xa347db35, 0x2462fca0, 0xf699777c, 0xd79f231a, 0x057b6ed4,
+	0x8455e7ce, 0xb6898724, 0x1a8d2857, 0x36b78bf0, 0x363d064d, 0xcbce29c0,
+	0x894ffa3b, 0x16febfe1, 0x77b9e705, 0x00331bf8, 0x6acfe72f, 0xddd5f604,
+	0x60ec5afb, 0x0a87ceef, 0x9adadf91, 0x191f76d7, 0xfee9eb05, 0x78a77f00,
+	0x4eabf63a, 0x9447d5d6, 0x5c00ba78, 0x4b8dcc13, 0xfef0095d, 0xf7d9858a,
+	0xdb3e4bef, 0xbf2c3b2b, 0x7afc8587, 0x7c41bf70, 0x378700ff, 0x13b37e46,
+	0x6592efc9, 0xf7986cce, 0xd99efa92, 0x7f25cf80, 0xe2253b50, 0x963e2dab,
+	0xee15b4df, 0xadf90017, 0x7d7bf2a1, 0x50d6fc8c, 0xff8bcdf9, 0xfbe5ae6e,
+	0x202ddf95, 0xfbe009bf, 0x90af9f09, 0x51d0fcdf, 0x1afcdf94, 0x4acef9fd,
+	0x2bf27b4b, 0xf2c29a75, 0x04d65c60, 0x00f01afa, 0x795bb2f9, 0x038c387e,
+	0x37c2fc72, 0x24bce394, 0x4a7fef06, 0x20ec978e, 0x4f128798, 0x4bc7266f,
+	0x4b1ca8ea, 0xc7267740, 0xcb09ea4b, 0x6fc83407, 0x65f9389f, 0xd98457df,
+	0xbcdadef7, 0xc741dcdf, 0x2eab702b, 0x20bf7d0b, 0x6283cf2a, 0xb685f9e5,
+	0xc760f9e4, 0x970779e4, 0x3da72a67, 0xeae8095c, 0x63f3e90a, 0xb27618de,
+	0xe5f03b7a, 0x275c659f, 0x0f79e21d, 0x7276197f, 0x3fc5f020, 0xff8a1e00,
+	0x9a6f9c3d, 0xbfec7e7c, 0xe853f874, 0xabec9547, 0x03ec35fa, 0x12a8de2e,
+	0x351587d0, 0x328fd147, 0x5f85bfae, 0x0b875c65, 0x7e5ed23d, 0x776e61cc,
+	0x088dca26, 0xc65b792f, 0xf38c3ffb, 0xc1c5ef17, 0x5deab079, 0xc042afb8,
+	0x407c5ed9, 0xb83da3cc, 0x15debcc4, 0x148032f2, 0xe286dc42, 0xffc5e511,
+	0xb5cf1947, 0xad5d1595, 0x23e51fbb, 0xaac2cf8e, 0x870812a2, 0x37e7e44f,
+	0xf7bdaee7, 0x18af9c00, 0x0059c474, 0xe188f4ee, 0xa7ac0f66, 0x416a4a5e,
+	0x49741679, 0xb974624d, 0xa379fbc4, 0xf031654b, 0x7256b79b, 0x86f1dd80,
+	0xc1f785ff, 0x8bea7dec, 0x05a3e052, 0x8cc392f5, 0x834e787c, 0xb5295e75,
+	0x854fc0bf, 0x7eb873a1, 0x245f505b, 0xea0bf7c3, 0xa3c0bc24, 0x9474fc3d,
+	0x0edc296e, 0x6df8016f, 0xb30b6700, 0xd13e1c6b, 0x3afc55ee, 0xbe9e4bf8,
+	0xc1972ab4, 0x571704ed, 0x02283bc0, 0xe7194cef, 0x057841d5, 0x3783dbde,
+	0xc74f4935, 0x679e47b7, 0xb4a110fb, 0xbe6bad0e, 0x97887ee9, 0x4cfae034,
+	0x90cc8bec, 0xac0b663d, 0xf3e97b6f, 0xc1690995, 0xa1efebbc, 0xf5d7ff7f,
+	0x0037d91f, 0xece41dff, 0xcbae2a4f, 0xb338a497, 0x1df1dce1, 0xb75d2af0,
+	0xfb336d34, 0xff313de1, 0xabb73e13, 0xc3710297, 0xf19e58b1, 0xf57ab995,
+	0x4f6fb18a, 0xbe43b79d, 0x821f7357, 0x294d2179, 0xda274e2c, 0x0b3249dd,
+	0x587e73fa, 0xf57c34a4, 0x1b49f821, 0x5c03061b, 0xbcf32e27, 0xb1e6cb92,
+	0xd3ddd689, 0xe7109db9, 0x3802ef28, 0x2fb2249d, 0xe59f377a, 0x927e705d,
+	0xad6667ae, 0xf8852b4b, 0xa0bcf1b2, 0x83403827, 0xde31a438, 0x11b88738,
+	0xd0f9718f, 0xbf1611e4, 0xeedff034, 0x273b0724, 0x706a54aa, 0x3ac4f65e,
+	0x8829f889, 0x8f88db0b, 0xf1e24c6f, 0x1fc20ec8, 0x46b1ce77, 0x11b978c2,
+	0x971b79ee, 0x6042eaad, 0xf15e6cbe, 0xb579b3f0, 0x78d20f35, 0x8bb58923,
+	0xbcec420b, 0x3223c576, 0xaf3c83b3, 0xe02bc044, 0x2a5dc233, 0xa7ff422f,
+	0xfa8cad79, 0x607f2ef7, 0xba479c15, 0xfb84a56b, 0x54f9c62e, 0x009cced4,
+	0xdf43c0fa, 0xf3693887, 0x69ed0447, 0x00296af3, 0xd73f8c78, 0x4bfb66e9,
+	0x427bbc7c, 0x7e7351f2, 0x1ea2f4db, 0x3665d20e, 0x849acba5, 0xd5c8a3fa,
+	0x168e462b, 0x7ab9c604, 0xe20bc6e9, 0xe6e17eba, 0xbfe7ba89, 0x74f1e249,
+	0x4f6fbf1e, 0x6265489e, 0x47e422bd, 0xaf9093c8, 0x817ce0dd, 0x0702519e,
+	0x4f00f372, 0x1a7ea18f, 0x4c5e7b27, 0x8f217962, 0x54afa74e, 0xe79e753c,
+	0x0cbc7085, 0x2399f465, 0x6de7c78f, 0x078a31e9, 0xb6c0fedd, 0xe27c7d24,
+	0xcc47cebb, 0x9881c843, 0x8c0e519f, 0x21e91bdf, 0xba46c0e2, 0x6721e918,
+	0x7e40fa9b, 0x5d939121, 0xb9cb77e0, 0xe8d7caa6, 0x5a7772b8, 0xcbf703f6,
+	0x65a4cf25, 0x7928c955, 0xa8572d16, 0x30ad2f4c, 0x1898b8b1, 0x970e5dff,
+	0xcb923328, 0x3ec1270e, 0x830ddbe3, 0x0c2f2cad, 0x5c7145f7, 0x9215a78a,
+	0xe83f809b, 0xfed04fdf, 0xbd926417, 0x712b7f80, 0xd6fdc50f, 0x9e3e7124,
+	0x5fd2e1e5, 0x582e942e, 0x35a70fc7, 0x83473fff, 0x8ab2ecbf, 0x4f5625fa,
+	0x153de135, 0xab25ffc4, 0xb93af367, 0xa2b9eacc, 0xfc29d69d, 0x7bfceeae,
+	0xef566ff2, 0xcdffda2b, 0xb45ebfde, 0x21dac57f, 0x7fb69481, 0xa357f0f5,
+	0xae7f30fd, 0xbbb548c7, 0x62bbf02f, 0xe529eb62, 0xdcf30d4d, 0x15fe3e44,
+	0x96ab97ce, 0x8562bf15, 0xb762a23c, 0x67c4f213, 0xdf3b73f0, 0x90f414b6,
+	0xf7bc26ad, 0x9ba04a01, 0xc0e2cc96, 0xc6a49e82, 0xcbce17dd, 0x2fbba091,
+	0x246635e4, 0xc192927b, 0x61ab7abf, 0xcd47c814, 0x060c1f3b, 0x0ee80bff,
+	0x05be14fd, 0xfa50da77, 0x436abced, 0x46705e6c, 0x193d9172, 0x6cdd1af2,
+	0x3a724adf, 0x14051dc8, 0xadf6b80e, 0xf4d68427, 0x8035b2fb, 0xfb0f6a3a,
+	0xc1659527, 0x986ba478, 0x9cb8fbc5, 0x2e9ecbe4, 0xb8b9bcf0, 0xe061d1f1,
+	0x7ed89997, 0xef48693b, 0xe5a13f90, 0x74798fdf, 0x47b5b9e9, 0x3e3ee166,
+	0xdc90cfdb, 0x6e838c6e, 0x24179c8d, 0x25603f98, 0x6fdc2ec9, 0x7f1c9177,
+	0x307fcf16, 0x1cccbe31, 0xccffb31c, 0x1fbf45e7, 0x870fe70d, 0x375e5958,
+	0x710ca79d, 0x89c401b8, 0x33e42165, 0xf55604f4, 0x09177d60, 0xed19f91e,
+	0x7624f1ff, 0x336585ef, 0x4c87e5d1, 0xd876664f, 0x21f9656f, 0x12706ac9,
+	0x73e4ede8, 0xbb7a01c4, 0xe0a6bd79, 0xf1c50ef9, 0x3c5fcc03, 0xef002260,
+	0x01c06a8b, 0xd72d7479, 0x48e7ce2a, 0x4e66bfb4, 0xe3efd17f, 0xc85af39a,
+	0xcad078cb, 0xb1162fea, 0xd396833c, 0x5af2005a, 0x952d73d3, 0x93b3f9b1,
+	0xf13f9992, 0x78f14a62, 0x3eb043a0, 0x43f8c099, 0xa2065376, 0xbaa579f4,
+	0x63f80920, 0xe288e02b, 0xb8ecdd21, 0x3f4bfa17, 0xc3cde14f, 0xafc46ee7,
+	0x4a901ce2, 0x66fc3bf1, 0x47ee139b, 0x7829317a, 0xcb41c833, 0x839e1316,
+	0x26b9aade, 0xab6b7d42, 0x257901d3, 0x9305e62f, 0x4c5b3e71, 0xc9ad9f38,
+	0xa12dd72d, 0xa3e9af70, 0x14f8058c, 0xe92c512b, 0xcec54f20, 0x99f3e97b,
+	0xacaf4cc9, 0x7a793134, 0xf589d93c, 0x7e3a25c7, 0x8457bc01, 0x4fce24f2,
+	0x75b0b734, 0x0d29a76b, 0xf275f032, 0x25de7144, 0x3c7f832c, 0xbda08cb2,
+	0x0353691c, 0xb9538c4e, 0xfd29336d, 0x07661147, 0x08b47d46, 0x0528c3f3,
+	0xfda08f0e, 0x04a9b5ed, 0x1e7845e6, 0x4f9c8de8, 0xb28d206e, 0x70dfb685,
+	0x74c2b889, 0xe92f1bbc, 0x2ddcbff3, 0x8d14d18e, 0xb62488a4, 0x78f863c7,
+	0x00adc4c5, 0xf45894ff, 0x7b018def, 0xe318e14c, 0x49339c30, 0x1bb476e6,
+	0x2e7b16e9, 0x8dede447, 0x853dfb91, 0xcd58c0fe, 0x66d87805, 0xfb07bf73,
+	0x9506fe7f, 0x46175a24, 0x6a3ed7d6, 0x61537a39, 0x798c94b9, 0x6d1f947b,
+	0xadd707c0, 0xfdc3f2ad, 0x62f5d787, 0x81df03fd, 0x7d66b95e, 0x87d80666,
+	0x511f4e02, 0x7e033e0a, 0x7e32a472, 0x951f5c72, 0x79c9f5c7, 0xbe429fc8,
+	0x9fa0cf80, 0x46606a5d, 0xe50738a2, 0x9fc0f5cf, 0xfd046cdb, 0xb6779fdb,
+	0x12b9034c, 0xfce31ce3, 0x2d572c6f, 0x8547b25b, 0x76645f79, 0x4b977466,
+	0x3c8c635a, 0xc1852bfe, 0x4d5675eb, 0x1bf1009e, 0x2dfb8a9b, 0x162e3c59,
+	0x639d1fcb, 0x59f043f8, 0x51e58d93, 0xa9297e08, 0xbbae2bfb, 0xe2122e4a,
+	0x3f1123c8, 0x28cbf04f, 0x61f3187f, 0xd7e61147, 0xe8a5f919, 0x7d015382,
+	0xb4bf1ed4, 0x089027c5, 0xc8d27605, 0xb992072f, 0xbf26d29f, 0x5e03b65c,
+	0x6497d696, 0xa4fc5d7f, 0x5b47f396, 0x4cacc3ed, 0x5f9f19ef, 0xa8f66148,
+	0x41faab4e, 0x5b85e83e, 0x7b902e4c, 0x70b90dca, 0x325d513f, 0x95288efb,
+	0x4dbea13d, 0x5b667a61, 0x667a8cc8, 0xc1e3cd9b, 0x161cd07d, 0xa78a241f,
+	0x5c9af211, 0x35ce2c5d, 0x0cfcfc31, 0x1aa0bfe7, 0xe212761e, 0x74607f0b,
+	0x014b7a1f, 0x9d25189c, 0x1fb93367, 0x00db4a78, 0x6c3dcbf7, 0x7c0d3a2d,
+	0x04b20c95, 0x35c74f79, 0x38ad77dc, 0xdbfea1ae, 0x2950fbb9, 0x9e0afdd8,
+	0xd76431e7, 0xd062d2a1, 0x37a65627, 0x29fd71cf, 0x4e60c3ea, 0xf70d0fa0,
+	0x4c230d15, 0x53ddc54f, 0x1f937c04, 0x77bb8f6d, 0x0635d1a8, 0xef771ac8,
+	0x98f711d0, 0x7bdc2714, 0x797bc1e3, 0xe1efeee2, 0x67bc60fc, 0xc7927fa6,
+	0xf83f0562, 0x36dd8695, 0xb3276be0, 0xb2ad41c7, 0xa92bcb2f, 0x23fc09ec,
+	0x147f0ce4, 0xf7bfc15f, 0x7cf62e79, 0xbe7c2dc3, 0xb614b88f, 0x61a8dd3f,
+	0x082fe81e, 0xedc82efb, 0x1e6b6f52, 0x5f4b0bd3, 0x5d7c019d, 0xf98ddf38,
+	0xc8a58fb6, 0x2497fcf8, 0x7409e96c, 0xaf182972, 0x92ceaa0a, 0x3df45293,
+	0x81f594a2, 0x4eaed4b8, 0xfe3085c9, 0xb5cf56c3, 0x5a513f00, 0x4377c20e,
+	0x82141786, 0x580db65e, 0x4a760199, 0x1ae14dc0, 0x8516c9bb, 0x6595c043,
+	0xb0f87c65, 0xd6b5e675, 0xd03bec4e, 0x4f47b53b, 0x417be058, 0x39fd60ff,
+	0x3e18f746, 0xf947ba00, 0xea1bba40, 0xdc050f02, 0xbff2dede, 0x25c60f87,
+	0x87e103e2, 0x76ecbf51, 0x1ff78fc0, 0x73b37b97, 0x6fcf2c1d, 0x24778c88,
+	0x9b25f246, 0xb05d8a72, 0xf9d1597f, 0x3e525887, 0xa067ff3a, 0xff766ee7,
+	0x04e7f0b0, 0x45cbe4df, 0x234bafd8, 0x2e83de78, 0xe21fb0d8, 0x823cc349,
+	0x63060eed, 0x744fdff3, 0x59282ea3, 0x72674d60, 0x1bb5efc9, 0xefd09610,
+	0xc715401c, 0x75d7e81d, 0xe2e8bd45, 0x242725f8, 0x03c5ce09, 0x05c95f2b,
+	0x019c33ff, 0x8f300fe3, 0xed3ec30f, 0xb951d66b, 0xeb029f30, 0x8fc09ec8,
+	0xfa59f7af, 0xaf693027, 0x7dbf9977, 0x4fea33a3, 0x97b8fd09, 0xbfc62b77,
+	0xcbf4cadf, 0xcfb66e8d, 0x163d7a0e, 0x801f610e, 0x1487008c, 0x62ba6a9e,
+	0x6a1dbd45, 0x11be78a4, 0xd8f4a0e0, 0xf01a9123, 0xee4c8a17, 0xfbf012cc,
+	0x8015853a, 0x5ffeb9b3, 0x340e20dd, 0x23fbf43e, 0x9178ef00, 0xe7d2eb1f,
+	0x6cfd1ac7, 0xf588ffe1, 0x148ff086, 0x47e1088a, 0x7874fb41, 0xe0e9f14e,
+	0x0c8de0f7, 0xfefe305b, 0x5af3ab78, 0x85779737, 0x98bc63ce, 0x572be5c7,
+	0x4cba5246, 0x5a3f3120, 0x0a4a05bf, 0x2483f8ff, 0x82f18c9c, 0xe7f80516,
+	0x8e449716, 0xadd9f0c5, 0xf875f543, 0xf68aca8d, 0x46fbded7, 0x2242323b,
+	0xff1f9b5f, 0x3e83cb4d, 0x47f62fb6, 0x18f7db1f, 0x2bf643de, 0xed0db0e0,
+	0x1c3bf6cc, 0xb5070f0c, 0x73ed76c4, 0x4dc3ef2e, 0xef10f98d, 0x4bebb3dd,
+	0xee6fa3b4, 0xf05df2fa, 0x569d871b, 0x8767c408, 0x8a981e9f, 0x39a1d271,
+	0xf442be78, 0xc6b4fdeb, 0xe473d84a, 0xdfff711f, 0x5b8f0a01, 0x582cf803,
+	0xf1f671be, 0x0bbfa027, 0xdbc615c7, 0xcd52fc7c, 0xebe20bd3, 0xc81efca5,
+	0xa7f31203, 0x7e01fb44, 0xd6bbdaef, 0x31105e48, 0x3497d2c7, 0xe820bfa0,
+	0xb1a9cb78, 0x78fc95a7, 0x98efc0ba, 0x39afbfe0, 0x295edccf, 0x714e97f6,
+	0xefbd91bc, 0x4ddb3ee3, 0xe02af735, 0x2394dfa7, 0xb8cc7713, 0xe2569ce3,
+	0xfc6b898e, 0x63f06ac3, 0xc4f4bef6, 0xbf8b8804, 0x2f91e325, 0x9fdb686f,
+	0x20dea1f2, 0x3377dc27, 0xd0a6f5e2, 0xfdf00abd, 0x97931b38, 0x7e1fc6f6,
+	0xeb8fe51d, 0xf984de81, 0x9425837f, 0xdf5d231f, 0x8cdbf5dc, 0xf8f17ec6,
+	0x02de8b6e, 0x6f0e82ef, 0x054e2d9c, 0x0747b47c, 0xa6ee9f23, 0x79d3f3f3,
+	0xfcfce985, 0xd37b4fd4, 0x3efac0e7, 0x05d60e5f, 0x9feb7a7f, 0x4f3f00cf,
+	0x0c98e5f0, 0xeac327c6, 0xb620c89e, 0xf88c9d28, 0x74636eea, 0xc0e9fc9b,
+	0x7e3e5414, 0x6ce700d2, 0x536c38ca, 0x57ef1389, 0xbdfcfc24, 0x954dbed1,
+	0xea0318ef, 0x8cdf68d2, 0x738a6d76, 0xd7c3de26, 0x6384bdde, 0x2052aadf,
+	0x5954b3dd, 0x80fbb0aa, 0xe02e9f99, 0x6ff3d231, 0x09b6eca4, 0x19f0db52,
+	0x67071de9, 0x3d353caf, 0x3fc2bc01, 0x4e10f717, 0x78f686f9, 0xd7cf4db8,
+	0x7e8d4fe7, 0x5b6ad41b, 0x8ba46472, 0x83e85db1, 0x00a01852, 0x23f83f4b,
+	0x48596ce3, 0x1a4bd2bf, 0xb579c371, 0x93e449ae, 0xc642bb18, 0xba7e4777,
+	0xb59ec1cf, 0x97154fc5, 0x1193c44c, 0xf6b52be5, 0x55f02b31, 0x8565529d,
+	0x6c1babfd, 0x52a45713, 0xb4fe8c3c, 0x5bee0a78, 0x84089a35, 0xce84c6c6,
+	0x813885ed, 0x1c5d5f38, 0x7c0253da, 0xc0694fb8, 0x05f35375, 0xf8c61bea,
+	0x15a1b599, 0x6ac3f056, 0x0b8c6533, 0xba23481c, 0x6dc7a073, 0x3daaefcc,
+	0x9dc13ade, 0x0a73bda0, 0xdc4c9f6b, 0x6768f999, 0x51be3053, 0xf384daba,
+	0xf6b92d1f, 0x321f8267, 0xd9afdd9d, 0x7f643f5c, 0xe9b78526, 0x1fd434ce,
+	0xba1f6a63, 0x110941d7, 0x3e2177fc, 0xb929996f, 0x95efa39e, 0x2f585d3a,
+	0xeb879d9d, 0x655f21b8, 0xed0f1e4c, 0xd75c54e3, 0x2917918c, 0x3cfa6f10,
+	0xf5802a47, 0x71a92ee1, 0xe4fc1999, 0xe21c6c89, 0x7ac067f0, 0x19f19df3,
+	0xcf20d603, 0x05d5fa88, 0x75ea186b, 0x3ba5b3eb, 0xbf81efce, 0x36f4b04f,
+	0x8e51cf09, 0x7f12e471, 0xae94ff0c, 0xfea4f4a6, 0x7650a4d7, 0xe4f1f031,
+	0xc40932b1, 0x656df1f0, 0x5d881256, 0xb1f3f20f, 0xde2af52d, 0xf8e42037,
+	0x53bfce66, 0xdee865a7, 0x0ede6fa7, 0x886267f6, 0xfcb841c7, 0xd9cb80fc,
+	0x3bfd7ccd, 0x1bab93dd, 0xbf308cb7, 0x65a4cfe6, 0x671d2a74, 0x960f4a76,
+	0x70d2d51f, 0xe5f07318, 0xcf4083ae, 0x618fa0fd, 0x5a2f2b3d, 0x2390fc89,
+	0xb77f00b9, 0xb7edf4b8, 0xe3d7d50e, 0xe8353fee, 0x74e0db81, 0x0faf1c94,
+	0x56dfdc84, 0xc710ab2f, 0xe288adb9, 0xbd61fd17, 0x5eaede7f, 0x107cecb8,
+	0x5bef86dc, 0xb2a4ec2f, 0xefca9436, 0xae1fa16e, 0x7e815722, 0x3f572318,
+	0x4cd2fc01, 0x01399eed, 0x21fb08bf, 0x54a5f3d6, 0xc83f815e, 0xab8c4cd9,
+	0xe544b98e, 0x9eba2336, 0xea0be88f, 0xf812b84f, 0x7ad81f39, 0x221df6ea,
+	0xd4fe83fa, 0x9fdddfc1, 0xb550a318, 0x75700052, 0xd442fe25, 0x2c571857,
+	0x10450385, 0x8739cbdf, 0x42af1b71, 0x5768a7d7, 0x8c7a679f, 0xd2ae0bfe,
+	0xf46ffc15, 0x789c4433, 0x7ff08a52, 0x89117cf4, 0x31514e17, 0xa1f8238e,
+	0xf208f014, 0x38c6453d, 0x21fc7b7e, 0xe29fffc6, 0xe3c2920b, 0xaf48bd7f,
+	0xe083f08e, 0x4617c103, 0x4347f1fb, 0xfb2da75c, 0xf5f29691, 0x1f7f63f6,
+	0x7b74f515, 0x6764887e, 0x7363d97a, 0xb5f37960, 0x0fc0d9b7, 0x83f5c499,
+	0xf89143b0, 0x17c65ad4, 0x29671bfb, 0x05e46fb0, 0xc51790bd, 0x0adf88bb,
+	0x4d58ac3e, 0x9e397e30, 0x7aa6a5eb, 0x8ef3cf16, 0x12900396, 0xec7a1252,
+	0xf9675609, 0xb79e822e, 0x5228be70, 0xf7f028f6, 0x49756e5f, 0xb913882e,
+	0x82703c79, 0x6078c25b, 0x7f823e9c, 0xc75bf6d0, 0xd758014a, 0xa093f519,
+	0x7fde64f7, 0x7df09f1c, 0xdc9f2357, 0xbee5c290, 0x9ffee93b, 0x03c60bf4,
+	0x4db53b5d, 0x7b7df157, 0xbe0d9d74, 0x5ed9c82d, 0xfe02bf96, 0xa6f5d035,
+	0x3bf60a4b, 0xe7ef4535, 0xf05b385c, 0xf20b5c4b, 0xaf798df7, 0x37b4f9c6,
+	0xc7e124fe, 0x8c5ebf73, 0xf2c2b37b, 0xa6768df9, 0xbfc2bb5d, 0x32abd233,
+	0x26bfb125, 0xde1c6bf4, 0x8b6dfeb9, 0x5fe9ed6e, 0x53feefa1, 0x2bf457fa,
+	0xcf1882ec, 0x5adb97a8, 0xbb3d09cf, 0xe309836a, 0xf5f03727, 0xae3c0df2,
+	0xc3f70cfb, 0xdede3c02, 0xf782d17c, 0xb6550653, 0x63dfe53f, 0xd6d038c3,
+	0xe97fccdd, 0xfbe2756d, 0xc6324a0f, 0x641c1e29, 0xd74c3d42, 0x3f308526,
+	0x3e9098a5, 0x85dacfb0, 0x2b56b3ed, 0xe3d432e1, 0x7e40852e, 0xf1959e54,
+	0xfa6f83c1, 0x0fff4067, 0x5a6deb1e, 0x70e37181, 0x4331f803, 0xa7dc140a,
+	0x313b334a, 0x6f54dcce, 0xe6aabcc6, 0x03daffba, 0xf9aa277f, 0xf2c17122,
+	0x39bd5f66, 0x7a044fb2, 0x96fc8c62, 0xcd11790c, 0x0ef3042d, 0x970b4e47,
+	0x06e97ac0, 0xd602b0d2, 0x7635c3e1, 0xa788dd8e, 0xc6de7ca3, 0xbec8847d,
+	0x1bcb3d41, 0x79e14929, 0x06fb7123, 0x6fb8250d, 0x15ae48a0, 0xd33ed124,
+	0x60df667f, 0xf29ce038, 0x7a0c5ff3, 0x7fb1b488, 0xfb6355b0, 0x54872d4c,
+	0x39327e02, 0x5e309995, 0x423eded3, 0xb764eedd, 0x82fb0ed3, 0xdbaf9fcc,
+	0xccedc5dd, 0xfcfdfb6f, 0x986296c6, 0xd36ad6c6, 0x47632b4a, 0xe246bb8c,
+	0x727c4671, 0xbfedc7bc, 0x61a35c84, 0x63c462dd, 0xe2194b71, 0x4fe29bf7,
+	0x13d6c5c8, 0xebe06ec9, 0xb8ba27ad, 0x3c63dc61, 0x3b32e70e, 0xbf7604fe,
+	0x326b78ee, 0x6e6ab3ee, 0x5a1bac27, 0x762e2cd2, 0xf3e3110f, 0x78dee760,
+	0x3c07fc6f, 0xb935b7ae, 0x337d8fe5, 0xd851f763, 0xc51349be, 0x0faa3eeb,
+	0x3cfd5f7f, 0xfe7e03e6, 0x13cc51f2, 0x4c86f85a, 0xadaf102c, 0x85e7e14e,
+	0x03f81a03, 0xee3235e5, 0x18b9534f, 0x5625393c, 0x7960acdf, 0x3dd95b34,
+	0x99afe59f, 0xf7e9e303, 0x927fafe5, 0x4df6f90f, 0xc00e5bcb, 0xbff01fdf,
+	0x33bfb12c, 0x87cc6fe7, 0x95cb9afb, 0x2eb271c4, 0x9d7fe676, 0x034eb3ad,
+	0xccda40ff, 0x79605efe, 0x8e6aaa70, 0xd9a5ef59, 0xd31ef155, 0xec492f0b,
+	0x5f0a097e, 0xe2bf7dec, 0x7d9c06ef, 0xd5f06249, 0xbbf801aa, 0xb79be583,
+	0xbe39c135, 0x5c6623f2, 0x9f3779bf, 0x4b33fbc3, 0xfcc16eb6, 0x4cad6f60,
+	0x09d619f6, 0x9b53f3ba, 0x45e430e5, 0xbfb0d455, 0x98eb4023, 0x4c115527,
+	0xbdb08e7c, 0xb70b3e73, 0x87eebfd6, 0x2feda83c, 0x9dbec1da, 0x0764d869,
+	0xbf7ed3bc, 0xdc62fcf6, 0xc3e0a979, 0xea5e7b5f, 0x5a72610c, 0x379c3762,
+	0x37c19cdb, 0x6383c223, 0x78a3faf3, 0x70b9fc64, 0xe067c5cb, 0xd7d9ef3c,
+	0xe0067b3e, 0xf9d5f45d, 0x0d7f6067, 0xbd60f512, 0x6ff97dea, 0x7c521e78,
+	0xdff72777, 0x117a5e9a, 0xbd3691cf, 0x6f41766f, 0xbb27f54d, 0xa6ca79c1,
+	0x82caff6d, 0xfad2deb8, 0x2e77e831, 0xcffd41dd, 0xa2bafb04, 0x83e8326c,
+	0x1ab1ceb6, 0x66b7b5e9, 0xf380376f, 0xf06f4a73, 0xdfe7237b, 0x2ecf2c82,
+	0x3aadee72, 0xb8e179f1, 0x93356e73, 0x2061bd6f, 0xd40baa94, 0xdde7bb46,
+	0x5e3a530e, 0x55fa01df, 0xeaf3cc3f, 0x73efd312, 0xd3a507f8, 0xf3faeccf,
+	0xe66b17e6, 0xb434fb3c, 0x79b464d5, 0x01dc2dde, 0xc8f389bc, 0x55eeed63,
+	0xd79b9076, 0x85f3c15e, 0x74f1b740, 0x2e5b4da2, 0xdc738376, 0xbb96d4a7,
+	0xda836f30, 0x940b8843, 0xf287bfd7, 0x10e3b4d3, 0x3f0969c6, 0x2244dc17,
+	0x538e763f, 0x98dcf3e2, 0x71c9cec2, 0x3a39c3fc, 0x39f4e3d0, 0xf8c72f3f,
+	0x6ba39c58, 0xbd5cfceb, 0xefbf07bd, 0x77bb13ba, 0x1a87e378, 0x4a20dfbb,
+	0x3468692f, 0x3a2dbd0f, 0x00383c09, 0xcf62430f, 0xa73e2683, 0xcc373918,
+	0x03f405e9, 0xde722fff, 0x6247d693, 0x248f7a5d, 0x1b4d07d0, 0x26d239d8,
+	0xf53df135, 0xad687ce2, 0x9e62ce81, 0x9087c96d, 0xffe0f6d3, 0xc979f8a6,
+	0x8bf1d4ee, 0xcbd9d3c0, 0x87f8f589, 0xddbbe976, 0x8c96e36b, 0x607cdaf7,
+	0xd3f704ef, 0x98f9e51e, 0xcc5acf2c, 0x96f82cf7, 0xbecf4f94, 0x17fde32c,
+	0x517b82ab, 0x85f51b8f, 0xc23597d8, 0xb78739fb, 0x89d996a2, 0x04a5afb3,
+	0xb39fb46c, 0x3f21d66f, 0xd51f5457, 0xc0ef8f20, 0x8b1aede6, 0x01f933d7,
+	0xc71b113b, 0x9eeafb3a, 0x24dbd47c, 0x97f27e71, 0x437a1f5c, 0x56f1c789,
+	0xb38dd39f, 0xfbc41fd0, 0xef1bdab1, 0x0a9ee28c, 0x50699dfd, 0x4cfa3f31,
+	0x8d41c590, 0xedcc7fe3, 0x203ee466, 0xcf412f6a, 0x5768fb88, 0x3cf30468,
+	0xe79626fd, 0x6a7efc69, 0x00d3c32a, 0xee04b387, 0x82d9750b, 0x52b8b9b8,
+	0xd4cdc32e, 0x6c7b9115, 0xf6f7f85c, 0xe4977922, 0x2f309b91, 0x925de452,
+	0x50bd5927, 0x613703d4, 0xfe10e51e, 0xdac03fbe, 0xeb4c2161, 0xecf97eb3,
+	0x2c7efe4d, 0x3320b26f, 0x9cffcfea, 0x829630fa, 0xd5a67ee8, 0xa049fc4c,
+	0x3ebdf82f, 0x22b27f22, 0x3575e449, 0x9ef7f3b2, 0xfa5f9d6c, 0x77a7bb47,
+	0x5b73ec2a, 0x552776c2, 0xc6eb4b29, 0x4e2a1da0, 0xfabe2075, 0x03911f6e,
+	0xc1ffc1fb, 0x76a707f1, 0xa57f1f8a, 0x6ab7faca, 0xe074dfb4, 0x4025be97,
+	0x09ddbee7, 0x831773e6, 0xa3bf4a13, 0x9e7c14cb, 0xfec1c944, 0x073b995e,
+	0xd22b1fe2, 0x2c63cc34, 0xdd00b0a6, 0x5e5c658c, 0x72a9aa68, 0xa5c72d1b,
+	0xcdc875b2, 0x1ebe4ea9, 0x277f6169, 0xffc2d1cc, 0xe7db184e, 0x9682e597,
+	0xe981d830, 0xecbfcc29, 0xdc31f2da, 0xb4eef1ff, 0x37582ebe, 0x7a4c0d4d,
+	0x21b7e610, 0xf00933d5, 0xeeb4befb, 0x9b5f946a, 0x34fc0527, 0xfcb1ddc9,
+	0x8871f8d0, 0xdbf4fc05, 0x61057187, 0xbc18f763, 0xe955943b, 0xa7c01e78,
+	0xae323f18, 0x03d8bf21, 0x13883d71, 0xf11efbf1, 0xce3a7afe, 0xcc26e3f6,
+	0xb64a6f8f, 0x59b47bf3, 0x00b608dc, 0x3fc7def0, 0xbef9972f, 0x5f775f96,
+	0xa7bb2d5c, 0xb89eee3e, 0x08129e83, 0x8bef1839, 0x3b37de33, 0x4f300494,
+	0xdc7f78d4, 0x307ebe08, 0x40f71bde, 0x5d0fb72a, 0xbddd1748, 0x5fd616db,
+	0xf03d5beb, 0x09dc1b9c, 0x6b3c5325, 0x4d0c994e, 0x0d73ca8f, 0x3e708fd5,
+	0x4e7e75f8, 0xc761ffb8, 0x2d7f003d, 0x79435fdc, 0x1dfc177b, 0x072a2d13,
+	0xbb930ffd, 0xb9a75ec1, 0xacd4ff99, 0xd3cb7bb2, 0xbf4058e8, 0xc98fb834,
+	0x89c4fa0e, 0x9bfe1b79, 0x3c7bee09, 0xa0499729, 0x92e9721f, 0x5c47ee1b,
+	0x3dc7cfc9, 0xf6e9326a, 0xca9d7e61, 0xf96fe3f4, 0xefb1633d, 0xb7a04a4d,
+	0xeb654d9f, 0x552cdb85, 0x9f198557, 0x9f740b3d, 0x70a8f26e, 0x599a9b4f,
+	0x331c42a9, 0xce4fca35, 0x13e2092b, 0xeecedfef, 0xab5f7f4b, 0x81ebf39a,
+	0xef66f3d1, 0xc7814600, 0x6fb8e92c, 0xbee38d3b, 0x4d4ce9d5, 0xd5a7818b,
+	0x99dbe426, 0xeadb720a, 0x91e589e9, 0x1f719a1f, 0xcfa688e9, 0x02f71783,
+	0xc480fb89, 0x7da2417d, 0x291586fe, 0xe91f1fe0, 0x74e77443, 0xd2ff5c42,
+	0xf9039ef5, 0xf3a3047f, 0xa67f73fb, 0xff15dfc2, 0x82f48fab, 0xc5f187bf,
+	0x4cdd73f4, 0x73cddf14, 0xd7f8b9ff, 0x79a3ddb4, 0xdcbc04de, 0x1578db9d,
+	0x531dd7c8, 0x4dfa41ce, 0xf9bf7cdc, 0x9e3f653c, 0x61250ecb, 0xef3bc23c,
+	0x4f180acb, 0xe6f687e8, 0x1d972f9b, 0x2a70bd06, 0x8bde09f0, 0xf626c646,
+	0x8dfe8d5f, 0xe71773e0, 0x0db231f3, 0x493356ed, 0x5cac7966, 0x73ee636e,
+	0xb37ee273, 0xe209fc9b, 0x98f62b3a, 0xd99fec08, 0xe1778941, 0x87754882,
+	0x8af6165d, 0xf5942db8, 0xda1fcd21, 0xb95ea54f, 0x17c42569, 0xc8682fe0,
+	0xfa80faf1, 0x56296e2b, 0x4f2bec0f, 0x79d607ac, 0x5a96c056, 0x58607a7b,
+	0xdb7e84d8, 0x5f07d537, 0xe281f419, 0x82be2270, 0x605033e0, 0xc43e1fff,
+	0x8000938b, 0x00008000, 0x00088b1f, 0x00000000, 0x59edff00, 0xe554707b,
+	0xef773f15, 0xc3cdddde, 0x210366e4, 0x804d8404, 0x85701020, 0x5c7c0188,
+	0x94422101, 0xd6da0300, 0x02101ba9, 0x16a52d79, 0x9b8cea9d, 0x8e233480,
+	0xd29dad13, 0x542cce96, 0x2ec4952a, 0x4dd0689a, 0xd15c04ba, 0x63e02711,
+	0xb46d63a0, 0xec083a2d, 0x98f8a71a, 0xe739ec76, 0x647dd7bb, 0xeffa7471,
+	0xe5f98617, 0x7cebdfbb, 0xe3cefce7, 0x648a12fb, 0x40a50307, 0x676a733f,
+	0xf1d31c02, 0xee016bb7, 0x0d1c442a, 0x00f250e0, 0x007f73e6, 0xd63c03c6,
+	0xe9b6dc06, 0xdb007b5a, 0x2ed8c9ae, 0x95d6c801, 0xfc76edf6, 0x06dbb8fd,
+	0xbab60173, 0xc00f77f0, 0xf9e9b1f0, 0xffbf8150, 0xbe956be7, 0x16bcea57,
+	0x373f4d7c, 0xb772b025, 0xecf7000d, 0xb5ddc372, 0x9d701e34, 0x913ac4a2,
+	0x46383668, 0x9dbac401, 0xef289d7b, 0xb1b0b870, 0xddc3db13, 0x5a64d759,
+	0x1e17c2cf, 0x40074ec2, 0xf3e36e8d, 0x91bdb05c, 0x35eb8157, 0x87973cf5,
+	0x33c685c7, 0xd752e6a7, 0x64df0e9b, 0x4e7dcf1d, 0x9aee5bd9, 0xc1ed1d84,
+	0x508fa459, 0x87df2ca4, 0x43b4ccf2, 0xe47b3ec0, 0xf8ddfefa, 0xcb400e71,
+	0xcce3376e, 0x987a9cf0, 0xe223df85, 0x8687b43c, 0x817ad35d, 0xb77b17db,
+	0xff3d69bb, 0x7c4afa5f, 0x470700b9, 0xc4219dc3, 0x2e98be67, 0x4ec1dbe6,
+	0x05e9e7e3, 0xd0402f2c, 0x39170ab6, 0x9c38e1a8, 0x1b0bf177, 0x9ff6b38f,
+	0xaab37bd9, 0x222a89a3, 0xfa80031d, 0x3d3b0ac8, 0xa7accf64, 0x676e1400,
+	0xfdc14105, 0x528297fa, 0x002bfa89, 0x5dfd82af, 0x01f7f1d9, 0xa73ef1db,
+	0x67f61f67, 0xc6e01de9, 0x421cbbf5, 0x8c00d3ff, 0xdf1372e7, 0xc2b2fdad,
+	0x65c806bf, 0xea411bbb, 0x0dc077b7, 0x9b7e89b9, 0xfdcb057e, 0xa7f05d43,
+	0xcb623b2b, 0x53e23945, 0x5cb1f600, 0xf7813909, 0x169ce4b5, 0x7fb4280c,
+	0xa303ecfc, 0x9f7d2e58, 0x7210e487, 0x67aa7842, 0x0dd7cd3f, 0xee96473e,
+	0x74f8d2f1, 0x20b3fcb9, 0x9e09f908, 0x55018fe2, 0xd13b3df8, 0xfb855d76,
+	0x42de0994, 0xe2fb7660, 0xc4da93eb, 0x7cc4aaf3, 0xfb6d7ebf, 0x1db9e40e,
+	0xa9a2afed, 0x70af1073, 0x7c3d39d3, 0x31bca43e, 0x83d0b67a, 0xe394f517,
+	0x8cdffd12, 0x52e6fe47, 0xf38c573b, 0x659cf53a, 0x400feca5, 0x8202d0fe,
+	0x912af7df, 0x00a937b3, 0xcdaf1fe7, 0x9d73c0af, 0xeb6c0db7, 0xd49f71c4,
+	0xd823ca85, 0xfdf3ceae, 0x5efdc75c, 0xaedd6f7c, 0x739fa899, 0x06b79a5d,
+	0xe050c1e7, 0x8fa87189, 0xcf346786, 0x007e4923, 0xc445e3e3, 0xfc7df4df,
+	0x189eda67, 0x58747e47, 0x1db0b8f1, 0x4813e2d3, 0x47f096ee, 0x7978830e,
+	0x078703f8, 0xf81715e7, 0x543f9276, 0x222eb6f5, 0x8e83cebd, 0x0ce23aed,
+	0xaf88f81b, 0x0f5c62a1, 0xbb7e1df7, 0x926b5f7c, 0x85753a1d, 0x7dc407e3,
+	0x0a916b13, 0xfaffd361, 0xf631721f, 0x24c1f91c, 0x1347bf5a, 0xb5e6b33c,
+	0xd238c0c2, 0x631c1b7b, 0x82c7beb4, 0xd9e26af6, 0x775d778c, 0x9fe3491b,
+	0x69f9fd36, 0x41710fda, 0xc90f6f81, 0x43e478db, 0xf39d1e47, 0x8009a19f,
+	0x4f764243, 0xf97b9ebe, 0xf34fe0f8, 0xee3dbf27, 0x8ffef1a0, 0x7415df91,
+	0x079f0f1e, 0x1f23beed, 0xddf078ed, 0xd875e9de, 0xfc2a53df, 0x1dab47b1,
+	0xdf5be347, 0x086b34b9, 0x123231f1, 0xd7d4bf8e, 0x86f49138, 0x4d985ffc,
+	0xf908767e, 0x69f849ef, 0x8a29f905, 0xaffc415e, 0x8e34f6a4, 0xc18e5dc3,
+	0x3d97ec65, 0x44bf2036, 0x203fb3fe, 0x40fdf5ff, 0x781fd1e3, 0xf3f654fe,
+	0x419b41ae, 0xf1c600ed, 0xb85edc29, 0x835dda9a, 0x73f6758b, 0x3679f21b,
+	0x80646bf9, 0x4c0109d7, 0x5029d321, 0xf648aa1b, 0xa3d63cdb, 0xdba7d4cd,
+	0x55d0dff4, 0x9e2f7c9e, 0x53554723, 0xb54f23fc, 0xb502f225, 0x3fdc2bb7,
+	0xed0df1e6, 0xc13fa24f, 0x740799a0, 0x2d383bc1, 0x5d4ff9e2, 0x7d6ffe22,
+	0x4e8afe65, 0x13d6e73c, 0x13f7f72a, 0xce31ca4f, 0x78193c9b, 0xe536e748,
+	0xe7da0f05, 0xfcc543d8, 0x1b6c5afd, 0x1ae74718, 0xba51165b, 0x38eeaaa9,
+	0x36bf384a, 0xbc4348b4, 0xa3c1cefe, 0x19f3709a, 0x81eebfc4, 0x29d8675b,
+	0x42719dee, 0xfdd88a16, 0x67fdfc55, 0xfadb0f50, 0xf219ff51, 0xf9871e06,
+	0x0e3b5007, 0xc7f6478a, 0xdc8e2b14, 0xbc7c4d5f, 0xda26add8, 0x120b4828,
+	0xf417da9b, 0x6c01ed6d, 0xc46057df, 0x71bf9789, 0x8d44e2fb, 0x8aafc9d8,
+	0xde7b2dc8, 0xe28f30fe, 0x98c3b77f, 0x2ee9fc41, 0x0ca14d83, 0xd74f7d3c,
+	0x4e954f98, 0xdfa992d8, 0xa0fc205d, 0x88bb003c, 0xaadd2d47, 0x5fbb441e,
+	0xc70d56e8, 0x431f00d5, 0x8a02ed60, 0xb0630137, 0x01bff518, 0xc8031fac,
+	0xa6894d1e, 0x68daf641, 0x6b1c3736, 0xd63ca1a8, 0x20e83ea4, 0xa4f8da3f,
+	0x55d0e1f6, 0xffc6ee66, 0x27686153, 0x53ff11c5, 0xed82378a, 0xfb527b4d,
+	0x7887c21b, 0xa953c35e, 0xb13a9bdf, 0xddb44aed, 0x3a8c5705, 0x997f033b,
+	0x42b09304, 0x5c711e20, 0xdc70ecd0, 0xd97dbc89, 0x6f4953ed, 0x3f426d07,
+	0x0fd94f18, 0x2cbf2b55, 0x7e287114, 0x2fec8201, 0xf3474207, 0x9df9ae50,
+	0x63da5ea5, 0xcff45dbe, 0xfce52def, 0xbaa0f602, 0x65500d12, 0xde9096f8,
+	0x176f2f51, 0x9937b9e3, 0xfc7411f1, 0xa6cded87, 0x1e91361e, 0x073635d0,
+	0x079ee553, 0x4fb4edc1, 0x3f3c01fd, 0x903c6bce, 0xd7e96fda, 0xb4d3ae6f,
+	0x2c7464fb, 0x75287362, 0xfd1e1f9c, 0xa17aa554, 0x73fbf537, 0x5cc09bc0,
+	0x60133f4b, 0xeb07e902, 0x97b1802b, 0x8ee4678f, 0x344f97a4, 0x329c5751,
+	0xd7fab1d7, 0x41697c9a, 0x344e28f1, 0xf120d505, 0xc8129365, 0xcfb79503,
+	0xd7b943ad, 0xe0f0bc5b, 0x3dc7d43f, 0x9bd67ea6, 0xe173e0bc, 0x4a3de6fc,
+	0xeff38230, 0xfe553479, 0x25baa10d, 0x79233ce3, 0xab53a9b3, 0x6e7e6073,
+	0xb40e2d80, 0x98de48f9, 0x70e800fa, 0xe74afe50, 0xb94f8b4f, 0x6e46dc8c,
+	0x5fffdcdd, 0xe22d46ee, 0xf8da7ca0, 0x201bc52b, 0xe27fbbf9, 0xe791b280,
+	0x60dbecb0, 0xd73cc4f1, 0x56cf3df7, 0x61d3becb, 0xbab7db3a, 0x37d93bf0,
+	0x6f463ebd, 0xc618ff63, 0x9150577c, 0xfcfa4fe0, 0xbd9d6625, 0x5e303774,
+	0xcf2cfde8, 0x8362f383, 0x68c079c3, 0x47bca6ab, 0xf149c9ed, 0xcec0ada6,
+	0xfe2ceefb, 0x500b63c7, 0x690f4b3f, 0xa0e0ac5e, 0x96ad4ae9, 0x95d373f2,
+	0x4e4d14af, 0xe6ed27ca, 0x3687a5f8, 0x9c5165b4, 0x25e293f8, 0x7f858bf7,
+	0x40ed4036, 0x829de2a7, 0xc383a3f3, 0x7b202e6f, 0xa5cce714, 0xb77dfcf5,
+	0x76c2e8cc, 0xaf1cdf74, 0x17ac49ea, 0x2df38b75, 0x7ce352bd, 0xe724be79,
+	0xbf68350f, 0xfb2f6306, 0x9e97f9f1, 0xc79f9077, 0xc62814ba, 0x47c5a28d,
+	0xbea00d9f, 0xd270bfcf, 0xa2f84541, 0xa18e93ee, 0x2e03a96f, 0x50885504,
+	0x3877b03c, 0xdb52bee8, 0x10c72ff3, 0x7d4f45d5, 0x34bd20e0, 0x471c2af4,
+	0x41a8f61a, 0xb47afe0f, 0xf10745ef, 0x8153a5a8, 0xfdc9a531, 0xe663f71a,
+	0xc23e7964, 0x092adffe, 0xf933ae7e, 0xe2d3f168, 0xce333aeb, 0xe54f6fac,
+	0x763046e3, 0x0dfce096, 0x4f1c759d, 0x8e647437, 0xaabc5633, 0x28ead88f,
+	0x4eefd7ee, 0x1bf9b71d, 0x941c061e, 0xd9e3d317, 0xd3878b51, 0x70f11a60,
+	0xa26eb238, 0xe97f791d, 0xa844c364, 0x37afa918, 0xe56d7ccc, 0xf5e9cbcb,
+	0x0e67ef62, 0xf1fe73ca, 0xfc17df1a, 0x48edffe3, 0xe209c9e6, 0x25b7e609,
+	0x3ed94fe1, 0x8b23e135, 0xc0dbbf69, 0xbdfa44d7, 0x53a39c2d, 0x24335bfd,
+	0xd85976fc, 0x242a0c19, 0xdbf38d9f, 0x41da106d, 0x3bf58ff6, 0x92f03fb9,
+	0x981075c2, 0xd4deac71, 0xdfa9bd77, 0x8d5fbc7a, 0x30f54d9b, 0xfae283ea,
+	0x6ea37b55, 0xbf316fec, 0x3476bd37, 0x0cbcf48f, 0x16599c44, 0xb2c67112,
+	0xd2279597, 0xd0331cfd, 0x8ddec98f, 0x823c2761, 0x719b97e3, 0xc6277966,
+	0xdfa46519, 0x30e6c6c7, 0x423d025f, 0xd8bcf4de, 0xfe3a9dd2, 0x2ba8814c,
+	0xf67acd87, 0xd57b34cb, 0x02a721c4, 0x51e65bd7, 0x359e41bd, 0x8ffda768,
+	0xe1fe42cf, 0x933dc7fa, 0x397b1ef8, 0xd669bd3b, 0x057b56e9, 0xc3ea8b12,
+	0x5b91de90, 0xf737e490, 0x3921d860, 0x01a9e61a, 0xbdd27fd2, 0xd2ecc565,
+	0xdd16f78c, 0x1d5fe38d, 0x6bd1febf, 0xc62def92, 0xe1e272de, 0x597f8a2f,
+	0x43e29b33, 0x172788a7, 0x7cdd70ab, 0xeb81c3aa, 0x33a77f5a, 0x3f0bf748,
+	0xadea88f2, 0xf213c4a5, 0x0b72b0cf, 0xfeeb04f1, 0x4afeb4f1, 0xe146192c,
+	0x8af657f9, 0xe2e5647a, 0xaf77994f, 0x46e7164d, 0x74c98dbd, 0xffeab00f,
+	0x9d442f96, 0xb16f7d69, 0xa03df10f, 0x6f30ac25, 0x9cbe07bb, 0xc6f28a50,
+	0x509f3efa, 0x16eb5887, 0x54f5b479, 0x797d5036, 0xa0ceaff1, 0x74bce513,
+	0x1f0fe5f8, 0x2ea5f6c0, 0xf2e63117, 0xc1d95c71, 0x0da2d37e, 0x721d7115,
+	0xebbceb18, 0x7c73bd68, 0xc81ee88f, 0xbc40dbba, 0x5137ceb8, 0xb5bb16ff,
+	0xa9d71b42, 0x9183b06d, 0x3ae499ea, 0xcac75eae, 0x8d864b14, 0x566bfe66,
+	0x7bc42373, 0x57346e2b, 0x883dc625, 0x6b88dffc, 0x533b38dc, 0x5cec08fc,
+	0x1b8c8ca0, 0x777b158d, 0xfd5321ba, 0x977ce113, 0xe3dcef5e, 0x24e6de46,
+	0x1e2a1b78, 0xde84b67a, 0xa13b4e71, 0xffce94b5, 0xafb204d5, 0x9ceb7e75,
+	0x9e7e4ce7, 0xc601fba7, 0x952ad69d, 0x279b65f5, 0x708baad0, 0x7626f31e,
+	0xf5483732, 0x17fe8cfe, 0x8d0788cb, 0xb25c7007, 0x7d1da1ff, 0x90438854,
+	0x421cdcf9, 0x3ce492a1, 0x74e2ffd8, 0x97d91be2, 0xa6fc9dde, 0xb2e5d37d,
+	0xfdc45150, 0xfb88a6d5, 0x287d8f6b, 0xfe4dee8f, 0x306d6ac3, 0xfcd67ec9,
+	0x80fe4d7b, 0xefd3dc5d, 0x7c4420a7, 0xd81e0b0a, 0xef2204fa, 0x1a778ad6,
+	0x334aca0d, 0x7e442ff4, 0x035e4b7c, 0xea6d0b92, 0x33cce2d7, 0xcb1b49ff,
+	0x5c393fe3, 0x03df9c94, 0x8a18eeb4, 0xc04dad7b, 0x28072471, 0x0b1c62ce,
+	0xf97b63e4, 0x97ae2ce9, 0x6abb65a2, 0xa9956e38, 0x5b6b7140, 0xc9a9e6ff,
+	0x05bfe1f9, 0xe91738aa, 0x95aa98a1, 0xa6fec527, 0x9560da1e, 0xbc17cf79,
+	0x7888140d, 0x27a3f782, 0x47ee19dd, 0x6ddeb5df, 0x3d8273ed, 0xfa974ec9,
+	0x0dfef9ea, 0x52a6f85f, 0xef9f019c, 0xfd6dc26a, 0xfbe953f5, 0x57cdfea5,
+	0xd2a4bbf3, 0x02bf9296, 0xd5315f24, 0x7914ef45, 0xefb9bfb0, 0xd460229d,
+	0x276a7e58, 0x6efdafe7, 0x8ec4e751, 0x84fbf164, 0x07826502, 0xdd653930,
+	0xb857eea6, 0x55ea688e, 0x583e648d, 0x22dc7af3, 0x1e261e0f, 0x1486b620,
+	0x487ced83, 0xf3ef5360, 0x16a34f2c, 0x76b63b62, 0x4bbfa26a, 0x3c97d620,
+	0x00ee002f, 0x39feafc9, 0x52754c3c, 0x87964f73, 0x3eb14bde, 0xc4b1f914,
+	0xfa8fdc19, 0x99c2f459, 0xcd5b8a22, 0xf50a682f, 0xbbea50c3, 0x679b736d,
+	0xbde88fc5, 0xe81575e0, 0xda55e1dd, 0xc4b0ed02, 0x4d54d181, 0x1d66efca,
+	0x5f9bb560, 0xc8d4abaa, 0xcb06dbef, 0xe6ce512b, 0x3b4abafe, 0x78b4f419,
+	0xb634ba0f, 0x3009eb91, 0x95b65d84, 0x27f0a634, 0xf7290300, 0xb14e506e,
+	0x8d61631e, 0x03a281df, 0xfde9ca67, 0x6fe4b9a8, 0x7347e4e5, 0x537bc50a,
+	0x554bbd33, 0x7de27c90, 0x4d7f9aca, 0xafb3f0a1, 0x0b7b7f4d, 0x7bcda3bc,
+	0x9e1b2ece, 0x4d1aadff, 0x05eb847f, 0x6edd9f51, 0xb7d5cdc6, 0xbc5f5c69,
+	0xc78d1e05, 0xcca18a1b, 0xfac8bce5, 0xa78cddb9, 0xe352aa2b, 0xf03dcadb,
+	0x91a11670, 0x635b4eb9, 0x0c0abfae, 0x7575775e, 0x6635b8ea, 0x36a6b69d,
+	0x56e7fbf5, 0x930c9cdc, 0xc2ea6e39, 0x66c75b9e, 0xb90c7e71, 0x0cdef93c,
+	0xb475e279, 0xf541c78f, 0x683c0b9a, 0x1c66e3a7, 0x01c92b6d, 0x177e28f7,
+	0xa3fd9209, 0x53bfbed9, 0x719f8ade, 0x34553bf2, 0x1565ce2d, 0x29cd3f0a,
+	0x514a7114, 0x9b276ca2, 0xa3ed9c72, 0xed9baf28, 0xe7ec5237, 0x533bc4d7,
+	0xfc937914, 0xacc27c8e, 0x788aa223, 0x7b974b47, 0x8778a17c, 0x34dfa11a,
+	0x5fd08ebd, 0x36a2de9a, 0x6ec5c751, 0x5ffef856, 0xc519f58e, 0x958bb5af,
+	0x14f141a0, 0x2bfdc0f5, 0x54c3d615, 0x62704f88, 0x8f981dff, 0x4d8beea8,
+	0xb6fd7513, 0xefe2f9b5, 0x9b780124, 0x64f168ec, 0x57e45d85, 0x286b0f8b,
+	0xdf12cebf, 0x6ede9f29, 0x7755d3e6, 0x850af35e, 0xadaf4b45, 0x2f9ca7e4,
+	0xb1463146, 0x475e20be, 0x53c7eafb, 0x917d23dc, 0xf32f479f, 0xadc2742e,
+	0xa72e7eb1, 0x2fd36abf, 0xa839476e, 0x7adefdae, 0xeba5e734, 0x9530bab3,
+	0x2b581887, 0xf2138fc8, 0xc7e6333b, 0x943ca6d5, 0x47e80363, 0xedfd4d6b,
+	0x2eadf7cd, 0x0477efc4, 0x7e4c3d57, 0xb3d94675, 0xce2410ff, 0x9d187a4d,
+	0x85addeb4, 0x090c3de2, 0x9b7ef192, 0x16f0179e, 0xd7b8e016, 0x54af718f,
+	0x241777bc, 0x933eebe9, 0xb161d7d7, 0x079c764a, 0x25de9501, 0xdc3b03ae,
+	0x443b6d45, 0x200b88ec, 0xfe81dfc2, 0x4fc45608, 0x9d73fcea, 0x9944723d,
+	0xeff8e3c5, 0x9061f343, 0x7108ddc7, 0x3b740346, 0x32f042e1, 0xcb22b956,
+	0x08af6ca9, 0x4baa5485, 0x82f64523, 0x5213a8b3, 0x2473a665, 0xb38763ce,
+	0xd2a2fd56, 0x0b6011df, 0x3cfd3dd9, 0x0c79d61d, 0x0d4a4bc1, 0x555ca177,
+	0x93695210, 0x1548601b, 0x39f32ff1, 0x4ddd6016, 0x01b8b77f, 0xfacd9fed,
+	0xebf8f208, 0xa66a3ce9, 0xaf2d0cf3, 0xeb1615e0, 0x29129f24, 0xf39d7db2,
+	0x2b11cfdd, 0x6f375e02, 0x03e2fc7d, 0xfbbf8995, 0x9eafc378, 0x2a3d3fa6,
+	0xd3f70186, 0x665c4ddf, 0x9fb11add, 0x788f86ff, 0xe17dfd3f, 0x5215dfba,
+	0x7ce8f23c, 0x8781f04e, 0xf60638f8, 0xb23f74e8, 0x7142b8d1, 0x128b8fdc,
+	0x5c103bbc, 0x2f21b1e5, 0xf2e518ee, 0x43cca5aa, 0xab8d675e, 0xcf7a4ae3,
+	0xe14e4d43, 0x91e4c1ba, 0x6aab7f25, 0x2af891fc, 0x8944a251, 0x944a2512,
+	0x44a25128, 0x4a251289, 0xa2512894, 0x25128944, 0x5128944a, 0x128944a2,
+	0x28944a25, 0x8944a251, 0x944a2512, 0x44a25128, 0x4a251289, 0xa2512894,
+	0x25128944, 0x5128944a, 0x128944a2, 0x28944a25, 0x8944a251, 0x944a2512,
+	0x44a25128, 0x4a251289, 0xa2512894, 0x25128944, 0x5128944a, 0x128944a2,
+	0x28944a25, 0xffe12251, 0x72255300, 0x008000ab, 0x00000000, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0x00088b1f,
+	0x00000000, 0xc5edff00, 0x30001131, 0xee300408, 0xd85aa12a, 0xaa66f6b1,
+	0x964d2113, 0x5dbbcce4, 0x6db6db15, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x3d017e3f, 0x009b1baa, 0x00009b1b, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0x00001000, 0x00002080, 0x00003100,
+	0x00004180, 0x00005200, 0x00006280, 0x00007300, 0x00008380, 0x00009400,
+	0x0000a480, 0x0000b500, 0x0000c580, 0x0000d600, 0x0000e680, 0x0000f700,
+	0x00010780, 0x00011800, 0x00012880, 0x00013900, 0x00014980, 0x00015a00,
+	0x00016a80, 0x00017b00, 0x00018b80, 0x00019c00, 0x0001ac80, 0x0001bd00,
+	0x0001cd80, 0x0001de00, 0x0001ee80, 0x0001ff00, 0x00000000, 0x00010001,
+	0x000e0004, 0xcccccccd, 0xffffffff, 0xffffffff, 0xcccc0201, 0xcccccccc,
+	0x00100000, 0x00000000, 0x00000000, 0xffffffff, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000, 0x40000000,
+	0x40000000, 0x40000000, 0x00088b1f, 0x00000000, 0x1113ff00, 0x51f86066,
+	0x423ec08f, 0xac9d0c0c, 0xc4b462a8, 0x1818990b, 0x12b102fe, 0x3c430333,
+	0x203aded0, 0x2388107d, 0x16181858, 0x2fd610b0, 0x022bd404, 0x2c4062c4,
+	0x19b7c401, 0x9cdfb348, 0x1f0f680b, 0xc8037f82, 0x3f4024be, 0x1c360fff,
+	0xfb5f40ad, 0x1819d502, 0x8aa06bfe, 0xf2a26831, 0x9bf13519, 0xcf2684c1,
+	0x2167c68c, 0x63247fa0, 0x0d75b600, 0x000400f1, 0x00000000, 0x00088b1f,
+	0x00000000, 0x7dd5ff00, 0xd554780b, 0x673ef0b5, 0xf3399cce, 0x00fde4cc,
+	0x108f0992, 0x2104e034, 0x0432b445, 0x3b69488c, 0xc514543c, 0xf791e109,
+	0xb14a3e44, 0x44033bd2, 0x350af808, 0x0380a050, 0xed168d02, 0x06823ca0,
+	0xfda2901c, 0x5a1bdedb, 0xcb6ab7bd, 0x880b851f, 0x52d11921, 0xf77ac5ea,
+	0xcc9f7b5a, 0xd0049339, 0x9fdffed6, 0xcfb3767e, 0xd6bdaf7e, 0xf7b5ebda,
+	0xa441c448, 0x84be421c, 0xfc8471bf, 0xa484205e, 0x83bf52c6, 0xa908a3a6,
+	0x2d0867d9, 0xc26425cf, 0x5c64633e, 0x136fcd0a, 0x179a2642, 0xbcb19b0f,
+	0xfbcb3373, 0xdd4f6d0d, 0x01c9cb4a, 0x1349d903, 0x49116f92, 0x1467211a,
+	0xd146fec2, 0xf321117c, 0xb35b2ccd, 0x426cc8ed, 0x98b797eb, 0x3fb69988,
+	0x81e0d7b3, 0x429dc2fc, 0xd439d088, 0x67255c1c, 0x8417fed1, 0x45d9b084,
+	0x47e93bf3, 0xd7b15e5a, 0x5dfa8210, 0xaf34b68e, 0x01f3908d, 0x9864b885,
+	0xed693bdf, 0x86548405, 0xd16494f6, 0x05bb957a, 0xa9c748b7, 0xc6442cdc,
+	0x42ef828d, 0xa8bb40c8, 0x5712b66a, 0x97c39b3e, 0x75c5c704, 0xc742dc2c,
+	0xa912fda5, 0x61daf651, 0xcbb2b5bc, 0xf9cf831e, 0x51c71380, 0xd3cf35f3,
+	0xdabb6871, 0x2c370497, 0xbe2456b1, 0xf3bf1d30, 0x73e679a0, 0x32df5836,
+	0x5daecf39, 0x587ee947, 0x9b686547, 0x7983625c, 0x17e7936d, 0x6aabfac4,
+	0x6bcfd64e, 0x9f74a0c3, 0x3d3e3ca6, 0x95c4201f, 0x1257cb17, 0x60db09e2,
+	0xea7921fe, 0x63f7d8f0, 0xb3f71124, 0x5c40d9aa, 0x427fac4a, 0x40ddf882,
+	0x2b8011dc, 0xb5bbb569, 0xa9d176fb, 0x8985b7df, 0xf31f6dbc, 0xc3ef9a79,
+	0x27a70e5a, 0x4d347e61, 0x499738f0, 0x1309fd74, 0x4ab85389, 0x3246b5fb,
+	0x7acbdc33, 0x186405ef, 0x888f884d, 0xe25adf38, 0x2ed21fbb, 0x4e9caeb1,
+	0x3245fe9e, 0xabbd74bc, 0xfbfed604, 0x57f585c4, 0xb03d900d, 0x2aef6baf,
+	0xcb90ce7a, 0xf90292d7, 0xd0d6bbf9, 0x7926932e, 0x0257c008, 0x614015b8,
+	0xcb40d07b, 0x4713bbed, 0xe16971ae, 0xc61f3c57, 0x75128c73, 0xcdbce3fb,
+	0x4d27cba2, 0xaf838a4c, 0x46e679cd, 0x34f3a79e, 0xd211d189, 0x67c64220,
+	0x8b6f882b, 0x9ba465b3, 0x79b3b7c5, 0x85b6ce2d, 0x38ff3482, 0xe685b834,
+	0xce4ddf65, 0x4741e05e, 0xa4064916, 0xe98c913e, 0x7000de59, 0x9c5078a4,
+	0x03ef2573, 0x68c5fa9a, 0x1f4d225e, 0x0107c616, 0x8e012b2e, 0x481ab534,
+	0x6adbac1d, 0x86a70822, 0x05cf4521, 0x9d5a7035, 0x1b94e14b, 0xb1b577eb,
+	0x09b8832e, 0x23bc1359, 0xf8526528, 0x1dcdd2e6, 0x23bdf30b, 0xdb4c1c12,
+	0x00ce2ee7, 0xa5322a78, 0x32f8ed03, 0xc700bf1d, 0xae38046f, 0x3f18fbe7,
+	0x463792be, 0x6079be37, 0xb37c6eb9, 0x38a7c74c, 0x5df829b9, 0x63853e3e,
+	0x233f2116, 0x95f1c5df, 0xfc704b81, 0xeb949906, 0x8f74b7c6, 0x7771821f,
+	0x437e31f5, 0xfafd58de, 0xd7ea5607, 0xbff5b32f, 0x8f9bbc10, 0xfff5c16f,
+	0xd6cddc82, 0x6c47f03f, 0xafda26fd, 0x37477c76, 0xf700c3fc, 0x5fc07dfd,
+	0x7e9b7a19, 0xf5aa83fd, 0xf1b137eb, 0xc83e0d5f, 0xf8e1b7c7, 0xd90791af,
+	0xc52d07fa, 0x3aa64df1, 0x991693b7, 0x21752c72, 0xfd176bc0, 0x074d3a5f,
+	0x45be71d3, 0x422482d3, 0x901ffb68, 0xc5870881, 0xa850df16, 0xf142c97f,
+	0x0be316cd, 0x0052b424, 0x38e5afba, 0xa977773d, 0x326d24fc, 0x4bf185b7,
+	0x10e9f942, 0x96cebf3a, 0x4fbd2f4f, 0x47773be7, 0x42e0dba1, 0x6ed0247a,
+	0xdf8a5f1e, 0x8841e0c0, 0x1daa7b37, 0x7b3793f8, 0xf22c70da, 0x40599f00,
+	0xf03d8e70, 0xf1a35e78, 0xbe86bb4c, 0xe7d06c16, 0xbcfa422b, 0x80f1ef3a,
+	0x160d3424, 0x3a3afc93, 0x8fca19ee, 0x1f93e508, 0xe9047e52, 0x08792359,
+	0x94d38dce, 0x4b59115f, 0x5f70cb57, 0xfe0c48ce, 0x91583667, 0x7dad2f1e,
+	0x60832658, 0x640a563e, 0xf24178cd, 0xff9d3e76, 0x089d7c3d, 0x57e022e7,
+	0xb4f73d6c, 0x47a14cd6, 0xb2779d2c, 0x31a75dae, 0xf5be0462, 0x963d0920,
+	0x7b2ffb41, 0x87d18f62, 0x4fd3c7fc, 0xfd26ba44, 0xf7d74a44, 0xd87e34ce,
+	0x733769ae, 0x3f635aef, 0xd3e7de9a, 0xae559f7f, 0x431a77cf, 0xe5ce1146,
+	0xa36c810a, 0xa7583ffd, 0xf9c7572f, 0x7db1248c, 0x43dde3e6, 0x049992d7,
+	0xfe91fa59, 0xeb64e7db, 0x497ae7a0, 0xca1f383f, 0x9a4ef704, 0x8dea107f,
+	0x4b7c7d84, 0xe7b3ef86, 0x90d7ba25, 0x5225f39e, 0x9273f606, 0x25271b9f,
+	0x08b753e3, 0x7edfe73d, 0x4711ead8, 0xfc475cf4, 0x4e221fbc, 0x968787dc,
+	0x0697bfe8, 0xf0c7d7e3, 0x3e32dfc0, 0x7be6b4f9, 0xe4aef84d, 0xa97c1ad3,
+	0xcbaa9e6a, 0xd3ee8457, 0xa1b05fd5, 0xaf3e5754, 0x5e5742b1, 0x2eb0f0d6,
+	0x447c1a5f, 0x25a1ff57, 0x0fe574cb, 0x95d6add6, 0xab5f2acf, 0xdbe7dfcb,
+	0xef7faba8, 0x72ba6dcc, 0x0e4570e1, 0xd97e37c8, 0xe1499397, 0x135de17e,
+	0xf0c42ed9, 0x73f3c558, 0xcf4c0c81, 0x2e985c07, 0xf4a57a03, 0xa9641e2f,
+	0x0e697b7f, 0xf78ab9d0, 0xd9758d67, 0xf8f12740, 0xf18f4f1d, 0x03780c78,
+	0xe75f50e7, 0xbc40b579, 0x025df740, 0xefd48f3b, 0x97fce67f, 0x6f48e590,
+	0xc7a332d5, 0x8b3397ea, 0xe5aa5e81, 0xad9d5e48, 0x58497eef, 0xb21075c0,
+	0x08a9fb1d, 0xedd0aa51, 0x3cdd20eb, 0x70188e3b, 0x3e105afc, 0xdef3816e,
+	0xf8c02a5b, 0xa7b23791, 0x78f9870f, 0x25be60e6, 0x4762f915, 0x69f25260,
+	0xc5e7e009, 0x26605cf4, 0x3e0267a6, 0xca074f4c, 0x5030fd31, 0x607b6987,
+	0x0327a609, 0x14ff4c41, 0xbdf4c068, 0x7fa62340, 0xf4c06c0c, 0x4c21033f,
+	0x4c1e033b, 0x14d4f1bb, 0x90f903cd, 0x20226fbc, 0x2e1492e2, 0x3372a97f,
+	0x5c85f961, 0x2e0bee6e, 0x27defbf1, 0xbc5048fc, 0x2c59beda, 0xa8e83f6c,
+	0xf3a50893, 0xbbf6c335, 0x44bcb0e7, 0xdd843f90, 0xbbb2f95a, 0xf483d85f,
+	0x9fc7ef6b, 0xf381dce1, 0x9e030def, 0x8af2247f, 0x0b5dba7b, 0xc047ba0f,
+	0x86b75e6f, 0xd5fe5276, 0x7c31705f, 0x2e5c06b9, 0x039cf5be, 0x0cc2e8f8,
+	0xe6fa79d1, 0x1357d7be, 0x34da75ce, 0xe8b7f1f3, 0x851b17f3, 0x741e4c49,
+	0x0e18cc25, 0x9e74e0f8, 0x96fb0c1e, 0xcfdaa981, 0x79d90cc2, 0x85ea193a,
+	0x77643d70, 0xd517e222, 0x3fdeb047, 0xf57d1e7b, 0x5ec79b13, 0x47b68a72,
+	0xdf507b19, 0x0bc4fbf5, 0xe41933e9, 0xbc6ad268, 0x85aae704, 0x40e5f7fe,
+	0xbcc257e8, 0x6c3fafda, 0x826e9a08, 0xe375bbef, 0x552e76d1, 0x7d7683d7,
+	0xce449fe3, 0x5941f8a8, 0x6fae09d2, 0x9d9b3556, 0xa89d7e7a, 0x325bfbcb,
+	0x0dfe3a9d, 0x4af4a0ff, 0x9c120a26, 0x2b1cd5bb, 0x835b24ba, 0x1bb7e740,
+	0x8bee86f0, 0x6af5605f, 0x47d97694, 0x67dc2d3f, 0xe17df3c7, 0x05efae5a,
+	0x9fb41b49, 0x70778633, 0xd34ace85, 0x449ce51f, 0x9e32f9a4, 0x39aef553,
+	0x7cf0ab7b, 0x94126fff, 0xc8ff3ce1, 0xe70458d0, 0x9b786553, 0x76fba1ec,
+	0x80938881, 0xbf205cef, 0x91acdc77, 0x3208afe3, 0x41acb7ae, 0x79516e7f,
+	0xf2a79747, 0x9f3f2e8e, 0x3d034b4c, 0x91937e6a, 0xff285e8a, 0x043bcae8,
+	0x80bbf627, 0xfe04add4, 0xcdef6a7e, 0x700adcf7, 0xc8a5b03c, 0x6873f347,
+	0xbf4a3f71, 0x6427cd95, 0x3aaf93a0, 0x1fb4057e, 0x755c73d9, 0x51f43d5d,
+	0x669a870e, 0x85ea13c1, 0xa6459b9f, 0x9f74a9d7, 0x72753a25, 0x7e5c5e5b,
+	0xf6fe5c64, 0xc6d7fcb8, 0x908a149f, 0xdefdb169, 0x44f2f82d, 0x478043e0,
+	0xf9bc3109, 0xf467fdda, 0x3fe47fe8, 0x44feffb5, 0xffb4ff87, 0xfda9ffdb,
+	0xff31ee0f, 0x5ff5bdc9, 0x149e8a6b, 0xdd7b1ce0, 0xe198e78c, 0x7a0839b0,
+	0x19ab7f82, 0xf637dcf5, 0xd01775fd, 0x26b3e75d, 0x8e7a6262, 0x93bff280,
+	0xdf767bf5, 0x0ffeb75f, 0x8d0b3f2d, 0xb60b7f69, 0xef80fad2, 0x64c8adbb,
+	0x184e5eba, 0x2f5d2841, 0xefd1fd78, 0x2e2f2a11, 0x91dee2b0, 0x0f4cbf05,
+	0x51fb00ad, 0x799fcb2a, 0x08284071, 0xed979e76, 0x1808c311, 0xa3ed40f7,
+	0xad2fee30, 0xa42d1e84, 0xd61375f3, 0xa305fceb, 0x2ee78a7f, 0x24415e9d,
+	0xe77e3757, 0xb5241c02, 0xe052f85c, 0xf0e22840, 0xd55ef4a1, 0x8f31e092,
+	0xfbeaf800, 0x2e8c60b9, 0x0eff2376, 0x5bc88966, 0x77205922, 0xcb53d7c7,
+	0x5540e2fd, 0xb6ec581e, 0xfd53eac5, 0x37cb3e79, 0xa0fd3166, 0x052feb0d,
+	0x958d49fa, 0xd597ecf7, 0xcec891ff, 0xac3f58db, 0x53d72d7d, 0x1572e9e3,
+	0xe6aacba7, 0xa6f6862f, 0x7f32f55f, 0xa65fbefc, 0xafdd0582, 0x10e6cbe0,
+	0x23b5616c, 0x6bef7a82, 0x6d511edc, 0x9e1ea089, 0x3905c7c5, 0x6944f20f,
+	0x8ced097f, 0xf404cc07, 0xedd79add, 0x9adfd81e, 0xf33d7ffd, 0x63bdfa33,
+	0xf80dd59f, 0x0ffaf350, 0x4c6bdf71, 0xdd02d991, 0x2a1ee8bf, 0x1dfad2ff,
+	0x4d9d7e7b, 0xc61d3f68, 0x55d27648, 0x613565ec, 0xcafc9c53, 0xe03e7de6,
+	0x98f2019e, 0x3b5447ca, 0x53fb7c5a, 0x45cff162, 0x4417970a, 0x7c88b34f,
+	0xa9df6f4b, 0x5d39525c, 0xfdabde7b, 0xbecc4a54, 0xc43f7ea4, 0x1ba87be1,
+	0x2e935fce, 0xf2e83d6d, 0x02ebcc10, 0xf3834b69, 0x13f9b6a8, 0x4aed4419,
+	0x2e813590, 0xbee032df, 0xa12b9b10, 0x4d4df937, 0x4cc605cf, 0x75de5097,
+	0xfedfea63, 0xc28b2381, 0x4f86fe7f, 0xe91ce1b2, 0x199e643a, 0x7f39fd42,
+	0xbf1834ba, 0xa0f5dca3, 0x51eb8ac1, 0xe3a4abc7, 0xf5b10bd5, 0x755c6b5f,
+	0xaefc753c, 0xcfac9f8d, 0x69f8e174, 0xb8fc7e30, 0x5c753b6a, 0xde7c572a,
+	0xd886978d, 0x414b96fe, 0x72c3ce19, 0x947ee122, 0x14797a93, 0xcd8841d2,
+	0x74ed4977, 0x97ca9f90, 0x4a5d3edc, 0xf900f366, 0x6901e60d, 0x5ef1c6d7,
+	0xfb7fa380, 0x3a520554, 0x305f5c0f, 0x221a16d3, 0x3ca0f9b0, 0x7d414c1f,
+	0x777be257, 0xe9fdf026, 0x17b87f4c, 0xb81d39fb, 0xf1c6f313, 0x1437c875,
+	0x85a988d2, 0x1ffd3184, 0x8f9c46f1, 0x90238a8d, 0x98f81aa7, 0x0e11c359,
+	0xb6245979, 0xe913f87e, 0xcdbf2b38, 0x572ddad7, 0x30e1c53d, 0x580679c5,
+	0xbe47d066, 0x08ae5da4, 0xa3f364cc, 0xbe5128f8, 0x517f84e5, 0xc79d3a78,
+	0x103826ff, 0x4a7ad9ef, 0x5d056848, 0x7c73248f, 0xbe184c99, 0x5e265c34,
+	0xdd3c1d8d, 0x090ef8c0, 0xe83e411e, 0x5fc1fde0, 0x9451e694, 0x245655a7,
+	0x7802df64, 0x089270e1, 0x91b7638c, 0xd8047588, 0x5af3a551, 0x1b79c1f9,
+	0xa8ecebcd, 0x3a78f2fa, 0xb14bed53, 0x7e70db71, 0xe2cfdfb4, 0xae2cfdfa,
+	0xd6aecfdf, 0xbf270aaf, 0x740bddb2, 0x87c058fc, 0xa78ea6fb, 0x7c28f1f0,
+	0x113f51f2, 0x429dee2d, 0xd616ae0c, 0x22633d15, 0xc0337fb8, 0xf4e14ce8,
+	0xe8e1c278, 0xb29050a3, 0xd7fe0081, 0xfd353b73, 0xed34daf3, 0x7a415388,
+	0x99c7cd5e, 0xa1fc6061, 0x9ecff39e, 0xee78cdf5, 0xee6fa8f7, 0xf7a5beab,
+	0xe6bef6fa, 0x3e75ed63, 0xd55f955f, 0xff5ffebe, 0xf869711e, 0x52829396,
+	0x30ccaf2f, 0x126b7ed0, 0x1dbdfccf, 0x0f5144bf, 0xf7ebfa51, 0x4af802cb,
+	0x5832c7f1, 0xa056ddff, 0xf05ee7df, 0xd74996c5, 0x63af9331, 0x14e97366,
+	0x86264738, 0x526496a6, 0x5abfbf3c, 0x17709b70, 0xbce8ffe9, 0xbaf3e3ee,
+	0x3a4051bf, 0x21917fb8, 0x6ec5eae4, 0x41fe0b35, 0x09d567cb, 0xd3d088a1,
+	0x1b4b057e, 0x3e760696, 0x3b4f6e36, 0xeaf1045c, 0x6cde3e21, 0x0c1a4a42,
+	0x845bed3d, 0xef2957e3, 0x35dede27, 0x2ed1b887, 0xffcf2f16, 0xe463e419,
+	0x91b921c7, 0x46c1afee, 0x31e416f1, 0xfe8dc583, 0x3037fe90, 0xdc2159fd,
+	0x2e5047ce, 0xa4040722, 0x573e8bfb, 0xec55d242, 0x33025b30, 0x62f0ced1,
+	0xd9a60360, 0xce043c8b, 0x412f6961, 0xa05b7f79, 0xdf0fbf1c, 0xd1942edb,
+	0x176ce4bc, 0x6774d3e6, 0x1f7e0960, 0x3fafaf58, 0x644baf7c, 0xd4225cf5,
+	0xc3e0367b, 0xea05cf7a, 0x7cf5bed0, 0x3ff301a0, 0xcd31040f, 0x95f3626b,
+	0x2fe82a96, 0x11a045fc, 0x18fed1eb, 0x41427aff, 0x2f7c3cfe, 0x79b797ab,
+	0x2dd6084c, 0x777e7939, 0x197fe639, 0x377f6108, 0xb5fde0ec, 0xb30b9412,
+	0xb3afdd17, 0xce96b863, 0x030fc9d1, 0xe5752beb, 0x3f063aea, 0x6f5750b1,
+	0x5890def8, 0xfdf30056, 0xe4bbee91, 0xffe0890c, 0x5cbcdfcb, 0x3b0dcfd7,
+	0x15eae8d6, 0xae89feec, 0x4ddec47c, 0x75bbbcba, 0xeeaf2ebb, 0xca65f54f,
+	0x5679e9d8, 0x4205f975, 0x2fdb3f28, 0xf4375e79, 0x075cb722, 0x76bca05e,
+	0x69f87c2d, 0x5c7083c0, 0x4840f545, 0xb3fcb833, 0x3493ec1a, 0x27d838ff,
+	0xecc82fd1, 0x3b734f54, 0x3e40dad5, 0xe0267f7e, 0xc0bafcc6, 0x0c6fcc18,
+	0xcffcc24c, 0xdfcc5e02, 0x09bf3123, 0xf00da9e0, 0xda3359f4, 0xe9671087,
+	0x525b5d7f, 0x923f2043, 0x32f1c68c, 0xa2e637e4, 0x3f709a5d, 0xb3a70543,
+	0x955687b3, 0x7122a6ac, 0x288beb55, 0x9939511f, 0x4e9adc80, 0xefd6bd79,
+	0x037f2d77, 0xfef58395, 0x0721b987, 0x258aadf0, 0x0a29c993, 0xcfcda3ef,
+	0xfae9da39, 0x432c762b, 0x3abcbb7d, 0x6d20bb28, 0xd3fd067e, 0x0dc4d5e5,
+	0xb6e547c7, 0x4b9eff6d, 0x99f55dc7, 0x0f80da40, 0x0107214b, 0x91b376bf,
+	0xde58ae72, 0x009cfcdc, 0xd93b34f3, 0xee9d98be, 0x0a79e208, 0xc5710f71,
+	0x519cdaf0, 0x7f6e7e87, 0xe825f47a, 0x3c87dcdf, 0x2ee92bfb, 0xe50d1a41,
+	0x8a30c46d, 0xf7d80cd1, 0x78582f1f, 0xe791ec04, 0x80b7201f, 0x6bf1e73f,
+	0xe74ecbcc, 0x4eeded05, 0x93bb0b06, 0xf36393fb, 0x1d96ff42, 0x9b1b95e6,
+	0xe5e89d97, 0x52bccd62, 0x772854a4, 0x4ece780a, 0xf04f1824, 0x79fa6df4,
+	0xb2f30d3d, 0x44af3df5, 0xebc02f3a, 0xc4af0e44, 0xfd42f09e, 0x5e0e3134,
+	0x6bc37d89, 0x912bcc28, 0xf4230578, 0xebc18333, 0x79fa2999, 0x780d733d,
+	0x4179d2a5, 0xaf0e54fb, 0xc2f09ed4, 0xbc1c6a7c, 0xd786fb52, 0xd2979858,
+	0x0df1d45f, 0x8e8b60cb, 0xefc0d82f, 0xfb5729a7, 0xd2ca8c73, 0xfddf4f7d,
+	0x005150b0, 0xdd40d3f2, 0x3789a4f7, 0x5329f2e8, 0x4ffaea46, 0x9756319b,
+	0x58a078cf, 0x3b9acf97, 0x7fbed759, 0x795d34f5, 0x191103fa, 0xcb9a8bf4,
+	0xd6f9e3db, 0x7c2f523d, 0x0728cbab, 0x35d7cec1, 0x4163c99c, 0xbd63ca81,
+	0x3e8baff0, 0x5b17db86, 0xff436f87, 0x39789fea, 0x1fb40bc7, 0xe0f03d83,
+	0x55f7f450, 0x13ddfa73, 0xf3be4668, 0x7c8c204a, 0x6fb71f68, 0x8738801d,
+	0x0381823c, 0xd378df68, 0x82788215, 0x19e6dfea, 0xfceb3ce0, 0xfc13329d,
+	0x8be69dae, 0xefc8230e, 0x3771d452, 0xa94107cd, 0x8905b881, 0x1f6dea4c,
+	0xb7f38dff, 0x9ecf07e5, 0x1d2f7e84, 0xb44cbb34, 0x56af09df, 0xf9a7643c,
+	0xdd98f7da, 0xfbfe2fa4, 0x63f7c92b, 0x97e62706, 0x93f97da9, 0x5f8e8411,
+	0x00d13bb8, 0x180aa7fc, 0x07f82a44, 0x1f41b3db, 0x05ec791a, 0xfd47fc1d,
+	0xfedd65f6, 0x230b1e14, 0x43f752df, 0x7ee84f3d, 0x8538c034, 0x56e3e07d,
+	0xe225db89, 0x067f74f9, 0xa7c0b3af, 0xbfe59569, 0x9697eca3, 0xda6cdfcf,
+	0xbf9ceb0e, 0xe071d961, 0x1cb767cb, 0x0bbefe38, 0xdbe3a1a7, 0xb9435f54,
+	0xe7f36bf0, 0x53dcf07f, 0xbd7e22f1, 0x119b9755, 0x70f9dd2e, 0xbdcef6ef,
+	0x7b53d312, 0x57b7d3c2, 0x18587f42, 0x2eaa8ee3, 0x7bdf84bf, 0xd42eb1ca,
+	0xd539172f, 0xfdc9b5f9, 0x601d0f80, 0x8545d97e, 0x2fc812e1, 0xbdd1317a,
+	0xf95faa02, 0x46834bf2, 0xda345ece, 0x2af0be93, 0xddea91f6, 0xc5ea1226,
+	0x4625d23e, 0x48d2996b, 0x578a50e5, 0x2768965d, 0xf5477fe4, 0x986349fa,
+	0x12e20abd, 0x62daa59a, 0xbf36ac69, 0xfaca58b0, 0x4d42faea, 0xdfc7c7eb,
+	0x817d3127, 0x427f07ea, 0xf5e814f8, 0xe84f5506, 0xae703d6b, 0x0cec3ea3,
+	0xbbaf5af7, 0x728d75dc, 0xd2e51ae5, 0x45ffca35, 0xcf29f7e3, 0x55ea66bf,
+	0xe058bf9e, 0xd03e074a, 0x76098f45, 0xe064468e, 0x2d0d555f, 0x1adf3f0c,
+	0x8ff9009d, 0x7d5abeb9, 0x72f8c439, 0x5cd79588, 0x788108a6, 0xd426e2ca,
+	0x49163f7e, 0xe17c6f79, 0x1fd32356, 0x20ff5a6b, 0x73cf80bb, 0x7ce1a93a,
+	0x4532e6a4, 0x695f4a28, 0x1bd61346, 0xfe233466, 0xef491915, 0xad8d9f72,
+	0x8027e54f, 0xce4fc093, 0x3cc5cc6e, 0xf5f227e5, 0x8e1a93f0, 0x9f955f67,
+	0x02f30adc, 0xf3da54bc, 0xf7905e30, 0xdaeebf13, 0xa9c20746, 0xfe80ce9b,
+	0xed7e066f, 0x02a63429, 0xf71809dd, 0x4262dd58, 0xa356e439, 0x5eebed8b,
+	0xbeac4fe6, 0x03f18bbc, 0x17d38d89, 0x538fc7be, 0xe3a2d8d5, 0x16f83d4b,
+	0xff8c578f, 0xf1919f7e, 0xe2848bef, 0xfc4f4147, 0xafa2f150, 0x5b2e6b7e,
+	0xdee8f374, 0x767deeef, 0xc6c73ee0, 0x953c7053, 0xa470aa1f, 0x38dbb9f0,
+	0x627c27ba, 0x38412970, 0x48fb7276, 0xc98d5f3c, 0xc55edc0f, 0x81e3d9f9,
+	0xb5adde3f, 0x38fb5af7, 0x387156c7, 0xd1040eac, 0xf51d242e, 0x5aaec138,
+	0x6f07766f, 0x49f0c437, 0xa164d8e6, 0x9d035e62, 0x849525a7, 0x16a6d196,
+	0xe7f02b98, 0x2af8825c, 0xfe1d24d9, 0xaf3429c8, 0x288db7e3, 0x13069b27,
+	0xe957f001, 0x615f2faf, 0x6b3af164, 0x4b3bb7f2, 0x927fbce9, 0xb820fbff,
+	0x6d601dd2, 0xebfda033, 0x63bfaea2, 0xca64dcd2, 0xfcf8132f, 0x72d1b79c,
+	0xc1e6f8cd, 0xade4014e, 0x76fa89a4, 0x0cb070f1, 0x8abfeba6, 0x1c524cdd,
+	0x20a197e8, 0xc2317662, 0x612230e5, 0x272ae8dc, 0xadcb47e3, 0x95d7a40e,
+	0x63ee2239, 0xadb7b713, 0xc81b3b07, 0xfc848a52, 0xeac894cb, 0xb44109cb,
+	0x4ec199a3, 0x8a7d806b, 0xda4ede60, 0xeb009f31, 0xb8f2041e, 0x4c9fd70b,
+	0x6f412b22, 0xbedf1aef, 0x164fbe18, 0xfe02cf7d, 0x3b6cca5c, 0xd53d8029,
+	0xf40a1beb, 0x2c10c1be, 0xcffad174, 0x5085a21a, 0xfd169b2f, 0x39a58931,
+	0xd3e4d710, 0xff987ff8, 0xf437963b, 0x912ce7cc, 0x5721fce2, 0x5fcb6a86,
+	0x012f3d10, 0x129ca79d, 0xba909e50, 0x09bd4de3, 0x2bf70f10, 0xb1878f23,
+	0x2fe93d31, 0x8f4cb9ad, 0xeac721ed, 0xf70537f9, 0x79ba63f4, 0x63f2e3b9,
+	0xd8d31fa6, 0x2797fdc3, 0x0ec1f8ea, 0x1d8d03a0, 0x11e4b07d, 0xa4355fc1,
+	0xcabf7e91, 0x6dd6bf7e, 0x601a0cbf, 0xf8f1f77e, 0xb3d5220b, 0x22f78f3c,
+	0xff744bb0, 0x6cd455df, 0x593bcb3d, 0xf9873fce, 0xc0f9baf3, 0xec045361,
+	0xce1eef7b, 0xdcfb062c, 0x51e8c66b, 0x3ff5c82b, 0xfdcdf260, 0x5a331cef,
+	0xedefba69, 0x4b9076ed, 0x9cfa724f, 0xd846a24d, 0x0f8f1f47, 0xaafaed53,
+	0x55ca8c5d, 0x0f9be046, 0x7f774244, 0x3c1ee697, 0x1c2efa16, 0x8adfb397,
+	0x9ad80afe, 0xb12cec2f, 0x16bb4d12, 0x679671f7, 0x45eb34ed, 0x57ad8e7c,
+	0xd668f5c1, 0x5b15fb8b, 0x50f082af, 0xfc19fbd7, 0x787ce69f, 0x14fbe8db,
+	0xdd7e33f0, 0x5b344ad5, 0xfbf8881a, 0x5a9bc057, 0x8d272f83, 0x00c2e3ae,
+	0x5854d0df, 0xf2325b4f, 0x6ff8bebf, 0xf466fc77, 0xcdb71606, 0xf386a7f1,
+	0x86fb15fb, 0xe11d8026, 0x220daffc, 0x21e7ddc8, 0xbf7e921e, 0x81343cd8,
+	0x73bec55c, 0x2e75db8e, 0xe8095d3a, 0x911d094f, 0xbec00b4b, 0x539e4471,
+	0x9a9eb3d0, 0x91ed0f2f, 0xb20ed14b, 0xa2388647, 0x1af71e1a, 0x993f7fdf,
+	0x6379c1d1, 0x20e804b8, 0x04e0a2fa, 0xeed55eff, 0x5ed1c73d, 0x247123b2,
+	0x710d2f8b, 0x08db473f, 0x908a9fd8, 0x146f1f41, 0x91057266, 0xb0cc46da,
+	0x8f52e16b, 0x63fd34ed, 0xed57dcfc, 0x31496cbf, 0xbb720539, 0xfa0f1cec,
+	0xc32b7043, 0x1e89b2cf, 0x33f69d83, 0x7f4041b6, 0xdba72de5, 0xaca98e31,
+	0xbf409129, 0x472c8f65, 0x9b6d533f, 0xf81da5ab, 0x0f704864, 0x794fe021,
+	0x14c7f0b8, 0xb28eeb1e, 0xc3fc807d, 0xb8f32fe8, 0x6fbb1a6c, 0xd0aeb187,
+	0xa0f0f409, 0x5cf203cd, 0xbbcf0ce9, 0x205957e0, 0xfe008e45, 0x7d2ceba8,
+	0x0e0635c0, 0xd9b0d285, 0x2442a983, 0xa56be50b, 0x4d205112, 0x86648997,
+	0xffcfeb7c, 0x6c289cc2, 0xa216e9bd, 0xfebfefb0, 0x8de1e01b, 0x8edeb2cd,
+	0x6fe7d619, 0x0be7d16a, 0xcdfcfa23, 0x4ff3e96f, 0xf74941c0, 0x0c2fb5f7,
+	0x37f088c4, 0x47fe0085, 0x8588990f, 0x97eaabae, 0x2703d466, 0x9cf0d954,
+	0x7305f8f8, 0xde47f208, 0xec08b57b, 0xb4bf954b, 0x6f01b8b3, 0xaf2825f8,
+	0x811fa97a, 0x0dff7273, 0x8cb7b544, 0xff64b900, 0x3ed80ddf, 0x9fed3cfa,
+	0xe0f8ffd6, 0x947a5fcf, 0xf9f8ffb6, 0x1210497e, 0xdf87d594, 0xef59aabb,
+	0x41ba1e1f, 0x2161ff3f, 0x664abe7d, 0x2f0bfafa, 0x187e2ac8, 0xdba5df18,
+	0x17f5c695, 0x907f9697, 0x6e20fd45, 0x8182e400, 0x4455efb1, 0xbf46afd3,
+	0xe1e9c89f, 0x019dc05c, 0xc88a63e7, 0x85cfeae2, 0x7a0348de, 0x1bc768fa,
+	0xe74f4069, 0x4fd04b03, 0x6767bd55, 0xd5e79c12, 0xf0e1bfcf, 0xaedf7a40,
+	0xdc82c6a7, 0x0c21bcf4, 0xd3dfa0f3, 0xf8477be0, 0xf4598e57, 0xc91db9fb,
+	0xd17def88, 0x30a8457b, 0x7c0e7b43, 0x8ee87c55, 0x383272a3, 0x9de3181c,
+	0x0fba5dfe, 0x780d9b55, 0x7c2aa65f, 0x038677f7, 0xa2366c1d, 0x8f7aabdd,
+	0xbe3ea091, 0x7bb456ed, 0x813eed55, 0x8dfec771, 0x5e5718a5, 0x61af1124,
+	0x664cba24, 0x75ae7a40, 0xb7ec039f, 0x926e7f55, 0xefc81725, 0x78a3fdfa,
+	0x548c7180, 0x069a28f2, 0xb2cf494e, 0xa9bff022, 0x57ae7bdf, 0x5f7e3edf,
+	0x78d5915e, 0xddaf9225, 0xc97af8df, 0x37688253, 0x5cb25ea9, 0x7d500f17,
+	0xfe0651ba, 0x4a0e3f1a, 0xfa739a90, 0x017f0835, 0xeb917f3a, 0x51ce5439,
+	0xc3cfa2e8, 0x3cb106bf, 0x80952b9c, 0x1bfeb047, 0x851744c8, 0x3234d547,
+	0x032cd209, 0xe42c7bc1, 0xefefe615, 0x7a0473a0, 0x237dd8e8, 0x84bab7da,
+	0xc33f5df6, 0xb7c8edfe, 0x07119c3a, 0xb6544dc4, 0xe2ee319a, 0xf2320be3,
+	0x37e1658d, 0xdc6f101c, 0x145992f1, 0xf1f5b0a9, 0x8362e49e, 0x37e7f003,
+	0x4e19aecc, 0x571d962e, 0x7b9a2fd1, 0x00f8b78f, 0xafcf49fe, 0x273d94ff,
+	0x7d678064, 0xe98b9e32, 0xbd8575c9, 0xedaafdc5, 0x42bae452, 0xb3d3c817,
+	0x5aefdc6d, 0x72f80b9f, 0x06e197d1, 0x6054a979, 0xf4480efd, 0xed7e849e,
+	0x924bd6cc, 0x076c32de, 0x70653a7c, 0xf852762f, 0x945e25cd, 0xf28ee60b,
+	0xa48740c9, 0x96fdc59e, 0x51222449, 0xf80cb039, 0xa3cc08ea, 0x59b65efc,
+	0xb9a2f8c0, 0xe775184f, 0x69e1c400, 0x1a4e57d7, 0x124adfb1, 0xb37cb4c6,
+	0xf802bcbe, 0xa5578534, 0x95e0fcc2, 0xf0718c3b, 0x5a3ea089, 0x930bfbbd,
+	0xfd751c76, 0xfaea217f, 0xd1e5973c, 0xfd45fd4c, 0xf70129be, 0xd8c8ffd9,
+	0xe65f380c, 0xf8e0f699, 0x985cfc55, 0x7e58ea4f, 0xa00c8922, 0xeba3cfe3,
+	0x9e607382, 0x918e8242, 0x92309f2c, 0xdf93c7ad, 0x0eb03d73, 0x36f277ed,
+	0x56aaafdb, 0xd61c3221, 0xf20de743, 0x5b47ac37, 0xb3af9aaf, 0xaefd377a,
+	0x7690ffb5, 0xf7bb5fac, 0x5fae930b, 0xe7a0cf6e, 0x7dd33110, 0x37cdd758,
+	0x71b4c343, 0xe1bdaa23, 0xbac78724, 0x0267e7bb, 0x720d3dbd, 0xd8071658,
+	0x8127aa18, 0x032c676f, 0xb7c742ac, 0xc79ddcdc, 0x2ce5a2f2, 0x76b4c83f,
+	0x54fe86c2, 0x32fa310f, 0x3329ae41, 0x68438f78, 0x37c54ece, 0x29c744d4,
+	0x12d93d93, 0x1e4f75f2, 0xb09659da, 0x6474fff5, 0x5ebab0dd, 0xbf7518e4,
+	0x08e10bd6, 0x641d43c6, 0xbd7c41f2, 0x4fc62671, 0x1fb68e12, 0xdd173e07,
+	0x5a11e547, 0xc1f0a36f, 0x6de3b071, 0x5645cbbc, 0xdd067cd0, 0x1b77e01f,
+	0x502f5205, 0x4bda8cb8, 0xa3d01be5, 0xd4378ef7, 0x4cdd70d8, 0x6dd6b06e,
+	0xfa47fb03, 0x0123f943, 0xb930ef5f, 0xdd741e70, 0xe5a34d0f, 0xb983cb19,
+	0x384bff60, 0x0ede6a1a, 0x8b70efe7, 0xa74bc61b, 0x6127b1b0, 0x7fe700da,
+	0x023c2906, 0xc5319574, 0x88ec1235, 0xffac1ceb, 0xd0f0d154, 0x46539054,
+	0x28d27cb2, 0xfb8ed9e2, 0x49c716b3, 0x30499137, 0xf20ef83f, 0xb3d6ed17,
+	0xc4349107, 0x5d4275f6, 0x3e9f8c21, 0x7064a588, 0x8bcbca06, 0x9eaa7cb8,
+	0xf7f59f6f, 0xea3f0024, 0x3f01f7e2, 0x201279b3, 0x1b1c8c8e, 0x9fed6a26,
+	0x0345fd6a, 0xdf50e93c, 0xc124597f, 0x287fe464, 0x5e67fa5f, 0xff857b42,
+	0x1a824cb2, 0x9cbfec3e, 0xd777cfa4, 0xb1f6c34b, 0x4a12167a, 0x839eadee,
+	0x11fd7484, 0x7f2d7f7d, 0xec24f5e5, 0x77698c8f, 0xe9d91c80, 0xdb3e476a,
+	0x46831215, 0x1174f7aa, 0x9dc1f182, 0x6d5df64e, 0x265339df, 0xa2679411,
+	0x40474fd8, 0xb2630881, 0xf413f5a9, 0x1eb7561f, 0x35ef8129, 0x7ce30da4,
+	0x1256c26b, 0xdbffc5ec, 0x970b0011, 0x3f29fce6, 0xfdd4ed8a, 0x9037b588,
+	0xb38fd690, 0x52222da2, 0xeab75a5a, 0xc00a2bd7, 0xd3947704, 0x2b37f04d,
+	0x72d1b73c, 0x23a3a883, 0xba6ea8eb, 0x275f3242, 0x061d181a, 0xe262bb7f,
+	0xaf3c9a6d, 0xf83edddf, 0x291309bb, 0x602a8ddd, 0xf55d9fed, 0x972c6fef,
+	0xae807c62, 0xd76bdb3f, 0xbcb895e4, 0x6b87e68d, 0xf2b8cef2, 0x8cf2b894,
+	0x567f7cb8, 0xc91ec3bf, 0x1ca9b836, 0x13f7eab7, 0x3a4fca24, 0xa9b32332,
+	0x309e4f04, 0x226133bc, 0x6a8f8dc4, 0x5a79c0f3, 0x59b82adb, 0x5f830fb8,
+	0xb9e19bad, 0x832eddcd, 0xaa3adf7f, 0x7688b8ed, 0x7209c12a, 0x27bc2da6,
+	0xe7687982, 0x330bb4d2, 0xfc3aa4fd, 0xb3ebb601, 0xe9117cff, 0x87fe4d7c,
+	0x765aef58, 0x8faf9274, 0x6e2bef83, 0x3be0b770, 0x92bd026a, 0x77bf9144,
+	0x992ff9ec, 0x7c633fdd, 0xf3d333ab, 0x43adf206, 0xb7cb5382, 0xa41f30fc,
+	0x0e5c65e7, 0x935c7062, 0xb886517f, 0xb787f78d, 0xa9c703be, 0xcb27d175,
+	0x73fb1e40, 0x20d1d9cd, 0x1f2d809f, 0x217af4f8, 0xd6f18664, 0x7861a18b,
+	0xc866adff, 0x7a05d111, 0xf016fb7c, 0x75574417, 0xda4ffc22, 0x284007db,
+	0x9755c5fb, 0x7db53e59, 0x5dc7ec0b, 0x009b7f0d, 0x0e4f1ef7, 0xd2201f68,
+	0x8355a5fc, 0x487e6227, 0xf7c816fb, 0x3f2c1c53, 0x81807dbc, 0x0921cfb6,
+	0x76ff6113, 0x8abf193a, 0x3967bfe7, 0xf7e755ff, 0x29cf7f4e, 0x66a90b80,
+	0x1c0b7dfc, 0xed9f94cf, 0x73b538ec, 0x9d9fdd17, 0x9fc8cc4b, 0x10bc6429,
+	0xe1ce938f, 0xf20ee78c, 0x8df50953, 0x926b384c, 0xcf68fb62, 0x7f21736e,
+	0x0f6ea1be, 0xb3e9d79c, 0xff3f900b, 0xac99ec87, 0xfdd2c6a4, 0xfad29a36,
+	0xe3271593, 0xed1106a7, 0xda8fe99e, 0xda78fe51, 0x0561d6cc, 0x1b534afe,
+	0x7c2b8fdf, 0x4c4c571d, 0xdf24b91f, 0xdacdfd81, 0x950cf946, 0x240966db,
+	0x309c80a8, 0x81394655, 0x3341a3aa, 0xdc5537cd, 0xddd27180, 0x335ad2fa,
+	0xd0a9bfce, 0x26760993, 0x9a693f55, 0xd44cc9ea, 0xdb4d65c8, 0x67cab958,
+	0x5e69729b, 0x12102e73, 0xad9779c6, 0xbb424dcf, 0xf3eec4bc, 0x743b5fac,
+	0x573c1afb, 0x9b3ae1a7, 0x68664312, 0x2fffc2a7, 0xf0c93dfd, 0xbef57efd,
+	0x7507df28, 0xd30ac9dc, 0xec0cca87, 0xf1527e3c, 0x44cb3ae1, 0x39a208cf,
+	0x4d03279f, 0xa448e068, 0xfdf0d1f6, 0x8c7843ff, 0xa3fdf586, 0xf1daf8f0,
+	0x84fc71c5, 0xa0bb8df2, 0x67259aff, 0xf30cdc79, 0x5df0a9bf, 0xe98c442f,
+	0x77b06f8c, 0xd077e804, 0x3e4c0abb, 0xe753b42f, 0x1ad1fb3d, 0xf5d4d794,
+	0x60787f58, 0x9d99dc12, 0xc5fe795d, 0xecfdb49d, 0xf775591f, 0xa4be5581,
+	0xae46bde2, 0x11eec618, 0x4fbb29ab, 0xe80a7208, 0x3aeaed77, 0xc4264f24,
+	0x18afaf8b, 0x609e8228, 0x828ff805, 0xb7432efb, 0x9c18132f, 0x9caa9e60,
+	0x08bb8843, 0x2953b3d6, 0xf97d86ae, 0xa05e2952, 0xf071b2a4, 0xa75e097c,
+	0x03d78dfd, 0x7aeae779, 0xfc72610d, 0xefa604a6, 0xd0f10249, 0x8dffcdf9,
+	0x811adb71, 0x13e21f03, 0xb0bb30b6, 0xc357972a, 0x79f74ee9, 0xe7e3973e,
+	0xc17e3973, 0x1deea306, 0x3e908a82, 0xbcd89a4f, 0x523eb9aa, 0x9fb0da45,
+	0xbd7d66aa, 0xd82faa6a, 0x83bf701d, 0x8690f1ca, 0xf9eae5f5, 0xc2f96fe1,
+	0xa0f77fa4, 0xa361f711, 0x09f1c999, 0x8fb93d23, 0xb5fc7f01, 0xffcde511,
+	0xc2a2fbe6, 0xf1624061, 0x8b5f00c3, 0x9ff84441, 0xec394916, 0x95c0676f,
+	0x93307831, 0x02f5a76f, 0x9aaf35bf, 0x361f01cb, 0xffcea22b, 0x64628f78,
+	0xcce5c37e, 0x4fbd1e9e, 0xc8133b46, 0xffbd5ac1, 0xe7d54539, 0x10ff72d7,
+	0x129a33d3, 0xa9bedc99, 0xe9077049, 0x2b4d277c, 0x5ff800af, 0xc021699c,
+	0x3d72e76a, 0x746a42f3, 0xfaadc031, 0x76bb4bbe, 0x7c37d011, 0x74be022f,
+	0xc7ecd18a, 0x967fd172, 0x0fb3fe00, 0x7cef8c43, 0xfdc0eb48, 0x9dab6098,
+	0x4f1355c5, 0xefda3ce6, 0x6ee9d5a9, 0xe3533b88, 0x641d4e49, 0x46f5cbfe,
+	0xf3dc096f, 0xfa63ee0d, 0x574687c3, 0xeb23dc36, 0xd011caf2, 0xd744cbbb,
+	0x7b2025d6, 0xaaf6b61d, 0x5a9ff85f, 0xfeae7180, 0x23dfa5ca, 0xfdd3d157,
+	0xbd3d7a2d, 0x61b34493, 0x07ca9b5d, 0xfc742386, 0x6e61d6ce, 0x40cad666,
+	0x93ad4be3, 0xd54d3dd5, 0x7187bc3a, 0xbdf088c8, 0xff91fede, 0xf0fc7ed1,
+	0x684bf06f, 0xef681fed, 0x7a43dbd0, 0xfabed9e6, 0x7e674d5f, 0xeff2b892,
+	0xb093becb, 0xf93fb50f, 0x0825c9eb, 0x37d228d7, 0x3e7a6469, 0xa144779d,
+	0xb46f4fed, 0xdda77f23, 0xf9fdbde1, 0xbe053654, 0xd194b393, 0xf486f55b,
+	0x4837606d, 0x57eac89c, 0x768a186f, 0xfedcb184, 0xea40df3b, 0x133ba015,
+	0xa1390069, 0x696ae4cc, 0x9c63c4d7, 0x5bd267b9, 0xbcb45f40, 0x0dabca1e,
+	0x82fcdeec, 0x8da2b6b6, 0xecc22f30, 0x02cde5c3, 0x78c53881, 0xdad115fb,
+	0x3560a889, 0x80cdbc36, 0x55d83f4a, 0x55df63f2, 0xbf79ee3a, 0x3d26efe5,
+	0xbf2a82dc, 0x7abd31f1, 0x9ff83a70, 0x901c7a54, 0x48cc95e8, 0xe074ae07,
+	0x3e0c67a7, 0xe02c97b4, 0xb82b1bf7, 0x478eaf9e, 0x3be30df9, 0x281fb70a,
+	0x2d70aecc, 0x47daa3fe, 0x0f403b54, 0xf294b3be, 0xf37d119d, 0x0aea421d,
+	0x80f84ed1, 0xd2740dcd, 0xfe70ff83, 0xba53eea6, 0xd8d0f8cc, 0xd10321a5,
+	0x6497b9c8, 0xcd3697cf, 0xa3ffe2b9, 0x826eb8a2, 0xb40c97bc, 0xbe365c81,
+	0xf70a91d2, 0xcebf98dc, 0xdb3c46da, 0x662df7e8, 0xfed02f2a, 0x90b871ee,
+	0x0cc8647f, 0x97928c0f, 0xa2f2d214, 0xfb9436e2, 0x09c4a9a3, 0xc6decefc,
+	0xa762dffa, 0x3b4246ed, 0xdbba77c6, 0xa2dea42b, 0x2afc5f69, 0xe9725fbb,
+	0xdfed1da0, 0x734b1297, 0xbe748f80, 0xa77e476a, 0xba53bbfa, 0x8958d5df,
+	0x1e1db9eb, 0xd5854a45, 0x3d03f715, 0xfb93088b, 0xd86277b1, 0xf46358b9,
+	0xe9ff00e5, 0x9644d778, 0xe08b7ed3, 0xbc54a231, 0xef0257b7, 0x2ebed536,
+	0xca486400, 0x8216772b, 0x5864d91d, 0xf7428f74, 0x34b0c4a7, 0x576d08fa,
+	0x04e6c033, 0x9cc4c293, 0x7de7fd3f, 0x5ed1ff34, 0xaba7de23, 0x266139d8,
+	0xa2fad1d8, 0xa2fbf8ff, 0x7975bf74, 0xbcbabfba, 0xbf9f45bf, 0x814f6cd7,
+	0x36942cfb, 0xd8cf70dd, 0xb73baa64, 0xf4d7ce8d, 0x354dc99a, 0xd9b37211,
+	0xe81980f7, 0x071e8c5d, 0x3a729978, 0xe17c8046, 0xc044fd6f, 0x559250f7,
+	0x072f80ff, 0x9992e7e8, 0xf3efb3c1, 0x9eec289b, 0x6967551d, 0xc36d94da,
+	0x597a68fd, 0x6f8c0908, 0x045d194f, 0xf7e130e7, 0x97183c4f, 0x265367e5,
+	0x4cfefc4d, 0x68437ed3, 0xded79c19, 0x5c27bc72, 0x856bcf4b, 0xb87177e6,
+	0x8b47d557, 0x53445ee0, 0x3061a4d2, 0xc90be92f, 0x0abbfb0b, 0x17c8d13a,
+	0x3fe2cdcb, 0x115a3f76, 0x2d89efe2, 0x7b873a07, 0xf8e9e39b, 0x0cf8bee3,
+	0xe8532ee3, 0xc9cfc030, 0xee376656, 0xfda21652, 0x43ae6fda, 0x44b8d5bf,
+	0x60b5d709, 0xa0242d80, 0x9f20991d, 0x3932260c, 0xcb41611a, 0xffb986cf,
+	0x0b4e0e3a, 0x9ab930b6, 0xbbe033ec, 0xc1400b31, 0xd5fbb902, 0x700f18a9,
+	0xc9a7f376, 0x5b6e01e3, 0xbb0270ef, 0xccbdf5a2, 0x8369e78f, 0xada4bbf7,
+	0xbd0172df, 0x6262db1f, 0x765448f3, 0xfb8f7ec1, 0x837280c2, 0xc81f1224,
+	0x67b24d0d, 0x6b91d018, 0x8012cef3, 0x59a9d9eb, 0x517fd622, 0x37184e20,
+	0x7c1a4971, 0x5e48205f, 0xfdfa3f46, 0xa26bf753, 0x05c99939, 0x7f782b86,
+	0x4db4d8fa, 0x3e3dcf18, 0xc6f7f367, 0x8d3b7f42, 0xd2846f78, 0x3e703dc7,
+	0xc0f5d0d4, 0xbcb66a7c, 0xb38cc693, 0x628a4484, 0xa3fcdbef, 0xaab26074,
+	0x07975c78, 0xf2c16c7c, 0x3bed3cba, 0x78f03d1e, 0x1b9ba533, 0x166078a9,
+	0xaf2097c8, 0x35ea4ff7, 0x1e5e293e, 0xc54af555, 0x6d3b3c4d, 0x08f3d768,
+	0x329d3f94, 0x4e47c84b, 0x27ac6599, 0xf3616dfd, 0x3f105cef, 0xd02dfbeb,
+	0x60f10e5f, 0xea2b853c, 0x0ee2a62c, 0x69cf0080, 0x9fc599b2, 0xf51f73d1,
+	0x44a9f5a0, 0xffd1aef7, 0x3a51fde5, 0xb2f0077e, 0xc72bddf3, 0xd9b5ae01,
+	0x8f40231d, 0x0cfe5ff0, 0x56b65eed, 0xb9b4f766, 0x0d9e64bd, 0x992a7be8,
+	0x83e3105b, 0xcfe3377e, 0xd1baebf1, 0x63f16462, 0xc7ec5129, 0xd7f4656a,
+	0x4df6f5c4, 0xf7a82c4b, 0x99094a6d, 0x955fd261, 0x573f54ce, 0x2a63b705,
+	0x7b415e3b, 0x4e09090e, 0x2e296bf0, 0xe55d2746, 0x3fcf11b3, 0x882831da,
+	0xd1e3bdf9, 0xf1db65af, 0x8040c35c, 0x83f30b53, 0x0eeddcd9, 0x3b47df58,
+	0x7d60394e, 0xb80c3be5, 0x27d0f2ee, 0x620af369, 0xf049bd74, 0xfaf3c15d,
+	0x46b59d0a, 0xc768fb4f, 0x6d7369a7, 0x730a4f4c, 0x706f417b, 0x2f1952f9,
+	0xc65c2858, 0x751d9839, 0xf7c453dc, 0x386de232, 0x4f6f6b5b, 0x6af3b0b5,
+	0x0dc9859c, 0x7ad0d6e3, 0x7c618788, 0x1ebe29cf, 0x57a32079, 0xf8fc6d3b,
+	0xb4441be7, 0x26ac56a3, 0xff3d6768, 0xf46c81fc, 0x903cee6b, 0x5f53473d,
+	0x6f8c3f8d, 0xf1616d9f, 0xa7f1a41f, 0xe20d251e, 0x7378d83d, 0x2007cabc,
+	0x3b86909f, 0x1e38fa3c, 0xbd4d130a, 0xa7c40dae, 0xefd58a0c, 0x01dcff2a,
+	0x5a736a71, 0xb413110d, 0xa70678f7, 0xebcbbc80, 0x2f77f7c7, 0x45a123c7,
+	0x5ac93d40, 0xfa3e4510, 0x1134f9b5, 0x4d3e5a6c, 0xe324f7f7, 0xf0e3d1d3,
+	0x19d813ee, 0x6b933ce1, 0xf5c8126d, 0x5e182b69, 0xa2207f45, 0xc435ddfc,
+	0xf9522f52, 0xb0cace45, 0xb14aff01, 0x01acb8b0, 0xbf58f959, 0x1a679efb,
+	0x4999e7bb, 0x8a7e30ea, 0x1f51db38, 0x38863fee, 0x580d0e41, 0x41911c98,
+	0x830c1bdc, 0x7a3c9820, 0xf10151ea, 0x8e7c03c4, 0xa9ea0a88, 0x3f185f14,
+	0xfa33e477, 0xf952f0e6, 0xa02ad0a8, 0x92db371c, 0x862bf5c8, 0xa3c2d757,
+	0x3c62ffd9, 0x778ea05a, 0x12e478bd, 0x5be1f5c8, 0x1fc2761f, 0x845fb828,
+	0xf0a771fb, 0x3628a67e, 0xb880af7c, 0xf1dd1d8d, 0xcd9fb464, 0xfd04de54,
+	0xf62bb541, 0x62ea515b, 0x76b6e07e, 0xdface790, 0x7c5cf052, 0x05cb6c5c,
+	0xcede3a79, 0x11367748, 0x533801ce, 0x1f94bd61, 0x5797fb30, 0x9d71fbc6,
+	0xe21b8f01, 0x44880607, 0xcfdf3e02, 0x35ef8b90, 0xdd380bf6, 0x59dc182b,
+	0xbb1e631c, 0xb12b5c0f, 0x075942fb, 0x1772c912, 0x5f4cc7c8, 0xcb8eface,
+	0x78b52e0d, 0x225346ff, 0x7ea987b3, 0x69ecbef7, 0x9f4f41a4, 0xcbad5daa,
+	0x537e174e, 0x0659ce0c, 0x85049f3e, 0x41ee4fdf, 0x35c00ae7, 0xff510c6d,
+	0x4b3e4036, 0x14fe37cd, 0xae957bcf, 0xbd9d7b55, 0x736b9bc0, 0x7abb9213,
+	0x8be031dd, 0xb8e7efd7, 0xfdd5e57f, 0x025dff1c, 0x9e7a0e7e, 0xb4c39776,
+	0xe6ab39e9, 0xae12203d, 0x21cac7b2, 0x11695ece, 0x2575fb78, 0xde5c4877,
+	0x3ee7f1c2, 0x3caaef06, 0xf1ba266d, 0xe293f3b4, 0x629087fa, 0xdfad4b9b,
+	0x5bfa1138, 0x83e79237, 0x5f83d41f, 0xbea369e7, 0x6f29043f, 0x029a990f,
+	0xc9f07fd6, 0x1a9fc741, 0xe1ef3b1c, 0xa97807fc, 0xf7aafee7, 0x44de34a6,
+	0x3f3ab2c7, 0xf50778c5, 0xf4a7f6fc, 0x24e22f7e, 0xff5fe3e7, 0x761e841d,
+	0x261693ca, 0xc90096df, 0xf9003d51, 0x17a97fb2, 0x9ca825e3, 0xb7cc21f1,
+	0x338eddf3, 0xe7796847, 0xf78fcfb4, 0x7d6e3c03, 0x31e83678, 0xfa03b7e0,
+	0x501b37f1, 0x055a55ce, 0x462d6f86, 0x7f1517a4, 0x7f1c9d2a, 0x7f788312,
+	0x3090e9b0, 0x5445ec1a, 0x9bbecd3f, 0xd5b165ee, 0x31c70173, 0x533fdfea,
+	0xeccad953, 0xe2fcfe56, 0xffd0798d, 0xf9c0810e, 0xc3ff26bf, 0xbfa8b9b4,
+	0x4f36907c, 0x65a2aed3, 0x691f8b1d, 0x470af636, 0xfa606ad9, 0x1e7dc1dc,
+	0x43e70be0, 0xca4b1618, 0x906ad0a5, 0x45cbcb57, 0x8546efd4, 0xbe43ef4a,
+	0x2026533d, 0x5c587787, 0xd3cd1e6c, 0xe42f1083, 0xbf81ffa7, 0x7e77936a,
+	0x3347ca29, 0xd5c41b15, 0xce2fcf9e, 0x5795fb44, 0x061c0276, 0x926537bc,
+	0x149ef80b, 0xadb4b8c6, 0x2091b6b5, 0x83efadd7, 0x63693ecf, 0x782f5cfb,
+	0x7e67df83, 0x5bb5a271, 0xf930b4e5, 0xe4cac32e, 0x867bf541, 0xb79dc995,
+	0x0026db05, 0xda961c3a, 0xfaf720ad, 0xd013e789, 0xe6d7894b, 0x0f7089da,
+	0x129ddea3, 0x4d3da170, 0xfc7f6949, 0x4dbf2826, 0x223b38c2, 0xe3033fb5,
+	0x894eae76, 0xfc850bdd, 0x6f680523, 0x1beacc7b, 0x14e3a562, 0xf3b27971,
+	0x10859de0, 0xe1ce38d7, 0xe824c9fa, 0xed0973e7, 0x8a9ee33a, 0xf8004db6,
+	0xd9903e7d, 0x08e895b9, 0x9376b9ed, 0x9e93ad95, 0x92ef286f, 0xfbf237c3,
+	0x1d9eb40a, 0x5a16f1ba, 0xb8de4153, 0xfd18f458, 0x24694c6c, 0x7d549f80,
+	0x9282bb40, 0x872025e2, 0x4a96084c, 0x271cb718, 0x71e04c87, 0xf4a4881d,
+	0xfbc7317b, 0xe1777421, 0x558a2dfd, 0xc7265ff5, 0xaeb792e4, 0xa3e090cc,
+	0x8a88af2a, 0x1399ed0f, 0xc6085a67, 0x1e8cafd3, 0x9e2ec117, 0x0a7af942,
+	0xf160cbde, 0xb32ff9a2, 0x466120fb, 0x3ffeddb7, 0xa8fd1da9, 0x7e56ad93,
+	0xfc33d7ae, 0xb7186d21, 0xb0900940, 0xd6cf68fd, 0xfd5057de, 0xf27f6a86,
+	0xad8f6672, 0x7c17f6e1, 0xdf9696f9, 0xb2ed556d, 0x467e75cb, 0xf0bedc75,
+	0xc6aac2ab, 0x804b4453, 0xd507a31e, 0x4a4f8a78, 0x8cfa2f00, 0x5778e0a7,
+	0x78f572db, 0xd195bf0a, 0x4f1d37c9, 0x913588f1, 0x11f44bc5, 0x471f114f,
+	0x911999d1, 0xda4b5ed0, 0x7d04eeb7, 0x7e84c5bc, 0x6f2d29db, 0xcbce11fe,
+	0x43090a29, 0x5cbc56dc, 0x5f4d35e1, 0x75def80e, 0xc056cce1, 0xef7775c3,
+	0xb03f3377, 0xd7885e5e, 0x7cabb60f, 0x1b73cb97, 0x1fa7d7ad, 0x29bef5a8,
+	0xcbe54eb9, 0x5187ae7e, 0xb9c072fa, 0x23d383b3, 0xd1c41250, 0xdd6dcbc5,
+	0x1c9c7ef2, 0xd19eff11, 0x9d6f69a4, 0x61a47bc1, 0xa87100b3, 0xde0919af,
+	0x63ee0dd3, 0x85b56dfa, 0xc3ea93f1, 0x2ca9fdec, 0xe3ed1793, 0xf22315e4,
+	0xf9e57f0d, 0xfd114497, 0xdc00af04, 0x27e5a653, 0x5faefd40, 0xd50e901b,
+	0x85b690f7, 0x2a74fc98, 0x94c88c5f, 0x79c1b7ce, 0xa13ce30c, 0x3d741bf7,
+	0x31f18fd9, 0x5ae837ee, 0x9706fccd, 0x6033e3cc, 0x330ecaf3, 0xb7197ffb,
+	0xf14aafff, 0xae8332fd, 0xc64ff16b, 0x1e75dda3, 0xd65a61f1, 0xc71bb7a0,
+	0xbfccbdec, 0x78af0554, 0xcc05fbb3, 0x8fb1a7df, 0xae523bc1, 0xce18a906,
+	0x5ce25ca5, 0x15d97c0e, 0x16fdd0a4, 0xbbf9eb4d, 0x802882c3, 0x4ead213c,
+	0xac7ec0b1, 0x012717aa, 0x45eb05bd, 0x5bb88f7c, 0xd7e7e00e, 0x29ffce25,
+	0x8e3042c8, 0x77f3d69a, 0x96fed57b, 0xa9023111, 0x9574b028, 0xdbd49a53,
+	0x57f6a8b9, 0xc653ac6f, 0x4034be83, 0x25ecbb44, 0xe01333d9, 0x8f29297a,
+	0xbd645c40, 0x1261aebd, 0xbf1c472b, 0xb17a8e64, 0x1312dbfc, 0x092b88d8,
+	0xfcdbd9f8, 0xdb878edc, 0x3838c6bc, 0xe3f3d9c6, 0x88cb3ee3, 0xa0e6686f,
+	0x6f92719e, 0x69733e73, 0xbfbf2e4d, 0xaeaee0c0, 0x067f597f, 0x4b007137,
+	0x9e280e07, 0x303ffb09, 0xa04d8651, 0xd223b329, 0x07fce095, 0xef053b7e,
+	0xb7cfc513, 0xc023841e, 0xbdecd5ab, 0xff3aaf8f, 0xddff79f8, 0xcce9c69a,
+	0x5f780193, 0x3df194e6, 0x17d96baf, 0x705cb972, 0xd6a1fe2d, 0x07f806fb,
+	0x1b578bd3, 0x13adbc78, 0x97bc08f9, 0x04a77f08, 0x9e08aef7, 0x79d5a1df,
+	0x06d5aeb0, 0xc27e1f7c, 0xef9e1b74, 0xfbd987b5, 0xdb4b3576, 0xad9ce1b3,
+	0xf7c76e2d, 0x0de1e36b, 0xfe5495b3, 0x49cdeb43, 0x1c6ef821, 0x79fbd069,
+	0x3cb3fa7a, 0xd3e03645, 0x37aec8c6, 0x654338e9, 0xd7c858df, 0x02ee8d2e,
+	0x91e3b579, 0x5e404af7, 0xd26f539b, 0x2792a7c5, 0xd2227ef4, 0x65768490,
+	0x5effa4d2, 0x5dae39be, 0x2351bf0f, 0xd97ae3e2, 0x7be80c37, 0xd3f9839c,
+	0xef88a549, 0x01db75fc, 0xe1ce1a9e, 0xeed3cee7, 0xee7d03ef, 0xef67ad2f,
+	0xf77aef59, 0x2fec02f7, 0xbe6ad7dd, 0x7df617bf, 0xb7f60f2b, 0xf2b7cf63,
+	0x1bd77f60, 0x3f403d1b, 0x0f9f153b, 0xefa73cfc, 0xa39ca2f0, 0x0b5c45c6,
+	0x275d172f, 0xe6fa2e5e, 0xe7aaf2f0, 0xcaa6c05a, 0x5ced5f49, 0x9ead3f83,
+	0x218af87d, 0x34bf4668, 0x3d98ffc3, 0xf077fcec, 0x0a2cb95c, 0xabe8f7e0,
+	0xae5c51bf, 0x104850cc, 0xc54513ec, 0xe8aed8ed, 0xdc809173, 0xf179caa4,
+	0x9f09f870, 0x4f36f3ff, 0x5f879c02, 0xce5aeed7, 0x4fa128eb, 0x43b9fc99,
+	0x6bbb3370, 0x1dd5c598, 0x9d82752a, 0xd9885cf5, 0xb1f76b5d, 0x422bea77,
+	0x5df0446f, 0x09c156f3, 0x8ec541fa, 0xeef01c0f, 0xae4e2690, 0x448ce819,
+	0x9deceff5, 0x40729d0b, 0xb79d1bbc, 0x6b78e415, 0x7a06544a, 0xf1172da8,
+	0x295ab3dd, 0xb5bcf18a, 0x241dc429, 0xf5ea5f20, 0x6c2e352d, 0xba465793,
+	0xdecad4d9, 0x7d934ca1, 0x0f11fdb9, 0x6dfbe3fb, 0xb7edfd2a, 0xf800fe79,
+	0x7ad3f31b, 0x8b576c5a, 0xc32ff91b, 0x617fe636, 0xccd97fe5, 0x6e3502e4,
+	0xce8ff544, 0xf3eef1e7, 0xf83bbc79, 0xf81a236d, 0x9b56df82, 0xd5b6fd57,
+	0x88322f10, 0xe5b56df9, 0x8d687edd, 0xae5b56df, 0xe889178d, 0x93234df8,
+	0x310b0d77, 0x25d2864e, 0x579163c6, 0x88ced505, 0x74a7e9f4, 0x0c6464b0,
+	0xf0fdd57f, 0x12b87f97, 0x3cfcf3e2, 0xf01ab2ee, 0x61c60477, 0x5fc04c46,
+	0x355f14c3, 0x4b6a2efc, 0x7d7374e4, 0x37e8c8f6, 0xedf1df4c, 0xe3c7573f,
+	0xb3aead74, 0x02ad5d7b, 0x75b8c3a3, 0xf189ebc5, 0xcf5569f6, 0x4bdf16bb,
+	0xdf1f5fb6, 0xabfdeb53, 0xfdb65ef9, 0xbef7c39f, 0x47fffb29, 0x3034c6ae,
+	0xe87b3e05, 0xd083df87, 0xad8312a4, 0x4c9477fb, 0xae3fe1d7, 0x437062d9,
+	0x21cd4ddc, 0xaaf21bbb, 0x7fde323f, 0x94f64958, 0xfde3c800, 0x5710d588,
+	0xfedae831, 0xb77b62c4, 0xfb573d5f, 0xb151785e, 0x61bf5f9e, 0xa7d934fb,
+	0x02e2d3c1, 0x82f7e02a, 0x24148ab7, 0xa9f638da, 0x7eb7e676, 0x9fefbf83,
+	0x88069321, 0x6eda7893, 0x14467bc5, 0xa62e306b, 0x6e9dead8, 0x9afc0d1f,
+	0x83d9efc1, 0x0b8812f7, 0xc618d67e, 0x1ac13efb, 0x5727de1b, 0xff6a23ed,
+	0xa97ea96f, 0xcd73e06b, 0xcd73e275, 0xf1cf8c37, 0xe3ab17ed, 0x61dfaf5d,
+	0x06918503, 0xb3f1e3df, 0xf17d76af, 0xef765581, 0x91d23b72, 0x5eb9ef07,
+	0x11e8ff2f, 0xf1b92987, 0x4dc41d5e, 0xfc559c42, 0x74f7a87e, 0x6b70b954,
+	0x97a1fb9f, 0xc57eb707, 0x3c96b8fd, 0x0de2a4f8, 0x8f4f7ac1, 0x3e81e4f7,
+	0xeccecabb, 0x0ccd9b2e, 0x0c0b372f, 0x13e07b1f, 0xe2a3efbf, 0xf5fb01d5,
+	0x7ad43bd9, 0x78fe6aff, 0x2d7c2ed4, 0x69fc077b, 0x886c90f1, 0x7d767be2,
+	0xfe9ec7f2, 0x5977162e, 0x517a86dc, 0xef8090f4, 0xf6f3ab33, 0xd63b31b7,
+	0xd45c39ef, 0x1cb8fbef, 0xfcf0e398, 0xf10a9ec1, 0x8f028a49, 0x59028c3b,
+	0x2f28e40e, 0x0da9378b, 0xda8fb8b3, 0x70e304a5, 0xb511fbd4, 0xc809532d,
+	0xbecd0b6d, 0x55fb433f, 0x8682d9ef, 0x4fa3b004, 0x75a72023, 0x37f2cbbe,
+	0x401f7d9b, 0x3b697fb9, 0xa69bfdda, 0xa138c79d, 0x5d3f2fa2, 0x3a5df609,
+	0xa40bb45f, 0xd3f7f498, 0xc7c60e1d, 0x50ead4d4, 0x7b41ebe1, 0xfb950778,
+	0xa7f2de40, 0x5f609c78, 0x56cded50, 0xe1da1b57, 0x6732f76b, 0xad81915f,
+	0x9fda3c87, 0x8ef6d1f9, 0xe415fc2f, 0x9042e6c3, 0xb7de8b57, 0xe3a8fda1,
+	0x138c2c91, 0xbbeacc78, 0x1f6af80c, 0xe40bba15, 0x0e3a1ccb, 0x7b1dadc8,
+	0x79842aee, 0xe0a17d77, 0xccecbaba, 0x24f5d0bf, 0xcaf88999, 0x665dd1ed,
+	0x9f022786, 0x666ad76f, 0xba1e0a9f, 0xdd03bf91, 0x38368f6d, 0xd1ecc746,
+	0xa5407cf1, 0x08fbeeff, 0x1ff414b3, 0xbd61fcab, 0xeaa39d91, 0xeece77f9,
+	0x3c16f1e2, 0x2a75aeae, 0x79e0d78b, 0xf8554f1b, 0xe3bb6f3e, 0x9cf3ac1c,
+	0xdec7fa5d, 0xe24cbec3, 0x2fd8e7fd, 0xe36939c4, 0xdfe0add5, 0x53897b01,
+	0x37b2bc46, 0xa1dfe610, 0xf9f20578, 0x09b6d16d, 0x839d7409, 0x01b7229b,
+	0x165f0fbc, 0x53a0bbd9, 0x92f78b7f, 0xde360413, 0x57da27a0, 0x8afb050f,
+	0xc36fedf3, 0x9d1abad6, 0x5b92e0c5, 0x870ef668, 0xd52241b8, 0xc578c3e5,
+	0x3c5918ef, 0xda2b21df, 0xabb29a09, 0xbb691273, 0x684cf7ce, 0x03e789af,
+	0xcb241a6a, 0x606f8e31, 0x0ed01fcb, 0xe82bdf6f, 0x93a6a6e0, 0xf7e6361c,
+	0x6095b609, 0x296dc9ac, 0x0e6e77ec, 0x7f041fbd, 0x5f1ab4bd, 0x46f52486,
+	0x7572088f, 0x7b411970, 0xa72690e4, 0xc395fb08, 0x0eba44e3, 0x1f08cfff,
+	0xddaa9f9e, 0x64a9843d, 0x7a095f6c, 0xfdacb495, 0xc7ec20f6, 0xb6b144de,
+	0xd68bfa00, 0x12b27c32, 0x5a739f38, 0x95bf2d06, 0x9c1a73e6, 0x7aea3fe8,
+	0x0de30ab5, 0x439d7483, 0xa167597a, 0xdce5c583, 0xa520e3c8, 0xfc115643,
+	0xb85ad2d2, 0x5341440d, 0xd29671ef, 0x129c43f6, 0xc38e7de2, 0xd4f00b37,
+	0x5902f9ca, 0x9c7f4162, 0x6d78625d, 0x8bdef25f, 0x55bfcf10, 0x1d4fef66,
+	0x34e37c3d, 0x96addfb4, 0xc6ab5ed0, 0x74e02e51, 0x47de13dc, 0xe0725e31,
+	0xbf303fa1, 0xfa0b1d0d, 0x3f0c8c97, 0xf829d5a4, 0xc87fe07e, 0xfd0b8c3e,
+	0x80c47d76, 0xe9e728fe, 0xaeb92667, 0xebc4b7ad, 0x891ccb29, 0x622e295f,
+	0x3d9be399, 0x3fdbe24f, 0x55f78eae, 0xd7a210f5, 0x63e87acf, 0x5c7572f1,
+	0x0ee4befb, 0xf542ee35, 0x504f1177, 0xcc526f17, 0xbe8a21c7, 0xef0b6abf,
+	0x50a3c78f, 0xf205c439, 0xd70ada4e, 0x8ebf8832, 0xdfa098fe, 0x09978d46,
+	0x9738b316, 0xbde0a8a0, 0x88d20cab, 0x48fd3271, 0x1ed01044, 0xd3b6e2c6,
+	0xb6865e21, 0x8818ef5b, 0xee216ffd, 0x41f20306, 0xa4abbe17, 0xfdd055c6,
+	0xd3243a41, 0xc13f1e64, 0x566b37f0, 0x70e2085a, 0x9f7edcfd, 0x7f1f2b35,
+	0x3f135567, 0x33feecc1, 0x7d4f882d, 0x10aa53de, 0xd6253df9, 0x1710d36f,
+	0x4d7ff3de, 0x693e6bb5, 0xe2e3ed67, 0xaa38542c, 0xfc2a1671, 0xf14c7bd4,
+	0x79314c70, 0x57bc02df, 0x7dd425e3, 0x0d8fdeee, 0x55d82ecc, 0x1cbbe221,
+	0x155d7f6a, 0xdc83679f, 0x626adc50, 0xefd712f8, 0xaefbeea6, 0x67c63321,
+	0x96bd7df3, 0xdfbeaa1c, 0x23bf9475, 0xd7bf8c75, 0xc21665de, 0x399f8006,
+	0xc9381b3a, 0xee4784f6, 0x5d817b84, 0xdc3668f6, 0x78dcdc2b, 0xa7cc24ff,
+	0xfe424f46, 0xc3ff84b5, 0x5ba243f2, 0x89e19ee1, 0x1e58b3a7, 0x908b2e70,
+	0x49848e33, 0xe07f82c1, 0xe3d73a5c, 0x022e1a77, 0xfb486b1c, 0xd3b436ca,
+	0x146973f0, 0xa26463e3, 0x44ecc429, 0x13ffdc2e, 0x631dec99, 0xbdc0fe5b,
+	0xdfde506d, 0x5a2df684, 0x627bf88e, 0xee5678c8, 0xfa4dfb62, 0x029c59db,
+	0x13d33dff, 0x1fd1e3b9, 0x4df8035e, 0x462b85e5, 0xc740989f, 0xe1b420bf,
+	0xee27593b, 0xe00161f8, 0x8ffd85eb, 0xe2b9060e, 0x77f7b94e, 0x72731ee1,
+	0x7ec8c239, 0xa34de5f7, 0xf9409acf, 0xb3ad1fc3, 0xff69124b, 0xcbc79333,
+	0xe0ef3089, 0x51eef8f3, 0xc4bfdf2a, 0x4bee4c19, 0x2015e89c, 0x3363f17e,
+	0x1f54b3a3, 0xe2063dfb, 0x3ddc219b, 0x12d7235c, 0xc4113bf6, 0x7b804ae7,
+	0x68eff108, 0x6b5e12dc, 0xf7f73900, 0x780ce881, 0xfefd5552, 0x38f7a107,
+	0xc604828b, 0x5c19a6c7, 0x141f01c3, 0x29cdf1b1, 0x2bc6c4b7, 0xfbf17048,
+	0x70a375c5, 0x51ba86dc, 0x28f4c7a0, 0xd072d1b7, 0xe5a09e93, 0x97297bc3,
+	0xd67afb78, 0x938ef35e, 0xa974e01b, 0x2fbc03f4, 0x04eb38b4, 0xc27647be,
+	0x0ef03c48, 0xfb93304e, 0x0739e8f5, 0x273c7dfb, 0x3b4fc591, 0x40704c37,
+	0xffdb107e, 0xfd0848bb, 0xb06653bc, 0x21df597b, 0xb4edf56d, 0xfa199dde,
+	0xcbb963df, 0x277fcfa5, 0xc29f9f5f, 0xebbe07e3, 0x1eff7fd6, 0x0f804a89,
+	0xffa31fdf, 0x441797dd, 0x0be0094f, 0x2f8c27e5, 0x2aee3c0a, 0xf9c20845,
+	0x2c81e37e, 0xc61a9ef5, 0xf5c0bdb3, 0x7fbf3f11, 0x91aa6701, 0x8586e3ae,
+	0x1bdc9dec, 0x93928833, 0xf92bfc04, 0xf07bbe08, 0xa729f711, 0xf4789bc0,
+	0xd713a97b, 0xf419e2af, 0x1f7c3b75, 0xe2e0cf9d, 0xd2538321, 0x418eefd8,
+	0x6b2309d9, 0x84b1fc1b, 0x7d684e7c, 0x3e2c6ff3, 0xf184bc62, 0xd7ab0277,
+	0x4ff86665, 0xfefc524f, 0xe21ede53, 0x27c89caa, 0x4abfdc2d, 0x8be85df0,
+	0x78d8d2a3, 0xfa8b0f1c, 0x5bc01629, 0x7cbb5df5, 0x1bf6e127, 0x7069a64a,
+	0xe794103e, 0xfd54bc54, 0x3c380348, 0x7f327e28, 0x1f872671, 0x06f7f091,
+	0xcfda8932, 0xadc61374, 0x04bafcf1, 0xde2a6f46, 0xb5ec7fae, 0x5f00a737,
+	0xc03f5c12, 0x3bb2ee7e, 0x74f9bd61, 0x804999ac, 0x78d829bf, 0xf449e707,
+	0x247927bb, 0x8c5d8798, 0xb09a1b18, 0xbf7f307c, 0x4fa4fa6c, 0xfe1a4866,
+	0x61adfb45, 0xfdf55afd, 0x459ba3c2, 0x68337e9a, 0x7e9a49bf, 0xf2c2689f,
+	0xbfe5fbbe, 0xe7e92147, 0xaf8fb6a7, 0xf3f5ea4f, 0x7c7dabec, 0x14f7f53d,
+	0x4df87326, 0xf329e356, 0x640f71f5, 0x7e615b8c, 0xc39c7d6c, 0x8edf1c9b,
+	0x306dff5b, 0x09250be8, 0x2844a70f, 0x1a524c3d, 0xf408dbeb, 0xd0fa461b,
+	0x802ff5d0, 0x7e652e2e, 0xb6fbbf80, 0xd989e078, 0x97a519d0, 0x7ba309dd,
+	0xe7c547ce, 0x3c5179ee, 0xf1543c06, 0xdd9e7e38, 0x30bdfc78, 0xa16d7206,
+	0xaaa407f2, 0x9d48423c, 0x0d70eb80, 0x2283ad88, 0x87528961, 0x22b37ff3,
+	0x33bdf388, 0xff8c19f1, 0x3be49319, 0x799af302, 0x3dec27d3, 0xa1d28c93,
+	0xf9d10720, 0xfd484d1e, 0x355f377f, 0xe1dc796e, 0xbd82373b, 0xe3ee9bb4,
+	0x5fed5ce8, 0xf234d395, 0x2af73b15, 0xf2be5545, 0x19642900, 0xe57cb832,
+	0xacd678c2, 0xf10dcb80, 0x8af91aaa, 0x44d547c9, 0xeece07bd, 0xa060fced,
+	0x987abe5e, 0xb76ebc60, 0x271a557c, 0xede3cd5f, 0xc42092a7, 0x93d936c5,
+	0x805e2c42, 0x162d39ce, 0x8db4dbff, 0xb74d8dc9, 0x82419cb8, 0x33d31672,
+	0xe87bf166, 0x6bde4cb9, 0x77c98fba, 0xbaea1ce3, 0x6b78fa2c, 0x133567ee,
+	0xb8b12f2a, 0x99abb40c, 0xa5d97d9d, 0xb53576f4, 0xf202ffde, 0x007f0832,
+	0x00007f08, 0x00088b1f, 0x00000000, 0x7de5ff00, 0xd5547c09, 0x73b9f8b9,
+	0x64cacb67, 0x109848df, 0x424e3b08, 0x875b3612, 0xe22948b0, 0x3cb888b0,
+	0x4240b21c, 0x3eb44196, 0xc33fedad, 0x0d220222, 0xc168d46d, 0x2a14180e,
+	0x0431a0d8, 0xa4587049, 0x141a87d0, 0x2f1f682d, 0x48145840, 0xad88a0c6,
+	0xbefbffcb, 0xef726e73, 0xf6b42264, 0xfa7fb6ff, 0xef7397b3, 0x6df3be59,
+	0x39ce5be7, 0xd78deec3, 0x1ec658b1, 0xacc630b4, 0x98eb458c, 0xb19436b3,
+	0x96eff0ef, 0x95e5e7ae, 0x6289e63a, 0xa31574ac, 0x47d7e5e7, 0x7a83cfa6,
+	0x4c8c7697, 0x66e783cf, 0x92d433b3, 0x50e158cd, 0x2fa18a7b, 0x7b46f963,
+	0x9b19933a, 0xbeb45e64, 0xcc9eba19, 0xd393f516, 0x5b09fb18, 0xcf074c74,
+	0x48ce9151, 0xc8673fac, 0x287a2967, 0x0379f8b3, 0xfd8c611c, 0xc28f675e,
+	0x19b98cf7, 0x718535c2, 0x4398a6b8, 0xf870f2bd, 0xa5c340f7, 0xfe8c8196,
+	0x1c7183be, 0x4b6f6726, 0x1155630c, 0x67971fb5, 0xa35cf631, 0xe2b6e6c9,
+	0xec1496d7, 0x11deb18f, 0x043086e7, 0x543ce185, 0x39e814f0, 0x4cb2d13c,
+	0x040bf4f0, 0xc65e630f, 0x68e0ba72, 0x94ae8437, 0xb1e0a97a, 0x00d17eb3,
+	0xc7be24a7, 0x06895ab1, 0x6a5383f3, 0x1257e3fc, 0x33d38fd4, 0x63265877,
+	0xea97981b, 0x1ce75efd, 0xb3c6e381, 0xb8e1894a, 0x112b6c2b, 0x96e0dd4f,
+	0xbf30c901, 0x1ec977d9, 0x0ae60c13, 0x7ace1f09, 0xd4fb3d61, 0xba7f1804,
+	0x2af67f18, 0xac0884c7, 0x78e25d37, 0x00ffc2f0, 0xfbe219fe, 0xa9b2c6f4,
+	0x3198e00c, 0x13ba70c1, 0xa66ff83e, 0x0181ba65, 0xc2a6b39c, 0x7bd7737b,
+	0x61d2209c, 0x2cffd28e, 0x6c39bb6d, 0x387267cd, 0x10191696, 0xd5ee735f,
+	0x09669f7e, 0x42173e8b, 0x077c076d, 0x60603af3, 0x79f68f34, 0xe609b3cc,
+	0x82cc56f1, 0x3347f7f5, 0x43b7e036, 0x7fe86533, 0xcd716c62, 0xf4bf283a,
+	0xbf2a60dc, 0x83cf7b15, 0xf3fbf147, 0xa09e54e9, 0x6e504683, 0xb3ce99ff,
+	0x9cd01529, 0x5494d773, 0x9040ff18, 0x467ef7e8, 0x9d83e3ef, 0x5ea1d355,
+	0x99bef0fc, 0x04d3e0cb, 0x0aede323, 0xde9ff7f9, 0x9ce54614, 0x191eacfb,
+	0x6c109cba, 0x30f7d375, 0xaa09b6c9, 0xd130c91e, 0x53e361f7, 0xccbef7f9,
+	0xe7df851e, 0x23fb5185, 0x0b5e2199, 0x47f2f78c, 0x01fbe883, 0x0eb9b3e9,
+	0x595b5120, 0x52e91ea8, 0xe336689e, 0xc940c873, 0x3df0083b, 0x5f8c5221,
+	0x353de7c6, 0xca226b73, 0xdf5c84cd, 0x6850c75c, 0x670e6897, 0xfe1f1137,
+	0x8879f90c, 0x34aadc6d, 0x5f16afe4, 0x150f9a8a, 0xf8574f3a, 0xcf96826f,
+	0xc51e7e40, 0xfa9f0af6, 0x87c2a7f3, 0xea0b3e1a, 0xb97f33e2, 0xd7009e57,
+	0xb18f924a, 0x97721f20, 0xe38e747c, 0xe3fe2727, 0xf8bf24f6, 0xdca2c527,
+	0xc2777f88, 0xd0f24b57, 0xb9e9b99b, 0xcb83e501, 0x8708b3cb, 0xc56cd751,
+	0xdb247df7, 0x33c20b60, 0x0c7666ca, 0x0a372eeb, 0x783ac97c, 0x2e1dd991,
+	0x3943d3fc, 0x45fe7589, 0xb8e0e747, 0xb3b43bfc, 0xb82ec8b3, 0xa4adfc60,
+	0x7870569f, 0xb88b4836, 0xd09763bc, 0x5768b94f, 0x41b7ccae, 0x0678df3b,
+	0x2ffa8778, 0xc4967ec4, 0xfe18dddb, 0x82b87995, 0x4f81ee41, 0x611d9966,
+	0x1dd39846, 0xb724f911, 0xebf73277, 0xb38f8011, 0xc2136706, 0xae643c77,
+	0xed7e1191, 0xf757a146, 0x74285f20, 0xc7f4e052, 0x7a0f794f, 0xfe1e7fcd,
+	0x57a2e978, 0x133e0273, 0x2b2a22d8, 0x79e1bdf3, 0xae5e0ff6, 0x69bdf983,
+	0x147c9df0, 0x6fcdebe4, 0xeb3736cc, 0x20cdf382, 0xbc3d12eb, 0xd992ff77,
+	0xb3e00736, 0x051f8168, 0xb32e772e, 0xcebeb19b, 0x8a9659ec, 0xe06af674,
+	0xcfe88fca, 0x7f174c58, 0x5eeb31c5, 0x2afb3154, 0x3d56b6ec, 0x57a487ad,
+	0xcf50c2d8, 0xab9f6866, 0x99c4e5e7, 0x50a59eb1, 0xdce8f76e, 0x3c6f1806,
+	0x347375e8, 0xefe74fbe, 0x38230b59, 0xc20b3147, 0x90011559, 0xbbaaf7c2,
+	0xd1c03dd2, 0xd95bfeac, 0x7ddf041d, 0x0fa4c1c5, 0xc16032aa, 0x83efc2cb,
+	0x896efbe9, 0xd62ee38f, 0xd355b8d1, 0xeaa8feb9, 0x53b8f16f, 0x54ceb271,
+	0x8daaf7e4, 0xd04884f6, 0x5b328923, 0xa47a0b54, 0xf0ca152a, 0xeec65d8c,
+	0xf4c47d43, 0xc3d3e1f6, 0x78fbf0ba, 0x11deb8d8, 0x84b71fc0, 0x96fe75f3,
+	0x2af9ccb5, 0x29b563d6, 0xd57587d3, 0x00008a8b, 0xd4305eb0, 0x596efe83,
+	0x6286f50d, 0xac2f507f, 0x83763d7b, 0x68a7c476, 0xac608c47, 0x9f11d858,
+	0x8299dd6e, 0x9b27c476, 0xa9e4a677, 0xd8945fea, 0xbfd71c1c, 0xbc656f7d,
+	0x66ec67ca, 0xa0926f18, 0xf98724fb, 0x50eb8494, 0xe77b44fc, 0x1a7c85fa,
+	0x0235bf61, 0xef9061f5, 0xb47ba445, 0xf90e77d1, 0x6261bd8a, 0xa3ae199b,
+	0x1a5d72be, 0xf7876faf, 0x1bc8efe5, 0xdb7e0357, 0xdbfe3862, 0xbe041976,
+	0x2dff9e5b, 0x0c34af94, 0x3c61ab60, 0x0231bc17, 0xd71cbbff, 0x853d702a,
+	0xbfce333f, 0x57112854, 0x55d98983, 0xfc1b60eb, 0xc5d99457, 0xd04ced2f,
+	0xe88421b7, 0x1b372e5b, 0x75caff91, 0x70875d92, 0x59ec1c2c, 0xa7e97526,
+	0xe8af4512, 0x132972d8, 0x262cede9, 0x9827bcbe, 0xfdf0966d, 0x80c5f4e9,
+	0x9af352af, 0xa18f5c66, 0xd70901de, 0xb407a41c, 0x53ed885f, 0xbcb3e3bb,
+	0xad54e00e, 0xb4dba5f7, 0xc6642f70, 0x0784efae, 0xd7cbc937, 0xc414e706,
+	0xb730596b, 0xcd4b2e43, 0xf03f0839, 0xf017b79d, 0x2649afa9, 0xc0c426be,
+	0xbe8a79f8, 0xdc7ce6a9, 0xb3c7d55e, 0x09e7be04, 0xca35e5ba, 0x0b3c135e,
+	0x5a7bbce3, 0x2d4addb0, 0xfbc74b1d, 0x3952ea0e, 0x1ea38e1e, 0x4dc152af,
+	0xa7054e7a, 0xa25cbddc, 0x7d45eef7, 0xc63f3fe0, 0xeba935be, 0x4ab689dd,
+	0xe709f236, 0xf495667b, 0x0797d066, 0x4ca15d05, 0xe62499f3, 0xf5f23466,
+	0x14ef9fdb, 0xd08eaefc, 0x8eba3971, 0xbe412dbc, 0xbcbeb98e, 0xb3fef529,
+	0xdff56de4, 0x06ec6f29, 0x8267c1d7, 0xf8083b98, 0xd8ca5ab3, 0xfab27a8b,
+	0x465b6636, 0x37ebb55f, 0x088673ac, 0xc035fd7c, 0x85987403, 0x9d7906ff,
+	0xad3f7d1d, 0xfc97607b, 0x9cd9f516, 0x776ec201, 0x2424bb70, 0x9b19f38b,
+	0x61fd3cc0, 0x1ca2610a, 0x23d9b932, 0x1eb91f03, 0x006765ce, 0xf8fb311f,
+	0x32f5f2fb, 0xe51375d9, 0xa865d8bf, 0x659431db, 0x5cc80582, 0xd53f4417,
+	0xa633f512, 0x24fee371, 0x6e4c4728, 0x7016464e, 0x6674bc3d, 0x9c0208eb,
+	0xd058b1e8, 0xdfe1f163, 0x67c5bd50, 0x7d827979, 0x57942f60, 0x799921d7,
+	0xa1c761a0, 0x2f2c7a99, 0x28b125d8, 0x104cf718, 0xd501a397, 0xc955663a,
+	0x304f20e3, 0x75f8aab3, 0x4d4896c8, 0xa26d5879, 0x0d4dfea6, 0x33df357d,
+	0xdf3583bc, 0xd4ca1c47, 0xc79bb394, 0x3c8fea68, 0x8f29ab9e, 0xa9a2996e,
+	0x0cc2f63f, 0xbe47f94d, 0x99f535bb, 0x103cdedb, 0xe82050f0, 0x8a0ff8af,
+	0x5ae5bafe, 0x2c3da69e, 0xe8095ec7, 0xb15fc85a, 0xc80b3582, 0x2d07193f,
+	0xc896fdd4, 0x8da20e6b, 0xf5a95856, 0x6b54164f, 0x1b0dc5cb, 0x5609408d,
+	0x2fbed1ec, 0x0f2da2d9, 0x43a79e88, 0x011b3abe, 0xf820c9fe, 0x428b1447,
+	0x2ec9eff9, 0x44efa7f1, 0x00758fe1, 0x7e8098df, 0x33dface9, 0x7d234737,
+	0xdf03cc0d, 0x7e05b03b, 0xf5c1be07, 0x03d0e3a3, 0x8523e43a, 0x3278fede,
+	0x1ee96bc5, 0x3dd2d564, 0xee96a064, 0xdd2d3661, 0xd2d28d7b, 0xa5aec23d,
+	0x2d64d47b, 0x5a1c63dd, 0xd1cdc7ba, 0xa9c13dd2, 0x91527ba5, 0x8bc9ee96,
+	0xf3ef74b4, 0xa9ae96b0, 0xbe5a85ee, 0xc503e3f0, 0xb95b4b4e, 0x8e9eafd8,
+	0xe0fcd4e9, 0x40ca9a28, 0xfcaff4cf, 0x7fffa6b9, 0x45a43f36, 0x92353f0a,
+	0x7e477e45, 0x645fbd86, 0x46ff7776, 0x7f6a6bd1, 0x65d39f42, 0xefa4f67f,
+	0xd3cbd9ba, 0x47a09c78, 0xdc9ad97b, 0xf5272f65, 0x982797cc, 0x4bd689bb,
+	0xf08746b6, 0x70e165dd, 0xa357c15c, 0xd59bbb19, 0x13dfb01a, 0xc0146750,
+	0x03e7027b, 0x9acbdfe3, 0xe65cfde9, 0x5993a7a3, 0x63ccf88c, 0x30167a36,
+	0xe8a73d07, 0x1e22b79c, 0x392dcd4a, 0xceaf6fa8, 0x42527a3b, 0x21448ea0,
+	0x2f23dfd0, 0x4ea75a6e, 0xd999cc57, 0xd6b18fda, 0xb44ce70a, 0x16e33ba7,
+	0x568d6676, 0xbc6f2de1, 0x7a8f5034, 0xa1b60e56, 0xa777943d, 0xde6cf644,
+	0x3739fc97, 0xfe43af92, 0x2b0f65cf, 0xfe14ab78, 0x1e43f707, 0xbb0ab48e,
+	0xc943399e, 0x7059b3be, 0xb8fe805d, 0x9182ff3d, 0x82582eec, 0x2e80d7f5,
+	0x4f4e24db, 0x4dda2ba4, 0x7cccf4c3, 0x9843ca0d, 0x27ccfc93, 0x7892a0f4,
+	0x1fb1833f, 0xa0dbda17, 0x75234a02, 0x8e394e34, 0x3cd4f007, 0xaa20cccb,
+	0xe98126ac, 0x9afb4207, 0xf12ca225, 0xbce1fe32, 0x454cdfc9, 0x4d2b6d78,
+	0x4858af64, 0x4ccd923e, 0xe3dc91c4, 0x493ef0dc, 0xed090d06, 0xbfc01bd6,
+	0x6b942488, 0x2198c3e6, 0xfaf99ce2, 0x934c157e, 0x51f8539d, 0x0a2f7c26,
+	0x1dbfdeb4, 0x9595edc9, 0x40efc715, 0xcdef297a, 0x84b3377a, 0xf7825ea2,
+	0xeb31b92a, 0xb0f11d99, 0x533dc2a2, 0xc8d9ff5d, 0xf1e7ea48, 0x4e2d8bde,
+	0xdf723ef8, 0x720428a2, 0xd9739067, 0x10a1658f, 0xe9458962, 0x069ad46f,
+	0x1ec3fb99, 0x919afd19, 0xd4e7f5c7, 0x2dd1a471, 0x0bd8f2da, 0x0f42b2da,
+	0xc177f376, 0x82650728, 0x00f14caf, 0x195eaa97, 0x67c0f63a, 0x0fdcf165,
+	0x84c1be63, 0x1b90a377, 0x0de49d53, 0xc130f902, 0x0a61925c, 0x9ccc0c2b,
+	0xedeb7480, 0x6e5da0b4, 0xd0ebd44c, 0x2ff2603f, 0xf800ce6c, 0x78a69e53,
+	0x7eb555fd, 0x17ea03f9, 0x48d86bb0, 0xdff11236, 0x41e02f09, 0x1a3a7802,
+	0x46ae36e2, 0xcdcc9c78, 0x613603cb, 0xb3fea7b9, 0x7a59becd, 0x0b6cc110,
+	0xbc3fc67f, 0x447e9674, 0xa3af7af7, 0x64e3075d, 0xd93db8eb, 0x8e5cde9d,
+	0x6bb3e93a, 0x01646c81, 0xb0d553e8, 0x7fd4e3ff, 0x25793e9c, 0x51e22ba9,
+	0x6fc092ee, 0xcd698787, 0xf8009612, 0xc75aac2e, 0x092e3f90, 0x8f325c38,
+	0x24d7b1c3, 0xb5e8a7aa, 0x10022c97, 0x875a1af3, 0x62f04407, 0x0eb72f7a,
+	0x6bd2f2e5, 0x2fae88bd, 0xf1474efe, 0xa3cc41b8, 0x610cf9c3, 0x8c3efa5e,
+	0x26f950b5, 0x67281514, 0xdbf12b30, 0xbb23e608, 0x2b5cf411, 0xd972b9ea,
+	0x2820bbf9, 0xbe764e5f, 0xe84e7a44, 0x4fc174fd, 0xaf7e8ae9, 0xddc7fbd6,
+	0xf8a48d3e, 0xad0b6f4a, 0x54ce4223, 0x3af82674, 0x7f91448e, 0xe768e209,
+	0x544db4e5, 0xc688303e, 0x587c1429, 0x8d88ec98, 0xc5ebd154, 0xf5336b8a,
+	0x7802f92a, 0x5bbf28ec, 0x4d66822a, 0xa397bd50, 0x32071e38, 0xa18fd5df,
+	0x2175f9ff, 0xe047b436, 0xf97ec68d, 0x4729c90f, 0x5b97bc01, 0x9a3af5ac,
+	0xf50ea567, 0x24efd962, 0x33c01cf6, 0x66b97180, 0x7035e2fd, 0x9848a96d,
+	0xf9f40135, 0x0fdc917b, 0x1c7cc971, 0xe28433b6, 0x8cc98531, 0x853f4385,
+	0x7c91a7af, 0x0bd0c40d, 0xcc32b1c2, 0x8c2f7683, 0xdd95151b, 0xf1ff1e1d,
+	0x35df43df, 0xb733f49b, 0x1ef543bf, 0xdd9973f4, 0x7bc7556e, 0x19bb0b56,
+	0xe39213fc, 0x3b9fd160, 0x7aff7197, 0x84fb8f30, 0x296ae6b2, 0x62f5e7ad,
+	0x9a651a20, 0x014c02f5, 0x36ce5718, 0x16ae5c93, 0x47acc472, 0xe18fe8ad,
+	0x3a184673, 0x8f0e9cfd, 0x1cfbb187, 0x24fd7a2a, 0x3f0baf8a, 0x6a78e289,
+	0x077b7337, 0x2a19f5e3, 0x31af387f, 0x251bec9f, 0xec2edfbe, 0x9503f6c9,
+	0x27e8c1b9, 0xbe388a57, 0x058a9fb0, 0xfa682d2b, 0x4ad78e4d, 0x3e91587d,
+	0xfe7de2f9, 0x535dad4a, 0xf649c5b8, 0x24ecf5cf, 0x09c514c0, 0x49a4edfd,
+	0xd96f4251, 0xefc60658, 0x3b49accf, 0x307a8c18, 0x1e305995, 0x9abf249f,
+	0xf11428b1, 0x871cace0, 0x4f543df2, 0x67afeb72, 0x1a87041e, 0xe245b9e9,
+	0xe7f774f4, 0xd92db4f1, 0xb2dcefdc, 0x7f5a63b7, 0x58cbe7e8, 0xdda52f8e,
+	0x6a06b197, 0x38e828f7, 0xd8cb5f7c, 0x6e7e40e4, 0xc977c091, 0x7a8b2096,
+	0x8dab2a7d, 0x0d5d94d0, 0x7af51609, 0x2c28f640, 0x4c146caa, 0x7e2ba721,
+	0x72d0d653, 0x768ac430, 0xca017381, 0xbfc1fb49, 0x3a250f86, 0xff5072f8,
+	0xf9c7183d, 0x02b6e113, 0x9f743b3c, 0xffd355cd, 0x80bd28d1, 0xffa7af9b,
+	0x2d765c30, 0x1e87975c, 0xe1c29d8b, 0xd5c239fa, 0x8b1d3de6, 0x8cb805f4,
+	0xe476b86a, 0x3f9ff8f2, 0xa1fc9e9e, 0x88f564f0, 0x87df35a7, 0xc0c61da7,
+	0x74126cf0, 0x6e80f97b, 0xfac027b2, 0xcfc849b4, 0x1aef0c3f, 0x6b6dfc3f,
+	0x15618d9e, 0xe1137780, 0x9635be2f, 0x7f021672, 0xe04e22b5, 0xc6fa12ce,
+	0xfd03ba06, 0x7a468fba, 0xead40f50, 0x2c54d459, 0x96d814e7, 0x46af3cc0,
+	0x5da0fb38, 0xfe21eae4, 0x7f8091f2, 0x7014a247, 0xef881ffe, 0xf5e48f7b,
+	0x89ff75c1, 0xf28626e3, 0x09ebe0af, 0x1339ad4b, 0xe6fa37fb, 0xf59cb0db,
+	0x1a1ad9b9, 0xef1d7fa0, 0xf27e57ff, 0x3caff4a1, 0xc81a8ddf, 0x01d247e3,
+	0xfb209419, 0x9ac7cdd5, 0x7f9cc3c7, 0x9d27a595, 0x79e30ffa, 0x79fe36b9,
+	0x762c7bd9, 0x778be71a, 0x4b97df51, 0x876fc367, 0xbdf24bb2, 0x97f31433,
+	0xd7cd2746, 0x6d8991f5, 0xbce7600c, 0x9942f821, 0x057f329f, 0x151fb0ee,
+	0xee0091f4, 0x5a2613b0, 0xf5103e0c, 0x97a07595, 0x5e71a3b0, 0xacdc0eb4,
+	0x6ca071e7, 0x57f7d13d, 0x1c50c6a3, 0x27c88768, 0x291b46fa, 0x870fe95e,
+	0xff8f9a21, 0x9bf7b5c4, 0x8d3f62fb, 0x33e78a06, 0x4cfc4b7d, 0xdbd717e0,
+	0x9427877a, 0x8ef266f8, 0x4763a450, 0x753c0b78, 0xa76a06d1, 0x7287bd43,
+	0x8c9e4d71, 0xc8fdae7a, 0xc55916e7, 0xccb472e7, 0xb9ac6be3, 0xbff42c79,
+	0x30599336, 0xd62a1cbe, 0xad23d0df, 0xe87f0845, 0x249b9fa8, 0x4af6fc55,
+	0xadda8994, 0x6d62fa35, 0x09ab9fd0, 0xf78a051c, 0xb1b118b7, 0x85970782,
+	0xeb3ca1e3, 0xe866e800, 0x4760d1a7, 0x2b68e578, 0x7700a7d7, 0x1e0fbc3d,
+	0x1f25b9f4, 0x78c167bf, 0x87dabb7d, 0x833d3859, 0x439e61b3, 0xb3de5f91,
+	0x97b3c014, 0xdf081cc4, 0x3f5cd183, 0xc455702b, 0xfc02fd71, 0xf114321e,
+	0x7ca3ac9a, 0x66fae3ea, 0x96e3dfc2, 0x8c25967e, 0x79c76ef0, 0x299b46f9,
+	0xed1d56ee, 0xf9c6898c, 0x1e1cd818, 0x8c766cf7, 0x32587c27, 0x07002356,
+	0xf0bd9106, 0xc38bc133, 0xb21cf358, 0x168cb20b, 0xf594475c, 0x8ede8bf9,
+	0xcd37685e, 0xd076e68d, 0xd742b7f9, 0xc0abbae0, 0x2c02b677, 0xb957633a,
+	0x8498e6d0, 0x9ccf9978, 0xe78e502b, 0x551e0157, 0xc0de7efc, 0x16bc2dfb,
+	0xaa51bccc, 0x36eea376, 0x6bc2bb62, 0x760bcce1, 0x5e20f9f7, 0x5f187cef,
+	0xf18ac6cf, 0x33fc156f, 0xea07fa33, 0x7c9f9b31, 0x8bf3f86d, 0x8d8f5c79,
+	0x4995bc45, 0x7bb4462d, 0xabf98ed7, 0xb3e9ed11, 0x4de61615, 0x4679beff,
+	0x9c6aaf31, 0xccd77ace, 0x3ca51f34, 0x5b24f917, 0xc250df3d, 0xf88ad9c3,
+	0x4931e0eb, 0xfeaec783, 0xd8f06afb, 0x1a74ffbd, 0xc0c56cff, 0xa43ff4eb,
+	0x1aadfe87, 0xedfabbf8, 0xedb6e347, 0x04297bcb, 0x779f7bf8, 0x31bb73e9,
+	0x64c7bb1e, 0x373ca131, 0x3f22e6d4, 0xbc0edd5e, 0xd0f4fc04, 0xff28af9e,
+	0xa326d0f8, 0x3768c70d, 0x02dae6f8, 0xcc7eefe8, 0xea8dd6e2, 0xa746167a,
+	0xdffd9d11, 0x0b1ef59b, 0x03cd95c6, 0x375c16f5, 0xb3d9accc, 0x4cd7f894,
+	0x3f8944fa, 0x8b4efd28, 0xccfa3592, 0x5fd9f425, 0x9047cef5, 0x5f3aedde,
+	0x89da1e7d, 0xb70f3f03, 0x41117f92, 0x3f12ba76, 0x892383d0, 0x3989ad76,
+	0xde4de618, 0xa3ba758e, 0x82dfbfa1, 0x411c62b8, 0xc4e3561e, 0x2293fb78,
+	0x5064ffee, 0xa54dbc73, 0xeb45d697, 0xb09d987f, 0xb35f8fd2, 0xcbd17c3f,
+	0xc47583d7, 0x2bf372af, 0xb58bbf08, 0xc55c7918, 0x5e67f64a, 0x94e3c1d6,
+	0xacfc520f, 0x05fbfe95, 0xbfd02392, 0xe11de252, 0x9ae2ace8, 0xd68c36bd,
+	0x5e10fceb, 0x28b7ffce, 0x8a819fca, 0x6ac43d26, 0x162f9e65, 0x310e9671,
+	0xdaf37687, 0xda1ff414, 0xa3ab1e45, 0x46e83f71, 0x4fecae78, 0xf37fcf74,
+	0x9f6869d1, 0x79e33e83, 0x5a30a6ab, 0x624df227, 0x9106279c, 0x9123b17f,
+	0x7ab57fdb, 0xcaa6786f, 0x77bc7c1e, 0x685f736e, 0x7f656977, 0x5dcbda1d,
+	0x0efbcdfa, 0x39dae7f6, 0x574df888, 0x39e9e83f, 0xd7489718, 0xdcebb0d9,
+	0x31eaaffa, 0x16b650de, 0xef1801ec, 0xa469e812, 0xc4feae93, 0x705f5c70,
+	0xcb3f5c6c, 0x9fc893ea, 0xd1f10fcc, 0x60b293fe, 0xea05ba5b, 0x4d8fd7e5,
+	0xf33fbe47, 0x5c14a0e4, 0x58c74e0f, 0xdc4f3053, 0x192024fe, 0x69ba79e6,
+	0x6abf7f00, 0x17d470e4, 0x9c436b61, 0xadfc009e, 0x95fcf7e8, 0x14767226,
+	0xcc48ef3a, 0x7c406b68, 0x7e93320c, 0xf6450507, 0xdf21fe53, 0x17786b0c,
+	0x473f159e, 0xa15ac7c1, 0x3ffe954f, 0x79517a16, 0xd299fac2, 0x31ef7082,
+	0x1ccffd11, 0xc38d7bd1, 0xf1b21ed1, 0xd08556f1, 0x0730f978, 0x29cbc80b,
+	0x8e9d8eb1, 0xbc470eb4, 0x2832ebe7, 0xe443ba0f, 0x7c5956e5, 0xe41eb8e1,
+	0x194ad9eb, 0xbe62afea, 0xf8d508f0, 0xd7b28df9, 0xfbddf448, 0x4d5ff90e,
+	0xabe90c61, 0xfc8c337b, 0xa69bd773, 0x018f684d, 0x6821ddfe, 0x91371e17,
+	0xc9afb2df, 0xfb617644, 0xd7f21089, 0x42c3f1fc, 0x3f48297d, 0x44499bd7,
+	0xdf64f6df, 0x0c8edcf3, 0x680f9862, 0x891cea3f, 0xd1694274, 0x9ec3e3c5,
+	0xd5cd7680, 0xda2f20a6, 0x60f4f035, 0x6acdd07c, 0x9efde9e2, 0x347e8610,
+	0xa8643d3c, 0xe0f17549, 0x1082a0b3, 0xfda99bed, 0x01ce2943, 0x4a1f1fa8,
+	0xb143fe83, 0xbbd52b7d, 0x9379d7da, 0x1221c785, 0xd48978c5, 0x5fb7326f,
+	0x8d39f499, 0xef89fd03, 0x65ddf4cd, 0x79a48e72, 0x36c61c7c, 0x61679806,
+	0xa0bcb6a5, 0xf62e5b5a, 0x5b25cb68, 0x745f65b4, 0x814fd114, 0x7f73e15b,
+	0x70c4a4bc, 0x87c142bb, 0xd2769f3f, 0x61f03af3, 0xcbf1013c, 0x571d391a,
+	0x929aba0f, 0x0f2de387, 0x9753e4d6, 0x3e279d34, 0x9fd8376c, 0x77c075c1,
+	0x9136c0d4, 0xd50360fe, 0xf8ffb6fe, 0xe7adf3dd, 0xbcf5be45, 0x26dadf26,
+	0x160c2ef4, 0x5084f18e, 0xe11c359b, 0xad4ba1d8, 0x3d5f22f5, 0xbe7c4419,
+	0x21cbf716, 0x401baddf, 0x054452f0, 0xf0b3cfc6, 0xfd08cff1, 0x54c5e677,
+	0x642d33ca, 0x4b4fa4b5, 0x44bef399, 0x5887e78c, 0x57ff9073, 0xcdfb02d9,
+	0x27baf8fe, 0x34e3d386, 0x7d8ac6b3, 0x6fcc564d, 0x80bfe114, 0x74fd4504,
+	0xdfbe20f6, 0xd175f48f, 0xd481137a, 0xd8ccf7c9, 0xf3ff9655, 0x0d0077cb,
+	0x3655fe11, 0x67ae3877, 0xde7c2c17, 0x91593f60, 0x9f3245be, 0x8875fa07,
+	0x2f120bc7, 0xd942fa2b, 0xdf00b12f, 0x989e7ccb, 0x518f2fa7, 0xc7ae38fc,
+	0xf75e925c, 0xd693e830, 0xf2386b08, 0xb4aaeb53, 0xb79825c8, 0x66c6b088,
+	0x3aecf0aa, 0xd026bcc0, 0x0946d65f, 0xf0ddbf63, 0xf39223b5, 0x085d6fa3,
+	0x8eaf2bad, 0xf6ab996b, 0x8635ff13, 0xc58f91db, 0x16a6ff70, 0x40ec826d,
+	0x5a74e078, 0xcab4e820, 0xf931fce6, 0x16f57e4b, 0xf412f754, 0xb95f7bf3,
+	0x2f295ffb, 0x0fc55eb5, 0x8a4edeb0, 0x385de3e5, 0x374b7464, 0xfb4a5f91,
+	0x82eb7c9a, 0x70bf6ec8, 0xae3a4a68, 0x82eb4543, 0x155f5122, 0x8b135e1d,
+	0x8db8c8f6, 0x3ae2c58f, 0x6269b055, 0x9fe2533a, 0x482a3e2a, 0x3da3a7b1,
+	0xb6567be4, 0xd5453f60, 0x4ab5c132, 0x167e695f, 0x36ab9fde, 0x24549d90,
+	0xdbd73ae0, 0x2f788e19, 0x11bd33d2, 0xef4e7ef0, 0xf03b270c, 0x2ffc9959,
+	0xbd40f9ff, 0x00ccfa63, 0xfa0b3278, 0x00b48be2, 0x47547bfc, 0x66082a74,
+	0x36e75724, 0xb9edfcf2, 0x587de764, 0x157f056e, 0xaaad77f7, 0x0eb70e2c,
+	0x73f78eed, 0xddf12766, 0x051f3102, 0x28e3feaf, 0xbc476d78, 0xe871fda2,
+	0x2768c9b9, 0x91929eb8, 0xc6ded933, 0x8e47e42b, 0x72789a6b, 0x075fbe2c,
+	0xd7158076, 0xdcefbec7, 0x4ad4eb83, 0x909da199, 0xcb1694d7, 0x57d76a86,
+	0xdef22fb4, 0xa3ae159e, 0x90d31df7, 0x6d5b4e9f, 0xfb2a99b1, 0xcfc11dee,
+	0x5fc974fe, 0xec70b7f7, 0x818d68af, 0xdff057e0, 0xfd93630e, 0x41e21953,
+	0xedef6f3e, 0x7a157ea3, 0x2116fd28, 0x81fe15be, 0x2c63577a, 0x7c795f82,
+	0x99d535b1, 0x0fc00b63, 0xe3037ff3, 0x61d57f62, 0xdcbfcc76, 0xbccab8e8,
+	0xc8aae93b, 0xafd399ad, 0x8a54f9f5, 0xea8eeadf, 0xe9e99369, 0x50535bca,
+	0xfd61f2de, 0x5f3a7f46, 0xfe449df2, 0xe315b285, 0x4aebf9d5, 0xf2f1c3ee,
+	0xf874984f, 0xadcf75bd, 0x3f2e09c3, 0x529e01fa, 0x7b75f102, 0xa6c52faa,
+	0xfbac1fc2, 0x7ccacf24, 0x1c137054, 0xe4307fe0, 0xa054f6ff, 0xbfee4a9b,
+	0xf50f7210, 0x63e2462d, 0x5fc470fb, 0x7e2d7ce7, 0xcdf8b5f3, 0x3e2318f7,
+	0xcd273666, 0x9323e0a7, 0xfc17df0e, 0x28e9a7fa, 0xc7abad9f, 0x235abcd3,
+	0x8e51c3ed, 0xffa0535a, 0x59e7dda1, 0xe7986bf0, 0x6799efd0, 0xabd11065,
+	0xfa4679c7, 0x33df679e, 0xed58c8cf, 0xadeacf3c, 0x5c5fa8e1, 0xfb8664bb,
+	0x8ec95cc3, 0xd97963b5, 0x768614b2, 0x05d0f589, 0xd5e912fa, 0xb8a068d6,
+	0x30ef22a1, 0x83f20dde, 0x35a1f88e, 0x28f4fad7, 0x0b656f5a, 0x575bd9c6,
+	0x51e7788f, 0x95aeb728, 0xb407682d, 0x81fc72e6, 0x0f50ebdc, 0xa638a6b4,
+	0xc18f7951, 0x39fa14b2, 0x0fb7f953, 0x0c0bf225, 0x73f34656, 0xbe3b45ac,
+	0xfd45e01d, 0xcd997805, 0xcd034aed, 0xa7cad57f, 0x7777e256, 0xfe964df5,
+	0xa7eb40d6, 0x6b45cf16, 0x321c8d64, 0x73c695bd, 0x8ac93afd, 0x0cf050f3,
+	0xc67d349a, 0x607c2eb0, 0x2eb0d679, 0xfd94b8bc, 0xeb3d626c, 0x9f885bd4,
+	0xc05d7cbe, 0xafa8ea7d, 0xa77d3f70, 0x57affb3c, 0xdfc67a7e, 0xd04db89e,
+	0x16ef5fd3, 0x7ca9cbab, 0x5cbaa15f, 0x4a37a7e2, 0xbac3befd, 0xfc502d3b,
+	0x78e77774, 0x02796f1c, 0xaf5c4a6f, 0xd68949fd, 0xd5bd3cd5, 0xabc17e88,
+	0x155c78e3, 0x8ff301fc, 0x683b0e49, 0xb67a793e, 0xef8994dc, 0x994f546d,
+	0xdcb7f427, 0x8ef46b04, 0xa077ae60, 0xaaa364de, 0x0557e8ac, 0xfe44479e,
+	0xc3fe7942, 0xad64f3fe, 0xb38aacff, 0x201b123f, 0xf7299ece, 0xc608ed7d,
+	0x64c976db, 0xf30eba37, 0x7d7d8575, 0xf06de9e5, 0x1a4f7cbe, 0xe6555bed,
+	0xf6c78fef, 0x7de12b86, 0x21db7de1, 0xcbc63063, 0x72a31af2, 0xf48a2fc9,
+	0xaffbc405, 0xfe94fcc4, 0xe12df4da, 0x2df6d5d3, 0xa19f096b, 0x86bc8fe2,
+	0x164fb42b, 0x5596e7f0, 0x534cfb7a, 0x79fa2e4c, 0xcc4a7ff9, 0x55f3c1d5,
+	0x06b5cf41, 0x506b0beb, 0xc899764e, 0x3c3b6f6f, 0xdbed8747, 0xe789170e,
+	0x9e7a7643, 0xe382933d, 0x7f38eb87, 0x14bde32b, 0x7ca5edcb, 0x7fe7a9dd,
+	0x8dfefa84, 0xaa65d4b8, 0x2ebe69f8, 0x6dc4fc93, 0x845cdf06, 0x95f1ec1b,
+	0xe943efb8, 0x2b58d5e7, 0xebb6d1c1, 0x0fa8f3fc, 0xce4dacce, 0xb6d23544,
+	0xcc70c06b, 0x189db7fb, 0x3fe78e91, 0xb04c9eda, 0x66d81ea1, 0xfc443eb0,
+	0x86cc2ab6, 0x87f4ae72, 0xac7e60e3, 0x1cc8d76d, 0x4edbb3f4, 0x4ca17870,
+	0x73d4377e, 0xa50efee3, 0xf59d70df, 0x8c8fb857, 0x7940c9fd, 0xa4dfc5b4,
+	0x1abfa1c5, 0x5f9c0c6e, 0xbb5dfb40, 0xc1296a7f, 0xf1e3ecad, 0x7d95efb7,
+	0xd17efe3d, 0xf80bd07c, 0x2b7fdfde, 0x702b9562, 0x1fbf917f, 0xdc7e40c7,
+	0xfa62beca, 0xbdf5a5ab, 0x2677cb64, 0x0fb13d4d, 0xbe031a88, 0x571e40f2,
+	0xc3c65c53, 0xe52b5fe3, 0x5f42f980, 0x3ec7cf2c, 0xdcedf213, 0x4f5d1516,
+	0x1e3e98eb, 0xc8da7af0, 0x7f99f1cc, 0xaf973fbf, 0xbacc57de, 0x73a777fb,
+	0x00b9953d, 0x32d5e9d7, 0xa1fd47ed, 0x7b7f9dfe, 0xf3e88480, 0x4effb099,
+	0x3e705cc3, 0x5ab3fb3d, 0x067c1578, 0xfe057d89, 0xfe22cdf5, 0x991bf03e,
+	0x9bee51d9, 0xcfe41e0c, 0x3f9f8d90, 0x67f20f7d, 0x6a92c1a3, 0xed017ca5,
+	0x64bb6573, 0xb75c7f4f, 0x2f84cf1a, 0xd76575c4, 0xd34e3a08, 0x92ecbfe9,
+	0xd6dfee29, 0xf3162570, 0xe50fd003, 0x4f4d5d3b, 0x7b7f90d9, 0x4e312382,
+	0xeffa61fe, 0xa1f8bf66, 0x6db67bc8, 0xfa8e973c, 0x6260f41d, 0xfa67ca1c,
+	0xff9e46d7, 0x7da1c96c, 0x273f1bbf, 0x73ddbcbb, 0x7349ec95, 0x940f6ca3,
+	0x671a1c57, 0x4df061fc, 0xf007c2d0, 0x1478450c, 0x3eadcb38, 0xbd737f3a,
+	0x5d84ffbc, 0x1cf14b85, 0xd1f9e608, 0x94682fbe, 0x582add69, 0x40ac7b2f,
+	0x7e030997, 0xed082fb1, 0x8a24a7d1, 0x6f924e0b, 0xd40e70fe, 0x4ca4f05f,
+	0x1f1bbde5, 0x7480d0fa, 0xbca06f5d, 0x99a682e1, 0xfc57da46, 0x122cc494,
+	0x37d9ab8d, 0x12972c50, 0x09f6dd5c, 0x680fedf8, 0x7d0b76cb, 0xe37ef7e0,
+	0xa2fb198a, 0x9267a9f1, 0x017efbb5, 0xcb103bfc, 0xdfe047cf, 0x126bfe96,
+	0x416d0e91, 0x5ca3358c, 0x5fc85bea, 0x3e7e5e61, 0xd9d76db3, 0xeb97a414,
+	0x5e292f8d, 0xb01c93d9, 0xf33fbe46, 0xfc8938ce, 0x5cdf81d2, 0x5008e699,
+	0x6fc8b67f, 0x13b8316f, 0xb7dd3ed1, 0xe38f2af1, 0xae71e2ec, 0xee67fb1e,
+	0xf26f1e20, 0x2c1c5106, 0x0532d6ab, 0x3cec01e5, 0x36b93fcc, 0x09b198ad,
+	0x0dbb41ca, 0xabab4456, 0x05c9287f, 0xaaa141ca, 0xb61e455f, 0xb2947c88,
+	0xd8b661bf, 0x6feca397, 0xf83bf650, 0x77eca1c3, 0x3bf62d98, 0x78eb1dbc,
+	0xc19e0e28, 0xd9385f3a, 0x5ffb7eb1, 0xe2cfda7a, 0x3fd1efbf, 0x4f9fee6e,
+	0xcbecad5b, 0x47146088, 0x8dfac237, 0xcd86c395, 0xabf51778, 0xfe4c98e1,
+	0xef23ded0, 0xf1f5aac5, 0x38737d0a, 0xd72f31c6, 0xd1fe4419, 0xc316f947,
+	0xa98dcf91, 0x862597a4, 0xca0bae76, 0xb375def8, 0x338eaede, 0xe7c71d3f,
+	0x9d2e22bb, 0xfb7ab7af, 0xb7afcc21, 0x688cf259, 0x148e1b8f, 0x367e464c,
+	0xf919e7c4, 0x54e6b9f8, 0x9e844ba4, 0xa3457ff8, 0x05b36e32, 0x7bf257fb,
+	0x2cc71161, 0x598577c8, 0xbf3fbeb3, 0x4a83f227, 0xdf33af7e, 0x7be33f17,
+	0xe33b7bf3, 0x8bf71d38, 0xff71bbe7, 0x3b7bf2cf, 0x5265fe91, 0x30f1a4cb,
+	0x8ef88ac4, 0x6ebdf9c0, 0xfcb1bcf3, 0x567147de, 0x9ceb7f9e, 0xbdf90f6f,
+	0x58dffd6e, 0xff7baf7e, 0xbaf7e43d, 0xf9837ff5, 0x797ecebd, 0x50f96f7e,
+	0x5b3fed1d, 0xe3c4dc7e, 0xffbe66e4, 0x65e424c3, 0x668277b9, 0x74f284ff,
+	0x184909b3, 0x275e5f27, 0xbcfe464f, 0xa0b3c96c, 0x78221a7c, 0xb064c83c,
+	0xd43f464f, 0x85191114, 0x711ff97c, 0xabaf3469, 0x3f888b11, 0xc87982af,
+	0x722d89cf, 0xef797960, 0xca649902, 0xd6177f2f, 0xd57cf324, 0x7758b368,
+	0x87e4b7df, 0xf3fbfe52, 0x3d1e5538, 0xed00fb2c, 0xce0d648b, 0x8fbf283e,
+	0x127f405b, 0x7c50b79f, 0x7f3928c6, 0x76217187, 0xbb639f82, 0x601ce41a,
+	0x3ea5c951, 0x2eec28de, 0x33ea126f, 0xa7a2ab37, 0xe10ae1ab, 0x2a478959,
+	0x47cc5ddf, 0xa5794778, 0x7a09b8c3, 0x706f7d2b, 0xb573a00e, 0xbb4e8e02,
+	0x6ea18fe0, 0x6a69a73c, 0x8ff286fa, 0x34df7aed, 0x36d4eb8c, 0xfc6a21ca,
+	0xb973c526, 0x9cfe4dc8, 0xd8e556f2, 0x57e617ae, 0x7ca0064b, 0xc37c9b96,
+	0x7b7dee11, 0xf51c36fe, 0x92a6d5ca, 0x0b7ef83e, 0xdc75a7d4, 0xe7572be8,
+	0xaaf3fd5d, 0xf381f1d1, 0x969a5bad, 0xdb1c2225, 0xf6005e30, 0x3fc3f955,
+	0x663a73a7, 0xea0dff66, 0x837a060f, 0x99868afc, 0x93b438f2, 0x01db003e,
+	0xf8bb59ca, 0xcc959bfa, 0x2d37983c, 0x0e60fa6f, 0xdb3cabd2, 0x32974168,
+	0x6de78338, 0x16cdb383, 0x45a6ff5c, 0xe2303e71, 0xed869b3a, 0xdd214627,
+	0x3031c41e, 0x7b73b9cf, 0xb7f01bbd, 0x854f0ff0, 0x27dfc033, 0xafda4afa,
+	0xb1f9ba51, 0x2f3c3b70, 0xc714dde8, 0x0723c4fc, 0x1794ffdc, 0xf6b7f8af,
+	0xfe754950, 0x3fdfd3e0, 0x6ef88d31, 0x7982c1b3, 0x1316d77f, 0xaf2b9f9d,
+	0x33f3a62b, 0xcbc53e50, 0xfe867f8f, 0xbb41ab95, 0xe97c2f4e, 0xad2b58a9,
+	0xb124ee0b, 0x941f11ee, 0x849d9cd7, 0xc1f857f8, 0x8f8e1dbf, 0xb55de5fb,
+	0xaf633cf2, 0x7b15dfee, 0x5dff5943, 0xfb884f75, 0x3fc85a01, 0x845faff0,
+	0xb19a3232, 0xfddaf199, 0x983e0ad1, 0xa3f48627, 0xf0c373bc, 0x54be5053,
+	0x92e16ff3, 0xbf8a4f0a, 0xff9b51d3, 0x3dac37bb, 0xb06defaf, 0x7574b02e,
+	0xb3b50f3b, 0x177e0836, 0xf06a5f7c, 0x955ca386, 0xdcd8346e, 0xfb74baf0,
+	0x973e2f0b, 0xc3667245, 0xfb64ac75, 0x818e1f5b, 0x430ec972, 0x8d3e54f4,
+	0xde754950, 0x61ff05cd, 0xc3a1405e, 0x32ded7c2, 0xf7f1875a, 0xa2ec88bf,
+	0xbd8ec947, 0xfc5e7446, 0x76be31f1, 0xbbf291af, 0x3da974fe, 0xef2b7245,
+	0xbbec86fc, 0x2c48f64f, 0x3e00352e, 0xee15be37, 0x3a227814, 0xd2753a0a,
+	0x92beafcd, 0xe4e8577f, 0xde3b3dff, 0xfd85f242, 0x3f023ca1, 0xc606f2bf,
+	0xd9323ffb, 0xaf45fe70, 0xf38e103c, 0xc45faf38, 0x63c37af9, 0xb71876b1,
+	0xb58ee0c8, 0x2fbf93d0, 0xc467fe7a, 0xfe8e0b6f, 0xcc7fdb8c, 0xfca3a09e,
+	0x7bf29ba5, 0xd0efeb85, 0x0da2ffe8, 0x85ffae3c, 0x11de4d9e, 0x0b8e2e40,
+	0x937e72bd, 0xe7d9f289, 0xf6860f31, 0xdf327bcb, 0xa7eec7b7, 0xdf629078,
+	0xe54d2780, 0xfe1e842f, 0x894ebc43, 0xfc359b7f, 0x3378f35a, 0x64d3e7d2,
+	0x25ea1c7b, 0x05bbd317, 0xa0efa43e, 0xc4e8f95f, 0x1e1f4e38, 0x6076a33f,
+	0xf7c78e8f, 0x40d9bf58, 0x3b5550f1, 0x8f966e71, 0x1f18a0ff, 0x7cc60ee0,
+	0x5658f9ca, 0xd44fc814, 0xe7c67427, 0x29e7359b, 0xcb9be4bb, 0x2e2acd7b,
+	0x7bbc8adf, 0xefc64fa6, 0xf2e31f9f, 0x3292ed83, 0x1c6f2e09, 0x4b3bdf7d,
+	0x27cfce1a, 0x75c9037b, 0xb5c8418d, 0x27c2bbf6, 0xee9eb700, 0x05f85021,
+	0xfc248ffb, 0xbad2d517, 0x1dd8d26d, 0xbc2f1c2e, 0xdc8a7f7b, 0xfff2102e,
+	0x7f38bc7e, 0xad9fe425, 0xf3a1b98f, 0xcf06a545, 0xbca1cf8b, 0xbef7e09e,
+	0x90376e94, 0xdb717e5d, 0x06ef0ffe, 0xab58b939, 0x7f56a9c8, 0xd169cbfa,
+	0x7f116fab, 0xc5e9cbfa, 0x0362b7da, 0xb78ee9ca, 0x2eee0ec8, 0xfab9ffa7,
+	0xdfdfc153, 0xe8a7f0fc, 0xe8f09ec3, 0xe3ab0faf, 0x21f501ad, 0x9f531dfc,
+	0xa5f0537b, 0x6fc29df0, 0xeb2e5f0b, 0x93a77a83, 0xebbe152f, 0xef854be4,
+	0xd73bedba, 0xfbffcfe0, 0xe00df85b, 0x60faec72, 0xb3d01252, 0xc9fb91b4,
+	0x9620eed0, 0x5a764f01, 0x41e558c6, 0x55db6f3c, 0x6b95f8f6, 0xfddbeafd,
+	0xfabf0eca, 0x22f2bf4e, 0x821efabf, 0x3f61b4ab, 0xbcc96fdf, 0xa586fa9a,
+	0xdd161dcc, 0xf31e0e75, 0xcfd02b58, 0xf4f3c6df, 0xb8f0258f, 0xb8c62fa9,
+	0x891ff682, 0xe13e2a7d, 0xb43bb467, 0x354e54fb, 0x5b7fa4ca, 0xf8f3813c,
+	0x3c9209e0, 0x3fb0f4b4, 0x4d3b895e, 0x1e534394, 0x2a3a3806, 0x60a0fc79,
+	0x94e19dc9, 0xbb4c69cf, 0x850ae31d, 0x2c31f987, 0x73143f65, 0xaa09f5db,
+	0x98170efd, 0x319be9fb, 0x4258d64f, 0x35828bf1, 0x0fcd7a5a, 0x0be2907d,
+	0xf7b4abf3, 0xad7c2900, 0xd68b0bc4, 0xc42fbf6b, 0x1ee8847b, 0x47da4a85,
+	0xfb46b0b4, 0x0e7c89dd, 0x41273c7c, 0x2c1909d8, 0xf9bde443, 0x57a11239,
+	0x6f43ca27, 0x21cbfca7, 0x1da5f299, 0xbde1328d, 0xfb449b63, 0x863b9869,
+	0xf99ca5e4, 0xf8299f48, 0x60f284b9, 0x5cdf59ec, 0x73dffd7a, 0xa52a851e,
+	0xff0f1ff5, 0xd52f3c2d, 0x8ddf489f, 0x4e7f9de5, 0xaf7598f9, 0x104f7e3f,
+	0xdf6a1fed, 0x35f74613, 0x9dbd37b5, 0x85e528f2, 0x07ba364c, 0x9497cf61,
+	0x6a5f6b7f, 0xd57d422a, 0xcafdcc9e, 0x0e5f4d65, 0x39657ae7, 0xc098df3f,
+	0x34744dcf, 0x94f28dfe, 0xca268d1d, 0x84f6b933, 0x7ae11f90, 0x9cfc8823,
+	0x292fa6f6, 0x9acffcbe, 0x3788b94f, 0x3cf187b7, 0xbfb8bb4f, 0xf47bb5c4,
+	0x2d87980b, 0x446bdbab, 0x9bfc65bf, 0xdf5d90df, 0x67b72afe, 0x1eded0da,
+	0x5bae7883, 0x7dc44e33, 0x4d0ecca8, 0x0db77b45, 0xb1912385, 0x9d5dcfa1,
+	0x081a1eea, 0x5e50df4f, 0xf3f146f5, 0xfc18ffd2, 0xb9dfb448, 0x63cc31b4,
+	0x5fde7096, 0xf518fc9d, 0xd79e1ec1, 0xb17be657, 0x1fd90976, 0xc1f79bd7,
+	0xfddfee04, 0xe4b799ef, 0x9f7991bd, 0xef4e6ffd, 0x98d43c62, 0xba2e5ddd,
+	0xcf1379f3, 0xb30f6d17, 0xc837bc11, 0xffc486ef, 0x157f7465, 0xfd7e1ee8,
+	0x6ffbf779, 0xe99eef3e, 0xa0ae787e, 0xf3e305bb, 0x21bf37ae, 0x3d5efef8,
+	0xe87fe137, 0xfd3b15fc, 0x6fd7c67d, 0x91c3a257, 0xee9efb88, 0x5f02bc38,
+	0x274e74b9, 0xcd12c3c5, 0x4d9fee74, 0x8e87e28d, 0x33618ef5, 0x57877586,
+	0x25fe1bf1, 0x31273ed1, 0x07feaba6, 0xa619efb3, 0x7d57779e, 0x58d1be3e,
+	0x379d3f7b, 0xfca50b29, 0xc18f6e2c, 0x36cc2f76, 0xf713fe71, 0x990f1c3c,
+	0xa606307b, 0x738347f2, 0xe739c23e, 0xc0ac9fce, 0xca2fb4b2, 0x35e718ba,
+	0xdc4725af, 0xc710b08b, 0xfb3b352f, 0x72f9aa60, 0x2f4073b2, 0x511c47b4,
+	0x1ccb9e3c, 0xe1011cc3, 0x37ce6b77, 0x994bf62a, 0xea3aa1b0, 0xbe517ad2,
+	0x645ed5e4, 0xb8e9b57f, 0xe026d87f, 0xb767e0bd, 0x1a8e7917, 0x5c58ff31,
+	0xcea8d473, 0x7389963f, 0x3fe3fea9, 0x23f3c891, 0x87f6be9f, 0x8bf748ab,
+	0xa3a7d45f, 0x5d2256bd, 0xdd5d79a2, 0x98ae747a, 0xff9c0adf, 0x14be1d43,
+	0xb4abcc23, 0xc791af76, 0xf4fcc523, 0x9beff4be, 0x79ae9b5f, 0xaabb9d12,
+	0x8e7ce897, 0xeeacaed2, 0x93b424e8, 0xe8226a8b, 0xa7dc473e, 0xca9dcfc2,
+	0x7860fb7e, 0x8eb2493f, 0x4bf7dbf8, 0xd21bf689, 0xe9dfb41f, 0xbee0f08e,
+	0xcaee936a, 0x676899b1, 0x8ce7e30f, 0x438307d6, 0xbbc77e0f, 0xc41b5f02,
+	0xbf780779, 0x07c7df12, 0x4ff1f6f3, 0xfedf68eb, 0x9409b90b, 0x7bdd543f,
+	0x72075291, 0x36b26494, 0x49b8c0b4, 0x0f28bd15, 0xb9d13dbc, 0x5f71f105,
+	0x18e52e94, 0x421ce1c1, 0x8e6f80e7, 0x753f21ba, 0xf818b306, 0x9e7e527e,
+	0x4f85867d, 0xb5d850a5, 0xee9d9933, 0x3983b8e1, 0x6d7bdea9, 0xd7da3b46,
+	0x336e61ef, 0xa3f7ee93, 0x0f3177f7, 0x5e61ef3d, 0xac2fffaa, 0x5f0aca85,
+	0x17de3f02, 0x3cf1f9cd, 0x8e8160a6, 0x47dc226f, 0x3f278643, 0xfe9ef80c,
+	0x94fe80cc, 0x76e2ad2f, 0x9ee0f5f0, 0x574bfc2e, 0xf210f3a2, 0x2fe4c77b,
+	0x7e82af9e, 0xf12f6674, 0x91ee7283, 0x7c11884d, 0xdb72fdf1, 0x0f99d1dc,
+	0xe2bf450d, 0x5bd24156, 0x509fdced, 0x2edc0dce, 0xd9ef01e4, 0x343a0489,
+	0xa7bdf95b, 0x3dbd2acc, 0xc793ca02, 0xe59fb4c9, 0x20d6e08f, 0x4331fc7f,
+	0xd9a5f716, 0xd134fda9, 0x3371d4ef, 0x7a82f797, 0xc96ebe93, 0x9327f3c2,
+	0x8afd6b7b, 0x5c6f449f, 0x29cf011e, 0x7fefb4fd, 0x88f59aca, 0xb543a9d7,
+	0x2cb4cdfe, 0x38bdbd6d, 0x2a962466, 0x4b845ee9, 0x193fc289, 0xfdab06cc,
+	0x1d3e6513, 0xd118af53, 0xe493ec17, 0x597482bc, 0xb9d9db7c, 0xcf4ede74,
+	0x4d111ede, 0xbcd1cff7, 0xe6f02b5a, 0x55cec646, 0xa179e998, 0x20ecc2c2,
+	0xc4569ff7, 0x8552f3d3, 0x55b7de95, 0xf8e7a40c, 0x90dfc724, 0x9f95053e,
+	0x670fd1e9, 0xee34e955, 0x9acf8203, 0x9d1d5e75, 0xd124d85f, 0x82079a7e,
+	0x8ef1739a, 0xa73607ac, 0x61739c52, 0x0c41ed43, 0x1d78141f, 0xf39c9093,
+	0x883a2d0f, 0xc06bb2cf, 0x158023e9, 0x24a61c12, 0xa763eb94, 0x7ac6df67,
+	0xe37b1be6, 0xd05183a9, 0x1b0af9ff, 0xc549f5c1, 0xac5e7a28, 0xb3dc13b0,
+	0xa4961c92, 0x67d2a46c, 0x6db73fc4, 0x824fa53b, 0x12ad13fa, 0xb63f2bad,
+	0xe4078f33, 0xd7c0daab, 0xb304ea85, 0xdc74e50d, 0x1e1b19b0, 0x9df7a864,
+	0x1dc70889, 0xe2278a53, 0xcd46d338, 0xeeb47ba3, 0x7bf19afc, 0x58dc2695,
+	0xe9f988dc, 0x247bcbdd, 0x3acbcbe2, 0xff18edeb, 0xa649ff75, 0xde0740fc,
+	0xe3fe80d3, 0xe32f7e71, 0xe2d3d7e2, 0x362f3f50, 0x5fd0e358, 0x79f62ddf,
+	0xe97a045c, 0xc5eceb55, 0x9cce7fed, 0x25bfd2a7, 0xfabaecaf, 0xf2fbf411,
+	0xea1b48c9, 0xe2918bbd, 0x28b6467e, 0x9b48f504, 0x19623bc8, 0xd65015ef,
+	0x5833d45d, 0xebcc18be, 0x9492bce7, 0xe382728b, 0x7a5d8d59, 0x6795fd28,
+	0xed6a3efa, 0x6d4b6f1a, 0x41cf47e1, 0xe747b2f9, 0x723ac036, 0x61b6d599,
+	0x135dd6dc, 0x066dfb4a, 0x87e0fa4a, 0x7c3d6f89, 0xd56488bf, 0xeb6c3c63,
+	0xd1121b2b, 0x392dd80e, 0xfa823ee6, 0xffaf037b, 0xe48fd3d2, 0x53e7c2ed,
+	0x823eaf65, 0xa08a4f78, 0x2b9ff376, 0x36fc8c1e, 0x5d5efd6c, 0xa9bfbc22,
+	0x7c52d4e6, 0xc2a240ce, 0x926d8cf7, 0x8cdd778c, 0x865d4ba8, 0x1fb898af,
+	0xfbc0b06e, 0xed9ff5d0, 0xf4107703, 0x08e2ed6b, 0x525757ca, 0x671bd7bb,
+	0xc10f7fbe, 0x7bc04ee1, 0x8cf3dd35, 0xb8f03edc, 0x5c9f2237, 0x51b40e76,
+	0x7ee1cd7e, 0xd973e916, 0xf5c36b19, 0x0ee75ea0, 0x4ff5831f, 0x76087bf2,
+	0xc70d1cd6, 0xeabd11fd, 0xed082c0c, 0xad1f0c95, 0x56c9654b, 0x259659bc,
+	0x64eb7f3f, 0xd7927946, 0xa53f5237, 0x99e61b66, 0x42bf5e8b, 0x13f61cf1,
+	0xfc8feb66, 0x4fcbc751, 0xfd13b04c, 0xc17e368a, 0x2a0ecdcb, 0x2a77643b,
+	0x0f945fdd, 0xfbf40c76, 0x5877588c, 0x47d270cb, 0x2ed672c4, 0x9f013f7c,
+	0x3abdf02d, 0xfb809dd5, 0x0af594c3, 0x219f99d6, 0x839ba37f, 0xef739bed,
+	0xf775a149, 0xa519d706, 0xfc8c679f, 0x46f5bd7e, 0xb66e1ebe, 0xf42778f0,
+	0x4fe514bf, 0x64931759, 0xae1d5869, 0x6c98cf73, 0x3e37accb, 0xfee2ddce,
+	0x2af813ae, 0x430d6c41, 0xf8ffba42, 0xe9ffc636, 0xfbf23dfe, 0x15675799,
+	0xf941ecf7, 0x4fc3a4e9, 0xe619bf2e, 0xe003f82d, 0x83013820, 0xf2a7f50c,
+	0xe38b2c3e, 0xb669d233, 0x2e2b7e61, 0x33dd116a, 0xd307f0a1, 0xdda0477b,
+	0x995bfab0, 0xffb94bce, 0x7f714a1a, 0x4ddf787f, 0x7af471e2, 0x639e07c0,
+	0x7bf07772, 0x5ac31a9f, 0xf403fe5e, 0x9bdc58df, 0x0dbbc6e2, 0xf756fefc,
+	0xdc5df174, 0x5778dd27, 0xa27dc92c, 0x7601eb89, 0xd38d304f, 0x7e6cf068,
+	0x8f4fbf1a, 0x85c7f6a7, 0xde671d9c, 0x955f707b, 0x7ba467db, 0x1822ab39,
+	0xb55367dd, 0xfdc50064, 0xe07dbfea, 0xbda186df, 0x639f84b2, 0xacf0caab,
+	0x55ff7dc4, 0x70e34cbf, 0x1724de48, 0x260e2fb6, 0xdd355d3d, 0x4071809e,
+	0xba10b4e7, 0xfba7ba57, 0x28323f5e, 0x09ec89e7, 0x0ae78f88, 0xfdf533fa,
+	0xca4e7f41, 0x57748c3d, 0x3a73d1f2, 0x8ecc22f3, 0x21de5bf7, 0x6e15770a,
+	0xbb7e420e, 0x14e828a3, 0x178f6bc6, 0xb8c30d2d, 0x7e4e1de3, 0x359b273f,
+	0x8f3eaed1, 0xfb24e7f9, 0x5657e461, 0xdeedd8a0, 0x1ed3c47c, 0xbdce999f,
+	0xa847f901, 0x7f8544ed, 0xbe77aafe, 0x8ddaed87, 0x92e35ee2, 0xb6cda6af,
+	0x6bf1faaa, 0xf27edfe7, 0x37bf330c, 0x7644cea9, 0x8ec27dc1, 0xb3bc9972,
+	0xad5a0ed1, 0xf607fe73, 0xe9f85aef, 0xa277b470, 0x17fbf067, 0x09be31db,
+	0x7a0a4f75, 0xff7c649d, 0x98517ba0, 0xb2f11bc8, 0x0f28d1f3, 0x9ec7a7bf,
+	0x07e58850, 0x1f70d886, 0x99e45390, 0x7c44b958, 0xf9c2d272, 0x146364e4,
+	0xe33e8fdf, 0x30e9d3da, 0xab8e938f, 0x7a848f6f, 0x79f81b27, 0xcc3cbf8e,
+	0x2f9619cf, 0x60d2913b, 0x7baf3eed, 0xd97bf297, 0x0bf85def, 0x93da6f3a,
+	0xaa5bddfc, 0xd21987dd, 0xce15fedb, 0xb5c67a33, 0x1287a849, 0x2d9f4a4b,
+	0xdbdabd44, 0x2fbe2df9, 0x2394b637, 0xff77ca3e, 0x7fda0a7b, 0xda5ebbed,
+	0xaf9e1071, 0xdeba9d19, 0xade97df8, 0x67618fa9, 0xd72f9ec2, 0xae403d03,
+	0x34edea9f, 0x717e06f6, 0x63a6f04d, 0xe6864bbb, 0xf6378849, 0x1bb31b8b,
+	0x5cdc6d3f, 0x115fa12f, 0x4f4e58dc, 0xe11c33d9, 0xc37928b8, 0xa71e1ef1,
+	0xc97e8fd8, 0x188f0de2, 0x72c7f78a, 0x1f3ff5fe, 0x8ac5a048, 0xaed0327d,
+	0xe71656a4, 0x27143c97, 0x025faff6, 0x0ddea9ef, 0xcbce57be, 0xac344495,
+	0x66f44b1d, 0xee166f5e, 0xf74feceb, 0xfd7e5d32, 0x5fb574a5, 0x0eecffe5,
+	0x9d3f9f2f, 0xa9bcc6e7, 0x7bf7df87, 0xe4068a09, 0x71266db3, 0x51f38f8e,
+	0xc104d6db, 0xda8f9c67, 0xa8f00eb6, 0x9cf3b6e5, 0x8736ada0, 0x6bdbedf6,
+	0x501ed073, 0x6b999fe4, 0x36677ee9, 0xda3876b0, 0xc23d8add, 0xf4e9b51c,
+	0x8e9b53de, 0x3c75ebf5, 0xbb8ca73f, 0x7e24e5f0, 0xeb9998df, 0x0f2f85c5,
+	0x94c6fba1, 0xcf37af7e, 0x4d46d501, 0x5b479079, 0xef9bd69e, 0xeffdf85b,
+	0xcbf1f73b, 0x93c61ec2, 0x40f79d85, 0xbd73de76, 0xf2f05255, 0xfa0e5260,
+	0xaaf28ca3, 0x0be07bbb, 0xbd21ae05, 0xf30b3fb9, 0xbbceb863, 0xbbfae23f,
+	0xf331c33e, 0x6fd72836, 0xb6fb2348, 0xe9f8e74d, 0x79e71c30, 0x3ebaafbe,
+	0x9cf947c2, 0xfdb967fa, 0x5de62e5b, 0x424ce11f, 0xc2a17a7d, 0xffee2f16,
+	0xe96c3b26, 0x98e11589, 0x61dfaff7, 0xd5f8c0c8, 0x63fb3527, 0xad3e2f91,
+	0x37ee9878, 0x75e3bc1c, 0x6e94a3ca, 0x1c7de07d, 0x8336d2f7, 0xd2acdd0d,
+	0x8b5c5ebb, 0x56e948f6, 0x5be716b9, 0xe53f75ba, 0xbed52fc7, 0x71fb8e15,
+	0x9144d42f, 0x53f08c1f, 0x1cf7fafe, 0x2ff8fe0f, 0xabc464f6, 0xc91591f7,
+	0x7f78b5ee, 0x5a2d90c9, 0x3c7d17bc, 0xedea10e9, 0x18cff50b, 0x1fbc56f5,
+	0xa6fdfb56, 0xeff96d06, 0xe4abaf3d, 0xd4e39091, 0xdcc85c74, 0xdda9f507,
+	0xb5645feb, 0x3f18edb3, 0x9727bce8, 0xdae8edcb, 0xbb379475, 0x6e6e78fe,
+	0xe9da853c, 0xa379ab9e, 0xe231b7f8, 0xe6b65efe, 0xdbc424e8, 0x5a2427bc,
+	0x36bde6ef, 0xcfea00e0, 0x9aee3b53, 0x93d9cf42, 0xe8f91f09, 0xd67ee085,
+	0x061bc946, 0x8e28cb65, 0x9e6ac2ef, 0xee5bca17, 0x8d69ee77, 0x2af8fedf,
+	0x0d05f7ed, 0xbcc2bf7b, 0x79f99176, 0x3f741f8a, 0xfba1e3c7, 0x605e177f,
+	0xac2dcfe4, 0x2814cef9, 0xcca851be, 0xda63fdc3, 0x264dc01f, 0xfff0fff0,
+	0x4daee346, 0xe1ea3d47, 0x651b5531, 0x731957c4, 0x2349643e, 0x021bc5bf,
+	0x8137cfe3, 0xeb4992dd, 0x4ae7cd21, 0x8f3a2f75, 0x5059ef0e, 0xda03de26,
+	0xe4eb839c, 0x1b30ba95, 0x9cd69fdf, 0x02d57683, 0x21f9d328, 0x0bce2956,
+	0xf1fee87b, 0xcd06d505, 0x1ec5e07d, 0x7255996d, 0xef14b0fb, 0xfa470ed3,
+	0x92bb50b7, 0xa57ada2d, 0x91e39f30, 0xfb51ffb8, 0x5b7cc5de, 0x2b7da45f,
+	0xefd498df, 0xb1547b15, 0x0e38958f, 0xe460b228, 0x7237bb97, 0x04b77a3d,
+	0x0c3b45f3, 0xafb234ed, 0x420e3b3c, 0x2878aab9, 0xdff54fd9, 0xca5bfdf2,
+	0x0caad4f9, 0x279e3a7b, 0x88375aab, 0xb25bbfe6, 0x7a863ea7, 0xd859ba60,
+	0xe1ee8c57, 0xa44e2aaf, 0x1c23fbf5, 0x2ced0132, 0xf729ef6e, 0xeea227f5,
+	0x18e371c2, 0xb53f8fbf, 0xb37a8bd6, 0x3fae29ea, 0x81563b14, 0x4f675f3b,
+	0xb36e75c1, 0xcdebe9d9, 0x23a56ff7, 0x09bff9dc, 0x4557bc22, 0xfce8c2f8,
+	0x0dd4517f, 0xc59ba7ad, 0xabcca0fa, 0x7dd07255, 0x293d4ae8, 0x9cdbeb99,
+	0xa28bfb5c, 0x0b39c51e, 0x6fab1d72, 0x3c147c01, 0x1fa2b3ad, 0xbb6789dd,
+	0x13ef1d3d, 0x94175d68, 0xcc9cf089, 0x47367ae0, 0xf6061bdc, 0x3c137dbb,
+	0x9a759383, 0x32706578, 0xc163a7b7, 0x5af1c27f, 0x5438f227, 0xfc839658,
+	0xe6a9c079, 0x19e53c1e, 0x302be02b, 0xa3f4fe3a, 0xd274f000, 0x7bc7027b,
+	0xad12fdeb, 0xa21f3047, 0x83afce5f, 0x2ad00e28, 0x4e8503d6, 0xb4d92fc0,
+	0x54bd09ce, 0x21f80ab5, 0x2da1ed15, 0x6afa835b, 0x0be286f8, 0x3ebe2eec,
+	0x7c704554, 0x4c393d56, 0x6f82e5d8, 0xbf120c9e, 0x6cedc0c7, 0xec51fd65,
+	0x3d3e6570, 0xef40e1d9, 0x7c2ac961, 0xe999d3f9, 0x7dead93d, 0x5e74fe94,
+	0xd2986e60, 0x0bc1550a, 0x5c067825, 0x6b7d91d2, 0xde0892ce, 0x2a7ad581,
+	0xb259d51f, 0x8da6ee51, 0x11f179dd, 0x7e7857d4, 0xebf47907, 0x39bbbeef,
+	0xa951f707, 0x7dc472fe, 0xeb0fc8ca, 0xef183799, 0xd894afd9, 0xa03acd69,
+	0x8fdfe32d, 0xb9fabb63, 0xf89efd5e, 0xdcfd9ff7, 0x69d8c159, 0xf85d8590,
+	0xebc347ec, 0x4665fbd5, 0xb93c7236, 0x5f4f308e, 0x597677d2, 0x0467ef8d,
+	0xcbfcee1f, 0x6e3023ab, 0x687dae31, 0xe9679e2f, 0xd3ef53a7, 0x5ed660ed,
+	0xa3b4efb8, 0xee17ad31, 0x2438eb4b, 0xf0033987, 0x038056f5, 0x572ff1e3,
+	0x5ed9e998, 0xf0652ac3, 0x8f99da2b, 0x6bba67f6, 0x5f902ccc, 0x73b934ba,
+	0x739319ff, 0x3ba5f7d4, 0xa4ec7d68, 0x9fa9a5d3, 0xc77787e8, 0x569dfe3f,
+	0x87f913d6, 0xebf715e3, 0x06b75e20, 0x740a4cb5, 0xe04aafde, 0xafbeda6d,
+	0xe0e33f68, 0x943a8dc6, 0x3a35c1f7, 0x26b93acd, 0xca43ec59, 0xd7089964,
+	0x61417db4, 0x438f4859, 0xe6082db2, 0xccfc6ae7, 0xd7e00ce8, 0xfb11e3f9,
+	0x8702fdc7, 0xa7842d93, 0xf25a86dd, 0x365f772b, 0xbb29fdd3, 0x8125958c,
+	0x23eccce0, 0xd8ebc405, 0x5f6833f3, 0x5e157fca, 0x7a87b5c7, 0x588e3173,
+	0xbb1c78c3, 0x7d71bf1b, 0xd185b1d4, 0x710ef739, 0x1b63a80e, 0xa01ebca3,
+	0x77dc49e2, 0x8b7df203, 0x80d63f7a, 0xe32f1d86, 0x28de5efa, 0x72e4fed1,
+	0xf7118b7d, 0x221aba5b, 0x65bfcc64, 0xae3dc558, 0x59b0cb7a, 0x9c38392a,
+	0x17e7e7ac, 0x07beecc6, 0xc972bdc7, 0xec87e748, 0xef747c07, 0x11c695c0,
+	0x3cefbebd, 0x6c156bce, 0x7b3bde0a, 0x5f119938, 0xd0c1f3f8, 0xf29ac16d,
+	0x8899d958, 0xd4cd34f2, 0x002da0f2, 0xc4528add, 0x7aa370fa, 0x89fff3cc,
+	0xe8a63b1e, 0x8eaca731, 0x74a46487, 0x817163d3, 0xc962d8f4, 0x5ef8871a,
+	0x871694dc, 0xd87c5cf4, 0x8f481310, 0xac38076d, 0x007f7720, 0x7b67c3f9,
+	0x9bb1e81d, 0x0eac7a54, 0xbf0058f4, 0x3ae52924, 0x5ff3cc7a, 0x51db6f23,
+	0xe798058f, 0x1fae14f0, 0xbc363d14, 0xb1e914f1, 0xcd3a75e1, 0x29feda39,
+	0x5263d3d7, 0x2c68accc, 0x01216c1b, 0x89f106f8, 0x27c47779, 0xd41c713e,
+	0x9629f967, 0xfd9f4bbf, 0xecfa7af8, 0x8abf08bf, 0x3afc5b3e, 0x3ba37f23,
+	0xdfe231df, 0xfd18bbc5, 0x26e2cfce, 0x3716cfae, 0xcf0fbe31, 0xec0d8f4f,
+	0xff18c7a6, 0x6f2f165d, 0x7d486aef, 0x2fe85ee6, 0x99b06ed3, 0x97667ea1,
+	0xeb56cb5e, 0x1e94fcb5, 0x1ba1daf4, 0x0baf6bd3, 0xaafa06bd, 0x7f51ea2f,
+	0x6a35b0be, 0xf763977f, 0x5edbf3e8, 0x5b78f943, 0x4701f3a6, 0x9dfebe45,
+	0x90b3e7d1, 0x1e11eaaf, 0x4235d390, 0xdc2adb5e, 0x03f8ffd4, 0xdb657e7c,
+	0xca0f5a24, 0x57bdb8eb, 0xf07b235e, 0x31a4e180, 0xff3ed15f, 0xe18926e2,
+	0xa6e9b2d1, 0xfef1a7ca, 0x8192f630, 0xd4e53889, 0x143f150c, 0xf8a12a8c,
+	0x999e8161, 0xc96ddf71, 0xb73fbe28, 0x7b959444, 0xf53b0ba2, 0xcdbedfa0,
+	0xf7e9fb86, 0x7bcffc34, 0xdcf7134e, 0xf2417b4d, 0xdd5ffde0, 0x712a7bcf,
+	0x97fbed38, 0xcba77cc5, 0xf61f842d, 0xde73e06e, 0xe5e3fb2f, 0x8d4bc090,
+	0x478bc5ac, 0xe517880a, 0x8782ad5c, 0x565bb7e2, 0xf8a75c60, 0x32fc8959,
+	0x782b79ff, 0x761f76ea, 0x909dff23, 0xdb62f7ee, 0xd25bbf35, 0x144477ed,
+	0x7fcdfb3c, 0x323fa2bb, 0x37bc08d8, 0xf482f81b, 0xfc4ffb87, 0xb8c7421e,
+	0x3d6a71fe, 0xc3c09afc, 0xe371ca88, 0x2e99add1, 0xd5f98afc, 0x804c1ff7,
+	0xb71f67f9, 0x6303fdd2, 0x9ee898c7, 0x3e41adbe, 0x9f0a6fd2, 0x3e70fa19,
+	0xecfd116b, 0x4b945c4e, 0x241a74fb, 0x2e13da0b, 0x37bf106f, 0x082df805,
+	0xf67fecaa, 0x478124ea, 0x689efd0b, 0x006f5883, 0x4e57e8ef, 0xc01fd652,
+	0x88cfca3b, 0xd7074e46, 0x2c874e16, 0xe567e3c6, 0xf30d92d5, 0xf70de5ea,
+	0x8a3e711b, 0x50bf85de, 0xfc545d35, 0x27f357a9, 0x7bad547d, 0x34fbe225,
+	0x1887e553, 0xfc06993f, 0x0cd891be, 0xfc3df7fe, 0x5f77eddb, 0x2f7bf3ae,
+	0xa78e1f72, 0xf44788ed, 0xdc7d12f3, 0xbac2f26e, 0xddf5d610, 0xf06b77a5,
+	0x6bb9fe38, 0x249abf9c, 0x3b01c62b, 0xefc354f1, 0xf056cf6d, 0x788afc84,
+	0x63279c6f, 0xf8837e79, 0xf9bf1fdd, 0xb5083ee2, 0x9d37171c, 0x906ffde2,
+	0xbcc02c6c, 0x7fdfe47b, 0x927b6f90, 0x2d1f435f, 0xe413beff, 0x2efb823b,
+	0x77e24ddc, 0x34231a19, 0x7dfec279, 0x35b01db8, 0xfd1cf808, 0x1b06c3e8,
+	0x2fe0a7de, 0x3d62b1ef, 0xaadf7b43, 0x420f97ca, 0xb12e3180, 0x28ceccc5,
+	0x76e5f470, 0x6edf5aa5, 0xb331620f, 0xbfee7ab2, 0x4f5ba09f, 0x1e528df6,
+	0x5ea1e386, 0x54f1806c, 0x7d7f2cf9, 0xd7e5f8af, 0x0fc6579b, 0xce32979c,
+	0xd87e7bdf, 0xfe799ac5, 0xd0245f02, 0x26dfe179, 0x9fa9bc63, 0x7bdf57a4,
+	0xfe6fdcae, 0x3b3fe84f, 0xfee7a625, 0xf685bc59, 0x4efb2937, 0x0e3cfd94,
+	0xf3fb130b, 0xae33fe6f, 0xfcec57e3, 0x8e9641fc, 0xae33707d, 0x3a06b90f,
+	0x88e57ac7, 0x573fe281, 0x2e7bf026, 0xbf7b7f6f, 0xedf2f51e, 0xff0114a8,
+	0xe44b67b7, 0xaec51d1f, 0x6d1da347, 0x80257da6, 0x1f9811bf, 0x1ca9a8c9,
+	0x333ee78f, 0xe19678b7, 0x17de36d3, 0x6dd20efd, 0x7bf74c9c, 0xc6a3db34,
+	0xc0fa3ae3, 0x71819659, 0xa9ef5877, 0x1621e67d, 0xd8c6f583, 0x18fdc863,
+	0xb1f9f75d, 0xe8c70099, 0x0acdb223, 0x0fefa6f5, 0xf77cc50d, 0xe63af5c0,
+	0x0c6be154, 0x73c32cf3, 0xed5ba33e, 0x22ee1133, 0x2e305fb0, 0x71b8d45f,
+	0x2e439e19, 0x7b512f5a, 0x2939db97, 0xbdb3e61c, 0x4349b3fa, 0x3d401f50,
+	0x9eb5c6ec, 0x48d573d7, 0xcf394c36, 0x24675c6f, 0xefc3517d, 0x91eed86f,
+	0xebf79998, 0xdbbf4331, 0x2cfb43d7, 0xd05b638a, 0x7cd5720e, 0x5cdffb46,
+	0xdfb52700, 0xccc5cfc6, 0x94567687, 0xa9fb9a97, 0x12f5e2bf, 0x457323c3,
+	0x543af8db, 0xc360c474, 0x9f7b75ef, 0x2a74f5a5, 0x8a90978c, 0x02778eeb,
+	0x884739c5, 0xffc77b7d, 0xed5df90f, 0xd58fd38c, 0x2feb92fc, 0x9fbf2989,
+	0x4f9f57ab, 0x97d3e8eb, 0x2db10bef, 0x3f414631, 0x498a8cef, 0x7d5f11d5,
+	0x71b4e746, 0xf013935f, 0x179c547d, 0x8cea3fdb, 0x409f7d51, 0x4f3f2b2f,
+	0x8e30d05b, 0x321551ce, 0x9aa39d07, 0x8fdb9a36, 0xc828feef, 0xde33fb83,
+	0xe37af835, 0xe23ffdfb, 0xb6dcd3bf, 0xb6dfeb9d, 0x1fda16e4, 0x307f1e7b,
+	0xb65c6009, 0x717d265e, 0xce819b7e, 0x7018c90d, 0xe76dfc44, 0xf5fec5f6,
+	0x5bf3dfbc, 0x7e361db8, 0x70931dc3, 0x59b05dbd, 0xf5c46739, 0xce3fe2b3,
+	0x62f388b7, 0xfc8d1b95, 0x5fefacde, 0xac3d1477, 0xbf911efc, 0xfceb2d9e,
+	0xc92f6cfd, 0xffad594f, 0x2ed0ccf9, 0xefe9043f, 0xb9f88afd, 0xa18188fc,
+	0xb99d221f, 0x4f8ff944, 0xef871d70, 0x3b251ba3, 0x3a2a3b9e, 0x1d90f26e,
+	0xdf7799ed, 0x393d71cf, 0x3afba261, 0x1cf055ce, 0xef9328f6, 0xf781bf2f,
+	0xe3351c9f, 0x97e03cce, 0x05ef8479, 0x3be1116c, 0x473f336b, 0x2ab47597,
+	0xe7ae3e76, 0x7fdc7813, 0x9cccb541, 0x89f0fd2a, 0x37799f64, 0x1ee8175f,
+	0x598e856b, 0xb8f9cac6, 0xe562de9e, 0xd9bad8fa, 0x2fcde740, 0x760cfd08,
+	0x5103dd3e, 0xac5d6bbc, 0xe2ff4551, 0xf8b175a6, 0xc42b2853, 0xe265b439,
+	0x91b17a9c, 0x47bc538f, 0x8d546cb2, 0xf4a3f4a3, 0xdbc5dc6b, 0x67f135a5,
+	0x2317fd4f, 0x2fcf347c, 0xfc27a88c, 0x7e36e9fb, 0x7a8cfd0f, 0xee155aff,
+	0xaef51817, 0x8c72974c, 0x2d35333a, 0x0f643ef0, 0xd31fcf8f, 0x747f7e69,
+	0x35825fd4, 0x6cba91f9, 0xdfa07790, 0x4edc0d6d, 0x653ecbab, 0xb2a2cb2c,
+	0x5cbbe505, 0x27cb85c4, 0x75f57df0, 0xa3974719, 0x92cd4fbc, 0xdb981f88,
+	0x29dfc469, 0x8accbb07, 0x5905a779, 0x6802fdbc, 0xf758728f, 0xd04b972c,
+	0x1fd4560f, 0x830017ff, 0x8000007b, 0x00008000, 0x00088b1f, 0x00000000,
+	0x7cedff00, 0x55537c7b, 0x393ef0b6, 0x526d3479, 0xa0fa5b42, 0xb4db4e50,
+	0x9494b14d, 0x27457897, 0x880b5a3c, 0x46107006, 0xf4228206, 0xbd185499,
+	0x35fde338, 0x9c414415, 0x0e7c570b, 0x42d2d37a, 0x1429a2c1, 0x20d5b16c,
+	0xb47441d2, 0x3bd15ef6, 0x0f8afea3, 0x52d25a04, 0x8ef4f987, 0x6b5adfa3,
+	0xa126d39f, 0xdf7ef515, 0x37f5375f, 0xef6758b3, 0xbdeb1fb3, 0xce275ef6,
+	0x620a3b5d, 0xd67318e3, 0x319e920a, 0x6abd54d6, 0xc614f273, 0x2c8ba68a,
+	0x67e7c242, 0xf01192b4, 0x2c7133e6, 0xd3e3b187, 0x1939db3f, 0x74339dda,
+	0x7ff41d26, 0xc637d357, 0xf3ec6064, 0x991cc64a, 0x5ea5b388, 0x698459c0,
+	0x8c09e2c2, 0x060e313d, 0xead8ca99, 0x0c0599e6, 0xeccddf9e, 0xf8bfb19b,
+	0x7981f436, 0x47a20d5d, 0x2e9c0027, 0xc60dba8e, 0xc680f77d, 0x7e3b280a,
+	0xfdda8e83, 0x153da007, 0x4faed8e0, 0xdcbfc004, 0xe86e61a5, 0xc89771b0,
+	0xb2fff418, 0xc5f7a05e, 0x90597ee7, 0x3adaf804, 0x2983aac1, 0xad9aef8f,
+	0x1efec24d, 0xd9bc61c0, 0xbc6869cf, 0x7f78fbb5, 0xc5e433f8, 0xf186b633,
+	0xe2a635bf, 0x632e599b, 0x65a7cce5, 0x69efd0e9, 0x115da7cd, 0xf87bedbc,
+	0x1827b15f, 0xc60fad2f, 0x4bb6d9cb, 0x6f403462, 0x750ef3f0, 0xef386dd7,
+	0xb4f41bfa, 0xb637d9ee, 0xeb8f5e9c, 0x5c30d06e, 0xef1bece7, 0xba02ef8f,
+	0x963351ae, 0x8e90693c, 0x0cc788e5, 0x559d8eb1, 0xb31fa4da, 0x1c75909e,
+	0x3b8c6719, 0x63fc0d7b, 0x3df49b12, 0xb8dcee90, 0x628d8c91, 0x740b5dcc,
+	0xb67e0049, 0xf00f360a, 0x4765b8a5, 0xf3d616bc, 0x02fdfe3d, 0xf11db6f0,
+	0xb6d73c22, 0xbd2fd408, 0xad62c38d, 0x60aff16d, 0x0aafc51f, 0xb0696aa6,
+	0xb9ce32bf, 0x7df10271, 0xbfef3373, 0xb5e015b1, 0xac5f8fc3, 0x0ac6bedd,
+	0x65dd4a70, 0x72b864ad, 0xb17e651c, 0x739c782d, 0x8e857cb2, 0x2c1a2f97,
+	0x10bba46a, 0xaf7e4569, 0xadbcb817, 0xdc799bbf, 0xacc94f20, 0x87406042,
+	0x9013e644, 0xbd29c0df, 0x86fce41d, 0xcb55903e, 0xf17edd20, 0x1d5b0475,
+	0x6c99c740, 0x3b3e1064, 0x7fa82922, 0xa83529be, 0xecc4a6fd, 0x52b9f6a0,
+	0xdf3e105e, 0xff505d72, 0x4199d605, 0xe7d3adf8, 0x685ff506, 0xdb84185b,
+	0xeb8d4998, 0xba5cad6f, 0x999073f8, 0x041d2b57, 0x70287a15, 0xd8c54e86,
+	0xee0bc13f, 0xa036ec9b, 0x19288ccf, 0x3e397367, 0xd8d1f002, 0x06ddb37e,
+	0x20919d74, 0xc7016ce3, 0x05f69593, 0xe538ffbc, 0xfb785bfd, 0x2dfb4a35,
+	0xd2a27dbc, 0xafc72b7e, 0xf1eed467, 0x03776adc, 0x8ed1ebfe, 0xb6f8096a,
+	0x138aebe4, 0xfde9a023, 0xa77745aa, 0x45583f02, 0x18d36306, 0x0034875b,
+	0xed182ed6, 0x351a4a78, 0x7ff4aed8, 0xe357b255, 0xd14bdef3, 0x63226cb9,
+	0x51dfc80d, 0xe6321f3f, 0x716779a3, 0x1159b2d2, 0x33ea1f06, 0x0cf4d08b,
+	0x631a51e8, 0x0db18d82, 0x906f4831, 0x0477a213, 0x89a55f41, 0x8995ebe0,
+	0x94a8df04, 0x63336ad8, 0x821695ed, 0xf4ad2b27, 0x74e554fc, 0xda576f82,
+	0x5953be08, 0xd2a3b048, 0xe290504e, 0x9f1872d8, 0xa23eb1ed, 0xd5b87a41,
+	0xfcc2cf23, 0x7e40cff8, 0xd4c64286, 0x1722acec, 0xcabf79d0, 0x65f48ad6,
+	0x6ce98c6c, 0x5ef3e0d4, 0x815243a1, 0x46a7bade, 0x0f40eb58, 0xf37a0f8c,
+	0x4617a04c, 0x11bac1a7, 0xe59c6168, 0xbb43524b, 0xbad69d71, 0x029fda11,
+	0xf587ab73, 0xda991ad4, 0x1cc9f602, 0x34f1100f, 0xf7aa8d0e, 0x7fec0552,
+	0x1ff5045f, 0x0236c603, 0x196825fa, 0x7de0f7c3, 0x1ffd8116, 0xdfd81163,
+	0x684024fc, 0x10e1adaf, 0x37f48224, 0x43cde741, 0xe2e09efb, 0xd555d8eb,
+	0xee304921, 0x5c7ca9ae, 0x29f13158, 0x8119cc5f, 0x499fb2f6, 0xb8394c23,
+	0x005abd1f, 0x83d5e881, 0x2bb009df, 0x45034f59, 0xdb24f402, 0x01d4c113,
+	0xbd4d757a, 0xe09f0829, 0x3fea0c4d, 0x6a0a59b1, 0x8259f3cf, 0xb49f27da,
+	0x3b53e106, 0xbff507a6, 0x105b43f6, 0x61575d7e, 0xcfebff50, 0xfef083d9,
+	0x618fd81c, 0x94defe78, 0x9ff50428, 0x703ade0b, 0x89aac67c, 0x232df33e,
+	0x1bf79e83, 0x1b869d38, 0x78358177, 0xb71f855c, 0x7f1e0f4e, 0xdc782da1,
+	0x3d87cb1f, 0x906a27a0, 0x13d07aff, 0x4f41fb84, 0x0de720d4, 0xbf8827a0,
+	0xd0827a08, 0xcf827a0b, 0x209e820f, 0x827a04de, 0x13d011f8, 0x4f419bc4,
+	0x3d051e10, 0xa357e7c1, 0xef3cbb57, 0x53de7949, 0xed8cbcf2, 0x8d5d3a20,
+	0xefcb6f2e, 0xdfbf23bf, 0x079bef81, 0x30bcb0e5, 0xc835f543, 0xf398ccf3,
+	0x87582eda, 0x01575d6d, 0x6f7d4dc6, 0x8246ac8a, 0xc9cf0e6e, 0x5aab8c4a,
+	0x9456d962, 0x5bdf7b5f, 0xa70fc42a, 0xfb42b69b, 0x78ffe696, 0x4dfddb07,
+	0xcdf184a2, 0x459fcdeb, 0x5f3d38e3, 0x2c753247, 0x3459f7be, 0xdf654bc6,
+	0x42b7c230, 0xb9c5fa3b, 0xe5f3228d, 0xb57bbaa0, 0x5663ded8, 0xe3168f70,
+	0xa3437b33, 0x5a52f916, 0x6fed48df, 0xedc1357a, 0xed41d5f5, 0xb00fec26,
+	0x4689ed54, 0x56685e7a, 0xff51f3c6, 0xc3f73332, 0x6f8fea17, 0x5bd61ebe,
+	0x562ab8de, 0x3697f584, 0xd77f17bc, 0xe2ffd00f, 0x1c721791, 0x0b797ef0,
+	0x1c18678c, 0x2345bcaf, 0x65fd7ce3, 0x1882c6e6, 0x621a1ba0, 0x87acfcb0,
+	0xc01b676d, 0xd736942e, 0x577c6195, 0xc343b96a, 0x5ebd3e80, 0x01d710b1,
+	0x71062a7f, 0x9189995b, 0x7ef1e87f, 0x8c177d15, 0xc4f79afb, 0x7a87e82d,
+	0xe3478f5b, 0xd04dbdbc, 0xabbf803f, 0xb11c7953, 0xfc4b553a, 0x1e3a69eb,
+	0x16fb412d, 0x774e71ef, 0xfff68fb5, 0x02f78cba, 0x8b377a24, 0x5afa2e3c,
+	0xc6507c45, 0x9a33a173, 0x4619ed8a, 0x97a757dc, 0xaa7f77cf, 0x579b9fc4,
+	0x9ff5c6ad, 0x55365c4a, 0xae05fb24, 0xf1def01a, 0xc2b6ebf3, 0x30bd4a52,
+	0xf7f210ee, 0xbb6e3ca8, 0x69bf471c, 0x1a3ed43e, 0x30aa7ee8, 0x807c8f36,
+	0xaa636fdc, 0x6815fa8a, 0xe80ae61d, 0x8c9069d7, 0x160fcf28, 0xbf911ba6,
+	0xe7c423af, 0x0eb5bb85, 0x7c4d2580, 0x47534ebe, 0x675dca34, 0xe31164d3,
+	0x7fbc6551, 0x901de791, 0xb9be01ef, 0x9e454f89, 0x338e036a, 0xfcfe3fc8,
+	0xbd1354fe, 0xdef3ca77, 0x7f603b92, 0x2b28969d, 0xcb5da5f2, 0xa5fd708a,
+	0xbe7160e8, 0xa46ae4d6, 0xfddf3283, 0x2d7a3f32, 0xbfabe22a, 0xbbf72359,
+	0xf5058ea3, 0xd57e70d3, 0xe4c7af06, 0x3bfbf339, 0x73d97fa0, 0xbef8c322,
+	0x8f11534c, 0x20fcc7fb, 0x9d32dd68, 0xf8be34cb, 0x2d7900f8, 0x1bd0196b,
+	0x8b79c903, 0xf11227a2, 0xa763e153, 0x4aebe445, 0x2c1879d4, 0x8eb1acb9,
+	0xced41ae4, 0x4d3907ad, 0xff20ea8e, 0xdccffc97, 0xfc0ad54f, 0xfb2559b9,
+	0xb5c7154d, 0x3e2fd100, 0x7853fb27, 0x1f00999f, 0x00146c0b, 0x4fa614d7,
+	0x917c4155, 0x21e6362c, 0xb6a9753f, 0xff9854a6, 0xc87ec97d, 0x6489f142,
+	0x553e43b4, 0x66e0d97e, 0x7efed023, 0xdbce24f4, 0x047904df, 0xd11e5bcc,
+	0x9073ef3c, 0x282b7ecf, 0x7fb7ac08, 0x5ce28697, 0x00833b10, 0xe99acd71,
+	0x94be5f5c, 0x8844d31c, 0x54abcb17, 0xc832afb4, 0x5f98f4bf, 0x62eef6c9,
+	0x28adea63, 0x452765a7, 0x2be60453, 0x9d7fd4de, 0x86837881, 0x57a72af2,
+	0xff30fab0, 0x255a326b, 0x9deacfc0, 0x9d379869, 0xe2d3eaad, 0x3e21677b,
+	0xcbcf9c7f, 0xa7cb4f50, 0xd4519f73, 0x193da3bc, 0x57e60960, 0xafc8cfe2,
+	0x3a99f242, 0x8c75c7fc, 0xf21539fd, 0xb26f8c1a, 0x37c63659, 0x048ce2f9,
+	0x7f995fd2, 0x1710276c, 0x13f5a612, 0xdfa809ac, 0x3cd2d45b, 0xab6f77dc,
+	0x71f484c6, 0xc62a1db5, 0x22ca7b75, 0xafd82aad, 0xa251c373, 0x815fef63,
+	0xf65710ce, 0x33ca3226, 0x67a7197b, 0x6797336f, 0x44f04697, 0x9e20fbe0,
+	0xe0845e34, 0x5ea682a9, 0xd9347f77, 0x510ac621, 0xcb2dd7fe, 0x68bc6150,
+	0x80337203, 0x57b91447, 0xf5c0a5fc, 0x977546fc, 0x7bde1238, 0xe7fdc343,
+	0x48d353df, 0x9968e7ee, 0x46f5f103, 0xc62bf92b, 0x887de8aa, 0x52fe28a7,
+	0x67c435fc, 0xc7254f14, 0x147e497a, 0x6d50bc3f, 0xab8e04d2, 0x821bdfbc,
+	0x0580f32b, 0xe0ede3e1, 0xf8c0def1, 0xa6fc7923, 0x1e740f8c, 0x6afe7481,
+	0x70fc958a, 0x91f042ca, 0xa41577a4, 0xf59e8a77, 0x01cd5c80, 0x3810cab9,
+	0x1d76cabf, 0x728073a3, 0x00e741d6, 0xde7cd105, 0xf11f9c02, 0xd2130e48,
+	0xedffda04, 0xdb5e9c69, 0xa7d8a864, 0xc4b66913, 0x698d8034, 0xa469318e,
+	0xb792261f, 0x408cb2f5, 0x35d0927c, 0xf75d7d84, 0xf06e7e27, 0xda3dc815,
+	0x08575f99, 0xfe203992, 0x6ed3fd29, 0x69c34b2d, 0x18161d6c, 0xbd7188af,
+	0xc5d9e451, 0x1b95b2bb, 0x303992f4, 0x7fc3b2fe, 0x72c567a0, 0x279e117b,
+	0xd8e0fe40, 0x27992f10, 0x8f627bc3, 0x4fd73c16, 0x73577396, 0x73d3ee30,
+	0x0519f379, 0xffc9c60f, 0xf5e72694, 0xc8a7fcb2, 0x10ed58f2, 0xabdf683c,
+	0xee1355bf, 0x90782e5e, 0x8784bf07, 0x45e4621b, 0x99a2bf62, 0xbf47f41e,
+	0x8ef29404, 0x2fa4719d, 0xb9fb08d4, 0x2fffde4d, 0x0263bed5, 0xc684e7ec,
+	0xc2f98a9b, 0x2b08f289, 0x418cbf8e, 0x64f60180, 0x7820eb39, 0xd879ca3a,
+	0xc7bb8347, 0x81ff3c62, 0xf8c56743, 0x8d537aa7, 0x2e66ed11, 0x3cc264db,
+	0xff514805, 0x9f1fe4fd, 0x0b8830cf, 0x5f0aff90, 0xcbecd31b, 0x073f00d2,
+	0x749c493c, 0xefa5fbe5, 0x3e468e1f, 0xd0e6ff76, 0xda19db92, 0x3bb23e9d,
+	0x1a551525, 0xa8c41c9c, 0x906d32c3, 0xe0bf0abf, 0x1f1052df, 0x90630f61,
+	0x42cf622e, 0x177ffbd2, 0xabb1fa19, 0xf3a151f1, 0xac7e642f, 0x21f6e16d,
+	0xdd0abf94, 0xa21e676f, 0x5daff81f, 0x23a617c4, 0xffca147b, 0x0407f1ab,
+	0xcd363971, 0xe431d38e, 0xdb8ab26b, 0x2035e4ab, 0x3beeb02f, 0x0bada398,
+	0x8fa892e0, 0x32e72434, 0x7ccb969f, 0x167980fa, 0xdfff5258, 0x9a44e9f1,
+	0xe91a7b63, 0xfd4869ad, 0x32c7d50c, 0xe2aefac9, 0x67a43c6f, 0x3579fac1,
+	0xf85664bd, 0x8e7f802f, 0x7fa4d10d, 0x625d7f0a, 0xd5f7e903, 0xad18f3cf,
+	0xf059bcab, 0x847f52f8, 0x10b00a5e, 0x884363e0, 0x1afc2a3e, 0xf6291ece,
+	0xfb96b4ec, 0xa27df763, 0x8ff2f307, 0x1309df05, 0xceeca7da, 0xd282640c,
+	0x8eaae3ff, 0x38c0f7b1, 0xf0e4cf1f, 0x0159c634, 0xad922db8, 0xa49e23b4,
+	0x3f7ee373, 0xa167b216, 0xf858fe7f, 0x9e385be7, 0x5f1095a0, 0x04cb3096,
+	0xef0d69f8, 0xc57dd8af, 0xecbc7fde, 0x4fd439da, 0xbee06f60, 0x47a25c60,
+	0xa782fddf, 0x256a4406, 0x5f5c6a98, 0x0aa7acea, 0x27f9e0d3, 0x43cf4dbf,
+	0x05d6d87e, 0x2e7e878f, 0x3ce22275, 0xebf1baef, 0x9caacd3e, 0xfc0679c1,
+	0xf44b9cd4, 0xfd53f15c, 0xb25a1965, 0x2833ab53, 0x69922fde, 0x1ec98b6d,
+	0x87f829f1, 0x08fa4f7e, 0x5fa0ce55, 0xc8793fe9, 0xefa008e0, 0x4c3bf079,
+	0x3f67801b, 0x97e81296, 0x3479ed8e, 0xa451f913, 0xb9e2e775, 0x3feab9e3,
+	0x2edfa0a4, 0x2670f3da, 0xb67b8dc5, 0x5ffc6264, 0xa668f354, 0x0728f08d,
+	0x4e6a1fe5, 0xd703ee87, 0x101adeed, 0xf26526a7, 0xc47979f8, 0x1f8c7cb1,
+	0xf1735de7, 0xcbf41321, 0x4260adf8, 0xeb35f831, 0xddfd0878, 0x77970b0a,
+	0xec8233b6, 0x43678c31, 0x2f65d7be, 0x3d7d45a8, 0xb34cf965, 0x74bfe31a,
+	0x056bd135, 0x804b6cf3, 0x6088f676, 0xfd8163aa, 0xff6356c7, 0x6d049b55,
+	0xb1fa1c62, 0x97987991, 0xd247d2e9, 0x247cee3a, 0xaf9f1daf, 0x9f3f8078,
+	0x7534f3a7, 0x694f38e2, 0x37a475ad, 0x278c34f6, 0xd2ebf595, 0xcadef4b5,
+	0xcbc5026d, 0x4f193cca, 0xee0757c6, 0xfdce9753, 0x8db29b32, 0x32adcfec,
+	0xe07e6e5f, 0x611237bd, 0x1d6349dc, 0x7f6874c6, 0x66b389e7, 0x139f3c1f,
+	0xf3073e73, 0xe3ad0c86, 0x77dc62d9, 0xfb8ace8d, 0x2c9b9298, 0xdfff6026,
+	0xe11c7ccc, 0x3d5607ca, 0xf2768c5b, 0x73ca08e9, 0x13adf594, 0x83e61524,
+	0xbf6fabc7, 0xee76e913, 0x778a0889, 0x256755d7, 0xdfc81fb0, 0xa5ff02f7,
+	0x41f6c64c, 0x05f08efe, 0xd825dfc8, 0x44ca81e2, 0x94aa5fc2, 0x5ca5b208,
+	0xb899d6be, 0xbc6e193d, 0x4b9790c9, 0x3ca64760, 0x7c41e302, 0xcd9c6f38,
+	0xcaf40e92, 0xb35e5f18, 0x0657de45, 0x4ee79fbc, 0x98748d9d, 0x769ee5b2,
+	0xd2e31d0c, 0x56f2fe79, 0xa82923e7, 0x381f6bc7, 0xb4317cf3, 0xe7b74aaf,
+	0xbb11fbe1, 0x7e3e512e, 0x15e1e5bc, 0x1aa2ef48, 0xbc9fdf1d, 0x644f73ef,
+	0xe78b3c26, 0x286a6f75, 0x3aed763f, 0x8016fcf3, 0x47bb6d77, 0x3fef281b,
+	0x7113bf76, 0xba7f533d, 0x97470d9e, 0x8f67ae9f, 0xd53e90a6, 0xfce59e80,
+	0x3d733d7c, 0xdcf44550, 0x15ff6e9c, 0x8d39dbca, 0xebf5053b, 0x240df65c,
+	0x8136eef7, 0x5fd8a1f8, 0xf2a5fde5, 0xde554bfe, 0xce02336d, 0xc6432f15,
+	0xd63ef486, 0xc62665f2, 0x979b97b5, 0x47dc088d, 0xfb023c28, 0x35864c86,
+	0xda7dbd03, 0x69a7db2b, 0xd856e91d, 0x8f1534df, 0xabdb4367, 0xbdb255d2,
+	0xcf0ffc85, 0xc5ff1433, 0x0f78b43a, 0xf82ad651, 0xaa3ef122, 0x6af3a6f9,
+	0xd81257ba, 0x7de741a7, 0xe01a7d8c, 0xe75b8777, 0xc713a0d5, 0xe3da0f51,
+	0x31543f6b, 0xbdbf805f, 0xf9922bae, 0xc81de602, 0x114f6023, 0x4b7f1fc0,
+	0xd87d9052, 0xe7042e9e, 0x36aef013, 0x15f21fa0, 0x2dfc65ef, 0x9f8835d9,
+	0xfb3e72d7, 0xc47c408f, 0x389849cf, 0xe71e906d, 0xd173ba5b, 0x70f42b2e,
+	0xb926ed1e, 0xf87e6b29, 0x99f44092, 0x7b9d5e7b, 0xdb1cfa75, 0x7c8b9dd3,
+	0x9ae9af3d, 0x5f947986, 0xfc8b95db, 0xedc2be71, 0x93a54ffe, 0x11f8bcf2,
+	0x05d800d8, 0x3ca260a7, 0xa1ec5f2f, 0x927fdcbc, 0x96cee5e7, 0xfb7764dd,
+	0x88b9f95b, 0x9051d54b, 0x1508e780, 0xeed1da67, 0xfcf227f5, 0x21b7c609,
+	0xcba466bd, 0x97dd53fe, 0xdfec10b2, 0xe1390189, 0xd439da79, 0xe38e2d73,
+	0xc17b432d, 0x1d58a372, 0xf501778c, 0xd8af9e1e, 0xe519ba90, 0xd4679cff,
+	0x6fa3b424, 0x49ddfee5, 0x45af68e7, 0x638a4499, 0x7f0069e0, 0x11e1f607,
+	0x1e282d99, 0x80b6628a, 0x74ad2e7e, 0xa7bfc52b, 0x74c7ba44, 0xc67ae78d,
+	0x2b3d728e, 0x9b4c74df, 0xa35e3018, 0x7894f76c, 0x7f6bde70, 0x345d7d27,
+	0xf023b991, 0x14f597bd, 0x2031b0e7, 0x7a2f80ff, 0xf283c139, 0xdca8f7e0,
+	0x178bc81a, 0x87fee64d, 0x7ef1d21d, 0xa454d289, 0x6af31f87, 0x4c794fa0,
+	0xc00c6647, 0x2b89df23, 0x4e4f98ed, 0x2be01c02, 0xf2cf9702, 0x23991d45,
+	0xd0495cf1, 0xa4bcf0ab, 0x77b786bc, 0x54bc78aa, 0xf7dfc69e, 0x7b57dc0f,
+	0xde821653, 0x9cbdb6f5, 0x5be02772, 0x9c35be0d, 0x7c4cbe0f, 0xb99eda0e,
+	0x46efe608, 0x5cf74d9f, 0x5ed08be5, 0xdc151b75, 0xf78c71c9, 0xb5cd9ee9,
+	0x5baae508, 0xbc15e0af, 0x5c64105d, 0xe5c15f79, 0x8c8547f7, 0xb4859237,
+	0x0a4ff487, 0x3dca35c1, 0x58ff3c9c, 0xb517fefe, 0x7bc22cf7, 0xcf3af9b7,
+	0x57f44ac9, 0xd218692f, 0xf0e07f55, 0x697fa459, 0xb33bf70b, 0x39ee1c0e,
+	0xd872f288, 0xb1e6f748, 0xe3e254e6, 0xdbbd96c6, 0x375ca126, 0xada073f2,
+	0xff3fef0d, 0xc3a7ea3b, 0x57183163, 0xe69eb759, 0xebca2f9d, 0x1e39eb66,
+	0xdfb3b725, 0xb2847c72, 0x3c71dfaf, 0x6b7fb796, 0xa75a17fe, 0xfb81f0ff,
+	0x9fa05e50, 0x17a5e29b, 0xa2605e40, 0x73d94392, 0x1bf59e30, 0xfce0563b,
+	0xab3afdce, 0x09e7c25b, 0x95be30b3, 0xa4dcfccc, 0x14f6fc78, 0x4754a7cc,
+	0x1df9edf7, 0xd7efbb9e, 0x2ba139d3, 0x12a5db86, 0x3dda7ab1, 0xf7bb3f24,
+	0xb5f701fc, 0xf6e74f4f, 0x1edcd1c5, 0xa0c9ef15, 0xfe579ef3, 0x940ad9f1,
+	0xd7717c83, 0xd7fe7c1e, 0xe90abd5a, 0x8db5a9d4, 0x4cfd01b8, 0xa9e286a7,
+	0x3e2f1962, 0x13d7cb1d, 0xf24055d5, 0x653db713, 0x4948c811, 0xe0b7fdec,
+	0xdb8be7ba, 0xce7f479f, 0x0ce8eaf6, 0xa03efceb, 0xbe77bc2b, 0x2da7434e,
+	0x35467dfe, 0xb0b76bca, 0x38b3c57d, 0xd7ef1d7e, 0xf44edfad, 0x3165d776,
+	0x31fa0f9c, 0x16ea9f3c, 0x3f3d0476, 0x1737b75f, 0xfb0bade3, 0xa5dc53e7,
+	0x97a293e8, 0x20bab525, 0x293fc7fd, 0xf963f31e, 0x8666e4cf, 0xc2e9fc80,
+	0xcffc6791, 0x85f5e806, 0x0b4f8d06, 0xb1d62a34, 0xa7af784d, 0xf24963e8,
+	0xbfb8854d, 0xe6755512, 0x278bc03a, 0xbd40ad77, 0x1167af3c, 0x2defd089,
+	0x2e0f3f15, 0x885aaa69, 0xf5b87b76, 0x7c799b14, 0x9f913979, 0xd66e5f97,
+	0xa9cf027b, 0x9547e143, 0x3d5b7a4c, 0x42baff88, 0x19473e47, 0xb7ad2def,
+	0x670ee30c, 0x9fc3f70c, 0x3306ff40, 0xee6661c4, 0x086e56a7, 0x1cdcbfee,
+	0xda9dcf43, 0xcbb44cae, 0xcccbf8ca, 0xd4cf6e26, 0x9fe8995d, 0x7923df43,
+	0x82f57b47, 0xa4aaf640, 0xf7433849, 0xf8d248c7, 0x44967a41, 0xee623db8,
+	0x9c487cd0, 0xc601f326, 0x77ac87fa, 0x9b7a0d25, 0x8d29c61e, 0x7f468aec,
+	0x8aec6655, 0xff6f73c6, 0xc8d59151, 0xc6538bec, 0x4f9c1b4f, 0xbd8b8fc9,
+	0xa87e9b1a, 0x183b60f9, 0x7fc5adef, 0x32622903, 0x3514deb8, 0xcc7cd147,
+	0xe50d35d4, 0xf05768b6, 0x534fb87b, 0x9bfa3e80, 0xc607c777, 0x75c7bfa1,
+	0x8677de53, 0x1ff7a6f4, 0x1ca0a69f, 0x26d8a894, 0x20bee73c, 0xc0617bf0,
+	0x82f85ee9, 0xf25e5e7c, 0xca244337, 0x65cb85cf, 0x01fec7a4, 0xd928fb3b,
+	0x6762bf07, 0x15c70d59, 0x7076cb3b, 0xeecec07c, 0x3762b8e2, 0x42a3fe95,
+	0x1d0a4fde, 0x9077a94f, 0xe6ed93c7, 0x6e7abaa2, 0x7f7e4537, 0x4c6b76c7,
+	0xfce27f3c, 0x19d32b5a, 0x95d3b412, 0x123577e4, 0x19c5b1fa, 0x76a579ca,
+	0xa0e27e90, 0x7499fa3e, 0xbbf509a0, 0x10f940c8, 0x070d15c4, 0x59791cf1,
+	0x07ef15bc, 0x9caee943, 0x7cc5c777, 0x7cf906b3, 0x8af1d6ab, 0x7269ef19,
+	0x82afc8ba, 0x74bf90b6, 0xb8a3aabe, 0xb7d357aa, 0x965fed07, 0xf61af1d6,
+	0x6c76d32b, 0xea4b029e, 0xe3b574d1, 0x943cd091, 0x5c43b257, 0xbb39fd5e,
+	0x39527ce9, 0xe6577f2f, 0x5ea70323, 0x7025f28d, 0x7cde3939, 0x57b46de8,
+	0xf2748efe, 0x03e45703, 0xa3478fcd, 0xe210f9dd, 0x5c407714, 0xbdaae1fc,
+	0x14b50f74, 0xde011fef, 0x7f7829b9, 0xa62edffd, 0x0b6d8838, 0xbcf28e78,
+	0xa50f1833, 0x2fb5ca0b, 0x24dc8f48, 0x9ba5dfc7, 0xeace9608, 0x061d2270,
+	0x8b8045f0, 0x21ebf801, 0x667f4878, 0x1e6266f7, 0x08d61dfa, 0xc6ca3d01,
+	0xbc534efa, 0xd0c3407a, 0xbf046c3f, 0x28e1f849, 0x1d532baa, 0xc718bf00,
+	0x9bfa22fd, 0xca0ee8d2, 0x7587fceb, 0x60f0df74, 0xb82997cb, 0x8ba6bf7f,
+	0x96d295f2, 0x9e31e397, 0x344b358d, 0x78dff83e, 0xdc3d8fba, 0xb6318f12,
+	0xe54cbcf2, 0x94eb94cd, 0x28fed172, 0x5f45cbcf, 0x54bfae26, 0xf6c5c8e8,
+	0xd453b458, 0x7111701f, 0x1328fb46, 0x2d60ddb8, 0x5089f922, 0x706745be,
+	0x25f60bbb, 0x8f11ee97, 0xb15978f2, 0x605f6427, 0x79f1d7bc, 0x261fb70e,
+	0x5e14fce5, 0x31bb462d, 0xa4d99d84, 0x41564dd8, 0x06644d79, 0x3ef2b75f,
+	0x30f3cb8d, 0xd2be776a, 0x83969b9d, 0x7bf249ef, 0x7d2cf601, 0x1da13ed0,
+	0xf23b2449, 0xf61e26a0, 0x60bcf255, 0x14af08dd, 0x802aaf2f, 0xfaad6cee,
+	0x2fca029e, 0x19ee9b99, 0x8575d742, 0x7f39997e, 0xd2f3e44b, 0xbcfc3d92,
+	0x9f8682f4, 0xf2bd832f, 0x3b5afda1, 0xf500aaa7, 0x72b8eeef, 0xdcb9e06d,
+	0xb35fc6d5, 0x5b5f1833, 0x30905551, 0x6c979e84, 0x8a3ed1d8, 0x1d17e10d,
+	0xd2f105d5, 0x7595cf95, 0xeb2bd05f, 0x1519a59c, 0x55a58f48, 0xae00a305,
+	0xf4b9f88d, 0xd0befbe0, 0xdfa1f3a2, 0xf9d2fa06, 0x575f8de9, 0x5c3f9d04,
+	0xe662bc6d, 0xe1c2f3a1, 0xcb0a1fd6, 0xcc69df04, 0x57e818a3, 0xe819e91d,
+	0x2f7cc04d, 0xb4fbfeca, 0x9e78a04a, 0x871aad15, 0xac7301f9, 0x6b7e871a,
+	0x0c0178b8, 0xdd1568e5, 0x8209ba49, 0x0eb0ae1e, 0xffa50f41, 0xbcfc191d,
+	0xaf5945e1, 0x7d2fcf46, 0x9fb420e6, 0x19456c96, 0x5c28a8c0, 0x8945e623,
+	0xda183c23, 0x7203be53, 0x036394ed, 0x5f7cab43, 0xbb522858, 0x586459ab,
+	0x8d12caa7, 0x57d80c9e, 0xf54edb21, 0x7798f37b, 0xa9791506, 0x3d7f7ebb,
+	0x4d7fcfb4, 0xd195bc9c, 0x37e50f7d, 0x330e9105, 0x5fb055c6, 0x4ef43f39,
+	0x7e701859, 0x37f647b7, 0xfc814433, 0x14525f40, 0x5dced39d, 0x503cbe5a,
+	0x10853b4e, 0x72fa79ba, 0xf5f1bf29, 0x6dd206e5, 0x15516bc9, 0xbcc80e74,
+	0xcc049653, 0xaf0672f3, 0x696af89f, 0xb8979d06, 0xd27bf32e, 0x27dbf17c,
+	0xd7588f31, 0x80af7906, 0x407ea7e3, 0x2f906675, 0x0dd09c62, 0xcc1ce9f9,
+	0x66cc1df5, 0x61387b43, 0xae7be514, 0x7ac30203, 0xae8b6733, 0x9d9bef06,
+	0x8c78ae9a, 0xbf34b673, 0xec3f45af, 0x569efc3a, 0x92abbcc2, 0x2637f5c9,
+	0x05d7d215, 0x5bbf40ec, 0x01f59cde, 0x93f2b5da, 0x5cb0d4fc, 0x3fdd1efd,
+	0xce14b9e2, 0x48d4da57, 0xcc658df7, 0xd65f7247, 0xf6ff9c10, 0x075d0e27,
+	0xa76cb2f4, 0xd3ee7c14, 0x225fcf3c, 0xb72f0ab2, 0xce9f148b, 0x73a04c4d,
+	0x38f7c1d8, 0xf7626d7d, 0xdb6e7843, 0x0558269b, 0xd237aa18, 0x26e309bd,
+	0xc73f2677, 0x85f57ded, 0x7ce86b29, 0xeb0533c4, 0x73e57ba4, 0xec8fb9fc,
+	0xeb018b50, 0xebcadb99, 0x59237a17, 0xbbe211d1, 0x23f20c63, 0x92ded52b,
+	0xb6967181, 0xf1eb9def, 0xaf8e59de, 0x57b1f743, 0x253e6fbe, 0xbacae9c7,
+	0xf32727e7, 0x21e618f3, 0xf3976a39, 0xda0a9b62, 0x4cd3ee03, 0x77528299,
+	0x046befc7, 0x9c3e8f7e, 0x17bd617e, 0xbd623bd6, 0x72fdc217, 0x7bd6245d,
+	0x5ca2fe21, 0x5ef58917, 0xbdeb1cf8, 0xeb926f10, 0x0bdeb122, 0x42f7ac71,
+	0x8bae5478, 0xa54b853c, 0xbfd2eef8, 0x3edb9e61, 0xfc721f10, 0x3e084c74,
+	0xe9ffb83a, 0x1f182118, 0x33ff502d, 0xcdfe9775, 0x1675839d, 0xf0dfd926,
+	0x9ce50b3a, 0xb670b3ab, 0xb1f20779, 0x3c24c275, 0x69266747, 0xb74bea04,
+	0x56ffe7cc, 0x1da02fdb, 0xe47d1160, 0x64c4e797, 0x78bb408f, 0xba62e4f2,
+	0xd38feb87, 0x93bc3e91, 0x7ec3e8ee, 0x9e6f98ae, 0xeb06cc3f, 0x265a66e6,
+	0xfc029a6f, 0xe4106998, 0x44fb7238, 0x059f72bb, 0xbadd99fc, 0xdb9a8a6a,
+	0x994e130b, 0xc6ad962f, 0xe925d771, 0xaa1e6327, 0x1dfc734b, 0x5fe855c7,
+	0x19ad7a30, 0x48d76fe7, 0xeabc601f, 0xf3cf13fe, 0xa501fa95, 0x33a927df,
+	0xdef099f5, 0x26fa4b5c, 0x2819ebcc, 0x4dbc8a1f, 0x3e501acb, 0x6128aa6c,
+	0x4933d923, 0x67d5fb8d, 0x92cefdf2, 0x4f80fdf8, 0x6063a92b, 0xcce45f7c,
+	0xdb77cc38, 0x949fbe65, 0xe64a7ef9, 0xcdd1d63b, 0x3a3af351, 0x87475884,
+	0x20ea10f0, 0xdd2fbfcc, 0x7149931e, 0xd20a5f72, 0x984ba95d, 0xedc67a87,
+	0xf5d20a4f, 0x27c72c90, 0x99d3fb64, 0xefe24f66, 0x54acf8f1, 0x96e81fae,
+	0x7d9473cf, 0xbe0def2a, 0xfb05e7cf, 0x6ffee12d, 0x5f645510, 0x5b38e543,
+	0x788efb96, 0xbd4abde3, 0x3739c608, 0x261fdc88, 0x41dffaab, 0x6f1c93bb,
+	0x90c23694, 0x78a3b3f9, 0x09bd36bf, 0xdb6f6c4d, 0xddce7437, 0x296190fb,
+	0x30314fbf, 0x3da11efc, 0xe04d4271, 0xe51fe979, 0x908fc079, 0x9a2cba5e,
+	0x3923f61f, 0x423f01c6, 0x144e66f9, 0x9bf923dd, 0xa4fca2bf, 0x09aa75b7,
+	0xe1dfd4e5, 0x715fa031, 0xc4ffe3fd, 0xb7f0bfa3, 0x6cf5417e, 0xa4bf6ffe,
+	0x722db73a, 0xcdebe1bd, 0x87589d72, 0xbde7c464, 0x60159f85, 0x1c32adbf,
+	0x3cfc23b4, 0x686f814f, 0x681de66c, 0x51509ecf, 0x8dd8a2fe, 0x98597d50,
+	0xf78e5239, 0x676831ec, 0xc4ec5f40, 0x597d01b7, 0xc1798316, 0x12917da0,
+	0x132ea87e, 0xcd598bed, 0xb4420bed, 0xb733662f, 0x7da3882f, 0x05f68841,
+	0x417da39f, 0xe20bed1c, 0x47105f68, 0xed1082fb, 0xebff3e0b, 0xd3a75e52,
+	0x3f29810b, 0x5c5035c7, 0x7f8ebebf, 0x8dbeb27f, 0xafaddbdf, 0xac0ea816,
+	0x94dfe607, 0x509d4439, 0xe4275c93, 0xa2f741dd, 0x15a2c3ae, 0xd6017859,
+	0x6dcf09bd, 0xd0277d35, 0x051ab82f, 0xcf84956b, 0xb63ad297, 0x60fd4eb4,
+	0xd3af3e39, 0x2fe85558, 0xa3c72f5a, 0xad692bdf, 0x2735ff23, 0xbfcb4da1,
+	0x9e181197, 0x71d12f1f, 0xe11728fd, 0x3fba24e9, 0xae5ffb26, 0xad73a0eb,
+	0x0a417ad2, 0x35c08efc, 0x7fc512fd, 0xf2b9fdca, 0x57b94bdd, 0xca0ead2f,
+	0x6ebc6c9f, 0x4aff718b, 0x9e406579, 0xc3277cab, 0x74d56fc4, 0xc08afe3f,
+	0xd7ab8562, 0x36b49573, 0xaffac66e, 0xb4767bd0, 0x16d75424, 0x44febf72,
+	0x95d75ca6, 0x6f30f2c7, 0xf5627c2e, 0x5f30c381, 0xe51d76ff, 0x774f609d,
+	0x409efea1, 0x9cba0439, 0xb9d3d368, 0xb23be7db, 0xde1f1046, 0xf6489914,
+	0xb7d90c72, 0xf2811a75, 0x3597b031, 0xc2ecdef8, 0xceaec8bb, 0xfc8abb06,
+	0x93eee454, 0xca2d7b54, 0x1209bae7, 0x1d7cb9f9, 0x93c697bb, 0x9200c5eb,
+	0x146fe5de, 0x57c02a3c, 0x2f7d2b31, 0x86c812b4, 0xc6af50f1, 0x75d608f0,
+	0xaebc64ad, 0x75a39143, 0x8eef943b, 0x577d10c6, 0xd95cf4c8, 0xb06fc7fa,
+	0x1d2ee36e, 0x56490f3f, 0xe4803af1, 0xe7bebcad, 0xf5e0e66e, 0x8b4d9872,
+	0xdabd2f94, 0x511dcf05, 0x79f85263, 0xe380948f, 0xa1a53aaf, 0x84fce33a,
+	0x8a0e9905, 0xfd50e64f, 0x43bbe086, 0x1f9afefc, 0xeb83878a, 0xec684e37,
+	0x3b573df4, 0x73d925ca, 0x9553c107, 0xbb150445, 0xf63fd4ad, 0x88dcbdb6,
+	0x0cbbe7f3, 0x62b7a456, 0x83a8ae07, 0xa950e48c, 0x100b1c83, 0xc0cc0aeb,
+	0xda0c5363, 0x44ec8967, 0xb273aab2, 0x1bfffb84, 0x3f200eff, 0xe30f2da1,
+	0xfc7974a6, 0x43fd92d3, 0x9a2d9fe4, 0xbc3df1b9, 0xf40eda4f, 0x080dc02c,
+	0x6d7b2f18, 0x6dbfc622, 0x1a46a982, 0x78ee8eca, 0x5c97eb08, 0x7184c166,
+	0xf6e44159, 0xbf1878fc, 0x8ad52d20, 0xccd5dc76, 0x403f2de3, 0x3debf7ce,
+	0x55dfc0a2, 0xce9c63c7, 0xea9596f9, 0x351ccee8, 0xacbe7cf8, 0x5fadbd4f,
+	0xabe3faf1, 0x38b563c7, 0xade99bbe, 0xc8f8f275, 0x8cd7f9f7, 0x914cb8a5,
+	0x8486fe9b, 0xef9b941a, 0x7cbbff65, 0xcc1a7bf9, 0x1cd33b03, 0xaff38dff,
+	0xa1e97d58, 0xd0e7da78, 0xd3a9507e, 0x4a83f50f, 0xabfff37c, 0xcaedff3e,
+	0x54ef820a, 0x51d8206e, 0x51f6a6e9, 0x735cfc59, 0xfc0ab8b8, 0x10d65c43,
+	0xbf4294f1, 0xe4aae221, 0xae3ca9ac, 0x89473cc2, 0xa388aeab, 0x0155df40,
+	0x75bb3efa, 0x608f9de4, 0x6b53b3dc, 0xeb0090c5, 0xc81bd7a8, 0x4d64643e,
+	0x529e2a1d, 0xf8a069e8, 0x27e99fab, 0x2cb3fce1, 0x1fbc6ae9, 0x0425989f,
+	0x3fa80be5, 0x1fd46fb8, 0xea3fa884, 0x39e7b880, 0x14f6a82d, 0x29b82f1e,
+	0xbf13d05e, 0x5d815737, 0x662efc01, 0x3d2bb20a, 0x31857621, 0x1d70affb,
+	0x9d795a6b, 0x3a5f5297, 0x93d7c00b, 0x5e33b086, 0x46c6676e, 0x3ce1f77a,
+	0x54919d63, 0xc3427caf, 0x16c4eee5, 0x9ab82f5e, 0x8ff7f9d9, 0x689bda57,
+	0xb15defa4, 0x01fc9f7d, 0xa8f5dffa, 0xcd9ee281, 0xfda66b49, 0xf8cdec4f,
+	0x7f0f4e2c, 0xe7de6fa5, 0x22731f20, 0x905cf7f9, 0xee2859bb, 0x7e924555,
+	0x8ed97e40, 0x6bac57df, 0x8ed0f211, 0x77d9d1cf, 0x6a9d3ae1, 0x27167ebe,
+	0x3baa27df, 0xbf42f78a, 0x2b64f8ce, 0x1263cfe6, 0xffac308f, 0xe61577ca,
+	0x64ecbda0, 0xbf7be9db, 0x5dc73a8d, 0x39c5e7c8, 0x7b9d0215, 0x5897faca,
+	0x95b35518, 0x153591f8, 0x2708bef9, 0xe6e473a3, 0x2e8f98c5, 0xf1c10a8a,
+	0x8dce8b1e, 0x6c78e55f, 0xd2827b48, 0x1cc9e32e, 0x2f356777, 0xd2e573c6,
+	0xf3e9bd7a, 0x91cffaa0, 0x39abb841, 0x6079f4c9, 0x03fe2a7d, 0x822daecf,
+	0x93dc059a, 0x078e0837, 0x85e53f70, 0x2e39f917, 0x5abf23eb, 0x7d112595,
+	0x7ccebd47, 0x737dff28, 0x902e748d, 0x87915b10, 0x472c09e7, 0x52fd8a9a,
+	0x17147d34, 0x96ab7ef6, 0x8be8b3bf, 0xef202dbe, 0x7f95be7f, 0xcfca117f,
+	0x7b970b3c, 0x0db0d997, 0x8c4ce6f9, 0x7ad82088, 0xa616e289, 0xdef19bb6,
+	0x3a0af214, 0x3f7d08b7, 0xb7efc2dd, 0x656b3753, 0x90ff813a, 0x8e337ae4,
+	0xc674d9bb, 0x948befaa, 0x266ccae9, 0x866d3ff8, 0xfcffc64a, 0x0113e638,
+	0x7fd86dfd, 0x6b5f9db0, 0xf77bef2b, 0x26f8fbbe, 0xca1f3cb3, 0xf1461ad7,
+	0xa0f7e873, 0x06625a9f, 0x2e5033ea, 0x53d38b9e, 0xf7583f9e, 0x6ba0ebcf,
+	0x7f3d72f5, 0xf83fbd68, 0x8ae7f43b, 0x725f7b9f, 0x52bf1d50, 0xf9d4aff7,
+	0xcd2f3127, 0xe027b4a7, 0xda88baed, 0x3847a1db, 0x2193f303, 0x659376e8,
+	0x91647fe6, 0x17d9b8de, 0xfee7c1ea, 0x810de8a8, 0xe486727c, 0x66593737,
+	0xfbca7bd1, 0xb1661be5, 0x26e3262f, 0xaccc38e5, 0x14e7ff22, 0xd9021bd9,
+	0xa28f5e11, 0xec1a77bc, 0x9ff7ce3c, 0x7e11f760, 0x0bfef608, 0xc19fe7a0,
+	0x08fc21de, 0xff3e1dec, 0xc21dec04, 0xf803f02e, 0x03f053fc, 0x61dc9708,
+	0xe57a829c, 0x5f62ae6a, 0x3b3dd704, 0xd3bfe302, 0x97968733, 0x9e2a598b,
+	0xdd0ea5bf, 0xe9e596ae, 0x6f9c8df4, 0x4f3377bd, 0xdfc92a4d, 0x86178c26,
+	0xd7b7eefb, 0x5afbedc8, 0x37bc1cf9, 0xb2e7e389, 0x473cfc34, 0xf7c69ec5,
+	0x3af066d7, 0x2b8ec539, 0x1fc82af9, 0x63acaf8d, 0x29d79c9c, 0x55663cfd,
+	0x0a48f7d0, 0xfde1639c, 0x5f313e48, 0x0d4c7947, 0x44a74f90, 0x172a19f2,
+	0x1d385a9f, 0x772e3acc, 0xa09768c9, 0x9e82727f, 0xe361c6e2, 0x51ef90b0,
+	0x394f90ae, 0x32c64b1e, 0xf3e1988f, 0xf7aca363, 0x7e3f3a3f, 0x943150d8,
+	0x3ff998fd, 0x3e80ffdd, 0x8affef06, 0x8e876b7d, 0xf7db873f, 0xca2bbd40,
+	0xc1f610cf, 0xb81f603a, 0xb164dcfc, 0x1d4cb78d, 0x05483bfe, 0xebe0cdae,
+	0x6a83e456, 0xc5ee3094, 0xee0fa57d, 0x91fd6067, 0xcb5bfb21, 0x94225ee8,
+	0x13df018f, 0x05fafb29, 0x1ba147ae, 0xa38a3e58, 0xa81994c7, 0xc590fbce,
+	0x4769efc4, 0x0762b97d, 0xf645feb4, 0xf1507f13, 0x6ec51a7a, 0xc4173c0f,
+	0x82730b95, 0x513339ec, 0xfd23ad9c, 0x7bbc41f2, 0x4f7830d0, 0xb7f7eca9,
+	0xcbf63eb2, 0x6fffb5f9, 0x8ff65f3c, 0x73dfa3ee, 0xefcd3ff0, 0x9bc460d1,
+	0x6eee0eb1, 0xcc253bf9, 0x9e5ad5af, 0xcf21ab37, 0x7d2f7bf6, 0xae01f782,
+	0xaa3c4467, 0x9733e262, 0xd3fc5097, 0x7e6131a6, 0xc7e9f1be, 0x9dbca24f,
+	0x7424126d, 0xd826fed9, 0x3ba7e636, 0xf43d1e8d, 0xe7e2682f, 0x45eb6b27,
+	0xfcff048f, 0x696efe42, 0xfa1a33f6, 0x234d699e, 0xfe3977af, 0x6bca7af4,
+	0xcf207acf, 0xc40b7186, 0xbcf1da07, 0xfe39e4be, 0xfb2512bb, 0x1efc5adc,
+	0xa05aa4af, 0x97c9537c, 0xaf75e4cb, 0x6c7909d9, 0x30483dd0, 0x14bccc5e,
+	0x06d4955f, 0x086b2f3f, 0xb5ef7fce, 0x94d77c0c, 0xf9c8ebc7, 0x3f62acbd,
+	0xdcab8caf, 0xfac1a7c4, 0xe3c6bfeb, 0x76d8be79, 0x0353e539, 0xf8ca17c7,
+	0xca2fc019, 0xfacaf6e7, 0x62e8fc95, 0x85e29b12, 0x2e313afd, 0x50efe495,
+	0xc8aedf81, 0x1f485bae, 0x8c268a8f, 0x8b34b6b3, 0x7967a46e, 0x7d377eab,
+	0xf4e78537, 0x3df3a446, 0xc26ff7e4, 0x273c267b, 0x83b67c84, 0xce83a6e4,
+	0x6fad99e5, 0xb697dd0d, 0x4fe84de8, 0x683f479d, 0xfd06efc7, 0x57860587,
+	0x7cefcaf8, 0xd78679ac, 0xcddfc199, 0x229afdf8, 0x5f006217, 0x436bef91,
+	0xc33e2af8, 0x8d35f3f3, 0xdedf2823, 0xf568b4f8, 0x39785ca1, 0x17be157f,
+	0x3ce8db88, 0x7c48f805, 0xdd4cfd9e, 0x07289193, 0x05fdf8fb, 0xfdf101c8,
+	0x77ce2eec, 0xc443db6d, 0x2cb57f77, 0xf7fc2d07, 0xf8ea4a49, 0x17e81079,
+	0xf1abc9f9, 0x5f7c6aff, 0x7c83db20, 0x3889c671, 0xe5ef778c, 0xe988adf3,
+	0x9a07bf2e, 0x20bb993e, 0x7975f515, 0x26c6eee1, 0x07891980, 0xce621999,
+	0x51db3a53, 0x5cb9e04b, 0x79d79e9b, 0xca045103, 0x4bfae433, 0x53dd4ce9,
+	0x5e47af68, 0xfe31f143, 0xf79c0445, 0x60cf3f00, 0xb23bd09e, 0xd8cbc0fb,
+	0x7e8e946f, 0x55bca58f, 0xa2e5d3e7, 0x9bf2e45f, 0x715bedaf, 0xafddc60f,
+	0x433ee8ea, 0x45dff1b0, 0xe3fd6d28, 0xc84f7887, 0x416df29a, 0xff209bbf,
+	0xfe2941ae, 0x37b602ae, 0xe8067a68, 0x3fa6ed80, 0x165d7bd1, 0x9e92875e,
+	0x71848370, 0xcc65c6ec, 0x943e9911, 0xe764679d, 0x8729ef95, 0x3986c38d,
+	0x2fe2506f, 0x12331ee6, 0x7df94cf5, 0xad0ec505, 0x3c5bbf95, 0x7529df71,
+	0x96afd22c, 0x714c8eff, 0x389fd82b, 0x7922df3a, 0x38aee73b, 0xbf9782a2,
+	0x468ab944, 0x0e68fab9, 0xc7db2ff6, 0x3c0683fb, 0xdf871919, 0x4f286b29,
+	0x7937d01b, 0xa3df867c, 0x05f3bb40, 0x0ea6c581, 0xab6a2f61, 0x421e8f37,
+	0x2e214a6e, 0xca8745c3, 0xb829b38f, 0x768377df, 0xf8f7c264, 0x9726567d,
+	0x662cf72f, 0x0d5ffdc1, 0x40d8f92a, 0x947e067e, 0x19e660ea, 0xc972df29,
+	0xd5ced2e7, 0x51273a26, 0xed538dfc, 0xf299b33d, 0x43972332, 0x63dfa335,
+	0xb68d54e9, 0xef380d6e, 0x4e979ce8, 0x6fc59fdf, 0xdc70faf4, 0xb58205be,
+	0x44e69a06, 0x5f2efcf1, 0x0bdb2f65, 0xb8a86fed, 0xf30e7ddf, 0x17fd1137,
+	0xff419397, 0x93cfcfc6, 0x7f3c8d0b, 0x9adca8f3, 0xa61df489, 0xdd94eb1b,
+	0x2bd52d76, 0x1d291ced, 0x677c0268, 0xefd90e71, 0x73f1df23, 0x5a65bef4,
+	0xf33bf3f2, 0xae51cbec, 0x4f1b2e77, 0xd208afed, 0xe33f7863, 0x7da48e2e,
+	0xcfc85cec, 0x4c2e771b, 0x5967f6a4, 0xdf1a955f, 0xd9fbe6ab, 0x3a04d8f9,
+	0x39d38dff, 0xbef1b372, 0xd8c7be58, 0xc24b6a99, 0x0527d37a, 0x71bbb2fb,
+	0x1a0609bc, 0xe24590ef, 0xc6cb12f7, 0xe907b63f, 0x8a9a4634, 0x7e0fdb76,
+	0x7ff00def, 0x4dd86c7f, 0xf71732f5, 0xc705b99f, 0xe4057d2f, 0x37f7c60f,
+	0xebbe8318, 0x66efbe57, 0xc5b4df6a, 0xb0bf40c6, 0xbcb5d791, 0xeaf6e464,
+	0x3373cea0, 0xe4725fbc, 0xebb45cb3, 0xdafb07fe, 0xdaee3193, 0x243af02f,
+	0xde2e5777, 0x87a89647, 0xce6faf8e, 0x4f4bf1e6, 0x8c52d44b, 0x63fa84e7,
+	0xe252df54, 0xe7435dbd, 0xd9bf7254, 0x866e3cd9, 0xed1f747e, 0xdfc827de,
+	0x05a6f8bb, 0x752b5bde, 0xb8fb3dcd, 0x6b34951c, 0x92e47be8, 0x51ff46fa,
+	0x59d43fc4, 0xf55d7c51, 0x783ee897, 0xe8074bf6, 0x3fef02de, 0x473a08de,
+	0x3b39dfe6, 0xaae63b65, 0x395ec917, 0x77f9ca9b, 0x19e2f55f, 0xd28fbdc2,
+	0xf15d34ef, 0x27731d5d, 0x5f4912ab, 0x29e6b193, 0xa5f4bdb4, 0x206e23b6,
+	0xbda1e961, 0x2ee7e8d1, 0x2345d474, 0x55dc61dd, 0x5e7c11da, 0x2b883dfc,
+	0x7d05b5c7, 0x6ef2e69f, 0x02ebc4b9, 0x845fc8bb, 0xff80bff6, 0x6f7f3f00,
+	0xd91dfcb6, 0xf25aecb5, 0x14b1f5ff, 0xa6fac157, 0x13b77e05, 0x0f4bffea,
+	0x0dde7e5c, 0xbb073ded, 0x96f5c79f, 0xd515f77d, 0xf557b123, 0xdae6b754,
+	0x73f93eff, 0xcd7fbde5, 0x7fa871e6, 0x583ddedf, 0x2fffea97, 0xf347159d,
+	0x22997db7, 0xe36839e1, 0xf10f7437, 0xfa7cb69b, 0x8737e6de, 0xd0fc2767,
+	0xb7281177, 0x75c0cc9e, 0x92bf8879, 0xac503bfc, 0xfc218655, 0x3c74de3a,
+	0x1179d705, 0xba754df8, 0x72f73f12, 0xbee9cb7a, 0x8007f101, 0x8786cebe,
+	0xb8ffa12f, 0x4d9cff1a, 0x61bd6f8b, 0x2b43d90a, 0xf58b95aa, 0xe87ffdd1,
+	0xfb7c037b, 0xc4e3e04f, 0x86bfc11f, 0x70ff342d, 0xff88bbfc, 0x081d3e00,
+	0xa6e5dfe1, 0xeb1dffbb, 0x6c535968, 0x8128f7d0, 0x08fdf374, 0x7c0915ba,
+	0xdb66dffc, 0xf4bed19a, 0x0f617c1c, 0xdf2309d6, 0x2eb0966e, 0x330a72ed,
+	0x7d30f7e2, 0x697f282b, 0xaed099d3, 0x407db495, 0xeb97e3fa, 0x80cfe029,
+	0x06cfce7e, 0x2fc83afe, 0x55d9b6c9, 0xa2e9f46b, 0xf4efe907, 0xa2942c2c,
+	0x7a19f9e9, 0x8c72e1ee, 0x2ddfc3f4, 0xdd8e7a73, 0x97c10c26, 0x9e34ecde,
+	0xd1467927, 0xff6bcbbe, 0x00c0e3c2, 0x91d3097e, 0x6e28674f, 0x0a4f8e1c,
+	0xeff24c1d, 0xadf00d88, 0x685feb96, 0x87f0177f, 0x7e653fee, 0xf7f0ff21,
+	0xceb85cd4, 0x20cf8e68, 0x0d80f85c, 0xd5fc0f61, 0xfce3ca3a, 0x83c9f721,
+	0xd79079fc, 0xa9ea855d, 0xf4e1cfbc, 0xabd3d143, 0xd2a7ffad, 0x52d92fe7,
+	0x7428bc90, 0x6876f9ea, 0xcf1c6e3b, 0x9f67fef5, 0x3504da3f, 0x7f3a0f8b,
+	0xf3c78c34, 0x4133a6c7, 0x7eca8fe7, 0xda73e47b, 0xf922fd9e, 0xeff2cdc7,
+	0x0bda84ac, 0xd0983be5, 0xd7938ef5, 0x7f7f94eb, 0xd1a0e7a1, 0x7a718063,
+	0x79216f5e, 0xd19c3676, 0x7b6a79e2, 0x9efa21fd, 0xf41fd14e, 0xf5809e13,
+	0x2c78bcf7, 0xce36ec97, 0x275f0433, 0x33cf978c, 0xc79e875b, 0x682970d5,
+	0x57b7529e, 0x57ab754b, 0xda82ef99, 0x778e3b76, 0x7eff83a1, 0xd0ef05e1,
+	0xf1a6a9e7, 0x615cf91a, 0x2fe0873b, 0xbcfe01ed, 0x1b4db053, 0xc30c2a3f,
+	0x30c30c30, 0x0c30c30c, 0xc30c30c3, 0x30c30c30, 0x0c30c30c, 0xc30c30c3,
+	0x30c30c30, 0x0c30c30c, 0xc30c30c3, 0x30c30c30, 0x0c30c30c, 0xc30c30c3,
+	0x30c30c30, 0x0c30c30c, 0xc30c30c3, 0x30c30c30, 0xc1b7ff0c, 0x8dca0bff,
+	0x8000e737, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x20000131,
+	0x22b0030c, 0xb0131302, 0x14e7ff1b, 0x93c9084d, 0x26ebaf39, 0x6db6db63,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xf6db6db6, 0x10192fc7,
+	0x8000dcb1, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x20000131,
+	0x22b0030c, 0xb0131302, 0x14e7ff1b, 0x93c9084d, 0x26ebaf39, 0x6db6db63,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xf6db6db6, 0x10192fc7,
+	0x8000dcb1, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x20000131,
+	0x22b0030c, 0xb0131302, 0x14e7ff1b, 0x93c9084d, 0x26ebaf39, 0x6db6db63,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xf6db6db6, 0x10192fc7,
+	0x8000dcb1, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x20000131,
+	0x22b0030c, 0xb0131302, 0x14e7ff1b, 0x93c9084d, 0x26ebaf39, 0x6db6db63,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xf6db6db6, 0x10192fc7,
+	0x8000dcb1, 0x00008000, 0x00088b1f, 0x00000000, 0xc5edff00, 0x20000131,
+	0x22b0030c, 0xb0131302, 0x14e7ff1b, 0x93c9084d, 0x26ebaf39, 0x6db6db63,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xb6db6db6, 0x6db6db6d,
+	0xdb6db6db, 0xb6db6db6, 0x6db6db6d, 0xdb6db6db, 0xf6db6db6, 0x10192fc7,
+	0x8000dcb1, 0x00008000, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0x00100000, 0x00000000, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+	0xffffffff, 0xffffffff, 0x00100000, 0x00000000, 0xfffffff3, 0x314fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd,
+	0xfffffff1, 0x30efffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0001cf3c, 0xcdcdcdcd, 0xfffffff6, 0x305fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0002cf3c, 0xcdcdcdcd, 0xfffff406, 0x1cbfffff,
+	0x0c30c305, 0xc30c30c3, 0xcf300014, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd,
+	0xfffffff2, 0x304fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0008cf3c, 0xcdcdcdcd, 0xfffffffa, 0x302fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0010cf3c, 0xcdcdcdcd, 0xfffffff7, 0x31efffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd,
+	0xfffffff5, 0x302fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0040cf3c, 0xcdcdcdcd, 0xfffffff3, 0x310fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd, 0xfffffff1, 0x310fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0001cf3c, 0xcdcdcdcd,
+	0xfffffff6, 0x305fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0002cf3c, 0xcdcdcdcd, 0xfffff406, 0x1cbfffff, 0x0c30c305, 0xc30c30c3,
+	0xcf300014, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd, 0xfffffff2, 0x304fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0008cf3c, 0xcdcdcdcd,
+	0xfffffffa, 0x302fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0010cf3c, 0xcdcdcdcd, 0xfffffff7, 0x30efffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd, 0xfffffff5, 0x304fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0040cf3c, 0xcdcdcdcd,
+	0xfffffff3, 0x31efffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0000cf3c, 0xcdcdcdcd, 0xfffffff1, 0x310fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0001cf3c, 0xcdcdcdcd, 0xfffffff6, 0x305fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0002cf3c, 0xcdcdcdcd,
+	0xfffff406, 0x1cbfffff, 0x0c30c305, 0xc30c30c3, 0xcf300014, 0xf3cf3cf3,
+	0x0004cf3c, 0xcdcdcdcd, 0xfffffff2, 0x304fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0008cf3c, 0xcdcdcdcd, 0xfffffffa, 0x302fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0010cf3c, 0xcdcdcdcd,
+	0xffffff97, 0x056fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cc000, 0xf3cf3cf3,
+	0x0020cf3c, 0xcdcdcdcd, 0xfffffff5, 0x310fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0040cf3c, 0xcdcdcdcd, 0xfffffff3, 0x320fffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd,
+	0xfffffff1, 0x310fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0001cf3c, 0xcdcdcdcd, 0xfffffff6, 0x305fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf300, 0xf3cf3cf3, 0x0002cf3c, 0xcdcdcdcd, 0xfffff406, 0x1cbfffff,
+	0x0c30c305, 0xc30c30c3, 0xcf300014, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd,
+	0xfffffff2, 0x304fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0008cf3c, 0xcdcdcdcd, 0xffffff8a, 0x042fffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cc000, 0xf3cf3cf3, 0x0010cf3c, 0xcdcdcdcd, 0xffffff97, 0x05cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cc000, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd,
+	0xfffffff5, 0x310fffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf300, 0xf3cf3cf3,
+	0x0040cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0001cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0002cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0008cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0010cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0040cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0000cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0001cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0002cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0004cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0008cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0010cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0020cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0040cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0001cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0002cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0008cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0010cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0040cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0000cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0001cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0002cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0004cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0008cf3c, 0xcdcdcdcd,
+	0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3,
+	0x0010cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff, 0x0c30c30c, 0xc30c30c3,
+	0xcf3cf3cc, 0xf3cf3cf3, 0x0020cf3c, 0xcdcdcdcd, 0xffffffff, 0x30cfffff,
+	0x0c30c30c, 0xc30c30c3, 0xcf3cf3cc, 0xf3cf3cf3, 0x0040cf3c, 0xcdcdcdcd,
+	0x000a0000, 0x000700a0, 0x00028110, 0x000b8138, 0x000201f0, 0x00010210,
+	0x000f0220, 0x00010310, 0x00080000, 0x00080080, 0x00028100, 0x000b8128,
+	0x000201e0, 0x00010200, 0x00070210, 0x00020280, 0x000f0000, 0x000800f0,
+	0x00028170, 0x000b8198, 0x00020250, 0x00010270, 0x000b8280, 0x00080338,
+	0x00100000, 0x00080100, 0x00028180, 0x000b81a8, 0x00020260, 0x00018280,
+	0x000e8298, 0x00080380, 0xcccccccc, 0xcccccccc, 0xcccccccc, 0xcccccccc,
+	0x00002000, 0xcccccccc, 0xcccccccc, 0xcccccccc, 0xcccccccc, 0x00002000,
+	0xcccccccc, 0xcccccccc, 0xcccccccc, 0xcccccccc, 0x00002000
+};
+
+#endif /*__BNX2X_INIT_VALUES_H__*/
diff --git a/drivers/net/bnx2x_reg.h b/drivers/net/bnx2x_reg.h
new file mode 100644
index 000000000000..86055297ab02
--- /dev/null
+++ b/drivers/net/bnx2x_reg.h
@@ -0,0 +1,4394 @@
+/* bnx2x_reg.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * The registers description starts with the regsister Access type followed
+ * by size in bits. For example [RW 32]. The access types are:
+ * R  - Read only
+ * RC - Clear on read
+ * RW - Read/Write
+ * ST - Statistics register (clear on read)
+ * W  - Write only
+ * WB - Wide bus register - the size is over 32 bits and it should be
+ *      read/write in consecutive 32 bits accesses
+ * WR - Write Clear (write 1 to clear the bit)
+ *
+ */
+
+
+/* [R 19] Interrupt register #0 read */
+#define BRB1_REG_BRB1_INT_STS					 0x6011c
+/* [RW 4] Parity mask register #0 read/write */
+#define BRB1_REG_BRB1_PRTY_MASK 				 0x60138
+/* [RW 10] At address BRB1_IND_FREE_LIST_PRS_CRDT initialize free head. At
+   address BRB1_IND_FREE_LIST_PRS_CRDT+1 initialize free tail. At address
+   BRB1_IND_FREE_LIST_PRS_CRDT+2 initialize parser initial credit. */
+#define BRB1_REG_FREE_LIST_PRS_CRDT				 0x60200
+/* [RW 23] LL RAM data. */
+#define BRB1_REG_LL_RAM 					 0x61000
+/* [R 24] The number of full blocks. */
+#define BRB1_REG_NUM_OF_FULL_BLOCKS				 0x60090
+/* [ST 32] The number of cycles that the write_full signal towards MAC #0
+   was asserted. */
+#define BRB1_REG_NUM_OF_FULL_CYCLES_0				 0x600c8
+#define BRB1_REG_NUM_OF_FULL_CYCLES_1				 0x600cc
+#define BRB1_REG_NUM_OF_FULL_CYCLES_2				 0x600d0
+#define BRB1_REG_NUM_OF_FULL_CYCLES_3				 0x600d4
+#define BRB1_REG_NUM_OF_FULL_CYCLES_4				 0x600d8
+/* [ST 32] The number of cycles that the pause signal towards MAC #0 was
+   asserted. */
+#define BRB1_REG_NUM_OF_PAUSE_CYCLES_0				 0x600b8
+#define BRB1_REG_NUM_OF_PAUSE_CYCLES_1				 0x600bc
+#define BRB1_REG_NUM_OF_PAUSE_CYCLES_2				 0x600c0
+#define BRB1_REG_NUM_OF_PAUSE_CYCLES_3				 0x600c4
+/* [RW 10] Write client 0: De-assert pause threshold. */
+#define BRB1_REG_PAUSE_HIGH_THRESHOLD_0 			 0x60078
+#define BRB1_REG_PAUSE_HIGH_THRESHOLD_1 			 0x6007c
+/* [RW 10] Write client 0: Assert pause threshold. */
+#define BRB1_REG_PAUSE_LOW_THRESHOLD_0				 0x60068
+#define BRB1_REG_PAUSE_LOW_THRESHOLD_1				 0x6006c
+/* [RW 1] Reset the design by software. */
+#define BRB1_REG_SOFT_RESET					 0x600dc
+/* [R 5] Used to read the value of the XX protection CAM occupancy counter. */
+#define CCM_REG_CAM_OCCUP					 0xd0188
+/* [RW 1] CM - CFC Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CCM_CFC_IFEN					 0xd003c
+/* [RW 1] CM - QM Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CCM_CQM_IFEN					 0xd000c
+/* [RW 1] If set the Q index; received from the QM is inserted to event ID.
+   Otherwise 0 is inserted. */
+#define CCM_REG_CCM_CQM_USE_Q					 0xd00c0
+/* [RW 11] Interrupt mask register #0 read/write */
+#define CCM_REG_CCM_INT_MASK					 0xd01e4
+/* [R 11] Interrupt register #0 read */
+#define CCM_REG_CCM_INT_STS					 0xd01d8
+/* [RW 3] The size of AG context region 0 in REG-pairs. Designates the MS
+   REG-pair number (e.g. if region 0 is 6 REG-pairs; the value should be 5).
+   Is used to determine the number of the AG context REG-pairs written back;
+   when the input message Reg1WbFlg isn't set. */
+#define CCM_REG_CCM_REG0_SZ					 0xd00c4
+/* [RW 1] CM - STORM 0 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CCM_STORM0_IFEN 				 0xd0004
+/* [RW 1] CM - STORM 1 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CCM_STORM1_IFEN 				 0xd0008
+/* [RW 1] CDU AG read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define CCM_REG_CDU_AG_RD_IFEN					 0xd0030
+/* [RW 1] CDU AG write Interface enable. If 0 - the request and valid input
+   are disregarded; all other signals are treated as usual; if 1 - normal
+   activity. */
+#define CCM_REG_CDU_AG_WR_IFEN					 0xd002c
+/* [RW 1] CDU STORM read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define CCM_REG_CDU_SM_RD_IFEN					 0xd0038
+/* [RW 1] CDU STORM write Interface enable. If 0 - the request and valid
+   input is disregarded; all other signals are treated as usual; if 1 -
+   normal activity. */
+#define CCM_REG_CDU_SM_WR_IFEN					 0xd0034
+/* [RW 4] CFC output initial credit. Max credit available - 15.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 1 at start-up. */
+#define CCM_REG_CFC_INIT_CRD					 0xd0204
+/* [RW 2] Auxillary counter flag Q number 1. */
+#define CCM_REG_CNT_AUX1_Q					 0xd00c8
+/* [RW 2] Auxillary counter flag Q number 2. */
+#define CCM_REG_CNT_AUX2_Q					 0xd00cc
+/* [RW 28] The CM header value for QM request (primary). */
+#define CCM_REG_CQM_CCM_HDR_P					 0xd008c
+/* [RW 28] The CM header value for QM request (secondary). */
+#define CCM_REG_CQM_CCM_HDR_S					 0xd0090
+/* [RW 1] QM - CM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CQM_CCM_IFEN					 0xd0014
+/* [RW 6] QM output initial credit. Max credit available - 32. Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 32 at start-up. */
+#define CCM_REG_CQM_INIT_CRD					 0xd020c
+/* [RW 3] The weight of the QM (primary) input in the WRR mechanism. 0
+   stands for weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define CCM_REG_CQM_P_WEIGHT					 0xd00b8
+/* [RW 1] Input SDM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_CSDM_IFEN					 0xd0018
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the SDM interface is detected. */
+#define CCM_REG_CSDM_LENGTH_MIS 				 0xd0170
+/* [RW 28] The CM header for QM formatting in case of an error in the QM
+   inputs. */
+#define CCM_REG_ERR_CCM_HDR					 0xd0094
+/* [RW 8] The Event ID in case the input message ErrorFlg is set. */
+#define CCM_REG_ERR_EVNT_ID					 0xd0098
+/* [RW 8] FIC0 output initial credit. Max credit available - 255. Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define CCM_REG_FIC0_INIT_CRD					 0xd0210
+/* [RW 8] FIC1 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define CCM_REG_FIC1_INIT_CRD					 0xd0214
+/* [RW 1] Arbitration between Input Arbiter groups: 0 - fair Round-Robin; 1
+   - strict priority defined by ~ccm_registers_gr_ag_pr.gr_ag_pr;
+   ~ccm_registers_gr_ld0_pr.gr_ld0_pr and
+   ~ccm_registers_gr_ld1_pr.gr_ld1_pr. Groups are according to channels and
+   outputs to STORM: aggregation; load FIC0; load FIC1 and store. */
+#define CCM_REG_GR_ARB_TYPE					 0xd015c
+/* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed; that the Store channel priority is
+   the compliment to 4 of the rest priorities - Aggregation channel; Load
+   (FIC0) channel and Load (FIC1). */
+#define CCM_REG_GR_LD0_PR					 0xd0164
+/* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed; that the Store channel priority is
+   the compliment to 4 of the rest priorities - Aggregation channel; Load
+   (FIC0) channel and Load (FIC1). */
+#define CCM_REG_GR_LD1_PR					 0xd0168
+/* [RW 2] General flags index. */
+#define CCM_REG_INV_DONE_Q					 0xd0108
+/* [RW 4] The number of double REG-pairs(128 bits); loaded from the STORM
+   context and sent to STORM; for a specific connection type. The double
+   REG-pairs are used in order to align to STORM context row size of 128
+   bits. The offset of these data in the STORM context is always 0. Index
+   _(0..15) stands for the connection type (one of 16). */
+#define CCM_REG_N_SM_CTX_LD_0					 0xd004c
+#define CCM_REG_N_SM_CTX_LD_1					 0xd0050
+#define CCM_REG_N_SM_CTX_LD_10					 0xd0074
+#define CCM_REG_N_SM_CTX_LD_11					 0xd0078
+#define CCM_REG_N_SM_CTX_LD_12					 0xd007c
+#define CCM_REG_N_SM_CTX_LD_13					 0xd0080
+#define CCM_REG_N_SM_CTX_LD_14					 0xd0084
+#define CCM_REG_N_SM_CTX_LD_15					 0xd0088
+#define CCM_REG_N_SM_CTX_LD_2					 0xd0054
+#define CCM_REG_N_SM_CTX_LD_3					 0xd0058
+#define CCM_REG_N_SM_CTX_LD_4					 0xd005c
+/* [RW 1] Input pbf Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define CCM_REG_PBF_IFEN					 0xd0028
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the pbf interface is detected. */
+#define CCM_REG_PBF_LENGTH_MIS					 0xd0180
+/* [RW 3] The weight of the input pbf in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define CCM_REG_PBF_WEIGHT					 0xd00ac
+/* [RW 6] The physical queue number of queue number 1 per port index. */
+#define CCM_REG_PHYS_QNUM1_0					 0xd0134
+#define CCM_REG_PHYS_QNUM1_1					 0xd0138
+/* [RW 6] The physical queue number of queue number 2 per port index. */
+#define CCM_REG_PHYS_QNUM2_0					 0xd013c
+#define CCM_REG_PHYS_QNUM2_1					 0xd0140
+/* [RW 6] The physical queue number of queue number 3 per port index. */
+#define CCM_REG_PHYS_QNUM3_0					 0xd0144
+/* [RW 6] The physical queue number of queue number 0 with QOS equal 0 port
+   index 0. */
+#define CCM_REG_QOS_PHYS_QNUM0_0				 0xd0114
+#define CCM_REG_QOS_PHYS_QNUM0_1				 0xd0118
+/* [RW 6] The physical queue number of queue number 0 with QOS equal 1 port
+   index 0. */
+#define CCM_REG_QOS_PHYS_QNUM1_0				 0xd011c
+#define CCM_REG_QOS_PHYS_QNUM1_1				 0xd0120
+/* [RW 6] The physical queue number of queue number 0 with QOS equal 2 port
+   index 0. */
+#define CCM_REG_QOS_PHYS_QNUM2_0				 0xd0124
+/* [RW 1] STORM - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define CCM_REG_STORM_CCM_IFEN					 0xd0010
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the STORM interface is detected. */
+#define CCM_REG_STORM_LENGTH_MIS				 0xd016c
+/* [RW 1] Input tsem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define CCM_REG_TSEM_IFEN					 0xd001c
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the tsem interface is detected. */
+#define CCM_REG_TSEM_LENGTH_MIS 				 0xd0174
+/* [RW 3] The weight of the input tsem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define CCM_REG_TSEM_WEIGHT					 0xd00a0
+/* [RW 1] Input usem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define CCM_REG_USEM_IFEN					 0xd0024
+/* [RC 1] Set when message length mismatch (relative to last indication) at
+   the usem interface is detected. */
+#define CCM_REG_USEM_LENGTH_MIS 				 0xd017c
+/* [RW 3] The weight of the input usem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define CCM_REG_USEM_WEIGHT					 0xd00a8
+/* [RW 1] Input xsem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define CCM_REG_XSEM_IFEN					 0xd0020
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the xsem interface is detected. */
+#define CCM_REG_XSEM_LENGTH_MIS 				 0xd0178
+/* [RW 3] The weight of the input xsem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define CCM_REG_XSEM_WEIGHT					 0xd00a4
+/* [RW 19] Indirect access to the descriptor table of the XX protection
+   mechanism. The fields are: [5:0] - message length; [12:6] - message
+   pointer; 18:13] - next pointer. */
+#define CCM_REG_XX_DESCR_TABLE					 0xd0300
+/* [R 7] Used to read the value of XX protection Free counter. */
+#define CCM_REG_XX_FREE 					 0xd0184
+/* [RW 6] Initial value for the credit counter; responsible for fulfilling
+   of the Input Stage XX protection buffer by the XX protection pending
+   messages. Max credit available - 127. Write writes the initial credit
+   value; read returns the current value of the credit counter. Must be
+   initialized to maximum XX protected message size - 2 at start-up. */
+#define CCM_REG_XX_INIT_CRD					 0xd0220
+/* [RW 7] The maximum number of pending messages; which may be stored in XX
+   protection. At read the ~ccm_registers_xx_free.xx_free counter is read.
+   At write comprises the start value of the ~ccm_registers_xx_free.xx_free
+   counter. */
+#define CCM_REG_XX_MSG_NUM					 0xd0224
+/* [RW 8] The Event ID; sent to the STORM in case of XX overflow. */
+#define CCM_REG_XX_OVFL_EVNT_ID 				 0xd0044
+/* [RW 18] Indirect access to the XX table of the XX protection mechanism.
+   The fields are: [5:0] - tail pointer; 11:6] - Link List size; 17:12] -
+   header pointer. */
+#define CCM_REG_XX_TABLE					 0xd0280
+#define CDU_REG_CDU_CHK_MASK0					 0x101000
+#define CDU_REG_CDU_CHK_MASK1					 0x101004
+#define CDU_REG_CDU_CONTROL0					 0x101008
+#define CDU_REG_CDU_DEBUG					 0x101010
+#define CDU_REG_CDU_GLOBAL_PARAMS				 0x101020
+/* [RW 7] Interrupt mask register #0 read/write */
+#define CDU_REG_CDU_INT_MASK					 0x10103c
+/* [R 7] Interrupt register #0 read */
+#define CDU_REG_CDU_INT_STS					 0x101030
+/* [RW 5] Parity mask register #0 read/write */
+#define CDU_REG_CDU_PRTY_MASK					 0x10104c
+/* [RC 32] logging of error data in case of a CDU load error:
+   {expected_cid[15:0]; xpected_type[2:0]; xpected_region[2:0]; ctive_error;
+   ype_error; ctual_active; ctual_compressed_context}; */
+#define CDU_REG_ERROR_DATA					 0x101014
+/* [WB 216] L1TT ram access. each entry has the following format :
+   {mrege_regions[7:0]; ffset12[5:0]...offset0[5:0];
+   ength12[5:0]...length0[5:0]; d12[3:0]...id0[3:0]} */
+#define CDU_REG_L1TT						 0x101800
+/* [WB 24] MATT ram access. each entry has the following
+   format:{RegionLength[11:0]; egionOffset[11:0]} */
+#define CDU_REG_MATT						 0x101100
+/* [R 1] indication the initializing the activity counter by the hardware
+   was done. */
+#define CFC_REG_AC_INIT_DONE					 0x104078
+/* [RW 13] activity counter ram access */
+#define CFC_REG_ACTIVITY_COUNTER				 0x104400
+#define CFC_REG_ACTIVITY_COUNTER_SIZE				 256
+/* [R 1] indication the initializing the cams by the hardware was done. */
+#define CFC_REG_CAM_INIT_DONE					 0x10407c
+/* [RW 2] Interrupt mask register #0 read/write */
+#define CFC_REG_CFC_INT_MASK					 0x104108
+/* [R 2] Interrupt register #0 read */
+#define CFC_REG_CFC_INT_STS					 0x1040fc
+/* [RC 2] Interrupt register #0 read clear */
+#define CFC_REG_CFC_INT_STS_CLR 				 0x104100
+/* [RW 4] Parity mask register #0 read/write */
+#define CFC_REG_CFC_PRTY_MASK					 0x104118
+/* [RW 21] CID cam access (21:1 - Data; alid - 0) */
+#define CFC_REG_CID_CAM 					 0x104800
+#define CFC_REG_CONTROL0					 0x104028
+#define CFC_REG_DEBUG0						 0x104050
+/* [RW 14] indicates per error (in #cfc_registers_cfc_error_vector.cfc_error
+   vector) whether the cfc should be disabled upon it */
+#define CFC_REG_DISABLE_ON_ERROR				 0x104044
+/* [RC 14] CFC error vector. when the CFC detects an internal error it will
+   set one of these bits. the bit description can be found in CFC
+   specifications */
+#define CFC_REG_ERROR_VECTOR					 0x10403c
+#define CFC_REG_INIT_REG					 0x10404c
+/* [RW 24] {weight_load_client7[2:0] to weight_load_client0[2:0]}. this
+   field allows changing the priorities of the weighted-round-robin arbiter
+   which selects which CFC load client should be served next */
+#define CFC_REG_LCREQ_WEIGHTS					 0x104084
+/* [R 1] indication the initializing the link list by the hardware was done. */
+#define CFC_REG_LL_INIT_DONE					 0x104074
+/* [R 9] Number of allocated LCIDs which are at empty state */
+#define CFC_REG_NUM_LCIDS_ALLOC 				 0x104020
+/* [R 9] Number of Arriving LCIDs in Link List Block */
+#define CFC_REG_NUM_LCIDS_ARRIVING				 0x104004
+/* [R 9] Number of Inside LCIDs in Link List Block */
+#define CFC_REG_NUM_LCIDS_INSIDE				 0x104008
+/* [R 9] Number of Leaving LCIDs in Link List Block */
+#define CFC_REG_NUM_LCIDS_LEAVING				 0x104018
+/* [RW 8] The event id for aggregated interrupt 0 */
+#define CSDM_REG_AGG_INT_EVENT_0				 0xc2038
+/* [RW 13] The start address in the internal RAM for the cfc_rsp lcid */
+#define CSDM_REG_CFC_RSP_START_ADDR				 0xc2008
+/* [RW 16] The maximum value of the competion counter #0 */
+#define CSDM_REG_CMP_COUNTER_MAX0				 0xc201c
+/* [RW 16] The maximum value of the competion counter #1 */
+#define CSDM_REG_CMP_COUNTER_MAX1				 0xc2020
+/* [RW 16] The maximum value of the competion counter #2 */
+#define CSDM_REG_CMP_COUNTER_MAX2				 0xc2024
+/* [RW 16] The maximum value of the competion counter #3 */
+#define CSDM_REG_CMP_COUNTER_MAX3				 0xc2028
+/* [RW 13] The start address in the internal RAM for the completion
+   counters. */
+#define CSDM_REG_CMP_COUNTER_START_ADDR 			 0xc200c
+/* [RW 32] Interrupt mask register #0 read/write */
+#define CSDM_REG_CSDM_INT_MASK_0				 0xc229c
+#define CSDM_REG_CSDM_INT_MASK_1				 0xc22ac
+/* [RW 11] Parity mask register #0 read/write */
+#define CSDM_REG_CSDM_PRTY_MASK 				 0xc22bc
+#define CSDM_REG_ENABLE_IN1					 0xc2238
+#define CSDM_REG_ENABLE_IN2					 0xc223c
+#define CSDM_REG_ENABLE_OUT1					 0xc2240
+#define CSDM_REG_ENABLE_OUT2					 0xc2244
+/* [RW 4] The initial number of messages that can be sent to the pxp control
+   interface without receiving any ACK. */
+#define CSDM_REG_INIT_CREDIT_PXP_CTRL				 0xc24bc
+/* [ST 32] The number of ACK after placement messages received */
+#define CSDM_REG_NUM_OF_ACK_AFTER_PLACE 			 0xc227c
+/* [ST 32] The number of packet end messages received from the parser */
+#define CSDM_REG_NUM_OF_PKT_END_MSG				 0xc2274
+/* [ST 32] The number of requests received from the pxp async if */
+#define CSDM_REG_NUM_OF_PXP_ASYNC_REQ				 0xc2278
+/* [ST 32] The number of commands received in queue 0 */
+#define CSDM_REG_NUM_OF_Q0_CMD					 0xc2248
+/* [ST 32] The number of commands received in queue 10 */
+#define CSDM_REG_NUM_OF_Q10_CMD 				 0xc226c
+/* [ST 32] The number of commands received in queue 11 */
+#define CSDM_REG_NUM_OF_Q11_CMD 				 0xc2270
+/* [ST 32] The number of commands received in queue 1 */
+#define CSDM_REG_NUM_OF_Q1_CMD					 0xc224c
+/* [ST 32] The number of commands received in queue 3 */
+#define CSDM_REG_NUM_OF_Q3_CMD					 0xc2250
+/* [ST 32] The number of commands received in queue 4 */
+#define CSDM_REG_NUM_OF_Q4_CMD					 0xc2254
+/* [ST 32] The number of commands received in queue 5 */
+#define CSDM_REG_NUM_OF_Q5_CMD					 0xc2258
+/* [ST 32] The number of commands received in queue 6 */
+#define CSDM_REG_NUM_OF_Q6_CMD					 0xc225c
+/* [ST 32] The number of commands received in queue 7 */
+#define CSDM_REG_NUM_OF_Q7_CMD					 0xc2260
+/* [ST 32] The number of commands received in queue 8 */
+#define CSDM_REG_NUM_OF_Q8_CMD					 0xc2264
+/* [ST 32] The number of commands received in queue 9 */
+#define CSDM_REG_NUM_OF_Q9_CMD					 0xc2268
+/* [RW 13] The start address in the internal RAM for queue counters */
+#define CSDM_REG_Q_COUNTER_START_ADDR				 0xc2010
+/* [R 1] pxp_ctrl rd_data fifo empty in sdm_dma_rsp block */
+#define CSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY			 0xc2548
+/* [R 1] parser fifo empty in sdm_sync block */
+#define CSDM_REG_SYNC_PARSER_EMPTY				 0xc2550
+/* [R 1] parser serial fifo empty in sdm_sync block */
+#define CSDM_REG_SYNC_SYNC_EMPTY				 0xc2558
+/* [RW 32] Tick for timer counter. Applicable only when
+   ~csdm_registers_timer_tick_enable.timer_tick_enable =1 */
+#define CSDM_REG_TIMER_TICK					 0xc2000
+/* [RW 5] The number of time_slots in the arbitration cycle */
+#define CSEM_REG_ARB_CYCLE_SIZE 				 0x200034
+/* [RW 3] The source that is associated with arbitration element 0. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2 */
+#define CSEM_REG_ARB_ELEMENT0					 0x200020
+/* [RW 3] The source that is associated with arbitration element 1. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~csem_registers_arb_element0.arb_element0 */
+#define CSEM_REG_ARB_ELEMENT1					 0x200024
+/* [RW 3] The source that is associated with arbitration element 2. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~csem_registers_arb_element0.arb_element0
+   and ~csem_registers_arb_element1.arb_element1 */
+#define CSEM_REG_ARB_ELEMENT2					 0x200028
+/* [RW 3] The source that is associated with arbitration element 3. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.Could
+   not be equal to register ~csem_registers_arb_element0.arb_element0 and
+   ~csem_registers_arb_element1.arb_element1 and
+   ~csem_registers_arb_element2.arb_element2 */
+#define CSEM_REG_ARB_ELEMENT3					 0x20002c
+/* [RW 3] The source that is associated with arbitration element 4. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~csem_registers_arb_element0.arb_element0
+   and ~csem_registers_arb_element1.arb_element1 and
+   ~csem_registers_arb_element2.arb_element2 and
+   ~csem_registers_arb_element3.arb_element3 */
+#define CSEM_REG_ARB_ELEMENT4					 0x200030
+/* [RW 32] Interrupt mask register #0 read/write */
+#define CSEM_REG_CSEM_INT_MASK_0				 0x200110
+#define CSEM_REG_CSEM_INT_MASK_1				 0x200120
+/* [RW 32] Parity mask register #0 read/write */
+#define CSEM_REG_CSEM_PRTY_MASK_0				 0x200130
+#define CSEM_REG_CSEM_PRTY_MASK_1				 0x200140
+#define CSEM_REG_ENABLE_IN					 0x2000a4
+#define CSEM_REG_ENABLE_OUT					 0x2000a8
+/* [RW 32] This address space contains all registers and memories that are
+   placed in SEM_FAST block. The SEM_FAST registers are described in
+   appendix B. In order to access the SEM_FAST registers the base address
+   CSEM_REGISTERS_FAST_MEMORY (Offset: 0x220000) should be added to each
+   SEM_FAST register offset. */
+#define CSEM_REG_FAST_MEMORY					 0x220000
+/* [RW 1] Disables input messages from FIC0 May be updated during run_time
+   by the microcode */
+#define CSEM_REG_FIC0_DISABLE					 0x200224
+/* [RW 1] Disables input messages from FIC1 May be updated during run_time
+   by the microcode */
+#define CSEM_REG_FIC1_DISABLE					 0x200234
+/* [RW 15] Interrupt table Read and write access to it is not possible in
+   the middle of the work */
+#define CSEM_REG_INT_TABLE					 0x200400
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC0 */
+#define CSEM_REG_MSG_NUM_FIC0					 0x200000
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC1 */
+#define CSEM_REG_MSG_NUM_FIC1					 0x200004
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC0 */
+#define CSEM_REG_MSG_NUM_FOC0					 0x200008
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC1 */
+#define CSEM_REG_MSG_NUM_FOC1					 0x20000c
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC2 */
+#define CSEM_REG_MSG_NUM_FOC2					 0x200010
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC3 */
+#define CSEM_REG_MSG_NUM_FOC3					 0x200014
+/* [RW 1] Disables input messages from the passive buffer May be updated
+   during run_time by the microcode */
+#define CSEM_REG_PAS_DISABLE					 0x20024c
+/* [WB 128] Debug only. Passive buffer memory */
+#define CSEM_REG_PASSIVE_BUFFER 				 0x202000
+/* [WB 46] pram memory. B45 is parity; b[44:0] - data. */
+#define CSEM_REG_PRAM						 0x240000
+/* [R 16] Valid sleeping threads indication have bit per thread */
+#define CSEM_REG_SLEEP_THREADS_VALID				 0x20026c
+/* [R 1] EXT_STORE FIFO is empty in sem_slow_ls_ext */
+#define CSEM_REG_SLOW_EXT_STORE_EMPTY				 0x2002a0
+/* [RW 16] List of free threads . There is a bit per thread. */
+#define CSEM_REG_THREADS_LIST					 0x2002e4
+/* [RW 3] The arbitration scheme of time_slot 0 */
+#define CSEM_REG_TS_0_AS					 0x200038
+/* [RW 3] The arbitration scheme of time_slot 10 */
+#define CSEM_REG_TS_10_AS					 0x200060
+/* [RW 3] The arbitration scheme of time_slot 11 */
+#define CSEM_REG_TS_11_AS					 0x200064
+/* [RW 3] The arbitration scheme of time_slot 12 */
+#define CSEM_REG_TS_12_AS					 0x200068
+/* [RW 3] The arbitration scheme of time_slot 13 */
+#define CSEM_REG_TS_13_AS					 0x20006c
+/* [RW 3] The arbitration scheme of time_slot 14 */
+#define CSEM_REG_TS_14_AS					 0x200070
+/* [RW 3] The arbitration scheme of time_slot 15 */
+#define CSEM_REG_TS_15_AS					 0x200074
+/* [RW 3] The arbitration scheme of time_slot 16 */
+#define CSEM_REG_TS_16_AS					 0x200078
+/* [RW 3] The arbitration scheme of time_slot 17 */
+#define CSEM_REG_TS_17_AS					 0x20007c
+/* [RW 3] The arbitration scheme of time_slot 18 */
+#define CSEM_REG_TS_18_AS					 0x200080
+/* [RW 3] The arbitration scheme of time_slot 1 */
+#define CSEM_REG_TS_1_AS					 0x20003c
+/* [RW 3] The arbitration scheme of time_slot 2 */
+#define CSEM_REG_TS_2_AS					 0x200040
+/* [RW 3] The arbitration scheme of time_slot 3 */
+#define CSEM_REG_TS_3_AS					 0x200044
+/* [RW 3] The arbitration scheme of time_slot 4 */
+#define CSEM_REG_TS_4_AS					 0x200048
+/* [RW 3] The arbitration scheme of time_slot 5 */
+#define CSEM_REG_TS_5_AS					 0x20004c
+/* [RW 3] The arbitration scheme of time_slot 6 */
+#define CSEM_REG_TS_6_AS					 0x200050
+/* [RW 3] The arbitration scheme of time_slot 7 */
+#define CSEM_REG_TS_7_AS					 0x200054
+/* [RW 3] The arbitration scheme of time_slot 8 */
+#define CSEM_REG_TS_8_AS					 0x200058
+/* [RW 3] The arbitration scheme of time_slot 9 */
+#define CSEM_REG_TS_9_AS					 0x20005c
+/* [RW 1] Parity mask register #0 read/write */
+#define DBG_REG_DBG_PRTY_MASK					 0xc0a8
+/* [RW 2] debug only: These bits indicate the credit for PCI request type 4
+   interface; MUST be configured AFTER pci_ext_buffer_strt_addr_lsb/msb are
+   configured */
+#define DBG_REG_PCI_REQ_CREDIT					 0xc120
+/* [RW 32] Commands memory. The address to command X; row Y is to calculated
+   as 14*X+Y. */
+#define DMAE_REG_CMD_MEM					 0x102400
+/* [RW 1] If 0 - the CRC-16c initial value is all zeroes; if 1 - the CRC-16c
+   initial value is all ones. */
+#define DMAE_REG_CRC16C_INIT					 0x10201c
+/* [RW 1] If 0 - the CRC-16 T10 initial value is all zeroes; if 1 - the
+   CRC-16 T10 initial value is all ones. */
+#define DMAE_REG_CRC16T10_INIT					 0x102020
+/* [RW 2] Interrupt mask register #0 read/write */
+#define DMAE_REG_DMAE_INT_MASK					 0x102054
+/* [RW 4] Parity mask register #0 read/write */
+#define DMAE_REG_DMAE_PRTY_MASK 				 0x102064
+/* [RW 1] Command 0 go. */
+#define DMAE_REG_GO_C0						 0x102080
+/* [RW 1] Command 1 go. */
+#define DMAE_REG_GO_C1						 0x102084
+/* [RW 1] Command 10 go. */
+#define DMAE_REG_GO_C10 					 0x102088
+#define DMAE_REG_GO_C10_SIZE					 1
+/* [RW 1] Command 11 go. */
+#define DMAE_REG_GO_C11 					 0x10208c
+#define DMAE_REG_GO_C11_SIZE					 1
+/* [RW 1] Command 12 go. */
+#define DMAE_REG_GO_C12 					 0x102090
+#define DMAE_REG_GO_C12_SIZE					 1
+/* [RW 1] Command 13 go. */
+#define DMAE_REG_GO_C13 					 0x102094
+#define DMAE_REG_GO_C13_SIZE					 1
+/* [RW 1] Command 14 go. */
+#define DMAE_REG_GO_C14 					 0x102098
+#define DMAE_REG_GO_C14_SIZE					 1
+/* [RW 1] Command 15 go. */
+#define DMAE_REG_GO_C15 					 0x10209c
+#define DMAE_REG_GO_C15_SIZE					 1
+/* [RW 1] Command 10 go. */
+#define DMAE_REG_GO_C10 					 0x102088
+/* [RW 1] Command 11 go. */
+#define DMAE_REG_GO_C11 					 0x10208c
+/* [RW 1] Command 12 go. */
+#define DMAE_REG_GO_C12 					 0x102090
+/* [RW 1] Command 13 go. */
+#define DMAE_REG_GO_C13 					 0x102094
+/* [RW 1] Command 14 go. */
+#define DMAE_REG_GO_C14 					 0x102098
+/* [RW 1] Command 15 go. */
+#define DMAE_REG_GO_C15 					 0x10209c
+/* [RW 1] Command 2 go. */
+#define DMAE_REG_GO_C2						 0x1020a0
+/* [RW 1] Command 3 go. */
+#define DMAE_REG_GO_C3						 0x1020a4
+/* [RW 1] Command 4 go. */
+#define DMAE_REG_GO_C4						 0x1020a8
+/* [RW 1] Command 5 go. */
+#define DMAE_REG_GO_C5						 0x1020ac
+/* [RW 1] Command 6 go. */
+#define DMAE_REG_GO_C6						 0x1020b0
+/* [RW 1] Command 7 go. */
+#define DMAE_REG_GO_C7						 0x1020b4
+/* [RW 1] Command 8 go. */
+#define DMAE_REG_GO_C8						 0x1020b8
+/* [RW 1] Command 9 go. */
+#define DMAE_REG_GO_C9						 0x1020bc
+/* [RW 1] DMAE GRC Interface (Target; aster) enable. If 0 - the acknowledge
+   input is disregarded; valid is deasserted; all other signals are treated
+   as usual; if 1 - normal activity. */
+#define DMAE_REG_GRC_IFEN					 0x102008
+/* [RW 1] DMAE PCI Interface (Request; ead; rite) enable. If 0 - the
+   acknowledge input is disregarded; valid is deasserted; full is asserted;
+   all other signals are treated as usual; if 1 - normal activity. */
+#define DMAE_REG_PCI_IFEN					 0x102004
+/* [RW 4] DMAE- PCI Request Interface initial credit. Write writes the
+   initial value to the credit counter; related to the address. Read returns
+   the current value of the counter. */
+#define DMAE_REG_PXP_REQ_INIT_CRD				 0x1020c0
+/* [RW 8] Aggregation command. */
+#define DORQ_REG_AGG_CMD0					 0x170060
+/* [RW 8] Aggregation command. */
+#define DORQ_REG_AGG_CMD1					 0x170064
+/* [RW 8] Aggregation command. */
+#define DORQ_REG_AGG_CMD2					 0x170068
+/* [RW 8] Aggregation command. */
+#define DORQ_REG_AGG_CMD3					 0x17006c
+/* [RW 28] UCM Header. */
+#define DORQ_REG_CMHEAD_RX					 0x170050
+/* [RW 5] Interrupt mask register #0 read/write */
+#define DORQ_REG_DORQ_INT_MASK					 0x170180
+/* [R 5] Interrupt register #0 read */
+#define DORQ_REG_DORQ_INT_STS					 0x170174
+/* [RC 5] Interrupt register #0 read clear */
+#define DORQ_REG_DORQ_INT_STS_CLR				 0x170178
+/* [RW 2] Parity mask register #0 read/write */
+#define DORQ_REG_DORQ_PRTY_MASK 				 0x170190
+/* [RW 8] The address to write the DPM CID to STORM. */
+#define DORQ_REG_DPM_CID_ADDR					 0x170044
+/* [RW 5] The DPM mode CID extraction offset. */
+#define DORQ_REG_DPM_CID_OFST					 0x170030
+/* [RW 12] The threshold of the DQ FIFO to send the almost full interrupt. */
+#define DORQ_REG_DQ_FIFO_AFULL_TH				 0x17007c
+/* [RW 12] The threshold of the DQ FIFO to send the full interrupt. */
+#define DORQ_REG_DQ_FIFO_FULL_TH				 0x170078
+/* [R 13] Current value of the DQ FIFO fill level according to following
+   pointer. The range is 0 - 256 FIFO rows; where each row stands for the
+   doorbell. */
+#define DORQ_REG_DQ_FILL_LVLF					 0x1700a4
+/* [R 1] DQ FIFO full status. Is set; when FIFO filling level is more or
+   equal to full threshold; reset on full clear. */
+#define DORQ_REG_DQ_FULL_ST					 0x1700c0
+/* [RW 28] The value sent to CM header in the case of CFC load error. */
+#define DORQ_REG_ERR_CMHEAD					 0x170058
+#define DORQ_REG_IF_EN						 0x170004
+#define DORQ_REG_MODE_ACT					 0x170008
+/* [RW 5] The normal mode CID extraction offset. */
+#define DORQ_REG_NORM_CID_OFST					 0x17002c
+/* [RW 28] TCM Header when only TCP context is loaded. */
+#define DORQ_REG_NORM_CMHEAD_TX 				 0x17004c
+/* [RW 3] The number of simultaneous outstanding requests to Context Fetch
+   Interface. */
+#define DORQ_REG_OUTST_REQ					 0x17003c
+#define DORQ_REG_REGN						 0x170038
+/* [R 4] Current value of response A counter credit. Initial credit is
+   configured through write to ~dorq_registers_rsp_init_crd.rsp_init_crd
+   register. */
+#define DORQ_REG_RSPA_CRD_CNT					 0x1700ac
+/* [R 4] Current value of response B counter credit. Initial credit is
+   configured through write to ~dorq_registers_rsp_init_crd.rsp_init_crd
+   register. */
+#define DORQ_REG_RSPB_CRD_CNT					 0x1700b0
+/* [RW 4] The initial credit at the Doorbell Response Interface. The write
+   writes the same initial credit to the rspa_crd_cnt and rspb_crd_cnt. The
+   read reads this written value. */
+#define DORQ_REG_RSP_INIT_CRD					 0x170048
+/* [RW 4] Initial activity counter value on the load request; when the
+   shortcut is done. */
+#define DORQ_REG_SHRT_ACT_CNT					 0x170070
+/* [RW 28] TCM Header when both ULP and TCP context is loaded. */
+#define DORQ_REG_SHRT_CMHEAD					 0x170054
+#define HC_CONFIG_0_REG_ATTN_BIT_EN_0				 (0x1<<4)
+#define HC_CONFIG_0_REG_INT_LINE_EN_0				 (0x1<<3)
+#define HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0			 (0x1<<2)
+#define HC_CONFIG_0_REG_SINGLE_ISR_EN_0 			 (0x1<<1)
+#define HC_REG_AGG_INT_0					 0x108050
+#define HC_REG_AGG_INT_1					 0x108054
+/* [RW 16] attention bit and attention acknowledge bits status for port 0
+   and 1 according to the following address map: addr 0 - attn_bit_0; addr 1
+   - attn_ack_bit_0; addr 2 - attn_bit_1; addr 3 - attn_ack_bit_1; */
+#define HC_REG_ATTN_BIT 					 0x108120
+/* [RW 16] attn bits status index for attn bit msg; addr 0 - function 0;
+   addr 1 - functin 1 */
+#define HC_REG_ATTN_IDX 					 0x108100
+/* [RW 32] port 0 lower 32 bits address field for attn messag. */
+#define HC_REG_ATTN_MSG0_ADDR_L 				 0x108018
+/* [RW 32] port 1 lower 32 bits address field for attn messag. */
+#define HC_REG_ATTN_MSG1_ADDR_L 				 0x108020
+/* [RW 8] status block number for attn bit msg - function 0; */
+#define HC_REG_ATTN_NUM_P0					 0x108038
+/* [RW 8] status block number for attn bit msg - function 1 */
+#define HC_REG_ATTN_NUM_P1					 0x10803c
+#define HC_REG_CONFIG_0 					 0x108000
+#define HC_REG_CONFIG_1 					 0x108004
+/* [RW 3] Parity mask register #0 read/write */
+#define HC_REG_HC_PRTY_MASK					 0x1080a0
+/* [RW 17] status block interrupt mask; one in each bit means unmask; zerow
+   in each bit means mask; bit 0 - default SB; bit 1 - SB_0; bit 2 - SB_1...
+   bit 16- SB_15; addr 0 - port 0; addr 1 - port 1 */
+#define HC_REG_INT_MASK 					 0x108108
+/* [RW 16] port 0 attn bit condition monitoring; each bit that is set will
+   lock a change fron 0 to 1 in the corresponding attention signals that
+   comes from the AEU */
+#define HC_REG_LEADING_EDGE_0					 0x108040
+#define HC_REG_LEADING_EDGE_1					 0x108048
+/* [RW 16] all producer and consumer of port 0 according to the following
+   addresses; U_prod: 0-15; C_prod: 16-31; U_cons: 32-47; C_cons:48-63;
+   Defoult_prod: U/C/X/T/Attn-64/65/66/67/68; Defoult_cons:
+   U/C/X/T/Attn-69/70/71/72/73 */
+#define HC_REG_P0_PROD_CONS					 0x108200
+/* [RW 16] all producer and consumer of port 1according to the following
+   addresses; U_prod: 0-15; C_prod: 16-31; U_cons: 32-47; C_cons:48-63;
+   Defoult_prod: U/C/X/T/Attn-64/65/66/67/68; Defoult_cons:
+   U/C/X/T/Attn-69/70/71/72/73 */
+#define HC_REG_P1_PROD_CONS					 0x108400
+/* [W 1] This register is write only and has 4 addresses as follow: 0 =
+   clear all PBA bits port 0; 1 = clear all pending interrupts request
+   port0; 2 = clear all PBA bits port 1; 3 = clear all pending interrupts
+   request port1; here is no meaning for the data in this register */
+#define HC_REG_PBA_COMMAND					 0x108140
+#define HC_REG_PCI_CONFIG_0					 0x108010
+#define HC_REG_PCI_CONFIG_1					 0x108014
+/* [RW 24] all counters acording to the following address: LSB: 0=read; 1=
+   read_clear; 0-71 = HW counters (the inside order is the same as the
+   interrupt table in the spec); 72-219 = SW counters 1 (stops after first
+   consumer upd) the inside order is: 72-103 - U_non_default_p0; 104-135
+   C_non_defaul_p0; 36-145 U/C/X/T/Attn_default_p0; 146-177
+   U_non_default_p1; 178-209 C_non_defaul_p1; 10-219 U/C/X/T/Attn_default_p1
+   ; 220-367 = SW counters 2 (stops when prod=cons) the inside order is:
+   220-251 - U_non_default_p0; 252-283 C_non_defaul_p0; 84-293
+   U/C/X/T/Attn_default_p0; 294-325 U_non_default_p1; 326-357
+   C_non_defaul_p1; 58-367 U/C/X/T/Attn_default_p1 ; 368-515 = mailbox
+   counters; (the inside order of the mailbox counter is 368-431 U and C
+   non_default_p0; 432-441 U/C/X/T/Attn_default_p0; 442-505 U and C
+   non_default_p1; 506-515 U/C/X/T/Attn_default_p1) */
+#define HC_REG_STATISTIC_COUNTERS				 0x109000
+/* [RW 16] port 0 attn bit condition monitoring; each bit that is set will
+   lock a change fron 1 to 0 in the corresponding attention signals that
+   comes from the AEU */
+#define HC_REG_TRAILING_EDGE_0					 0x108044
+#define HC_REG_TRAILING_EDGE_1					 0x10804c
+#define HC_REG_UC_RAM_ADDR_0					 0x108028
+#define HC_REG_UC_RAM_ADDR_1					 0x108030
+/* [RW 16] ustorm address for coalesc now message */
+#define HC_REG_USTORM_ADDR_FOR_COALESCE 			 0x108068
+#define HC_REG_VQID_0						 0x108008
+#define HC_REG_VQID_1						 0x10800c
+#define MCP_REG_MCPR_NVM_ACCESS_ENABLE				 0x86424
+#define MCP_REG_MCPR_NVM_ADDR					 0x8640c
+#define MCP_REG_MCPR_NVM_CFG4					 0x8642c
+#define MCP_REG_MCPR_NVM_COMMAND				 0x86400
+#define MCP_REG_MCPR_NVM_READ					 0x86410
+#define MCP_REG_MCPR_NVM_SW_ARB 				 0x86420
+#define MCP_REG_MCPR_NVM_WRITE					 0x86408
+#define MCP_REG_MCPR_NVM_WRITE1 				 0x86428
+#define MCP_REG_MCPR_SCRATCH					 0xa0000
+/* [R 32] read first 32 bit after inversion of function 0. mapped as
+   follows: [0] NIG attention for function0; [1] NIG attention for
+   function1; [2] GPIO1 mcp; [3] GPIO2 mcp; [4] GPIO3 mcp; [5] GPIO4 mcp;
+   [6] GPIO1 function 1; [7] GPIO2 function 1; [8] GPIO3 function 1; [9]
+   GPIO4 function 1; [10] PCIE glue/PXP VPD event function0; [11] PCIE
+   glue/PXP VPD event function1; [12] PCIE glue/PXP Expansion ROM event0;
+   [13] PCIE glue/PXP Expansion ROM event1; [14] SPIO4; [15] SPIO5; [16]
+   MSI/X indication for mcp; [17] MSI/X indication for function 1; [18] BRB
+   Parity error; [19] BRB Hw interrupt; [20] PRS Parity error; [21] PRS Hw
+   interrupt; [22] SRC Parity error; [23] SRC Hw interrupt; [24] TSDM Parity
+   error; [25] TSDM Hw interrupt; [26] TCM Parity error; [27] TCM Hw
+   interrupt; [28] TSEMI Parity error; [29] TSEMI Hw interrupt; [30] PBF
+   Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_AFTER_INVERT_1_FUNC_0			 0xa42c
+#define MISC_REG_AEU_AFTER_INVERT_1_FUNC_1			 0xa430
+/* [R 32] read first 32 bit after inversion of mcp. mapped as follows: [0]
+   NIG attention for function0; [1] NIG attention for function1; [2] GPIO1
+   mcp; [3] GPIO2 mcp; [4] GPIO3 mcp; [5] GPIO4 mcp; [6] GPIO1 function 1;
+   [7] GPIO2 function 1; [8] GPIO3 function 1; [9] GPIO4 function 1; [10]
+   PCIE glue/PXP VPD event function0; [11] PCIE glue/PXP VPD event
+   function1; [12] PCIE glue/PXP Expansion ROM event0; [13] PCIE glue/PXP
+   Expansion ROM event1; [14] SPIO4; [15] SPIO5; [16] MSI/X indication for
+   mcp; [17] MSI/X indication for function 1; [18] BRB Parity error; [19]
+   BRB Hw interrupt; [20] PRS Parity error; [21] PRS Hw interrupt; [22] SRC
+   Parity error; [23] SRC Hw interrupt; [24] TSDM Parity error; [25] TSDM Hw
+   interrupt; [26] TCM Parity error; [27] TCM Hw interrupt; [28] TSEMI
+   Parity error; [29] TSEMI Hw interrupt; [30] PBF Parity error; [31] PBF Hw
+   interrupt; */
+#define MISC_REG_AEU_AFTER_INVERT_1_MCP 			 0xa434
+/* [R 32] read second 32 bit after inversion of function 0. mapped as
+   follows: [0] PBClient Parity error; [1] PBClient Hw interrupt; [2] QM
+   Parity error; [3] QM Hw interrupt; [4] Timers Parity error; [5] Timers Hw
+   interrupt; [6] XSDM Parity error; [7] XSDM Hw interrupt; [8] XCM Parity
+   error; [9] XCM Hw interrupt; [10] XSEMI Parity error; [11] XSEMI Hw
+   interrupt; [12] DoorbellQ Parity error; [13] DoorbellQ Hw interrupt; [14]
+   NIG Parity error; [15] NIG Hw interrupt; [16] Vaux PCI core Parity error;
+   [17] Vaux PCI core Hw interrupt; [18] Debug Parity error; [19] Debug Hw
+   interrupt; [20] USDM Parity error; [21] USDM Hw interrupt; [22] UCM
+   Parity error; [23] UCM Hw interrupt; [24] USEMI Parity error; [25] USEMI
+   Hw interrupt; [26] UPB Parity error; [27] UPB Hw interrupt; [28] CSDM
+   Parity error; [29] CSDM Hw interrupt; [30] CCM Parity error; [31] CCM Hw
+   interrupt; */
+#define MISC_REG_AEU_AFTER_INVERT_2_FUNC_0			 0xa438
+#define MISC_REG_AEU_AFTER_INVERT_2_FUNC_1			 0xa43c
+/* [R 32] read second 32 bit after inversion of mcp. mapped as follows: [0]
+   PBClient Parity error; [1] PBClient Hw interrupt; [2] QM Parity error;
+   [3] QM Hw interrupt; [4] Timers Parity error; [5] Timers Hw interrupt;
+   [6] XSDM Parity error; [7] XSDM Hw interrupt; [8] XCM Parity error; [9]
+   XCM Hw interrupt; [10] XSEMI Parity error; [11] XSEMI Hw interrupt; [12]
+   DoorbellQ Parity error; [13] DoorbellQ Hw interrupt; [14] NIG Parity
+   error; [15] NIG Hw interrupt; [16] Vaux PCI core Parity error; [17] Vaux
+   PCI core Hw interrupt; [18] Debug Parity error; [19] Debug Hw interrupt;
+   [20] USDM Parity error; [21] USDM Hw interrupt; [22] UCM Parity error;
+   [23] UCM Hw interrupt; [24] USEMI Parity error; [25] USEMI Hw interrupt;
+   [26] UPB Parity error; [27] UPB Hw interrupt; [28] CSDM Parity error;
+   [29] CSDM Hw interrupt; [30] CCM Parity error; [31] CCM Hw interrupt; */
+#define MISC_REG_AEU_AFTER_INVERT_2_MCP 			 0xa440
+/* [R 32] read third 32 bit after inversion of function 0. mapped as
+   follows: [0] CSEMI Parity error; [1] CSEMI Hw interrupt; [2] PXP Parity
+   error; [3] PXP Hw interrupt; [4] PXPpciClockClient Parity error; [5]
+   PXPpciClockClient Hw interrupt; [6] CFC Parity error; [7] CFC Hw
+   interrupt; [8] CDU Parity error; [9] CDU Hw interrupt; [10] DMAE Parity
+   error; [11] DMAE Hw interrupt; [12] IGU (HC) Parity error; [13] IGU (HC)
+   Hw interrupt; [14] MISC Parity error; [15] MISC Hw interrupt; [16]
+   pxp_misc_mps_attn; [17] Flash event; [18] SMB event; [19] MCP attn0; [20]
+   MCP attn1; [21] SW timers attn_1 func0; [22] SW timers attn_2 func0; [23]
+   SW timers attn_3 func0; [24] SW timers attn_4 func0; [25] PERST; [26] SW
+   timers attn_1 func1; [27] SW timers attn_2 func1; [28] SW timers attn_3
+   func1; [29] SW timers attn_4 func1; [30] General attn0; [31] General
+   attn1; */
+#define MISC_REG_AEU_AFTER_INVERT_3_FUNC_0			 0xa444
+#define MISC_REG_AEU_AFTER_INVERT_3_FUNC_1			 0xa448
+/* [R 32] read third 32 bit after inversion of mcp. mapped as follows: [0]
+   CSEMI Parity error; [1] CSEMI Hw interrupt; [2] PXP Parity error; [3] PXP
+   Hw interrupt; [4] PXPpciClockClient Parity error; [5] PXPpciClockClient
+   Hw interrupt; [6] CFC Parity error; [7] CFC Hw interrupt; [8] CDU Parity
+   error; [9] CDU Hw interrupt; [10] DMAE Parity error; [11] DMAE Hw
+   interrupt; [12] IGU (HC) Parity error; [13] IGU (HC) Hw interrupt; [14]
+   MISC Parity error; [15] MISC Hw interrupt; [16] pxp_misc_mps_attn; [17]
+   Flash event; [18] SMB event; [19] MCP attn0; [20] MCP attn1; [21] SW
+   timers attn_1 func0; [22] SW timers attn_2 func0; [23] SW timers attn_3
+   func0; [24] SW timers attn_4 func0; [25] PERST; [26] SW timers attn_1
+   func1; [27] SW timers attn_2 func1; [28] SW timers attn_3 func1; [29] SW
+   timers attn_4 func1; [30] General attn0; [31] General attn1; */
+#define MISC_REG_AEU_AFTER_INVERT_3_MCP 			 0xa44c
+/* [R 32] read fourth 32 bit after inversion of function 0. mapped as
+   follows: [0] General attn2; [1] General attn3; [2] General attn4; [3]
+   General attn5; [4] General attn6; [5] General attn7; [6] General attn8;
+   [7] General attn9; [8] General attn10; [9] General attn11; [10] General
+   attn12; [11] General attn13; [12] General attn14; [13] General attn15;
+   [14] General attn16; [15] General attn17; [16] General attn18; [17]
+   General attn19; [18] General attn20; [19] General attn21; [20] Main power
+   interrupt; [21] RBCR Latched attn; [22] RBCT Latched attn; [23] RBCN
+   Latched attn; [24] RBCU Latched attn; [25] RBCP Latched attn; [26] GRC
+   Latched timeout attention; [27] GRC Latched reserved access attention;
+   [28] MCP Latched rom_parity; [29] MCP Latched ump_rx_parity; [30] MCP
+   Latched ump_tx_parity; [31] MCP Latched scpad_parity; */
+#define MISC_REG_AEU_AFTER_INVERT_4_FUNC_0			 0xa450
+#define MISC_REG_AEU_AFTER_INVERT_4_FUNC_1			 0xa454
+/* [R 32] read fourth 32 bit after inversion of mcp. mapped as follows: [0]
+   General attn2; [1] General attn3; [2] General attn4; [3] General attn5;
+   [4] General attn6; [5] General attn7; [6] General attn8; [7] General
+   attn9; [8] General attn10; [9] General attn11; [10] General attn12; [11]
+   General attn13; [12] General attn14; [13] General attn15; [14] General
+   attn16; [15] General attn17; [16] General attn18; [17] General attn19;
+   [18] General attn20; [19] General attn21; [20] Main power interrupt; [21]
+   RBCR Latched attn; [22] RBCT Latched attn; [23] RBCN Latched attn; [24]
+   RBCU Latched attn; [25] RBCP Latched attn; [26] GRC Latched timeout
+   attention; [27] GRC Latched reserved access attention; [28] MCP Latched
+   rom_parity; [29] MCP Latched ump_rx_parity; [30] MCP Latched
+   ump_tx_parity; [31] MCP Latched scpad_parity; */
+#define MISC_REG_AEU_AFTER_INVERT_4_MCP 			 0xa458
+/* [W 11] write to this register results with the clear of the latched
+   signals; one in d0 clears RBCR latch; one in d1 clears RBCT latch; one in
+   d2 clears RBCN latch; one in d3 clears RBCU latch; one in d4 clears RBCP
+   latch; one in d5 clears GRC Latched timeout attention; one in d6 clears
+   GRC Latched reserved access attention; one in d7 clears Latched
+   rom_parity; one in d8 clears Latched ump_rx_parity; one in d9 clears
+   Latched ump_tx_parity; one in d10 clears Latched scpad_parity; read from
+   this register return zero */
+#define MISC_REG_AEU_CLR_LATCH_SIGNAL				 0xa45c
+/* [RW 32] first 32b for enabling the output for function 0 output0. mapped
+   as follows: [0] NIG attention for function0; [1] NIG attention for
+   function1; [2] GPIO1 function 0; [3] GPIO2 function 0; [4] GPIO3 function
+   0; [5] GPIO4 function 0; [6] GPIO1 function 1; [7] GPIO2 function 1; [8]
+   GPIO3 function 1; [9] GPIO4 function 1; [10] PCIE glue/PXP VPD event
+   function0; [11] PCIE glue/PXP VPD event function1; [12] PCIE glue/PXP
+   Expansion ROM event0; [13] PCIE glue/PXP Expansion ROM event1; [14]
+   SPIO4; [15] SPIO5; [16] MSI/X indication for function 0; [17] MSI/X
+   indication for function 1; [18] BRB Parity error; [19] BRB Hw interrupt;
+   [20] PRS Parity error; [21] PRS Hw interrupt; [22] SRC Parity error; [23]
+   SRC Hw interrupt; [24] TSDM Parity error; [25] TSDM Hw interrupt; [26]
+   TCM Parity error; [27] TCM Hw interrupt; [28] TSEMI Parity error; [29]
+   TSEMI Hw interrupt; [30] PBF Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0			 0xa06c
+#define MISC_REG_AEU_ENABLE1_FUNC_0_OUT_1			 0xa07c
+#define MISC_REG_AEU_ENABLE1_FUNC_0_OUT_3			 0xa09c
+/* [RW 32] first 32b for enabling the output for function 1 output0. mapped
+   as follows: [0] NIG attention for function0; [1] NIG attention for
+   function1; [2] GPIO1 function 1; [3] GPIO2 function 1; [4] GPIO3 function
+   1; [5] GPIO4 function 1; [6] GPIO1 function 1; [7] GPIO2 function 1; [8]
+   GPIO3 function 1; [9] GPIO4 function 1; [10] PCIE glue/PXP VPD event
+   function0; [11] PCIE glue/PXP VPD event function1; [12] PCIE glue/PXP
+   Expansion ROM event0; [13] PCIE glue/PXP Expansion ROM event1; [14]
+   SPIO4; [15] SPIO5; [16] MSI/X indication for function 1; [17] MSI/X
+   indication for function 1; [18] BRB Parity error; [19] BRB Hw interrupt;
+   [20] PRS Parity error; [21] PRS Hw interrupt; [22] SRC Parity error; [23]
+   SRC Hw interrupt; [24] TSDM Parity error; [25] TSDM Hw interrupt; [26]
+   TCM Parity error; [27] TCM Hw interrupt; [28] TSEMI Parity error; [29]
+   TSEMI Hw interrupt; [30] PBF Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0			 0xa10c
+#define MISC_REG_AEU_ENABLE1_FUNC_1_OUT_1			 0xa11c
+#define MISC_REG_AEU_ENABLE1_FUNC_1_OUT_3			 0xa13c
+/* [RW 32] first 32b for enabling the output for close the gate nig 0.
+   mapped as follows: [0] NIG attention for function0; [1] NIG attention for
+   function1; [2] GPIO1 function 0; [3] GPIO2 function 0; [4] GPIO3 function
+   0; [5] GPIO4 function 0; [6] GPIO1 function 1; [7] GPIO2 function 1; [8]
+   GPIO3 function 1; [9] GPIO4 function 1; [10] PCIE glue/PXP VPD event
+   function0; [11] PCIE glue/PXP VPD event function1; [12] PCIE glue/PXP
+   Expansion ROM event0; [13] PCIE glue/PXP Expansion ROM event1; [14]
+   SPIO4; [15] SPIO5; [16] MSI/X indication for function 0; [17] MSI/X
+   indication for function 1; [18] BRB Parity error; [19] BRB Hw interrupt;
+   [20] PRS Parity error; [21] PRS Hw interrupt; [22] SRC Parity error; [23]
+   SRC Hw interrupt; [24] TSDM Parity error; [25] TSDM Hw interrupt; [26]
+   TCM Parity error; [27] TCM Hw interrupt; [28] TSEMI Parity error; [29]
+   TSEMI Hw interrupt; [30] PBF Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_ENABLE1_NIG_0				 0xa0ec
+#define MISC_REG_AEU_ENABLE1_NIG_1				 0xa18c
+/* [RW 32] first 32b for enabling the output for close the gate pxp 0.
+   mapped as follows: [0] NIG attention for function0; [1] NIG attention for
+   function1; [2] GPIO1 function 0; [3] GPIO2 function 0; [4] GPIO3 function
+   0; [5] GPIO4 function 0; [6] GPIO1 function 1; [7] GPIO2 function 1; [8]
+   GPIO3 function 1; [9] GPIO4 function 1; [10] PCIE glue/PXP VPD event
+   function0; [11] PCIE glue/PXP VPD event function1; [12] PCIE glue/PXP
+   Expansion ROM event0; [13] PCIE glue/PXP Expansion ROM event1; [14]
+   SPIO4; [15] SPIO5; [16] MSI/X indication for function 0; [17] MSI/X
+   indication for function 1; [18] BRB Parity error; [19] BRB Hw interrupt;
+   [20] PRS Parity error; [21] PRS Hw interrupt; [22] SRC Parity error; [23]
+   SRC Hw interrupt; [24] TSDM Parity error; [25] TSDM Hw interrupt; [26]
+   TCM Parity error; [27] TCM Hw interrupt; [28] TSEMI Parity error; [29]
+   TSEMI Hw interrupt; [30] PBF Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_ENABLE1_PXP_0				 0xa0fc
+#define MISC_REG_AEU_ENABLE1_PXP_1				 0xa19c
+/* [RW 32] second 32b for enabling the output for function 0 output0. mapped
+   as follows: [0] PBClient Parity error; [1] PBClient Hw interrupt; [2] QM
+   Parity error; [3] QM Hw interrupt; [4] Timers Parity error; [5] Timers Hw
+   interrupt; [6] XSDM Parity error; [7] XSDM Hw interrupt; [8] XCM Parity
+   error; [9] XCM Hw interrupt; [10] XSEMI Parity error; [11] XSEMI Hw
+   interrupt; [12] DoorbellQ Parity error; [13] DoorbellQ Hw interrupt; [14]
+   NIG Parity error; [15] NIG Hw interrupt; [16] Vaux PCI core Parity error;
+   [17] Vaux PCI core Hw interrupt; [18] Debug Parity error; [19] Debug Hw
+   interrupt; [20] USDM Parity error; [21] USDM Hw interrupt; [22] UCM
+   Parity error; [23] UCM Hw interrupt; [24] USEMI Parity error; [25] USEMI
+   Hw interrupt; [26] UPB Parity error; [27] UPB Hw interrupt; [28] CSDM
+   Parity error; [29] CSDM Hw interrupt; [30] CCM Parity error; [31] CCM Hw
+   interrupt; */
+#define MISC_REG_AEU_ENABLE2_FUNC_0_OUT_0			 0xa070
+#define MISC_REG_AEU_ENABLE2_FUNC_0_OUT_1			 0xa080
+/* [RW 32] second 32b for enabling the output for function 1 output0. mapped
+   as follows: [0] PBClient Parity error; [1] PBClient Hw interrupt; [2] QM
+   Parity error; [3] QM Hw interrupt; [4] Timers Parity error; [5] Timers Hw
+   interrupt; [6] XSDM Parity error; [7] XSDM Hw interrupt; [8] XCM Parity
+   error; [9] XCM Hw interrupt; [10] XSEMI Parity error; [11] XSEMI Hw
+   interrupt; [12] DoorbellQ Parity error; [13] DoorbellQ Hw interrupt; [14]
+   NIG Parity error; [15] NIG Hw interrupt; [16] Vaux PCI core Parity error;
+   [17] Vaux PCI core Hw interrupt; [18] Debug Parity error; [19] Debug Hw
+   interrupt; [20] USDM Parity error; [21] USDM Hw interrupt; [22] UCM
+   Parity error; [23] UCM Hw interrupt; [24] USEMI Parity error; [25] USEMI
+   Hw interrupt; [26] UPB Parity error; [27] UPB Hw interrupt; [28] CSDM
+   Parity error; [29] CSDM Hw interrupt; [30] CCM Parity error; [31] CCM Hw
+   interrupt; */
+#define MISC_REG_AEU_ENABLE2_FUNC_1_OUT_0			 0xa110
+#define MISC_REG_AEU_ENABLE2_FUNC_1_OUT_1			 0xa120
+/* [RW 32] second 32b for enabling the output for close the gate nig 0.
+   mapped as follows: [0] PBClient Parity error; [1] PBClient Hw interrupt;
+   [2] QM Parity error; [3] QM Hw interrupt; [4] Timers Parity error; [5]
+   Timers Hw interrupt; [6] XSDM Parity error; [7] XSDM Hw interrupt; [8]
+   XCM Parity error; [9] XCM Hw interrupt; [10] XSEMI Parity error; [11]
+   XSEMI Hw interrupt; [12] DoorbellQ Parity error; [13] DoorbellQ Hw
+   interrupt; [14] NIG Parity error; [15] NIG Hw interrupt; [16] Vaux PCI
+   core Parity error; [17] Vaux PCI core Hw interrupt; [18] Debug Parity
+   error; [19] Debug Hw interrupt; [20] USDM Parity error; [21] USDM Hw
+   interrupt; [22] UCM Parity error; [23] UCM Hw interrupt; [24] USEMI
+   Parity error; [25] USEMI Hw interrupt; [26] UPB Parity error; [27] UPB Hw
+   interrupt; [28] CSDM Parity error; [29] CSDM Hw interrupt; [30] CCM
+   Parity error; [31] CCM Hw interrupt; */
+#define MISC_REG_AEU_ENABLE2_NIG_0				 0xa0f0
+#define MISC_REG_AEU_ENABLE2_NIG_1				 0xa190
+/* [RW 32] second 32b for enabling the output for close the gate pxp 0.
+   mapped as follows: [0] PBClient Parity error; [1] PBClient Hw interrupt;
+   [2] QM Parity error; [3] QM Hw interrupt; [4] Timers Parity error; [5]
+   Timers Hw interrupt; [6] XSDM Parity error; [7] XSDM Hw interrupt; [8]
+   XCM Parity error; [9] XCM Hw interrupt; [10] XSEMI Parity error; [11]
+   XSEMI Hw interrupt; [12] DoorbellQ Parity error; [13] DoorbellQ Hw
+   interrupt; [14] NIG Parity error; [15] NIG Hw interrupt; [16] Vaux PCI
+   core Parity error; [17] Vaux PCI core Hw interrupt; [18] Debug Parity
+   error; [19] Debug Hw interrupt; [20] USDM Parity error; [21] USDM Hw
+   interrupt; [22] UCM Parity error; [23] UCM Hw interrupt; [24] USEMI
+   Parity error; [25] USEMI Hw interrupt; [26] UPB Parity error; [27] UPB Hw
+   interrupt; [28] CSDM Parity error; [29] CSDM Hw interrupt; [30] CCM
+   Parity error; [31] CCM Hw interrupt; */
+#define MISC_REG_AEU_ENABLE2_PXP_0				 0xa100
+#define MISC_REG_AEU_ENABLE2_PXP_1				 0xa1a0
+/* [RW 32] third 32b for enabling the output for function 0 output0. mapped
+   as follows: [0] CSEMI Parity error; [1] CSEMI Hw interrupt; [2] PXP
+   Parity error; [3] PXP Hw interrupt; [4] PXPpciClockClient Parity error;
+   [5] PXPpciClockClient Hw interrupt; [6] CFC Parity error; [7] CFC Hw
+   interrupt; [8] CDU Parity error; [9] CDU Hw interrupt; [10] DMAE Parity
+   error; [11] DMAE Hw interrupt; [12] IGU (HC) Parity error; [13] IGU (HC)
+   Hw interrupt; [14] MISC Parity error; [15] MISC Hw interrupt; [16]
+   pxp_misc_mps_attn; [17] Flash event; [18] SMB event; [19] MCP attn0; [20]
+   MCP attn1; [21] SW timers attn_1 func0; [22] SW timers attn_2 func0; [23]
+   SW timers attn_3 func0; [24] SW timers attn_4 func0; [25] PERST; [26] SW
+   timers attn_1 func1; [27] SW timers attn_2 func1; [28] SW timers attn_3
+   func1; [29] SW timers attn_4 func1; [30] General attn0; [31] General
+   attn1; */
+#define MISC_REG_AEU_ENABLE3_FUNC_0_OUT_0			 0xa074
+#define MISC_REG_AEU_ENABLE3_FUNC_0_OUT_1			 0xa084
+/* [RW 32] third 32b for enabling the output for function 1 output0. mapped
+   as follows: [0] CSEMI Parity error; [1] CSEMI Hw interrupt; [2] PXP
+   Parity error; [3] PXP Hw interrupt; [4] PXPpciClockClient Parity error;
+   [5] PXPpciClockClient Hw interrupt; [6] CFC Parity error; [7] CFC Hw
+   interrupt; [8] CDU Parity error; [9] CDU Hw interrupt; [10] DMAE Parity
+   error; [11] DMAE Hw interrupt; [12] IGU (HC) Parity error; [13] IGU (HC)
+   Hw interrupt; [14] MISC Parity error; [15] MISC Hw interrupt; [16]
+   pxp_misc_mps_attn; [17] Flash event; [18] SMB event; [19] MCP attn0; [20]
+   MCP attn1; [21] SW timers attn_1 func0; [22] SW timers attn_2 func0; [23]
+   SW timers attn_3 func0; [24] SW timers attn_4 func0; [25] PERST; [26] SW
+   timers attn_1 func1; [27] SW timers attn_2 func1; [28] SW timers attn_3
+   func1; [29] SW timers attn_4 func1; [30] General attn0; [31] General
+   attn1; */
+#define MISC_REG_AEU_ENABLE3_FUNC_1_OUT_0			 0xa114
+#define MISC_REG_AEU_ENABLE3_FUNC_1_OUT_1			 0xa124
+/* [RW 32] third 32b for enabling the output for close the gate nig 0.
+   mapped as follows: [0] CSEMI Parity error; [1] CSEMI Hw interrupt; [2]
+   PXP Parity error; [3] PXP Hw interrupt; [4] PXPpciClockClient Parity
+   error; [5] PXPpciClockClient Hw interrupt; [6] CFC Parity error; [7] CFC
+   Hw interrupt; [8] CDU Parity error; [9] CDU Hw interrupt; [10] DMAE
+   Parity error; [11] DMAE Hw interrupt; [12] IGU (HC) Parity error; [13]
+   IGU (HC) Hw interrupt; [14] MISC Parity error; [15] MISC Hw interrupt;
+   [16] pxp_misc_mps_attn; [17] Flash event; [18] SMB event; [19] MCP attn0;
+   [20] MCP attn1; [21] SW timers attn_1 func0; [22] SW timers attn_2 func0;
+   [23] SW timers attn_3 func0; [24] SW timers attn_4 func0; [25] PERST;
+   [26] SW timers attn_1 func1; [27] SW timers attn_2 func1; [28] SW timers
+   attn_3 func1; [29] SW timers attn_4 func1; [30] General attn0; [31]
+   General attn1; */
+#define MISC_REG_AEU_ENABLE3_NIG_0				 0xa0f4
+#define MISC_REG_AEU_ENABLE3_NIG_1				 0xa194
+/* [RW 32] third 32b for enabling the output for close the gate pxp 0.
+   mapped as follows: [0] CSEMI Parity error; [1] CSEMI Hw interrupt; [2]
+   PXP Parity error; [3] PXP Hw interrupt; [4] PXPpciClockClient Parity
+   error; [5] PXPpciClockClient Hw interrupt; [6] CFC Parity error; [7] CFC
+   Hw interrupt; [8] CDU Parity error; [9] CDU Hw interrupt; [10] DMAE
+   Parity error; [11] DMAE Hw interrupt; [12] IGU (HC) Parity error; [13]
+   IGU (HC) Hw interrupt; [14] MISC Parity error; [15] MISC Hw interrupt;
+   [16] pxp_misc_mps_attn; [17] Flash event; [18] SMB event; [19] MCP attn0;
+   [20] MCP attn1; [21] SW timers attn_1 func0; [22] SW timers attn_2 func0;
+   [23] SW timers attn_3 func0; [24] SW timers attn_4 func0; [25] PERST;
+   [26] SW timers attn_1 func1; [27] SW timers attn_2 func1; [28] SW timers
+   attn_3 func1; [29] SW timers attn_4 func1; [30] General attn0; [31]
+   General attn1; */
+#define MISC_REG_AEU_ENABLE3_PXP_0				 0xa104
+#define MISC_REG_AEU_ENABLE3_PXP_1				 0xa1a4
+/* [RW 32] fourth 32b for enabling the output for function 0 output0.mapped
+   as follows: [0] General attn2; [1] General attn3; [2] General attn4; [3]
+   General attn5; [4] General attn6; [5] General attn7; [6] General attn8;
+   [7] General attn9; [8] General attn10; [9] General attn11; [10] General
+   attn12; [11] General attn13; [12] General attn14; [13] General attn15;
+   [14] General attn16; [15] General attn17; [16] General attn18; [17]
+   General attn19; [18] General attn20; [19] General attn21; [20] Main power
+   interrupt; [21] RBCR Latched attn; [22] RBCT Latched attn; [23] RBCN
+   Latched attn; [24] RBCU Latched attn; [25] RBCP Latched attn; [26] GRC
+   Latched timeout attention; [27] GRC Latched reserved access attention;
+   [28] MCP Latched rom_parity; [29] MCP Latched ump_rx_parity; [30] MCP
+   Latched ump_tx_parity; [31] MCP Latched scpad_parity; */
+#define MISC_REG_AEU_ENABLE4_FUNC_0_OUT_0			 0xa078
+#define MISC_REG_AEU_ENABLE4_FUNC_0_OUT_2			 0xa098
+/* [RW 32] fourth 32b for enabling the output for function 1 output0.mapped
+   as follows: [0] General attn2; [1] General attn3; [2] General attn4; [3]
+   General attn5; [4] General attn6; [5] General attn7; [6] General attn8;
+   [7] General attn9; [8] General attn10; [9] General attn11; [10] General
+   attn12; [11] General attn13; [12] General attn14; [13] General attn15;
+   [14] General attn16; [15] General attn17; [16] General attn18; [17]
+   General attn19; [18] General attn20; [19] General attn21; [20] Main power
+   interrupt; [21] RBCR Latched attn; [22] RBCT Latched attn; [23] RBCN
+   Latched attn; [24] RBCU Latched attn; [25] RBCP Latched attn; [26] GRC
+   Latched timeout attention; [27] GRC Latched reserved access attention;
+   [28] MCP Latched rom_parity; [29] MCP Latched ump_rx_parity; [30] MCP
+   Latched ump_tx_parity; [31] MCP Latched scpad_parity; */
+#define MISC_REG_AEU_ENABLE4_FUNC_1_OUT_0			 0xa118
+#define MISC_REG_AEU_ENABLE4_FUNC_1_OUT_2			 0xa138
+/* [RW 32] fourth 32b for enabling the output for close the gate nig
+   0.mapped as follows: [0] General attn2; [1] General attn3; [2] General
+   attn4; [3] General attn5; [4] General attn6; [5] General attn7; [6]
+   General attn8; [7] General attn9; [8] General attn10; [9] General attn11;
+   [10] General attn12; [11] General attn13; [12] General attn14; [13]
+   General attn15; [14] General attn16; [15] General attn17; [16] General
+   attn18; [17] General attn19; [18] General attn20; [19] General attn21;
+   [20] Main power interrupt; [21] RBCR Latched attn; [22] RBCT Latched
+   attn; [23] RBCN Latched attn; [24] RBCU Latched attn; [25] RBCP Latched
+   attn; [26] GRC Latched timeout attention; [27] GRC Latched reserved
+   access attention; [28] MCP Latched rom_parity; [29] MCP Latched
+   ump_rx_parity; [30] MCP Latched ump_tx_parity; [31] MCP Latched
+   scpad_parity; */
+#define MISC_REG_AEU_ENABLE4_NIG_0				 0xa0f8
+#define MISC_REG_AEU_ENABLE4_NIG_1				 0xa198
+/* [RW 32] fourth 32b for enabling the output for close the gate pxp
+   0.mapped as follows: [0] General attn2; [1] General attn3; [2] General
+   attn4; [3] General attn5; [4] General attn6; [5] General attn7; [6]
+   General attn8; [7] General attn9; [8] General attn10; [9] General attn11;
+   [10] General attn12; [11] General attn13; [12] General attn14; [13]
+   General attn15; [14] General attn16; [15] General attn17; [16] General
+   attn18; [17] General attn19; [18] General attn20; [19] General attn21;
+   [20] Main power interrupt; [21] RBCR Latched attn; [22] RBCT Latched
+   attn; [23] RBCN Latched attn; [24] RBCU Latched attn; [25] RBCP Latched
+   attn; [26] GRC Latched timeout attention; [27] GRC Latched reserved
+   access attention; [28] MCP Latched rom_parity; [29] MCP Latched
+   ump_rx_parity; [30] MCP Latched ump_tx_parity; [31] MCP Latched
+   scpad_parity; */
+#define MISC_REG_AEU_ENABLE4_PXP_0				 0xa108
+#define MISC_REG_AEU_ENABLE4_PXP_1				 0xa1a8
+/* [RW 1] set/clr general attention 0; this will set/clr bit 94 in the aeu
+   128 bit vector */
+#define MISC_REG_AEU_GENERAL_ATTN_0				 0xa000
+#define MISC_REG_AEU_GENERAL_ATTN_1				 0xa004
+#define MISC_REG_AEU_GENERAL_ATTN_10				 0xa028
+#define MISC_REG_AEU_GENERAL_ATTN_11				 0xa02c
+#define MISC_REG_AEU_GENERAL_ATTN_12				 0xa030
+#define MISC_REG_AEU_GENERAL_ATTN_13				 0xa034
+#define MISC_REG_AEU_GENERAL_ATTN_14				 0xa038
+#define MISC_REG_AEU_GENERAL_ATTN_15				 0xa03c
+#define MISC_REG_AEU_GENERAL_ATTN_16				 0xa040
+#define MISC_REG_AEU_GENERAL_ATTN_17				 0xa044
+#define MISC_REG_AEU_GENERAL_ATTN_18				 0xa048
+#define MISC_REG_AEU_GENERAL_ATTN_19				 0xa04c
+#define MISC_REG_AEU_GENERAL_ATTN_11				 0xa02c
+#define MISC_REG_AEU_GENERAL_ATTN_2				 0xa008
+#define MISC_REG_AEU_GENERAL_ATTN_20				 0xa050
+#define MISC_REG_AEU_GENERAL_ATTN_21				 0xa054
+#define MISC_REG_AEU_GENERAL_ATTN_3				 0xa00c
+#define MISC_REG_AEU_GENERAL_ATTN_4				 0xa010
+#define MISC_REG_AEU_GENERAL_ATTN_5				 0xa014
+#define MISC_REG_AEU_GENERAL_ATTN_6				 0xa018
+/* [RW 32] first 32b for inverting the input for function 0; for each bit:
+   0= do not invert; 1= invert; mapped as follows: [0] NIG attention for
+   function0; [1] NIG attention for function1; [2] GPIO1 mcp; [3] GPIO2 mcp;
+   [4] GPIO3 mcp; [5] GPIO4 mcp; [6] GPIO1 function 1; [7] GPIO2 function 1;
+   [8] GPIO3 function 1; [9] GPIO4 function 1; [10] PCIE glue/PXP VPD event
+   function0; [11] PCIE glue/PXP VPD event function1; [12] PCIE glue/PXP
+   Expansion ROM event0; [13] PCIE glue/PXP Expansion ROM event1; [14]
+   SPIO4; [15] SPIO5; [16] MSI/X indication for mcp; [17] MSI/X indication
+   for function 1; [18] BRB Parity error; [19] BRB Hw interrupt; [20] PRS
+   Parity error; [21] PRS Hw interrupt; [22] SRC Parity error; [23] SRC Hw
+   interrupt; [24] TSDM Parity error; [25] TSDM Hw interrupt; [26] TCM
+   Parity error; [27] TCM Hw interrupt; [28] TSEMI Parity error; [29] TSEMI
+   Hw interrupt; [30] PBF Parity error; [31] PBF Hw interrupt; */
+#define MISC_REG_AEU_INVERTER_1_FUNC_0				 0xa22c
+#define MISC_REG_AEU_INVERTER_1_FUNC_1				 0xa23c
+/* [RW 32] second 32b for inverting the input for function 0; for each bit:
+   0= do not invert; 1= invert. mapped as follows: [0] PBClient Parity
+   error; [1] PBClient Hw interrupt; [2] QM Parity error; [3] QM Hw
+   interrupt; [4] Timers Parity error; [5] Timers Hw interrupt; [6] XSDM
+   Parity error; [7] XSDM Hw interrupt; [8] XCM Parity error; [9] XCM Hw
+   interrupt; [10] XSEMI Parity error; [11] XSEMI Hw interrupt; [12]
+   DoorbellQ Parity error; [13] DoorbellQ Hw interrupt; [14] NIG Parity
+   error; [15] NIG Hw interrupt; [16] Vaux PCI core Parity error; [17] Vaux
+   PCI core Hw interrupt; [18] Debug Parity error; [19] Debug Hw interrupt;
+   [20] USDM Parity error; [21] USDM Hw interrupt; [22] UCM Parity error;
+   [23] UCM Hw interrupt; [24] USEMI Parity error; [25] USEMI Hw interrupt;
+   [26] UPB Parity error; [27] UPB Hw interrupt; [28] CSDM Parity error;
+   [29] CSDM Hw interrupt; [30] CCM Parity error; [31] CCM Hw interrupt; */
+#define MISC_REG_AEU_INVERTER_2_FUNC_0				 0xa230
+#define MISC_REG_AEU_INVERTER_2_FUNC_1				 0xa240
+/* [RW 10] [7:0] = mask 8 attention output signals toward IGU function0;
+   [9:8] = mask close the gates signals of function 0 toward PXP [8] and NIG
+   [9]. Zero = mask; one = unmask */
+#define MISC_REG_AEU_MASK_ATTN_FUNC_0				 0xa060
+#define MISC_REG_AEU_MASK_ATTN_FUNC_1				 0xa064
+/* [R 4] This field indicates the type of the device. '0' - 2 Ports; '1' - 1
+   Port. */
+#define MISC_REG_BOND_ID					 0xa400
+/* [R 8] These bits indicate the metal revision of the chip. This value
+   starts at 0x00 for each all-layer tape-out and increments by one for each
+   tape-out. */
+#define MISC_REG_CHIP_METAL					 0xa404
+/* [R 16] These bits indicate the part number for the chip. */
+#define MISC_REG_CHIP_NUM					 0xa408
+/* [R 4] These bits indicate the base revision of the chip. This value
+   starts at 0x0 for the A0 tape-out and increments by one for each
+   all-layer tape-out. */
+#define MISC_REG_CHIP_REV					 0xa40c
+/* [RW 1] Setting this bit enables a timer in the GRC block to timeout any
+   access that does not finish within
+   ~misc_registers_grc_timout_val.grc_timeout_val cycles. When this bit is
+   cleared; this timeout is disabled. If this timeout occurs; the GRC shall
+   assert it attention output. */
+#define MISC_REG_GRC_TIMEOUT_EN 				 0xa280
+/* [RW 28] 28 LSB of LCPLL first register; reset val = 521. inside order of
+   the bits is: [2:0] OAC reset value 001) CML output buffer bias control;
+   111 for +40%; 011 for +20%; 001 for 0%; 000 for -20%. [5:3] Icp_ctrl
+   (reset value 001) Charge pump current control; 111 for 720u; 011 for
+   600u; 001 for 480u and 000 for 360u. [7:6] Bias_ctrl (reset value 00)
+   Global bias control; When bit 7 is high bias current will be 10 0gh; When
+   bit 6 is high bias will be 100w; Valid values are 00; 10; 01. [10:8]
+   Pll_observe (reset value 010) Bits to control observability. bit 10 is
+   for test bias; bit 9 is for test CK; bit 8 is test Vc. [12:11] Vth_ctrl
+   (reset value 00) Comparator threshold control. 00 for 0.6V; 01 for 0.54V
+   and 10 for 0.66V. [13] pllSeqStart (reset value 0) Enables VCO tuning
+   sequencer: 1= sequencer disabled; 0= sequencer enabled (inverted
+   internally). [14] reserved (reset value 0) Reset for VCO sequencer is
+   connected to RESET input directly. [15] capRetry_en (reset value 0)
+   enable retry on cap search failure (inverted). [16] freqMonitor_e (reset
+   value 0) bit to continuously monitor vco freq (inverted). [17]
+   freqDetRestart_en (reset value 0) bit to enable restart when not freq
+   locked (inverted). [18] freqDetRetry_en (reset value 0) bit to enable
+   retry on freq det failure(inverted). [19] pllForceFdone_en (reset value
+   0) bit to enable pllForceFdone & pllForceFpass into pllSeq. [20]
+   pllForceFdone (reset value 0) bit to force freqDone. [21] pllForceFpass
+   (reset value 0) bit to force freqPass. [22] pllForceDone_en (reset value
+   0) bit to enable pllForceCapDone. [23] pllForceCapDone (reset value 0)
+   bit to force capDone. [24] pllForceCapPass_en (reset value 0) bit to
+   enable pllForceCapPass. [25] pllForceCapPass (reset value 0) bit to force
+   capPass. [26] capRestart (reset value 0) bit to force cap sequencer to
+   restart. [27] capSelectM_en (reset value 0) bit to enable cap select
+   register bits. */
+#define MISC_REG_LCPLL_CTRL_1					 0xa2a4
+#define MISC_REG_LCPLL_CTRL_REG_2				 0xa2a8
+/* [RW 4] Interrupt mask register #0 read/write */
+#define MISC_REG_MISC_INT_MASK					 0xa388
+/* [RW 1] Parity mask register #0 read/write */
+#define MISC_REG_MISC_PRTY_MASK 				 0xa398
+/* [RW 32] 32 LSB of storm PLL first register; reset val = 0x 071d2911.
+   inside order of the bits is: [0] P1 divider[0] (reset value 1); [1] P1
+   divider[1] (reset value 0); [2] P1 divider[2] (reset value 0); [3] P1
+   divider[3] (reset value 0); [4] P2 divider[0] (reset value 1); [5] P2
+   divider[1] (reset value 0); [6] P2 divider[2] (reset value 0); [7] P2
+   divider[3] (reset value 0); [8] ph_det_dis (reset value 1); [9]
+   freq_det_dis (reset value 0); [10] Icpx[0] (reset value 0); [11] Icpx[1]
+   (reset value 1); [12] Icpx[2] (reset value 0); [13] Icpx[3] (reset value
+   1); [14] Icpx[4] (reset value 0); [15] Icpx[5] (reset value 0); [16]
+   Rx[0] (reset value 1); [17] Rx[1] (reset value 0); [18] vc_en (reset
+   value 1); [19] vco_rng[0] (reset value 1); [20] vco_rng[1] (reset value
+   1); [21] Kvco_xf[0] (reset value 0); [22] Kvco_xf[1] (reset value 0);
+   [23] Kvco_xf[2] (reset value 0); [24] Kvco_xs[0] (reset value 1); [25]
+   Kvco_xs[1] (reset value 1); [26] Kvco_xs[2] (reset value 1); [27]
+   testd_en (reset value 0); [28] testd_sel[0] (reset value 0); [29]
+   testd_sel[1] (reset value 0); [30] testd_sel[2] (reset value 0); [31]
+   testa_en (reset value 0); */
+#define MISC_REG_PLL_STORM_CTRL_1				 0xa294
+#define MISC_REG_PLL_STORM_CTRL_2				 0xa298
+#define MISC_REG_PLL_STORM_CTRL_3				 0xa29c
+#define MISC_REG_PLL_STORM_CTRL_4				 0xa2a0
+/* [RW 32] reset reg#1; rite/read one = the specific block is out of reset;
+   write/read zero = the specific block is in reset; addr 0-wr- the write
+   value will be written to the register; addr 1-set - one will be written
+   to all the bits that have the value of one in the data written (bits that
+   have the value of zero will not be change) ; addr 2-clear - zero will be
+   written to all the bits that have the value of one in the data written
+   (bits that have the value of zero will not be change); addr 3-ignore;
+   read ignore from all addr except addr 00; inside order of the bits is:
+   [0] rst_brb1; [1] rst_prs; [2] rst_src; [3] rst_tsdm; [4] rst_tsem; [5]
+   rst_tcm; [6] rst_rbcr; [7] rst_nig; [8] rst_usdm; [9] rst_ucm; [10]
+   rst_usem; [11] rst_upb; [12] rst_ccm; [13] rst_csem; [14] rst_csdm; [15]
+   rst_rbcu; [16] rst_pbf; [17] rst_qm; [18] rst_tm; [19] rst_dorq; [20]
+   rst_xcm; [21] rst_xsdm; [22] rst_xsem; [23] rst_rbct; [24] rst_cdu; [25]
+   rst_cfc; [26] rst_pxp; [27] rst_pxpv; [28] rst_rbcp; [29] rst_hc; [30]
+   rst_dmae; [31] rst_semi_rtc; */
+#define MISC_REG_RESET_REG_1					 0xa580
+#define MISC_REG_RESET_REG_2					 0xa590
+/* [RW 20] 20 bit GRC address where the scratch-pad of the MCP that is
+   shared with the driver resides */
+#define MISC_REG_SHARED_MEM_ADDR				 0xa2b4
+#define NIG_MASK_INTERRUPT_PORT0_REG_MASK_EMAC0_MISC_MI_INT	 (0x1<<0)
+#define NIG_MASK_INTERRUPT_PORT0_REG_MASK_SERDES0_LINK_STATUS	 (0x1<<9)
+#define NIG_MASK_INTERRUPT_PORT0_REG_MASK_XGXS0_LINK10G 	 (0x1<<15)
+#define NIG_MASK_INTERRUPT_PORT0_REG_MASK_XGXS0_LINK_STATUS	 (0xf<<18)
+/* [RW 1] Input enable for RX_BMAC0 IF */
+#define NIG_REG_BMAC0_IN_EN					 0x100ac
+/* [RW 1] output enable for TX_BMAC0 IF */
+#define NIG_REG_BMAC0_OUT_EN					 0x100e0
+/* [RW 1] output enable for TX BMAC pause port 0 IF */
+#define NIG_REG_BMAC0_PAUSE_OUT_EN				 0x10110
+/* [RW 1] output enable for RX_BMAC0_REGS IF */
+#define NIG_REG_BMAC0_REGS_OUT_EN				 0x100e8
+/* [RW 1] output enable for RX BRB1 port0 IF */
+#define NIG_REG_BRB0_OUT_EN					 0x100f8
+/* [RW 1] Input enable for TX BRB1 pause port 0 IF */
+#define NIG_REG_BRB0_PAUSE_IN_EN				 0x100c4
+/* [RW 1] output enable for RX BRB1 port1 IF */
+#define NIG_REG_BRB1_OUT_EN					 0x100fc
+/* [RW 1] Input enable for TX BRB1 pause port 1 IF */
+#define NIG_REG_BRB1_PAUSE_IN_EN				 0x100c8
+/* [RW 1] output enable for RX BRB1 LP IF */
+#define NIG_REG_BRB_LB_OUT_EN					 0x10100
+/* [WB_W 72] Debug packet to LP from RBC; Data spelling:[63:0] data; 64]
+   error; [67:65]eop_bvalid; [68]eop; [69]sop; [70]port_id; 71]flush */
+#define NIG_REG_DEBUG_PACKET_LB 				 0x10800
+/* [RW 1] Input enable for TX Debug packet */
+#define NIG_REG_EGRESS_DEBUG_IN_EN				 0x100dc
+/* [RW 1] If 1 - egress drain mode for port0 is active. In this mode all
+   packets from PBFare not forwarded to the MAC and just deleted from FIFO.
+   First packet may be deleted from the middle. And last packet will be
+   always deleted till the end. */
+#define NIG_REG_EGRESS_DRAIN0_MODE				 0x10060
+/* [RW 1] Output enable to EMAC0 */
+#define NIG_REG_EGRESS_EMAC0_OUT_EN				 0x10120
+/* [RW 1] MAC configuration for packets of port0. If 1 - all packet outputs
+   to emac for port0; other way to bmac for port0 */
+#define NIG_REG_EGRESS_EMAC0_PORT				 0x10058
+/* [RW 1] Input enable for TX PBF user packet port0 IF */
+#define NIG_REG_EGRESS_PBF0_IN_EN				 0x100cc
+/* [RW 1] Input enable for TX PBF user packet port1 IF */
+#define NIG_REG_EGRESS_PBF1_IN_EN				 0x100d0
+/* [RW 1] Input enable for RX_EMAC0 IF */
+#define NIG_REG_EMAC0_IN_EN					 0x100a4
+/* [RW 1] output enable for TX EMAC pause port 0 IF */
+#define NIG_REG_EMAC0_PAUSE_OUT_EN				 0x10118
+/* [R 1] status from emac0. This bit is set when MDINT from either the
+   EXT_MDINT pin or from the Copper PHY is driven low. This condition must
+   be cleared in the attached PHY device that is driving the MINT pin. */
+#define NIG_REG_EMAC0_STATUS_MISC_MI_INT			 0x10494
+/* [WB 48] This address space contains BMAC0 registers. The BMAC registers
+   are described in appendix A. In order to access the BMAC0 registers; the
+   base address; NIG_REGISTERS_INGRESS_BMAC0_MEM; Offset: 0x10c00; should be
+   added to each BMAC register offset */
+#define NIG_REG_INGRESS_BMAC0_MEM				 0x10c00
+/* [WB 48] This address space contains BMAC1 registers. The BMAC registers
+   are described in appendix A. In order to access the BMAC0 registers; the
+   base address; NIG_REGISTERS_INGRESS_BMAC1_MEM; Offset: 0x11000; should be
+   added to each BMAC register offset */
+#define NIG_REG_INGRESS_BMAC1_MEM				 0x11000
+/* [R 1] FIFO empty in EOP descriptor FIFO of LP in NIG_RX_EOP */
+#define NIG_REG_INGRESS_EOP_LB_EMPTY				 0x104e0
+/* [RW 17] Debug only. RX_EOP_DSCR_lb_FIFO in NIG_RX_EOP. Data
+   packet_length[13:0]; mac_error[14]; trunc_error[15]; parity[16] */
+#define NIG_REG_INGRESS_EOP_LB_FIFO				 0x104e4
+/* [RW 1] led 10g for port 0 */
+#define NIG_REG_LED_10G_P0					 0x10320
+/* [RW 1] Port0: This bit is set to enable the use of the
+   ~nig_registers_led_control_blink_rate_p0.led_control_blink_rate_p0 field
+   defined below. If this bit is cleared; then the blink rate will be about
+   8Hz. */
+#define NIG_REG_LED_CONTROL_BLINK_RATE_ENA_P0			 0x10318
+/* [RW 12] Port0: Specifies the period of each blink cycle (on + off) for
+   Traffic LED in milliseconds. Must be a non-zero value. This 12-bit field
+   is reset to 0x080; giving a default blink period of approximately 8Hz. */
+#define NIG_REG_LED_CONTROL_BLINK_RATE_P0			 0x10310
+/* [RW 1] Port0: If set along with the
+   nig_registers_led_control_override_traffic_p0.led_control_override_traffic_p0
+   bit and ~nig_registers_led_control_traffic_p0.led_control_traffic_p0 LED
+   bit; the Traffic LED will blink with the blink rate specified in
+   ~nig_registers_led_control_blink_rate_p0.led_control_blink_rate_p0 and
+   ~nig_registers_led_control_blink_rate_ena_p0.led_control_blink_rate_ena_p0
+   fields. */
+#define NIG_REG_LED_CONTROL_BLINK_TRAFFIC_P0			 0x10308
+/* [RW 1] Port0: If set overrides hardware control of the Traffic LED. The
+   Traffic LED will then be controlled via bit ~nig_registers_
+   led_control_traffic_p0.led_control_traffic_p0 and bit
+   ~nig_registers_led_control_blink_traffic_p0.led_control_blink_traffic_p0 */
+#define NIG_REG_LED_CONTROL_OVERRIDE_TRAFFIC_P0 		 0x102f8
+/* [RW 1] Port0: If set along with the led_control_override_trafic_p0 bit;
+   turns on the Traffic LED. If the led_control_blink_traffic_p0 bit is also
+   set; the LED will blink with blink rate specified in
+   ~nig_registers_led_control_blink_rate_p0.led_control_blink_rate_p0 and
+   ~nig_regsters_led_control_blink_rate_ena_p0.led_control_blink_rate_ena_p0
+   fields. */
+#define NIG_REG_LED_CONTROL_TRAFFIC_P0				 0x10300
+/* [RW 4] led mode for port0: 0 MAC; 1-3 PHY1; 4 MAC2; 5-7 PHY4; 8-MAC3;
+   9-11PHY7; 12 MAC4; 13-15 PHY10; */
+#define NIG_REG_LED_MODE_P0					 0x102f0
+#define NIG_REG_LLH0_BRB1_DRV_MASK				 0x10244
+/* [RW 1] send to BRB1 if no match on any of RMP rules. */
+#define NIG_REG_LLH0_BRB1_NOT_MCP				 0x1025c
+/* [RW 32] cm header for llh0 */
+#define NIG_REG_LLH0_CM_HEADER					 0x1007c
+#define NIG_REG_LLH0_ERROR_MASK 				 0x1008c
+/* [RW 8] event id for llh0 */
+#define NIG_REG_LLH0_EVENT_ID					 0x10084
+/* [RW 8] init credit counter for port0 in LLH */
+#define NIG_REG_LLH0_XCM_INIT_CREDIT				 0x10554
+#define NIG_REG_LLH0_XCM_MASK					 0x10130
+/* [RW 1] send to BRB1 if no match on any of RMP rules. */
+#define NIG_REG_LLH1_BRB1_NOT_MCP				 0x102dc
+/* [RW 32] cm header for llh1 */
+#define NIG_REG_LLH1_CM_HEADER					 0x10080
+#define NIG_REG_LLH1_ERROR_MASK 				 0x10090
+/* [RW 8] event id for llh1 */
+#define NIG_REG_LLH1_EVENT_ID					 0x10088
+/* [RW 8] init credit counter for port1 in LLH */
+#define NIG_REG_LLH1_XCM_INIT_CREDIT				 0x10564
+#define NIG_REG_LLH1_XCM_MASK					 0x10134
+#define NIG_REG_MASK_INTERRUPT_PORT0				 0x10330
+#define NIG_REG_MASK_INTERRUPT_PORT1				 0x10334
+/* [RW 1] Output signal from NIG to EMAC0. When set enables the EMAC0 block. */
+#define NIG_REG_NIG_EMAC0_EN					 0x1003c
+/* [RW 1] Output signal from NIG to TX_EMAC0. When set indicates to the
+   EMAC0 to strip the CRC from the ingress packets. */
+#define NIG_REG_NIG_INGRESS_EMAC0_NO_CRC			 0x10044
+/* [RW 1] Input enable for RX PBF LP IF */
+#define NIG_REG_PBF_LB_IN_EN					 0x100b4
+/* [RW 1] output enable for RX parser descriptor IF */
+#define NIG_REG_PRS_EOP_OUT_EN					 0x10104
+/* [RW 1] Input enable for RX parser request IF */
+#define NIG_REG_PRS_REQ_IN_EN					 0x100b8
+/* [RW 5] control to serdes - CL22 PHY_ADD and CL45 PRTAD */
+#define NIG_REG_SERDES0_CTRL_PHY_ADDR				 0x10374
+/* [R 1] status from serdes0 that inputs to interrupt logic of link status */
+#define NIG_REG_SERDES0_STATUS_LINK_STATUS			 0x10578
+/* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
+   for port0 */
+#define NIG_REG_STAT0_BRB_DISCARD				 0x105f0
+/* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
+   for port1 */
+#define NIG_REG_STAT1_BRB_DISCARD				 0x10628
+/* [WB_R 64] Rx statistics : User octets received for LP */
+#define NIG_REG_STAT2_BRB_OCTET 				 0x107e0
+#define NIG_REG_STATUS_INTERRUPT_PORT0				 0x10328
+#define NIG_REG_STATUS_INTERRUPT_PORT1				 0x1032c
+/* [RW 1] output enable for RX_XCM0 IF */
+#define NIG_REG_XCM0_OUT_EN					 0x100f0
+/* [RW 1] output enable for RX_XCM1 IF */
+#define NIG_REG_XCM1_OUT_EN					 0x100f4
+/* [RW 5] control to xgxs - CL45 DEVAD */
+#define NIG_REG_XGXS0_CTRL_MD_DEVAD				 0x1033c
+/* [RW 5] control to xgxs - CL22 PHY_ADD and CL45 PRTAD */
+#define NIG_REG_XGXS0_CTRL_PHY_ADDR				 0x10340
+/* [R 1] status from xgxs0 that inputs to interrupt logic of link10g. */
+#define NIG_REG_XGXS0_STATUS_LINK10G				 0x10680
+/* [R 4] status from xgxs0 that inputs to interrupt logic of link status */
+#define NIG_REG_XGXS0_STATUS_LINK_STATUS			 0x10684
+/* [RW 2] selection for XGXS lane of port 0 in NIG_MUX block */
+#define NIG_REG_XGXS_LANE_SEL_P0				 0x102e8
+/* [RW 1] selection for port0 for NIG_MUX block : 0 = SerDes; 1 = XGXS */
+#define NIG_REG_XGXS_SERDES0_MODE_SEL				 0x102e0
+#define NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_SERDES0_LINK_STATUS (0x1<<9)
+#define NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK10G	 (0x1<<15)
+#define NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK_STATUS  (0xf<<18)
+#define NIG_STATUS_INTERRUPT_PORT0_REG_STATUS_XGXS0_LINK_STATUS_SIZE 18
+/* [RW 1] Disable processing further tasks from port 0 (after ending the
+   current task in process). */
+#define PBF_REG_DISABLE_NEW_TASK_PROC_P0			 0x14005c
+/* [RW 1] Disable processing further tasks from port 1 (after ending the
+   current task in process). */
+#define PBF_REG_DISABLE_NEW_TASK_PROC_P1			 0x140060
+/* [RW 1] Disable processing further tasks from port 4 (after ending the
+   current task in process). */
+#define PBF_REG_DISABLE_NEW_TASK_PROC_P4			 0x14006c
+#define PBF_REG_IF_ENABLE_REG					 0x140044
+/* [RW 1] Init bit. When set the initial credits are copied to the credit
+   registers (except the port credits). Should be set and then reset after
+   the configuration of the block has ended. */
+#define PBF_REG_INIT						 0x140000
+/* [RW 1] Init bit for port 0. When set the initial credit of port 0 is
+   copied to the credit register. Should be set and then reset after the
+   configuration of the port has ended. */
+#define PBF_REG_INIT_P0 					 0x140004
+/* [RW 1] Init bit for port 1. When set the initial credit of port 1 is
+   copied to the credit register. Should be set and then reset after the
+   configuration of the port has ended. */
+#define PBF_REG_INIT_P1 					 0x140008
+/* [RW 1] Init bit for port 4. When set the initial credit of port 4 is
+   copied to the credit register. Should be set and then reset after the
+   configuration of the port has ended. */
+#define PBF_REG_INIT_P4 					 0x14000c
+/* [RW 1] Enable for mac interface 0. */
+#define PBF_REG_MAC_IF0_ENABLE					 0x140030
+/* [RW 1] Enable for mac interface 1. */
+#define PBF_REG_MAC_IF1_ENABLE					 0x140034
+/* [RW 1] Enable for the loopback interface. */
+#define PBF_REG_MAC_LB_ENABLE					 0x140040
+/* [RW 10] Port 0 threshold used by arbiter in 16 byte lines used when pause
+   not suppoterd. */
+#define PBF_REG_P0_ARB_THRSH					 0x1400e4
+/* [R 11] Current credit for port 0 in the tx port buffers in 16 byte lines. */
+#define PBF_REG_P0_CREDIT					 0x140200
+/* [RW 11] Initial credit for port 0 in the tx port buffers in 16 byte
+   lines. */
+#define PBF_REG_P0_INIT_CRD					 0x1400d0
+/* [RW 1] Indication that pause is enabled for port 0. */
+#define PBF_REG_P0_PAUSE_ENABLE 				 0x140014
+/* [R 8] Number of tasks in port 0 task queue. */
+#define PBF_REG_P0_TASK_CNT					 0x140204
+/* [R 11] Current credit for port 1 in the tx port buffers in 16 byte lines. */
+#define PBF_REG_P1_CREDIT					 0x140208
+/* [RW 11] Initial credit for port 1 in the tx port buffers in 16 byte
+   lines. */
+#define PBF_REG_P1_INIT_CRD					 0x1400d4
+/* [R 8] Number of tasks in port 1 task queue. */
+#define PBF_REG_P1_TASK_CNT					 0x14020c
+/* [R 11] Current credit for port 4 in the tx port buffers in 16 byte lines. */
+#define PBF_REG_P4_CREDIT					 0x140210
+/* [RW 11] Initial credit for port 4 in the tx port buffers in 16 byte
+   lines. */
+#define PBF_REG_P4_INIT_CRD					 0x1400e0
+/* [R 8] Number of tasks in port 4 task queue. */
+#define PBF_REG_P4_TASK_CNT					 0x140214
+/* [RW 5] Interrupt mask register #0 read/write */
+#define PBF_REG_PBF_INT_MASK					 0x1401d4
+/* [R 5] Interrupt register #0 read */
+#define PBF_REG_PBF_INT_STS					 0x1401c8
+#define PB_REG_CONTROL						 0
+/* [RW 2] Interrupt mask register #0 read/write */
+#define PB_REG_PB_INT_MASK					 0x28
+/* [R 2] Interrupt register #0 read */
+#define PB_REG_PB_INT_STS					 0x1c
+/* [RW 4] Parity mask register #0 read/write */
+#define PB_REG_PB_PRTY_MASK					 0x38
+#define PRS_REG_A_PRSU_20					 0x40134
+/* [R 8] debug only: CFC load request current credit. Transaction based. */
+#define PRS_REG_CFC_LD_CURRENT_CREDIT				 0x40164
+/* [R 8] debug only: CFC search request current credit. Transaction based. */
+#define PRS_REG_CFC_SEARCH_CURRENT_CREDIT			 0x40168
+/* [RW 6] The initial credit for the search message to the CFC interface.
+   Credit is transaction based. */
+#define PRS_REG_CFC_SEARCH_INITIAL_CREDIT			 0x4011c
+/* [RW 24] CID for port 0 if no match */
+#define PRS_REG_CID_PORT_0					 0x400fc
+#define PRS_REG_CID_PORT_1					 0x40100
+/* [RW 32] The CM header for flush message where 'load existed' bit in CFC
+   load response is reset and packet type is 0. Used in packet start message
+   to TCM. */
+#define PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_0			 0x400dc
+#define PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_1			 0x400e0
+#define PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_2			 0x400e4
+#define PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_3			 0x400e8
+#define PRS_REG_CM_HDR_FLUSH_LOAD_TYPE_4			 0x400ec
+/* [RW 32] The CM header for flush message where 'load existed' bit in CFC
+   load response is set and packet type is 0. Used in packet start message
+   to TCM. */
+#define PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_0			 0x400bc
+#define PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_1			 0x400c0
+#define PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_2			 0x400c4
+#define PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_3			 0x400c8
+#define PRS_REG_CM_HDR_FLUSH_NO_LOAD_TYPE_4			 0x400cc
+/* [RW 32] The CM header for a match and packet type 1 for loopback port.
+   Used in packet start message to TCM. */
+#define PRS_REG_CM_HDR_LOOPBACK_TYPE_1				 0x4009c
+#define PRS_REG_CM_HDR_LOOPBACK_TYPE_2				 0x400a0
+#define PRS_REG_CM_HDR_LOOPBACK_TYPE_3				 0x400a4
+#define PRS_REG_CM_HDR_LOOPBACK_TYPE_4				 0x400a8
+/* [RW 32] The CM header for a match and packet type 0. Used in packet start
+   message to TCM. */
+#define PRS_REG_CM_HDR_TYPE_0					 0x40078
+#define PRS_REG_CM_HDR_TYPE_1					 0x4007c
+#define PRS_REG_CM_HDR_TYPE_2					 0x40080
+#define PRS_REG_CM_HDR_TYPE_3					 0x40084
+#define PRS_REG_CM_HDR_TYPE_4					 0x40088
+/* [RW 32] The CM header in case there was not a match on the connection */
+#define PRS_REG_CM_NO_MATCH_HDR 				 0x400b8
+/* [RW 8] The 8-bit event ID for a match and packet type 1. Used in packet
+   start message to TCM. */
+#define PRS_REG_EVENT_ID_1					 0x40054
+#define PRS_REG_EVENT_ID_2					 0x40058
+#define PRS_REG_EVENT_ID_3					 0x4005c
+/* [RW 8] Context region for flush packet with packet type 0. Used in CFC
+   load request message. */
+#define PRS_REG_FLUSH_REGIONS_TYPE_0				 0x40004
+#define PRS_REG_FLUSH_REGIONS_TYPE_1				 0x40008
+#define PRS_REG_FLUSH_REGIONS_TYPE_2				 0x4000c
+#define PRS_REG_FLUSH_REGIONS_TYPE_3				 0x40010
+#define PRS_REG_FLUSH_REGIONS_TYPE_4				 0x40014
+#define PRS_REG_FLUSH_REGIONS_TYPE_5				 0x40018
+#define PRS_REG_FLUSH_REGIONS_TYPE_6				 0x4001c
+#define PRS_REG_FLUSH_REGIONS_TYPE_7				 0x40020
+/* [RW 4] The increment value to send in the CFC load request message */
+#define PRS_REG_INC_VALUE					 0x40048
+/* [RW 1] If set indicates not to send messages to CFC on received packets */
+#define PRS_REG_NIC_MODE					 0x40138
+/* [RW 8] The 8-bit event ID for cases where there is no match on the
+   connection. Used in packet start message to TCM. */
+#define PRS_REG_NO_MATCH_EVENT_ID				 0x40070
+/* [ST 24] The number of input CFC flush packets */
+#define PRS_REG_NUM_OF_CFC_FLUSH_MESSAGES			 0x40128
+/* [ST 32] The number of cycles the Parser halted its operation since it
+   could not allocate the next serial number */
+#define PRS_REG_NUM_OF_DEAD_CYCLES				 0x40130
+/* [ST 24] The number of input packets */
+#define PRS_REG_NUM_OF_PACKETS					 0x40124
+/* [ST 24] The number of input transparent flush packets */
+#define PRS_REG_NUM_OF_TRANSPARENT_FLUSH_MESSAGES		 0x4012c
+/* [RW 8] Context region for received Ethernet packet with a match and
+   packet type 0. Used in CFC load request message */
+#define PRS_REG_PACKET_REGIONS_TYPE_0				 0x40028
+#define PRS_REG_PACKET_REGIONS_TYPE_1				 0x4002c
+#define PRS_REG_PACKET_REGIONS_TYPE_2				 0x40030
+#define PRS_REG_PACKET_REGIONS_TYPE_3				 0x40034
+#define PRS_REG_PACKET_REGIONS_TYPE_4				 0x40038
+#define PRS_REG_PACKET_REGIONS_TYPE_5				 0x4003c
+#define PRS_REG_PACKET_REGIONS_TYPE_6				 0x40040
+#define PRS_REG_PACKET_REGIONS_TYPE_7				 0x40044
+/* [R 2] debug only: Number of pending requests for CAC on port 0. */
+#define PRS_REG_PENDING_BRB_CAC0_RQ				 0x40174
+/* [R 2] debug only: Number of pending requests for header parsing. */
+#define PRS_REG_PENDING_BRB_PRS_RQ				 0x40170
+/* [R 1] Interrupt register #0 read */
+#define PRS_REG_PRS_INT_STS					 0x40188
+/* [RW 8] Parity mask register #0 read/write */
+#define PRS_REG_PRS_PRTY_MASK					 0x401a4
+/* [RW 8] Context region for pure acknowledge packets. Used in CFC load
+   request message */
+#define PRS_REG_PURE_REGIONS					 0x40024
+/* [R 32] debug only: Serial number status lsb 32 bits. '1' indicates this
+   serail number was released by SDM but cannot be used because a previous
+   serial number was not released. */
+#define PRS_REG_SERIAL_NUM_STATUS_LSB				 0x40154
+/* [R 32] debug only: Serial number status msb 32 bits. '1' indicates this
+   serail number was released by SDM but cannot be used because a previous
+   serial number was not released. */
+#define PRS_REG_SERIAL_NUM_STATUS_MSB				 0x40158
+/* [R 4] debug only: SRC current credit. Transaction based. */
+#define PRS_REG_SRC_CURRENT_CREDIT				 0x4016c
+/* [R 8] debug only: TCM current credit. Cycle based. */
+#define PRS_REG_TCM_CURRENT_CREDIT				 0x40160
+/* [R 8] debug only: TSDM current credit. Transaction based. */
+#define PRS_REG_TSDM_CURRENT_CREDIT				 0x4015c
+/* [R 6] Debug only: Number of used entries in the data FIFO */
+#define PXP2_REG_HST_DATA_FIFO_STATUS				 0x12047c
+/* [R 7] Debug only: Number of used entries in the header FIFO */
+#define PXP2_REG_HST_HEADER_FIFO_STATUS 			 0x120478
+#define PXP2_REG_PGL_CONTROL0					 0x120490
+#define PXP2_REG_PGL_CONTROL1					 0x120514
+/* [RW 32] Inbound interrupt table for CSDM: bits[31:16]-mask;
+   its[15:0]-address */
+#define PXP2_REG_PGL_INT_CSDM_0 				 0x1204f4
+#define PXP2_REG_PGL_INT_CSDM_1 				 0x1204f8
+#define PXP2_REG_PGL_INT_CSDM_2 				 0x1204fc
+#define PXP2_REG_PGL_INT_CSDM_3 				 0x120500
+#define PXP2_REG_PGL_INT_CSDM_4 				 0x120504
+#define PXP2_REG_PGL_INT_CSDM_5 				 0x120508
+#define PXP2_REG_PGL_INT_CSDM_6 				 0x12050c
+#define PXP2_REG_PGL_INT_CSDM_7 				 0x120510
+/* [RW 32] Inbound interrupt table for TSDM: bits[31:16]-mask;
+   its[15:0]-address */
+#define PXP2_REG_PGL_INT_TSDM_0 				 0x120494
+#define PXP2_REG_PGL_INT_TSDM_1 				 0x120498
+#define PXP2_REG_PGL_INT_TSDM_2 				 0x12049c
+#define PXP2_REG_PGL_INT_TSDM_3 				 0x1204a0
+#define PXP2_REG_PGL_INT_TSDM_4 				 0x1204a4
+#define PXP2_REG_PGL_INT_TSDM_5 				 0x1204a8
+#define PXP2_REG_PGL_INT_TSDM_6 				 0x1204ac
+#define PXP2_REG_PGL_INT_TSDM_7 				 0x1204b0
+/* [RW 32] Inbound interrupt table for USDM: bits[31:16]-mask;
+   its[15:0]-address */
+#define PXP2_REG_PGL_INT_USDM_0 				 0x1204b4
+#define PXP2_REG_PGL_INT_USDM_1 				 0x1204b8
+#define PXP2_REG_PGL_INT_USDM_2 				 0x1204bc
+#define PXP2_REG_PGL_INT_USDM_3 				 0x1204c0
+#define PXP2_REG_PGL_INT_USDM_4 				 0x1204c4
+#define PXP2_REG_PGL_INT_USDM_5 				 0x1204c8
+#define PXP2_REG_PGL_INT_USDM_6 				 0x1204cc
+#define PXP2_REG_PGL_INT_USDM_7 				 0x1204d0
+/* [RW 32] Inbound interrupt table for XSDM: bits[31:16]-mask;
+   its[15:0]-address */
+#define PXP2_REG_PGL_INT_XSDM_0 				 0x1204d4
+#define PXP2_REG_PGL_INT_XSDM_1 				 0x1204d8
+#define PXP2_REG_PGL_INT_XSDM_2 				 0x1204dc
+#define PXP2_REG_PGL_INT_XSDM_3 				 0x1204e0
+#define PXP2_REG_PGL_INT_XSDM_4 				 0x1204e4
+#define PXP2_REG_PGL_INT_XSDM_5 				 0x1204e8
+#define PXP2_REG_PGL_INT_XSDM_6 				 0x1204ec
+#define PXP2_REG_PGL_INT_XSDM_7 				 0x1204f0
+/* [R 1] this bit indicates that a read request was blocked because of
+   bus_master_en was deasserted */
+#define PXP2_REG_PGL_READ_BLOCKED				 0x120568
+/* [R 6] debug only */
+#define PXP2_REG_PGL_TXR_CDTS					 0x120528
+/* [R 18] debug only */
+#define PXP2_REG_PGL_TXW_CDTS					 0x12052c
+/* [R 1] this bit indicates that a write request was blocked because of
+   bus_master_en was deasserted */
+#define PXP2_REG_PGL_WRITE_BLOCKED				 0x120564
+#define PXP2_REG_PSWRQ_BW_ADD1					 0x1201c0
+#define PXP2_REG_PSWRQ_BW_ADD10 				 0x1201e4
+#define PXP2_REG_PSWRQ_BW_ADD11 				 0x1201e8
+#define PXP2_REG_PSWRQ_BW_ADD10 				 0x1201e4
+#define PXP2_REG_PSWRQ_BW_ADD11 				 0x1201e8
+#define PXP2_REG_PSWRQ_BW_ADD2					 0x1201c4
+#define PXP2_REG_PSWRQ_BW_ADD28 				 0x120228
+#define PXP2_REG_PSWRQ_BW_ADD28 				 0x120228
+#define PXP2_REG_PSWRQ_BW_ADD3					 0x1201c8
+#define PXP2_REG_PSWRQ_BW_ADD6					 0x1201d4
+#define PXP2_REG_PSWRQ_BW_ADD7					 0x1201d8
+#define PXP2_REG_PSWRQ_BW_ADD8					 0x1201dc
+#define PXP2_REG_PSWRQ_BW_ADD9					 0x1201e0
+#define PXP2_REG_PSWRQ_BW_CREDIT				 0x12032c
+#define PXP2_REG_PSWRQ_BW_L1					 0x1202b0
+#define PXP2_REG_PSWRQ_BW_L10					 0x1202d4
+#define PXP2_REG_PSWRQ_BW_L11					 0x1202d8
+#define PXP2_REG_PSWRQ_BW_L10					 0x1202d4
+#define PXP2_REG_PSWRQ_BW_L11					 0x1202d8
+#define PXP2_REG_PSWRQ_BW_L2					 0x1202b4
+#define PXP2_REG_PSWRQ_BW_L28					 0x120318
+#define PXP2_REG_PSWRQ_BW_L28					 0x120318
+#define PXP2_REG_PSWRQ_BW_L3					 0x1202b8
+#define PXP2_REG_PSWRQ_BW_L6					 0x1202c4
+#define PXP2_REG_PSWRQ_BW_L7					 0x1202c8
+#define PXP2_REG_PSWRQ_BW_L8					 0x1202cc
+#define PXP2_REG_PSWRQ_BW_L9					 0x1202d0
+#define PXP2_REG_PSWRQ_BW_RD					 0x120324
+#define PXP2_REG_PSWRQ_BW_UB1					 0x120238
+#define PXP2_REG_PSWRQ_BW_UB10					 0x12025c
+#define PXP2_REG_PSWRQ_BW_UB11					 0x120260
+#define PXP2_REG_PSWRQ_BW_UB10					 0x12025c
+#define PXP2_REG_PSWRQ_BW_UB11					 0x120260
+#define PXP2_REG_PSWRQ_BW_UB2					 0x12023c
+#define PXP2_REG_PSWRQ_BW_UB28					 0x1202a0
+#define PXP2_REG_PSWRQ_BW_UB28					 0x1202a0
+#define PXP2_REG_PSWRQ_BW_UB3					 0x120240
+#define PXP2_REG_PSWRQ_BW_UB6					 0x12024c
+#define PXP2_REG_PSWRQ_BW_UB7					 0x120250
+#define PXP2_REG_PSWRQ_BW_UB8					 0x120254
+#define PXP2_REG_PSWRQ_BW_UB9					 0x120258
+#define PXP2_REG_PSWRQ_BW_WR					 0x120328
+#define PXP2_REG_PSWRQ_CDU0_L2P 				 0x120000
+#define PXP2_REG_PSWRQ_QM0_L2P					 0x120038
+#define PXP2_REG_PSWRQ_SRC0_L2P 				 0x120054
+#define PXP2_REG_PSWRQ_TM0_L2P					 0x12001c
+/* [RW 25] Interrupt mask register #0 read/write */
+#define PXP2_REG_PXP2_INT_MASK					 0x120578
+/* [R 25] Interrupt register #0 read */
+#define PXP2_REG_PXP2_INT_STS					 0x12056c
+/* [RC 25] Interrupt register #0 read clear */
+#define PXP2_REG_PXP2_INT_STS_CLR				 0x120570
+/* [RW 32] Parity mask register #0 read/write */
+#define PXP2_REG_PXP2_PRTY_MASK_0				 0x120588
+#define PXP2_REG_PXP2_PRTY_MASK_1				 0x120598
+/* [R 1] Debug only: The 'almost full' indication from each fifo (gives
+   indication about backpressure) */
+#define PXP2_REG_RD_ALMOST_FULL_0				 0x120424
+/* [R 8] Debug only: The blocks counter - number of unused block ids */
+#define PXP2_REG_RD_BLK_CNT					 0x120418
+/* [RW 8] Debug only: Total number of available blocks in Tetris Buffer.
+   Must be bigger than 6. Normally should not be changed. */
+#define PXP2_REG_RD_BLK_NUM_CFG 				 0x12040c
+/* [RW 2] CDU byte swapping mode configuration for master read requests */
+#define PXP2_REG_RD_CDURD_SWAP_MODE				 0x120404
+/* [RW 1] When '1'; inputs to the PSWRD block are ignored */
+#define PXP2_REG_RD_DISABLE_INPUTS				 0x120374
+/* [R 1] PSWRD internal memories initialization is done */
+#define PXP2_REG_RD_INIT_DONE					 0x120370
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq10 */
+#define PXP2_REG_RD_MAX_BLKS_VQ10				 0x1203a0
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq11 */
+#define PXP2_REG_RD_MAX_BLKS_VQ11				 0x1203a4
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq17 */
+#define PXP2_REG_RD_MAX_BLKS_VQ17				 0x1203bc
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq18 */
+#define PXP2_REG_RD_MAX_BLKS_VQ18				 0x1203c0
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq19 */
+#define PXP2_REG_RD_MAX_BLKS_VQ19				 0x1203c4
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq22 */
+#define PXP2_REG_RD_MAX_BLKS_VQ22				 0x1203d0
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq6 */
+#define PXP2_REG_RD_MAX_BLKS_VQ6				 0x120390
+/* [RW 8] The maximum number of blocks in Tetris Buffer that can be
+   allocated for vq9 */
+#define PXP2_REG_RD_MAX_BLKS_VQ9				 0x12039c
+/* [RW 2] PBF byte swapping mode configuration for master read requests */
+#define PXP2_REG_RD_PBF_SWAP_MODE				 0x1203f4
+/* [R 1] Debug only: Indication if delivery ports are idle */
+#define PXP2_REG_RD_PORT_IS_IDLE_0				 0x12041c
+#define PXP2_REG_RD_PORT_IS_IDLE_1				 0x120420
+/* [RW 2] QM byte swapping mode configuration for master read requests */
+#define PXP2_REG_RD_QM_SWAP_MODE				 0x1203f8
+/* [R 7] Debug only: The SR counter - number of unused sub request ids */
+#define PXP2_REG_RD_SR_CNT					 0x120414
+/* [RW 2] SRC byte swapping mode configuration for master read requests */
+#define PXP2_REG_RD_SRC_SWAP_MODE				 0x120400
+/* [RW 7] Debug only: Total number of available PCI read sub-requests. Must
+   be bigger than 1. Normally should not be changed. */
+#define PXP2_REG_RD_SR_NUM_CFG					 0x120408
+/* [RW 1] Signals the PSWRD block to start initializing internal memories */
+#define PXP2_REG_RD_START_INIT					 0x12036c
+/* [RW 2] TM byte swapping mode configuration for master read requests */
+#define PXP2_REG_RD_TM_SWAP_MODE				 0x1203fc
+/* [RW 10] Bandwidth addition to VQ0 write requests */
+#define PXP2_REG_RQ_BW_RD_ADD0					 0x1201bc
+/* [RW 10] Bandwidth addition to VQ12 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD12 				 0x1201ec
+/* [RW 10] Bandwidth addition to VQ13 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD13 				 0x1201f0
+/* [RW 10] Bandwidth addition to VQ14 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD14 				 0x1201f4
+/* [RW 10] Bandwidth addition to VQ15 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD15 				 0x1201f8
+/* [RW 10] Bandwidth addition to VQ16 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD16 				 0x1201fc
+/* [RW 10] Bandwidth addition to VQ17 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD17 				 0x120200
+/* [RW 10] Bandwidth addition to VQ18 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD18 				 0x120204
+/* [RW 10] Bandwidth addition to VQ19 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD19 				 0x120208
+/* [RW 10] Bandwidth addition to VQ20 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD20 				 0x12020c
+/* [RW 10] Bandwidth addition to VQ22 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD22 				 0x120210
+/* [RW 10] Bandwidth addition to VQ23 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD23 				 0x120214
+/* [RW 10] Bandwidth addition to VQ24 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD24 				 0x120218
+/* [RW 10] Bandwidth addition to VQ25 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD25 				 0x12021c
+/* [RW 10] Bandwidth addition to VQ26 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD26 				 0x120220
+/* [RW 10] Bandwidth addition to VQ27 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD27 				 0x120224
+/* [RW 10] Bandwidth addition to VQ4 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD4					 0x1201cc
+/* [RW 10] Bandwidth addition to VQ5 read requests */
+#define PXP2_REG_RQ_BW_RD_ADD5					 0x1201d0
+/* [RW 10] Bandwidth Typical L for VQ0 Read requests */
+#define PXP2_REG_RQ_BW_RD_L0					 0x1202ac
+/* [RW 10] Bandwidth Typical L for VQ12 Read requests */
+#define PXP2_REG_RQ_BW_RD_L12					 0x1202dc
+/* [RW 10] Bandwidth Typical L for VQ13 Read requests */
+#define PXP2_REG_RQ_BW_RD_L13					 0x1202e0
+/* [RW 10] Bandwidth Typical L for VQ14 Read requests */
+#define PXP2_REG_RQ_BW_RD_L14					 0x1202e4
+/* [RW 10] Bandwidth Typical L for VQ15 Read requests */
+#define PXP2_REG_RQ_BW_RD_L15					 0x1202e8
+/* [RW 10] Bandwidth Typical L for VQ16 Read requests */
+#define PXP2_REG_RQ_BW_RD_L16					 0x1202ec
+/* [RW 10] Bandwidth Typical L for VQ17 Read requests */
+#define PXP2_REG_RQ_BW_RD_L17					 0x1202f0
+/* [RW 10] Bandwidth Typical L for VQ18 Read requests */
+#define PXP2_REG_RQ_BW_RD_L18					 0x1202f4
+/* [RW 10] Bandwidth Typical L for VQ19 Read requests */
+#define PXP2_REG_RQ_BW_RD_L19					 0x1202f8
+/* [RW 10] Bandwidth Typical L for VQ20 Read requests */
+#define PXP2_REG_RQ_BW_RD_L20					 0x1202fc
+/* [RW 10] Bandwidth Typical L for VQ22 Read requests */
+#define PXP2_REG_RQ_BW_RD_L22					 0x120300
+/* [RW 10] Bandwidth Typical L for VQ23 Read requests */
+#define PXP2_REG_RQ_BW_RD_L23					 0x120304
+/* [RW 10] Bandwidth Typical L for VQ24 Read requests */
+#define PXP2_REG_RQ_BW_RD_L24					 0x120308
+/* [RW 10] Bandwidth Typical L for VQ25 Read requests */
+#define PXP2_REG_RQ_BW_RD_L25					 0x12030c
+/* [RW 10] Bandwidth Typical L for VQ26 Read requests */
+#define PXP2_REG_RQ_BW_RD_L26					 0x120310
+/* [RW 10] Bandwidth Typical L for VQ27 Read requests */
+#define PXP2_REG_RQ_BW_RD_L27					 0x120314
+/* [RW 10] Bandwidth Typical L for VQ4 Read requests */
+#define PXP2_REG_RQ_BW_RD_L4					 0x1202bc
+/* [RW 10] Bandwidth Typical L for VQ5 Read- currently not used */
+#define PXP2_REG_RQ_BW_RD_L5					 0x1202c0
+/* [RW 7] Bandwidth upper bound for VQ0 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND0				 0x120234
+/* [RW 7] Bandwidth upper bound for VQ12 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND12				 0x120264
+/* [RW 7] Bandwidth upper bound for VQ13 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND13				 0x120268
+/* [RW 7] Bandwidth upper bound for VQ14 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND14				 0x12026c
+/* [RW 7] Bandwidth upper bound for VQ15 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND15				 0x120270
+/* [RW 7] Bandwidth upper bound for VQ16 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND16				 0x120274
+/* [RW 7] Bandwidth upper bound for VQ17 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND17				 0x120278
+/* [RW 7] Bandwidth upper bound for VQ18 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND18				 0x12027c
+/* [RW 7] Bandwidth upper bound for VQ19 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND19				 0x120280
+/* [RW 7] Bandwidth upper bound for VQ20 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND20				 0x120284
+/* [RW 7] Bandwidth upper bound for VQ22 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND22				 0x120288
+/* [RW 7] Bandwidth upper bound for VQ23 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND23				 0x12028c
+/* [RW 7] Bandwidth upper bound for VQ24 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND24				 0x120290
+/* [RW 7] Bandwidth upper bound for VQ25 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND25				 0x120294
+/* [RW 7] Bandwidth upper bound for VQ26 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND26				 0x120298
+/* [RW 7] Bandwidth upper bound for VQ27 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND27				 0x12029c
+/* [RW 7] Bandwidth upper bound for VQ4 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND4				 0x120244
+/* [RW 7] Bandwidth upper bound for VQ5 read requests */
+#define PXP2_REG_RQ_BW_RD_UBOUND5				 0x120248
+/* [RW 10] Bandwidth addition to VQ29 write requests */
+#define PXP2_REG_RQ_BW_WR_ADD29 				 0x12022c
+/* [RW 10] Bandwidth addition to VQ30 write requests */
+#define PXP2_REG_RQ_BW_WR_ADD30 				 0x120230
+/* [RW 10] Bandwidth Typical L for VQ29 Write requests */
+#define PXP2_REG_RQ_BW_WR_L29					 0x12031c
+/* [RW 10] Bandwidth Typical L for VQ30 Write requests */
+#define PXP2_REG_RQ_BW_WR_L30					 0x120320
+/* [RW 7] Bandwidth upper bound for VQ29 */
+#define PXP2_REG_RQ_BW_WR_UBOUND29				 0x1202a4
+/* [RW 7] Bandwidth upper bound for VQ30 */
+#define PXP2_REG_RQ_BW_WR_UBOUND30				 0x1202a8
+/* [RW 2] Endian mode for cdu */
+#define PXP2_REG_RQ_CDU_ENDIAN_M				 0x1201a0
+/* [RW 3] page size in L2P table for CDU module; -4k; -8k; -16k; -32k; -64k;
+   -128k */
+#define PXP2_REG_RQ_CDU_P_SIZE					 0x120018
+/* [R 1] 1' indicates that the requester has finished its internal
+   configuration */
+#define PXP2_REG_RQ_CFG_DONE					 0x1201b4
+/* [RW 2] Endian mode for debug */
+#define PXP2_REG_RQ_DBG_ENDIAN_M				 0x1201a4
+/* [RW 1] When '1'; requests will enter input buffers but wont get out
+   towards the glue */
+#define PXP2_REG_RQ_DISABLE_INPUTS				 0x120330
+/* [RW 2] Endian mode for hc */
+#define PXP2_REG_RQ_HC_ENDIAN_M 				 0x1201a8
+/* [WB 53] Onchip address table */
+#define PXP2_REG_RQ_ONCHIP_AT					 0x122000
+/* [RW 2] Endian mode for qm */
+#define PXP2_REG_RQ_QM_ENDIAN_M 				 0x120194
+/* [RW 3] page size in L2P table for QM module; -4k; -8k; -16k; -32k; -64k;
+   -128k */
+#define PXP2_REG_RQ_QM_P_SIZE					 0x120050
+/* [RW 1] 1' indicates that the RBC has finished configurating the PSWRQ */
+#define PXP2_REG_RQ_RBC_DONE					 0x1201b0
+/* [RW 3] Max burst size filed for read requests port 0; 000 - 128B;
+   001:256B; 010: 512B; 11:1K:100:2K; 01:4K */
+#define PXP2_REG_RQ_RD_MBS0					 0x120160
+/* [RW 2] Endian mode for src */
+#define PXP2_REG_RQ_SRC_ENDIAN_M				 0x12019c
+/* [RW 3] page size in L2P table for SRC module; -4k; -8k; -16k; -32k; -64k;
+   -128k */
+#define PXP2_REG_RQ_SRC_P_SIZE					 0x12006c
+/* [RW 2] Endian mode for tm */
+#define PXP2_REG_RQ_TM_ENDIAN_M 				 0x120198
+/* [RW 3] page size in L2P table for TM module; -4k; -8k; -16k; -32k; -64k;
+   -128k */
+#define PXP2_REG_RQ_TM_P_SIZE					 0x120034
+/* [R 5] Number of entries in the ufifo; his fifo has l2p completions */
+#define PXP2_REG_RQ_UFIFO_NUM_OF_ENTRY				 0x12080c
+/* [R 8] Number of entries occupied by vq 0 in pswrq memory */
+#define PXP2_REG_RQ_VQ0_ENTRY_CNT				 0x120810
+/* [R 8] Number of entries occupied by vq 10 in pswrq memory */
+#define PXP2_REG_RQ_VQ10_ENTRY_CNT				 0x120818
+/* [R 8] Number of entries occupied by vq 11 in pswrq memory */
+#define PXP2_REG_RQ_VQ11_ENTRY_CNT				 0x120820
+/* [R 8] Number of entries occupied by vq 12 in pswrq memory */
+#define PXP2_REG_RQ_VQ12_ENTRY_CNT				 0x120828
+/* [R 8] Number of entries occupied by vq 13 in pswrq memory */
+#define PXP2_REG_RQ_VQ13_ENTRY_CNT				 0x120830
+/* [R 8] Number of entries occupied by vq 14 in pswrq memory */
+#define PXP2_REG_RQ_VQ14_ENTRY_CNT				 0x120838
+/* [R 8] Number of entries occupied by vq 15 in pswrq memory */
+#define PXP2_REG_RQ_VQ15_ENTRY_CNT				 0x120840
+/* [R 8] Number of entries occupied by vq 16 in pswrq memory */
+#define PXP2_REG_RQ_VQ16_ENTRY_CNT				 0x120848
+/* [R 8] Number of entries occupied by vq 17 in pswrq memory */
+#define PXP2_REG_RQ_VQ17_ENTRY_CNT				 0x120850
+/* [R 8] Number of entries occupied by vq 18 in pswrq memory */
+#define PXP2_REG_RQ_VQ18_ENTRY_CNT				 0x120858
+/* [R 8] Number of entries occupied by vq 19 in pswrq memory */
+#define PXP2_REG_RQ_VQ19_ENTRY_CNT				 0x120860
+/* [R 8] Number of entries occupied by vq 1 in pswrq memory */
+#define PXP2_REG_RQ_VQ1_ENTRY_CNT				 0x120868
+/* [R 8] Number of entries occupied by vq 20 in pswrq memory */
+#define PXP2_REG_RQ_VQ20_ENTRY_CNT				 0x120870
+/* [R 8] Number of entries occupied by vq 21 in pswrq memory */
+#define PXP2_REG_RQ_VQ21_ENTRY_CNT				 0x120878
+/* [R 8] Number of entries occupied by vq 22 in pswrq memory */
+#define PXP2_REG_RQ_VQ22_ENTRY_CNT				 0x120880
+/* [R 8] Number of entries occupied by vq 23 in pswrq memory */
+#define PXP2_REG_RQ_VQ23_ENTRY_CNT				 0x120888
+/* [R 8] Number of entries occupied by vq 24 in pswrq memory */
+#define PXP2_REG_RQ_VQ24_ENTRY_CNT				 0x120890
+/* [R 8] Number of entries occupied by vq 25 in pswrq memory */
+#define PXP2_REG_RQ_VQ25_ENTRY_CNT				 0x120898
+/* [R 8] Number of entries occupied by vq 26 in pswrq memory */
+#define PXP2_REG_RQ_VQ26_ENTRY_CNT				 0x1208a0
+/* [R 8] Number of entries occupied by vq 27 in pswrq memory */
+#define PXP2_REG_RQ_VQ27_ENTRY_CNT				 0x1208a8
+/* [R 8] Number of entries occupied by vq 28 in pswrq memory */
+#define PXP2_REG_RQ_VQ28_ENTRY_CNT				 0x1208b0
+/* [R 8] Number of entries occupied by vq 29 in pswrq memory */
+#define PXP2_REG_RQ_VQ29_ENTRY_CNT				 0x1208b8
+/* [R 8] Number of entries occupied by vq 2 in pswrq memory */
+#define PXP2_REG_RQ_VQ2_ENTRY_CNT				 0x1208c0
+/* [R 8] Number of entries occupied by vq 30 in pswrq memory */
+#define PXP2_REG_RQ_VQ30_ENTRY_CNT				 0x1208c8
+/* [R 8] Number of entries occupied by vq 31 in pswrq memory */
+#define PXP2_REG_RQ_VQ31_ENTRY_CNT				 0x1208d0
+/* [R 8] Number of entries occupied by vq 3 in pswrq memory */
+#define PXP2_REG_RQ_VQ3_ENTRY_CNT				 0x1208d8
+/* [R 8] Number of entries occupied by vq 4 in pswrq memory */
+#define PXP2_REG_RQ_VQ4_ENTRY_CNT				 0x1208e0
+/* [R 8] Number of entries occupied by vq 5 in pswrq memory */
+#define PXP2_REG_RQ_VQ5_ENTRY_CNT				 0x1208e8
+/* [R 8] Number of entries occupied by vq 6 in pswrq memory */
+#define PXP2_REG_RQ_VQ6_ENTRY_CNT				 0x1208f0
+/* [R 8] Number of entries occupied by vq 7 in pswrq memory */
+#define PXP2_REG_RQ_VQ7_ENTRY_CNT				 0x1208f8
+/* [R 8] Number of entries occupied by vq 8 in pswrq memory */
+#define PXP2_REG_RQ_VQ8_ENTRY_CNT				 0x120900
+/* [R 8] Number of entries occupied by vq 9 in pswrq memory */
+#define PXP2_REG_RQ_VQ9_ENTRY_CNT				 0x120908
+/* [RW 3] Max burst size filed for write requests port 0; 000 - 128B;
+   001:256B; 010: 512B; */
+#define PXP2_REG_RQ_WR_MBS0					 0x12015c
+/* [RW 10] if Number of entries in dmae fifo will be higer than this
+   threshold then has_payload indication will be asserted; the default value
+   should be equal to &gt;  write MBS size! */
+#define PXP2_REG_WR_DMAE_TH					 0x120368
+/* [R 1] debug only: Indication if PSWHST arbiter is idle */
+#define PXP_REG_HST_ARB_IS_IDLE 				 0x103004
+/* [R 8] debug only: A bit mask for all PSWHST arbiter clients. '1' means
+   this client is waiting for the arbiter. */
+#define PXP_REG_HST_CLIENTS_WAITING_TO_ARB			 0x103008
+/* [WB 160] Used for initialization of the inbound interrupts memory */
+#define PXP_REG_HST_INBOUND_INT 				 0x103800
+/* [RW 32] Interrupt mask register #0 read/write */
+#define PXP_REG_PXP_INT_MASK_0					 0x103074
+#define PXP_REG_PXP_INT_MASK_1					 0x103084
+/* [R 32] Interrupt register #0 read */
+#define PXP_REG_PXP_INT_STS_0					 0x103068
+#define PXP_REG_PXP_INT_STS_1					 0x103078
+/* [RC 32] Interrupt register #0 read clear */
+#define PXP_REG_PXP_INT_STS_CLR_0				 0x10306c
+/* [RW 26] Parity mask register #0 read/write */
+#define PXP_REG_PXP_PRTY_MASK					 0x103094
+/* [RW 4] The activity counter initial increment value sent in the load
+   request */
+#define QM_REG_ACTCTRINITVAL_0					 0x168040
+#define QM_REG_ACTCTRINITVAL_1					 0x168044
+#define QM_REG_ACTCTRINITVAL_2					 0x168048
+#define QM_REG_ACTCTRINITVAL_3					 0x16804c
+/* [RW 32] The base logical address (in bytes) of each physical queue. The
+   index I represents the physical queue number. The 12 lsbs are ignore and
+   considered zero so practically there are only 20 bits in this register. */
+#define QM_REG_BASEADDR 					 0x168900
+/* [RW 16] The byte credit cost for each task. This value is for both ports */
+#define QM_REG_BYTECRDCOST					 0x168234
+/* [RW 16] The initial byte credit value for both ports. */
+#define QM_REG_BYTECRDINITVAL					 0x168238
+/* [RW 32] A bit per physical queue. If the bit is cleared then the physical
+   queue uses port 0 else it uses port 1. */
+#define QM_REG_BYTECRDPORT_LSB					 0x168228
+/* [RW 32] A bit per physical queue. If the bit is cleared then the physical
+   queue uses port 0 else it uses port 1. */
+#define QM_REG_BYTECRDPORT_MSB					 0x168224
+/* [RW 16] The byte credit value that if above the QM is considered almost
+   full */
+#define QM_REG_BYTECREDITAFULLTHR				 0x168094
+/* [RW 4] The initial credit for interface */
+#define QM_REG_CMINITCRD_0					 0x1680cc
+#define QM_REG_CMINITCRD_1					 0x1680d0
+#define QM_REG_CMINITCRD_2					 0x1680d4
+#define QM_REG_CMINITCRD_3					 0x1680d8
+#define QM_REG_CMINITCRD_4					 0x1680dc
+#define QM_REG_CMINITCRD_5					 0x1680e0
+#define QM_REG_CMINITCRD_6					 0x1680e4
+#define QM_REG_CMINITCRD_7					 0x1680e8
+/* [RW 8] A mask bit per CM interface. If this bit is 0 then this interface
+   is masked */
+#define QM_REG_CMINTEN						 0x1680ec
+/* [RW 12] A bit vector which indicates which one of the queues are tied to
+   interface 0 */
+#define QM_REG_CMINTVOQMASK_0					 0x1681f4
+#define QM_REG_CMINTVOQMASK_1					 0x1681f8
+#define QM_REG_CMINTVOQMASK_2					 0x1681fc
+#define QM_REG_CMINTVOQMASK_3					 0x168200
+#define QM_REG_CMINTVOQMASK_4					 0x168204
+#define QM_REG_CMINTVOQMASK_5					 0x168208
+#define QM_REG_CMINTVOQMASK_6					 0x16820c
+#define QM_REG_CMINTVOQMASK_7					 0x168210
+/* [RW 20] The number of connections divided by 16 which dictates the size
+   of each queue per port 0 */
+#define QM_REG_CONNNUM_0					 0x168020
+/* [R 6] Keep the fill level of the fifo from write client 4 */
+#define QM_REG_CQM_WRC_FIFOLVL					 0x168018
+/* [RW 8] The context regions sent in the CFC load request */
+#define QM_REG_CTXREG_0 					 0x168030
+#define QM_REG_CTXREG_1 					 0x168034
+#define QM_REG_CTXREG_2 					 0x168038
+#define QM_REG_CTXREG_3 					 0x16803c
+/* [RW 12] The VOQ mask used to select the VOQs which needs to be full for
+   bypass enable */
+#define QM_REG_ENBYPVOQMASK					 0x16823c
+/* [RW 32] A bit mask per each physical queue. If a bit is set then the
+   physical queue uses the byte credit */
+#define QM_REG_ENBYTECRD_LSB					 0x168220
+/* [RW 32] A bit mask per each physical queue. If a bit is set then the
+   physical queue uses the byte credit */
+#define QM_REG_ENBYTECRD_MSB					 0x16821c
+/* [RW 4] If cleared then the secondary interface will not be served by the
+   RR arbiter */
+#define QM_REG_ENSEC						 0x1680f0
+/* [RW 32] A bit vector per each physical queue which selects which function
+   number to use on PCI access for that queue. */
+#define QM_REG_FUNCNUMSEL_LSB					 0x168230
+/* [RW 32] A bit vector per each physical queue which selects which function
+   number to use on PCI access for that queue. */
+#define QM_REG_FUNCNUMSEL_MSB					 0x16822c
+/* [RW 32] A mask register to mask the Almost empty signals which will not
+   be use for the almost empty indication to the HW block */
+#define QM_REG_HWAEMPTYMASK_LSB 				 0x168218
+/* [RW 32] A mask register to mask the Almost empty signals which will not
+   be use for the almost empty indication to the HW block */
+#define QM_REG_HWAEMPTYMASK_MSB 				 0x168214
+/* [RW 4] The number of outstanding request to CFC */
+#define QM_REG_OUTLDREQ 					 0x168804
+/* [RC 1] A flag to indicate that overflow error occurred in one of the
+   queues. */
+#define QM_REG_OVFERROR 					 0x16805c
+/* [RC 6] the Q were the qverflow occurs */
+#define QM_REG_OVFQNUM						 0x168058
+/* [R 32] Pause state for physical queues 31-0 */
+#define QM_REG_PAUSESTATE0					 0x168410
+/* [R 32] Pause state for physical queues 64-32 */
+#define QM_REG_PAUSESTATE1					 0x168414
+/* [RW 2] The PCI attributes field used in the PCI request. */
+#define QM_REG_PCIREQAT 					 0x168054
+/* [R 16] The byte credit of port 0 */
+#define QM_REG_PORT0BYTECRD					 0x168300
+/* [R 16] The byte credit of port 1 */
+#define QM_REG_PORT1BYTECRD					 0x168304
+/* [WB 54] Pointer Table Memory; The mapping is as follow: ptrtbl[53:30]
+   read pointer; ptrtbl[29:6] write pointer; ptrtbl[5:4] read bank0;
+   ptrtbl[3:2] read bank 1; ptrtbl[1:0] write bank; */
+#define QM_REG_PTRTBL						 0x168a00
+/* [RW 2] Interrupt mask register #0 read/write */
+#define QM_REG_QM_INT_MASK					 0x168444
+/* [R 2] Interrupt register #0 read */
+#define QM_REG_QM_INT_STS					 0x168438
+/* [RW 9] Parity mask register #0 read/write */
+#define QM_REG_QM_PRTY_MASK					 0x168454
+/* [R 32] Current queues in pipeline: Queues from 32 to 63 */
+#define QM_REG_QSTATUS_HIGH					 0x16802c
+/* [R 32] Current queues in pipeline: Queues from 0 to 31 */
+#define QM_REG_QSTATUS_LOW					 0x168028
+/* [R 24] The number of tasks queued for each queue */
+#define QM_REG_QTASKCTR_0					 0x168308
+/* [RW 4] Queue tied to VOQ */
+#define QM_REG_QVOQIDX_0					 0x1680f4
+#define QM_REG_QVOQIDX_10					 0x16811c
+#define QM_REG_QVOQIDX_11					 0x168120
+#define QM_REG_QVOQIDX_12					 0x168124
+#define QM_REG_QVOQIDX_13					 0x168128
+#define QM_REG_QVOQIDX_14					 0x16812c
+#define QM_REG_QVOQIDX_15					 0x168130
+#define QM_REG_QVOQIDX_16					 0x168134
+#define QM_REG_QVOQIDX_17					 0x168138
+#define QM_REG_QVOQIDX_21					 0x168148
+#define QM_REG_QVOQIDX_25					 0x168158
+#define QM_REG_QVOQIDX_29					 0x168168
+#define QM_REG_QVOQIDX_32					 0x168174
+#define QM_REG_QVOQIDX_33					 0x168178
+#define QM_REG_QVOQIDX_34					 0x16817c
+#define QM_REG_QVOQIDX_35					 0x168180
+#define QM_REG_QVOQIDX_36					 0x168184
+#define QM_REG_QVOQIDX_37					 0x168188
+#define QM_REG_QVOQIDX_38					 0x16818c
+#define QM_REG_QVOQIDX_39					 0x168190
+#define QM_REG_QVOQIDX_40					 0x168194
+#define QM_REG_QVOQIDX_41					 0x168198
+#define QM_REG_QVOQIDX_42					 0x16819c
+#define QM_REG_QVOQIDX_43					 0x1681a0
+#define QM_REG_QVOQIDX_44					 0x1681a4
+#define QM_REG_QVOQIDX_45					 0x1681a8
+#define QM_REG_QVOQIDX_46					 0x1681ac
+#define QM_REG_QVOQIDX_47					 0x1681b0
+#define QM_REG_QVOQIDX_48					 0x1681b4
+#define QM_REG_QVOQIDX_49					 0x1681b8
+#define QM_REG_QVOQIDX_5					 0x168108
+#define QM_REG_QVOQIDX_50					 0x1681bc
+#define QM_REG_QVOQIDX_51					 0x1681c0
+#define QM_REG_QVOQIDX_52					 0x1681c4
+#define QM_REG_QVOQIDX_53					 0x1681c8
+#define QM_REG_QVOQIDX_54					 0x1681cc
+#define QM_REG_QVOQIDX_55					 0x1681d0
+#define QM_REG_QVOQIDX_56					 0x1681d4
+#define QM_REG_QVOQIDX_57					 0x1681d8
+#define QM_REG_QVOQIDX_58					 0x1681dc
+#define QM_REG_QVOQIDX_59					 0x1681e0
+#define QM_REG_QVOQIDX_50					 0x1681bc
+#define QM_REG_QVOQIDX_51					 0x1681c0
+#define QM_REG_QVOQIDX_52					 0x1681c4
+#define QM_REG_QVOQIDX_53					 0x1681c8
+#define QM_REG_QVOQIDX_54					 0x1681cc
+#define QM_REG_QVOQIDX_55					 0x1681d0
+#define QM_REG_QVOQIDX_56					 0x1681d4
+#define QM_REG_QVOQIDX_57					 0x1681d8
+#define QM_REG_QVOQIDX_58					 0x1681dc
+#define QM_REG_QVOQIDX_59					 0x1681e0
+#define QM_REG_QVOQIDX_6					 0x16810c
+#define QM_REG_QVOQIDX_60					 0x1681e4
+#define QM_REG_QVOQIDX_61					 0x1681e8
+#define QM_REG_QVOQIDX_62					 0x1681ec
+#define QM_REG_QVOQIDX_63					 0x1681f0
+#define QM_REG_QVOQIDX_60					 0x1681e4
+#define QM_REG_QVOQIDX_61					 0x1681e8
+#define QM_REG_QVOQIDX_62					 0x1681ec
+#define QM_REG_QVOQIDX_63					 0x1681f0
+#define QM_REG_QVOQIDX_7					 0x168110
+#define QM_REG_QVOQIDX_8					 0x168114
+#define QM_REG_QVOQIDX_9					 0x168118
+/* [R 24] Remaining pause timeout for port 0 */
+#define QM_REG_REMAINPAUSETM0					 0x168418
+/* [R 24] Remaining pause timeout for port 1 */
+#define QM_REG_REMAINPAUSETM1					 0x16841c
+/* [RW 1] Initialization bit command */
+#define QM_REG_SOFT_RESET					 0x168428
+/* [RW 8] The credit cost per every task in the QM. A value per each VOQ */
+#define QM_REG_TASKCRDCOST_0					 0x16809c
+#define QM_REG_TASKCRDCOST_1					 0x1680a0
+#define QM_REG_TASKCRDCOST_10					 0x1680c4
+#define QM_REG_TASKCRDCOST_11					 0x1680c8
+#define QM_REG_TASKCRDCOST_2					 0x1680a4
+#define QM_REG_TASKCRDCOST_4					 0x1680ac
+#define QM_REG_TASKCRDCOST_5					 0x1680b0
+/* [R 6] Keep the fill level of the fifo from write client 3 */
+#define QM_REG_TQM_WRC_FIFOLVL					 0x168010
+/* [R 6] Keep the fill level of the fifo from write client 2 */
+#define QM_REG_UQM_WRC_FIFOLVL					 0x168008
+/* [RC 32] Credit update error register */
+#define QM_REG_VOQCRDERRREG					 0x168408
+/* [R 16] The credit value for each VOQ */
+#define QM_REG_VOQCREDIT_0					 0x1682d0
+#define QM_REG_VOQCREDIT_1					 0x1682d4
+#define QM_REG_VOQCREDIT_10					 0x1682f8
+#define QM_REG_VOQCREDIT_11					 0x1682fc
+#define QM_REG_VOQCREDIT_4					 0x1682e0
+/* [RW 16] The credit value that if above the QM is considered almost full */
+#define QM_REG_VOQCREDITAFULLTHR				 0x168090
+/* [RW 16] The init and maximum credit for each VoQ */
+#define QM_REG_VOQINITCREDIT_0					 0x168060
+#define QM_REG_VOQINITCREDIT_1					 0x168064
+#define QM_REG_VOQINITCREDIT_10 				 0x168088
+#define QM_REG_VOQINITCREDIT_11 				 0x16808c
+#define QM_REG_VOQINITCREDIT_2					 0x168068
+#define QM_REG_VOQINITCREDIT_4					 0x168070
+#define QM_REG_VOQINITCREDIT_5					 0x168074
+/* [RW 1] The port of which VOQ belongs */
+#define QM_REG_VOQPORT_1					 0x1682a4
+#define QM_REG_VOQPORT_10					 0x1682c8
+#define QM_REG_VOQPORT_11					 0x1682cc
+#define QM_REG_VOQPORT_2					 0x1682a8
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_0_LSB					 0x168240
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_0_MSB					 0x168244
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_1_MSB					 0x16824c
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_2_LSB					 0x168250
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_2_MSB					 0x168254
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_3_LSB					 0x168258
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_4_LSB					 0x168260
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_4_MSB					 0x168264
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_5_LSB					 0x168268
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_5_MSB					 0x16826c
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_6_LSB					 0x168270
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_6_MSB					 0x168274
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_7_LSB					 0x168278
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_7_MSB					 0x16827c
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_8_LSB					 0x168280
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_8_MSB					 0x168284
+/* [RW 32] The physical queue number associated with each VOQ */
+#define QM_REG_VOQQMASK_9_LSB					 0x168288
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_0					 0x16880c
+#define QM_REG_WRRWEIGHTS_1					 0x168810
+#define QM_REG_WRRWEIGHTS_10					 0x168814
+#define QM_REG_WRRWEIGHTS_10_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_11					 0x168818
+#define QM_REG_WRRWEIGHTS_11_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_12					 0x16881c
+#define QM_REG_WRRWEIGHTS_12_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_13					 0x168820
+#define QM_REG_WRRWEIGHTS_13_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_14					 0x168824
+#define QM_REG_WRRWEIGHTS_14_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_15					 0x168828
+#define QM_REG_WRRWEIGHTS_15_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_10					 0x168814
+#define QM_REG_WRRWEIGHTS_11					 0x168818
+#define QM_REG_WRRWEIGHTS_12					 0x16881c
+#define QM_REG_WRRWEIGHTS_13					 0x168820
+#define QM_REG_WRRWEIGHTS_14					 0x168824
+#define QM_REG_WRRWEIGHTS_15					 0x168828
+#define QM_REG_WRRWEIGHTS_2					 0x16882c
+#define QM_REG_WRRWEIGHTS_3					 0x168830
+#define QM_REG_WRRWEIGHTS_4					 0x168834
+#define QM_REG_WRRWEIGHTS_5					 0x168838
+#define QM_REG_WRRWEIGHTS_6					 0x16883c
+#define QM_REG_WRRWEIGHTS_7					 0x168840
+#define QM_REG_WRRWEIGHTS_8					 0x168844
+#define QM_REG_WRRWEIGHTS_9					 0x168848
+/* [R 6] Keep the fill level of the fifo from write client 1 */
+#define QM_REG_XQM_WRC_FIFOLVL					 0x168000
+#define DORQ_DORQ_INT_STS_REG_ADDRESS_ERROR			 (0x1<<0)
+#define DORQ_DORQ_INT_STS_REG_ADDRESS_ERROR_SIZE		 0
+#define DORQ_DORQ_INT_STS_CLR_REG_ADDRESS_ERROR 		 (0x1<<0)
+#define DORQ_DORQ_INT_STS_CLR_REG_ADDRESS_ERROR_SIZE		 0
+#define DORQ_DORQ_INT_STS_WR_REG_ADDRESS_ERROR			 (0x1<<0)
+#define DORQ_DORQ_INT_STS_WR_REG_ADDRESS_ERROR_SIZE		 0
+#define DORQ_DORQ_INT_MASK_REG_ADDRESS_ERROR			 (0x1<<0)
+#define DORQ_DORQ_INT_MASK_REG_ADDRESS_ERROR_SIZE		 0
+#define NIG_NIG_INT_STS_0_REG_ADDRESS_ERROR			 (0x1<<0)
+#define NIG_NIG_INT_STS_0_REG_ADDRESS_ERROR_SIZE		 0
+#define NIG_NIG_INT_STS_CLR_0_REG_ADDRESS_ERROR 		 (0x1<<0)
+#define NIG_NIG_INT_STS_CLR_0_REG_ADDRESS_ERROR_SIZE		 0
+#define NIG_NIG_INT_STS_WR_0_REG_ADDRESS_ERROR			 (0x1<<0)
+#define NIG_NIG_INT_STS_WR_0_REG_ADDRESS_ERROR_SIZE		 0
+#define NIG_NIG_INT_MASK_0_REG_ADDRESS_ERROR			 (0x1<<0)
+#define NIG_NIG_INT_MASK_0_REG_ADDRESS_ERROR_SIZE		 0
+#define TCM_TCM_INT_STS_REG_ADDRESS_ERROR			 (0x1<<0)
+#define TCM_TCM_INT_STS_REG_ADDRESS_ERROR_SIZE			 0
+#define TCM_TCM_INT_STS_CLR_REG_ADDRESS_ERROR			 (0x1<<0)
+#define TCM_TCM_INT_STS_CLR_REG_ADDRESS_ERROR_SIZE		 0
+#define TCM_TCM_INT_STS_WR_REG_ADDRESS_ERROR			 (0x1<<0)
+#define TCM_TCM_INT_STS_WR_REG_ADDRESS_ERROR_SIZE		 0
+#define TCM_TCM_INT_MASK_REG_ADDRESS_ERROR			 (0x1<<0)
+#define TCM_TCM_INT_MASK_REG_ADDRESS_ERROR_SIZE 		 0
+#define CFC_DEBUG1_REG_WRITE_AC 				 (0x1<<4)
+#define CFC_DEBUG1_REG_WRITE_AC_SIZE				 4
+/* [R 1] debug only: This bit indicates wheter indicates that external
+   buffer was wrapped (oldest data was thrown); Relevant only when
+   ~dbg_registers_debug_target=2 (PCI) & ~dbg_registers_full_mode=1 (wrap); */
+#define DBG_REG_WRAP_ON_EXT_BUFFER				 0xc124
+#define DBG_REG_WRAP_ON_EXT_BUFFER_SIZE 			 1
+/* [R 1] debug only: This bit indicates wheter the internal buffer was
+   wrapped (oldest data was thrown) Relevant only when
+   ~dbg_registers_debug_target=0 (internal buffer) */
+#define DBG_REG_WRAP_ON_INT_BUFFER				 0xc128
+#define DBG_REG_WRAP_ON_INT_BUFFER_SIZE 			 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_0					 0x16880c
+#define QM_REG_WRRWEIGHTS_0_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_1					 0x168810
+#define QM_REG_WRRWEIGHTS_1_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_10					 0x168814
+#define QM_REG_WRRWEIGHTS_10_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_11					 0x168818
+#define QM_REG_WRRWEIGHTS_11_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_12					 0x16881c
+#define QM_REG_WRRWEIGHTS_12_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_13					 0x168820
+#define QM_REG_WRRWEIGHTS_13_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_14					 0x168824
+#define QM_REG_WRRWEIGHTS_14_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_15					 0x168828
+#define QM_REG_WRRWEIGHTS_15_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_2					 0x16882c
+#define QM_REG_WRRWEIGHTS_2_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_3					 0x168830
+#define QM_REG_WRRWEIGHTS_3_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_4					 0x168834
+#define QM_REG_WRRWEIGHTS_4_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_5					 0x168838
+#define QM_REG_WRRWEIGHTS_5_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_6					 0x16883c
+#define QM_REG_WRRWEIGHTS_6_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_7					 0x168840
+#define QM_REG_WRRWEIGHTS_7_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_8					 0x168844
+#define QM_REG_WRRWEIGHTS_8_SIZE				 1
+/* [RW 32] Wrr weights */
+#define QM_REG_WRRWEIGHTS_9					 0x168848
+#define QM_REG_WRRWEIGHTS_9_SIZE				 1
+/* [RW 22] Number of free element in the free list of T2 entries - port 0. */
+#define SRC_REG_COUNTFREE0					 0x40500
+/* [WB 64] First free element in the free list of T2 entries - port 0. */
+#define SRC_REG_FIRSTFREE0					 0x40510
+#define SRC_REG_KEYRSS0_0					 0x40408
+#define SRC_REG_KEYRSS1_9					 0x40454
+/* [WB 64] Last free element in the free list of T2 entries - port 0. */
+#define SRC_REG_LASTFREE0					 0x40530
+/* [RW 5] The number of hash bits used for the search (h); Values can be 8
+   to 24. */
+#define SRC_REG_NUMBER_HASH_BITS0				 0x40400
+/* [RW 1] Reset internal state machines. */
+#define SRC_REG_SOFT_RST					 0x4049c
+/* [R 1] Interrupt register #0 read */
+#define SRC_REG_SRC_INT_STS					 0x404ac
+/* [RW 3] Parity mask register #0 read/write */
+#define SRC_REG_SRC_PRTY_MASK					 0x404c8
+/* [R 4] Used to read the value of the XX protection CAM occupancy counter. */
+#define TCM_REG_CAM_OCCUP					 0x5017c
+/* [RW 1] CDU AG read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define TCM_REG_CDU_AG_RD_IFEN					 0x50034
+/* [RW 1] CDU AG write Interface enable. If 0 - the request and valid input
+   are disregarded; all other signals are treated as usual; if 1 - normal
+   activity. */
+#define TCM_REG_CDU_AG_WR_IFEN					 0x50030
+/* [RW 1] CDU STORM read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define TCM_REG_CDU_SM_RD_IFEN					 0x5003c
+/* [RW 1] CDU STORM write Interface enable. If 0 - the request and valid
+   input is disregarded; all other signals are treated as usual; if 1 -
+   normal activity. */
+#define TCM_REG_CDU_SM_WR_IFEN					 0x50038
+/* [RW 4] CFC output initial credit. Max credit available - 15.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 1 at start-up. */
+#define TCM_REG_CFC_INIT_CRD					 0x50204
+/* [RW 3] The weight of the CP input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define TCM_REG_CP_WEIGHT					 0x500c0
+/* [RW 1] Input csem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define TCM_REG_CSEM_IFEN					 0x5002c
+/* [RC 1] Message length mismatch (relative to last indication) at the In#9
+   interface. */
+#define TCM_REG_CSEM_LENGTH_MIS 				 0x50174
+/* [RW 8] The Event ID in case of ErrorFlg is set in the input message. */
+#define TCM_REG_ERR_EVNT_ID					 0x500a0
+/* [RW 28] The CM erroneous header for QM and Timers formatting. */
+#define TCM_REG_ERR_TCM_HDR					 0x5009c
+/* [RW 8] The Event ID for Timers expiration. */
+#define TCM_REG_EXPR_EVNT_ID					 0x500a4
+/* [RW 8] FIC0 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define TCM_REG_FIC0_INIT_CRD					 0x5020c
+/* [RW 8] FIC1 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define TCM_REG_FIC1_INIT_CRD					 0x50210
+/* [RW 1] Arbitration between Input Arbiter groups: 0 - fair Round-Robin; 1
+   - strict priority defined by ~tcm_registers_gr_ag_pr.gr_ag_pr;
+   ~tcm_registers_gr_ld0_pr.gr_ld0_pr and
+   ~tcm_registers_gr_ld1_pr.gr_ld1_pr. */
+#define TCM_REG_GR_ARB_TYPE					 0x50114
+/* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Store channel is the
+   compliment of the other 3 groups. */
+#define TCM_REG_GR_LD0_PR					 0x5011c
+/* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Store channel is the
+   compliment of the other 3 groups. */
+#define TCM_REG_GR_LD1_PR					 0x50120
+/* [RW 4] The number of double REG-pairs; loaded from the STORM context and
+   sent to STORM; for a specific connection type. The double REG-pairs are
+   used to align to STORM context row size of 128 bits. The offset of these
+   data in the STORM context is always 0. Index _i stands for the connection
+   type (one of 16). */
+#define TCM_REG_N_SM_CTX_LD_0					 0x50050
+#define TCM_REG_N_SM_CTX_LD_1					 0x50054
+#define TCM_REG_N_SM_CTX_LD_10					 0x50078
+#define TCM_REG_N_SM_CTX_LD_11					 0x5007c
+#define TCM_REG_N_SM_CTX_LD_12					 0x50080
+#define TCM_REG_N_SM_CTX_LD_13					 0x50084
+#define TCM_REG_N_SM_CTX_LD_14					 0x50088
+#define TCM_REG_N_SM_CTX_LD_15					 0x5008c
+#define TCM_REG_N_SM_CTX_LD_2					 0x50058
+#define TCM_REG_N_SM_CTX_LD_3					 0x5005c
+#define TCM_REG_N_SM_CTX_LD_4					 0x50060
+/* [RW 1] Input pbf Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_PBF_IFEN					 0x50024
+/* [RC 1] Message length mismatch (relative to last indication) at the In#7
+   interface. */
+#define TCM_REG_PBF_LENGTH_MIS					 0x5016c
+/* [RW 3] The weight of the input pbf in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define TCM_REG_PBF_WEIGHT					 0x500b4
+/* [RW 6] The physical queue number 0 per port index. */
+#define TCM_REG_PHYS_QNUM0_0					 0x500e0
+#define TCM_REG_PHYS_QNUM0_1					 0x500e4
+/* [RW 6] The physical queue number 1 per port index. */
+#define TCM_REG_PHYS_QNUM1_0					 0x500e8
+/* [RW 1] Input prs Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_PRS_IFEN					 0x50020
+/* [RC 1] Message length mismatch (relative to last indication) at the In#6
+   interface. */
+#define TCM_REG_PRS_LENGTH_MIS					 0x50168
+/* [RW 3] The weight of the input prs in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define TCM_REG_PRS_WEIGHT					 0x500b0
+/* [RW 8] The Event ID for Timers formatting in case of stop done. */
+#define TCM_REG_STOP_EVNT_ID					 0x500a8
+/* [RC 1] Message length mismatch (relative to last indication) at the STORM
+   interface. */
+#define TCM_REG_STORM_LENGTH_MIS				 0x50160
+/* [RW 1] STORM - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define TCM_REG_STORM_TCM_IFEN					 0x50010
+/* [RW 1] CM - CFC Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TCM_CFC_IFEN					 0x50040
+/* [RW 11] Interrupt mask register #0 read/write */
+#define TCM_REG_TCM_INT_MASK					 0x501dc
+/* [R 11] Interrupt register #0 read */
+#define TCM_REG_TCM_INT_STS					 0x501d0
+/* [RW 3] The size of AG context region 0 in REG-pairs. Designates the MS
+   REG-pair number (e.g. if region 0 is 6 REG-pairs; the value should be 5).
+   Is used to determine the number of the AG context REG-pairs written back;
+   when the input message Reg1WbFlg isn't set. */
+#define TCM_REG_TCM_REG0_SZ					 0x500d8
+/* [RW 1] CM - STORM 0 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TCM_STORM0_IFEN 				 0x50004
+/* [RW 1] CM - STORM 1 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TCM_STORM1_IFEN 				 0x50008
+/* [RW 1] CM - QM Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TCM_TQM_IFEN					 0x5000c
+/* [RW 1] If set the Q index; received from the QM is inserted to event ID. */
+#define TCM_REG_TCM_TQM_USE_Q					 0x500d4
+/* [RW 28] The CM header for Timers expiration command. */
+#define TCM_REG_TM_TCM_HDR					 0x50098
+/* [RW 1] Timers - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define TCM_REG_TM_TCM_IFEN					 0x5001c
+/* [RW 6] QM output initial credit. Max credit available - 32.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 32 at start-up. */
+#define TCM_REG_TQM_INIT_CRD					 0x5021c
+/* [RW 28] The CM header value for QM request (primary). */
+#define TCM_REG_TQM_TCM_HDR_P					 0x50090
+/* [RW 28] The CM header value for QM request (secondary). */
+#define TCM_REG_TQM_TCM_HDR_S					 0x50094
+/* [RW 1] QM - CM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TQM_TCM_IFEN					 0x50014
+/* [RW 1] Input SDM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define TCM_REG_TSDM_IFEN					 0x50018
+/* [RC 1] Message length mismatch (relative to last indication) at the SDM
+   interface. */
+#define TCM_REG_TSDM_LENGTH_MIS 				 0x50164
+/* [RW 3] The weight of the SDM input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define TCM_REG_TSDM_WEIGHT					 0x500c4
+/* [RW 1] Input usem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define TCM_REG_USEM_IFEN					 0x50028
+/* [RC 1] Message length mismatch (relative to last indication) at the In#8
+   interface. */
+#define TCM_REG_USEM_LENGTH_MIS 				 0x50170
+/* [RW 21] Indirect access to the descriptor table of the XX protection
+   mechanism. The fields are: [5:0] - length of the message; 15:6] - message
+   pointer; 20:16] - next pointer. */
+#define TCM_REG_XX_DESCR_TABLE					 0x50280
+/* [R 6] Use to read the value of XX protection Free counter. */
+#define TCM_REG_XX_FREE 					 0x50178
+/* [RW 6] Initial value for the credit counter; responsible for fulfilling
+   of the Input Stage XX protection buffer by the XX protection pending
+   messages. Max credit available - 127.Write writes the initial credit
+   value; read returns the current value of the credit counter. Must be
+   initialized to 19 at start-up. */
+#define TCM_REG_XX_INIT_CRD					 0x50220
+/* [RW 6] Maximum link list size (messages locked) per connection in the XX
+   protection. */
+#define TCM_REG_XX_MAX_LL_SZ					 0x50044
+/* [RW 6] The maximum number of pending messages; which may be stored in XX
+   protection. ~tcm_registers_xx_free.xx_free is read on read. */
+#define TCM_REG_XX_MSG_NUM					 0x50224
+/* [RW 8] The Event ID; sent to the STORM in case of XX overflow. */
+#define TCM_REG_XX_OVFL_EVNT_ID 				 0x50048
+/* [RW 16] Indirect access to the XX table of the XX protection mechanism.
+   The fields are:[4:0] - tail pointer; [10:5] - Link List size; 15:11] -
+   header pointer. */
+#define TCM_REG_XX_TABLE					 0x50240
+/* [RW 4] Load value for for cfc ac credit cnt. */
+#define TM_REG_CFC_AC_CRDCNT_VAL				 0x164208
+/* [RW 4] Load value for cfc cld credit cnt. */
+#define TM_REG_CFC_CLD_CRDCNT_VAL				 0x164210
+/* [RW 8] Client0 context region. */
+#define TM_REG_CL0_CONT_REGION					 0x164030
+/* [RW 8] Client1 context region. */
+#define TM_REG_CL1_CONT_REGION					 0x164034
+/* [RW 8] Client2 context region. */
+#define TM_REG_CL2_CONT_REGION					 0x164038
+/* [RW 2] Client in High priority client number. */
+#define TM_REG_CLIN_PRIOR0_CLIENT				 0x164024
+/* [RW 4] Load value for clout0 cred cnt. */
+#define TM_REG_CLOUT_CRDCNT0_VAL				 0x164220
+/* [RW 4] Load value for clout1 cred cnt. */
+#define TM_REG_CLOUT_CRDCNT1_VAL				 0x164228
+/* [RW 4] Load value for clout2 cred cnt. */
+#define TM_REG_CLOUT_CRDCNT2_VAL				 0x164230
+/* [RW 1] Enable client0 input. */
+#define TM_REG_EN_CL0_INPUT					 0x164008
+/* [RW 1] Enable client1 input. */
+#define TM_REG_EN_CL1_INPUT					 0x16400c
+/* [RW 1] Enable client2 input. */
+#define TM_REG_EN_CL2_INPUT					 0x164010
+/* [RW 1] Enable real time counter. */
+#define TM_REG_EN_REAL_TIME_CNT 				 0x1640d8
+/* [RW 1] Enable for Timers state machines. */
+#define TM_REG_EN_TIMERS					 0x164000
+/* [RW 4] Load value for expiration credit cnt. CFC max number of
+   outstanding load requests for timers (expiration) context loading. */
+#define TM_REG_EXP_CRDCNT_VAL					 0x164238
+/* [RW 18] Linear0 Max active cid. */
+#define TM_REG_LIN0_MAX_ACTIVE_CID				 0x164048
+/* [WB 64] Linear0 phy address. */
+#define TM_REG_LIN0_PHY_ADDR					 0x164270
+/* [RW 24] Linear0 array scan timeout. */
+#define TM_REG_LIN0_SCAN_TIME					 0x16403c
+/* [WB 64] Linear1 phy address. */
+#define TM_REG_LIN1_PHY_ADDR					 0x164280
+/* [RW 6] Linear timer set_clear fifo threshold. */
+#define TM_REG_LIN_SETCLR_FIFO_ALFULL_THR			 0x164070
+/* [RW 2] Load value for pci arbiter credit cnt. */
+#define TM_REG_PCIARB_CRDCNT_VAL				 0x164260
+/* [RW 1] Timer software reset - active high. */
+#define TM_REG_TIMER_SOFT_RST					 0x164004
+/* [RW 20] The amount of hardware cycles for each timer tick. */
+#define TM_REG_TIMER_TICK_SIZE					 0x16401c
+/* [RW 8] Timers Context region. */
+#define TM_REG_TM_CONTEXT_REGION				 0x164044
+/* [RW 1] Interrupt mask register #0 read/write */
+#define TM_REG_TM_INT_MASK					 0x1640fc
+/* [R 1] Interrupt register #0 read */
+#define TM_REG_TM_INT_STS					 0x1640f0
+/* [RW 8] The event id for aggregated interrupt 0 */
+#define TSDM_REG_AGG_INT_EVENT_0				 0x42038
+/* [RW 13] The start address in the internal RAM for the cfc_rsp lcid */
+#define TSDM_REG_CFC_RSP_START_ADDR				 0x42008
+/* [RW 16] The maximum value of the competion counter #0 */
+#define TSDM_REG_CMP_COUNTER_MAX0				 0x4201c
+/* [RW 16] The maximum value of the competion counter #1 */
+#define TSDM_REG_CMP_COUNTER_MAX1				 0x42020
+/* [RW 16] The maximum value of the competion counter #2 */
+#define TSDM_REG_CMP_COUNTER_MAX2				 0x42024
+/* [RW 16] The maximum value of the competion counter #3 */
+#define TSDM_REG_CMP_COUNTER_MAX3				 0x42028
+/* [RW 13] The start address in the internal RAM for the completion
+   counters. */
+#define TSDM_REG_CMP_COUNTER_START_ADDR 			 0x4200c
+#define TSDM_REG_ENABLE_IN1					 0x42238
+#define TSDM_REG_ENABLE_IN2					 0x4223c
+#define TSDM_REG_ENABLE_OUT1					 0x42240
+#define TSDM_REG_ENABLE_OUT2					 0x42244
+/* [RW 4] The initial number of messages that can be sent to the pxp control
+   interface without receiving any ACK. */
+#define TSDM_REG_INIT_CREDIT_PXP_CTRL				 0x424bc
+/* [ST 32] The number of ACK after placement messages received */
+#define TSDM_REG_NUM_OF_ACK_AFTER_PLACE 			 0x4227c
+/* [ST 32] The number of packet end messages received from the parser */
+#define TSDM_REG_NUM_OF_PKT_END_MSG				 0x42274
+/* [ST 32] The number of requests received from the pxp async if */
+#define TSDM_REG_NUM_OF_PXP_ASYNC_REQ				 0x42278
+/* [ST 32] The number of commands received in queue 0 */
+#define TSDM_REG_NUM_OF_Q0_CMD					 0x42248
+/* [ST 32] The number of commands received in queue 10 */
+#define TSDM_REG_NUM_OF_Q10_CMD 				 0x4226c
+/* [ST 32] The number of commands received in queue 11 */
+#define TSDM_REG_NUM_OF_Q11_CMD 				 0x42270
+/* [ST 32] The number of commands received in queue 1 */
+#define TSDM_REG_NUM_OF_Q1_CMD					 0x4224c
+/* [ST 32] The number of commands received in queue 3 */
+#define TSDM_REG_NUM_OF_Q3_CMD					 0x42250
+/* [ST 32] The number of commands received in queue 4 */
+#define TSDM_REG_NUM_OF_Q4_CMD					 0x42254
+/* [ST 32] The number of commands received in queue 5 */
+#define TSDM_REG_NUM_OF_Q5_CMD					 0x42258
+/* [ST 32] The number of commands received in queue 6 */
+#define TSDM_REG_NUM_OF_Q6_CMD					 0x4225c
+/* [ST 32] The number of commands received in queue 7 */
+#define TSDM_REG_NUM_OF_Q7_CMD					 0x42260
+/* [ST 32] The number of commands received in queue 8 */
+#define TSDM_REG_NUM_OF_Q8_CMD					 0x42264
+/* [ST 32] The number of commands received in queue 9 */
+#define TSDM_REG_NUM_OF_Q9_CMD					 0x42268
+/* [RW 13] The start address in the internal RAM for the packet end message */
+#define TSDM_REG_PCK_END_MSG_START_ADDR 			 0x42014
+/* [RW 13] The start address in the internal RAM for queue counters */
+#define TSDM_REG_Q_COUNTER_START_ADDR				 0x42010
+/* [R 1] pxp_ctrl rd_data fifo empty in sdm_dma_rsp block */
+#define TSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY			 0x42548
+/* [R 1] parser fifo empty in sdm_sync block */
+#define TSDM_REG_SYNC_PARSER_EMPTY				 0x42550
+/* [R 1] parser serial fifo empty in sdm_sync block */
+#define TSDM_REG_SYNC_SYNC_EMPTY				 0x42558
+/* [RW 32] Tick for timer counter. Applicable only when
+   ~tsdm_registers_timer_tick_enable.timer_tick_enable =1 */
+#define TSDM_REG_TIMER_TICK					 0x42000
+/* [RW 32] Interrupt mask register #0 read/write */
+#define TSDM_REG_TSDM_INT_MASK_0				 0x4229c
+#define TSDM_REG_TSDM_INT_MASK_1				 0x422ac
+/* [RW 11] Parity mask register #0 read/write */
+#define TSDM_REG_TSDM_PRTY_MASK 				 0x422bc
+/* [RW 5] The number of time_slots in the arbitration cycle */
+#define TSEM_REG_ARB_CYCLE_SIZE 				 0x180034
+/* [RW 3] The source that is associated with arbitration element 0. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2 */
+#define TSEM_REG_ARB_ELEMENT0					 0x180020
+/* [RW 3] The source that is associated with arbitration element 1. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~tsem_registers_arb_element0.arb_element0 */
+#define TSEM_REG_ARB_ELEMENT1					 0x180024
+/* [RW 3] The source that is associated with arbitration element 2. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~tsem_registers_arb_element0.arb_element0
+   and ~tsem_registers_arb_element1.arb_element1 */
+#define TSEM_REG_ARB_ELEMENT2					 0x180028
+/* [RW 3] The source that is associated with arbitration element 3. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.Could
+   not be equal to register ~tsem_registers_arb_element0.arb_element0 and
+   ~tsem_registers_arb_element1.arb_element1 and
+   ~tsem_registers_arb_element2.arb_element2 */
+#define TSEM_REG_ARB_ELEMENT3					 0x18002c
+/* [RW 3] The source that is associated with arbitration element 4. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~tsem_registers_arb_element0.arb_element0
+   and ~tsem_registers_arb_element1.arb_element1 and
+   ~tsem_registers_arb_element2.arb_element2 and
+   ~tsem_registers_arb_element3.arb_element3 */
+#define TSEM_REG_ARB_ELEMENT4					 0x180030
+#define TSEM_REG_ENABLE_IN					 0x1800a4
+#define TSEM_REG_ENABLE_OUT					 0x1800a8
+/* [RW 32] This address space contains all registers and memories that are
+   placed in SEM_FAST block. The SEM_FAST registers are described in
+   appendix B. In order to access the SEM_FAST registers the base address
+   TSEM_REGISTERS_FAST_MEMORY (Offset: 0x1a0000) should be added to each
+   SEM_FAST register offset. */
+#define TSEM_REG_FAST_MEMORY					 0x1a0000
+/* [RW 1] Disables input messages from FIC0 May be updated during run_time
+   by the microcode */
+#define TSEM_REG_FIC0_DISABLE					 0x180224
+/* [RW 1] Disables input messages from FIC1 May be updated during run_time
+   by the microcode */
+#define TSEM_REG_FIC1_DISABLE					 0x180234
+/* [RW 15] Interrupt table Read and write access to it is not possible in
+   the middle of the work */
+#define TSEM_REG_INT_TABLE					 0x180400
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC0 */
+#define TSEM_REG_MSG_NUM_FIC0					 0x180000
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC1 */
+#define TSEM_REG_MSG_NUM_FIC1					 0x180004
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC0 */
+#define TSEM_REG_MSG_NUM_FOC0					 0x180008
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC1 */
+#define TSEM_REG_MSG_NUM_FOC1					 0x18000c
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC2 */
+#define TSEM_REG_MSG_NUM_FOC2					 0x180010
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC3 */
+#define TSEM_REG_MSG_NUM_FOC3					 0x180014
+/* [RW 1] Disables input messages from the passive buffer May be updated
+   during run_time by the microcode */
+#define TSEM_REG_PAS_DISABLE					 0x18024c
+/* [WB 128] Debug only. Passive buffer memory */
+#define TSEM_REG_PASSIVE_BUFFER 				 0x181000
+/* [WB 46] pram memory. B45 is parity; b[44:0] - data. */
+#define TSEM_REG_PRAM						 0x1c0000
+/* [R 8] Valid sleeping threads indication have bit per thread */
+#define TSEM_REG_SLEEP_THREADS_VALID				 0x18026c
+/* [R 1] EXT_STORE FIFO is empty in sem_slow_ls_ext */
+#define TSEM_REG_SLOW_EXT_STORE_EMPTY				 0x1802a0
+/* [RW 8] List of free threads . There is a bit per thread. */
+#define TSEM_REG_THREADS_LIST					 0x1802e4
+/* [RW 3] The arbitration scheme of time_slot 0 */
+#define TSEM_REG_TS_0_AS					 0x180038
+/* [RW 3] The arbitration scheme of time_slot 10 */
+#define TSEM_REG_TS_10_AS					 0x180060
+/* [RW 3] The arbitration scheme of time_slot 11 */
+#define TSEM_REG_TS_11_AS					 0x180064
+/* [RW 3] The arbitration scheme of time_slot 12 */
+#define TSEM_REG_TS_12_AS					 0x180068
+/* [RW 3] The arbitration scheme of time_slot 13 */
+#define TSEM_REG_TS_13_AS					 0x18006c
+/* [RW 3] The arbitration scheme of time_slot 14 */
+#define TSEM_REG_TS_14_AS					 0x180070
+/* [RW 3] The arbitration scheme of time_slot 15 */
+#define TSEM_REG_TS_15_AS					 0x180074
+/* [RW 3] The arbitration scheme of time_slot 16 */
+#define TSEM_REG_TS_16_AS					 0x180078
+/* [RW 3] The arbitration scheme of time_slot 17 */
+#define TSEM_REG_TS_17_AS					 0x18007c
+/* [RW 3] The arbitration scheme of time_slot 18 */
+#define TSEM_REG_TS_18_AS					 0x180080
+/* [RW 3] The arbitration scheme of time_slot 1 */
+#define TSEM_REG_TS_1_AS					 0x18003c
+/* [RW 3] The arbitration scheme of time_slot 2 */
+#define TSEM_REG_TS_2_AS					 0x180040
+/* [RW 3] The arbitration scheme of time_slot 3 */
+#define TSEM_REG_TS_3_AS					 0x180044
+/* [RW 3] The arbitration scheme of time_slot 4 */
+#define TSEM_REG_TS_4_AS					 0x180048
+/* [RW 3] The arbitration scheme of time_slot 5 */
+#define TSEM_REG_TS_5_AS					 0x18004c
+/* [RW 3] The arbitration scheme of time_slot 6 */
+#define TSEM_REG_TS_6_AS					 0x180050
+/* [RW 3] The arbitration scheme of time_slot 7 */
+#define TSEM_REG_TS_7_AS					 0x180054
+/* [RW 3] The arbitration scheme of time_slot 8 */
+#define TSEM_REG_TS_8_AS					 0x180058
+/* [RW 3] The arbitration scheme of time_slot 9 */
+#define TSEM_REG_TS_9_AS					 0x18005c
+/* [RW 32] Interrupt mask register #0 read/write */
+#define TSEM_REG_TSEM_INT_MASK_0				 0x180100
+#define TSEM_REG_TSEM_INT_MASK_1				 0x180110
+/* [RW 32] Parity mask register #0 read/write */
+#define TSEM_REG_TSEM_PRTY_MASK_0				 0x180120
+#define TSEM_REG_TSEM_PRTY_MASK_1				 0x180130
+/* [R 5] Used to read the XX protection CAM occupancy counter. */
+#define UCM_REG_CAM_OCCUP					 0xe0170
+/* [RW 1] CDU AG read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define UCM_REG_CDU_AG_RD_IFEN					 0xe0038
+/* [RW 1] CDU AG write Interface enable. If 0 - the request and valid input
+   are disregarded; all other signals are treated as usual; if 1 - normal
+   activity. */
+#define UCM_REG_CDU_AG_WR_IFEN					 0xe0034
+/* [RW 1] CDU STORM read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define UCM_REG_CDU_SM_RD_IFEN					 0xe0040
+/* [RW 1] CDU STORM write Interface enable. If 0 - the request and valid
+   input is disregarded; all other signals are treated as usual; if 1 -
+   normal activity. */
+#define UCM_REG_CDU_SM_WR_IFEN					 0xe003c
+/* [RW 4] CFC output initial credit. Max credit available - 15.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 1 at start-up. */
+#define UCM_REG_CFC_INIT_CRD					 0xe0204
+/* [RW 3] The weight of the CP input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define UCM_REG_CP_WEIGHT					 0xe00c4
+/* [RW 1] Input csem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_CSEM_IFEN					 0xe0028
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the csem interface is detected. */
+#define UCM_REG_CSEM_LENGTH_MIS 				 0xe0160
+/* [RW 3] The weight of the input csem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define UCM_REG_CSEM_WEIGHT					 0xe00b8
+/* [RW 1] Input dorq Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_DORQ_IFEN					 0xe0030
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the dorq interface is detected. */
+#define UCM_REG_DORQ_LENGTH_MIS 				 0xe0168
+/* [RW 8] The Event ID in case ErrorFlg input message bit is set. */
+#define UCM_REG_ERR_EVNT_ID					 0xe00a4
+/* [RW 28] The CM erroneous header for QM and Timers formatting. */
+#define UCM_REG_ERR_UCM_HDR					 0xe00a0
+/* [RW 8] The Event ID for Timers expiration. */
+#define UCM_REG_EXPR_EVNT_ID					 0xe00a8
+/* [RW 8] FIC0 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define UCM_REG_FIC0_INIT_CRD					 0xe020c
+/* [RW 8] FIC1 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define UCM_REG_FIC1_INIT_CRD					 0xe0210
+/* [RW 1] Arbitration between Input Arbiter groups: 0 - fair Round-Robin; 1
+   - strict priority defined by ~ucm_registers_gr_ag_pr.gr_ag_pr;
+   ~ucm_registers_gr_ld0_pr.gr_ld0_pr and
+   ~ucm_registers_gr_ld1_pr.gr_ld1_pr. */
+#define UCM_REG_GR_ARB_TYPE					 0xe0144
+/* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Store channel group is
+   compliment to the others. */
+#define UCM_REG_GR_LD0_PR					 0xe014c
+/* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Store channel group is
+   compliment to the others. */
+#define UCM_REG_GR_LD1_PR					 0xe0150
+/* [RW 2] The queue index for invalidate counter flag decision. */
+#define UCM_REG_INV_CFLG_Q					 0xe00e4
+/* [RW 5] The number of double REG-pairs; loaded from the STORM context and
+   sent to STORM; for a specific connection type. the double REG-pairs are
+   used in order to align to STORM context row size of 128 bits. The offset
+   of these data in the STORM context is always 0. Index _i stands for the
+   connection type (one of 16). */
+#define UCM_REG_N_SM_CTX_LD_0					 0xe0054
+#define UCM_REG_N_SM_CTX_LD_1					 0xe0058
+#define UCM_REG_N_SM_CTX_LD_10					 0xe007c
+#define UCM_REG_N_SM_CTX_LD_11					 0xe0080
+#define UCM_REG_N_SM_CTX_LD_12					 0xe0084
+#define UCM_REG_N_SM_CTX_LD_13					 0xe0088
+#define UCM_REG_N_SM_CTX_LD_14					 0xe008c
+#define UCM_REG_N_SM_CTX_LD_15					 0xe0090
+#define UCM_REG_N_SM_CTX_LD_2					 0xe005c
+#define UCM_REG_N_SM_CTX_LD_3					 0xe0060
+#define UCM_REG_N_SM_CTX_LD_4					 0xe0064
+/* [RW 6] The physical queue number 0 per port index (CID[23]) */
+#define UCM_REG_PHYS_QNUM0_0					 0xe0110
+#define UCM_REG_PHYS_QNUM0_1					 0xe0114
+/* [RW 6] The physical queue number 1 per port index (CID[23]) */
+#define UCM_REG_PHYS_QNUM1_0					 0xe0118
+#define UCM_REG_PHYS_QNUM1_1					 0xe011c
+/* [RW 8] The Event ID for Timers formatting in case of stop done. */
+#define UCM_REG_STOP_EVNT_ID					 0xe00ac
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the STORM interface is detected. */
+#define UCM_REG_STORM_LENGTH_MIS				 0xe0154
+/* [RW 1] STORM - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_STORM_UCM_IFEN					 0xe0010
+/* [RW 4] Timers output initial credit. Max credit available - 15.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 4 at start-up. */
+#define UCM_REG_TM_INIT_CRD					 0xe021c
+/* [RW 28] The CM header for Timers expiration command. */
+#define UCM_REG_TM_UCM_HDR					 0xe009c
+/* [RW 1] Timers - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_TM_UCM_IFEN					 0xe001c
+/* [RW 1] Input tsem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_TSEM_IFEN					 0xe0024
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the tsem interface is detected. */
+#define UCM_REG_TSEM_LENGTH_MIS 				 0xe015c
+/* [RW 3] The weight of the input tsem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define UCM_REG_TSEM_WEIGHT					 0xe00b4
+/* [RW 1] CM - CFC Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_UCM_CFC_IFEN					 0xe0044
+/* [RW 11] Interrupt mask register #0 read/write */
+#define UCM_REG_UCM_INT_MASK					 0xe01d4
+/* [R 11] Interrupt register #0 read */
+#define UCM_REG_UCM_INT_STS					 0xe01c8
+/* [RW 2] The size of AG context region 0 in REG-pairs. Designates the MS
+   REG-pair number (e.g. if region 0 is 6 REG-pairs; the value should be 5).
+   Is used to determine the number of the AG context REG-pairs written back;
+   when the Reg1WbFlg isn't set. */
+#define UCM_REG_UCM_REG0_SZ					 0xe00dc
+/* [RW 1] CM - STORM 0 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_UCM_STORM0_IFEN 				 0xe0004
+/* [RW 1] CM - STORM 1 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_UCM_STORM1_IFEN 				 0xe0008
+/* [RW 1] CM - Timers Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_UCM_TM_IFEN					 0xe0020
+/* [RW 1] CM - QM Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_UCM_UQM_IFEN					 0xe000c
+/* [RW 1] If set the Q index; received from the QM is inserted to event ID. */
+#define UCM_REG_UCM_UQM_USE_Q					 0xe00d8
+/* [RW 6] QM output initial credit. Max credit available - 32.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 32 at start-up. */
+#define UCM_REG_UQM_INIT_CRD					 0xe0220
+/* [RW 3] The weight of the QM (primary) input in the WRR mechanism. 0
+   stands for weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define UCM_REG_UQM_P_WEIGHT					 0xe00cc
+/* [RW 28] The CM header value for QM request (primary). */
+#define UCM_REG_UQM_UCM_HDR_P					 0xe0094
+/* [RW 28] The CM header value for QM request (secondary). */
+#define UCM_REG_UQM_UCM_HDR_S					 0xe0098
+/* [RW 1] QM - CM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_UQM_UCM_IFEN					 0xe0014
+/* [RW 1] Input SDM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define UCM_REG_USDM_IFEN					 0xe0018
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the SDM interface is detected. */
+#define UCM_REG_USDM_LENGTH_MIS 				 0xe0158
+/* [RW 1] Input xsem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define UCM_REG_XSEM_IFEN					 0xe002c
+/* [RC 1] Set when the message length mismatch (relative to last indication)
+   at the xsem interface isdetected. */
+#define UCM_REG_XSEM_LENGTH_MIS 				 0xe0164
+/* [RW 20] Indirect access to the descriptor table of the XX protection
+   mechanism. The fields are:[5:0] - message length; 14:6] - message
+   pointer; 19:15] - next pointer. */
+#define UCM_REG_XX_DESCR_TABLE					 0xe0280
+/* [R 6] Use to read the XX protection Free counter. */
+#define UCM_REG_XX_FREE 					 0xe016c
+/* [RW 6] Initial value for the credit counter; responsible for fulfilling
+   of the Input Stage XX protection buffer by the XX protection pending
+   messages. Write writes the initial credit value; read returns the current
+   value of the credit counter. Must be initialized to 12 at start-up. */
+#define UCM_REG_XX_INIT_CRD					 0xe0224
+/* [RW 6] The maximum number of pending messages; which may be stored in XX
+   protection. ~ucm_registers_xx_free.xx_free read on read. */
+#define UCM_REG_XX_MSG_NUM					 0xe0228
+/* [RW 8] The Event ID; sent to the STORM in case of XX overflow. */
+#define UCM_REG_XX_OVFL_EVNT_ID 				 0xe004c
+/* [RW 16] Indirect access to the XX table of the XX protection mechanism.
+   The fields are: [4:0] - tail pointer; 10:5] - Link List size; 15:11] -
+   header pointer. */
+#define UCM_REG_XX_TABLE					 0xe0300
+/* [RW 8] The event id for aggregated interrupt 0 */
+#define USDM_REG_AGG_INT_EVENT_0				 0xc4038
+#define USDM_REG_AGG_INT_EVENT_1				 0xc403c
+#define USDM_REG_AGG_INT_EVENT_10				 0xc4060
+#define USDM_REG_AGG_INT_EVENT_11				 0xc4064
+#define USDM_REG_AGG_INT_EVENT_12				 0xc4068
+#define USDM_REG_AGG_INT_EVENT_13				 0xc406c
+#define USDM_REG_AGG_INT_EVENT_14				 0xc4070
+#define USDM_REG_AGG_INT_EVENT_15				 0xc4074
+#define USDM_REG_AGG_INT_EVENT_16				 0xc4078
+#define USDM_REG_AGG_INT_EVENT_17				 0xc407c
+#define USDM_REG_AGG_INT_EVENT_18				 0xc4080
+#define USDM_REG_AGG_INT_EVENT_19				 0xc4084
+/* [RW 1] For each aggregated interrupt index whether the mode is normal (0)
+   or auto-mask-mode (1) */
+#define USDM_REG_AGG_INT_MODE_0 				 0xc41b8
+#define USDM_REG_AGG_INT_MODE_1 				 0xc41bc
+#define USDM_REG_AGG_INT_MODE_10				 0xc41e0
+#define USDM_REG_AGG_INT_MODE_11				 0xc41e4
+#define USDM_REG_AGG_INT_MODE_12				 0xc41e8
+#define USDM_REG_AGG_INT_MODE_13				 0xc41ec
+#define USDM_REG_AGG_INT_MODE_14				 0xc41f0
+#define USDM_REG_AGG_INT_MODE_15				 0xc41f4
+#define USDM_REG_AGG_INT_MODE_16				 0xc41f8
+#define USDM_REG_AGG_INT_MODE_17				 0xc41fc
+#define USDM_REG_AGG_INT_MODE_18				 0xc4200
+#define USDM_REG_AGG_INT_MODE_19				 0xc4204
+/* [RW 13] The start address in the internal RAM for the cfc_rsp lcid */
+#define USDM_REG_CFC_RSP_START_ADDR				 0xc4008
+/* [RW 16] The maximum value of the competion counter #0 */
+#define USDM_REG_CMP_COUNTER_MAX0				 0xc401c
+/* [RW 16] The maximum value of the competion counter #1 */
+#define USDM_REG_CMP_COUNTER_MAX1				 0xc4020
+/* [RW 16] The maximum value of the competion counter #2 */
+#define USDM_REG_CMP_COUNTER_MAX2				 0xc4024
+/* [RW 16] The maximum value of the competion counter #3 */
+#define USDM_REG_CMP_COUNTER_MAX3				 0xc4028
+/* [RW 13] The start address in the internal RAM for the completion
+   counters. */
+#define USDM_REG_CMP_COUNTER_START_ADDR 			 0xc400c
+#define USDM_REG_ENABLE_IN1					 0xc4238
+#define USDM_REG_ENABLE_IN2					 0xc423c
+#define USDM_REG_ENABLE_OUT1					 0xc4240
+#define USDM_REG_ENABLE_OUT2					 0xc4244
+/* [RW 4] The initial number of messages that can be sent to the pxp control
+   interface without receiving any ACK. */
+#define USDM_REG_INIT_CREDIT_PXP_CTRL				 0xc44c0
+/* [ST 32] The number of ACK after placement messages received */
+#define USDM_REG_NUM_OF_ACK_AFTER_PLACE 			 0xc4280
+/* [ST 32] The number of packet end messages received from the parser */
+#define USDM_REG_NUM_OF_PKT_END_MSG				 0xc4278
+/* [ST 32] The number of requests received from the pxp async if */
+#define USDM_REG_NUM_OF_PXP_ASYNC_REQ				 0xc427c
+/* [ST 32] The number of commands received in queue 0 */
+#define USDM_REG_NUM_OF_Q0_CMD					 0xc4248
+/* [ST 32] The number of commands received in queue 10 */
+#define USDM_REG_NUM_OF_Q10_CMD 				 0xc4270
+/* [ST 32] The number of commands received in queue 11 */
+#define USDM_REG_NUM_OF_Q11_CMD 				 0xc4274
+/* [ST 32] The number of commands received in queue 1 */
+#define USDM_REG_NUM_OF_Q1_CMD					 0xc424c
+/* [ST 32] The number of commands received in queue 2 */
+#define USDM_REG_NUM_OF_Q2_CMD					 0xc4250
+/* [ST 32] The number of commands received in queue 3 */
+#define USDM_REG_NUM_OF_Q3_CMD					 0xc4254
+/* [ST 32] The number of commands received in queue 4 */
+#define USDM_REG_NUM_OF_Q4_CMD					 0xc4258
+/* [ST 32] The number of commands received in queue 5 */
+#define USDM_REG_NUM_OF_Q5_CMD					 0xc425c
+/* [ST 32] The number of commands received in queue 6 */
+#define USDM_REG_NUM_OF_Q6_CMD					 0xc4260
+/* [ST 32] The number of commands received in queue 7 */
+#define USDM_REG_NUM_OF_Q7_CMD					 0xc4264
+/* [ST 32] The number of commands received in queue 8 */
+#define USDM_REG_NUM_OF_Q8_CMD					 0xc4268
+/* [ST 32] The number of commands received in queue 9 */
+#define USDM_REG_NUM_OF_Q9_CMD					 0xc426c
+/* [RW 13] The start address in the internal RAM for the packet end message */
+#define USDM_REG_PCK_END_MSG_START_ADDR 			 0xc4014
+/* [RW 13] The start address in the internal RAM for queue counters */
+#define USDM_REG_Q_COUNTER_START_ADDR				 0xc4010
+/* [R 1] pxp_ctrl rd_data fifo empty in sdm_dma_rsp block */
+#define USDM_REG_RSP_PXP_CTRL_RDATA_EMPTY			 0xc4550
+/* [R 1] parser fifo empty in sdm_sync block */
+#define USDM_REG_SYNC_PARSER_EMPTY				 0xc4558
+/* [R 1] parser serial fifo empty in sdm_sync block */
+#define USDM_REG_SYNC_SYNC_EMPTY				 0xc4560
+/* [RW 32] Tick for timer counter. Applicable only when
+   ~usdm_registers_timer_tick_enable.timer_tick_enable =1 */
+#define USDM_REG_TIMER_TICK					 0xc4000
+/* [RW 32] Interrupt mask register #0 read/write */
+#define USDM_REG_USDM_INT_MASK_0				 0xc42a0
+#define USDM_REG_USDM_INT_MASK_1				 0xc42b0
+/* [RW 11] Parity mask register #0 read/write */
+#define USDM_REG_USDM_PRTY_MASK 				 0xc42c0
+/* [RW 5] The number of time_slots in the arbitration cycle */
+#define USEM_REG_ARB_CYCLE_SIZE 				 0x300034
+/* [RW 3] The source that is associated with arbitration element 0. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2 */
+#define USEM_REG_ARB_ELEMENT0					 0x300020
+/* [RW 3] The source that is associated with arbitration element 1. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~usem_registers_arb_element0.arb_element0 */
+#define USEM_REG_ARB_ELEMENT1					 0x300024
+/* [RW 3] The source that is associated with arbitration element 2. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~usem_registers_arb_element0.arb_element0
+   and ~usem_registers_arb_element1.arb_element1 */
+#define USEM_REG_ARB_ELEMENT2					 0x300028
+/* [RW 3] The source that is associated with arbitration element 3. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.Could
+   not be equal to register ~usem_registers_arb_element0.arb_element0 and
+   ~usem_registers_arb_element1.arb_element1 and
+   ~usem_registers_arb_element2.arb_element2 */
+#define USEM_REG_ARB_ELEMENT3					 0x30002c
+/* [RW 3] The source that is associated with arbitration element 4. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~usem_registers_arb_element0.arb_element0
+   and ~usem_registers_arb_element1.arb_element1 and
+   ~usem_registers_arb_element2.arb_element2 and
+   ~usem_registers_arb_element3.arb_element3 */
+#define USEM_REG_ARB_ELEMENT4					 0x300030
+#define USEM_REG_ENABLE_IN					 0x3000a4
+#define USEM_REG_ENABLE_OUT					 0x3000a8
+/* [RW 32] This address space contains all registers and memories that are
+   placed in SEM_FAST block. The SEM_FAST registers are described in
+   appendix B. In order to access the SEM_FAST registers... the base address
+   USEM_REGISTERS_FAST_MEMORY (Offset: 0x320000) should be added to each
+   SEM_FAST register offset. */
+#define USEM_REG_FAST_MEMORY					 0x320000
+/* [RW 1] Disables input messages from FIC0 May be updated during run_time
+   by the microcode */
+#define USEM_REG_FIC0_DISABLE					 0x300224
+/* [RW 1] Disables input messages from FIC1 May be updated during run_time
+   by the microcode */
+#define USEM_REG_FIC1_DISABLE					 0x300234
+/* [RW 15] Interrupt table Read and write access to it is not possible in
+   the middle of the work */
+#define USEM_REG_INT_TABLE					 0x300400
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC0 */
+#define USEM_REG_MSG_NUM_FIC0					 0x300000
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC1 */
+#define USEM_REG_MSG_NUM_FIC1					 0x300004
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC0 */
+#define USEM_REG_MSG_NUM_FOC0					 0x300008
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC1 */
+#define USEM_REG_MSG_NUM_FOC1					 0x30000c
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC2 */
+#define USEM_REG_MSG_NUM_FOC2					 0x300010
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC3 */
+#define USEM_REG_MSG_NUM_FOC3					 0x300014
+/* [RW 1] Disables input messages from the passive buffer May be updated
+   during run_time by the microcode */
+#define USEM_REG_PAS_DISABLE					 0x30024c
+/* [WB 128] Debug only. Passive buffer memory */
+#define USEM_REG_PASSIVE_BUFFER 				 0x302000
+/* [WB 46] pram memory. B45 is parity; b[44:0] - data. */
+#define USEM_REG_PRAM						 0x340000
+/* [R 16] Valid sleeping threads indication have bit per thread */
+#define USEM_REG_SLEEP_THREADS_VALID				 0x30026c
+/* [R 1] EXT_STORE FIFO is empty in sem_slow_ls_ext */
+#define USEM_REG_SLOW_EXT_STORE_EMPTY				 0x3002a0
+/* [RW 16] List of free threads . There is a bit per thread. */
+#define USEM_REG_THREADS_LIST					 0x3002e4
+/* [RW 3] The arbitration scheme of time_slot 0 */
+#define USEM_REG_TS_0_AS					 0x300038
+/* [RW 3] The arbitration scheme of time_slot 10 */
+#define USEM_REG_TS_10_AS					 0x300060
+/* [RW 3] The arbitration scheme of time_slot 11 */
+#define USEM_REG_TS_11_AS					 0x300064
+/* [RW 3] The arbitration scheme of time_slot 12 */
+#define USEM_REG_TS_12_AS					 0x300068
+/* [RW 3] The arbitration scheme of time_slot 13 */
+#define USEM_REG_TS_13_AS					 0x30006c
+/* [RW 3] The arbitration scheme of time_slot 14 */
+#define USEM_REG_TS_14_AS					 0x300070
+/* [RW 3] The arbitration scheme of time_slot 15 */
+#define USEM_REG_TS_15_AS					 0x300074
+/* [RW 3] The arbitration scheme of time_slot 16 */
+#define USEM_REG_TS_16_AS					 0x300078
+/* [RW 3] The arbitration scheme of time_slot 17 */
+#define USEM_REG_TS_17_AS					 0x30007c
+/* [RW 3] The arbitration scheme of time_slot 18 */
+#define USEM_REG_TS_18_AS					 0x300080
+/* [RW 3] The arbitration scheme of time_slot 1 */
+#define USEM_REG_TS_1_AS					 0x30003c
+/* [RW 3] The arbitration scheme of time_slot 2 */
+#define USEM_REG_TS_2_AS					 0x300040
+/* [RW 3] The arbitration scheme of time_slot 3 */
+#define USEM_REG_TS_3_AS					 0x300044
+/* [RW 3] The arbitration scheme of time_slot 4 */
+#define USEM_REG_TS_4_AS					 0x300048
+/* [RW 3] The arbitration scheme of time_slot 5 */
+#define USEM_REG_TS_5_AS					 0x30004c
+/* [RW 3] The arbitration scheme of time_slot 6 */
+#define USEM_REG_TS_6_AS					 0x300050
+/* [RW 3] The arbitration scheme of time_slot 7 */
+#define USEM_REG_TS_7_AS					 0x300054
+/* [RW 3] The arbitration scheme of time_slot 8 */
+#define USEM_REG_TS_8_AS					 0x300058
+/* [RW 3] The arbitration scheme of time_slot 9 */
+#define USEM_REG_TS_9_AS					 0x30005c
+/* [RW 32] Interrupt mask register #0 read/write */
+#define USEM_REG_USEM_INT_MASK_0				 0x300110
+#define USEM_REG_USEM_INT_MASK_1				 0x300120
+/* [RW 32] Parity mask register #0 read/write */
+#define USEM_REG_USEM_PRTY_MASK_0				 0x300130
+#define USEM_REG_USEM_PRTY_MASK_1				 0x300140
+/* [RW 2] The queue index for registration on Aux1 counter flag. */
+#define XCM_REG_AUX1_Q						 0x20134
+/* [RW 2] Per each decision rule the queue index to register to. */
+#define XCM_REG_AUX_CNT_FLG_Q_19				 0x201b0
+/* [R 5] Used to read the XX protection CAM occupancy counter. */
+#define XCM_REG_CAM_OCCUP					 0x20244
+/* [RW 1] CDU AG read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define XCM_REG_CDU_AG_RD_IFEN					 0x20044
+/* [RW 1] CDU AG write Interface enable. If 0 - the request and valid input
+   are disregarded; all other signals are treated as usual; if 1 - normal
+   activity. */
+#define XCM_REG_CDU_AG_WR_IFEN					 0x20040
+/* [RW 1] CDU STORM read Interface enable. If 0 - the request input is
+   disregarded; valid output is deasserted; all other signals are treated as
+   usual; if 1 - normal activity. */
+#define XCM_REG_CDU_SM_RD_IFEN					 0x2004c
+/* [RW 1] CDU STORM write Interface enable. If 0 - the request and valid
+   input is disregarded; all other signals are treated as usual; if 1 -
+   normal activity. */
+#define XCM_REG_CDU_SM_WR_IFEN					 0x20048
+/* [RW 4] CFC output initial credit. Max credit available - 15.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 1 at start-up. */
+#define XCM_REG_CFC_INIT_CRD					 0x20404
+/* [RW 3] The weight of the CP input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_CP_WEIGHT					 0x200dc
+/* [RW 1] Input csem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_CSEM_IFEN					 0x20028
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the csem interface. */
+#define XCM_REG_CSEM_LENGTH_MIS 				 0x20228
+/* [RW 3] The weight of the input csem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_CSEM_WEIGHT					 0x200c4
+/* [RW 1] Input dorq Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_DORQ_IFEN					 0x20030
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the dorq interface. */
+#define XCM_REG_DORQ_LENGTH_MIS 				 0x20230
+/* [RW 8] The Event ID in case the ErrorFlg input message bit is set. */
+#define XCM_REG_ERR_EVNT_ID					 0x200b0
+/* [RW 28] The CM erroneous header for QM and Timers formatting. */
+#define XCM_REG_ERR_XCM_HDR					 0x200ac
+/* [RW 8] The Event ID for Timers expiration. */
+#define XCM_REG_EXPR_EVNT_ID					 0x200b4
+/* [RW 8] FIC0 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define XCM_REG_FIC0_INIT_CRD					 0x2040c
+/* [RW 8] FIC1 output initial credit. Max credit available - 255.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 64 at start-up. */
+#define XCM_REG_FIC1_INIT_CRD					 0x20410
+/* [RW 8] The maximum delayed ACK counter value.Must be at least 2. Per port
+   value. */
+#define XCM_REG_GLB_DEL_ACK_MAX_CNT_0				 0x20118
+#define XCM_REG_GLB_DEL_ACK_MAX_CNT_1				 0x2011c
+/* [RW 28] The delayed ACK timeout in ticks. Per port value. */
+#define XCM_REG_GLB_DEL_ACK_TMR_VAL_0				 0x20108
+#define XCM_REG_GLB_DEL_ACK_TMR_VAL_1				 0x2010c
+/* [RW 1] Arbitratiojn between Input Arbiter groups: 0 - fair Round-Robin; 1
+   - strict priority defined by ~xcm_registers_gr_ag_pr.gr_ag_pr;
+   ~xcm_registers_gr_ld0_pr.gr_ld0_pr and
+   ~xcm_registers_gr_ld1_pr.gr_ld1_pr. */
+#define XCM_REG_GR_ARB_TYPE					 0x2020c
+/* [RW 2] Load (FIC0) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Channel group is the
+   compliment of the other 3 groups. */
+#define XCM_REG_GR_LD0_PR					 0x20214
+/* [RW 2] Load (FIC1) channel group priority. The lowest priority is 0; the
+   highest priority is 3. It is supposed that the Channel group is the
+   compliment of the other 3 groups. */
+#define XCM_REG_GR_LD1_PR					 0x20218
+/* [RW 1] Input nig0 Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_NIG0_IFEN					 0x20038
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the nig0 interface. */
+#define XCM_REG_NIG0_LENGTH_MIS 				 0x20238
+/* [RW 1] Input nig1 Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_NIG1_IFEN					 0x2003c
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the nig1 interface. */
+#define XCM_REG_NIG1_LENGTH_MIS 				 0x2023c
+/* [RW 3] The weight of the input nig1 in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_NIG1_WEIGHT					 0x200d8
+/* [RW 5] The number of double REG-pairs; loaded from the STORM context and
+   sent to STORM; for a specific connection type. The double REG-pairs are
+   used in order to align to STORM context row size of 128 bits. The offset
+   of these data in the STORM context is always 0. Index _i stands for the
+   connection type (one of 16). */
+#define XCM_REG_N_SM_CTX_LD_0					 0x20060
+#define XCM_REG_N_SM_CTX_LD_1					 0x20064
+#define XCM_REG_N_SM_CTX_LD_10					 0x20088
+#define XCM_REG_N_SM_CTX_LD_11					 0x2008c
+#define XCM_REG_N_SM_CTX_LD_12					 0x20090
+#define XCM_REG_N_SM_CTX_LD_13					 0x20094
+#define XCM_REG_N_SM_CTX_LD_14					 0x20098
+#define XCM_REG_N_SM_CTX_LD_15					 0x2009c
+#define XCM_REG_N_SM_CTX_LD_2					 0x20068
+#define XCM_REG_N_SM_CTX_LD_3					 0x2006c
+#define XCM_REG_N_SM_CTX_LD_4					 0x20070
+/* [RW 1] Input pbf Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_PBF_IFEN					 0x20034
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the pbf interface. */
+#define XCM_REG_PBF_LENGTH_MIS					 0x20234
+/* [RW 3] The weight of the input pbf in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_PBF_WEIGHT					 0x200d0
+/* [RW 8] The Event ID for Timers formatting in case of stop done. */
+#define XCM_REG_STOP_EVNT_ID					 0x200b8
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the STORM interface. */
+#define XCM_REG_STORM_LENGTH_MIS				 0x2021c
+/* [RW 3] The weight of the STORM input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_STORM_WEIGHT					 0x200bc
+/* [RW 1] STORM - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_STORM_XCM_IFEN					 0x20010
+/* [RW 4] Timers output initial credit. Max credit available - 15.Write
+   writes the initial credit value; read returns the current value of the
+   credit counter. Must be initialized to 4 at start-up. */
+#define XCM_REG_TM_INIT_CRD					 0x2041c
+/* [RW 28] The CM header for Timers expiration command. */
+#define XCM_REG_TM_XCM_HDR					 0x200a8
+/* [RW 1] Timers - CM Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_TM_XCM_IFEN					 0x2001c
+/* [RW 1] Input tsem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_TSEM_IFEN					 0x20024
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the tsem interface. */
+#define XCM_REG_TSEM_LENGTH_MIS 				 0x20224
+/* [RW 3] The weight of the input tsem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_TSEM_WEIGHT					 0x200c0
+/* [RW 2] The queue index for registration on UNA greater NXT decision rule. */
+#define XCM_REG_UNA_GT_NXT_Q					 0x20120
+/* [RW 1] Input usem Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_USEM_IFEN					 0x2002c
+/* [RC 1] Message length mismatch (relative to last indication) at the usem
+   interface. */
+#define XCM_REG_USEM_LENGTH_MIS 				 0x2022c
+/* [RW 3] The weight of the input usem in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_USEM_WEIGHT					 0x200c8
+/* [RW 2] DA counter command; used in case of window update doorbell.The
+   first index stands for the value DaEnable of that connection. The second
+   index stands for port number. */
+#define XCM_REG_WU_DA_CNT_CMD00 				 0x201d4
+/* [RW 2] DA counter command; used in case of window update doorbell.The
+   first index stands for the value DaEnable of that connection. The second
+   index stands for port number. */
+#define XCM_REG_WU_DA_CNT_CMD01 				 0x201d8
+/* [RW 2] DA counter command; used in case of window update doorbell.The
+   first index stands for the value DaEnable of that connection. The second
+   index stands for port number. */
+#define XCM_REG_WU_DA_CNT_CMD10 				 0x201dc
+/* [RW 2] DA counter command; used in case of window update doorbell.The
+   first index stands for the value DaEnable of that connection. The second
+   index stands for port number. */
+#define XCM_REG_WU_DA_CNT_CMD11 				 0x201e0
+/* [RW 8] DA counter update value used in case of window update doorbell.The
+   first index stands for the value DaEnable of that connection. The second
+   index stands for port number. */
+#define XCM_REG_WU_DA_CNT_UPD_VAL00				 0x201e4
+/* [RW 8] DA counter update value; used in case of window update
+   doorbell.The first index stands for the value DaEnable of that
+   connection. The second index stands for port number. */
+#define XCM_REG_WU_DA_CNT_UPD_VAL01				 0x201e8
+/* [RW 8] DA counter update value; used in case of window update
+   doorbell.The first index stands for the value DaEnable of that
+   connection. The second index stands for port number. */
+#define XCM_REG_WU_DA_CNT_UPD_VAL10				 0x201ec
+/* [RW 8] DA counter update value; used in case of window update
+   doorbell.The first index stands for the value DaEnable of that
+   connection. The second index stands for port number. */
+#define XCM_REG_WU_DA_CNT_UPD_VAL11				 0x201f0
+/* [RW 1] DA timer command; used in case of window update doorbell.The first
+   index stands for the value DaEnable of that connection. The second index
+   stands for port number. */
+#define XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD00			 0x201c4
+/* [RW 1] DA timer command; used in case of window update doorbell.The first
+   index stands for the value DaEnable of that connection. The second index
+   stands for port number. */
+#define XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD01			 0x201c8
+/* [RW 1] DA timer command; used in case of window update doorbell.The first
+   index stands for the value DaEnable of that connection. The second index
+   stands for port number. */
+#define XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD10			 0x201cc
+/* [RW 1] DA timer command; used in case of window update doorbell.The first
+   index stands for the value DaEnable of that connection. The second index
+   stands for port number. */
+#define XCM_REG_WU_DA_SET_TMR_CNT_FLG_CMD11			 0x201d0
+/* [RW 1] CM - CFC Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XCM_CFC_IFEN					 0x20050
+/* [RW 14] Interrupt mask register #0 read/write */
+#define XCM_REG_XCM_INT_MASK					 0x202b4
+/* [R 14] Interrupt register #0 read */
+#define XCM_REG_XCM_INT_STS					 0x202a8
+/* [RW 4] The size of AG context region 0 in REG-pairs. Designates the MS
+   REG-pair number (e.g. if region 0 is 6 REG-pairs; the value should be 5).
+   Is used to determine the number of the AG context REG-pairs written back;
+   when the Reg1WbFlg isn't set. */
+#define XCM_REG_XCM_REG0_SZ					 0x200f4
+/* [RW 1] CM - STORM 0 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XCM_STORM0_IFEN 				 0x20004
+/* [RW 1] CM - STORM 1 Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XCM_STORM1_IFEN 				 0x20008
+/* [RW 1] CM - Timers Interface enable. If 0 - the valid input is
+   disregarded; acknowledge output is deasserted; all other signals are
+   treated as usual; if 1 - normal activity. */
+#define XCM_REG_XCM_TM_IFEN					 0x20020
+/* [RW 1] CM - QM Interface enable. If 0 - the acknowledge input is
+   disregarded; valid is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XCM_XQM_IFEN					 0x2000c
+/* [RW 1] If set the Q index; received from the QM is inserted to event ID. */
+#define XCM_REG_XCM_XQM_USE_Q					 0x200f0
+/* [RW 4] The value by which CFC updates the activity counter at QM bypass. */
+#define XCM_REG_XQM_BYP_ACT_UPD 				 0x200fc
+/* [RW 6] QM output initial credit. Max credit available - 32.Write writes
+   the initial credit value; read returns the current value of the credit
+   counter. Must be initialized to 32 at start-up. */
+#define XCM_REG_XQM_INIT_CRD					 0x20420
+/* [RW 3] The weight of the QM (primary) input in the WRR mechanism. 0
+   stands for weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_XQM_P_WEIGHT					 0x200e4
+/* [RW 28] The CM header value for QM request (primary). */
+#define XCM_REG_XQM_XCM_HDR_P					 0x200a0
+/* [RW 28] The CM header value for QM request (secondary). */
+#define XCM_REG_XQM_XCM_HDR_S					 0x200a4
+/* [RW 1] QM - CM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XQM_XCM_IFEN					 0x20014
+/* [RW 1] Input SDM Interface enable. If 0 - the valid input is disregarded;
+   acknowledge output is deasserted; all other signals are treated as usual;
+   if 1 - normal activity. */
+#define XCM_REG_XSDM_IFEN					 0x20018
+/* [RC 1] Set at message length mismatch (relative to last indication) at
+   the SDM interface. */
+#define XCM_REG_XSDM_LENGTH_MIS 				 0x20220
+/* [RW 3] The weight of the SDM input in the WRR mechanism. 0 stands for
+   weight 8 (the most prioritised); 1 stands for weight 1(least
+   prioritised); 2 stands for weight 2; tc. */
+#define XCM_REG_XSDM_WEIGHT					 0x200e0
+/* [RW 17] Indirect access to the descriptor table of the XX protection
+   mechanism. The fields are: [5:0] - message length; 11:6] - message
+   pointer; 16:12] - next pointer. */
+#define XCM_REG_XX_DESCR_TABLE					 0x20480
+/* [R 6] Used to read the XX protection Free counter. */
+#define XCM_REG_XX_FREE 					 0x20240
+/* [RW 6] Initial value for the credit counter; responsible for fulfilling
+   of the Input Stage XX protection buffer by the XX protection pending
+   messages. Max credit available - 3.Write writes the initial credit value;
+   read returns the current value of the credit counter. Must be initialized
+   to 2 at start-up. */
+#define XCM_REG_XX_INIT_CRD					 0x20424
+/* [RW 6] The maximum number of pending messages; which may be stored in XX
+   protection. ~xcm_registers_xx_free.xx_free read on read. */
+#define XCM_REG_XX_MSG_NUM					 0x20428
+/* [RW 8] The Event ID; sent to the STORM in case of XX overflow. */
+#define XCM_REG_XX_OVFL_EVNT_ID 				 0x20058
+/* [RW 15] Indirect access to the XX table of the XX protection mechanism.
+   The fields are:[4:0] - tail pointer; 9:5] - Link List size; 14:10] -
+   header pointer. */
+#define XCM_REG_XX_TABLE					 0x20500
+/* [RW 8] The event id for aggregated interrupt 0 */
+#define XSDM_REG_AGG_INT_EVENT_0				 0x166038
+#define XSDM_REG_AGG_INT_EVENT_1				 0x16603c
+#define XSDM_REG_AGG_INT_EVENT_10				 0x166060
+#define XSDM_REG_AGG_INT_EVENT_11				 0x166064
+#define XSDM_REG_AGG_INT_EVENT_12				 0x166068
+#define XSDM_REG_AGG_INT_EVENT_13				 0x16606c
+#define XSDM_REG_AGG_INT_EVENT_14				 0x166070
+#define XSDM_REG_AGG_INT_EVENT_15				 0x166074
+#define XSDM_REG_AGG_INT_EVENT_16				 0x166078
+#define XSDM_REG_AGG_INT_EVENT_17				 0x16607c
+#define XSDM_REG_AGG_INT_EVENT_18				 0x166080
+#define XSDM_REG_AGG_INT_EVENT_19				 0x166084
+#define XSDM_REG_AGG_INT_EVENT_2				 0x166040
+#define XSDM_REG_AGG_INT_EVENT_20				 0x166088
+#define XSDM_REG_AGG_INT_EVENT_21				 0x16608c
+#define XSDM_REG_AGG_INT_EVENT_22				 0x166090
+#define XSDM_REG_AGG_INT_EVENT_23				 0x166094
+#define XSDM_REG_AGG_INT_EVENT_24				 0x166098
+#define XSDM_REG_AGG_INT_EVENT_25				 0x16609c
+#define XSDM_REG_AGG_INT_EVENT_26				 0x1660a0
+#define XSDM_REG_AGG_INT_EVENT_27				 0x1660a4
+#define XSDM_REG_AGG_INT_EVENT_28				 0x1660a8
+#define XSDM_REG_AGG_INT_EVENT_29				 0x1660ac
+/* [RW 1] For each aggregated interrupt index whether the mode is normal (0)
+   or auto-mask-mode (1) */
+#define XSDM_REG_AGG_INT_MODE_0 				 0x1661b8
+#define XSDM_REG_AGG_INT_MODE_1 				 0x1661bc
+#define XSDM_REG_AGG_INT_MODE_10				 0x1661e0
+#define XSDM_REG_AGG_INT_MODE_11				 0x1661e4
+#define XSDM_REG_AGG_INT_MODE_12				 0x1661e8
+#define XSDM_REG_AGG_INT_MODE_13				 0x1661ec
+#define XSDM_REG_AGG_INT_MODE_14				 0x1661f0
+#define XSDM_REG_AGG_INT_MODE_15				 0x1661f4
+#define XSDM_REG_AGG_INT_MODE_16				 0x1661f8
+#define XSDM_REG_AGG_INT_MODE_17				 0x1661fc
+#define XSDM_REG_AGG_INT_MODE_18				 0x166200
+#define XSDM_REG_AGG_INT_MODE_19				 0x166204
+/* [RW 13] The start address in the internal RAM for the cfc_rsp lcid */
+#define XSDM_REG_CFC_RSP_START_ADDR				 0x166008
+/* [RW 16] The maximum value of the competion counter #0 */
+#define XSDM_REG_CMP_COUNTER_MAX0				 0x16601c
+/* [RW 16] The maximum value of the competion counter #1 */
+#define XSDM_REG_CMP_COUNTER_MAX1				 0x166020
+/* [RW 16] The maximum value of the competion counter #2 */
+#define XSDM_REG_CMP_COUNTER_MAX2				 0x166024
+/* [RW 16] The maximum value of the competion counter #3 */
+#define XSDM_REG_CMP_COUNTER_MAX3				 0x166028
+/* [RW 13] The start address in the internal RAM for the completion
+   counters. */
+#define XSDM_REG_CMP_COUNTER_START_ADDR 			 0x16600c
+#define XSDM_REG_ENABLE_IN1					 0x166238
+#define XSDM_REG_ENABLE_IN2					 0x16623c
+#define XSDM_REG_ENABLE_OUT1					 0x166240
+#define XSDM_REG_ENABLE_OUT2					 0x166244
+/* [RW 4] The initial number of messages that can be sent to the pxp control
+   interface without receiving any ACK. */
+#define XSDM_REG_INIT_CREDIT_PXP_CTRL				 0x1664bc
+/* [ST 32] The number of ACK after placement messages received */
+#define XSDM_REG_NUM_OF_ACK_AFTER_PLACE 			 0x16627c
+/* [ST 32] The number of packet end messages received from the parser */
+#define XSDM_REG_NUM_OF_PKT_END_MSG				 0x166274
+/* [ST 32] The number of requests received from the pxp async if */
+#define XSDM_REG_NUM_OF_PXP_ASYNC_REQ				 0x166278
+/* [ST 32] The number of commands received in queue 0 */
+#define XSDM_REG_NUM_OF_Q0_CMD					 0x166248
+/* [ST 32] The number of commands received in queue 10 */
+#define XSDM_REG_NUM_OF_Q10_CMD 				 0x16626c
+/* [ST 32] The number of commands received in queue 11 */
+#define XSDM_REG_NUM_OF_Q11_CMD 				 0x166270
+/* [ST 32] The number of commands received in queue 1 */
+#define XSDM_REG_NUM_OF_Q1_CMD					 0x16624c
+/* [ST 32] The number of commands received in queue 3 */
+#define XSDM_REG_NUM_OF_Q3_CMD					 0x166250
+/* [ST 32] The number of commands received in queue 4 */
+#define XSDM_REG_NUM_OF_Q4_CMD					 0x166254
+/* [ST 32] The number of commands received in queue 5 */
+#define XSDM_REG_NUM_OF_Q5_CMD					 0x166258
+/* [ST 32] The number of commands received in queue 6 */
+#define XSDM_REG_NUM_OF_Q6_CMD					 0x16625c
+/* [ST 32] The number of commands received in queue 7 */
+#define XSDM_REG_NUM_OF_Q7_CMD					 0x166260
+/* [ST 32] The number of commands received in queue 8 */
+#define XSDM_REG_NUM_OF_Q8_CMD					 0x166264
+/* [ST 32] The number of commands received in queue 9 */
+#define XSDM_REG_NUM_OF_Q9_CMD					 0x166268
+/* [RW 13] The start address in the internal RAM for queue counters */
+#define XSDM_REG_Q_COUNTER_START_ADDR				 0x166010
+/* [R 1] pxp_ctrl rd_data fifo empty in sdm_dma_rsp block */
+#define XSDM_REG_RSP_PXP_CTRL_RDATA_EMPTY			 0x166548
+/* [R 1] parser fifo empty in sdm_sync block */
+#define XSDM_REG_SYNC_PARSER_EMPTY				 0x166550
+/* [R 1] parser serial fifo empty in sdm_sync block */
+#define XSDM_REG_SYNC_SYNC_EMPTY				 0x166558
+/* [RW 32] Tick for timer counter. Applicable only when
+   ~xsdm_registers_timer_tick_enable.timer_tick_enable =1 */
+#define XSDM_REG_TIMER_TICK					 0x166000
+/* [RW 32] Interrupt mask register #0 read/write */
+#define XSDM_REG_XSDM_INT_MASK_0				 0x16629c
+#define XSDM_REG_XSDM_INT_MASK_1				 0x1662ac
+/* [RW 11] Parity mask register #0 read/write */
+#define XSDM_REG_XSDM_PRTY_MASK 				 0x1662bc
+/* [RW 5] The number of time_slots in the arbitration cycle */
+#define XSEM_REG_ARB_CYCLE_SIZE 				 0x280034
+/* [RW 3] The source that is associated with arbitration element 0. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2 */
+#define XSEM_REG_ARB_ELEMENT0					 0x280020
+/* [RW 3] The source that is associated with arbitration element 1. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~xsem_registers_arb_element0.arb_element0 */
+#define XSEM_REG_ARB_ELEMENT1					 0x280024
+/* [RW 3] The source that is associated with arbitration element 2. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~xsem_registers_arb_element0.arb_element0
+   and ~xsem_registers_arb_element1.arb_element1 */
+#define XSEM_REG_ARB_ELEMENT2					 0x280028
+/* [RW 3] The source that is associated with arbitration element 3. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.Could
+   not be equal to register ~xsem_registers_arb_element0.arb_element0 and
+   ~xsem_registers_arb_element1.arb_element1 and
+   ~xsem_registers_arb_element2.arb_element2 */
+#define XSEM_REG_ARB_ELEMENT3					 0x28002c
+/* [RW 3] The source that is associated with arbitration element 4. Source
+   decoding is: 0- foc0; 1-fic1; 2-sleeping thread with priority 0; 3-
+   sleeping thread with priority 1; 4- sleeping thread with priority 2.
+   Could not be equal to register ~xsem_registers_arb_element0.arb_element0
+   and ~xsem_registers_arb_element1.arb_element1 and
+   ~xsem_registers_arb_element2.arb_element2 and
+   ~xsem_registers_arb_element3.arb_element3 */
+#define XSEM_REG_ARB_ELEMENT4					 0x280030
+#define XSEM_REG_ENABLE_IN					 0x2800a4
+#define XSEM_REG_ENABLE_OUT					 0x2800a8
+/* [RW 32] This address space contains all registers and memories that are
+   placed in SEM_FAST block. The SEM_FAST registers are described in
+   appendix B. In order to access the SEM_FAST registers the base address
+   XSEM_REGISTERS_FAST_MEMORY (Offset: 0x2a0000) should be added to each
+   SEM_FAST register offset. */
+#define XSEM_REG_FAST_MEMORY					 0x2a0000
+/* [RW 1] Disables input messages from FIC0 May be updated during run_time
+   by the microcode */
+#define XSEM_REG_FIC0_DISABLE					 0x280224
+/* [RW 1] Disables input messages from FIC1 May be updated during run_time
+   by the microcode */
+#define XSEM_REG_FIC1_DISABLE					 0x280234
+/* [RW 15] Interrupt table Read and write access to it is not possible in
+   the middle of the work */
+#define XSEM_REG_INT_TABLE					 0x280400
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC0 */
+#define XSEM_REG_MSG_NUM_FIC0					 0x280000
+/* [ST 24] Statistics register. The number of messages that entered through
+   FIC1 */
+#define XSEM_REG_MSG_NUM_FIC1					 0x280004
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC0 */
+#define XSEM_REG_MSG_NUM_FOC0					 0x280008
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC1 */
+#define XSEM_REG_MSG_NUM_FOC1					 0x28000c
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC2 */
+#define XSEM_REG_MSG_NUM_FOC2					 0x280010
+/* [ST 24] Statistics register. The number of messages that were sent to
+   FOC3 */
+#define XSEM_REG_MSG_NUM_FOC3					 0x280014
+/* [RW 1] Disables input messages from the passive buffer May be updated
+   during run_time by the microcode */
+#define XSEM_REG_PAS_DISABLE					 0x28024c
+/* [WB 128] Debug only. Passive buffer memory */
+#define XSEM_REG_PASSIVE_BUFFER 				 0x282000
+/* [WB 46] pram memory. B45 is parity; b[44:0] - data. */
+#define XSEM_REG_PRAM						 0x2c0000
+/* [R 16] Valid sleeping threads indication have bit per thread */
+#define XSEM_REG_SLEEP_THREADS_VALID				 0x28026c
+/* [R 1] EXT_STORE FIFO is empty in sem_slow_ls_ext */
+#define XSEM_REG_SLOW_EXT_STORE_EMPTY				 0x2802a0
+/* [RW 16] List of free threads . There is a bit per thread. */
+#define XSEM_REG_THREADS_LIST					 0x2802e4
+/* [RW 3] The arbitration scheme of time_slot 0 */
+#define XSEM_REG_TS_0_AS					 0x280038
+/* [RW 3] The arbitration scheme of time_slot 10 */
+#define XSEM_REG_TS_10_AS					 0x280060
+/* [RW 3] The arbitration scheme of time_slot 11 */
+#define XSEM_REG_TS_11_AS					 0x280064
+/* [RW 3] The arbitration scheme of time_slot 12 */
+#define XSEM_REG_TS_12_AS					 0x280068
+/* [RW 3] The arbitration scheme of time_slot 13 */
+#define XSEM_REG_TS_13_AS					 0x28006c
+/* [RW 3] The arbitration scheme of time_slot 14 */
+#define XSEM_REG_TS_14_AS					 0x280070
+/* [RW 3] The arbitration scheme of time_slot 15 */
+#define XSEM_REG_TS_15_AS					 0x280074
+/* [RW 3] The arbitration scheme of time_slot 16 */
+#define XSEM_REG_TS_16_AS					 0x280078
+/* [RW 3] The arbitration scheme of time_slot 17 */
+#define XSEM_REG_TS_17_AS					 0x28007c
+/* [RW 3] The arbitration scheme of time_slot 18 */
+#define XSEM_REG_TS_18_AS					 0x280080
+/* [RW 3] The arbitration scheme of time_slot 1 */
+#define XSEM_REG_TS_1_AS					 0x28003c
+/* [RW 3] The arbitration scheme of time_slot 2 */
+#define XSEM_REG_TS_2_AS					 0x280040
+/* [RW 3] The arbitration scheme of time_slot 3 */
+#define XSEM_REG_TS_3_AS					 0x280044
+/* [RW 3] The arbitration scheme of time_slot 4 */
+#define XSEM_REG_TS_4_AS					 0x280048
+/* [RW 3] The arbitration scheme of time_slot 5 */
+#define XSEM_REG_TS_5_AS					 0x28004c
+/* [RW 3] The arbitration scheme of time_slot 6 */
+#define XSEM_REG_TS_6_AS					 0x280050
+/* [RW 3] The arbitration scheme of time_slot 7 */
+#define XSEM_REG_TS_7_AS					 0x280054
+/* [RW 3] The arbitration scheme of time_slot 8 */
+#define XSEM_REG_TS_8_AS					 0x280058
+/* [RW 3] The arbitration scheme of time_slot 9 */
+#define XSEM_REG_TS_9_AS					 0x28005c
+/* [RW 32] Interrupt mask register #0 read/write */
+#define XSEM_REG_XSEM_INT_MASK_0				 0x280110
+#define XSEM_REG_XSEM_INT_MASK_1				 0x280120
+/* [RW 32] Parity mask register #0 read/write */
+#define XSEM_REG_XSEM_PRTY_MASK_0				 0x280130
+#define XSEM_REG_XSEM_PRTY_MASK_1				 0x280140
+#define MCPR_NVM_ACCESS_ENABLE_EN				 (1L<<0)
+#define MCPR_NVM_ACCESS_ENABLE_WR_EN				 (1L<<1)
+#define MCPR_NVM_ADDR_NVM_ADDR_VALUE				 (0xffffffL<<0)
+#define MCPR_NVM_CFG4_FLASH_SIZE				 (0x7L<<0)
+#define MCPR_NVM_COMMAND_DOIT					 (1L<<4)
+#define MCPR_NVM_COMMAND_DONE					 (1L<<3)
+#define MCPR_NVM_COMMAND_FIRST					 (1L<<7)
+#define MCPR_NVM_COMMAND_LAST					 (1L<<8)
+#define MCPR_NVM_COMMAND_WR					 (1L<<5)
+#define MCPR_NVM_COMMAND_WREN					 (1L<<16)
+#define MCPR_NVM_COMMAND_WREN_BITSHIFT				 16
+#define MCPR_NVM_COMMAND_WRDI					 (1L<<17)
+#define MCPR_NVM_COMMAND_WRDI_BITSHIFT				 17
+#define MCPR_NVM_SW_ARB_ARB_ARB1				 (1L<<9)
+#define MCPR_NVM_SW_ARB_ARB_REQ_CLR1				 (1L<<5)
+#define MCPR_NVM_SW_ARB_ARB_REQ_SET1				 (1L<<1)
+#define BIGMAC_REGISTER_BMAC_CONTROL				 (0x00<<3)
+#define BIGMAC_REGISTER_BMAC_XGXS_CONTROL			 (0x01<<3)
+#define BIGMAC_REGISTER_CNT_MAX_SIZE				 (0x05<<3)
+#define BIGMAC_REGISTER_RX_CONTROL				 (0x21<<3)
+#define BIGMAC_REGISTER_RX_LLFC_MSG_FLDS			 (0x46<<3)
+#define BIGMAC_REGISTER_RX_MAX_SIZE				 (0x23<<3)
+#define BIGMAC_REGISTER_RX_STAT_GR64				 (0x26<<3)
+#define BIGMAC_REGISTER_RX_STAT_GRIPJ				 (0x42<<3)
+#define BIGMAC_REGISTER_TX_CONTROL				 (0x07<<3)
+#define BIGMAC_REGISTER_TX_MAX_SIZE				 (0x09<<3)
+#define BIGMAC_REGISTER_TX_PAUSE_THRESHOLD			 (0x0A<<3)
+#define BIGMAC_REGISTER_TX_SOURCE_ADDR				 (0x08<<3)
+#define BIGMAC_REGISTER_TX_STAT_GTBYT				 (0x20<<3)
+#define BIGMAC_REGISTER_TX_STAT_GTPKT				 (0x0C<<3)
+#define EMAC_MDIO_COMM_COMMAND_ADDRESS				 (0L<<26)
+#define EMAC_MDIO_COMM_COMMAND_READ_22				 (2L<<26)
+#define EMAC_MDIO_COMM_COMMAND_READ_45				 (3L<<26)
+#define EMAC_MDIO_COMM_COMMAND_WRITE_22 			 (1L<<26)
+#define EMAC_MDIO_COMM_COMMAND_WRITE_45 			 (1L<<26)
+#define EMAC_MDIO_COMM_DATA					 (0xffffL<<0)
+#define EMAC_MDIO_COMM_START_BUSY				 (1L<<29)
+#define EMAC_MDIO_MODE_AUTO_POLL				 (1L<<4)
+#define EMAC_MDIO_MODE_CLAUSE_45				 (1L<<31)
+#define EMAC_MODE_25G_MODE					 (1L<<5)
+#define EMAC_MODE_ACPI_RCVD					 (1L<<20)
+#define EMAC_MODE_HALF_DUPLEX					 (1L<<1)
+#define EMAC_MODE_MPKT						 (1L<<18)
+#define EMAC_MODE_MPKT_RCVD					 (1L<<19)
+#define EMAC_MODE_PORT_GMII					 (2L<<2)
+#define EMAC_MODE_PORT_MII					 (1L<<2)
+#define EMAC_MODE_PORT_MII_10M					 (3L<<2)
+#define EMAC_MODE_RESET 					 (1L<<0)
+#define EMAC_REG_EMAC_MAC_MATCH 				 0x10
+#define EMAC_REG_EMAC_MDIO_COMM 				 0xac
+#define EMAC_REG_EMAC_MDIO_MODE 				 0xb4
+#define EMAC_REG_EMAC_MODE					 0x0
+#define EMAC_REG_EMAC_RX_MODE					 0xc8
+#define EMAC_REG_EMAC_RX_MTU_SIZE				 0x9c
+#define EMAC_REG_EMAC_RX_STAT_AC				 0x180
+#define EMAC_REG_EMAC_RX_STAT_AC_28				 0x1f4
+#define EMAC_REG_EMAC_RX_STAT_AC_COUNT				 23
+#define EMAC_REG_EMAC_TX_MODE					 0xbc
+#define EMAC_REG_EMAC_TX_STAT_AC				 0x280
+#define EMAC_REG_EMAC_TX_STAT_AC_COUNT				 22
+#define EMAC_RX_MODE_FLOW_EN					 (1L<<2)
+#define EMAC_RX_MODE_KEEP_VLAN_TAG				 (1L<<10)
+#define EMAC_RX_MODE_PROMISCUOUS				 (1L<<8)
+#define EMAC_RX_MTU_SIZE_JUMBO_ENA				 (1L<<31)
+#define EMAC_TX_MODE_EXT_PAUSE_EN				 (1L<<3)
+#define EMAC_TX_MODE_RESET					 (1L<<0)
+#define MISC_REGISTERS_RESET_REG_1_CLEAR			 0x588
+#define MISC_REGISTERS_RESET_REG_1_SET				 0x584
+#define MISC_REGISTERS_RESET_REG_2_CLEAR			 0x598
+#define MISC_REGISTERS_RESET_REG_2_RST_BMAC0			 (0x1<<0)
+#define MISC_REGISTERS_RESET_REG_2_RST_EMAC0_HARD_CORE		 (0x1<<14)
+#define MISC_REGISTERS_RESET_REG_2_SET				 0x594
+#define MISC_REGISTERS_RESET_REG_3_CLEAR			 0x5a8
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_IDDQ	 (0x1<<1)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_PWRDWN	 (0x1<<2)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_PWRDWN_SD (0x1<<3)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_SERDES0_RSTB_HW  (0x1<<0)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_IDDQ	 (0x1<<5)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_PWRDWN	 (0x1<<6)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_PWRDWN_SD  (0x1<<7)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_RSTB_HW	 (0x1<<4)
+#define MISC_REGISTERS_RESET_REG_3_MISC_NIG_MUX_XGXS0_TXD_FIFO_RSTB (0x1<<8)
+#define MISC_REGISTERS_RESET_REG_3_SET				 0x5a4
+#define AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR		      (1<<18)
+#define AEU_INPUTS_ATTN_BITS_CCM_HW_INTERRUPT		      (1<<31)
+#define AEU_INPUTS_ATTN_BITS_CDU_HW_INTERRUPT		      (1<<9)
+#define AEU_INPUTS_ATTN_BITS_CDU_PARITY_ERROR		      (1<<8)
+#define AEU_INPUTS_ATTN_BITS_CFC_HW_INTERRUPT		      (1<<7)
+#define AEU_INPUTS_ATTN_BITS_CFC_PARITY_ERROR		      (1<<6)
+#define AEU_INPUTS_ATTN_BITS_CSDM_HW_INTERRUPT		      (1<<29)
+#define AEU_INPUTS_ATTN_BITS_CSDM_PARITY_ERROR		      (1<<28)
+#define AEU_INPUTS_ATTN_BITS_CSEMI_HW_INTERRUPT 	      (1<<1)
+#define AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR 	      (1<<0)
+#define AEU_INPUTS_ATTN_BITS_DEBUG_PARITY_ERROR 	      (1<<18)
+#define AEU_INPUTS_ATTN_BITS_DMAE_HW_INTERRUPT		      (1<<11)
+#define AEU_INPUTS_ATTN_BITS_DOORBELLQ_HW_INTERRUPT	      (1<<13)
+#define AEU_INPUTS_ATTN_BITS_DOORBELLQ_PARITY_ERROR	      (1<<12)
+#define AEU_INPUTS_ATTN_BITS_IGU_PARITY_ERROR		      (1<<12)
+#define AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT		      (1<<15)
+#define AEU_INPUTS_ATTN_BITS_MISC_PARITY_ERROR		      (1<<14)
+#define AEU_INPUTS_ATTN_BITS_PARSER_PARITY_ERROR	      (1<<20)
+#define AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR	      (1<<0)
+#define AEU_INPUTS_ATTN_BITS_PBF_HW_INTERRUPT		      (1<<31)
+#define AEU_INPUTS_ATTN_BITS_PXP_HW_INTERRUPT		      (1<<3)
+#define AEU_INPUTS_ATTN_BITS_PXP_PARITY_ERROR		      (1<<2)
+#define AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_HW_INTERRUPT   (1<<5)
+#define AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_PARITY_ERROR   (1<<4)
+#define AEU_INPUTS_ATTN_BITS_QM_HW_INTERRUPT		      (1<<3)
+#define AEU_INPUTS_ATTN_BITS_QM_PARITY_ERROR		      (1<<2)
+#define AEU_INPUTS_ATTN_BITS_SEARCHER_PARITY_ERROR	      (1<<22)
+#define AEU_INPUTS_ATTN_BITS_TCM_HW_INTERRUPT		      (1<<27)
+#define AEU_INPUTS_ATTN_BITS_TIMERS_HW_INTERRUPT	      (1<<5)
+#define AEU_INPUTS_ATTN_BITS_TSDM_HW_INTERRUPT		      (1<<25)
+#define AEU_INPUTS_ATTN_BITS_TSDM_PARITY_ERROR		      (1<<24)
+#define AEU_INPUTS_ATTN_BITS_TSEMI_HW_INTERRUPT 	      (1<<29)
+#define AEU_INPUTS_ATTN_BITS_TSEMI_PARITY_ERROR 	      (1<<28)
+#define AEU_INPUTS_ATTN_BITS_UCM_HW_INTERRUPT		      (1<<23)
+#define AEU_INPUTS_ATTN_BITS_UPB_HW_INTERRUPT		      (1<<27)
+#define AEU_INPUTS_ATTN_BITS_UPB_PARITY_ERROR		      (1<<26)
+#define AEU_INPUTS_ATTN_BITS_USDM_HW_INTERRUPT		      (1<<21)
+#define AEU_INPUTS_ATTN_BITS_USDM_PARITY_ERROR		      (1<<20)
+#define AEU_INPUTS_ATTN_BITS_USEMI_HW_INTERRUPT 	      (1<<25)
+#define AEU_INPUTS_ATTN_BITS_USEMI_PARITY_ERROR 	      (1<<24)
+#define AEU_INPUTS_ATTN_BITS_VAUX_PCI_CORE_PARITY_ERROR       (1<<16)
+#define AEU_INPUTS_ATTN_BITS_XCM_HW_INTERRUPT		      (1<<9)
+#define AEU_INPUTS_ATTN_BITS_XSDM_HW_INTERRUPT		      (1<<7)
+#define AEU_INPUTS_ATTN_BITS_XSDM_PARITY_ERROR		      (1<<6)
+#define AEU_INPUTS_ATTN_BITS_XSEMI_HW_INTERRUPT 	      (1<<11)
+#define AEU_INPUTS_ATTN_BITS_XSEMI_PARITY_ERROR 	      (1<<10)
+#define RESERVED_GENERAL_ATTENTION_BIT_0	0
+
+#define EVEREST_GEN_ATTN_IN_USE_MASK		0x3e0
+#define EVEREST_LATCHED_ATTN_IN_USE_MASK	0xffe00000
+
+#define RESERVED_GENERAL_ATTENTION_BIT_6	6
+#define RESERVED_GENERAL_ATTENTION_BIT_7	7
+#define RESERVED_GENERAL_ATTENTION_BIT_8	8
+#define RESERVED_GENERAL_ATTENTION_BIT_9	9
+#define RESERVED_GENERAL_ATTENTION_BIT_10	10
+#define RESERVED_GENERAL_ATTENTION_BIT_11	11
+#define RESERVED_GENERAL_ATTENTION_BIT_12	12
+#define RESERVED_GENERAL_ATTENTION_BIT_13	13
+#define RESERVED_GENERAL_ATTENTION_BIT_14	14
+#define RESERVED_GENERAL_ATTENTION_BIT_15	15
+#define RESERVED_GENERAL_ATTENTION_BIT_16	16
+#define RESERVED_GENERAL_ATTENTION_BIT_17	17
+#define RESERVED_GENERAL_ATTENTION_BIT_18	18
+#define RESERVED_GENERAL_ATTENTION_BIT_19	19
+#define RESERVED_GENERAL_ATTENTION_BIT_20	20
+#define RESERVED_GENERAL_ATTENTION_BIT_21	21
+
+/* storm asserts attention bits */
+#define TSTORM_FATAL_ASSERT_ATTENTION_BIT     RESERVED_GENERAL_ATTENTION_BIT_7
+#define USTORM_FATAL_ASSERT_ATTENTION_BIT     RESERVED_GENERAL_ATTENTION_BIT_8
+#define CSTORM_FATAL_ASSERT_ATTENTION_BIT     RESERVED_GENERAL_ATTENTION_BIT_9
+#define XSTORM_FATAL_ASSERT_ATTENTION_BIT     RESERVED_GENERAL_ATTENTION_BIT_10
+
+/* mcp error attention bit */
+#define MCP_FATAL_ASSERT_ATTENTION_BIT	      RESERVED_GENERAL_ATTENTION_BIT_11
+
+#define LATCHED_ATTN_RBCR			23
+#define LATCHED_ATTN_RBCT			24
+#define LATCHED_ATTN_RBCN			25
+#define LATCHED_ATTN_RBCU			26
+#define LATCHED_ATTN_RBCP			27
+#define LATCHED_ATTN_TIMEOUT_GRC		28
+#define LATCHED_ATTN_RSVD_GRC			29
+#define LATCHED_ATTN_ROM_PARITY_MCP		30
+#define LATCHED_ATTN_UM_RX_PARITY_MCP		31
+#define LATCHED_ATTN_UM_TX_PARITY_MCP		32
+#define LATCHED_ATTN_SCPAD_PARITY_MCP		33
+
+#define GENERAL_ATTEN_WORD(atten_name)	       ((94 + atten_name) / 32)
+#define GENERAL_ATTEN_OFFSET(atten_name)       (1 << ((94 + atten_name) % 32))
+/*
+ * This file defines GRC base address for every block.
+ * This file is included by chipsim, asm microcode and cpp microcode.
+ * These values are used in Design.xml on regBase attribute
+ * Use the base with the generated offsets of specific registers.
+ */
+
+#define GRCBASE_PXPCS		0x000000
+#define GRCBASE_PCICONFIG	0x002000
+#define GRCBASE_PCIREG		0x002400
+#define GRCBASE_EMAC0		0x008000
+#define GRCBASE_EMAC1		0x008400
+#define GRCBASE_DBU		0x008800
+#define GRCBASE_MISC		0x00A000
+#define GRCBASE_DBG		0x00C000
+#define GRCBASE_NIG		0x010000
+#define GRCBASE_XCM		0x020000
+#define GRCBASE_PRS		0x040000
+#define GRCBASE_SRCH		0x040400
+#define GRCBASE_TSDM		0x042000
+#define GRCBASE_TCM		0x050000
+#define GRCBASE_BRB1		0x060000
+#define GRCBASE_MCP		0x080000
+#define GRCBASE_UPB		0x0C1000
+#define GRCBASE_CSDM		0x0C2000
+#define GRCBASE_USDM		0x0C4000
+#define GRCBASE_CCM		0x0D0000
+#define GRCBASE_UCM		0x0E0000
+#define GRCBASE_CDU		0x101000
+#define GRCBASE_DMAE		0x102000
+#define GRCBASE_PXP		0x103000
+#define GRCBASE_CFC		0x104000
+#define GRCBASE_HC		0x108000
+#define GRCBASE_PXP2		0x120000
+#define GRCBASE_PBF		0x140000
+#define GRCBASE_XPB		0x161000
+#define GRCBASE_TIMERS		0x164000
+#define GRCBASE_XSDM		0x166000
+#define GRCBASE_QM		0x168000
+#define GRCBASE_DQ		0x170000
+#define GRCBASE_TSEM		0x180000
+#define GRCBASE_CSEM		0x200000
+#define GRCBASE_XSEM		0x280000
+#define GRCBASE_USEM		0x300000
+#define GRCBASE_MISC_AEU	GRCBASE_MISC
+
+
+/*the offset of the configuration space in the pci core register*/
+#define PCICFG_OFFSET					0x2000
+#define PCICFG_VENDOR_ID_OFFSET 			0x00
+#define PCICFG_DEVICE_ID_OFFSET 			0x02
+#define PCICFG_SUBSYSTEM_VENDOR_ID_OFFSET		0x2c
+#define PCICFG_SUBSYSTEM_ID_OFFSET			0x2e
+#define PCICFG_INT_LINE 				0x3c
+#define PCICFG_INT_PIN					0x3d
+#define PCICFG_CACHE_LINE_SIZE				0x0c
+#define PCICFG_LATENCY_TIMER				0x0d
+#define PCICFG_REVESION_ID				0x08
+#define PCICFG_BAR_1_LOW				0x10
+#define PCICFG_BAR_1_HIGH				0x14
+#define PCICFG_BAR_2_LOW				0x18
+#define PCICFG_BAR_2_HIGH				0x1c
+#define PCICFG_GRC_ADDRESS				0x78
+#define PCICFG_GRC_DATA 				0x80
+#define PCICFG_DEVICE_CONTROL				0xb4
+#define PCICFG_LINK_CONTROL				0xbc
+
+#define BAR_USTRORM_INTMEM				0x400000
+#define BAR_CSTRORM_INTMEM				0x410000
+#define BAR_XSTRORM_INTMEM				0x420000
+#define BAR_TSTRORM_INTMEM				0x430000
+
+#define BAR_IGU_INTMEM					0x440000
+
+#define BAR_DOORBELL_OFFSET				0x800000
+
+#define BAR_ME_REGISTER 				0x450000
+
+
+#define GRC_CONFIG_2_SIZE_REG		    0x408 /* config_2 offset */
+#define PCI_CONFIG_2_BAR1_SIZE			    (0xfL<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_DISABLED 	(0L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_64K		(1L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_128K		(2L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_256K		(3L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_512K		(4L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_1M		(5L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_2M		(6L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_4M		(7L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_8M		(8L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_16M		(9L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_32M		(10L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_64M		(11L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_128M		(12L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_256M		(13L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_512M		(14L<<0)
+#define PCI_CONFIG_2_BAR1_SIZE_1G		(15L<<0)
+#define PCI_CONFIG_2_BAR1_64ENA 		    (1L<<4)
+#define PCI_CONFIG_2_EXP_ROM_RETRY		    (1L<<5)
+#define PCI_CONFIG_2_CFG_CYCLE_RETRY		    (1L<<6)
+#define PCI_CONFIG_2_FIRST_CFG_DONE		    (1L<<7)
+#define PCI_CONFIG_2_EXP_ROM_SIZE		    (0xffL<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_DISABLED	(0L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_2K		(1L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_4K		(2L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_8K		(3L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_16K		(4L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_32K		(5L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_64K		(6L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_128K		(7L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_256K		(8L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_512K		(9L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_1M		(10L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_2M		(11L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_4M		(12L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_8M		(13L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_16M		(14L<<8)
+#define PCI_CONFIG_2_EXP_ROM_SIZE_32M		(15L<<8)
+#define PCI_CONFIG_2_BAR_PREFETCH		    (1L<<16)
+#define PCI_CONFIG_2_RESERVED0			    (0x7fffL<<17)
+
+/* config_3 offset */
+#define GRC_CONFIG_3_SIZE_REG				(0x40c)
+#define PCI_CONFIG_3_STICKY_BYTE		    (0xffL<<0)
+#define PCI_CONFIG_3_FORCE_PME			    (1L<<24)
+#define PCI_CONFIG_3_PME_STATUS 		    (1L<<25)
+#define PCI_CONFIG_3_PME_ENABLE 		    (1L<<26)
+#define PCI_CONFIG_3_PM_STATE			    (0x3L<<27)
+#define PCI_CONFIG_3_VAUX_PRESET		    (1L<<30)
+#define PCI_CONFIG_3_PCI_POWER			    (1L<<31)
+
+/* config_2 offset */
+#define GRC_CONFIG_2_SIZE_REG		    0x408
+
+#define GRC_BAR2_CONFIG 				0x4e0
+#define PCI_CONFIG_2_BAR2_SIZE			    (0xfL<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_DISABLED 	    (0L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_64K		    (1L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_128K		    (2L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_256K		    (3L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_512K		    (4L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_1M		    (5L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_2M		    (6L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_4M		    (7L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_8M		    (8L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_16M		    (9L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_32M		    (10L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_64M		    (11L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_128M		    (12L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_256M		    (13L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_512M		    (14L<<0)
+#define PCI_CONFIG_2_BAR2_SIZE_1G		    (15L<<0)
+#define PCI_CONFIG_2_BAR2_64ENA 		    (1L<<4)
+
+#define PCI_PM_DATA_A					(0x410)
+#define PCI_PM_DATA_B					(0x414)
+#define PCI_ID_VAL1					(0x434)
+#define PCI_ID_VAL2					(0x438)
+
+#define MDIO_REG_BANK_CL73_IEEEB0			0x0
+#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL		0x0
+#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL_RESTART_AN	0x0200
+#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL_AN_EN		0x1000
+#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL_MAIN_RST	0x8000
+
+#define MDIO_REG_BANK_CL73_IEEEB1			0x10
+#define MDIO_CL73_IEEEB1_AN_ADV2			0x01
+#define MDIO_CL73_IEEEB1_AN_ADV2_ADVR_1000M		0x0000
+#define MDIO_CL73_IEEEB1_AN_ADV2_ADVR_1000M_KX		0x0020
+#define MDIO_CL73_IEEEB1_AN_ADV2_ADVR_10G_KX4		0x0040
+#define MDIO_CL73_IEEEB1_AN_ADV2_ADVR_10G_KR		0x0080
+
+#define MDIO_REG_BANK_RX0				0x80b0
+#define MDIO_RX0_RX_EQ_BOOST				0x1c
+#define MDIO_RX0_RX_EQ_BOOST_EQUALIZER_CTRL_MASK	0x7
+#define MDIO_RX0_RX_EQ_BOOST_OFFSET_CTRL		0x10
+
+#define MDIO_REG_BANK_RX1				0x80c0
+#define MDIO_RX1_RX_EQ_BOOST				0x1c
+#define MDIO_RX1_RX_EQ_BOOST_EQUALIZER_CTRL_MASK	0x7
+#define MDIO_RX1_RX_EQ_BOOST_OFFSET_CTRL		0x10
+
+#define MDIO_REG_BANK_RX2				0x80d0
+#define MDIO_RX2_RX_EQ_BOOST				0x1c
+#define MDIO_RX2_RX_EQ_BOOST_EQUALIZER_CTRL_MASK	0x7
+#define MDIO_RX2_RX_EQ_BOOST_OFFSET_CTRL		0x10
+
+#define MDIO_REG_BANK_RX3				0x80e0
+#define MDIO_RX3_RX_EQ_BOOST				0x1c
+#define MDIO_RX3_RX_EQ_BOOST_EQUALIZER_CTRL_MASK	0x7
+#define MDIO_RX3_RX_EQ_BOOST_OFFSET_CTRL		0x10
+
+#define MDIO_REG_BANK_RX_ALL				0x80f0
+#define MDIO_RX_ALL_RX_EQ_BOOST 			0x1c
+#define MDIO_RX_ALL_RX_EQ_BOOST_EQUALIZER_CTRL_MASK	0x7
+#define MDIO_RX_ALL_RX_EQ_BOOST_OFFSET_CTRL		0x10
+
+#define MDIO_REG_BANK_TX0				0x8060
+#define MDIO_TX0_TX_DRIVER				0x17
+#define MDIO_TX0_TX_DRIVER_PREEMPHASIS_MASK		0xf000
+#define MDIO_TX0_TX_DRIVER_PREEMPHASIS_SHIFT		12
+#define MDIO_TX0_TX_DRIVER_IDRIVER_MASK 		0x0f00
+#define MDIO_TX0_TX_DRIVER_IDRIVER_SHIFT		8
+#define MDIO_TX0_TX_DRIVER_IPREDRIVER_MASK		0x00f0
+#define MDIO_TX0_TX_DRIVER_IPREDRIVER_SHIFT		4
+#define MDIO_TX0_TX_DRIVER_IFULLSPD_MASK		0x000e
+#define MDIO_TX0_TX_DRIVER_IFULLSPD_SHIFT		1
+#define MDIO_TX0_TX_DRIVER_ICBUF1T			1
+
+#define MDIO_REG_BANK_XGXS_BLOCK0			0x8000
+#define MDIO_BLOCK0_XGXS_CONTROL			0x10
+
+#define MDIO_REG_BANK_XGXS_BLOCK1			0x8010
+#define MDIO_BLOCK1_LANE_CTRL0				0x15
+#define MDIO_BLOCK1_LANE_CTRL1				0x16
+#define MDIO_BLOCK1_LANE_CTRL2				0x17
+#define MDIO_BLOCK1_LANE_PRBS				0x19
+
+#define MDIO_REG_BANK_XGXS_BLOCK2			0x8100
+#define MDIO_XGXS_BLOCK2_RX_LN_SWAP			0x10
+#define MDIO_XGXS_BLOCK2_RX_LN_SWAP_ENABLE		0x8000
+#define MDIO_XGXS_BLOCK2_RX_LN_SWAP_FORCE_ENABLE	0x4000
+#define MDIO_XGXS_BLOCK2_TX_LN_SWAP			0x11
+#define MDIO_XGXS_BLOCK2_TX_LN_SWAP_ENABLE		0x8000
+#define MDIO_XGXS_BLOCK2_TEST_MODE_LANE 		0x15
+
+#define MDIO_REG_BANK_GP_STATUS 			0x8120
+#define MDIO_GP_STATUS_TOP_AN_STATUS1			    0x1B
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_CL73_AUTONEG_COMPLETE 0x0001
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_CL37_AUTONEG_COMPLETE 0x0002
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_LINK_STATUS	    0x0004
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_DUPLEX_STATUS	    0x0008
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_CL73_MR_LP_NP_AN_ABLE 0x0010
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_CL73_LP_NP_BAM_ABLE   0x0020
+
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_PAUSE_RSOLUTION_TXSIDE 0x0040
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_PAUSE_RSOLUTION_RXSIDE 0x0080
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_MASK     0x3f00
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10M	    0x0000
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_100M     0x0100
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_1G	    0x0200
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_2_5G     0x0300
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_5G	    0x0400
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_6G	    0x0500
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_HIG  0x0600
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_CX4  0x0700
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_12G_HIG  0x0800
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_12_5G    0x0900
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_13G	    0x0A00
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_15G	    0x0B00
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_16G	    0x0C00
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_1G_KX    0x0D00
+#define MDIO_GP_STATUS_TOP_AN_STATUS1_ACTUAL_SPEED_10G_KX4  0x0E00
+
+
+#define MDIO_REG_BANK_10G_PARALLEL_DETECT		0x8130
+#define MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL	    0x11
+#define MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_CONTROL_PARDET10G_EN 0x1
+#define MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_LINK	    0x13
+#define MDIO_10G_PARALLEL_DETECT_PAR_DET_10G_LINK_CNT	    (0xb71<<1)
+
+#define MDIO_REG_BANK_SERDES_DIGITAL			0x8300
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1		0x10
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_FIBER_MODE 0x0001
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_TBI_IF	0x0002
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_SIGNAL_DETECT_EN	  0x0004
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_INVERT_SIGNAL_DETECT 0x0008
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_AUTODET	0x0010
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL1_MSTR_MODE	0x0020
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL2		0x11
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL2_PRL_DT_EN	0x0001
+#define MDIO_SERDES_DIGITAL_A_1000X_CONTROL2_AN_FST_TMR 0x0040
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1		0x14
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_DUPLEX	0x0004
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_MASK	0x0018
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_SHIFT 3
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_2_5G	0x0018
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_1G	0x0010
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_100M	0x0008
+#define MDIO_SERDES_DIGITAL_A_1000X_STATUS1_SPEED_10M	0x0000
+#define MDIO_SERDES_DIGITAL_MISC1			0x18
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_MASK	0xE000
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_25M	0x0000
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_100M	0x2000
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_125M	0x4000
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_156_25M	0x6000
+#define MDIO_SERDES_DIGITAL_MISC1_REFCLK_SEL_187_5M	0x8000
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_SEL	0x0010
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_MASK	0x000f
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_2_5G	0x0000
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_5G	0x0001
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_6G	0x0002
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_10G_HIG	0x0003
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_10G_CX4	0x0004
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_12G	0x0005
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_12_5G	0x0006
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_13G	0x0007
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_15G	0x0008
+#define MDIO_SERDES_DIGITAL_MISC1_FORCE_SPEED_16G	0x0009
+
+#define MDIO_REG_BANK_OVER_1G				0x8320
+#define MDIO_OVER_1G_DIGCTL_3_4 			0x14
+#define MDIO_OVER_1G_DIGCTL_3_4_MP_ID_MASK		0xffe0
+#define MDIO_OVER_1G_DIGCTL_3_4_MP_ID_SHIFT		5
+#define MDIO_OVER_1G_UP1				0x19
+#define MDIO_OVER_1G_UP1_2_5G				0x0001
+#define MDIO_OVER_1G_UP1_5G				0x0002
+#define MDIO_OVER_1G_UP1_6G				0x0004
+#define MDIO_OVER_1G_UP1_10G				0x0010
+#define MDIO_OVER_1G_UP1_10GH				0x0008
+#define MDIO_OVER_1G_UP1_12G				0x0020
+#define MDIO_OVER_1G_UP1_12_5G				0x0040
+#define MDIO_OVER_1G_UP1_13G				0x0080
+#define MDIO_OVER_1G_UP1_15G				0x0100
+#define MDIO_OVER_1G_UP1_16G				0x0200
+#define MDIO_OVER_1G_UP2				0x1A
+#define MDIO_OVER_1G_UP2_IPREDRIVER_MASK		0x0007
+#define MDIO_OVER_1G_UP2_IDRIVER_MASK			0x0038
+#define MDIO_OVER_1G_UP2_PREEMPHASIS_MASK		0x03C0
+#define MDIO_OVER_1G_UP3				0x1B
+#define MDIO_OVER_1G_UP3_HIGIG2 			0x0001
+#define MDIO_OVER_1G_LP_UP1				0x1C
+#define MDIO_OVER_1G_LP_UP2				0x1D
+#define MDIO_OVER_1G_LP_UP2_MR_ADV_OVER_1G_MASK 	0x03ff
+#define MDIO_OVER_1G_LP_UP2_PREEMPHASIS_MASK		0x0780
+#define MDIO_OVER_1G_LP_UP2_PREEMPHASIS_SHIFT		7
+#define MDIO_OVER_1G_LP_UP3				0x1E
+
+#define MDIO_REG_BANK_BAM_NEXT_PAGE			0x8350
+#define MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL		0x10
+#define MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_BAM_MODE	0x0001
+#define MDIO_BAM_NEXT_PAGE_MP5_NEXT_PAGE_CTRL_TETON_AN	0x0002
+
+#define MDIO_REG_BANK_CL73_USERB0			0x8370
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL1 		0x12
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_EN			0x8000
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_STATION_MNGR_EN	0x4000
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL1_BAM_NP_AFTER_BP_EN	0x2000
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL3 		0x14
+#define MDIO_CL73_USERB0_CL73_BAM_CTRL3_USE_CL73_HCD_MR 0x0001
+
+#define MDIO_REG_BANK_AER_BLOCK 			0xFFD0
+#define MDIO_AER_BLOCK_AER_REG				0x1E
+
+#define MDIO_REG_BANK_COMBO_IEEE0			0xFFE0
+#define MDIO_COMBO_IEEE0_MII_CONTROL			0x10
+#define MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_MASK	0x2040
+#define MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_10	0x0000
+#define MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_100	0x2000
+#define MDIO_COMBO_IEEO_MII_CONTROL_MAN_SGMII_SP_1000	0x0040
+#define MDIO_COMBO_IEEO_MII_CONTROL_FULL_DUPLEX 	0x0100
+#define MDIO_COMBO_IEEO_MII_CONTROL_RESTART_AN		0x0200
+#define MDIO_COMBO_IEEO_MII_CONTROL_AN_EN		0x1000
+#define MDIO_COMBO_IEEO_MII_CONTROL_LOOPBACK		0x4000
+#define MDIO_COMBO_IEEO_MII_CONTROL_RESET		0x8000
+#define MDIO_COMBO_IEEE0_MII_STATUS			0x11
+#define MDIO_COMBO_IEEE0_MII_STATUS_LINK_PASS		0x0004
+#define MDIO_COMBO_IEEE0_MII_STATUS_AUTONEG_COMPLETE	0x0020
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV			0x14
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_FULL_DUPLEX	0x0020
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_HALF_DUPLEX	0x0040
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_MASK	0x0180
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_NONE	0x0000
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_SYMMETRIC	0x0080
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_ASYMMETRIC	0x0100
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_BOTH	0x0180
+#define MDIO_COMBO_IEEE0_AUTO_NEG_ADV_NEXT_PAGE 	0x8000
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1 0x15
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_NEXT_PAGE    0x8000
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_ACK	     0x4000
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_PAUSE_MASK   0x0180
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_PAUSE_NONE\
+	0x0000
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_PAUSE_BOTH\
+	0x0180
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_HALF_DUP_CAP 0x0040
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_FULL_DUP_CAP 0x0020
+#define MDIO_COMBO_IEEE0_AUTO_NEG_LINK_PARTNER_ABILITY1_SGMII_MODE   0x0001
+
+
+#define EXT_PHY_OPT_PMA_PMD_DEVAD			0x1
+#define EXT_PHY_OPT_WIS_DEVAD				0x2
+#define EXT_PHY_OPT_PCS_DEVAD				0x3
+#define EXT_PHY_OPT_PHY_XS_DEVAD			0x4
+#define EXT_PHY_OPT_CNTL				0x0
+#define EXT_PHY_OPT_PMD_RX_SD				0xa
+#define EXT_PHY_OPT_PMD_MISC_CNTL			0xca0a
+#define EXT_PHY_OPT_PHY_IDENTIFIER			0xc800
+#define EXT_PHY_OPT_PMD_DIGITAL_CNT			0xc808
+#define EXT_PHY_OPT_PMD_DIGITAL_SATUS			0xc809
+#define EXT_PHY_OPT_CMU_PLL_BYPASS			0xca09
+#define EXT_PHY_OPT_LASI_CNTL				0x9002
+#define EXT_PHY_OPT_RX_ALARM				0x9003
+#define EXT_PHY_OPT_LASI_STATUS 			0x9005
+#define EXT_PHY_OPT_PCS_STATUS				0x0020
+#define EXT_PHY_OPT_XGXS_LANE_STATUS			0x0018
+
+#define EXT_PHY_KR_PMA_PMD_DEVAD			0x1
+#define EXT_PHY_KR_PCS_DEVAD				0x3
+#define EXT_PHY_KR_AUTO_NEG_DEVAD			0x7
+#define EXT_PHY_KR_CTRL 				0x0000
+#define EXT_PHY_KR_CTRL2				0x0007
+#define EXT_PHY_KR_PCS_STATUS				0x0020
+#define EXT_PHY_KR_PMD_CTRL				0x0096
+#define EXT_PHY_KR_LASI_CNTL				0x9002
+#define EXT_PHY_KR_LASI_STATUS				0x9005
+#define EXT_PHY_KR_MISC_CTRL1				0xca85
+#define EXT_PHY_KR_GEN_CTRL				0xca10
+#define EXT_PHY_KR_ROM_CODE				0xca19
+
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 1fbd0256e86b..f162d9c1226c 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1943,6 +1943,7 @@
 #define PCI_DEVICE_ID_NX2_5706		0x164a
 #define PCI_DEVICE_ID_NX2_5708		0x164c
 #define PCI_DEVICE_ID_TIGON3_5702FE	0x164d
+#define PCI_DEVICE_ID_NX2_57710		0x164e
 #define PCI_DEVICE_ID_TIGON3_5705	0x1653
 #define PCI_DEVICE_ID_TIGON3_5705_2	0x1654
 #define PCI_DEVICE_ID_TIGON3_5720	0x1658
-- 
cgit v1.2.3


From 092427be8cef341c957a93ec2469890501a09bff Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Fri, 23 Nov 2007 21:49:27 -0500
Subject: drivers/net/r6040: fix obvious problems (but more remain)

- checkpatch fixes

- fix bogus and uninitialized return codes in r6040_start_xmit()

- netdev_get_settings() fix obvious locking bug flagged by compiler warning

- set DMA consistent mask

- remove unnecessary setting of dev->base_addr

Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/r6040.c     | 31 ++++++++++++++-----------------
 include/linux/pci_ids.h |  3 +++
 2 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c
index edce5a4c7f20..1d7efa2b6cbc 100644
--- a/drivers/net/r6040.c
+++ b/drivers/net/r6040.c
@@ -42,12 +42,12 @@
 #include <linux/ethtool.h>
 #include <linux/crc32.h>
 #include <linux/spinlock.h>
+#include <linux/bitops.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/uaccess.h>
 
 #include <asm/processor.h>
-#include <asm/bitops.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
 
 #define DRV_NAME	"r6040"
 #define DRV_VERSION	"0.16"
@@ -181,7 +181,7 @@ static char version[] __devinitdata = KERN_INFO DRV_NAME
 	": RDC R6040 NAPI net driver,"
 	"version "DRV_VERSION " (" DRV_RELDATE ")\n";
 
-static int phy_table[] = { PHY1_ADDR, PHY2_ADDR};
+static int phy_table[] = { PHY1_ADDR, PHY2_ADDR };
 
 /* Read a word data from PHY Chip */
 static int phy_read(void __iomem *ioaddr, int phy_addr, int reg)
@@ -771,15 +771,7 @@ r6040_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct r6040_descriptor *descptr;
 	void __iomem *ioaddr = lp->base;
 	unsigned long flags;
-	int ret;
-
-	if (!skb)	/* NULL skb directly return */
-		return ret;
-
-	if (skb->len >= MAX_BUF_SIZE) {	/* Packet too long, drop it */
-		dev_kfree_skb(skb);
-		return ret;
-	}
+	int ret = NETDEV_TX_OK;
 
 	/* Critical Section */
 	spin_lock_irqsave(&lp->lock, flags);
@@ -787,8 +779,9 @@ r6040_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* TX resource check */
 	if (!lp->tx_free_desc) {
 		spin_unlock_irqrestore(&lp->lock, flags);
+		netif_stop_queue(dev);
 		printk(KERN_ERR DRV_NAME ": no tx descriptor\n");
-		ret = 1;
+		ret = NETDEV_TX_BUSY;
 		return ret;
 	}
 
@@ -916,7 +909,7 @@ static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 	spin_lock_irq(&rp->lock);
 	rc = mii_ethtool_gset(&rp->mii_if, cmd);
-	spin_unlock_irq(&rp->mii_if);
+	spin_unlock_irq(&rp->lock);
 
 	return rc;
 }
@@ -973,6 +966,11 @@ static int __devinit r6040_init_one(struct pci_dev *pdev,
 				"not supported by the card\n");
 		return -ENODEV;
 	}
+	if (pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK)) {
+		printk(KERN_ERR DRV_NAME "32-bit PCI DMA addresses"
+				"not supported by the card\n");
+		return -ENODEV;
+	}
 
 	/* IO Size check */
 	if (pci_resource_len(pdev, 0) < io_size) {
@@ -1006,7 +1004,6 @@ static int __devinit r6040_init_one(struct pci_dev *pdev,
 	}
 
 	/* Init system & device */
-	dev->base_addr = (unsigned long)ioaddr;
 	lp->base = ioaddr;
 	dev->irq = pdev->irq;
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index f162d9c1226c..1280b0c726ea 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2110,6 +2110,9 @@
 #define PCI_DEVICE_ID_HERC_WIN		0x5732
 #define PCI_DEVICE_ID_HERC_UNI		0x5832
 
+#define PCI_VENDOR_ID_RDC		0x17f3
+#define PCI_DEVICE_ID_RDC_R6040		0x6040
+
 #define PCI_VENDOR_ID_SITECOM		0x182d
 #define PCI_DEVICE_ID_SITECOM_DC105V2	0x3069
 
-- 
cgit v1.2.3


From 5ac5d616327bdbdf632bdf4dc9ae09477f79b6b3 Mon Sep 17 00:00:00 2001
From: Francois Romieu <romieu@fr.zoreil.com>
Date: Wed, 28 Nov 2007 23:02:33 +0100
Subject: r6040: cleanups

- whitespaces vs tabs
- use 80 cols
- use if_mii
- use netdev_priv
- remove useless cast to void *
- PCI device id does not need to be globally available

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
---
 drivers/net/r6040.c     | 33 +++++++++++++--------------------
 include/linux/pci_ids.h |  1 -
 2 files changed, 13 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c
index 71aa4783efd9..2334f4ebf907 100644
--- a/drivers/net/r6040.c
+++ b/drivers/net/r6040.c
@@ -3,7 +3,7 @@
  *
  * Copyright (C) 2004 Sten Wang <sten.wang@rdc.com.tw>
  * Copyright (C) 2007
- * 	Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
+ *	Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
  *	Florian Fainelli <florian@openwrt.org>
  *
  * This program is free software; you can redistribute it and/or
@@ -60,7 +60,7 @@
 #define PHY_CAP		0x01E1	/* PHY CHIP Register 4 */
 
 /* Time in jiffies before concluding the transmitter is hung. */
-#define TX_TIMEOUT  	(6000 * HZ / 1000)
+#define TX_TIMEOUT	(6000 * HZ / 1000)
 #define TIMER_WUT	(jiffies + HZ * 1)/* timer wakeup time : 1 second */
 
 /* RDC MAC I/O Size */
@@ -235,8 +235,7 @@ static void mdio_write(struct net_device *dev, int mii_id, int reg, int val)
 	phy_write(ioaddr, lp->phy_addr, reg, val);
 }
 
-static void
-r6040_tx_timeout(struct net_device *dev)
+static void r6040_tx_timeout(struct net_device *dev)
 {
 	struct r6040_private *priv = netdev_priv(dev);
 
@@ -341,8 +340,7 @@ static void r6040_down(struct net_device *dev)
 	pci_free_consistent(pdev, TX_DESC_SIZE, lp->tx_ring, lp->tx_ring_dma);
 }
 
-static int
-r6040_close(struct net_device *dev)
+static int r6040_close(struct net_device *dev)
 {
 	struct r6040_private *lp = netdev_priv(dev);
 
@@ -405,7 +403,7 @@ static void r6040_set_carrier(struct mii_if_info *mii)
 static int r6040_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 {
 	struct r6040_private *lp = netdev_priv(dev);
-	struct mii_ioctl_data *data = (struct mii_ioctl_data *) &rq->ifr_data;
+	struct mii_ioctl_data *data = if_mii(rq);
 	int rc;
 
 	if (!netif_running(dev))
@@ -574,12 +572,11 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id)
 static void r6040_poll_controller(struct net_device *dev)
 {
 	disable_irq(dev->irq);
-	r6040_interrupt(dev->irq, (void *)dev);
+	r6040_interrupt(dev->irq, dev);
 	enable_irq(dev->irq);
 }
 #endif
 
-
 static void r6040_init_ring_desc(struct r6040_descriptor *desc_ring,
 				 dma_addr_t desc_dma, int size)
 {
@@ -716,10 +713,9 @@ static void r6040_mac_address(struct net_device *dev)
 	iowrite16(adrp[2], ioaddr + MID_0H);
 }
 
-static int
-r6040_open(struct net_device *dev)
+static int r6040_open(struct net_device *dev)
 {
-	struct r6040_private *lp = dev->priv;
+	struct r6040_private *lp = netdev_priv(dev);
 	int ret;
 
 	/* Request IRQ and Register interrupt handler */
@@ -761,8 +757,7 @@ r6040_open(struct net_device *dev)
 	return 0;
 }
 
-static int
-r6040_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static int r6040_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct r6040_private *lp = netdev_priv(dev);
 	struct r6040_descriptor *descptr;
@@ -810,8 +805,7 @@ r6040_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return ret;
 }
 
-static void
-r6040_multicast_list(struct net_device *dev)
+static void r6040_multicast_list(struct net_device *dev)
 {
 	struct r6040_private *lp = netdev_priv(dev);
 	void __iomem *ioaddr = lp->base;
@@ -938,7 +932,6 @@ static struct ethtool_ops netdev_ethtool_ops = {
 	.get_link		= netdev_get_link,
 };
 
-
 static int __devinit r6040_init_one(struct pci_dev *pdev,
 					 const struct pci_device_id *ent)
 {
@@ -1075,13 +1068,13 @@ static void __devexit r6040_remove_one(struct pci_dev *pdev)
 
 
 static struct pci_device_id r6040_pci_tbl[] = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_RDC, PCI_DEVICE_ID_RDC_R6040) },
-	{0 }
+	{ PCI_DEVICE(PCI_VENDOR_ID_RDC, 0x6040) },
+	{ 0 }
 };
 MODULE_DEVICE_TABLE(pci, r6040_pci_tbl);
 
 static struct pci_driver r6040_driver = {
-	.name		= "r6040",
+	.name		= DRV_NAME,
 	.id_table	= r6040_pci_tbl,
 	.probe		= r6040_init_one,
 	.remove		= __devexit_p(r6040_remove_one),
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 1280b0c726ea..6b4a132bf861 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2111,7 +2111,6 @@
 #define PCI_DEVICE_ID_HERC_UNI		0x5832
 
 #define PCI_VENDOR_ID_RDC		0x17f3
-#define PCI_DEVICE_ID_RDC_R6040		0x6040
 
 #define PCI_VENDOR_ID_SITECOM		0x182d
 #define PCI_DEVICE_ID_SITECOM_DC105V2	0x3069
-- 
cgit v1.2.3


From 40afa5315823761b174926235dc38be24dc3ea63 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 28 Nov 2007 20:56:04 -0600
Subject: pasemi_mac: Move register definitions to include/asm-powerpc

pasemi_mac: Move register definitions to include/asm-powerpc

Move the common register formats and descriptor layouts from
drivers/net/pasemi_mac.h to include/asm-poewrpc/pasemi_dma.h

Previously only the ethernet driver was using them, but other drivers
are coming up that will also use them, so it makes sense to share the
constants.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/pasemi_mac.c         |   1 +
 drivers/net/pasemi_mac.h         | 336 +--------------------------------
 include/asm-powerpc/pasemi_dma.h | 391 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 394 insertions(+), 334 deletions(-)
 create mode 100644 include/asm-powerpc/pasemi_dma.h

(limited to 'include')

diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c
index a2527b53da8b..8319bc18bfa2 100644
--- a/drivers/net/pasemi_mac.c
+++ b/drivers/net/pasemi_mac.c
@@ -35,6 +35,7 @@
 
 #include <asm/irq.h>
 #include <asm/firmware.h>
+#include <asm/pasemi_dma.h>
 
 #include "pasemi_mac.h"
 
diff --git a/drivers/net/pasemi_mac.h b/drivers/net/pasemi_mac.h
index 1269015fa5da..c61c0110e085 100644
--- a/drivers/net/pasemi_mac.h
+++ b/drivers/net/pasemi_mac.h
@@ -96,11 +96,8 @@ struct pasemi_mac_buffer {
 };
 
 
-/* status register layout in IOB region, at 0xfb800000 */
-struct pasdma_status {
-	u64 rx_sta[64];
-	u64 tx_sta[20];
-};
+/* PCI register offsets and formats */
+
 
 /* MAC CFG register offsets */
 enum {
@@ -174,333 +171,4 @@ enum {
 #define PAS_MAC_IPC_CHNL_BCH(x)		(((x) << PAS_MAC_IPC_CHNL_BCH_S) & \
 					 PAS_MAC_IPC_CHNL_BCH_M)
 
-/* All these registers live in the PCI configuration space for the DMA PCI
- * device. Use the normal PCI config access functions for them.
- */
-enum {
-	PAS_DMA_COM_TXCMD = 0x100,	/* Transmit Command Register  */
-	PAS_DMA_COM_TXSTA = 0x104,	/* Transmit Status Register   */
-	PAS_DMA_COM_RXCMD = 0x108,	/* Receive Command Register   */
-	PAS_DMA_COM_RXSTA = 0x10c,	/* Receive Status Register    */
-};
-#define PAS_DMA_COM_TXCMD_EN	0x00000001 /* enable */
-#define PAS_DMA_COM_TXSTA_ACT	0x00000001 /* active */
-#define PAS_DMA_COM_RXCMD_EN	0x00000001 /* enable */
-#define PAS_DMA_COM_RXSTA_ACT	0x00000001 /* active */
-
-
-/* Per-interface and per-channel registers */
-#define _PAS_DMA_RXINT_STRIDE		0x20
-#define PAS_DMA_RXINT_RCMDSTA(i)	(0x200+(i)*_PAS_DMA_RXINT_STRIDE)
-#define    PAS_DMA_RXINT_RCMDSTA_EN	0x00000001
-#define    PAS_DMA_RXINT_RCMDSTA_ST	0x00000002
-#define    PAS_DMA_RXINT_RCMDSTA_MBT	0x00000008
-#define    PAS_DMA_RXINT_RCMDSTA_MDR	0x00000010
-#define    PAS_DMA_RXINT_RCMDSTA_MOO	0x00000020
-#define    PAS_DMA_RXINT_RCMDSTA_MBP	0x00000040
-#define    PAS_DMA_RXINT_RCMDSTA_BT	0x00000800
-#define    PAS_DMA_RXINT_RCMDSTA_DR	0x00001000
-#define    PAS_DMA_RXINT_RCMDSTA_OO	0x00002000
-#define    PAS_DMA_RXINT_RCMDSTA_BP	0x00004000
-#define    PAS_DMA_RXINT_RCMDSTA_TB	0x00008000
-#define    PAS_DMA_RXINT_RCMDSTA_ACT	0x00010000
-#define    PAS_DMA_RXINT_RCMDSTA_DROPS_M	0xfffe0000
-#define    PAS_DMA_RXINT_RCMDSTA_DROPS_S	17
-#define PAS_DMA_RXINT_CFG(i)		(0x204+(i)*_PAS_DMA_RXINT_STRIDE)
-#define    PAS_DMA_RXINT_CFG_RBP	0x80000000
-#define    PAS_DMA_RXINT_CFG_ITRR	0x40000000
-#define    PAS_DMA_RXINT_CFG_DHL_M	0x07000000
-#define    PAS_DMA_RXINT_CFG_DHL_S	24
-#define    PAS_DMA_RXINT_CFG_DHL(x)	(((x) << PAS_DMA_RXINT_CFG_DHL_S) & \
-					 PAS_DMA_RXINT_CFG_DHL_M)
-#define    PAS_DMA_RXINT_CFG_ITR	0x00400000
-#define    PAS_DMA_RXINT_CFG_LW		0x00200000
-#define    PAS_DMA_RXINT_CFG_L2		0x00100000
-#define    PAS_DMA_RXINT_CFG_HEN	0x00080000
-#define    PAS_DMA_RXINT_CFG_WIF	0x00000002
-#define    PAS_DMA_RXINT_CFG_WIL	0x00000001
-
-#define PAS_DMA_RXINT_INCR(i)		(0x210+(i)*_PAS_DMA_RXINT_STRIDE)
-#define    PAS_DMA_RXINT_INCR_INCR_M	0x0000ffff
-#define    PAS_DMA_RXINT_INCR_INCR_S	0
-#define    PAS_DMA_RXINT_INCR_INCR(x)	((x) & 0x0000ffff)
-#define PAS_DMA_RXINT_BASEL(i)		(0x218+(i)*_PAS_DMA_RXINT_STRIDE)
-#define    PAS_DMA_RXINT_BASEL_BRBL(x)	((x) & ~0x3f)
-#define PAS_DMA_RXINT_BASEU(i)		(0x21c+(i)*_PAS_DMA_RXINT_STRIDE)
-#define    PAS_DMA_RXINT_BASEU_BRBH(x)	((x) & 0xfff)
-#define    PAS_DMA_RXINT_BASEU_SIZ_M	0x3fff0000	/* # of cache lines worth of buffer ring */
-#define    PAS_DMA_RXINT_BASEU_SIZ_S	16		/* 0 = 16K */
-#define    PAS_DMA_RXINT_BASEU_SIZ(x)	(((x) << PAS_DMA_RXINT_BASEU_SIZ_S) & \
-					 PAS_DMA_RXINT_BASEU_SIZ_M)
-
-
-#define _PAS_DMA_TXCHAN_STRIDE	0x20    /* Size per channel		*/
-#define _PAS_DMA_TXCHAN_TCMDSTA	0x300	/* Command / Status		*/
-#define _PAS_DMA_TXCHAN_CFG	0x304	/* Configuration		*/
-#define _PAS_DMA_TXCHAN_DSCRBU	0x308	/* Descriptor BU Allocation	*/
-#define _PAS_DMA_TXCHAN_INCR	0x310	/* Descriptor increment		*/
-#define _PAS_DMA_TXCHAN_CNT	0x314	/* Descriptor count/offset	*/
-#define _PAS_DMA_TXCHAN_BASEL	0x318	/* Descriptor ring base (low)	*/
-#define _PAS_DMA_TXCHAN_BASEU	0x31c	/*			(high)	*/
-#define PAS_DMA_TXCHAN_TCMDSTA(c) (0x300+(c)*_PAS_DMA_TXCHAN_STRIDE)
-#define    PAS_DMA_TXCHAN_TCMDSTA_EN	0x00000001	/* Enabled */
-#define    PAS_DMA_TXCHAN_TCMDSTA_ST	0x00000002	/* Stop interface */
-#define    PAS_DMA_TXCHAN_TCMDSTA_ACT	0x00010000	/* Active */
-#define    PAS_DMA_TXCHAN_TCMDSTA_SZ	0x00000800
-#define    PAS_DMA_TXCHAN_TCMDSTA_DB	0x00000400
-#define    PAS_DMA_TXCHAN_TCMDSTA_DE	0x00000200
-#define    PAS_DMA_TXCHAN_TCMDSTA_DA	0x00000100
-#define PAS_DMA_TXCHAN_CFG(c)     (0x304+(c)*_PAS_DMA_TXCHAN_STRIDE)
-#define    PAS_DMA_TXCHAN_CFG_TY_IFACE	0x00000000	/* Type = interface */
-#define    PAS_DMA_TXCHAN_CFG_TATTR_M	0x0000003c
-#define    PAS_DMA_TXCHAN_CFG_TATTR_S	2
-#define    PAS_DMA_TXCHAN_CFG_TATTR(x)	(((x) << PAS_DMA_TXCHAN_CFG_TATTR_S) & \
-					 PAS_DMA_TXCHAN_CFG_TATTR_M)
-#define    PAS_DMA_TXCHAN_CFG_WT_M	0x000001c0
-#define    PAS_DMA_TXCHAN_CFG_WT_S	6
-#define    PAS_DMA_TXCHAN_CFG_WT(x)	(((x) << PAS_DMA_TXCHAN_CFG_WT_S) & \
-					 PAS_DMA_TXCHAN_CFG_WT_M)
-#define    PAS_DMA_TXCHAN_CFG_TRD	0x00010000	/* translate data */
-#define    PAS_DMA_TXCHAN_CFG_TRR	0x00008000	/* translate rings */
-#define    PAS_DMA_TXCHAN_CFG_UP	0x00004000	/* update tx descr when sent */
-#define    PAS_DMA_TXCHAN_CFG_CL	0x00002000	/* Clean last line */
-#define    PAS_DMA_TXCHAN_CFG_CF	0x00001000	/* Clean first line */
-#define PAS_DMA_TXCHAN_INCR(c)    (0x310+(c)*_PAS_DMA_TXCHAN_STRIDE)
-#define PAS_DMA_TXCHAN_BASEL(c)   (0x318+(c)*_PAS_DMA_TXCHAN_STRIDE)
-#define    PAS_DMA_TXCHAN_BASEL_BRBL_M	0xffffffc0
-#define    PAS_DMA_TXCHAN_BASEL_BRBL_S	0
-#define    PAS_DMA_TXCHAN_BASEL_BRBL(x)	(((x) << PAS_DMA_TXCHAN_BASEL_BRBL_S) & \
-					 PAS_DMA_TXCHAN_BASEL_BRBL_M)
-#define PAS_DMA_TXCHAN_BASEU(c)   (0x31c+(c)*_PAS_DMA_TXCHAN_STRIDE)
-#define    PAS_DMA_TXCHAN_BASEU_BRBH_M	0x00000fff
-#define    PAS_DMA_TXCHAN_BASEU_BRBH_S	0
-#define    PAS_DMA_TXCHAN_BASEU_BRBH(x)	(((x) << PAS_DMA_TXCHAN_BASEU_BRBH_S) & \
-					 PAS_DMA_TXCHAN_BASEU_BRBH_M)
-/* # of cache lines worth of buffer ring */
-#define    PAS_DMA_TXCHAN_BASEU_SIZ_M	0x3fff0000
-#define    PAS_DMA_TXCHAN_BASEU_SIZ_S	16		/* 0 = 16K */
-#define    PAS_DMA_TXCHAN_BASEU_SIZ(x)	(((x) << PAS_DMA_TXCHAN_BASEU_SIZ_S) & \
-					 PAS_DMA_TXCHAN_BASEU_SIZ_M)
-
-#define _PAS_DMA_RXCHAN_STRIDE	0x20    /* Size per channel		*/
-#define _PAS_DMA_RXCHAN_CCMDSTA	0x800	/* Command / Status		*/
-#define _PAS_DMA_RXCHAN_CFG	0x804	/* Configuration		*/
-#define _PAS_DMA_RXCHAN_INCR	0x810	/* Descriptor increment		*/
-#define _PAS_DMA_RXCHAN_CNT	0x814	/* Descriptor count/offset	*/
-#define _PAS_DMA_RXCHAN_BASEL	0x818	/* Descriptor ring base (low)	*/
-#define _PAS_DMA_RXCHAN_BASEU	0x81c	/*			(high)	*/
-#define PAS_DMA_RXCHAN_CCMDSTA(c) (0x800+(c)*_PAS_DMA_RXCHAN_STRIDE)
-#define    PAS_DMA_RXCHAN_CCMDSTA_EN	0x00000001	/* Enabled */
-#define    PAS_DMA_RXCHAN_CCMDSTA_ST	0x00000002	/* Stop interface */
-#define    PAS_DMA_RXCHAN_CCMDSTA_ACT	0x00010000	/* Active */
-#define    PAS_DMA_RXCHAN_CCMDSTA_DU	0x00020000
-#define    PAS_DMA_RXCHAN_CCMDSTA_OD	0x00002000
-#define    PAS_DMA_RXCHAN_CCMDSTA_FD	0x00001000
-#define    PAS_DMA_RXCHAN_CCMDSTA_DT	0x00000800
-#define PAS_DMA_RXCHAN_CFG(c)     (0x804+(c)*_PAS_DMA_RXCHAN_STRIDE)
-#define    PAS_DMA_RXCHAN_CFG_CTR	0x00000400
-#define    PAS_DMA_RXCHAN_CFG_HBU_M	0x00000380
-#define    PAS_DMA_RXCHAN_CFG_HBU_S	7
-#define    PAS_DMA_RXCHAN_CFG_HBU(x)	(((x) << PAS_DMA_RXCHAN_CFG_HBU_S) & \
-					 PAS_DMA_RXCHAN_CFG_HBU_M)
-#define PAS_DMA_RXCHAN_INCR(c)    (0x810+(c)*_PAS_DMA_RXCHAN_STRIDE)
-#define PAS_DMA_RXCHAN_BASEL(c)   (0x818+(c)*_PAS_DMA_RXCHAN_STRIDE)
-#define    PAS_DMA_RXCHAN_BASEL_BRBL_M	0xffffffc0
-#define    PAS_DMA_RXCHAN_BASEL_BRBL_S	0
-#define    PAS_DMA_RXCHAN_BASEL_BRBL(x)	(((x) << PAS_DMA_RXCHAN_BASEL_BRBL_S) & \
-					 PAS_DMA_RXCHAN_BASEL_BRBL_M)
-#define PAS_DMA_RXCHAN_BASEU(c)   (0x81c+(c)*_PAS_DMA_RXCHAN_STRIDE)
-#define    PAS_DMA_RXCHAN_BASEU_BRBH_M	0x00000fff
-#define    PAS_DMA_RXCHAN_BASEU_BRBH_S	0
-#define    PAS_DMA_RXCHAN_BASEU_BRBH(x)	(((x) << PAS_DMA_RXCHAN_BASEU_BRBH_S) & \
-					 PAS_DMA_RXCHAN_BASEU_BRBH_M)
-/* # of cache lines worth of buffer ring */
-#define    PAS_DMA_RXCHAN_BASEU_SIZ_M	0x3fff0000
-#define    PAS_DMA_RXCHAN_BASEU_SIZ_S	16		/* 0 = 16K */
-#define    PAS_DMA_RXCHAN_BASEU_SIZ(x)	(((x) << PAS_DMA_RXCHAN_BASEU_SIZ_S) & \
-					 PAS_DMA_RXCHAN_BASEU_SIZ_M)
-
-#define    PAS_STATUS_PCNT_M		0x000000000000ffffull
-#define    PAS_STATUS_PCNT_S		0
-#define    PAS_STATUS_DCNT_M		0x00000000ffff0000ull
-#define    PAS_STATUS_DCNT_S		16
-#define    PAS_STATUS_BPCNT_M		0x0000ffff00000000ull
-#define    PAS_STATUS_BPCNT_S		32
-#define    PAS_STATUS_CAUSE_M		0xf000000000000000ull
-#define    PAS_STATUS_TIMER		0x1000000000000000ull
-#define    PAS_STATUS_ERROR		0x2000000000000000ull
-#define    PAS_STATUS_SOFT		0x4000000000000000ull
-#define    PAS_STATUS_INT		0x8000000000000000ull
-
-#define PAS_IOB_COM_PKTHDRCNT		0x120
-#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR1_M	0x0fff0000
-#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR1_S	16
-#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR0_M	0x00000fff
-#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR0_S	0
-
-#define PAS_IOB_DMA_RXCH_CFG(i)		(0x1100 + (i)*4)
-#define    PAS_IOB_DMA_RXCH_CFG_CNTTH_M		0x00000fff
-#define    PAS_IOB_DMA_RXCH_CFG_CNTTH_S		0
-#define    PAS_IOB_DMA_RXCH_CFG_CNTTH(x)	(((x) << PAS_IOB_DMA_RXCH_CFG_CNTTH_S) & \
-						 PAS_IOB_DMA_RXCH_CFG_CNTTH_M)
-#define PAS_IOB_DMA_TXCH_CFG(i)		(0x1200 + (i)*4)
-#define    PAS_IOB_DMA_TXCH_CFG_CNTTH_M		0x00000fff
-#define    PAS_IOB_DMA_TXCH_CFG_CNTTH_S		0
-#define    PAS_IOB_DMA_TXCH_CFG_CNTTH(x)	(((x) << PAS_IOB_DMA_TXCH_CFG_CNTTH_S) & \
-						 PAS_IOB_DMA_TXCH_CFG_CNTTH_M)
-#define PAS_IOB_DMA_RXCH_STAT(i)	(0x1300 + (i)*4)
-#define    PAS_IOB_DMA_RXCH_STAT_INTGEN	0x00001000
-#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL_M	0x00000fff
-#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL_S	0
-#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL(x)	(((x) << PAS_IOB_DMA_RXCH_STAT_CNTDEL_S) &\
-						 PAS_IOB_DMA_RXCH_STAT_CNTDEL_M)
-#define PAS_IOB_DMA_TXCH_STAT(i)	(0x1400 + (i)*4)
-#define    PAS_IOB_DMA_TXCH_STAT_INTGEN	0x00001000
-#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL_M	0x00000fff
-#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL_S	0
-#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL(x)	(((x) << PAS_IOB_DMA_TXCH_STAT_CNTDEL_S) &\
-						 PAS_IOB_DMA_TXCH_STAT_CNTDEL_M)
-#define PAS_IOB_DMA_RXCH_RESET(i)	(0x1500 + (i)*4)
-#define    PAS_IOB_DMA_RXCH_RESET_PCNT_M	0xffff0000
-#define    PAS_IOB_DMA_RXCH_RESET_PCNT_S	16
-#define    PAS_IOB_DMA_RXCH_RESET_PCNT(x)	(((x) << PAS_IOB_DMA_RXCH_RESET_PCNT_S) & \
-						 PAS_IOB_DMA_RXCH_RESET_PCNT_M)
-#define    PAS_IOB_DMA_RXCH_RESET_PCNTRST	0x00000020
-#define    PAS_IOB_DMA_RXCH_RESET_DCNTRST	0x00000010
-#define    PAS_IOB_DMA_RXCH_RESET_TINTC		0x00000008
-#define    PAS_IOB_DMA_RXCH_RESET_DINTC		0x00000004
-#define    PAS_IOB_DMA_RXCH_RESET_SINTC		0x00000002
-#define    PAS_IOB_DMA_RXCH_RESET_PINTC		0x00000001
-#define PAS_IOB_DMA_TXCH_RESET(i)	(0x1600 + (i)*4)
-#define    PAS_IOB_DMA_TXCH_RESET_PCNT_M	0xffff0000
-#define    PAS_IOB_DMA_TXCH_RESET_PCNT_S	16
-#define    PAS_IOB_DMA_TXCH_RESET_PCNT(x)	(((x) << PAS_IOB_DMA_TXCH_RESET_PCNT_S) & \
-						 PAS_IOB_DMA_TXCH_RESET_PCNT_M)
-#define    PAS_IOB_DMA_TXCH_RESET_PCNTRST	0x00000020
-#define    PAS_IOB_DMA_TXCH_RESET_DCNTRST	0x00000010
-#define    PAS_IOB_DMA_TXCH_RESET_TINTC		0x00000008
-#define    PAS_IOB_DMA_TXCH_RESET_DINTC		0x00000004
-#define    PAS_IOB_DMA_TXCH_RESET_SINTC		0x00000002
-#define    PAS_IOB_DMA_TXCH_RESET_PINTC		0x00000001
-
-#define PAS_IOB_DMA_COM_TIMEOUTCFG		0x1700
-#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_M	0x00ffffff
-#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_S	0
-#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT(x)	(((x) << PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_S) & \
-						 PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_M)
-
-/* Transmit descriptor fields */
-#define	XCT_MACTX_T		0x8000000000000000ull
-#define	XCT_MACTX_ST		0x4000000000000000ull
-#define XCT_MACTX_NORES		0x0000000000000000ull
-#define XCT_MACTX_8BRES		0x1000000000000000ull
-#define XCT_MACTX_24BRES	0x2000000000000000ull
-#define XCT_MACTX_40BRES	0x3000000000000000ull
-#define XCT_MACTX_I		0x0800000000000000ull
-#define XCT_MACTX_O		0x0400000000000000ull
-#define XCT_MACTX_E		0x0200000000000000ull
-#define XCT_MACTX_VLAN_M	0x0180000000000000ull
-#define XCT_MACTX_VLAN_NOP	0x0000000000000000ull
-#define XCT_MACTX_VLAN_REMOVE	0x0080000000000000ull
-#define XCT_MACTX_VLAN_INSERT   0x0100000000000000ull
-#define XCT_MACTX_VLAN_REPLACE  0x0180000000000000ull
-#define XCT_MACTX_CRC_M		0x0060000000000000ull
-#define XCT_MACTX_CRC_NOP	0x0000000000000000ull
-#define XCT_MACTX_CRC_INSERT	0x0020000000000000ull
-#define XCT_MACTX_CRC_PAD	0x0040000000000000ull
-#define XCT_MACTX_CRC_REPLACE	0x0060000000000000ull
-#define XCT_MACTX_SS		0x0010000000000000ull
-#define XCT_MACTX_LLEN_M	0x00007fff00000000ull
-#define XCT_MACTX_LLEN_S	32ull
-#define XCT_MACTX_LLEN(x)	((((long)(x)) << XCT_MACTX_LLEN_S) & \
-				 XCT_MACTX_LLEN_M)
-#define XCT_MACTX_IPH_M		0x00000000f8000000ull
-#define XCT_MACTX_IPH_S		27ull
-#define XCT_MACTX_IPH(x)	((((long)(x)) << XCT_MACTX_IPH_S) & \
-				 XCT_MACTX_IPH_M)
-#define XCT_MACTX_IPO_M		0x0000000007c00000ull
-#define XCT_MACTX_IPO_S		22ull
-#define XCT_MACTX_IPO(x)	((((long)(x)) << XCT_MACTX_IPO_S) & \
-				 XCT_MACTX_IPO_M)
-#define XCT_MACTX_CSUM_M	0x0000000000000060ull
-#define XCT_MACTX_CSUM_NOP	0x0000000000000000ull
-#define XCT_MACTX_CSUM_TCP	0x0000000000000040ull
-#define XCT_MACTX_CSUM_UDP	0x0000000000000060ull
-#define XCT_MACTX_V6		0x0000000000000010ull
-#define XCT_MACTX_C		0x0000000000000004ull
-#define XCT_MACTX_AL2		0x0000000000000002ull
-
-/* Receive descriptor fields */
-#define	XCT_MACRX_T		0x8000000000000000ull
-#define	XCT_MACRX_ST		0x4000000000000000ull
-#define XCT_MACRX_RR_M		0x3000000000000000ull
-#define XCT_MACRX_RR_NORES	0x0000000000000000ull
-#define XCT_MACRX_RR_8BRES	0x1000000000000000ull
-#define XCT_MACRX_O		0x0400000000000000ull
-#define XCT_MACRX_E		0x0200000000000000ull
-#define XCT_MACRX_FF		0x0100000000000000ull
-#define XCT_MACRX_PF		0x0080000000000000ull
-#define XCT_MACRX_OB		0x0040000000000000ull
-#define XCT_MACRX_OD		0x0020000000000000ull
-#define XCT_MACRX_FS		0x0010000000000000ull
-#define XCT_MACRX_NB_M		0x000fc00000000000ull
-#define XCT_MACRX_NB_S		46ULL
-#define XCT_MACRX_NB(x)		((((long)(x)) << XCT_MACRX_NB_S) & \
-				 XCT_MACRX_NB_M)
-#define XCT_MACRX_LLEN_M	0x00003fff00000000ull
-#define XCT_MACRX_LLEN_S	32ULL
-#define XCT_MACRX_LLEN(x)	((((long)(x)) << XCT_MACRX_LLEN_S) & \
-				 XCT_MACRX_LLEN_M)
-#define XCT_MACRX_CRC		0x0000000080000000ull
-#define XCT_MACRX_LEN_M		0x0000000060000000ull
-#define XCT_MACRX_LEN_TOOSHORT	0x0000000020000000ull
-#define XCT_MACRX_LEN_BELOWMIN	0x0000000040000000ull
-#define XCT_MACRX_LEN_TRUNC	0x0000000060000000ull
-#define XCT_MACRX_CAST_M	0x0000000018000000ull
-#define XCT_MACRX_CAST_UNI	0x0000000000000000ull
-#define XCT_MACRX_CAST_MULTI	0x0000000008000000ull
-#define XCT_MACRX_CAST_BROAD	0x0000000010000000ull
-#define XCT_MACRX_CAST_PAUSE	0x0000000018000000ull
-#define XCT_MACRX_VLC_M		0x0000000006000000ull
-#define XCT_MACRX_FM		0x0000000001000000ull
-#define XCT_MACRX_HTY_M		0x0000000000c00000ull
-#define XCT_MACRX_HTY_IPV4_OK	0x0000000000000000ull
-#define XCT_MACRX_HTY_IPV6 	0x0000000000400000ull
-#define XCT_MACRX_HTY_IPV4_BAD	0x0000000000800000ull
-#define XCT_MACRX_HTY_NONIP	0x0000000000c00000ull
-#define XCT_MACRX_IPP_M		0x00000000003f0000ull
-#define XCT_MACRX_IPP_S		16
-#define XCT_MACRX_CSUM_M	0x000000000000ffffull
-#define XCT_MACRX_CSUM_S	0
-
-#define XCT_PTR_T		0x8000000000000000ull
-#define XCT_PTR_LEN_M		0x7ffff00000000000ull
-#define XCT_PTR_LEN_S		44
-#define XCT_PTR_LEN(x)		((((long)(x)) << XCT_PTR_LEN_S) & \
-				 XCT_PTR_LEN_M)
-#define XCT_PTR_ADDR_M		0x00000fffffffffffull
-#define XCT_PTR_ADDR_S		0
-#define XCT_PTR_ADDR(x)		((((long)(x)) << XCT_PTR_ADDR_S) & \
-				 XCT_PTR_ADDR_M)
-
-/* Receive interface 8byte result fields */
-#define XCT_RXRES_8B_L4O_M	0xff00000000000000ull
-#define XCT_RXRES_8B_L4O_S	56
-#define XCT_RXRES_8B_RULE_M	0x00ffff0000000000ull
-#define XCT_RXRES_8B_RULE_S	40
-#define XCT_RXRES_8B_EVAL_M	0x000000ffff000000ull
-#define XCT_RXRES_8B_EVAL_S	24
-#define XCT_RXRES_8B_HTYPE_M	0x0000000000f00000ull
-#define XCT_RXRES_8B_HASH_M	0x00000000000fffffull
-#define XCT_RXRES_8B_HASH_S	0
-
-/* Receive interface buffer fields */
-#define XCT_RXB_LEN_M		0x0ffff00000000000ull
-#define XCT_RXB_LEN_S		44
-#define XCT_RXB_LEN(x)		((((long)(x)) << XCT_PTR_LEN_S) & XCT_PTR_LEN_M)
-#define XCT_RXB_ADDR_M		0x00000fffffffffffull
-#define XCT_RXB_ADDR_S		0
-#define XCT_RXB_ADDR(x)		((((long)(x)) << XCT_PTR_ADDR_S) & XCT_PTR_ADDR_M)
-
-
 #endif /* PASEMI_MAC_H */
diff --git a/include/asm-powerpc/pasemi_dma.h b/include/asm-powerpc/pasemi_dma.h
new file mode 100644
index 000000000000..8ef80d8bdecd
--- /dev/null
+++ b/include/asm-powerpc/pasemi_dma.h
@@ -0,0 +1,391 @@
+/*
+ * Copyright (C) 2006 PA Semi, Inc
+ *
+ * Hardware register layout and descriptor formats for the on-board
+ * DMA engine on PA Semi PWRficient. Used by ethernet, function and security
+ * drivers.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#ifndef ASM_PASEMI_DMA_H
+#define ASM_PASEMI_DMA_H
+
+/* status register layout in IOB region, at 0xfb800000 */
+struct pasdma_status {
+	u64 rx_sta[64];		/* RX channel status */
+	u64 tx_sta[20];		/* TX channel status */
+};
+
+
+/* All these registers live in the PCI configuration space for the DMA PCI
+ * device. Use the normal PCI config access functions for them.
+ */
+enum {
+	PAS_DMA_COM_TXCMD = 0x100,	/* Transmit Command Register  */
+	PAS_DMA_COM_TXSTA = 0x104,	/* Transmit Status Register   */
+	PAS_DMA_COM_RXCMD = 0x108,	/* Receive Command Register   */
+	PAS_DMA_COM_RXSTA = 0x10c,	/* Receive Status Register    */
+};
+#define PAS_DMA_COM_TXCMD_EN	0x00000001 /* enable */
+#define PAS_DMA_COM_TXSTA_ACT	0x00000001 /* active */
+#define PAS_DMA_COM_RXCMD_EN	0x00000001 /* enable */
+#define PAS_DMA_COM_RXSTA_ACT	0x00000001 /* active */
+
+
+/* Per-interface and per-channel registers */
+#define _PAS_DMA_RXINT_STRIDE		0x20
+#define PAS_DMA_RXINT_RCMDSTA(i)	(0x200+(i)*_PAS_DMA_RXINT_STRIDE)
+#define    PAS_DMA_RXINT_RCMDSTA_EN	0x00000001
+#define    PAS_DMA_RXINT_RCMDSTA_ST	0x00000002
+#define    PAS_DMA_RXINT_RCMDSTA_MBT	0x00000008
+#define    PAS_DMA_RXINT_RCMDSTA_MDR	0x00000010
+#define    PAS_DMA_RXINT_RCMDSTA_MOO	0x00000020
+#define    PAS_DMA_RXINT_RCMDSTA_MBP	0x00000040
+#define    PAS_DMA_RXINT_RCMDSTA_BT	0x00000800
+#define    PAS_DMA_RXINT_RCMDSTA_DR	0x00001000
+#define    PAS_DMA_RXINT_RCMDSTA_OO	0x00002000
+#define    PAS_DMA_RXINT_RCMDSTA_BP	0x00004000
+#define    PAS_DMA_RXINT_RCMDSTA_TB	0x00008000
+#define    PAS_DMA_RXINT_RCMDSTA_ACT	0x00010000
+#define    PAS_DMA_RXINT_RCMDSTA_DROPS_M	0xfffe0000
+#define    PAS_DMA_RXINT_RCMDSTA_DROPS_S	17
+#define PAS_DMA_RXINT_CFG(i)		(0x204+(i)*_PAS_DMA_RXINT_STRIDE)
+#define    PAS_DMA_RXINT_CFG_RBP	0x80000000
+#define    PAS_DMA_RXINT_CFG_ITRR	0x40000000
+#define    PAS_DMA_RXINT_CFG_DHL_M	0x07000000
+#define    PAS_DMA_RXINT_CFG_DHL_S	24
+#define    PAS_DMA_RXINT_CFG_DHL(x)	(((x) << PAS_DMA_RXINT_CFG_DHL_S) & \
+					 PAS_DMA_RXINT_CFG_DHL_M)
+#define    PAS_DMA_RXINT_CFG_ITR	0x00400000
+#define    PAS_DMA_RXINT_CFG_LW		0x00200000
+#define    PAS_DMA_RXINT_CFG_L2		0x00100000
+#define    PAS_DMA_RXINT_CFG_HEN	0x00080000
+#define    PAS_DMA_RXINT_CFG_WIF	0x00000002
+#define    PAS_DMA_RXINT_CFG_WIL	0x00000001
+
+#define PAS_DMA_RXINT_INCR(i)		(0x210+(i)*_PAS_DMA_RXINT_STRIDE)
+#define    PAS_DMA_RXINT_INCR_INCR_M	0x0000ffff
+#define    PAS_DMA_RXINT_INCR_INCR_S	0
+#define    PAS_DMA_RXINT_INCR_INCR(x)	((x) & 0x0000ffff)
+#define PAS_DMA_RXINT_BASEL(i)		(0x218+(i)*_PAS_DMA_RXINT_STRIDE)
+#define    PAS_DMA_RXINT_BASEL_BRBL(x)	((x) & ~0x3f)
+#define PAS_DMA_RXINT_BASEU(i)		(0x21c+(i)*_PAS_DMA_RXINT_STRIDE)
+#define    PAS_DMA_RXINT_BASEU_BRBH(x)	((x) & 0xfff)
+#define    PAS_DMA_RXINT_BASEU_SIZ_M	0x3fff0000	/* # of cache lines worth of buffer ring */
+#define    PAS_DMA_RXINT_BASEU_SIZ_S	16		/* 0 = 16K */
+#define    PAS_DMA_RXINT_BASEU_SIZ(x)	(((x) << PAS_DMA_RXINT_BASEU_SIZ_S) & \
+					 PAS_DMA_RXINT_BASEU_SIZ_M)
+
+
+#define _PAS_DMA_TXCHAN_STRIDE	0x20    /* Size per channel		*/
+#define _PAS_DMA_TXCHAN_TCMDSTA	0x300	/* Command / Status		*/
+#define _PAS_DMA_TXCHAN_CFG	0x304	/* Configuration		*/
+#define _PAS_DMA_TXCHAN_DSCRBU	0x308	/* Descriptor BU Allocation	*/
+#define _PAS_DMA_TXCHAN_INCR	0x310	/* Descriptor increment		*/
+#define _PAS_DMA_TXCHAN_CNT	0x314	/* Descriptor count/offset	*/
+#define _PAS_DMA_TXCHAN_BASEL	0x318	/* Descriptor ring base (low)	*/
+#define _PAS_DMA_TXCHAN_BASEU	0x31c	/*			(high)	*/
+#define PAS_DMA_TXCHAN_TCMDSTA(c) (0x300+(c)*_PAS_DMA_TXCHAN_STRIDE)
+#define    PAS_DMA_TXCHAN_TCMDSTA_EN	0x00000001	/* Enabled */
+#define    PAS_DMA_TXCHAN_TCMDSTA_ST	0x00000002	/* Stop interface */
+#define    PAS_DMA_TXCHAN_TCMDSTA_ACT	0x00010000	/* Active */
+#define    PAS_DMA_TXCHAN_TCMDSTA_SZ	0x00000800
+#define    PAS_DMA_TXCHAN_TCMDSTA_DB	0x00000400
+#define    PAS_DMA_TXCHAN_TCMDSTA_DE	0x00000200
+#define    PAS_DMA_TXCHAN_TCMDSTA_DA	0x00000100
+#define PAS_DMA_TXCHAN_CFG(c)     (0x304+(c)*_PAS_DMA_TXCHAN_STRIDE)
+#define    PAS_DMA_TXCHAN_CFG_TY_IFACE	0x00000000	/* Type = interface */
+#define    PAS_DMA_TXCHAN_CFG_TATTR_M	0x0000003c
+#define    PAS_DMA_TXCHAN_CFG_TATTR_S	2
+#define    PAS_DMA_TXCHAN_CFG_TATTR(x)	(((x) << PAS_DMA_TXCHAN_CFG_TATTR_S) & \
+					 PAS_DMA_TXCHAN_CFG_TATTR_M)
+#define    PAS_DMA_TXCHAN_CFG_WT_M	0x000001c0
+#define    PAS_DMA_TXCHAN_CFG_WT_S	6
+#define    PAS_DMA_TXCHAN_CFG_WT(x)	(((x) << PAS_DMA_TXCHAN_CFG_WT_S) & \
+					 PAS_DMA_TXCHAN_CFG_WT_M)
+#define    PAS_DMA_TXCHAN_CFG_TRD	0x00010000	/* translate data */
+#define    PAS_DMA_TXCHAN_CFG_TRR	0x00008000	/* translate rings */
+#define    PAS_DMA_TXCHAN_CFG_UP	0x00004000	/* update tx descr when sent */
+#define    PAS_DMA_TXCHAN_CFG_CL	0x00002000	/* Clean last line */
+#define    PAS_DMA_TXCHAN_CFG_CF	0x00001000	/* Clean first line */
+#define PAS_DMA_TXCHAN_INCR(c)    (0x310+(c)*_PAS_DMA_TXCHAN_STRIDE)
+#define PAS_DMA_TXCHAN_BASEL(c)   (0x318+(c)*_PAS_DMA_TXCHAN_STRIDE)
+#define    PAS_DMA_TXCHAN_BASEL_BRBL_M	0xffffffc0
+#define    PAS_DMA_TXCHAN_BASEL_BRBL_S	0
+#define    PAS_DMA_TXCHAN_BASEL_BRBL(x)	(((x) << PAS_DMA_TXCHAN_BASEL_BRBL_S) & \
+					 PAS_DMA_TXCHAN_BASEL_BRBL_M)
+#define PAS_DMA_TXCHAN_BASEU(c)   (0x31c+(c)*_PAS_DMA_TXCHAN_STRIDE)
+#define    PAS_DMA_TXCHAN_BASEU_BRBH_M	0x00000fff
+#define    PAS_DMA_TXCHAN_BASEU_BRBH_S	0
+#define    PAS_DMA_TXCHAN_BASEU_BRBH(x)	(((x) << PAS_DMA_TXCHAN_BASEU_BRBH_S) & \
+					 PAS_DMA_TXCHAN_BASEU_BRBH_M)
+/* # of cache lines worth of buffer ring */
+#define    PAS_DMA_TXCHAN_BASEU_SIZ_M	0x3fff0000
+#define    PAS_DMA_TXCHAN_BASEU_SIZ_S	16		/* 0 = 16K */
+#define    PAS_DMA_TXCHAN_BASEU_SIZ(x)	(((x) << PAS_DMA_TXCHAN_BASEU_SIZ_S) & \
+					 PAS_DMA_TXCHAN_BASEU_SIZ_M)
+
+#define _PAS_DMA_RXCHAN_STRIDE	0x20    /* Size per channel		*/
+#define _PAS_DMA_RXCHAN_CCMDSTA	0x800	/* Command / Status		*/
+#define _PAS_DMA_RXCHAN_CFG	0x804	/* Configuration		*/
+#define _PAS_DMA_RXCHAN_INCR	0x810	/* Descriptor increment		*/
+#define _PAS_DMA_RXCHAN_CNT	0x814	/* Descriptor count/offset	*/
+#define _PAS_DMA_RXCHAN_BASEL	0x818	/* Descriptor ring base (low)	*/
+#define _PAS_DMA_RXCHAN_BASEU	0x81c	/*			(high)	*/
+#define PAS_DMA_RXCHAN_CCMDSTA(c) (0x800+(c)*_PAS_DMA_RXCHAN_STRIDE)
+#define    PAS_DMA_RXCHAN_CCMDSTA_EN	0x00000001	/* Enabled */
+#define    PAS_DMA_RXCHAN_CCMDSTA_ST	0x00000002	/* Stop interface */
+#define    PAS_DMA_RXCHAN_CCMDSTA_ACT	0x00010000	/* Active */
+#define    PAS_DMA_RXCHAN_CCMDSTA_DU	0x00020000
+#define    PAS_DMA_RXCHAN_CCMDSTA_OD	0x00002000
+#define    PAS_DMA_RXCHAN_CCMDSTA_FD	0x00001000
+#define    PAS_DMA_RXCHAN_CCMDSTA_DT	0x00000800
+#define PAS_DMA_RXCHAN_CFG(c)     (0x804+(c)*_PAS_DMA_RXCHAN_STRIDE)
+#define    PAS_DMA_RXCHAN_CFG_CTR	0x00000400
+#define    PAS_DMA_RXCHAN_CFG_HBU_M	0x00000380
+#define    PAS_DMA_RXCHAN_CFG_HBU_S	7
+#define    PAS_DMA_RXCHAN_CFG_HBU(x)	(((x) << PAS_DMA_RXCHAN_CFG_HBU_S) & \
+					 PAS_DMA_RXCHAN_CFG_HBU_M)
+#define PAS_DMA_RXCHAN_INCR(c)    (0x810+(c)*_PAS_DMA_RXCHAN_STRIDE)
+#define PAS_DMA_RXCHAN_BASEL(c)   (0x818+(c)*_PAS_DMA_RXCHAN_STRIDE)
+#define    PAS_DMA_RXCHAN_BASEL_BRBL_M	0xffffffc0
+#define    PAS_DMA_RXCHAN_BASEL_BRBL_S	0
+#define    PAS_DMA_RXCHAN_BASEL_BRBL(x)	(((x) << PAS_DMA_RXCHAN_BASEL_BRBL_S) & \
+					 PAS_DMA_RXCHAN_BASEL_BRBL_M)
+#define PAS_DMA_RXCHAN_BASEU(c)   (0x81c+(c)*_PAS_DMA_RXCHAN_STRIDE)
+#define    PAS_DMA_RXCHAN_BASEU_BRBH_M	0x00000fff
+#define    PAS_DMA_RXCHAN_BASEU_BRBH_S	0
+#define    PAS_DMA_RXCHAN_BASEU_BRBH(x)	(((x) << PAS_DMA_RXCHAN_BASEU_BRBH_S) & \
+					 PAS_DMA_RXCHAN_BASEU_BRBH_M)
+/* # of cache lines worth of buffer ring */
+#define    PAS_DMA_RXCHAN_BASEU_SIZ_M	0x3fff0000
+#define    PAS_DMA_RXCHAN_BASEU_SIZ_S	16		/* 0 = 16K */
+#define    PAS_DMA_RXCHAN_BASEU_SIZ(x)	(((x) << PAS_DMA_RXCHAN_BASEU_SIZ_S) & \
+					 PAS_DMA_RXCHAN_BASEU_SIZ_M)
+
+#define    PAS_STATUS_PCNT_M		0x000000000000ffffull
+#define    PAS_STATUS_PCNT_S		0
+#define    PAS_STATUS_DCNT_M		0x00000000ffff0000ull
+#define    PAS_STATUS_DCNT_S		16
+#define    PAS_STATUS_BPCNT_M		0x0000ffff00000000ull
+#define    PAS_STATUS_BPCNT_S		32
+#define    PAS_STATUS_CAUSE_M		0xf000000000000000ull
+#define    PAS_STATUS_TIMER		0x1000000000000000ull
+#define    PAS_STATUS_ERROR		0x2000000000000000ull
+#define    PAS_STATUS_SOFT		0x4000000000000000ull
+#define    PAS_STATUS_INT		0x8000000000000000ull
+
+#define PAS_IOB_COM_PKTHDRCNT		0x120
+#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR1_M	0x0fff0000
+#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR1_S	16
+#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR0_M	0x00000fff
+#define    PAS_IOB_COM_PKTHDRCNT_PKTHDR0_S	0
+
+#define PAS_IOB_DMA_RXCH_CFG(i)		(0x1100 + (i)*4)
+#define    PAS_IOB_DMA_RXCH_CFG_CNTTH_M		0x00000fff
+#define    PAS_IOB_DMA_RXCH_CFG_CNTTH_S		0
+#define    PAS_IOB_DMA_RXCH_CFG_CNTTH(x)	(((x) << PAS_IOB_DMA_RXCH_CFG_CNTTH_S) & \
+						 PAS_IOB_DMA_RXCH_CFG_CNTTH_M)
+#define PAS_IOB_DMA_TXCH_CFG(i)		(0x1200 + (i)*4)
+#define    PAS_IOB_DMA_TXCH_CFG_CNTTH_M		0x00000fff
+#define    PAS_IOB_DMA_TXCH_CFG_CNTTH_S		0
+#define    PAS_IOB_DMA_TXCH_CFG_CNTTH(x)	(((x) << PAS_IOB_DMA_TXCH_CFG_CNTTH_S) & \
+						 PAS_IOB_DMA_TXCH_CFG_CNTTH_M)
+#define PAS_IOB_DMA_RXCH_STAT(i)	(0x1300 + (i)*4)
+#define    PAS_IOB_DMA_RXCH_STAT_INTGEN	0x00001000
+#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL_M	0x00000fff
+#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL_S	0
+#define    PAS_IOB_DMA_RXCH_STAT_CNTDEL(x)	(((x) << PAS_IOB_DMA_RXCH_STAT_CNTDEL_S) &\
+						 PAS_IOB_DMA_RXCH_STAT_CNTDEL_M)
+#define PAS_IOB_DMA_TXCH_STAT(i)	(0x1400 + (i)*4)
+#define    PAS_IOB_DMA_TXCH_STAT_INTGEN	0x00001000
+#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL_M	0x00000fff
+#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL_S	0
+#define    PAS_IOB_DMA_TXCH_STAT_CNTDEL(x)	(((x) << PAS_IOB_DMA_TXCH_STAT_CNTDEL_S) &\
+						 PAS_IOB_DMA_TXCH_STAT_CNTDEL_M)
+#define PAS_IOB_DMA_RXCH_RESET(i)	(0x1500 + (i)*4)
+#define    PAS_IOB_DMA_RXCH_RESET_PCNT_M	0xffff0000
+#define    PAS_IOB_DMA_RXCH_RESET_PCNT_S	16
+#define    PAS_IOB_DMA_RXCH_RESET_PCNT(x)	(((x) << PAS_IOB_DMA_RXCH_RESET_PCNT_S) & \
+						 PAS_IOB_DMA_RXCH_RESET_PCNT_M)
+#define    PAS_IOB_DMA_RXCH_RESET_PCNTRST	0x00000020
+#define    PAS_IOB_DMA_RXCH_RESET_DCNTRST	0x00000010
+#define    PAS_IOB_DMA_RXCH_RESET_TINTC		0x00000008
+#define    PAS_IOB_DMA_RXCH_RESET_DINTC		0x00000004
+#define    PAS_IOB_DMA_RXCH_RESET_SINTC		0x00000002
+#define    PAS_IOB_DMA_RXCH_RESET_PINTC		0x00000001
+#define PAS_IOB_DMA_TXCH_RESET(i)	(0x1600 + (i)*4)
+#define    PAS_IOB_DMA_TXCH_RESET_PCNT_M	0xffff0000
+#define    PAS_IOB_DMA_TXCH_RESET_PCNT_S	16
+#define    PAS_IOB_DMA_TXCH_RESET_PCNT(x)	(((x) << PAS_IOB_DMA_TXCH_RESET_PCNT_S) & \
+						 PAS_IOB_DMA_TXCH_RESET_PCNT_M)
+#define    PAS_IOB_DMA_TXCH_RESET_PCNTRST	0x00000020
+#define    PAS_IOB_DMA_TXCH_RESET_DCNTRST	0x00000010
+#define    PAS_IOB_DMA_TXCH_RESET_TINTC		0x00000008
+#define    PAS_IOB_DMA_TXCH_RESET_DINTC		0x00000004
+#define    PAS_IOB_DMA_TXCH_RESET_SINTC		0x00000002
+#define    PAS_IOB_DMA_TXCH_RESET_PINTC		0x00000001
+
+#define PAS_IOB_DMA_COM_TIMEOUTCFG		0x1700
+#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_M	0x00ffffff
+#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_S	0
+#define    PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT(x)	(((x) << PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_S) & \
+						 PAS_IOB_DMA_COM_TIMEOUTCFG_TCNT_M)
+
+/* Transmit descriptor fields */
+#define	XCT_MACTX_T		0x8000000000000000ull
+#define	XCT_MACTX_ST		0x4000000000000000ull
+#define XCT_MACTX_NORES		0x0000000000000000ull
+#define XCT_MACTX_8BRES		0x1000000000000000ull
+#define XCT_MACTX_24BRES	0x2000000000000000ull
+#define XCT_MACTX_40BRES	0x3000000000000000ull
+#define XCT_MACTX_I		0x0800000000000000ull
+#define XCT_MACTX_O		0x0400000000000000ull
+#define XCT_MACTX_E		0x0200000000000000ull
+#define XCT_MACTX_VLAN_M	0x0180000000000000ull
+#define XCT_MACTX_VLAN_NOP	0x0000000000000000ull
+#define XCT_MACTX_VLAN_REMOVE	0x0080000000000000ull
+#define XCT_MACTX_VLAN_INSERT   0x0100000000000000ull
+#define XCT_MACTX_VLAN_REPLACE  0x0180000000000000ull
+#define XCT_MACTX_CRC_M		0x0060000000000000ull
+#define XCT_MACTX_CRC_NOP	0x0000000000000000ull
+#define XCT_MACTX_CRC_INSERT	0x0020000000000000ull
+#define XCT_MACTX_CRC_PAD	0x0040000000000000ull
+#define XCT_MACTX_CRC_REPLACE	0x0060000000000000ull
+#define XCT_MACTX_SS		0x0010000000000000ull
+#define XCT_MACTX_LLEN_M	0x00007fff00000000ull
+#define XCT_MACTX_LLEN_S	32ull
+#define XCT_MACTX_LLEN(x)	((((long)(x)) << XCT_MACTX_LLEN_S) & \
+				 XCT_MACTX_LLEN_M)
+#define XCT_MACTX_IPH_M		0x00000000f8000000ull
+#define XCT_MACTX_IPH_S		27ull
+#define XCT_MACTX_IPH(x)	((((long)(x)) << XCT_MACTX_IPH_S) & \
+				 XCT_MACTX_IPH_M)
+#define XCT_MACTX_IPO_M		0x0000000007c00000ull
+#define XCT_MACTX_IPO_S		22ull
+#define XCT_MACTX_IPO(x)	((((long)(x)) << XCT_MACTX_IPO_S) & \
+				 XCT_MACTX_IPO_M)
+#define XCT_MACTX_CSUM_M	0x0000000000000060ull
+#define XCT_MACTX_CSUM_NOP	0x0000000000000000ull
+#define XCT_MACTX_CSUM_TCP	0x0000000000000040ull
+#define XCT_MACTX_CSUM_UDP	0x0000000000000060ull
+#define XCT_MACTX_V6		0x0000000000000010ull
+#define XCT_MACTX_C		0x0000000000000004ull
+#define XCT_MACTX_AL2		0x0000000000000002ull
+
+/* Receive descriptor fields */
+#define	XCT_MACRX_T		0x8000000000000000ull
+#define	XCT_MACRX_ST		0x4000000000000000ull
+#define XCT_MACRX_RR_M		0x3000000000000000ull
+#define XCT_MACRX_RR_NORES	0x0000000000000000ull
+#define XCT_MACRX_RR_8BRES	0x1000000000000000ull
+#define XCT_MACRX_O		0x0400000000000000ull
+#define XCT_MACRX_E		0x0200000000000000ull
+#define XCT_MACRX_FF		0x0100000000000000ull
+#define XCT_MACRX_PF		0x0080000000000000ull
+#define XCT_MACRX_OB		0x0040000000000000ull
+#define XCT_MACRX_OD		0x0020000000000000ull
+#define XCT_MACRX_FS		0x0010000000000000ull
+#define XCT_MACRX_NB_M		0x000fc00000000000ull
+#define XCT_MACRX_NB_S		46ULL
+#define XCT_MACRX_NB(x)		((((long)(x)) << XCT_MACRX_NB_S) & \
+				 XCT_MACRX_NB_M)
+#define XCT_MACRX_LLEN_M	0x00003fff00000000ull
+#define XCT_MACRX_LLEN_S	32ULL
+#define XCT_MACRX_LLEN(x)	((((long)(x)) << XCT_MACRX_LLEN_S) & \
+				 XCT_MACRX_LLEN_M)
+#define XCT_MACRX_CRC		0x0000000080000000ull
+#define XCT_MACRX_LEN_M		0x0000000060000000ull
+#define XCT_MACRX_LEN_TOOSHORT	0x0000000020000000ull
+#define XCT_MACRX_LEN_BELOWMIN	0x0000000040000000ull
+#define XCT_MACRX_LEN_TRUNC	0x0000000060000000ull
+#define XCT_MACRX_CAST_M	0x0000000018000000ull
+#define XCT_MACRX_CAST_UNI	0x0000000000000000ull
+#define XCT_MACRX_CAST_MULTI	0x0000000008000000ull
+#define XCT_MACRX_CAST_BROAD	0x0000000010000000ull
+#define XCT_MACRX_CAST_PAUSE	0x0000000018000000ull
+#define XCT_MACRX_VLC_M		0x0000000006000000ull
+#define XCT_MACRX_FM		0x0000000001000000ull
+#define XCT_MACRX_HTY_M		0x0000000000c00000ull
+#define XCT_MACRX_HTY_IPV4_OK	0x0000000000000000ull
+#define XCT_MACRX_HTY_IPV6 	0x0000000000400000ull
+#define XCT_MACRX_HTY_IPV4_BAD	0x0000000000800000ull
+#define XCT_MACRX_HTY_NONIP	0x0000000000c00000ull
+#define XCT_MACRX_IPP_M		0x00000000003f0000ull
+#define XCT_MACRX_IPP_S		16
+#define XCT_MACRX_CSUM_M	0x000000000000ffffull
+#define XCT_MACRX_CSUM_S	0
+
+#define XCT_PTR_T		0x8000000000000000ull
+#define XCT_PTR_LEN_M		0x7ffff00000000000ull
+#define XCT_PTR_LEN_S		44
+#define XCT_PTR_LEN(x)		((((long)(x)) << XCT_PTR_LEN_S) & \
+				 XCT_PTR_LEN_M)
+#define XCT_PTR_ADDR_M		0x00000fffffffffffull
+#define XCT_PTR_ADDR_S		0
+#define XCT_PTR_ADDR(x)		((((long)(x)) << XCT_PTR_ADDR_S) & \
+				 XCT_PTR_ADDR_M)
+
+/* Receive interface 8byte result fields */
+#define XCT_RXRES_8B_L4O_M	0xff00000000000000ull
+#define XCT_RXRES_8B_L4O_S	56
+#define XCT_RXRES_8B_RULE_M	0x00ffff0000000000ull
+#define XCT_RXRES_8B_RULE_S	40
+#define XCT_RXRES_8B_EVAL_M	0x000000ffff000000ull
+#define XCT_RXRES_8B_EVAL_S	24
+#define XCT_RXRES_8B_HTYPE_M	0x0000000000f00000ull
+#define XCT_RXRES_8B_HASH_M	0x00000000000fffffull
+#define XCT_RXRES_8B_HASH_S	0
+
+/* Receive interface buffer fields */
+#define XCT_RXB_LEN_M		0x0ffff00000000000ull
+#define XCT_RXB_LEN_S		44
+#define XCT_RXB_LEN(x)		((((long)(x)) << XCT_RXB_LEN_S) & \
+				 XCT_RXB_LEN_M)
+#define XCT_RXB_ADDR_M		0x00000fffffffffffull
+#define XCT_RXB_ADDR_S		0
+#define XCT_RXB_ADDR(x)		((((long)(x)) << XCT_RXB_ADDR_S) & \
+				 XCT_RXB_ADDR_M)
+
+/* Copy descriptor fields */
+#define XCT_COPY_T		0x8000000000000000ull
+#define XCT_COPY_ST		0x4000000000000000ull
+#define XCT_COPY_RR_M		0x3000000000000000ull
+#define XCT_COPY_RR_NORES	0x0000000000000000ull
+#define XCT_COPY_RR_8BRES	0x1000000000000000ull
+#define XCT_COPY_RR_24BRES	0x2000000000000000ull
+#define XCT_COPY_RR_40BRES	0x3000000000000000ull
+#define XCT_COPY_I		0x0800000000000000ull
+#define XCT_COPY_O		0x0400000000000000ull
+#define XCT_COPY_E		0x0200000000000000ull
+#define XCT_COPY_STY_ZERO	0x01c0000000000000ull
+#define XCT_COPY_DTY_PREF	0x0038000000000000ull
+#define XCT_COPY_LLEN_M		0x0007ffff00000000ull
+#define XCT_COPY_LLEN_S		32
+#define XCT_COPY_LLEN(x)	((((long)(x)) << XCT_COPY_LLEN_S) & \
+				 XCT_COPY_LLEN_M)
+#define XCT_COPY_SE		0x0000000000000001ull
+
+/* Control descriptor fields */
+#define CTRL_CMD_T		0x8000000000000000ull
+#define CTRL_CMD_META_EVT	0x2000000000000000ull
+#define CTRL_CMD_O		0x0400000000000000ull
+#define CTRL_CMD_REG_M		0x000000000000000full
+#define CTRL_CMD_REG_S		0
+#define CTRL_CMD_REG(x)		((((long)(x)) << CTRL_CMD_REG_S) & \
+				 CTRL_CMD_REG_M)
+
+
+#endif /* ASM_PASEMI_DMA_H */
-- 
cgit v1.2.3


From 8ee9d85779356c1dc2ba87aca27fbf9414f2d82b Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 28 Nov 2007 20:56:20 -0600
Subject: pasemi: DMA engine management library

pasemi: DMA engine management library

Introduce a DMA management library to manage the various DMA resources
on the PA Semi SoCs. Since several drivers need to allocate these shared
resources, provide some abstractions as well as allocation/free functions
for channels, etc.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 arch/powerpc/platforms/pasemi/Makefile  |   2 +-
 arch/powerpc/platforms/pasemi/dma_lib.c | 487 ++++++++++++++++++++++++++++++++
 arch/powerpc/platforms/pasemi/pasemi.h  |   1 +
 include/asm-powerpc/pasemi_dma.h        |  76 +++++
 4 files changed, 565 insertions(+), 1 deletion(-)
 create mode 100644 arch/powerpc/platforms/pasemi/dma_lib.c

(limited to 'include')

diff --git a/arch/powerpc/platforms/pasemi/Makefile b/arch/powerpc/platforms/pasemi/Makefile
index f47fcac7e581..e636daa7a80e 100644
--- a/arch/powerpc/platforms/pasemi/Makefile
+++ b/arch/powerpc/platforms/pasemi/Makefile
@@ -1,4 +1,4 @@
-obj-y	+= setup.o pci.o time.o idle.o powersave.o iommu.o
+obj-y	+= setup.o pci.o time.o idle.o powersave.o iommu.o dma_lib.o
 obj-$(CONFIG_PPC_PASEMI_MDIO)	+= gpio_mdio.o
 obj-$(CONFIG_ELECTRA_IDE) += electra_ide.o
 obj-$(CONFIG_PPC_PASEMI_CPUFREQ) += cpufreq.o
diff --git a/arch/powerpc/platforms/pasemi/dma_lib.c b/arch/powerpc/platforms/pasemi/dma_lib.c
new file mode 100644
index 000000000000..e6e474247dc9
--- /dev/null
+++ b/arch/powerpc/platforms/pasemi/dma_lib.c
@@ -0,0 +1,487 @@
+/*
+ * Copyright (C) 2006-2007 PA Semi, Inc
+ *
+ * Common functions for DMA access on PA Semi PWRficient
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/of.h>
+
+#include <asm/pasemi_dma.h>
+
+#define MAX_TXCH 64
+#define MAX_RXCH 64
+
+static struct pasdma_status *dma_status;
+
+static void __iomem *iob_regs;
+static void __iomem *mac_regs[6];
+static void __iomem *dma_regs;
+
+static int base_hw_irq;
+
+static int num_txch, num_rxch;
+
+static struct pci_dev *dma_pdev;
+
+/* Bitmaps to handle allocation of channels */
+
+static DECLARE_BITMAP(txch_free, MAX_TXCH);
+static DECLARE_BITMAP(rxch_free, MAX_RXCH);
+
+/* pasemi_read_iob_reg - read IOB register
+ * @reg: Register to read (offset into PCI CFG space)
+ */
+unsigned int pasemi_read_iob_reg(unsigned int reg)
+{
+	return in_le32(iob_regs+reg);
+}
+EXPORT_SYMBOL(pasemi_read_iob_reg);
+
+/* pasemi_write_iob_reg - write IOB register
+ * @reg: Register to write to (offset into PCI CFG space)
+ * @val: Value to write
+ */
+void pasemi_write_iob_reg(unsigned int reg, unsigned int val)
+{
+	out_le32(iob_regs+reg, val);
+}
+EXPORT_SYMBOL(pasemi_write_iob_reg);
+
+/* pasemi_read_mac_reg - read MAC register
+ * @intf: MAC interface
+ * @reg: Register to read (offset into PCI CFG space)
+ */
+unsigned int pasemi_read_mac_reg(int intf, unsigned int reg)
+{
+	return in_le32(mac_regs[intf]+reg);
+}
+EXPORT_SYMBOL(pasemi_read_mac_reg);
+
+/* pasemi_write_mac_reg - write MAC register
+ * @intf: MAC interface
+ * @reg: Register to write to (offset into PCI CFG space)
+ * @val: Value to write
+ */
+void pasemi_write_mac_reg(int intf, unsigned int reg, unsigned int val)
+{
+	out_le32(mac_regs[intf]+reg, val);
+}
+EXPORT_SYMBOL(pasemi_write_mac_reg);
+
+/* pasemi_read_dma_reg - read DMA register
+ * @reg: Register to read (offset into PCI CFG space)
+ */
+unsigned int pasemi_read_dma_reg(unsigned int reg)
+{
+	return in_le32(dma_regs+reg);
+}
+EXPORT_SYMBOL(pasemi_read_dma_reg);
+
+/* pasemi_write_dma_reg - write DMA register
+ * @reg: Register to write to (offset into PCI CFG space)
+ * @val: Value to write
+ */
+void pasemi_write_dma_reg(unsigned int reg, unsigned int val)
+{
+	out_le32(dma_regs+reg, val);
+}
+EXPORT_SYMBOL(pasemi_write_dma_reg);
+
+static int pasemi_alloc_tx_chan(enum pasemi_dmachan_type type)
+{
+	int bit;
+	int start, limit;
+
+	switch (type & (TXCHAN_EVT0|TXCHAN_EVT1)) {
+	case TXCHAN_EVT0:
+		start = 0;
+		limit = 10;
+		break;
+	case TXCHAN_EVT1:
+		start = 10;
+		limit = MAX_TXCH;
+		break;
+	default:
+		start = 0;
+		limit = MAX_TXCH;
+		break;
+	}
+retry:
+	bit = find_next_bit(txch_free, MAX_TXCH, start);
+	if (bit >= limit)
+		return -ENOSPC;
+	if (!test_and_clear_bit(bit, txch_free))
+		goto retry;
+
+	return bit;
+}
+
+static void pasemi_free_tx_chan(int chan)
+{
+	BUG_ON(test_bit(chan, txch_free));
+	set_bit(chan, txch_free);
+}
+
+static int pasemi_alloc_rx_chan(void)
+{
+	int bit;
+retry:
+	bit = find_first_bit(rxch_free, MAX_RXCH);
+	if (bit >= MAX_TXCH)
+		return -ENOSPC;
+	if (!test_and_clear_bit(bit, rxch_free))
+		goto retry;
+
+	return bit;
+}
+
+static void pasemi_free_rx_chan(int chan)
+{
+	BUG_ON(test_bit(chan, rxch_free));
+	set_bit(chan, rxch_free);
+}
+
+/* pasemi_dma_alloc_chan - Allocate a DMA channel
+ * @type: Type of channel to allocate
+ * @total_size: Total size of structure to allocate (to allow for more
+ *		room behind the structure to be used by the client)
+ * @offset: Offset in bytes from start of the total structure to the beginning
+ *	    of struct pasemi_dmachan. Needed when struct pasemi_dmachan is
+ *	    not the first member of the client structure.
+ *
+ * pasemi_dma_alloc_chan allocates a DMA channel for use by a client. The
+ * type argument specifies whether it's a RX or TX channel, and in the case
+ * of TX channels which group it needs to belong to (if any).
+ *
+ * Returns a pointer to the total structure allocated on success, NULL
+ * on failure.
+ */
+void *pasemi_dma_alloc_chan(enum pasemi_dmachan_type type,
+			    int total_size, int offset)
+{
+	void *buf;
+	struct pasemi_dmachan *chan;
+	int chno;
+
+	BUG_ON(total_size < sizeof(struct pasemi_dmachan));
+
+	buf = kzalloc(total_size, GFP_KERNEL);
+
+	if (!buf)
+		return NULL;
+	chan = buf + offset;
+
+	chan->priv = buf;
+
+	switch (type & (TXCHAN|RXCHAN)) {
+	case RXCHAN:
+		chno = pasemi_alloc_rx_chan();
+		chan->chno = chno;
+		chan->irq = irq_create_mapping(NULL,
+					       base_hw_irq + num_txch + chno);
+		chan->status = &dma_status->rx_sta[chno];
+		break;
+	case TXCHAN:
+		chno = pasemi_alloc_tx_chan(type);
+		chan->chno = chno;
+		chan->irq = irq_create_mapping(NULL, base_hw_irq + chno);
+		chan->status = &dma_status->tx_sta[chno];
+		break;
+	}
+
+	chan->chan_type = type;
+
+	return chan;
+}
+EXPORT_SYMBOL(pasemi_dma_alloc_chan);
+
+/* pasemi_dma_free_chan - Free a previously allocated channel
+ * @chan: Channel to free
+ *
+ * Frees a previously allocated channel. It will also deallocate any
+ * descriptor ring associated with the channel, if allocated.
+ */
+void pasemi_dma_free_chan(struct pasemi_dmachan *chan)
+{
+	if (chan->ring_virt)
+		pasemi_dma_free_ring(chan);
+
+	switch (chan->chan_type & (RXCHAN|TXCHAN)) {
+	case RXCHAN:
+		pasemi_free_rx_chan(chan->chno);
+		break;
+	case TXCHAN:
+		pasemi_free_tx_chan(chan->chno);
+		break;
+	}
+
+	kfree(chan->priv);
+}
+EXPORT_SYMBOL(pasemi_dma_free_chan);
+
+/* pasemi_dma_alloc_ring - Allocate descriptor ring for a channel
+ * @chan: Channel for which to allocate
+ * @ring_size: Ring size in 64-bit (8-byte) words
+ *
+ * Allocate a descriptor ring for a channel. Returns 0 on success, errno
+ * on failure. The passed in struct pasemi_dmachan is updated with the
+ * virtual and DMA addresses of the ring.
+ */
+int pasemi_dma_alloc_ring(struct pasemi_dmachan *chan, int ring_size)
+{
+	BUG_ON(chan->ring_virt);
+
+	chan->ring_size = ring_size;
+
+	chan->ring_virt = dma_alloc_coherent(&dma_pdev->dev,
+					     ring_size * sizeof(u64),
+					     &chan->ring_dma, GFP_KERNEL);
+
+	if (!chan->ring_virt)
+		return -ENOMEM;
+
+	memset(chan->ring_virt, 0, ring_size * sizeof(u64));
+
+	return 0;
+}
+EXPORT_SYMBOL(pasemi_dma_alloc_ring);
+
+/* pasemi_dma_free_ring - Free an allocated descriptor ring for a channel
+ * @chan: Channel for which to free the descriptor ring
+ *
+ * Frees a previously allocated descriptor ring for a channel.
+ */
+void pasemi_dma_free_ring(struct pasemi_dmachan *chan)
+{
+	BUG_ON(!chan->ring_virt);
+
+	dma_free_coherent(&dma_pdev->dev, chan->ring_size * sizeof(u64),
+			  chan->ring_virt, chan->ring_dma);
+	chan->ring_virt = NULL;
+	chan->ring_size = 0;
+	chan->ring_dma = 0;
+}
+EXPORT_SYMBOL(pasemi_dma_free_ring);
+
+/* pasemi_dma_start_chan - Start a DMA channel
+ * @chan: Channel to start
+ * @cmdsta: Additional CCMDSTA/TCMDSTA bits to write
+ *
+ * Enables (starts) a DMA channel with optional additional arguments.
+ */
+void pasemi_dma_start_chan(const struct pasemi_dmachan *chan, const u32 cmdsta)
+{
+	if (chan->chan_type == RXCHAN)
+		pasemi_write_dma_reg(PAS_DMA_RXCHAN_CCMDSTA(chan->chno),
+				     cmdsta | PAS_DMA_RXCHAN_CCMDSTA_EN);
+	else
+		pasemi_write_dma_reg(PAS_DMA_TXCHAN_TCMDSTA(chan->chno),
+				     cmdsta | PAS_DMA_TXCHAN_TCMDSTA_EN);
+}
+EXPORT_SYMBOL(pasemi_dma_start_chan);
+
+/* pasemi_dma_stop_chan - Stop a DMA channel
+ * @chan: Channel to stop
+ *
+ * Stops (disables) a DMA channel. This is done by setting the ST bit in the
+ * CMDSTA register and waiting on the ACT (active) bit to clear, then
+ * finally disabling the whole channel.
+ *
+ * This function will only try for a short while for the channel to stop, if
+ * it doesn't it will return failure.
+ *
+ * Returns 1 on success, 0 on failure.
+ */
+#define MAX_RETRIES 5000
+int pasemi_dma_stop_chan(const struct pasemi_dmachan *chan)
+{
+	int reg, retries;
+	u32 sta;
+
+	if (chan->chan_type == RXCHAN) {
+		reg = PAS_DMA_RXCHAN_CCMDSTA(chan->chno);
+		pasemi_write_dma_reg(reg, PAS_DMA_RXCHAN_CCMDSTA_ST);
+		for (retries = 0; retries < MAX_RETRIES; retries++) {
+			sta = pasemi_read_dma_reg(reg);
+			if (!(sta & PAS_DMA_RXCHAN_CCMDSTA_ACT)) {
+				pasemi_write_dma_reg(reg, 0);
+				return 1;
+			}
+			cond_resched();
+		}
+	} else {
+		reg = PAS_DMA_TXCHAN_TCMDSTA(chan->chno);
+		pasemi_write_dma_reg(reg, PAS_DMA_TXCHAN_TCMDSTA_ST);
+		for (retries = 0; retries < MAX_RETRIES; retries++) {
+			sta = pasemi_read_dma_reg(reg);
+			if (!(sta & PAS_DMA_TXCHAN_TCMDSTA_ACT)) {
+				pasemi_write_dma_reg(reg, 0);
+				return 1;
+			}
+			cond_resched();
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(pasemi_dma_stop_chan);
+
+/* pasemi_dma_alloc_buf - Allocate a buffer to use for DMA
+ * @chan: Channel to allocate for
+ * @size: Size of buffer in bytes
+ * @handle: DMA handle
+ *
+ * Allocate a buffer to be used by the DMA engine for read/write,
+ * similar to dma_alloc_coherent().
+ *
+ * Returns the virtual address of the buffer, or NULL in case of failure.
+ */
+void *pasemi_dma_alloc_buf(struct pasemi_dmachan *chan, int size,
+			   dma_addr_t *handle)
+{
+	return dma_alloc_coherent(&dma_pdev->dev, size, handle, GFP_KERNEL);
+}
+EXPORT_SYMBOL(pasemi_dma_alloc_buf);
+
+/* pasemi_dma_free_buf - Free a buffer used for DMA
+ * @chan: Channel the buffer was allocated for
+ * @size: Size of buffer in bytes
+ * @handle: DMA handle
+ *
+ * Frees a previously allocated buffer.
+ */
+void pasemi_dma_free_buf(struct pasemi_dmachan *chan, int size,
+			 dma_addr_t *handle)
+{
+	dma_free_coherent(&dma_pdev->dev, size, handle, GFP_KERNEL);
+}
+EXPORT_SYMBOL(pasemi_dma_free_buf);
+
+static void *map_onedev(struct pci_dev *p, int index)
+{
+	struct device_node *dn;
+	void __iomem *ret;
+
+	dn = pci_device_to_OF_node(p);
+	if (!dn)
+		goto fallback;
+
+	ret = of_iomap(dn, index);
+	if (!ret)
+		goto fallback;
+
+	return ret;
+fallback:
+	/* This is hardcoded and ugly, but we have some firmware versions
+	 * that don't provide the register space in the device tree. Luckily
+	 * they are at well-known locations so we can just do the math here.
+	 */
+	return ioremap(0xe0000000 + (p->devfn << 12), 0x2000);
+}
+
+/* pasemi_dma_init - Initialize the PA Semi DMA library
+ *
+ * This function initializes the DMA library. It must be called before
+ * any other function in the library.
+ *
+ * Returns 0 on success, errno on failure.
+ */
+int pasemi_dma_init(void)
+{
+	static spinlock_t init_lock = SPIN_LOCK_UNLOCKED;
+	struct pci_dev *iob_pdev;
+	struct pci_dev *pdev;
+	struct resource res;
+	struct device_node *dn;
+	int i, intf, err = 0;
+	u32 tmp;
+
+	if (!machine_is(pasemi))
+		return -ENODEV;
+
+	spin_lock(&init_lock);
+
+	/* Make sure we haven't already initialized */
+	if (dma_pdev)
+		goto out;
+
+	iob_pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa001, NULL);
+	if (!iob_pdev) {
+		BUG();
+		printk(KERN_WARNING "Can't find I/O Bridge\n");
+		err = -ENODEV;
+		goto out;
+	}
+	iob_regs = map_onedev(iob_pdev, 0);
+
+	dma_pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa007, NULL);
+	if (!dma_pdev) {
+		BUG();
+		printk(KERN_WARNING "Can't find DMA controller\n");
+		err = -ENODEV;
+		goto out;
+	}
+	dma_regs = map_onedev(dma_pdev, 0);
+	base_hw_irq = virq_to_hw(dma_pdev->irq);
+
+	pci_read_config_dword(dma_pdev, PAS_DMA_CAP_TXCH, &tmp);
+	num_txch = (tmp & PAS_DMA_CAP_TXCH_TCHN_M) >> PAS_DMA_CAP_TXCH_TCHN_S;
+
+	pci_read_config_dword(dma_pdev, PAS_DMA_CAP_RXCH, &tmp);
+	num_rxch = (tmp & PAS_DMA_CAP_RXCH_RCHN_M) >> PAS_DMA_CAP_RXCH_RCHN_S;
+
+	intf = 0;
+	for (pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa006, NULL);
+	     pdev;
+	     pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa006, pdev))
+		mac_regs[intf++] = map_onedev(pdev, 0);
+
+	pci_dev_put(pdev);
+
+	for (pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa005, NULL);
+	     pdev;
+	     pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa005, pdev))
+		mac_regs[intf++] = map_onedev(pdev, 0);
+
+	pci_dev_put(pdev);
+
+	dn = pci_device_to_OF_node(iob_pdev);
+	if (dn)
+		err = of_address_to_resource(dn, 1, &res);
+	if (!dn || err) {
+		/* Fallback for old firmware */
+		res.start = 0xfd800000;
+		res.end = res.start + 0x1000;
+	}
+	dma_status = __ioremap(res.start, res.end-res.start, 0);
+	pci_dev_put(iob_pdev);
+
+	for (i = 0; i < MAX_TXCH; i++)
+		__set_bit(i, txch_free);
+
+	for (i = 0; i < MAX_RXCH; i++)
+		__set_bit(i, rxch_free);
+
+	printk(KERN_INFO "PA Semi PWRficient DMA library initialized "
+		"(%d tx, %d rx channels)\n", num_txch, num_rxch);
+
+out:
+	spin_unlock(&init_lock);
+	return err;
+}
diff --git a/arch/powerpc/platforms/pasemi/pasemi.h b/arch/powerpc/platforms/pasemi/pasemi.h
index 516acabb4e96..58b344c6fc38 100644
--- a/arch/powerpc/platforms/pasemi/pasemi.h
+++ b/arch/powerpc/platforms/pasemi/pasemi.h
@@ -9,6 +9,7 @@ extern void __devinit pas_pci_dma_dev_setup(struct pci_dev *dev);
 extern void __iomem *pasemi_pci_getcfgaddr(struct pci_dev *dev, int offset);
 
 extern void __init alloc_iobmap_l2(void);
+extern void __init pasemi_map_registers(void);
 
 /* Power savings modes, implemented in asm */
 extern void idle_spin(void);
diff --git a/include/asm-powerpc/pasemi_dma.h b/include/asm-powerpc/pasemi_dma.h
index 8ef80d8bdecd..b4526ff3a50d 100644
--- a/include/asm-powerpc/pasemi_dma.h
+++ b/include/asm-powerpc/pasemi_dma.h
@@ -33,11 +33,27 @@ struct pasdma_status {
  * device. Use the normal PCI config access functions for them.
  */
 enum {
+	PAS_DMA_CAP_TXCH  = 0x44,	/* Transmit Channel Info      */
+	PAS_DMA_CAP_RXCH  = 0x48,	/* Transmit Channel Info      */
+	PAS_DMA_CAP_IFI	  = 0x4c,	/* Interface Info	      */
 	PAS_DMA_COM_TXCMD = 0x100,	/* Transmit Command Register  */
 	PAS_DMA_COM_TXSTA = 0x104,	/* Transmit Status Register   */
 	PAS_DMA_COM_RXCMD = 0x108,	/* Receive Command Register   */
 	PAS_DMA_COM_RXSTA = 0x10c,	/* Receive Status Register    */
 };
+
+
+#define PAS_DMA_CAP_TXCH_TCHN_M	0x00ff0000 /* # of TX channels */
+#define PAS_DMA_CAP_TXCH_TCHN_S	16
+
+#define PAS_DMA_CAP_RXCH_RCHN_M	0x00ff0000 /* # of RX channels */
+#define PAS_DMA_CAP_RXCH_RCHN_S	16
+
+#define PAS_DMA_CAP_IFI_IOFF_M	0xff000000 /* Cfg reg for intf pointers */
+#define PAS_DMA_CAP_IFI_IOFF_S	24
+#define PAS_DMA_CAP_IFI_NIN_M	0x00ff0000 /* # of interfaces */
+#define PAS_DMA_CAP_IFI_NIN_S	16
+
 #define PAS_DMA_COM_TXCMD_EN	0x00000001 /* enable */
 #define PAS_DMA_COM_TXSTA_ACT	0x00000001 /* active */
 #define PAS_DMA_COM_RXCMD_EN	0x00000001 /* enable */
@@ -388,4 +404,64 @@ enum {
 				 CTRL_CMD_REG_M)
 
 
+
+/* Prototypes for the shared DMA functions in the platform code. */
+
+/* DMA TX Channel type. Right now only limitations used are event types 0/1,
+ * for event-triggered DMA transactions.
+ */
+
+enum pasemi_dmachan_type {
+	RXCHAN = 0,		/* Any RX chan */
+	TXCHAN = 1,		/* Any TX chan */
+	TXCHAN_EVT0 = 0x1001,	/* TX chan in event class 0 (chan 0-9) */
+	TXCHAN_EVT1 = 0x2001,	/* TX chan in event class 1 (chan 10-19) */
+};
+
+struct pasemi_dmachan {
+	int		 chno;		/* Channel number */
+	enum pasemi_dmachan_type chan_type;	/* TX / RX */
+	u64		*status;	/* Ptr to cacheable status */
+	int		 irq;		/* IRQ used by channel */
+	unsigned int	 ring_size;	/* size of allocated ring */
+	dma_addr_t	 ring_dma;	/* DMA address for ring */
+	u64		*ring_virt;	/* Virt address for ring */
+	void		*priv;		/* Ptr to start of client struct */
+};
+
+/* Read/write the different registers in the I/O Bridge, Ethernet
+ * and DMA Controller
+ */
+extern unsigned int pasemi_read_iob_reg(unsigned int reg);
+extern void pasemi_write_iob_reg(unsigned int reg, unsigned int val);
+
+extern unsigned int pasemi_read_mac_reg(int intf, unsigned int reg);
+extern void pasemi_write_mac_reg(int intf, unsigned int reg, unsigned int val);
+
+extern unsigned int pasemi_read_dma_reg(unsigned int reg);
+extern void pasemi_write_dma_reg(unsigned int reg, unsigned int val);
+
+/* Channel management routines */
+
+extern void *pasemi_dma_alloc_chan(enum pasemi_dmachan_type type,
+				   int total_size, int offset);
+extern void pasemi_dma_free_chan(struct pasemi_dmachan *chan);
+
+extern void pasemi_dma_start_chan(const struct pasemi_dmachan *chan,
+				  const u32 cmdsta);
+extern int pasemi_dma_stop_chan(const struct pasemi_dmachan *chan);
+
+/* Common routines to allocate rings and buffers */
+
+extern int pasemi_dma_alloc_ring(struct pasemi_dmachan *chan, int ring_size);
+extern void pasemi_dma_free_ring(struct pasemi_dmachan *chan);
+
+extern void *pasemi_dma_alloc_buf(struct pasemi_dmachan *chan, int size,
+				  dma_addr_t *handle);
+extern void pasemi_dma_free_buf(struct pasemi_dmachan *chan, int size,
+				dma_addr_t *handle);
+
+/* Initialize the library, must be called before any other functions */
+extern int pasemi_dma_init(void);
+
 #endif /* ASM_PASEMI_DMA_H */
-- 
cgit v1.2.3


From ac82fab44f6b981e3f6b53784e2f23838f4908e1 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Fri, 9 Nov 2007 16:54:45 -0600
Subject: ssb: Add new SPROM structure while keeping the old

The SPROM's for various devices utilizing the Sonics Silicon Backplane come
with various revisions. The Revision 2 SPROM inherited the data layout of 1, and
Revision 3 inherited the layout of 2. The first instance of Revision 4 has
now been found in a BCM4328 wireless LAN card. This device does not inherit any
layout from previous versions. Although it was possible to create a data
structure that kept all the old layouts, we decided to start fresh, keep only
those SPROM variables that are used by the drivers that utilize ssb, and to
do the conversion in such a manner that neither compilation or execution will
be affected if a bisection lands in the middle of these changes, while keeping
the patches as small as possible.

In this patch, the sprom structures are changed while maintaining the old ones.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/pci.c            |  2 --
 include/linux/ssb/ssb.h      | 32 ++++++++++++++++++++++++++------
 include/linux/ssb/ssb_regs.h | 32 ++++++++++++++++++++++++++++++++
 3 files changed, 58 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index 0ab095c6581a..7226a716acdf 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -423,8 +423,6 @@ static int sprom_extract(struct ssb_bus *bus,
 	memset(out, 0, sizeof(*out));
 
 	SPEX(revision, SSB_SPROM_REVISION, SSB_SPROM_REVISION_REV, 0);
-	SPEX(crc, SSB_SPROM_REVISION, SSB_SPROM_REVISION_CRC,
-	     SSB_SPROM_REVISION_CRC_SHIFT);
 
 	if ((bus->chip_id & 0xFF00) == 0x4400) {
 		/* Workaround: The BCM44XX chip has a stupid revision
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 2b5c312c4960..cdd8a2fd4a69 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -78,13 +78,34 @@ struct ssb_sprom_r3 {
 	u32 ofdmgpo;		/* G-PHY OFDM Power Offset */
 };
 
-struct ssb_sprom_r4 {
-	/* TODO */
-};
-
 struct ssb_sprom {
 	u8 revision;
-	u8 crc;
+	u8 temp_fill[2 * sizeof(struct ssb_sprom_r1)];
+	u8 il0mac[6];		/* MAC address for 802.11b/g */
+	u8 et0mac[6];		/* MAC address for Ethernet */
+	u8 et1mac[6];		/* MAC address for 802.11a */
+	u8 et0phyaddr;		/* MII address for enet0 */
+	u8 et1phyaddr;		/* MII address for enet1 */
+	u8 country_code;	/* Country Code */
+	u16 pa0b0;
+	u16 pa0b1;
+	u16 pa0b2;
+	u16 pa1b0;
+	u16 pa1b1;
+	u16 pa1b2;
+	u8 gpio0;		/* GPIO pin 0 */
+	u8 gpio1;		/* GPIO pin 1 */
+	u8 gpio2;		/* GPIO pin 2 */
+	u8 gpio3;		/* GPIO pin 3 */
+	u16 maxpwr_a;		/* A-PHY Amplifier Max Power (in dBm Q5.2) */
+	u16 maxpwr_bg;		/* B/G-PHY Amplifier Max Power (in dBm Q5.2) */
+	u8 itssi_a;		/* Idle TSSI Target for A-PHY */
+	u8 itssi_bg;		/* Idle TSSI Target for B/G-PHY */
+	u16 boardflags_lo;	/* Boardflags (low 16 bits) */
+	u8 antenna_gain_a;	/* A-PHY Antenna gain (in dBm Q5.2) */
+	u8 antenna_gain_bg;	/* B/G-PHY Antenna gain (in dBm Q5.2) */
+
+	/* TODO - add any parameters needed from rev 2, 3, or 4 SPROMs */
 	/* The valid r# fields are selected by the "revision".
 	 * Revision 3 and lower inherit from lower revisions.
 	 */
@@ -94,7 +115,6 @@ struct ssb_sprom {
 			struct ssb_sprom_r2 r2;
 			struct ssb_sprom_r3 r3;
 		};
-		struct ssb_sprom_r4 r4;
 	};
 };
 
diff --git a/include/linux/ssb/ssb_regs.h b/include/linux/ssb/ssb_regs.h
index 47c7c71a5acf..bcebcffd448b 100644
--- a/include/linux/ssb/ssb_regs.h
+++ b/include/linux/ssb/ssb_regs.h
@@ -250,6 +250,38 @@
 #define  SSB_SPROM3_CCKPO_11M		0xF000	/* 11M Rate PO */
 #define  SSB_SPROM3_CCKPO_11M_SHIFT	12
 #define  SSB_SPROM3_OFDMGPO		0x107A	/* G-PHY OFDM Power Offset (4 bytes, BigEndian) */
+/* SPROM Revision 4 */
+#define SSB_SPROM4_IL0MAC		0x104C	/* 6 byte MAC address for b/g */
+#define SSB_SPROM4_ETHPHY		0x105A	/* Ethernet PHY settings */
+#define  SSB_SPROM4_ETHPHY_ET0A		0x001F	/* MII Address for enet0 */
+#define  SSB_SPROM4_ETHPHY_ET1A		0x03E0	/* MII Address for enet1 */
+#define  SSB_SPROM4_ETHPHY_ET1A_SHIFT	5
+#define  SSB_SPROM4_ETHPHY_ET0M		(1<<14)	/* MDIO for enet0 */
+#define  SSB_SPROM4_ETHPHY_ET1M		(1<<15)	/* MDIO for enet1 */
+#define SSB_SPROM4_CCODE		0x1052	/* Country Code (2 bytes) */
+#define SSB_SPROM4_ANT_A		0x105D  /* A Antennas */
+#define SSB_SPROM4_ANT_BG		0x105C  /* B/G Antennas */
+#define SSB_SPROM4_BFLLO		0x1044	/* Boardflags (low 16 bits) */
+#define SSB_SPROM4_AGAIN		0x105E	/* Antenna Gain (in dBm Q5.2) */
+#define SSB_SPROM4_BFLHI		0x1046  /* Board Flags Hi */
+#define SSB_SPROM4_MAXP_A		0x1000  /* Max Power A */
+#define SSB_SPROM4_MAXP_A_HI		0x00FF  /* Mask for Hi */
+#define SSB_SPROM4_MAXP_A_LO		0xFF00  /* Mask for Lo */
+#define SSB_SPROM4_MAXP_A_LO_SHIFT	16	/* Shift for Lo */
+#define SSB_SPROM4_PA1LOB0		0x1000
+#define SSB_SPROM4_PA1LOB1		0x1000
+#define SSB_SPROM4_PA1LOB2		0x1000
+#define SSB_SPROM4_PA1HIB0		0x1000
+#define SSB_SPROM4_PA1HIB1		0x1000
+#define SSB_SPROM4_PA1HIB2		0x1000
+#define SSB_SPROM4_OPO			0x1000
+#define SSB_SPROM4_OPO_VALUE		0x0000
+#define SSB_SPROM4_GPIOLDC		0x105A	/* LED Powersave Duty Cycle */
+#define  SSB_SPROM4_GPIOLDC_OFF		0x0000FF00	/* Off Count */
+#define  SSB_SPROM4_GPIOLDC_OFF_SHIFT	8
+#define  SSB_SPROM4_GPIOLDC_ON		0x00FF0000	/* On Count */
+#define  SSB_SPROM4_GPIOLDC_ON_SHIFT	16
+
 
 /* Values for SSB_SPROM1_BINF_CCODE */
 enum {
-- 
cgit v1.2.3


From c272ef4403c271799a7f09a4ab7a236c86643843 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Fri, 9 Nov 2007 16:56:25 -0600
Subject: ssb: Convert to use of the new SPROM structure

In disagreement with the SPROM specs, revision 3 devices appear to have
moved the MAC address.

Change ssb to handle the revision 4 SPROM, which is a different size.
This change in size is handled by adding a new variable to the ssb_sprom
struct and using it whenever possible. For those routines that do not
have access to this structure, a 'u16 size' argument is added.

The new PCI_ID for the BCM4328 is also added.

Testing of the Revision 4 SPROM, which is used on the BCM4328, was done
by Michael Gerdau <mgerdau@tiscali.de>.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/b43_pci_bridge.c |   1 +
 drivers/ssb/main.c           |  10 +-
 drivers/ssb/pci.c            | 224 +++++++++++++++++++++++++------------------
 include/linux/ssb/ssb.h      |   1 +
 include/linux/ssb/ssb_regs.h |  38 +++++---
 5 files changed, 170 insertions(+), 104 deletions(-)

(limited to 'include')

diff --git a/drivers/ssb/b43_pci_bridge.c b/drivers/ssb/b43_pci_bridge.c
index f145d8a4cfde..b8b7cb0b436a 100644
--- a/drivers/ssb/b43_pci_bridge.c
+++ b/drivers/ssb/b43_pci_bridge.c
@@ -27,6 +27,7 @@ static const struct pci_device_id b43_pci_bridge_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4321) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4324) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4325) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4328) },
 	{ 0, },
 };
 MODULE_DEVICE_TABLE(pci, b43_pci_bridge_tbl);
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 85a20546e827..9028ed5715a1 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -872,14 +872,22 @@ EXPORT_SYMBOL(ssb_clockspeed);
 
 static u32 ssb_tmslow_reject_bitmask(struct ssb_device *dev)
 {
+	u32 rev = ssb_read32(dev, SSB_IDLOW) & SSB_IDLOW_SSBREV;
+
 	/* The REJECT bit changed position in TMSLOW between
 	 * Backplane revisions. */
-	switch (ssb_read32(dev, SSB_IDLOW) & SSB_IDLOW_SSBREV) {
+	switch (rev) {
 	case SSB_IDLOW_SSBREV_22:
 		return SSB_TMSLOW_REJECT_22;
 	case SSB_IDLOW_SSBREV_23:
 		return SSB_TMSLOW_REJECT_23;
+	case SSB_IDLOW_SSBREV_24:     /* TODO - find the proper REJECT bits */
+	case SSB_IDLOW_SSBREV_25:     /* same here */
+	case SSB_IDLOW_SSBREV_26:     /* same here */
+	case SSB_IDLOW_SSBREV_27:     /* same here */
+		return SSB_TMSLOW_REJECT_23;	/* this is a guess */
 	default:
+		printk(KERN_INFO "ssb: Backplane Revision 0x%.8X\n", rev);
 		WARN_ON(1);
 	}
 	return (SSB_TMSLOW_REJECT_22 | SSB_TMSLOW_REJECT_23);
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index 7226a716acdf..8f600fcd2430 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -212,29 +212,29 @@ static inline u8 ssb_crc8(u8 crc, u8 data)
 	return t[crc ^ data];
 }
 
-static u8 ssb_sprom_crc(const u16 *sprom)
+static u8 ssb_sprom_crc(const u16 *sprom, u16 size)
 {
 	int word;
 	u8 crc = 0xFF;
 
-	for (word = 0; word < SSB_SPROMSIZE_WORDS - 1; word++) {
+	for (word = 0; word < size - 1; word++) {
 		crc = ssb_crc8(crc, sprom[word] & 0x00FF);
 		crc = ssb_crc8(crc, (sprom[word] & 0xFF00) >> 8);
 	}
-	crc = ssb_crc8(crc, sprom[SPOFF(SSB_SPROM_REVISION)] & 0x00FF);
+	crc = ssb_crc8(crc, sprom[size - 1] & 0x00FF);
 	crc ^= 0xFF;
 
 	return crc;
 }
 
-static int sprom_check_crc(const u16 *sprom)
+static int sprom_check_crc(const u16 *sprom, u16 size)
 {
 	u8 crc;
 	u8 expected_crc;
 	u16 tmp;
 
-	crc = ssb_sprom_crc(sprom);
-	tmp = sprom[SPOFF(SSB_SPROM_REVISION)] & SSB_SPROM_REVISION_CRC;
+	crc = ssb_sprom_crc(sprom, size);
+	tmp = sprom[size - 1] & SSB_SPROM_REVISION_CRC;
 	expected_crc = tmp >> SSB_SPROM_REVISION_CRC_SHIFT;
 	if (crc != expected_crc)
 		return -EPROTO;
@@ -246,7 +246,7 @@ static void sprom_do_read(struct ssb_bus *bus, u16 *sprom)
 {
 	int i;
 
-	for (i = 0; i < SSB_SPROMSIZE_WORDS; i++)
+	for (i = 0; i < bus->sprom_size; i++)
 		sprom[i] = readw(bus->mmio + SSB_SPROM_BASE + (i * 2));
 }
 
@@ -255,6 +255,7 @@ static int sprom_do_write(struct ssb_bus *bus, const u16 *sprom)
 	struct pci_dev *pdev = bus->host_pci;
 	int i, err;
 	u32 spromctl;
+	u16 size = bus->sprom_size;
 
 	ssb_printk(KERN_NOTICE PFX "Writing SPROM. Do NOT turn off the power! Please stand by...\n");
 	err = pci_read_config_dword(pdev, SSB_SPROMCTL, &spromctl);
@@ -266,12 +267,12 @@ static int sprom_do_write(struct ssb_bus *bus, const u16 *sprom)
 		goto err_ctlreg;
 	ssb_printk(KERN_NOTICE PFX "[ 0%%");
 	msleep(500);
-	for (i = 0; i < SSB_SPROMSIZE_WORDS; i++) {
-		if (i == SSB_SPROMSIZE_WORDS / 4)
+	for (i = 0; i < size; i++) {
+		if (i == size / 4)
 			ssb_printk("25%%");
-		else if (i == SSB_SPROMSIZE_WORDS / 2)
+		else if (i == size / 2)
 			ssb_printk("50%%");
-		else if (i == (SSB_SPROMSIZE_WORDS / 4) * 3)
+		else if (i == (size * 3) / 4)
 			ssb_printk("75%%");
 		else if (i % 2)
 			ssb_printk(".");
@@ -350,95 +351,120 @@ static void sprom_extract_r1(struct ssb_sprom_r1 *out, const u16 *in)
 	SPEX(antenna_gain_a, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_A, 0);
 	SPEX(antenna_gain_bg, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_BG,
 	     SSB_SPROM1_AGAIN_BG_SHIFT);
-	for (i = 0; i < 4; i++) {
-		v = in[SPOFF(SSB_SPROM1_OEM) + i];
-		*(((__le16 *)out->oem) + i) = cpu_to_le16(v);
-	}
 }
 
-static void sprom_extract_r2(struct ssb_sprom_r2 *out, const u16 *in)
+static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 {
 	int i;
 	u16 v;
+	u16 loc[3];
 
-	SPEX(boardflags_hi, SSB_SPROM2_BFLHI,  0xFFFF, 0);
-	SPEX(maxpwr_a_hi, SSB_SPROM2_MAXP_A, SSB_SPROM2_MAXP_A_HI, 0);
-	SPEX(maxpwr_a_lo, SSB_SPROM2_MAXP_A, SSB_SPROM2_MAXP_A_LO,
-	     SSB_SPROM2_MAXP_A_LO_SHIFT);
-	SPEX(pa1lob0, SSB_SPROM2_PA1LOB0, 0xFFFF, 0);
-	SPEX(pa1lob1, SSB_SPROM2_PA1LOB1, 0xFFFF, 0);
-	SPEX(pa1lob2, SSB_SPROM2_PA1LOB2, 0xFFFF, 0);
-	SPEX(pa1hib0, SSB_SPROM2_PA1HIB0, 0xFFFF, 0);
-	SPEX(pa1hib1, SSB_SPROM2_PA1HIB1, 0xFFFF, 0);
-	SPEX(pa1hib2, SSB_SPROM2_PA1HIB2, 0xFFFF, 0);
-	SPEX(ofdm_pwr_off, SSB_SPROM2_OPO, SSB_SPROM2_OPO_VALUE, 0);
-	for (i = 0; i < 4; i++) {
-		v = in[SPOFF(SSB_SPROM2_CCODE) + i];
-		*(((__le16 *)out->country_str) + i) = cpu_to_le16(v);
+	if (out->revision == 3) {			/* rev 3 moved MAC */
+		loc[0] = SSB_SPROM3_IL0MAC;
+		loc[1] = SSB_SPROM3_ET0MAC;
+		loc[2] = SSB_SPROM3_ET1MAC;
+	} else {
+		loc[0] = SSB_SPROM1_IL0MAC;
+		loc[1] = SSB_SPROM1_ET0MAC;
+		loc[2] = SSB_SPROM1_ET1MAC;
+	}
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(loc[0]) + i];
+		*(((__be16 *)out->il0mac) + i) = cpu_to_be16(v);
+	}
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(loc[1]) + i];
+		*(((__be16 *)out->et0mac) + i) = cpu_to_be16(v);
+	}
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(loc[2]) + i];
+		*(((__be16 *)out->et1mac) + i) = cpu_to_be16(v);
 	}
+	SPEX(et0phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET0A, 0);
+	SPEX(et1phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET1A,
+	     SSB_SPROM1_ETHPHY_ET1A_SHIFT);
+	SPEX(country_code, SSB_SPROM1_BINF, SSB_SPROM1_BINF_CCODE,
+	     SSB_SPROM1_BINF_CCODE_SHIFT);
+	SPEX(pa0b0, SSB_SPROM1_PA0B0, 0xFFFF, 0);
+	SPEX(pa0b1, SSB_SPROM1_PA0B1, 0xFFFF, 0);
+	SPEX(pa0b2, SSB_SPROM1_PA0B2, 0xFFFF, 0);
+	SPEX(pa1b0, SSB_SPROM1_PA1B0, 0xFFFF, 0);
+	SPEX(pa1b1, SSB_SPROM1_PA1B1, 0xFFFF, 0);
+	SPEX(pa1b2, SSB_SPROM1_PA1B2, 0xFFFF, 0);
+	SPEX(gpio0, SSB_SPROM1_GPIOA, SSB_SPROM1_GPIOA_P0, 0);
+	SPEX(gpio1, SSB_SPROM1_GPIOA, SSB_SPROM1_GPIOA_P1,
+	     SSB_SPROM1_GPIOA_P1_SHIFT);
+	SPEX(gpio2, SSB_SPROM1_GPIOB, SSB_SPROM1_GPIOB_P2, 0);
+	SPEX(gpio3, SSB_SPROM1_GPIOB, SSB_SPROM1_GPIOB_P3,
+	     SSB_SPROM1_GPIOB_P3_SHIFT);
+	SPEX(maxpwr_a, SSB_SPROM1_MAXPWR, SSB_SPROM1_MAXPWR_A,
+	     SSB_SPROM1_MAXPWR_A_SHIFT);
+	SPEX(maxpwr_bg, SSB_SPROM1_MAXPWR, SSB_SPROM1_MAXPWR_BG, 0);
+	SPEX(itssi_a, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_A,
+	     SSB_SPROM1_ITSSI_A_SHIFT);
+	SPEX(itssi_bg, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_BG, 0);
+	SPEX(boardflags_lo, SSB_SPROM1_BFLLO, 0xFFFF, 0);
+	SPEX(antenna_gain_a, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_A, 0);
+	SPEX(antenna_gain_bg, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_BG,
+	     SSB_SPROM1_AGAIN_BG_SHIFT);
 }
 
-static void sprom_extract_r3(struct ssb_sprom_r3 *out, const u16 *in)
+static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 {
-	out->ofdmapo  = (in[SPOFF(SSB_SPROM3_OFDMAPO) + 0] & 0xFF00) >> 8;
-	out->ofdmapo |= (in[SPOFF(SSB_SPROM3_OFDMAPO) + 0] & 0x00FF) << 8;
-	out->ofdmapo <<= 16;
-	out->ofdmapo |= (in[SPOFF(SSB_SPROM3_OFDMAPO) + 1] & 0xFF00) >> 8;
-	out->ofdmapo |= (in[SPOFF(SSB_SPROM3_OFDMAPO) + 1] & 0x00FF) << 8;
-
-	out->ofdmalpo  = (in[SPOFF(SSB_SPROM3_OFDMALPO) + 0] & 0xFF00) >> 8;
-	out->ofdmalpo |= (in[SPOFF(SSB_SPROM3_OFDMALPO) + 0] & 0x00FF) << 8;
-	out->ofdmalpo <<= 16;
-	out->ofdmalpo |= (in[SPOFF(SSB_SPROM3_OFDMALPO) + 1] & 0xFF00) >> 8;
-	out->ofdmalpo |= (in[SPOFF(SSB_SPROM3_OFDMALPO) + 1] & 0x00FF) << 8;
-
-	out->ofdmahpo  = (in[SPOFF(SSB_SPROM3_OFDMAHPO) + 0] & 0xFF00) >> 8;
-	out->ofdmahpo |= (in[SPOFF(SSB_SPROM3_OFDMAHPO) + 0] & 0x00FF) << 8;
-	out->ofdmahpo <<= 16;
-	out->ofdmahpo |= (in[SPOFF(SSB_SPROM3_OFDMAHPO) + 1] & 0xFF00) >> 8;
-	out->ofdmahpo |= (in[SPOFF(SSB_SPROM3_OFDMAHPO) + 1] & 0x00FF) << 8;
-
-	SPEX(gpioldc_on_cnt, SSB_SPROM3_GPIOLDC, SSB_SPROM3_GPIOLDC_ON,
-	     SSB_SPROM3_GPIOLDC_ON_SHIFT);
-	SPEX(gpioldc_off_cnt, SSB_SPROM3_GPIOLDC, SSB_SPROM3_GPIOLDC_OFF,
-	     SSB_SPROM3_GPIOLDC_OFF_SHIFT);
-	SPEX(cckpo_1M, SSB_SPROM3_CCKPO, SSB_SPROM3_CCKPO_1M, 0);
-	SPEX(cckpo_2M, SSB_SPROM3_CCKPO, SSB_SPROM3_CCKPO_2M,
-	     SSB_SPROM3_CCKPO_2M_SHIFT);
-	SPEX(cckpo_55M, SSB_SPROM3_CCKPO, SSB_SPROM3_CCKPO_55M,
-	     SSB_SPROM3_CCKPO_55M_SHIFT);
-	SPEX(cckpo_11M, SSB_SPROM3_CCKPO, SSB_SPROM3_CCKPO_11M,
-	     SSB_SPROM3_CCKPO_11M_SHIFT);
-
-	out->ofdmgpo  = (in[SPOFF(SSB_SPROM3_OFDMGPO) + 0] & 0xFF00) >> 8;
-	out->ofdmgpo |= (in[SPOFF(SSB_SPROM3_OFDMGPO) + 0] & 0x00FF) << 8;
-	out->ofdmgpo <<= 16;
-	out->ofdmgpo |= (in[SPOFF(SSB_SPROM3_OFDMGPO) + 1] & 0xFF00) >> 8;
-	out->ofdmgpo |= (in[SPOFF(SSB_SPROM3_OFDMGPO) + 1] & 0x00FF) << 8;
+	int i;
+	u16 v;
+
+	/* extract the r1 variables */
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(SSB_SPROM4_IL0MAC) + i];
+		*(((__be16 *)out->il0mac) + i) = cpu_to_be16(v);
+	}
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(SSB_SPROM4_ET0MAC) + i];
+		*(((__be16 *)out->et0mac) + i) = cpu_to_be16(v);
+	}
+	for (i = 0; i < 3; i++) {
+		v = in[SPOFF(SSB_SPROM4_ET1MAC) + i];
+		*(((__be16 *)out->et1mac) + i) = cpu_to_be16(v);
+	}
+	SPEX(et0phyaddr, SSB_SPROM4_ETHPHY, SSB_SPROM4_ETHPHY_ET0A, 0);
+	SPEX(et1phyaddr, SSB_SPROM4_ETHPHY, SSB_SPROM4_ETHPHY_ET1A,
+	     SSB_SPROM4_ETHPHY_ET1A_SHIFT);
+	SPEX(country_code, SSB_SPROM4_CCODE, 0xFFFF, 0);
+	SPEX(boardflags_lo, SSB_SPROM4_BFLLO, 0xFFFF, 0);
+	SPEX(antenna_gain_a, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_0, 0);
+	SPEX(antenna_gain_bg, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_1,
+	     SSB_SPROM4_AGAIN_1_SHIFT);
+	/* TODO - get remaining rev 4 stuff needed */
 }
 
-static int sprom_extract(struct ssb_bus *bus,
-			 struct ssb_sprom *out, const u16 *in)
+static int sprom_extract(struct ssb_bus *bus, struct ssb_sprom *out,
+			 const u16 *in, u16 size)
 {
 	memset(out, 0, sizeof(*out));
 
-	SPEX(revision, SSB_SPROM_REVISION, SSB_SPROM_REVISION_REV, 0);
-
+	out->revision = in[size - 1] & 0x00FF;
 	if ((bus->chip_id & 0xFF00) == 0x4400) {
 		/* Workaround: The BCM44XX chip has a stupid revision
 		 * number stored in the SPROM.
 		 * Always extract r1. */
+		out->revision = 1;
+		sprom_extract_r123(out, in);
 		sprom_extract_r1(&out->r1, in);
+	} else if (bus->chip_id == 0x4321) {
+		/* the BCM4328 has a chipid == 0x4321 and a rev 4 SPROM */
+		out->revision = 4;
+		sprom_extract_r4(out, in);
 	} else {
 		if (out->revision == 0)
 			goto unsupported;
-		if (out->revision >= 1 && out->revision <= 3)
+		if (out->revision >= 1 && out->revision <= 3) {
+			sprom_extract_r123(out, in);
 			sprom_extract_r1(&out->r1, in);
-		if (out->revision >= 2 && out->revision <= 3)
-			sprom_extract_r2(&out->r2, in);
-		if (out->revision == 3)
-			sprom_extract_r3(&out->r3, in);
-		if (out->revision >= 4)
+		}
+		if (out->revision == 4)
+			sprom_extract_r4(out, in);
+		if (out->revision >= 5)
 			goto unsupported;
 	}
 
@@ -456,16 +482,31 @@ static int ssb_pci_sprom_get(struct ssb_bus *bus,
 	int err = -ENOMEM;
 	u16 *buf;
 
-	buf = kcalloc(SSB_SPROMSIZE_WORDS, sizeof(u16), GFP_KERNEL);
+	buf = kcalloc(SSB_SPROMSIZE_WORDS_R123, sizeof(u16), GFP_KERNEL);
 	if (!buf)
 		goto out;
+	bus->sprom_size = SSB_SPROMSIZE_WORDS_R123;
 	sprom_do_read(bus, buf);
-	err = sprom_check_crc(buf);
+	err = sprom_check_crc(buf, bus->sprom_size);
 	if (err) {
-		ssb_printk(KERN_WARNING PFX
-			   "WARNING: Invalid SPROM CRC (corrupt SPROM)\n");
+		/* check for rev 4 sprom - has special signature */
+		if (buf [32] == 0x5372) {
+			ssb_printk(KERN_WARNING PFX "Extracting a rev 4"
+				   " SPROM\n");
+			kfree(buf);
+			buf = kcalloc(SSB_SPROMSIZE_WORDS_R4, sizeof(u16),
+				      GFP_KERNEL);
+			if (!buf)
+				goto out;
+			bus->sprom_size = SSB_SPROMSIZE_WORDS_R4;
+			sprom_do_read(bus, buf);
+			err = sprom_check_crc(buf, bus->sprom_size);
+		}
+		if (err)
+			ssb_printk(KERN_WARNING PFX "WARNING: Invalid"
+				   " SPROM CRC (corrupt SPROM)\n");
 	}
-	err = sprom_extract(bus, sprom, buf);
+	err = sprom_extract(bus, sprom, buf, bus->sprom_size);
 
 	kfree(buf);
 out:
@@ -579,29 +620,28 @@ const struct ssb_bus_ops ssb_pci_ops = {
 	.write32	= ssb_pci_write32,
 };
 
-static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len)
+static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len, u16 size)
 {
 	int i, pos = 0;
 
-	for (i = 0; i < SSB_SPROMSIZE_WORDS; i++) {
+	for (i = 0; i < size; i++)
 		pos += snprintf(buf + pos, buf_len - pos - 1,
 				"%04X", swab16(sprom[i]) & 0xFFFF);
-	}
 	pos += snprintf(buf + pos, buf_len - pos - 1, "\n");
 
 	return pos + 1;
 }
 
-static int hex2sprom(u16 *sprom, const char *dump, size_t len)
+static int hex2sprom(u16 *sprom, const char *dump, size_t len, u16 size)
 {
 	char tmp[5] = { 0 };
 	int cnt = 0;
 	unsigned long parsed;
 
-	if (len < SSB_SPROMSIZE_BYTES * 2)
+	if (len < size * 2)
 		return -EINVAL;
 
-	while (cnt < SSB_SPROMSIZE_WORDS) {
+	while (cnt < size) {
 		memcpy(tmp, dump, 4);
 		dump += 4;
 		parsed = simple_strtoul(tmp, NULL, 16);
@@ -625,7 +665,7 @@ static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev,
 	if (!bus)
 		goto out;
 	err = -ENOMEM;
-	sprom = kcalloc(SSB_SPROMSIZE_WORDS, sizeof(u16), GFP_KERNEL);
+	sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL);
 	if (!sprom)
 		goto out;
 
@@ -638,7 +678,7 @@ static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev,
 	sprom_do_read(bus, sprom);
 	mutex_unlock(&bus->pci_sprom_mutex);
 
-	count = sprom2hex(sprom, buf, PAGE_SIZE);
+	count = sprom2hex(sprom, buf, PAGE_SIZE, bus->sprom_size);
 	err = 0;
 
 out_kfree:
@@ -660,15 +700,15 @@ static ssize_t ssb_pci_attr_sprom_store(struct device *pcidev,
 	if (!bus)
 		goto out;
 	err = -ENOMEM;
-	sprom = kcalloc(SSB_SPROMSIZE_WORDS, sizeof(u16), GFP_KERNEL);
+	sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL);
 	if (!sprom)
 		goto out;
-	err = hex2sprom(sprom, buf, count);
+	err = hex2sprom(sprom, buf, count, bus->sprom_size);
 	if (err) {
 		err = -EINVAL;
 		goto out_kfree;
 	}
-	err = sprom_check_crc(sprom);
+	err = sprom_check_crc(sprom, bus->sprom_size);
 	if (err) {
 		err = -EINVAL;
 		goto out_kfree;
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index cdd8a2fd4a69..745de2aac85f 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -308,6 +308,7 @@ struct ssb_bus {
 	/* ID information about the Chip. */
 	u16 chip_id;
 	u16 chip_rev;
+	u16 sprom_size;		/* number of words in sprom */
 	u8 chip_package;
 
 	/* List of devices (cores) on the backplane. */
diff --git a/include/linux/ssb/ssb_regs.h b/include/linux/ssb/ssb_regs.h
index bcebcffd448b..96bba69b1271 100644
--- a/include/linux/ssb/ssb_regs.h
+++ b/include/linux/ssb/ssb_regs.h
@@ -147,6 +147,10 @@
 #define  SSB_IDLOW_SSBREV	0xF0000000 /* Sonics Backplane Revision code */
 #define  SSB_IDLOW_SSBREV_22	0x00000000 /* <= 2.2 */
 #define  SSB_IDLOW_SSBREV_23	0x10000000 /* 2.3 */
+#define  SSB_IDLOW_SSBREV_24	0x40000000 /* ?? Found in BCM4328 */
+#define  SSB_IDLOW_SSBREV_25	0x50000000 /* ?? Not Found yet */
+#define  SSB_IDLOW_SSBREV_26	0x60000000 /* ?? Found in some BCM4311/2 */
+#define  SSB_IDLOW_SSBREV_27	0x70000000 /* ?? Found in some BCM4311/2 */
 #define SSB_IDHIGH		0x0FFC     /* SB Identification High */
 #define  SSB_IDHIGH_RCLO	0x0000000F /* Revision Code (low part) */
 #define  SSB_IDHIGH_CC		0x00008FF0 /* Core Code */
@@ -162,6 +166,10 @@
  */
 #define SSB_SPROMSIZE_WORDS		64
 #define SSB_SPROMSIZE_BYTES		(SSB_SPROMSIZE_WORDS * sizeof(u16))
+#define SSB_SPROMSIZE_WORDS_R123	64
+#define SSB_SPROMSIZE_WORDS_R4		220
+#define SSB_SPROMSIZE_BYTES_R123	(SSB_SPROMSIZE_WORDS_R123 * sizeof(u16))
+#define SSB_SPROMSIZE_BYTES_R4		(SSB_SPROMSIZE_WORDS_R4 * sizeof(u16))
 #define SSB_SPROM_BASE			0x1000
 #define SSB_SPROM_REVISION		0x107E
 #define  SSB_SPROM_REVISION_REV		0x00FF	/* SPROM Revision number */
@@ -232,7 +240,10 @@
 #define  SSB_SPROM2_OPO_VALUE		0x00FF
 #define  SSB_SPROM2_OPO_UNUSED		0xFF00
 #define SSB_SPROM2_CCODE		0x107C	/* Two char Country Code */
-/* SPROM Revision 3 (inherits from rev 2) */
+/* SPROM Revision 3 (inherits most data from rev 2) */
+#define SSB_SPROM3_IL0MAC		0x104A	/* 6 bytes MAC address for 802.11b/g */
+#define SSB_SPROM3_ET0MAC		0x1050	/* 6 bytes MAC address for Ethernet ?? */
+#define SSB_SPROM3_ET1MAC		0x1050	/* 6 bytes MAC address for 802.11a ?? */
 #define SSB_SPROM3_OFDMAPO		0x102C	/* A-PHY OFDM Mid Power Offset (4 bytes, BigEndian) */
 #define SSB_SPROM3_OFDMALPO		0x1030	/* A-PHY OFDM Low Power Offset (4 bytes, BigEndian) */
 #define SSB_SPROM3_OFDMAHPO		0x1034	/* A-PHY OFDM High Power Offset (4 bytes, BigEndian) */
@@ -250,8 +261,10 @@
 #define  SSB_SPROM3_CCKPO_11M		0xF000	/* 11M Rate PO */
 #define  SSB_SPROM3_CCKPO_11M_SHIFT	12
 #define  SSB_SPROM3_OFDMGPO		0x107A	/* G-PHY OFDM Power Offset (4 bytes, BigEndian) */
-/* SPROM Revision 4 */
+/* SPROM Revision 4 		entries with ?? in comment are unknown */
 #define SSB_SPROM4_IL0MAC		0x104C	/* 6 byte MAC address for b/g */
+#define SSB_SPROM4_ET0MAC		0x1018	/* 6 bytes MAC address for Ethernet ?? */
+#define SSB_SPROM4_ET1MAC		0x1018	/* 6 bytes MAC address for 802.11a ?? */
 #define SSB_SPROM4_ETHPHY		0x105A	/* Ethernet PHY settings */
 #define  SSB_SPROM4_ETHPHY_ET0A		0x001F	/* MII Address for enet0 */
 #define  SSB_SPROM4_ETHPHY_ET1A		0x03E0	/* MII Address for enet1 */
@@ -263,19 +276,22 @@
 #define SSB_SPROM4_ANT_BG		0x105C  /* B/G Antennas */
 #define SSB_SPROM4_BFLLO		0x1044	/* Boardflags (low 16 bits) */
 #define SSB_SPROM4_AGAIN		0x105E	/* Antenna Gain (in dBm Q5.2) */
+#define  SSB_SPROM4_AGAIN_0		0x00FF	/* Antenna 0 */
+#define  SSB_SPROM4_AGAIN_1		0xFF00	/* Antenna 1 */
+#define  SSB_SPROM4_AGAIN_1_SHIFT	8
 #define SSB_SPROM4_BFLHI		0x1046  /* Board Flags Hi */
-#define SSB_SPROM4_MAXP_A		0x1000  /* Max Power A */
+#define SSB_SPROM4_MAXP_A		0x1000  /* Max Power A ?? */
 #define SSB_SPROM4_MAXP_A_HI		0x00FF  /* Mask for Hi */
 #define SSB_SPROM4_MAXP_A_LO		0xFF00  /* Mask for Lo */
 #define SSB_SPROM4_MAXP_A_LO_SHIFT	16	/* Shift for Lo */
-#define SSB_SPROM4_PA1LOB0		0x1000
-#define SSB_SPROM4_PA1LOB1		0x1000
-#define SSB_SPROM4_PA1LOB2		0x1000
-#define SSB_SPROM4_PA1HIB0		0x1000
-#define SSB_SPROM4_PA1HIB1		0x1000
-#define SSB_SPROM4_PA1HIB2		0x1000
-#define SSB_SPROM4_OPO			0x1000
-#define SSB_SPROM4_OPO_VALUE		0x0000
+#define SSB_SPROM4_PA1LOB0		0x1000  /* ?? */
+#define SSB_SPROM4_PA1LOB1		0x1000  /* ?? */
+#define SSB_SPROM4_PA1LOB2		0x1000  /* ?? */
+#define SSB_SPROM4_PA1HIB0		0x1000	/* ?? */
+#define SSB_SPROM4_PA1HIB1		0x1000	/* ?? */
+#define SSB_SPROM4_PA1HIB2		0x1000	/* ?? */
+#define SSB_SPROM4_OPO			0x1000	/* ?? */
+#define SSB_SPROM4_OPO_VALUE		0x0000	/* ?? */
 #define SSB_SPROM4_GPIOLDC		0x105A	/* LED Powersave Duty Cycle */
 #define  SSB_SPROM4_GPIOLDC_OFF		0x0000FF00	/* Off Count */
 #define  SSB_SPROM4_GPIOLDC_OFF_SHIFT	8
-- 
cgit v1.2.3


From d3c319f9c8d9ee2c042c60b8a1bbd909dcc42782 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Fri, 9 Nov 2007 16:58:20 -0600
Subject: ssb: Remove the old, now unused, data structures

The old, now unused, data structures and SPROM extraction routines
are removed.

Signed-off-by: Larry Finger<Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/pci.c            | 75 +++++++++-----------------------------------
 include/linux/ssb/ssb.h      | 74 -------------------------------------------
 include/linux/ssb/ssb_regs.h | 49 ++++++++++++++++-------------
 3 files changed, 43 insertions(+), 155 deletions(-)

(limited to 'include')

diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index 8f600fcd2430..9777dcb5bfe5 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -297,62 +297,6 @@ err_ctlreg:
 	return err;
 }
 
-static void sprom_extract_r1(struct ssb_sprom_r1 *out, const u16 *in)
-{
-	int i;
-	u16 v;
-
-	SPEX(pci_spid, SSB_SPROM1_SPID, 0xFFFF, 0);
-	SPEX(pci_svid, SSB_SPROM1_SVID, 0xFFFF, 0);
-	SPEX(pci_pid, SSB_SPROM1_PID, 0xFFFF, 0);
-	for (i = 0; i < 3; i++) {
-		v = in[SPOFF(SSB_SPROM1_IL0MAC) + i];
-		*(((__be16 *)out->il0mac) + i) = cpu_to_be16(v);
-	}
-	for (i = 0; i < 3; i++) {
-		v = in[SPOFF(SSB_SPROM1_ET0MAC) + i];
-		*(((__be16 *)out->et0mac) + i) = cpu_to_be16(v);
-	}
-	for (i = 0; i < 3; i++) {
-		v = in[SPOFF(SSB_SPROM1_ET1MAC) + i];
-		*(((__be16 *)out->et1mac) + i) = cpu_to_be16(v);
-	}
-	SPEX(et0phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET0A, 0);
-	SPEX(et1phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET1A,
-	     SSB_SPROM1_ETHPHY_ET1A_SHIFT);
-	SPEX(et0mdcport, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET0M, 14);
-	SPEX(et1mdcport, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET1M, 15);
-	SPEX(board_rev, SSB_SPROM1_BINF, SSB_SPROM1_BINF_BREV, 0);
-	SPEX(country_code, SSB_SPROM1_BINF, SSB_SPROM1_BINF_CCODE,
-	     SSB_SPROM1_BINF_CCODE_SHIFT);
-	SPEX(antenna_a, SSB_SPROM1_BINF, SSB_SPROM1_BINF_ANTA,
-	     SSB_SPROM1_BINF_ANTA_SHIFT);
-	SPEX(antenna_bg, SSB_SPROM1_BINF, SSB_SPROM1_BINF_ANTBG,
-	     SSB_SPROM1_BINF_ANTBG_SHIFT);
-	SPEX(pa0b0, SSB_SPROM1_PA0B0, 0xFFFF, 0);
-	SPEX(pa0b1, SSB_SPROM1_PA0B1, 0xFFFF, 0);
-	SPEX(pa0b2, SSB_SPROM1_PA0B2, 0xFFFF, 0);
-	SPEX(pa1b0, SSB_SPROM1_PA1B0, 0xFFFF, 0);
-	SPEX(pa1b1, SSB_SPROM1_PA1B1, 0xFFFF, 0);
-	SPEX(pa1b2, SSB_SPROM1_PA1B2, 0xFFFF, 0);
-	SPEX(gpio0, SSB_SPROM1_GPIOA, SSB_SPROM1_GPIOA_P0, 0);
-	SPEX(gpio1, SSB_SPROM1_GPIOA, SSB_SPROM1_GPIOA_P1,
-	     SSB_SPROM1_GPIOA_P1_SHIFT);
-	SPEX(gpio2, SSB_SPROM1_GPIOB, SSB_SPROM1_GPIOB_P2, 0);
-	SPEX(gpio3, SSB_SPROM1_GPIOB, SSB_SPROM1_GPIOB_P3,
-	     SSB_SPROM1_GPIOB_P3_SHIFT);
-	SPEX(maxpwr_a, SSB_SPROM1_MAXPWR, SSB_SPROM1_MAXPWR_A,
-	     SSB_SPROM1_MAXPWR_A_SHIFT);
-	SPEX(maxpwr_bg, SSB_SPROM1_MAXPWR, SSB_SPROM1_MAXPWR_BG, 0);
-	SPEX(itssi_a, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_A,
-	     SSB_SPROM1_ITSSI_A_SHIFT);
-	SPEX(itssi_bg, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_BG, 0);
-	SPEX(boardflags_lo, SSB_SPROM1_BFLLO, 0xFFFF, 0);
-	SPEX(antenna_gain_a, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_A, 0);
-	SPEX(antenna_gain_bg, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_BG,
-	     SSB_SPROM1_AGAIN_BG_SHIFT);
-}
-
 static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 {
 	int i;
@@ -414,7 +358,7 @@ static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 	int i;
 	u16 v;
 
-	/* extract the r1 variables */
+	/* extract the equivalent of the r1 variables */
 	for (i = 0; i < 3; i++) {
 		v = in[SPOFF(SSB_SPROM4_IL0MAC) + i];
 		*(((__be16 *)out->il0mac) + i) = cpu_to_be16(v);
@@ -435,6 +379,18 @@ static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 	SPEX(antenna_gain_a, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_0, 0);
 	SPEX(antenna_gain_bg, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_1,
 	     SSB_SPROM4_AGAIN_1_SHIFT);
+	SPEX(maxpwr_bg, SSB_SPROM4_MAXP_BG, SSB_SPROM4_MAXP_BG_MASK, 0);
+	SPEX(itssi_bg, SSB_SPROM4_MAXP_BG, SSB_SPROM4_ITSSI_BG,
+	     SSB_SPROM4_ITSSI_BG_SHIFT);
+	SPEX(maxpwr_a, SSB_SPROM4_MAXP_A, SSB_SPROM4_MAXP_A_MASK, 0);
+	SPEX(itssi_a, SSB_SPROM4_MAXP_A, SSB_SPROM4_ITSSI_A,
+	     SSB_SPROM4_ITSSI_A_SHIFT);
+	SPEX(gpio0, SSB_SPROM4_GPIOA, SSB_SPROM4_GPIOA_P0, 0);
+	SPEX(gpio1, SSB_SPROM4_GPIOA, SSB_SPROM4_GPIOA_P1,
+	     SSB_SPROM4_GPIOA_P1_SHIFT);
+	SPEX(gpio2, SSB_SPROM4_GPIOB, SSB_SPROM4_GPIOB_P2, 0);
+	SPEX(gpio3, SSB_SPROM4_GPIOB, SSB_SPROM4_GPIOB_P3,
+	     SSB_SPROM4_GPIOB_P3_SHIFT);
 	/* TODO - get remaining rev 4 stuff needed */
 }
 
@@ -444,13 +400,13 @@ static int sprom_extract(struct ssb_bus *bus, struct ssb_sprom *out,
 	memset(out, 0, sizeof(*out));
 
 	out->revision = in[size - 1] & 0x00FF;
+	ssb_printk(KERN_INFO PFX "SPROM revision %d detected.\n", out->revision);
 	if ((bus->chip_id & 0xFF00) == 0x4400) {
 		/* Workaround: The BCM44XX chip has a stupid revision
 		 * number stored in the SPROM.
 		 * Always extract r1. */
 		out->revision = 1;
 		sprom_extract_r123(out, in);
-		sprom_extract_r1(&out->r1, in);
 	} else if (bus->chip_id == 0x4321) {
 		/* the BCM4328 has a chipid == 0x4321 and a rev 4 SPROM */
 		out->revision = 4;
@@ -460,7 +416,6 @@ static int sprom_extract(struct ssb_bus *bus, struct ssb_sprom *out,
 			goto unsupported;
 		if (out->revision >= 1 && out->revision <= 3) {
 			sprom_extract_r123(out, in);
-			sprom_extract_r1(&out->r1, in);
 		}
 		if (out->revision == 4)
 			sprom_extract_r4(out, in);
@@ -472,7 +427,7 @@ static int sprom_extract(struct ssb_bus *bus, struct ssb_sprom *out,
 unsupported:
 	ssb_printk(KERN_WARNING PFX "Unsupported SPROM revision %d "
 		   "detected. Will extract v1\n", out->revision);
-	sprom_extract_r1(&out->r1, in);
+	sprom_extract_r123(out, in);
 	return 0;
 }
 
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 745de2aac85f..a21ab29ff363 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -15,72 +15,8 @@ struct pcmcia_device;
 struct ssb_bus;
 struct ssb_driver;
 
-
-struct ssb_sprom_r1 {
-	u16 pci_spid;		/* Subsystem Product ID for PCI */
-	u16 pci_svid;		/* Subsystem Vendor ID for PCI */
-	u16 pci_pid;		/* Product ID for PCI */
-	u8 il0mac[6];		/* MAC address for 802.11b/g */
-	u8 et0mac[6];		/* MAC address for Ethernet */
-	u8 et1mac[6];		/* MAC address for 802.11a */
-	u8 et0phyaddr:5;	/* MII address for enet0 */
-	u8 et1phyaddr:5;	/* MII address for enet1 */
-	u8 et0mdcport:1;	/* MDIO for enet0 */
-	u8 et1mdcport:1;	/* MDIO for enet1 */
-	u8 board_rev;		/* Board revision */
-	u8 country_code:4;	/* Country Code */
-	u8 antenna_a:2;		/* Antenna 0/1 available for A-PHY */
-	u8 antenna_bg:2;	/* Antenna 0/1 available for B-PHY and G-PHY */
-	u16 pa0b0;
-	u16 pa0b1;
-	u16 pa0b2;
-	u16 pa1b0;
-	u16 pa1b1;
-	u16 pa1b2;
-	u8 gpio0;		/* GPIO pin 0 */
-	u8 gpio1;		/* GPIO pin 1 */
-	u8 gpio2;		/* GPIO pin 2 */
-	u8 gpio3;		/* GPIO pin 3 */
-	u16 maxpwr_a;		/* A-PHY Power Amplifier Max Power (in dBm Q5.2) */
-	u16 maxpwr_bg;		/* B/G-PHY Power Amplifier Max Power (in dBm Q5.2) */
-	u8 itssi_a;		/* Idle TSSI Target for A-PHY */
-	u8 itssi_bg;		/* Idle TSSI Target for B/G-PHY */
-	u16 boardflags_lo;	/* Boardflags (low 16 bits) */
-	u8 antenna_gain_a;	/* A-PHY Antenna gain (in dBm Q5.2) */
-	u8 antenna_gain_bg;	/* B/G-PHY Antenna gain (in dBm Q5.2) */
-	u8 oem[8];		/* OEM string (rev 1 only) */
-};
-
-struct ssb_sprom_r2 {
-	u16 boardflags_hi;	/* Boardflags (high 16 bits) */
-	u8 maxpwr_a_lo;		/* A-PHY Max Power Low */
-	u8 maxpwr_a_hi;		/* A-PHY Max Power High */
-	u16 pa1lob0;		/* A-PHY PA Low Settings */
-	u16 pa1lob1;		/* A-PHY PA Low Settings */
-	u16 pa1lob2;		/* A-PHY PA Low Settings */
-	u16 pa1hib0;		/* A-PHY PA High Settings */
-	u16 pa1hib1;		/* A-PHY PA High Settings */
-	u16 pa1hib2;		/* A-PHY PA High Settings */
-	u8 ofdm_pwr_off;	/* OFDM Power Offset from CCK Level */
-	u8 country_str[2];	/* Two char Country Code */
-};
-
-struct ssb_sprom_r3 {
-	u32 ofdmapo;		/* A-PHY OFDM Mid Power Offset */
-	u32 ofdmalpo;		/* A-PHY OFDM Low Power Offset */
-	u32 ofdmahpo;		/* A-PHY OFDM High Power Offset */
-	u8 gpioldc_on_cnt;	/* GPIO LED Powersave Duty Cycle ON count */
-	u8 gpioldc_off_cnt;	/* GPIO LED Powersave Duty Cycle OFF count */
-	u8 cckpo_1M:4;		/* CCK Power Offset for Rate 1M */
-	u8 cckpo_2M:4;		/* CCK Power Offset for Rate 2M */
-	u8 cckpo_55M:4;		/* CCK Power Offset for Rate 5.5M */
-	u8 cckpo_11M:4;		/* CCK Power Offset for Rate 11M */
-	u32 ofdmgpo;		/* G-PHY OFDM Power Offset */
-};
-
 struct ssb_sprom {
 	u8 revision;
-	u8 temp_fill[2 * sizeof(struct ssb_sprom_r1)];
 	u8 il0mac[6];		/* MAC address for 802.11b/g */
 	u8 et0mac[6];		/* MAC address for Ethernet */
 	u8 et1mac[6];		/* MAC address for 802.11a */
@@ -106,16 +42,6 @@ struct ssb_sprom {
 	u8 antenna_gain_bg;	/* B/G-PHY Antenna gain (in dBm Q5.2) */
 
 	/* TODO - add any parameters needed from rev 2, 3, or 4 SPROMs */
-	/* The valid r# fields are selected by the "revision".
-	 * Revision 3 and lower inherit from lower revisions.
-	 */
-	union {
-		struct {
-			struct ssb_sprom_r1 r1;
-			struct ssb_sprom_r2 r2;
-			struct ssb_sprom_r3 r3;
-		};
-	};
 };
 
 /* Information about the PCB the circuitry is soldered on. */
diff --git a/include/linux/ssb/ssb_regs.h b/include/linux/ssb/ssb_regs.h
index 96bba69b1271..30222e89ad16 100644
--- a/include/linux/ssb/ssb_regs.h
+++ b/include/linux/ssb/ssb_regs.h
@@ -175,6 +175,7 @@
 #define  SSB_SPROM_REVISION_REV		0x00FF	/* SPROM Revision number */
 #define  SSB_SPROM_REVISION_CRC		0xFF00	/* SPROM CRC8 value */
 #define  SSB_SPROM_REVISION_CRC_SHIFT	8
+
 /* SPROM Revision 1 */
 #define SSB_SPROM1_SPID			0x1004	/* Subsystem Product ID for PCI */
 #define SSB_SPROM1_SVID			0x1006	/* Subsystem Vendor ID for PCI */
@@ -223,7 +224,7 @@
 #define  SSB_SPROM1_AGAIN_A		0x00FF	/* A-PHY */
 #define  SSB_SPROM1_AGAIN_BG		0xFF00	/* B-PHY and G-PHY */
 #define  SSB_SPROM1_AGAIN_BG_SHIFT	8
-#define SSB_SPROM1_OEM			0x1076	/* 8 bytes OEM string (rev 1 only) */
+
 /* SPROM Revision 2 (inherits from rev 1) */
 #define SSB_SPROM2_BFLHI		0x1038	/* Boardflags (high 16 bits) */
 #define SSB_SPROM2_MAXP_A		0x103A	/* A-PHY Max Power */
@@ -240,6 +241,7 @@
 #define  SSB_SPROM2_OPO_VALUE		0x00FF
 #define  SSB_SPROM2_OPO_UNUSED		0xFF00
 #define SSB_SPROM2_CCODE		0x107C	/* Two char Country Code */
+
 /* SPROM Revision 3 (inherits most data from rev 2) */
 #define SSB_SPROM3_IL0MAC		0x104A	/* 6 bytes MAC address for 802.11b/g */
 #define SSB_SPROM3_ET0MAC		0x1050	/* 6 bytes MAC address for Ethernet ?? */
@@ -261,11 +263,12 @@
 #define  SSB_SPROM3_CCKPO_11M		0xF000	/* 11M Rate PO */
 #define  SSB_SPROM3_CCKPO_11M_SHIFT	12
 #define  SSB_SPROM3_OFDMGPO		0x107A	/* G-PHY OFDM Power Offset (4 bytes, BigEndian) */
+
 /* SPROM Revision 4 		entries with ?? in comment are unknown */
-#define SSB_SPROM4_IL0MAC		0x104C	/* 6 byte MAC address for b/g */
+#define SSB_SPROM4_IL0MAC		0x104C	/* 6 byte MAC address for a/b/g/n */
 #define SSB_SPROM4_ET0MAC		0x1018	/* 6 bytes MAC address for Ethernet ?? */
 #define SSB_SPROM4_ET1MAC		0x1018	/* 6 bytes MAC address for 802.11a ?? */
-#define SSB_SPROM4_ETHPHY		0x105A	/* Ethernet PHY settings */
+#define SSB_SPROM4_ETHPHY		0x105A	/* Ethernet PHY settings ?? */
 #define  SSB_SPROM4_ETHPHY_ET0A		0x001F	/* MII Address for enet0 */
 #define  SSB_SPROM4_ETHPHY_ET1A		0x03E0	/* MII Address for enet1 */
 #define  SSB_SPROM4_ETHPHY_ET1A_SHIFT	5
@@ -280,24 +283,28 @@
 #define  SSB_SPROM4_AGAIN_1		0xFF00	/* Antenna 1 */
 #define  SSB_SPROM4_AGAIN_1_SHIFT	8
 #define SSB_SPROM4_BFLHI		0x1046  /* Board Flags Hi */
-#define SSB_SPROM4_MAXP_A		0x1000  /* Max Power A ?? */
-#define SSB_SPROM4_MAXP_A_HI		0x00FF  /* Mask for Hi */
-#define SSB_SPROM4_MAXP_A_LO		0xFF00  /* Mask for Lo */
-#define SSB_SPROM4_MAXP_A_LO_SHIFT	16	/* Shift for Lo */
-#define SSB_SPROM4_PA1LOB0		0x1000  /* ?? */
-#define SSB_SPROM4_PA1LOB1		0x1000  /* ?? */
-#define SSB_SPROM4_PA1LOB2		0x1000  /* ?? */
-#define SSB_SPROM4_PA1HIB0		0x1000	/* ?? */
-#define SSB_SPROM4_PA1HIB1		0x1000	/* ?? */
-#define SSB_SPROM4_PA1HIB2		0x1000	/* ?? */
-#define SSB_SPROM4_OPO			0x1000	/* ?? */
-#define SSB_SPROM4_OPO_VALUE		0x0000	/* ?? */
-#define SSB_SPROM4_GPIOLDC		0x105A	/* LED Powersave Duty Cycle */
-#define  SSB_SPROM4_GPIOLDC_OFF		0x0000FF00	/* Off Count */
-#define  SSB_SPROM4_GPIOLDC_OFF_SHIFT	8
-#define  SSB_SPROM4_GPIOLDC_ON		0x00FF0000	/* On Count */
-#define  SSB_SPROM4_GPIOLDC_ON_SHIFT	16
-
+#define SSB_SPROM4_MAXP_BG		0x1080  /* Max Power BG in path 1 */
+#define  SSB_SPROM4_MAXP_BG_MASK	0x00FF  /* Mask for Max Power BG */
+#define  SSB_SPROM4_ITSSI_BG		0xFF00	/* Mask for path 1 itssi_bg */
+#define  SSB_SPROM4_ITSSI_BG_SHIFT	8
+#define SSB_SPROM4_MAXP_A		0x108A  /* Max Power A in path 1 */
+#define  SSB_SPROM4_MAXP_A_MASK		0x00FF  /* Mask for Max Power A */
+#define  SSB_SPROM4_ITSSI_A		0xFF00	/* Mask for path 1 itssi_a */
+#define  SSB_SPROM4_ITSSI_A_SHIFT	8
+#define SSB_SPROM4_GPIOA		0x1056	/* Gen. Purpose IO # 0 and 1 */
+#define  SSB_SPROM4_GPIOA_P0		0x00FF	/* Pin 0 */
+#define  SSB_SPROM4_GPIOA_P1		0xFF00	/* Pin 1 */
+#define  SSB_SPROM4_GPIOA_P1_SHIFT	8
+#define SSB_SPROM4_GPIOB		0x1058	/* Gen. Purpose IO # 2 and 3 */
+#define  SSB_SPROM4_GPIOB_P2		0x00FF	/* Pin 2 */
+#define  SSB_SPROM4_GPIOB_P3		0xFF00	/* Pin 3 */
+#define  SSB_SPROM4_GPIOB_P3_SHIFT	8
+#define SSB_SPROM4_PA0B0		0x1082	/* The paXbY locations are */
+#define SSB_SPROM4_PA0B1		0x1084	/*   only guesses */
+#define SSB_SPROM4_PA0B2		0x1086
+#define SSB_SPROM4_PA1B0		0x108E
+#define SSB_SPROM4_PA1B1		0x1090
+#define SSB_SPROM4_PA1B2		0x1092
 
 /* Values for SSB_SPROM1_BINF_CCODE */
 enum {
-- 
cgit v1.2.3


From a3edb08311fc559652ffc959e93eb5be9294443f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 22 Dec 2007 17:52:42 +0000
Subject: annotate tun

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/tun.c      | 2 +-
 include/linux/if_tun.h | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 5db4df460049..46339f6bcd00 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -506,7 +506,7 @@ static int tun_set_iff(struct file *file, struct ifreq *ifr)
 		/* Be promiscuous by default to maintain previous behaviour. */
 		tun->if_flags = IFF_PROMISC;
 		/* Generate random Ethernet address. */
-		*(u16 *)tun->dev_addr = htons(0x00FF);
+		*(__be16 *)tun->dev_addr = htons(0x00FF);
 		get_random_bytes(tun->dev_addr + sizeof(u16), 4);
 		memset(tun->chr_filter, 0, sizeof tun->chr_filter);
 
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 33e489d5bb33..72f1c5f47be3 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -21,6 +21,8 @@
 /* Uncomment to enable debugging */
 /* #define TUN_DEBUG 1 */
 
+#include <linux/types.h>
+
 #ifdef __KERNEL__
 
 #ifdef TUN_DEBUG
@@ -88,7 +90,7 @@ struct tun_struct {
 
 struct tun_pi {
 	unsigned short flags;
-	unsigned short proto;
+	__be16 proto;
 };
 #define TUN_PKT_STRIP	0x0001
 
-- 
cgit v1.2.3


From 904584018e9ba30a3e562d86ee7dfb6239105664 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 22 Dec 2007 17:52:52 +0000
Subject: annotate the rest of drivers/net/wan

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/wan/hdlc_ppp.c      | 2 +-
 drivers/net/wan/hdlc_raw_eth.c  | 2 +-
 drivers/net/wan/lmc/lmc_proto.c | 2 +-
 drivers/net/wan/lmc/lmc_proto.h | 2 +-
 drivers/net/wan/sbni.c          | 4 ++--
 drivers/net/wan/wanxl.c         | 2 +-
 include/linux/if_frad.h         | 2 +-
 7 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c
index 3caeb528eace..519e1550e2e7 100644
--- a/drivers/net/wan/hdlc_ppp.c
+++ b/drivers/net/wan/hdlc_ppp.c
@@ -42,7 +42,7 @@ static inline struct ppp_state* state(hdlc_device *hdlc)
 static int ppp_open(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	void *old_ioctl;
+	int (*old_ioctl)(struct net_device *, struct ifreq *, int);
 	int result;
 
 	dev->priv = &state(hdlc)->syncppp_ptr;
diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c
index 1a69a9aaa9b9..8895394e6006 100644
--- a/drivers/net/wan/hdlc_raw_eth.c
+++ b/drivers/net/wan/hdlc_raw_eth.c
@@ -59,7 +59,7 @@ static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 	raw_hdlc_proto new_settings;
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	int result;
-	void *old_ch_mtu;
+	int (*old_ch_mtu)(struct net_device *, int);
 	int old_qlen;
 
 	switch (ifr->ifr_settings.type) {
diff --git a/drivers/net/wan/lmc/lmc_proto.c b/drivers/net/wan/lmc/lmc_proto.c
index 426c0678d983..85315758198d 100644
--- a/drivers/net/wan/lmc/lmc_proto.c
+++ b/drivers/net/wan/lmc/lmc_proto.c
@@ -208,7 +208,7 @@ void lmc_proto_close(lmc_softc_t *sc) /*FOLD00*/
     lmc_trace(sc->lmc_device, "lmc_proto_close out");
 }
 
-unsigned short lmc_proto_type(lmc_softc_t *sc, struct sk_buff *skb) /*FOLD00*/
+__be16 lmc_proto_type(lmc_softc_t *sc, struct sk_buff *skb) /*FOLD00*/
 {
     lmc_trace(sc->lmc_device, "lmc_proto_type in");
     switch(sc->if_type){
diff --git a/drivers/net/wan/lmc/lmc_proto.h b/drivers/net/wan/lmc/lmc_proto.h
index 080a55773349..ccaa69e8b3c7 100644
--- a/drivers/net/wan/lmc/lmc_proto.h
+++ b/drivers/net/wan/lmc/lmc_proto.h
@@ -8,7 +8,7 @@ void lmc_proto_reopen(lmc_softc_t *sc);
 int lmc_proto_ioctl(lmc_softc_t *sc, struct ifreq *ifr, int cmd);
 void lmc_proto_open(lmc_softc_t *sc);
 void lmc_proto_close(lmc_softc_t *sc);
-unsigned short lmc_proto_type(lmc_softc_t *sc, struct sk_buff *skb);
+__be16 lmc_proto_type(lmc_softc_t *sc, struct sk_buff *skb);
 void lmc_proto_netif(lmc_softc_t *sc, struct sk_buff *skb);
 int lmc_skb_rawpackets(char *buf, char **start, off_t offset, int len, int unused);
 
diff --git a/drivers/net/wan/sbni.c b/drivers/net/wan/sbni.c
index 2e8b5c2de887..15d5c58e57bc 100644
--- a/drivers/net/wan/sbni.c
+++ b/drivers/net/wan/sbni.c
@@ -391,8 +391,8 @@ sbni_probe1( struct net_device  *dev,  unsigned long  ioaddr,  int  irq )
 	spin_lock_init( &nl->lock );
 
 	/* store MAC address (generate if that isn't known) */
-	*(u16 *)dev->dev_addr = htons( 0x00ff );
-	*(u32 *)(dev->dev_addr + 2) = htonl( 0x01000000 |
+	*(__be16 *)dev->dev_addr = htons( 0x00ff );
+	*(__be32 *)(dev->dev_addr + 2) = htonl( 0x01000000 |
 		( (mac[num]  ?  mac[num]  :  (u32)((long)dev->priv)) & 0x00ffffff) );
 
 	/* store link settings (speed, receive level ) */
diff --git a/drivers/net/wan/wanxl.c b/drivers/net/wan/wanxl.c
index ad8c8651d298..d4aab8a28b61 100644
--- a/drivers/net/wan/wanxl.c
+++ b/drivers/net/wan/wanxl.c
@@ -715,7 +715,7 @@ static int __devinit wanxl_pci_init_one(struct pci_dev *pdev,
 	}
 
 	for (i = 0; i < sizeof(firmware); i += 4)
-		writel(htonl(*(u32*)(firmware + i)), mem + PDM_OFFSET + i);
+		writel(ntohl(*(__be32*)(firmware + i)), mem + PDM_OFFSET + i);
 
 	for (i = 0; i < ports; i++)
 		writel(card->status_address +
diff --git a/include/linux/if_frad.h b/include/linux/if_frad.h
index f272a80caa3e..5c34240de746 100644
--- a/include/linux/if_frad.h
+++ b/include/linux/if_frad.h
@@ -137,7 +137,7 @@ struct frhdr
 
    unsigned char  NLPID;
    unsigned char  OUI[3];
-   unsigned short PID;
+   __be16 PID;
 
 #define IP_NLPID pad 
 } __attribute__((packed));
-- 
cgit v1.2.3


From b7c6ba6eb1234e35a74fb8ba8123232a7b1ba9e4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 28 Jan 2008 14:41:19 -0800
Subject: [NETNS]: Consolidate kernel netlink socket destruction.

Create a specific helper for netlink kernel socket disposal. This just
let the code look better and provides a ground for proper disposal
inside a namespace.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Tested-by: Alexey Dobriyan <adobriyan@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/connector/connector.c       |  9 +++------
 drivers/scsi/scsi_netlink.c         |  2 +-
 drivers/scsi/scsi_transport_iscsi.c |  4 ++--
 fs/ecryptfs/netlink.c               |  3 +--
 include/linux/netlink.h             |  1 +
 net/bridge/netfilter/ebt_ulog.c     |  4 ++--
 net/core/rtnetlink.c                |  2 +-
 net/decnet/netfilter/dn_rtmsg.c     |  4 ++--
 net/ipv4/fib_frontend.c             |  2 +-
 net/ipv4/inet_diag.c                |  2 +-
 net/ipv4/netfilter/ip_queue.c       |  4 ++--
 net/ipv4/netfilter/ipt_ULOG.c       |  4 ++--
 net/ipv6/netfilter/ip6_queue.c      |  4 ++--
 net/netfilter/nfnetlink.c           |  2 +-
 net/netlink/af_netlink.c            | 11 +++++++++++
 net/xfrm/xfrm_user.c                |  2 +-
 16 files changed, 34 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c
index 37976dcf044b..fea2d3ed9cbd 100644
--- a/drivers/connector/connector.c
+++ b/drivers/connector/connector.c
@@ -420,8 +420,7 @@ static int __devinit cn_init(void)
 
 	dev->cbdev = cn_queue_alloc_dev("cqueue", dev->nls);
 	if (!dev->cbdev) {
-		if (dev->nls->sk_socket)
-			sock_release(dev->nls->sk_socket);
+		netlink_kernel_release(dev->nls);
 		return -EINVAL;
 	}
 	
@@ -431,8 +430,7 @@ static int __devinit cn_init(void)
 	if (err) {
 		cn_already_initialized = 0;
 		cn_queue_free_dev(dev->cbdev);
-		if (dev->nls->sk_socket)
-			sock_release(dev->nls->sk_socket);
+		netlink_kernel_release(dev->nls);
 		return -EINVAL;
 	}
 
@@ -447,8 +445,7 @@ static void __devexit cn_fini(void)
 
 	cn_del_callback(&dev->id);
 	cn_queue_free_dev(dev->cbdev);
-	if (dev->nls->sk_socket)
-		sock_release(dev->nls->sk_socket);
+	netlink_kernel_release(dev->nls);
 }
 
 subsys_initcall(cn_init);
diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c
index 3e1591828171..370c78cc1cb5 100644
--- a/drivers/scsi/scsi_netlink.c
+++ b/drivers/scsi/scsi_netlink.c
@@ -164,7 +164,7 @@ void
 scsi_netlink_exit(void)
 {
 	if (scsi_nl_sock) {
-		sock_release(scsi_nl_sock->sk_socket);
+		netlink_kernel_release(scsi_nl_sock);
 		netlink_unregister_notifier(&scsi_netlink_notifier);
 	}
 
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index ef0e74264880..0d7b4e79415c 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -1558,7 +1558,7 @@ static __init int iscsi_transport_init(void)
 	return 0;
 
 release_nls:
-	sock_release(nls->sk_socket);
+	netlink_kernel_release(nls);
 unregister_session_class:
 	transport_class_unregister(&iscsi_session_class);
 unregister_conn_class:
@@ -1573,7 +1573,7 @@ unregister_transport_class:
 static void __exit iscsi_transport_exit(void)
 {
 	destroy_workqueue(iscsi_eh_timer_workq);
-	sock_release(nls->sk_socket);
+	netlink_kernel_release(nls);
 	transport_class_unregister(&iscsi_connection_class);
 	transport_class_unregister(&iscsi_session_class);
 	transport_class_unregister(&iscsi_host_class);
diff --git a/fs/ecryptfs/netlink.c b/fs/ecryptfs/netlink.c
index 9aa345121e09..f638a698dc52 100644
--- a/fs/ecryptfs/netlink.c
+++ b/fs/ecryptfs/netlink.c
@@ -237,7 +237,6 @@ out:
  */
 void ecryptfs_release_netlink(void)
 {
-	if (ecryptfs_nl_sock && ecryptfs_nl_sock->sk_socket)
-		sock_release(ecryptfs_nl_sock->sk_socket);
+	netlink_kernel_release(ecryptfs_nl_sock);
 	ecryptfs_nl_sock = NULL;
 }
diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 2aee0f510876..bd13b6f4a98e 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -178,6 +178,7 @@ extern struct sock *netlink_kernel_create(struct net *net,
 					  void (*input)(struct sk_buff *skb),
 					  struct mutex *cb_mutex,
 					  struct module *module);
+extern void netlink_kernel_release(struct sock *sk);
 extern int netlink_change_ngroups(struct sock *sk, unsigned int groups);
 extern void netlink_clear_multicast_users(struct sock *sk, unsigned int group);
 extern void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err);
diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c
index b73ba28bcbe8..8e7b00b68d38 100644
--- a/net/bridge/netfilter/ebt_ulog.c
+++ b/net/bridge/netfilter/ebt_ulog.c
@@ -307,7 +307,7 @@ static int __init ebt_ulog_init(void)
 	if (!ebtulognl)
 		ret = -ENOMEM;
 	else if ((ret = ebt_register_watcher(&ulog)))
-		sock_release(ebtulognl->sk_socket);
+		netlink_kernel_release(ebtulognl);
 
 	if (ret == 0)
 		nf_log_register(PF_BRIDGE, &ebt_ulog_logger);
@@ -333,7 +333,7 @@ static void __exit ebt_ulog_fini(void)
 		}
 		spin_unlock_bh(&ub->lock);
 	}
-	sock_release(ebtulognl->sk_socket);
+	netlink_kernel_release(ebtulognl);
 }
 
 module_init(ebt_ulog_init);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index a5f4f661fa62..02cf848f71d2 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1384,7 +1384,7 @@ static void rtnetlink_net_exit(struct net *net)
 		 * free.
 		 */
 		sk->sk_net = get_net(&init_net);
-		sock_release(net->rtnl->sk_socket);
+		netlink_kernel_release(net->rtnl);
 		net->rtnl = NULL;
 	}
 }
diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c
index 96375f2e64ff..6d2bd3202048 100644
--- a/net/decnet/netfilter/dn_rtmsg.c
+++ b/net/decnet/netfilter/dn_rtmsg.c
@@ -137,7 +137,7 @@ static int __init dn_rtmsg_init(void)
 
 	rv = nf_register_hook(&dnrmg_ops);
 	if (rv) {
-		sock_release(dnrmg->sk_socket);
+		netlink_kernel_release(dnrmg);
 	}
 
 	return rv;
@@ -146,7 +146,7 @@ static int __init dn_rtmsg_init(void)
 static void __exit dn_rtmsg_fini(void)
 {
 	nf_unregister_hook(&dnrmg_ops);
-	sock_release(dnrmg->sk_socket);
+	netlink_kernel_release(dnrmg);
 }
 
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 4e5216e9aacb..e787d2151152 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -881,7 +881,7 @@ static void nl_fib_lookup_exit(struct net *net)
 	 * initial network namespace. So the socket will  be safe to free.
 	 */
 	net->ipv4.fibnl->sk_net = get_net(&init_net);
-	sock_release(net->ipv4.fibnl->sk_socket);
+	netlink_kernel_release(net->ipv4.fibnl);
 }
 
 static void fib_disable_ip(struct net_device *dev, int force)
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index e468e7a7aac4..605ed2cd7972 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -935,7 +935,7 @@ out_free_table:
 
 static void __exit inet_diag_exit(void)
 {
-	sock_release(idiagnl->sk_socket);
+	netlink_kernel_release(idiagnl);
 	kfree(inet_diag_table);
 }
 
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 7361315f20c6..5109839da222 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -605,7 +605,7 @@ cleanup_sysctl:
 	unregister_netdevice_notifier(&ipq_dev_notifier);
 	proc_net_remove(&init_net, IPQ_PROC_FS_NAME);
 cleanup_ipqnl:
-	sock_release(ipqnl->sk_socket);
+	netlink_kernel_release(ipqnl);
 	mutex_lock(&ipqnl_mutex);
 	mutex_unlock(&ipqnl_mutex);
 
@@ -624,7 +624,7 @@ static void __exit ip_queue_fini(void)
 	unregister_netdevice_notifier(&ipq_dev_notifier);
 	proc_net_remove(&init_net, IPQ_PROC_FS_NAME);
 
-	sock_release(ipqnl->sk_socket);
+	netlink_kernel_release(ipqnl);
 	mutex_lock(&ipqnl_mutex);
 	mutex_unlock(&ipqnl_mutex);
 
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index fa24efaf3eaa..b192756c6d0d 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -415,7 +415,7 @@ static int __init ulog_tg_init(void)
 
 	ret = xt_register_target(&ulog_tg_reg);
 	if (ret < 0) {
-		sock_release(nflognl->sk_socket);
+		netlink_kernel_release(nflognl);
 		return ret;
 	}
 	if (nflog)
@@ -434,7 +434,7 @@ static void __exit ulog_tg_exit(void)
 	if (nflog)
 		nf_log_unregister(&ipt_ulog_logger);
 	xt_unregister_target(&ulog_tg_reg);
-	sock_release(nflognl->sk_socket);
+	netlink_kernel_release(nflognl);
 
 	/* remove pending timers and free allocated skb's */
 	for (i = 0; i < ULOG_MAXNLGROUPS; i++) {
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index a20db0bb5a1f..56b4ea6d29ed 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -609,7 +609,7 @@ cleanup_sysctl:
 	proc_net_remove(&init_net, IPQ_PROC_FS_NAME);
 
 cleanup_ipqnl:
-	sock_release(ipqnl->sk_socket);
+	netlink_kernel_release(ipqnl);
 	mutex_lock(&ipqnl_mutex);
 	mutex_unlock(&ipqnl_mutex);
 
@@ -628,7 +628,7 @@ static void __exit ip6_queue_fini(void)
 	unregister_netdevice_notifier(&ipq_dev_notifier);
 	proc_net_remove(&init_net, IPQ_PROC_FS_NAME);
 
-	sock_release(ipqnl->sk_socket);
+	netlink_kernel_release(ipqnl);
 	mutex_lock(&ipqnl_mutex);
 	mutex_unlock(&ipqnl_mutex);
 
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index 2128542995f7..b75c9c4a995d 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -179,7 +179,7 @@ static void nfnetlink_rcv(struct sk_buff *skb)
 static void __exit nfnetlink_exit(void)
 {
 	printk("Removing netfilter NETLINK layer.\n");
-	sock_release(nfnl->sk_socket);
+	netlink_kernel_release(nfnl);
 	return;
 }
 
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 29fef558aab6..626a58206298 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1405,6 +1405,17 @@ out_sock_release:
 }
 EXPORT_SYMBOL(netlink_kernel_create);
 
+
+void
+netlink_kernel_release(struct sock *sk)
+{
+	if (sk == NULL || sk->sk_socket == NULL)
+		return;
+	sock_release(sk->sk_socket);
+}
+EXPORT_SYMBOL(netlink_kernel_release);
+
+
 /**
  * netlink_change_ngroups - change number of multicast groups
  *
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 35fc16ae50ac..e0ccdf267813 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -2420,7 +2420,7 @@ static void __exit xfrm_user_exit(void)
 	xfrm_unregister_km(&netlink_mgr);
 	rcu_assign_pointer(xfrm_nl, NULL);
 	synchronize_rcu();
-	sock_release(nlsk->sk_socket);
+	netlink_kernel_release(nlsk);
 }
 
 module_init(xfrm_user_init);
-- 
cgit v1.2.3


From e9888f5498083f5e4d873cbbe16aa97d89aa1342 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sat, 19 Jan 2008 00:00:42 -0800
Subject: [IrDA]: Irport removal - part 1

This patch removes IrPORT and the old dongle drivers (all off them
have replacement drivers).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/irda/Kconfig       |  141 -----
 drivers/net/irda/Makefile      |   17 +-
 drivers/net/irda/irport.c      | 1123 ----------------------------------------
 drivers/net/irda/irport.h      |   80 ---
 include/net/irda/irda_device.h |   13 -
 net/irda/irda_device.c         |  169 +-----
 6 files changed, 3 insertions(+), 1540 deletions(-)
 delete mode 100644 drivers/net/irda/irport.c
 delete mode 100644 drivers/net/irda/irport.h

(limited to 'include')

diff --git a/drivers/net/irda/Kconfig b/drivers/net/irda/Kconfig
index 65806956728a..ce816ba9c40d 100644
--- a/drivers/net/irda/Kconfig
+++ b/drivers/net/irda/Kconfig
@@ -190,147 +190,6 @@ config KS959_DONGLE
 	  To compile it as a module, choose M here: the module will be called
 	  ks959-sir.
 
-comment "Old SIR device drivers"
-
-config IRPORT_SIR
-	tristate "IrPORT (IrDA serial driver)"
-	depends on IRDA && BROKEN_ON_SMP
-	---help---
-	  Say Y here if you want to build support for the IrPORT IrDA device
-	  driver.  To compile it as a module, choose M here: the module will be
-	  called irport. IrPORT can be used instead of IrTTY and sometimes
-	  this can be better.  One example is if your IrDA port does not
-	  have echo-canceling, which will work OK with IrPORT since this
-	  driver is working in half-duplex mode only.  You don't need to use
-	  irattach with IrPORT, but you just insert it the same way as FIR
-	  drivers (insmod irport io=0x3e8 irq=11).  Notice that IrPORT is a
-	  SIR device driver which means that speed is limited to 115200 bps.
-
-	  If unsure, say Y.
-
-comment "Old Serial dongle support"
-
-config DONGLE_OLD
-	bool "Old Serial dongle support"
-	depends on IRPORT_SIR && BROKEN_ON_SMP
-	help
-	  Say Y here if you have an infrared device that connects to your
-	  computer's serial port. These devices are called dongles. Then say Y
-	  or M to the driver for your particular dongle below.
-
-	  Note that the answer to this question won't directly affect the
-	  kernel: saying N will just cause the configurator to skip all
-	  the questions about serial dongles.
-
-config ESI_DONGLE_OLD
-	tristate "ESI JetEye PC dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Extended Systems
-	  JetEye PC dongle.  To compile it as a module, choose M here.  The ESI
-	  dongle attaches to the normal 9-pin serial port connector, and can
-	  currently only be used by IrTTY.  To activate support for ESI
-	  dongles you will have to start irattach like this:
-	  "irattach -d esi".
-
-config ACTISYS_DONGLE_OLD
-	tristate "ACTiSYS IR-220L and IR220L+ dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the ACTiSYS IR-220L and
-	  IR220L+ dongles.  To compile it as a module, choose M here.  The
-	  ACTiSYS dongles attaches to the normal 9-pin serial port connector,
-	  and can currently only be used by IrTTY.  To activate support for
-	  ACTiSYS dongles you will have to start irattach like this:
-	  "irattach -d actisys" or "irattach -d actisys+".
-
-config TEKRAM_DONGLE_OLD
-	tristate "Tekram IrMate 210B dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Tekram IrMate 210B
-	  dongle.  To compile it as a module, choose M here.  The Tekram dongle
-	  attaches to the normal 9-pin serial port connector, and can
-	  currently only be used by IrTTY.  To activate support for Tekram
-	  dongles you will have to start irattach like this:
-	  "irattach -d tekram".
-
-config GIRBIL_DONGLE_OLD
-	tristate "Greenwich GIrBIL dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Greenwich GIrBIL
-	  dongle.  To compile it as a module, choose M here.  The Greenwich
-	  dongle attaches to the normal 9-pin serial port connector, and can
-	  currently only be used by IrTTY.  To activate support for Greenwich
-	  dongles you will have to insert "irattach -d girbil" in the
-	  /etc/irda/drivers script.
-
-config LITELINK_DONGLE_OLD
-	tristate "Parallax LiteLink dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Parallax Litelink
-	  dongle.  To compile it as a module, choose M here.  The Parallax
-	  dongle attaches to the normal 9-pin serial port connector, and can
-	  currently only be used by IrTTY.  To activate support for Parallax
-	  dongles you will have to start irattach like this:
-	  "irattach -d litelink".
-
-config MCP2120_DONGLE_OLD
-	tristate "Microchip MCP2120"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Microchip MCP2120
-	  dongle.  To compile it as a module, choose M here.  The MCP2120 dongle
-	  attaches to the normal 9-pin serial port connector, and can
-	  currently only be used by IrTTY.  To activate support for MCP2120
-	  dongles you will have to insert "irattach -d mcp2120" in the
-	  /etc/irda/drivers script.
-
-	  You must build this dongle yourself.  For more information see:
-	  <http://www.eyetap.org/~tangf/irda_sir_linux.html>
-
-config OLD_BELKIN_DONGLE_OLD
-	tristate "Old Belkin dongle"
-	depends on DONGLE_OLD && IRDA
-	help
-	  Say Y here if you want to build support for the Adaptec Airport 1000
-	  and 2000 dongles.  To compile it as a module, choose M here: the module
-	  will be called old_belkin.  Some information is contained in the
-	  comments at the top of <file:drivers/net/irda/old_belkin.c>.
-
-config ACT200L_DONGLE_OLD
-	tristate "ACTiSYS IR-200L dongle (EXPERIMENTAL)"
-	depends on DONGLE_OLD && EXPERIMENTAL && IRDA
-	help
-	  Say Y here if you want to build support for the ACTiSYS IR-200L
-	  dongle.  To compile it as a module, choose M here.  The ACTiSYS
-	  IR-200L dongle attaches to the normal 9-pin serial port connector,
-	  and can currently only be used by IrTTY. To activate support for
-	  ACTiSYS IR-200L dongles you will have to start irattach like this:
-	  "irattach -d act200l".
-
-config MA600_DONGLE_OLD
-	tristate "Mobile Action MA600 dongle (EXPERIMENTAL)"
-	depends on DONGLE_OLD && EXPERIMENTAL && IRDA
-	---help---
-	  Say Y here if you want to build support for the Mobile Action MA600
-	  dongle.  To compile it as a module, choose M here. The MA600 dongle
-	  attaches to the normal 9-pin serial port connector, and can
-	  currently only be tested on IrCOMM.  To activate support for MA600
-	  dongles you will have to insert "irattach -d ma600" in the
-	  /etc/irda/drivers script.  Note: irutils 0.9.15 requires no
-	  modification. irutils 0.9.9 needs modification. For more
-	  information, download the following tar gzip file.
-
-	  There is a pre-compiled module on
-	  <http://engsvr.ust.hk/~eetwl95/ma600.html>
-
-config EP7211_IR
-	tristate "EP7211 I/R support"
-	depends on DONGLE_OLD && ARCH_EP7211 && IRDA
-
 comment "FIR device drivers"
 
 config USB_IRDA
diff --git a/drivers/net/irda/Makefile b/drivers/net/irda/Makefile
index fefbb5909081..5d20fde32a24 100644
--- a/drivers/net/irda/Makefile
+++ b/drivers/net/irda/Makefile
@@ -5,8 +5,6 @@
 # Rewritten to use lists instead of if-statements.
 #
 
-# Old SIR drivers 
-obj-$(CONFIG_IRPORT_SIR)	+= 		irport.o
 # FIR drivers
 obj-$(CONFIG_USB_IRDA)		+= irda-usb.o
 obj-$(CONFIG_SIGMATEL_FIR)	+= stir4200.o
@@ -20,21 +18,10 @@ obj-$(CONFIG_VLSI_FIR)		+= vlsi_ir.o
 obj-$(CONFIG_VIA_FIR)		+= via-ircc.o
 obj-$(CONFIG_PXA_FICP)	        += pxaficp_ir.o
 obj-$(CONFIG_MCS_FIR)	        += mcs7780.o
-# Old dongle drivers for old SIR drivers
-obj-$(CONFIG_ESI_DONGLE_OLD)		+= esi.o
-obj-$(CONFIG_TEKRAM_DONGLE_OLD)	+= tekram.o
-obj-$(CONFIG_ACTISYS_DONGLE_OLD)	+= actisys.o
-obj-$(CONFIG_GIRBIL_DONGLE_OLD)	+= girbil.o
-obj-$(CONFIG_LITELINK_DONGLE_OLD)	+= litelink.o
-obj-$(CONFIG_OLD_BELKIN_DONGLE_OLD)	+= old_belkin.o
-obj-$(CONFIG_MCP2120_DONGLE_OLD)	+= mcp2120.o
-obj-$(CONFIG_ACT200L_DONGLE_OLD)	+= act200l.o
-obj-$(CONFIG_MA600_DONGLE_OLD)	+= ma600.o
-obj-$(CONFIG_EP7211_IR)		+= ep7211_ir.o
 obj-$(CONFIG_AU1000_FIR)	+= au1k_ir.o
-# New SIR drivers
+# SIR drivers
 obj-$(CONFIG_IRTTY_SIR)		+= irtty-sir.o	sir-dev.o
-# New dongles drivers for new SIR drivers
+# dongle drivers for SIR drivers
 obj-$(CONFIG_ESI_DONGLE)	+= esi-sir.o
 obj-$(CONFIG_TEKRAM_DONGLE)	+= tekram-sir.o
 obj-$(CONFIG_ACTISYS_DONGLE)	+= actisys-sir.o
diff --git a/drivers/net/irda/irport.c b/drivers/net/irda/irport.c
deleted file mode 100644
index c79caa5d3d71..000000000000
--- a/drivers/net/irda/irport.c
+++ /dev/null
@@ -1,1123 +0,0 @@
-/*********************************************************************
- * 
- * Filename:	  irport.c
- * Version:	  1.0
- * Description:   Half duplex serial port SIR driver for IrDA. 
- * Status:	  Experimental.
- * Author:	  Dag Brattli <dagb@cs.uit.no>
- * Created at:	  Sun Aug  3 13:49:59 1997
- * Modified at:   Fri Jan 28 20:22:38 2000
- * Modified by:   Dag Brattli <dagb@cs.uit.no>
- * Sources:	  serial.c by Linus Torvalds 
- * 
- *     Copyright (c) 1997, 1998, 1999-2000 Dag Brattli, All Rights Reserved.
- *     Copyright (c) 2000-2003 Jean Tourrilhes, All Rights Reserved.
- *     
- *     This program is free software; you can redistribute it and/or 
- *     modify it under the terms of the GNU General Public License as 
- *     published by the Free Software Foundation; either version 2 of 
- *     the License, or (at your option) any later version.
- * 
- *     This program is distributed in the hope that it will be useful,
- *     but WITHOUT ANY WARRANTY; without even the implied warranty of
- *     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- *     GNU General Public License for more details.
- * 
- *     You should have received a copy of the GNU General Public License 
- *     along with this program; if not, write to the Free Software 
- *     Foundation, Inc., 59 Temple Place, Suite 330, Boston, 
- *     MA 02111-1307 USA
- *
- *     This driver is ment to be a small half duplex serial driver to be
- *     used for IR-chipsets that has a UART (16550) compatibility mode. 
- *     Eventually it will replace irtty, because of irtty has some 
- *     problems that is hard to get around when we don't have control
- *     over the serial driver. This driver may also be used by FIR 
- *     drivers to handle SIR mode for them.
- *
- ********************************************************************/
-
-#include <linux/module.h>
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/ioport.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/skbuff.h>
-#include <linux/serial_reg.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/delay.h>
-#include <linux/rtnetlink.h>
-#include <linux/bitops.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-
-#include <net/irda/irda.h>
-#include <net/irda/wrapper.h>
-#include "irport.h"
-
-#define IO_EXTENT 8
-
-/* 
- * Currently you'll need to set these values using insmod like this:
- * insmod irport io=0x3e8 irq=11
- */
-static unsigned int io[]  = { ~0, ~0, ~0, ~0 };
-static unsigned int irq[] = { 0, 0, 0, 0 };
-
-static unsigned int qos_mtt_bits = 0x03;
-
-static struct irport_cb *dev_self[] = { NULL, NULL, NULL, NULL};
-static char *driver_name = "irport";
-
-static inline void irport_write_wakeup(struct irport_cb *self);
-static inline int  irport_write(int iobase, int fifo_size, __u8 *buf, int len);
-static inline void irport_receive(struct irport_cb *self);
-
-static int  irport_net_ioctl(struct net_device *dev, struct ifreq *rq, 
-			     int cmd);
-static inline int  irport_is_receiving(struct irport_cb *self);
-static int  irport_set_dtr_rts(struct net_device *dev, int dtr, int rts);
-static int  irport_raw_write(struct net_device *dev, __u8 *buf, int len);
-static struct net_device_stats *irport_net_get_stats(struct net_device *dev);
-static int irport_change_speed_complete(struct irda_task *task);
-static void irport_timeout(struct net_device *dev);
-
-static irqreturn_t irport_interrupt(int irq, void *dev_id);
-static int irport_hard_xmit(struct sk_buff *skb, struct net_device *dev);
-static void irport_change_speed(void *priv, __u32 speed);
-static int irport_net_open(struct net_device *dev);
-static int irport_net_close(struct net_device *dev);
-
-static struct irport_cb *
-irport_open(int i, unsigned int iobase, unsigned int irq)
-{
-	struct net_device *dev;
-	struct irport_cb *self;
-
-	IRDA_DEBUG(1, "%s()\n", __FUNCTION__);
-
-	/* Lock the port that we need */
-	if (!request_region(iobase, IO_EXTENT, driver_name)) {
-		IRDA_DEBUG(0, "%s(), can't get iobase of 0x%03x\n",
-			   __FUNCTION__, iobase);
-		goto err_out1;
-	}
-
-	/*
-	 *  Allocate new instance of the driver
-	 */
-	dev = alloc_irdadev(sizeof(struct irport_cb));
-	if (!dev) {
-		IRDA_ERROR("%s(), can't allocate memory for "
-			   "irda device!\n", __FUNCTION__);
-		goto err_out2;
-	}
-
-	self = dev->priv;
-	spin_lock_init(&self->lock);
-
-	/* Need to store self somewhere */
-	dev_self[i] = self;
-	self->priv = self;
-	self->index = i;
-
-	/* Initialize IO */
-	self->io.sir_base  = iobase;
-        self->io.sir_ext   = IO_EXTENT;
-        self->io.irq       = irq;
-        self->io.fifo_size = 16;		/* 16550A and compatible */
-
-	/* Initialize QoS for this device */
-	irda_init_max_qos_capabilies(&self->qos);
-	
-	self->qos.baud_rate.bits = IR_9600|IR_19200|IR_38400|IR_57600|
-		IR_115200;
-
-	self->qos.min_turn_time.bits = qos_mtt_bits;
-	irda_qos_bits_to_value(&self->qos);
-	
-	/* Bootstrap ZeroCopy Rx */
-	self->rx_buff.truesize = IRDA_SKB_MAX_MTU;
-	self->rx_buff.skb = __dev_alloc_skb(self->rx_buff.truesize,
-					    GFP_KERNEL);
-	if (self->rx_buff.skb == NULL) {
-		IRDA_ERROR("%s(), can't allocate memory for "
-			   "receive buffer!\n", __FUNCTION__);
-		goto err_out3;
-	}
-	skb_reserve(self->rx_buff.skb, 1);
-	self->rx_buff.head = self->rx_buff.skb->data;
-	/* No need to memset the buffer, unless you are really pedantic */
-
-	/* Finish setup the Rx buffer descriptor */
-	self->rx_buff.in_frame = FALSE;
-	self->rx_buff.state = OUTSIDE_FRAME;
-	self->rx_buff.data = self->rx_buff.head;
-
-	/* Specify how much memory we want */
-	self->tx_buff.truesize = 4000;
-	
-	/* Allocate memory if needed */
-	if (self->tx_buff.truesize > 0) {
-		self->tx_buff.head = kzalloc(self->tx_buff.truesize,
-						      GFP_KERNEL);
-		if (self->tx_buff.head == NULL) {
-			IRDA_ERROR("%s(), can't allocate memory for "
-				   "transmit buffer!\n", __FUNCTION__);
-			goto err_out4;
-		}
-	}	
-	self->tx_buff.data = self->tx_buff.head;
-
-	self->netdev = dev;
-
-	/* May be overridden by piggyback drivers */
-	self->interrupt    = irport_interrupt;
-	self->change_speed = irport_change_speed;
-
-	/* Override the network functions we need to use */
-	dev->hard_start_xmit = irport_hard_xmit;
-	dev->tx_timeout	     = irport_timeout;
-	dev->watchdog_timeo  = HZ;  /* Allow time enough for speed change */
-	dev->open            = irport_net_open;
-	dev->stop            = irport_net_close;
-	dev->get_stats	     = irport_net_get_stats;
-	dev->do_ioctl        = irport_net_ioctl;
-
-	/* Make ifconfig display some details */
-	dev->base_addr = iobase;
-	dev->irq = irq;
-
-	if (register_netdev(dev)) {
-		IRDA_ERROR("%s(), register_netdev() failed!\n", __FUNCTION__);
-		goto err_out5;
-	}
-	IRDA_MESSAGE("IrDA: Registered device %s (irport io=0x%X irq=%d)\n",
-		dev->name, iobase, irq);
-
-	return self;
- err_out5:
-	kfree(self->tx_buff.head);
- err_out4:
-	kfree_skb(self->rx_buff.skb);
- err_out3:
-	free_netdev(dev);
-	dev_self[i] = NULL;
- err_out2:
-	release_region(iobase, IO_EXTENT);
- err_out1:
-	return NULL;
-}
-
-static int irport_close(struct irport_cb *self)
-{
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	/* We are not using any dongle anymore! */
-	if (self->dongle)
-		irda_device_dongle_cleanup(self->dongle);
-	self->dongle = NULL;
-	
-	/* Remove netdevice */
-	unregister_netdev(self->netdev);
-
-	/* Release the IO-port that this driver is using */
-	IRDA_DEBUG(0 , "%s(), Releasing Region %03x\n", 
-		   __FUNCTION__, self->io.sir_base);
-	release_region(self->io.sir_base, self->io.sir_ext);
-
-	kfree(self->tx_buff.head);
-	
-	if (self->rx_buff.skb)
-		kfree_skb(self->rx_buff.skb);
-	self->rx_buff.skb = NULL;
-	
-	/* Remove ourselves */
-	dev_self[self->index] = NULL;
-	free_netdev(self->netdev);
-	
-	return 0;
-}
-
-static void irport_stop(struct irport_cb *self)
-{
-	int iobase;
-
-	iobase = self->io.sir_base;
-
-	/* We can't lock, we may be called from a FIR driver - Jean II */
-
-	/* We are not transmitting any more */
-	self->transmitting = 0;
-
-	/* Reset UART */
-	outb(0, iobase+UART_MCR);
-	
-	/* Turn off interrupts */
-	outb(0, iobase+UART_IER);
-}
-
-static void irport_start(struct irport_cb *self)
-{
-	int iobase;
-
-	iobase = self->io.sir_base;
-
-	irport_stop(self);
-	
-	/* We can't lock, we may be called from a FIR driver - Jean II */
-
-	/* Initialize UART */
-	outb(UART_LCR_WLEN8, iobase+UART_LCR);  /* Reset DLAB */
-	outb((UART_MCR_DTR | UART_MCR_RTS | UART_MCR_OUT2), iobase+UART_MCR);
-	
-	/* Turn on interrups */
-	outb(UART_IER_RLSI | UART_IER_RDI |UART_IER_THRI, iobase+UART_IER);
-}
-
-/*
- * Function irport_get_fcr (speed)
- *
- *    Compute value of fcr
- *
- */
-static inline unsigned int irport_get_fcr(__u32 speed)
-{
-	unsigned int fcr;    /* FIFO control reg */
-
-	/* Enable fifos */
-	fcr = UART_FCR_ENABLE_FIFO;
-
-	/* 
-	 * Use trigger level 1 to avoid 3 ms. timeout delay at 9600 bps, and
-	 * almost 1,7 ms at 19200 bps. At speeds above that we can just forget
-	 * about this timeout since it will always be fast enough. 
-	 */
-	if (speed < 38400)
-		fcr |= UART_FCR_TRIGGER_1;
-	else 
-		//fcr |= UART_FCR_TRIGGER_14;
-		fcr |= UART_FCR_TRIGGER_8;
-
-	return(fcr);
-}
- 
-/*
- * Function irport_change_speed (self, speed)
- *
- *    Set speed of IrDA port to specified baudrate
- *
- * This function should be called with irq off and spin-lock.
- */
-static void irport_change_speed(void *priv, __u32 speed)
-{
-	struct irport_cb *self = (struct irport_cb *) priv;
-	int iobase; 
-	unsigned int fcr;    /* FIFO control reg */
-	unsigned int lcr;    /* Line control reg */
-	int divisor;
-
-	IRDA_ASSERT(self != NULL, return;);
-	IRDA_ASSERT(speed != 0, return;);
-
-	IRDA_DEBUG(1, "%s(), Setting speed to: %d - iobase=%#x\n",
-		    __FUNCTION__, speed, self->io.sir_base);
-
-	/* We can't lock, we may be called from a FIR driver - Jean II */
-
-	iobase = self->io.sir_base;
-	
-	/* Update accounting for new speed */
-	self->io.speed = speed;
-
-	/* Turn off interrupts */
-	outb(0, iobase+UART_IER); 
-
-	divisor = SPEED_MAX/speed;
-	
-	/* Get proper fifo configuration */
-	fcr = irport_get_fcr(speed);
-
-	/* IrDA ports use 8N1 */
-	lcr = UART_LCR_WLEN8;
-	
-	outb(UART_LCR_DLAB | lcr, iobase+UART_LCR); /* Set DLAB */
-	outb(divisor & 0xff,      iobase+UART_DLL); /* Set speed */
-	outb(divisor >> 8,	  iobase+UART_DLM);
-	outb(lcr,		  iobase+UART_LCR); /* Set 8N1	*/
-	outb(fcr,		  iobase+UART_FCR); /* Enable FIFO's */
-
-	/* Turn on interrups */
-	/* This will generate a fatal interrupt storm.
-	 * People calling us will do that properly - Jean II */
-	//outb(/*UART_IER_RLSI|*/UART_IER_RDI/*|UART_IER_THRI*/, iobase+UART_IER);
-}
-
-/*
- * Function __irport_change_speed (instance, state, param)
- *
- *    State machine for changing speed of the device. We do it this way since
- *    we cannot use schedule_timeout() when we are in interrupt context
- *
- */
-static int __irport_change_speed(struct irda_task *task)
-{
-	struct irport_cb *self;
-	__u32 speed = (__u32) task->param;
-	unsigned long flags = 0;
-	int wasunlocked = 0;
-	int ret = 0;
-
-	IRDA_DEBUG(2, "%s(), <%ld>\n", __FUNCTION__, jiffies); 
-
-	self = (struct irport_cb *) task->instance;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	/* Locking notes : this function may be called from irq context with
-	 * spinlock, via irport_write_wakeup(), or from non-interrupt without
-	 * spinlock (from the task timer). Yuck !
-	 * This is ugly, and unsafe is the spinlock is not already acquired.
-	 * This will be fixed when irda-task get rewritten.
-	 * Jean II */
-	if (!spin_is_locked(&self->lock)) {
-		spin_lock_irqsave(&self->lock, flags);
-		wasunlocked = 1;
-	}
-
-	switch (task->state) {
-	case IRDA_TASK_INIT:
-	case IRDA_TASK_WAIT:
-		/* Are we ready to change speed yet? */
-		if (self->tx_buff.len > 0) {
-			task->state = IRDA_TASK_WAIT;
-
-			/* Try again later */
-			ret = msecs_to_jiffies(20);
-			break;
-		}
-
-		if (self->dongle)
-			irda_task_next_state(task, IRDA_TASK_CHILD_INIT);
-		else
-			irda_task_next_state(task, IRDA_TASK_CHILD_DONE);
-		break;
-	case IRDA_TASK_CHILD_INIT:
-		/* Go to default speed */
-		self->change_speed(self->priv, 9600);
-
-		/* Change speed of dongle */
-		if (irda_task_execute(self->dongle,
-				      self->dongle->issue->change_speed, 
-				      NULL, task, (void *) speed))
-		{
-			/* Dongle need more time to change its speed */
-			irda_task_next_state(task, IRDA_TASK_CHILD_WAIT);
-
-			/* Give dongle 1 sec to finish */
-			ret = msecs_to_jiffies(1000);
-		} else
-			/* Child finished immediately */
-			irda_task_next_state(task, IRDA_TASK_CHILD_DONE);
-		break;
-	case IRDA_TASK_CHILD_WAIT:
-		IRDA_WARNING("%s(), changing speed of dongle timed out!\n", __FUNCTION__);
-		ret = -1;		
-		break;
-	case IRDA_TASK_CHILD_DONE:
-		/* Finally we are ready to change the speed */
-		self->change_speed(self->priv, speed);
-		
-		irda_task_next_state(task, IRDA_TASK_DONE);
-		break;
-	default:
-		IRDA_ERROR("%s(), unknown state %d\n",
-			   __FUNCTION__, task->state);
-		irda_task_next_state(task, IRDA_TASK_DONE);
-		ret = -1;
-		break;
-	}
-	/* Put stuff in the state we found them - Jean II */
-	if(wasunlocked) {
-		spin_unlock_irqrestore(&self->lock, flags);
-	}
-
-	return ret;
-}
-
-/*
- * Function irport_change_speed_complete (task)
- *
- *    Called when the change speed operation completes
- *
- */
-static int irport_change_speed_complete(struct irda_task *task)
-{
-	struct irport_cb *self;
-
-	IRDA_DEBUG(1, "%s()\n", __FUNCTION__);
-
-	self = (struct irport_cb *) task->instance;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-	IRDA_ASSERT(self->netdev != NULL, return -1;);
-
-	/* Finished changing speed, so we are not busy any longer */
-	/* Signal network layer so it can try to send the frame */
-
-	netif_wake_queue(self->netdev);
-	
-	return 0;
-}
-
-/*
- * Function irport_timeout (struct net_device *dev)
- *
- *    The networking layer thinks we timed out.
- *
- */
-
-static void irport_timeout(struct net_device *dev)
-{
-	struct irport_cb *self;
-	int iobase;
-	int iir, lsr;
-	unsigned long flags;
-
-	self = (struct irport_cb *) dev->priv;
-	IRDA_ASSERT(self != NULL, return;);
-	iobase = self->io.sir_base;
-	
-	IRDA_WARNING("%s: transmit timed out, jiffies = %ld, trans_start = %ld\n",
-		dev->name, jiffies, dev->trans_start);
-	spin_lock_irqsave(&self->lock, flags);
-
-	/* Debug what's happening... */
-
-	/* Get interrupt status */
-	lsr = inb(iobase+UART_LSR);
-	/* Read interrupt register */
-	iir = inb(iobase+UART_IIR);
-	IRDA_DEBUG(0, "%s(), iir=%02x, lsr=%02x, iobase=%#x\n", 
-		   __FUNCTION__, iir, lsr, iobase);
-
-	IRDA_DEBUG(0, "%s(), transmitting=%d, remain=%d, done=%td\n",
-		   __FUNCTION__, self->transmitting, self->tx_buff.len,
-		   self->tx_buff.data - self->tx_buff.head);
-
-	/* Now, restart the port */
-	irport_start(self);
-	self->change_speed(self->priv, self->io.speed);
-	/* This will re-enable irqs */
-	outb(/*UART_IER_RLSI|*/UART_IER_RDI/*|UART_IER_THRI*/, iobase+UART_IER);
-	dev->trans_start = jiffies;
-	spin_unlock_irqrestore(&self->lock, flags);
-
-	netif_wake_queue(dev);
-}
- 
-/*
- * Function irport_wait_hw_transmitter_finish ()
- *
- *    Wait for the real end of HW transmission
- *
- * The UART is a strict FIFO, and we get called only when we have finished
- * pushing data to the FIFO, so the maximum amount of time we must wait
- * is only for the FIFO to drain out.
- *
- * We use a simple calibrated loop. We may need to adjust the loop
- * delay (udelay) to balance I/O traffic and latency. And we also need to
- * adjust the maximum timeout.
- * It would probably be better to wait for the proper interrupt,
- * but it doesn't seem to be available.
- *
- * We can't use jiffies or kernel timers because :
- * 1) We are called from the interrupt handler, which disable softirqs,
- * so jiffies won't be increased
- * 2) Jiffies granularity is usually very coarse (10ms), and we don't
- * want to wait that long to detect stuck hardware.
- * Jean II
- */
-
-static void irport_wait_hw_transmitter_finish(struct irport_cb *self)
-{
-	int iobase;
-	int count = 1000;	/* 1 ms */
-	
-	iobase = self->io.sir_base;
-
-	/* Calibrated busy loop */
-	while((count-- > 0) && !(inb(iobase+UART_LSR) & UART_LSR_TEMT))
-		udelay(1);
-
-	if(count == 0)
-		IRDA_DEBUG(0, "%s(): stuck transmitter\n", __FUNCTION__);
-}
-
-/*
- * Function irport_hard_start_xmit (struct sk_buff *skb, struct net_device *dev)
- *
- *    Transmits the current frame until FIFO is full, then
- *    waits until the next transmitt interrupt, and continues until the
- *    frame is transmitted.
- */
-static int irport_hard_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct irport_cb *self;
-	unsigned long flags;
-	int iobase;
-	s32 speed;
-
-	IRDA_DEBUG(1, "%s()\n", __FUNCTION__);
-
-	IRDA_ASSERT(dev != NULL, return 0;);
-	
-	self = (struct irport_cb *) dev->priv;
-	IRDA_ASSERT(self != NULL, return 0;);
-
-	iobase = self->io.sir_base;
-
-	netif_stop_queue(dev);
-
-	/* Make sure tests & speed change are atomic */
-	spin_lock_irqsave(&self->lock, flags);
-
-	/* Check if we need to change the speed */
-	speed = irda_get_next_speed(skb);
-	if ((speed != self->io.speed) && (speed != -1)) {
-		/* Check for empty frame */
-		if (!skb->len) {
-			/*
-			 * We send frames one by one in SIR mode (no
-			 * pipelining), so at this point, if we were sending
-			 * a previous frame, we just received the interrupt
-			 * telling us it is finished (UART_IIR_THRI).
-			 * Therefore, waiting for the transmitter to really
-			 * finish draining the fifo won't take too long.
-			 * And the interrupt handler is not expected to run.
-			 * - Jean II */
-			irport_wait_hw_transmitter_finish(self);
-			/* Better go there already locked - Jean II */
-			irda_task_execute(self, __irport_change_speed, 
-					  irport_change_speed_complete, 
-					  NULL, (void *) speed);
-			dev->trans_start = jiffies;
-			spin_unlock_irqrestore(&self->lock, flags);
-			dev_kfree_skb(skb);
-			return 0;
-		} else
-			self->new_speed = speed;
-	}
-
-	/* Init tx buffer */
-	self->tx_buff.data = self->tx_buff.head;
-
-        /* Copy skb to tx_buff while wrapping, stuffing and making CRC */
-	self->tx_buff.len = async_wrap_skb(skb, self->tx_buff.data, 
-					   self->tx_buff.truesize);
-	
-	self->stats.tx_bytes += self->tx_buff.len;
-
-	/* We are transmitting */
-	self->transmitting = 1;
-
-	/* Turn on transmit finished interrupt. Will fire immediately!  */
-	outb(UART_IER_THRI, iobase+UART_IER); 
-
-	dev->trans_start = jiffies;
-	spin_unlock_irqrestore(&self->lock, flags);
-
-	dev_kfree_skb(skb);
-	
-	return 0;
-}
-        
-/*
- * Function irport_write (driver)
- *
- *    Fill Tx FIFO with transmit data
- *
- * Called only from irport_write_wakeup()
- */
-static inline int irport_write(int iobase, int fifo_size, __u8 *buf, int len)
-{
-	int actual = 0;
-
-	/* Fill FIFO with current frame */
-	while ((actual < fifo_size) && (actual < len)) {
-		/* Transmit next byte */
-		outb(buf[actual], iobase+UART_TX);
-
-		actual++;
-	}
-        
-	return actual;
-}
-
-/*
- * Function irport_write_wakeup (tty)
- *
- *    Called by the driver when there's room for more data.  If we have
- *    more packets to send, we send them here.
- *
- * Called only from irport_interrupt()
- * Make sure this function is *not* called while we are receiving,
- * otherwise we will reset fifo and loose data :-(
- */
-static inline void irport_write_wakeup(struct irport_cb *self)
-{
-	int actual = 0;
-	int iobase;
-	unsigned int fcr;
-
-	IRDA_ASSERT(self != NULL, return;);
-
-	IRDA_DEBUG(4, "%s()\n", __FUNCTION__);
-
-	iobase = self->io.sir_base;
-
-	/* Finished with frame?  */
-	if (self->tx_buff.len > 0)  {
-		/* Write data left in transmit buffer */
-		actual = irport_write(iobase, self->io.fifo_size, 
-				      self->tx_buff.data, self->tx_buff.len);
-		self->tx_buff.data += actual;
-		self->tx_buff.len  -= actual;
-	} else {
-		/* 
-		 *  Now serial buffer is almost free & we can start 
-		 *  transmission of another packet. But first we must check
-		 *  if we need to change the speed of the hardware
-		 */
-		if (self->new_speed) {
-			irport_wait_hw_transmitter_finish(self);
-			irda_task_execute(self, __irport_change_speed, 
-					  irport_change_speed_complete, 
-					  NULL, (void *) self->new_speed);
-			self->new_speed = 0;
-		} else {
-			/* Tell network layer that we want more frames */
-			netif_wake_queue(self->netdev);
-		}
-		self->stats.tx_packets++;
-
-		/* 
-		 * Reset Rx FIFO to make sure that all reflected transmit data
-		 * is discarded. This is needed for half duplex operation
-		 */
-		fcr = irport_get_fcr(self->io.speed);
-		fcr |= UART_FCR_CLEAR_RCVR;
-		outb(fcr, iobase+UART_FCR);
-
-		/* Finished transmitting */
-		self->transmitting = 0;
-
-		/* Turn on receive interrupts */
-		outb(UART_IER_RDI, iobase+UART_IER);
-
-		IRDA_DEBUG(1, "%s() : finished Tx\n", __FUNCTION__);
-	}
-}
-
-/*
- * Function irport_receive (self)
- *
- *    Receive one frame from the infrared port
- *
- * Called only from irport_interrupt()
- */
-static inline void irport_receive(struct irport_cb *self) 
-{
-	int boguscount = 0;
-	int iobase;
-
-	IRDA_ASSERT(self != NULL, return;);
-
-	iobase = self->io.sir_base;
-
-	/*  
-	 * Receive all characters in Rx FIFO, unwrap and unstuff them. 
-         * async_unwrap_char will deliver all found frames  
-	 */
-	do {
-		async_unwrap_char(self->netdev, &self->stats, &self->rx_buff, 
-				  inb(iobase+UART_RX));
-
-		/* Make sure we don't stay here too long */
-		if (boguscount++ > 32) {
-			IRDA_DEBUG(2,"%s(), breaking!\n", __FUNCTION__);
-			break;
-		}
-	} while (inb(iobase+UART_LSR) & UART_LSR_DR);	
-}
-
-/*
- * Function irport_interrupt (irq, dev_id)
- *
- *    Interrupt handler
- */
-static irqreturn_t irport_interrupt(int irq, void *dev_id) 
-{
-	struct net_device *dev = dev_id;
-	struct irport_cb *self;
-	int boguscount = 0;
-	int iobase;
-	int iir, lsr;
-	int handled = 0;
-
-	self = dev->priv;
-
-	spin_lock(&self->lock);
-
-	iobase = self->io.sir_base;
-
-	/* Cut'n'paste interrupt routine from serial.c
-	 * This version try to minimise latency and I/O operations.
-	 * Simplified and modified to enforce half duplex operation.
-	 * - Jean II */
-
-	/* Check status even is iir reg is cleared, more robust and
-	 * eliminate a read on the I/O bus - Jean II */
-	do {
-		/* Get interrupt status ; Clear interrupt */
-		lsr = inb(iobase+UART_LSR);
-		
-		/* Are we receiving or transmitting ? */
-		if(!self->transmitting) {
-			/* Received something ? */
-			if (lsr & UART_LSR_DR)
-				irport_receive(self);
-		} else {
-			/* Room in Tx fifo ? */
-			if (lsr & (UART_LSR_THRE | UART_LSR_TEMT))
-				irport_write_wakeup(self);
-		}
-
-		/* A bit hackish, but working as expected... Jean II */
-		if(lsr & (UART_LSR_THRE | UART_LSR_TEMT | UART_LSR_DR))
-			handled = 1;
-
-		/* Make sure we don't stay here to long */
-		if (boguscount++ > 10) {
-			IRDA_WARNING("%s() irq handler looping : lsr=%02x\n",
-				     __FUNCTION__, lsr);
-			break;
-		}
-
-		/* Read interrupt register */
- 	        iir = inb(iobase+UART_IIR);
-
-		/* Enable this debug only when no other options and at low
-		 * bit rates, otherwise it may cause Rx overruns (lsr=63).
-		 * - Jean II */
-		IRDA_DEBUG(6, "%s(), iir=%02x, lsr=%02x, iobase=%#x\n", 
-			    __FUNCTION__, iir, lsr, iobase);
-
-		/* As long as interrupt pending... */
-	} while ((iir & UART_IIR_NO_INT) == 0);
-
-	spin_unlock(&self->lock);
-	return IRQ_RETVAL(handled);
-}
-
-/*
- * Function irport_net_open (dev)
- *
- *    Network device is taken up. Usually this is done by "ifconfig irda0 up" 
- *   
- */
-static int irport_net_open(struct net_device *dev)
-{
-	struct irport_cb *self;
-	int iobase;
-	char hwname[16];
-	unsigned long flags;
-
-	IRDA_DEBUG(2, "%s()\n", __FUNCTION__);
-
-	IRDA_ASSERT(dev != NULL, return -1;);
-	self = (struct irport_cb *) dev->priv;
-
-	iobase = self->io.sir_base;
-
-	if (request_irq(self->io.irq, self->interrupt, 0, dev->name, 
-			(void *) dev)) {
-		IRDA_DEBUG(0, "%s(), unable to allocate irq=%d\n",
-			   __FUNCTION__, self->io.irq);
-		return -EAGAIN;
-	}
-
-	spin_lock_irqsave(&self->lock, flags);
-	/* Init uart */
-	irport_start(self);
-	/* Set 9600 bauds per default, including at the dongle */
-	irda_task_execute(self, __irport_change_speed, 
-			  irport_change_speed_complete, 
-			  NULL, (void *) 9600);
-	spin_unlock_irqrestore(&self->lock, flags);
-
-
-	/* Give self a hardware name */
-	sprintf(hwname, "SIR @ 0x%03x", self->io.sir_base);
-
-	/* 
-	 * Open new IrLAP layer instance, now that everything should be
-	 * initialized properly 
-	 */
-	self->irlap = irlap_open(dev, &self->qos, hwname);
-
-	/* Ready to play! */
-
-	netif_start_queue(dev);
-
-	return 0;
-}
-
-/*
- * Function irport_net_close (self)
- *
- *    Network device is taken down. Usually this is done by 
- *    "ifconfig irda0 down" 
- */
-static int irport_net_close(struct net_device *dev)
-{
-	struct irport_cb *self;
-	int iobase;
-	unsigned long flags;
-
-	IRDA_DEBUG(4, "%s()\n", __FUNCTION__);
-
-	IRDA_ASSERT(dev != NULL, return -1;);
-	self = (struct irport_cb *) dev->priv;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	iobase = self->io.sir_base;
-
-	/* Stop device */
-	netif_stop_queue(dev);
-	
-	/* Stop and remove instance of IrLAP */
-	if (self->irlap)
-		irlap_close(self->irlap);
-	self->irlap = NULL;
-
-	spin_lock_irqsave(&self->lock, flags);
-	irport_stop(self);
-	spin_unlock_irqrestore(&self->lock, flags);
-
-	free_irq(self->io.irq, dev);
-
-	return 0;
-}
-
-/*
- * Function irport_is_receiving (self)
- *
- *    Returns true is we are currently receiving data
- *
- */
-static inline int irport_is_receiving(struct irport_cb *self)
-{
-	return (self->rx_buff.state != OUTSIDE_FRAME);
-}
-
-/*
- * Function irport_set_dtr_rts (tty, dtr, rts)
- *
- *    This function can be used by dongles etc. to set or reset the status
- *    of the dtr and rts lines
- */
-static int irport_set_dtr_rts(struct net_device *dev, int dtr, int rts)
-{
-	struct irport_cb *self = dev->priv;
-	int iobase;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	iobase = self->io.sir_base;
-
-	if (dtr)
-		dtr = UART_MCR_DTR;
-	if (rts)
-		rts = UART_MCR_RTS;
-
-	outb(dtr|rts|UART_MCR_OUT2, iobase+UART_MCR);
-
-	return 0;
-}
-
-static int irport_raw_write(struct net_device *dev, __u8 *buf, int len)
-{
-	struct irport_cb *self = (struct irport_cb *) dev->priv;
-	int actual = 0;
-	int iobase;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	iobase = self->io.sir_base;
-
-	/* Tx FIFO should be empty! */
-	if (!(inb(iobase+UART_LSR) & UART_LSR_THRE)) {
-		IRDA_DEBUG( 0, "%s(), failed, fifo not empty!\n", __FUNCTION__);
-		return -1;
-	}
-        
-	/* Fill FIFO with current frame */
-	while (actual < len) {
-		/* Transmit next byte */
-		outb(buf[actual], iobase+UART_TX);
-		actual++;
-	}
-
-	return actual;
-}
-
-/*
- * Function irport_net_ioctl (dev, rq, cmd)
- *
- *    Process IOCTL commands for this device
- *
- */
-static int irport_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct if_irda_req *irq = (struct if_irda_req *) rq;
-	struct irport_cb *self;
-	dongle_t *dongle;
-	unsigned long flags;
-	int ret = 0;
-
-	IRDA_ASSERT(dev != NULL, return -1;);
-
-	self = dev->priv;
-
-	IRDA_ASSERT(self != NULL, return -1;);
-
-	IRDA_DEBUG(2, "%s(), %s, (cmd=0x%X)\n", __FUNCTION__, dev->name, cmd);
-	
-	switch (cmd) {
-	case SIOCSBANDWIDTH: /* Set bandwidth */
-		if (!capable(CAP_NET_ADMIN))
-			ret = -EPERM;
-                else
-			irda_task_execute(self, __irport_change_speed, NULL, 
-					  NULL, (void *) irq->ifr_baudrate);
-		break;
-	case SIOCSDONGLE: /* Set dongle */
-		if (!capable(CAP_NET_ADMIN)) {
-			ret = -EPERM;
-			break;
-		}
-
-		/* Locking :
-		 * irda_device_dongle_init() can't be locked.
-		 * irda_task_execute() doesn't need to be locked.
-		 * Jean II
-		 */
-
-		/* Initialize dongle */
-		dongle = irda_device_dongle_init(dev, irq->ifr_dongle);
-		if (!dongle)
-			break;
-		
-		dongle->set_mode    = NULL;
-		dongle->read        = NULL;
-		dongle->write       = irport_raw_write;
-		dongle->set_dtr_rts = irport_set_dtr_rts;
-		
-		/* Now initialize the dongle!  */
-		dongle->issue->open(dongle, &self->qos);
-		
-		/* Reset dongle */
-		irda_task_execute(dongle, dongle->issue->reset, NULL, NULL, 
-				  NULL);	
-
-		/* Make dongle available to driver only now to avoid
-		 * race conditions - Jean II */
-		self->dongle = dongle;
-		break;
-	case SIOCSMEDIABUSY: /* Set media busy */
-		if (!capable(CAP_NET_ADMIN)) {
-			ret = -EPERM;
-			break;
-		}
-
-		irda_device_set_media_busy(self->netdev, TRUE);
-		break;
-	case SIOCGRECEIVING: /* Check if we are receiving right now */
-		irq->ifr_receiving = irport_is_receiving(self);
-		break;
-	case SIOCSDTRRTS:
-		if (!capable(CAP_NET_ADMIN)) {
-			ret = -EPERM;
-			break;
-		}
-
-		/* No real need to lock... */
-		spin_lock_irqsave(&self->lock, flags);
-		irport_set_dtr_rts(dev, irq->ifr_dtr, irq->ifr_rts);
-		spin_unlock_irqrestore(&self->lock, flags);
-		break;
-	default:
-		ret = -EOPNOTSUPP;
-	}
-	
-	return ret;
-}
-
-static struct net_device_stats *irport_net_get_stats(struct net_device *dev)
-{
-	struct irport_cb *self = (struct irport_cb *) dev->priv;
-	
-	return &self->stats;
-}
-
-static int __init irport_init(void)
-{
- 	int i;
-
- 	for (i=0; (io[i] < 2000) && (i < ARRAY_SIZE(dev_self)); i++) {
- 		if (irport_open(i, io[i], irq[i]) != NULL)
- 			return 0;
- 	}
-	/* 
-	 * Maybe something failed, but we can still be usable for FIR drivers 
-	 */
- 	return 0;
-}
-
-/*
- * Function irport_cleanup ()
- *
- *    Close all configured ports
- *
- */
-static void __exit irport_cleanup(void)
-{
- 	int i;
-
-        IRDA_DEBUG( 4, "%s()\n", __FUNCTION__);
-
-	for (i=0; i < ARRAY_SIZE(dev_self); i++) {
- 		if (dev_self[i])
- 			irport_close(dev_self[i]);
- 	}
-}
-
-module_param_array(io, int, NULL, 0);
-MODULE_PARM_DESC(io, "Base I/O addresses");
-module_param_array(irq, int, NULL, 0);
-MODULE_PARM_DESC(irq, "IRQ lines");
-
-MODULE_AUTHOR("Dag Brattli <dagb@cs.uit.no>");
-MODULE_DESCRIPTION("Half duplex serial driver for IrDA SIR mode");
-MODULE_LICENSE("GPL");
-
-module_init(irport_init);
-module_exit(irport_cleanup);
-
diff --git a/drivers/net/irda/irport.h b/drivers/net/irda/irport.h
deleted file mode 100644
index 39829cffba13..000000000000
--- a/drivers/net/irda/irport.h
+++ /dev/null
@@ -1,80 +0,0 @@
-/*********************************************************************
- *                
- * Filename:      irport.h
- * Version:       0.1
- * Description:   Serial driver for IrDA
- * Status:        Experimental.
- * Author:        Dag Brattli <dagb@cs.uit.no>
- * Created at:    Sun Aug  3 13:49:59 1997
- * Modified at:   Fri Jan 14 10:21:10 2000
- * Modified by:   Dag Brattli <dagb@cs.uit.no>
- * 
- *     Copyright (c) 1997, 1998-2000 Dag Brattli <dagb@cs.uit.no>
- *     All Rights Reserved.
- *     
- *     This program is free software; you can redistribute it and/or 
- *     modify it under the terms of the GNU General Public License as 
- *     published by the Free Software Foundation; either version 2 of 
- *     the License, or (at your option) any later version.
- *
- *     Neither Dag Brattli nor University of Tromsø admit liability nor
- *     provide warranty for any of this software. This material is 
- *     provided "AS-IS" and at no charge.
- *
- ********************************************************************/
-
-#ifndef IRPORT_H
-#define IRPORT_H
-
-#include <linux/netdevice.h>
-#include <linux/skbuff.h>
-#include <linux/types.h>
-#include <linux/spinlock.h>
-
-#include <net/irda/irda_device.h>
-
-#define SPEED_DEFAULT 9600
-#define SPEED_MAX     115200
-
-/*
- * These are the supported serial types.
- */
-#define PORT_UNKNOWN    0
-#define PORT_8250       1
-#define PORT_16450      2
-#define PORT_16550      3
-#define PORT_16550A     4
-#define PORT_CIRRUS     5
-#define PORT_16650      6
-#define PORT_MAX        6  
-
-#define FRAME_MAX_SIZE 2048
-
-struct irport_cb {
-	struct net_device *netdev; /* Yes! we are some kind of netdevice */
-	struct net_device_stats stats;
-
-	struct irlap_cb *irlap;    /* The link layer we are attached to */
-
-	chipio_t io;               /* IrDA controller information */
-	iobuff_t tx_buff;          /* Transmit buffer */
-	iobuff_t rx_buff;          /* Receive buffer */
-
-	struct qos_info qos;       /* QoS capabilities for this device */
-	dongle_t *dongle;          /* Dongle driver */
-
- 	__u32 flags;               /* Interface flags */
-	__u32 new_speed;
-	int mode;
-	int index;                 /* Instance index */
-	int transmitting;	   /* Are we transmitting ? */
-
-	spinlock_t lock;           /* For serializing operations */
-
-	/* For piggyback drivers */
-	void *priv;                
-	void (*change_speed)(void *priv, __u32 speed);
-	irq_handler_t interrupt;
-};
-
-#endif /* IRPORT_H */
diff --git a/include/net/irda/irda_device.h b/include/net/irda/irda_device.h
index bca19ca7bdd4..f70e9b39ebaf 100644
--- a/include/net/irda/irda_device.h
+++ b/include/net/irda/irda_device.h
@@ -228,21 +228,8 @@ static inline int irda_device_txqueue_empty(const struct net_device *dev)
 int  irda_device_set_raw_mode(struct net_device* self, int status);
 struct net_device *alloc_irdadev(int sizeof_priv);
 
-/* Dongle interface */
-void irda_device_unregister_dongle(struct dongle_reg *dongle);
-int  irda_device_register_dongle(struct dongle_reg *dongle);
-dongle_t *irda_device_dongle_init(struct net_device *dev, int type);
-int irda_device_dongle_cleanup(dongle_t *dongle);
-
 void irda_setup_dma(int channel, dma_addr_t buffer, int count, int mode);
 
-void irda_task_delete(struct irda_task *task);
-struct irda_task *irda_task_execute(void *instance, 
-				    IRDA_TASK_CALLBACK function, 
-				    IRDA_TASK_CALLBACK finished, 
-				    struct irda_task *parent, void *param);
-void irda_task_next_state(struct irda_task *task, IRDA_TASK_STATE state);
-
 /*
  * Function irda_get_mtt (skb)
  *
diff --git a/net/irda/irda_device.c b/net/irda/irda_device.c
index 435b563d29a6..87185910d0ee 100644
--- a/net/irda/irda_device.c
+++ b/net/irda/irda_device.c
@@ -57,20 +57,6 @@ static void __irda_task_delete(struct irda_task *task);
 static hashbin_t *dongles = NULL;
 static hashbin_t *tasks = NULL;
 
-#ifdef CONFIG_IRDA_DEBUG
-static const char *task_state[] = {
-	"IRDA_TASK_INIT",
-	"IRDA_TASK_DONE",
-	"IRDA_TASK_WAIT",
-	"IRDA_TASK_WAIT1",
-	"IRDA_TASK_WAIT2",
-	"IRDA_TASK_WAIT3",
-	"IRDA_TASK_CHILD_INIT",
-	"IRDA_TASK_CHILD_WAIT",
-	"IRDA_TASK_CHILD_DONE",
-};
-#endif	/* CONFIG_IRDA_DEBUG */
-
 static void irda_task_timer_expired(void *data);
 
 int __init irda_device_init( void)
@@ -176,14 +162,6 @@ int irda_device_is_receiving(struct net_device *dev)
 	return req.ifr_receiving;
 }
 
-void irda_task_next_state(struct irda_task *task, IRDA_TASK_STATE state)
-{
-	IRDA_DEBUG(2, "%s(), state = %s\n", __FUNCTION__, task_state[state]);
-
-	task->state = state;
-}
-EXPORT_SYMBOL(irda_task_next_state);
-
 static void __irda_task_delete(struct irda_task *task)
 {
 	del_timer(&task->timer);
@@ -191,14 +169,13 @@ static void __irda_task_delete(struct irda_task *task)
 	kfree(task);
 }
 
-void irda_task_delete(struct irda_task *task)
+static void irda_task_delete(struct irda_task *task)
 {
 	/* Unregister task */
 	hashbin_remove(tasks, (long) task, NULL);
 
 	__irda_task_delete(task);
 }
-EXPORT_SYMBOL(irda_task_delete);
 
 /*
  * Function irda_task_kick (task)
@@ -271,51 +248,6 @@ static int irda_task_kick(struct irda_task *task)
 	return finished;
 }
 
-/*
- * Function irda_task_execute (instance, function, finished)
- *
- *    This function registers and tries to execute tasks that may take some
- *    time to complete. We do it this hairy way since we may have been
- *    called from interrupt context, so it's not possible to use
- *    schedule_timeout()
- * Two important notes :
- *	o Make sure you irda_task_delete(task); in case you delete the
- *	  calling instance.
- *	o No real need to lock when calling this function, but you may
- *	  want to lock within the task handler.
- * Jean II
- */
-struct irda_task *irda_task_execute(void *instance,
-				    IRDA_TASK_CALLBACK function,
-				    IRDA_TASK_CALLBACK finished,
-				    struct irda_task *parent, void *param)
-{
-	struct irda_task *task;
-
-	IRDA_DEBUG(2, "%s()\n", __FUNCTION__);
-
-	task = kmalloc(sizeof(struct irda_task), GFP_ATOMIC);
-	if (!task)
-		return NULL;
-
-	task->state    = IRDA_TASK_INIT;
-	task->instance = instance;
-	task->function = function;
-	task->finished = finished;
-	task->parent   = parent;
-	task->param    = param;
-	task->magic    = IRDA_TASK_MAGIC;
-
-	init_timer(&task->timer);
-
-	/* Register task */
-	hashbin_insert(tasks, (irda_queue_t *) task, (long) task, NULL);
-
-	/* No time to waste, so lets get going! */
-	return irda_task_kick(task) ? NULL : task;
-}
-EXPORT_SYMBOL(irda_task_execute);
-
 /*
  * Function irda_task_timer_expired (data)
  *
@@ -364,105 +296,6 @@ struct net_device *alloc_irdadev(int sizeof_priv)
 }
 EXPORT_SYMBOL(alloc_irdadev);
 
-/*
- * Function irda_device_init_dongle (self, type, qos)
- *
- *    Initialize attached dongle.
- *
- * Important : request_module require us to call this function with
- * a process context and irq enabled. - Jean II
- */
-dongle_t *irda_device_dongle_init(struct net_device *dev, int type)
-{
-	struct dongle_reg *reg;
-	dongle_t *dongle = kzalloc(sizeof(dongle_t), GFP_KERNEL);
-
-	might_sleep();
-
-	spin_lock(&dongles->hb_spinlock);
-	reg = hashbin_find(dongles, type, NULL);
-
-#ifdef CONFIG_KMOD
-	/* Try to load the module needed */
-	if (!reg && capable(CAP_SYS_MODULE)) {
-		spin_unlock(&dongles->hb_spinlock);
-
-		request_module("irda-dongle-%d", type);
-
-		spin_lock(&dongles->hb_spinlock);
-		reg = hashbin_find(dongles, type, NULL);
-	}
-#endif
-
-	if (!reg || !try_module_get(reg->owner) ) {
-		IRDA_ERROR("IrDA: Unable to find requested dongle type %x\n",
-			   type);
-		kfree(dongle);
-		dongle = NULL;
-	}
-	if (dongle) {
-		/* Bind the registration info to this particular instance */
-		dongle->issue = reg;
-		dongle->dev = dev;
-	}
-	spin_unlock(&dongles->hb_spinlock);
-	return dongle;
-}
-EXPORT_SYMBOL(irda_device_dongle_init);
-
-/*
- * Function irda_device_dongle_cleanup (dongle)
- */
-int irda_device_dongle_cleanup(dongle_t *dongle)
-{
-	IRDA_ASSERT(dongle != NULL, return -1;);
-
-	dongle->issue->close(dongle);
-	module_put(dongle->issue->owner);
-	kfree(dongle);
-
-	return 0;
-}
-EXPORT_SYMBOL(irda_device_dongle_cleanup);
-
-/*
- * Function irda_device_register_dongle (dongle)
- */
-int irda_device_register_dongle(struct dongle_reg *new)
-{
-	spin_lock(&dongles->hb_spinlock);
-	/* Check if this dongle has been registered before */
-	if (hashbin_find(dongles, new->type, NULL)) {
-		IRDA_MESSAGE("%s: Dongle type %x already registered\n",
-			     __FUNCTION__, new->type);
-	} else {
-		/* Insert IrDA dongle into hashbin */
-		hashbin_insert(dongles, (irda_queue_t *) new, new->type, NULL);
-	}
-	spin_unlock(&dongles->hb_spinlock);
-
-	return 0;
-}
-EXPORT_SYMBOL(irda_device_register_dongle);
-
-/*
- * Function irda_device_unregister_dongle (dongle)
- *
- *    Unregister dongle, and remove dongle from list of registered dongles
- *
- */
-void irda_device_unregister_dongle(struct dongle_reg *dongle)
-{
-	struct dongle *node;
-
-	spin_lock(&dongles->hb_spinlock);
-	node = hashbin_remove(dongles, dongle->type, NULL);
-	if (!node)
-		IRDA_ERROR("%s: dongle not found!\n", __FUNCTION__);
-	spin_unlock(&dongles->hb_spinlock);
-}
-EXPORT_SYMBOL(irda_device_unregister_dongle);
-
 #ifdef CONFIG_ISA_DMA_API
 /*
  * Function setup_dma (idev, buffer, count, mode)
-- 
cgit v1.2.3


From 035923833369e4da5d3c4ad0700bc7c367a0fa37 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 20 Jan 2008 16:46:01 -0800
Subject: [FIB]: Add netns to fib_rules_ops.

The backward link from FIB rules operations to the network namespace
will allow to simplify the API a bit.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 1 +
 net/decnet/dn_rules.c   | 1 +
 net/ipv4/fib_rules.c    | 2 ++
 net/ipv6/fib6_rules.c   | 1 +
 4 files changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 4f472507129e..6910e0195ca5 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -67,6 +67,7 @@ struct fib_rules_ops
 	const struct nla_policy	*policy;
 	struct list_head	rules_list;
 	struct module		*owner;
+	struct net		*fro_net;
 };
 
 #define FRA_GENERIC_POLICY \
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index c1fae23d2263..964e65887603 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -249,6 +249,7 @@ static struct fib_rules_ops dn_fib_rules_ops = {
 	.policy		= dn_fib_rule_policy,
 	.rules_list	= LIST_HEAD_INIT(dn_fib_rules_ops.rules_list),
 	.owner		= THIS_MODULE,
+	.fro_net	= &init_net,
 };
 
 void __init dn_fib_rules_init(void)
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 72232ab4ecb1..8d0ebe7f360d 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -315,6 +315,8 @@ int __net_init fib4_rules_init(struct net *net)
 	if (ops == NULL)
 		return -ENOMEM;
 	INIT_LIST_HEAD(&ops->rules_list);
+	ops->fro_net = net;
+
 	fib_rules_register(net, ops);
 
 	err = fib_default_rules_init(ops);
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 76437a1fcab0..ead5ab2da9a2 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -249,6 +249,7 @@ static struct fib_rules_ops fib6_rules_ops = {
 	.policy			= fib6_rule_policy,
 	.rules_list		= LIST_HEAD_INIT(fib6_rules_ops.rules_list),
 	.owner			= THIS_MODULE,
+	.fro_net		= &init_net,
 };
 
 static int __init fib6_default_rules_init(void)
-- 
cgit v1.2.3


From 9e3a548781fc1c0da617fc65769a515f074be740 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 20 Jan 2008 16:46:41 -0800
Subject: [NETNS]: FIB rules API cleanup.

Remove struct net from fib_rules_register(unregister)/notify_change
paths and diet code size a bit.

add/remove: 0/0 grow/shrink: 10/12 up/down: 35/-100 (-65)
function                                     old     new   delta
notify_rule_change                           273     280      +7
trie_show_stats                              471     475      +4
fn_trie_delete                               473     477      +4
fib_rules_unregister                         144     148      +4
fib4_rule_compare                            119     123      +4
resize                                      2842    2845      +3
fn_trie_select_default                       515     518      +3
inet_sk_rebuild_header                       836     838      +2
fib_trie_seq_show                            764     766      +2
__devinet_sysctl_register                    276     278      +2
fn_trie_lookup                              1124    1123      -1
ip_fib_check_default                         133     131      -2
devinet_conf_sysctl                          223     221      -2
snmp_fold_field                              126     123      -3
fn_trie_insert                              2091    2086      -5
inet_create                                  876     870      -6
fib4_rules_init                              197     191      -6
fib_sync_down                                452     444      -8
inet_gso_send_check                          334     325      -9
fib_create_info                             3003    2991     -12
fib_nl_delrule                               568     553     -15
fib_nl_newrule                               883     852     -31

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h |  4 ++--
 net/core/fib_rules.c    | 20 +++++++++++++-------
 net/decnet/dn_rules.c   |  4 ++--
 net/ipv4/fib_rules.c    |  6 +++---
 net/ipv6/fib6_rules.c   |  4 ++--
 5 files changed, 22 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 6910e0195ca5..7f9f4aec8fc8 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -102,8 +102,8 @@ static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla)
 	return frh->table;
 }
 
-extern int fib_rules_register(struct net *, struct fib_rules_ops *);
-extern void fib_rules_unregister(struct net *, struct fib_rules_ops *);
+extern int fib_rules_register(struct fib_rules_ops *);
+extern void fib_rules_unregister(struct fib_rules_ops *);
 extern void                     fib_rules_cleanup_ops(struct fib_rules_ops *);
 
 extern int			fib_rules_lookup(struct fib_rules_ops *,
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 541728aa2bae..3cd4f13413f6 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -37,8 +37,7 @@ int fib_default_rule_add(struct fib_rules_ops *ops,
 }
 EXPORT_SYMBOL(fib_default_rule_add);
 
-static void notify_rule_change(struct net *net, int event,
-			       struct fib_rule *rule,
+static void notify_rule_change(int event, struct fib_rule *rule,
 			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
 			       u32 pid);
 
@@ -72,10 +71,13 @@ static void flush_route_cache(struct fib_rules_ops *ops)
 		ops->flush_cache();
 }
 
-int fib_rules_register(struct net *net, struct fib_rules_ops *ops)
+int fib_rules_register(struct fib_rules_ops *ops)
 {
 	int err = -EEXIST;
 	struct fib_rules_ops *o;
+	struct net *net;
+
+	net = ops->fro_net;
 
 	if (ops->rule_size < sizeof(struct fib_rule))
 		return -EINVAL;
@@ -112,8 +114,9 @@ void fib_rules_cleanup_ops(struct fib_rules_ops *ops)
 }
 EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops);
 
-void fib_rules_unregister(struct net *net, struct fib_rules_ops *ops)
+void fib_rules_unregister(struct fib_rules_ops *ops)
 {
+	struct net *net = ops->fro_net;
 
 	spin_lock(&net->rules_mod_lock);
 	list_del_rcu(&ops->list);
@@ -333,7 +336,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	else
 		list_add_rcu(&rule->list, &ops->rules_list);
 
-	notify_rule_change(net, RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
+	notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).pid);
 	flush_route_cache(ops);
 	rules_ops_put(ops);
 	return 0;
@@ -423,7 +426,7 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		}
 
 		synchronize_rcu();
-		notify_rule_change(net, RTM_DELRULE, rule, ops, nlh,
+		notify_rule_change(RTM_DELRULE, rule, ops, nlh,
 				   NETLINK_CB(skb).pid);
 		fib_rule_put(rule);
 		flush_route_cache(ops);
@@ -561,13 +564,15 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
-static void notify_rule_change(struct net *net, int event, struct fib_rule *rule,
+static void notify_rule_change(int event, struct fib_rule *rule,
 			       struct fib_rules_ops *ops, struct nlmsghdr *nlh,
 			       u32 pid)
 {
+	struct net *net;
 	struct sk_buff *skb;
 	int err = -ENOBUFS;
 
+	net = ops->fro_net;
 	skb = nlmsg_new(fib_rule_nlmsg_size(ops, rule), GFP_KERNEL);
 	if (skb == NULL)
 		goto errout;
@@ -579,6 +584,7 @@ static void notify_rule_change(struct net *net, int event, struct fib_rule *rule
 		kfree_skb(skb);
 		goto errout;
 	}
+
 	err = rtnl_notify(skb, net, pid, ops->nlgroup, nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 964e65887603..5b7539b7fe0c 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -256,12 +256,12 @@ void __init dn_fib_rules_init(void)
 {
 	BUG_ON(fib_default_rule_add(&dn_fib_rules_ops, 0x7fff,
 			            RT_TABLE_MAIN, 0));
-	fib_rules_register(&init_net, &dn_fib_rules_ops);
+	fib_rules_register(&dn_fib_rules_ops);
 }
 
 void __exit dn_fib_rules_cleanup(void)
 {
-	fib_rules_unregister(&init_net, &dn_fib_rules_ops);
+	fib_rules_unregister(&dn_fib_rules_ops);
 }
 
 
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 8d0ebe7f360d..3b7affd5cb3b 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -317,7 +317,7 @@ int __net_init fib4_rules_init(struct net *net)
 	INIT_LIST_HEAD(&ops->rules_list);
 	ops->fro_net = net;
 
-	fib_rules_register(net, ops);
+	fib_rules_register(ops);
 
 	err = fib_default_rules_init(ops);
 	if (err < 0)
@@ -327,13 +327,13 @@ int __net_init fib4_rules_init(struct net *net)
 
 fail:
 	/* also cleans all rules already added */
-	fib_rules_unregister(net, ops);
+	fib_rules_unregister(ops);
 	kfree(ops);
 	return err;
 }
 
 void __net_exit fib4_rules_exit(struct net *net)
 {
-	fib_rules_unregister(net, net->ipv4.rules_ops);
+	fib_rules_unregister(net->ipv4.rules_ops);
 	kfree(net->ipv4.rules_ops);
 }
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index ead5ab2da9a2..695c0ca8a417 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -274,7 +274,7 @@ int __init fib6_rules_init(void)
 	if (ret)
 		goto out;
 
-	ret = fib_rules_register(&init_net, &fib6_rules_ops);
+	ret = fib_rules_register(&fib6_rules_ops);
 	if (ret)
 		goto out_default_rules_init;
 out:
@@ -287,5 +287,5 @@ out_default_rules_init:
 
 void fib6_rules_cleanup(void)
 {
-	fib_rules_unregister(&init_net, &fib6_rules_ops);
+	fib_rules_unregister(&fib6_rules_ops);
 }
-- 
cgit v1.2.3


From 51314a17baabc710e5fb12975fe8983dedd5ac0d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 20 Jan 2008 16:47:09 -0800
Subject: [NETNS]: Process FIB rule action in the context of the namespace.

Save namespace context on the fib rule at the rule creation time and
call routing lookup in the correct namespace.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 1 +
 net/core/fib_rules.c    | 2 ++
 net/ipv4/fib_rules.c    | 2 +-
 3 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 7f9f4aec8fc8..34349f9f4331 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -22,6 +22,7 @@ struct fib_rule
 	u32			target;
 	struct fib_rule *	ctarget;
 	struct rcu_head		rcu;
+	struct net *		fr_net;
 };
 
 struct fib_lookup_arg
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 3cd4f13413f6..42ccaf5b8509 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -29,6 +29,7 @@ int fib_default_rule_add(struct fib_rules_ops *ops,
 	r->pref = pref;
 	r->table = table;
 	r->flags = flags;
+	r->fr_net = ops->fro_net;
 
 	/* The lock is not required here, the list in unreacheable
 	 * at the moment this function is called */
@@ -242,6 +243,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		err = -ENOMEM;
 		goto errout;
 	}
+	rule->fr_net = net;
 
 	if (tb[FRA_PRIORITY])
 		rule->pref = nla_get_u32(tb[FRA_PRIORITY]);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 3b7affd5cb3b..d2001f1c28a2 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -91,7 +91,7 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto errout;
 	}
 
-	if ((tbl = fib_get_table(&init_net, rule->table)) == NULL)
+	if ((tbl = fib_get_table(rule->fr_net, rule->table)) == NULL)
 		goto errout;
 
 	err = tbl->tb_lookup(tbl, flp, (struct fib_result *) arg->result);
-- 
cgit v1.2.3


From 476bcea67f9a1ca6f2c0028e75fb2129272c8398 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:18:26 -0800
Subject: [VLAN]: Remove unnecessary structure declarations

- struct packet_type is not used
- struct vlan_group is declared later in the file before the first use
- struct net_device is not needed since netdevice.h is included
- struct vlan_collection does not exist
- struct vlan_dev_info is declared later in the file before the first use

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 4562105fdb2b..a26805198b14 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -16,11 +16,6 @@
 #ifdef __KERNEL__
 
 /* externally defined structs */
-struct vlan_group;
-struct net_device;
-struct packet_type;
-struct vlan_collection;
-struct vlan_dev_info;
 struct hlist_node;
 
 #include <linux/netdevice.h>
-- 
cgit v1.2.3


From 740c15d0dd281c0cbe1a9ab1abc4f332e0df29bc Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:18:53 -0800
Subject: [VLAN]: Clean up vlan_hdr/vlan_ethhdr structs

Fix 3 space indentation and some overly long lines by moving the
comments to a kdoc structure description.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 33 +++++++++++++++++++++++----------
 1 file changed, 23 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index a26805198b14..a1b0066ec0d9 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -34,12 +34,30 @@ struct hlist_node;
 #define VLAN_ETH_DATA_LEN	1500	/* Max. octets in payload	 */
 #define VLAN_ETH_FRAME_LEN	1518	/* Max. octets in frame sans FCS */
 
+/*
+ * 	struct vlan_hdr - vlan header
+ * 	@h_vlan_TCI: priority and VLAN ID
+ *	@h_vlan_encapsulated_proto: packet type ID or len
+ */
+struct vlan_hdr {
+	__be16	h_vlan_TCI;
+	__be16	h_vlan_encapsulated_proto;
+};
+
+/**
+ *	struct vlan_ethhdr - vlan ethernet header (ethhdr + vlan_hdr)
+ *	@h_dest: destination ethernet address
+ *	@h_source: source ethernet address
+ *	@h_vlan_proto: ethernet protocol (always 0x8100)
+ *	@h_vlan_TCI: priority and VLAN ID
+ *	@h_vlan_encapsulated_proto: packet type ID or len
+ */
 struct vlan_ethhdr {
-   unsigned char	h_dest[ETH_ALEN];	   /* destination eth addr	*/
-   unsigned char	h_source[ETH_ALEN];	   /* source ether addr	*/
-   __be16               h_vlan_proto;              /* Should always be 0x8100 */
-   __be16               h_vlan_TCI;                /* Encapsulates priority and VLAN ID */
-   __be16		h_vlan_encapsulated_proto; /* packet type ID field (or len) */
+	unsigned char	h_dest[ETH_ALEN];
+	unsigned char	h_source[ETH_ALEN];
+	__be16		h_vlan_proto;
+	__be16		h_vlan_TCI;
+	__be16		h_vlan_encapsulated_proto;
 };
 
 #include <linux/skbuff.h>
@@ -49,11 +67,6 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb)
 	return (struct vlan_ethhdr *)skb_mac_header(skb);
 }
 
-struct vlan_hdr {
-   __be16               h_vlan_TCI;                /* Encapsulates priority and VLAN ID */
-   __be16               h_vlan_encapsulated_proto; /* packet type ID field (or len) */
-};
-
 #define VLAN_VID_MASK	0xfff
 
 /* found in socket.c */
-- 
cgit v1.2.3


From b7a4a83629c1ddde8c2e6a872618c66577cb20f0 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:19:16 -0800
Subject: [VLAN]: Kill useless VLAN_NAME define

The only user already includes __FUNCTION__ (vlan_proto_init) in the
output, which is enough to identify what the message is about.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 2 --
 net/8021q/vlan.c        | 4 ++--
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index a1b0066ec0d9..0325d6b17e07 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -72,8 +72,6 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb)
 /* found in socket.c */
 extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
 
-#define VLAN_NAME "vlan"
-
 /* if this changes, algorithm will have to be reworked because this
  * depends on completely exhausting the VLAN identifier space.  Thus
  * it gives constant time look-up, but in many cases it wastes memory.
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 032bf44eca5e..af2525569424 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -89,8 +89,8 @@ static int __init vlan_proto_init(void)
 	err = vlan_proc_init();
 	if (err < 0) {
 		printk(KERN_ERR
-		       "%s %s: can't create entry in proc filesystem!\n",
-		       __FUNCTION__, VLAN_NAME);
+		       "%s: can't create entry in proc filesystem!\n",
+		       __FUNCTION__);
 		return err;
 	}
 
-- 
cgit v1.2.3


From 7bd38d778e3f2250e96fc277040879d66c30ecb4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:19:31 -0800
Subject: [VLAN]: Use dev->stats

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 9 +--------
 net/8021q/vlan.c        | 2 --
 net/8021q/vlan_dev.c    | 8 ++++----
 net/8021q/vlanproc.c    | 5 +----
 4 files changed, 6 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 0325d6b17e07..07db4169463e 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -140,18 +140,11 @@ struct vlan_dev_info {
 	struct proc_dir_entry *dent;    /* Holds the proc data */
 	unsigned long cnt_inc_headroom_on_tx; /* How many times did we have to grow the skb on TX. */
 	unsigned long cnt_encap_on_xmit;      /* How many times did we have to encapsulate the skb on TX. */
-	struct net_device_stats dev_stats; /* Device stats (rx-bytes, tx-pkts, etc...) */
 };
 
 #define VLAN_DEV_INFO(x) ((struct vlan_dev_info *)(x->priv))
 
 /* inline functions */
-
-static inline struct net_device_stats *vlan_dev_get_stats(struct net_device *dev)
-{
-	return &(VLAN_DEV_INFO(dev)->dev_stats);
-}
-
 static inline __u32 vlan_get_ingress_priority(struct net_device *dev,
 					      unsigned short vlan_tag)
 {
@@ -196,7 +189,7 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb,
 
 	skb->dev->last_rx = jiffies;
 
-	stats = vlan_dev_get_stats(skb->dev);
+	stats = &skb->dev->stats;
 	stats->rx_packets++;
 	stats->rx_bytes += skb->len;
 
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index af2525569424..54f234696411 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -366,8 +366,6 @@ void vlan_setup(struct net_device *new_dev)
 	 * the global list.
 	 * iflink is set as well.
 	 */
-	new_dev->get_stats = vlan_dev_get_stats;
-
 	/* Make this thing known as a VLAN device */
 	new_dev->priv_flags |= IFF_802_1Q_VLAN;
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 4f99bb86af5c..9543e9182796 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -174,7 +174,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 	skb->dev->last_rx = jiffies;
 
 	/* Bump the rx counters for the VLAN device. */
-	stats = vlan_dev_get_stats(skb->dev);
+	stats = &skb->dev->stats;
 	stats->rx_packets++;
 	stats->rx_bytes += skb->len;
 
@@ -422,7 +422,7 @@ int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 		skb = skb_realloc_headroom(sk_tmp, dev->hard_header_len);
 		kfree_skb(sk_tmp);
 		if (skb == NULL) {
-			struct net_device_stats *stats = vlan_dev_get_stats(vdev);
+			struct net_device_stats *stats = &vdev->stats;
 			stats->tx_dropped++;
 			return -ENOMEM;
 		}
@@ -453,7 +453,7 @@ int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 
 int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-	struct net_device_stats *stats = vlan_dev_get_stats(dev);
+	struct net_device_stats *stats = &dev->stats;
 	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)(skb->data);
 
 	/* Handle non-VLAN frames if they are sent to us, for example by DHCP.
@@ -514,7 +514,7 @@ int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-	struct net_device_stats *stats = vlan_dev_get_stats(dev);
+	struct net_device_stats *stats = &dev->stats;
 	unsigned short veth_TCI;
 
 	/* Construct the second two bytes. This field looks something
diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c
index 6cefdf8e381a..1972d5cc34e4 100644
--- a/net/8021q/vlanproc.c
+++ b/net/8021q/vlanproc.c
@@ -316,7 +316,7 @@ static int vlandev_seq_show(struct seq_file *seq, void *offset)
 {
 	struct net_device *vlandev = (struct net_device *) seq->private;
 	const struct vlan_dev_info *dev_info = VLAN_DEV_INFO(vlandev);
-	struct net_device_stats *stats;
+	struct net_device_stats *stats = &vlandev->stats;
 	static const char fmt[] = "%30s %12lu\n";
 	int i;
 
@@ -327,9 +327,6 @@ static int vlandev_seq_show(struct seq_file *seq, void *offset)
 		       vlandev->name, dev_info->vlan_id,
 		       (int)(dev_info->flags & 1), vlandev->priv_flags);
 
-
-	stats = vlan_dev_get_stats(vlandev);
-
 	seq_printf(seq, fmt, "total frames received", stats->rx_packets);
 	seq_printf(seq, fmt, "total bytes received", stats->rx_bytes);
 	seq_printf(seq, fmt, "Broadcast/Multicast Rcvd", stats->multicast);
-- 
cgit v1.2.3


From a5250a36954c6658e28cc2e7e07e314e0c79e8bb Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:24:13 -0800
Subject: [ETHER]: Bring back MAC_FMT

The print_mac function is not very suitable for debugging printks
in performance critical paths since without ifdefs it will always
get called. MAC_FMT can be used with pr_debug without any overhead
when debugging is disabled.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_ether.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h
index 7a1e011b8a2c..e157c1399b61 100644
--- a/include/linux/if_ether.h
+++ b/include/linux/if_ether.h
@@ -130,6 +130,7 @@ extern ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len);
  *	Display a 6 byte device address (MAC) in a readable format.
  */
 extern char *print_mac(char *buf, const unsigned char *addr);
+#define MAC_FMT "%02x:%02x:%02x:%02x:%02x:%02x"
 #define MAC_BUF_SIZE	18
 #define DECLARE_MAC_BUF(var) char var[MAC_BUF_SIZE] __maybe_unused
 
-- 
cgit v1.2.3


From af30151709bcace1ca844d4bb8b7e2e392ff81eb Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:25:50 -0800
Subject: [VLAN]: Simplify vlan unregistration

Keep track of the number of VLAN devices in a vlan group. This allows
to have the caller sense when the group is going to be destroyed and
stop using it, which in turn allows to remove the wrapper around
unregister_vlan_dev for the NETDEV_UNREGISTER notifier and avoid
iterating over all possible VLAN ids whenever a device in unregistered.

Also fix what looks like a use-after-free (but is actually safe since
we're holding the RTNL), the real_dev reference should not be dropped
while we still use it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h  |  1 +
 net/8021q/vlan.c         | 76 +++++++++++++-----------------------------------
 net/8021q/vlan.h         |  2 +-
 net/8021q/vlan_netlink.c |  7 +----
 4 files changed, 23 insertions(+), 63 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 07db4169463e..129fa876dbe4 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -82,6 +82,7 @@ extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
 
 struct vlan_group {
 	int real_dev_ifindex; /* The ifindex of the ethernet(like) device the vlan is attached to. */
+	unsigned int		nr_vlans;
 	struct hlist_node	hlist;	/* linked list */
 	struct net_device **vlan_devices_arrays[VLAN_GROUP_ARRAY_SPLIT_PARTS];
 	struct rcu_head		rcu;
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index ad34e4a0326a..ac7963854103 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -132,33 +132,17 @@ static void vlan_rcu_free(struct rcu_head *rcu)
 	vlan_group_free(container_of(rcu, struct vlan_group, rcu));
 }
 
-
-/* This returns 0 if everything went fine.
- * It will return 1 if the group was killed as a result.
- * A negative return indicates failure.
- *
- * The RTNL lock must be held.
- */
-static int unregister_vlan_dev(struct net_device *real_dev,
-			       unsigned short vlan_id)
+void unregister_vlan_dev(struct net_device *dev)
 {
-	struct net_device *dev;
-	int real_dev_ifindex = real_dev->ifindex;
+	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct net_device *real_dev = vlan->real_dev;
 	struct vlan_group *grp;
-	unsigned int i;
-	int ret;
-
-	if (vlan_id >= VLAN_VID_MASK)
-		return -EINVAL;
+	unsigned short vlan_id = vlan->vlan_id;
 
 	ASSERT_RTNL();
-	grp = __vlan_find_group(real_dev_ifindex);
-	if (!grp)
-		return -ENOENT;
 
-	dev = vlan_group_get_device(grp, vlan_id);
-	if (!dev)
-		return -ENOENT;
+	grp = __vlan_find_group(real_dev->ifindex);
+	BUG_ON(!grp);
 
 	vlan_proc_rem_dev(dev);
 
@@ -169,20 +153,12 @@ static int unregister_vlan_dev(struct net_device *real_dev,
 		real_dev->vlan_rx_kill_vid(real_dev, vlan_id);
 
 	vlan_group_set_device(grp, vlan_id, NULL);
-	synchronize_net();
+	grp->nr_vlans--;
 
-	/* Caller unregisters (and if necessary, puts) VLAN device, but we
-	 * get rid of the reference to real_dev here.
-	 */
-	dev_put(real_dev);
+	synchronize_net();
 
 	/* If the group is now empty, kill off the group. */
-	ret = 0;
-	for (i = 0; i < VLAN_VID_MASK; i++)
-		if (vlan_group_get_device(grp, i))
-			break;
-
-	if (i == VLAN_VID_MASK) {
+	if (grp->nr_vlans == 0) {
 		if (real_dev->features & NETIF_F_HW_VLAN_RX)
 			real_dev->vlan_rx_register(real_dev, NULL);
 
@@ -190,23 +166,12 @@ static int unregister_vlan_dev(struct net_device *real_dev,
 
 		/* Free the group, after all cpu's are done. */
 		call_rcu(&grp->rcu, vlan_rcu_free);
-		ret = 1;
 	}
 
-	return ret;
-}
-
-int unregister_vlan_device(struct net_device *dev)
-{
-	int ret;
+	/* Get rid of the vlan's reference to real_dev */
+	dev_put(real_dev);
 
-	ret = unregister_vlan_dev(VLAN_DEV_INFO(dev)->real_dev,
-				  VLAN_DEV_INFO(dev)->vlan_id);
 	unregister_netdevice(dev);
-
-	if (ret == 1)
-		ret = 0;
-	return ret;
 }
 
 static void vlan_transfer_operstate(const struct net_device *dev, struct net_device *vlandev)
@@ -291,6 +256,8 @@ int register_vlan_dev(struct net_device *dev)
 	 * it into our local structure.
 	 */
 	vlan_group_set_device(grp, vlan_id, dev);
+	grp->nr_vlans++;
+
 	if (ngrp && real_dev->features & NETIF_F_HW_VLAN_RX)
 		real_dev->vlan_rx_register(real_dev, ngrp);
 	if (real_dev->features & NETIF_F_HW_VLAN_FILTER)
@@ -479,20 +446,16 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 	case NETDEV_UNREGISTER:
 		/* Delete all VLANs for this dev. */
 		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
-			int ret;
-
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
 
-			ret = unregister_vlan_dev(dev,
-						  VLAN_DEV_INFO(vlandev)->vlan_id);
-
-			unregister_netdevice(vlandev);
+			/* unregistration of last vlan destroys group, abort
+			 * afterwards */
+			if (grp->nr_vlans == 1)
+				i = VLAN_GROUP_ARRAY_LEN;
 
-			/* Group was destroyed? */
-			if (ret == 1)
-				break;
+			unregister_vlan_dev(vlandev);
 		}
 		break;
 	}
@@ -598,7 +561,8 @@ static int vlan_ioctl_handler(struct net *net, void __user *arg)
 		err = -EPERM;
 		if (!capable(CAP_NET_ADMIN))
 			break;
-		err = unregister_vlan_device(dev);
+		unregister_vlan_dev(dev);
+		err = 0;
 		break;
 
 	case GET_VLAN_REALDEV_NAME_CMD:
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 56378651cc46..0cfdf77b497c 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -38,7 +38,7 @@ void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result);
 int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id);
 void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
-int unregister_vlan_device(struct net_device *dev);
+void unregister_vlan_dev(struct net_device *dev);
 
 int vlan_netlink_init(void);
 void vlan_netlink_fini(void);
diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c
index 0996185e2ed5..9ee63583ed27 100644
--- a/net/8021q/vlan_netlink.c
+++ b/net/8021q/vlan_netlink.c
@@ -137,11 +137,6 @@ static int vlan_newlink(struct net_device *dev,
 	return register_vlan_dev(dev);
 }
 
-static void vlan_dellink(struct net_device *dev)
-{
-	unregister_vlan_device(dev);
-}
-
 static inline size_t vlan_qos_map_size(unsigned int n)
 {
 	if (n == 0)
@@ -226,7 +221,7 @@ struct rtnl_link_ops vlan_link_ops __read_mostly = {
 	.validate	= vlan_validate,
 	.newlink	= vlan_newlink,
 	.changelink	= vlan_changelink,
-	.dellink	= vlan_dellink,
+	.dellink	= unregister_vlan_dev,
 	.get_size	= vlan_get_size,
 	.fill_info	= vlan_fill_info,
 };
-- 
cgit v1.2.3


From 9dfebcc6479c55c001e4bb5fe7cc16b6799c43a7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:26:07 -0800
Subject: [VLAN]: Turn VLAN_DEV_INFO into inline function

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cxgb3/l2t.c      |  2 +-
 drivers/s390/net/qeth_main.c |  4 ++--
 include/linux/if_vlan.h      |  7 ++++--
 net/8021q/vlan.c             | 14 +++++------
 net/8021q/vlan_dev.c         | 56 ++++++++++++++++++++++----------------------
 net/8021q/vlan_netlink.c     | 10 ++++----
 net/8021q/vlanproc.c         | 12 +++++-----
 7 files changed, 54 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/drivers/net/cxgb3/l2t.c b/drivers/net/cxgb3/l2t.c
index d660af74606e..17ed4c3527b7 100644
--- a/drivers/net/cxgb3/l2t.c
+++ b/drivers/net/cxgb3/l2t.c
@@ -337,7 +337,7 @@ struct l2t_entry *t3_l2t_get(struct t3cdev *cdev, struct neighbour *neigh,
 		atomic_set(&e->refcnt, 1);
 		neigh_replace(e, neigh);
 		if (neigh->dev->priv_flags & IFF_802_1Q_VLAN)
-			e->vlan = VLAN_DEV_INFO(neigh->dev)->vlan_id;
+			e->vlan = vlan_dev_info(neigh->dev)->vlan_id;
 		else
 			e->vlan = VLAN_NONE;
 		spin_unlock(&e->lock);
diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c
index f1866a283850..62606ce26e55 100644
--- a/drivers/s390/net/qeth_main.c
+++ b/drivers/s390/net/qeth_main.c
@@ -3890,7 +3890,7 @@ qeth_verify_vlan_dev(struct net_device *dev, struct qeth_card *card)
 			break;
 		}
 	}
-	if (rc && !(VLAN_DEV_INFO(dev)->real_dev->priv == (void *)card))
+	if (rc && !(vlan_dev_info(dev)->real_dev->priv == (void *)card))
 		return 0;
 
 #endif
@@ -3930,7 +3930,7 @@ qeth_get_card_from_dev(struct net_device *dev)
 		card = (struct qeth_card *)dev->priv;
 	else if (rc == QETH_VLAN_CARD)
 		card = (struct qeth_card *)
-			VLAN_DEV_INFO(dev)->real_dev->priv;
+			vlan_dev_info(dev)->real_dev->priv;
 
 	QETH_DBF_TEXT_(trace, 4, "%d", rc);
 	return card ;
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 129fa876dbe4..82c23522a46d 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -143,13 +143,16 @@ struct vlan_dev_info {
 	unsigned long cnt_encap_on_xmit;      /* How many times did we have to encapsulate the skb on TX. */
 };
 
-#define VLAN_DEV_INFO(x) ((struct vlan_dev_info *)(x->priv))
+static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
+{
+	return netdev_priv(dev);
+}
 
 /* inline functions */
 static inline __u32 vlan_get_ingress_priority(struct net_device *dev,
 					      unsigned short vlan_tag)
 {
-	struct vlan_dev_info *vip = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vip = vlan_dev_info(dev);
 
 	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
 }
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index ac7963854103..d058c0ef3b6e 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -134,7 +134,7 @@ static void vlan_rcu_free(struct rcu_head *rcu)
 
 void unregister_vlan_dev(struct net_device *dev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
 	struct vlan_group *grp;
 	unsigned short vlan_id = vlan->vlan_id;
@@ -229,7 +229,7 @@ int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id)
 
 int register_vlan_dev(struct net_device *dev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
 	unsigned short vlan_id = vlan->vlan_id;
 	struct vlan_group *grp, *ngrp = NULL;
@@ -328,10 +328,10 @@ static int register_vlan_device(struct net_device *real_dev,
 	 */
 	new_dev->mtu = real_dev->mtu;
 
-	VLAN_DEV_INFO(new_dev)->vlan_id = VLAN_ID; /* 1 through VLAN_VID_MASK */
-	VLAN_DEV_INFO(new_dev)->real_dev = real_dev;
-	VLAN_DEV_INFO(new_dev)->dent = NULL;
-	VLAN_DEV_INFO(new_dev)->flags = VLAN_FLAG_REORDER_HDR;
+	vlan_dev_info(new_dev)->vlan_id = VLAN_ID; /* 1 through VLAN_VID_MASK */
+	vlan_dev_info(new_dev)->real_dev = real_dev;
+	vlan_dev_info(new_dev)->dent = NULL;
+	vlan_dev_info(new_dev)->flags = VLAN_FLAG_REORDER_HDR;
 
 	new_dev->rtnl_link_ops = &vlan_link_ops;
 	err = register_vlan_dev(new_dev);
@@ -348,7 +348,7 @@ out_free_newdev:
 static void vlan_sync_address(struct net_device *dev,
 			      struct net_device *vlandev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(vlandev);
+	struct vlan_dev_info *vlan = vlan_dev_info(vlandev);
 
 	/* May be called without an actual change */
 	if (!compare_ether_addr(vlan->real_dev_addr, dev->dev_addr))
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 756a71c1f4a5..a846559da85e 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -72,7 +72,7 @@ static int vlan_dev_rebuild_header(struct sk_buff *skb)
 
 static inline struct sk_buff *vlan_check_reorder_header(struct sk_buff *skb)
 {
-	if (VLAN_DEV_INFO(skb->dev)->flags & VLAN_FLAG_REORDER_HDR) {
+	if (vlan_dev_info(skb->dev)->flags & VLAN_FLAG_REORDER_HDR) {
 		if (skb_shared(skb) || skb_cloned(skb)) {
 			struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC);
 			kfree_skb(skb);
@@ -290,7 +290,7 @@ static inline unsigned short vlan_dev_get_egress_qos_mask(struct net_device* dev
 							  struct sk_buff* skb)
 {
 	struct vlan_priority_tci_mapping *mp =
-		VLAN_DEV_INFO(dev)->egress_priority_map[(skb->priority & 0xF)];
+		vlan_dev_info(dev)->egress_priority_map[(skb->priority & 0xF)];
 
 	while (mp) {
 		if (mp->priority == skb->priority) {
@@ -324,7 +324,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 	struct net_device *vdev = dev; /* save this for the bottom of the method */
 
 	pr_debug("%s: skb: %p type: %hx len: %u vlan_id: %hx, daddr: %p\n",
-		 __FUNCTION__, skb, type, len, VLAN_DEV_INFO(dev)->vlan_id, daddr);
+		 __FUNCTION__, skb, type, len, vlan_dev_info(dev)->vlan_id, daddr);
 
 	/* build vlan header only if re_order_header flag is NOT set.  This
 	 * fixes some programs that get confused when they see a VLAN device
@@ -334,7 +334,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 	 * header shuffling in the hard_start_xmit.  Users can turn off this
 	 * REORDER behaviour with the vconfig tool.
 	 */
-	if (!(VLAN_DEV_INFO(dev)->flags & VLAN_FLAG_REORDER_HDR))
+	if (!(vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR))
 		build_vlan_header = 1;
 
 	if (build_vlan_header) {
@@ -349,7 +349,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 		 * VLAN ID	 12 bits (low bits)
 		 *
 		 */
-		veth_TCI = VLAN_DEV_INFO(dev)->vlan_id;
+		veth_TCI = vlan_dev_info(dev)->vlan_id;
 		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
 
 		vhdr->h_vlan_TCI = htons(veth_TCI);
@@ -374,7 +374,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 	if (saddr == NULL)
 		saddr = dev->dev_addr;
 
-	dev = VLAN_DEV_INFO(dev)->real_dev;
+	dev = vlan_dev_info(dev)->real_dev;
 
 	/* MPLS can send us skbuffs w/out enough space.	 This check will grow the
 	 * skb if it doesn't have enough headroom.  Not a beautiful solution, so
@@ -395,7 +395,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 			stats->tx_dropped++;
 			return -ENOMEM;
 		}
-		VLAN_DEV_INFO(vdev)->cnt_inc_headroom_on_tx++;
+		vlan_dev_info(vdev)->cnt_inc_headroom_on_tx++;
 		pr_debug("%s: %s: had to grow skb.\n", __FUNCTION__, vdev->name);
 	}
 
@@ -430,12 +430,12 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 */
 
 	if (veth->h_vlan_proto != htons(ETH_P_8021Q) ||
-		VLAN_DEV_INFO(dev)->flags & VLAN_FLAG_REORDER_HDR) {
+		vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR) {
 		int orig_headroom = skb_headroom(skb);
 		unsigned short veth_TCI;
 
 		/* This is not a VLAN frame...but we can fix that! */
-		VLAN_DEV_INFO(dev)->cnt_encap_on_xmit++;
+		vlan_dev_info(dev)->cnt_encap_on_xmit++;
 
 		pr_debug("%s: proto to encap: 0x%hx\n",
 			 __FUNCTION__, htons(veth->h_vlan_proto));
@@ -445,7 +445,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * CFI		 1 bit
 		 * VLAN ID	 12 bits (low bits)
 		 */
-		veth_TCI = VLAN_DEV_INFO(dev)->vlan_id;
+		veth_TCI = vlan_dev_info(dev)->vlan_id;
 		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
 
 		skb = __vlan_put_tag(skb, veth_TCI);
@@ -455,7 +455,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 
 		if (orig_headroom < VLAN_HLEN) {
-			VLAN_DEV_INFO(dev)->cnt_inc_headroom_on_tx++;
+			vlan_dev_info(dev)->cnt_inc_headroom_on_tx++;
 		}
 	}
 
@@ -472,7 +472,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	stats->tx_packets++; /* for statics only */
 	stats->tx_bytes += skb->len;
 
-	skb->dev = VLAN_DEV_INFO(dev)->real_dev;
+	skb->dev = vlan_dev_info(dev)->real_dev;
 	dev_queue_xmit(skb);
 
 	return 0;
@@ -490,14 +490,14 @@ static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
 	 * CFI		 1 bit
 	 * VLAN ID	 12 bits (low bits)
 	 */
-	veth_TCI = VLAN_DEV_INFO(dev)->vlan_id;
+	veth_TCI = vlan_dev_info(dev)->vlan_id;
 	veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
 	skb = __vlan_hwaccel_put_tag(skb, veth_TCI);
 
 	stats->tx_packets++;
 	stats->tx_bytes += skb->len;
 
-	skb->dev = VLAN_DEV_INFO(dev)->real_dev;
+	skb->dev = vlan_dev_info(dev)->real_dev;
 	dev_queue_xmit(skb);
 
 	return 0;
@@ -508,7 +508,7 @@ static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu)
 	/* TODO: gotta make sure the underlying layer can handle it,
 	 * maybe an IFF_VLAN_CAPABLE flag for devices?
 	 */
-	if (VLAN_DEV_INFO(dev)->real_dev->mtu < new_mtu)
+	if (vlan_dev_info(dev)->real_dev->mtu < new_mtu)
 		return -ERANGE;
 
 	dev->mtu = new_mtu;
@@ -519,7 +519,7 @@ static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu)
 void vlan_dev_set_ingress_priority(const struct net_device *dev,
 				   u32 skb_prio, short vlan_prio)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 
 	if (vlan->ingress_priority_map[vlan_prio & 0x7] && !skb_prio)
 		vlan->nr_ingress_mappings--;
@@ -532,7 +532,7 @@ void vlan_dev_set_ingress_priority(const struct net_device *dev,
 int vlan_dev_set_egress_priority(const struct net_device *dev,
 				 u32 skb_prio, short vlan_prio)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct vlan_priority_tci_mapping *mp = NULL;
 	struct vlan_priority_tci_mapping *np;
 	u32 vlan_qos = (vlan_prio << 13) & 0xE000;
@@ -573,9 +573,9 @@ int vlan_dev_set_vlan_flag(const struct net_device *dev,
 	/* verify flag is supported */
 	if (flag == VLAN_FLAG_REORDER_HDR) {
 		if (flag_val) {
-			VLAN_DEV_INFO(dev)->flags |= VLAN_FLAG_REORDER_HDR;
+			vlan_dev_info(dev)->flags |= VLAN_FLAG_REORDER_HDR;
 		} else {
-			VLAN_DEV_INFO(dev)->flags &= ~VLAN_FLAG_REORDER_HDR;
+			vlan_dev_info(dev)->flags &= ~VLAN_FLAG_REORDER_HDR;
 		}
 		return 0;
 	}
@@ -584,17 +584,17 @@ int vlan_dev_set_vlan_flag(const struct net_device *dev,
 
 void vlan_dev_get_realdev_name(const struct net_device *dev, char *result)
 {
-	strncpy(result, VLAN_DEV_INFO(dev)->real_dev->name, 23);
+	strncpy(result, vlan_dev_info(dev)->real_dev->name, 23);
 }
 
 void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result)
 {
-	*result = VLAN_DEV_INFO(dev)->vlan_id;
+	*result = vlan_dev_info(dev)->vlan_id;
 }
 
 static int vlan_dev_open(struct net_device *dev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
 	int err;
 
@@ -618,7 +618,7 @@ static int vlan_dev_open(struct net_device *dev)
 
 static int vlan_dev_stop(struct net_device *dev)
 {
-	struct net_device *real_dev = VLAN_DEV_INFO(dev)->real_dev;
+	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
 
 	dev_mc_unsync(real_dev, dev);
 	if (dev->flags & IFF_ALLMULTI)
@@ -634,7 +634,7 @@ static int vlan_dev_stop(struct net_device *dev)
 
 static int vlan_dev_set_mac_address(struct net_device *dev, void *p)
 {
-	struct net_device *real_dev = VLAN_DEV_INFO(dev)->real_dev;
+	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
 	struct sockaddr *addr = p;
 	int err;
 
@@ -660,7 +660,7 @@ out:
 
 static int vlan_dev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
-	struct net_device *real_dev = VLAN_DEV_INFO(dev)->real_dev;
+	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
 	struct ifreq ifrr;
 	int err = -EOPNOTSUPP;
 
@@ -684,7 +684,7 @@ static int vlan_dev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 static void vlan_dev_change_rx_flags(struct net_device *dev, int change)
 {
-	struct net_device *real_dev = VLAN_DEV_INFO(dev)->real_dev;
+	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
 
 	if (change & IFF_ALLMULTI)
 		dev_set_allmulti(real_dev, dev->flags & IFF_ALLMULTI ? 1 : -1);
@@ -694,7 +694,7 @@ static void vlan_dev_change_rx_flags(struct net_device *dev, int change)
 
 static void vlan_dev_set_multicast_list(struct net_device *vlan_dev)
 {
-	dev_mc_sync(VLAN_DEV_INFO(vlan_dev)->real_dev, vlan_dev);
+	dev_mc_sync(vlan_dev_info(vlan_dev)->real_dev, vlan_dev);
 }
 
 /*
@@ -712,7 +712,7 @@ static const struct header_ops vlan_header_ops = {
 
 static int vlan_dev_init(struct net_device *dev)
 {
-	struct net_device *real_dev = VLAN_DEV_INFO(dev)->real_dev;
+	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
 	int subclass = 0;
 
 	/* IFF_BROADCAST|IFF_MULTICAST; ??? */
diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c
index 9ee63583ed27..e32eeb37987e 100644
--- a/net/8021q/vlan_netlink.c
+++ b/net/8021q/vlan_netlink.c
@@ -75,7 +75,7 @@ static int vlan_validate(struct nlattr *tb[], struct nlattr *data[])
 static int vlan_changelink(struct net_device *dev,
 			   struct nlattr *tb[], struct nlattr *data[])
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct ifla_vlan_flags *flags;
 	struct ifla_vlan_qos_mapping *m;
 	struct nlattr *attr;
@@ -104,7 +104,7 @@ static int vlan_changelink(struct net_device *dev,
 static int vlan_newlink(struct net_device *dev,
 			struct nlattr *tb[], struct nlattr *data[])
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev;
 	int err;
 
@@ -148,7 +148,7 @@ static inline size_t vlan_qos_map_size(unsigned int n)
 
 static size_t vlan_get_size(const struct net_device *dev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 
 	return nla_total_size(2) +	/* IFLA_VLAN_ID */
 	       vlan_qos_map_size(vlan->nr_ingress_mappings) +
@@ -157,14 +157,14 @@ static size_t vlan_get_size(const struct net_device *dev)
 
 static int vlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 {
-	struct vlan_dev_info *vlan = VLAN_DEV_INFO(dev);
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct vlan_priority_tci_mapping *pm;
 	struct ifla_vlan_flags f;
 	struct ifla_vlan_qos_mapping m;
 	struct nlattr *nest;
 	unsigned int i;
 
-	NLA_PUT_U16(skb, IFLA_VLAN_ID, VLAN_DEV_INFO(dev)->vlan_id);
+	NLA_PUT_U16(skb, IFLA_VLAN_ID, vlan_dev_info(dev)->vlan_id);
 	if (vlan->flags) {
 		f.flags = vlan->flags;
 		f.mask  = ~0;
diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c
index 971e6233801e..b5202443b1a1 100644
--- a/net/8021q/vlanproc.c
+++ b/net/8021q/vlanproc.c
@@ -180,7 +180,7 @@ err:
 
 int vlan_proc_add_dev (struct net_device *vlandev)
 {
-	struct vlan_dev_info *dev_info = VLAN_DEV_INFO(vlandev);
+	struct vlan_dev_info *dev_info = vlan_dev_info(vlandev);
 
 	dev_info->dent = create_proc_entry(vlandev->name,
 					   S_IFREG|S_IRUSR|S_IWUSR,
@@ -199,9 +199,9 @@ int vlan_proc_add_dev (struct net_device *vlandev)
 int vlan_proc_rem_dev(struct net_device *vlandev)
 {
 	/** NOTE:  This will consume the memory pointed to by dent, it seems. */
-	if (VLAN_DEV_INFO(vlandev)->dent) {
-		remove_proc_entry(VLAN_DEV_INFO(vlandev)->dent->name, proc_vlan_dir);
-		VLAN_DEV_INFO(vlandev)->dent = NULL;
+	if (vlan_dev_info(vlandev)->dent) {
+		remove_proc_entry(vlan_dev_info(vlandev)->dent->name, proc_vlan_dir);
+		vlan_dev_info(vlandev)->dent = NULL;
 	}
 	return 0;
 }
@@ -278,7 +278,7 @@ static int vlan_seq_show(struct seq_file *seq, void *v)
 			   nmtype ? nmtype :  "UNKNOWN" );
 	} else {
 		const struct net_device *vlandev = v;
-		const struct vlan_dev_info *dev_info = VLAN_DEV_INFO(vlandev);
+		const struct vlan_dev_info *dev_info = vlan_dev_info(vlandev);
 
 		seq_printf(seq, "%-15s| %d  | %s\n",  vlandev->name,
 			   dev_info->vlan_id,    dev_info->real_dev->name);
@@ -289,7 +289,7 @@ static int vlan_seq_show(struct seq_file *seq, void *v)
 static int vlandev_seq_show(struct seq_file *seq, void *offset)
 {
 	struct net_device *vlandev = (struct net_device *) seq->private;
-	const struct vlan_dev_info *dev_info = VLAN_DEV_INFO(vlandev);
+	const struct vlan_dev_info *dev_info = vlan_dev_info(vlandev);
 	struct net_device_stats *stats = &vlandev->stats;
 	static const char fmt[] = "%30s %12lu\n";
 	int i;
-- 
cgit v1.2.3


From 57d3ae847d4403c5e4a35ae5f38665fff1a94c02 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 21 Jan 2008 00:26:25 -0800
Subject: [VLAN]: Turn __constant_htons into htons where possible

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 82c23522a46d..34f40efc7607 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -271,12 +271,12 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 	memmove(skb->data, skb->data + VLAN_HLEN, 2 * VLAN_ETH_ALEN);
 
 	/* first, the ethernet type */
-	veth->h_vlan_proto = __constant_htons(ETH_P_8021Q);
+	veth->h_vlan_proto = htons(ETH_P_8021Q);
 
 	/* now, the tag */
 	veth->h_vlan_TCI = htons(tag);
 
-	skb->protocol = __constant_htons(ETH_P_8021Q);
+	skb->protocol = htons(ETH_P_8021Q);
 	skb->mac_header -= VLAN_HLEN;
 	skb->network_header -= VLAN_HLEN;
 
@@ -331,7 +331,7 @@ static inline int __vlan_get_tag(struct sk_buff *skb, unsigned short *tag)
 {
 	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data;
 
-	if (veth->h_vlan_proto != __constant_htons(ETH_P_8021Q)) {
+	if (veth->h_vlan_proto != htons(ETH_P_8021Q)) {
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3


From 1e637c74b0f84eaca02b914c0b8c6f67276e9697 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 21 Jan 2008 03:18:08 -0800
Subject: [IPV4]: Enable use of 240/4 address space.

This short patch modifies the IPv4 networking to enable use of the
240.0.0.0/4 (aka "class-E") address space as propsed in the internet
draft draft-fuller-240space-00.txt.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h      |  5 +++--
 include/net/addrconf.h  |  2 +-
 net/core/pktgen.c       |  2 +-
 net/ipv4/fib_frontend.c |  2 +-
 net/ipv4/route.c        | 12 ++++++------
 5 files changed, 12 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index 27d8a5ae9f75..70c6df882694 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -262,9 +262,10 @@ static inline bool ipv4_is_local_multicast(__be32 addr)
 	return (addr & htonl(0xffffff00)) == htonl(0xe0000000);
 }
 
-static inline bool ipv4_is_badclass(__be32 addr)
+static inline bool ipv4_is_lbcast(__be32 addr)
 {
-	return (addr & htonl(0xf0000000)) == htonl(0xf0000000);
+	/* limited broadcast */
+	return addr == INADDR_BROADCAST;
 }
 
 static inline bool ipv4_is_zeronet(__be32 addr)
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 8b1509bfc695..496503c03846 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -262,7 +262,7 @@ static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr)
 		  ipv4_is_private_172(addr) || ipv4_is_test_192(addr) ||
 		  ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) ||
 		  ipv4_is_test_198(addr) || ipv4_is_multicast(addr) ||
-		  ipv4_is_badclass(addr)) ? 0x00 : 0x02;
+		  ipv4_is_lbcast(addr)) ? 0x00 : 0x02;
 	eui[1] = 0;
 	eui[2] = 0x5E;
 	eui[3] = 0xFE;
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index d18fdb102360..eebccdbdbaca 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2266,7 +2266,7 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 
 					while (ipv4_is_loopback(s) ||
 					       ipv4_is_multicast(s) ||
-					       ipv4_is_badclass(s) ||
+					       ipv4_is_lbcast(s) ||
 					       ipv4_is_zeronet(s) ||
 					       ipv4_is_local_multicast(s)) {
 						t = random32() % (imx - imn) + imn;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 62bd791c204e..8987046d97f3 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -176,7 +176,7 @@ static inline unsigned __inet_dev_addr_type(struct net *net,
 	unsigned ret = RTN_BROADCAST;
 	struct fib_table *local_table;
 
-	if (ipv4_is_zeronet(addr) || ipv4_is_badclass(addr))
+	if (ipv4_is_zeronet(addr) || ipv4_is_lbcast(addr))
 		return RTN_BROADCAST;
 	if (ipv4_is_multicast(addr))
 		return RTN_MULTICAST;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index fc0145385e80..f80c761ea0bb 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1154,7 +1154,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 		return;
 
 	if (new_gw == old_gw || !IN_DEV_RX_REDIRECTS(in_dev)
-	    || ipv4_is_multicast(new_gw) || ipv4_is_badclass(new_gw)
+	    || ipv4_is_multicast(new_gw) || ipv4_is_lbcast(new_gw)
 	    || ipv4_is_zeronet(new_gw))
 		goto reject_redirect;
 
@@ -1634,7 +1634,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	if (in_dev == NULL)
 		return -EINVAL;
 
-	if (ipv4_is_multicast(saddr) || ipv4_is_badclass(saddr) ||
+	if (ipv4_is_multicast(saddr) || ipv4_is_lbcast(saddr) ||
 	    ipv4_is_loopback(saddr) || skb->protocol != htons(ETH_P_IP))
 		goto e_inval;
 
@@ -1891,7 +1891,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	   by fib_lookup.
 	 */
 
-	if (ipv4_is_multicast(saddr) || ipv4_is_badclass(saddr) ||
+	if (ipv4_is_multicast(saddr) || ipv4_is_lbcast(saddr) ||
 	    ipv4_is_loopback(saddr))
 		goto martian_source;
 
@@ -1904,7 +1904,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	if (ipv4_is_zeronet(saddr))
 		goto martian_source;
 
-	if (ipv4_is_badclass(daddr) || ipv4_is_zeronet(daddr) ||
+	if (ipv4_is_lbcast(daddr) || ipv4_is_zeronet(daddr) ||
 	    ipv4_is_loopback(daddr))
 		goto martian_destination;
 
@@ -2125,7 +2125,7 @@ static inline int __mkroute_output(struct rtable **result,
 		res->type = RTN_BROADCAST;
 	else if (ipv4_is_multicast(fl->fl4_dst))
 		res->type = RTN_MULTICAST;
-	else if (ipv4_is_badclass(fl->fl4_dst) || ipv4_is_zeronet(fl->fl4_dst))
+	else if (ipv4_is_lbcast(fl->fl4_dst) || ipv4_is_zeronet(fl->fl4_dst))
 		return -EINVAL;
 
 	if (dev_out->flags & IFF_LOOPBACK)
@@ -2276,7 +2276,7 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
 	if (oldflp->fl4_src) {
 		err = -EINVAL;
 		if (ipv4_is_multicast(oldflp->fl4_src) ||
-		    ipv4_is_badclass(oldflp->fl4_src) ||
+		    ipv4_is_lbcast(oldflp->fl4_src) ||
 		    ipv4_is_zeronet(oldflp->fl4_src))
 			goto out;
 
-- 
cgit v1.2.3


From d9e94d5647ee6700773d81514a8ccb7dc6342fb4 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 29 Dec 2007 05:01:07 -0500
Subject: ieee80211: fix misannotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h              |  6 +++---
 net/ieee80211/ieee80211_crypt_tkip.c | 22 +++++++++++-----------
 net/ieee80211/ieee80211_rx.c         |  2 +-
 net/ieee80211/ieee80211_tx.c         | 14 +++++++-------
 4 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index d8ae48439f12..285b2adfa648 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -677,7 +677,7 @@ struct ieee80211_probe_request {
 
 struct ieee80211_probe_response {
 	struct ieee80211_hdr_3addr header;
-	u32 time_stamp[2];
+	__le32 time_stamp[2];
 	__le16 beacon_interval;
 	__le16 capability;
 	/* SSID, supported rates, FH params, DS params,
@@ -718,8 +718,8 @@ struct ieee80211_txb {
 	u8 encrypted;
 	u8 rts_included;
 	u8 reserved;
-	__le16 frag_size;
-	__le16 payload_size;
+	u16 frag_size;
+	u16 payload_size;
 	struct sk_buff *fragments[0];
 };
 
diff --git a/net/ieee80211/ieee80211_crypt_tkip.c b/net/ieee80211/ieee80211_crypt_tkip.c
index 8e146949fc6f..bba0152e2d71 100644
--- a/net/ieee80211/ieee80211_crypt_tkip.c
+++ b/net/ieee80211/ieee80211_crypt_tkip.c
@@ -189,7 +189,7 @@ static inline u16 Mk16(u8 hi, u8 lo)
 	return lo | (((u16) hi) << 8);
 }
 
-static inline u16 Mk16_le(u16 * v)
+static inline u16 Mk16_le(__le16 * v)
 {
 	return le16_to_cpu(*v);
 }
@@ -275,15 +275,15 @@ static void tkip_mixing_phase2(u8 * WEPSeed, const u8 * TK, const u16 * TTAK,
 	PPK[5] = TTAK[4] + IV16;
 
 	/* Step 2 - 96-bit bijective mixing using S-box */
-	PPK[0] += _S_(PPK[5] ^ Mk16_le((u16 *) & TK[0]));
-	PPK[1] += _S_(PPK[0] ^ Mk16_le((u16 *) & TK[2]));
-	PPK[2] += _S_(PPK[1] ^ Mk16_le((u16 *) & TK[4]));
-	PPK[3] += _S_(PPK[2] ^ Mk16_le((u16 *) & TK[6]));
-	PPK[4] += _S_(PPK[3] ^ Mk16_le((u16 *) & TK[8]));
-	PPK[5] += _S_(PPK[4] ^ Mk16_le((u16 *) & TK[10]));
-
-	PPK[0] += RotR1(PPK[5] ^ Mk16_le((u16 *) & TK[12]));
-	PPK[1] += RotR1(PPK[0] ^ Mk16_le((u16 *) & TK[14]));
+	PPK[0] += _S_(PPK[5] ^ Mk16_le((__le16 *) & TK[0]));
+	PPK[1] += _S_(PPK[0] ^ Mk16_le((__le16 *) & TK[2]));
+	PPK[2] += _S_(PPK[1] ^ Mk16_le((__le16 *) & TK[4]));
+	PPK[3] += _S_(PPK[2] ^ Mk16_le((__le16 *) & TK[6]));
+	PPK[4] += _S_(PPK[3] ^ Mk16_le((__le16 *) & TK[8]));
+	PPK[5] += _S_(PPK[4] ^ Mk16_le((__le16 *) & TK[10]));
+
+	PPK[0] += RotR1(PPK[5] ^ Mk16_le((__le16 *) & TK[12]));
+	PPK[1] += RotR1(PPK[0] ^ Mk16_le((__le16 *) & TK[14]));
 	PPK[2] += RotR1(PPK[1]);
 	PPK[3] += RotR1(PPK[2]);
 	PPK[4] += RotR1(PPK[3]);
@@ -294,7 +294,7 @@ static void tkip_mixing_phase2(u8 * WEPSeed, const u8 * TK, const u16 * TTAK,
 	WEPSeed[0] = Hi8(IV16);
 	WEPSeed[1] = (Hi8(IV16) | 0x20) & 0x7F;
 	WEPSeed[2] = Lo8(IV16);
-	WEPSeed[3] = Lo8((PPK[5] ^ Mk16_le((u16 *) & TK[0])) >> 1);
+	WEPSeed[3] = Lo8((PPK[5] ^ Mk16_le((__le16 *) & TK[0])) >> 1);
 
 #ifdef __BIG_ENDIAN
 	{
diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c
index 0a18edbdfcb5..6a9c07f9b10f 100644
--- a/net/ieee80211/ieee80211_rx.c
+++ b/net/ieee80211/ieee80211_rx.c
@@ -754,7 +754,7 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
 		memcpy(skb_push(skb, ETH_ALEN), src, ETH_ALEN);
 		memcpy(skb_push(skb, ETH_ALEN), dst, ETH_ALEN);
 	} else {
-		u16 len;
+		__be16 len;
 		/* Leave Ethernet header part of hdr and full payload */
 		skb_pull(skb, hdrlen);
 		len = htons(skb->len);
diff --git a/net/ieee80211/ieee80211_tx.c b/net/ieee80211/ieee80211_tx.c
index 6d06f1385e28..d8b02603cbe5 100644
--- a/net/ieee80211/ieee80211_tx.c
+++ b/net/ieee80211/ieee80211_tx.c
@@ -126,7 +126,7 @@ payload of each frame is reduced to 492 bytes.
 static u8 P802_1H_OUI[P80211_OUI_LEN] = { 0x00, 0x00, 0xf8 };
 static u8 RFC1042_OUI[P80211_OUI_LEN] = { 0x00, 0x00, 0x00 };
 
-static int ieee80211_copy_snap(u8 * data, u16 h_proto)
+static int ieee80211_copy_snap(u8 * data, __be16 h_proto)
 {
 	struct ieee80211_snap_hdr *snap;
 	u8 *oui;
@@ -136,7 +136,7 @@ static int ieee80211_copy_snap(u8 * data, u16 h_proto)
 	snap->ssap = 0xaa;
 	snap->ctrl = 0x03;
 
-	if (h_proto == 0x8137 || h_proto == 0x80f3)
+	if (h_proto == htons(ETH_P_AARP) || h_proto == htons(ETH_P_IPX))
 		oui = P802_1H_OUI;
 	else
 		oui = RFC1042_OUI;
@@ -144,7 +144,6 @@ static int ieee80211_copy_snap(u8 * data, u16 h_proto)
 	snap->oui[1] = oui[1];
 	snap->oui[2] = oui[2];
 
-	h_proto = htons(h_proto);
 	memcpy(data + SNAP_SIZE, &h_proto, sizeof(u16));
 
 	return SNAP_SIZE + sizeof(u16);
@@ -261,7 +260,8 @@ int ieee80211_xmit(struct sk_buff *skb, struct net_device *dev)
 	    rts_required;
 	unsigned long flags;
 	struct net_device_stats *stats = &ieee->stats;
-	int ether_type, encrypt, host_encrypt, host_encrypt_msdu, host_build_iv;
+	int encrypt, host_encrypt, host_encrypt_msdu, host_build_iv;
+	__be16 ether_type;
 	int bytes, fc, hdr_len;
 	struct sk_buff *skb_frag;
 	struct ieee80211_hdr_3addrqos header = {/* Ensure zero initialized */
@@ -292,11 +292,11 @@ int ieee80211_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto success;
 	}
 
-	ether_type = ntohs(((struct ethhdr *)skb->data)->h_proto);
+	ether_type = ((struct ethhdr *)skb->data)->h_proto;
 
 	crypt = ieee->crypt[ieee->tx_keyidx];
 
-	encrypt = !(ether_type == ETH_P_PAE && ieee->ieee802_1x) &&
+	encrypt = !(ether_type == htons(ETH_P_PAE) && ieee->ieee802_1x) &&
 	    ieee->sec.encrypt;
 
 	host_encrypt = ieee->host_encrypt && encrypt && crypt;
@@ -304,7 +304,7 @@ int ieee80211_xmit(struct sk_buff *skb, struct net_device *dev)
 	host_build_iv = ieee->host_build_iv && encrypt && crypt;
 
 	if (!encrypt && ieee->ieee802_1x &&
-	    ieee->drop_unencrypted && ether_type != ETH_P_PAE) {
+	    ieee->drop_unencrypted && ether_type != htons(ETH_P_PAE)) {
 		stats->tx_dropped++;
 		goto success;
 	}
-- 
cgit v1.2.3


From e861b98d5e1be769ca6483b6df97149b956ea834 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Sat, 22 Dec 2007 21:51:30 +0100
Subject: ssb: Fix extraction of values from SPROM

This fixes extraction of some values from the SPROM.
It mainly fixes extraction of antenna related values, which
is needed for another b43 fix sent later.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/b43/main.c       | 10 -----
 drivers/net/wireless/b43legacy/main.c |  5 ---
 drivers/net/wireless/b43legacy/phy.c  |  2 +-
 drivers/ssb/pci.c                     | 76 ++++++++++++++++++++++++++++++-----
 include/linux/ssb/ssb.h               | 19 ++++++++-
 include/linux/ssb/ssb_regs.h          | 38 +++++++++++-------
 6 files changed, 107 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index d7ea671394a8..68bbe8eafd6d 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -3884,16 +3884,6 @@ static void b43_sprom_fixup(struct ssb_bus *bus)
 	if (bus->boardinfo.vendor == PCI_VENDOR_ID_APPLE &&
 	    bus->boardinfo.type == 0x4E && bus->boardinfo.rev > 0x40)
 		bus->sprom.boardflags_lo |= B43_BFL_PACTRL;
-
-	/* Handle case when gain is not set in sprom */
-	if (bus->sprom.antenna_gain_a == 0xFF)
-		bus->sprom.antenna_gain_a = 2;
-	if (bus->sprom.antenna_gain_bg == 0xFF)
-		bus->sprom.antenna_gain_bg = 2;
-
-	/* Convert Antennagain values to Q5.2 */
-	bus->sprom.antenna_gain_a <<= 2;
-	bus->sprom.antenna_gain_bg <<= 2;
 }
 
 static void b43_wireless_exit(struct ssb_device *dev, struct b43_wl *wl)
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index 14087fc20f3a..575fd9a5874d 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -3572,11 +3572,6 @@ static void b43legacy_sprom_fixup(struct ssb_bus *bus)
 	    bus->boardinfo.type == 0x4E &&
 	    bus->boardinfo.rev > 0x40)
 		bus->sprom.boardflags_lo |= B43legacy_BFL_PACTRL;
-
-	/* Convert Antennagain values to Q5.2 */
-	if (bus->sprom.antenna_gain_bg == 0xFF)
-		bus->sprom.antenna_gain_bg = 2; /* if unset, use 2 dBm */
-	bus->sprom.antenna_gain_bg <<= 2;
 }
 
 static void b43legacy_wireless_exit(struct ssb_device *dev,
diff --git a/drivers/net/wireless/b43legacy/phy.c b/drivers/net/wireless/b43legacy/phy.c
index 9d527e6d6ced..57c668f575fc 100644
--- a/drivers/net/wireless/b43legacy/phy.c
+++ b/drivers/net/wireless/b43legacy/phy.c
@@ -1859,7 +1859,7 @@ void b43legacy_phy_xmitpower(struct b43legacy_wldev *dev)
 	 * which accounts for the factor of 4 */
 #define REG_MAX_PWR 20
 	max_pwr = min(REG_MAX_PWR * 4
-		      - dev->dev->bus->sprom.antenna_gain_bg
+		      - dev->dev->bus->sprom.antenna_gain.ghz24.a0
 		      - 0x6, max_pwr);
 
 	/* find the desired power in Q5.2 - power_level is in dBm
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index 9777dcb5bfe5..ed2a38752274 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -247,7 +247,7 @@ static void sprom_do_read(struct ssb_bus *bus, u16 *sprom)
 	int i;
 
 	for (i = 0; i < bus->sprom_size; i++)
-		sprom[i] = readw(bus->mmio + SSB_SPROM_BASE + (i * 2));
+		sprom[i] = ioread16(bus->mmio + SSB_SPROM_BASE + (i * 2));
 }
 
 static int sprom_do_write(struct ssb_bus *bus, const u16 *sprom)
@@ -297,10 +297,32 @@ err_ctlreg:
 	return err;
 }
 
+static s8 r123_extract_antgain(u8 sprom_revision, const u16 *in,
+			       u16 mask, u16 shift)
+{
+	u16 v;
+	u8 gain;
+
+	v = in[SPOFF(SSB_SPROM1_AGAIN)];
+	gain = (v & mask) >> shift;
+	if (gain == 0xFF)
+		gain = 2; /* If unset use 2dBm */
+	if (sprom_revision == 1) {
+		/* Convert to Q5.2 */
+		gain <<= 2;
+	} else {
+		/* Q5.2 Fractional part is stored in 0xC0 */
+		gain = ((gain & 0xC0) >> 6) | ((gain & 0x3F) << 2);
+	}
+
+	return (s8)gain;
+}
+
 static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 {
 	int i;
 	u16 v;
+	s8 gain;
 	u16 loc[3];
 
 	if (out->revision == 3) {			/* rev 3 moved MAC */
@@ -327,8 +349,15 @@ static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 	SPEX(et0phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET0A, 0);
 	SPEX(et1phyaddr, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET1A,
 	     SSB_SPROM1_ETHPHY_ET1A_SHIFT);
+	SPEX(et0mdcport, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET0M, 14);
+	SPEX(et1mdcport, SSB_SPROM1_ETHPHY, SSB_SPROM1_ETHPHY_ET1M, 15);
+	SPEX(board_rev, SSB_SPROM1_BINF, SSB_SPROM1_BINF_BREV, 0);
 	SPEX(country_code, SSB_SPROM1_BINF, SSB_SPROM1_BINF_CCODE,
 	     SSB_SPROM1_BINF_CCODE_SHIFT);
+	SPEX(ant_available_a, SSB_SPROM1_BINF, SSB_SPROM1_BINF_ANTA,
+	     SSB_SPROM1_BINF_ANTA_SHIFT);
+	SPEX(ant_available_bg, SSB_SPROM1_BINF, SSB_SPROM1_BINF_ANTBG,
+	     SSB_SPROM1_BINF_ANTBG_SHIFT);
 	SPEX(pa0b0, SSB_SPROM1_PA0B0, 0xFFFF, 0);
 	SPEX(pa0b1, SSB_SPROM1_PA0B1, 0xFFFF, 0);
 	SPEX(pa0b2, SSB_SPROM1_PA0B2, 0xFFFF, 0);
@@ -348,9 +377,22 @@ static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 	     SSB_SPROM1_ITSSI_A_SHIFT);
 	SPEX(itssi_bg, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_BG, 0);
 	SPEX(boardflags_lo, SSB_SPROM1_BFLLO, 0xFFFF, 0);
-	SPEX(antenna_gain_a, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_A, 0);
-	SPEX(antenna_gain_bg, SSB_SPROM1_AGAIN, SSB_SPROM1_AGAIN_BG,
-	     SSB_SPROM1_AGAIN_BG_SHIFT);
+
+	/* Extract the antenna gain values. */
+	gain = r123_extract_antgain(out->revision, in,
+				    SSB_SPROM1_AGAIN_BG,
+				    SSB_SPROM1_AGAIN_BG_SHIFT);
+	out->antenna_gain.ghz24.a0 = gain;
+	out->antenna_gain.ghz24.a1 = gain;
+	out->antenna_gain.ghz24.a2 = gain;
+	out->antenna_gain.ghz24.a3 = gain;
+	gain = r123_extract_antgain(out->revision, in,
+				    SSB_SPROM1_AGAIN_A,
+				    SSB_SPROM1_AGAIN_A_SHIFT);
+	out->antenna_gain.ghz5.a0 = gain;
+	out->antenna_gain.ghz5.a1 = gain;
+	out->antenna_gain.ghz5.a2 = gain;
+	out->antenna_gain.ghz5.a3 = gain;
 }
 
 static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
@@ -376,9 +418,10 @@ static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 	     SSB_SPROM4_ETHPHY_ET1A_SHIFT);
 	SPEX(country_code, SSB_SPROM4_CCODE, 0xFFFF, 0);
 	SPEX(boardflags_lo, SSB_SPROM4_BFLLO, 0xFFFF, 0);
-	SPEX(antenna_gain_a, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_0, 0);
-	SPEX(antenna_gain_bg, SSB_SPROM4_AGAIN, SSB_SPROM4_AGAIN_1,
-	     SSB_SPROM4_AGAIN_1_SHIFT);
+	SPEX(ant_available_a, SSB_SPROM4_ANTAVAIL, SSB_SPROM4_ANTAVAIL_A,
+	     SSB_SPROM4_ANTAVAIL_A_SHIFT);
+	SPEX(ant_available_bg, SSB_SPROM4_ANTAVAIL, SSB_SPROM4_ANTAVAIL_BG,
+	     SSB_SPROM4_ANTAVAIL_BG_SHIFT);
 	SPEX(maxpwr_bg, SSB_SPROM4_MAXP_BG, SSB_SPROM4_MAXP_BG_MASK, 0);
 	SPEX(itssi_bg, SSB_SPROM4_MAXP_BG, SSB_SPROM4_ITSSI_BG,
 	     SSB_SPROM4_ITSSI_BG_SHIFT);
@@ -391,6 +434,19 @@ static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 	SPEX(gpio2, SSB_SPROM4_GPIOB, SSB_SPROM4_GPIOB_P2, 0);
 	SPEX(gpio3, SSB_SPROM4_GPIOB, SSB_SPROM4_GPIOB_P3,
 	     SSB_SPROM4_GPIOB_P3_SHIFT);
+
+	/* Extract the antenna gain values. */
+	SPEX(antenna_gain.ghz24.a0, SSB_SPROM4_AGAIN01,
+	     SSB_SPROM4_AGAIN0, SSB_SPROM4_AGAIN0_SHIFT);
+	SPEX(antenna_gain.ghz24.a1, SSB_SPROM4_AGAIN01,
+	     SSB_SPROM4_AGAIN1, SSB_SPROM4_AGAIN1_SHIFT);
+	SPEX(antenna_gain.ghz24.a2, SSB_SPROM4_AGAIN23,
+	     SSB_SPROM4_AGAIN2, SSB_SPROM4_AGAIN2_SHIFT);
+	SPEX(antenna_gain.ghz24.a3, SSB_SPROM4_AGAIN23,
+	     SSB_SPROM4_AGAIN3, SSB_SPROM4_AGAIN3_SHIFT);
+	memcpy(&out->antenna_gain.ghz5, &out->antenna_gain.ghz24,
+	       sizeof(out->antenna_gain.ghz5));
+
 	/* TODO - get remaining rev 4 stuff needed */
 }
 
@@ -400,7 +456,7 @@ static int sprom_extract(struct ssb_bus *bus, struct ssb_sprom *out,
 	memset(out, 0, sizeof(*out));
 
 	out->revision = in[size - 1] & 0x00FF;
-	ssb_printk(KERN_INFO PFX "SPROM revision %d detected.\n", out->revision);
+	ssb_dprintk(KERN_DEBUG PFX "SPROM revision %d detected.\n", out->revision);
 	if ((bus->chip_id & 0xFF00) == 0x4400) {
 		/* Workaround: The BCM44XX chip has a stupid revision
 		 * number stored in the SPROM.
@@ -445,9 +501,7 @@ static int ssb_pci_sprom_get(struct ssb_bus *bus,
 	err = sprom_check_crc(buf, bus->sprom_size);
 	if (err) {
 		/* check for rev 4 sprom - has special signature */
-		if (buf [32] == 0x5372) {
-			ssb_printk(KERN_WARNING PFX "Extracting a rev 4"
-				   " SPROM\n");
+		if (buf[32] == 0x5372) {
 			kfree(buf);
 			buf = kcalloc(SSB_SPROMSIZE_WORDS_R4, sizeof(u16),
 				      GFP_KERNEL);
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index a21ab29ff363..0eaa98424f0a 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -22,7 +22,12 @@ struct ssb_sprom {
 	u8 et1mac[6];		/* MAC address for 802.11a */
 	u8 et0phyaddr;		/* MII address for enet0 */
 	u8 et1phyaddr;		/* MII address for enet1 */
+	u8 et0mdcport;		/* MDIO for enet0 */
+	u8 et1mdcport;		/* MDIO for enet1 */
+	u8 board_rev;		/* Board revision number from SPROM. */
 	u8 country_code;	/* Country Code */
+	u8 ant_available_a;	/* A-PHY antenna available bits (up to 4) */
+	u8 ant_available_bg;	/* B/G-PHY antenna available bits (up to 4) */
 	u16 pa0b0;
 	u16 pa0b1;
 	u16 pa0b2;
@@ -38,8 +43,18 @@ struct ssb_sprom {
 	u8 itssi_a;		/* Idle TSSI Target for A-PHY */
 	u8 itssi_bg;		/* Idle TSSI Target for B/G-PHY */
 	u16 boardflags_lo;	/* Boardflags (low 16 bits) */
-	u8 antenna_gain_a;	/* A-PHY Antenna gain (in dBm Q5.2) */
-	u8 antenna_gain_bg;	/* B/G-PHY Antenna gain (in dBm Q5.2) */
+
+	/* Antenna gain values for up to 4 antennas
+	 * on each band. Values in dBm/4 (Q5.2). Negative gain means the
+	 * loss in the connectors is bigger than the gain. */
+	struct {
+		struct {
+			s8 a0, a1, a2, a3;
+		} ghz24;	/* 2.4GHz band */
+		struct {
+			s8 a0, a1, a2, a3;
+		} ghz5;		/* 5GHz band */
+	} antenna_gain;
 
 	/* TODO - add any parameters needed from rev 2, 3, or 4 SPROMs */
 };
diff --git a/include/linux/ssb/ssb_regs.h b/include/linux/ssb/ssb_regs.h
index 30222e89ad16..ebad0bac9801 100644
--- a/include/linux/ssb/ssb_regs.h
+++ b/include/linux/ssb/ssb_regs.h
@@ -193,10 +193,10 @@
 #define  SSB_SPROM1_BINF_BREV		0x00FF	/* Board Revision */
 #define  SSB_SPROM1_BINF_CCODE		0x0F00	/* Country Code */
 #define  SSB_SPROM1_BINF_CCODE_SHIFT	8
-#define  SSB_SPROM1_BINF_ANTA		0x3000	/* Available A-PHY antennas */
-#define  SSB_SPROM1_BINF_ANTA_SHIFT	12
-#define  SSB_SPROM1_BINF_ANTBG		0xC000	/* Available B-PHY antennas */
-#define  SSB_SPROM1_BINF_ANTBG_SHIFT	14
+#define  SSB_SPROM1_BINF_ANTBG		0x3000	/* Available B-PHY and G-PHY antennas */
+#define  SSB_SPROM1_BINF_ANTBG_SHIFT	12
+#define  SSB_SPROM1_BINF_ANTA		0xC000	/* Available A-PHY antennas */
+#define  SSB_SPROM1_BINF_ANTA_SHIFT	14
 #define SSB_SPROM1_PA0B0		0x105E
 #define SSB_SPROM1_PA0B1		0x1060
 #define SSB_SPROM1_PA0B2		0x1062
@@ -221,9 +221,10 @@
 #define  SSB_SPROM1_ITSSI_A_SHIFT	8
 #define SSB_SPROM1_BFLLO		0x1072	/* Boardflags (low 16 bits) */
 #define SSB_SPROM1_AGAIN		0x1074	/* Antenna Gain (in dBm Q5.2) */
-#define  SSB_SPROM1_AGAIN_A		0x00FF	/* A-PHY */
-#define  SSB_SPROM1_AGAIN_BG		0xFF00	/* B-PHY and G-PHY */
-#define  SSB_SPROM1_AGAIN_BG_SHIFT	8
+#define  SSB_SPROM1_AGAIN_BG		0x00FF	/* B-PHY and G-PHY */
+#define  SSB_SPROM1_AGAIN_BG_SHIFT	0
+#define  SSB_SPROM1_AGAIN_A		0xFF00	/* A-PHY */
+#define  SSB_SPROM1_AGAIN_A_SHIFT	8
 
 /* SPROM Revision 2 (inherits from rev 1) */
 #define SSB_SPROM2_BFLHI		0x1038	/* Boardflags (high 16 bits) */
@@ -264,7 +265,7 @@
 #define  SSB_SPROM3_CCKPO_11M_SHIFT	12
 #define  SSB_SPROM3_OFDMGPO		0x107A	/* G-PHY OFDM Power Offset (4 bytes, BigEndian) */
 
-/* SPROM Revision 4 		entries with ?? in comment are unknown */
+/* SPROM Revision 4 */
 #define SSB_SPROM4_IL0MAC		0x104C	/* 6 byte MAC address for a/b/g/n */
 #define SSB_SPROM4_ET0MAC		0x1018	/* 6 bytes MAC address for Ethernet ?? */
 #define SSB_SPROM4_ET1MAC		0x1018	/* 6 bytes MAC address for 802.11a ?? */
@@ -275,13 +276,22 @@
 #define  SSB_SPROM4_ETHPHY_ET0M		(1<<14)	/* MDIO for enet0 */
 #define  SSB_SPROM4_ETHPHY_ET1M		(1<<15)	/* MDIO for enet1 */
 #define SSB_SPROM4_CCODE		0x1052	/* Country Code (2 bytes) */
-#define SSB_SPROM4_ANT_A		0x105D  /* A Antennas */
-#define SSB_SPROM4_ANT_BG		0x105C  /* B/G Antennas */
+#define SSB_SPROM4_ANTAVAIL		0x105D  /* Antenna available bitfields */
+#define SSB_SPROM4_ANTAVAIL_A		0x00FF	/* A-PHY bitfield */
+#define SSB_SPROM4_ANTAVAIL_A_SHIFT	0
+#define SSB_SPROM4_ANTAVAIL_BG		0xFF00	/* B-PHY and G-PHY bitfield */
+#define SSB_SPROM4_ANTAVAIL_BG_SHIFT	8
 #define SSB_SPROM4_BFLLO		0x1044	/* Boardflags (low 16 bits) */
-#define SSB_SPROM4_AGAIN		0x105E	/* Antenna Gain (in dBm Q5.2) */
-#define  SSB_SPROM4_AGAIN_0		0x00FF	/* Antenna 0 */
-#define  SSB_SPROM4_AGAIN_1		0xFF00	/* Antenna 1 */
-#define  SSB_SPROM4_AGAIN_1_SHIFT	8
+#define SSB_SPROM4_AGAIN01		0x105E	/* Antenna Gain (in dBm Q5.2) */
+#define  SSB_SPROM4_AGAIN0		0x00FF	/* Antenna 0 */
+#define  SSB_SPROM4_AGAIN0_SHIFT	0
+#define  SSB_SPROM4_AGAIN1		0xFF00	/* Antenna 1 */
+#define  SSB_SPROM4_AGAIN1_SHIFT	8
+#define SSB_SPROM4_AGAIN23		0x1060
+#define  SSB_SPROM4_AGAIN2		0x00FF	/* Antenna 2 */
+#define  SSB_SPROM4_AGAIN2_SHIFT	0
+#define  SSB_SPROM4_AGAIN3		0xFF00	/* Antenna 3 */
+#define  SSB_SPROM4_AGAIN3_SHIFT	8
 #define SSB_SPROM4_BFLHI		0x1046  /* Board Flags Hi */
 #define SSB_SPROM4_MAXP_BG		0x1080  /* Max Power BG in path 1 */
 #define  SSB_SPROM4_MAXP_BG_MASK	0x00FF  /* Mask for Max Power BG */
-- 
cgit v1.2.3


From 993e1c780b323736a2cdc24564f35e80ce8d3337 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Sat, 22 Dec 2007 22:01:36 +0100
Subject: ssb: Fix PCMCIA lowlevel register access

This fixes lowlevel register access for PCMCIA based devices.

The patch also adds a temporary workaround for the device mac address.
It simply adds generation of a random address. The real SPROM extraction
will follow in another patch.
The temporary workaround will be removed then, but for now it's OK.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/pcmcia.c    | 71 ++++++++++++++++++++++++++++++-------------------
 include/linux/ssb/ssb.h |  3 ++-
 2 files changed, 45 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index bb44a76b3eb5..46816cda8b98 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -94,7 +94,6 @@ int ssb_pcmcia_switch_core(struct ssb_bus *bus,
 			   struct ssb_device *dev)
 {
 	int err;
-	unsigned long flags;
 
 #if SSB_VERBOSE_PCMCIACORESWITCH_DEBUG
 	ssb_printk(KERN_INFO PFX
@@ -103,11 +102,9 @@ int ssb_pcmcia_switch_core(struct ssb_bus *bus,
 		   dev->core_index);
 #endif
 
-	spin_lock_irqsave(&bus->bar_lock, flags);
 	err = ssb_pcmcia_switch_coreidx(bus, dev->core_index);
 	if (!err)
 		bus->mapped_device = dev;
-	spin_unlock_irqrestore(&bus->bar_lock, flags);
 
 	return err;
 }
@@ -115,14 +112,12 @@ int ssb_pcmcia_switch_core(struct ssb_bus *bus,
 int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg)
 {
 	int attempts = 0;
-	unsigned long flags;
 	conf_reg_t reg;
-	int res, err = 0;
+	int res;
 
 	SSB_WARN_ON((seg != 0) && (seg != 1));
 	reg.Offset = 0x34;
 	reg.Function = 0;
-	spin_lock_irqsave(&bus->bar_lock, flags);
 	while (1) {
 		reg.Action = CS_WRITE;
 		reg.Value = seg;
@@ -143,13 +138,11 @@ int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg)
 		udelay(10);
 	}
 	bus->mapped_pcmcia_seg = seg;
-out_unlock:
-	spin_unlock_irqrestore(&bus->bar_lock, flags);
-	return err;
+
+	return 0;
 error:
 	ssb_printk(KERN_ERR PFX "Failed to switch pcmcia segment\n");
-	err = -ENODEV;
-	goto out_unlock;
+	return -ENODEV;
 }
 
 static int select_core_and_segment(struct ssb_device *dev,
@@ -182,22 +175,33 @@ static int select_core_and_segment(struct ssb_device *dev,
 static u16 ssb_pcmcia_read16(struct ssb_device *dev, u16 offset)
 {
 	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	int err;
+	u16 value = 0xFFFF;
 
-	if (unlikely(select_core_and_segment(dev, &offset)))
-		return 0xFFFF;
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err))
+		value = readw(bus->mmio + offset);
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
 
-	return readw(bus->mmio + offset);
+	return value;
 }
 
 static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset)
 {
 	struct ssb_bus *bus = dev->bus;
-	u32 lo, hi;
+	unsigned long flags;
+	int err;
+	u32 lo = 0xFFFFFFFF, hi = 0xFFFFFFFF;
 
-	if (unlikely(select_core_and_segment(dev, &offset)))
-		return 0xFFFFFFFF;
-	lo = readw(bus->mmio + offset);
-	hi = readw(bus->mmio + offset + 2);
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err)) {
+		lo = readw(bus->mmio + offset);
+		hi = readw(bus->mmio + offset + 2);
+	}
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
 
 	return (lo | (hi << 16));
 }
@@ -205,22 +209,31 @@ static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset)
 static void ssb_pcmcia_write16(struct ssb_device *dev, u16 offset, u16 value)
 {
 	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	int err;
 
-	if (unlikely(select_core_and_segment(dev, &offset)))
-		return;
-	writew(value, bus->mmio + offset);
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err))
+		writew(value, bus->mmio + offset);
+	mmiowb();
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
 }
 
 static void ssb_pcmcia_write32(struct ssb_device *dev, u16 offset, u32 value)
 {
 	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	int err;
 
-	if (unlikely(select_core_and_segment(dev, &offset)))
-		return;
-	writeb((value & 0xFF000000) >> 24, bus->mmio + offset + 3);
-	writeb((value & 0x00FF0000) >> 16, bus->mmio + offset + 2);
-	writeb((value & 0x0000FF00) >> 8, bus->mmio + offset + 1);
-	writeb((value & 0x000000FF) >> 0, bus->mmio + offset + 0);
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err)) {
+		writew((value & 0x0000FFFF), bus->mmio + offset);
+		writew(((value & 0xFFFF0000) >> 16), bus->mmio + offset + 2);
+	}
+	mmiowb();
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
 }
 
 /* Not "static", as it's used in main.c */
@@ -231,10 +244,12 @@ const struct ssb_bus_ops ssb_pcmcia_ops = {
 	.write32	= ssb_pcmcia_write32,
 };
 
+#include <linux/etherdevice.h>
 int ssb_pcmcia_get_invariants(struct ssb_bus *bus,
 			      struct ssb_init_invariants *iv)
 {
 	//TODO
+	random_ether_addr(iv->sprom.il0mac);
 	return 0;
 }
 
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 0eaa98424f0a..cacbae531945 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -231,7 +231,8 @@ struct ssb_bus {
 	struct ssb_device *mapped_device;
 	/* Currently mapped PCMCIA segment. (bustype == SSB_BUSTYPE_PCMCIA only) */
 	u8 mapped_pcmcia_seg;
-	/* Lock for core and segment switching. */
+	/* Lock for core and segment switching.
+	 * On PCMCIA-host busses this is used to protect the whole MMIO access. */
 	spinlock_t bar_lock;
 
 	/* The bus this backplane is running on. */
-- 
cgit v1.2.3


From 961d57c883198831503c7be5c088a26101dfb16c Mon Sep 17 00:00:00 2001
From: Miguel Botón <mboton.lkml@gmail.com>
Date: Tue, 1 Jan 2008 01:16:46 +0100
Subject: ssb: add 'ssb_pcihost_set_power_state' function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds the 'ssb_pcihost_set_power_state' function.

This function allows us to set the power state of a PCI device
(for example b44 ethernet device).

Signed-off-by: Miguel Botón <mboton@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ssb/ssb.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index cacbae531945..1ab4688c678b 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -365,6 +365,13 @@ static inline void ssb_pcihost_unregister(struct pci_driver *driver)
 {
 	pci_unregister_driver(driver);
 }
+
+static inline
+void ssb_pcihost_set_power_state(struct ssb_device *sdev, pci_power_t state)
+{
+	if (sdev->bus->bustype == SSB_BUSTYPE_PCI)
+		pci_set_power_state(sdev->bus->host_pci, state);
+}
 #endif /* CONFIG_SSB_PCIHOST */
 
 
-- 
cgit v1.2.3


From f653211197f3841f383fa9757ef8ce182c6cf627 Mon Sep 17 00:00:00 2001
From: Michael Wu <flamingice@sourmilk.net>
Date: Sun, 14 Oct 2007 14:43:16 -0400
Subject: Add rtl8180 wireless driver

This patch adds a mac80211 based wireless driver for the rtl8180 and
rtl8185 PCI wireless cards.  Also included are some rtl8187 changes
required due to the relationship between that driver and this one.

Michael Wu is primarily responsible for the initial driver and rtl8185
support.  Andreas Merello provided the additional rtl8180 support.

Thanks to Jukka Ruohonen for the donating a rtl8185 card! It was very
helpful for the rtl8225z2 code.

The Signed-off-by information below is collected from the individual
patches submitted to wireless-2.6 before merging this driver upstream.

Signed-off-by: Andrea Merello <andreamrl@tiscali.it>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Pavel Roskin <proski@gnu.org>
Signed-off-by: Michael Wu <flamingice@sourmilk.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/Kconfig           |   56 ++
 drivers/net/wireless/Makefile          |    3 +
 drivers/net/wireless/rtl8180.h         |  151 +++++
 drivers/net/wireless/rtl8180_dev.c     | 1048 ++++++++++++++++++++++++++++++++
 drivers/net/wireless/rtl8180_grf5101.c |  179 ++++++
 drivers/net/wireless/rtl8180_grf5101.h |   28 +
 drivers/net/wireless/rtl8180_max2820.c |  150 +++++
 drivers/net/wireless/rtl8180_max2820.h |   28 +
 drivers/net/wireless/rtl8180_rtl8225.c |  779 ++++++++++++++++++++++++
 drivers/net/wireless/rtl8180_rtl8225.h |   23 +
 drivers/net/wireless/rtl8180_sa2400.c  |  201 ++++++
 drivers/net/wireless/rtl8180_sa2400.h  |   36 ++
 drivers/net/wireless/rtl8187.h         |    2 +-
 drivers/net/wireless/rtl8187_dev.c     |   60 +-
 drivers/net/wireless/rtl8187_rtl8225.c |   52 +-
 drivers/net/wireless/rtl8187_rtl8225.h |    9 +-
 drivers/net/wireless/rtl818x.h         |   29 +-
 include/linux/pci_ids.h                |    3 +
 18 files changed, 2777 insertions(+), 60 deletions(-)
 create mode 100644 drivers/net/wireless/rtl8180.h
 create mode 100644 drivers/net/wireless/rtl8180_dev.c
 create mode 100644 drivers/net/wireless/rtl8180_grf5101.c
 create mode 100644 drivers/net/wireless/rtl8180_grf5101.h
 create mode 100644 drivers/net/wireless/rtl8180_max2820.c
 create mode 100644 drivers/net/wireless/rtl8180_max2820.h
 create mode 100644 drivers/net/wireless/rtl8180_rtl8225.c
 create mode 100644 drivers/net/wireless/rtl8180_rtl8225.h
 create mode 100644 drivers/net/wireless/rtl8180_sa2400.c
 create mode 100644 drivers/net/wireless/rtl8180_sa2400.h

(limited to 'include')

diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index 50a8b6c2fa00..f372960904b2 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -545,6 +545,62 @@ config USB_ZD1201
 	  To compile this driver as a module, choose M here: the
 	  module will be called zd1201.
 
+config RTL8180
+	tristate "Realtek 8180/8185 PCI support"
+	depends on MAC80211 && PCI && WLAN_80211 && EXPERIMENTAL
+	select EEPROM_93CX6
+	---help---
+	  This is a driver for RTL8180 and RTL8185 based cards.
+	  These are PCI based chips found in cards such as:
+
+	  (RTL8185 802.11g)
+	  A-Link WL54PC
+
+	  (RTL8180 802.11b)
+	  Belkin F5D6020 v3
+	  Belkin F5D6020 v3
+	  Dlink DWL-610
+	  Dlink DWL-510
+	  Netgear MA521
+	  Level-One WPC-0101
+	  Acer Aspire 1357 LMi
+	  VCTnet PC-11B1
+	  Ovislink AirLive WL-1120PCM
+	  Mentor WL-PCI
+	  Linksys WPC11 v4
+	  TrendNET TEW-288PI
+	  D-Link DWL-520 Rev D
+	  Repotec RP-WP7126
+	  TP-Link TL-WN250/251
+	  Zonet ZEW1000
+	  Longshine LCS-8031-R
+	  HomeLine HLW-PCC200
+	  GigaFast WF721-AEX
+	  Planet WL-3553
+	  Encore ENLWI-PCI1-NT
+	  TrendNET TEW-266PC
+	  Gigabyte GN-WLMR101
+	  Siemens-fujitsu Amilo D1840W
+	  Edimax EW-7126
+	  PheeNet WL-11PCIR
+	  Tonze PC-2100T
+	  Planet WL-8303
+	  Dlink DWL-650 v M1
+	  Edimax EW-7106
+	  Q-Tec 770WC
+	  Topcom Skyr@cer 4011b
+	  Roper FreeLan 802.11b (edition 2004)
+	  Wistron Neweb Corp CB-200B
+	  Pentagram HorNET
+	  QTec 775WC
+	  TwinMOS Booming B Series
+	  Micronet SP906BB
+	  Sweex LC700010
+	  Surecom EP-9428
+	  Safecom SWLCR-1100
+
+	  Thanks to Realtek for their support!
+
 config RTL8187
 	tristate "Realtek 8187 USB support"
 	depends on MAC80211 && USB && WLAN_80211 && EXPERIMENTAL
diff --git a/drivers/net/wireless/Makefile b/drivers/net/wireless/Makefile
index 7e1535ece17a..6af7b158624e 100644
--- a/drivers/net/wireless/Makefile
+++ b/drivers/net/wireless/Makefile
@@ -47,7 +47,10 @@ obj-$(CONFIG_PCMCIA_WL3501)	+= wl3501_cs.o
 obj-$(CONFIG_USB_ZD1201)	+= zd1201.o
 obj-$(CONFIG_LIBERTAS)		+= libertas/
 
+rtl8180-objs		:= rtl8180_dev.o rtl8180_rtl8225.o rtl8180_sa2400.o rtl8180_max2820.o rtl8180_grf5101.o
 rtl8187-objs		:= rtl8187_dev.o rtl8187_rtl8225.o
+
+obj-$(CONFIG_RTL8180)	+= rtl8180.o
 obj-$(CONFIG_RTL8187)	+= rtl8187.o
 
 obj-$(CONFIG_ADM8211)	+= adm8211.o
diff --git a/drivers/net/wireless/rtl8180.h b/drivers/net/wireless/rtl8180.h
new file mode 100644
index 000000000000..0b333cacbd85
--- /dev/null
+++ b/drivers/net/wireless/rtl8180.h
@@ -0,0 +1,151 @@
+#ifndef RTL8180_H
+#define RTL8180_H
+
+#include "rtl818x.h"
+
+#define MAX_RX_SIZE IEEE80211_MAX_RTS_THRESHOLD
+
+#define RF_PARAM_ANALOGPHY	(1 << 0)
+#define RF_PARAM_ANTBDEFAULT	(1 << 1)
+#define RF_PARAM_CARRIERSENSE1	(1 << 2)
+#define RF_PARAM_CARRIERSENSE2	(1 << 3)
+
+#define BB_ANTATTEN_CHAN14	0x0C
+#define BB_ANTENNA_B 		0x40
+
+#define BB_HOST_BANG 		(1 << 30)
+#define BB_HOST_BANG_EN 	(1 << 2)
+#define BB_HOST_BANG_CLK 	(1 << 1)
+#define BB_HOST_BANG_DATA	1
+
+#define ANAPARAM_TXDACOFF_SHIFT	27
+#define ANAPARAM_PWR0_SHIFT	28
+#define ANAPARAM_PWR0_MASK 	(0x07 << ANAPARAM_PWR0_SHIFT)
+#define ANAPARAM_PWR1_SHIFT	20
+#define ANAPARAM_PWR1_MASK	(0x7F << ANAPARAM_PWR1_SHIFT)
+
+enum rtl8180_tx_desc_flags {
+	RTL8180_TX_DESC_FLAG_NO_ENC	= (1 << 15),
+	RTL8180_TX_DESC_FLAG_TX_OK	= (1 << 15),
+	RTL8180_TX_DESC_FLAG_SPLCP	= (1 << 16),
+	RTL8180_TX_DESC_FLAG_RX_UNDER	= (1 << 16),
+	RTL8180_TX_DESC_FLAG_MOREFRAG	= (1 << 17),
+	RTL8180_TX_DESC_FLAG_CTS	= (1 << 18),
+	RTL8180_TX_DESC_FLAG_RTS	= (1 << 23),
+	RTL8180_TX_DESC_FLAG_LS		= (1 << 28),
+	RTL8180_TX_DESC_FLAG_FS		= (1 << 29),
+	RTL8180_TX_DESC_FLAG_DMA	= (1 << 30),
+	RTL8180_TX_DESC_FLAG_OWN	= (1 << 31)
+};
+
+struct rtl8180_tx_desc {
+	__le32 flags;
+	__le16 rts_duration;
+	__le16 plcp_len;
+	__le32 tx_buf;
+	__le32 frame_len;
+	__le32 next_tx_desc;
+	u8 cw;
+	u8 retry_limit;
+	u8 agc;
+	u8 flags2;
+	u32 reserved[2];
+} __attribute__ ((packed));
+
+enum rtl8180_rx_desc_flags {
+	RTL8180_RX_DESC_FLAG_ICV_ERR	= (1 << 12),
+	RTL8180_RX_DESC_FLAG_CRC32_ERR	= (1 << 13),
+	RTL8180_RX_DESC_FLAG_PM		= (1 << 14),
+	RTL8180_RX_DESC_FLAG_RX_ERR	= (1 << 15),
+	RTL8180_RX_DESC_FLAG_BCAST	= (1 << 16),
+	RTL8180_RX_DESC_FLAG_PAM	= (1 << 17),
+	RTL8180_RX_DESC_FLAG_MCAST	= (1 << 18),
+	RTL8180_RX_DESC_FLAG_SPLCP	= (1 << 25),
+	RTL8180_RX_DESC_FLAG_FOF	= (1 << 26),
+	RTL8180_RX_DESC_FLAG_DMA_FAIL	= (1 << 27),
+	RTL8180_RX_DESC_FLAG_LS		= (1 << 28),
+	RTL8180_RX_DESC_FLAG_FS		= (1 << 29),
+	RTL8180_RX_DESC_FLAG_EOR	= (1 << 30),
+	RTL8180_RX_DESC_FLAG_OWN	= (1 << 31)
+};
+
+struct rtl8180_rx_desc {
+	__le32 flags;
+	__le32 flags2;
+	union {
+		__le32 rx_buf;
+		__le64 tsft;
+	};
+} __attribute__ ((packed));
+
+struct rtl8180_tx_ring {
+	struct rtl8180_tx_desc *desc;
+	dma_addr_t dma;
+	unsigned int idx;
+	unsigned int entries;
+	struct sk_buff_head queue;
+};
+
+struct rtl8180_priv {
+	/* common between rtl818x drivers */
+	struct rtl818x_csr __iomem *map;
+	const struct rtl818x_rf_ops *rf;
+	int mode;
+	int if_id;
+
+	/* rtl8180 driver specific */
+	spinlock_t lock;
+	struct rtl8180_rx_desc *rx_ring;
+	dma_addr_t rx_ring_dma;
+	unsigned int rx_idx;
+	struct sk_buff *rx_buf[32];
+	struct rtl8180_tx_ring tx_ring[4];
+	struct ieee80211_channel channels[14];
+	struct ieee80211_rate rates[12];
+	struct ieee80211_hw_mode modes[2];
+	struct pci_dev *pdev;
+	u32 rx_conf;
+
+	int r8185;
+	u32 anaparam;
+	u16 rfparam;
+	u8 csthreshold;
+};
+
+void rtl8180_write_phy(struct ieee80211_hw *dev, u8 addr, u32 data);
+void rtl8180_set_anaparam(struct rtl8180_priv *priv, u32 anaparam);
+
+static inline u8 rtl818x_ioread8(struct rtl8180_priv *priv, u8 __iomem *addr)
+{
+	return ioread8(addr);
+}
+
+static inline u16 rtl818x_ioread16(struct rtl8180_priv *priv, __le16 __iomem *addr)
+{
+	return ioread16(addr);
+}
+
+static inline u32 rtl818x_ioread32(struct rtl8180_priv *priv, __le32 __iomem *addr)
+{
+	return ioread32(addr);
+}
+
+static inline void rtl818x_iowrite8(struct rtl8180_priv *priv,
+				    u8 __iomem *addr, u8 val)
+{
+	iowrite8(val, addr);
+}
+
+static inline void rtl818x_iowrite16(struct rtl8180_priv *priv,
+				     __le16 __iomem *addr, u16 val)
+{
+	iowrite16(val, addr);
+}
+
+static inline void rtl818x_iowrite32(struct rtl8180_priv *priv,
+				     __le32 __iomem *addr, u32 val)
+{
+	iowrite32(val, addr);
+}
+
+#endif /* RTL8180_H */
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
new file mode 100644
index 000000000000..4b7b032c194a
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -0,0 +1,1048 @@
+
+/*
+ * Linux device driver for RTL8180 / RTL8185
+ *
+ * Copyright 2007 Michael Wu <flamingice@sourmilk.net>
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Based on the r8180 driver, which is:
+ * Copyright 2004-2005 Andrea Merello <andreamrl@tiscali.it>, et al.
+ *
+ * Thanks to Realtek for their support!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/eeprom_93cx6.h>
+#include <net/mac80211.h>
+
+#include "rtl8180.h"
+#include "rtl8180_rtl8225.h"
+#include "rtl8180_sa2400.h"
+#include "rtl8180_max2820.h"
+#include "rtl8180_grf5101.h"
+
+MODULE_AUTHOR("Michael Wu <flamingice@sourmilk.net>");
+MODULE_AUTHOR("Andrea Merello <andreamrl@tiscali.it>");
+MODULE_DESCRIPTION("RTL8180 / RTL8185 PCI wireless driver");
+MODULE_LICENSE("GPL");
+
+static struct pci_device_id rtl8180_table[] __devinitdata = {
+	/* rtl8185 */
+	{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8185) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_BELKIN, 0x701f) },
+
+	/* rtl8180 */
+	{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8180) },
+	{ PCI_DEVICE(0x1799, 0x6001) },
+	{ PCI_DEVICE(0x1799, 0x6020) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_DLINK, 0x3300) },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(pci, rtl8180_table);
+
+void rtl8180_write_phy(struct ieee80211_hw *dev, u8 addr, u32 data)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int i = 10;
+	u32 buf;
+
+	buf = (data << 8) | addr;
+
+	rtl818x_iowrite32(priv, (__le32 __iomem *)&priv->map->PHY[0], buf | 0x80);
+	while (i--) {
+		rtl818x_iowrite32(priv, (__le32 __iomem *)&priv->map->PHY[0], buf);
+		if (rtl818x_ioread8(priv, &priv->map->PHY[2]) == (data & 0xFF))
+			return;
+	}
+}
+
+static void rtl8180_handle_rx(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	unsigned int count = 32;
+
+	while (count--) {
+		struct rtl8180_rx_desc *entry = &priv->rx_ring[priv->rx_idx];
+		struct sk_buff *skb = priv->rx_buf[priv->rx_idx];
+		u32 flags = le32_to_cpu(entry->flags);
+
+		if (flags & RTL8180_RX_DESC_FLAG_OWN)
+			return;
+
+		if (unlikely(flags & (RTL8180_RX_DESC_FLAG_DMA_FAIL |
+				      RTL8180_RX_DESC_FLAG_FOF |
+				      RTL8180_RX_DESC_FLAG_RX_ERR)))
+			goto done;
+		else {
+			u32 flags2 = le32_to_cpu(entry->flags2);
+			struct ieee80211_rx_status rx_status = {0};
+			struct sk_buff *new_skb = dev_alloc_skb(MAX_RX_SIZE);
+
+			if (unlikely(!new_skb))
+				goto done;
+
+			pci_unmap_single(priv->pdev,
+					 *((dma_addr_t *)skb->cb),
+					 MAX_RX_SIZE, PCI_DMA_FROMDEVICE);
+			skb_put(skb, flags & 0xFFF);
+
+			rx_status.antenna = (flags2 >> 15) & 1;
+			/* TODO: improve signal/rssi reporting */
+			rx_status.signal = flags2 & 0xFF;
+			rx_status.ssi = (flags2 >> 8) & 0x7F;
+			rx_status.rate = (flags >> 20) & 0xF;
+			rx_status.freq = dev->conf.freq;
+			rx_status.channel = dev->conf.channel;
+			rx_status.phymode = dev->conf.phymode;
+			rx_status.mactime = le64_to_cpu(entry->tsft);
+			rx_status.flag |= RX_FLAG_TSFT;
+			if (flags & RTL8180_RX_DESC_FLAG_CRC32_ERR)
+				rx_status.flag |= RX_FLAG_FAILED_FCS_CRC;
+
+			ieee80211_rx_irqsafe(dev, skb, &rx_status);
+
+			skb = new_skb;
+			priv->rx_buf[priv->rx_idx] = skb;
+			*((dma_addr_t *) skb->cb) =
+				pci_map_single(priv->pdev, skb_tail_pointer(skb),
+					       MAX_RX_SIZE, PCI_DMA_FROMDEVICE);
+		}
+
+	done:
+		entry->rx_buf = cpu_to_le32(*((dma_addr_t *)skb->cb));
+		entry->flags = cpu_to_le32(RTL8180_RX_DESC_FLAG_OWN |
+					   MAX_RX_SIZE);
+		if (priv->rx_idx == 31)
+			entry->flags |= cpu_to_le32(RTL8180_RX_DESC_FLAG_EOR);
+		priv->rx_idx = (priv->rx_idx + 1) % 32;
+	}
+}
+
+static void rtl8180_handle_tx(struct ieee80211_hw *dev, unsigned int prio)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	struct rtl8180_tx_ring *ring = &priv->tx_ring[prio];
+
+	while (skb_queue_len(&ring->queue)) {
+		struct rtl8180_tx_desc *entry = &ring->desc[ring->idx];
+		struct sk_buff *skb;
+		struct ieee80211_tx_status status = { {0} };
+		struct ieee80211_tx_control *control;
+		u32 flags = le32_to_cpu(entry->flags);
+
+		if (flags & RTL8180_TX_DESC_FLAG_OWN)
+			return;
+
+		ring->idx = (ring->idx + 1) % ring->entries;
+		skb = __skb_dequeue(&ring->queue);
+		pci_unmap_single(priv->pdev, le32_to_cpu(entry->tx_buf),
+				 skb->len, PCI_DMA_TODEVICE);
+
+		control = *((struct ieee80211_tx_control **)skb->cb);
+		if (control)
+			memcpy(&status.control, control, sizeof(*control));
+		kfree(control);
+
+		if (!(status.control.flags & IEEE80211_TXCTL_NO_ACK)) {
+			if (flags & RTL8180_TX_DESC_FLAG_TX_OK)
+				status.flags = IEEE80211_TX_STATUS_ACK;
+			else
+				status.excessive_retries = 1;
+		}
+		status.retry_count = flags & 0xFF;
+
+		ieee80211_tx_status_irqsafe(dev, skb, &status);
+		if (ring->entries - skb_queue_len(&ring->queue) == 2)
+			ieee80211_wake_queue(dev, prio);
+	}
+}
+
+static irqreturn_t rtl8180_interrupt(int irq, void *dev_id)
+{
+	struct ieee80211_hw *dev = dev_id;
+	struct rtl8180_priv *priv = dev->priv;
+	u16 reg;
+
+	spin_lock(&priv->lock);
+	reg = rtl818x_ioread16(priv, &priv->map->INT_STATUS);
+	if (unlikely(reg == 0xFFFF)) {
+		spin_unlock(&priv->lock);
+		return IRQ_HANDLED;
+	}
+
+	rtl818x_iowrite16(priv, &priv->map->INT_STATUS, reg);
+
+	if (reg & (RTL818X_INT_TXB_OK | RTL818X_INT_TXB_ERR))
+		rtl8180_handle_tx(dev, 3);
+
+	if (reg & (RTL818X_INT_TXH_OK | RTL818X_INT_TXH_ERR))
+		rtl8180_handle_tx(dev, 2);
+
+	if (reg & (RTL818X_INT_TXN_OK | RTL818X_INT_TXN_ERR))
+		rtl8180_handle_tx(dev, 1);
+
+	if (reg & (RTL818X_INT_TXL_OK | RTL818X_INT_TXL_ERR))
+		rtl8180_handle_tx(dev, 0);
+
+	if (reg & (RTL818X_INT_RX_OK | RTL818X_INT_RX_ERR))
+		rtl8180_handle_rx(dev);
+
+	spin_unlock(&priv->lock);
+
+	return IRQ_HANDLED;
+}
+
+static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
+		      struct ieee80211_tx_control *control)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	struct rtl8180_tx_ring *ring;
+	struct rtl8180_tx_desc *entry;
+	unsigned long flags;
+	unsigned int idx, prio;
+	dma_addr_t mapping;
+	u32 tx_flags;
+	u16 plcp_len = 0;
+	__le16 rts_duration = 0;
+
+	prio = control->queue;
+	ring = &priv->tx_ring[prio];
+
+	mapping = pci_map_single(priv->pdev, skb->data,
+				 skb->len, PCI_DMA_TODEVICE);
+
+	tx_flags = RTL8180_TX_DESC_FLAG_OWN | RTL8180_TX_DESC_FLAG_FS |
+		   RTL8180_TX_DESC_FLAG_LS | (control->tx_rate << 24) |
+		   (control->rts_cts_rate << 19) | skb->len;
+
+	if (priv->r8185)
+		tx_flags |= RTL8180_TX_DESC_FLAG_DMA |
+			    RTL8180_TX_DESC_FLAG_NO_ENC;
+
+	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+		tx_flags |= RTL8180_TX_DESC_FLAG_RTS;
+	else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
+		tx_flags |= RTL8180_TX_DESC_FLAG_CTS;
+
+	*((struct ieee80211_tx_control **) skb->cb) =
+		kmemdup(control, sizeof(*control), GFP_ATOMIC);
+
+	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+		rts_duration = ieee80211_rts_duration(dev, priv->if_id, skb->len, control);
+
+	if (!priv->r8185) {
+		unsigned int remainder;
+
+		plcp_len = DIV_ROUND_UP(16 * (skb->len + 4),
+					(control->rate->rate * 2) / 10);
+		remainder = (16 * (skb->len + 4)) %
+			    ((control->rate->rate * 2) / 10);
+		if (remainder > 0 && remainder <= 6)
+			plcp_len |= 1 << 15;
+	}
+
+	spin_lock_irqsave(&priv->lock, flags);
+	idx = (ring->idx + skb_queue_len(&ring->queue)) % ring->entries;
+	entry = &ring->desc[idx];
+
+	entry->rts_duration = rts_duration;
+	entry->plcp_len = cpu_to_le16(plcp_len);
+	entry->tx_buf = cpu_to_le32(mapping);
+	entry->frame_len = cpu_to_le32(skb->len);
+	entry->flags2 = control->alt_retry_rate != -1 ?
+			control->alt_retry_rate << 4 : 0;
+	entry->retry_limit = control->retry_limit;
+	entry->flags = cpu_to_le32(tx_flags);
+	__skb_queue_tail(&ring->queue, skb);
+	if (ring->entries - skb_queue_len(&ring->queue) < 2)
+		ieee80211_stop_queue(dev, control->queue);
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	rtl818x_iowrite8(priv, &priv->map->TX_DMA_POLLING, (1 << (prio + 4)));
+
+	return 0;
+}
+
+void rtl8180_set_anaparam(struct rtl8180_priv *priv, u32 anaparam)
+{
+	u8 reg;
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	reg = rtl818x_ioread8(priv, &priv->map->CONFIG3);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3,
+		 reg | RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite32(priv, &priv->map->ANAPARAM, anaparam);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3,
+		 reg & ~RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+}
+
+static int rtl8180_init_hw(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u16 reg;
+
+	rtl818x_iowrite8(priv, &priv->map->CMD, 0);
+	rtl818x_ioread8(priv, &priv->map->CMD);
+	msleep(10);
+
+	/* reset */
+	rtl818x_iowrite16(priv, &priv->map->INT_MASK, 0);
+	rtl818x_ioread8(priv, &priv->map->CMD);
+
+	reg = rtl818x_ioread8(priv, &priv->map->CMD);
+	reg &= (1 << 1);
+	reg |= RTL818X_CMD_RESET;
+	rtl818x_iowrite8(priv, &priv->map->CMD, RTL818X_CMD_RESET);
+	rtl818x_ioread8(priv, &priv->map->CMD);
+	msleep(200);
+
+	/* check success of reset */
+	if (rtl818x_ioread8(priv, &priv->map->CMD) & RTL818X_CMD_RESET) {
+		printk(KERN_ERR "%s: reset timeout!\n", wiphy_name(dev->wiphy));
+		return -ETIMEDOUT;
+	}
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_LOAD);
+	rtl818x_ioread8(priv, &priv->map->CMD);
+	msleep(200);
+
+	if (rtl818x_ioread8(priv, &priv->map->CONFIG3) & (1 << 3)) {
+		/* For cardbus */
+		reg = rtl818x_ioread8(priv, &priv->map->CONFIG3);
+		reg |= 1 << 1;
+		rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg);
+		reg = rtl818x_ioread16(priv, &priv->map->FEMR);
+		reg |= (1 << 15) | (1 << 14) | (1 << 4);
+		rtl818x_iowrite16(priv, &priv->map->FEMR, reg);
+	}
+
+	rtl818x_iowrite8(priv, &priv->map->MSR, 0);
+
+	if (!priv->r8185)
+		rtl8180_set_anaparam(priv, priv->anaparam);
+
+	rtl818x_iowrite32(priv, &priv->map->RDSAR, priv->rx_ring_dma);
+	rtl818x_iowrite32(priv, &priv->map->TBDA, priv->tx_ring[3].dma);
+	rtl818x_iowrite32(priv, &priv->map->THPDA, priv->tx_ring[2].dma);
+	rtl818x_iowrite32(priv, &priv->map->TNPDA, priv->tx_ring[1].dma);
+	rtl818x_iowrite32(priv, &priv->map->TLPDA, priv->tx_ring[0].dma);
+
+	/* TODO: necessary? specs indicate not */
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	reg = rtl818x_ioread8(priv, &priv->map->CONFIG2);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG2, reg & ~(1 << 3));
+	if (priv->r8185) {
+		reg = rtl818x_ioread8(priv, &priv->map->CONFIG2);
+		rtl818x_iowrite8(priv, &priv->map->CONFIG2, reg | (1 << 4));
+	}
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	/* TODO: set CONFIG5 for calibrating AGC on rtl8180 + philips radio? */
+
+	/* TODO: turn off hw wep on rtl8180 */
+
+	rtl818x_iowrite32(priv, &priv->map->INT_TIMEOUT, 0);
+
+	if (priv->r8185) {
+		rtl818x_iowrite8(priv, &priv->map->WPA_CONF, 0);
+		rtl818x_iowrite8(priv, &priv->map->RATE_FALLBACK, 0x81);
+		rtl818x_iowrite8(priv, &priv->map->RESP_RATE, (8 << 4) | 0);
+
+		rtl818x_iowrite16(priv, &priv->map->BRSR, 0x01F3);
+
+		/* TODO: set ClkRun enable? necessary? */
+		reg = rtl818x_ioread8(priv, &priv->map->GP_ENABLE);
+		rtl818x_iowrite8(priv, &priv->map->GP_ENABLE, reg & ~(1 << 6));
+		rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+		reg = rtl818x_ioread8(priv, &priv->map->CONFIG3);
+		rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg | (1 << 2));
+		rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+	} else {
+		rtl818x_iowrite16(priv, &priv->map->BRSR, 0x1);
+		rtl818x_iowrite8(priv, &priv->map->SECURITY, 0);
+
+		rtl818x_iowrite8(priv, &priv->map->PHY_DELAY, 0x6);
+		rtl818x_iowrite8(priv, &priv->map->CARRIER_SENSE_COUNTER, 0x4C);
+	}
+
+	priv->rf->init(dev);
+	if (priv->r8185)
+		rtl818x_iowrite16(priv, &priv->map->BRSR, 0x01F3);
+	return 0;
+}
+
+static int rtl8180_init_rx_ring(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	struct rtl8180_rx_desc *entry;
+	int i;
+
+	priv->rx_ring = pci_alloc_consistent(priv->pdev,
+					     sizeof(*priv->rx_ring) * 32,
+					     &priv->rx_ring_dma);
+
+	if (!priv->rx_ring || (unsigned long)priv->rx_ring & 0xFF) {
+		printk(KERN_ERR "%s: Cannot allocate RX ring\n",
+		       wiphy_name(dev->wiphy));
+		return -ENOMEM;
+	}
+
+	memset(priv->rx_ring, 0, sizeof(*priv->rx_ring) * 32);
+	priv->rx_idx = 0;
+
+	for (i = 0; i < 32; i++) {
+		struct sk_buff *skb = dev_alloc_skb(MAX_RX_SIZE);
+		dma_addr_t *mapping;
+		entry = &priv->rx_ring[i];
+		if (!skb)
+			return 0;
+
+		priv->rx_buf[i] = skb;
+		mapping = (dma_addr_t *)skb->cb;
+		*mapping = pci_map_single(priv->pdev, skb_tail_pointer(skb),
+					  MAX_RX_SIZE, PCI_DMA_FROMDEVICE);
+		entry->rx_buf = cpu_to_le32(*mapping);
+		entry->flags = cpu_to_le32(RTL8180_RX_DESC_FLAG_OWN |
+					   MAX_RX_SIZE);
+	}
+	entry->flags |= cpu_to_le32(RTL8180_RX_DESC_FLAG_EOR);
+	return 0;
+}
+
+static void rtl8180_free_rx_ring(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int i;
+
+	for (i = 0; i < 32; i++) {
+		struct sk_buff *skb = priv->rx_buf[i];
+		if (!skb)
+			continue;
+
+		pci_unmap_single(priv->pdev,
+				 *((dma_addr_t *)skb->cb),
+				 MAX_RX_SIZE, PCI_DMA_FROMDEVICE);
+		kfree_skb(skb);
+	}
+
+	pci_free_consistent(priv->pdev, sizeof(*priv->rx_ring) * 32,
+			    priv->rx_ring, priv->rx_ring_dma);
+	priv->rx_ring = NULL;
+}
+
+static int rtl8180_init_tx_ring(struct ieee80211_hw *dev,
+				unsigned int prio, unsigned int entries)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	struct rtl8180_tx_desc *ring;
+	dma_addr_t dma;
+	int i;
+
+	ring = pci_alloc_consistent(priv->pdev, sizeof(*ring) * entries, &dma);
+	if (!ring || (unsigned long)ring & 0xFF) {
+		printk(KERN_ERR "%s: Cannot allocate TX ring (prio = %d)\n",
+		       wiphy_name(dev->wiphy), prio);
+		return -ENOMEM;
+	}
+
+	memset(ring, 0, sizeof(*ring)*entries);
+	priv->tx_ring[prio].desc = ring;
+	priv->tx_ring[prio].dma = dma;
+	priv->tx_ring[prio].idx = 0;
+	priv->tx_ring[prio].entries = entries;
+	skb_queue_head_init(&priv->tx_ring[prio].queue);
+
+	for (i = 0; i < entries; i++)
+		ring[i].next_tx_desc =
+			cpu_to_le32((u32)dma + ((i + 1) % entries) * sizeof(*ring));
+
+	return 0;
+}
+
+static void rtl8180_free_tx_ring(struct ieee80211_hw *dev, unsigned int prio)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	struct rtl8180_tx_ring *ring = &priv->tx_ring[prio];
+
+	while (skb_queue_len(&ring->queue)) {
+		struct rtl8180_tx_desc *entry = &ring->desc[ring->idx];
+		struct sk_buff *skb = __skb_dequeue(&ring->queue);
+
+		pci_unmap_single(priv->pdev, le32_to_cpu(entry->tx_buf),
+				 skb->len, PCI_DMA_TODEVICE);
+		kfree(*((struct ieee80211_tx_control **) skb->cb));
+		kfree_skb(skb);
+		ring->idx = (ring->idx + 1) % ring->entries;
+	}
+
+	pci_free_consistent(priv->pdev, sizeof(*ring->desc)*ring->entries,
+			    ring->desc, ring->dma);
+	ring->desc = NULL;
+}
+
+static int rtl8180_start(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int ret, i;
+	u32 reg;
+
+	ret = rtl8180_init_rx_ring(dev);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < 4; i++)
+		if ((ret = rtl8180_init_tx_ring(dev, i, 16)))
+			goto err_free_rings;
+
+	ret = rtl8180_init_hw(dev);
+	if (ret)
+		goto err_free_rings;
+
+	rtl818x_iowrite32(priv, &priv->map->RDSAR, priv->rx_ring_dma);
+	rtl818x_iowrite32(priv, &priv->map->TBDA, priv->tx_ring[3].dma);
+	rtl818x_iowrite32(priv, &priv->map->THPDA, priv->tx_ring[2].dma);
+	rtl818x_iowrite32(priv, &priv->map->TNPDA, priv->tx_ring[1].dma);
+	rtl818x_iowrite32(priv, &priv->map->TLPDA, priv->tx_ring[0].dma);
+
+	ret = request_irq(priv->pdev->irq, &rtl8180_interrupt,
+			  IRQF_SHARED, KBUILD_MODNAME, dev);
+	if (ret) {
+		printk(KERN_ERR "%s: failed to register IRQ handler\n",
+		       wiphy_name(dev->wiphy));
+		goto err_free_rings;
+	}
+
+	rtl818x_iowrite16(priv, &priv->map->INT_MASK, 0xFFFF);
+
+	rtl818x_iowrite32(priv, &priv->map->MAR[0], ~0);
+	rtl818x_iowrite32(priv, &priv->map->MAR[1], ~0);
+
+	reg = RTL818X_RX_CONF_ONLYERLPKT |
+	      RTL818X_RX_CONF_RX_AUTORESETPHY |
+	      RTL818X_RX_CONF_MGMT |
+	      RTL818X_RX_CONF_DATA |
+	      (7 << 8 /* MAX RX DMA */) |
+	      RTL818X_RX_CONF_BROADCAST |
+	      RTL818X_RX_CONF_NICMAC;
+
+	if (priv->r8185)
+		reg |= RTL818X_RX_CONF_CSDM1 | RTL818X_RX_CONF_CSDM2;
+	else {
+		reg |= (priv->rfparam & RF_PARAM_CARRIERSENSE1)
+			? RTL818X_RX_CONF_CSDM1 : 0;
+		reg |= (priv->rfparam & RF_PARAM_CARRIERSENSE2)
+			? RTL818X_RX_CONF_CSDM2 : 0;
+	}
+
+	priv->rx_conf = reg;
+	rtl818x_iowrite32(priv, &priv->map->RX_CONF, reg);
+
+	if (priv->r8185) {
+		reg = rtl818x_ioread8(priv, &priv->map->CW_CONF);
+		reg &= ~RTL818X_CW_CONF_PERPACKET_CW_SHIFT;
+		reg |= RTL818X_CW_CONF_PERPACKET_RETRY_SHIFT;
+		rtl818x_iowrite8(priv, &priv->map->CW_CONF, reg);
+
+		reg = rtl818x_ioread8(priv, &priv->map->TX_AGC_CTL);
+		reg &= ~RTL818X_TX_AGC_CTL_PERPACKET_GAIN_SHIFT;
+		reg &= ~RTL818X_TX_AGC_CTL_PERPACKET_ANTSEL_SHIFT;
+		reg |=  RTL818X_TX_AGC_CTL_FEEDBACK_ANT;
+		rtl818x_iowrite8(priv, &priv->map->TX_AGC_CTL, reg);
+
+		/* disable early TX */
+		rtl818x_iowrite8(priv, (u8 __iomem *)priv->map + 0xec, 0x3f);
+	}
+
+	reg = rtl818x_ioread32(priv, &priv->map->TX_CONF);
+	reg |= (6 << 21 /* MAX TX DMA */) |
+	       RTL818X_TX_CONF_NO_ICV;
+
+	if (priv->r8185)
+		reg &= ~RTL818X_TX_CONF_PROBE_DTS;
+	else
+		reg &= ~RTL818X_TX_CONF_HW_SEQNUM;
+
+	/* different meaning, same value on both rtl8185 and rtl8180 */
+	reg &= ~RTL818X_TX_CONF_SAT_HWPLCP;
+
+	rtl818x_iowrite32(priv, &priv->map->TX_CONF, reg);
+
+	reg = rtl818x_ioread8(priv, &priv->map->CMD);
+	reg |= RTL818X_CMD_RX_ENABLE;
+	reg |= RTL818X_CMD_TX_ENABLE;
+	rtl818x_iowrite8(priv, &priv->map->CMD, reg);
+
+	priv->mode = IEEE80211_IF_TYPE_MNTR;
+	return 0;
+
+ err_free_rings:
+	rtl8180_free_rx_ring(dev);
+	for (i = 0; i < 4; i++)
+		if (priv->tx_ring[i].desc)
+			rtl8180_free_tx_ring(dev, i);
+
+	return ret;
+}
+
+static void rtl8180_stop(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 reg;
+	int i;
+
+	priv->mode = IEEE80211_IF_TYPE_INVALID;
+
+	rtl818x_iowrite16(priv, &priv->map->INT_MASK, 0);
+
+	reg = rtl818x_ioread8(priv, &priv->map->CMD);
+	reg &= ~RTL818X_CMD_TX_ENABLE;
+	reg &= ~RTL818X_CMD_RX_ENABLE;
+	rtl818x_iowrite8(priv, &priv->map->CMD, reg);
+
+	priv->rf->stop(dev);
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	reg = rtl818x_ioread8(priv, &priv->map->CONFIG4);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG4, reg | RTL818X_CONFIG4_VCOOFF);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	free_irq(priv->pdev->irq, dev);
+
+	rtl8180_free_rx_ring(dev);
+	for (i = 0; i < 4; i++)
+		rtl8180_free_tx_ring(dev, i);
+}
+
+static int rtl8180_add_interface(struct ieee80211_hw *dev,
+				 struct ieee80211_if_init_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	if (priv->mode != IEEE80211_IF_TYPE_MNTR)
+		return -EOPNOTSUPP;
+
+	switch (conf->type) {
+	case IEEE80211_IF_TYPE_STA:
+		priv->mode = conf->type;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	rtl818x_iowrite32(priv, (__le32 __iomem *)&priv->map->MAC[0],
+			  cpu_to_le32(*(u32 *)conf->mac_addr));
+	rtl818x_iowrite16(priv, (__le16 __iomem *)&priv->map->MAC[4],
+			  cpu_to_le16(*(u16 *)(conf->mac_addr + 4)));
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	return 0;
+}
+
+static void rtl8180_remove_interface(struct ieee80211_hw *dev,
+				     struct ieee80211_if_init_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	priv->mode = IEEE80211_IF_TYPE_MNTR;
+}
+
+static int rtl8180_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	priv->rf->set_chan(dev, conf);
+
+	return 0;
+}
+
+static int rtl8180_config_interface(struct ieee80211_hw *dev, int if_id,
+				    struct ieee80211_if_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int i;
+
+	priv->if_id = if_id;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		rtl818x_iowrite8(priv, &priv->map->BSSID[i], conf->bssid[i]);
+
+	if (is_valid_ether_addr(conf->bssid))
+		rtl818x_iowrite8(priv, &priv->map->MSR, RTL818X_MSR_INFRA);
+	else
+		rtl818x_iowrite8(priv, &priv->map->MSR, RTL818X_MSR_NO_LINK);
+
+	return 0;
+}
+
+static void rtl8180_configure_filter(struct ieee80211_hw *dev,
+				     unsigned int changed_flags,
+				     unsigned int *total_flags,
+				     int mc_count, struct dev_addr_list *mclist)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	if (changed_flags & FIF_FCSFAIL)
+		priv->rx_conf ^= RTL818X_RX_CONF_FCS;
+	if (changed_flags & FIF_CONTROL)
+		priv->rx_conf ^= RTL818X_RX_CONF_CTRL;
+	if (changed_flags & FIF_OTHER_BSS)
+		priv->rx_conf ^= RTL818X_RX_CONF_MONITOR;
+	if (*total_flags & FIF_ALLMULTI || mc_count > 0)
+		priv->rx_conf |= RTL818X_RX_CONF_MULTICAST;
+	else
+		priv->rx_conf &= ~RTL818X_RX_CONF_MULTICAST;
+
+	*total_flags = 0;
+
+	if (priv->rx_conf & RTL818X_RX_CONF_FCS)
+		*total_flags |= FIF_FCSFAIL;
+	if (priv->rx_conf & RTL818X_RX_CONF_CTRL)
+		*total_flags |= FIF_CONTROL;
+	if (priv->rx_conf & RTL818X_RX_CONF_MONITOR)
+		*total_flags |= FIF_OTHER_BSS;
+	if (priv->rx_conf & RTL818X_RX_CONF_MULTICAST)
+		*total_flags |= FIF_ALLMULTI;
+
+	rtl818x_iowrite32(priv, &priv->map->RX_CONF, priv->rx_conf);
+}
+
+static const struct ieee80211_ops rtl8180_ops = {
+	.tx			= rtl8180_tx,
+	.start			= rtl8180_start,
+	.stop			= rtl8180_stop,
+	.add_interface		= rtl8180_add_interface,
+	.remove_interface	= rtl8180_remove_interface,
+	.config			= rtl8180_config,
+	.config_interface	= rtl8180_config_interface,
+	.configure_filter	= rtl8180_configure_filter,
+};
+
+static void rtl8180_eeprom_register_read(struct eeprom_93cx6 *eeprom)
+{
+	struct ieee80211_hw *dev = eeprom->data;
+	struct rtl8180_priv *priv = dev->priv;
+	u8 reg = rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+
+	eeprom->reg_data_in = reg & RTL818X_EEPROM_CMD_WRITE;
+	eeprom->reg_data_out = reg & RTL818X_EEPROM_CMD_READ;
+	eeprom->reg_data_clock = reg & RTL818X_EEPROM_CMD_CK;
+	eeprom->reg_chip_select = reg & RTL818X_EEPROM_CMD_CS;
+}
+
+static void rtl8180_eeprom_register_write(struct eeprom_93cx6 *eeprom)
+{
+	struct ieee80211_hw *dev = eeprom->data;
+	struct rtl8180_priv *priv = dev->priv;
+	u8 reg = 2 << 6;
+
+	if (eeprom->reg_data_in)
+		reg |= RTL818X_EEPROM_CMD_WRITE;
+	if (eeprom->reg_data_out)
+		reg |= RTL818X_EEPROM_CMD_READ;
+	if (eeprom->reg_data_clock)
+		reg |= RTL818X_EEPROM_CMD_CK;
+	if (eeprom->reg_chip_select)
+		reg |= RTL818X_EEPROM_CMD_CS;
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, reg);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(10);
+}
+
+static int __devinit rtl8180_probe(struct pci_dev *pdev,
+				   const struct pci_device_id *id)
+{
+	struct ieee80211_hw *dev;
+	struct rtl8180_priv *priv;
+	unsigned long mem_addr, mem_len;
+	unsigned int io_addr, io_len;
+	int err, i;
+	struct eeprom_93cx6 eeprom;
+	const char *chip_name, *rf_name = NULL;
+	u32 reg;
+	u16 eeprom_val;
+	DECLARE_MAC_BUF(mac);
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		printk(KERN_ERR "%s (rtl8180): Cannot enable new PCI device\n",
+		       pci_name(pdev));
+		return err;
+	}
+
+	err = pci_request_regions(pdev, KBUILD_MODNAME);
+	if (err) {
+		printk(KERN_ERR "%s (rtl8180): Cannot obtain PCI resources\n",
+		       pci_name(pdev));
+		return err;
+	}
+
+	io_addr = pci_resource_start(pdev, 0);
+	io_len = pci_resource_len(pdev, 0);
+	mem_addr = pci_resource_start(pdev, 1);
+	mem_len = pci_resource_len(pdev, 1);
+
+	if (mem_len < sizeof(struct rtl818x_csr) ||
+	    io_len < sizeof(struct rtl818x_csr)) {
+		printk(KERN_ERR "%s (rtl8180): Too short PCI resources\n",
+		       pci_name(pdev));
+		err = -ENOMEM;
+		goto err_free_reg;
+	}
+
+	if ((err = pci_set_dma_mask(pdev, 0xFFFFFF00ULL)) ||
+	    (err = pci_set_consistent_dma_mask(pdev, 0xFFFFFF00ULL))) {
+		printk(KERN_ERR "%s (rtl8180): No suitable DMA available\n",
+		       pci_name(pdev));
+		goto err_free_reg;
+	}
+
+	pci_set_master(pdev);
+
+	dev = ieee80211_alloc_hw(sizeof(*priv), &rtl8180_ops);
+	if (!dev) {
+		printk(KERN_ERR "%s (rtl8180): ieee80211 alloc failed\n",
+		       pci_name(pdev));
+		err = -ENOMEM;
+		goto err_free_reg;
+	}
+
+	priv = dev->priv;
+	priv->pdev = pdev;
+
+	SET_IEEE80211_DEV(dev, &pdev->dev);
+	pci_set_drvdata(pdev, dev);
+
+	priv->map = pci_iomap(pdev, 1, mem_len);
+	if (!priv->map)
+		priv->map = pci_iomap(pdev, 0, io_len);
+
+	if (!priv->map) {
+		printk(KERN_ERR "%s (rtl8180): Cannot map device memory\n",
+		       pci_name(pdev));
+		goto err_free_dev;
+	}
+
+	memcpy(priv->channels, rtl818x_channels, sizeof(rtl818x_channels));
+	memcpy(priv->rates, rtl818x_rates, sizeof(rtl818x_rates));
+	priv->modes[0].mode = MODE_IEEE80211G;
+	priv->modes[0].num_rates = ARRAY_SIZE(rtl818x_rates);
+	priv->modes[0].rates = priv->rates;
+	priv->modes[0].num_channels = ARRAY_SIZE(rtl818x_channels);
+	priv->modes[0].channels = priv->channels;
+	priv->modes[1].mode = MODE_IEEE80211B;
+	priv->modes[1].num_rates = 4;
+	priv->modes[1].rates = priv->rates;
+	priv->modes[1].num_channels = ARRAY_SIZE(rtl818x_channels);
+	priv->modes[1].channels = priv->channels;
+	priv->mode = IEEE80211_IF_TYPE_INVALID;
+	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+		     IEEE80211_HW_RX_INCLUDES_FCS;
+	dev->queues = 1;
+	dev->max_rssi = 65;
+
+	reg = rtl818x_ioread32(priv, &priv->map->TX_CONF);
+	reg &= RTL818X_TX_CONF_HWVER_MASK;
+	switch (reg) {
+	case RTL818X_TX_CONF_R8180_ABCD:
+		chip_name = "RTL8180";
+		break;
+	case RTL818X_TX_CONF_R8180_F:
+		chip_name = "RTL8180vF";
+		break;
+	case RTL818X_TX_CONF_R8185_ABC:
+		chip_name = "RTL8185";
+		break;
+	case RTL818X_TX_CONF_R8185_D:
+		chip_name = "RTL8185vD";
+		break;
+	default:
+		printk(KERN_ERR "%s (rtl8180): Unknown chip! (0x%x)\n",
+		       pci_name(pdev), reg >> 25);
+		goto err_iounmap;
+	}
+
+	priv->r8185 = reg & RTL818X_TX_CONF_R8185_ABC;
+	if (priv->r8185) {
+		if ((err = ieee80211_register_hwmode(dev, &priv->modes[0])))
+			goto err_iounmap;
+
+		pci_try_set_mwi(pdev);
+	}
+
+	if ((err = ieee80211_register_hwmode(dev, &priv->modes[1])))
+		goto err_iounmap;
+
+	eeprom.data = dev;
+	eeprom.register_read = rtl8180_eeprom_register_read;
+	eeprom.register_write = rtl8180_eeprom_register_write;
+	if (rtl818x_ioread32(priv, &priv->map->RX_CONF) & (1 << 6))
+		eeprom.width = PCI_EEPROM_WIDTH_93C66;
+	else
+		eeprom.width = PCI_EEPROM_WIDTH_93C46;
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_PROGRAM);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(10);
+
+	eeprom_93cx6_read(&eeprom, 0x06, &eeprom_val);
+	eeprom_val &= 0xFF;
+	switch (eeprom_val) {
+	case 1:	rf_name = "Intersil";
+		break;
+	case 2:	rf_name = "RFMD";
+		break;
+	case 3:	priv->rf = &sa2400_rf_ops;
+		break;
+	case 4:	priv->rf = &max2820_rf_ops;
+		break;
+	case 5:	priv->rf = &grf5101_rf_ops;
+		break;
+	case 9:	priv->rf = rtl8180_detect_rf(dev);
+		break;
+	case 10:
+		rf_name = "RTL8255";
+		break;
+	default:
+		printk(KERN_ERR "%s (rtl8180): Unknown RF! (0x%x)\n",
+		       pci_name(pdev), eeprom_val);
+		goto err_iounmap;
+	}
+
+	if (!priv->rf) {
+		printk(KERN_ERR "%s (rtl8180): %s RF frontend not supported!\n",
+		       pci_name(pdev), rf_name);
+		goto err_iounmap;
+	}
+
+	eeprom_93cx6_read(&eeprom, 0x17, &eeprom_val);
+	priv->csthreshold = eeprom_val >> 8;
+	if (!priv->r8185) {
+		__le32 anaparam;
+		eeprom_93cx6_multiread(&eeprom, 0xD, (__le16 *)&anaparam, 2);
+		priv->anaparam = le32_to_cpu(anaparam);
+		eeprom_93cx6_read(&eeprom, 0x19, &priv->rfparam);
+	}
+
+	eeprom_93cx6_multiread(&eeprom, 0x7, (__le16 *)dev->wiphy->perm_addr, 3);
+	if (!is_valid_ether_addr(dev->wiphy->perm_addr)) {
+		printk(KERN_WARNING "%s (rtl8180): Invalid hwaddr! Using"
+		       " randomly generated MAC addr\n", pci_name(pdev));
+		random_ether_addr(dev->wiphy->perm_addr);
+	}
+
+	/* CCK TX power */
+	for (i = 0; i < 14; i += 2) {
+		u16 txpwr;
+		eeprom_93cx6_read(&eeprom, 0x10 + (i >> 1), &txpwr);
+		priv->channels[i].val = txpwr & 0xFF;
+		priv->channels[i + 1].val = txpwr >> 8;
+	}
+
+	/* OFDM TX power */
+	if (priv->r8185) {
+		for (i = 0; i < 14; i += 2) {
+			u16 txpwr;
+			eeprom_93cx6_read(&eeprom, 0x20 + (i >> 1), &txpwr);
+			priv->channels[i].val |= (txpwr & 0xFF) << 8;
+			priv->channels[i + 1].val |= txpwr & 0xFF00;
+		}
+	}
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	spin_lock_init(&priv->lock);
+
+	err = ieee80211_register_hw(dev);
+	if (err) {
+		printk(KERN_ERR "%s (rtl8180): Cannot register device\n",
+		       pci_name(pdev));
+		goto err_iounmap;
+	}
+
+	printk(KERN_INFO "%s: hwaddr %s, %s + %s\n",
+	       wiphy_name(dev->wiphy), print_mac(mac, dev->wiphy->perm_addr),
+	       chip_name, priv->rf->name);
+
+	return 0;
+
+ err_iounmap:
+	iounmap(priv->map);
+
+ err_free_dev:
+	pci_set_drvdata(pdev, NULL);
+	ieee80211_free_hw(dev);
+
+ err_free_reg:
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	return err;
+}
+
+static void __devexit rtl8180_remove(struct pci_dev *pdev)
+{
+	struct ieee80211_hw *dev = pci_get_drvdata(pdev);
+	struct rtl8180_priv *priv;
+
+	if (!dev)
+		return;
+
+	ieee80211_unregister_hw(dev);
+
+	priv = dev->priv;
+
+	pci_iounmap(pdev, priv->map);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	ieee80211_free_hw(dev);
+}
+
+#ifdef CONFIG_PM
+static int rtl8180_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	pci_save_state(pdev);
+	pci_set_power_state(pdev, pci_choose_state(pdev, state));
+	return 0;
+}
+
+static int rtl8180_resume(struct pci_dev *pdev)
+{
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+	return 0;
+}
+
+#endif /* CONFIG_PM */
+
+static struct pci_driver rtl8180_driver = {
+	.name		= KBUILD_MODNAME,
+	.id_table	= rtl8180_table,
+	.probe		= rtl8180_probe,
+	.remove		= __devexit_p(rtl8180_remove),
+#ifdef CONFIG_PM
+	.suspend	= rtl8180_suspend,
+	.resume		= rtl8180_resume,
+#endif /* CONFIG_PM */
+};
+
+static int __init rtl8180_init(void)
+{
+	return pci_register_driver(&rtl8180_driver);
+}
+
+static void __exit rtl8180_exit(void)
+{
+	pci_unregister_driver(&rtl8180_driver);
+}
+
+module_init(rtl8180_init);
+module_exit(rtl8180_exit);
diff --git a/drivers/net/wireless/rtl8180_grf5101.c b/drivers/net/wireless/rtl8180_grf5101.c
new file mode 100644
index 000000000000..8293e19c4c59
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_grf5101.c
@@ -0,0 +1,179 @@
+
+/*
+ * Radio tuning for GCT GRF5101 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <net/mac80211.h>
+
+#include "rtl8180.h"
+#include "rtl8180_grf5101.h"
+
+static const int grf5101_encode[] = {
+	0x0, 0x8, 0x4, 0xC,
+	0x2, 0xA, 0x6, 0xE,
+	0x1, 0x9, 0x5, 0xD,
+	0x3, 0xB, 0x7, 0xF
+};
+
+static void write_grf5101(struct ieee80211_hw *dev, u8 addr, u32 data)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 phy_config;
+
+	phy_config =  grf5101_encode[(data >> 8) & 0xF];
+	phy_config |= grf5101_encode[(data >> 4) & 0xF] << 4;
+	phy_config |= grf5101_encode[data & 0xF] << 8;
+	phy_config |= grf5101_encode[(addr >> 1) & 0xF] << 12;
+	phy_config |= (addr & 1) << 16;
+	phy_config |= grf5101_encode[(data & 0xf000) >> 12] << 24;
+
+	/* MAC will bang bits to the chip */
+	phy_config |= 0x90000000;
+
+	rtl818x_iowrite32(priv,
+		(__le32 __iomem *) &priv->map->RFPinsOutput, phy_config);
+
+	msleep(3);
+}
+
+static void grf5101_write_phy_antenna(struct ieee80211_hw *dev, short chan)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 ant = GRF5101_ANTENNA;
+
+	if (priv->rfparam & RF_PARAM_ANTBDEFAULT)
+		ant |= BB_ANTENNA_B;
+
+	if (chan == 14)
+		ant |= BB_ANTATTEN_CHAN14;
+
+	rtl8180_write_phy(dev, 0x10, ant);
+}
+
+static void grf5101_rf_set_channel(struct ieee80211_hw *dev,
+				   struct ieee80211_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 txpw = priv->channels[conf->channel - 1].val & 0xFF;
+	u32 chan = conf->channel - 1;
+
+	/* set TX power */
+	write_grf5101(dev, 0x15, 0x0);
+	write_grf5101(dev, 0x06, txpw);
+	write_grf5101(dev, 0x15, 0x10);
+	write_grf5101(dev, 0x15, 0x0);
+
+	/* set frequency */
+	write_grf5101(dev, 0x07, 0x0);
+	write_grf5101(dev, 0x0B, chan);
+	write_grf5101(dev, 0x07, 0x1000);
+
+	grf5101_write_phy_antenna(dev, chan);
+}
+
+static void grf5101_rf_stop(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 anaparam;
+
+	anaparam = priv->anaparam;
+	anaparam &= 0x000fffff;
+	anaparam |= 0x3f900000;
+	rtl8180_set_anaparam(priv, anaparam);
+
+	write_grf5101(dev, 0x07, 0x0);
+	write_grf5101(dev, 0x1f, 0x45);
+	write_grf5101(dev, 0x1f, 0x5);
+	write_grf5101(dev, 0x00, 0x8e4);
+}
+
+static void grf5101_rf_init(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	rtl8180_set_anaparam(priv, priv->anaparam);
+
+	write_grf5101(dev, 0x1f, 0x0);
+	write_grf5101(dev, 0x1f, 0x0);
+	write_grf5101(dev, 0x1f, 0x40);
+	write_grf5101(dev, 0x1f, 0x60);
+	write_grf5101(dev, 0x1f, 0x61);
+	write_grf5101(dev, 0x1f, 0x61);
+	write_grf5101(dev, 0x00, 0xae4);
+	write_grf5101(dev, 0x1f, 0x1);
+	write_grf5101(dev, 0x1f, 0x41);
+	write_grf5101(dev, 0x1f, 0x61);
+
+	write_grf5101(dev, 0x01, 0x1a23);
+	write_grf5101(dev, 0x02, 0x4971);
+	write_grf5101(dev, 0x03, 0x41de);
+	write_grf5101(dev, 0x04, 0x2d80);
+	write_grf5101(dev, 0x05, 0x68ff);	/* 0x61ff original value */
+	write_grf5101(dev, 0x06, 0x0);
+	write_grf5101(dev, 0x07, 0x0);
+	write_grf5101(dev, 0x08, 0x7533);
+	write_grf5101(dev, 0x09, 0xc401);
+	write_grf5101(dev, 0x0a, 0x0);
+	write_grf5101(dev, 0x0c, 0x1c7);
+	write_grf5101(dev, 0x0d, 0x29d3);
+	write_grf5101(dev, 0x0e, 0x2e8);
+	write_grf5101(dev, 0x10, 0x192);
+	write_grf5101(dev, 0x11, 0x248);
+	write_grf5101(dev, 0x12, 0x0);
+	write_grf5101(dev, 0x13, 0x20c4);
+	write_grf5101(dev, 0x14, 0xf4fc);
+	write_grf5101(dev, 0x15, 0x0);
+	write_grf5101(dev, 0x16, 0x1500);
+
+	write_grf5101(dev, 0x07, 0x1000);
+
+	/* baseband configuration */
+	rtl8180_write_phy(dev, 0, 0xa8);
+	rtl8180_write_phy(dev, 3, 0x0);
+	rtl8180_write_phy(dev, 4, 0xc0);
+	rtl8180_write_phy(dev, 5, 0x90);
+	rtl8180_write_phy(dev, 6, 0x1e);
+	rtl8180_write_phy(dev, 7, 0x64);
+
+	grf5101_write_phy_antenna(dev, 1);
+
+	rtl8180_write_phy(dev, 0x11, 0x88);
+
+	if (rtl818x_ioread8(priv, &priv->map->CONFIG2) &
+	    RTL818X_CONFIG2_ANTENNA_DIV)
+		rtl8180_write_phy(dev, 0x12, 0xc0); /* enable ant diversity */
+	else
+		rtl8180_write_phy(dev, 0x12, 0x40); /* disable ant diversity */
+
+	rtl8180_write_phy(dev, 0x13, 0x90 | priv->csthreshold);
+
+	rtl8180_write_phy(dev, 0x19, 0x0);
+	rtl8180_write_phy(dev, 0x1a, 0xa0);
+	rtl8180_write_phy(dev, 0x1b, 0x44);
+}
+
+const struct rtl818x_rf_ops grf5101_rf_ops = {
+	.name		= "GCT",
+	.init		= grf5101_rf_init,
+	.stop		= grf5101_rf_stop,
+	.set_chan	= grf5101_rf_set_channel
+};
diff --git a/drivers/net/wireless/rtl8180_grf5101.h b/drivers/net/wireless/rtl8180_grf5101.h
new file mode 100644
index 000000000000..76647111bcff
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_grf5101.h
@@ -0,0 +1,28 @@
+#ifndef RTL8180_GRF5101_H
+#define RTL8180_GRF5101_H
+
+/*
+ * Radio tuning for GCT GRF5101 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define GRF5101_ANTENNA 0xA3
+
+extern const struct rtl818x_rf_ops grf5101_rf_ops;
+
+#endif /* RTL8180_GRF5101_H */
diff --git a/drivers/net/wireless/rtl8180_max2820.c b/drivers/net/wireless/rtl8180_max2820.c
new file mode 100644
index 000000000000..98fe9fd64968
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_max2820.c
@@ -0,0 +1,150 @@
+/*
+ * Radio tuning for Maxim max2820 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <net/mac80211.h>
+
+#include "rtl8180.h"
+#include "rtl8180_max2820.h"
+
+static const u32 max2820_chan[] = {
+	12, /* CH 1 */
+	17,
+	22,
+	27,
+	32,
+	37,
+	42,
+	47,
+	52,
+	57,
+	62,
+	67,
+	72,
+	84, /* CH 14 */
+};
+
+static void write_max2820(struct ieee80211_hw *dev, u8 addr, u32 data)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 phy_config;
+
+	phy_config = 0x90 + (data & 0xf);
+	phy_config <<= 16;
+	phy_config += addr;
+	phy_config <<= 8;
+	phy_config += (data >> 4) & 0xff;
+
+	rtl818x_iowrite32(priv,
+		(__le32 __iomem *) &priv->map->RFPinsOutput, phy_config);
+
+	msleep(1);
+}
+
+static void max2820_write_phy_antenna(struct ieee80211_hw *dev, short chan)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 ant;
+
+	ant = MAXIM_ANTENNA;
+	if (priv->rfparam & RF_PARAM_ANTBDEFAULT)
+		ant |= BB_ANTENNA_B;
+	if (chan == 14)
+		ant |= BB_ANTATTEN_CHAN14;
+
+	rtl8180_write_phy(dev, 0x10, ant);
+}
+
+static void max2820_rf_set_channel(struct ieee80211_hw *dev,
+				   struct ieee80211_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	unsigned int chan_idx = conf ? conf->channel - 1 : 0;
+	u32 txpw = priv->channels[chan_idx].val & 0xFF;
+	u32 chan = max2820_chan[chan_idx];
+
+	/* While philips SA2400 drive the PA bias from
+	 * sa2400, for MAXIM we do this directly from BB */
+	rtl8180_write_phy(dev, 3, txpw);
+
+	max2820_write_phy_antenna(dev, chan);
+	write_max2820(dev, 3, chan);
+}
+
+static void max2820_rf_stop(struct ieee80211_hw *dev)
+{
+	rtl8180_write_phy(dev, 3, 0x8);
+	write_max2820(dev, 1, 0);
+}
+
+
+static void max2820_rf_init(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	/* MAXIM from netbsd driver */
+	write_max2820(dev, 0, 0x007); /* test mode as indicated in datasheet */
+	write_max2820(dev, 1, 0x01e); /* enable register */
+	write_max2820(dev, 2, 0x001); /* synt register */
+
+	max2820_rf_set_channel(dev, NULL);
+
+	write_max2820(dev, 4, 0x313); /* rx register */
+
+	/* PA is driven directly by the BB, we keep the MAXIM bias
+	 * at the highest value in case that setting it to lower
+	 * values may introduce some further attenuation somewhere..
+	 */
+	write_max2820(dev, 5, 0x00f);
+
+	/* baseband configuration */
+	rtl8180_write_phy(dev, 0, 0x88); /* sys1       */
+	rtl8180_write_phy(dev, 3, 0x08); /* txagc      */
+	rtl8180_write_phy(dev, 4, 0xf8); /* lnadet     */
+	rtl8180_write_phy(dev, 5, 0x90); /* ifagcinit  */
+	rtl8180_write_phy(dev, 6, 0x1a); /* ifagclimit */
+	rtl8180_write_phy(dev, 7, 0x64); /* ifagcdet   */
+
+	max2820_write_phy_antenna(dev, 1);
+
+	rtl8180_write_phy(dev, 0x11, 0x88); /* trl */
+
+	if (rtl818x_ioread8(priv, &priv->map->CONFIG2) &
+	    RTL818X_CONFIG2_ANTENNA_DIV)
+		rtl8180_write_phy(dev, 0x12, 0xc7);
+	else
+		rtl8180_write_phy(dev, 0x12, 0x47);
+
+	rtl8180_write_phy(dev, 0x13, 0x9b);
+
+	rtl8180_write_phy(dev, 0x19, 0x0);  /* CHESTLIM */
+	rtl8180_write_phy(dev, 0x1a, 0x9f); /* CHSQLIM  */
+
+	max2820_rf_set_channel(dev, NULL);
+}
+
+const struct rtl818x_rf_ops max2820_rf_ops = {
+	.name		= "Maxim",
+	.init		= max2820_rf_init,
+	.stop		= max2820_rf_stop,
+	.set_chan	= max2820_rf_set_channel
+};
diff --git a/drivers/net/wireless/rtl8180_max2820.h b/drivers/net/wireless/rtl8180_max2820.h
new file mode 100644
index 000000000000..61cf6d1e7d57
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_max2820.h
@@ -0,0 +1,28 @@
+#ifndef RTL8180_MAX2820_H
+#define RTL8180_MAX2820_H
+
+/*
+ * Radio tuning for Maxim max2820 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define MAXIM_ANTENNA 0xb3
+
+extern const struct rtl818x_rf_ops max2820_rf_ops;
+
+#endif /* RTL8180_MAX2820_H */
diff --git a/drivers/net/wireless/rtl8180_rtl8225.c b/drivers/net/wireless/rtl8180_rtl8225.c
new file mode 100644
index 000000000000..ef3832bee85c
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_rtl8225.c
@@ -0,0 +1,779 @@
+
+/*
+ * Radio tuning for RTL8225 on RTL8180
+ *
+ * Copyright 2007 Michael Wu <flamingice@sourmilk.net>
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Based on the r8180 driver, which is:
+ * Copyright 2005 Andrea Merello <andreamrl@tiscali.it>, et al.
+ *
+ * Thanks to Realtek for their support!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <net/mac80211.h>
+
+#include "rtl8180.h"
+#include "rtl8180_rtl8225.h"
+
+static void rtl8225_write(struct ieee80211_hw *dev, u8 addr, u16 data)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u16 reg80, reg84, reg82;
+	u32 bangdata;
+	int i;
+
+	bangdata = (data << 4) | (addr & 0xf);
+
+	reg80 = rtl818x_ioread16(priv, &priv->map->RFPinsOutput) & 0xfff3;
+	reg82 = rtl818x_ioread16(priv, &priv->map->RFPinsEnable);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, reg82 | 0x7);
+
+	reg84 = rtl818x_ioread16(priv, &priv->map->RFPinsSelect);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, reg84 | 0x7 | 0x400);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(10);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80 | (1 << 2));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(2);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(10);
+
+	for (i = 15; i >= 0; i--) {
+		u16 reg = reg80 | !!(bangdata & (1 << i));
+
+		if (i & 1)
+			rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg);
+
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg | (1 << 1));
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg | (1 << 1));
+
+		if (!(i & 1))
+			rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg);
+	}
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80 | (1 << 2));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(10);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80 | (1 << 2));
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, reg84 | 0x400);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+}
+
+static u16 rtl8225_read(struct ieee80211_hw *dev, u8 addr)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u16 reg80, reg82, reg84, out;
+	int i;
+
+	reg80 = rtl818x_ioread16(priv, &priv->map->RFPinsOutput);
+	reg82 = rtl818x_ioread16(priv, &priv->map->RFPinsEnable);
+	reg84 = rtl818x_ioread16(priv, &priv->map->RFPinsSelect) | 0x400;
+
+	reg80 &= ~0xF;
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, reg82 | 0x000F);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, reg84 | 0x000F);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80 | (1 << 2));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(4);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg80);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(5);
+
+	for (i = 4; i >= 0; i--) {
+		u16 reg = reg80 | ((addr >> i) & 1);
+
+		if (!(i & 1)) {
+			rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg);
+			rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+			udelay(1);
+		}
+
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg | (1 << 1));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg | (1 << 1));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+
+		if (i & 1) {
+			rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, reg);
+			rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+			udelay(1);
+		}
+	}
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x000E);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, 0x040E);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+			  reg80 | (1 << 3) | (1 << 1));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(2);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+			  reg80 | (1 << 3));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(2);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+			  reg80 | (1 << 3));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(2);
+
+	out = 0;
+	for (i = 11; i >= 0; i--) {
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg80 | (1 << 3));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(1);
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg80 | (1 << 3) | (1 << 1));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg80 | (1 << 3) | (1 << 1));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg80 | (1 << 3) | (1 << 1));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+
+		if (rtl818x_ioread16(priv, &priv->map->RFPinsInput) & (1 << 1))
+			out |= 1 << i;
+
+		rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+				  reg80 | (1 << 3));
+		rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+		udelay(2);
+	}
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput,
+			  reg80 | (1 << 3) | (1 << 2));
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	udelay(2);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, reg82);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, reg84);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, 0x03A0);
+
+	return out;
+}
+
+static const u16 rtl8225bcd_rxgain[] = {
+	0x0400, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0408, 0x0409,
+	0x040a, 0x040b, 0x0502, 0x0503, 0x0504, 0x0505, 0x0540, 0x0541,
+	0x0542, 0x0543, 0x0544, 0x0545, 0x0580, 0x0581, 0x0582, 0x0583,
+	0x0584, 0x0585, 0x0588, 0x0589, 0x058a, 0x058b, 0x0643, 0x0644,
+	0x0645, 0x0680, 0x0681, 0x0682, 0x0683, 0x0684, 0x0685, 0x0688,
+	0x0689, 0x068a, 0x068b, 0x068c, 0x0742, 0x0743, 0x0744, 0x0745,
+	0x0780, 0x0781, 0x0782, 0x0783, 0x0784, 0x0785, 0x0788, 0x0789,
+	0x078a, 0x078b, 0x078c, 0x078d, 0x0790, 0x0791, 0x0792, 0x0793,
+	0x0794, 0x0795, 0x0798, 0x0799, 0x079a, 0x079b, 0x079c, 0x079d,
+	0x07a0, 0x07a1, 0x07a2, 0x07a3, 0x07a4, 0x07a5, 0x07a8, 0x07a9,
+	0x07aa, 0x07ab, 0x07ac, 0x07ad, 0x07b0, 0x07b1, 0x07b2, 0x07b3,
+	0x07b4, 0x07b5, 0x07b8, 0x07b9, 0x07ba, 0x07bb, 0x07bb
+};
+
+static const u8 rtl8225_agc[] = {
+	0x9e, 0x9e, 0x9e, 0x9e, 0x9e, 0x9e, 0x9e, 0x9e,
+	0x9d, 0x9c, 0x9b, 0x9a, 0x99, 0x98, 0x97, 0x96,
+	0x95, 0x94, 0x93, 0x92, 0x91, 0x90, 0x8f, 0x8e,
+	0x8d, 0x8c, 0x8b, 0x8a, 0x89, 0x88, 0x87, 0x86,
+	0x85, 0x84, 0x83, 0x82, 0x81, 0x80, 0x3f, 0x3e,
+	0x3d, 0x3c, 0x3b, 0x3a, 0x39, 0x38, 0x37, 0x36,
+	0x35, 0x34, 0x33, 0x32, 0x31, 0x30, 0x2f, 0x2e,
+	0x2d, 0x2c, 0x2b, 0x2a, 0x29, 0x28, 0x27, 0x26,
+	0x25, 0x24, 0x23, 0x22, 0x21, 0x20, 0x1f, 0x1e,
+	0x1d, 0x1c, 0x1b, 0x1a, 0x19, 0x18, 0x17, 0x16,
+	0x15, 0x14, 0x13, 0x12, 0x11, 0x10, 0x0f, 0x0e,
+	0x0d, 0x0c, 0x0b, 0x0a, 0x09, 0x08, 0x07, 0x06,
+	0x05, 0x04, 0x03, 0x02, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01
+};
+
+static const u8 rtl8225_gain[] = {
+	0x23, 0x88, 0x7c, 0xa5, /* -82dbm */
+	0x23, 0x88, 0x7c, 0xb5, /* -82dbm */
+	0x23, 0x88, 0x7c, 0xc5, /* -82dbm */
+	0x33, 0x80, 0x79, 0xc5, /* -78dbm */
+	0x43, 0x78, 0x76, 0xc5, /* -74dbm */
+	0x53, 0x60, 0x73, 0xc5, /* -70dbm */
+	0x63, 0x58, 0x70, 0xc5, /* -66dbm */
+};
+
+static const u8 rtl8225_threshold[] = {
+	0x8d, 0x8d, 0x8d, 0x8d, 0x9d, 0xad, 0xbd
+};
+
+static const u8 rtl8225_tx_gain_cck_ofdm[] = {
+	0x02, 0x06, 0x0e, 0x1e, 0x3e, 0x7e
+};
+
+static const u8 rtl8225_tx_power_cck[] = {
+	0x18, 0x17, 0x15, 0x11, 0x0c, 0x08, 0x04, 0x02,
+	0x1b, 0x1a, 0x17, 0x13, 0x0e, 0x09, 0x04, 0x02,
+	0x1f, 0x1e, 0x1a, 0x15, 0x10, 0x0a, 0x05, 0x02,
+	0x22, 0x21, 0x1d, 0x18, 0x11, 0x0b, 0x06, 0x02,
+	0x26, 0x25, 0x21, 0x1b, 0x14, 0x0d, 0x06, 0x03,
+	0x2b, 0x2a, 0x25, 0x1e, 0x16, 0x0e, 0x07, 0x03
+};
+
+static const u8 rtl8225_tx_power_cck_ch14[] = {
+	0x18, 0x17, 0x15, 0x0c, 0x00, 0x00, 0x00, 0x00,
+	0x1b, 0x1a, 0x17, 0x0e, 0x00, 0x00, 0x00, 0x00,
+	0x1f, 0x1e, 0x1a, 0x0f, 0x00, 0x00, 0x00, 0x00,
+	0x22, 0x21, 0x1d, 0x11, 0x00, 0x00, 0x00, 0x00,
+	0x26, 0x25, 0x21, 0x13, 0x00, 0x00, 0x00, 0x00,
+	0x2b, 0x2a, 0x25, 0x15, 0x00, 0x00, 0x00, 0x00
+};
+
+static const u8 rtl8225_tx_power_ofdm[] = {
+	0x80, 0x90, 0xa2, 0xb5, 0xcb, 0xe4
+};
+
+static const u32 rtl8225_chan[] = {
+	0x085c, 0x08dc, 0x095c, 0x09dc, 0x0a5c, 0x0adc, 0x0b5c,
+	0x0bdc, 0x0c5c, 0x0cdc, 0x0d5c, 0x0ddc, 0x0e5c, 0x0f72
+};
+
+static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 cck_power, ofdm_power;
+	const u8 *tmp;
+	u32 reg;
+	int i;
+
+	cck_power = priv->channels[channel - 1].val & 0xFF;
+	ofdm_power = priv->channels[channel - 1].val >> 8;
+
+	cck_power = min(cck_power, (u8)35);
+	ofdm_power = min(ofdm_power, (u8)35);
+
+	rtl818x_iowrite8(priv, &priv->map->TX_GAIN_CCK,
+			 rtl8225_tx_gain_cck_ofdm[cck_power / 6] >> 1);
+
+	if (channel == 14)
+		tmp = &rtl8225_tx_power_cck_ch14[(cck_power % 6) * 8];
+	else
+		tmp = &rtl8225_tx_power_cck[(cck_power % 6) * 8];
+
+	for (i = 0; i < 8; i++)
+		rtl8225_write_phy_cck(dev, 0x44 + i, *tmp++);
+
+	msleep(1); /* FIXME: optional? */
+
+	/* anaparam2 on */
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	reg = rtl818x_ioread8(priv, &priv->map->CONFIG3);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg | RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite32(priv, &priv->map->ANAPARAM2, RTL8225_ANAPARAM2_ON);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg & ~RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	rtl818x_iowrite8(priv, &priv->map->TX_GAIN_OFDM,
+			 rtl8225_tx_gain_cck_ofdm[ofdm_power/6] >> 1);
+
+	tmp = &rtl8225_tx_power_ofdm[ofdm_power % 6];
+
+	rtl8225_write_phy_ofdm(dev, 5, *tmp);
+	rtl8225_write_phy_ofdm(dev, 7, *tmp);
+
+	msleep(1);
+}
+
+static void rtl8225_rf_init(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int i;
+
+	rtl8180_set_anaparam(priv, RTL8225_ANAPARAM_ON);
+
+	/* host_pci_init */
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, 0x0480);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, 0x0488);
+	rtl818x_iowrite8(priv, &priv->map->GP_ENABLE, 0);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	msleep(200);	/* FIXME: ehh?? */
+	rtl818x_iowrite8(priv, &priv->map->GP_ENABLE, 0xFF & ~(1 << 6));
+
+	rtl818x_iowrite32(priv, &priv->map->RF_TIMING, 0x000a8008);
+
+	/* TODO: check if we need really to change BRSR to do RF config */
+	rtl818x_ioread16(priv, &priv->map->BRSR);
+	rtl818x_iowrite16(priv, &priv->map->BRSR, 0xFFFF);
+	rtl818x_iowrite32(priv, &priv->map->RF_PARA, 0x00100044);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, 0x44);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	rtl8225_write(dev, 0x0, 0x067);
+	rtl8225_write(dev, 0x1, 0xFE0);
+	rtl8225_write(dev, 0x2, 0x44D);
+	rtl8225_write(dev, 0x3, 0x441);
+	rtl8225_write(dev, 0x4, 0x8BE);
+	rtl8225_write(dev, 0x5, 0xBF0);		/* TODO: minipci */
+	rtl8225_write(dev, 0x6, 0xAE6);
+	rtl8225_write(dev, 0x7, rtl8225_chan[0]);
+	rtl8225_write(dev, 0x8, 0x01F);
+	rtl8225_write(dev, 0x9, 0x334);
+	rtl8225_write(dev, 0xA, 0xFD4);
+	rtl8225_write(dev, 0xB, 0x391);
+	rtl8225_write(dev, 0xC, 0x050);
+	rtl8225_write(dev, 0xD, 0x6DB);
+	rtl8225_write(dev, 0xE, 0x029);
+	rtl8225_write(dev, 0xF, 0x914); msleep(1);
+
+	rtl8225_write(dev, 0x2, 0xC4D); msleep(100);
+
+	rtl8225_write(dev, 0x0, 0x127);
+
+	for (i = 0; i < ARRAY_SIZE(rtl8225bcd_rxgain); i++) {
+		rtl8225_write(dev, 0x1, i + 1);
+		rtl8225_write(dev, 0x2, rtl8225bcd_rxgain[i]);
+	}
+
+	rtl8225_write(dev, 0x0, 0x027);
+	rtl8225_write(dev, 0x0, 0x22F);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+
+	for (i = 0; i < ARRAY_SIZE(rtl8225_agc); i++) {
+		rtl8225_write_phy_ofdm(dev, 0xB, rtl8225_agc[i]);
+		msleep(1);
+		rtl8225_write_phy_ofdm(dev, 0xA, 0x80 + i);
+		msleep(1);
+	}
+
+	msleep(1);
+
+	rtl8225_write_phy_ofdm(dev, 0x00, 0x01); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x01, 0x02); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x02, 0x62); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x03, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x04, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x05, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x06, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x07, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x08, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x09, 0xfe); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0a, 0x09); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0b, 0x80); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0c, 0x01); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0e, 0xd3); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0f, 0x38); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x10, 0x84); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x11, 0x03); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x12, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x13, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x14, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x15, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x16, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x17, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x18, 0xef); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x19, 0x19); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1a, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1b, 0x76); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1c, 0x04); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1e, 0x95); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1f, 0x75); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x20, 0x1f); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x21, 0x27); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x22, 0x16); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x24, 0x46); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x25, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x26, 0x90); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x27, 0x88); msleep(1);
+
+	rtl8225_write_phy_cck(dev, 0x00, 0x98); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x03, 0x20); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x04, 0x7e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x05, 0x12); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x06, 0xfc); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x07, 0x78); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x08, 0x2e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x10, 0x93); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x11, 0x88); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x12, 0x47); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x13, 0xd0);
+	rtl8225_write_phy_cck(dev, 0x19, 0x00);
+	rtl8225_write_phy_cck(dev, 0x1a, 0xa0);
+	rtl8225_write_phy_cck(dev, 0x1b, 0x08);
+	rtl8225_write_phy_cck(dev, 0x40, 0x86);
+	rtl8225_write_phy_cck(dev, 0x41, 0x8d); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x42, 0x15); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x43, 0x18); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x44, 0x1f); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x45, 0x1e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x46, 0x1a); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x47, 0x15); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x48, 0x10); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x49, 0x0a); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4a, 0x05); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4b, 0x02); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4c, 0x05); msleep(1);
+
+	rtl818x_iowrite8(priv, &priv->map->TESTR, 0x0D); msleep(1);
+
+	rtl8225_rf_set_tx_power(dev, 1);
+
+	/* RX antenna default to A */
+	rtl8225_write_phy_cck(dev, 0x10, 0x9b); msleep(1);	/* B: 0xDB */
+	rtl8225_write_phy_ofdm(dev, 0x26, 0x90); msleep(1);	/* B: 0x10 */
+
+	rtl818x_iowrite8(priv, &priv->map->TX_ANTENNA, 0x03);	/* B: 0x00 */
+	msleep(1);
+	rtl818x_iowrite32(priv, (__le32 __iomem *)((void __iomem *)priv->map + 0x94), 0x15c00002);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+
+	rtl8225_write(dev, 0x0c, 0x50);
+	/* set OFDM initial gain */
+	rtl8225_write_phy_ofdm(dev, 0x0d, rtl8225_gain[4 * 4]);
+	rtl8225_write_phy_ofdm(dev, 0x23, rtl8225_gain[4 * 4 + 1]);
+	rtl8225_write_phy_ofdm(dev, 0x1b, rtl8225_gain[4 * 4 + 2]);
+	rtl8225_write_phy_ofdm(dev, 0x1d, rtl8225_gain[4 * 4 + 3]);
+	/* set CCK threshold */
+	rtl8225_write_phy_cck(dev, 0x41, rtl8225_threshold[0]);
+}
+
+static const u8 rtl8225z2_tx_power_cck_ch14[] = {
+	0x36, 0x35, 0x2e, 0x1b, 0x00, 0x00, 0x00, 0x00
+};
+
+static const u8 rtl8225z2_tx_power_cck_B[] = {
+	0x30, 0x2f, 0x29, 0x21, 0x19, 0x10, 0x08, 0x04
+};
+
+static const u8 rtl8225z2_tx_power_cck_A[] = {
+	0x33, 0x32, 0x2b, 0x23, 0x1a, 0x11, 0x08, 0x04
+};
+
+static const u8 rtl8225z2_tx_power_cck[] = {
+	0x36, 0x35, 0x2e, 0x25, 0x1c, 0x12, 0x09, 0x04
+};
+
+static void rtl8225z2_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 cck_power, ofdm_power;
+	const u8 *tmp;
+	int i;
+
+	cck_power = priv->channels[channel - 1].val & 0xFF;
+	ofdm_power = priv->channels[channel - 1].val >> 8;
+
+	if (channel == 14)
+		tmp = rtl8225z2_tx_power_cck_ch14;
+	else if (cck_power == 12)
+		tmp = rtl8225z2_tx_power_cck_B;
+	else if (cck_power == 13)
+		tmp = rtl8225z2_tx_power_cck_A;
+	else
+		tmp = rtl8225z2_tx_power_cck;
+
+	for (i = 0; i < 8; i++)
+		rtl8225_write_phy_cck(dev, 0x44 + i, *tmp++);
+
+	cck_power = min(cck_power, (u8)35);
+	if (cck_power == 13 || cck_power == 14)
+		cck_power = 12;
+	if (cck_power >= 15)
+		cck_power -= 2;
+
+	rtl818x_iowrite8(priv, &priv->map->TX_GAIN_CCK, cck_power);
+	rtl818x_ioread8(priv, &priv->map->TX_GAIN_CCK);
+	msleep(1);
+
+	ofdm_power = min(ofdm_power, (u8)35);
+	rtl818x_iowrite8(priv, &priv->map->TX_GAIN_OFDM, ofdm_power);
+
+	rtl8225_write_phy_ofdm(dev, 2, 0x62);
+	rtl8225_write_phy_ofdm(dev, 5, 0x00);
+	rtl8225_write_phy_ofdm(dev, 6, 0x40);
+	rtl8225_write_phy_ofdm(dev, 7, 0x00);
+	rtl8225_write_phy_ofdm(dev, 8, 0x40);
+
+	msleep(1);
+}
+
+static const u16 rtl8225z2_rxgain[] = {
+	0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0008, 0x0009,
+	0x000a, 0x000b, 0x0102, 0x0103, 0x0104, 0x0105, 0x0140, 0x0141,
+	0x0142, 0x0143, 0x0144, 0x0145, 0x0180, 0x0181, 0x0182, 0x0183,
+	0x0184, 0x0185, 0x0188, 0x0189, 0x018a, 0x018b, 0x0243, 0x0244,
+	0x0245, 0x0280, 0x0281, 0x0282, 0x0283, 0x0284, 0x0285, 0x0288,
+	0x0289, 0x028a, 0x028b, 0x028c, 0x0342, 0x0343, 0x0344, 0x0345,
+	0x0380, 0x0381, 0x0382, 0x0383, 0x0384, 0x0385, 0x0388, 0x0389,
+	0x038a, 0x038b, 0x038c, 0x038d, 0x0390, 0x0391, 0x0392, 0x0393,
+	0x0394, 0x0395, 0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d,
+	0x03a0, 0x03a1, 0x03a2, 0x03a3, 0x03a4, 0x03a5, 0x03a8, 0x03a9,
+	0x03aa, 0x03ab, 0x03ac, 0x03ad, 0x03b0, 0x03b1, 0x03b2, 0x03b3,
+	0x03b4, 0x03b5, 0x03b8, 0x03b9, 0x03ba, 0x03bb, 0x03bb
+};
+
+static void rtl8225z2_rf_init(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	int i;
+
+	rtl8180_set_anaparam(priv, RTL8225_ANAPARAM_ON);
+
+	/* host_pci_init */
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, 0x0480);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, 0x0488);
+	rtl818x_iowrite8(priv, &priv->map->GP_ENABLE, 0);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	msleep(200);	/* FIXME: ehh?? */
+	rtl818x_iowrite8(priv, &priv->map->GP_ENABLE, 0xFF & ~(1 << 6));
+
+	rtl818x_iowrite32(priv, &priv->map->RF_TIMING, 0x00088008);
+
+	/* TODO: check if we need really to change BRSR to do RF config */
+	rtl818x_ioread16(priv, &priv->map->BRSR);
+	rtl818x_iowrite16(priv, &priv->map->BRSR, 0xFFFF);
+	rtl818x_iowrite32(priv, &priv->map->RF_PARA, 0x00100044);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, 0x44);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+
+	rtl8225_write(dev, 0x0, 0x0B7); msleep(1);
+	rtl8225_write(dev, 0x1, 0xEE0); msleep(1);
+	rtl8225_write(dev, 0x2, 0x44D); msleep(1);
+	rtl8225_write(dev, 0x3, 0x441); msleep(1);
+	rtl8225_write(dev, 0x4, 0x8C3); msleep(1);
+	rtl8225_write(dev, 0x5, 0xC72); msleep(1);
+	rtl8225_write(dev, 0x6, 0x0E6); msleep(1);
+	rtl8225_write(dev, 0x7, 0x82A); msleep(1);
+	rtl8225_write(dev, 0x8, 0x03F); msleep(1);
+	rtl8225_write(dev, 0x9, 0x335); msleep(1);
+	rtl8225_write(dev, 0xa, 0x9D4); msleep(1);
+	rtl8225_write(dev, 0xb, 0x7BB); msleep(1);
+	rtl8225_write(dev, 0xc, 0x850); msleep(1);
+	rtl8225_write(dev, 0xd, 0xCDF); msleep(1);
+	rtl8225_write(dev, 0xe, 0x02B); msleep(1);
+	rtl8225_write(dev, 0xf, 0x114); msleep(100);
+
+	if (!(rtl8225_read(dev, 6) & (1 << 7))) {
+		rtl8225_write(dev, 0x02, 0x0C4D);
+		msleep(200);
+		rtl8225_write(dev, 0x02, 0x044D);
+		msleep(100);
+		/* TODO: readd calibration failure message when the calibration
+		   check works */
+	}
+
+	rtl8225_write(dev, 0x0, 0x1B7);
+	rtl8225_write(dev, 0x3, 0x002);
+	rtl8225_write(dev, 0x5, 0x004);
+
+	for (i = 0; i < ARRAY_SIZE(rtl8225z2_rxgain); i++) {
+		rtl8225_write(dev, 0x1, i + 1);
+		rtl8225_write(dev, 0x2, rtl8225z2_rxgain[i]);
+	}
+
+	rtl8225_write(dev, 0x0, 0x0B7); msleep(100);
+	rtl8225_write(dev, 0x2, 0xC4D);
+
+	msleep(200);
+	rtl8225_write(dev, 0x2, 0x44D);
+	msleep(100);
+
+	rtl8225_write(dev, 0x00, 0x2BF);
+	rtl8225_write(dev, 0xFF, 0xFFFF);
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+
+	for (i = 0; i < ARRAY_SIZE(rtl8225_agc); i++) {
+		rtl8225_write_phy_ofdm(dev, 0xB, rtl8225_agc[i]);
+		msleep(1);
+		rtl8225_write_phy_ofdm(dev, 0xA, 0x80 + i);
+		msleep(1);
+	}
+
+	msleep(1);
+
+	rtl8225_write_phy_ofdm(dev, 0x00, 0x01); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x01, 0x02); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x02, 0x62); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x03, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x04, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x05, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x06, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x07, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x08, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x09, 0xfe); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0a, 0x09); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x18, 0xef); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0b, 0x80); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0c, 0x01); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0d, 0x43);
+	rtl8225_write_phy_ofdm(dev, 0x0e, 0xd3); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x0f, 0x38); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x10, 0x84); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x11, 0x06); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x12, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x13, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x14, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x15, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x16, 0x00); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x17, 0x40); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x18, 0xef); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x19, 0x19); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1a, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1b, 0x11); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1c, 0x04); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1d, 0xc5); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1e, 0xb3); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x1f, 0x75); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x20, 0x1f); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x21, 0x27); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x22, 0x16); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x23, 0x80); msleep(1); /* FIXME: not needed? */
+	rtl8225_write_phy_ofdm(dev, 0x24, 0x46); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x25, 0x20); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x26, 0x90); msleep(1);
+	rtl8225_write_phy_ofdm(dev, 0x27, 0x88); msleep(1);
+
+	rtl8225_write_phy_cck(dev, 0x00, 0x98); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x03, 0x20); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x04, 0x7e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x05, 0x12); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x06, 0xfc); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x07, 0x78); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x08, 0x2e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x10, 0x93); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x11, 0x88); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x12, 0x47); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x13, 0xd0);
+	rtl8225_write_phy_cck(dev, 0x19, 0x00);
+	rtl8225_write_phy_cck(dev, 0x1a, 0xa0);
+	rtl8225_write_phy_cck(dev, 0x1b, 0x08);
+	rtl8225_write_phy_cck(dev, 0x40, 0x86);
+	rtl8225_write_phy_cck(dev, 0x41, 0x8a); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x42, 0x15); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x43, 0x18); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x44, 0x36); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x45, 0x35); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x46, 0x2e); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x47, 0x25); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x48, 0x1c); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x49, 0x12); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4a, 0x09); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4b, 0x04); msleep(1);
+	rtl8225_write_phy_cck(dev, 0x4c, 0x05); msleep(1);
+
+	rtl818x_iowrite8(priv, (u8 __iomem *)((void __iomem *)priv->map + 0x5B), 0x0D); msleep(1);
+
+	rtl8225z2_rf_set_tx_power(dev, 1);
+
+	/* RX antenna default to A */
+	rtl8225_write_phy_cck(dev, 0x10, 0x9b); msleep(1);	/* B: 0xDB */
+	rtl8225_write_phy_ofdm(dev, 0x26, 0x90); msleep(1);	/* B: 0x10 */
+
+	rtl818x_iowrite8(priv, &priv->map->TX_ANTENNA, 0x03);	/* B: 0x00 */
+	msleep(1);
+	rtl818x_iowrite32(priv, (__le32 __iomem *)((void __iomem *)priv->map + 0x94), 0x15c00002);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+}
+
+static void rtl8225_rf_stop(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 reg;
+
+	rtl8225_write(dev, 0x4, 0x1f); msleep(1);
+
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
+	reg = rtl818x_ioread8(priv, &priv->map->CONFIG3);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg | RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite32(priv, &priv->map->ANAPARAM2, RTL8225_ANAPARAM2_OFF);
+	rtl818x_iowrite32(priv, &priv->map->ANAPARAM, RTL8225_ANAPARAM_OFF);
+	rtl818x_iowrite8(priv, &priv->map->CONFIG3, reg & ~RTL818X_CONFIG3_ANAPARAM_WRITE);
+	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
+}
+
+static void rtl8225_rf_set_channel(struct ieee80211_hw *dev,
+				   struct ieee80211_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+
+	if (priv->rf->init == rtl8225_rf_init)
+		rtl8225_rf_set_tx_power(dev, conf->channel);
+	else
+		rtl8225z2_rf_set_tx_power(dev, conf->channel);
+
+	rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]);
+	msleep(10);
+
+	if (conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME) {
+		rtl818x_iowrite8(priv, &priv->map->SLOT, 0x9);
+		rtl818x_iowrite8(priv, &priv->map->SIFS, 0x22);
+		rtl818x_iowrite8(priv, &priv->map->DIFS, 0x14);
+		rtl818x_iowrite8(priv, &priv->map->EIFS, 81);
+		rtl818x_iowrite8(priv, &priv->map->CW_VAL, 0x73);
+	} else {
+		rtl818x_iowrite8(priv, &priv->map->SLOT, 0x14);
+		rtl818x_iowrite8(priv, &priv->map->SIFS, 0x44);
+		rtl818x_iowrite8(priv, &priv->map->DIFS, 0x24);
+		rtl818x_iowrite8(priv, &priv->map->EIFS, 81);
+		rtl818x_iowrite8(priv, &priv->map->CW_VAL, 0xa5);
+	}
+}
+
+static const struct rtl818x_rf_ops rtl8225_ops = {
+	.name		= "rtl8225",
+	.init		= rtl8225_rf_init,
+	.stop		= rtl8225_rf_stop,
+	.set_chan	= rtl8225_rf_set_channel
+};
+
+static const struct rtl818x_rf_ops rtl8225z2_ops = {
+	.name		= "rtl8225z2",
+	.init		= rtl8225z2_rf_init,
+	.stop		= rtl8225_rf_stop,
+	.set_chan	= rtl8225_rf_set_channel
+};
+
+const struct rtl818x_rf_ops * rtl8180_detect_rf(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u16 reg8, reg9;
+
+	rtl818x_iowrite16(priv, &priv->map->RFPinsOutput, 0x0480);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsSelect, 0x0488);
+	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FFF);
+	rtl818x_ioread8(priv, &priv->map->EEPROM_CMD);
+	msleep(100);
+
+	rtl8225_write(dev, 0, 0x1B7);
+
+	reg8 = rtl8225_read(dev, 8);
+	reg9 = rtl8225_read(dev, 9);
+
+	rtl8225_write(dev, 0, 0x0B7);
+
+	if (reg8 != 0x588 || reg9 != 0x700)
+		return &rtl8225_ops;
+
+	return &rtl8225z2_ops;
+}
diff --git a/drivers/net/wireless/rtl8180_rtl8225.h b/drivers/net/wireless/rtl8180_rtl8225.h
new file mode 100644
index 000000000000..310013a2d726
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_rtl8225.h
@@ -0,0 +1,23 @@
+#ifndef RTL8180_RTL8225_H
+#define RTL8180_RTL8225_H
+
+#define RTL8225_ANAPARAM_ON	0xa0000b59
+#define RTL8225_ANAPARAM2_ON	0x860dec11
+#define RTL8225_ANAPARAM_OFF	0xa00beb59
+#define RTL8225_ANAPARAM2_OFF	0x840dec11
+
+const struct rtl818x_rf_ops * rtl8180_detect_rf(struct ieee80211_hw *);
+
+static inline void rtl8225_write_phy_ofdm(struct ieee80211_hw *dev,
+					  u8 addr, u8 data)
+{
+	rtl8180_write_phy(dev, addr, data);
+}
+
+static inline void rtl8225_write_phy_cck(struct ieee80211_hw *dev,
+					 u8 addr, u8 data)
+{
+	rtl8180_write_phy(dev, addr, data | 0x10000);
+}
+
+#endif /* RTL8180_RTL8225_H */
diff --git a/drivers/net/wireless/rtl8180_sa2400.c b/drivers/net/wireless/rtl8180_sa2400.c
new file mode 100644
index 000000000000..e08ace7b1cb7
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_sa2400.c
@@ -0,0 +1,201 @@
+
+/*
+ * Radio tuning for Philips SA2400 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <net/mac80211.h>
+
+#include "rtl8180.h"
+#include "rtl8180_sa2400.h"
+
+static const u32 sa2400_chan[] = {
+	0x00096c, /* ch1 */
+	0x080970,
+	0x100974,
+	0x180978,
+	0x000980,
+	0x080984,
+	0x100988,
+	0x18098c,
+	0x000994,
+	0x080998,
+	0x10099c,
+	0x1809a0,
+	0x0009a8,
+	0x0009b4, /* ch 14 */
+};
+
+static void write_sa2400(struct ieee80211_hw *dev, u8 addr, u32 data)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 phy_config;
+
+	/* MAC will bang bits to the sa2400. sw 3-wire is NOT used */
+	phy_config = 0xb0000000;
+
+	phy_config |= ((u32)(addr & 0xf)) << 24;
+	phy_config |= data & 0xffffff;
+
+	rtl818x_iowrite32(priv,
+		(__le32 __iomem *) &priv->map->RFPinsOutput, phy_config);
+
+	msleep(3);
+}
+
+static void sa2400_write_phy_antenna(struct ieee80211_hw *dev, short chan)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u8 ant = SA2400_ANTENNA;
+
+	if (priv->rfparam & RF_PARAM_ANTBDEFAULT)
+		ant |= BB_ANTENNA_B;
+
+	if (chan == 14)
+		ant |= BB_ANTATTEN_CHAN14;
+
+	rtl8180_write_phy(dev, 0x10, ant);
+
+}
+
+static void sa2400_rf_set_channel(struct ieee80211_hw *dev,
+				  struct ieee80211_conf *conf)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 txpw = priv->channels[conf->channel - 1].val & 0xFF;
+	u32 chan = sa2400_chan[conf->channel - 1];
+
+	write_sa2400(dev, 7, txpw);
+
+	sa2400_write_phy_antenna(dev, chan);
+
+	write_sa2400(dev, 0, chan);
+	write_sa2400(dev, 1, 0xbb50);
+	write_sa2400(dev, 2, 0x80);
+	write_sa2400(dev, 3, 0);
+}
+
+static void sa2400_rf_stop(struct ieee80211_hw *dev)
+{
+	write_sa2400(dev, 4, 0);
+}
+
+static void sa2400_rf_init(struct ieee80211_hw *dev)
+{
+	struct rtl8180_priv *priv = dev->priv;
+	u32 anaparam, txconf;
+	u8 firdac;
+	int analogphy = priv->rfparam & RF_PARAM_ANALOGPHY;
+
+	anaparam = priv->anaparam;
+	anaparam &= ~(1 << ANAPARAM_TXDACOFF_SHIFT);
+	anaparam &= ~ANAPARAM_PWR1_MASK;
+	anaparam &= ~ANAPARAM_PWR0_MASK;
+
+	if (analogphy) {
+		anaparam |= SA2400_ANA_ANAPARAM_PWR1_ON << ANAPARAM_PWR1_SHIFT;
+		firdac = 0;
+	} else {
+		anaparam |= (SA2400_DIG_ANAPARAM_PWR1_ON << ANAPARAM_PWR1_SHIFT);
+		anaparam |= (SA2400_ANAPARAM_PWR0_ON << ANAPARAM_PWR0_SHIFT);
+		firdac = 1 << SA2400_REG4_FIRDAC_SHIFT;
+	}
+
+	rtl8180_set_anaparam(priv, anaparam);
+
+	write_sa2400(dev, 0, sa2400_chan[0]);
+	write_sa2400(dev, 1, 0xbb50);
+	write_sa2400(dev, 2, 0x80);
+	write_sa2400(dev, 3, 0);
+	write_sa2400(dev, 4, 0x19340 | firdac);
+	write_sa2400(dev, 5, 0x1dfb | (SA2400_MAX_SENS - 54) << 15);
+	write_sa2400(dev, 4, 0x19348 | firdac); /* calibrate VCO */
+
+	if (!analogphy)
+		write_sa2400(dev, 4, 0x1938c); /*???*/
+
+	write_sa2400(dev, 4, 0x19340 | firdac);
+
+	write_sa2400(dev, 0, sa2400_chan[0]);
+	write_sa2400(dev, 1, 0xbb50);
+	write_sa2400(dev, 2, 0x80);
+	write_sa2400(dev, 3, 0);
+	write_sa2400(dev, 4, 0x19344 | firdac); /* calibrate filter */
+
+	/* new from rtl8180 embedded driver (rtl8181 project) */
+	write_sa2400(dev, 6, 0x13ff | (1 << 23)); /* MANRX */
+	write_sa2400(dev, 8, 0); /* VCO */
+
+	if (analogphy) {
+		rtl8180_set_anaparam(priv, anaparam |
+				     (1 << ANAPARAM_TXDACOFF_SHIFT));
+
+		txconf = rtl818x_ioread32(priv, &priv->map->TX_CONF);
+		rtl818x_iowrite32(priv, &priv->map->TX_CONF,
+			txconf | RTL818X_TX_CONF_LOOPBACK_CONT);
+
+		write_sa2400(dev, 4, 0x19341); /* calibrates DC */
+
+		/* a 5us sleep is required here,
+		 * we rely on the 3ms delay introduced in write_sa2400 */
+		write_sa2400(dev, 4, 0x19345);
+
+		/* a 20us sleep is required here,
+		 * we rely on the 3ms delay introduced in write_sa2400 */
+
+		rtl818x_iowrite32(priv, &priv->map->TX_CONF, txconf);
+
+		rtl8180_set_anaparam(priv, anaparam);
+	}
+	/* end new code */
+
+	write_sa2400(dev, 4, 0x19341 | firdac); /* RTX MODE */
+
+	/* baseband configuration */
+	rtl8180_write_phy(dev, 0, 0x98);
+	rtl8180_write_phy(dev, 3, 0x38);
+	rtl8180_write_phy(dev, 4, 0xe0);
+	rtl8180_write_phy(dev, 5, 0x90);
+	rtl8180_write_phy(dev, 6, 0x1a);
+	rtl8180_write_phy(dev, 7, 0x64);
+
+	sa2400_write_phy_antenna(dev, 1);
+
+	rtl8180_write_phy(dev, 0x11, 0x80);
+
+	if (rtl818x_ioread8(priv, &priv->map->CONFIG2) &
+	    RTL818X_CONFIG2_ANTENNA_DIV)
+		rtl8180_write_phy(dev, 0x12, 0xc7); /* enable ant diversity */
+	else
+		rtl8180_write_phy(dev, 0x12, 0x47); /* disable ant diversity */
+
+	rtl8180_write_phy(dev, 0x13, 0x90 | priv->csthreshold);
+
+	rtl8180_write_phy(dev, 0x19, 0x0);
+	rtl8180_write_phy(dev, 0x1a, 0xa0);
+}
+
+const struct rtl818x_rf_ops sa2400_rf_ops = {
+	.name		= "Philips",
+	.init		= sa2400_rf_init,
+	.stop		= sa2400_rf_stop,
+	.set_chan	= sa2400_rf_set_channel
+};
diff --git a/drivers/net/wireless/rtl8180_sa2400.h b/drivers/net/wireless/rtl8180_sa2400.h
new file mode 100644
index 000000000000..a4aaa0d413f1
--- /dev/null
+++ b/drivers/net/wireless/rtl8180_sa2400.h
@@ -0,0 +1,36 @@
+#ifndef RTL8180_SA2400_H
+#define RTL8180_SA2400_H
+
+/*
+ * Radio tuning for Philips SA2400 on RTL8180
+ *
+ * Copyright 2007 Andrea Merello <andreamrl@tiscali.it>
+ *
+ * Code from the BSD driver and the rtl8181 project have been
+ * very useful to understand certain things
+ *
+ * I want to thanks the Authors of such projects and the Ndiswrapper
+ * project Authors.
+ *
+ * A special Big Thanks also is for all people who donated me cards,
+ * making possible the creation of the original rtl8180 driver
+ * from which this code is derived!
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define SA2400_ANTENNA 0x91
+#define SA2400_DIG_ANAPARAM_PWR1_ON 0x8
+#define SA2400_ANA_ANAPARAM_PWR1_ON 0x28
+#define SA2400_ANAPARAM_PWR0_ON 0x3
+
+/* RX sensitivity in dbm */
+#define SA2400_MAX_SENS 85
+
+#define SA2400_REG4_FIRDAC_SHIFT 7
+
+extern const struct rtl818x_rf_ops sa2400_rf_ops;
+
+#endif /* RTL8180_SA2400_H */
diff --git a/drivers/net/wireless/rtl8187.h b/drivers/net/wireless/rtl8187.h
index 6ad322ef0da1..26a4f1097b4a 100644
--- a/drivers/net/wireless/rtl8187.h
+++ b/drivers/net/wireless/rtl8187.h
@@ -64,7 +64,7 @@ struct rtl8187_tx_hdr {
 struct rtl8187_priv {
 	/* common between rtl818x drivers */
 	struct rtl818x_csr *map;
-	void (*rf_init)(struct ieee80211_hw *);
+	const struct rtl818x_rf_ops *rf;
 	int mode;
 	int if_id;
 
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index 09e48d3bef36..8dab25a65990 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -393,37 +393,19 @@ static int rtl8187_init_hw(struct ieee80211_hw *dev)
 	rtl818x_iowrite16(priv, &priv->map->RFPinsEnable, 0x1FF7);
 	msleep(100);
 
-	priv->rf_init(dev);
+	priv->rf->init(dev);
 
 	rtl818x_iowrite16(priv, &priv->map->BRSR, 0x01F3);
-	reg = rtl818x_ioread16(priv, &priv->map->PGSELECT) & 0xfffe;
-	rtl818x_iowrite16(priv, &priv->map->PGSELECT, reg | 0x1);
+	reg = rtl818x_ioread8(priv, &priv->map->PGSELECT) & ~1;
+	rtl818x_iowrite8(priv, &priv->map->PGSELECT, reg | 1);
 	rtl818x_iowrite16(priv, (__le16 *)0xFFFE, 0x10);
 	rtl818x_iowrite8(priv, &priv->map->TALLY_SEL, 0x80);
 	rtl818x_iowrite8(priv, (u8 *)0xFFFF, 0x60);
-	rtl818x_iowrite16(priv, &priv->map->PGSELECT, reg);
+	rtl818x_iowrite8(priv, &priv->map->PGSELECT, reg);
 
 	return 0;
 }
 
-static void rtl8187_set_channel(struct ieee80211_hw *dev, int channel)
-{
-	u32 reg;
-	struct rtl8187_priv *priv = dev->priv;
-
-	reg = rtl818x_ioread32(priv, &priv->map->TX_CONF);
-	/* Enable TX loopback on MAC level to avoid TX during channel
-	 * changes, as this has be seen to causes problems and the
-	 * card will stop work until next reset
-	 */
-	rtl818x_iowrite32(priv, &priv->map->TX_CONF,
-			  reg | RTL818X_TX_CONF_LOOPBACK_MAC);
-	msleep(10);
-	rtl8225_rf_set_channel(dev, channel);
-	msleep(10);
-	rtl818x_iowrite32(priv, &priv->map->TX_CONF, reg);
-}
-
 static int rtl8187_start(struct ieee80211_hw *dev)
 {
 	struct rtl8187_priv *priv = dev->priv;
@@ -492,7 +474,7 @@ static void rtl8187_stop(struct ieee80211_hw *dev)
 	reg &= ~RTL818X_CMD_RX_ENABLE;
 	rtl818x_iowrite8(priv, &priv->map->CMD, reg);
 
-	rtl8225_rf_stop(dev);
+	priv->rf->stop(dev);
 
 	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
 	reg = rtl818x_ioread8(priv, &priv->map->CONFIG4);
@@ -543,7 +525,19 @@ static void rtl8187_remove_interface(struct ieee80211_hw *dev,
 static int rtl8187_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 {
 	struct rtl8187_priv *priv = dev->priv;
-	rtl8187_set_channel(dev, conf->channel);
+	u32 reg;
+
+	reg = rtl818x_ioread32(priv, &priv->map->TX_CONF);
+	/* Enable TX loopback on MAC level to avoid TX during channel
+	 * changes, as this has be seen to causes problems and the
+	 * card will stop work until next reset
+	 */
+	rtl818x_iowrite32(priv, &priv->map->TX_CONF,
+			  reg | RTL818X_TX_CONF_LOOPBACK_MAC);
+	msleep(10);
+	priv->rf->set_chan(dev, conf);
+	msleep(10);
+	rtl818x_iowrite32(priv, &priv->map->TX_CONF, reg);
 
 	rtl818x_iowrite8(priv, &priv->map->SIFS, 0x22);
 
@@ -753,23 +747,16 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 	eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_BASE,
 			  &priv->txpwr_base);
 
-	reg = rtl818x_ioread16(priv, &priv->map->PGSELECT) & ~1;
-	rtl818x_iowrite16(priv, &priv->map->PGSELECT, reg | 1);
+	reg = rtl818x_ioread8(priv, &priv->map->PGSELECT) & ~1;
+	rtl818x_iowrite8(priv, &priv->map->PGSELECT, reg | 1);
 	/* 0 means asic B-cut, we should use SW 3 wire
 	 * bit-by-bit banging for radio. 1 means we can use
 	 * USB specific request to write radio registers */
 	priv->asic_rev = rtl818x_ioread8(priv, (u8 *)0xFFFE) & 0x3;
-	rtl818x_iowrite16(priv, &priv->map->PGSELECT, reg);
+	rtl818x_iowrite8(priv, &priv->map->PGSELECT, reg);
 	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
 
-	rtl8225_write(dev, 0, 0x1B7);
-
-	if (rtl8225_read(dev, 8) != 0x588 || rtl8225_read(dev, 9) != 0x700)
-		priv->rf_init = rtl8225_rf_init;
-	else
-		priv->rf_init = rtl8225z2_rf_init;
-
-	rtl8225_write(dev, 0, 0x0B7);
+	priv->rf = rtl8187_detect_rf(dev);
 
 	err = ieee80211_register_hw(dev);
 	if (err) {
@@ -779,8 +766,7 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 
 	printk(KERN_INFO "%s: hwaddr %s, rtl8187 V%d + %s\n",
 	       wiphy_name(dev->wiphy), print_mac(mac, dev->wiphy->perm_addr),
-	       priv->asic_rev, priv->rf_init == rtl8225_rf_init ?
-	       "rtl8225" : "rtl8225z2");
+	       priv->asic_rev, priv->rf->name);
 
 	return 0;
 
diff --git a/drivers/net/wireless/rtl8187_rtl8225.c b/drivers/net/wireless/rtl8187_rtl8225.c
index efc41207780e..b713de17ba0a 100644
--- a/drivers/net/wireless/rtl8187_rtl8225.c
+++ b/drivers/net/wireless/rtl8187_rtl8225.c
@@ -101,7 +101,7 @@ static void rtl8225_write_8051(struct ieee80211_hw *dev, u8 addr, __le16 data)
 	msleep(2);
 }
 
-void rtl8225_write(struct ieee80211_hw *dev, u8 addr, u16 data)
+static void rtl8225_write(struct ieee80211_hw *dev, u8 addr, u16 data)
 {
 	struct rtl8187_priv *priv = dev->priv;
 
@@ -111,7 +111,7 @@ void rtl8225_write(struct ieee80211_hw *dev, u8 addr, u16 data)
 		rtl8225_write_bitbang(dev, addr, data);
 }
 
-u16 rtl8225_read(struct ieee80211_hw *dev, u8 addr)
+static u16 rtl8225_read(struct ieee80211_hw *dev, u8 addr)
 {
 	struct rtl8187_priv *priv = dev->priv;
 	u16 reg80, reg82, reg84, out;
@@ -325,7 +325,7 @@ static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
 	msleep(1);
 }
 
-void rtl8225_rf_init(struct ieee80211_hw *dev)
+static void rtl8225_rf_init(struct ieee80211_hw *dev)
 {
 	struct rtl8187_priv *priv = dev->priv;
 	int i;
@@ -567,7 +567,7 @@ static const u8 rtl8225z2_gain_bg[] = {
 	0x63, 0x15, 0xc5  /* -66dBm */
 };
 
-void rtl8225z2_rf_init(struct ieee80211_hw *dev)
+static void rtl8225z2_rf_init(struct ieee80211_hw *dev)
 {
 	struct rtl8187_priv *priv = dev->priv;
 	int i;
@@ -715,7 +715,7 @@ void rtl8225z2_rf_init(struct ieee80211_hw *dev)
 	rtl818x_iowrite32(priv, (__le32 *)0xFF94, 0x3dc00002);
 }
 
-void rtl8225_rf_stop(struct ieee80211_hw *dev)
+static void rtl8225_rf_stop(struct ieee80211_hw *dev)
 {
 	u8 reg;
 	struct rtl8187_priv *priv = dev->priv;
@@ -731,15 +731,47 @@ void rtl8225_rf_stop(struct ieee80211_hw *dev)
 	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL);
 }
 
-void rtl8225_rf_set_channel(struct ieee80211_hw *dev, int channel)
+static void rtl8225_rf_set_channel(struct ieee80211_hw *dev,
+				   struct ieee80211_conf *conf)
 {
 	struct rtl8187_priv *priv = dev->priv;
 
-	if (priv->rf_init == rtl8225_rf_init)
-		rtl8225_rf_set_tx_power(dev, channel);
+	if (priv->rf->init == rtl8225_rf_init)
+		rtl8225_rf_set_tx_power(dev, conf->channel);
 	else
-		rtl8225z2_rf_set_tx_power(dev, channel);
+		rtl8225z2_rf_set_tx_power(dev, conf->channel);
 
-	rtl8225_write(dev, 0x7, rtl8225_chan[channel - 1]);
+	rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]);
 	msleep(10);
 }
+
+static const struct rtl818x_rf_ops rtl8225_ops = {
+	.name		= "rtl8225",
+	.init		= rtl8225_rf_init,
+	.stop		= rtl8225_rf_stop,
+	.set_chan	= rtl8225_rf_set_channel
+};
+
+static const struct rtl818x_rf_ops rtl8225z2_ops = {
+	.name		= "rtl8225z2",
+	.init		= rtl8225z2_rf_init,
+	.stop		= rtl8225_rf_stop,
+	.set_chan	= rtl8225_rf_set_channel
+};
+
+const struct rtl818x_rf_ops * rtl8187_detect_rf(struct ieee80211_hw *dev)
+{
+	u16 reg8, reg9;
+
+	rtl8225_write(dev, 0, 0x1B7);
+
+	reg8 = rtl8225_read(dev, 8);
+	reg9 = rtl8225_read(dev, 9);
+
+	rtl8225_write(dev, 0, 0x0B7);
+
+	if (reg8 != 0x588 || reg9 != 0x700)
+		return &rtl8225_ops;
+
+	return &rtl8225z2_ops;
+}
diff --git a/drivers/net/wireless/rtl8187_rtl8225.h b/drivers/net/wireless/rtl8187_rtl8225.h
index 798ba4a97376..d39ed0295b6e 100644
--- a/drivers/net/wireless/rtl8187_rtl8225.h
+++ b/drivers/net/wireless/rtl8187_rtl8225.h
@@ -20,14 +20,7 @@
 #define RTL8225_ANAPARAM_OFF	0xa00beb59
 #define RTL8225_ANAPARAM2_OFF	0x840dec11
 
-void rtl8225_write(struct ieee80211_hw *, u8 addr, u16 data);
-u16  rtl8225_read(struct ieee80211_hw *, u8 addr);
-
-void rtl8225_rf_init(struct ieee80211_hw *);
-void rtl8225z2_rf_init(struct ieee80211_hw *);
-void rtl8225_rf_stop(struct ieee80211_hw *);
-void rtl8225_rf_set_channel(struct ieee80211_hw *, int);
-
+const struct rtl818x_rf_ops * rtl8187_detect_rf(struct ieee80211_hw *);
 
 static inline void rtl8225_write_phy_ofdm(struct ieee80211_hw *dev,
 					  u8 addr, u32 data)
diff --git a/drivers/net/wireless/rtl818x.h b/drivers/net/wireless/rtl818x.h
index 880d4becae31..1e7d6f8278d7 100644
--- a/drivers/net/wireless/rtl818x.h
+++ b/drivers/net/wireless/rtl818x.h
@@ -58,13 +58,17 @@ struct rtl818x_csr {
 #define RTL818X_INT_TX_FO		(1 << 15)
 	__le32	TX_CONF;
 #define RTL818X_TX_CONF_LOOPBACK_MAC	(1 << 17)
+#define RTL818X_TX_CONF_LOOPBACK_CONT	(3 << 17)
 #define RTL818X_TX_CONF_NO_ICV		(1 << 19)
 #define RTL818X_TX_CONF_DISCW		(1 << 20)
+#define RTL818X_TX_CONF_SAT_HWPLCP	(1 << 24)
 #define RTL818X_TX_CONF_R8180_ABCD	(2 << 25)
 #define RTL818X_TX_CONF_R8180_F		(3 << 25)
 #define RTL818X_TX_CONF_R8185_ABC	(4 << 25)
 #define RTL818X_TX_CONF_R8185_D		(5 << 25)
 #define RTL818X_TX_CONF_HWVER_MASK	(7 << 25)
+#define RTL818X_TX_CONF_PROBE_DTS	(1 << 29)
+#define RTL818X_TX_CONF_HW_SEQNUM	(1 << 30)
 #define RTL818X_TX_CONF_CW_MIN		(1 << 31)
 	__le32	RX_CONF;
 #define RTL818X_RX_CONF_MONITOR		(1 <<  0)
@@ -75,8 +79,12 @@ struct rtl818x_csr {
 #define RTL818X_RX_CONF_DATA		(1 << 18)
 #define RTL818X_RX_CONF_CTRL		(1 << 19)
 #define RTL818X_RX_CONF_MGMT		(1 << 20)
+#define RTL818X_RX_CONF_ADDR3		(1 << 21)
+#define RTL818X_RX_CONF_PM		(1 << 22)
 #define RTL818X_RX_CONF_BSSID		(1 << 23)
 #define RTL818X_RX_CONF_RX_AUTORESETPHY	(1 << 28)
+#define RTL818X_RX_CONF_CSDM1		(1 << 29)
+#define RTL818X_RX_CONF_CSDM2		(1 << 30)
 #define RTL818X_RX_CONF_ONLYERLPKT	(1 << 31)
 	__le32	INT_TIMEOUT;
 	__le32	TBDA;
@@ -92,6 +100,7 @@ struct rtl818x_csr {
 	u8	CONFIG0;
 	u8	CONFIG1;
 	u8	CONFIG2;
+#define RTL818X_CONFIG2_ANTENNA_DIV	(1 << 6)
 	__le32	ANAPARAM;
 	u8	MSR;
 #define RTL818X_MSR_NO_LINK		(0 << 2)
@@ -104,14 +113,17 @@ struct rtl818x_csr {
 #define RTL818X_CONFIG4_VCOOFF		(1 << 7)
 	u8	TESTR;
 	u8	reserved_9[2];
-	__le16	PGSELECT;
+	u8	PGSELECT;
+	u8	SECURITY;
 	__le32	ANAPARAM2;
 	u8	reserved_10[12];
 	__le16	BEACON_INTERVAL;
 	__le16	ATIM_WND;
 	__le16	BEACON_INTERVAL_TIME;
 	__le16	ATIMTR_INTERVAL;
-	u8	reserved_11[4];
+	u8	PHY_DELAY;
+	u8	CARRIER_SENSE_COUNTER;
+	u8	reserved_11[2];
 	u8	PHY[4];
 	__le16	RFPinsOutput;
 	__le16	RFPinsEnable;
@@ -149,11 +161,20 @@ struct rtl818x_csr {
 	u8	RETRY_CTR;
 	u8	reserved_18[5];
 	__le32	RDSAR;
-	u8	reserved_19[18];
-	u16	TALLY_CNT;
+	u8	reserved_19[12];
+	__le16	FEMR;
+	u8	reserved_20[4];
+	__le16	TALLY_CNT;
 	u8	TALLY_SEL;
 } __attribute__((packed));
 
+struct rtl818x_rf_ops {
+	char *name;
+	void (*init)(struct ieee80211_hw *);
+	void (*stop)(struct ieee80211_hw *);
+	void (*set_chan)(struct ieee80211_hw *, struct ieee80211_conf *);
+};
+
 static const struct ieee80211_rate rtl818x_rates[] = {
 	{ .rate = 10,
 	  .val = 0,
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 6b4a132bf861..c69531348363 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2082,6 +2082,9 @@
 #define PCI_DEVICE_ID_ALTIMA_AC9100	0x03ea
 #define PCI_DEVICE_ID_ALTIMA_AC1003	0x03eb
 
+#define PCI_VENDOR_ID_BELKIN		0x1799
+#define PCI_DEVICE_ID_BELKIN_F5D7010V7	0x701f
+
 #define PCI_VENDOR_ID_LENOVO		0x17aa
 
 #define PCI_VENDOR_ID_ARECA		0x17d3
-- 
cgit v1.2.3


From 32bfd35d4b63bd63de4bb0d791ef049c3c868726 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 01:31:26 +0100
Subject: mac80211: dont use interface indices in drivers

This patch gets rid of the if_id stuff where possible in favour of
a new per-virtual-interface structure "struct ieee80211_vif". This
structure is located at the end of the per-interface structure and
contains a variable length driver-use data area.

This has two advantages:
 * removes the need to look up interfaces by if_id, this is better
   for working with network namespaces and performance
 * allows drivers to store and retrieve per-interface data without
   having to allocate own lists/hash tables

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |  3 +-
 drivers/net/wireless/ath5k/base.c           | 15 ++++---
 drivers/net/wireless/ath5k/base.h           |  2 +-
 drivers/net/wireless/ath5k/hw.c             |  2 +-
 drivers/net/wireless/b43/b43.h              |  5 +--
 drivers/net/wireless/b43/main.c             | 13 +++---
 drivers/net/wireless/b43/xmit.c             |  6 +--
 drivers/net/wireless/b43legacy/b43legacy.h  |  5 +--
 drivers/net/wireless/b43legacy/main.c       | 13 +++---
 drivers/net/wireless/b43legacy/xmit.c       |  6 +--
 drivers/net/wireless/iwlwifi/iwl-3945.h     |  2 +-
 drivers/net/wireless/iwlwifi/iwl-4965.h     |  2 +-
 drivers/net/wireless/iwlwifi/iwl3945-base.c | 28 ++++++------
 drivers/net/wireless/iwlwifi/iwl4965-base.c | 28 ++++++------
 drivers/net/wireless/p54common.c            |  3 +-
 drivers/net/wireless/rt2x00/rt2x00.h        |  5 ++-
 drivers/net/wireless/rt2x00/rt2x00mac.c     |  5 ++-
 drivers/net/wireless/rtl8180.h              |  2 +-
 drivers/net/wireless/rtl8180_dev.c          | 11 +++--
 drivers/net/wireless/rtl8187.h              |  2 +-
 drivers/net/wireless/rtl8187_dev.c          |  8 ++--
 drivers/net/wireless/zd1211rw/zd_mac.c      |  3 +-
 include/net/mac80211.h                      | 70 ++++++++++++++++++-----------
 net/mac80211/ieee80211.c                    | 12 ++---
 net/mac80211/ieee80211_i.h                  |  8 ++++
 net/mac80211/ieee80211_iface.c              |  2 +-
 net/mac80211/sta_info.c                     | 27 ++++++++---
 net/mac80211/tx.c                           | 46 +++++++++----------
 net/mac80211/util.c                         | 45 +++++++------------
 29 files changed, 205 insertions(+), 174 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 14d90a9409aa..79796186713e 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -1312,7 +1312,8 @@ static int adm8211_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 	return 0;
 }
 
-static int adm8211_config_interface(struct ieee80211_hw *dev, int if_id,
+static int adm8211_config_interface(struct ieee80211_hw *dev,
+				    struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
 {
 	struct adm8211_priv *priv = dev->priv;
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index d3d37282f3dc..742616a03d5f 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -178,7 +178,8 @@ static void ath5k_remove_interface(struct ieee80211_hw *hw,
 		struct ieee80211_if_init_conf *conf);
 static int ath5k_config(struct ieee80211_hw *hw,
 		struct ieee80211_conf *conf);
-static int ath5k_config_interface(struct ieee80211_hw *hw, int if_id,
+static int ath5k_config_interface(struct ieee80211_hw *hw,
+		struct ieee80211_vif *vif,
 		struct ieee80211_if_conf *conf);
 static void ath5k_configure_filter(struct ieee80211_hw *hw,
 		unsigned int changed_flags,
@@ -2498,12 +2499,12 @@ static int ath5k_add_interface(struct ieee80211_hw *hw,
 	int ret;
 
 	mutex_lock(&sc->lock);
-	if (sc->iface_id) {
+	if (sc->vif) {
 		ret = 0;
 		goto end;
 	}
 
-	sc->iface_id = conf->if_id;
+	sc->vif = conf->vif;
 
 	switch (conf->type) {
 	case IEEE80211_IF_TYPE_STA:
@@ -2528,10 +2529,10 @@ ath5k_remove_interface(struct ieee80211_hw *hw,
 	struct ath5k_softc *sc = hw->priv;
 
 	mutex_lock(&sc->lock);
-	if (sc->iface_id != conf->if_id)
+	if (sc->vif != conf->vif)
 		goto end;
 
-	sc->iface_id = 0;
+	sc->vif = NULL;
 end:
 	mutex_unlock(&sc->lock);
 }
@@ -2549,7 +2550,7 @@ ath5k_config(struct ieee80211_hw *hw,
 }
 
 static int
-ath5k_config_interface(struct ieee80211_hw *hw, int if_id,
+ath5k_config_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			struct ieee80211_if_conf *conf)
 {
 	struct ath5k_softc *sc = hw->priv;
@@ -2560,7 +2561,7 @@ ath5k_config_interface(struct ieee80211_hw *hw, int if_id,
 	 * be set to mac80211's value at ath5k_config(). */
 	sc->bintval = 1000 * 1000 / 1024;
 	mutex_lock(&sc->lock);
-	if (sc->iface_id != if_id) {
+	if (sc->vif != vif) {
 		ret = -EIO;
 		goto unlock;
 	}
diff --git a/drivers/net/wireless/ath5k/base.h b/drivers/net/wireless/ath5k/base.h
index 927d67db3dc2..7ba2223d2247 100644
--- a/drivers/net/wireless/ath5k/base.h
+++ b/drivers/net/wireless/ath5k/base.h
@@ -123,7 +123,7 @@ struct ath5k_softc {
 	unsigned int		curmode;	/* current phy mode */
 	struct ieee80211_channel *curchan;	/* current h/w channel */
 
-	int 			iface_id;	/* add/remove_interface id */
+	struct ieee80211_vif *vif;
 
 	struct {
 		u8	rxflags;	/* radiotap rx flags */
diff --git a/drivers/net/wireless/ath5k/hw.c b/drivers/net/wireless/ath5k/hw.c
index 5623d7dc738e..3b9336387df0 100644
--- a/drivers/net/wireless/ath5k/hw.c
+++ b/drivers/net/wireless/ath5k/hw.c
@@ -533,7 +533,7 @@ static inline void ath5k_hw_write_rate_duration(struct ath5k_hw *ah,
 		 * ieee80211_duration() for a brief description of
 		 * what rate we should choose to TX ACKs. */
 		tx_time = ieee80211_generic_frame_duration(sc->hw,
-			sc->iface_id, 10, control_rate->rate_kbps/100);
+			sc->vif, 10, control_rate->rate_kbps/100);
 
 		ath5k_hw_reg_write(ah, tx_time, reg);
 
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index 5a1a790a105b..82bff51d5595 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -625,10 +625,7 @@ struct b43_wl {
 	 * at a time. General information about this interface follows.
 	 */
 
-	/* Opaque ID of the operating interface from the ieee80211
-	 * subsystem. Do not modify.
-	 */
-	int if_id;
+	struct ieee80211_vif *vif;
 	/* The MAC address of the operating interface. */
 	u8 mac_addr[ETH_ALEN];
 	/* Current BSSID */
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index ea63a9928803..af3d24c559c0 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -1169,7 +1169,7 @@ static void b43_write_probe_resp_plcp(struct b43_wldev *dev,
 	plcp.data = 0;
 	b43_generate_plcp_hdr(&plcp, size + FCS_LEN, rate);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
-					       dev->wl->if_id, size,
+					       dev->wl->vif, size,
 					       B43_RATE_TO_BASE100KBPS(rate));
 	/* Write PLCP in two parts and timing for packet transfer */
 	tmp = le32_to_cpu(plcp.data);
@@ -1226,7 +1226,7 @@ static u8 *b43_generate_probe_resp(struct b43_wldev *dev,
 	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT |
 					 IEEE80211_STYPE_PROBE_RESP);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
-					       dev->wl->if_id, *dest_size,
+					       dev->wl->vif, *dest_size,
 					       B43_RATE_TO_BASE100KBPS(rate));
 	hdr->duration_id = dur;
 
@@ -2928,7 +2928,7 @@ static void b43_op_configure_filter(struct ieee80211_hw *hw,
 }
 
 static int b43_op_config_interface(struct ieee80211_hw *hw,
-				   int if_id,
+				   struct ieee80211_vif *vif,
 				   struct ieee80211_if_conf *conf)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
@@ -2939,7 +2939,7 @@ static int b43_op_config_interface(struct ieee80211_hw *hw,
 		return -ENODEV;
 	mutex_lock(&wl->mutex);
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	B43_WARN_ON(wl->if_id != if_id);
+	B43_WARN_ON(wl->vif != vif);
 	if (conf->bssid)
 		memcpy(wl->bssid, conf->bssid, ETH_ALEN);
 	else
@@ -3445,7 +3445,7 @@ static int b43_op_add_interface(struct ieee80211_hw *hw,
 
 	dev = wl->current_dev;
 	wl->operating = 1;
-	wl->if_id = conf->if_id;
+	wl->vif = conf->vif;
 	wl->if_type = conf->type;
 	memcpy(wl->mac_addr, conf->mac_addr, ETH_ALEN);
 
@@ -3473,7 +3473,8 @@ static void b43_op_remove_interface(struct ieee80211_hw *hw,
 	mutex_lock(&wl->mutex);
 
 	B43_WARN_ON(!wl->operating);
-	B43_WARN_ON(wl->if_id != conf->if_id);
+	B43_WARN_ON(wl->vif != conf->vif);
+	wl->vif = NULL;
 
 	wl->operating = 0;
 
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 419aca1fc4fe..5014213b7752 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -221,7 +221,7 @@ static void generate_txhdr_fw4(struct b43_wldev *dev,
 	} else {
 		int fbrate_base100kbps = B43_RATE_TO_BASE100KBPS(rate_fb);
 		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
-								 dev->wl->if_id,
+								 txctl->vif,
 								 fragment_len,
 								 fbrate_base100kbps);
 	}
@@ -312,7 +312,7 @@ static void generate_txhdr_fw4(struct b43_wldev *dev,
 		rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb);
 
 		if (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
-			ieee80211_ctstoself_get(dev->wl->hw, dev->wl->if_id,
+			ieee80211_ctstoself_get(dev->wl->hw, txctl->vif,
 						fragment_data, fragment_len,
 						txctl,
 						(struct ieee80211_cts *)(txhdr->
@@ -320,7 +320,7 @@ static void generate_txhdr_fw4(struct b43_wldev *dev,
 			mac_ctl |= B43_TX4_MAC_SENDCTS;
 			len = sizeof(struct ieee80211_cts);
 		} else {
-			ieee80211_rts_get(dev->wl->hw, dev->wl->if_id,
+			ieee80211_rts_get(dev->wl->hw, txctl->vif,
 					  fragment_data, fragment_len, txctl,
 					  (struct ieee80211_rts *)(txhdr->
 								   rts_frame));
diff --git a/drivers/net/wireless/b43legacy/b43legacy.h b/drivers/net/wireless/b43legacy/b43legacy.h
index e4de437b7896..8352a4e1cfc9 100644
--- a/drivers/net/wireless/b43legacy/b43legacy.h
+++ b/drivers/net/wireless/b43legacy/b43legacy.h
@@ -577,10 +577,7 @@ struct b43legacy_wl {
 	 * at a time. General information about this interface follows.
 	 */
 
-	/* Opaque ID of the operating interface from the ieee80211
-	 * subsystem. Do not modify.
-	 */
-	int if_id;
+	struct ieee80211_vif *vif;
 	/* MAC address (can be NULL). */
 	u8 mac_addr[ETH_ALEN];
 	/* Current BSSID (can be NULL). */
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index 575fd9a5874d..2d5735ddad8c 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -976,7 +976,7 @@ static void b43legacy_write_probe_resp_plcp(struct b43legacy_wldev *dev,
 	plcp.data = 0;
 	b43legacy_generate_plcp_hdr(&plcp, size + FCS_LEN, rate);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
-					       dev->wl->if_id,
+					       dev->wl->vif,
 					       size,
 					       B43legacy_RATE_TO_100KBPS(rate));
 	/* Write PLCP in two parts and timing for packet transfer */
@@ -1042,7 +1042,7 @@ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev,
 	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT |
 					 IEEE80211_STYPE_PROBE_RESP);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
-					       dev->wl->if_id,
+					       dev->wl->vif,
 					       *dest_size,
 					       B43legacy_RATE_TO_100KBPS(rate));
 	hdr->duration_id = dur;
@@ -2647,7 +2647,7 @@ static void b43legacy_op_configure_filter(struct ieee80211_hw *hw,
 }
 
 static int b43legacy_op_config_interface(struct ieee80211_hw *hw,
-					 int if_id,
+					 struct ieee80211_vif *vif,
 					 struct ieee80211_if_conf *conf)
 {
 	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
@@ -2658,7 +2658,7 @@ static int b43legacy_op_config_interface(struct ieee80211_hw *hw,
 		return -ENODEV;
 	mutex_lock(&wl->mutex);
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	B43legacy_WARN_ON(wl->if_id != if_id);
+	B43legacy_WARN_ON(wl->vif != vif);
 	if (conf->bssid)
 		memcpy(wl->bssid, conf->bssid, ETH_ALEN);
 	else
@@ -3177,7 +3177,7 @@ static int b43legacy_op_add_interface(struct ieee80211_hw *hw,
 
 	dev = wl->current_dev;
 	wl->operating = 1;
-	wl->if_id = conf->if_id;
+	wl->vif = conf->vif;
 	wl->if_type = conf->type;
 	memcpy(wl->mac_addr, conf->mac_addr, ETH_ALEN);
 
@@ -3205,7 +3205,8 @@ static void b43legacy_op_remove_interface(struct ieee80211_hw *hw,
 	mutex_lock(&wl->mutex);
 
 	B43legacy_WARN_ON(!wl->operating);
-	B43legacy_WARN_ON(wl->if_id != conf->if_id);
+	B43legacy_WARN_ON(wl->vif != conf->vif);
+	wl->vif = NULL;
 
 	wl->operating = 0;
 
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index fa095d43237f..e20c552442d5 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -223,7 +223,7 @@ static void generate_txhdr_fw3(struct b43legacy_wldev *dev,
 	} else {
 		int fbrate_base100kbps = B43legacy_RATE_TO_100KBPS(rate_fb);
 		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
-							 dev->wl->if_id,
+							 txctl->vif,
 							 fragment_len,
 							 fbrate_base100kbps);
 	}
@@ -312,7 +312,7 @@ static void generate_txhdr_fw3(struct b43legacy_wldev *dev,
 
 		if (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
 			ieee80211_ctstoself_get(dev->wl->hw,
-						dev->wl->if_id,
+						txctl->vif,
 						fragment_data,
 						fragment_len, txctl,
 						(struct ieee80211_cts *)
@@ -321,7 +321,7 @@ static void generate_txhdr_fw3(struct b43legacy_wldev *dev,
 			len = sizeof(struct ieee80211_cts);
 		} else {
 			ieee80211_rts_get(dev->wl->hw,
-					  dev->wl->if_id,
+					  txctl->vif,
 					  fragment_data, fragment_len, txctl,
 					  (struct ieee80211_rts *)
 					  (txhdr->rts_frame));
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.h b/drivers/net/wireless/iwlwifi/iwl-3945.h
index e97f1fb48b97..20b925f57e35 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.h
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.h
@@ -863,7 +863,7 @@ struct iwl3945_priv {
 	u32 timestamp1;
 	u16 beacon_int;
 	struct iwl3945_driver_hw_info hw_setting;
-	int interface_id;
+	struct ieee80211_vif *vif;
 
 	/* Current association information needed to configure the
 	 * hardware */
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.h b/drivers/net/wireless/iwlwifi/iwl-4965.h
index 241f4448a4ad..aad7f70bab76 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.h
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.h
@@ -1212,7 +1212,7 @@ struct iwl4965_priv {
 	u32 timestamp1;
 	u16 beacon_int;
 	struct iwl4965_driver_hw_info hw_setting;
-	int interface_id;
+	struct ieee80211_vif *vif;
 
 	/* Current association information needed to configure the
 	 * hardware */
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index b662ff3b9141..1830e13d5cfd 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2767,8 +2767,8 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 		goto drop_unlock;
 	}
 
-	if (!priv->interface_id) {
-		IWL_DEBUG_DROP("Dropping - !priv->interface_id\n");
+	if (!priv->vif) {
+		IWL_DEBUG_DROP("Dropping - !priv->vif\n");
 		goto drop_unlock;
 	}
 
@@ -3549,7 +3549,7 @@ static void iwl3945_bg_beacon_update(struct work_struct *work)
 	struct sk_buff *beacon;
 
 	/* Pull updated AP beacon from mac80211. will fail if not in AP mode */
-	beacon = ieee80211_beacon_get(priv->hw, priv->interface_id, NULL);
+	beacon = ieee80211_beacon_get(priv->hw, priv->vif, NULL);
 
 	if (!beacon) {
 		IWL_ERROR("update beacon failed\n");
@@ -6780,7 +6780,7 @@ static void iwl3945_bg_post_associate(struct work_struct *data)
 
 	mutex_lock(&priv->mutex);
 
-	if (!priv->interface_id || !priv->is_open) {
+	if (!priv->vif || !priv->is_open) {
 		mutex_unlock(&priv->mutex);
 		return;
 	}
@@ -6981,15 +6981,15 @@ static int iwl3945_mac_add_interface(struct ieee80211_hw *hw,
 	unsigned long flags;
 	DECLARE_MAC_BUF(mac);
 
-	IWL_DEBUG_MAC80211("enter: id %d, type %d\n", conf->if_id, conf->type);
+	IWL_DEBUG_MAC80211("enter: type %d\n", conf->type);
 
-	if (priv->interface_id) {
-		IWL_DEBUG_MAC80211("leave - interface_id != 0\n");
+	if (priv->vif) {
+		IWL_DEBUG_MAC80211("leave - vif != NULL\n");
 		return -EOPNOTSUPP;
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
-	priv->interface_id = conf->if_id;
+	priv->vif = conf->vif;
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
@@ -7157,7 +7157,8 @@ static void iwl3945_config_ap(struct iwl3945_priv *priv)
 	 * clear sta table, add BCAST sta... */
 }
 
-static int iwl3945_mac_config_interface(struct ieee80211_hw *hw, int if_id,
+static int iwl3945_mac_config_interface(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
 {
 	struct iwl3945_priv *priv = hw->priv;
@@ -7179,7 +7180,6 @@ static int iwl3945_mac_config_interface(struct ieee80211_hw *hw, int if_id,
 
 	mutex_lock(&priv->mutex);
 
-	IWL_DEBUG_MAC80211("enter: interface id %d\n", if_id);
 	if (conf->bssid)
 		IWL_DEBUG_MAC80211("bssid: %s\n",
 				   print_mac(mac, conf->bssid));
@@ -7196,8 +7196,8 @@ static int iwl3945_mac_config_interface(struct ieee80211_hw *hw, int if_id,
 		return 0;
 	}
 
-	if (priv->interface_id != if_id) {
-		IWL_DEBUG_MAC80211("leave - interface_id != if_id\n");
+	if (priv->vif != vif) {
+		IWL_DEBUG_MAC80211("leave - priv->vif != vif\n");
 		mutex_unlock(&priv->mutex);
 		return 0;
 	}
@@ -7295,8 +7295,8 @@ static void iwl3945_mac_remove_interface(struct ieee80211_hw *hw,
 		priv->staging_rxon.filter_flags &= ~RXON_FILTER_ASSOC_MSK;
 		iwl3945_commit_rxon(priv);
 	}
-	if (priv->interface_id == conf->if_id) {
-		priv->interface_id = 0;
+	if (priv->vif == conf->vif) {
+		priv->vif = NULL;
 		memset(priv->bssid, 0, ETH_ALEN);
 		memset(priv->essid, 0, IW_ESSID_MAX_SIZE);
 		priv->essid_len = 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index ce0574b087fb..52eb37f4ac78 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -2897,8 +2897,8 @@ static int iwl4965_tx_skb(struct iwl4965_priv *priv,
 		goto drop_unlock;
 	}
 
-	if (!priv->interface_id) {
-		IWL_DEBUG_DROP("Dropping - !priv->interface_id\n");
+	if (!priv->vif) {
+		IWL_DEBUG_DROP("Dropping - !priv->vif\n");
 		goto drop_unlock;
 	}
 
@@ -3893,7 +3893,7 @@ static void iwl4965_bg_beacon_update(struct work_struct *work)
 	struct sk_buff *beacon;
 
 	/* Pull updated AP beacon from mac80211. will fail if not in AP mode */
-	beacon = ieee80211_beacon_get(priv->hw, priv->interface_id, NULL);
+	beacon = ieee80211_beacon_get(priv->hw, priv->vif, NULL);
 
 	if (!beacon) {
 		IWL_ERROR("update beacon failed\n");
@@ -7214,7 +7214,7 @@ static void iwl4965_bg_post_associate(struct work_struct *data)
 
 	mutex_lock(&priv->mutex);
 
-	if (!priv->interface_id || !priv->is_open) {
+	if (!priv->vif || !priv->is_open) {
 		mutex_unlock(&priv->mutex);
 		return;
 	}
@@ -7425,15 +7425,15 @@ static int iwl4965_mac_add_interface(struct ieee80211_hw *hw,
 	unsigned long flags;
 	DECLARE_MAC_BUF(mac);
 
-	IWL_DEBUG_MAC80211("enter: id %d, type %d\n", conf->if_id, conf->type);
+	IWL_DEBUG_MAC80211("enter: type %d\n", conf->type);
 
-	if (priv->interface_id) {
-		IWL_DEBUG_MAC80211("leave - interface_id != 0\n");
+	if (priv->vif) {
+		IWL_DEBUG_MAC80211("leave - vif != NULL\n");
 		return 0;
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
-	priv->interface_id = conf->if_id;
+	priv->vif = conf->vif;
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
@@ -7617,7 +7617,8 @@ static void iwl4965_config_ap(struct iwl4965_priv *priv)
 	 * clear sta table, add BCAST sta... */
 }
 
-static int iwl4965_mac_config_interface(struct ieee80211_hw *hw, int if_id,
+static int iwl4965_mac_config_interface(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
 {
 	struct iwl4965_priv *priv = hw->priv;
@@ -7637,7 +7638,6 @@ static int iwl4965_mac_config_interface(struct ieee80211_hw *hw, int if_id,
 
 	mutex_lock(&priv->mutex);
 
-	IWL_DEBUG_MAC80211("enter: interface id %d\n", if_id);
 	if (conf->bssid)
 		IWL_DEBUG_MAC80211("bssid: %s\n",
 				   print_mac(mac, conf->bssid));
@@ -7654,8 +7654,8 @@ static int iwl4965_mac_config_interface(struct ieee80211_hw *hw, int if_id,
 		return 0;
 	}
 
-	if (priv->interface_id != if_id) {
-		IWL_DEBUG_MAC80211("leave - interface_id != if_id\n");
+	if (priv->vif != vif) {
+		IWL_DEBUG_MAC80211("leave - priv->vif != vif\n");
 		mutex_unlock(&priv->mutex);
 		return 0;
 	}
@@ -7753,8 +7753,8 @@ static void iwl4965_mac_remove_interface(struct ieee80211_hw *hw,
 		priv->staging_rxon.filter_flags &= ~RXON_FILTER_ASSOC_MSK;
 		iwl4965_commit_rxon(priv);
 	}
-	if (priv->interface_id == conf->if_id) {
-		priv->interface_id = 0;
+	if (priv->vif == conf->vif) {
+		priv->vif = NULL;
 		memset(priv->bssid, 0, ETH_ALEN);
 		memset(priv->essid, 0, IW_ESSID_MAX_SIZE);
 		priv->essid_len = 0;
diff --git a/drivers/net/wireless/p54common.c b/drivers/net/wireless/p54common.c
index 9660fdd873ac..5cda49aff3a8 100644
--- a/drivers/net/wireless/p54common.c
+++ b/drivers/net/wireless/p54common.c
@@ -854,7 +854,8 @@ static int p54_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 	return ret;
 }
 
-static int p54_config_interface(struct ieee80211_hw *dev, int if_id,
+static int p54_config_interface(struct ieee80211_hw *dev,
+				struct ieee80211_vif *vif,
 				struct ieee80211_if_conf *conf)
 {
 	struct p54_common *priv = dev->priv;
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 5fb9734ade85..94a8a7ce5382 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -372,7 +372,7 @@ struct interface {
 	 * to us by the 80211 stack, and is used to request
 	 * new beacons.
 	 */
-	int id;
+	struct ieee80211_vif *id;
 
 	/*
 	 * Current working type (IEEE80211_IF_TYPE_*).
@@ -929,7 +929,8 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw,
 void rt2x00mac_remove_interface(struct ieee80211_hw *hw,
 				struct ieee80211_if_init_conf *conf);
 int rt2x00mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
-int rt2x00mac_config_interface(struct ieee80211_hw *hw, int if_id,
+int rt2x00mac_config_interface(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif,
 			       struct ieee80211_if_conf *conf);
 int rt2x00mac_get_stats(struct ieee80211_hw *hw,
 			struct ieee80211_low_level_stats *stats);
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index e99d167d7df6..1d67bcd46bcb 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -181,7 +181,7 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw,
 	    is_interface_present(intf))
 		return -ENOBUFS;
 
-	intf->id = conf->if_id;
+	intf->id = conf->vif;
 	intf->type = conf->type;
 	if (conf->type == IEEE80211_IF_TYPE_AP)
 		memcpy(&intf->bssid, conf->mac_addr, ETH_ALEN);
@@ -265,7 +265,8 @@ int rt2x00mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf)
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_config);
 
-int rt2x00mac_config_interface(struct ieee80211_hw *hw, int if_id,
+int rt2x00mac_config_interface(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif,
 			       struct ieee80211_if_conf *conf)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
diff --git a/drivers/net/wireless/rtl8180.h b/drivers/net/wireless/rtl8180.h
index 0b333cacbd85..2cbfe3c8081f 100644
--- a/drivers/net/wireless/rtl8180.h
+++ b/drivers/net/wireless/rtl8180.h
@@ -90,8 +90,8 @@ struct rtl8180_priv {
 	/* common between rtl818x drivers */
 	struct rtl818x_csr __iomem *map;
 	const struct rtl818x_rf_ops *rf;
+	struct ieee80211_vif *vif;
 	int mode;
-	int if_id;
 
 	/* rtl8180 driver specific */
 	spinlock_t lock;
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index 4b7b032c194a..07f37b0ccf91 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -236,7 +236,8 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 		kmemdup(control, sizeof(*control), GFP_ATOMIC);
 
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
-		rts_duration = ieee80211_rts_duration(dev, priv->if_id, skb->len, control);
+		rts_duration = ieee80211_rts_duration(dev, priv->vif, skb->len,
+						      control);
 
 	if (!priv->r8185) {
 		unsigned int remainder;
@@ -638,6 +639,8 @@ static int rtl8180_add_interface(struct ieee80211_hw *dev,
 		return -EOPNOTSUPP;
 	}
 
+	priv->vif = conf->vif;
+
 	rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG);
 	rtl818x_iowrite32(priv, (__le32 __iomem *)&priv->map->MAC[0],
 			  cpu_to_le32(*(u32 *)conf->mac_addr));
@@ -653,6 +656,7 @@ static void rtl8180_remove_interface(struct ieee80211_hw *dev,
 {
 	struct rtl8180_priv *priv = dev->priv;
 	priv->mode = IEEE80211_IF_TYPE_MNTR;
+	priv->vif = NULL;
 }
 
 static int rtl8180_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
@@ -664,14 +668,13 @@ static int rtl8180_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 	return 0;
 }
 
-static int rtl8180_config_interface(struct ieee80211_hw *dev, int if_id,
+static int rtl8180_config_interface(struct ieee80211_hw *dev,
+				    struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
 {
 	struct rtl8180_priv *priv = dev->priv;
 	int i;
 
-	priv->if_id = if_id;
-
 	for (i = 0; i < ETH_ALEN; i++)
 		rtl818x_iowrite8(priv, &priv->map->BSSID[i], conf->bssid[i]);
 
diff --git a/drivers/net/wireless/rtl8187.h b/drivers/net/wireless/rtl8187.h
index 26a4f1097b4a..8680a0b6433c 100644
--- a/drivers/net/wireless/rtl8187.h
+++ b/drivers/net/wireless/rtl8187.h
@@ -65,8 +65,8 @@ struct rtl8187_priv {
 	/* common between rtl818x drivers */
 	struct rtl818x_csr *map;
 	const struct rtl818x_rf_ops *rf;
+	struct ieee80211_vif *vif;
 	int mode;
-	int if_id;
 
 	/* rtl8187 specific */
 	struct ieee80211_channel channels[14];
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index 8dab25a65990..0d71716d750d 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -150,7 +150,8 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
 		flags |= RTL8187_TX_FLAG_RTS;
-		rts_dur = ieee80211_rts_duration(dev, priv->if_id, skb->len, control);
+		rts_dur = ieee80211_rts_duration(dev, priv->vif,
+						 skb->len, control);
 	}
 	if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
 		flags |= RTL8187_TX_FLAG_CTS;
@@ -560,14 +561,13 @@ static int rtl8187_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 	return 0;
 }
 
-static int rtl8187_config_interface(struct ieee80211_hw *dev, int if_id,
+static int rtl8187_config_interface(struct ieee80211_hw *dev,
+				    struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
 {
 	struct rtl8187_priv *priv = dev->priv;
 	int i;
 
-	priv->if_id = if_id;
-
 	for (i = 0; i < ETH_ALEN; i++)
 		rtl818x_iowrite8(priv, &priv->map->BSSID[i], conf->bssid[i]);
 
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 7b8693050480..3409cf97f5f8 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -733,7 +733,8 @@ static int zd_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf)
 	return zd_chip_set_channel(&mac->chip, conf->channel);
 }
 
-static int zd_op_config_interface(struct ieee80211_hw *hw, int if_id,
+static int zd_op_config_interface(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif,
 				   struct ieee80211_if_conf *conf)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index be2a383b4776..24a8ad3f5073 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -280,6 +280,7 @@ struct ieee80211_low_level_stats {
  * the hardware to use given values (depending on what is supported). */
 
 struct ieee80211_tx_control {
+	struct ieee80211_vif *vif;
 	int tx_rate; /* Transmit rate, given as the hw specific value for the
 		      * rate (from struct ieee80211_rate) */
 	int rts_cts_rate; /* Transmit rate for RTS/CTS frame, given as the hw
@@ -332,7 +333,6 @@ struct ieee80211_tx_control {
 			     * packet dropping when probing higher rates, if hw
 			     * supports multiple retry rates. -1 = not used */
 	int type;	/* internal */
-	int ifindex;	/* internal */
 };
 
 
@@ -529,14 +529,26 @@ enum ieee80211_if_types {
 	IEEE80211_IF_TYPE_VLAN,
 };
 
+/**
+ * struct ieee80211_vif - per-interface data
+ *
+ * Data in this structure is continually present for driver
+ * use during the life of a virtual interface.
+ *
+ * @drv_priv: data area for driver use, will always be aligned to
+ *	sizeof(void *).
+ */
+struct ieee80211_vif {
+	/* must be last */
+	u8 drv_priv[0] __attribute__((__aligned__(sizeof(void *))));
+};
+
 /**
  * struct ieee80211_if_init_conf - initial configuration of an interface
  *
- * @if_id: internal interface ID. This number has no particular meaning to
- *	drivers and the only allowed usage is to pass it to
- *	ieee80211_beacon_get() and ieee80211_get_buffered_bc() functions.
- *	This field is not valid for monitor interfaces
- *	(interfaces of %IEEE80211_IF_TYPE_MNTR type).
+ * @vif: pointer to a driver-use per-interface structure. The pointer
+ *	itself is also used for various functions including
+ *	ieee80211_beacon_get() and ieee80211_get_buffered_bc().
  * @type: one of &enum ieee80211_if_types constants. Determines the type of
  *	added/removed interface.
  * @mac_addr: pointer to MAC address of the interface. This pointer is valid
@@ -553,8 +565,8 @@ enum ieee80211_if_types {
  * in pure monitor mode.
  */
 struct ieee80211_if_init_conf {
-	int if_id;
 	enum ieee80211_if_types type;
+	struct ieee80211_vif *vif;
 	void *mac_addr;
 };
 
@@ -757,6 +769,9 @@ enum ieee80211_hw_flags {
  * @rate_control_algorithm: rate control algorithm for this hardware.
  *	If unset (NULL), the default algorithm will be used. Must be
  *	set before calling ieee80211_register_hw().
+ *
+ * @vif_data_size: size (in bytes) of the drv_priv data area
+ *	within &struct ieee80211_vif.
  */
 struct ieee80211_hw {
 	struct ieee80211_conf conf;
@@ -767,6 +782,7 @@ struct ieee80211_hw {
 	u32 flags;
 	unsigned int extra_tx_headroom;
 	int channel_change_time;
+	int vif_data_size;
 	u8 queues;
 	s8 max_rssi;
 	s8 max_signal;
@@ -1076,7 +1092,8 @@ struct ieee80211_ops {
 				 struct ieee80211_if_init_conf *conf);
 	int (*config)(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
 	int (*config_interface)(struct ieee80211_hw *hw,
-				int if_id, struct ieee80211_if_conf *conf);
+				struct ieee80211_vif *vif,
+				struct ieee80211_if_conf *conf);
 	void (*configure_filter)(struct ieee80211_hw *hw,
 				 unsigned int changed_flags,
 				 unsigned int *total_flags,
@@ -1094,7 +1111,7 @@ struct ieee80211_ops {
 	int (*set_frag_threshold)(struct ieee80211_hw *hw, u32 value);
 	int (*set_retry_limit)(struct ieee80211_hw *hw,
 			       u32 short_retry, u32 long_retr);
-	void (*sta_notify)(struct ieee80211_hw *hw, int if_id,
+	void (*sta_notify)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			enum sta_notify_cmd, const u8 *addr);
 	void (*erp_ie_changed)(struct ieee80211_hw *hw, u8 changes,
 			       int cts_protection, int preamble);
@@ -1309,7 +1326,7 @@ void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
 /**
  * ieee80211_beacon_get - beacon generation function
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @control: will be filled with information needed to send this beacon.
  *
  * If the beacon frames are generated by the host system (i.e., not in
@@ -1320,13 +1337,13 @@ void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
  * is responsible of freeing it.
  */
 struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
-				     int if_id,
+				     struct ieee80211_vif *vif,
 				     struct ieee80211_tx_control *control);
 
 /**
  * ieee80211_rts_get - RTS frame generation function
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame: pointer to the frame that is going to be protected by the RTS.
  * @frame_len: the frame length (in octets).
  * @frame_txctl: &struct ieee80211_tx_control of the frame.
@@ -1337,7 +1354,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
  * the next RTS frame from the 802.11 code. The low-level is responsible
  * for calling this function before and RTS frame is needed.
  */
-void ieee80211_rts_get(struct ieee80211_hw *hw, int if_id,
+void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       const void *frame, size_t frame_len,
 		       const struct ieee80211_tx_control *frame_txctl,
 		       struct ieee80211_rts *rts);
@@ -1345,7 +1362,7 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, int if_id,
 /**
  * ieee80211_rts_duration - Get the duration field for an RTS frame
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame that is going to be protected by the RTS.
  * @frame_txctl: &struct ieee80211_tx_control of the frame.
  *
@@ -1353,14 +1370,14 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, int if_id,
  * the duration field, the low-level driver uses this function to receive
  * the duration field value in little-endian byteorder.
  */
-__le16 ieee80211_rts_duration(struct ieee80211_hw *hw, int if_id,
-			      size_t frame_len,
+__le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
+			      struct ieee80211_vif *vif, size_t frame_len,
 			      const struct ieee80211_tx_control *frame_txctl);
 
 /**
  * ieee80211_ctstoself_get - CTS-to-self frame generation function
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame: pointer to the frame that is going to be protected by the CTS-to-self.
  * @frame_len: the frame length (in octets).
  * @frame_txctl: &struct ieee80211_tx_control of the frame.
@@ -1371,7 +1388,8 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw, int if_id,
  * the next CTS-to-self frame from the 802.11 code. The low-level is responsible
  * for calling this function before and CTS-to-self frame is needed.
  */
-void ieee80211_ctstoself_get(struct ieee80211_hw *hw, int if_id,
+void ieee80211_ctstoself_get(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
 			     const void *frame, size_t frame_len,
 			     const struct ieee80211_tx_control *frame_txctl,
 			     struct ieee80211_cts *cts);
@@ -1379,7 +1397,7 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw, int if_id,
 /**
  * ieee80211_ctstoself_duration - Get the duration field for a CTS-to-self frame
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame that is going to be protected by the CTS-to-self.
  * @frame_txctl: &struct ieee80211_tx_control of the frame.
  *
@@ -1387,28 +1405,30 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw, int if_id,
  * the duration field, the low-level driver uses this function to receive
  * the duration field value in little-endian byteorder.
  */
-__le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, int if_id,
+__le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
 				    size_t frame_len,
 				    const struct ieee80211_tx_control *frame_txctl);
 
 /**
  * ieee80211_generic_frame_duration - Calculate the duration field for a frame
  * @hw: pointer obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame.
  * @rate: the rate (in 100kbps) at which the frame is going to be transmitted.
  *
  * Calculate the duration field of some generic frame, given its
  * length and transmission rate (in 100kbps).
  */
-__le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, int if_id,
+__le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
 					size_t frame_len,
 					int rate);
 
 /**
  * ieee80211_get_buffered_bc - accessing buffered broadcast and multicast frames
  * @hw: pointer as obtained from ieee80211_alloc_hw().
- * @if_id: interface ID from &struct ieee80211_if_init_conf.
+ * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @control: will be filled with information needed to send returned frame.
  *
  * Function for accessing buffered broadcast and multicast frames. If
@@ -1427,7 +1447,7 @@ __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, int if_id,
  * use common code for all beacons.
  */
 struct sk_buff *
-ieee80211_get_buffered_bc(struct ieee80211_hw *hw, int if_id,
+ieee80211_get_buffered_bc(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			  struct ieee80211_tx_control *control);
 
 /**
@@ -1517,7 +1537,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw);
  */
 void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
 					 void (*iterator)(void *data, u8 *mac,
-							  int if_id),
+						struct ieee80211_vif *vif),
 					 void *data);
 
 #endif /* MAC80211_H */
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 4807e5215a79..42c27089f006 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -243,7 +243,7 @@ static int ieee80211_open(struct net_device *dev)
 		sdata->u.sta.flags &= ~IEEE80211_STA_PREV_BSSID_SET;
 		/* fall through */
 	default:
-		conf.if_id = dev->ifindex;
+		conf.vif = &sdata->vif;
 		conf.type = sdata->type;
 		conf.mac_addr = dev->dev_addr;
 		res = local->ops->add_interface(local_to_hw(local), &conf);
@@ -378,7 +378,7 @@ static int ieee80211_stop(struct net_device *dev)
 		sdata->u.sta.extra_ie_len = 0;
 		/* fall through */
 	default:
-		conf.if_id = dev->ifindex;
+		conf.vif = &sdata->vif;
 		conf.type = sdata->type;
 		conf.mac_addr = dev->dev_addr;
 		/* disable all keys for as long as this netdev is down */
@@ -515,7 +515,7 @@ static int __ieee80211_if_config(struct net_device *dev,
 		conf.beacon_control = control;
 	}
 	return local->ops->config_interface(local_to_hw(local),
-					   dev->ifindex, &conf);
+					    &sdata->vif, &conf);
 }
 
 int ieee80211_if_config(struct net_device *dev)
@@ -527,11 +527,13 @@ int ieee80211_if_config_beacon(struct net_device *dev)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_tx_control control;
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct sk_buff *skb;
 
 	if (!(local->hw.flags & IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE))
 		return 0;
-	skb = ieee80211_beacon_get(local_to_hw(local), dev->ifindex, &control);
+	skb = ieee80211_beacon_get(local_to_hw(local), &sdata->vif,
+				   &control);
 	if (!skb)
 		return -ENOMEM;
 	return __ieee80211_if_config(dev, skb, &control);
@@ -736,7 +738,7 @@ static void ieee80211_remove_tx_extra(struct ieee80211_local *local,
 	struct ieee80211_tx_packet_data *pkt_data;
 
 	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	pkt_data->ifindex = control->ifindex;
+	pkt_data->ifindex = vif_to_sdata(control->vif)->dev->ifindex;
 	pkt_data->flags = 0;
 	if (control->flags & IEEE80211_TXCTL_REQ_TX_STATUS)
 		pkt_data->flags |= IEEE80211_TXPD_REQ_TX_STATUS;
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index b898b316bda1..c551a7f379e0 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -387,8 +387,16 @@ struct ieee80211_sub_if_data {
 		struct dentry *default_key;
 	} debugfs;
 #endif
+	/* must be last, dynamically sized area in this! */
+	struct ieee80211_vif vif;
 };
 
+static inline
+struct ieee80211_sub_if_data *vif_to_sdata(struct ieee80211_vif *p)
+{
+	return container_of(p, struct ieee80211_sub_if_data, vif);
+}
+
 #define IEEE80211_DEV_TO_SUB_IF(dev) netdev_priv(dev)
 
 enum {
diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c
index 7cfd8660b231..b72066378805 100644
--- a/net/mac80211/ieee80211_iface.c
+++ b/net/mac80211/ieee80211_iface.c
@@ -47,7 +47,7 @@ int ieee80211_if_add(struct net_device *dev, const char *name,
 	int ret;
 
 	ASSERT_RTNL();
-	ndev = alloc_netdev(sizeof(struct ieee80211_sub_if_data),
+	ndev = alloc_netdev(sizeof(*sdata) + local->hw.vif_data_size,
 			    name, ieee80211_if_setup);
 	if (!ndev)
 		return -ENOMEM;
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 1257c7aab2a5..32e241764725 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -177,9 +177,16 @@ struct sta_info * sta_info_add(struct ieee80211_local *local,
 	list_add(&sta->list, &local->sta_list);
 	local->num_sta++;
 	sta_info_hash_add(local, sta);
-	if (local->ops->sta_notify)
-		local->ops->sta_notify(local_to_hw(local), dev->ifindex,
-					STA_NOTIFY_ADD, addr);
+	if (local->ops->sta_notify) {
+		struct ieee80211_sub_if_data *sdata;
+
+		sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+		if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+			sdata = sdata->u.vlan.ap;
+
+		local->ops->sta_notify(local_to_hw(local), &sdata->vif,
+				       STA_NOTIFY_ADD, addr);
+	}
 	write_unlock_bh(&local->sta_lock);
 
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
@@ -247,9 +254,17 @@ void sta_info_free(struct sta_info *sta)
 	ieee80211_key_free(sta->key);
 	sta->key = NULL;
 
-	if (local->ops->sta_notify)
-		local->ops->sta_notify(local_to_hw(local), sta->dev->ifindex,
-					STA_NOTIFY_REMOVE, sta->addr);
+	if (local->ops->sta_notify) {
+		struct ieee80211_sub_if_data *sdata;
+
+		sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
+
+		if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+			sdata = sdata->u.vlan.ap;
+
+		local->ops->sta_notify(local_to_hw(local), &sdata->vif,
+				       STA_NOTIFY_REMOVE, sta->addr);
+	}
 
 	rate_control_remove_sta_debugfs(sta);
 	ieee80211_sta_debugfs_remove(sta);
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index f6194167253d..1b772ee2fe14 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -999,9 +999,7 @@ __ieee80211_tx_prepare(struct ieee80211_txrx_data *tx,
 	return TXRX_CONTINUE;
 }
 
-/* Device in tx->dev has a reference added; use dev_put(tx->dev) when
- * finished with it.
- *
+/*
  * NB: @tx is uninitialised when passed in here
  */
 static int ieee80211_tx_prepare(struct ieee80211_txrx_data *tx,
@@ -1022,6 +1020,7 @@ static int ieee80211_tx_prepare(struct ieee80211_txrx_data *tx,
 		return -ENODEV;
 	/* initialises tx with control */
 	__ieee80211_tx_prepare(tx, skb, dev, control);
+	dev_put(dev);
 	return 0;
 }
 
@@ -1252,7 +1251,7 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 		}
 	}
 
-	control.ifindex = odev->ifindex;
+	control.vif = &osdata->vif;
 	control.type = osdata->type;
 	if (pkt_data->flags & IEEE80211_TXPD_REQ_TX_STATUS)
 		control.flags |= IEEE80211_TXCTL_REQ_TX_STATUS;
@@ -1691,7 +1690,8 @@ static void ieee80211_beacon_add_tim(struct ieee80211_local *local,
 	read_unlock_bh(&local->sta_lock);
 }
 
-struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
+struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
 				     struct ieee80211_tx_control *control)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
@@ -1703,19 +1703,16 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
 	u8 *b_head, *b_tail;
 	int bh_len, bt_len;
 
-	bdev = dev_get_by_index(&init_net, if_id);
-	if (bdev) {
-		sdata = IEEE80211_DEV_TO_SUB_IF(bdev);
-		ap = &sdata->u.ap;
-		dev_put(bdev);
-	}
+	sdata = vif_to_sdata(vif);
+	bdev = sdata->dev;
+	ap = &sdata->u.ap;
 
 	if (!ap || sdata->type != IEEE80211_IF_TYPE_AP ||
 	    !ap->beacon_head) {
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
 		if (net_ratelimit())
-			printk(KERN_DEBUG "no beacon data avail for idx=%d "
-			       "(%s)\n", if_id, bdev ? bdev->name : "N/A");
+			printk(KERN_DEBUG "no beacon data avail for %s\n",
+			       bdev->name);
 #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
 		return NULL;
 	}
@@ -1771,7 +1768,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
 }
 EXPORT_SYMBOL(ieee80211_beacon_get);
 
-void ieee80211_rts_get(struct ieee80211_hw *hw, int if_id,
+void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       const void *frame, size_t frame_len,
 		       const struct ieee80211_tx_control *frame_txctl,
 		       struct ieee80211_rts *rts)
@@ -1781,13 +1778,14 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, int if_id,
 
 	fctl = IEEE80211_FTYPE_CTL | IEEE80211_STYPE_RTS;
 	rts->frame_control = cpu_to_le16(fctl);
-	rts->duration = ieee80211_rts_duration(hw, if_id, frame_len, frame_txctl);
+	rts->duration = ieee80211_rts_duration(hw, vif, frame_len,
+					       frame_txctl);
 	memcpy(rts->ra, hdr->addr1, sizeof(rts->ra));
 	memcpy(rts->ta, hdr->addr2, sizeof(rts->ta));
 }
 EXPORT_SYMBOL(ieee80211_rts_get);
 
-void ieee80211_ctstoself_get(struct ieee80211_hw *hw, int if_id,
+void ieee80211_ctstoself_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			     const void *frame, size_t frame_len,
 			     const struct ieee80211_tx_control *frame_txctl,
 			     struct ieee80211_cts *cts)
@@ -1797,13 +1795,15 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw, int if_id,
 
 	fctl = IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CTS;
 	cts->frame_control = cpu_to_le16(fctl);
-	cts->duration = ieee80211_ctstoself_duration(hw, if_id, frame_len, frame_txctl);
+	cts->duration = ieee80211_ctstoself_duration(hw, vif,
+						     frame_len, frame_txctl);
 	memcpy(cts->ra, hdr->addr1, sizeof(cts->ra));
 }
 EXPORT_SYMBOL(ieee80211_ctstoself_get);
 
 struct sk_buff *
-ieee80211_get_buffered_bc(struct ieee80211_hw *hw, int if_id,
+ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
+			  struct ieee80211_vif *vif,
 			  struct ieee80211_tx_control *control)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
@@ -1816,12 +1816,9 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw, int if_id,
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_if_ap *bss = NULL;
 
-	bdev = dev_get_by_index(&init_net, if_id);
-	if (bdev) {
-		sdata = IEEE80211_DEV_TO_SUB_IF(bdev);
-		bss = &sdata->u.ap;
-		dev_put(bdev);
-	}
+	sdata = vif_to_sdata(vif);
+	bdev = sdata->dev;
+
 	if (!bss || sdata->type != IEEE80211_IF_TYPE_AP || !bss->beacon_head)
 		return NULL;
 
@@ -1857,7 +1854,6 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw, int if_id,
 		if (res == TXRX_DROP || res == TXRX_QUEUED)
 			break;
 	}
-	dev_put(tx.dev);
 	skb = tx.skb; /* handlers are allowed to change skb */
 
 	if (res == TXRX_DROP) {
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index adb85dd5098d..15503ca3e945 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -302,44 +302,34 @@ int ieee80211_frame_duration(struct ieee80211_local *local, size_t len,
 }
 
 /* Exported duration function for driver use */
-__le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, int if_id,
+__le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
 					size_t frame_len, int rate)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
-	struct net_device *bdev = dev_get_by_index(&init_net, if_id);
-	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
 	u16 dur;
 	int erp;
 
-	if (unlikely(!bdev))
-		return 0;
-
-	sdata = IEEE80211_DEV_TO_SUB_IF(bdev);
 	erp = ieee80211_is_erp_rate(hw->conf.phymode, rate);
 	dur = ieee80211_frame_duration(local, frame_len, rate,
 		       erp, sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE);
 
-	dev_put(bdev);
 	return cpu_to_le16(dur);
 }
 EXPORT_SYMBOL(ieee80211_generic_frame_duration);
 
-__le16 ieee80211_rts_duration(struct ieee80211_hw *hw, int if_id,
-			      size_t frame_len,
+__le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
+			      struct ieee80211_vif *vif, size_t frame_len,
 			      const struct ieee80211_tx_control *frame_txctl)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
-	struct net_device *bdev = dev_get_by_index(&init_net, if_id);
-	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
 	int short_preamble;
 	int erp;
 	u16 dur;
 
-	if (unlikely(!bdev))
-		return 0;
-
-	sdata = IEEE80211_DEV_TO_SUB_IF(bdev);
 	short_preamble = sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE;
 
 	rate = frame_txctl->rts_rate;
@@ -355,27 +345,22 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw, int if_id,
 	dur += ieee80211_frame_duration(local, 10, rate->rate,
 					erp, short_preamble);
 
-	dev_put(bdev);
 	return cpu_to_le16(dur);
 }
 EXPORT_SYMBOL(ieee80211_rts_duration);
 
-__le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, int if_id,
+__le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
 				    size_t frame_len,
 				    const struct ieee80211_tx_control *frame_txctl)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
-	struct net_device *bdev = dev_get_by_index(&init_net, if_id);
-	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
 	int short_preamble;
 	int erp;
 	u16 dur;
 
-	if (unlikely(!bdev))
-		return 0;
-
-	sdata = IEEE80211_DEV_TO_SUB_IF(bdev);
 	short_preamble = sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE;
 
 	rate = frame_txctl->rts_rate;
@@ -390,7 +375,6 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, int if_id,
 						erp, short_preamble);
 	}
 
-	dev_put(bdev);
 	return cpu_to_le16(dur);
 }
 EXPORT_SYMBOL(ieee80211_ctstoself_duration);
@@ -475,10 +459,11 @@ void ieee80211_wake_queues(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL(ieee80211_wake_queues);
 
-void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
-					 void (*iterator)(void *data, u8 *mac,
-							  int if_id),
-					 void *data)
+void ieee80211_iterate_active_interfaces(
+	struct ieee80211_hw *hw,
+	void (*iterator)(void *data, u8 *mac,
+			 struct ieee80211_vif *vif),
+	void *data)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata;
@@ -501,7 +486,7 @@ void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
 			continue;
 		if (netif_running(sdata->dev))
 			iterator(data, sdata->dev->dev_addr,
-				 sdata->dev->ifindex);
+				 &sdata->vif);
 	}
 
 	rcu_read_unlock();
-- 
cgit v1.2.3


From 51fb61e76d952e6bc2fbdd9f0d38425fbab1cf31 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 19 Dec 2007 01:31:27 +0100
Subject: mac80211: move interface type to vif structure

Drivers that support mixed AP/STA operation may well need to
know the type of a virtual interface when iterating over them.
The easiest way to support that is to move the interface type
variable into the vif structure.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h         |  2 ++
 net/mac80211/cfg.c             |  2 +-
 net/mac80211/debugfs_netdev.c  |  4 +--
 net/mac80211/ieee80211.c       | 36 +++++++++++------------
 net/mac80211/ieee80211_i.h     |  1 -
 net/mac80211/ieee80211_iface.c | 10 +++----
 net/mac80211/ieee80211_ioctl.c | 65 +++++++++++++++++++++---------------------
 net/mac80211/ieee80211_sta.c   | 48 +++++++++++++++----------------
 net/mac80211/key.c             |  6 ++--
 net/mac80211/rx.c              | 38 ++++++++++++------------
 net/mac80211/sta_info.c        |  4 +--
 net/mac80211/tx.c              | 17 +++++------
 net/mac80211/util.c            |  2 +-
 13 files changed, 118 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 24a8ad3f5073..8a49c06dfd38 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -535,10 +535,12 @@ enum ieee80211_if_types {
  * Data in this structure is continually present for driver
  * use during the life of a virtual interface.
  *
+ * @type: type of this virtual interface
  * @drv_priv: data area for driver use, will always be aligned to
  *	sizeof(void *).
  */
 struct ieee80211_vif {
+	enum ieee80211_if_types type;
 	/* must be last */
 	u8 drv_priv[0] __attribute__((__aligned__(sizeof(void *))));
 };
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 11156b381ec3..d02d9ef6b1ef 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -91,7 +91,7 @@ static int ieee80211_change_iface(struct wiphy *wiphy, int ifindex,
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
-        if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_VLAN)
 		return -EOPNOTSUPP;
 
 	ieee80211_if_reinit(dev);
diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c
index bf715d28643b..d2d3c076c853 100644
--- a/net/mac80211/debugfs_netdev.c
+++ b/net/mac80211/debugfs_netdev.c
@@ -226,7 +226,7 @@ static void add_files(struct ieee80211_sub_if_data *sdata)
 	if (!sdata->debugfsdir)
 		return;
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_STA:
 	case IEEE80211_IF_TYPE_IBSS:
 		add_sta_files(sdata);
@@ -353,7 +353,7 @@ void ieee80211_debugfs_add_netdev(struct ieee80211_sub_if_data *sdata)
 
 void ieee80211_debugfs_remove_netdev(struct ieee80211_sub_if_data *sdata)
 {
-	del_files(sdata, sdata->type);
+	del_files(sdata, sdata->vif.type);
 	debugfs_remove(sdata->debugfsdir);
 	sdata->debugfsdir = NULL;
 }
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 42c27089f006..c9981701ef68 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -177,21 +177,21 @@ static int ieee80211_open(struct net_device *dev)
 			/*
 			 * check whether it may have the same address
 			 */
-			if (!identical_mac_addr_allowed(sdata->type,
-							nsdata->type))
+			if (!identical_mac_addr_allowed(sdata->vif.type,
+							nsdata->vif.type))
 				return -ENOTUNIQ;
 
 			/*
 			 * can only add VLANs to enabled APs
 			 */
-			if (sdata->type == IEEE80211_IF_TYPE_VLAN &&
-			    nsdata->type == IEEE80211_IF_TYPE_AP &&
+			if (sdata->vif.type == IEEE80211_IF_TYPE_VLAN &&
+			    nsdata->vif.type == IEEE80211_IF_TYPE_AP &&
 			    netif_running(nsdata->dev))
 				sdata->u.vlan.ap = nsdata;
 		}
 	}
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_WDS:
 		if (is_zero_ether_addr(sdata->u.wds.remote_addr))
 			return -ENOLINK;
@@ -222,7 +222,7 @@ static int ieee80211_open(struct net_device *dev)
 		ieee80211_led_radio(local, local->hw.conf.radio_enabled);
 	}
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_VLAN:
 		list_add(&sdata->u.vlan.list, &sdata->u.vlan.ap->u.ap.vlans);
 		/* no need to tell driver */
@@ -244,7 +244,7 @@ static int ieee80211_open(struct net_device *dev)
 		/* fall through */
 	default:
 		conf.vif = &sdata->vif;
-		conf.type = sdata->type;
+		conf.type = sdata->vif.type;
 		conf.mac_addr = dev->dev_addr;
 		res = local->ops->add_interface(local_to_hw(local), &conf);
 		if (res && !local->open_count && local->ops->stop)
@@ -256,7 +256,7 @@ static int ieee80211_open(struct net_device *dev)
 		ieee80211_reset_erp_info(dev);
 		ieee80211_enable_keys(sdata);
 
-		if (sdata->type == IEEE80211_IF_TYPE_STA &&
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA &&
 		    !(sdata->flags & IEEE80211_SDATA_USERSPACE_MLME))
 			netif_carrier_off(dev);
 		else
@@ -322,7 +322,7 @@ static int ieee80211_stop(struct net_device *dev)
 	dev_mc_unsync(local->mdev, dev);
 
 	/* down all dependent devices, that is VLANs */
-	if (sdata->type == IEEE80211_IF_TYPE_AP) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		struct ieee80211_sub_if_data *vlan, *tmp;
 
 		list_for_each_entry_safe(vlan, tmp, &sdata->u.ap.vlans,
@@ -333,7 +333,7 @@ static int ieee80211_stop(struct net_device *dev)
 
 	local->open_count--;
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_VLAN:
 		list_del(&sdata->u.vlan.list);
 		sdata->u.vlan.ap = NULL;
@@ -379,7 +379,7 @@ static int ieee80211_stop(struct net_device *dev)
 		/* fall through */
 	default:
 		conf.vif = &sdata->vif;
-		conf.type = sdata->type;
+		conf.type = sdata->vif.type;
 		conf.mac_addr = dev->dev_addr;
 		/* disable all keys for as long as this netdev is down */
 		ieee80211_disable_keys(sdata);
@@ -502,13 +502,13 @@ static int __ieee80211_if_config(struct net_device *dev,
 		return 0;
 
 	memset(&conf, 0, sizeof(conf));
-	conf.type = sdata->type;
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	conf.type = sdata->vif.type;
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		conf.bssid = sdata->u.sta.bssid;
 		conf.ssid = sdata->u.sta.ssid;
 		conf.ssid_len = sdata->u.sta.ssid_len;
-	} else if (sdata->type == IEEE80211_IF_TYPE_AP) {
+	} else if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		conf.ssid = sdata->u.ap.ssid;
 		conf.ssid_len = sdata->u.ap.ssid_len;
 		conf.beacon = beacon;
@@ -703,7 +703,7 @@ static void ieee80211_tasklet_handler(unsigned long data)
 		case IEEE80211_RX_MSG:
 			/* status is in skb->cb */
 			memcpy(&rx_status, skb->cb, sizeof(rx_status));
-			/* Clear skb->type in order to not confuse kernel
+			/* Clear skb->pkt_type in order to not confuse kernel
 			 * netstack. */
 			skb->pkt_type = 0;
 			__ieee80211_rx(local_to_hw(local), skb, &rx_status);
@@ -962,7 +962,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 		if (!monitors || !skb)
 			goto out;
 
-		if (sdata->type == IEEE80211_IF_TYPE_MNTR) {
+		if (sdata->vif.type == IEEE80211_IF_TYPE_MNTR) {
 			if (!netif_running(sdata->dev))
 				continue;
 			monitors--;
@@ -1084,7 +1084,7 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len,
 	mdev->header_ops = &ieee80211_header_ops;
 	mdev->set_multicast_list = ieee80211_master_set_multicast_list;
 
-	sdata->type = IEEE80211_IF_TYPE_AP;
+	sdata->vif.type = IEEE80211_IF_TYPE_AP;
 	sdata->dev = mdev;
 	sdata->local = local;
 	sdata->u.ap.force_unicast_rateidx = -1;
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index c551a7f379e0..91edaad4c620 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -294,7 +294,6 @@ struct ieee80211_if_sta {
 #define IEEE80211_SDATA_USERSPACE_MLME	BIT(4)
 struct ieee80211_sub_if_data {
 	struct list_head list;
-	enum ieee80211_if_types type;
 
 	struct wireless_dev wdev;
 
diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c
index b72066378805..be96aa84c79a 100644
--- a/net/mac80211/ieee80211_iface.c
+++ b/net/mac80211/ieee80211_iface.c
@@ -66,7 +66,7 @@ int ieee80211_if_add(struct net_device *dev, const char *name,
 	sdata = IEEE80211_DEV_TO_SUB_IF(ndev);
 	ndev->ieee80211_ptr = &sdata->wdev;
 	sdata->wdev.wiphy = local->hw.wiphy;
-	sdata->type = IEEE80211_IF_TYPE_AP;
+	sdata->vif.type = IEEE80211_IF_TYPE_AP;
 	sdata->dev = ndev;
 	sdata->local = local;
 	ieee80211_if_sdata_init(sdata);
@@ -98,7 +98,7 @@ fail:
 void ieee80211_if_set_type(struct net_device *dev, int type)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	int oldtype = sdata->type;
+	int oldtype = sdata->vif.type;
 
 	/*
 	 * We need to call this function on the master interface
@@ -116,7 +116,7 @@ void ieee80211_if_set_type(struct net_device *dev, int type)
 
 	/* most have no BSS pointer */
 	sdata->bss = NULL;
-	sdata->type = type;
+	sdata->vif.type = type;
 
 	switch (type) {
 	case IEEE80211_IF_TYPE_WDS:
@@ -181,7 +181,7 @@ void ieee80211_if_reinit(struct net_device *dev)
 
 	ieee80211_if_sdata_deinit(sdata);
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_INVALID:
 		/* cannot happen */
 		WARN_ON(1);
@@ -279,7 +279,7 @@ int ieee80211_if_remove(struct net_device *dev, const char *name, int id)
 	ASSERT_RTNL();
 
 	list_for_each_entry_safe(sdata, n, &local->interfaces, list) {
-		if ((sdata->type == id || id == -1) &&
+		if ((sdata->vif.type == id || id == -1) &&
 		    strcmp(name, sdata->dev->name) == 0 &&
 		    sdata->dev != local->mdev) {
 			list_del_rcu(&sdata->list);
diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c
index 2604e21c05a8..5024d3733834 100644
--- a/net/mac80211/ieee80211_ioctl.c
+++ b/net/mac80211/ieee80211_ioctl.c
@@ -112,8 +112,8 @@ static int ieee80211_ioctl_siwgenie(struct net_device *dev,
 	if (sdata->flags & IEEE80211_SDATA_USERSPACE_MLME)
 		return -EOPNOTSUPP;
 
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		int ret = ieee80211_sta_set_extra_ie(dev, extra, data->length);
 		if (ret)
 			return ret;
@@ -232,7 +232,7 @@ static int ieee80211_ioctl_siwmode(struct net_device *dev,
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	int type;
 
-	if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_VLAN)
 		return -EOPNOTSUPP;
 
 	switch (*mode) {
@@ -249,7 +249,7 @@ static int ieee80211_ioctl_siwmode(struct net_device *dev,
 		return -EINVAL;
 	}
 
-	if (type == sdata->type)
+	if (type == sdata->vif.type)
 		return 0;
 	if (netif_running(dev))
 		return -EBUSY;
@@ -268,7 +268,7 @@ static int ieee80211_ioctl_giwmode(struct net_device *dev,
 	struct ieee80211_sub_if_data *sdata;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_AP:
 		*mode = IW_MODE_MASTER;
 		break;
@@ -336,13 +336,13 @@ static int ieee80211_ioctl_siwfreq(struct net_device *dev,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
-	if (sdata->type == IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA)
 		sdata->u.sta.flags &= ~IEEE80211_STA_AUTO_CHANNEL_SEL;
 
 	/* freq->e == 0: freq->m = channel; otherwise freq = m * 10^e */
 	if (freq->e == 0) {
 		if (freq->m < 0) {
-			if (sdata->type == IEEE80211_IF_TYPE_STA)
+			if (sdata->vif.type == IEEE80211_IF_TYPE_STA)
 				sdata->u.sta.flags |=
 					IEEE80211_STA_AUTO_CHANNEL_SEL;
 			return 0;
@@ -388,8 +388,8 @@ static int ieee80211_ioctl_siwessid(struct net_device *dev,
 		len--;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		int ret;
 		if (sdata->flags & IEEE80211_SDATA_USERSPACE_MLME) {
 			if (len > IEEE80211_MAX_SSID_LEN)
@@ -409,7 +409,7 @@ static int ieee80211_ioctl_siwessid(struct net_device *dev,
 		return 0;
 	}
 
-	if (sdata->type == IEEE80211_IF_TYPE_AP) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		memcpy(sdata->u.ap.ssid, ssid, len);
 		memset(sdata->u.ap.ssid + len, 0,
 		       IEEE80211_MAX_SSID_LEN - len);
@@ -428,8 +428,8 @@ static int ieee80211_ioctl_giwessid(struct net_device *dev,
 
 	struct ieee80211_sub_if_data *sdata;
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		int res = ieee80211_sta_get_ssid(dev, ssid, &len);
 		if (res == 0) {
 			data->length = len;
@@ -439,7 +439,7 @@ static int ieee80211_ioctl_giwessid(struct net_device *dev,
 		return res;
 	}
 
-	if (sdata->type == IEEE80211_IF_TYPE_AP) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		len = sdata->u.ap.ssid_len;
 		if (len > IW_ESSID_MAX_SIZE)
 			len = IW_ESSID_MAX_SIZE;
@@ -459,8 +459,8 @@ static int ieee80211_ioctl_siwap(struct net_device *dev,
 	struct ieee80211_sub_if_data *sdata;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		int ret;
 		if (sdata->flags & IEEE80211_SDATA_USERSPACE_MLME) {
 			memcpy(sdata->u.sta.bssid, (u8 *) &ap_addr->sa_data,
@@ -479,7 +479,7 @@ static int ieee80211_ioctl_siwap(struct net_device *dev,
 			return ret;
 		ieee80211_sta_req_auth(dev, &sdata->u.sta);
 		return 0;
-	} else if (sdata->type == IEEE80211_IF_TYPE_WDS) {
+	} else if (sdata->vif.type == IEEE80211_IF_TYPE_WDS) {
 		if (memcmp(sdata->u.wds.remote_addr, (u8 *) &ap_addr->sa_data,
 			   ETH_ALEN) == 0)
 			return 0;
@@ -497,12 +497,12 @@ static int ieee80211_ioctl_giwap(struct net_device *dev,
 	struct ieee80211_sub_if_data *sdata;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		ap_addr->sa_family = ARPHRD_ETHER;
 		memcpy(&ap_addr->sa_data, sdata->u.sta.bssid, ETH_ALEN);
 		return 0;
-	} else if (sdata->type == IEEE80211_IF_TYPE_WDS) {
+	} else if (sdata->vif.type == IEEE80211_IF_TYPE_WDS) {
 		ap_addr->sa_family = ARPHRD_ETHER;
 		memcpy(&ap_addr->sa_data, sdata->u.wds.remote_addr, ETH_ALEN);
 		return 0;
@@ -524,11 +524,10 @@ static int ieee80211_ioctl_siwscan(struct net_device *dev,
 	if (!netif_running(dev))
 		return -ENETDOWN;
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA &&
-	    sdata->type != IEEE80211_IF_TYPE_IBSS &&
-	    sdata->type != IEEE80211_IF_TYPE_AP) {
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA &&
+	    sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
+	    sdata->vif.type != IEEE80211_IF_TYPE_AP)
 		return -EOPNOTSUPP;
-	}
 
 	/* if SSID was specified explicitly then use that */
 	if (wrqu->data.length == sizeof(struct iw_scan_req) &&
@@ -606,7 +605,7 @@ static int ieee80211_ioctl_giwrate(struct net_device *dev,
 	struct ieee80211_sub_if_data *sdata;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA)
 		sta = sta_info_get(local, sdata->u.sta.bssid);
 	else
 		return -EOPNOTSUPP;
@@ -820,8 +819,8 @@ static int ieee80211_ioctl_siwmlme(struct net_device *dev,
 	struct iw_mlme *mlme = (struct iw_mlme *) extra;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type != IEEE80211_IF_TYPE_STA &&
-	    sdata->type != IEEE80211_IF_TYPE_IBSS)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA &&
+	    sdata->vif.type != IEEE80211_IF_TYPE_IBSS)
 		return -EINVAL;
 
 	switch (mlme->cmd) {
@@ -938,7 +937,7 @@ static int ieee80211_ioctl_siwauth(struct net_device *dev,
 		sdata->drop_unencrypted = !!data->value;
 		break;
 	case IW_AUTH_PRIVACY_INVOKED:
-		if (sdata->type != IEEE80211_IF_TYPE_STA)
+		if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 			ret = -EINVAL;
 		else {
 			sdata->u.sta.flags &= ~IEEE80211_STA_PRIVACY_INVOKED;
@@ -953,8 +952,8 @@ static int ieee80211_ioctl_siwauth(struct net_device *dev,
 		}
 		break;
 	case IW_AUTH_80211_AUTH_ALG:
-		if (sdata->type == IEEE80211_IF_TYPE_STA ||
-		    sdata->type == IEEE80211_IF_TYPE_IBSS)
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+		    sdata->vif.type == IEEE80211_IF_TYPE_IBSS)
 			sdata->u.sta.auth_algs = data->value;
 		else
 			ret = -EOPNOTSUPP;
@@ -974,8 +973,8 @@ static struct iw_statistics *ieee80211_get_wireless_stats(struct net_device *dev
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct sta_info *sta = NULL;
 
-	if (sdata->type == IEEE80211_IF_TYPE_STA ||
-	    sdata->type == IEEE80211_IF_TYPE_IBSS)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS)
 		sta = sta_info_get(local, sdata->u.sta.bssid);
 	if (!sta) {
 		wstats->discard.fragment = 0;
@@ -1003,8 +1002,8 @@ static int ieee80211_ioctl_giwauth(struct net_device *dev,
 
 	switch (data->flags & IW_AUTH_INDEX) {
 	case IW_AUTH_80211_AUTH_ALG:
-		if (sdata->type == IEEE80211_IF_TYPE_STA ||
-		    sdata->type == IEEE80211_IF_TYPE_IBSS)
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+		    sdata->vif.type == IEEE80211_IF_TYPE_IBSS)
 			data->value = sdata->u.sta.auth_algs;
 		else
 			ret = -EOPNOTSUPP;
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index d1f7199a2083..b1e7d17ee253 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -471,7 +471,7 @@ static void ieee80211_set_associated(struct net_device *dev,
 		ifsta->flags |= IEEE80211_STA_ASSOCIATED;
 
 		sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-		if (sdata->type != IEEE80211_IF_TYPE_STA)
+		if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 			return;
 
 		bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
@@ -1025,7 +1025,7 @@ static void ieee80211_send_addba_resp(struct net_device *dev, u8 *da, u16 tid,
 	memset(mgmt, 0, 24);
 	memcpy(mgmt->da, da, ETH_ALEN);
 	memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN);
-	if (sdata->type == IEEE80211_IF_TYPE_AP)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP)
 		memcpy(mgmt->bssid, dev->dev_addr, ETH_ALEN);
 	else
 		memcpy(mgmt->bssid, ifsta->bssid, ETH_ALEN);
@@ -1184,7 +1184,7 @@ void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
 	memset(mgmt, 0, 24);
 	memcpy(mgmt->da, da, ETH_ALEN);
 	memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN);
-	if (sdata->type == IEEE80211_IF_TYPE_AP)
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP)
 		memcpy(mgmt->bssid, dev->dev_addr, ETH_ALEN);
 	else
 		memcpy(mgmt->bssid, ifsta->bssid, ETH_ALEN);
@@ -1329,7 +1329,7 @@ static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 	DECLARE_MAC_BUF(mac);
 
 	if (ifsta->state != IEEE80211_AUTHENTICATE &&
-	    sdata->type != IEEE80211_IF_TYPE_IBSS) {
+	    sdata->vif.type != IEEE80211_IF_TYPE_IBSS) {
 		printk(KERN_DEBUG "%s: authentication frame received from "
 		       "%s, but not in authenticate state - ignored\n",
 		       dev->name, print_mac(mac, mgmt->sa));
@@ -1343,7 +1343,7 @@ static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 		return;
 	}
 
-	if (sdata->type != IEEE80211_IF_TYPE_IBSS &&
+	if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
 	    memcmp(ifsta->bssid, mgmt->sa, ETH_ALEN) != 0) {
 		printk(KERN_DEBUG "%s: authentication frame received from "
 		       "unknown AP (SA=%s BSSID=%s) - "
@@ -1352,7 +1352,7 @@ static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 		return;
 	}
 
-	if (sdata->type != IEEE80211_IF_TYPE_IBSS &&
+	if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
 	    memcmp(ifsta->bssid, mgmt->bssid, ETH_ALEN) != 0) {
 		printk(KERN_DEBUG "%s: authentication frame received from "
 		       "unknown BSSID (SA=%s BSSID=%s) - "
@@ -1370,7 +1370,7 @@ static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 	       dev->name, print_mac(mac, mgmt->sa), auth_alg,
 	       auth_transaction, status_code);
 
-	if (sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		/* IEEE 802.11 standard does not require authentication in IBSS
 		 * networks and most implementations do not seem to use it.
 		 * However, try to reply to authentication attempts if someone
@@ -1849,7 +1849,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 
 	timestamp = le64_to_cpu(mgmt->u.beacon.timestamp);
 
-	if (sdata->type == IEEE80211_IF_TYPE_IBSS && beacon &&
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && beacon &&
 	    memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0) {
 #ifdef CONFIG_MAC80211_IBSS_DEBUG
 		static unsigned long last_tsf_debug = 0;
@@ -1874,7 +1874,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 
 	ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems);
 
-	if (sdata->type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates &&
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates &&
 	    memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0 &&
 	    (sta = sta_info_get(local, mgmt->sa))) {
 		struct ieee80211_hw_mode *mode;
@@ -2103,7 +2103,7 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 	ieee80211_rx_bss_info(dev, mgmt, len, rx_status, 1);
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type != IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 		return;
 	ifsta = &sdata->u.sta;
 
@@ -2163,7 +2163,7 @@ static void ieee80211_rx_mgmt_probe_req(struct net_device *dev,
 	DECLARE_MAC_BUF(mac3);
 #endif
 
-	if (sdata->type != IEEE80211_IF_TYPE_IBSS ||
+	if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS ||
 	    ifsta->state != IEEE80211_IBSS_JOINED ||
 	    len < 24 + 2 || !ifsta->probe_resp)
 		return;
@@ -2474,10 +2474,10 @@ void ieee80211_sta_work(struct work_struct *work)
 	if (local->sta_sw_scanning || local->sta_hw_scanning)
 		return;
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA &&
-	    sdata->type != IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA &&
+	    sdata->vif.type != IEEE80211_IF_TYPE_IBSS) {
 		printk(KERN_DEBUG "%s: ieee80211_sta_work: non-STA interface "
-		       "(type=%d)\n", dev->name, sdata->type);
+		       "(type=%d)\n", dev->name, sdata->vif.type);
 		return;
 	}
 	ifsta = &sdata->u.sta;
@@ -2572,7 +2572,7 @@ void ieee80211_sta_req_auth(struct net_device *dev,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 		return;
 
 	if ((ifsta->flags & (IEEE80211_STA_BSSID_SET |
@@ -3039,7 +3039,7 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 		ifsta->flags |= IEEE80211_STA_SSID_SET;
 	else
 		ifsta->flags &= ~IEEE80211_STA_SSID_SET;
-	if (sdata->type == IEEE80211_IF_TYPE_IBSS &&
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS &&
 	    !(ifsta->flags & IEEE80211_STA_BSSID_SET)) {
 		ifsta->ibss_join_req = jiffies;
 		ifsta->state = IEEE80211_IBSS_SEARCH;
@@ -3157,7 +3157,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 		if (sdata->dev == local->mdev)
 			continue;
 
-		if (sdata->type == IEEE80211_IF_TYPE_STA) {
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA) {
 			if (sdata->u.sta.flags & IEEE80211_STA_ASSOCIATED)
 				ieee80211_send_nullfunc(local, sdata, 0);
 			ieee80211_sta_timer((unsigned long)sdata);
@@ -3169,7 +3169,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 
 done:
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		struct ieee80211_if_sta *ifsta = &sdata->u.sta;
 		if (!(ifsta->flags & IEEE80211_STA_BSSID_SET) ||
 		    (!ifsta->state == IEEE80211_IBSS_JOINED &&
@@ -3204,7 +3204,7 @@ void ieee80211_sta_scan_work(struct work_struct *work)
 		skip = !(local->enabled_modes & (1 << mode->mode));
 		chan = &mode->channels[local->scan_channel_idx];
 		if (!(chan->flag & IEEE80211_CHAN_W_SCAN) ||
-		    (sdata->type == IEEE80211_IF_TYPE_IBSS &&
+		    (sdata->vif.type == IEEE80211_IF_TYPE_IBSS &&
 		     !(chan->flag & IEEE80211_CHAN_W_IBSS)) ||
 		    (local->hw_modes & local->enabled_modes &
 		     (1 << MODE_IEEE80211G) && mode->mode == MODE_IEEE80211B))
@@ -3312,7 +3312,7 @@ static int ieee80211_sta_start_scan(struct net_device *dev,
 			continue;
 
 		netif_stop_queue(sdata->dev);
-		if (sdata->type == IEEE80211_IF_TYPE_STA &&
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA &&
 		    (sdata->u.sta.flags & IEEE80211_STA_ASSOCIATED))
 			ieee80211_send_nullfunc(local, sdata, 1);
 	}
@@ -3353,7 +3353,7 @@ int ieee80211_sta_req_scan(struct net_device *dev, u8 *ssid, size_t ssid_len)
 	struct ieee80211_if_sta *ifsta = &sdata->u.sta;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 		return ieee80211_sta_start_scan(dev, ssid, ssid_len);
 
 	if (local->sta_sw_scanning || local->sta_hw_scanning) {
@@ -3576,8 +3576,8 @@ int ieee80211_sta_deauthenticate(struct net_device *dev, u16 reason)
 	printk(KERN_DEBUG "%s: deauthenticate(reason=%d)\n",
 	       dev->name, reason);
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA &&
-	    sdata->type != IEEE80211_IF_TYPE_IBSS)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA &&
+	    sdata->vif.type != IEEE80211_IF_TYPE_IBSS)
 		return -EINVAL;
 
 	ieee80211_send_deauth(dev, ifsta, reason);
@@ -3594,7 +3594,7 @@ int ieee80211_sta_disassociate(struct net_device *dev, u16 reason)
 	printk(KERN_DEBUG "%s: disassociate(reason=%d)\n",
 	       dev->name, reason);
 
-	if (sdata->type != IEEE80211_IF_TYPE_STA)
+	if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 		return -EINVAL;
 
 	if (!(ifsta->flags & IEEE80211_STA_ASSOCIATED))
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index 0b2328f7d67c..ed57fb8e82fc 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -49,8 +49,8 @@ static const u8 *get_mac_for_key(struct ieee80211_key *key)
 	 * address to indicate a transmit-only key.
 	 */
 	if (key->conf.alg != ALG_WEP &&
-	    (key->sdata->type == IEEE80211_IF_TYPE_AP ||
-	     key->sdata->type == IEEE80211_IF_TYPE_VLAN))
+	    (key->sdata->vif.type == IEEE80211_IF_TYPE_AP ||
+	     key->sdata->vif.type == IEEE80211_IF_TYPE_VLAN))
 		addr = zero_addr;
 
 	if (key->sta)
@@ -172,7 +172,7 @@ struct ieee80211_key *ieee80211_key_alloc(struct ieee80211_sub_if_data *sdata,
 		if (sta->flags & WLAN_STA_WME)
 			key->conf.flags |= IEEE80211_KEY_FLAG_WMM_STA;
 	} else {
-		if (sdata->type == IEEE80211_IF_TYPE_STA) {
+		if (sdata->vif.type == IEEE80211_IF_TYPE_STA) {
 			struct sta_info *ap;
 
 			/* same here, the AP could be using QoS */
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index ed3b81639203..465fce0f62ad 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -223,7 +223,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 		if (!netif_running(sdata->dev))
 			continue;
 
-		if (sdata->type != IEEE80211_IF_TYPE_MNTR)
+		if (sdata->vif.type != IEEE80211_IF_TYPE_MNTR)
 			continue;
 
 		if (prev_dev) {
@@ -419,7 +419,7 @@ ieee80211_rx_h_check(struct ieee80211_txrx_data *rx)
 	if (unlikely(((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA ||
 		      ((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL &&
 		       (rx->fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PSPOLL)) &&
-		     rx->sdata->type != IEEE80211_IF_TYPE_IBSS &&
+		     rx->sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
 		     (!rx->sta || !(rx->sta->flags & WLAN_STA_ASSOC)))) {
 		if ((!(rx->fc & IEEE80211_FCTL_FROMDS) &&
 		     !(rx->fc & IEEE80211_FCTL_TODS) &&
@@ -639,14 +639,14 @@ ieee80211_rx_h_sta_process(struct ieee80211_txrx_data *rx)
 	/* Update last_rx only for IBSS packets which are for the current
 	 * BSSID to avoid keeping the current IBSS network alive in cases where
 	 * other STAs are using different BSSID. */
-	if (rx->sdata->type == IEEE80211_IF_TYPE_IBSS) {
+	if (rx->sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		u8 *bssid = ieee80211_get_bssid(hdr, rx->skb->len,
 						IEEE80211_IF_TYPE_IBSS);
 		if (compare_ether_addr(bssid, rx->sdata->u.sta.bssid) == 0)
 			sta->last_rx = jiffies;
 	} else
 	if (!is_multicast_ether_addr(hdr->addr1) ||
-	    rx->sdata->type == IEEE80211_IF_TYPE_STA) {
+	    rx->sdata->vif.type == IEEE80211_IF_TYPE_STA) {
 		/* Update last_rx only for unicast frames in order to prevent
 		 * the Probe Request frames (the only broadcast frames from a
 		 * STA in infrastructure mode) from keeping a connection alive.
@@ -901,8 +901,8 @@ ieee80211_rx_h_ps_poll(struct ieee80211_txrx_data *rx)
 		   !(rx->flags & IEEE80211_TXRXD_RXRA_MATCH)))
 		return TXRX_CONTINUE;
 
-	if ((sdata->type != IEEE80211_IF_TYPE_AP) &&
-	    (sdata->type != IEEE80211_IF_TYPE_VLAN))
+	if ((sdata->vif.type != IEEE80211_IF_TYPE_AP) &&
+	    (sdata->vif.type != IEEE80211_IF_TYPE_VLAN))
 		return TXRX_DROP;
 
 	skb = skb_dequeue(&rx->sta->tx_filtered);
@@ -1058,8 +1058,8 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 		memcpy(dst, hdr->addr3, ETH_ALEN);
 		memcpy(src, hdr->addr2, ETH_ALEN);
 
-		if (unlikely(sdata->type != IEEE80211_IF_TYPE_AP &&
-			     sdata->type != IEEE80211_IF_TYPE_VLAN)) {
+		if (unlikely(sdata->vif.type != IEEE80211_IF_TYPE_AP &&
+			     sdata->vif.type != IEEE80211_IF_TYPE_VLAN)) {
 			if (net_ratelimit())
 				printk(KERN_DEBUG "%s: dropped ToDS frame "
 				       "(BSSID=%s SA=%s DA=%s)\n",
@@ -1075,7 +1075,7 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 		memcpy(dst, hdr->addr3, ETH_ALEN);
 		memcpy(src, hdr->addr4, ETH_ALEN);
 
-		if (unlikely(sdata->type != IEEE80211_IF_TYPE_WDS)) {
+		if (unlikely(sdata->vif.type != IEEE80211_IF_TYPE_WDS)) {
 			if (net_ratelimit())
 				printk(KERN_DEBUG "%s: dropped FromDS&ToDS "
 				       "frame (RA=%s TA=%s DA=%s SA=%s)\n",
@@ -1092,7 +1092,7 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 		memcpy(dst, hdr->addr1, ETH_ALEN);
 		memcpy(src, hdr->addr3, ETH_ALEN);
 
-		if (sdata->type != IEEE80211_IF_TYPE_STA ||
+		if (sdata->vif.type != IEEE80211_IF_TYPE_STA ||
 		    (is_multicast_ether_addr(dst) &&
 		     !compare_ether_addr(src, dev->dev_addr)))
 			return -1;
@@ -1102,7 +1102,7 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 		memcpy(dst, hdr->addr1, ETH_ALEN);
 		memcpy(src, hdr->addr2, ETH_ALEN);
 
-		if (sdata->type != IEEE80211_IF_TYPE_IBSS) {
+		if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS) {
 			if (net_ratelimit()) {
 				printk(KERN_DEBUG "%s: dropped IBSS frame "
 				       "(DA=%s SA=%s BSSID=%s)\n",
@@ -1190,8 +1190,8 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx)
 	skb = rx->skb;
 	xmit_skb = NULL;
 
-	if (local->bridge_packets && (sdata->type == IEEE80211_IF_TYPE_AP ||
-				      sdata->type == IEEE80211_IF_TYPE_VLAN) &&
+	if (local->bridge_packets && (sdata->vif.type == IEEE80211_IF_TYPE_AP ||
+				      sdata->vif.type == IEEE80211_IF_TYPE_VLAN) &&
 	    (rx->flags & IEEE80211_TXRXD_RXRA_MATCH)) {
 		if (is_multicast_ether_addr(ehdr->h_dest)) {
 			/*
@@ -1435,8 +1435,8 @@ ieee80211_rx_h_mgmt(struct ieee80211_txrx_data *rx)
 		return TXRX_DROP;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(rx->dev);
-	if ((sdata->type == IEEE80211_IF_TYPE_STA ||
-	     sdata->type == IEEE80211_IF_TYPE_IBSS) &&
+	if ((sdata->vif.type == IEEE80211_IF_TYPE_STA ||
+	     sdata->vif.type == IEEE80211_IF_TYPE_IBSS) &&
 	    !(sdata->flags & IEEE80211_SDATA_USERSPACE_MLME))
 		ieee80211_sta_rx_mgmt(rx->dev, rx->skb, rx->u.rx.status);
 	else
@@ -1528,7 +1528,7 @@ static void ieee80211_rx_michael_mic_report(struct net_device *dev,
 		goto ignore;
 	}
 
-	if (rx->sdata->type == IEEE80211_IF_TYPE_AP && keyidx) {
+	if (rx->sdata->vif.type == IEEE80211_IF_TYPE_AP && keyidx) {
 		/*
 		 * APs with pairwise keys should never receive Michael MIC
 		 * errors for non-zero keyidx because these are reserved for
@@ -1590,7 +1590,7 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata,
 {
 	int multicast = is_multicast_ether_addr(hdr->addr1);
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_STA:
 		if (!bssid)
 			return 0;
@@ -1738,10 +1738,10 @@ void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, struct sk_buff *skb,
 		if (!netif_running(sdata->dev))
 			continue;
 
-		if (sdata->type == IEEE80211_IF_TYPE_MNTR)
+		if (sdata->vif.type == IEEE80211_IF_TYPE_MNTR)
 			continue;
 
-		bssid = ieee80211_get_bssid(hdr, skb->len, sdata->type);
+		bssid = ieee80211_get_bssid(hdr, skb->len, sdata->vif.type);
 		rx.flags |= IEEE80211_TXRXD_RXRA_MATCH;
 		prepares = prepare_for_handlers(sdata, bssid, &rx, hdr);
 		/* prepare_for_handlers can change sta */
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 32e241764725..1f74bd296357 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -181,7 +181,7 @@ struct sta_info * sta_info_add(struct ieee80211_local *local,
 		struct ieee80211_sub_if_data *sdata;
 
 		sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-		if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+		if (sdata->vif.type == IEEE80211_IF_TYPE_VLAN)
 			sdata = sdata->u.vlan.ap;
 
 		local->ops->sta_notify(local_to_hw(local), &sdata->vif,
@@ -259,7 +259,7 @@ void sta_info_free(struct sta_info *sta)
 
 		sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
 
-		if (sdata->type == IEEE80211_IF_TYPE_VLAN)
+		if (sdata->vif.type == IEEE80211_IF_TYPE_VLAN)
 			sdata = sdata->u.vlan.ap;
 
 		local->ops->sta_notify(local_to_hw(local), &sdata->vif,
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 1b772ee2fe14..99590e4ce30b 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -237,7 +237,7 @@ ieee80211_tx_h_check_assoc(struct ieee80211_txrx_data *tx)
 
 	if (likely(tx->flags & IEEE80211_TXRXD_TXUNICAST)) {
 		if (unlikely(!(sta_flags & WLAN_STA_ASSOC) &&
-			     tx->sdata->type != IEEE80211_IF_TYPE_IBSS &&
+			     tx->sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
 			     (tx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA)) {
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
 			DECLARE_MAC_BUF(mac);
@@ -251,7 +251,7 @@ ieee80211_tx_h_check_assoc(struct ieee80211_txrx_data *tx)
 	} else {
 		if (unlikely((tx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA &&
 			     tx->local->num_sta == 0 &&
-			     tx->sdata->type != IEEE80211_IF_TYPE_IBSS)) {
+			     tx->sdata->vif.type != IEEE80211_IF_TYPE_IBSS)) {
 			/*
 			 * No associated STAs - no need to send multicast
 			 * frames.
@@ -294,7 +294,7 @@ static void purge_old_ps_buffers(struct ieee80211_local *local)
 	list_for_each_entry_rcu(sdata, &local->interfaces, list) {
 		struct ieee80211_if_ap *ap;
 		if (sdata->dev == local->mdev ||
-		    sdata->type != IEEE80211_IF_TYPE_AP)
+		    sdata->vif.type != IEEE80211_IF_TYPE_AP)
 			continue;
 		ap = &sdata->u.ap;
 		skb = skb_dequeue(&ap->ps_bc_buf);
@@ -949,7 +949,7 @@ __ieee80211_tx_prepare(struct ieee80211_txrx_data *tx,
 
 	/* process and remove the injection radiotap header */
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (unlikely(sdata->type == IEEE80211_IF_TYPE_MNTR)) {
+	if (unlikely(sdata->vif.type == IEEE80211_IF_TYPE_MNTR)) {
 		if (__ieee80211_parse_tx_radiotap(tx, skb) == TXRX_DROP)
 			return TXRX_DROP;
 
@@ -1252,7 +1252,7 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 	}
 
 	control.vif = &osdata->vif;
-	control.type = osdata->type;
+	control.type = osdata->vif.type;
 	if (pkt_data->flags & IEEE80211_TXPD_REQ_TX_STATUS)
 		control.flags |= IEEE80211_TXCTL_REQ_TX_STATUS;
 	if (pkt_data->flags & IEEE80211_TXPD_DO_NOT_ENCRYPT)
@@ -1371,7 +1371,7 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 	ethertype = (skb->data[12] << 8) | skb->data[13];
 	fc = IEEE80211_FTYPE_DATA | IEEE80211_STYPE_DATA;
 
-	switch (sdata->type) {
+	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_AP:
 	case IEEE80211_IF_TYPE_VLAN:
 		fc |= IEEE80211_FCTL_FROMDS;
@@ -1707,7 +1707,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	bdev = sdata->dev;
 	ap = &sdata->u.ap;
 
-	if (!ap || sdata->type != IEEE80211_IF_TYPE_AP ||
+	if (!ap || sdata->vif.type != IEEE80211_IF_TYPE_AP ||
 	    !ap->beacon_head) {
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
 		if (net_ratelimit())
@@ -1819,7 +1819,8 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 	sdata = vif_to_sdata(vif);
 	bdev = sdata->dev;
 
-	if (!bss || sdata->type != IEEE80211_IF_TYPE_AP || !bss->beacon_head)
+	if (!bss || sdata->vif.type != IEEE80211_IF_TYPE_AP ||
+	    !bss->beacon_head)
 		return NULL;
 
 	if (bss->dtim_count != 0)
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 15503ca3e945..ba81cf542818 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -471,7 +471,7 @@ void ieee80211_iterate_active_interfaces(
 	rcu_read_lock();
 
 	list_for_each_entry_rcu(sdata, &local->interfaces, list) {
-		switch (sdata->type) {
+		switch (sdata->vif.type) {
 		case IEEE80211_IF_TYPE_INVALID:
 		case IEEE80211_IF_TYPE_MNTR:
 		case IEEE80211_IF_TYPE_VLAN:
-- 
cgit v1.2.3


From 471b3efdfccc257591331724145f8ccf8b3217e1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 28 Dec 2007 14:32:58 +0100
Subject: mac80211: add unified BSS configuration

This patch (based on Ron Rindjunsky's) creates a framework for
a unified way to pass BSS configuration to drivers that require
the information, e.g. for implementing power save mode.

This patch introduces new ieee80211_bss_conf structure that is
passed to the driver via the new bss_info_changed() callback
when the BSS configuration changes.

This new BSS configuration infrastructure adds the following
new features:
 * drivers are notified of their association AID
 * drivers are notified of association status

and replaces the erp_ie_changed() callback. The patch also does
the relevant driver updates for the latter change.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-3945.c     |  6 +--
 drivers/net/wireless/iwlwifi/iwl-4965.c     |  6 +--
 drivers/net/wireless/iwlwifi/iwl4965-base.c | 24 +++++++---
 drivers/net/wireless/rt2x00/rt2400pci.c     |  2 +-
 drivers/net/wireless/rt2x00/rt2500pci.c     |  2 +-
 drivers/net/wireless/rt2x00/rt2500usb.c     |  2 +-
 drivers/net/wireless/rt2x00/rt2x00.h        |  6 ++-
 drivers/net/wireless/rt2x00/rt2x00dev.c     | 13 ++++--
 drivers/net/wireless/rt2x00/rt2x00mac.c     | 16 ++++---
 drivers/net/wireless/rt2x00/rt61pci.c       |  2 +-
 drivers/net/wireless/rt2x00/rt73usb.c       |  2 +-
 drivers/net/wireless/zd1211rw/zd_mac.c      | 12 ++---
 include/net/mac80211.h                      | 66 ++++++++++++++++++++-------
 net/mac80211/debugfs_netdev.c               |  2 +-
 net/mac80211/ieee80211.c                    | 29 +++++++-----
 net/mac80211/ieee80211_i.h                  | 19 +++++---
 net/mac80211/ieee80211_sta.c                | 69 +++++++++++++++--------------
 net/mac80211/tx.c                           | 13 +++---
 net/mac80211/util.c                         | 12 ++---
 19 files changed, 183 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index a793cd11f738..77e7202c026b 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -520,10 +520,8 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 			break;
 
 			/*
-			 * TODO: There is no callback function from upper
-			 * stack to inform us when associated status. this
-			 * work around to sniff assoc_resp management frame
-			 * and finish the association process.
+			 * TODO: Use the new callback function from
+			 * mac80211 instead of sniffing these packets.
 			 */
 		case IEEE80211_STYPE_ASSOC_RESP:
 		case IEEE80211_STYPE_REASSOC_RESP:{
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index 89e2c44bfcca..ed3f119b56cd 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -4095,10 +4095,8 @@ static void iwl4965_rx_reply_rx(struct iwl4965_priv *priv,
 			break;
 
 			/*
-			 * TODO: There is no callback function from upper
-			 * stack to inform us when associated status. this
-			 * work around to sniff assoc_resp management frame
-			 * and finish the association process.
+			 * TODO: Use the new callback function from
+			 * mac80211 instead of sniffing these packets.
 			 */
 		case IEEE80211_STYPE_ASSOC_RESP:
 		case IEEE80211_STYPE_REASSOC_RESP:
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 92bb7e13a1f0..2597c08a2395 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -7764,25 +7764,35 @@ static void iwl4965_mac_remove_interface(struct ieee80211_hw *hw,
 	IWL_DEBUG_MAC80211("leave\n");
 
 }
-static void iwl4965_mac_erp_ie_changed(struct ieee80211_hw *hw,
-		u8 changes, int cts_protection, int preamble)
+
+static void iwl4965_bss_info_changed(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *bss_conf,
+				     u32 changes)
 {
 	struct iwl4965_priv *priv = hw->priv;
 
-	if (changes & IEEE80211_ERP_CHANGE_PREAMBLE) {
-		if (preamble == WLAN_ERP_PREAMBLE_SHORT)
+	if (changes & BSS_CHANGED_ERP_PREAMBLE) {
+		if (bss_conf->use_short_preamble)
 			priv->staging_rxon.flags |= RXON_FLG_SHORT_PREAMBLE_MSK;
 		else
 			priv->staging_rxon.flags &= ~RXON_FLG_SHORT_PREAMBLE_MSK;
 	}
 
-	if (changes & IEEE80211_ERP_CHANGE_PROTECTION) {
-		if (cts_protection && (priv->phymode != MODE_IEEE80211A))
+	if (changes & BSS_CHANGED_ERP_CTS_PROT) {
+		if (bss_conf->use_cts_prot && (priv->phymode != MODE_IEEE80211A))
 			priv->staging_rxon.flags |= RXON_FLG_TGG_PROTECT_MSK;
 		else
 			priv->staging_rxon.flags &= ~RXON_FLG_TGG_PROTECT_MSK;
 	}
 
+	if (changes & BSS_CHANGED_ASSOC) {
+		/*
+		 * TODO:
+		 * do stuff instead of sniffing assoc resp
+		 */
+	}
+
 	if (iwl4965_is_associated(priv))
 		iwl4965_send_rxon_assoc(priv);
 }
@@ -8952,7 +8962,7 @@ static struct ieee80211_ops iwl4965_hw_ops = {
 	.get_tsf = iwl4965_mac_get_tsf,
 	.reset_tsf = iwl4965_mac_reset_tsf,
 	.beacon_update = iwl4965_mac_beacon_update,
-	.erp_ie_changed = iwl4965_mac_erp_ie_changed,
+	.bss_info_changed = iwl4965_bss_info_changed,
 #ifdef CONFIG_IWL4965_HT
 	.conf_ht = iwl4965_mac_conf_ht,
 	.ampdu_action = iwl4965_mac_ampdu_action,
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index f4d0c779ac78..d6cba138c7ab 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -1524,7 +1524,7 @@ static const struct ieee80211_ops rt2400pci_mac80211_ops = {
 	.configure_filter	= rt2400pci_configure_filter,
 	.get_stats		= rt2x00mac_get_stats,
 	.set_retry_limit	= rt2400pci_set_retry_limit,
-	.erp_ie_changed		= rt2x00mac_erp_ie_changed,
+	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2400pci_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt2400pci_get_tsf,
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 206b50ff831f..e874fdcae204 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -1835,7 +1835,7 @@ static const struct ieee80211_ops rt2500pci_mac80211_ops = {
 	.configure_filter	= rt2500pci_configure_filter,
 	.get_stats		= rt2x00mac_get_stats,
 	.set_retry_limit	= rt2500pci_set_retry_limit,
-	.erp_ie_changed		= rt2x00mac_erp_ie_changed,
+	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt2500pci_get_tsf,
diff --git a/drivers/net/wireless/rt2x00/rt2500usb.c b/drivers/net/wireless/rt2x00/rt2500usb.c
index 6579718f36ad..1933113d7baf 100644
--- a/drivers/net/wireless/rt2x00/rt2500usb.c
+++ b/drivers/net/wireless/rt2x00/rt2500usb.c
@@ -1776,7 +1776,7 @@ static const struct ieee80211_ops rt2500usb_mac80211_ops = {
 	.config_interface	= rt2x00mac_config_interface,
 	.configure_filter	= rt2500usb_configure_filter,
 	.get_stats		= rt2x00mac_get_stats,
-	.erp_ie_changed		= rt2x00mac_erp_ie_changed,
+	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.beacon_update		= rt2500usb_beacon_update,
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 94a8a7ce5382..05927b908f80 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -936,8 +936,10 @@ int rt2x00mac_get_stats(struct ieee80211_hw *hw,
 			struct ieee80211_low_level_stats *stats);
 int rt2x00mac_get_tx_stats(struct ieee80211_hw *hw,
 			   struct ieee80211_tx_queue_stats *stats);
-void rt2x00mac_erp_ie_changed(struct ieee80211_hw *hw, u8 changes,
-			      int cts_protection, int preamble);
+void rt2x00mac_bss_info_changed(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
+				u32 changes);
 int rt2x00mac_conf_tx(struct ieee80211_hw *hw, int queue,
 		      const struct ieee80211_tx_queue_params *params);
 
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index b26c634f210b..72cfe6f866f8 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -481,10 +481,17 @@ static void rt2x00lib_configuration_scheduled(struct work_struct *work)
 {
 	struct rt2x00_dev *rt2x00dev =
 	    container_of(work, struct rt2x00_dev, config_work);
-	int preamble = !test_bit(CONFIG_SHORT_PREAMBLE, &rt2x00dev->flags);
+	struct ieee80211_bss_conf bss_conf;
 
-	rt2x00mac_erp_ie_changed(rt2x00dev->hw,
-				 IEEE80211_ERP_CHANGE_PREAMBLE, 0, preamble);
+	bss_conf.use_short_preamble =
+		test_bit(CONFIG_SHORT_PREAMBLE, &rt2x00dev->flags);
+
+	/*
+	 * FIXME: shouldn't invoke it this way because all other contents
+	 *	  of bss_conf is invalid.
+	 */
+	rt2x00mac_bss_info_changed(rt2x00dev->hw, rt2x00dev->interface.id,
+				   &bss_conf, BSS_CHANGED_ERP_PREAMBLE);
 }
 
 /*
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index 1d67bcd46bcb..e3f15e518c76 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -342,23 +342,27 @@ int rt2x00mac_get_tx_stats(struct ieee80211_hw *hw,
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_get_tx_stats);
 
-void rt2x00mac_erp_ie_changed(struct ieee80211_hw *hw, u8 changes,
-			      int cts_protection, int preamble)
+void rt2x00mac_bss_info_changed(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
+				u32 changes)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 	int short_preamble;
 	int ack_timeout;
 	int ack_consume_time;
 	int difs;
+	int preamble;
 
 	/*
 	 * We only support changing preamble mode.
 	 */
-	if (!(changes & IEEE80211_ERP_CHANGE_PREAMBLE))
+	if (!(changes & BSS_CHANGED_ERP_PREAMBLE))
 		return;
 
-	short_preamble = !preamble;
-	preamble = !!(preamble) ? PREAMBLE : SHORT_PREAMBLE;
+	short_preamble = bss_conf->use_short_preamble;
+	preamble = bss_conf->use_short_preamble ?
+				SHORT_PREAMBLE : PREAMBLE;
 
 	difs = (hw->conf.flags & IEEE80211_CONF_SHORT_SLOT_TIME) ?
 		SHORT_DIFS : DIFS;
@@ -374,7 +378,7 @@ void rt2x00mac_erp_ie_changed(struct ieee80211_hw *hw, u8 changes,
 	rt2x00dev->ops->lib->config_preamble(rt2x00dev, short_preamble,
 					     ack_timeout, ack_consume_time);
 }
-EXPORT_SYMBOL_GPL(rt2x00mac_erp_ie_changed);
+EXPORT_SYMBOL_GPL(rt2x00mac_bss_info_changed);
 
 int rt2x00mac_conf_tx(struct ieee80211_hw *hw, int queue,
 		      const struct ieee80211_tx_queue_params *params)
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index 88d169800886..ab52f221cd71 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -2441,7 +2441,7 @@ static const struct ieee80211_ops rt61pci_mac80211_ops = {
 	.configure_filter	= rt61pci_configure_filter,
 	.get_stats		= rt2x00mac_get_stats,
 	.set_retry_limit	= rt61pci_set_retry_limit,
-	.erp_ie_changed		= rt2x00mac_erp_ie_changed,
+	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt61pci_get_tsf,
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index 47ed3070ae0b..f9518764aeb0 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -2024,7 +2024,7 @@ static const struct ieee80211_ops rt73usb_mac80211_ops = {
 	.configure_filter	= rt73usb_configure_filter,
 	.get_stats		= rt2x00mac_get_stats,
 	.set_retry_limit	= rt73usb_set_retry_limit,
-	.erp_ie_changed		= rt2x00mac_erp_ie_changed,
+	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt73usb_get_tsf,
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 3409cf97f5f8..49127e4b42c2 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -849,17 +849,19 @@ static void set_rts_cts_work(struct work_struct *work)
 	mutex_unlock(&mac->chip.mutex);
 }
 
-static void zd_op_erp_ie_changed(struct ieee80211_hw *hw, u8 changes,
-				 int cts_protection, int preamble)
+static void zd_op_bss_info_changed(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_bss_conf *bss_conf,
+				   u32 changes)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
 	unsigned long flags;
 
 	dev_dbg_f(zd_mac_dev(mac), "changes: %x\n", changes);
 
-	if (changes & IEEE80211_ERP_CHANGE_PREAMBLE) {
+	if (changes & BSS_CHANGED_ERP_PREAMBLE) {
 		spin_lock_irqsave(&mac->lock, flags);
-		mac->short_preamble = !preamble;
+		mac->short_preamble = bss_conf->use_short_preamble;
 		if (!mac->updating_rts_rate) {
 			mac->updating_rts_rate = 1;
 			/* FIXME: should disable TX here, until work has
@@ -879,7 +881,7 @@ static const struct ieee80211_ops zd_ops = {
 	.config			= zd_op_config,
 	.config_interface	= zd_op_config_interface,
 	.configure_filter	= zd_op_configure_filter,
-	.erp_ie_changed		= zd_op_erp_ie_changed,
+	.bss_info_changed	= zd_op_bss_info_changed,
 };
 
 struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 8a49c06dfd38..9083bafb63ca 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -275,6 +275,43 @@ struct ieee80211_low_level_stats {
 	unsigned int dot11RTSSuccessCount;
 };
 
+/**
+ * enum ieee80211_bss_change - BSS change notification flags
+ *
+ * These flags are used with the bss_info_changed() callback
+ * to indicate which BSS parameter changed.
+ *
+ * @BSS_CHANGED_ASSOC: association status changed (associated/disassociated),
+ *	also implies a change in the AID.
+ * @BSS_CHANGED_ERP_CTS_PROT: CTS protection changed
+ * @BSS_CHANGED_ERP_PREAMBLE: preamble changed
+ */
+enum ieee80211_bss_change {
+	BSS_CHANGED_ASSOC		= 1<<0,
+	BSS_CHANGED_ERP_CTS_PROT	= 1<<1,
+	BSS_CHANGED_ERP_PREAMBLE	= 1<<2,
+};
+
+/**
+ * struct ieee80211_bss_conf - holds the BSS's changing parameters
+ *
+ * This structure keeps information about a BSS (and an association
+ * to that BSS) that can change during the lifetime of the BSS.
+ *
+ * @assoc: association status
+ * @aid: association ID number, valid only when @assoc is true
+ * @use_cts_prot: use CTS protection
+ * @use_short_preamble: use 802.11b short preamble
+ */
+struct ieee80211_bss_conf {
+	/* association related data */
+	bool assoc;
+	u16 aid;
+	/* erp related data */
+	bool use_cts_prot;
+	bool use_short_preamble;
+};
+
 /* Transmit control fields. This data structure is passed to low-level driver
  * with each TX frame. The low-level driver is responsible for configuring
  * the hardware to use given values (depending on what is supported). */
@@ -923,19 +960,6 @@ enum ieee80211_filter_flags {
 	FIF_OTHER_BSS		= 1<<6,
 };
 
-/**
- * enum ieee80211_erp_change_flags - erp change flags
- *
- * These flags are used with the erp_ie_changed() callback in
- * &struct ieee80211_ops to indicate which parameter(s) changed.
- * @IEEE80211_ERP_CHANGE_PROTECTION: protection changed
- * @IEEE80211_ERP_CHANGE_PREAMBLE: barker preamble mode changed
- */
-enum ieee80211_erp_change_flags {
-	IEEE80211_ERP_CHANGE_PROTECTION	= 1<<0,
-	IEEE80211_ERP_CHANGE_PREAMBLE	= 1<<1,
-};
-
 /**
  * enum ieee80211_ampdu_mlme_action - A-MPDU actions
  *
@@ -1004,6 +1028,14 @@ enum ieee80211_ampdu_mlme_action {
  * @config_interface: Handler for configuration requests related to interfaces
  *	(e.g. BSSID changes.)
  *
+ * @bss_info_changed: Handler for configuration requests related to BSS
+ *	parameters that may vary during BSS's lifespan, and may affect low
+ *	level driver (e.g. assoc/disassoc status, erp parameters).
+ *	This function should not be used if no BSS has been set, unless
+ *	for association indication. The @changed parameter indicates which
+ *	of the bss parameters has changed when a call is made. This callback
+ *	has to be atomic.
+ *
  * @configure_filter: Configure the device's RX filter.
  *	See the section "Frame filtering" for more information.
  *	This callback must be implemented and atomic.
@@ -1038,8 +1070,6 @@ enum ieee80211_ampdu_mlme_action {
  * @sta_notify: Notifies low level driver about addition or removal
  *	of assocaited station or AP.
  *
- * @erp_ie_changed: Handle ERP IE change notifications. Must be atomic.
- *
  * @conf_tx: Configure TX queue parameters (EDCF (aifs, cw_min, cw_max),
  *	bursting) for a hardware TX queue. The @queue parameter uses the
  *	%IEEE80211_TX_QUEUE_* constants. Must be atomic.
@@ -1096,6 +1126,10 @@ struct ieee80211_ops {
 	int (*config_interface)(struct ieee80211_hw *hw,
 				struct ieee80211_vif *vif,
 				struct ieee80211_if_conf *conf);
+	void (*bss_info_changed)(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_bss_conf *info,
+				 u32 changed);
 	void (*configure_filter)(struct ieee80211_hw *hw,
 				 unsigned int changed_flags,
 				 unsigned int *total_flags,
@@ -1115,8 +1149,6 @@ struct ieee80211_ops {
 			       u32 short_retry, u32 long_retr);
 	void (*sta_notify)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			enum sta_notify_cmd, const u8 *addr);
-	void (*erp_ie_changed)(struct ieee80211_hw *hw, u8 changes,
-			       int cts_protection, int preamble);
 	int (*conf_tx)(struct ieee80211_hw *hw, int queue,
 		       const struct ieee80211_tx_queue_params *params);
 	int (*get_tx_stats)(struct ieee80211_hw *hw,
diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c
index 3500fe0d380b..829872a3ae81 100644
--- a/net/mac80211/debugfs_netdev.c
+++ b/net/mac80211/debugfs_netdev.c
@@ -118,7 +118,7 @@ static ssize_t ieee80211_if_fmt_flags(
 		 sdata->u.sta.flags & IEEE80211_STA_AUTHENTICATED ? "AUTH\n" : "",
 		 sdata->u.sta.flags & IEEE80211_STA_ASSOCIATED ? "ASSOC\n" : "",
 		 sdata->u.sta.flags & IEEE80211_STA_PROBEREQ_POLL ? "PROBEREQ POLL\n" : "",
-		 sdata->flags & IEEE80211_SDATA_USE_PROTECTION ? "CTS prot\n" : "");
+		 sdata->bss_conf.use_cts_prot ? "CTS prot\n" : "");
 }
 __IEEE80211_IF_FILE(flags);
 
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 859682eee547..177040218232 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -635,25 +635,30 @@ int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
 	return 0;
 }
 
-void ieee80211_erp_info_change_notify(struct net_device *dev, u8 changes)
+void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata,
+				      u32 changed)
 {
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	if (local->ops->erp_ie_changed)
-		local->ops->erp_ie_changed(local_to_hw(local), changes,
-			!!(sdata->flags & IEEE80211_SDATA_USE_PROTECTION),
-			!(sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE));
+	struct ieee80211_local *local = sdata->local;
+
+	if (!changed)
+		return;
+
+	if (local->ops->bss_info_changed)
+		local->ops->bss_info_changed(local_to_hw(local),
+					     &sdata->vif,
+					     &sdata->bss_conf,
+					     changed);
 }
 
 void ieee80211_reset_erp_info(struct net_device *dev)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
-	sdata->flags &= ~(IEEE80211_SDATA_USE_PROTECTION |
-			IEEE80211_SDATA_SHORT_PREAMBLE);
-	ieee80211_erp_info_change_notify(dev,
-					 IEEE80211_ERP_CHANGE_PROTECTION |
-					 IEEE80211_ERP_CHANGE_PREAMBLE);
+	sdata->bss_conf.use_cts_prot = 0;
+	sdata->bss_conf.use_short_preamble = 0;
+	ieee80211_bss_info_change_notify(sdata,
+					 BSS_CHANGED_ERP_CTS_PROT |
+					 BSS_CHANGED_ERP_PREAMBLE);
 }
 
 void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 08a6c0cff690..72ecbf7bf962 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -291,12 +291,7 @@ struct ieee80211_if_sta {
 /* flags used in struct ieee80211_sub_if_data.flags */
 #define IEEE80211_SDATA_ALLMULTI	BIT(0)
 #define IEEE80211_SDATA_PROMISC		BIT(1)
-#define IEEE80211_SDATA_USE_PROTECTION	BIT(2) /* CTS protect ERP frames */
-/* use short preamble with IEEE 802.11b: this flag is set when the AP or beacon
- * generator reports that there are no present stations that cannot support short
- * preambles */
-#define IEEE80211_SDATA_SHORT_PREAMBLE	BIT(3)
-#define IEEE80211_SDATA_USERSPACE_MLME	BIT(4)
+#define IEEE80211_SDATA_USERSPACE_MLME	BIT(2)
 struct ieee80211_sub_if_data {
 	struct list_head list;
 
@@ -327,6 +322,15 @@ struct ieee80211_sub_if_data {
 	struct ieee80211_key *keys[NUM_DEFAULT_KEYS];
 	struct ieee80211_key *default_key;
 
+	/*
+	 * BSS configuration for this interface.
+	 *
+	 * FIXME: I feel bad putting this here when we already have a
+	 *	  bss pointer, but the bss pointer is just wrong when
+	 *	  you have multiple virtual STA mode interfaces...
+	 *	  This needs to be fixed.
+	 */
+	struct ieee80211_bss_conf bss_conf;
 	struct ieee80211_if_ap *bss; /* BSS that this device belongs to */
 
 	union {
@@ -770,7 +774,8 @@ struct sta_info * ieee80211_ibss_add_sta(struct net_device *dev,
 					 u8 *addr);
 int ieee80211_sta_deauthenticate(struct net_device *dev, u16 reason);
 int ieee80211_sta_disassociate(struct net_device *dev, u16 reason);
-void ieee80211_erp_info_change_notify(struct net_device *dev, u8 changes);
+void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata,
+				      u32 changed);
 void ieee80211_reset_erp_info(struct net_device *dev);
 int ieee80211_ht_cap_ie_to_ht_info(struct ieee80211_ht_cap *ht_cap_ie,
 				   struct ieee80211_ht_info *ht_info);
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index b1e7d17ee253..866eb8078121 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -313,48 +313,42 @@ static void ieee80211_sta_wmm_params(struct net_device *dev,
 }
 
 
-static void ieee80211_handle_erp_ie(struct net_device *dev, u8 erp_value)
+static u32 ieee80211_handle_erp_ie(struct ieee80211_sub_if_data *sdata,
+				   u8 erp_value)
 {
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_bss_conf *bss_conf = &sdata->bss_conf;
 	struct ieee80211_if_sta *ifsta = &sdata->u.sta;
-	int use_protection = (erp_value & WLAN_ERP_USE_PROTECTION) != 0;
-	int preamble_mode = (erp_value & WLAN_ERP_BARKER_PREAMBLE) != 0;
-	u8 changes = 0;
+	bool use_protection = (erp_value & WLAN_ERP_USE_PROTECTION) != 0;
+	bool preamble_mode = (erp_value & WLAN_ERP_BARKER_PREAMBLE) != 0;
 	DECLARE_MAC_BUF(mac);
+	u32 changed = 0;
 
-	if (use_protection != !!(sdata->flags & IEEE80211_SDATA_USE_PROTECTION)) {
+	if (use_protection != bss_conf->use_cts_prot) {
 		if (net_ratelimit()) {
 			printk(KERN_DEBUG "%s: CTS protection %s (BSSID="
 			       "%s)\n",
-			       dev->name,
+			       sdata->dev->name,
 			       use_protection ? "enabled" : "disabled",
 			       print_mac(mac, ifsta->bssid));
 		}
-		if (use_protection)
-			sdata->flags |= IEEE80211_SDATA_USE_PROTECTION;
-		else
-			sdata->flags &= ~IEEE80211_SDATA_USE_PROTECTION;
-		changes |= IEEE80211_ERP_CHANGE_PROTECTION;
+		bss_conf->use_cts_prot = use_protection;
+		changed |= BSS_CHANGED_ERP_CTS_PROT;
 	}
 
-	if (preamble_mode != !(sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE)) {
+	if (preamble_mode != bss_conf->use_short_preamble) {
 		if (net_ratelimit()) {
 			printk(KERN_DEBUG "%s: switched to %s barker preamble"
 			       " (BSSID=%s)\n",
-			       dev->name,
+			       sdata->dev->name,
 			       (preamble_mode == WLAN_ERP_PREAMBLE_SHORT) ?
 					"short" : "long",
 			       print_mac(mac, ifsta->bssid));
 		}
-		if (preamble_mode)
-			sdata->flags &= ~IEEE80211_SDATA_SHORT_PREAMBLE;
-		else
-			sdata->flags |= IEEE80211_SDATA_SHORT_PREAMBLE;
-		changes |= IEEE80211_ERP_CHANGE_PREAMBLE;
+		bss_conf->use_short_preamble = preamble_mode;
+		changed |= BSS_CHANGED_ERP_PREAMBLE;
 	}
 
-	if (changes)
-		ieee80211_erp_info_change_notify(dev, changes);
+	return changed;
 }
 
 int ieee80211_ht_cap_ie_to_ht_info(struct ieee80211_ht_cap *ht_cap_ie,
@@ -458,19 +452,16 @@ static void ieee80211_set_associated(struct net_device *dev,
 				     struct ieee80211_if_sta *ifsta,
 				     bool assoc)
 {
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = sdata->local;
 	union iwreq_data wrqu;
-
-	if (!!(ifsta->flags & IEEE80211_STA_ASSOCIATED) == assoc)
-		return;
+	u32 changed = BSS_CHANGED_ASSOC;
 
 	if (assoc) {
-		struct ieee80211_sub_if_data *sdata;
 		struct ieee80211_sta_bss *bss;
 
 		ifsta->flags |= IEEE80211_STA_ASSOCIATED;
 
-		sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 		if (sdata->vif.type != IEEE80211_IF_TYPE_STA)
 			return;
 
@@ -479,7 +470,8 @@ static void ieee80211_set_associated(struct net_device *dev,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
 			if (bss->has_erp_value)
-				ieee80211_handle_erp_ie(dev, bss->erp_value);
+				changed |= ieee80211_handle_erp_ie(
+						sdata, bss->erp_value);
 			ieee80211_rx_bss_put(dev, bss);
 		}
 
@@ -499,6 +491,8 @@ static void ieee80211_set_associated(struct net_device *dev,
 	wireless_send_event(dev, SIOCGIWAP, &wrqu, NULL);
 	ifsta->last_probe = jiffies;
 	ieee80211_led_assoc(local, assoc);
+
+	ieee80211_bss_info_change_notify(sdata, changed);
 }
 
 static void ieee80211_set_disassoc(struct net_device *dev,
@@ -1536,18 +1530,20 @@ static void ieee80211_rx_mgmt_disassoc(struct net_device *dev,
 }
 
 
-static void ieee80211_rx_mgmt_assoc_resp(struct net_device *dev,
+static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 					 struct ieee80211_if_sta *ifsta,
 					 struct ieee80211_mgmt *mgmt,
 					 size_t len,
 					 int reassoc)
 {
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_local *local = sdata->local;
+	struct net_device *dev = sdata->dev;
 	struct ieee80211_hw_mode *mode;
 	struct sta_info *sta;
 	u32 rates;
 	u16 capab_info, status_code, aid;
 	struct ieee802_11_elems elems;
+	struct ieee80211_bss_conf *bss_conf = &sdata->bss_conf;
 	u8 *pos;
 	int i, j;
 	DECLARE_MAC_BUF(mac);
@@ -1620,6 +1616,8 @@ static void ieee80211_rx_mgmt_assoc_resp(struct net_device *dev,
 	if (ifsta->assocresp_ies)
 		memcpy(ifsta->assocresp_ies, pos, ifsta->assocresp_ies_len);
 
+	/* set AID, ieee80211_set_associated() will tell the driver */
+	bss_conf->aid = aid;
 	ieee80211_set_associated(dev, ifsta, 1);
 
 	/* Add STA entry for the AP */
@@ -2099,6 +2097,7 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 	struct ieee802_11_elems elems;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_conf *conf = &local->hw.conf;
+	u32 changed = 0;
 
 	ieee80211_rx_bss_info(dev, mgmt, len, rx_status, 1);
 
@@ -2119,7 +2118,7 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 	ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems);
 
 	if (elems.erp_info && elems.erp_info_len >= 1)
-		ieee80211_handle_erp_ie(dev, elems.erp_info[0]);
+		changed |= ieee80211_handle_erp_ie(sdata, elems.erp_info[0]);
 
 	if (elems.ht_cap_elem && elems.ht_info_elem &&
 	    elems.wmm_param && local->ops->conf_ht &&
@@ -2142,6 +2141,8 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 		ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param,
 					 elems.wmm_param_len);
 	}
+
+	ieee80211_bss_info_change_notify(sdata, changed);
 }
 
 
@@ -2329,10 +2330,10 @@ static void ieee80211_sta_rx_queued_mgmt(struct net_device *dev,
 		ieee80211_rx_mgmt_auth(dev, ifsta, mgmt, skb->len);
 		break;
 	case IEEE80211_STYPE_ASSOC_RESP:
-		ieee80211_rx_mgmt_assoc_resp(dev, ifsta, mgmt, skb->len, 0);
+		ieee80211_rx_mgmt_assoc_resp(sdata, ifsta, mgmt, skb->len, 0);
 		break;
 	case IEEE80211_STYPE_REASSOC_RESP:
-		ieee80211_rx_mgmt_assoc_resp(dev, ifsta, mgmt, skb->len, 1);
+		ieee80211_rx_mgmt_assoc_resp(sdata, ifsta, mgmt, skb->len, 1);
 		break;
 	case IEEE80211_STYPE_DEAUTH:
 		ieee80211_rx_mgmt_deauth(dev, ifsta, mgmt, skb->len);
@@ -2787,7 +2788,7 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 			break;
 		}
 		control.tx_rate =
-			((sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE) &&
+			(sdata->bss_conf.use_short_preamble &&
 			(ratesel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ?
 			ratesel.rate->val2 : ratesel.rate->val;
 		control.antenna_sel_tx = local->hw.conf.antenna_sel_tx;
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 51c0f00d02d1..f9088fe34d59 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -176,7 +176,7 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr,
 	 * to closest integer */
 
 	dur = ieee80211_frame_duration(local, 10, rate, erp,
-		       tx->sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE);
+				tx->sdata->bss_conf.use_short_preamble);
 
 	if (next_frag_len) {
 		/* Frame is fragmented: duration increases with time needed to
@@ -185,8 +185,7 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr,
 		/* next fragment */
 		dur += ieee80211_frame_duration(local, next_frag_len,
 				txrate->rate, erp,
-				tx->sdata->flags &
-					IEEE80211_SDATA_SHORT_PREAMBLE);
+				tx->sdata->bss_conf.use_short_preamble);
 	}
 
 	return dur;
@@ -605,7 +604,7 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_txrx_data *tx)
 		tx->u.tx.control->alt_retry_rate = -1;
 
 	if (tx->u.tx.mode->mode == MODE_IEEE80211G &&
-	    (tx->sdata->flags & IEEE80211_SDATA_USE_PROTECTION) &&
+	    tx->sdata->bss_conf.use_cts_prot &&
 	    (tx->flags & IEEE80211_TXRXD_FRAGMENTED) && rsel.nonerp) {
 		tx->u.tx.last_frag_rate = tx->u.tx.rate;
 		if (rsel.probe)
@@ -667,7 +666,7 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 	if (mode->mode == MODE_IEEE80211G &&
 	    (tx->u.tx.rate->flags & IEEE80211_RATE_ERP) &&
 	    (tx->flags & IEEE80211_TXRXD_TXUNICAST) &&
-	    (tx->sdata->flags & IEEE80211_SDATA_USE_PROTECTION) &&
+	    tx->sdata->bss_conf.use_cts_prot &&
 	    !(control->flags & IEEE80211_TXCTL_USE_RTS_CTS))
 		control->flags |= IEEE80211_TXCTL_USE_CTS_PROTECT;
 
@@ -676,7 +675,7 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 	 * available on the network at the current point in time. */
 	if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
 	    (tx->u.tx.rate->flags & IEEE80211_RATE_PREAMBLE2) &&
-	    (tx->sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE) &&
+	    tx->sdata->bss_conf.use_short_preamble &&
 	    (!tx->sta || (tx->sta->flags & WLAN_STA_SHORT_PREAMBLE))) {
 		tx->u.tx.control->tx_rate = tx->u.tx.rate->val2;
 	}
@@ -1754,7 +1753,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		}
 
 		control->tx_rate =
-			((sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE) &&
+			(sdata->bss_conf.use_short_preamble &&
 			(rsel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ?
 			rsel.rate->val2 : rsel.rate->val;
 		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index ba81cf542818..5e631ce98d7e 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -312,8 +312,8 @@ __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
 	int erp;
 
 	erp = ieee80211_is_erp_rate(hw->conf.phymode, rate);
-	dur = ieee80211_frame_duration(local, frame_len, rate,
-		       erp, sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE);
+	dur = ieee80211_frame_duration(local, frame_len, rate, erp,
+				       sdata->bss_conf.use_short_preamble);
 
 	return cpu_to_le16(dur);
 }
@@ -326,11 +326,11 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
 	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
-	int short_preamble;
+	bool short_preamble;
 	int erp;
 	u16 dur;
 
-	short_preamble = sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE;
+	short_preamble = sdata->bss_conf.use_short_preamble;
 
 	rate = frame_txctl->rts_rate;
 	erp = !!(rate->flags & IEEE80211_RATE_ERP);
@@ -357,11 +357,11 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
 	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
-	int short_preamble;
+	bool short_preamble;
 	int erp;
 	u16 dur;
 
-	short_preamble = sdata->flags & IEEE80211_SDATA_SHORT_PREAMBLE;
+	short_preamble = sdata->bss_conf.use_short_preamble;
 
 	rate = frame_txctl->rts_rate;
 	erp = !!(rate->flags & IEEE80211_RATE_ERP);
-- 
cgit v1.2.3


From af4b7450788426a113057ce2d85c25b4f4e440d1 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Sun, 13 Jan 2008 21:08:24 +0100
Subject: ssb: Add boardflags_hi field to the sprom data structure

Add boardflags-high.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/pci.c       | 3 +++
 include/linux/ssb/ssb.h | 1 +
 2 files changed, 4 insertions(+)

(limited to 'include')

diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index ed2a38752274..b434df75047f 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -377,6 +377,8 @@ static void sprom_extract_r123(struct ssb_sprom *out, const u16 *in)
 	     SSB_SPROM1_ITSSI_A_SHIFT);
 	SPEX(itssi_bg, SSB_SPROM1_ITSSI, SSB_SPROM1_ITSSI_BG, 0);
 	SPEX(boardflags_lo, SSB_SPROM1_BFLLO, 0xFFFF, 0);
+	if (out->revision >= 2)
+		SPEX(boardflags_hi, SSB_SPROM2_BFLHI, 0xFFFF, 0);
 
 	/* Extract the antenna gain values. */
 	gain = r123_extract_antgain(out->revision, in,
@@ -418,6 +420,7 @@ static void sprom_extract_r4(struct ssb_sprom *out, const u16 *in)
 	     SSB_SPROM4_ETHPHY_ET1A_SHIFT);
 	SPEX(country_code, SSB_SPROM4_CCODE, 0xFFFF, 0);
 	SPEX(boardflags_lo, SSB_SPROM4_BFLLO, 0xFFFF, 0);
+	SPEX(boardflags_hi, SSB_SPROM4_BFLHI, 0xFFFF, 0);
 	SPEX(ant_available_a, SSB_SPROM4_ANTAVAIL, SSB_SPROM4_ANTAVAIL_A,
 	     SSB_SPROM4_ANTAVAIL_A_SHIFT);
 	SPEX(ant_available_bg, SSB_SPROM4_ANTAVAIL, SSB_SPROM4_ANTAVAIL_BG,
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 1ab4688c678b..e18f5c23b930 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -43,6 +43,7 @@ struct ssb_sprom {
 	u8 itssi_a;		/* Idle TSSI Target for A-PHY */
 	u8 itssi_bg;		/* Idle TSSI Target for B/G-PHY */
 	u16 boardflags_lo;	/* Boardflags (low 16 bits) */
+	u16 boardflags_hi;	/* Boardflags (high 16 bits) */
 
 	/* Antenna gain values for up to 4 antennas
 	 * on each band. Values in dBm/4 (Q5.2). Negative gain means the
-- 
cgit v1.2.3


From da0e28cb68a7e22b47c6ae1a5b12cb538c13c69f Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 21 Jan 2008 17:31:55 -0800
Subject: [NETNS]: Add netns parameter to fib_lookup.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     | 9 +++++----
 net/ipv4/fib_frontend.c  | 4 ++--
 net/ipv4/fib_rules.c     | 4 ++--
 net/ipv4/fib_semantics.c | 2 +-
 net/ipv4/route.c         | 6 +++---
 5 files changed, 13 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index f580542f08ed..a85912434189 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -178,15 +178,16 @@ static inline struct fib_table *fib_new_table(struct net *net, u32 id)
 	return fib_get_table(net, id);
 }
 
-static inline int fib_lookup(const struct flowi *flp, struct fib_result *res)
+static inline int fib_lookup(struct net *net, const struct flowi *flp,
+			     struct fib_result *res)
 {
 	struct fib_table *table;
 
-	table = fib_get_table(&init_net, RT_TABLE_LOCAL);
+	table = fib_get_table(net, RT_TABLE_LOCAL);
 	if (!table->tb_lookup(table, flp, res))
 		return 0;
 
-	table = fib_get_table(&init_net, RT_TABLE_MAIN);
+	table = fib_get_table(net, RT_TABLE_MAIN);
 	if (!table->tb_lookup(table, flp, res))
 		return 0;
 	return -ENETUNREACH;
@@ -208,7 +209,7 @@ extern void __net_exit fib4_rules_exit(struct net *net);
 extern u32 fib_rules_tclass(struct fib_result *res);
 #endif
 
-extern int fib_lookup(struct flowi *flp, struct fib_result *res);
+extern int fib_lookup(struct net *n, struct flowi *flp, struct fib_result *res);
 
 extern struct fib_table *fib_new_table(struct net *net, u32 id);
 extern struct fib_table *fib_get_table(struct net *net, u32 id);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 8987046d97f3..e056154076b3 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -241,7 +241,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 	if (in_dev == NULL)
 		goto e_inval;
 
-	if (fib_lookup(&fl, &res))
+	if (fib_lookup(&init_net, &fl, &res))
 		goto last_resort;
 	if (res.type != RTN_UNICAST)
 		goto e_inval_res;
@@ -265,7 +265,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 	fl.oif = dev->ifindex;
 
 	ret = 0;
-	if (fib_lookup(&fl, &res) == 0) {
+	if (fib_lookup(&init_net, &fl, &res) == 0) {
 		if (res.type == RTN_UNICAST) {
 			*spec_dst = FIB_RES_PREFSRC(res);
 			ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index d2001f1c28a2..1effb4ab688c 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -54,14 +54,14 @@ u32 fib_rules_tclass(struct fib_result *res)
 }
 #endif
 
-int fib_lookup(struct flowi *flp, struct fib_result *res)
+int fib_lookup(struct net *net, struct flowi *flp, struct fib_result *res)
 {
 	struct fib_lookup_arg arg = {
 		.result = res,
 	};
 	int err;
 
-	err = fib_rules_lookup(init_net.ipv4.rules_ops, flp, 0, &arg);
+	err = fib_rules_lookup(net->ipv4.rules_ops, flp, 0, &arg);
 	res->r = arg.rule;
 
 	return err;
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 0e08df4d6f9a..ecd91c60975f 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -559,7 +559,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi,
 			/* It is not necessary, but requires a bit of thinking */
 			if (fl.fl4_scope < RT_SCOPE_LINK)
 				fl.fl4_scope = RT_SCOPE_LINK;
-			if ((err = fib_lookup(&fl, &res)) != 0)
+			if ((err = fib_lookup(&init_net, &fl, &res)) != 0)
 				return err;
 		}
 		err = -EINVAL;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index f80c761ea0bb..a7651c64bb48 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1559,7 +1559,7 @@ void ip_rt_get_source(u8 *addr, struct rtable *rt)
 
 	if (rt->fl.iif == 0)
 		src = rt->rt_src;
-	else if (fib_lookup(&rt->fl, &res) == 0) {
+	else if (fib_lookup(&init_net, &rt->fl, &res) == 0) {
 		src = FIB_RES_PREFSRC(res);
 		fib_res_put(&res);
 	} else
@@ -1911,7 +1911,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	/*
 	 *	Now we are ready to route packet.
 	 */
-	if ((err = fib_lookup(&fl, &res)) != 0) {
+	if ((err = fib_lookup(&init_net, &fl, &res)) != 0) {
 		if (!IN_DEV_FORWARD(in_dev))
 			goto e_hostunreach;
 		goto no_route;
@@ -2363,7 +2363,7 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
 		goto make_route;
 	}
 
-	if (fib_lookup(&fl, &res)) {
+	if (fib_lookup(&init_net, &fl, &res)) {
 		res.fi = NULL;
 		if (oldflp->oif) {
 			/* Apparently, routing tables are wrong. Assume,
-- 
cgit v1.2.3


From 7fee0ca23711ce1a6b13d3ab78915809a72a59ec Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 21 Jan 2008 17:32:38 -0800
Subject: [NETNS]: Add netns parameter to inetdev_by_index.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h | 2 +-
 net/ipv4/devinet.c         | 6 +++---
 net/ipv4/fib_semantics.c   | 2 +-
 net/ipv4/igmp.c            | 4 ++--
 net/ipv4/ip_gre.c          | 3 ++-
 5 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 45f373107530..e74a2ee8ee2a 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -133,7 +133,7 @@ extern struct net_device 	*ip_dev_find(__be32 addr);
 extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
 extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
-extern struct in_device	*inetdev_by_index(int);
+extern struct in_device	*inetdev_by_index(struct net *, int);
 extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
 extern __be32		inet_confirm_addr(struct in_device *in_dev, __be32 dst, __be32 local, int scope);
 extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, __be32 mask);
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index e381edb19b23..21f71bf912d5 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -409,12 +409,12 @@ static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa)
 	return inet_insert_ifa(ifa);
 }
 
-struct in_device *inetdev_by_index(int ifindex)
+struct in_device *inetdev_by_index(struct net *net, int ifindex)
 {
 	struct net_device *dev;
 	struct in_device *in_dev = NULL;
 	read_lock(&dev_base_lock);
-	dev = __dev_get_by_index(&init_net, ifindex);
+	dev = __dev_get_by_index(net, ifindex);
 	if (dev)
 		in_dev = in_dev_get(dev);
 	read_unlock(&dev_base_lock);
@@ -454,7 +454,7 @@ static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg
 		goto errout;
 
 	ifm = nlmsg_data(nlh);
-	in_dev = inetdev_by_index(ifm->ifa_index);
+	in_dev = inetdev_by_index(net, ifm->ifa_index);
 	if (in_dev == NULL) {
 		err = -ENODEV;
 		goto errout;
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index ecd91c60975f..8b47e112ae5c 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -583,7 +583,7 @@ out:
 		if (nh->nh_flags&(RTNH_F_PERVASIVE|RTNH_F_ONLINK))
 			return -EINVAL;
 
-		in_dev = inetdev_by_index(nh->nh_oif);
+		in_dev = inetdev_by_index(&init_net, nh->nh_oif);
 		if (in_dev == NULL)
 			return -ENODEV;
 		if (!(in_dev->dev->flags&IFF_UP)) {
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 016cfdb184f3..928bc3284554 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1389,7 +1389,7 @@ static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr)
 	struct in_device *idev = NULL;
 
 	if (imr->imr_ifindex) {
-		idev = inetdev_by_index(imr->imr_ifindex);
+		idev = inetdev_by_index(&init_net, imr->imr_ifindex);
 		if (idev)
 			__in_dev_put(idev);
 		return idev;
@@ -2222,7 +2222,7 @@ void ip_mc_drop_socket(struct sock *sk)
 		struct in_device *in_dev;
 		inet->mc_list = iml->next;
 
-		in_dev = inetdev_by_index(iml->multi.imr_ifindex);
+		in_dev = inetdev_by_index(&init_net, iml->multi.imr_ifindex);
 		(void) ip_mc_leave_src(sk, iml, in_dev);
 		if (in_dev != NULL) {
 			ip_mc_dec_group(in_dev, iml->multi.imr_multiaddr.s_addr);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 8b81deb8ff1f..a74983d8c89c 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -1193,7 +1193,8 @@ static int ipgre_close(struct net_device *dev)
 {
 	struct ip_tunnel *t = netdev_priv(dev);
 	if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) {
-		struct in_device *in_dev = inetdev_by_index(t->mlink);
+		struct in_device *in_dev;
+		in_dev = inetdev_by_index(dev->nd_net, t->mlink);
 		if (in_dev) {
 			ip_mc_dec_group(in_dev, t->parms.iph.daddr);
 			in_dev_put(in_dev);
-- 
cgit v1.2.3


From 2334ecbdb27bd1745c0fc6d05cce09ed9585e4c1 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 22 Jan 2008 17:18:38 +0900
Subject: [IPV6]: Sparse: Declare non-static
 ipv6_{route,icmp,frag}_sysctl_init() in header.

Fix the following sparse warnings:
| net/ipv6/route.c:2491:18: warning: symbol 'ipv6_route_sysctl_init' was not declared. Should it be static?
| net/ipv6/icmp.c:922:18: warning: symbol 'ipv6_icmp_sysctl_init' was not declared. Should it be static?
| net/ipv6/reassembly.c:628:6: warning: symbol 'ipv6_frag_sysctl_init' was not declared. Should it be static?

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/ipv6.h         | 4 ++++
 net/ipv6/af_inet6.c        | 2 --
 net/ipv6/sysctl_net_ipv6.c | 3 ---
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index c8e8cb241090..3712caeae74e 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -586,6 +586,10 @@ extern int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 			 int __user *optlen);
 
 #ifdef CONFIG_PROC_FS
+extern struct ctl_table *ipv6_icmp_sysctl_init(struct net *net);
+extern void ipv6_frag_sysctl_init(struct net *net);
+extern struct ctl_table *ipv6_route_sysctl_init(struct net *net);
+
 extern int  ac6_proc_init(void);
 extern void ac6_proc_exit(void);
 extern int  raw6_proc_init(void);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3150c4be3c0c..6738a7b0e67f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -72,8 +72,6 @@ MODULE_LICENSE("GPL");
 static struct list_head inetsw6[SOCK_MAX];
 static DEFINE_SPINLOCK(inetsw6_lock);
 
-void ipv6_frag_sysctl_init(struct net *net);
-
 static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk)
 {
 	const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo);
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 5e0af4d46324..7197eb74a755 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -14,9 +14,6 @@
 #include <net/addrconf.h>
 #include <net/inet_frag.h>
 
-extern struct ctl_table *ipv6_route_sysctl_init(struct net *net);
-extern struct ctl_table *ipv6_icmp_sysctl_init(struct net *net);
-
 static ctl_table ipv6_table_template[] = {
 	{
 		.ctl_name	= NET_IPV6_ROUTE,
-- 
cgit v1.2.3


From 8d8354d2fb9277f165715a6e1cb92bcc89259975 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 05:58:31 -0800
Subject: [NETNS][FRAGS]: Move ctl tables around.

This is a preparation for sysctl netns-ization.
Move the ctl tables to the files, where the tuning
variables reside. Plus make the helpers to register
the tables.

This will simplify the later patches and will keep
similar things closer to each other.

ipv4, ipv6 and conntrack_reasm are patched differently,
but the result is all the tables are in appropriate files.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h                               |  5 --
 include/net/ipv6.h                             |  1 -
 include/net/netfilter/ipv6/nf_conntrack_ipv6.h |  4 +-
 net/ipv4/ip_fragment.c                         | 74 +++++++++++++++++++++++++-
 net/ipv4/sysctl_net_ipv4.c                     | 42 ---------------
 net/ipv6/af_inet6.c                            |  5 --
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 29 ----------
 net/ipv6/netfilter/nf_conntrack_reasm.c        | 31 ++++++++++-
 net/ipv6/reassembly.c                          | 66 +++++++++++++++++++++--
 net/ipv6/sysctl_net_ipv6.c                     | 40 +-------------
 10 files changed, 169 insertions(+), 128 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 6850a80a6886..9d43ac222d15 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -179,11 +179,6 @@ extern int sysctl_ip_nonlocal_bind;
 
 extern struct ctl_path net_ipv4_ctl_path[];
 
-/* From ip_fragment.c */
-struct inet_frags_ctl;
-extern struct inet_frags_ctl ip4_frags_ctl;
-extern int sysctl_ipfrag_max_dist;
-
 /* From inetpeer.c */
 extern int inet_peer_threshold;
 extern int inet_peer_minttl;
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 3712caeae74e..87ca1bf17d71 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -587,7 +587,6 @@ extern int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 
 #ifdef CONFIG_PROC_FS
 extern struct ctl_table *ipv6_icmp_sysctl_init(struct net *net);
-extern void ipv6_frag_sysctl_init(struct net *net);
 extern struct ctl_table *ipv6_route_sysctl_init(struct net *net);
 
 extern int  ac6_proc_init(void);
diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
index f703533fb4db..abc55ad75c2b 100644
--- a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
+++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
@@ -16,6 +16,8 @@ extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb,
 			       int (*okfn)(struct sk_buff *));
 
 struct inet_frags_ctl;
-extern struct inet_frags_ctl nf_frags_ctl;
+
+#include <linux/sysctl.h>
+extern struct ctl_table nf_ct_ipv6_sysctl_table[];
 
 #endif /* _NF_CONNTRACK_IPV6_H*/
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 2143bf30597a..a53463e594b9 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -50,7 +50,7 @@
  * as well. Or notify me, at least. --ANK
  */
 
-int sysctl_ipfrag_max_dist __read_mostly = 64;
+static int sysctl_ipfrag_max_dist __read_mostly = 64;
 
 struct ipfrag_skb_cb
 {
@@ -74,7 +74,7 @@ struct ipq {
 	struct inet_peer *peer;
 };
 
-struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
+static struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
 	/*
 	 * Fragment cache limits. We will commit 256K at one time. Should we
 	 * cross that limit we will prune down to 192K. This should cope with
@@ -607,8 +607,78 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 	return -ENOMEM;
 }
 
+#ifdef CONFIG_SYSCTL
+static int zero;
+
+static struct ctl_table ip4_frags_ctl_table[] = {
+	{
+		.ctl_name	= NET_IPV4_IPFRAG_HIGH_THRESH,
+		.procname	= "ipfrag_high_thresh",
+		.data		= &ip4_frags_ctl.high_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_IPFRAG_LOW_THRESH,
+		.procname	= "ipfrag_low_thresh",
+		.data		= &ip4_frags_ctl.low_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_IPFRAG_TIME,
+		.procname	= "ipfrag_time",
+		.data		= &ip4_frags_ctl.timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_jiffies,
+		.strategy	= &sysctl_jiffies
+	},
+	{
+		.ctl_name	= NET_IPV4_IPFRAG_SECRET_INTERVAL,
+		.procname	= "ipfrag_secret_interval",
+		.data		= &ip4_frags_ctl.secret_interval,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_jiffies,
+		.strategy	= &sysctl_jiffies
+	},
+	{
+		.procname	= "ipfrag_max_dist",
+		.data		= &sysctl_ipfrag_max_dist,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.extra1		= &zero
+	},
+	{ }
+};
+
+static int ip4_frags_ctl_register(struct net *net)
+{
+	struct ctl_table_header *hdr;
+
+	hdr = register_net_sysctl_table(net, net_ipv4_ctl_path,
+			ip4_frags_ctl_table);
+	return hdr == NULL ? -ENOMEM : 0;
+}
+#else
+static inline int ip4_frags_ctl_register(struct net *net)
+{
+	return 0;
+}
+#endif
+
+static int ipv4_frags_init_net(struct net *net)
+{
+	return ip4_frags_ctl_register(net);
+}
+
 void __init ipfrag_init(void)
 {
+	ipv4_frags_init_net(&init_net);
 	ip4_frags.ctl = &ip4_frags_ctl;
 	ip4_frags.hashfn = ip4_hashfn;
 	ip4_frags.constructor = ip4_frag_init;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 45536a91266a..82cdf23837e3 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -283,22 +283,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV4_IPFRAG_HIGH_THRESH,
-		.procname	= "ipfrag_high_thresh",
-		.data		= &ip4_frags_ctl.high_thresh,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV4_IPFRAG_LOW_THRESH,
-		.procname	= "ipfrag_low_thresh",
-		.data		= &ip4_frags_ctl.low_thresh,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
 	{
 		.ctl_name	= NET_IPV4_DYNADDR,
 		.procname	= "ip_dynaddr",
@@ -307,15 +291,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV4_IPFRAG_TIME,
-		.procname	= "ipfrag_time",
-		.data		= &ip4_frags_ctl.timeout,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_jiffies,
-		.strategy	= &sysctl_jiffies
-	},
 	{
 		.ctl_name	= NET_IPV4_TCP_KEEPALIVE_TIME,
 		.procname	= "tcp_keepalive_time",
@@ -658,23 +633,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV4_IPFRAG_SECRET_INTERVAL,
-		.procname	= "ipfrag_secret_interval",
-		.data		= &ip4_frags_ctl.secret_interval,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_jiffies,
-		.strategy	= &sysctl_jiffies
-	},
-	{
-		.procname	= "ipfrag_max_dist",
-		.data		= &sysctl_ipfrag_max_dist,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
-		.extra1		= &zero
-	},
 	{
 		.ctl_name	= NET_TCP_NO_METRICS_SAVE,
 		.procname	= "tcp_no_metrics_save",
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 6738a7b0e67f..bddac0e8780f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -721,10 +721,6 @@ static void cleanup_ipv6_mibs(void)
 static int inet6_net_init(struct net *net)
 {
 	net->ipv6.sysctl.bindv6only = 0;
-	net->ipv6.sysctl.frags.high_thresh = 256 * 1024;
-	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
-	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
-	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
 	net->ipv6.sysctl.flush_delay = 0;
 	net->ipv6.sysctl.ip6_rt_max_size = 4096;
 	net->ipv6.sysctl.ip6_rt_gc_min_interval = HZ / 2;
@@ -734,7 +730,6 @@ static int inet6_net_init(struct net *net)
 	net->ipv6.sysctl.ip6_rt_mtu_expires = 10*60*HZ;
 	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
 	net->ipv6.sysctl.icmpv6_time = 1*HZ;
-	ipv6_frag_sysctl_init(net);
 
 	return 0;
 }
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index cf42f5cfc338..2d7b0246475d 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -297,35 +297,6 @@ static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = {
 	},
 };
 
-#ifdef CONFIG_SYSCTL
-static ctl_table nf_ct_ipv6_sysctl_table[] = {
-	{
-		.procname	= "nf_conntrack_frag6_timeout",
-		.data		= &nf_frags_ctl.timeout,
-		.maxlen		= sizeof(unsigned int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_jiffies,
-	},
-	{
-		.ctl_name	= NET_NF_CONNTRACK_FRAG6_LOW_THRESH,
-		.procname	= "nf_conntrack_frag6_low_thresh",
-		.data		= &nf_frags_ctl.low_thresh,
-		.maxlen		= sizeof(unsigned int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
-		.ctl_name	= NET_NF_CONNTRACK_FRAG6_HIGH_THRESH,
-		.procname	= "nf_conntrack_frag6_high_thresh",
-		.data		= &nf_frags_ctl.high_thresh,
-		.maxlen		= sizeof(unsigned int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{ .ctl_name = 0 }
-};
-#endif
-
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 
 #include <linux/netfilter/nfnetlink.h>
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index e170c67c47a5..d631631189b6 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -70,7 +70,7 @@ struct nf_ct_frag6_queue
 	__u16			nhoffset;
 };
 
-struct inet_frags_ctl nf_frags_ctl __read_mostly = {
+static struct inet_frags_ctl nf_frags_ctl __read_mostly = {
 	.high_thresh	 = 256 * 1024,
 	.low_thresh	 = 192 * 1024,
 	.timeout	 = IPV6_FRAG_TIMEOUT,
@@ -79,6 +79,35 @@ struct inet_frags_ctl nf_frags_ctl __read_mostly = {
 
 static struct inet_frags nf_frags;
 
+#ifdef CONFIG_SYSCTL
+struct ctl_table nf_ct_ipv6_sysctl_table[] = {
+	{
+		.procname	= "nf_conntrack_frag6_timeout",
+		.data		= &nf_frags_ctl.timeout,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= NET_NF_CONNTRACK_FRAG6_LOW_THRESH,
+		.procname	= "nf_conntrack_frag6_low_thresh",
+		.data		= &nf_frags_ctl.low_thresh,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= NET_NF_CONNTRACK_FRAG6_HIGH_THRESH,
+		.procname	= "nf_conntrack_frag6_high_thresh",
+		.data		= &nf_frags_ctl.high_thresh,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{ .ctl_name = 0 }
+};
+#endif
+
 static unsigned int ip6qhashfn(__be32 id, struct in6_addr *saddr,
 			       struct in6_addr *daddr)
 {
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 4dfcddc871ce..1815ff0cf628 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -625,12 +625,70 @@ static struct inet6_protocol frag_protocol =
 	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
-void ipv6_frag_sysctl_init(struct net *net)
+#ifdef CONFIG_SYSCTL
+static struct ctl_table ip6_frags_ctl_table[] = {
+	{
+		.ctl_name	= NET_IPV6_IP6FRAG_HIGH_THRESH,
+		.procname	= "ip6frag_high_thresh",
+		.data		= &init_net.ipv6.sysctl.frags.high_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV6_IP6FRAG_LOW_THRESH,
+		.procname	= "ip6frag_low_thresh",
+		.data		= &init_net.ipv6.sysctl.frags.low_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV6_IP6FRAG_TIME,
+		.procname	= "ip6frag_time",
+		.data		= &init_net.ipv6.sysctl.frags.timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_jiffies,
+		.strategy	= &sysctl_jiffies,
+	},
+	{
+		.ctl_name	= NET_IPV6_IP6FRAG_SECRET_INTERVAL,
+		.procname	= "ip6frag_secret_interval",
+		.data		= &init_net.ipv6.sysctl.frags.secret_interval,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_jiffies,
+		.strategy	= &sysctl_jiffies
+	},
+	{ }
+};
+
+static int ip6_frags_sysctl_register(struct net *net)
+{
+	struct ctl_table_header *hdr;
+
+	hdr = register_net_sysctl_table(net, net_ipv6_ctl_path,
+			ip6_frags_ctl_table);
+	return hdr == NULL ? -ENOMEM : 0;
+}
+#else
+static inline int ip6_frags_sysctl_register(struct net *net)
 {
-	if (net != &init_net)
-		return;
+	return 0;
+}
+#endif
 
+static int ipv6_frags_init_net(struct net *net)
+{
 	ip6_frags.ctl = &net->ipv6.sysctl.frags;
+
+	net->ipv6.sysctl.frags.high_thresh = 256 * 1024;
+	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
+	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
+	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
+
+	return ip6_frags_sysctl_register(net);
 }
 
 int __init ipv6_frag_init(void)
@@ -641,6 +699,8 @@ int __init ipv6_frag_init(void)
 	if (ret)
 		goto out;
 
+	ipv6_frags_init_net(&init_net);
+
 	ip6_frags.hashfn = ip6_hashfn;
 	ip6_frags.constructor = ip6_frag_init;
 	ip6_frags.destructor = NULL;
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 7197eb74a755..408691b777c2 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -37,40 +37,6 @@ static ctl_table ipv6_table_template[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV6_IP6FRAG_HIGH_THRESH,
-		.procname	= "ip6frag_high_thresh",
-		.data		= &init_net.ipv6.sysctl.frags.high_thresh,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV6_IP6FRAG_LOW_THRESH,
-		.procname	= "ip6frag_low_thresh",
-		.data		= &init_net.ipv6.sysctl.frags.low_thresh,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV6_IP6FRAG_TIME,
-		.procname	= "ip6frag_time",
-		.data		= &init_net.ipv6.sysctl.frags.timeout,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_jiffies,
-		.strategy	= &sysctl_jiffies,
-	},
-	{
-		.ctl_name	= NET_IPV6_IP6FRAG_SECRET_INTERVAL,
-		.procname	= "ip6frag_secret_interval",
-		.data		= &init_net.ipv6.sysctl.frags.secret_interval,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_jiffies,
-		.strategy	= &sysctl_jiffies
-	},
 	{
 		.ctl_name	= NET_IPV6_MLD_MAX_MSF,
 		.procname	= "mld_max_msf",
@@ -126,16 +92,12 @@ static int ipv6_sysctl_net_init(struct net *net)
 	ipv6_table[1].child = ipv6_icmp_table;
 
 	ipv6_table[2].data = &net->ipv6.sysctl.bindv6only;
-	ipv6_table[3].data = &net->ipv6.sysctl.frags.high_thresh;
-	ipv6_table[4].data = &net->ipv6.sysctl.frags.low_thresh;
-	ipv6_table[5].data = &net->ipv6.sysctl.frags.timeout;
-	ipv6_table[6].data = &net->ipv6.sysctl.frags.secret_interval;
 
 	/* We don't want this value to be per namespace, it should be global
 	   to all namespaces, so make it read-only when we are not in the
 	   init network namespace */
 	if (net != &init_net)
-		ipv6_table[7].mode = 0444;
+		ipv6_table[3].mode = 0444;
 
 	net->ipv6.sysctl.table = register_net_sysctl_table(net, net_ipv6_ctl_path,
 							   ipv6_table);
-- 
cgit v1.2.3


From ac18e7509e7df327e30d6e073a787d922eaf211d Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:02:14 -0800
Subject: [NETNS][FRAGS]: Make the inet_frag_queue lookup work in namespaces.

Since fragment management code is consolidated, we cannot have the
pointer from inet_frag_queue to struct net, since we must know what
king of fragment this is.

So, I introduce the netns_frags structure. This one is currently
empty, but will be eventually filled with per-namespace
attributes. Each inet_frag_queue is tagged with this one.

The conntrack_reasm is not "netns-izated", so it has one static
netns_frags instance to keep working in init namespace.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  8 ++++++--
 include/net/netns/ipv4.h                |  4 ++++
 include/net/netns/ipv6.h                |  1 +
 net/ipv4/inet_fragment.c                | 27 +++++++++++++++------------
 net/ipv4/ip_fragment.c                  |  8 +++++---
 net/ipv6/netfilter/nf_conntrack_reasm.c |  3 ++-
 net/ipv6/reassembly.c                   |  8 +++++---
 7 files changed, 38 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 954def408975..8ab6df64a12a 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -1,8 +1,12 @@
 #ifndef __NET_FRAG_H__
 #define __NET_FRAG_H__
 
+struct netns_frags {
+};
+
 struct inet_frag_queue {
 	struct hlist_node	list;
+	struct netns_frags	*net;
 	struct list_head	lru_list;   /* lru list member */
 	spinlock_t		lock;
 	atomic_t		refcnt;
@@ -55,8 +59,8 @@ void inet_frag_kill(struct inet_frag_queue *q, struct inet_frags *f);
 void inet_frag_destroy(struct inet_frag_queue *q,
 				struct inet_frags *f, int *work);
 int inet_frag_evictor(struct inet_frags *f);
-struct inet_frag_queue *inet_frag_find(struct inet_frags *f, void *key,
-		unsigned int hash);
+struct inet_frag_queue *inet_frag_find(struct netns_frags *nf,
+		struct inet_frags *f, void *key, unsigned int hash);
 
 static inline void inet_frag_put(struct inet_frag_queue *q, struct inet_frags *f)
 {
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 3872aa7e294b..80680e09443c 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -5,6 +5,8 @@
 #ifndef __NETNS_IPV4_H__
 #define __NETNS_IPV4_H__
 
+#include <net/inet_frag.h>
+
 struct ctl_table_header;
 struct ipv4_devconf;
 struct fib_rules_ops;
@@ -22,5 +24,7 @@ struct netns_ipv4 {
 #endif
 	struct hlist_head	*fib_table_hash;
 	struct sock		*fibnl;
+
+	struct netns_frags	frags;
 };
 #endif
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 06b4dc034bbd..057c8e473a75 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -30,5 +30,6 @@ struct netns_ipv6 {
 	struct netns_sysctl_ipv6 sysctl;
 	struct ipv6_devconf	*devconf_all;
 	struct ipv6_devconf	*devconf_dflt;
+	struct netns_frags	frags;
 };
 #endif
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 737910767ff1..158c5f60d023 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -174,8 +174,9 @@ int inet_frag_evictor(struct inet_frags *f)
 }
 EXPORT_SYMBOL(inet_frag_evictor);
 
-static struct inet_frag_queue *inet_frag_intern(struct inet_frag_queue *qp_in,
-		struct inet_frags *f, unsigned int hash, void *arg)
+static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf,
+		struct inet_frag_queue *qp_in, struct inet_frags *f,
+		unsigned int hash, void *arg)
 {
 	struct inet_frag_queue *qp;
 #ifdef CONFIG_SMP
@@ -189,7 +190,7 @@ static struct inet_frag_queue *inet_frag_intern(struct inet_frag_queue *qp_in,
 	 * promoted read lock to write lock.
 	 */
 	hlist_for_each_entry(qp, n, &f->hash[hash], list) {
-		if (f->match(qp, arg)) {
+		if (qp->net == nf && f->match(qp, arg)) {
 			atomic_inc(&qp->refcnt);
 			write_unlock(&f->lock);
 			qp_in->last_in |= COMPLETE;
@@ -210,7 +211,8 @@ static struct inet_frag_queue *inet_frag_intern(struct inet_frag_queue *qp_in,
 	return qp;
 }
 
-static struct inet_frag_queue *inet_frag_alloc(struct inet_frags *f, void *arg)
+static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
+		struct inet_frags *f, void *arg)
 {
 	struct inet_frag_queue *q;
 
@@ -223,31 +225,32 @@ static struct inet_frag_queue *inet_frag_alloc(struct inet_frags *f, void *arg)
 	setup_timer(&q->timer, f->frag_expire, (unsigned long)q);
 	spin_lock_init(&q->lock);
 	atomic_set(&q->refcnt, 1);
+	q->net = nf;
 
 	return q;
 }
 
-static struct inet_frag_queue *inet_frag_create(struct inet_frags *f,
-		void *arg, unsigned int hash)
+static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
+		struct inet_frags *f, void *arg, unsigned int hash)
 {
 	struct inet_frag_queue *q;
 
-	q = inet_frag_alloc(f, arg);
+	q = inet_frag_alloc(nf, f, arg);
 	if (q == NULL)
 		return NULL;
 
-	return inet_frag_intern(q, f, hash, arg);
+	return inet_frag_intern(nf, q, f, hash, arg);
 }
 
-struct inet_frag_queue *inet_frag_find(struct inet_frags *f, void *key,
-		unsigned int hash)
+struct inet_frag_queue *inet_frag_find(struct netns_frags *nf,
+		struct inet_frags *f, void *key, unsigned int hash)
 {
 	struct inet_frag_queue *q;
 	struct hlist_node *n;
 
 	read_lock(&f->lock);
 	hlist_for_each_entry(q, n, &f->hash[hash], list) {
-		if (f->match(q, key)) {
+		if (q->net == nf && f->match(q, key)) {
 			atomic_inc(&q->refcnt);
 			read_unlock(&f->lock);
 			return q;
@@ -255,6 +258,6 @@ struct inet_frag_queue *inet_frag_find(struct inet_frags *f, void *key,
 	}
 	read_unlock(&f->lock);
 
-	return inet_frag_create(f, key, hash);
+	return inet_frag_create(nf, f, key, hash);
 }
 EXPORT_SYMBOL(inet_frag_find);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index a53463e594b9..56211ef46eed 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -236,7 +236,7 @@ out:
 /* Find the correct entry in the "incomplete datagrams" queue for
  * this IP datagram, and create new one, if nothing is found.
  */
-static inline struct ipq *ip_find(struct iphdr *iph, u32 user)
+static inline struct ipq *ip_find(struct net *net, struct iphdr *iph, u32 user)
 {
 	struct inet_frag_queue *q;
 	struct ip4_create_arg arg;
@@ -246,7 +246,7 @@ static inline struct ipq *ip_find(struct iphdr *iph, u32 user)
 	arg.user = user;
 	hash = ipqhashfn(iph->id, iph->saddr, iph->daddr, iph->protocol);
 
-	q = inet_frag_find(&ip4_frags, &arg, hash);
+	q = inet_frag_find(&net->ipv4.frags, &ip4_frags, &arg, hash);
 	if (q == NULL)
 		goto out_nomem;
 
@@ -582,15 +582,17 @@ out_fail:
 int ip_defrag(struct sk_buff *skb, u32 user)
 {
 	struct ipq *qp;
+	struct net *net;
 
 	IP_INC_STATS_BH(IPSTATS_MIB_REASMREQDS);
 
+	net = skb->dev->nd_net;
 	/* Start by cleaning up the memory. */
 	if (atomic_read(&ip4_frags.mem) > ip4_frags_ctl.high_thresh)
 		ip_evictor();
 
 	/* Lookup (or create) queue header */
-	if ((qp = ip_find(ip_hdr(skb), user)) != NULL) {
+	if ((qp = ip_find(net, ip_hdr(skb), user)) != NULL) {
 		int ret;
 
 		spin_lock(&qp->q.lock);
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index d631631189b6..18accd4eab06 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -78,6 +78,7 @@ static struct inet_frags_ctl nf_frags_ctl __read_mostly = {
 };
 
 static struct inet_frags nf_frags;
+static struct netns_frags nf_init_frags;
 
 #ifdef CONFIG_SYSCTL
 struct ctl_table nf_ct_ipv6_sysctl_table[] = {
@@ -212,7 +213,7 @@ fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst)
 	arg.dst = dst;
 	hash = ip6qhashfn(id, src, dst);
 
-	q = inet_frag_find(&nf_frags, &arg, hash);
+	q = inet_frag_find(&nf_init_frags, &nf_frags, &arg, hash);
 	if (q == NULL)
 		goto oom;
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 1815ff0cf628..ab2d53b81b70 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -234,7 +234,7 @@ out:
 }
 
 static __inline__ struct frag_queue *
-fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst,
+fq_find(struct net *net, __be32 id, struct in6_addr *src, struct in6_addr *dst,
 	struct inet6_dev *idev)
 {
 	struct inet_frag_queue *q;
@@ -246,7 +246,7 @@ fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst,
 	arg.dst = dst;
 	hash = ip6qhashfn(id, src, dst);
 
-	q = inet_frag_find(&ip6_frags, &arg, hash);
+	q = inet_frag_find(&net->ipv6.frags, &ip6_frags, &arg, hash);
 	if (q == NULL)
 		goto oom;
 
@@ -568,6 +568,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 	struct frag_hdr *fhdr;
 	struct frag_queue *fq;
 	struct ipv6hdr *hdr = ipv6_hdr(skb);
+	struct net *net;
 
 	IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMREQDS);
 
@@ -598,10 +599,11 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 		return 1;
 	}
 
+	net = skb->dev->nd_net;
 	if (atomic_read(&ip6_frags.mem) > init_net.ipv6.sysctl.frags.high_thresh)
 		ip6_evictor(ip6_dst_idev(skb->dst));
 
-	if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr,
+	if ((fq = fq_find(net, fhdr->identification, &hdr->saddr, &hdr->daddr,
 			  ip6_dst_idev(skb->dst))) != NULL) {
 		int ret;
 
-- 
cgit v1.2.3


From e5a2bb842cd9681d00d4ca963e63e4d3647e66f8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:06:23 -0800
Subject: [NETNS][FRAGS]: Make the nqueues counter per-namespace.

This is simple - just move the variable from struct inet_frags
to struct netns_frags and adjust the usage appropriately.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  4 +++-
 include/net/ip.h                        |  2 +-
 include/net/ipv6.h                      |  2 +-
 net/ipv4/inet_fragment.c                | 11 ++++++++---
 net/ipv4/ip_fragment.c                  |  6 ++++--
 net/ipv4/proc.c                         |  2 +-
 net/ipv6/netfilter/nf_conntrack_reasm.c |  1 +
 net/ipv6/proc.c                         |  2 +-
 net/ipv6/reassembly.c                   |  6 ++++--
 9 files changed, 24 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 8ab6df64a12a..d36f3a6e6d4e 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -2,6 +2,7 @@
 #define __NET_FRAG_H__
 
 struct netns_frags {
+	int			nqueues;
 };
 
 struct inet_frag_queue {
@@ -36,7 +37,6 @@ struct inet_frags {
 	struct hlist_head	hash[INETFRAGS_HASHSZ];
 	rwlock_t		lock;
 	u32			rnd;
-	int			nqueues;
 	int			qsize;
 	atomic_t		mem;
 	struct timer_list	secret_timer;
@@ -55,6 +55,8 @@ struct inet_frags {
 void inet_frags_init(struct inet_frags *);
 void inet_frags_fini(struct inet_frags *);
 
+void inet_frags_init_net(struct netns_frags *nf);
+
 void inet_frag_kill(struct inet_frag_queue *q, struct inet_frags *f);
 void inet_frag_destroy(struct inet_frag_queue *q,
 				struct inet_frags *f, int *work);
diff --git a/include/net/ip.h b/include/net/ip.h
index 9d43ac222d15..fbe28308bf73 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -332,7 +332,7 @@ enum ip_defrag_users
 
 int ip_defrag(struct sk_buff *skb, u32 user);
 int ip_frag_mem(void);
-int ip_frag_nqueues(void);
+int ip_frag_nqueues(struct net *net);
 
 /*
  *	Functions provided by ip_forward.c
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 87ca1bf17d71..da1c089680d8 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -245,7 +245,7 @@ struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
 
 extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb);
 
-int ip6_frag_nqueues(void);
+int ip6_frag_nqueues(struct net *net);
 int ip6_frag_mem(void);
 
 #define IPV6_FRAG_TIMEOUT	(60*HZ)		/* 60 seconds */
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 158c5f60d023..4fec0b911f85 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -63,7 +63,6 @@ void inet_frags_init(struct inet_frags *f)
 	f->rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^
 				   (jiffies ^ (jiffies >> 6)));
 
-	f->nqueues = 0;
 	atomic_set(&f->mem, 0);
 
 	setup_timer(&f->secret_timer, inet_frag_secret_rebuild,
@@ -73,6 +72,12 @@ void inet_frags_init(struct inet_frags *f)
 }
 EXPORT_SYMBOL(inet_frags_init);
 
+void inet_frags_init_net(struct netns_frags *nf)
+{
+	nf->nqueues = 0;
+}
+EXPORT_SYMBOL(inet_frags_init_net);
+
 void inet_frags_fini(struct inet_frags *f)
 {
 	del_timer(&f->secret_timer);
@@ -84,7 +89,7 @@ static inline void fq_unlink(struct inet_frag_queue *fq, struct inet_frags *f)
 	write_lock(&f->lock);
 	hlist_del(&fq->list);
 	list_del(&fq->lru_list);
-	f->nqueues--;
+	fq->net->nqueues--;
 	write_unlock(&f->lock);
 }
 
@@ -206,7 +211,7 @@ static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf,
 	atomic_inc(&qp->refcnt);
 	hlist_add_head(&qp->list, &f->hash[hash]);
 	list_add_tail(&qp->lru_list, &f->lru_list);
-	f->nqueues++;
+	nf->nqueues++;
 	write_unlock(&f->lock);
 	return qp;
 }
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 56211ef46eed..cd8c83025b48 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -95,9 +95,9 @@ static struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
 
 static struct inet_frags ip4_frags;
 
-int ip_frag_nqueues(void)
+int ip_frag_nqueues(struct net *net)
 {
-	return ip4_frags.nqueues;
+	return net->ipv4.frags.nqueues;
 }
 
 int ip_frag_mem(void)
@@ -675,6 +675,8 @@ static inline int ip4_frags_ctl_register(struct net *net)
 
 static int ipv4_frags_init_net(struct net *net)
 {
+	inet_frags_init_net(&net->ipv4.frags);
+
 	return ip4_frags_ctl_register(net);
 }
 
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index cb3787fbeb90..bae32808616e 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -62,7 +62,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
 	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
-			ip_frag_nqueues(), ip_frag_mem());
+			ip_frag_nqueues(&init_net), ip_frag_mem());
 	return 0;
 }
 
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 18accd4eab06..0b9d0097b68d 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -712,6 +712,7 @@ int nf_ct_frag6_init(void)
 	nf_frags.qsize = sizeof(struct nf_ct_frag6_queue);
 	nf_frags.match = ip6_frag_match;
 	nf_frags.frag_expire = nf_ct_frag6_expire;
+	inet_frags_init_net(&nf_init_frags);
 	inet_frags_init(&nf_frags);
 
 	return 0;
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 571d95a21c15..dec34c87cb4a 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -44,7 +44,7 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v)
 	seq_printf(seq, "RAW6: inuse %d\n",
 		       sock_prot_inuse_get(&rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
-		       ip6_frag_nqueues(), ip6_frag_mem());
+		       ip6_frag_nqueues(&init_net), ip6_frag_mem());
 	return 0;
 }
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index ab2d53b81b70..77a874020f3a 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -84,9 +84,9 @@ struct frag_queue
 
 static struct inet_frags ip6_frags;
 
-int ip6_frag_nqueues(void)
+int ip6_frag_nqueues(struct net *net)
 {
-	return ip6_frags.nqueues;
+	return net->ipv6.frags.nqueues;
 }
 
 int ip6_frag_mem(void)
@@ -690,6 +690,8 @@ static int ipv6_frags_init_net(struct net *net)
 	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
 	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
 
+	inet_frags_init_net(&net->ipv6.frags);
+
 	return ip6_frags_sysctl_register(net);
 }
 
-- 
cgit v1.2.3


From 6ddc082223ef0f73717b4133fa7e648842bbfd02 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:07:25 -0800
Subject: [NETNS][FRAGS]: Make the mem counter per-namespace.

This is also simple, but introduces more changes, since
then mem counter is altered in more places.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  4 ++--
 include/net/ip.h                        |  2 +-
 include/net/ipv6.h                      |  2 +-
 net/ipv4/inet_fragment.c                | 21 +++++++++++----------
 net/ipv4/ip_fragment.c                  | 29 +++++++++++++++--------------
 net/ipv4/proc.c                         |  2 +-
 net/ipv6/netfilter/nf_conntrack_reasm.c | 14 +++++++-------
 net/ipv6/proc.c                         |  2 +-
 net/ipv6/reassembly.c                   | 28 +++++++++++++++-------------
 9 files changed, 54 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index d36f3a6e6d4e..6edce7b2ff13 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -3,6 +3,7 @@
 
 struct netns_frags {
 	int			nqueues;
+	atomic_t		mem;
 };
 
 struct inet_frag_queue {
@@ -38,7 +39,6 @@ struct inet_frags {
 	rwlock_t		lock;
 	u32			rnd;
 	int			qsize;
-	atomic_t		mem;
 	struct timer_list	secret_timer;
 	struct inet_frags_ctl	*ctl;
 
@@ -60,7 +60,7 @@ void inet_frags_init_net(struct netns_frags *nf);
 void inet_frag_kill(struct inet_frag_queue *q, struct inet_frags *f);
 void inet_frag_destroy(struct inet_frag_queue *q,
 				struct inet_frags *f, int *work);
-int inet_frag_evictor(struct inet_frags *f);
+int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f);
 struct inet_frag_queue *inet_frag_find(struct netns_frags *nf,
 		struct inet_frags *f, void *key, unsigned int hash);
 
diff --git a/include/net/ip.h b/include/net/ip.h
index fbe28308bf73..9f50d4f1f157 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -331,7 +331,7 @@ enum ip_defrag_users
 };
 
 int ip_defrag(struct sk_buff *skb, u32 user);
-int ip_frag_mem(void);
+int ip_frag_mem(struct net *net);
 int ip_frag_nqueues(struct net *net);
 
 /*
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index da1c089680d8..fa80ea48639d 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -246,7 +246,7 @@ struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
 extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb);
 
 int ip6_frag_nqueues(struct net *net);
-int ip6_frag_mem(void);
+int ip6_frag_mem(struct net *net);
 
 #define IPV6_FRAG_TIMEOUT	(60*HZ)		/* 60 seconds */
 
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 4fec0b911f85..ad79ae0c0264 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -63,8 +63,6 @@ void inet_frags_init(struct inet_frags *f)
 	f->rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^
 				   (jiffies ^ (jiffies >> 6)));
 
-	atomic_set(&f->mem, 0);
-
 	setup_timer(&f->secret_timer, inet_frag_secret_rebuild,
 			(unsigned long)f);
 	f->secret_timer.expires = jiffies + f->ctl->secret_interval;
@@ -75,6 +73,7 @@ EXPORT_SYMBOL(inet_frags_init);
 void inet_frags_init_net(struct netns_frags *nf)
 {
 	nf->nqueues = 0;
+	atomic_set(&nf->mem, 0);
 }
 EXPORT_SYMBOL(inet_frags_init_net);
 
@@ -107,13 +106,13 @@ void inet_frag_kill(struct inet_frag_queue *fq, struct inet_frags *f)
 
 EXPORT_SYMBOL(inet_frag_kill);
 
-static inline void frag_kfree_skb(struct inet_frags *f, struct sk_buff *skb,
-						int *work)
+static inline void frag_kfree_skb(struct netns_frags *nf, struct inet_frags *f,
+		struct sk_buff *skb, int *work)
 {
 	if (work)
 		*work -= skb->truesize;
 
-	atomic_sub(skb->truesize, &f->mem);
+	atomic_sub(skb->truesize, &nf->mem);
 	if (f->skb_free)
 		f->skb_free(skb);
 	kfree_skb(skb);
@@ -123,22 +122,24 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f,
 					int *work)
 {
 	struct sk_buff *fp;
+	struct netns_frags *nf;
 
 	BUG_TRAP(q->last_in & COMPLETE);
 	BUG_TRAP(del_timer(&q->timer) == 0);
 
 	/* Release all fragment data. */
 	fp = q->fragments;
+	nf = q->net;
 	while (fp) {
 		struct sk_buff *xp = fp->next;
 
-		frag_kfree_skb(f, fp, work);
+		frag_kfree_skb(nf, f, fp, work);
 		fp = xp;
 	}
 
 	if (work)
 		*work -= f->qsize;
-	atomic_sub(f->qsize, &f->mem);
+	atomic_sub(f->qsize, &nf->mem);
 
 	if (f->destructor)
 		f->destructor(q);
@@ -147,12 +148,12 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f,
 }
 EXPORT_SYMBOL(inet_frag_destroy);
 
-int inet_frag_evictor(struct inet_frags *f)
+int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f)
 {
 	struct inet_frag_queue *q;
 	int work, evicted = 0;
 
-	work = atomic_read(&f->mem) - f->ctl->low_thresh;
+	work = atomic_read(&nf->mem) - f->ctl->low_thresh;
 	while (work > 0) {
 		read_lock(&f->lock);
 		if (list_empty(&f->lru_list)) {
@@ -226,7 +227,7 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
 		return NULL;
 
 	f->constructor(q, arg);
-	atomic_add(f->qsize, &f->mem);
+	atomic_add(f->qsize, &nf->mem);
 	setup_timer(&q->timer, f->frag_expire, (unsigned long)q);
 	spin_lock_init(&q->lock);
 	atomic_set(&q->refcnt, 1);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index cd8c83025b48..4f013343cef7 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -100,9 +100,9 @@ int ip_frag_nqueues(struct net *net)
 	return net->ipv4.frags.nqueues;
 }
 
-int ip_frag_mem(void)
+int ip_frag_mem(struct net *net)
 {
-	return atomic_read(&ip4_frags.mem);
+	return atomic_read(&net->ipv4.frags.mem);
 }
 
 static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
@@ -142,11 +142,12 @@ static int ip4_frag_match(struct inet_frag_queue *q, void *a)
 }
 
 /* Memory Tracking Functions. */
-static __inline__ void frag_kfree_skb(struct sk_buff *skb, int *work)
+static __inline__ void frag_kfree_skb(struct netns_frags *nf,
+		struct sk_buff *skb, int *work)
 {
 	if (work)
 		*work -= skb->truesize;
-	atomic_sub(skb->truesize, &ip4_frags.mem);
+	atomic_sub(skb->truesize, &nf->mem);
 	kfree_skb(skb);
 }
 
@@ -192,11 +193,11 @@ static void ipq_kill(struct ipq *ipq)
 /* Memory limiting on fragments.  Evictor trashes the oldest
  * fragment queue until we are back under the threshold.
  */
-static void ip_evictor(void)
+static void ip_evictor(struct net *net)
 {
 	int evicted;
 
-	evicted = inet_frag_evictor(&ip4_frags);
+	evicted = inet_frag_evictor(&net->ipv4.frags, &ip4_frags);
 	if (evicted)
 		IP_ADD_STATS_BH(IPSTATS_MIB_REASMFAILS, evicted);
 }
@@ -294,7 +295,7 @@ static int ip_frag_reinit(struct ipq *qp)
 	fp = qp->q.fragments;
 	do {
 		struct sk_buff *xp = fp->next;
-		frag_kfree_skb(fp, NULL);
+		frag_kfree_skb(qp->q.net, fp, NULL);
 		fp = xp;
 	} while (fp);
 
@@ -431,7 +432,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 				qp->q.fragments = next;
 
 			qp->q.meat -= free_it->len;
-			frag_kfree_skb(free_it, NULL);
+			frag_kfree_skb(qp->q.net, free_it, NULL);
 		}
 	}
 
@@ -451,7 +452,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 	}
 	qp->q.stamp = skb->tstamp;
 	qp->q.meat += skb->len;
-	atomic_add(skb->truesize, &ip4_frags.mem);
+	atomic_add(skb->truesize, &qp->q.net->mem);
 	if (offset == 0)
 		qp->q.last_in |= FIRST_IN;
 
@@ -534,12 +535,12 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
 		head->len -= clone->len;
 		clone->csum = 0;
 		clone->ip_summed = head->ip_summed;
-		atomic_add(clone->truesize, &ip4_frags.mem);
+		atomic_add(clone->truesize, &qp->q.net->mem);
 	}
 
 	skb_shinfo(head)->frag_list = head->next;
 	skb_push(head, head->data - skb_network_header(head));
-	atomic_sub(head->truesize, &ip4_frags.mem);
+	atomic_sub(head->truesize, &qp->q.net->mem);
 
 	for (fp=head->next; fp; fp = fp->next) {
 		head->data_len += fp->len;
@@ -549,7 +550,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
 		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
-		atomic_sub(fp->truesize, &ip4_frags.mem);
+		atomic_sub(fp->truesize, &qp->q.net->mem);
 	}
 
 	head->next = NULL;
@@ -588,8 +589,8 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 
 	net = skb->dev->nd_net;
 	/* Start by cleaning up the memory. */
-	if (atomic_read(&ip4_frags.mem) > ip4_frags_ctl.high_thresh)
-		ip_evictor();
+	if (atomic_read(&net->ipv4.frags.mem) > ip4_frags_ctl.high_thresh)
+		ip_evictor(net);
 
 	/* Lookup (or create) queue header */
 	if ((qp = ip_find(net, ip_hdr(skb), user)) != NULL) {
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index bae32808616e..d63474c6b400 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -62,7 +62,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
 	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
-			ip_frag_nqueues(&init_net), ip_frag_mem());
+			ip_frag_nqueues(&init_net), ip_frag_mem(&init_net));
 	return 0;
 }
 
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 0b9d0097b68d..cb826bea4b1b 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -155,7 +155,7 @@ static inline void frag_kfree_skb(struct sk_buff *skb, unsigned int *work)
 {
 	if (work)
 		*work -= skb->truesize;
-	atomic_sub(skb->truesize, &nf_frags.mem);
+	atomic_sub(skb->truesize, &nf_init_frags.mem);
 	nf_skb_free(skb);
 	kfree_skb(skb);
 }
@@ -177,7 +177,7 @@ static __inline__ void fq_kill(struct nf_ct_frag6_queue *fq)
 
 static void nf_ct_frag6_evictor(void)
 {
-	inet_frag_evictor(&nf_frags);
+	inet_frag_evictor(&nf_init_frags, &nf_frags);
 }
 
 static void nf_ct_frag6_expire(unsigned long data)
@@ -382,7 +382,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 	skb->dev = NULL;
 	fq->q.stamp = skb->tstamp;
 	fq->q.meat += skb->len;
-	atomic_add(skb->truesize, &nf_frags.mem);
+	atomic_add(skb->truesize, &nf_init_frags.mem);
 
 	/* The first fragment.
 	 * nhoffset is obtained from the first fragment, of course.
@@ -459,7 +459,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 		clone->ip_summed = head->ip_summed;
 
 		NFCT_FRAG6_CB(clone)->orig = NULL;
-		atomic_add(clone->truesize, &nf_frags.mem);
+		atomic_add(clone->truesize, &nf_init_frags.mem);
 	}
 
 	/* We have to remove fragment header from datagram and to relocate
@@ -473,7 +473,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 	skb_shinfo(head)->frag_list = head->next;
 	skb_reset_transport_header(head);
 	skb_push(head, head->data - skb_network_header(head));
-	atomic_sub(head->truesize, &nf_frags.mem);
+	atomic_sub(head->truesize, &nf_init_frags.mem);
 
 	for (fp=head->next; fp; fp = fp->next) {
 		head->data_len += fp->len;
@@ -483,7 +483,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
-		atomic_sub(fp->truesize, &nf_frags.mem);
+		atomic_sub(fp->truesize, &nf_init_frags.mem);
 	}
 
 	head->next = NULL;
@@ -633,7 +633,7 @@ struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb)
 		goto ret_orig;
 	}
 
-	if (atomic_read(&nf_frags.mem) > nf_frags_ctl.high_thresh)
+	if (atomic_read(&nf_init_frags.mem) > nf_frags_ctl.high_thresh)
 		nf_ct_frag6_evictor();
 
 	fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr);
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index dec34c87cb4a..35e502a72495 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -44,7 +44,7 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v)
 	seq_printf(seq, "RAW6: inuse %d\n",
 		       sock_prot_inuse_get(&rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
-		       ip6_frag_nqueues(&init_net), ip6_frag_mem());
+		       ip6_frag_nqueues(&init_net), ip6_frag_mem(&init_net));
 	return 0;
 }
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 77a874020f3a..241b2cc49bf5 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -89,9 +89,9 @@ int ip6_frag_nqueues(struct net *net)
 	return net->ipv6.frags.nqueues;
 }
 
-int ip6_frag_mem(void)
+int ip6_frag_mem(struct net *net)
 {
-	return atomic_read(&ip6_frags.mem);
+	return atomic_read(&net->ipv6.frags.mem);
 }
 
 static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
@@ -149,11 +149,12 @@ int ip6_frag_match(struct inet_frag_queue *q, void *a)
 EXPORT_SYMBOL(ip6_frag_match);
 
 /* Memory Tracking Functions. */
-static inline void frag_kfree_skb(struct sk_buff *skb, int *work)
+static inline void frag_kfree_skb(struct netns_frags *nf,
+		struct sk_buff *skb, int *work)
 {
 	if (work)
 		*work -= skb->truesize;
-	atomic_sub(skb->truesize, &ip6_frags.mem);
+	atomic_sub(skb->truesize, &nf->mem);
 	kfree_skb(skb);
 }
 
@@ -183,11 +184,11 @@ static __inline__ void fq_kill(struct frag_queue *fq)
 	inet_frag_kill(&fq->q, &ip6_frags);
 }
 
-static void ip6_evictor(struct inet6_dev *idev)
+static void ip6_evictor(struct net *net, struct inet6_dev *idev)
 {
 	int evicted;
 
-	evicted = inet_frag_evictor(&ip6_frags);
+	evicted = inet_frag_evictor(&net->ipv6.frags, &ip6_frags);
 	if (evicted)
 		IP6_ADD_STATS_BH(idev, IPSTATS_MIB_REASMFAILS, evicted);
 }
@@ -389,7 +390,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 				fq->q.fragments = next;
 
 			fq->q.meat -= free_it->len;
-			frag_kfree_skb(free_it, NULL);
+			frag_kfree_skb(fq->q.net, free_it, NULL);
 		}
 	}
 
@@ -409,7 +410,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 	}
 	fq->q.stamp = skb->tstamp;
 	fq->q.meat += skb->len;
-	atomic_add(skb->truesize, &ip6_frags.mem);
+	atomic_add(skb->truesize, &fq->q.net->mem);
 
 	/* The first fragment.
 	 * nhoffset is obtained from the first fragment, of course.
@@ -503,7 +504,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 		head->len -= clone->len;
 		clone->csum = 0;
 		clone->ip_summed = head->ip_summed;
-		atomic_add(clone->truesize, &ip6_frags.mem);
+		atomic_add(clone->truesize, &fq->q.net->mem);
 	}
 
 	/* We have to remove fragment header from datagram and to relocate
@@ -518,7 +519,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 	skb_shinfo(head)->frag_list = head->next;
 	skb_reset_transport_header(head);
 	skb_push(head, head->data - skb_network_header(head));
-	atomic_sub(head->truesize, &ip6_frags.mem);
+	atomic_sub(head->truesize, &fq->q.net->mem);
 
 	for (fp=head->next; fp; fp = fp->next) {
 		head->data_len += fp->len;
@@ -528,7 +529,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
-		atomic_sub(fp->truesize, &ip6_frags.mem);
+		atomic_sub(fp->truesize, &fq->q.net->mem);
 	}
 
 	head->next = NULL;
@@ -600,8 +601,9 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 	}
 
 	net = skb->dev->nd_net;
-	if (atomic_read(&ip6_frags.mem) > init_net.ipv6.sysctl.frags.high_thresh)
-		ip6_evictor(ip6_dst_idev(skb->dst));
+	if (atomic_read(&net->ipv6.frags.mem) >
+			init_net.ipv6.sysctl.frags.high_thresh)
+		ip6_evictor(net, ip6_dst_idev(skb->dst));
 
 	if ((fq = fq_find(net, fhdr->identification, &hdr->saddr, &hdr->daddr,
 			  ip6_dst_idev(skb->dst))) != NULL) {
-- 
cgit v1.2.3


From e4a2d5c2bccd5bd29de5ae4f14ff4448fac9cfc8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:08:36 -0800
Subject: [NETNS][FRAGS]: Duplicate sysctl tables for new namespaces.

Each namespace has to have own tables to tune their
different parameters, so duplicate the tables and
register them.

All the tables in sub-namespaces are temporarily made
read-only.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  1 +
 include/net/netns/ipv6.h |  1 +
 net/ipv4/ip_fragment.c   | 42 +++++++++++++++++++++++++++++++++++++++---
 net/ipv6/reassembly.c    | 41 ++++++++++++++++++++++++++++++++++++++---
 4 files changed, 79 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 80680e09443c..15a0b052df22 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -16,6 +16,7 @@ struct sock;
 struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*forw_hdr;
+	struct ctl_table_header	*frags_hdr;
 #endif
 	struct ipv4_devconf	*devconf_all;
 	struct ipv4_devconf	*devconf_dflt;
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 057c8e473a75..87ab56ab93fc 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -12,6 +12,7 @@ struct ctl_table_header;
 struct netns_sysctl_ipv6 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header *table;
+	struct ctl_table_header *frags_hdr;
 #endif
 	struct inet_frags_ctl frags;
 	int bindv6only;
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 4f013343cef7..c51e1a11dc6b 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -661,17 +661,53 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 
 static int ip4_frags_ctl_register(struct net *net)
 {
+	struct ctl_table *table;
 	struct ctl_table_header *hdr;
 
-	hdr = register_net_sysctl_table(net, net_ipv4_ctl_path,
-			ip4_frags_ctl_table);
-	return hdr == NULL ? -ENOMEM : 0;
+	table = ip4_frags_ctl_table;
+	if (net != &init_net) {
+		table = kmemdup(table, sizeof(ip4_frags_ctl_table), GFP_KERNEL);
+		if (table == NULL)
+			goto err_alloc;
+
+		table[0].mode &= ~0222;
+		table[1].mode &= ~0222;
+		table[2].mode &= ~0222;
+		table[3].mode &= ~0222;
+		table[4].mode &= ~0222;
+	}
+
+	hdr = register_net_sysctl_table(net, net_ipv4_ctl_path, table);
+	if (hdr == NULL)
+		goto err_reg;
+
+	net->ipv4.frags_hdr = hdr;
+	return 0;
+
+err_reg:
+	if (net != &init_net)
+		kfree(table);
+err_alloc:
+	return -ENOMEM;
+}
+
+static void ip4_frags_ctl_unregister(struct net *net)
+{
+	struct ctl_table *table;
+
+	table = net->ipv4.frags_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->ipv4.frags_hdr);
+	kfree(table);
 }
 #else
 static inline int ip4_frags_ctl_register(struct net *net)
 {
 	return 0;
 }
+
+static inline void ip4_frags_ctl_unregister(struct net *net)
+{
+}
 #endif
 
 static int ipv4_frags_init_net(struct net *net)
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 241b2cc49bf5..0300dcbf1a75 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -670,17 +670,52 @@ static struct ctl_table ip6_frags_ctl_table[] = {
 
 static int ip6_frags_sysctl_register(struct net *net)
 {
+	struct ctl_table *table;
 	struct ctl_table_header *hdr;
 
-	hdr = register_net_sysctl_table(net, net_ipv6_ctl_path,
-			ip6_frags_ctl_table);
-	return hdr == NULL ? -ENOMEM : 0;
+	table = ip6_frags_ctl_table;
+	if (net != &init_net) {
+		table = kmemdup(table, sizeof(ip6_frags_ctl_table), GFP_KERNEL);
+		if (table == NULL)
+			goto err_alloc;
+
+		table[0].mode &= ~0222;
+		table[1].mode &= ~0222;
+		table[2].mode &= ~0222;
+		table[3].mode &= ~0222;
+	}
+
+	hdr = register_net_sysctl_table(net, net_ipv6_ctl_path, table);
+	if (hdr == NULL)
+		goto err_reg;
+
+	net->ipv6.sysctl.frags_hdr = hdr;
+	return 0;
+
+err_reg:
+	if (net != &init_net)
+		kfree(table);
+err_alloc:
+	return -ENOMEM;
+}
+
+static void ip6_frags_sysctl_unregister(struct net *net)
+{
+	struct ctl_table *table;
+
+	table = net->ipv6.sysctl.frags_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->ipv6.sysctl.frags_hdr);
+	kfree(table);
 }
 #else
 static inline int ip6_frags_sysctl_register(struct net *net)
 {
 	return 0;
 }
+
+static inline void ip6_frags_sysctl_unregister(struct net *net)
+{
+}
 #endif
 
 static int ipv6_frags_init_net(struct net *net)
-- 
cgit v1.2.3


From b2fd5321dd160ef309dfb6cfc78ed8de4a830659 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:09:37 -0800
Subject: [NETNS][FRAGS]: Make the net.ipv4.ipfrag_timeout work in namespaces.

Move it to the netns_frags, adjust the usage and
make the appropriate ctl table writable.

Now fragment, that live in different namespaces can
live for different times.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  4 +++-
 net/ipv4/inet_fragment.c                |  2 +-
 net/ipv4/ip_fragment.c                  | 20 ++++++++++----------
 net/ipv6/netfilter/nf_conntrack_reasm.c |  4 ++--
 net/ipv6/reassembly.c                   |  6 +++---
 5 files changed, 19 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 6edce7b2ff13..f56e296e6227 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -4,6 +4,9 @@
 struct netns_frags {
 	int			nqueues;
 	atomic_t		mem;
+
+	/* sysctls */
+	int			timeout;
 };
 
 struct inet_frag_queue {
@@ -29,7 +32,6 @@ struct inet_frag_queue {
 struct inet_frags_ctl {
 	int high_thresh;
 	int low_thresh;
-	int timeout;
 	int secret_interval;
 };
 
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index ad79ae0c0264..9da96792fffb 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -206,7 +206,7 @@ static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf,
 	}
 #endif
 	qp = qp_in;
-	if (!mod_timer(&qp->timer, jiffies + f->ctl->timeout))
+	if (!mod_timer(&qp->timer, jiffies + nf->timeout))
 		atomic_inc(&qp->refcnt);
 
 	atomic_inc(&qp->refcnt);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index c51e1a11dc6b..70d241c8d2a8 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -83,13 +83,6 @@ static struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
 	 */
 	.high_thresh	 = 256 * 1024,
 	.low_thresh	 = 192 * 1024,
-
-	/*
-	 * Important NOTE! Fragment queue must be destroyed before MSL expires.
-	 * RFC791 is wrong proposing to prolongate timer each fragment arrival
-	 * by TTL.
-	 */
-	.timeout	 = IP_FRAG_TIME,
 	.secret_interval = 10 * 60 * HZ,
 };
 
@@ -287,7 +280,7 @@ static int ip_frag_reinit(struct ipq *qp)
 {
 	struct sk_buff *fp;
 
-	if (!mod_timer(&qp->q.timer, jiffies + ip4_frags_ctl.timeout)) {
+	if (!mod_timer(&qp->q.timer, jiffies + qp->q.net->timeout)) {
 		atomic_inc(&qp->q.refcnt);
 		return -ETIMEDOUT;
 	}
@@ -633,7 +626,7 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV4_IPFRAG_TIME,
 		.procname	= "ipfrag_time",
-		.data		= &ip4_frags_ctl.timeout,
+		.data		= &init_net.ipv4.frags.timeout,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -672,7 +665,7 @@ static int ip4_frags_ctl_register(struct net *net)
 
 		table[0].mode &= ~0222;
 		table[1].mode &= ~0222;
-		table[2].mode &= ~0222;
+		table[2].data = &net->ipv4.frags.timeout;
 		table[3].mode &= ~0222;
 		table[4].mode &= ~0222;
 	}
@@ -712,6 +705,13 @@ static inline void ip4_frags_ctl_unregister(struct net *net)
 
 static int ipv4_frags_init_net(struct net *net)
 {
+	/*
+	 * Important NOTE! Fragment queue must be destroyed before MSL expires.
+	 * RFC791 is wrong proposing to prolongate timer each fragment arrival
+	 * by TTL.
+	 */
+	net->ipv4.frags.timeout = IP_FRAG_TIME;
+
 	inet_frags_init_net(&net->ipv4.frags);
 
 	return ip4_frags_ctl_register(net);
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index cb826bea4b1b..92a311ff79c7 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -73,7 +73,6 @@ struct nf_ct_frag6_queue
 static struct inet_frags_ctl nf_frags_ctl __read_mostly = {
 	.high_thresh	 = 256 * 1024,
 	.low_thresh	 = 192 * 1024,
-	.timeout	 = IPV6_FRAG_TIMEOUT,
 	.secret_interval = 10 * 60 * HZ,
 };
 
@@ -84,7 +83,7 @@ static struct netns_frags nf_init_frags;
 struct ctl_table nf_ct_ipv6_sysctl_table[] = {
 	{
 		.procname	= "nf_conntrack_frag6_timeout",
-		.data		= &nf_frags_ctl.timeout,
+		.data		= &nf_init_frags.timeout,
 		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -712,6 +711,7 @@ int nf_ct_frag6_init(void)
 	nf_frags.qsize = sizeof(struct nf_ct_frag6_queue);
 	nf_frags.match = ip6_frag_match;
 	nf_frags.frag_expire = nf_ct_frag6_expire;
+	nf_init_frags.timeout = IPV6_FRAG_TIMEOUT;
 	inet_frags_init_net(&nf_init_frags);
 	inet_frags_init(&nf_frags);
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 0300dcbf1a75..91761365b181 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -650,7 +650,7 @@ static struct ctl_table ip6_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_TIME,
 		.procname	= "ip6frag_time",
-		.data		= &init_net.ipv6.sysctl.frags.timeout,
+		.data		= &init_net.ipv6.frags.timeout,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -681,7 +681,7 @@ static int ip6_frags_sysctl_register(struct net *net)
 
 		table[0].mode &= ~0222;
 		table[1].mode &= ~0222;
-		table[2].mode &= ~0222;
+		table[2].data = &net->ipv6.frags.timeout;
 		table[3].mode &= ~0222;
 	}
 
@@ -724,7 +724,7 @@ static int ipv6_frags_init_net(struct net *net)
 
 	net->ipv6.sysctl.frags.high_thresh = 256 * 1024;
 	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
-	net->ipv6.sysctl.frags.timeout = IPV6_FRAG_TIMEOUT;
+	net->ipv6.frags.timeout = IPV6_FRAG_TIMEOUT;
 	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
 
 	inet_frags_init_net(&net->ipv6.frags);
-- 
cgit v1.2.3


From e31e0bdc7e7fb9a4b09d2f3266c035a18fdcee9d Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:10:13 -0800
Subject: [NETNS][FRAGS]: Make thresholds work in namespaces.

This is the same as with the timeout variable.

Currently, after exceeding the high threshold _all_
the fragments are evicted, but it will be fixed in
later patch.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  4 ++--
 net/ipv4/inet_fragment.c                |  2 +-
 net/ipv4/ip_fragment.c                  | 26 +++++++++++++-------------
 net/ipv6/netfilter/nf_conntrack_reasm.c | 12 ++++++------
 net/ipv6/reassembly.c                   | 15 +++++++--------
 5 files changed, 29 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index f56e296e6227..de4135925490 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -7,6 +7,8 @@ struct netns_frags {
 
 	/* sysctls */
 	int			timeout;
+	int			high_thresh;
+	int			low_thresh;
 };
 
 struct inet_frag_queue {
@@ -30,8 +32,6 @@ struct inet_frag_queue {
 #define INETFRAGS_HASHSZ		64
 
 struct inet_frags_ctl {
-	int high_thresh;
-	int low_thresh;
 	int secret_interval;
 };
 
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 9da96792fffb..5ab399c15282 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -153,7 +153,7 @@ int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f)
 	struct inet_frag_queue *q;
 	int work, evicted = 0;
 
-	work = atomic_read(&nf->mem) - f->ctl->low_thresh;
+	work = atomic_read(&nf->mem) - nf->low_thresh;
 	while (work > 0) {
 		read_lock(&f->lock);
 		if (list_empty(&f->lru_list)) {
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 70d241c8d2a8..80c2c19196cd 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -75,14 +75,6 @@ struct ipq {
 };
 
 static struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
-	/*
-	 * Fragment cache limits. We will commit 256K at one time. Should we
-	 * cross that limit we will prune down to 192K. This should cope with
-	 * even the most extreme cases without allowing an attacker to
-	 * measurably harm machine performance.
-	 */
-	.high_thresh	 = 256 * 1024,
-	.low_thresh	 = 192 * 1024,
 	.secret_interval = 10 * 60 * HZ,
 };
 
@@ -582,7 +574,7 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 
 	net = skb->dev->nd_net;
 	/* Start by cleaning up the memory. */
-	if (atomic_read(&net->ipv4.frags.mem) > ip4_frags_ctl.high_thresh)
+	if (atomic_read(&net->ipv4.frags.mem) > net->ipv4.frags.high_thresh)
 		ip_evictor(net);
 
 	/* Lookup (or create) queue header */
@@ -610,7 +602,7 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV4_IPFRAG_HIGH_THRESH,
 		.procname	= "ipfrag_high_thresh",
-		.data		= &ip4_frags_ctl.high_thresh,
+		.data		= &init_net.ipv4.frags.high_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -618,7 +610,7 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV4_IPFRAG_LOW_THRESH,
 		.procname	= "ipfrag_low_thresh",
-		.data		= &ip4_frags_ctl.low_thresh,
+		.data		= &init_net.ipv4.frags.low_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -663,8 +655,8 @@ static int ip4_frags_ctl_register(struct net *net)
 		if (table == NULL)
 			goto err_alloc;
 
-		table[0].mode &= ~0222;
-		table[1].mode &= ~0222;
+		table[0].data = &net->ipv4.frags.high_thresh;
+		table[1].data = &net->ipv4.frags.low_thresh;
 		table[2].data = &net->ipv4.frags.timeout;
 		table[3].mode &= ~0222;
 		table[4].mode &= ~0222;
@@ -705,6 +697,14 @@ static inline void ip4_frags_ctl_unregister(struct net *net)
 
 static int ipv4_frags_init_net(struct net *net)
 {
+	/*
+	 * Fragment cache limits. We will commit 256K at one time. Should we
+	 * cross that limit we will prune down to 192K. This should cope with
+	 * even the most extreme cases without allowing an attacker to
+	 * measurably harm machine performance.
+	 */
+	net->ipv4.frags.high_thresh = 256 * 1024;
+	net->ipv4.frags.low_thresh = 192 * 1024;
 	/*
 	 * Important NOTE! Fragment queue must be destroyed before MSL expires.
 	 * RFC791 is wrong proposing to prolongate timer each fragment arrival
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 92a311ff79c7..c75ac17e3945 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -71,8 +71,6 @@ struct nf_ct_frag6_queue
 };
 
 static struct inet_frags_ctl nf_frags_ctl __read_mostly = {
-	.high_thresh	 = 256 * 1024,
-	.low_thresh	 = 192 * 1024,
 	.secret_interval = 10 * 60 * HZ,
 };
 
@@ -91,7 +89,7 @@ struct ctl_table nf_ct_ipv6_sysctl_table[] = {
 	{
 		.ctl_name	= NET_NF_CONNTRACK_FRAG6_LOW_THRESH,
 		.procname	= "nf_conntrack_frag6_low_thresh",
-		.data		= &nf_frags_ctl.low_thresh,
+		.data		= &nf_init_frags.low_thresh,
 		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
@@ -99,7 +97,7 @@ struct ctl_table nf_ct_ipv6_sysctl_table[] = {
 	{
 		.ctl_name	= NET_NF_CONNTRACK_FRAG6_HIGH_THRESH,
 		.procname	= "nf_conntrack_frag6_high_thresh",
-		.data		= &nf_frags_ctl.high_thresh,
+		.data		= &nf_init_frags.high_thresh,
 		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
@@ -632,7 +630,7 @@ struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb)
 		goto ret_orig;
 	}
 
-	if (atomic_read(&nf_init_frags.mem) > nf_frags_ctl.high_thresh)
+	if (atomic_read(&nf_init_frags.mem) > nf_init_frags.high_thresh)
 		nf_ct_frag6_evictor();
 
 	fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr);
@@ -712,6 +710,8 @@ int nf_ct_frag6_init(void)
 	nf_frags.match = ip6_frag_match;
 	nf_frags.frag_expire = nf_ct_frag6_expire;
 	nf_init_frags.timeout = IPV6_FRAG_TIMEOUT;
+	nf_init_frags.high_thresh = 256 * 1024;
+	nf_init_frags.low_thresh = 192 * 1024;
 	inet_frags_init_net(&nf_init_frags);
 	inet_frags_init(&nf_frags);
 
@@ -722,6 +722,6 @@ void nf_ct_frag6_cleanup(void)
 {
 	inet_frags_fini(&nf_frags);
 
-	nf_frags_ctl.low_thresh = 0;
+	nf_init_frags.low_thresh = 0;
 	nf_ct_frag6_evictor();
 }
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 91761365b181..85f3fa382230 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -601,8 +601,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 	}
 
 	net = skb->dev->nd_net;
-	if (atomic_read(&net->ipv6.frags.mem) >
-			init_net.ipv6.sysctl.frags.high_thresh)
+	if (atomic_read(&net->ipv6.frags.mem) > net->ipv6.frags.high_thresh)
 		ip6_evictor(net, ip6_dst_idev(skb->dst));
 
 	if ((fq = fq_find(net, fhdr->identification, &hdr->saddr, &hdr->daddr,
@@ -634,7 +633,7 @@ static struct ctl_table ip6_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_HIGH_THRESH,
 		.procname	= "ip6frag_high_thresh",
-		.data		= &init_net.ipv6.sysctl.frags.high_thresh,
+		.data		= &init_net.ipv6.frags.high_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -642,7 +641,7 @@ static struct ctl_table ip6_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_LOW_THRESH,
 		.procname	= "ip6frag_low_thresh",
-		.data		= &init_net.ipv6.sysctl.frags.low_thresh,
+		.data		= &init_net.ipv6.frags.low_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -679,8 +678,8 @@ static int ip6_frags_sysctl_register(struct net *net)
 		if (table == NULL)
 			goto err_alloc;
 
-		table[0].mode &= ~0222;
-		table[1].mode &= ~0222;
+		table[0].data = &net->ipv6.frags.high_thresh;
+		table[1].data = &net->ipv6.frags.low_thresh;
 		table[2].data = &net->ipv6.frags.timeout;
 		table[3].mode &= ~0222;
 	}
@@ -722,8 +721,8 @@ static int ipv6_frags_init_net(struct net *net)
 {
 	ip6_frags.ctl = &net->ipv6.sysctl.frags;
 
-	net->ipv6.sysctl.frags.high_thresh = 256 * 1024;
-	net->ipv6.sysctl.frags.low_thresh = 192 * 1024;
+	net->ipv6.frags.high_thresh = 256 * 1024;
+	net->ipv6.frags.low_thresh = 192 * 1024;
 	net->ipv6.frags.timeout = IPV6_FRAG_TIMEOUT;
 	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
 
-- 
cgit v1.2.3


From 3b4bc4a2bfe80d01ebd4f2b6dcc58986c970ed16 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:11:04 -0800
Subject: [NETNS][FRAGS]: Isolate the secret interval from namespaces.

Since we have one hashtable to lookup the fragment, having
different secret_interval-s for hash rebuild doesn't make
sense, so move this one to inet_frags.

The inet_frags_ctl becomes empty after this, so remove it.
The appropriate ctl table is kept read-only in namespaces.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 | 6 +-----
 include/net/netns/ipv6.h                | 1 -
 net/ipv4/inet_fragment.c                | 4 ++--
 net/ipv4/ip_fragment.c                  | 8 ++------
 net/ipv6/netfilter/nf_conntrack_reasm.c | 6 +-----
 net/ipv6/reassembly.c                   | 6 ++----
 6 files changed, 8 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index de4135925490..1917fbeb362b 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -31,18 +31,14 @@ struct inet_frag_queue {
 
 #define INETFRAGS_HASHSZ		64
 
-struct inet_frags_ctl {
-	int secret_interval;
-};
-
 struct inet_frags {
 	struct list_head	lru_list;
 	struct hlist_head	hash[INETFRAGS_HASHSZ];
 	rwlock_t		lock;
 	u32			rnd;
 	int			qsize;
+	int			secret_interval;
 	struct timer_list	secret_timer;
-	struct inet_frags_ctl	*ctl;
 
 	unsigned int		(*hashfn)(struct inet_frag_queue *);
 	void			(*constructor)(struct inet_frag_queue *q,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 87ab56ab93fc..187c4248df22 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -14,7 +14,6 @@ struct netns_sysctl_ipv6 {
 	struct ctl_table_header *table;
 	struct ctl_table_header *frags_hdr;
 #endif
-	struct inet_frags_ctl frags;
 	int bindv6only;
 	int flush_delay;
 	int ip6_rt_max_size;
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 5ab399c15282..fcf5252166fa 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -47,7 +47,7 @@ static void inet_frag_secret_rebuild(unsigned long dummy)
 	}
 	write_unlock(&f->lock);
 
-	mod_timer(&f->secret_timer, now + f->ctl->secret_interval);
+	mod_timer(&f->secret_timer, now + f->secret_interval);
 }
 
 void inet_frags_init(struct inet_frags *f)
@@ -65,7 +65,7 @@ void inet_frags_init(struct inet_frags *f)
 
 	setup_timer(&f->secret_timer, inet_frag_secret_rebuild,
 			(unsigned long)f);
-	f->secret_timer.expires = jiffies + f->ctl->secret_interval;
+	f->secret_timer.expires = jiffies + f->secret_interval;
 	add_timer(&f->secret_timer);
 }
 EXPORT_SYMBOL(inet_frags_init);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 80c2c19196cd..00646ed451f5 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -74,10 +74,6 @@ struct ipq {
 	struct inet_peer *peer;
 };
 
-static struct inet_frags_ctl ip4_frags_ctl __read_mostly = {
-	.secret_interval = 10 * 60 * HZ,
-};
-
 static struct inet_frags ip4_frags;
 
 int ip_frag_nqueues(struct net *net)
@@ -627,7 +623,7 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV4_IPFRAG_SECRET_INTERVAL,
 		.procname	= "ipfrag_secret_interval",
-		.data		= &ip4_frags_ctl.secret_interval,
+		.data		= &ip4_frags.secret_interval,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -720,7 +716,6 @@ static int ipv4_frags_init_net(struct net *net)
 void __init ipfrag_init(void)
 {
 	ipv4_frags_init_net(&init_net);
-	ip4_frags.ctl = &ip4_frags_ctl;
 	ip4_frags.hashfn = ip4_hashfn;
 	ip4_frags.constructor = ip4_frag_init;
 	ip4_frags.destructor = ip4_frag_free;
@@ -728,6 +723,7 @@ void __init ipfrag_init(void)
 	ip4_frags.qsize = sizeof(struct ipq);
 	ip4_frags.match = ip4_frag_match;
 	ip4_frags.frag_expire = ip_expire;
+	ip4_frags.secret_interval = 10 * 60 * HZ;
 	inet_frags_init(&ip4_frags);
 }
 
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index c75ac17e3945..6eed991a4a3f 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -70,10 +70,6 @@ struct nf_ct_frag6_queue
 	__u16			nhoffset;
 };
 
-static struct inet_frags_ctl nf_frags_ctl __read_mostly = {
-	.secret_interval = 10 * 60 * HZ,
-};
-
 static struct inet_frags nf_frags;
 static struct netns_frags nf_init_frags;
 
@@ -701,7 +697,6 @@ int nf_ct_frag6_kfree_frags(struct sk_buff *skb)
 
 int nf_ct_frag6_init(void)
 {
-	nf_frags.ctl = &nf_frags_ctl;
 	nf_frags.hashfn = nf_hashfn;
 	nf_frags.constructor = ip6_frag_init;
 	nf_frags.destructor = NULL;
@@ -709,6 +704,7 @@ int nf_ct_frag6_init(void)
 	nf_frags.qsize = sizeof(struct nf_ct_frag6_queue);
 	nf_frags.match = ip6_frag_match;
 	nf_frags.frag_expire = nf_ct_frag6_expire;
+	nf_frags.secret_interval = 10 * 60 * HZ;
 	nf_init_frags.timeout = IPV6_FRAG_TIMEOUT;
 	nf_init_frags.high_thresh = 256 * 1024;
 	nf_init_frags.low_thresh = 192 * 1024;
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 85f3fa382230..852070087307 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -658,7 +658,7 @@ static struct ctl_table ip6_frags_ctl_table[] = {
 	{
 		.ctl_name	= NET_IPV6_IP6FRAG_SECRET_INTERVAL,
 		.procname	= "ip6frag_secret_interval",
-		.data		= &init_net.ipv6.sysctl.frags.secret_interval,
+		.data		= &ip6_frags.secret_interval,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -719,12 +719,9 @@ static inline void ip6_frags_sysctl_unregister(struct net *net)
 
 static int ipv6_frags_init_net(struct net *net)
 {
-	ip6_frags.ctl = &net->ipv6.sysctl.frags;
-
 	net->ipv6.frags.high_thresh = 256 * 1024;
 	net->ipv6.frags.low_thresh = 192 * 1024;
 	net->ipv6.frags.timeout = IPV6_FRAG_TIMEOUT;
-	net->ipv6.sysctl.frags.secret_interval = 10 * 60 * HZ;
 
 	inet_frags_init_net(&net->ipv6.frags);
 
@@ -748,6 +745,7 @@ int __init ipv6_frag_init(void)
 	ip6_frags.qsize = sizeof(struct frag_queue);
 	ip6_frags.match = ip6_frag_match;
 	ip6_frags.frag_expire = ip6_frag_expire;
+	ip6_frags.secret_interval = 10 * 60 * HZ;
 	inet_frags_init(&ip6_frags);
 out:
 	return ret;
-- 
cgit v1.2.3


From 3140c25c82106645a6b1fc469dab7006a1d09fd0 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:11:48 -0800
Subject: [NETNS][FRAGS]: Make the LRU list per namespace.

The inet_frags.lru_list is used for evicting only, so we have
to make it per-namespace, to evict only those fragments, who's
namespace exceeded its high threshold, but not the whole hash.
Besides, this helps to avoid long loops  in evictor.

The spinlock is not per-namespace because it protects the
hash table as well, which is global.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 | 2 +-
 net/ipv4/inet_fragment.c                | 8 ++++----
 net/ipv4/ip_fragment.c                  | 2 +-
 net/ipv6/netfilter/nf_conntrack_reasm.c | 2 +-
 net/ipv6/reassembly.c                   | 2 +-
 5 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 1917fbeb362b..3695ff4cfe63 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -4,6 +4,7 @@
 struct netns_frags {
 	int			nqueues;
 	atomic_t		mem;
+	struct list_head	lru_list;
 
 	/* sysctls */
 	int			timeout;
@@ -32,7 +33,6 @@ struct inet_frag_queue {
 #define INETFRAGS_HASHSZ		64
 
 struct inet_frags {
-	struct list_head	lru_list;
 	struct hlist_head	hash[INETFRAGS_HASHSZ];
 	rwlock_t		lock;
 	u32			rnd;
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index fcf5252166fa..f1b95e128772 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -57,7 +57,6 @@ void inet_frags_init(struct inet_frags *f)
 	for (i = 0; i < INETFRAGS_HASHSZ; i++)
 		INIT_HLIST_HEAD(&f->hash[i]);
 
-	INIT_LIST_HEAD(&f->lru_list);
 	rwlock_init(&f->lock);
 
 	f->rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^
@@ -74,6 +73,7 @@ void inet_frags_init_net(struct netns_frags *nf)
 {
 	nf->nqueues = 0;
 	atomic_set(&nf->mem, 0);
+	INIT_LIST_HEAD(&nf->lru_list);
 }
 EXPORT_SYMBOL(inet_frags_init_net);
 
@@ -156,12 +156,12 @@ int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f)
 	work = atomic_read(&nf->mem) - nf->low_thresh;
 	while (work > 0) {
 		read_lock(&f->lock);
-		if (list_empty(&f->lru_list)) {
+		if (list_empty(&nf->lru_list)) {
 			read_unlock(&f->lock);
 			break;
 		}
 
-		q = list_first_entry(&f->lru_list,
+		q = list_first_entry(&nf->lru_list,
 				struct inet_frag_queue, lru_list);
 		atomic_inc(&q->refcnt);
 		read_unlock(&f->lock);
@@ -211,7 +211,7 @@ static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf,
 
 	atomic_inc(&qp->refcnt);
 	hlist_add_head(&qp->list, &f->hash[hash]);
-	list_add_tail(&qp->lru_list, &f->lru_list);
+	list_add_tail(&qp->lru_list, &nf->lru_list);
 	nf->nqueues++;
 	write_unlock(&f->lock);
 	return qp;
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 00646ed451f5..29b4b0972e44 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -441,7 +441,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		return ip_frag_reasm(qp, prev, dev);
 
 	write_lock(&ip4_frags.lock);
-	list_move_tail(&qp->q.lru_list, &ip4_frags.lru_list);
+	list_move_tail(&qp->q.lru_list, &qp->q.net->lru_list);
 	write_unlock(&ip4_frags.lock);
 	return -EINPROGRESS;
 
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 6eed991a4a3f..022da6ce4c0f 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -385,7 +385,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 		fq->q.last_in |= FIRST_IN;
 	}
 	write_lock(&nf_frags.lock);
-	list_move_tail(&fq->q.lru_list, &nf_frags.lru_list);
+	list_move_tail(&fq->q.lru_list, &nf_init_frags.lru_list);
 	write_unlock(&nf_frags.lock);
 	return 0;
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 852070087307..0c4bc46dee0c 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -424,7 +424,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 		return ip6_frag_reasm(fq, prev, dev);
 
 	write_lock(&ip6_frags.lock);
-	list_move_tail(&fq->q.lru_list, &ip6_frags.lru_list);
+	list_move_tail(&fq->q.lru_list, &fq->q.net->lru_list);
 	write_unlock(&ip6_frags.lock);
 	return -1;
 
-- 
cgit v1.2.3


From 81566e8322c3f6c6f9a2277fe0e440fee8d917bd Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 22 Jan 2008 06:12:39 -0800
Subject: [NETNS][FRAGS]: Make the pernet subsystem for fragments.

On namespace start we mainly prepare the ctl variables.

When the namespace is stopped we have to kill all the fragments that
point to this namespace.  The inet_frags_exit_net() handles it.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h  |  1 +
 net/ipv4/inet_fragment.c |  7 +++++++
 net/ipv4/ip_fragment.c   | 13 ++++++++++++-
 net/ipv6/reassembly.c    | 14 +++++++++++++-
 4 files changed, 33 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 3695ff4cfe63..7374251b9787 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -54,6 +54,7 @@ void inet_frags_init(struct inet_frags *);
 void inet_frags_fini(struct inet_frags *);
 
 void inet_frags_init_net(struct netns_frags *nf);
+void inet_frags_exit_net(struct netns_frags *nf, struct inet_frags *f);
 
 void inet_frag_kill(struct inet_frag_queue *q, struct inet_frags *f);
 void inet_frag_destroy(struct inet_frag_queue *q,
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index f1b95e128772..724d69aed031 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -83,6 +83,13 @@ void inet_frags_fini(struct inet_frags *f)
 }
 EXPORT_SYMBOL(inet_frags_fini);
 
+void inet_frags_exit_net(struct netns_frags *nf, struct inet_frags *f)
+{
+	nf->low_thresh = 0;
+	inet_frag_evictor(nf, f);
+}
+EXPORT_SYMBOL(inet_frags_exit_net);
+
 static inline void fq_unlink(struct inet_frag_queue *fq, struct inet_frags *f)
 {
 	write_lock(&f->lock);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 29b4b0972e44..a2e92f9709db 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -713,9 +713,20 @@ static int ipv4_frags_init_net(struct net *net)
 	return ip4_frags_ctl_register(net);
 }
 
+static void ipv4_frags_exit_net(struct net *net)
+{
+	ip4_frags_ctl_unregister(net);
+	inet_frags_exit_net(&net->ipv4.frags, &ip4_frags);
+}
+
+static struct pernet_operations ip4_frags_ops = {
+	.init = ipv4_frags_init_net,
+	.exit = ipv4_frags_exit_net,
+};
+
 void __init ipfrag_init(void)
 {
-	ipv4_frags_init_net(&init_net);
+	register_pernet_subsys(&ip4_frags_ops);
 	ip4_frags.hashfn = ip4_hashfn;
 	ip4_frags.constructor = ip4_frag_init;
 	ip4_frags.destructor = ip4_frag_free;
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 0c4bc46dee0c..f936d045a39d 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -728,6 +728,17 @@ static int ipv6_frags_init_net(struct net *net)
 	return ip6_frags_sysctl_register(net);
 }
 
+static void ipv6_frags_exit_net(struct net *net)
+{
+	ip6_frags_sysctl_unregister(net);
+	inet_frags_exit_net(&net->ipv6.frags, &ip6_frags);
+}
+
+static struct pernet_operations ip6_frags_ops = {
+	.init = ipv6_frags_init_net,
+	.exit = ipv6_frags_exit_net,
+};
+
 int __init ipv6_frag_init(void)
 {
 	int ret;
@@ -736,7 +747,7 @@ int __init ipv6_frag_init(void)
 	if (ret)
 		goto out;
 
-	ipv6_frags_init_net(&init_net);
+	register_pernet_subsys(&ip6_frags_ops);
 
 	ip6_frags.hashfn = ip6_hashfn;
 	ip6_frags.constructor = ip6_frag_init;
@@ -754,5 +765,6 @@ out:
 void ipv6_frag_exit(void)
 {
 	inet_frags_fini(&ip6_frags);
+	unregister_pernet_subsys(&ip6_frags_ops);
 	inet6_del_protocol(&frag_protocol, IPPROTO_FRAGMENT);
 }
-- 
cgit v1.2.3


From 69a73829dbb10e7c8554e66a80cb4fde57347fff Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 22 Jan 2008 06:18:34 -0800
Subject: [DST]: shrinks sizeof(struct rtable) by 64 bytes on x86_64

On x86_64, sizeof(struct rtable) is 0x148, which is rounded up to
0x180 bytes by SLAB allocator.

We can reduce this to exactly 0x140 bytes, without alignment overhead,
and store 12 struct rtable per PAGE instead of 10.

rate_tokens is currently defined as an "unsigned long", while its
content should not exceed 6*HZ. It can safely be converted to an
unsigned int.

Moving tclassid right after rate_tokens to fill the 4 bytes hole
permits to save 8 bytes on 'struct dst_entry', which finally permits
to save 8 bytes on 'struct rtable'

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h | 10 +++++-----
 net/ipv4/icmp.c   | 13 +++++++------
 2 files changed, 12 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index c45dcc31b3bb..e3ac7d0fc4e1 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -56,7 +56,11 @@ struct dst_entry
 	struct dst_entry	*path;
 
 	unsigned long		rate_last;	/* rate limiting for ICMP */
-	unsigned long		rate_tokens;
+	unsigned int		rate_tokens;
+
+#ifdef CONFIG_NET_CLS_ROUTE
+	__u32			tclassid;
+#endif
 
 	struct neighbour	*neighbour;
 	struct hh_cache		*hh;
@@ -65,10 +69,6 @@ struct dst_entry
 	int			(*input)(struct sk_buff*);
 	int			(*output)(struct sk_buff*);
 
-#ifdef CONFIG_NET_CLS_ROUTE
-	__u32			tclassid;
-#endif
-
 	struct  dst_ops	        *ops;
 		
 	unsigned long		lastuse;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index e57f1673bf6b..37cdea0c26b4 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -275,18 +275,19 @@ static inline void icmp_xmit_unlock(void)
 #define XRLIM_BURST_FACTOR 6
 int xrlim_allow(struct dst_entry *dst, int timeout)
 {
-	unsigned long now;
+	unsigned long now, token = dst->rate_tokens;
 	int rc = 0;
 
 	now = jiffies;
-	dst->rate_tokens += now - dst->rate_last;
+	token += now - dst->rate_last;
 	dst->rate_last = now;
-	if (dst->rate_tokens > XRLIM_BURST_FACTOR * timeout)
-		dst->rate_tokens = XRLIM_BURST_FACTOR * timeout;
-	if (dst->rate_tokens >= timeout) {
-		dst->rate_tokens -= timeout;
+	if (token > XRLIM_BURST_FACTOR * timeout)
+		token = XRLIM_BURST_FACTOR * timeout;
+	if (token >= timeout) {
+		token -= timeout;
 		rc = 1;
 	}
+	dst->rate_tokens = token;
 	return rc;
 }
 
-- 
cgit v1.2.3


From a8b47ea3c583645977a916ab3e2d323c7504aa7b Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 21 Jan 2008 12:39:11 +0200
Subject: mac80211: fixing ieee80211_bar types

This patch changes ieee80211_bar control and start_seq_num to
match the proper bitwise attribute expected from ieee 802.11 frame

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 4d5a4c9dcba7..5de6d911cdf7 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -237,8 +237,8 @@ struct ieee80211_bar {
 	__le16 duration;
 	__u8 ra[6];
 	__u8 ta[6];
-	__u16 control;
-	__u16 start_seq_num;
+	__le16 control;
+	__le16 start_seq_num;
 } __attribute__((packed));
 
 /**
-- 
cgit v1.2.3


From e30d3a0ccddbd458b3fb14c9b0229b01d97bad5b Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:03:03 -0800
Subject: [IPV4]: Declarations cleanup in ip_fib.h.

Two small issues fixed:
- fib_select_multipath is exported from fib_semantics.c rather than from
  fib_frontend.c. So, move the declaration below appropriate comment.
- struct rt_entry declaration is not used. Drop it.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index a85912434189..be70b33b744c 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -222,15 +222,13 @@ extern const struct nla_policy rtm_ipv4_policy[];
 extern void		ip_fib_init(void);
 extern int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 			       struct net_device *dev, __be32 *spec_dst, u32 *itag);
-extern void fib_select_multipath(const struct flowi *flp, struct fib_result *res);
-
-struct rtentry;
 
 /* Exported by fib_semantics.c */
 extern int ip_fib_check_default(__be32 gw, struct net_device *dev);
 extern int fib_sync_down(__be32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
 extern __be32  __fib_res_prefsrc(struct fib_result *res);
+extern void fib_select_multipath(const struct flowi *flp, struct fib_result *res);
 
 /* Exported by fib_{hash|trie}.c */
 extern void fib_hash_init(void);
-- 
cgit v1.2.3


From 64c2d5382954ccf6054424653f4c7f4f04c1ff21 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:03:33 -0800
Subject: [IPV4]: Consolidate fib_select_default.

The difference in the implementation of the fib_select_default when
CONFIG_IP_MULTIPLE_TABLES is (not) defined looks
negligible. Consolidate it and place into fib_frontend.c.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 10 +---------
 net/ipv4/fib_frontend.c | 14 ++++++++++++++
 net/ipv4/fib_rules.c    | 10 ----------
 3 files changed, 15 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index be70b33b744c..39f944a096d9 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -193,14 +193,6 @@ static inline int fib_lookup(struct net *net, const struct flowi *flp,
 	return -ENETUNREACH;
 }
 
-static inline void fib_select_default(const struct flowi *flp,
-				      struct fib_result *res)
-{
-	struct fib_table *table = fib_get_table(&init_net, RT_TABLE_MAIN);
-	if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
-		table->tb_select_default(table, flp, res);
-}
-
 #else /* CONFIG_IP_MULTIPLE_TABLES */
 extern int __net_init fib4_rules_init(struct net *net);
 extern void __net_exit fib4_rules_exit(struct net *net);
@@ -213,7 +205,6 @@ extern int fib_lookup(struct net *n, struct flowi *flp, struct fib_result *res);
 
 extern struct fib_table *fib_new_table(struct net *net, u32 id);
 extern struct fib_table *fib_get_table(struct net *net, u32 id);
-extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
 
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
@@ -222,6 +213,7 @@ extern const struct nla_policy rtm_ipv4_policy[];
 extern void		ip_fib_init(void);
 extern int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 			       struct net_device *dev, __be32 *spec_dst, u32 *itag);
+extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
 
 /* Exported by fib_semantics.c */
 extern int ip_fib_check_default(__be32 gw, struct net_device *dev);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 6761639dd0e9..13bf01de0ed8 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -116,6 +116,20 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
 }
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
+void fib_select_default(const struct flowi *flp, struct fib_result *res)
+{
+	struct fib_table *tb;
+	int table = RT_TABLE_MAIN;
+#ifdef CONFIG_IP_MULTIPLE_TABLES
+	if (res->r == NULL || res->r->action != FR_ACT_TO_TBL)
+		return;
+	table = res->r->table;
+#endif
+	tb = fib_get_table(&init_net, table);
+	if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
+		tb->tb_select_default(tb, flp, res);
+}
+
 static void fib_flush(struct net *net)
 {
 	int flushed = 0;
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 1effb4ab688c..19274d01afa4 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -102,16 +102,6 @@ errout:
 }
 
 
-void fib_select_default(const struct flowi *flp, struct fib_result *res)
-{
-	if (res->r && res->r->action == FR_ACT_TO_TBL &&
-	    FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK) {
-		struct fib_table *tb;
-		if ((tb = fib_get_table(&init_net, res->r->table)) != NULL)
-			tb->tb_select_default(tb, flp, res);
-	}
-}
-
 static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 {
 	struct fib4_rule *r = (struct fib4_rule *) rule;
-- 
cgit v1.2.3


From 010278ec4cdf404aefc0bbd5e7406674fec95286 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:04:04 -0800
Subject: [NETNS]: Add netns parameter to fib_select_default.

Currently fib_select_default calls fib_get_table() with the
init_net. Prepare it to provide a correct namespace to lookup default
route.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 3 ++-
 net/ipv4/fib_frontend.c | 5 +++--
 net/ipv4/route.c        | 2 +-
 3 files changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 39f944a096d9..9daa60b544ba 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -213,7 +213,8 @@ extern const struct nla_policy rtm_ipv4_policy[];
 extern void		ip_fib_init(void);
 extern int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 			       struct net_device *dev, __be32 *spec_dst, u32 *itag);
-extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
+extern void fib_select_default(struct net *net, const struct flowi *flp,
+			       struct fib_result *res);
 
 /* Exported by fib_semantics.c */
 extern int ip_fib_check_default(__be32 gw, struct net_device *dev);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 13bf01de0ed8..7e3e7329dacf 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -116,7 +116,8 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
 }
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
-void fib_select_default(const struct flowi *flp, struct fib_result *res)
+void fib_select_default(struct net *net,
+			const struct flowi *flp, struct fib_result *res)
 {
 	struct fib_table *tb;
 	int table = RT_TABLE_MAIN;
@@ -125,7 +126,7 @@ void fib_select_default(const struct flowi *flp, struct fib_result *res)
 		return;
 	table = res->r->table;
 #endif
-	tb = fib_get_table(&init_net, table);
+	tb = fib_get_table(net, table);
 	if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
 		tb->tb_select_default(tb, flp, res);
 }
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 27e0f81060a7..4313255e5a14 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2419,7 +2419,7 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
 	else
 #endif
 	if (!res.prefixlen && res.type == RTN_UNICAST && !fl.oif)
-		fib_select_default(&fl, &res);
+		fib_select_default(&init_net, &fl, &res);
 
 	if (!fl.fl4_src)
 		fl.fl4_src = FIB_RES_PREFSRC(res);
-- 
cgit v1.2.3


From 1ab352768fc73838b062776ca5d1add3876a019f Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:04:30 -0800
Subject: [NETNS]: Add namespace parameter to ip_dev_find.

in_dev_find() need a namespace to pass it to fib_get_table(), so add
an argument.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/core/addr.c | 4 ++--
 drivers/infiniband/core/cma.c  | 2 +-
 include/linux/inetdevice.h     | 2 +-
 net/ipv4/fib_frontend.c        | 4 ++--
 net/ipv4/igmp.c                | 2 +-
 net/ipv4/ip_sockglue.c         | 2 +-
 net/ipv4/ipmr.c                | 2 +-
 net/ipv4/route.c               | 6 +++---
 8 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 0802b79c552d..963177e1c9de 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -110,7 +110,7 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr)
 	__be32 ip = ((struct sockaddr_in *) addr)->sin_addr.s_addr;
 	int ret;
 
-	dev = ip_dev_find(ip);
+	dev = ip_dev_find(&init_net, ip);
 	if (!dev)
 		return -EADDRNOTAVAIL;
 
@@ -261,7 +261,7 @@ static int addr_resolve_local(struct sockaddr_in *src_in,
 	__be32 dst_ip = dst_in->sin_addr.s_addr;
 	int ret;
 
-	dev = ip_dev_find(dst_ip);
+	dev = ip_dev_find(&init_net, dst_ip);
 	if (!dev)
 		return -EADDRNOTAVAIL;
 
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 15b7e9fb615a..1eff1b2c0e08 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -1289,7 +1289,7 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 	atomic_inc(&conn_id->dev_remove);
 	conn_id->state = CMA_CONNECT;
 
-	dev = ip_dev_find(iw_event->local_addr.sin_addr.s_addr);
+	dev = ip_dev_find(&init_net, iw_event->local_addr.sin_addr.s_addr);
 	if (!dev) {
 		ret = -EADDRNOTAVAIL;
 		cma_enable_remove(conn_id);
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index e74a2ee8ee2a..8d9eaaebded7 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -129,7 +129,7 @@ struct in_ifaddr
 extern int register_inetaddr_notifier(struct notifier_block *nb);
 extern int unregister_inetaddr_notifier(struct notifier_block *nb);
 
-extern struct net_device 	*ip_dev_find(__be32 addr);
+extern struct net_device *ip_dev_find(struct net *net, __be32 addr);
 extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
 extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 7e3e7329dacf..d28261826bc2 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -153,7 +153,7 @@ static void fib_flush(struct net *net)
  *	Find the first device with a given source address.
  */
 
-struct net_device * ip_dev_find(__be32 addr)
+struct net_device * ip_dev_find(struct net *net, __be32 addr)
 {
 	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } } };
 	struct fib_result res;
@@ -164,7 +164,7 @@ struct net_device * ip_dev_find(__be32 addr)
 	res.r = NULL;
 #endif
 
-	local_table = fib_get_table(&init_net, RT_TABLE_LOCAL);
+	local_table = fib_get_table(net, RT_TABLE_LOCAL);
 	if (!local_table || local_table->tb_lookup(local_table, &fl, &res))
 		return NULL;
 	if (res.type != RTN_LOCAL)
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 928bc3284554..1f5314ca109c 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1395,7 +1395,7 @@ static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr)
 		return idev;
 	}
 	if (imr->imr_address.s_addr) {
-		dev = ip_dev_find(imr->imr_address.s_addr);
+		dev = ip_dev_find(&init_net, imr->imr_address.s_addr);
 		if (!dev)
 			return NULL;
 		dev_put(dev);
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 82817e554363..754b0a5bbfe9 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -594,7 +594,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 				err = 0;
 				break;
 			}
-			dev = ip_dev_find(mreq.imr_address.s_addr);
+			dev = ip_dev_find(&init_net, mreq.imr_address.s_addr);
 			if (dev) {
 				mreq.imr_ifindex = dev->ifindex;
 				dev_put(dev);
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 4198615ca678..221271758b93 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -423,7 +423,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock)
 			return -ENOBUFS;
 		break;
 	case 0:
-		dev = ip_dev_find(vifc->vifc_lcl_addr.s_addr);
+		dev = ip_dev_find(&init_net, vifc->vifc_lcl_addr.s_addr);
 		if (!dev)
 			return -EADDRNOTAVAIL;
 		dev_put(dev);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 4313255e5a14..674575b622ad 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2282,14 +2282,14 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
 			goto out;
 
 		/* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */
-		dev_out = ip_dev_find(oldflp->fl4_src);
+		dev_out = ip_dev_find(&init_net, oldflp->fl4_src);
 		if (dev_out == NULL)
 			goto out;
 
 		/* I removed check for oif == dev_out->oif here.
 		   It was wrong for two reasons:
-		   1. ip_dev_find(saddr) can return wrong iface, if saddr is
-		      assigned to multiple interfaces.
+		   1. ip_dev_find(net, saddr) can return wrong iface, if saddr
+		      is assigned to multiple interfaces.
 		   2. Moreover, we are allowed to send packets with saddr
 		      of another iface. --ANK
 		 */
-- 
cgit v1.2.3


From 611c183ebcb5af384df3a4ddb391034a1b6ac255 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:06:48 -0800
Subject: [NETNS]: Add namespace parameter to __ip_route_output_key.

This is only required to propagate it down to the
ip_route_output_slow.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h     | 4 ++--
 net/ipv4/icmp.c         | 4 ++--
 net/ipv4/route.c        | 7 ++++---
 net/ipv4/xfrm4_policy.c | 2 +-
 4 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 5847e6fabe8e..3e3b14e060f9 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -110,7 +110,7 @@ extern int		ip_rt_init(void);
 extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
 				       __be32 src, struct net_device *dev);
 extern void		rt_cache_flush(int how);
-extern int		__ip_route_output_key(struct rtable **, const struct flowi *flp);
+extern int		__ip_route_output_key(struct net *, struct rtable **, const struct flowi *flp);
 extern int		ip_route_output_key(struct rtable **, struct flowi *flp);
 extern int		ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
 extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, struct net_device *devin);
@@ -158,7 +158,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 
 	int err;
 	if (!dst || !src) {
-		err = __ip_route_output_key(rp, &fl);
+		err = __ip_route_output_key(&init_net, rp, &fl);
 		if (err)
 			return err;
 		fl.fl4_dst = (*rp)->rt_dst;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 37cdea0c26b4..11760310f917 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -569,7 +569,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 		struct rtable *rt2;
 
 		security_skb_classify_flow(skb_in, &fl);
-		if (__ip_route_output_key(&rt, &fl))
+		if (__ip_route_output_key(&init_net, &rt, &fl))
 			goto out_unlock;
 
 		/* No need to clone since we're just using its address. */
@@ -592,7 +592,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 			goto out_unlock;
 
 		if (inet_addr_type(&init_net, fl.fl4_src) == RTN_LOCAL)
-			err = __ip_route_output_key(&rt2, &fl);
+			err = __ip_route_output_key(&init_net, &rt2, &fl);
 		else {
 			struct flowi fl2 = {};
 			struct dst_entry *odst;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 0576ff79c550..971ab0253a09 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2443,7 +2443,8 @@ make_route:
 out:	return err;
 }
 
-int __ip_route_output_key(struct rtable **rp, const struct flowi *flp)
+int __ip_route_output_key(struct net *net, struct rtable **rp,
+			  const struct flowi *flp)
 {
 	unsigned hash;
 	struct rtable *rth;
@@ -2470,7 +2471,7 @@ int __ip_route_output_key(struct rtable **rp, const struct flowi *flp)
 	}
 	rcu_read_unlock_bh();
 
-	return ip_route_output_slow(&init_net, rp, flp);
+	return ip_route_output_slow(net, rp, flp);
 }
 
 EXPORT_SYMBOL_GPL(__ip_route_output_key);
@@ -2536,7 +2537,7 @@ int ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk,
 {
 	int err;
 
-	if ((err = __ip_route_output_key(rp, flp)) != 0)
+	if ((err = __ip_route_output_key(&init_net, rp, flp)) != 0)
 		return err;
 
 	if (flp->proto) {
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index f04516c880fb..3783e3ee56a4 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -36,7 +36,7 @@ static struct dst_entry *xfrm4_dst_lookup(int tos, xfrm_address_t *saddr,
 	if (saddr)
 		fl.fl4_src = saddr->a4;
 
-	err = __ip_route_output_key(&rt, &fl);
+	err = __ip_route_output_key(&init_net, &rt, &fl);
 	dst = &rt->u.dst;
 	if (err)
 		dst = ERR_PTR(err);
-- 
cgit v1.2.3


From f1b050bf7a88910f9f00c9c8989c1bf5a67dd140 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:07:10 -0800
Subject: [NETNS]: Add namespace parameter to ip_route_output_flow.

Needed to propagate it down to the __ip_route_output_key.

Signed_off_by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/hw/cxgb3/iwch_cm.c | 2 +-
 include/net/route.h                   | 6 +++---
 net/dccp/ipv4.c                       | 2 +-
 net/ipv4/af_inet.c                    | 2 +-
 net/ipv4/inet_connection_sock.c       | 2 +-
 net/ipv4/ip_output.c                  | 2 +-
 net/ipv4/raw.c                        | 2 +-
 net/ipv4/route.c                      | 7 ++++---
 net/ipv4/udp.c                        | 2 +-
 9 files changed, 14 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index f8cb0fe748c3..e9a08fa3dffe 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -332,7 +332,7 @@ static struct rtable *find_route(struct t3cdev *dev, __be32 local_ip,
 			  }
 	};
 
-	if (ip_route_output_flow(&rt, &fl, NULL, 0))
+	if (ip_route_output_flow(&init_net, &rt, &fl, NULL, 0))
 		return NULL;
 	return rt;
 }
diff --git a/include/net/route.h b/include/net/route.h
index 3e3b14e060f9..6b970d7d07e0 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -112,7 +112,7 @@ extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
 extern void		rt_cache_flush(int how);
 extern int		__ip_route_output_key(struct net *, struct rtable **, const struct flowi *flp);
 extern int		ip_route_output_key(struct rtable **, struct flowi *flp);
-extern int		ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
+extern int		ip_route_output_flow(struct net *, struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
 extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, struct net_device *devin);
 extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
@@ -167,7 +167,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 		*rp = NULL;
 	}
 	security_sk_classify_flow(sk, &fl);
-	return ip_route_output_flow(rp, &fl, sk, flags);
+	return ip_route_output_flow(&init_net, rp, &fl, sk, flags);
 }
 
 static inline int ip_route_newports(struct rtable **rp, u8 protocol,
@@ -184,7 +184,7 @@ static inline int ip_route_newports(struct rtable **rp, u8 protocol,
 		ip_rt_put(*rp);
 		*rp = NULL;
 		security_sk_classify_flow(sk, &fl);
-		return ip_route_output_flow(rp, &fl, sk, 0);
+		return ip_route_output_flow(&init_net, rp, &fl, sk, 0);
 	}
 	return 0;
 }
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index f450df2fc863..9e38b0d6195c 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -469,7 +469,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
 			  };
 
 	security_skb_classify_flow(skb, &fl);
-	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
+	if (ip_route_output_flow(&init_net, &rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index bcf8c8a4a3ad..09ca5293d08f 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1113,7 +1113,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 	};
 
 	security_sk_classify_flow(sk, &fl);
-	err = ip_route_output_flow(&rt, &fl, sk, 0);
+	err = ip_route_output_flow(&init_net, &rt, &fl, sk, 0);
 }
 	if (!err)
 		sk_setup_caps(sk, &rt->u.dst);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 1c2a32f6bfc4..7801cceb2d1b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -333,7 +333,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 					 .dport = ireq->rmt_port } } };
 
 	security_req_classify_flow(req, &fl);
-	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
+	if (ip_route_output_flow(&init_net, &rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 8950d18001f1..6d78e1d6b785 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -350,7 +350,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
 			 * itself out.
 			 */
 			security_sk_classify_flow(sk, &fl);
-			if (ip_route_output_flow(&rt, &fl, sk, 0))
+			if (ip_route_output_flow(&init_net, &rt, &fl, sk, 0))
 				goto no_route;
 		}
 		sk_setup_caps(sk, &rt->u.dst);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 91a52184351d..85c08696abbe 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -558,7 +558,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		}
 
 		security_sk_classify_flow(sk, &fl);
-		err = ip_route_output_flow(&rt, &fl, sk, 1);
+		err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
 	}
 	if (err)
 		goto done;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 971ab0253a09..c75fc20b07e2 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2533,11 +2533,12 @@ static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp, struct sock
 	return (rt ? 0 : -ENOMEM);
 }
 
-int ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk, int flags)
+int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp,
+			 struct sock *sk, int flags)
 {
 	int err;
 
-	if ((err = __ip_route_output_key(&init_net, rp, flp)) != 0)
+	if ((err = __ip_route_output_key(net, rp, flp)) != 0)
 		return err;
 
 	if (flp->proto) {
@@ -2560,7 +2561,7 @@ EXPORT_SYMBOL_GPL(ip_route_output_flow);
 
 int ip_route_output_key(struct rtable **rp, struct flowi *flp)
 {
-	return ip_route_output_flow(rp, flp, NULL, 0);
+	return ip_route_output_flow(&init_net, rp, flp, NULL, 0);
 }
 
 static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index ecd9d91b30c6..2fb8d731026b 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -660,7 +660,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 					       { .sport = inet->sport,
 						 .dport = dport } } };
 		security_sk_classify_flow(sk, &fl);
-		err = ip_route_output_flow(&rt, &fl, sk, 1);
+		err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
 		if (err) {
 			if (err == -ENETUNREACH)
 				IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
-- 
cgit v1.2.3


From f206351a50ea86250fabea96b9af8d8f8fc02603 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 22:07:34 -0800
Subject: [NETNS]: Add namespace parameter to ip_route_output_key.

Needed to propagate it down to the ip_route_output_flow.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/core/addr.c   |  4 ++--
 drivers/net/bonding/bond_main.c  |  2 +-
 include/net/route.h              |  2 +-
 net/atm/clip.c                   |  2 +-
 net/bridge/br_netfilter.c        |  2 +-
 net/ipv4/arp.c                   |  6 +++---
 net/ipv4/icmp.c                  |  4 ++--
 net/ipv4/igmp.c                  |  6 +++---
 net/ipv4/ip_gre.c                | 10 +++++-----
 net/ipv4/ip_output.c             |  2 +-
 net/ipv4/ipip.c                  |  8 ++++----
 net/ipv4/ipmr.c                  |  4 ++--
 net/ipv4/ipvs/ip_vs_xmit.c       |  6 +++---
 net/ipv4/netfilter.c             |  6 +++---
 net/ipv4/netfilter/nf_nat_rule.c |  2 +-
 net/ipv4/route.c                 |  6 +++---
 net/ipv4/syncookies.c            |  2 +-
 net/ipv6/ip6_tunnel.c            |  4 ++--
 net/ipv6/sit.c                   |  4 ++--
 net/rxrpc/ar-peer.c              |  2 +-
 net/sctp/protocol.c              |  4 ++--
 21 files changed, 44 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 963177e1c9de..a58ad8a470f9 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -158,7 +158,7 @@ static void addr_send_arp(struct sockaddr_in *dst_in)
 
 	memset(&fl, 0, sizeof fl);
 	fl.nl_u.ip4_u.daddr = dst_ip;
-	if (ip_route_output_key(&rt, &fl))
+	if (ip_route_output_key(&init_net, &rt, &fl))
 		return;
 
 	neigh_event_send(rt->u.dst.neighbour, NULL);
@@ -179,7 +179,7 @@ static int addr_resolve_remote(struct sockaddr_in *src_in,
 	memset(&fl, 0, sizeof fl);
 	fl.nl_u.ip4_u.daddr = dst_ip;
 	fl.nl_u.ip4_u.saddr = src_ip;
-	ret = ip_route_output_key(&rt, &fl);
+	ret = ip_route_output_key(&init_net, &rt, &fl);
 	if (ret)
 		goto out;
 
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 49a198206e3d..2039f7838f2d 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2517,7 +2517,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
 		fl.fl4_dst = targets[i];
 		fl.fl4_tos = RTO_ONLINK;
 
-		rv = ip_route_output_key(&rt, &fl);
+		rv = ip_route_output_key(&init_net, &rt, &fl);
 		if (rv) {
 			if (net_ratelimit()) {
 				printk(KERN_WARNING DRV_NAME
diff --git a/include/net/route.h b/include/net/route.h
index 6b970d7d07e0..d9b876adbf42 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -111,7 +111,7 @@ extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
 				       __be32 src, struct net_device *dev);
 extern void		rt_cache_flush(int how);
 extern int		__ip_route_output_key(struct net *, struct rtable **, const struct flowi *flp);
-extern int		ip_route_output_key(struct rtable **, struct flowi *flp);
+extern int		ip_route_output_key(struct net *, struct rtable **, struct flowi *flp);
 extern int		ip_route_output_flow(struct net *, struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
 extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, struct net_device *devin);
 extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
diff --git a/net/atm/clip.c b/net/atm/clip.c
index 45e08620c8ca..86b885ec1cbd 100644
--- a/net/atm/clip.c
+++ b/net/atm/clip.c
@@ -534,7 +534,7 @@ static int clip_setentry(struct atm_vcc *vcc, __be32 ip)
 		unlink_clip_vcc(clip_vcc);
 		return 0;
 	}
-	error = ip_route_output_key(&rt, &fl);
+	error = ip_route_output_key(&init_net, &rt, &fl);
 	if (error)
 		return error;
 	neigh = __neigh_lookup(&clip_tbl, &ip, rt->u.dst.dev, 1);
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 141f069e77a0..80014bab81b0 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -353,7 +353,7 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb)
 			if (err != -EHOSTUNREACH || !in_dev || IN_DEV_FORWARD(in_dev))
 				goto free_skb;
 
-			if (!ip_route_output_key(&rt, &fl)) {
+			if (!ip_route_output_key(&init_net, &rt, &fl)) {
 				/* - Bridged-and-DNAT'ed traffic doesn't
 				 *   require ip_forwarding. */
 				if (((struct dst_entry *)rt)->dev == dev) {
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index b2c19cb12063..5976c598cc4b 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -424,7 +424,7 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev)
 	int flag = 0;
 	/*unsigned long now; */
 
-	if (ip_route_output_key(&rt, &fl) < 0)
+	if (ip_route_output_key(&init_net, &rt, &fl) < 0)
 		return 1;
 	if (rt->u.dst.dev != dev) {
 		NET_INC_STATS_BH(LINUX_MIB_ARPFILTER);
@@ -1002,7 +1002,7 @@ static int arp_req_set(struct net *net, struct arpreq *r,
 		struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ip,
 							 .tos = RTO_ONLINK } } };
 		struct rtable * rt;
-		if ((err = ip_route_output_key(&rt, &fl)) != 0)
+		if ((err = ip_route_output_key(net, &rt, &fl)) != 0)
 			return err;
 		dev = rt->u.dst.dev;
 		ip_rt_put(rt);
@@ -1109,7 +1109,7 @@ static int arp_req_delete(struct net *net, struct arpreq *r,
 		struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ip,
 							 .tos = RTO_ONLINK } } };
 		struct rtable * rt;
-		if ((err = ip_route_output_key(&rt, &fl)) != 0)
+		if ((err = ip_route_output_key(net, &rt, &fl)) != 0)
 			return err;
 		dev = rt->u.dst.dev;
 		ip_rt_put(rt);
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 11760310f917..a142f19fec45 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -405,7 +405,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 						.tos = RT_TOS(ip_hdr(skb)->tos) } },
 				    .proto = IPPROTO_ICMP };
 		security_skb_classify_flow(skb, &fl);
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			goto out_unlock;
 	}
 	if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type,
@@ -598,7 +598,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 			struct dst_entry *odst;
 
 			fl2.fl4_dst = fl.fl4_src;
-			if (ip_route_output_key(&rt2, &fl2))
+			if (ip_route_output_key(&init_net, &rt2, &fl2))
 				goto out_unlock;
 
 			/* Ugh! */
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 1f5314ca109c..994648be80ab 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -301,7 +301,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size)
 				    .nl_u = { .ip4_u = {
 				    .daddr = IGMPV3_ALL_MCR } },
 				    .proto = IPPROTO_IGMP };
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			kfree_skb(skb);
 			return NULL;
 		}
@@ -645,7 +645,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
 		struct flowi fl = { .oif = dev->ifindex,
 				    .nl_u = { .ip4_u = { .daddr = dst } },
 				    .proto = IPPROTO_IGMP };
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			return -1;
 	}
 	if (rt->rt_src == 0) {
@@ -1401,7 +1401,7 @@ static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr)
 		dev_put(dev);
 	}
 
-	if (!dev && !ip_route_output_key(&rt, &fl)) {
+	if (!dev && !ip_route_output_key(&init_net, &rt, &fl)) {
 		dev = rt->u.dst.dev;
 		ip_rt_put(rt);
 	}
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index a74983d8c89c..63f691719353 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -480,7 +480,7 @@ out:
 	fl.fl4_dst = eiph->saddr;
 	fl.fl4_tos = RT_TOS(eiph->tos);
 	fl.proto = IPPROTO_GRE;
-	if (ip_route_output_key(&rt, &fl)) {
+	if (ip_route_output_key(&init_net, &rt, &fl)) {
 		kfree_skb(skb2);
 		return;
 	}
@@ -493,7 +493,7 @@ out:
 		fl.fl4_dst = eiph->daddr;
 		fl.fl4_src = eiph->saddr;
 		fl.fl4_tos = eiph->tos;
-		if (ip_route_output_key(&rt, &fl) ||
+		if (ip_route_output_key(&init_net, &rt, &fl) ||
 		    rt->u.dst.dev->type != ARPHRD_IPGRE) {
 			ip_rt_put(rt);
 			kfree_skb(skb2);
@@ -748,7 +748,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 						.saddr = tiph->saddr,
 						.tos = RT_TOS(tos) } },
 				    .proto = IPPROTO_GRE };
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			tunnel->stat.tx_carrier_errors++;
 			goto tx_error;
 		}
@@ -921,7 +921,7 @@ static void ipgre_tunnel_bind_dev(struct net_device *dev)
 						.tos = RT_TOS(iph->tos) } },
 				    .proto = IPPROTO_GRE };
 		struct rtable *rt;
-		if (!ip_route_output_key(&rt, &fl)) {
+		if (!ip_route_output_key(&init_net, &rt, &fl)) {
 			tdev = rt->u.dst.dev;
 			ip_rt_put(rt);
 		}
@@ -1177,7 +1177,7 @@ static int ipgre_open(struct net_device *dev)
 						.tos = RT_TOS(t->parms.iph.tos) } },
 				    .proto = IPPROTO_GRE };
 		struct rtable *rt;
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			return -EADDRNOTAVAIL;
 		dev = rt->u.dst.dev;
 		ip_rt_put(rt);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 6d78e1d6b785..1725e0613982 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1379,7 +1379,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 						 .dport = tcp_hdr(skb)->source } },
 				    .proto = sk->sk_protocol };
 		security_skb_classify_flow(skb, &fl);
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			return;
 	}
 
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 160535b51705..da281581692c 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -405,7 +405,7 @@ out:
 	fl.fl4_daddr = eiph->saddr;
 	fl.fl4_tos = RT_TOS(eiph->tos);
 	fl.proto = IPPROTO_IPIP;
-	if (ip_route_output_key(&rt, &key)) {
+	if (ip_route_output_key(&init_net, &rt, &key)) {
 		kfree_skb(skb2);
 		return 0;
 	}
@@ -418,7 +418,7 @@ out:
 		fl.fl4_daddr = eiph->daddr;
 		fl.fl4_src = eiph->saddr;
 		fl.fl4_tos = eiph->tos;
-		if (ip_route_output_key(&rt, &fl) ||
+		if (ip_route_output_key(&init_net, &rt, &fl) ||
 		    rt->u.dst.dev->type != ARPHRD_TUNNEL) {
 			ip_rt_put(rt);
 			kfree_skb(skb2);
@@ -547,7 +547,7 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 						.saddr = tiph->saddr,
 						.tos = RT_TOS(tos) } },
 				    .proto = IPPROTO_IPIP };
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			tunnel->stat.tx_carrier_errors++;
 			goto tx_error_icmp;
 		}
@@ -668,7 +668,7 @@ static void ipip_tunnel_bind_dev(struct net_device *dev)
 						.tos = RT_TOS(iph->tos) } },
 				    .proto = IPPROTO_IPIP };
 		struct rtable *rt;
-		if (!ip_route_output_key(&rt, &fl)) {
+		if (!ip_route_output_key(&init_net, &rt, &fl)) {
 			tdev = rt->u.dst.dev;
 			ip_rt_put(rt);
 		}
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 221271758b93..a94f52c207a7 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1185,7 +1185,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi)
 						.saddr = vif->local,
 						.tos = RT_TOS(iph->tos) } },
 				    .proto = IPPROTO_IPIP };
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			goto out_free;
 		encap = sizeof(struct iphdr);
 	} else {
@@ -1194,7 +1194,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi)
 					      { .daddr = iph->daddr,
 						.tos = RT_TOS(iph->tos) } },
 				    .proto = IPPROTO_IPIP };
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(&init_net, &rt, &fl))
 			goto out_free;
 	}
 
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index 8436bf818590..f63006caea03 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -78,7 +78,7 @@ __ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos)
 						.tos = rtos, } },
 			};
 
-			if (ip_route_output_key(&rt, &fl)) {
+			if (ip_route_output_key(&init_net, &rt, &fl)) {
 				spin_unlock(&dest->dst_lock);
 				IP_VS_DBG_RL("ip_route_output error, "
 					     "dest: %u.%u.%u.%u\n",
@@ -101,7 +101,7 @@ __ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos)
 					.tos = rtos, } },
 		};
 
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			IP_VS_DBG_RL("ip_route_output error, dest: "
 				     "%u.%u.%u.%u\n", NIPQUAD(cp->daddr));
 			return NULL;
@@ -170,7 +170,7 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	EnterFunction(10);
 
-	if (ip_route_output_key(&rt, &fl)) {
+	if (ip_route_output_key(&init_net, &rt, &fl)) {
 		IP_VS_DBG_RL("ip_vs_bypass_xmit(): ip_route_output error, "
 			     "dest: %u.%u.%u.%u\n", NIPQUAD(iph->daddr));
 		goto tx_error_icmp;
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 63221553d26b..9a904c6c0dc8 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -33,7 +33,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
 		fl.nl_u.ip4_u.tos = RT_TOS(iph->tos);
 		fl.oif = skb->sk ? skb->sk->sk_bound_dev_if : 0;
 		fl.mark = skb->mark;
-		if (ip_route_output_key(&rt, &fl) != 0)
+		if (ip_route_output_key(&init_net, &rt, &fl) != 0)
 			return -1;
 
 		/* Drop old route. */
@@ -43,7 +43,7 @@ int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
 		/* non-local src, find valid iif to satisfy
 		 * rp-filter when calling ip_route_input. */
 		fl.nl_u.ip4_u.daddr = iph->saddr;
-		if (ip_route_output_key(&rt, &fl) != 0)
+		if (ip_route_output_key(&init_net, &rt, &fl) != 0)
 			return -1;
 
 		odst = skb->dst;
@@ -187,7 +187,7 @@ EXPORT_SYMBOL(nf_ip_checksum);
 
 static int nf_ip_route(struct dst_entry **dst, struct flowi *fl)
 {
-	return ip_route_output_key((struct rtable **)dst, fl);
+	return ip_route_output_key(&init_net, (struct rtable **)dst, fl);
 }
 
 static const struct nf_afinfo nf_ip_afinfo = {
diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c
index 4391aec56abd..519182269e76 100644
--- a/net/ipv4/netfilter/nf_nat_rule.c
+++ b/net/ipv4/netfilter/nf_nat_rule.c
@@ -97,7 +97,7 @@ static void warn_if_extra_mangle(__be32 dstip, __be32 srcip)
 	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = dstip } } };
 	struct rtable *rt;
 
-	if (ip_route_output_key(&rt, &fl) != 0)
+	if (ip_route_output_key(&init_net, &rt, &fl) != 0)
 		return;
 
 	if (rt->rt_src != srcip && !warned) {
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index c75fc20b07e2..39a40342142f 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2559,9 +2559,9 @@ int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp,
 
 EXPORT_SYMBOL_GPL(ip_route_output_flow);
 
-int ip_route_output_key(struct rtable **rp, struct flowi *flp)
+int ip_route_output_key(struct net *net, struct rtable **rp, struct flowi *flp)
 {
-	return ip_route_output_flow(&init_net, rp, flp, NULL, 0);
+	return ip_route_output_flow(net, rp, flp, NULL, 0);
 }
 
 static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
@@ -2728,7 +2728,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 			},
 			.oif = tb[RTA_OIF] ? nla_get_u32(tb[RTA_OIF]) : 0,
 		};
-		err = ip_route_output_key(&rt, &fl);
+		err = ip_route_output_key(&init_net, &rt, &fl);
 	}
 
 	if (err)
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 2da1be0589a9..f470fe4511db 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -264,7 +264,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 					       { .sport = th->dest,
 						 .dport = th->source } } };
 		security_req_classify_flow(req, &fl);
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			reqsk_free(req);
 			goto out;
 		}
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 425c9ae8b315..9031e521c1df 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -533,7 +533,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	fl.fl4_dst = eiph->saddr;
 	fl.fl4_tos = RT_TOS(eiph->tos);
 	fl.proto = IPPROTO_IPIP;
-	if (ip_route_output_key(&rt, &fl))
+	if (ip_route_output_key(&init_net, &rt, &fl))
 		goto out;
 
 	skb2->dev = rt->u.dst.dev;
@@ -545,7 +545,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		fl.fl4_dst = eiph->daddr;
 		fl.fl4_src = eiph->saddr;
 		fl.fl4_tos = eiph->tos;
-		if (ip_route_output_key(&rt, &fl) ||
+		if (ip_route_output_key(&init_net, &rt, &fl) ||
 		    rt->u.dst.dev->type != ARPHRD_TUNNEL) {
 			ip_rt_put(rt);
 			goto out;
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 1c6fddb80b37..e77239d02bf5 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -557,7 +557,7 @@ static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 						.tos = RT_TOS(tos) } },
 				    .oif = tunnel->parms.link,
 				    .proto = IPPROTO_IPV6 };
-		if (ip_route_output_key(&rt, &fl)) {
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
 			tunnel->stat.tx_carrier_errors++;
 			goto tx_error_icmp;
 		}
@@ -686,7 +686,7 @@ static void ipip6_tunnel_bind_dev(struct net_device *dev)
 				    .oif = tunnel->parms.link,
 				    .proto = IPPROTO_IPV6 };
 		struct rtable *rt;
-		if (!ip_route_output_key(&rt, &fl)) {
+		if (!ip_route_output_key(&init_net, &rt, &fl)) {
 			tdev = rt->u.dst.dev;
 			ip_rt_put(rt);
 		}
diff --git a/net/rxrpc/ar-peer.c b/net/rxrpc/ar-peer.c
index 90fa107a8af9..2abe2081a5e8 100644
--- a/net/rxrpc/ar-peer.c
+++ b/net/rxrpc/ar-peer.c
@@ -57,7 +57,7 @@ static void rxrpc_assess_MTU_size(struct rxrpc_peer *peer)
 		BUG();
 	}
 
-	ret = ip_route_output_key(&rt, &fl);
+	ret = ip_route_output_key(&init_net, &rt, &fl);
 	if (ret < 0) {
 		_leave(" [route err %d]", ret);
 		return;
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 3f7def2936b4..1339742e49f1 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -454,7 +454,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc,
 			  __FUNCTION__, NIPQUAD(fl.fl4_dst),
 			  NIPQUAD(fl.fl4_src));
 
-	if (!ip_route_output_key(&rt, &fl)) {
+	if (!ip_route_output_key(&init_net, &rt, &fl)) {
 		dst = &rt->u.dst;
 	}
 
@@ -497,7 +497,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc,
 		if ((laddr->state == SCTP_ADDR_SRC) &&
 		    (AF_INET == laddr->a.sa.sa_family)) {
 			fl.fl4_src = laddr->a.v4.sin_addr.s_addr;
-			if (!ip_route_output_key(&rt, &fl)) {
+			if (!ip_route_output_key(&init_net, &rt, &fl)) {
 				dst = &rt->u.dst;
 				goto out_unlock;
 			}
-- 
cgit v1.2.3


From 01480e1cf5e2118eba8a8968239f3242072f9563 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Jan 2008 22:10:59 -0800
Subject: [NETLINK]: Add nla_append()

Used to append data to a message without a header or padding.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h |  3 +++
 net/netlink/attr.c    | 19 +++++++++++++++++++
 2 files changed, 22 insertions(+)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index a5e6d0ef51dd..a5506c42f03c 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -91,6 +91,7 @@
  *   nla_reserve_nohdr(skb, len)	reserve room for an attribute w/o hdr
  *   nla_put(skb, type, len, data)	add attribute to skb
  *   nla_put_nohdr(skb, len, data)	add attribute w/o hdr
+ *   nla_append(skb, len, data)		append data to skb
  *
  * Attribute Construction for Basic Types:
  *   nla_put_u8(skb, type, value)	add u8 attribute to skb
@@ -254,6 +255,8 @@ extern int		nla_put(struct sk_buff *skb, int attrtype,
 				int attrlen, const void *data);
 extern int		nla_put_nohdr(struct sk_buff *skb, int attrlen,
 				      const void *data);
+extern int		nla_append(struct sk_buff *skb, int attrlen,
+				   const void *data);
 
 /**************************************************************************
  * Netlink Messages
diff --git a/net/netlink/attr.c b/net/netlink/attr.c
index ec39d12c2423..feb326f4a752 100644
--- a/net/netlink/attr.c
+++ b/net/netlink/attr.c
@@ -430,6 +430,24 @@ int nla_put_nohdr(struct sk_buff *skb, int attrlen, const void *data)
 	return 0;
 }
 
+/**
+ * nla_append - Add a netlink attribute without header or padding
+ * @skb: socket buffer to add attribute to
+ * @attrlen: length of attribute payload
+ * @data: head of attribute payload
+ *
+ * Returns -1 if the tailroom of the skb is insufficient to store
+ * the attribute payload.
+ */
+int nla_append(struct sk_buff *skb, int attrlen, const void *data)
+{
+	if (unlikely(skb_tailroom(skb) < NLA_ALIGN(attrlen)))
+		return -1;
+
+	memcpy(skb_put(skb, attrlen), data, attrlen);
+	return 0;
+}
+
 EXPORT_SYMBOL(nla_validate);
 EXPORT_SYMBOL(nla_parse);
 EXPORT_SYMBOL(nla_find);
@@ -445,3 +463,4 @@ EXPORT_SYMBOL(nla_put_nohdr);
 EXPORT_SYMBOL(nla_memcpy);
 EXPORT_SYMBOL(nla_memcmp);
 EXPORT_SYMBOL(nla_strcmp);
+EXPORT_SYMBOL(nla_append);
-- 
cgit v1.2.3


From 1e90474c377e92db7262a8968a45c1dd980ca9e5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Jan 2008 22:11:17 -0800
Subject: [NET_SCHED]: Convert packet schedulers from rtnetlink to new netlink
 API

Convert packet schedulers to use the netlink API. Unfortunately a gradual
conversion is not possible without breaking compilation in the middle or
adding lots of casts, so this patch converts them all in one step. The
patch has been mostly generated automatically with some minor edits to
at least allow seperate conversion of classifiers and actions.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/gen_stats.h    |   6 +-
 include/net/pkt_sched.h    |   2 +-
 include/net/sch_generic.h  |   6 +-
 net/core/gen_estimator.c   |   8 +--
 net/core/gen_stats.c       |   9 +--
 net/mac80211/wme.c         |  14 ++--
 net/netfilter/xt_RATEEST.c |   7 +-
 net/sched/act_api.c        |   2 +-
 net/sched/act_police.c     |   6 +-
 net/sched/sch_api.c        |  87 ++++++++++++++-----------
 net/sched/sch_atm.c        |  48 +++++++-------
 net/sched/sch_cbq.c        | 156 ++++++++++++++++++++++-----------------------
 net/sched/sch_dsmark.c     |  85 +++++++++++++-----------
 net/sched/sch_fifo.c       |  10 +--
 net/sched/sch_generic.c    |   6 +-
 net/sched/sch_gred.c       |  59 +++++++++--------
 net/sched/sch_hfsc.c       |  72 ++++++++++-----------
 net/sched/sch_htb.c        |  68 ++++++++++----------
 net/sched/sch_ingress.c    |  14 ++--
 net/sched/sch_netem.c      | 100 ++++++++++++++---------------
 net/sched/sch_prio.c       |  30 +++++----
 net/sched/sch_red.c        |  52 +++++++--------
 net/sched/sch_sfq.c        |  12 ++--
 net/sched/sch_tbf.c        |  49 +++++++-------
 net/sched/sch_teql.c       |   2 +-
 25 files changed, 474 insertions(+), 436 deletions(-)

(limited to 'include')

diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index 0b95cf031d6e..8cd8185fa2ed 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -10,7 +10,7 @@ struct gnet_dump
 {
 	spinlock_t *      lock;
 	struct sk_buff *  skb;
-	struct rtattr *   tail;
+	struct nlattr *   tail;
 
 	/* Backward compatability */
 	int               compat_tc_stats;
@@ -39,11 +39,11 @@ extern int gnet_stats_finish_copy(struct gnet_dump *d);
 
 extern int gen_new_estimator(struct gnet_stats_basic *bstats,
 			     struct gnet_stats_rate_est *rate_est,
-			     spinlock_t *stats_lock, struct rtattr *opt);
+			     spinlock_t *stats_lock, struct nlattr *opt);
 extern void gen_kill_estimator(struct gnet_stats_basic *bstats,
 			       struct gnet_stats_rate_est *rate_est);
 extern int gen_replace_estimator(struct gnet_stats_basic *bstats,
 				 struct gnet_stats_rate_est *rate_est,
-				 spinlock_t *stats_lock, struct rtattr *opt);
+				 spinlock_t *stats_lock, struct nlattr *opt);
 
 #endif
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index ab61809a9616..46fb4d80c74a 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -77,7 +77,7 @@ extern int unregister_qdisc(struct Qdisc_ops *qops);
 extern struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle);
 extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle);
 extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
-		struct rtattr *tab);
+		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
 
 extern void __qdisc_run(struct net_device *dev);
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 9418ac26c986..8cacdff24594 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -66,7 +66,7 @@ struct Qdisc_class_ops
 	unsigned long		(*get)(struct Qdisc *, u32 classid);
 	void			(*put)(struct Qdisc *, unsigned long);
 	int			(*change)(struct Qdisc *, u32, u32,
-					struct rtattr **, unsigned long *);
+					struct nlattr **, unsigned long *);
 	int			(*delete)(struct Qdisc *, unsigned long);
 	void			(*walk)(struct Qdisc *, struct qdisc_walker * arg);
 
@@ -95,10 +95,10 @@ struct Qdisc_ops
 	int 			(*requeue)(struct sk_buff *, struct Qdisc *);
 	unsigned int		(*drop)(struct Qdisc *);
 
-	int			(*init)(struct Qdisc *, struct rtattr *arg);
+	int			(*init)(struct Qdisc *, struct nlattr *arg);
 	void			(*reset)(struct Qdisc *);
 	void			(*destroy)(struct Qdisc *);
-	int			(*change)(struct Qdisc *, struct rtattr *arg);
+	int			(*change)(struct Qdisc *, struct nlattr *arg);
 
 	int			(*dump)(struct Qdisc *, struct sk_buff *);
 	int			(*dump_stats)(struct Qdisc *, struct gnet_dump *);
diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c
index 7ab9060bccd0..57abe8266be1 100644
--- a/net/core/gen_estimator.c
+++ b/net/core/gen_estimator.c
@@ -159,13 +159,13 @@ skip:
 int gen_new_estimator(struct gnet_stats_basic *bstats,
 		      struct gnet_stats_rate_est *rate_est,
 		      spinlock_t *stats_lock,
-		      struct rtattr *opt)
+		      struct nlattr *opt)
 {
 	struct gen_estimator *est;
-	struct gnet_estimator *parm = RTA_DATA(opt);
+	struct gnet_estimator *parm = nla_data(opt);
 	int idx;
 
-	if (RTA_PAYLOAD(opt) < sizeof(*parm))
+	if (nla_len(opt) < sizeof(*parm))
 		return -EINVAL;
 
 	if (parm->interval < -2 || parm->interval > 3)
@@ -254,7 +254,7 @@ void gen_kill_estimator(struct gnet_stats_basic *bstats,
  */
 int gen_replace_estimator(struct gnet_stats_basic *bstats,
 			  struct gnet_stats_rate_est *rate_est,
-			  spinlock_t *stats_lock, struct rtattr *opt)
+			  spinlock_t *stats_lock, struct nlattr *opt)
 {
 	gen_kill_estimator(bstats, rate_est);
 	return gen_new_estimator(bstats, rate_est, stats_lock, opt);
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index 8073561f7c63..c3d0ffeac243 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -20,16 +20,17 @@
 #include <linux/socket.h>
 #include <linux/rtnetlink.h>
 #include <linux/gen_stats.h>
+#include <net/netlink.h>
 #include <net/gen_stats.h>
 
 
 static inline int
 gnet_stats_copy(struct gnet_dump *d, int type, void *buf, int size)
 {
-	RTA_PUT(d->skb, type, size, buf);
+	NLA_PUT(d->skb, type, size, buf);
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	spin_unlock_bh(d->lock);
 	return -1;
 }
@@ -62,7 +63,7 @@ gnet_stats_start_copy_compat(struct sk_buff *skb, int type, int tc_stats_type,
 	spin_lock_bh(lock);
 	d->lock = lock;
 	if (type)
-		d->tail = (struct rtattr *)skb_tail_pointer(skb);
+		d->tail = (struct nlattr *)skb_tail_pointer(skb);
 	d->skb = skb;
 	d->compat_tc_stats = tc_stats_type;
 	d->compat_xstats = xstats_type;
@@ -213,7 +214,7 @@ int
 gnet_stats_finish_copy(struct gnet_dump *d)
 {
 	if (d->tail)
-		d->tail->rta_len = skb_tail_pointer(d->skb) - (u8 *)d->tail;
+		d->tail->nla_len = skb_tail_pointer(d->skb) - (u8 *)d->tail;
 
 	if (d->compat_tc_stats)
 		if (gnet_stats_copy(d, d->compat_tc_stats, &d->tc_stats,
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 024519522d37..4e236599dd31 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -297,16 +297,16 @@ static void wme_qdiscop_destroy(struct Qdisc* qd)
 
 
 /* called whenever parameters are updated on existing qdisc */
-static int wme_qdiscop_tune(struct Qdisc *qd, struct rtattr *opt)
+static int wme_qdiscop_tune(struct Qdisc *qd, struct nlattr *opt)
 {
 /*	struct ieee80211_sched_data *q = qdisc_priv(qd);
 */
 	/* check our options block is the right size */
 	/* copy any options to our local structure */
 /*	Ignore options block for now - always use static mapping
-	struct tc_ieee80211_qopt *qopt = RTA_DATA(opt);
+	struct tc_ieee80211_qopt *qopt = nla_data(opt);
 
-	if (opt->rta_len < RTA_LENGTH(sizeof(*qopt)))
+	if (opt->nla_len < nla_attr_size(sizeof(*qopt)))
 		return -EINVAL;
 	memcpy(q->tag2queue, qopt->tag2queue, sizeof(qopt->tag2queue));
 */
@@ -315,7 +315,7 @@ static int wme_qdiscop_tune(struct Qdisc *qd, struct rtattr *opt)
 
 
 /* called during initial creation of qdisc on device */
-static int wme_qdiscop_init(struct Qdisc *qd, struct rtattr *opt)
+static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
 	struct net_device *dev = qd->dev;
@@ -370,10 +370,10 @@ static int wme_qdiscop_dump(struct Qdisc *qd, struct sk_buff *skb)
 	struct tc_ieee80211_qopt opt;
 
 	memcpy(&opt.tag2queue, q->tag2queue, TC_80211_MAX_TAG + 1);
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 */	return skb->len;
 /*
-rtattr_failure:
+nla_put_failure:
 	skb_trim(skb, p - skb->data);*/
 	return -1;
 }
@@ -444,7 +444,7 @@ static void wme_classop_put(struct Qdisc *q, unsigned long cl)
 
 
 static int wme_classop_change(struct Qdisc *qd, u32 handle, u32 parent,
-			      struct rtattr **tca, unsigned long *arg)
+			      struct nlattr **tca, unsigned long *arg)
 {
 	unsigned long cl = *arg;
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c
index c5ba525dc32f..24c73ba31eaa 100644
--- a/net/netfilter/xt_RATEEST.c
+++ b/net/netfilter/xt_RATEEST.c
@@ -12,6 +12,7 @@
 #include <linux/rtnetlink.h>
 #include <linux/random.h>
 #include <net/gen_stats.h>
+#include <net/netlink.h>
 
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_RATEEST.h>
@@ -98,7 +99,7 @@ xt_rateest_tg_checkentry(const char *tablename,
 	struct xt_rateest_target_info *info = (void *)targinfo;
 	struct xt_rateest *est;
 	struct {
-		struct rtattr		opt;
+		struct nlattr		opt;
 		struct gnet_estimator	est;
 	} cfg;
 
@@ -128,8 +129,8 @@ xt_rateest_tg_checkentry(const char *tablename,
 	est->params.interval	= info->interval;
 	est->params.ewma_log	= info->ewma_log;
 
-	cfg.opt.rta_len		= RTA_LENGTH(sizeof(cfg.est));
-	cfg.opt.rta_type	= TCA_STATS_RATE_EST;
+	cfg.opt.nla_len		= nla_attr_size(sizeof(cfg.est));
+	cfg.opt.nla_type	= TCA_STATS_RATE_EST;
 	cfg.est.interval	= info->interval;
 	cfg.est.ewma_log	= info->ewma_log;
 
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 3825508fdcd1..11f3097a6912 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -227,7 +227,7 @@ struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est, struct tc_acti
 	p->tcfc_tm.lastuse = jiffies;
 	if (est)
 		gen_new_estimator(&p->tcfc_bstats, &p->tcfc_rate_est,
-				  &p->tcfc_lock, est);
+				  &p->tcfc_lock, (struct nlattr *)est);
 	a->priv = (void *) p;
 	return p;
 }
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index a73e3e6d87ea..07ffdf9c5e59 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -174,12 +174,12 @@ static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
 override:
 	if (parm->rate.rate) {
 		err = -ENOMEM;
-		R_tab = qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE-1]);
+		R_tab = qdisc_get_rtab(&parm->rate, (struct nlattr *)tb[TCA_POLICE_RATE-1]);
 		if (R_tab == NULL)
 			goto failure;
 		if (parm->peakrate.rate) {
 			P_tab = qdisc_get_rtab(&parm->peakrate,
-					       tb[TCA_POLICE_PEAKRATE-1]);
+					       (struct nlattr *)tb[TCA_POLICE_PEAKRATE-1]);
 			if (P_tab == NULL) {
 				qdisc_put_rtab(R_tab);
 				goto failure;
@@ -216,7 +216,7 @@ override:
 	if (est)
 		gen_replace_estimator(&police->tcf_bstats,
 				      &police->tcf_rate_est,
-				      &police->tcf_lock, est);
+				      &police->tcf_lock, (struct nlattr *)est);
 
 	spin_unlock_bh(&police->tcf_lock);
 	if (ret != ACT_P_CREATED)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index dc89a9343f30..7abb028dd96b 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -213,14 +213,14 @@ static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 
 /* Find queueing discipline by name */
 
-static struct Qdisc_ops *qdisc_lookup_ops(struct rtattr *kind)
+static struct Qdisc_ops *qdisc_lookup_ops(struct nlattr *kind)
 {
 	struct Qdisc_ops *q = NULL;
 
 	if (kind) {
 		read_lock(&qdisc_mod_lock);
 		for (q = qdisc_base; q; q = q->next) {
-			if (rtattr_strcmp(kind, q->id) == 0) {
+			if (nla_strcmp(kind, q->id) == 0) {
 				if (!try_module_get(q->owner))
 					q = NULL;
 				break;
@@ -233,7 +233,7 @@ static struct Qdisc_ops *qdisc_lookup_ops(struct rtattr *kind)
 
 static struct qdisc_rate_table *qdisc_rtab_list;
 
-struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct rtattr *tab)
+struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *tab)
 {
 	struct qdisc_rate_table *rtab;
 
@@ -244,14 +244,14 @@ struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct rtattr *ta
 		}
 	}
 
-	if (tab == NULL || r->rate == 0 || r->cell_log == 0 || RTA_PAYLOAD(tab) != 1024)
+	if (tab == NULL || r->rate == 0 || r->cell_log == 0 || nla_len(tab) != 1024)
 		return NULL;
 
 	rtab = kmalloc(sizeof(*rtab), GFP_KERNEL);
 	if (rtab) {
 		rtab->rate = *r;
 		rtab->refcnt = 1;
-		memcpy(rtab->data, RTA_DATA(tab), 1024);
+		memcpy(rtab->data, nla_data(tab), 1024);
 		rtab->next = qdisc_rtab_list;
 		qdisc_rtab_list = rtab;
 	}
@@ -445,10 +445,10 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
 
 static struct Qdisc *
 qdisc_create(struct net_device *dev, u32 parent, u32 handle,
-	   struct rtattr **tca, int *errp)
+	   struct nlattr **tca, int *errp)
 {
 	int err;
-	struct rtattr *kind = tca[TCA_KIND-1];
+	struct nlattr *kind = tca[TCA_KIND];
 	struct Qdisc *sch;
 	struct Qdisc_ops *ops;
 
@@ -456,7 +456,7 @@ qdisc_create(struct net_device *dev, u32 parent, u32 handle,
 #ifdef CONFIG_KMOD
 	if (ops == NULL && kind != NULL) {
 		char name[IFNAMSIZ];
-		if (rtattr_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) {
+		if (nla_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) {
 			/* We dropped the RTNL semaphore in order to
 			 * perform the module load.  So, even if we
 			 * succeeded in loading the module we have to
@@ -509,11 +509,11 @@ qdisc_create(struct net_device *dev, u32 parent, u32 handle,
 
 	sch->handle = handle;
 
-	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS-1])) == 0) {
-		if (tca[TCA_RATE-1]) {
+	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+		if (tca[TCA_RATE]) {
 			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
 						sch->stats_lock,
-						tca[TCA_RATE-1]);
+						tca[TCA_RATE]);
 			if (err) {
 				/*
 				 * Any broken qdiscs that would require
@@ -541,20 +541,20 @@ err_out:
 	return NULL;
 }
 
-static int qdisc_change(struct Qdisc *sch, struct rtattr **tca)
+static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
 {
-	if (tca[TCA_OPTIONS-1]) {
+	if (tca[TCA_OPTIONS]) {
 		int err;
 
 		if (sch->ops->change == NULL)
 			return -EINVAL;
-		err = sch->ops->change(sch, tca[TCA_OPTIONS-1]);
+		err = sch->ops->change(sch, tca[TCA_OPTIONS]);
 		if (err)
 			return err;
 	}
-	if (tca[TCA_RATE-1])
+	if (tca[TCA_RATE])
 		gen_replace_estimator(&sch->bstats, &sch->rate_est,
-			sch->stats_lock, tca[TCA_RATE-1]);
+			sch->stats_lock, tca[TCA_RATE]);
 	return 0;
 }
 
@@ -606,7 +606,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = skb->sk->sk_net;
 	struct tcmsg *tcm = NLMSG_DATA(n);
-	struct rtattr **tca = arg;
+	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
 	u32 clid = tcm->tcm_parent;
 	struct Qdisc *q = NULL;
@@ -619,6 +619,10 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 	if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return -ENODEV;
 
+	err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
+	if (err < 0)
+		return err;
+
 	if (clid) {
 		if (clid != TC_H_ROOT) {
 			if (TC_H_MAJ(clid) != TC_H_MAJ(TC_H_INGRESS)) {
@@ -641,7 +645,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 			return -ENOENT;
 	}
 
-	if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))
+	if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id))
 		return -EINVAL;
 
 	if (n->nlmsg_type == RTM_DELQDISC) {
@@ -671,7 +675,7 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = skb->sk->sk_net;
 	struct tcmsg *tcm;
-	struct rtattr **tca;
+	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
 	u32 clid;
 	struct Qdisc *q, *p;
@@ -683,13 +687,16 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 replay:
 	/* Reinit, just in case something touches this. */
 	tcm = NLMSG_DATA(n);
-	tca = arg;
 	clid = tcm->tcm_parent;
 	q = p = NULL;
 
 	if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return -ENODEV;
 
+	err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
+	if (err < 0)
+		return err;
+
 	if (clid) {
 		if (clid != TC_H_ROOT) {
 			if (clid != TC_H_INGRESS) {
@@ -717,7 +724,7 @@ replay:
 					goto create_n_graft;
 				if (n->nlmsg_flags&NLM_F_EXCL)
 					return -EEXIST;
-				if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))
+				if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id))
 					return -EINVAL;
 				if (q == p ||
 				    (p && check_loop(q, p, 0)))
@@ -750,8 +757,8 @@ replay:
 				if ((n->nlmsg_flags&NLM_F_CREATE) &&
 				    (n->nlmsg_flags&NLM_F_REPLACE) &&
 				    ((n->nlmsg_flags&NLM_F_EXCL) ||
-				     (tca[TCA_KIND-1] &&
-				      rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))))
+				     (tca[TCA_KIND] &&
+				      nla_strcmp(tca[TCA_KIND], q->ops->id))))
 					goto create_n_graft;
 			}
 		}
@@ -766,7 +773,7 @@ replay:
 		return -ENOENT;
 	if (n->nlmsg_flags&NLM_F_EXCL)
 		return -EEXIST;
-	if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))
+	if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id))
 		return -EINVAL;
 	err = qdisc_change(q, tca);
 	if (err == 0)
@@ -827,31 +834,31 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 	tcm->tcm_parent = clid;
 	tcm->tcm_handle = q->handle;
 	tcm->tcm_info = atomic_read(&q->refcnt);
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
+	NLA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
 	if (q->ops->dump && q->ops->dump(q, skb) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 	q->qstats.qlen = q->q.qlen;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
 			TCA_XSTATS, q->stats_lock, &d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (gnet_stats_copy_basic(&d, &q->bstats) < 0 ||
 	    gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 ||
 	    gnet_stats_copy_queue(&d, &q->qstats) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (gnet_stats_finish_copy(&d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
 nlmsg_failure:
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -939,7 +946,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = skb->sk->sk_net;
 	struct tcmsg *tcm = NLMSG_DATA(n);
-	struct rtattr **tca = arg;
+	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
 	struct Qdisc *q = NULL;
 	const struct Qdisc_class_ops *cops;
@@ -956,6 +963,10 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 	if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return -ENODEV;
 
+	err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL);
+	if (err < 0)
+		return err;
+
 	/*
 	   parent == TC_H_UNSPEC - unspecified parent.
 	   parent == TC_H_ROOT   - class is root, which has no parent.
@@ -1069,25 +1080,25 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
 	tcm->tcm_parent = q->handle;
 	tcm->tcm_handle = q->handle;
 	tcm->tcm_info = 0;
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
+	NLA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
 	if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
 			TCA_XSTATS, q->stats_lock, &d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (gnet_stats_finish_copy(&d) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
 nlmsg_failure:
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 734be9d37d46..eb01aae117df 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -196,13 +196,13 @@ static const u8 llc_oui_ip[] = {
 };				/* Ethertype IP (0800) */
 
 static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
-			 struct rtattr **tca, unsigned long *arg)
+			 struct nlattr **tca, unsigned long *arg)
 {
 	struct atm_qdisc_data *p = qdisc_priv(sch);
 	struct atm_flow_data *flow = (struct atm_flow_data *)*arg;
 	struct atm_flow_data *excess = NULL;
-	struct rtattr *opt = tca[TCA_OPTIONS - 1];
-	struct rtattr *tb[TCA_ATM_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_ATM_MAX + 1];
 	struct socket *sock;
 	int fd, error, hdr_len;
 	void *hdr;
@@ -223,31 +223,31 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 	 */
 	if (flow)
 		return -EBUSY;
-	if (opt == NULL || rtattr_parse_nested(tb, TCA_ATM_MAX, opt))
+	if (opt == NULL || nla_parse_nested(tb, TCA_ATM_MAX, opt, NULL))
 		return -EINVAL;
-	if (!tb[TCA_ATM_FD - 1] || RTA_PAYLOAD(tb[TCA_ATM_FD - 1]) < sizeof(fd))
+	if (!tb[TCA_ATM_FD] || nla_len(tb[TCA_ATM_FD]) < sizeof(fd))
 		return -EINVAL;
-	fd = *(int *)RTA_DATA(tb[TCA_ATM_FD - 1]);
+	fd = *(int *)nla_data(tb[TCA_ATM_FD]);
 	pr_debug("atm_tc_change: fd %d\n", fd);
-	if (tb[TCA_ATM_HDR - 1]) {
-		hdr_len = RTA_PAYLOAD(tb[TCA_ATM_HDR - 1]);
-		hdr = RTA_DATA(tb[TCA_ATM_HDR - 1]);
+	if (tb[TCA_ATM_HDR]) {
+		hdr_len = nla_len(tb[TCA_ATM_HDR]);
+		hdr = nla_data(tb[TCA_ATM_HDR]);
 	} else {
 		hdr_len = RFC1483LLC_LEN;
 		hdr = NULL;	/* default LLC/SNAP for IP */
 	}
-	if (!tb[TCA_ATM_EXCESS - 1])
+	if (!tb[TCA_ATM_EXCESS])
 		excess = NULL;
 	else {
-		if (RTA_PAYLOAD(tb[TCA_ATM_EXCESS - 1]) != sizeof(u32))
+		if (nla_len(tb[TCA_ATM_EXCESS]) != sizeof(u32))
 			return -EINVAL;
 		excess = (struct atm_flow_data *)
-			atm_tc_get(sch, *(u32 *)RTA_DATA(tb[TCA_ATM_EXCESS - 1]));
+			atm_tc_get(sch, *(u32 *)nla_data(tb[TCA_ATM_EXCESS]));
 		if (!excess)
 			return -ENOENT;
 	}
 	pr_debug("atm_tc_change: type %d, payload %lu, hdr_len %d\n",
-		 opt->rta_type, RTA_PAYLOAD(opt), hdr_len);
+		 opt->nla_type, nla_len(opt), hdr_len);
 	sock = sockfd_lookup(fd, &error);
 	if (!sock)
 		return error;	/* f_count++ */
@@ -541,7 +541,7 @@ static unsigned int atm_tc_drop(struct Qdisc *sch)
 	return 0;
 }
 
-static int atm_tc_init(struct Qdisc *sch, struct rtattr *opt)
+static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct atm_qdisc_data *p = qdisc_priv(sch);
 
@@ -602,7 +602,7 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl,
 	struct atm_qdisc_data *p = qdisc_priv(sch);
 	struct atm_flow_data *flow = (struct atm_flow_data *)cl;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	pr_debug("atm_tc_dump_class(sch %p,[qdisc %p],flow %p,skb %p,tcm %p)\n",
 		sch, p, flow, skb, tcm);
@@ -610,9 +610,9 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl,
 		return -EINVAL;
 	tcm->tcm_handle = flow->classid;
 	tcm->tcm_info = flow->q->handle;
-	rta = (struct rtattr *)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
-	RTA_PUT(skb, TCA_ATM_HDR, flow->hdr_len, flow->hdr);
+	nla = (struct nlattr *)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	NLA_PUT(skb, TCA_ATM_HDR, flow->hdr_len, flow->hdr);
 	if (flow->vcc) {
 		struct sockaddr_atmpvc pvc;
 		int state;
@@ -621,21 +621,21 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl,
 		pvc.sap_addr.itf = flow->vcc->dev ? flow->vcc->dev->number : -1;
 		pvc.sap_addr.vpi = flow->vcc->vpi;
 		pvc.sap_addr.vci = flow->vcc->vci;
-		RTA_PUT(skb, TCA_ATM_ADDR, sizeof(pvc), &pvc);
+		NLA_PUT(skb, TCA_ATM_ADDR, sizeof(pvc), &pvc);
 		state = ATM_VF2VS(flow->vcc->flags);
-		RTA_PUT(skb, TCA_ATM_STATE, sizeof(state), &state);
+		NLA_PUT(skb, TCA_ATM_STATE, sizeof(state), &state);
 	}
 	if (flow->excess)
-		RTA_PUT(skb, TCA_ATM_EXCESS, sizeof(u32), &flow->classid);
+		NLA_PUT(skb, TCA_ATM_EXCESS, sizeof(u32), &flow->classid);
 	else {
 		static u32 zero;
 
-		RTA_PUT(skb, TCA_ATM_EXCESS, sizeof(zero), &zero);
+		NLA_PUT(skb, TCA_ATM_EXCESS, sizeof(zero), &zero);
 	}
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index bea123fc24a4..5c8667ef4ba7 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1377,24 +1377,24 @@ static int cbq_set_fopt(struct cbq_class *cl, struct tc_cbq_fopt *fopt)
 	return 0;
 }
 
-static int cbq_init(struct Qdisc *sch, struct rtattr *opt)
+static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
-	struct rtattr *tb[TCA_CBQ_MAX];
+	struct nlattr *tb[TCA_CBQ_MAX + 1];
 	struct tc_ratespec *r;
 
-	if (rtattr_parse_nested(tb, TCA_CBQ_MAX, opt) < 0 ||
-	    tb[TCA_CBQ_RTAB-1] == NULL || tb[TCA_CBQ_RATE-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_CBQ_RATE-1]) < sizeof(struct tc_ratespec))
+	if (nla_parse_nested(tb, TCA_CBQ_MAX, opt, NULL) < 0 ||
+	    tb[TCA_CBQ_RTAB] == NULL || tb[TCA_CBQ_RATE] == NULL ||
+	    nla_len(tb[TCA_CBQ_RATE]) < sizeof(struct tc_ratespec))
 		return -EINVAL;
 
-	if (tb[TCA_CBQ_LSSOPT-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_LSSOPT-1]) < sizeof(struct tc_cbq_lssopt))
+	if (tb[TCA_CBQ_LSSOPT] &&
+	    nla_len(tb[TCA_CBQ_LSSOPT]) < sizeof(struct tc_cbq_lssopt))
 		return -EINVAL;
 
-	r = RTA_DATA(tb[TCA_CBQ_RATE-1]);
+	r = nla_data(tb[TCA_CBQ_RATE]);
 
-	if ((q->link.R_tab = qdisc_get_rtab(r, tb[TCA_CBQ_RTAB-1])) == NULL)
+	if ((q->link.R_tab = qdisc_get_rtab(r, tb[TCA_CBQ_RTAB])) == NULL)
 		return -EINVAL;
 
 	q->link.refcnt = 1;
@@ -1427,8 +1427,8 @@ static int cbq_init(struct Qdisc *sch, struct rtattr *opt)
 
 	cbq_link_class(&q->link);
 
-	if (tb[TCA_CBQ_LSSOPT-1])
-		cbq_set_lss(&q->link, RTA_DATA(tb[TCA_CBQ_LSSOPT-1]));
+	if (tb[TCA_CBQ_LSSOPT])
+		cbq_set_lss(&q->link, nla_data(tb[TCA_CBQ_LSSOPT]));
 
 	cbq_addprio(q, &q->link);
 	return 0;
@@ -1438,10 +1438,10 @@ static __inline__ int cbq_dump_rate(struct sk_buff *skb, struct cbq_class *cl)
 {
 	unsigned char *b = skb_tail_pointer(skb);
 
-	RTA_PUT(skb, TCA_CBQ_RATE, sizeof(cl->R_tab->rate), &cl->R_tab->rate);
+	NLA_PUT(skb, TCA_CBQ_RATE, sizeof(cl->R_tab->rate), &cl->R_tab->rate);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1463,10 +1463,10 @@ static __inline__ int cbq_dump_lss(struct sk_buff *skb, struct cbq_class *cl)
 	opt.minidle = (u32)(-cl->minidle);
 	opt.offtime = cl->offtime;
 	opt.change = ~0;
-	RTA_PUT(skb, TCA_CBQ_LSSOPT, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_CBQ_LSSOPT, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1481,10 +1481,10 @@ static __inline__ int cbq_dump_wrr(struct sk_buff *skb, struct cbq_class *cl)
 	opt.priority = cl->priority+1;
 	opt.cpriority = cl->cpriority+1;
 	opt.weight = cl->weight;
-	RTA_PUT(skb, TCA_CBQ_WRROPT, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_CBQ_WRROPT, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1498,10 +1498,10 @@ static __inline__ int cbq_dump_ovl(struct sk_buff *skb, struct cbq_class *cl)
 	opt.priority2 = cl->priority2+1;
 	opt.pad = 0;
 	opt.penalty = cl->penalty;
-	RTA_PUT(skb, TCA_CBQ_OVL_STRATEGY, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_CBQ_OVL_STRATEGY, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1515,11 +1515,11 @@ static __inline__ int cbq_dump_fopt(struct sk_buff *skb, struct cbq_class *cl)
 		opt.split = cl->split ? cl->split->classid : 0;
 		opt.defmap = cl->defmap;
 		opt.defchange = ~0;
-		RTA_PUT(skb, TCA_CBQ_FOPT, sizeof(opt), &opt);
+		NLA_PUT(skb, TCA_CBQ_FOPT, sizeof(opt), &opt);
 	}
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1534,11 +1534,11 @@ static __inline__ int cbq_dump_police(struct sk_buff *skb, struct cbq_class *cl)
 		opt.police = cl->police;
 		opt.__res1 = 0;
 		opt.__res2 = 0;
-		RTA_PUT(skb, TCA_CBQ_POLICE, sizeof(opt), &opt);
+		NLA_PUT(skb, TCA_CBQ_POLICE, sizeof(opt), &opt);
 	}
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1562,16 +1562,16 @@ static int cbq_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 	if (cbq_dump_attr(skb, &q->link) < 0)
-		goto rtattr_failure;
-	rta->rta_len = skb_tail_pointer(skb) - b;
+		goto nla_put_failure;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1591,7 +1591,7 @@ cbq_dump_class(struct Qdisc *sch, unsigned long arg,
 {
 	struct cbq_class *cl = (struct cbq_class*)arg;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	if (cl->tparent)
 		tcm->tcm_parent = cl->tparent->classid;
@@ -1600,14 +1600,14 @@ cbq_dump_class(struct Qdisc *sch, unsigned long arg,
 	tcm->tcm_handle = cl->classid;
 	tcm->tcm_info = cl->q->handle;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 	if (cbq_dump_attr(skb, cl) < 0)
-		goto rtattr_failure;
-	rta->rta_len = skb_tail_pointer(skb) - b;
+		goto nla_put_failure;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1753,43 +1753,43 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg)
 }
 
 static int
-cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **tca,
+cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **tca,
 		 unsigned long *arg)
 {
 	int err;
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	struct cbq_class *cl = (struct cbq_class*)*arg;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_CBQ_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_CBQ_MAX + 1];
 	struct cbq_class *parent;
 	struct qdisc_rate_table *rtab = NULL;
 
-	if (opt==NULL || rtattr_parse_nested(tb, TCA_CBQ_MAX, opt))
+	if (opt==NULL || nla_parse_nested(tb, TCA_CBQ_MAX, opt, NULL))
 		return -EINVAL;
 
-	if (tb[TCA_CBQ_OVL_STRATEGY-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_OVL_STRATEGY-1]) < sizeof(struct tc_cbq_ovl))
+	if (tb[TCA_CBQ_OVL_STRATEGY] &&
+	    nla_len(tb[TCA_CBQ_OVL_STRATEGY]) < sizeof(struct tc_cbq_ovl))
 		return -EINVAL;
 
-	if (tb[TCA_CBQ_FOPT-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_FOPT-1]) < sizeof(struct tc_cbq_fopt))
+	if (tb[TCA_CBQ_FOPT] &&
+	    nla_len(tb[TCA_CBQ_FOPT]) < sizeof(struct tc_cbq_fopt))
 		return -EINVAL;
 
-	if (tb[TCA_CBQ_RATE-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_RATE-1]) < sizeof(struct tc_ratespec))
+	if (tb[TCA_CBQ_RATE] &&
+	    nla_len(tb[TCA_CBQ_RATE]) < sizeof(struct tc_ratespec))
 			return -EINVAL;
 
-	if (tb[TCA_CBQ_LSSOPT-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_LSSOPT-1]) < sizeof(struct tc_cbq_lssopt))
+	if (tb[TCA_CBQ_LSSOPT] &&
+	    nla_len(tb[TCA_CBQ_LSSOPT]) < sizeof(struct tc_cbq_lssopt))
 			return -EINVAL;
 
-	if (tb[TCA_CBQ_WRROPT-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_WRROPT-1]) < sizeof(struct tc_cbq_wrropt))
+	if (tb[TCA_CBQ_WRROPT] &&
+	    nla_len(tb[TCA_CBQ_WRROPT]) < sizeof(struct tc_cbq_wrropt))
 			return -EINVAL;
 
 #ifdef CONFIG_NET_CLS_ACT
-	if (tb[TCA_CBQ_POLICE-1] &&
-	    RTA_PAYLOAD(tb[TCA_CBQ_POLICE-1]) < sizeof(struct tc_cbq_police))
+	if (tb[TCA_CBQ_POLICE] &&
+	    nla_len(tb[TCA_CBQ_POLICE]) < sizeof(struct tc_cbq_police))
 			return -EINVAL;
 #endif
 
@@ -1802,8 +1802,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t
 				return -EINVAL;
 		}
 
-		if (tb[TCA_CBQ_RATE-1]) {
-			rtab = qdisc_get_rtab(RTA_DATA(tb[TCA_CBQ_RATE-1]), tb[TCA_CBQ_RTAB-1]);
+		if (tb[TCA_CBQ_RATE]) {
+			rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), tb[TCA_CBQ_RTAB]);
 			if (rtab == NULL)
 				return -EINVAL;
 		}
@@ -1819,45 +1819,45 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t
 			qdisc_put_rtab(rtab);
 		}
 
-		if (tb[TCA_CBQ_LSSOPT-1])
-			cbq_set_lss(cl, RTA_DATA(tb[TCA_CBQ_LSSOPT-1]));
+		if (tb[TCA_CBQ_LSSOPT])
+			cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT]));
 
-		if (tb[TCA_CBQ_WRROPT-1]) {
+		if (tb[TCA_CBQ_WRROPT]) {
 			cbq_rmprio(q, cl);
-			cbq_set_wrr(cl, RTA_DATA(tb[TCA_CBQ_WRROPT-1]));
+			cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT]));
 		}
 
-		if (tb[TCA_CBQ_OVL_STRATEGY-1])
-			cbq_set_overlimit(cl, RTA_DATA(tb[TCA_CBQ_OVL_STRATEGY-1]));
+		if (tb[TCA_CBQ_OVL_STRATEGY])
+			cbq_set_overlimit(cl, nla_data(tb[TCA_CBQ_OVL_STRATEGY]));
 
 #ifdef CONFIG_NET_CLS_ACT
-		if (tb[TCA_CBQ_POLICE-1])
-			cbq_set_police(cl, RTA_DATA(tb[TCA_CBQ_POLICE-1]));
+		if (tb[TCA_CBQ_POLICE])
+			cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE]));
 #endif
 
-		if (tb[TCA_CBQ_FOPT-1])
-			cbq_set_fopt(cl, RTA_DATA(tb[TCA_CBQ_FOPT-1]));
+		if (tb[TCA_CBQ_FOPT])
+			cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT]));
 
 		if (cl->q->q.qlen)
 			cbq_activate_class(cl);
 
 		sch_tree_unlock(sch);
 
-		if (tca[TCA_RATE-1])
+		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
 					      &sch->dev->queue_lock,
-					      tca[TCA_RATE-1]);
+					      tca[TCA_RATE]);
 		return 0;
 	}
 
 	if (parentid == TC_H_ROOT)
 		return -EINVAL;
 
-	if (tb[TCA_CBQ_WRROPT-1] == NULL || tb[TCA_CBQ_RATE-1] == NULL ||
-	    tb[TCA_CBQ_LSSOPT-1] == NULL)
+	if (tb[TCA_CBQ_WRROPT] == NULL || tb[TCA_CBQ_RATE] == NULL ||
+	    tb[TCA_CBQ_LSSOPT] == NULL)
 		return -EINVAL;
 
-	rtab = qdisc_get_rtab(RTA_DATA(tb[TCA_CBQ_RATE-1]), tb[TCA_CBQ_RTAB-1]);
+	rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), tb[TCA_CBQ_RTAB]);
 	if (rtab == NULL)
 		return -EINVAL;
 
@@ -1912,8 +1912,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t
 		cl->share = cl->tparent;
 	cbq_adjust_levels(parent);
 	cl->minidle = -0x7FFFFFFF;
-	cbq_set_lss(cl, RTA_DATA(tb[TCA_CBQ_LSSOPT-1]));
-	cbq_set_wrr(cl, RTA_DATA(tb[TCA_CBQ_WRROPT-1]));
+	cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT]));
+	cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT]));
 	if (cl->ewma_log==0)
 		cl->ewma_log = q->link.ewma_log;
 	if (cl->maxidle==0)
@@ -1921,19 +1921,19 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t
 	if (cl->avpkt==0)
 		cl->avpkt = q->link.avpkt;
 	cl->overlimit = cbq_ovl_classic;
-	if (tb[TCA_CBQ_OVL_STRATEGY-1])
-		cbq_set_overlimit(cl, RTA_DATA(tb[TCA_CBQ_OVL_STRATEGY-1]));
+	if (tb[TCA_CBQ_OVL_STRATEGY])
+		cbq_set_overlimit(cl, nla_data(tb[TCA_CBQ_OVL_STRATEGY]));
 #ifdef CONFIG_NET_CLS_ACT
-	if (tb[TCA_CBQ_POLICE-1])
-		cbq_set_police(cl, RTA_DATA(tb[TCA_CBQ_POLICE-1]));
+	if (tb[TCA_CBQ_POLICE])
+		cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE]));
 #endif
-	if (tb[TCA_CBQ_FOPT-1])
-		cbq_set_fopt(cl, RTA_DATA(tb[TCA_CBQ_FOPT-1]));
+	if (tb[TCA_CBQ_FOPT])
+		cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT]));
 	sch_tree_unlock(sch);
 
-	if (tca[TCA_RATE-1])
+	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev->queue_lock, tca[TCA_RATE-1]);
+				  &sch->dev->queue_lock, tca[TCA_RATE]);
 
 	*arg = (unsigned long)cl;
 	return 0;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 40e06a6890da..f183ab768873 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -100,11 +100,11 @@ static void dsmark_put(struct Qdisc *sch, unsigned long cl)
 }
 
 static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent,
-			 struct rtattr **tca, unsigned long *arg)
+			 struct nlattr **tca, unsigned long *arg)
 {
 	struct dsmark_qdisc_data *p = qdisc_priv(sch);
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_DSMARK_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_DSMARK_MAX + 1];
 	int err = -EINVAL;
 	u8 mask = 0;
 
@@ -113,24 +113,29 @@ static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent,
 
 	if (!dsmark_valid_index(p, *arg)) {
 		err = -ENOENT;
-		goto rtattr_failure;
+		goto errout;
 	}
 
-	if (!opt || rtattr_parse_nested(tb, TCA_DSMARK_MAX, opt))
-		goto rtattr_failure;
-
-	if (tb[TCA_DSMARK_MASK-1])
-		mask = RTA_GET_U8(tb[TCA_DSMARK_MASK-1]);
+	if (!opt || nla_parse_nested(tb, TCA_DSMARK_MAX, opt, NULL))
+		goto errout;
 
-	if (tb[TCA_DSMARK_VALUE-1])
-		p->value[*arg-1] = RTA_GET_U8(tb[TCA_DSMARK_VALUE-1]);
+	if (tb[TCA_DSMARK_MASK]) {
+		if (nla_len(tb[TCA_DSMARK_MASK]) < sizeof(u8))
+			goto errout;
+		mask = nla_get_u8(tb[TCA_DSMARK_MASK]);
+	}
+	if (tb[TCA_DSMARK_VALUE]) {
+		if (nla_len(tb[TCA_DSMARK_VALUE]) < sizeof(u8))
+			goto errout;
+		p->value[*arg-1] = nla_get_u8(tb[TCA_DSMARK_VALUE]);
+	}
 
-	if (tb[TCA_DSMARK_MASK-1])
+	if (tb[TCA_DSMARK_MASK])
 		p->mask[*arg-1] = mask;
 
 	err = 0;
 
-rtattr_failure:
+errout:
 	return err;
 }
 
@@ -335,10 +340,10 @@ static unsigned int dsmark_drop(struct Qdisc *sch)
 	return len;
 }
 
-static int dsmark_init(struct Qdisc *sch, struct rtattr *opt)
+static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct dsmark_qdisc_data *p = qdisc_priv(sch);
-	struct rtattr *tb[TCA_DSMARK_MAX];
+	struct nlattr *tb[TCA_DSMARK_MAX + 1];
 	int err = -EINVAL;
 	u32 default_index = NO_DEFAULT_INDEX;
 	u16 indices;
@@ -346,16 +351,21 @@ static int dsmark_init(struct Qdisc *sch, struct rtattr *opt)
 
 	pr_debug("dsmark_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt);
 
-	if (!opt || rtattr_parse_nested(tb, TCA_DSMARK_MAX, opt) < 0)
+	if (!opt || nla_parse_nested(tb, TCA_DSMARK_MAX, opt, NULL) < 0)
 		goto errout;
 
-	indices = RTA_GET_U16(tb[TCA_DSMARK_INDICES-1]);
+	if (nla_len(tb[TCA_DSMARK_INDICES]) < sizeof(u16))
+		goto errout;
+	indices = nla_get_u16(tb[TCA_DSMARK_INDICES]);
 
 	if (hweight32(indices) != 1)
 		goto errout;
 
-	if (tb[TCA_DSMARK_DEFAULT_INDEX-1])
-		default_index = RTA_GET_U16(tb[TCA_DSMARK_DEFAULT_INDEX-1]);
+	if (tb[TCA_DSMARK_DEFAULT_INDEX]) {
+		if (nla_len(tb[TCA_DSMARK_DEFAULT_INDEX]) < sizeof(u16))
+			goto errout;
+		default_index = nla_get_u16(tb[TCA_DSMARK_DEFAULT_INDEX]);
+	}
 
 	mask = kmalloc(indices * 2, GFP_KERNEL);
 	if (mask == NULL) {
@@ -371,7 +381,7 @@ static int dsmark_init(struct Qdisc *sch, struct rtattr *opt)
 
 	p->indices = indices;
 	p->default_index = default_index;
-	p->set_tc_index = RTA_GET_FLAG(tb[TCA_DSMARK_SET_TC_INDEX-1]);
+	p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
 
 	p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
 	if (p->q == NULL)
@@ -381,7 +391,6 @@ static int dsmark_init(struct Qdisc *sch, struct rtattr *opt)
 
 	err = 0;
 errout:
-rtattr_failure:
 	return err;
 }
 
@@ -409,7 +418,7 @@ static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl,
 			     struct sk_buff *skb, struct tcmsg *tcm)
 {
 	struct dsmark_qdisc_data *p = qdisc_priv(sch);
-	struct rtattr *opts = NULL;
+	struct nlattr *opts = NULL;
 
 	pr_debug("dsmark_dump_class(sch %p,[qdisc %p],class %ld\n", sch, p, cl);
 
@@ -419,34 +428,38 @@ static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl,
 	tcm->tcm_handle = TC_H_MAKE(TC_H_MAJ(sch->handle), cl-1);
 	tcm->tcm_info = p->q->handle;
 
-	opts = RTA_NEST(skb, TCA_OPTIONS);
-	RTA_PUT_U8(skb, TCA_DSMARK_MASK, p->mask[cl-1]);
-	RTA_PUT_U8(skb, TCA_DSMARK_VALUE, p->value[cl-1]);
+	opts = nla_nest_start(skb, TCA_OPTIONS);
+	if (opts == NULL)
+		goto nla_put_failure;
+	NLA_PUT_U8(skb, TCA_DSMARK_MASK, p->mask[cl-1]);
+	NLA_PUT_U8(skb, TCA_DSMARK_VALUE, p->value[cl-1]);
 
-	return RTA_NEST_END(skb, opts);
+	return nla_nest_end(skb, opts);
 
-rtattr_failure:
-	return RTA_NEST_CANCEL(skb, opts);
+nla_put_failure:
+	return nla_nest_cancel(skb, opts);
 }
 
 static int dsmark_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct dsmark_qdisc_data *p = qdisc_priv(sch);
-	struct rtattr *opts = NULL;
+	struct nlattr *opts = NULL;
 
-	opts = RTA_NEST(skb, TCA_OPTIONS);
-	RTA_PUT_U16(skb, TCA_DSMARK_INDICES, p->indices);
+	opts = nla_nest_start(skb, TCA_OPTIONS);
+	if (opts == NULL)
+		goto nla_put_failure;
+	NLA_PUT_U16(skb, TCA_DSMARK_INDICES, p->indices);
 
 	if (p->default_index != NO_DEFAULT_INDEX)
-		RTA_PUT_U16(skb, TCA_DSMARK_DEFAULT_INDEX, p->default_index);
+		NLA_PUT_U16(skb, TCA_DSMARK_DEFAULT_INDEX, p->default_index);
 
 	if (p->set_tc_index)
-		RTA_PUT_FLAG(skb, TCA_DSMARK_SET_TC_INDEX);
+		NLA_PUT_FLAG(skb, TCA_DSMARK_SET_TC_INDEX);
 
-	return RTA_NEST_END(skb, opts);
+	return nla_nest_end(skb, opts);
 
-rtattr_failure:
-	return RTA_NEST_CANCEL(skb, opts);
+nla_put_failure:
+	return nla_nest_cancel(skb, opts);
 }
 
 static const struct Qdisc_class_ops dsmark_class_ops = {
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index fd0591903c8a..95ed48221652 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -43,7 +43,7 @@ static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 	return qdisc_reshape_fail(skb, sch);
 }
 
-static int fifo_init(struct Qdisc *sch, struct rtattr *opt)
+static int fifo_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct fifo_sched_data *q = qdisc_priv(sch);
 
@@ -55,9 +55,9 @@ static int fifo_init(struct Qdisc *sch, struct rtattr *opt)
 
 		q->limit = limit;
 	} else {
-		struct tc_fifo_qopt *ctl = RTA_DATA(opt);
+		struct tc_fifo_qopt *ctl = nla_data(opt);
 
-		if (RTA_PAYLOAD(opt) < sizeof(*ctl))
+		if (nla_len(opt) < sizeof(*ctl))
 			return -EINVAL;
 
 		q->limit = ctl->limit;
@@ -71,10 +71,10 @@ static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct fifo_sched_data *q = qdisc_priv(sch);
 	struct tc_fifo_qopt opt = { .limit = q->limit };
 
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 51e64acd5098..10b5c0887fff 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -397,14 +397,14 @@ static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb)
 	struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS };
 
 	memcpy(&opt.priomap, prio2band, TC_PRIO_MAX+1);
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
-static int pfifo_fast_init(struct Qdisc *qdisc, struct rtattr *opt)
+static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt)
 {
 	int prio;
 	struct sk_buff_head *list = qdisc_priv(qdisc);
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index e2bcd6682c70..6b784838a534 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -350,16 +350,16 @@ static inline void gred_destroy_vq(struct gred_sched_data *q)
 	kfree(q);
 }
 
-static inline int gred_change_table_def(struct Qdisc *sch, struct rtattr *dps)
+static inline int gred_change_table_def(struct Qdisc *sch, struct nlattr *dps)
 {
 	struct gred_sched *table = qdisc_priv(sch);
 	struct tc_gred_sopt *sopt;
 	int i;
 
-	if (dps == NULL || RTA_PAYLOAD(dps) < sizeof(*sopt))
+	if (dps == NULL || nla_len(dps) < sizeof(*sopt))
 		return -EINVAL;
 
-	sopt = RTA_DATA(dps);
+	sopt = nla_data(dps);
 
 	if (sopt->DPs > MAX_DPs || sopt->DPs == 0 || sopt->def_DP >= sopt->DPs)
 		return -EINVAL;
@@ -425,28 +425,28 @@ static inline int gred_change_vq(struct Qdisc *sch, int dp,
 	return 0;
 }
 
-static int gred_change(struct Qdisc *sch, struct rtattr *opt)
+static int gred_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct gred_sched *table = qdisc_priv(sch);
 	struct tc_gred_qopt *ctl;
-	struct rtattr *tb[TCA_GRED_MAX];
+	struct nlattr *tb[TCA_GRED_MAX + 1];
 	int err = -EINVAL, prio = GRED_DEF_PRIO;
 	u8 *stab;
 
-	if (opt == NULL || rtattr_parse_nested(tb, TCA_GRED_MAX, opt))
+	if (opt == NULL || nla_parse_nested(tb, TCA_GRED_MAX, opt, NULL))
 		return -EINVAL;
 
-	if (tb[TCA_GRED_PARMS-1] == NULL && tb[TCA_GRED_STAB-1] == NULL)
+	if (tb[TCA_GRED_PARMS] == NULL && tb[TCA_GRED_STAB] == NULL)
 		return gred_change_table_def(sch, opt);
 
-	if (tb[TCA_GRED_PARMS-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_GRED_PARMS-1]) < sizeof(*ctl) ||
-	    tb[TCA_GRED_STAB-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_GRED_STAB-1]) < 256)
+	if (tb[TCA_GRED_PARMS] == NULL ||
+	    nla_len(tb[TCA_GRED_PARMS]) < sizeof(*ctl) ||
+	    tb[TCA_GRED_STAB] == NULL ||
+	    nla_len(tb[TCA_GRED_STAB]) < 256)
 		return -EINVAL;
 
-	ctl = RTA_DATA(tb[TCA_GRED_PARMS-1]);
-	stab = RTA_DATA(tb[TCA_GRED_STAB-1]);
+	ctl = nla_data(tb[TCA_GRED_PARMS]);
+	stab = nla_data(tb[TCA_GRED_STAB]);
 
 	if (ctl->DP >= table->DPs)
 		goto errout;
@@ -486,23 +486,23 @@ errout:
 	return err;
 }
 
-static int gred_init(struct Qdisc *sch, struct rtattr *opt)
+static int gred_init(struct Qdisc *sch, struct nlattr *opt)
 {
-	struct rtattr *tb[TCA_GRED_MAX];
+	struct nlattr *tb[TCA_GRED_MAX + 1];
 
-	if (opt == NULL || rtattr_parse_nested(tb, TCA_GRED_MAX, opt))
+	if (opt == NULL || nla_parse_nested(tb, TCA_GRED_MAX, opt, NULL))
 		return -EINVAL;
 
-	if (tb[TCA_GRED_PARMS-1] || tb[TCA_GRED_STAB-1])
+	if (tb[TCA_GRED_PARMS] || tb[TCA_GRED_STAB])
 		return -EINVAL;
 
-	return gred_change_table_def(sch, tb[TCA_GRED_DPS-1]);
+	return gred_change_table_def(sch, tb[TCA_GRED_DPS]);
 }
 
 static int gred_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct gred_sched *table = qdisc_priv(sch);
-	struct rtattr *parms, *opts = NULL;
+	struct nlattr *parms, *opts = NULL;
 	int i;
 	struct tc_gred_sopt sopt = {
 		.DPs	= table->DPs,
@@ -511,9 +511,13 @@ static int gred_dump(struct Qdisc *sch, struct sk_buff *skb)
 		.flags	= table->red_flags,
 	};
 
-	opts = RTA_NEST(skb, TCA_OPTIONS);
-	RTA_PUT(skb, TCA_GRED_DPS, sizeof(sopt), &sopt);
-	parms = RTA_NEST(skb, TCA_GRED_PARMS);
+	opts = nla_nest_start(skb, TCA_OPTIONS);
+	if (opts == NULL)
+		goto nla_put_failure;
+	NLA_PUT(skb, TCA_GRED_DPS, sizeof(sopt), &sopt);
+	parms = nla_nest_start(skb, TCA_GRED_PARMS);
+	if (parms == NULL)
+		goto nla_put_failure;
 
 	for (i = 0; i < MAX_DPs; i++) {
 		struct gred_sched_data *q = table->tab[i];
@@ -555,15 +559,16 @@ static int gred_dump(struct Qdisc *sch, struct sk_buff *skb)
 		opt.qave = red_calc_qavg(&q->parms, q->parms.qavg);
 
 append_opt:
-		RTA_APPEND(skb, sizeof(opt), &opt);
+		if (nla_append(skb, sizeof(opt), &opt) < 0)
+			goto nla_put_failure;
 	}
 
-	RTA_NEST_END(skb, parms);
+	nla_nest_end(skb, parms);
 
-	return RTA_NEST_END(skb, opts);
+	return nla_nest_end(skb, opts);
 
-rtattr_failure:
-	return RTA_NEST_CANCEL(skb, opts);
+nla_put_failure:
+	return nla_nest_cancel(skb, opts);
 }
 
 static void gred_destroy(struct Qdisc *sch)
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 69dc3bccf024..4e6a164d3058 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -988,39 +988,39 @@ hfsc_change_usc(struct hfsc_class *cl, struct tc_service_curve *usc,
 
 static int
 hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
-		  struct rtattr **tca, unsigned long *arg)
+		  struct nlattr **tca, unsigned long *arg)
 {
 	struct hfsc_sched *q = qdisc_priv(sch);
 	struct hfsc_class *cl = (struct hfsc_class *)*arg;
 	struct hfsc_class *parent = NULL;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_HFSC_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_HFSC_MAX + 1];
 	struct tc_service_curve *rsc = NULL, *fsc = NULL, *usc = NULL;
 	u64 cur_time;
 
-	if (opt == NULL || rtattr_parse_nested(tb, TCA_HFSC_MAX, opt))
+	if (opt == NULL || nla_parse_nested(tb, TCA_HFSC_MAX, opt, NULL))
 		return -EINVAL;
 
-	if (tb[TCA_HFSC_RSC-1]) {
-		if (RTA_PAYLOAD(tb[TCA_HFSC_RSC-1]) < sizeof(*rsc))
+	if (tb[TCA_HFSC_RSC]) {
+		if (nla_len(tb[TCA_HFSC_RSC]) < sizeof(*rsc))
 			return -EINVAL;
-		rsc = RTA_DATA(tb[TCA_HFSC_RSC-1]);
+		rsc = nla_data(tb[TCA_HFSC_RSC]);
 		if (rsc->m1 == 0 && rsc->m2 == 0)
 			rsc = NULL;
 	}
 
-	if (tb[TCA_HFSC_FSC-1]) {
-		if (RTA_PAYLOAD(tb[TCA_HFSC_FSC-1]) < sizeof(*fsc))
+	if (tb[TCA_HFSC_FSC]) {
+		if (nla_len(tb[TCA_HFSC_FSC]) < sizeof(*fsc))
 			return -EINVAL;
-		fsc = RTA_DATA(tb[TCA_HFSC_FSC-1]);
+		fsc = nla_data(tb[TCA_HFSC_FSC]);
 		if (fsc->m1 == 0 && fsc->m2 == 0)
 			fsc = NULL;
 	}
 
-	if (tb[TCA_HFSC_USC-1]) {
-		if (RTA_PAYLOAD(tb[TCA_HFSC_USC-1]) < sizeof(*usc))
+	if (tb[TCA_HFSC_USC]) {
+		if (nla_len(tb[TCA_HFSC_USC]) < sizeof(*usc))
 			return -EINVAL;
-		usc = RTA_DATA(tb[TCA_HFSC_USC-1]);
+		usc = nla_data(tb[TCA_HFSC_USC]);
 		if (usc->m1 == 0 && usc->m2 == 0)
 			usc = NULL;
 	}
@@ -1050,10 +1050,10 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 		}
 		sch_tree_unlock(sch);
 
-		if (tca[TCA_RATE-1])
+		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
 					      &sch->dev->queue_lock,
-					      tca[TCA_RATE-1]);
+					      tca[TCA_RATE]);
 		return 0;
 	}
 
@@ -1106,9 +1106,9 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	cl->cl_pcvtoff = parent->cl_cvtoff;
 	sch_tree_unlock(sch);
 
-	if (tca[TCA_RATE-1])
+	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev->queue_lock, tca[TCA_RATE-1]);
+				  &sch->dev->queue_lock, tca[TCA_RATE]);
 	*arg = (unsigned long)cl;
 	return 0;
 }
@@ -1304,11 +1304,11 @@ hfsc_dump_sc(struct sk_buff *skb, int attr, struct internal_sc *sc)
 	tsc.m1 = sm2m(sc->sm1);
 	tsc.d  = dx2d(sc->dx);
 	tsc.m2 = sm2m(sc->sm2);
-	RTA_PUT(skb, attr, sizeof(tsc), &tsc);
+	NLA_PUT(skb, attr, sizeof(tsc), &tsc);
 
 	return skb->len;
 
- rtattr_failure:
+ nla_put_failure:
 	return -1;
 }
 
@@ -1317,19 +1317,19 @@ hfsc_dump_curves(struct sk_buff *skb, struct hfsc_class *cl)
 {
 	if ((cl->cl_flags & HFSC_RSC) &&
 	    (hfsc_dump_sc(skb, TCA_HFSC_RSC, &cl->cl_rsc) < 0))
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if ((cl->cl_flags & HFSC_FSC) &&
 	    (hfsc_dump_sc(skb, TCA_HFSC_FSC, &cl->cl_fsc) < 0))
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if ((cl->cl_flags & HFSC_USC) &&
 	    (hfsc_dump_sc(skb, TCA_HFSC_USC, &cl->cl_usc) < 0))
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	return skb->len;
 
- rtattr_failure:
+ nla_put_failure:
 	return -1;
 }
 
@@ -1339,20 +1339,20 @@ hfsc_dump_class(struct Qdisc *sch, unsigned long arg, struct sk_buff *skb,
 {
 	struct hfsc_class *cl = (struct hfsc_class *)arg;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta = (struct rtattr *)b;
+	struct nlattr *nla = (struct nlattr *)b;
 
 	tcm->tcm_parent = cl->cl_parent ? cl->cl_parent->classid : TC_H_ROOT;
 	tcm->tcm_handle = cl->classid;
 	if (cl->level == 0)
 		tcm->tcm_info = cl->qdisc->handle;
 
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 	if (hfsc_dump_curves(skb, cl) < 0)
-		goto rtattr_failure;
-	rta->rta_len = skb_tail_pointer(skb) - b;
+		goto nla_put_failure;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
- rtattr_failure:
+ nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1423,15 +1423,15 @@ hfsc_schedule_watchdog(struct Qdisc *sch)
 }
 
 static int
-hfsc_init_qdisc(struct Qdisc *sch, struct rtattr *opt)
+hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct hfsc_sched *q = qdisc_priv(sch);
 	struct tc_hfsc_qopt *qopt;
 	unsigned int i;
 
-	if (opt == NULL || RTA_PAYLOAD(opt) < sizeof(*qopt))
+	if (opt == NULL || nla_len(opt) < sizeof(*qopt))
 		return -EINVAL;
-	qopt = RTA_DATA(opt);
+	qopt = nla_data(opt);
 
 	q->defcls = qopt->defcls;
 	for (i = 0; i < HFSC_HSIZE; i++)
@@ -1459,14 +1459,14 @@ hfsc_init_qdisc(struct Qdisc *sch, struct rtattr *opt)
 }
 
 static int
-hfsc_change_qdisc(struct Qdisc *sch, struct rtattr *opt)
+hfsc_change_qdisc(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct hfsc_sched *q = qdisc_priv(sch);
 	struct tc_hfsc_qopt *qopt;
 
-	if (opt == NULL || RTA_PAYLOAD(opt) < sizeof(*qopt))
+	if (opt == NULL || nla_len(opt) < sizeof(*qopt))
 		return -EINVAL;
-	qopt = RTA_DATA(opt);
+	qopt = nla_data(opt);
 
 	sch_tree_lock(sch);
 	q->defcls = qopt->defcls;
@@ -1550,10 +1550,10 @@ hfsc_dump_qdisc(struct Qdisc *sch, struct sk_buff *skb)
 	struct tc_hfsc_qopt qopt;
 
 	qopt.defcls = q->defcls;
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(qopt), &qopt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(qopt), &qopt);
 	return skb->len;
 
- rtattr_failure:
+ nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 6a2352cd9c2e..3b3ff641b6d7 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -992,19 +992,19 @@ static void htb_reset(struct Qdisc *sch)
 		INIT_LIST_HEAD(q->drops + i);
 }
 
-static int htb_init(struct Qdisc *sch, struct rtattr *opt)
+static int htb_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct htb_sched *q = qdisc_priv(sch);
-	struct rtattr *tb[TCA_HTB_INIT];
+	struct nlattr *tb[TCA_HTB_INIT + 1];
 	struct tc_htb_glob *gopt;
 	int i;
-	if (!opt || rtattr_parse_nested(tb, TCA_HTB_INIT, opt) ||
-	    tb[TCA_HTB_INIT - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_HTB_INIT - 1]) < sizeof(*gopt)) {
+	if (!opt || nla_parse_nested(tb, TCA_HTB_INIT, opt, NULL) ||
+	    tb[TCA_HTB_INIT] == NULL ||
+	    nla_len(tb[TCA_HTB_INIT]) < sizeof(*gopt)) {
 		printk(KERN_ERR "HTB: hey probably you have bad tc tool ?\n");
 		return -EINVAL;
 	}
-	gopt = RTA_DATA(tb[TCA_HTB_INIT - 1]);
+	gopt = nla_data(tb[TCA_HTB_INIT]);
 	if (gopt->version != HTB_VER >> 16) {
 		printk(KERN_ERR
 		       "HTB: need tc/htb version %d (minor is %d), you have %d\n",
@@ -1036,7 +1036,7 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct htb_sched *q = qdisc_priv(sch);
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 	struct tc_htb_glob gopt;
 	spin_lock_bh(&sch->dev->queue_lock);
 	gopt.direct_pkts = q->direct_pkts;
@@ -1045,13 +1045,13 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	gopt.rate2quantum = q->rate2quantum;
 	gopt.defcls = q->defcls;
 	gopt.debug = 0;
-	rta = (struct rtattr *)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
-	RTA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla = (struct nlattr *)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	NLA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	spin_unlock_bh(&sch->dev->queue_lock);
 	return skb->len;
-rtattr_failure:
+nla_put_failure:
 	spin_unlock_bh(&sch->dev->queue_lock);
 	nlmsg_trim(skb, skb_tail_pointer(skb));
 	return -1;
@@ -1062,7 +1062,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 {
 	struct htb_class *cl = (struct htb_class *)arg;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 	struct tc_htb_opt opt;
 
 	spin_lock_bh(&sch->dev->queue_lock);
@@ -1071,8 +1071,8 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	if (!cl->level && cl->un.leaf.q)
 		tcm->tcm_info = cl->un.leaf.q->handle;
 
-	rta = (struct rtattr *)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr *)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	memset(&opt, 0, sizeof(opt));
 
@@ -1083,11 +1083,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	opt.quantum = cl->un.leaf.quantum;
 	opt.prio = cl->un.leaf.prio;
 	opt.level = cl->level;
-	RTA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	NLA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	spin_unlock_bh(&sch->dev->queue_lock);
 	return skb->len;
-rtattr_failure:
+nla_put_failure:
 	spin_unlock_bh(&sch->dev->queue_lock);
 	nlmsg_trim(skb, b);
 	return -1;
@@ -1290,29 +1290,29 @@ static void htb_put(struct Qdisc *sch, unsigned long arg)
 }
 
 static int htb_change_class(struct Qdisc *sch, u32 classid,
-			    u32 parentid, struct rtattr **tca,
+			    u32 parentid, struct nlattr **tca,
 			    unsigned long *arg)
 {
 	int err = -EINVAL;
 	struct htb_sched *q = qdisc_priv(sch);
 	struct htb_class *cl = (struct htb_class *)*arg, *parent;
-	struct rtattr *opt = tca[TCA_OPTIONS - 1];
+	struct nlattr *opt = tca[TCA_OPTIONS];
 	struct qdisc_rate_table *rtab = NULL, *ctab = NULL;
-	struct rtattr *tb[TCA_HTB_RTAB];
+	struct nlattr *tb[TCA_HTB_RTAB + 1];
 	struct tc_htb_opt *hopt;
 
 	/* extract all subattrs from opt attr */
-	if (!opt || rtattr_parse_nested(tb, TCA_HTB_RTAB, opt) ||
-	    tb[TCA_HTB_PARMS - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_HTB_PARMS - 1]) < sizeof(*hopt))
+	if (!opt || nla_parse_nested(tb, TCA_HTB_RTAB, opt, NULL) ||
+	    tb[TCA_HTB_PARMS] == NULL ||
+	    nla_len(tb[TCA_HTB_PARMS]) < sizeof(*hopt))
 		goto failure;
 
 	parent = parentid == TC_H_ROOT ? NULL : htb_find(parentid, sch);
 
-	hopt = RTA_DATA(tb[TCA_HTB_PARMS - 1]);
+	hopt = nla_data(tb[TCA_HTB_PARMS]);
 
-	rtab = qdisc_get_rtab(&hopt->rate, tb[TCA_HTB_RTAB - 1]);
-	ctab = qdisc_get_rtab(&hopt->ceil, tb[TCA_HTB_CTAB - 1]);
+	rtab = qdisc_get_rtab(&hopt->rate, tb[TCA_HTB_RTAB]);
+	ctab = qdisc_get_rtab(&hopt->ceil, tb[TCA_HTB_CTAB]);
 	if (!rtab || !ctab)
 		goto failure;
 
@@ -1320,12 +1320,12 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		struct Qdisc *new_q;
 		int prio;
 		struct {
-			struct rtattr		rta;
+			struct nlattr		nla;
 			struct gnet_estimator	opt;
 		} est = {
-			.rta = {
-				.rta_len	= RTA_LENGTH(sizeof(est.opt)),
-				.rta_type	= TCA_RATE,
+			.nla = {
+				.nla_len	= nla_attr_size(sizeof(est.opt)),
+				.nla_type	= TCA_RATE,
 			},
 			.opt = {
 				/* 4s interval, 16s averaging constant */
@@ -1350,7 +1350,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
 				  &sch->dev->queue_lock,
-				  tca[TCA_RATE-1] ? : &est.rta);
+				  tca[TCA_RATE] ? : &est.nla);
 		cl->refcnt = 1;
 		INIT_LIST_HEAD(&cl->sibling);
 		INIT_HLIST_NODE(&cl->hlist);
@@ -1403,10 +1403,10 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		list_add_tail(&cl->sibling,
 			      parent ? &parent->children : &q->root);
 	} else {
-		if (tca[TCA_RATE-1])
+		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
 					      &sch->dev->queue_lock,
-					      tca[TCA_RATE-1]);
+					      tca[TCA_RATE]);
 		sch_tree_lock(sch);
 	}
 
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index 72525710b66f..f6decbb56645 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -57,7 +57,7 @@ static void ingress_put(struct Qdisc *sch, unsigned long cl)
 }
 
 static int ingress_change(struct Qdisc *sch, u32 classid, u32 parent,
-			  struct rtattr **tca, unsigned long *arg)
+			  struct nlattr **tca, unsigned long *arg)
 {
 	return 0;
 }
@@ -156,7 +156,7 @@ static struct nf_hook_ops ing_ops[] __read_mostly = {
 };
 #endif
 
-static int ingress_init(struct Qdisc *sch, struct rtattr *opt)
+static int ingress_init(struct Qdisc *sch, struct nlattr *opt)
 {
 #if !defined(CONFIG_NET_CLS_ACT) && defined(CONFIG_NETFILTER)
 	printk("Ingress scheduler: Classifier actions prefered over netfilter\n");
@@ -184,14 +184,14 @@ static void ingress_destroy(struct Qdisc *sch)
 static int ingress_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
-	rta = (struct rtattr *)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla = (struct nlattr *)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 6c344ade33c2..a7b58df4546d 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -313,21 +313,21 @@ static void netem_reset(struct Qdisc *sch)
 /* Pass size change message down to embedded FIFO */
 static int set_fifo_limit(struct Qdisc *q, int limit)
 {
-	struct rtattr *rta;
+	struct nlattr *nla;
 	int ret = -ENOMEM;
 
 	/* Hack to avoid sending change message to non-FIFO */
 	if (strncmp(q->ops->id + 1, "fifo", 4) != 0)
 		return 0;
 
-	rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
-	if (rta) {
-		rta->rta_type = RTM_NEWQDISC;
-		rta->rta_len = RTA_LENGTH(sizeof(struct tc_fifo_qopt));
-		((struct tc_fifo_qopt *)RTA_DATA(rta))->limit = limit;
+	nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
+	if (nla) {
+		nla->nla_type = RTM_NEWQDISC;
+		nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
+		((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
 
-		ret = q->ops->change(q, rta);
-		kfree(rta);
+		ret = q->ops->change(q, nla);
+		kfree(nla);
 	}
 	return ret;
 }
@@ -336,11 +336,11 @@ static int set_fifo_limit(struct Qdisc *q, int limit)
  * Distribution data is a variable size payload containing
  * signed 16 bit values.
  */
-static int get_dist_table(struct Qdisc *sch, const struct rtattr *attr)
+static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
-	unsigned long n = RTA_PAYLOAD(attr)/sizeof(__s16);
-	const __s16 *data = RTA_DATA(attr);
+	unsigned long n = nla_len(attr)/sizeof(__s16);
+	const __s16 *data = nla_data(attr);
 	struct disttable *d;
 	int i;
 
@@ -363,12 +363,12 @@ static int get_dist_table(struct Qdisc *sch, const struct rtattr *attr)
 	return 0;
 }
 
-static int get_correlation(struct Qdisc *sch, const struct rtattr *attr)
+static int get_correlation(struct Qdisc *sch, const struct nlattr *attr)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
-	const struct tc_netem_corr *c = RTA_DATA(attr);
+	const struct tc_netem_corr *c = nla_data(attr);
 
-	if (RTA_PAYLOAD(attr) != sizeof(*c))
+	if (nla_len(attr) != sizeof(*c))
 		return -EINVAL;
 
 	init_crandom(&q->delay_cor, c->delay_corr);
@@ -377,12 +377,12 @@ static int get_correlation(struct Qdisc *sch, const struct rtattr *attr)
 	return 0;
 }
 
-static int get_reorder(struct Qdisc *sch, const struct rtattr *attr)
+static int get_reorder(struct Qdisc *sch, const struct nlattr *attr)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
-	const struct tc_netem_reorder *r = RTA_DATA(attr);
+	const struct tc_netem_reorder *r = nla_data(attr);
 
-	if (RTA_PAYLOAD(attr) != sizeof(*r))
+	if (nla_len(attr) != sizeof(*r))
 		return -EINVAL;
 
 	q->reorder = r->probability;
@@ -390,12 +390,12 @@ static int get_reorder(struct Qdisc *sch, const struct rtattr *attr)
 	return 0;
 }
 
-static int get_corrupt(struct Qdisc *sch, const struct rtattr *attr)
+static int get_corrupt(struct Qdisc *sch, const struct nlattr *attr)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
-	const struct tc_netem_corrupt *r = RTA_DATA(attr);
+	const struct tc_netem_corrupt *r = nla_data(attr);
 
-	if (RTA_PAYLOAD(attr) != sizeof(*r))
+	if (nla_len(attr) != sizeof(*r))
 		return -EINVAL;
 
 	q->corrupt = r->probability;
@@ -404,16 +404,16 @@ static int get_corrupt(struct Qdisc *sch, const struct rtattr *attr)
 }
 
 /* Parse netlink message to set options */
-static int netem_change(struct Qdisc *sch, struct rtattr *opt)
+static int netem_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
 	struct tc_netem_qopt *qopt;
 	int ret;
 
-	if (opt == NULL || RTA_PAYLOAD(opt) < sizeof(*qopt))
+	if (opt == NULL || nla_len(opt) < sizeof(*qopt))
 		return -EINVAL;
 
-	qopt = RTA_DATA(opt);
+	qopt = nla_data(opt);
 	ret = set_fifo_limit(q->qdisc, qopt->limit);
 	if (ret) {
 		pr_debug("netem: can't set fifo limit\n");
@@ -437,33 +437,33 @@ static int netem_change(struct Qdisc *sch, struct rtattr *opt)
 	/* Handle nested options after initial queue options.
 	 * Should have put all options in nested format but too late now.
 	 */
-	if (RTA_PAYLOAD(opt) > sizeof(*qopt)) {
-		struct rtattr *tb[TCA_NETEM_MAX];
-		if (rtattr_parse(tb, TCA_NETEM_MAX,
-				 RTA_DATA(opt) + sizeof(*qopt),
-				 RTA_PAYLOAD(opt) - sizeof(*qopt)))
+	if (nla_len(opt) > sizeof(*qopt)) {
+		struct nlattr *tb[TCA_NETEM_MAX + 1];
+		if (nla_parse(tb, TCA_NETEM_MAX,
+			      nla_data(opt) + sizeof(*qopt),
+			      nla_len(opt) - sizeof(*qopt), NULL))
 			return -EINVAL;
 
-		if (tb[TCA_NETEM_CORR-1]) {
-			ret = get_correlation(sch, tb[TCA_NETEM_CORR-1]);
+		if (tb[TCA_NETEM_CORR]) {
+			ret = get_correlation(sch, tb[TCA_NETEM_CORR]);
 			if (ret)
 				return ret;
 		}
 
-		if (tb[TCA_NETEM_DELAY_DIST-1]) {
-			ret = get_dist_table(sch, tb[TCA_NETEM_DELAY_DIST-1]);
+		if (tb[TCA_NETEM_DELAY_DIST]) {
+			ret = get_dist_table(sch, tb[TCA_NETEM_DELAY_DIST]);
 			if (ret)
 				return ret;
 		}
 
-		if (tb[TCA_NETEM_REORDER-1]) {
-			ret = get_reorder(sch, tb[TCA_NETEM_REORDER-1]);
+		if (tb[TCA_NETEM_REORDER]) {
+			ret = get_reorder(sch, tb[TCA_NETEM_REORDER]);
 			if (ret)
 				return ret;
 		}
 
-		if (tb[TCA_NETEM_CORRUPT-1]) {
-			ret = get_corrupt(sch, tb[TCA_NETEM_CORRUPT-1]);
+		if (tb[TCA_NETEM_CORRUPT]) {
+			ret = get_corrupt(sch, tb[TCA_NETEM_CORRUPT]);
 			if (ret)
 				return ret;
 		}
@@ -515,13 +515,13 @@ static int tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch)
 	return qdisc_reshape_fail(nskb, sch);
 }
 
-static int tfifo_init(struct Qdisc *sch, struct rtattr *opt)
+static int tfifo_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct fifo_sched_data *q = qdisc_priv(sch);
 
 	if (opt) {
-		struct tc_fifo_qopt *ctl = RTA_DATA(opt);
-		if (RTA_PAYLOAD(opt) < sizeof(*ctl))
+		struct tc_fifo_qopt *ctl = nla_data(opt);
+		if (nla_len(opt) < sizeof(*ctl))
 			return -EINVAL;
 
 		q->limit = ctl->limit;
@@ -537,10 +537,10 @@ static int tfifo_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct fifo_sched_data *q = qdisc_priv(sch);
 	struct tc_fifo_qopt opt = { .limit = q->limit };
 
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
@@ -557,7 +557,7 @@ static struct Qdisc_ops tfifo_qdisc_ops __read_mostly = {
 	.dump		=	tfifo_dump,
 };
 
-static int netem_init(struct Qdisc *sch, struct rtattr *opt)
+static int netem_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
 	int ret;
@@ -595,7 +595,7 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	const struct netem_sched_data *q = qdisc_priv(sch);
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta = (struct rtattr *) b;
+	struct nlattr *nla = (struct nlattr *) b;
 	struct tc_netem_qopt qopt;
 	struct tc_netem_corr cor;
 	struct tc_netem_reorder reorder;
@@ -607,26 +607,26 @@ static int netem_dump(struct Qdisc *sch, struct sk_buff *skb)
 	qopt.loss = q->loss;
 	qopt.gap = q->gap;
 	qopt.duplicate = q->duplicate;
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(qopt), &qopt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(qopt), &qopt);
 
 	cor.delay_corr = q->delay_cor.rho;
 	cor.loss_corr = q->loss_cor.rho;
 	cor.dup_corr = q->dup_cor.rho;
-	RTA_PUT(skb, TCA_NETEM_CORR, sizeof(cor), &cor);
+	NLA_PUT(skb, TCA_NETEM_CORR, sizeof(cor), &cor);
 
 	reorder.probability = q->reorder;
 	reorder.correlation = q->reorder_cor.rho;
-	RTA_PUT(skb, TCA_NETEM_REORDER, sizeof(reorder), &reorder);
+	NLA_PUT(skb, TCA_NETEM_REORDER, sizeof(reorder), &reorder);
 
 	corrupt.probability = q->corrupt;
 	corrupt.correlation = q->corrupt_cor.rho;
-	RTA_PUT(skb, TCA_NETEM_CORRUPT, sizeof(corrupt), &corrupt);
+	NLA_PUT(skb, TCA_NETEM_CORRUPT, sizeof(corrupt), &corrupt);
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -678,7 +678,7 @@ static void netem_put(struct Qdisc *sch, unsigned long arg)
 }
 
 static int netem_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
-			    struct rtattr **tca, unsigned long *arg)
+			    struct nlattr **tca, unsigned long *arg)
 {
 	return -ENOSYS;
 }
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 2243aaa8d851..a4f932df86e9 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -224,15 +224,15 @@ prio_destroy(struct Qdisc* sch)
 		qdisc_destroy(q->queues[prio]);
 }
 
-static int prio_tune(struct Qdisc *sch, struct rtattr *opt)
+static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
 	struct tc_prio_qopt *qopt;
-	struct rtattr *tb[TCA_PRIO_MAX];
+	struct nlattr *tb[TCA_PRIO_MAX + 1];
 	int i;
 
-	if (rtattr_parse_nested_compat(tb, TCA_PRIO_MAX, opt, qopt,
-				       sizeof(*qopt)))
+	if (nla_parse_nested_compat(tb, TCA_PRIO_MAX, opt, NULL, qopt,
+				    sizeof(*qopt)))
 		return -EINVAL;
 	q->bands = qopt->bands;
 	/* If we're multiqueue, make sure the number of incoming bands
@@ -242,7 +242,7 @@ static int prio_tune(struct Qdisc *sch, struct rtattr *opt)
 	 * only one that is enabled for multiqueue, since it's the only one
 	 * that interacts with the underlying device.
 	 */
-	q->mq = RTA_GET_FLAG(tb[TCA_PRIO_MQ - 1]);
+	q->mq = nla_get_flag(tb[TCA_PRIO_MQ]);
 	if (q->mq) {
 		if (sch->parent != TC_H_ROOT)
 			return -EINVAL;
@@ -296,7 +296,7 @@ static int prio_tune(struct Qdisc *sch, struct rtattr *opt)
 	return 0;
 }
 
-static int prio_init(struct Qdisc *sch, struct rtattr *opt)
+static int prio_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
 	int i;
@@ -319,20 +319,24 @@ static int prio_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *nest;
+	struct nlattr *nest;
 	struct tc_prio_qopt opt;
 
 	opt.bands = q->bands;
 	memcpy(&opt.priomap, q->prio2band, TC_PRIO_MAX+1);
 
-	nest = RTA_NEST_COMPAT(skb, TCA_OPTIONS, sizeof(opt), &opt);
-	if (q->mq)
-		RTA_PUT_FLAG(skb, TCA_PRIO_MQ);
-	RTA_NEST_COMPAT_END(skb, nest);
+	nest = nla_nest_compat_start(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	if (nest == NULL)
+		goto nla_put_failure;
+	if (q->mq) {
+		if (nla_put_flag(skb, TCA_PRIO_MQ) < 0)
+			goto nla_put_failure;
+	}
+	nla_nest_compat_end(skb, nest);
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -392,7 +396,7 @@ static void prio_put(struct Qdisc *q, unsigned long cl)
 	return;
 }
 
-static int prio_change(struct Qdisc *sch, u32 handle, u32 parent, struct rtattr **tca, unsigned long *arg)
+static int prio_change(struct Qdisc *sch, u32 handle, u32 parent, struct nlattr **tca, unsigned long *arg)
 {
 	unsigned long cl = *arg;
 	struct prio_sched_data *q = qdisc_priv(sch);
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index f1e9647f7db7..6ce8da5aca0b 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -177,21 +177,21 @@ static void red_destroy(struct Qdisc *sch)
 static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit)
 {
 	struct Qdisc *q;
-	struct rtattr *rta;
+	struct nlattr *nla;
 	int ret;
 
 	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
 			      TC_H_MAKE(sch->handle, 1));
 	if (q) {
-		rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)),
+		nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)),
 			      GFP_KERNEL);
-		if (rta) {
-			rta->rta_type = RTM_NEWQDISC;
-			rta->rta_len = RTA_LENGTH(sizeof(struct tc_fifo_qopt));
-			((struct tc_fifo_qopt *)RTA_DATA(rta))->limit = limit;
+		if (nla) {
+			nla->nla_type = RTM_NEWQDISC;
+			nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
+			((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
 
-			ret = q->ops->change(q, rta);
-			kfree(rta);
+			ret = q->ops->change(q, nla);
+			kfree(nla);
 
 			if (ret == 0)
 				return q;
@@ -201,23 +201,23 @@ static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit)
 	return NULL;
 }
 
-static int red_change(struct Qdisc *sch, struct rtattr *opt)
+static int red_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct red_sched_data *q = qdisc_priv(sch);
-	struct rtattr *tb[TCA_RED_MAX];
+	struct nlattr *tb[TCA_RED_MAX + 1];
 	struct tc_red_qopt *ctl;
 	struct Qdisc *child = NULL;
 
-	if (opt == NULL || rtattr_parse_nested(tb, TCA_RED_MAX, opt))
+	if (opt == NULL || nla_parse_nested(tb, TCA_RED_MAX, opt, NULL))
 		return -EINVAL;
 
-	if (tb[TCA_RED_PARMS-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_RED_PARMS-1]) < sizeof(*ctl) ||
-	    tb[TCA_RED_STAB-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_RED_STAB-1]) < RED_STAB_SIZE)
+	if (tb[TCA_RED_PARMS] == NULL ||
+	    nla_len(tb[TCA_RED_PARMS]) < sizeof(*ctl) ||
+	    tb[TCA_RED_STAB] == NULL ||
+	    nla_len(tb[TCA_RED_STAB]) < RED_STAB_SIZE)
 		return -EINVAL;
 
-	ctl = RTA_DATA(tb[TCA_RED_PARMS-1]);
+	ctl = nla_data(tb[TCA_RED_PARMS]);
 
 	if (ctl->limit > 0) {
 		child = red_create_dflt(sch, ctl->limit);
@@ -235,7 +235,7 @@ static int red_change(struct Qdisc *sch, struct rtattr *opt)
 
 	red_set_parms(&q->parms, ctl->qth_min, ctl->qth_max, ctl->Wlog,
 				 ctl->Plog, ctl->Scell_log,
-				 RTA_DATA(tb[TCA_RED_STAB-1]));
+				 nla_data(tb[TCA_RED_STAB]));
 
 	if (skb_queue_empty(&sch->q))
 		red_end_of_idle_period(&q->parms);
@@ -244,7 +244,7 @@ static int red_change(struct Qdisc *sch, struct rtattr *opt)
 	return 0;
 }
 
-static int red_init(struct Qdisc* sch, struct rtattr *opt)
+static int red_init(struct Qdisc* sch, struct nlattr *opt)
 {
 	struct red_sched_data *q = qdisc_priv(sch);
 
@@ -255,7 +255,7 @@ static int red_init(struct Qdisc* sch, struct rtattr *opt)
 static int red_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct red_sched_data *q = qdisc_priv(sch);
-	struct rtattr *opts = NULL;
+	struct nlattr *opts = NULL;
 	struct tc_red_qopt opt = {
 		.limit		= q->limit,
 		.flags		= q->flags,
@@ -266,12 +266,14 @@ static int red_dump(struct Qdisc *sch, struct sk_buff *skb)
 		.Scell_log	= q->parms.Scell_log,
 	};
 
-	opts = RTA_NEST(skb, TCA_OPTIONS);
-	RTA_PUT(skb, TCA_RED_PARMS, sizeof(opt), &opt);
-	return RTA_NEST_END(skb, opts);
+	opts = nla_nest_start(skb, TCA_OPTIONS);
+	if (opts == NULL)
+		goto nla_put_failure;
+	NLA_PUT(skb, TCA_RED_PARMS, sizeof(opt), &opt);
+	return nla_nest_end(skb, opts);
 
-rtattr_failure:
-	return RTA_NEST_CANCEL(skb, opts);
+nla_put_failure:
+	return nla_nest_cancel(skb, opts);
 }
 
 static int red_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
@@ -332,7 +334,7 @@ static void red_put(struct Qdisc *sch, unsigned long arg)
 }
 
 static int red_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
-			    struct rtattr **tca, unsigned long *arg)
+			    struct nlattr **tca, unsigned long *arg)
 {
 	return -ENOSYS;
 }
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 25afe0f1d83a..91af539ab6e6 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -397,13 +397,13 @@ static void sfq_perturbation(unsigned long arg)
 		mod_timer(&q->perturb_timer, jiffies + q->perturb_period);
 }
 
-static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
+static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct sfq_sched_data *q = qdisc_priv(sch);
-	struct tc_sfq_qopt *ctl = RTA_DATA(opt);
+	struct tc_sfq_qopt *ctl = nla_data(opt);
 	unsigned int qlen;
 
-	if (opt->rta_len < RTA_LENGTH(sizeof(*ctl)))
+	if (opt->nla_len < nla_attr_size(sizeof(*ctl)))
 		return -EINVAL;
 
 	sch_tree_lock(sch);
@@ -426,7 +426,7 @@ static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
 	return 0;
 }
 
-static int sfq_init(struct Qdisc *sch, struct rtattr *opt)
+static int sfq_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct sfq_sched_data *q = qdisc_priv(sch);
 	int i;
@@ -481,11 +481,11 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
 	opt.divisor = SFQ_HASH_DIVISOR;
 	opt.flows = q->limit;
 
-	RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index d88fea9d6b61..6c4ad7e677b3 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -245,20 +245,21 @@ static void tbf_reset(struct Qdisc* sch)
 static struct Qdisc *tbf_create_dflt_qdisc(struct Qdisc *sch, u32 limit)
 {
 	struct Qdisc *q;
-	struct rtattr *rta;
+	struct nlattr *nla;
 	int ret;
 
 	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
 			      TC_H_MAKE(sch->handle, 1));
 	if (q) {
-		rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
-		if (rta) {
-			rta->rta_type = RTM_NEWQDISC;
-			rta->rta_len = RTA_LENGTH(sizeof(struct tc_fifo_qopt));
-			((struct tc_fifo_qopt *)RTA_DATA(rta))->limit = limit;
+		nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)),
+			      GFP_KERNEL);
+		if (nla) {
+			nla->nla_type = RTM_NEWQDISC;
+			nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
+			((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
 
-			ret = q->ops->change(q, rta);
-			kfree(rta);
+			ret = q->ops->change(q, nla);
+			kfree(nla);
 
 			if (ret == 0)
 				return q;
@@ -269,30 +270,30 @@ static struct Qdisc *tbf_create_dflt_qdisc(struct Qdisc *sch, u32 limit)
 	return NULL;
 }
 
-static int tbf_change(struct Qdisc* sch, struct rtattr *opt)
+static int tbf_change(struct Qdisc* sch, struct nlattr *opt)
 {
 	int err = -EINVAL;
 	struct tbf_sched_data *q = qdisc_priv(sch);
-	struct rtattr *tb[TCA_TBF_PTAB];
+	struct nlattr *tb[TCA_TBF_PTAB + 1];
 	struct tc_tbf_qopt *qopt;
 	struct qdisc_rate_table *rtab = NULL;
 	struct qdisc_rate_table *ptab = NULL;
 	struct Qdisc *child = NULL;
 	int max_size,n;
 
-	if (rtattr_parse_nested(tb, TCA_TBF_PTAB, opt) ||
-	    tb[TCA_TBF_PARMS-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_TBF_PARMS-1]) < sizeof(*qopt))
+	if (nla_parse_nested(tb, TCA_TBF_PTAB, opt, NULL) ||
+	    tb[TCA_TBF_PARMS] == NULL ||
+	    nla_len(tb[TCA_TBF_PARMS]) < sizeof(*qopt))
 		goto done;
 
-	qopt = RTA_DATA(tb[TCA_TBF_PARMS-1]);
-	rtab = qdisc_get_rtab(&qopt->rate, tb[TCA_TBF_RTAB-1]);
+	qopt = nla_data(tb[TCA_TBF_PARMS]);
+	rtab = qdisc_get_rtab(&qopt->rate, tb[TCA_TBF_RTAB]);
 	if (rtab == NULL)
 		goto done;
 
 	if (qopt->peakrate.rate) {
 		if (qopt->peakrate.rate > qopt->rate.rate)
-			ptab = qdisc_get_rtab(&qopt->peakrate, tb[TCA_TBF_PTAB-1]);
+			ptab = qdisc_get_rtab(&qopt->peakrate, tb[TCA_TBF_PTAB]);
 		if (ptab == NULL)
 			goto done;
 	}
@@ -339,7 +340,7 @@ done:
 	return err;
 }
 
-static int tbf_init(struct Qdisc* sch, struct rtattr *opt)
+static int tbf_init(struct Qdisc* sch, struct nlattr *opt)
 {
 	struct tbf_sched_data *q = qdisc_priv(sch);
 
@@ -371,11 +372,11 @@ static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 	struct tc_tbf_qopt opt;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	opt.limit = q->limit;
 	opt.rate = q->R_tab->rate;
@@ -385,12 +386,12 @@ static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
 		memset(&opt.peakrate, 0, sizeof(opt.peakrate));
 	opt.mtu = q->mtu;
 	opt.buffer = q->buffer;
-	RTA_PUT(skb, TCA_TBF_PARMS, sizeof(opt), &opt);
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	NLA_PUT(skb, TCA_TBF_PARMS, sizeof(opt), &opt);
+	nla->nla_len = skb_tail_pointer(skb) - b;
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -442,7 +443,7 @@ static void tbf_put(struct Qdisc *sch, unsigned long arg)
 }
 
 static int tbf_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
-			    struct rtattr **tca, unsigned long *arg)
+			    struct nlattr **tca, unsigned long *arg)
 {
 	return -ENOSYS;
 }
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index c0ed06d4a504..1411c7b1fbdc 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -168,7 +168,7 @@ teql_destroy(struct Qdisc* sch)
 	}
 }
 
-static int teql_qdisc_init(struct Qdisc *sch, struct rtattr *opt)
+static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct net_device *dev = sch->dev;
 	struct teql_master *m = (struct teql_master*)sch->ops;
-- 
cgit v1.2.3


From add93b610a4e66d36d0cf0b2596c3d3bcfdaee39 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Jan 2008 22:11:33 -0800
Subject: [NET_SCHED]: Convert classifiers from rtnetlink to new netlink API

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_cls.h     | 10 +++---
 include/net/sch_generic.h |  2 +-
 net/sched/cls_api.c       | 65 +++++++++++++++++++++------------------
 net/sched/cls_basic.c     | 40 ++++++++++++------------
 net/sched/cls_fw.c        | 54 ++++++++++++++++----------------
 net/sched/cls_route.c     | 70 +++++++++++++++++++++---------------------
 net/sched/cls_rsvp.h      | 48 ++++++++++++++---------------
 net/sched/cls_tcindex.c   | 66 +++++++++++++++++++--------------------
 net/sched/cls_u32.c       | 78 +++++++++++++++++++++++------------------------
 net/sched/em_meta.c       | 56 +++++++++++++++++-----------------
 net/sched/em_text.c       |  9 ++++--
 net/sched/ematch.c        | 74 ++++++++++++++++++++++----------------------
 12 files changed, 290 insertions(+), 282 deletions(-)

(limited to 'include')

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 2eaf20480024..8716eb757d51 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -129,8 +129,8 @@ tcf_exts_exec(struct sk_buff *skb, struct tcf_exts *exts,
 	return 0;
 }
 
-extern int tcf_exts_validate(struct tcf_proto *tp, struct rtattr **tb,
-	                     struct rtattr *rate_tlv, struct tcf_exts *exts,
+extern int tcf_exts_validate(struct tcf_proto *tp, struct nlattr **tb,
+	                     struct nlattr *rate_tlv, struct tcf_exts *exts,
 	                     struct tcf_ext_map *map);
 extern void tcf_exts_destroy(struct tcf_proto *tp, struct tcf_exts *exts);
 extern void tcf_exts_change(struct tcf_proto *tp, struct tcf_exts *dst,
@@ -247,7 +247,7 @@ struct tcf_ematch_ops
 
 extern int tcf_em_register(struct tcf_ematch_ops *);
 extern int tcf_em_unregister(struct tcf_ematch_ops *);
-extern int tcf_em_tree_validate(struct tcf_proto *, struct rtattr *,
+extern int tcf_em_tree_validate(struct tcf_proto *, struct nlattr *,
 				struct tcf_ematch_tree *);
 extern void tcf_em_tree_destroy(struct tcf_proto *, struct tcf_ematch_tree *);
 extern int tcf_em_tree_dump(struct sk_buff *, struct tcf_ematch_tree *, int);
@@ -338,9 +338,9 @@ static inline int tcf_valid_offset(const struct sk_buff *skb,
 #include <net/net_namespace.h>
 
 static inline int
-tcf_change_indev(struct tcf_proto *tp, char *indev, struct rtattr *indev_tlv)
+tcf_change_indev(struct tcf_proto *tp, char *indev, struct nlattr *indev_tlv)
 {
-	if (rtattr_strlcpy(indev, indev_tlv, IFNAMSIZ) >= IFNAMSIZ)
+	if (nla_strlcpy(indev, indev_tlv, IFNAMSIZ) >= IFNAMSIZ)
 		return -EINVAL;
 	return 0;
 }
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8cacdff24594..ab502ec1c61c 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -126,7 +126,7 @@ struct tcf_proto_ops
 	unsigned long		(*get)(struct tcf_proto*, u32 handle);
 	void			(*put)(struct tcf_proto*, unsigned long);
 	int			(*change)(struct tcf_proto*, unsigned long,
-					u32 handle, struct rtattr **,
+					u32 handle, struct nlattr **,
 					unsigned long *);
 	int			(*delete)(struct tcf_proto*, unsigned long);
 	void			(*walk)(struct tcf_proto*, struct tcf_walker *arg);
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 9eeb3c6c82f4..87be2b2fc29a 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -38,14 +38,14 @@ static DEFINE_RWLOCK(cls_mod_lock);
 
 /* Find classifier type by string name */
 
-static struct tcf_proto_ops *tcf_proto_lookup_ops(struct rtattr *kind)
+static struct tcf_proto_ops *tcf_proto_lookup_ops(struct nlattr *kind)
 {
 	struct tcf_proto_ops *t = NULL;
 
 	if (kind) {
 		read_lock(&cls_mod_lock);
 		for (t = tcf_proto_base; t; t = t->next) {
-			if (rtattr_strcmp(kind, t->kind) == 0) {
+			if (nla_strcmp(kind, t->kind) == 0) {
 				if (!try_module_get(t->owner))
 					t = NULL;
 				break;
@@ -118,7 +118,7 @@ static inline u32 tcf_auto_prio(struct tcf_proto *tp)
 static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = skb->sk->sk_net;
-	struct rtattr **tca;
+	struct nlattr *tca[TCA_MAX + 1];
 	struct tcmsg *t;
 	u32 protocol;
 	u32 prio;
@@ -138,7 +138,6 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 		return -EINVAL;
 
 replay:
-	tca = arg;
 	t = NLMSG_DATA(n);
 	protocol = TC_H_MIN(t->tcm_info);
 	prio = TC_H_MAJ(t->tcm_info);
@@ -160,6 +159,10 @@ replay:
 	if (dev == NULL)
 		return -ENODEV;
 
+	err = nlmsg_parse(n, sizeof(*t), tca, TCA_MAX, NULL);
+	if (err < 0)
+		return err;
+
 	/* Find qdisc */
 	if (!parent) {
 		q = dev->qdisc_sleeping;
@@ -202,7 +205,7 @@ replay:
 	if (tp == NULL) {
 		/* Proto-tcf does not exist, create new one */
 
-		if (tca[TCA_KIND-1] == NULL || !protocol)
+		if (tca[TCA_KIND] == NULL || !protocol)
 			goto errout;
 
 		err = -ENOENT;
@@ -217,14 +220,14 @@ replay:
 		if (tp == NULL)
 			goto errout;
 		err = -EINVAL;
-		tp_ops = tcf_proto_lookup_ops(tca[TCA_KIND-1]);
+		tp_ops = tcf_proto_lookup_ops(tca[TCA_KIND]);
 		if (tp_ops == NULL) {
 #ifdef CONFIG_KMOD
-			struct rtattr *kind = tca[TCA_KIND-1];
+			struct nlattr *kind = tca[TCA_KIND];
 			char name[IFNAMSIZ];
 
 			if (kind != NULL &&
-			    rtattr_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) {
+			    nla_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) {
 				rtnl_unlock();
 				request_module("cls_%s", name);
 				rtnl_lock();
@@ -263,7 +266,7 @@ replay:
 		*back = tp;
 		qdisc_unlock_tree(dev);
 
-	} else if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], tp->ops->kind))
+	} else if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], tp->ops->kind))
 		goto errout;
 
 	fh = tp->ops->get(tp, t->tcm_handle);
@@ -333,18 +336,18 @@ static int tcf_fill_node(struct sk_buff *skb, struct tcf_proto *tp,
 	tcm->tcm_ifindex = tp->q->dev->ifindex;
 	tcm->tcm_parent = tp->classid;
 	tcm->tcm_info = TC_H_MAKE(tp->prio, tp->protocol);
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, tp->ops->kind);
+	NLA_PUT(skb, TCA_KIND, IFNAMSIZ, tp->ops->kind);
 	tcm->tcm_handle = fh;
 	if (RTM_DELTFILTER != event) {
 		tcm->tcm_handle = 0;
 		if (tp->ops->dump && tp->ops->dump(tp, fh, skb, tcm) < 0)
-			goto rtattr_failure;
+			goto nla_put_failure;
 	}
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
 nlmsg_failure:
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -476,8 +479,8 @@ void tcf_exts_destroy(struct tcf_proto *tp, struct tcf_exts *exts)
 }
 EXPORT_SYMBOL(tcf_exts_destroy);
 
-int tcf_exts_validate(struct tcf_proto *tp, struct rtattr **tb,
-		  struct rtattr *rate_tlv, struct tcf_exts *exts,
+int tcf_exts_validate(struct tcf_proto *tp, struct nlattr **tb,
+		  struct nlattr *rate_tlv, struct tcf_exts *exts,
 		  struct tcf_ext_map *map)
 {
 	memset(exts, 0, sizeof(*exts));
@@ -487,8 +490,9 @@ int tcf_exts_validate(struct tcf_proto *tp, struct rtattr **tb,
 		int err;
 		struct tc_action *act;
 
-		if (map->police && tb[map->police-1]) {
-			act = tcf_action_init_1(tb[map->police-1], rate_tlv,
+		if (map->police && tb[map->police]) {
+			act = tcf_action_init_1((struct rtattr *)tb[map->police],
+						(struct rtattr *)rate_tlv,
 						"police", TCA_ACT_NOREPLACE,
 						TCA_ACT_BIND, &err);
 			if (act == NULL)
@@ -496,8 +500,9 @@ int tcf_exts_validate(struct tcf_proto *tp, struct rtattr **tb,
 
 			act->type = TCA_OLD_COMPAT;
 			exts->action = act;
-		} else if (map->action && tb[map->action-1]) {
-			act = tcf_action_init(tb[map->action-1], rate_tlv, NULL,
+		} else if (map->action && tb[map->action]) {
+			act = tcf_action_init((struct rtattr *)tb[map->action],
+					      (struct rtattr *)rate_tlv, NULL,
 				TCA_ACT_NOREPLACE, TCA_ACT_BIND, &err);
 			if (act == NULL)
 				return err;
@@ -506,8 +511,8 @@ int tcf_exts_validate(struct tcf_proto *tp, struct rtattr **tb,
 		}
 	}
 #else
-	if ((map->action && tb[map->action-1]) ||
-	    (map->police && tb[map->police-1]))
+	if ((map->action && tb[map->action]) ||
+	    (map->police && tb[map->police]))
 		return -EOPNOTSUPP;
 #endif
 
@@ -541,23 +546,23 @@ int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts,
 		 * to work with both old and new modes of entering
 		 * tc data even if iproute2  was newer - jhs
 		 */
-		struct rtattr *p_rta = (struct rtattr *)skb_tail_pointer(skb);
+		struct nlattr *p_rta = (struct nlattr *)skb_tail_pointer(skb);
 
 		if (exts->action->type != TCA_OLD_COMPAT) {
-			RTA_PUT(skb, map->action, 0, NULL);
+			NLA_PUT(skb, map->action, 0, NULL);
 			if (tcf_action_dump(skb, exts->action, 0, 0) < 0)
-				goto rtattr_failure;
-			p_rta->rta_len = skb_tail_pointer(skb) - (u8 *)p_rta;
+				goto nla_put_failure;
+			p_rta->nla_len = skb_tail_pointer(skb) - (u8 *)p_rta;
 		} else if (map->police) {
-			RTA_PUT(skb, map->police, 0, NULL);
+			NLA_PUT(skb, map->police, 0, NULL);
 			if (tcf_action_dump_old(skb, exts->action, 0, 0) < 0)
-				goto rtattr_failure;
-			p_rta->rta_len = skb_tail_pointer(skb) - (u8 *)p_rta;
+				goto nla_put_failure;
+			p_rta->nla_len = skb_tail_pointer(skb) - (u8 *)p_rta;
 		}
 	}
 #endif
 	return 0;
-rtattr_failure: __attribute__ ((unused))
+nla_put_failure: __attribute__ ((unused))
 	return -1;
 }
 EXPORT_SYMBOL(tcf_exts_dump);
@@ -569,10 +574,10 @@ int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts,
 #ifdef CONFIG_NET_CLS_ACT
 	if (exts->action)
 		if (tcf_action_copy_stats(skb, exts->action, 1) < 0)
-			goto rtattr_failure;
+			goto nla_put_failure;
 #endif
 	return 0;
-rtattr_failure: __attribute__ ((unused))
+nla_put_failure: __attribute__ ((unused))
 	return -1;
 }
 EXPORT_SYMBOL(tcf_exts_dump_stats);
diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c
index b31f9f971987..3953da33956f 100644
--- a/net/sched/cls_basic.c
+++ b/net/sched/cls_basic.c
@@ -130,27 +130,27 @@ static int basic_delete(struct tcf_proto *tp, unsigned long arg)
 }
 
 static inline int basic_set_parms(struct tcf_proto *tp, struct basic_filter *f,
-				  unsigned long base, struct rtattr **tb,
-				  struct rtattr *est)
+				  unsigned long base, struct nlattr **tb,
+				  struct nlattr *est)
 {
 	int err = -EINVAL;
 	struct tcf_exts e;
 	struct tcf_ematch_tree t;
 
-	if (tb[TCA_BASIC_CLASSID-1])
-		if (RTA_PAYLOAD(tb[TCA_BASIC_CLASSID-1]) < sizeof(u32))
+	if (tb[TCA_BASIC_CLASSID])
+		if (nla_len(tb[TCA_BASIC_CLASSID]) < sizeof(u32))
 			return err;
 
 	err = tcf_exts_validate(tp, tb, est, &e, &basic_ext_map);
 	if (err < 0)
 		return err;
 
-	err = tcf_em_tree_validate(tp, tb[TCA_BASIC_EMATCHES-1], &t);
+	err = tcf_em_tree_validate(tp, tb[TCA_BASIC_EMATCHES], &t);
 	if (err < 0)
 		goto errout;
 
-	if (tb[TCA_BASIC_CLASSID-1]) {
-		f->res.classid = *(u32*)RTA_DATA(tb[TCA_BASIC_CLASSID-1]);
+	if (tb[TCA_BASIC_CLASSID]) {
+		f->res.classid = *(u32*)nla_data(tb[TCA_BASIC_CLASSID]);
 		tcf_bind_filter(tp, &f->res, base);
 	}
 
@@ -164,23 +164,23 @@ errout:
 }
 
 static int basic_change(struct tcf_proto *tp, unsigned long base, u32 handle,
-			struct rtattr **tca, unsigned long *arg)
+			struct nlattr **tca, unsigned long *arg)
 {
 	int err = -EINVAL;
 	struct basic_head *head = (struct basic_head *) tp->root;
-	struct rtattr *tb[TCA_BASIC_MAX];
+	struct nlattr *tb[TCA_BASIC_MAX + 1];
 	struct basic_filter *f = (struct basic_filter *) *arg;
 
-	if (tca[TCA_OPTIONS-1] == NULL)
+	if (tca[TCA_OPTIONS] == NULL)
 		return -EINVAL;
 
-	if (rtattr_parse_nested(tb, TCA_BASIC_MAX, tca[TCA_OPTIONS-1]) < 0)
+	if (nla_parse_nested(tb, TCA_BASIC_MAX, tca[TCA_OPTIONS], NULL) < 0)
 		return -EINVAL;
 
 	if (f != NULL) {
 		if (handle && f->handle != handle)
 			return -EINVAL;
-		return basic_set_parms(tp, f, base, tb, tca[TCA_RATE-1]);
+		return basic_set_parms(tp, f, base, tb, tca[TCA_RATE]);
 	}
 
 	err = -ENOBUFS;
@@ -206,7 +206,7 @@ static int basic_change(struct tcf_proto *tp, unsigned long base, u32 handle,
 		f->handle = head->hgenerator;
 	}
 
-	err = basic_set_parms(tp, f, base, tb, tca[TCA_RATE-1]);
+	err = basic_set_parms(tp, f, base, tb, tca[TCA_RATE]);
 	if (err < 0)
 		goto errout;
 
@@ -246,27 +246,27 @@ static int basic_dump(struct tcf_proto *tp, unsigned long fh,
 {
 	struct basic_filter *f = (struct basic_filter *) fh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	if (f == NULL)
 		return skb->len;
 
 	t->tcm_handle = f->handle;
 
-	rta = (struct rtattr *) b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr *) b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	if (f->res.classid)
-		RTA_PUT(skb, TCA_BASIC_CLASSID, sizeof(u32), &f->res.classid);
+		NLA_PUT(skb, TCA_BASIC_CLASSID, sizeof(u32), &f->res.classid);
 
 	if (tcf_exts_dump(skb, &f->exts, &basic_ext_map) < 0 ||
 	    tcf_em_tree_dump(skb, &f->ematches, TCA_BASIC_EMATCHES) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c
index b45038770e79..db6e90a37846 100644
--- a/net/sched/cls_fw.c
+++ b/net/sched/cls_fw.c
@@ -188,37 +188,37 @@ out:
 
 static int
 fw_change_attrs(struct tcf_proto *tp, struct fw_filter *f,
-	struct rtattr **tb, struct rtattr **tca, unsigned long base)
+	struct nlattr **tb, struct nlattr **tca, unsigned long base)
 {
 	struct fw_head *head = (struct fw_head *)tp->root;
 	struct tcf_exts e;
 	u32 mask;
 	int err;
 
-	err = tcf_exts_validate(tp, tb, tca[TCA_RATE-1], &e, &fw_ext_map);
+	err = tcf_exts_validate(tp, tb, tca[TCA_RATE], &e, &fw_ext_map);
 	if (err < 0)
 		return err;
 
 	err = -EINVAL;
-	if (tb[TCA_FW_CLASSID-1]) {
-		if (RTA_PAYLOAD(tb[TCA_FW_CLASSID-1]) != sizeof(u32))
+	if (tb[TCA_FW_CLASSID]) {
+		if (nla_len(tb[TCA_FW_CLASSID]) != sizeof(u32))
 			goto errout;
-		f->res.classid = *(u32*)RTA_DATA(tb[TCA_FW_CLASSID-1]);
+		f->res.classid = *(u32*)nla_data(tb[TCA_FW_CLASSID]);
 		tcf_bind_filter(tp, &f->res, base);
 	}
 
 #ifdef CONFIG_NET_CLS_IND
-	if (tb[TCA_FW_INDEV-1]) {
-		err = tcf_change_indev(tp, f->indev, tb[TCA_FW_INDEV-1]);
+	if (tb[TCA_FW_INDEV]) {
+		err = tcf_change_indev(tp, f->indev, tb[TCA_FW_INDEV]);
 		if (err < 0)
 			goto errout;
 	}
 #endif /* CONFIG_NET_CLS_IND */
 
-	if (tb[TCA_FW_MASK-1]) {
-		if (RTA_PAYLOAD(tb[TCA_FW_MASK-1]) != sizeof(u32))
+	if (tb[TCA_FW_MASK]) {
+		if (nla_len(tb[TCA_FW_MASK]) != sizeof(u32))
 			goto errout;
-		mask = *(u32*)RTA_DATA(tb[TCA_FW_MASK-1]);
+		mask = *(u32*)nla_data(tb[TCA_FW_MASK]);
 		if (mask != head->mask)
 			goto errout;
 	} else if (head->mask != 0xFFFFFFFF)
@@ -234,19 +234,19 @@ errout:
 
 static int fw_change(struct tcf_proto *tp, unsigned long base,
 		     u32 handle,
-		     struct rtattr **tca,
+		     struct nlattr **tca,
 		     unsigned long *arg)
 {
 	struct fw_head *head = (struct fw_head*)tp->root;
 	struct fw_filter *f = (struct fw_filter *) *arg;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_FW_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_FW_MAX + 1];
 	int err;
 
 	if (!opt)
 		return handle ? -EINVAL : 0;
 
-	if (rtattr_parse_nested(tb, TCA_FW_MAX, opt) < 0)
+	if (nla_parse_nested(tb, TCA_FW_MAX, opt, NULL) < 0)
 		return -EINVAL;
 
 	if (f != NULL) {
@@ -260,10 +260,10 @@ static int fw_change(struct tcf_proto *tp, unsigned long base,
 
 	if (head == NULL) {
 		u32 mask = 0xFFFFFFFF;
-		if (tb[TCA_FW_MASK-1]) {
-			if (RTA_PAYLOAD(tb[TCA_FW_MASK-1]) != sizeof(u32))
+		if (tb[TCA_FW_MASK]) {
+			if (nla_len(tb[TCA_FW_MASK]) != sizeof(u32))
 				return -EINVAL;
-			mask = *(u32*)RTA_DATA(tb[TCA_FW_MASK-1]);
+			mask = *(u32*)nla_data(tb[TCA_FW_MASK]);
 		}
 
 		head = kzalloc(sizeof(struct fw_head), GFP_KERNEL);
@@ -333,7 +333,7 @@ static int fw_dump(struct tcf_proto *tp, unsigned long fh,
 	struct fw_head *head = (struct fw_head *)tp->root;
 	struct fw_filter *f = (struct fw_filter*)fh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	if (f == NULL)
 		return skb->len;
@@ -343,29 +343,29 @@ static int fw_dump(struct tcf_proto *tp, unsigned long fh,
 	if (!f->res.classid && !tcf_exts_is_available(&f->exts))
 		return skb->len;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	if (f->res.classid)
-		RTA_PUT(skb, TCA_FW_CLASSID, 4, &f->res.classid);
+		NLA_PUT(skb, TCA_FW_CLASSID, 4, &f->res.classid);
 #ifdef CONFIG_NET_CLS_IND
 	if (strlen(f->indev))
-		RTA_PUT(skb, TCA_FW_INDEV, IFNAMSIZ, f->indev);
+		NLA_PUT(skb, TCA_FW_INDEV, IFNAMSIZ, f->indev);
 #endif /* CONFIG_NET_CLS_IND */
 	if (head->mask != 0xFFFFFFFF)
-		RTA_PUT(skb, TCA_FW_MASK, 4, &head->mask);
+		NLA_PUT(skb, TCA_FW_MASK, 4, &head->mask);
 
 	if (tcf_exts_dump(skb, &f->exts, &fw_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 
 	if (tcf_exts_dump_stats(skb, &f->exts, &fw_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c
index e70edd0f7bc8..b1aae84cbadc 100644
--- a/net/sched/cls_route.c
+++ b/net/sched/cls_route.c
@@ -325,7 +325,7 @@ static int route4_delete(struct tcf_proto *tp, unsigned long arg)
 
 static int route4_set_parms(struct tcf_proto *tp, unsigned long base,
 	struct route4_filter *f, u32 handle, struct route4_head *head,
-	struct rtattr **tb, struct rtattr *est, int new)
+	struct nlattr **tb, struct nlattr *est, int new)
 {
 	int err;
 	u32 id = 0, to = 0, nhandle = 0x8000;
@@ -339,34 +339,34 @@ static int route4_set_parms(struct tcf_proto *tp, unsigned long base,
 		return err;
 
 	err = -EINVAL;
-	if (tb[TCA_ROUTE4_CLASSID-1])
-		if (RTA_PAYLOAD(tb[TCA_ROUTE4_CLASSID-1]) < sizeof(u32))
+	if (tb[TCA_ROUTE4_CLASSID])
+		if (nla_len(tb[TCA_ROUTE4_CLASSID]) < sizeof(u32))
 			goto errout;
 
-	if (tb[TCA_ROUTE4_TO-1]) {
+	if (tb[TCA_ROUTE4_TO]) {
 		if (new && handle & 0x8000)
 			goto errout;
-		if (RTA_PAYLOAD(tb[TCA_ROUTE4_TO-1]) < sizeof(u32))
+		if (nla_len(tb[TCA_ROUTE4_TO]) < sizeof(u32))
 			goto errout;
-		to = *(u32*)RTA_DATA(tb[TCA_ROUTE4_TO-1]);
+		to = *(u32*)nla_data(tb[TCA_ROUTE4_TO]);
 		if (to > 0xFF)
 			goto errout;
 		nhandle = to;
 	}
 
-	if (tb[TCA_ROUTE4_FROM-1]) {
-		if (tb[TCA_ROUTE4_IIF-1])
+	if (tb[TCA_ROUTE4_FROM]) {
+		if (tb[TCA_ROUTE4_IIF])
 			goto errout;
-		if (RTA_PAYLOAD(tb[TCA_ROUTE4_FROM-1]) < sizeof(u32))
+		if (nla_len(tb[TCA_ROUTE4_FROM]) < sizeof(u32))
 			goto errout;
-		id = *(u32*)RTA_DATA(tb[TCA_ROUTE4_FROM-1]);
+		id = *(u32*)nla_data(tb[TCA_ROUTE4_FROM]);
 		if (id > 0xFF)
 			goto errout;
 		nhandle |= id << 16;
-	} else if (tb[TCA_ROUTE4_IIF-1]) {
-		if (RTA_PAYLOAD(tb[TCA_ROUTE4_IIF-1]) < sizeof(u32))
+	} else if (tb[TCA_ROUTE4_IIF]) {
+		if (nla_len(tb[TCA_ROUTE4_IIF]) < sizeof(u32))
 			goto errout;
-		id = *(u32*)RTA_DATA(tb[TCA_ROUTE4_IIF-1]);
+		id = *(u32*)nla_data(tb[TCA_ROUTE4_IIF]);
 		if (id > 0x7FFF)
 			goto errout;
 		nhandle |= (id | 0x8000) << 16;
@@ -398,20 +398,20 @@ static int route4_set_parms(struct tcf_proto *tp, unsigned long base,
 	}
 
 	tcf_tree_lock(tp);
-	if (tb[TCA_ROUTE4_TO-1])
+	if (tb[TCA_ROUTE4_TO])
 		f->id = to;
 
-	if (tb[TCA_ROUTE4_FROM-1])
+	if (tb[TCA_ROUTE4_FROM])
 		f->id = to | id<<16;
-	else if (tb[TCA_ROUTE4_IIF-1])
+	else if (tb[TCA_ROUTE4_IIF])
 		f->iif = id;
 
 	f->handle = nhandle;
 	f->bkt = b;
 	tcf_tree_unlock(tp);
 
-	if (tb[TCA_ROUTE4_CLASSID-1]) {
-		f->res.classid = *(u32*)RTA_DATA(tb[TCA_ROUTE4_CLASSID-1]);
+	if (tb[TCA_ROUTE4_CLASSID]) {
+		f->res.classid = *(u32*)nla_data(tb[TCA_ROUTE4_CLASSID]);
 		tcf_bind_filter(tp, &f->res, base);
 	}
 
@@ -425,14 +425,14 @@ errout:
 
 static int route4_change(struct tcf_proto *tp, unsigned long base,
 		       u32 handle,
-		       struct rtattr **tca,
+		       struct nlattr **tca,
 		       unsigned long *arg)
 {
 	struct route4_head *head = tp->root;
 	struct route4_filter *f, *f1, **fp;
 	struct route4_bucket *b;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_ROUTE4_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_ROUTE4_MAX + 1];
 	unsigned int h, th;
 	u32 old_handle = 0;
 	int err;
@@ -440,7 +440,7 @@ static int route4_change(struct tcf_proto *tp, unsigned long base,
 	if (opt == NULL)
 		return handle ? -EINVAL : 0;
 
-	if (rtattr_parse_nested(tb, TCA_ROUTE4_MAX, opt) < 0)
+	if (nla_parse_nested(tb, TCA_ROUTE4_MAX, opt, NULL) < 0)
 		return -EINVAL;
 
 	if ((f = (struct route4_filter*)*arg) != NULL) {
@@ -451,7 +451,7 @@ static int route4_change(struct tcf_proto *tp, unsigned long base,
 			old_handle = f->handle;
 
 		err = route4_set_parms(tp, base, f, handle, head, tb,
-			tca[TCA_RATE-1], 0);
+			tca[TCA_RATE], 0);
 		if (err < 0)
 			return err;
 
@@ -474,7 +474,7 @@ static int route4_change(struct tcf_proto *tp, unsigned long base,
 		goto errout;
 
 	err = route4_set_parms(tp, base, f, handle, head, tb,
-		tca[TCA_RATE-1], 1);
+		tca[TCA_RATE], 1);
 	if (err < 0)
 		goto errout;
 
@@ -550,7 +550,7 @@ static int route4_dump(struct tcf_proto *tp, unsigned long fh,
 {
 	struct route4_filter *f = (struct route4_filter*)fh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 	u32 id;
 
 	if (f == NULL)
@@ -558,34 +558,34 @@ static int route4_dump(struct tcf_proto *tp, unsigned long fh,
 
 	t->tcm_handle = f->handle;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	if (!(f->handle&0x8000)) {
 		id = f->id&0xFF;
-		RTA_PUT(skb, TCA_ROUTE4_TO, sizeof(id), &id);
+		NLA_PUT(skb, TCA_ROUTE4_TO, sizeof(id), &id);
 	}
 	if (f->handle&0x80000000) {
 		if ((f->handle>>16) != 0xFFFF)
-			RTA_PUT(skb, TCA_ROUTE4_IIF, sizeof(f->iif), &f->iif);
+			NLA_PUT(skb, TCA_ROUTE4_IIF, sizeof(f->iif), &f->iif);
 	} else {
 		id = f->id>>16;
-		RTA_PUT(skb, TCA_ROUTE4_FROM, sizeof(id), &id);
+		NLA_PUT(skb, TCA_ROUTE4_FROM, sizeof(id), &id);
 	}
 	if (f->res.classid)
-		RTA_PUT(skb, TCA_ROUTE4_CLASSID, 4, &f->res.classid);
+		NLA_PUT(skb, TCA_ROUTE4_CLASSID, 4, &f->res.classid);
 
 	if (tcf_exts_dump(skb, &f->exts, &route_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 
 	if (tcf_exts_dump_stats(skb, &f->exts, &route_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h
index 22f9ede70e8f..2364c79d0837 100644
--- a/net/sched/cls_rsvp.h
+++ b/net/sched/cls_rsvp.h
@@ -399,15 +399,15 @@ static u32 gen_tunnel(struct rsvp_head *data)
 
 static int rsvp_change(struct tcf_proto *tp, unsigned long base,
 		       u32 handle,
-		       struct rtattr **tca,
+		       struct nlattr **tca,
 		       unsigned long *arg)
 {
 	struct rsvp_head *data = tp->root;
 	struct rsvp_filter *f, **fp;
 	struct rsvp_session *s, **sp;
 	struct tc_rsvp_pinfo *pinfo = NULL;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_RSVP_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS-1];
+	struct nlattr *tb[TCA_RSVP_MAX + 1];
 	struct tcf_exts e;
 	unsigned h1, h2;
 	__be32 *dst;
@@ -416,7 +416,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base,
 	if (opt == NULL)
 		return handle ? -EINVAL : 0;
 
-	if (rtattr_parse_nested(tb, TCA_RSVP_MAX, opt) < 0)
+	if (nla_parse_nested(tb, TCA_RSVP_MAX, opt, NULL) < 0)
 		return -EINVAL;
 
 	err = tcf_exts_validate(tp, tb, tca[TCA_RATE-1], &e, &rsvp_ext_map);
@@ -429,7 +429,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base,
 		if (f->handle != handle && handle)
 			goto errout2;
 		if (tb[TCA_RSVP_CLASSID-1]) {
-			f->res.classid = *(u32*)RTA_DATA(tb[TCA_RSVP_CLASSID-1]);
+			f->res.classid = *(u32*)nla_data(tb[TCA_RSVP_CLASSID-1]);
 			tcf_bind_filter(tp, &f->res, base);
 		}
 
@@ -452,30 +452,30 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base,
 	h2 = 16;
 	if (tb[TCA_RSVP_SRC-1]) {
 		err = -EINVAL;
-		if (RTA_PAYLOAD(tb[TCA_RSVP_SRC-1]) != sizeof(f->src))
+		if (nla_len(tb[TCA_RSVP_SRC-1]) != sizeof(f->src))
 			goto errout;
-		memcpy(f->src, RTA_DATA(tb[TCA_RSVP_SRC-1]), sizeof(f->src));
+		memcpy(f->src, nla_data(tb[TCA_RSVP_SRC-1]), sizeof(f->src));
 		h2 = hash_src(f->src);
 	}
 	if (tb[TCA_RSVP_PINFO-1]) {
 		err = -EINVAL;
-		if (RTA_PAYLOAD(tb[TCA_RSVP_PINFO-1]) < sizeof(struct tc_rsvp_pinfo))
+		if (nla_len(tb[TCA_RSVP_PINFO-1]) < sizeof(struct tc_rsvp_pinfo))
 			goto errout;
-		pinfo = RTA_DATA(tb[TCA_RSVP_PINFO-1]);
+		pinfo = nla_data(tb[TCA_RSVP_PINFO-1]);
 		f->spi = pinfo->spi;
 		f->tunnelhdr = pinfo->tunnelhdr;
 	}
 	if (tb[TCA_RSVP_CLASSID-1]) {
 		err = -EINVAL;
-		if (RTA_PAYLOAD(tb[TCA_RSVP_CLASSID-1]) != 4)
+		if (nla_len(tb[TCA_RSVP_CLASSID-1]) != 4)
 			goto errout;
-		f->res.classid = *(u32*)RTA_DATA(tb[TCA_RSVP_CLASSID-1]);
+		f->res.classid = *(u32*)nla_data(tb[TCA_RSVP_CLASSID-1]);
 	}
 
 	err = -EINVAL;
-	if (RTA_PAYLOAD(tb[TCA_RSVP_DST-1]) != sizeof(f->src))
+	if (nla_len(tb[TCA_RSVP_DST-1]) != sizeof(f->src))
 		goto errout;
-	dst = RTA_DATA(tb[TCA_RSVP_DST-1]);
+	dst = nla_data(tb[TCA_RSVP_DST-1]);
 	h1 = hash_dst(dst, pinfo ? pinfo->protocol : 0, pinfo ? pinfo->tunnelid : 0);
 
 	err = -ENOMEM;
@@ -594,7 +594,7 @@ static int rsvp_dump(struct tcf_proto *tp, unsigned long fh,
 	struct rsvp_filter *f = (struct rsvp_filter*)fh;
 	struct rsvp_session *s;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 	struct tc_rsvp_pinfo pinfo;
 
 	if (f == NULL)
@@ -604,32 +604,32 @@ static int rsvp_dump(struct tcf_proto *tp, unsigned long fh,
 	t->tcm_handle = f->handle;
 
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
-	RTA_PUT(skb, TCA_RSVP_DST, sizeof(s->dst), &s->dst);
+	NLA_PUT(skb, TCA_RSVP_DST, sizeof(s->dst), &s->dst);
 	pinfo.dpi = s->dpi;
 	pinfo.spi = f->spi;
 	pinfo.protocol = s->protocol;
 	pinfo.tunnelid = s->tunnelid;
 	pinfo.tunnelhdr = f->tunnelhdr;
 	pinfo.pad = 0;
-	RTA_PUT(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo);
+	NLA_PUT(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo);
 	if (f->res.classid)
-		RTA_PUT(skb, TCA_RSVP_CLASSID, 4, &f->res.classid);
+		NLA_PUT(skb, TCA_RSVP_CLASSID, 4, &f->res.classid);
 	if (((f->handle>>8)&0xFF) != 16)
-		RTA_PUT(skb, TCA_RSVP_SRC, sizeof(f->src), f->src);
+		NLA_PUT(skb, TCA_RSVP_SRC, sizeof(f->src), f->src);
 
 	if (tcf_exts_dump(skb, &f->exts, &rsvp_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 
 	if (tcf_exts_dump_stats(skb, &f->exts, &rsvp_ext_map) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c
index e36977b17fa6..ed8023944fe5 100644
--- a/net/sched/cls_tcindex.c
+++ b/net/sched/cls_tcindex.c
@@ -196,7 +196,7 @@ valid_perfect_hash(struct tcindex_data *p)
 static int
 tcindex_set_parms(struct tcf_proto *tp, unsigned long base, u32 handle,
 		  struct tcindex_data *p, struct tcindex_filter_result *r,
-		  struct rtattr **tb, struct rtattr *est)
+		  struct nlattr **tb, struct nlattr *est)
 {
 	int err, balloc = 0;
 	struct tcindex_filter_result new_filter_result, *old_r = r;
@@ -218,22 +218,22 @@ tcindex_set_parms(struct tcf_proto *tp, unsigned long base, u32 handle,
 		memset(&cr, 0, sizeof(cr));
 
 	err = -EINVAL;
-	if (tb[TCA_TCINDEX_HASH-1]) {
-		if (RTA_PAYLOAD(tb[TCA_TCINDEX_HASH-1]) < sizeof(u32))
+	if (tb[TCA_TCINDEX_HASH]) {
+		if (nla_len(tb[TCA_TCINDEX_HASH]) < sizeof(u32))
 			goto errout;
-		cp.hash = *(u32 *) RTA_DATA(tb[TCA_TCINDEX_HASH-1]);
+		cp.hash = *(u32 *) nla_data(tb[TCA_TCINDEX_HASH]);
 	}
 
-	if (tb[TCA_TCINDEX_MASK-1]) {
-		if (RTA_PAYLOAD(tb[TCA_TCINDEX_MASK-1]) < sizeof(u16))
+	if (tb[TCA_TCINDEX_MASK]) {
+		if (nla_len(tb[TCA_TCINDEX_MASK]) < sizeof(u16))
 			goto errout;
-		cp.mask = *(u16 *) RTA_DATA(tb[TCA_TCINDEX_MASK-1]);
+		cp.mask = *(u16 *) nla_data(tb[TCA_TCINDEX_MASK]);
 	}
 
-	if (tb[TCA_TCINDEX_SHIFT-1]) {
-		if (RTA_PAYLOAD(tb[TCA_TCINDEX_SHIFT-1]) < sizeof(int))
+	if (tb[TCA_TCINDEX_SHIFT]) {
+		if (nla_len(tb[TCA_TCINDEX_SHIFT]) < sizeof(int))
 			goto errout;
-		cp.shift = *(int *) RTA_DATA(tb[TCA_TCINDEX_SHIFT-1]);
+		cp.shift = *(int *) nla_data(tb[TCA_TCINDEX_SHIFT]);
 	}
 
 	err = -EBUSY;
@@ -248,11 +248,11 @@ tcindex_set_parms(struct tcf_proto *tp, unsigned long base, u32 handle,
 		goto errout;
 
 	err = -EINVAL;
-	if (tb[TCA_TCINDEX_FALL_THROUGH-1]) {
-		if (RTA_PAYLOAD(tb[TCA_TCINDEX_FALL_THROUGH-1]) < sizeof(u32))
+	if (tb[TCA_TCINDEX_FALL_THROUGH]) {
+		if (nla_len(tb[TCA_TCINDEX_FALL_THROUGH]) < sizeof(u32))
 			goto errout;
 		cp.fall_through =
-			*(u32 *) RTA_DATA(tb[TCA_TCINDEX_FALL_THROUGH-1]);
+			*(u32 *) nla_data(tb[TCA_TCINDEX_FALL_THROUGH]);
 	}
 
 	if (!cp.hash) {
@@ -304,8 +304,8 @@ tcindex_set_parms(struct tcf_proto *tp, unsigned long base, u32 handle,
 			goto errout_alloc;
 	}
 
-	if (tb[TCA_TCINDEX_CLASSID-1]) {
-		cr.res.classid = *(u32 *) RTA_DATA(tb[TCA_TCINDEX_CLASSID-1]);
+	if (tb[TCA_TCINDEX_CLASSID]) {
+		cr.res.classid = *(u32 *) nla_data(tb[TCA_TCINDEX_CLASSID]);
 		tcf_bind_filter(tp, &cr.res, base);
 	}
 
@@ -344,10 +344,10 @@ errout:
 
 static int
 tcindex_change(struct tcf_proto *tp, unsigned long base, u32 handle,
-	       struct rtattr **tca, unsigned long *arg)
+	       struct nlattr **tca, unsigned long *arg)
 {
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_TCINDEX_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_TCINDEX_MAX + 1];
 	struct tcindex_data *p = PRIV(tp);
 	struct tcindex_filter_result *r = (struct tcindex_filter_result *) *arg;
 
@@ -358,10 +358,10 @@ tcindex_change(struct tcf_proto *tp, unsigned long base, u32 handle,
 	if (!opt)
 		return 0;
 
-	if (rtattr_parse_nested(tb, TCA_TCINDEX_MAX, opt) < 0)
+	if (nla_parse_nested(tb, TCA_TCINDEX_MAX, opt, NULL) < 0)
 		return -EINVAL;
 
-	return tcindex_set_parms(tp, base, handle, p, r, tb, tca[TCA_RATE-1]);
+	return tcindex_set_parms(tp, base, handle, p, r, tb, tca[TCA_RATE]);
 }
 
 
@@ -435,21 +435,21 @@ static int tcindex_dump(struct tcf_proto *tp, unsigned long fh,
 	struct tcindex_data *p = PRIV(tp);
 	struct tcindex_filter_result *r = (struct tcindex_filter_result *) fh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	pr_debug("tcindex_dump(tp %p,fh 0x%lx,skb %p,t %p),p %p,r %p,b %p\n",
 		 tp, fh, skb, t, p, r, b);
 	pr_debug("p->perfect %p p->h %p\n", p->perfect, p->h);
-	rta = (struct rtattr *) b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr *) b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 	if (!fh) {
 		t->tcm_handle = ~0; /* whatever ... */
-		RTA_PUT(skb, TCA_TCINDEX_HASH, sizeof(p->hash), &p->hash);
-		RTA_PUT(skb, TCA_TCINDEX_MASK, sizeof(p->mask), &p->mask);
-		RTA_PUT(skb, TCA_TCINDEX_SHIFT, sizeof(p->shift), &p->shift);
-		RTA_PUT(skb, TCA_TCINDEX_FALL_THROUGH, sizeof(p->fall_through),
+		NLA_PUT(skb, TCA_TCINDEX_HASH, sizeof(p->hash), &p->hash);
+		NLA_PUT(skb, TCA_TCINDEX_MASK, sizeof(p->mask), &p->mask);
+		NLA_PUT(skb, TCA_TCINDEX_SHIFT, sizeof(p->shift), &p->shift);
+		NLA_PUT(skb, TCA_TCINDEX_FALL_THROUGH, sizeof(p->fall_through),
 		    &p->fall_through);
-		rta->rta_len = skb_tail_pointer(skb) - b;
+		nla->nla_len = skb_tail_pointer(skb) - b;
 	} else {
 		if (p->perfect) {
 			t->tcm_handle = r-p->perfect;
@@ -468,19 +468,19 @@ static int tcindex_dump(struct tcf_proto *tp, unsigned long fh,
 		}
 		pr_debug("handle = %d\n", t->tcm_handle);
 		if (r->res.class)
-			RTA_PUT(skb, TCA_TCINDEX_CLASSID, 4, &r->res.classid);
+			NLA_PUT(skb, TCA_TCINDEX_CLASSID, 4, &r->res.classid);
 
 		if (tcf_exts_dump(skb, &r->exts, &tcindex_ext_map) < 0)
-			goto rtattr_failure;
-		rta->rta_len = skb_tail_pointer(skb) - b;
+			goto nla_put_failure;
+		nla->nla_len = skb_tail_pointer(skb) - b;
 
 		if (tcf_exts_dump_stats(skb, &r->exts, &tcindex_ext_map) < 0)
-			goto rtattr_failure;
+			goto nla_put_failure;
 	}
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 7bf3cd4e7318..aaf5049f951c 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -462,8 +462,8 @@ static u32 gen_new_kid(struct tc_u_hnode *ht, u32 handle)
 
 static int u32_set_parms(struct tcf_proto *tp, unsigned long base,
 			 struct tc_u_hnode *ht,
-			 struct tc_u_knode *n, struct rtattr **tb,
-			 struct rtattr *est)
+			 struct tc_u_knode *n, struct nlattr **tb,
+			 struct nlattr *est)
 {
 	int err;
 	struct tcf_exts e;
@@ -473,8 +473,8 @@ static int u32_set_parms(struct tcf_proto *tp, unsigned long base,
 		return err;
 
 	err = -EINVAL;
-	if (tb[TCA_U32_LINK-1]) {
-		u32 handle = *(u32*)RTA_DATA(tb[TCA_U32_LINK-1]);
+	if (tb[TCA_U32_LINK]) {
+		u32 handle = *(u32*)nla_data(tb[TCA_U32_LINK]);
 		struct tc_u_hnode *ht_down = NULL;
 
 		if (TC_U32_KEY(handle))
@@ -495,14 +495,14 @@ static int u32_set_parms(struct tcf_proto *tp, unsigned long base,
 		if (ht_down)
 			ht_down->refcnt--;
 	}
-	if (tb[TCA_U32_CLASSID-1]) {
-		n->res.classid = *(u32*)RTA_DATA(tb[TCA_U32_CLASSID-1]);
+	if (tb[TCA_U32_CLASSID]) {
+		n->res.classid = *(u32*)nla_data(tb[TCA_U32_CLASSID]);
 		tcf_bind_filter(tp, &n->res, base);
 	}
 
 #ifdef CONFIG_NET_CLS_IND
-	if (tb[TCA_U32_INDEV-1]) {
-		err = tcf_change_indev(tp, n->indev, tb[TCA_U32_INDEV-1]);
+	if (tb[TCA_U32_INDEV]) {
+		err = tcf_change_indev(tp, n->indev, tb[TCA_U32_INDEV]);
 		if (err < 0)
 			goto errout;
 	}
@@ -516,33 +516,33 @@ errout:
 }
 
 static int u32_change(struct tcf_proto *tp, unsigned long base, u32 handle,
-		      struct rtattr **tca,
+		      struct nlattr **tca,
 		      unsigned long *arg)
 {
 	struct tc_u_common *tp_c = tp->data;
 	struct tc_u_hnode *ht;
 	struct tc_u_knode *n;
 	struct tc_u32_sel *s;
-	struct rtattr *opt = tca[TCA_OPTIONS-1];
-	struct rtattr *tb[TCA_U32_MAX];
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_U32_MAX + 1];
 	u32 htid;
 	int err;
 
 	if (opt == NULL)
 		return handle ? -EINVAL : 0;
 
-	if (rtattr_parse_nested(tb, TCA_U32_MAX, opt) < 0)
+	if (nla_parse_nested(tb, TCA_U32_MAX, opt, NULL) < 0)
 		return -EINVAL;
 
 	if ((n = (struct tc_u_knode*)*arg) != NULL) {
 		if (TC_U32_KEY(n->handle) == 0)
 			return -EINVAL;
 
-		return u32_set_parms(tp, base, n->ht_up, n, tb, tca[TCA_RATE-1]);
+		return u32_set_parms(tp, base, n->ht_up, n, tb, tca[TCA_RATE]);
 	}
 
-	if (tb[TCA_U32_DIVISOR-1]) {
-		unsigned divisor = *(unsigned*)RTA_DATA(tb[TCA_U32_DIVISOR-1]);
+	if (tb[TCA_U32_DIVISOR]) {
+		unsigned divisor = *(unsigned*)nla_data(tb[TCA_U32_DIVISOR]);
 
 		if (--divisor > 0x100)
 			return -EINVAL;
@@ -567,8 +567,8 @@ static int u32_change(struct tcf_proto *tp, unsigned long base, u32 handle,
 		return 0;
 	}
 
-	if (tb[TCA_U32_HASH-1]) {
-		htid = *(unsigned*)RTA_DATA(tb[TCA_U32_HASH-1]);
+	if (tb[TCA_U32_HASH]) {
+		htid = *(unsigned*)nla_data(tb[TCA_U32_HASH]);
 		if (TC_U32_HTID(htid) == TC_U32_ROOT) {
 			ht = tp->root;
 			htid = ht->handle;
@@ -592,11 +592,11 @@ static int u32_change(struct tcf_proto *tp, unsigned long base, u32 handle,
 	} else
 		handle = gen_new_kid(ht, htid);
 
-	if (tb[TCA_U32_SEL-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_U32_SEL-1]) < sizeof(struct tc_u32_sel))
+	if (tb[TCA_U32_SEL] == NULL ||
+	    nla_len(tb[TCA_U32_SEL]) < sizeof(struct tc_u32_sel))
 		return -EINVAL;
 
-	s = RTA_DATA(tb[TCA_U32_SEL-1]);
+	s = nla_data(tb[TCA_U32_SEL]);
 
 	n = kzalloc(sizeof(*n) + s->nkeys*sizeof(struct tc_u32_key), GFP_KERNEL);
 	if (n == NULL)
@@ -616,23 +616,23 @@ static int u32_change(struct tcf_proto *tp, unsigned long base, u32 handle,
 	n->fshift = s->hmask ? ffs(ntohl(s->hmask)) - 1 : 0;
 
 #ifdef CONFIG_CLS_U32_MARK
-	if (tb[TCA_U32_MARK-1]) {
+	if (tb[TCA_U32_MARK]) {
 		struct tc_u32_mark *mark;
 
-		if (RTA_PAYLOAD(tb[TCA_U32_MARK-1]) < sizeof(struct tc_u32_mark)) {
+		if (nla_len(tb[TCA_U32_MARK]) < sizeof(struct tc_u32_mark)) {
 #ifdef CONFIG_CLS_U32_PERF
 			kfree(n->pf);
 #endif
 			kfree(n);
 			return -EINVAL;
 		}
-		mark = RTA_DATA(tb[TCA_U32_MARK-1]);
+		mark = nla_data(tb[TCA_U32_MARK]);
 		memcpy(&n->mark, mark, sizeof(struct tc_u32_mark));
 		n->mark.success = 0;
 	}
 #endif
 
-	err = u32_set_parms(tp, base, ht, n, tb, tca[TCA_RATE-1]);
+	err = u32_set_parms(tp, base, ht, n, tb, tca[TCA_RATE]);
 	if (err == 0) {
 		struct tc_u_knode **ins;
 		for (ins = &ht->ht[TC_U32_HASH(handle)]; *ins; ins = &(*ins)->next)
@@ -694,59 +694,59 @@ static int u32_dump(struct tcf_proto *tp, unsigned long fh,
 {
 	struct tc_u_knode *n = (struct tc_u_knode*)fh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *rta;
+	struct nlattr *nla;
 
 	if (n == NULL)
 		return skb->len;
 
 	t->tcm_handle = n->handle;
 
-	rta = (struct rtattr*)b;
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+	nla = (struct nlattr*)b;
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 
 	if (TC_U32_KEY(n->handle) == 0) {
 		struct tc_u_hnode *ht = (struct tc_u_hnode*)fh;
 		u32 divisor = ht->divisor+1;
-		RTA_PUT(skb, TCA_U32_DIVISOR, 4, &divisor);
+		NLA_PUT(skb, TCA_U32_DIVISOR, 4, &divisor);
 	} else {
-		RTA_PUT(skb, TCA_U32_SEL,
+		NLA_PUT(skb, TCA_U32_SEL,
 			sizeof(n->sel) + n->sel.nkeys*sizeof(struct tc_u32_key),
 			&n->sel);
 		if (n->ht_up) {
 			u32 htid = n->handle & 0xFFFFF000;
-			RTA_PUT(skb, TCA_U32_HASH, 4, &htid);
+			NLA_PUT(skb, TCA_U32_HASH, 4, &htid);
 		}
 		if (n->res.classid)
-			RTA_PUT(skb, TCA_U32_CLASSID, 4, &n->res.classid);
+			NLA_PUT(skb, TCA_U32_CLASSID, 4, &n->res.classid);
 		if (n->ht_down)
-			RTA_PUT(skb, TCA_U32_LINK, 4, &n->ht_down->handle);
+			NLA_PUT(skb, TCA_U32_LINK, 4, &n->ht_down->handle);
 
 #ifdef CONFIG_CLS_U32_MARK
 		if (n->mark.val || n->mark.mask)
-			RTA_PUT(skb, TCA_U32_MARK, sizeof(n->mark), &n->mark);
+			NLA_PUT(skb, TCA_U32_MARK, sizeof(n->mark), &n->mark);
 #endif
 
 		if (tcf_exts_dump(skb, &n->exts, &u32_ext_map) < 0)
-			goto rtattr_failure;
+			goto nla_put_failure;
 
 #ifdef CONFIG_NET_CLS_IND
 		if(strlen(n->indev))
-			RTA_PUT(skb, TCA_U32_INDEV, IFNAMSIZ, n->indev);
+			NLA_PUT(skb, TCA_U32_INDEV, IFNAMSIZ, n->indev);
 #endif
 #ifdef CONFIG_CLS_U32_PERF
-		RTA_PUT(skb, TCA_U32_PCNT,
+		NLA_PUT(skb, TCA_U32_PCNT,
 		sizeof(struct tc_u32_pcnt) + n->sel.nkeys*sizeof(u64),
 			n->pf);
 #endif
 	}
 
-	rta->rta_len = skb_tail_pointer(skb) - b;
+	nla->nla_len = skb_tail_pointer(skb) - b;
 	if (TC_U32_KEY(n->handle))
 		if (tcf_exts_dump_stats(skb, &n->exts, &u32_ext_map) < 0)
-			goto rtattr_failure;
+			goto nla_put_failure;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index ceda8890ab0e..92b6863e928d 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -542,11 +542,11 @@ static int meta_var_compare(struct meta_obj *a, struct meta_obj *b)
 	return r;
 }
 
-static int meta_var_change(struct meta_value *dst, struct rtattr *rta)
+static int meta_var_change(struct meta_value *dst, struct nlattr *nla)
 {
-	int len = RTA_PAYLOAD(rta);
+	int len = nla_len(nla);
 
-	dst->val = (unsigned long)kmemdup(RTA_DATA(rta), len, GFP_KERNEL);
+	dst->val = (unsigned long)kmemdup(nla_data(nla), len, GFP_KERNEL);
 	if (dst->val == 0UL)
 		return -ENOMEM;
 	dst->len = len;
@@ -570,10 +570,10 @@ static void meta_var_apply_extras(struct meta_value *v,
 static int meta_var_dump(struct sk_buff *skb, struct meta_value *v, int tlv)
 {
 	if (v->val && v->len)
-		RTA_PUT(skb, tlv, v->len, (void *) v->val);
+		NLA_PUT(skb, tlv, v->len, (void *) v->val);
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
@@ -594,13 +594,13 @@ static int meta_int_compare(struct meta_obj *a, struct meta_obj *b)
 		return 1;
 }
 
-static int meta_int_change(struct meta_value *dst, struct rtattr *rta)
+static int meta_int_change(struct meta_value *dst, struct nlattr *nla)
 {
-	if (RTA_PAYLOAD(rta) >= sizeof(unsigned long)) {
-		dst->val = *(unsigned long *) RTA_DATA(rta);
+	if (nla_len(nla) >= sizeof(unsigned long)) {
+		dst->val = *(unsigned long *) nla_data(nla);
 		dst->len = sizeof(unsigned long);
-	} else if (RTA_PAYLOAD(rta) == sizeof(u32)) {
-		dst->val = *(u32 *) RTA_DATA(rta);
+	} else if (nla_len(nla) == sizeof(u32)) {
+		dst->val = *(u32 *) nla_data(nla);
 		dst->len = sizeof(u32);
 	} else
 		return -EINVAL;
@@ -621,15 +621,15 @@ static void meta_int_apply_extras(struct meta_value *v,
 static int meta_int_dump(struct sk_buff *skb, struct meta_value *v, int tlv)
 {
 	if (v->len == sizeof(unsigned long))
-		RTA_PUT(skb, tlv, sizeof(unsigned long), &v->val);
+		NLA_PUT(skb, tlv, sizeof(unsigned long), &v->val);
 	else if (v->len == sizeof(u32)) {
 		u32 d = v->val;
-		RTA_PUT(skb, tlv, sizeof(d), &d);
+		NLA_PUT(skb, tlv, sizeof(d), &d);
 	}
 
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
@@ -641,7 +641,7 @@ struct meta_type_ops
 {
 	void	(*destroy)(struct meta_value *);
 	int	(*compare)(struct meta_obj *, struct meta_obj *);
-	int	(*change)(struct meta_value *, struct rtattr *);
+	int	(*change)(struct meta_value *, struct nlattr *);
 	void	(*apply_extras)(struct meta_value *, struct meta_obj *);
 	int	(*dump)(struct sk_buff *, struct meta_value *, int);
 };
@@ -729,13 +729,13 @@ static inline void meta_delete(struct meta_match *meta)
 	kfree(meta);
 }
 
-static inline int meta_change_data(struct meta_value *dst, struct rtattr *rta)
+static inline int meta_change_data(struct meta_value *dst, struct nlattr *nla)
 {
-	if (rta) {
-		if (RTA_PAYLOAD(rta) == 0)
+	if (nla) {
+		if (nla_len(nla) == 0)
 			return -EINVAL;
 
-		return meta_type_ops(dst)->change(dst, rta);
+		return meta_type_ops(dst)->change(dst, nla);
 	}
 
 	return 0;
@@ -750,17 +750,17 @@ static int em_meta_change(struct tcf_proto *tp, void *data, int len,
 			  struct tcf_ematch *m)
 {
 	int err = -EINVAL;
-	struct rtattr *tb[TCA_EM_META_MAX];
+	struct nlattr *tb[TCA_EM_META_MAX + 1];
 	struct tcf_meta_hdr *hdr;
 	struct meta_match *meta = NULL;
 
-	if (rtattr_parse(tb, TCA_EM_META_MAX, data, len) < 0)
+	if (nla_parse(tb, TCA_EM_META_MAX, data, len, NULL) < 0)
 		goto errout;
 
-	if (tb[TCA_EM_META_HDR-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_EM_META_HDR-1]) < sizeof(*hdr))
+	if (tb[TCA_EM_META_HDR] == NULL ||
+	    nla_len(tb[TCA_EM_META_HDR]) < sizeof(*hdr))
 		goto errout;
-	hdr = RTA_DATA(tb[TCA_EM_META_HDR-1]);
+	hdr = nla_data(tb[TCA_EM_META_HDR]);
 
 	if (TCF_META_TYPE(hdr->left.kind) != TCF_META_TYPE(hdr->right.kind) ||
 	    TCF_META_TYPE(hdr->left.kind) > TCF_META_TYPE_MAX ||
@@ -781,8 +781,8 @@ static int em_meta_change(struct tcf_proto *tp, void *data, int len,
 		goto errout;
 	}
 
-	if (meta_change_data(&meta->lvalue, tb[TCA_EM_META_LVALUE-1]) < 0 ||
-	    meta_change_data(&meta->rvalue, tb[TCA_EM_META_RVALUE-1]) < 0)
+	if (meta_change_data(&meta->lvalue, tb[TCA_EM_META_LVALUE]) < 0 ||
+	    meta_change_data(&meta->rvalue, tb[TCA_EM_META_RVALUE]) < 0)
 		goto errout;
 
 	m->datalen = sizeof(*meta);
@@ -811,16 +811,16 @@ static int em_meta_dump(struct sk_buff *skb, struct tcf_ematch *em)
 	memcpy(&hdr.left, &meta->lvalue.hdr, sizeof(hdr.left));
 	memcpy(&hdr.right, &meta->rvalue.hdr, sizeof(hdr.right));
 
-	RTA_PUT(skb, TCA_EM_META_HDR, sizeof(hdr), &hdr);
+	NLA_PUT(skb, TCA_EM_META_HDR, sizeof(hdr), &hdr);
 
 	ops = meta_type_ops(&meta->lvalue);
 	if (ops->dump(skb, &meta->lvalue, TCA_EM_META_LVALUE) < 0 ||
 	    ops->dump(skb, &meta->rvalue, TCA_EM_META_RVALUE) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
diff --git a/net/sched/em_text.c b/net/sched/em_text.c
index d5cd86efb7d0..853c5ead87fd 100644
--- a/net/sched/em_text.c
+++ b/net/sched/em_text.c
@@ -118,11 +118,14 @@ static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m)
 	conf.pattern_len = textsearch_get_pattern_len(tm->config);
 	conf.pad = 0;
 
-	RTA_PUT_NOHDR(skb, sizeof(conf), &conf);
-	RTA_APPEND(skb, conf.pattern_len, textsearch_get_pattern(tm->config));
+	if (nla_put_nohdr(skb, sizeof(conf), &conf) < 0)
+		goto nla_put_failure;
+	if (nla_append(skb, conf.pattern_len,
+		       textsearch_get_pattern(tm->config)) < 0)
+		goto nla_put_failure;
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 
diff --git a/net/sched/ematch.c b/net/sched/ematch.c
index 27941cfc0ab5..72d9b2735245 100644
--- a/net/sched/ematch.c
+++ b/net/sched/ematch.c
@@ -183,11 +183,11 @@ static inline struct tcf_ematch * tcf_em_get_match(struct tcf_ematch_tree *tree,
 
 static int tcf_em_validate(struct tcf_proto *tp,
 			   struct tcf_ematch_tree_hdr *tree_hdr,
-			   struct tcf_ematch *em, struct rtattr *rta, int idx)
+			   struct tcf_ematch *em, struct nlattr *nla, int idx)
 {
 	int err = -EINVAL;
-	struct tcf_ematch_hdr *em_hdr = RTA_DATA(rta);
-	int data_len = RTA_PAYLOAD(rta) - sizeof(*em_hdr);
+	struct tcf_ematch_hdr *em_hdr = nla_data(nla);
+	int data_len = nla_len(nla) - sizeof(*em_hdr);
 	void *data = (void *) em_hdr + sizeof(*em_hdr);
 
 	if (!TCF_EM_REL_VALID(em_hdr->flags))
@@ -286,11 +286,11 @@ errout:
  * tcf_em_tree_validate - validate ematch config TLV and build ematch tree
  *
  * @tp: classifier kind handle
- * @rta: ematch tree configuration TLV
+ * @nla: ematch tree configuration TLV
  * @tree: destination ematch tree variable to store the resulting
  *        ematch tree.
  *
- * This function validates the given configuration TLV @rta and builds an
+ * This function validates the given configuration TLV @nla and builds an
  * ematch tree in @tree. The resulting tree must later be copied into
  * the private classifier data using tcf_em_tree_change(). You MUST NOT
  * provide the ematch tree variable of the private classifier data directly,
@@ -298,45 +298,45 @@ errout:
  *
  * Returns a negative error code if the configuration TLV contains errors.
  */
-int tcf_em_tree_validate(struct tcf_proto *tp, struct rtattr *rta,
+int tcf_em_tree_validate(struct tcf_proto *tp, struct nlattr *nla,
 			 struct tcf_ematch_tree *tree)
 {
 	int idx, list_len, matches_len, err = -EINVAL;
-	struct rtattr *tb[TCA_EMATCH_TREE_MAX];
-	struct rtattr *rt_match, *rt_hdr, *rt_list;
+	struct nlattr *tb[TCA_EMATCH_TREE_MAX + 1];
+	struct nlattr *rt_match, *rt_hdr, *rt_list;
 	struct tcf_ematch_tree_hdr *tree_hdr;
 	struct tcf_ematch *em;
 
-	if (!rta) {
+	if (!nla) {
 		memset(tree, 0, sizeof(*tree));
 		return 0;
 	}
 
-	if (rtattr_parse_nested(tb, TCA_EMATCH_TREE_MAX, rta) < 0)
+	if (nla_parse_nested(tb, TCA_EMATCH_TREE_MAX, nla, NULL) < 0)
 		goto errout;
 
-	rt_hdr = tb[TCA_EMATCH_TREE_HDR-1];
-	rt_list = tb[TCA_EMATCH_TREE_LIST-1];
+	rt_hdr = tb[TCA_EMATCH_TREE_HDR];
+	rt_list = tb[TCA_EMATCH_TREE_LIST];
 
 	if (rt_hdr == NULL || rt_list == NULL)
 		goto errout;
 
-	if (RTA_PAYLOAD(rt_hdr) < sizeof(*tree_hdr) ||
-	    RTA_PAYLOAD(rt_list) < sizeof(*rt_match))
+	if (nla_len(rt_hdr) < sizeof(*tree_hdr) ||
+	    nla_len(rt_list) < sizeof(*rt_match))
 		goto errout;
 
-	tree_hdr = RTA_DATA(rt_hdr);
+	tree_hdr = nla_data(rt_hdr);
 	memcpy(&tree->hdr, tree_hdr, sizeof(*tree_hdr));
 
-	rt_match = RTA_DATA(rt_list);
-	list_len = RTA_PAYLOAD(rt_list);
+	rt_match = nla_data(rt_list);
+	list_len = nla_len(rt_list);
 	matches_len = tree_hdr->nmatches * sizeof(*em);
 
 	tree->matches = kzalloc(matches_len, GFP_KERNEL);
 	if (tree->matches == NULL)
 		goto errout;
 
-	/* We do not use rtattr_parse_nested here because the maximum
+	/* We do not use nla_parse_nested here because the maximum
 	 * number of attributes is unknown. This saves us the allocation
 	 * for a tb buffer which would serve no purpose at all.
 	 *
@@ -344,16 +344,16 @@ int tcf_em_tree_validate(struct tcf_proto *tp, struct rtattr *rta,
 	 * provided, their type must be incremental from 1 to n. Even
 	 * if it does not serve any real purpose, a failure of sticking
 	 * to this policy will result in parsing failure. */
-	for (idx = 0; RTA_OK(rt_match, list_len); idx++) {
+	for (idx = 0; nla_ok(rt_match, list_len); idx++) {
 		err = -EINVAL;
 
-		if (rt_match->rta_type != (idx + 1))
+		if (rt_match->nla_type != (idx + 1))
 			goto errout_abort;
 
 		if (idx >= tree_hdr->nmatches)
 			goto errout_abort;
 
-		if (RTA_PAYLOAD(rt_match) < sizeof(struct tcf_ematch_hdr))
+		if (nla_len(rt_match) < sizeof(struct tcf_ematch_hdr))
 			goto errout_abort;
 
 		em = tcf_em_get_match(tree, idx);
@@ -362,7 +362,7 @@ int tcf_em_tree_validate(struct tcf_proto *tp, struct rtattr *rta,
 		if (err < 0)
 			goto errout_abort;
 
-		rt_match = RTA_NEXT(rt_match, list_len);
+		rt_match = nla_next(rt_match, &list_len);
 	}
 
 	/* Check if the number of matches provided by userspace actually
@@ -434,18 +434,18 @@ int tcf_em_tree_dump(struct sk_buff *skb, struct tcf_ematch_tree *tree, int tlv)
 {
 	int i;
 	u8 *tail;
-	struct rtattr *top_start = (struct rtattr *)skb_tail_pointer(skb);
-	struct rtattr *list_start;
+	struct nlattr *top_start = (struct nlattr *)skb_tail_pointer(skb);
+	struct nlattr *list_start;
 
-	RTA_PUT(skb, tlv, 0, NULL);
-	RTA_PUT(skb, TCA_EMATCH_TREE_HDR, sizeof(tree->hdr), &tree->hdr);
+	NLA_PUT(skb, tlv, 0, NULL);
+	NLA_PUT(skb, TCA_EMATCH_TREE_HDR, sizeof(tree->hdr), &tree->hdr);
 
-	list_start = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_EMATCH_TREE_LIST, 0, NULL);
+	list_start = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_EMATCH_TREE_LIST, 0, NULL);
 
 	tail = skb_tail_pointer(skb);
 	for (i = 0; i < tree->hdr.nmatches; i++) {
-		struct rtattr *match_start = (struct rtattr *)tail;
+		struct nlattr *match_start = (struct nlattr *)tail;
 		struct tcf_ematch *em = tcf_em_get_match(tree, i);
 		struct tcf_ematch_hdr em_hdr = {
 			.kind = em->ops ? em->ops->kind : TCF_EM_CONTAINER,
@@ -453,27 +453,27 @@ int tcf_em_tree_dump(struct sk_buff *skb, struct tcf_ematch_tree *tree, int tlv)
 			.flags = em->flags
 		};
 
-		RTA_PUT(skb, i+1, sizeof(em_hdr), &em_hdr);
+		NLA_PUT(skb, i+1, sizeof(em_hdr), &em_hdr);
 
 		if (em->ops && em->ops->dump) {
 			if (em->ops->dump(skb, em) < 0)
-				goto rtattr_failure;
+				goto nla_put_failure;
 		} else if (tcf_em_is_container(em) || tcf_em_is_simple(em)) {
 			u32 u = em->data;
-			RTA_PUT_NOHDR(skb, sizeof(u), &u);
+			nla_put_nohdr(skb, sizeof(u), &u);
 		} else if (em->datalen > 0)
-			RTA_PUT_NOHDR(skb, em->datalen, (void *) em->data);
+			nla_put_nohdr(skb, em->datalen, (void *) em->data);
 
 		tail = skb_tail_pointer(skb);
-		match_start->rta_len = tail - (u8 *)match_start;
+		match_start->nla_len = tail - (u8 *)match_start;
 	}
 
-	list_start->rta_len = tail - (u8 *)list_start;
-	top_start->rta_len = tail - (u8 *)top_start;
+	list_start->nla_len = tail - (u8 *)list_start;
+	top_start->nla_len = tail - (u8 *)top_start;
 
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	return -1;
 }
 EXPORT_SYMBOL(tcf_em_tree_dump);
-- 
cgit v1.2.3


From 7ba699c604ab811972eee2e041fd6b07659a2e6e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Jan 2008 22:11:50 -0800
Subject: [NET_SCHED]: Convert actions from rtnetlink to new netlink API

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/act_api.h  |   8 +-
 net/sched/act_api.c    | 214 +++++++++++++++++++++++++------------------------
 net/sched/act_gact.c   |  28 +++----
 net/sched/act_ipt.c    |  46 +++++------
 net/sched/act_mirred.c |  18 ++---
 net/sched/act_nat.c    |  18 ++---
 net/sched/act_pedit.c  |  20 ++---
 net/sched/act_police.c |  52 ++++++------
 net/sched/act_simple.c |  24 +++---
 net/sched/cls_api.c    |   6 +-
 10 files changed, 218 insertions(+), 216 deletions(-)

(limited to 'include')

diff --git a/include/net/act_api.h b/include/net/act_api.h
index 68b4eaf7719d..c5ac61ada94f 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -89,7 +89,7 @@ struct tc_action_ops {
 	int     (*dump)(struct sk_buff *, struct tc_action *, int, int);
 	int     (*cleanup)(struct tc_action *, int bind);
 	int     (*lookup)(struct tc_action *, u32);
-	int     (*init)(struct rtattr *, struct rtattr *, struct tc_action *, int , int);
+	int     (*init)(struct nlattr *, struct nlattr *, struct tc_action *, int , int);
 	int     (*walk)(struct sk_buff *, struct netlink_callback *, int, struct tc_action *);
 };
 
@@ -104,7 +104,7 @@ extern u32 tcf_hash_new_index(u32 *idx_gen, struct tcf_hashinfo *hinfo);
 extern int tcf_hash_search(struct tc_action *a, u32 index);
 extern struct tcf_common *tcf_hash_check(u32 index, struct tc_action *a,
 					 int bind, struct tcf_hashinfo *hinfo);
-extern struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est,
+extern struct tcf_common *tcf_hash_create(u32 index, struct nlattr *est,
 					  struct tc_action *a, int size,
 					  int bind, u32 *idx_gen,
 					  struct tcf_hashinfo *hinfo);
@@ -114,8 +114,8 @@ extern int tcf_register_action(struct tc_action_ops *a);
 extern int tcf_unregister_action(struct tc_action_ops *a);
 extern void tcf_action_destroy(struct tc_action *a, int bind);
 extern int tcf_action_exec(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res);
-extern struct tc_action *tcf_action_init(struct rtattr *rta, struct rtattr *est, char *n, int ovr, int bind, int *err);
-extern struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est, char *n, int ovr, int bind, int *err);
+extern struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind, int *err);
+extern struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind, int *err);
 extern int tcf_action_dump(struct sk_buff *skb, struct tc_action *a, int, int);
 extern int tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int, int);
 extern int tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int, int);
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 11f3097a6912..ebd21d2cb5f1 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -68,7 +68,7 @@ static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,
 {
 	struct tcf_common *p;
 	int err = 0, index = -1,i = 0, s_i = 0, n_i = 0;
-	struct rtattr *r ;
+	struct nlattr *r ;
 
 	read_lock_bh(hinfo->lock);
 
@@ -83,15 +83,15 @@ static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,
 				continue;
 			a->priv = p;
 			a->order = n_i;
-			r = (struct rtattr *)skb_tail_pointer(skb);
-			RTA_PUT(skb, a->order, 0, NULL);
+			r = (struct nlattr *)skb_tail_pointer(skb);
+			NLA_PUT(skb, a->order, 0, NULL);
 			err = tcf_action_dump_1(skb, a, 0, 0);
 			if (err < 0) {
 				index--;
 				nlmsg_trim(skb, r);
 				goto done;
 			}
-			r->rta_len = skb_tail_pointer(skb) - (u8 *)r;
+			r->nla_len = skb_tail_pointer(skb) - (u8 *)r;
 			n_i++;
 			if (n_i >= TCA_ACT_MAX_PRIO)
 				goto done;
@@ -103,7 +103,7 @@ done:
 		cb->args[0] += n_i;
 	return n_i;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, r);
 	goto done;
 }
@@ -112,12 +112,12 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a,
 			  struct tcf_hashinfo *hinfo)
 {
 	struct tcf_common *p, *s_p;
-	struct rtattr *r ;
+	struct nlattr *r ;
 	int i= 0, n_i = 0;
 
-	r = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, a->order, 0, NULL);
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
+	r = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, a->order, 0, NULL);
+	NLA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
 	for (i = 0; i < (hinfo->hmask + 1); i++) {
 		p = hinfo->htab[tcf_hash(i, hinfo->hmask)];
 
@@ -129,11 +129,11 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a,
 			p = s_p;
 		}
 	}
-	RTA_PUT(skb, TCA_FCNT, 4, &n_i);
-	r->rta_len = skb_tail_pointer(skb) - (u8 *)r;
+	NLA_PUT(skb, TCA_FCNT, 4, &n_i);
+	r->nla_len = skb_tail_pointer(skb) - (u8 *)r;
 
 	return n_i;
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, r);
 	return -EINVAL;
 }
@@ -211,7 +211,7 @@ struct tcf_common *tcf_hash_check(u32 index, struct tc_action *a, int bind,
 }
 EXPORT_SYMBOL(tcf_hash_check);
 
-struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est, struct tc_action *a, int size, int bind, u32 *idx_gen, struct tcf_hashinfo *hinfo)
+struct tcf_common *tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a, int size, int bind, u32 *idx_gen, struct tcf_hashinfo *hinfo)
 {
 	struct tcf_common *p = kzalloc(size, GFP_KERNEL);
 
@@ -227,7 +227,7 @@ struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est, struct tc_acti
 	p->tcfc_tm.lastuse = jiffies;
 	if (est)
 		gen_new_estimator(&p->tcfc_bstats, &p->tcfc_rate_est,
-				  &p->tcfc_lock, (struct nlattr *)est);
+				  &p->tcfc_lock, est);
 	a->priv = (void *) p;
 	return p;
 }
@@ -305,15 +305,15 @@ static struct tc_action_ops *tc_lookup_action_n(char *kind)
 	return a;
 }
 
-/* lookup by rtattr */
-static struct tc_action_ops *tc_lookup_action(struct rtattr *kind)
+/* lookup by nlattr */
+static struct tc_action_ops *tc_lookup_action(struct nlattr *kind)
 {
 	struct tc_action_ops *a = NULL;
 
 	if (kind) {
 		read_lock(&act_mod_lock);
 		for (a = act_base; a; a = a->next) {
-			if (rtattr_strcmp(kind, a->kind) == 0) {
+			if (nla_strcmp(kind, a->kind) == 0) {
 				if (!try_module_get(a->owner)) {
 					read_unlock(&act_mod_lock);
 					return NULL;
@@ -414,22 +414,22 @@ tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 {
 	int err = -EINVAL;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *r;
+	struct nlattr *r;
 
 	if (a->ops == NULL || a->ops->dump == NULL)
 		return err;
 
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
+	NLA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
 	if (tcf_action_copy_stats(skb, a, 0))
-		goto rtattr_failure;
-	r = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
+		goto nla_put_failure;
+	r = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_OPTIONS, 0, NULL);
 	if ((err = tcf_action_dump_old(skb, a, bind, ref)) > 0) {
-		r->rta_len = skb_tail_pointer(skb) - (u8 *)r;
+		r->nla_len = skb_tail_pointer(skb) - (u8 *)r;
 		return err;
 	}
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -441,45 +441,45 @@ tcf_action_dump(struct sk_buff *skb, struct tc_action *act, int bind, int ref)
 	struct tc_action *a;
 	int err = -EINVAL;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *r ;
+	struct nlattr *r ;
 
 	while ((a = act) != NULL) {
-		r = (struct rtattr *)skb_tail_pointer(skb);
+		r = (struct nlattr *)skb_tail_pointer(skb);
 		act = a->next;
-		RTA_PUT(skb, a->order, 0, NULL);
+		NLA_PUT(skb, a->order, 0, NULL);
 		err = tcf_action_dump_1(skb, a, bind, ref);
 		if (err < 0)
 			goto errout;
-		r->rta_len = skb_tail_pointer(skb) - (u8 *)r;
+		r->nla_len = skb_tail_pointer(skb) - (u8 *)r;
 	}
 
 	return 0;
 
-rtattr_failure:
+nla_put_failure:
 	err = -EINVAL;
 errout:
 	nlmsg_trim(skb, b);
 	return err;
 }
 
-struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est,
+struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est,
 				    char *name, int ovr, int bind, int *err)
 {
 	struct tc_action *a;
 	struct tc_action_ops *a_o;
 	char act_name[IFNAMSIZ];
-	struct rtattr *tb[TCA_ACT_MAX+1];
-	struct rtattr *kind;
+	struct nlattr *tb[TCA_ACT_MAX+1];
+	struct nlattr *kind;
 
 	*err = -EINVAL;
 
 	if (name == NULL) {
-		if (rtattr_parse_nested(tb, TCA_ACT_MAX, rta) < 0)
+		if (nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL) < 0)
 			goto err_out;
-		kind = tb[TCA_ACT_KIND-1];
+		kind = tb[TCA_ACT_KIND];
 		if (kind == NULL)
 			goto err_out;
-		if (rtattr_strlcpy(act_name, kind, IFNAMSIZ) >= IFNAMSIZ)
+		if (nla_strlcpy(act_name, kind, IFNAMSIZ) >= IFNAMSIZ)
 			goto err_out;
 	} else {
 		if (strlcpy(act_name, name, IFNAMSIZ) >= IFNAMSIZ)
@@ -517,9 +517,9 @@ struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est,
 
 	/* backward compatibility for policer */
 	if (name == NULL)
-		*err = a_o->init(tb[TCA_ACT_OPTIONS-1], est, a, ovr, bind);
+		*err = a_o->init(tb[TCA_ACT_OPTIONS], est, a, ovr, bind);
 	else
-		*err = a_o->init(rta, est, a, ovr, bind);
+		*err = a_o->init(nla, est, a, ovr, bind);
 	if (*err < 0)
 		goto err_free;
 
@@ -542,23 +542,23 @@ err_out:
 	return NULL;
 }
 
-struct tc_action *tcf_action_init(struct rtattr *rta, struct rtattr *est,
+struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est,
 				  char *name, int ovr, int bind, int *err)
 {
-	struct rtattr *tb[TCA_ACT_MAX_PRIO+1];
+	struct nlattr *tb[TCA_ACT_MAX_PRIO+1];
 	struct tc_action *head = NULL, *act, *act_prev = NULL;
 	int i;
 
-	if (rtattr_parse_nested(tb, TCA_ACT_MAX_PRIO, rta) < 0) {
+	if (nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL) < 0) {
 		*err = -EINVAL;
 		return head;
 	}
 
-	for (i=0; i < TCA_ACT_MAX_PRIO && tb[i]; i++) {
+	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) {
 		act = tcf_action_init_1(tb[i], est, name, ovr, bind, err);
 		if (act == NULL)
 			goto err;
-		act->order = i+1;
+		act->order = i;
 
 		if (head == NULL)
 			head = act;
@@ -625,7 +625,7 @@ tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq,
 	struct tcamsg *t;
 	struct nlmsghdr *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *x;
+	struct nlattr *x;
 
 	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags);
 
@@ -634,18 +634,18 @@ tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq,
 	t->tca__pad1 = 0;
 	t->tca__pad2 = 0;
 
-	x = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_ACT_TAB, 0, NULL);
+	x = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_ACT_TAB, 0, NULL);
 
 	if (tcf_action_dump(skb, a, bind, ref) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	x->rta_len = skb_tail_pointer(skb) - (u8 *)x;
+	x->nla_len = skb_tail_pointer(skb) - (u8 *)x;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 nlmsg_failure:
 	nlmsg_trim(skb, b);
 	return -1;
@@ -668,20 +668,20 @@ act_get_notify(u32 pid, struct nlmsghdr *n, struct tc_action *a, int event)
 }
 
 static struct tc_action *
-tcf_action_get_1(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int *err)
+tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int *err)
 {
-	struct rtattr *tb[TCA_ACT_MAX+1];
+	struct nlattr *tb[TCA_ACT_MAX+1];
 	struct tc_action *a;
 	int index;
 
 	*err = -EINVAL;
-	if (rtattr_parse_nested(tb, TCA_ACT_MAX, rta) < 0)
+	if (nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL) < 0)
 		return NULL;
 
-	if (tb[TCA_ACT_INDEX - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_ACT_INDEX - 1]) < sizeof(index))
+	if (tb[TCA_ACT_INDEX] == NULL ||
+	    nla_len(tb[TCA_ACT_INDEX]) < sizeof(index))
 		return NULL;
-	index = *(int *)RTA_DATA(tb[TCA_ACT_INDEX - 1]);
+	index = *(int *)nla_data(tb[TCA_ACT_INDEX]);
 
 	*err = -ENOMEM;
 	a = kzalloc(sizeof(struct tc_action), GFP_KERNEL);
@@ -689,7 +689,7 @@ tcf_action_get_1(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int *err)
 		return NULL;
 
 	*err = -EINVAL;
-	a->ops = tc_lookup_action(tb[TCA_ACT_KIND - 1]);
+	a->ops = tc_lookup_action(tb[TCA_ACT_KIND]);
 	if (a->ops == NULL)
 		goto err_free;
 	if (a->ops->lookup == NULL)
@@ -731,16 +731,16 @@ static struct tc_action *create_a(int i)
 	return act;
 }
 
-static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid)
+static int tca_action_flush(struct nlattr *nla, struct nlmsghdr *n, u32 pid)
 {
 	struct sk_buff *skb;
 	unsigned char *b;
 	struct nlmsghdr *nlh;
 	struct tcamsg *t;
 	struct netlink_callback dcb;
-	struct rtattr *x;
-	struct rtattr *tb[TCA_ACT_MAX+1];
-	struct rtattr *kind;
+	struct nlattr *x;
+	struct nlattr *tb[TCA_ACT_MAX+1];
+	struct nlattr *kind;
 	struct tc_action *a = create_a(0);
 	int err = -EINVAL;
 
@@ -758,10 +758,10 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid)
 
 	b = skb_tail_pointer(skb);
 
-	if (rtattr_parse_nested(tb, TCA_ACT_MAX, rta) < 0)
+	if (nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL) < 0)
 		goto err_out;
 
-	kind = tb[TCA_ACT_KIND-1];
+	kind = tb[TCA_ACT_KIND];
 	a->ops = tc_lookup_action(kind);
 	if (a->ops == NULL)
 		goto err_out;
@@ -772,14 +772,14 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid)
 	t->tca__pad1 = 0;
 	t->tca__pad2 = 0;
 
-	x = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_ACT_TAB, 0, NULL);
+	x = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_ACT_TAB, 0, NULL);
 
 	err = a->ops->walk(skb, &dcb, RTM_DELACTION, a);
 	if (err < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	x->rta_len = skb_tail_pointer(skb) - (u8 *)x;
+	x->nla_len = skb_tail_pointer(skb) - (u8 *)x;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	nlh->nlmsg_flags |= NLM_F_ROOT;
@@ -791,7 +791,7 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid)
 
 	return err;
 
-rtattr_failure:
+nla_put_failure:
 nlmsg_failure:
 	module_put(a->ops->owner);
 err_out:
@@ -801,13 +801,13 @@ err_out:
 }
 
 static int
-tca_action_gd(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int event)
+tca_action_gd(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int event)
 {
 	int i, ret = 0;
-	struct rtattr *tb[TCA_ACT_MAX_PRIO+1];
+	struct nlattr *tb[TCA_ACT_MAX_PRIO+1];
 	struct tc_action *head = NULL, *act, *act_prev = NULL;
 
-	if (rtattr_parse_nested(tb, TCA_ACT_MAX_PRIO, rta) < 0)
+	if (nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL) < 0)
 		return -EINVAL;
 
 	if (event == RTM_DELACTION && n->nlmsg_flags&NLM_F_ROOT) {
@@ -815,11 +815,11 @@ tca_action_gd(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int event)
 			return tca_action_flush(tb[0], n, pid);
 	}
 
-	for (i=0; i < TCA_ACT_MAX_PRIO && tb[i]; i++) {
+	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) {
 		act = tcf_action_get_1(tb[i], n, pid, &ret);
 		if (act == NULL)
 			goto err;
-		act->order = i+1;
+		act->order = i;
 
 		if (head == NULL)
 			head = act;
@@ -865,7 +865,7 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event,
 	struct tcamsg *t;
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
-	struct rtattr *x;
+	struct nlattr *x;
 	unsigned char *b;
 	int err = 0;
 
@@ -881,13 +881,13 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event,
 	t->tca__pad1 = 0;
 	t->tca__pad2 = 0;
 
-	x = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_ACT_TAB, 0, NULL);
+	x = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_ACT_TAB, 0, NULL);
 
 	if (tcf_action_dump(skb, a, 0, 0) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
-	x->rta_len = skb_tail_pointer(skb) - (u8 *)x;
+	x->nla_len = skb_tail_pointer(skb) - (u8 *)x;
 
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	NETLINK_CB(skb).dst_group = RTNLGRP_TC;
@@ -897,7 +897,7 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event,
 		err = 0;
 	return err;
 
-rtattr_failure:
+nla_put_failure:
 nlmsg_failure:
 	kfree_skb(skb);
 	return -1;
@@ -905,14 +905,14 @@ nlmsg_failure:
 
 
 static int
-tcf_action_add(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int ovr)
+tcf_action_add(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int ovr)
 {
 	int ret = 0;
 	struct tc_action *act;
 	struct tc_action *a;
 	u32 seq = n->nlmsg_seq;
 
-	act = tcf_action_init(rta, NULL, NULL, ovr, 0, &ret);
+	act = tcf_action_init(nla, NULL, NULL, ovr, 0, &ret);
 	if (act == NULL)
 		goto done;
 
@@ -931,14 +931,18 @@ done:
 static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = skb->sk->sk_net;
-	struct rtattr **tca = arg;
+	struct nlattr *tca[TCA_ACT_MAX + 1];
 	u32 pid = skb ? NETLINK_CB(skb).pid : 0;
 	int ret = 0, ovr = 0;
 
 	if (net != &init_net)
 		return -EINVAL;
 
-	if (tca[TCA_ACT_TAB-1] == NULL) {
+	ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL);
+	if (ret < 0)
+		return ret;
+
+	if (tca[TCA_ACT_TAB] == NULL) {
 		printk("tc_ctl_action: received NO action attribs\n");
 		return -EINVAL;
 	}
@@ -956,15 +960,15 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 		if (n->nlmsg_flags&NLM_F_REPLACE)
 			ovr = 1;
 replay:
-		ret = tcf_action_add(tca[TCA_ACT_TAB-1], n, pid, ovr);
+		ret = tcf_action_add(tca[TCA_ACT_TAB], n, pid, ovr);
 		if (ret == -EAGAIN)
 			goto replay;
 		break;
 	case RTM_DELACTION:
-		ret = tca_action_gd(tca[TCA_ACT_TAB-1], n, pid, RTM_DELACTION);
+		ret = tca_action_gd(tca[TCA_ACT_TAB], n, pid, RTM_DELACTION);
 		break;
 	case RTM_GETACTION:
-		ret = tca_action_gd(tca[TCA_ACT_TAB-1], n, pid, RTM_GETACTION);
+		ret = tca_action_gd(tca[TCA_ACT_TAB], n, pid, RTM_GETACTION);
 		break;
 	default:
 		BUG();
@@ -973,33 +977,33 @@ replay:
 	return ret;
 }
 
-static struct rtattr *
+static struct nlattr *
 find_dump_kind(struct nlmsghdr *n)
 {
-	struct rtattr *tb1, *tb2[TCA_ACT_MAX+1];
-	struct rtattr *tb[TCA_ACT_MAX_PRIO + 1];
-	struct rtattr *rta[TCAA_MAX + 1];
-	struct rtattr *kind;
+	struct nlattr *tb1, *tb2[TCA_ACT_MAX+1];
+	struct nlattr *tb[TCA_ACT_MAX_PRIO + 1];
+	struct nlattr *nla[TCAA_MAX + 1];
+	struct nlattr *kind;
 	int min_len = NLMSG_LENGTH(sizeof(struct tcamsg));
 	int attrlen = n->nlmsg_len - NLMSG_ALIGN(min_len);
-	struct rtattr *attr = (void *) n + NLMSG_ALIGN(min_len);
+	struct nlattr *attr = (void *) n + NLMSG_ALIGN(min_len);
 
-	if (rtattr_parse(rta, TCAA_MAX, attr, attrlen) < 0)
+	if (nla_parse(nla, TCAA_MAX, attr, attrlen, NULL) < 0)
 		return NULL;
-	tb1 = rta[TCA_ACT_TAB - 1];
+	tb1 = nla[TCA_ACT_TAB];
 	if (tb1 == NULL)
 		return NULL;
 
-	if (rtattr_parse(tb, TCA_ACT_MAX_PRIO, RTA_DATA(tb1),
-			 NLMSG_ALIGN(RTA_PAYLOAD(tb1))) < 0)
+	if (nla_parse(tb, TCA_ACT_MAX_PRIO, nla_data(tb1),
+		      NLMSG_ALIGN(nla_len(tb1)), NULL) < 0)
 		return NULL;
 	if (tb[0] == NULL)
 		return NULL;
 
-	if (rtattr_parse(tb2, TCA_ACT_MAX, RTA_DATA(tb[0]),
-			 RTA_PAYLOAD(tb[0])) < 0)
+	if (nla_parse(tb2, TCA_ACT_MAX, nla_data(tb[0]),
+		      nla_len(tb[0]), NULL) < 0)
 		return NULL;
-	kind = tb2[TCA_ACT_KIND-1];
+	kind = tb2[TCA_ACT_KIND];
 
 	return kind;
 }
@@ -1010,12 +1014,12 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 	struct net *net = skb->sk->sk_net;
 	struct nlmsghdr *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
-	struct rtattr *x;
+	struct nlattr *x;
 	struct tc_action_ops *a_o;
 	struct tc_action a;
 	int ret = 0;
 	struct tcamsg *t = (struct tcamsg *) NLMSG_DATA(cb->nlh);
-	struct rtattr *kind = find_dump_kind(cb->nlh);
+	struct nlattr *kind = find_dump_kind(cb->nlh);
 
 	if (net != &init_net)
 		return 0;
@@ -1035,7 +1039,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 
 	if (a_o->walk == NULL) {
 		printk("tc_dump_action: %s !capable of dumping table\n", a_o->kind);
-		goto rtattr_failure;
+		goto nla_put_failure;
 	}
 
 	nlh = NLMSG_PUT(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq,
@@ -1045,15 +1049,15 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 	t->tca__pad1 = 0;
 	t->tca__pad2 = 0;
 
-	x = (struct rtattr *)skb_tail_pointer(skb);
-	RTA_PUT(skb, TCA_ACT_TAB, 0, NULL);
+	x = (struct nlattr *)skb_tail_pointer(skb);
+	NLA_PUT(skb, TCA_ACT_TAB, 0, NULL);
 
 	ret = a_o->walk(skb, cb, RTM_GETACTION, &a);
 	if (ret < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	if (ret > 0) {
-		x->rta_len = skb_tail_pointer(skb) - (u8 *)x;
+		x->nla_len = skb_tail_pointer(skb) - (u8 *)x;
 		ret = skb->len;
 	} else
 		nlmsg_trim(skb, x);
@@ -1064,7 +1068,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 	module_put(a_o->owner);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 nlmsg_failure:
 	module_put(a_o->owner);
 	nlmsg_trim(skb, b);
diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c
index a9631e426d91..5402cf885f95 100644
--- a/net/sched/act_gact.c
+++ b/net/sched/act_gact.c
@@ -53,26 +53,26 @@ typedef int (*g_rand)(struct tcf_gact *gact);
 static g_rand gact_rand[MAX_RAND]= { NULL, gact_net_rand, gact_determ };
 #endif /* CONFIG_GACT_PROB */
 
-static int tcf_gact_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_gact_init(struct nlattr *nla, struct nlattr *est,
 			 struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_GACT_MAX];
+	struct nlattr *tb[TCA_GACT_MAX + 1];
 	struct tc_gact *parm;
 	struct tcf_gact *gact;
 	struct tcf_common *pc;
 	int ret = 0;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_GACT_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_GACT_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_GACT_PARMS - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_GACT_PARMS - 1]) < sizeof(*parm))
+	if (tb[TCA_GACT_PARMS] == NULL ||
+	    nla_len(tb[TCA_GACT_PARMS]) < sizeof(*parm))
 		return -EINVAL;
-	parm = RTA_DATA(tb[TCA_GACT_PARMS - 1]);
+	parm = nla_data(tb[TCA_GACT_PARMS]);
 
-	if (tb[TCA_GACT_PROB-1] != NULL)
+	if (tb[TCA_GACT_PROB] != NULL)
 #ifdef CONFIG_GACT_PROB
-		if (RTA_PAYLOAD(tb[TCA_GACT_PROB-1]) < sizeof(struct tc_gact_p))
+		if (nla_len(tb[TCA_GACT_PROB]) < sizeof(struct tc_gact_p))
 			return -EINVAL;
 #else
 		return -EOPNOTSUPP;
@@ -97,8 +97,8 @@ static int tcf_gact_init(struct rtattr *rta, struct rtattr *est,
 	spin_lock_bh(&gact->tcf_lock);
 	gact->tcf_action = parm->action;
 #ifdef CONFIG_GACT_PROB
-	if (tb[TCA_GACT_PROB-1] != NULL) {
-		struct tc_gact_p *p_parm = RTA_DATA(tb[TCA_GACT_PROB-1]);
+	if (tb[TCA_GACT_PROB] != NULL) {
+		struct tc_gact_p *p_parm = nla_data(tb[TCA_GACT_PROB]);
 		gact->tcfg_paction = p_parm->paction;
 		gact->tcfg_pval    = p_parm->pval;
 		gact->tcfg_ptype   = p_parm->ptype;
@@ -154,23 +154,23 @@ static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int
 	opt.refcnt = gact->tcf_refcnt - ref;
 	opt.bindcnt = gact->tcf_bindcnt - bind;
 	opt.action = gact->tcf_action;
-	RTA_PUT(skb, TCA_GACT_PARMS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_GACT_PARMS, sizeof(opt), &opt);
 #ifdef CONFIG_GACT_PROB
 	if (gact->tcfg_ptype) {
 		struct tc_gact_p p_opt;
 		p_opt.paction = gact->tcfg_paction;
 		p_opt.pval = gact->tcfg_pval;
 		p_opt.ptype = gact->tcfg_ptype;
-		RTA_PUT(skb, TCA_GACT_PROB, sizeof(p_opt), &p_opt);
+		NLA_PUT(skb, TCA_GACT_PROB, sizeof(p_opt), &p_opt);
 	}
 #endif
 	t.install = jiffies_to_clock_t(jiffies - gact->tcf_tm.install);
 	t.lastuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.lastuse);
 	t.expires = jiffies_to_clock_t(gact->tcf_tm.expires);
-	RTA_PUT(skb, TCA_GACT_TM, sizeof(t), &t);
+	NLA_PUT(skb, TCA_GACT_TM, sizeof(t), &t);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index fa006e06ce33..fee5282637cc 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -92,10 +92,10 @@ static int tcf_ipt_release(struct tcf_ipt *ipt, int bind)
 	return ret;
 }
 
-static int tcf_ipt_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est,
 			struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_IPT_MAX];
+	struct nlattr *tb[TCA_IPT_MAX + 1];
 	struct tcf_ipt *ipt;
 	struct tcf_common *pc;
 	struct ipt_entry_target *td, *t;
@@ -104,22 +104,22 @@ static int tcf_ipt_init(struct rtattr *rta, struct rtattr *est,
 	u32 hook = 0;
 	u32 index = 0;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_IPT_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_IPT_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_IPT_HOOK-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_IPT_HOOK-1]) < sizeof(u32))
+	if (tb[TCA_IPT_HOOK] == NULL ||
+	    nla_len(tb[TCA_IPT_HOOK]) < sizeof(u32))
 		return -EINVAL;
-	if (tb[TCA_IPT_TARG-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_IPT_TARG-1]) < sizeof(*t))
+	if (tb[TCA_IPT_TARG] == NULL ||
+	    nla_len(tb[TCA_IPT_TARG]) < sizeof(*t))
 		return -EINVAL;
-	td = (struct ipt_entry_target *)RTA_DATA(tb[TCA_IPT_TARG-1]);
-	if (RTA_PAYLOAD(tb[TCA_IPT_TARG-1]) < td->u.target_size)
+	td = (struct ipt_entry_target *)nla_data(tb[TCA_IPT_TARG]);
+	if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size)
 		return -EINVAL;
 
-	if (tb[TCA_IPT_INDEX-1] != NULL &&
-	    RTA_PAYLOAD(tb[TCA_IPT_INDEX-1]) >= sizeof(u32))
-		index = *(u32 *)RTA_DATA(tb[TCA_IPT_INDEX-1]);
+	if (tb[TCA_IPT_INDEX] != NULL &&
+	    nla_len(tb[TCA_IPT_INDEX]) >= sizeof(u32))
+		index = *(u32 *)nla_data(tb[TCA_IPT_INDEX]);
 
 	pc = tcf_hash_check(index, a, bind, &ipt_hash_info);
 	if (!pc) {
@@ -136,14 +136,14 @@ static int tcf_ipt_init(struct rtattr *rta, struct rtattr *est,
 	}
 	ipt = to_ipt(pc);
 
-	hook = *(u32 *)RTA_DATA(tb[TCA_IPT_HOOK-1]);
+	hook = *(u32 *)nla_data(tb[TCA_IPT_HOOK]);
 
 	err = -ENOMEM;
 	tname = kmalloc(IFNAMSIZ, GFP_KERNEL);
 	if (unlikely(!tname))
 		goto err1;
-	if (tb[TCA_IPT_TABLE - 1] == NULL ||
-	    rtattr_strlcpy(tname, tb[TCA_IPT_TABLE-1], IFNAMSIZ) >= IFNAMSIZ)
+	if (tb[TCA_IPT_TABLE] == NULL ||
+	    nla_strlcpy(tname, tb[TCA_IPT_TABLE], IFNAMSIZ) >= IFNAMSIZ)
 		strcpy(tname, "mangle");
 
 	t = kmemdup(td, td->u.target_size, GFP_KERNEL);
@@ -243,25 +243,25 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int
 
 	t = kmemdup(ipt->tcfi_t, ipt->tcfi_t->u.user.target_size, GFP_ATOMIC);
 	if (unlikely(!t))
-		goto rtattr_failure;
+		goto nla_put_failure;
 
 	c.bindcnt = ipt->tcf_bindcnt - bind;
 	c.refcnt = ipt->tcf_refcnt - ref;
 	strcpy(t->u.user.name, ipt->tcfi_t->u.kernel.target->name);
 
-	RTA_PUT(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t);
-	RTA_PUT(skb, TCA_IPT_INDEX, 4, &ipt->tcf_index);
-	RTA_PUT(skb, TCA_IPT_HOOK, 4, &ipt->tcfi_hook);
-	RTA_PUT(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c);
-	RTA_PUT(skb, TCA_IPT_TABLE, IFNAMSIZ, ipt->tcfi_tname);
+	NLA_PUT(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t);
+	NLA_PUT(skb, TCA_IPT_INDEX, 4, &ipt->tcf_index);
+	NLA_PUT(skb, TCA_IPT_HOOK, 4, &ipt->tcfi_hook);
+	NLA_PUT(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c);
+	NLA_PUT(skb, TCA_IPT_TABLE, IFNAMSIZ, ipt->tcfi_tname);
 	tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install);
 	tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse);
 	tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires);
-	RTA_PUT(skb, TCA_IPT_TM, sizeof (tm), &tm);
+	NLA_PUT(skb, TCA_IPT_TM, sizeof (tm), &tm);
 	kfree(t);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	kfree(t);
 	return -1;
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index c3fde9180f9d..db943a8c7388 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -54,10 +54,10 @@ static inline int tcf_mirred_release(struct tcf_mirred *m, int bind)
 	return 0;
 }
 
-static int tcf_mirred_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_mirred_init(struct nlattr *nla, struct nlattr *est,
 			   struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_MIRRED_MAX];
+	struct nlattr *tb[TCA_MIRRED_MAX + 1];
 	struct tc_mirred *parm;
 	struct tcf_mirred *m;
 	struct tcf_common *pc;
@@ -65,13 +65,13 @@ static int tcf_mirred_init(struct rtattr *rta, struct rtattr *est,
 	int ret = 0;
 	int ok_push = 0;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_MIRRED_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_MIRRED_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_MIRRED_PARMS-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_MIRRED_PARMS-1]) < sizeof(*parm))
+	if (tb[TCA_MIRRED_PARMS] == NULL ||
+	    nla_len(tb[TCA_MIRRED_PARMS]) < sizeof(*parm))
 		return -EINVAL;
-	parm = RTA_DATA(tb[TCA_MIRRED_PARMS-1]);
+	parm = nla_data(tb[TCA_MIRRED_PARMS]);
 
 	if (parm->ifindex) {
 		dev = __dev_get_by_index(&init_net, parm->ifindex);
@@ -207,14 +207,14 @@ static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, i
 	opt.bindcnt = m->tcf_bindcnt - bind;
 	opt.eaction = m->tcfm_eaction;
 	opt.ifindex = m->tcfm_ifindex;
-	RTA_PUT(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt);
 	t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install);
 	t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse);
 	t.expires = jiffies_to_clock_t(m->tcf_tm.expires);
-	RTA_PUT(skb, TCA_MIRRED_TM, sizeof(t), &t);
+	NLA_PUT(skb, TCA_MIRRED_TM, sizeof(t), &t);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c
index da5c1eae4222..be007bb31b59 100644
--- a/net/sched/act_nat.c
+++ b/net/sched/act_nat.c
@@ -40,22 +40,22 @@ static struct tcf_hashinfo nat_hash_info = {
 	.lock	=	&nat_lock,
 };
 
-static int tcf_nat_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_nat_init(struct nlattr *nla, struct nlattr *est,
 			struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_NAT_MAX];
+	struct nlattr *tb[TCA_NAT_MAX + 1];
 	struct tc_nat *parm;
 	int ret = 0;
 	struct tcf_nat *p;
 	struct tcf_common *pc;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_NAT_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_NAT_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_NAT_PARMS - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_NAT_PARMS - 1]) < sizeof(*parm))
+	if (tb[TCA_NAT_PARMS] == NULL ||
+	    nla_len(tb[TCA_NAT_PARMS]) < sizeof(*parm))
 		return -EINVAL;
-	parm = RTA_DATA(tb[TCA_NAT_PARMS - 1]);
+	parm = nla_data(tb[TCA_NAT_PARMS]);
 
 	pc = tcf_hash_check(parm->index, a, bind, &nat_hash_info);
 	if (!pc) {
@@ -275,17 +275,17 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a,
 	opt->refcnt = p->tcf_refcnt - ref;
 	opt->bindcnt = p->tcf_bindcnt - bind;
 
-	RTA_PUT(skb, TCA_NAT_PARMS, s, opt);
+	NLA_PUT(skb, TCA_NAT_PARMS, s, opt);
 	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
 	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
 	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
-	RTA_PUT(skb, TCA_NAT_TM, sizeof(t), &t);
+	NLA_PUT(skb, TCA_NAT_TM, sizeof(t), &t);
 
 	kfree(opt);
 
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	kfree(opt);
 	return -1;
diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index b46fab5fb323..88d8a15a1921 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -33,10 +33,10 @@ static struct tcf_hashinfo pedit_hash_info = {
 	.lock	=	&pedit_lock,
 };
 
-static int tcf_pedit_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_pedit_init(struct nlattr *nla, struct nlattr *est,
 			  struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_PEDIT_MAX];
+	struct nlattr *tb[TCA_PEDIT_MAX + 1];
 	struct tc_pedit *parm;
 	int ret = 0;
 	struct tcf_pedit *p;
@@ -44,15 +44,15 @@ static int tcf_pedit_init(struct rtattr *rta, struct rtattr *est,
 	struct tc_pedit_key *keys = NULL;
 	int ksize;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_PEDIT_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_PEDIT_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_PEDIT_PARMS - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_PEDIT_PARMS-1]) < sizeof(*parm))
+	if (tb[TCA_PEDIT_PARMS] == NULL ||
+	    nla_len(tb[TCA_PEDIT_PARMS]) < sizeof(*parm))
 		return -EINVAL;
-	parm = RTA_DATA(tb[TCA_PEDIT_PARMS-1]);
+	parm = nla_data(tb[TCA_PEDIT_PARMS]);
 	ksize = parm->nkeys * sizeof(struct tc_pedit_key);
-	if (RTA_PAYLOAD(tb[TCA_PEDIT_PARMS-1]) < sizeof(*parm) + ksize)
+	if (nla_len(tb[TCA_PEDIT_PARMS]) < sizeof(*parm) + ksize)
 		return -EINVAL;
 
 	pc = tcf_hash_check(parm->index, a, bind, &pedit_hash_info);
@@ -206,15 +206,15 @@ static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a,
 	opt->refcnt = p->tcf_refcnt - ref;
 	opt->bindcnt = p->tcf_bindcnt - bind;
 
-	RTA_PUT(skb, TCA_PEDIT_PARMS, s, opt);
+	NLA_PUT(skb, TCA_PEDIT_PARMS, s, opt);
 	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
 	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
 	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
-	RTA_PUT(skb, TCA_PEDIT_TM, sizeof(t), &t);
+	NLA_PUT(skb, TCA_PEDIT_TM, sizeof(t), &t);
 	kfree(opt);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	kfree(opt);
 	return -1;
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 07ffdf9c5e59..3af5759aac26 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -54,7 +54,7 @@ static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *c
 {
 	struct tcf_common *p;
 	int err = 0, index = -1, i = 0, s_i = 0, n_i = 0;
-	struct rtattr *r;
+	struct nlattr *r;
 
 	read_lock_bh(&police_lock);
 
@@ -69,8 +69,8 @@ static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *c
 				continue;
 			a->priv = p;
 			a->order = index;
-			r = (struct rtattr *)skb_tail_pointer(skb);
-			RTA_PUT(skb, a->order, 0, NULL);
+			r = (struct nlattr *)skb_tail_pointer(skb);
+			NLA_PUT(skb, a->order, 0, NULL);
 			if (type == RTM_DELACTION)
 				err = tcf_action_dump_1(skb, a, 0, 1);
 			else
@@ -80,7 +80,7 @@ static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *c
 				nlmsg_trim(skb, r);
 				goto done;
 			}
-			r->rta_len = skb_tail_pointer(skb) - (u8 *)r;
+			r->nla_len = skb_tail_pointer(skb) - (u8 *)r;
 			n_i++;
 		}
 	}
@@ -90,7 +90,7 @@ done:
 		cb->args[0] += n_i;
 	return n_i;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, r);
 	goto done;
 }
@@ -118,32 +118,32 @@ static void tcf_police_destroy(struct tcf_police *p)
 	BUG_TRAP(0);
 }
 
-static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
+static int tcf_act_police_locate(struct nlattr *nla, struct nlattr *est,
 				 struct tc_action *a, int ovr, int bind)
 {
 	unsigned h;
 	int ret = 0, err;
-	struct rtattr *tb[TCA_POLICE_MAX];
+	struct nlattr *tb[TCA_POLICE_MAX + 1];
 	struct tc_police *parm;
 	struct tcf_police *police;
 	struct qdisc_rate_table *R_tab = NULL, *P_tab = NULL;
 	int size;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_POLICE_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_POLICE_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_POLICE_TBF-1] == NULL)
+	if (tb[TCA_POLICE_TBF] == NULL)
 		return -EINVAL;
-	size = RTA_PAYLOAD(tb[TCA_POLICE_TBF-1]);
+	size = nla_len(tb[TCA_POLICE_TBF]);
 	if (size != sizeof(*parm) && size != sizeof(struct tc_police_compat))
 		return -EINVAL;
-	parm = RTA_DATA(tb[TCA_POLICE_TBF-1]);
+	parm = nla_data(tb[TCA_POLICE_TBF]);
 
-	if (tb[TCA_POLICE_RESULT-1] != NULL &&
-	    RTA_PAYLOAD(tb[TCA_POLICE_RESULT-1]) != sizeof(u32))
+	if (tb[TCA_POLICE_RESULT] != NULL &&
+	    nla_len(tb[TCA_POLICE_RESULT]) != sizeof(u32))
 		return -EINVAL;
-	if (tb[TCA_POLICE_RESULT-1] != NULL &&
-	    RTA_PAYLOAD(tb[TCA_POLICE_RESULT-1]) != sizeof(u32))
+	if (tb[TCA_POLICE_RESULT] != NULL &&
+	    nla_len(tb[TCA_POLICE_RESULT]) != sizeof(u32))
 		return -EINVAL;
 
 	if (parm->index) {
@@ -174,12 +174,12 @@ static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
 override:
 	if (parm->rate.rate) {
 		err = -ENOMEM;
-		R_tab = qdisc_get_rtab(&parm->rate, (struct nlattr *)tb[TCA_POLICE_RATE-1]);
+		R_tab = qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE]);
 		if (R_tab == NULL)
 			goto failure;
 		if (parm->peakrate.rate) {
 			P_tab = qdisc_get_rtab(&parm->peakrate,
-					       (struct nlattr *)tb[TCA_POLICE_PEAKRATE-1]);
+					       tb[TCA_POLICE_PEAKRATE]);
 			if (P_tab == NULL) {
 				qdisc_put_rtab(R_tab);
 				goto failure;
@@ -197,8 +197,8 @@ override:
 		police->tcfp_P_tab = P_tab;
 	}
 
-	if (tb[TCA_POLICE_RESULT-1])
-		police->tcfp_result = *(u32*)RTA_DATA(tb[TCA_POLICE_RESULT-1]);
+	if (tb[TCA_POLICE_RESULT])
+		police->tcfp_result = *(u32*)nla_data(tb[TCA_POLICE_RESULT]);
 	police->tcfp_toks = police->tcfp_burst = parm->burst;
 	police->tcfp_mtu = parm->mtu;
 	if (police->tcfp_mtu == 0) {
@@ -210,13 +210,13 @@ override:
 		police->tcfp_ptoks = L2T_P(police, police->tcfp_mtu);
 	police->tcf_action = parm->action;
 
-	if (tb[TCA_POLICE_AVRATE-1])
+	if (tb[TCA_POLICE_AVRATE])
 		police->tcfp_ewma_rate =
-			*(u32*)RTA_DATA(tb[TCA_POLICE_AVRATE-1]);
+			*(u32*)nla_data(tb[TCA_POLICE_AVRATE]);
 	if (est)
 		gen_replace_estimator(&police->tcf_bstats,
 				      &police->tcf_rate_est,
-				      &police->tcf_lock, (struct nlattr *)est);
+				      &police->tcf_lock, est);
 
 	spin_unlock_bh(&police->tcf_lock);
 	if (ret != ACT_P_CREATED)
@@ -332,15 +332,15 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 		opt.peakrate = police->tcfp_P_tab->rate;
 	else
 		memset(&opt.peakrate, 0, sizeof(opt.peakrate));
-	RTA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt);
 	if (police->tcfp_result)
-		RTA_PUT(skb, TCA_POLICE_RESULT, sizeof(int),
+		NLA_PUT(skb, TCA_POLICE_RESULT, sizeof(int),
 			&police->tcfp_result);
 	if (police->tcfp_ewma_rate)
-		RTA_PUT(skb, TCA_POLICE_AVRATE, 4, &police->tcfp_ewma_rate);
+		NLA_PUT(skb, TCA_POLICE_AVRATE, 4, &police->tcfp_ewma_rate);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c
index fb84ef33d14f..d3226e24070b 100644
--- a/net/sched/act_simple.c
+++ b/net/sched/act_simple.c
@@ -84,10 +84,10 @@ static int realloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
 	return alloc_defdata(d, datalen, defdata);
 }
 
-static int tcf_simp_init(struct rtattr *rta, struct rtattr *est,
+static int tcf_simp_init(struct nlattr *nla, struct nlattr *est,
 			 struct tc_action *a, int ovr, int bind)
 {
-	struct rtattr *tb[TCA_DEF_MAX];
+	struct nlattr *tb[TCA_DEF_MAX + 1];
 	struct tc_defact *parm;
 	struct tcf_defact *d;
 	struct tcf_common *pc;
@@ -95,19 +95,19 @@ static int tcf_simp_init(struct rtattr *rta, struct rtattr *est,
 	u32 datalen = 0;
 	int ret = 0;
 
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_DEF_MAX, rta) < 0)
+	if (nla == NULL || nla_parse_nested(tb, TCA_DEF_MAX, nla, NULL) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_DEF_PARMS - 1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_DEF_PARMS - 1]) < sizeof(*parm))
+	if (tb[TCA_DEF_PARMS] == NULL ||
+	    nla_len(tb[TCA_DEF_PARMS]) < sizeof(*parm))
 		return -EINVAL;
 
-	parm = RTA_DATA(tb[TCA_DEF_PARMS - 1]);
-	defdata = RTA_DATA(tb[TCA_DEF_DATA - 1]);
+	parm = nla_data(tb[TCA_DEF_PARMS]);
+	defdata = nla_data(tb[TCA_DEF_DATA]);
 	if (defdata == NULL)
 		return -EINVAL;
 
-	datalen = RTA_PAYLOAD(tb[TCA_DEF_DATA - 1]);
+	datalen = nla_len(tb[TCA_DEF_DATA]);
 	if (datalen <= 0)
 		return -EINVAL;
 
@@ -164,15 +164,15 @@ static inline int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a,
 	opt.refcnt = d->tcf_refcnt - ref;
 	opt.bindcnt = d->tcf_bindcnt - bind;
 	opt.action = d->tcf_action;
-	RTA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt);
-	RTA_PUT(skb, TCA_DEF_DATA, d->tcfd_datalen, d->tcfd_defdata);
+	NLA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt);
+	NLA_PUT(skb, TCA_DEF_DATA, d->tcfd_datalen, d->tcfd_defdata);
 	t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install);
 	t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse);
 	t.expires = jiffies_to_clock_t(d->tcf_tm.expires);
-	RTA_PUT(skb, TCA_DEF_TM, sizeof(t), &t);
+	NLA_PUT(skb, TCA_DEF_TM, sizeof(t), &t);
 	return skb->len;
 
-rtattr_failure:
+nla_put_failure:
 	nlmsg_trim(skb, b);
 	return -1;
 }
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 87be2b2fc29a..d870478e3640 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -491,8 +491,7 @@ int tcf_exts_validate(struct tcf_proto *tp, struct nlattr **tb,
 		struct tc_action *act;
 
 		if (map->police && tb[map->police]) {
-			act = tcf_action_init_1((struct rtattr *)tb[map->police],
-						(struct rtattr *)rate_tlv,
+			act = tcf_action_init_1(tb[map->police], rate_tlv,
 						"police", TCA_ACT_NOREPLACE,
 						TCA_ACT_BIND, &err);
 			if (act == NULL)
@@ -501,8 +500,7 @@ int tcf_exts_validate(struct tcf_proto *tp, struct nlattr **tb,
 			act->type = TCA_OLD_COMPAT;
 			exts->action = act;
 		} else if (map->action && tb[map->action]) {
-			act = tcf_action_init((struct rtattr *)tb[map->action],
-					      (struct rtattr *)rate_tlv, NULL,
+			act = tcf_action_init(tb[map->action], rate_tlv, NULL,
 				TCA_ACT_NOREPLACE, TCA_ACT_BIND, &err);
 			if (act == NULL)
 				return err;
-- 
cgit v1.2.3


From eee80592c3c1f7381c04913d9d3eb6e3c3c87628 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 23:49:35 -0800
Subject: [NETNS]: Correct namespace for connect-time routing.

ip_route_connect and ip_route_newports are a part of routing API
presented to the socket layer. The namespace is available inside them
through a socket.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index d9b876adbf42..1985d820edea 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -33,6 +33,7 @@
 #include <linux/ip.h>
 #include <linux/cache.h>
 #include <linux/security.h>
+#include <net/sock.h>
 
 #ifndef __KERNEL__
 #warning This file is not supposed to be used outside of kernel.
@@ -157,8 +158,9 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 					 .dport = dport } } };
 
 	int err;
+	struct net *net = sk->sk_net;
 	if (!dst || !src) {
-		err = __ip_route_output_key(&init_net, rp, &fl);
+		err = __ip_route_output_key(net, rp, &fl);
 		if (err)
 			return err;
 		fl.fl4_dst = (*rp)->rt_dst;
@@ -167,7 +169,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 		*rp = NULL;
 	}
 	security_sk_classify_flow(sk, &fl);
-	return ip_route_output_flow(&init_net, rp, &fl, sk, flags);
+	return ip_route_output_flow(net, rp, &fl, sk, flags);
 }
 
 static inline int ip_route_newports(struct rtable **rp, u8 protocol,
@@ -184,7 +186,7 @@ static inline int ip_route_newports(struct rtable **rp, u8 protocol,
 		ip_rt_put(*rp);
 		*rp = NULL;
 		security_sk_classify_flow(sk, &fl);
-		return ip_route_output_flow(&init_net, rp, &fl, sk, 0);
+		return ip_route_output_flow(sk->sk_net, rp, &fl, sk, 0);
 	}
 	return 0;
 }
-- 
cgit v1.2.3


From b5921910a1de4ba82add59154976c3dc7352c8c2 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Tue, 22 Jan 2008 23:50:25 -0800
Subject: [NETNS]: Routing cache virtualization.

Basically, this piece looks relatively easy. Namespace is already
available on the dst entry via device and the device is safe to
dereferrence. Compare it with one of a searcher and skip entry if
appropriate.

The only exception is ip_rt_frag_needed. So, add namespace parameter to it.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h |  2 +-
 net/ipv4/icmp.c     |  2 +-
 net/ipv4/route.c    | 21 ++++++++++++++++-----
 3 files changed, 18 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 1985d820edea..4eabf008413b 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -115,7 +115,7 @@ extern int		__ip_route_output_key(struct net *, struct rtable **, const struct f
 extern int		ip_route_output_key(struct net *, struct rtable **, struct flowi *flp);
 extern int		ip_route_output_flow(struct net *, struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
 extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, struct net_device *devin);
-extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
+extern unsigned short	ip_rt_frag_needed(struct net *net, struct iphdr *iph, unsigned short new_mtu);
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
 extern unsigned		inet_addr_type(struct net *net, __be32 addr);
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index a142f19fec45..63ffc7d86f98 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -696,7 +696,7 @@ static void icmp_unreach(struct sk_buff *skb)
 							 "and DF set.\n",
 					       NIPQUAD(iph->daddr));
 			} else {
-				info = ip_rt_frag_needed(iph,
+				info = ip_rt_frag_needed(&init_net, iph,
 						     ntohs(icmph->un.frag.mtu));
 				if (!info)
 					goto out;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 39a40342142f..896c768e41a2 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -648,6 +648,11 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2)
 		(fl1->iif ^ fl2->iif)) == 0;
 }
 
+static inline int compare_netns(struct rtable *rt1, struct rtable *rt2)
+{
+	return rt1->u.dst.dev->nd_net == rt2->u.dst.dev->nd_net;
+}
+
 /*
  * Perform a full scan of hash table and free all entries.
  * Can be called by a softirq or a process.
@@ -961,7 +966,7 @@ restart:
 
 	spin_lock_bh(rt_hash_lock_addr(hash));
 	while ((rth = *rthp) != NULL) {
-		if (compare_keys(&rth->fl, &rt->fl)) {
+		if (compare_keys(&rth->fl, &rt->fl) && compare_netns(rth, rt)) {
 			/* Put it first */
 			*rthp = rth->u.dst.rt_next;
 			/*
@@ -1415,7 +1420,8 @@ static __inline__ unsigned short guess_mtu(unsigned short old_mtu)
 	return 68;
 }
 
-unsigned short ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu)
+unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
+				 unsigned short new_mtu)
 {
 	int i;
 	unsigned short old_mtu = ntohs(iph->tot_len);
@@ -1438,7 +1444,8 @@ unsigned short ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu)
 			    rth->rt_dst  == daddr &&
 			    rth->rt_src  == iph->saddr &&
 			    rth->fl.iif == 0 &&
-			    !(dst_metric_locked(&rth->u.dst, RTAX_MTU))) {
+			    !(dst_metric_locked(&rth->u.dst, RTAX_MTU)) &&
+			    rth->u.dst.dev->nd_net == net) {
 				unsigned short mtu = new_mtu;
 
 				if (new_mtu < 68 || new_mtu >= old_mtu) {
@@ -2049,7 +2056,9 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	struct rtable * rth;
 	unsigned	hash;
 	int iif = dev->ifindex;
+	struct net *net;
 
+	net = skb->dev->nd_net;
 	tos &= IPTOS_RT_MASK;
 	hash = rt_hash(daddr, saddr, iif);
 
@@ -2061,7 +2070,8 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		    rth->fl.iif == iif &&
 		    rth->fl.oif == 0 &&
 		    rth->fl.mark == skb->mark &&
-		    rth->fl.fl4_tos == tos) {
+		    rth->fl.fl4_tos == tos &&
+		    rth->u.dst.dev->nd_net == net) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(in_hit);
 			rcu_read_unlock();
@@ -2460,7 +2470,8 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
 		    rth->fl.oif == flp->oif &&
 		    rth->fl.mark == flp->mark &&
 		    !((rth->fl.fl4_tos ^ flp->fl4_tos) &
-			    (IPTOS_RT_MASK | RTO_ONLINK))) {
+			    (IPTOS_RT_MASK | RTO_ONLINK)) &&
+		    rth->u.dst.dev->nd_net == net) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(out_hit);
 			rcu_read_unlock_bh();
-- 
cgit v1.2.3


From ab27cfb85c5778400740ad0c401bde65616774eb Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 23 Jan 2008 20:33:13 -0800
Subject: [NET_SCHED]: act_api: use PTR_ERR in tcf_action_init/tcf_action_get

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/act_api.h |  4 +--
 net/sched/act_api.c   | 71 ++++++++++++++++++++++++++++-----------------------
 net/sched/cls_api.c   | 14 +++++-----
 3 files changed, 48 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/include/net/act_api.h b/include/net/act_api.h
index c5ac61ada94f..565eed8fe496 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -114,8 +114,8 @@ extern int tcf_register_action(struct tc_action_ops *a);
 extern int tcf_unregister_action(struct tc_action_ops *a);
 extern void tcf_action_destroy(struct tc_action *a, int bind);
 extern int tcf_action_exec(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res);
-extern struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind, int *err);
-extern struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind, int *err);
+extern struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind);
+extern struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est, char *n, int ovr, int bind);
 extern int tcf_action_dump(struct sk_buff *skb, struct tc_action *a, int, int);
 extern int tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int, int);
 extern int tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int, int);
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 2fe0345ddcb1..ea80f82dbb6a 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -18,6 +18,7 @@
 #include <linux/skbuff.h>
 #include <linux/init.h>
 #include <linux/kmod.h>
+#include <linux/err.h>
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/sch_generic.h>
@@ -463,15 +464,16 @@ errout:
 }
 
 struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est,
-				    char *name, int ovr, int bind, int *err)
+				    char *name, int ovr, int bind)
 {
 	struct tc_action *a;
 	struct tc_action_ops *a_o;
 	char act_name[IFNAMSIZ];
 	struct nlattr *tb[TCA_ACT_MAX+1];
 	struct nlattr *kind;
+	int err;
 
-	*err = -EINVAL;
+	err = -EINVAL;
 
 	if (name == NULL) {
 		if (nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL) < 0)
@@ -502,36 +504,35 @@ struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est,
 		 * indicate this using -EAGAIN.
 		 */
 		if (a_o != NULL) {
-			*err = -EAGAIN;
+			err = -EAGAIN;
 			goto err_mod;
 		}
 #endif
-		*err = -ENOENT;
+		err = -ENOENT;
 		goto err_out;
 	}
 
-	*err = -ENOMEM;
+	err = -ENOMEM;
 	a = kzalloc(sizeof(*a), GFP_KERNEL);
 	if (a == NULL)
 		goto err_mod;
 
 	/* backward compatibility for policer */
 	if (name == NULL)
-		*err = a_o->init(tb[TCA_ACT_OPTIONS], est, a, ovr, bind);
+		err = a_o->init(tb[TCA_ACT_OPTIONS], est, a, ovr, bind);
 	else
-		*err = a_o->init(nla, est, a, ovr, bind);
-	if (*err < 0)
+		err = a_o->init(nla, est, a, ovr, bind);
+	if (err < 0)
 		goto err_free;
 
 	/* module count goes up only when brand new policy is created
 	   if it exists and is only bound to in a_o->init() then
 	   ACT_P_CREATED is not returned (a zero is).
 	*/
-	if (*err != ACT_P_CREATED)
+	if (err != ACT_P_CREATED)
 		module_put(a_o->owner);
 	a->ops = a_o;
 
-	*err = 0;
 	return a;
 
 err_free:
@@ -539,24 +540,22 @@ err_free:
 err_mod:
 	module_put(a_o->owner);
 err_out:
-	return NULL;
+	return ERR_PTR(err);
 }
 
 struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est,
-				  char *name, int ovr, int bind, int *err)
+				  char *name, int ovr, int bind)
 {
 	struct nlattr *tb[TCA_ACT_MAX_PRIO+1];
 	struct tc_action *head = NULL, *act, *act_prev = NULL;
 	int i;
 
-	if (nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL) < 0) {
-		*err = -EINVAL;
-		return head;
-	}
+	if (nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL) < 0)
+		return ERR_PTR(-EINVAL);
 
 	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) {
-		act = tcf_action_init_1(tb[i], est, name, ovr, bind, err);
-		if (act == NULL)
+		act = tcf_action_init_1(tb[i], est, name, ovr, bind);
+		if (IS_ERR(act))
 			goto err;
 		act->order = i;
 
@@ -571,7 +570,7 @@ struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est,
 err:
 	if (head != NULL)
 		tcf_action_destroy(head, bind);
-	return NULL;
+	return act;
 }
 
 int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
@@ -668,44 +667,46 @@ act_get_notify(u32 pid, struct nlmsghdr *n, struct tc_action *a, int event)
 }
 
 static struct tc_action *
-tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int *err)
+tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 pid)
 {
 	struct nlattr *tb[TCA_ACT_MAX+1];
 	struct tc_action *a;
 	int index;
+	int err;
 
-	*err = -EINVAL;
+	err = -EINVAL;
 	if (nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL) < 0)
-		return NULL;
+		goto err_out;
 
 	if (tb[TCA_ACT_INDEX] == NULL ||
 	    nla_len(tb[TCA_ACT_INDEX]) < sizeof(index))
-		return NULL;
+		goto err_out;
 	index = *(int *)nla_data(tb[TCA_ACT_INDEX]);
 
-	*err = -ENOMEM;
+	err = -ENOMEM;
 	a = kzalloc(sizeof(struct tc_action), GFP_KERNEL);
 	if (a == NULL)
-		return NULL;
+		goto err_out;
 
-	*err = -EINVAL;
+	err = -EINVAL;
 	a->ops = tc_lookup_action(tb[TCA_ACT_KIND]);
 	if (a->ops == NULL)
 		goto err_free;
 	if (a->ops->lookup == NULL)
 		goto err_mod;
-	*err = -ENOENT;
+	err = -ENOENT;
 	if (a->ops->lookup(a, index) == 0)
 		goto err_mod;
 
 	module_put(a->ops->owner);
-	*err = 0;
 	return a;
+
 err_mod:
 	module_put(a->ops->owner);
 err_free:
 	kfree(a);
-	return NULL;
+err_out:
+	return ERR_PTR(err);
 }
 
 static void cleanup_a(struct tc_action *act)
@@ -816,9 +817,11 @@ tca_action_gd(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int event)
 	}
 
 	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) {
-		act = tcf_action_get_1(tb[i], n, pid, &ret);
-		if (act == NULL)
+		act = tcf_action_get_1(tb[i], n, pid);
+		if (IS_ERR(act)) {
+			ret = PTR_ERR(act);
 			goto err;
+		}
 		act->order = i;
 
 		if (head == NULL)
@@ -912,9 +915,13 @@ tcf_action_add(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int ovr)
 	struct tc_action *a;
 	u32 seq = n->nlmsg_seq;
 
-	act = tcf_action_init(nla, NULL, NULL, ovr, 0, &ret);
+	act = tcf_action_init(nla, NULL, NULL, ovr, 0);
 	if (act == NULL)
 		goto done;
+	if (IS_ERR(act)) {
+		ret = PTR_ERR(act);
+		goto done;
+	}
 
 	/* dump then free all the actions after update; inserted policy
 	 * stays intact
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index d870478e3640..92fa1559c211 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -23,6 +23,7 @@
 #include <linux/init.h>
 #include <linux/kmod.h>
 #include <linux/netlink.h>
+#include <linux/err.h>
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/netlink.h>
@@ -487,23 +488,22 @@ int tcf_exts_validate(struct tcf_proto *tp, struct nlattr **tb,
 
 #ifdef CONFIG_NET_CLS_ACT
 	{
-		int err;
 		struct tc_action *act;
 
 		if (map->police && tb[map->police]) {
 			act = tcf_action_init_1(tb[map->police], rate_tlv,
 						"police", TCA_ACT_NOREPLACE,
-						TCA_ACT_BIND, &err);
-			if (act == NULL)
-				return err;
+						TCA_ACT_BIND);
+			if (IS_ERR(act))
+				return PTR_ERR(act);
 
 			act->type = TCA_OLD_COMPAT;
 			exts->action = act;
 		} else if (map->action && tb[map->action]) {
 			act = tcf_action_init(tb[map->action], rate_tlv, NULL,
-				TCA_ACT_NOREPLACE, TCA_ACT_BIND, &err);
-			if (act == NULL)
-				return err;
+					      TCA_ACT_NOREPLACE, TCA_ACT_BIND);
+			if (IS_ERR(act))
+				return PTR_ERR(act);
 
 			exts->action = act;
 		}
-- 
cgit v1.2.3


From 5feb5e1aaa887f6427b8290bce48bfb6b7010fc6 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 23 Jan 2008 20:35:19 -0800
Subject: [NET_SCHED]: sch_api: introduce constant for rate table size

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_sched.h | 2 ++
 net/sched/sch_api.c       | 3 ++-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index 919af93b7059..32761352e858 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -83,6 +83,8 @@ struct tc_ratespec
 	__u32		rate;
 };
 
+#define TC_RTAB_SIZE	1024
+
 /* FIFO section */
 
 struct tc_fifo_qopt
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 8db554d54854..7e3c048ba9b1 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -244,7 +244,8 @@ struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *ta
 		}
 	}
 
-	if (tab == NULL || r->rate == 0 || r->cell_log == 0 || nla_len(tab) != 1024)
+	if (tab == NULL || r->rate == 0 || r->cell_log == 0 ||
+	    nla_len(tab) != TC_RTAB_SIZE)
 		return NULL;
 
 	rtab = kmalloc(sizeof(*rtab), GFP_KERNEL);
-- 
cgit v1.2.3


From afc7cbca5bfd556c3e12d3acefbee5ab0cbd4670 Mon Sep 17 00:00:00 2001
From: Takashi Sato <sho@tnes.nec.co.jp>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4:  Support large blocksize up to PAGESIZE

This patch set supports large block size(>4k, <=64k) in ext4,
just enlarging the block size limit. But it is NOT possible to have 64kB
blocksize on ext4 without some changes to the directory handling
code.  The reason is that an empty 64kB directory block would have a
rec_len == (__u16)2^16 == 0, and this would cause an error to be hit in
the filesystem.  The proposed solution is treat 64k rec_len
with a an impossible value like rec_len = 0xffff to handle this.

The Patch-set consists of the following 2 patches.
  [1/2]  ext4: enlarge blocksize
         - Allow blocksize up to pagesize

  [2/2]  ext4: fix rec_len overflow
         - prevent rec_len from overflow with 64KB blocksize

Now on 64k page ppc64 box runs with this patch set we could create a 64k
block size ext4dev, and able to handle empty directory block.

Signed-off-by: Takashi Sato <sho@tnes.nec.co.jp>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---
 fs/ext4/super.c         | 5 +++++
 include/linux/ext4_fs.h | 4 ++--
 2 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 1ca0f546c466..ab7010dde1b5 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1624,6 +1624,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		goto out_fail;
 	}
 
+	if (!sb_set_blocksize(sb, blocksize)) {
+		printk(KERN_ERR "EXT4-fs: bad blocksize %d.\n", blocksize);
+		goto out_fail;
+	}
+
 	/*
 	 * The ext4 superblock will not be buffer aligned for other than 1kB
 	 * block sizes.  We need to calculate the offset from buffer start.
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 97dd409d5f4a..dfe4487fc739 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -73,8 +73,8 @@
  * Macro-instructions used to manage several block sizes
  */
 #define EXT4_MIN_BLOCK_SIZE		1024
-#define	EXT4_MAX_BLOCK_SIZE		4096
-#define EXT4_MIN_BLOCK_LOG_SIZE		  10
+#define	EXT4_MAX_BLOCK_SIZE		65536
+#define EXT4_MIN_BLOCK_LOG_SIZE		10
 #ifdef __KERNEL__
 # define EXT4_BLOCK_SIZE(s)		((s)->s_blocksize)
 #else
-- 
cgit v1.2.3


From a72d7f834e1afa08421938d7eb06bd8e56b0e58c Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Avoid rec_len overflow with 64KB block size

With 64KB blocksize, a directory entry can have size 64KB which does not fit
into 16 bits we have for entry lenght. So we store 0xffff instead and convert
value when read from / written to disk. The patch also converts some places
to use ext4_next_entry() when we are changing them anyway.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---
 fs/ext4/dir.c           | 12 ++++----
 fs/ext4/namei.c         | 77 ++++++++++++++++++++++++-------------------------
 include/linux/ext4_fs.h | 20 +++++++++++++
 3 files changed, 63 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index f612bef98315..145a9c0c972d 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -67,7 +67,7 @@ int ext4_check_dir_entry (const char * function, struct inode * dir,
 			  unsigned long offset)
 {
 	const char * error_msg = NULL;
-	const int rlen = le16_to_cpu(de->rec_len);
+	const int rlen = ext4_rec_len_from_disk(de->rec_len);
 
 	if (rlen < EXT4_DIR_REC_LEN(1))
 		error_msg = "rec_len is smaller than minimal";
@@ -172,10 +172,10 @@ revalidate:
 				 * least that it is non-zero.  A
 				 * failure will be detected in the
 				 * dirent test below. */
-				if (le16_to_cpu(de->rec_len) <
-						EXT4_DIR_REC_LEN(1))
+				if (ext4_rec_len_from_disk(de->rec_len)
+						< EXT4_DIR_REC_LEN(1))
 					break;
-				i += le16_to_cpu(de->rec_len);
+				i += ext4_rec_len_from_disk(de->rec_len);
 			}
 			offset = i;
 			filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1))
@@ -197,7 +197,7 @@ revalidate:
 				ret = stored;
 				goto out;
 			}
-			offset += le16_to_cpu(de->rec_len);
+			offset += ext4_rec_len_from_disk(de->rec_len);
 			if (le32_to_cpu(de->inode)) {
 				/* We might block in the next section
 				 * if the data destination is
@@ -219,7 +219,7 @@ revalidate:
 					goto revalidate;
 				stored ++;
 			}
-			filp->f_pos += le16_to_cpu(de->rec_len);
+			filp->f_pos += ext4_rec_len_from_disk(de->rec_len);
 		}
 		offset = 0;
 		brelse (bh);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 94ee6f315dc1..d9a3a2fc5b0d 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -280,7 +280,7 @@ static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext4_dir_ent
 			space += EXT4_DIR_REC_LEN(de->name_len);
 			names++;
 		}
-		de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
+		de = ext4_next_entry(de);
 	}
 	printk("(%i)\n", names);
 	return (struct stats) { names, space, 1 };
@@ -551,7 +551,8 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash,
  */
 static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *p)
 {
-	return (struct ext4_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len));
+	return (struct ext4_dir_entry_2 *)((char *)p +
+		ext4_rec_len_from_disk(p->rec_len));
 }
 
 /*
@@ -720,7 +721,7 @@ static int dx_make_map (struct ext4_dir_entry_2 *de, int size,
 			cond_resched();
 		}
 		/* XXX: do we need to check rec_len == 0 case? -Chris */
-		de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
+		de = ext4_next_entry(de);
 	}
 	return count;
 }
@@ -820,7 +821,7 @@ static inline int search_dirblock(struct buffer_head * bh,
 			return 1;
 		}
 		/* prevent looping on a bad block */
-		de_len = le16_to_cpu(de->rec_len);
+		de_len = ext4_rec_len_from_disk(de->rec_len);
 		if (de_len <= 0)
 			return -1;
 		offset += de_len;
@@ -1128,7 +1129,7 @@ dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count)
 		rec_len = EXT4_DIR_REC_LEN(de->name_len);
 		memcpy (to, de, rec_len);
 		((struct ext4_dir_entry_2 *) to)->rec_len =
-				cpu_to_le16(rec_len);
+				ext4_rec_len_to_disk(rec_len);
 		de->inode = 0;
 		map++;
 		to += rec_len;
@@ -1147,13 +1148,12 @@ static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size)
 
 	prev = to = de;
 	while ((char*)de < base + size) {
-		next = (struct ext4_dir_entry_2 *) ((char *) de +
-						    le16_to_cpu(de->rec_len));
+		next = ext4_next_entry(de);
 		if (de->inode && de->name_len) {
 			rec_len = EXT4_DIR_REC_LEN(de->name_len);
 			if (de > to)
 				memmove(to, de, rec_len);
-			to->rec_len = cpu_to_le16(rec_len);
+			to->rec_len = ext4_rec_len_to_disk(rec_len);
 			prev = to;
 			to = (struct ext4_dir_entry_2 *) (((char *) to) + rec_len);
 		}
@@ -1227,8 +1227,8 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 	/* Fancy dance to stay within two buffers */
 	de2 = dx_move_dirents(data1, data2, map + split, count - split);
 	de = dx_pack_dirents(data1,blocksize);
-	de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de);
-	de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2);
+	de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de);
+	de2->rec_len = ext4_rec_len_to_disk(data2 + blocksize - (char *) de2);
 	dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1));
 	dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data2, blocksize, 1));
 
@@ -1297,7 +1297,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 				return -EEXIST;
 			}
 			nlen = EXT4_DIR_REC_LEN(de->name_len);
-			rlen = le16_to_cpu(de->rec_len);
+			rlen = ext4_rec_len_from_disk(de->rec_len);
 			if ((de->inode? rlen - nlen: rlen) >= reclen)
 				break;
 			de = (struct ext4_dir_entry_2 *)((char *)de + rlen);
@@ -1316,11 +1316,11 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 
 	/* By now the buffer is marked for journaling */
 	nlen = EXT4_DIR_REC_LEN(de->name_len);
-	rlen = le16_to_cpu(de->rec_len);
+	rlen = ext4_rec_len_from_disk(de->rec_len);
 	if (de->inode) {
 		struct ext4_dir_entry_2 *de1 = (struct ext4_dir_entry_2 *)((char *)de + nlen);
-		de1->rec_len = cpu_to_le16(rlen - nlen);
-		de->rec_len = cpu_to_le16(nlen);
+		de1->rec_len = ext4_rec_len_to_disk(rlen - nlen);
+		de->rec_len = ext4_rec_len_to_disk(nlen);
 		de = de1;
 	}
 	de->file_type = EXT4_FT_UNKNOWN;
@@ -1397,17 +1397,18 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 
 	/* The 0th block becomes the root, move the dirents out */
 	fde = &root->dotdot;
-	de = (struct ext4_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len));
+	de = (struct ext4_dir_entry_2 *)((char *)fde +
+		ext4_rec_len_from_disk(fde->rec_len));
 	len = ((char *) root) + blocksize - (char *) de;
 	memcpy (data1, de, len);
 	de = (struct ext4_dir_entry_2 *) data1;
 	top = data1 + len;
-	while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top)
+	while ((char *)(de2 = ext4_next_entry(de)) < top)
 		de = de2;
-	de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de);
+	de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de);
 	/* Initialize the root; the dot dirents already exist */
 	de = (struct ext4_dir_entry_2 *) (&root->dotdot);
-	de->rec_len = cpu_to_le16(blocksize - EXT4_DIR_REC_LEN(2));
+	de->rec_len = ext4_rec_len_to_disk(blocksize - EXT4_DIR_REC_LEN(2));
 	memset (&root->info, 0, sizeof(root->info));
 	root->info.info_length = sizeof(root->info);
 	root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version;
@@ -1487,7 +1488,7 @@ static int ext4_add_entry (handle_t *handle, struct dentry *dentry,
 		return retval;
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
 	de->inode = 0;
-	de->rec_len = cpu_to_le16(blocksize);
+	de->rec_len = ext4_rec_len_to_disk(blocksize);
 	return add_dirent_to_buf(handle, dentry, inode, de, bh);
 }
 
@@ -1550,7 +1551,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 			goto cleanup;
 		node2 = (struct dx_node *)(bh2->b_data);
 		entries2 = node2->entries;
-		node2->fake.rec_len = cpu_to_le16(sb->s_blocksize);
+		node2->fake.rec_len = ext4_rec_len_to_disk(sb->s_blocksize);
 		node2->fake.inode = 0;
 		BUFFER_TRACE(frame->bh, "get_write_access");
 		err = ext4_journal_get_write_access(handle, frame->bh);
@@ -1648,9 +1649,9 @@ static int ext4_delete_entry (handle_t *handle,
 			BUFFER_TRACE(bh, "get_write_access");
 			ext4_journal_get_write_access(handle, bh);
 			if (pde)
-				pde->rec_len =
-					cpu_to_le16(le16_to_cpu(pde->rec_len) +
-						    le16_to_cpu(de->rec_len));
+				pde->rec_len = ext4_rec_len_to_disk(
+					ext4_rec_len_from_disk(pde->rec_len) +
+					ext4_rec_len_from_disk(de->rec_len));
 			else
 				de->inode = 0;
 			dir->i_version++;
@@ -1658,10 +1659,9 @@ static int ext4_delete_entry (handle_t *handle,
 			ext4_journal_dirty_metadata(handle, bh);
 			return 0;
 		}
-		i += le16_to_cpu(de->rec_len);
+		i += ext4_rec_len_from_disk(de->rec_len);
 		pde = de;
-		de = (struct ext4_dir_entry_2 *)
-			((char *) de + le16_to_cpu(de->rec_len));
+		de = ext4_next_entry(de);
 	}
 	return -ENOENT;
 }
@@ -1824,13 +1824,13 @@ retry:
 	de = (struct ext4_dir_entry_2 *) dir_block->b_data;
 	de->inode = cpu_to_le32(inode->i_ino);
 	de->name_len = 1;
-	de->rec_len = cpu_to_le16(EXT4_DIR_REC_LEN(de->name_len));
+	de->rec_len = ext4_rec_len_to_disk(EXT4_DIR_REC_LEN(de->name_len));
 	strcpy (de->name, ".");
 	ext4_set_de_type(dir->i_sb, de, S_IFDIR);
-	de = (struct ext4_dir_entry_2 *)
-			((char *) de + le16_to_cpu(de->rec_len));
+	de = ext4_next_entry(de);
 	de->inode = cpu_to_le32(dir->i_ino);
-	de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT4_DIR_REC_LEN(1));
+	de->rec_len = ext4_rec_len_to_disk(inode->i_sb->s_blocksize -
+						EXT4_DIR_REC_LEN(1));
 	de->name_len = 2;
 	strcpy (de->name, "..");
 	ext4_set_de_type(dir->i_sb, de, S_IFDIR);
@@ -1882,8 +1882,7 @@ static int empty_dir (struct inode * inode)
 		return 1;
 	}
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
-	de1 = (struct ext4_dir_entry_2 *)
-			((char *) de + le16_to_cpu(de->rec_len));
+	de1 = ext4_next_entry(de);
 	if (le32_to_cpu(de->inode) != inode->i_ino ||
 			!le32_to_cpu(de1->inode) ||
 			strcmp (".", de->name) ||
@@ -1894,9 +1893,9 @@ static int empty_dir (struct inode * inode)
 		brelse (bh);
 		return 1;
 	}
-	offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len);
-	de = (struct ext4_dir_entry_2 *)
-			((char *) de1 + le16_to_cpu(de1->rec_len));
+	offset = ext4_rec_len_from_disk(de->rec_len) +
+		 ext4_rec_len_from_disk(de1->rec_len);
+	de = ext4_next_entry(de1);
 	while (offset < inode->i_size ) {
 		if (!bh ||
 			(void *) de >= (void *) (bh->b_data+sb->s_blocksize)) {
@@ -1925,9 +1924,8 @@ static int empty_dir (struct inode * inode)
 			brelse (bh);
 			return 0;
 		}
-		offset += le16_to_cpu(de->rec_len);
-		de = (struct ext4_dir_entry_2 *)
-				((char *) de + le16_to_cpu(de->rec_len));
+		offset += ext4_rec_len_from_disk(de->rec_len);
+		de = ext4_next_entry(de);
 	}
 	brelse (bh);
 	return 1;
@@ -2282,8 +2280,7 @@ retry:
 }
 
 #define PARENT_INO(buffer) \
-	((struct ext4_dir_entry_2 *) ((char *) buffer + \
-	le16_to_cpu(((struct ext4_dir_entry_2 *) buffer)->rec_len)))->inode
+	(ext4_next_entry((struct ext4_dir_entry_2 *)(buffer))->inode)
 
 /*
  * Anybody can rename anything with this: the permission checks are left to the
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index dfe4487fc739..fb31c1aba989 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -767,6 +767,26 @@ struct ext4_dir_entry_2 {
 #define EXT4_DIR_ROUND			(EXT4_DIR_PAD - 1)
 #define EXT4_DIR_REC_LEN(name_len)	(((name_len) + 8 + EXT4_DIR_ROUND) & \
 					 ~EXT4_DIR_ROUND)
+#define EXT4_MAX_REC_LEN		((1<<16)-1)
+
+static inline unsigned ext4_rec_len_from_disk(__le16 dlen)
+{
+	unsigned len = le16_to_cpu(dlen);
+
+	if (len == EXT4_MAX_REC_LEN)
+		return 1 << 16;
+	return len;
+}
+
+static inline __le16 ext4_rec_len_to_disk(unsigned len)
+{
+	if (len == (1 << 16))
+		return cpu_to_le16(EXT4_MAX_REC_LEN);
+	else if (len > (1 << 16))
+		BUG();
+	return cpu_to_le16(len);
+}
+
 /*
  * Hash Tree Directory indexing
  * (c) Daniel Phillips, 2001
-- 
cgit v1.2.3


From 725d26d3f09ccb5bac4b4293096b985a312a0d67 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Introduce ext4_lblk_t

This patch adds a new data type ext4_lblk_t to represent
the logical file blocks.

This is the preparatory patch to support large files in ext4
The follow up patch with convert the ext4_inode i_blocks to
represent the number of blocks in file system block size. This
changes makes it possible to have a block number 2**32 -1 which
will result in overflow if the block number is represented by
signed long. This patch convert all the block number to type
ext4_lblk_t which is typedef to __u32

Also remove dead code ext4_ext_walk_space

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---
 fs/ext4/dir.c                   |   2 +-
 fs/ext4/extents.c               | 218 ++++++++++++----------------------------
 fs/ext4/inode.c                 |  34 ++++---
 fs/ext4/namei.c                 |  54 +++++-----
 fs/ext4/super.c                 |   4 +-
 include/linux/ext4_fs.h         |  29 ++++--
 include/linux/ext4_fs_extents.h |  19 +---
 include/linux/ext4_fs_i.h       |   9 +-
 8 files changed, 143 insertions(+), 226 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 145a9c0c972d..33888bb58144 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -124,7 +124,7 @@ static int ext4_readdir(struct file * filp,
 	offset = filp->f_pos & (sb->s_blocksize - 1);
 
 	while (!error && !stored && filp->f_pos < inode->i_size) {
-		unsigned long blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb);
+		ext4_lblk_t blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb);
 		struct buffer_head map_bh;
 		struct buffer_head *bh = NULL;
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 85287742f2ae..19d8059b58aa 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -144,7 +144,7 @@ static int ext4_ext_dirty(handle_t *handle, struct inode *inode,
 
 static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
 			      struct ext4_ext_path *path,
-			      ext4_fsblk_t block)
+			      ext4_lblk_t block)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	ext4_fsblk_t bg_start;
@@ -367,13 +367,14 @@ static void ext4_ext_drop_refs(struct ext4_ext_path *path)
  * the header must be checked before calling this
  */
 static void
-ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int block)
+ext4_ext_binsearch_idx(struct inode *inode,
+			struct ext4_ext_path *path, ext4_lblk_t block)
 {
 	struct ext4_extent_header *eh = path->p_hdr;
 	struct ext4_extent_idx *r, *l, *m;
 
 
-	ext_debug("binsearch for %d(idx):  ", block);
+	ext_debug("binsearch for %lu(idx):  ", (unsigned long)block);
 
 	l = EXT_FIRST_INDEX(eh) + 1;
 	r = EXT_LAST_INDEX(eh);
@@ -425,7 +426,8 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc
  * the header must be checked before calling this
  */
 static void
-ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block)
+ext4_ext_binsearch(struct inode *inode,
+		struct ext4_ext_path *path, ext4_lblk_t block)
 {
 	struct ext4_extent_header *eh = path->p_hdr;
 	struct ext4_extent *r, *l, *m;
@@ -438,7 +440,7 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block)
 		return;
 	}
 
-	ext_debug("binsearch for %d:  ", block);
+	ext_debug("binsearch for %lu:  ", (unsigned long)block);
 
 	l = EXT_FIRST_EXTENT(eh) + 1;
 	r = EXT_LAST_EXTENT(eh);
@@ -494,7 +496,8 @@ int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
 }
 
 struct ext4_ext_path *
-ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path)
+ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
+					struct ext4_ext_path *path)
 {
 	struct ext4_extent_header *eh;
 	struct buffer_head *bh;
@@ -979,8 +982,8 @@ repeat:
 		/* refill path */
 		ext4_ext_drop_refs(path);
 		path = ext4_ext_find_extent(inode,
-					    le32_to_cpu(newext->ee_block),
-					    path);
+				    (ext4_lblk_t)le32_to_cpu(newext->ee_block),
+				    path);
 		if (IS_ERR(path))
 			err = PTR_ERR(path);
 	} else {
@@ -992,8 +995,8 @@ repeat:
 		/* refill path */
 		ext4_ext_drop_refs(path);
 		path = ext4_ext_find_extent(inode,
-					    le32_to_cpu(newext->ee_block),
-					    path);
+				   (ext4_lblk_t)le32_to_cpu(newext->ee_block),
+				    path);
 		if (IS_ERR(path)) {
 			err = PTR_ERR(path);
 			goto out;
@@ -1021,7 +1024,7 @@ out:
  * allocated block. Thus, index entries have to be consistent
  * with leaves.
  */
-static unsigned long
+static ext4_lblk_t
 ext4_ext_next_allocated_block(struct ext4_ext_path *path)
 {
 	int depth;
@@ -1054,7 +1057,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path)
  * ext4_ext_next_leaf_block:
  * returns first allocated block from next leaf or EXT_MAX_BLOCK
  */
-static unsigned ext4_ext_next_leaf_block(struct inode *inode,
+static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode,
 					struct ext4_ext_path *path)
 {
 	int depth;
@@ -1072,7 +1075,8 @@ static unsigned ext4_ext_next_leaf_block(struct inode *inode,
 	while (depth >= 0) {
 		if (path[depth].p_idx !=
 				EXT_LAST_INDEX(path[depth].p_hdr))
-		  return le32_to_cpu(path[depth].p_idx[1].ei_block);
+			return (ext4_lblk_t)
+				le32_to_cpu(path[depth].p_idx[1].ei_block);
 		depth--;
 	}
 
@@ -1239,7 +1243,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode,
 				    struct ext4_extent *newext,
 				    struct ext4_ext_path *path)
 {
-	unsigned long b1, b2;
+	ext4_lblk_t b1, b2;
 	unsigned int depth, len1;
 	unsigned int ret = 0;
 
@@ -1260,7 +1264,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode,
 			goto out;
 	}
 
-	/* check for wrap through zero */
+	/* check for wrap through zero on extent logical start block*/
 	if (b1 + len1 < b1) {
 		len1 = EXT_MAX_BLOCK - b1;
 		newext->ee_len = cpu_to_le16(len1);
@@ -1290,7 +1294,8 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 	struct ext4_extent *ex, *fex;
 	struct ext4_extent *nearex; /* nearest extent */
 	struct ext4_ext_path *npath = NULL;
-	int depth, len, err, next;
+	int depth, len, err;
+	ext4_lblk_t next;
 	unsigned uninitialized = 0;
 
 	BUG_ON(ext4_ext_get_actual_len(newext) == 0);
@@ -1435,114 +1440,8 @@ cleanup:
 	return err;
 }
 
-int ext4_ext_walk_space(struct inode *inode, unsigned long block,
-			unsigned long num, ext_prepare_callback func,
-			void *cbdata)
-{
-	struct ext4_ext_path *path = NULL;
-	struct ext4_ext_cache cbex;
-	struct ext4_extent *ex;
-	unsigned long next, start = 0, end = 0;
-	unsigned long last = block + num;
-	int depth, exists, err = 0;
-
-	BUG_ON(func == NULL);
-	BUG_ON(inode == NULL);
-
-	while (block < last && block != EXT_MAX_BLOCK) {
-		num = last - block;
-		/* find extent for this block */
-		path = ext4_ext_find_extent(inode, block, path);
-		if (IS_ERR(path)) {
-			err = PTR_ERR(path);
-			path = NULL;
-			break;
-		}
-
-		depth = ext_depth(inode);
-		BUG_ON(path[depth].p_hdr == NULL);
-		ex = path[depth].p_ext;
-		next = ext4_ext_next_allocated_block(path);
-
-		exists = 0;
-		if (!ex) {
-			/* there is no extent yet, so try to allocate
-			 * all requested space */
-			start = block;
-			end = block + num;
-		} else if (le32_to_cpu(ex->ee_block) > block) {
-			/* need to allocate space before found extent */
-			start = block;
-			end = le32_to_cpu(ex->ee_block);
-			if (block + num < end)
-				end = block + num;
-		} else if (block >= le32_to_cpu(ex->ee_block)
-					+ ext4_ext_get_actual_len(ex)) {
-			/* need to allocate space after found extent */
-			start = block;
-			end = block + num;
-			if (end >= next)
-				end = next;
-		} else if (block >= le32_to_cpu(ex->ee_block)) {
-			/*
-			 * some part of requested space is covered
-			 * by found extent
-			 */
-			start = block;
-			end = le32_to_cpu(ex->ee_block)
-				+ ext4_ext_get_actual_len(ex);
-			if (block + num < end)
-				end = block + num;
-			exists = 1;
-		} else {
-			BUG();
-		}
-		BUG_ON(end <= start);
-
-		if (!exists) {
-			cbex.ec_block = start;
-			cbex.ec_len = end - start;
-			cbex.ec_start = 0;
-			cbex.ec_type = EXT4_EXT_CACHE_GAP;
-		} else {
-			cbex.ec_block = le32_to_cpu(ex->ee_block);
-			cbex.ec_len = ext4_ext_get_actual_len(ex);
-			cbex.ec_start = ext_pblock(ex);
-			cbex.ec_type = EXT4_EXT_CACHE_EXTENT;
-		}
-
-		BUG_ON(cbex.ec_len == 0);
-		err = func(inode, path, &cbex, cbdata);
-		ext4_ext_drop_refs(path);
-
-		if (err < 0)
-			break;
-		if (err == EXT_REPEAT)
-			continue;
-		else if (err == EXT_BREAK) {
-			err = 0;
-			break;
-		}
-
-		if (ext_depth(inode) != depth) {
-			/* depth was changed. we have to realloc path */
-			kfree(path);
-			path = NULL;
-		}
-
-		block = cbex.ec_block + cbex.ec_len;
-	}
-
-	if (path) {
-		ext4_ext_drop_refs(path);
-		kfree(path);
-	}
-
-	return err;
-}
-
 static void
-ext4_ext_put_in_cache(struct inode *inode, __u32 block,
+ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block,
 			__u32 len, ext4_fsblk_t start, int type)
 {
 	struct ext4_ext_cache *cex;
@@ -1561,10 +1460,11 @@ ext4_ext_put_in_cache(struct inode *inode, __u32 block,
  */
 static void
 ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
-				unsigned long block)
+				ext4_lblk_t block)
 {
 	int depth = ext_depth(inode);
-	unsigned long lblock, len;
+	unsigned long len;
+	ext4_lblk_t lblock;
 	struct ext4_extent *ex;
 
 	ex = path[depth].p_ext;
@@ -1582,15 +1482,17 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
 				(unsigned long) ext4_ext_get_actual_len(ex));
 	} else if (block >= le32_to_cpu(ex->ee_block)
 			+ ext4_ext_get_actual_len(ex)) {
+		ext4_lblk_t next;
 		lblock = le32_to_cpu(ex->ee_block)
 			+ ext4_ext_get_actual_len(ex);
-		len = ext4_ext_next_allocated_block(path);
+
+		next = ext4_ext_next_allocated_block(path);
 		ext_debug("cache gap(after): [%lu:%lu] %lu",
 				(unsigned long) le32_to_cpu(ex->ee_block),
 				(unsigned long) ext4_ext_get_actual_len(ex),
 				(unsigned long) block);
-		BUG_ON(len == lblock);
-		len = len - lblock;
+		BUG_ON(next == lblock);
+		len = next - lblock;
 	} else {
 		lblock = len = 0;
 		BUG();
@@ -1601,7 +1503,7 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
 }
 
 static int
-ext4_ext_in_cache(struct inode *inode, unsigned long block,
+ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
 			struct ext4_extent *ex)
 {
 	struct ext4_ext_cache *cex;
@@ -1714,7 +1616,7 @@ int ext4_ext_calc_credits_for_insert(struct inode *inode,
 
 static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 				struct ext4_extent *ex,
-				unsigned long from, unsigned long to)
+				ext4_lblk_t from, ext4_lblk_t to)
 {
 	struct buffer_head *bh;
 	unsigned short ee_len =  ext4_ext_get_actual_len(ex);
@@ -1738,11 +1640,12 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 	if (from >= le32_to_cpu(ex->ee_block)
 	    && to == le32_to_cpu(ex->ee_block) + ee_len - 1) {
 		/* tail removal */
-		unsigned long num;
+		ext4_lblk_t num;
 		ext4_fsblk_t start;
+
 		num = le32_to_cpu(ex->ee_block) + ee_len - from;
 		start = ext_pblock(ex) + ee_len - num;
-		ext_debug("free last %lu blocks starting %llu\n", num, start);
+		ext_debug("free last %u blocks starting %llu\n", num, start);
 		for (i = 0; i < num; i++) {
 			bh = sb_find_get_block(inode->i_sb, start + i);
 			ext4_forget(handle, 0, inode, bh, start + i);
@@ -1750,30 +1653,32 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 		ext4_free_blocks(handle, inode, start, num);
 	} else if (from == le32_to_cpu(ex->ee_block)
 		   && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) {
-		printk("strange request: removal %lu-%lu from %u:%u\n",
+		printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n",
 			from, to, le32_to_cpu(ex->ee_block), ee_len);
 	} else {
-		printk("strange request: removal(2) %lu-%lu from %u:%u\n",
-			from, to, le32_to_cpu(ex->ee_block), ee_len);
+		printk(KERN_INFO "strange request: removal(2) "
+				"%u-%u from %u:%u\n",
+				from, to, le32_to_cpu(ex->ee_block), ee_len);
 	}
 	return 0;
 }
 
 static int
 ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
-		struct ext4_ext_path *path, unsigned long start)
+		struct ext4_ext_path *path, ext4_lblk_t start)
 {
 	int err = 0, correct_index = 0;
 	int depth = ext_depth(inode), credits;
 	struct ext4_extent_header *eh;
-	unsigned a, b, block, num;
-	unsigned long ex_ee_block;
+	ext4_lblk_t a, b, block;
+	unsigned num;
+	ext4_lblk_t ex_ee_block;
 	unsigned short ex_ee_len;
 	unsigned uninitialized = 0;
 	struct ext4_extent *ex;
 
 	/* the header must be checked already in ext4_ext_remove_space() */
-	ext_debug("truncate since %lu in leaf\n", start);
+	ext_debug("truncate since %u in leaf\n", start);
 	if (!path[depth].p_hdr)
 		path[depth].p_hdr = ext_block_hdr(path[depth].p_bh);
 	eh = path[depth].p_hdr;
@@ -1904,7 +1809,7 @@ ext4_ext_more_to_rm(struct ext4_ext_path *path)
 	return 1;
 }
 
-int ext4_ext_remove_space(struct inode *inode, unsigned long start)
+int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
 {
 	struct super_block *sb = inode->i_sb;
 	int depth = ext_depth(inode);
@@ -1912,7 +1817,7 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start)
 	handle_t *handle;
 	int i = 0, err = 0;
 
-	ext_debug("truncate since %lu\n", start);
+	ext_debug("truncate since %u\n", start);
 
 	/* probably first extent we're gonna free will be last in block */
 	handle = ext4_journal_start(inode, depth + 1);
@@ -2094,17 +1999,19 @@ void ext4_ext_release(struct super_block *sb)
  *   b> Splits in two extents: Write is happening at either end of the extent
  *   c> Splits in three extents: Somone is writing in middle of the extent
  */
-int ext4_ext_convert_to_initialized(handle_t *handle, struct inode *inode,
-					struct ext4_ext_path *path,
-					ext4_fsblk_t iblock,
-					unsigned long max_blocks)
+static int ext4_ext_convert_to_initialized(handle_t *handle,
+						struct inode *inode,
+						struct ext4_ext_path *path,
+						ext4_lblk_t iblock,
+						unsigned long max_blocks)
 {
 	struct ext4_extent *ex, newex;
 	struct ext4_extent *ex1 = NULL;
 	struct ext4_extent *ex2 = NULL;
 	struct ext4_extent *ex3 = NULL;
 	struct ext4_extent_header *eh;
-	unsigned int allocated, ee_block, ee_len, depth;
+	ext4_lblk_t ee_block;
+	unsigned int allocated, ee_len, depth;
 	ext4_fsblk_t newblock;
 	int err = 0;
 	int ret = 0;
@@ -2226,7 +2133,7 @@ out:
 }
 
 int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t iblock,
+			ext4_lblk_t iblock,
 			unsigned long max_blocks, struct buffer_head *bh_result,
 			int create, int extend_disksize)
 {
@@ -2238,8 +2145,9 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	unsigned long allocated = 0;
 
 	__clear_bit(BH_New, &bh_result->b_state);
-	ext_debug("blocks %d/%lu requested for inode %u\n", (int) iblock,
-			max_blocks, (unsigned) inode->i_ino);
+	ext_debug("blocks %lu/%lu requested for inode %u\n",
+			(unsigned long) iblock, max_blocks,
+			(unsigned) inode->i_ino);
 	mutex_lock(&EXT4_I(inode)->truncate_mutex);
 
 	/* check in cache */
@@ -2288,7 +2196,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 
 	ex = path[depth].p_ext;
 	if (ex) {
-		unsigned long ee_block = le32_to_cpu(ex->ee_block);
+		ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block);
 		ext4_fsblk_t ee_start = ext_pblock(ex);
 		unsigned short ee_len;
 
@@ -2423,7 +2331,7 @@ void ext4_ext_truncate(struct inode * inode, struct page *page)
 {
 	struct address_space *mapping = inode->i_mapping;
 	struct super_block *sb = inode->i_sb;
-	unsigned long last_block;
+	ext4_lblk_t last_block;
 	handle_t *handle;
 	int err = 0;
 
@@ -2516,7 +2424,8 @@ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num)
 long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
 {
 	handle_t *handle;
-	ext4_fsblk_t block, max_blocks;
+	ext4_lblk_t block;
+	unsigned long max_blocks;
 	ext4_fsblk_t nblocks = 0;
 	int ret = 0;
 	int ret2 = 0;
@@ -2561,8 +2470,9 @@ retry:
 		if (!ret) {
 			ext4_error(inode->i_sb, "ext4_fallocate",
 				   "ext4_ext_get_blocks returned 0! inode#%lu"
-				   ", block=%llu, max_blocks=%llu",
-				   inode->i_ino, block, max_blocks);
+				   ", block=%lu, max_blocks=%lu",
+				   inode->i_ino, (unsigned long)block,
+				   (unsigned long)max_blocks);
 			ret = -EIO;
 			ext4_mark_inode_dirty(handle, inode);
 			ret2 = ext4_journal_stop(handle);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 5489703d9573..488f829a8879 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -105,7 +105,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
  */
 static unsigned long blocks_for_truncate(struct inode *inode)
 {
-	unsigned long needed;
+	ext4_lblk_t needed;
 
 	needed = inode->i_blocks >> (inode->i_sb->s_blocksize_bits - 9);
 
@@ -282,7 +282,8 @@ static int verify_chain(Indirect *from, Indirect *to)
  */
 
 static int ext4_block_to_path(struct inode *inode,
-			long i_block, int offsets[4], int *boundary)
+			ext4_lblk_t i_block,
+			ext4_lblk_t offsets[4], int *boundary)
 {
 	int ptrs = EXT4_ADDR_PER_BLOCK(inode->i_sb);
 	int ptrs_bits = EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb);
@@ -349,7 +350,8 @@ static int ext4_block_to_path(struct inode *inode,
  *	or when it reads all @depth-1 indirect blocks successfully and finds
  *	the whole chain, all way to the data (returns %NULL, *err == 0).
  */
-static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets,
+static Indirect *ext4_get_branch(struct inode *inode, int depth,
+				 ext4_lblk_t  *offsets,
 				 Indirect chain[4], int *err)
 {
 	struct super_block *sb = inode->i_sb;
@@ -445,7 +447,7 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind)
  *	stores it in *@goal and returns zero.
  */
 
-static ext4_fsblk_t ext4_find_goal(struct inode *inode, long block,
+static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block,
 		Indirect chain[4], Indirect *partial)
 {
 	struct ext4_block_alloc_info *block_i;
@@ -590,7 +592,7 @@ failed_out:
  */
 static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 			int indirect_blks, int *blks, ext4_fsblk_t goal,
-			int *offsets, Indirect *branch)
+			ext4_lblk_t *offsets, Indirect *branch)
 {
 	int blocksize = inode->i_sb->s_blocksize;
 	int i, n = 0;
@@ -680,7 +682,7 @@ failed:
  * chain to new block and return 0.
  */
 static int ext4_splice_branch(handle_t *handle, struct inode *inode,
-			long block, Indirect *where, int num, int blks)
+			ext4_lblk_t block, Indirect *where, int num, int blks)
 {
 	int i;
 	int err = 0;
@@ -784,12 +786,12 @@ err_out:
  * return < 0, error case.
  */
 int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
-		sector_t iblock, unsigned long maxblocks,
+		ext4_lblk_t iblock, unsigned long maxblocks,
 		struct buffer_head *bh_result,
 		int create, int extend_disksize)
 {
 	int err = -EIO;
-	int offsets[4];
+	ext4_lblk_t offsets[4];
 	Indirect chain[4];
 	Indirect *partial;
 	ext4_fsblk_t goal;
@@ -803,7 +805,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 
 	J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL));
 	J_ASSERT(handle != NULL || create == 0);
-	depth = ext4_block_to_path(inode,iblock,offsets,&blocks_to_boundary);
+	depth = ext4_block_to_path(inode, iblock, offsets,
+					&blocks_to_boundary);
 
 	if (depth == 0)
 		goto out;
@@ -996,7 +999,7 @@ get_block:
  * `handle' can be NULL if create is zero
  */
 struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
-				long block, int create, int *errp)
+				ext4_lblk_t block, int create, int *errp)
 {
 	struct buffer_head dummy;
 	int fatal = 0, err;
@@ -1063,7 +1066,7 @@ err:
 }
 
 struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode,
-			       int block, int create, int *err)
+			       ext4_lblk_t block, int create, int *err)
 {
 	struct buffer_head * bh;
 
@@ -1828,7 +1831,8 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page,
 {
 	ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT;
 	unsigned offset = from & (PAGE_CACHE_SIZE-1);
-	unsigned blocksize, iblock, length, pos;
+	unsigned blocksize, length, pos;
+	ext4_lblk_t iblock;
 	struct inode *inode = mapping->host;
 	struct buffer_head *bh;
 	int err = 0;
@@ -1964,7 +1968,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q)
  *			(no partially truncated stuff there).  */
 
 static Indirect *ext4_find_shared(struct inode *inode, int depth,
-			int offsets[4], Indirect chain[4], __le32 *top)
+			ext4_lblk_t offsets[4], Indirect chain[4], __le32 *top)
 {
 	Indirect *partial, *p;
 	int k, err;
@@ -2289,12 +2293,12 @@ void ext4_truncate(struct inode *inode)
 	__le32 *i_data = ei->i_data;
 	int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb);
 	struct address_space *mapping = inode->i_mapping;
-	int offsets[4];
+	ext4_lblk_t offsets[4];
 	Indirect chain[4];
 	Indirect *partial;
 	__le32 nr = 0;
 	int n;
-	long last_block;
+	ext4_lblk_t last_block;
 	unsigned blocksize = inode->i_sb->s_blocksize;
 	struct page *page;
 
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index d9a3a2fc5b0d..fb673b14ccd5 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -51,7 +51,7 @@
 
 static struct buffer_head *ext4_append(handle_t *handle,
 					struct inode *inode,
-					u32 *block, int *err)
+					ext4_lblk_t *block, int *err)
 {
 	struct buffer_head *bh;
 
@@ -144,8 +144,8 @@ struct dx_map_entry
 	u16 size;
 };
 
-static inline unsigned dx_get_block (struct dx_entry *entry);
-static void dx_set_block (struct dx_entry *entry, unsigned value);
+static inline ext4_lblk_t dx_get_block(struct dx_entry *entry);
+static void dx_set_block(struct dx_entry *entry, ext4_lblk_t value);
 static inline unsigned dx_get_hash (struct dx_entry *entry);
 static void dx_set_hash (struct dx_entry *entry, unsigned value);
 static unsigned dx_get_count (struct dx_entry *entries);
@@ -166,7 +166,8 @@ static void dx_sort_map(struct dx_map_entry *map, unsigned count);
 static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to,
 		struct dx_map_entry *offsets, int count);
 static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size);
-static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block);
+static void dx_insert_block(struct dx_frame *frame,
+					u32 hash, ext4_lblk_t block);
 static int ext4_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
 				 struct dx_frame *frames,
@@ -181,12 +182,12 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
  * Mask them off for now.
  */
 
-static inline unsigned dx_get_block (struct dx_entry *entry)
+static inline ext4_lblk_t dx_get_block(struct dx_entry *entry)
 {
 	return le32_to_cpu(entry->block) & 0x00ffffff;
 }
 
-static inline void dx_set_block (struct dx_entry *entry, unsigned value)
+static inline void dx_set_block(struct dx_entry *entry, ext4_lblk_t value)
 {
 	entry->block = cpu_to_le32(value);
 }
@@ -243,8 +244,8 @@ static void dx_show_index (char * label, struct dx_entry *entries)
 	int i, n = dx_get_count (entries);
 	printk("%s index ", label);
 	for (i = 0; i < n; i++) {
-		printk("%x->%u ", i? dx_get_hash(entries + i) :
-				0, dx_get_block(entries + i));
+		printk("%x->%lu ", i? dx_get_hash(entries + i) :
+				0, (unsigned long)dx_get_block(entries + i));
 	}
 	printk("\n");
 }
@@ -297,7 +298,8 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir,
 	printk("%i indexed blocks...\n", count);
 	for (i = 0; i < count; i++, entries++)
 	{
-		u32 block = dx_get_block(entries), hash = i? dx_get_hash(entries): 0;
+		ext4_lblk_t block = dx_get_block(entries);
+		ext4_lblk_t hash  = i ? dx_get_hash(entries): 0;
 		u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash;
 		struct stats stats;
 		printk("%s%3u:%03u hash %8x/%8x ",levels?"":"   ", i, block, hash, range);
@@ -561,7 +563,7 @@ static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *
  * into the tree.  If there is an error it is returned in err.
  */
 static int htree_dirblock_to_tree(struct file *dir_file,
-				  struct inode *dir, int block,
+				  struct inode *dir, ext4_lblk_t block,
 				  struct dx_hash_info *hinfo,
 				  __u32 start_hash, __u32 start_minor_hash)
 {
@@ -569,7 +571,8 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 	struct ext4_dir_entry_2 *de, *top;
 	int err, count = 0;
 
-	dxtrace(printk("In htree dirblock_to_tree: block %d\n", block));
+	dxtrace(printk(KERN_INFO "In htree dirblock_to_tree: block %lu\n",
+							(unsigned long)block));
 	if (!(bh = ext4_bread (NULL, dir, block, 0, &err)))
 		return err;
 
@@ -621,9 +624,9 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 	struct ext4_dir_entry_2 *de;
 	struct dx_frame frames[2], *frame;
 	struct inode *dir;
-	int block, err;
+	ext4_lblk_t block;
 	int count = 0;
-	int ret;
+	int ret, err;
 	__u32 hashval;
 
 	dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash,
@@ -753,7 +756,7 @@ static void dx_sort_map (struct dx_map_entry *map, unsigned count)
 	} while(more);
 }
 
-static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block)
+static void dx_insert_block(struct dx_frame *frame, u32 hash, ext4_lblk_t block)
 {
 	struct dx_entry *entries = frame->entries;
 	struct dx_entry *old = frame->at, *new = old + 1;
@@ -848,13 +851,14 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry,
 	struct super_block * sb;
 	struct buffer_head * bh_use[NAMEI_RA_SIZE];
 	struct buffer_head * bh, *ret = NULL;
-	unsigned long start, block, b;
+	ext4_lblk_t start, block, b;
 	int ra_max = 0;		/* Number of bh's in the readahead
 				   buffer, bh_use[] */
 	int ra_ptr = 0;		/* Current index into readahead
 				   buffer */
 	int num = 0;
-	int nblocks, i, err;
+	ext4_lblk_t  nblocks;
+	int i, err;
 	struct inode *dir = dentry->d_parent->d_inode;
 	int namelen;
 	const u8 *name;
@@ -915,7 +919,8 @@ restart:
 		if (!buffer_uptodate(bh)) {
 			/* read error, skip block & hope for the best */
 			ext4_error(sb, __FUNCTION__, "reading directory #%lu "
-				   "offset %lu", dir->i_ino, block);
+				   "offset %lu", dir->i_ino,
+				   (unsigned long)block);
 			brelse(bh);
 			goto next;
 		}
@@ -962,7 +967,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
 	struct dx_frame frames[2], *frame;
 	struct ext4_dir_entry_2 *de, *top;
 	struct buffer_head *bh;
-	unsigned long block;
+	ext4_lblk_t block;
 	int retval;
 	int namelen = dentry->d_name.len;
 	const u8 *name = dentry->d_name.name;
@@ -1174,7 +1179,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 	unsigned blocksize = dir->i_sb->s_blocksize;
 	unsigned count, continued;
 	struct buffer_head *bh2;
-	u32 newblock;
+	ext4_lblk_t newblock;
 	u32 hash2;
 	struct dx_map_entry *map;
 	char *data1 = (*bh)->b_data, *data2;
@@ -1221,8 +1226,9 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 	split = count - move;
 	hash2 = map[split].hash;
 	continued = hash2 == map[split - 1].hash;
-	dxtrace(printk("Split block %i at %x, %i/%i\n",
-		dx_get_block(frame->at), hash2, split, count-split));
+	dxtrace(printk(KERN_INFO "Split block %lu at %x, %i/%i\n",
+			(unsigned long)dx_get_block(frame->at),
+					hash2, split, count-split));
 
 	/* Fancy dance to stay within two buffers */
 	de2 = dx_move_dirents(data1, data2, map + split, count - split);
@@ -1374,7 +1380,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 	int		retval;
 	unsigned	blocksize;
 	struct dx_hash_info hinfo;
-	u32		block;
+	ext4_lblk_t  block;
 	struct fake_dirent *fde;
 
 	blocksize =  dir->i_sb->s_blocksize;
@@ -1455,7 +1461,7 @@ static int ext4_add_entry (handle_t *handle, struct dentry *dentry,
 	int	retval;
 	int	dx_fallback=0;
 	unsigned blocksize;
-	u32 block, blocks;
+	ext4_lblk_t block, blocks;
 
 	sb = dir->i_sb;
 	blocksize = sb->s_blocksize;
@@ -1532,7 +1538,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 		       dx_get_count(entries), dx_get_limit(entries)));
 	/* Need to split index? */
 	if (dx_get_count(entries) == dx_get_limit(entries)) {
-		u32 newblock;
+		ext4_lblk_t newblock;
 		unsigned icount = dx_get_count(entries);
 		int levels = frame - frames;
 		struct dx_entry *entries2;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index ab7010dde1b5..6302b036c121 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -2914,7 +2914,7 @@ static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
 			       size_t len, loff_t off)
 {
 	struct inode *inode = sb_dqopt(sb)->files[type];
-	sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
+	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
 	int err = 0;
 	int offset = off & (sb->s_blocksize - 1);
 	int tocopy;
@@ -2952,7 +2952,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 				const char *data, size_t len, loff_t off)
 {
 	struct inode *inode = sb_dqopt(sb)->files[type];
-	sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
+	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
 	int err = 0;
 	int offset = off & (sb->s_blocksize - 1);
 	int tocopy;
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index fb31c1aba989..5e2da0974d16 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -935,11 +935,14 @@ extern unsigned long ext4_count_free (struct buffer_head *, unsigned);
 /* inode.c */
 int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
 		struct buffer_head *bh, ext4_fsblk_t blocknr);
-struct buffer_head * ext4_getblk (handle_t *, struct inode *, long, int, int *);
-struct buffer_head * ext4_bread (handle_t *, struct inode *, int, int, int *);
+struct buffer_head *ext4_getblk(handle_t *, struct inode *,
+						ext4_lblk_t, int, int *);
+struct buffer_head *ext4_bread(handle_t *, struct inode *,
+						ext4_lblk_t, int, int *);
 int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
-	sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result,
-	int create, int extend_disksize);
+				ext4_lblk_t iblock, unsigned long maxblocks,
+				struct buffer_head *bh_result,
+				int create, int extend_disksize);
 
 extern void ext4_read_inode (struct inode *);
 extern int  ext4_write_inode (struct inode *, int);
@@ -1068,7 +1071,7 @@ extern const struct inode_operations ext4_fast_symlink_inode_operations;
 extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
 extern int ext4_ext_writepage_trans_blocks(struct inode *, int);
 extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t iblock,
+			ext4_lblk_t iblock,
 			unsigned long max_blocks, struct buffer_head *bh_result,
 			int create, int extend_disksize);
 extern void ext4_ext_truncate(struct inode *, struct page *);
@@ -1081,11 +1084,17 @@ ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 			unsigned long max_blocks, struct buffer_head *bh,
 			int create, int extend_disksize)
 {
-	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)
-		return ext4_ext_get_blocks(handle, inode, block, max_blocks,
-					bh, create, extend_disksize);
-	return ext4_get_blocks_handle(handle, inode, block, max_blocks, bh,
-					create, extend_disksize);
+	int retval;
+	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) {
+		retval = ext4_ext_get_blocks(handle, inode,
+						(ext4_lblk_t)block, max_blocks,
+						bh, create, extend_disksize);
+	} else {
+		retval = ext4_get_blocks_handle(handle, inode,
+						(ext4_lblk_t)block, max_blocks,
+						bh, create, extend_disksize);
+	}
+	return retval;
 }
 
 
diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h
index d2045a26195d..023683b9d883 100644
--- a/include/linux/ext4_fs_extents.h
+++ b/include/linux/ext4_fs_extents.h
@@ -124,20 +124,6 @@ struct ext4_ext_path {
 #define EXT4_EXT_CACHE_GAP	1
 #define EXT4_EXT_CACHE_EXTENT	2
 
-/*
- * to be called by ext4_ext_walk_space()
- * negative retcode - error
- * positive retcode - signal for ext4_ext_walk_space(), see below
- * callback must return valid extent (passed or newly created)
- */
-typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *,
-					struct ext4_ext_cache *,
-					void *);
-
-#define EXT_CONTINUE	0
-#define EXT_BREAK	1
-#define EXT_REPEAT	2
-
 
 #define EXT_MAX_BLOCK	0xffffffff
 
@@ -233,8 +219,7 @@ extern int ext4_ext_try_to_merge(struct inode *inode,
 				 struct ext4_extent *);
 extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *);
 extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *);
-extern int ext4_ext_walk_space(struct inode *, unsigned long, unsigned long, ext_prepare_callback, void *);
-extern struct ext4_ext_path * ext4_ext_find_extent(struct inode *, int, struct ext4_ext_path *);
-
+extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
+							struct ext4_ext_path *);
 #endif /* _LINUX_EXT4_EXTENTS */
 
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index 86ddfe2089f3..6c610b67b047 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -27,6 +27,9 @@ typedef int ext4_grpblk_t;
 /* data type for filesystem-wide blocks number */
 typedef unsigned long long ext4_fsblk_t;
 
+/* data type for file logical block number */
+typedef __u32 ext4_lblk_t;
+
 struct ext4_reserve_window {
 	ext4_fsblk_t	_rsv_start;	/* First byte reserved */
 	ext4_fsblk_t	_rsv_end;	/* Last byte reserved or 0 */
@@ -48,7 +51,7 @@ struct ext4_block_alloc_info {
 	 * most-recently-allocated block in this file.
 	 * We use this for detecting linearly ascending allocation requests.
 	 */
-	__u32 last_alloc_logical_block;
+	ext4_lblk_t last_alloc_logical_block;
 	/*
 	 * Was i_next_alloc_goal in ext4_inode_info
 	 * is the *physical* companion to i_next_alloc_block.
@@ -67,7 +70,7 @@ struct ext4_block_alloc_info {
  */
 struct ext4_ext_cache {
 	ext4_fsblk_t	ec_start;
-	__u32		ec_block;
+	ext4_lblk_t	ec_block;
 	__u32		ec_len; /* must be 32bit to return holes */
 	__u32		ec_type;
 };
@@ -95,7 +98,7 @@ struct ext4_inode_info {
 	/* block reservation info */
 	struct ext4_block_alloc_info *i_block_alloc_info;
 
-	__u32	i_dir_start_lookup;
+	ext4_lblk_t		i_dir_start_lookup;
 #ifdef CONFIG_EXT4DEV_FS_XATTR
 	/*
 	 * Extended attributes can be read independently of the main file
-- 
cgit v1.2.3


From fd2d42912f9f09e5250cb3b024ee0625704e9cb7 Mon Sep 17 00:00:00 2001
From: Avantika Mathur <mathur@us.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: add ext4_group_t, and change all group variables to this type.

In many places variables for block group are of type int, which limits the
maximum number of block groups to 2^31.  Each block group can have up to
2^15 blocks, with a 4K block size,  and the max filesystem size is limited to
2^31 * (2^15 * 2^12) = 2^58  -- or 256 PB

This patch introduces a new type ext4_group_t, of type unsigned long, to
represent block group numbers in ext4.
All occurrences of block group variables are converted to type ext4_group_t.

Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
---
 fs/ext4/balloc.c           | 69 ++++++++++++++++++++++------------------------
 fs/ext4/group.h            |  8 ++++--
 fs/ext4/ialloc.c           | 46 ++++++++++++++++---------------
 fs/ext4/inode.c            |  5 ++--
 fs/ext4/resize.c           | 12 ++++----
 fs/ext4/super.c            | 20 ++++++--------
 include/linux/ext4_fs.h    | 11 ++++----
 include/linux/ext4_fs_i.h  |  5 +++-
 include/linux/ext4_fs_sb.h |  2 +-
 9 files changed, 91 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 71ee95e534fd..9568a57c607c 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -29,7 +29,7 @@
  * Calculate the block group number and offset, given a block number
  */
 void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
-		unsigned long *blockgrpp, ext4_grpblk_t *offsetp)
+		ext4_group_t *blockgrpp, ext4_grpblk_t *offsetp)
 {
 	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
 	ext4_grpblk_t offset;
@@ -46,7 +46,7 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
 /* Initializes an uninitialized block bitmap if given, and returns the
  * number of blocks free in the group. */
 unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
-				int block_group, struct ext4_group_desc *gdp)
+		 ext4_group_t block_group, struct ext4_group_desc *gdp)
 {
 	unsigned long start;
 	int bit, bit_max;
@@ -60,7 +60,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
 		 * essentially implementing a per-group read-only flag. */
 		if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
 			ext4_error(sb, __FUNCTION__,
-				   "Checksum bad for group %u\n", block_group);
+				  "Checksum bad for group %lu\n", block_group);
 			gdp->bg_free_blocks_count = 0;
 			gdp->bg_free_inodes_count = 0;
 			gdp->bg_itable_unused = 0;
@@ -153,7 +153,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
  *			group descriptor
  */
 struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
-					     unsigned int block_group,
+					     ext4_group_t block_group,
 					     struct buffer_head ** bh)
 {
 	unsigned long group_desc;
@@ -164,7 +164,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
 	if (block_group >= sbi->s_groups_count) {
 		ext4_error (sb, "ext4_get_group_desc",
 			    "block_group >= groups_count - "
-			    "block_group = %d, groups_count = %lu",
+			    "block_group = %lu, groups_count = %lu",
 			    block_group, sbi->s_groups_count);
 
 		return NULL;
@@ -176,7 +176,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
 	if (!sbi->s_group_desc[group_desc]) {
 		ext4_error (sb, "ext4_get_group_desc",
 			    "Group descriptor not loaded - "
-			    "block_group = %d, group_desc = %lu, desc = %lu",
+			    "block_group = %lu, group_desc = %lu, desc = %lu",
 			     block_group, group_desc, offset);
 		return NULL;
 	}
@@ -200,7 +200,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
  * Return buffer_head on success or NULL in case of failure.
  */
 struct buffer_head *
-read_block_bitmap(struct super_block *sb, unsigned int block_group)
+read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 {
 	struct ext4_group_desc * desc;
 	struct buffer_head * bh = NULL;
@@ -227,7 +227,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group)
 	if (!bh)
 		ext4_error (sb, __FUNCTION__,
 			    "Cannot read block bitmap - "
-			    "block_group = %d, block_bitmap = %llu",
+			    "block_group = %lu, block_bitmap = %llu",
 			    block_group, bitmap_blk);
 	return bh;
 }
@@ -320,7 +320,7 @@ restart:
  */
 static int
 goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal,
-			unsigned int group, struct super_block * sb)
+			ext4_group_t group, struct super_block *sb)
 {
 	ext4_fsblk_t group_first_block, group_last_block;
 
@@ -540,7 +540,7 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
 {
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *gd_bh;
-	unsigned long block_group;
+	ext4_group_t block_group;
 	ext4_grpblk_t bit;
 	unsigned long i;
 	unsigned long overflow;
@@ -920,9 +920,10 @@ claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh)
  * ext4_journal_release_buffer(), else we'll run out of credits.
  */
 static ext4_grpblk_t
-ext4_try_to_allocate(struct super_block *sb, handle_t *handle, int group,
-			struct buffer_head *bitmap_bh, ext4_grpblk_t grp_goal,
-			unsigned long *count, struct ext4_reserve_window *my_rsv)
+ext4_try_to_allocate(struct super_block *sb, handle_t *handle,
+			ext4_group_t group, struct buffer_head *bitmap_bh,
+			ext4_grpblk_t grp_goal, unsigned long *count,
+			struct ext4_reserve_window *my_rsv)
 {
 	ext4_fsblk_t group_first_block;
 	ext4_grpblk_t start, end;
@@ -1156,7 +1157,7 @@ static int find_next_reservable_window(
  */
 static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv,
 		ext4_grpblk_t grp_goal, struct super_block *sb,
-		unsigned int group, struct buffer_head *bitmap_bh)
+		ext4_group_t group, struct buffer_head *bitmap_bh)
 {
 	struct ext4_reserve_window_node *search_head;
 	ext4_fsblk_t group_first_block, group_end_block, start_block;
@@ -1354,7 +1355,7 @@ static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv,
  */
 static ext4_grpblk_t
 ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
-			unsigned int group, struct buffer_head *bitmap_bh,
+			ext4_group_t group, struct buffer_head *bitmap_bh,
 			ext4_grpblk_t grp_goal,
 			struct ext4_reserve_window_node * my_rsv,
 			unsigned long *count, int *errp)
@@ -1528,12 +1529,12 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode,
 {
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *gdp_bh;
-	unsigned long group_no;
-	int goal_group;
+	ext4_group_t group_no;
+	ext4_group_t goal_group;
 	ext4_grpblk_t grp_target_blk;	/* blockgroup relative goal block */
 	ext4_grpblk_t grp_alloc_blk;	/* blockgroup-relative allocated block*/
 	ext4_fsblk_t ret_block;		/* filesyetem-wide allocated block */
-	int bgi;			/* blockgroup iteration index */
+	ext4_group_t bgi;			/* blockgroup iteration index */
 	int fatal = 0, err;
 	int performed_allocation = 0;
 	ext4_grpblk_t free_blocks;	/* number of free blocks in a group */
@@ -1544,10 +1545,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode,
 	struct ext4_reserve_window_node *my_rsv = NULL;
 	struct ext4_block_alloc_info *block_i;
 	unsigned short windowsz = 0;
-#ifdef EXT4FS_DEBUG
-	static int goal_hits, goal_attempts;
-#endif
-	unsigned long ngroups;
+	ext4_group_t ngroups;
 	unsigned long num = *count;
 
 	*errp = -ENOSPC;
@@ -1743,9 +1741,6 @@ allocated:
 	 * list of some description.  We don't know in advance whether
 	 * the caller wants to use it as metadata or data.
 	 */
-	ext4_debug("allocating block %lu. Goal hits %d of %d.\n",
-			ret_block, goal_hits, goal_attempts);
-
 	spin_lock(sb_bgl_lock(sbi, group_no));
 	if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))
 		gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
@@ -1804,8 +1799,8 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb)
 {
 	ext4_fsblk_t desc_count;
 	struct ext4_group_desc *gdp;
-	int i;
-	unsigned long ngroups = EXT4_SB(sb)->s_groups_count;
+	ext4_group_t i;
+	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
 #ifdef EXT4FS_DEBUG
 	struct ext4_super_block *es;
 	ext4_fsblk_t bitmap_count;
@@ -1829,7 +1824,7 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb)
 			continue;
 
 		x = ext4_count_free(bitmap_bh, sb->s_blocksize);
-		printk("group %d: stored = %d, counted = %lu\n",
+		printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n",
 			i, le16_to_cpu(gdp->bg_free_blocks_count), x);
 		bitmap_count += x;
 	}
@@ -1853,7 +1848,7 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb)
 #endif
 }
 
-static inline int test_root(int a, int b)
+static inline int test_root(ext4_group_t a, int b)
 {
 	int num = b;
 
@@ -1862,7 +1857,7 @@ static inline int test_root(int a, int b)
 	return num == a;
 }
 
-static int ext4_group_sparse(int group)
+static int ext4_group_sparse(ext4_group_t group)
 {
 	if (group <= 1)
 		return 1;
@@ -1880,7 +1875,7 @@ static int ext4_group_sparse(int group)
  *	Return the number of blocks used by the superblock (primary or backup)
  *	in this group.  Currently this will be only 0 or 1.
  */
-int ext4_bg_has_super(struct super_block *sb, int group)
+int ext4_bg_has_super(struct super_block *sb, ext4_group_t group)
 {
 	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
 				EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) &&
@@ -1889,18 +1884,20 @@ int ext4_bg_has_super(struct super_block *sb, int group)
 	return 1;
 }
 
-static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb, int group)
+static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb,
+					ext4_group_t group)
 {
 	unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb);
-	unsigned long first = metagroup * EXT4_DESC_PER_BLOCK(sb);
-	unsigned long last = first + EXT4_DESC_PER_BLOCK(sb) - 1;
+	ext4_group_t first = metagroup * EXT4_DESC_PER_BLOCK(sb);
+	ext4_group_t last = first + EXT4_DESC_PER_BLOCK(sb) - 1;
 
 	if (group == first || group == first + 1 || group == last)
 		return 1;
 	return 0;
 }
 
-static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group)
+static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb,
+					ext4_group_t group)
 {
 	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
 				EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) &&
@@ -1918,7 +1915,7 @@ static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group)
  *	(primary or backup) in this group.  In the future there may be a
  *	different number of descriptor blocks in each group.
  */
-unsigned long ext4_bg_num_gdb(struct super_block *sb, int group)
+unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group)
 {
 	unsigned long first_meta_bg =
 			le32_to_cpu(EXT4_SB(sb)->s_es->s_first_meta_bg);
diff --git a/fs/ext4/group.h b/fs/ext4/group.h
index 1577910bb58b..7eb0604e7eea 100644
--- a/fs/ext4/group.h
+++ b/fs/ext4/group.h
@@ -14,14 +14,16 @@ extern __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 group,
 extern int ext4_group_desc_csum_verify(struct ext4_sb_info *sbi, __u32 group,
 				       struct ext4_group_desc *gdp);
 struct buffer_head *read_block_bitmap(struct super_block *sb,
-				      unsigned int block_group);
+				      ext4_group_t block_group);
 extern unsigned ext4_init_block_bitmap(struct super_block *sb,
-				       struct buffer_head *bh, int group,
+				       struct buffer_head *bh,
+				       ext4_group_t group,
 				       struct ext4_group_desc *desc);
 #define ext4_free_blocks_after_init(sb, group, desc)			\
 		ext4_init_block_bitmap(sb, NULL, group, desc)
 extern unsigned ext4_init_inode_bitmap(struct super_block *sb,
-				       struct buffer_head *bh, int group,
+				       struct buffer_head *bh,
+				       ext4_group_t group,
 				       struct ext4_group_desc *desc);
 extern void mark_bitmap_end(int start_bit, int end_bit, char *bitmap);
 #endif /* _LINUX_EXT4_GROUP_H */
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index c61f37fd3f05..64dea8689e1f 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -64,8 +64,8 @@ void mark_bitmap_end(int start_bit, int end_bit, char *bitmap)
 }
 
 /* Initializes an uninitialized inode bitmap */
-unsigned ext4_init_inode_bitmap(struct super_block *sb,
-				struct buffer_head *bh, int block_group,
+unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh,
+				ext4_group_t block_group,
 				struct ext4_group_desc *gdp)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -75,7 +75,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb,
 	/* If checksum is bad mark all blocks and inodes use to prevent
 	 * allocation, essentially implementing a per-group read-only flag. */
 	if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
-		ext4_error(sb, __FUNCTION__, "Checksum bad for group %u\n",
+		ext4_error(sb, __FUNCTION__, "Checksum bad for group %lu\n",
 			   block_group);
 		gdp->bg_free_blocks_count = 0;
 		gdp->bg_free_inodes_count = 0;
@@ -98,7 +98,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb,
  * Return buffer_head of bitmap on success or NULL.
  */
 static struct buffer_head *
-read_inode_bitmap(struct super_block * sb, unsigned long block_group)
+read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 {
 	struct ext4_group_desc *desc;
 	struct buffer_head *bh = NULL;
@@ -152,7 +152,7 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
 	unsigned long ino;
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *bh2;
-	unsigned long block_group;
+	ext4_group_t block_group;
 	unsigned long bit;
 	struct ext4_group_desc * gdp;
 	struct ext4_super_block * es;
@@ -260,12 +260,12 @@ error_return:
  * For other inodes, search forward from the parent directory\'s block
  * group to find a free inode.
  */
-static int find_group_dir(struct super_block *sb, struct inode *parent)
+static ext4_group_t find_group_dir(struct super_block *sb, struct inode *parent)
 {
-	int ngroups = EXT4_SB(sb)->s_groups_count;
+	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
 	unsigned int freei, avefreei;
 	struct ext4_group_desc *desc, *best_desc = NULL;
-	int group, best_group = -1;
+	ext4_group_t group, best_group = -1;
 
 	freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter);
 	avefreei = freei / ngroups;
@@ -314,12 +314,13 @@ static int find_group_dir(struct super_block *sb, struct inode *parent)
 #define INODE_COST 64
 #define BLOCK_COST 256
 
-static int find_group_orlov(struct super_block *sb, struct inode *parent)
+static ext4_group_t find_group_orlov(struct super_block *sb,
+				      struct inode *parent)
 {
-	int parent_group = EXT4_I(parent)->i_block_group;
+	ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_super_block *es = sbi->s_es;
-	int ngroups = sbi->s_groups_count;
+	ext4_group_t ngroups = sbi->s_groups_count;
 	int inodes_per_group = EXT4_INODES_PER_GROUP(sb);
 	unsigned int freei, avefreei;
 	ext4_fsblk_t freeb, avefreeb;
@@ -327,7 +328,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent)
 	unsigned int ndirs;
 	int max_debt, max_dirs, min_inodes;
 	ext4_grpblk_t min_blocks;
-	int group = -1, i;
+	ext4_group_t group = -1, i;
 	struct ext4_group_desc *desc;
 
 	freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
@@ -340,7 +341,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent)
 	if ((parent == sb->s_root->d_inode) ||
 	    (EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL)) {
 		int best_ndir = inodes_per_group;
-		int best_group = -1;
+		ext4_group_t best_group = -1;
 
 		get_random_bytes(&group, sizeof(group));
 		parent_group = (unsigned)group % ngroups;
@@ -415,12 +416,13 @@ fallback:
 	return -1;
 }
 
-static int find_group_other(struct super_block *sb, struct inode *parent)
+static ext4_group_t find_group_other(struct super_block *sb,
+					struct inode *parent)
 {
-	int parent_group = EXT4_I(parent)->i_block_group;
-	int ngroups = EXT4_SB(sb)->s_groups_count;
+	ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
+	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
 	struct ext4_group_desc *desc;
-	int group, i;
+	ext4_group_t group, i;
 
 	/*
 	 * Try to place the inode in its parent directory
@@ -487,7 +489,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
 	struct super_block *sb;
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *bh2;
-	int group;
+	ext4_group_t group;
 	unsigned long ino = 0;
 	struct inode * inode;
 	struct ext4_group_desc * gdp = NULL;
@@ -583,7 +585,7 @@ got:
 	    ino > EXT4_INODES_PER_GROUP(sb)) {
 		ext4_error(sb, __FUNCTION__,
 			   "reserved inode or inode > inodes count - "
-			   "block_group = %d, inode=%lu", group,
+			   "block_group = %lu, inode=%lu", group,
 			   ino + group * EXT4_INODES_PER_GROUP(sb));
 		err = -EIO;
 		goto fail;
@@ -777,7 +779,7 @@ fail_drop:
 struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 {
 	unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
-	unsigned long block_group;
+	ext4_group_t block_group;
 	int bit;
 	struct buffer_head *bitmap_bh = NULL;
 	struct inode *inode = NULL;
@@ -833,7 +835,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 {
 	unsigned long desc_count;
 	struct ext4_group_desc *gdp;
-	int i;
+	ext4_group_t i;
 #ifdef EXT4FS_DEBUG
 	struct ext4_super_block *es;
 	unsigned long bitmap_count, x;
@@ -879,7 +881,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 unsigned long ext4_count_dirs (struct super_block * sb)
 {
 	unsigned long count = 0;
-	int i;
+	ext4_group_t i;
 
 	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
 		struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 488f829a8879..1ee19c918686 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2464,7 +2464,8 @@ out_stop:
 static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
 		unsigned long ino, struct ext4_iloc *iloc)
 {
-	unsigned long desc, group_desc, block_group;
+	unsigned long desc, group_desc;
+	ext4_group_t block_group;
 	unsigned long offset;
 	ext4_fsblk_t block;
 	struct buffer_head *bh;
@@ -2551,7 +2552,7 @@ static int __ext4_get_inode_loc(struct inode *inode,
 			struct ext4_group_desc *desc;
 			int inodes_per_buffer;
 			int inode_offset, i;
-			int block_group;
+			ext4_group_t block_group;
 			int start;
 
 			block_group = (inode->i_ino - 1) /
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index bd8a52bb3999..7090c2d25c76 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -28,7 +28,7 @@ static int verify_group_input(struct super_block *sb,
 	struct ext4_super_block *es = sbi->s_es;
 	ext4_fsblk_t start = ext4_blocks_count(es);
 	ext4_fsblk_t end = start + input->blocks_count;
-	unsigned group = input->group;
+	ext4_group_t group = input->group;
 	ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group;
 	unsigned overhead = ext4_bg_has_super(sb, group) ?
 		(1 + ext4_bg_num_gdb(sb, group) +
@@ -357,7 +357,7 @@ static int verify_reserved_gdb(struct super_block *sb,
 			       struct buffer_head *primary)
 {
 	const ext4_fsblk_t blk = primary->b_blocknr;
-	const unsigned long end = EXT4_SB(sb)->s_groups_count;
+	const ext4_group_t end = EXT4_SB(sb)->s_groups_count;
 	unsigned three = 1;
 	unsigned five = 5;
 	unsigned seven = 7;
@@ -656,12 +656,12 @@ static void update_backups(struct super_block *sb,
 			   int blk_off, char *data, int size)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	const unsigned long last = sbi->s_groups_count;
+	const ext4_group_t last = sbi->s_groups_count;
 	const int bpg = EXT4_BLOCKS_PER_GROUP(sb);
 	unsigned three = 1;
 	unsigned five = 5;
 	unsigned seven = 7;
-	unsigned group;
+	ext4_group_t group;
 	int rest = sb->s_blocksize - size;
 	handle_t *handle;
 	int err = 0, err2;
@@ -716,7 +716,7 @@ static void update_backups(struct super_block *sb,
 exit_err:
 	if (err) {
 		ext4_warning(sb, __FUNCTION__,
-			     "can't update backup for group %d (err %d), "
+			     "can't update backup for group %lu (err %d), "
 			     "forcing fsck on next reboot", group, err);
 		sbi->s_mount_state &= ~EXT4_VALID_FS;
 		sbi->s_es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
@@ -952,7 +952,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 		      ext4_fsblk_t n_blocks_count)
 {
 	ext4_fsblk_t o_blocks_count;
-	unsigned long o_groups_count;
+	ext4_group_t o_groups_count;
 	ext4_grpblk_t last;
 	ext4_grpblk_t add;
 	struct buffer_head * bh;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 6302b036c121..df8842b43544 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1364,7 +1364,7 @@ static int ext4_check_descriptors (struct super_block * sb)
 	struct ext4_group_desc * gdp = NULL;
 	int desc_block = 0;
 	int flexbg_flag = 0;
-	int i;
+	ext4_group_t i;
 
 	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
 		flexbg_flag = 1;
@@ -1386,7 +1386,7 @@ static int ext4_check_descriptors (struct super_block * sb)
 		if (block_bitmap < first_block || block_bitmap > last_block)
 		{
 			ext4_error (sb, "ext4_check_descriptors",
-				    "Block bitmap for group %d"
+				    "Block bitmap for group %lu"
 				    " not in group (block %llu)!",
 				    i, block_bitmap);
 			return 0;
@@ -1395,7 +1395,7 @@ static int ext4_check_descriptors (struct super_block * sb)
 		if (inode_bitmap < first_block || inode_bitmap > last_block)
 		{
 			ext4_error (sb, "ext4_check_descriptors",
-				    "Inode bitmap for group %d"
+				    "Inode bitmap for group %lu"
 				    " not in group (block %llu)!",
 				    i, inode_bitmap);
 			return 0;
@@ -1405,17 +1405,16 @@ static int ext4_check_descriptors (struct super_block * sb)
 		    inode_table + sbi->s_itb_per_group - 1 > last_block)
 		{
 			ext4_error (sb, "ext4_check_descriptors",
-				    "Inode table for group %d"
+				    "Inode table for group %lu"
 				    " not in group (block %llu)!",
 				    i, inode_table);
 			return 0;
 		}
 		if (!ext4_group_desc_csum_verify(sbi, i, gdp)) {
 			ext4_error(sb, __FUNCTION__,
-				   "Checksum for group %d failed (%u!=%u)\n", i,
-				   le16_to_cpu(ext4_group_desc_csum(sbi, i,
-								    gdp)),
-				   le16_to_cpu(gdp->bg_checksum));
+				   "Checksum for group %lu failed (%u!=%u)\n",
+				    i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
+				    gdp)), le16_to_cpu(gdp->bg_checksum));
 			return 0;
 		}
 		if (!flexbg_flag)
@@ -1429,7 +1428,6 @@ static int ext4_check_descriptors (struct super_block * sb)
 	return 1;
 }
 
-
 /* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
  * the superblock) which were deleted from all directories, but held open by
  * a process at the time of a crash.  We walk the list and try to delete these
@@ -1570,7 +1568,7 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb,
 				ext4_fsblk_t logical_sb_block, int nr)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	unsigned long bg, first_meta_bg;
+	ext4_group_t bg, first_meta_bg;
 	int has_super = 0;
 
 	first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg);
@@ -2678,7 +2676,7 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf)
 	if (test_opt(sb, MINIX_DF)) {
 		sbi->s_overhead_last = 0;
 	} else if (sbi->s_blocks_last != ext4_blocks_count(es)) {
-		unsigned long ngroups = sbi->s_groups_count, i;
+		ext4_group_t ngroups = sbi->s_groups_count, i;
 		ext4_fsblk_t overhead = 0;
 		smp_rmb();
 
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 5e2da0974d16..e1103c2a0d42 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -830,7 +830,7 @@ struct ext4_iloc
 {
 	struct buffer_head *bh;
 	unsigned long offset;
-	unsigned long block_group;
+	ext4_group_t block_group;
 };
 
 static inline struct ext4_inode *ext4_raw_inode(struct ext4_iloc *iloc)
@@ -855,7 +855,7 @@ struct dir_private_info {
 
 /* calculate the first block number of the group */
 static inline ext4_fsblk_t
-ext4_group_first_block_no(struct super_block *sb, unsigned long group_no)
+ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no)
 {
 	return group_no * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) +
 		le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
@@ -886,8 +886,9 @@ extern unsigned int ext4_block_group(struct super_block *sb,
 			ext4_fsblk_t blocknr);
 extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb,
 			ext4_fsblk_t blocknr);
-extern int ext4_bg_has_super(struct super_block *sb, int group);
-extern unsigned long ext4_bg_num_gdb(struct super_block *sb, int group);
+extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group);
+extern unsigned long ext4_bg_num_gdb(struct super_block *sb,
+			ext4_group_t group);
 extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, int *errp);
 extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode,
@@ -900,7 +901,7 @@ extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb,
 extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *);
 extern void ext4_check_blocks_bitmap (struct super_block *);
 extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
-						    unsigned int block_group,
+						    ext4_group_t block_group,
 						    struct buffer_head ** bh);
 extern int ext4_should_retry_alloc(struct super_block *sb, int *retries);
 extern void ext4_init_block_alloc_info(struct inode *);
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index 6c610b67b047..2b4e3700c725 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -30,6 +30,9 @@ typedef unsigned long long ext4_fsblk_t;
 /* data type for file logical block number */
 typedef __u32 ext4_lblk_t;
 
+/* data type for block group number */
+typedef unsigned long ext4_group_t;
+
 struct ext4_reserve_window {
 	ext4_fsblk_t	_rsv_start;	/* First byte reserved */
 	ext4_fsblk_t	_rsv_end;	/* Last byte reserved or 0 */
@@ -92,7 +95,7 @@ struct ext4_inode_info {
 	 * place a file's data blocks near its inode block, and new inodes
 	 * near to their parent directory's inode.
 	 */
-	__u32	i_block_group;
+	ext4_group_t	i_block_group;
 	__u32	i_state;		/* Dynamic state flags for ext4 */
 
 	/* block reservation info */
diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h
index b40e827cd495..f15821c5e4c4 100644
--- a/include/linux/ext4_fs_sb.h
+++ b/include/linux/ext4_fs_sb.h
@@ -35,7 +35,7 @@ struct ext4_sb_info {
 	unsigned long s_itb_per_group;	/* Number of inode table blocks per group */
 	unsigned long s_gdb_count;	/* Number of group descriptor blocks */
 	unsigned long s_desc_per_block;	/* Number of group descriptors per block */
-	unsigned long s_groups_count;	/* Number of groups in the fs */
+	ext4_group_t s_groups_count;	/* Number of groups in the fs */
 	unsigned long s_overhead_last;  /* Last calculated overhead */
 	unsigned long s_blocks_last;    /* Last seen block count */
 	struct buffer_head * s_sbh;	/* Buffer containing the super block */
-- 
cgit v1.2.3


From 99e6f829a854daa6d56006cad51156e98863e73a Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Introduce ext4_update_*_feature

Introduce ext4_update_*_feature and use them instead
of opencoding.


Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/ialloc.c        | 11 ++++-----
 fs/ext4/super.c         | 60 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/ext4_fs.h |  6 +++++
 3 files changed, 70 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 7b5cfa62b663..00b152b92480 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -748,13 +748,10 @@ got:
 	if (test_opt(sb, EXTENTS)) {
 		EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
 		ext4_ext_tree_init(handle, inode);
-		if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
-			err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh);
-			if (err) goto fail;
-			EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS);
-			BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "call ext4_journal_dirty_metadata");
-			err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh);
-		}
+		err = ext4_update_incompat_feature(handle, sb,
+						EXT4_FEATURE_INCOMPAT_EXTENTS);
+		if (err)
+			goto fail;
 	}
 
 	ext4_debug("allocating inode %lu\n", inode->i_ino);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index df8842b43544..4d7f33f79552 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -373,6 +373,66 @@ void ext4_update_dynamic_rev(struct super_block *sb)
 	 */
 }
 
+int ext4_update_compat_feature(handle_t *handle,
+					struct super_block *sb, __u32 compat)
+{
+	int err = 0;
+	if (!EXT4_HAS_COMPAT_FEATURE(sb, compat)) {
+		err = ext4_journal_get_write_access(handle,
+				EXT4_SB(sb)->s_sbh);
+		if (err)
+			return err;
+		EXT4_SET_COMPAT_FEATURE(sb, compat);
+		sb->s_dirt = 1;
+		handle->h_sync = 1;
+		BUFFER_TRACE(EXT4_SB(sb)->s_sbh,
+					"call ext4_journal_dirty_met adata");
+		err = ext4_journal_dirty_metadata(handle,
+				EXT4_SB(sb)->s_sbh);
+	}
+	return err;
+}
+
+int ext4_update_rocompat_feature(handle_t *handle,
+					struct super_block *sb, __u32 rocompat)
+{
+	int err = 0;
+	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, rocompat)) {
+		err = ext4_journal_get_write_access(handle,
+				EXT4_SB(sb)->s_sbh);
+		if (err)
+			return err;
+		EXT4_SET_RO_COMPAT_FEATURE(sb, rocompat);
+		sb->s_dirt = 1;
+		handle->h_sync = 1;
+		BUFFER_TRACE(EXT4_SB(sb)->s_sbh,
+					"call ext4_journal_dirty_met adata");
+		err = ext4_journal_dirty_metadata(handle,
+				EXT4_SB(sb)->s_sbh);
+	}
+	return err;
+}
+
+int ext4_update_incompat_feature(handle_t *handle,
+					struct super_block *sb, __u32 incompat)
+{
+	int err = 0;
+	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, incompat)) {
+		err = ext4_journal_get_write_access(handle,
+				EXT4_SB(sb)->s_sbh);
+		if (err)
+			return err;
+		EXT4_SET_INCOMPAT_FEATURE(sb, incompat);
+		sb->s_dirt = 1;
+		handle->h_sync = 1;
+		BUFFER_TRACE(EXT4_SB(sb)->s_sbh,
+					"call ext4_journal_dirty_met adata");
+		err = ext4_journal_dirty_metadata(handle,
+				EXT4_SB(sb)->s_sbh);
+	}
+	return err;
+}
+
 /*
  * Open the external journal device
  */
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index e1103c2a0d42..429dbfc851ea 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -989,6 +989,12 @@ extern void ext4_abort (struct super_block *, const char *, const char *, ...)
 extern void ext4_warning (struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
 extern void ext4_update_dynamic_rev (struct super_block *sb);
+extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb,
+					__u32 compat);
+extern int ext4_update_rocompat_feature(handle_t *handle,
+					struct super_block *sb,	__u32 rocompat);
+extern int ext4_update_incompat_feature(handle_t *handle,
+					struct super_block *sb,	__u32 incompat);
 extern ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
 				      struct ext4_group_desc *bg);
 extern ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
-- 
cgit v1.2.3


From 1d03ec984ca41ba184822d1101babb3fa3e26c77 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4:  Fix sparse warnings.

Fix sparse warnings related to static functions
and local variables.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/extents.c       |  6 +++---
 fs/ext4/inode.c         | 18 +++++++++++-------
 fs/ext4/super.c         |  3 +++
 include/linux/ext4_fs.h |  2 ++
 4 files changed, 19 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 68537229ee1c..754c0d36d162 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1088,7 +1088,7 @@ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode,
  * then we have to correct all indexes above.
  * TODO: do we need to correct tree in all cases?
  */
-int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
+static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
 				struct ext4_ext_path *path)
 {
 	struct ext4_extent_header *eh;
@@ -1535,7 +1535,7 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
  * It's used in truncate case only, thus all requests are for
  * last index in the block only.
  */
-int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
+static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
 			struct ext4_ext_path *path)
 {
 	struct buffer_head *bh;
@@ -1806,7 +1806,7 @@ ext4_ext_more_to_rm(struct ext4_ext_path *path)
 	return 1;
 }
 
-int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
+static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
 {
 	struct super_block *sb = inode->i_sb;
 	int depth = ext_depth(inode);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 1ee19c918686..76ceba2718b9 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2052,11 +2052,11 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode,
 	for (p = first; p < last; p++) {
 		u32 nr = le32_to_cpu(*p);
 		if (nr) {
-			struct buffer_head *bh;
+			struct buffer_head *tbh;
 
 			*p = 0;
-			bh = sb_find_get_block(inode->i_sb, nr);
-			ext4_forget(handle, 0, inode, bh, nr);
+			tbh = sb_find_get_block(inode->i_sb, nr);
+			ext4_forget(handle, 0, inode, tbh, nr);
 		}
 	}
 
@@ -2324,8 +2324,10 @@ void ext4_truncate(struct inode *inode)
 			return;
 	}
 
-	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)
-		return ext4_ext_truncate(inode, page);
+	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) {
+		ext4_ext_truncate(inode, page);
+		return;
+	}
 
 	handle = start_transaction(inode);
 	if (IS_ERR(handle)) {
@@ -3163,8 +3165,10 @@ ext4_reserve_inode_write(handle_t *handle, struct inode *inode,
  * Expand an inode by new_extra_isize bytes.
  * Returns 0 on success or negative error number on failure.
  */
-int ext4_expand_extra_isize(struct inode *inode, unsigned int new_extra_isize,
-			struct ext4_iloc iloc, handle_t *handle)
+static int ext4_expand_extra_isize(struct inode *inode,
+				   unsigned int new_extra_isize,
+				   struct ext4_iloc iloc,
+				   handle_t *handle)
 {
 	struct ext4_inode *raw_inode;
 	struct ext4_xattr_ibody_header *header;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 4d7f33f79552..7be27dbe76bf 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1644,6 +1644,9 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb,
 
 
 static int ext4_fill_super (struct super_block *sb, void *data, int silent)
+				__releases(kernel_sem)
+				__acquires(kernel_sem)
+
 {
 	struct buffer_head * bh;
 	struct ext4_super_block *es = NULL;
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 429dbfc851ea..1a27433b159a 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -893,6 +893,8 @@ extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, int *errp);
 extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, unsigned long *count, int *errp);
+extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
+			ext4_fsblk_t goal, unsigned long *count, int *errp);
 extern void ext4_free_blocks (handle_t *handle, struct inode *inode,
 			ext4_fsblk_t block, unsigned long count);
 extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb,
-- 
cgit v1.2.3


From 7973c0c19ecba92f113488045005f8e7ce1cd7c8 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Rename i_file_acl to i_file_acl_lo

Rename i_file_acl to i_file_acl_lo. This helps
in finding bugs where we use i_file_acl instead
of the combined i_file_acl_lo and i_file_acl_high

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/inode.c         | 4 ++--
 include/linux/ext4_fs.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 76ceba2718b9..7bcec1860084 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2718,7 +2718,7 @@ void ext4_read_inode(struct inode * inode)
 	}
 	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
-	ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl);
+	ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_HURD))
 		ei->i_file_acl |=
@@ -2866,7 +2866,7 @@ static int ext4_do_update_inode(handle_t *handle,
 	    cpu_to_le32(EXT4_OS_HURD))
 		raw_inode->i_file_acl_high =
 			cpu_to_le16(ei->i_file_acl >> 32);
-	raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl);
+	raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl);
 	if (!S_ISREG(inode->i_mode)) {
 		raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl);
 	} else {
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 1a27433b159a..6894f361d01d 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -297,7 +297,7 @@ struct ext4_inode {
 	} osd1;				/* OS dependent 1 */
 	__le32	i_block[EXT4_N_BLOCKS];/* Pointers to blocks */
 	__le32	i_generation;	/* File version (for NFS) */
-	__le32	i_file_acl;	/* File ACL */
+	__le32	i_file_acl_lo;	/* File ACL */
 	__le32	i_dir_acl;	/* Directory ACL */
 	__le32	i_obso_faddr;	/* Obsoleted fragment address */
 	union {
-- 
cgit v1.2.3


From a48380f769dfed6163fb82a68b13bd562ea1e027 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Rename i_dir_acl to i_size_high

Rename ext4_inode.i_dir_acl to i_size_high
drop ext4_inode_info.i_dir_acl as it is not used
Rename ext4_inode.i_size to ext4_inode.i_size_lo
Add helper function for accessing the ext4_inode combined i_size.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/ialloc.c          |  1 -
 fs/ext4/inode.c           | 55 +++++++++++++++++++----------------------------
 include/linux/ext4_fs.h   | 15 ++++++++++---
 include/linux/ext4_fs_i.h |  1 -
 4 files changed, 34 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 00b152b92480..17b5df14f85b 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -709,7 +709,6 @@ got:
 	if (!S_ISDIR(mode))
 		ei->i_flags &= ~EXT4_DIRSYNC_FL;
 	ei->i_file_acl = 0;
-	ei->i_dir_acl = 0;
 	ei->i_dtime = 0;
 	ei->i_block_alloc_info = NULL;
 	ei->i_block_group = group;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 7bcec1860084..e6634550cfc8 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2694,7 +2694,6 @@ void ext4_read_inode(struct inode * inode)
 		inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16;
 	}
 	inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
-	inode->i_size = le32_to_cpu(raw_inode->i_size);
 
 	ei->i_state = 0;
 	ei->i_dir_start_lookup = 0;
@@ -2720,15 +2719,11 @@ void ext4_read_inode(struct inode * inode)
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 	ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
-	    cpu_to_le32(EXT4_OS_HURD))
+	    cpu_to_le32(EXT4_OS_HURD)) {
 		ei->i_file_acl |=
 			((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
-	if (!S_ISREG(inode->i_mode)) {
-		ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl);
-	} else {
-		inode->i_size |=
-			((__u64)le32_to_cpu(raw_inode->i_size_high)) << 32;
 	}
+	inode->i_size = ext4_isize(raw_inode);
 	ei->i_disksize = inode->i_size;
 	inode->i_generation = le32_to_cpu(raw_inode->i_generation);
 	ei->i_block_group = iloc.block_group;
@@ -2852,7 +2847,6 @@ static int ext4_do_update_inode(handle_t *handle,
 		raw_inode->i_gid_high = 0;
 	}
 	raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
-	raw_inode->i_size = cpu_to_le32(ei->i_disksize);
 
 	EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
 	EXT4_INODE_SET_XTIME(i_mtime, inode, raw_inode);
@@ -2867,32 +2861,27 @@ static int ext4_do_update_inode(handle_t *handle,
 		raw_inode->i_file_acl_high =
 			cpu_to_le16(ei->i_file_acl >> 32);
 	raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl);
-	if (!S_ISREG(inode->i_mode)) {
-		raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl);
-	} else {
-		raw_inode->i_size_high =
-			cpu_to_le32(ei->i_disksize >> 32);
-		if (ei->i_disksize > 0x7fffffffULL) {
-			struct super_block *sb = inode->i_sb;
-			if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-					EXT4_FEATURE_RO_COMPAT_LARGE_FILE) ||
-			    EXT4_SB(sb)->s_es->s_rev_level ==
-					cpu_to_le32(EXT4_GOOD_OLD_REV)) {
-			       /* If this is the first large file
-				* created, add a flag to the superblock.
-				*/
-				err = ext4_journal_get_write_access(handle,
-						EXT4_SB(sb)->s_sbh);
-				if (err)
-					goto out_brelse;
-				ext4_update_dynamic_rev(sb);
-				EXT4_SET_RO_COMPAT_FEATURE(sb,
+	ext4_isize_set(raw_inode, ei->i_disksize);
+	if (ei->i_disksize > 0x7fffffffULL) {
+		struct super_block *sb = inode->i_sb;
+		if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
+				EXT4_FEATURE_RO_COMPAT_LARGE_FILE) ||
+				EXT4_SB(sb)->s_es->s_rev_level ==
+				cpu_to_le32(EXT4_GOOD_OLD_REV)) {
+			/* If this is the first large file
+			 * created, add a flag to the superblock.
+			 */
+			err = ext4_journal_get_write_access(handle,
+					EXT4_SB(sb)->s_sbh);
+			if (err)
+				goto out_brelse;
+			ext4_update_dynamic_rev(sb);
+			EXT4_SET_RO_COMPAT_FEATURE(sb,
 					EXT4_FEATURE_RO_COMPAT_LARGE_FILE);
-				sb->s_dirt = 1;
-				handle->h_sync = 1;
-				err = ext4_journal_dirty_metadata(handle,
-						EXT4_SB(sb)->s_sbh);
-			}
+			sb->s_dirt = 1;
+			handle->h_sync = 1;
+			err = ext4_journal_dirty_metadata(handle,
+					EXT4_SB(sb)->s_sbh);
 		}
 	}
 	raw_inode->i_generation = cpu_to_le32(inode->i_generation);
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 6894f361d01d..a8f3faea8eff 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -275,7 +275,7 @@ struct ext4_mount_options {
 struct ext4_inode {
 	__le16	i_mode;		/* File mode */
 	__le16	i_uid;		/* Low 16 bits of Owner Uid */
-	__le32	i_size;		/* Size in bytes */
+	__le32	i_size_lo;	/* Size in bytes */
 	__le32	i_atime;	/* Access time */
 	__le32	i_ctime;	/* Inode Change time */
 	__le32	i_mtime;	/* Modification time */
@@ -298,7 +298,7 @@ struct ext4_inode {
 	__le32	i_block[EXT4_N_BLOCKS];/* Pointers to blocks */
 	__le32	i_generation;	/* File version (for NFS) */
 	__le32	i_file_acl_lo;	/* File ACL */
-	__le32	i_dir_acl;	/* Directory ACL */
+	__le32	i_size_high;
 	__le32	i_obso_faddr;	/* Obsoleted fragment address */
 	union {
 		struct {
@@ -330,7 +330,6 @@ struct ext4_inode {
 	__le32  i_crtime_extra; /* extra FileCreationtime (nsec << 2 | epoch) */
 };
 
-#define i_size_high	i_dir_acl
 
 #define EXT4_EPOCH_BITS 2
 #define EXT4_EPOCH_MASK ((1 << EXT4_EPOCH_BITS) - 1)
@@ -1049,7 +1048,17 @@ static inline void ext4_r_blocks_count_set(struct ext4_super_block *es,
 	es->s_r_blocks_count_hi = cpu_to_le32(blk >> 32);
 }
 
+static inline loff_t ext4_isize(struct ext4_inode *raw_inode)
+{
+	return ((loff_t)le32_to_cpu(raw_inode->i_size_high) << 32) |
+		le32_to_cpu(raw_inode->i_size_lo);
+}
 
+static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size)
+{
+	raw_inode->i_size_lo = cpu_to_le32(i_size);
+	raw_inode->i_size_high = cpu_to_le32(i_size >> 32);
+}
 
 #define ext4_std_error(sb, errno)				\
 do {								\
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index 2b4e3700c725..f1cd4934e46f 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -85,7 +85,6 @@ struct ext4_inode_info {
 	__le32	i_data[15];	/* unconverted */
 	__u32	i_flags;
 	ext4_fsblk_t	i_file_acl;
-	__u32	i_dir_acl;
 	__u32	i_dtime;
 
 	/*
-- 
cgit v1.2.3


From 0fc1b451471dfc3cabd6e99ef441df9804616e63 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:26 -0500
Subject: ext4: Add support for 48 bit inode i_blocks.

Use the __le16 l_i_reserved1 field of the linux2 struct of ext4_inode
to represet the higher 16 bits for i_blocks. With this change max_file
size becomes (2**48 -1 )* 512 bytes.

We add a RO_COMPAT feature to the super block to indicate that inode
have i_blocks represented as a split 48 bits. Super block with this
feature set cannot be mounted read write on a kernel with CONFIG_LSF
disabled.

Super block flag EXT4_FEATURE_RO_COMPAT_HUGE_FILE

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/inode.c         | 58 +++++++++++++++++++++++++++++++++++++++++++--
 fs/ext4/super.c         | 62 ++++++++++++++++++++++++++++++++++++++++++++-----
 include/linux/ext4_fs.h | 10 +++++---
 3 files changed, 119 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index e6634550cfc8..bb89fe727bb1 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2667,6 +2667,22 @@ void ext4_get_inode_flags(struct ext4_inode_info *ei)
 	if (flags & S_DIRSYNC)
 		ei->i_flags |= EXT4_DIRSYNC_FL;
 }
+static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
+					struct ext4_inode_info *ei)
+{
+	blkcnt_t i_blocks ;
+	struct super_block *sb = ei->vfs_inode.i_sb;
+
+	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
+				EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
+		/* we are using combined 48 bit field */
+		i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 |
+					le32_to_cpu(raw_inode->i_blocks_lo);
+		return i_blocks;
+	} else {
+		return le32_to_cpu(raw_inode->i_blocks_lo);
+	}
+}
 
 void ext4_read_inode(struct inode * inode)
 {
@@ -2715,8 +2731,8 @@ void ext4_read_inode(struct inode * inode)
 		 * recovery code: that's fine, we're about to complete
 		 * the process of deleting those. */
 	}
-	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
+	inode->i_blocks = ext4_inode_blocks(raw_inode, ei);
 	ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_HURD)) {
@@ -2799,6 +2815,43 @@ bad_inode:
 	return;
 }
 
+static int ext4_inode_blocks_set(handle_t *handle,
+				struct ext4_inode *raw_inode,
+				struct ext4_inode_info *ei)
+{
+	struct inode *inode = &(ei->vfs_inode);
+	u64 i_blocks = inode->i_blocks;
+	struct super_block *sb = inode->i_sb;
+	int err = 0;
+
+	if (i_blocks <= ~0U) {
+		/*
+		 * i_blocks can be represnted in a 32 bit variable
+		 * as multiple of 512 bytes
+		 */
+		raw_inode->i_blocks_lo   = cpu_to_le32((u32)i_blocks);
+		raw_inode->i_blocks_high = 0;
+	} else if (i_blocks <= 0xffffffffffffULL) {
+		/*
+		 * i_blocks can be represented in a 48 bit variable
+		 * as multiple of 512 bytes
+		 */
+		err = ext4_update_rocompat_feature(handle, sb,
+					    EXT4_FEATURE_RO_COMPAT_HUGE_FILE);
+		if (err)
+			goto  err_out;
+		/* i_block is stored in the split  48 bit fields */
+		raw_inode->i_blocks_lo   = cpu_to_le32((u32)i_blocks);
+		raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32);
+	} else {
+		ext4_error(sb, __FUNCTION__,
+				"Wrong inode i_blocks count  %llu\n",
+				(unsigned long long)inode->i_blocks);
+	}
+err_out:
+	return err;
+}
+
 /*
  * Post the struct inode info into an on-disk inode location in the
  * buffer-cache.  This gobbles the caller's reference to the
@@ -2853,7 +2906,8 @@ static int ext4_do_update_inode(handle_t *handle,
 	EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode);
 	EXT4_EINODE_SET_XTIME(i_crtime, ei, raw_inode);
 
-	raw_inode->i_blocks = cpu_to_le32(inode->i_blocks);
+	if (ext4_inode_blocks_set(handle, raw_inode, ei))
+		goto out_brelse;
 	raw_inode->i_dtime = cpu_to_le32(ei->i_dtime);
 	raw_inode->i_flags = cpu_to_le32(ei->i_flags);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 7be27dbe76bf..2b9dc96ec43e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1603,17 +1603,50 @@ static void ext4_orphan_cleanup (struct super_block * sb,
 
 /*
  * Maximal file size.  There is a direct, and {,double-,triple-}indirect
- * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks.
- * We need to be 1 filesystem block less than the 2^32 sector limit.
+ * block limit, and also a limit of (2^48 - 1) 512-byte sectors in i_blocks.
+ * We need to be 1 filesystem block less than the 2^48 sector limit.
  */
 static loff_t ext4_max_size(int bits)
 {
 	loff_t res = EXT4_NDIR_BLOCKS;
-	/* This constant is calculated to be the largest file size for a
-	 * dense, 4k-blocksize file such that the total number of
+	int meta_blocks;
+	loff_t upper_limit;
+	/* This is calculated to be the largest file size for a
+	 * dense, file such that the total number of
 	 * sectors in the file, including data and all indirect blocks,
-	 * does not exceed 2^32. */
-	const loff_t upper_limit = 0x1ff7fffd000LL;
+	 * does not exceed 2^48 -1
+	 * __u32 i_blocks_lo and _u16 i_blocks_high representing the
+	 * total number of  512 bytes blocks of the file
+	 */
+
+	if (sizeof(blkcnt_t) < sizeof(u64)) {
+		/*
+		 * CONFIG_LSF is not enabled implies the inode
+		 * i_block represent total blocks in 512 bytes
+		 * 32 == size of vfs inode i_blocks * 8
+		 */
+		upper_limit = (1LL << 32) - 1;
+
+		/* total blocks in file system block size */
+		upper_limit >>= (bits - 9);
+
+	} else {
+		/* We use 48 bit ext4_inode i_blocks */
+		upper_limit = (1LL << 48) - 1;
+
+		/* total blocks in file system block size */
+		upper_limit >>= (bits - 9);
+	}
+
+	/* indirect blocks */
+	meta_blocks = 1;
+	/* double indirect blocks */
+	meta_blocks += 1 + (1LL << (bits-2));
+	/* tripple indirect blocks */
+	meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2)));
+
+	upper_limit -= meta_blocks;
+	upper_limit <<= bits;
 
 	res += 1LL << (bits-2);
 	res += 1LL << (2*(bits-2));
@@ -1621,6 +1654,10 @@ static loff_t ext4_max_size(int bits)
 	res <<= bits;
 	if (res > upper_limit)
 		res = upper_limit;
+
+	if (res > MAX_LFS_FILESIZE)
+		res = MAX_LFS_FILESIZE;
+
 	return res;
 }
 
@@ -1789,6 +1826,19 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		       sb->s_id, le32_to_cpu(features));
 		goto failed_mount;
 	}
+	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
+		/*
+		 * Large file size enabled file system can only be
+		 * mount if kernel is build with CONFIG_LSF
+		 */
+		if (sizeof(root->i_blocks) < sizeof(u64) &&
+				!(sb->s_flags & MS_RDONLY)) {
+			printk(KERN_ERR "EXT4-fs: %s: Filesystem with huge "
+					"files cannot be mounted read-write "
+					"without CONFIG_LSF.\n", sb->s_id);
+			goto failed_mount;
+		}
+	}
 	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
 
 	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index a8f3faea8eff..be25eca9c040 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -282,7 +282,7 @@ struct ext4_inode {
 	__le32	i_dtime;	/* Deletion Time */
 	__le16	i_gid;		/* Low 16 bits of Group Id */
 	__le16	i_links_count;	/* Links count */
-	__le32	i_blocks;	/* Blocks count */
+	__le32	i_blocks_lo;	/* Blocks count */
 	__le32	i_flags;	/* File flags */
 	union {
 		struct {
@@ -302,7 +302,7 @@ struct ext4_inode {
 	__le32	i_obso_faddr;	/* Obsoleted fragment address */
 	union {
 		struct {
-			__le16	l_i_reserved1;	/* Obsoleted fragment number/size which are removed in ext4 */
+			__le16	l_i_blocks_high; /* were l_i_reserved1 */
 			__le16	l_i_file_acl_high;
 			__le16	l_i_uid_high;	/* these 2 fields */
 			__le16	l_i_gid_high;	/* were reserved2[0] */
@@ -404,6 +404,7 @@ do {									       \
 #if defined(__KERNEL__) || defined(__linux__)
 #define i_reserved1	osd1.linux1.l_i_reserved1
 #define i_file_acl_high	osd2.linux2.l_i_file_acl_high
+#define i_blocks_high	osd2.linux2.l_i_blocks_high
 #define i_uid_low	i_uid
 #define i_gid_low	i_gid
 #define i_uid_high	osd2.linux2.l_i_uid_high
@@ -670,6 +671,7 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 #define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER	0x0001
 #define EXT4_FEATURE_RO_COMPAT_LARGE_FILE	0x0002
 #define EXT4_FEATURE_RO_COMPAT_BTREE_DIR	0x0004
+#define EXT4_FEATURE_RO_COMPAT_HUGE_FILE        0x0008
 #define EXT4_FEATURE_RO_COMPAT_GDT_CSUM		0x0010
 #define EXT4_FEATURE_RO_COMPAT_DIR_NLINK	0x0020
 #define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE	0x0040
@@ -681,6 +683,7 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 #define EXT4_FEATURE_INCOMPAT_META_BG		0x0010
 #define EXT4_FEATURE_INCOMPAT_EXTENTS		0x0040 /* extents support */
 #define EXT4_FEATURE_INCOMPAT_64BIT		0x0080
+#define EXT4_FEATURE_INCOMPAT_MMP               0x0100
 #define EXT4_FEATURE_INCOMPAT_FLEX_BG		0x0200
 
 #define EXT4_FEATURE_COMPAT_SUPP	EXT2_FEATURE_COMPAT_EXT_ATTR
@@ -695,7 +698,8 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 					 EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \
 					 EXT4_FEATURE_RO_COMPAT_DIR_NLINK | \
 					 EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE | \
-					 EXT4_FEATURE_RO_COMPAT_BTREE_DIR)
+					 EXT4_FEATURE_RO_COMPAT_BTREE_DIR |\
+					 EXT4_FEATURE_RO_COMPAT_HUGE_FILE)
 
 /*
  * Default values for user and/or group using reserved blocks
-- 
cgit v1.2.3


From 8180a5627d126362c2f64e4fa886d6f608d9632a Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Support large files

This patch converts ext4_inode i_blocks to represent total
blocks occupied by the inode in file system block size.
Earlier the variable used to represent this in 512 byte
block size. This actually limited the total size of the file.

The feature is enabled transparently when we write an inode
whose i_blocks cannot be represnted as 512 byte units in a
48 bit variable.

inode flag  EXT4_HUGE_FILE_FL

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/inode.c         | 32 +++++++++++++++++++++++++-------
 fs/ext4/super.c         |  9 ++++++---
 include/linux/ext4_fs.h |  3 ++-
 3 files changed, 33 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index bb89fe727bb1..9cf85721d83c 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2671,14 +2671,20 @@ static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
 					struct ext4_inode_info *ei)
 {
 	blkcnt_t i_blocks ;
-	struct super_block *sb = ei->vfs_inode.i_sb;
+	struct inode *inode = &(ei->vfs_inode);
+	struct super_block *sb = inode->i_sb;
 
 	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
 				EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
 		/* we are using combined 48 bit field */
 		i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 |
 					le32_to_cpu(raw_inode->i_blocks_lo);
-		return i_blocks;
+		if (ei->i_flags & EXT4_HUGE_FILE_FL) {
+			/* i_blocks represent file system block size */
+			return i_blocks  << (inode->i_blkbits - 9);
+		} else {
+			return i_blocks;
+		}
 	} else {
 		return le32_to_cpu(raw_inode->i_blocks_lo);
 	}
@@ -2829,8 +2835,9 @@ static int ext4_inode_blocks_set(handle_t *handle,
 		 * i_blocks can be represnted in a 32 bit variable
 		 * as multiple of 512 bytes
 		 */
-		raw_inode->i_blocks_lo   = cpu_to_le32((u32)i_blocks);
+		raw_inode->i_blocks_lo   = cpu_to_le32(i_blocks);
 		raw_inode->i_blocks_high = 0;
+		ei->i_flags &= ~EXT4_HUGE_FILE_FL;
 	} else if (i_blocks <= 0xffffffffffffULL) {
 		/*
 		 * i_blocks can be represented in a 48 bit variable
@@ -2841,12 +2848,23 @@ static int ext4_inode_blocks_set(handle_t *handle,
 		if (err)
 			goto  err_out;
 		/* i_block is stored in the split  48 bit fields */
-		raw_inode->i_blocks_lo   = cpu_to_le32((u32)i_blocks);
+		raw_inode->i_blocks_lo   = cpu_to_le32(i_blocks);
 		raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32);
+		ei->i_flags &= ~EXT4_HUGE_FILE_FL;
 	} else {
-		ext4_error(sb, __FUNCTION__,
-				"Wrong inode i_blocks count  %llu\n",
-				(unsigned long long)inode->i_blocks);
+		/*
+		 * i_blocks should be represented in a 48 bit variable
+		 * as multiple of  file system block size
+		 */
+		err = ext4_update_rocompat_feature(handle, sb,
+					    EXT4_FEATURE_RO_COMPAT_HUGE_FILE);
+		if (err)
+			goto  err_out;
+		ei->i_flags |= EXT4_HUGE_FILE_FL;
+		/* i_block is stored in file system block size */
+		i_blocks = i_blocks >> (inode->i_blkbits - 9);
+		raw_inode->i_blocks_lo   = cpu_to_le32(i_blocks);
+		raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32);
 	}
 err_out:
 	return err;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 2b9dc96ec43e..64067de70c6f 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1631,11 +1631,14 @@ static loff_t ext4_max_size(int bits)
 		upper_limit >>= (bits - 9);
 
 	} else {
-		/* We use 48 bit ext4_inode i_blocks */
+		/*
+		 * We use 48 bit ext4_inode i_blocks
+		 * With EXT4_HUGE_FILE_FL set the i_blocks
+		 * represent total number of blocks in
+		 * file system block size
+		 */
 		upper_limit = (1LL << 48) - 1;
 
-		/* total blocks in file system block size */
-		upper_limit >>= (bits - 9);
 	}
 
 	/* indirect blocks */
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index be25eca9c040..6ae91f40aaa2 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -178,8 +178,9 @@ struct ext4_group_desc
 #define EXT4_NOTAIL_FL			0x00008000 /* file tail should not be merged */
 #define EXT4_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
 #define EXT4_TOPDIR_FL			0x00020000 /* Top of directory hierarchies*/
-#define EXT4_RESERVED_FL		0x80000000 /* reserved for ext4 lib */
+#define EXT4_HUGE_FILE_FL               0x00040000 /* Set to each huge file */
 #define EXT4_EXTENTS_FL			0x00080000 /* Inode uses extents */
+#define EXT4_RESERVED_FL		0x80000000 /* reserved for ext4 lib */
 
 #define EXT4_FL_USER_VISIBLE		0x000BDFFF /* User visible flags */
 #define EXT4_FL_USER_MODIFIABLE		0x000380FF /* User modifiable flags */
-- 
cgit v1.2.3


From e2b4657453c0d5571bd3c7256585c486ed42d364 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@redhat.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: store maxbytes for bitmapped  files and return EFBIG as
 appropriate

Calculate & store the max offset for bitmapped files, and
catch too-large seeks, truncates, and writes in ext4, shortening
or rejecting as appropriate.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---
 fs/ext4/file.c             | 19 ++++++++++++++++++-
 fs/ext4/inode.c            | 16 +++++++++++++++-
 fs/ext4/super.c            |  1 +
 include/linux/ext4_fs_sb.h |  1 +
 4 files changed, 35 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 1a81cd66d63b..a6b2aa14626e 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -56,8 +56,25 @@ ext4_file_write(struct kiocb *iocb, const struct iovec *iov,
 	ssize_t ret;
 	int err;
 
-	ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
+	/*
+	 * If we have encountered a bitmap-format file, the size limit
+	 * is smaller than s_maxbytes, which is for extent-mapped files.
+	 */
+
+	if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) {
+		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+		size_t length = iov_length(iov, nr_segs);
 
+		if (pos > sbi->s_bitmap_maxbytes)
+			return -EFBIG;
+
+		if (pos + length > sbi->s_bitmap_maxbytes) {
+			nr_segs = iov_shorten((struct iovec *)iov, nr_segs,
+					      sbi->s_bitmap_maxbytes - pos);
+		}
+	}
+
+	ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
 	/*
 	 * Skip flushing if there was an error, or if nothing was written.
 	 */
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 9cf85721d83c..eaace1373ccb 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -314,7 +314,10 @@ static int ext4_block_to_path(struct inode *inode,
 		offsets[n++] = i_block & (ptrs - 1);
 		final = ptrs;
 	} else {
-		ext4_warning(inode->i_sb, "ext4_block_to_path", "block > big");
+		ext4_warning(inode->i_sb, "ext4_block_to_path",
+				"block %u > max",
+				i_block + direct_blocks +
+				indirect_blocks + double_blocks);
 	}
 	if (boundary)
 		*boundary = final - 1 - (i_block & (ptrs - 1));
@@ -3092,6 +3095,17 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 		ext4_journal_stop(handle);
 	}
 
+	if (attr->ia_valid & ATTR_SIZE) {
+		if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) {
+			struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+
+			if (attr->ia_size > sbi->s_bitmap_maxbytes) {
+				error = -EFBIG;
+				goto err_out;
+			}
+		}
+	}
+
 	if (S_ISREG(inode->i_mode) &&
 	    attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) {
 		handle_t *handle;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index c79e46b7f159..0931831537a2 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1922,6 +1922,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		}
 	}
 
+	sbi->s_bitmap_maxbytes = ext4_max_bitmap_size(sb->s_blocksize_bits);
 	sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits);
 
 	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h
index f15821c5e4c4..38a47ec06df9 100644
--- a/include/linux/ext4_fs_sb.h
+++ b/include/linux/ext4_fs_sb.h
@@ -38,6 +38,7 @@ struct ext4_sb_info {
 	ext4_group_t s_groups_count;	/* Number of groups in the fs */
 	unsigned long s_overhead_last;  /* Last calculated overhead */
 	unsigned long s_blocks_last;    /* Last seen block count */
+	loff_t s_bitmap_maxbytes;	/* max bytes for bitmap files */
 	struct buffer_head * s_sbh;	/* Buffer containing the super block */
 	struct ext4_super_block * s_es;	/* Pointer to the super block in the buffer */
 	struct buffer_head ** s_group_desc;
-- 
cgit v1.2.3


From 91b51a018d7711b20e9e0bb14c3d790de4e310d4 Mon Sep 17 00:00:00 2001
From: Coly Li <coyli@suse.de>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: sync up block group descriptor with e2fsprogs.

This patch extends bg_itable_unused of ext4 group descriptor
from 16bit into 32bit. In order to add bg_itable_unused_hi into
struct ext4_group_desc, some extra fields which are already introduced into
e2fsprogs are also added in for consistency.

Signed-off-by: Coly Li <coyli@suse.de>
Cc: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---
 include/linux/ext4_fs.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 6ae91f40aaa2..55a376e7bca0 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -118,6 +118,11 @@ struct ext4_group_desc
 	__le32	bg_block_bitmap_hi;	/* Blocks bitmap block MSB */
 	__le32	bg_inode_bitmap_hi;	/* Inodes bitmap block MSB */
 	__le32	bg_inode_table_hi;	/* Inodes table block MSB */
+	__le16	bg_free_blocks_count_hi;/* Free blocks count MSB */
+	__le16	bg_free_inodes_count_hi;/* Free inodes count MSB */
+	__le16	bg_used_dirs_count_hi;	/* Directories count MSB */
+	__le16	bg_itable_unused_hi;	/* Unused inodes count MSB */
+	__u32	bg_reserved2[3];
 };
 
 #define EXT4_BG_INODE_UNINIT	0x0001 /* Inode table/bitmap not in use */
-- 
cgit v1.2.3


From 389d1b083c767a360ec84b27a95da06244becec8 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:26 -0500
Subject: Add buffer head related helper functions

Add buffer head related helper function bh_uptodate_or_lock and
bh_submit_read which can be used by file system

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/buffer.c                 | 44 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/buffer_head.h |  2 ++
 2 files changed, 46 insertions(+)

(limited to 'include')

diff --git a/fs/buffer.c b/fs/buffer.c
index 7249e014819e..456c9ab7705b 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -3213,6 +3213,50 @@ static int buffer_cpu_notify(struct notifier_block *self,
 	return NOTIFY_OK;
 }
 
+/**
+ * bh_uptodate_or_lock: Test whether the buffer is uptodate
+ * @bh: struct buffer_head
+ *
+ * Return true if the buffer is up-to-date and false,
+ * with the buffer locked, if not.
+ */
+int bh_uptodate_or_lock(struct buffer_head *bh)
+{
+	if (!buffer_uptodate(bh)) {
+		lock_buffer(bh);
+		if (!buffer_uptodate(bh))
+			return 0;
+		unlock_buffer(bh);
+	}
+	return 1;
+}
+EXPORT_SYMBOL(bh_uptodate_or_lock);
+
+/**
+ * bh_submit_read: Submit a locked buffer for reading
+ * @bh: struct buffer_head
+ *
+ * Returns zero on success and -EIO on error.
+ */
+int bh_submit_read(struct buffer_head *bh)
+{
+	BUG_ON(!buffer_locked(bh));
+
+	if (buffer_uptodate(bh)) {
+		unlock_buffer(bh);
+		return 0;
+	}
+
+	get_bh(bh);
+	bh->b_end_io = end_buffer_read_sync;
+	submit_bh(READ, bh);
+	wait_on_buffer(bh);
+	if (buffer_uptodate(bh))
+		return 0;
+	return -EIO;
+}
+EXPORT_SYMBOL(bh_submit_read);
+
 void __init buffer_init(void)
 {
 	int nrpages;
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index da0d83fbadc0..e98801f06dcc 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -192,6 +192,8 @@ int sync_dirty_buffer(struct buffer_head *bh);
 int submit_bh(int, struct buffer_head *);
 void write_boundary_block(struct block_device *bdev,
 			sector_t bblock, unsigned blocksize);
+int bh_uptodate_or_lock(struct buffer_head *bh);
+int bh_submit_read(struct buffer_head *bh);
 
 extern int buffer_heads_over_limit;
 
-- 
cgit v1.2.3


From 36df53f4a3e445175fc1e9d7f433599482ec6d7f Mon Sep 17 00:00:00 2001
From: Chris Snook <csnook@redhat.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: jbd2: Remove printk from J_ASSERT to preserve registers during BUG

Signed-off-by: Chris Snook <csnook@redhat.com>
Cc: "Stephen C. Tweedie" <sct@redhat.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 include/linux/jbd2.h | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 06ef11457051..d5f7cff4cb28 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -256,17 +256,7 @@ typedef struct journal_superblock_s
 #include <linux/fs.h>
 #include <linux/sched.h>
 
-#define JBD2_ASSERTIONS
-#ifdef JBD2_ASSERTIONS
-#define J_ASSERT(assert)						\
-do {									\
-	if (!(assert)) {						\
-		printk (KERN_EMERG					\
-			"Assertion failure in %s() at %s:%d: \"%s\"\n",	\
-			__FUNCTION__, __FILE__, __LINE__, # assert);	\
-		BUG();							\
-	}								\
-} while (0)
+#define J_ASSERT(assert)	BUG_ON(!(assert))
 
 #if defined(CONFIG_BUFFER_DEBUG)
 void buffer_assertion_failure(struct buffer_head *bh);
@@ -282,10 +272,6 @@ void buffer_assertion_failure(struct buffer_head *bh);
 #define J_ASSERT_JH(jh, expr)	J_ASSERT(expr)
 #endif
 
-#else
-#define J_ASSERT(assert)	do { } while (0)
-#endif		/* JBD2_ASSERTIONS */
-
 #if defined(JBD2_PARANOID_IOFAIL)
 #define J_EXPECT(expr, why...)		J_ASSERT(expr)
 #define J_EXPECT_BH(bh, expr, why...)	J_ASSERT_BH(bh, expr)
-- 
cgit v1.2.3


From f5a7a6b0d9b6af7d46124ed3f6b3995225cb62d0 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: jbd2: Fix assertion failure in fs/jbd2/checkpoint.c

Before we start committing a transaction, we call
__journal_clean_checkpoint_list() to cleanup transaction's written-back
buffers.

If this call happens to remove all of them (and there were already some
buffers), __journal_remove_checkpoint() will decide to free the transaction
because it isn't (yet) a committing transaction and soon we fail some
assertion - the transaction really isn't ready to be freed :).

We change the check in __journal_remove_checkpoint() to free only a
transaction in T_FINISHED state.  The locking there is subtle though (as
everywhere in JBD ;().  We use j_list_lock to protect the check and a
subsequent call to __journal_drop_transaction() and do the same in the end
of journal_commit_transaction() which is the only place where a transaction
can get to T_FINISHED state.

Probably I'm too paranoid here and such locking is not really necessary -
checkpoint lists are processed only from log_do_checkpoint() where a
transaction must be already committed to be processed or from
__journal_clean_checkpoint_list() where kjournald itself calls it and thus
transaction cannot change state either.  Better be safe if something
changes in future...

Signed-off-by: Jan Kara <jack@suse.cz>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 fs/jbd2/checkpoint.c | 12 ++++++------
 fs/jbd2/commit.c     |  8 ++++----
 include/linux/jbd2.h |  2 ++
 3 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 3fccde7ba008..7e958c86242f 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -602,15 +602,15 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh)
 
 	/*
 	 * There is one special case to worry about: if we have just pulled the
-	 * buffer off a committing transaction's forget list, then even if the
-	 * checkpoint list is empty, the transaction obviously cannot be
-	 * dropped!
+	 * buffer off a running or committing transaction's checkpoing list,
+	 * then even if the checkpoint list is empty, the transaction obviously
+	 * cannot be dropped!
 	 *
-	 * The locking here around j_committing_transaction is a bit sleazy.
+	 * The locking here around t_state is a bit sleazy.
 	 * See the comment at the end of jbd2_journal_commit_transaction().
 	 */
-	if (transaction == journal->j_committing_transaction) {
-		JBUFFER_TRACE(jh, "belongs to committing transaction");
+	if (transaction->t_state != T_FINISHED) {
+		JBUFFER_TRACE(jh, "belongs to running/committing transaction");
 		goto out;
 	}
 
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 6986f334c643..39b5cee3dd8a 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -867,10 +867,10 @@ restart_loop:
 	}
 	spin_unlock(&journal->j_list_lock);
 	/*
-	 * This is a bit sleazy.  We borrow j_list_lock to protect
-	 * journal->j_committing_transaction in __jbd2_journal_remove_checkpoint.
-	 * Really, __jbd2_journal_remove_checkpoint should be using j_state_lock but
-	 * it's a bit hassle to hold that across __jbd2_journal_remove_checkpoint
+	 * This is a bit sleazy.  We use j_list_lock to protect transition
+	 * of a transaction into T_FINISHED state and calling
+	 * __jbd2_journal_drop_transaction(). Otherwise we could race with
+	 * other checkpointing code processing the transaction...
 	 */
 	spin_lock(&journal->j_state_lock);
 	spin_lock(&journal->j_list_lock);
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index d5f7cff4cb28..d861ffd49821 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -442,6 +442,8 @@ struct transaction_s
 	/*
 	 * Transaction's current state
 	 * [no locking - only kjournald2 alters this]
+	 * [j_list_lock] guards transition of a transaction into T_FINISHED
+	 * state and subsequent call of __jbd2_journal_drop_transaction()
 	 * FIXME: needs barriers
 	 * KLUDGE: [use j_state_lock]
 	 */
-- 
cgit v1.2.3


From c278bfecebfb1ed67c326ef472660878baa745cd Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Make ext4_get_blocks_wrap take the truncate_mutex early.

When doing a migrate from ext3 to ext4 inode we need to make sure the test
for inode type and walking inode data happens inside  lock. To make this
happen move truncate_mutex early before checking the i_flags.


This actually should enable us to remove the verify_chain().

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/extents.c       |  9 ++++---
 fs/ext4/inode.c         | 69 ++++++-------------------------------------------
 include/linux/ext4_fs.h |  2 ++
 3 files changed, 16 insertions(+), 64 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 8593e59020fe..ec5019fa552f 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -2129,6 +2129,10 @@ out:
 	return err ? err : allocated;
 }
 
+/*
+ * Need to be called with
+ * mutex_lock(&EXT4_I(inode)->truncate_mutex);
+ */
 int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 			ext4_lblk_t iblock,
 			unsigned long max_blocks, struct buffer_head *bh_result,
@@ -2144,7 +2148,6 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	__clear_bit(BH_New, &bh_result->b_state);
 	ext_debug("blocks %u/%lu requested for inode %u\n",
 			iblock, max_blocks, inode->i_ino);
-	mutex_lock(&EXT4_I(inode)->truncate_mutex);
 
 	/* check in cache */
 	goal = ext4_ext_in_cache(inode, iblock, &newex);
@@ -2318,8 +2321,6 @@ out2:
 		ext4_ext_drop_refs(path);
 		kfree(path);
 	}
-	mutex_unlock(&EXT4_I(inode)->truncate_mutex);
-
 	return err ? err : allocated;
 }
 
@@ -2449,6 +2450,7 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
 	 * modify 1 super block, 1 block bitmap and 1 group descriptor.
 	 */
 	credits = EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + 3;
+	mutex_lock(&EXT4_I(inode)->truncate_mutex)
 retry:
 	while (ret >= 0 && ret < max_blocks) {
 		block = block + ret;
@@ -2505,6 +2507,7 @@ retry:
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
 
+	mutex_unlock(&EXT4_I(inode)->truncate_mutex)
 	/*
 	 * Time to update the file size.
 	 * Update only when preallocation was requested beyond the file size.
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index eaace1373ccb..71c7ad0c6723 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -243,13 +243,6 @@ static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v)
 	p->bh = bh;
 }
 
-static int verify_chain(Indirect *from, Indirect *to)
-{
-	while (from <= to && from->key == *from->p)
-		from++;
-	return (from > to);
-}
-
 /**
  *	ext4_block_to_path - parse the block number into array of offsets
  *	@inode: inode in question (we are only interested in its superblock)
@@ -348,10 +341,11 @@ static int ext4_block_to_path(struct inode *inode,
  *		(pointer to last triple returned, *@err == 0)
  *	or when it gets an IO error reading an indirect block
  *		(ditto, *@err == -EIO)
- *	or when it notices that chain had been changed while it was reading
- *		(ditto, *@err == -EAGAIN)
  *	or when it reads all @depth-1 indirect blocks successfully and finds
  *	the whole chain, all way to the data (returns %NULL, *err == 0).
+ *
+ *      Need to be called with
+ *      mutex_lock(&EXT4_I(inode)->truncate_mutex)
  */
 static Indirect *ext4_get_branch(struct inode *inode, int depth,
 				 ext4_lblk_t  *offsets,
@@ -370,9 +364,6 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth,
 		bh = sb_bread(sb, le32_to_cpu(p->key));
 		if (!bh)
 			goto failure;
-		/* Reader: pointers */
-		if (!verify_chain(chain, p))
-			goto changed;
 		add_chain(++p, bh, (__le32*)bh->b_data + *++offsets);
 		/* Reader: end */
 		if (!p->key)
@@ -380,10 +371,6 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth,
 	}
 	return NULL;
 
-changed:
-	brelse(bh);
-	*err = -EAGAIN;
-	goto no_block;
 failure:
 	*err = -EIO;
 no_block:
@@ -787,6 +774,10 @@ err_out:
  * return > 0, # of blocks mapped or allocated.
  * return = 0, if plain lookup failed.
  * return < 0, error case.
+ *
+ *
+ * Need to be called with
+ * mutex_lock(&EXT4_I(inode)->truncate_mutex)
  */
 int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 		ext4_lblk_t iblock, unsigned long maxblocks,
@@ -825,18 +816,6 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 		while (count < maxblocks && count <= blocks_to_boundary) {
 			ext4_fsblk_t blk;
 
-			if (!verify_chain(chain, partial)) {
-				/*
-				 * Indirect block might be removed by
-				 * truncate while we were reading it.
-				 * Handling of that case: forget what we've
-				 * got now. Flag the err as EAGAIN, so it
-				 * will reread.
-				 */
-				err = -EAGAIN;
-				count = 0;
-				break;
-			}
 			blk = le32_to_cpu(*(chain[depth-1].p + count));
 
 			if (blk == first_block + count)
@@ -844,44 +823,13 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 			else
 				break;
 		}
-		if (err != -EAGAIN)
-			goto got_it;
+		goto got_it;
 	}
 
 	/* Next simple case - plain lookup or failed read of indirect block */
 	if (!create || err == -EIO)
 		goto cleanup;
 
-	mutex_lock(&ei->truncate_mutex);
-
-	/*
-	 * If the indirect block is missing while we are reading
-	 * the chain(ext4_get_branch() returns -EAGAIN err), or
-	 * if the chain has been changed after we grab the semaphore,
-	 * (either because another process truncated this branch, or
-	 * another get_block allocated this branch) re-grab the chain to see if
-	 * the request block has been allocated or not.
-	 *
-	 * Since we already block the truncate/other get_block
-	 * at this point, we will have the current copy of the chain when we
-	 * splice the branch into the tree.
-	 */
-	if (err == -EAGAIN || !verify_chain(chain, partial)) {
-		while (partial > chain) {
-			brelse(partial->bh);
-			partial--;
-		}
-		partial = ext4_get_branch(inode, depth, offsets, chain, &err);
-		if (!partial) {
-			count++;
-			mutex_unlock(&ei->truncate_mutex);
-			if (err)
-				goto cleanup;
-			clear_buffer_new(bh_result);
-			goto got_it;
-		}
-	}
-
 	/*
 	 * Okay, we need to do block allocation.  Lazily initialize the block
 	 * allocation info here if necessary
@@ -923,7 +871,6 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 	*/
 	if (!err && extend_disksize && inode->i_size > ei->i_disksize)
 		ei->i_disksize = inode->i_size;
-	mutex_unlock(&ei->truncate_mutex);
 	if (err)
 		goto cleanup;
 
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 55a376e7bca0..583049c1d366 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -1113,6 +1113,7 @@ ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 			int create, int extend_disksize)
 {
 	int retval;
+	mutex_lock(&EXT4_I(inode)->truncate_mutex);
 	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) {
 		retval = ext4_ext_get_blocks(handle, inode,
 						(ext4_lblk_t)block, max_blocks,
@@ -1122,6 +1123,7 @@ ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 						(ext4_lblk_t)block, max_blocks,
 						bh, create, extend_disksize);
 	}
+	mutex_unlock(&EXT4_I(inode)->truncate_mutex);
 	return retval;
 }
 
-- 
cgit v1.2.3


From 0e855ac8b103ef579052936b59fe7c599ac422a4 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:26 -0500
Subject: ext4: Convert truncate_mutex to read write semaphore.

We are currently taking the truncate_mutex for every read. This would have
performance impact on large CPU configuration. Convert the lock to read write
semaphore and take read lock when we are trying to read the file.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/balloc.c          |  2 +-
 fs/ext4/extents.c         | 13 +++++++------
 fs/ext4/file.c            |  4 ++--
 fs/ext4/inode.c           | 40 +++++++++++++++++++++++++++++++++-------
 fs/ext4/ioctl.c           |  4 ++--
 fs/ext4/super.c           |  2 +-
 include/linux/ext4_fs.h   | 25 ++++---------------------
 include/linux/ext4_fs_i.h |  6 +++---
 8 files changed, 53 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index d460223b8e1d..7ae223ed152f 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -526,7 +526,7 @@ static inline int rsv_is_empty(struct ext4_reserve_window *rsv)
  * when setting the reservation window size through ioctl before the file
  * is open for write (needs block allocation).
  *
- * Needs truncate_mutex protection prior to call this function.
+ * Needs down_write(i_data_sem) protection prior to call this function.
  */
 void ext4_init_block_alloc_info(struct inode *inode)
 {
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index ec5019fa552f..03d1bbb78a2f 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1565,7 +1565,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
  * This routine returns max. credits that the extent tree can consume.
  * It should be OK for low-performance paths like ->writepage()
  * To allow many writing processes to fit into a single transaction,
- * the caller should calculate credits under truncate_mutex and
+ * the caller should calculate credits under i_data_sem and
  * pass the actual path.
  */
 int ext4_ext_calc_credits_for_insert(struct inode *inode,
@@ -2131,7 +2131,8 @@ out:
 
 /*
  * Need to be called with
- * mutex_lock(&EXT4_I(inode)->truncate_mutex);
+ * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block
+ * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem)
  */
 int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 			ext4_lblk_t iblock,
@@ -2350,7 +2351,7 @@ void ext4_ext_truncate(struct inode * inode, struct page *page)
 	if (page)
 		ext4_block_truncate_page(handle, page, mapping, inode->i_size);
 
-	mutex_lock(&EXT4_I(inode)->truncate_mutex);
+	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_ext_invalidate_cache(inode);
 
 	/*
@@ -2386,7 +2387,7 @@ out_stop:
 	if (inode->i_nlink)
 		ext4_orphan_del(handle, inode);
 
-	mutex_unlock(&EXT4_I(inode)->truncate_mutex);
+	up_write(&EXT4_I(inode)->i_data_sem);
 	ext4_journal_stop(handle);
 }
 
@@ -2450,7 +2451,7 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
 	 * modify 1 super block, 1 block bitmap and 1 group descriptor.
 	 */
 	credits = EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + 3;
-	mutex_lock(&EXT4_I(inode)->truncate_mutex)
+	down_write((&EXT4_I(inode)->i_data_sem));
 retry:
 	while (ret >= 0 && ret < max_blocks) {
 		block = block + ret;
@@ -2507,7 +2508,7 @@ retry:
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
 
-	mutex_unlock(&EXT4_I(inode)->truncate_mutex)
+	up_write((&EXT4_I(inode)->i_data_sem));
 	/*
 	 * Time to update the file size.
 	 * Update only when preallocation was requested beyond the file size.
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index a6b2aa14626e..ac35ec58db55 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -37,9 +37,9 @@ static int ext4_release_file (struct inode * inode, struct file * filp)
 	if ((filp->f_mode & FMODE_WRITE) &&
 			(atomic_read(&inode->i_writecount) == 1))
 	{
-		mutex_lock(&EXT4_I(inode)->truncate_mutex);
+		down_write(&EXT4_I(inode)->i_data_sem);
 		ext4_discard_reservation(inode);
-		mutex_unlock(&EXT4_I(inode)->truncate_mutex);
+		up_write(&EXT4_I(inode)->i_data_sem);
 	}
 	if (is_dx(inode) && filp->private_data)
 		ext4_htree_free_dir_info(filp->private_data);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 71c7ad0c6723..a7eb8bb4bdd4 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -308,7 +308,7 @@ static int ext4_block_to_path(struct inode *inode,
 		final = ptrs;
 	} else {
 		ext4_warning(inode->i_sb, "ext4_block_to_path",
-				"block %u > max",
+				"block %lu > max",
 				i_block + direct_blocks +
 				indirect_blocks + double_blocks);
 	}
@@ -345,7 +345,7 @@ static int ext4_block_to_path(struct inode *inode,
  *	the whole chain, all way to the data (returns %NULL, *err == 0).
  *
  *      Need to be called with
- *      mutex_lock(&EXT4_I(inode)->truncate_mutex)
+ *      down_read(&EXT4_I(inode)->i_data_sem)
  */
 static Indirect *ext4_get_branch(struct inode *inode, int depth,
 				 ext4_lblk_t  *offsets,
@@ -777,7 +777,8 @@ err_out:
  *
  *
  * Need to be called with
- * mutex_lock(&EXT4_I(inode)->truncate_mutex)
+ * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block
+ * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem)
  */
 int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 		ext4_lblk_t iblock, unsigned long maxblocks,
@@ -865,7 +866,7 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 		err = ext4_splice_branch(handle, inode, iblock,
 					partial, indirect_blks, count);
 	/*
-	 * i_disksize growing is protected by truncate_mutex.  Don't forget to
+	 * i_disksize growing is protected by i_data_sem.  Don't forget to
 	 * protect it if you're about to implement concurrent
 	 * ext4_get_block() -bzzz
 	*/
@@ -895,6 +896,31 @@ out:
 
 #define DIO_CREDITS (EXT4_RESERVE_TRANS_BLOCKS + 32)
 
+int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
+			unsigned long max_blocks, struct buffer_head *bh,
+			int create, int extend_disksize)
+{
+	int retval;
+	if (create) {
+		down_write((&EXT4_I(inode)->i_data_sem));
+	} else {
+		down_read((&EXT4_I(inode)->i_data_sem));
+	}
+	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) {
+		retval =  ext4_ext_get_blocks(handle, inode, block, max_blocks,
+				bh, create, extend_disksize);
+	} else {
+		retval = ext4_get_blocks_handle(handle, inode, block,
+				max_blocks, bh, create, extend_disksize);
+	}
+	if (create) {
+		up_write((&EXT4_I(inode)->i_data_sem));
+	} else {
+		up_read((&EXT4_I(inode)->i_data_sem));
+	}
+	return retval;
+}
+
 static int ext4_get_block(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create)
 {
@@ -1399,7 +1425,7 @@ static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh)
  *	ext4_file_write() -> generic_file_write() -> __alloc_pages() -> ...
  *
  * Same applies to ext4_get_block().  We will deadlock on various things like
- * lock_journal and i_truncate_mutex.
+ * lock_journal and i_data_sem
  *
  * Setting PF_MEMALLOC here doesn't work - too many internal memory
  * allocations fail.
@@ -2325,7 +2351,7 @@ void ext4_truncate(struct inode *inode)
 	 * From here we block out all ext4_get_block() callers who want to
 	 * modify the block allocation tree.
 	 */
-	mutex_lock(&ei->truncate_mutex);
+	down_write(&ei->i_data_sem);
 
 	if (n == 1) {		/* direct blocks */
 		ext4_free_data(handle, inode, NULL, i_data+offsets[0],
@@ -2389,7 +2415,7 @@ do_indirects:
 
 	ext4_discard_reservation(inode);
 
-	mutex_unlock(&ei->truncate_mutex);
+	up_write(&ei->i_data_sem);
 	inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
 	ext4_mark_inode_dirty(handle, inode);
 
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index e7f894bdb420..c0e5b8cf635c 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -199,7 +199,7 @@ flags_err:
 		 * need to allocate reservation structure for this inode
 		 * before set the window size
 		 */
-		mutex_lock(&ei->truncate_mutex);
+		down_write(&ei->i_data_sem);
 		if (!ei->i_block_alloc_info)
 			ext4_init_block_alloc_info(inode);
 
@@ -207,7 +207,7 @@ flags_err:
 			struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
 			rsv->rsv_goal_size = rsv_window_size;
 		}
-		mutex_unlock(&ei->truncate_mutex);
+		up_write(&ei->i_data_sem);
 		return 0;
 	}
 	case EXT4_IOC_GROUP_EXTEND: {
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index effd375ece80..c7305443e100 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -593,7 +593,7 @@ static void init_once(struct kmem_cache *cachep, void *foo)
 #ifdef CONFIG_EXT4DEV_FS_XATTR
 	init_rwsem(&ei->xattr_sem);
 #endif
-	mutex_init(&ei->truncate_mutex);
+	init_rwsem(&ei->i_data_sem);
 	inode_init_once(&ei->vfs_inode);
 }
 
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 583049c1d366..300cc5a5adb9 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -1107,27 +1107,10 @@ extern void ext4_ext_init(struct super_block *);
 extern void ext4_ext_release(struct super_block *);
 extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset,
 			  loff_t len);
-static inline int
-ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
-			unsigned long max_blocks, struct buffer_head *bh,
-			int create, int extend_disksize)
-{
-	int retval;
-	mutex_lock(&EXT4_I(inode)->truncate_mutex);
-	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) {
-		retval = ext4_ext_get_blocks(handle, inode,
-						(ext4_lblk_t)block, max_blocks,
-						bh, create, extend_disksize);
-	} else {
-		retval = ext4_get_blocks_handle(handle, inode,
-						(ext4_lblk_t)block, max_blocks,
-						bh, create, extend_disksize);
-	}
-	mutex_unlock(&EXT4_I(inode)->truncate_mutex);
-	return retval;
-}
-
-
+extern int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode,
+			sector_t block, unsigned long max_blocks,
+			struct buffer_head *bh, int create,
+			int extend_disksize);
 #endif	/* __KERNEL__ */
 
 #endif	/* _LINUX_EXT4_FS_H */
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index f1cd4934e46f..4377d249d378 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -139,16 +139,16 @@ struct ext4_inode_info {
 	__u16 i_extra_isize;
 
 	/*
-	 * truncate_mutex is for serialising ext4_truncate() against
+	 * i_data_sem is for serialising ext4_truncate() against
 	 * ext4_getblock().  In the 2.4 ext2 design, great chunks of inode's
 	 * data tree are chopped off during truncate. We can't do that in
 	 * ext4 because whenever we perform intermediate commits during
 	 * truncate, the inode and all the metadata blocks *must* be in a
 	 * consistent state which allows truncation of the orphans to restart
 	 * during recovery.  Hence we must fix the get_block-vs-truncate race
-	 * by other means, so we have truncate_mutex.
+	 * by other means, so we have i_data_sem.
 	 */
-	struct mutex truncate_mutex;
+	struct rw_semaphore i_data_sem;
 	struct inode vfs_inode;
 
 	unsigned long i_ext_generation;
-- 
cgit v1.2.3


From 8e85fb3f305b24b79c6d9cb7a56d22b062335ad3 Mon Sep 17 00:00:00 2001
From: Johann Lombardi <johann.lombardi@bull.net>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: jbd2: jbd2 stats through procfs

The patch below updates the jbd stats patch to 2.6.20/jbd2.
The initial patch was posted by Alex Tomas in December 2005
(http://marc.info/?l=linux-ext4&m=113538565128617&w=2).
It provides statistics via procfs such as transaction lifetime and size.

Sometimes, investigating performance problems, i find useful to have
stats from jbd about transaction's lifetime, size, etc. here is a
patch for review and inclusion probably.

for example, stats after creation of 3M files in htree directory:

[root@bob ~]# cat /proc/fs/jbd/sda/history
R/C  tid   wait  run   lock  flush log   hndls  block inlog ctime write drop  close
R    261   8260  2720  0     0     750   9892   8170  8187
C    259                                                    750   0     4885  1
R    262   20    2200  10    0     770   9836   8170  8187
R    263   30    2200  10    0     3070  9812   8170  8187
R    264   0     5000  10    0     1340  0      0     0
C    261                                                    8240  3212  4957  0
R    265   8260  1470  0     0     4640  9854   8170  8187
R    266   0     5000  10    0     1460  0      0     0
C    262                                                    8210  2989  4868  0
R    267   8230  1490  10    0     4440  9875   8171  8188
R    268   0     5000  10    0     1260  0      0     0
C    263                                                    7710  2937  4908  0
R    269   7730  1470  10    0     3330  9841   8170  8187
R    270   0     5000  10    0     830   0      0     0
C    265                                                    8140  3234  4898  0
C    267                                                    720   0     4849  1
R    271   8630  2740  20    0     740   9819   8170  8187
C    269                                                    800   0     4214  1
R    272   40    2170  10    0     830   9716   8170  8187
R    273   40    2280  0     0     3530  9799   8170  8187
R    274   0     5000  10    0     990   0      0     0


where,

R     - line for transaction's life from T_RUNNING to T_FINISHED
C     - line for transaction's checkpointing
tid   - transaction's id
wait  - for how long we were waiting for new transaction to start
         (the longest period journal_start() took in this transaction)
run   - real transaction's lifetime (from T_RUNNING to T_LOCKED
lock  - how long we were waiting for all handles to close
         (time the transaction was in T_LOCKED)
flush - how long it took to flush all data (data=ordered)
log   - how long it took to write the transaction to the log
hndls - how many handles got to the transaction
block - how many blocks got to the transaction
inlog - how many blocks are written to the log (block + descriptors)
ctime - how long it took to checkpoint the transaction
write - how many blocks have been written during checkpointing
drop  - how many blocks have been dropped during checkpointing
close - how many running transactions have been closed to checkpoint this one

all times are in msec.


[root@bob ~]# cat /proc/fs/jbd/sda/info
280 transaction, each upto 8192 blocks
average:
  1633ms waiting for transaction
  3616ms running transaction
  5ms transaction was being locked
  1ms flushing data (in ordered mode)
  1799ms logging transaction
  11781 handles per transaction
  5629 blocks per transaction
  5641 logged blocks per transaction

Signed-off-by: Johann Lombardi <johann.lombardi@bull.net>
Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---
 fs/jbd2/checkpoint.c  |  10 +-
 fs/jbd2/commit.c      |  49 ++++++++
 fs/jbd2/journal.c     | 338 ++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/jbd2/transaction.c |   9 ++
 include/linux/jbd2.h  |  77 ++++++++++++
 5 files changed, 481 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 7e958c86242f..1b7f282c1ae9 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -232,7 +232,8 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
 static int __process_buffer(journal_t *journal, struct journal_head *jh,
-			struct buffer_head **bhs, int *batch_count)
+			struct buffer_head **bhs, int *batch_count,
+			transaction_t *transaction)
 {
 	struct buffer_head *bh = jh2bh(jh);
 	int ret = 0;
@@ -250,6 +251,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		transaction_t *t = jh->b_transaction;
 		tid_t tid = t->t_tid;
 
+		transaction->t_chp_stats.cs_forced_to_close++;
 		spin_unlock(&journal->j_list_lock);
 		jbd_unlock_bh_state(bh);
 		jbd2_log_start_commit(journal, tid);
@@ -279,6 +281,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		bhs[*batch_count] = bh;
 		__buffer_relink_io(jh);
 		jbd_unlock_bh_state(bh);
+		transaction->t_chp_stats.cs_written++;
 		(*batch_count)++;
 		if (*batch_count == NR_BATCH) {
 			spin_unlock(&journal->j_list_lock);
@@ -322,6 +325,8 @@ int jbd2_log_do_checkpoint(journal_t *journal)
 	if (!journal->j_checkpoint_transactions)
 		goto out;
 	transaction = journal->j_checkpoint_transactions;
+	if (transaction->t_chp_stats.cs_chp_time == 0)
+		transaction->t_chp_stats.cs_chp_time = jiffies;
 	this_tid = transaction->t_tid;
 restart:
 	/*
@@ -346,7 +351,8 @@ restart:
 				retry = 1;
 				break;
 			}
-			retry = __process_buffer(journal, jh, bhs,&batch_count);
+			retry = __process_buffer(journal, jh, bhs, &batch_count,
+						 transaction);
 			if (!retry && lock_need_resched(&journal->j_list_lock)){
 				spin_unlock(&journal->j_list_lock);
 				retry = 1;
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 39b5cee3dd8a..8749a86f4175 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -20,6 +20,7 @@
 #include <linux/slab.h>
 #include <linux/mm.h>
 #include <linux/pagemap.h>
+#include <linux/jiffies.h>
 
 /*
  * Default IO end handler for temporary BJ_IO buffer_heads.
@@ -290,6 +291,7 @@ static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag,
  */
 void jbd2_journal_commit_transaction(journal_t *journal)
 {
+	struct transaction_stats_s stats;
 	transaction_t *commit_transaction;
 	struct journal_head *jh, *new_jh, *descriptor;
 	struct buffer_head **wbuf = journal->j_wbuf;
@@ -337,6 +339,11 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	spin_lock(&journal->j_state_lock);
 	commit_transaction->t_state = T_LOCKED;
 
+	stats.u.run.rs_wait = commit_transaction->t_max_wait;
+	stats.u.run.rs_locked = jiffies;
+	stats.u.run.rs_running = jbd2_time_diff(commit_transaction->t_start,
+						stats.u.run.rs_locked);
+
 	spin_lock(&commit_transaction->t_handle_lock);
 	while (commit_transaction->t_updates) {
 		DEFINE_WAIT(wait);
@@ -407,6 +414,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 */
 	jbd2_journal_switch_revoke_table(journal);
 
+	stats.u.run.rs_flushing = jiffies;
+	stats.u.run.rs_locked = jbd2_time_diff(stats.u.run.rs_locked,
+					       stats.u.run.rs_flushing);
+
 	commit_transaction->t_state = T_FLUSH;
 	journal->j_committing_transaction = commit_transaction;
 	journal->j_running_transaction = NULL;
@@ -498,6 +509,12 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 */
 	commit_transaction->t_state = T_COMMIT;
 
+	stats.u.run.rs_logging = jiffies;
+	stats.u.run.rs_flushing = jbd2_time_diff(stats.u.run.rs_flushing,
+						 stats.u.run.rs_logging);
+	stats.u.run.rs_blocks = commit_transaction->t_outstanding_credits;
+	stats.u.run.rs_blocks_logged = 0;
+
 	descriptor = NULL;
 	bufs = 0;
 	while (commit_transaction->t_buffers) {
@@ -646,6 +663,7 @@ start_journal_io:
 				submit_bh(WRITE, bh);
 			}
 			cond_resched();
+			stats.u.run.rs_blocks_logged += bufs;
 
 			/* Force a new descriptor to be generated next
                            time round the loop. */
@@ -816,6 +834,7 @@ restart_loop:
 		cp_transaction = jh->b_cp_transaction;
 		if (cp_transaction) {
 			JBUFFER_TRACE(jh, "remove from old cp transaction");
+			cp_transaction->t_chp_stats.cs_dropped++;
 			__jbd2_journal_remove_checkpoint(jh);
 		}
 
@@ -890,6 +909,36 @@ restart_loop:
 
 	J_ASSERT(commit_transaction->t_state == T_COMMIT);
 
+	commit_transaction->t_start = jiffies;
+	stats.u.run.rs_logging = jbd2_time_diff(stats.u.run.rs_logging,
+						commit_transaction->t_start);
+
+	/*
+	 * File the transaction for history
+	 */
+	stats.ts_type = JBD2_STATS_RUN;
+	stats.ts_tid = commit_transaction->t_tid;
+	stats.u.run.rs_handle_count = commit_transaction->t_handle_count;
+	spin_lock(&journal->j_history_lock);
+	memcpy(journal->j_history + journal->j_history_cur, &stats,
+			sizeof(stats));
+	if (++journal->j_history_cur == journal->j_history_max)
+		journal->j_history_cur = 0;
+
+	/*
+	 * Calculate overall stats
+	 */
+	journal->j_stats.ts_tid++;
+	journal->j_stats.u.run.rs_wait += stats.u.run.rs_wait;
+	journal->j_stats.u.run.rs_running += stats.u.run.rs_running;
+	journal->j_stats.u.run.rs_locked += stats.u.run.rs_locked;
+	journal->j_stats.u.run.rs_flushing += stats.u.run.rs_flushing;
+	journal->j_stats.u.run.rs_logging += stats.u.run.rs_logging;
+	journal->j_stats.u.run.rs_handle_count += stats.u.run.rs_handle_count;
+	journal->j_stats.u.run.rs_blocks += stats.u.run.rs_blocks;
+	journal->j_stats.u.run.rs_blocks_logged += stats.u.run.rs_blocks_logged;
+	spin_unlock(&journal->j_history_lock);
+
 	commit_transaction->t_state = T_FINISHED;
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);
 	journal->j_commit_sequence = commit_transaction->t_tid;
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 6ddc5531587c..3667c91bc786 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -36,6 +36,7 @@
 #include <linux/poison.h>
 #include <linux/proc_fs.h>
 #include <linux/debugfs.h>
+#include <linux/seq_file.h>
 
 #include <asm/uaccess.h>
 #include <asm/page.h>
@@ -640,6 +641,312 @@ struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal)
 	return jbd2_journal_add_journal_head(bh);
 }
 
+struct jbd2_stats_proc_session {
+	journal_t *journal;
+	struct transaction_stats_s *stats;
+	int start;
+	int max;
+};
+
+static void *jbd2_history_skip_empty(struct jbd2_stats_proc_session *s,
+					struct transaction_stats_s *ts,
+					int first)
+{
+	if (ts == s->stats + s->max)
+		ts = s->stats;
+	if (!first && ts == s->stats + s->start)
+		return NULL;
+	while (ts->ts_type == 0) {
+		ts++;
+		if (ts == s->stats + s->max)
+			ts = s->stats;
+		if (ts == s->stats + s->start)
+			return NULL;
+	}
+	return ts;
+
+}
+
+static void *jbd2_seq_history_start(struct seq_file *seq, loff_t *pos)
+{
+	struct jbd2_stats_proc_session *s = seq->private;
+	struct transaction_stats_s *ts;
+	int l = *pos;
+
+	if (l == 0)
+		return SEQ_START_TOKEN;
+	ts = jbd2_history_skip_empty(s, s->stats + s->start, 1);
+	if (!ts)
+		return NULL;
+	l--;
+	while (l) {
+		ts = jbd2_history_skip_empty(s, ++ts, 0);
+		if (!ts)
+			break;
+		l--;
+	}
+	return ts;
+}
+
+static void *jbd2_seq_history_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct jbd2_stats_proc_session *s = seq->private;
+	struct transaction_stats_s *ts = v;
+
+	++*pos;
+	if (v == SEQ_START_TOKEN)
+		return jbd2_history_skip_empty(s, s->stats + s->start, 1);
+	else
+		return jbd2_history_skip_empty(s, ++ts, 0);
+}
+
+static int jbd2_seq_history_show(struct seq_file *seq, void *v)
+{
+	struct transaction_stats_s *ts = v;
+	if (v == SEQ_START_TOKEN) {
+		seq_printf(seq, "%-4s %-5s %-5s %-5s %-5s %-5s %-5s %-6s %-5s "
+				"%-5s %-5s %-5s %-5s %-5s\n", "R/C", "tid",
+				"wait", "run", "lock", "flush", "log", "hndls",
+				"block", "inlog", "ctime", "write", "drop",
+				"close");
+		return 0;
+	}
+	if (ts->ts_type == JBD2_STATS_RUN)
+		seq_printf(seq, "%-4s %-5lu %-5u %-5u %-5u %-5u %-5u "
+				"%-6lu %-5lu %-5lu\n", "R", ts->ts_tid,
+				jiffies_to_msecs(ts->u.run.rs_wait),
+				jiffies_to_msecs(ts->u.run.rs_running),
+				jiffies_to_msecs(ts->u.run.rs_locked),
+				jiffies_to_msecs(ts->u.run.rs_flushing),
+				jiffies_to_msecs(ts->u.run.rs_logging),
+				ts->u.run.rs_handle_count,
+				ts->u.run.rs_blocks,
+				ts->u.run.rs_blocks_logged);
+	else if (ts->ts_type == JBD2_STATS_CHECKPOINT)
+		seq_printf(seq, "%-4s %-5lu %48s %-5u %-5lu %-5lu %-5lu\n",
+				"C", ts->ts_tid, " ",
+				jiffies_to_msecs(ts->u.chp.cs_chp_time),
+				ts->u.chp.cs_written, ts->u.chp.cs_dropped,
+				ts->u.chp.cs_forced_to_close);
+	else
+		J_ASSERT(0);
+	return 0;
+}
+
+static void jbd2_seq_history_stop(struct seq_file *seq, void *v)
+{
+}
+
+static struct seq_operations jbd2_seq_history_ops = {
+	.start  = jbd2_seq_history_start,
+	.next   = jbd2_seq_history_next,
+	.stop   = jbd2_seq_history_stop,
+	.show   = jbd2_seq_history_show,
+};
+
+static int jbd2_seq_history_open(struct inode *inode, struct file *file)
+{
+	journal_t *journal = PDE(inode)->data;
+	struct jbd2_stats_proc_session *s;
+	int rc, size;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s == NULL)
+		return -ENOMEM;
+	size = sizeof(struct transaction_stats_s) * journal->j_history_max;
+	s->stats = kmalloc(size, GFP_KERNEL);
+	if (s->stats == NULL) {
+		kfree(s);
+		return -ENOMEM;
+	}
+	spin_lock(&journal->j_history_lock);
+	memcpy(s->stats, journal->j_history, size);
+	s->max = journal->j_history_max;
+	s->start = journal->j_history_cur % s->max;
+	spin_unlock(&journal->j_history_lock);
+
+	rc = seq_open(file, &jbd2_seq_history_ops);
+	if (rc == 0) {
+		struct seq_file *m = file->private_data;
+		m->private = s;
+	} else {
+		kfree(s->stats);
+		kfree(s);
+	}
+	return rc;
+
+}
+
+static int jbd2_seq_history_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct jbd2_stats_proc_session *s = seq->private;
+
+	kfree(s->stats);
+	kfree(s);
+	return seq_release(inode, file);
+}
+
+static struct file_operations jbd2_seq_history_fops = {
+	.owner		= THIS_MODULE,
+	.open           = jbd2_seq_history_open,
+	.read           = seq_read,
+	.llseek         = seq_lseek,
+	.release        = jbd2_seq_history_release,
+};
+
+static void *jbd2_seq_info_start(struct seq_file *seq, loff_t *pos)
+{
+	return *pos ? NULL : SEQ_START_TOKEN;
+}
+
+static void *jbd2_seq_info_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	return NULL;
+}
+
+static int jbd2_seq_info_show(struct seq_file *seq, void *v)
+{
+	struct jbd2_stats_proc_session *s = seq->private;
+
+	if (v != SEQ_START_TOKEN)
+		return 0;
+	seq_printf(seq, "%lu transaction, each upto %u blocks\n",
+			s->stats->ts_tid,
+			s->journal->j_max_transaction_buffers);
+	if (s->stats->ts_tid == 0)
+		return 0;
+	seq_printf(seq, "average: \n  %ums waiting for transaction\n",
+	    jiffies_to_msecs(s->stats->u.run.rs_wait / s->stats->ts_tid));
+	seq_printf(seq, "  %ums running transaction\n",
+	    jiffies_to_msecs(s->stats->u.run.rs_running / s->stats->ts_tid));
+	seq_printf(seq, "  %ums transaction was being locked\n",
+	    jiffies_to_msecs(s->stats->u.run.rs_locked / s->stats->ts_tid));
+	seq_printf(seq, "  %ums flushing data (in ordered mode)\n",
+	    jiffies_to_msecs(s->stats->u.run.rs_flushing / s->stats->ts_tid));
+	seq_printf(seq, "  %ums logging transaction\n",
+	    jiffies_to_msecs(s->stats->u.run.rs_logging / s->stats->ts_tid));
+	seq_printf(seq, "  %lu handles per transaction\n",
+	    s->stats->u.run.rs_handle_count / s->stats->ts_tid);
+	seq_printf(seq, "  %lu blocks per transaction\n",
+	    s->stats->u.run.rs_blocks / s->stats->ts_tid);
+	seq_printf(seq, "  %lu logged blocks per transaction\n",
+	    s->stats->u.run.rs_blocks_logged / s->stats->ts_tid);
+	return 0;
+}
+
+static void jbd2_seq_info_stop(struct seq_file *seq, void *v)
+{
+}
+
+static struct seq_operations jbd2_seq_info_ops = {
+	.start  = jbd2_seq_info_start,
+	.next   = jbd2_seq_info_next,
+	.stop   = jbd2_seq_info_stop,
+	.show   = jbd2_seq_info_show,
+};
+
+static int jbd2_seq_info_open(struct inode *inode, struct file *file)
+{
+	journal_t *journal = PDE(inode)->data;
+	struct jbd2_stats_proc_session *s;
+	int rc, size;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s == NULL)
+		return -ENOMEM;
+	size = sizeof(struct transaction_stats_s);
+	s->stats = kmalloc(size, GFP_KERNEL);
+	if (s->stats == NULL) {
+		kfree(s);
+		return -ENOMEM;
+	}
+	spin_lock(&journal->j_history_lock);
+	memcpy(s->stats, &journal->j_stats, size);
+	s->journal = journal;
+	spin_unlock(&journal->j_history_lock);
+
+	rc = seq_open(file, &jbd2_seq_info_ops);
+	if (rc == 0) {
+		struct seq_file *m = file->private_data;
+		m->private = s;
+	} else {
+		kfree(s->stats);
+		kfree(s);
+	}
+	return rc;
+
+}
+
+static int jbd2_seq_info_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct jbd2_stats_proc_session *s = seq->private;
+	kfree(s->stats);
+	kfree(s);
+	return seq_release(inode, file);
+}
+
+static struct file_operations jbd2_seq_info_fops = {
+	.owner		= THIS_MODULE,
+	.open           = jbd2_seq_info_open,
+	.read           = seq_read,
+	.llseek         = seq_lseek,
+	.release        = jbd2_seq_info_release,
+};
+
+static struct proc_dir_entry *proc_jbd2_stats;
+
+static void jbd2_stats_proc_init(journal_t *journal)
+{
+	char name[BDEVNAME_SIZE];
+
+	snprintf(name, sizeof(name) - 1, "%s", bdevname(journal->j_dev, name));
+	journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats);
+	if (journal->j_proc_entry) {
+		struct proc_dir_entry *p;
+		p = create_proc_entry("history", S_IRUGO,
+				journal->j_proc_entry);
+		if (p) {
+			p->proc_fops = &jbd2_seq_history_fops;
+			p->data = journal;
+			p = create_proc_entry("info", S_IRUGO,
+						journal->j_proc_entry);
+			if (p) {
+				p->proc_fops = &jbd2_seq_info_fops;
+				p->data = journal;
+			}
+		}
+	}
+}
+
+static void jbd2_stats_proc_exit(journal_t *journal)
+{
+	char name[BDEVNAME_SIZE];
+
+	snprintf(name, sizeof(name) - 1, "%s", bdevname(journal->j_dev, name));
+	remove_proc_entry("info", journal->j_proc_entry);
+	remove_proc_entry("history", journal->j_proc_entry);
+	remove_proc_entry(name, proc_jbd2_stats);
+}
+
+static void journal_init_stats(journal_t *journal)
+{
+	int size;
+
+	if (!proc_jbd2_stats)
+		return;
+
+	journal->j_history_max = 100;
+	size = sizeof(struct transaction_stats_s) * journal->j_history_max;
+	journal->j_history = kzalloc(size, GFP_KERNEL);
+	if (!journal->j_history) {
+		journal->j_history_max = 0;
+		return;
+	}
+	spin_lock_init(&journal->j_history_lock);
+}
+
 /*
  * Management for journal control blocks: functions to create and
  * destroy journal_t structures, and to initialise and read existing
@@ -681,6 +988,9 @@ static journal_t * journal_init_common (void)
 		kfree(journal);
 		goto fail;
 	}
+
+	journal_init_stats(journal);
+
 	return journal;
 fail:
 	return NULL;
@@ -735,6 +1045,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
 	journal->j_fs_dev = fs_dev;
 	journal->j_blk_offset = start;
 	journal->j_maxlen = len;
+	jbd2_stats_proc_init(journal);
 
 	bh = __getblk(journal->j_dev, start, journal->j_blocksize);
 	J_ASSERT(bh != NULL);
@@ -773,6 +1084,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
 
 	journal->j_maxlen = inode->i_size >> inode->i_sb->s_blocksize_bits;
 	journal->j_blocksize = inode->i_sb->s_blocksize;
+	jbd2_stats_proc_init(journal);
 
 	/* journal descriptor can store up to n blocks -bzzz */
 	n = journal->j_blocksize / sizeof(journal_block_tag_t);
@@ -1153,6 +1465,8 @@ void jbd2_journal_destroy(journal_t *journal)
 		brelse(journal->j_sb_buffer);
 	}
 
+	if (journal->j_proc_entry)
+		jbd2_stats_proc_exit(journal);
 	if (journal->j_inode)
 		iput(journal->j_inode);
 	if (journal->j_revoke)
@@ -1900,6 +2214,28 @@ static void __exit jbd2_remove_debugfs_entry(void)
 
 #endif
 
+#ifdef CONFIG_PROC_FS
+
+#define JBD2_STATS_PROC_NAME "fs/jbd2"
+
+static void __init jbd2_create_jbd_stats_proc_entry(void)
+{
+	proc_jbd2_stats = proc_mkdir(JBD2_STATS_PROC_NAME, NULL);
+}
+
+static void __exit jbd2_remove_jbd_stats_proc_entry(void)
+{
+	if (proc_jbd2_stats)
+		remove_proc_entry(JBD2_STATS_PROC_NAME, NULL);
+}
+
+#else
+
+#define jbd2_create_jbd_stats_proc_entry() do {} while (0)
+#define jbd2_remove_jbd_stats_proc_entry() do {} while (0)
+
+#endif
+
 struct kmem_cache *jbd2_handle_cache;
 
 static int __init journal_init_handle_cache(void)
@@ -1955,6 +2291,7 @@ static int __init journal_init(void)
 	if (ret != 0)
 		jbd2_journal_destroy_caches();
 	jbd2_create_debugfs_entry();
+	jbd2_create_jbd_stats_proc_entry();
 	return ret;
 }
 
@@ -1966,6 +2303,7 @@ static void __exit journal_exit(void)
 		printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n);
 #endif
 	jbd2_remove_debugfs_entry();
+	jbd2_remove_jbd_stats_proc_entry();
 	jbd2_journal_destroy_caches();
 }
 
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index b1fcf2b3dca3..f30802aeefae 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -59,6 +59,8 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
 
 	J_ASSERT(journal->j_running_transaction == NULL);
 	journal->j_running_transaction = transaction;
+	transaction->t_max_wait = 0;
+	transaction->t_start = jiffies;
 
 	return transaction;
 }
@@ -85,6 +87,7 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 	int nblocks = handle->h_buffer_credits;
 	transaction_t *new_transaction = NULL;
 	int ret = 0;
+	unsigned long ts = jiffies;
 
 	if (nblocks > journal->j_max_transaction_buffers) {
 		printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n",
@@ -217,6 +220,12 @@ repeat_locked:
 	/* OK, account for the buffers that this operation expects to
 	 * use and add the handle to the running transaction. */
 
+	if (time_after(transaction->t_start, ts)) {
+		ts = jbd2_time_diff(ts, transaction->t_start);
+		if (ts > transaction->t_max_wait)
+			transaction->t_max_wait = ts;
+	}
+
 	handle->h_transaction = transaction;
 	transaction->t_outstanding_credits += nblocks;
 	transaction->t_updates++;
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index d861ffd49821..685640036e81 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -395,6 +395,16 @@ struct handle_s
 };
 
 
+/*
+ * Some stats for checkpoint phase
+ */
+struct transaction_chp_stats_s {
+	unsigned long		cs_chp_time;
+	unsigned long		cs_forced_to_close;
+	unsigned long		cs_written;
+	unsigned long		cs_dropped;
+};
+
 /* The transaction_t type is the guts of the journaling mechanism.  It
  * tracks a compound transaction through its various states:
  *
@@ -531,6 +541,21 @@ struct transaction_s
 	 */
 	spinlock_t		t_handle_lock;
 
+	/*
+	 * Longest time some handle had to wait for running transaction
+	 */
+	unsigned long		t_max_wait;
+
+	/*
+	 * When transaction started
+	 */
+	unsigned long		t_start;
+
+	/*
+	 * Checkpointing stats [j_checkpoint_sem]
+	 */
+	struct transaction_chp_stats_s t_chp_stats;
+
 	/*
 	 * Number of outstanding updates running on this transaction
 	 * [t_handle_lock]
@@ -562,6 +587,39 @@ struct transaction_s
 
 };
 
+struct transaction_run_stats_s {
+	unsigned long		rs_wait;
+	unsigned long		rs_running;
+	unsigned long		rs_locked;
+	unsigned long		rs_flushing;
+	unsigned long		rs_logging;
+
+	unsigned long		rs_handle_count;
+	unsigned long		rs_blocks;
+	unsigned long		rs_blocks_logged;
+};
+
+struct transaction_stats_s {
+	int 			ts_type;
+	unsigned long		ts_tid;
+	union {
+		struct transaction_run_stats_s run;
+		struct transaction_chp_stats_s chp;
+	} u;
+};
+
+#define JBD2_STATS_RUN		1
+#define JBD2_STATS_CHECKPOINT	2
+
+static inline unsigned long
+jbd2_time_diff(unsigned long start, unsigned long end)
+{
+	if (end >= start)
+		return end - start;
+
+	return end + (MAX_JIFFY_OFFSET - start);
+}
+
 /**
  * struct journal_s - The journal_s type is the concrete type associated with
  *     journal_t.
@@ -623,6 +681,12 @@ struct transaction_s
  * @j_wbufsize: maximum number of buffer_heads allowed in j_wbuf, the
  *	number that will fit in j_blocksize
  * @j_last_sync_writer: most recent pid which did a synchronous write
+ * @j_history: Buffer storing the transactions statistics history
+ * @j_history_max: Maximum number of transactions in the statistics history
+ * @j_history_cur: Current number of transactions in the statistics history
+ * @j_history_lock: Protect the transactions statistics history
+ * @j_proc_entry: procfs entry for the jbd statistics directory
+ * @j_stats: Overall statistics
  * @j_private: An opaque pointer to fs-private information.
  */
 
@@ -814,6 +878,19 @@ struct journal_s
 
 	pid_t			j_last_sync_writer;
 
+	/*
+	 * Journal statistics
+	 */
+	struct transaction_stats_s *j_history;
+	int			j_history_max;
+	int			j_history_cur;
+	/*
+	 * Protect the transactions statistics history
+	 */
+	spinlock_t		j_history_lock;
+	struct proc_dir_entry	*j_proc_entry;
+	struct transaction_stats_s j_stats;
+
 	/*
 	 * An opaque pointer to fs-private information.  ext3 puts its
 	 * superblock pointer here
-- 
cgit v1.2.3


From 818d276ceb83aa9fdebb5e0a53188290312de987 Mon Sep 17 00:00:00 2001
From: Girish Shilamkar <girish@clusterfs.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Add the journal checksum feature

The journal checksum feature adds two new flags i.e
JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT and JBD2_FEATURE_COMPAT_CHECKSUM.

JBD2_FEATURE_CHECKSUM flag indicates that the commit block contains the
checksum for the blocks described by the descriptor blocks.
Due to checksums, writing of the commit record no longer needs to be
synchronous. Now commit record can be sent to disk without waiting for
descriptor blocks to be written to disk. This behavior is controlled
using JBD2_FEATURE_ASYNC_COMMIT flag. Older kernels/e2fsck should not be
able to recover the journal with _ASYNC_COMMIT hence it is made
incompat.
The commit header has been extended to hold the checksum along with the
type of the checksum.

For recovery in pass scan checksums are verified to ensure the sanity
and completeness(in case of _ASYNC_COMMIT) of every transaction.

Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Girish Shilamkar <girish@clusterfs.com>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---
 Documentation/filesystems/ext4.txt |  10 ++
 fs/Kconfig                         |   1 +
 fs/ext4/super.c                    |  25 +++++
 fs/jbd2/commit.c                   | 198 ++++++++++++++++++++++++++++---------
 fs/jbd2/journal.c                  |  26 +++++
 fs/jbd2/recovery.c                 | 151 ++++++++++++++++++++++++++--
 include/linux/ext4_fs.h            |   3 +-
 include/linux/jbd2.h               |  36 ++++++-
 8 files changed, 388 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index 6a4adcae9f9a..4f329afe20ec 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -89,6 +89,16 @@ When mounting an ext4 filesystem, the following option are accepted:
 extents			ext4 will use extents to address file data.  The
 			file system will no longer be mountable by ext3.
 
+journal_checksum	Enable checksumming of the journal transactions.
+			This will allow the recovery code in e2fsck and the
+			kernel to detect corruption in the kernel.  It is a
+			compatible change and will be ignored by older kernels.
+
+journal_async_commit	Commit block can be written to disk without waiting
+			for descriptor blocks. If enabled older kernels cannot
+			mount the device. This will enable 'journal_checksum'
+			internally.
+
 journal=update		Update the ext4 file system's journal to the current
 			format.
 
diff --git a/fs/Kconfig b/fs/Kconfig
index 9656139d2e99..219ec06a8c7e 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -236,6 +236,7 @@ config JBD_DEBUG
 
 config JBD2
 	tristate
+	select CRC32
 	help
 	  This is a generic journaling layer for block devices that support
 	  both 32-bit and 64-bit block numbers.  It is currently used by
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index c7305443e100..f7479d30735e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -869,6 +869,7 @@ enum {
 	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
 	Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh,
 	Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
+	Opt_journal_checksum, Opt_journal_async_commit,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
@@ -908,6 +909,8 @@ static match_table_t tokens = {
 	{Opt_journal_update, "journal=update"},
 	{Opt_journal_inum, "journal=%u"},
 	{Opt_journal_dev, "journal_dev=%u"},
+	{Opt_journal_checksum, "journal_checksum"},
+	{Opt_journal_async_commit, "journal_async_commit"},
 	{Opt_abort, "abort"},
 	{Opt_data_journal, "data=journal"},
 	{Opt_data_ordered, "data=ordered"},
@@ -1095,6 +1098,13 @@ static int parse_options (char *options, struct super_block *sb,
 				return 0;
 			*journal_devnum = option;
 			break;
+		case Opt_journal_checksum:
+			set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM);
+			break;
+		case Opt_journal_async_commit:
+			set_opt(sbi->s_mount_opt, JOURNAL_ASYNC_COMMIT);
+			set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM);
+			break;
 		case Opt_noload:
 			set_opt (sbi->s_mount_opt, NOLOAD);
 			break;
@@ -2114,6 +2124,21 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		goto failed_mount4;
 	}
 
+	if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
+		jbd2_journal_set_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	} else if (test_opt(sb, JOURNAL_CHECKSUM)) {
+		jbd2_journal_set_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0);
+		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	} else {
+		jbd2_journal_clear_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	}
+
 	/* We have now updated the journal if required, so we can
 	 * validate the data journaling mode. */
 	switch (test_opt(sb, DATA_FLAGS)) {
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 8749a86f4175..da8d0eb3b7b9 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -21,6 +21,7 @@
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/jiffies.h>
+#include <linux/crc32.h>
 
 /*
  * Default IO end handler for temporary BJ_IO buffer_heads.
@@ -93,19 +94,23 @@ static int inverted_lock(journal_t *journal, struct buffer_head *bh)
 	return 1;
 }
 
-/* Done it all: now write the commit record.  We should have
+/*
+ * Done it all: now submit the commit record.  We should have
  * cleaned up our previous buffers by now, so if we are in abort
  * mode we can now just skip the rest of the journal write
  * entirely.
  *
  * Returns 1 if the journal needs to be aborted or 0 on success
  */
-static int journal_write_commit_record(journal_t *journal,
-					transaction_t *commit_transaction)
+static int journal_submit_commit_record(journal_t *journal,
+					transaction_t *commit_transaction,
+					struct buffer_head **cbh,
+					__u32 crc32_sum)
 {
 	struct journal_head *descriptor;
+	struct commit_header *tmp;
 	struct buffer_head *bh;
-	int i, ret;
+	int ret;
 	int barrier_done = 0;
 
 	if (is_journal_aborted(journal))
@@ -117,21 +122,33 @@ static int journal_write_commit_record(journal_t *journal,
 
 	bh = jh2bh(descriptor);
 
-	/* AKPM: buglet - add `i' to tmp! */
-	for (i = 0; i < bh->b_size; i += 512) {
-		journal_header_t *tmp = (journal_header_t*)bh->b_data;
-		tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER);
-		tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK);
-		tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid);
+	tmp = (struct commit_header *)bh->b_data;
+	tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER);
+	tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK);
+	tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid);
+
+	if (JBD2_HAS_COMPAT_FEATURE(journal,
+				    JBD2_FEATURE_COMPAT_CHECKSUM)) {
+		tmp->h_chksum_type 	= JBD2_CRC32_CHKSUM;
+		tmp->h_chksum_size 	= JBD2_CRC32_CHKSUM_SIZE;
+		tmp->h_chksum[0] 	= cpu_to_be32(crc32_sum);
 	}
 
-	JBUFFER_TRACE(descriptor, "write commit block");
+	JBUFFER_TRACE(descriptor, "submit commit block");
+	lock_buffer(bh);
+
 	set_buffer_dirty(bh);
-	if (journal->j_flags & JBD2_BARRIER) {
+	set_buffer_uptodate(bh);
+	bh->b_end_io = journal_end_buffer_io_sync;
+
+	if (journal->j_flags & JBD2_BARRIER &&
+		!JBD2_HAS_COMPAT_FEATURE(journal,
+					 JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
 		set_buffer_ordered(bh);
 		barrier_done = 1;
 	}
-	ret = sync_dirty_buffer(bh);
+	ret = submit_bh(WRITE, bh);
+
 	/* is it possible for another commit to fail at roughly
 	 * the same time as this one?  If so, we don't want to
 	 * trust the barrier flag in the super, but instead want
@@ -152,14 +169,72 @@ static int journal_write_commit_record(journal_t *journal,
 		clear_buffer_ordered(bh);
 		set_buffer_uptodate(bh);
 		set_buffer_dirty(bh);
-		ret = sync_dirty_buffer(bh);
+		ret = submit_bh(WRITE, bh);
 	}
-	put_bh(bh);		/* One for getblk() */
-	jbd2_journal_put_journal_head(descriptor);
+	*cbh = bh;
+	return ret;
+}
+
+/*
+ * This function along with journal_submit_commit_record
+ * allows to write the commit record asynchronously.
+ */
+static int journal_wait_on_commit_record(struct buffer_head *bh)
+{
+	int ret = 0;
+
+	clear_buffer_dirty(bh);
+	wait_on_buffer(bh);
 
-	return (ret == -EIO);
+	if (unlikely(!buffer_uptodate(bh)))
+		ret = -EIO;
+	put_bh(bh);            /* One for getblk() */
+	jbd2_journal_put_journal_head(bh2jh(bh));
+
+	return ret;
 }
 
+/*
+ * Wait for all submitted IO to complete.
+ */
+static int journal_wait_on_locked_list(journal_t *journal,
+				       transaction_t *commit_transaction)
+{
+	int ret = 0;
+	struct journal_head *jh;
+
+	while (commit_transaction->t_locked_list) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_locked_list->b_tprev;
+		bh = jh2bh(jh);
+		get_bh(bh);
+		if (buffer_locked(bh)) {
+			spin_unlock(&journal->j_list_lock);
+			wait_on_buffer(bh);
+			if (unlikely(!buffer_uptodate(bh)))
+				ret = -EIO;
+			spin_lock(&journal->j_list_lock);
+		}
+		if (!inverted_lock(journal, bh)) {
+			put_bh(bh);
+			spin_lock(&journal->j_list_lock);
+			continue;
+		}
+		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
+			__jbd2_journal_unfile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+			jbd2_journal_remove_journal_head(bh);
+			put_bh(bh);
+		} else {
+			jbd_unlock_bh_state(bh);
+		}
+		put_bh(bh);
+		cond_resched_lock(&journal->j_list_lock);
+	}
+	return ret;
+  }
+
 static void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
 {
 	int i;
@@ -275,7 +350,21 @@ write_out_data:
 	journal_do_submit_data(wbuf, bufs);
 }
 
-static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag,
+static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh)
+{
+	struct page *page = bh->b_page;
+	char *addr;
+	__u32 checksum;
+
+	addr = kmap_atomic(page, KM_USER0);
+	checksum = crc32_be(crc32_sum,
+		(void *)(addr + offset_in_page(bh->b_data)), bh->b_size);
+	kunmap_atomic(addr, KM_USER0);
+
+	return checksum;
+}
+
+static void write_tag_block(int tag_bytes, journal_block_tag_t *tag,
 				   unsigned long long block)
 {
 	tag->t_blocknr = cpu_to_be32(block & (u32)~0);
@@ -307,6 +396,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	int tag_flag;
 	int i;
 	int tag_bytes = journal_tag_bytes(journal);
+	struct buffer_head *cbh = NULL; /* For transactional checksums */
+	__u32 crc32_sum = ~0;
 
 	/*
 	 * First job: lock down the current transaction and wait for
@@ -451,38 +542,15 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	journal_submit_data_buffers(journal, commit_transaction);
 
 	/*
-	 * Wait for all previously submitted IO to complete.
+	 * Wait for all previously submitted IO to complete if commit
+	 * record is to be written synchronously.
 	 */
 	spin_lock(&journal->j_list_lock);
-	while (commit_transaction->t_locked_list) {
-		struct buffer_head *bh;
+	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT))
+		err = journal_wait_on_locked_list(journal,
+						commit_transaction);
 
-		jh = commit_transaction->t_locked_list->b_tprev;
-		bh = jh2bh(jh);
-		get_bh(bh);
-		if (buffer_locked(bh)) {
-			spin_unlock(&journal->j_list_lock);
-			wait_on_buffer(bh);
-			if (unlikely(!buffer_uptodate(bh)))
-				err = -EIO;
-			spin_lock(&journal->j_list_lock);
-		}
-		if (!inverted_lock(journal, bh)) {
-			put_bh(bh);
-			spin_lock(&journal->j_list_lock);
-			continue;
-		}
-		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
-			__jbd2_journal_unfile_buffer(jh);
-			jbd_unlock_bh_state(bh);
-			jbd2_journal_remove_journal_head(bh);
-			put_bh(bh);
-		} else {
-			jbd_unlock_bh_state(bh);
-		}
-		put_bh(bh);
-		cond_resched_lock(&journal->j_list_lock);
-	}
 	spin_unlock(&journal->j_list_lock);
 
 	if (err)
@@ -656,6 +724,15 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 start_journal_io:
 			for (i = 0; i < bufs; i++) {
 				struct buffer_head *bh = wbuf[i];
+				/*
+				 * Compute checksum.
+				 */
+				if (JBD2_HAS_COMPAT_FEATURE(journal,
+					JBD2_FEATURE_COMPAT_CHECKSUM)) {
+					crc32_sum =
+					    jbd2_checksum_data(crc32_sum, bh);
+				}
+
 				lock_buffer(bh);
 				clear_buffer_dirty(bh);
 				set_buffer_uptodate(bh);
@@ -672,6 +749,23 @@ start_journal_io:
 		}
 	}
 
+	/* Done it all: now write the commit record asynchronously. */
+
+	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
+		err = journal_submit_commit_record(journal, commit_transaction,
+						 &cbh, crc32_sum);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+
+		spin_lock(&journal->j_list_lock);
+		err = journal_wait_on_locked_list(journal,
+						commit_transaction);
+		spin_unlock(&journal->j_list_lock);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+	}
+
 	/* Lo and behold: we have just managed to send a transaction to
            the log.  Before we can commit it, wait for the IO so far to
            complete.  Control buffers being written are on the
@@ -771,8 +865,14 @@ wait_for_iobuf:
 
 	jbd_debug(3, "JBD: commit phase 6\n");
 
-	if (journal_write_commit_record(journal, commit_transaction))
-		err = -EIO;
+	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
+		err = journal_submit_commit_record(journal, commit_transaction,
+						&cbh, crc32_sum);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+	}
+	err = journal_wait_on_commit_record(cbh);
 
 	if (err)
 		jbd2_journal_abort(journal, err);
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 3667c91bc786..59ba2494dcaf 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1578,6 +1578,32 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 	return 1;
 }
 
+/*
+ * jbd2_journal_clear_features () - Clear a given journal feature in the
+ * 				    superblock
+ * @journal: Journal to act on.
+ * @compat: bitmask of compatible features
+ * @ro: bitmask of features that force read-only mount
+ * @incompat: bitmask of incompatible features
+ *
+ * Clear a given journal feature as present on the
+ * superblock.
+ */
+void jbd2_journal_clear_features(journal_t *journal, unsigned long compat,
+				unsigned long ro, unsigned long incompat)
+{
+	journal_superblock_t *sb;
+
+	jbd_debug(1, "Clear features 0x%lx/0x%lx/0x%lx\n",
+		  compat, ro, incompat);
+
+	sb = journal->j_superblock;
+
+	sb->s_feature_compat    &= ~cpu_to_be32(compat);
+	sb->s_feature_ro_compat &= ~cpu_to_be32(ro);
+	sb->s_feature_incompat  &= ~cpu_to_be32(incompat);
+}
+EXPORT_SYMBOL(jbd2_journal_clear_features);
 
 /**
  * int jbd2_journal_update_format () - Update on-disk journal structure.
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index d0ce627539ef..921680663fa2 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -21,6 +21,7 @@
 #include <linux/jbd2.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
+#include <linux/crc32.h>
 #endif
 
 /*
@@ -316,6 +317,37 @@ static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag
 	return block;
 }
 
+/*
+ * calc_chksums calculates the checksums for the blocks described in the
+ * descriptor block.
+ */
+static int calc_chksums(journal_t *journal, struct buffer_head *bh,
+			unsigned long *next_log_block, __u32 *crc32_sum)
+{
+	int i, num_blks, err;
+	unsigned long io_block;
+	struct buffer_head *obh;
+
+	num_blks = count_tags(journal, bh);
+	/* Calculate checksum of the descriptor block. */
+	*crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size);
+
+	for (i = 0; i < num_blks; i++) {
+		io_block = (*next_log_block)++;
+		wrap(journal, *next_log_block);
+		err = jread(&obh, journal, io_block);
+		if (err) {
+			printk(KERN_ERR "JBD: IO error %d recovering block "
+				"%lu in log\n", err, io_block);
+			return 1;
+		} else {
+			*crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data,
+				     obh->b_size);
+		}
+	}
+	return 0;
+}
+
 static int do_one_pass(journal_t *journal,
 			struct recovery_info *info, enum passtype pass)
 {
@@ -328,6 +360,7 @@ static int do_one_pass(journal_t *journal,
 	unsigned int		sequence;
 	int			blocktype;
 	int			tag_bytes = journal_tag_bytes(journal);
+	__u32			crc32_sum = ~0; /* Transactional Checksums */
 
 	/* Precompute the maximum metadata descriptors in a descriptor block */
 	int			MAX_BLOCKS_PER_DESC;
@@ -419,12 +452,26 @@ static int do_one_pass(journal_t *journal,
 		switch(blocktype) {
 		case JBD2_DESCRIPTOR_BLOCK:
 			/* If it is a valid descriptor block, replay it
-			 * in pass REPLAY; otherwise, just skip over the
-			 * blocks it describes. */
+			 * in pass REPLAY; if journal_checksums enabled, then
+			 * calculate checksums in PASS_SCAN, otherwise,
+			 * just skip over the blocks it describes. */
 			if (pass != PASS_REPLAY) {
+				if (pass == PASS_SCAN &&
+				    JBD2_HAS_COMPAT_FEATURE(journal,
+					    JBD2_FEATURE_COMPAT_CHECKSUM) &&
+				    !info->end_transaction) {
+					if (calc_chksums(journal, bh,
+							&next_log_block,
+							&crc32_sum)) {
+						put_bh(bh);
+						break;
+					}
+					put_bh(bh);
+					continue;
+				}
 				next_log_block += count_tags(journal, bh);
 				wrap(journal, next_log_block);
-				brelse(bh);
+				put_bh(bh);
 				continue;
 			}
 
@@ -516,9 +563,96 @@ static int do_one_pass(journal_t *journal,
 			continue;
 
 		case JBD2_COMMIT_BLOCK:
-			/* Found an expected commit block: not much to
-			 * do other than move on to the next sequence
+			/*     How to differentiate between interrupted commit
+			 *               and journal corruption ?
+			 *
+			 * {nth transaction}
+			 *        Checksum Verification Failed
+			 *			 |
+			 *		 ____________________
+			 *		|		     |
+			 * 	async_commit             sync_commit
+			 *     		|                    |
+			 *		| GO TO NEXT    "Journal Corruption"
+			 *		| TRANSACTION
+			 *		|
+			 * {(n+1)th transanction}
+			 *		|
+			 * 	 _______|______________
+			 * 	|	 	      |
+			 * Commit block found	Commit block not found
+			 *      |		      |
+			 * "Journal Corruption"       |
+			 *		 _____________|_________
+			 *     		|	           	|
+			 *	nth trans corrupt	OR   nth trans
+			 *	and (n+1)th interrupted     interrupted
+			 *	before commit block
+			 *      could reach the disk.
+			 *	(Cannot find the difference in above
+			 *	 mentioned conditions. Hence assume
+			 *	 "Interrupted Commit".)
+			 */
+
+			/* Found an expected commit block: if checksums
+			 * are present verify them in PASS_SCAN; else not
+			 * much to do other than move on to the next sequence
 			 * number. */
+			if (pass == PASS_SCAN &&
+			    JBD2_HAS_COMPAT_FEATURE(journal,
+				    JBD2_FEATURE_COMPAT_CHECKSUM)) {
+				int chksum_err, chksum_seen;
+				struct commit_header *cbh =
+					(struct commit_header *)bh->b_data;
+				unsigned found_chksum =
+					be32_to_cpu(cbh->h_chksum[0]);
+
+				chksum_err = chksum_seen = 0;
+
+				if (info->end_transaction) {
+					printk(KERN_ERR "JBD: Transaction %u "
+						"found to be corrupt.\n",
+						next_commit_ID - 1);
+					brelse(bh);
+					break;
+				}
+
+				if (crc32_sum == found_chksum &&
+				    cbh->h_chksum_type == JBD2_CRC32_CHKSUM &&
+				    cbh->h_chksum_size ==
+						JBD2_CRC32_CHKSUM_SIZE)
+				       chksum_seen = 1;
+				else if (!(cbh->h_chksum_type == 0 &&
+					     cbh->h_chksum_size == 0 &&
+					     found_chksum == 0 &&
+					     !chksum_seen))
+				/*
+				 * If fs is mounted using an old kernel and then
+				 * kernel with journal_chksum is used then we
+				 * get a situation where the journal flag has
+				 * checksum flag set but checksums are not
+				 * present i.e chksum = 0, in the individual
+				 * commit blocks.
+				 * Hence to avoid checksum failures, in this
+				 * situation, this extra check is added.
+				 */
+						chksum_err = 1;
+
+				if (chksum_err) {
+					info->end_transaction = next_commit_ID;
+
+					if (!JBD2_HAS_COMPAT_FEATURE(journal,
+					   JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)){
+						printk(KERN_ERR
+						       "JBD: Transaction %u "
+						       "found to be corrupt.\n",
+						       next_commit_ID);
+						brelse(bh);
+						break;
+					}
+				}
+				crc32_sum = ~0;
+			}
 			brelse(bh);
 			next_commit_ID++;
 			continue;
@@ -554,9 +688,10 @@ static int do_one_pass(journal_t *journal,
 	 * transaction marks the end of the valid log.
 	 */
 
-	if (pass == PASS_SCAN)
-		info->end_transaction = next_commit_ID;
-	else {
+	if (pass == PASS_SCAN) {
+		if (!info->end_transaction)
+			info->end_transaction = next_commit_ID;
+	} else {
 		/* It's really bad news if different passes end up at
 		 * different places (but possible due to IO errors). */
 		if (info->end_transaction != next_commit_ID) {
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 300cc5a5adb9..cd406dba0e64 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -467,7 +467,8 @@ do {									       \
 #define EXT4_MOUNT_USRQUOTA		0x100000 /* "old" user quota */
 #define EXT4_MOUNT_GRPQUOTA		0x200000 /* "old" group quota */
 #define EXT4_MOUNT_EXTENTS		0x400000 /* Extents support */
-
+#define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
+#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
 #define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 685640036e81..98a2bc5d3e3f 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -149,6 +149,28 @@ typedef struct journal_header_s
 	__be32		h_sequence;
 } journal_header_t;
 
+/*
+ * Checksum types.
+ */
+#define JBD2_CRC32_CHKSUM   1
+#define JBD2_MD5_CHKSUM     2
+#define JBD2_SHA1_CHKSUM    3
+
+#define JBD2_CRC32_CHKSUM_SIZE 4
+
+#define JBD2_CHECKSUM_BYTES (32 / sizeof(u32))
+/*
+ * Commit block header for storing transactional checksums:
+ */
+struct commit_header {
+	__be32		h_magic;
+	__be32          h_blocktype;
+	__be32          h_sequence;
+	unsigned char   h_chksum_type;
+	unsigned char   h_chksum_size;
+	unsigned char 	h_padding[2];
+	__be32 		h_chksum[JBD2_CHECKSUM_BYTES];
+};
 
 /*
  * The block tag: used to describe a single buffer in the journal.
@@ -242,14 +264,18 @@ typedef struct journal_superblock_s
 	((j)->j_format_version >= 2 &&					\
 	 ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask))))
 
-#define JBD2_FEATURE_INCOMPAT_REVOKE	0x00000001
-#define JBD2_FEATURE_INCOMPAT_64BIT	0x00000002
+#define JBD2_FEATURE_COMPAT_CHECKSUM	0x00000001
+
+#define JBD2_FEATURE_INCOMPAT_REVOKE		0x00000001
+#define JBD2_FEATURE_INCOMPAT_64BIT		0x00000002
+#define JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT	0x00000004
 
 /* Features known to this kernel version: */
-#define JBD2_KNOWN_COMPAT_FEATURES	0
+#define JBD2_KNOWN_COMPAT_FEATURES	JBD2_FEATURE_COMPAT_CHECKSUM
 #define JBD2_KNOWN_ROCOMPAT_FEATURES	0
 #define JBD2_KNOWN_INCOMPAT_FEATURES	(JBD2_FEATURE_INCOMPAT_REVOKE | \
-					 JBD2_FEATURE_INCOMPAT_64BIT)
+					JBD2_FEATURE_INCOMPAT_64BIT | \
+					JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)
 
 #ifdef __KERNEL__
 
@@ -997,6 +1023,8 @@ extern int	   jbd2_journal_check_available_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   jbd2_journal_set_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
+extern void	   jbd2_journal_clear_features
+		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   jbd2_journal_create     (journal_t *);
 extern int	   jbd2_journal_load       (journal_t *journal);
 extern void	   jbd2_journal_destroy    (journal_t *);
-- 
cgit v1.2.3


From 7a224228ed79d587ece2304869000aad1b8e97dd Mon Sep 17 00:00:00 2001
From: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: vfs: Add 64 bit i_version support

The i_version field of the inode is changed to be a 64-bit counter that
is set on every inode creation and that is incremented every time the
inode data is modified (similarly to the "ctime" time-stamp).
The aim is to fulfill a NFSv4 requirement for rfc3530.
This first part concerns the vfs, it converts the 32-bit i_version in
the generic inode to a 64-bit, a flag is added in the super block in
order to check if the feature is enabled and the i_version is
incremented in the vfs.

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
Signed-off-by: Kalpak Shah <kalpak@clusterfs.com>
---
 fs/afs/dir.c       |  9 +++++----
 fs/afs/inode.c     |  3 ++-
 fs/inode.c         | 22 ++++++++++++++++++++++
 include/linux/fs.h |  5 ++++-
 4 files changed, 33 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index 33fe39ad4e03..0cc3597c1197 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -546,11 +546,11 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry,
 	dentry->d_op = &afs_fs_dentry_operations;
 
 	d_add(dentry, inode);
-	_leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%lu }",
+	_leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%llu }",
 	       fid.vnode,
 	       fid.unique,
 	       dentry->d_inode->i_ino,
-	       dentry->d_inode->i_version);
+	       (unsigned long long)dentry->d_inode->i_version);
 
 	return NULL;
 }
@@ -630,9 +630,10 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
 		 * been deleted and replaced, and the original vnode ID has
 		 * been reused */
 		if (fid.unique != vnode->fid.unique) {
-			_debug("%s: file deleted (uq %u -> %u I:%lu)",
+			_debug("%s: file deleted (uq %u -> %u I:%llu)",
 			       dentry->d_name.name, fid.unique,
-			       vnode->fid.unique, dentry->d_inode->i_version);
+			       vnode->fid.unique,
+			       (unsigned long long)dentry->d_inode->i_version);
 			spin_lock(&vnode->lock);
 			set_bit(AFS_VNODE_DELETED, &vnode->flags);
 			spin_unlock(&vnode->lock);
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index d196840127c6..84750c8e9f95 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -301,7 +301,8 @@ int afs_getattr(struct vfsmount *mnt, struct dentry *dentry,
 
 	inode = dentry->d_inode;
 
-	_enter("{ ino=%lu v=%lu }", inode->i_ino, inode->i_version);
+	_enter("{ ino=%lu v=%llu }", inode->i_ino,
+		(unsigned long long)inode->i_version);
 
 	generic_fillattr(inode, stat);
 	return 0;
diff --git a/fs/inode.c b/fs/inode.c
index ed35383d0b6c..b48324a94c2b 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1242,6 +1242,23 @@ void touch_atime(struct vfsmount *mnt, struct dentry *dentry)
 }
 EXPORT_SYMBOL(touch_atime);
 
+/**
+ *     inode_inc_iversion      -       increments i_version
+ *     @inode: inode that need to be updated
+ *
+ *     Every time the inode is modified, the i_version field
+ *     will be incremented.
+ *     The filesystem has to be mounted with i_version flag
+ *
+ */
+
+void inode_inc_iversion(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	inode->i_version++;
+	spin_unlock(&inode->i_lock);
+}
+
 /**
  *	file_update_time	-	update mtime and ctime time
  *	@file: file accessed
@@ -1276,6 +1293,11 @@ void file_update_time(struct file *file)
 		sync_it = 1;
 	}
 
+	if (IS_I_VERSION(inode)) {
+		inode_inc_iversion(inode);
+		sync_it = 1;
+	}
+
 	if (sync_it)
 		mark_inode_dirty_sync(inode);
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 21398a5d688d..9608839552b3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -124,6 +124,7 @@ extern int dir_notify_enable;
 #define MS_SHARED	(1<<20)	/* change to shared */
 #define MS_RELATIME	(1<<21)	/* Update atime relative to mtime/ctime. */
 #define MS_KERNMOUNT	(1<<22) /* this is a kern_mount call */
+#define MS_I_VERSION	(1<<23) /* Update inode I_version field */
 #define MS_ACTIVE	(1<<30)
 #define MS_NOUSER	(1<<31)
 
@@ -173,6 +174,7 @@ extern int dir_notify_enable;
 					((inode)->i_flags & (S_SYNC|S_DIRSYNC)))
 #define IS_MANDLOCK(inode)	__IS_FLG(inode, MS_MANDLOCK)
 #define IS_NOATIME(inode)   __IS_FLG(inode, MS_RDONLY|MS_NOATIME)
+#define IS_I_VERSION(inode)   __IS_FLG(inode, MS_I_VERSION)
 
 #define IS_NOQUOTA(inode)	((inode)->i_flags & S_NOQUOTA)
 #define IS_APPEND(inode)	((inode)->i_flags & S_APPEND)
@@ -599,7 +601,7 @@ struct inode {
 	uid_t			i_uid;
 	gid_t			i_gid;
 	dev_t			i_rdev;
-	unsigned long		i_version;
+	u64			i_version;
 	loff_t			i_size;
 #ifdef __NEED_I_SIZE_ORDERED
 	seqcount_t		i_size_seqcount;
@@ -1394,6 +1396,7 @@ static inline void inode_dec_link_count(struct inode *inode)
 	mark_inode_dirty(inode);
 }
 
+extern void inode_inc_iversion(struct inode *inode);
 extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry);
 static inline void file_accessed(struct file *file)
 {
-- 
cgit v1.2.3


From 25ec56b518257a56d2ff41a941d288e4b5ff9488 Mon Sep 17 00:00:00 2001
From: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Add inode version support in ext4

This patch adds 64-bit inode version support to ext4. The lower 32 bits
are stored in the osd1.linux1.l_i_version field while the high 32 bits
are stored in the i_version_hi field newly created in the ext4_inode.
This field is incremented in case the ext4_inode is large enough. A
i_version mount option has been added to enable the feature.

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Kalpak Shah <kalpak@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Jean Noel Cordenner <jean-noel.cordenner@bull.net>
---
 fs/ext4/inode.c         | 18 +++++++++++++++++-
 fs/ext4/super.c         | 10 ++++++++--
 fs/inode.c              | 17 -----------------
 include/linux/ext4_fs.h |  6 +++++-
 include/linux/fs.h      | 16 +++++++++++++++-
 5 files changed, 45 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 89cd35386ff5..a06a3b7cfc34 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2781,6 +2781,13 @@ void ext4_read_inode(struct inode * inode)
 	EXT4_INODE_GET_XTIME(i_atime, inode, raw_inode);
 	EXT4_EINODE_GET_XTIME(i_crtime, ei, raw_inode);
 
+	inode->i_version = le32_to_cpu(raw_inode->i_disk_version);
+	if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
+		if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
+			inode->i_version |=
+			(__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32;
+	}
+
 	if (S_ISREG(inode->i_mode)) {
 		inode->i_op = &ext4_file_inode_operations;
 		inode->i_fop = &ext4_file_operations;
@@ -2963,8 +2970,14 @@ static int ext4_do_update_inode(handle_t *handle,
 	} else for (block = 0; block < EXT4_N_BLOCKS; block++)
 		raw_inode->i_block[block] = ei->i_data[block];
 
-	if (ei->i_extra_isize)
+	raw_inode->i_disk_version = cpu_to_le32(inode->i_version);
+	if (ei->i_extra_isize) {
+		if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
+			raw_inode->i_version_hi =
+			cpu_to_le32(inode->i_version >> 32);
 		raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize);
+	}
+
 
 	BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata");
 	rc = ext4_journal_dirty_metadata(handle, bh);
@@ -3191,6 +3204,9 @@ int ext4_mark_iloc_dirty(handle_t *handle,
 {
 	int err = 0;
 
+	if (test_opt(inode->i_sb, I_VERSION))
+		inode_inc_iversion(inode);
+
 	/* the do_update_inode consumes one bh->b_count */
 	get_bh(iloc->bh);
 
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index f7479d30735e..aa22acd6eb06 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -732,6 +732,8 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",nobh");
 	if (!test_opt(sb, EXTENTS))
 		seq_puts(seq, ",noextents");
+	if (test_opt(sb, I_VERSION))
+		seq_puts(seq, ",i_version");
 
 	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
 		seq_puts(seq, ",data=journal");
@@ -874,7 +876,7 @@ enum {
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
 	Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
-	Opt_grpquota, Opt_extents, Opt_noextents,
+	Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version,
 };
 
 static match_table_t tokens = {
@@ -928,6 +930,7 @@ static match_table_t tokens = {
 	{Opt_barrier, "barrier=%u"},
 	{Opt_extents, "extents"},
 	{Opt_noextents, "noextents"},
+	{Opt_i_version, "i_version"},
 	{Opt_err, NULL},
 	{Opt_resize, "resize"},
 };
@@ -1273,6 +1276,10 @@ clear_qf_name:
 		case Opt_noextents:
 			clear_opt (sbi->s_mount_opt, EXTENTS);
 			break;
+		case Opt_i_version:
+			set_opt(sbi->s_mount_opt, I_VERSION);
+			sb->s_flags |= MS_I_VERSION;
+			break;
 		default:
 			printk (KERN_ERR
 				"EXT4-fs: Unrecognized mount option \"%s\" "
@@ -3197,7 +3204,6 @@ out:
 		i_size_write(inode, off+len-towrite);
 		EXT4_I(inode)->i_disksize = inode->i_size;
 	}
-	inode->i_version++;
 	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	ext4_mark_inode_dirty(handle, inode);
 	mutex_unlock(&inode->i_mutex);
diff --git a/fs/inode.c b/fs/inode.c
index b48324a94c2b..276ffd6b6fdd 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1242,23 +1242,6 @@ void touch_atime(struct vfsmount *mnt, struct dentry *dentry)
 }
 EXPORT_SYMBOL(touch_atime);
 
-/**
- *     inode_inc_iversion      -       increments i_version
- *     @inode: inode that need to be updated
- *
- *     Every time the inode is modified, the i_version field
- *     will be incremented.
- *     The filesystem has to be mounted with i_version flag
- *
- */
-
-void inode_inc_iversion(struct inode *inode)
-{
-	spin_lock(&inode->i_lock);
-	inode->i_version++;
-	spin_unlock(&inode->i_lock);
-}
-
 /**
  *	file_update_time	-	update mtime and ctime time
  *	@file: file accessed
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index cd406dba0e64..b6092940c8b8 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -292,7 +292,7 @@ struct ext4_inode {
 	__le32	i_flags;	/* File flags */
 	union {
 		struct {
-			__u32  l_i_reserved1;
+			__le32  l_i_version;
 		} linux1;
 		struct {
 			__u32  h_i_translator;
@@ -334,6 +334,7 @@ struct ext4_inode {
 	__le32  i_atime_extra;  /* extra Access time      (nsec << 2 | epoch) */
 	__le32  i_crtime;       /* File Creation time */
 	__le32  i_crtime_extra; /* extra FileCreationtime (nsec << 2 | epoch) */
+	__le32  i_version_hi;	/* high 32 bits for 64-bit version */
 };
 
 
@@ -407,6 +408,8 @@ do {									       \
 				       raw_inode->xtime ## _extra);	       \
 } while (0)
 
+#define i_disk_version osd1.linux1.l_i_version
+
 #if defined(__KERNEL__) || defined(__linux__)
 #define i_reserved1	osd1.linux1.l_i_reserved1
 #define i_file_acl_high	osd2.linux2.l_i_file_acl_high
@@ -469,6 +472,7 @@ do {									       \
 #define EXT4_MOUNT_EXTENTS		0x400000 /* Extents support */
 #define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
 #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
+#define EXT4_MOUNT_I_VERSION            0x2000000 /* i_version support */
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
 #define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9608839552b3..a516b6716870 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1396,7 +1396,21 @@ static inline void inode_dec_link_count(struct inode *inode)
 	mark_inode_dirty(inode);
 }
 
-extern void inode_inc_iversion(struct inode *inode);
+/**
+ * inode_inc_iversion - increments i_version
+ * @inode: inode that need to be updated
+ *
+ * Every time the inode is modified, the i_version field will be incremented.
+ * The filesystem has to be mounted with i_version flag
+ */
+
+static inline void inode_inc_iversion(struct inode *inode)
+{
+       spin_lock(&inode->i_lock);
+       inode->i_version++;
+       spin_unlock(&inode->i_lock);
+}
+
 extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry);
 static inline void file_accessed(struct file *file)
 {
-- 
cgit v1.2.3


From c14c6fd5c56a0d0495d8a7c0f2bc330be658663e Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:26 -0500
Subject: ext4: Add EXT4_IOC_MIGRATE ioctl

The below patch add ioctl for migrating ext3 indirect block mapped inode
to ext4 extent mapped inode.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
 fs/ext4/Makefile                |   2 +-
 fs/ext4/extents.c               |   4 +-
 fs/ext4/ioctl.c                 |   3 +
 fs/ext4/migrate.c               | 560 ++++++++++++++++++++++++++++++++++++++++
 include/linux/ext4_fs.h         |   4 +
 include/linux/ext4_fs_extents.h |   2 +
 6 files changed, 572 insertions(+), 3 deletions(-)
 create mode 100644 fs/ext4/migrate.c

(limited to 'include')

diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile
index ae6e7e502ac9..d5fd80bc0d04 100644
--- a/fs/ext4/Makefile
+++ b/fs/ext4/Makefile
@@ -6,7 +6,7 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o
 
 ext4dev-y	:= balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \
 		   ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \
-		   ext4_jbd2.o
+		   ext4_jbd2.o migrate.o
 
 ext4dev-$(CONFIG_EXT4DEV_FS_XATTR)	+= xattr.o xattr_user.o xattr_trusted.o
 ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL)	+= acl.o
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 03d1bbb78a2f..01eda5c5281e 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -61,7 +61,7 @@ static ext4_fsblk_t ext_pblock(struct ext4_extent *ex)
  * idx_pblock:
  * combine low and high parts of a leaf physical block number into ext4_fsblk_t
  */
-static ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix)
+ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix)
 {
 	ext4_fsblk_t block;
 
@@ -75,7 +75,7 @@ static ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix)
  * stores a large physical block number into an extent struct,
  * breaking it into parts
  */
-static void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb)
+void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb)
 {
 	ex->ee_start_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff));
 	ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff);
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index c0e5b8cf635c..2ed7c37f897e 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -254,6 +254,9 @@ flags_err:
 		return err;
 	}
 
+	case EXT4_IOC_MIGRATE:
+		return ext4_ext_migrate(inode, filp, cmd, arg);
+
 	default:
 		return -ENOTTY;
 	}
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
new file mode 100644
index 000000000000..ec7cb567a7da
--- /dev/null
+++ b/fs/ext4/migrate.c
@@ -0,0 +1,560 @@
+/*
+ * Copyright IBM Corporation, 2007
+ * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/ext4_jbd2.h>
+#include <linux/ext4_fs_extents.h>
+
+/*
+ * The contiguous blocks details which can be
+ * represented by a single extent
+ */
+struct list_blocks_struct {
+	ext4_lblk_t first_block, last_block;
+	ext4_fsblk_t first_pblock, last_pblock;
+};
+
+static int finish_range(handle_t *handle, struct inode *inode,
+				struct list_blocks_struct *lb)
+
+{
+	int retval = 0, needed;
+	struct ext4_extent newext;
+	struct ext4_ext_path *path;
+	if (lb->first_pblock == 0)
+		return 0;
+
+	/* Add the extent to temp inode*/
+	newext.ee_block = cpu_to_le32(lb->first_block);
+	newext.ee_len   = cpu_to_le16(lb->last_block - lb->first_block + 1);
+	ext4_ext_store_pblock(&newext, lb->first_pblock);
+	path = ext4_ext_find_extent(inode, lb->first_block, NULL);
+
+	if (IS_ERR(path)) {
+		retval = PTR_ERR(path);
+		goto err_out;
+	}
+
+	/*
+	 * Calculate the credit needed to inserting this extent
+	 * Since we are doing this in loop we may accumalate extra
+	 * credit. But below we try to not accumalate too much
+	 * of them by restarting the journal.
+	 */
+	needed = ext4_ext_calc_credits_for_insert(inode, path);
+
+	/*
+	 * Make sure the credit we accumalated is not really high
+	 */
+	if (needed && handle->h_buffer_credits >= EXT4_RESERVE_TRANS_BLOCKS) {
+		retval = ext4_journal_restart(handle, needed);
+		if (retval)
+			goto err_out;
+	}
+	if (needed) {
+		retval = ext4_journal_extend(handle, needed);
+		if (retval != 0) {
+			/*
+			 * IF not able to extend the journal restart the journal
+			 */
+			retval = ext4_journal_restart(handle, needed);
+			if (retval)
+				goto err_out;
+		}
+	}
+	retval = ext4_ext_insert_extent(handle, inode, path, &newext);
+err_out:
+	lb->first_pblock = 0;
+	return retval;
+}
+
+static int update_extent_range(handle_t *handle, struct inode *inode,
+				ext4_fsblk_t pblock, ext4_lblk_t blk_num,
+				struct list_blocks_struct *lb)
+{
+	int retval;
+	/*
+	 * See if we can add on to the existing range (if it exists)
+	 */
+	if (lb->first_pblock &&
+		(lb->last_pblock+1 == pblock) &&
+		(lb->last_block+1 == blk_num)) {
+		lb->last_pblock = pblock;
+		lb->last_block = blk_num;
+		return 0;
+	}
+	/*
+	 * Start a new range.
+	 */
+	retval = finish_range(handle, inode, lb);
+	lb->first_pblock = lb->last_pblock = pblock;
+	lb->first_block = lb->last_block = blk_num;
+
+	return retval;
+}
+
+static int update_ind_extent_range(handle_t *handle, struct inode *inode,
+				   ext4_fsblk_t pblock, ext4_lblk_t *blk_nump,
+				   struct list_blocks_struct *lb)
+{
+	struct buffer_head *bh;
+	__le32 *i_data;
+	int i, retval = 0;
+	ext4_lblk_t blk_count = *blk_nump;
+	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+
+	if (!pblock) {
+		/* Only update the file block number */
+		*blk_nump += max_entries;
+		return 0;
+	}
+
+	bh = sb_bread(inode->i_sb, pblock);
+	if (!bh)
+		return -EIO;
+
+	i_data = (__le32 *)bh->b_data;
+	for (i = 0; i < max_entries; i++, blk_count++) {
+		if (i_data[i]) {
+			retval = update_extent_range(handle, inode,
+						le32_to_cpu(i_data[i]),
+						blk_count, lb);
+			if (retval)
+				break;
+		}
+	}
+
+	/* Update the file block number */
+	*blk_nump = blk_count;
+	put_bh(bh);
+	return retval;
+
+}
+
+static int update_dind_extent_range(handle_t *handle, struct inode *inode,
+				    ext4_fsblk_t pblock, ext4_lblk_t *blk_nump,
+				    struct list_blocks_struct *lb)
+{
+	struct buffer_head *bh;
+	__le32 *i_data;
+	int i, retval = 0;
+	ext4_lblk_t blk_count = *blk_nump;
+	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+
+	if (!pblock) {
+		/* Only update the file block number */
+		*blk_nump += max_entries * max_entries;
+		return 0;
+	}
+	bh = sb_bread(inode->i_sb, pblock);
+	if (!bh)
+		return -EIO;
+
+	i_data = (__le32 *)bh->b_data;
+	for (i = 0; i < max_entries; i++) {
+		if (i_data[i]) {
+			retval = update_ind_extent_range(handle, inode,
+						le32_to_cpu(i_data[i]),
+						&blk_count, lb);
+			if (retval)
+				break;
+		} else {
+			/* Only update the file block number */
+			blk_count += max_entries;
+		}
+	}
+
+	/* Update the file block number */
+	*blk_nump = blk_count;
+	put_bh(bh);
+	return retval;
+
+}
+
+static int update_tind_extent_range(handle_t *handle, struct inode *inode,
+				     ext4_fsblk_t pblock, ext4_lblk_t *blk_nump,
+				     struct list_blocks_struct *lb)
+{
+	struct buffer_head *bh;
+	__le32 *i_data;
+	int i, retval = 0;
+	ext4_lblk_t blk_count = *blk_nump;
+	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+
+	if (!pblock) {
+		/* Only update the file block number */
+		*blk_nump += max_entries * max_entries * max_entries;
+		return 0;
+	}
+	bh = sb_bread(inode->i_sb, pblock);
+	if (!bh)
+		return -EIO;
+
+	i_data = (__le32 *)bh->b_data;
+	for (i = 0; i < max_entries; i++) {
+		if (i_data[i]) {
+			retval = update_dind_extent_range(handle, inode,
+						le32_to_cpu(i_data[i]),
+						&blk_count, lb);
+			if (retval)
+				break;
+		} else
+			/* Only update the file block number */
+			blk_count += max_entries * max_entries;
+	}
+	/* Update the file block number */
+	*blk_nump = blk_count;
+	put_bh(bh);
+	return retval;
+
+}
+
+static int free_dind_blocks(handle_t *handle,
+				struct inode *inode, __le32 i_data)
+{
+	int i;
+	__le32 *tmp_idata;
+	struct buffer_head *bh;
+	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+
+	bh = sb_bread(inode->i_sb, le32_to_cpu(i_data));
+	if (!bh)
+		return -EIO;
+
+	tmp_idata = (__le32 *)bh->b_data;
+	for (i = 0; i < max_entries; i++) {
+		if (tmp_idata[i])
+			ext4_free_blocks(handle, inode,
+					le32_to_cpu(tmp_idata[i]), 1);
+	}
+	put_bh(bh);
+	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1);
+	return 0;
+}
+
+static int free_tind_blocks(handle_t *handle,
+				struct inode *inode, __le32 i_data)
+{
+	int i, retval = 0;
+	__le32 *tmp_idata;
+	struct buffer_head *bh;
+	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+
+	bh = sb_bread(inode->i_sb, le32_to_cpu(i_data));
+	if (!bh)
+		return -EIO;
+
+	tmp_idata = (__le32 *)bh->b_data;
+	for (i = 0; i < max_entries; i++) {
+		if (tmp_idata[i]) {
+			retval = free_dind_blocks(handle,
+					inode, tmp_idata[i]);
+			if (retval) {
+				put_bh(bh);
+				return retval;
+			}
+		}
+	}
+	put_bh(bh);
+	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1);
+	return 0;
+}
+
+static int free_ind_block(handle_t *handle, struct inode *inode)
+{
+	int retval;
+	struct ext4_inode_info *ei = EXT4_I(inode);
+
+	if (ei->i_data[EXT4_IND_BLOCK])
+		ext4_free_blocks(handle, inode,
+				le32_to_cpu(ei->i_data[EXT4_IND_BLOCK]), 1);
+
+	if (ei->i_data[EXT4_DIND_BLOCK]) {
+		retval = free_dind_blocks(handle, inode,
+						ei->i_data[EXT4_DIND_BLOCK]);
+		if (retval)
+			return retval;
+	}
+
+	if (ei->i_data[EXT4_TIND_BLOCK]) {
+		retval = free_tind_blocks(handle, inode,
+						ei->i_data[EXT4_TIND_BLOCK]);
+		if (retval)
+			return retval;
+	}
+	return 0;
+}
+
+static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
+				struct inode *tmp_inode, int retval)
+{
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	struct ext4_inode_info *tmp_ei = EXT4_I(tmp_inode);
+
+	retval = free_ind_block(handle, inode);
+	if (retval)
+		goto err_out;
+
+	/*
+	 * One credit accounted for writing the
+	 * i_data field of the original inode
+	 */
+	retval = ext4_journal_extend(handle, 1);
+	if (retval != 0) {
+		retval = ext4_journal_restart(handle, 1);
+		if (retval)
+			goto err_out;
+	}
+
+	/*
+	 * We have the extent map build with the tmp inode.
+	 * Now copy the i_data across
+	 */
+	ei->i_flags |= EXT4_EXTENTS_FL;
+	memcpy(ei->i_data, tmp_ei->i_data, sizeof(ei->i_data));
+
+	/*
+	 * Update i_blocks with the new blocks that got
+	 * allocated while adding extents for extent index
+	 * blocks.
+	 *
+	 * While converting to extents we need not
+	 * update the orignal inode i_blocks for extent blocks
+	 * via quota APIs. The quota update happened via tmp_inode already.
+	 */
+	spin_lock(&inode->i_lock);
+	inode->i_blocks += tmp_inode->i_blocks;
+	spin_unlock(&inode->i_lock);
+
+	ext4_mark_inode_dirty(handle, inode);
+err_out:
+	return retval;
+}
+
+static int free_ext_idx(handle_t *handle, struct inode *inode,
+					struct ext4_extent_idx *ix)
+{
+	int i, retval = 0;
+	ext4_fsblk_t block;
+	struct buffer_head *bh;
+	struct ext4_extent_header *eh;
+
+	block = idx_pblock(ix);
+	bh = sb_bread(inode->i_sb, block);
+	if (!bh)
+		return -EIO;
+
+	eh = (struct ext4_extent_header *)bh->b_data;
+	if (eh->eh_depth != 0) {
+		ix = EXT_FIRST_INDEX(eh);
+		for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ix++) {
+			retval = free_ext_idx(handle, inode, ix);
+			if (retval)
+				break;
+		}
+	}
+	put_bh(bh);
+	ext4_free_blocks(handle, inode, block, 1);
+	return retval;
+}
+
+/*
+ * Free the extent meta data blocks only
+ */
+static int free_ext_block(handle_t *handle, struct inode *inode)
+{
+	int i, retval = 0;
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	struct ext4_extent_header *eh = (struct ext4_extent_header *)ei->i_data;
+	struct ext4_extent_idx *ix;
+	if (eh->eh_depth == 0)
+		/*
+		 * No extra blocks allocated for extent meta data
+		 */
+		return 0;
+	ix = EXT_FIRST_INDEX(eh);
+	for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ix++) {
+		retval = free_ext_idx(handle, inode, ix);
+		if (retval)
+			return retval;
+	}
+	return retval;
+
+}
+
+int ext4_ext_migrate(struct inode *inode, struct file *filp,
+				unsigned int cmd, unsigned long arg)
+{
+	handle_t *handle;
+	int retval = 0, i;
+	__le32 *i_data;
+	ext4_lblk_t blk_count = 0;
+	struct ext4_inode_info *ei;
+	struct inode *tmp_inode = NULL;
+	struct list_blocks_struct lb;
+	unsigned long max_entries;
+
+	if (!test_opt(inode->i_sb, EXTENTS))
+		/*
+		 * if mounted with noextents we don't allow the migrate
+		 */
+		return -EINVAL;
+
+	if ((EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
+		return -EINVAL;
+
+	down_write(&EXT4_I(inode)->i_data_sem);
+	handle = ext4_journal_start(inode,
+					EXT4_DATA_TRANS_BLOCKS(inode->i_sb) +
+					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+					2 * EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)
+					+ 1);
+	if (IS_ERR(handle)) {
+		retval = PTR_ERR(handle);
+		goto err_out;
+	}
+	tmp_inode = ext4_new_inode(handle,
+				inode->i_sb->s_root->d_inode,
+				S_IFREG);
+	if (IS_ERR(tmp_inode)) {
+		retval = -ENOMEM;
+		ext4_journal_stop(handle);
+		tmp_inode = NULL;
+		goto err_out;
+	}
+	i_size_write(tmp_inode, i_size_read(inode));
+	/*
+	 * We don't want the inode to be reclaimed
+	 * if we got interrupted in between. We have
+	 * this tmp inode carrying reference to the
+	 * data blocks of the original file. We set
+	 * the i_nlink to zero at the last stage after
+	 * switching the original file to extent format
+	 */
+	tmp_inode->i_nlink = 1;
+
+	ext4_ext_tree_init(handle, tmp_inode);
+	ext4_orphan_add(handle, tmp_inode);
+	ext4_journal_stop(handle);
+
+	ei = EXT4_I(inode);
+	i_data = ei->i_data;
+	memset(&lb, 0, sizeof(lb));
+
+	/* 32 bit block address 4 bytes */
+	max_entries = inode->i_sb->s_blocksize >> 2;
+
+	/*
+	 * start with one credit accounted for
+	 * superblock modification.
+	 *
+	 * For the tmp_inode we already have commited the
+	 * trascation that created the inode. Later as and
+	 * when we add extents we extent the journal
+	 */
+	handle = ext4_journal_start(inode, 1);
+	for (i = 0; i < EXT4_NDIR_BLOCKS; i++, blk_count++) {
+		if (i_data[i]) {
+			retval = update_extent_range(handle, tmp_inode,
+						le32_to_cpu(i_data[i]),
+						blk_count, &lb);
+			if (retval)
+				goto err_out;
+		}
+	}
+	if (i_data[EXT4_IND_BLOCK]) {
+		retval = update_ind_extent_range(handle, tmp_inode,
+					le32_to_cpu(i_data[EXT4_IND_BLOCK]),
+					&blk_count, &lb);
+			if (retval)
+				goto err_out;
+	} else
+		blk_count +=  max_entries;
+	if (i_data[EXT4_DIND_BLOCK]) {
+		retval = update_dind_extent_range(handle, tmp_inode,
+					le32_to_cpu(i_data[EXT4_DIND_BLOCK]),
+					&blk_count, &lb);
+			if (retval)
+				goto err_out;
+	} else
+		blk_count += max_entries * max_entries;
+	if (i_data[EXT4_TIND_BLOCK]) {
+		retval = update_tind_extent_range(handle, tmp_inode,
+					le32_to_cpu(i_data[EXT4_TIND_BLOCK]),
+					&blk_count, &lb);
+			if (retval)
+				goto err_out;
+	}
+	/*
+	 * Build the last extent
+	 */
+	retval = finish_range(handle, tmp_inode, &lb);
+err_out:
+	/*
+	 * We are either freeing extent information or indirect
+	 * blocks. During this we touch superblock, group descriptor
+	 * and block bitmap. Later we mark the tmp_inode dirty
+	 * via ext4_ext_tree_init. So allocate a credit of 4
+	 * We may update quota (user and group).
+	 *
+	 * FIXME!! we may be touching bitmaps in different block groups.
+	 */
+	if (ext4_journal_extend(handle,
+			4 + 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb)) != 0)
+		ext4_journal_restart(handle,
+				4 + 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb));
+	if (retval)
+		/*
+		 * Failure case delete the extent information with the
+		 * tmp_inode
+		 */
+		free_ext_block(handle, tmp_inode);
+	else
+		retval = ext4_ext_swap_inode_data(handle, inode,
+							tmp_inode, retval);
+
+	/*
+	 * Mark the tmp_inode as of size zero
+	 */
+	i_size_write(tmp_inode, 0);
+
+	/*
+	 * set the  i_blocks count to zero
+	 * so that the ext4_delete_inode does the
+	 * right job
+	 *
+	 * We don't need to take the i_lock because
+	 * the inode is not visible to user space.
+	 */
+	tmp_inode->i_blocks = 0;
+
+	/* Reset the extent details */
+	ext4_ext_tree_init(handle, tmp_inode);
+
+	/*
+	 * Set the i_nlink to zero so that
+	 * generic_drop_inode really deletes the
+	 * inode
+	 */
+	tmp_inode->i_nlink = 0;
+
+	ext4_journal_stop(handle);
+
+	up_write(&EXT4_I(inode)->i_data_sem);
+
+	if (tmp_inode)
+		iput(tmp_inode);
+
+	return retval;
+}
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index b6092940c8b8..213974f7c5d8 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -243,6 +243,7 @@ struct ext4_new_group_data {
 #endif
 #define EXT4_IOC_GETRSVSZ		_IOR('f', 5, long)
 #define EXT4_IOC_SETRSVSZ		_IOW('f', 6, long)
+#define EXT4_IOC_MIGRATE		_IO('f', 7)
 
 /*
  * ioctl commands in 32 bit emulation
@@ -983,6 +984,9 @@ extern int ext4_ioctl (struct inode *, struct file *, unsigned int,
 		       unsigned long);
 extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long);
 
+/* migrate.c */
+extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int,
+		       unsigned long);
 /* namei.c */
 extern int ext4_orphan_add(handle_t *, struct inode *);
 extern int ext4_orphan_del(handle_t *, struct inode *);
diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h
index 023683b9d883..1cfb4854c964 100644
--- a/include/linux/ext4_fs_extents.h
+++ b/include/linux/ext4_fs_extents.h
@@ -212,6 +212,8 @@ static inline int ext4_ext_get_actual_len(struct ext4_extent *ext)
 		(le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN));
 }
 
+extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *);
+extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t);
 extern int ext4_extent_tree_init(handle_t *, struct inode *);
 extern int ext4_ext_calc_credits_for_insert(struct inode *, struct ext4_ext_path *);
 extern int ext4_ext_try_to_merge(struct inode *inode,
-- 
cgit v1.2.3


From aa02ad67d9b308290fde390682cd039b29f7ab85 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Add ext4_find_next_bit()

This function is used by the ext4 multi block allocator patches.

Also add generic_find_next_le_bit

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 include/asm-arm/bitops.h                     |  2 ++
 include/asm-generic/bitops/ext2-non-atomic.h |  2 ++
 include/asm-generic/bitops/le.h              |  4 +++
 include/asm-m68k/bitops.h                    |  2 ++
 include/asm-m68knommu/bitops.h               |  2 ++
 include/asm-powerpc/bitops.h                 |  4 +++
 include/asm-s390/bitops.h                    |  2 ++
 include/linux/ext4_fs.h                      |  1 +
 lib/find_next_bit.c                          | 43 ++++++++++++++++++++++++++++
 9 files changed, 62 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/bitops.h b/include/asm-arm/bitops.h
index 47a6b086eee2..5c60bfc1a84d 100644
--- a/include/asm-arm/bitops.h
+++ b/include/asm-arm/bitops.h
@@ -310,6 +310,8 @@ static inline int constant_fls(int x)
 		_find_first_zero_bit_le(p,sz)
 #define ext2_find_next_zero_bit(p,sz,off)	\
 		_find_next_zero_bit_le(p,sz,off)
+#define ext2_find_next_bit(p, sz, off) \
+		_find_next_bit_le(p, sz, off)
 
 /*
  * Minix is defined to use little-endian byte ordering.
diff --git a/include/asm-generic/bitops/ext2-non-atomic.h b/include/asm-generic/bitops/ext2-non-atomic.h
index 1697404afa05..63cf822431a2 100644
--- a/include/asm-generic/bitops/ext2-non-atomic.h
+++ b/include/asm-generic/bitops/ext2-non-atomic.h
@@ -14,5 +14,7 @@
 	generic_find_first_zero_le_bit((unsigned long *)(addr), (size))
 #define ext2_find_next_zero_bit(addr, size, off) \
 	generic_find_next_zero_le_bit((unsigned long *)(addr), (size), (off))
+#define ext2_find_next_bit(addr, size, off) \
+	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 
 #endif /* _ASM_GENERIC_BITOPS_EXT2_NON_ATOMIC_H_ */
diff --git a/include/asm-generic/bitops/le.h b/include/asm-generic/bitops/le.h
index b9c7e5d2d2ad..80e3bf13b2b9 100644
--- a/include/asm-generic/bitops/le.h
+++ b/include/asm-generic/bitops/le.h
@@ -20,6 +20,8 @@
 #define generic___test_and_clear_le_bit(nr, addr) __test_and_clear_bit(nr, addr)
 
 #define generic_find_next_zero_le_bit(addr, size, offset) find_next_zero_bit(addr, size, offset)
+#define generic_find_next_le_bit(addr, size, offset) \
+			find_next_bit(addr, size, offset)
 
 #elif defined(__BIG_ENDIAN)
 
@@ -42,6 +44,8 @@
 
 extern unsigned long generic_find_next_zero_le_bit(const unsigned long *addr,
 		unsigned long size, unsigned long offset);
+extern unsigned long generic_find_next_le_bit(const unsigned long *addr,
+		unsigned long size, unsigned long offset);
 
 #else
 #error "Please fix <asm/byteorder.h>"
diff --git a/include/asm-m68k/bitops.h b/include/asm-m68k/bitops.h
index 2976b5d68e96..83d1f286230b 100644
--- a/include/asm-m68k/bitops.h
+++ b/include/asm-m68k/bitops.h
@@ -410,6 +410,8 @@ static inline int ext2_find_next_zero_bit(const void *vaddr, unsigned size,
 	res = ext2_find_first_zero_bit (p, size - 32 * (p - addr));
 	return (p - addr) * 32 + res;
 }
+#define ext2_find_next_bit(addr, size, off) \
+	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 
 #endif /* __KERNEL__ */
 
diff --git a/include/asm-m68knommu/bitops.h b/include/asm-m68knommu/bitops.h
index f8dfb7ba2e25..f43afe1fc3b3 100644
--- a/include/asm-m68knommu/bitops.h
+++ b/include/asm-m68knommu/bitops.h
@@ -294,6 +294,8 @@ found_middle:
 	return result + ffz(__swab32(tmp));
 }
 
+#define ext2_find_next_bit(addr, size, off) \
+	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 #include <asm-generic/bitops/minix.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-powerpc/bitops.h b/include/asm-powerpc/bitops.h
index 733b4af7f4f1..220d9a781ab9 100644
--- a/include/asm-powerpc/bitops.h
+++ b/include/asm-powerpc/bitops.h
@@ -359,6 +359,8 @@ static __inline__ int test_le_bit(unsigned long nr,
 unsigned long generic_find_next_zero_le_bit(const unsigned long *addr,
 				    unsigned long size, unsigned long offset);
 
+unsigned long generic_find_next_le_bit(const unsigned long *addr,
+				    unsigned long size, unsigned long offset);
 /* Bitmap functions for the ext2 filesystem */
 
 #define ext2_set_bit(nr,addr) \
@@ -378,6 +380,8 @@ unsigned long generic_find_next_zero_le_bit(const unsigned long *addr,
 #define ext2_find_next_zero_bit(addr, size, off) \
 	generic_find_next_zero_le_bit((unsigned long*)addr, size, off)
 
+#define ext2_find_next_bit(addr, size, off) \
+	generic_find_next_le_bit((unsigned long *)addr, size, off)
 /* Bitmap functions for the minix filesystem.  */
 
 #define minix_test_and_set_bit(nr,addr) \
diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index 34d9a6357c38..dba6fecad0be 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -772,6 +772,8 @@ static inline int sched_find_first_bit(unsigned long *b)
 	test_and_clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (unsigned long *)addr)
 #define ext2_test_bit(nr, addr)      \
 	test_bit((nr)^(__BITOPS_WORDSIZE - 8), (unsigned long *)addr)
+#define ext2_find_next_bit(addr, size, off) \
+	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 
 #ifndef __s390x__
 
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 213974f7c5d8..d0b7ca99b91f 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -493,6 +493,7 @@ do {									       \
 #define ext4_test_bit			ext2_test_bit
 #define ext4_find_first_zero_bit	ext2_find_first_zero_bit
 #define ext4_find_next_zero_bit		ext2_find_next_zero_bit
+#define ext4_find_next_bit		ext2_find_next_bit
 
 /*
  * Maximal mount counts between two filesystem checks
diff --git a/lib/find_next_bit.c b/lib/find_next_bit.c
index bda0d71a2514..78ccd73a8841 100644
--- a/lib/find_next_bit.c
+++ b/lib/find_next_bit.c
@@ -178,4 +178,47 @@ found_middle_swap:
 
 EXPORT_SYMBOL(generic_find_next_zero_le_bit);
 
+unsigned long generic_find_next_le_bit(const unsigned long *addr, unsigned
+		long size, unsigned long offset)
+{
+	const unsigned long *p = addr + BITOP_WORD(offset);
+	unsigned long result = offset & ~(BITS_PER_LONG - 1);
+	unsigned long tmp;
+
+	if (offset >= size)
+		return size;
+	size -= result;
+	offset &= (BITS_PER_LONG - 1UL);
+	if (offset) {
+		tmp = ext2_swabp(p++);
+		tmp &= (~0UL << offset);
+		if (size < BITS_PER_LONG)
+			goto found_first;
+		if (tmp)
+			goto found_middle;
+		size -= BITS_PER_LONG;
+		result += BITS_PER_LONG;
+	}
+
+	while (size & ~(BITS_PER_LONG - 1)) {
+		tmp = *(p++);
+		if (tmp)
+			goto found_middle_swap;
+		result += BITS_PER_LONG;
+		size -= BITS_PER_LONG;
+	}
+	if (!size)
+		return result;
+	tmp = ext2_swabp(p);
+found_first:
+	tmp &= (~0UL >> (BITS_PER_LONG - size));
+	if (tmp == 0UL)		/* Are any bits set? */
+		return result + size; /* Nope. */
+found_middle:
+	return result + __ffs(tmp);
+
+found_middle_swap:
+	return result + __ffs(ext2_swab(tmp));
+}
+EXPORT_SYMBOL(generic_find_next_le_bit);
 #endif /* __BIG_ENDIAN */
-- 
cgit v1.2.3


From 1988b51e476bd097d910c9245b53f2e38aedaf0d Mon Sep 17 00:00:00 2001
From: Alex Tomas <alex@clusterfs.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: ext4: Add new functions for searching extent tree

Add the functions ext4_ext_search_left() and ext4_ext_search_right(),
which are used by mballoc during ext4_ext_get_blocks to decided whether
to merge extent information.

Signed-off-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Johann Lombardi <johann@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext4/extents.c               | 142 ++++++++++++++++++++++++++++++++++++++++
 include/linux/ext4_fs_extents.h |   4 ++
 2 files changed, 146 insertions(+)

(limited to 'include')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 01eda5c5281e..f5cf2a94b6fc 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1016,6 +1016,148 @@ out:
 	return err;
 }
 
+/*
+ * search the closest allocated block to the left for *logical
+ * and returns it at @logical + it's physical address at @phys
+ * if *logical is the smallest allocated block, the function
+ * returns 0 at @phys
+ * return value contains 0 (success) or error code
+ */
+int
+ext4_ext_search_left(struct inode *inode, struct ext4_ext_path *path,
+			ext4_lblk_t *logical, ext4_fsblk_t *phys)
+{
+	struct ext4_extent_idx *ix;
+	struct ext4_extent *ex;
+	int depth;
+
+	BUG_ON(path == NULL);
+	depth = path->p_depth;
+	*phys = 0;
+
+	if (depth == 0 && path->p_ext == NULL)
+		return 0;
+
+	/* usually extent in the path covers blocks smaller
+	 * then *logical, but it can be that extent is the
+	 * first one in the file */
+
+	ex = path[depth].p_ext;
+	if (*logical < le32_to_cpu(ex->ee_block)) {
+		BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
+		while (--depth >= 0) {
+			ix = path[depth].p_idx;
+			BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
+		}
+		return 0;
+	}
+
+	BUG_ON(*logical < le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len));
+
+	*logical = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1;
+	*phys = ext_pblock(ex) + le16_to_cpu(ex->ee_len) - 1;
+	return 0;
+}
+
+/*
+ * search the closest allocated block to the right for *logical
+ * and returns it at @logical + it's physical address at @phys
+ * if *logical is the smallest allocated block, the function
+ * returns 0 at @phys
+ * return value contains 0 (success) or error code
+ */
+int
+ext4_ext_search_right(struct inode *inode, struct ext4_ext_path *path,
+			ext4_lblk_t *logical, ext4_fsblk_t *phys)
+{
+	struct buffer_head *bh = NULL;
+	struct ext4_extent_header *eh;
+	struct ext4_extent_idx *ix;
+	struct ext4_extent *ex;
+	ext4_fsblk_t block;
+	int depth;
+
+	BUG_ON(path == NULL);
+	depth = path->p_depth;
+	*phys = 0;
+
+	if (depth == 0 && path->p_ext == NULL)
+		return 0;
+
+	/* usually extent in the path covers blocks smaller
+	 * then *logical, but it can be that extent is the
+	 * first one in the file */
+
+	ex = path[depth].p_ext;
+	if (*logical < le32_to_cpu(ex->ee_block)) {
+		BUG_ON(EXT_FIRST_EXTENT(path[depth].p_hdr) != ex);
+		while (--depth >= 0) {
+			ix = path[depth].p_idx;
+			BUG_ON(ix != EXT_FIRST_INDEX(path[depth].p_hdr));
+		}
+		*logical = le32_to_cpu(ex->ee_block);
+		*phys = ext_pblock(ex);
+		return 0;
+	}
+
+	BUG_ON(*logical < le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len));
+
+	if (ex != EXT_LAST_EXTENT(path[depth].p_hdr)) {
+		/* next allocated block in this leaf */
+		ex++;
+		*logical = le32_to_cpu(ex->ee_block);
+		*phys = ext_pblock(ex);
+		return 0;
+	}
+
+	/* go up and search for index to the right */
+	while (--depth >= 0) {
+		ix = path[depth].p_idx;
+		if (ix != EXT_LAST_INDEX(path[depth].p_hdr))
+			break;
+	}
+
+	if (depth < 0) {
+		/* we've gone up to the root and
+		 * found no index to the right */
+		return 0;
+	}
+
+	/* we've found index to the right, let's
+	 * follow it and find the closest allocated
+	 * block to the right */
+	ix++;
+	block = idx_pblock(ix);
+	while (++depth < path->p_depth) {
+		bh = sb_bread(inode->i_sb, block);
+		if (bh == NULL)
+			return -EIO;
+		eh = ext_block_hdr(bh);
+		if (ext4_ext_check_header(inode, eh, depth)) {
+			put_bh(bh);
+			return -EIO;
+		}
+		ix = EXT_FIRST_INDEX(eh);
+		block = idx_pblock(ix);
+		put_bh(bh);
+	}
+
+	bh = sb_bread(inode->i_sb, block);
+	if (bh == NULL)
+		return -EIO;
+	eh = ext_block_hdr(bh);
+	if (ext4_ext_check_header(inode, eh, path->p_depth - depth)) {
+		put_bh(bh);
+		return -EIO;
+	}
+	ex = EXT_FIRST_EXTENT(eh);
+	*logical = le32_to_cpu(ex->ee_block);
+	*phys = ext_pblock(ex);
+	put_bh(bh);
+	return 0;
+
+}
+
 /*
  * ext4_ext_next_allocated_block:
  * returns allocated block in subsequent extent or EXT_MAX_BLOCK.
diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h
index 1cfb4854c964..697da4bce6c5 100644
--- a/include/linux/ext4_fs_extents.h
+++ b/include/linux/ext4_fs_extents.h
@@ -223,5 +223,9 @@ extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *,
 extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *);
 extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
 							struct ext4_ext_path *);
+extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *,
+						ext4_lblk_t *, ext4_fsblk_t *);
+extern int ext4_ext_search_right(struct inode *, struct ext4_ext_path *,
+						ext4_lblk_t *, ext4_fsblk_t *);
 #endif /* _LINUX_EXT4_EXTENTS */
 
-- 
cgit v1.2.3


From c9de560ded61faa5b754137b7753da252391c55a Mon Sep 17 00:00:00 2001
From: Alex Tomas <alex@clusterfs.com>
Date: Tue, 29 Jan 2008 00:19:52 -0500
Subject: ext4: Add multi block allocator for ext4

Signed-off-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 Documentation/filesystems/ext4.txt |   10 +-
 Documentation/filesystems/proc.txt |   39 +
 fs/ext4/Makefile                   |    2 +-
 fs/ext4/balloc.c                   |   67 +-
 fs/ext4/extents.c                  |   45 +-
 fs/ext4/inode.c                    |   15 +-
 fs/ext4/mballoc.c                  | 4552 ++++++++++++++++++++++++++++++++++++
 fs/ext4/migrate.c                  |   10 +-
 fs/ext4/super.c                    |   62 +-
 fs/ext4/xattr.c                    |    4 +-
 include/linux/ext4_fs.h            |   76 +-
 include/linux/ext4_fs_i.h          |    4 +
 include/linux/ext4_fs_sb.h         |   52 +
 13 files changed, 4900 insertions(+), 38 deletions(-)
 create mode 100644 fs/ext4/mballoc.c

(limited to 'include')

diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index 4f329afe20ec..560f88dc7090 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -86,9 +86,11 @@ Alex is working on a new set of patches right now.
 When mounting an ext4 filesystem, the following option are accepted:
 (*) == default
 
-extents			ext4 will use extents to address file data.  The
+extents		(*)	ext4 will use extents to address file data.  The
 			file system will no longer be mountable by ext3.
 
+noextents		ext4 will not use extents for newly created files
+
 journal_checksum	Enable checksumming of the journal transactions.
 			This will allow the recovery code in e2fsck and the
 			kernel to detect corruption in the kernel.  It is a
@@ -206,6 +208,12 @@ nobh			(a) cache disk block mapping information
 			"nobh" option tries to avoid associating buffer
 			heads (supported only for "writeback" mode).
 
+mballoc		(*)	Use the multiple block allocator for block allocation
+nomballoc		disabled multiple block allocator for block allocation.
+stripe=n		Number of filesystem blocks that mballoc will try
+			to use for allocation size and alignment. For RAID5/6
+			systems this should be the number of data
+			disks *  RAID chunk size in file system blocks.
 
 Data Mode
 ---------
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index dec99455321f..4413a2d4646f 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -857,6 +857,45 @@ CPUs.
 The   "procs_blocked" line gives  the  number of  processes currently blocked,
 waiting for I/O to complete.
 
+1.9 Ext4 file system parameters
+------------------------------
+Ext4 file system have one directory per partition under /proc/fs/ext4/
+# ls /proc/fs/ext4/hdc/
+group_prealloc  max_to_scan  mb_groups  mb_history  min_to_scan  order2_req
+stats  stream_req
+
+mb_groups:
+This file gives the details of mutiblock allocator buddy cache of free blocks
+
+mb_history:
+Multiblock allocation history.
+
+stats:
+This file indicate whether the multiblock allocator should start collecting
+statistics. The statistics are shown during unmount
+
+group_prealloc:
+The multiblock allocator normalize the block allocation request to
+group_prealloc filesystem blocks if we don't have strip value set.
+The stripe value can be specified at mount time or during mke2fs.
+
+max_to_scan:
+How long multiblock allocator can look for a best extent (in found extents)
+
+min_to_scan:
+How long multiblock allocator  must look for a best extent
+
+order2_req:
+Multiblock allocator use  2^N search using buddies only for requests greater
+than or equal to order2_req. The request size is specfied in file system
+blocks. A value of 2 indicate only if the requests are greater than or equal
+to 4 blocks.
+
+stream_req:
+Files smaller than stream_req are served by the stream allocator, whose
+purpose is to pack requests as close each to other as possible to
+produce smooth I/O traffic. Avalue of 16 indicate that file smaller than 16
+filesystem block size will use group based preallocation.
 
 ------------------------------------------------------------------------------
 Summary
diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile
index d5fd80bc0d04..ac6fa8ca0a2f 100644
--- a/fs/ext4/Makefile
+++ b/fs/ext4/Makefile
@@ -6,7 +6,7 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o
 
 ext4dev-y	:= balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \
 		   ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \
-		   ext4_jbd2.o migrate.o
+		   ext4_jbd2.o migrate.o mballoc.o
 
 ext4dev-$(CONFIG_EXT4DEV_FS_XATTR)	+= xattr.o xattr_user.o xattr_trusted.o
 ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL)	+= acl.o
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 80a4616c8244..ac75ea953d83 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -577,6 +577,8 @@ void ext4_discard_reservation(struct inode *inode)
 	struct ext4_reserve_window_node *rsv;
 	spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock;
 
+	ext4_mb_discard_inode_preallocations(inode);
+
 	if (!block_i)
 		return;
 
@@ -785,19 +787,29 @@ error_return:
  * @inode:		inode
  * @block:		start physical block to free
  * @count:		number of blocks to count
+ * @metadata: 		Are these metadata blocks
  */
 void ext4_free_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t block, unsigned long count)
+			ext4_fsblk_t block, unsigned long count,
+			int metadata)
 {
 	struct super_block * sb;
 	unsigned long dquot_freed_blocks;
 
+	/* this isn't the right place to decide whether block is metadata
+	 * inode.c/extents.c knows better, but for safety ... */
+	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode) ||
+			ext4_should_journal_data(inode))
+		metadata = 1;
+
 	sb = inode->i_sb;
-	if (!sb) {
-		printk ("ext4_free_blocks: nonexistent device");
-		return;
-	}
-	ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks);
+
+	if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info)
+		ext4_free_blocks_sb(handle, sb, block, count,
+						&dquot_freed_blocks);
+	else
+		ext4_mb_free_blocks(handle, inode, block, count,
+						metadata, &dquot_freed_blocks);
 	if (dquot_freed_blocks)
 		DQUOT_FREE_BLOCK(inode, dquot_freed_blocks);
 	return;
@@ -1576,7 +1588,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries)
 }
 
 /**
- * ext4_new_blocks() -- core block(s) allocation function
+ * ext4_new_blocks_old() -- core block(s) allocation function
  * @handle:		handle to this transaction
  * @inode:		file inode
  * @goal:		given target block(filesystem wide)
@@ -1589,7 +1601,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries)
  * any specific goal block.
  *
  */
-ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode,
+ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, unsigned long *count, int *errp)
 {
 	struct buffer_head *bitmap_bh = NULL;
@@ -1849,13 +1861,46 @@ out:
 }
 
 ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t goal, int *errp)
+		ext4_fsblk_t goal, int *errp)
+{
+	struct ext4_allocation_request ar;
+	ext4_fsblk_t ret;
+
+	if (!test_opt(inode->i_sb, MBALLOC)) {
+		unsigned long count = 1;
+		ret = ext4_new_blocks_old(handle, inode, goal, &count, errp);
+		return ret;
+	}
+
+	memset(&ar, 0, sizeof(ar));
+	ar.inode = inode;
+	ar.goal = goal;
+	ar.len = 1;
+	ret = ext4_mb_new_blocks(handle, &ar, errp);
+	return ret;
+}
+
+ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode,
+		ext4_fsblk_t goal, unsigned long *count, int *errp)
 {
-	unsigned long count = 1;
+	struct ext4_allocation_request ar;
+	ext4_fsblk_t ret;
 
-	return ext4_new_blocks(handle, inode, goal, &count, errp);
+	if (!test_opt(inode->i_sb, MBALLOC)) {
+		ret = ext4_new_blocks_old(handle, inode, goal, count, errp);
+		return ret;
+	}
+
+	memset(&ar, 0, sizeof(ar));
+	ar.inode = inode;
+	ar.goal = goal;
+	ar.len = *count;
+	ret = ext4_mb_new_blocks(handle, &ar, errp);
+	*count = ar.len;
+	return ret;
 }
 
+
 /**
  * ext4_count_free_blocks() -- count filesystem free blocks
  * @sb:		superblock
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index f5cf2a94b6fc..0cffb59fff46 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -853,7 +853,7 @@ cleanup:
 		for (i = 0; i < depth; i++) {
 			if (!ablocks[i])
 				continue;
-			ext4_free_blocks(handle, inode, ablocks[i], 1);
+			ext4_free_blocks(handle, inode, ablocks[i], 1, 1);
 		}
 	}
 	kfree(ablocks);
@@ -1698,7 +1698,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
 	ext_debug("index is empty, remove it, free block %llu\n", leaf);
 	bh = sb_find_get_block(inode->i_sb, leaf);
 	ext4_forget(handle, 1, inode, bh, leaf);
-	ext4_free_blocks(handle, inode, leaf, 1);
+	ext4_free_blocks(handle, inode, leaf, 1, 1);
 	return err;
 }
 
@@ -1759,8 +1759,10 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 {
 	struct buffer_head *bh;
 	unsigned short ee_len =  ext4_ext_get_actual_len(ex);
-	int i;
+	int i, metadata = 0;
 
+	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
+		metadata = 1;
 #ifdef EXTENTS_STATS
 	{
 		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
@@ -1789,7 +1791,7 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 			bh = sb_find_get_block(inode->i_sb, start + i);
 			ext4_forget(handle, 0, inode, bh, start + i);
 		}
-		ext4_free_blocks(handle, inode, start, num);
+		ext4_free_blocks(handle, inode, start, num, metadata);
 	} else if (from == le32_to_cpu(ex->ee_block)
 		   && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) {
 		printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n",
@@ -2287,6 +2289,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	ext4_fsblk_t goal, newblock;
 	int err = 0, depth, ret;
 	unsigned long allocated = 0;
+	struct ext4_allocation_request ar;
 
 	__clear_bit(BH_New, &bh_result->b_state);
 	ext_debug("blocks %u/%lu requested for inode %u\n",
@@ -2397,8 +2400,15 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info))
 		ext4_init_block_alloc_info(inode);
 
-	/* allocate new block */
-	goal = ext4_ext_find_goal(inode, path, iblock);
+	/* find neighbour allocated blocks */
+	ar.lleft = iblock;
+	err = ext4_ext_search_left(inode, path, &ar.lleft, &ar.pleft);
+	if (err)
+		goto out2;
+	ar.lright = iblock;
+	err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright);
+	if (err)
+		goto out2;
 
 	/*
 	 * See if request is beyond maximum number of blocks we can have in
@@ -2421,7 +2431,18 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 		allocated = le16_to_cpu(newex.ee_len);
 	else
 		allocated = max_blocks;
-	newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err);
+
+	/* allocate new block */
+	ar.inode = inode;
+	ar.goal = ext4_ext_find_goal(inode, path, iblock);
+	ar.logical = iblock;
+	ar.len = allocated;
+	if (S_ISREG(inode->i_mode))
+		ar.flags = EXT4_MB_HINT_DATA;
+	else
+		/* disable in-core preallocation for non-regular files */
+		ar.flags = 0;
+	newblock = ext4_mb_new_blocks(handle, &ar, &err);
 	if (!newblock)
 		goto out2;
 	ext_debug("allocate new block: goal %llu, found %llu/%lu\n",
@@ -2429,14 +2450,17 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 
 	/* try to insert new extent into found leaf and return */
 	ext4_ext_store_pblock(&newex, newblock);
-	newex.ee_len = cpu_to_le16(allocated);
+	newex.ee_len = cpu_to_le16(ar.len);
 	if (create == EXT4_CREATE_UNINITIALIZED_EXT)  /* Mark uninitialized */
 		ext4_ext_mark_uninitialized(&newex);
 	err = ext4_ext_insert_extent(handle, inode, path, &newex);
 	if (err) {
 		/* free data blocks we just allocated */
+		/* not a good idea to call discard here directly,
+		 * but otherwise we'd need to call it every free() */
+		ext4_mb_discard_inode_preallocations(inode);
 		ext4_free_blocks(handle, inode, ext_pblock(&newex),
-					le16_to_cpu(newex.ee_len));
+					le16_to_cpu(newex.ee_len), 0);
 		goto out2;
 	}
 
@@ -2445,6 +2469,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 
 	/* previous routine could use block we allocated */
 	newblock = ext_pblock(&newex);
+	allocated = le16_to_cpu(newex.ee_len);
 outnew:
 	__set_bit(BH_New, &bh_result->b_state);
 
@@ -2496,6 +2521,8 @@ void ext4_ext_truncate(struct inode * inode, struct page *page)
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_ext_invalidate_cache(inode);
 
+	ext4_mb_discard_inode_preallocations(inode);
+
 	/*
 	 * TODO: optimization is possible here.
 	 * Probably we need not scan at all,
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index a06a3b7cfc34..bb717cbb749c 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -551,7 +551,7 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode,
 	return ret;
 failed_out:
 	for (i = 0; i <index; i++)
-		ext4_free_blocks(handle, inode, new_blocks[i], 1);
+		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
 	return ret;
 }
 
@@ -650,9 +650,9 @@ failed:
 		ext4_journal_forget(handle, branch[i].bh);
 	}
 	for (i = 0; i <indirect_blks; i++)
-		ext4_free_blocks(handle, inode, new_blocks[i], 1);
+		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
 
-	ext4_free_blocks(handle, inode, new_blocks[i], num);
+	ext4_free_blocks(handle, inode, new_blocks[i], num, 0);
 
 	return err;
 }
@@ -749,9 +749,10 @@ err_out:
 	for (i = 1; i <= num; i++) {
 		BUFFER_TRACE(where[i].bh, "call jbd2_journal_forget");
 		ext4_journal_forget(handle, where[i].bh);
-		ext4_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1);
+		ext4_free_blocks(handle, inode,
+					le32_to_cpu(where[i-1].key), 1, 0);
 	}
-	ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks);
+	ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks, 0);
 
 	return err;
 }
@@ -2052,7 +2053,7 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode,
 		}
 	}
 
-	ext4_free_blocks(handle, inode, block_to_free, count);
+	ext4_free_blocks(handle, inode, block_to_free, count, 0);
 }
 
 /**
@@ -2225,7 +2226,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 				ext4_journal_test_restart(handle, inode);
 			}
 
-			ext4_free_blocks(handle, inode, nr, 1);
+			ext4_free_blocks(handle, inode, nr, 1, 1);
 
 			if (parent_bh) {
 				/*
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
new file mode 100644
index 000000000000..76e5fedc0a0b
--- /dev/null
+++ b/fs/ext4/mballoc.c
@@ -0,0 +1,4552 @@
+/*
+ * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com
+ * Written by Alex Tomas <alex@clusterfs.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public Licens
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
+ */
+
+
+/*
+ * mballoc.c contains the multiblocks allocation routines
+ */
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/ext4_jbd2.h>
+#include <linux/ext4_fs.h>
+#include <linux/quotaops.h>
+#include <linux/buffer_head.h>
+#include <linux/module.h>
+#include <linux/swap.h>
+#include <linux/proc_fs.h>
+#include <linux/pagemap.h>
+#include <linux/seq_file.h>
+#include <linux/version.h>
+#include "group.h"
+
+/*
+ * MUSTDO:
+ *   - test ext4_ext_search_left() and ext4_ext_search_right()
+ *   - search for metadata in few groups
+ *
+ * TODO v4:
+ *   - normalization should take into account whether file is still open
+ *   - discard preallocations if no free space left (policy?)
+ *   - don't normalize tails
+ *   - quota
+ *   - reservation for superuser
+ *
+ * TODO v3:
+ *   - bitmap read-ahead (proposed by Oleg Drokin aka green)
+ *   - track min/max extents in each group for better group selection
+ *   - mb_mark_used() may allocate chunk right after splitting buddy
+ *   - tree of groups sorted by number of free blocks
+ *   - error handling
+ */
+
+/*
+ * The allocation request involve request for multiple number of blocks
+ * near to the goal(block) value specified.
+ *
+ * During initialization phase of the allocator we decide to use the group
+ * preallocation or inode preallocation depending on the size file. The
+ * size of the file could be the resulting file size we would have after
+ * allocation or the current file size which ever is larger. If the size is
+ * less that sbi->s_mb_stream_request we select the group
+ * preallocation. The default value of s_mb_stream_request is 16
+ * blocks. This can also be tuned via
+ * /proc/fs/ext4/<partition>/stream_req. The value is represented in terms
+ * of number of blocks.
+ *
+ * The main motivation for having small file use group preallocation is to
+ * ensure that we have small file closer in the disk.
+ *
+ * First stage the allocator looks at the inode prealloc list
+ * ext4_inode_info->i_prealloc_list contain list of prealloc spaces for
+ * this particular inode. The inode prealloc space is represented as:
+ *
+ * pa_lstart -> the logical start block for this prealloc space
+ * pa_pstart -> the physical start block for this prealloc space
+ * pa_len    -> lenght for this prealloc space
+ * pa_free   ->  free space available in this prealloc space
+ *
+ * The inode preallocation space is used looking at the _logical_ start
+ * block. If only the logical file block falls within the range of prealloc
+ * space we will consume the particular prealloc space. This make sure that
+ * that the we have contiguous physical blocks representing the file blocks
+ *
+ * The important thing to be noted in case of inode prealloc space is that
+ * we don't modify the values associated to inode prealloc space except
+ * pa_free.
+ *
+ * If we are not able to find blocks in the inode prealloc space and if we
+ * have the group allocation flag set then we look at the locality group
+ * prealloc space. These are per CPU prealloc list repreasented as
+ *
+ * ext4_sb_info.s_locality_groups[smp_processor_id()]
+ *
+ * The reason for having a per cpu locality group is to reduce the contention
+ * between CPUs. It is possible to get scheduled at this point.
+ *
+ * The locality group prealloc space is used looking at whether we have
+ * enough free space (pa_free) withing the prealloc space.
+ *
+ * If we can't allocate blocks via inode prealloc or/and locality group
+ * prealloc then we look at the buddy cache. The buddy cache is represented
+ * by ext4_sb_info.s_buddy_cache (struct inode) whose file offset gets
+ * mapped to the buddy and bitmap information regarding different
+ * groups. The buddy information is attached to buddy cache inode so that
+ * we can access them through the page cache. The information regarding
+ * each group is loaded via ext4_mb_load_buddy.  The information involve
+ * block bitmap and buddy information. The information are stored in the
+ * inode as:
+ *
+ *  {                        page                        }
+ *  [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]...
+ *
+ *
+ * one block each for bitmap and buddy information.  So for each group we
+ * take up 2 blocks. A page can contain blocks_per_page (PAGE_CACHE_SIZE /
+ * blocksize) blocks.  So it can have information regarding groups_per_page
+ * which is blocks_per_page/2
+ *
+ * The buddy cache inode is not stored on disk. The inode is thrown
+ * away when the filesystem is unmounted.
+ *
+ * We look for count number of blocks in the buddy cache. If we were able
+ * to locate that many free blocks we return with additional information
+ * regarding rest of the contiguous physical block available
+ *
+ * Before allocating blocks via buddy cache we normalize the request
+ * blocks. This ensure we ask for more blocks that we needed. The extra
+ * blocks that we get after allocation is added to the respective prealloc
+ * list. In case of inode preallocation we follow a list of heuristics
+ * based on file size. This can be found in ext4_mb_normalize_request. If
+ * we are doing a group prealloc we try to normalize the request to
+ * sbi->s_mb_group_prealloc. Default value of s_mb_group_prealloc is set to
+ * 512 blocks. This can be tuned via
+ * /proc/fs/ext4/<partition/group_prealloc. The value is represented in
+ * terms of number of blocks. If we have mounted the file system with -O
+ * stripe=<value> option the group prealloc request is normalized to the
+ * stripe value (sbi->s_stripe)
+ *
+ * The regular allocator(using the buddy cache) support few tunables.
+ *
+ * /proc/fs/ext4/<partition>/min_to_scan
+ * /proc/fs/ext4/<partition>/max_to_scan
+ * /proc/fs/ext4/<partition>/order2_req
+ *
+ * The regular allocator use buddy scan only if the request len is power of
+ * 2 blocks and the order of allocation is >= sbi->s_mb_order2_reqs. The
+ * value of s_mb_order2_reqs can be tuned via
+ * /proc/fs/ext4/<partition>/order2_req.  If the request len is equal to
+ * stripe size (sbi->s_stripe), we try to search for contigous block in
+ * stripe size. This should result in better allocation on RAID setup. If
+ * not we search in the specific group using bitmap for best extents. The
+ * tunable min_to_scan and max_to_scan controll the behaviour here.
+ * min_to_scan indicate how long the mballoc __must__ look for a best
+ * extent and max_to_scanindicate how long the mballoc __can__ look for a
+ * best extent in the found extents. Searching for the blocks starts with
+ * the group specified as the goal value in allocation context via
+ * ac_g_ex. Each group is first checked based on the criteria whether it
+ * can used for allocation. ext4_mb_good_group explains how the groups are
+ * checked.
+ *
+ * Both the prealloc space are getting populated as above. So for the first
+ * request we will hit the buddy cache which will result in this prealloc
+ * space getting filled. The prealloc space is then later used for the
+ * subsequent request.
+ */
+
+/*
+ * mballoc operates on the following data:
+ *  - on-disk bitmap
+ *  - in-core buddy (actually includes buddy and bitmap)
+ *  - preallocation descriptors (PAs)
+ *
+ * there are two types of preallocations:
+ *  - inode
+ *    assiged to specific inode and can be used for this inode only.
+ *    it describes part of inode's space preallocated to specific
+ *    physical blocks. any block from that preallocated can be used
+ *    independent. the descriptor just tracks number of blocks left
+ *    unused. so, before taking some block from descriptor, one must
+ *    make sure corresponded logical block isn't allocated yet. this
+ *    also means that freeing any block within descriptor's range
+ *    must discard all preallocated blocks.
+ *  - locality group
+ *    assigned to specific locality group which does not translate to
+ *    permanent set of inodes: inode can join and leave group. space
+ *    from this type of preallocation can be used for any inode. thus
+ *    it's consumed from the beginning to the end.
+ *
+ * relation between them can be expressed as:
+ *    in-core buddy = on-disk bitmap + preallocation descriptors
+ *
+ * this mean blocks mballoc considers used are:
+ *  - allocated blocks (persistent)
+ *  - preallocated blocks (non-persistent)
+ *
+ * consistency in mballoc world means that at any time a block is either
+ * free or used in ALL structures. notice: "any time" should not be read
+ * literally -- time is discrete and delimited by locks.
+ *
+ *  to keep it simple, we don't use block numbers, instead we count number of
+ *  blocks: how many blocks marked used/free in on-disk bitmap, buddy and PA.
+ *
+ * all operations can be expressed as:
+ *  - init buddy:			buddy = on-disk + PAs
+ *  - new PA:				buddy += N; PA = N
+ *  - use inode PA:			on-disk += N; PA -= N
+ *  - discard inode PA			buddy -= on-disk - PA; PA = 0
+ *  - use locality group PA		on-disk += N; PA -= N
+ *  - discard locality group PA		buddy -= PA; PA = 0
+ *  note: 'buddy -= on-disk - PA' is used to show that on-disk bitmap
+ *        is used in real operation because we can't know actual used
+ *        bits from PA, only from on-disk bitmap
+ *
+ * if we follow this strict logic, then all operations above should be atomic.
+ * given some of them can block, we'd have to use something like semaphores
+ * killing performance on high-end SMP hardware. let's try to relax it using
+ * the following knowledge:
+ *  1) if buddy is referenced, it's already initialized
+ *  2) while block is used in buddy and the buddy is referenced,
+ *     nobody can re-allocate that block
+ *  3) we work on bitmaps and '+' actually means 'set bits'. if on-disk has
+ *     bit set and PA claims same block, it's OK. IOW, one can set bit in
+ *     on-disk bitmap if buddy has same bit set or/and PA covers corresponded
+ *     block
+ *
+ * so, now we're building a concurrency table:
+ *  - init buddy vs.
+ *    - new PA
+ *      blocks for PA are allocated in the buddy, buddy must be referenced
+ *      until PA is linked to allocation group to avoid concurrent buddy init
+ *    - use inode PA
+ *      we need to make sure that either on-disk bitmap or PA has uptodate data
+ *      given (3) we care that PA-=N operation doesn't interfere with init
+ *    - discard inode PA
+ *      the simplest way would be to have buddy initialized by the discard
+ *    - use locality group PA
+ *      again PA-=N must be serialized with init
+ *    - discard locality group PA
+ *      the simplest way would be to have buddy initialized by the discard
+ *  - new PA vs.
+ *    - use inode PA
+ *      i_data_sem serializes them
+ *    - discard inode PA
+ *      discard process must wait until PA isn't used by another process
+ *    - use locality group PA
+ *      some mutex should serialize them
+ *    - discard locality group PA
+ *      discard process must wait until PA isn't used by another process
+ *  - use inode PA
+ *    - use inode PA
+ *      i_data_sem or another mutex should serializes them
+ *    - discard inode PA
+ *      discard process must wait until PA isn't used by another process
+ *    - use locality group PA
+ *      nothing wrong here -- they're different PAs covering different blocks
+ *    - discard locality group PA
+ *      discard process must wait until PA isn't used by another process
+ *
+ * now we're ready to make few consequences:
+ *  - PA is referenced and while it is no discard is possible
+ *  - PA is referenced until block isn't marked in on-disk bitmap
+ *  - PA changes only after on-disk bitmap
+ *  - discard must not compete with init. either init is done before
+ *    any discard or they're serialized somehow
+ *  - buddy init as sum of on-disk bitmap and PAs is done atomically
+ *
+ * a special case when we've used PA to emptiness. no need to modify buddy
+ * in this case, but we should care about concurrent init
+ *
+ */
+
+ /*
+ * Logic in few words:
+ *
+ *  - allocation:
+ *    load group
+ *    find blocks
+ *    mark bits in on-disk bitmap
+ *    release group
+ *
+ *  - use preallocation:
+ *    find proper PA (per-inode or group)
+ *    load group
+ *    mark bits in on-disk bitmap
+ *    release group
+ *    release PA
+ *
+ *  - free:
+ *    load group
+ *    mark bits in on-disk bitmap
+ *    release group
+ *
+ *  - discard preallocations in group:
+ *    mark PAs deleted
+ *    move them onto local list
+ *    load on-disk bitmap
+ *    load group
+ *    remove PA from object (inode or locality group)
+ *    mark free blocks in-core
+ *
+ *  - discard inode's preallocations:
+ */
+
+/*
+ * Locking rules
+ *
+ * Locks:
+ *  - bitlock on a group	(group)
+ *  - object (inode/locality)	(object)
+ *  - per-pa lock		(pa)
+ *
+ * Paths:
+ *  - new pa
+ *    object
+ *    group
+ *
+ *  - find and use pa:
+ *    pa
+ *
+ *  - release consumed pa:
+ *    pa
+ *    group
+ *    object
+ *
+ *  - generate in-core bitmap:
+ *    group
+ *        pa
+ *
+ *  - discard all for given object (inode, locality group):
+ *    object
+ *        pa
+ *    group
+ *
+ *  - discard all for given group:
+ *    group
+ *        pa
+ *    group
+ *        object
+ *
+ */
+
+/*
+ * with AGGRESSIVE_CHECK allocator runs consistency checks over
+ * structures. these checks slow things down a lot
+ */
+#define AGGRESSIVE_CHECK__
+
+/*
+ * with DOUBLE_CHECK defined mballoc creates persistent in-core
+ * bitmaps, maintains and uses them to check for double allocations
+ */
+#define DOUBLE_CHECK__
+
+/*
+ */
+#define MB_DEBUG__
+#ifdef MB_DEBUG
+#define mb_debug(fmt, a...)	printk(fmt, ##a)
+#else
+#define mb_debug(fmt, a...)
+#endif
+
+/*
+ * with EXT4_MB_HISTORY mballoc stores last N allocations in memory
+ * and you can monitor it in /proc/fs/ext4/<dev>/mb_history
+ */
+#define EXT4_MB_HISTORY
+#define EXT4_MB_HISTORY_ALLOC		1	/* allocation */
+#define EXT4_MB_HISTORY_PREALLOC	2	/* preallocated blocks used */
+#define EXT4_MB_HISTORY_DISCARD		4	/* preallocation discarded */
+#define EXT4_MB_HISTORY_FREE		8	/* free */
+
+#define EXT4_MB_HISTORY_DEFAULT		(EXT4_MB_HISTORY_ALLOC | \
+					 EXT4_MB_HISTORY_PREALLOC)
+
+/*
+ * How long mballoc can look for a best extent (in found extents)
+ */
+#define MB_DEFAULT_MAX_TO_SCAN		200
+
+/*
+ * How long mballoc must look for a best extent
+ */
+#define MB_DEFAULT_MIN_TO_SCAN		10
+
+/*
+ * How many groups mballoc will scan looking for the best chunk
+ */
+#define MB_DEFAULT_MAX_GROUPS_TO_SCAN	5
+
+/*
+ * with 'ext4_mb_stats' allocator will collect stats that will be
+ * shown at umount. The collecting costs though!
+ */
+#define MB_DEFAULT_STATS		1
+
+/*
+ * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served
+ * by the stream allocator, which purpose is to pack requests
+ * as close each to other as possible to produce smooth I/O traffic
+ * We use locality group prealloc space for stream request.
+ * We can tune the same via /proc/fs/ext4/<parition>/stream_req
+ */
+#define MB_DEFAULT_STREAM_THRESHOLD	16	/* 64K */
+
+/*
+ * for which requests use 2^N search using buddies
+ */
+#define MB_DEFAULT_ORDER2_REQS		2
+
+/*
+ * default group prealloc size 512 blocks
+ */
+#define MB_DEFAULT_GROUP_PREALLOC	512
+
+static struct kmem_cache *ext4_pspace_cachep;
+
+#ifdef EXT4_BB_MAX_BLOCKS
+#undef EXT4_BB_MAX_BLOCKS
+#endif
+#define EXT4_BB_MAX_BLOCKS	30
+
+struct ext4_free_metadata {
+	ext4_group_t group;
+	unsigned short num;
+	ext4_grpblk_t  blocks[EXT4_BB_MAX_BLOCKS];
+	struct list_head list;
+};
+
+struct ext4_group_info {
+	unsigned long	bb_state;
+	unsigned long	bb_tid;
+	struct ext4_free_metadata *bb_md_cur;
+	unsigned short	bb_first_free;
+	unsigned short	bb_free;
+	unsigned short	bb_fragments;
+	struct		list_head bb_prealloc_list;
+#ifdef DOUBLE_CHECK
+	void		*bb_bitmap;
+#endif
+	unsigned short	bb_counters[];
+};
+
+#define EXT4_GROUP_INFO_NEED_INIT_BIT	0
+#define EXT4_GROUP_INFO_LOCKED_BIT	1
+
+#define EXT4_MB_GRP_NEED_INIT(grp)	\
+	(test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state)))
+
+
+struct ext4_prealloc_space {
+	struct list_head	pa_inode_list;
+	struct list_head	pa_group_list;
+	union {
+		struct list_head pa_tmp_list;
+		struct rcu_head	pa_rcu;
+	} u;
+	spinlock_t		pa_lock;
+	atomic_t		pa_count;
+	unsigned		pa_deleted;
+	ext4_fsblk_t		pa_pstart;	/* phys. block */
+	ext4_lblk_t		pa_lstart;	/* log. block */
+	unsigned short		pa_len;		/* len of preallocated chunk */
+	unsigned short		pa_free;	/* how many blocks are free */
+	unsigned short		pa_linear;	/* consumed in one direction
+						 * strictly, for grp prealloc */
+	spinlock_t		*pa_obj_lock;
+	struct inode		*pa_inode;	/* hack, for history only */
+};
+
+
+struct ext4_free_extent {
+	ext4_lblk_t fe_logical;
+	ext4_grpblk_t fe_start;
+	ext4_group_t fe_group;
+	int fe_len;
+};
+
+/*
+ * Locality group:
+ *   we try to group all related changes together
+ *   so that writeback can flush/allocate them together as well
+ */
+struct ext4_locality_group {
+	/* for allocator */
+	struct mutex		lg_mutex;	/* to serialize allocates */
+	struct list_head	lg_prealloc_list;/* list of preallocations */
+	spinlock_t		lg_prealloc_lock;
+};
+
+struct ext4_allocation_context {
+	struct inode *ac_inode;
+	struct super_block *ac_sb;
+
+	/* original request */
+	struct ext4_free_extent ac_o_ex;
+
+	/* goal request (after normalization) */
+	struct ext4_free_extent ac_g_ex;
+
+	/* the best found extent */
+	struct ext4_free_extent ac_b_ex;
+
+	/* copy of the bext found extent taken before preallocation efforts */
+	struct ext4_free_extent ac_f_ex;
+
+	/* number of iterations done. we have to track to limit searching */
+	unsigned long ac_ex_scanned;
+	__u16 ac_groups_scanned;
+	__u16 ac_found;
+	__u16 ac_tail;
+	__u16 ac_buddy;
+	__u16 ac_flags;		/* allocation hints */
+	__u8 ac_status;
+	__u8 ac_criteria;
+	__u8 ac_repeats;
+	__u8 ac_2order;		/* if request is to allocate 2^N blocks and
+				 * N > 0, the field stores N, otherwise 0 */
+	__u8 ac_op;		/* operation, for history only */
+	struct page *ac_bitmap_page;
+	struct page *ac_buddy_page;
+	struct ext4_prealloc_space *ac_pa;
+	struct ext4_locality_group *ac_lg;
+};
+
+#define AC_STATUS_CONTINUE	1
+#define AC_STATUS_FOUND		2
+#define AC_STATUS_BREAK		3
+
+struct ext4_mb_history {
+	struct ext4_free_extent orig;	/* orig allocation */
+	struct ext4_free_extent goal;	/* goal allocation */
+	struct ext4_free_extent result;	/* result allocation */
+	unsigned pid;
+	unsigned ino;
+	__u16 found;	/* how many extents have been found */
+	__u16 groups;	/* how many groups have been scanned */
+	__u16 tail;	/* what tail broke some buddy */
+	__u16 buddy;	/* buddy the tail ^^^ broke */
+	__u16 flags;
+	__u8 cr:3;	/* which phase the result extent was found at */
+	__u8 op:4;
+	__u8 merged:1;
+};
+
+struct ext4_buddy {
+	struct page *bd_buddy_page;
+	void *bd_buddy;
+	struct page *bd_bitmap_page;
+	void *bd_bitmap;
+	struct ext4_group_info *bd_info;
+	struct super_block *bd_sb;
+	__u16 bd_blkbits;
+	ext4_group_t bd_group;
+};
+#define EXT4_MB_BITMAP(e4b)	((e4b)->bd_bitmap)
+#define EXT4_MB_BUDDY(e4b)	((e4b)->bd_buddy)
+
+#ifndef EXT4_MB_HISTORY
+static inline void ext4_mb_store_history(struct ext4_allocation_context *ac)
+{
+	return;
+}
+#else
+static void ext4_mb_store_history(struct ext4_allocation_context *ac);
+#endif
+
+#define in_range(b, first, len)	((b) >= (first) && (b) <= (first) + (len) - 1)
+
+static struct proc_dir_entry *proc_root_ext4;
+struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
+ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
+			ext4_fsblk_t goal, unsigned long *count, int *errp);
+
+static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+					ext4_group_t group);
+static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *);
+static void ext4_mb_free_committed_blocks(struct super_block *);
+static void ext4_mb_return_to_preallocation(struct inode *inode,
+					struct ext4_buddy *e4b, sector_t block,
+					int count);
+static void ext4_mb_put_pa(struct ext4_allocation_context *,
+			struct super_block *, struct ext4_prealloc_space *pa);
+static int ext4_mb_init_per_dev_proc(struct super_block *sb);
+static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
+
+
+static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
+}
+
+static inline void ext4_unlock_group(struct super_block *sb,
+					ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
+}
+
+static inline int ext4_is_group_locked(struct super_block *sb,
+					ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT,
+						&(grinfo->bb_state));
+}
+
+static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb,
+					struct ext4_free_extent *fex)
+{
+	ext4_fsblk_t block;
+
+	block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb)
+			+ fex->fe_start
+			+ le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
+	return block;
+}
+
+#if BITS_PER_LONG == 64
+#define mb_correct_addr_and_bit(bit, addr)		\
+{							\
+	bit += ((unsigned long) addr & 7UL) << 3;	\
+	addr = (void *) ((unsigned long) addr & ~7UL);	\
+}
+#elif BITS_PER_LONG == 32
+#define mb_correct_addr_and_bit(bit, addr)		\
+{							\
+	bit += ((unsigned long) addr & 3UL) << 3;	\
+	addr = (void *) ((unsigned long) addr & ~3UL);	\
+}
+#else
+#error "how many bits you are?!"
+#endif
+
+static inline int mb_test_bit(int bit, void *addr)
+{
+	/*
+	 * ext4_test_bit on architecture like powerpc
+	 * needs unsigned long aligned address
+	 */
+	mb_correct_addr_and_bit(bit, addr);
+	return ext4_test_bit(bit, addr);
+}
+
+static inline void mb_set_bit(int bit, void *addr)
+{
+	mb_correct_addr_and_bit(bit, addr);
+	ext4_set_bit(bit, addr);
+}
+
+static inline void mb_set_bit_atomic(spinlock_t *lock, int bit, void *addr)
+{
+	mb_correct_addr_and_bit(bit, addr);
+	ext4_set_bit_atomic(lock, bit, addr);
+}
+
+static inline void mb_clear_bit(int bit, void *addr)
+{
+	mb_correct_addr_and_bit(bit, addr);
+	ext4_clear_bit(bit, addr);
+}
+
+static inline void mb_clear_bit_atomic(spinlock_t *lock, int bit, void *addr)
+{
+	mb_correct_addr_and_bit(bit, addr);
+	ext4_clear_bit_atomic(lock, bit, addr);
+}
+
+static void *mb_find_buddy(struct ext4_buddy *e4b, int order, int *max)
+{
+	char *bb;
+
+	/* FIXME!! is this needed */
+	BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b));
+	BUG_ON(max == NULL);
+
+	if (order > e4b->bd_blkbits + 1) {
+		*max = 0;
+		return NULL;
+	}
+
+	/* at order 0 we see each particular block */
+	*max = 1 << (e4b->bd_blkbits + 3);
+	if (order == 0)
+		return EXT4_MB_BITMAP(e4b);
+
+	bb = EXT4_MB_BUDDY(e4b) + EXT4_SB(e4b->bd_sb)->s_mb_offsets[order];
+	*max = EXT4_SB(e4b->bd_sb)->s_mb_maxs[order];
+
+	return bb;
+}
+
+#ifdef DOUBLE_CHECK
+static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b,
+			   int first, int count)
+{
+	int i;
+	struct super_block *sb = e4b->bd_sb;
+
+	if (unlikely(e4b->bd_info->bb_bitmap == NULL))
+		return;
+	BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group));
+	for (i = 0; i < count; i++) {
+		if (!mb_test_bit(first + i, e4b->bd_info->bb_bitmap)) {
+			ext4_fsblk_t blocknr;
+			blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb);
+			blocknr += first + i;
+			blocknr +=
+			    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
+
+			ext4_error(sb, __FUNCTION__, "double-free of inode"
+				   " %lu's block %llu(bit %u in group %lu)\n",
+				   inode ? inode->i_ino : 0, blocknr,
+				   first + i, e4b->bd_group);
+		}
+		mb_clear_bit(first + i, e4b->bd_info->bb_bitmap);
+	}
+}
+
+static void mb_mark_used_double(struct ext4_buddy *e4b, int first, int count)
+{
+	int i;
+
+	if (unlikely(e4b->bd_info->bb_bitmap == NULL))
+		return;
+	BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group));
+	for (i = 0; i < count; i++) {
+		BUG_ON(mb_test_bit(first + i, e4b->bd_info->bb_bitmap));
+		mb_set_bit(first + i, e4b->bd_info->bb_bitmap);
+	}
+}
+
+static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap)
+{
+	if (memcmp(e4b->bd_info->bb_bitmap, bitmap, e4b->bd_sb->s_blocksize)) {
+		unsigned char *b1, *b2;
+		int i;
+		b1 = (unsigned char *) e4b->bd_info->bb_bitmap;
+		b2 = (unsigned char *) bitmap;
+		for (i = 0; i < e4b->bd_sb->s_blocksize; i++) {
+			if (b1[i] != b2[i]) {
+				printk("corruption in group %lu at byte %u(%u):"
+				       " %x in copy != %x on disk/prealloc\n",
+					e4b->bd_group, i, i * 8, b1[i], b2[i]);
+				BUG();
+			}
+		}
+	}
+}
+
+#else
+static inline void mb_free_blocks_double(struct inode *inode,
+				struct ext4_buddy *e4b, int first, int count)
+{
+	return;
+}
+static inline void mb_mark_used_double(struct ext4_buddy *e4b,
+						int first, int count)
+{
+	return;
+}
+static inline void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap)
+{
+	return;
+}
+#endif
+
+#ifdef AGGRESSIVE_CHECK
+
+#define MB_CHECK_ASSERT(assert)						\
+do {									\
+	if (!(assert)) {						\
+		printk(KERN_EMERG					\
+			"Assertion failure in %s() at %s:%d: \"%s\"\n",	\
+			function, file, line, # assert);		\
+		BUG();							\
+	}								\
+} while (0)
+
+static int __mb_check_buddy(struct ext4_buddy *e4b, char *file,
+				const char *function, int line)
+{
+	struct super_block *sb = e4b->bd_sb;
+	int order = e4b->bd_blkbits + 1;
+	int max;
+	int max2;
+	int i;
+	int j;
+	int k;
+	int count;
+	struct ext4_group_info *grp;
+	int fragments = 0;
+	int fstart;
+	struct list_head *cur;
+	void *buddy;
+	void *buddy2;
+
+	if (!test_opt(sb, MBALLOC))
+		return 0;
+
+	{
+		static int mb_check_counter;
+		if (mb_check_counter++ % 100 != 0)
+			return 0;
+	}
+
+	while (order > 1) {
+		buddy = mb_find_buddy(e4b, order, &max);
+		MB_CHECK_ASSERT(buddy);
+		buddy2 = mb_find_buddy(e4b, order - 1, &max2);
+		MB_CHECK_ASSERT(buddy2);
+		MB_CHECK_ASSERT(buddy != buddy2);
+		MB_CHECK_ASSERT(max * 2 == max2);
+
+		count = 0;
+		for (i = 0; i < max; i++) {
+
+			if (mb_test_bit(i, buddy)) {
+				/* only single bit in buddy2 may be 1 */
+				if (!mb_test_bit(i << 1, buddy2)) {
+					MB_CHECK_ASSERT(
+						mb_test_bit((i<<1)+1, buddy2));
+				} else if (!mb_test_bit((i << 1) + 1, buddy2)) {
+					MB_CHECK_ASSERT(
+						mb_test_bit(i << 1, buddy2));
+				}
+				continue;
+			}
+
+			/* both bits in buddy2 must be 0 */
+			MB_CHECK_ASSERT(mb_test_bit(i << 1, buddy2));
+			MB_CHECK_ASSERT(mb_test_bit((i << 1) + 1, buddy2));
+
+			for (j = 0; j < (1 << order); j++) {
+				k = (i * (1 << order)) + j;
+				MB_CHECK_ASSERT(
+					!mb_test_bit(k, EXT4_MB_BITMAP(e4b)));
+			}
+			count++;
+		}
+		MB_CHECK_ASSERT(e4b->bd_info->bb_counters[order] == count);
+		order--;
+	}
+
+	fstart = -1;
+	buddy = mb_find_buddy(e4b, 0, &max);
+	for (i = 0; i < max; i++) {
+		if (!mb_test_bit(i, buddy)) {
+			MB_CHECK_ASSERT(i >= e4b->bd_info->bb_first_free);
+			if (fstart == -1) {
+				fragments++;
+				fstart = i;
+			}
+			continue;
+		}
+		fstart = -1;
+		/* check used bits only */
+		for (j = 0; j < e4b->bd_blkbits + 1; j++) {
+			buddy2 = mb_find_buddy(e4b, j, &max2);
+			k = i >> j;
+			MB_CHECK_ASSERT(k < max2);
+			MB_CHECK_ASSERT(mb_test_bit(k, buddy2));
+		}
+	}
+	MB_CHECK_ASSERT(!EXT4_MB_GRP_NEED_INIT(e4b->bd_info));
+	MB_CHECK_ASSERT(e4b->bd_info->bb_fragments == fragments);
+
+	grp = ext4_get_group_info(sb, e4b->bd_group);
+	buddy = mb_find_buddy(e4b, 0, &max);
+	list_for_each(cur, &grp->bb_prealloc_list) {
+		ext4_group_t groupnr;
+		struct ext4_prealloc_space *pa;
+		pa = list_entry(cur, struct ext4_prealloc_space, group_list);
+		ext4_get_group_no_and_offset(sb, pa->pstart, &groupnr, &k);
+		MB_CHECK_ASSERT(groupnr == e4b->bd_group);
+		for (i = 0; i < pa->len; i++)
+			MB_CHECK_ASSERT(mb_test_bit(k + i, buddy));
+	}
+	return 0;
+}
+#undef MB_CHECK_ASSERT
+#define mb_check_buddy(e4b) __mb_check_buddy(e4b,	\
+					__FILE__, __FUNCTION__, __LINE__)
+#else
+#define mb_check_buddy(e4b)
+#endif
+
+/* FIXME!! need more doc */
+static void ext4_mb_mark_free_simple(struct super_block *sb,
+				void *buddy, unsigned first, int len,
+					struct ext4_group_info *grp)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	unsigned short min;
+	unsigned short max;
+	unsigned short chunk;
+	unsigned short border;
+
+	BUG_ON(len >= EXT4_BLOCKS_PER_GROUP(sb));
+
+	border = 2 << sb->s_blocksize_bits;
+
+	while (len > 0) {
+		/* find how many blocks can be covered since this position */
+		max = ffs(first | border) - 1;
+
+		/* find how many blocks of power 2 we need to mark */
+		min = fls(len) - 1;
+
+		if (max < min)
+			min = max;
+		chunk = 1 << min;
+
+		/* mark multiblock chunks only */
+		grp->bb_counters[min]++;
+		if (min > 0)
+			mb_clear_bit(first >> min,
+				     buddy + sbi->s_mb_offsets[min]);
+
+		len -= chunk;
+		first += chunk;
+	}
+}
+
+static void ext4_mb_generate_buddy(struct super_block *sb,
+				void *buddy, void *bitmap, ext4_group_t group)
+{
+	struct ext4_group_info *grp = ext4_get_group_info(sb, group);
+	unsigned short max = EXT4_BLOCKS_PER_GROUP(sb);
+	unsigned short i = 0;
+	unsigned short first;
+	unsigned short len;
+	unsigned free = 0;
+	unsigned fragments = 0;
+	unsigned long long period = get_cycles();
+
+	/* initialize buddy from bitmap which is aggregation
+	 * of on-disk bitmap and preallocations */
+	i = ext4_find_next_zero_bit(bitmap, max, 0);
+	grp->bb_first_free = i;
+	while (i < max) {
+		fragments++;
+		first = i;
+		i = ext4_find_next_bit(bitmap, max, i);
+		len = i - first;
+		free += len;
+		if (len > 1)
+			ext4_mb_mark_free_simple(sb, buddy, first, len, grp);
+		else
+			grp->bb_counters[0]++;
+		if (i < max)
+			i = ext4_find_next_zero_bit(bitmap, max, i);
+	}
+	grp->bb_fragments = fragments;
+
+	if (free != grp->bb_free) {
+		printk(KERN_DEBUG
+			"EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n",
+			group, free, grp->bb_free);
+		grp->bb_free = free;
+	}
+
+	clear_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state));
+
+	period = get_cycles() - period;
+	spin_lock(&EXT4_SB(sb)->s_bal_lock);
+	EXT4_SB(sb)->s_mb_buddies_generated++;
+	EXT4_SB(sb)->s_mb_generation_time += period;
+	spin_unlock(&EXT4_SB(sb)->s_bal_lock);
+}
+
+/* The buddy information is attached the buddy cache inode
+ * for convenience. The information regarding each group
+ * is loaded via ext4_mb_load_buddy. The information involve
+ * block bitmap and buddy information. The information are
+ * stored in the inode as
+ *
+ * {                        page                        }
+ * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]...
+ *
+ *
+ * one block each for bitmap and buddy information.
+ * So for each group we take up 2 blocks. A page can
+ * contain blocks_per_page (PAGE_CACHE_SIZE / blocksize)  blocks.
+ * So it can have information regarding groups_per_page which
+ * is blocks_per_page/2
+ */
+
+static int ext4_mb_init_cache(struct page *page, char *incore)
+{
+	int blocksize;
+	int blocks_per_page;
+	int groups_per_page;
+	int err = 0;
+	int i;
+	ext4_group_t first_group;
+	int first_block;
+	struct super_block *sb;
+	struct buffer_head *bhs;
+	struct buffer_head **bh;
+	struct inode *inode;
+	char *data;
+	char *bitmap;
+
+	mb_debug("init page %lu\n", page->index);
+
+	inode = page->mapping->host;
+	sb = inode->i_sb;
+	blocksize = 1 << inode->i_blkbits;
+	blocks_per_page = PAGE_CACHE_SIZE / blocksize;
+
+	groups_per_page = blocks_per_page >> 1;
+	if (groups_per_page == 0)
+		groups_per_page = 1;
+
+	/* allocate buffer_heads to read bitmaps */
+	if (groups_per_page > 1) {
+		err = -ENOMEM;
+		i = sizeof(struct buffer_head *) * groups_per_page;
+		bh = kzalloc(i, GFP_NOFS);
+		if (bh == NULL)
+			goto out;
+	} else
+		bh = &bhs;
+
+	first_group = page->index * blocks_per_page / 2;
+
+	/* read all groups the page covers into the cache */
+	for (i = 0; i < groups_per_page; i++) {
+		struct ext4_group_desc *desc;
+
+		if (first_group + i >= EXT4_SB(sb)->s_groups_count)
+			break;
+
+		err = -EIO;
+		desc = ext4_get_group_desc(sb, first_group + i, NULL);
+		if (desc == NULL)
+			goto out;
+
+		err = -ENOMEM;
+		bh[i] = sb_getblk(sb, ext4_block_bitmap(sb, desc));
+		if (bh[i] == NULL)
+			goto out;
+
+		if (bh_uptodate_or_lock(bh[i]))
+			continue;
+
+		if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
+			ext4_init_block_bitmap(sb, bh[i],
+						first_group + i, desc);
+			set_buffer_uptodate(bh[i]);
+			unlock_buffer(bh[i]);
+			continue;
+		}
+		get_bh(bh[i]);
+		bh[i]->b_end_io = end_buffer_read_sync;
+		submit_bh(READ, bh[i]);
+		mb_debug("read bitmap for group %lu\n", first_group + i);
+	}
+
+	/* wait for I/O completion */
+	for (i = 0; i < groups_per_page && bh[i]; i++)
+		wait_on_buffer(bh[i]);
+
+	err = -EIO;
+	for (i = 0; i < groups_per_page && bh[i]; i++)
+		if (!buffer_uptodate(bh[i]))
+			goto out;
+
+	first_block = page->index * blocks_per_page;
+	for (i = 0; i < blocks_per_page; i++) {
+		int group;
+		struct ext4_group_info *grinfo;
+
+		group = (first_block + i) >> 1;
+		if (group >= EXT4_SB(sb)->s_groups_count)
+			break;
+
+		/*
+		 * data carry information regarding this
+		 * particular group in the format specified
+		 * above
+		 *
+		 */
+		data = page_address(page) + (i * blocksize);
+		bitmap = bh[group - first_group]->b_data;
+
+		/*
+		 * We place the buddy block and bitmap block
+		 * close together
+		 */
+		if ((first_block + i) & 1) {
+			/* this is block of buddy */
+			BUG_ON(incore == NULL);
+			mb_debug("put buddy for group %u in page %lu/%x\n",
+				group, page->index, i * blocksize);
+			memset(data, 0xff, blocksize);
+			grinfo = ext4_get_group_info(sb, group);
+			grinfo->bb_fragments = 0;
+			memset(grinfo->bb_counters, 0,
+			       sizeof(unsigned short)*(sb->s_blocksize_bits+2));
+			/*
+			 * incore got set to the group block bitmap below
+			 */
+			ext4_mb_generate_buddy(sb, data, incore, group);
+			incore = NULL;
+		} else {
+			/* this is block of bitmap */
+			BUG_ON(incore != NULL);
+			mb_debug("put bitmap for group %u in page %lu/%x\n",
+				group, page->index, i * blocksize);
+
+			/* see comments in ext4_mb_put_pa() */
+			ext4_lock_group(sb, group);
+			memcpy(data, bitmap, blocksize);
+
+			/* mark all preallocated blks used in in-core bitmap */
+			ext4_mb_generate_from_pa(sb, data, group);
+			ext4_unlock_group(sb, group);
+
+			/* set incore so that the buddy information can be
+			 * generated using this
+			 */
+			incore = data;
+		}
+	}
+	SetPageUptodate(page);
+
+out:
+	if (bh) {
+		for (i = 0; i < groups_per_page && bh[i]; i++)
+			brelse(bh[i]);
+		if (bh != &bhs)
+			kfree(bh);
+	}
+	return err;
+}
+
+static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
+		struct ext4_buddy *e4b)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct inode *inode = sbi->s_buddy_cache;
+	int blocks_per_page;
+	int block;
+	int pnum;
+	int poff;
+	struct page *page;
+
+	mb_debug("load group %lu\n", group);
+
+	blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize;
+
+	e4b->bd_blkbits = sb->s_blocksize_bits;
+	e4b->bd_info = ext4_get_group_info(sb, group);
+	e4b->bd_sb = sb;
+	e4b->bd_group = group;
+	e4b->bd_buddy_page = NULL;
+	e4b->bd_bitmap_page = NULL;
+
+	/*
+	 * the buddy cache inode stores the block bitmap
+	 * and buddy information in consecutive blocks.
+	 * So for each group we need two blocks.
+	 */
+	block = group * 2;
+	pnum = block / blocks_per_page;
+	poff = block % blocks_per_page;
+
+	/* we could use find_or_create_page(), but it locks page
+	 * what we'd like to avoid in fast path ... */
+	page = find_get_page(inode->i_mapping, pnum);
+	if (page == NULL || !PageUptodate(page)) {
+		if (page)
+			page_cache_release(page);
+		page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+		if (page) {
+			BUG_ON(page->mapping != inode->i_mapping);
+			if (!PageUptodate(page)) {
+				ext4_mb_init_cache(page, NULL);
+				mb_cmp_bitmaps(e4b, page_address(page) +
+					       (poff * sb->s_blocksize));
+			}
+			unlock_page(page);
+		}
+	}
+	if (page == NULL || !PageUptodate(page))
+		goto err;
+	e4b->bd_bitmap_page = page;
+	e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize);
+	mark_page_accessed(page);
+
+	block++;
+	pnum = block / blocks_per_page;
+	poff = block % blocks_per_page;
+
+	page = find_get_page(inode->i_mapping, pnum);
+	if (page == NULL || !PageUptodate(page)) {
+		if (page)
+			page_cache_release(page);
+		page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+		if (page) {
+			BUG_ON(page->mapping != inode->i_mapping);
+			if (!PageUptodate(page))
+				ext4_mb_init_cache(page, e4b->bd_bitmap);
+
+			unlock_page(page);
+		}
+	}
+	if (page == NULL || !PageUptodate(page))
+		goto err;
+	e4b->bd_buddy_page = page;
+	e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize);
+	mark_page_accessed(page);
+
+	BUG_ON(e4b->bd_bitmap_page == NULL);
+	BUG_ON(e4b->bd_buddy_page == NULL);
+
+	return 0;
+
+err:
+	if (e4b->bd_bitmap_page)
+		page_cache_release(e4b->bd_bitmap_page);
+	if (e4b->bd_buddy_page)
+		page_cache_release(e4b->bd_buddy_page);
+	e4b->bd_buddy = NULL;
+	e4b->bd_bitmap = NULL;
+	return -EIO;
+}
+
+static void ext4_mb_release_desc(struct ext4_buddy *e4b)
+{
+	if (e4b->bd_bitmap_page)
+		page_cache_release(e4b->bd_bitmap_page);
+	if (e4b->bd_buddy_page)
+		page_cache_release(e4b->bd_buddy_page);
+}
+
+
+static int mb_find_order_for_block(struct ext4_buddy *e4b, int block)
+{
+	int order = 1;
+	void *bb;
+
+	BUG_ON(EXT4_MB_BITMAP(e4b) == EXT4_MB_BUDDY(e4b));
+	BUG_ON(block >= (1 << (e4b->bd_blkbits + 3)));
+
+	bb = EXT4_MB_BUDDY(e4b);
+	while (order <= e4b->bd_blkbits + 1) {
+		block = block >> 1;
+		if (!mb_test_bit(block, bb)) {
+			/* this block is part of buddy of order 'order' */
+			return order;
+		}
+		bb += 1 << (e4b->bd_blkbits - order);
+		order++;
+	}
+	return 0;
+}
+
+static void mb_clear_bits(spinlock_t *lock, void *bm, int cur, int len)
+{
+	__u32 *addr;
+
+	len = cur + len;
+	while (cur < len) {
+		if ((cur & 31) == 0 && (len - cur) >= 32) {
+			/* fast path: clear whole word at once */
+			addr = bm + (cur >> 3);
+			*addr = 0;
+			cur += 32;
+			continue;
+		}
+		mb_clear_bit_atomic(lock, cur, bm);
+		cur++;
+	}
+}
+
+static void mb_set_bits(spinlock_t *lock, void *bm, int cur, int len)
+{
+	__u32 *addr;
+
+	len = cur + len;
+	while (cur < len) {
+		if ((cur & 31) == 0 && (len - cur) >= 32) {
+			/* fast path: set whole word at once */
+			addr = bm + (cur >> 3);
+			*addr = 0xffffffff;
+			cur += 32;
+			continue;
+		}
+		mb_set_bit_atomic(lock, cur, bm);
+		cur++;
+	}
+}
+
+static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
+			  int first, int count)
+{
+	int block = 0;
+	int max = 0;
+	int order;
+	void *buddy;
+	void *buddy2;
+	struct super_block *sb = e4b->bd_sb;
+
+	BUG_ON(first + count > (sb->s_blocksize << 3));
+	BUG_ON(!ext4_is_group_locked(sb, e4b->bd_group));
+	mb_check_buddy(e4b);
+	mb_free_blocks_double(inode, e4b, first, count);
+
+	e4b->bd_info->bb_free += count;
+	if (first < e4b->bd_info->bb_first_free)
+		e4b->bd_info->bb_first_free = first;
+
+	/* let's maintain fragments counter */
+	if (first != 0)
+		block = !mb_test_bit(first - 1, EXT4_MB_BITMAP(e4b));
+	if (first + count < EXT4_SB(sb)->s_mb_maxs[0])
+		max = !mb_test_bit(first + count, EXT4_MB_BITMAP(e4b));
+	if (block && max)
+		e4b->bd_info->bb_fragments--;
+	else if (!block && !max)
+		e4b->bd_info->bb_fragments++;
+
+	/* let's maintain buddy itself */
+	while (count-- > 0) {
+		block = first++;
+		order = 0;
+
+		if (!mb_test_bit(block, EXT4_MB_BITMAP(e4b))) {
+			ext4_fsblk_t blocknr;
+			blocknr = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb);
+			blocknr += block;
+			blocknr +=
+			    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
+
+			ext4_error(sb, __FUNCTION__, "double-free of inode"
+				   " %lu's block %llu(bit %u in group %lu)\n",
+				   inode ? inode->i_ino : 0, blocknr, block,
+				   e4b->bd_group);
+		}
+		mb_clear_bit(block, EXT4_MB_BITMAP(e4b));
+		e4b->bd_info->bb_counters[order]++;
+
+		/* start of the buddy */
+		buddy = mb_find_buddy(e4b, order, &max);
+
+		do {
+			block &= ~1UL;
+			if (mb_test_bit(block, buddy) ||
+					mb_test_bit(block + 1, buddy))
+				break;
+
+			/* both the buddies are free, try to coalesce them */
+			buddy2 = mb_find_buddy(e4b, order + 1, &max);
+
+			if (!buddy2)
+				break;
+
+			if (order > 0) {
+				/* for special purposes, we don't set
+				 * free bits in bitmap */
+				mb_set_bit(block, buddy);
+				mb_set_bit(block + 1, buddy);
+			}
+			e4b->bd_info->bb_counters[order]--;
+			e4b->bd_info->bb_counters[order]--;
+
+			block = block >> 1;
+			order++;
+			e4b->bd_info->bb_counters[order]++;
+
+			mb_clear_bit(block, buddy2);
+			buddy = buddy2;
+		} while (1);
+	}
+	mb_check_buddy(e4b);
+
+	return 0;
+}
+
+static int mb_find_extent(struct ext4_buddy *e4b, int order, int block,
+				int needed, struct ext4_free_extent *ex)
+{
+	int next = block;
+	int max;
+	int ord;
+	void *buddy;
+
+	BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group));
+	BUG_ON(ex == NULL);
+
+	buddy = mb_find_buddy(e4b, order, &max);
+	BUG_ON(buddy == NULL);
+	BUG_ON(block >= max);
+	if (mb_test_bit(block, buddy)) {
+		ex->fe_len = 0;
+		ex->fe_start = 0;
+		ex->fe_group = 0;
+		return 0;
+	}
+
+	/* FIXME dorp order completely ? */
+	if (likely(order == 0)) {
+		/* find actual order */
+		order = mb_find_order_for_block(e4b, block);
+		block = block >> order;
+	}
+
+	ex->fe_len = 1 << order;
+	ex->fe_start = block << order;
+	ex->fe_group = e4b->bd_group;
+
+	/* calc difference from given start */
+	next = next - ex->fe_start;
+	ex->fe_len -= next;
+	ex->fe_start += next;
+
+	while (needed > ex->fe_len &&
+	       (buddy = mb_find_buddy(e4b, order, &max))) {
+
+		if (block + 1 >= max)
+			break;
+
+		next = (block + 1) * (1 << order);
+		if (mb_test_bit(next, EXT4_MB_BITMAP(e4b)))
+			break;
+
+		ord = mb_find_order_for_block(e4b, next);
+
+		order = ord;
+		block = next >> order;
+		ex->fe_len += 1 << order;
+	}
+
+	BUG_ON(ex->fe_start + ex->fe_len > (1 << (e4b->bd_blkbits + 3)));
+	return ex->fe_len;
+}
+
+static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex)
+{
+	int ord;
+	int mlen = 0;
+	int max = 0;
+	int cur;
+	int start = ex->fe_start;
+	int len = ex->fe_len;
+	unsigned ret = 0;
+	int len0 = len;
+	void *buddy;
+
+	BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3));
+	BUG_ON(e4b->bd_group != ex->fe_group);
+	BUG_ON(!ext4_is_group_locked(e4b->bd_sb, e4b->bd_group));
+	mb_check_buddy(e4b);
+	mb_mark_used_double(e4b, start, len);
+
+	e4b->bd_info->bb_free -= len;
+	if (e4b->bd_info->bb_first_free == start)
+		e4b->bd_info->bb_first_free += len;
+
+	/* let's maintain fragments counter */
+	if (start != 0)
+		mlen = !mb_test_bit(start - 1, EXT4_MB_BITMAP(e4b));
+	if (start + len < EXT4_SB(e4b->bd_sb)->s_mb_maxs[0])
+		max = !mb_test_bit(start + len, EXT4_MB_BITMAP(e4b));
+	if (mlen && max)
+		e4b->bd_info->bb_fragments++;
+	else if (!mlen && !max)
+		e4b->bd_info->bb_fragments--;
+
+	/* let's maintain buddy itself */
+	while (len) {
+		ord = mb_find_order_for_block(e4b, start);
+
+		if (((start >> ord) << ord) == start && len >= (1 << ord)) {
+			/* the whole chunk may be allocated at once! */
+			mlen = 1 << ord;
+			buddy = mb_find_buddy(e4b, ord, &max);
+			BUG_ON((start >> ord) >= max);
+			mb_set_bit(start >> ord, buddy);
+			e4b->bd_info->bb_counters[ord]--;
+			start += mlen;
+			len -= mlen;
+			BUG_ON(len < 0);
+			continue;
+		}
+
+		/* store for history */
+		if (ret == 0)
+			ret = len | (ord << 16);
+
+		/* we have to split large buddy */
+		BUG_ON(ord <= 0);
+		buddy = mb_find_buddy(e4b, ord, &max);
+		mb_set_bit(start >> ord, buddy);
+		e4b->bd_info->bb_counters[ord]--;
+
+		ord--;
+		cur = (start >> ord) & ~1U;
+		buddy = mb_find_buddy(e4b, ord, &max);
+		mb_clear_bit(cur, buddy);
+		mb_clear_bit(cur + 1, buddy);
+		e4b->bd_info->bb_counters[ord]++;
+		e4b->bd_info->bb_counters[ord]++;
+	}
+
+	mb_set_bits(sb_bgl_lock(EXT4_SB(e4b->bd_sb), ex->fe_group),
+			EXT4_MB_BITMAP(e4b), ex->fe_start, len0);
+	mb_check_buddy(e4b);
+
+	return ret;
+}
+
+/*
+ * Must be called under group lock!
+ */
+static void ext4_mb_use_best_found(struct ext4_allocation_context *ac,
+					struct ext4_buddy *e4b)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+	int ret;
+
+	BUG_ON(ac->ac_b_ex.fe_group != e4b->bd_group);
+	BUG_ON(ac->ac_status == AC_STATUS_FOUND);
+
+	ac->ac_b_ex.fe_len = min(ac->ac_b_ex.fe_len, ac->ac_g_ex.fe_len);
+	ac->ac_b_ex.fe_logical = ac->ac_g_ex.fe_logical;
+	ret = mb_mark_used(e4b, &ac->ac_b_ex);
+
+	/* preallocation can change ac_b_ex, thus we store actually
+	 * allocated blocks for history */
+	ac->ac_f_ex = ac->ac_b_ex;
+
+	ac->ac_status = AC_STATUS_FOUND;
+	ac->ac_tail = ret & 0xffff;
+	ac->ac_buddy = ret >> 16;
+
+	/* XXXXXXX: SUCH A HORRIBLE **CK */
+	/*FIXME!! Why ? */
+	ac->ac_bitmap_page = e4b->bd_bitmap_page;
+	get_page(ac->ac_bitmap_page);
+	ac->ac_buddy_page = e4b->bd_buddy_page;
+	get_page(ac->ac_buddy_page);
+
+	/* store last allocated for subsequent stream allocation */
+	if ((ac->ac_flags & EXT4_MB_HINT_DATA)) {
+		spin_lock(&sbi->s_md_lock);
+		sbi->s_mb_last_group = ac->ac_f_ex.fe_group;
+		sbi->s_mb_last_start = ac->ac_f_ex.fe_start;
+		spin_unlock(&sbi->s_md_lock);
+	}
+}
+
+/*
+ * regular allocator, for general purposes allocation
+ */
+
+static void ext4_mb_check_limits(struct ext4_allocation_context *ac,
+					struct ext4_buddy *e4b,
+					int finish_group)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+	struct ext4_free_extent *bex = &ac->ac_b_ex;
+	struct ext4_free_extent *gex = &ac->ac_g_ex;
+	struct ext4_free_extent ex;
+	int max;
+
+	/*
+	 * We don't want to scan for a whole year
+	 */
+	if (ac->ac_found > sbi->s_mb_max_to_scan &&
+			!(ac->ac_flags & EXT4_MB_HINT_FIRST)) {
+		ac->ac_status = AC_STATUS_BREAK;
+		return;
+	}
+
+	/*
+	 * Haven't found good chunk so far, let's continue
+	 */
+	if (bex->fe_len < gex->fe_len)
+		return;
+
+	if ((finish_group || ac->ac_found > sbi->s_mb_min_to_scan)
+			&& bex->fe_group == e4b->bd_group) {
+		/* recheck chunk's availability - we don't know
+		 * when it was found (within this lock-unlock
+		 * period or not) */
+		max = mb_find_extent(e4b, 0, bex->fe_start, gex->fe_len, &ex);
+		if (max >= gex->fe_len) {
+			ext4_mb_use_best_found(ac, e4b);
+			return;
+		}
+	}
+}
+
+/*
+ * The routine checks whether found extent is good enough. If it is,
+ * then the extent gets marked used and flag is set to the context
+ * to stop scanning. Otherwise, the extent is compared with the
+ * previous found extent and if new one is better, then it's stored
+ * in the context. Later, the best found extent will be used, if
+ * mballoc can't find good enough extent.
+ *
+ * FIXME: real allocation policy is to be designed yet!
+ */
+static void ext4_mb_measure_extent(struct ext4_allocation_context *ac,
+					struct ext4_free_extent *ex,
+					struct ext4_buddy *e4b)
+{
+	struct ext4_free_extent *bex = &ac->ac_b_ex;
+	struct ext4_free_extent *gex = &ac->ac_g_ex;
+
+	BUG_ON(ex->fe_len <= 0);
+	BUG_ON(ex->fe_len >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
+	BUG_ON(ex->fe_start >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
+	BUG_ON(ac->ac_status != AC_STATUS_CONTINUE);
+
+	ac->ac_found++;
+
+	/*
+	 * The special case - take what you catch first
+	 */
+	if (unlikely(ac->ac_flags & EXT4_MB_HINT_FIRST)) {
+		*bex = *ex;
+		ext4_mb_use_best_found(ac, e4b);
+		return;
+	}
+
+	/*
+	 * Let's check whether the chuck is good enough
+	 */
+	if (ex->fe_len == gex->fe_len) {
+		*bex = *ex;
+		ext4_mb_use_best_found(ac, e4b);
+		return;
+	}
+
+	/*
+	 * If this is first found extent, just store it in the context
+	 */
+	if (bex->fe_len == 0) {
+		*bex = *ex;
+		return;
+	}
+
+	/*
+	 * If new found extent is better, store it in the context
+	 */
+	if (bex->fe_len < gex->fe_len) {
+		/* if the request isn't satisfied, any found extent
+		 * larger than previous best one is better */
+		if (ex->fe_len > bex->fe_len)
+			*bex = *ex;
+	} else if (ex->fe_len > gex->fe_len) {
+		/* if the request is satisfied, then we try to find
+		 * an extent that still satisfy the request, but is
+		 * smaller than previous one */
+		if (ex->fe_len < bex->fe_len)
+			*bex = *ex;
+	}
+
+	ext4_mb_check_limits(ac, e4b, 0);
+}
+
+static int ext4_mb_try_best_found(struct ext4_allocation_context *ac,
+					struct ext4_buddy *e4b)
+{
+	struct ext4_free_extent ex = ac->ac_b_ex;
+	ext4_group_t group = ex.fe_group;
+	int max;
+	int err;
+
+	BUG_ON(ex.fe_len <= 0);
+	err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
+	if (err)
+		return err;
+
+	ext4_lock_group(ac->ac_sb, group);
+	max = mb_find_extent(e4b, 0, ex.fe_start, ex.fe_len, &ex);
+
+	if (max > 0) {
+		ac->ac_b_ex = ex;
+		ext4_mb_use_best_found(ac, e4b);
+	}
+
+	ext4_unlock_group(ac->ac_sb, group);
+	ext4_mb_release_desc(e4b);
+
+	return 0;
+}
+
+static int ext4_mb_find_by_goal(struct ext4_allocation_context *ac,
+				struct ext4_buddy *e4b)
+{
+	ext4_group_t group = ac->ac_g_ex.fe_group;
+	int max;
+	int err;
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+	struct ext4_super_block *es = sbi->s_es;
+	struct ext4_free_extent ex;
+
+	if (!(ac->ac_flags & EXT4_MB_HINT_TRY_GOAL))
+		return 0;
+
+	err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
+	if (err)
+		return err;
+
+	ext4_lock_group(ac->ac_sb, group);
+	max = mb_find_extent(e4b, 0, ac->ac_g_ex.fe_start,
+			     ac->ac_g_ex.fe_len, &ex);
+
+	if (max >= ac->ac_g_ex.fe_len && ac->ac_g_ex.fe_len == sbi->s_stripe) {
+		ext4_fsblk_t start;
+
+		start = (e4b->bd_group * EXT4_BLOCKS_PER_GROUP(ac->ac_sb)) +
+			ex.fe_start + le32_to_cpu(es->s_first_data_block);
+		/* use do_div to get remainder (would be 64-bit modulo) */
+		if (do_div(start, sbi->s_stripe) == 0) {
+			ac->ac_found++;
+			ac->ac_b_ex = ex;
+			ext4_mb_use_best_found(ac, e4b);
+		}
+	} else if (max >= ac->ac_g_ex.fe_len) {
+		BUG_ON(ex.fe_len <= 0);
+		BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group);
+		BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start);
+		ac->ac_found++;
+		ac->ac_b_ex = ex;
+		ext4_mb_use_best_found(ac, e4b);
+	} else if (max > 0 && (ac->ac_flags & EXT4_MB_HINT_MERGE)) {
+		/* Sometimes, caller may want to merge even small
+		 * number of blocks to an existing extent */
+		BUG_ON(ex.fe_len <= 0);
+		BUG_ON(ex.fe_group != ac->ac_g_ex.fe_group);
+		BUG_ON(ex.fe_start != ac->ac_g_ex.fe_start);
+		ac->ac_found++;
+		ac->ac_b_ex = ex;
+		ext4_mb_use_best_found(ac, e4b);
+	}
+	ext4_unlock_group(ac->ac_sb, group);
+	ext4_mb_release_desc(e4b);
+
+	return 0;
+}
+
+/*
+ * The routine scans buddy structures (not bitmap!) from given order
+ * to max order and tries to find big enough chunk to satisfy the req
+ */
+static void ext4_mb_simple_scan_group(struct ext4_allocation_context *ac,
+					struct ext4_buddy *e4b)
+{
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_group_info *grp = e4b->bd_info;
+	void *buddy;
+	int i;
+	int k;
+	int max;
+
+	BUG_ON(ac->ac_2order <= 0);
+	for (i = ac->ac_2order; i <= sb->s_blocksize_bits + 1; i++) {
+		if (grp->bb_counters[i] == 0)
+			continue;
+
+		buddy = mb_find_buddy(e4b, i, &max);
+		BUG_ON(buddy == NULL);
+
+		k = ext4_find_next_zero_bit(buddy, max, 0);
+		BUG_ON(k >= max);
+
+		ac->ac_found++;
+
+		ac->ac_b_ex.fe_len = 1 << i;
+		ac->ac_b_ex.fe_start = k << i;
+		ac->ac_b_ex.fe_group = e4b->bd_group;
+
+		ext4_mb_use_best_found(ac, e4b);
+
+		BUG_ON(ac->ac_b_ex.fe_len != ac->ac_g_ex.fe_len);
+
+		if (EXT4_SB(sb)->s_mb_stats)
+			atomic_inc(&EXT4_SB(sb)->s_bal_2orders);
+
+		break;
+	}
+}
+
+/*
+ * The routine scans the group and measures all found extents.
+ * In order to optimize scanning, caller must pass number of
+ * free blocks in the group, so the routine can know upper limit.
+ */
+static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
+					struct ext4_buddy *e4b)
+{
+	struct super_block *sb = ac->ac_sb;
+	void *bitmap = EXT4_MB_BITMAP(e4b);
+	struct ext4_free_extent ex;
+	int i;
+	int free;
+
+	free = e4b->bd_info->bb_free;
+	BUG_ON(free <= 0);
+
+	i = e4b->bd_info->bb_first_free;
+
+	while (free && ac->ac_status == AC_STATUS_CONTINUE) {
+		i = ext4_find_next_zero_bit(bitmap,
+						EXT4_BLOCKS_PER_GROUP(sb), i);
+		if (i >= EXT4_BLOCKS_PER_GROUP(sb)) {
+			BUG_ON(free != 0);
+			break;
+		}
+
+		mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex);
+		BUG_ON(ex.fe_len <= 0);
+		BUG_ON(free < ex.fe_len);
+
+		ext4_mb_measure_extent(ac, &ex, e4b);
+
+		i += ex.fe_len;
+		free -= ex.fe_len;
+	}
+
+	ext4_mb_check_limits(ac, e4b, 1);
+}
+
+/*
+ * This is a special case for storages like raid5
+ * we try to find stripe-aligned chunks for stripe-size requests
+ * XXX should do so at least for multiples of stripe size as well
+ */
+static void ext4_mb_scan_aligned(struct ext4_allocation_context *ac,
+				 struct ext4_buddy *e4b)
+{
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	void *bitmap = EXT4_MB_BITMAP(e4b);
+	struct ext4_free_extent ex;
+	ext4_fsblk_t first_group_block;
+	ext4_fsblk_t a;
+	ext4_grpblk_t i;
+	int max;
+
+	BUG_ON(sbi->s_stripe == 0);
+
+	/* find first stripe-aligned block in group */
+	first_group_block = e4b->bd_group * EXT4_BLOCKS_PER_GROUP(sb)
+		+ le32_to_cpu(sbi->s_es->s_first_data_block);
+	a = first_group_block + sbi->s_stripe - 1;
+	do_div(a, sbi->s_stripe);
+	i = (a * sbi->s_stripe) - first_group_block;
+
+	while (i < EXT4_BLOCKS_PER_GROUP(sb)) {
+		if (!mb_test_bit(i, bitmap)) {
+			max = mb_find_extent(e4b, 0, i, sbi->s_stripe, &ex);
+			if (max >= sbi->s_stripe) {
+				ac->ac_found++;
+				ac->ac_b_ex = ex;
+				ext4_mb_use_best_found(ac, e4b);
+				break;
+			}
+		}
+		i += sbi->s_stripe;
+	}
+}
+
+static int ext4_mb_good_group(struct ext4_allocation_context *ac,
+				ext4_group_t group, int cr)
+{
+	unsigned free, fragments;
+	unsigned i, bits;
+	struct ext4_group_desc *desc;
+	struct ext4_group_info *grp = ext4_get_group_info(ac->ac_sb, group);
+
+	BUG_ON(cr < 0 || cr >= 4);
+	BUG_ON(EXT4_MB_GRP_NEED_INIT(grp));
+
+	free = grp->bb_free;
+	fragments = grp->bb_fragments;
+	if (free == 0)
+		return 0;
+	if (fragments == 0)
+		return 0;
+
+	switch (cr) {
+	case 0:
+		BUG_ON(ac->ac_2order == 0);
+		/* If this group is uninitialized, skip it initially */
+		desc = ext4_get_group_desc(ac->ac_sb, group, NULL);
+		if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))
+			return 0;
+
+		bits = ac->ac_sb->s_blocksize_bits + 1;
+		for (i = ac->ac_2order; i <= bits; i++)
+			if (grp->bb_counters[i] > 0)
+				return 1;
+		break;
+	case 1:
+		if ((free / fragments) >= ac->ac_g_ex.fe_len)
+			return 1;
+		break;
+	case 2:
+		if (free >= ac->ac_g_ex.fe_len)
+			return 1;
+		break;
+	case 3:
+		return 1;
+	default:
+		BUG();
+	}
+
+	return 0;
+}
+
+static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
+{
+	ext4_group_t group;
+	ext4_group_t i;
+	int cr;
+	int err = 0;
+	int bsbits;
+	struct ext4_sb_info *sbi;
+	struct super_block *sb;
+	struct ext4_buddy e4b;
+	loff_t size, isize;
+
+	sb = ac->ac_sb;
+	sbi = EXT4_SB(sb);
+	BUG_ON(ac->ac_status == AC_STATUS_FOUND);
+
+	/* first, try the goal */
+	err = ext4_mb_find_by_goal(ac, &e4b);
+	if (err || ac->ac_status == AC_STATUS_FOUND)
+		goto out;
+
+	if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
+		goto out;
+
+	/*
+	 * ac->ac2_order is set only if the fe_len is a power of 2
+	 * if ac2_order is set we also set criteria to 0 so that we
+	 * try exact allocation using buddy.
+	 */
+	i = fls(ac->ac_g_ex.fe_len);
+	ac->ac_2order = 0;
+	/*
+	 * We search using buddy data only if the order of the request
+	 * is greater than equal to the sbi_s_mb_order2_reqs
+	 * You can tune it via /proc/fs/ext4/<partition>/order2_req
+	 */
+	if (i >= sbi->s_mb_order2_reqs) {
+		/*
+		 * This should tell if fe_len is exactly power of 2
+		 */
+		if ((ac->ac_g_ex.fe_len & (~(1 << (i - 1)))) == 0)
+			ac->ac_2order = i - 1;
+	}
+
+	bsbits = ac->ac_sb->s_blocksize_bits;
+	/* if stream allocation is enabled, use global goal */
+	size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len;
+	isize = i_size_read(ac->ac_inode) >> bsbits;
+	if (size < isize)
+		size = isize;
+
+	if (size < sbi->s_mb_stream_request &&
+			(ac->ac_flags & EXT4_MB_HINT_DATA)) {
+		/* TBD: may be hot point */
+		spin_lock(&sbi->s_md_lock);
+		ac->ac_g_ex.fe_group = sbi->s_mb_last_group;
+		ac->ac_g_ex.fe_start = sbi->s_mb_last_start;
+		spin_unlock(&sbi->s_md_lock);
+	}
+
+	/* searching for the right group start from the goal value specified */
+	group = ac->ac_g_ex.fe_group;
+
+	/* Let's just scan groups to find more-less suitable blocks */
+	cr = ac->ac_2order ? 0 : 1;
+	/*
+	 * cr == 0 try to get exact allocation,
+	 * cr == 3  try to get anything
+	 */
+repeat:
+	for (; cr < 4 && ac->ac_status == AC_STATUS_CONTINUE; cr++) {
+		ac->ac_criteria = cr;
+		for (i = 0; i < EXT4_SB(sb)->s_groups_count; group++, i++) {
+			struct ext4_group_info *grp;
+			struct ext4_group_desc *desc;
+
+			if (group == EXT4_SB(sb)->s_groups_count)
+				group = 0;
+
+			/* quick check to skip empty groups */
+			grp = ext4_get_group_info(ac->ac_sb, group);
+			if (grp->bb_free == 0)
+				continue;
+
+			/*
+			 * if the group is already init we check whether it is
+			 * a good group and if not we don't load the buddy
+			 */
+			if (EXT4_MB_GRP_NEED_INIT(grp)) {
+				/*
+				 * we need full data about the group
+				 * to make a good selection
+				 */
+				err = ext4_mb_load_buddy(sb, group, &e4b);
+				if (err)
+					goto out;
+				ext4_mb_release_desc(&e4b);
+			}
+
+			/*
+			 * If the particular group doesn't satisfy our
+			 * criteria we continue with the next group
+			 */
+			if (!ext4_mb_good_group(ac, group, cr))
+				continue;
+
+			err = ext4_mb_load_buddy(sb, group, &e4b);
+			if (err)
+				goto out;
+
+			ext4_lock_group(sb, group);
+			if (!ext4_mb_good_group(ac, group, cr)) {
+				/* someone did allocation from this group */
+				ext4_unlock_group(sb, group);
+				ext4_mb_release_desc(&e4b);
+				continue;
+			}
+
+			ac->ac_groups_scanned++;
+			desc = ext4_get_group_desc(sb, group, NULL);
+			if (cr == 0 || (desc->bg_flags &
+					cpu_to_le16(EXT4_BG_BLOCK_UNINIT) &&
+					ac->ac_2order != 0))
+				ext4_mb_simple_scan_group(ac, &e4b);
+			else if (cr == 1 &&
+					ac->ac_g_ex.fe_len == sbi->s_stripe)
+				ext4_mb_scan_aligned(ac, &e4b);
+			else
+				ext4_mb_complex_scan_group(ac, &e4b);
+
+			ext4_unlock_group(sb, group);
+			ext4_mb_release_desc(&e4b);
+
+			if (ac->ac_status != AC_STATUS_CONTINUE)
+				break;
+		}
+	}
+
+	if (ac->ac_b_ex.fe_len > 0 && ac->ac_status != AC_STATUS_FOUND &&
+	    !(ac->ac_flags & EXT4_MB_HINT_FIRST)) {
+		/*
+		 * We've been searching too long. Let's try to allocate
+		 * the best chunk we've found so far
+		 */
+
+		ext4_mb_try_best_found(ac, &e4b);
+		if (ac->ac_status != AC_STATUS_FOUND) {
+			/*
+			 * Someone more lucky has already allocated it.
+			 * The only thing we can do is just take first
+			 * found block(s)
+			printk(KERN_DEBUG "EXT4-fs: someone won our chunk\n");
+			 */
+			ac->ac_b_ex.fe_group = 0;
+			ac->ac_b_ex.fe_start = 0;
+			ac->ac_b_ex.fe_len = 0;
+			ac->ac_status = AC_STATUS_CONTINUE;
+			ac->ac_flags |= EXT4_MB_HINT_FIRST;
+			cr = 3;
+			atomic_inc(&sbi->s_mb_lost_chunks);
+			goto repeat;
+		}
+	}
+out:
+	return err;
+}
+
+#ifdef EXT4_MB_HISTORY
+struct ext4_mb_proc_session {
+	struct ext4_mb_history *history;
+	struct super_block *sb;
+	int start;
+	int max;
+};
+
+static void *ext4_mb_history_skip_empty(struct ext4_mb_proc_session *s,
+					struct ext4_mb_history *hs,
+					int first)
+{
+	if (hs == s->history + s->max)
+		hs = s->history;
+	if (!first && hs == s->history + s->start)
+		return NULL;
+	while (hs->orig.fe_len == 0) {
+		hs++;
+		if (hs == s->history + s->max)
+			hs = s->history;
+		if (hs == s->history + s->start)
+			return NULL;
+	}
+	return hs;
+}
+
+static void *ext4_mb_seq_history_start(struct seq_file *seq, loff_t *pos)
+{
+	struct ext4_mb_proc_session *s = seq->private;
+	struct ext4_mb_history *hs;
+	int l = *pos;
+
+	if (l == 0)
+		return SEQ_START_TOKEN;
+	hs = ext4_mb_history_skip_empty(s, s->history + s->start, 1);
+	if (!hs)
+		return NULL;
+	while (--l && (hs = ext4_mb_history_skip_empty(s, ++hs, 0)) != NULL);
+	return hs;
+}
+
+static void *ext4_mb_seq_history_next(struct seq_file *seq, void *v,
+				      loff_t *pos)
+{
+	struct ext4_mb_proc_session *s = seq->private;
+	struct ext4_mb_history *hs = v;
+
+	++*pos;
+	if (v == SEQ_START_TOKEN)
+		return ext4_mb_history_skip_empty(s, s->history + s->start, 1);
+	else
+		return ext4_mb_history_skip_empty(s, ++hs, 0);
+}
+
+static int ext4_mb_seq_history_show(struct seq_file *seq, void *v)
+{
+	char buf[25], buf2[25], buf3[25], *fmt;
+	struct ext4_mb_history *hs = v;
+
+	if (v == SEQ_START_TOKEN) {
+		seq_printf(seq, "%-5s %-8s %-23s %-23s %-23s %-5s "
+				"%-5s %-2s %-5s %-5s %-5s %-6s\n",
+			  "pid", "inode", "original", "goal", "result", "found",
+			   "grps", "cr", "flags", "merge", "tail", "broken");
+		return 0;
+	}
+
+	if (hs->op == EXT4_MB_HISTORY_ALLOC) {
+		fmt = "%-5u %-8u %-23s %-23s %-23s %-5u %-5u %-2u "
+			"%-5u %-5s %-5u %-6u\n";
+		sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group,
+			hs->result.fe_start, hs->result.fe_len,
+			hs->result.fe_logical);
+		sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group,
+			hs->orig.fe_start, hs->orig.fe_len,
+			hs->orig.fe_logical);
+		sprintf(buf3, "%lu/%d/%u@%u", hs->goal.fe_group,
+			hs->goal.fe_start, hs->goal.fe_len,
+			hs->goal.fe_logical);
+		seq_printf(seq, fmt, hs->pid, hs->ino, buf, buf3, buf2,
+				hs->found, hs->groups, hs->cr, hs->flags,
+				hs->merged ? "M" : "", hs->tail,
+				hs->buddy ? 1 << hs->buddy : 0);
+	} else if (hs->op == EXT4_MB_HISTORY_PREALLOC) {
+		fmt = "%-5u %-8u %-23s %-23s %-23s\n";
+		sprintf(buf2, "%lu/%d/%u@%u", hs->result.fe_group,
+			hs->result.fe_start, hs->result.fe_len,
+			hs->result.fe_logical);
+		sprintf(buf, "%lu/%d/%u@%u", hs->orig.fe_group,
+			hs->orig.fe_start, hs->orig.fe_len,
+			hs->orig.fe_logical);
+		seq_printf(seq, fmt, hs->pid, hs->ino, buf, "", buf2);
+	} else if (hs->op == EXT4_MB_HISTORY_DISCARD) {
+		sprintf(buf2, "%lu/%d/%u", hs->result.fe_group,
+			hs->result.fe_start, hs->result.fe_len);
+		seq_printf(seq, "%-5u %-8u %-23s discard\n",
+				hs->pid, hs->ino, buf2);
+	} else if (hs->op == EXT4_MB_HISTORY_FREE) {
+		sprintf(buf2, "%lu/%d/%u", hs->result.fe_group,
+			hs->result.fe_start, hs->result.fe_len);
+		seq_printf(seq, "%-5u %-8u %-23s free\n",
+				hs->pid, hs->ino, buf2);
+	}
+	return 0;
+}
+
+static void ext4_mb_seq_history_stop(struct seq_file *seq, void *v)
+{
+}
+
+static struct seq_operations ext4_mb_seq_history_ops = {
+	.start  = ext4_mb_seq_history_start,
+	.next   = ext4_mb_seq_history_next,
+	.stop   = ext4_mb_seq_history_stop,
+	.show   = ext4_mb_seq_history_show,
+};
+
+static int ext4_mb_seq_history_open(struct inode *inode, struct file *file)
+{
+	struct super_block *sb = PDE(inode)->data;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct ext4_mb_proc_session *s;
+	int rc;
+	int size;
+
+	s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s == NULL)
+		return -ENOMEM;
+	s->sb = sb;
+	size = sizeof(struct ext4_mb_history) * sbi->s_mb_history_max;
+	s->history = kmalloc(size, GFP_KERNEL);
+	if (s->history == NULL) {
+		kfree(s);
+		return -ENOMEM;
+	}
+
+	spin_lock(&sbi->s_mb_history_lock);
+	memcpy(s->history, sbi->s_mb_history, size);
+	s->max = sbi->s_mb_history_max;
+	s->start = sbi->s_mb_history_cur % s->max;
+	spin_unlock(&sbi->s_mb_history_lock);
+
+	rc = seq_open(file, &ext4_mb_seq_history_ops);
+	if (rc == 0) {
+		struct seq_file *m = (struct seq_file *)file->private_data;
+		m->private = s;
+	} else {
+		kfree(s->history);
+		kfree(s);
+	}
+	return rc;
+
+}
+
+static int ext4_mb_seq_history_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = (struct seq_file *)file->private_data;
+	struct ext4_mb_proc_session *s = seq->private;
+	kfree(s->history);
+	kfree(s);
+	return seq_release(inode, file);
+}
+
+static ssize_t ext4_mb_seq_history_write(struct file *file,
+				const char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	struct seq_file *seq = (struct seq_file *)file->private_data;
+	struct ext4_mb_proc_session *s = seq->private;
+	struct super_block *sb = s->sb;
+	char str[32];
+	int value;
+
+	if (count >= sizeof(str)) {
+		printk(KERN_ERR "EXT4-fs: %s string too long, max %u bytes\n",
+				"mb_history", (int)sizeof(str));
+		return -EOVERFLOW;
+	}
+
+	if (copy_from_user(str, buffer, count))
+		return -EFAULT;
+
+	value = simple_strtol(str, NULL, 0);
+	if (value < 0)
+		return -ERANGE;
+	EXT4_SB(sb)->s_mb_history_filter = value;
+
+	return count;
+}
+
+static struct file_operations ext4_mb_seq_history_fops = {
+	.owner		= THIS_MODULE,
+	.open		= ext4_mb_seq_history_open,
+	.read		= seq_read,
+	.write		= ext4_mb_seq_history_write,
+	.llseek		= seq_lseek,
+	.release	= ext4_mb_seq_history_release,
+};
+
+static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos)
+{
+	struct super_block *sb = seq->private;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	ext4_group_t group;
+
+	if (*pos < 0 || *pos >= sbi->s_groups_count)
+		return NULL;
+
+	group = *pos + 1;
+	return (void *) group;
+}
+
+static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct super_block *sb = seq->private;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	ext4_group_t group;
+
+	++*pos;
+	if (*pos < 0 || *pos >= sbi->s_groups_count)
+		return NULL;
+	group = *pos + 1;
+	return (void *) group;;
+}
+
+static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
+{
+	struct super_block *sb = seq->private;
+	long group = (long) v;
+	int i;
+	int err;
+	struct ext4_buddy e4b;
+	struct sg {
+		struct ext4_group_info info;
+		unsigned short counters[16];
+	} sg;
+
+	group--;
+	if (group == 0)
+		seq_printf(seq, "#%-5s: %-5s %-5s %-5s "
+				"[ %-5s %-5s %-5s %-5s %-5s %-5s %-5s "
+				  "%-5s %-5s %-5s %-5s %-5s %-5s %-5s ]\n",
+			   "group", "free", "frags", "first",
+			   "2^0", "2^1", "2^2", "2^3", "2^4", "2^5", "2^6",
+			   "2^7", "2^8", "2^9", "2^10", "2^11", "2^12", "2^13");
+
+	i = (sb->s_blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) +
+		sizeof(struct ext4_group_info);
+	err = ext4_mb_load_buddy(sb, group, &e4b);
+	if (err) {
+		seq_printf(seq, "#%-5lu: I/O error\n", group);
+		return 0;
+	}
+	ext4_lock_group(sb, group);
+	memcpy(&sg, ext4_get_group_info(sb, group), i);
+	ext4_unlock_group(sb, group);
+	ext4_mb_release_desc(&e4b);
+
+	seq_printf(seq, "#%-5lu: %-5u %-5u %-5u [", group, sg.info.bb_free,
+			sg.info.bb_fragments, sg.info.bb_first_free);
+	for (i = 0; i <= 13; i++)
+		seq_printf(seq, " %-5u", i <= sb->s_blocksize_bits + 1 ?
+				sg.info.bb_counters[i] : 0);
+	seq_printf(seq, " ]\n");
+
+	return 0;
+}
+
+static void ext4_mb_seq_groups_stop(struct seq_file *seq, void *v)
+{
+}
+
+static struct seq_operations ext4_mb_seq_groups_ops = {
+	.start  = ext4_mb_seq_groups_start,
+	.next   = ext4_mb_seq_groups_next,
+	.stop   = ext4_mb_seq_groups_stop,
+	.show   = ext4_mb_seq_groups_show,
+};
+
+static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file)
+{
+	struct super_block *sb = PDE(inode)->data;
+	int rc;
+
+	rc = seq_open(file, &ext4_mb_seq_groups_ops);
+	if (rc == 0) {
+		struct seq_file *m = (struct seq_file *)file->private_data;
+		m->private = sb;
+	}
+	return rc;
+
+}
+
+static struct file_operations ext4_mb_seq_groups_fops = {
+	.owner		= THIS_MODULE,
+	.open		= ext4_mb_seq_groups_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static void ext4_mb_history_release(struct super_block *sb)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	remove_proc_entry("mb_groups", sbi->s_mb_proc);
+	remove_proc_entry("mb_history", sbi->s_mb_proc);
+
+	kfree(sbi->s_mb_history);
+}
+
+static void ext4_mb_history_init(struct super_block *sb)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	int i;
+
+	if (sbi->s_mb_proc != NULL) {
+		struct proc_dir_entry *p;
+		p = create_proc_entry("mb_history", S_IRUGO, sbi->s_mb_proc);
+		if (p) {
+			p->proc_fops = &ext4_mb_seq_history_fops;
+			p->data = sb;
+		}
+		p = create_proc_entry("mb_groups", S_IRUGO, sbi->s_mb_proc);
+		if (p) {
+			p->proc_fops = &ext4_mb_seq_groups_fops;
+			p->data = sb;
+		}
+	}
+
+	sbi->s_mb_history_max = 1000;
+	sbi->s_mb_history_cur = 0;
+	spin_lock_init(&sbi->s_mb_history_lock);
+	i = sbi->s_mb_history_max * sizeof(struct ext4_mb_history);
+	sbi->s_mb_history = kmalloc(i, GFP_KERNEL);
+	if (likely(sbi->s_mb_history != NULL))
+		memset(sbi->s_mb_history, 0, i);
+	/* if we can't allocate history, then we simple won't use it */
+}
+
+static void ext4_mb_store_history(struct ext4_allocation_context *ac)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+	struct ext4_mb_history h;
+
+	if (unlikely(sbi->s_mb_history == NULL))
+		return;
+
+	if (!(ac->ac_op & sbi->s_mb_history_filter))
+		return;
+
+	h.op = ac->ac_op;
+	h.pid = current->pid;
+	h.ino = ac->ac_inode ? ac->ac_inode->i_ino : 0;
+	h.orig = ac->ac_o_ex;
+	h.result = ac->ac_b_ex;
+	h.flags = ac->ac_flags;
+	h.found = ac->ac_found;
+	h.groups = ac->ac_groups_scanned;
+	h.cr = ac->ac_criteria;
+	h.tail = ac->ac_tail;
+	h.buddy = ac->ac_buddy;
+	h.merged = 0;
+	if (ac->ac_op == EXT4_MB_HISTORY_ALLOC) {
+		if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start &&
+				ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group)
+			h.merged = 1;
+		h.goal = ac->ac_g_ex;
+		h.result = ac->ac_f_ex;
+	}
+
+	spin_lock(&sbi->s_mb_history_lock);
+	memcpy(sbi->s_mb_history + sbi->s_mb_history_cur, &h, sizeof(h));
+	if (++sbi->s_mb_history_cur >= sbi->s_mb_history_max)
+		sbi->s_mb_history_cur = 0;
+	spin_unlock(&sbi->s_mb_history_lock);
+}
+
+#else
+#define ext4_mb_history_release(sb)
+#define ext4_mb_history_init(sb)
+#endif
+
+static int ext4_mb_init_backend(struct super_block *sb)
+{
+	ext4_group_t i;
+	int j, len, metalen;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	int num_meta_group_infos =
+		(sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) >>
+			EXT4_DESC_PER_BLOCK_BITS(sb);
+	struct ext4_group_info **meta_group_info;
+
+	/* An 8TB filesystem with 64-bit pointers requires a 4096 byte
+	 * kmalloc. A 128kb malloc should suffice for a 256TB filesystem.
+	 * So a two level scheme suffices for now. */
+	sbi->s_group_info = kmalloc(sizeof(*sbi->s_group_info) *
+				    num_meta_group_infos, GFP_KERNEL);
+	if (sbi->s_group_info == NULL) {
+		printk(KERN_ERR "EXT4-fs: can't allocate buddy meta group\n");
+		return -ENOMEM;
+	}
+	sbi->s_buddy_cache = new_inode(sb);
+	if (sbi->s_buddy_cache == NULL) {
+		printk(KERN_ERR "EXT4-fs: can't get new inode\n");
+		goto err_freesgi;
+	}
+	EXT4_I(sbi->s_buddy_cache)->i_disksize = 0;
+
+	metalen = sizeof(*meta_group_info) << EXT4_DESC_PER_BLOCK_BITS(sb);
+	for (i = 0; i < num_meta_group_infos; i++) {
+		if ((i + 1) == num_meta_group_infos)
+			metalen = sizeof(*meta_group_info) *
+				(sbi->s_groups_count -
+					(i << EXT4_DESC_PER_BLOCK_BITS(sb)));
+		meta_group_info = kmalloc(metalen, GFP_KERNEL);
+		if (meta_group_info == NULL) {
+			printk(KERN_ERR "EXT4-fs: can't allocate mem for a "
+			       "buddy group\n");
+			goto err_freemeta;
+		}
+		sbi->s_group_info[i] = meta_group_info;
+	}
+
+	/*
+	 * calculate needed size. if change bb_counters size,
+	 * don't forget about ext4_mb_generate_buddy()
+	 */
+	len = sizeof(struct ext4_group_info);
+	len += sizeof(unsigned short) * (sb->s_blocksize_bits + 2);
+	for (i = 0; i < sbi->s_groups_count; i++) {
+		struct ext4_group_desc *desc;
+
+		meta_group_info =
+			sbi->s_group_info[i >> EXT4_DESC_PER_BLOCK_BITS(sb)];
+		j = i & (EXT4_DESC_PER_BLOCK(sb) - 1);
+
+		meta_group_info[j] = kzalloc(len, GFP_KERNEL);
+		if (meta_group_info[j] == NULL) {
+			printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
+			i--;
+			goto err_freebuddy;
+		}
+		desc = ext4_get_group_desc(sb, i, NULL);
+		if (desc == NULL) {
+			printk(KERN_ERR
+				"EXT4-fs: can't read descriptor %lu\n", i);
+			goto err_freebuddy;
+		}
+		memset(meta_group_info[j], 0, len);
+		set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT,
+			&(meta_group_info[j]->bb_state));
+
+		/*
+		 * initialize bb_free to be able to skip
+		 * empty groups without initialization
+		 */
+		if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
+			meta_group_info[j]->bb_free =
+				ext4_free_blocks_after_init(sb, i, desc);
+		} else {
+			meta_group_info[j]->bb_free =
+				le16_to_cpu(desc->bg_free_blocks_count);
+		}
+
+		INIT_LIST_HEAD(&meta_group_info[j]->bb_prealloc_list);
+
+#ifdef DOUBLE_CHECK
+		{
+			struct buffer_head *bh;
+			meta_group_info[j]->bb_bitmap =
+				kmalloc(sb->s_blocksize, GFP_KERNEL);
+			BUG_ON(meta_group_info[j]->bb_bitmap == NULL);
+			bh = read_block_bitmap(sb, i);
+			BUG_ON(bh == NULL);
+			memcpy(meta_group_info[j]->bb_bitmap, bh->b_data,
+					sb->s_blocksize);
+			put_bh(bh);
+		}
+#endif
+
+	}
+
+	return 0;
+
+err_freebuddy:
+	while (i >= 0) {
+		kfree(ext4_get_group_info(sb, i));
+		i--;
+	}
+	i = num_meta_group_infos;
+err_freemeta:
+	while (--i >= 0)
+		kfree(sbi->s_group_info[i]);
+	iput(sbi->s_buddy_cache);
+err_freesgi:
+	kfree(sbi->s_group_info);
+	return -ENOMEM;
+}
+
+int ext4_mb_init(struct super_block *sb, int needs_recovery)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	unsigned i;
+	unsigned offset;
+	unsigned max;
+
+	if (!test_opt(sb, MBALLOC))
+		return 0;
+
+	i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short);
+
+	sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL);
+	if (sbi->s_mb_offsets == NULL) {
+		clear_opt(sbi->s_mount_opt, MBALLOC);
+		return -ENOMEM;
+	}
+	sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL);
+	if (sbi->s_mb_maxs == NULL) {
+		clear_opt(sbi->s_mount_opt, MBALLOC);
+		kfree(sbi->s_mb_maxs);
+		return -ENOMEM;
+	}
+
+	/* order 0 is regular bitmap */
+	sbi->s_mb_maxs[0] = sb->s_blocksize << 3;
+	sbi->s_mb_offsets[0] = 0;
+
+	i = 1;
+	offset = 0;
+	max = sb->s_blocksize << 2;
+	do {
+		sbi->s_mb_offsets[i] = offset;
+		sbi->s_mb_maxs[i] = max;
+		offset += 1 << (sb->s_blocksize_bits - i);
+		max = max >> 1;
+		i++;
+	} while (i <= sb->s_blocksize_bits + 1);
+
+	/* init file for buddy data */
+	i = ext4_mb_init_backend(sb);
+	if (i) {
+		clear_opt(sbi->s_mount_opt, MBALLOC);
+		kfree(sbi->s_mb_offsets);
+		kfree(sbi->s_mb_maxs);
+		return i;
+	}
+
+	spin_lock_init(&sbi->s_md_lock);
+	INIT_LIST_HEAD(&sbi->s_active_transaction);
+	INIT_LIST_HEAD(&sbi->s_closed_transaction);
+	INIT_LIST_HEAD(&sbi->s_committed_transaction);
+	spin_lock_init(&sbi->s_bal_lock);
+
+	sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN;
+	sbi->s_mb_min_to_scan = MB_DEFAULT_MIN_TO_SCAN;
+	sbi->s_mb_stats = MB_DEFAULT_STATS;
+	sbi->s_mb_stream_request = MB_DEFAULT_STREAM_THRESHOLD;
+	sbi->s_mb_order2_reqs = MB_DEFAULT_ORDER2_REQS;
+	sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT;
+	sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC;
+
+	i = sizeof(struct ext4_locality_group) * NR_CPUS;
+	sbi->s_locality_groups = kmalloc(i, GFP_KERNEL);
+	if (sbi->s_locality_groups == NULL) {
+		clear_opt(sbi->s_mount_opt, MBALLOC);
+		kfree(sbi->s_mb_offsets);
+		kfree(sbi->s_mb_maxs);
+		return -ENOMEM;
+	}
+	for (i = 0; i < NR_CPUS; i++) {
+		struct ext4_locality_group *lg;
+		lg = &sbi->s_locality_groups[i];
+		mutex_init(&lg->lg_mutex);
+		INIT_LIST_HEAD(&lg->lg_prealloc_list);
+		spin_lock_init(&lg->lg_prealloc_lock);
+	}
+
+	ext4_mb_init_per_dev_proc(sb);
+	ext4_mb_history_init(sb);
+
+	printk("EXT4-fs: mballoc enabled\n");
+	return 0;
+}
+
+/* need to called with ext4 group lock (ext4_lock_group) */
+static void ext4_mb_cleanup_pa(struct ext4_group_info *grp)
+{
+	struct ext4_prealloc_space *pa;
+	struct list_head *cur, *tmp;
+	int count = 0;
+
+	list_for_each_safe(cur, tmp, &grp->bb_prealloc_list) {
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list);
+		list_del(&pa->pa_group_list);
+		count++;
+		kfree(pa);
+	}
+	if (count)
+		mb_debug("mballoc: %u PAs left\n", count);
+
+}
+
+int ext4_mb_release(struct super_block *sb)
+{
+	ext4_group_t i;
+	int num_meta_group_infos;
+	struct ext4_group_info *grinfo;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	if (!test_opt(sb, MBALLOC))
+		return 0;
+
+	/* release freed, non-committed blocks */
+	spin_lock(&sbi->s_md_lock);
+	list_splice_init(&sbi->s_closed_transaction,
+			&sbi->s_committed_transaction);
+	list_splice_init(&sbi->s_active_transaction,
+			&sbi->s_committed_transaction);
+	spin_unlock(&sbi->s_md_lock);
+	ext4_mb_free_committed_blocks(sb);
+
+	if (sbi->s_group_info) {
+		for (i = 0; i < sbi->s_groups_count; i++) {
+			grinfo = ext4_get_group_info(sb, i);
+#ifdef DOUBLE_CHECK
+			kfree(grinfo->bb_bitmap);
+#endif
+			ext4_lock_group(sb, i);
+			ext4_mb_cleanup_pa(grinfo);
+			ext4_unlock_group(sb, i);
+			kfree(grinfo);
+		}
+		num_meta_group_infos = (sbi->s_groups_count +
+				EXT4_DESC_PER_BLOCK(sb) - 1) >>
+			EXT4_DESC_PER_BLOCK_BITS(sb);
+		for (i = 0; i < num_meta_group_infos; i++)
+			kfree(sbi->s_group_info[i]);
+		kfree(sbi->s_group_info);
+	}
+	kfree(sbi->s_mb_offsets);
+	kfree(sbi->s_mb_maxs);
+	if (sbi->s_buddy_cache)
+		iput(sbi->s_buddy_cache);
+	if (sbi->s_mb_stats) {
+		printk(KERN_INFO
+		       "EXT4-fs: mballoc: %u blocks %u reqs (%u success)\n",
+				atomic_read(&sbi->s_bal_allocated),
+				atomic_read(&sbi->s_bal_reqs),
+				atomic_read(&sbi->s_bal_success));
+		printk(KERN_INFO
+		      "EXT4-fs: mballoc: %u extents scanned, %u goal hits, "
+				"%u 2^N hits, %u breaks, %u lost\n",
+				atomic_read(&sbi->s_bal_ex_scanned),
+				atomic_read(&sbi->s_bal_goals),
+				atomic_read(&sbi->s_bal_2orders),
+				atomic_read(&sbi->s_bal_breaks),
+				atomic_read(&sbi->s_mb_lost_chunks));
+		printk(KERN_INFO
+		       "EXT4-fs: mballoc: %lu generated and it took %Lu\n",
+				sbi->s_mb_buddies_generated++,
+				sbi->s_mb_generation_time);
+		printk(KERN_INFO
+		       "EXT4-fs: mballoc: %u preallocated, %u discarded\n",
+				atomic_read(&sbi->s_mb_preallocated),
+				atomic_read(&sbi->s_mb_discarded));
+	}
+
+	kfree(sbi->s_locality_groups);
+
+	ext4_mb_history_release(sb);
+	ext4_mb_destroy_per_dev_proc(sb);
+
+	return 0;
+}
+
+static void ext4_mb_free_committed_blocks(struct super_block *sb)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	int err;
+	int i;
+	int count = 0;
+	int count2 = 0;
+	struct ext4_free_metadata *md;
+	struct ext4_buddy e4b;
+
+	if (list_empty(&sbi->s_committed_transaction))
+		return;
+
+	/* there is committed blocks to be freed yet */
+	do {
+		/* get next array of blocks */
+		md = NULL;
+		spin_lock(&sbi->s_md_lock);
+		if (!list_empty(&sbi->s_committed_transaction)) {
+			md = list_entry(sbi->s_committed_transaction.next,
+					struct ext4_free_metadata, list);
+			list_del(&md->list);
+		}
+		spin_unlock(&sbi->s_md_lock);
+
+		if (md == NULL)
+			break;
+
+		mb_debug("gonna free %u blocks in group %lu (0x%p):",
+				md->num, md->group, md);
+
+		err = ext4_mb_load_buddy(sb, md->group, &e4b);
+		/* we expect to find existing buddy because it's pinned */
+		BUG_ON(err != 0);
+
+		/* there are blocks to put in buddy to make them really free */
+		count += md->num;
+		count2++;
+		ext4_lock_group(sb, md->group);
+		for (i = 0; i < md->num; i++) {
+			mb_debug(" %u", md->blocks[i]);
+			err = mb_free_blocks(NULL, &e4b, md->blocks[i], 1);
+			BUG_ON(err != 0);
+		}
+		mb_debug("\n");
+		ext4_unlock_group(sb, md->group);
+
+		/* balance refcounts from ext4_mb_free_metadata() */
+		page_cache_release(e4b.bd_buddy_page);
+		page_cache_release(e4b.bd_bitmap_page);
+
+		kfree(md);
+		ext4_mb_release_desc(&e4b);
+
+	} while (md);
+
+	mb_debug("freed %u blocks in %u structures\n", count, count2);
+}
+
+#define EXT4_ROOT			"ext4"
+#define EXT4_MB_STATS_NAME		"stats"
+#define EXT4_MB_MAX_TO_SCAN_NAME	"max_to_scan"
+#define EXT4_MB_MIN_TO_SCAN_NAME	"min_to_scan"
+#define EXT4_MB_ORDER2_REQ		"order2_req"
+#define EXT4_MB_STREAM_REQ		"stream_req"
+#define EXT4_MB_GROUP_PREALLOC		"group_prealloc"
+
+
+
+#define MB_PROC_VALUE_READ(name)				\
+static int ext4_mb_read_##name(char *page, char **start,	\
+		off_t off, int count, int *eof, void *data)	\
+{								\
+	struct ext4_sb_info *sbi = data;			\
+	int len;						\
+	*eof = 1;						\
+	if (off != 0)						\
+		return 0;					\
+	len = sprintf(page, "%ld\n", sbi->s_mb_##name);		\
+	*start = page;						\
+	return len;						\
+}
+
+#define MB_PROC_VALUE_WRITE(name)				\
+static int ext4_mb_write_##name(struct file *file,		\
+		const char __user *buf, unsigned long cnt, void *data)	\
+{								\
+	struct ext4_sb_info *sbi = data;			\
+	char str[32];						\
+	long value;						\
+	if (cnt >= sizeof(str))					\
+		return -EINVAL;					\
+	if (copy_from_user(str, buf, cnt))			\
+		return -EFAULT;					\
+	value = simple_strtol(str, NULL, 0);			\
+	if (value <= 0)						\
+		return -ERANGE;					\
+	sbi->s_mb_##name = value;				\
+	return cnt;						\
+}
+
+MB_PROC_VALUE_READ(stats);
+MB_PROC_VALUE_WRITE(stats);
+MB_PROC_VALUE_READ(max_to_scan);
+MB_PROC_VALUE_WRITE(max_to_scan);
+MB_PROC_VALUE_READ(min_to_scan);
+MB_PROC_VALUE_WRITE(min_to_scan);
+MB_PROC_VALUE_READ(order2_reqs);
+MB_PROC_VALUE_WRITE(order2_reqs);
+MB_PROC_VALUE_READ(stream_request);
+MB_PROC_VALUE_WRITE(stream_request);
+MB_PROC_VALUE_READ(group_prealloc);
+MB_PROC_VALUE_WRITE(group_prealloc);
+
+#define	MB_PROC_HANDLER(name, var)					\
+do {									\
+	proc = create_proc_entry(name, mode, sbi->s_mb_proc);		\
+	if (proc == NULL) {						\
+		printk(KERN_ERR "EXT4-fs: can't to create %s\n", name);	\
+		goto err_out;						\
+	}								\
+	proc->data = sbi;						\
+	proc->read_proc  = ext4_mb_read_##var ;				\
+	proc->write_proc = ext4_mb_write_##var;				\
+} while (0)
+
+static int ext4_mb_init_per_dev_proc(struct super_block *sb)
+{
+	mode_t mode = S_IFREG | S_IRUGO | S_IWUSR;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct proc_dir_entry *proc;
+	char devname[64];
+
+	snprintf(devname, sizeof(devname) - 1, "%s",
+		bdevname(sb->s_bdev, devname));
+	sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4);
+
+	MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats);
+	MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan);
+	MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan);
+	MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs);
+	MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request);
+	MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc);
+
+	return 0;
+
+err_out:
+	printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname);
+	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc);
+	remove_proc_entry(devname, proc_root_ext4);
+	sbi->s_mb_proc = NULL;
+
+	return -ENOMEM;
+}
+
+static int ext4_mb_destroy_per_dev_proc(struct super_block *sb)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	char devname[64];
+
+	if (sbi->s_mb_proc == NULL)
+		return -EINVAL;
+
+	snprintf(devname, sizeof(devname) - 1, "%s",
+		bdevname(sb->s_bdev, devname));
+	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc);
+	remove_proc_entry(devname, proc_root_ext4);
+
+	return 0;
+}
+
+int __init init_ext4_mballoc(void)
+{
+	ext4_pspace_cachep =
+		kmem_cache_create("ext4_prealloc_space",
+				     sizeof(struct ext4_prealloc_space),
+				     0, SLAB_RECLAIM_ACCOUNT, NULL);
+	if (ext4_pspace_cachep == NULL)
+		return -ENOMEM;
+
+#ifdef CONFIG_PROC_FS
+	proc_root_ext4 = proc_mkdir(EXT4_ROOT, proc_root_fs);
+	if (proc_root_ext4 == NULL)
+		printk(KERN_ERR "EXT4-fs: Unable to create %s\n", EXT4_ROOT);
+#endif
+
+	return 0;
+}
+
+void exit_ext4_mballoc(void)
+{
+	/* XXX: synchronize_rcu(); */
+	kmem_cache_destroy(ext4_pspace_cachep);
+#ifdef CONFIG_PROC_FS
+	remove_proc_entry(EXT4_ROOT, proc_root_fs);
+#endif
+}
+
+
+/*
+ * Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps
+ * Returns 0 if success or error code
+ */
+static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
+				handle_t *handle)
+{
+	struct buffer_head *bitmap_bh = NULL;
+	struct ext4_super_block *es;
+	struct ext4_group_desc *gdp;
+	struct buffer_head *gdp_bh;
+	struct ext4_sb_info *sbi;
+	struct super_block *sb;
+	ext4_fsblk_t block;
+	int err;
+
+	BUG_ON(ac->ac_status != AC_STATUS_FOUND);
+	BUG_ON(ac->ac_b_ex.fe_len <= 0);
+
+	sb = ac->ac_sb;
+	sbi = EXT4_SB(sb);
+	es = sbi->s_es;
+
+	ext4_debug("using block group %lu(%d)\n", ac->ac_b_ex.fe_group,
+			gdp->bg_free_blocks_count);
+
+	err = -EIO;
+	bitmap_bh = read_block_bitmap(sb, ac->ac_b_ex.fe_group);
+	if (!bitmap_bh)
+		goto out_err;
+
+	err = ext4_journal_get_write_access(handle, bitmap_bh);
+	if (err)
+		goto out_err;
+
+	err = -EIO;
+	gdp = ext4_get_group_desc(sb, ac->ac_b_ex.fe_group, &gdp_bh);
+	if (!gdp)
+		goto out_err;
+
+	err = ext4_journal_get_write_access(handle, gdp_bh);
+	if (err)
+		goto out_err;
+
+	block = ac->ac_b_ex.fe_group * EXT4_BLOCKS_PER_GROUP(sb)
+		+ ac->ac_b_ex.fe_start
+		+ le32_to_cpu(es->s_first_data_block);
+
+	if (block == ext4_block_bitmap(sb, gdp) ||
+			block == ext4_inode_bitmap(sb, gdp) ||
+			in_range(block, ext4_inode_table(sb, gdp),
+				EXT4_SB(sb)->s_itb_per_group)) {
+
+		ext4_error(sb, __FUNCTION__,
+			   "Allocating block in system zone - block = %llu",
+			   block);
+	}
+#ifdef AGGRESSIVE_CHECK
+	{
+		int i;
+		for (i = 0; i < ac->ac_b_ex.fe_len; i++) {
+			BUG_ON(mb_test_bit(ac->ac_b_ex.fe_start + i,
+						bitmap_bh->b_data));
+		}
+	}
+#endif
+	mb_set_bits(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group), bitmap_bh->b_data,
+				ac->ac_b_ex.fe_start, ac->ac_b_ex.fe_len);
+
+	spin_lock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
+	if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
+		gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
+		gdp->bg_free_blocks_count =
+			cpu_to_le16(ext4_free_blocks_after_init(sb,
+						ac->ac_b_ex.fe_group,
+						gdp));
+	}
+	gdp->bg_free_blocks_count =
+		cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)
+				- ac->ac_b_ex.fe_len);
+	gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp);
+	spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
+	percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len);
+
+	err = ext4_journal_dirty_metadata(handle, bitmap_bh);
+	if (err)
+		goto out_err;
+	err = ext4_journal_dirty_metadata(handle, gdp_bh);
+
+out_err:
+	sb->s_dirt = 1;
+	put_bh(bitmap_bh);
+	return err;
+}
+
+/*
+ * here we normalize request for locality group
+ * Group request are normalized to s_strip size if we set the same via mount
+ * option. If not we set it to s_mb_group_prealloc which can be configured via
+ * /proc/fs/ext4/<partition>/group_prealloc
+ *
+ * XXX: should we try to preallocate more than the group has now?
+ */
+static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac)
+{
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_locality_group *lg = ac->ac_lg;
+
+	BUG_ON(lg == NULL);
+	if (EXT4_SB(sb)->s_stripe)
+		ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_stripe;
+	else
+		ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc;
+	mb_debug("#%u: goal %lu blocks for locality group\n",
+		current->pid, ac->ac_g_ex.fe_len);
+}
+
+/*
+ * Normalization means making request better in terms of
+ * size and alignment
+ */
+static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
+				struct ext4_allocation_request *ar)
+{
+	int bsbits, max;
+	ext4_lblk_t end;
+	struct list_head *cur;
+	loff_t size, orig_size, start_off;
+	ext4_lblk_t start, orig_start;
+	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
+
+	/* do normalize only data requests, metadata requests
+	   do not need preallocation */
+	if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
+		return;
+
+	/* sometime caller may want exact blocks */
+	if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
+		return;
+
+	/* caller may indicate that preallocation isn't
+	 * required (it's a tail, for example) */
+	if (ac->ac_flags & EXT4_MB_HINT_NOPREALLOC)
+		return;
+
+	if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) {
+		ext4_mb_normalize_group_request(ac);
+		return ;
+	}
+
+	bsbits = ac->ac_sb->s_blocksize_bits;
+
+	/* first, let's learn actual file size
+	 * given current request is allocated */
+	size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len;
+	size = size << bsbits;
+	if (size < i_size_read(ac->ac_inode))
+		size = i_size_read(ac->ac_inode);
+
+	/* max available blocks in a free group */
+	max = EXT4_BLOCKS_PER_GROUP(ac->ac_sb) - 1 - 1 -
+				EXT4_SB(ac->ac_sb)->s_itb_per_group;
+
+#define NRL_CHECK_SIZE(req, size, max,bits)	\
+		(req <= (size) || max <= ((size) >> bits))
+
+	/* first, try to predict filesize */
+	/* XXX: should this table be tunable? */
+	start_off = 0;
+	if (size <= 16 * 1024) {
+		size = 16 * 1024;
+	} else if (size <= 32 * 1024) {
+		size = 32 * 1024;
+	} else if (size <= 64 * 1024) {
+		size = 64 * 1024;
+	} else if (size <= 128 * 1024) {
+		size = 128 * 1024;
+	} else if (size <= 256 * 1024) {
+		size = 256 * 1024;
+	} else if (size <= 512 * 1024) {
+		size = 512 * 1024;
+	} else if (size <= 1024 * 1024) {
+		size = 1024 * 1024;
+	} else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, bsbits)) {
+		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
+						(20 - bsbits)) << 20;
+		size = 1024 * 1024;
+	} else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, bsbits)) {
+		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
+							(22 - bsbits)) << 22;
+		size = 4 * 1024 * 1024;
+	} else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len,
+					(8<<20)>>bsbits, max, bsbits)) {
+		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
+							(23 - bsbits)) << 23;
+		size = 8 * 1024 * 1024;
+	} else {
+		start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits;
+		size	  = ac->ac_o_ex.fe_len << bsbits;
+	}
+	orig_size = size = size >> bsbits;
+	orig_start = start = start_off >> bsbits;
+
+	/* don't cover already allocated blocks in selected range */
+	if (ar->pleft && start <= ar->lleft) {
+		size -= ar->lleft + 1 - start;
+		start = ar->lleft + 1;
+	}
+	if (ar->pright && start + size - 1 >= ar->lright)
+		size -= start + size - ar->lright;
+
+	end = start + size;
+
+	/* check we don't cross already preallocated blocks */
+	rcu_read_lock();
+	list_for_each_rcu(cur, &ei->i_prealloc_list) {
+		struct ext4_prealloc_space *pa;
+		unsigned long pa_end;
+
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+
+		if (pa->pa_deleted)
+			continue;
+		spin_lock(&pa->pa_lock);
+		if (pa->pa_deleted) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+
+		pa_end = pa->pa_lstart + pa->pa_len;
+
+		/* PA must not overlap original request */
+		BUG_ON(!(ac->ac_o_ex.fe_logical >= pa_end ||
+			ac->ac_o_ex.fe_logical < pa->pa_lstart));
+
+		/* skip PA normalized request doesn't overlap with */
+		if (pa->pa_lstart >= end) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+		if (pa_end <= start) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+		BUG_ON(pa->pa_lstart <= start && pa_end >= end);
+
+		if (pa_end <= ac->ac_o_ex.fe_logical) {
+			BUG_ON(pa_end < start);
+			start = pa_end;
+		}
+
+		if (pa->pa_lstart > ac->ac_o_ex.fe_logical) {
+			BUG_ON(pa->pa_lstart > end);
+			end = pa->pa_lstart;
+		}
+		spin_unlock(&pa->pa_lock);
+	}
+	rcu_read_unlock();
+	size = end - start;
+
+	/* XXX: extra loop to check we really don't overlap preallocations */
+	rcu_read_lock();
+	list_for_each_rcu(cur, &ei->i_prealloc_list) {
+		struct ext4_prealloc_space *pa;
+		unsigned long pa_end;
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+		spin_lock(&pa->pa_lock);
+		if (pa->pa_deleted == 0) {
+			pa_end = pa->pa_lstart + pa->pa_len;
+			BUG_ON(!(start >= pa_end || end <= pa->pa_lstart));
+		}
+		spin_unlock(&pa->pa_lock);
+	}
+	rcu_read_unlock();
+
+	if (start + size <= ac->ac_o_ex.fe_logical &&
+			start > ac->ac_o_ex.fe_logical) {
+		printk(KERN_ERR "start %lu, size %lu, fe_logical %lu\n",
+			(unsigned long) start, (unsigned long) size,
+			(unsigned long) ac->ac_o_ex.fe_logical);
+	}
+	BUG_ON(start + size <= ac->ac_o_ex.fe_logical &&
+			start > ac->ac_o_ex.fe_logical);
+	BUG_ON(size <= 0 || size >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
+
+	/* now prepare goal request */
+
+	/* XXX: is it better to align blocks WRT to logical
+	 * placement or satisfy big request as is */
+	ac->ac_g_ex.fe_logical = start;
+	ac->ac_g_ex.fe_len = size;
+
+	/* define goal start in order to merge */
+	if (ar->pright && (ar->lright == (start + size))) {
+		/* merge to the right */
+		ext4_get_group_no_and_offset(ac->ac_sb, ar->pright - size,
+						&ac->ac_f_ex.fe_group,
+						&ac->ac_f_ex.fe_start);
+		ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL;
+	}
+	if (ar->pleft && (ar->lleft + 1 == start)) {
+		/* merge to the left */
+		ext4_get_group_no_and_offset(ac->ac_sb, ar->pleft + 1,
+						&ac->ac_f_ex.fe_group,
+						&ac->ac_f_ex.fe_start);
+		ac->ac_flags |= EXT4_MB_HINT_TRY_GOAL;
+	}
+
+	mb_debug("goal: %u(was %u) blocks at %u\n", (unsigned) size,
+		(unsigned) orig_size, (unsigned) start);
+}
+
+static void ext4_mb_collect_stats(struct ext4_allocation_context *ac)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+
+	if (sbi->s_mb_stats && ac->ac_g_ex.fe_len > 1) {
+		atomic_inc(&sbi->s_bal_reqs);
+		atomic_add(ac->ac_b_ex.fe_len, &sbi->s_bal_allocated);
+		if (ac->ac_o_ex.fe_len >= ac->ac_g_ex.fe_len)
+			atomic_inc(&sbi->s_bal_success);
+		atomic_add(ac->ac_found, &sbi->s_bal_ex_scanned);
+		if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start &&
+				ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group)
+			atomic_inc(&sbi->s_bal_goals);
+		if (ac->ac_found > sbi->s_mb_max_to_scan)
+			atomic_inc(&sbi->s_bal_breaks);
+	}
+
+	ext4_mb_store_history(ac);
+}
+
+/*
+ * use blocks preallocated to inode
+ */
+static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac,
+				struct ext4_prealloc_space *pa)
+{
+	ext4_fsblk_t start;
+	ext4_fsblk_t end;
+	int len;
+
+	/* found preallocated blocks, use them */
+	start = pa->pa_pstart + (ac->ac_o_ex.fe_logical - pa->pa_lstart);
+	end = min(pa->pa_pstart + pa->pa_len, start + ac->ac_o_ex.fe_len);
+	len = end - start;
+	ext4_get_group_no_and_offset(ac->ac_sb, start, &ac->ac_b_ex.fe_group,
+					&ac->ac_b_ex.fe_start);
+	ac->ac_b_ex.fe_len = len;
+	ac->ac_status = AC_STATUS_FOUND;
+	ac->ac_pa = pa;
+
+	BUG_ON(start < pa->pa_pstart);
+	BUG_ON(start + len > pa->pa_pstart + pa->pa_len);
+	BUG_ON(pa->pa_free < len);
+	pa->pa_free -= len;
+
+	mb_debug("use %llu/%lu from inode pa %p\n", start, len, pa);
+}
+
+/*
+ * use blocks preallocated to locality group
+ */
+static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
+				struct ext4_prealloc_space *pa)
+{
+	unsigned len = ac->ac_o_ex.fe_len;
+
+	ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart,
+					&ac->ac_b_ex.fe_group,
+					&ac->ac_b_ex.fe_start);
+	ac->ac_b_ex.fe_len = len;
+	ac->ac_status = AC_STATUS_FOUND;
+	ac->ac_pa = pa;
+
+	/* we don't correct pa_pstart or pa_plen here to avoid
+	 * possible race when tte group is being loaded concurrently
+	 * instead we correct pa later, after blocks are marked
+	 * in on-disk bitmap -- see ext4_mb_release_context() */
+	/*
+	 * FIXME!! but the other CPUs can look at this particular
+	 * pa and think that it have enought free blocks if we
+	 * don't update pa_free here right ?
+	 */
+	mb_debug("use %u/%u from group pa %p\n", pa->pa_lstart-len, len, pa);
+}
+
+/*
+ * search goal blocks in preallocated space
+ */
+static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
+{
+	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
+	struct ext4_locality_group *lg;
+	struct ext4_prealloc_space *pa;
+	struct list_head *cur;
+
+	/* only data can be preallocated */
+	if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
+		return 0;
+
+	/* first, try per-file preallocation */
+	rcu_read_lock();
+	list_for_each_rcu(cur, &ei->i_prealloc_list) {
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+
+		/* all fields in this condition don't change,
+		 * so we can skip locking for them */
+		if (ac->ac_o_ex.fe_logical < pa->pa_lstart ||
+			ac->ac_o_ex.fe_logical >= pa->pa_lstart + pa->pa_len)
+			continue;
+
+		/* found preallocated blocks, use them */
+		spin_lock(&pa->pa_lock);
+		if (pa->pa_deleted == 0 && pa->pa_free) {
+			atomic_inc(&pa->pa_count);
+			ext4_mb_use_inode_pa(ac, pa);
+			spin_unlock(&pa->pa_lock);
+			ac->ac_criteria = 10;
+			rcu_read_unlock();
+			return 1;
+		}
+		spin_unlock(&pa->pa_lock);
+	}
+	rcu_read_unlock();
+
+	/* can we use group allocation? */
+	if (!(ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC))
+		return 0;
+
+	/* inode may have no locality group for some reason */
+	lg = ac->ac_lg;
+	if (lg == NULL)
+		return 0;
+
+	rcu_read_lock();
+	list_for_each_rcu(cur, &lg->lg_prealloc_list) {
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+		spin_lock(&pa->pa_lock);
+		if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) {
+			atomic_inc(&pa->pa_count);
+			ext4_mb_use_group_pa(ac, pa);
+			spin_unlock(&pa->pa_lock);
+			ac->ac_criteria = 20;
+			rcu_read_unlock();
+			return 1;
+		}
+		spin_unlock(&pa->pa_lock);
+	}
+	rcu_read_unlock();
+
+	return 0;
+}
+
+/*
+ * the function goes through all preallocation in this group and marks them
+ * used in in-core bitmap. buddy must be generated from this bitmap
+ * Need to be called with ext4 group lock (ext4_lock_group)
+ */
+static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+					ext4_group_t group)
+{
+	struct ext4_group_info *grp = ext4_get_group_info(sb, group);
+	struct ext4_prealloc_space *pa;
+	struct list_head *cur;
+	ext4_group_t groupnr;
+	ext4_grpblk_t start;
+	int preallocated = 0;
+	int count = 0;
+	int len;
+
+	/* all form of preallocation discards first load group,
+	 * so the only competing code is preallocation use.
+	 * we don't need any locking here
+	 * notice we do NOT ignore preallocations with pa_deleted
+	 * otherwise we could leave used blocks available for
+	 * allocation in buddy when concurrent ext4_mb_put_pa()
+	 * is dropping preallocation
+	 */
+	list_for_each(cur, &grp->bb_prealloc_list) {
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list);
+		spin_lock(&pa->pa_lock);
+		ext4_get_group_no_and_offset(sb, pa->pa_pstart,
+					     &groupnr, &start);
+		len = pa->pa_len;
+		spin_unlock(&pa->pa_lock);
+		if (unlikely(len == 0))
+			continue;
+		BUG_ON(groupnr != group);
+		mb_set_bits(sb_bgl_lock(EXT4_SB(sb), group),
+						bitmap, start, len);
+		preallocated += len;
+		count++;
+	}
+	mb_debug("prellocated %u for group %lu\n", preallocated, group);
+}
+
+static void ext4_mb_pa_callback(struct rcu_head *head)
+{
+	struct ext4_prealloc_space *pa;
+	pa = container_of(head, struct ext4_prealloc_space, u.pa_rcu);
+	kmem_cache_free(ext4_pspace_cachep, pa);
+}
+
+/*
+ * drops a reference to preallocated space descriptor
+ * if this was the last reference and the space is consumed
+ */
+static void ext4_mb_put_pa(struct ext4_allocation_context *ac,
+			struct super_block *sb, struct ext4_prealloc_space *pa)
+{
+	unsigned long grp;
+
+	if (!atomic_dec_and_test(&pa->pa_count) || pa->pa_free != 0)
+		return;
+
+	/* in this short window concurrent discard can set pa_deleted */
+	spin_lock(&pa->pa_lock);
+	if (pa->pa_deleted == 1) {
+		spin_unlock(&pa->pa_lock);
+		return;
+	}
+
+	pa->pa_deleted = 1;
+	spin_unlock(&pa->pa_lock);
+
+	/* -1 is to protect from crossing allocation group */
+	ext4_get_group_no_and_offset(sb, pa->pa_pstart - 1, &grp, NULL);
+
+	/*
+	 * possible race:
+	 *
+	 *  P1 (buddy init)			P2 (regular allocation)
+	 *					find block B in PA
+	 *  copy on-disk bitmap to buddy
+	 *  					mark B in on-disk bitmap
+	 *					drop PA from group
+	 *  mark all PAs in buddy
+	 *
+	 * thus, P1 initializes buddy with B available. to prevent this
+	 * we make "copy" and "mark all PAs" atomic and serialize "drop PA"
+	 * against that pair
+	 */
+	ext4_lock_group(sb, grp);
+	list_del(&pa->pa_group_list);
+	ext4_unlock_group(sb, grp);
+
+	spin_lock(pa->pa_obj_lock);
+	list_del_rcu(&pa->pa_inode_list);
+	spin_unlock(pa->pa_obj_lock);
+
+	call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
+}
+
+/*
+ * creates new preallocated space for given inode
+ */
+static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
+{
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_prealloc_space *pa;
+	struct ext4_group_info *grp;
+	struct ext4_inode_info *ei;
+
+	/* preallocate only when found space is larger then requested */
+	BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len);
+	BUG_ON(ac->ac_status != AC_STATUS_FOUND);
+	BUG_ON(!S_ISREG(ac->ac_inode->i_mode));
+
+	pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS);
+	if (pa == NULL)
+		return -ENOMEM;
+
+	if (ac->ac_b_ex.fe_len < ac->ac_g_ex.fe_len) {
+		int winl;
+		int wins;
+		int win;
+		int offs;
+
+		/* we can't allocate as much as normalizer wants.
+		 * so, found space must get proper lstart
+		 * to cover original request */
+		BUG_ON(ac->ac_g_ex.fe_logical > ac->ac_o_ex.fe_logical);
+		BUG_ON(ac->ac_g_ex.fe_len < ac->ac_o_ex.fe_len);
+
+		/* we're limited by original request in that
+		 * logical block must be covered any way
+		 * winl is window we can move our chunk within */
+		winl = ac->ac_o_ex.fe_logical - ac->ac_g_ex.fe_logical;
+
+		/* also, we should cover whole original request */
+		wins = ac->ac_b_ex.fe_len - ac->ac_o_ex.fe_len;
+
+		/* the smallest one defines real window */
+		win = min(winl, wins);
+
+		offs = ac->ac_o_ex.fe_logical % ac->ac_b_ex.fe_len;
+		if (offs && offs < win)
+			win = offs;
+
+		ac->ac_b_ex.fe_logical = ac->ac_o_ex.fe_logical - win;
+		BUG_ON(ac->ac_o_ex.fe_logical < ac->ac_b_ex.fe_logical);
+		BUG_ON(ac->ac_o_ex.fe_len > ac->ac_b_ex.fe_len);
+	}
+
+	/* preallocation can change ac_b_ex, thus we store actually
+	 * allocated blocks for history */
+	ac->ac_f_ex = ac->ac_b_ex;
+
+	pa->pa_lstart = ac->ac_b_ex.fe_logical;
+	pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
+	pa->pa_len = ac->ac_b_ex.fe_len;
+	pa->pa_free = pa->pa_len;
+	atomic_set(&pa->pa_count, 1);
+	spin_lock_init(&pa->pa_lock);
+	pa->pa_deleted = 0;
+	pa->pa_linear = 0;
+
+	mb_debug("new inode pa %p: %llu/%u for %u\n", pa,
+			pa->pa_pstart, pa->pa_len, pa->pa_lstart);
+
+	ext4_mb_use_inode_pa(ac, pa);
+	atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated);
+
+	ei = EXT4_I(ac->ac_inode);
+	grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group);
+
+	pa->pa_obj_lock = &ei->i_prealloc_lock;
+	pa->pa_inode = ac->ac_inode;
+
+	ext4_lock_group(sb, ac->ac_b_ex.fe_group);
+	list_add(&pa->pa_group_list, &grp->bb_prealloc_list);
+	ext4_unlock_group(sb, ac->ac_b_ex.fe_group);
+
+	spin_lock(pa->pa_obj_lock);
+	list_add_rcu(&pa->pa_inode_list, &ei->i_prealloc_list);
+	spin_unlock(pa->pa_obj_lock);
+
+	return 0;
+}
+
+/*
+ * creates new preallocated space for locality group inodes belongs to
+ */
+static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
+{
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_locality_group *lg;
+	struct ext4_prealloc_space *pa;
+	struct ext4_group_info *grp;
+
+	/* preallocate only when found space is larger then requested */
+	BUG_ON(ac->ac_o_ex.fe_len >= ac->ac_b_ex.fe_len);
+	BUG_ON(ac->ac_status != AC_STATUS_FOUND);
+	BUG_ON(!S_ISREG(ac->ac_inode->i_mode));
+
+	BUG_ON(ext4_pspace_cachep == NULL);
+	pa = kmem_cache_alloc(ext4_pspace_cachep, GFP_NOFS);
+	if (pa == NULL)
+		return -ENOMEM;
+
+	/* preallocation can change ac_b_ex, thus we store actually
+	 * allocated blocks for history */
+	ac->ac_f_ex = ac->ac_b_ex;
+
+	pa->pa_pstart = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
+	pa->pa_lstart = pa->pa_pstart;
+	pa->pa_len = ac->ac_b_ex.fe_len;
+	pa->pa_free = pa->pa_len;
+	atomic_set(&pa->pa_count, 1);
+	spin_lock_init(&pa->pa_lock);
+	pa->pa_deleted = 0;
+	pa->pa_linear = 1;
+
+	mb_debug("new group pa %p: %llu/%u for %u\n", pa,
+			pa->pa_pstart, pa->pa_len, pa->pa_lstart);
+
+	ext4_mb_use_group_pa(ac, pa);
+	atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated);
+
+	grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group);
+	lg = ac->ac_lg;
+	BUG_ON(lg == NULL);
+
+	pa->pa_obj_lock = &lg->lg_prealloc_lock;
+	pa->pa_inode = NULL;
+
+	ext4_lock_group(sb, ac->ac_b_ex.fe_group);
+	list_add(&pa->pa_group_list, &grp->bb_prealloc_list);
+	ext4_unlock_group(sb, ac->ac_b_ex.fe_group);
+
+	spin_lock(pa->pa_obj_lock);
+	list_add_tail_rcu(&pa->pa_inode_list, &lg->lg_prealloc_list);
+	spin_unlock(pa->pa_obj_lock);
+
+	return 0;
+}
+
+static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac)
+{
+	int err;
+
+	if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC)
+		err = ext4_mb_new_group_pa(ac);
+	else
+		err = ext4_mb_new_inode_pa(ac);
+	return err;
+}
+
+/*
+ * finds all unused blocks in on-disk bitmap, frees them in
+ * in-core bitmap and buddy.
+ * @pa must be unlinked from inode and group lists, so that
+ * nobody else can find/use it.
+ * the caller MUST hold group/inode locks.
+ * TODO: optimize the case when there are no in-core structures yet
+ */
+static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
+				struct buffer_head *bitmap_bh,
+				struct ext4_prealloc_space *pa)
+{
+	struct ext4_allocation_context ac;
+	struct super_block *sb = e4b->bd_sb;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	unsigned long end;
+	unsigned long next;
+	ext4_group_t group;
+	ext4_grpblk_t bit;
+	sector_t start;
+	int err = 0;
+	int free = 0;
+
+	BUG_ON(pa->pa_deleted == 0);
+	ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit);
+	BUG_ON(group != e4b->bd_group && pa->pa_len != 0);
+	end = bit + pa->pa_len;
+
+	ac.ac_sb = sb;
+	ac.ac_inode = pa->pa_inode;
+	ac.ac_op = EXT4_MB_HISTORY_DISCARD;
+
+	while (bit < end) {
+		bit = ext4_find_next_zero_bit(bitmap_bh->b_data, end, bit);
+		if (bit >= end)
+			break;
+		next = ext4_find_next_bit(bitmap_bh->b_data, end, bit);
+		if (next > end)
+			next = end;
+		start = group * EXT4_BLOCKS_PER_GROUP(sb) + bit +
+				le32_to_cpu(sbi->s_es->s_first_data_block);
+		mb_debug("    free preallocated %u/%u in group %u\n",
+				(unsigned) start, (unsigned) next - bit,
+				(unsigned) group);
+		free += next - bit;
+
+		ac.ac_b_ex.fe_group = group;
+		ac.ac_b_ex.fe_start = bit;
+		ac.ac_b_ex.fe_len = next - bit;
+		ac.ac_b_ex.fe_logical = 0;
+		ext4_mb_store_history(&ac);
+
+		mb_free_blocks(pa->pa_inode, e4b, bit, next - bit);
+		bit = next + 1;
+	}
+	if (free != pa->pa_free) {
+		printk(KERN_ERR "pa %p: logic %lu, phys. %lu, len %lu\n",
+			pa, (unsigned long) pa->pa_lstart,
+			(unsigned long) pa->pa_pstart,
+			(unsigned long) pa->pa_len);
+		printk(KERN_ERR "free %u, pa_free %u\n", free, pa->pa_free);
+	}
+	BUG_ON(free != pa->pa_free);
+	atomic_add(free, &sbi->s_mb_discarded);
+
+	return err;
+}
+
+static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
+				struct ext4_prealloc_space *pa)
+{
+	struct ext4_allocation_context ac;
+	struct super_block *sb = e4b->bd_sb;
+	ext4_group_t group;
+	ext4_grpblk_t bit;
+
+	ac.ac_op = EXT4_MB_HISTORY_DISCARD;
+
+	BUG_ON(pa->pa_deleted == 0);
+	ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit);
+	BUG_ON(group != e4b->bd_group && pa->pa_len != 0);
+	mb_free_blocks(pa->pa_inode, e4b, bit, pa->pa_len);
+	atomic_add(pa->pa_len, &EXT4_SB(sb)->s_mb_discarded);
+
+	ac.ac_sb = sb;
+	ac.ac_inode = NULL;
+	ac.ac_b_ex.fe_group = group;
+	ac.ac_b_ex.fe_start = bit;
+	ac.ac_b_ex.fe_len = pa->pa_len;
+	ac.ac_b_ex.fe_logical = 0;
+	ext4_mb_store_history(&ac);
+
+	return 0;
+}
+
+/*
+ * releases all preallocations in given group
+ *
+ * first, we need to decide discard policy:
+ * - when do we discard
+ *   1) ENOSPC
+ * - how many do we discard
+ *   1) how many requested
+ */
+static int ext4_mb_discard_group_preallocations(struct super_block *sb,
+					ext4_group_t group, int needed)
+{
+	struct ext4_group_info *grp = ext4_get_group_info(sb, group);
+	struct buffer_head *bitmap_bh = NULL;
+	struct ext4_prealloc_space *pa, *tmp;
+	struct list_head list;
+	struct ext4_buddy e4b;
+	int err;
+	int busy = 0;
+	int free = 0;
+
+	mb_debug("discard preallocation for group %lu\n", group);
+
+	if (list_empty(&grp->bb_prealloc_list))
+		return 0;
+
+	bitmap_bh = read_block_bitmap(sb, group);
+	if (bitmap_bh == NULL) {
+		/* error handling here */
+		ext4_mb_release_desc(&e4b);
+		BUG_ON(bitmap_bh == NULL);
+	}
+
+	err = ext4_mb_load_buddy(sb, group, &e4b);
+	BUG_ON(err != 0); /* error handling here */
+
+	if (needed == 0)
+		needed = EXT4_BLOCKS_PER_GROUP(sb) + 1;
+
+	grp = ext4_get_group_info(sb, group);
+	INIT_LIST_HEAD(&list);
+
+repeat:
+	ext4_lock_group(sb, group);
+	list_for_each_entry_safe(pa, tmp,
+				&grp->bb_prealloc_list, pa_group_list) {
+		spin_lock(&pa->pa_lock);
+		if (atomic_read(&pa->pa_count)) {
+			spin_unlock(&pa->pa_lock);
+			busy = 1;
+			continue;
+		}
+		if (pa->pa_deleted) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+
+		/* seems this one can be freed ... */
+		pa->pa_deleted = 1;
+
+		/* we can trust pa_free ... */
+		free += pa->pa_free;
+
+		spin_unlock(&pa->pa_lock);
+
+		list_del(&pa->pa_group_list);
+		list_add(&pa->u.pa_tmp_list, &list);
+	}
+
+	/* if we still need more blocks and some PAs were used, try again */
+	if (free < needed && busy) {
+		busy = 0;
+		ext4_unlock_group(sb, group);
+		/*
+		 * Yield the CPU here so that we don't get soft lockup
+		 * in non preempt case.
+		 */
+		yield();
+		goto repeat;
+	}
+
+	/* found anything to free? */
+	if (list_empty(&list)) {
+		BUG_ON(free != 0);
+		goto out;
+	}
+
+	/* now free all selected PAs */
+	list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) {
+
+		/* remove from object (inode or locality group) */
+		spin_lock(pa->pa_obj_lock);
+		list_del_rcu(&pa->pa_inode_list);
+		spin_unlock(pa->pa_obj_lock);
+
+		if (pa->pa_linear)
+			ext4_mb_release_group_pa(&e4b, pa);
+		else
+			ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa);
+
+		list_del(&pa->u.pa_tmp_list);
+		call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
+	}
+
+out:
+	ext4_unlock_group(sb, group);
+	ext4_mb_release_desc(&e4b);
+	put_bh(bitmap_bh);
+	return free;
+}
+
+/*
+ * releases all non-used preallocated blocks for given inode
+ *
+ * It's important to discard preallocations under i_data_sem
+ * We don't want another block to be served from the prealloc
+ * space when we are discarding the inode prealloc space.
+ *
+ * FIXME!! Make sure it is valid at all the call sites
+ */
+void ext4_mb_discard_inode_preallocations(struct inode *inode)
+{
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	struct super_block *sb = inode->i_sb;
+	struct buffer_head *bitmap_bh = NULL;
+	struct ext4_prealloc_space *pa, *tmp;
+	ext4_group_t group = 0;
+	struct list_head list;
+	struct ext4_buddy e4b;
+	int err;
+
+	if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) {
+		/*BUG_ON(!list_empty(&ei->i_prealloc_list));*/
+		return;
+	}
+
+	mb_debug("discard preallocation for inode %lu\n", inode->i_ino);
+
+	INIT_LIST_HEAD(&list);
+
+repeat:
+	/* first, collect all pa's in the inode */
+	spin_lock(&ei->i_prealloc_lock);
+	while (!list_empty(&ei->i_prealloc_list)) {
+		pa = list_entry(ei->i_prealloc_list.next,
+				struct ext4_prealloc_space, pa_inode_list);
+		BUG_ON(pa->pa_obj_lock != &ei->i_prealloc_lock);
+		spin_lock(&pa->pa_lock);
+		if (atomic_read(&pa->pa_count)) {
+			/* this shouldn't happen often - nobody should
+			 * use preallocation while we're discarding it */
+			spin_unlock(&pa->pa_lock);
+			spin_unlock(&ei->i_prealloc_lock);
+			printk(KERN_ERR "uh-oh! used pa while discarding\n");
+			WARN_ON(1);
+			schedule_timeout_uninterruptible(HZ);
+			goto repeat;
+
+		}
+		if (pa->pa_deleted == 0) {
+			pa->pa_deleted = 1;
+			spin_unlock(&pa->pa_lock);
+			list_del_rcu(&pa->pa_inode_list);
+			list_add(&pa->u.pa_tmp_list, &list);
+			continue;
+		}
+
+		/* someone is deleting pa right now */
+		spin_unlock(&pa->pa_lock);
+		spin_unlock(&ei->i_prealloc_lock);
+
+		/* we have to wait here because pa_deleted
+		 * doesn't mean pa is already unlinked from
+		 * the list. as we might be called from
+		 * ->clear_inode() the inode will get freed
+		 * and concurrent thread which is unlinking
+		 * pa from inode's list may access already
+		 * freed memory, bad-bad-bad */
+
+		/* XXX: if this happens too often, we can
+		 * add a flag to force wait only in case
+		 * of ->clear_inode(), but not in case of
+		 * regular truncate */
+		schedule_timeout_uninterruptible(HZ);
+		goto repeat;
+	}
+	spin_unlock(&ei->i_prealloc_lock);
+
+	list_for_each_entry_safe(pa, tmp, &list, u.pa_tmp_list) {
+		BUG_ON(pa->pa_linear != 0);
+		ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL);
+
+		err = ext4_mb_load_buddy(sb, group, &e4b);
+		BUG_ON(err != 0); /* error handling here */
+
+		bitmap_bh = read_block_bitmap(sb, group);
+		if (bitmap_bh == NULL) {
+			/* error handling here */
+			ext4_mb_release_desc(&e4b);
+			BUG_ON(bitmap_bh == NULL);
+		}
+
+		ext4_lock_group(sb, group);
+		list_del(&pa->pa_group_list);
+		ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa);
+		ext4_unlock_group(sb, group);
+
+		ext4_mb_release_desc(&e4b);
+		put_bh(bitmap_bh);
+
+		list_del(&pa->u.pa_tmp_list);
+		call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
+	}
+}
+
+/*
+ * finds all preallocated spaces and return blocks being freed to them
+ * if preallocated space becomes full (no block is used from the space)
+ * then the function frees space in buddy
+ * XXX: at the moment, truncate (which is the only way to free blocks)
+ * discards all preallocations
+ */
+static void ext4_mb_return_to_preallocation(struct inode *inode,
+					struct ext4_buddy *e4b,
+					sector_t block, int count)
+{
+	BUG_ON(!list_empty(&EXT4_I(inode)->i_prealloc_list));
+}
+#ifdef MB_DEBUG
+static void ext4_mb_show_ac(struct ext4_allocation_context *ac)
+{
+	struct super_block *sb = ac->ac_sb;
+	ext4_group_t i;
+
+	printk(KERN_ERR "EXT4-fs: Can't allocate:"
+			" Allocation context details:\n");
+	printk(KERN_ERR "EXT4-fs: status %d flags %d\n",
+			ac->ac_status, ac->ac_flags);
+	printk(KERN_ERR "EXT4-fs: orig %lu/%lu/%lu@%lu, goal %lu/%lu/%lu@%lu, "
+			"best %lu/%lu/%lu@%lu cr %d\n",
+			(unsigned long)ac->ac_o_ex.fe_group,
+			(unsigned long)ac->ac_o_ex.fe_start,
+			(unsigned long)ac->ac_o_ex.fe_len,
+			(unsigned long)ac->ac_o_ex.fe_logical,
+			(unsigned long)ac->ac_g_ex.fe_group,
+			(unsigned long)ac->ac_g_ex.fe_start,
+			(unsigned long)ac->ac_g_ex.fe_len,
+			(unsigned long)ac->ac_g_ex.fe_logical,
+			(unsigned long)ac->ac_b_ex.fe_group,
+			(unsigned long)ac->ac_b_ex.fe_start,
+			(unsigned long)ac->ac_b_ex.fe_len,
+			(unsigned long)ac->ac_b_ex.fe_logical,
+			(int)ac->ac_criteria);
+	printk(KERN_ERR "EXT4-fs: %lu scanned, %d found\n", ac->ac_ex_scanned,
+		ac->ac_found);
+	printk(KERN_ERR "EXT4-fs: groups: \n");
+	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
+		struct ext4_group_info *grp = ext4_get_group_info(sb, i);
+		struct ext4_prealloc_space *pa;
+		ext4_grpblk_t start;
+		struct list_head *cur;
+		ext4_lock_group(sb, i);
+		list_for_each(cur, &grp->bb_prealloc_list) {
+			pa = list_entry(cur, struct ext4_prealloc_space,
+					pa_group_list);
+			spin_lock(&pa->pa_lock);
+			ext4_get_group_no_and_offset(sb, pa->pa_pstart,
+						     NULL, &start);
+			spin_unlock(&pa->pa_lock);
+			printk(KERN_ERR "PA:%lu:%d:%u \n", i,
+							start, pa->pa_len);
+		}
+		ext4_lock_group(sb, i);
+
+		if (grp->bb_free == 0)
+			continue;
+		printk(KERN_ERR "%lu: %d/%d \n",
+		       i, grp->bb_free, grp->bb_fragments);
+	}
+	printk(KERN_ERR "\n");
+}
+#else
+static inline void ext4_mb_show_ac(struct ext4_allocation_context *ac)
+{
+	return;
+}
+#endif
+
+/*
+ * We use locality group preallocation for small size file. The size of the
+ * file is determined by the current size or the resulting size after
+ * allocation which ever is larger
+ *
+ * One can tune this size via /proc/fs/ext4/<partition>/stream_req
+ */
+static void ext4_mb_group_or_file(struct ext4_allocation_context *ac)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
+	int bsbits = ac->ac_sb->s_blocksize_bits;
+	loff_t size, isize;
+
+	if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
+		return;
+
+	size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len;
+	isize = i_size_read(ac->ac_inode) >> bsbits;
+	size = max(size, isize);
+
+	/* don't use group allocation for large files */
+	if (size >= sbi->s_mb_stream_request)
+		return;
+
+	if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
+		return;
+
+	BUG_ON(ac->ac_lg != NULL);
+	/*
+	 * locality group prealloc space are per cpu. The reason for having
+	 * per cpu locality group is to reduce the contention between block
+	 * request from multiple CPUs.
+	 */
+	ac->ac_lg = &sbi->s_locality_groups[get_cpu()];
+	put_cpu();
+
+	/* we're going to use group allocation */
+	ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC;
+
+	/* serialize all allocations in the group */
+	mutex_lock(&ac->ac_lg->lg_mutex);
+}
+
+static int ext4_mb_initialize_context(struct ext4_allocation_context *ac,
+				struct ext4_allocation_request *ar)
+{
+	struct super_block *sb = ar->inode->i_sb;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct ext4_super_block *es = sbi->s_es;
+	ext4_group_t group;
+	unsigned long len;
+	unsigned long goal;
+	ext4_grpblk_t block;
+
+	/* we can't allocate > group size */
+	len = ar->len;
+
+	/* just a dirty hack to filter too big requests  */
+	if (len >= EXT4_BLOCKS_PER_GROUP(sb) - 10)
+		len = EXT4_BLOCKS_PER_GROUP(sb) - 10;
+
+	/* start searching from the goal */
+	goal = ar->goal;
+	if (goal < le32_to_cpu(es->s_first_data_block) ||
+			goal >= ext4_blocks_count(es))
+		goal = le32_to_cpu(es->s_first_data_block);
+	ext4_get_group_no_and_offset(sb, goal, &group, &block);
+
+	/* set up allocation goals */
+	ac->ac_b_ex.fe_logical = ar->logical;
+	ac->ac_b_ex.fe_group = 0;
+	ac->ac_b_ex.fe_start = 0;
+	ac->ac_b_ex.fe_len = 0;
+	ac->ac_status = AC_STATUS_CONTINUE;
+	ac->ac_groups_scanned = 0;
+	ac->ac_ex_scanned = 0;
+	ac->ac_found = 0;
+	ac->ac_sb = sb;
+	ac->ac_inode = ar->inode;
+	ac->ac_o_ex.fe_logical = ar->logical;
+	ac->ac_o_ex.fe_group = group;
+	ac->ac_o_ex.fe_start = block;
+	ac->ac_o_ex.fe_len = len;
+	ac->ac_g_ex.fe_logical = ar->logical;
+	ac->ac_g_ex.fe_group = group;
+	ac->ac_g_ex.fe_start = block;
+	ac->ac_g_ex.fe_len = len;
+	ac->ac_f_ex.fe_len = 0;
+	ac->ac_flags = ar->flags;
+	ac->ac_2order = 0;
+	ac->ac_criteria = 0;
+	ac->ac_pa = NULL;
+	ac->ac_bitmap_page = NULL;
+	ac->ac_buddy_page = NULL;
+	ac->ac_lg = NULL;
+
+	/* we have to define context: we'll we work with a file or
+	 * locality group. this is a policy, actually */
+	ext4_mb_group_or_file(ac);
+
+	mb_debug("init ac: %u blocks @ %u, goal %u, flags %x, 2^%d, "
+			"left: %u/%u, right %u/%u to %swritable\n",
+			(unsigned) ar->len, (unsigned) ar->logical,
+			(unsigned) ar->goal, ac->ac_flags, ac->ac_2order,
+			(unsigned) ar->lleft, (unsigned) ar->pleft,
+			(unsigned) ar->lright, (unsigned) ar->pright,
+			atomic_read(&ar->inode->i_writecount) ? "" : "non-");
+	return 0;
+
+}
+
+/*
+ * release all resource we used in allocation
+ */
+static int ext4_mb_release_context(struct ext4_allocation_context *ac)
+{
+	if (ac->ac_pa) {
+		if (ac->ac_pa->pa_linear) {
+			/* see comment in ext4_mb_use_group_pa() */
+			spin_lock(&ac->ac_pa->pa_lock);
+			ac->ac_pa->pa_pstart += ac->ac_b_ex.fe_len;
+			ac->ac_pa->pa_lstart += ac->ac_b_ex.fe_len;
+			ac->ac_pa->pa_free -= ac->ac_b_ex.fe_len;
+			ac->ac_pa->pa_len -= ac->ac_b_ex.fe_len;
+			spin_unlock(&ac->ac_pa->pa_lock);
+		}
+		ext4_mb_put_pa(ac, ac->ac_sb, ac->ac_pa);
+	}
+	if (ac->ac_bitmap_page)
+		page_cache_release(ac->ac_bitmap_page);
+	if (ac->ac_buddy_page)
+		page_cache_release(ac->ac_buddy_page);
+	if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC)
+		mutex_unlock(&ac->ac_lg->lg_mutex);
+	ext4_mb_collect_stats(ac);
+	return 0;
+}
+
+static int ext4_mb_discard_preallocations(struct super_block *sb, int needed)
+{
+	ext4_group_t i;
+	int ret;
+	int freed = 0;
+
+	for (i = 0; i < EXT4_SB(sb)->s_groups_count && needed > 0; i++) {
+		ret = ext4_mb_discard_group_preallocations(sb, i, needed);
+		freed += ret;
+		needed -= ret;
+	}
+
+	return freed;
+}
+
+/*
+ * Main entry point into mballoc to allocate blocks
+ * it tries to use preallocation first, then falls back
+ * to usual allocation
+ */
+ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
+				 struct ext4_allocation_request *ar, int *errp)
+{
+	struct ext4_allocation_context ac;
+	struct ext4_sb_info *sbi;
+	struct super_block *sb;
+	ext4_fsblk_t block = 0;
+	int freed;
+	int inquota;
+
+	sb = ar->inode->i_sb;
+	sbi = EXT4_SB(sb);
+
+	if (!test_opt(sb, MBALLOC)) {
+		block = ext4_new_blocks_old(handle, ar->inode, ar->goal,
+					    &(ar->len), errp);
+		return block;
+	}
+
+	while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) {
+		ar->flags |= EXT4_MB_HINT_NOPREALLOC;
+		ar->len--;
+	}
+	if (ar->len == 0) {
+		*errp = -EDQUOT;
+		return 0;
+	}
+	inquota = ar->len;
+
+	ext4_mb_poll_new_transaction(sb, handle);
+
+	*errp = ext4_mb_initialize_context(&ac, ar);
+	if (*errp) {
+		ar->len = 0;
+		goto out;
+	}
+
+	ac.ac_op = EXT4_MB_HISTORY_PREALLOC;
+	if (!ext4_mb_use_preallocated(&ac)) {
+
+		ac.ac_op = EXT4_MB_HISTORY_ALLOC;
+		ext4_mb_normalize_request(&ac, ar);
+
+repeat:
+		/* allocate space in core */
+		ext4_mb_regular_allocator(&ac);
+
+		/* as we've just preallocated more space than
+		 * user requested orinally, we store allocated
+		 * space in a special descriptor */
+		if (ac.ac_status == AC_STATUS_FOUND &&
+				ac.ac_o_ex.fe_len < ac.ac_b_ex.fe_len)
+			ext4_mb_new_preallocation(&ac);
+	}
+
+	if (likely(ac.ac_status == AC_STATUS_FOUND)) {
+		ext4_mb_mark_diskspace_used(&ac, handle);
+		*errp = 0;
+		block = ext4_grp_offs_to_block(sb, &ac.ac_b_ex);
+		ar->len = ac.ac_b_ex.fe_len;
+	} else {
+		freed  = ext4_mb_discard_preallocations(sb, ac.ac_o_ex.fe_len);
+		if (freed)
+			goto repeat;
+		*errp = -ENOSPC;
+		ac.ac_b_ex.fe_len = 0;
+		ar->len = 0;
+		ext4_mb_show_ac(&ac);
+	}
+
+	ext4_mb_release_context(&ac);
+
+out:
+	if (ar->len < inquota)
+		DQUOT_FREE_BLOCK(ar->inode, inquota - ar->len);
+
+	return block;
+}
+static void ext4_mb_poll_new_transaction(struct super_block *sb,
+						handle_t *handle)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	if (sbi->s_last_transaction == handle->h_transaction->t_tid)
+		return;
+
+	/* new transaction! time to close last one and free blocks for
+	 * committed transaction. we know that only transaction can be
+	 * active, so previos transaction can be being logged and we
+	 * know that transaction before previous is known to be already
+	 * logged. this means that now we may free blocks freed in all
+	 * transactions before previous one. hope I'm clear enough ... */
+
+	spin_lock(&sbi->s_md_lock);
+	if (sbi->s_last_transaction != handle->h_transaction->t_tid) {
+		mb_debug("new transaction %lu, old %lu\n",
+				(unsigned long) handle->h_transaction->t_tid,
+				(unsigned long) sbi->s_last_transaction);
+		list_splice_init(&sbi->s_closed_transaction,
+				&sbi->s_committed_transaction);
+		list_splice_init(&sbi->s_active_transaction,
+				&sbi->s_closed_transaction);
+		sbi->s_last_transaction = handle->h_transaction->t_tid;
+	}
+	spin_unlock(&sbi->s_md_lock);
+
+	ext4_mb_free_committed_blocks(sb);
+}
+
+static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
+			  ext4_group_t group, ext4_grpblk_t block, int count)
+{
+	struct ext4_group_info *db = e4b->bd_info;
+	struct super_block *sb = e4b->bd_sb;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct ext4_free_metadata *md;
+	int i;
+
+	BUG_ON(e4b->bd_bitmap_page == NULL);
+	BUG_ON(e4b->bd_buddy_page == NULL);
+
+	ext4_lock_group(sb, group);
+	for (i = 0; i < count; i++) {
+		md = db->bb_md_cur;
+		if (md && db->bb_tid != handle->h_transaction->t_tid) {
+			db->bb_md_cur = NULL;
+			md = NULL;
+		}
+
+		if (md == NULL) {
+			ext4_unlock_group(sb, group);
+			md = kmalloc(sizeof(*md), GFP_NOFS);
+			if (md == NULL)
+				return -ENOMEM;
+			md->num = 0;
+			md->group = group;
+
+			ext4_lock_group(sb, group);
+			if (db->bb_md_cur == NULL) {
+				spin_lock(&sbi->s_md_lock);
+				list_add(&md->list, &sbi->s_active_transaction);
+				spin_unlock(&sbi->s_md_lock);
+				/* protect buddy cache from being freed,
+				 * otherwise we'll refresh it from
+				 * on-disk bitmap and lose not-yet-available
+				 * blocks */
+				page_cache_get(e4b->bd_buddy_page);
+				page_cache_get(e4b->bd_bitmap_page);
+				db->bb_md_cur = md;
+				db->bb_tid = handle->h_transaction->t_tid;
+				mb_debug("new md 0x%p for group %lu\n",
+						md, md->group);
+			} else {
+				kfree(md);
+				md = db->bb_md_cur;
+			}
+		}
+
+		BUG_ON(md->num >= EXT4_BB_MAX_BLOCKS);
+		md->blocks[md->num] = block + i;
+		md->num++;
+		if (md->num == EXT4_BB_MAX_BLOCKS) {
+			/* no more space, put full container on a sb's list */
+			db->bb_md_cur = NULL;
+		}
+	}
+	ext4_unlock_group(sb, group);
+	return 0;
+}
+
+/*
+ * Main entry point into mballoc to free blocks
+ */
+void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
+			unsigned long block, unsigned long count,
+			int metadata, unsigned long *freed)
+{
+	struct buffer_head *bitmap_bh = 0;
+	struct super_block *sb = inode->i_sb;
+	struct ext4_allocation_context ac;
+	struct ext4_group_desc *gdp;
+	struct ext4_super_block *es;
+	unsigned long overflow;
+	ext4_grpblk_t bit;
+	struct buffer_head *gd_bh;
+	ext4_group_t block_group;
+	struct ext4_sb_info *sbi;
+	struct ext4_buddy e4b;
+	int err = 0;
+	int ret;
+
+	*freed = 0;
+
+	ext4_mb_poll_new_transaction(sb, handle);
+
+	sbi = EXT4_SB(sb);
+	es = EXT4_SB(sb)->s_es;
+	if (block < le32_to_cpu(es->s_first_data_block) ||
+	    block + count < block ||
+	    block + count > ext4_blocks_count(es)) {
+		ext4_error(sb, __FUNCTION__,
+			    "Freeing blocks not in datazone - "
+			    "block = %lu, count = %lu", block, count);
+		goto error_return;
+	}
+
+	ext4_debug("freeing block %lu\n", block);
+
+	ac.ac_op = EXT4_MB_HISTORY_FREE;
+	ac.ac_inode = inode;
+	ac.ac_sb = sb;
+
+do_more:
+	overflow = 0;
+	ext4_get_group_no_and_offset(sb, block, &block_group, &bit);
+
+	/*
+	 * Check to see if we are freeing blocks across a group
+	 * boundary.
+	 */
+	if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) {
+		overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb);
+		count -= overflow;
+	}
+	bitmap_bh = read_block_bitmap(sb, block_group);
+	if (!bitmap_bh)
+		goto error_return;
+	gdp = ext4_get_group_desc(sb, block_group, &gd_bh);
+	if (!gdp)
+		goto error_return;
+
+	if (in_range(ext4_block_bitmap(sb, gdp), block, count) ||
+	    in_range(ext4_inode_bitmap(sb, gdp), block, count) ||
+	    in_range(block, ext4_inode_table(sb, gdp),
+		      EXT4_SB(sb)->s_itb_per_group) ||
+	    in_range(block + count - 1, ext4_inode_table(sb, gdp),
+		      EXT4_SB(sb)->s_itb_per_group)) {
+
+		ext4_error(sb, __FUNCTION__,
+			   "Freeing blocks in system zone - "
+			   "Block = %lu, count = %lu", block, count);
+	}
+
+	BUFFER_TRACE(bitmap_bh, "getting write access");
+	err = ext4_journal_get_write_access(handle, bitmap_bh);
+	if (err)
+		goto error_return;
+
+	/*
+	 * We are about to modify some metadata.  Call the journal APIs
+	 * to unshare ->b_data if a currently-committing transaction is
+	 * using it
+	 */
+	BUFFER_TRACE(gd_bh, "get_write_access");
+	err = ext4_journal_get_write_access(handle, gd_bh);
+	if (err)
+		goto error_return;
+
+	err = ext4_mb_load_buddy(sb, block_group, &e4b);
+	if (err)
+		goto error_return;
+
+#ifdef AGGRESSIVE_CHECK
+	{
+		int i;
+		for (i = 0; i < count; i++)
+			BUG_ON(!mb_test_bit(bit + i, bitmap_bh->b_data));
+	}
+#endif
+	mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data,
+			bit, count);
+
+	/* We dirtied the bitmap block */
+	BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
+	err = ext4_journal_dirty_metadata(handle, bitmap_bh);
+
+	ac.ac_b_ex.fe_group = block_group;
+	ac.ac_b_ex.fe_start = bit;
+	ac.ac_b_ex.fe_len = count;
+	ext4_mb_store_history(&ac);
+
+	if (metadata) {
+		/* blocks being freed are metadata. these blocks shouldn't
+		 * be used until this transaction is committed */
+		ext4_mb_free_metadata(handle, &e4b, block_group, bit, count);
+	} else {
+		ext4_lock_group(sb, block_group);
+		err = mb_free_blocks(inode, &e4b, bit, count);
+		ext4_mb_return_to_preallocation(inode, &e4b, block, count);
+		ext4_unlock_group(sb, block_group);
+		BUG_ON(err != 0);
+	}
+
+	spin_lock(sb_bgl_lock(sbi, block_group));
+	gdp->bg_free_blocks_count =
+		cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count);
+	gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp);
+	spin_unlock(sb_bgl_lock(sbi, block_group));
+	percpu_counter_add(&sbi->s_freeblocks_counter, count);
+
+	ext4_mb_release_desc(&e4b);
+
+	*freed += count;
+
+	/* And the group descriptor block */
+	BUFFER_TRACE(gd_bh, "dirtied group descriptor block");
+	ret = ext4_journal_dirty_metadata(handle, gd_bh);
+	if (!err)
+		err = ret;
+
+	if (overflow && !err) {
+		block += count;
+		count = overflow;
+		put_bh(bitmap_bh);
+		goto do_more;
+	}
+	sb->s_dirt = 1;
+error_return:
+	brelse(bitmap_bh);
+	ext4_std_error(sb, err);
+	return;
+}
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index ec7cb567a7da..3ebc2332f52e 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -236,10 +236,10 @@ static int free_dind_blocks(handle_t *handle,
 	for (i = 0; i < max_entries; i++) {
 		if (tmp_idata[i])
 			ext4_free_blocks(handle, inode,
-					le32_to_cpu(tmp_idata[i]), 1);
+					le32_to_cpu(tmp_idata[i]), 1, 1);
 	}
 	put_bh(bh);
-	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1);
+	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1);
 	return 0;
 }
 
@@ -267,7 +267,7 @@ static int free_tind_blocks(handle_t *handle,
 		}
 	}
 	put_bh(bh);
-	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1);
+	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1);
 	return 0;
 }
 
@@ -278,7 +278,7 @@ static int free_ind_block(handle_t *handle, struct inode *inode)
 
 	if (ei->i_data[EXT4_IND_BLOCK])
 		ext4_free_blocks(handle, inode,
-				le32_to_cpu(ei->i_data[EXT4_IND_BLOCK]), 1);
+				le32_to_cpu(ei->i_data[EXT4_IND_BLOCK]), 1, 1);
 
 	if (ei->i_data[EXT4_DIND_BLOCK]) {
 		retval = free_dind_blocks(handle, inode,
@@ -365,7 +365,7 @@ static int free_ext_idx(handle_t *handle, struct inode *inode,
 		}
 	}
 	put_bh(bh);
-	ext4_free_blocks(handle, inode, block, 1);
+	ext4_free_blocks(handle, inode, block, 1, 1);
 	return retval;
 }
 
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 64fc7f111734..3a51ffc47790 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -503,6 +503,7 @@ static void ext4_put_super (struct super_block * sb)
 	struct ext4_super_block *es = sbi->s_es;
 	int i;
 
+	ext4_mb_release(sb);
 	ext4_ext_release(sb);
 	ext4_xattr_put_super(sb);
 	jbd2_journal_destroy(sbi->s_journal);
@@ -569,6 +570,8 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei->i_block_alloc_info = NULL;
 	ei->vfs_inode.i_version = 1;
 	memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache));
+	INIT_LIST_HEAD(&ei->i_prealloc_list);
+	spin_lock_init(&ei->i_prealloc_lock);
 	return &ei->vfs_inode;
 }
 
@@ -881,6 +884,7 @@ enum {
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
 	Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
 	Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version,
+	Opt_mballoc, Opt_nomballoc, Opt_stripe,
 };
 
 static match_table_t tokens = {
@@ -935,6 +939,9 @@ static match_table_t tokens = {
 	{Opt_extents, "extents"},
 	{Opt_noextents, "noextents"},
 	{Opt_i_version, "i_version"},
+	{Opt_mballoc, "mballoc"},
+	{Opt_nomballoc, "nomballoc"},
+	{Opt_stripe, "stripe=%u"},
 	{Opt_err, NULL},
 	{Opt_resize, "resize"},
 };
@@ -1284,6 +1291,19 @@ clear_qf_name:
 			set_opt(sbi->s_mount_opt, I_VERSION);
 			sb->s_flags |= MS_I_VERSION;
 			break;
+		case Opt_mballoc:
+			set_opt(sbi->s_mount_opt, MBALLOC);
+			break;
+		case Opt_nomballoc:
+			clear_opt(sbi->s_mount_opt, MBALLOC);
+			break;
+		case Opt_stripe:
+			if (match_int(&args[0], &option))
+				return 0;
+			if (option < 0)
+				return 0;
+			sbi->s_stripe = option;
+			break;
 		default:
 			printk (KERN_ERR
 				"EXT4-fs: Unrecognized mount option \"%s\" "
@@ -1742,6 +1762,34 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb,
 	return (has_super + ext4_group_first_block_no(sb, bg));
 }
 
+/**
+ * ext4_get_stripe_size: Get the stripe size.
+ * @sbi: In memory super block info
+ *
+ * If we have specified it via mount option, then
+ * use the mount option value. If the value specified at mount time is
+ * greater than the blocks per group use the super block value.
+ * If the super block value is greater than blocks per group return 0.
+ * Allocator needs it be less than blocks per group.
+ *
+ */
+static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi)
+{
+	unsigned long stride = le16_to_cpu(sbi->s_es->s_raid_stride);
+	unsigned long stripe_width =
+			le32_to_cpu(sbi->s_es->s_raid_stripe_width);
+
+	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
+		return sbi->s_stripe;
+
+	if (stripe_width <= sbi->s_blocks_per_group)
+		return stripe_width;
+
+	if (stride <= sbi->s_blocks_per_group)
+		return stride;
+
+	return 0;
+}
 
 static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 				__releases(kernel_sem)
@@ -2091,6 +2139,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	sbi->s_rsv_window_head.rsv_goal_size = 0;
 	ext4_rsv_window_add(sb, &sbi->s_rsv_window_head);
 
+	sbi->s_stripe = ext4_get_stripe_size(sbi);
+
 	/*
 	 * set up enough so that it can read an inode
 	 */
@@ -2250,6 +2300,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		"writeback");
 
 	ext4_ext_init(sb);
+	ext4_mb_init(sb, needs_recovery);
 
 	lock_kernel();
 	return 0;
@@ -3232,9 +3283,15 @@ static struct file_system_type ext4dev_fs_type = {
 
 static int __init init_ext4_fs(void)
 {
-	int err = init_ext4_xattr();
+	int err;
+
+	err = init_ext4_mballoc();
 	if (err)
 		return err;
+
+	err = init_ext4_xattr();
+	if (err)
+		goto out2;
 	err = init_inodecache();
 	if (err)
 		goto out1;
@@ -3246,6 +3303,8 @@ out:
 	destroy_inodecache();
 out1:
 	exit_ext4_xattr();
+out2:
+	exit_ext4_mballoc();
 	return err;
 }
 
@@ -3254,6 +3313,7 @@ static void __exit exit_ext4_fs(void)
 	unregister_filesystem(&ext4dev_fs_type);
 	destroy_inodecache();
 	exit_ext4_xattr();
+	exit_ext4_mballoc();
 }
 
 MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 86387302c2a9..d7962139c010 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -480,7 +480,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 		ea_bdebug(bh, "refcount now=0; freeing");
 		if (ce)
 			mb_cache_entry_free(ce);
-		ext4_free_blocks(handle, inode, bh->b_blocknr, 1);
+		ext4_free_blocks(handle, inode, bh->b_blocknr, 1, 1);
 		get_bh(bh);
 		ext4_forget(handle, 1, inode, bh, bh->b_blocknr);
 	} else {
@@ -821,7 +821,7 @@ inserted:
 			new_bh = sb_getblk(sb, block);
 			if (!new_bh) {
 getblk_failed:
-				ext4_free_blocks(handle, inode, block, 1);
+				ext4_free_blocks(handle, inode, block, 1, 1);
 				error = -EIO;
 				goto cleanup;
 			}
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index d0b7ca99b91f..1852313fc7c7 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -20,6 +20,8 @@
 #include <linux/blkdev.h>
 #include <linux/magic.h>
 
+#include <linux/ext4_fs_i.h>
+
 /*
  * The second extended filesystem constants/structures
  */
@@ -51,6 +53,50 @@
 #define ext4_debug(f, a...)	do {} while (0)
 #endif
 
+#define EXT4_MULTIBLOCK_ALLOCATOR	1
+
+/* prefer goal again. length */
+#define EXT4_MB_HINT_MERGE		1
+/* blocks already reserved */
+#define EXT4_MB_HINT_RESERVED		2
+/* metadata is being allocated */
+#define EXT4_MB_HINT_METADATA		4
+/* first blocks in the file */
+#define EXT4_MB_HINT_FIRST		8
+/* search for the best chunk */
+#define EXT4_MB_HINT_BEST		16
+/* data is being allocated */
+#define EXT4_MB_HINT_DATA		32
+/* don't preallocate (for tails) */
+#define EXT4_MB_HINT_NOPREALLOC		64
+/* allocate for locality group */
+#define EXT4_MB_HINT_GROUP_ALLOC	128
+/* allocate goal blocks or none */
+#define EXT4_MB_HINT_GOAL_ONLY		256
+/* goal is meaningful */
+#define EXT4_MB_HINT_TRY_GOAL		512
+
+struct ext4_allocation_request {
+	/* target inode for block we're allocating */
+	struct inode *inode;
+	/* logical block in target inode */
+	ext4_lblk_t logical;
+	/* phys. target (a hint) */
+	ext4_fsblk_t goal;
+	/* the closest logical allocated block to the left */
+	ext4_lblk_t lleft;
+	/* phys. block for ^^^ */
+	ext4_fsblk_t pleft;
+	/* the closest logical allocated block to the right */
+	ext4_lblk_t lright;
+	/* phys. block for ^^^ */
+	ext4_fsblk_t pright;
+	/* how many blocks we want to allocate */
+	unsigned long len;
+	/* flags. see above EXT4_MB_HINT_* */
+	unsigned long flags;
+};
+
 /*
  * Special inodes numbers
  */
@@ -474,6 +520,7 @@ do {									       \
 #define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
 #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
 #define EXT4_MOUNT_I_VERSION            0x2000000 /* i_version support */
+#define EXT4_MOUNT_MBALLOC		0x4000000 /* Buddy allocation support */
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
 #define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
@@ -912,7 +959,7 @@ extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode,
 extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, unsigned long *count, int *errp);
 extern void ext4_free_blocks (handle_t *handle, struct inode *inode,
-			ext4_fsblk_t block, unsigned long count);
+			ext4_fsblk_t block, unsigned long count, int metadata);
 extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb,
 				 ext4_fsblk_t block, unsigned long count,
 				unsigned long *pdquot_freed_blocks);
@@ -950,6 +997,20 @@ extern unsigned long ext4_count_dirs (struct super_block *);
 extern void ext4_check_inodes_bitmap (struct super_block *);
 extern unsigned long ext4_count_free (struct buffer_head *, unsigned);
 
+/* mballoc.c */
+extern long ext4_mb_stats;
+extern long ext4_mb_max_to_scan;
+extern int ext4_mb_init(struct super_block *, int);
+extern int ext4_mb_release(struct super_block *);
+extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *,
+				struct ext4_allocation_request *, int *);
+extern int ext4_mb_reserve_blocks(struct super_block *, int);
+extern void ext4_mb_discard_inode_preallocations(struct inode *);
+extern int __init init_ext4_mballoc(void);
+extern void exit_ext4_mballoc(void);
+extern void ext4_mb_free_blocks(handle_t *, struct inode *,
+		unsigned long, unsigned long, int, unsigned long *);
+
 
 /* inode.c */
 int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
@@ -1080,6 +1141,19 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size)
 	raw_inode->i_size_high = cpu_to_le32(i_size >> 32);
 }
 
+static inline
+struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
+							ext4_group_t group)
+{
+	 struct ext4_group_info ***grp_info;
+	 long indexv, indexh;
+	 grp_info = EXT4_SB(sb)->s_group_info;
+	 indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb));
+	 indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1);
+	 return grp_info[indexv][indexh];
+}
+
+
 #define ext4_std_error(sb, errno)				\
 do {								\
 	if ((errno))						\
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index 4377d249d378..d5508d3cf290 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -158,6 +158,10 @@ struct ext4_inode_info {
 	 * struct timespec i_{a,c,m}time in the generic inode.
 	 */
 	struct timespec i_crtime;
+
+	/* mballoc */
+	struct list_head i_prealloc_list;
+	spinlock_t i_prealloc_lock;
 };
 
 #endif	/* _LINUX_EXT4_FS_I */
diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h
index 38a47ec06df9..abaae2c8cccf 100644
--- a/include/linux/ext4_fs_sb.h
+++ b/include/linux/ext4_fs_sb.h
@@ -91,6 +91,58 @@ struct ext4_sb_info {
 	unsigned long s_ext_blocks;
 	unsigned long s_ext_extents;
 #endif
+
+	/* for buddy allocator */
+	struct ext4_group_info ***s_group_info;
+	struct inode *s_buddy_cache;
+	long s_blocks_reserved;
+	spinlock_t s_reserve_lock;
+	struct list_head s_active_transaction;
+	struct list_head s_closed_transaction;
+	struct list_head s_committed_transaction;
+	spinlock_t s_md_lock;
+	tid_t s_last_transaction;
+	unsigned short *s_mb_offsets, *s_mb_maxs;
+
+	/* tunables */
+	unsigned long s_stripe;
+	unsigned long s_mb_stream_request;
+	unsigned long s_mb_max_to_scan;
+	unsigned long s_mb_min_to_scan;
+	unsigned long s_mb_stats;
+	unsigned long s_mb_order2_reqs;
+	unsigned long s_mb_group_prealloc;
+	/* where last allocation was done - for stream allocation */
+	unsigned long s_mb_last_group;
+	unsigned long s_mb_last_start;
+
+	/* history to debug policy */
+	struct ext4_mb_history *s_mb_history;
+	int s_mb_history_cur;
+	int s_mb_history_max;
+	int s_mb_history_num;
+	struct proc_dir_entry *s_mb_proc;
+	spinlock_t s_mb_history_lock;
+	int s_mb_history_filter;
+
+	/* stats for buddy allocator */
+	spinlock_t s_mb_pa_lock;
+	atomic_t s_bal_reqs;	/* number of reqs with len > 1 */
+	atomic_t s_bal_success;	/* we found long enough chunks */
+	atomic_t s_bal_allocated;	/* in blocks */
+	atomic_t s_bal_ex_scanned;	/* total extents scanned */
+	atomic_t s_bal_goals;	/* goal hits */
+	atomic_t s_bal_breaks;	/* too long searches */
+	atomic_t s_bal_2orders;	/* 2^order hits */
+	spinlock_t s_bal_lock;
+	unsigned long s_mb_buddies_generated;
+	unsigned long long s_mb_generation_time;
+	atomic_t s_mb_lost_chunks;
+	atomic_t s_mb_preallocated;
+	atomic_t s_mb_discarded;
+
+	/* locality groups */
+	struct ext4_locality_group *s_locality_groups;
 };
 
 #endif	/* _LINUX_EXT4_FS_SB */
-- 
cgit v1.2.3


From 7b7510662f4d05ddcc45d435769860e73e6aa20e Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Mon, 28 Jan 2008 23:58:27 -0500
Subject: jbd2: add lockdep support

Ported from similar patch for the jbd layer.

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/jbd2/transaction.c | 11 +++++++++++
 include/linux/jbd2.h  |  4 ++++
 2 files changed, 15 insertions(+)

(limited to 'include')

diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index f30802aeefae..70b3199e69dc 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -241,6 +241,8 @@ out:
 	return ret;
 }
 
+static struct lock_class_key jbd2_handle_key;
+
 /* Allocate a new handle.  This should probably be in a slab... */
 static handle_t *new_handle(int nblocks)
 {
@@ -251,6 +253,9 @@ static handle_t *new_handle(int nblocks)
 	handle->h_buffer_credits = nblocks;
 	handle->h_ref = 1;
 
+	lockdep_init_map(&handle->h_lockdep_map, "jbd2_handle",
+						&jbd2_handle_key, 0);
+
 	return handle;
 }
 
@@ -293,7 +298,11 @@ handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
 		jbd2_free_handle(handle);
 		current->journal_info = NULL;
 		handle = ERR_PTR(err);
+		goto out;
 	}
+
+	lock_acquire(&handle->h_lockdep_map, 0, 0, 0, 2, _THIS_IP_);
+out:
 	return handle;
 }
 
@@ -1419,6 +1428,8 @@ int jbd2_journal_stop(handle_t *handle)
 		spin_unlock(&journal->j_state_lock);
 	}
 
+	lock_release(&handle->h_lockdep_map, 1, _THIS_IP_);
+
 	jbd2_free_handle(handle);
 	return err;
 }
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 98a2bc5d3e3f..2cbf6fdb1799 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -418,6 +418,10 @@ struct handle_s
 	unsigned int	h_sync:		1;	/* sync-on-close */
 	unsigned int	h_jdata:	1;	/* force data journaling */
 	unsigned int	h_aborted:	1;	/* fatal error on handle */
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map	h_lockdep_map;
+#endif
 };
 
 
-- 
cgit v1.2.3


From 6dd06c9fbe025f542bce4cdb91790c0f91962722 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 29 Jan 2008 17:13:22 -0500
Subject: module: make module_address_lookup safe

module_address_lookup releases preemption then returns a pointer into
the module space.  The only user (kallsyms) copies the result, so just
do that under the preempt disable.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/module.h | 22 +++++++++++++---------
 kernel/kallsyms.c      | 11 ++++-------
 kernel/module.c        | 22 +++++++++++++---------
 3 files changed, 30 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/module.h b/include/linux/module.h
index c97bdb7eb957..aedc06be1de8 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -446,11 +446,14 @@ static inline void __module_get(struct module *module)
 	__mod ? __mod->name : "kernel";		\
 })
 
-/* For kallsyms to ask for address resolution.  NULL means not found. */
-const char *module_address_lookup(unsigned long addr,
-				  unsigned long *symbolsize,
-				  unsigned long *offset,
-				  char **modname);
+/* For kallsyms to ask for address resolution.  namebuf should be at
+ * least KSYM_NAME_LEN long: a pointer to namebuf is returned if
+ * found, otherwise NULL. */
+char *module_address_lookup(unsigned long addr,
+			    unsigned long *symbolsize,
+			    unsigned long *offset,
+			    char **modname,
+			    char *namebuf);
 int lookup_module_symbol_name(unsigned long addr, char *symname);
 int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name);
 
@@ -516,10 +519,11 @@ static inline void module_put(struct module *module)
 #define module_name(mod) "kernel"
 
 /* For kallsyms to ask for address resolution.  NULL means not found. */
-static inline const char *module_address_lookup(unsigned long addr,
-						unsigned long *symbolsize,
-						unsigned long *offset,
-						char **modname)
+static inline char *module_address_lookup(unsigned long addr,
+					  unsigned long *symbolsize,
+					  unsigned long *offset,
+					  char **modname,
+					  char *namebuf)
 {
 	return NULL;
 }
diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
index 2fc25810509e..7dadc71ce516 100644
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -233,10 +233,11 @@ static unsigned long get_symbol_pos(unsigned long addr,
 int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize,
 				unsigned long *offset)
 {
+	char namebuf[KSYM_NAME_LEN];
 	if (is_ksym_addr(addr))
 		return !!get_symbol_pos(addr, symbolsize, offset);
 
-	return !!module_address_lookup(addr, symbolsize, offset, NULL);
+	return !!module_address_lookup(addr, symbolsize, offset, NULL, namebuf);
 }
 
 /*
@@ -251,8 +252,6 @@ const char *kallsyms_lookup(unsigned long addr,
 			    unsigned long *offset,
 			    char **modname, char *namebuf)
 {
-	const char *msym;
-
 	namebuf[KSYM_NAME_LEN - 1] = 0;
 	namebuf[0] = 0;
 
@@ -268,10 +267,8 @@ const char *kallsyms_lookup(unsigned long addr,
 	}
 
 	/* see if it's in a module */
-	msym = module_address_lookup(addr, symbolsize, offset, modname);
-	if (msym)
-		return strncpy(namebuf, msym, KSYM_NAME_LEN - 1);
-
+	return module_address_lookup(addr, symbolsize, offset, modname,
+				     namebuf);
 	return NULL;
 }
 
diff --git a/kernel/module.c b/kernel/module.c
index 12067ff34d01..e814cd7da634 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2230,14 +2230,13 @@ static const char *get_ksymbol(struct module *mod,
 	return mod->strtab + mod->symtab[best].st_name;
 }
 
-/* For kallsyms to ask for address resolution.  NULL means not found.
-   We don't lock, as this is used for oops resolution and races are a
-   lesser concern. */
-/* FIXME: Risky: returns a pointer into a module w/o lock */
-const char *module_address_lookup(unsigned long addr,
-				  unsigned long *size,
-				  unsigned long *offset,
-				  char **modname)
+/* For kallsyms to ask for address resolution.  NULL means not found.  Careful
+ * not to lock to avoid deadlock on oopses, simply disable preemption. */
+char *module_address_lookup(unsigned long addr,
+			    unsigned long *size,
+			    unsigned long *offset,
+			    char **modname,
+			    char *namebuf)
 {
 	struct module *mod;
 	const char *ret = NULL;
@@ -2252,8 +2251,13 @@ const char *module_address_lookup(unsigned long addr,
 			break;
 		}
 	}
+	/* Make a copy in here where it's safe */
+	if (ret) {
+		strncpy(namebuf, ret, KSYM_NAME_LEN - 1);
+		ret = namebuf;
+	}
 	preempt_enable();
-	return ret;
+	return (char *)ret;
 }
 
 int lookup_module_symbol_name(unsigned long addr, char *symname)
-- 
cgit v1.2.3


From 20d60d9973c3b441902b0a3f4f6f7e7ade08f77d Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Tue, 23 Oct 2007 12:43:11 +0100
Subject: [MIPS] R4000/R4400 errata workarounds

 This is the gereric part of R4000/R4400 errata workarounds.  They include
compiler and assembler support as well as some source code modifications
to address the problems with some combinations of multiply/divide+shift
instructions as well as the daddi and daddiu instructions.

 Changes included are as follows:

1. New Kconfig options to select workarounds by platforms as necessary.

2. Arch top-level Makefile to pass necessary options to the compiler; also
   incompatible configurations are detected (-mno-sym32 unsupported as
   horribly intrusive for little gain).

3. Bug detection updated and shuffled -- the multiply/divide+shift problem
   is lethal enough that if not worked around it makes the kernel crash in
   time_init() because of a division by zero; the daddiu erratum might
   also trigger early potentially, though I have not observed it.  On the
   other hand the daddi detection code requires the exception subsystem to
   have been initialised (and is there mainly for information).

4. r4k_daddiu_bug() added so that the existence of the erratum can be
   queried by code at the run time as necessary; useful for generated code
   like TLB fault and copy/clear page handlers.

5. __udelay() updated as it uses multiplication in inline assembly.

 Note that -mdaddi requires modified toolchain (which has been maintained
by myself and available from my site for ~4years now -- versions covered
are GCC 2.95.4 - 4.1.2 and binutils from 2.13 onwards).  The -mfix-r4000
and -mfix-r4400 have been standard for a while though.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig             | 16 +++++++++++
 arch/mips/Makefile            | 12 ++++++---
 arch/mips/kernel/cpu-bugs64.c | 47 ++++++++++++++++----------------
 arch/mips/kernel/setup.c      |  4 ++-
 include/asm-mips/bugs.h       | 25 +++++++++++++++++
 include/asm-mips/delay.h      | 12 +++++++--
 include/asm-mips/war.h        | 62 +++++++++++++++++++++++++++++++++++++++++++
 7 files changed, 148 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 12bf96334174..11bc17ce0ebf 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -91,6 +91,9 @@ config MACH_DECSTATION
 	select BOOT_ELF32
 	select CEVT_R4K
 	select CSRC_R4K
+	select CPU_DADDI_WORKAROUNDS if 64BIT
+	select CPU_R4000_WORKAROUNDS if 64BIT
+	select CPU_R4400_WORKAROUNDS if 64BIT
 	select DMA_NONCOHERENT
 	select NO_IOPORT
 	select IRQ_CPU
@@ -1605,6 +1608,19 @@ config CPU_HAS_SYNC
 config GENERIC_CLOCKEVENTS_BROADCAST
 	bool
 
+#
+# CPU non-features
+#
+config CPU_DADDI_WORKAROUNDS
+	bool
+
+config CPU_R4000_WORKAROUNDS
+	bool
+	select CPU_R4400_WORKAROUNDS
+
+config CPU_R4400_WORKAROUNDS
+	bool
+
 #
 # Use the generic interrupt handling code in kernel/irq/:
 #
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index a1f8d8b96b03..a49127af33c8 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -141,6 +141,10 @@ cflags-$(CONFIG_CPU_R8000)	+= -march=r8000 -Wa,--trap
 cflags-$(CONFIG_CPU_R10000)	+= $(call cc-option,-march=r10000,-march=r8000) \
 			-Wa,--trap
 
+cflags-$(CONFIG_CPU_R4000_WORKAROUNDS)	+= $(call cc-option,-mfix-r4000,)
+cflags-$(CONFIG_CPU_R4400_WORKAROUNDS)	+= $(call cc-option,-mfix-r4400,)
+cflags-$(CONFIG_CPU_DADDI_WORKAROUNDS)	+= $(call cc-option,-mno-daddi,)
+
 ifdef CONFIG_CPU_SB1
 ifdef CONFIG_SB1_PASS_1_WORKAROUNDS
 MODFLAGS	+= -msb1-pass1-workarounds
@@ -602,9 +606,11 @@ ifdef CONFIG_64BIT
     endif
   endif
 
-  ifeq ($(KBUILD_SYM32), y)
-    ifeq ($(call cc-option-yn,-msym32), y)
-      cflags-y += -msym32 -DKBUILD_64BIT_SYM32
+  ifeq ($(KBUILD_SYM32)$(call cc-option-yn,-msym32), yy)
+    cflags-y += -msym32 -DKBUILD_64BIT_SYM32
+  else
+    ifeq ($(CONFIG_CPU_DADDI_WORKAROUNDS), y)
+      $(error CONFIG_CPU_DADDI_WORKAROUNDS unsupported without -msym32)
     endif
   endif
 endif
diff --git a/arch/mips/kernel/cpu-bugs64.c b/arch/mips/kernel/cpu-bugs64.c
index af78456d4138..417bb3e336ac 100644
--- a/arch/mips/kernel/cpu-bugs64.c
+++ b/arch/mips/kernel/cpu-bugs64.c
@@ -18,6 +18,15 @@
 #include <asm/mipsregs.h>
 #include <asm/system.h>
 
+static char bug64hit[] __initdata =
+	"reliable operation impossible!\n%s";
+static char nowar[] __initdata =
+	"Please report to <linux-mips@linux-mips.org>.";
+static char r4kwar[] __initdata =
+	"Enable CPU_R4000_WORKAROUNDS to rectify.";
+static char daddiwar[] __initdata =
+	"Enable CPU_DADDI_WORKAROUNDS to rectify.";
+
 static inline void align_mod(const int align, const int mod)
 {
 	asm volatile(
@@ -155,13 +164,7 @@ static inline void check_mult_sh(void)
 	}
 
 	printk("no.\n");
-	panic("Reliable operation impossible!\n"
-#ifndef CONFIG_CPU_R4000
-	      "Configure for R4000 to enable the workaround."
-#else
-	      "Please report to <linux-mips@linux-mips.org>."
-#endif
-	      );
+	panic(bug64hit, !R4000_WAR ? r4kwar : nowar);
 }
 
 static volatile int daddi_ov __initdata = 0;
@@ -233,15 +236,11 @@ static inline void check_daddi(void)
 	}
 
 	printk("no.\n");
-	panic("Reliable operation impossible!\n"
-#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400)
-	      "Configure for R4000 or R4400 to enable the workaround."
-#else
-	      "Please report to <linux-mips@linux-mips.org>."
-#endif
-	      );
+	panic(bug64hit, !DADDI_WAR ? daddiwar : nowar);
 }
 
+int daddiu_bug __initdata = -1;
+
 static inline void check_daddiu(void)
 {
 	long v, w, tmp;
@@ -281,7 +280,9 @@ static inline void check_daddiu(void)
 		: "=&r" (v), "=&r" (w), "=&r" (tmp)
 		: "I" (0xffffffffffffdb9aUL), "I" (0x1234));
 
-	if (v == w) {
+	daddiu_bug = v != w;
+
+	if (!daddiu_bug) {
 		printk("no.\n");
 		return;
 	}
@@ -303,18 +304,16 @@ static inline void check_daddiu(void)
 	}
 
 	printk("no.\n");
-	panic("Reliable operation impossible!\n"
-#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400)
-	      "Configure for R4000 or R4400 to enable the workaround."
-#else
-	      "Please report to <linux-mips@linux-mips.org>."
-#endif
-	      );
+	panic(bug64hit, !DADDI_WAR ? daddiwar : nowar);
 }
 
-void __init check_bugs64(void)
+void __init check_bugs64_early(void)
 {
 	check_mult_sh();
-	check_daddi();
 	check_daddiu();
 }
+
+void __init check_bugs64(void)
+{
+	check_daddi();
+}
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index f8a535afce39..7b4418dd5857 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -8,7 +8,7 @@
  * Copyright (C) 1994, 95, 96, 97, 98, 99, 2000, 01, 02, 03  Ralf Baechle
  * Copyright (C) 1996 Stoned Elipot
  * Copyright (C) 1999 Silicon Graphics, Inc.
- * Copyright (C) 2000 2001, 2002  Maciej W. Rozycki
+ * Copyright (C) 2000, 2001, 2002, 2007  Maciej W. Rozycki
  */
 #include <linux/init.h>
 #include <linux/ioport.h>
@@ -24,6 +24,7 @@
 
 #include <asm/addrspace.h>
 #include <asm/bootinfo.h>
+#include <asm/bugs.h>
 #include <asm/cache.h>
 #include <asm/cpu.h>
 #include <asm/sections.h>
@@ -561,6 +562,7 @@ void __init setup_arch(char **cmdline_p)
 	}
 #endif
 	cpu_report();
+	check_bugs_early();
 
 #if defined(CONFIG_VT)
 #if defined(CONFIG_VGA_CONSOLE)
diff --git a/include/asm-mips/bugs.h b/include/asm-mips/bugs.h
index 0d7f9c1f5546..9dc10df32078 100644
--- a/include/asm-mips/bugs.h
+++ b/include/asm-mips/bugs.h
@@ -1,19 +1,34 @@
 /*
  * This is included by init/main.c to check for architecture-dependent bugs.
  *
+ * Copyright (C) 2007  Maciej W. Rozycki
+ *
  * Needs:
  *	void check_bugs(void);
  */
 #ifndef _ASM_BUGS_H
 #define _ASM_BUGS_H
 
+#include <linux/bug.h>
 #include <linux/delay.h>
+
 #include <asm/cpu.h>
 #include <asm/cpu-info.h>
 
+extern int daddiu_bug;
+
+extern void check_bugs64_early(void);
+
 extern void check_bugs32(void);
 extern void check_bugs64(void);
 
+static inline void check_bugs_early(void)
+{
+#ifdef CONFIG_64BIT
+	check_bugs64_early();
+#endif
+}
+
 static inline void check_bugs(void)
 {
 	unsigned int cpu = smp_processor_id();
@@ -25,4 +40,14 @@ static inline void check_bugs(void)
 #endif
 }
 
+static inline int r4k_daddiu_bug(void)
+{
+#ifdef CONFIG_64BIT
+	WARN_ON(daddiu_bug < 0);
+	return daddiu_bug != 0;
+#else
+	return 0;
+#endif
+}
+
 #endif /* _ASM_BUGS_H */
diff --git a/include/asm-mips/delay.h b/include/asm-mips/delay.h
index fab32131e9b4..de5105d05f1e 100644
--- a/include/asm-mips/delay.h
+++ b/include/asm-mips/delay.h
@@ -6,13 +6,16 @@
  * Copyright (C) 1994 by Waldorf Electronics
  * Copyright (C) 1995 - 2000, 01, 03 by Ralf Baechle
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #ifndef _ASM_DELAY_H
 #define _ASM_DELAY_H
 
 #include <linux/param.h>
 #include <linux/smp.h>
+
 #include <asm/compiler.h>
+#include <asm/war.h>
 
 static inline void __delay(unsigned long loops)
 {
@@ -50,7 +53,7 @@ static inline void __delay(unsigned long loops)
 
 static inline void __udelay(unsigned long usecs, unsigned long lpj)
 {
-	unsigned long lo;
+	unsigned long hi, lo;
 
 	/*
 	 * The rates of 128 is rounded wrongly by the catchall case
@@ -70,11 +73,16 @@ static inline void __udelay(unsigned long usecs, unsigned long lpj)
 		: "=h" (usecs), "=l" (lo)
 		: "r" (usecs), "r" (lpj)
 		: GCC_REG_ACCUM);
-	else if (sizeof(long) == 8)
+	else if (sizeof(long) == 8 && !R4000_WAR)
 		__asm__("dmultu\t%2, %3"
 		: "=h" (usecs), "=l" (lo)
 		: "r" (usecs), "r" (lpj)
 		: GCC_REG_ACCUM);
+	else if (sizeof(long) == 8 && R4000_WAR)
+		__asm__("dmultu\t%3, %4\n\tmfhi\t%0"
+		: "=r" (usecs), "=h" (hi), "=l" (lo)
+		: "r" (usecs), "r" (lpj)
+		: GCC_REG_ACCUM);
 
 	__delay(usecs);
 }
diff --git a/include/asm-mips/war.h b/include/asm-mips/war.h
index d2808edfd4e9..22361d5e3bf0 100644
--- a/include/asm-mips/war.h
+++ b/include/asm-mips/war.h
@@ -4,12 +4,74 @@
  * for more details.
  *
  * Copyright (C) 2002, 2004, 2007 by Ralf Baechle
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #ifndef _ASM_WAR_H
 #define _ASM_WAR_H
 
 #include <war.h>
 
+/*
+ * Work around certain R4000 CPU errata (as implemented by GCC):
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ *   if executed immediately after starting an integer division:
+ *   "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ *   erratum #28
+ *   "MIPS R4000MC Errata, Processor Revision 2.2 and 3.0", erratum
+ *   #19
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ *   if executed while an integer multiplication is in progress:
+ *   "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ *   errata #16 & #28
+ *
+ * - An integer division may give an incorrect result if started in
+ *   a delay slot of a taken branch or a jump:
+ *   "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ *   erratum #52
+ */
+#ifdef CONFIG_CPU_R4000_WORKAROUNDS
+#define R4000_WAR 1
+#else
+#define R4000_WAR 0
+#endif
+
+/*
+ * Work around certain R4400 CPU errata (as implemented by GCC):
+ *
+ * - A double-word or a variable shift may give an incorrect result
+ *   if executed immediately after starting an integer division:
+ *   "MIPS R4400MC Errata, Processor Revision 1.0", erratum #10
+ *   "MIPS R4400MC Errata, Processor Revision 2.0 & 3.0", erratum #4
+ */
+#ifdef CONFIG_CPU_R4400_WORKAROUNDS
+#define R4400_WAR 1
+#else
+#define R4400_WAR 0
+#endif
+
+/*
+ * Work around the "daddi" and "daddiu" CPU errata:
+ *
+ * - The `daddi' instruction fails to trap on overflow.
+ *   "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ *   erratum #23
+ *
+ * - The `daddiu' instruction can produce an incorrect result.
+ *   "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0",
+ *   erratum #41
+ *   "MIPS R4000MC Errata, Processor Revision 2.2 and 3.0", erratum
+ *   #15
+ *   "MIPS R4400PC/SC Errata, Processor Revision 1.0", erratum #7
+ *   "MIPS R4400MC Errata, Processor Revision 1.0", erratum #5
+ */
+#ifdef CONFIG_CPU_DADDI_WORKAROUNDS
+#define DADDI_WAR 1
+#else
+#define DADDI_WAR 0
+#endif
+
 /*
  * Another R4600 erratum.  Due to the lack of errata information the exact
  * technical details aren't known.  I've experimentally found that disabling
-- 
cgit v1.2.3


From 619b6e18fce20e4b2d0082cde989f37e1be7b3e1 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Tue, 23 Oct 2007 12:43:25 +0100
Subject: [MIPS] R4000/R4400 daddiu erratum workaround

 This complements the generic R4000/R4400 errata workaround code and adds
bits for the daddiu problem.  In most places it just modifies handwritten
assembly code so that the assembler is allowed to use a temporary register
as daddiu may now be treated as a macro that expands to a sequence of li
and daddu.  It is the AT register or, where AT is unavailable or used
explicitly for another purpose, an explicitly-named register is selected,
using the .set at=<reg> feature added recently to gas.  This feature is
only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the
workaround remains disabled, the required version of binutils stays
unchanged.

 Similarly, daddiu instructions put in branch delay slots in noreorder
fragments are now taken out of them and the assembler is allowed to
reorder them itself as possible (which it does making the whole idea of
scheduling them into delay slots manually questionable).

 Also in the very few places where such a simple conversion was not
possible, a handcoded longer sequence is implemented.

 Other than that there are changes to code responsible for building the
TLB fault and page clear/copy handlers to avoid daddiu as appropriate.
These are only effective if the erratum is verified to be present at the
run time.

 Finally there is a trivial update to __delay(), because it uses daddiu in
a branch delay slot.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/genex.S        |  8 ++++-
 arch/mips/lib/csum_partial.S    | 61 ++++++++++++++++++++++++++++++-------
 arch/mips/lib/memcpy-inatomic.S | 25 ++++++++++++----
 arch/mips/lib/memcpy.S          | 60 +++++++++++++++++++++++++++++--------
 arch/mips/lib/memset.S          | 11 ++++++-
 arch/mips/lib/strncpy_user.S    |  4 +--
 arch/mips/mm/pg-r4k.c           | 66 +++++++++++++++++++----------------------
 arch/mips/mm/tlbex.c            | 42 ++++++++++++++++++--------
 include/asm-mips/delay.h        | 11 ++++++-
 include/asm-mips/stackframe.h   |  9 ++++++
 include/asm-mips/uaccess.h      | 13 ++++++--
 11 files changed, 227 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S
index e76a76bf0b3d..c6ada98ee042 100644
--- a/arch/mips/kernel/genex.S
+++ b/arch/mips/kernel/genex.S
@@ -6,7 +6,7 @@
  * Copyright (C) 1994 - 2000, 2001, 2003 Ralf Baechle
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
  * Copyright (C) 2001 MIPS Technologies, Inc.
- * Copyright (C) 2002 Maciej W. Rozycki
+ * Copyright (C) 2002, 2007  Maciej W. Rozycki
  */
 #include <linux/init.h>
 
@@ -471,7 +471,13 @@ NESTED(nmi_handler, PT_SIZE, sp)
 	jr	k0
 	 rfe
 #else
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	LONG_ADDIU	k0, 4		/* stall on $k0 */
+#else
+	.set	at=v1
+	LONG_ADDIU	k0, 4
+	.set	noat
+#endif
 	MTC0	k0, CP0_EPC
 	/* I hope three instructions between MTC0 and ERET are enough... */
 	ori	k1, _THREAD_MASK
diff --git a/arch/mips/lib/csum_partial.S b/arch/mips/lib/csum_partial.S
index c0a77fe038be..957a82484e3e 100644
--- a/arch/mips/lib/csum_partial.S
+++ b/arch/mips/lib/csum_partial.S
@@ -7,6 +7,7 @@
  *
  * Copyright (C) 1998, 1999 Ralf Baechle
  * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #include <linux/errno.h>
 #include <asm/asm.h>
@@ -52,9 +53,12 @@
 #define UNIT(unit)  ((unit)*NBYTES)
 
 #define ADDC(sum,reg)						\
+	.set	push;						\
+	.set	noat;						\
 	ADD	sum, reg;					\
 	sltu	v1, sum, reg;					\
-	ADD	sum, v1
+	ADD	sum, v1;					\
+	.set	pop
 
 #define CSUM_BIGCHUNK1(src, offset, sum, _t0, _t1, _t2, _t3)	\
 	LOAD	_t0, (offset + UNIT(0))(src);			\
@@ -178,8 +182,10 @@ move_128bytes:
 	CSUM_BIGCHUNK(src, 0x40, sum, t0, t1, t3, t4)
 	CSUM_BIGCHUNK(src, 0x60, sum, t0, t1, t3, t4)
 	LONG_SUBU	t8, t8, 0x01
+	.set	reorder				/* DADDI_WAR */
+	PTR_ADDU	src, src, 0x80
 	bnez	t8, move_128bytes
-	 PTR_ADDU	src, src, 0x80
+	.set	noreorder
 
 1:
 	beqz	t2, 1f
@@ -208,8 +214,10 @@ end_words:
 	lw	t0, (src)
 	LONG_SUBU	t8, t8, 0x1
 	ADDC(sum, t0)
+	.set	reorder				/* DADDI_WAR */
+	PTR_ADDU	src, src, 0x4
 	bnez	t8, end_words
-	 PTR_ADDU	src, src, 0x4
+	.set	noreorder
 
 /* unknown src alignment and < 8 bytes to go  */
 small_csumcpy:
@@ -246,6 +254,8 @@ small_csumcpy:
 1:	ADDC(sum, t1)
 
 	/* fold checksum */
+	.set	push
+	.set	noat
 #ifdef USE_DOUBLE
 	dsll32	v1, sum, 0
 	daddu	sum, v1
@@ -266,6 +276,7 @@ small_csumcpy:
 	srl	sum, sum, 8
 	or	sum, v1
 	andi	sum, 0xffff
+	.set	pop
 1:
 	.set	reorder
 	/* Add the passed partial csum.  */
@@ -373,7 +384,11 @@ small_csumcpy:
 
 #define ADDRMASK (NBYTES-1)
 
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	.set	noat
+#else
+	.set	at=v1
+#endif
 
 LEAF(__csum_partial_copy_user)
 	PTR_ADDU	AT, src, len	/* See (1) above. */
@@ -441,8 +456,10 @@ EXC(	STORE	t6, UNIT(6)(dst),	s_exc)
 	ADDC(sum, t6)
 EXC(	STORE	t7, UNIT(7)(dst),	s_exc)
 	ADDC(sum, t7)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 8*NBYTES
 	bgez	len, 1b
-	 ADD	dst, dst, 8*NBYTES
+	.set	noreorder
 	ADD	len, 8*NBYTES		# revert len (see above)
 
 	/*
@@ -471,8 +488,10 @@ EXC(	STORE	t2, UNIT(2)(dst),	s_exc)
 	ADDC(sum, t2)
 EXC(	STORE	t3, UNIT(3)(dst),	s_exc)
 	ADDC(sum, t3)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	beqz	len, done
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 less_than_4units:
 	/*
 	 * rem = len % NBYTES
@@ -485,8 +504,10 @@ EXC(	LOAD	t0, 0(src),		l_exc)
 	SUB	len, len, NBYTES
 EXC(	STORE	t0, 0(dst),		s_exc)
 	ADDC(sum, t0)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	rem, len, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 	/*
 	 * src and dst are aligned, need to copy rem bytes (rem < NBYTES)
@@ -572,8 +593,10 @@ EXC(	STORE	t2, UNIT(2)(dst),	s_exc)
 	ADDC(sum, t2)
 EXC(	STORE	t3, UNIT(3)(dst),	s_exc)
 	ADDC(sum, t3)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 
 cleanup_src_unaligned:
 	beqz	len, done
@@ -587,8 +610,10 @@ EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
 	SUB	len, len, NBYTES
 EXC(	STORE	t0, 0(dst),		s_exc)
 	ADDC(sum, t0)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 copy_bytes_checklen:
 	beqz	len, done
@@ -631,6 +656,8 @@ copy_bytes_done:
 	ADDC(sum, t2)
 done:
 	/* fold checksum */
+	.set	push
+	.set	noat
 #ifdef USE_DOUBLE
 	dsll32	v1, sum, 0
 	daddu	sum, v1
@@ -651,6 +678,7 @@ done:
 	srl	sum, sum, 8
 	or	sum, v1
 	andi	sum, 0xffff
+	.set	pop
 1:
 	.set reorder
 	ADDC(sum, psum)
@@ -678,8 +706,10 @@ EXC(	lbu	t1, 0(src),	l_exc)
 	SLLV	t1, t1, t2
 	addu	t2, SHIFT_INC
 	ADDC(sum, t1)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 1
 	bne	src, t0, 1b
-	 ADD	dst, dst, 1
+	.set	noreorder
 l_exc:
 	LOAD	t0, TI_TASK($28)
 	 nop
@@ -697,12 +727,22 @@ l_exc:
 	 * Clear len bytes starting at dst.  Can't call __bzero because it
 	 * might modify len.  An inefficient loop for these rare times...
 	 */
+	.set	reorder				/* DADDI_WAR */
+	SUB	src, len, 1
 	beqz	len, done
-	 SUB	src, len, 1
+	.set	noreorder
 1:	sb	zero, 0(dst)
 	ADD	dst, dst, 1
+	.set	push
+	.set	noat
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	bnez	src, 1b
 	 SUB	src, src, 1
+#else
+	li	v1, 1
+	bnez	src, 1b
+	 SUB	src, src, v1
+#endif
 	li	v1, -EFAULT
 	b	done
 	 sw	v1, (errptr)
@@ -712,4 +752,5 @@ s_exc:
 	li	v1, -EFAULT
 	jr	ra
 	 sw	v1, (errptr)
+	.set	pop
 	END(__csum_partial_copy_user)
diff --git a/arch/mips/lib/memcpy-inatomic.S b/arch/mips/lib/memcpy-inatomic.S
index 3a534b2baa0f..d1b08f5d6860 100644
--- a/arch/mips/lib/memcpy-inatomic.S
+++ b/arch/mips/lib/memcpy-inatomic.S
@@ -9,6 +9,7 @@
  * Copyright (C) 1999, 2000, 01, 2002 Silicon Graphics, Inc.
  * Copyright (C) 2002 Broadcom, Inc.
  *   memcpy/copy_user author: Mark Vandevoorde
+ * Copyright (C) 2007  Maciej W. Rozycki
  *
  * Mnemonic names for arguments to memcpy/__copy_user
  */
@@ -175,7 +176,11 @@
 
 	.text
 	.set	noreorder
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	.set	noat
+#else
+	.set	at=v1
+#endif
 
 /*
  * A combined memcpy/__copy_user
@@ -268,8 +273,10 @@ EXC(	LOAD	t3, UNIT(3)(src),	l_exc_copy)
 	STORE	t1, UNIT(1)(dst)
 	STORE	t2, UNIT(2)(dst)
 	STORE	t3, UNIT(3)(dst)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	beqz	len, done
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 less_than_4units:
 	/*
 	 * rem = len % NBYTES
@@ -281,8 +288,10 @@ EXC(	LOAD	t0, 0(src),		l_exc)
 	ADD	src, src, NBYTES
 	SUB	len, len, NBYTES
 	STORE	t0, 0(dst)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	rem, len, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 	/*
 	 * src and dst are aligned, need to copy rem bytes (rem < NBYTES)
@@ -361,8 +370,10 @@ EXC(	LDREST	t3, REST(3)(src),	l_exc_copy)
 	STORE	t2, UNIT(2)(dst)
 	STORE	t3, UNIT(3)(dst)
 	PREF(	1, 9*32(dst) )     	# 1 is PREF_STORE (not streamed)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 
 cleanup_src_unaligned:
 	beqz	len, done
@@ -375,8 +386,10 @@ EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
 	ADD	src, src, NBYTES
 	SUB	len, len, NBYTES
 	STORE	t0, 0(dst)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 copy_bytes_checklen:
 	beqz	len, done
@@ -424,8 +437,10 @@ l_exc_copy:
 EXC(	lb	t1, 0(src),	l_exc)
 	ADD	src, src, 1
 	sb	t1, 0(dst)	# can't fault -- we're copy_from_user
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 1
 	bne	src, t0, 1b
-	 ADD	dst, dst, 1
+	.set	noreorder
 l_exc:
 	LOAD	t0, TI_TASK($28)
 	 nop
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
index a526c62cb76a..aded7b159052 100644
--- a/arch/mips/lib/memcpy.S
+++ b/arch/mips/lib/memcpy.S
@@ -9,6 +9,7 @@
  * Copyright (C) 1999, 2000, 01, 2002 Silicon Graphics, Inc.
  * Copyright (C) 2002 Broadcom, Inc.
  *   memcpy/copy_user author: Mark Vandevoorde
+ * Copyright (C) 2007  Maciej W. Rozycki
  *
  * Mnemonic names for arguments to memcpy/__copy_user
  */
@@ -175,7 +176,11 @@
 
 	.text
 	.set	noreorder
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	.set	noat
+#else
+	.set	at=v1
+#endif
 
 /*
  * A combined memcpy/__copy_user
@@ -271,8 +276,10 @@ EXC(	STORE	t0, UNIT(0)(dst),	s_exc_p4u)
 EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p3u)
 EXC(	STORE	t2, UNIT(2)(dst),	s_exc_p2u)
 EXC(	STORE	t3, UNIT(3)(dst),	s_exc_p1u)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	beqz	len, done
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 less_than_4units:
 	/*
 	 * rem = len % NBYTES
@@ -284,8 +291,10 @@ EXC(	LOAD	t0, 0(src),		l_exc)
 	ADD	src, src, NBYTES
 	SUB	len, len, NBYTES
 EXC(	STORE	t0, 0(dst),		s_exc_p1u)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	rem, len, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 	/*
 	 * src and dst are aligned, need to copy rem bytes (rem < NBYTES)
@@ -364,8 +373,10 @@ EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p3u)
 EXC(	STORE	t2, UNIT(2)(dst),	s_exc_p2u)
 EXC(	STORE	t3, UNIT(3)(dst),	s_exc_p1u)
 	PREF(	1, 9*32(dst) )     	# 1 is PREF_STORE (not streamed)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 4*NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, 4*NBYTES
+	.set	noreorder
 
 cleanup_src_unaligned:
 	beqz	len, done
@@ -378,8 +389,10 @@ EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
 	ADD	src, src, NBYTES
 	SUB	len, len, NBYTES
 EXC(	STORE	t0, 0(dst),		s_exc_p1u)
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, NBYTES
 	bne	len, rem, 1b
-	 ADD	dst, dst, NBYTES
+	.set	noreorder
 
 copy_bytes_checklen:
 	beqz	len, done
@@ -427,8 +440,10 @@ l_exc_copy:
 EXC(	lb	t1, 0(src),	l_exc)
 	ADD	src, src, 1
 	sb	t1, 0(dst)	# can't fault -- we're copy_from_user
+	.set	reorder				/* DADDI_WAR */
+	ADD	dst, dst, 1
 	bne	src, t0, 1b
-	 ADD	dst, dst, 1
+	.set	noreorder
 l_exc:
 	LOAD	t0, TI_TASK($28)
 	 nop
@@ -446,20 +461,33 @@ l_exc:
 	 * Clear len bytes starting at dst.  Can't call __bzero because it
 	 * might modify len.  An inefficient loop for these rare times...
 	 */
+	.set	reorder				/* DADDI_WAR */
+	SUB	src, len, 1
 	beqz	len, done
-	 SUB	src, len, 1
+	.set	noreorder
 1:	sb	zero, 0(dst)
 	ADD	dst, dst, 1
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	bnez	src, 1b
 	 SUB	src, src, 1
+#else
+	.set	push
+	.set	noat
+	li	v1, 1
+	bnez	src, 1b
+	 SUB	src, src, v1
+	.set	pop
+#endif
 	jr	ra
 	 nop
 
 
-#define SEXC(n)				\
-s_exc_p ## n ## u:			\
-	jr	ra;			\
-	 ADD	len, len, n*NBYTES
+#define SEXC(n)							\
+	.set	reorder;			/* DADDI_WAR */	\
+s_exc_p ## n ## u:						\
+	ADD	len, len, n*NBYTES;				\
+	jr	ra;						\
+	.set	noreorder
 
 SEXC(8)
 SEXC(7)
@@ -471,8 +499,10 @@ SEXC(2)
 SEXC(1)
 
 s_exc_p1:
+	.set	reorder				/* DADDI_WAR */
+	ADD	len, len, 1
 	jr	ra
-	 ADD	len, len, 1
+	.set	noreorder
 s_exc:
 	jr	ra
 	 nop
@@ -502,8 +532,10 @@ r_end_bytes:
 	SUB	a2, a2, 0x1
 	sb	t0, -1(a0)
 	SUB	a1, a1, 0x1
+	.set	reorder				/* DADDI_WAR */
+	SUB	a0, a0, 0x1
 	bnez	a2, r_end_bytes
-	 SUB	a0, a0, 0x1
+	.set	noreorder
 
 r_out:
 	jr	ra
@@ -514,8 +546,10 @@ r_end_bytes_up:
 	SUB	a2, a2, 0x1
 	sb	t0, (a0)
 	ADD	a1, a1, 0x1
+	.set	reorder				/* DADDI_WAR */
+	ADD	a0, a0, 0x1
 	bnez	a2, r_end_bytes_up
-	 ADD	a0, a0, 0x1
+	.set	noreorder
 
 	jr	ra
 	 move	a2, zero
diff --git a/arch/mips/lib/memset.S b/arch/mips/lib/memset.S
index 3f8b8b3d0b23..3bf38422342f 100644
--- a/arch/mips/lib/memset.S
+++ b/arch/mips/lib/memset.S
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 1998, 1999, 2000 by Ralf Baechle
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #include <asm/asm.h>
 #include <asm/asm-offsets.h>
@@ -74,8 +75,16 @@ FEXPORT(__bzero)
 	bnez		t0, small_memset
 	 andi		t0, a0, LONGMASK	/* aligned? */
 
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 	beqz		t0, 1f
 	 PTR_SUBU	t0, LONGSIZE		/* alignment in bytes */
+#else
+	.set		noat
+	li		AT, LONGSIZE
+	beqz		t0, 1f
+	 PTR_SUBU	t0, AT			/* alignment in bytes */
+	.set		at
+#endif
 
 #ifdef __MIPSEB__
 	EX(LONG_S_L, a1, (a0), first_fixup)	/* make word/dword aligned */
@@ -106,7 +115,7 @@ memset_partial:
 	.set		noat
 	LONG_SRL		AT, t0, 1
 	PTR_SUBU	t1, AT
-	.set		noat
+	.set		at
 #endif
 	jr		t1
 	 PTR_ADDU	a0, t0			/* dest ptr */
diff --git a/arch/mips/lib/strncpy_user.S b/arch/mips/lib/strncpy_user.S
index d16c76fbfac7..5c8fb9d6b7f9 100644
--- a/arch/mips/lib/strncpy_user.S
+++ b/arch/mips/lib/strncpy_user.S
@@ -41,9 +41,9 @@ FEXPORT(__strncpy_from_user_nocheck_asm)
 	beqz		t0, 2f
 	 sb		t0, (a0)
 	PTR_ADDIU	v0, 1
-	bne		v0, a2, 1b
-	 PTR_ADDIU	a0, 1
 	.set		reorder
+	PTR_ADDIU	a0, 1
+	bne		v0, a2, 1b
 2:	PTR_ADDU	t0, a1, v0
 	xor		t0, a1
 	bltz		t0, fault
diff --git a/arch/mips/mm/pg-r4k.c b/arch/mips/mm/pg-r4k.c
index 4f770ac885ce..9185fbf37c0d 100644
--- a/arch/mips/mm/pg-r4k.c
+++ b/arch/mips/mm/pg-r4k.c
@@ -4,6 +4,7 @@
  * for more details.
  *
  * Copyright (C) 2003, 04, 05 Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -12,6 +13,7 @@
 #include <linux/module.h>
 #include <linux/proc_fs.h>
 
+#include <asm/bugs.h>
 #include <asm/cacheops.h>
 #include <asm/inst.h>
 #include <asm/io.h>
@@ -255,64 +257,58 @@ static inline void build_store_reg(int reg)
 	__build_store_reg(reg);
 }
 
-static inline void build_addiu_a2_a0(unsigned long offset)
+static inline void build_addiu_rt_rs(unsigned int rt, unsigned int rs,
+				     unsigned long offset)
 {
 	union mips_instruction mi;
 
 	BUG_ON(offset > 0x7fff);
 
-	mi.i_format.opcode     = cpu_has_64bit_gp_regs ? daddiu_op : addiu_op;
-	mi.i_format.rs         = 4;		/* $a0 */
-	mi.i_format.rt         = 6;		/* $a2 */
-	mi.i_format.simmediate = offset;
+	if (cpu_has_64bit_gp_regs && DADDI_WAR && r4k_daddiu_bug()) {
+		mi.i_format.opcode     = addiu_op;
+		mi.i_format.rs         = 0;	/* $zero */
+		mi.i_format.rt         = 25;	/* $t9 */
+		mi.i_format.simmediate = offset;
+		emit_instruction(mi);
 
+		mi.r_format.opcode     = spec_op;
+		mi.r_format.rs         = rs;
+		mi.r_format.rt         = 25;	/* $t9 */
+		mi.r_format.rd         = rt;
+		mi.r_format.re         = 0;
+		mi.r_format.func       = daddu_op;
+	} else {
+		mi.i_format.opcode     = cpu_has_64bit_gp_regs ?
+					 daddiu_op : addiu_op;
+		mi.i_format.rs         = rs;
+		mi.i_format.rt         = rt;
+		mi.i_format.simmediate = offset;
+	}
 	emit_instruction(mi);
 }
 
-static inline void build_addiu_a2(unsigned long offset)
+static inline void build_addiu_a2_a0(unsigned long offset)
 {
-	union mips_instruction mi;
-
-	BUG_ON(offset > 0x7fff);
-
-	mi.i_format.opcode     = cpu_has_64bit_gp_regs ? daddiu_op : addiu_op;
-	mi.i_format.rs         = 6;		/* $a2 */
-	mi.i_format.rt         = 6;		/* $a2 */
-	mi.i_format.simmediate = offset;
+	build_addiu_rt_rs(6, 4, offset);	/* $a2, $a0, offset */
+}
 
-	emit_instruction(mi);
+static inline void build_addiu_a2(unsigned long offset)
+{
+	build_addiu_rt_rs(6, 6, offset);	/* $a2, $a2, offset */
 }
 
 static inline void build_addiu_a1(unsigned long offset)
 {
-	union mips_instruction mi;
-
-	BUG_ON(offset > 0x7fff);
-
-	mi.i_format.opcode     = cpu_has_64bit_gp_regs ? daddiu_op : addiu_op;
-	mi.i_format.rs         = 5;		/* $a1 */
-	mi.i_format.rt         = 5;		/* $a1 */
-	mi.i_format.simmediate = offset;
+	build_addiu_rt_rs(5, 5, offset);	/* $a1, $a1, offset */
 
 	load_offset -= offset;
-
-	emit_instruction(mi);
 }
 
 static inline void build_addiu_a0(unsigned long offset)
 {
-	union mips_instruction mi;
-
-	BUG_ON(offset > 0x7fff);
-
-	mi.i_format.opcode     = cpu_has_64bit_gp_regs ? daddiu_op : addiu_op;
-	mi.i_format.rs         = 4;		/* $a0 */
-	mi.i_format.rt         = 4;		/* $a0 */
-	mi.i_format.simmediate = offset;
+	build_addiu_rt_rs(4, 4, offset);	/* $a0, $a0, offset */
 
 	store_offset -= offset;
-
-	emit_instruction(mi);
 }
 
 static inline void build_bne(unsigned int *dest)
diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c
index 511107f92d9c..f8925ba0b39e 100644
--- a/arch/mips/mm/tlbex.c
+++ b/arch/mips/mm/tlbex.c
@@ -6,7 +6,7 @@
  * Synthesize TLB refill handlers at runtime.
  *
  * Copyright (C) 2004,2005,2006 by Thiemo Seufer
- * Copyright (C) 2005  Maciej W. Rozycki
+ * Copyright (C) 2005, 2007  Maciej W. Rozycki
  * Copyright (C) 2006  Ralf Baechle (ralf@linux-mips.org)
  *
  * ... and the days got worse and worse and now you see
@@ -27,6 +27,7 @@
 #include <linux/string.h>
 #include <linux/init.h>
 
+#include <asm/bugs.h>
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 #include <asm/mmu_context.h>
@@ -293,7 +294,7 @@ static void __init build_insn(u32 **buf, enum opcode opc, ...)
 			break;
 		}
 
-	if (!ip)
+	if (!ip || (opc == insn_daddiu && r4k_daddiu_bug()))
 		panic("Unsupported TLB synthesizer instruction %d", opc);
 
 	op = ip->match;
@@ -525,23 +526,33 @@ L_LA(_r3000_write_probe_fail)
 #define i_ssnop(buf) i_sll(buf, 0, 0, 1)
 #define i_ehb(buf) i_sll(buf, 0, 0, 3)
 
-#ifdef CONFIG_64BIT
 static __init int __maybe_unused in_compat_space_p(long addr)
 {
 	/* Is this address in 32bit compat space? */
+#ifdef CONFIG_64BIT
 	return (((addr) & 0xffffffff00000000L) == 0xffffffff00000000L);
+#else
+	return 1;
+#endif
 }
 
 static __init int __maybe_unused rel_highest(long val)
 {
+#ifdef CONFIG_64BIT
 	return ((((val + 0x800080008000L) >> 48) & 0xffff) ^ 0x8000) - 0x8000;
+#else
+	return 0;
+#endif
 }
 
 static __init int __maybe_unused rel_higher(long val)
 {
+#ifdef CONFIG_64BIT
 	return ((((val + 0x80008000L) >> 32) & 0xffff) ^ 0x8000) - 0x8000;
-}
+#else
+	return 0;
 #endif
+}
 
 static __init int rel_hi(long val)
 {
@@ -555,7 +566,6 @@ static __init int rel_lo(long val)
 
 static __init void i_LA_mostly(u32 **buf, unsigned int rs, long addr)
 {
-#ifdef CONFIG_64BIT
 	if (!in_compat_space_p(addr)) {
 		i_lui(buf, rs, rel_highest(addr));
 		if (rel_higher(addr))
@@ -567,16 +577,18 @@ static __init void i_LA_mostly(u32 **buf, unsigned int rs, long addr)
 		} else
 			i_dsll32(buf, rs, rs, 0);
 	} else
-#endif
 		i_lui(buf, rs, rel_hi(addr));
 }
 
-static __init void __maybe_unused i_LA(u32 **buf, unsigned int rs,
-					     long addr)
+static __init void __maybe_unused i_LA(u32 **buf, unsigned int rs, long addr)
 {
 	i_LA_mostly(buf, rs, addr);
-	if (rel_lo(addr))
-		i_ADDIU(buf, rs, rs, rel_lo(addr));
+	if (rel_lo(addr)) {
+		if (!in_compat_space_p(addr))
+			i_daddiu(buf, rs, rs, rel_lo(addr));
+		else
+			i_addiu(buf, rs, rs, rel_lo(addr));
+	}
 }
 
 /*
@@ -1085,7 +1097,10 @@ build_get_pgd_vmalloc64(u32 **p, struct label **l, struct reloc **r,
 	} else {
 		i_LA_mostly(p, ptr, modd);
 		il_b(p, r, label_vmalloc_done);
-		i_daddiu(p, ptr, ptr, rel_lo(modd));
+		if (in_compat_space_p(modd))
+			i_addiu(p, ptr, ptr, rel_lo(modd));
+		else
+			i_daddiu(p, ptr, ptr, rel_lo(modd));
 	}
 
 	l_vmalloc(l, *p);
@@ -1106,7 +1121,10 @@ build_get_pgd_vmalloc64(u32 **p, struct label **l, struct reloc **r,
 	} else {
 		i_LA_mostly(p, ptr, swpd);
 		il_b(p, r, label_vmalloc_done);
-		i_daddiu(p, ptr, ptr, rel_lo(swpd));
+		if (in_compat_space_p(swpd))
+			i_addiu(p, ptr, ptr, rel_lo(swpd));
+		else
+			i_daddiu(p, ptr, ptr, rel_lo(swpd));
 	}
 }
 
diff --git a/include/asm-mips/delay.h b/include/asm-mips/delay.h
index de5105d05f1e..b0bccd2c4ed5 100644
--- a/include/asm-mips/delay.h
+++ b/include/asm-mips/delay.h
@@ -28,7 +28,7 @@ static inline void __delay(unsigned long loops)
 		"	.set	reorder					\n"
 		: "=r" (loops)
 		: "0" (loops));
-	else if (sizeof(long) == 8)
+	else if (sizeof(long) == 8 && !DADDI_WAR)
 		__asm__ __volatile__ (
 		"	.set	noreorder				\n"
 		"	.align	3					\n"
@@ -37,6 +37,15 @@ static inline void __delay(unsigned long loops)
 		"	.set	reorder					\n"
 		: "=r" (loops)
 		: "0" (loops));
+	else if (sizeof(long) == 8 && DADDI_WAR)
+		__asm__ __volatile__ (
+		"	.set	noreorder				\n"
+		"	.align	3					\n"
+		"1:	bnez	%0, 1b					\n"
+		"	dsubu	%0, %2					\n"
+		"	.set	reorder					\n"
+		: "=r" (loops)
+		: "0" (loops), "r" (1));
 }
 
 
diff --git a/include/asm-mips/stackframe.h b/include/asm-mips/stackframe.h
index fb41a8d76392..051e1af0bb95 100644
--- a/include/asm-mips/stackframe.h
+++ b/include/asm-mips/stackframe.h
@@ -6,6 +6,7 @@
  * Copyright (C) 1994, 95, 96, 99, 2001 Ralf Baechle
  * Copyright (C) 1994, 1995, 1996 Paul M. Antoine.
  * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #ifndef _ASM_STACKFRAME_H
 #define _ASM_STACKFRAME_H
@@ -145,8 +146,16 @@
 		.set	reorder
 		/* Called from user mode, new stack. */
 		get_saved_sp
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
 8:		move	k0, sp
 		PTR_SUBU sp, k1, PT_SIZE
+#else
+		.set	at=k0
+8:		PTR_SUBU k1, PT_SIZE
+		.set	noat
+		move	k0, sp
+		move	sp, k1
+#endif
 		LONG_S	k0, PT_R29(sp)
 		LONG_S	$3, PT_R3(sp)
 		/*
diff --git a/include/asm-mips/uaccess.h b/include/asm-mips/uaccess.h
index c30c718994c9..66523d610950 100644
--- a/include/asm-mips/uaccess.h
+++ b/include/asm-mips/uaccess.h
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 1996, 1997, 1998, 1999, 2000, 03, 04 by Ralf Baechle
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2007  Maciej W. Rozycki
  */
 #ifndef _ASM_UACCESS_H
 #define _ASM_UACCESS_H
@@ -387,6 +388,12 @@ extern void __put_user_unknown(void);
 	"jal\t" #destination "\n\t"
 #endif
 
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
+#define DADDI_SCRATCH "$0"
+#else
+#define DADDI_SCRATCH "$3"
+#endif
+
 extern size_t __copy_user(void *__to, const void *__from, size_t __n);
 
 #define __invoke_copy_to_user(to, from, n)				\
@@ -403,7 +410,7 @@ extern size_t __copy_user(void *__to, const void *__from, size_t __n);
 	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
 	:								\
 	: "$8", "$9", "$10", "$11", "$12", "$15", "$24", "$31",		\
-	  "memory");							\
+	  DADDI_SCRATCH, "memory");					\
 	__cu_len_r;							\
 })
 
@@ -512,7 +519,7 @@ extern size_t __copy_user_inatomic(void *__to, const void *__from, size_t __n);
 	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
 	:								\
 	: "$8", "$9", "$10", "$11", "$12", "$15", "$24", "$31",		\
-	  "memory");							\
+	  DADDI_SCRATCH, "memory");					\
 	__cu_len_r;							\
 })
 
@@ -535,7 +542,7 @@ extern size_t __copy_user_inatomic(void *__to, const void *__from, size_t __n);
 	: "+r" (__cu_to_r), "+r" (__cu_from_r), "+r" (__cu_len_r)	\
 	:								\
 	: "$8", "$9", "$10", "$11", "$12", "$15", "$24", "$31",		\
-	  "memory");							\
+	  DADDI_SCRATCH, "memory");					\
 	__cu_len_r;							\
 })
 
-- 
cgit v1.2.3


From 417345d8f5bacc291138c99a292a8ca69947530e Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 29 Oct 2007 14:23:43 +0000
Subject: [MIPS] CFE: Make code remotely resemble Linux code.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/fw/cfe/cfe_api.c          | 184 +++++++++++------------------------
 arch/mips/fw/cfe/cfe_api_int.h      | 186 ++++++++++++++++++------------------
 include/asm-mips/fw/cfe/cfe_api.h   |  87 +++--------------
 include/asm-mips/fw/cfe/cfe_error.h |  19 ++--
 4 files changed, 167 insertions(+), 309 deletions(-)

(limited to 'include')

diff --git a/arch/mips/fw/cfe/cfe_api.c b/arch/mips/fw/cfe/cfe_api.c
index a9f69e4e40ac..717db74f7c6e 100644
--- a/arch/mips/fw/cfe/cfe_api.c
+++ b/arch/mips/fw/cfe/cfe_api.c
@@ -16,19 +16,16 @@
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
-/*  *********************************************************************
-    *
-    *  Broadcom Common Firmware Environment (CFE)
-    *
-    *  Device Function stubs			File: cfe_api.c
-    *
-    *  This module contains device function stubs (small routines to
-    *  call the standard "iocb" interface entry point to CFE).
-    *  There should be one routine here per iocb function call.
-    *
-    *  Authors:  Mitch Lichtenberg, Chris Demetriou
-    *
-    ********************************************************************* */
+/*
+ *
+ * Broadcom Common Firmware Environment (CFE)
+ *
+ * This module contains device function stubs (small routines to
+ * call the standard "iocb" interface entry point to CFE).
+ * There should be one routine here per iocb function call.
+ *
+ * Authors:  Mitch Lichtenberg, Chris Demetriou
+ */
 
 #include <asm/fw/cfe/cfe_api.h>
 #include "cfe_api_int.h"
@@ -37,12 +34,8 @@
 #define XPTR_FROM_NATIVE(n)	((cfe_xptr_t) (intptr_t) (n))
 #define NATIVE_FROM_XPTR(x)	((void *) (intptr_t) (x))
 
-#ifdef CFE_API_IMPL_NAMESPACE
-#define cfe_iocb_dispatch(a)		__cfe_iocb_dispatch(a)
-#endif
-int cfe_iocb_dispatch(cfe_xiocb_t * xiocb);
+int cfe_iocb_dispatch(struct cfe_xiocb *xiocb);
 
-#if defined(CFE_API_common) || defined(CFE_API_ALL)
 /*
  * Declare the dispatch function with args of "intptr_t".
  * This makes sure whatever model we're compiling in
@@ -53,27 +46,25 @@ int cfe_iocb_dispatch(cfe_xiocb_t * xiocb);
  */
 
 static int (*cfe_dispfunc) (intptr_t handle, intptr_t xiocb) = 0;
-static cfe_xuint_t cfe_handle = 0;
+static u64 cfe_handle = 0;
 
-int cfe_init(cfe_xuint_t handle, cfe_xuint_t ept)
+int cfe_init(u64 handle, u64 ept)
 {
 	cfe_dispfunc = NATIVE_FROM_XPTR(ept);
 	cfe_handle = handle;
 	return 0;
 }
 
-int cfe_iocb_dispatch(cfe_xiocb_t * xiocb)
+int cfe_iocb_dispatch(struct cfe_xiocb * xiocb)
 {
 	if (!cfe_dispfunc)
 		return -1;
 	return (*cfe_dispfunc) ((intptr_t) cfe_handle, (intptr_t) xiocb);
 }
-#endif				/* CFE_API_common || CFE_API_ALL */
 
-#if defined(CFE_API_close) || defined(CFE_API_ALL)
 int cfe_close(int handle)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_CLOSE;
 	xiocb.xiocb_status = 0;
@@ -86,18 +77,16 @@ int cfe_close(int handle)
 	return xiocb.xiocb_status;
 
 }
-#endif				/* CFE_API_close || CFE_API_ALL */
 
-#if defined(CFE_API_cpu_start) || defined(CFE_API_ALL)
 int cfe_cpu_start(int cpu, void (*fn) (void), long sp, long gp, long a1)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_CPUCTL;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_cpuctl_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_cpuctl);
 	xiocb.plist.xiocb_cpuctl.cpu_number = cpu;
 	xiocb.plist.xiocb_cpuctl.cpu_command = CFE_CPU_CMD_START;
 	xiocb.plist.xiocb_cpuctl.gp_val = gp;
@@ -109,18 +98,16 @@ int cfe_cpu_start(int cpu, void (*fn) (void), long sp, long gp, long a1)
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_cpu_start || CFE_API_ALL */
 
-#if defined(CFE_API_cpu_stop) || defined(CFE_API_ALL)
 int cfe_cpu_stop(int cpu)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_CPUCTL;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_cpuctl_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_cpuctl);
 	xiocb.plist.xiocb_cpuctl.cpu_number = cpu;
 	xiocb.plist.xiocb_cpuctl.cpu_command = CFE_CPU_CMD_STOP;
 
@@ -128,18 +115,16 @@ int cfe_cpu_stop(int cpu)
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_cpu_stop || CFE_API_ALL */
 
-#if defined(CFE_API_enumenv) || defined(CFE_API_ALL)
 int cfe_enumenv(int idx, char *name, int namelen, char *val, int vallen)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_ENV_SET;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_envbuf_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_envbuf);
 	xiocb.plist.xiocb_envbuf.enum_idx = idx;
 	xiocb.plist.xiocb_envbuf.name_ptr = XPTR_FROM_NATIVE(name);
 	xiocb.plist.xiocb_envbuf.name_length = namelen;
@@ -150,20 +135,17 @@ int cfe_enumenv(int idx, char *name, int namelen, char *val, int vallen)
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_enumenv || CFE_API_ALL */
 
-#if defined(CFE_API_enummem) || defined(CFE_API_ALL)
 int
-cfe_enummem(int idx, int flags, cfe_xuint_t * start, cfe_xuint_t * length,
-	    cfe_xuint_t * type)
+cfe_enummem(int idx, int flags, u64 *start, u64 *length, u64 *type)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_MEMENUM;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = flags;
-	xiocb.xiocb_psize = sizeof(xiocb_meminfo_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_meminfo);
 	xiocb.plist.xiocb_meminfo.mi_idx = idx;
 
 	cfe_iocb_dispatch(&xiocb);
@@ -177,30 +159,26 @@ cfe_enummem(int idx, int flags, cfe_xuint_t * start, cfe_xuint_t * length,
 
 	return 0;
 }
-#endif				/* CFE_API_enummem || CFE_API_ALL */
 
-#if defined(CFE_API_exit) || defined(CFE_API_ALL)
 int cfe_exit(int warm, int status)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_RESTART;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = warm ? CFE_FLG_WARMSTART : 0;
-	xiocb.xiocb_psize = sizeof(xiocb_exitstat_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_exitstat);
 	xiocb.plist.xiocb_exitstat.status = status;
 
 	cfe_iocb_dispatch(&xiocb);
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_exit || CFE_API_ALL */
 
-#if defined(CFE_API_flushcache) || defined(CFE_API_ALL)
 int cfe_flushcache(int flg)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_FLUSHCACHE;
 	xiocb.xiocb_status = 0;
@@ -212,34 +190,30 @@ int cfe_flushcache(int flg)
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_flushcache || CFE_API_ALL */
 
-#if defined(CFE_API_getdevinfo) || defined(CFE_API_ALL)
 int cfe_getdevinfo(char *name)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_GETINFO;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_buffer_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_buffer);
 	xiocb.plist.xiocb_buffer.buf_offset = 0;
 	xiocb.plist.xiocb_buffer.buf_ptr = XPTR_FROM_NATIVE(name);
-	xiocb.plist.xiocb_buffer.buf_length = cfe_strlen(name);
+	xiocb.plist.xiocb_buffer.buf_length = strlen(name);
 
 	cfe_iocb_dispatch(&xiocb);
 
 	if (xiocb.xiocb_status < 0)
 		return xiocb.xiocb_status;
-	return xiocb.plist.xiocb_buffer.buf_devflags;
+	return xiocb.plist.xiocb_buffer.buf_ioctlcmd;
 }
-#endif				/* CFE_API_getdevinfo || CFE_API_ALL */
 
-#if defined(CFE_API_getenv) || defined(CFE_API_ALL)
 int cfe_getenv(char *name, char *dest, int destlen)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	*dest = 0;
 
@@ -247,10 +221,10 @@ int cfe_getenv(char *name, char *dest, int destlen)
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_envbuf_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_envbuf);
 	xiocb.plist.xiocb_envbuf.enum_idx = 0;
 	xiocb.plist.xiocb_envbuf.name_ptr = XPTR_FROM_NATIVE(name);
-	xiocb.plist.xiocb_envbuf.name_length = cfe_strlen(name);
+	xiocb.plist.xiocb_envbuf.name_length = strlen(name);
 	xiocb.plist.xiocb_envbuf.val_ptr = XPTR_FROM_NATIVE(dest);
 	xiocb.plist.xiocb_envbuf.val_length = destlen;
 
@@ -258,18 +232,16 @@ int cfe_getenv(char *name, char *dest, int destlen)
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_getenv || CFE_API_ALL */
 
-#if defined(CFE_API_getfwinfo) || defined(CFE_API_ALL)
 int cfe_getfwinfo(cfe_fwinfo_t * info)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_GETINFO;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_fwinfo_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_fwinfo);
 
 	cfe_iocb_dispatch(&xiocb);
 
@@ -292,12 +264,10 @@ int cfe_getfwinfo(cfe_fwinfo_t * info)
 
 	return 0;
 }
-#endif				/* CFE_API_getfwinfo || CFE_API_ALL */
 
-#if defined(CFE_API_getstdhandle) || defined(CFE_API_ALL)
 int cfe_getstdhandle(int flg)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_GETHANDLE;
 	xiocb.xiocb_status = 0;
@@ -311,23 +281,17 @@ int cfe_getstdhandle(int flg)
 		return xiocb.xiocb_status;
 	return xiocb.xiocb_handle;
 }
-#endif				/* CFE_API_getstdhandle || CFE_API_ALL */
 
-#if defined(CFE_API_getticks) || defined(CFE_API_ALL)
 int64_t
-#ifdef CFE_API_IMPL_NAMESPACE
-__cfe_getticks(void)
-#else
 cfe_getticks(void)
-#endif
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_FW_GETTIME;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_time_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_time);
 	xiocb.plist.xiocb_time.ticks = 0;
 
 	cfe_iocb_dispatch(&xiocb);
@@ -335,18 +299,16 @@ cfe_getticks(void)
 	return xiocb.plist.xiocb_time.ticks;
 
 }
-#endif				/* CFE_API_getticks || CFE_API_ALL */
 
-#if defined(CFE_API_inpstat) || defined(CFE_API_ALL)
 int cfe_inpstat(int handle)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_INPSTAT;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = handle;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_inpstat_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_inpstat);
 	xiocb.plist.xiocb_inpstat.inp_status = 0;
 
 	cfe_iocb_dispatch(&xiocb);
@@ -355,20 +317,18 @@ int cfe_inpstat(int handle)
 		return xiocb.xiocb_status;
 	return xiocb.plist.xiocb_inpstat.inp_status;
 }
-#endif				/* CFE_API_inpstat || CFE_API_ALL */
 
-#if defined(CFE_API_ioctl) || defined(CFE_API_ALL)
 int
 cfe_ioctl(int handle, unsigned int ioctlnum, unsigned char *buffer,
-	  int length, int *retlen, cfe_xuint_t offset)
+	  int length, int *retlen, u64 offset)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_IOCTL;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = handle;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_buffer_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_buffer);
 	xiocb.plist.xiocb_buffer.buf_offset = offset;
 	xiocb.plist.xiocb_buffer.buf_ioctlcmd = ioctlnum;
 	xiocb.plist.xiocb_buffer.buf_ptr = XPTR_FROM_NATIVE(buffer);
@@ -380,21 +340,19 @@ cfe_ioctl(int handle, unsigned int ioctlnum, unsigned char *buffer,
 		*retlen = xiocb.plist.xiocb_buffer.buf_retlen;
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_ioctl || CFE_API_ALL */
 
-#if defined(CFE_API_open) || defined(CFE_API_ALL)
 int cfe_open(char *name)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_OPEN;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_buffer_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_buffer);
 	xiocb.plist.xiocb_buffer.buf_offset = 0;
 	xiocb.plist.xiocb_buffer.buf_ptr = XPTR_FROM_NATIVE(name);
-	xiocb.plist.xiocb_buffer.buf_length = cfe_strlen(name);
+	xiocb.plist.xiocb_buffer.buf_length = strlen(name);
 
 	cfe_iocb_dispatch(&xiocb);
 
@@ -402,27 +360,21 @@ int cfe_open(char *name)
 		return xiocb.xiocb_status;
 	return xiocb.xiocb_handle;
 }
-#endif				/* CFE_API_open || CFE_API_ALL */
 
-#if defined(CFE_API_read) || defined(CFE_API_ALL)
 int cfe_read(int handle, unsigned char *buffer, int length)
 {
 	return cfe_readblk(handle, 0, buffer, length);
 }
-#endif				/* CFE_API_read || CFE_API_ALL */
 
-#if defined(CFE_API_readblk) || defined(CFE_API_ALL)
-int
-cfe_readblk(int handle, cfe_xint_t offset, unsigned char *buffer,
-	    int length)
+int cfe_readblk(int handle, s64 offset, unsigned char *buffer, int length)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_READ;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = handle;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_buffer_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_buffer);
 	xiocb.plist.xiocb_buffer.buf_offset = offset;
 	xiocb.plist.xiocb_buffer.buf_ptr = XPTR_FROM_NATIVE(buffer);
 	xiocb.plist.xiocb_buffer.buf_length = length;
@@ -433,62 +385,41 @@ cfe_readblk(int handle, cfe_xint_t offset, unsigned char *buffer,
 		return xiocb.xiocb_status;
 	return xiocb.plist.xiocb_buffer.buf_retlen;
 }
-#endif				/* CFE_API_readblk || CFE_API_ALL */
 
-#if defined(CFE_API_setenv) || defined(CFE_API_ALL)
 int cfe_setenv(char *name, char *val)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_ENV_SET;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = 0;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_envbuf_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_envbuf);
 	xiocb.plist.xiocb_envbuf.enum_idx = 0;
 	xiocb.plist.xiocb_envbuf.name_ptr = XPTR_FROM_NATIVE(name);
-	xiocb.plist.xiocb_envbuf.name_length = cfe_strlen(name);
+	xiocb.plist.xiocb_envbuf.name_length = strlen(name);
 	xiocb.plist.xiocb_envbuf.val_ptr = XPTR_FROM_NATIVE(val);
-	xiocb.plist.xiocb_envbuf.val_length = cfe_strlen(val);
+	xiocb.plist.xiocb_envbuf.val_length = strlen(val);
 
 	cfe_iocb_dispatch(&xiocb);
 
 	return xiocb.xiocb_status;
 }
-#endif				/* CFE_API_setenv || CFE_API_ALL */
-
-#if (defined(CFE_API_strlen) || defined(CFE_API_ALL)) \
-    && !defined(CFE_API_STRLEN_CUSTOM)
-int cfe_strlen(char *name)
-{
-	int count = 0;
-
-	while (*name++)
-		count++;
 
-	return count;
-}
-#endif				/* CFE_API_strlen || CFE_API_ALL */
-
-#if defined(CFE_API_write) || defined(CFE_API_ALL)
 int cfe_write(int handle, unsigned char *buffer, int length)
 {
 	return cfe_writeblk(handle, 0, buffer, length);
 }
-#endif				/* CFE_API_write || CFE_API_ALL */
 
-#if defined(CFE_API_writeblk) || defined(CFE_API_ALL)
-int
-cfe_writeblk(int handle, cfe_xint_t offset, unsigned char *buffer,
-	     int length)
+int cfe_writeblk(int handle, s64 offset, unsigned char *buffer, int length)
 {
-	cfe_xiocb_t xiocb;
+	struct cfe_xiocb xiocb;
 
 	xiocb.xiocb_fcode = CFE_CMD_DEV_WRITE;
 	xiocb.xiocb_status = 0;
 	xiocb.xiocb_handle = handle;
 	xiocb.xiocb_flags = 0;
-	xiocb.xiocb_psize = sizeof(xiocb_buffer_t);
+	xiocb.xiocb_psize = sizeof(struct xiocb_buffer);
 	xiocb.plist.xiocb_buffer.buf_offset = offset;
 	xiocb.plist.xiocb_buffer.buf_ptr = XPTR_FROM_NATIVE(buffer);
 	xiocb.plist.xiocb_buffer.buf_length = length;
@@ -499,4 +430,3 @@ cfe_writeblk(int handle, cfe_xint_t offset, unsigned char *buffer,
 		return xiocb.xiocb_status;
 	return xiocb.plist.xiocb_buffer.buf_retlen;
 }
-#endif				/* CFE_API_writeblk || CFE_API_ALL */
diff --git a/arch/mips/fw/cfe/cfe_api_int.h b/arch/mips/fw/cfe/cfe_api_int.h
index f7e5a64b55f3..d9759e646956 100644
--- a/arch/mips/fw/cfe/cfe_api_int.h
+++ b/arch/mips/fw/cfe/cfe_api_int.h
@@ -15,28 +15,12 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
-
-/*  *********************************************************************
-    *
-    *  Broadcom Common Firmware Environment (CFE)
-    *
-    *  Device function prototypes		File: cfe_api_int.h
-    *
-    *  This header defines all internal types and macros for the
-    *  library.  This is stuff that's not exported to an app
-    *  using the library.
-    *
-    *  Authors:  Mitch Lichtenberg, Chris Demetriou
-    *
-    ********************************************************************* */
-
 #ifndef CFE_API_INT_H
 #define CFE_API_INT_H
 
-/*  *********************************************************************
-    *  Constants
-    ********************************************************************* */
-
+/*
+ * Constants.
+ */
 #define CFE_CMD_FW_GETINFO	0
 #define CFE_CMD_FW_RESTART	1
 #define CFE_CMD_FW_BOOT		2
@@ -64,89 +48,101 @@
 
 #define CFE_CMD_VENDOR_USE	0x8000	/* codes above this are for customer use */
 
-/*  *********************************************************************
-    *  Structures
-    ********************************************************************* */
+/*
+ * Structures.
+ */
 
-typedef uint64_t cfe_xuint_t;
-typedef int64_t cfe_xint_t;
-typedef int64_t cfe_xptr_t;
+/* eeek, signed "pointers" */
+typedef s64 cfe_xptr_t;
 
-typedef struct xiocb_buffer_s {
-	cfe_xuint_t buf_offset;		/* offset on device (bytes) */
+struct xiocb_buffer {
+	u64 buf_offset;		/* offset on device (bytes) */
 	cfe_xptr_t  buf_ptr;		/* pointer to a buffer */
-	cfe_xuint_t buf_length;		/* length of this buffer */
-	cfe_xuint_t buf_retlen;		/* returned length (for read ops) */
-	cfe_xuint_t buf_ioctlcmd;	/* IOCTL command (used only for IOCTLs) */
-} xiocb_buffer_t;
-
-#define buf_devflags buf_ioctlcmd	/* returned device info flags */
+	u64 buf_length;		/* length of this buffer */
+	u64 buf_retlen;		/* returned length (for read ops) */
+	u64 buf_ioctlcmd;	/* IOCTL command (used only for IOCTLs) */
+};
 
-typedef struct xiocb_inpstat_s {
-	cfe_xuint_t inp_status;		/* 1 means input available */
-} xiocb_inpstat_t;
+struct xiocb_inpstat {
+	u64 inp_status;		/* 1 means input available */
+};
 
-typedef struct xiocb_envbuf_s {
-	cfe_xint_t enum_idx;		/* 0-based enumeration index */
+struct xiocb_envbuf {
+	s64 enum_idx;		/* 0-based enumeration index */
 	cfe_xptr_t name_ptr;		/* name string buffer */
-	cfe_xint_t name_length;		/* size of name buffer */
+	s64 name_length;		/* size of name buffer */
 	cfe_xptr_t val_ptr;		/* value string buffer */
-	cfe_xint_t val_length;		/* size of value string buffer */
-} xiocb_envbuf_t;
-
-typedef struct xiocb_cpuctl_s {
-	cfe_xuint_t cpu_number;		/* cpu number to control */
-	cfe_xuint_t cpu_command;	/* command to issue to CPU */
-	cfe_xuint_t start_addr;		/* CPU start address */
-	cfe_xuint_t gp_val;		/* starting GP value */
-	cfe_xuint_t sp_val;		/* starting SP value */
-	cfe_xuint_t a1_val;		/* starting A1 value */
-} xiocb_cpuctl_t;
-
-typedef struct xiocb_time_s {
-	cfe_xint_t ticks;		/* current time in ticks */
-} xiocb_time_t;
-
-typedef struct xiocb_exitstat_s {
-	cfe_xint_t status;
-} xiocb_exitstat_t;
-
-typedef struct xiocb_meminfo_s {
-	cfe_xint_t mi_idx;		/* 0-based enumeration index */
-	cfe_xint_t mi_type;		/* type of memory block */
-	cfe_xuint_t mi_addr;		/* physical start address */
-	cfe_xuint_t mi_size;		/* block size */
-} xiocb_meminfo_t;
-
-typedef struct xiocb_fwinfo_s {
-	cfe_xint_t fwi_version;		/* major, minor, eco version */
-	cfe_xint_t fwi_totalmem;	/* total installed mem */
-	cfe_xint_t fwi_flags;		/* various flags */
-	cfe_xint_t fwi_boardid;		/* board ID */
-	cfe_xint_t fwi_bootarea_va;	/* VA of boot area */
-	cfe_xint_t fwi_bootarea_pa;	/* PA of boot area */
-	cfe_xint_t fwi_bootarea_size;	/* size of boot area */
-	cfe_xint_t fwi_reserved1;
-	cfe_xint_t fwi_reserved2;
-	cfe_xint_t fwi_reserved3;
-} xiocb_fwinfo_t;
-
-typedef struct cfe_xiocb_s {
-	cfe_xuint_t xiocb_fcode;	/* IOCB function code */
-	cfe_xint_t xiocb_status;	/* return status */
-	cfe_xint_t xiocb_handle;	/* file/device handle */
-	cfe_xuint_t xiocb_flags;	/* flags for this IOCB */
-	cfe_xuint_t xiocb_psize;	/* size of parameter list */
+	s64 val_length;		/* size of value string buffer */
+};
+
+struct xiocb_cpuctl {
+	u64 cpu_number;		/* cpu number to control */
+	u64 cpu_command;	/* command to issue to CPU */
+	u64 start_addr;		/* CPU start address */
+	u64 gp_val;		/* starting GP value */
+	u64 sp_val;		/* starting SP value */
+	u64 a1_val;		/* starting A1 value */
+};
+
+struct xiocb_time {
+	s64 ticks;		/* current time in ticks */
+};
+
+struct xiocb_exitstat{
+	s64 status;
+};
+
+struct xiocb_meminfo {
+	s64 mi_idx;		/* 0-based enumeration index */
+	s64 mi_type;		/* type of memory block */
+	u64 mi_addr;		/* physical start address */
+	u64 mi_size;		/* block size */
+};
+
+struct xiocb_fwinfo {
+	s64 fwi_version;		/* major, minor, eco version */
+	s64 fwi_totalmem;	/* total installed mem */
+	s64 fwi_flags;		/* various flags */
+	s64 fwi_boardid;		/* board ID */
+	s64 fwi_bootarea_va;	/* VA of boot area */
+	s64 fwi_bootarea_pa;	/* PA of boot area */
+	s64 fwi_bootarea_size;	/* size of boot area */
+	s64 fwi_reserved1;
+	s64 fwi_reserved2;
+	s64 fwi_reserved3;
+};
+
+struct cfe_xiocb {
+	u64 xiocb_fcode;	/* IOCB function code */
+	s64 xiocb_status;	/* return status */
+	s64 xiocb_handle;	/* file/device handle */
+	u64 xiocb_flags;	/* flags for this IOCB */
+	u64 xiocb_psize;	/* size of parameter list */
 	union {
-		xiocb_buffer_t xiocb_buffer;	/* buffer parameters */
-		xiocb_inpstat_t xiocb_inpstat;	/* input status parameters */
-		xiocb_envbuf_t xiocb_envbuf;	/* environment function parameters */
-		xiocb_cpuctl_t xiocb_cpuctl;	/* CPU control parameters */
-		xiocb_time_t xiocb_time;	/* timer parameters */
-		xiocb_meminfo_t xiocb_meminfo;	/* memory arena info parameters */
-		xiocb_fwinfo_t xiocb_fwinfo;	/* firmware information */
-		xiocb_exitstat_t xiocb_exitstat;	/* Exit Status */
+		/* buffer parameters */
+		struct xiocb_buffer xiocb_buffer;
+
+		/* input status parameters */
+		struct xiocb_inpstat xiocb_inpstat;
+
+		/* environment function parameters */
+		struct xiocb_envbuf xiocb_envbuf;
+
+		/* CPU control parameters */
+		struct xiocb_cpuctl xiocb_cpuctl;
+
+		/* timer parameters */
+		struct xiocb_time xiocb_time;
+
+		/* memory arena info parameters */
+		struct xiocb_meminfo xiocb_meminfo;
+
+		/* firmware information */
+		struct xiocb_fwinfo xiocb_fwinfo;
+
+		/* Exit Status */
+		struct xiocb_exitstat xiocb_exitstat;
 	} plist;
-} cfe_xiocb_t;
+};
 
-#endif				/* CFE_API_INT_H */
+#endif /* CFE_API_INT_H */
diff --git a/include/asm-mips/fw/cfe/cfe_api.h b/include/asm-mips/fw/cfe/cfe_api.h
index 1003e7156bfc..0995575db320 100644
--- a/include/asm-mips/fw/cfe/cfe_api.h
+++ b/include/asm-mips/fw/cfe/cfe_api.h
@@ -15,49 +15,27 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
-
-/*  *********************************************************************
-    *
-    *  Broadcom Common Firmware Environment (CFE)
-    *
-    *  Device function prototypes		File: cfe_api.h
-    *
-    *  This file contains declarations for doing callbacks to
-    *  cfe from an application.  It should be the only header
-    *  needed by the application to use this library
-    *
-    *  Authors:  Mitch Lichtenberg, Chris Demetriou
-    *
-    ********************************************************************* */
-
+/*
+ * Broadcom Common Firmware Environment (CFE)
+ *
+ * This file contains declarations for doing callbacks to
+ * cfe from an application.  It should be the only header
+ * needed by the application to use this library
+ *
+ * Authors:  Mitch Lichtenberg, Chris Demetriou
+ */
 #ifndef CFE_API_H
 #define CFE_API_H
 
-/*
- * Apply customizations here for different OSes.  These need to:
- *	* typedef uint64_t, int64_t, intptr_t, uintptr_t.
- *	* define cfe_strlen() if use of an existing function is desired.
- *	* define CFE_API_IMPL_NAMESPACE if API functions are to use
- *	  names in the implementation namespace.
- * Also, optionally, if the build environment does not do so automatically,
- * CFE_API_* can be defined here as desired.
- */
-/* Begin customization. */
 #include <linux/types.h>
 #include <linux/string.h>
 
 typedef long intptr_t;
 
-#define cfe_strlen strlen
 
-#define CFE_API_ALL
-#define CFE_API_STRLEN_CUSTOM
-/* End customization. */
-
-
-/*  *********************************************************************
-    *  Constants
-    ********************************************************************* */
+/*
+ * Constants
+ */
 
 /* Seal indicating CFE's presence, passed to user program. */
 #define CFE_EPTSEAL 0x43464531
@@ -108,55 +86,14 @@ typedef struct {
 } cfe_fwinfo_t;
 
 
-/*
- * cfe_strlen is handled specially: If already defined, it has been
- * overridden in this environment with a standard strlen-like function.
- */
-#ifdef cfe_strlen
-# define CFE_API_STRLEN_CUSTOM
-#else
-# ifdef CFE_API_IMPL_NAMESPACE
-#  define cfe_strlen(a)			__cfe_strlen(a)
-# endif
-int cfe_strlen(char *name);
-#endif
-
 /*
  * Defines and prototypes for functions which take no arguments.
  */
-#ifdef CFE_API_IMPL_NAMESPACE
-int64_t __cfe_getticks(void);
-#define cfe_getticks()			__cfe_getticks()
-#else
 int64_t cfe_getticks(void);
-#endif
 
 /*
  * Defines and prototypes for the rest of the functions.
  */
-#ifdef CFE_API_IMPL_NAMESPACE
-#define cfe_close(a)			__cfe_close(a)
-#define cfe_cpu_start(a, b, c, d, e)	__cfe_cpu_start(a, b, c, d, e)
-#define cfe_cpu_stop(a)			__cfe_cpu_stop(a)
-#define cfe_enumenv(a, b, d, e, f)	__cfe_enumenv(a, b, d, e, f)
-#define cfe_enummem(a, b, c, d, e)	__cfe_enummem(a, b, c, d, e)
-#define cfe_exit(a, b)			__cfe_exit(a, b)
-#define cfe_flushcache(a)		__cfe_cacheflush(a)
-#define cfe_getdevinfo(a)		__cfe_getdevinfo(a)
-#define cfe_getenv(a, b, c)		__cfe_getenv(a, b, c)
-#define cfe_getfwinfo(a)		__cfe_getfwinfo(a)
-#define cfe_getstdhandle(a)		__cfe_getstdhandle(a)
-#define cfe_init(a, b)			__cfe_init(a, b)
-#define cfe_inpstat(a)			__cfe_inpstat(a)
-#define cfe_ioctl(a, b, c, d, e, f)	__cfe_ioctl(a, b, c, d, e, f)
-#define cfe_open(a)			__cfe_open(a)
-#define cfe_read(a, b, c)		__cfe_read(a, b, c)
-#define cfe_readblk(a, b, c, d)		__cfe_readblk(a, b, c, d)
-#define cfe_setenv(a, b)		__cfe_setenv(a, b)
-#define cfe_write(a, b, c)		__cfe_write(a, b, c)
-#define cfe_writeblk(a, b, c, d)	__cfe_writeblk(a, b, c, d)
-#endif				/* CFE_API_IMPL_NAMESPACE */
-
 int cfe_close(int handle);
 int cfe_cpu_start(int cpu, void (*fn) (void), long sp, long gp, long a1);
 int cfe_cpu_stop(int cpu);
diff --git a/include/asm-mips/fw/cfe/cfe_error.h b/include/asm-mips/fw/cfe/cfe_error.h
index 975f00002cbe..b80374636279 100644
--- a/include/asm-mips/fw/cfe/cfe_error.h
+++ b/include/asm-mips/fw/cfe/cfe_error.h
@@ -16,18 +16,13 @@
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
-/*  *********************************************************************
-    *
-    *  Broadcom Common Firmware Environment (CFE)
-    *
-    *  Error codes				File: cfe_error.h
-    *
-    *  CFE's global error code list is here.
-    *
-    *  Author:  Mitch Lichtenberg
-    *
-    ********************************************************************* */
-
+/*
+ * Broadcom Common Firmware Environment (CFE)
+ *
+ * CFE's global error code list is here.
+ *
+ * Author:  Mitch Lichtenberg
+ */
 
 #define CFE_OK			 0
 #define CFE_ERR                 -1	/* generic error */
-- 
cgit v1.2.3


From 7f3f4900619743f4e3af21e9228748c3d692a2c2 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Fri, 26 Oct 2007 22:42:31 +0900
Subject: [MIPS] remove unused mips_machtype

Removed unused mips_machtype.  These are only set but not used.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/au1000/db1x00/init.c               |  11 ---
 arch/mips/au1000/mtx-1/init.c                |   2 -
 arch/mips/au1000/pb1000/init.c               |   2 -
 arch/mips/au1000/pb1100/init.c               |   2 -
 arch/mips/au1000/pb1200/init.c               |   2 -
 arch/mips/au1000/pb1500/init.c               |   2 -
 arch/mips/au1000/pb1550/init.c               |   2 -
 arch/mips/au1000/xxs1500/init.c              |   2 -
 arch/mips/basler/excite/excite_prom.c        |   2 -
 arch/mips/gt64120/wrppmc/setup.c             |   2 -
 arch/mips/jmr3927/rbhma3100/init.c           |   4 --
 arch/mips/lemote/lm2e/prom.c                 |   2 -
 arch/mips/philips/pnx8550/jbs/init.c         |   3 -
 arch/mips/philips/pnx8550/stb810/prom_init.c |   2 -
 arch/mips/pmc-sierra/yosemite/prom.c         |   2 -
 arch/mips/tx4938/toshiba_rbtx4938/prom.c     |   1 -
 arch/mips/vr41xx/nec-cmbvr4133/setup.c       |   2 -
 include/asm-mips/bootinfo.h                  | 104 ---------------------------
 18 files changed, 149 deletions(-)

(limited to 'include')

diff --git a/arch/mips/au1000/db1x00/init.c b/arch/mips/au1000/db1x00/init.c
index 43298fd9459c..e822c123eab8 100644
--- a/arch/mips/au1000/db1x00/init.c
+++ b/arch/mips/au1000/db1x00/init.c
@@ -57,17 +57,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	/* Set the platform # */
-#if	defined(CONFIG_MIPS_DB1550)
-	mips_machtype = MACH_DB1550;
-#elif	defined(CONFIG_MIPS_DB1500)
-	mips_machtype = MACH_DB1500;
-#elif	defined(CONFIG_MIPS_DB1100)
-	mips_machtype = MACH_DB1100;
-#else
-	mips_machtype = MACH_DB1000;
-#endif
-
 	prom_init_cmdline();
 
 	memsize_str = prom_getenv("memsize");
diff --git a/arch/mips/au1000/mtx-1/init.c b/arch/mips/au1000/mtx-1/init.c
index cdeae3212a2d..e700fd312a24 100644
--- a/arch/mips/au1000/mtx-1/init.c
+++ b/arch/mips/au1000/mtx-1/init.c
@@ -54,8 +54,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_MTX1;	/* set the platform # */
-
 	prom_init_cmdline();
 
 	memsize_str = prom_getenv("memsize");
diff --git a/arch/mips/au1000/pb1000/init.c b/arch/mips/au1000/pb1000/init.c
index ddccaf6997d0..2515b9fb24af 100644
--- a/arch/mips/au1000/pb1000/init.c
+++ b/arch/mips/au1000/pb1000/init.c
@@ -52,8 +52,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_PB1000;
-
 	prom_init_cmdline();
 	memsize_str = prom_getenv("memsize");
 	if (!memsize_str) {
diff --git a/arch/mips/au1000/pb1100/init.c b/arch/mips/au1000/pb1100/init.c
index c93fd39b4aba..490c3801c275 100644
--- a/arch/mips/au1000/pb1100/init.c
+++ b/arch/mips/au1000/pb1100/init.c
@@ -53,8 +53,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg3;
 
-	mips_machtype = MACH_PB1100;
-
 	prom_init_cmdline();
 
 	memsize_str = prom_getenv("memsize");
diff --git a/arch/mips/au1000/pb1200/init.c b/arch/mips/au1000/pb1200/init.c
index c251570749ee..069ed45f04f2 100644
--- a/arch/mips/au1000/pb1200/init.c
+++ b/arch/mips/au1000/pb1200/init.c
@@ -53,8 +53,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_PB1200;
-
 	prom_init_cmdline();
 	memsize_str = prom_getenv("memsize");
 	if (!memsize_str) {
diff --git a/arch/mips/au1000/pb1500/init.c b/arch/mips/au1000/pb1500/init.c
index 507d4b204161..db558c967048 100644
--- a/arch/mips/au1000/pb1500/init.c
+++ b/arch/mips/au1000/pb1500/init.c
@@ -53,8 +53,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_PB1500;
-
 	prom_init_cmdline();
 	memsize_str = prom_getenv("memsize");
 	if (!memsize_str) {
diff --git a/arch/mips/au1000/pb1550/init.c b/arch/mips/au1000/pb1550/init.c
index b03eee601e36..b716363ea564 100644
--- a/arch/mips/au1000/pb1550/init.c
+++ b/arch/mips/au1000/pb1550/init.c
@@ -53,8 +53,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_PB1550;
-
 	prom_init_cmdline();
 	memsize_str = prom_getenv("memsize");
 	if (!memsize_str) {
diff --git a/arch/mips/au1000/xxs1500/init.c b/arch/mips/au1000/xxs1500/init.c
index 6532939f377a..7e6878c1b0a5 100644
--- a/arch/mips/au1000/xxs1500/init.c
+++ b/arch/mips/au1000/xxs1500/init.c
@@ -52,8 +52,6 @@ void __init prom_init(void)
 	prom_argv = (char **) fw_arg1;
 	prom_envp = (char **) fw_arg2;
 
-	mips_machtype = MACH_XXS1500;	/* set the platform # */
-
 	prom_init_cmdline();
 
 	memsize_str = prom_getenv("memsize");
diff --git a/arch/mips/basler/excite/excite_prom.c b/arch/mips/basler/excite/excite_prom.c
index 2d752c2f6e59..68d8bc597e34 100644
--- a/arch/mips/basler/excite/excite_prom.c
+++ b/arch/mips/basler/excite/excite_prom.c
@@ -135,8 +135,6 @@ void __init prom_init(void)
 #ifdef CONFIG_64BIT
 #	error 64 bit support not implemented
 #endif /* CONFIG_64BIT */
-
-	mips_machtype = MACH_TITAN_EXCITE;
 }
 
 /* This is called from free_initmem(), so we need to provide it */
diff --git a/arch/mips/gt64120/wrppmc/setup.c b/arch/mips/gt64120/wrppmc/setup.c
index 51f6b7862460..728ef6a80edd 100644
--- a/arch/mips/gt64120/wrppmc/setup.c
+++ b/arch/mips/gt64120/wrppmc/setup.c
@@ -121,8 +121,6 @@ const char *get_system_type(void)
  */
 void __init prom_init(void)
 {
-	mips_machtype = MACH_WRPPMC;
-
 	add_memory_region(WRPPMC_SDRAM_SCS0_BASE, WRPPMC_SDRAM_SCS0_SIZE, BOOT_MEM_RAM);
 	add_memory_region(WRPPMC_BOOTROM_BASE, WRPPMC_BOOTROM_SIZE, BOOT_MEM_ROM_DATA);
 
diff --git a/arch/mips/jmr3927/rbhma3100/init.c b/arch/mips/jmr3927/rbhma3100/init.c
index b643f75ec9a5..700b9cf8eb9d 100644
--- a/arch/mips/jmr3927/rbhma3100/init.c
+++ b/arch/mips/jmr3927/rbhma3100/init.c
@@ -52,10 +52,6 @@ void __init prom_init(void)
 		puts("Warning: TX3927 TLB off\n");
 #endif
 
-#ifdef CONFIG_TOSHIBA_JMR3927
-	mips_machtype = MACH_TOSHIBA_JMR3927;
-#endif
-
 	prom_init_cmdline();
 	add_memory_region(0, JMR3927_SDRAM_SIZE, BOOT_MEM_RAM);
 }
diff --git a/arch/mips/lemote/lm2e/prom.c b/arch/mips/lemote/lm2e/prom.c
index 824336812198..7edc15dfed6c 100644
--- a/arch/mips/lemote/lm2e/prom.c
+++ b/arch/mips/lemote/lm2e/prom.c
@@ -57,8 +57,6 @@ void __init prom_init(void)
 	arg = (int *)fw_arg1;
 	env = (int *)fw_arg2;
 
-	mips_machtype = MACH_LEMOTE_FULONG;
-
 	prom_init_cmdline();
 
 	if ((strstr(arcs_cmdline, "console=")) == NULL)
diff --git a/arch/mips/philips/pnx8550/jbs/init.c b/arch/mips/philips/pnx8550/jbs/init.c
index cfd90fa3d799..90b4d35f3ece 100644
--- a/arch/mips/philips/pnx8550/jbs/init.c
+++ b/arch/mips/philips/pnx8550/jbs/init.c
@@ -45,11 +45,8 @@ const char *get_system_type(void)
 
 void __init prom_init(void)
 {
-
 	unsigned long memsize;
 
-	mips_machtype = MACH_PHILIPS_JBS;
-
 	//memsize = 0x02800000; /* Trimedia uses memory above */
 	memsize = 0x08000000; /* Trimedia uses memory above */
 	add_memory_region(0, memsize, BOOT_MEM_RAM);
diff --git a/arch/mips/philips/pnx8550/stb810/prom_init.c b/arch/mips/philips/pnx8550/stb810/prom_init.c
index fdb33ed089b9..832dd60b0a7a 100644
--- a/arch/mips/philips/pnx8550/stb810/prom_init.c
+++ b/arch/mips/philips/pnx8550/stb810/prom_init.c
@@ -41,8 +41,6 @@ void __init prom_init(void)
 
 	prom_init_cmdline();
 
-	mips_machtype = MACH_PHILIPS_STB810;
-
 	memsize = 0x08000000; /* Trimedia uses memory above */
 	add_memory_region(0, memsize, BOOT_MEM_RAM);
 }
diff --git a/arch/mips/pmc-sierra/yosemite/prom.c b/arch/mips/pmc-sierra/yosemite/prom.c
index 9b9936de6589..96d3ff051d3d 100644
--- a/arch/mips/pmc-sierra/yosemite/prom.c
+++ b/arch/mips/pmc-sierra/yosemite/prom.c
@@ -126,8 +126,6 @@ void __init prom_init(void)
 		env++;
 	}
 
-	mips_machtype = MACH_TITAN_YOSEMITE;
-
 	prom_grab_secondary();
 }
 
diff --git a/arch/mips/tx4938/toshiba_rbtx4938/prom.c b/arch/mips/tx4938/toshiba_rbtx4938/prom.c
index 69f21c1b7942..1644bffa501a 100644
--- a/arch/mips/tx4938/toshiba_rbtx4938/prom.c
+++ b/arch/mips/tx4938/toshiba_rbtx4938/prom.c
@@ -47,7 +47,6 @@ void __init prom_init(void)
 #ifndef CONFIG_TX4938_NAND_BOOT
 	prom_init_cmdline();
 #endif
-	mips_machtype = MACH_TOSHIBA_RBTX4938;
 
 	msize = tx4938_get_mem_size();
 	add_memory_region(0, msize << 20, BOOT_MEM_RAM);
diff --git a/arch/mips/vr41xx/nec-cmbvr4133/setup.c b/arch/mips/vr41xx/nec-cmbvr4133/setup.c
index f14bea8daf90..7723d2011b08 100644
--- a/arch/mips/vr41xx/nec-cmbvr4133/setup.c
+++ b/arch/mips/vr41xx/nec-cmbvr4133/setup.c
@@ -64,8 +64,6 @@ static void __init nec_cmbvr4133_setup(void)
 #endif
 	set_io_port_base(KSEG1ADDR(0x16000000));
 
-	mips_machtype = MACH_NEC_CMBVR4133;
-
 #ifdef CONFIG_PCI
 #ifdef CONFIG_ROCKHOPPER
 	ali_m5229_preinit();
diff --git a/include/asm-mips/bootinfo.h b/include/asm-mips/bootinfo.h
index b2dd9b33de8f..8e9bb16f0aa2 100644
--- a/include/asm-mips/bootinfo.h
+++ b/include/asm-mips/bootinfo.h
@@ -47,22 +47,11 @@
 #define  MACH_DS5800		9	/* DECsystem 5800		*/
 #define  MACH_DS5900		10	/* DECsystem 5900		*/
 
-/*
- * Valid machtype for group ARC
- */
-#define MACH_DESKSTATION_RPC44  0	/* Deskstation rPC44 */
-#define MACH_DESKSTATION_TYNE	1	/* Deskstation Tyne */
-
 /*
  * Valid machtype for group SNI_RM
  */
 #define  MACH_SNI_RM200_PCI	0	/* RM200/RM300/RM400 PCI series */
 
-/*
- * Valid machtype for group ACN
- */
-#define  MACH_ACN_MIPS_BOARD	0       /* ACN MIPS single board        */
-
 /*
  * Valid machtype for group SGI
  */
@@ -72,44 +61,6 @@
 #define  MACH_SGI_IP32		3	/* O2				*/
 #define  MACH_SGI_IP30		4	/* Octane, Octane2              */
 
-/*
- * Valid machtype for group COBALT
- */
-#define  MACH_COBALT_27		0	/* Proto "27" hardware		*/
-
-/*
- * Valid machtype for group BAGET
- */
-#define  MACH_BAGET201		0	/* BT23-201 */
-#define  MACH_BAGET202		1	/* BT23-202 */
-
-/*
- * Cosine boards.
- */
-#define  MACH_COSINE_ORION	0
-
-/*
- * Valid machtype for group MOMENCO
- */
-#define  MACH_MOMENCO_OCELOT	0
-#define  MACH_MOMENCO_OCELOT_G	1	/* no more supported (may 2007) */
-#define  MACH_MOMENCO_OCELOT_C	2	/* no more supported (jun 2007) */
-#define  MACH_MOMENCO_JAGUAR_ATX 3	/* no more supported (may 2007) */
-#define  MACH_MOMENCO_OCELOT_3	4
-
-/*
- * Valid machtype for group PHILIPS
- */
-#define  MACH_PHILIPS_NINO	0	/* Nino */
-#define  MACH_PHILIPS_VELO	1	/* Velo */
-#define  MACH_PHILIPS_JBS	2	/* JBS */
-#define  MACH_PHILIPS_STB810	3	/* STB810 */
-
-/*
- * Valid machtype for group SIBYTE
- */
-#define  MACH_SWARM              0
-
 /*
  * Valid machtypes for group Toshiba
  */
@@ -121,64 +72,17 @@
 #define  MACH_TOSHIBA_RBTX4937	5
 #define  MACH_TOSHIBA_RBTX4938	6
 
-/*
- * Valid machtype for group Alchemy
- */
-#define  MACH_PB1000		0	/* Au1000-based eval board */
-#define  MACH_PB1100		1	/* Au1100-based eval board */
-#define  MACH_PB1500		2	/* Au1500-based eval board */
-#define  MACH_DB1000		3       /* Au1000-based eval board */
-#define  MACH_DB1100		4       /* Au1100-based eval board */
-#define  MACH_DB1500		5       /* Au1500-based eval board */
-#define  MACH_XXS1500		6       /* Au1500-based eval board */
-#define  MACH_MTX1		7       /* 4G MTX-1 Au1500-based board */
-#define  MACH_PB1550		8       /* Au1550-based eval board */
-#define  MACH_DB1550		9       /* Au1550-based eval board */
-#define  MACH_PB1200		10       /* Au1200-based eval board */
-#define  MACH_DB1200		11       /* Au1200-based eval board */
-
-/*
- * Valid machtype for group NEC_VR41XX
- *
- * Various NEC-based devices.
- *
- * FIXME: MACH_GROUPs should be by _MANUFACTURER_ of * the device, not by
- *        technical properties, so no new additions to this group.
- */
-#define  MACH_NEC_OSPREY	0	/* Osprey eval board */
-#define  MACH_NEC_EAGLE		1	/* NEC Eagle/Hawk board */
-#define  MACH_ZAO_CAPCELLA	2	/* ZAO Networks Capcella */
-#define  MACH_VICTOR_MPC30X	3	/* Victor MP-C303/304 */
-#define  MACH_IBM_WORKPAD	4	/* IBM WorkPad z50 */
-#define  MACH_CASIO_E55		5	/* CASIO CASSIOPEIA E-10/15/55/65 */
-#define  MACH_TANBAC_TB0226	6	/* TANBAC TB0226 (Mbase) */
-#define  MACH_TANBAC_TB0229	7	/* TANBAC TB0229 (VR4131DIMM) */
-#define  MACH_NEC_CMBVR4133	8	/* CMB VR4133 Board */
-
-#define  MACH_HP_LASERJET	1
-
 /*
  * Valid machtype for group LASAT
  */
 #define  MACH_LASAT_100		0	/* Masquerade II/SP100/SP50/SP25 */
 #define  MACH_LASAT_200		1	/* Masquerade PRO/SP200 */
 
-/*
- * Valid machtype for group TITAN
- */
-#define  MACH_TITAN_YOSEMITE	1	/* PMC-Sierra Yosemite		*/
-#define  MACH_TITAN_EXCITE	2	/* Basler eXcite		*/
-
 /*
  * Valid machtype for group NEC EMMA2RH
  */
 #define  MACH_NEC_MARKEINS	0	/* NEC EMMA2RH Mark-eins	*/
 
-/*
- * Valid machtype for group LEMOTE
- */
-#define  MACH_LEMOTE_FULONG        0
-
 /*
  * Valid machtype for group PMC-MSP
  */
@@ -190,14 +94,6 @@
 #define MACH_MSP7120_FPGA       5	/* PMC-Sierra MSP7120 Emulation */
 #define MACH_MSP_OTHER        255	/* PMC-Sierra unknown board type */
 
-#define MACH_WRPPMC             1
-
-/*
- * Valid machtype for group Broadcom
- */
-#define MACH_GROUP_BRCM		23	/* Broadcom			*/
-#define  MACH_BCM47XX		1	/* Broadcom BCM47XX		*/
-
 #define CL_SIZE			COMMAND_LINE_SIZE
 
 const char *get_system_type(void);
-- 
cgit v1.2.3


From 48ef2626aeecac3b160d1aee38ac46d6c3540122 Mon Sep 17 00:00:00 2001
From: Andrew Sharp <andy.sharp@onstor.com>
Date: Wed, 31 Oct 2007 14:11:24 -0700
Subject: [MIPS] Put cast inside macro instead of all the callers

Since all the callers of the PHYS_TO_XKPHYS macro call with a constant,
put the cast to LL inside the macro where it really should be rather
than in all the callers.  This makes macros like PHYS_TO_XKSEG_UNCACHED
work without gcc whining.

Signed-off-by: Andrew Sharp <andy.sharp@onstor.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/lib/uncached.c     | 12 ++++++------
 include/asm-mips/addrspace.h |  2 +-
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/mips/lib/uncached.c b/arch/mips/lib/uncached.c
index 58d14f4d9349..27b012d4341c 100644
--- a/arch/mips/lib/uncached.c
+++ b/arch/mips/lib/uncached.c
@@ -46,9 +46,9 @@ unsigned long __init run_uncached(void *func)
 	if (sp >= (long)CKSEG0 && sp < (long)CKSEG2)
 		usp = CKSEG1ADDR(sp);
 #ifdef CONFIG_64BIT
-	else if ((long long)sp >= (long long)PHYS_TO_XKPHYS(0LL, 0) &&
-		 (long long)sp < (long long)PHYS_TO_XKPHYS(8LL, 0))
-		usp = PHYS_TO_XKPHYS((long long)K_CALG_UNCACHED,
+	else if ((long long)sp >= (long long)PHYS_TO_XKPHYS(0, 0) &&
+		 (long long)sp < (long long)PHYS_TO_XKPHYS(8, 0))
+		usp = PHYS_TO_XKPHYS(K_CALG_UNCACHED,
 				     XKPHYS_TO_PHYS((long long)sp));
 #endif
 	else {
@@ -58,9 +58,9 @@ unsigned long __init run_uncached(void *func)
 	if (lfunc >= (long)CKSEG0 && lfunc < (long)CKSEG2)
 		ufunc = CKSEG1ADDR(lfunc);
 #ifdef CONFIG_64BIT
-	else if ((long long)lfunc >= (long long)PHYS_TO_XKPHYS(0LL, 0) &&
-		 (long long)lfunc < (long long)PHYS_TO_XKPHYS(8LL, 0))
-		ufunc = PHYS_TO_XKPHYS((long long)K_CALG_UNCACHED,
+	else if ((long long)lfunc >= (long long)PHYS_TO_XKPHYS(0, 0) &&
+		 (long long)lfunc < (long long)PHYS_TO_XKPHYS(8, 0))
+		ufunc = PHYS_TO_XKPHYS(K_CALG_UNCACHED,
 				       XKPHYS_TO_PHYS((long long)lfunc));
 #endif
 	else {
diff --git a/include/asm-mips/addrspace.h b/include/asm-mips/addrspace.h
index 0bb7a93b7a5e..569f80aacbd2 100644
--- a/include/asm-mips/addrspace.h
+++ b/include/asm-mips/addrspace.h
@@ -127,7 +127,7 @@
 #define PHYS_TO_XKSEG_CACHED(p)		PHYS_TO_XKPHYS(K_CALG_COH_SHAREABLE, (p))
 #define XKPHYS_TO_PHYS(p)		((p) & TO_PHYS_MASK)
 #define PHYS_TO_XKPHYS(cm, a)		(_CONST64_(0x8000000000000000) | \
-					 ((cm)<<59) | (a))
+					 (_CONST64_(cm) << 59) | (a))
 
 /*
  * The ultimate limited of the 64-bit MIPS architecture:  2 bits for selecting
-- 
cgit v1.2.3


From 0ab7aefc4d43a6dee26c891b41ef9c7a67d2379b Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 2 Mar 2007 20:42:04 +0000
Subject: [MIPS] MT: Scheduler support for SMT

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig           | 14 ++++++++++++++
 arch/mips/kernel/proc.c     |  1 +
 arch/mips/kernel/smp-mt.c   |  6 +++++-
 arch/mips/kernel/smp.c      | 30 ++++++++++++++++++++++++++++++
 include/asm-mips/cpu-info.h |  5 ++++-
 include/asm-mips/smp.h      |  3 +++
 include/asm-mips/topology.h | 16 ++++++++++++++++
 7 files changed, 73 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 11bc17ce0ebf..d5a89f3fdfd3 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1442,6 +1442,7 @@ config MIPS_MT_SMP
 	select MIPS_MT
 	select NR_CPUS_DEFAULT_2
 	select SMP
+	select SYS_SUPPORTS_SCHED_SMT if SMP
 	select SYS_SUPPORTS_SMP
 	help
 	  This is a kernel model which is also known a VSMP or lately
@@ -1468,6 +1469,19 @@ endchoice
 config MIPS_MT
 	bool
 
+config SCHED_SMT
+	bool "SMT (multithreading) scheduler support"
+	depends on SYS_SUPPORTS_SCHED_SMT
+	default n
+	help
+	  SMT scheduler support improves the CPU scheduler's decision making
+	  when dealing with MIPS MT enabled cores at a cost of slightly
+	  increased overhead in some places. If unsure say N here.
+
+config SYS_SUPPORTS_SCHED_SMT
+	bool
+
+
 config SYS_SUPPORTS_MULTITHREADING
 	bool
 
diff --git a/arch/mips/kernel/proc.c b/arch/mips/kernel/proc.c
index 6e6e947cce1e..34dd22838fdb 100644
--- a/arch/mips/kernel/proc.c
+++ b/arch/mips/kernel/proc.c
@@ -62,6 +62,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		);
 	seq_printf(m, "shadow register sets\t: %d\n",
 		       cpu_data[n].srsets);
+	seq_printf(m, "core\t\t\t: %d\n", cpu_data[n].core);
 
 	sprintf(fmt, "VCE%%c exceptions\t\t: %s\n",
 	        cpu_has_vce ? "%u" : "not available");
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index 94e210cc6cb6..2ab0b7eeaa7e 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -22,6 +22,7 @@
 #include <linux/cpumask.h>
 #include <linux/interrupt.h>
 #include <linux/compiler.h>
+#include <linux/smp.h>
 
 #include <asm/atomic.h>
 #include <asm/cacheflush.h>
@@ -30,7 +31,6 @@
 #include <asm/system.h>
 #include <asm/hardirq.h>
 #include <asm/mmu_context.h>
-#include <asm/smp.h>
 #include <asm/time.h>
 #include <asm/mipsregs.h>
 #include <asm/mipsmtregs.h>
@@ -223,6 +223,7 @@ static void __init smp_tc_init(unsigned int tc, unsigned int mvpconf0)
 void __init plat_smp_setup(void)
 {
 	unsigned int mvpconf0, ntc, tc, ncpu = 0;
+	unsigned int nvpe;
 
 #ifdef CONFIG_MIPS_MT_FPAFF
 	/* If we have an FPU, enroll ourselves in the FPU-full mask */
@@ -242,6 +243,9 @@ void __init plat_smp_setup(void)
 	mvpconf0 = read_c0_mvpconf0();
 	ntc = (mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT;
 
+	nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+	smp_num_siblings = nvpe;
+
 	/* we'll always have more TC's than VPE's, so loop setting everything
 	   to a sensible state */
 	for (tc = 0; tc <= ntc; tc++) {
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 63989e9df4f9..335be9bcf0dc 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -56,6 +56,34 @@ EXPORT_SYMBOL(cpu_online_map);
 extern void __init calibrate_delay(void);
 extern void cpu_idle(void);
 
+/* Number of TCs (or siblings in Intel speak) per CPU core */
+int smp_num_siblings = 1;
+EXPORT_SYMBOL(smp_num_siblings);
+
+/* representing the TCs (or siblings in Intel speak) of each logical CPU */
+cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(cpu_sibling_map);
+
+/* representing cpus for which sibling maps can be computed */
+static cpumask_t cpu_sibling_setup_map;
+
+static inline void set_cpu_sibling_map(int cpu)
+{
+	int i;
+
+	cpu_set(cpu, cpu_sibling_setup_map);
+
+	if (smp_num_siblings > 1) {
+		for_each_cpu_mask(i, cpu_sibling_setup_map) {
+			if (cpu_data[cpu].core == cpu_data[i].core) {
+				cpu_set(i, cpu_sibling_map[cpu]);
+				cpu_set(cpu, cpu_sibling_map[i]);
+			}
+		}
+	} else
+		cpu_set(cpu, cpu_sibling_map[cpu]);
+}
+
 /*
  * First C code run on the secondary CPUs after being started up by
  * the master.
@@ -85,6 +113,7 @@ asmlinkage __cpuinit void start_secondary(void)
 	cpu_data[cpu].udelay_val = loops_per_jiffy;
 
 	prom_smp_finish();
+	set_cpu_sibling_map(cpu);
 
 	cpu_set(cpu, cpu_callin_map);
 
@@ -258,6 +287,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	init_new_context(current, &init_mm);
 	current_thread_info()->cpu = 0;
 	plat_prepare_cpus(max_cpus);
+	set_cpu_sibling_map(0);
 #ifndef CONFIG_HOTPLUG_CPU
 	cpu_present_map = cpu_possible_map;
 #endif
diff --git a/include/asm-mips/cpu-info.h b/include/asm-mips/cpu-info.h
index ed5c02c6afbb..0c5a358863f3 100644
--- a/include/asm-mips/cpu-info.h
+++ b/include/asm-mips/cpu-info.h
@@ -55,6 +55,7 @@ struct cpuinfo_mips {
 	struct cache_desc	scache;	/* Secondary cache */
 	struct cache_desc	tcache;	/* Tertiary/split secondary cache */
 	int			srsets;	/* Shadow register sets */
+	int			core;	/* physical core number */
 #if defined(CONFIG_MIPS_MT_SMTC)
 	/*
 	 * In the MIPS MT "SMTC" model, each TC is considered
@@ -63,8 +64,10 @@ struct cpuinfo_mips {
 	 * to all TCs within the same VPE.
 	 */
 	int			vpe_id;  /* Virtual Processor number */
-	int			tc_id;   /* Thread Context number */
 #endif /* CONFIG_MIPS_MT */
+#ifdef CONFIG_MIPS_MT_SMTC
+	int			tc_id;   /* Thread Context number */
+#endif
 	void 			*data;	/* Additional data */
 } __attribute__((aligned(SMP_CACHE_BYTES)));
 
diff --git a/include/asm-mips/smp.h b/include/asm-mips/smp.h
index dc770025a9b0..23265879cee9 100644
--- a/include/asm-mips/smp.h
+++ b/include/asm-mips/smp.h
@@ -20,6 +20,9 @@
 #include <linux/cpumask.h>
 #include <asm/atomic.h>
 
+extern int smp_num_siblings;
+extern cpumask_t cpu_sibling_map[];
+
 #define raw_smp_processor_id() (current_thread_info()->cpu)
 
 /* Map from cpu id to sequential logical cpu number.  This will only
diff --git a/include/asm-mips/topology.h b/include/asm-mips/topology.h
index 0440fb9f2180..259145e07e97 100644
--- a/include/asm-mips/topology.h
+++ b/include/asm-mips/topology.h
@@ -1 +1,17 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2007 by Ralf Baechle
+ */
+#ifndef __ASM_TOPOLOGY_H
+#define __ASM_TOPOLOGY_H
+
 #include <topology.h>
+
+#ifdef CONFIG_SMP
+#define smt_capable()   (smp_num_siblings > 1)
+#endif
+
+#endif /* __ASM_TOPOLOGY_H */
-- 
cgit v1.2.3


From 87353d8ac39c52784da605ecbe965ecdfad609ad Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 19 Nov 2007 12:23:51 +0000
Subject: [MIPS] SMP: Call platform methods via ops structure.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig                        |   5 +
 arch/mips/fw/arc/init.c                  |   8 ++
 arch/mips/kernel/mips-mt.c               |   1 -
 arch/mips/kernel/setup.c                 |   3 +-
 arch/mips/kernel/smp-mt.c                | 193 +++++++++++++++++--------------
 arch/mips/kernel/smp.c                   |  23 ++--
 arch/mips/kernel/smtc-proc.c             |   1 -
 arch/mips/kernel/smtc.c                  |   1 -
 arch/mips/mips-boards/generic/init.c     |   8 ++
 arch/mips/mips-boards/malta/malta_smtc.c |  66 ++++++-----
 arch/mips/mipssim/Makefile               |   2 +-
 arch/mips/mipssim/sim_setup.c            |  16 ++-
 arch/mips/mipssim/sim_smp.c              | 123 --------------------
 arch/mips/mipssim/sim_smtc.c             | 117 +++++++++++++++++++
 arch/mips/pmc-sierra/yosemite/prom.c     |   5 +
 arch/mips/pmc-sierra/yosemite/smp.c      | 149 +++++++++++++-----------
 arch/mips/qemu/q-smp.c                   |  40 ++++---
 arch/mips/sgi-ip27/ip27-init.c           |   1 -
 arch/mips/sgi-ip27/ip27-klnuma.c         |   1 -
 arch/mips/sgi-ip27/ip27-smp.c            | 109 ++++++++++-------
 arch/mips/sibyte/bcm1480/smp.c           | 101 ++++++++++++++--
 arch/mips/sibyte/cfe/Makefile            |   1 -
 arch/mips/sibyte/cfe/setup.c             |  11 ++
 arch/mips/sibyte/cfe/smp.c               | 110 ------------------
 arch/mips/sibyte/sb1250/smp.c            | 100 ++++++++++++++--
 include/asm-mips/sibyte/sb1250.h         |   2 -
 include/asm-mips/smp-ops.h               |  56 +++++++++
 include/asm-mips/smp.h                   |  61 +---------
 28 files changed, 755 insertions(+), 559 deletions(-)
 delete mode 100644 arch/mips/mipssim/sim_smp.c
 create mode 100644 arch/mips/mipssim/sim_smtc.c
 delete mode 100644 arch/mips/sibyte/cfe/smp.c
 create mode 100644 include/asm-mips/smp-ops.h

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 8cbdfd255670..b211e7961f28 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1441,6 +1441,7 @@ config MIPS_MT_SMP
 	select SMP
 	select SYS_SUPPORTS_SCHED_SMT if SMP
 	select SYS_SUPPORTS_SMP
+	select SMP_UP
 	help
 	  This is a kernel model which is also known a VSMP or lately
 	  has been marketesed into SMVP.
@@ -1457,6 +1458,7 @@ config MIPS_MT_SMTC
 	select NR_CPUS_DEFAULT_8
 	select SMP
 	select SYS_SUPPORTS_SMP
+	select SMP_UP
 	help
 	  This is a kernel model which is known a SMTC or lately has been
 	  marketesed into SMVP.
@@ -1735,6 +1737,9 @@ config SMP
 
 	  If you don't know what to do here, say N.
 
+config SMP_UP
+	bool
+
 config SYS_SUPPORTS_SMP
 	bool
 
diff --git a/arch/mips/fw/arc/init.c b/arch/mips/fw/arc/init.c
index e2f75b13312f..3ad8788b6eaa 100644
--- a/arch/mips/fw/arc/init.c
+++ b/arch/mips/fw/arc/init.c
@@ -12,6 +12,7 @@
 
 #include <asm/bootinfo.h>
 #include <asm/sgialib.h>
+#include <asm/smp-ops.h>
 
 #undef DEBUG_PROM_INIT
 
@@ -48,4 +49,11 @@ void __init prom_init(void)
 	ArcRead(0, &c, 1, &cnt);
 	ArcEnterInteractiveMode();
 #endif
+#ifdef CONFIG_SGI_IP27
+	{
+		extern struct plat_smp_ops ip27_smp_ops;
+
+		register_smp_ops(&ip27_smp_ops);
+	}
+#endif
 }
diff --git a/arch/mips/kernel/mips-mt.c b/arch/mips/kernel/mips-mt.c
index 3d6b1ec1f328..640fb0cc6e39 100644
--- a/arch/mips/kernel/mips-mt.c
+++ b/arch/mips/kernel/mips-mt.c
@@ -17,7 +17,6 @@
 #include <asm/system.h>
 #include <asm/hardirq.h>
 #include <asm/mmu_context.h>
-#include <asm/smp.h>
 #include <asm/mipsmtregs.h>
 #include <asm/r4kcache.h>
 #include <asm/cacheflush.h>
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 7b4418dd5857..269c252d956f 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -29,6 +29,7 @@
 #include <asm/cpu.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/smp-ops.h>
 #include <asm/system.h>
 
 struct cpuinfo_mips cpu_data[NR_CPUS] __read_mostly;
@@ -575,9 +576,7 @@ void __init setup_arch(char **cmdline_p)
 	arch_mem_init(cmdline_p);
 
 	resource_init();
-#ifdef CONFIG_SMP
 	plat_smp_setup();
-#endif
 }
 
 static int __init fpu_disable(char *s)
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index 2ab0b7eeaa7e..89e6f6aa5166 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -215,72 +215,67 @@ static void __init smp_tc_init(unsigned int tc, unsigned int mvpconf0)
 	write_tc_c0_tchalt(TCHALT_H);
 }
 
-/*
- * Common setup before any secondaries are started
- * Make sure all CPU's are in a sensible state before we boot any of the
- * secondarys
- */
-void __init plat_smp_setup(void)
+static void vsmp_send_ipi_single(int cpu, unsigned int action)
 {
-	unsigned int mvpconf0, ntc, tc, ncpu = 0;
-	unsigned int nvpe;
+	int i;
+	unsigned long flags;
+	int vpflags;
 
-#ifdef CONFIG_MIPS_MT_FPAFF
-	/* If we have an FPU, enroll ourselves in the FPU-full mask */
-	if (cpu_has_fpu)
-		cpu_set(0, mt_fpu_cpumask);
-#endif /* CONFIG_MIPS_MT_FPAFF */
-	if (!cpu_has_mipsmt)
-		return;
+	local_irq_save(flags);
 
-	/* disable MT so we can configure */
-	dvpe();
-	dmt();
+	vpflags = dvpe();	/* cant access the other CPU's registers whilst MVPE enabled */
 
-	/* Put MVPE's into 'configuration state' */
-	set_c0_mvpcontrol(MVPCONTROL_VPC);
+	switch (action) {
+	case SMP_CALL_FUNCTION:
+		i = C_SW1;
+		break;
 
-	mvpconf0 = read_c0_mvpconf0();
-	ntc = (mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT;
+	case SMP_RESCHEDULE_YOURSELF:
+	default:
+		i = C_SW0;
+		break;
+	}
 
-	nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
-	smp_num_siblings = nvpe;
+	/* 1:1 mapping of vpe and tc... */
+	settc(cpu);
+	write_vpe_c0_cause(read_vpe_c0_cause() | i);
+	evpe(vpflags);
 
-	/* we'll always have more TC's than VPE's, so loop setting everything
-	   to a sensible state */
-	for (tc = 0; tc <= ntc; tc++) {
-		settc(tc);
+	local_irq_restore(flags);
+}
 
-		smp_tc_init(tc, mvpconf0);
-		ncpu = smp_vpe_init(tc, mvpconf0, ncpu);
-	}
+static void vsmp_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
 
-	/* Release config state */
-	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+	for_each_cpu_mask(i, mask)
+		vsmp_send_ipi_single(i, action);
+}
 
-	/* We'll wait until starting the secondaries before starting MVPE */
+static void __cpuinit vsmp_init_secondary(void)
+{
+	/* Enable per-cpu interrupts */
 
-	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", ncpu);
+	/* This is Malta specific: IPI,performance and timer inetrrupts */
+	write_c0_status((read_c0_status() & ~ST0_IM ) |
+	                (STATUSF_IP0 | STATUSF_IP1 | STATUSF_IP6 | STATUSF_IP7));
 }
 
-void __init plat_prepare_cpus(unsigned int max_cpus)
+static void __cpuinit vsmp_smp_finish(void)
 {
-	mips_mt_set_cpuoptions();
-
-	/* set up ipi interrupts */
-	if (cpu_has_vint) {
-		set_vi_handler(MIPS_CPU_IPI_RESCHED_IRQ, ipi_resched_dispatch);
-		set_vi_handler(MIPS_CPU_IPI_CALL_IRQ, ipi_call_dispatch);
-	}
+	write_c0_compare(read_c0_count() + (8* mips_hpt_frequency/HZ));
 
-	cpu_ipi_resched_irq = MIPS_CPU_IRQ_BASE + MIPS_CPU_IPI_RESCHED_IRQ;
-	cpu_ipi_call_irq = MIPS_CPU_IRQ_BASE + MIPS_CPU_IPI_CALL_IRQ;
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpu_set(smp_processor_id(), mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
 
-	setup_irq(cpu_ipi_resched_irq, &irq_resched);
-	setup_irq(cpu_ipi_call_irq, &irq_call);
+	local_irq_enable();
+}
 
-	set_irq_handler(cpu_ipi_resched_irq, handle_percpu_irq);
-	set_irq_handler(cpu_ipi_call_irq, handle_percpu_irq);
+static void vsmp_cpus_done(void)
+{
 }
 
 /*
@@ -291,7 +286,7 @@ void __init plat_prepare_cpus(unsigned int max_cpus)
  * (unsigned long)idle->thread_info the gp
  * assumes a 1:1 mapping of TC => VPE
  */
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
+static void __cpuinit vsmp_boot_secondary(int cpu, struct task_struct *idle)
 {
 	struct thread_info *gp = task_thread_info(idle);
 	dvpe();
@@ -325,57 +320,81 @@ void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
 	evpe(EVPE_ENABLE);
 }
 
-void __cpuinit prom_init_secondary(void)
-{
-	/* Enable per-cpu interrupts */
-
-	/* This is Malta specific: IPI,performance and timer inetrrupts */
-	write_c0_status((read_c0_status() & ~ST0_IM ) |
-	                (STATUSF_IP0 | STATUSF_IP1 | STATUSF_IP6 | STATUSF_IP7));
-}
-
-void __cpuinit prom_smp_finish(void)
+/*
+ * Common setup before any secondaries are started
+ * Make sure all CPU's are in a sensible state before we boot any of the
+ * secondarys
+ */
+static void __init vsmp_smp_setup(void)
 {
-	write_c0_compare(read_c0_count() + (8* mips_hpt_frequency/HZ));
+	unsigned int mvpconf0, ntc, tc, ncpu = 0;
+	unsigned int nvpe;
 
 #ifdef CONFIG_MIPS_MT_FPAFF
 	/* If we have an FPU, enroll ourselves in the FPU-full mask */
 	if (cpu_has_fpu)
-		cpu_set(smp_processor_id(), mt_fpu_cpumask);
+		cpu_set(0, mt_fpu_cpumask);
 #endif /* CONFIG_MIPS_MT_FPAFF */
+	if (!cpu_has_mipsmt)
+		return;
 
-	local_irq_enable();
-}
+	/* disable MT so we can configure */
+	dvpe();
+	dmt();
 
-void prom_cpus_done(void)
-{
-}
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
 
-void core_send_ipi(int cpu, unsigned int action)
-{
-	int i;
-	unsigned long flags;
-	int vpflags;
+	mvpconf0 = read_c0_mvpconf0();
+	ntc = (mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT;
 
-	local_irq_save(flags);
+	nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+	smp_num_siblings = nvpe;
 
-	vpflags = dvpe();	/* cant access the other CPU's registers whilst MVPE enabled */
+	/* we'll always have more TC's than VPE's, so loop setting everything
+	   to a sensible state */
+	for (tc = 0; tc <= ntc; tc++) {
+		settc(tc);
 
-	switch (action) {
-	case SMP_CALL_FUNCTION:
-		i = C_SW1;
-		break;
+		smp_tc_init(tc, mvpconf0);
+		ncpu = smp_vpe_init(tc, mvpconf0, ncpu);
+	}
 
-	case SMP_RESCHEDULE_YOURSELF:
-	default:
-		i = C_SW0;
-		break;
+	/* Release config state */
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	/* We'll wait until starting the secondaries before starting MVPE */
+
+	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", ncpu);
+}
+
+static void __init vsmp_prepare_cpus(unsigned int max_cpus)
+{
+	mips_mt_set_cpuoptions();
+
+	/* set up ipi interrupts */
+	if (cpu_has_vint) {
+		set_vi_handler(MIPS_CPU_IPI_RESCHED_IRQ, ipi_resched_dispatch);
+		set_vi_handler(MIPS_CPU_IPI_CALL_IRQ, ipi_call_dispatch);
 	}
 
-	/* 1:1 mapping of vpe and tc... */
-	settc(cpu);
-	write_vpe_c0_cause(read_vpe_c0_cause() | i);
-	evpe(vpflags);
+	cpu_ipi_resched_irq = MIPS_CPU_IRQ_BASE + MIPS_CPU_IPI_RESCHED_IRQ;
+	cpu_ipi_call_irq = MIPS_CPU_IRQ_BASE + MIPS_CPU_IPI_CALL_IRQ;
 
-	local_irq_restore(flags);
+	setup_irq(cpu_ipi_resched_irq, &irq_resched);
+	setup_irq(cpu_ipi_call_irq, &irq_call);
+
+	set_irq_handler(cpu_ipi_resched_irq, handle_percpu_irq);
+	set_irq_handler(cpu_ipi_call_irq, handle_percpu_irq);
 }
+
+struct plat_smp_ops vsmp_smp_ops = {
+	.send_ipi_single	= vsmp_send_ipi_single,
+	.send_ipi_mask		= vsmp_send_ipi_mask,
+	.init_secondary		= vsmp_init_secondary,
+	.smp_finish		= vsmp_smp_finish,
+	.cpus_done		= vsmp_cpus_done,
+	.boot_secondary		= vsmp_boot_secondary,
+	.smp_setup		= vsmp_smp_setup,
+	.prepare_cpus		= vsmp_prepare_cpus,
+};
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 335be9bcf0dc..1e5dfc28294a 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -37,7 +37,6 @@
 #include <asm/processor.h>
 #include <asm/system.h>
 #include <asm/mmu_context.h>
-#include <asm/smp.h>
 #include <asm/time.h>
 
 #ifdef CONFIG_MIPS_MT_SMTC
@@ -84,6 +83,16 @@ static inline void set_cpu_sibling_map(int cpu)
 		cpu_set(cpu, cpu_sibling_map[cpu]);
 }
 
+struct plat_smp_ops *mp_ops;
+
+__cpuinit void register_smp_ops(struct plat_smp_ops *ops)
+{
+	if (ops)
+		printk(KERN_WARNING "Overriding previous set SMP ops\n");
+
+	mp_ops = ops;
+}
+
 /*
  * First C code run on the secondary CPUs after being started up by
  * the master.
@@ -100,7 +109,7 @@ asmlinkage __cpuinit void start_secondary(void)
 	cpu_report();
 	per_cpu_trap_init();
 	mips_clockevent_init();
-	prom_init_secondary();
+	mp_ops->init_secondary();
 
 	/*
 	 * XXX parity protection should be folded in here when it's converted
@@ -112,7 +121,7 @@ asmlinkage __cpuinit void start_secondary(void)
 	cpu = smp_processor_id();
 	cpu_data[cpu].udelay_val = loops_per_jiffy;
 
-	prom_smp_finish();
+	mp_ops->smp_finish();
 	set_cpu_sibling_map(cpu);
 
 	cpu_set(cpu, cpu_callin_map);
@@ -184,7 +193,7 @@ int smp_call_function_mask(cpumask_t mask, void (*func) (void *info),
 	smp_mb();
 
 	/* Send a message to all other CPUs and wait for them to respond */
-	core_send_ipi_mask(mask, SMP_CALL_FUNCTION);
+	mp_ops->send_ipi_mask(mask, SMP_CALL_FUNCTION);
 
 	/* Wait for response */
 	/* FIXME: lock-up detection, backtrace on lock-up */
@@ -278,7 +287,7 @@ void smp_send_stop(void)
 
 void __init smp_cpus_done(unsigned int max_cpus)
 {
-	prom_cpus_done();
+	mp_ops->cpus_done();
 }
 
 /* called from main before smp_init() */
@@ -286,7 +295,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 	init_new_context(current, &init_mm);
 	current_thread_info()->cpu = 0;
-	plat_prepare_cpus(max_cpus);
+	mp_ops->prepare_cpus(max_cpus);
 	set_cpu_sibling_map(0);
 #ifndef CONFIG_HOTPLUG_CPU
 	cpu_present_map = cpu_possible_map;
@@ -325,7 +334,7 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	if (IS_ERR(idle))
 		panic(KERN_ERR "Fork failed for CPU %d", cpu);
 
-	prom_boot_secondary(cpu, idle);
+	mp_ops->boot_secondary(cpu, idle);
 
 	/*
 	 * Trust is futile.  We should really have timeouts ...
diff --git a/arch/mips/kernel/smtc-proc.c b/arch/mips/kernel/smtc-proc.c
index 6f3709996172..fe256559c997 100644
--- a/arch/mips/kernel/smtc-proc.c
+++ b/arch/mips/kernel/smtc-proc.c
@@ -14,7 +14,6 @@
 #include <asm/system.h>
 #include <asm/hardirq.h>
 #include <asm/mmu_context.h>
-#include <asm/smp.h>
 #include <asm/mipsregs.h>
 #include <asm/cacheflush.h>
 #include <linux/proc_fs.h>
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index 9c92d42996cb..85f700e58131 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -16,7 +16,6 @@
 #include <asm/hazards.h>
 #include <asm/irq.h>
 #include <asm/mmu_context.h>
-#include <asm/smp.h>
 #include <asm/mipsregs.h>
 #include <asm/cacheflush.h>
 #include <asm/time.h>
diff --git a/arch/mips/mips-boards/generic/init.c b/arch/mips/mips-boards/generic/init.c
index 30f1f54cb68b..1695dca5506b 100644
--- a/arch/mips/mips-boards/generic/init.c
+++ b/arch/mips/mips-boards/generic/init.c
@@ -250,6 +250,8 @@ void __init mips_ejtag_setup(void)
 	flush_icache_range((unsigned long)base, (unsigned long)base + 0x80);
 }
 
+extern struct plat_smp_ops msmtc_smp_ops;
+
 void __init prom_init(void)
 {
 	prom_argc = fw_arg0;
@@ -416,4 +418,10 @@ void __init prom_init(void)
 #ifdef CONFIG_SERIAL_8250_CONSOLE
 	console_config();
 #endif
+#ifdef CONFIG_MIPS_MT_SMP
+	register_smp_ops(&vsmp_smp_ops);
+#endif
+#ifdef CONFIG_MIPS_MT_SMTC
+	register_smp_ops(&msmtc_smp_ops);
+#endif
 }
diff --git a/arch/mips/mips-boards/malta/malta_smtc.c b/arch/mips/mips-boards/malta/malta_smtc.c
index 5c980f4a48fe..6f051ca243fa 100644
--- a/arch/mips/mips-boards/malta/malta_smtc.c
+++ b/arch/mips/mips-boards/malta/malta_smtc.c
@@ -15,26 +15,24 @@
  * Cause the specified action to be performed on a targeted "CPU"
  */
 
-void core_send_ipi(int cpu, unsigned int action)
+static void msmtc_send_ipi_single(int cpu, unsigned int action)
 {
 	/* "CPU" may be TC of same VPE, VPE of same CPU, or different CPU */
 	smtc_send_ipi(cpu, LINUX_SMP_IPI, action);
 }
 
-/*
- * Platform "CPU" startup hook
- */
-
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
+static void msmtc_send_ipi_mask(cpumask_t mask, unsigned int action)
 {
-	smtc_boot_secondary(cpu, idle);
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		msmtc_send_ipi_single(i, action);
 }
 
 /*
  * Post-config but pre-boot cleanup entry point
  */
-
-void __cpuinit prom_init_secondary(void)
+static void __cpuinit msmtc_init_secondary(void)
 {
         void smtc_init_secondary(void);
 	int myvpe;
@@ -50,45 +48,61 @@ void __cpuinit prom_init_secondary(void)
 			set_c0_status(0x100 << cp0_perfcount_irq);
 	}
 
-        smtc_init_secondary();
+	smtc_init_secondary();
 }
 
 /*
- * Platform SMP pre-initialization
- *
- * As noted above, we can assume a single CPU for now
- * but it may be multithreaded.
+ * Platform "CPU" startup hook
  */
-
-void __cpuinit plat_smp_setup(void)
+static void __cpuinit msmtc_boot_secondary(int cpu, struct task_struct *idle)
 {
-	if (read_c0_config3() & (1<<2))
-		mipsmt_build_cpu_map(0);
+	smtc_boot_secondary(cpu, idle);
 }
 
-void __init plat_prepare_cpus(unsigned int max_cpus)
+/*
+ * SMP initialization finalization entry point
+ */
+static void __cpuinit msmtc_smp_finish(void)
 {
-	if (read_c0_config3() & (1<<2))
-		mipsmt_prepare_cpus();
+	smtc_smp_finish();
 }
 
 /*
- * SMP initialization finalization entry point
+ * Hook for after all CPUs are online
  */
 
-void __cpuinit prom_smp_finish(void)
+static void msmtc_cpus_done(void)
 {
-	smtc_smp_finish();
 }
 
 /*
- * Hook for after all CPUs are online
+ * Platform SMP pre-initialization
+ *
+ * As noted above, we can assume a single CPU for now
+ * but it may be multithreaded.
  */
 
-void prom_cpus_done(void)
+static void __init msmtc_smp_setup(void)
 {
+	mipsmt_build_cpu_map(0);
 }
 
+static void __init msmtc_prepare_cpus(unsigned int max_cpus)
+{
+	mipsmt_prepare_cpus();
+}
+
+struct plat_smp_ops msmtc_smp_ops = {
+	.send_ipi_single	= msmtc_send_ipi_single,
+	.send_ipi_mask		= msmtc_send_ipi_mask,
+	.init_secondary		= msmtc_init_secondary,
+	.smp_finish		= msmtc_smp_finish,
+	.cpus_done		= msmtc_cpus_done,
+	.boot_secondary		= msmtc_boot_secondary,
+	.smp_setup		= msmtc_smp_setup,
+	.prepare_cpus		= msmtc_prepare_cpus,
+};
+
 #ifdef CONFIG_MIPS_MT_SMTC_IRQAFF
 /*
  * IRQ affinity hook
diff --git a/arch/mips/mipssim/Makefile b/arch/mips/mipssim/Makefile
index 75568b584df4..57f43c1c7882 100644
--- a/arch/mips/mipssim/Makefile
+++ b/arch/mips/mipssim/Makefile
@@ -21,6 +21,6 @@ obj-y := sim_platform.o sim_setup.o sim_mem.o sim_time.o sim_int.o \
 	 sim_cmdline.o
 
 obj-$(CONFIG_EARLY_PRINTK) += sim_console.o
-obj-$(CONFIG_SMP) += sim_smp.o
+obj-$(CONFIG_MIPS_MT_SMTC) += sim_smtc.o
 
 EXTRA_CFLAGS += -Werror
diff --git a/arch/mips/mipssim/sim_setup.c b/arch/mips/mipssim/sim_setup.c
index 452c129d02c1..d49fe73426b7 100644
--- a/arch/mips/mipssim/sim_setup.c
+++ b/arch/mips/mipssim/sim_setup.c
@@ -60,6 +60,8 @@ void __init plat_mem_setup(void)
 #endif
 }
 
+extern struct plat_smp_ops ssmtc_smp_ops;
+
 void __init prom_init(void)
 {
 	set_io_port_base(0xbfd00000);
@@ -67,8 +69,20 @@ void __init prom_init(void)
 	pr_info("\nLINUX started...\n");
 	prom_init_cmdline();
 	prom_meminit();
-}
 
+#ifdef CONFIG_MIPS_MT_SMP
+	if (cpu_has_mipsmt)
+		register_smp_ops(&vsmp_smp_ops);
+	else
+		register_smp_ops(&up_smp_ops);
+#endif
+#ifdef CONFIG_MIPS_MT_SMTC
+	if (cpu_has_mipsmt)
+		register_smp_ops(&ssmtc_smp_ops);
+	else
+		register_smp_ops(&up_smp_ops);
+#endif
+}
 
 static void __init serial_init(void)
 {
diff --git a/arch/mips/mipssim/sim_smp.c b/arch/mips/mipssim/sim_smp.c
deleted file mode 100644
index ccbbccac23ef..000000000000
--- a/arch/mips/mipssim/sim_smp.c
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * Copyright (C) 2005 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- */
-/*
- * Simulator Platform-specific hooks for SMP operation
- */
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/cpumask.h>
-#include <linux/interrupt.h>
-#include <linux/smp.h>
-
-#include <asm/atomic.h>
-#include <asm/cpu.h>
-#include <asm/processor.h>
-#include <asm/system.h>
-#include <asm/mmu_context.h>
-#ifdef CONFIG_MIPS_MT_SMTC
-#include <asm/smtc_ipi.h>
-#endif /* CONFIG_MIPS_MT_SMTC */
-
-/* VPE/SMP Prototype implements platform interfaces directly */
-#if !defined(CONFIG_MIPS_MT_SMP)
-
-/*
- * Cause the specified action to be performed on a targeted "CPU"
- */
-
-void core_send_ipi(int cpu, unsigned int action)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	smtc_send_ipi(cpu, LINUX_SMP_IPI, action);
-#endif /* CONFIG_MIPS_MT_SMTC */
-/* "CPU" may be TC of same VPE, VPE of same CPU, or different CPU */
-
-}
-
-/*
- * Platform "CPU" startup hook
- */
-
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	smtc_boot_secondary(cpu, idle);
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-
-/*
- * Post-config but pre-boot cleanup entry point
- */
-
-void __cpuinit prom_init_secondary(void)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	void smtc_init_secondary(void);
-
-	smtc_init_secondary();
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-
-void plat_smp_setup(void)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	if (read_c0_config3() & (1 << 2))
-		mipsmt_build_cpu_map(0);
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-
-/*
- * Platform SMP pre-initialization
- */
-
-void plat_prepare_cpus(unsigned int max_cpus)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	/*
-	 * As noted above, we can assume a single CPU for now
-	 * but it may be multithreaded.
-	 */
-
-	if (read_c0_config3() & (1 << 2)) {
-		mipsmt_prepare_cpus();
-	}
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-
-/*
- * SMP initialization finalization entry point
- */
-
-void __cpuinit prom_smp_finish(void)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-	smtc_smp_finish();
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-
-/*
- * Hook for after all CPUs are online
- */
-
-void prom_cpus_done(void)
-{
-#ifdef CONFIG_MIPS_MT_SMTC
-
-#endif /* CONFIG_MIPS_MT_SMTC */
-}
-#endif /* CONFIG_MIPS32R2_MT_SMP */
diff --git a/arch/mips/mipssim/sim_smtc.c b/arch/mips/mipssim/sim_smtc.c
new file mode 100644
index 000000000000..d6e4f656ad14
--- /dev/null
+++ b/arch/mips/mipssim/sim_smtc.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright (C) 2005 MIPS Technologies, Inc.  All rights reserved.
+ *
+ *  This program is free software; you can distribute it and/or modify it
+ *  under the terms of the GNU General Public License (Version 2) as
+ *  published by the Free Software Foundation.
+ *
+ *  This program is distributed in the hope it will be useful, but WITHOUT
+ *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ *  for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, write to the Free Software Foundation, Inc.,
+ *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
+ *
+ */
+/*
+ * Simulator Platform-specific hooks for SMTC operation
+ */
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/smp.h>
+
+#include <asm/atomic.h>
+#include <asm/cpu.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/mmu_context.h>
+#include <asm/smtc_ipi.h>
+
+/* VPE/SMP Prototype implements platform interfaces directly */
+
+/*
+ * Cause the specified action to be performed on a targeted "CPU"
+ */
+
+static void ssmtc_send_ipi_single(int cpu, unsigned int action)
+{
+	smtc_send_ipi(cpu, LINUX_SMP_IPI, action);
+	/* "CPU" may be TC of same VPE, VPE of same CPU, or different CPU */
+}
+
+static inline void ssmtc_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		ssmtc_send_ipi_single(i, action);
+}
+
+/*
+ * Post-config but pre-boot cleanup entry point
+ */
+static void __cpuinit ssmtc_init_secondary(void)
+{
+	void smtc_init_secondary(void);
+
+	smtc_init_secondary();
+}
+
+/*
+ * SMP initialization finalization entry point
+ */
+static void __cpuinit ssmtc_smp_finish(void)
+{
+	smtc_smp_finish();
+}
+
+/*
+ * Hook for after all CPUs are online
+ */
+static void ssmtc_cpus_done(void)
+{
+}
+
+/*
+ * Platform "CPU" startup hook
+ */
+static void __cpuinit ssmtc_boot_secondary(int cpu, struct task_struct *idle)
+{
+	smtc_boot_secondary(cpu, idle);
+}
+
+static void __init ssmtc_smp_setup(void)
+{
+	if (read_c0_config3() & (1 << 2))
+		mipsmt_build_cpu_map(0);
+}
+
+/*
+ * Platform SMP pre-initialization
+ */
+static void ssmtc_prepare_cpus(unsigned int max_cpus)
+{
+	/*
+	 * As noted above, we can assume a single CPU for now
+	 * but it may be multithreaded.
+	 */
+
+	if (read_c0_config3() & (1 << 2)) {
+		mipsmt_prepare_cpus();
+	}
+}
+
+struct plat_smp_ops ssmtc_smp_ops = {
+	.send_ipi_single	= ssmtc_send_ipi_single,
+	.send_ipi_mask		= ssmtc_send_ipi_mask,
+	.init_secondary		= ssmtc_init_secondary,
+	.smp_finish		= ssmtc_smp_finish,
+	.cpus_done		= ssmtc_cpus_done,
+	.boot_secondary		= ssmtc_boot_secondary,
+	.smp_setup		= ssmtc_smp_setup,
+	.prepare_cpus		= ssmtc_prepare_cpus,
+};
diff --git a/arch/mips/pmc-sierra/yosemite/prom.c b/arch/mips/pmc-sierra/yosemite/prom.c
index 96d3ff051d3d..35dc435846a6 100644
--- a/arch/mips/pmc-sierra/yosemite/prom.c
+++ b/arch/mips/pmc-sierra/yosemite/prom.c
@@ -19,6 +19,7 @@
 #include <asm/pgtable.h>
 #include <asm/processor.h>
 #include <asm/reboot.h>
+#include <asm/smp-ops.h>
 #include <asm/system.h>
 #include <asm/bootinfo.h>
 #include <asm/pmon.h>
@@ -78,6 +79,8 @@ static void prom_halt(void)
 		__asm__(".set\tmips3\n\t" "wait\n\t" ".set\tmips0");
 }
 
+extern struct plat_smp_ops yos_smp_ops;
+
 /*
  * Init routine which accepts the variables from PMON
  */
@@ -127,6 +130,8 @@ void __init prom_init(void)
 	}
 
 	prom_grab_secondary();
+
+	register_smp_ops(&yos_smp_ops);
 }
 
 void __init prom_free_prom_memory(void)
diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index b0f12cd2968a..653f3ec61cab 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -42,70 +42,6 @@ void __init prom_grab_secondary(void)
 	              launchstack + LAUNCHSTACK_SIZE, 0);
 }
 
-/*
- * Detect available CPUs, populate phys_cpu_present_map before smp_init
- *
- * We don't want to start the secondary CPU yet nor do we have a nice probing
- * feature in PMON so we just assume presence of the secondary core.
- */
-void __init plat_smp_setup(void)
-{
-	int i;
-
-	cpus_clear(phys_cpu_present_map);
-
-	for (i = 0; i < 2; i++) {
-		cpu_set(i, phys_cpu_present_map);
-		__cpu_number_map[i]	= i;
-		__cpu_logical_map[i]	= i;
-	}
-}
-
-void __init plat_prepare_cpus(unsigned int max_cpus)
-{
-	/*
-	 * Be paranoid.  Enable the IPI only if we're really about to go SMP.
-	 */
-	if (cpus_weight(cpu_possible_map))
-		set_c0_status(STATUSF_IP5);
-}
-
-/*
- * Firmware CPU startup hook
- * Complicated by PMON's weird interface which tries to minimic the UNIX fork.
- * It launches the next * available CPU and copies some information on the
- * stack so the first thing we do is throw away that stuff and load useful
- * values into the registers ...
- */
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
-{
-	unsigned long gp = (unsigned long) task_thread_info(idle);
-	unsigned long sp = __KSTK_TOS(idle);
-
-	secondary_sp = sp;
-	secondary_gp = gp;
-
-	spin_unlock(&launch_lock);
-}
-
-/* Hook for after all CPUs are online */
-void prom_cpus_done(void)
-{
-}
-
-/*
- *  After we've done initial boot, this function is called to allow the
- *  board code to clean up state, if needed
- */
-void __cpuinit prom_init_secondary(void)
-{
-	set_c0_status(ST0_CO | ST0_IE | ST0_IM);
-}
-
-void __cpuinit prom_smp_finish(void)
-{
-}
-
 void titan_mailbox_irq(void)
 {
 	int cpu = smp_processor_id();
@@ -133,7 +69,7 @@ void titan_mailbox_irq(void)
 /*
  * Send inter-processor interrupt
  */
-void core_send_ipi(int cpu, unsigned int action)
+static void yos_send_ipi_single(int cpu, unsigned int action)
 {
 	/*
 	 * Generate an INTMSG so that it can be sent over to the
@@ -159,3 +95,86 @@ void core_send_ipi(int cpu, unsigned int action)
 		break;
 	}
 }
+
+static void yos_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		yos_send_ipi_single(i, action);
+}
+
+/*
+ *  After we've done initial boot, this function is called to allow the
+ *  board code to clean up state, if needed
+ */
+static void __cpuinit yos_init_secondary(void)
+{
+	set_c0_status(ST0_CO | ST0_IE | ST0_IM);
+}
+
+static void __cpuinit yos_smp_finish(void)
+{
+}
+
+/* Hook for after all CPUs are online */
+static void yos_cpus_done(void)
+{
+}
+
+/*
+ * Firmware CPU startup hook
+ * Complicated by PMON's weird interface which tries to minimic the UNIX fork.
+ * It launches the next * available CPU and copies some information on the
+ * stack so the first thing we do is throw away that stuff and load useful
+ * values into the registers ...
+ */
+static void __cpuinit yos_boot_secondary(int cpu, struct task_struct *idle)
+{
+	unsigned long gp = (unsigned long) task_thread_info(idle);
+	unsigned long sp = __KSTK_TOS(idle);
+
+	secondary_sp = sp;
+	secondary_gp = gp;
+
+	spin_unlock(&launch_lock);
+}
+
+/*
+ * Detect available CPUs, populate phys_cpu_present_map before smp_init
+ *
+ * We don't want to start the secondary CPU yet nor do we have a nice probing
+ * feature in PMON so we just assume presence of the secondary core.
+ */
+static void __init yos_smp_setup(void)
+{
+	int i;
+
+	cpus_clear(phys_cpu_present_map);
+
+	for (i = 0; i < 2; i++) {
+		cpu_set(i, phys_cpu_present_map);
+		__cpu_number_map[i]	= i;
+		__cpu_logical_map[i]	= i;
+	}
+}
+
+static void __init yos_prepare_cpus(unsigned int max_cpus)
+{
+	/*
+	 * Be paranoid.  Enable the IPI only if we're really about to go SMP.
+	 */
+	if (cpus_weight(cpu_possible_map))
+		set_c0_status(STATUSF_IP5);
+}
+
+struct plat_smp_ops yos_smp_ops = {
+	.send_ipi_single	= yos_send_ipi_single,
+	.send_ipi_mask		= yos_send_ipi_mask,
+	.init_secondary		= yos_init_secondary,
+	.smp_finish		= yos_smp_finish,
+	.cpus_done		= yos_cpus_done,
+	.boot_secondary		= yos_boot_secondary,
+	.smp_setup		= yos_smp_setup,
+	.prepare_cpus		= yos_prepare_cpus,
+};
diff --git a/arch/mips/qemu/q-smp.c b/arch/mips/qemu/q-smp.c
index 4b0178d0df0b..ead6c30eeb14 100644
--- a/arch/mips/qemu/q-smp.c
+++ b/arch/mips/qemu/q-smp.c
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2006 by Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2006, 07 by Ralf Baechle (ralf@linux-mips.org)
  *
  * Symmetric Uniprocessor (TM) Support
  */
@@ -13,43 +13,55 @@
 /*
  * Send inter-processor interrupt
  */
-void core_send_ipi(int cpu, unsigned int action)
+void up_send_ipi_single(int cpu, unsigned int action)
 {
-	panic(KERN_ERR "%s called", __FUNCTION__);
+	panic(KERN_ERR "%s called", __func__);
+}
+
+static inline void up_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	panic(KERN_ERR "%s called", __func__);
 }
 
 /*
  *  After we've done initial boot, this function is called to allow the
  *  board code to clean up state, if needed
  */
-void __cpuinit prom_init_secondary(void)
+void __cpuinit up_init_secondary(void)
 {
 }
 
-void __cpuinit prom_smp_finish(void)
+void __cpuinit up_smp_finish(void)
 {
 }
 
 /* Hook for after all CPUs are online */
-void prom_cpus_done(void)
-{
-}
-
-void __init prom_prepare_cpus(unsigned int max_cpus)
+void up_cpus_done(void)
 {
-	cpus_clear(phys_cpu_present_map);
 }
 
 /*
  * Firmware CPU startup hook
  */
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
+void __cpuinit up_boot_secondary(int cpu, struct task_struct *idle)
 {
 }
 
-void __init plat_smp_setup(void)
+void __init up_smp_setup(void)
 {
 }
-void __init plat_prepare_cpus(unsigned int max_cpus)
+
+void __init up_prepare_cpus(unsigned int max_cpus)
 {
 }
+
+struct plat_smp_ops up_smp_ops = {
+	.send_ipi_single	= up_send_ipi_single,
+	.send_ipi_mask		= up_send_ipi_mask,
+	.init_secondary		= up_init_secondary,
+	.smp_finish		= up_smp_finish,
+	.cpus_done		= up_cpus_done,
+	.boot_secondary		= up_boot_secondary,
+	.smp_setup		= up_smp_setup,
+	.prepare_cpus		= up_prepare_cpus,
+};
diff --git a/arch/mips/sgi-ip27/ip27-init.c b/arch/mips/sgi-ip27/ip27-init.c
index 3305fa9ae66d..a49e7c85f724 100644
--- a/arch/mips/sgi-ip27/ip27-init.c
+++ b/arch/mips/sgi-ip27/ip27-init.c
@@ -27,7 +27,6 @@
 #include <asm/sn/hub.h>
 #include <asm/sn/intr.h>
 #include <asm/current.h>
-#include <asm/smp.h>
 #include <asm/processor.h>
 #include <asm/mmu_context.h>
 #include <asm/thread_info.h>
diff --git a/arch/mips/sgi-ip27/ip27-klnuma.c b/arch/mips/sgi-ip27/ip27-klnuma.c
index f10d9839006d..48932ce1d730 100644
--- a/arch/mips/sgi-ip27/ip27-klnuma.c
+++ b/arch/mips/sgi-ip27/ip27-klnuma.c
@@ -11,7 +11,6 @@
 
 #include <asm/page.h>
 #include <asm/sections.h>
-#include <asm/smp.h>
 #include <asm/sn/types.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/gda.h>
diff --git a/arch/mips/sgi-ip27/ip27-smp.c b/arch/mips/sgi-ip27/ip27-smp.c
index a70656d42191..f15fc93d6b35 100644
--- a/arch/mips/sgi-ip27/ip27-smp.c
+++ b/arch/mips/sgi-ip27/ip27-smp.c
@@ -140,30 +140,51 @@ static __init void intr_clear_all(nasid_t nasid)
 		REMOTE_HUB_CLR_INTR(nasid, i);
 }
 
-void __init plat_smp_setup(void)
+static void ip27_send_ipi_single(int destid, unsigned int action)
 {
-	cnodeid_t	cnode;
+	int irq;
 
-	for_each_online_node(cnode) {
-		if (cnode == 0)
-			continue;
-		intr_clear_all(COMPACT_TO_NASID_NODEID(cnode));
+	switch (action) {
+	case SMP_RESCHEDULE_YOURSELF:
+		irq = CPU_RESCHED_A_IRQ;
+		break;
+	case SMP_CALL_FUNCTION:
+		irq = CPU_CALL_A_IRQ;
+		break;
+	default:
+		panic("sendintr");
 	}
 
-	replicate_kernel_text();
+	irq += cputoslice(destid);
 
 	/*
-	 * Assumption to be fixed: we're always booted on logical / physical
-	 * processor 0.  While we're always running on logical processor 0
-	 * this still means this is physical processor zero; it might for
-	 * example be disabled in the firwware.
+	 * Convert the compact hub number to the NASID to get the correct
+	 * part of the address space.  Then set the interrupt bit associated
+	 * with the CPU we want to send the interrupt to.
 	 */
-	alloc_cpupda(0, 0);
+	REMOTE_HUB_SEND_INTR(COMPACT_TO_NASID_NODEID(cpu_to_node(destid)), irq);
 }
 
-void __init plat_prepare_cpus(unsigned int max_cpus)
+static void ip27_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		ip27_send_ipi_single(i, action);
+}
+
+static void __cpuinit ip27_init_secondary(void)
+{
+	per_cpu_init();
+	local_irq_enable();
+}
+
+static void __cpuinit ip27_smp_finish(void)
+{
+}
+
+static void __init ip27_cpus_done(void)
 {
-	/* We already did everything necessary earlier */
 }
 
 /*
@@ -171,7 +192,7 @@ void __init plat_prepare_cpus(unsigned int max_cpus)
  * set sp to the kernel stack of the newly created idle process, gp to the proc
  * struct so that current_thread_info() will work.
  */
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
+static void __cpuinit ip27_boot_secondary(int cpu, struct task_struct *idle)
 {
 	unsigned long gp = (unsigned long)task_thread_info(idle);
 	unsigned long sp = __KSTK_TOS(idle);
@@ -181,41 +202,39 @@ void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
 		0, (void *) sp, (void *) gp);
 }
 
-void __cpuinit prom_init_secondary(void)
+static void __init ip27_smp_setup(void)
 {
-	per_cpu_init();
-	local_irq_enable();
-}
-
-void __init prom_cpus_done(void)
-{
-}
-
-void __cpuinit prom_smp_finish(void)
-{
-}
-
-void core_send_ipi(int destid, unsigned int action)
-{
-	int irq;
+	cnodeid_t	cnode;
 
-	switch (action) {
-	case SMP_RESCHEDULE_YOURSELF:
-		irq = CPU_RESCHED_A_IRQ;
-		break;
-	case SMP_CALL_FUNCTION:
-		irq = CPU_CALL_A_IRQ;
-		break;
-	default:
-		panic("sendintr");
+	for_each_online_node(cnode) {
+		if (cnode == 0)
+			continue;
+		intr_clear_all(COMPACT_TO_NASID_NODEID(cnode));
 	}
 
-	irq += cputoslice(destid);
+	replicate_kernel_text();
 
 	/*
-	 * Convert the compact hub number to the NASID to get the correct
-	 * part of the address space.  Then set the interrupt bit associated
-	 * with the CPU we want to send the interrupt to.
+	 * Assumption to be fixed: we're always booted on logical / physical
+	 * processor 0.  While we're always running on logical processor 0
+	 * this still means this is physical processor zero; it might for
+	 * example be disabled in the firwware.
 	 */
-	REMOTE_HUB_SEND_INTR(COMPACT_TO_NASID_NODEID(cpu_to_node(destid)), irq);
+	alloc_cpupda(0, 0);
 }
+
+static void __init ip27_prepare_cpus(unsigned int max_cpus)
+{
+	/* We already did everything necessary earlier */
+}
+
+struct plat_smp_ops ip27_smp_ops = {
+	.send_ipi_single	= ip27_send_ipi_single,
+	.send_ipi_mask		= ip27_send_ipi_mask,
+	.init_secondary		= ip27_init_secondary,
+	.smp_finish		= ip27_smp_finish,
+	.cpus_done		= ip27_cpus_done,
+	.boot_secondary		= ip27_boot_secondary,
+	.smp_setup		= ip27_smp_setup,
+	.prepare_cpus		= ip27_prepare_cpus,
+};
diff --git a/arch/mips/sibyte/bcm1480/smp.c b/arch/mips/sibyte/bcm1480/smp.c
index 436ba78359ab..183c460b9ca1 100644
--- a/arch/mips/sibyte/bcm1480/smp.c
+++ b/arch/mips/sibyte/bcm1480/smp.c
@@ -23,6 +23,7 @@
 
 #include <asm/mmu_context.h>
 #include <asm/io.h>
+#include <asm/fw/cfe/cfe_api.h>
 #include <asm/sibyte/sb1250.h>
 #include <asm/sibyte/bcm1480_regs.h>
 #include <asm/sibyte/bcm1480_int.h>
@@ -67,28 +68,114 @@ void __cpuinit bcm1480_smp_init(void)
 	change_c0_status(ST0_IM, imask);
 }
 
-void __cpuinit bcm1480_smp_finish(void)
+/*
+ * These are routines for dealing with the sb1250 smp capabilities
+ * independent of board/firmware
+ */
+
+/*
+ * Simple enough; everything is set up, so just poke the appropriate mailbox
+ * register, and we should be set
+ */
+static void bcm1480_send_ipi_single(int cpu, unsigned int action)
+{
+	__raw_writeq((((u64)action)<< 48), mailbox_0_set_regs[cpu]);
+}
+
+static void bcm1480_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		bcm1480_send_ipi_single(i, action);
+}
+
+/*
+ * Code to run on secondary just after probing the CPU
+ */
+static void __cpuinit bcm1480_init_secondary(void)
+{
+	extern void bcm1480_smp_init(void);
+
+	bcm1480_smp_init();
+}
+
+/*
+ * Do any tidying up before marking online and running the idle
+ * loop
+ */
+static void __cpuinit bcm1480_smp_finish(void)
 {
 	extern void sb1480_clockevent_init(void);
 
 	sb1480_clockevent_init();
 	local_irq_enable();
+	bcm1480_smp_finish();
 }
 
 /*
- * These are routines for dealing with the sb1250 smp capabilities
- * independent of board/firmware
+ * Final cleanup after all secondaries booted
  */
+static void bcm1480_cpus_done(void)
+{
+}
 
 /*
- * Simple enough; everything is set up, so just poke the appropriate mailbox
- * register, and we should be set
+ * Setup the PC, SP, and GP of a secondary processor and start it
+ * running!
  */
-void core_send_ipi(int cpu, unsigned int action)
+static void __cpuinit bcm1480_boot_secondary(int cpu, struct task_struct *idle)
 {
-	__raw_writeq((((u64)action)<< 48), mailbox_0_set_regs[cpu]);
+	int retval;
+
+	retval = cfe_cpu_start(cpu_logical_map(cpu), &smp_bootstrap,
+			       __KSTK_TOS(idle),
+			       (unsigned long)task_thread_info(idle), 0);
+	if (retval != 0)
+		printk("cfe_start_cpu(%i) returned %i\n" , cpu, retval);
 }
 
+/*
+ * Use CFE to find out how many CPUs are available, setting up
+ * phys_cpu_present_map and the logical/physical mappings.
+ * XXXKW will the boot CPU ever not be physical 0?
+ *
+ * Common setup before any secondaries are started
+ */
+static void __init bcm1480_smp_setup(void)
+{
+	int i, num;
+
+	cpus_clear(phys_cpu_present_map);
+	cpu_set(0, phys_cpu_present_map);
+	__cpu_number_map[0] = 0;
+	__cpu_logical_map[0] = 0;
+
+	for (i = 1, num = 0; i < NR_CPUS; i++) {
+		if (cfe_cpu_stop(i) == 0) {
+			cpu_set(i, phys_cpu_present_map);
+			__cpu_number_map[i] = ++num;
+			__cpu_logical_map[num] = i;
+		}
+	}
+	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", num);
+}
+
+static void __init bcm1480_prepare_cpus(unsigned int max_cpus)
+{
+}
+
+struct plat_smp_ops bcm1480_smp_ops = {
+	.send_ipi_single	= bcm1480_send_ipi_single,
+	.send_ipi_mask		= bcm1480_send_ipi_mask,
+	.init_secondary		= bcm1480_init_secondary,
+	.smp_finish		= bcm1480_smp_finish,
+	.cpus_done		= bcm1480_cpus_done,
+	.boot_secondary		= bcm1480_boot_secondary,
+	.smp_setup		= bcm1480_smp_setup,
+	.prepare_cpus		= bcm1480_prepare_cpus,
+};
+
 void bcm1480_mailbox_interrupt(void)
 {
 	int cpu = smp_processor_id();
diff --git a/arch/mips/sibyte/cfe/Makefile b/arch/mips/sibyte/cfe/Makefile
index a1214937b705..02b32e142adf 100644
--- a/arch/mips/sibyte/cfe/Makefile
+++ b/arch/mips/sibyte/cfe/Makefile
@@ -1,3 +1,2 @@
 lib-y					= setup.o
-lib-$(CONFIG_SMP)			+= smp.o
 lib-$(CONFIG_SIBYTE_CFE_CONSOLE)	+= console.o
diff --git a/arch/mips/sibyte/cfe/setup.c b/arch/mips/sibyte/cfe/setup.c
index dbd6e6fdd3f9..50d7c05e15b8 100644
--- a/arch/mips/sibyte/cfe/setup.c
+++ b/arch/mips/sibyte/cfe/setup.c
@@ -28,6 +28,7 @@
 #include <asm/bootinfo.h>
 #include <asm/reboot.h>
 #include <asm/sibyte/board.h>
+#include <asm/smp-ops.h>
 
 #include <asm/fw/cfe/cfe_api.h>
 #include <asm/fw/cfe/cfe_error.h>
@@ -232,6 +233,9 @@ static int __init initrd_setup(char *str)
 
 #endif
 
+extern struct plat_smp_ops sb_smp_ops;
+extern struct plat_smp_ops bcm1480_smp_ops;
+
 /*
  * prom_init is called just after the cpu type is determined, from setup_arch()
  */
@@ -340,6 +344,13 @@ void __init prom_init(void)
 	arcs_cmdline[CL_SIZE-1] = 0;
 
 	prom_meminit();
+
+#if defined(CONFIG_SIBYTE_BCM112X) || defined(CONFIG_SIBYTE_SB1250)
+	register_smp_ops(&sb_smp_ops);
+#endif
+#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+	register_smp_ops(&bcm1480_smp_ops);
+#endif
 }
 
 void __init prom_free_prom_memory(void)
diff --git a/arch/mips/sibyte/cfe/smp.c b/arch/mips/sibyte/cfe/smp.c
deleted file mode 100644
index 534a62912f21..000000000000
--- a/arch/mips/sibyte/cfe/smp.c
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * Copyright (C) 2000, 2001, 2002, 2003 Broadcom Corporation
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- */
-
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/smp.h>
-#include <asm/processor.h>
-
-#include <asm/fw/cfe/cfe_api.h>
-#include <asm/fw/cfe/cfe_error.h>
-
-/*
- * Use CFE to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
- * XXXKW will the boot CPU ever not be physical 0?
- *
- * Common setup before any secondaries are started
- */
-void __init plat_smp_setup(void)
-{
-	int i, num;
-
-	cpus_clear(phys_cpu_present_map);
-	cpu_set(0, phys_cpu_present_map);
-	__cpu_number_map[0] = 0;
-	__cpu_logical_map[0] = 0;
-
-	for (i = 1, num = 0; i < NR_CPUS; i++) {
-		if (cfe_cpu_stop(i) == 0) {
-			cpu_set(i, phys_cpu_present_map);
-			__cpu_number_map[i] = ++num;
-			__cpu_logical_map[num] = i;
-		}
-	}
-	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", num);
-}
-
-void __init plat_prepare_cpus(unsigned int max_cpus)
-{
-}
-
-/*
- * Setup the PC, SP, and GP of a secondary processor and start it
- * running!
- */
-void __cpuinit prom_boot_secondary(int cpu, struct task_struct *idle)
-{
-	int retval;
-
-	retval = cfe_cpu_start(cpu_logical_map(cpu), &smp_bootstrap,
-			       __KSTK_TOS(idle),
-			       (unsigned long)task_thread_info(idle), 0);
-	if (retval != 0)
-		printk("cfe_start_cpu(%i) returned %i\n" , cpu, retval);
-}
-
-/*
- * Code to run on secondary just after probing the CPU
- */
-void __cpuinit prom_init_secondary(void)
-{
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
-	extern void bcm1480_smp_init(void);
-	bcm1480_smp_init();
-#elif defined(CONFIG_SIBYTE_SB1250)
-	extern void sb1250_smp_init(void);
-	sb1250_smp_init();
-#else
-#error invalid SMP configuration
-#endif
-}
-
-/*
- * Do any tidying up before marking online and running the idle
- * loop
- */
-void __cpuinit prom_smp_finish(void)
-{
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
-	extern void bcm1480_smp_finish(void);
-	bcm1480_smp_finish();
-#elif defined(CONFIG_SIBYTE_SB1250)
-	extern void sb1250_smp_finish(void);
-	sb1250_smp_finish();
-#else
-#error invalid SMP configuration
-#endif
-}
-
-/*
- * Final cleanup after all secondaries booted
- */
-void prom_cpus_done(void)
-{
-}
diff --git a/arch/mips/sibyte/sb1250/smp.c b/arch/mips/sibyte/sb1250/smp.c
index 3f52c95a4eb8..0734b933e969 100644
--- a/arch/mips/sibyte/sb1250/smp.c
+++ b/arch/mips/sibyte/sb1250/smp.c
@@ -24,6 +24,7 @@
 
 #include <asm/mmu_context.h>
 #include <asm/io.h>
+#include <asm/fw/cfe/cfe_api.h>
 #include <asm/sibyte/sb1250.h>
 #include <asm/sibyte/sb1250_regs.h>
 #include <asm/sibyte/sb1250_int.h>
@@ -55,7 +56,43 @@ void __cpuinit sb1250_smp_init(void)
 	change_c0_status(ST0_IM, imask);
 }
 
-void __cpuinit sb1250_smp_finish(void)
+/*
+ * These are routines for dealing with the sb1250 smp capabilities
+ * independent of board/firmware
+ */
+
+/*
+ * Simple enough; everything is set up, so just poke the appropriate mailbox
+ * register, and we should be set
+ */
+static void sb1250_send_ipi_single(int cpu, unsigned int action)
+{
+	__raw_writeq((((u64)action) << 48), mailbox_set_regs[cpu]);
+}
+
+static inline void sb1250_send_ipi_mask(cpumask_t mask, unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu_mask(i, mask)
+		sb1250_send_ipi_single(i, action);
+}
+
+/*
+ * Code to run on secondary just after probing the CPU
+ */
+static void __cpuinit sb1250_init_secondary(void)
+{
+	extern void sb1250_smp_init(void);
+
+	sb1250_smp_init();
+}
+
+/*
+ * Do any tidying up before marking online and running the idle
+ * loop
+ */
+static void __cpuinit sb1250_smp_finish(void)
 {
 	extern void sb1250_clockevent_init(void);
 
@@ -64,19 +101,68 @@ void __cpuinit sb1250_smp_finish(void)
 }
 
 /*
- * These are routines for dealing with the sb1250 smp capabilities
- * independent of board/firmware
+ * Final cleanup after all secondaries booted
  */
+static void sb1250_cpus_done(void)
+{
+}
 
 /*
- * Simple enough; everything is set up, so just poke the appropriate mailbox
- * register, and we should be set
+ * Setup the PC, SP, and GP of a secondary processor and start it
+ * running!
  */
-void core_send_ipi(int cpu, unsigned int action)
+static void __cpuinit sb1250_boot_secondary(int cpu, struct task_struct *idle)
 {
-	__raw_writeq((((u64)action) << 48), mailbox_set_regs[cpu]);
+	int retval;
+
+	retval = cfe_cpu_start(cpu_logical_map(cpu), &smp_bootstrap,
+			       __KSTK_TOS(idle),
+			       (unsigned long)task_thread_info(idle), 0);
+	if (retval != 0)
+		printk("cfe_start_cpu(%i) returned %i\n" , cpu, retval);
 }
 
+/*
+ * Use CFE to find out how many CPUs are available, setting up
+ * phys_cpu_present_map and the logical/physical mappings.
+ * XXXKW will the boot CPU ever not be physical 0?
+ *
+ * Common setup before any secondaries are started
+ */
+static void __init sb1250_smp_setup(void)
+{
+	int i, num;
+
+	cpus_clear(phys_cpu_present_map);
+	cpu_set(0, phys_cpu_present_map);
+	__cpu_number_map[0] = 0;
+	__cpu_logical_map[0] = 0;
+
+	for (i = 1, num = 0; i < NR_CPUS; i++) {
+		if (cfe_cpu_stop(i) == 0) {
+			cpu_set(i, phys_cpu_present_map);
+			__cpu_number_map[i] = ++num;
+			__cpu_logical_map[num] = i;
+		}
+	}
+	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", num);
+}
+
+static void __init sb1250_prepare_cpus(unsigned int max_cpus)
+{
+}
+
+struct plat_smp_ops sb_smp_ops = {
+	.send_ipi_single	= sb1250_send_ipi_single,
+	.send_ipi_mask		= sb1250_send_ipi_mask,
+	.init_secondary		= sb1250_init_secondary,
+	.smp_finish		= sb1250_smp_finish,
+	.cpus_done		= sb1250_cpus_done,
+	.boot_secondary		= sb1250_boot_secondary,
+	.smp_setup		= sb1250_smp_setup,
+	.prepare_cpus		= sb1250_prepare_cpus,
+};
+
 void sb1250_mailbox_interrupt(void)
 {
 	int cpu = smp_processor_id();
diff --git a/include/asm-mips/sibyte/sb1250.h b/include/asm-mips/sibyte/sb1250.h
index 0dad844a3b5b..80c1a052662a 100644
--- a/include/asm-mips/sibyte/sb1250.h
+++ b/include/asm-mips/sibyte/sb1250.h
@@ -48,12 +48,10 @@ extern unsigned int zbbus_mhz;
 extern void sb1250_time_init(void);
 extern void sb1250_mask_irq(int cpu, int irq);
 extern void sb1250_unmask_irq(int cpu, int irq);
-extern void sb1250_smp_finish(void);
 
 extern void bcm1480_time_init(void);
 extern void bcm1480_mask_irq(int cpu, int irq);
 extern void bcm1480_unmask_irq(int cpu, int irq);
-extern void bcm1480_smp_finish(void);
 
 #define AT_spin \
 	__asm__ __volatile__ (		\
diff --git a/include/asm-mips/smp-ops.h b/include/asm-mips/smp-ops.h
new file mode 100644
index 000000000000..b17fdfb5d818
--- /dev/null
+++ b/include/asm-mips/smp-ops.h
@@ -0,0 +1,56 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of this
+ * archive for more details.
+ *
+ * Copyright (C) 2000 - 2001 by Kanoj Sarcar (kanoj@sgi.com)
+ * Copyright (C) 2000 - 2001 by Silicon Graphics, Inc.
+ * Copyright (C) 2000, 2001, 2002 Ralf Baechle
+ * Copyright (C) 2000, 2001 Broadcom Corporation
+ */
+#ifndef __ASM_SMP_OPS_H
+#define __ASM_SMP_OPS_H
+
+#ifdef CONFIG_SMP
+
+#include <linux/cpumask.h>
+
+struct plat_smp_ops {
+	void (*send_ipi_single)(int cpu, unsigned int action);
+	void (*send_ipi_mask)(cpumask_t mask, unsigned int action);
+	void (*init_secondary)(void);
+	void (*smp_finish)(void);
+	void (*cpus_done)(void);
+	void (*boot_secondary)(int cpu, struct task_struct *idle);
+	void (*smp_setup)(void);
+	void (*prepare_cpus)(unsigned int max_cpus);
+};
+
+extern void register_smp_ops(struct plat_smp_ops *ops);
+
+static inline void plat_smp_setup(void)
+{
+	extern struct plat_smp_ops *mp_ops;	/* private */
+
+	mp_ops->smp_setup();
+}
+
+#else /* !CONFIG_SMP */
+
+struct plat_smp_ops;
+
+static inline void plat_smp_setup(void)
+{
+	/* UP, nothing to do ...  */
+}
+
+static inline void register_smp_ops(struct plat_smp_ops *ops)
+{
+}
+
+#endif /* !CONFIG_SMP */
+
+extern struct plat_smp_ops up_smp_ops;
+extern struct plat_smp_ops vsmp_smp_ops;
+
+#endif /* __ASM_SMP_OPS_H */
diff --git a/include/asm-mips/smp.h b/include/asm-mips/smp.h
index 23265879cee9..84fef1aeec0c 100644
--- a/include/asm-mips/smp.h
+++ b/include/asm-mips/smp.h
@@ -11,14 +11,13 @@
 #ifndef __ASM_SMP_H
 #define __ASM_SMP_H
 
-
-#ifdef CONFIG_SMP
-
 #include <linux/bitops.h>
 #include <linux/linkage.h>
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+
 #include <asm/atomic.h>
+#include <asm/smp-ops.h>
 
 extern int smp_num_siblings;
 extern cpumask_t cpu_sibling_map[];
@@ -52,56 +51,6 @@ extern struct call_data_struct *call_data;
 extern cpumask_t phys_cpu_present_map;
 #define cpu_possible_map	phys_cpu_present_map
 
-/*
- * These are defined by the board-specific code.
- */
-
-/*
- * Cause the function described by call_data to be executed on the passed
- * cpu.  When the function has finished, increment the finished field of
- * call_data.
- */
-extern void core_send_ipi(int cpu, unsigned int action);
-
-static inline void core_send_ipi_mask(cpumask_t mask, unsigned int action)
-{
-	unsigned int i;
-
-	for_each_cpu_mask(i, mask)
-		core_send_ipi(i, action);
-}
-
-
-/*
- * Firmware CPU startup hook
- */
-extern void prom_boot_secondary(int cpu, struct task_struct *idle);
-
-/*
- *  After we've done initial boot, this function is called to allow the
- *  board code to clean up state, if needed
- */
-extern void prom_init_secondary(void);
-
-/*
- * Populate cpu_possible_map before smp_init, called from setup_arch.
- */
-extern void plat_smp_setup(void);
-
-/*
- * Called in smp_prepare_cpus.
- */
-extern void plat_prepare_cpus(unsigned int max_cpus);
-
-/*
- * Last chance for the board code to finish SMP initialization before
- * the CPU is "online".
- */
-extern void prom_smp_finish(void);
-
-/* Hook for after all CPUs are online */
-extern void prom_cpus_done(void);
-
 extern void asmlinkage smp_bootstrap(void);
 
 /*
@@ -111,11 +60,11 @@ extern void asmlinkage smp_bootstrap(void);
  */
 static inline void smp_send_reschedule(int cpu)
 {
-	core_send_ipi(cpu, SMP_RESCHEDULE_YOURSELF);
+	extern struct plat_smp_ops *mp_ops;	/* private */
+
+	mp_ops->send_ipi_single(cpu, SMP_RESCHEDULE_YOURSELF);
 }
 
 extern asmlinkage void smp_call_function_interrupt(void);
 
-#endif /* CONFIG_SMP */
-
 #endif /* __ASM_SMP_H */
-- 
cgit v1.2.3


From e9c33572a92aa2ee565da5136745ea304ac23d9a Mon Sep 17 00:00:00 2001
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Date: Mon, 26 Nov 2007 23:40:01 +0100
Subject: [MIPS] Use real cache invalidate

R10k non coherent machines need a real dma cache invalidate to get rid of
speculative stores in cache.  For other machines this promises a slight
speedup.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/c-r4k.c        | 4 ++--
 include/asm-mips/r4kcache.h | 7 +++++++
 2 files changed, 9 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index c91b59680695..06074948450d 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -589,7 +589,7 @@ static void r4k_dma_cache_inv(unsigned long addr, unsigned long size)
 		if (size >= scache_size)
 			r4k_blast_scache();
 		else
-			blast_scache_range(addr, addr + size);
+			blast_inv_scache_range(addr, addr + size);
 		return;
 	}
 
@@ -597,7 +597,7 @@ static void r4k_dma_cache_inv(unsigned long addr, unsigned long size)
 		r4k_blast_dcache();
 	} else {
 		R4600_HIT_CACHEOP_WAR_IMPL;
-		blast_dcache_range(addr, addr + size);
+		blast_inv_dcache_range(addr, addr + size);
 	}
 
 	bc_inv(addr, size);
diff --git a/include/asm-mips/r4kcache.h b/include/asm-mips/r4kcache.h
index 2b8466ffd3ca..4c140db36786 100644
--- a/include/asm-mips/r4kcache.h
+++ b/include/asm-mips/r4kcache.h
@@ -403,6 +403,13 @@ __BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 64)
 __BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 64)
 __BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 128)
 
+__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 16)
+__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 32)
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 16)
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 32)
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 64)
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 128)
+
 /* build blast_xxx_range, protected_blast_xxx_range */
 #define __BUILD_BLAST_CACHE_RANGE(pfx, desc, hitop, prot) \
 static inline void prot##blast_##pfx##cache##_range(unsigned long start, \
-- 
cgit v1.2.3


From 68efdb81c6c3b5c1df7169a4b460907198b19103 Mon Sep 17 00:00:00 2001
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Date: Sun, 25 Nov 2007 11:27:06 +0100
Subject: [MIPS] IP22/IP28: fix extracting board/chip rev

Taken from Peter Fuersts IP28 patches

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/sgi/ioc.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/sgi/ioc.h b/include/asm-mips/sgi/ioc.h
index f3e3dc9bb732..343ed15f8dc4 100644
--- a/include/asm-mips/sgi/ioc.h
+++ b/include/asm-mips/sgi/ioc.h
@@ -138,8 +138,8 @@ struct sgioc_regs {
 	u8 _sysid[3];
 	volatile u8 sysid;
 #define SGIOC_SYSID_FULLHOUSE	0x01
-#define SGIOC_SYSID_BOARDREV(x)	((x & 0xe0) > 5)
-#define SGIOC_SYSID_CHIPREV(x)	((x & 0x1e) > 1)
+#define SGIOC_SYSID_BOARDREV(x)	(((x) & 0x1e) >> 1)
+#define SGIOC_SYSID_CHIPREV(x)	(((x) & 0xe0) >> 5)
 	u32 _unused2;
 	u8 _read[3];
 	volatile u8 read;
-- 
cgit v1.2.3


From 2064ba23e58daa929eec6f5e7a2abc24574a95b9 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 24 Nov 2007 01:20:27 +0900
Subject: [MIPS] TXx9 watchdog support for rbhma3100,rbhma4200,rbhma4500

This patch adds support for txx9wdt driver to rbhma3100, rbhma4200 and
rbhma4500 platform.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/jmr3927_defconfig                | 15 +++++-
 arch/mips/configs/rbhma4200_defconfig              | 15 +++++-
 arch/mips/configs/rbhma4500_defconfig              | 15 +++++-
 arch/mips/jmr3927/rbhma3100/setup.c                | 55 ++++++++++++++++++++++
 .../toshiba_rbtx4927/toshiba_rbtx4927_setup.c      | 55 ++++++++++++++++++++++
 arch/mips/tx4938/toshiba_rbtx4938/setup.c          | 25 ++++++++++
 include/asm-mips/tx4927/tx4927_pci.h               |  1 +
 7 files changed, 178 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/mips/configs/jmr3927_defconfig b/arch/mips/configs/jmr3927_defconfig
index eb96791c33ea..4ace37838993 100644
--- a/arch/mips/configs/jmr3927_defconfig
+++ b/arch/mips/configs/jmr3927_defconfig
@@ -464,7 +464,6 @@ CONFIG_SERIAL_TXX9_STDSERIAL=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_IPMI_HANDLER is not set
-# CONFIG_WATCHDOG is not set
 # CONFIG_HW_RANDOM is not set
 # CONFIG_RTC is not set
 # CONFIG_R3964 is not set
@@ -482,6 +481,20 @@ CONFIG_DEVPORT=y
 # CONFIG_W1 is not set
 # CONFIG_POWER_SUPPLY is not set
 # CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_TXX9_WDT=y
+
+#
+# PCI-based Watchdog Cards
+#
+# CONFIG_PCIPCWATCHDOG is not set
+# CONFIG_WDTPCI is not set
 
 #
 # Multifunction device drivers
diff --git a/arch/mips/configs/rbhma4200_defconfig b/arch/mips/configs/rbhma4200_defconfig
index 9383a598094b..a67c698a82d9 100644
--- a/arch/mips/configs/rbhma4200_defconfig
+++ b/arch/mips/configs/rbhma4200_defconfig
@@ -431,7 +431,6 @@ CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_IPMI_HANDLER is not set
-# CONFIG_WATCHDOG is not set
 # CONFIG_HW_RANDOM is not set
 # CONFIG_RTC is not set
 # CONFIG_R3964 is not set
@@ -449,6 +448,20 @@ CONFIG_DEVPORT=y
 # CONFIG_W1 is not set
 # CONFIG_POWER_SUPPLY is not set
 # CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_TXX9_WDT=m
+
+#
+# PCI-based Watchdog Cards
+#
+# CONFIG_PCIPCWATCHDOG is not set
+# CONFIG_WDTPCI is not set
 
 #
 # Multifunction device drivers
diff --git a/arch/mips/configs/rbhma4500_defconfig b/arch/mips/configs/rbhma4500_defconfig
index d1b56cc0fd7c..ebc8ad4c6439 100644
--- a/arch/mips/configs/rbhma4500_defconfig
+++ b/arch/mips/configs/rbhma4500_defconfig
@@ -450,7 +450,6 @@ CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
 # CONFIG_IPMI_HANDLER is not set
-# CONFIG_WATCHDOG is not set
 # CONFIG_HW_RANDOM is not set
 # CONFIG_RTC is not set
 # CONFIG_R3964 is not set
@@ -479,6 +478,20 @@ CONFIG_SPI_AT25=y
 # CONFIG_W1 is not set
 # CONFIG_POWER_SUPPLY is not set
 # CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_TXX9_WDT=m
+
+#
+# PCI-based Watchdog Cards
+#
+# CONFIG_PCIPCWATCHDOG is not set
+# CONFIG_WDTPCI is not set
 
 #
 # Multifunction device drivers
diff --git a/arch/mips/jmr3927/rbhma3100/setup.c b/arch/mips/jmr3927/rbhma3100/setup.c
index 75cfe6559005..c886d804d303 100644
--- a/arch/mips/jmr3927/rbhma3100/setup.c
+++ b/arch/mips/jmr3927/rbhma3100/setup.c
@@ -35,6 +35,7 @@
 #include <linux/delay.h>
 #include <linux/pm.h>
 #include <linux/platform_device.h>
+#include <linux/clk.h>
 #ifdef CONFIG_SERIAL_TXX9
 #include <linux/serial_core.h>
 #endif
@@ -233,6 +234,8 @@ static void __init tx3927_setup(void)
 	tx3927_ccfgptr->ccfg &= ~TX3927_CCFG_BEOW;
 	/* Disable PCI snoop */
 	tx3927_ccfgptr->ccfg &= ~TX3927_CCFG_PSNP;
+	/* do reset on watchdog */
+	tx3927_ccfgptr->ccfg |= TX3927_CCFG_WR;
 
 #ifdef DO_WRITE_THROUGH
 	/* Enable PCI SNOOP - with write through only */
@@ -383,3 +386,55 @@ static int __init jmr3927_rtc_init(void)
 	return IS_ERR(dev) ? PTR_ERR(dev) : 0;
 }
 device_initcall(jmr3927_rtc_init);
+
+/* Watchdog support */
+
+static int __init txx9_wdt_init(unsigned long base)
+{
+	struct resource res = {
+		.start	= base,
+		.end	= base + 0x100 - 1,
+		.flags	= IORESOURCE_MEM,
+	};
+	struct platform_device *dev =
+		platform_device_register_simple("txx9wdt", -1, &res, 1);
+	return IS_ERR(dev) ? PTR_ERR(dev) : 0;
+}
+
+static int __init jmr3927_wdt_init(void)
+{
+	return txx9_wdt_init(TX3927_TMR_REG(2));
+}
+device_initcall(jmr3927_wdt_init);
+
+/* Minimum CLK support */
+
+struct clk *clk_get(struct device *dev, const char *id)
+{
+	if (!strcmp(id, "imbus_clk"))
+		return (struct clk *)JMR3927_IMCLK;
+	return ERR_PTR(-ENOENT);
+}
+EXPORT_SYMBOL(clk_get);
+
+int clk_enable(struct clk *clk)
+{
+	return 0;
+}
+EXPORT_SYMBOL(clk_enable);
+
+void clk_disable(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_disable);
+
+unsigned long clk_get_rate(struct clk *clk)
+{
+	return (unsigned long)clk;
+}
+EXPORT_SYMBOL(clk_get_rate);
+
+void clk_put(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_put);
diff --git a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c
index c29a5283237b..e466e5e711d8 100644
--- a/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c
+++ b/arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c
@@ -50,6 +50,7 @@
 #include <linux/pci.h>
 #include <linux/pm.h>
 #include <linux/platform_device.h>
+#include <linux/clk.h>
 
 #include <asm/bootinfo.h>
 #include <asm/io.h>
@@ -803,6 +804,8 @@ void __init plat_mem_setup(void)
 		}
 
 	/* CCFG */
+	/* do reset on watchdog */
+	tx4927_ccfgptr->ccfg |= TX4927_CCFG_WR;
 	/* enable Timeout BusError */
 	if (tx4927_ccfg_toeon)
 		tx4927_ccfgptr->ccfg |= TX4927_CCFG_TOE;
@@ -944,3 +947,55 @@ static int __init rbtx4927_ne_init(void)
 	return IS_ERR(dev) ? PTR_ERR(dev) : 0;
 }
 device_initcall(rbtx4927_ne_init);
+
+/* Watchdog support */
+
+static int __init txx9_wdt_init(unsigned long base)
+{
+	struct resource res = {
+		.start	= base,
+		.end	= base + 0x100 - 1,
+		.flags	= IORESOURCE_MEM,
+	};
+	struct platform_device *dev =
+		platform_device_register_simple("txx9wdt", -1, &res, 1);
+	return IS_ERR(dev) ? PTR_ERR(dev) : 0;
+}
+
+static int __init rbtx4927_wdt_init(void)
+{
+	return txx9_wdt_init(TX4927_TMR_REG(2) & 0xfffffffffULL);
+}
+device_initcall(rbtx4927_wdt_init);
+
+/* Minimum CLK support */
+
+struct clk *clk_get(struct device *dev, const char *id)
+{
+	if (!strcmp(id, "imbus_clk"))
+		return (struct clk *)50000000;
+	return ERR_PTR(-ENOENT);
+}
+EXPORT_SYMBOL(clk_get);
+
+int clk_enable(struct clk *clk)
+{
+	return 0;
+}
+EXPORT_SYMBOL(clk_enable);
+
+void clk_disable(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_disable);
+
+unsigned long clk_get_rate(struct clk *clk)
+{
+	return (unsigned long)clk;
+}
+EXPORT_SYMBOL(clk_get_rate);
+
+void clk_put(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_put);
diff --git a/arch/mips/tx4938/toshiba_rbtx4938/setup.c b/arch/mips/tx4938/toshiba_rbtx4938/setup.c
index 2333d9ff879d..61249f049cd6 100644
--- a/arch/mips/tx4938/toshiba_rbtx4938/setup.c
+++ b/arch/mips/tx4938/toshiba_rbtx4938/setup.c
@@ -724,6 +724,8 @@ void __init tx4938_board_setup(void)
 	/* CCFG */
 	/* clear WatchDogReset,BusErrorOnWrite flag (W1C) */
 	tx4938_ccfgptr->ccfg |= TX4938_CCFG_WDRST | TX4938_CCFG_BEOW;
+	/* do reset on watchdog */
+	tx4938_ccfgptr->ccfg |= TX4938_CCFG_WR;
 	/* clear PCIC1 reset */
 	if (tx4938_ccfgptr->clkctr & TX4938_CLKCTR_PCIC1RST)
 		tx4938_ccfgptr->clkctr &= ~TX4938_CLKCTR_PCIC1RST;
@@ -1121,12 +1123,35 @@ static int __init rbtx4938_spi_init(void)
 }
 arch_initcall(rbtx4938_spi_init);
 
+/* Watchdog support */
+
+static int __init txx9_wdt_init(unsigned long base)
+{
+	struct resource res = {
+		.start	= base,
+		.end	= base + 0x100 - 1,
+		.flags	= IORESOURCE_MEM,
+		.parent	= &tx4938_reg_resource,
+	};
+	struct platform_device *dev =
+		platform_device_register_simple("txx9wdt", -1, &res, 1);
+	return IS_ERR(dev) ? PTR_ERR(dev) : 0;
+}
+
+static int __init rbtx4938_wdt_init(void)
+{
+	return txx9_wdt_init(TX4938_TMR_REG(2) & 0xfffffffffULL);
+}
+device_initcall(rbtx4938_wdt_init);
+
 /* Minimum CLK support */
 
 struct clk *clk_get(struct device *dev, const char *id)
 {
 	if (!strcmp(id, "spi-baseclk"))
 		return (struct clk *)(txx9_gbus_clock / 2 / 4);
+	if (!strcmp(id, "imbus_clk"))
+		return (struct clk *)(txx9_gbus_clock / 2);
 	return ERR_PTR(-ENOENT);
 }
 EXPORT_SYMBOL(clk_get);
diff --git a/include/asm-mips/tx4927/tx4927_pci.h b/include/asm-mips/tx4927/tx4927_pci.h
index 3f1e470192e3..0be77df70f2b 100644
--- a/include/asm-mips/tx4927/tx4927_pci.h
+++ b/include/asm-mips/tx4927/tx4927_pci.h
@@ -9,6 +9,7 @@
 #define __ASM_TX4927_TX4927_PCI_H
 
 #define TX4927_CCFG_TOE 0x00004000
+#define TX4927_CCFG_WR	0x00008000
 #define TX4927_CCFG_TINTDIS	0x01000000
 
 #define TX4927_PCIMEM      0x08000000
-- 
cgit v1.2.3


From 930bff882296c02ca81db108672ef4ca06c37db5 Mon Sep 17 00:00:00 2001
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Date: Sun, 25 Nov 2007 11:47:56 +0100
Subject: [MIPS] IP28: added cache barrier to assembly routines

IP28 needs special treatment to avoid speculative accesses. gcc
takes care for .c code, but for assembly code we need to do it
manually.

This is taken from Peter Fuersts IP28 patches.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/lib/memcpy.S       | 10 ++++++++++
 arch/mips/lib/memset.S       |  5 +++++
 arch/mips/lib/strncpy_user.S |  1 +
 include/asm-mips/asm.h       |  8 ++++++++
 4 files changed, 24 insertions(+)

(limited to 'include')

diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
index aded7b159052..01e450b1ebc9 100644
--- a/arch/mips/lib/memcpy.S
+++ b/arch/mips/lib/memcpy.S
@@ -199,6 +199,7 @@ FEXPORT(__copy_user)
 	 */
 #define rem t8
 
+	R10KCBARRIER(0(ra))
 	/*
 	 * The "issue break"s below are very approximate.
 	 * Issue delays for dcache fills will perturb the schedule, as will
@@ -231,6 +232,7 @@ both_aligned:
 	PREF(	1, 3*32(dst) )
 	.align	4
 1:
+	R10KCBARRIER(0(ra))
 EXC(	LOAD	t0, UNIT(0)(src),	l_exc)
 EXC(	LOAD	t1, UNIT(1)(src),	l_exc_copy)
 EXC(	LOAD	t2, UNIT(2)(src),	l_exc_copy)
@@ -272,6 +274,7 @@ EXC(	LOAD	t2, UNIT(2)(src),	l_exc_copy)
 EXC(	LOAD	t3, UNIT(3)(src),	l_exc_copy)
 	SUB	len, len, 4*NBYTES
 	ADD	src, src, 4*NBYTES
+	R10KCBARRIER(0(ra))
 EXC(	STORE	t0, UNIT(0)(dst),	s_exc_p4u)
 EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p3u)
 EXC(	STORE	t2, UNIT(2)(dst),	s_exc_p2u)
@@ -287,6 +290,7 @@ less_than_4units:
 	beq	rem, len, copy_bytes
 	 nop
 1:
+	R10KCBARRIER(0(ra))
 EXC(	LOAD	t0, 0(src),		l_exc)
 	ADD	src, src, NBYTES
 	SUB	len, len, NBYTES
@@ -334,6 +338,7 @@ EXC(	LDFIRST	t3, FIRST(0)(src),	l_exc)
 EXC(	LDREST	t3, REST(0)(src),	l_exc_copy)
 	SUB	t2, t2, t1	# t2 = number of bytes copied
 	xor	match, t0, t1
+	R10KCBARRIER(0(ra))
 EXC(	STFIRST t3, FIRST(0)(dst),	s_exc)
 	beq	len, t2, done
 	 SUB	len, len, t2
@@ -354,6 +359,7 @@ src_unaligned_dst_aligned:
  * It's OK to load FIRST(N+1) before REST(N) because the two addresses
  * are to the same unit (unless src is aligned, but it's not).
  */
+	R10KCBARRIER(0(ra))
 EXC(	LDFIRST	t0, FIRST(0)(src),	l_exc)
 EXC(	LDFIRST	t1, FIRST(1)(src),	l_exc_copy)
 	SUB     len, len, 4*NBYTES
@@ -384,6 +390,7 @@ cleanup_src_unaligned:
 	beq	rem, len, copy_bytes
 	 nop
 1:
+	R10KCBARRIER(0(ra))
 EXC(	LDFIRST t0, FIRST(0)(src),	l_exc)
 EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
 	ADD	src, src, NBYTES
@@ -399,6 +406,7 @@ copy_bytes_checklen:
 	 nop
 copy_bytes:
 	/* 0 < len < NBYTES  */
+	R10KCBARRIER(0(ra))
 #define COPY_BYTE(N)			\
 EXC(	lb	t0, N(src), l_exc);	\
 	SUB	len, len, 1;		\
@@ -528,6 +536,7 @@ LEAF(__rmemcpy)					/* a0=dst a1=src a2=len */
 	ADD	a1, a2				# src = src + len
 
 r_end_bytes:
+	R10KCBARRIER(0(ra))
 	lb	t0, -1(a1)
 	SUB	a2, a2, 0x1
 	sb	t0, -1(a0)
@@ -542,6 +551,7 @@ r_out:
 	 move	a2, zero
 
 r_end_bytes_up:
+	R10KCBARRIER(0(ra))
 	lb	t0, (a1)
 	SUB	a2, a2, 0x1
 	sb	t0, (a0)
diff --git a/arch/mips/lib/memset.S b/arch/mips/lib/memset.S
index 3bf38422342f..c018a4721693 100644
--- a/arch/mips/lib/memset.S
+++ b/arch/mips/lib/memset.S
@@ -86,6 +86,7 @@ FEXPORT(__bzero)
 	.set		at
 #endif
 
+	R10KCBARRIER(0(ra))
 #ifdef __MIPSEB__
 	EX(LONG_S_L, a1, (a0), first_fixup)	/* make word/dword aligned */
 #endif
@@ -103,11 +104,13 @@ FEXPORT(__bzero)
 	PTR_ADDU	t1, a0			/* end address */
 	.set		reorder
 1:	PTR_ADDIU	a0, 64
+	R10KCBARRIER(0(ra))
 	f_fill64 a0, -64, a1, fwd_fixup
 	bne		t1, a0, 1b
 	.set		noreorder
 
 memset_partial:
+	R10KCBARRIER(0(ra))
 	PTR_LA		t1, 2f			/* where to start */
 #if LONGSIZE == 4
 	PTR_SUBU	t1, t0
@@ -129,6 +132,7 @@ memset_partial:
 
 	beqz		a2, 1f
 	 PTR_ADDU	a0, a2			/* What's left */
+	R10KCBARRIER(0(ra))
 #ifdef __MIPSEB__
 	EX(LONG_S_R, a1, -1(a0), last_fixup)
 #endif
@@ -143,6 +147,7 @@ small_memset:
 	 PTR_ADDU	t1, a0, a2
 
 1:	PTR_ADDIU	a0, 1			/* fill bytewise */
+	R10KCBARRIER(0(ra))
 	bne		t1, a0, 1b
 	 sb		a1, -1(a0)
 
diff --git a/arch/mips/lib/strncpy_user.S b/arch/mips/lib/strncpy_user.S
index 5c8fb9d6b7f9..8a63f72b81d3 100644
--- a/arch/mips/lib/strncpy_user.S
+++ b/arch/mips/lib/strncpy_user.S
@@ -38,6 +38,7 @@ FEXPORT(__strncpy_from_user_nocheck_asm)
 	.set		noreorder
 1:	EX(lbu, t0, (v1), fault)
 	PTR_ADDIU	v1, 1
+	R10KCBARRIER(0(ra))
 	beqz		t0, 2f
 	 sb		t0, (a0)
 	PTR_ADDIU	v0, 1
diff --git a/include/asm-mips/asm.h b/include/asm-mips/asm.h
index 12e17581b823..608cfcfbb3ea 100644
--- a/include/asm-mips/asm.h
+++ b/include/asm-mips/asm.h
@@ -398,4 +398,12 @@ symbol		=	value
 
 #define SSNOP		sll zero, zero, 1
 
+#ifdef CONFIG_SGI_IP28
+/* Inhibit speculative stores to volatile (e.g.DMA) or invalid addresses. */
+#include <asm/cacheops.h>
+#define R10KCBARRIER(addr)  cache   Cache_Barrier, addr;
+#else
+#define R10KCBARRIER(addr)
+#endif
+
 #endif /* __ASM_ASM_H */
-- 
cgit v1.2.3


From 81149be11327cbad006f82318f46e0b68a7b14ad Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 29 Jan 2008 10:14:58 +0000
Subject: [MIPS] Remove CONFIG_SIBYTE_PT{1120,1125,SWARM}

According to Broadcom the PT systems are production test systems which
never reached customers so no need to keep the fragmentary support we
currently have.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig                          | 13 -------------
 arch/mips/configs/atlas_defconfig          |  1 -
 arch/mips/configs/bigsur_defconfig         |  1 -
 arch/mips/configs/capcella_defconfig       |  1 -
 arch/mips/configs/cobalt_defconfig         |  1 -
 arch/mips/configs/db1000_defconfig         |  1 -
 arch/mips/configs/db1100_defconfig         |  1 -
 arch/mips/configs/db1200_defconfig         |  1 -
 arch/mips/configs/db1500_defconfig         |  1 -
 arch/mips/configs/db1550_defconfig         |  1 -
 arch/mips/configs/decstation_defconfig     |  1 -
 arch/mips/configs/e55_defconfig            |  1 -
 arch/mips/configs/emma2rh_defconfig        |  1 -
 arch/mips/configs/excite_defconfig         |  1 -
 arch/mips/configs/fulong_defconfig         |  1 -
 arch/mips/configs/ip22_defconfig           |  1 -
 arch/mips/configs/ip27_defconfig           |  1 -
 arch/mips/configs/ip32_defconfig           |  1 -
 arch/mips/configs/jazz_defconfig           |  1 -
 arch/mips/configs/jmr3927_defconfig        |  1 -
 arch/mips/configs/lasat_defconfig          |  1 -
 arch/mips/configs/malta_defconfig          |  1 -
 arch/mips/configs/mipssim_defconfig        |  1 -
 arch/mips/configs/mpc30x_defconfig         |  1 -
 arch/mips/configs/msp71xx_defconfig        |  1 -
 arch/mips/configs/mtx1_defconfig           |  1 -
 arch/mips/configs/pb1100_defconfig         |  1 -
 arch/mips/configs/pb1500_defconfig         |  1 -
 arch/mips/configs/pb1550_defconfig         |  1 -
 arch/mips/configs/pnx8550-jbs_defconfig    |  1 -
 arch/mips/configs/pnx8550-stb810_defconfig |  1 -
 arch/mips/configs/qemu_defconfig           |  1 -
 arch/mips/configs/rbhma4200_defconfig      |  1 -
 arch/mips/configs/rbhma4500_defconfig      |  1 -
 arch/mips/configs/rm200_defconfig          |  1 -
 arch/mips/configs/sb1250-swarm_defconfig   |  1 -
 arch/mips/configs/sead_defconfig           |  1 -
 arch/mips/configs/tb0219_defconfig         |  1 -
 arch/mips/configs/tb0226_defconfig         |  1 -
 arch/mips/configs/tb0287_defconfig         |  1 -
 arch/mips/configs/workpad_defconfig        |  1 -
 arch/mips/configs/wrppmc_defconfig         |  1 -
 arch/mips/configs/yosemite_defconfig       |  1 -
 arch/mips/defconfig                        |  1 -
 arch/mips/sibyte/cfe/setup.c               |  3 ---
 include/asm-mips/sibyte/board.h            |  6 ++----
 include/asm-mips/sibyte/swarm.h            | 18 ------------------
 47 files changed, 2 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index b211e7961f28..00d6940d4ed7 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -546,19 +546,6 @@ config SIBYTE_SENTOSA
 	select SYS_SUPPORTS_BIG_ENDIAN
 	select SYS_SUPPORTS_LITTLE_ENDIAN
 
-config SIBYTE_PTSWARM
-	bool "Sibyte BCM91250PT-PTSWARM"
-	depends on EXPERIMENTAL
-	select BOOT_ELF32
-	select DMA_COHERENT
-	select NR_CPUS_DEFAULT_2
-	select SIBYTE_SB1250
-	select SWAP_IO_SPACE
-	select SYS_HAS_CPU_SB1
-	select SYS_SUPPORTS_BIG_ENDIAN
-	select SYS_SUPPORTS_HIGHMEM
-	select SYS_SUPPORTS_LITTLE_ENDIAN
-
 config SIBYTE_BIGSUR
 	bool "Sibyte BCM91480B-BigSur"
 	select BOOT_ELF32
diff --git a/arch/mips/configs/atlas_defconfig b/arch/mips/configs/atlas_defconfig
index 62bcc887f2ca..c28db00fe469 100644
--- a/arch/mips/configs/atlas_defconfig
+++ b/arch/mips/configs/atlas_defconfig
@@ -47,7 +47,6 @@ CONFIG_MIPS_ATLAS=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig
index 3c70c9d16d01..4bb97a08c9e6 100644
--- a/arch/mips/configs/bigsur_defconfig
+++ b/arch/mips/configs/bigsur_defconfig
@@ -47,7 +47,6 @@ CONFIG_SIBYTE_BIGSUR=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/capcella_defconfig b/arch/mips/configs/capcella_defconfig
index 8ecbbb226c76..128908a542db 100644
--- a/arch/mips/configs/capcella_defconfig
+++ b/arch/mips/configs/capcella_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/cobalt_defconfig b/arch/mips/configs/cobalt_defconfig
index 36c13039e237..cb56ebc760e9 100644
--- a/arch/mips/configs/cobalt_defconfig
+++ b/arch/mips/configs/cobalt_defconfig
@@ -35,7 +35,6 @@ CONFIG_MIPS_COBALT=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/db1000_defconfig b/arch/mips/configs/db1000_defconfig
index 5a8b7acb7dd7..184f119ff29d 100644
--- a/arch/mips/configs/db1000_defconfig
+++ b/arch/mips/configs/db1000_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_DB1000=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/db1100_defconfig b/arch/mips/configs/db1100_defconfig
index d4ed90bca269..e61dbee45560 100644
--- a/arch/mips/configs/db1100_defconfig
+++ b/arch/mips/configs/db1100_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_DB1100=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/db1200_defconfig b/arch/mips/configs/db1200_defconfig
index a055657e6983..508faf44ca72 100644
--- a/arch/mips/configs/db1200_defconfig
+++ b/arch/mips/configs/db1200_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_DB1200=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/db1500_defconfig b/arch/mips/configs/db1500_defconfig
index 0ad08cf446ec..5abadf34924d 100644
--- a/arch/mips/configs/db1500_defconfig
+++ b/arch/mips/configs/db1500_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_DB1500=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/db1550_defconfig b/arch/mips/configs/db1550_defconfig
index 057c7d429c80..9ccdcf0848a6 100644
--- a/arch/mips/configs/db1550_defconfig
+++ b/arch/mips/configs/db1550_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_DB1550=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/decstation_defconfig b/arch/mips/configs/decstation_defconfig
index 2fb350432669..b661cd61ea51 100644
--- a/arch/mips/configs/decstation_defconfig
+++ b/arch/mips/configs/decstation_defconfig
@@ -47,7 +47,6 @@ CONFIG_MACH_DECSTATION=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/e55_defconfig b/arch/mips/configs/e55_defconfig
index d0d07faeb844..93d19a575008 100644
--- a/arch/mips/configs/e55_defconfig
+++ b/arch/mips/configs/e55_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/emma2rh_defconfig b/arch/mips/configs/emma2rh_defconfig
index d73d965f7615..f931bcb73ae6 100644
--- a/arch/mips/configs/emma2rh_defconfig
+++ b/arch/mips/configs/emma2rh_defconfig
@@ -47,7 +47,6 @@ CONFIG_MARKEINS=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/excite_defconfig b/arch/mips/configs/excite_defconfig
index 17a866057fd4..dc3058645acc 100644
--- a/arch/mips/configs/excite_defconfig
+++ b/arch/mips/configs/excite_defconfig
@@ -48,7 +48,6 @@ CONFIG_BASLER_EXCITE=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/fulong_defconfig b/arch/mips/configs/fulong_defconfig
index 4ef39a0527cc..d89eeb006c67 100644
--- a/arch/mips/configs/fulong_defconfig
+++ b/arch/mips/configs/fulong_defconfig
@@ -33,7 +33,6 @@ CONFIG_LEMOTE_FULONG=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index 670039bb1a7c..a6957a66838d 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -36,7 +36,6 @@ CONFIG_SGI_IP22=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 892d4c38fd0d..8feb16a55b56 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -35,7 +35,6 @@ CONFIG_SGI_IP27=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/ip32_defconfig b/arch/mips/configs/ip32_defconfig
index 47f49b60c5d6..ad547c7c6f81 100644
--- a/arch/mips/configs/ip32_defconfig
+++ b/arch/mips/configs/ip32_defconfig
@@ -47,7 +47,6 @@ CONFIG_SGI_IP32=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/jazz_defconfig b/arch/mips/configs/jazz_defconfig
index fa655e247ecc..500daa9c3d82 100644
--- a/arch/mips/configs/jazz_defconfig
+++ b/arch/mips/configs/jazz_defconfig
@@ -47,7 +47,6 @@ CONFIG_MACH_JAZZ=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/jmr3927_defconfig b/arch/mips/configs/jmr3927_defconfig
index 4ace37838993..8dbfa19496cb 100644
--- a/arch/mips/configs/jmr3927_defconfig
+++ b/arch/mips/configs/jmr3927_defconfig
@@ -35,7 +35,6 @@ CONFIG_MIPS=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 CONFIG_TOSHIBA_JMR3927=y
diff --git a/arch/mips/configs/lasat_defconfig b/arch/mips/configs/lasat_defconfig
index 2c665fcef089..aa67a3baf53f 100644
--- a/arch/mips/configs/lasat_defconfig
+++ b/arch/mips/configs/lasat_defconfig
@@ -36,7 +36,6 @@ CONFIG_LASAT=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/malta_defconfig b/arch/mips/configs/malta_defconfig
index 4b7e43c9f69a..9e32b029501e 100644
--- a/arch/mips/configs/malta_defconfig
+++ b/arch/mips/configs/malta_defconfig
@@ -36,7 +36,6 @@ CONFIG_MIPS_MALTA=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/mipssim_defconfig b/arch/mips/configs/mipssim_defconfig
index 61b72f5a953e..5a7b4ed76906 100644
--- a/arch/mips/configs/mipssim_defconfig
+++ b/arch/mips/configs/mipssim_defconfig
@@ -37,7 +37,6 @@ CONFIG_MIPS_SIM=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/mpc30x_defconfig b/arch/mips/configs/mpc30x_defconfig
index 8334350d7229..c2f7ce9ff603 100644
--- a/arch/mips/configs/mpc30x_defconfig
+++ b/arch/mips/configs/mpc30x_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/msp71xx_defconfig b/arch/mips/configs/msp71xx_defconfig
index 69278999c9a2..af4a41e06e87 100644
--- a/arch/mips/configs/msp71xx_defconfig
+++ b/arch/mips/configs/msp71xx_defconfig
@@ -48,7 +48,6 @@ CONFIG_PMC_MSP=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/mtx1_defconfig b/arch/mips/configs/mtx1_defconfig
index b536d7c63790..be82f123324d 100644
--- a/arch/mips/configs/mtx1_defconfig
+++ b/arch/mips/configs/mtx1_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_ALCHEMY=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/pb1100_defconfig b/arch/mips/configs/pb1100_defconfig
index 703d28db05b9..2d394578b969 100644
--- a/arch/mips/configs/pb1100_defconfig
+++ b/arch/mips/configs/pb1100_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_PB1100=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/pb1500_defconfig b/arch/mips/configs/pb1500_defconfig
index 82f0c5cee0dc..0d645621ce79 100644
--- a/arch/mips/configs/pb1500_defconfig
+++ b/arch/mips/configs/pb1500_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_PB1500=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/pb1550_defconfig b/arch/mips/configs/pb1550_defconfig
index 147a4fc7fdd8..210a06543116 100644
--- a/arch/mips/configs/pb1550_defconfig
+++ b/arch/mips/configs/pb1550_defconfig
@@ -48,7 +48,6 @@ CONFIG_MIPS_PB1550=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/pnx8550-jbs_defconfig b/arch/mips/configs/pnx8550-jbs_defconfig
index f6906b069e04..bee11ebed8ec 100644
--- a/arch/mips/configs/pnx8550-jbs_defconfig
+++ b/arch/mips/configs/pnx8550-jbs_defconfig
@@ -47,7 +47,6 @@ CONFIG_PNX8550_JBS=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/pnx8550-stb810_defconfig b/arch/mips/configs/pnx8550-stb810_defconfig
index b741f81696fb..b088a5ce268f 100644
--- a/arch/mips/configs/pnx8550-stb810_defconfig
+++ b/arch/mips/configs/pnx8550-stb810_defconfig
@@ -47,7 +47,6 @@ CONFIG_PNX8550_STB810=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/qemu_defconfig b/arch/mips/configs/qemu_defconfig
index b3caf5125c15..ec1a6a796599 100644
--- a/arch/mips/configs/qemu_defconfig
+++ b/arch/mips/configs/qemu_defconfig
@@ -47,7 +47,6 @@ CONFIG_QEMU=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/rbhma4200_defconfig b/arch/mips/configs/rbhma4200_defconfig
index a67c698a82d9..16494d0942b8 100644
--- a/arch/mips/configs/rbhma4200_defconfig
+++ b/arch/mips/configs/rbhma4200_defconfig
@@ -35,7 +35,6 @@ CONFIG_MIPS=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/rbhma4500_defconfig b/arch/mips/configs/rbhma4500_defconfig
index ebc8ad4c6439..aa9e0cf66143 100644
--- a/arch/mips/configs/rbhma4500_defconfig
+++ b/arch/mips/configs/rbhma4500_defconfig
@@ -35,7 +35,6 @@ CONFIG_MIPS=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/rm200_defconfig b/arch/mips/configs/rm200_defconfig
index fc388118b114..7aaa6d6513f1 100644
--- a/arch/mips/configs/rm200_defconfig
+++ b/arch/mips/configs/rm200_defconfig
@@ -47,7 +47,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/sb1250-swarm_defconfig b/arch/mips/configs/sb1250-swarm_defconfig
index c2798229cbfb..4fe6a0b5c357 100644
--- a/arch/mips/configs/sb1250-swarm_defconfig
+++ b/arch/mips/configs/sb1250-swarm_defconfig
@@ -47,7 +47,6 @@ CONFIG_SIBYTE_SWARM=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/sead_defconfig b/arch/mips/configs/sead_defconfig
index 2b6282d132a8..aa5733779014 100644
--- a/arch/mips/configs/sead_defconfig
+++ b/arch/mips/configs/sead_defconfig
@@ -47,7 +47,6 @@ CONFIG_MIPS_SEAD=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/tb0219_defconfig b/arch/mips/configs/tb0219_defconfig
index 326aa7aa40ea..0ae553923f98 100644
--- a/arch/mips/configs/tb0219_defconfig
+++ b/arch/mips/configs/tb0219_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/tb0226_defconfig b/arch/mips/configs/tb0226_defconfig
index 9fd0faeacf53..e1f22023969d 100644
--- a/arch/mips/configs/tb0226_defconfig
+++ b/arch/mips/configs/tb0226_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/tb0287_defconfig b/arch/mips/configs/tb0287_defconfig
index 499b6bd7ee68..790e62bc0860 100644
--- a/arch/mips/configs/tb0287_defconfig
+++ b/arch/mips/configs/tb0287_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/workpad_defconfig b/arch/mips/configs/workpad_defconfig
index b52256ca0b53..1412c47517ff 100644
--- a/arch/mips/configs/workpad_defconfig
+++ b/arch/mips/configs/workpad_defconfig
@@ -35,7 +35,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/configs/wrppmc_defconfig b/arch/mips/configs/wrppmc_defconfig
index 7e410e10fed7..bbfeeeacc486 100644
--- a/arch/mips/configs/wrppmc_defconfig
+++ b/arch/mips/configs/wrppmc_defconfig
@@ -47,7 +47,6 @@ CONFIG_WR_PPMC=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/configs/yosemite_defconfig b/arch/mips/configs/yosemite_defconfig
index acaf0e21bb00..980dfb6417e9 100644
--- a/arch/mips/configs/yosemite_defconfig
+++ b/arch/mips/configs/yosemite_defconfig
@@ -47,7 +47,6 @@ CONFIG_PMC_YOSEMITE=y
 # CONFIG_SIBYTE_SENTOSA is not set
 # CONFIG_SIBYTE_RHONE is not set
 # CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_CRHINE is not set
 # CONFIG_SIBYTE_CRHONE is not set
diff --git a/arch/mips/defconfig b/arch/mips/defconfig
index 670039bb1a7c..a6957a66838d 100644
--- a/arch/mips/defconfig
+++ b/arch/mips/defconfig
@@ -36,7 +36,6 @@ CONFIG_SGI_IP22=y
 # CONFIG_SIBYTE_SWARM is not set
 # CONFIG_SIBYTE_LITTLESUR is not set
 # CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_PTSWARM is not set
 # CONFIG_SIBYTE_BIGSUR is not set
 # CONFIG_SNI_RM is not set
 # CONFIG_TOSHIBA_JMR3927 is not set
diff --git a/arch/mips/sibyte/cfe/setup.c b/arch/mips/sibyte/cfe/setup.c
index 50d7c05e15b8..33fce826f8bf 100644
--- a/arch/mips/sibyte/cfe/setup.c
+++ b/arch/mips/sibyte/cfe/setup.c
@@ -301,9 +301,6 @@ void __init prom_init(void)
 			 *  command line
 			 */
 			strcpy(arcs_cmdline, "root=/dev/ram0 ");
-#ifdef CONFIG_SIBYTE_PTSWARM
-			strcat(arcs_cmdline, "console=ttyS0,115200 ");
-#endif
 		} else {
 			/* The loader should have set the command line */
 			/* too early for panic to do any good */
diff --git a/include/asm-mips/sibyte/board.h b/include/asm-mips/sibyte/board.h
index da198a1c8c81..25372ae0e814 100644
--- a/include/asm-mips/sibyte/board.h
+++ b/include/asm-mips/sibyte/board.h
@@ -19,10 +19,8 @@
 #ifndef _SIBYTE_BOARD_H
 #define _SIBYTE_BOARD_H
 
-#if defined(CONFIG_SIBYTE_SWARM) || defined(CONFIG_SIBYTE_PTSWARM) || \
-    defined(CONFIG_SIBYTE_PT1120) || defined(CONFIG_SIBYTE_PT1125) || \
-    defined(CONFIG_SIBYTE_CRHONE) || defined(CONFIG_SIBYTE_CRHINE) || \
-    defined(CONFIG_SIBYTE_LITTLESUR)
+#if defined(CONFIG_SIBYTE_SWARM) || defined(CONFIG_SIBYTE_CRHONE) || \
+    defined(CONFIG_SIBYTE_CRHINE) || defined(CONFIG_SIBYTE_LITTLESUR)
 #include <asm/sibyte/swarm.h>
 #endif
 
diff --git a/include/asm-mips/sibyte/swarm.h b/include/asm-mips/sibyte/swarm.h
index 540865fa7ec3..114d9d29ca9d 100644
--- a/include/asm-mips/sibyte/swarm.h
+++ b/include/asm-mips/sibyte/swarm.h
@@ -26,24 +26,6 @@
 #define SIBYTE_HAVE_PCMCIA 1
 #define SIBYTE_HAVE_IDE    1
 #endif
-#ifdef CONFIG_SIBYTE_PTSWARM
-#define SIBYTE_BOARD_NAME "PTSWARM"
-#define SIBYTE_HAVE_PCMCIA 1
-#define SIBYTE_HAVE_IDE    1
-#define SIBYTE_DEFAULT_CONSOLE "ttyS0,115200"
-#endif
-#ifdef CONFIG_SIBYTE_PT1120
-#define SIBYTE_BOARD_NAME "PT1120"
-#define SIBYTE_HAVE_PCMCIA 1
-#define SIBYTE_HAVE_IDE    1
-#define SIBYTE_DEFAULT_CONSOLE "ttyS0,115200"
-#endif
-#ifdef CONFIG_SIBYTE_PT1125
-#define SIBYTE_BOARD_NAME "PT1125"
-#define SIBYTE_HAVE_PCMCIA 1
-#define SIBYTE_HAVE_IDE    1
-#define SIBYTE_DEFAULT_CONSOLE "ttyS0,115200"
-#endif
 #ifdef CONFIG_SIBYTE_LITTLESUR
 #define SIBYTE_BOARD_NAME "BCM91250C2 (LittleSur)"
 #define SIBYTE_HAVE_PCMCIA 0
-- 
cgit v1.2.3


From e2defae5a9b4f8d1acb058be212ef89c8763dc5b Mon Sep 17 00:00:00 2001
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Date: Sun, 2 Dec 2007 13:00:32 +0100
Subject: [MIPS] IP28 support

Add support for SGI IP28 machines (Indigo 2 with R10k CPUs)
This work is mainly based on Peter Fuersts work.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig                                  |  67 ++-
 arch/mips/Makefile                                 |  14 +
 arch/mips/sgi-ip22/Makefile                        |   8 +-
 arch/mips/sgi-ip22/ip22-mc.c                       |   4 +
 arch/mips/sgi-ip22/ip28-berr.c                     | 502 +++++++++++++++++++++
 include/asm-mips/dma.h                             |   7 +-
 include/asm-mips/mach-ip28/cpu-feature-overrides.h |  50 ++
 include/asm-mips/mach-ip28/ds1286.h                |   4 +
 include/asm-mips/mach-ip28/spaces.h                |  22 +
 include/asm-mips/mach-ip28/war.h                   |  25 +
 10 files changed, 693 insertions(+), 10 deletions(-)
 create mode 100644 arch/mips/sgi-ip22/ip28-berr.c
 create mode 100644 include/asm-mips/mach-ip28/cpu-feature-overrides.h
 create mode 100644 include/asm-mips/mach-ip28/ds1286.h
 create mode 100644 include/asm-mips/mach-ip28/spaces.h
 create mode 100644 include/asm-mips/mach-ip28/war.h

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 00d6940d4ed7..01740ef77824 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -127,6 +127,7 @@ config MACH_JAZZ
 	select ARCH_MAY_HAVE_PC_FDC
 	select CEVT_R4K
 	select CSRC_R4K
+	select DEFAULT_SGI_PARTITION if CPU_BIG_ENDIAN
 	select GENERIC_ISA_DMA
 	select IRQ_CPU
 	select I8253
@@ -401,6 +402,7 @@ config SGI_IP22
 	select BOOT_ELF32
 	select CEVT_R4K
 	select CSRC_R4K
+	select DEFAULT_SGI_PARTITION
 	select DMA_NONCOHERENT
 	select HW_HAS_EISA
 	select I8253
@@ -408,6 +410,12 @@ config SGI_IP22
 	select IP22_CPU_SCACHE
 	select IRQ_CPU
 	select GENERIC_ISA_DMA_SUPPORT_BROKEN
+	select SGI_HAS_DS1286
+	select SGI_HAS_I8042
+	select SGI_HAS_INDYDOG
+	select SGI_HAS_SEEQ
+	select SGI_HAS_WD93
+	select SGI_HAS_ZILOG
 	select SWAP_IO_SPACE
 	select SYS_HAS_CPU_R4X00
 	select SYS_HAS_CPU_R5000
@@ -425,6 +433,7 @@ config SGI_IP27
 	select ARC
 	select ARC64
 	select BOOT_ELF64
+	select DEFAULT_SGI_PARTITION
 	select DMA_IP27
 	select SYS_HAS_EARLY_PRINTK
 	select HW_HAS_PCI
@@ -441,6 +450,36 @@ config SGI_IP27
 	  workstations.  To compile a Linux kernel that runs on these, say Y
 	  here.
 
+config SGI_IP28
+	bool "SGI IP28 (Indigo2 R10k) (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	select ARC
+	select ARC64
+	select BOOT_ELF64
+	select CEVT_R4K
+	select CSRC_R4K
+	select DEFAULT_SGI_PARTITION
+	select DMA_NONCOHERENT
+	select GENERIC_ISA_DMA_SUPPORT_BROKEN
+	select IRQ_CPU
+	select HW_HAS_EISA
+	select I8253
+	select I8259
+	select SGI_HAS_DS1286
+	select SGI_HAS_I8042
+	select SGI_HAS_INDYDOG
+	select SGI_HAS_SEEQ
+	select SGI_HAS_WD93
+	select SGI_HAS_ZILOG
+	select SWAP_IO_SPACE
+	select SYS_HAS_CPU_R10000
+	select SYS_HAS_EARLY_PRINTK
+	select SYS_SUPPORTS_64BIT_KERNEL
+	select SYS_SUPPORTS_BIG_ENDIAN
+      help
+        This is the SGI Indigo2 with R10000 processor.  To compile a Linux
+        kernel that runs on these, say Y here.
+
 config SGI_IP32
 	bool "SGI IP32 (O2)"
 	select ARC
@@ -567,6 +606,7 @@ config SNI_RM
 	select BOOT_ELF32
 	select CEVT_R4K
 	select CSRC_R4K
+	select DEFAULT_SGI_PARTITION if CPU_BIG_ENDIAN
 	select DMA_NONCOHERENT
 	select GENERIC_ISA_DMA
 	select HW_HAS_EISA
@@ -939,6 +979,27 @@ config EMMA2RH
 config SERIAL_RM9000
 	bool
 
+config SGI_HAS_DS1286
+	bool
+
+config SGI_HAS_INDYDOG
+	bool
+
+config SGI_HAS_SEEQ
+	bool
+
+config SGI_HAS_WD93
+	bool
+
+config SGI_HAS_ZILOG
+	bool
+
+config SGI_HAS_I8042
+	bool
+
+config DEFAULT_SGI_PARTITION
+	bool
+
 config ARC32
 	bool
 
@@ -948,7 +1009,7 @@ config BOOT_ELF32
 config MIPS_L1_CACHE_SHIFT
 	int
 	default "4" if MACH_DECSTATION
-	default "7" if SGI_IP27 || SNI_RM
+	default "7" if SGI_IP27 || SGI_IP28 || SNI_RM
 	default "4" if PMC_MSP4200_EVAL
 	default "5"
 
@@ -957,7 +1018,7 @@ config HAVE_STD_PC_SERIAL_PORT
 
 config ARC_CONSOLE
 	bool "ARC console support"
-	depends on SGI_IP22 || (SNI_RM && CPU_LITTLE_ENDIAN)
+	depends on SGI_IP22 || SGI_IP28 || (SNI_RM && CPU_LITTLE_ENDIAN)
 
 config ARC_MEMORY
 	bool
@@ -966,7 +1027,7 @@ config ARC_MEMORY
 
 config ARC_PROMLIB
 	bool
-	depends on MACH_JAZZ || SNI_RM || SGI_IP22 || SGI_IP32
+	depends on MACH_JAZZ || SNI_RM || SGI_IP22 || SGI_IP28 || SGI_IP32
 	default y
 
 config ARC64
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index a49127af33c8..dd668f3a8394 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -478,6 +478,20 @@ OBJCOPYFLAGS			:= --change-addresses=0x57ffffff80000000
 endif
 endif
 
+#
+# SGI IP28 (Indigo2 R10k)
+#
+# Set the load address to >= 0xa800000020080000 if you want to leave space for
+# symmon, 0xa800000020004000 for production kernels ?  Note that the value must
+# be 16kb aligned or the handling of the current variable will break.
+# Simplified: what IP22 does at 128MB+ in ksegN, IP28 does at 512MB+ in xkphys
+#
+#core-$(CONFIG_SGI_IP28)		+= arch/mips/sgi-ip22/ arch/mips/arc/arc_con.o
+core-$(CONFIG_SGI_IP28)		+= arch/mips/sgi-ip22/
+cflags-$(CONFIG_SGI_IP28)	+= -mr10k-cache-barrier=1 -Iinclude/asm-mips/mach-ip28
+#cflags-$(CONFIG_SGI_IP28)	+= -Iinclude/asm-mips/mach-ip28
+load-$(CONFIG_SGI_IP28)		+= 0xa800000020004000
+
 #
 # SGI-IP32 (O2)
 #
diff --git a/arch/mips/sgi-ip22/Makefile b/arch/mips/sgi-ip22/Makefile
index e3acb51b70b5..ef1564e40c8d 100644
--- a/arch/mips/sgi-ip22/Makefile
+++ b/arch/mips/sgi-ip22/Makefile
@@ -3,9 +3,11 @@
 # under Linux.
 #
 
-obj-y	+= ip22-mc.o ip22-hpc.o ip22-int.o ip22-berr.o \
-	   ip22-time.o ip22-nvram.o ip22-platform.o ip22-reset.o ip22-setup.o
+obj-y	+= ip22-mc.o ip22-hpc.o ip22-int.o ip22-time.o ip22-nvram.o \
+	   ip22-platform.o ip22-reset.o ip22-setup.o
 
+obj-$(CONFIG_SGI_IP22) += ip22-berr.o
+obj-$(CONFIG_SGI_IP28) += ip28-berr.o
 obj-$(CONFIG_EISA)	+= ip22-eisa.o
 
-EXTRA_CFLAGS += -Werror
+# EXTRA_CFLAGS += -Werror
diff --git a/arch/mips/sgi-ip22/ip22-mc.c b/arch/mips/sgi-ip22/ip22-mc.c
index 01a805dcc67c..3f35d6367bec 100644
--- a/arch/mips/sgi-ip22/ip22-mc.c
+++ b/arch/mips/sgi-ip22/ip22-mc.c
@@ -4,6 +4,7 @@
  * Copyright (C) 1996 David S. Miller (dm@engr.sgi.com)
  * Copyright (C) 1999 Andrew R. Baker (andrewb@uab.edu) - Indigo2 changes
  * Copyright (C) 2003 Ladislav Michl  (ladis@linux-mips.org)
+ * Copyright (C) 2004 Peter Fuerst    (pf@net.alphadv.de) - IP28
  */
 
 #include <linux/init.h>
@@ -137,9 +138,12 @@ void __init sgimc_init(void)
 	/* Step 2: Enable all parity checking in cpu control register
 	 *         zero.
 	 */
+	/* don't touch parity settings for IP28 */
+#ifndef CONFIG_SGI_IP28
 	tmp = sgimc->cpuctrl0;
 	tmp |= (SGIMC_CCTRL0_EPERRGIO | SGIMC_CCTRL0_EPERRMEM |
 		SGIMC_CCTRL0_R4KNOCHKPARR);
+#endif
 	sgimc->cpuctrl0 = tmp;
 
 	/* Step 3: Setup the MC write buffer depth, this is controlled
diff --git a/arch/mips/sgi-ip22/ip28-berr.c b/arch/mips/sgi-ip22/ip28-berr.c
new file mode 100644
index 000000000000..30e12e2ec4b5
--- /dev/null
+++ b/arch/mips/sgi-ip22/ip28-berr.c
@@ -0,0 +1,502 @@
+/*
+ * ip28-berr.c: Bus error handling.
+ *
+ * Copyright (C) 2002, 2003 Ladislav Michl (ladis@linux-mips.org)
+ * Copyright (C) 2005 Peter Fuerst (pf@net.alphadv.de) - IP28
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+
+#include <asm/addrspace.h>
+#include <asm/system.h>
+#include <asm/traps.h>
+#include <asm/branch.h>
+#include <asm/irq_regs.h>
+#include <asm/sgi/mc.h>
+#include <asm/sgi/hpc3.h>
+#include <asm/sgi/ioc.h>
+#include <asm/sgi/ip22.h>
+#include <asm/r4kcache.h>
+#include <asm/uaccess.h>
+#include <asm/bootinfo.h>
+
+static unsigned int count_be_is_fixup;
+static unsigned int count_be_handler;
+static unsigned int count_be_interrupt;
+static int debug_be_interrupt;
+
+static unsigned int cpu_err_stat;	/* Status reg for CPU */
+static unsigned int gio_err_stat;	/* Status reg for GIO */
+static unsigned int cpu_err_addr;	/* Error address reg for CPU */
+static unsigned int gio_err_addr;	/* Error address reg for GIO */
+static unsigned int extio_stat;
+static unsigned int hpc3_berr_stat;	/* Bus error interrupt status */
+
+struct hpc3_stat {
+	unsigned long addr;
+	unsigned int ctrl;
+	unsigned int cbp;
+	unsigned int ndptr;
+};
+
+static struct {
+	struct hpc3_stat pbdma[8];
+	struct hpc3_stat scsi[2];
+	struct hpc3_stat ethrx, ethtx;
+} hpc3;
+
+static struct {
+	unsigned long err_addr;
+	struct {
+		u32 lo;
+		u32 hi;
+	} tags[1][2], tagd[4][2], tagi[4][2]; /* Way 0/1 */
+} cache_tags;
+
+static inline void save_cache_tags(unsigned busaddr)
+{
+	unsigned long addr = CAC_BASE | busaddr;
+	int i;
+	cache_tags.err_addr = addr;
+
+	/*
+	 * Starting with a bus-address, save secondary cache (indexed by
+	 * PA[23..18:7..6]) tags first.
+	 */
+	addr &= ~1L;
+#define tag cache_tags.tags[0]
+	cache_op(Index_Load_Tag_S, addr);
+	tag[0].lo = read_c0_taglo();	/* PA[35:18], VA[13:12] */
+	tag[0].hi = read_c0_taghi();	/* PA[39:36] */
+	cache_op(Index_Load_Tag_S, addr | 1L);
+	tag[1].lo = read_c0_taglo();	/* PA[35:18], VA[13:12] */
+	tag[1].hi = read_c0_taghi();	/* PA[39:36] */
+#undef tag
+
+	/*
+	 * Save all primary data cache (indexed by VA[13:5]) tags which
+	 * might fit to this bus-address, knowing that VA[11:0] == PA[11:0].
+	 * Saving all tags and evaluating them later is easier and safer
+	 * than relying on VA[13:12] from the secondary cache tags to pick
+	 * matching primary tags here already.
+	 */
+	addr &= (0xffL << 56) | ((1 << 12) - 1);
+#define tag cache_tags.tagd[i]
+	for (i = 0; i < 4; ++i, addr += (1 << 12)) {
+		cache_op(Index_Load_Tag_D, addr);
+		tag[0].lo = read_c0_taglo();	/* PA[35:12] */
+		tag[0].hi = read_c0_taghi();	/* PA[39:36] */
+		cache_op(Index_Load_Tag_D, addr | 1L);
+		tag[1].lo = read_c0_taglo();	/* PA[35:12] */
+		tag[1].hi = read_c0_taghi();	/* PA[39:36] */
+	}
+#undef tag
+
+	/*
+	 * Save primary instruction cache (indexed by VA[13:6]) tags
+	 * the same way.
+	 */
+	addr &= (0xffL << 56) | ((1 << 12) - 1);
+#define tag cache_tags.tagi[i]
+	for (i = 0; i < 4; ++i, addr += (1 << 12)) {
+		cache_op(Index_Load_Tag_I, addr);
+		tag[0].lo = read_c0_taglo();	/* PA[35:12] */
+		tag[0].hi = read_c0_taghi();	/* PA[39:36] */
+		cache_op(Index_Load_Tag_I, addr | 1L);
+		tag[1].lo = read_c0_taglo();	/* PA[35:12] */
+		tag[1].hi = read_c0_taghi();	/* PA[39:36] */
+	}
+#undef tag
+}
+
+#define GIO_ERRMASK	0xff00
+#define CPU_ERRMASK	0x3f00
+
+static void save_and_clear_buserr(void)
+{
+	int i;
+
+	/* save status registers */
+	cpu_err_addr = sgimc->cerr;
+	cpu_err_stat = sgimc->cstat;
+	gio_err_addr = sgimc->gerr;
+	gio_err_stat = sgimc->gstat;
+	extio_stat = sgioc->extio;
+	hpc3_berr_stat = hpc3c0->bestat;
+
+	hpc3.scsi[0].addr  = (unsigned long)&hpc3c0->scsi_chan0;
+	hpc3.scsi[0].ctrl  = hpc3c0->scsi_chan0.ctrl; /* HPC3_SCTRL_ACTIVE ? */
+	hpc3.scsi[0].cbp   = hpc3c0->scsi_chan0.cbptr;
+	hpc3.scsi[0].ndptr = hpc3c0->scsi_chan0.ndptr;
+
+	hpc3.scsi[1].addr  = (unsigned long)&hpc3c0->scsi_chan1;
+	hpc3.scsi[1].ctrl  = hpc3c0->scsi_chan1.ctrl; /* HPC3_SCTRL_ACTIVE ? */
+	hpc3.scsi[1].cbp   = hpc3c0->scsi_chan1.cbptr;
+	hpc3.scsi[1].ndptr = hpc3c0->scsi_chan1.ndptr;
+
+	hpc3.ethrx.addr  = (unsigned long)&hpc3c0->ethregs.rx_cbptr;
+	hpc3.ethrx.ctrl  = hpc3c0->ethregs.rx_ctrl; /* HPC3_ERXCTRL_ACTIVE ? */
+	hpc3.ethrx.cbp   = hpc3c0->ethregs.rx_cbptr;
+	hpc3.ethrx.ndptr = hpc3c0->ethregs.rx_ndptr;
+
+	hpc3.ethtx.addr  = (unsigned long)&hpc3c0->ethregs.tx_cbptr;
+	hpc3.ethtx.ctrl  = hpc3c0->ethregs.tx_ctrl; /* HPC3_ETXCTRL_ACTIVE ? */
+	hpc3.ethtx.cbp   = hpc3c0->ethregs.tx_cbptr;
+	hpc3.ethtx.ndptr = hpc3c0->ethregs.tx_ndptr;
+
+	for (i = 0; i < 8; ++i) {
+		/* HPC3_PDMACTRL_ISACT ? */
+		hpc3.pbdma[i].addr  = (unsigned long)&hpc3c0->pbdma[i];
+		hpc3.pbdma[i].ctrl  = hpc3c0->pbdma[i].pbdma_ctrl;
+		hpc3.pbdma[i].cbp   = hpc3c0->pbdma[i].pbdma_bptr;
+		hpc3.pbdma[i].ndptr = hpc3c0->pbdma[i].pbdma_dptr;
+	}
+	i = 0;
+	if (gio_err_stat & CPU_ERRMASK)
+		i = gio_err_addr;
+	if (cpu_err_stat & CPU_ERRMASK)
+		i = cpu_err_addr;
+	save_cache_tags(i);
+
+	sgimc->cstat = sgimc->gstat = 0;
+}
+
+static void print_cache_tags(void)
+{
+	u32 scb, scw;
+	int i;
+
+	printk(KERN_ERR "Cache tags @ %08x:\n", (unsigned)cache_tags.err_addr);
+
+	/* PA[31:12] shifted to PTag0 (PA[35:12]) format */
+	scw = (cache_tags.err_addr >> 4) & 0x0fffff00;
+
+	scb = cache_tags.err_addr & ((1 << 12) - 1) & ~((1 << 5) - 1);
+	for (i = 0; i < 4; ++i) { /* for each possible VA[13:12] value */
+		if ((cache_tags.tagd[i][0].lo & 0x0fffff00) != scw &&
+		    (cache_tags.tagd[i][1].lo & 0x0fffff00) != scw)
+		    continue;
+		printk(KERN_ERR
+		       "D: 0: %08x %08x, 1: %08x %08x  (VA[13:5]  %04x)\n",
+			cache_tags.tagd[i][0].hi, cache_tags.tagd[i][0].lo,
+			cache_tags.tagd[i][1].hi, cache_tags.tagd[i][1].lo,
+			scb | (1 << 12)*i);
+	}
+	scb = cache_tags.err_addr & ((1 << 12) - 1) & ~((1 << 6) - 1);
+	for (i = 0; i < 4; ++i) { /* for each possible VA[13:12] value */
+		if ((cache_tags.tagi[i][0].lo & 0x0fffff00) != scw &&
+		    (cache_tags.tagi[i][1].lo & 0x0fffff00) != scw)
+		    continue;
+		printk(KERN_ERR
+		       "I: 0: %08x %08x, 1: %08x %08x  (VA[13:6]  %04x)\n",
+			cache_tags.tagi[i][0].hi, cache_tags.tagi[i][0].lo,
+			cache_tags.tagi[i][1].hi, cache_tags.tagi[i][1].lo,
+			scb | (1 << 12)*i);
+	}
+	i = read_c0_config();
+	scb = i & (1 << 13) ? 7:6;      /* scblksize = 2^[7..6] */
+	scw = ((i >> 16) & 7) + 19 - 1; /* scwaysize = 2^[24..19] / 2 */
+
+	i = ((1 << scw) - 1) & ~((1 << scb) - 1);
+	printk(KERN_ERR "S: 0: %08x %08x, 1: %08x %08x  (PA[%u:%u] %05x)\n",
+		cache_tags.tags[0][0].hi, cache_tags.tags[0][0].lo,
+		cache_tags.tags[0][1].hi, cache_tags.tags[0][1].lo,
+		scw-1, scb, i & (unsigned)cache_tags.err_addr);
+}
+
+static inline const char *cause_excode_text(int cause)
+{
+	static const char *txt[32] =
+	{	"Interrupt",
+		"TLB modification",
+		"TLB (load or instruction fetch)",
+		"TLB (store)",
+		"Address error (load or instruction fetch)",
+		"Address error (store)",
+		"Bus error (instruction fetch)",
+		"Bus error (data: load or store)",
+		"Syscall",
+		"Breakpoint",
+		"Reserved instruction",
+		"Coprocessor unusable",
+		"Arithmetic Overflow",
+		"Trap",
+		"14",
+		"Floating-Point",
+		"16", "17", "18", "19", "20", "21", "22",
+		"Watch Hi/Lo",
+		"24", "25", "26", "27", "28", "29", "30", "31",
+	};
+	return txt[(cause & 0x7c) >> 2];
+}
+
+static void print_buserr(const struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	int error = 0;
+
+	if (extio_stat & EXTIO_MC_BUSERR) {
+		printk(KERN_ERR "MC Bus Error\n");
+		error |= 1;
+	}
+	if (extio_stat & EXTIO_HPC3_BUSERR) {
+		printk(KERN_ERR "HPC3 Bus Error 0x%x:<id=0x%x,%s,lane=0x%x>\n",
+			hpc3_berr_stat,
+			(hpc3_berr_stat & HPC3_BESTAT_PIDMASK) >>
+					  HPC3_BESTAT_PIDSHIFT,
+			(hpc3_berr_stat & HPC3_BESTAT_CTYPE) ? "PIO" : "DMA",
+			hpc3_berr_stat & HPC3_BESTAT_BLMASK);
+		error |= 2;
+	}
+	if (extio_stat & EXTIO_EISA_BUSERR) {
+		printk(KERN_ERR "EISA Bus Error\n");
+		error |= 4;
+	}
+	if (cpu_err_stat & CPU_ERRMASK) {
+		printk(KERN_ERR "CPU error 0x%x<%s%s%s%s%s%s> @ 0x%08x\n",
+			cpu_err_stat,
+			cpu_err_stat & SGIMC_CSTAT_RD ? "RD " : "",
+			cpu_err_stat & SGIMC_CSTAT_PAR ? "PAR " : "",
+			cpu_err_stat & SGIMC_CSTAT_ADDR ? "ADDR " : "",
+			cpu_err_stat & SGIMC_CSTAT_SYSAD_PAR ? "SYSAD " : "",
+			cpu_err_stat & SGIMC_CSTAT_SYSCMD_PAR ? "SYSCMD " : "",
+			cpu_err_stat & SGIMC_CSTAT_BAD_DATA ? "BAD_DATA " : "",
+			cpu_err_addr);
+		error |= 8;
+	}
+	if (gio_err_stat & GIO_ERRMASK) {
+		printk(KERN_ERR "GIO error 0x%x:<%s%s%s%s%s%s%s%s> @ 0x%08x\n",
+			gio_err_stat,
+			gio_err_stat & SGIMC_GSTAT_RD ? "RD " : "",
+			gio_err_stat & SGIMC_GSTAT_WR ? "WR " : "",
+			gio_err_stat & SGIMC_GSTAT_TIME ? "TIME " : "",
+			gio_err_stat & SGIMC_GSTAT_PROM ? "PROM " : "",
+			gio_err_stat & SGIMC_GSTAT_ADDR ? "ADDR " : "",
+			gio_err_stat & SGIMC_GSTAT_BC ? "BC " : "",
+			gio_err_stat & SGIMC_GSTAT_PIO_RD ? "PIO_RD " : "",
+			gio_err_stat & SGIMC_GSTAT_PIO_WR ? "PIO_WR " : "",
+			gio_err_addr);
+		error |= 16;
+	}
+	if (!error)
+		printk(KERN_ERR "MC: Hmm, didn't find any error condition.\n");
+	else {
+		printk(KERN_ERR "CP0: config %08x,  "
+			"MC: cpuctrl0/1: %08x/%05x, giopar: %04x\n"
+			"MC: cpu/gio_memacc: %08x/%05x, memcfg0/1: %08x/%08x\n",
+			read_c0_config(),
+			sgimc->cpuctrl0, sgimc->cpuctrl0, sgimc->giopar,
+			sgimc->cmacc, sgimc->gmacc,
+			sgimc->mconfig0, sgimc->mconfig1);
+		print_cache_tags();
+	}
+	printk(KERN_ALERT "%s, epc == %0*lx, ra == %0*lx\n",
+	       cause_excode_text(regs->cp0_cause),
+	       field, regs->cp0_epc, field, regs->regs[31]);
+}
+
+/*
+ * Check, whether MC's (virtual) DMA address caused the bus error.
+ * See "Virtual DMA Specification", Draft 1.5, Feb 13 1992, SGI
+ */
+
+static int addr_is_ram(unsigned long addr, unsigned sz)
+{
+	int i;
+
+	for (i = 0; i < boot_mem_map.nr_map; i++) {
+		unsigned long a = boot_mem_map.map[i].addr;
+		if (a <= addr && addr+sz <= a+boot_mem_map.map[i].size)
+			return 1;
+	}
+	return 0;
+}
+
+static int check_microtlb(u32 hi, u32 lo, unsigned long vaddr)
+{
+	/* This is likely rather similar to correct code ;-) */
+
+	vaddr &= 0x7fffffff; /* Doc. states that top bit is ignored */
+
+	/* If tlb-entry is valid and VPN-high (bits [30:21] ?) matches... */
+	if ((lo & 2) && (vaddr >> 21) == ((hi<<1) >> 22)) {
+		u32 ctl = sgimc->dma_ctrl;
+		if (ctl & 1) {
+			unsigned int pgsz = (ctl & 2) ? 14:12; /* 16k:4k */
+			/* PTEIndex is VPN-low (bits [22:14]/[20:12] ?) */
+			unsigned long pte = (lo >> 6) << 12; /* PTEBase */
+			pte += 8*((vaddr >> pgsz) & 0x1ff);
+			if (addr_is_ram(pte, 8)) {
+				/*
+				 * Note: Since DMA hardware does look up
+				 * translation on its own, this PTE *must*
+				 * match the TLB/EntryLo-register format !
+				 */
+				unsigned long a = *(unsigned long *)
+						PHYS_TO_XKSEG_UNCACHED(pte);
+				a = (a & 0x3f) << 6; /* PFN */
+				a += vaddr & ((1 << pgsz) - 1);
+				return (cpu_err_addr == a);
+			}
+		}
+	}
+	return 0;
+}
+
+static int check_vdma_memaddr(void)
+{
+	if (cpu_err_stat & CPU_ERRMASK) {
+		u32 a = sgimc->maddronly;
+
+		if (!(sgimc->dma_ctrl & 0x100)) /* Xlate-bit clear ? */
+			return (cpu_err_addr == a);
+
+		if (check_microtlb(sgimc->dtlb_hi0, sgimc->dtlb_lo0, a) ||
+		    check_microtlb(sgimc->dtlb_hi1, sgimc->dtlb_lo1, a) ||
+		    check_microtlb(sgimc->dtlb_hi2, sgimc->dtlb_lo2, a) ||
+		    check_microtlb(sgimc->dtlb_hi3, sgimc->dtlb_lo3, a))
+			return 1;
+	}
+	return 0;
+}
+
+static int check_vdma_gioaddr(void)
+{
+	if (gio_err_stat & GIO_ERRMASK) {
+		u32 a = sgimc->gio_dma_trans;
+		a = (sgimc->gmaddronly & ~a) | (sgimc->gio_dma_sbits & a);
+		return (gio_err_addr == a);
+	}
+	return 0;
+}
+
+/*
+ * MC sends an interrupt whenever bus or parity errors occur. In addition,
+ * if the error happened during a CPU read, it also asserts the bus error
+ * pin on the R4K. Code in bus error handler save the MC bus error registers
+ * and then clear the interrupt when this happens.
+ */
+
+static int ip28_be_interrupt(const struct pt_regs *regs)
+{
+	int i;
+
+	save_and_clear_buserr();
+	/*
+	 * Try to find out, whether we got here by a mispredicted speculative
+	 * load/store operation.  If so, it's not fatal, we can go on.
+	 */
+	/* Any cause other than "Interrupt" (ExcCode 0) is fatal. */
+	if (regs->cp0_cause & CAUSEF_EXCCODE)
+		goto mips_be_fatal;
+
+	/* Any cause other than "Bus error interrupt" (IP6) is weird. */
+	if ((regs->cp0_cause & CAUSEF_IP6) != CAUSEF_IP6)
+		goto mips_be_fatal;
+
+	if (extio_stat & (EXTIO_HPC3_BUSERR | EXTIO_EISA_BUSERR))
+		goto mips_be_fatal;
+
+	/* Any state other than "Memory bus error" is fatal. */
+	if (cpu_err_stat & CPU_ERRMASK & ~SGIMC_CSTAT_ADDR)
+		goto mips_be_fatal;
+
+	/* GIO errors other than timeouts are fatal */
+	if (gio_err_stat & GIO_ERRMASK & ~SGIMC_GSTAT_TIME)
+		goto mips_be_fatal;
+
+	/*
+	 * Now we have an asynchronous bus error, speculatively or DMA caused.
+	 * Need to search all DMA descriptors for the error address.
+	 */
+	for (i = 0; i < sizeof(hpc3)/sizeof(struct hpc3_stat); ++i) {
+		struct hpc3_stat *hp = (struct hpc3_stat *)&hpc3 + i;
+		if ((cpu_err_stat & CPU_ERRMASK) &&
+		    (cpu_err_addr == hp->ndptr || cpu_err_addr == hp->cbp))
+			break;
+		if ((gio_err_stat & GIO_ERRMASK) &&
+		    (gio_err_addr == hp->ndptr || gio_err_addr == hp->cbp))
+			break;
+	}
+	if (i < sizeof(hpc3)/sizeof(struct hpc3_stat)) {
+		struct hpc3_stat *hp = (struct hpc3_stat *)&hpc3 + i;
+		printk(KERN_ERR "at DMA addresses: HPC3 @ %08lx:"
+		       " ctl %08x, ndp %08x, cbp %08x\n",
+		       CPHYSADDR(hp->addr), hp->ctrl, hp->ndptr, hp->cbp);
+		goto mips_be_fatal;
+	}
+	/* Check MC's virtual DMA stuff. */
+	if (check_vdma_memaddr()) {
+		printk(KERN_ERR "at GIO DMA: mem address 0x%08x.\n",
+			sgimc->maddronly);
+		goto mips_be_fatal;
+	}
+	if (check_vdma_gioaddr()) {
+		printk(KERN_ERR "at GIO DMA: gio address 0x%08x.\n",
+			sgimc->gmaddronly);
+		goto mips_be_fatal;
+	}
+	/* A speculative bus error... */
+	if (debug_be_interrupt) {
+		print_buserr(regs);
+		printk(KERN_ERR "discarded!\n");
+	}
+	return MIPS_BE_DISCARD;
+
+mips_be_fatal:
+	print_buserr(regs);
+	return MIPS_BE_FATAL;
+}
+
+void ip22_be_interrupt(int irq)
+{
+	const struct pt_regs *regs = get_irq_regs();
+
+	count_be_interrupt++;
+
+	if (ip28_be_interrupt(regs) != MIPS_BE_DISCARD) {
+		/* Assume it would be too dangerous to continue ... */
+		die_if_kernel("Oops", regs);
+		force_sig(SIGBUS, current);
+	} else if (debug_be_interrupt)
+		show_regs((struct pt_regs *)regs);
+}
+
+static int ip28_be_handler(struct pt_regs *regs, int is_fixup)
+{
+	/*
+	 * We arrive here only in the unusual case of do_be() invocation,
+	 * i.e. by a bus error exception without a bus error interrupt.
+	 */
+	if (is_fixup) {
+		count_be_is_fixup++;
+		save_and_clear_buserr();
+		return MIPS_BE_FIXUP;
+	}
+	count_be_handler++;
+	return ip28_be_interrupt(regs);
+}
+
+void __init ip22_be_init(void)
+{
+	board_be_handler = ip28_be_handler;
+}
+
+int ip28_show_be_info(struct seq_file *m)
+{
+	seq_printf(m, "IP28 be fixups\t\t: %u\n", count_be_is_fixup);
+	seq_printf(m, "IP28 be interrupts\t: %u\n", count_be_interrupt);
+	seq_printf(m, "IP28 be handler\t\t: %u\n", count_be_handler);
+
+	return 0;
+}
+
+static int __init debug_be_setup(char *str)
+{
+	debug_be_interrupt++;
+	return 1;
+}
+__setup("ip28_debug_be", debug_be_setup);
diff --git a/include/asm-mips/dma.h b/include/asm-mips/dma.h
index d6a6c21f16db..1353c81065d1 100644
--- a/include/asm-mips/dma.h
+++ b/include/asm-mips/dma.h
@@ -84,10 +84,9 @@
  * Deskstations or Acer PICA but not the much more versatile DMA logic used
  * for the local devices on Acer PICA or Magnums.
  */
-#ifdef CONFIG_SGI_IP22
-/* Horrible hack to have a correct DMA window on IP22 */
-#include <asm/sgi/mc.h>
-#define MAX_DMA_ADDRESS		(PAGE_OFFSET + SGIMC_SEG0_BADDR + 0x01000000)
+#if defined(CONFIG_SGI_IP22) || defined(CONFIG_SGI_IP28)
+/* don't care; ISA bus master won't work, ISA slave DMA supports 32bit addr */
+#define MAX_DMA_ADDRESS		PAGE_OFFSET
 #else
 #define MAX_DMA_ADDRESS		(PAGE_OFFSET + 0x01000000)
 #endif
diff --git a/include/asm-mips/mach-ip28/cpu-feature-overrides.h b/include/asm-mips/mach-ip28/cpu-feature-overrides.h
new file mode 100644
index 000000000000..9a53b326f848
--- /dev/null
+++ b/include/asm-mips/mach-ip28/cpu-feature-overrides.h
@@ -0,0 +1,50 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003 Ralf Baechle
+ * 6/2004	pf
+ */
+#ifndef __ASM_MACH_IP28_CPU_FEATURE_OVERRIDES_H
+#define __ASM_MACH_IP28_CPU_FEATURE_OVERRIDES_H
+
+/*
+ * IP28 only comes with R10000 family processors all using the same config
+ */
+#define cpu_has_watch		1
+#define cpu_has_mips16		0
+#define cpu_has_divec		0
+#define cpu_has_vce		0
+#define cpu_has_cache_cdex_p	0
+#define cpu_has_cache_cdex_s	0
+#define cpu_has_prefetch	1
+#define cpu_has_mcheck		0
+#define cpu_has_ejtag		0
+
+#define cpu_has_llsc		1
+#define cpu_has_vtag_icache	0
+#define cpu_has_dc_aliases	0 /* see probe_pcache() */
+#define cpu_has_ic_fills_f_dc	0
+#define cpu_has_dsp		0
+#define cpu_icache_snoops_remote_store  1
+#define cpu_has_mipsmt		0
+#define cpu_has_userlocal	0
+
+#define cpu_has_nofpuex		0
+#define cpu_has_64bits		1
+
+#define cpu_has_4kex		1
+#define cpu_has_4k_cache	1
+
+#define cpu_has_inclusive_pcaches	1
+
+#define cpu_dcache_line_size()	32
+#define cpu_icache_line_size()	64
+
+#define cpu_has_mips32r1	0
+#define cpu_has_mips32r2	0
+#define cpu_has_mips64r1	0
+#define cpu_has_mips64r2	0
+
+#endif /* __ASM_MACH_IP28_CPU_FEATURE_OVERRIDES_H */
diff --git a/include/asm-mips/mach-ip28/ds1286.h b/include/asm-mips/mach-ip28/ds1286.h
new file mode 100644
index 000000000000..471bb9a33e0f
--- /dev/null
+++ b/include/asm-mips/mach-ip28/ds1286.h
@@ -0,0 +1,4 @@
+#ifndef __ASM_MACH_IP28_DS1286_H
+#define __ASM_MACH_IP28_DS1286_H
+#include <asm/mach-ip22/ds1286.h>
+#endif /* __ASM_MACH_IP28_DS1286_H */
diff --git a/include/asm-mips/mach-ip28/spaces.h b/include/asm-mips/mach-ip28/spaces.h
new file mode 100644
index 000000000000..05aabb27e5e7
--- /dev/null
+++ b/include/asm-mips/mach-ip28/spaces.h
@@ -0,0 +1,22 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 1999, 2000, 03, 04 Ralf Baechle
+ * Copyright (C) 2000, 2002  Maciej W. Rozycki
+ * Copyright (C) 1990, 1999, 2000 Silicon Graphics, Inc.
+ * 2004	pf
+ */
+#ifndef _ASM_MACH_IP28_SPACES_H
+#define _ASM_MACH_IP28_SPACES_H
+
+#define CAC_BASE		0xa800000000000000
+
+#define HIGHMEM_START		(~0UL)
+
+#define PHYS_OFFSET		_AC(0x20000000, UL)
+
+#include <asm/mach-generic/spaces.h>
+
+#endif /* _ASM_MACH_IP28_SPACES_H */
diff --git a/include/asm-mips/mach-ip28/war.h b/include/asm-mips/mach-ip28/war.h
new file mode 100644
index 000000000000..a1baafab486a
--- /dev/null
+++ b/include/asm-mips/mach-ip28/war.h
@@ -0,0 +1,25 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002, 2004, 2007 by Ralf Baechle <ralf@linux-mips.org>
+ */
+#ifndef __ASM_MIPS_MACH_IP28_WAR_H
+#define __ASM_MIPS_MACH_IP28_WAR_H
+
+#define R4600_V1_INDEX_ICACHEOP_WAR	0
+#define R4600_V1_HIT_CACHEOP_WAR	0
+#define R4600_V2_HIT_CACHEOP_WAR	0
+#define R5432_CP0_INTERRUPT_WAR		0
+#define BCM1250_M3_WAR			0
+#define SIBYTE_1956_WAR			0
+#define MIPS4K_ICACHE_REFILL_WAR	0
+#define MIPS_CACHE_SYNC_WAR		0
+#define TX49XX_ICACHE_INDEX_INV_WAR	0
+#define RM9000_CDEX_SMP_WAR		0
+#define ICACHE_REFILLS_WORKAROUND_WAR	0
+#define R10000_LLSC_WAR			1
+#define MIPS34K_MISSED_ITLB_WAR		0
+
+#endif /* __ASM_MIPS_MACH_IP28_WAR_H */
-- 
cgit v1.2.3


From 237cfee1db66147aef4457f02b56a41e6f84bfd3 Mon Sep 17 00:00:00 2001
From: Manuel Lauss <mano@roarinelk.homelinux.net>
Date: Thu, 6 Dec 2007 09:07:55 +0100
Subject: [MIPS] Alchemy: Au1210/Au1250 CPU support

This patch adds IDs for new Au1200 variants: Au1210 and Au1250.
They are essentially identical to the Au1200 except for the Au1210
which has a different SoC-ID in the PRId register [bits 31:24].
The Au1250 is a "Au1200 V0.2".

Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/cpu-probe.c | 9 +++++++++
 arch/mips/mm/c-r4k.c         | 2 ++
 arch/mips/mm/tlbex.c         | 2 ++
 include/asm-mips/cpu.h       | 4 ++--
 4 files changed, 15 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index 5c2794391bf5..5861a432a52f 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -188,6 +188,8 @@ static inline void check_wait(void)
 	case CPU_AU1500:
 	case CPU_AU1550:
 	case CPU_AU1200:
+	case CPU_AU1210:
+	case CPU_AU1250:
 		if (allow_au1k_wait)
 			cpu_wait = au1k_wait;
 		break;
@@ -733,6 +735,11 @@ static inline void cpu_probe_alchemy(struct cpuinfo_mips *c)
 			break;
 		case 4:
 			c->cputype = CPU_AU1200;
+			if (2 == (c->processor_id & 0xff))
+				c->cputype = CPU_AU1250;
+			break;
+		case 5:
+			c->cputype = CPU_AU1210;
 			break;
 		default:
 			panic("Unknown Au Core!");
@@ -858,6 +865,8 @@ static __init const char *cpu_to_name(struct cpuinfo_mips *c)
 	case CPU_AU1100:	name = "Au1100"; break;
 	case CPU_AU1550:	name = "Au1550"; break;
 	case CPU_AU1200:	name = "Au1200"; break;
+	case CPU_AU1210:	name = "Au1210"; break;
+	case CPU_AU1250:	name = "Au1250"; break;
 	case CPU_4KEC:		name = "MIPS 4KEc"; break;
 	case CPU_4KSC:		name = "MIPS 4KSc"; break;
 	case CPU_VR41XX:	name = "NEC Vr41xx"; break;
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 06074948450d..02bd180f0e02 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -980,6 +980,8 @@ static void __init probe_pcache(void)
 	case CPU_AU1100:
 	case CPU_AU1550:
 	case CPU_AU1200:
+	case CPU_AU1210:
+	case CPU_AU1250:
 		c->icache.flags |= MIPS_CACHE_IC_F_DC;
 		break;
 	}
diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c
index c298344fcb71..d026302e0ecc 100644
--- a/arch/mips/mm/tlbex.c
+++ b/arch/mips/mm/tlbex.c
@@ -917,6 +917,8 @@ static void __init build_tlb_write_entry(u32 **p, struct label **l,
 	case CPU_AU1500:
 	case CPU_AU1550:
 	case CPU_AU1200:
+	case CPU_AU1210:
+	case CPU_AU1250:
 	case CPU_PR4450:
 		i_nop(p);
 		tlbw(p);
diff --git a/include/asm-mips/cpu.h b/include/asm-mips/cpu.h
index 54fc18a4e5a8..bf5bbc78a9f7 100644
--- a/include/asm-mips/cpu.h
+++ b/include/asm-mips/cpu.h
@@ -195,8 +195,8 @@ enum cpu_type_enum {
 	 * MIPS32 class processors
 	 */
 	CPU_4KC, CPU_4KEC, CPU_4KSC, CPU_24K, CPU_34K, CPU_74K, CPU_AU1000,
-	CPU_AU1100, CPU_AU1200, CPU_AU1500, CPU_AU1550, CPU_PR4450,
-	CPU_BCM3302, CPU_BCM4710,
+	CPU_AU1100, CPU_AU1200, CPU_AU1210, CPU_AU1250, CPU_AU1500, CPU_AU1550,
+	CPU_PR4450, CPU_BCM3302, CPU_BCM4710,
 
 	/*
 	 * MIPS64 class processors
-- 
cgit v1.2.3


From 231a35d37293ab88d325a9cb94e5474c156282c0 Mon Sep 17 00:00:00 2001
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Date: Fri, 4 Jan 2008 23:31:07 +0100
Subject: [MIPS] RM: Collected changes

- EISA support for non PCI RMs (RM200 and RM400-xxx). The major part
  is the splitting of the EISA and onboard ISA of the RM200, which
  makes the EISA bus on the RM200 look like on other RMs.
- 64bit kernel support
- system type detection is now common for big and little endian
- moved sniprom code to arch/mips/fw
- added call_o32 function to arch/mips/fw/lib, which uses a private
  stack for calling prom functions
- fix problem with ISA interrupts, which makes using PIT clockevent
  possible

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig           |   4 +
 arch/mips/Makefile          |   2 +
 arch/mips/fw/lib/Makefile   |   5 +
 arch/mips/fw/lib/call_o32.S |  97 +++++++++++++
 arch/mips/fw/sni/Makefile   |   5 +
 arch/mips/fw/sni/sniprom.c  | 151 ++++++++++++++++++++
 arch/mips/sni/Makefile      |   2 +-
 arch/mips/sni/a20r.c        |  13 +-
 arch/mips/sni/eisa.c        |  50 +++++++
 arch/mips/sni/irq.c         |   4 +-
 arch/mips/sni/rm200.c       | 326 ++++++++++++++++++++++++++++++++++++++++++--
 arch/mips/sni/setup.c       | 143 ++++++++++++++++++-
 arch/mips/sni/sniprom.c     | 251 ----------------------------------
 arch/mips/sni/time.c        |   1 +
 include/asm-mips/bootinfo.h |   1 +
 include/asm-mips/mipsprom.h |   2 +
 include/asm-mips/sni.h      | 159 ++++++++++++---------
 17 files changed, 878 insertions(+), 338 deletions(-)
 create mode 100644 arch/mips/fw/lib/Makefile
 create mode 100644 arch/mips/fw/lib/call_o32.S
 create mode 100644 arch/mips/fw/sni/Makefile
 create mode 100644 arch/mips/fw/sni/sniprom.c
 create mode 100644 arch/mips/sni/eisa.c
 delete mode 100644 arch/mips/sni/sniprom.c

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 01740ef77824..d2d9cd8e4912 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -602,6 +602,7 @@ config SNI_RM
 	bool "SNI RM200/300/400"
 	select ARC if CPU_LITTLE_ENDIAN
 	select ARC32 if CPU_LITTLE_ENDIAN
+	select SNIPROM if CPU_BIG_ENDIAN
 	select ARCH_MAY_HAVE_PC_FDC
 	select BOOT_ELF32
 	select CEVT_R4K
@@ -1003,6 +1004,9 @@ config DEFAULT_SGI_PARTITION
 config ARC32
 	bool
 
+config SNIPROM
+	bool
+
 config BOOT_ELF32
 	bool
 
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index dd668f3a8394..2df68cd0ed40 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -156,6 +156,8 @@ endif
 #
 libs-$(CONFIG_ARC)		+= arch/mips/fw/arc/
 libs-$(CONFIG_CFE)		+= arch/mips/fw/cfe/
+libs-$(CONFIG_SNIPROM)		+= arch/mips/fw/sni/
+libs-y				+= arch/mips/fw/lib/
 libs-$(CONFIG_SIBYTE_CFE)	+= arch/mips/sibyte/cfe/
 
 #
diff --git a/arch/mips/fw/lib/Makefile b/arch/mips/fw/lib/Makefile
new file mode 100644
index 000000000000..84befc968fc4
--- /dev/null
+++ b/arch/mips/fw/lib/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for generic prom monitor library routines under Linux.
+#
+
+lib-$(CONFIG_64BIT)	+= call_o32.o
diff --git a/arch/mips/fw/lib/call_o32.S b/arch/mips/fw/lib/call_o32.S
new file mode 100644
index 000000000000..bdf7d1d4081a
--- /dev/null
+++ b/arch/mips/fw/lib/call_o32.S
@@ -0,0 +1,97 @@
+/*
+ *	arch/mips/dec/prom/call_o32.S
+ *
+ *	O32 interface for the 64 (or N32) ABI.
+ *
+ *	Copyright (C) 2002  Maciej W. Rozycki
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ */
+
+#include <asm/asm.h>
+#include <asm/regdef.h>
+
+/* Maximum number of arguments supported.  Must be even!  */
+#define O32_ARGC	32
+/* Number of static registers we save.  */
+#define O32_STATC	11
+/* Frame size for static register  */
+#define O32_FRAMESZ	(SZREG * O32_STATC)
+/* Frame size on new stack */
+#define O32_FRAMESZ_NEW (SZREG + 4 * O32_ARGC)
+
+		.text
+
+/*
+ * O32 function call dispatcher, for interfacing 32-bit ROM routines.
+ *
+ * The standard 64 (N32) calling sequence is supported, with a0
+ * holding a function pointer, a1 a new stack pointer, a2-a7 -- its
+ * first six arguments and the stack -- remaining ones (up to O32_ARGC,
+ * including a2-a7). Static registers, gp and fp are preserved, v0 holds
+ * a result. This code relies on the called o32 function for sp and ra
+ * restoration and this dispatcher has to be placed in a KSEGx (or KUSEG)
+ * address space.  Any pointers passed have to point to addresses within
+ * one of these spaces as well.
+ */
+NESTED(call_o32, O32_FRAMESZ, ra)
+		REG_SUBU	sp,O32_FRAMESZ
+
+		REG_S		ra,O32_FRAMESZ-1*SZREG(sp)
+		REG_S		fp,O32_FRAMESZ-2*SZREG(sp)
+		REG_S		gp,O32_FRAMESZ-3*SZREG(sp)
+		REG_S		s7,O32_FRAMESZ-4*SZREG(sp)
+		REG_S		s6,O32_FRAMESZ-5*SZREG(sp)
+		REG_S		s5,O32_FRAMESZ-6*SZREG(sp)
+		REG_S		s4,O32_FRAMESZ-7*SZREG(sp)
+		REG_S		s3,O32_FRAMESZ-8*SZREG(sp)
+		REG_S		s2,O32_FRAMESZ-9*SZREG(sp)
+		REG_S		s1,O32_FRAMESZ-10*SZREG(sp)
+		REG_S		s0,O32_FRAMESZ-11*SZREG(sp)
+
+		move		jp,a0
+		REG_SUBU	s0,a1,O32_FRAMESZ_NEW
+		REG_S		sp,O32_FRAMESZ_NEW-1*SZREG(s0)
+
+		sll		a0,a2,zero
+		sll		a1,a3,zero
+		sll		a2,a4,zero
+		sll		a3,a5,zero
+		sw		a6,0x10(s0)
+		sw		a7,0x14(s0)
+
+		PTR_LA		t0,O32_FRAMESZ(sp)
+		PTR_LA		t1,0x18(s0)
+		li		t2,O32_ARGC-6
+1:
+		lw		t3,(t0)
+		REG_ADDU	t0,SZREG
+		sw		t3,(t1)
+		REG_SUBU	t2,1
+		REG_ADDU	t1,4
+		bnez		t2,1b
+
+		move		sp,s0
+
+		jalr		jp
+
+		REG_L		sp,O32_FRAMESZ_NEW-1*SZREG(sp)
+
+		REG_L		s0,O32_FRAMESZ-11*SZREG(sp)
+		REG_L		s1,O32_FRAMESZ-10*SZREG(sp)
+		REG_L		s2,O32_FRAMESZ-9*SZREG(sp)
+		REG_L		s3,O32_FRAMESZ-8*SZREG(sp)
+		REG_L		s4,O32_FRAMESZ-7*SZREG(sp)
+		REG_L		s5,O32_FRAMESZ-6*SZREG(sp)
+		REG_L		s6,O32_FRAMESZ-5*SZREG(sp)
+		REG_L		s7,O32_FRAMESZ-4*SZREG(sp)
+		REG_L		gp,O32_FRAMESZ-3*SZREG(sp)
+		REG_L		fp,O32_FRAMESZ-2*SZREG(sp)
+		REG_L		ra,O32_FRAMESZ-1*SZREG(sp)
+
+		REG_ADDU	sp,O32_FRAMESZ
+		jr		ra
+END(call_o32)
diff --git a/arch/mips/fw/sni/Makefile b/arch/mips/fw/sni/Makefile
new file mode 100644
index 000000000000..d9740a3788e2
--- /dev/null
+++ b/arch/mips/fw/sni/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the SNI prom monitor routines under Linux.
+#
+
+lib-$(CONFIG_SNIPROM)	+= sniprom.o
diff --git a/arch/mips/fw/sni/sniprom.c b/arch/mips/fw/sni/sniprom.c
new file mode 100644
index 000000000000..96ba99202758
--- /dev/null
+++ b/arch/mips/fw/sni/sniprom.c
@@ -0,0 +1,151 @@
+/*
+ * Big Endian PROM code for SNI RM machines
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005-2006 Florian Lohoff (flo@rfc822.org)
+ * Copyright (C) 2005-2006 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/console.h>
+
+#include <asm/addrspace.h>
+#include <asm/sni.h>
+#include <asm/mipsprom.h>
+#include <asm/mipsregs.h>
+#include <asm/bootinfo.h>
+
+/* special SNI prom calls */
+/*
+ * This does not exist in all proms - SINIX compares
+ * the prom env variable "version" against "2.0008"
+ * or greater. If lesser it tries to probe interesting
+ * registers
+ */
+#define PROM_GET_MEMCONF	58
+#define PROM_GET_HWCONF         61
+
+#define PROM_VEC		(u64 *)CKSEG1ADDR(0x1fc00000)
+#define PROM_ENTRY(x)		(PROM_VEC + (x))
+
+#define ___prom_putchar         ((int *(*)(int))PROM_ENTRY(PROM_PUTCHAR))
+#define ___prom_getenv          ((char *(*)(char *))PROM_ENTRY(PROM_GETENV))
+#define ___prom_get_memconf     ((void (*)(void *))PROM_ENTRY(PROM_GET_MEMCONF))
+#define ___prom_get_hwconf      ((u32 (*)(void))PROM_ENTRY(PROM_GET_HWCONF))
+
+#ifdef CONFIG_64BIT
+
+static u8 o32_stk[16384];
+#define O32_STK   &o32_stk[sizeof(o32_stk)]
+
+#define __PROM_O32(fun, arg) fun arg __asm__(#fun); \
+				     __asm__(#fun " = call_o32")
+
+int   __PROM_O32(__prom_putchar, (int *(*)(int), void *, int));
+char *__PROM_O32(__prom_getenv, (char *(*)(char *), void *, char *));
+void  __PROM_O32(__prom_get_memconf, (void (*)(void *), void *, void *));
+u32   __PROM_O32(__prom_get_hwconf, (u32 (*)(void), void *));
+
+#define _prom_putchar(x)     __prom_putchar(___prom_putchar, O32_STK, x)
+#define _prom_getenv(x)      __prom_getenv(___prom_getenv, O32_STK, x)
+#define _prom_get_memconf(x) __prom_get_memconf(___prom_get_memconf, O32_STK, x)
+#define _prom_get_hwconf()   __prom_get_hwconf(___prom_get_hwconf, O32_STK)
+
+#else
+#define _prom_putchar(x)     ___prom_putchar(x)
+#define _prom_getenv(x)      ___prom_getenv(x)
+#define _prom_get_memconf(x) ___prom_get_memconf(x)
+#define _prom_get_hwconf(x)  ___prom_get_hwconf(x)
+#endif
+
+void prom_putchar(char c)
+{
+	_prom_putchar(c);
+}
+
+
+char *prom_getenv(char *s)
+{
+	return _prom_getenv(s);
+}
+
+void *prom_get_hwconf(void)
+{
+	u32 hwconf = _prom_get_hwconf();
+
+	if (hwconf == 0xffffffff)
+		return NULL;
+
+	return (void *)CKSEG1ADDR(hwconf);
+}
+
+void __init prom_free_prom_memory(void)
+{
+}
+
+/*
+ * /proc/cpuinfo system type
+ *
+ */
+char *system_type = "Unknown";
+const char *get_system_type(void)
+{
+	return system_type;
+}
+
+static void __init sni_mem_init(void)
+{
+	int i, memsize;
+	struct membank {
+		u32		size;
+		u32		base;
+		u32		size2;
+		u32		pad1;
+		u32		pad2;
+	} memconf[8];
+	int brd_type = *(unsigned char *)SNI_IDPROM_BRDTYPE;
+
+
+	/* MemSIZE from prom in 16MByte chunks */
+	memsize = *((unsigned char *) SNI_IDPROM_MEMSIZE) * 16;
+
+	pr_debug("IDProm memsize: %u MByte\n", memsize);
+
+	/* get memory bank layout from prom */
+	_prom_get_memconf(&memconf);
+
+	pr_debug("prom_get_mem_conf memory configuration:\n");
+	for (i = 0; i < 8 && memconf[i].size; i++) {
+		if (brd_type == SNI_BRD_PCI_TOWER ||
+		    brd_type == SNI_BRD_PCI_TOWER_CPLUS) {
+			if (memconf[i].base >= 0x20000000 &&
+			    memconf[i].base <  0x30000000)
+				memconf[i].base -= 0x20000000;
+		}
+		pr_debug("Bank%d: %08x @ %08x\n", i,
+			memconf[i].size, memconf[i].base);
+		add_memory_region(memconf[i].base, memconf[i].size,
+				  BOOT_MEM_RAM);
+	}
+}
+
+void __init prom_init(void)
+{
+	int argc = fw_arg0;
+	u32 *argv = (u32 *)CKSEG0ADDR(fw_arg1);
+	int i;
+
+	sni_mem_init();
+
+	/* copy prom cmdline parameters to kernel cmdline */
+	for (i = 1; i < argc; i++) {
+		strcat(arcs_cmdline, (char *)CKSEG0ADDR(argv[i]));
+		if (i < (argc - 1))
+			strcat(arcs_cmdline, " ");
+	}
+}
diff --git a/arch/mips/sni/Makefile b/arch/mips/sni/Makefile
index 3a99cd62c0bd..a7dbeebe7fe6 100644
--- a/arch/mips/sni/Makefile
+++ b/arch/mips/sni/Makefile
@@ -3,6 +3,6 @@
 #
 
 obj-y += irq.o reset.o setup.o a20r.o rm200.o pcimt.o pcit.o time.o
-obj-$(CONFIG_CPU_BIG_ENDIAN) += sniprom.o
+obj-$(CONFIG_EISA) += eisa.o
 
 EXTRA_CFLAGS += -Werror
diff --git a/arch/mips/sni/a20r.c b/arch/mips/sni/a20r.c
index b74607599971..3f8cf5eb2f06 100644
--- a/arch/mips/sni/a20r.c
+++ b/arch/mips/sni/a20r.c
@@ -117,10 +117,19 @@ static struct resource sc26xx_rsrc[] = {
 	}
 };
 
+static unsigned int sc26xx_data[2] = {
+	/* DTR   |   RTS    |   DSR    |   CTS     |   DCD     |   RI    */
+	(8 << 0) | (4 << 4) | (6 << 8) | (0 << 12) | (6 << 16) | (0 << 20),
+	(3 << 0) | (2 << 4) | (1 << 8) | (2 << 12) | (3 << 16) | (4 << 20)
+};
+
 static struct platform_device sc26xx_pdev = {
 	.name           = "SC26xx",
 	.num_resources  = ARRAY_SIZE(sc26xx_rsrc),
-	.resource       = sc26xx_rsrc
+	.resource       = sc26xx_rsrc,
+	.dev			= {
+		.platform_data	= sc26xx_data,
+	}
 };
 
 static u32 a20r_ack_hwint(void)
@@ -231,9 +240,9 @@ static int __init snirm_a20r_setup_devinit(void)
 	        platform_device_register(&sc26xx_pdev);
 	        platform_device_register(&a20r_serial8250_device);
 	        platform_device_register(&a20r_ds1216_device);
+		sni_eisa_root_init();
 	        break;
 	}
-
 	return 0;
 }
 
diff --git a/arch/mips/sni/eisa.c b/arch/mips/sni/eisa.c
new file mode 100644
index 000000000000..7396cd719900
--- /dev/null
+++ b/arch/mips/sni/eisa.c
@@ -0,0 +1,50 @@
+/*
+ * Virtual EISA root driver.
+ * Acts as a placeholder if we don't have a proper EISA bridge.
+ *
+ * (C) 2003 Marc Zyngier <maz@wild-wind.fr.eu.org>
+ * modified for SNI usage by Thomas Bogendoerfer
+ *
+ * This code is released under the GPL version 2.
+ */
+
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/eisa.h>
+#include <linux/init.h>
+
+/* The default EISA device parent (virtual root device).
+ * Now use a platform device, since that's the obvious choice. */
+
+static struct platform_device eisa_root_dev = {
+	.name = "eisa",
+	.id   = 0,
+};
+
+static struct eisa_root_device eisa_bus_root = {
+	.dev           = &eisa_root_dev.dev,
+	.bus_base_addr = 0,
+	.res	       = &ioport_resource,
+	.slots	       = EISA_MAX_SLOTS,
+	.dma_mask      = 0xffffffff,
+	.force_probe   = 1,
+};
+
+int __init sni_eisa_root_init(void)
+{
+	int r;
+
+	r = platform_device_register(&eisa_root_dev);
+	if (!r)
+		return r;
+
+	eisa_root_dev.dev.driver_data = &eisa_bus_root;
+
+	if (eisa_root_register(&eisa_bus_root)) {
+		/* A real bridge may have been registered before
+		 * us. So quietly unregister. */
+		platform_device_unregister(&eisa_root_dev);
+		return -1;
+	}
+	return 0;
+}
diff --git a/arch/mips/sni/irq.c b/arch/mips/sni/irq.c
index 9ccffdfb8289..e8e72bb3a9af 100644
--- a/arch/mips/sni/irq.c
+++ b/arch/mips/sni/irq.c
@@ -35,14 +35,14 @@ static irqreturn_t sni_isa_irq_handler(int dummy, void *p)
 	if (unlikely(irq < 0))
 		return IRQ_NONE;
 
-	do_IRQ(irq);
+	generic_handle_irq(irq);
 	return IRQ_HANDLED;
 }
 
 struct irqaction sni_isa_irq = {
 	.handler = sni_isa_irq_handler,
 	.name = "ISA",
-	.flags = IRQF_SHARED
+	.flags = IRQF_SHARED | IRQF_DISABLED
 };
 
 /*
diff --git a/arch/mips/sni/rm200.c b/arch/mips/sni/rm200.c
index 67b061eef6cd..5310aa75afa4 100644
--- a/arch/mips/sni/rm200.c
+++ b/arch/mips/sni/rm200.c
@@ -5,30 +5,36 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2006 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
+ * Copyright (C) 2006,2007 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
+ *
+ * i8259 parts ripped out of arch/mips/kernel/i8259.c
  */
 
+#include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/platform_device.h>
 #include <linux/serial_8250.h>
+#include <linux/io.h>
 
 #include <asm/sni.h>
 #include <asm/time.h>
 #include <asm/irq_cpu.h>
 
-#define PORT(_base,_irq)				\
+#define RM200_I8259A_IRQ_BASE 32
+
+#define MEMPORT(_base,_irq)				\
 	{						\
-		.iobase		= _base,		\
+		.mapbase	= _base,		\
 		.irq		= _irq,			\
 		.uartclk	= 1843200,		\
-		.iotype		= UPIO_PORT,		\
-		.flags		= UPF_BOOT_AUTOCONF,	\
+		.iotype		= UPIO_MEM,		\
+		.flags		= UPF_BOOT_AUTOCONF|UPF_IOREMAP, \
 	}
 
 static struct plat_serial8250_port rm200_data[] = {
-	PORT(0x3f8, 4),
-	PORT(0x2f8, 3),
+	MEMPORT(0x160003f8, RM200_I8259A_IRQ_BASE + 4),
+	MEMPORT(0x160002f8, RM200_I8259A_IRQ_BASE + 3),
 	{ },
 };
 
@@ -112,15 +118,311 @@ static int __init snirm_setup_devinit(void)
 		platform_device_register(&rm200_ds1216_device);
 		platform_device_register(&snirm_82596_rm200_pdev);
 		platform_device_register(&snirm_53c710_rm200_pdev);
+		sni_eisa_root_init();
 	}
 	return 0;
 }
 
 device_initcall(snirm_setup_devinit);
 
+/*
+ * RM200 has an ISA and an EISA bus. The iSA bus is only used
+ * for onboard devices and also has twi i8259 PICs. Since these
+ * PICs are no accessible via inb/outb the following code uses
+ * readb/writeb to access them
+ */
+
+DEFINE_SPINLOCK(sni_rm200_i8259A_lock);
+#define PIC_CMD    0x00
+#define PIC_IMR    0x01
+#define PIC_ISR    PIC_CMD
+#define PIC_POLL   PIC_ISR
+#define PIC_OCW3   PIC_ISR
+
+/* i8259A PIC related value */
+#define PIC_CASCADE_IR		2
+#define MASTER_ICW4_DEFAULT	0x01
+#define SLAVE_ICW4_DEFAULT	0x01
+
+/*
+ * This contains the irq mask for both 8259A irq controllers,
+ */
+static unsigned int rm200_cached_irq_mask = 0xffff;
+static __iomem u8 *rm200_pic_master;
+static __iomem u8 *rm200_pic_slave;
+
+#define cached_master_mask	(rm200_cached_irq_mask)
+#define cached_slave_mask	(rm200_cached_irq_mask >> 8)
+
+static void sni_rm200_disable_8259A_irq(unsigned int irq)
+{
+	unsigned int mask;
+	unsigned long flags;
+
+	irq -= RM200_I8259A_IRQ_BASE;
+	mask = 1 << irq;
+	spin_lock_irqsave(&sni_rm200_i8259A_lock, flags);
+	rm200_cached_irq_mask |= mask;
+	if (irq & 8)
+		writeb(cached_slave_mask, rm200_pic_slave + PIC_IMR);
+	else
+		writeb(cached_master_mask, rm200_pic_master + PIC_IMR);
+	spin_unlock_irqrestore(&sni_rm200_i8259A_lock, flags);
+}
+
+static void sni_rm200_enable_8259A_irq(unsigned int irq)
+{
+	unsigned int mask;
+	unsigned long flags;
+
+	irq -= RM200_I8259A_IRQ_BASE;
+	mask = ~(1 << irq);
+	spin_lock_irqsave(&sni_rm200_i8259A_lock, flags);
+	rm200_cached_irq_mask &= mask;
+	if (irq & 8)
+		writeb(cached_slave_mask, rm200_pic_slave + PIC_IMR);
+	else
+		writeb(cached_master_mask, rm200_pic_master + PIC_IMR);
+	spin_unlock_irqrestore(&sni_rm200_i8259A_lock, flags);
+}
+
+static inline int sni_rm200_i8259A_irq_real(unsigned int irq)
+{
+	int value;
+	int irqmask = 1 << irq;
+
+	if (irq < 8) {
+		writeb(0x0B, rm200_pic_master + PIC_CMD);
+		value = readb(rm200_pic_master + PIC_CMD) & irqmask;
+		writeb(0x0A, rm200_pic_master + PIC_CMD);
+		return value;
+	}
+	writeb(0x0B, rm200_pic_slave + PIC_CMD); /* ISR register */
+	value = readb(rm200_pic_slave + PIC_CMD) & (irqmask >> 8);
+	writeb(0x0A, rm200_pic_slave + PIC_CMD);
+	return value;
+}
+
+/*
+ * Careful! The 8259A is a fragile beast, it pretty
+ * much _has_ to be done exactly like this (mask it
+ * first, _then_ send the EOI, and the order of EOI
+ * to the two 8259s is important!
+ */
+void sni_rm200_mask_and_ack_8259A(unsigned int irq)
+{
+	unsigned int irqmask;
+	unsigned long flags;
+
+	irq -= RM200_I8259A_IRQ_BASE;
+	irqmask = 1 << irq;
+	spin_lock_irqsave(&sni_rm200_i8259A_lock, flags);
+	/*
+	 * Lightweight spurious IRQ detection. We do not want
+	 * to overdo spurious IRQ handling - it's usually a sign
+	 * of hardware problems, so we only do the checks we can
+	 * do without slowing down good hardware unnecessarily.
+	 *
+	 * Note that IRQ7 and IRQ15 (the two spurious IRQs
+	 * usually resulting from the 8259A-1|2 PICs) occur
+	 * even if the IRQ is masked in the 8259A. Thus we
+	 * can check spurious 8259A IRQs without doing the
+	 * quite slow i8259A_irq_real() call for every IRQ.
+	 * This does not cover 100% of spurious interrupts,
+	 * but should be enough to warn the user that there
+	 * is something bad going on ...
+	 */
+	if (rm200_cached_irq_mask & irqmask)
+		goto spurious_8259A_irq;
+	rm200_cached_irq_mask |= irqmask;
+
+handle_real_irq:
+	if (irq & 8) {
+		readb(rm200_pic_slave + PIC_IMR);
+		writeb(cached_slave_mask, rm200_pic_slave + PIC_IMR);
+		writeb(0x60+(irq & 7), rm200_pic_slave + PIC_CMD);
+		writeb(0x60+PIC_CASCADE_IR, rm200_pic_master + PIC_CMD);
+	} else {
+		readb(rm200_pic_master + PIC_IMR);
+		writeb(cached_master_mask, rm200_pic_master + PIC_IMR);
+		writeb(0x60+irq, rm200_pic_master + PIC_CMD);
+	}
+	spin_unlock_irqrestore(&sni_rm200_i8259A_lock, flags);
+	return;
+
+spurious_8259A_irq:
+	/*
+	 * this is the slow path - should happen rarely.
+	 */
+	if (sni_rm200_i8259A_irq_real(irq))
+		/*
+		 * oops, the IRQ _is_ in service according to the
+		 * 8259A - not spurious, go handle it.
+		 */
+		goto handle_real_irq;
+
+	{
+		static int spurious_irq_mask;
+		/*
+		 * At this point we can be sure the IRQ is spurious,
+		 * lets ACK and report it. [once per IRQ]
+		 */
+		if (!(spurious_irq_mask & irqmask)) {
+			printk(KERN_DEBUG
+			       "spurious RM200 8259A interrupt: IRQ%d.\n", irq);
+			spurious_irq_mask |= irqmask;
+		}
+		atomic_inc(&irq_err_count);
+		/*
+		 * Theoretically we do not have to handle this IRQ,
+		 * but in Linux this does not cause problems and is
+		 * simpler for us.
+		 */
+		goto handle_real_irq;
+	}
+}
+
+static struct irq_chip sni_rm200_i8259A_chip = {
+	.name		= "RM200-XT-PIC",
+	.mask		= sni_rm200_disable_8259A_irq,
+	.unmask		= sni_rm200_enable_8259A_irq,
+	.mask_ack	= sni_rm200_mask_and_ack_8259A,
+};
+
+/*
+ * Do the traditional i8259 interrupt polling thing.  This is for the few
+ * cases where no better interrupt acknowledge method is available and we
+ * absolutely must touch the i8259.
+ */
+static inline int sni_rm200_i8259_irq(void)
+{
+	int irq;
+
+	spin_lock(&sni_rm200_i8259A_lock);
+
+	/* Perform an interrupt acknowledge cycle on controller 1. */
+	writeb(0x0C, rm200_pic_master + PIC_CMD);	/* prepare for poll */
+	irq = readb(rm200_pic_master + PIC_CMD) & 7;
+	if (irq == PIC_CASCADE_IR) {
+		/*
+		 * Interrupt is cascaded so perform interrupt
+		 * acknowledge on controller 2.
+		 */
+		writeb(0x0C, rm200_pic_slave + PIC_CMD); /* prepare for poll */
+		irq = (readb(rm200_pic_slave + PIC_CMD) & 7) + 8;
+	}
+
+	if (unlikely(irq == 7)) {
+		/*
+		 * This may be a spurious interrupt.
+		 *
+		 * Read the interrupt status register (ISR). If the most
+		 * significant bit is not set then there is no valid
+		 * interrupt.
+		 */
+		writeb(0x0B, rm200_pic_master + PIC_ISR); /* ISR register */
+		if (~readb(rm200_pic_master + PIC_ISR) & 0x80)
+			irq = -1;
+	}
+
+	spin_unlock(&sni_rm200_i8259A_lock);
+
+	return likely(irq >= 0) ? irq + RM200_I8259A_IRQ_BASE : irq;
+}
+
+void sni_rm200_init_8259A(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sni_rm200_i8259A_lock, flags);
+
+	writeb(0xff, rm200_pic_master + PIC_IMR);
+	writeb(0xff, rm200_pic_slave + PIC_IMR);
+
+	writeb(0x11, rm200_pic_master + PIC_CMD);
+	writeb(0, rm200_pic_master + PIC_IMR);
+	writeb(1U << PIC_CASCADE_IR, rm200_pic_master + PIC_IMR);
+	writeb(MASTER_ICW4_DEFAULT, rm200_pic_master + PIC_IMR);
+	writeb(0x11, rm200_pic_slave + PIC_CMD);
+	writeb(8, rm200_pic_slave + PIC_IMR);
+	writeb(PIC_CASCADE_IR, rm200_pic_slave + PIC_IMR);
+	writeb(SLAVE_ICW4_DEFAULT, rm200_pic_slave + PIC_IMR);
+	udelay(100);		/* wait for 8259A to initialize */
+
+	writeb(cached_master_mask, rm200_pic_master + PIC_IMR);
+	writeb(cached_slave_mask, rm200_pic_slave + PIC_IMR);
+
+	spin_unlock_irqrestore(&sni_rm200_i8259A_lock, flags);
+}
+
+/*
+ * IRQ2 is cascade interrupt to second interrupt controller
+ */
+static struct irqaction sni_rm200_irq2 = {
+	no_action, 0, CPU_MASK_NONE, "cascade", NULL, NULL
+};
+
+static struct resource sni_rm200_pic1_resource = {
+	.name = "onboard ISA pic1",
+	.start = 0x16000020,
+	.end = 0x16000023,
+	.flags = IORESOURCE_BUSY
+};
+
+static struct resource sni_rm200_pic2_resource = {
+	.name = "onboard ISA pic2",
+	.start = 0x160000a0,
+	.end = 0x160000a3,
+	.flags = IORESOURCE_BUSY
+};
+
+/* ISA irq handler */
+static irqreturn_t sni_rm200_i8259A_irq_handler(int dummy, void *p)
+{
+	int irq;
+
+	irq = sni_rm200_i8259_irq();
+	if (unlikely(irq < 0))
+		return IRQ_NONE;
+
+	do_IRQ(irq);
+	return IRQ_HANDLED;
+}
+
+struct irqaction sni_rm200_i8259A_irq = {
+	.handler = sni_rm200_i8259A_irq_handler,
+	.name = "onboard ISA",
+	.flags = IRQF_SHARED
+};
+
+void __init sni_rm200_i8259_irqs(void)
+{
+	int i;
+
+	rm200_pic_master = ioremap_nocache(0x16000020, 4);
+	if (!rm200_pic_master)
+		return;
+	rm200_pic_slave = ioremap_nocache(0x160000a0, 4);
+	if (!rm200_pic_master) {
+		iounmap(rm200_pic_master);
+		return;
+	}
+
+	insert_resource(&iomem_resource, &sni_rm200_pic1_resource);
+	insert_resource(&iomem_resource, &sni_rm200_pic2_resource);
+
+	sni_rm200_init_8259A();
+
+	for (i = RM200_I8259A_IRQ_BASE; i < RM200_I8259A_IRQ_BASE + 16; i++)
+		set_irq_chip_and_handler(i, &sni_rm200_i8259A_chip,
+					 handle_level_irq);
+
+	setup_irq(RM200_I8259A_IRQ_BASE + PIC_CASCADE_IR, &sni_rm200_irq2);
+}
+
 
-#define SNI_RM200_INT_STAT_REG  0xbc000000
-#define SNI_RM200_INT_ENA_REG   0xbc080000
+#define SNI_RM200_INT_STAT_REG  CKSEG1ADDR(0xbc000000)
+#define SNI_RM200_INT_ENA_REG   CKSEG1ADDR(0xbc080000)
 
 #define SNI_RM200_INT_START  24
 #define SNI_RM200_INT_END    28
@@ -181,17 +483,17 @@ void __init sni_rm200_irq_init(void)
 
 	* (volatile u8 *)SNI_RM200_INT_ENA_REG = 0x1f;
 
+	sni_rm200_i8259_irqs();
 	mips_cpu_irq_init();
 	/* Actually we've got more interrupts to handle ...  */
 	for (i = SNI_RM200_INT_START; i <= SNI_RM200_INT_END; i++)
 		set_irq_chip(i, &rm200_irq_type);
 	sni_hwint = sni_rm200_hwint;
 	change_c0_status(ST0_IM, IE_IRQ0);
-	setup_irq(SNI_RM200_INT_START + 0, &sni_isa_irq);
+	setup_irq(SNI_RM200_INT_START + 0, &sni_rm200_i8259A_irq);
+	setup_irq(SNI_RM200_INT_START + 1, &sni_isa_irq);
 }
 
 void __init sni_rm200_init(void)
 {
-	set_io_port_base(SNI_PORT_BASE + 0x02000000);
-	ioport_resource.end += 0x02000000;
 }
diff --git a/arch/mips/sni/setup.c b/arch/mips/sni/setup.c
index e8b26bdee24c..5484e1c62054 100644
--- a/arch/mips/sni/setup.c
+++ b/arch/mips/sni/setup.c
@@ -19,11 +19,17 @@
 #include <asm/sgialib.h>
 #endif
 
+#ifdef CONFIG_SNIPROM
+#include <asm/mipsprom.h>
+#endif
+
+#include <asm/bootinfo.h>
 #include <asm/io.h>
 #include <asm/reboot.h>
 #include <asm/sni.h>
 
 unsigned int sni_brd_type;
+EXPORT_SYMBOL(sni_brd_type);
 
 extern void sni_machine_restart(char *command);
 extern void sni_machine_power_off(void);
@@ -47,20 +53,152 @@ static void __init sni_display_setup(void)
 #endif
 }
 
+static void __init sni_console_setup(void)
+{
+#ifndef CONFIG_ARC
+	char *ctype;
+	char *cdev;
+	char *baud;
+	int port;
+	static char options[8];
+
+	cdev = prom_getenv("console_dev");
+	if (strncmp(cdev, "tty", 3) == 0) {
+		ctype = prom_getenv("console");
+		switch (*ctype) {
+		default:
+		case 'l':
+			port = 0;
+			baud = prom_getenv("lbaud");
+			break;
+		case 'r':
+			port = 1;
+			baud = prom_getenv("rbaud");
+			break;
+		}
+		if (baud)
+			strcpy(options, baud);
+		if (strncmp(cdev, "tty552", 6) == 0)
+			add_preferred_console("ttyS", port,
+					      baud ? options : NULL);
+		else
+			add_preferred_console("ttySC", port,
+					      baud ? options : NULL);
+	}
+#endif
+}
+
+#ifdef DEBUG
+static void __init sni_idprom_dump(void)
+{
+	int	i;
+
+	pr_debug("SNI IDProm dump:\n");
+	for (i = 0; i < 256; i++) {
+		if (i%16 == 0)
+			pr_debug("%04x ", i);
+
+		printk("%02x ", *(unsigned char *) (SNI_IDPROM_BASE + i));
+
+		if (i % 16 == 15)
+			printk("\n");
+	}
+}
+#endif
 
 void __init plat_mem_setup(void)
 {
+	int cputype;
+
 	set_io_port_base(SNI_PORT_BASE);
 //	ioport_resource.end = sni_io_resource.end;
 
 	/*
 	 * Setup (E)ISA I/O memory access stuff
 	 */
-	isa_slot_offset = 0xb0000000;
+	isa_slot_offset = CKSEG1ADDR(0xb0000000);
 #ifdef CONFIG_EISA
 	EISA_bus = 1;
 #endif
 
+	sni_brd_type = *(unsigned char *)SNI_IDPROM_BRDTYPE;
+	cputype = *(unsigned char *)SNI_IDPROM_CPUTYPE;
+	switch (sni_brd_type) {
+	case SNI_BRD_TOWER_OASIC:
+		switch (cputype) {
+		case SNI_CPU_M8030:
+			system_type = "RM400-330";
+			break;
+		case SNI_CPU_M8031:
+			system_type = "RM400-430";
+			break;
+		case SNI_CPU_M8037:
+			system_type = "RM400-530";
+			break;
+		case SNI_CPU_M8034:
+			system_type = "RM400-730";
+			break;
+		default:
+			system_type = "RM400-xxx";
+			break;
+		}
+		break;
+	case SNI_BRD_MINITOWER:
+		switch (cputype) {
+		case SNI_CPU_M8021:
+		case SNI_CPU_M8043:
+			system_type = "RM400-120";
+			break;
+		case SNI_CPU_M8040:
+			system_type = "RM400-220";
+			break;
+		case SNI_CPU_M8053:
+			system_type = "RM400-225";
+			break;
+		case SNI_CPU_M8050:
+			system_type = "RM400-420";
+			break;
+		default:
+			system_type = "RM400-xxx";
+			break;
+		}
+		break;
+	case SNI_BRD_PCI_TOWER:
+		system_type = "RM400-Cxx";
+		break;
+	case SNI_BRD_RM200:
+		system_type = "RM200-xxx";
+		break;
+	case SNI_BRD_PCI_MTOWER:
+		system_type = "RM300-Cxx";
+		break;
+	case SNI_BRD_PCI_DESKTOP:
+		switch (read_c0_prid() & 0xff00) {
+		case PRID_IMP_R4600:
+		case PRID_IMP_R4700:
+			system_type = "RM200-C20";
+			break;
+		case PRID_IMP_R5000:
+			system_type = "RM200-C40";
+			break;
+		default:
+			system_type = "RM200-Cxx";
+			break;
+		}
+		break;
+	case SNI_BRD_PCI_TOWER_CPLUS:
+		system_type = "RM400-Exx";
+		break;
+	case SNI_BRD_PCI_MTOWER_CPLUS:
+		system_type = "RM300-Exx";
+		break;
+	}
+	pr_debug("Found SNI brdtype %02x name %s\n", sni_brd_type, system_type);
+
+#ifdef DEBUG
+	sni_idprom_dump();
+#endif
+
 	switch (sni_brd_type) {
 	case SNI_BRD_10:
 	case SNI_BRD_10NEW:
@@ -89,9 +227,10 @@ void __init plat_mem_setup(void)
 	pm_power_off = sni_machine_power_off;
 
 	sni_display_setup();
+	sni_console_setup();
 }
 
-#if CONFIG_PCI
+#ifdef CONFIG_PCI
 
 #include <linux/pci.h>
 #include <video/vga.h>
diff --git a/arch/mips/sni/sniprom.c b/arch/mips/sni/sniprom.c
deleted file mode 100644
index eff4b89d7b75..000000000000
--- a/arch/mips/sni/sniprom.c
+++ /dev/null
@@ -1,251 +0,0 @@
-/*
- * Big Endian PROM code for SNI RM machines
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2005-2006 Florian Lohoff (flo@rfc822.org)
- * Copyright (C) 2005-2006 Thomas Bogendoerfer (tsbogend@alpha.franken.de)
- */
-
-#define DEBUG
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/console.h>
-
-#include <asm/addrspace.h>
-#include <asm/sni.h>
-#include <asm/mipsprom.h>
-#include <asm/mipsregs.h>
-#include <asm/bootinfo.h>
-
-/* special SNI prom calls */
-/*
- * This does not exist in all proms - SINIX compares
- * the prom env variable "version" against "2.0008"
- * or greater. If lesser it tries to probe interesting
- * registers
- */
-#define PROM_GET_MEMCONF	58
-
-#define PROM_VEC		(u64 *)CKSEG1ADDR(0x1fc00000)
-#define PROM_ENTRY(x)		(PROM_VEC + (x))
-
-
-static int *(*__prom_putchar)(int)        = (int *(*)(int))PROM_ENTRY(PROM_PUTCHAR);
-
-void prom_putchar(char c)
-{
-	__prom_putchar(c);
-}
-
-static char *(*__prom_getenv)(char *)     = (char *(*)(char *))PROM_ENTRY(PROM_GETENV);
-static void (*__prom_get_memconf)(void *) = (void (*)(void *))PROM_ENTRY(PROM_GET_MEMCONF);
-
-char *prom_getenv(char *s)
-{
-	return __prom_getenv(s);
-}
-
-void __init prom_free_prom_memory(void)
-{
-}
-
-/*
- * /proc/cpuinfo system type
- *
- */
-static const char *systype = "Unknown";
-const char *get_system_type(void)
-{
-	return systype;
-}
-
-#define SNI_IDPROM_BASE                0xbff00000
-#define SNI_IDPROM_MEMSIZE             (SNI_IDPROM_BASE+0x28)  /* Memsize in 16MB quantities */
-#define SNI_IDPROM_BRDTYPE             (SNI_IDPROM_BASE+0x29)  /* Board Type */
-#define SNI_IDPROM_CPUTYPE             (SNI_IDPROM_BASE+0x30)  /* CPU Type */
-
-#define SNI_IDPROM_SIZE	0x1000
-
-#ifdef DEBUG
-static void __init sni_idprom_dump(void)
-{
-	int	i;
-
-	pr_debug("SNI IDProm dump:\n");
-	for (i = 0; i < 256; i++) {
-		if (i%16 == 0)
-			pr_debug("%04x ", i);
-
-		printk("%02x ", *(unsigned char *) (SNI_IDPROM_BASE + i));
-
-		if (i % 16 == 15)
-			printk("\n");
-	}
-}
-#endif
-
-static void __init sni_mem_init(void )
-{
-	int i, memsize;
-	struct membank {
-	        u32		size;
-	        u32		base;
-	        u32		size2;
-	        u32		pad1;
-	        u32		pad2;
-	} memconf[8];
-
-	/* MemSIZE from prom in 16MByte chunks */
-	memsize = *((unsigned char *) SNI_IDPROM_MEMSIZE) * 16;
-
-	pr_debug("IDProm memsize: %lu MByte\n", memsize);
-
-	/* get memory bank layout from prom */
-	__prom_get_memconf(&memconf);
-
-	pr_debug("prom_get_mem_conf memory configuration:\n");
-	for (i = 0;i < 8 && memconf[i].size; i++) {
-		if (sni_brd_type == SNI_BRD_PCI_TOWER ||
-		    sni_brd_type == SNI_BRD_PCI_TOWER_CPLUS) {
-			if (memconf[i].base >= 0x20000000 &&
-			    memconf[i].base <  0x30000000) {
-				memconf[i].base -= 0x20000000;
-			}
-	}
-		pr_debug("Bank%d: %08x @ %08x\n", i,
-			memconf[i].size, memconf[i].base);
-		add_memory_region(memconf[i].base, memconf[i].size, BOOT_MEM_RAM);
-	}
-}
-
-static void __init sni_console_setup(void)
-{
-	char *ctype;
-	char *cdev;
-	char *baud;
-	int port;
-	static char options[8];
-
-	cdev = prom_getenv("console_dev");
-	if (strncmp (cdev, "tty", 3) == 0) {
-		ctype = prom_getenv("console");
-		switch (*ctype) {
-		default:
-		case 'l':
-	                port = 0;
-	                baud = prom_getenv("lbaud");
-	                break;
-		case 'r':
-	                port = 1;
-	                baud = prom_getenv("rbaud");
-	                break;
-		}
-		if (baud)
-			strcpy(options, baud);
-		if (strncmp (cdev, "tty552", 6) == 0)
-			add_preferred_console("ttyS", port, baud ? options : NULL);
-		else
-			add_preferred_console("ttySC", port, baud ? options : NULL);
-	}
-}
-
-void __init prom_init(void)
-{
-	int argc = fw_arg0;
-	char **argv = (void *)fw_arg1;
-	int i;
-	int cputype;
-
-	sni_brd_type = *(unsigned char *)SNI_IDPROM_BRDTYPE;
-	cputype = *(unsigned char *)SNI_IDPROM_CPUTYPE;
-	switch (sni_brd_type) {
-	case SNI_BRD_TOWER_OASIC:
-	        switch (cputype) {
-		case SNI_CPU_M8030:
-		        systype = "RM400-330";
-		        break;
-		case SNI_CPU_M8031:
-		        systype = "RM400-430";
-		        break;
-		case SNI_CPU_M8037:
-		        systype = "RM400-530";
-		        break;
-		case SNI_CPU_M8034:
-		        systype = "RM400-730";
-		        break;
-		default:
-			systype = "RM400-xxx";
-			break;
-		}
-	        break;
-	case SNI_BRD_MINITOWER:
-	        switch (cputype) {
-		case SNI_CPU_M8021:
-		case SNI_CPU_M8043:
-		        systype = "RM400-120";
-		        break;
-		case SNI_CPU_M8040:
-		        systype = "RM400-220";
-		        break;
-		case SNI_CPU_M8053:
-		        systype = "RM400-225";
-		        break;
-		case SNI_CPU_M8050:
-		        systype = "RM400-420";
-		        break;
-		default:
-			systype = "RM400-xxx";
-			break;
-		}
-	        break;
-	case SNI_BRD_PCI_TOWER:
-	        systype = "RM400-Cxx";
-	        break;
-	case SNI_BRD_RM200:
-	        systype = "RM200-xxx";
-	        break;
-	case SNI_BRD_PCI_MTOWER:
-	        systype = "RM300-Cxx";
-	        break;
-	case SNI_BRD_PCI_DESKTOP:
-	        switch (read_c0_prid() & 0xff00) {
-		case PRID_IMP_R4600:
-		case PRID_IMP_R4700:
-		        systype = "RM200-C20";
-		        break;
-		case PRID_IMP_R5000:
-		        systype = "RM200-C40";
-		        break;
-		default:
-		        systype = "RM200-Cxx";
-		        break;
-		}
-	        break;
-	case SNI_BRD_PCI_TOWER_CPLUS:
-	        systype = "RM400-Exx";
-	        break;
-	case SNI_BRD_PCI_MTOWER_CPLUS:
-	        systype = "RM300-Exx";
-	        break;
-	}
-	pr_debug("Found SNI brdtype %02x name %s\n", sni_brd_type, systype);
-
-#ifdef DEBUG
-	sni_idprom_dump();
-#endif
-	sni_mem_init();
-	sni_console_setup();
-
-	/* copy prom cmdline parameters to kernel cmdline */
-	for (i = 1; i < argc; i++) {
-		strcat(arcs_cmdline, argv[i]);
-		if (i < (argc - 1))
-			strcat(arcs_cmdline, " ");
-	}
-}
-
diff --git a/arch/mips/sni/time.c b/arch/mips/sni/time.c
index 6f339af08d22..796e3ce28720 100644
--- a/arch/mips/sni/time.c
+++ b/arch/mips/sni/time.c
@@ -178,6 +178,7 @@ void __init plat_time_init(void)
 		sni_a20r_timer_setup();
 		break;
 	}
+	setup_pit_timer();
 }
 
 unsigned long read_persistent_clock(void)
diff --git a/include/asm-mips/bootinfo.h b/include/asm-mips/bootinfo.h
index 8e9bb16f0aa2..e031bdff9920 100644
--- a/include/asm-mips/bootinfo.h
+++ b/include/asm-mips/bootinfo.h
@@ -96,6 +96,7 @@
 
 #define CL_SIZE			COMMAND_LINE_SIZE
 
+extern char *system_type;
 const char *get_system_type(void);
 
 extern unsigned long mips_machtype;
diff --git a/include/asm-mips/mipsprom.h b/include/asm-mips/mipsprom.h
index ce7cff7f1e8e..146d41b67adc 100644
--- a/include/asm-mips/mipsprom.h
+++ b/include/asm-mips/mipsprom.h
@@ -71,4 +71,6 @@
 #define PROM_NV_GET		53	/* XXX */
 #define PROM_NV_SET		54	/* XXX */
 
+extern char *prom_getenv(char *);
+
 #endif /* __ASM_MIPS_PROM_H */
diff --git a/include/asm-mips/sni.h b/include/asm-mips/sni.h
index af081457f847..e716447e5e03 100644
--- a/include/asm-mips/sni.h
+++ b/include/asm-mips/sni.h
@@ -35,23 +35,23 @@ extern unsigned int sni_brd_type;
 #define SNI_CPU_M8050           0x0b
 #define SNI_CPU_M8053           0x0d
 
-#define SNI_PORT_BASE		0xb4000000
+#define SNI_PORT_BASE		CKSEG1ADDR(0xb4000000)
 
 #ifndef __MIPSEL__
 /*
  * ASIC PCI registers for big endian configuration.
  */
-#define PCIMT_UCONF		0xbfff0004
-#define PCIMT_IOADTIMEOUT2	0xbfff000c
-#define PCIMT_IOMEMCONF		0xbfff0014
-#define PCIMT_IOMMU		0xbfff001c
-#define PCIMT_IOADTIMEOUT1	0xbfff0024
-#define PCIMT_DMAACCESS		0xbfff002c
-#define PCIMT_DMAHIT		0xbfff0034
-#define PCIMT_ERRSTATUS		0xbfff003c
-#define PCIMT_ERRADDR		0xbfff0044
-#define PCIMT_SYNDROME		0xbfff004c
-#define PCIMT_ITPEND		0xbfff0054
+#define PCIMT_UCONF		CKSEG1ADDR(0xbfff0004)
+#define PCIMT_IOADTIMEOUT2	CKSEG1ADDR(0xbfff000c)
+#define PCIMT_IOMEMCONF		CKSEG1ADDR(0xbfff0014)
+#define PCIMT_IOMMU		CKSEG1ADDR(0xbfff001c)
+#define PCIMT_IOADTIMEOUT1	CKSEG1ADDR(0xbfff0024)
+#define PCIMT_DMAACCESS		CKSEG1ADDR(0xbfff002c)
+#define PCIMT_DMAHIT		CKSEG1ADDR(0xbfff0034)
+#define PCIMT_ERRSTATUS		CKSEG1ADDR(0xbfff003c)
+#define PCIMT_ERRADDR		CKSEG1ADDR(0xbfff0044)
+#define PCIMT_SYNDROME		CKSEG1ADDR(0xbfff004c)
+#define PCIMT_ITPEND		CKSEG1ADDR(0xbfff0054)
 #define  IT_INT2		0x01
 #define  IT_INTD		0x02
 #define  IT_INTC		0x04
@@ -60,32 +60,32 @@ extern unsigned int sni_brd_type;
 #define  IT_EISA		0x20
 #define  IT_SCSI		0x40
 #define  IT_ETH			0x80
-#define PCIMT_IRQSEL		0xbfff005c
-#define PCIMT_TESTMEM		0xbfff0064
-#define PCIMT_ECCREG		0xbfff006c
-#define PCIMT_CONFIG_ADDRESS	0xbfff0074
-#define PCIMT_ASIC_ID		0xbfff007c	/* read */
-#define PCIMT_SOFT_RESET	0xbfff007c	/* write */
-#define PCIMT_PIA_OE		0xbfff0084
-#define PCIMT_PIA_DATAOUT	0xbfff008c
-#define PCIMT_PIA_DATAIN	0xbfff0094
-#define PCIMT_CACHECONF		0xbfff009c
-#define PCIMT_INVSPACE		0xbfff00a4
+#define PCIMT_IRQSEL		CKSEG1ADDR(0xbfff005c)
+#define PCIMT_TESTMEM		CKSEG1ADDR(0xbfff0064)
+#define PCIMT_ECCREG		CKSEG1ADDR(0xbfff006c)
+#define PCIMT_CONFIG_ADDRESS	CKSEG1ADDR(0xbfff0074)
+#define PCIMT_ASIC_ID		CKSEG1ADDR(0xbfff007c)	/* read */
+#define PCIMT_SOFT_RESET	CKSEG1ADDR(0xbfff007c)	/* write */
+#define PCIMT_PIA_OE		CKSEG1ADDR(0xbfff0084)
+#define PCIMT_PIA_DATAOUT	CKSEG1ADDR(0xbfff008c)
+#define PCIMT_PIA_DATAIN	CKSEG1ADDR(0xbfff0094)
+#define PCIMT_CACHECONF		CKSEG1ADDR(0xbfff009c)
+#define PCIMT_INVSPACE		CKSEG1ADDR(0xbfff00a4)
 #else
 /*
  * ASIC PCI registers for little endian configuration.
  */
-#define PCIMT_UCONF		0xbfff0000
-#define PCIMT_IOADTIMEOUT2	0xbfff0008
-#define PCIMT_IOMEMCONF		0xbfff0010
-#define PCIMT_IOMMU		0xbfff0018
-#define PCIMT_IOADTIMEOUT1	0xbfff0020
-#define PCIMT_DMAACCESS		0xbfff0028
-#define PCIMT_DMAHIT		0xbfff0030
-#define PCIMT_ERRSTATUS		0xbfff0038
-#define PCIMT_ERRADDR		0xbfff0040
-#define PCIMT_SYNDROME		0xbfff0048
-#define PCIMT_ITPEND		0xbfff0050
+#define PCIMT_UCONF		CKSEG1ADDR(0xbfff0000)
+#define PCIMT_IOADTIMEOUT2	CKSEG1ADDR(0xbfff0008)
+#define PCIMT_IOMEMCONF		CKSEG1ADDR(0xbfff0010)
+#define PCIMT_IOMMU		CKSEG1ADDR(0xbfff0018)
+#define PCIMT_IOADTIMEOUT1	CKSEG1ADDR(0xbfff0020)
+#define PCIMT_DMAACCESS		CKSEG1ADDR(0xbfff0028)
+#define PCIMT_DMAHIT		CKSEG1ADDR(0xbfff0030)
+#define PCIMT_ERRSTATUS		CKSEG1ADDR(0xbfff0038)
+#define PCIMT_ERRADDR		CKSEG1ADDR(0xbfff0040)
+#define PCIMT_SYNDROME		CKSEG1ADDR(0xbfff0048)
+#define PCIMT_ITPEND		CKSEG1ADDR(0xbfff0050)
 #define  IT_INT2		0x01
 #define  IT_INTD		0x02
 #define  IT_INTC		0x04
@@ -94,20 +94,20 @@ extern unsigned int sni_brd_type;
 #define  IT_EISA		0x20
 #define  IT_SCSI		0x40
 #define  IT_ETH			0x80
-#define PCIMT_IRQSEL		0xbfff0058
-#define PCIMT_TESTMEM		0xbfff0060
-#define PCIMT_ECCREG		0xbfff0068
-#define PCIMT_CONFIG_ADDRESS	0xbfff0070
-#define PCIMT_ASIC_ID		0xbfff0078	/* read */
-#define PCIMT_SOFT_RESET	0xbfff0078	/* write */
-#define PCIMT_PIA_OE		0xbfff0080
-#define PCIMT_PIA_DATAOUT	0xbfff0088
-#define PCIMT_PIA_DATAIN	0xbfff0090
-#define PCIMT_CACHECONF		0xbfff0098
-#define PCIMT_INVSPACE		0xbfff00a0
+#define PCIMT_IRQSEL		CKSEG1ADDR(0xbfff0058)
+#define PCIMT_TESTMEM		CKSEG1ADDR(0xbfff0060)
+#define PCIMT_ECCREG		CKSEG1ADDR(0xbfff0068)
+#define PCIMT_CONFIG_ADDRESS	CKSEG1ADDR(0xbfff0070)
+#define PCIMT_ASIC_ID		CKSEG1ADDR(0xbfff0078)	/* read */
+#define PCIMT_SOFT_RESET	CKSEG1ADDR(0xbfff0078)	/* write */
+#define PCIMT_PIA_OE		CKSEG1ADDR(0xbfff0080)
+#define PCIMT_PIA_DATAOUT	CKSEG1ADDR(0xbfff0088)
+#define PCIMT_PIA_DATAIN	CKSEG1ADDR(0xbfff0090)
+#define PCIMT_CACHECONF		CKSEG1ADDR(0xbfff0098)
+#define PCIMT_INVSPACE		CKSEG1ADDR(0xbfff00a0)
 #endif
 
-#define PCIMT_PCI_CONF		0xbfff0100
+#define PCIMT_PCI_CONF		CKSEG1ADDR(0xbfff0100)
 
 /*
  * Data port for the PCI bus in IO space
@@ -117,34 +117,34 @@ extern unsigned int sni_brd_type;
 /*
  * Board specific registers
  */
-#define PCIMT_CSMSR		0xbfd00000
-#define PCIMT_CSSWITCH		0xbfd10000
-#define PCIMT_CSITPEND		0xbfd20000
-#define PCIMT_AUTO_PO_EN	0xbfd30000
-#define PCIMT_CLR_TEMP		0xbfd40000
-#define PCIMT_AUTO_PO_DIS	0xbfd50000
-#define PCIMT_EXMSR		0xbfd60000
-#define PCIMT_UNUSED1		0xbfd70000
-#define PCIMT_CSWCSM		0xbfd80000
-#define PCIMT_UNUSED2		0xbfd90000
-#define PCIMT_CSLED		0xbfda0000
-#define PCIMT_CSMAPISA		0xbfdb0000
-#define PCIMT_CSRSTBP		0xbfdc0000
-#define PCIMT_CLRPOFF		0xbfdd0000
-#define PCIMT_CSTIMER		0xbfde0000
-#define PCIMT_PWDN		0xbfdf0000
+#define PCIMT_CSMSR		CKSEG1ADDR(0xbfd00000)
+#define PCIMT_CSSWITCH		CKSEG1ADDR(0xbfd10000)
+#define PCIMT_CSITPEND		CKSEG1ADDR(0xbfd20000)
+#define PCIMT_AUTO_PO_EN	CKSEG1ADDR(0xbfd30000)
+#define PCIMT_CLR_TEMP		CKSEG1ADDR(0xbfd40000)
+#define PCIMT_AUTO_PO_DIS	CKSEG1ADDR(0xbfd50000)
+#define PCIMT_EXMSR		CKSEG1ADDR(0xbfd60000)
+#define PCIMT_UNUSED1		CKSEG1ADDR(0xbfd70000)
+#define PCIMT_CSWCSM		CKSEG1ADDR(0xbfd80000)
+#define PCIMT_UNUSED2		CKSEG1ADDR(0xbfd90000)
+#define PCIMT_CSLED		CKSEG1ADDR(0xbfda0000)
+#define PCIMT_CSMAPISA		CKSEG1ADDR(0xbfdb0000)
+#define PCIMT_CSRSTBP		CKSEG1ADDR(0xbfdc0000)
+#define PCIMT_CLRPOFF		CKSEG1ADDR(0xbfdd0000)
+#define PCIMT_CSTIMER		CKSEG1ADDR(0xbfde0000)
+#define PCIMT_PWDN		CKSEG1ADDR(0xbfdf0000)
 
 /*
  * A20R based boards
  */
-#define A20R_PT_CLOCK_BASE      0xbc040000
-#define A20R_PT_TIM0_ACK        0xbc050000
-#define A20R_PT_TIM1_ACK        0xbc060000
+#define A20R_PT_CLOCK_BASE      CKSEG1ADDR(0xbc040000)
+#define A20R_PT_TIM0_ACK        CKSEG1ADDR(0xbc050000)
+#define A20R_PT_TIM1_ACK        CKSEG1ADDR(0xbc060000)
 
 #define SNI_A20R_IRQ_BASE       MIPS_CPU_IRQ_BASE
 #define SNI_A20R_IRQ_TIMER      (SNI_A20R_IRQ_BASE+5)
 
-#define SNI_PCIT_INT_REG        0xbfff000c
+#define SNI_PCIT_INT_REG        CKSEG1ADDR(0xbfff000c)
 
 #define SNI_PCIT_INT_START      24
 #define SNI_PCIT_INT_END        30
@@ -186,10 +186,30 @@ extern unsigned int sni_brd_type;
 /*
  * Base address for the mapped 16mb EISA bus segment.
  */
-#define PCIMT_EISA_BASE		0xb0000000
+#define PCIMT_EISA_BASE		CKSEG1ADDR(0xb0000000)
 
 /* PCI EISA Interrupt acknowledge  */
-#define PCIMT_INT_ACKNOWLEDGE	0xba000000
+#define PCIMT_INT_ACKNOWLEDGE	CKSEG1ADDR(0xba000000)
+
+/*
+ *  SNI ID PROM
+ *
+ * SNI_IDPROM_MEMSIZE  Memsize in 16MB quantities
+ * SNI_IDPROM_BRDTYPE  Board Type
+ * SNI_IDPROM_CPUTYPE  CPU Type on RM400
+ */
+#ifdef CONFIG_CPU_BIG_ENDIAN
+#define __SNI_END 0
+#endif
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+#define __SNI_END 3
+#endif
+#define SNI_IDPROM_BASE        CKSEG1ADDR(0x1ff00000)
+#define SNI_IDPROM_MEMSIZE     (SNI_IDPROM_BASE + (0x28 ^ __SNI_END))
+#define SNI_IDPROM_BRDTYPE     (SNI_IDPROM_BASE + (0x29 ^ __SNI_END))
+#define SNI_IDPROM_CPUTYPE     (SNI_IDPROM_BASE + (0x30 ^ __SNI_END))
+
+#define SNI_IDPROM_SIZE	0x1000
 
 /* board specific init functions */
 extern void sni_a20r_init(void);
@@ -207,6 +227,9 @@ extern void sni_pcimt_irq_init(void);
 /* timer inits */
 extern void sni_cpu_time_init(void);
 
+/* eisa init for RM200/400 */
+extern int sni_eisa_root_init(void);
+
 /* common irq stuff */
 extern void (*sni_hwint)(void);
 extern struct irqaction sni_isa_irq;
-- 
cgit v1.2.3


From ae0c69acd61918cecd8e2949d9575271bc2fdfff Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 29 Jan 2008 10:15:00 +0000
Subject: [MIPS] fixmap: delete unused __set_fixmap, set_fixmap and
 set_fixmap_nocache

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/fixmap.h | 10 ----------
 1 file changed, 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/fixmap.h b/include/asm-mips/fixmap.h
index f27b96cfac2e..9cc8522a394f 100644
--- a/include/asm-mips/fixmap.h
+++ b/include/asm-mips/fixmap.h
@@ -60,16 +60,6 @@ enum fixed_addresses {
 	__end_of_fixed_addresses
 };
 
-extern void __set_fixmap(enum fixed_addresses idx,
-			 unsigned long phys, pgprot_t flags);
-
-#define set_fixmap(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL)
-/*
- * Some hardware wants to get fixmapped without caching.
- */
-#define set_fixmap_nocache(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
 /*
  * used by vmalloc.c.
  *
-- 
cgit v1.2.3


From c9662341f8a09cffd7234a89cc066c342fe343bf Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Sun, 9 Dec 2007 21:19:36 +0900
Subject: [MIPS] time: remove unused mips_timer_state()

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/dec/time.c    | 1 -
 arch/mips/kernel/time.c | 2 --
 include/asm-mips/time.h | 9 +--------
 3 files changed, 1 insertion(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/mips/dec/time.c b/arch/mips/dec/time.c
index 820e5331205f..60349062595a 100644
--- a/arch/mips/dec/time.c
+++ b/arch/mips/dec/time.c
@@ -161,7 +161,6 @@ static cycle_t dec_ioasic_hpt_read(void)
 
 void __init plat_time_init(void)
 {
-	mips_timer_state = dec_timer_state;
 	mips_timer_ack = dec_timer_ack;
 
 	if (!cpu_has_counter && IOASIC)
diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
index 2995be1ab3ca..9f85d4cecc5b 100644
--- a/arch/mips/kernel/time.c
+++ b/arch/mips/kernel/time.c
@@ -50,8 +50,6 @@ int update_persistent_clock(struct timespec now)
 	return rtc_mips_set_mmss(now.tv_sec);
 }
 
-int (*mips_timer_state)(void);
-
 int null_perf_irq(void)
 {
 	return 0;
diff --git a/include/asm-mips/time.h b/include/asm-mips/time.h
index 7717934f94c3..a8fd16e1981f 100644
--- a/include/asm-mips/time.h
+++ b/include/asm-mips/time.h
@@ -30,12 +30,6 @@ extern spinlock_t rtc_lock;
 extern int rtc_mips_set_time(unsigned long);
 extern int rtc_mips_set_mmss(unsigned long);
 
-/*
- * Timer interrupt functions.
- * mips_timer_state is needed for high precision timer calibration.
- */
-extern int (*mips_timer_state)(void);
-
 /*
  * board specific routines required by time_init().
  */
@@ -43,8 +37,7 @@ extern void plat_time_init(void);
 
 /*
  * mips_hpt_frequency - must be set if you intend to use an R4k-compatible
- * counter as a timer interrupt source; otherwise it can be set up
- * automagically with an aid of mips_timer_state.
+ * counter as a timer interrupt source.
  */
 extern unsigned int mips_hpt_frequency;
 
-- 
cgit v1.2.3


From 71466dd477c7a67d188d1d8310c7016a88531a04 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 12 Dec 2007 22:20:19 +0900
Subject: [MIPS] remove unneeded button check for reset

Removed unneeded button check for reset.
Because, the Cobalt has power switch.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/cobalt/reset.c              | 21 ++-------------------
 include/asm-mips/mach-cobalt/cobalt.h | 15 +--------------
 2 files changed, 3 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/arch/mips/cobalt/reset.c b/arch/mips/cobalt/reset.c
index 71eb4ccc4bc1..c3d1baa5bcf1 100644
--- a/arch/mips/cobalt/reset.c
+++ b/arch/mips/cobalt/reset.c
@@ -10,7 +10,6 @@
  */
 #include <linux/init.h>
 #include <linux/io.h>
-#include <linux/jiffies.h>
 #include <linux/leds.h>
 
 #include <cobalt.h>
@@ -29,29 +28,13 @@ device_initcall(ledtrig_power_off_init);
 
 void cobalt_machine_halt(void)
 {
-	int state, last, diff;
-	unsigned long mark;
-
 	/*
 	 * turn on power off LED on RaQ
-	 *
-	 * restart if ENTER and SELECT are pressed
 	 */
-
-	last = COBALT_KEY_PORT;
-
 	led_trigger_event(power_off_led_trigger, LED_FULL);
 
-	for (state = 0;;) {
-		diff = COBALT_KEY_PORT ^ last;
-		last ^= diff;
-
-		if((diff & (COBALT_KEY_ENTER | COBALT_KEY_SELECT)) && !(~last & (COBALT_KEY_ENTER | COBALT_KEY_SELECT)))
-			writeb(RESET, RESET_PORT);
-
-		for (mark = jiffies; jiffies - mark < HZ;)
-			;
-	}
+	local_irq_disable();
+	while (1) ;
 }
 
 void cobalt_machine_restart(char *command)
diff --git a/include/asm-mips/mach-cobalt/cobalt.h b/include/asm-mips/mach-cobalt/cobalt.h
index a79e7caf3a86..5b9fce73f11d 100644
--- a/include/asm-mips/mach-cobalt/cobalt.h
+++ b/include/asm-mips/mach-cobalt/cobalt.h
@@ -1,5 +1,5 @@
 /*
- * Lowlevel hardware stuff for the MIPS based Cobalt microservers.
+ * The Cobalt board ID information.
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -12,9 +12,6 @@
 #ifndef __ASM_COBALT_H
 #define __ASM_COBALT_H
 
-/*
- * The Cobalt board ID information.
- */
 extern int cobalt_board_id;
 
 #define COBALT_BRD_ID_QUBE1    0x3
@@ -22,14 +19,4 @@ extern int cobalt_board_id;
 #define COBALT_BRD_ID_QUBE2    0x5
 #define COBALT_BRD_ID_RAQ2     0x6
 
-#define COBALT_KEY_PORT		((~*(volatile unsigned int *) CKSEG1ADDR(0x1d000000) >> 24) & COBALT_KEY_MASK)
-# define COBALT_KEY_CLEAR	(1 << 1)
-# define COBALT_KEY_LEFT	(1 << 2)
-# define COBALT_KEY_UP		(1 << 3)
-# define COBALT_KEY_DOWN	(1 << 4)
-# define COBALT_KEY_RIGHT	(1 << 5)
-# define COBALT_KEY_ENTER	(1 << 6)
-# define COBALT_KEY_SELECT	(1 << 7)
-# define COBALT_KEY_MASK	0xfe
-
 #endif /* __ASM_COBALT_H */
-- 
cgit v1.2.3


From 2d47c58efadd33b8399f3f440b7e9cf93692f262 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 29 Jan 2008 10:15:01 +0000
Subject: [MIPS] Delete CONFIG_MSP_FPGA

Originally noticed by Jiri Olsa <olsajiri@gmail.com>.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/pmc-sierra/msp71xx/msp_regs.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/pmc-sierra/msp71xx/msp_regs.h b/include/asm-mips/pmc-sierra/msp71xx/msp_regs.h
index 0b56f55206c6..603eb737b4a8 100644
--- a/include/asm-mips/pmc-sierra/msp71xx/msp_regs.h
+++ b/include/asm-mips/pmc-sierra/msp71xx/msp_regs.h
@@ -585,11 +585,7 @@
  * UART defines                                                            *
  ***************************************************************************
  */
-#ifndef CONFIG_MSP_FPGA
 #define MSP_BASE_BAUD		25000000
-#else
-#define MSP_BASE_BAUD		6000000
-#endif
 #define MSP_UART_REG_LEN	0x20
 
 /*
-- 
cgit v1.2.3


From 302922e5f6901eb6f29c58539631f71b3d9746b8 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 29 Jan 2008 10:15:02 +0000
Subject: [MIPS] Qemu: Remove platform.

The Qemu platform was originally implemented to have an easily supportable
platform until Qemu reaches a state where it emulates a real world system.
Since the latest release Qemu is capable of emulating the MIPSsim and
Malta platforms, so this goal has been reached.  The Qemu plaform is also
rather underfeatured so less useful than a Malta emulation.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig                                  | 28 ---------
 arch/mips/Makefile                                 |  8 ---
 arch/mips/configs/atlas_defconfig                  |  1 -
 arch/mips/configs/bigsur_defconfig                 |  1 -
 arch/mips/configs/capcella_defconfig               |  1 -
 arch/mips/configs/cobalt_defconfig                 |  1 -
 arch/mips/configs/db1000_defconfig                 |  1 -
 arch/mips/configs/db1100_defconfig                 |  1 -
 arch/mips/configs/db1200_defconfig                 |  1 -
 arch/mips/configs/db1500_defconfig                 |  1 -
 arch/mips/configs/db1550_defconfig                 |  1 -
 arch/mips/configs/decstation_defconfig             |  1 -
 arch/mips/configs/e55_defconfig                    |  1 -
 arch/mips/configs/emma2rh_defconfig                |  1 -
 arch/mips/configs/excite_defconfig                 |  1 -
 arch/mips/configs/fulong_defconfig                 |  1 -
 arch/mips/configs/ip22_defconfig                   |  1 -
 arch/mips/configs/ip27_defconfig                   |  1 -
 arch/mips/configs/ip32_defconfig                   |  1 -
 arch/mips/configs/jazz_defconfig                   |  1 -
 arch/mips/configs/jmr3927_defconfig                |  1 -
 arch/mips/configs/lasat_defconfig                  |  1 -
 arch/mips/configs/malta_defconfig                  |  1 -
 arch/mips/configs/mipssim_defconfig                |  1 -
 arch/mips/configs/mpc30x_defconfig                 |  1 -
 arch/mips/configs/msp71xx_defconfig                |  1 -
 arch/mips/configs/mtx1_defconfig                   |  1 -
 arch/mips/configs/pb1100_defconfig                 |  1 -
 arch/mips/configs/pb1500_defconfig                 |  1 -
 arch/mips/configs/pb1550_defconfig                 |  1 -
 arch/mips/configs/pnx8550-jbs_defconfig            |  1 -
 arch/mips/configs/pnx8550-stb810_defconfig         |  1 -
 arch/mips/configs/qemu_defconfig                   |  1 -
 arch/mips/configs/rbhma4200_defconfig              |  1 -
 arch/mips/configs/rbhma4500_defconfig              |  1 -
 arch/mips/configs/rm200_defconfig                  |  1 -
 arch/mips/configs/sb1250-swarm_defconfig           |  1 -
 arch/mips/configs/sead_defconfig                   |  1 -
 arch/mips/configs/tb0219_defconfig                 |  1 -
 arch/mips/configs/tb0226_defconfig                 |  1 -
 arch/mips/configs/tb0287_defconfig                 |  1 -
 arch/mips/configs/workpad_defconfig                |  1 -
 arch/mips/configs/wrppmc_defconfig                 |  1 -
 arch/mips/configs/yosemite_defconfig               |  1 -
 arch/mips/defconfig                                |  1 -
 arch/mips/qemu/Makefile                            | 10 ----
 arch/mips/qemu/q-console.c                         | 26 ---------
 arch/mips/qemu/q-firmware.c                        | 24 --------
 arch/mips/qemu/q-irq.c                             | 37 ------------
 arch/mips/qemu/q-mem.c                             |  5 --
 arch/mips/qemu/q-reset.c                           | 33 -----------
 arch/mips/qemu/q-setup.c                           | 37 ------------
 arch/mips/qemu/q-smp.c                             | 67 ----------------------
 include/asm-mips/mach-qemu/cpu-feature-overrides.h | 32 -----------
 include/asm-mips/mach-qemu/war.h                   | 25 --------
 55 files changed, 375 deletions(-)
 delete mode 100644 arch/mips/qemu/Makefile
 delete mode 100644 arch/mips/qemu/q-console.c
 delete mode 100644 arch/mips/qemu/q-firmware.c
 delete mode 100644 arch/mips/qemu/q-irq.c
 delete mode 100644 arch/mips/qemu/q-mem.c
 delete mode 100644 arch/mips/qemu/q-reset.c
 delete mode 100644 arch/mips/qemu/q-setup.c
 delete mode 100644 arch/mips/qemu/q-smp.c
 delete mode 100644 include/asm-mips/mach-qemu/cpu-feature-overrides.h
 delete mode 100644 include/asm-mips/mach-qemu/war.h

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index ff9cc18d456a..a97bee108d3c 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -359,34 +359,6 @@ config PMC_YOSEMITE
 	  Yosemite is an evaluation board for the RM9000x2 processor
 	  manufactured by PMC-Sierra.
 
-config QEMU
-	bool "Qemu"
-	select CEVT_R4K
-	select CSRC_R4K
-	select DMA_COHERENT
-	select GENERIC_ISA_DMA
-	select HAVE_STD_PC_SERIAL_PORT
-	select I8253
-	select I8259
-	select IRQ_CPU
-	select ISA
-	select SWAP_IO_SPACE
-	select SYS_HAS_CPU_MIPS32_R1
-	select SYS_HAS_EARLY_PRINTK
-	select SYS_SUPPORTS_32BIT_KERNEL
-	select SYS_SUPPORTS_BIG_ENDIAN
-	select SYS_SUPPORTS_LITTLE_ENDIAN
-	select GENERIC_HARDIRQS_NO__DO_IRQ
-	select NR_CPUS_DEFAULT_1
-	select SYS_SUPPORTS_SMP
-	help
-	  Qemu is a software emulator which among other architectures also
-	  can simulate a MIPS32 4Kc system.  This patch adds support for the
-	  system architecture that currently is being simulated by Qemu.  It
-	  will eventually be removed again when Qemu has the capability to
-	  simulate actual MIPS hardware platforms.  More information on Qemu
-	  can be found at http://www.linux-mips.org/wiki/Qemu.
-
 config SGI_IP22
 	bool "SGI IP22 (Indy/Indigo2)"
 	select ARC
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index b96af0669513..3fb7f3065c92 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -354,14 +354,6 @@ core-$(CONFIG_PMC_YOSEMITE)	+= arch/mips/pmc-sierra/yosemite/
 cflags-$(CONFIG_PMC_YOSEMITE)	+= -Iinclude/asm-mips/mach-yosemite
 load-$(CONFIG_PMC_YOSEMITE)	+= 0xffffffff80100000
 
-#
-# Qemu simulating MIPS32 4Kc
-#
-core-$(CONFIG_QEMU)		+= arch/mips/qemu/
-cflags-$(CONFIG_QEMU)		+= -Iinclude/asm-mips/mach-qemu
-load-$(CONFIG_QEMU)		+= 0xffffffff80010000
-all-$(CONFIG_QEMU)		:= vmlinux.bin
-
 #
 # Basler eXcite
 #
diff --git a/arch/mips/configs/atlas_defconfig b/arch/mips/configs/atlas_defconfig
index c28db00fe469..3443f6cd57bb 100644
--- a/arch/mips/configs/atlas_defconfig
+++ b/arch/mips/configs/atlas_defconfig
@@ -37,7 +37,6 @@ CONFIG_MIPS_ATLAS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig
index 4bb97a08c9e6..abf70d74e9d7 100644
--- a/arch/mips/configs/bigsur_defconfig
+++ b/arch/mips/configs/bigsur_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/capcella_defconfig b/arch/mips/configs/capcella_defconfig
index 128908a542db..a94f14b5c8fa 100644
--- a/arch/mips/configs/capcella_defconfig
+++ b/arch/mips/configs/capcella_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/cobalt_defconfig b/arch/mips/configs/cobalt_defconfig
index cb56ebc760e9..b7295e988381 100644
--- a/arch/mips/configs/cobalt_defconfig
+++ b/arch/mips/configs/cobalt_defconfig
@@ -24,7 +24,6 @@ CONFIG_MIPS_COBALT=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/db1000_defconfig b/arch/mips/configs/db1000_defconfig
index 184f119ff29d..36578968d386 100644
--- a/arch/mips/configs/db1000_defconfig
+++ b/arch/mips/configs/db1000_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_DB1000=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/db1100_defconfig b/arch/mips/configs/db1100_defconfig
index e61dbee45560..5a90740c363a 100644
--- a/arch/mips/configs/db1100_defconfig
+++ b/arch/mips/configs/db1100_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_DB1100=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/db1200_defconfig b/arch/mips/configs/db1200_defconfig
index 508faf44ca72..76f37a1159fe 100644
--- a/arch/mips/configs/db1200_defconfig
+++ b/arch/mips/configs/db1200_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_DB1200=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/db1500_defconfig b/arch/mips/configs/db1500_defconfig
index 5abadf34924d..508c91944f30 100644
--- a/arch/mips/configs/db1500_defconfig
+++ b/arch/mips/configs/db1500_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_DB1500=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/db1550_defconfig b/arch/mips/configs/db1550_defconfig
index 9ccdcf0848a6..0c2c70d21db9 100644
--- a/arch/mips/configs/db1550_defconfig
+++ b/arch/mips/configs/db1550_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_DB1550=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/decstation_defconfig b/arch/mips/configs/decstation_defconfig
index b661cd61ea51..58c2cd68c3a7 100644
--- a/arch/mips/configs/decstation_defconfig
+++ b/arch/mips/configs/decstation_defconfig
@@ -37,7 +37,6 @@ CONFIG_MACH_DECSTATION=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/e55_defconfig b/arch/mips/configs/e55_defconfig
index 93d19a575008..90d81f5dcebc 100644
--- a/arch/mips/configs/e55_defconfig
+++ b/arch/mips/configs/e55_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/emma2rh_defconfig b/arch/mips/configs/emma2rh_defconfig
index f931bcb73ae6..f9a003c2b3a1 100644
--- a/arch/mips/configs/emma2rh_defconfig
+++ b/arch/mips/configs/emma2rh_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 CONFIG_MARKEINS=y
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/excite_defconfig b/arch/mips/configs/excite_defconfig
index dc3058645acc..15efacc75d73 100644
--- a/arch/mips/configs/excite_defconfig
+++ b/arch/mips/configs/excite_defconfig
@@ -38,7 +38,6 @@ CONFIG_BASLER_EXCITE=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/fulong_defconfig b/arch/mips/configs/fulong_defconfig
index d89eeb006c67..5887a1735fba 100644
--- a/arch/mips/configs/fulong_defconfig
+++ b/arch/mips/configs/fulong_defconfig
@@ -23,7 +23,6 @@ CONFIG_LEMOTE_FULONG=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index a6957a66838d..4f5e56c9335e 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -25,7 +25,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 CONFIG_SGI_IP22=y
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 8feb16a55b56..f40e437bd9e5 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -24,7 +24,6 @@ CONFIG_MIPS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 CONFIG_SGI_IP27=y
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/ip32_defconfig b/arch/mips/configs/ip32_defconfig
index ad547c7c6f81..2c5c624c5d42 100644
--- a/arch/mips/configs/ip32_defconfig
+++ b/arch/mips/configs/ip32_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/jazz_defconfig b/arch/mips/configs/jazz_defconfig
index 500daa9c3d82..56148745e8f2 100644
--- a/arch/mips/configs/jazz_defconfig
+++ b/arch/mips/configs/jazz_defconfig
@@ -37,7 +37,6 @@ CONFIG_MACH_JAZZ=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/jmr3927_defconfig b/arch/mips/configs/jmr3927_defconfig
index 8dbfa19496cb..a7cd67753aac 100644
--- a/arch/mips/configs/jmr3927_defconfig
+++ b/arch/mips/configs/jmr3927_defconfig
@@ -24,7 +24,6 @@ CONFIG_MIPS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/lasat_defconfig b/arch/mips/configs/lasat_defconfig
index aa67a3baf53f..e6aef999854c 100644
--- a/arch/mips/configs/lasat_defconfig
+++ b/arch/mips/configs/lasat_defconfig
@@ -25,7 +25,6 @@ CONFIG_LASAT=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/malta_defconfig b/arch/mips/configs/malta_defconfig
index 9e32b029501e..3d0da952811c 100644
--- a/arch/mips/configs/malta_defconfig
+++ b/arch/mips/configs/malta_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_MALTA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/mipssim_defconfig b/arch/mips/configs/mipssim_defconfig
index 5a7b4ed76906..6db0bdaefb27 100644
--- a/arch/mips/configs/mipssim_defconfig
+++ b/arch/mips/configs/mipssim_defconfig
@@ -26,7 +26,6 @@ CONFIG_MIPS_SIM=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/mpc30x_defconfig b/arch/mips/configs/mpc30x_defconfig
index c2f7ce9ff603..27e23fc9363a 100644
--- a/arch/mips/configs/mpc30x_defconfig
+++ b/arch/mips/configs/mpc30x_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/msp71xx_defconfig b/arch/mips/configs/msp71xx_defconfig
index af4a41e06e87..b12b73f6d74f 100644
--- a/arch/mips/configs/msp71xx_defconfig
+++ b/arch/mips/configs/msp71xx_defconfig
@@ -38,7 +38,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_MACH_VR41XX is not set
 CONFIG_PMC_MSP=y
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/mtx1_defconfig b/arch/mips/configs/mtx1_defconfig
index 3a34208600d1..fa3aa3919448 100644
--- a/arch/mips/configs/mtx1_defconfig
+++ b/arch/mips/configs/mtx1_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_ALCHEMY=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/pb1100_defconfig b/arch/mips/configs/pb1100_defconfig
index 2d394578b969..1d0157d3a5bb 100644
--- a/arch/mips/configs/pb1100_defconfig
+++ b/arch/mips/configs/pb1100_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_PB1100=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/pb1500_defconfig b/arch/mips/configs/pb1500_defconfig
index 0d645621ce79..d0491a05ee58 100644
--- a/arch/mips/configs/pb1500_defconfig
+++ b/arch/mips/configs/pb1500_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_PB1500=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/pb1550_defconfig b/arch/mips/configs/pb1550_defconfig
index 210a06543116..16d78d3cd2aa 100644
--- a/arch/mips/configs/pb1550_defconfig
+++ b/arch/mips/configs/pb1550_defconfig
@@ -38,7 +38,6 @@ CONFIG_MIPS_PB1550=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/pnx8550-jbs_defconfig b/arch/mips/configs/pnx8550-jbs_defconfig
index bee11ebed8ec..518a60892b78 100644
--- a/arch/mips/configs/pnx8550-jbs_defconfig
+++ b/arch/mips/configs/pnx8550-jbs_defconfig
@@ -37,7 +37,6 @@ CONFIG_PNX8550_JBS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/pnx8550-stb810_defconfig b/arch/mips/configs/pnx8550-stb810_defconfig
index b088a5ce268f..68351eb81bc8 100644
--- a/arch/mips/configs/pnx8550-stb810_defconfig
+++ b/arch/mips/configs/pnx8550-stb810_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 CONFIG_PNX8550_STB810=y
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/qemu_defconfig b/arch/mips/configs/qemu_defconfig
index ec1a6a796599..72ca147f9422 100644
--- a/arch/mips/configs/qemu_defconfig
+++ b/arch/mips/configs/qemu_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-CONFIG_QEMU=y
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/rbhma4200_defconfig b/arch/mips/configs/rbhma4200_defconfig
index 16494d0942b8..470f6f4d3ea2 100644
--- a/arch/mips/configs/rbhma4200_defconfig
+++ b/arch/mips/configs/rbhma4200_defconfig
@@ -24,7 +24,6 @@ CONFIG_MIPS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/rbhma4500_defconfig b/arch/mips/configs/rbhma4500_defconfig
index aa9e0cf66143..5a39f56b175e 100644
--- a/arch/mips/configs/rbhma4500_defconfig
+++ b/arch/mips/configs/rbhma4500_defconfig
@@ -24,7 +24,6 @@ CONFIG_MIPS=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/rm200_defconfig b/arch/mips/configs/rm200_defconfig
index 7aaa6d6513f1..56371b860eb0 100644
--- a/arch/mips/configs/rm200_defconfig
+++ b/arch/mips/configs/rm200_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/sb1250-swarm_defconfig b/arch/mips/configs/sb1250-swarm_defconfig
index 4fe6a0b5c357..117470b60e34 100644
--- a/arch/mips/configs/sb1250-swarm_defconfig
+++ b/arch/mips/configs/sb1250-swarm_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/sead_defconfig b/arch/mips/configs/sead_defconfig
index aa5733779014..3ee75b15c0b0 100644
--- a/arch/mips/configs/sead_defconfig
+++ b/arch/mips/configs/sead_defconfig
@@ -37,7 +37,6 @@ CONFIG_MIPS_SEAD=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/tb0219_defconfig b/arch/mips/configs/tb0219_defconfig
index 0ae553923f98..af82e1a1823c 100644
--- a/arch/mips/configs/tb0219_defconfig
+++ b/arch/mips/configs/tb0219_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/tb0226_defconfig b/arch/mips/configs/tb0226_defconfig
index e1f22023969d..a95385b24546 100644
--- a/arch/mips/configs/tb0226_defconfig
+++ b/arch/mips/configs/tb0226_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/tb0287_defconfig b/arch/mips/configs/tb0287_defconfig
index 790e62bc0860..40d4a40a970e 100644
--- a/arch/mips/configs/tb0287_defconfig
+++ b/arch/mips/configs/tb0287_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/workpad_defconfig b/arch/mips/configs/workpad_defconfig
index 1412c47517ff..edf90b321fe6 100644
--- a/arch/mips/configs/workpad_defconfig
+++ b/arch/mips/configs/workpad_defconfig
@@ -24,7 +24,6 @@ CONFIG_MACH_VR41XX=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/configs/wrppmc_defconfig b/arch/mips/configs/wrppmc_defconfig
index bbfeeeacc486..2e3c683b2052 100644
--- a/arch/mips/configs/wrppmc_defconfig
+++ b/arch/mips/configs/wrppmc_defconfig
@@ -37,7 +37,6 @@ CONFIG_WR_PPMC=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/configs/yosemite_defconfig b/arch/mips/configs/yosemite_defconfig
index 980dfb6417e9..b6178ffbc523 100644
--- a/arch/mips/configs/yosemite_defconfig
+++ b/arch/mips/configs/yosemite_defconfig
@@ -37,7 +37,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_MACH_VR41XX is not set
 CONFIG_PMC_YOSEMITE=y
-# CONFIG_QEMU is not set
 # CONFIG_MARKEINS is not set
 # CONFIG_SGI_IP22 is not set
 # CONFIG_SGI_IP27 is not set
diff --git a/arch/mips/defconfig b/arch/mips/defconfig
index a6957a66838d..4f5e56c9335e 100644
--- a/arch/mips/defconfig
+++ b/arch/mips/defconfig
@@ -25,7 +25,6 @@ CONFIG_ZONE_DMA=y
 # CONFIG_PNX8550_STB810 is not set
 # CONFIG_PMC_MSP is not set
 # CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
 CONFIG_SGI_IP22=y
 # CONFIG_SGI_IP27 is not set
 # CONFIG_SGI_IP32 is not set
diff --git a/arch/mips/qemu/Makefile b/arch/mips/qemu/Makefile
deleted file mode 100644
index 2ba4ef34b4a7..000000000000
--- a/arch/mips/qemu/Makefile
+++ /dev/null
@@ -1,10 +0,0 @@
-#
-# Makefile for Qemu specific kernel interface routines under Linux.
-#
-
-obj-y		= q-firmware.o q-irq.o q-mem.o q-setup.o q-reset.o
-
-obj-$(CONFIG_EARLY_PRINTK)	+= q-console.o
-obj-$(CONFIG_SMP)		+= q-smp.o
-
-EXTRA_CFLAGS += -Werror
diff --git a/arch/mips/qemu/q-console.c b/arch/mips/qemu/q-console.c
deleted file mode 100644
index 81101ae5017a..000000000000
--- a/arch/mips/qemu/q-console.c
+++ /dev/null
@@ -1,26 +0,0 @@
-#include <linux/console.h>
-#include <linux/init.h>
-#include <linux/serial_reg.h>
-#include <asm/io.h>
-
-#define PORT(offset) (0x3f8 + (offset))
-
-static inline unsigned int serial_in(int offset)
-{
-	return inb(PORT(offset));
-}
-
-static inline void serial_out(int offset, int value)
-{
-	outb(value, PORT(offset));
-}
-
-int prom_putchar(char c)
-{
-	while ((serial_in(UART_LSR) & UART_LSR_THRE) == 0)
-		;
-
-	serial_out(UART_TX, c);
-
-	return 1;
-}
diff --git a/arch/mips/qemu/q-firmware.c b/arch/mips/qemu/q-firmware.c
deleted file mode 100644
index 3ed43f416cd1..000000000000
--- a/arch/mips/qemu/q-firmware.c
+++ /dev/null
@@ -1,24 +0,0 @@
-#include <linux/init.h>
-#include <linux/string.h>
-#include <asm/addrspace.h>
-#include <asm/bootinfo.h>
-#include <asm/io.h>
-
-#define QEMU_PORT_BASE 0xb4000000
-
-void __init prom_init(void)
-{
-	int *cmdline;
-
-	cmdline = (int *) (CKSEG0 + (0x10 << 20) - 260);
-	if (*cmdline == 0x12345678) {
-		if (*(char *)(cmdline + 1))
-			strcpy(arcs_cmdline, (char *)(cmdline + 1));
-		add_memory_region(0x0<<20, cmdline[-1], BOOT_MEM_RAM);
-	} else {
-		add_memory_region(0x0<<20, 0x10<<20, BOOT_MEM_RAM);
-	}
-
-
-	set_io_port_base(QEMU_PORT_BASE);
-}
diff --git a/arch/mips/qemu/q-irq.c b/arch/mips/qemu/q-irq.c
deleted file mode 100644
index 7df36dbe65c7..000000000000
--- a/arch/mips/qemu/q-irq.c
+++ /dev/null
@@ -1,37 +0,0 @@
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/linkage.h>
-
-#include <asm/i8259.h>
-#include <asm/irq_cpu.h>
-#include <asm/mipsregs.h>
-#include <asm/qemu.h>
-#include <asm/system.h>
-#include <asm/time.h>
-
-asmlinkage void plat_irq_dispatch(void)
-{
-	unsigned int pending = read_c0_status() & read_c0_cause();
-
-	if (pending & 0x8000) {
-		do_IRQ(Q_COUNT_COMPARE_IRQ);
-		return;
-	}
-	if (pending & 0x0400) {
-		int irq = i8259_irq();
-
-		if (likely(irq >= 0))
-			do_IRQ(irq);
-
-		return;
-	}
-}
-
-void __init arch_init_irq(void)
-{
-	mips_hpt_frequency = QEMU_C0_COUNTER_CLOCK;		/* 100MHz */
-
-	mips_cpu_irq_init();
-	init_i8259_irqs();
-	set_c0_status(0x400);
-}
diff --git a/arch/mips/qemu/q-mem.c b/arch/mips/qemu/q-mem.c
deleted file mode 100644
index dae39b59de15..000000000000
--- a/arch/mips/qemu/q-mem.c
+++ /dev/null
@@ -1,5 +0,0 @@
-#include <linux/init.h>
-
-void __init prom_free_prom_memory(void)
-{
-}
diff --git a/arch/mips/qemu/q-reset.c b/arch/mips/qemu/q-reset.c
deleted file mode 100644
index dbbe44ad7e89..000000000000
--- a/arch/mips/qemu/q-reset.c
+++ /dev/null
@@ -1,33 +0,0 @@
-
-#include <asm/io.h>
-#include <asm/reboot.h>
-#include <asm/cacheflush.h>
-#include <asm/qemu.h>
-
-static void qemu_machine_restart(char *command)
-{
-	volatile unsigned int *reg = (unsigned int *)QEMU_RESTART_REG;
-
-	set_c0_status(ST0_BEV | ST0_ERL);
-	change_c0_config(CONF_CM_CMASK, CONF_CM_UNCACHED);
-	flush_cache_all();
-	write_c0_wired(0);
-	*reg = 42;
-	while (1)
-		cpu_wait();
-}
-
-static void qemu_machine_halt(void)
-{
-	volatile unsigned int *reg = (unsigned int *)QEMU_HALT_REG;
-
-	*reg = 42;
-	while (1)
-		cpu_wait();
-}
-
-void qemu_reboot_setup(void)
-{
-	_machine_restart = qemu_machine_restart;
-	_machine_halt = qemu_machine_halt;
-}
diff --git a/arch/mips/qemu/q-setup.c b/arch/mips/qemu/q-setup.c
deleted file mode 100644
index 9f0c2d3c67ea..000000000000
--- a/arch/mips/qemu/q-setup.c
+++ /dev/null
@@ -1,37 +0,0 @@
-#include <linux/init.h>
-#include <linux/platform_device.h>
-
-#include <asm/i8253.h>
-#include <asm/io.h>
-#include <asm/time.h>
-
-extern void qemu_reboot_setup(void);
-
-const char *get_system_type(void)
-{
-	return "Qemu";
-}
-
-void __init plat_time_init(void)
-{
-	setup_pit_timer();
-}
-
-void __init plat_mem_setup(void)
-{
-	qemu_reboot_setup();
-}
-
-static struct platform_device pcspeaker_pdev = {
-	.name	= "pcspkr",
-	.id	= -1,
-};
-
-static int __init qemu_platform_devinit(void)
-{
-	platform_device_register(&pcspeaker_pdev);
-
-	return 0;
-}
-
-device_initcall(qemu_platform_devinit);
diff --git a/arch/mips/qemu/q-smp.c b/arch/mips/qemu/q-smp.c
deleted file mode 100644
index ead6c30eeb14..000000000000
--- a/arch/mips/qemu/q-smp.c
+++ /dev/null
@@ -1,67 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2006, 07 by Ralf Baechle (ralf@linux-mips.org)
- *
- * Symmetric Uniprocessor (TM) Support
- */
-#include <linux/kernel.h>
-#include <linux/sched.h>
-
-/*
- * Send inter-processor interrupt
- */
-void up_send_ipi_single(int cpu, unsigned int action)
-{
-	panic(KERN_ERR "%s called", __func__);
-}
-
-static inline void up_send_ipi_mask(cpumask_t mask, unsigned int action)
-{
-	panic(KERN_ERR "%s called", __func__);
-}
-
-/*
- *  After we've done initial boot, this function is called to allow the
- *  board code to clean up state, if needed
- */
-void __cpuinit up_init_secondary(void)
-{
-}
-
-void __cpuinit up_smp_finish(void)
-{
-}
-
-/* Hook for after all CPUs are online */
-void up_cpus_done(void)
-{
-}
-
-/*
- * Firmware CPU startup hook
- */
-void __cpuinit up_boot_secondary(int cpu, struct task_struct *idle)
-{
-}
-
-void __init up_smp_setup(void)
-{
-}
-
-void __init up_prepare_cpus(unsigned int max_cpus)
-{
-}
-
-struct plat_smp_ops up_smp_ops = {
-	.send_ipi_single	= up_send_ipi_single,
-	.send_ipi_mask		= up_send_ipi_mask,
-	.init_secondary		= up_init_secondary,
-	.smp_finish		= up_smp_finish,
-	.cpus_done		= up_cpus_done,
-	.boot_secondary		= up_boot_secondary,
-	.smp_setup		= up_smp_setup,
-	.prepare_cpus		= up_prepare_cpus,
-};
diff --git a/include/asm-mips/mach-qemu/cpu-feature-overrides.h b/include/asm-mips/mach-qemu/cpu-feature-overrides.h
deleted file mode 100644
index d2daaed235d5..000000000000
--- a/include/asm-mips/mach-qemu/cpu-feature-overrides.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2003, 07 Ralf Baechle
- */
-#ifndef __ASM_MACH_QEMU_CPU_FEATURE_OVERRIDES_H
-#define __ASM_MACH_QEMU_CPU_FEATURE_OVERRIDES_H
-
-/*
- * QEMU only comes with a hazard-free MIPS32 processor, so things are easy.
- */
-#define cpu_has_mips16		0
-#define cpu_has_divec		0
-#define cpu_has_cache_cdex_p	0
-#define cpu_has_prefetch	0
-#define cpu_has_mcheck		0
-#define cpu_has_ejtag		0
-
-#define cpu_has_llsc		1
-#define cpu_has_vtag_icache	0
-#define cpu_has_dc_aliases	0
-#define cpu_has_ic_fills_f_dc	0
-
-#define cpu_has_dsp		0
-#define cpu_has_mipsmt		0
-
-#define cpu_has_nofpuex		0
-#define cpu_has_64bits		0
-
-#endif /* __ASM_MACH_QEMU_CPU_FEATURE_OVERRIDES_H */
diff --git a/include/asm-mips/mach-qemu/war.h b/include/asm-mips/mach-qemu/war.h
deleted file mode 100644
index 0eaf0c548a47..000000000000
--- a/include/asm-mips/mach-qemu/war.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2002, 2004, 2007 by Ralf Baechle <ralf@linux-mips.org>
- */
-#ifndef __ASM_MIPS_MACH_QEMU_WAR_H
-#define __ASM_MIPS_MACH_QEMU_WAR_H
-
-#define R4600_V1_INDEX_ICACHEOP_WAR	0
-#define R4600_V1_HIT_CACHEOP_WAR	0
-#define R4600_V2_HIT_CACHEOP_WAR	0
-#define R5432_CP0_INTERRUPT_WAR		0
-#define BCM1250_M3_WAR			0
-#define SIBYTE_1956_WAR			0
-#define MIPS4K_ICACHE_REFILL_WAR	0
-#define MIPS_CACHE_SYNC_WAR		0
-#define TX49XX_ICACHE_INDEX_INV_WAR	0
-#define RM9000_CDEX_SMP_WAR		0
-#define ICACHE_REFILLS_WORKAROUND_WAR	0
-#define R10000_LLSC_WAR			0
-#define MIPS34K_MISSED_ITLB_WAR		0
-
-#endif /* __ASM_MIPS_MACH_QEMU_WAR_H */
-- 
cgit v1.2.3


From 1997267072d6bfadb37d336779e6122c2d44343d Mon Sep 17 00:00:00 2001
From: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Date: Thu, 24 Jan 2008 19:52:48 +0300
Subject: [MIPS] Malta, Atlas, Sead: remove an extern from .c files

This patch moves the "extern" declaration for the function
mips_reboot_setup() from the board setup .c files to the
header file include/asm-mips/mips-boards/generic.h.

This fixes a warning produced by the checkpatch.pl script.

No functional changes introduced.

This was compile-tested by building the kernel for all
three boards affected by this change. All builds finished
successfully.

Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mips-boards/atlas/atlas_setup.c | 2 --
 arch/mips/mips-boards/malta/malta_setup.c | 2 --
 arch/mips/mips-boards/sead/sead_setup.c   | 2 --
 include/asm-mips/mips-boards/generic.h    | 2 ++
 4 files changed, 2 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/mips/mips-boards/atlas/atlas_setup.c b/arch/mips/mips-boards/atlas/atlas_setup.c
index e405d112a067..5ce800420012 100644
--- a/arch/mips/mips-boards/atlas/atlas_setup.c
+++ b/arch/mips/mips-boards/atlas/atlas_setup.c
@@ -34,8 +34,6 @@
 #include <asm/time.h>
 #include <asm/traps.h>
 
-extern void mips_reboot_setup(void);
-
 #ifdef CONFIG_KGDB
 extern void kgdb_config(void);
 #endif
diff --git a/arch/mips/mips-boards/malta/malta_setup.c b/arch/mips/mips-boards/malta/malta_setup.c
index 480521f05b8c..e5ac079e47e0 100644
--- a/arch/mips/mips-boards/malta/malta_setup.c
+++ b/arch/mips/mips-boards/malta/malta_setup.c
@@ -35,8 +35,6 @@
 #include <linux/console.h>
 #endif
 
-extern void mips_reboot_setup(void);
-
 #ifdef CONFIG_KGDB
 extern void kgdb_config(void);
 #endif
diff --git a/arch/mips/mips-boards/sead/sead_setup.c b/arch/mips/mips-boards/sead/sead_setup.c
index 1fb61b852304..8aa8e5b7b074 100644
--- a/arch/mips/mips-boards/sead/sead_setup.c
+++ b/arch/mips/mips-boards/sead/sead_setup.c
@@ -34,8 +34,6 @@
 #include <asm/mips-boards/seadint.h>
 #include <asm/time.h>
 
-extern void mips_reboot_setup(void);
-
 static void __init serial_init(void);
 
 const char *get_system_type(void)
diff --git a/include/asm-mips/mips-boards/generic.h b/include/asm-mips/mips-boards/generic.h
index d58977483534..2ca6bda37d1c 100644
--- a/include/asm-mips/mips-boards/generic.h
+++ b/include/asm-mips/mips-boards/generic.h
@@ -97,6 +97,8 @@ extern int mips_revision_corid;
 
 extern int mips_revision_sconid;
 
+extern void mips_reboot_setup(void);
+
 #ifdef CONFIG_PCI
 extern void mips_pcibios_init(void);
 #else
-- 
cgit v1.2.3


From bbdda5e1fe9c1a5b34a2b552910f5f4c0dc24485 Mon Sep 17 00:00:00 2001
From: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Date: Thu, 24 Jan 2008 19:52:53 +0300
Subject: [MIPS] Malta, Atlas: move an extern function declaration to the
 header file

This was compile-tested using default configs for the boards
affected by this change.

This patch does not introduce any functional changes.

Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mips-boards/atlas/atlas_setup.c | 4 ----
 arch/mips/mips-boards/malta/malta_setup.c | 4 ----
 include/asm-mips/mips-boards/generic.h    | 4 ++++
 3 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/mips/mips-boards/atlas/atlas_setup.c b/arch/mips/mips-boards/atlas/atlas_setup.c
index 5ce800420012..5c500802271e 100644
--- a/arch/mips/mips-boards/atlas/atlas_setup.c
+++ b/arch/mips/mips-boards/atlas/atlas_setup.c
@@ -34,10 +34,6 @@
 #include <asm/time.h>
 #include <asm/traps.h>
 
-#ifdef CONFIG_KGDB
-extern void kgdb_config(void);
-#endif
-
 static void __init serial_init(void);
 
 const char *get_system_type(void)
diff --git a/arch/mips/mips-boards/malta/malta_setup.c b/arch/mips/mips-boards/malta/malta_setup.c
index 3e2de0b8e2cf..8dacb6adb6db 100644
--- a/arch/mips/mips-boards/malta/malta_setup.c
+++ b/arch/mips/mips-boards/malta/malta_setup.c
@@ -35,10 +35,6 @@
 #include <linux/console.h>
 #endif
 
-#ifdef CONFIG_KGDB
-extern void kgdb_config(void);
-#endif
-
 struct resource standard_io_resources[] = {
 	{
 		.name = "dma1",
diff --git a/include/asm-mips/mips-boards/generic.h b/include/asm-mips/mips-boards/generic.h
index 2ca6bda37d1c..1c39d339521e 100644
--- a/include/asm-mips/mips-boards/generic.h
+++ b/include/asm-mips/mips-boards/generic.h
@@ -105,4 +105,8 @@ extern void mips_pcibios_init(void);
 #define mips_pcibios_init() do { } while (0)
 #endif
 
+#ifdef CONFIG_KGDB
+extern void kgdb_config(void);
+#endif
+
 #endif  /* __ASM_MIPS_BOARDS_GENERIC_H */
-- 
cgit v1.2.3


From 023ccde109b995bb99862bf9c87efd006b1d1885 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 29 Jan 2008 14:08:06 +0100
Subject: block: fix warning on compile with CONFIG_BLOCK

struct io_context was not defined, just add an empty forward decl.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 71e7a847dffc..e18d4192f6e8 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -825,6 +825,7 @@ static inline void exit_io_context(void)
 {
 }
 
+struct io_context;
 static inline int put_io_context(struct io_context *ioc)
 {
 	return 1;
-- 
cgit v1.2.3


From 7da975a231ae8c11593fd79d1083215321be213a Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Tue, 29 Jan 2008 19:12:06 +0100
Subject: Fix blktrace compile warning

request_queue_t is deprecated

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blktrace_api.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index 06dadba349ac..cfc3147e5cf9 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -282,10 +282,10 @@ static inline void blk_add_trace_remap(struct request_queue *q, struct bio *bio,
 	__blk_add_trace(bt, from, bio->bi_size, bio->bi_rw, BLK_TA_REMAP, !bio_flagged(bio, BIO_UPTODATE), sizeof(r), &r);
 }
 
-extern int blk_trace_setup(request_queue_t *q, char *name, dev_t dev,
+extern int blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
 			   char __user *arg);
-extern int blk_trace_startstop(request_queue_t *q, int start);
-extern int blk_trace_remove(request_queue_t *q);
+extern int blk_trace_startstop(struct request_queue *q, int start);
+extern int blk_trace_remove(struct request_queue *q);
 
 #else /* !CONFIG_BLK_DEV_IO_TRACE */
 #define blk_trace_ioctl(bdev, cmd, arg)		(-ENOTTY)
-- 
cgit v1.2.3


From 16efd45435fa695b501b7f73c3259bd7c77cc12c Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Jan 2008 08:37:59 -0500
Subject: NetLabel: Add secid token support to the NetLabel secattr struct

This patch adds support to the NetLabel LSM secattr struct for a secid token
and a type field, paving the way for full LSM/SELinux context support and
"static" or "fallback" labels.  In addition, this patch adds a fair amount
of documentation to the core NetLabel structures used as part of the
NetLabel kernel API.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/net/netlabel.h            | 91 +++++++++++++++++++++++++++++++--------
 net/ipv4/cipso_ipv4.c             | 59 +++++++++++++++----------
 net/netlabel/netlabel_unlabeled.c |  1 +
 security/selinux/ss/mls.c         | 10 ++---
 security/selinux/ss/services.c    |  5 ++-
 5 files changed, 120 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 2e5b2f6f9fa0..18b73cf507df 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -105,17 +105,49 @@ struct netlbl_dom_map;
 /* Domain mapping operations */
 int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info);
 
-/* LSM security attributes */
+/*
+ * LSM security attributes
+ */
+
+/**
+ * struct netlbl_lsm_cache - NetLabel LSM security attribute cache
+ * @refcount: atomic reference counter
+ * @free: LSM supplied function to free the cache data
+ * @data: LSM supplied cache data
+ *
+ * Description:
+ * This structure is provided for LSMs which wish to make use of the NetLabel
+ * caching mechanism to store LSM specific data/attributes in the NetLabel
+ * cache.  If the LSM has to perform a lot of translation from the NetLabel
+ * security attributes into it's own internal representation then the cache
+ * mechanism can provide a way to eliminate some or all of that translation
+ * overhead on a cache hit.
+ *
+ */
 struct netlbl_lsm_cache {
 	atomic_t refcount;
 	void (*free) (const void *data);
 	void *data;
 };
-/* The catmap bitmap field MUST be a power of two in length and large
+
+/**
+ * struct netlbl_lsm_secattr_catmap - NetLabel LSM secattr category bitmap
+ * @startbit: the value of the lowest order bit in the bitmap
+ * @bitmap: the category bitmap
+ * @next: pointer to the next bitmap "node" or NULL
+ *
+ * Description:
+ * This structure is used to represent category bitmaps.  Due to the large
+ * number of categories supported by most labeling protocols it is not
+ * practical to transfer a full bitmap internally so NetLabel adopts a sparse
+ * bitmap structure modeled after SELinux's ebitmap structure.
+ * The catmap bitmap field MUST be a power of two in length and large
  * enough to hold at least 240 bits.  Special care (i.e. check the code!)
  * should be used when changing these values as the LSM implementation
  * probably has functions which rely on the sizes of these types to speed
- * processing. */
+ * processing.
+ *
+ */
 #define NETLBL_CATMAP_MAPTYPE           u64
 #define NETLBL_CATMAP_MAPCNT            4
 #define NETLBL_CATMAP_MAPSIZE           (sizeof(NETLBL_CATMAP_MAPTYPE) * 8)
@@ -127,22 +159,48 @@ struct netlbl_lsm_secattr_catmap {
 	NETLBL_CATMAP_MAPTYPE bitmap[NETLBL_CATMAP_MAPCNT];
 	struct netlbl_lsm_secattr_catmap *next;
 };
+
+/**
+ * struct netlbl_lsm_secattr - NetLabel LSM security attributes
+ * @flags: indicate which attributes are contained in this structure
+ * @type: indicate the NLTYPE of the attributes
+ * @domain: the NetLabel LSM domain
+ * @cache: NetLabel LSM specific cache
+ * @attr.mls: MLS sensitivity label
+ * @attr.mls.cat: MLS category bitmap
+ * @attr.mls.lvl: MLS sensitivity level
+ * @attr.secid: LSM specific secid token
+ *
+ * Description:
+ * This structure is used to pass security attributes between NetLabel and the
+ * LSM modules.  The flags field is used to specify which fields within the
+ * struct are valid and valid values can be created by bitwise OR'ing the
+ * NETLBL_SECATTR_* defines.  The domain field is typically set by the LSM to
+ * specify domain specific configuration settings and is not usually used by
+ * NetLabel itself when returning security attributes to the LSM.
+ *
+ */
 #define NETLBL_SECATTR_NONE             0x00000000
 #define NETLBL_SECATTR_DOMAIN           0x00000001
 #define NETLBL_SECATTR_CACHE            0x00000002
 #define NETLBL_SECATTR_MLS_LVL          0x00000004
 #define NETLBL_SECATTR_MLS_CAT          0x00000008
+#define NETLBL_SECATTR_SECID            0x00000010
 #define NETLBL_SECATTR_CACHEABLE        (NETLBL_SECATTR_MLS_LVL | \
-					 NETLBL_SECATTR_MLS_CAT)
+					 NETLBL_SECATTR_MLS_CAT | \
+					 NETLBL_SECATTR_SECID)
 struct netlbl_lsm_secattr {
 	u32 flags;
-
+	u32 type;
 	char *domain;
-
-	u32 mls_lvl;
-	struct netlbl_lsm_secattr_catmap *mls_cat;
-
 	struct netlbl_lsm_cache *cache;
+	union {
+		struct {
+			struct netlbl_lsm_secattr_catmap *cat;
+			u32 lvl;
+		} mls;
+		u32 secid;
+	} attr;
 };
 
 /*
@@ -231,10 +289,7 @@ static inline void netlbl_secattr_catmap_free(
  */
 static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr)
 {
-	secattr->flags = 0;
-	secattr->domain = NULL;
-	secattr->mls_cat = NULL;
-	secattr->cache = NULL;
+	memset(secattr, 0, sizeof(*secattr));
 }
 
 /**
@@ -248,11 +303,11 @@ static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr)
  */
 static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr)
 {
-	if (secattr->cache)
-		netlbl_secattr_cache_free(secattr->cache);
 	kfree(secattr->domain);
-	if (secattr->mls_cat)
-		netlbl_secattr_catmap_free(secattr->mls_cat);
+	if (secattr->flags & NETLBL_SECATTR_CACHE)
+		netlbl_secattr_cache_free(secattr->cache);
+	if (secattr->flags & NETLBL_SECATTR_MLS_CAT)
+		netlbl_secattr_catmap_free(secattr->attr.mls.cat);
 }
 
 /**
@@ -300,7 +355,7 @@ int netlbl_secattr_catmap_setrng(struct netlbl_lsm_secattr_catmap *catmap,
 				 gfp_t flags);
 
 /*
- * LSM protocol operations
+ * LSM protocol operations (NetLabel LSM/kernel API)
  */
 int netlbl_enabled(void);
 int netlbl_sock_setattr(struct sock *sk,
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index d4dc4eb48d95..a2241060113b 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -348,6 +348,7 @@ static int cipso_v4_cache_check(const unsigned char *key,
 			atomic_inc(&entry->lsm_data->refcount);
 			secattr->cache = entry->lsm_data;
 			secattr->flags |= NETLBL_SECATTR_CACHE;
+			secattr->type = NETLBL_NLTYPE_CIPSOV4;
 			if (prev_entry == NULL) {
 				spin_unlock_bh(&cipso_v4_cache[bkt].lock);
 				return 0;
@@ -865,7 +866,7 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def,
 	}
 
 	for (;;) {
-		host_spot = netlbl_secattr_catmap_walk(secattr->mls_cat,
+		host_spot = netlbl_secattr_catmap_walk(secattr->attr.mls.cat,
 						       host_spot + 1);
 		if (host_spot < 0)
 			break;
@@ -948,7 +949,7 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def,
 				return -EPERM;
 			break;
 		}
-		ret_val = netlbl_secattr_catmap_setbit(secattr->mls_cat,
+		ret_val = netlbl_secattr_catmap_setbit(secattr->attr.mls.cat,
 						       host_spot,
 						       GFP_ATOMIC);
 		if (ret_val != 0)
@@ -1014,7 +1015,8 @@ static int cipso_v4_map_cat_enum_hton(const struct cipso_v4_doi *doi_def,
 	u32 cat_iter = 0;
 
 	for (;;) {
-		cat = netlbl_secattr_catmap_walk(secattr->mls_cat, cat + 1);
+		cat = netlbl_secattr_catmap_walk(secattr->attr.mls.cat,
+						 cat + 1);
 		if (cat < 0)
 			break;
 		if ((cat_iter + 2) > net_cat_len)
@@ -1049,7 +1051,7 @@ static int cipso_v4_map_cat_enum_ntoh(const struct cipso_v4_doi *doi_def,
 	u32 iter;
 
 	for (iter = 0; iter < net_cat_len; iter += 2) {
-		ret_val = netlbl_secattr_catmap_setbit(secattr->mls_cat,
+		ret_val = netlbl_secattr_catmap_setbit(secattr->attr.mls.cat,
 				ntohs(get_unaligned((__be16 *)&net_cat[iter])),
 				GFP_ATOMIC);
 		if (ret_val != 0)
@@ -1130,7 +1132,8 @@ static int cipso_v4_map_cat_rng_hton(const struct cipso_v4_doi *doi_def,
 		return -ENOSPC;
 
 	for (;;) {
-		iter = netlbl_secattr_catmap_walk(secattr->mls_cat, iter + 1);
+		iter = netlbl_secattr_catmap_walk(secattr->attr.mls.cat,
+						  iter + 1);
 		if (iter < 0)
 			break;
 		cat_size += (iter == 0 ? 0 : sizeof(u16));
@@ -1138,7 +1141,8 @@ static int cipso_v4_map_cat_rng_hton(const struct cipso_v4_doi *doi_def,
 			return -ENOSPC;
 		array[array_cnt++] = iter;
 
-		iter = netlbl_secattr_catmap_walk_rng(secattr->mls_cat, iter);
+		iter = netlbl_secattr_catmap_walk_rng(secattr->attr.mls.cat,
+						      iter);
 		if (iter < 0)
 			return -EFAULT;
 		cat_size += sizeof(u16);
@@ -1191,7 +1195,7 @@ static int cipso_v4_map_cat_rng_ntoh(const struct cipso_v4_doi *doi_def,
 		else
 			cat_low = 0;
 
-		ret_val = netlbl_secattr_catmap_setrng(secattr->mls_cat,
+		ret_val = netlbl_secattr_catmap_setrng(secattr->attr.mls.cat,
 						       cat_low,
 						       cat_high,
 						       GFP_ATOMIC);
@@ -1251,7 +1255,9 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def,
 	if ((secattr->flags & NETLBL_SECATTR_MLS_LVL) == 0)
 		return -EPERM;
 
-	ret_val = cipso_v4_map_lvl_hton(doi_def, secattr->mls_lvl, &level);
+	ret_val = cipso_v4_map_lvl_hton(doi_def,
+					secattr->attr.mls.lvl,
+					&level);
 	if (ret_val != 0)
 		return ret_val;
 
@@ -1303,12 +1309,13 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def,
 	ret_val = cipso_v4_map_lvl_ntoh(doi_def, tag[3], &level);
 	if (ret_val != 0)
 		return ret_val;
-	secattr->mls_lvl = level;
+	secattr->attr.mls.lvl = level;
 	secattr->flags |= NETLBL_SECATTR_MLS_LVL;
 
 	if (tag_len > 4) {
-		secattr->mls_cat = netlbl_secattr_catmap_alloc(GFP_ATOMIC);
-		if (secattr->mls_cat == NULL)
+		secattr->attr.mls.cat =
+		                       netlbl_secattr_catmap_alloc(GFP_ATOMIC);
+		if (secattr->attr.mls.cat == NULL)
 			return -ENOMEM;
 
 		ret_val = cipso_v4_map_cat_rbm_ntoh(doi_def,
@@ -1316,7 +1323,7 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def,
 						    tag_len - 4,
 						    secattr);
 		if (ret_val != 0) {
-			netlbl_secattr_catmap_free(secattr->mls_cat);
+			netlbl_secattr_catmap_free(secattr->attr.mls.cat);
 			return ret_val;
 		}
 
@@ -1350,7 +1357,9 @@ static int cipso_v4_gentag_enum(const struct cipso_v4_doi *doi_def,
 	if (!(secattr->flags & NETLBL_SECATTR_MLS_LVL))
 		return -EPERM;
 
-	ret_val = cipso_v4_map_lvl_hton(doi_def, secattr->mls_lvl, &level);
+	ret_val = cipso_v4_map_lvl_hton(doi_def,
+					secattr->attr.mls.lvl,
+					&level);
 	if (ret_val != 0)
 		return ret_val;
 
@@ -1396,12 +1405,13 @@ static int cipso_v4_parsetag_enum(const struct cipso_v4_doi *doi_def,
 	ret_val = cipso_v4_map_lvl_ntoh(doi_def, tag[3], &level);
 	if (ret_val != 0)
 		return ret_val;
-	secattr->mls_lvl = level;
+	secattr->attr.mls.lvl = level;
 	secattr->flags |= NETLBL_SECATTR_MLS_LVL;
 
 	if (tag_len > 4) {
-		secattr->mls_cat = netlbl_secattr_catmap_alloc(GFP_ATOMIC);
-		if (secattr->mls_cat == NULL)
+		secattr->attr.mls.cat =
+			               netlbl_secattr_catmap_alloc(GFP_ATOMIC);
+		if (secattr->attr.mls.cat == NULL)
 			return -ENOMEM;
 
 		ret_val = cipso_v4_map_cat_enum_ntoh(doi_def,
@@ -1409,7 +1419,7 @@ static int cipso_v4_parsetag_enum(const struct cipso_v4_doi *doi_def,
 						     tag_len - 4,
 						     secattr);
 		if (ret_val != 0) {
-			netlbl_secattr_catmap_free(secattr->mls_cat);
+			netlbl_secattr_catmap_free(secattr->attr.mls.cat);
 			return ret_val;
 		}
 
@@ -1443,7 +1453,9 @@ static int cipso_v4_gentag_rng(const struct cipso_v4_doi *doi_def,
 	if (!(secattr->flags & NETLBL_SECATTR_MLS_LVL))
 		return -EPERM;
 
-	ret_val = cipso_v4_map_lvl_hton(doi_def, secattr->mls_lvl, &level);
+	ret_val = cipso_v4_map_lvl_hton(doi_def,
+					secattr->attr.mls.lvl,
+					&level);
 	if (ret_val != 0)
 		return ret_val;
 
@@ -1488,12 +1500,13 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def,
 	ret_val = cipso_v4_map_lvl_ntoh(doi_def, tag[3], &level);
 	if (ret_val != 0)
 		return ret_val;
-	secattr->mls_lvl = level;
+	secattr->attr.mls.lvl = level;
 	secattr->flags |= NETLBL_SECATTR_MLS_LVL;
 
 	if (tag_len > 4) {
-		secattr->mls_cat = netlbl_secattr_catmap_alloc(GFP_ATOMIC);
-		if (secattr->mls_cat == NULL)
+		secattr->attr.mls.cat =
+			               netlbl_secattr_catmap_alloc(GFP_ATOMIC);
+		if (secattr->attr.mls.cat == NULL)
 			return -ENOMEM;
 
 		ret_val = cipso_v4_map_cat_rng_ntoh(doi_def,
@@ -1501,7 +1514,7 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def,
 						    tag_len - 4,
 						    secattr);
 		if (ret_val != 0) {
-			netlbl_secattr_catmap_free(secattr->mls_cat);
+			netlbl_secattr_catmap_free(secattr->attr.mls.cat);
 			return ret_val;
 		}
 
@@ -1850,6 +1863,8 @@ static int cipso_v4_getattr(const unsigned char *cipso,
 		ret_val = cipso_v4_parsetag_rng(doi_def, &cipso[6], secattr);
 		break;
 	}
+	if (ret_val == 0)
+		secattr->type = NETLBL_NLTYPE_CIPSOV4;
 
 getattr_return:
 	rcu_read_unlock();
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index fd53c7ae2977..7f5df0cbc63f 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -241,6 +241,7 @@ int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr)
 	if (netlabel_unlabel_acceptflg == 0)
 		return -ENOMSG;
 	netlbl_secattr_init(secattr);
+	secattr->type = NETLBL_NLTYPE_UNLABELED;
 	return 0;
 }
 
diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c
index 3bbcb5369af9..feaf0a5b828f 100644
--- a/security/selinux/ss/mls.c
+++ b/security/selinux/ss/mls.c
@@ -562,7 +562,7 @@ void mls_export_netlbl_lvl(struct context *context,
 	if (!selinux_mls_enabled)
 		return;
 
-	secattr->mls_lvl = context->range.level[0].sens - 1;
+	secattr->attr.mls.lvl = context->range.level[0].sens - 1;
 	secattr->flags |= NETLBL_SECATTR_MLS_LVL;
 }
 
@@ -582,7 +582,7 @@ void mls_import_netlbl_lvl(struct context *context,
 	if (!selinux_mls_enabled)
 		return;
 
-	context->range.level[0].sens = secattr->mls_lvl + 1;
+	context->range.level[0].sens = secattr->attr.mls.lvl + 1;
 	context->range.level[1].sens = context->range.level[0].sens;
 }
 
@@ -605,8 +605,8 @@ int mls_export_netlbl_cat(struct context *context,
 		return 0;
 
 	rc = ebitmap_netlbl_export(&context->range.level[0].cat,
-				   &secattr->mls_cat);
-	if (rc == 0 && secattr->mls_cat != NULL)
+				   &secattr->attr.mls.cat);
+	if (rc == 0 && secattr->attr.mls.cat != NULL)
 		secattr->flags |= NETLBL_SECATTR_MLS_CAT;
 
 	return rc;
@@ -633,7 +633,7 @@ int mls_import_netlbl_cat(struct context *context,
 		return 0;
 
 	rc = ebitmap_netlbl_import(&context->range.level[0].cat,
-				   secattr->mls_cat);
+				   secattr->attr.mls.cat);
 	if (rc != 0)
 		goto import_netlbl_cat_failure;
 
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 4bf715d4cf29..0f97ef578370 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2550,6 +2550,9 @@ int security_netlbl_secattr_to_sid(struct netlbl_lsm_secattr *secattr,
 		default:
 			goto netlbl_secattr_to_sid_return;
 		}
+	} else if (secattr->flags & NETLBL_SECATTR_SECID) {
+		*sid = secattr->attr.secid;
+		rc = 0;
 	} else if (secattr->flags & NETLBL_SECATTR_MLS_LVL) {
 		ctx = sidtab_search(&sidtab, base_sid);
 		if (ctx == NULL)
@@ -2561,7 +2564,7 @@ int security_netlbl_secattr_to_sid(struct netlbl_lsm_secattr *secattr,
 		mls_import_netlbl_lvl(&ctx_new, secattr);
 		if (secattr->flags & NETLBL_SECATTR_MLS_CAT) {
 			if (ebitmap_netlbl_import(&ctx_new.range.level[0].cat,
-						  secattr->mls_cat) != 0)
+						  secattr->attr.mls.cat) != 0)
 				goto netlbl_secattr_to_sid_return;
 			ctx_new.range.level[1].cat.highbit =
 				ctx_new.range.level[0].cat.highbit;
-- 
cgit v1.2.3


From 75e22910cf0c26802b09dac2e34c13e648d3ed02 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Jan 2008 08:38:04 -0500
Subject: NetLabel: Add IP address family information to the
 netlbl_skbuff_getattr() function

In order to do any sort of IP header inspection of incoming packets we need to
know which address family, AF_INET/AF_INET6/etc., it belongs to and since the
sk_buff structure does not store this information we need to pass along the
address family separate from the packet itself.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/net/netlabel.h              |  2 ++
 net/netlabel/netlabel_kapi.c        |  2 ++
 security/selinux/hooks.c            | 33 ++++++++++++++++++++++-----------
 security/selinux/include/netlabel.h |  8 +++++++-
 security/selinux/netlabel.c         | 12 +++++++++---
 5 files changed, 42 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 18b73cf507df..a3bffb4bb749 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -363,6 +363,7 @@ int netlbl_sock_setattr(struct sock *sk,
 int netlbl_sock_getattr(struct sock *sk,
 			struct netlbl_lsm_secattr *secattr);
 int netlbl_skbuff_getattr(const struct sk_buff *skb,
+			  u16 family,
 			  struct netlbl_lsm_secattr *secattr);
 void netlbl_skbuff_err(struct sk_buff *skb, int error);
 
@@ -415,6 +416,7 @@ static inline int netlbl_sock_getattr(struct sock *sk,
 	return -ENOSYS;
 }
 static inline int netlbl_skbuff_getattr(const struct sk_buff *skb,
+					u16 family,
 					struct netlbl_lsm_secattr *secattr)
 {
 	return -ENOSYS;
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index d3762ea24957..49146153f235 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -332,6 +332,7 @@ int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 /**
  * netlbl_skbuff_getattr - Determine the security attributes of a packet
  * @skb: the packet
+ * @family: protocol family
  * @secattr: the security attributes
  *
  * Description:
@@ -342,6 +343,7 @@ int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
  *
  */
 int netlbl_skbuff_getattr(const struct sk_buff *skb,
+			  u16 family,
 			  struct netlbl_lsm_secattr *secattr)
 {
 	if (CIPSO_V4_OPTEXIST(skb) &&
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 64d414efb404..5df12072c8d5 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3429,6 +3429,7 @@ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad,
 /**
  * selinux_skb_extlbl_sid - Determine the external label of a packet
  * @skb: the packet
+ * @family: protocol family
  * @sid: the packet's SID
  *
  * Description:
@@ -3441,13 +3442,16 @@ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad,
  * selinux_netlbl_skbuff_getsid().
  *
  */
-static void selinux_skb_extlbl_sid(struct sk_buff *skb, u32 *sid)
+static void selinux_skb_extlbl_sid(struct sk_buff *skb,
+				   u16 family,
+				   u32 *sid)
 {
 	u32 xfrm_sid;
 	u32 nlbl_sid;
 
 	selinux_skb_xfrm_sid(skb, &xfrm_sid);
 	if (selinux_netlbl_skbuff_getsid(skb,
+					 family,
 					 (xfrm_sid == SECSID_NULL ?
 					  SECINITSID_NETMSG : xfrm_sid),
 					 &nlbl_sid) != 0)
@@ -3940,7 +3944,7 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (err)
 		goto out;
 
-	err = selinux_netlbl_sock_rcv_skb(sksec, skb, &ad);
+	err = selinux_netlbl_sock_rcv_skb(sksec, skb, family, &ad);
 	if (err)
 		goto out;
 
@@ -3996,18 +4000,25 @@ out:
 static int selinux_socket_getpeersec_dgram(struct socket *sock, struct sk_buff *skb, u32 *secid)
 {
 	u32 peer_secid = SECSID_NULL;
-	int err = 0;
+	u16 family;
 
-	if (sock && sock->sk->sk_family == PF_UNIX)
+	if (sock)
+		family = sock->sk->sk_family;
+	else if (skb && skb->sk)
+		family = skb->sk->sk_family;
+	else
+		goto out;
+
+	if (sock && family == PF_UNIX)
 		selinux_get_inode_sid(SOCK_INODE(sock), &peer_secid);
 	else if (skb)
-		selinux_skb_extlbl_sid(skb, &peer_secid);
+		selinux_skb_extlbl_sid(skb, family, &peer_secid);
 
-	if (peer_secid == SECSID_NULL)
-		err = -EINVAL;
+out:
 	*secid = peer_secid;
-
-	return err;
+	if (peer_secid == SECSID_NULL)
+		return -EINVAL;
+	return 0;
 }
 
 static int selinux_sk_alloc_security(struct sock *sk, int family, gfp_t priority)
@@ -4062,7 +4073,7 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
 	u32 newsid;
 	u32 peersid;
 
-	selinux_skb_extlbl_sid(skb, &peersid);
+	selinux_skb_extlbl_sid(skb, sk->sk_family, &peersid);
 	if (peersid == SECSID_NULL) {
 		req->secid = sksec->sid;
 		req->peer_secid = SECSID_NULL;
@@ -4100,7 +4111,7 @@ static void selinux_inet_conn_established(struct sock *sk,
 {
 	struct sk_security_struct *sksec = sk->sk_security;
 
-	selinux_skb_extlbl_sid(skb, &sksec->peer_sid);
+	selinux_skb_extlbl_sid(skb, sk->sk_family, &sksec->peer_sid);
 }
 
 static void selinux_req_classify_flow(const struct request_sock *req,
diff --git a/security/selinux/include/netlabel.h b/security/selinux/include/netlabel.h
index 218e3f77c350..272769a1cb96 100644
--- a/security/selinux/include/netlabel.h
+++ b/security/selinux/include/netlabel.h
@@ -46,13 +46,17 @@ void selinux_netlbl_sk_security_init(struct sk_security_struct *ssec,
 void selinux_netlbl_sk_security_clone(struct sk_security_struct *ssec,
 				      struct sk_security_struct *newssec);
 
-int selinux_netlbl_skbuff_getsid(struct sk_buff *skb, u32 base_sid, u32 *sid);
+int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
+				 u16 family,
+				 u32 base_sid,
+				 u32 *sid);
 
 void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock);
 int selinux_netlbl_socket_post_create(struct socket *sock);
 int selinux_netlbl_inode_permission(struct inode *inode, int mask);
 int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 				struct sk_buff *skb,
+				u16 family,
 				struct avc_audit_data *ad);
 int selinux_netlbl_socket_setsockopt(struct socket *sock,
 				     int level,
@@ -83,6 +87,7 @@ static inline void selinux_netlbl_sk_security_clone(
 }
 
 static inline int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
+					       u16 family,
 					       u32 base_sid,
 					       u32 *sid)
 {
@@ -106,6 +111,7 @@ static inline int selinux_netlbl_inode_permission(struct inode *inode,
 }
 static inline int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 					      struct sk_buff *skb,
+					      u16 family,
 					      struct avc_audit_data *ad)
 {
 	return 0;
diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c
index 66e013d6f6f6..f4bcbf12a4c9 100644
--- a/security/selinux/netlabel.c
+++ b/security/selinux/netlabel.c
@@ -144,6 +144,7 @@ void selinux_netlbl_sk_security_clone(struct sk_security_struct *ssec,
 /**
  * selinux_netlbl_skbuff_getsid - Get the sid of a packet using NetLabel
  * @skb: the packet
+ * @family: protocol family
  * @base_sid: the SELinux SID to use as a context for MLS only attributes
  * @sid: the SID
  *
@@ -153,7 +154,10 @@ void selinux_netlbl_sk_security_clone(struct sk_security_struct *ssec,
  * assign to the packet.  Returns zero on success, negative values on failure.
  *
  */
-int selinux_netlbl_skbuff_getsid(struct sk_buff *skb, u32 base_sid, u32 *sid)
+int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
+				 u16 family,
+				 u32 base_sid,
+				 u32 *sid)
 {
 	int rc;
 	struct netlbl_lsm_secattr secattr;
@@ -164,7 +168,7 @@ int selinux_netlbl_skbuff_getsid(struct sk_buff *skb, u32 base_sid, u32 *sid)
 	}
 
 	netlbl_secattr_init(&secattr);
-	rc = netlbl_skbuff_getattr(skb, &secattr);
+	rc = netlbl_skbuff_getattr(skb, family, &secattr);
 	if (rc == 0 && secattr.flags != NETLBL_SECATTR_NONE) {
 		rc = security_netlbl_secattr_to_sid(&secattr, base_sid, sid);
 		if (rc == 0 &&
@@ -292,6 +296,7 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
  * selinux_netlbl_sock_rcv_skb - Do an inbound access check using NetLabel
  * @sksec: the sock's sk_security_struct
  * @skb: the packet
+ * @family: protocol family
  * @ad: the audit data
  *
  * Description:
@@ -302,6 +307,7 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
  */
 int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 				struct sk_buff *skb,
+				u16 family,
 				struct avc_audit_data *ad)
 {
 	int rc;
@@ -313,7 +319,7 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 		return 0;
 
 	netlbl_secattr_init(&secattr);
-	rc = netlbl_skbuff_getattr(skb, &secattr);
+	rc = netlbl_skbuff_getattr(skb, family, &secattr);
 	if (rc == 0 && secattr.flags != NETLBL_SECATTR_NONE) {
 		rc = security_netlbl_secattr_to_sid(&secattr,
 						    SECINITSID_NETMSG,
-- 
cgit v1.2.3


From d621d35e576aa20a0ddae8022c3810f38357c8ff Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Jan 2008 08:43:36 -0500
Subject: SELinux: Enable dynamic enable/disable of the network access checks

This patch introduces a mechanism for checking when labeled IPsec or SECMARK
are in use by keeping introducing a configuration reference counter for each
subsystem.  In the case of labeled IPsec, whenever a labeled SA or SPD entry
is created the labeled IPsec/XFRM reference count is increased and when the
entry is removed it is decreased.  In the case of SECMARK, when a SECMARK
target is created the reference count is increased and later decreased when the
target is removed.  These reference counters allow SELinux to quickly determine
if either of these subsystems are enabled.

NetLabel already has a similar mechanism which provides the netlbl_enabled()
function.

This patch also renames the selinux_relabel_packet_permission() function to
selinux_secmark_relabel_packet_permission() as the original name and
description were misleading in that they referenced a single packet label which
is not the case.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/selinux.h         | 45 +++++++++++++++++++++++++++++++++-------
 net/netfilter/xt_SECMARK.c      | 13 +++++++++++-
 security/selinux/exports.c      | 20 ++++++++++++++++--
 security/selinux/hooks.c        | 46 +++++++++++++++++++++++++++++++++--------
 security/selinux/include/xfrm.h | 12 +++++++++++
 security/selinux/xfrm.c         | 18 ++++++++++++++--
 6 files changed, 132 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 6080f73fc85f..8c2cc4c02526 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -120,16 +120,35 @@ void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
 int selinux_string_to_sid(char *str, u32 *sid);
 
 /**
- *     selinux_relabel_packet_permission - check permission to relabel a packet
- *     @sid: ID value to be applied to network packet (via SECMARK, most likely)
+ *     selinux_secmark_relabel_packet_permission - secmark permission check
+ *     @sid: SECMARK ID value to be applied to network packet
  *
- *     Returns 0 if the current task is allowed to label packets with the
- *     supplied security ID.  Note that it is implicit that the packet is always
- *     being relabeled from the default unlabled value, and that the access
- *     control decision is made in the AVC.
+ *     Returns 0 if the current task is allowed to set the SECMARK label of
+ *     packets with the supplied security ID.  Note that it is implicit that
+ *     the packet is always being relabeled from the default unlabeled value,
+ *     and that the access control decision is made in the AVC.
  */
-int selinux_relabel_packet_permission(u32 sid);
+int selinux_secmark_relabel_packet_permission(u32 sid);
 
+/**
+ *     selinux_secmark_refcount_inc - increments the secmark use counter
+ *
+ *     SELinux keeps track of the current SECMARK targets in use so it knows
+ *     when to apply SECMARK label access checks to network packets.  This
+ *     function incements this reference count to indicate that a new SECMARK
+ *     target has been configured.
+ */
+void selinux_secmark_refcount_inc(void);
+
+/**
+ *     selinux_secmark_refcount_dec - decrements the secmark use counter
+ *
+ *     SELinux keeps track of the current SECMARK targets in use so it knows
+ *     when to apply SECMARK label access checks to network packets.  This
+ *     function decements this reference count to indicate that one of the
+ *     existing SECMARK targets has been removed/flushed.
+ */
+void selinux_secmark_refcount_dec(void);
 #else
 
 static inline int selinux_audit_rule_init(u32 field, u32 op,
@@ -184,11 +203,21 @@ static inline int selinux_string_to_sid(const char *str, u32 *sid)
        return 0;
 }
 
-static inline int selinux_relabel_packet_permission(u32 sid)
+static inline int selinux_secmark_relabel_packet_permission(u32 sid)
 {
 	return 0;
 }
 
+static inline void selinux_secmark_refcount_inc(void)
+{
+	return;
+}
+
+static inline void selinux_secmark_refcount_dec(void)
+{
+	return;
+}
+
 #endif	/* CONFIG_SECURITY_SELINUX */
 
 #endif /* _LINUX_SELINUX_H */
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index b11b3ecbb39d..7708e2084ce2 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -72,12 +72,13 @@ static bool checkentry_selinux(struct xt_secmark_target_info *info)
 		return false;
 	}
 
-	err = selinux_relabel_packet_permission(sel->selsid);
+	err = selinux_secmark_relabel_packet_permission(sel->selsid);
 	if (err) {
 		printk(KERN_INFO PFX "unable to obtain relabeling permission\n");
 		return false;
 	}
 
+	selinux_secmark_refcount_inc();
 	return true;
 }
 
@@ -110,11 +111,20 @@ secmark_tg_check(const char *tablename, const void *entry,
 	return true;
 }
 
+void secmark_tg_destroy(const struct xt_target *target, void *targinfo)
+{
+	switch (mode) {
+	case SECMARK_MODE_SEL:
+		selinux_secmark_refcount_dec();
+	}
+}
+
 static struct xt_target secmark_tg_reg[] __read_mostly = {
 	{
 		.name		= "SECMARK",
 		.family		= AF_INET,
 		.checkentry	= secmark_tg_check,
+		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
 		.table		= "mangle",
@@ -124,6 +134,7 @@ static struct xt_target secmark_tg_reg[] __read_mostly = {
 		.name		= "SECMARK",
 		.family		= AF_INET6,
 		.checkentry	= secmark_tg_check,
+		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
 		.table		= "mangle",
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index b6f96943be1f..87d2bb3ea355 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -17,10 +17,14 @@
 #include <linux/selinux.h>
 #include <linux/fs.h>
 #include <linux/ipc.h>
+#include <asm/atomic.h>
 
 #include "security.h"
 #include "objsec.h"
 
+/* SECMARK reference count */
+extern atomic_t selinux_secmark_refcount;
+
 int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
 {
 	if (selinux_enabled)
@@ -74,7 +78,7 @@ int selinux_string_to_sid(char *str, u32 *sid)
 }
 EXPORT_SYMBOL_GPL(selinux_string_to_sid);
 
-int selinux_relabel_packet_permission(u32 sid)
+int selinux_secmark_relabel_packet_permission(u32 sid)
 {
 	if (selinux_enabled) {
 		struct task_security_struct *tsec = current->security;
@@ -84,4 +88,16 @@ int selinux_relabel_packet_permission(u32 sid)
 	}
 	return 0;
 }
-EXPORT_SYMBOL_GPL(selinux_relabel_packet_permission);
+EXPORT_SYMBOL_GPL(selinux_secmark_relabel_packet_permission);
+
+void selinux_secmark_refcount_inc(void)
+{
+	atomic_inc(&selinux_secmark_refcount);
+}
+EXPORT_SYMBOL_GPL(selinux_secmark_refcount_inc);
+
+void selinux_secmark_refcount_dec(void)
+{
+	atomic_dec(&selinux_secmark_refcount);
+}
+EXPORT_SYMBOL_GPL(selinux_secmark_refcount_dec);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index bfe9a05db3a2..6156241c8770 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -51,8 +51,10 @@
 #include <net/ip.h>		/* for local_port_range[] */
 #include <net/tcp.h>		/* struct or_callable used in sock_rcv_skb */
 #include <net/net_namespace.h>
+#include <net/netlabel.h>
 #include <asm/uaccess.h>
 #include <asm/ioctls.h>
+#include <asm/atomic.h>
 #include <linux/bitops.h>
 #include <linux/interrupt.h>
 #include <linux/netdevice.h>	/* for network interface checks */
@@ -91,6 +93,9 @@ extern int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm);
 extern int selinux_compat_net;
 extern struct security_operations *security_ops;
 
+/* SECMARK reference count */
+atomic_t selinux_secmark_refcount = ATOMIC_INIT(0);
+
 #ifdef CONFIG_SECURITY_SELINUX_DEVELOP
 int selinux_enforcing = 0;
 
@@ -157,6 +162,21 @@ getsecurity_exit:
 	return len;
 }
 
+/**
+ * selinux_secmark_enabled - Check to see if SECMARK is currently enabled
+ *
+ * Description:
+ * This function checks the SECMARK reference counter to see if any SECMARK
+ * targets are currently configured, if the reference counter is greater than
+ * zero SECMARK is considered to be enabled.  Returns true (1) if SECMARK is
+ * enabled, false (0) if SECMARK is disabled.
+ *
+ */
+static int selinux_secmark_enabled(void)
+{
+	return (atomic_read(&selinux_secmark_refcount) > 0);
+}
+
 /* Allocate and free functions for each kind of security blob. */
 
 static int task_alloc_security(struct task_struct *task)
@@ -3931,7 +3951,6 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	struct sk_security_struct *sksec = sk->sk_security;
 	u16 family = sk->sk_family;
 	u32 sk_sid = sksec->sid;
-	u32 peer_sid;
 	struct avc_audit_data ad;
 	char *addrp;
 
@@ -3957,15 +3976,24 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 		return selinux_sock_rcv_skb_compat(sk, skb, &ad,
 						   family, addrp);
 
-	err = avc_has_perm(sk_sid, skb->secmark, SECCLASS_PACKET,
-			   PACKET__RECV, &ad);
-	if (err)
-		return err;
+	if (selinux_secmark_enabled()) {
+		err = avc_has_perm(sk_sid, skb->secmark, SECCLASS_PACKET,
+				   PACKET__RECV, &ad);
+		if (err)
+			return err;
+	}
 
-	err = selinux_skb_peerlbl_sid(skb, family, &peer_sid);
-	if (err)
-		return err;
-	return avc_has_perm(sk_sid, peer_sid, SECCLASS_PEER, PEER__RECV, &ad);
+	if (netlbl_enabled() || selinux_xfrm_enabled()) {
+		u32 peer_sid;
+
+		err = selinux_skb_peerlbl_sid(skb, family, &peer_sid);
+		if (err)
+			return err;
+		err = avc_has_perm(sk_sid, peer_sid, SECCLASS_PEER,
+				   PEER__RECV, &ad);
+	}
+
+	return err;
 }
 
 static int selinux_socket_getpeersec_stream(struct socket *sock, char __user *optval,
diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h
index 31929e39f5ca..36b0510efa7b 100644
--- a/security/selinux/include/xfrm.h
+++ b/security/selinux/include/xfrm.h
@@ -32,6 +32,13 @@ static inline struct inode_security_struct *get_sock_isec(struct sock *sk)
 }
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
+extern atomic_t selinux_xfrm_refcount;
+
+static inline int selinux_xfrm_enabled(void)
+{
+	return (atomic_read(&selinux_xfrm_refcount) > 0);
+}
+
 int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb,
 			struct avc_audit_data *ad);
 int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb,
@@ -43,6 +50,11 @@ static inline void selinux_xfrm_notify_policyload(void)
 	atomic_inc(&flow_cache_genid);
 }
 #else
+static inline int selinux_xfrm_enabled(void)
+{
+	return 0;
+}
+
 static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb,
 			struct avc_audit_data *ad)
 {
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index e07603969033..7e158205d081 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -46,11 +46,14 @@
 #include <net/checksum.h>
 #include <net/udp.h>
 #include <asm/semaphore.h>
+#include <asm/atomic.h>
 
 #include "avc.h"
 #include "objsec.h"
 #include "xfrm.h"
 
+/* Labeled XFRM instance counter */
+atomic_t selinux_xfrm_refcount = ATOMIC_INIT(0);
 
 /*
  * Returns true if an LSM/SELinux context
@@ -293,6 +296,9 @@ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
 	BUG_ON(!uctx);
 
 	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, 0);
+	if (err == 0)
+		atomic_inc(&selinux_xfrm_refcount);
+
 	return err;
 }
 
@@ -340,10 +346,13 @@ int selinux_xfrm_policy_delete(struct xfrm_policy *xp)
 	struct xfrm_sec_ctx *ctx = xp->security;
 	int rc = 0;
 
-	if (ctx)
+	if (ctx) {
 		rc = avc_has_perm(tsec->sid, ctx->ctx_sid,
 				  SECCLASS_ASSOCIATION,
 				  ASSOCIATION__SETCONTEXT, NULL);
+		if (rc == 0)
+			atomic_dec(&selinux_xfrm_refcount);
+	}
 
 	return rc;
 }
@@ -360,6 +369,8 @@ int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *uct
 	BUG_ON(!x);
 
 	err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, secid);
+	if (err == 0)
+		atomic_inc(&selinux_xfrm_refcount);
 	return err;
 }
 
@@ -382,10 +393,13 @@ int selinux_xfrm_state_delete(struct xfrm_state *x)
 	struct xfrm_sec_ctx *ctx = x->security;
 	int rc = 0;
 
-	if (ctx)
+	if (ctx) {
 		rc = avc_has_perm(tsec->sid, ctx->ctx_sid,
 				  SECCLASS_ASSOCIATION,
 				  ASSOCIATION__SETCONTEXT, NULL);
+		if (rc == 0)
+			atomic_dec(&selinux_xfrm_refcount);
+	}
 
 	return rc;
 }
-- 
cgit v1.2.3


From 8cc44579d1bd77ba3a32f2cb76fd9669c229c5fd Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Jan 2008 08:44:21 -0500
Subject: NetLabel: Introduce static network labels for unlabeled connections

Most trusted OSs, with the exception of Linux, have the ability to specify
static security labels for unlabeled networks.  This patch adds this ability to
the NetLabel packet labeling framework.

If the NetLabel subsystem is called to determine the security attributes of an
incoming packet it first checks to see if any recognized NetLabel packet
labeling protocols are in-use on the packet.  If none can be found then the
unlabled connection table is queried and based on the packets incoming
interface and address it is matched with a security label as configured by the
administrator using the netlabel_tools package.  The matching security label is
returned to the caller just as if the packet was explicitly labeled using a
labeling protocol.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/net/netlabel.h            |    6 +-
 net/netlabel/netlabel_kapi.c      |   16 +-
 net/netlabel/netlabel_unlabeled.c | 1375 ++++++++++++++++++++++++++++++++++++-
 net/netlabel/netlabel_unlabeled.h |  145 +++-
 4 files changed, 1524 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index a3bffb4bb749..b3213c7c5309 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -67,7 +67,11 @@
  * NetLabel NETLINK protocol
  */
 
-#define NETLBL_PROTO_VERSION            1
+/* NetLabel NETLINK protocol version
+ *  1: initial version
+ *  2: added static labels for unlabeled connections
+ */
+#define NETLBL_PROTO_VERSION            2
 
 /* NetLabel NETLINK types/families */
 #define NETLBL_NLTYPE_NONE              0
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 49146153f235..c69e3e1f05c3 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -312,7 +312,7 @@ socket_setattr_return:
  * @secattr: the security attributes
  *
  * Description:
- * Examines the given sock to see any NetLabel style labeling has been
+ * Examines the given sock to see if any NetLabel style labeling has been
  * applied to the sock, if so it parses the socket label and returns the
  * security attributes in @secattr.  Returns zero on success, negative values
  * on failure.
@@ -320,13 +320,7 @@ socket_setattr_return:
  */
 int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 {
-	int ret_val;
-
-	ret_val = cipso_v4_sock_getattr(sk, secattr);
-	if (ret_val == 0)
-		return 0;
-
-	return netlbl_unlabel_getattr(secattr);
+	return cipso_v4_sock_getattr(sk, secattr);
 }
 
 /**
@@ -350,7 +344,7 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb,
 	    cipso_v4_skbuff_getattr(skb, secattr) == 0)
 		return 0;
 
-	return netlbl_unlabel_getattr(secattr);
+	return netlbl_unlabel_getattr(skb, family, secattr);
 }
 
 /**
@@ -434,6 +428,10 @@ static int __init netlbl_init(void)
 	if (ret_val != 0)
 		goto init_failure;
 
+	ret_val = netlbl_unlabel_init(NETLBL_UNLHSH_BITSIZE);
+	if (ret_val != 0)
+		goto init_failure;
+
 	ret_val = netlbl_netlink_init();
 	if (ret_val != 0)
 		goto init_failure;
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 7f5df0cbc63f..d0c628c875ae 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -10,7 +10,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 - 2007
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -29,27 +29,99 @@
  */
 
 #include <linux/types.h>
+#include <linux/rcupdate.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/socket.h>
 #include <linux/string.h>
 #include <linux/skbuff.h>
 #include <linux/audit.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/notifier.h>
+#include <linux/netdevice.h>
+#include <linux/security.h>
 #include <net/sock.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
-
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <net/net_namespace.h>
 #include <net/netlabel.h>
 #include <asm/bug.h>
+#include <asm/atomic.h>
 
 #include "netlabel_user.h"
 #include "netlabel_domainhash.h"
 #include "netlabel_unlabeled.h"
+#include "netlabel_mgmt.h"
+
+/* NOTE: at present we always use init's network namespace since we don't
+ *       presently support different namespaces even though the majority of
+ *       the functions in this file are "namespace safe" */
+
+/* The unlabeled connection hash table which we use to map network interfaces
+ * and addresses of unlabeled packets to a user specified secid value for the
+ * LSM.  The hash table is used to lookup the network interface entry
+ * (struct netlbl_unlhsh_iface) and then the interface entry is used to
+ * lookup an IP address match from an ordered list.  If a network interface
+ * match can not be found in the hash table then the default entry
+ * (netlbl_unlhsh_def) is used.  The IP address entry list
+ * (struct netlbl_unlhsh_addr) is ordered such that the entries with a
+ * larger netmask come first.
+ */
+struct netlbl_unlhsh_tbl {
+	struct list_head *tbl;
+	u32 size;
+};
+struct netlbl_unlhsh_addr4 {
+	__be32 addr;
+	__be32 mask;
+	u32 secid;
+
+	u32 valid;
+	struct list_head list;
+	struct rcu_head rcu;
+};
+struct netlbl_unlhsh_addr6 {
+	struct in6_addr addr;
+	struct in6_addr mask;
+	u32 secid;
+
+	u32 valid;
+	struct list_head list;
+	struct rcu_head rcu;
+};
+struct netlbl_unlhsh_iface {
+	int ifindex;
+	struct list_head addr4_list;
+	struct list_head addr6_list;
+
+	u32 valid;
+	struct list_head list;
+	struct rcu_head rcu;
+};
+
+/* Argument struct for netlbl_unlhsh_walk() */
+struct netlbl_unlhsh_walk_arg {
+	struct netlink_callback *nl_cb;
+	struct sk_buff *skb;
+	u32 seq;
+};
+
+/* Unlabeled connection hash table */
+/* updates should be so rare that having one spinlock for the entire
+ * hash table should be okay */
+static DEFINE_SPINLOCK(netlbl_unlhsh_lock);
+static struct netlbl_unlhsh_tbl *netlbl_unlhsh = NULL;
+static struct netlbl_unlhsh_iface *netlbl_unlhsh_def = NULL;
 
 /* Accept unlabeled packets flag */
 static u8 netlabel_unlabel_acceptflg = 0;
 
-/* NetLabel Generic NETLINK CIPSOv4 family */
+/* NetLabel Generic NETLINK unlabeled family */
 static struct genl_family netlbl_unlabel_gnl_family = {
 	.id = GENL_ID_GENERATE,
 	.hdrsize = 0,
@@ -61,11 +133,681 @@ static struct genl_family netlbl_unlabel_gnl_family = {
 /* NetLabel Netlink attribute policy */
 static const struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1] = {
 	[NLBL_UNLABEL_A_ACPTFLG] = { .type = NLA_U8 },
+	[NLBL_UNLABEL_A_IPV6ADDR] = { .type = NLA_BINARY,
+				      .len = sizeof(struct in6_addr) },
+	[NLBL_UNLABEL_A_IPV6MASK] = { .type = NLA_BINARY,
+				      .len = sizeof(struct in6_addr) },
+	[NLBL_UNLABEL_A_IPV4ADDR] = { .type = NLA_BINARY,
+				      .len = sizeof(struct in_addr) },
+	[NLBL_UNLABEL_A_IPV4MASK] = { .type = NLA_BINARY,
+				      .len = sizeof(struct in_addr) },
+	[NLBL_UNLABEL_A_IFACE] = { .type = NLA_NUL_STRING,
+				   .len = IFNAMSIZ - 1 },
+	[NLBL_UNLABEL_A_SECCTX] = { .type = NLA_BINARY }
 };
 
 /*
- * Helper Functions
+ * Unlabeled Connection Hash Table Functions
+ */
+
+/**
+ * netlbl_unlhsh_free_addr4 - Frees an IPv4 address entry from the hash table
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that memory allocated to a hash table address entry can be
+ * released safely.
+ *
+ */
+static void netlbl_unlhsh_free_addr4(struct rcu_head *entry)
+{
+	struct netlbl_unlhsh_addr4 *ptr;
+
+	ptr = container_of(entry, struct netlbl_unlhsh_addr4, rcu);
+	kfree(ptr);
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_unlhsh_free_addr6 - Frees an IPv6 address entry from the hash table
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that memory allocated to a hash table address entry can be
+ * released safely.
+ *
+ */
+static void netlbl_unlhsh_free_addr6(struct rcu_head *entry)
+{
+	struct netlbl_unlhsh_addr6 *ptr;
+
+	ptr = container_of(entry, struct netlbl_unlhsh_addr6, rcu);
+	kfree(ptr);
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_unlhsh_free_iface - Frees an interface entry from the hash table
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that memory allocated to a hash table interface entry can be
+ * released safely.  It is important to note that this function does not free
+ * the IPv4 and IPv6 address lists contained as part of an interface entry.  It
+ * is up to the rest of the code to make sure an interface entry is only freed
+ * once it's address lists are empty.
+ *
+ */
+static void netlbl_unlhsh_free_iface(struct rcu_head *entry)
+{
+	struct netlbl_unlhsh_iface *iface;
+	struct netlbl_unlhsh_addr4 *iter4;
+	struct netlbl_unlhsh_addr4 *tmp4;
+	struct netlbl_unlhsh_addr6 *iter6;
+	struct netlbl_unlhsh_addr6 *tmp6;
+
+	iface = container_of(entry, struct netlbl_unlhsh_iface, rcu);
+
+	/* no need for locks here since we are the only one with access to this
+	 * structure */
+
+	list_for_each_entry_safe(iter4, tmp4, &iface->addr4_list, list)
+		if (iter4->valid) {
+			list_del_rcu(&iter4->list);
+			kfree(iter4);
+		}
+	list_for_each_entry_safe(iter6, tmp6, &iface->addr6_list, list)
+		if (iter6->valid) {
+			list_del_rcu(&iter6->list);
+			kfree(iter6);
+		}
+	kfree(iface);
+}
+
+/**
+ * netlbl_unlhsh_hash - Hashing function for the hash table
+ * @ifindex: the network interface/device to hash
+ *
+ * Description:
+ * This is the hashing function for the unlabeled hash table, it returns the
+ * bucket number for the given device/interface.  The caller is responsible for
+ * calling the rcu_read_[un]lock() functions.
+ *
+ */
+static u32 netlbl_unlhsh_hash(int ifindex)
+{
+	/* this is taken _almost_ directly from
+	 * security/selinux/netif.c:sel_netif_hasfn() as they do pretty much
+	 * the same thing */
+	return ifindex & (rcu_dereference(netlbl_unlhsh)->size - 1);
+}
+
+/**
+ * netlbl_unlhsh_search_addr4 - Search for a matching IPv4 address entry
+ * @addr: IPv4 address
+ * @iface: the network interface entry
+ *
+ * Description:
+ * Searches the IPv4 address list of the network interface specified by @iface.
+ * If a matching address entry is found it is returned, otherwise NULL is
+ * returned.  The caller is responsible for calling the rcu_read_[un]lock()
+ * functions.
+ *
+ */
+static struct netlbl_unlhsh_addr4 *netlbl_unlhsh_search_addr4(
+	                               __be32 addr,
+	                               const struct netlbl_unlhsh_iface *iface)
+{
+	struct netlbl_unlhsh_addr4 *iter;
+
+	list_for_each_entry_rcu(iter, &iface->addr4_list, list)
+		if (iter->valid && (addr & iter->mask) == iter->addr)
+			return iter;
+
+	return NULL;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_unlhsh_search_addr6 - Search for a matching IPv6 address entry
+ * @addr: IPv6 address
+ * @iface: the network interface entry
+ *
+ * Description:
+ * Searches the IPv6 address list of the network interface specified by @iface.
+ * If a matching address entry is found it is returned, otherwise NULL is
+ * returned.  The caller is responsible for calling the rcu_read_[un]lock()
+ * functions.
+ *
+ */
+static struct netlbl_unlhsh_addr6 *netlbl_unlhsh_search_addr6(
+	                               const struct in6_addr *addr,
+	                               const struct netlbl_unlhsh_iface *iface)
+{
+	struct netlbl_unlhsh_addr6 *iter;
+
+	list_for_each_entry_rcu(iter, &iface->addr6_list, list)
+		if (iter->valid &&
+		    ipv6_masked_addr_cmp(&iter->addr, &iter->mask, addr) == 0)
+		return iter;
+
+	return NULL;
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_unlhsh_search_iface - Search for a matching interface entry
+ * @ifindex: the network interface
+ *
+ * Description:
+ * Searches the unlabeled connection hash table and returns a pointer to the
+ * interface entry which matches @ifindex, otherwise NULL is returned.  The
+ * caller is responsible for calling the rcu_read_[un]lock() functions.
+ *
+ */
+static struct netlbl_unlhsh_iface *netlbl_unlhsh_search_iface(int ifindex)
+{
+	u32 bkt;
+	struct netlbl_unlhsh_iface *iter;
+
+	bkt = netlbl_unlhsh_hash(ifindex);
+	list_for_each_entry_rcu(iter,
+				&rcu_dereference(netlbl_unlhsh)->tbl[bkt],
+				list)
+		if (iter->valid && iter->ifindex == ifindex)
+			return iter;
+
+	return NULL;
+}
+
+/**
+ * netlbl_unlhsh_search_iface_def - Search for a matching interface entry
+ * @ifindex: the network interface
+ *
+ * Description:
+ * Searches the unlabeled connection hash table and returns a pointer to the
+ * interface entry which matches @ifindex.  If an exact match can not be found
+ * and there is a valid default entry, the default entry is returned, otherwise
+ * NULL is returned.  The caller is responsible for calling the
+ * rcu_read_[un]lock() functions.
+ *
+ */
+static struct netlbl_unlhsh_iface *netlbl_unlhsh_search_iface_def(int ifindex)
+{
+	struct netlbl_unlhsh_iface *entry;
+
+	entry = netlbl_unlhsh_search_iface(ifindex);
+	if (entry != NULL)
+		return entry;
+
+	entry = rcu_dereference(netlbl_unlhsh_def);
+	if (entry != NULL && entry->valid)
+		return entry;
+
+	return NULL;
+}
+
+/**
+ * netlbl_unlhsh_add_addr4 - Add a new IPv4 address entry to the hash table
+ * @iface: the associated interface entry
+ * @addr: IPv4 address in network byte order
+ * @mask: IPv4 address mask in network byte order
+ * @secid: LSM secid value for entry
+ *
+ * Description:
+ * Add a new address entry into the unlabeled connection hash table using the
+ * interface entry specified by @iface.  On success zero is returned, otherwise
+ * a negative value is returned.  The caller is responsible for calling the
+ * rcu_read_[un]lock() functions.
+ *
  */
+static int netlbl_unlhsh_add_addr4(struct netlbl_unlhsh_iface *iface,
+				   const struct in_addr *addr,
+				   const struct in_addr *mask,
+				   u32 secid)
+{
+	struct netlbl_unlhsh_addr4 *entry;
+	struct netlbl_unlhsh_addr4 *iter;
+
+	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
+	if (entry == NULL)
+		return -ENOMEM;
+
+	entry->addr = addr->s_addr & mask->s_addr;
+	entry->mask = mask->s_addr;
+	entry->secid = secid;
+	entry->valid = 1;
+	INIT_RCU_HEAD(&entry->rcu);
+
+	spin_lock(&netlbl_unlhsh_lock);
+	iter = netlbl_unlhsh_search_addr4(entry->addr, iface);
+	if (iter != NULL &&
+	    iter->addr == addr->s_addr && iter->mask == mask->s_addr) {
+		spin_unlock(&netlbl_unlhsh_lock);
+		kfree(entry);
+		return -EEXIST;
+	}
+	/* in order to speed up address searches through the list (the common
+	 * case) we need to keep the list in order based on the size of the
+	 * address mask such that the entry with the widest mask (smallest
+	 * numerical value) appears first in the list */
+	list_for_each_entry_rcu(iter, &iface->addr4_list, list)
+		if (iter->valid &&
+		    ntohl(entry->mask) > ntohl(iter->mask)) {
+			__list_add_rcu(&entry->list,
+				       iter->list.prev,
+				       &iter->list);
+			spin_unlock(&netlbl_unlhsh_lock);
+			return 0;
+		}
+	list_add_tail_rcu(&entry->list, &iface->addr4_list);
+	spin_unlock(&netlbl_unlhsh_lock);
+	return 0;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_unlhsh_add_addr6 - Add a new IPv6 address entry to the hash table
+ * @iface: the associated interface entry
+ * @addr: IPv6 address in network byte order
+ * @mask: IPv6 address mask in network byte order
+ * @secid: LSM secid value for entry
+ *
+ * Description:
+ * Add a new address entry into the unlabeled connection hash table using the
+ * interface entry specified by @iface.  On success zero is returned, otherwise
+ * a negative value is returned.  The caller is responsible for calling the
+ * rcu_read_[un]lock() functions.
+ *
+ */
+static int netlbl_unlhsh_add_addr6(struct netlbl_unlhsh_iface *iface,
+				   const struct in6_addr *addr,
+				   const struct in6_addr *mask,
+				   u32 secid)
+{
+	struct netlbl_unlhsh_addr6 *entry;
+	struct netlbl_unlhsh_addr6 *iter;
+
+	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
+	if (entry == NULL)
+		return -ENOMEM;
+
+	ipv6_addr_copy(&entry->addr, addr);
+	entry->addr.s6_addr32[0] &= mask->s6_addr32[0];
+	entry->addr.s6_addr32[1] &= mask->s6_addr32[1];
+	entry->addr.s6_addr32[2] &= mask->s6_addr32[2];
+	entry->addr.s6_addr32[3] &= mask->s6_addr32[3];
+	ipv6_addr_copy(&entry->mask, mask);
+	entry->secid = secid;
+	entry->valid = 1;
+	INIT_RCU_HEAD(&entry->rcu);
+
+	spin_lock(&netlbl_unlhsh_lock);
+	iter = netlbl_unlhsh_search_addr6(&entry->addr, iface);
+	if (iter != NULL &&
+	    (ipv6_addr_equal(&iter->addr, addr) &&
+	     ipv6_addr_equal(&iter->mask, mask))) {
+		spin_unlock(&netlbl_unlhsh_lock);
+		kfree(entry);
+		return -EEXIST;
+	}
+	/* in order to speed up address searches through the list (the common
+	 * case) we need to keep the list in order based on the size of the
+	 * address mask such that the entry with the widest mask (smallest
+	 * numerical value) appears first in the list */
+	list_for_each_entry_rcu(iter, &iface->addr6_list, list)
+		if (iter->valid &&
+		    ipv6_addr_cmp(&entry->mask, &iter->mask) > 0) {
+			__list_add_rcu(&entry->list,
+				       iter->list.prev,
+				       &iter->list);
+			spin_unlock(&netlbl_unlhsh_lock);
+			return 0;
+		}
+	list_add_tail_rcu(&entry->list, &iface->addr6_list);
+	spin_unlock(&netlbl_unlhsh_lock);
+	return 0;
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_unlhsh_add_iface - Adds a new interface entry to the hash table
+ * @ifindex: network interface
+ *
+ * Description:
+ * Add a new, empty, interface entry into the unlabeled connection hash table.
+ * On success a pointer to the new interface entry is returned, on failure NULL
+ * is returned.  The caller is responsible for calling the rcu_read_[un]lock()
+ * functions.
+ *
+ */
+static struct netlbl_unlhsh_iface *netlbl_unlhsh_add_iface(int ifindex)
+{
+	u32 bkt;
+	struct netlbl_unlhsh_iface *iface;
+
+	iface = kzalloc(sizeof(*iface), GFP_ATOMIC);
+	if (iface == NULL)
+		return NULL;
+
+	iface->ifindex = ifindex;
+	INIT_LIST_HEAD(&iface->addr4_list);
+	INIT_LIST_HEAD(&iface->addr6_list);
+	iface->valid = 1;
+	INIT_RCU_HEAD(&iface->rcu);
+
+	spin_lock(&netlbl_unlhsh_lock);
+	if (ifindex > 0) {
+		bkt = netlbl_unlhsh_hash(ifindex);
+		if (netlbl_unlhsh_search_iface(ifindex) != NULL)
+			goto add_iface_failure;
+		list_add_tail_rcu(&iface->list,
+				  &rcu_dereference(netlbl_unlhsh)->tbl[bkt]);
+	} else {
+		INIT_LIST_HEAD(&iface->list);
+		if (rcu_dereference(netlbl_unlhsh_def) != NULL)
+			goto add_iface_failure;
+		rcu_assign_pointer(netlbl_unlhsh_def, iface);
+	}
+	spin_unlock(&netlbl_unlhsh_lock);
+
+	return iface;
+
+add_iface_failure:
+	spin_unlock(&netlbl_unlhsh_lock);
+	kfree(iface);
+	return NULL;
+}
+
+/**
+ * netlbl_unlhsh_add - Adds a new entry to the unlabeled connection hash table
+ * @net: network namespace
+ * @dev_name: interface name
+ * @addr: IP address in network byte order
+ * @mask: address mask in network byte order
+ * @addr_len: length of address/mask (4 for IPv4, 16 for IPv6)
+ * @secid: LSM secid value for the entry
+ *
+ * Description:
+ * Adds a new entry to the unlabeled connection hash table.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int netlbl_unlhsh_add(struct net *net,
+			     const char *dev_name,
+			     const void *addr,
+			     const void *mask,
+			     u32 addr_len,
+			     u32 secid)
+{
+	int ret_val;
+	int ifindex;
+	struct net_device *dev;
+	struct netlbl_unlhsh_iface *iface;
+
+	if (addr_len != sizeof(struct in_addr) &&
+	    addr_len != sizeof(struct in6_addr))
+		return -EINVAL;
+
+	rcu_read_lock();
+	if (dev_name != NULL) {
+		dev = dev_get_by_name(net, dev_name);
+		if (dev == NULL) {
+			ret_val = -ENODEV;
+			goto unlhsh_add_return;
+		}
+		ifindex = dev->ifindex;
+		dev_put(dev);
+		iface = netlbl_unlhsh_search_iface(ifindex);
+	} else {
+		ifindex = 0;
+		iface = rcu_dereference(netlbl_unlhsh_def);
+	}
+	if (iface == NULL) {
+		iface = netlbl_unlhsh_add_iface(ifindex);
+		if (iface == NULL) {
+			ret_val = -ENOMEM;
+			goto unlhsh_add_return;
+		}
+	}
+	switch (addr_len) {
+	case sizeof(struct in_addr):
+		ret_val = netlbl_unlhsh_add_addr4(iface, addr, mask, secid);
+		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	case sizeof(struct in6_addr):
+		ret_val = netlbl_unlhsh_add_addr6(iface, addr, mask, secid);
+		break;
+#endif /* IPv6 */
+	default:
+		ret_val = -EINVAL;
+	}
+	if (ret_val == 0)
+		atomic_inc(&netlabel_mgmt_protocount);
+
+unlhsh_add_return:
+	rcu_read_unlock();
+	return ret_val;
+}
+
+/**
+ * netlbl_unlhsh_remove_addr4 - Remove an IPv4 address entry
+ * @iface: interface entry
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Remove an IP address entry from the unlabeled connection hash table.
+ * Returns zero on success, negative values on failure.  The caller is
+ * responsible for calling the rcu_read_[un]lock() functions.
+ *
+ */
+static int netlbl_unlhsh_remove_addr4(struct netlbl_unlhsh_iface *iface,
+				      const struct in_addr *addr,
+				      const struct in_addr *mask)
+{
+	int ret_val = -ENOENT;
+	struct netlbl_unlhsh_addr4 *entry;
+
+	spin_lock(&netlbl_unlhsh_lock);
+	entry = netlbl_unlhsh_search_addr4(addr->s_addr, iface);
+	if (entry != NULL &&
+	    entry->addr == addr->s_addr && entry->mask == mask->s_addr) {
+		entry->valid = 0;
+		list_del_rcu(&entry->list);
+		ret_val = 0;
+	}
+	spin_unlock(&netlbl_unlhsh_lock);
+
+	if (ret_val == 0)
+		call_rcu(&entry->rcu, netlbl_unlhsh_free_addr4);
+	return ret_val;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_unlhsh_remove_addr6 - Remove an IPv6 address entry
+ * @iface: interface entry
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Remove an IP address entry from the unlabeled connection hash table.
+ * Returns zero on success, negative values on failure.  The caller is
+ * responsible for calling the rcu_read_[un]lock() functions.
+ *
+ */
+static int netlbl_unlhsh_remove_addr6(struct netlbl_unlhsh_iface *iface,
+				      const struct in6_addr *addr,
+				      const struct in6_addr *mask)
+{
+	int ret_val = -ENOENT;
+	struct netlbl_unlhsh_addr6 *entry;
+
+	spin_lock(&netlbl_unlhsh_lock);
+	entry = netlbl_unlhsh_search_addr6(addr, iface);
+	if (entry != NULL &&
+	    (ipv6_addr_equal(&entry->addr, addr) &&
+	     ipv6_addr_equal(&entry->mask, mask))) {
+		entry->valid = 0;
+		list_del_rcu(&entry->list);
+		ret_val = 0;
+	}
+	spin_unlock(&netlbl_unlhsh_lock);
+
+	if (ret_val == 0)
+		call_rcu(&entry->rcu, netlbl_unlhsh_free_addr6);
+	return ret_val;
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_unlhsh_condremove_iface - Remove an interface entry
+ * @iface: the interface entry
+ *
+ * Description:
+ * Remove an interface entry from the unlabeled connection hash table if it is
+ * empty.  An interface entry is considered to be empty if there are no
+ * address entries assigned to it.
+ *
+ */
+static void netlbl_unlhsh_condremove_iface(struct netlbl_unlhsh_iface *iface)
+{
+	struct netlbl_unlhsh_addr4 *iter4;
+	struct netlbl_unlhsh_addr6 *iter6;
+
+	spin_lock(&netlbl_unlhsh_lock);
+	list_for_each_entry_rcu(iter4, &iface->addr4_list, list)
+		if (iter4->valid)
+			goto unlhsh_condremove_failure;
+	list_for_each_entry_rcu(iter6, &iface->addr6_list, list)
+		if (iter6->valid)
+			goto unlhsh_condremove_failure;
+	iface->valid = 0;
+	if (iface->ifindex > 0)
+		list_del_rcu(&iface->list);
+	else
+		rcu_assign_pointer(netlbl_unlhsh_def, NULL);
+	spin_unlock(&netlbl_unlhsh_lock);
+
+	call_rcu(&iface->rcu, netlbl_unlhsh_free_iface);
+	return;
+
+unlhsh_condremove_failure:
+	spin_unlock(&netlbl_unlhsh_lock);
+	return;
+}
+
+/**
+ * netlbl_unlhsh_remove - Remove an entry from the unlabeled hash table
+ * @net: network namespace
+ * @dev_name: interface name
+ * @addr: IP address in network byte order
+ * @mask: address mask in network byte order
+ * @addr_len: length of address/mask (4 for IPv4, 16 for IPv6)
+ *
+ * Description:
+ * Removes and existing entry from the unlabeled connection hash table.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_unlhsh_remove(struct net *net,
+				const char *dev_name,
+				const void *addr,
+				const void *mask,
+				u32 addr_len)
+{
+	int ret_val;
+	struct net_device *dev;
+	struct netlbl_unlhsh_iface *iface;
+
+	if (addr_len != sizeof(struct in_addr) &&
+	    addr_len != sizeof(struct in6_addr))
+		return -EINVAL;
+
+	rcu_read_lock();
+	if (dev_name != NULL) {
+		dev = dev_get_by_name(net, dev_name);
+		if (dev == NULL) {
+			ret_val = -ENODEV;
+			goto unlhsh_remove_return;
+		}
+		iface = netlbl_unlhsh_search_iface(dev->ifindex);
+		dev_put(dev);
+	} else
+		iface = rcu_dereference(netlbl_unlhsh_def);
+	if (iface == NULL) {
+		ret_val = -ENOENT;
+		goto unlhsh_remove_return;
+	}
+	switch (addr_len) {
+	case sizeof(struct in_addr):
+		ret_val = netlbl_unlhsh_remove_addr4(iface, addr, mask);
+		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	case sizeof(struct in6_addr):
+		ret_val = netlbl_unlhsh_remove_addr6(iface, addr, mask);
+		break;
+#endif /* IPv6 */
+	default:
+		ret_val = -EINVAL;
+	}
+	if (ret_val == 0) {
+		netlbl_unlhsh_condremove_iface(iface);
+		atomic_dec(&netlabel_mgmt_protocount);
+	}
+
+unlhsh_remove_return:
+	rcu_read_unlock();
+	return ret_val;
+}
+
+/*
+ * General Helper Functions
+ */
+
+/**
+ * netlbl_unlhsh_netdev_handler - Network device notification handler
+ * @this: notifier block
+ * @event: the event
+ * @ptr: the network device (cast to void)
+ *
+ * Description:
+ * Handle network device events, although at present all we care about is a
+ * network device going away.  In the case of a device going away we clear any
+ * related entries from the unlabeled connection hash table.
+ *
+ */
+static int netlbl_unlhsh_netdev_handler(struct notifier_block *this,
+					unsigned long event,
+					void *ptr)
+{
+	struct net_device *dev = ptr;
+	struct netlbl_unlhsh_iface *iface = NULL;
+
+	if (dev->nd_net != &init_net)
+		return NOTIFY_DONE;
+
+	/* XXX - should this be a check for NETDEV_DOWN or _UNREGISTER? */
+	if (event == NETDEV_DOWN) {
+		spin_lock(&netlbl_unlhsh_lock);
+		iface = netlbl_unlhsh_search_iface(dev->ifindex);
+		if (iface != NULL && iface->valid) {
+			iface->valid = 0;
+			list_del_rcu(&iface->list);
+		} else
+			iface = NULL;
+		spin_unlock(&netlbl_unlhsh_lock);
+	}
+
+	if (iface != NULL)
+		call_rcu(&iface->rcu, netlbl_unlhsh_free_iface);
+
+	return NOTIFY_DONE;
+}
 
 /**
  * netlbl_unlabel_acceptflg_set - Set the unlabeled accept flag
@@ -93,6 +835,48 @@ static void netlbl_unlabel_acceptflg_set(u8 value,
 	}
 }
 
+/**
+ * netlbl_unlabel_addrinfo_get - Get the IPv4/6 address information
+ * @info: the Generic NETLINK info block
+ * @addr: the IP address
+ * @mask: the IP address mask
+ * @len: the address length
+ *
+ * Description:
+ * Examine the Generic NETLINK message and extract the IP address information.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_unlabel_addrinfo_get(struct genl_info *info,
+				       void **addr,
+				       void **mask,
+				       u32 *len)
+{
+	u32 addr_len;
+
+	if (info->attrs[NLBL_UNLABEL_A_IPV4ADDR]) {
+		addr_len = nla_len(info->attrs[NLBL_UNLABEL_A_IPV4ADDR]);
+		if (addr_len != sizeof(struct in_addr) &&
+		    addr_len != nla_len(info->attrs[NLBL_UNLABEL_A_IPV4MASK]))
+			return -EINVAL;
+		*len = addr_len;
+		*addr = nla_data(info->attrs[NLBL_UNLABEL_A_IPV4ADDR]);
+		*mask = nla_data(info->attrs[NLBL_UNLABEL_A_IPV4MASK]);
+		return 0;
+	} else if (info->attrs[NLBL_UNLABEL_A_IPV6ADDR]) {
+		addr_len = nla_len(info->attrs[NLBL_UNLABEL_A_IPV6ADDR]);
+		if (addr_len != sizeof(struct in6_addr) &&
+		    addr_len != nla_len(info->attrs[NLBL_UNLABEL_A_IPV6MASK]))
+			return -EINVAL;
+		*len = addr_len;
+		*addr = nla_data(info->attrs[NLBL_UNLABEL_A_IPV6ADDR]);
+		*mask = nla_data(info->attrs[NLBL_UNLABEL_A_IPV6MASK]);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
 /*
  * NetLabel Command Handlers
  */
@@ -168,11 +952,470 @@ list_failure:
 	return ret_val;
 }
 
+/**
+ * netlbl_unlabel_staticadd - Handle a STATICADD message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated STATICADD message and add a new unlabeled
+ * connection entry to the hash table.  Returns zero on success, negative
+ * values on failure.
+ *
+ */
+static int netlbl_unlabel_staticadd(struct sk_buff *skb,
+				    struct genl_info *info)
+{
+	int ret_val;
+	char *dev_name;
+	void *addr;
+	void *mask;
+	u32 addr_len;
+	u32 secid;
+
+	/* Don't allow users to add both IPv4 and IPv6 addresses for a
+	 * single entry.  However, allow users to create two entries, one each
+	 * for IPv4 and IPv4, with the same LSM security context which should
+	 * achieve the same result. */
+	if (!info->attrs[NLBL_UNLABEL_A_SECCTX] ||
+	    !info->attrs[NLBL_UNLABEL_A_IFACE] ||
+	    !((!info->attrs[NLBL_UNLABEL_A_IPV4ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV4MASK]) ^
+	      (!info->attrs[NLBL_UNLABEL_A_IPV6ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
+		return -EINVAL;
+
+	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
+	if (ret_val != 0)
+		return ret_val;
+	dev_name = nla_data(info->attrs[NLBL_UNLABEL_A_IFACE]);
+	ret_val = security_secctx_to_secid(
+		                  nla_data(info->attrs[NLBL_UNLABEL_A_SECCTX]),
+				  nla_len(info->attrs[NLBL_UNLABEL_A_SECCTX]),
+				  &secid);
+	if (ret_val != 0)
+		return ret_val;
+
+	return netlbl_unlhsh_add(&init_net,
+				 dev_name, addr, mask, addr_len, secid);
+}
+
+/**
+ * netlbl_unlabel_staticadddef - Handle a STATICADDDEF message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated STATICADDDEF message and add a new default
+ * unlabeled connection entry.  Returns zero on success, negative values on
+ * failure.
+ *
+ */
+static int netlbl_unlabel_staticadddef(struct sk_buff *skb,
+				       struct genl_info *info)
+{
+	int ret_val;
+	void *addr;
+	void *mask;
+	u32 addr_len;
+	u32 secid;
+
+	/* Don't allow users to add both IPv4 and IPv6 addresses for a
+	 * single entry.  However, allow users to create two entries, one each
+	 * for IPv4 and IPv6, with the same LSM security context which should
+	 * achieve the same result. */
+	if (!info->attrs[NLBL_UNLABEL_A_SECCTX] ||
+	    !((!info->attrs[NLBL_UNLABEL_A_IPV4ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV4MASK]) ^
+	      (!info->attrs[NLBL_UNLABEL_A_IPV6ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
+		return -EINVAL;
+
+	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
+	if (ret_val != 0)
+		return ret_val;
+	ret_val = security_secctx_to_secid(
+		                  nla_data(info->attrs[NLBL_UNLABEL_A_SECCTX]),
+				  nla_len(info->attrs[NLBL_UNLABEL_A_SECCTX]),
+				  &secid);
+	if (ret_val != 0)
+		return ret_val;
+
+	return netlbl_unlhsh_add(&init_net, NULL, addr, mask, addr_len, secid);
+}
+
+/**
+ * netlbl_unlabel_staticremove - Handle a STATICREMOVE message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated STATICREMOVE message and remove the specified
+ * unlabeled connection entry.  Returns zero on success, negative values on
+ * failure.
+ *
+ */
+static int netlbl_unlabel_staticremove(struct sk_buff *skb,
+				       struct genl_info *info)
+{
+	int ret_val;
+	char *dev_name;
+	void *addr;
+	void *mask;
+	u32 addr_len;
+
+	/* See the note in netlbl_unlabel_staticadd() about not allowing both
+	 * IPv4 and IPv6 in the same entry. */
+	if (!info->attrs[NLBL_UNLABEL_A_IFACE] ||
+	    !((!info->attrs[NLBL_UNLABEL_A_IPV4ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV4MASK]) ^
+	      (!info->attrs[NLBL_UNLABEL_A_IPV6ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
+		return -EINVAL;
+
+	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
+	if (ret_val != 0)
+		return ret_val;
+	dev_name = nla_data(info->attrs[NLBL_UNLABEL_A_IFACE]);
+
+	return netlbl_unlhsh_remove(&init_net, dev_name, addr, mask, addr_len);
+}
+
+/**
+ * netlbl_unlabel_staticremovedef - Handle a STATICREMOVEDEF message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated STATICREMOVEDEF message and remove the default
+ * unlabeled connection entry.  Returns zero on success, negative values on
+ * failure.
+ *
+ */
+static int netlbl_unlabel_staticremovedef(struct sk_buff *skb,
+					  struct genl_info *info)
+{
+	int ret_val;
+	void *addr;
+	void *mask;
+	u32 addr_len;
+
+	/* See the note in netlbl_unlabel_staticadd() about not allowing both
+	 * IPv4 and IPv6 in the same entry. */
+	if (!((!info->attrs[NLBL_UNLABEL_A_IPV4ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV4MASK]) ^
+	      (!info->attrs[NLBL_UNLABEL_A_IPV6ADDR] ||
+	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
+		return -EINVAL;
+
+	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
+	if (ret_val != 0)
+		return ret_val;
+
+	return netlbl_unlhsh_remove(&init_net, NULL, addr, mask, addr_len);
+}
+
+
+/**
+ * netlbl_unlabel_staticlist_gen - Generate messages for STATICLIST[DEF]
+ * @cmd: command/message
+ * @iface: the interface entry
+ * @addr4: the IPv4 address entry
+ * @addr6: the IPv6 address entry
+ * @arg: the netlbl_unlhsh_walk_arg structure
+ *
+ * Description:
+ * This function is designed to be used to generate a response for a
+ * STATICLIST or STATICLISTDEF message.  When called either @addr4 or @addr6
+ * can be specified, not both, the other unspecified entry should be set to
+ * NULL by the caller.  Returns the size of the message on success, negative
+ * values on failure.
+ *
+ */
+static int netlbl_unlabel_staticlist_gen(u32 cmd,
+				       const struct netlbl_unlhsh_iface *iface,
+				       const struct netlbl_unlhsh_addr4 *addr4,
+				       const struct netlbl_unlhsh_addr6 *addr6,
+				       void *arg)
+{
+	int ret_val = -ENOMEM;
+	struct netlbl_unlhsh_walk_arg *cb_arg = arg;
+	struct net_device *dev;
+	void *data;
+	u32 secid;
+	char *secctx;
+	u32 secctx_len;
+
+	data = genlmsg_put(cb_arg->skb, NETLINK_CB(cb_arg->nl_cb->skb).pid,
+			   cb_arg->seq, &netlbl_unlabel_gnl_family,
+			   NLM_F_MULTI, cmd);
+	if (data == NULL)
+		goto list_cb_failure;
+
+	if (iface->ifindex > 0) {
+		dev = dev_get_by_index(&init_net, iface->ifindex);
+		ret_val = nla_put_string(cb_arg->skb,
+					 NLBL_UNLABEL_A_IFACE, dev->name);
+		dev_put(dev);
+		if (ret_val != 0)
+			goto list_cb_failure;
+	}
+
+	if (addr4) {
+		struct in_addr addr_struct;
+
+		addr_struct.s_addr = addr4->addr;
+		ret_val = nla_put(cb_arg->skb,
+				  NLBL_UNLABEL_A_IPV4ADDR,
+				  sizeof(struct in_addr),
+				  &addr_struct);
+		if (ret_val != 0)
+			goto list_cb_failure;
+
+		addr_struct.s_addr = addr4->mask;
+		ret_val = nla_put(cb_arg->skb,
+				  NLBL_UNLABEL_A_IPV4MASK,
+				  sizeof(struct in_addr),
+				  &addr_struct);
+		if (ret_val != 0)
+			goto list_cb_failure;
+
+		secid = addr4->secid;
+	} else {
+		ret_val = nla_put(cb_arg->skb,
+				  NLBL_UNLABEL_A_IPV6ADDR,
+				  sizeof(struct in6_addr),
+				  &addr6->addr);
+		if (ret_val != 0)
+			goto list_cb_failure;
+
+		ret_val = nla_put(cb_arg->skb,
+				  NLBL_UNLABEL_A_IPV6MASK,
+				  sizeof(struct in6_addr),
+				  &addr6->mask);
+		if (ret_val != 0)
+			goto list_cb_failure;
+
+		secid = addr6->secid;
+	}
+
+	ret_val = security_secid_to_secctx(secid, &secctx, &secctx_len);
+	if (ret_val != 0)
+		goto list_cb_failure;
+	ret_val = nla_put(cb_arg->skb,
+			  NLBL_UNLABEL_A_SECCTX,
+			  secctx_len,
+			  secctx);
+	security_release_secctx(secctx, secctx_len);
+	if (ret_val != 0)
+		goto list_cb_failure;
+
+	cb_arg->seq++;
+	return genlmsg_end(cb_arg->skb, data);
+
+list_cb_failure:
+	genlmsg_cancel(cb_arg->skb, data);
+	return ret_val;
+}
+
+/**
+ * netlbl_unlabel_staticlist - Handle a STATICLIST message
+ * @skb: the NETLINK buffer
+ * @cb: the NETLINK callback
+ *
+ * Description:
+ * Process a user generated STATICLIST message and dump the unlabeled
+ * connection hash table in a form suitable for use in a kernel generated
+ * STATICLIST message.  Returns the length of @skb.
+ *
+ */
+static int netlbl_unlabel_staticlist(struct sk_buff *skb,
+				     struct netlink_callback *cb)
+{
+	struct netlbl_unlhsh_walk_arg cb_arg;
+	u32 skip_bkt = cb->args[0];
+	u32 skip_chain = cb->args[1];
+	u32 skip_addr4 = cb->args[2];
+	u32 skip_addr6 = cb->args[3];
+	u32 iter_bkt;
+	u32 iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0;
+	struct netlbl_unlhsh_iface *iface;
+	struct netlbl_unlhsh_addr4 *addr4;
+	struct netlbl_unlhsh_addr6 *addr6;
+
+	cb_arg.nl_cb = cb;
+	cb_arg.skb = skb;
+	cb_arg.seq = cb->nlh->nlmsg_seq;
+
+	rcu_read_lock();
+	for (iter_bkt = skip_bkt;
+	     iter_bkt < rcu_dereference(netlbl_unlhsh)->size;
+	     iter_bkt++, iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0) {
+		list_for_each_entry_rcu(iface,
+			        &rcu_dereference(netlbl_unlhsh)->tbl[iter_bkt],
+				list) {
+			if (!iface->valid ||
+			    iter_chain++ < skip_chain)
+				continue;
+			list_for_each_entry_rcu(addr4,
+						&iface->addr4_list,
+						list) {
+				if (!addr4->valid || iter_addr4++ < skip_addr4)
+					continue;
+				if (netlbl_unlabel_staticlist_gen(
+					             NLBL_UNLABEL_C_STATICLIST,
+						     iface,
+						     addr4,
+						     NULL,
+						     &cb_arg) < 0) {
+					iter_addr4--;
+					iter_chain--;
+					goto unlabel_staticlist_return;
+				}
+			}
+			list_for_each_entry_rcu(addr6,
+						&iface->addr6_list,
+						list) {
+				if (!addr6->valid || iter_addr6++ < skip_addr6)
+					continue;
+				if (netlbl_unlabel_staticlist_gen(
+						     NLBL_UNLABEL_C_STATICLIST,
+						     iface,
+						     NULL,
+						     addr6,
+						     &cb_arg) < 0) {
+					iter_addr6--;
+					iter_chain--;
+					goto unlabel_staticlist_return;
+				}
+			}
+		}
+	}
+
+unlabel_staticlist_return:
+	rcu_read_unlock();
+	cb->args[0] = skip_bkt;
+	cb->args[1] = skip_chain;
+	cb->args[2] = skip_addr4;
+	cb->args[3] = skip_addr6;
+	return skb->len;
+}
+
+/**
+ * netlbl_unlabel_staticlistdef - Handle a STATICLISTDEF message
+ * @skb: the NETLINK buffer
+ * @cb: the NETLINK callback
+ *
+ * Description:
+ * Process a user generated STATICLISTDEF message and dump the default
+ * unlabeled connection entry in a form suitable for use in a kernel generated
+ * STATICLISTDEF message.  Returns the length of @skb.
+ *
+ */
+static int netlbl_unlabel_staticlistdef(struct sk_buff *skb,
+					struct netlink_callback *cb)
+{
+	struct netlbl_unlhsh_walk_arg cb_arg;
+	struct netlbl_unlhsh_iface *iface;
+	u32 skip_addr4 = cb->args[0];
+	u32 skip_addr6 = cb->args[1];
+	u32 iter_addr4 = 0, iter_addr6 = 0;
+	struct netlbl_unlhsh_addr4 *addr4;
+	struct netlbl_unlhsh_addr6 *addr6;
+
+	cb_arg.nl_cb = cb;
+	cb_arg.skb = skb;
+	cb_arg.seq = cb->nlh->nlmsg_seq;
+
+	rcu_read_lock();
+	iface = rcu_dereference(netlbl_unlhsh_def);
+	if (iface == NULL || !iface->valid)
+		goto unlabel_staticlistdef_return;
+
+	list_for_each_entry_rcu(addr4, &iface->addr4_list, list) {
+		if (!addr4->valid || iter_addr4++ < skip_addr4)
+			continue;
+		if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF,
+					   iface,
+					   addr4,
+					   NULL,
+					   &cb_arg) < 0) {
+			iter_addr4--;
+			goto unlabel_staticlistdef_return;
+		}
+	}
+	list_for_each_entry_rcu(addr6, &iface->addr6_list, list) {
+		if (addr6->valid || iter_addr6++ < skip_addr6)
+			continue;
+		if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF,
+					   iface,
+					   NULL,
+					   addr6,
+					   &cb_arg) < 0) {
+			iter_addr6--;
+			goto unlabel_staticlistdef_return;
+		}
+	}
+
+unlabel_staticlistdef_return:
+	rcu_read_unlock();
+	cb->args[0] = skip_addr4;
+	cb->args[1] = skip_addr6;
+	return skb->len;
+}
 
 /*
  * NetLabel Generic NETLINK Command Definitions
  */
 
+static struct genl_ops netlbl_unlabel_genl_c_staticadd = {
+	.cmd = NLBL_UNLABEL_C_STATICADD,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = netlbl_unlabel_staticadd,
+	.dumpit = NULL,
+};
+
+static struct genl_ops netlbl_unlabel_genl_c_staticremove = {
+	.cmd = NLBL_UNLABEL_C_STATICREMOVE,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = netlbl_unlabel_staticremove,
+	.dumpit = NULL,
+};
+
+static struct genl_ops netlbl_unlabel_genl_c_staticlist = {
+	.cmd = NLBL_UNLABEL_C_STATICLIST,
+	.flags = 0,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = NULL,
+	.dumpit = netlbl_unlabel_staticlist,
+};
+
+static struct genl_ops netlbl_unlabel_genl_c_staticadddef = {
+	.cmd = NLBL_UNLABEL_C_STATICADDDEF,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = netlbl_unlabel_staticadddef,
+	.dumpit = NULL,
+};
+
+static struct genl_ops netlbl_unlabel_genl_c_staticremovedef = {
+	.cmd = NLBL_UNLABEL_C_STATICREMOVEDEF,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = netlbl_unlabel_staticremovedef,
+	.dumpit = NULL,
+};
+
+static struct genl_ops netlbl_unlabel_genl_c_staticlistdef = {
+	.cmd = NLBL_UNLABEL_C_STATICLISTDEF,
+	.flags = 0,
+	.policy = netlbl_unlabel_genl_policy,
+	.doit = NULL,
+	.dumpit = netlbl_unlabel_staticlistdef,
+};
+
 static struct genl_ops netlbl_unlabel_genl_c_accept = {
 	.cmd = NLBL_UNLABEL_C_ACCEPT,
 	.flags = GENL_ADMIN_PERM,
@@ -189,7 +1432,6 @@ static struct genl_ops netlbl_unlabel_genl_c_list = {
 	.dumpit = NULL,
 };
 
-
 /*
  * NetLabel Generic NETLINK Protocol Functions
  */
@@ -210,6 +1452,36 @@ int netlbl_unlabel_genl_init(void)
 	if (ret_val != 0)
 		return ret_val;
 
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticadd);
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticremove);
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticlist);
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticadddef);
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticremovedef);
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+				    &netlbl_unlabel_genl_c_staticlistdef);
+	if (ret_val != 0)
+		return ret_val;
+
 	ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
 				    &netlbl_unlabel_genl_c_accept);
 	if (ret_val != 0)
@@ -227,8 +1499,58 @@ int netlbl_unlabel_genl_init(void)
  * NetLabel KAPI Hooks
  */
 
+static struct notifier_block netlbl_unlhsh_netdev_notifier = {
+	.notifier_call = netlbl_unlhsh_netdev_handler,
+};
+
+/**
+ * netlbl_unlabel_init - Initialize the unlabeled connection hash table
+ * @size: the number of bits to use for the hash buckets
+ *
+ * Description:
+ * Initializes the unlabeled connection hash table and registers a network
+ * device notification handler.  This function should only be called by the
+ * NetLabel subsystem itself during initialization.  Returns zero on success,
+ * non-zero values on error.
+ *
+ */
+int netlbl_unlabel_init(u32 size)
+{
+	u32 iter;
+	struct netlbl_unlhsh_tbl *hsh_tbl;
+
+	if (size == 0)
+		return -EINVAL;
+
+	hsh_tbl = kmalloc(sizeof(*hsh_tbl), GFP_KERNEL);
+	if (hsh_tbl == NULL)
+		return -ENOMEM;
+	hsh_tbl->size = 1 << size;
+	hsh_tbl->tbl = kcalloc(hsh_tbl->size,
+			       sizeof(struct list_head),
+			       GFP_KERNEL);
+	if (hsh_tbl->tbl == NULL) {
+		kfree(hsh_tbl);
+		return -ENOMEM;
+	}
+	for (iter = 0; iter < hsh_tbl->size; iter++)
+		INIT_LIST_HEAD(&hsh_tbl->tbl[iter]);
+
+	rcu_read_lock();
+	spin_lock(&netlbl_unlhsh_lock);
+	rcu_assign_pointer(netlbl_unlhsh, hsh_tbl);
+	spin_unlock(&netlbl_unlhsh_lock);
+	rcu_read_unlock();
+
+	register_netdevice_notifier(&netlbl_unlhsh_netdev_notifier);
+
+	return 0;
+}
+
 /**
  * netlbl_unlabel_getattr - Get the security attributes for an unlabled packet
+ * @skb: the packet
+ * @family: protocol family
  * @secattr: the security attributes
  *
  * Description:
@@ -236,11 +1558,50 @@ int netlbl_unlabel_genl_init(void)
  * them in @secattr.  Returns zero on success and negative values on failure.
  *
  */
-int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr)
+int netlbl_unlabel_getattr(const struct sk_buff *skb,
+			   u16 family,
+			   struct netlbl_lsm_secattr *secattr)
 {
+	struct iphdr *hdr4;
+	struct ipv6hdr *hdr6;
+	struct netlbl_unlhsh_addr4 *addr4;
+	struct netlbl_unlhsh_addr6 *addr6;
+	struct netlbl_unlhsh_iface *iface;
+
+	rcu_read_lock();
+	iface = netlbl_unlhsh_search_iface_def(skb->iif);
+	if (iface == NULL)
+		goto unlabel_getattr_nolabel;
+	switch (family) {
+	case PF_INET:
+		hdr4 = ip_hdr(skb);
+		addr4 = netlbl_unlhsh_search_addr4(hdr4->saddr, iface);
+		if (addr4 == NULL)
+			goto unlabel_getattr_nolabel;
+		secattr->attr.secid = addr4->secid;
+		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	case PF_INET6:
+		hdr6 = ipv6_hdr(skb);
+		addr6 = netlbl_unlhsh_search_addr6(&hdr6->saddr, iface);
+		if (addr6 == NULL)
+			goto unlabel_getattr_nolabel;
+		secattr->attr.secid = addr6->secid;
+		break;
+#endif /* IPv6 */
+	default:
+		goto unlabel_getattr_nolabel;
+	}
+	rcu_read_unlock();
+
+	secattr->flags |= NETLBL_SECATTR_SECID;
+	secattr->type = NETLBL_NLTYPE_UNLABELED;
+	return 0;
+
+unlabel_getattr_nolabel:
+	rcu_read_unlock();
 	if (netlabel_unlabel_acceptflg == 0)
 		return -ENOMSG;
-	netlbl_secattr_init(secattr);
 	secattr->type = NETLBL_NLTYPE_UNLABELED;
 	return 0;
 }
diff --git a/net/netlabel/netlabel_unlabeled.h b/net/netlabel/netlabel_unlabeled.h
index c2917fbb42cf..06b1301ac072 100644
--- a/net/netlabel/netlabel_unlabeled.h
+++ b/net/netlabel/netlabel_unlabeled.h
@@ -36,6 +36,116 @@
 /*
  * The following NetLabel payloads are supported by the Unlabeled subsystem.
  *
+ * o STATICADD
+ *   This message is sent from an application to add a new static label for
+ *   incoming unlabeled connections.
+ *
+ *   Required attributes:
+ *
+ *     NLBL_UNLABEL_A_IFACE
+ *     NLBL_UNLABEL_A_SECCTX
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
+ * o STATICREMOVE
+ *   This message is sent from an application to remove an existing static
+ *   label for incoming unlabeled connections.
+ *
+ *   Required attributes:
+ *
+ *     NLBL_UNLABEL_A_IFACE
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
+ * o STATICLIST
+ *   This message can be sent either from an application or by the kernel in
+ *   response to an application generated STATICLIST message.  When sent by an
+ *   application there is no payload and the NLM_F_DUMP flag should be set.
+ *   The kernel should response with a series of the following messages.
+ *
+ *   Required attributes:
+ *
+ *     NLBL_UNLABEL_A_IFACE
+ *     NLBL_UNLABEL_A_SECCTX
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
+ * o STATICADDDEF
+ *   This message is sent from an application to set the default static
+ *   label for incoming unlabeled connections.
+ *
+ *   Required attribute:
+ *
+ *     NLBL_UNLABEL_A_SECCTX
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
+ * o STATICREMOVEDEF
+ *   This message is sent from an application to remove the existing default
+ *   static label for incoming unlabeled connections.
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
+ * o STATICLISTDEF
+ *   This message can be sent either from an application or by the kernel in
+ *   response to an application generated STATICLISTDEF message.  When sent by
+ *   an application there is no payload and the NLM_F_DUMP flag should be set.
+ *   The kernel should response with the following message.
+ *
+ *   Required attribute:
+ *
+ *     NLBL_UNLABEL_A_SECCTX
+ *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV4ADDR
+ *     NLBL_UNLABEL_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_UNLABEL_A_IPV6ADDR
+ *     NLBL_UNLABEL_A_IPV6MASK
+ *
  * o ACCEPT
  *   This message is sent from an application to specify if the kernel should
  *   allow unlabled packets to pass if they do not match any of the static
@@ -62,6 +172,12 @@ enum {
 	NLBL_UNLABEL_C_UNSPEC,
 	NLBL_UNLABEL_C_ACCEPT,
 	NLBL_UNLABEL_C_LIST,
+	NLBL_UNLABEL_C_STATICADD,
+	NLBL_UNLABEL_C_STATICREMOVE,
+	NLBL_UNLABEL_C_STATICLIST,
+	NLBL_UNLABEL_C_STATICADDDEF,
+	NLBL_UNLABEL_C_STATICREMOVEDEF,
+	NLBL_UNLABEL_C_STATICLISTDEF,
 	__NLBL_UNLABEL_C_MAX,
 };
 #define NLBL_UNLABEL_C_MAX (__NLBL_UNLABEL_C_MAX - 1)
@@ -73,6 +189,24 @@ enum {
 	/* (NLA_U8)
 	 * if true then unlabeled packets are allowed to pass, else unlabeled
 	 * packets are rejected */
+	NLBL_UNLABEL_A_IPV6ADDR,
+	/* (NLA_BINARY, struct in6_addr)
+	 * an IPv6 address */
+	NLBL_UNLABEL_A_IPV6MASK,
+	/* (NLA_BINARY, struct in6_addr)
+	 * an IPv6 address mask */
+	NLBL_UNLABEL_A_IPV4ADDR,
+	/* (NLA_BINARY, struct in_addr)
+	 * an IPv4 address */
+	NLBL_UNLABEL_A_IPV4MASK,
+	/* (NLA_BINARY, struct in_addr)
+	 * and IPv4 address mask */
+	NLBL_UNLABEL_A_IFACE,
+	/* (NLA_NULL_STRING)
+	 * network interface */
+	NLBL_UNLABEL_A_SECCTX,
+	/* (NLA_BINARY)
+	 * a LSM specific security context */
 	__NLBL_UNLABEL_A_MAX,
 };
 #define NLBL_UNLABEL_A_MAX (__NLBL_UNLABEL_A_MAX - 1)
@@ -80,8 +214,17 @@ enum {
 /* NetLabel protocol functions */
 int netlbl_unlabel_genl_init(void);
 
+/* Unlabeled connection hash table size */
+/* XXX - currently this number is an uneducated guess */
+#define NETLBL_UNLHSH_BITSIZE       7
+
+/* General Unlabeled init function */
+int netlbl_unlabel_init(u32 size);
+
 /* Process Unlabeled incoming network packets */
-int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr);
+int netlbl_unlabel_getattr(const struct sk_buff *skb,
+			   u16 family,
+			   struct netlbl_lsm_secattr *secattr);
 
 /* Set the default configuration to allow Unlabeled packets */
 int netlbl_unlabel_defconf(void);
-- 
cgit v1.2.3


From 13541b3adad2dc2f56761c5193c2b88db3597f0e Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Jan 2008 08:44:23 -0500
Subject: NetLabel: Add auditing to the static labeling mechanism

This patch adds auditing support to the NetLabel static labeling mechanism.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/audit.h             |   2 +
 net/netlabel/netlabel_unlabeled.c | 207 +++++++++++++++++++++++++++++++++++---
 2 files changed, 195 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index c68781692838..bdd6f5de5fc4 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -115,6 +115,8 @@
 #define AUDIT_MAC_IPSEC_ADDSPD	1413	/* Not used */
 #define AUDIT_MAC_IPSEC_DELSPD	1414	/* Not used */
 #define AUDIT_MAC_IPSEC_EVENT	1415	/* Audit an IPSec event */
+#define AUDIT_MAC_UNLBL_STCADD	1416	/* NetLabel: add a static label */
+#define AUDIT_MAC_UNLBL_STCDEL	1417	/* NetLabel: del a static label */
 
 #define AUDIT_FIRST_KERN_ANOM_MSG   1700
 #define AUDIT_LAST_KERN_ANOM_MSG    1799
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index d0c628c875ae..42e81fd8cc49 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -146,6 +146,74 @@ static const struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1
 	[NLBL_UNLABEL_A_SECCTX] = { .type = NLA_BINARY }
 };
 
+/*
+ * Audit Helper Functions
+ */
+
+/**
+ * netlbl_unlabel_audit_addr4 - Audit an IPv4 address
+ * @audit_buf: audit buffer
+ * @dev: network interface
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Write the IPv4 address and address mask, if necessary, to @audit_buf.
+ *
+ */
+static void netlbl_unlabel_audit_addr4(struct audit_buffer *audit_buf,
+				     const char *dev,
+				     __be32 addr, __be32 mask)
+{
+	u32 mask_val = ntohl(mask);
+
+	if (dev != NULL)
+		audit_log_format(audit_buf, " netif=%s", dev);
+	audit_log_format(audit_buf, " src=" NIPQUAD_FMT, NIPQUAD(addr));
+	if (mask_val != 0xffffffff) {
+		u32 mask_len = 0;
+		while (mask_val > 0) {
+			mask_val <<= 1;
+			mask_len++;
+		}
+		audit_log_format(audit_buf, " src_prefixlen=%d", mask_len);
+	}
+}
+
+/**
+ * netlbl_unlabel_audit_addr6 - Audit an IPv6 address
+ * @audit_buf: audit buffer
+ * @dev: network interface
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Write the IPv6 address and address mask, if necessary, to @audit_buf.
+ *
+ */
+static void netlbl_unlabel_audit_addr6(struct audit_buffer *audit_buf,
+				     const char *dev,
+				     const struct in6_addr *addr,
+				     const struct in6_addr *mask)
+{
+	if (dev != NULL)
+		audit_log_format(audit_buf, " netif=%s", dev);
+	audit_log_format(audit_buf, " src=" NIP6_FMT, NIP6(*addr));
+	if (ntohl(mask->s6_addr32[3]) != 0xffffffff) {
+		u32 mask_len = 0;
+		u32 mask_val;
+		int iter = -1;
+		while (ntohl(mask->s6_addr32[++iter]) == 0xffffffff)
+			mask_len += 32;
+		mask_val = ntohl(mask->s6_addr32[iter]);
+		while (mask_val > 0) {
+			mask_val <<= 1;
+			mask_len++;
+		}
+		audit_log_format(audit_buf, " src_prefixlen=%d", mask_len);
+	}
+}
+
 /*
  * Unlabeled Connection Hash Table Functions
  */
@@ -530,6 +598,7 @@ add_iface_failure:
  * @mask: address mask in network byte order
  * @addr_len: length of address/mask (4 for IPv4, 16 for IPv6)
  * @secid: LSM secid value for the entry
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Adds a new entry to the unlabeled connection hash table.  Returns zero on
@@ -541,12 +610,18 @@ static int netlbl_unlhsh_add(struct net *net,
 			     const void *addr,
 			     const void *mask,
 			     u32 addr_len,
-			     u32 secid)
+			     u32 secid,
+			     struct netlbl_audit *audit_info)
 {
 	int ret_val;
 	int ifindex;
 	struct net_device *dev;
 	struct netlbl_unlhsh_iface *iface;
+	struct in_addr *addr4, *mask4;
+	struct in6_addr *addr6, *mask6;
+	struct audit_buffer *audit_buf = NULL;
+	char *secctx = NULL;
+	u32 secctx_len;
 
 	if (addr_len != sizeof(struct in_addr) &&
 	    addr_len != sizeof(struct in6_addr))
@@ -573,13 +648,28 @@ static int netlbl_unlhsh_add(struct net *net,
 			goto unlhsh_add_return;
 		}
 	}
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCADD,
+					      audit_info);
 	switch (addr_len) {
 	case sizeof(struct in_addr):
-		ret_val = netlbl_unlhsh_add_addr4(iface, addr, mask, secid);
+		addr4 = (struct in_addr *)addr;
+		mask4 = (struct in_addr *)mask;
+		ret_val = netlbl_unlhsh_add_addr4(iface, addr4, mask4, secid);
+		if (audit_buf != NULL)
+			netlbl_unlabel_audit_addr4(audit_buf,
+						   dev_name,
+						   addr4->s_addr,
+						   mask4->s_addr);
 		break;
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	case sizeof(struct in6_addr):
-		ret_val = netlbl_unlhsh_add_addr6(iface, addr, mask, secid);
+		addr6 = (struct in6_addr *)addr;
+		mask6 = (struct in6_addr *)mask;
+		ret_val = netlbl_unlhsh_add_addr6(iface, addr6, mask6, secid);
+		if (audit_buf != NULL)
+			netlbl_unlabel_audit_addr6(audit_buf,
+						   dev_name,
+						   addr6, mask6);
 		break;
 #endif /* IPv6 */
 	default:
@@ -590,14 +680,26 @@ static int netlbl_unlhsh_add(struct net *net,
 
 unlhsh_add_return:
 	rcu_read_unlock();
+	if (audit_buf != NULL) {
+		if (security_secid_to_secctx(secid,
+					     &secctx,
+					     &secctx_len) == 0) {
+			audit_log_format(audit_buf, " sec_obj=%s", secctx);
+			security_release_secctx(secctx, secctx_len);
+		}
+		audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0);
+		audit_log_end(audit_buf);
+	}
 	return ret_val;
 }
 
 /**
  * netlbl_unlhsh_remove_addr4 - Remove an IPv4 address entry
+ * @net: network namespace
  * @iface: interface entry
  * @addr: IP address
  * @mask: IP address mask
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Remove an IP address entry from the unlabeled connection hash table.
@@ -605,12 +707,18 @@ unlhsh_add_return:
  * responsible for calling the rcu_read_[un]lock() functions.
  *
  */
-static int netlbl_unlhsh_remove_addr4(struct netlbl_unlhsh_iface *iface,
+static int netlbl_unlhsh_remove_addr4(struct net *net,
+				      struct netlbl_unlhsh_iface *iface,
 				      const struct in_addr *addr,
-				      const struct in_addr *mask)
+				      const struct in_addr *mask,
+				      struct netlbl_audit *audit_info)
 {
 	int ret_val = -ENOENT;
 	struct netlbl_unlhsh_addr4 *entry;
+	struct audit_buffer *audit_buf = NULL;
+	struct net_device *dev;
+	char *secctx = NULL;
+	u32 secctx_len;
 
 	spin_lock(&netlbl_unlhsh_lock);
 	entry = netlbl_unlhsh_search_addr4(addr->s_addr, iface);
@@ -622,6 +730,25 @@ static int netlbl_unlhsh_remove_addr4(struct netlbl_unlhsh_iface *iface,
 	}
 	spin_unlock(&netlbl_unlhsh_lock);
 
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL,
+					      audit_info);
+	if (audit_buf != NULL) {
+		dev = dev_get_by_index(net, iface->ifindex);
+		netlbl_unlabel_audit_addr4(audit_buf,
+					   (dev != NULL ? dev->name : NULL),
+					   entry->addr, entry->mask);
+		if (dev != NULL)
+			dev_put(dev);
+		if (security_secid_to_secctx(entry->secid,
+					     &secctx,
+					     &secctx_len) == 0) {
+			audit_log_format(audit_buf, " sec_obj=%s", secctx);
+			security_release_secctx(secctx, secctx_len);
+		}
+		audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0);
+		audit_log_end(audit_buf);
+	}
+
 	if (ret_val == 0)
 		call_rcu(&entry->rcu, netlbl_unlhsh_free_addr4);
 	return ret_val;
@@ -630,9 +757,11 @@ static int netlbl_unlhsh_remove_addr4(struct netlbl_unlhsh_iface *iface,
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 /**
  * netlbl_unlhsh_remove_addr6 - Remove an IPv6 address entry
+ * @net: network namespace
  * @iface: interface entry
  * @addr: IP address
  * @mask: IP address mask
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Remove an IP address entry from the unlabeled connection hash table.
@@ -640,12 +769,18 @@ static int netlbl_unlhsh_remove_addr4(struct netlbl_unlhsh_iface *iface,
  * responsible for calling the rcu_read_[un]lock() functions.
  *
  */
-static int netlbl_unlhsh_remove_addr6(struct netlbl_unlhsh_iface *iface,
+static int netlbl_unlhsh_remove_addr6(struct net *net,
+				      struct netlbl_unlhsh_iface *iface,
 				      const struct in6_addr *addr,
-				      const struct in6_addr *mask)
+				      const struct in6_addr *mask,
+				      struct netlbl_audit *audit_info)
 {
 	int ret_val = -ENOENT;
 	struct netlbl_unlhsh_addr6 *entry;
+	struct audit_buffer *audit_buf = NULL;
+	struct net_device *dev;
+	char *secctx = NULL;
+	u32 secctx_len;
 
 	spin_lock(&netlbl_unlhsh_lock);
 	entry = netlbl_unlhsh_search_addr6(addr, iface);
@@ -658,6 +793,25 @@ static int netlbl_unlhsh_remove_addr6(struct netlbl_unlhsh_iface *iface,
 	}
 	spin_unlock(&netlbl_unlhsh_lock);
 
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL,
+					      audit_info);
+	if (audit_buf != NULL) {
+		dev = dev_get_by_index(net, iface->ifindex);
+		netlbl_unlabel_audit_addr6(audit_buf,
+					   (dev != NULL ? dev->name : NULL),
+					   addr, mask);
+		if (dev != NULL)
+			dev_put(dev);
+		if (security_secid_to_secctx(entry->secid,
+					     &secctx,
+					     &secctx_len) == 0) {
+			audit_log_format(audit_buf, " sec_obj=%s", secctx);
+			security_release_secctx(secctx, secctx_len);
+		}
+		audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0);
+		audit_log_end(audit_buf);
+	}
+
 	if (ret_val == 0)
 		call_rcu(&entry->rcu, netlbl_unlhsh_free_addr6);
 	return ret_val;
@@ -708,6 +862,7 @@ unlhsh_condremove_failure:
  * @addr: IP address in network byte order
  * @mask: address mask in network byte order
  * @addr_len: length of address/mask (4 for IPv4, 16 for IPv6)
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Removes and existing entry from the unlabeled connection hash table.
@@ -718,7 +873,8 @@ static int netlbl_unlhsh_remove(struct net *net,
 				const char *dev_name,
 				const void *addr,
 				const void *mask,
-				u32 addr_len)
+				u32 addr_len,
+				struct netlbl_audit *audit_info)
 {
 	int ret_val;
 	struct net_device *dev;
@@ -745,11 +901,15 @@ static int netlbl_unlhsh_remove(struct net *net,
 	}
 	switch (addr_len) {
 	case sizeof(struct in_addr):
-		ret_val = netlbl_unlhsh_remove_addr4(iface, addr, mask);
+		ret_val = netlbl_unlhsh_remove_addr4(net,
+						     iface, addr, mask,
+						     audit_info);
 		break;
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	case sizeof(struct in6_addr):
-		ret_val = netlbl_unlhsh_remove_addr6(iface, addr, mask);
+		ret_val = netlbl_unlhsh_remove_addr6(net,
+						     iface, addr, mask,
+						     audit_info);
 		break;
 #endif /* IPv6 */
 	default:
@@ -972,6 +1132,7 @@ static int netlbl_unlabel_staticadd(struct sk_buff *skb,
 	void *mask;
 	u32 addr_len;
 	u32 secid;
+	struct netlbl_audit audit_info;
 
 	/* Don't allow users to add both IPv4 and IPv6 addresses for a
 	 * single entry.  However, allow users to create two entries, one each
@@ -985,6 +1146,8 @@ static int netlbl_unlabel_staticadd(struct sk_buff *skb,
 	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
 		return -EINVAL;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
 	if (ret_val != 0)
 		return ret_val;
@@ -997,7 +1160,8 @@ static int netlbl_unlabel_staticadd(struct sk_buff *skb,
 		return ret_val;
 
 	return netlbl_unlhsh_add(&init_net,
-				 dev_name, addr, mask, addr_len, secid);
+				 dev_name, addr, mask, addr_len, secid,
+				 &audit_info);
 }
 
 /**
@@ -1019,6 +1183,7 @@ static int netlbl_unlabel_staticadddef(struct sk_buff *skb,
 	void *mask;
 	u32 addr_len;
 	u32 secid;
+	struct netlbl_audit audit_info;
 
 	/* Don't allow users to add both IPv4 and IPv6 addresses for a
 	 * single entry.  However, allow users to create two entries, one each
@@ -1031,6 +1196,8 @@ static int netlbl_unlabel_staticadddef(struct sk_buff *skb,
 	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
 		return -EINVAL;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
 	if (ret_val != 0)
 		return ret_val;
@@ -1041,7 +1208,9 @@ static int netlbl_unlabel_staticadddef(struct sk_buff *skb,
 	if (ret_val != 0)
 		return ret_val;
 
-	return netlbl_unlhsh_add(&init_net, NULL, addr, mask, addr_len, secid);
+	return netlbl_unlhsh_add(&init_net,
+				 NULL, addr, mask, addr_len, secid,
+				 &audit_info);
 }
 
 /**
@@ -1063,6 +1232,7 @@ static int netlbl_unlabel_staticremove(struct sk_buff *skb,
 	void *addr;
 	void *mask;
 	u32 addr_len;
+	struct netlbl_audit audit_info;
 
 	/* See the note in netlbl_unlabel_staticadd() about not allowing both
 	 * IPv4 and IPv6 in the same entry. */
@@ -1073,12 +1243,16 @@ static int netlbl_unlabel_staticremove(struct sk_buff *skb,
 	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
 		return -EINVAL;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
 	if (ret_val != 0)
 		return ret_val;
 	dev_name = nla_data(info->attrs[NLBL_UNLABEL_A_IFACE]);
 
-	return netlbl_unlhsh_remove(&init_net, dev_name, addr, mask, addr_len);
+	return netlbl_unlhsh_remove(&init_net,
+				    dev_name, addr, mask, addr_len,
+				    &audit_info);
 }
 
 /**
@@ -1099,6 +1273,7 @@ static int netlbl_unlabel_staticremovedef(struct sk_buff *skb,
 	void *addr;
 	void *mask;
 	u32 addr_len;
+	struct netlbl_audit audit_info;
 
 	/* See the note in netlbl_unlabel_staticadd() about not allowing both
 	 * IPv4 and IPv6 in the same entry. */
@@ -1108,11 +1283,15 @@ static int netlbl_unlabel_staticremovedef(struct sk_buff *skb,
 	       !info->attrs[NLBL_UNLABEL_A_IPV6MASK])))
 		return -EINVAL;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	ret_val = netlbl_unlabel_addrinfo_get(info, &addr, &mask, &addr_len);
 	if (ret_val != 0)
 		return ret_val;
 
-	return netlbl_unlhsh_remove(&init_net, NULL, addr, mask, addr_len);
+	return netlbl_unlhsh_remove(&init_net,
+				    NULL, addr, mask, addr_len,
+				    &audit_info);
 }
 
 
-- 
cgit v1.2.3


From acee478afc6ff7e1b8852d9a4dca1ff36021414d Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 22 Jan 2008 17:13:07 -0500
Subject: NFS: Clean up the write request locking.

Ensure that we set/clear NFS_PAGE_TAG_LOCKED when the nfs_page is hashed.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/pagelist.c        | 13 ++++++++-----
 fs/nfs/write.c           | 16 +++++++---------
 include/linux/nfs_page.h | 13 +------------
 3 files changed, 16 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 345bb9b4765b..3b3dbb94393d 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -111,13 +111,14 @@ void nfs_unlock_request(struct nfs_page *req)
  * nfs_set_page_tag_locked - Tag a request as locked
  * @req:
  */
-static int nfs_set_page_tag_locked(struct nfs_page *req)
+int nfs_set_page_tag_locked(struct nfs_page *req)
 {
 	struct nfs_inode *nfsi = NFS_I(req->wb_context->path.dentry->d_inode);
 
-	if (!nfs_lock_request(req))
+	if (!nfs_lock_request_dontget(req))
 		return 0;
-	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
+	if (req->wb_page != NULL)
+		radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
 	return 1;
 }
 
@@ -132,9 +133,10 @@ void nfs_clear_page_tag_locked(struct nfs_page *req)
 	if (req->wb_page != NULL) {
 		spin_lock(&inode->i_lock);
 		radix_tree_tag_clear(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
+		nfs_unlock_request(req);
 		spin_unlock(&inode->i_lock);
-	}
-	nfs_unlock_request(req);
+	} else
+		nfs_unlock_request(req);
 }
 
 /**
@@ -421,6 +423,7 @@ int nfs_scan_list(struct nfs_inode *nfsi,
 				goto out;
 			idx_start = req->wb_index + 1;
 			if (nfs_set_page_tag_locked(req)) {
+				kref_get(&req->wb_kref);
 				nfs_list_remove_request(req);
 				radix_tree_tag_clear(&nfsi->nfs_page_tree,
 						req->wb_index, tag);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 51cc1bd6a116..092e79c6d962 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -196,7 +196,7 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page,
 	}
 	/* Update file length */
 	nfs_grow_file(page, offset, count);
-	nfs_unlock_request(req);
+	nfs_clear_page_tag_locked(req);
 	return 0;
 }
 
@@ -252,7 +252,6 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 				struct page *page)
 {
 	struct inode *inode = page->mapping->host;
-	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_page *req;
 	int ret;
 
@@ -263,10 +262,10 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 			spin_unlock(&inode->i_lock);
 			return 0;
 		}
-		if (nfs_lock_request_dontget(req))
+		if (nfs_set_page_tag_locked(req))
 			break;
 		/* Note: If we hold the page lock, as is the case in nfs_writepage,
-		 *	 then the call to nfs_lock_request_dontget() will always
+		 *	 then the call to nfs_set_page_tag_locked() will always
 		 *	 succeed provided that someone hasn't already marked the
 		 *	 request as dirty (in which case we don't care).
 		 */
@@ -280,7 +279,7 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 	if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) {
 		/* This request is marked for commit */
 		spin_unlock(&inode->i_lock);
-		nfs_unlock_request(req);
+		nfs_clear_page_tag_locked(req);
 		nfs_pageio_complete(pgio);
 		return 0;
 	}
@@ -288,8 +287,6 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 		spin_unlock(&inode->i_lock);
 		BUG();
 	}
-	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index,
-			NFS_PAGE_TAG_LOCKED);
 	spin_unlock(&inode->i_lock);
 	nfs_pageio_add_request(pgio, req);
 	return 0;
@@ -381,6 +378,7 @@ static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
 	set_page_private(req->wb_page, (unsigned long)req);
 	nfsi->npages++;
 	kref_get(&req->wb_kref);
+	radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
 	return 0;
 }
 
@@ -596,7 +594,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx,
 		spin_lock(&inode->i_lock);
 		req = nfs_page_find_request_locked(page);
 		if (req) {
-			if (!nfs_lock_request_dontget(req)) {
+			if (!nfs_set_page_tag_locked(req)) {
 				int error;
 
 				spin_unlock(&inode->i_lock);
@@ -646,7 +644,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx,
 	    || req->wb_page != page
 	    || !nfs_dirty_request(req)
 	    || offset > rqend || end < req->wb_offset) {
-		nfs_unlock_request(req);
+		nfs_clear_page_tag_locked(req);
 		return ERR_PTR(-EBUSY);
 	}
 
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index 30dbcc185e69..a1676e19e491 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -83,6 +83,7 @@ extern	void nfs_pageio_complete(struct nfs_pageio_descriptor *desc);
 extern	void nfs_pageio_cond_complete(struct nfs_pageio_descriptor *, pgoff_t);
 extern  int nfs_wait_on_request(struct nfs_page *);
 extern	void nfs_unlock_request(struct nfs_page *req);
+extern	int nfs_set_page_tag_locked(struct nfs_page *req);
 extern  void nfs_clear_page_tag_locked(struct nfs_page *req);
 
 
@@ -95,18 +96,6 @@ nfs_lock_request_dontget(struct nfs_page *req)
 	return !test_and_set_bit(PG_BUSY, &req->wb_flags);
 }
 
-/*
- * Lock the page of an asynchronous request and take a reference
- */
-static inline int
-nfs_lock_request(struct nfs_page *req)
-{
-	if (test_and_set_bit(PG_BUSY, &req->wb_flags))
-		return 0;
-	kref_get(&req->wb_kref);
-	return 1;
-}
-
 /**
  * nfs_list_add_request - Insert a request into a list
  * @req: request
-- 
cgit v1.2.3


From ef818a28fac9bd214e676986d8301db0582b92a9 Mon Sep 17 00:00:00 2001
From: Steve Dickson <SteveD@redhat.com>
Date: Thu, 8 Nov 2007 04:05:04 -0500
Subject: NFS: Stop sillyname renames and unmounts from racing

Added an active/deactive mechanism to the nfs_server structure
allowing async operations to hold off umount until the
operations are done.

Signed-off-by: Steve Dickson <steved@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           |  3 +++
 fs/nfs/internal.h         |  2 ++
 fs/nfs/super.c            | 24 ++++++++++++++++++++++++
 fs/nfs/unlink.c           |  4 ++++
 include/linux/nfs_fs_sb.h |  6 ++++++
 5 files changed, 39 insertions(+)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index a6f625497612..c3740f5ab978 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -729,6 +729,9 @@ static struct nfs_server *nfs_alloc_server(void)
 	INIT_LIST_HEAD(&server->client_link);
 	INIT_LIST_HEAD(&server->master_link);
 
+	init_waitqueue_head(&server->active_wq);
+	atomic_set(&server->active, 0);
+
 	server->io_stats = nfs_alloc_iostats();
 	if (!server->io_stats) {
 		kfree(server);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index f3acf48412be..75793794aefe 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -160,6 +160,8 @@ extern struct rpc_stat nfs_rpcstat;
 
 extern int __init register_nfs_fs(void);
 extern void __exit unregister_nfs_fs(void);
+extern void nfs_sb_active(struct nfs_server *server);
+extern void nfs_sb_deactive(struct nfs_server *server);
 
 /* namespace.c */
 extern char *nfs_path(const char *base,
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 0b0c72a072ff..fda1635dd133 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -202,6 +202,7 @@ static int nfs_get_sb(struct file_system_type *, int, const char *, void *, stru
 static int nfs_xdev_get_sb(struct file_system_type *fs_type,
 		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static void nfs_kill_super(struct super_block *);
+static void nfs_put_super(struct super_block *);
 
 static struct file_system_type nfs_fs_type = {
 	.owner		= THIS_MODULE,
@@ -223,6 +224,7 @@ static const struct super_operations nfs_sops = {
 	.alloc_inode	= nfs_alloc_inode,
 	.destroy_inode	= nfs_destroy_inode,
 	.write_inode	= nfs_write_inode,
+	.put_super	= nfs_put_super,
 	.statfs		= nfs_statfs,
 	.clear_inode	= nfs_clear_inode,
 	.umount_begin	= nfs_umount_begin,
@@ -325,6 +327,28 @@ void __exit unregister_nfs_fs(void)
 	unregister_filesystem(&nfs_fs_type);
 }
 
+void nfs_sb_active(struct nfs_server *server)
+{
+	atomic_inc(&server->active);
+}
+
+void nfs_sb_deactive(struct nfs_server *server)
+{
+	if (atomic_dec_and_test(&server->active))
+		wake_up(&server->active_wq);
+}
+
+static void nfs_put_super(struct super_block *sb)
+{
+	struct nfs_server *server = NFS_SB(sb);
+	/*
+	 * Make sure there are no outstanding ops to this server.
+	 * If so, wait for them to finish before allowing the
+	 * unmount to continue.
+	 */
+	wait_event(server->active_wq, atomic_read(&server->active) == 0);
+}
+
 /*
  * Deliver file system statistics to userspace
  */
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 431981d0265f..8e5428e0b86f 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -14,6 +14,8 @@
 #include <linux/sched.h>
 #include <linux/wait.h>
 
+#include "internal.h"
+
 struct nfs_unlinkdata {
 	struct hlist_node list;
 	struct nfs_removeargs args;
@@ -113,6 +115,7 @@ static void nfs_async_unlink_release(void *calldata)
 	struct nfs_unlinkdata	*data = calldata;
 
 	nfs_dec_sillycount(data->dir);
+	nfs_sb_deactive(NFS_SERVER(data->dir));
 	nfs_free_unlinkdata(data);
 }
 
@@ -153,6 +156,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
 		nfs_dec_sillycount(dir);
 		return 0;
 	}
+	nfs_sb_active(NFS_SERVER(dir));
 	data->args.fh = NFS_FH(dir);
 	nfs_fattr_init(&data->res.dir_attr);
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 0cac49bc0955..9f949b587684 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -3,6 +3,9 @@
 
 #include <linux/list.h>
 #include <linux/backing-dev.h>
+#include <linux/wait.h>
+
+#include <asm/atomic.h>
 
 struct nfs_iostats;
 
@@ -110,6 +113,9 @@ struct nfs_server {
 						   filesystem */
 #endif
 	void (*destroy)(struct nfs_server *);
+
+	atomic_t active; /* Keep trace of any activity to this server */
+	wait_queue_head_t active_wq;  /* Wait for any activity to stop  */
 };
 
 /* Server capabilities */
-- 
cgit v1.2.3


From 66af1e558538137080615e7ad6d1f2f80862de01 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 6 Nov 2007 10:18:36 -0500
Subject: SUNRPC: Fix a race in xs_tcp_state_change()

When scheduling the autoclose RPC call, we want to ensure that we don't
race against the test_bit() call in xprt_clear_locked().

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h |  1 +
 net/sunrpc/xprt.c           | 20 ++++++++++++++++++++
 net/sunrpc/xprtsock.c       |  5 +----
 3 files changed, 22 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 30b17b3bc1a9..6f524a9e7fd0 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -246,6 +246,7 @@ struct rpc_rqst *	xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid);
 void			xprt_complete_rqst(struct rpc_task *task, int copied);
 void			xprt_release_rqst_cong(struct rpc_task *task);
 void			xprt_disconnect(struct rpc_xprt *xprt);
+void			xprt_force_disconnect(struct rpc_xprt *xprt);
 
 /*
  * Reserved bit positions in xprt->state
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index fb92f51405c5..a3af02168af1 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -570,6 +570,7 @@ static void xprt_autoclose(struct work_struct *work)
 
 	xprt_disconnect(xprt);
 	xprt->ops->close(xprt);
+	clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
 	xprt_release_write(xprt, NULL);
 }
 
@@ -588,6 +589,25 @@ void xprt_disconnect(struct rpc_xprt *xprt)
 }
 EXPORT_SYMBOL_GPL(xprt_disconnect);
 
+/**
+ * xprt_force_disconnect - force a transport to disconnect
+ * @xprt: transport to disconnect
+ *
+ */
+void xprt_force_disconnect(struct rpc_xprt *xprt)
+{
+	/* Don't race with the test_bit() in xprt_clear_locked() */
+	spin_lock_bh(&xprt->transport_lock);
+	set_bit(XPRT_CLOSE_WAIT, &xprt->state);
+	/* Try to schedule an autoclose RPC call */
+	if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
+		queue_work(rpciod_workqueue, &xprt->task_cleanup);
+	else if (xprt->snd_task != NULL)
+		rpc_wake_up_task(xprt->snd_task);
+	spin_unlock_bh(&xprt->transport_lock);
+}
+EXPORT_SYMBOL_GPL(xprt_force_disconnect);
+
 static void
 xprt_init_autodisconnect(unsigned long data)
 {
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 6fa52f44de0f..abb40c140738 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1122,10 +1122,7 @@ static void xs_tcp_state_change(struct sock *sk)
 	case TCP_SYN_RECV:
 		break;
 	case TCP_CLOSE_WAIT:
-		/* Try to schedule an autoclose RPC calls */
-		set_bit(XPRT_CLOSE_WAIT, &xprt->state);
-		if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
-			queue_work(rpciod_workqueue, &xprt->task_cleanup);
+		xprt_force_disconnect(xprt);
 	default:
 		xprt_disconnect(xprt);
 	}
-- 
cgit v1.2.3


From 3b948ae5be5e22532584113e2e02029519bbad8f Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 5 Nov 2007 17:42:39 -0500
Subject: SUNRPC: Allow the client to detect if the TCP connection is closed

Add an xprt->state bit to enable the TCP ->state_change() method to signal
whether or not the TCP connection is in the process of closing down.
This will to be used by the reconnection logic in a separate patch.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h |  1 +
 net/sunrpc/xprtsock.c       | 24 +++++++++++++++++++++---
 2 files changed, 22 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 6f524a9e7fd0..afb9e6ad7fe0 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -257,6 +257,7 @@ void			xprt_force_disconnect(struct rpc_xprt *xprt);
 #define XPRT_CLOSE_WAIT		(3)
 #define XPRT_BOUND		(4)
 #define XPRT_BINDING		(5)
+#define XPRT_CLOSING		(6)
 
 static inline void xprt_set_connected(struct rpc_xprt *xprt)
 {
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index b5544514ee57..721c5419765e 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -758,7 +758,9 @@ static void xs_close(struct rpc_xprt *xprt)
 	sock_release(sock);
 clear_close_wait:
 	smp_mb__before_clear_bit();
+	clear_bit(XPRT_CONNECTED, &xprt->state);
 	clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
+	clear_bit(XPRT_CLOSING, &xprt->state);
 	smp_mb__after_clear_bit();
 }
 
@@ -1118,12 +1120,28 @@ static void xs_tcp_state_change(struct sock *sk)
 		}
 		spin_unlock_bh(&xprt->transport_lock);
 		break;
-	case TCP_SYN_SENT:
-	case TCP_SYN_RECV:
+	case TCP_FIN_WAIT1:
+		/* The client initiated a shutdown of the socket */
+		set_bit(XPRT_CLOSING, &xprt->state);
+		smp_mb__before_clear_bit();
+		clear_bit(XPRT_CONNECTED, &xprt->state);
+		smp_mb__after_clear_bit();
 		break;
 	case TCP_CLOSE_WAIT:
+		/* The server initiated a shutdown of the socket */
+		set_bit(XPRT_CLOSING, &xprt->state);
 		xprt_force_disconnect(xprt);
-	default:
+		break;
+	case TCP_LAST_ACK:
+		smp_mb__before_clear_bit();
+		clear_bit(XPRT_CONNECTED, &xprt->state);
+		smp_mb__after_clear_bit();
+		break;
+	case TCP_CLOSE:
+		smp_mb__before_clear_bit();
+		clear_bit(XPRT_CLOSING, &xprt->state);
+		smp_mb__after_clear_bit();
+		/* Mark transport as closed and wake up all pending tasks */
 		xprt_disconnect(xprt);
 	}
  out:
-- 
cgit v1.2.3


From 62da3b24880bccd4ffc32cf8d9a7e23fab475bdd Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 6 Nov 2007 18:44:20 -0500
Subject: SUNRPC: Rename xprt_disconnect()

xprt_disconnect() should really only be called when the transport shutdown
is completed, and it is time to wake up any pending tasks. Rename it to
xprt_disconnect_done() in order to reflect the semantical change.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h     | 2 +-
 net/sunrpc/xprt.c               | 6 +++---
 net/sunrpc/xprtrdma/transport.c | 4 ++--
 net/sunrpc/xprtsock.c           | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index afb9e6ad7fe0..2554cd2b016f 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -245,7 +245,7 @@ void			xprt_adjust_cwnd(struct rpc_task *task, int result);
 struct rpc_rqst *	xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid);
 void			xprt_complete_rqst(struct rpc_task *task, int copied);
 void			xprt_release_rqst_cong(struct rpc_task *task);
-void			xprt_disconnect(struct rpc_xprt *xprt);
+void			xprt_disconnect_done(struct rpc_xprt *xprt);
 void			xprt_force_disconnect(struct rpc_xprt *xprt);
 
 /*
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 80742995e02e..592c2ee63e0f 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -574,11 +574,11 @@ static void xprt_autoclose(struct work_struct *work)
 }
 
 /**
- * xprt_disconnect - mark a transport as disconnected
+ * xprt_disconnect_done - mark a transport as disconnected
  * @xprt: transport to flag for disconnect
  *
  */
-void xprt_disconnect(struct rpc_xprt *xprt)
+void xprt_disconnect_done(struct rpc_xprt *xprt)
 {
 	dprintk("RPC:       disconnected transport %p\n", xprt);
 	spin_lock_bh(&xprt->transport_lock);
@@ -586,7 +586,7 @@ void xprt_disconnect(struct rpc_xprt *xprt)
 	xprt_wake_pending_tasks(xprt, -ENOTCONN);
 	spin_unlock_bh(&xprt->transport_lock);
 }
-EXPORT_SYMBOL_GPL(xprt_disconnect);
+EXPORT_SYMBOL_GPL(xprt_disconnect_done);
 
 /**
  * xprt_force_disconnect - force a transport to disconnect
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 6f2112dd9f78..73033d824eed 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -449,7 +449,7 @@ xprt_rdma_close(struct rpc_xprt *xprt)
 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 
 	dprintk("RPC:       %s: closing\n", __func__);
-	xprt_disconnect(xprt);
+	xprt_disconnect_done(xprt);
 	(void) rpcrdma_ep_disconnect(&r_xprt->rx_ep, &r_xprt->rx_ia);
 }
 
@@ -682,7 +682,7 @@ xprt_rdma_send_request(struct rpc_task *task)
 	}
 
 	if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req)) {
-		xprt_disconnect(xprt);
+		xprt_disconnect_done(xprt);
 		return -ENOTCONN;	/* implies disconnect */
 	}
 
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 8c9af3d92c67..741ab8ad1f3a 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -777,7 +777,7 @@ clear_close_wait:
 	clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
 	clear_bit(XPRT_CLOSING, &xprt->state);
 	smp_mb__after_clear_bit();
-	xprt_disconnect(xprt);
+	xprt_disconnect_done(xprt);
 }
 
 /**
@@ -1159,7 +1159,7 @@ static void xs_tcp_state_change(struct sock *sk)
 		clear_bit(XPRT_CLOSING, &xprt->state);
 		smp_mb__after_clear_bit();
 		/* Mark transport as closed and wake up all pending tasks */
-		xprt_disconnect(xprt);
+		xprt_disconnect_done(xprt);
 	}
  out:
 	read_unlock(&sk->sk_callback_lock);
-- 
cgit v1.2.3


From 84115e1cd4a3614c4e566d4cce31381dce3dbef9 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat, 14 Jul 2007 15:39:59 -0400
Subject: SUNRPC: Cleanup of rpc_task initialisation

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c              | 36 ++++++++++++++++++++++++-------
 fs/nfs/read.c                | 15 ++++++++-----
 fs/nfs/write.c               | 30 ++++++++++++++++----------
 include/linux/sunrpc/clnt.h  |  2 +-
 include/linux/sunrpc/sched.h | 14 +++++++-----
 net/sunrpc/clnt.c            | 51 ++++++++++++++++++++++++++++++++------------
 net/sunrpc/sched.c           | 27 +++++++++++------------
 7 files changed, 117 insertions(+), 58 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 3c9d16b4f80c..f9f5fc13dc7d 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -272,6 +272,11 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
 	size_t rsize = NFS_SERVER(inode)->rsize;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = &nfs_read_direct_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	unsigned int pgbase;
 	int result;
 	ssize_t started = 0;
@@ -322,8 +327,8 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		data->res.eof = 0;
 		data->res.count = bytes;
 
-		rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC,
-				&nfs_read_direct_ops, data);
+		task_setup_data.callback_data = data;
+		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->read_setup(data);
 
 		data->task.tk_cookie = (unsigned long) inode;
@@ -431,6 +436,11 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	struct inode *inode = dreq->inode;
 	struct list_head *p;
 	struct nfs_write_data *data;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = &nfs_write_direct_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 
 	dreq->count = 0;
 	get_dreq(dreq);
@@ -451,8 +461,8 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 		 * Reuse data->task; data->args should not have changed
 		 * since the original request was sent.
 		 */
-		rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC,
-				&nfs_write_direct_ops, data);
+		task_setup_data.callback_data = data;
+		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE);
 
 		data->task.tk_priority = RPC_PRIORITY_NORMAL;
@@ -504,6 +514,12 @@ static const struct rpc_call_ops nfs_commit_direct_ops = {
 static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 {
 	struct nfs_write_data *data = dreq->commit_data;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(dreq->inode),
+		.callback_ops = &nfs_commit_direct_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC,
+	};
 
 	data->inode = dreq->inode;
 	data->cred = dreq->ctx->cred;
@@ -515,8 +531,7 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 	data->res.fattr = &data->fattr;
 	data->res.verf = &data->verf;
 
-	rpc_init_task(&data->task, NFS_CLIENT(dreq->inode), RPC_TASK_ASYNC,
-				&nfs_commit_direct_ops, data);
+	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(data->inode)->commit_setup(data, 0);
 
 	data->task.tk_priority = RPC_PRIORITY_NORMAL;
@@ -641,6 +656,11 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 	struct inode *inode = ctx->path.dentry->d_inode;
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = &nfs_write_direct_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	size_t wsize = NFS_SERVER(inode)->wsize;
 	unsigned int pgbase;
 	int result;
@@ -694,8 +714,8 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 		data->res.count = bytes;
 		data->res.verf = &data->verf;
 
-		rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC,
-				&nfs_write_direct_ops, data);
+		task_setup_data.callback_data = data;
+		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->write_setup(data, sync);
 
 		data->task.tk_priority = RPC_PRIORITY_NORMAL;
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 4587a86adaac..c7f0d5ebd451 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -160,11 +160,17 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 		const struct rpc_call_ops *call_ops,
 		unsigned int count, unsigned int offset)
 {
-	struct inode		*inode;
-	int flags;
+	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = call_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC | swap_flags,
+	};
 
 	data->req	  = req;
-	data->inode	  = inode = req->wb_context->path.dentry->d_inode;
+	data->inode	  = inode;
 	data->cred	  = req->wb_context->cred;
 
 	data->args.fh     = NFS_FH(inode);
@@ -180,8 +186,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct. */
-	flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0);
-	rpc_init_task(&data->task, NFS_CLIENT(inode), flags, call_ops, data);
+	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->read_setup(data);
 
 	data->task.tk_cookie = (unsigned long)inode;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 092e79c6d962..c4376606f106 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -773,8 +773,14 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 		unsigned int count, unsigned int offset,
 		int how)
 {
-	struct inode		*inode;
-	int flags;
+	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = call_ops,
+		.callback_data = data,
+		.flags = flags,
+	};
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
@@ -796,8 +802,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct.  */
-	flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
-	rpc_init_task(&data->task, NFS_CLIENT(inode), flags, call_ops, data);
+	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->write_setup(data, how);
 
 	data->task.tk_priority = flush_task_priority(how);
@@ -1144,16 +1149,20 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 		struct nfs_write_data *data,
 		int how)
 {
-	struct nfs_page		*first;
-	struct inode		*inode;
-	int flags;
+	struct nfs_page *first = nfs_list_entry(head->next);
+	struct inode *inode = first->wb_context->path.dentry->d_inode;
+	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(inode),
+		.callback_ops = &nfs_commit_ops,
+		.callback_data = data,
+		.flags = flags,
+	};
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
 
 	list_splice_init(head, &data->pages);
-	first = nfs_list_entry(data->pages.next);
-	inode = first->wb_context->path.dentry->d_inode;
 
 	data->inode	  = inode;
 	data->cred	  = first->wb_context->cred;
@@ -1168,8 +1177,7 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct.  */
-	flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
-	rpc_init_task(&data->task, NFS_CLIENT(inode), flags, &nfs_commit_ops, data);
+	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->commit_setup(data, how);
 
 	data->task.tk_priority = flush_task_priority(how);
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index d9d5c5ad826c..ec9704181f94 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -126,7 +126,7 @@ int		rpcb_register(u32, u32, int, unsigned short, int *);
 int		rpcb_getport_sync(struct sockaddr_in *, __u32, __u32, int);
 void		rpcb_getport_async(struct rpc_task *);
 
-void		rpc_call_setup(struct rpc_task *, struct rpc_message *, int);
+void		rpc_call_setup(struct rpc_task *, const struct rpc_message *, int);
 
 int		rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg,
 			       int flags, const struct rpc_call_ops *tk_ops,
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 8ea077db0099..9efe045fc376 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -117,6 +117,13 @@ struct rpc_call_ops {
 	void (*rpc_release)(void *);
 };
 
+struct rpc_task_setup {
+	struct rpc_clnt *rpc_client;
+	const struct rpc_message *rpc_message;
+	const struct rpc_call_ops *callback_ops;
+	void *callback_data;
+	unsigned short flags;
+};
 
 /*
  * RPC task flags
@@ -236,13 +243,10 @@ struct rpc_wait_queue {
 /*
  * Function prototypes
  */
-struct rpc_task *rpc_new_task(struct rpc_clnt *, int flags,
-				const struct rpc_call_ops *ops, void *data);
+struct rpc_task *rpc_new_task(const struct rpc_task_setup *);
 struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
 				const struct rpc_call_ops *ops, void *data);
-void		rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt,
-				int flags, const struct rpc_call_ops *ops,
-				void *data);
+void		rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *);
 void		rpc_put_task(struct rpc_task *);
 void		rpc_exit_task(struct rpc_task *);
 void		rpc_release_calldata(const struct rpc_call_ops *, void *);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 22092b91dd85..7c80abd9263f 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -524,25 +524,22 @@ void rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset)
 EXPORT_SYMBOL_GPL(rpc_clnt_sigunmask);
 
 static
-struct rpc_task *rpc_do_run_task(struct rpc_clnt *clnt,
-		struct rpc_message *msg,
-		int flags,
-		const struct rpc_call_ops *ops,
-		void *data)
+struct rpc_task *rpc_do_run_task(const struct rpc_task_setup *task_setup_data)
 {
 	struct rpc_task *task, *ret;
 	sigset_t oldset;
 
-	task = rpc_new_task(clnt, flags, ops, data);
+	task = rpc_new_task(task_setup_data);
 	if (task == NULL) {
-		rpc_release_calldata(ops, data);
+		rpc_release_calldata(task_setup_data->callback_ops,
+				task_setup_data->callback_data);
 		return ERR_PTR(-ENOMEM);
 	}
 
 	/* Mask signals on synchronous RPC calls and RPCSEC_GSS upcalls */
 	rpc_task_sigmask(task, &oldset);
-	if (msg != NULL) {
-		rpc_call_setup(task, msg, 0);
+	if (task_setup_data->rpc_message != NULL) {
+		rpc_call_setup(task, task_setup_data->rpc_message, 0);
 		if (task->tk_status != 0) {
 			ret = ERR_PTR(task->tk_status);
 			rpc_put_task(task);
@@ -566,11 +563,17 @@ out:
 int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
 {
 	struct rpc_task	*task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.rpc_message = msg,
+		.callback_ops = &rpc_default_ops,
+		.flags = flags,
+	};
 	int status;
 
 	BUG_ON(flags & RPC_TASK_ASYNC);
 
-	task = rpc_do_run_task(clnt, msg, flags, &rpc_default_ops, NULL);
+	task = rpc_do_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = task->tk_status;
@@ -592,8 +595,15 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
 	       const struct rpc_call_ops *tk_ops, void *data)
 {
 	struct rpc_task	*task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.rpc_message = msg,
+		.callback_ops = tk_ops,
+		.callback_data = data,
+		.flags = flags|RPC_TASK_ASYNC,
+	};
 
-	task = rpc_do_run_task(clnt, msg, flags|RPC_TASK_ASYNC, tk_ops, data);
+	task = rpc_do_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	rpc_put_task(task);
@@ -612,12 +622,19 @@ struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
 					const struct rpc_call_ops *tk_ops,
 					void *data)
 {
-	return rpc_do_run_task(clnt, NULL, flags, tk_ops, data);
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.callback_ops = tk_ops,
+		.callback_data = data,
+		.flags = flags,
+	};
+
+	return rpc_do_run_task(&task_setup_data);
 }
 EXPORT_SYMBOL_GPL(rpc_run_task);
 
 void
-rpc_call_setup(struct rpc_task *task, struct rpc_message *msg, int flags)
+rpc_call_setup(struct rpc_task *task, const struct rpc_message *msg, int flags)
 {
 	task->tk_msg   = *msg;
 	task->tk_flags |= flags;
@@ -1527,7 +1544,13 @@ struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred, int
 		.rpc_proc = &rpcproc_null,
 		.rpc_cred = cred,
 	};
-	return rpc_do_run_task(clnt, &msg, flags, &rpc_default_ops, NULL);
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.rpc_message = &msg,
+		.callback_ops = &rpc_default_ops,
+		.flags = flags,
+	};
+	return rpc_do_run_task(&task_setup_data);
 }
 EXPORT_SYMBOL_GPL(rpc_call_null);
 
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index d0b4c7e11e06..10216989309c 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -815,18 +815,15 @@ EXPORT_SYMBOL_GPL(rpc_free);
 /*
  * Creation and deletion of RPC task structures
  */
-void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *tk_ops, void *calldata)
+void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setup_data)
 {
 	memset(task, 0, sizeof(*task));
 	setup_timer(&task->tk_timer, (void (*)(unsigned long))rpc_run_timer,
 			(unsigned long)task);
 	atomic_set(&task->tk_count, 1);
-	task->tk_client = clnt;
-	task->tk_flags  = flags;
-	task->tk_ops = tk_ops;
-	if (tk_ops->rpc_call_prepare != NULL)
-		task->tk_action = rpc_prepare_task;
-	task->tk_calldata = calldata;
+	task->tk_flags  = task_setup_data->flags;
+	task->tk_ops = task_setup_data->callback_ops;
+	task->tk_calldata = task_setup_data->callback_data;
 	INIT_LIST_HEAD(&task->tk_task);
 
 	/* Initialize retry counters */
@@ -839,15 +836,17 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, cons
 	/* Initialize workqueue for async tasks */
 	task->tk_workqueue = rpciod_workqueue;
 
-	if (clnt) {
-		kref_get(&clnt->cl_kref);
-		if (clnt->cl_softrtry)
+	task->tk_client = task_setup_data->rpc_client;
+	if (task->tk_client != NULL) {
+		kref_get(&task->tk_client->cl_kref);
+		if (task->tk_client->cl_softrtry)
 			task->tk_flags |= RPC_TASK_SOFT;
-		if (!clnt->cl_intr)
+		if (!task->tk_client->cl_intr)
 			task->tk_flags |= RPC_TASK_NOINTR;
 	}
 
-	BUG_ON(task->tk_ops == NULL);
+	if (task->tk_ops->rpc_call_prepare != NULL)
+		task->tk_action = rpc_prepare_task;
 
 	/* starting timestamp */
 	task->tk_start = jiffies;
@@ -873,7 +872,7 @@ static void rpc_free_task(struct rcu_head *rcu)
 /*
  * Create a new task for the specified client.
  */
-struct rpc_task *rpc_new_task(struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *tk_ops, void *calldata)
+struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data)
 {
 	struct rpc_task	*task;
 
@@ -881,7 +880,7 @@ struct rpc_task *rpc_new_task(struct rpc_clnt *clnt, int flags, const struct rpc
 	if (!task)
 		goto out;
 
-	rpc_init_task(task, clnt, flags, tk_ops, calldata);
+	rpc_init_task(task, setup_data);
 
 	dprintk("RPC:       allocated task %p\n", task);
 	task->tk_flags |= RPC_TASK_DYNAMIC;
-- 
cgit v1.2.3


From c970aa85e71bd581726c42df843f6f129db275ac Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat, 14 Jul 2007 15:39:59 -0400
Subject: SUNRPC: Clean up rpc_run_task

Make it use the new task initialiser structure instead of acting as a
wrapper.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c            | 49 +++++++++++++++++++++++++++++++++++++-------
 fs/nfs/unlink.c              |  9 +++++++-
 include/linux/sunrpc/sched.h |  3 +--
 net/sunrpc/clnt.c            | 36 ++++++++------------------------
 net/sunrpc/rpcb_clnt.c       |  8 +++++++-
 5 files changed, 67 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a51a7537f3f6..ff2c5f83ce87 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -779,12 +779,18 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data)
 {
 	struct nfs_server *server = NFS_SERVER(data->dir->d_inode);
 	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = server->client,
+		.callback_ops = &nfs4_open_confirm_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC,
+	};
 	int status;
 
 	kref_get(&data->kref);
 	data->rpc_done = 0;
 	data->rpc_status = 0;
-	task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_open_confirm_ops, data);
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = nfs4_wait_for_completion_rpc_task(task);
@@ -908,13 +914,19 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 	struct nfs_openargs *o_arg = &data->o_arg;
 	struct nfs_openres *o_res = &data->o_res;
 	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = server->client,
+		.callback_ops = &nfs4_open_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC,
+	};
 	int status;
 
 	kref_get(&data->kref);
 	data->rpc_done = 0;
 	data->rpc_status = 0;
 	data->cancelled = 0;
-	task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_open_ops, data);
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = nfs4_wait_for_completion_rpc_task(task);
@@ -1309,6 +1321,11 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	struct nfs4_closedata *calldata;
 	struct nfs4_state_owner *sp = state->owner;
 	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = server->client,
+		.callback_ops = &nfs4_close_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	int status = -ENOMEM;
 
 	calldata = kmalloc(sizeof(*calldata), GFP_KERNEL);
@@ -1328,7 +1345,8 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->path.mnt = mntget(path->mnt);
 	calldata->path.dentry = dget(path->dentry);
 
-	task = rpc_run_task(server->client, RPC_TASK_ASYNC, &nfs4_close_ops, calldata);
+	task_setup_data.callback_data = calldata;
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = 0;
@@ -3027,6 +3045,11 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	struct nfs4_delegreturndata *data;
 	struct nfs_server *server = NFS_SERVER(inode);
 	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = server->client,
+		.callback_ops = &nfs4_delegreturn_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	int status;
 
 	data = kmalloc(sizeof(*data), GFP_KERNEL);
@@ -3043,7 +3066,8 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	data->timestamp = jiffies;
 	data->rpc_status = 0;
 
-	task = rpc_run_task(NFS_CLIENT(inode), RPC_TASK_ASYNC, &nfs4_delegreturn_ops, data);
+	task_setup_data.callback_data = data;
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = nfs4_wait_for_completion_rpc_task(task);
@@ -3260,6 +3284,11 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl,
 		struct nfs_seqid *seqid)
 {
 	struct nfs4_unlockdata *data;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(lsp->ls_state->inode),
+		.callback_ops = &nfs4_locku_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 
 	/* Ensure this is an unlock - when canceling a lock, the
 	 * canceled lock is passed in, and it won't be an unlock.
@@ -3272,7 +3301,8 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl,
 		return ERR_PTR(-ENOMEM);
 	}
 
-	return rpc_run_task(NFS_CLIENT(lsp->ls_state->inode), RPC_TASK_ASYNC, &nfs4_locku_ops, data);
+	task_setup_data.callback_data = data;
+	return rpc_run_task(&task_setup_data);
 }
 
 static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
@@ -3438,6 +3468,11 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 {
 	struct nfs4_lockdata *data;
 	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = NFS_CLIENT(state->inode),
+		.callback_ops = &nfs4_lock_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	int ret;
 
 	dprintk("%s: begin!\n", __FUNCTION__);
@@ -3449,8 +3484,8 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		data->arg.block = 1;
 	if (reclaim != 0)
 		data->arg.reclaim = 1;
-	task = rpc_run_task(NFS_CLIENT(state->inode), RPC_TASK_ASYNC,
-			&nfs4_lock_ops, data);
+	task_setup_data.callback_data = data;
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	ret = nfs4_wait_for_completion_rpc_task(task);
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 8e5428e0b86f..6660d9a53345 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -127,6 +127,11 @@ static const struct rpc_call_ops nfs_unlink_ops = {
 
 static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data)
 {
+	struct rpc_task_setup task_setup_data = {
+		.callback_ops = &nfs_unlink_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC,
+	};
 	struct rpc_task *task;
 	struct dentry *alias;
 
@@ -160,7 +165,9 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
 	data->args.fh = NFS_FH(dir);
 	nfs_fattr_init(&data->res.dir_attr);
 
-	task = rpc_run_task(NFS_CLIENT(dir), RPC_TASK_ASYNC, &nfs_unlink_ops, data);
+	task_setup_data.rpc_client = NFS_CLIENT(dir);
+
+	task = rpc_run_task(&task_setup_data);
 	if (!IS_ERR(task))
 		rpc_put_task(task);
 	return 1;
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 9efe045fc376..d974421d7647 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -244,8 +244,7 @@ struct rpc_wait_queue {
  * Function prototypes
  */
 struct rpc_task *rpc_new_task(const struct rpc_task_setup *);
-struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
-				const struct rpc_call_ops *ops, void *data);
+struct rpc_task *rpc_run_task(const struct rpc_task_setup *);
 void		rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *);
 void		rpc_put_task(struct rpc_task *);
 void		rpc_exit_task(struct rpc_task *);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 7c80abd9263f..9aad45946d32 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -523,8 +523,11 @@ void rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset)
 }
 EXPORT_SYMBOL_GPL(rpc_clnt_sigunmask);
 
-static
-struct rpc_task *rpc_do_run_task(const struct rpc_task_setup *task_setup_data)
+/**
+ * rpc_run_task - Allocate a new RPC task, then run rpc_execute against it
+ * @task_setup_data: pointer to task initialisation data
+ */
+struct rpc_task *rpc_run_task(const struct rpc_task_setup *task_setup_data)
 {
 	struct rpc_task *task, *ret;
 	sigset_t oldset;
@@ -553,6 +556,7 @@ out:
 	rpc_restore_sigmask(&oldset);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(rpc_run_task);
 
 /**
  * rpc_call_sync - Perform a synchronous RPC call
@@ -573,7 +577,7 @@ int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
 
 	BUG_ON(flags & RPC_TASK_ASYNC);
 
-	task = rpc_do_run_task(&task_setup_data);
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	status = task->tk_status;
@@ -603,7 +607,7 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
 		.flags = flags|RPC_TASK_ASYNC,
 	};
 
-	task = rpc_do_run_task(&task_setup_data);
+	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
 	rpc_put_task(task);
@@ -611,28 +615,6 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
 }
 EXPORT_SYMBOL_GPL(rpc_call_async);
 
-/**
- * rpc_run_task - Allocate a new RPC task, then run rpc_execute against it
- * @clnt: pointer to RPC client
- * @flags: RPC flags
- * @ops: RPC call ops
- * @data: user call data
- */
-struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
-					const struct rpc_call_ops *tk_ops,
-					void *data)
-{
-	struct rpc_task_setup task_setup_data = {
-		.rpc_client = clnt,
-		.callback_ops = tk_ops,
-		.callback_data = data,
-		.flags = flags,
-	};
-
-	return rpc_do_run_task(&task_setup_data);
-}
-EXPORT_SYMBOL_GPL(rpc_run_task);
-
 void
 rpc_call_setup(struct rpc_task *task, const struct rpc_message *msg, int flags)
 {
@@ -1550,7 +1532,7 @@ struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred, int
 		.callback_ops = &rpc_default_ops,
 		.flags = flags,
 	};
-	return rpc_do_run_task(&task_setup_data);
+	return rpc_run_task(&task_setup_data);
 }
 EXPORT_SYMBOL_GPL(rpc_call_null);
 
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index a05493aedb68..7c362e5f6e1b 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -310,6 +310,10 @@ void rpcb_getport_async(struct rpc_task *task)
 	struct rpc_clnt	*rpcb_clnt;
 	static struct rpcbind_args *map;
 	struct rpc_task	*child;
+	struct rpc_task_setup task_setup_data = {
+		.callback_ops = &rpcb_getport_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
 	struct sockaddr addr;
 	int status;
 	struct rpcb_info *info;
@@ -395,7 +399,9 @@ void rpcb_getport_async(struct rpc_task *task)
 	       sizeof(map->r_addr));
 	map->r_owner = RPCB_OWNER_STRING;	/* ignored for GETADDR */
 
-	child = rpc_run_task(rpcb_clnt, RPC_TASK_ASYNC, &rpcb_getport_ops, map);
+	task_setup_data.rpc_client = rpcb_clnt;
+	task_setup_data.callback_data = map;
+	child = rpc_run_task(&task_setup_data);
 	rpc_release_client(rpcb_clnt);
 	if (IS_ERR(child)) {
 		status = -EIO;
-- 
cgit v1.2.3


From 3ff7576ddac06c3d07089e241b40826d24bbf1ac Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat, 14 Jul 2007 15:40:00 -0400
Subject: SUNRPC: Clean up the initialisation of priority queue scheduling
 info.

We want the default scheduling priority (priority == 0) to remain
RPC_PRIORITY_NORMAL.

Also ensure that the priority wait queue scheduling is per process id
instead of sometimes being per thread, and sometimes being per inode.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c              | 10 ----------
 fs/nfs/read.c                |  2 --
 fs/nfs/write.c               | 12 +++++-------
 include/linux/sunrpc/sched.h | 17 +++++++++--------
 net/sunrpc/sched.c           | 30 +++++++++++++++---------------
 5 files changed, 29 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index f9f5fc13dc7d..5bcc764e501a 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -331,8 +331,6 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->read_setup(data);
 
-		data->task.tk_cookie = (unsigned long) inode;
-
 		rpc_execute(&data->task);
 
 		dprintk("NFS: %5u initiated direct read call "
@@ -465,9 +463,6 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE);
 
-		data->task.tk_priority = RPC_PRIORITY_NORMAL;
-		data->task.tk_cookie = (unsigned long) inode;
-
 		/*
 		 * We're called via an RPC callback, so BKL is already held.
 		 */
@@ -534,8 +529,6 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(data->inode)->commit_setup(data, 0);
 
-	data->task.tk_priority = RPC_PRIORITY_NORMAL;
-	data->task.tk_cookie = (unsigned long)data->inode;
 	/* Note: task.tk_ops->rpc_release will free dreq->commit_data */
 	dreq->commit_data = NULL;
 
@@ -718,9 +711,6 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 		rpc_init_task(&data->task, &task_setup_data);
 		NFS_PROTO(inode)->write_setup(data, sync);
 
-		data->task.tk_priority = RPC_PRIORITY_NORMAL;
-		data->task.tk_cookie = (unsigned long) inode;
-
 		rpc_execute(&data->task);
 
 		dprintk("NFS: %5u initiated direct write call "
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index c7f0d5ebd451..8f1eb08ccffa 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -189,8 +189,6 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->read_setup(data);
 
-	data->task.tk_cookie = (unsigned long)inode;
-
 	dprintk("NFS: %5u initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n",
 			data->task.tk_pid,
 			inode->i_sb->s_id,
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index c4376606f106..8d90e90ccd47 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -753,7 +753,7 @@ static void nfs_writepage_release(struct nfs_page *req)
 	nfs_clear_page_tag_locked(req);
 }
 
-static inline int flush_task_priority(int how)
+static int flush_task_priority(int how)
 {
 	switch (how & (FLUSH_HIGHPRI|FLUSH_LOWPRI)) {
 		case FLUSH_HIGHPRI:
@@ -775,11 +775,13 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 {
 	struct inode *inode = req->wb_context->path.dentry->d_inode;
 	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
+	int priority = flush_task_priority(how);
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
 		.callback_ops = call_ops,
 		.callback_data = data,
 		.flags = flags,
+		.priority = priority,
 	};
 
 	/* Set up the RPC argument and reply structs
@@ -805,9 +807,6 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->write_setup(data, how);
 
-	data->task.tk_priority = flush_task_priority(how);
-	data->task.tk_cookie = (unsigned long)inode;
-
 	dprintk("NFS: %5u initiated write call "
 		"(req %s/%Ld, %u bytes @ offset %Lu)\n",
 		data->task.tk_pid,
@@ -1152,11 +1151,13 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	struct nfs_page *first = nfs_list_entry(head->next);
 	struct inode *inode = first->wb_context->path.dentry->d_inode;
 	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
+	int priority = flush_task_priority(how);
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
 		.callback_ops = &nfs_commit_ops,
 		.callback_data = data,
 		.flags = flags,
+		.priority = priority,
 	};
 
 	/* Set up the RPC argument and reply structs
@@ -1180,9 +1181,6 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	rpc_init_task(&data->task, &task_setup_data);
 	NFS_PROTO(inode)->commit_setup(data, how);
 
-	data->task.tk_priority = flush_task_priority(how);
-	data->task.tk_cookie = (unsigned long)inode;
-	
 	dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
 }
 
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index d974421d7647..c9444fdc23ac 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -56,8 +56,6 @@ struct rpc_task {
 	__u8			tk_garb_retry;
 	__u8			tk_cred_retry;
 
-	unsigned long		tk_cookie;	/* Cookie for batching tasks */
-
 	/*
 	 * timeout_fn   to be executed by timer bottom half
 	 * callback	to be executed after waking up
@@ -78,7 +76,6 @@ struct rpc_task {
 	struct timer_list	tk_timer;	/* kernel timer */
 	unsigned long		tk_timeout;	/* timeout for rpc_sleep() */
 	unsigned short		tk_flags;	/* misc flags */
-	unsigned char		tk_priority : 2;/* Task priority */
 	unsigned long		tk_runstate;	/* Task run status */
 	struct workqueue_struct	*tk_workqueue;	/* Normally rpciod, but could
 						 * be any workqueue
@@ -94,6 +91,9 @@ struct rpc_task {
 	unsigned long		tk_start;	/* RPC task init timestamp */
 	long			tk_rtt;		/* round-trip time (jiffies) */
 
+	pid_t			tk_owner;	/* Process id for batching tasks */
+	unsigned char		tk_priority : 2;/* Task priority */
+
 #ifdef RPC_DEBUG
 	unsigned short		tk_pid;		/* debugging aid */
 #endif
@@ -123,6 +123,7 @@ struct rpc_task_setup {
 	const struct rpc_call_ops *callback_ops;
 	void *callback_data;
 	unsigned short flags;
+	signed char priority;
 };
 
 /*
@@ -187,10 +188,10 @@ struct rpc_task_setup {
  * Note: if you change these, you must also change
  * the task initialization definitions below.
  */
-#define RPC_PRIORITY_LOW	0
-#define RPC_PRIORITY_NORMAL	1
-#define RPC_PRIORITY_HIGH	2
-#define RPC_NR_PRIORITY		(RPC_PRIORITY_HIGH+1)
+#define RPC_PRIORITY_LOW	(-1)
+#define RPC_PRIORITY_NORMAL	(0)
+#define RPC_PRIORITY_HIGH	(1)
+#define RPC_NR_PRIORITY		(1 + RPC_PRIORITY_HIGH - RPC_PRIORITY_LOW)
 
 /*
  * RPC synchronization objects
@@ -198,7 +199,7 @@ struct rpc_task_setup {
 struct rpc_wait_queue {
 	spinlock_t		lock;
 	struct list_head	tasks[RPC_NR_PRIORITY];	/* task queue for each priority level */
-	unsigned long		cookie;			/* cookie of last task serviced */
+	pid_t			owner;			/* process id of last task serviced */
 	unsigned char		maxpriority;		/* maximum priority (0 if queue is not a priority queue) */
 	unsigned char		priority;		/* current priority */
 	unsigned char		count;			/* # task groups remaining serviced so far */
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 10216989309c..b9061bcf6fc1 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -135,7 +135,7 @@ static void __rpc_add_wait_queue_priority(struct rpc_wait_queue *queue, struct r
 	if (unlikely(task->tk_priority > queue->maxpriority))
 		q = &queue->tasks[queue->maxpriority];
 	list_for_each_entry(t, q, u.tk_wait.list) {
-		if (t->tk_cookie == task->tk_cookie) {
+		if (t->tk_owner == task->tk_owner) {
 			list_add_tail(&task->u.tk_wait.list, &t->u.tk_wait.links);
 			return;
 		}
@@ -208,26 +208,26 @@ static inline void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int
 	queue->count = 1 << (priority * 2);
 }
 
-static inline void rpc_set_waitqueue_cookie(struct rpc_wait_queue *queue, unsigned long cookie)
+static inline void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid)
 {
-	queue->cookie = cookie;
+	queue->owner = pid;
 	queue->nr = RPC_BATCH_COUNT;
 }
 
 static inline void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue)
 {
 	rpc_set_waitqueue_priority(queue, queue->maxpriority);
-	rpc_set_waitqueue_cookie(queue, 0);
+	rpc_set_waitqueue_owner(queue, 0);
 }
 
-static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, int maxprio)
+static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, unsigned char nr_queues)
 {
 	int i;
 
 	spin_lock_init(&queue->lock);
 	for (i = 0; i < ARRAY_SIZE(queue->tasks); i++)
 		INIT_LIST_HEAD(&queue->tasks[i]);
-	queue->maxpriority = maxprio;
+	queue->maxpriority = nr_queues - 1;
 	rpc_reset_waitqueue_priority(queue);
 #ifdef RPC_DEBUG
 	queue->name = qname;
@@ -236,12 +236,12 @@ static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const c
 
 void rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname)
 {
-	__rpc_init_priority_wait_queue(queue, qname, RPC_PRIORITY_HIGH);
+	__rpc_init_priority_wait_queue(queue, qname, RPC_NR_PRIORITY);
 }
 
 void rpc_init_wait_queue(struct rpc_wait_queue *queue, const char *qname)
 {
-	__rpc_init_priority_wait_queue(queue, qname, 0);
+	__rpc_init_priority_wait_queue(queue, qname, 1);
 }
 EXPORT_SYMBOL_GPL(rpc_init_wait_queue);
 
@@ -456,12 +456,12 @@ static struct rpc_task * __rpc_wake_up_next_priority(struct rpc_wait_queue *queu
 	struct rpc_task *task;
 
 	/*
-	 * Service a batch of tasks from a single cookie.
+	 * Service a batch of tasks from a single owner.
 	 */
 	q = &queue->tasks[queue->priority];
 	if (!list_empty(q)) {
 		task = list_entry(q->next, struct rpc_task, u.tk_wait.list);
-		if (queue->cookie == task->tk_cookie) {
+		if (queue->owner == task->tk_owner) {
 			if (--queue->nr)
 				goto out;
 			list_move_tail(&task->u.tk_wait.list, q);
@@ -470,7 +470,7 @@ static struct rpc_task * __rpc_wake_up_next_priority(struct rpc_wait_queue *queu
 		 * Check if we need to switch queues.
 		 */
 		if (--queue->count)
-			goto new_cookie;
+			goto new_owner;
 	}
 
 	/*
@@ -492,8 +492,8 @@ static struct rpc_task * __rpc_wake_up_next_priority(struct rpc_wait_queue *queu
 
 new_queue:
 	rpc_set_waitqueue_priority(queue, (unsigned int)(q - &queue->tasks[0]));
-new_cookie:
-	rpc_set_waitqueue_cookie(queue, task->tk_cookie);
+new_owner:
+	rpc_set_waitqueue_owner(queue, task->tk_owner);
 out:
 	__rpc_wake_up_task(task);
 	return task;
@@ -830,8 +830,8 @@ void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setu
 	task->tk_garb_retry = 2;
 	task->tk_cred_retry = 2;
 
-	task->tk_priority = RPC_PRIORITY_NORMAL;
-	task->tk_cookie = (unsigned long)current;
+	task->tk_priority = task_setup_data->priority - RPC_PRIORITY_LOW;
+	task->tk_owner = current->tgid;
 
 	/* Initialize workqueue for async tasks */
 	task->tk_workqueue = rpciod_workqueue;
-- 
cgit v1.2.3


From 77de2c590ec72828156d85fa13a96db87301cc68 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 25 Oct 2007 18:40:21 -0400
Subject: SUNRPC: Add a helper rpc_call_start() that initialises
 task->tk_action

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h | 2 +-
 net/sunrpc/clnt.c           | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index ec9704181f94..c79f2edf0323 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -127,7 +127,7 @@ int		rpcb_getport_sync(struct sockaddr_in *, __u32, __u32, int);
 void		rpcb_getport_async(struct rpc_task *);
 
 void		rpc_call_setup(struct rpc_task *, const struct rpc_message *, int);
-
+void		rpc_call_start(struct rpc_task *);
 int		rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg,
 			       int flags, const struct rpc_call_ops *tk_ops,
 			       void *calldata);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index aefe3ae218f7..7aeffeebf42d 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -634,6 +634,13 @@ rpc_call_setup(struct rpc_task *task, const struct rpc_message *msg, int flags)
 }
 EXPORT_SYMBOL_GPL(rpc_call_setup);
 
+void
+rpc_call_start(struct rpc_task *task)
+{
+	task->tk_action = call_start;
+}
+EXPORT_SYMBOL_GPL(rpc_call_start);
+
 /**
  * rpc_peeraddr - extract remote peer address from clnt's xprt
  * @clnt: RPC client structure
-- 
cgit v1.2.3


From bdc7f021f3a1fade77adf3c2d7f65690566fddfe Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat, 14 Jul 2007 15:40:00 -0400
Subject: NFS: Clean up the (commit|read|write)_setup() callback routines

Move the common code for setting up the nfs_write_data and nfs_read_data
structures into fs/nfs/read.c, fs/nfs/write.c and fs/nfs/direct.c.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c         | 45 ++++++++++++++++++++++++++++++++++---------
 fs/nfs/nfs3proc.c       | 41 ++++++---------------------------------
 fs/nfs/nfs4proc.c       | 49 ++++++++---------------------------------------
 fs/nfs/proc.c           | 26 +++++--------------------
 fs/nfs/read.c           | 38 +++++++++++++++++++-----------------
 fs/nfs/write.c          | 51 +++++++++++++++++++++++++++++++++----------------
 include/linux/nfs_xdr.h |  6 +++---
 7 files changed, 113 insertions(+), 143 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 5bcc764e501a..244d1bd7002c 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -272,8 +272,12 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
 	size_t rsize = NFS_SERVER(inode)->rsize;
+	struct rpc_message msg = {
+		.rpc_cred = ctx->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
+		.rpc_message = &msg,
 		.callback_ops = &nfs_read_direct_ops,
 		.flags = RPC_TASK_ASYNC,
 	};
@@ -316,7 +320,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 
 		data->req = (struct nfs_page *) dreq;
 		data->inode = inode;
-		data->cred = ctx->cred;
+		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
 		data->args.context = ctx;
 		data->args.offset = pos;
@@ -326,10 +330,12 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		data->res.fattr = &data->fattr;
 		data->res.eof = 0;
 		data->res.count = bytes;
+		msg.rpc_argp = &data->args;
+		msg.rpc_resp = &data->res;
 
 		task_setup_data.callback_data = data;
+		NFS_PROTO(inode)->read_setup(data, &msg);
 		rpc_init_task(&data->task, &task_setup_data);
-		NFS_PROTO(inode)->read_setup(data);
 
 		rpc_execute(&data->task);
 
@@ -434,6 +440,9 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	struct inode *inode = dreq->inode;
 	struct list_head *p;
 	struct nfs_write_data *data;
+	struct rpc_message msg = {
+		.rpc_cred = dreq->ctx->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
 		.callback_ops = &nfs_write_direct_ops,
@@ -448,6 +457,9 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 
 		get_dreq(dreq);
 
+		/* Use stable writes */
+		data->args.stable = NFS_FILE_SYNC;
+
 		/*
 		 * Reset data->res.
 		 */
@@ -460,8 +472,10 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 		 * since the original request was sent.
 		 */
 		task_setup_data.callback_data = data;
+		msg.rpc_argp = &data->args;
+		msg.rpc_resp = &data->res;
+		NFS_PROTO(inode)->write_setup(data, &msg);
 		rpc_init_task(&data->task, &task_setup_data);
-		NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE);
 
 		/*
 		 * We're called via an RPC callback, so BKL is already held.
@@ -509,15 +523,21 @@ static const struct rpc_call_ops nfs_commit_direct_ops = {
 static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 {
 	struct nfs_write_data *data = dreq->commit_data;
+	struct rpc_message msg = {
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+		.rpc_cred = dreq->ctx->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(dreq->inode),
+		.rpc_message = &msg,
 		.callback_ops = &nfs_commit_direct_ops,
 		.callback_data = data,
 		.flags = RPC_TASK_ASYNC,
 	};
 
 	data->inode = dreq->inode;
-	data->cred = dreq->ctx->cred;
+	data->cred = msg.rpc_cred;
 
 	data->args.fh = NFS_FH(data->inode);
 	data->args.offset = 0;
@@ -526,8 +546,8 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 	data->res.fattr = &data->fattr;
 	data->res.verf = &data->verf;
 
+	NFS_PROTO(data->inode)->commit_setup(data, &msg);
 	rpc_init_task(&data->task, &task_setup_data);
-	NFS_PROTO(data->inode)->commit_setup(data, 0);
 
 	/* Note: task.tk_ops->rpc_release will free dreq->commit_data */
 	dreq->commit_data = NULL;
@@ -649,8 +669,12 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 	struct inode *inode = ctx->path.dentry->d_inode;
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
+	struct rpc_message msg = {
+		.rpc_cred = ctx->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
+		.rpc_message = &msg,
 		.callback_ops = &nfs_write_direct_ops,
 		.flags = RPC_TASK_ASYNC,
 	};
@@ -696,20 +720,23 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 
 		data->req = (struct nfs_page *) dreq;
 		data->inode = inode;
-		data->cred = ctx->cred;
+		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
 		data->args.context = ctx;
 		data->args.offset = pos;
 		data->args.pgbase = pgbase;
 		data->args.pages = data->pagevec;
 		data->args.count = bytes;
+		data->args.stable = sync;
 		data->res.fattr = &data->fattr;
 		data->res.count = bytes;
 		data->res.verf = &data->verf;
 
 		task_setup_data.callback_data = data;
+		msg.rpc_argp = &data->args;
+		msg.rpc_resp = &data->res;
+		NFS_PROTO(inode)->write_setup(data, &msg);
 		rpc_init_task(&data->task, &task_setup_data);
-		NFS_PROTO(inode)->write_setup(data, sync);
 
 		rpc_execute(&data->task);
 
@@ -782,7 +809,7 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov,
 	struct rpc_clnt *clnt = NFS_CLIENT(inode);
 	struct nfs_direct_req *dreq;
 	size_t wsize = NFS_SERVER(inode)->wsize;
-	int sync = 0;
+	int sync = NFS_UNSTABLE;
 
 	dreq = nfs_direct_req_alloc();
 	if (!dreq)
@@ -790,7 +817,7 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov,
 	nfs_alloc_commit_data(dreq);
 
 	if (dreq->commit_data == NULL || count < wsize)
-		sync = FLUSH_STABLE;
+		sync = NFS_FILE_SYNC;
 
 	dreq->inode = inode;
 	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 4cdc2361a669..e68580ebaa47 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -732,16 +732,9 @@ static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data)
 	return 0;
 }
 
-static void nfs3_proc_read_setup(struct nfs_read_data *data)
+static void nfs3_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg)
 {
-	struct rpc_message	msg = {
-		.rpc_proc	= &nfs3_procedures[NFS3PROC_READ],
-		.rpc_argp	= &data->args,
-		.rpc_resp	= &data->res,
-		.rpc_cred	= data->cred,
-	};
-
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ];
 }
 
 static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -753,24 +746,9 @@ static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data)
 	return 0;
 }
 
-static void nfs3_proc_write_setup(struct nfs_write_data *data, int how)
+static void nfs3_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
-	struct rpc_message	msg = {
-		.rpc_proc	= &nfs3_procedures[NFS3PROC_WRITE],
-		.rpc_argp	= &data->args,
-		.rpc_resp	= &data->res,
-		.rpc_cred	= data->cred,
-	};
-
-	data->args.stable = NFS_UNSTABLE;
-	if (how & FLUSH_STABLE) {
-		data->args.stable = NFS_FILE_SYNC;
-		if (NFS_I(data->inode)->ncommit)
-			data->args.stable = NFS_DATA_SYNC;
-	}
-
-	/* Finalize the task. */
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE];
 }
 
 static int nfs3_commit_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -781,16 +759,9 @@ static int nfs3_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	return 0;
 }
 
-static void nfs3_proc_commit_setup(struct nfs_write_data *data, int how)
+static void nfs3_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
-	struct rpc_message	msg = {
-		.rpc_proc	= &nfs3_procedures[NFS3PROC_COMMIT],
-		.rpc_argp	= &data->args,
-		.rpc_resp	= &data->res,
-		.rpc_cred	= data->cred,
-	};
-
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT];
 }
 
 static int
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ff2c5f83ce87..7c0baf23abdc 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2432,18 +2432,10 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 	return 0;
 }
 
-static void nfs4_proc_read_setup(struct nfs_read_data *data)
+static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg)
 {
-	struct rpc_message msg = {
-		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ],
-		.rpc_argp = &data->args,
-		.rpc_resp = &data->res,
-		.rpc_cred = data->cred,
-	};
-
 	data->timestamp   = jiffies;
-
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ];
 }
 
 static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -2461,33 +2453,15 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 	return 0;
 }
 
-static void nfs4_proc_write_setup(struct nfs_write_data *data, int how)
+static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
-	struct rpc_message msg = {
-		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE],
-		.rpc_argp = &data->args,
-		.rpc_resp = &data->res,
-		.rpc_cred = data->cred,
-	};
-	struct inode *inode = data->inode;
-	struct nfs_server *server = NFS_SERVER(inode);
-	int stable;
-	
-	if (how & FLUSH_STABLE) {
-		if (!NFS_I(inode)->ncommit)
-			stable = NFS_FILE_SYNC;
-		else
-			stable = NFS_DATA_SYNC;
-	} else
-		stable = NFS_UNSTABLE;
-	data->args.stable = stable;
+	struct nfs_server *server = NFS_SERVER(data->inode);
+
 	data->args.bitmask = server->attr_bitmask;
 	data->res.server = server;
-
 	data->timestamp   = jiffies;
 
-	/* Finalize the task. */
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE];
 }
 
 static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -2502,20 +2476,13 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	return 0;
 }
 
-static void nfs4_proc_commit_setup(struct nfs_write_data *data, int how)
+static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
-	struct rpc_message msg = {
-		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT],
-		.rpc_argp = &data->args,
-		.rpc_resp = &data->res,
-		.rpc_cred = data->cred,
-	};	
 	struct nfs_server *server = NFS_SERVER(data->inode);
 	
 	data->args.bitmask = server->attr_bitmask;
 	data->res.server = server;
-
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT];
 }
 
 /*
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 4f80d88e9fee..c9f46a24e75c 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -565,16 +565,9 @@ static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data)
 	return 0;
 }
 
-static void nfs_proc_read_setup(struct nfs_read_data *data)
+static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg)
 {
-	struct rpc_message	msg = {
-		.rpc_proc	= &nfs_procedures[NFSPROC_READ],
-		.rpc_argp	= &data->args,
-		.rpc_resp	= &data->res,
-		.rpc_cred	= data->cred,
-	};
-
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs_procedures[NFSPROC_READ];
 }
 
 static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -584,24 +577,15 @@ static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data)
 	return 0;
 }
 
-static void nfs_proc_write_setup(struct nfs_write_data *data, int how)
+static void nfs_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
-	struct rpc_message	msg = {
-		.rpc_proc	= &nfs_procedures[NFSPROC_WRITE],
-		.rpc_argp	= &data->args,
-		.rpc_resp	= &data->res,
-		.rpc_cred	= data->cred,
-	};
-
 	/* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */
 	data->args.stable = NFS_FILE_SYNC;
-
-	/* Finalize the task. */
-	rpc_call_setup(&data->task, &msg, 0);
+	msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE];
 }
 
 static void
-nfs_proc_commit_setup(struct nfs_write_data *data, int how)
+nfs_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
 	BUG();
 }
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 8f1eb08ccffa..e9dbdc8eafe6 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -153,6 +153,16 @@ static void nfs_readpage_release(struct nfs_page *req)
 	nfs_release_request(req);
 }
 
+static void nfs_execute_read(struct nfs_read_data *data)
+{
+	struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
+	sigset_t oldset;
+
+	rpc_clnt_sigmask(clnt, &oldset);
+	rpc_execute(&data->task);
+	rpc_clnt_sigunmask(clnt, &oldset);
+}
+
 /*
  * Set up the NFS read request struct
  */
@@ -162,8 +172,14 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 {
 	struct inode *inode = req->wb_context->path.dentry->d_inode;
 	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
+	struct rpc_message msg = {
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+		.rpc_cred = req->wb_context->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
+		.rpc_message = &msg,
 		.callback_ops = call_ops,
 		.callback_data = data,
 		.flags = RPC_TASK_ASYNC | swap_flags,
@@ -171,7 +187,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 
 	data->req	  = req;
 	data->inode	  = inode;
-	data->cred	  = req->wb_context->cred;
+	data->cred	  = msg.rpc_cred;
 
 	data->args.fh     = NFS_FH(inode);
 	data->args.offset = req_offset(req) + offset;
@@ -186,8 +202,8 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct. */
+	NFS_PROTO(inode)->read_setup(data, &msg);
 	rpc_init_task(&data->task, &task_setup_data);
-	NFS_PROTO(inode)->read_setup(data);
 
 	dprintk("NFS: %5u initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n",
 			data->task.tk_pid,
@@ -195,6 +211,8 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 			(long long)NFS_FILEID(inode),
 			count,
 			(unsigned long long)data->args.offset);
+
+	nfs_execute_read(data);
 }
 
 static void
@@ -210,19 +228,6 @@ nfs_async_read_error(struct list_head *head)
 	}
 }
 
-/*
- * Start an async read operation
- */
-static void nfs_execute_read(struct nfs_read_data *data)
-{
-	struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
-	sigset_t oldset;
-
-	rpc_clnt_sigmask(clnt, &oldset);
-	rpc_execute(&data->task);
-	rpc_clnt_sigunmask(clnt, &oldset);
-}
-
 /*
  * Generate multiple requests to fill a single page.
  *
@@ -277,7 +282,6 @@ static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigne
 				  rsize, offset);
 		offset += rsize;
 		nbytes -= rsize;
-		nfs_execute_read(data);
 	} while (nbytes != 0);
 
 	return 0;
@@ -315,8 +319,6 @@ static int nfs_pagein_one(struct inode *inode, struct list_head *head, unsigned
 	req = nfs_list_entry(data->pages.next);
 
 	nfs_read_rpcsetup(req, data, &nfs_read_full_ops, count, 0);
-
-	nfs_execute_read(data);
 	return 0;
 out_bad:
 	nfs_async_read_error(head);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 8d90e90ccd47..9a69469274ae 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -764,6 +764,16 @@ static int flush_task_priority(int how)
 	return RPC_PRIORITY_NORMAL;
 }
 
+static void nfs_execute_write(struct nfs_write_data *data)
+{
+	struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
+	sigset_t oldset;
+
+	rpc_clnt_sigmask(clnt, &oldset);
+	rpc_execute(&data->task);
+	rpc_clnt_sigunmask(clnt, &oldset);
+}
+
 /*
  * Set up the argument/result storage required for the RPC call.
  */
@@ -776,8 +786,14 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	struct inode *inode = req->wb_context->path.dentry->d_inode;
 	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
 	int priority = flush_task_priority(how);
+	struct rpc_message msg = {
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+		.rpc_cred = req->wb_context->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
+		.rpc_message = &msg,
 		.callback_ops = call_ops,
 		.callback_data = data,
 		.flags = flags,
@@ -789,7 +805,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 
 	data->req = req;
 	data->inode = inode = req->wb_context->path.dentry->d_inode;
-	data->cred = req->wb_context->cred;
+	data->cred = msg.rpc_cred;
 
 	data->args.fh     = NFS_FH(inode);
 	data->args.offset = req_offset(req) + offset;
@@ -797,6 +813,12 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	data->args.pages  = data->pagevec;
 	data->args.count  = count;
 	data->args.context = req->wb_context;
+	data->args.stable  = NFS_UNSTABLE;
+	if (how & FLUSH_STABLE) {
+		data->args.stable = NFS_DATA_SYNC;
+		if (!NFS_I(inode)->ncommit)
+			data->args.stable = NFS_FILE_SYNC;
+	}
 
 	data->res.fattr   = &data->fattr;
 	data->res.count   = count;
@@ -804,8 +826,8 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct.  */
+	NFS_PROTO(inode)->write_setup(data, &msg);
 	rpc_init_task(&data->task, &task_setup_data);
-	NFS_PROTO(inode)->write_setup(data, how);
 
 	dprintk("NFS: %5u initiated write call "
 		"(req %s/%Ld, %u bytes @ offset %Lu)\n",
@@ -814,16 +836,8 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
 		(long long)NFS_FILEID(inode),
 		count,
 		(unsigned long long)data->args.offset);
-}
 
-static void nfs_execute_write(struct nfs_write_data *data)
-{
-	struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
-	sigset_t oldset;
-
-	rpc_clnt_sigmask(clnt, &oldset);
-	rpc_execute(&data->task);
-	rpc_clnt_sigunmask(clnt, &oldset);
+	nfs_execute_write(data);
 }
 
 /*
@@ -870,7 +884,6 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, unsigned
 				   wsize, offset, how);
 		offset += wsize;
 		nbytes -= wsize;
-		nfs_execute_write(data);
 	} while (nbytes != 0);
 
 	return 0;
@@ -918,7 +931,6 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, unsigned i
 	/* Set up the argument struct */
 	nfs_write_rpcsetup(req, data, &nfs_write_full_ops, count, 0, how);
 
-	nfs_execute_write(data);
 	return 0;
  out_bad:
 	while (!list_empty(head)) {
@@ -1152,8 +1164,14 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	struct inode *inode = first->wb_context->path.dentry->d_inode;
 	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
 	int priority = flush_task_priority(how);
+	struct rpc_message msg = {
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+		.rpc_cred = first->wb_context->cred,
+	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = NFS_CLIENT(inode),
+		.rpc_message = &msg,
 		.callback_ops = &nfs_commit_ops,
 		.callback_data = data,
 		.flags = flags,
@@ -1166,7 +1184,7 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	list_splice_init(head, &data->pages);
 
 	data->inode	  = inode;
-	data->cred	  = first->wb_context->cred;
+	data->cred	  = msg.rpc_cred;
 
 	data->args.fh     = NFS_FH(data->inode);
 	/* Note: we always request a commit of the entire inode */
@@ -1178,10 +1196,12 @@ static void nfs_commit_rpcsetup(struct list_head *head,
 	nfs_fattr_init(&data->fattr);
 
 	/* Set up the initial task struct.  */
+	NFS_PROTO(inode)->commit_setup(data, &msg);
 	rpc_init_task(&data->task, &task_setup_data);
-	NFS_PROTO(inode)->commit_setup(data, how);
 
 	dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
+
+	nfs_execute_write(data);
 }
 
 /*
@@ -1201,7 +1221,6 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how)
 	/* Set up the argument struct */
 	nfs_commit_rpcsetup(head, data, how);
 
-	nfs_execute_write(data);
 	return 0;
  out_bad:
 	while (!list_empty(head)) {
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index daab252f2e5c..6b213a64b15f 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -816,11 +816,11 @@ struct nfs_rpc_ops {
 			     struct nfs_pathconf *);
 	int	(*set_capabilities)(struct nfs_server *, struct nfs_fh *);
 	__be32 *(*decode_dirent)(__be32 *, struct nfs_entry *, int plus);
-	void	(*read_setup)   (struct nfs_read_data *);
+	void	(*read_setup)   (struct nfs_read_data *, struct rpc_message *);
 	int	(*read_done)  (struct rpc_task *, struct nfs_read_data *);
-	void	(*write_setup)  (struct nfs_write_data *, int how);
+	void	(*write_setup)  (struct nfs_write_data *, struct rpc_message *);
 	int	(*write_done)  (struct rpc_task *, struct nfs_write_data *);
-	void	(*commit_setup) (struct nfs_write_data *, int how);
+	void	(*commit_setup) (struct nfs_write_data *, struct rpc_message *);
 	int	(*commit_done) (struct rpc_task *, struct nfs_write_data *);
 	int	(*file_open)   (struct inode *, struct file *);
 	int	(*file_release) (struct inode *, struct file *);
-- 
cgit v1.2.3


From b5627943ab6fabbc13a45d92683363a3d08a249f Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 25 Oct 2007 18:42:21 -0400
Subject: SUNRPC: Remove the now unused function rpc_call_setup()

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  1 -
 net/sunrpc/clnt.c           | 18 ------------------
 2 files changed, 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index c79f2edf0323..4fc268ab6f3e 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -126,7 +126,6 @@ int		rpcb_register(u32, u32, int, unsigned short, int *);
 int		rpcb_getport_sync(struct sockaddr_in *, __u32, __u32, int);
 void		rpcb_getport_async(struct rpc_task *);
 
-void		rpc_call_setup(struct rpc_task *, const struct rpc_message *, int);
 void		rpc_call_start(struct rpc_task *);
 int		rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg,
 			       int flags, const struct rpc_call_ops *tk_ops,
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 6eb79c49c937..e0a5e47a1d4b 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -613,24 +613,6 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags,
 }
 EXPORT_SYMBOL_GPL(rpc_call_async);
 
-void
-rpc_call_setup(struct rpc_task *task, const struct rpc_message *msg, int flags)
-{
-	task->tk_msg   = *msg;
-	task->tk_flags |= flags;
-	/* Bind the user cred */
-	if (task->tk_msg.rpc_cred != NULL)
-		rpcauth_holdcred(task);
-	else
-		rpcauth_bindcred(task);
-
-	if (task->tk_status == 0)
-		task->tk_action = call_start;
-	else
-		task->tk_action = rpc_exit_task;
-}
-EXPORT_SYMBOL_GPL(rpc_call_setup);
-
 void
 rpc_call_start(struct rpc_task *task)
 {
-- 
cgit v1.2.3


From e8f5d77c8029ff8f5dcd1dfc133aac0bbbffd92b Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 25 Oct 2007 18:42:53 -0400
Subject: SUNRPC: allow the caller of rpc_run_task to preallocate the struct
 rpc_task

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h |  1 +
 net/sunrpc/sched.c           | 16 ++++++++++------
 2 files changed, 11 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index c9444fdc23ac..60a05c71637e 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -118,6 +118,7 @@ struct rpc_call_ops {
 };
 
 struct rpc_task_setup {
+	struct rpc_task *task;
 	struct rpc_clnt *rpc_client;
 	const struct rpc_message *rpc_message;
 	const struct rpc_call_ops *callback_ops;
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index fa53a88b2c5b..c03e7bf6e9bc 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -885,16 +885,20 @@ static void rpc_free_task(struct rcu_head *rcu)
  */
 struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data)
 {
-	struct rpc_task	*task;
-
-	task = rpc_alloc_task();
-	if (!task)
-		goto out;
+	struct rpc_task	*task = setup_data->task;
+	unsigned short flags = 0;
+
+	if (task == NULL) {
+		task = rpc_alloc_task();
+		if (task == NULL)
+			goto out;
+		flags = RPC_TASK_DYNAMIC;
+	}
 
 	rpc_init_task(task, setup_data);
 
+	task->tk_flags |= flags;
 	dprintk("RPC:       allocated task %p\n", task);
-	task->tk_flags |= RPC_TASK_DYNAMIC;
 out:
 	return task;
 }
-- 
cgit v1.2.3


From 47fe064831a2a949f6c1e0086f61a105e99ea867 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 25 Oct 2007 18:42:55 -0400
Subject: SUNRPC: Unexport rpc_init_task() and rpc_execute()

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h | 1 -
 net/sunrpc/sched.c           | 4 +---
 2 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 60a05c71637e..23481a5e1f53 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -247,7 +247,6 @@ struct rpc_wait_queue {
  */
 struct rpc_task *rpc_new_task(const struct rpc_task_setup *);
 struct rpc_task *rpc_run_task(const struct rpc_task_setup *);
-void		rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *);
 void		rpc_put_task(struct rpc_task *);
 void		rpc_exit_task(struct rpc_task *);
 void		rpc_release_calldata(const struct rpc_call_ops *, void *);
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index c03e7bf6e9bc..ce6cfae91e88 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -737,7 +737,6 @@ void rpc_execute(struct rpc_task *task)
 	rpc_set_running(task);
 	__rpc_execute(task);
 }
-EXPORT_SYMBOL_GPL(rpc_execute);
 
 static void rpc_async_schedule(struct work_struct *work)
 {
@@ -815,7 +814,7 @@ EXPORT_SYMBOL_GPL(rpc_free);
 /*
  * Creation and deletion of RPC task structures
  */
-void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setup_data)
+static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setup_data)
 {
 	memset(task, 0, sizeof(*task));
 	setup_timer(&task->tk_timer, (void (*)(unsigned long))rpc_run_timer,
@@ -865,7 +864,6 @@ void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setu
 	dprintk("RPC:       new task initialized, procpid %u\n",
 				task_pid_nr(current));
 }
-EXPORT_SYMBOL_GPL(rpc_init_task);
 
 static struct rpc_task *
 rpc_alloc_task(void)
-- 
cgit v1.2.3


From c087567d3ffb2c7c61e091982e6ca45478394f1a Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 18 Jul 2007 18:32:38 -0400
Subject: SUNRPC: Remove the obsolete RPC_WAITQ macro

Now that we've killed off all the users.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h | 23 -----------------------
 1 file changed, 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 23481a5e1f53..ce3d1b132729 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -217,29 +217,6 @@ struct rpc_wait_queue {
  * performance of NFS operations such as read/write.
  */
 #define RPC_BATCH_COUNT			16
-
-#ifndef RPC_DEBUG
-# define RPC_WAITQ_INIT(var,qname) { \
-		.lock = __SPIN_LOCK_UNLOCKED(var.lock), \
-		.tasks = { \
-			[0] = LIST_HEAD_INIT(var.tasks[0]), \
-			[1] = LIST_HEAD_INIT(var.tasks[1]), \
-			[2] = LIST_HEAD_INIT(var.tasks[2]), \
-		}, \
-	}
-#else
-# define RPC_WAITQ_INIT(var,qname) { \
-		.lock = __SPIN_LOCK_UNLOCKED(var.lock), \
-		.tasks = { \
-			[0] = LIST_HEAD_INIT(var.tasks[0]), \
-			[1] = LIST_HEAD_INIT(var.tasks[1]), \
-			[2] = LIST_HEAD_INIT(var.tasks[2]), \
-		}, \
-		.name = qname, \
-	}
-#endif
-# define RPC_WAITQ(var,qname)      struct rpc_wait_queue var = RPC_WAITQ_INIT(var,qname)
-
 #define RPC_IS_PRIORITY(q)		((q)->maxpriority > 0)
 
 /*
-- 
cgit v1.2.3


From bfc69a456642a51c89dfd8e5184468857cb44f32 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 15 Oct 2007 18:18:29 -0400
Subject: NFS: define a function to update nfsi->cache_change_attribute

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 15 +++++++++++++++
 fs/nfs/inode.c         |  6 ++++--
 fs/nfs/nfs4proc.c      |  2 +-
 include/linux/nfs_fs.h |  1 +
 4 files changed, 21 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 32c666c612a1..72d141a0dbd8 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -638,6 +638,21 @@ static int nfs_fsync_dir(struct file *filp, struct dentry *dentry, int datasync)
 	return 0;
 }
 
+/**
+ * nfs_force_lookup_revalidate - Mark the directory as having changed
+ * @dir - pointer to directory inode
+ *
+ * This forces the revalidation code in nfs_lookup_revalidate() to do a
+ * full lookup on all child dentries of 'dir' whenever a change occurs
+ * on the server that might have invalidated our dcache.
+ *
+ * The caller should be holding dir->i_lock
+ */
+void nfs_force_lookup_revalidate(struct inode *dir)
+{
+	NFS_I(dir)->cache_change_attribute = jiffies;
+}
+
 /*
  * A check for whether or not the parent directory has changed.
  * In the case it has, we assume that the dentries are untrustworthy
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index cc3a09db41a9..5747d49bdd76 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1029,7 +1029,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 			dprintk("NFS: mtime change on server for file %s/%ld\n",
 					inode->i_sb->s_id, inode->i_ino);
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
-			nfsi->cache_change_attribute = now;
+			if (S_ISDIR(inode->i_mode))
+				nfs_force_lookup_revalidate(inode);
 		}
 		/* If ctime has changed we should definitely clear access+acl caches */
 		if (!timespec_equal(&inode->i_ctime, &fattr->ctime))
@@ -1038,7 +1039,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		dprintk("NFS: change_attr change on server for file %s/%ld\n",
 				inode->i_sb->s_id, inode->i_ino);
 		invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
-		nfsi->cache_change_attribute = now;
+		if (S_ISDIR(inode->i_mode))
+			nfs_force_lookup_revalidate(inode);
 	}
 
 	/* Check if our cached file size is stale */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 826b445b8c70..26192a703129 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -210,7 +210,7 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 	spin_lock(&dir->i_lock);
 	nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA;
 	if (!cinfo->atomic || cinfo->before != nfsi->change_attr)
-		nfsi->cache_change_attribute = jiffies;
+		nfs_force_lookup_revalidate(dir);
 	nfsi->change_attr = cinfo->after;
 	spin_unlock(&dir->i_lock);
 }
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 2d15d4aac094..7095aaa087d8 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -366,6 +366,7 @@ extern const struct inode_operations nfs3_dir_inode_operations;
 extern const struct file_operations nfs_dir_operations;
 extern struct dentry_operations nfs_dentry_operations;
 
+extern void nfs_force_lookup_revalidate(struct inode *dir);
 extern int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fh, struct nfs_fattr *fattr);
 extern int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags);
 extern void nfs_access_zap_cache(struct inode *inode);
-- 
cgit v1.2.3


From 40c553193df41920de659f0446e5d214c862e827 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Fri, 14 Dec 2007 14:56:07 -0500
Subject: NFS: Remove the redundant nfs_client->cl_nfsversion

We can get the same information from the rpc_ops structure instead.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 41 ++++++++++++++++++-----------------------
 fs/nfs/namespace.c        |  2 +-
 fs/nfs/super.c            |  2 +-
 include/linux/nfs_fs_sb.h |  1 -
 4 files changed, 20 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index ff778ecee0bd..3b21731ae571 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -96,7 +96,7 @@ struct rpc_program		nfsacl_program = {
 struct nfs_client_initdata {
 	const char *hostname;
 	const struct sockaddr_in *addr;
-	int version;
+	const struct nfs_rpc_ops *rpc_ops;
 };
 
 /*
@@ -112,7 +112,9 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL)
 		goto error_0;
 
-	if (cl_init->version == 4) {
+	clp->rpc_ops = cl_init->rpc_ops;
+
+	if (cl_init->rpc_ops->version == 4) {
 		if (nfs_callback_up() < 0)
 			goto error_2;
 		__set_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
@@ -121,7 +123,6 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	atomic_set(&clp->cl_count, 1);
 	clp->cl_cons_state = NFS_CS_INITING;
 
-	clp->cl_nfsversion = cl_init->version;
 	memcpy(&clp->cl_addr, cl_init->addr, sizeof(clp->cl_addr));
 
 	if (cl_init->hostname) {
@@ -170,7 +171,7 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
  */
 static void nfs_free_client(struct nfs_client *clp)
 {
-	dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion);
+	dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version);
 
 	nfs4_shutdown_client(clp);
 
@@ -222,7 +223,7 @@ struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversio
 			continue;
 
 		/* Different NFS versions cannot share the same nfs_client */
-		if (clp->cl_nfsversion != nfsversion)
+		if (clp->rpc_ops->version != nfsversion)
 			continue;
 
 		/* Match only the IP address, not the port number */
@@ -251,7 +252,7 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 			continue;
 
 		/* Different NFS versions cannot share the same nfs_client */
-		if (clp->cl_nfsversion != data->version)
+		if (clp->rpc_ops != data->rpc_ops)
 			continue;
 
 		/* Match the full socket address */
@@ -273,9 +274,9 @@ static struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_in
 	struct nfs_client *clp, *new = NULL;
 	int error;
 
-	dprintk("--> nfs_get_client(%s,"NIPQUAD_FMT":%d,%d)\n",
+	dprintk("--> nfs_get_client(%s,"NIPQUAD_FMT":%d,%u)\n",
 		cl_init->hostname ?: "", NIPQUAD(cl_init->addr->sin_addr),
-		cl_init->addr->sin_port, cl_init->version);
+		cl_init->addr->sin_port, cl_init->rpc_ops->version);
 
 	/* see if the client already exists */
 	do {
@@ -430,7 +431,7 @@ static int nfs_start_lockd(struct nfs_server *server)
 {
 	int error = 0;
 
-	if (server->nfs_client->cl_nfsversion > 3)
+	if (server->nfs_client->rpc_ops->version > 3)
 		goto out;
 	if (server->flags & NFS_MOUNT_NONLM)
 		goto out;
@@ -450,7 +451,7 @@ out:
 #ifdef CONFIG_NFS_V3_ACL
 static void nfs_init_server_aclclient(struct nfs_server *server)
 {
-	if (server->nfs_client->cl_nfsversion != 3)
+	if (server->nfs_client->rpc_ops->version != 3)
 		goto out_noacl;
 	if (server->flags & NFS_MOUNT_NOACL)
 		goto out_noacl;
@@ -521,12 +522,6 @@ static int nfs_init_client(struct nfs_client *clp,
 		return 0;
 	}
 
-	/* Check NFS protocol revision and initialize RPC op vector */
-	clp->rpc_ops = &nfs_v2_clientops;
-#ifdef CONFIG_NFS_V3
-	if (clp->cl_nfsversion == 3)
-		clp->rpc_ops = &nfs_v3_clientops;
-#endif
 	/*
 	 * Create a client RPC handle for doing FSSTAT with UNIX auth only
 	 * - RFC 2623, sec 2.3.2
@@ -553,7 +548,7 @@ static int nfs_init_server(struct nfs_server *server,
 	struct nfs_client_initdata cl_init = {
 		.hostname = data->nfs_server.hostname,
 		.addr = &data->nfs_server.address,
-		.version = 2,
+		.rpc_ops = &nfs_v2_clientops,
 	};
 	struct nfs_client *clp;
 	int error;
@@ -562,7 +557,7 @@ static int nfs_init_server(struct nfs_server *server,
 
 #ifdef CONFIG_NFS_V3
 	if (data->flags & NFS_MOUNT_VER3)
-		cl_init.version = 3;
+		cl_init.rpc_ops = &nfs_v3_clientops;
 #endif
 
 	/* Allocate or find a client reference we can use */
@@ -906,7 +901,7 @@ static int nfs4_set_client(struct nfs_server *server,
 	struct nfs_client_initdata cl_init = {
 		.hostname = hostname,
 		.addr = addr,
-		.version = 4,
+		.rpc_ops = &nfs_v4_clientops,
 	};
 	struct nfs_client *clp;
 	int error;
@@ -1284,8 +1279,8 @@ static int nfs_server_list_show(struct seq_file *m, void *v)
 	/* display one transport per line on subsequent lines */
 	clp = list_entry(v, struct nfs_client, cl_share_link);
 
-	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %3d %s\n",
-		   clp->cl_nfsversion,
+	seq_printf(m, "v%u %02x%02x%02x%02x %4hx %3d %s\n",
+		   clp->rpc_ops->version,
 		   NIPQUAD(clp->cl_addr.sin_addr),
 		   ntohs(clp->cl_addr.sin_port),
 		   atomic_read(&clp->cl_count),
@@ -1363,8 +1358,8 @@ static int nfs_volume_list_show(struct seq_file *m, void *v)
 		 (unsigned long long) server->fsid.major,
 		 (unsigned long long) server->fsid.minor);
 
-	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s\n",
-		   clp->cl_nfsversion,
+	seq_printf(m, "v%u %02x%02x%02x%02x %4hx %-7s %-17s\n",
+		   clp->rpc_ops->version,
 		   NIPQUAD(clp->cl_addr.sin_addr),
 		   ntohs(clp->cl_addr.sin_port),
 		   dev,
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index acfc56f9edc0..be4ce1c3a3d8 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -188,7 +188,7 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
 {
 #ifdef CONFIG_NFS_V4
 	struct vfsmount *mnt = NULL;
-	switch (server->nfs_client->cl_nfsversion) {
+	switch (server->nfs_client->rpc_ops->version) {
 		case 2:
 		case 3:
 			mnt = vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index a3492d6f8f9b..5608e6a4c1e1 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -529,7 +529,7 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt)
 	seq_printf(m, ",namelen=%d", nfss->namelen);
 
 #ifdef CONFIG_NFS_V4
-	if (nfss->nfs_client->cl_nfsversion == 4) {
+	if (nfss->nfs_client->rpc_ops->version == 4) {
 		seq_printf(m, "\n\tnfsv4:\t");
 		seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]);
 		seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]);
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 9f949b587684..97b325756667 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -17,7 +17,6 @@ struct nfs_client {
 	int			cl_cons_state;	/* current construction state (-ve: init error) */
 #define NFS_CS_READY		0		/* ready to be used */
 #define NFS_CS_INITING		1		/* busy initialising */
-	int			cl_nfsversion;	/* NFS protocol version */
 	unsigned long		cl_res_state;	/* NFS resources state */
 #define NFS_CS_CALLBACK		1		/* - callback started */
 #define NFS_CS_IDMAP		2		/* - idmap started */
-- 
cgit v1.2.3


From 0fb2b7e945f55a8317e5f58db7c068aab5b825a1 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 10 Dec 2007 14:56:46 -0500
Subject: SUNRPC: Move universal address definitions to global header

Universal addresses are defined in RFC 1833 and clarified in RFC 3530.  We
need to use them in several places in the NFS and RPC clients, so move the
relevant definition and block comment to an appropriate global include
file.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/msg_prot.h | 39 +++++++++++++++++++++++++++++++++++
 net/sunrpc/rpcb_clnt.c          | 45 +++--------------------------------------
 2 files changed, 42 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h
index c4beb5775111..70df4f1d8847 100644
--- a/include/linux/sunrpc/msg_prot.h
+++ b/include/linux/sunrpc/msg_prot.h
@@ -152,5 +152,44 @@ typedef __be32	rpc_fraghdr;
  */
 #define RPCBIND_MAXNETIDLEN	(4u)
 
+/*
+ * Universal addresses are introduced in RFC 1833 and further spelled
+ * out in RFC 3530.  RPCBIND_MAXUADDRLEN defines a maximum byte length
+ * of a universal address for use in allocating buffers and character
+ * arrays.
+ *
+ * Quoting RFC 3530, section 2.2:
+ *
+ * For TCP over IPv4 and for UDP over IPv4, the format of r_addr is the
+ * US-ASCII string:
+ *
+ *	h1.h2.h3.h4.p1.p2
+ *
+ * The prefix, "h1.h2.h3.h4", is the standard textual form for
+ * representing an IPv4 address, which is always four octets long.
+ * Assuming big-endian ordering, h1, h2, h3, and h4, are respectively,
+ * the first through fourth octets each converted to ASCII-decimal.
+ * Assuming big-endian ordering, p1 and p2 are, respectively, the first
+ * and second octets each converted to ASCII-decimal.  For example, if a
+ * host, in big-endian order, has an address of 0x0A010307 and there is
+ * a service listening on, in big endian order, port 0x020F (decimal
+ * 527), then the complete universal address is "10.1.3.7.2.15".
+ *
+ * ...
+ *
+ * For TCP over IPv6 and for UDP over IPv6, the format of r_addr is the
+ * US-ASCII string:
+ *
+ *	x1:x2:x3:x4:x5:x6:x7:x8.p1.p2
+ *
+ * The suffix "p1.p2" is the service port, and is computed the same way
+ * as with universal addresses for TCP and UDP over IPv4.  The prefix,
+ * "x1:x2:x3:x4:x5:x6:x7:x8", is the standard textual form for
+ * representing an IPv6 address as defined in Section 2.2 of [RFC2373].
+ * Additionally, the two alternative forms specified in Section 2.2 of
+ * [RFC2373] are also acceptable.
+ */
+#define RPCBIND_MAXUADDRLEN	(56u)
+
 #endif /* __KERNEL__ */
 #endif /* _LINUX_SUNRPC_MSGPROT_H_ */
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 9696b5127060..f494e58910ec 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -54,45 +54,6 @@ enum {
 #define RPCB_HIGHPROC_3		RPCBPROC_TADDR2UADDR
 #define RPCB_HIGHPROC_4		RPCBPROC_GETSTAT
 
-/*
- * r_addr
- *
- * Quoting RFC 3530, section 2.2:
- *
- * For TCP over IPv4 and for UDP over IPv4, the format of r_addr is the
- * US-ASCII string:
- *
- *	h1.h2.h3.h4.p1.p2
- *
- * The prefix, "h1.h2.h3.h4", is the standard textual form for
- * representing an IPv4 address, which is always four octets long.
- * Assuming big-endian ordering, h1, h2, h3, and h4, are respectively,
- * the first through fourth octets each converted to ASCII-decimal.
- * Assuming big-endian ordering, p1 and p2 are, respectively, the first
- * and second octets each converted to ASCII-decimal.  For example, if a
- * host, in big-endian order, has an address of 0x0A010307 and there is
- * a service listening on, in big endian order, port 0x020F (decimal
- * 527), then the complete universal address is "10.1.3.7.2.15".
- *
- * ...
- *
- * For TCP over IPv6 and for UDP over IPv6, the format of r_addr is the
- * US-ASCII string:
- *
- *	x1:x2:x3:x4:x5:x6:x7:x8.p1.p2
- *
- * The suffix "p1.p2" is the service port, and is computed the same way
- * as with universal addresses for TCP and UDP over IPv4.  The prefix,
- * "x1:x2:x3:x4:x5:x6:x7:x8", is the standard textual form for
- * representing an IPv6 address as defined in Section 2.2 of [RFC2373].
- * Additionally, the two alternative forms specified in Section 2.2 of
- * [RFC2373] are also acceptable.
- *
- * XXX: Currently this implementation does not explicitly convert the
- *      stored address to US-ASCII on non-ASCII systems.
- */
-#define RPCB_MAXADDRLEN		(128u)
-
 /*
  * r_owner
  *
@@ -113,7 +74,7 @@ struct rpcbind_args {
 	u32			r_prot;
 	unsigned short		r_port;
 	char *			r_netid;
-	char			r_addr[RPCB_MAXADDRLEN];
+	char			r_addr[RPCBIND_MAXUADDRLEN];
 	char *			r_owner;
 };
 
@@ -526,7 +487,7 @@ static int rpcb_decode_getaddr(struct rpc_rqst *req, __be32 *p,
 	 * Simple sanity check.  The smallest possible universal
 	 * address is an IPv4 address string containing 11 bytes.
 	 */
-	if (addr_len < 11 || addr_len > RPCB_MAXADDRLEN)
+	if (addr_len < 11 || addr_len > RPCBIND_MAXUADDRLEN)
 		goto out_err;
 
 	/*
@@ -577,7 +538,7 @@ out_err:
 #define RPCB_boolean_sz		(1u)
 
 #define RPCB_netid_sz		(1+XDR_QUADLEN(RPCBIND_MAXNETIDLEN))
-#define RPCB_addr_sz		(1+XDR_QUADLEN(RPCB_MAXADDRLEN))
+#define RPCB_addr_sz		(1+XDR_QUADLEN(RPCBIND_MAXUADDRLEN))
 #define RPCB_ownerstring_sz	(1+XDR_QUADLEN(RPCB_MAXOWNERLEN))
 
 #define RPCB_mappingargs_sz	RPCB_program_sz+RPCB_version_sz+	\
-- 
cgit v1.2.3


From cc38bac3a0093b3b7928efc6ff8e9faf9e75f41d Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 10 Dec 2007 14:56:54 -0500
Subject: NFS: Ensure NFSv4 SETCLIENTID send buffer is large enough

Ensure that the RPC buffer size specified for NFSv4 SETCLIENTID procedures
matches what we are encoding into the buffer.  See the definition of
struct nfs4_setclientid {} and the encode_setclientid() function.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c        | 10 ++++++----
 include/linux/nfs_xdr.h | 13 +++++++------
 2 files changed, 13 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index eae46f008da7..db1ed9c46ede 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -116,10 +116,12 @@ static int nfs4_stat_to_errno(int);
 #define decode_renew_maxsz	(op_decode_hdr_maxsz)
 #define encode_setclientid_maxsz \
 				(op_encode_hdr_maxsz + \
-				4 /*server->ip_addr*/ + \
-				1 /*Netid*/ + \
-				6 /*uaddr*/ + \
-				6 + (NFS4_VERIFIER_SIZE >> 2))
+				XDR_QUADLEN(NFS4_VERIFIER_SIZE) + \
+				XDR_QUADLEN(NFS4_SETCLIENTID_NAMELEN) + \
+				1 /* sc_prog */ + \
+				XDR_QUADLEN(RPCBIND_MAXNETIDLEN) + \
+				XDR_QUADLEN(RPCBIND_MAXUADDRLEN) + \
+				1) /* sc_cb_ident */
 #define decode_setclientid_maxsz \
 				(op_decode_hdr_maxsz + \
 				2 + \
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 6b213a64b15f..d8e395d0c7c9 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -666,16 +666,17 @@ struct nfs4_rename_res {
 	struct nfs_fattr *		new_fattr;
 };
 
+#define NFS4_SETCLIENTID_NAMELEN	(48)
 struct nfs4_setclientid {
-	const nfs4_verifier *		sc_verifier;      /* request */
+	const nfs4_verifier *		sc_verifier;
 	unsigned int			sc_name_len;
-	char				sc_name[48];	  /* request */
-	u32				sc_prog;          /* request */
+	char				sc_name[NFS4_SETCLIENTID_NAMELEN];
+	u32				sc_prog;
 	unsigned int			sc_netid_len;
-	char				sc_netid[4];	  /* request */
+	char				sc_netid[RPCBIND_MAXNETIDLEN];
 	unsigned int			sc_uaddr_len;
-	char				sc_uaddr[24];     /* request */
-	u32				sc_cb_ident;      /* request */
+	char				sc_uaddr[RPCBIND_MAXUADDRLEN];
+	u32				sc_cb_ident;
 };
 
 struct nfs4_statfs_arg {
-- 
cgit v1.2.3


From 4392f2592297876967191238a341667a6d4fc456 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 10 Dec 2007 14:57:01 -0500
Subject: NFS: Increase size of cl_ipaddr field to hold IPv6 addresses

The nfs_client's cl_ipaddr field needs to be larger to hold strings that
represent IPv6 addresses.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Cc: Aurelien Charbon <aurelien.charbon@ext.bull.net>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/nfs_fs_sb.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 97b325756667..0b3dcba3d97d 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -64,7 +64,7 @@ struct nfs_client {
 	/* Our own IP address, as a null-terminated string.
 	 * This is used to generate the clientid, and the callback address.
 	 */
-	char			cl_ipaddr[16];
+	char			cl_ipaddr[48];
 	unsigned char		cl_id_uniquifier;
 #endif
 };
-- 
cgit v1.2.3


From 6e4cffd7b2cf86022dcf9cceeb63f16ff852caa1 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 10 Dec 2007 14:58:15 -0500
Subject: NFS: Expand server address storage in nfs_client struct

Prepare for managing larger addresses in the NFS client by widening the
nfs_client struct's cl_addr field.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Cc: Aurelien Charbon <aurelien.charbon@ext.bull.net>

(Modified to work with the new parameters for nfs_alloc_client)
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 8 ++++++--
 include/linux/nfs_fs_sb.h | 3 ++-
 2 files changed, 8 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 876162cddf1e..44fe7fd7bfbf 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -98,6 +98,7 @@ struct rpc_program		nfsacl_program = {
 struct nfs_client_initdata {
 	const char *hostname;
 	const struct sockaddr_in *addr;
+	size_t addrlen;
 	const struct nfs_rpc_ops *rpc_ops;
 };
 
@@ -125,7 +126,8 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	atomic_set(&clp->cl_count, 1);
 	clp->cl_cons_state = NFS_CS_INITING;
 
-	memcpy(&clp->cl_addr, cl_init->addr, sizeof(clp->cl_addr));
+	memcpy(&clp->cl_addr, cl_init->addr, cl_init->addrlen);
+	clp->cl_addrlen = cl_init->addrlen;
 
 	if (cl_init->hostname) {
 		clp->cl_hostname = kstrdup(cl_init->hostname, GFP_KERNEL);
@@ -425,7 +427,7 @@ static int nfs_create_rpc_client(struct nfs_client *clp, int proto,
 	struct rpc_create_args args = {
 		.protocol	= proto,
 		.address	= (struct sockaddr *)&clp->cl_addr,
-		.addrsize	= sizeof(clp->cl_addr),
+		.addrsize	= clp->cl_addrlen,
 		.timeout	= &timeparms,
 		.servername	= clp->cl_hostname,
 		.program	= &nfs_program,
@@ -585,6 +587,7 @@ static int nfs_init_server(struct nfs_server *server,
 	struct nfs_client_initdata cl_init = {
 		.hostname = data->nfs_server.hostname,
 		.addr = &data->nfs_server.address,
+		.addrlen = sizeof(data->nfs_server.address),
 		.rpc_ops = &nfs_v2_clientops,
 	};
 	struct nfs_client *clp;
@@ -938,6 +941,7 @@ static int nfs4_set_client(struct nfs_server *server,
 	struct nfs_client_initdata cl_init = {
 		.hostname = hostname,
 		.addr = addr,
+		.addrlen = sizeof(*addr),
 		.rpc_ops = &nfs_v4_clientops,
 	};
 	struct nfs_client *clp;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 0b3dcba3d97d..912cfe84ea86 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -21,7 +21,8 @@ struct nfs_client {
 #define NFS_CS_CALLBACK		1		/* - callback started */
 #define NFS_CS_IDMAP		2		/* - idmap started */
 #define NFS_CS_RENEWD		3		/* - renewd started */
-	struct sockaddr_in	cl_addr;	/* server identifier */
+	struct sockaddr_storage	cl_addr;	/* server identifier */
+	size_t			cl_addrlen;
 	char *			cl_hostname;	/* hostname of server */
 	struct list_head	cl_share_link;	/* link in global client list */
 	struct list_head	cl_superblocks;	/* List of nfs_server structs */
-- 
cgit v1.2.3


From 69dd716c5ffd89f5ba14ffb871d633ecea74d13a Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Fri, 14 Dec 2007 14:56:07 -0500
Subject: NFSv4: Add socket proto argument to setclientid

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 4 +++-
 include/linux/nfs_xdr.h | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5e8c4cf7959e..b3d4e8e5696a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2891,10 +2891,12 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po
 
 	for(;;) {
 		setclientid.sc_name_len = scnprintf(setclientid.sc_name,
-				sizeof(setclientid.sc_name), "%s/%s %s %u",
+				sizeof(setclientid.sc_name), "%s/%s %s %s %u",
 				clp->cl_ipaddr,
 				rpc_peeraddr2str(clp->cl_rpcclient,
 							RPC_DISPLAY_ADDR),
+				rpc_peeraddr2str(clp->cl_rpcclient,
+							RPC_DISPLAY_PROTO),
 				cred->cr_ops->cr_name,
 				clp->cl_id_uniquifier);
 		setclientid.sc_netid_len = scnprintf(setclientid.sc_netid,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d8e395d0c7c9..d342ae5c76ab 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -666,7 +666,7 @@ struct nfs4_rename_res {
 	struct nfs_fattr *		new_fattr;
 };
 
-#define NFS4_SETCLIENTID_NAMELEN	(48)
+#define NFS4_SETCLIENTID_NAMELEN	(56)
 struct nfs4_setclientid {
 	const nfs4_verifier *		sc_verifier;
 	unsigned int			sc_name_len;
-- 
cgit v1.2.3


From 2881ae74e68ecfe3b32a90936e5d93a9ba598c3a Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 20 Dec 2007 16:03:54 -0500
Subject: SUNRPC: Clean up the transport timeout initialisation

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h     |  5 -----
 net/sunrpc/xprt.c               | 17 -----------------
 net/sunrpc/xprtrdma/transport.c |  7 ++++++-
 net/sunrpc/xprtsock.c           | 29 +++++++++++++++++++++--------
 4 files changed, 27 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 2554cd2b016f..a00d4a4ba6ee 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -202,11 +202,6 @@ struct xprt_class {
 	char			name[32];
 };
 
-/*
- * Transport operations used by ULPs
- */
-void			xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long incr);
-
 /*
  * Generic internal transport functions
  */
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 7520c6623c46..2bc99569c586 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -977,23 +977,6 @@ void xprt_release(struct rpc_task *task)
 	spin_unlock(&xprt->reserve_lock);
 }
 
-/**
- * xprt_set_timeout - set constant RPC timeout
- * @to: RPC timeout parameters to set up
- * @retr: number of retries
- * @incr: amount of increase after each retry
- *
- */
-void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long incr)
-{
-	to->to_initval   =
-	to->to_increment = incr;
-	to->to_maxval    = to->to_initval + (incr * retr);
-	to->to_retries   = retr;
-	to->to_exponential = 0;
-}
-EXPORT_SYMBOL_GPL(xprt_set_timeout);
-
 /**
  * xprt_create_transport - create an RPC transport
  * @args: rpc transport creation arguments
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 73033d824eed..39f10016c86b 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -289,6 +289,11 @@ xprt_rdma_destroy(struct rpc_xprt *xprt)
 	module_put(THIS_MODULE);
 }
 
+static const struct rpc_timeout xprt_rdma_default_timeout = {
+	.to_initval = 60 * HZ,
+	.to_maxval = 60 * HZ,
+};
+
 /**
  * xprt_setup_rdma - Set up transport to use RDMA
  *
@@ -327,7 +332,7 @@ xprt_setup_rdma(struct xprt_create *args)
 	}
 
 	/* 60 second timeout, no retries */
-	xprt_set_timeout(&xprt->timeout, 0, 60UL * HZ);
+	memcpy(&xprt->timeout, &xprt_rdma_default_timeout, sizeof(xprt->timeout));
 	xprt->bind_timeout = (60U * HZ);
 	xprt->connect_timeout = (60U * HZ);
 	xprt->reestablish_timeout = (5U * HZ);
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index a4cfdc5b2648..d7b07ac5b710 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1895,6 +1895,13 @@ static struct rpc_xprt *xs_setup_xprt(struct xprt_create *args,
 	return xprt;
 }
 
+static const struct rpc_timeout xs_udp_default_timeout = {
+	.to_initval = 5 * HZ,
+	.to_maxval = 30 * HZ,
+	.to_increment = 5 * HZ,
+	.to_retries = 5,
+};
+
 /**
  * xs_setup_udp - Set up transport to use a UDP socket
  * @args: rpc transport creation arguments
@@ -1905,6 +1912,7 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 	struct sockaddr *addr = args->dstaddr;
 	struct rpc_xprt *xprt;
 	struct sock_xprt *transport;
+	const struct rpc_timeout *timeo = &xs_udp_default_timeout;
 
 	xprt = xs_setup_xprt(args, xprt_udp_slot_table_entries);
 	if (IS_ERR(xprt))
@@ -1923,10 +1931,9 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 
 	xprt->ops = &xs_udp_ops;
 
-	if (args->timeout)
-		xprt->timeout = *args->timeout;
-	else
-		xprt_set_timeout(&xprt->timeout, 5, 5 * HZ);
+	if (args->timeout != NULL)
+		timeo = args->timeout;
+	memcpy(&xprt->timeout, timeo, sizeof(xprt->timeout));
 
 	switch (addr->sa_family) {
 	case AF_INET:
@@ -1961,6 +1968,12 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 	return ERR_PTR(-EINVAL);
 }
 
+static const struct rpc_timeout xs_tcp_default_timeout = {
+	.to_initval = 60 * HZ,
+	.to_maxval = 60 * HZ,
+	.to_retries = 2,
+};
+
 /**
  * xs_setup_tcp - Set up transport to use a TCP socket
  * @args: rpc transport creation arguments
@@ -1971,6 +1984,7 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 	struct sockaddr *addr = args->dstaddr;
 	struct rpc_xprt *xprt;
 	struct sock_xprt *transport;
+	const struct rpc_timeout *timeo = &xs_tcp_default_timeout;
 
 	xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries);
 	if (IS_ERR(xprt))
@@ -1988,10 +2002,9 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 
 	xprt->ops = &xs_tcp_ops;
 
-	if (args->timeout)
-		xprt->timeout = *args->timeout;
-	else
-		xprt_set_timeout(&xprt->timeout, 2, 60 * HZ);
+	if (args->timeout != NULL)
+		timeo = args->timeout;
+	memcpy(&xprt->timeout, timeo, sizeof(xprt->timeout));
 
 	switch (addr->sa_family) {
 	case AF_INET:
-- 
cgit v1.2.3


From ba7392bb37cb12781890f45d7ddee1618e33a036 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 20 Dec 2007 16:03:55 -0500
Subject: SUNRPC: Add support for per-client timeout values

In order to be able to support setting the timeo and retrans parameters on
a per-mountpoint basis, we move the rpc_timeout structure into the
rpc_clnt.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h     |  4 +++-
 include/linux/sunrpc/xprt.h     |  3 +--
 net/sunrpc/clnt.c               | 12 +++++++++---
 net/sunrpc/xprt.c               | 11 ++++++-----
 net/sunrpc/xprtrdma/transport.c |  2 +-
 net/sunrpc/xprtsock.c           | 11 ++---------
 6 files changed, 22 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 4fc268ab6f3e..51a338189a23 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -46,6 +46,7 @@ struct rpc_clnt {
 				cl_autobind : 1;/* use getport() */
 
 	struct rpc_rtt *	cl_rtt;		/* RTO estimator data */
+	const struct rpc_timeout *cl_timeout;	/* Timeout strategy */
 
 	int			cl_nodelen;	/* nodename length */
 	char 			cl_nodename[UNX_MAXNODENAME];
@@ -54,6 +55,7 @@ struct rpc_clnt {
 	struct dentry *		cl_dentry;	/* inode */
 	struct rpc_clnt *	cl_parent;	/* Points to parent of clones */
 	struct rpc_rtt		cl_rtt_default;
+	struct rpc_timeout	cl_timeout_default;
 	struct rpc_program *	cl_program;
 	char			cl_inline_name[32];
 };
@@ -99,7 +101,7 @@ struct rpc_create_args {
 	struct sockaddr		*address;
 	size_t			addrsize;
 	struct sockaddr		*saddress;
-	struct rpc_timeout	*timeout;
+	const struct rpc_timeout *timeout;
 	char			*servername;
 	struct rpc_program	*program;
 	u32			version;
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index a00d4a4ba6ee..c3364d88d83e 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -120,7 +120,7 @@ struct rpc_xprt {
 	struct kref		kref;		/* Reference count */
 	struct rpc_xprt_ops *	ops;		/* transport methods */
 
-	struct rpc_timeout	timeout;	/* timeout parms */
+	const struct rpc_timeout *timeout;	/* timeout parms */
 	struct sockaddr_storage	addr;		/* server address */
 	size_t			addrlen;	/* size of server address */
 	int			prot;		/* IP protocol */
@@ -191,7 +191,6 @@ struct xprt_create {
 	struct sockaddr *	srcaddr;	/* optional local address */
 	struct sockaddr *	dstaddr;	/* remote peer address */
 	size_t			addrlen;
-	struct rpc_timeout *	timeout;	/* optional timeout parameters */
 };
 
 struct xprt_class {
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 6a6b96e8dcb2..a3c00da9ce21 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -188,8 +188,15 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru
 	if (!xprt_bound(clnt->cl_xprt))
 		clnt->cl_autobind = 1;
 
+	clnt->cl_timeout = xprt->timeout;
+	if (args->timeout != NULL) {
+		memcpy(&clnt->cl_timeout_default, args->timeout,
+				sizeof(clnt->cl_timeout_default));
+		clnt->cl_timeout = &clnt->cl_timeout_default;
+	}
+
 	clnt->cl_rtt = &clnt->cl_rtt_default;
-	rpc_init_rtt(&clnt->cl_rtt_default, xprt->timeout.to_initval);
+	rpc_init_rtt(&clnt->cl_rtt_default, clnt->cl_timeout->to_initval);
 
 	kref_init(&clnt->cl_kref);
 
@@ -251,7 +258,6 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 		.srcaddr = args->saddress,
 		.dstaddr = args->address,
 		.addrlen = args->addrsize,
-		.timeout = args->timeout
 	};
 	char servername[48];
 
@@ -348,7 +354,7 @@ rpc_clone_client(struct rpc_clnt *clnt)
 	new->cl_autobind = 0;
 	INIT_LIST_HEAD(&new->cl_tasks);
 	spin_lock_init(&new->cl_lock);
-	rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval);
+	rpc_init_rtt(&new->cl_rtt_default, clnt->cl_timeout->to_initval);
 	new->cl_metrics = rpc_alloc_iostats(clnt);
 	if (new->cl_metrics == NULL)
 		goto out_no_stats;
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 2bc99569c586..cfcade906a56 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -501,9 +501,10 @@ EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_def);
 void xprt_set_retrans_timeout_rtt(struct rpc_task *task)
 {
 	int timer = task->tk_msg.rpc_proc->p_timer;
-	struct rpc_rtt *rtt = task->tk_client->cl_rtt;
+	struct rpc_clnt *clnt = task->tk_client;
+	struct rpc_rtt *rtt = clnt->cl_rtt;
 	struct rpc_rqst *req = task->tk_rqstp;
-	unsigned long max_timeout = req->rq_xprt->timeout.to_maxval;
+	unsigned long max_timeout = clnt->cl_timeout->to_maxval;
 
 	task->tk_timeout = rpc_calc_rto(rtt, timer);
 	task->tk_timeout <<= rpc_ntimeo(rtt, timer) + req->rq_retries;
@@ -514,7 +515,7 @@ EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_rtt);
 
 static void xprt_reset_majortimeo(struct rpc_rqst *req)
 {
-	struct rpc_timeout *to = &req->rq_xprt->timeout;
+	const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 
 	req->rq_majortimeo = req->rq_timeout;
 	if (to->to_exponential)
@@ -534,7 +535,7 @@ static void xprt_reset_majortimeo(struct rpc_rqst *req)
 int xprt_adjust_timeout(struct rpc_rqst *req)
 {
 	struct rpc_xprt *xprt = req->rq_xprt;
-	struct rpc_timeout *to = &xprt->timeout;
+	const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 	int status = 0;
 
 	if (time_before(jiffies, req->rq_majortimeo)) {
@@ -928,7 +929,7 @@ static void xprt_request_init(struct rpc_task *task, struct rpc_xprt *xprt)
 {
 	struct rpc_rqst	*req = task->tk_rqstp;
 
-	req->rq_timeout = xprt->timeout.to_initval;
+	req->rq_timeout = task->tk_client->cl_timeout->to_initval;
 	req->rq_task	= task;
 	req->rq_xprt    = xprt;
 	req->rq_buffer  = NULL;
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 39f10016c86b..d1389afc8342 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -332,7 +332,7 @@ xprt_setup_rdma(struct xprt_create *args)
 	}
 
 	/* 60 second timeout, no retries */
-	memcpy(&xprt->timeout, &xprt_rdma_default_timeout, sizeof(xprt->timeout));
+	xprt->timeout = &xprt_rdma_default_timeout;
 	xprt->bind_timeout = (60U * HZ);
 	xprt->connect_timeout = (60U * HZ);
 	xprt->reestablish_timeout = (5U * HZ);
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index d7b07ac5b710..6ba329d339a2 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1912,7 +1912,6 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 	struct sockaddr *addr = args->dstaddr;
 	struct rpc_xprt *xprt;
 	struct sock_xprt *transport;
-	const struct rpc_timeout *timeo = &xs_udp_default_timeout;
 
 	xprt = xs_setup_xprt(args, xprt_udp_slot_table_entries);
 	if (IS_ERR(xprt))
@@ -1931,9 +1930,7 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 
 	xprt->ops = &xs_udp_ops;
 
-	if (args->timeout != NULL)
-		timeo = args->timeout;
-	memcpy(&xprt->timeout, timeo, sizeof(xprt->timeout));
+	xprt->timeout = &xs_udp_default_timeout;
 
 	switch (addr->sa_family) {
 	case AF_INET:
@@ -1984,7 +1981,6 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 	struct sockaddr *addr = args->dstaddr;
 	struct rpc_xprt *xprt;
 	struct sock_xprt *transport;
-	const struct rpc_timeout *timeo = &xs_tcp_default_timeout;
 
 	xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries);
 	if (IS_ERR(xprt))
@@ -2001,10 +1997,7 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 	xprt->idle_timeout = XS_IDLE_DISC_TO;
 
 	xprt->ops = &xs_tcp_ops;
-
-	if (args->timeout != NULL)
-		timeo = args->timeout;
-	memcpy(&xprt->timeout, timeo, sizeof(xprt->timeout));
+	xprt->timeout = &xs_tcp_default_timeout;
 
 	switch (addr->sa_family) {
 	case AF_INET:
-- 
cgit v1.2.3


From 331702337f2b2e7cef40366ee207a25604df4671 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 20 Dec 2007 16:03:59 -0500
Subject: NFS: Support per-mountpoint timeout parameters.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 82 +++++++++++++++++++++++++++--------------------
 fs/nfs/super.c            |  4 +--
 include/linux/nfs_fs_sb.h |  2 --
 3 files changed, 49 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 59a6dccab548..03d9bed7849a 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -415,18 +415,16 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
  * Create an RPC client handle
  */
 static int nfs_create_rpc_client(struct nfs_client *clp, int proto,
-						unsigned int timeo,
-						unsigned int retrans,
-						rpc_authflavor_t flavor,
-						int flags)
+				 const struct rpc_timeout *timeparms,
+				 rpc_authflavor_t flavor,
+				 int flags)
 {
-	struct rpc_timeout	timeparms;
 	struct rpc_clnt		*clnt = NULL;
 	struct rpc_create_args args = {
 		.protocol	= proto,
 		.address	= (struct sockaddr *)&clp->cl_addr,
 		.addrsize	= clp->cl_addrlen,
-		.timeout	= &timeparms,
+		.timeout	= timeparms,
 		.servername	= clp->cl_hostname,
 		.program	= &nfs_program,
 		.version	= clp->rpc_ops->version,
@@ -437,10 +435,6 @@ static int nfs_create_rpc_client(struct nfs_client *clp, int proto,
 	if (!IS_ERR(clp->cl_rpcclient))
 		return 0;
 
-	nfs_init_timeout_values(&timeparms, proto, timeo, retrans);
-	clp->retrans_timeo = timeparms.to_initval;
-	clp->retrans_count = timeparms.to_retries;
-
 	clnt = rpc_create(&args);
 	if (IS_ERR(clnt)) {
 		dprintk("%s: cannot create RPC client. Error = %ld\n",
@@ -515,7 +509,9 @@ static inline void nfs_init_server_aclclient(struct nfs_server *server)
 /*
  * Create a general RPC client
  */
-static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t pseudoflavour)
+static int nfs_init_server_rpcclient(struct nfs_server *server,
+		const struct rpc_timeout *timeo,
+		rpc_authflavor_t pseudoflavour)
 {
 	struct nfs_client *clp = server->nfs_client;
 
@@ -525,6 +521,11 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t
 		return PTR_ERR(server->client);
 	}
 
+	memcpy(&server->client->cl_timeout_default,
+			timeo,
+			sizeof(server->client->cl_timeout_default));
+	server->client->cl_timeout = &server->client->cl_timeout_default;
+
 	if (pseudoflavour != clp->cl_rpcclient->cl_auth->au_flavor) {
 		struct rpc_auth *auth;
 
@@ -549,6 +550,7 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t
  * Initialise an NFS2 or NFS3 client
  */
 static int nfs_init_client(struct nfs_client *clp,
+			   const struct rpc_timeout *timeparms,
 			   const struct nfs_parsed_mount_data *data)
 {
 	int error;
@@ -564,7 +566,7 @@ static int nfs_init_client(struct nfs_client *clp,
 	 * - RFC 2623, sec 2.3.2
 	 */
 	error = nfs_create_rpc_client(clp, data->nfs_server.protocol,
-				data->timeo, data->retrans, RPC_AUTH_UNIX, 0);
+				timeparms, RPC_AUTH_UNIX, 0);
 	if (error < 0)
 		goto error;
 	nfs_mark_client_ready(clp, NFS_CS_READY);
@@ -588,6 +590,7 @@ static int nfs_init_server(struct nfs_server *server,
 		.addrlen = data->nfs_server.addrlen,
 		.rpc_ops = &nfs_v2_clientops,
 	};
+	struct rpc_timeout timeparms;
 	struct nfs_client *clp;
 	int error;
 
@@ -605,7 +608,9 @@ static int nfs_init_server(struct nfs_server *server,
 		return PTR_ERR(clp);
 	}
 
-	error = nfs_init_client(clp, data);
+	nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
+			data->timeo, data->retrans);
+	error = nfs_init_client(clp, &timeparms, data);
 	if (error < 0)
 		goto error;
 
@@ -629,7 +634,7 @@ static int nfs_init_server(struct nfs_server *server,
 	if (error < 0)
 		goto error;
 
-	error = nfs_init_server_rpcclient(server, data->auth_flavors[0]);
+	error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]);
 	if (error < 0)
 		goto error;
 
@@ -889,7 +894,8 @@ error:
  * Initialise an NFS4 client record
  */
 static int nfs4_init_client(struct nfs_client *clp,
-		int proto, int timeo, int retrans,
+		int proto,
+		const struct rpc_timeout *timeparms,
 		const char *ip_addr,
 		rpc_authflavor_t authflavour)
 {
@@ -904,7 +910,7 @@ static int nfs4_init_client(struct nfs_client *clp,
 	/* Check NFS protocol revision and initialize RPC op vector */
 	clp->rpc_ops = &nfs_v4_clientops;
 
-	error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour,
+	error = nfs_create_rpc_client(clp, proto, timeparms, authflavour,
 					RPC_CLNT_CREATE_DISCRTRY);
 	if (error < 0)
 		goto error;
@@ -936,7 +942,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		const size_t addrlen,
 		const char *ip_addr,
 		rpc_authflavor_t authflavour,
-		int proto, int timeo, int retrans)
+		int proto, const struct rpc_timeout *timeparms)
 {
 	struct nfs_client_initdata cl_init = {
 		.hostname = hostname,
@@ -955,7 +961,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		error = PTR_ERR(clp);
 		goto error;
 	}
-	error = nfs4_init_client(clp, proto, timeo, retrans, ip_addr, authflavour);
+	error = nfs4_init_client(clp, proto, timeparms, ip_addr, authflavour);
 	if (error < 0)
 		goto error_put;
 
@@ -976,10 +982,26 @@ error:
 static int nfs4_init_server(struct nfs_server *server,
 		const struct nfs_parsed_mount_data *data)
 {
+	struct rpc_timeout timeparms;
 	int error;
 
 	dprintk("--> nfs4_init_server()\n");
 
+	nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
+			data->timeo, data->retrans);
+
+	/* Get a client record */
+	error = nfs4_set_client(server,
+			data->nfs_server.hostname,
+			(const struct sockaddr *)&data->nfs_server.address,
+			data->nfs_server.addrlen,
+			data->client_address,
+			data->auth_flavors[0],
+			data->nfs_server.protocol,
+			&timeparms);
+	if (error < 0)
+		goto error;
+
 	/* Initialise the client representation from the mount data */
 	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
 	server->caps |= NFS_CAP_ATOMIC_OPEN;
@@ -994,8 +1016,9 @@ static int nfs4_init_server(struct nfs_server *server,
 	server->acdirmin = data->acdirmin * HZ;
 	server->acdirmax = data->acdirmax * HZ;
 
-	error = nfs_init_server_rpcclient(server, data->auth_flavors[0]);
+	error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]);
 
+error:
 	/* Done */
 	dprintk("<-- nfs4_init_server() = %d\n", error);
 	return error;
@@ -1018,18 +1041,6 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
-	/* Get a client record */
-	error = nfs4_set_client(server,
-			data->nfs_server.hostname,
-			(struct sockaddr *)&data->nfs_server.address,
-			data->nfs_server.addrlen,
-			data->client_address,
-			data->auth_flavors[0],
-			data->nfs_server.protocol,
-			data->timeo, data->retrans);
-	if (error < 0)
-		goto error;
-
 	/* set up the general RPC client */
 	error = nfs4_init_server(server, data);
 	if (error < 0)
@@ -1103,8 +1114,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 				parent_client->cl_ipaddr,
 				data->authflavor,
 				parent_server->client->cl_xprt->prot,
-				parent_client->retrans_timeo,
-				parent_client->retrans_count);
+				parent_server->client->cl_timeout);
 	if (error < 0)
 		goto error;
 
@@ -1112,7 +1122,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 	nfs_server_copy_userdata(server, parent_server);
 	server->caps |= NFS_CAP_ATOMIC_OPEN;
 
-	error = nfs_init_server_rpcclient(server, data->authflavor);
+	error = nfs_init_server_rpcclient(server, parent_server->client->cl_timeout, data->authflavor);
 	if (error < 0)
 		goto error;
 
@@ -1181,7 +1191,9 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source,
 
 	server->fsid = fattr->fsid;
 
-	error = nfs_init_server_rpcclient(server, source->client->cl_auth->au_flavor);
+	error = nfs_init_server_rpcclient(server,
+			source->client->cl_timeout,
+			source->client->cl_auth->au_flavor);
 	if (error < 0)
 		goto out_free_server;
 	if (!IS_ERR(source->client_acl))
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 3cbe32f3e88b..0d1bc61d0b68 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -479,8 +479,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 	}
 	seq_printf(m, ",proto=%s",
 		   rpc_peeraddr2str(nfss->client, RPC_DISPLAY_PROTO));
-	seq_printf(m, ",timeo=%lu", 10U * clp->retrans_timeo / HZ);
-	seq_printf(m, ",retrans=%u", clp->retrans_count);
+	seq_printf(m, ",timeo=%lu", 10U * nfss->client->cl_timeout->to_initval / HZ);
+	seq_printf(m, ",retrans=%u", nfss->client->cl_timeout->to_retries);
 	seq_printf(m, ",sec=%s", nfs_pseudoflavour_to_name(nfss->client->cl_auth->au_flavor));
 }
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 912cfe84ea86..d15c9487b8f1 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -29,8 +29,6 @@ struct nfs_client {
 
 	struct rpc_clnt *	cl_rpcclient;
 	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
-	unsigned long		retrans_timeo;	/* retransmit timeout */
-	unsigned int		retrans_count;	/* number of retransmit tries */
 
 #ifdef CONFIG_NFS_V4
 	u64			cl_clientid;	/* constant */
-- 
cgit v1.2.3


From 59dca3b28cb915745019d4f4c27d97b6b6ab12c6 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 3 Jan 2008 16:29:06 -0500
Subject: NFS: Fix the 'proto=' mount option

Currently, if you have a server mounted using networking protocol, you
cannot specify a different value using the 'proto=' option on another
mountpoint.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 20 +++++++++++++-------
 include/linux/nfs_fs_sb.h |  1 +
 2 files changed, 14 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 03d9bed7849a..18fcb05a0707 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -100,6 +100,7 @@ struct nfs_client_initdata {
 	const struct sockaddr *addr;
 	size_t addrlen;
 	const struct nfs_rpc_ops *rpc_ops;
+	int proto;
 };
 
 /*
@@ -138,6 +139,8 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	INIT_LIST_HEAD(&clp->cl_superblocks);
 	clp->cl_rpcclient = ERR_PTR(-EINVAL);
 
+	clp->cl_proto = cl_init->proto;
+
 #ifdef CONFIG_NFS_V4
 	init_rwsem(&clp->cl_sem);
 	INIT_LIST_HEAD(&clp->cl_delegations);
@@ -289,6 +292,9 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 		if (clp->rpc_ops != data->rpc_ops)
 			continue;
 
+		if (clp->cl_proto != data->proto)
+			continue;
+
 		/* Match the full socket address */
 		if (memcmp(&clp->cl_addr, data->addr, sizeof(clp->cl_addr)) != 0)
 			continue;
@@ -414,14 +420,14 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 /*
  * Create an RPC client handle
  */
-static int nfs_create_rpc_client(struct nfs_client *clp, int proto,
+static int nfs_create_rpc_client(struct nfs_client *clp,
 				 const struct rpc_timeout *timeparms,
 				 rpc_authflavor_t flavor,
 				 int flags)
 {
 	struct rpc_clnt		*clnt = NULL;
 	struct rpc_create_args args = {
-		.protocol	= proto,
+		.protocol	= clp->cl_proto,
 		.address	= (struct sockaddr *)&clp->cl_addr,
 		.addrsize	= clp->cl_addrlen,
 		.timeout	= timeparms,
@@ -565,8 +571,7 @@ static int nfs_init_client(struct nfs_client *clp,
 	 * Create a client RPC handle for doing FSSTAT with UNIX auth only
 	 * - RFC 2623, sec 2.3.2
 	 */
-	error = nfs_create_rpc_client(clp, data->nfs_server.protocol,
-				timeparms, RPC_AUTH_UNIX, 0);
+	error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_UNIX, 0);
 	if (error < 0)
 		goto error;
 	nfs_mark_client_ready(clp, NFS_CS_READY);
@@ -589,6 +594,7 @@ static int nfs_init_server(struct nfs_server *server,
 		.addr = (const struct sockaddr *)&data->nfs_server.address,
 		.addrlen = data->nfs_server.addrlen,
 		.rpc_ops = &nfs_v2_clientops,
+		.proto = data->nfs_server.protocol,
 	};
 	struct rpc_timeout timeparms;
 	struct nfs_client *clp;
@@ -894,7 +900,6 @@ error:
  * Initialise an NFS4 client record
  */
 static int nfs4_init_client(struct nfs_client *clp,
-		int proto,
 		const struct rpc_timeout *timeparms,
 		const char *ip_addr,
 		rpc_authflavor_t authflavour)
@@ -910,7 +915,7 @@ static int nfs4_init_client(struct nfs_client *clp,
 	/* Check NFS protocol revision and initialize RPC op vector */
 	clp->rpc_ops = &nfs_v4_clientops;
 
-	error = nfs_create_rpc_client(clp, proto, timeparms, authflavour,
+	error = nfs_create_rpc_client(clp, timeparms, authflavour,
 					RPC_CLNT_CREATE_DISCRTRY);
 	if (error < 0)
 		goto error;
@@ -949,6 +954,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		.addr = addr,
 		.addrlen = addrlen,
 		.rpc_ops = &nfs_v4_clientops,
+		.proto = proto,
 	};
 	struct nfs_client *clp;
 	int error;
@@ -961,7 +967,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		error = PTR_ERR(clp);
 		goto error;
 	}
-	error = nfs4_init_client(clp, proto, timeparms, ip_addr, authflavour);
+	error = nfs4_init_client(clp, timeparms, ip_addr, authflavour);
 	if (error < 0)
 		goto error_put;
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index d15c9487b8f1..b5ba5f79485d 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -29,6 +29,7 @@ struct nfs_client {
 
 	struct rpc_clnt *	cl_rpcclient;
 	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
+	int			cl_proto;	/* Network transport protocol */
 
 #ifdef CONFIG_NFS_V4
 	u64			cl_clientid;	/* constant */
-- 
cgit v1.2.3


From b454ae906085cf7774fb4756746680c9b03b6f84 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 7 Jan 2008 18:34:48 -0500
Subject: SUNRPC: fewer conditionals in the format_ip_address routines

Clean up: have the set up routines explicitly pass the strings to be used
for the transport name and NETID.  This removes a number of conditionals
and dependencies on rpc_xprt.prot, which is overloaded.

Tighten up type checking on the address_strings array while we're at it.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  2 +-
 include/linux/sunrpc/xprt.h |  2 +-
 net/sunrpc/clnt.c           |  3 ++-
 net/sunrpc/rpcb_clnt.c      |  2 +-
 net/sunrpc/xprtsock.c       | 56 ++++++++++++++++++---------------------------
 5 files changed, 27 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 51a338189a23..5a13da2573b0 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -143,7 +143,7 @@ void		rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int);
 size_t		rpc_max_payload(struct rpc_clnt *);
 void		rpc_force_rebind(struct rpc_clnt *);
 size_t		rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
-char *		rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
+const char	*rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_SUNRPC_CLNT_H */
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index c3364d88d83e..b3ff9a815e6f 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -183,7 +183,7 @@ struct rpc_xprt {
 					bklog_u;	/* backlog queue utilization */
 	} stat;
 
-	char *			address_strings[RPC_DISPLAY_MAX];
+	const char		*address_strings[RPC_DISPLAY_MAX];
 };
 
 struct xprt_create {
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index a3c00da9ce21..e775ca793249 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -679,7 +679,8 @@ EXPORT_SYMBOL_GPL(rpc_peeraddr);
  * @format: address format
  *
  */
-char *rpc_peeraddr2str(struct rpc_clnt *clnt, enum rpc_display_format_t format)
+const char *rpc_peeraddr2str(struct rpc_clnt *clnt,
+			     enum rpc_display_format_t format)
 {
 	struct rpc_xprt *xprt = clnt->cl_xprt;
 
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index f494e58910ec..b60fa92321a5 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -358,7 +358,7 @@ void rpcb_getport_async(struct rpc_task *task)
 	map->r_prot = xprt->prot;
 	map->r_port = 0;
 	map->r_xprt = xprt_get(xprt);
-	map->r_netid = rpc_peeraddr2str(clnt, RPC_DISPLAY_NETID);
+	map->r_netid = (char *)rpc_peeraddr2str(clnt, RPC_DISPLAY_NETID);
 	memcpy(map->r_addr,
 	       rpc_peeraddr2str(rpcb_clnt, RPC_DISPLAY_UNIVERSAL_ADDR),
 	       sizeof(map->r_addr));
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 6ba329d339a2..b9b94f49c620 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -280,7 +280,9 @@ static inline struct sockaddr_in6 *xs_addr_in6(struct rpc_xprt *xprt)
 	return (struct sockaddr_in6 *) &xprt->addr;
 }
 
-static void xs_format_ipv4_peer_addresses(struct rpc_xprt *xprt)
+static void xs_format_ipv4_peer_addresses(struct rpc_xprt *xprt,
+					  const char *protocol,
+					  const char *netid)
 {
 	struct sockaddr_in *addr = xs_addr_in(xprt);
 	char *buf;
@@ -299,21 +301,14 @@ static void xs_format_ipv4_peer_addresses(struct rpc_xprt *xprt)
 	}
 	xprt->address_strings[RPC_DISPLAY_PORT] = buf;
 
-	buf = kzalloc(8, GFP_KERNEL);
-	if (buf) {
-		if (xprt->prot == IPPROTO_UDP)
-			snprintf(buf, 8, "udp");
-		else
-			snprintf(buf, 8, "tcp");
-	}
-	xprt->address_strings[RPC_DISPLAY_PROTO] = buf;
+	xprt->address_strings[RPC_DISPLAY_PROTO] = protocol;
 
 	buf = kzalloc(48, GFP_KERNEL);
 	if (buf) {
 		snprintf(buf, 48, "addr="NIPQUAD_FMT" port=%u proto=%s",
 			NIPQUAD(addr->sin_addr.s_addr),
 			ntohs(addr->sin_port),
-			xprt->prot == IPPROTO_UDP ? "udp" : "tcp");
+			protocol);
 	}
 	xprt->address_strings[RPC_DISPLAY_ALL] = buf;
 
@@ -340,12 +335,12 @@ static void xs_format_ipv4_peer_addresses(struct rpc_xprt *xprt)
 	}
 	xprt->address_strings[RPC_DISPLAY_UNIVERSAL_ADDR] = buf;
 
-	xprt->address_strings[RPC_DISPLAY_NETID] =
-		kstrdup(xprt->prot == IPPROTO_UDP ?
-			RPCBIND_NETID_UDP : RPCBIND_NETID_TCP, GFP_KERNEL);
+	xprt->address_strings[RPC_DISPLAY_NETID] = netid;
 }
 
-static void xs_format_ipv6_peer_addresses(struct rpc_xprt *xprt)
+static void xs_format_ipv6_peer_addresses(struct rpc_xprt *xprt,
+					  const char *protocol,
+					  const char *netid)
 {
 	struct sockaddr_in6 *addr = xs_addr_in6(xprt);
 	char *buf;
@@ -364,21 +359,14 @@ static void xs_format_ipv6_peer_addresses(struct rpc_xprt *xprt)
 	}
 	xprt->address_strings[RPC_DISPLAY_PORT] = buf;
 
-	buf = kzalloc(8, GFP_KERNEL);
-	if (buf) {
-		if (xprt->prot == IPPROTO_UDP)
-			snprintf(buf, 8, "udp");
-		else
-			snprintf(buf, 8, "tcp");
-	}
-	xprt->address_strings[RPC_DISPLAY_PROTO] = buf;
+	xprt->address_strings[RPC_DISPLAY_PROTO] = protocol;
 
 	buf = kzalloc(64, GFP_KERNEL);
 	if (buf) {
 		snprintf(buf, 64, "addr="NIP6_FMT" port=%u proto=%s",
 				NIP6(addr->sin6_addr),
 				ntohs(addr->sin6_port),
-				xprt->prot == IPPROTO_UDP ? "udp" : "tcp");
+				protocol);
 	}
 	xprt->address_strings[RPC_DISPLAY_ALL] = buf;
 
@@ -405,17 +393,17 @@ static void xs_format_ipv6_peer_addresses(struct rpc_xprt *xprt)
 	}
 	xprt->address_strings[RPC_DISPLAY_UNIVERSAL_ADDR] = buf;
 
-	xprt->address_strings[RPC_DISPLAY_NETID] =
-		kstrdup(xprt->prot == IPPROTO_UDP ?
-			RPCBIND_NETID_UDP6 : RPCBIND_NETID_TCP6, GFP_KERNEL);
+	xprt->address_strings[RPC_DISPLAY_NETID] = netid;
 }
 
 static void xs_free_peer_addresses(struct rpc_xprt *xprt)
 {
-	int i;
-
-	for (i = 0; i < RPC_DISPLAY_MAX; i++)
-		kfree(xprt->address_strings[i]);
+	kfree(xprt->address_strings[RPC_DISPLAY_ADDR]);
+	kfree(xprt->address_strings[RPC_DISPLAY_PORT]);
+	kfree(xprt->address_strings[RPC_DISPLAY_ALL]);
+	kfree(xprt->address_strings[RPC_DISPLAY_HEX_ADDR]);
+	kfree(xprt->address_strings[RPC_DISPLAY_HEX_PORT]);
+	kfree(xprt->address_strings[RPC_DISPLAY_UNIVERSAL_ADDR]);
 }
 
 #define XS_SENDMSG_FLAGS	(MSG_DONTWAIT | MSG_NOSIGNAL)
@@ -1939,7 +1927,7 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 
 		INIT_DELAYED_WORK(&transport->connect_worker,
 					xs_udp_connect_worker4);
-		xs_format_ipv4_peer_addresses(xprt);
+		xs_format_ipv4_peer_addresses(xprt, "udp", RPCBIND_NETID_UDP);
 		break;
 	case AF_INET6:
 		if (((struct sockaddr_in6 *)addr)->sin6_port != htons(0))
@@ -1947,7 +1935,7 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 
 		INIT_DELAYED_WORK(&transport->connect_worker,
 					xs_udp_connect_worker6);
-		xs_format_ipv6_peer_addresses(xprt);
+		xs_format_ipv6_peer_addresses(xprt, "udp", RPCBIND_NETID_UDP6);
 		break;
 	default:
 		kfree(xprt);
@@ -2005,14 +1993,14 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 			xprt_set_bound(xprt);
 
 		INIT_DELAYED_WORK(&transport->connect_worker, xs_tcp_connect_worker4);
-		xs_format_ipv4_peer_addresses(xprt);
+		xs_format_ipv4_peer_addresses(xprt, "tcp", RPCBIND_NETID_TCP);
 		break;
 	case AF_INET6:
 		if (((struct sockaddr_in6 *)addr)->sin6_port != htons(0))
 			xprt_set_bound(xprt);
 
 		INIT_DELAYED_WORK(&transport->connect_worker, xs_tcp_connect_worker6);
-		xs_format_ipv6_peer_addresses(xprt);
+		xs_format_ipv6_peer_addresses(xprt, "tcp", RPCBIND_NETID_TCP6);
 		break;
 	default:
 		kfree(xprt);
-- 
cgit v1.2.3


From 52c4044d00fe703eb3fb18e0d8dfd1c196eb28be Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 11 Jan 2008 17:09:44 -0500
Subject: NLM: Introduce external nlm_host set-up and tear-down functions

We would like to remove the per-lock-operation nlm_lookup_host() call from
nlmclnt_proc().

The new architecture pins an nlm_host structure to each NFS client
superblock that has the "lock" mount option set.  The NFS client passes
in the pinned nlm_host structure during each call to nlmclnt_proc().  NFS
client unmount processing "puts" the nlm_host so it can be garbage-
collected later.

This patch introduces externally callable NLM functions that handle
mount-time nlm_host set up and tear-down.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/clntlock.c        | 48 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/lockd/bind.h |  7 +++++++
 2 files changed, 55 insertions(+)

(limited to 'include')

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index d070b18e539d..9a8f4f45c19e 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -41,6 +41,54 @@ struct nlm_wait {
 
 static LIST_HEAD(nlm_blocked);
 
+/**
+ * nlmclnt_init - Set up per-NFS mount point lockd data structures
+ * @server_name: server's hostname
+ * @server_address: server's network address
+ * @server_addrlen: length of server's address
+ * @protocol: transport protocol lockd should use
+ * @nfs_version: NFS protocol version for this mount point
+ *
+ * Returns pointer to an appropriate nlm_host struct,
+ * or an ERR_PTR value.
+ */
+struct nlm_host *nlmclnt_init(const char *server_name,
+			      const struct sockaddr *server_address,
+			      size_t server_addrlen,
+			      unsigned short protocol, u32 nfs_version)
+{
+	struct nlm_host *host;
+	u32 nlm_version = (nfs_version == 2) ? 1 : 4;
+	int status;
+
+	status = lockd_up(protocol);
+	if (status < 0)
+		return ERR_PTR(status);
+
+	host = nlmclnt_lookup_host((struct sockaddr_in *)server_address,
+				   protocol, nlm_version,
+				   server_name, strlen(server_name));
+	if (host == NULL) {
+		lockd_down();
+		return ERR_PTR(-ENOLCK);
+	}
+
+	return host;
+}
+EXPORT_SYMBOL_GPL(nlmclnt_init);
+
+/**
+ * nlmclnt_done - Release resources allocated by nlmclnt_init()
+ * @host: nlm_host structure reserved by nlmclnt_init()
+ *
+ */
+void nlmclnt_done(struct nlm_host *host)
+{
+	nlm_release_host(host);
+	lockd_down();
+}
+EXPORT_SYMBOL_GPL(nlmclnt_done);
+
 /*
  * Queue up a lock for blocking so that the GRANTED request can see it
  */
diff --git a/include/linux/lockd/bind.h b/include/linux/lockd/bind.h
index 6f1637c61e10..ad5402f5456b 100644
--- a/include/linux/lockd/bind.h
+++ b/include/linux/lockd/bind.h
@@ -35,6 +35,13 @@ extern struct nlmsvc_binding *	nlmsvc_ops;
 /*
  * Functions exported by the lockd module
  */
+extern struct nlm_host *nlmclnt_init(const char *server_name,
+					const struct sockaddr *server_address,
+					size_t server_addrlen,
+					unsigned short protocol,
+					u32 nfs_version);
+extern void	nlmclnt_done(struct nlm_host *host);
+
 extern int	nlmclnt_proc(struct inode *, int, struct file_lock *);
 extern int	lockd_up(int proto);
 extern void	lockd_down(void);
-- 
cgit v1.2.3


From 9289e7f91add1c09c3ec8571a2080f7507730b8d Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 11 Jan 2008 17:09:52 -0500
Subject: NFS: Invoke nlmclnt_init during NFS mount processing

Cache an appropriate nlm_host structure in the NFS client's mount point
metadata for later use.

Note that there is no need to set NFS_MOUNT_NONLM in the error case -- if
nfs_start_lockd() returns a non-zero value, its callers ensure that the
mount request fails outright.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 32 +++++++++++++++++++-------------
 include/linux/nfs_fs_sb.h |  2 ++
 2 files changed, 21 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 18fcb05a0707..0b3ce86f6fc9 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -458,7 +458,7 @@ static int nfs_create_rpc_client(struct nfs_client *clp,
 static void nfs_destroy_server(struct nfs_server *server)
 {
 	if (!(server->flags & NFS_MOUNT_NONLM))
-		lockd_down();	/* release rpc.lockd */
+		nlmclnt_done(server->nlm_host);
 }
 
 /*
@@ -466,20 +466,26 @@ static void nfs_destroy_server(struct nfs_server *server)
  */
 static int nfs_start_lockd(struct nfs_server *server)
 {
-	int error = 0;
+	struct nlm_host *host;
+	struct nfs_client *clp = server->nfs_client;
+	u32 nfs_version = clp->rpc_ops->version;
+	unsigned short protocol = server->flags & NFS_MOUNT_TCP ?
+						IPPROTO_TCP : IPPROTO_UDP;
 
-	if (server->nfs_client->rpc_ops->version > 3)
-		goto out;
+	if (nfs_version > 3)
+		return 0;
 	if (server->flags & NFS_MOUNT_NONLM)
-		goto out;
-	error = lockd_up((server->flags & NFS_MOUNT_TCP) ?
-			IPPROTO_TCP : IPPROTO_UDP);
-	if (error < 0)
-		server->flags |= NFS_MOUNT_NONLM;
-	else
-		server->destroy = nfs_destroy_server;
-out:
-	return error;
+		return 0;
+
+	host = nlmclnt_init(clp->cl_hostname,
+			    (struct sockaddr *)&clp->cl_addr,
+			    clp->cl_addrlen, protocol, nfs_version);
+	if (IS_ERR(host))
+		return PTR_ERR(host);
+
+	server->nlm_host = host;
+	server->destroy = nfs_destroy_server;
+	return 0;
 }
 
 /*
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index b5ba5f79485d..3423c6761bf7 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -8,6 +8,7 @@
 #include <asm/atomic.h>
 
 struct nfs_iostats;
+struct nlm_host;
 
 /*
  * The nfs_client identifies our client state to the server.
@@ -80,6 +81,7 @@ struct nfs_server {
 	struct list_head	master_link;	/* link in master servers list */
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
+	struct nlm_host		*nlm_host;	/* NLM client handle */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
 	atomic_long_t		writeback;	/* number of writeback pages */
-- 
cgit v1.2.3


From 1093a60ef34bb12010fe7ea4b780bee1c57cfbbe Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 11 Jan 2008 17:09:59 -0500
Subject: NLM/NFS: Use cached nlm_host when calling nlmclnt_proc()

Now that each NFS mount point caches its own nlm_host structure, it can be
passed to nlmclnt_proc() for each lock request.  By pinning an nlm_host for
each mount point, we trade the overhead of looking up or creating a fresh
nlm_host struct during every NLM procedure call for a little extra memory.

We also restrict the nlmclnt_proc symbol to limit the use of this call to
in-tree modules.

Note that nlm_lookup_host() (just removed from the client's per-request
NLM processing) could also trigger an nlm_host garbage collection.  Now
client-side nlm_host garbage collection occurs only during NFS mount
processing.  Since the NFS client now holds a reference on these nlm_host
structures, they wouldn't have been affected by garbage collection
anyway.

Given that nlm_lookup_host() reorders the global nlm_host chain after
every successful lookup, and that a garbage collection could be triggered
during the call, we've removed a significant amount of per-NLM-request
CPU processing overhead.

Sidebar: there are only a few remaining references to the internals of
NFS inodes in the client-side NLM code.  The only references I found are
related to extracting or comparing the inode's file handle via NFS_FH().
One is in nlmclnt_grant(); the other is in nlmclnt_setlockargs().

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/clntproc.c        | 33 ++++++++++-----------------------
 fs/nfs/nfs3proc.c          |  4 +++-
 fs/nfs/proc.c              |  4 +++-
 include/linux/lockd/bind.h |  3 ++-
 4 files changed, 18 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index a10343bed160..b1a4dba443bc 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -145,34 +145,21 @@ static void nlmclnt_release_lockargs(struct nlm_rqst *req)
 	BUG_ON(req->a_args.lock.fl.fl_ops != NULL);
 }
 
-/*
- * This is the main entry point for the NLM client.
+/**
+ * nlmclnt_proc - Perform a single client-side lock request
+ * @host: address of a valid nlm_host context representing the NLM server
+ * @cmd: fcntl-style file lock operation to perform
+ * @fl: address of arguments for the lock operation
+ *
  */
-int
-nlmclnt_proc(struct inode *inode, int cmd, struct file_lock *fl)
+int nlmclnt_proc(struct nlm_host *host, int cmd, struct file_lock *fl)
 {
-	struct rpc_clnt		*client = NFS_CLIENT(inode);
-	struct sockaddr_in	addr;
-	struct nfs_server	*nfssrv = NFS_SERVER(inode);
-	struct nlm_host		*host;
 	struct nlm_rqst		*call;
 	sigset_t		oldset;
 	unsigned long		flags;
-	int			status, vers;
-
-	vers = (NFS_PROTO(inode)->version == 3) ? 4 : 1;
-	if (NFS_PROTO(inode)->version > 3) {
-		printk(KERN_NOTICE "NFSv4 file locking not implemented!\n");
-		return -ENOLCK;
-	}
-
-	rpc_peeraddr(client, (struct sockaddr *) &addr, sizeof(addr));
-	host = nlmclnt_lookup_host(&addr, client->cl_xprt->prot, vers,
-				   nfssrv->nfs_client->cl_hostname,
-				   strlen(nfssrv->nfs_client->cl_hostname));
-	if (host == NULL)
-		return -ENOLCK;
+	int			status;
 
+	nlm_get_host(host);
 	call = nlm_alloc_call(host);
 	if (call == NULL)
 		return -ENOMEM;
@@ -219,7 +206,7 @@ nlmclnt_proc(struct inode *inode, int cmd, struct file_lock *fl)
 	dprintk("lockd: clnt proc returns %d\n", status);
 	return status;
 }
-EXPORT_SYMBOL(nlmclnt_proc);
+EXPORT_SYMBOL_GPL(nlmclnt_proc);
 
 /*
  * Allocate an NLM RPC call struct
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index e68580ebaa47..b353c1a05bfd 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -767,7 +767,9 @@ static void nfs3_proc_commit_setup(struct nfs_write_data *data, struct rpc_messa
 static int
 nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 {
-	return nlmclnt_proc(filp->f_path.dentry->d_inode, cmd, fl);
+	struct inode *inode = filp->f_path.dentry->d_inode;
+
+	return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl);
 }
 
 const struct nfs_rpc_ops nfs_v3_clientops = {
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index c9f46a24e75c..5ccf7faee19c 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -593,7 +593,9 @@ nfs_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg)
 static int
 nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 {
-	return nlmclnt_proc(filp->f_path.dentry->d_inode, cmd, fl);
+	struct inode *inode = filp->f_path.dentry->d_inode;
+
+	return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl);
 }
 
 
diff --git a/include/linux/lockd/bind.h b/include/linux/lockd/bind.h
index ad5402f5456b..73368075af03 100644
--- a/include/linux/lockd/bind.h
+++ b/include/linux/lockd/bind.h
@@ -42,7 +42,8 @@ extern struct nlm_host *nlmclnt_init(const char *server_name,
 					u32 nfs_version);
 extern void	nlmclnt_done(struct nlm_host *host);
 
-extern int	nlmclnt_proc(struct inode *, int, struct file_lock *);
+extern int	nlmclnt_proc(struct nlm_host *host, int cmd,
+					struct file_lock *fl);
 extern int	lockd_up(int proto);
 extern void	lockd_down(void);
 
-- 
cgit v1.2.3


From 883bb163f84e0a54b29846c61621f52db3f27393 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 15 Jan 2008 16:04:20 -0500
Subject: NLM: Introduce an arguments structure for nlmclnt_init()

Clean up: pass 5 arguments to nlmclnt_init() in a structure similar to the
new nfs_client_initdata structure.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/lockd/clntlock.c        | 22 ++++++++--------------
 fs/nfs/client.c            | 17 ++++++++++-------
 include/linux/lockd/bind.h | 19 ++++++++++++++-----
 3 files changed, 32 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index 9a8f4f45c19e..0b45fd3a4bfd 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -43,31 +43,25 @@ static LIST_HEAD(nlm_blocked);
 
 /**
  * nlmclnt_init - Set up per-NFS mount point lockd data structures
- * @server_name: server's hostname
- * @server_address: server's network address
- * @server_addrlen: length of server's address
- * @protocol: transport protocol lockd should use
- * @nfs_version: NFS protocol version for this mount point
+ * @nlm_init: pointer to arguments structure
  *
  * Returns pointer to an appropriate nlm_host struct,
  * or an ERR_PTR value.
  */
-struct nlm_host *nlmclnt_init(const char *server_name,
-			      const struct sockaddr *server_address,
-			      size_t server_addrlen,
-			      unsigned short protocol, u32 nfs_version)
+struct nlm_host *nlmclnt_init(const struct nlmclnt_initdata *nlm_init)
 {
 	struct nlm_host *host;
-	u32 nlm_version = (nfs_version == 2) ? 1 : 4;
+	u32 nlm_version = (nlm_init->nfs_version == 2) ? 1 : 4;
 	int status;
 
-	status = lockd_up(protocol);
+	status = lockd_up(nlm_init->protocol);
 	if (status < 0)
 		return ERR_PTR(status);
 
-	host = nlmclnt_lookup_host((struct sockaddr_in *)server_address,
-				   protocol, nlm_version,
-				   server_name, strlen(server_name));
+	host = nlmclnt_lookup_host((struct sockaddr_in *)nlm_init->address,
+				   nlm_init->protocol, nlm_version,
+				   nlm_init->hostname,
+				   strlen(nlm_init->hostname));
 	if (host == NULL) {
 		lockd_down();
 		return ERR_PTR(-ENOLCK);
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 0b3ce86f6fc9..7a15832369e9 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -468,18 +468,21 @@ static int nfs_start_lockd(struct nfs_server *server)
 {
 	struct nlm_host *host;
 	struct nfs_client *clp = server->nfs_client;
-	u32 nfs_version = clp->rpc_ops->version;
-	unsigned short protocol = server->flags & NFS_MOUNT_TCP ?
-						IPPROTO_TCP : IPPROTO_UDP;
+	struct nlmclnt_initdata nlm_init = {
+		.hostname	= clp->cl_hostname,
+		.address	= (struct sockaddr *)&clp->cl_addr,
+		.addrlen	= clp->cl_addrlen,
+		.protocol	= server->flags & NFS_MOUNT_TCP ?
+						IPPROTO_TCP : IPPROTO_UDP,
+		.nfs_version	= clp->rpc_ops->version,
+	};
 
-	if (nfs_version > 3)
+	if (nlm_init.nfs_version > 3)
 		return 0;
 	if (server->flags & NFS_MOUNT_NONLM)
 		return 0;
 
-	host = nlmclnt_init(clp->cl_hostname,
-			    (struct sockaddr *)&clp->cl_addr,
-			    clp->cl_addrlen, protocol, nfs_version);
+	host = nlmclnt_init(&nlm_init);
 	if (IS_ERR(host))
 		return PTR_ERR(host);
 
diff --git a/include/linux/lockd/bind.h b/include/linux/lockd/bind.h
index 73368075af03..3d25bcd139d1 100644
--- a/include/linux/lockd/bind.h
+++ b/include/linux/lockd/bind.h
@@ -32,14 +32,23 @@ struct nlmsvc_binding {
 
 extern struct nlmsvc_binding *	nlmsvc_ops;
 
+/*
+ * Similar to nfs_client_initdata, but without the NFS-specific
+ * rpc_ops field.
+ */
+struct nlmclnt_initdata {
+	const char		*hostname;
+	const struct sockaddr	*address;
+	size_t			addrlen;
+	unsigned short		protocol;
+	u32			nfs_version;
+};
+
 /*
  * Functions exported by the lockd module
  */
-extern struct nlm_host *nlmclnt_init(const char *server_name,
-					const struct sockaddr *server_address,
-					size_t server_addrlen,
-					unsigned short protocol,
-					u32 nfs_version);
+
+extern struct nlm_host *nlmclnt_init(const struct nlmclnt_initdata *nlm_init);
 extern void	nlmclnt_done(struct nlm_host *host);
 
 extern int	nlmclnt_proc(struct nlm_host *host, int cmd,
-- 
cgit v1.2.3


From c0e07cb68db353c0ffbb0f82401cf6d79c253aed Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 14 Jan 2008 12:32:05 -0500
Subject: NFS: NFS version number is unsigned

RPC protocol version numbers are unsigned.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/nfs_xdr.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d342ae5c76ab..f301d0b8babc 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -774,7 +774,7 @@ struct nfs_access_entry;
  * RPC procedure vector for NFSv2/NFSv3 demuxing
  */
 struct nfs_rpc_ops {
-	int	version;		/* Protocol version */
+	u32	version;		/* Protocol version */
 	struct dentry_operations *dentry_ops;
 	const struct inode_operations *dir_inode_ops;
 	const struct inode_operations *file_inode_ops;
-- 
cgit v1.2.3


From f1ec08cb9492cab579f85f9d937c79788b1dfde3 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 14 Jan 2008 15:11:53 -0500
Subject: SUNRPC: Use appropriate argument types in rpcb client

Clean up: Follow recommendations of Chapter 5 of Documentation/CodingStyle
and use "u32" instead of "__u32" for types in definitions that are not
shared with user space.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h | 2 +-
 net/sunrpc/rpcb_clnt.c      | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 5a13da2573b0..3e9addc741c1 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -125,7 +125,7 @@ void		rpc_shutdown_client(struct rpc_clnt *);
 void		rpc_release_client(struct rpc_clnt *);
 
 int		rpcb_register(u32, u32, int, unsigned short, int *);
-int		rpcb_getport_sync(struct sockaddr_in *, __u32, __u32, int);
+int		rpcb_getport_sync(struct sockaddr_in *, u32, u32, int);
 void		rpcb_getport_async(struct rpc_task *);
 
 void		rpc_call_start(struct rpc_task *);
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 8962ac6b4c11..77af989ecc9c 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -205,8 +205,7 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay)
  *
  * XXX: Needs to support IPv6, and rpcbind versions 3 and 4
  */
-int rpcb_getport_sync(struct sockaddr_in *sin, __u32 prog,
-		      __u32 vers, int prot)
+int rpcb_getport_sync(struct sockaddr_in *sin, u32 prog, u32 vers, int prot)
 {
 	struct rpcbind_args map = {
 		.r_prog		= prog,
-- 
cgit v1.2.3


From 3a10c30acc4821ca000b52ed0edafd0d3bf26a52 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 23 Jan 2008 08:58:59 +0200
Subject: nfs: obliterate NFS_FLAGS macro

use NFS_I(inode)->flags instead

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 8 ++++----
 fs/nfs/inode.c         | 6 +++---
 fs/nfs/read.c          | 2 +-
 include/linux/nfs_fs.h | 5 ++---
 4 files changed, 10 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 72d141a0dbd8..c578d942f000 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -192,7 +192,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page)
 		/* We requested READDIRPLUS, but the server doesn't grok it */
 		if (error == -ENOTSUPP && desc->plus) {
 			NFS_SERVER(inode)->caps &= ~NFS_CAP_READDIRPLUS;
-			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
+			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
 			desc->plus = 0;
 			goto again;
 		}
@@ -577,7 +577,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 			break;
 		}
 		if (res == -ETOOSMALL && desc->plus) {
-			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
+			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
 			nfs_zap_caches(inode);
 			desc->plus = 0;
 			desc->entry->eof = 0;
@@ -1760,7 +1760,7 @@ static void __nfs_access_zap_cache(struct inode *inode)
 void nfs_access_zap_cache(struct inode *inode)
 {
 	/* Remove from global LRU init */
-	if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) {
+	if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) {
 		spin_lock(&nfs_access_lru_lock);
 		list_del_init(&NFS_I(inode)->access_cache_inode_lru);
 		spin_unlock(&nfs_access_lru_lock);
@@ -1874,7 +1874,7 @@ static void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *s
 	smp_mb__after_atomic_inc();
 
 	/* Add inode to global LRU list */
-	if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) {
+	if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) {
 		spin_lock(&nfs_access_lru_lock);
 		list_add_tail(&NFS_I(inode)->access_cache_inode_lru, &nfs_access_lru_list);
 		spin_unlock(&nfs_access_lru_lock);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 5747d49bdd76..9d7b08c43865 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -192,7 +192,7 @@ void nfs_invalidate_atime(struct inode *inode)
  */
 static void nfs_invalidate_inode(struct inode *inode)
 {
-	set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
+	set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
 	nfs_zap_caches_locked(inode);
 }
 
@@ -291,7 +291,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 			inode->i_fop = &nfs_dir_operations;
 			if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS)
 			    && fattr->size <= NFS_LIMIT_READDIRPLUS)
-				set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
+				set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
 			/* Deal with crossing mountpoints */
 			if (!nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) {
 				if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
@@ -668,7 +668,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		if (status == -ESTALE) {
 			nfs_zap_caches(inode);
 			if (!S_ISDIR(inode->i_mode))
-				set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
+				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
 		}
 		goto out;
 	}
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index efc121c494fe..8fd6dfbe1bc3 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -336,7 +336,7 @@ int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data)
 	nfs_add_stats(data->inode, NFSIOS_SERVERREADBYTES, data->res.count);
 
 	if (task->tk_status == -ESTALE) {
-		set_bit(NFS_INO_STALE, &NFS_FLAGS(data->inode));
+		set_bit(NFS_INO_STALE, &NFS_I(data->inode)->flags);
 		nfs_mark_for_revalidate(data->inode);
 	}
 	return 0;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 7095aaa087d8..9e7c24a2aca9 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -214,8 +214,7 @@ static inline struct nfs_inode *NFS_I(struct inode *inode)
 	(S_ISDIR(inode->i_mode)? NFS_SERVER(inode)->acdirmax \
 			       : NFS_SERVER(inode)->acregmax)
 
-#define NFS_FLAGS(inode)		(NFS_I(inode)->flags)
-#define NFS_STALE(inode)		(test_bit(NFS_INO_STALE, &NFS_FLAGS(inode)))
+#define NFS_STALE(inode)		(test_bit(NFS_INO_STALE, &NFS_I(inode)->flags))
 
 #define NFS_FILEID(inode)		(NFS_I(inode)->fileid)
 
@@ -237,7 +236,7 @@ static inline int nfs_server_capable(struct inode *inode, int cap)
 
 static inline int NFS_USE_READDIRPLUS(struct inode *inode)
 {
-	return test_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
+	return test_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
 }
 
 static inline void nfs_set_verifier(struct dentry * dentry, unsigned long verf)
-- 
cgit v1.2.3


From 99fadcd76465842c014c88b8c9c19b457e9debc0 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 23 Jan 2008 08:59:08 +0200
Subject: nfs: convert NFS_*(inode) helpers to static inline

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/inode.c         |  2 +-
 include/linux/nfs_fs.h | 76 ++++++++++++++++++++++++++++++++++++++------------
 2 files changed, 59 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 9d7b08c43865..5d381cfbfe7e 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -229,7 +229,7 @@ nfs_init_locked(struct inode *inode, void *opaque)
 	struct nfs_find_desc	*desc = (struct nfs_find_desc *)opaque;
 	struct nfs_fattr	*fattr = desc->fattr;
 
-	NFS_FILEID(inode) = fattr->fileid;
+	set_nfs_fileid(inode, fattr->fileid);
 	nfs_copy_fh(NFS_FH(inode), desc->fh);
 	return 0;
 }
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 9e7c24a2aca9..099ddb4481c0 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -196,27 +196,67 @@ struct nfs_inode {
 #define NFS_INO_STALE		(2)		/* possible stale inode */
 #define NFS_INO_ACL_LRU_SET	(3)		/* Inode is on the LRU list */
 
-static inline struct nfs_inode *NFS_I(struct inode *inode)
+static inline struct nfs_inode *NFS_I(const struct inode *inode)
 {
 	return container_of(inode, struct nfs_inode, vfs_inode);
 }
-#define NFS_SB(s)		((struct nfs_server *)(s->s_fs_info))
-
-#define NFS_FH(inode)			(&NFS_I(inode)->fh)
-#define NFS_SERVER(inode)		(NFS_SB(inode->i_sb))
-#define NFS_CLIENT(inode)		(NFS_SERVER(inode)->client)
-#define NFS_PROTO(inode)		(NFS_SERVER(inode)->nfs_client->rpc_ops)
-#define NFS_COOKIEVERF(inode)		(NFS_I(inode)->cookieverf)
-#define NFS_MINATTRTIMEO(inode) \
-	(S_ISDIR(inode->i_mode)? NFS_SERVER(inode)->acdirmin \
-			       : NFS_SERVER(inode)->acregmin)
-#define NFS_MAXATTRTIMEO(inode) \
-	(S_ISDIR(inode->i_mode)? NFS_SERVER(inode)->acdirmax \
-			       : NFS_SERVER(inode)->acregmax)
-
-#define NFS_STALE(inode)		(test_bit(NFS_INO_STALE, &NFS_I(inode)->flags))
-
-#define NFS_FILEID(inode)		(NFS_I(inode)->fileid)
+
+static inline struct nfs_server *NFS_SB(const struct super_block *s)
+{
+	return (struct nfs_server *)(s->s_fs_info);
+}
+
+static inline struct nfs_fh *NFS_FH(const struct inode *inode)
+{
+	return &NFS_I(inode)->fh;
+}
+
+static inline struct nfs_server *NFS_SERVER(const struct inode *inode)
+{
+	return NFS_SB(inode->i_sb);
+}
+
+static inline struct rpc_clnt *NFS_CLIENT(const struct inode *inode)
+{
+	return NFS_SERVER(inode)->client;
+}
+
+static inline const struct nfs_rpc_ops *NFS_PROTO(const struct inode *inode)
+{
+	return NFS_SERVER(inode)->nfs_client->rpc_ops;
+}
+
+static inline __be32 *NFS_COOKIEVERF(const struct inode *inode)
+{
+	return NFS_I(inode)->cookieverf;
+}
+
+static inline unsigned NFS_MINATTRTIMEO(const struct inode *inode)
+{
+	struct nfs_server *nfss = NFS_SERVER(inode);
+	return S_ISDIR(inode->i_mode) ? nfss->acdirmin : nfss->acregmin;
+}
+
+static inline unsigned NFS_MAXATTRTIMEO(const struct inode *inode)
+{
+	struct nfs_server *nfss = NFS_SERVER(inode);
+	return S_ISDIR(inode->i_mode) ? nfss->acdirmax : nfss->acregmax;
+}
+
+static inline int NFS_STALE(const struct inode *inode)
+{
+	return test_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+}
+
+static inline __u64 NFS_FILEID(const struct inode *inode)
+{
+	return NFS_I(inode)->fileid;
+}
+
+static inline void set_nfs_fileid(struct inode *inode, __u64 fileid)
+{
+	NFS_I(inode)->fileid = fileid;
+}
 
 static inline void nfs_mark_for_revalidate(struct inode *inode)
 {
-- 
cgit v1.2.3


From ec04b13f67be3c90b38c625f4b8bdfea54c1ff60 Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@gmail.com>
Date: Fri, 28 Dec 2007 14:26:24 +0530
Subject: lguest: Reboot support

Reboot Implemented

(Prevent fd leak, fix style and fix documentation --RR)

Signed-off-by: Balaji Rao <balajirrao@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 Documentation/lguest/lguest.c  | 32 ++++++++++++++++++++++++++++++--
 arch/x86/lguest/boot.c         | 11 +++++++++--
 drivers/lguest/core.c          |  2 ++
 drivers/lguest/hypercalls.c    |  6 ++++--
 include/asm-x86/lguest_hcall.h |  6 +++++-
 5 files changed, 50 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c
index 9b0e322118b5..86cac3e622ab 100644
--- a/Documentation/lguest/lguest.c
+++ b/Documentation/lguest/lguest.c
@@ -153,6 +153,9 @@ struct virtqueue
 	void (*handle_output)(int fd, struct virtqueue *me);
 };
 
+/* Remember the arguments to the program so we can "reboot" */
+static char **main_args;
+
 /* Since guest is UP and we don't run at the same time, we don't need barriers.
  * But I include them in the code in case others copy it. */
 #define wmb()
@@ -1489,7 +1492,9 @@ static void setup_block_file(const char *filename)
 
 	/* Create stack for thread and run it */
 	stack = malloc(32768);
-	if (clone(io_thread, stack + 32768, CLONE_VM, dev) == -1)
+	/* SIGCHLD - We dont "wait" for our cloned thread, so prevent it from
+	 * becoming a zombie. */
+	if (clone(io_thread, stack + 32768,  CLONE_VM | SIGCHLD, dev) == -1)
 		err(1, "Creating clone");
 
 	/* We don't need to keep the I/O thread's end of the pipes open. */
@@ -1499,7 +1504,21 @@ static void setup_block_file(const char *filename)
 	verbose("device %u: virtblock %llu sectors\n",
 		devices.device_num, cap);
 }
-/* That's the end of device setup. */
+/* That's the end of device setup. :*/
+
+/* Reboot */
+static void __attribute__((noreturn)) restart_guest(void)
+{
+	unsigned int i;
+
+	/* Closing pipes causes the waker thread and io_threads to die, and
+	 * closing /dev/lguest cleans up the Guest.  Since we don't track all
+	 * open fds, we simply close everything beyond stderr. */
+	for (i = 3; i < FD_SETSIZE; i++)
+		close(i);
+	execv(main_args[0], main_args);
+	err(1, "Could not exec %s", main_args[0]);
+}
 
 /*L:220 Finally we reach the core of the Launcher, which runs the Guest, serves
  * its input and output, and finally, lays it to rest. */
@@ -1523,6 +1542,9 @@ static void __attribute__((noreturn)) run_guest(int lguest_fd)
 			char reason[1024] = { 0 };
 			read(lguest_fd, reason, sizeof(reason)-1);
 			errx(1, "%s", reason);
+		/* ERESTART means that we need to reboot the guest */
+		} else if (errno == ERESTART) {
+			restart_guest();
 		/* EAGAIN means the Waker wanted us to look at some input.
 		 * Anything else means a bug or incompatible change. */
 		} else if (errno != EAGAIN)
@@ -1571,6 +1593,12 @@ int main(int argc, char *argv[])
 	/* If they specify an initrd file to load. */
 	const char *initrd_name = NULL;
 
+	/* Save the args: we "reboot" by execing ourselves again. */
+	main_args = argv;
+	/* We don't "wait" for the children, so prevent them from becoming
+	 * zombies. */
+	signal(SIGCHLD, SIG_IGN);
+
 	/* First we initialize the device list.  Since console and network
 	 * device receive input from a file descriptor, we keep an fdset
 	 * (infds) and the maximum fd number (max_infd) with the head of the
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index 92c56117eae5..d6b18e2e5431 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -67,6 +67,7 @@
 #include <asm/mce.h>
 #include <asm/io.h>
 #include <asm/i387.h>
+#include <asm/reboot.h>		/* for struct machine_ops */
 
 /*G:010 Welcome to the Guest!
  *
@@ -812,7 +813,7 @@ static void lguest_safe_halt(void)
  * rather than virtual addresses, so we use __pa() here. */
 static void lguest_power_off(void)
 {
-	hcall(LHCALL_CRASH, __pa("Power down"), 0, 0);
+	hcall(LHCALL_SHUTDOWN, __pa("Power down"), LGUEST_SHUTDOWN_POWEROFF, 0);
 }
 
 /*
@@ -822,7 +823,7 @@ static void lguest_power_off(void)
  */
 static int lguest_panic(struct notifier_block *nb, unsigned long l, void *p)
 {
-	hcall(LHCALL_CRASH, __pa(p), 0, 0);
+	hcall(LHCALL_SHUTDOWN, __pa(p), LGUEST_SHUTDOWN_POWEROFF, 0);
 	/* The hcall won't return, but to keep gcc happy, we're "done". */
 	return NOTIFY_DONE;
 }
@@ -926,6 +927,11 @@ static unsigned lguest_patch(u8 type, u16 clobber, void *ibuf,
 	return insn_len;
 }
 
+static void lguest_restart(char *reason)
+{
+	hcall(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART, 0);
+}
+
 /*G:030 Once we get to lguest_init(), we know we're a Guest.  The pv_ops
  * structures in the kernel provide points for (almost) every routine we have
  * to override to avoid privileged instructions. */
@@ -1059,6 +1065,7 @@ __init void lguest_init(void)
 	 * the Guest routine to power off. */
 	pm_power_off = lguest_power_off;
 
+	machine_ops.restart = lguest_restart;
 	/* Now we're set up, call start_kernel() in init/main.c and we proceed
 	 * to boot as normal.  It never returns. */
 	start_kernel();
diff --git a/drivers/lguest/core.c b/drivers/lguest/core.c
index f10abc8d7635..c1069bceba11 100644
--- a/drivers/lguest/core.c
+++ b/drivers/lguest/core.c
@@ -235,6 +235,8 @@ int run_guest(struct lguest *lg, unsigned long __user *user)
 		lguest_arch_handle_trap(lg);
 	}
 
+	if (lg->dead == ERR_PTR(-ERESTART))
+		return -ERESTART;
 	/* The Guest is dead => "No such file or directory" */
 	return -ENOENT;
 }
diff --git a/drivers/lguest/hypercalls.c b/drivers/lguest/hypercalls.c
index b478affe8f91..05fad6fa8049 100644
--- a/drivers/lguest/hypercalls.c
+++ b/drivers/lguest/hypercalls.c
@@ -41,8 +41,8 @@ static void do_hcall(struct lguest *lg, struct hcall_args *args)
 		 * do that. */
 		kill_guest(lg, "already have lguest_data");
 		break;
-	case LHCALL_CRASH: {
-		/* Crash is such a trivial hypercall that we do it in four
+	case LHCALL_SHUTDOWN: {
+		/* Shutdown is such a trivial hypercall that we do it in four
 		 * lines right here. */
 		char msg[128];
 		/* If the lgread fails, it will call kill_guest() itself; the
@@ -50,6 +50,8 @@ static void do_hcall(struct lguest *lg, struct hcall_args *args)
 		__lgread(lg, msg, args->arg1, sizeof(msg));
 		msg[sizeof(msg)-1] = '\0';
 		kill_guest(lg, "CRASH: %s", msg);
+		if (args->arg2 == LGUEST_SHUTDOWN_RESTART)
+			lg->dead = ERR_PTR(-ERESTART);
 		break;
 	}
 	case LHCALL_FLUSH_TLB:
diff --git a/include/asm-x86/lguest_hcall.h b/include/asm-x86/lguest_hcall.h
index 2091779e91fb..758b9a5d4539 100644
--- a/include/asm-x86/lguest_hcall.h
+++ b/include/asm-x86/lguest_hcall.h
@@ -4,7 +4,7 @@
 
 #define LHCALL_FLUSH_ASYNC	0
 #define LHCALL_LGUEST_INIT	1
-#define LHCALL_CRASH		2
+#define LHCALL_SHUTDOWN		2
 #define LHCALL_LOAD_GDT		3
 #define LHCALL_NEW_PGTABLE	4
 #define LHCALL_FLUSH_TLB	5
@@ -20,6 +20,10 @@
 
 #define LGUEST_TRAP_ENTRY 0x1F
 
+/* Argument number 3 to LHCALL_LGUEST_SHUTDOWN */
+#define LGUEST_SHUTDOWN_POWEROFF	1
+#define LGUEST_SHUTDOWN_RESTART		2
+
 #ifndef __ASSEMBLY__
 #include <asm/hw_irq.h>
 
-- 
cgit v1.2.3


From fc708b3e407dfd2e12ba9a6cf35bd0bffad1796d Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Mon, 7 Jan 2008 11:05:33 -0200
Subject: lguest: replace lguest_arch with lg_cpu_arch.

The fields found in lguest_arch are not really per-guest,
but per-cpu (gdt, idt, etc). So this patch turns lguest_arch
into lg_cpu_arch.

It makes sense to have a per-guest per-arch struct, but this
can be addressed later, when the need arrives.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/lguest/interrupts_and_traps.c | 28 +++++++++++------------
 drivers/lguest/lg.h                   | 19 ++++++++--------
 drivers/lguest/segments.c             | 42 ++++++++++++++++++-----------------
 drivers/lguest/x86/core.c             | 24 +++++++++-----------
 include/asm-x86/lguest.h              |  2 +-
 5 files changed, 58 insertions(+), 57 deletions(-)

(limited to 'include')

diff --git a/drivers/lguest/interrupts_and_traps.c b/drivers/lguest/interrupts_and_traps.c
index 468faf8233d6..306b93c71dcc 100644
--- a/drivers/lguest/interrupts_and_traps.c
+++ b/drivers/lguest/interrupts_and_traps.c
@@ -178,7 +178,7 @@ void maybe_do_interrupt(struct lg_cpu *cpu)
 	/* Look at the IDT entry the Guest gave us for this interrupt.  The
 	 * first 32 (FIRST_EXTERNAL_VECTOR) entries are for traps, so we skip
 	 * over them. */
-	idt = &lg->arch.idt[FIRST_EXTERNAL_VECTOR+irq];
+	idt = &cpu->arch.idt[FIRST_EXTERNAL_VECTOR+irq];
 	/* If they don't have a handler (yet?), we just ignore it */
 	if (idt_present(idt->a, idt->b)) {
 		/* OK, mark it no longer pending and deliver it. */
@@ -251,15 +251,15 @@ int deliver_trap(struct lg_cpu *cpu, unsigned int num)
 {
 	/* Trap numbers are always 8 bit, but we set an impossible trap number
 	 * for traps inside the Switcher, so check that here. */
-	if (num >= ARRAY_SIZE(cpu->lg->arch.idt))
+	if (num >= ARRAY_SIZE(cpu->arch.idt))
 		return 0;
 
 	/* Early on the Guest hasn't set the IDT entries (or maybe it put a
 	 * bogus one in): if we fail here, the Guest will be killed. */
-	if (!idt_present(cpu->lg->arch.idt[num].a, cpu->lg->arch.idt[num].b))
+	if (!idt_present(cpu->arch.idt[num].a, cpu->arch.idt[num].b))
 		return 0;
-	set_guest_interrupt(cpu, cpu->lg->arch.idt[num].a,
-			    cpu->lg->arch.idt[num].b, has_err(num));
+	set_guest_interrupt(cpu, cpu->arch.idt[num].a,
+			    cpu->arch.idt[num].b, has_err(num));
 	return 1;
 }
 
@@ -385,7 +385,7 @@ static void set_trap(struct lguest *lg, struct desc_struct *trap,
  *
  * We saw the Guest setting Interrupt Descriptor Table (IDT) entries with the
  * LHCALL_LOAD_IDT_ENTRY hypercall before: that comes here. */
-void load_guest_idt_entry(struct lguest *lg, unsigned int num, u32 lo, u32 hi)
+void load_guest_idt_entry(struct lg_cpu *cpu, unsigned int num, u32 lo, u32 hi)
 {
 	/* Guest never handles: NMI, doublefault, spurious interrupt or
 	 * hypercall.  We ignore when it tries to set them. */
@@ -394,13 +394,13 @@ void load_guest_idt_entry(struct lguest *lg, unsigned int num, u32 lo, u32 hi)
 
 	/* Mark the IDT as changed: next time the Guest runs we'll know we have
 	 * to copy this again. */
-	lg->changed |= CHANGED_IDT;
+	cpu->lg->changed |= CHANGED_IDT;
 
 	/* Check that the Guest doesn't try to step outside the bounds. */
-	if (num >= ARRAY_SIZE(lg->arch.idt))
-		kill_guest(lg, "Setting idt entry %u", num);
+	if (num >= ARRAY_SIZE(cpu->arch.idt))
+		kill_guest(cpu->lg, "Setting idt entry %u", num);
 	else
-		set_trap(lg, &lg->arch.idt[num], num, lo, hi);
+		set_trap(cpu->lg, &cpu->arch.idt[num], num, lo, hi);
 }
 
 /* The default entry for each interrupt points into the Switcher routines which
@@ -436,14 +436,14 @@ void setup_default_idt_entries(struct lguest_ro_state *state,
 /*H:240 We don't use the IDT entries in the "struct lguest" directly, instead
  * we copy them into the IDT which we've set up for Guests on this CPU, just
  * before we run the Guest.  This routine does that copy. */
-void copy_traps(const struct lguest *lg, struct desc_struct *idt,
+void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt,
 		const unsigned long *def)
 {
 	unsigned int i;
 
 	/* We can simply copy the direct traps, otherwise we use the default
 	 * ones in the Switcher: they will return to the Host. */
-	for (i = 0; i < ARRAY_SIZE(lg->arch.idt); i++) {
+	for (i = 0; i < ARRAY_SIZE(cpu->arch.idt); i++) {
 		/* If no Guest can ever override this trap, leave it alone. */
 		if (!direct_trap(i))
 			continue;
@@ -452,8 +452,8 @@ void copy_traps(const struct lguest *lg, struct desc_struct *idt,
 		 * Interrupt gates (type 14) disable interrupts as they are
 		 * entered, which we never let the Guest do.  Not present
 		 * entries (type 0x0) also can't go direct, of course. */
-		if (idt_type(lg->arch.idt[i].a, lg->arch.idt[i].b) == 0xF)
-			idt[i] = lg->arch.idt[i];
+		if (idt_type(cpu->arch.idt[i].a, cpu->arch.idt[i].b) == 0xF)
+			idt[i] = cpu->arch.idt[i];
 		else
 			/* Reset it to the default. */
 			default_idt_entry(&idt[i], i, def[i]);
diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h
index 35b331230c55..d08b85342b92 100644
--- a/drivers/lguest/lg.h
+++ b/drivers/lguest/lg.h
@@ -57,6 +57,8 @@ struct lg_cpu {
 
 	/* Pending virtual interrupts */
 	DECLARE_BITMAP(irqs_pending, LGUEST_IRQS);
+
+	struct lg_cpu_arch arch;
 };
 
 /* The private info the thread maintains about the guest. */
@@ -99,8 +101,6 @@ struct lguest
 
 	/* Dead? */
 	const char *dead;
-
-	struct lguest_arch arch;
 };
 
 extern struct mutex lguest_lock;
@@ -139,12 +139,13 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user);
 /* interrupts_and_traps.c: */
 void maybe_do_interrupt(struct lg_cpu *cpu);
 int deliver_trap(struct lg_cpu *cpu, unsigned int num);
-void load_guest_idt_entry(struct lguest *lg, unsigned int i, u32 low, u32 hi);
+void load_guest_idt_entry(struct lg_cpu *cpu, unsigned int i,
+			  u32 low, u32 hi);
 void guest_set_stack(struct lguest *lg, u32 seg, u32 esp, unsigned int pages);
 void pin_stack_pages(struct lguest *lg);
 void setup_default_idt_entries(struct lguest_ro_state *state,
 			       const unsigned long *def);
-void copy_traps(const struct lguest *lg, struct desc_struct *idt,
+void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt,
 		const unsigned long *def);
 void guest_set_clockevent(struct lg_cpu *cpu, unsigned long delta);
 void init_clockdev(struct lg_cpu *cpu);
@@ -154,11 +155,11 @@ void free_interrupts(void);
 
 /* segments.c: */
 void setup_default_gdt_entries(struct lguest_ro_state *state);
-void setup_guest_gdt(struct lguest *lg);
-void load_guest_gdt(struct lguest *lg, unsigned long table, u32 num);
-void guest_load_tls(struct lguest *lg, unsigned long tls_array);
-void copy_gdt(const struct lguest *lg, struct desc_struct *gdt);
-void copy_gdt_tls(const struct lguest *lg, struct desc_struct *gdt);
+void setup_guest_gdt(struct lg_cpu *cpu);
+void load_guest_gdt(struct lg_cpu *cpu, unsigned long table, u32 num);
+void guest_load_tls(struct lg_cpu *cpu, unsigned long tls_array);
+void copy_gdt(const struct lg_cpu *cpu, struct desc_struct *gdt);
+void copy_gdt_tls(const struct lg_cpu *cpu, struct desc_struct *gdt);
 
 /* page_tables.c: */
 int init_guest_pagetable(struct lguest *lg, unsigned long pgtable);
diff --git a/drivers/lguest/segments.c b/drivers/lguest/segments.c
index 9e189cbec7dd..02138450ecf5 100644
--- a/drivers/lguest/segments.c
+++ b/drivers/lguest/segments.c
@@ -58,7 +58,7 @@ static int ignored_gdt(unsigned int num)
  * Protection Fault in the Switcher when it restores a Guest segment register
  * which tries to use that entry.  Then we kill the Guest for causing such a
  * mess: the message will be "unhandled trap 256". */
-static void fixup_gdt_table(struct lguest *lg, unsigned start, unsigned end)
+static void fixup_gdt_table(struct lg_cpu *cpu, unsigned start, unsigned end)
 {
 	unsigned int i;
 
@@ -71,14 +71,14 @@ static void fixup_gdt_table(struct lguest *lg, unsigned start, unsigned end)
 		/* Segment descriptors contain a privilege level: the Guest is
 		 * sometimes careless and leaves this as 0, even though it's
 		 * running at privilege level 1.  If so, we fix it here. */
-		if ((lg->arch.gdt[i].b & 0x00006000) == 0)
-			lg->arch.gdt[i].b |= (GUEST_PL << 13);
+		if ((cpu->arch.gdt[i].b & 0x00006000) == 0)
+			cpu->arch.gdt[i].b |= (GUEST_PL << 13);
 
 		/* Each descriptor has an "accessed" bit.  If we don't set it
 		 * now, the CPU will try to set it when the Guest first loads
 		 * that entry into a segment register.  But the GDT isn't
 		 * writable by the Guest, so bad things can happen. */
-		lg->arch.gdt[i].b |= 0x00000100;
+		cpu->arch.gdt[i].b |= 0x00000100;
 	}
 }
 
@@ -109,31 +109,31 @@ void setup_default_gdt_entries(struct lguest_ro_state *state)
 
 /* This routine sets up the initial Guest GDT for booting.  All entries start
  * as 0 (unusable). */
-void setup_guest_gdt(struct lguest *lg)
+void setup_guest_gdt(struct lg_cpu *cpu)
 {
 	/* Start with full 0-4G segments... */
-	lg->arch.gdt[GDT_ENTRY_KERNEL_CS] = FULL_EXEC_SEGMENT;
-	lg->arch.gdt[GDT_ENTRY_KERNEL_DS] = FULL_SEGMENT;
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_CS] = FULL_EXEC_SEGMENT;
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_DS] = FULL_SEGMENT;
 	/* ...except the Guest is allowed to use them, so set the privilege
 	 * level appropriately in the flags. */
-	lg->arch.gdt[GDT_ENTRY_KERNEL_CS].b |= (GUEST_PL << 13);
-	lg->arch.gdt[GDT_ENTRY_KERNEL_DS].b |= (GUEST_PL << 13);
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_CS].b |= (GUEST_PL << 13);
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_DS].b |= (GUEST_PL << 13);
 }
 
 /*H:650 An optimization of copy_gdt(), for just the three "thead-local storage"
  * entries. */
-void copy_gdt_tls(const struct lguest *lg, struct desc_struct *gdt)
+void copy_gdt_tls(const struct lg_cpu *cpu, struct desc_struct *gdt)
 {
 	unsigned int i;
 
 	for (i = GDT_ENTRY_TLS_MIN; i <= GDT_ENTRY_TLS_MAX; i++)
-		gdt[i] = lg->arch.gdt[i];
+		gdt[i] = cpu->arch.gdt[i];
 }
 
 /*H:640 When the Guest is run on a different CPU, or the GDT entries have
  * changed, copy_gdt() is called to copy the Guest's GDT entries across to this
  * CPU's GDT. */
-void copy_gdt(const struct lguest *lg, struct desc_struct *gdt)
+void copy_gdt(const struct lg_cpu *cpu, struct desc_struct *gdt)
 {
 	unsigned int i;
 
@@ -141,21 +141,22 @@ void copy_gdt(const struct lguest *lg, struct desc_struct *gdt)
 	 * replaced.  See ignored_gdt() above. */
 	for (i = 0; i < GDT_ENTRIES; i++)
 		if (!ignored_gdt(i))
-			gdt[i] = lg->arch.gdt[i];
+			gdt[i] = cpu->arch.gdt[i];
 }
 
 /*H:620 This is where the Guest asks us to load a new GDT (LHCALL_LOAD_GDT).
  * We copy it from the Guest and tweak the entries. */
-void load_guest_gdt(struct lguest *lg, unsigned long table, u32 num)
+void load_guest_gdt(struct lg_cpu *cpu, unsigned long table, u32 num)
 {
+	struct lguest *lg = cpu->lg;
 	/* We assume the Guest has the same number of GDT entries as the
 	 * Host, otherwise we'd have to dynamically allocate the Guest GDT. */
-	if (num > ARRAY_SIZE(lg->arch.gdt))
+	if (num > ARRAY_SIZE(cpu->arch.gdt))
 		kill_guest(lg, "too many gdt entries %i", num);
 
 	/* We read the whole thing in, then fix it up. */
-	__lgread(lg, lg->arch.gdt, table, num * sizeof(lg->arch.gdt[0]));
-	fixup_gdt_table(lg, 0, ARRAY_SIZE(lg->arch.gdt));
+	__lgread(lg, cpu->arch.gdt, table, num * sizeof(cpu->arch.gdt[0]));
+	fixup_gdt_table(cpu, 0, ARRAY_SIZE(cpu->arch.gdt));
 	/* Mark that the GDT changed so the core knows it has to copy it again,
 	 * even if the Guest is run on the same CPU. */
 	lg->changed |= CHANGED_GDT;
@@ -165,12 +166,13 @@ void load_guest_gdt(struct lguest *lg, unsigned long table, u32 num)
  * Remember that this happens on every context switch, so it's worth
  * optimizing.  But wouldn't it be neater to have a single hypercall to cover
  * both cases? */
-void guest_load_tls(struct lguest *lg, unsigned long gtls)
+void guest_load_tls(struct lg_cpu *cpu, unsigned long gtls)
 {
-	struct desc_struct *tls = &lg->arch.gdt[GDT_ENTRY_TLS_MIN];
+	struct desc_struct *tls = &cpu->arch.gdt[GDT_ENTRY_TLS_MIN];
+	struct lguest *lg = cpu->lg;
 
 	__lgread(lg, tls, gtls, sizeof(*tls)*GDT_ENTRY_TLS_ENTRIES);
-	fixup_gdt_table(lg, GDT_ENTRY_TLS_MIN, GDT_ENTRY_TLS_MAX+1);
+	fixup_gdt_table(cpu, GDT_ENTRY_TLS_MIN, GDT_ENTRY_TLS_MAX+1);
 	/* Note that just the TLS entries have changed. */
 	lg->changed |= CHANGED_GDT_TLS;
 }
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index d96a93d95aea..e989b8358864 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -100,14 +100,14 @@ static void copy_in_guest_info(struct lg_cpu *cpu, struct lguest_pages *pages)
 
 	/* Copy direct-to-Guest trap entries. */
 	if (lg->changed & CHANGED_IDT)
-		copy_traps(lg, pages->state.guest_idt, default_idt_entries);
+		copy_traps(cpu, pages->state.guest_idt, default_idt_entries);
 
 	/* Copy all GDT entries which the Guest can change. */
 	if (lg->changed & CHANGED_GDT)
-		copy_gdt(lg, pages->state.guest_gdt);
+		copy_gdt(cpu, pages->state.guest_gdt);
 	/* If only the TLS entries have changed, copy them. */
 	else if (lg->changed & CHANGED_GDT_TLS)
-		copy_gdt_tls(lg, pages->state.guest_gdt);
+		copy_gdt_tls(cpu, pages->state.guest_gdt);
 
 	/* Mark the Guest as unchanged for next time. */
 	lg->changed = 0;
@@ -196,7 +196,7 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
 	 * re-enable interrupts an interrupt could fault and thus overwrite
 	 * cr2, or we could even move off to a different CPU. */
 	if (cpu->regs->trapnum == 14)
-		lg->arch.last_pagefault = read_cr2();
+		cpu->arch.last_pagefault = read_cr2();
 	/* Similarly, if we took a trap because the Guest used the FPU,
 	 * we have to restore the FPU it expects to see. */
 	else if (cpu->regs->trapnum == 7)
@@ -307,7 +307,7 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu)
 		 *
 		 * The errcode tells whether this was a read or a write, and
 		 * whether kernel or userspace code. */
-		if (demand_page(lg, lg->arch.last_pagefault, cpu->regs->errcode))
+		if (demand_page(lg,cpu->arch.last_pagefault,cpu->regs->errcode))
 			return;
 
 		/* OK, it's really not there (or not OK): the Guest needs to
@@ -318,7 +318,7 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu)
 		 * happen before it's done the LHCALL_LGUEST_INIT hypercall, so
 		 * lg->lguest_data could be NULL */
 		if (lg->lguest_data &&
-		    put_user(lg->arch.last_pagefault, &lg->lguest_data->cr2))
+		    put_user(cpu->arch.last_pagefault, &lg->lguest_data->cr2))
 			kill_guest(lg, "Writing cr2");
 		break;
 	case 7: /* We've intercepted a Device Not Available fault. */
@@ -349,7 +349,7 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu)
 		 * it handle), it dies with a cryptic error message. */
 		kill_guest(lg, "unhandled trap %li at %#lx (%#lx)",
 			   cpu->regs->trapnum, cpu->regs->eip,
-			   cpu->regs->trapnum == 14 ? lg->arch.last_pagefault
+			   cpu->regs->trapnum == 14 ? cpu->arch.last_pagefault
 			   : cpu->regs->errcode);
 }
 
@@ -495,17 +495,15 @@ void __exit lguest_arch_host_fini(void)
 /*H:122 The i386-specific hypercalls simply farm out to the right functions. */
 int lguest_arch_do_hcall(struct lg_cpu *cpu, struct hcall_args *args)
 {
-	struct lguest *lg = cpu->lg;
-
 	switch (args->arg0) {
 	case LHCALL_LOAD_GDT:
-		load_guest_gdt(lg, args->arg1, args->arg2);
+		load_guest_gdt(cpu, args->arg1, args->arg2);
 		break;
 	case LHCALL_LOAD_IDT_ENTRY:
-		load_guest_idt_entry(lg, args->arg1, args->arg2, args->arg3);
+		load_guest_idt_entry(cpu, args->arg1, args->arg2, args->arg3);
 		break;
 	case LHCALL_LOAD_TLS:
-		guest_load_tls(lg, args->arg1);
+		guest_load_tls(cpu, args->arg1);
 		break;
 	default:
 		/* Bad Guest.  Bad! */
@@ -586,5 +584,5 @@ void lguest_arch_setup_regs(struct lg_cpu *cpu, unsigned long start)
 
 	/* There are a couple of GDT entries the Guest expects when first
 	 * booting. */
-	setup_guest_gdt(cpu->lg);
+	setup_guest_gdt(cpu);
 }
diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h
index ccd338460811..b9d003b8005e 100644
--- a/include/asm-x86/lguest.h
+++ b/include/asm-x86/lguest.h
@@ -56,7 +56,7 @@ struct lguest_ro_state
 	struct desc_struct guest_gdt[GDT_ENTRIES];
 };
 
-struct lguest_arch
+struct lg_cpu_arch
 {
 	/* The GDT entries copied into lguest_ro_state when running. */
 	struct desc_struct gdt[GDT_ENTRIES];
-- 
cgit v1.2.3


From a6fa8e5a6172a5a5bc06ed04f34e50b36c978127 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Wed, 30 Jan 2008 13:30:00 +0100
Subject: time: clean hungarian notation from timers

Clean up hungarian notation from timer code.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/timer.h |  6 ++--
 kernel/timer.c        | 80 +++++++++++++++++++++++++--------------------------
 2 files changed, 42 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/linux/timer.h b/include/linux/timer.h
index 78cf899b4409..de0e71359ede 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -5,7 +5,7 @@
 #include <linux/ktime.h>
 #include <linux/stddef.h>
 
-struct tvec_t_base_s;
+struct tvec_base;
 
 struct timer_list {
 	struct list_head entry;
@@ -14,7 +14,7 @@ struct timer_list {
 	void (*function)(unsigned long);
 	unsigned long data;
 
-	struct tvec_t_base_s *base;
+	struct tvec_base *base;
 #ifdef CONFIG_TIMER_STATS
 	void *start_site;
 	char start_comm[16];
@@ -22,7 +22,7 @@ struct timer_list {
 #endif
 };
 
-extern struct tvec_t_base_s boot_tvec_bases;
+extern struct tvec_base boot_tvec_bases;
 
 #define TIMER_INITIALIZER(_function, _expires, _data) {		\
 		.function = (_function),			\
diff --git a/kernel/timer.c b/kernel/timer.c
index f739dfb539ce..aadfbc8367f5 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -58,59 +58,57 @@ EXPORT_SYMBOL(jiffies_64);
 #define TVN_MASK (TVN_SIZE - 1)
 #define TVR_MASK (TVR_SIZE - 1)
 
-typedef struct tvec_s {
+struct tvec {
 	struct list_head vec[TVN_SIZE];
-} tvec_t;
+};
 
-typedef struct tvec_root_s {
+struct tvec_root {
 	struct list_head vec[TVR_SIZE];
-} tvec_root_t;
+};
 
-struct tvec_t_base_s {
+struct tvec_base {
 	spinlock_t lock;
 	struct timer_list *running_timer;
 	unsigned long timer_jiffies;
-	tvec_root_t tv1;
-	tvec_t tv2;
-	tvec_t tv3;
-	tvec_t tv4;
-	tvec_t tv5;
+	struct tvec_root tv1;
+	struct tvec tv2;
+	struct tvec tv3;
+	struct tvec tv4;
+	struct tvec tv5;
 } ____cacheline_aligned;
 
-typedef struct tvec_t_base_s tvec_base_t;
-
-tvec_base_t boot_tvec_bases;
+struct tvec_base boot_tvec_bases;
 EXPORT_SYMBOL(boot_tvec_bases);
-static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases;
+static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases;
 
 /*
- * Note that all tvec_bases is 2 byte aligned and lower bit of
+ * Note that all tvec_bases are 2 byte aligned and lower bit of
  * base in timer_list is guaranteed to be zero. Use the LSB for
  * the new flag to indicate whether the timer is deferrable
  */
 #define TBASE_DEFERRABLE_FLAG		(0x1)
 
 /* Functions below help us manage 'deferrable' flag */
-static inline unsigned int tbase_get_deferrable(tvec_base_t *base)
+static inline unsigned int tbase_get_deferrable(struct tvec_base *base)
 {
 	return ((unsigned int)(unsigned long)base & TBASE_DEFERRABLE_FLAG);
 }
 
-static inline tvec_base_t *tbase_get_base(tvec_base_t *base)
+static inline struct tvec_base *tbase_get_base(struct tvec_base *base)
 {
-	return ((tvec_base_t *)((unsigned long)base & ~TBASE_DEFERRABLE_FLAG));
+	return ((struct tvec_base *)((unsigned long)base & ~TBASE_DEFERRABLE_FLAG));
 }
 
 static inline void timer_set_deferrable(struct timer_list *timer)
 {
-	timer->base = ((tvec_base_t *)((unsigned long)(timer->base) |
+	timer->base = ((struct tvec_base *)((unsigned long)(timer->base) |
 				       TBASE_DEFERRABLE_FLAG));
 }
 
 static inline void
-timer_set_base(struct timer_list *timer, tvec_base_t *new_base)
+timer_set_base(struct timer_list *timer, struct tvec_base *new_base)
 {
-	timer->base = (tvec_base_t *)((unsigned long)(new_base) |
+	timer->base = (struct tvec_base *)((unsigned long)(new_base) |
 				      tbase_get_deferrable(timer->base));
 }
 
@@ -246,7 +244,7 @@ unsigned long round_jiffies_relative(unsigned long j)
 EXPORT_SYMBOL_GPL(round_jiffies_relative);
 
 
-static inline void set_running_timer(tvec_base_t *base,
+static inline void set_running_timer(struct tvec_base *base,
 					struct timer_list *timer)
 {
 #ifdef CONFIG_SMP
@@ -254,7 +252,7 @@ static inline void set_running_timer(tvec_base_t *base,
 #endif
 }
 
-static void internal_add_timer(tvec_base_t *base, struct timer_list *timer)
+static void internal_add_timer(struct tvec_base *base, struct timer_list *timer)
 {
 	unsigned long expires = timer->expires;
 	unsigned long idx = expires - base->timer_jiffies;
@@ -371,14 +369,14 @@ static inline void detach_timer(struct timer_list *timer,
  * possible to set timer->base = NULL and drop the lock: the timer remains
  * locked.
  */
-static tvec_base_t *lock_timer_base(struct timer_list *timer,
+static struct tvec_base *lock_timer_base(struct timer_list *timer,
 					unsigned long *flags)
 	__acquires(timer->base->lock)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 
 	for (;;) {
-		tvec_base_t *prelock_base = timer->base;
+		struct tvec_base *prelock_base = timer->base;
 		base = tbase_get_base(prelock_base);
 		if (likely(base != NULL)) {
 			spin_lock_irqsave(&base->lock, *flags);
@@ -393,7 +391,7 @@ static tvec_base_t *lock_timer_base(struct timer_list *timer,
 
 int __mod_timer(struct timer_list *timer, unsigned long expires)
 {
-	tvec_base_t *base, *new_base;
+	struct tvec_base *base, *new_base;
 	unsigned long flags;
 	int ret = 0;
 
@@ -445,7 +443,7 @@ EXPORT_SYMBOL(__mod_timer);
  */
 void add_timer_on(struct timer_list *timer, int cpu)
 {
-	tvec_base_t *base = per_cpu(tvec_bases, cpu);
+	struct tvec_base *base = per_cpu(tvec_bases, cpu);
 	unsigned long flags;
 
 	timer_stats_timer_set_start_info(timer);
@@ -508,7 +506,7 @@ EXPORT_SYMBOL(mod_timer);
  */
 int del_timer(struct timer_list *timer)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 	unsigned long flags;
 	int ret = 0;
 
@@ -539,7 +537,7 @@ EXPORT_SYMBOL(del_timer);
  */
 int try_to_del_timer_sync(struct timer_list *timer)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 	unsigned long flags;
 	int ret = -1;
 
@@ -591,7 +589,7 @@ int del_timer_sync(struct timer_list *timer)
 EXPORT_SYMBOL(del_timer_sync);
 #endif
 
-static int cascade(tvec_base_t *base, tvec_t *tv, int index)
+static int cascade(struct tvec_base *base, struct tvec *tv, int index)
 {
 	/* cascade all the timers from tv up one level */
 	struct timer_list *timer, *tmp;
@@ -620,7 +618,7 @@ static int cascade(tvec_base_t *base, tvec_t *tv, int index)
  * This function cascades all vectors and executes all expired timer
  * vectors.
  */
-static inline void __run_timers(tvec_base_t *base)
+static inline void __run_timers(struct tvec_base *base)
 {
 	struct timer_list *timer;
 
@@ -678,13 +676,13 @@ static inline void __run_timers(tvec_base_t *base)
  * is used on S/390 to stop all activity when a cpus is idle.
  * This functions needs to be called disabled.
  */
-static unsigned long __next_timer_interrupt(tvec_base_t *base)
+static unsigned long __next_timer_interrupt(struct tvec_base *base)
 {
 	unsigned long timer_jiffies = base->timer_jiffies;
 	unsigned long expires = timer_jiffies + NEXT_TIMER_MAX_DELTA;
 	int index, slot, array, found = 0;
 	struct timer_list *nte;
-	tvec_t *varray[4];
+	struct tvec *varray[4];
 
 	/* Look for timer events in tv1. */
 	index = slot = timer_jiffies & TVR_MASK;
@@ -716,7 +714,7 @@ cascade:
 	varray[3] = &base->tv5;
 
 	for (array = 0; array < 4; array++) {
-		tvec_t *varp = varray[array];
+		struct tvec *varp = varray[array];
 
 		index = slot = timer_jiffies & TVN_MASK;
 		do {
@@ -795,7 +793,7 @@ static unsigned long cmp_next_hrtimer_event(unsigned long now,
  */
 unsigned long get_next_timer_interrupt(unsigned long now)
 {
-	tvec_base_t *base = __get_cpu_var(tvec_bases);
+	struct tvec_base *base = __get_cpu_var(tvec_bases);
 	unsigned long expires;
 
 	spin_lock(&base->lock);
@@ -894,7 +892,7 @@ static inline void calc_load(unsigned long ticks)
  */
 static void run_timer_softirq(struct softirq_action *h)
 {
-	tvec_base_t *base = __get_cpu_var(tvec_bases);
+	struct tvec_base *base = __get_cpu_var(tvec_bases);
 
 	hrtimer_run_pending();
 
@@ -1223,7 +1221,7 @@ static struct lock_class_key base_lock_keys[NR_CPUS];
 static int __cpuinit init_timers_cpu(int cpu)
 {
 	int j;
-	tvec_base_t *base;
+	struct tvec_base *base;
 	static char __cpuinitdata tvec_base_done[NR_CPUS];
 
 	if (!tvec_base_done[cpu]) {
@@ -1278,7 +1276,7 @@ static int __cpuinit init_timers_cpu(int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
+static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head)
 {
 	struct timer_list *timer;
 
@@ -1292,8 +1290,8 @@ static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
 
 static void __cpuinit migrate_timers(int cpu)
 {
-	tvec_base_t *old_base;
-	tvec_base_t *new_base;
+	struct tvec_base *old_base;
+	struct tvec_base *new_base;
 	int i;
 
 	BUG_ON(cpu_online(cpu));
-- 
cgit v1.2.3


From 1d76c2622813fbc692b0d323028cfef9ee36051a Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 30 Jan 2008 13:30:01 +0100
Subject: clocksource: make CLOCKSOURCE_MASK bullet-proof

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/clocksource.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index 107787aacb64..07b42153de24 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -103,7 +103,7 @@ struct clocksource {
 #define CLOCK_SOURCE_VALID_FOR_HRES		0x20
 
 /* simplify initialization of mask field */
-#define CLOCKSOURCE_MASK(bits) (cycle_t)(bits<64 ? ((1ULL<<bits)-1) : -1)
+#define CLOCKSOURCE_MASK(bits) (cycle_t)((bits) < 64 ? ((1ULL<<(bits))-1) : -1)
 
 /**
  * clocksource_khz2mult - calculates mult from khz and shift
-- 
cgit v1.2.3


From df619e6bafeb8e129aba3173d384544d7fa6e3c2 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:30:02 +0100
Subject: x86: offer is_hpet_enabled() on !CONFIG_HPET_TIMER too

offer is_hpet_enabled() on !CONFIG_HPET_TIMER too.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/hpet.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/hpet.h b/include/asm-x86/hpet.h
index ad8d6e758785..624f600f7161 100644
--- a/include/asm-x86/hpet.h
+++ b/include/asm-x86/hpet.h
@@ -80,10 +80,11 @@ extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id);
 
 #endif /* CONFIG_HPET_EMULATE_RTC */
 
-#else
+#else /* CONFIG_HPET_TIMER */
 
 static inline int hpet_enable(void) { return 0; }
 static inline unsigned long hpet_readl(unsigned long a) { return 0; }
+static inline int is_hpet_enabled(void) { return 0; }
 
-#endif /* CONFIG_HPET_TIMER */
+#endif
 #endif /* ASM_X86_HPET_H */
-- 
cgit v1.2.3


From 4713e22ce81eb8b3353e16435362eb3d0ec95640 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:02 +0100
Subject: clocksource: add unregister function to disable unusable clocksources

On x86 the PIT might become an unusable clocksource. Add an unregister
function to provide a possibilty to remove the PIT from the list of
available clock sources.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/i8253.c     |  1 +
 include/linux/clocksource.h |  1 +
 kernel/time/clocksource.c   | 15 +++++++++++++++
 3 files changed, 17 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index 0f8f35458a8f..decc5d294d76 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -13,6 +13,7 @@
 #include <asm/delay.h>
 #include <asm/i8253.h>
 #include <asm/io.h>
+#include <asm/hpet.h>
 
 DEFINE_SPINLOCK(i8253_lock);
 EXPORT_SYMBOL(i8253_lock);
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index 07b42153de24..85778a4b1209 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -215,6 +215,7 @@ static inline void clocksource_calculate_interval(struct clocksource *c,
 
 /* used to install a new clocksource */
 extern int clocksource_register(struct clocksource*);
+extern void clocksource_unregister(struct clocksource*);
 extern struct clocksource* clocksource_get_next(void);
 extern void clocksource_change_rating(struct clocksource *cs, int rating);
 extern void clocksource_resume(void);
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index edd5ef8e1765..6e9259a5d501 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -337,6 +337,21 @@ void clocksource_change_rating(struct clocksource *cs, int rating)
 	spin_unlock_irqrestore(&clocksource_lock, flags);
 }
 
+/**
+ * clocksource_unregister - remove a registered clocksource
+ */
+void clocksource_unregister(struct clocksource *cs)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clocksource_lock, flags);
+	list_del(&cs->list);
+	if (clocksource_override == cs)
+		clocksource_override = NULL;
+	next_clocksource = select_clocksource();
+	spin_unlock_irqrestore(&clocksource_lock, flags);
+}
+
 #ifdef CONFIG_SYSFS
 /**
  * sysfs_show_current_clocksources - sysfs interface for current clocksource
-- 
cgit v1.2.3


From e3f37a54f690d3e64995ea7ecea08c5ab3070faf Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@gmail.com>
Date: Wed, 30 Jan 2008 13:30:03 +0100
Subject: x86: assign IRQs to HPET timers

The userspace API for the HPET (see Documentation/hpet.txt) did not work. The
HPET_IE_ON ioctl was failing as there was no IRQ assigned to the timer
device. This patch fixes it by allocating IRQs to timer blocks in the HPET.

arch/x86/kernel/hpet.c |   13 +++++--------
drivers/char/hpet.c    |   45 ++++++++++++++++++++++++++++++++++++++-------
include/linux/hpet.h   |    2 +-
3 files changed, 44 insertions(+), 16 deletions(-)

Signed-off-by: Balaji Rao <balajirrao@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/hpet.c | 13 +++++--------
 drivers/char/hpet.c    | 45 ++++++++++++++++++++++++++++++++++++++-------
 include/linux/hpet.h   |  2 +-
 3 files changed, 44 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 9ec2ab793042..786aa227afdf 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -117,8 +117,7 @@ int is_hpet_enabled(void)
 static void hpet_reserve_platform_timers(unsigned long id)
 {
 	struct hpet __iomem *hpet = hpet_virt_address;
-	struct hpet_timer __iomem *timer = &hpet->hpet_timers[2];
-	unsigned int nrtimers, i;
+	unsigned int nrtimers;
 	struct hpet_data hd;
 
 	nrtimers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1;
@@ -133,16 +132,14 @@ static void hpet_reserve_platform_timers(unsigned long id)
 #ifdef CONFIG_HPET_EMULATE_RTC
 	hpet_reserve_timer(&hd, 1);
 #endif
-
 	hd.hd_irq[0] = HPET_LEGACY_8254;
 	hd.hd_irq[1] = HPET_LEGACY_RTC;
 
-	for (i = 2; i < nrtimers; timer++, i++)
-		hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >>
-			Tn_INT_ROUTE_CNF_SHIFT;
-
+	/*
+	 * IRQs for the other timers are assigned dynamically
+	 * in hpet_alloc
+	 */
 	hpet_alloc(&hd);
-
 }
 #else
 static void hpet_reserve_platform_timers(unsigned long id) { }
diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 4c16778e3f84..593b32cfbc33 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -806,14 +806,14 @@ static unsigned long hpet_calibrate(struct hpets *hpetp)
 
 int hpet_alloc(struct hpet_data *hdp)
 {
-	u64 cap, mcfg;
+	u64 cap, mcfg, hpet_config;
 	struct hpet_dev *devp;
-	u32 i, ntimer;
+	u32 i, ntimer, irq;
 	struct hpets *hpetp;
 	size_t siz;
 	struct hpet __iomem *hpet;
 	static struct hpets *last = NULL;
-	unsigned long period;
+	unsigned long period, irq_bitmap;
 	unsigned long long temp;
 
 	/*
@@ -840,11 +840,41 @@ int hpet_alloc(struct hpet_data *hdp)
 	hpetp->hp_hpet_phys = hdp->hd_phys_address;
 
 	hpetp->hp_ntimer = hdp->hd_nirqs;
+	hpet = hpetp->hp_hpet;
 
-	for (i = 0; i < hdp->hd_nirqs; i++)
-		hpetp->hp_dev[i].hd_hdwirq = hdp->hd_irq[i];
+	/* Assign IRQs statically for legacy devices */
+	hpetp->hp_dev[0].hd_hdwirq = hdp->hd_irq[0];
+	hpetp->hp_dev[1].hd_hdwirq = hdp->hd_irq[1];
 
-	hpet = hpetp->hp_hpet;
+	/* Assign IRQs dynamically for the others */
+	for (i = 2, devp = &hpetp->hp_dev[2]; i < hdp->hd_nirqs; i++, devp++) {
+		struct hpet_timer __iomem *timer;
+
+		timer = &hpet->hpet_timers[devp - hpetp->hp_dev];
+
+		hpet_config = readq(&timer->hpet_config);
+		irq_bitmap = (hpet_config & Tn_INT_ROUTE_CAP_MASK)
+			>> Tn_INT_ROUTE_CAP_SHIFT;
+		if (!irq_bitmap)
+			irq = 0;        /* No valid IRQ Assignable */
+		else {
+			irq = find_first_bit(&irq_bitmap, 32);
+			do {
+				hpet_config |= irq << Tn_INT_ROUTE_CNF_SHIFT;
+				writeq(hpet_config, &timer->hpet_config);
+
+				/*
+				 * Verify whether we have written a valid
+				 * IRQ number by reading it back again
+				 */
+				hpet_config = readq(&timer->hpet_config);
+				if (irq == (hpet_config & Tn_INT_ROUTE_CNF_MASK)
+						>> Tn_INT_ROUTE_CNF_SHIFT)
+					break;  /* Success */
+			} while ((irq = (find_next_bit(&irq_bitmap, 32, irq))));
+		}
+		hpetp->hp_dev[i].hd_hdwirq = irq;
+	}
 
 	cap = readq(&hpet->hpet_cap);
 
@@ -875,7 +905,8 @@ int hpet_alloc(struct hpet_data *hdp)
 		hpetp->hp_which, hdp->hd_phys_address,
 		hpetp->hp_ntimer > 1 ? "s" : "");
 	for (i = 0; i < hpetp->hp_ntimer; i++)
-		printk("%s %d", i > 0 ? "," : "", hdp->hd_irq[i]);
+		printk("%s %d", i > 0 ? "," : "",
+				hpetp->hp_dev[i].hd_hdwirq);
 	printk("\n");
 
 	printk(KERN_INFO "hpet%u: %u %d-bit timers, %Lu Hz\n",
diff --git a/include/linux/hpet.h b/include/linux/hpet.h
index 707f7cb9e795..e3c0b2aa944c 100644
--- a/include/linux/hpet.h
+++ b/include/linux/hpet.h
@@ -64,7 +64,7 @@ struct hpet {
  */
 
 #define	Tn_INT_ROUTE_CAP_MASK		(0xffffffff00000000ULL)
-#define	Tn_INI_ROUTE_CAP_SHIFT		(32UL)
+#define	Tn_INT_ROUTE_CAP_SHIFT		(32UL)
 #define	Tn_FSB_INT_DELCAP_MASK		(0x8000UL)
 #define	Tn_FSB_INT_DELCAP_SHIFT		(15)
 #define	Tn_FSB_EN_CNF_MASK		(0x4000UL)
-- 
cgit v1.2.3


From 6378ddb592158db4b42197f1bc8666228800e379 Mon Sep 17 00:00:00 2001
From: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Date: Wed, 30 Jan 2008 13:30:04 +0100
Subject: time: track accurate idle time with tick_sched.idle_sleeptime

Current idle time in kstat is based on jiffies and is coarse grained.
tick_sched.idle_sleeptime is making some attempt to keep track of idle time
in a fine grained manner.  But, it is not handling the time spent in
interrupts fully.

Make tick_sched.idle_sleeptime accurate with respect to time spent on
handling interrupts and also add tick_sched.idle_lastupdate, which keeps
track of last time when idle_sleeptime was updated.

This statistics will be crucial for cpufreq-ondemand governor, which can
shed some conservative gaurd band that is uses today while setting the
frequency.  The ondemand changes that uses the exact idle time is coming
soon.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/tick.h     |  6 +++++
 kernel/softirq.c         |  7 ++++-
 kernel/time/tick-sched.c | 70 +++++++++++++++++++++++++++++++++---------------
 3 files changed, 60 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/tick.h b/include/linux/tick.h
index f4a1395e05ff..0fadf95debe1 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -51,8 +51,10 @@ struct tick_sched {
 	unsigned long			idle_jiffies;
 	unsigned long			idle_calls;
 	unsigned long			idle_sleeps;
+	int				idle_active;
 	ktime_t				idle_entrytime;
 	ktime_t				idle_sleeptime;
+	ktime_t				idle_lastupdate;
 	ktime_t				sleep_length;
 	unsigned long			last_jiffies;
 	unsigned long			next_jiffies;
@@ -103,6 +105,8 @@ extern void tick_nohz_stop_sched_tick(void);
 extern void tick_nohz_restart_sched_tick(void);
 extern void tick_nohz_update_jiffies(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
+extern void tick_nohz_stop_idle(int cpu);
+extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
 # else
 static inline void tick_nohz_stop_sched_tick(void) { }
 static inline void tick_nohz_restart_sched_tick(void) { }
@@ -113,6 +117,8 @@ static inline ktime_t tick_nohz_get_sleep_length(void)
 
 	return len;
 }
+static inline void tick_nohz_stop_idle(int cpu) { }
+static inline u64 get_cpu_idle_time_us(int cpu, u64 *unused) { return 0; }
 # endif /* !NO_HZ */
 
 #endif
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 8fe1ff40102d..d7837d45419e 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -280,9 +280,14 @@ asmlinkage void do_softirq(void)
  */
 void irq_enter(void)
 {
+#ifdef CONFIG_NO_HZ
+	int cpu = smp_processor_id();
+	if (idle_cpu(cpu) && !in_interrupt())
+		tick_nohz_stop_idle(cpu);
+#endif
 	__irq_enter();
 #ifdef CONFIG_NO_HZ
-	if (idle_cpu(smp_processor_id()))
+	if (idle_cpu(cpu))
 		tick_nohz_update_jiffies();
 #endif
 }
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 49e12f6a4bab..63f24b550695 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -143,6 +143,44 @@ void tick_nohz_update_jiffies(void)
 	local_irq_restore(flags);
 }
 
+void tick_nohz_stop_idle(int cpu)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+
+	if (ts->idle_active) {
+		ktime_t now, delta;
+		now = ktime_get();
+		delta = ktime_sub(now, ts->idle_entrytime);
+		ts->idle_lastupdate = now;
+		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
+		ts->idle_active = 0;
+	}
+}
+
+static ktime_t tick_nohz_start_idle(int cpu)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+	ktime_t now, delta;
+
+	now = ktime_get();
+	if (ts->idle_active) {
+		delta = ktime_sub(now, ts->idle_entrytime);
+		ts->idle_lastupdate = now;
+		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
+	}
+	ts->idle_entrytime = now;
+	ts->idle_active = 1;
+	return now;
+}
+
+u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+
+	*last_update_time = ktime_to_us(ts->idle_lastupdate);
+	return ktime_to_us(ts->idle_sleeptime);
+}
+
 /**
  * tick_nohz_stop_sched_tick - stop the idle tick from the idle task
  *
@@ -155,13 +193,14 @@ void tick_nohz_stop_sched_tick(void)
 	unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
 	unsigned long rt_jiffies;
 	struct tick_sched *ts;
-	ktime_t last_update, expires, now, delta;
+	ktime_t last_update, expires, now;
 	struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
 	int cpu;
 
 	local_irq_save(flags);
 
 	cpu = smp_processor_id();
+	now = tick_nohz_start_idle(cpu);
 	ts = &per_cpu(tick_cpu_sched, cpu);
 
 	/*
@@ -193,19 +232,7 @@ void tick_nohz_stop_sched_tick(void)
 		}
 	}
 
-	now = ktime_get();
-	/*
-	 * When called from irq_exit we need to account the idle sleep time
-	 * correctly.
-	 */
-	if (ts->tick_stopped) {
-		delta = ktime_sub(now, ts->idle_entrytime);
-		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-	}
-
-	ts->idle_entrytime = now;
 	ts->idle_calls++;
-
 	/* Read jiffies and the time when jiffies were updated last */
 	do {
 		seq = read_seqbegin(&xtime_lock);
@@ -337,23 +364,22 @@ void tick_nohz_restart_sched_tick(void)
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 	unsigned long ticks;
-	ktime_t now, delta;
+	ktime_t now;
 
-	if (!ts->tick_stopped)
+	local_irq_disable();
+	tick_nohz_stop_idle(cpu);
+
+	if (!ts->tick_stopped) {
+		local_irq_enable();
 		return;
+	}
 
 	/* Update jiffies first */
-	now = ktime_get();
-
-	local_irq_disable();
 	select_nohz_load_balancer(0);
+	now = ktime_get();
 	tick_do_update_jiffies64(now);
 	cpu_clear(cpu, nohz_cpu_mask);
 
-	/* Account the idle time */
-	delta = ktime_sub(now, ts->idle_entrytime);
-	ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-
 	/*
 	 * We stopped the tick in idle. Update process times would miss the
 	 * time we slept as update_process_times does only a 1 tick
-- 
cgit v1.2.3


From 5c9c9bec0589be696c70c5efb448b17d5ab720e2 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 30 Jan 2008 13:30:04 +0100
Subject: x86: hibernation: document __save_processor_state() on x86

Document the fact that __save_processor_state() has to save all CPU
registers referred to by the kernel in case a different kernel is
used to load and restore a hibernation image containing it.

Sigend-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/suspend_64.c | 20 ++++++++++++++++++++
 include/asm-x86/suspend_64.h |  9 ++++++++-
 2 files changed, 28 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/suspend_64.c b/arch/x86/kernel/suspend_64.c
index 2e5efaaf8800..569f1b540e36 100644
--- a/arch/x86/kernel/suspend_64.c
+++ b/arch/x86/kernel/suspend_64.c
@@ -19,6 +19,21 @@ extern const void __nosave_begin, __nosave_end;
 
 struct saved_context saved_context;
 
+/**
+ *	__save_processor_state - save CPU registers before creating a
+ *		hibernation image and before restoring the memory state from it
+ *	@ctxt - structure to store the registers contents in
+ *
+ *	NOTE: If there is a CPU register the modification of which by the
+ *	boot kernel (ie. the kernel used for loading the hibernation image)
+ *	might affect the operations of the restored target kernel (ie. the one
+ *	saved in the hibernation image), then its contents must be saved by this
+ *	function.  In other words, if kernel A is hibernated and different
+ *	kernel B is used for loading the hibernation image into memory, the
+ *	kernel A's __save_processor_state() function must save all registers
+ *	needed by kernel A, so that it can operate correctly after the resume
+ *	regardless of what kernel B does in the meantime.
+ */
 void __save_processor_state(struct saved_context *ctxt)
 {
 	kernel_fpu_begin();
@@ -69,6 +84,11 @@ static void do_fpu_end(void)
 	kernel_fpu_end();
 }
 
+/**
+ *	__restore_processor_state - restore the contents of CPU registers saved
+ *		by __save_processor_state()
+ *	@ctxt - structure to load the registers contents from
+ */
 void __restore_processor_state(struct saved_context *ctxt)
 {
 	/*
diff --git a/include/asm-x86/suspend_64.h b/include/asm-x86/suspend_64.h
index c505a76bcf6e..4404668f9aa4 100644
--- a/include/asm-x86/suspend_64.h
+++ b/include/asm-x86/suspend_64.h
@@ -15,7 +15,14 @@ arch_prepare_suspend(void)
 	return 0;
 }
 
-/* Image of the saved processor state. If you touch this, fix acpi/wakeup.S. */
+/*
+ * Image of the saved processor state, used by the low level ACPI suspend to
+ * RAM code and by the low level hibernation code.
+ *
+ * If you modify it, fix arch/x86/kernel/acpi/wakeup_64.S and make sure that
+ * __save/__restore_processor_state(), defined in arch/x86/kernel/suspend_64.c,
+ * still work as required.
+ */
 struct saved_context {
 	struct pt_regs regs;
   	u16 ds, es, fs, gs, ss;
-- 
cgit v1.2.3


From b02aae9cf52956dfe1bec73f77f81a3d05d3902b Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@gmail.com>
Date: Wed, 30 Jan 2008 13:30:05 +0100
Subject: x86: provide a DMI based port 0x80 I/O delay override.

x86: provide a DMI based port 0x80 I/O delay override.

Certain (HP) laptops experience trouble from our port 0x80 I/O delay
writes. This patch provides for a DMI based switch to the "alternate
diagnostic port" 0xed (as used by some BIOSes as well) for these.

David P. Reed confirmed that port 0xed works for him and provides a
proper delay. The symptoms of _not_ working are a hanging machine,
with "hwclock" use being a direct trigger.

Earlier versions of this attempted to simply use udelay(2), with the
2 being a value tested to be a nicely conservative upper-bound with
help from many on the linux-kernel mailinglist but that approach has
two problems.

First, pre-loops_per_jiffy calibration (which is post PIT init while
some implementations of the PIT are actually one of the historically
problematic devices that need the delay) udelay() isn't particularly
well-defined. We could initialise loops_per_jiffy conservatively (and
based on CPU family so as to not unduly delay old machines) which
would sort of work, but...

Second, delaying isn't the only effect that a write to port 0x80 has.
It's also a PCI posting barrier which some devices may be explicitly
or implicitly relying on. Alan Cox did a survey and found evidence
that additionally some drivers may be racy on SMP without the bus
locking outb.

Switching to an inb() makes the timing too unpredictable and as such,
this DMI based switch should be the safest approach for now. Any more
invasive changes should get more rigid testing first. It's moreover
only very few machines with the problem and a DMI based hack seems
to fit that situation.

This also introduces a command-line parameter "io_delay" to override
the DMI based choice again:

	io_delay=<standard|alternate>

where "standard" means using the standard port 0x80 and "alternate"
port 0xed.

This retains the udelay method as a config (CONFIG_UDELAY_IO_DELAY) and
command-line ("io_delay=udelay") choice for testing purposes as well.

This does not change the io_delay() in the boot code which is using
the same port 0x80 I/O delay but those do not appear to be a problem
as David P. Reed reported the problem was already gone after using the
udelay version. He moreover reported that booting with "acpi=off" also
fixed things and seeing as how ACPI isn't touched until after this DMI
based I/O port switch I believe it's safe to leave the ones in the boot
code be.

The DMI strings from David's HP Pavilion dv9000z are in there already
and we need to get/verify the DMI info from other machines with the
problem, notably the HP Pavilion dv6000z.

This patch is partly based on earlier patches from Pavel Machek and
David P. Reed.

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |   8 +++
 arch/x86/Kconfig.debug              |   9 +++
 arch/x86/boot/compressed/misc_32.c  |   8 +--
 arch/x86/boot/compressed/misc_64.c  |   8 +--
 arch/x86/kernel/Makefile_32         |   2 +-
 arch/x86/kernel/Makefile_64         |   2 +-
 arch/x86/kernel/io_delay.c          | 106 ++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/setup_32.c          |   2 +
 arch/x86/kernel/setup_64.c          |   2 +
 include/asm-x86/io_32.h             |   8 ++-
 include/asm-x86/io_64.h             |  33 +++++++----
 11 files changed, 165 insertions(+), 23 deletions(-)
 create mode 100644 arch/x86/kernel/io_delay.c

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 880f882160e2..9e6056058425 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -794,6 +794,14 @@ and is between 256 and 4096 characters. It is defined in the file
 			for translation below 32 bit and if not available
 			then look in the higher range.
 
+	io_delay=	[X86-32,X86-64] I/O delay method
+		standard
+			Standard port 0x80 delay
+		alternate
+			Alternate port 0xed delay
+		udelay
+			Simple two microsecond delay
+
 	io7=		[HW] IO7 for Marvel based alpha systems
 			See comment before marvel_specify_io7 in
 			arch/alpha/kernel/core_marvel.c.
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 761ca7b5f120..40aba670fb37 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -112,4 +112,13 @@ config IOMMU_LEAK
 	  Add a simple leak tracer to the IOMMU code. This is useful when you
 	  are debugging a buggy device driver that leaks IOMMU mappings.
 
+config UDELAY_IO_DELAY
+	bool "Delay I/O through udelay instead of outb"
+	depends on DEBUG_KERNEL
+	help
+	  Make inb_p/outb_p use udelay() based delays by default. Please note
+	  that udelay() does not have the same bus-level side-effects that
+	  the normal outb based delay does meaning this could cause drivers
+	  to change behaviour and/or bugs to surface.
+
 endmenu
diff --git a/arch/x86/boot/compressed/misc_32.c b/arch/x86/boot/compressed/misc_32.c
index b74d60d1b2fa..288e16283ef9 100644
--- a/arch/x86/boot/compressed/misc_32.c
+++ b/arch/x86/boot/compressed/misc_32.c
@@ -276,10 +276,10 @@ static void putstr(const char *s)
 	RM_SCREEN_INFO.orig_y = y;
 
 	pos = (x + cols * y) * 2;	/* Update cursor position */
-	outb_p(14, vidport);
-	outb_p(0xff & (pos >> 9), vidport+1);
-	outb_p(15, vidport);
-	outb_p(0xff & (pos >> 1), vidport+1);
+	outb(14, vidport);
+	outb(0xff & (pos >> 9), vidport+1);
+	outb(15, vidport);
+	outb(0xff & (pos >> 1), vidport+1);
 }
 
 static void* memset(void* s, int c, unsigned n)
diff --git a/arch/x86/boot/compressed/misc_64.c b/arch/x86/boot/compressed/misc_64.c
index 6ea015aa65e4..43e5fcc37be9 100644
--- a/arch/x86/boot/compressed/misc_64.c
+++ b/arch/x86/boot/compressed/misc_64.c
@@ -269,10 +269,10 @@ static void putstr(const char *s)
 	RM_SCREEN_INFO.orig_y = y;
 
 	pos = (x + cols * y) * 2;	/* Update cursor position */
-	outb_p(14, vidport);
-	outb_p(0xff & (pos >> 9), vidport+1);
-	outb_p(15, vidport);
-	outb_p(0xff & (pos >> 1), vidport+1);
+	outb(14, vidport);
+	outb(0xff & (pos >> 9), vidport+1);
+	outb(15, vidport);
+	outb(0xff & (pos >> 1), vidport+1);
 }
 
 static void* memset(void* s, int c, unsigned n)
diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index a7bc93c27662..0cc1981d1e38 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -8,7 +8,7 @@ CPPFLAGS_vmlinux.lds += -Ui386
 obj-y	:= process_32.o signal_32.o entry_32.o traps_32.o irq_32.o \
 		ptrace_32.o time_32.o ioport_32.o ldt_32.o setup_32.o i8259_32.o sys_i386_32.o \
 		pci-dma_32.o i386_ksyms_32.o i387_32.o bootflag.o e820_32.o\
-		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o
+		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index 5a88890d8ee9..08a68f0d8fda 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -11,7 +11,7 @@ obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
 		x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \
 		setup64.o bootflag.o e820_64.o reboot_64.o quirks.o i8237.o \
 		pci-dma_64.o pci-nommu_64.o alternative.o hpet.o tsc_64.o bugs_64.o \
-		i8253.o
+		i8253.o io_delay.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c
new file mode 100644
index 000000000000..4d955e74b974
--- /dev/null
+++ b/arch/x86/kernel/io_delay.c
@@ -0,0 +1,106 @@
+/*
+ * I/O delay strategies for inb_p/outb_p
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/dmi.h>
+#include <asm/io.h>
+
+/*
+ * Allow for a DMI based override of port 0x80 needed for certain HP laptops
+ */
+#define IO_DELAY_PORT_STD 0x80
+#define IO_DELAY_PORT_ALT 0xed
+
+static void standard_io_delay(void)
+{
+	asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_STD));
+}
+
+static void alternate_io_delay(void)
+{
+	asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_ALT));
+}
+
+/*
+ * 2 usecs is an upper-bound for the outb delay but note that udelay doesn't
+ * have the bus-level side-effects that outb does
+ */
+#define IO_DELAY_USECS 2
+
+/*
+ * High on a hill was a lonely goatherd
+ */
+static void udelay_io_delay(void)
+{
+	udelay(IO_DELAY_USECS);
+}
+
+#ifndef CONFIG_UDELAY_IO_DELAY
+static void (*io_delay)(void) = standard_io_delay;
+#else
+static void (*io_delay)(void) = udelay_io_delay;
+#endif
+
+/*
+ * Paravirt wants native_io_delay to be a constant.
+ */
+void native_io_delay(void)
+{
+	io_delay();
+}
+EXPORT_SYMBOL(native_io_delay);
+
+#ifndef CONFIG_UDELAY_IO_DELAY
+static int __init dmi_alternate_io_delay_port(const struct dmi_system_id *id)
+{
+	printk(KERN_NOTICE "%s: using alternate I/O delay port\n", id->ident);
+	io_delay = alternate_io_delay;
+	return 0;
+}
+
+static struct dmi_system_id __initdata alternate_io_delay_port_dmi_table[] = {
+	{
+		.callback	= dmi_alternate_io_delay_port,
+		.ident		= "HP Pavilion dv9000z",
+		.matches	= {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_NAME, "30B9")
+		}
+	},
+	{
+	}
+};
+
+static int __initdata io_delay_override;
+
+void __init io_delay_init(void)
+{
+	if (!io_delay_override)
+		dmi_check_system(alternate_io_delay_port_dmi_table);
+}
+#endif
+
+static int __init io_delay_param(char *s)
+{
+	if (!s)
+		return -EINVAL;
+
+	if (!strcmp(s, "standard"))
+		io_delay = standard_io_delay;
+	else if (!strcmp(s, "alternate"))
+		io_delay = alternate_io_delay;
+	else if (!strcmp(s, "udelay"))
+		io_delay = udelay_io_delay;
+	else
+		return -EINVAL;
+
+#ifndef CONFIG_UDELAY_IO_DELAY
+	io_delay_override = 1;
+#endif
+	return 0;
+}
+
+early_param("io_delay", io_delay_param);
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 9c24b45b513c..51bdc0b1b72e 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -648,6 +648,8 @@ void __init setup_arch(char **cmdline_p)
 
 	dmi_scan_machine();
 
+	io_delay_init();;
+
 #ifdef CONFIG_X86_GENERICARCH
 	generic_apic_probe();
 #endif	
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 30d94d1d5f5f..ec976edf0399 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -311,6 +311,8 @@ void __init setup_arch(char **cmdline_p)
 
 	dmi_scan_machine();
 
+	io_delay_init();
+
 #ifdef CONFIG_SMP
 	/* setup to use the static apicid table during kernel startup */
 	x86_cpu_to_apicid_ptr = (void *)&x86_cpu_to_apicid_init;
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index fe881cd1e6f4..a8d25c38b91c 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -250,10 +250,14 @@ static inline void flush_write_buffers(void)
 
 #endif /* __KERNEL__ */
 
-static inline void native_io_delay(void)
+#ifndef CONFIG_UDELAY_IO_DELAY
+extern void io_delay_init(void);
+#else
+static inline void io_delay_init(void)
 {
-	asm volatile("outb %%al,$0x80" : : : "memory");
 }
+#endif
+extern void native_io_delay(void);
 
 #if defined(CONFIG_PARAVIRT)
 #include <asm/paravirt.h>
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index a037b0794332..5bebaf961692 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -35,13 +35,24 @@
   *  - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
   */
 
-#define __SLOW_DOWN_IO "\noutb %%al,$0x80"
+#ifndef CONFIG_UDELAY_IO_DELAY
+extern void io_delay_init(void);
+#else
+static inline void io_delay_init(void)
+{
+}
+#endif
+extern void native_io_delay(void);
 
+static inline void slow_down_io(void)
+{
+	native_io_delay();
 #ifdef REALLY_SLOW_IO
-#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO
-#else
-#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO
+	native_io_delay();
+	native_io_delay();
+	native_io_delay();
 #endif
+}
 
 /*
  * Talk about misusing macros..
@@ -50,21 +61,21 @@
 static inline void out##s(unsigned x value, unsigned short port) {
 
 #define __OUT2(s,s1,s2) \
-__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1"
+__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1" : : "a" (value), "Nd" (port))
 
 #define __OUT(s,s1,x) \
-__OUT1(s,x) __OUT2(s,s1,"w") : : "a" (value), "Nd" (port)); } \
-__OUT1(s##_p,x) __OUT2(s,s1,"w") __FULL_SLOW_DOWN_IO : : "a" (value), "Nd" (port));} \
+__OUT1(s,x) __OUT2(s,s1,"w"); } \
+__OUT1(s##_p,x) __OUT2(s,s1,"w"); slow_down_io(); }
 
 #define __IN1(s) \
 static inline RETURN_TYPE in##s(unsigned short port) { RETURN_TYPE _v;
 
 #define __IN2(s,s1,s2) \
-__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0"
+__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0" : "=a" (_v) : "Nd" (port))
 
-#define __IN(s,s1,i...) \
-__IN1(s) __IN2(s,s1,"w") : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \
-__IN1(s##_p) __IN2(s,s1,"w") __FULL_SLOW_DOWN_IO : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \
+#define __IN(s,s1) \
+__IN1(s) __IN2(s,s1,"w"); return _v; } \
+__IN1(s##_p) __IN2(s,s1,"w"); slow_down_io(); return _v; }
 
 #define __INS(s) \
 static inline void ins##s(unsigned short port, void * addr, unsigned long count) \
-- 
cgit v1.2.3


From 6e7c402590b75b6b45138792445ee0f0315a8473 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:30:05 +0100
Subject: x86: various changes and cleanups to in_p/out_p delay details

various changes to the in_p/out_p delay details:

- add the io_delay=none method
- make each method selectable from the kernel config
- simplify the delay code a bit by getting rid of an indirect function call
- add the /proc/sys/kernel/io_delay_type sysctl
- change 'io_delay=standard|alternate' to io_delay=0x80 and io_delay=0xed
- make the io delay config not depend on CONFIG_DEBUG_KERNEL

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: "David P. Reed" <dpreed@reed.com>
---
 Documentation/kernel-parameters.txt |  12 ++--
 arch/x86/Kconfig.debug              |  79 ++++++++++++++++++++++++---
 arch/x86/kernel/io_delay.c          | 106 +++++++++++++++++-------------------
 include/asm-x86/io_32.h             |  10 +---
 include/asm-x86/io_64.h             |  10 +---
 kernel/sysctl.c                     |   9 +++
 6 files changed, 143 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 9e6056058425..b427b7c0e5d0 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -795,12 +795,14 @@ and is between 256 and 4096 characters. It is defined in the file
 			then look in the higher range.
 
 	io_delay=	[X86-32,X86-64] I/O delay method
-		standard
-			Standard port 0x80 delay
-		alternate
-			Alternate port 0xed delay
+		0x80
+			Standard port 0x80 based delay
+		0xed
+			Alternate port 0xed based delay (needed on some systems)
 		udelay
-			Simple two microsecond delay
+			Simple two microseconds delay
+		none
+			No delay
 
 	io7=		[HW] IO7 for Marvel based alpha systems
 			See comment before marvel_specify_io7 in
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 40aba670fb37..77eda46f97b8 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -112,13 +112,78 @@ config IOMMU_LEAK
 	  Add a simple leak tracer to the IOMMU code. This is useful when you
 	  are debugging a buggy device driver that leaks IOMMU mappings.
 
-config UDELAY_IO_DELAY
-	bool "Delay I/O through udelay instead of outb"
-	depends on DEBUG_KERNEL
+#
+# IO delay types:
+#
+
+config IO_DELAY_TYPE_0X80
+	int
+	default "0"
+
+config IO_DELAY_TYPE_0XED
+	int
+	default "1"
+
+config IO_DELAY_TYPE_UDELAY
+	int
+	default "2"
+
+config IO_DELAY_TYPE_NONE
+	int
+	default "3"
+
+choice
+	prompt "IO delay type"
+	default IO_DELAY_0X80
+
+config IO_DELAY_0X80
+	bool "port 0x80 based port-IO delay [recommended]"
+	help
+	  This is the traditional Linux IO delay used for in/out_p.
+	  It is the most tested hence safest selection here.
+
+config IO_DELAY_0XED
+	bool "port 0xed based port-IO delay"
+	help
+	  Use port 0xed as the IO delay. This frees up port 0x80 which is
+	  often used as a hardware-debug port.
+
+config IO_DELAY_UDELAY
+	bool "udelay based port-IO delay"
+	help
+	  Use udelay(2) as the IO delay method. This provides the delay
+	  while not having any side-effect on the IO port space.
+
+config IO_DELAY_NONE
+	bool "no port-IO delay"
 	help
-	  Make inb_p/outb_p use udelay() based delays by default. Please note
-	  that udelay() does not have the same bus-level side-effects that
-	  the normal outb based delay does meaning this could cause drivers
-	  to change behaviour and/or bugs to surface.
+	  No port-IO delay. Will break on old boxes that require port-IO
+	  delay for certain operations. Should work on most new machines.
+
+endchoice
+
+if IO_DELAY_0X80
+config DEFAULT_IO_DELAY_TYPE
+	int
+	default IO_DELAY_TYPE_0X80
+endif
+
+if IO_DELAY_0XED
+config DEFAULT_IO_DELAY_TYPE
+	int
+	default IO_DELAY_TYPE_0XED
+endif
+
+if IO_DELAY_UDELAY
+config DEFAULT_IO_DELAY_TYPE
+	int
+	default IO_DELAY_TYPE_UDELAY
+endif
+
+if IO_DELAY_NONE
+config DEFAULT_IO_DELAY_TYPE
+	int
+	default IO_DELAY_TYPE_NONE
+endif
 
 endmenu
diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c
index 4d955e74b974..f052e34dc94c 100644
--- a/arch/x86/kernel/io_delay.c
+++ b/arch/x86/kernel/io_delay.c
@@ -1,5 +1,9 @@
 /*
  * I/O delay strategies for inb_p/outb_p
+ *
+ * Allow for a DMI based override of port 0x80, needed for certain HP laptops
+ * and possibly other systems. Also allow for the gradual elimination of
+ * outb_p/inb_p API uses.
  */
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -8,98 +12,86 @@
 #include <linux/dmi.h>
 #include <asm/io.h>
 
-/*
- * Allow for a DMI based override of port 0x80 needed for certain HP laptops
- */
-#define IO_DELAY_PORT_STD 0x80
-#define IO_DELAY_PORT_ALT 0xed
-
-static void standard_io_delay(void)
-{
-	asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_STD));
-}
-
-static void alternate_io_delay(void)
-{
-	asm volatile ("outb %%al, %0" : : "N" (IO_DELAY_PORT_ALT));
-}
-
-/*
- * 2 usecs is an upper-bound for the outb delay but note that udelay doesn't
- * have the bus-level side-effects that outb does
- */
-#define IO_DELAY_USECS 2
-
-/*
- * High on a hill was a lonely goatherd
- */
-static void udelay_io_delay(void)
-{
-	udelay(IO_DELAY_USECS);
-}
+int io_delay_type __read_mostly = CONFIG_DEFAULT_IO_DELAY_TYPE;
+EXPORT_SYMBOL_GPL(io_delay_type);
 
-#ifndef CONFIG_UDELAY_IO_DELAY
-static void (*io_delay)(void) = standard_io_delay;
-#else
-static void (*io_delay)(void) = udelay_io_delay;
-#endif
+static int __initdata io_delay_override;
 
 /*
  * Paravirt wants native_io_delay to be a constant.
  */
 void native_io_delay(void)
 {
-	io_delay();
+	switch (io_delay_type) {
+	default:
+	case CONFIG_IO_DELAY_TYPE_0X80:
+		asm volatile ("outb %al, $0x80");
+		break;
+	case CONFIG_IO_DELAY_TYPE_0XED:
+		asm volatile ("outb %al, $0xed");
+		break;
+	case CONFIG_IO_DELAY_TYPE_UDELAY:
+		/*
+		 * 2 usecs is an upper-bound for the outb delay but
+		 * note that udelay doesn't have the bus-level
+		 * side-effects that outb does, nor does udelay() have
+		 * precise timings during very early bootup (the delays
+		 * are shorter until calibrated):
+		 */
+		udelay(2);
+	case CONFIG_IO_DELAY_TYPE_NONE:
+		break;
+	}
 }
 EXPORT_SYMBOL(native_io_delay);
 
-#ifndef CONFIG_UDELAY_IO_DELAY
-static int __init dmi_alternate_io_delay_port(const struct dmi_system_id *id)
+static int __init dmi_io_delay_0xed_port(const struct dmi_system_id *id)
 {
-	printk(KERN_NOTICE "%s: using alternate I/O delay port\n", id->ident);
-	io_delay = alternate_io_delay;
+	if (io_delay_type == CONFIG_IO_DELAY_TYPE_0X80) {
+		printk(KERN_NOTICE "%s: using 0xed I/O delay port\n",
+			id->ident);
+		io_delay_type = CONFIG_IO_DELAY_TYPE_0XED;
+	}
+
 	return 0;
 }
 
-static struct dmi_system_id __initdata alternate_io_delay_port_dmi_table[] = {
+/*
+ * Quirk table for systems that misbehave (lock up, etc.) if port
+ * 0x80 is used:
+ */
+static struct dmi_system_id __initdata io_delay_0xed_port_dmi_table[] = {
 	{
-		.callback	= dmi_alternate_io_delay_port,
+		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "HP Pavilion dv9000z",
 		.matches	= {
 			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
 			DMI_MATCH(DMI_BOARD_NAME, "30B9")
 		}
 	},
-	{
-	}
+	{ }
 };
 
-static int __initdata io_delay_override;
-
 void __init io_delay_init(void)
 {
 	if (!io_delay_override)
-		dmi_check_system(alternate_io_delay_port_dmi_table);
+		dmi_check_system(io_delay_0xed_port_dmi_table);
 }
-#endif
 
 static int __init io_delay_param(char *s)
 {
-	if (!s)
-		return -EINVAL;
-
-	if (!strcmp(s, "standard"))
-		io_delay = standard_io_delay;
-	else if (!strcmp(s, "alternate"))
-		io_delay = alternate_io_delay;
+	if (!strcmp(s, "0x80"))
+		io_delay_type = CONFIG_IO_DELAY_TYPE_0X80;
+	else if (!strcmp(s, "0xed"))
+		io_delay_type = CONFIG_IO_DELAY_TYPE_0XED;
 	else if (!strcmp(s, "udelay"))
-		io_delay = udelay_io_delay;
+		io_delay_type = CONFIG_IO_DELAY_TYPE_UDELAY;
+	else if (!strcmp(s, "none"))
+		io_delay_type = CONFIG_IO_DELAY_TYPE_NONE;
 	else
 		return -EINVAL;
 
-#ifndef CONFIG_UDELAY_IO_DELAY
 	io_delay_override = 1;
-#endif
 	return 0;
 }
 
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index a8d25c38b91c..2a04bd17eac5 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -250,15 +250,11 @@ static inline void flush_write_buffers(void)
 
 #endif /* __KERNEL__ */
 
-#ifndef CONFIG_UDELAY_IO_DELAY
-extern void io_delay_init(void);
-#else
-static inline void io_delay_init(void)
-{
-}
-#endif
 extern void native_io_delay(void);
 
+extern int io_delay_type;
+extern void io_delay_init(void);
+
 #if defined(CONFIG_PARAVIRT)
 #include <asm/paravirt.h>
 #else
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index 5bebaf961692..dbcc03aa1c6a 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -35,15 +35,11 @@
   *  - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
   */
 
-#ifndef CONFIG_UDELAY_IO_DELAY
-extern void io_delay_init(void);
-#else
-static inline void io_delay_init(void)
-{
-}
-#endif
 extern void native_io_delay(void);
 
+extern int io_delay_type;
+extern void io_delay_init(void);
+
 static inline void slow_down_io(void)
 {
 	native_io_delay();
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 4bc8e48434a7..357b68ba23ec 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -53,6 +53,7 @@
 #ifdef CONFIG_X86
 #include <asm/nmi.h>
 #include <asm/stacktrace.h>
+#include <asm/io.h>
 #endif
 
 static int deprecated_sysctl_warning(struct __sysctl_args *args);
@@ -727,6 +728,14 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "io_delay_type",
+		.data		= &io_delay_type,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 #endif
 #if defined(CONFIG_MMU)
 	{
-- 
cgit v1.2.3


From 53d517cdbaac704352b3d0c10fecb99e0b54572e Mon Sep 17 00:00:00 2001
From: Guillaume Chazarain <guichaz@yahoo.fr>
Date: Wed, 30 Jan 2008 13:30:06 +0100
Subject: x86: scale cyc_2_nsec according to CPU frequency

scale the sched_clock() cyc_2_nsec scaling factor according to
CPU frequency changes.

[ mingo@elte.hu: simplified it and fixed it for SMP. ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/tsc_32.c | 43 +++++++++++++++++++++++++++++++-----
 arch/x86/kernel/tsc_64.c | 57 +++++++++++++++++++++++++++++++++++++++---------
 include/asm-x86/timer.h  | 23 ++++++++++++++-----
 3 files changed, 102 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index 9ebc0dab66b4..00bb4c1c0593 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -5,6 +5,7 @@
 #include <linux/jiffies.h>
 #include <linux/init.h>
 #include <linux/dmi.h>
+#include <linux/percpu.h>
 
 #include <asm/delay.h>
 #include <asm/tsc.h>
@@ -80,13 +81,31 @@ EXPORT_SYMBOL_GPL(check_tsc_unstable);
  *
  *			-johnstul@us.ibm.com "math is hard, lets go shopping!"
  */
-unsigned long cyc2ns_scale __read_mostly;
 
-#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */
+DEFINE_PER_CPU(unsigned long, cyc2ns);
 
-static inline void set_cyc2ns_scale(unsigned long cpu_khz)
+static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu)
 {
-	cyc2ns_scale = (1000000 << CYC2NS_SCALE_FACTOR)/cpu_khz;
+	unsigned long flags, prev_scale, *scale;
+	unsigned long long tsc_now, ns_now;
+
+	local_irq_save(flags);
+	sched_clock_idle_sleep_event();
+
+	scale = &per_cpu(cyc2ns, cpu);
+
+	rdtscll(tsc_now);
+	ns_now = __cycles_2_ns(tsc_now);
+
+	prev_scale = *scale;
+	if (cpu_khz)
+		*scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz;
+
+	/*
+	 * Start smoothly with the new frequency:
+	 */
+	sched_clock_idle_wakeup_event(0);
+	local_irq_restore(flags);
 }
 
 /*
@@ -239,7 +258,9 @@ time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data)
 						ref_freq, freq->new);
 			if (!(freq->flags & CPUFREQ_CONST_LOOPS)) {
 				tsc_khz = cpu_khz;
-				set_cyc2ns_scale(cpu_khz);
+				preempt_disable();
+				set_cyc2ns_scale(cpu_khz, smp_processor_id());
+				preempt_enable();
 				/*
 				 * TSC based sched_clock turns
 				 * to junk w/ cpufreq
@@ -367,6 +388,8 @@ static inline void check_geode_tsc_reliable(void) { }
 
 void __init tsc_init(void)
 {
+	int cpu;
+
 	if (!cpu_has_tsc || tsc_disable)
 		goto out_no_tsc;
 
@@ -380,7 +403,15 @@ void __init tsc_init(void)
 				(unsigned long)cpu_khz / 1000,
 				(unsigned long)cpu_khz % 1000);
 
-	set_cyc2ns_scale(cpu_khz);
+	/*
+	 * Secondary CPUs do not run through tsc_init(), so set up
+	 * all the scale factors for all CPUs, assuming the same
+	 * speed as the bootup CPU. (cpufreq notifiers will fix this
+	 * up if their speed diverges)
+	 */
+	for_each_possible_cpu(cpu)
+		set_cyc2ns_scale(cpu_khz, cpu);
+
 	use_tsc_delay();
 
 	/* Check and install the TSC clocksource */
diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c
index 9c70af45b42b..32edd2c50e94 100644
--- a/arch/x86/kernel/tsc_64.c
+++ b/arch/x86/kernel/tsc_64.c
@@ -10,6 +10,7 @@
 
 #include <asm/hpet.h>
 #include <asm/timex.h>
+#include <asm/timer.h>
 
 static int notsc __initdata = 0;
 
@@ -18,16 +19,48 @@ EXPORT_SYMBOL(cpu_khz);
 unsigned int tsc_khz;
 EXPORT_SYMBOL(tsc_khz);
 
-static unsigned int cyc2ns_scale __read_mostly;
+/* Accelerators for sched_clock()
+ * convert from cycles(64bits) => nanoseconds (64bits)
+ *  basic equation:
+ *		ns = cycles / (freq / ns_per_sec)
+ *		ns = cycles * (ns_per_sec / freq)
+ *		ns = cycles * (10^9 / (cpu_khz * 10^3))
+ *		ns = cycles * (10^6 / cpu_khz)
+ *
+ *	Then we use scaling math (suggested by george@mvista.com) to get:
+ *		ns = cycles * (10^6 * SC / cpu_khz) / SC
+ *		ns = cycles * cyc2ns_scale / SC
+ *
+ *	And since SC is a constant power of two, we can convert the div
+ *  into a shift.
+ *
+ *  We can use khz divisor instead of mhz to keep a better precision, since
+ *  cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits.
+ *  (mathieu.desnoyers@polymtl.ca)
+ *
+ *			-johnstul@us.ibm.com "math is hard, lets go shopping!"
+ */
+DEFINE_PER_CPU(unsigned long, cyc2ns);
 
-static inline void set_cyc2ns_scale(unsigned long khz)
+static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu)
 {
-	cyc2ns_scale = (NSEC_PER_MSEC << NS_SCALE) / khz;
-}
+	unsigned long flags, prev_scale, *scale;
+	unsigned long long tsc_now, ns_now;
 
-static unsigned long long cycles_2_ns(unsigned long long cyc)
-{
-	return (cyc * cyc2ns_scale) >> NS_SCALE;
+	local_irq_save(flags);
+	sched_clock_idle_sleep_event();
+
+	scale = &per_cpu(cyc2ns, cpu);
+
+	rdtscll(tsc_now);
+	ns_now = __cycles_2_ns(tsc_now);
+
+	prev_scale = *scale;
+	if (cpu_khz)
+		*scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz;
+
+	sched_clock_idle_wakeup_event(0);
+	local_irq_restore(flags);
 }
 
 unsigned long long sched_clock(void)
@@ -100,7 +133,9 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
 			mark_tsc_unstable("cpufreq changes");
 	}
 
-	set_cyc2ns_scale(tsc_khz_ref);
+	preempt_disable();
+	set_cyc2ns_scale(tsc_khz_ref, smp_processor_id());
+	preempt_enable();
 
 	return 0;
 }
@@ -151,7 +186,7 @@ static unsigned long __init tsc_read_refs(unsigned long *pm,
 void __init tsc_calibrate(void)
 {
 	unsigned long flags, tsc1, tsc2, tr1, tr2, pm1, pm2, hpet1, hpet2;
-	int hpet = is_hpet_enabled();
+	int hpet = is_hpet_enabled(), cpu;
 
 	local_irq_save(flags);
 
@@ -206,7 +241,9 @@ void __init tsc_calibrate(void)
 	}
 
 	tsc_khz = tsc2 / tsc1;
-	set_cyc2ns_scale(tsc_khz);
+
+	for_each_possible_cpu(cpu)
+		set_cyc2ns_scale(tsc_khz, cpu);
 }
 
 /*
diff --git a/include/asm-x86/timer.h b/include/asm-x86/timer.h
index 0db7e994fb8b..4f6fcb050c11 100644
--- a/include/asm-x86/timer.h
+++ b/include/asm-x86/timer.h
@@ -2,6 +2,7 @@
 #define _ASMi386_TIMER_H
 #include <linux/init.h>
 #include <linux/pm.h>
+#include <linux/percpu.h>
 
 #define TICK_SIZE (tick_nsec / 1000)
 
@@ -16,7 +17,7 @@ extern int recalibrate_cpu_khz(void);
 #define calculate_cpu_khz() native_calculate_cpu_khz()
 #endif
 
-/* Accellerators for sched_clock()
+/* Accelerators for sched_clock()
  * convert from cycles(64bits) => nanoseconds (64bits)
  *  basic equation:
  *		ns = cycles / (freq / ns_per_sec)
@@ -31,20 +32,32 @@ extern int recalibrate_cpu_khz(void);
  *	And since SC is a constant power of two, we can convert the div
  *  into a shift.
  *
- *  We can use khz divisor instead of mhz to keep a better percision, since
+ *  We can use khz divisor instead of mhz to keep a better precision, since
  *  cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits.
  *  (mathieu.desnoyers@polymtl.ca)
  *
  *			-johnstul@us.ibm.com "math is hard, lets go shopping!"
  */
-extern unsigned long cyc2ns_scale __read_mostly;
+
+DECLARE_PER_CPU(unsigned long, cyc2ns);
 
 #define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */
 
-static inline unsigned long long cycles_2_ns(unsigned long long cyc)
+static inline unsigned long long __cycles_2_ns(unsigned long long cyc)
 {
-	return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR;
+	return cyc * per_cpu(cyc2ns, smp_processor_id()) >> CYC2NS_SCALE_FACTOR;
 }
 
+static inline unsigned long long cycles_2_ns(unsigned long long cyc)
+{
+	unsigned long long ns;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	ns = __cycles_2_ns(cyc);
+	local_irq_restore(flags);
+
+	return ns;
+}
 
 #endif
-- 
cgit v1.2.3


From e845c06bced7f5f325e0f9aefd3207cd45c21ff2 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:06 +0100
Subject: x86: add <asm/asm.h>

Create <asm/asm.h>, with common definitions suitable for assembly
unification.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/asm.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
 create mode 100644 include/asm-x86/asm.h

(limited to 'include')

diff --git a/include/asm-x86/asm.h b/include/asm-x86/asm.h
new file mode 100644
index 000000000000..b5006eb9acba
--- /dev/null
+++ b/include/asm-x86/asm.h
@@ -0,0 +1,18 @@
+#ifndef _ASM_X86_ASM_H
+#define _ASM_X86_ASM_H
+
+#ifdef CONFIG_X86_32
+/* 32 bits */
+
+# define _ASM_PTR	" .long "
+# define _ASM_ALIGN	" .balign 4 "
+
+#else
+/* 64 bits */
+
+# define _ASM_PTR	" .quad "
+# define _ASM_ALIGN	" .balign 8 "
+
+#endif /* CONFIG_X86_32 */
+
+#endif /* _ASM_X86_ASM_H */
-- 
cgit v1.2.3


From 7b11fb51567dedeaf6dc03f0135c0a8bb2399818 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:07 +0100
Subject: x86: unify asm/cpufeature.h

asm/cpufeature.h was already almost unified; this completes the job.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/cpufeature.h    | 191 +++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/cpufeature_32.h | 176 ------------------------------------
 include/asm-x86/cpufeature_64.h |  30 -------
 3 files changed, 187 insertions(+), 210 deletions(-)
 delete mode 100644 include/asm-x86/cpufeature_32.h
 delete mode 100644 include/asm-x86/cpufeature_64.h

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index b7160a4598d7..4c7875554d01 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -1,5 +1,188 @@
-#ifdef CONFIG_X86_32
-# include "cpufeature_32.h"
-#else
-# include "cpufeature_64.h"
+/*
+ * Defines x86 CPU feature bits
+ */
+#ifndef _ASM_X86_CPUFEATURE_H
+#define _ASM_X86_CPUFEATURE_H
+
+#ifndef __ASSEMBLY__
+#include <linux/bitops.h>
 #endif
+#include <asm/required-features.h>
+
+#define NCAPINTS	8	/* N 32-bit words worth of info */
+
+/* Intel-defined CPU features, CPUID level 0x00000001 (edx), word 0 */
+#define X86_FEATURE_FPU		(0*32+ 0) /* Onboard FPU */
+#define X86_FEATURE_VME		(0*32+ 1) /* Virtual Mode Extensions */
+#define X86_FEATURE_DE		(0*32+ 2) /* Debugging Extensions */
+#define X86_FEATURE_PSE		(0*32+ 3) /* Page Size Extensions */
+#define X86_FEATURE_TSC		(0*32+ 4) /* Time Stamp Counter */
+#define X86_FEATURE_MSR		(0*32+ 5) /* Model-Specific Registers, RDMSR, WRMSR */
+#define X86_FEATURE_PAE		(0*32+ 6) /* Physical Address Extensions */
+#define X86_FEATURE_MCE		(0*32+ 7) /* Machine Check Architecture */
+#define X86_FEATURE_CX8		(0*32+ 8) /* CMPXCHG8 instruction */
+#define X86_FEATURE_APIC	(0*32+ 9) /* Onboard APIC */
+#define X86_FEATURE_SEP		(0*32+11) /* SYSENTER/SYSEXIT */
+#define X86_FEATURE_MTRR	(0*32+12) /* Memory Type Range Registers */
+#define X86_FEATURE_PGE		(0*32+13) /* Page Global Enable */
+#define X86_FEATURE_MCA		(0*32+14) /* Machine Check Architecture */
+#define X86_FEATURE_CMOV	(0*32+15) /* CMOV instruction (FCMOVCC and FCOMI too if FPU present) */
+#define X86_FEATURE_PAT		(0*32+16) /* Page Attribute Table */
+#define X86_FEATURE_PSE36	(0*32+17) /* 36-bit PSEs */
+#define X86_FEATURE_PN		(0*32+18) /* Processor serial number */
+#define X86_FEATURE_CLFLSH	(0*32+19) /* Supports the CLFLUSH instruction */
+#define X86_FEATURE_DS		(0*32+21) /* Debug Store */
+#define X86_FEATURE_ACPI	(0*32+22) /* ACPI via MSR */
+#define X86_FEATURE_MMX		(0*32+23) /* Multimedia Extensions */
+#define X86_FEATURE_FXSR	(0*32+24) /* FXSAVE and FXRSTOR instructions (fast save and restore */
+					  /* of FPU context), and CR4.OSFXSR available */
+#define X86_FEATURE_XMM		(0*32+25) /* Streaming SIMD Extensions */
+#define X86_FEATURE_XMM2	(0*32+26) /* Streaming SIMD Extensions-2 */
+#define X86_FEATURE_SELFSNOOP	(0*32+27) /* CPU self snoop */
+#define X86_FEATURE_HT		(0*32+28) /* Hyper-Threading */
+#define X86_FEATURE_ACC		(0*32+29) /* Automatic clock control */
+#define X86_FEATURE_IA64	(0*32+30) /* IA-64 processor */
+
+/* AMD-defined CPU features, CPUID level 0x80000001, word 1 */
+/* Don't duplicate feature flags which are redundant with Intel! */
+#define X86_FEATURE_SYSCALL	(1*32+11) /* SYSCALL/SYSRET */
+#define X86_FEATURE_MP		(1*32+19) /* MP Capable. */
+#define X86_FEATURE_NX		(1*32+20) /* Execute Disable */
+#define X86_FEATURE_MMXEXT	(1*32+22) /* AMD MMX extensions */
+#define X86_FEATURE_RDTSCP	(1*32+27) /* RDTSCP */
+#define X86_FEATURE_LM		(1*32+29) /* Long Mode (x86-64) */
+#define X86_FEATURE_3DNOWEXT	(1*32+30) /* AMD 3DNow! extensions */
+#define X86_FEATURE_3DNOW	(1*32+31) /* 3DNow! */
+
+/* Transmeta-defined CPU features, CPUID level 0x80860001, word 2 */
+#define X86_FEATURE_RECOVERY	(2*32+ 0) /* CPU in recovery mode */
+#define X86_FEATURE_LONGRUN	(2*32+ 1) /* Longrun power control */
+#define X86_FEATURE_LRTI	(2*32+ 3) /* LongRun table interface */
+
+/* Other features, Linux-defined mapping, word 3 */
+/* This range is used for feature bits which conflict or are synthesized */
+#define X86_FEATURE_CXMMX	(3*32+ 0) /* Cyrix MMX extensions */
+#define X86_FEATURE_K6_MTRR	(3*32+ 1) /* AMD K6 nonstandard MTRRs */
+#define X86_FEATURE_CYRIX_ARR	(3*32+ 2) /* Cyrix ARRs (= MTRRs) */
+#define X86_FEATURE_CENTAUR_MCR	(3*32+ 3) /* Centaur MCRs (= MTRRs) */
+/* cpu types for specific tunings: */
+#define X86_FEATURE_K8		(3*32+ 4) /* Opteron, Athlon64 */
+#define X86_FEATURE_K7		(3*32+ 5) /* Athlon */
+#define X86_FEATURE_P3		(3*32+ 6) /* P3 */
+#define X86_FEATURE_P4		(3*32+ 7) /* P4 */
+#define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */
+#define X86_FEATURE_UP		(3*32+ 9) /* smp kernel running on up */
+#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */
+#define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */
+#define X86_FEATURE_PEBS	(3*32+12)  /* Precise-Event Based Sampling */
+#define X86_FEATURE_BTS		(3*32+13)  /* Branch Trace Store */
+/* 14 free */
+#define X86_FEATURE_SYNC_RDTSC	(3*32+15)  /* RDTSC synchronizes the CPU */
+#define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well on this CPU */
+
+/* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
+#define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
+#define X86_FEATURE_MWAIT	(4*32+ 3) /* Monitor/Mwait support */
+#define X86_FEATURE_DSCPL	(4*32+ 4) /* CPL Qualified Debug Store */
+#define X86_FEATURE_EST		(4*32+ 7) /* Enhanced SpeedStep */
+#define X86_FEATURE_TM2		(4*32+ 8) /* Thermal Monitor 2 */
+#define X86_FEATURE_CID		(4*32+10) /* Context ID */
+#define X86_FEATURE_CX16	(4*32+13) /* CMPXCHG16B */
+#define X86_FEATURE_XTPR	(4*32+14) /* Send Task Priority Messages */
+#define X86_FEATURE_DCA		(4*32+18) /* Direct Cache Access */
+
+/* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */
+#define X86_FEATURE_XSTORE	(5*32+ 2) /* on-CPU RNG present (xstore insn) */
+#define X86_FEATURE_XSTORE_EN	(5*32+ 3) /* on-CPU RNG enabled */
+#define X86_FEATURE_XCRYPT	(5*32+ 6) /* on-CPU crypto (xcrypt insn) */
+#define X86_FEATURE_XCRYPT_EN	(5*32+ 7) /* on-CPU crypto enabled */
+#define X86_FEATURE_ACE2	(5*32+ 8) /* Advanced Cryptography Engine v2 */
+#define X86_FEATURE_ACE2_EN	(5*32+ 9) /* ACE v2 enabled */
+#define X86_FEATURE_PHE		(5*32+ 10) /* PadLock Hash Engine */
+#define X86_FEATURE_PHE_EN	(5*32+ 11) /* PHE enabled */
+#define X86_FEATURE_PMM		(5*32+ 12) /* PadLock Montgomery Multiplier */
+#define X86_FEATURE_PMM_EN	(5*32+ 13) /* PMM enabled */
+
+/* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */
+#define X86_FEATURE_LAHF_LM	(6*32+ 0) /* LAHF/SAHF in long mode */
+#define X86_FEATURE_CMP_LEGACY	(6*32+ 1) /* If yes HyperThreading not valid */
+
+/*
+ * Auxiliary flags: Linux defined - For features scattered in various
+ * CPUID levels like 0x6, 0xA etc
+ */
+#define X86_FEATURE_IDA		(7*32+ 0) /* Intel Dynamic Acceleration */
+
+#define cpu_has(c, bit)							\
+	(__builtin_constant_p(bit) &&					\
+	 ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) ||	\
+	   (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) ||	\
+	   (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) ||	\
+	   (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3)) ||	\
+	   (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4)) ||	\
+	   (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5)) ||	\
+	   (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) ||	\
+	   (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) )	\
+	  ? 1 :								\
+	  test_bit(bit, (c)->x86_capability))
+#define boot_cpu_has(bit)	cpu_has(&boot_cpu_data, bit)
+
+#define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
+#define cpu_has_vme		boot_cpu_has(X86_FEATURE_VME)
+#define cpu_has_de		boot_cpu_has(X86_FEATURE_DE)
+#define cpu_has_pse		boot_cpu_has(X86_FEATURE_PSE)
+#define cpu_has_tsc		boot_cpu_has(X86_FEATURE_TSC)
+#define cpu_has_pae		boot_cpu_has(X86_FEATURE_PAE)
+#define cpu_has_pge		boot_cpu_has(X86_FEATURE_PGE)
+#define cpu_has_apic		boot_cpu_has(X86_FEATURE_APIC)
+#define cpu_has_sep		boot_cpu_has(X86_FEATURE_SEP)
+#define cpu_has_mtrr		boot_cpu_has(X86_FEATURE_MTRR)
+#define cpu_has_mmx		boot_cpu_has(X86_FEATURE_MMX)
+#define cpu_has_fxsr		boot_cpu_has(X86_FEATURE_FXSR)
+#define cpu_has_xmm		boot_cpu_has(X86_FEATURE_XMM)
+#define cpu_has_xmm2		boot_cpu_has(X86_FEATURE_XMM2)
+#define cpu_has_xmm3		boot_cpu_has(X86_FEATURE_XMM3)
+#define cpu_has_ht		boot_cpu_has(X86_FEATURE_HT)
+#define cpu_has_mp		boot_cpu_has(X86_FEATURE_MP)
+#define cpu_has_nx		boot_cpu_has(X86_FEATURE_NX)
+#define cpu_has_k6_mtrr		boot_cpu_has(X86_FEATURE_K6_MTRR)
+#define cpu_has_cyrix_arr	boot_cpu_has(X86_FEATURE_CYRIX_ARR)
+#define cpu_has_centaur_mcr	boot_cpu_has(X86_FEATURE_CENTAUR_MCR)
+#define cpu_has_xstore		boot_cpu_has(X86_FEATURE_XSTORE)
+#define cpu_has_xstore_enabled	boot_cpu_has(X86_FEATURE_XSTORE_EN)
+#define cpu_has_xcrypt		boot_cpu_has(X86_FEATURE_XCRYPT)
+#define cpu_has_xcrypt_enabled	boot_cpu_has(X86_FEATURE_XCRYPT_EN)
+#define cpu_has_ace2		boot_cpu_has(X86_FEATURE_ACE2)
+#define cpu_has_ace2_enabled	boot_cpu_has(X86_FEATURE_ACE2_EN)
+#define cpu_has_phe		boot_cpu_has(X86_FEATURE_PHE)
+#define cpu_has_phe_enabled	boot_cpu_has(X86_FEATURE_PHE_EN)
+#define cpu_has_pmm		boot_cpu_has(X86_FEATURE_PMM)
+#define cpu_has_pmm_enabled	boot_cpu_has(X86_FEATURE_PMM_EN)
+#define cpu_has_ds		boot_cpu_has(X86_FEATURE_DS)
+#define cpu_has_pebs		boot_cpu_has(X86_FEATURE_PEBS)
+#define cpu_has_clflush		boot_cpu_has(X86_FEATURE_CLFLSH)
+#define cpu_has_bts		boot_cpu_has(X86_FEATURE_BTS)
+
+#ifdef CONFIG_X86_64
+
+#undef  cpu_has_vme
+#define cpu_has_vme		0
+
+#undef  cpu_has_pae
+#define cpu_has_pae		___BUG___
+
+#undef  cpu_has_mp
+#define cpu_has_mp		1
+
+#undef  cpu_has_k6_mtrr
+#define cpu_has_k6_mtrr		0
+
+#undef  cpu_has_cyrix_arr
+#define cpu_has_cyrix_arr	0
+
+#undef  cpu_has_centaur_mcr
+#define cpu_has_centaur_mcr	0
+
+#endif /* CONFIG_X86_64 */
+
+#endif /* _ASM_X86_CPUFEATURE_H */
diff --git a/include/asm-x86/cpufeature_32.h b/include/asm-x86/cpufeature_32.h
deleted file mode 100644
index f17e688dfb05..000000000000
--- a/include/asm-x86/cpufeature_32.h
+++ /dev/null
@@ -1,176 +0,0 @@
-/*
- * cpufeature.h
- *
- * Defines x86 CPU feature bits
- */
-
-#ifndef __ASM_I386_CPUFEATURE_H
-#define __ASM_I386_CPUFEATURE_H
-
-#ifndef __ASSEMBLY__
-#include <linux/bitops.h>
-#endif
-#include <asm/required-features.h>
-
-#define NCAPINTS	8	/* N 32-bit words worth of info */
-
-/* Intel-defined CPU features, CPUID level 0x00000001 (edx), word 0 */
-#define X86_FEATURE_FPU		(0*32+ 0) /* Onboard FPU */
-#define X86_FEATURE_VME		(0*32+ 1) /* Virtual Mode Extensions */
-#define X86_FEATURE_DE		(0*32+ 2) /* Debugging Extensions */
-#define X86_FEATURE_PSE 	(0*32+ 3) /* Page Size Extensions */
-#define X86_FEATURE_TSC		(0*32+ 4) /* Time Stamp Counter */
-#define X86_FEATURE_MSR		(0*32+ 5) /* Model-Specific Registers, RDMSR, WRMSR */
-#define X86_FEATURE_PAE		(0*32+ 6) /* Physical Address Extensions */
-#define X86_FEATURE_MCE		(0*32+ 7) /* Machine Check Architecture */
-#define X86_FEATURE_CX8		(0*32+ 8) /* CMPXCHG8 instruction */
-#define X86_FEATURE_APIC	(0*32+ 9) /* Onboard APIC */
-#define X86_FEATURE_SEP		(0*32+11) /* SYSENTER/SYSEXIT */
-#define X86_FEATURE_MTRR	(0*32+12) /* Memory Type Range Registers */
-#define X86_FEATURE_PGE		(0*32+13) /* Page Global Enable */
-#define X86_FEATURE_MCA		(0*32+14) /* Machine Check Architecture */
-#define X86_FEATURE_CMOV	(0*32+15) /* CMOV instruction (FCMOVCC and FCOMI too if FPU present) */
-#define X86_FEATURE_PAT		(0*32+16) /* Page Attribute Table */
-#define X86_FEATURE_PSE36	(0*32+17) /* 36-bit PSEs */
-#define X86_FEATURE_PN		(0*32+18) /* Processor serial number */
-#define X86_FEATURE_CLFLSH	(0*32+19) /* Supports the CLFLUSH instruction */
-#define X86_FEATURE_DS		(0*32+21) /* Debug Store */
-#define X86_FEATURE_ACPI	(0*32+22) /* ACPI via MSR */
-#define X86_FEATURE_MMX		(0*32+23) /* Multimedia Extensions */
-#define X86_FEATURE_FXSR	(0*32+24) /* FXSAVE and FXRSTOR instructions (fast save and restore */
-				          /* of FPU context), and CR4.OSFXSR available */
-#define X86_FEATURE_XMM		(0*32+25) /* Streaming SIMD Extensions */
-#define X86_FEATURE_XMM2	(0*32+26) /* Streaming SIMD Extensions-2 */
-#define X86_FEATURE_SELFSNOOP	(0*32+27) /* CPU self snoop */
-#define X86_FEATURE_HT		(0*32+28) /* Hyper-Threading */
-#define X86_FEATURE_ACC		(0*32+29) /* Automatic clock control */
-#define X86_FEATURE_IA64	(0*32+30) /* IA-64 processor */
-
-/* AMD-defined CPU features, CPUID level 0x80000001, word 1 */
-/* Don't duplicate feature flags which are redundant with Intel! */
-#define X86_FEATURE_SYSCALL	(1*32+11) /* SYSCALL/SYSRET */
-#define X86_FEATURE_MP		(1*32+19) /* MP Capable. */
-#define X86_FEATURE_NX		(1*32+20) /* Execute Disable */
-#define X86_FEATURE_MMXEXT	(1*32+22) /* AMD MMX extensions */
-#define X86_FEATURE_RDTSCP	(1*32+27) /* RDTSCP */
-#define X86_FEATURE_LM		(1*32+29) /* Long Mode (x86-64) */
-#define X86_FEATURE_3DNOWEXT	(1*32+30) /* AMD 3DNow! extensions */
-#define X86_FEATURE_3DNOW	(1*32+31) /* 3DNow! */
-
-/* Transmeta-defined CPU features, CPUID level 0x80860001, word 2 */
-#define X86_FEATURE_RECOVERY	(2*32+ 0) /* CPU in recovery mode */
-#define X86_FEATURE_LONGRUN	(2*32+ 1) /* Longrun power control */
-#define X86_FEATURE_LRTI	(2*32+ 3) /* LongRun table interface */
-
-/* Other features, Linux-defined mapping, word 3 */
-/* This range is used for feature bits which conflict or are synthesized */
-#define X86_FEATURE_CXMMX	(3*32+ 0) /* Cyrix MMX extensions */
-#define X86_FEATURE_K6_MTRR	(3*32+ 1) /* AMD K6 nonstandard MTRRs */
-#define X86_FEATURE_CYRIX_ARR	(3*32+ 2) /* Cyrix ARRs (= MTRRs) */
-#define X86_FEATURE_CENTAUR_MCR	(3*32+ 3) /* Centaur MCRs (= MTRRs) */
-/* cpu types for specific tunings: */
-#define X86_FEATURE_K8		(3*32+ 4) /* Opteron, Athlon64 */
-#define X86_FEATURE_K7		(3*32+ 5) /* Athlon */
-#define X86_FEATURE_P3		(3*32+ 6) /* P3 */
-#define X86_FEATURE_P4		(3*32+ 7) /* P4 */
-#define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */
-#define X86_FEATURE_UP		(3*32+ 9) /* smp kernel running on up */
-#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */
-#define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */
-#define X86_FEATURE_PEBS	(3*32+12)  /* Precise-Event Based Sampling */
-#define X86_FEATURE_BTS		(3*32+13)  /* Branch Trace Store */
-/* 14 free */
-#define X86_FEATURE_SYNC_RDTSC	(3*32+15)  /* RDTSC synchronizes the CPU */
-#define X86_FEATURE_REP_GOOD   (3*32+16) /* rep microcode works well on this CPU */
-
-/* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
-#define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
-#define X86_FEATURE_MWAIT	(4*32+ 3) /* Monitor/Mwait support */
-#define X86_FEATURE_DSCPL	(4*32+ 4) /* CPL Qualified Debug Store */
-#define X86_FEATURE_EST		(4*32+ 7) /* Enhanced SpeedStep */
-#define X86_FEATURE_TM2		(4*32+ 8) /* Thermal Monitor 2 */
-#define X86_FEATURE_CID		(4*32+10) /* Context ID */
-#define X86_FEATURE_CX16        (4*32+13) /* CMPXCHG16B */
-#define X86_FEATURE_XTPR	(4*32+14) /* Send Task Priority Messages */
-#define X86_FEATURE_DCA		(4*32+18) /* Direct Cache Access */
-
-/* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */
-#define X86_FEATURE_XSTORE	(5*32+ 2) /* on-CPU RNG present (xstore insn) */
-#define X86_FEATURE_XSTORE_EN	(5*32+ 3) /* on-CPU RNG enabled */
-#define X86_FEATURE_XCRYPT	(5*32+ 6) /* on-CPU crypto (xcrypt insn) */
-#define X86_FEATURE_XCRYPT_EN	(5*32+ 7) /* on-CPU crypto enabled */
-#define X86_FEATURE_ACE2	(5*32+ 8) /* Advanced Cryptography Engine v2 */
-#define X86_FEATURE_ACE2_EN	(5*32+ 9) /* ACE v2 enabled */
-#define X86_FEATURE_PHE		(5*32+ 10) /* PadLock Hash Engine */
-#define X86_FEATURE_PHE_EN	(5*32+ 11) /* PHE enabled */
-#define X86_FEATURE_PMM		(5*32+ 12) /* PadLock Montgomery Multiplier */
-#define X86_FEATURE_PMM_EN	(5*32+ 13) /* PMM enabled */
-
-/* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */
-#define X86_FEATURE_LAHF_LM	(6*32+ 0) /* LAHF/SAHF in long mode */
-#define X86_FEATURE_CMP_LEGACY	(6*32+ 1) /* If yes HyperThreading not valid */
-
-/*
- * Auxiliary flags: Linux defined - For features scattered in various
- * CPUID levels like 0x6, 0xA etc
- */
-#define X86_FEATURE_IDA		(7*32+ 0) /* Intel Dynamic Acceleration */
-
-#define cpu_has(c, bit)							\
-	(__builtin_constant_p(bit) &&					\
-	 ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) ||	\
-	   (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) ||	\
-	   (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) ||	\
-	   (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3)) ||	\
-	   (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4)) ||	\
-	   (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5)) ||	\
-	   (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) ||	\
-	   (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) )	\
-	  ? 1 :								\
-	  test_bit(bit, (c)->x86_capability))
-#define boot_cpu_has(bit)	cpu_has(&boot_cpu_data, bit)
-
-#define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
-#define cpu_has_vme		boot_cpu_has(X86_FEATURE_VME)
-#define cpu_has_de		boot_cpu_has(X86_FEATURE_DE)
-#define cpu_has_pse		boot_cpu_has(X86_FEATURE_PSE)
-#define cpu_has_tsc		boot_cpu_has(X86_FEATURE_TSC)
-#define cpu_has_pae		boot_cpu_has(X86_FEATURE_PAE)
-#define cpu_has_pge		boot_cpu_has(X86_FEATURE_PGE)
-#define cpu_has_apic		boot_cpu_has(X86_FEATURE_APIC)
-#define cpu_has_sep		boot_cpu_has(X86_FEATURE_SEP)
-#define cpu_has_mtrr		boot_cpu_has(X86_FEATURE_MTRR)
-#define cpu_has_mmx		boot_cpu_has(X86_FEATURE_MMX)
-#define cpu_has_fxsr		boot_cpu_has(X86_FEATURE_FXSR)
-#define cpu_has_xmm		boot_cpu_has(X86_FEATURE_XMM)
-#define cpu_has_xmm2		boot_cpu_has(X86_FEATURE_XMM2)
-#define cpu_has_xmm3		boot_cpu_has(X86_FEATURE_XMM3)
-#define cpu_has_ht		boot_cpu_has(X86_FEATURE_HT)
-#define cpu_has_mp		boot_cpu_has(X86_FEATURE_MP)
-#define cpu_has_nx		boot_cpu_has(X86_FEATURE_NX)
-#define cpu_has_k6_mtrr		boot_cpu_has(X86_FEATURE_K6_MTRR)
-#define cpu_has_cyrix_arr	boot_cpu_has(X86_FEATURE_CYRIX_ARR)
-#define cpu_has_centaur_mcr	boot_cpu_has(X86_FEATURE_CENTAUR_MCR)
-#define cpu_has_xstore		boot_cpu_has(X86_FEATURE_XSTORE)
-#define cpu_has_xstore_enabled	boot_cpu_has(X86_FEATURE_XSTORE_EN)
-#define cpu_has_xcrypt		boot_cpu_has(X86_FEATURE_XCRYPT)
-#define cpu_has_xcrypt_enabled	boot_cpu_has(X86_FEATURE_XCRYPT_EN)
-#define cpu_has_ace2		boot_cpu_has(X86_FEATURE_ACE2)
-#define cpu_has_ace2_enabled	boot_cpu_has(X86_FEATURE_ACE2_EN)
-#define cpu_has_phe		boot_cpu_has(X86_FEATURE_PHE)
-#define cpu_has_phe_enabled	boot_cpu_has(X86_FEATURE_PHE_EN)
-#define cpu_has_pmm		boot_cpu_has(X86_FEATURE_PMM)
-#define cpu_has_pmm_enabled	boot_cpu_has(X86_FEATURE_PMM_EN)
-#define cpu_has_ds		boot_cpu_has(X86_FEATURE_DS)
-#define cpu_has_pebs 		boot_cpu_has(X86_FEATURE_PEBS)
-#define cpu_has_clflush		boot_cpu_has(X86_FEATURE_CLFLSH)
-#define cpu_has_bts 		boot_cpu_has(X86_FEATURE_BTS)
-
-#endif /* __ASM_I386_CPUFEATURE_H */
-
-/* 
- * Local Variables:
- * mode:c
- * comment-column:42
- * End:
- */
diff --git a/include/asm-x86/cpufeature_64.h b/include/asm-x86/cpufeature_64.h
deleted file mode 100644
index e18496b7b850..000000000000
--- a/include/asm-x86/cpufeature_64.h
+++ /dev/null
@@ -1,30 +0,0 @@
-/*
- * cpufeature_32.h
- *
- * Defines x86 CPU feature bits
- */
-
-#ifndef __ASM_X8664_CPUFEATURE_H
-#define __ASM_X8664_CPUFEATURE_H
-
-#include "cpufeature_32.h"
-
-#undef  cpu_has_vme
-#define cpu_has_vme            0
-
-#undef  cpu_has_pae
-#define cpu_has_pae            ___BUG___
-
-#undef  cpu_has_mp
-#define cpu_has_mp             1 /* XXX */
-
-#undef  cpu_has_k6_mtrr
-#define cpu_has_k6_mtrr        0
-
-#undef  cpu_has_cyrix_arr
-#define cpu_has_cyrix_arr      0
-
-#undef  cpu_has_centaur_mcr
-#define cpu_has_centaur_mcr    0
-
-#endif /* __ASM_X8664_CPUFEATURE_H */
-- 
cgit v1.2.3


From c202f298de59c17c0a9799dc0e1b9e0629347935 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:08 +0100
Subject: x86: clean up arch/x86/ia32/sys_ia32.c

White space and coding style clenaup.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/ia32/sys_ia32.c | 498 ++++++++++++++++++++++++-----------------------
 include/linux/compat.h   |   4 +
 2 files changed, 257 insertions(+), 245 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index bee96d614432..58991abc5b59 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -1,29 +1,29 @@
 /*
  * sys_ia32.c: Conversion between 32bit and 64bit native syscalls. Based on
- *             sys_sparc32 
+ *             sys_sparc32
  *
  * Copyright (C) 2000		VA Linux Co
  * Copyright (C) 2000		Don Dugger <n0ano@valinux.com>
- * Copyright (C) 1999 		Arun Sharma <arun.sharma@intel.com>
- * Copyright (C) 1997,1998 	Jakub Jelinek (jj@sunsite.mff.cuni.cz)
- * Copyright (C) 1997 		David S. Miller (davem@caip.rutgers.edu)
+ * Copyright (C) 1999		Arun Sharma <arun.sharma@intel.com>
+ * Copyright (C) 1997,1998	Jakub Jelinek (jj@sunsite.mff.cuni.cz)
+ * Copyright (C) 1997		David S. Miller (davem@caip.rutgers.edu)
  * Copyright (C) 2000		Hewlett-Packard Co.
  * Copyright (C) 2000		David Mosberger-Tang <davidm@hpl.hp.com>
- * Copyright (C) 2000,2001,2002	Andi Kleen, SuSE Labs (x86-64 port) 
+ * Copyright (C) 2000,2001,2002	Andi Kleen, SuSE Labs (x86-64 port)
  *
  * These routines maintain argument size conversion between 32bit and 64bit
- * environment. In 2.5 most of this should be moved to a generic directory. 
+ * environment. In 2.5 most of this should be moved to a generic directory.
  *
  * This file assumes that there is a hole at the end of user address space.
- * 
- * Some of the functions are LE specific currently. These are hopefully all marked.
- * This should be fixed.
+ *
+ * Some of the functions are LE specific currently. These are
+ * hopefully all marked.  This should be fixed.
  */
 
 #include <linux/kernel.h>
 #include <linux/sched.h>
-#include <linux/fs.h> 
-#include <linux/file.h> 
+#include <linux/fs.h>
+#include <linux/file.h>
 #include <linux/signal.h>
 #include <linux/syscalls.h>
 #include <linux/resource.h>
@@ -90,43 +90,44 @@ int cp_compat_stat(struct kstat *kbuf, struct compat_stat __user *ubuf)
 	if (sizeof(ino) < sizeof(kbuf->ino) && ino != kbuf->ino)
 		return -EOVERFLOW;
 	if (!access_ok(VERIFY_WRITE, ubuf, sizeof(struct compat_stat)) ||
-	    __put_user (old_encode_dev(kbuf->dev), &ubuf->st_dev) ||
-	    __put_user (ino, &ubuf->st_ino) ||
-	    __put_user (kbuf->mode, &ubuf->st_mode) ||
-	    __put_user (kbuf->nlink, &ubuf->st_nlink) ||
-	    __put_user (uid, &ubuf->st_uid) ||
-	    __put_user (gid, &ubuf->st_gid) ||
-	    __put_user (old_encode_dev(kbuf->rdev), &ubuf->st_rdev) ||
-	    __put_user (kbuf->size, &ubuf->st_size) ||
-	    __put_user (kbuf->atime.tv_sec, &ubuf->st_atime) ||
-	    __put_user (kbuf->atime.tv_nsec, &ubuf->st_atime_nsec) ||
-	    __put_user (kbuf->mtime.tv_sec, &ubuf->st_mtime) ||
-	    __put_user (kbuf->mtime.tv_nsec, &ubuf->st_mtime_nsec) ||
-	    __put_user (kbuf->ctime.tv_sec, &ubuf->st_ctime) ||
-	    __put_user (kbuf->ctime.tv_nsec, &ubuf->st_ctime_nsec) ||
-	    __put_user (kbuf->blksize, &ubuf->st_blksize) ||
-	    __put_user (kbuf->blocks, &ubuf->st_blocks))
+	    __put_user(old_encode_dev(kbuf->dev), &ubuf->st_dev) ||
+	    __put_user(ino, &ubuf->st_ino) ||
+	    __put_user(kbuf->mode, &ubuf->st_mode) ||
+	    __put_user(kbuf->nlink, &ubuf->st_nlink) ||
+	    __put_user(uid, &ubuf->st_uid) ||
+	    __put_user(gid, &ubuf->st_gid) ||
+	    __put_user(old_encode_dev(kbuf->rdev), &ubuf->st_rdev) ||
+	    __put_user(kbuf->size, &ubuf->st_size) ||
+	    __put_user(kbuf->atime.tv_sec, &ubuf->st_atime) ||
+	    __put_user(kbuf->atime.tv_nsec, &ubuf->st_atime_nsec) ||
+	    __put_user(kbuf->mtime.tv_sec, &ubuf->st_mtime) ||
+	    __put_user(kbuf->mtime.tv_nsec, &ubuf->st_mtime_nsec) ||
+	    __put_user(kbuf->ctime.tv_sec, &ubuf->st_ctime) ||
+	    __put_user(kbuf->ctime.tv_nsec, &ubuf->st_ctime_nsec) ||
+	    __put_user(kbuf->blksize, &ubuf->st_blksize) ||
+	    __put_user(kbuf->blocks, &ubuf->st_blocks))
 		return -EFAULT;
 	return 0;
 }
 
-asmlinkage long
-sys32_truncate64(char __user * filename, unsigned long offset_low, unsigned long offset_high)
+asmlinkage long sys32_truncate64(char __user *filename,
+				 unsigned long offset_low,
+				 unsigned long offset_high)
 {
        return sys_truncate(filename, ((loff_t) offset_high << 32) | offset_low);
 }
 
-asmlinkage long
-sys32_ftruncate64(unsigned int fd, unsigned long offset_low, unsigned long offset_high)
+asmlinkage long sys32_ftruncate64(unsigned int fd, unsigned long offset_low,
+				  unsigned long offset_high)
 {
        return sys_ftruncate(fd, ((loff_t) offset_high << 32) | offset_low);
 }
 
-/* Another set for IA32/LFS -- x86_64 struct stat is different due to 
-   support for 64bit inode numbers. */
-
-static int
-cp_stat64(struct stat64 __user *ubuf, struct kstat *stat)
+/*
+ * Another set for IA32/LFS -- x86_64 struct stat is different due to
+ * support for 64bit inode numbers.
+ */
+static int cp_stat64(struct stat64 __user *ubuf, struct kstat *stat)
 {
 	typeof(ubuf->st_uid) uid = 0;
 	typeof(ubuf->st_gid) gid = 0;
@@ -134,38 +135,39 @@ cp_stat64(struct stat64 __user *ubuf, struct kstat *stat)
 	SET_GID(gid, stat->gid);
 	if (!access_ok(VERIFY_WRITE, ubuf, sizeof(struct stat64)) ||
 	    __put_user(huge_encode_dev(stat->dev), &ubuf->st_dev) ||
-	    __put_user (stat->ino, &ubuf->__st_ino) ||
-	    __put_user (stat->ino, &ubuf->st_ino) ||
-	    __put_user (stat->mode, &ubuf->st_mode) ||
-	    __put_user (stat->nlink, &ubuf->st_nlink) ||
-	    __put_user (uid, &ubuf->st_uid) ||
-	    __put_user (gid, &ubuf->st_gid) ||
-	    __put_user (huge_encode_dev(stat->rdev), &ubuf->st_rdev) ||
-	    __put_user (stat->size, &ubuf->st_size) ||
-	    __put_user (stat->atime.tv_sec, &ubuf->st_atime) ||
-	    __put_user (stat->atime.tv_nsec, &ubuf->st_atime_nsec) ||
-	    __put_user (stat->mtime.tv_sec, &ubuf->st_mtime) ||
-	    __put_user (stat->mtime.tv_nsec, &ubuf->st_mtime_nsec) ||
-	    __put_user (stat->ctime.tv_sec, &ubuf->st_ctime) ||
-	    __put_user (stat->ctime.tv_nsec, &ubuf->st_ctime_nsec) ||
-	    __put_user (stat->blksize, &ubuf->st_blksize) ||
-	    __put_user (stat->blocks, &ubuf->st_blocks))
+	    __put_user(stat->ino, &ubuf->__st_ino) ||
+	    __put_user(stat->ino, &ubuf->st_ino) ||
+	    __put_user(stat->mode, &ubuf->st_mode) ||
+	    __put_user(stat->nlink, &ubuf->st_nlink) ||
+	    __put_user(uid, &ubuf->st_uid) ||
+	    __put_user(gid, &ubuf->st_gid) ||
+	    __put_user(huge_encode_dev(stat->rdev), &ubuf->st_rdev) ||
+	    __put_user(stat->size, &ubuf->st_size) ||
+	    __put_user(stat->atime.tv_sec, &ubuf->st_atime) ||
+	    __put_user(stat->atime.tv_nsec, &ubuf->st_atime_nsec) ||
+	    __put_user(stat->mtime.tv_sec, &ubuf->st_mtime) ||
+	    __put_user(stat->mtime.tv_nsec, &ubuf->st_mtime_nsec) ||
+	    __put_user(stat->ctime.tv_sec, &ubuf->st_ctime) ||
+	    __put_user(stat->ctime.tv_nsec, &ubuf->st_ctime_nsec) ||
+	    __put_user(stat->blksize, &ubuf->st_blksize) ||
+	    __put_user(stat->blocks, &ubuf->st_blocks))
 		return -EFAULT;
 	return 0;
 }
 
-asmlinkage long
-sys32_stat64(char __user * filename, struct stat64 __user *statbuf)
+asmlinkage long sys32_stat64(char __user *filename,
+			     struct stat64 __user *statbuf)
 {
 	struct kstat stat;
 	int ret = vfs_stat(filename, &stat);
+
 	if (!ret)
 		ret = cp_stat64(statbuf, &stat);
 	return ret;
 }
 
-asmlinkage long
-sys32_lstat64(char __user * filename, struct stat64 __user *statbuf)
+asmlinkage long sys32_lstat64(char __user *filename,
+			      struct stat64 __user *statbuf)
 {
 	struct kstat stat;
 	int ret = vfs_lstat(filename, &stat);
@@ -174,8 +176,7 @@ sys32_lstat64(char __user * filename, struct stat64 __user *statbuf)
 	return ret;
 }
 
-asmlinkage long
-sys32_fstat64(unsigned int fd, struct stat64 __user *statbuf)
+asmlinkage long sys32_fstat64(unsigned int fd, struct stat64 __user *statbuf)
 {
 	struct kstat stat;
 	int ret = vfs_fstat(fd, &stat);
@@ -184,9 +185,8 @@ sys32_fstat64(unsigned int fd, struct stat64 __user *statbuf)
 	return ret;
 }
 
-asmlinkage long
-sys32_fstatat(unsigned int dfd, char __user *filename,
-	      struct stat64 __user* statbuf, int flag)
+asmlinkage long sys32_fstatat(unsigned int dfd, char __user *filename,
+			      struct stat64 __user *statbuf, int flag)
 {
 	struct kstat stat;
 	int error = -EINVAL;
@@ -221,8 +221,7 @@ struct mmap_arg_struct {
 	unsigned int offset;
 };
 
-asmlinkage long
-sys32_mmap(struct mmap_arg_struct __user *arg)
+asmlinkage long sys32_mmap(struct mmap_arg_struct __user *arg)
 {
 	struct mmap_arg_struct a;
 	struct file *file = NULL;
@@ -233,33 +232,33 @@ sys32_mmap(struct mmap_arg_struct __user *arg)
 		return -EFAULT;
 
 	if (a.offset & ~PAGE_MASK)
-		return -EINVAL; 
+		return -EINVAL;
 
 	if (!(a.flags & MAP_ANONYMOUS)) {
 		file = fget(a.fd);
 		if (!file)
 			return -EBADF;
 	}
-	
-	mm = current->mm; 
-	down_write(&mm->mmap_sem); 
-	retval = do_mmap_pgoff(file, a.addr, a.len, a.prot, a.flags, a.offset>>PAGE_SHIFT);
+
+	mm = current->mm;
+	down_write(&mm->mmap_sem);
+	retval = do_mmap_pgoff(file, a.addr, a.len, a.prot, a.flags,
+			       a.offset>>PAGE_SHIFT);
 	if (file)
 		fput(file);
 
-	up_write(&mm->mmap_sem); 
+	up_write(&mm->mmap_sem);
 
 	return retval;
 }
 
-asmlinkage long 
-sys32_mprotect(unsigned long start, size_t len, unsigned long prot)
+asmlinkage long sys32_mprotect(unsigned long start, size_t len,
+			       unsigned long prot)
 {
-	return sys_mprotect(start,len,prot); 
+	return sys_mprotect(start, len, prot);
 }
 
-asmlinkage long
-sys32_pipe(int __user *fd)
+asmlinkage long sys32_pipe(int __user *fd)
 {
 	int retval;
 	int fds[2];
@@ -269,13 +268,13 @@ sys32_pipe(int __user *fd)
 		goto out;
 	if (copy_to_user(fd, fds, sizeof(fds)))
 		retval = -EFAULT;
-  out:
+out:
 	return retval;
 }
 
-asmlinkage long
-sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
-		   struct sigaction32 __user *oact,  unsigned int sigsetsize)
+asmlinkage long sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
+				   struct sigaction32 __user *oact,
+				   unsigned int sigsetsize)
 {
 	struct k_sigaction new_ka, old_ka;
 	int ret;
@@ -291,12 +290,17 @@ sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
 		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
 		    __get_user(handler, &act->sa_handler) ||
 		    __get_user(new_ka.sa.sa_flags, &act->sa_flags) ||
-		    __get_user(restorer, &act->sa_restorer)||
-		    __copy_from_user(&set32, &act->sa_mask, sizeof(compat_sigset_t)))
+		    __get_user(restorer, &act->sa_restorer) ||
+		    __copy_from_user(&set32, &act->sa_mask,
+				     sizeof(compat_sigset_t)))
 			return -EFAULT;
 		new_ka.sa.sa_handler = compat_ptr(handler);
 		new_ka.sa.sa_restorer = compat_ptr(restorer);
-		/* FIXME: here we rely on _COMPAT_NSIG_WORS to be >= than _NSIG_WORDS << 1 */
+
+		/*
+		 * FIXME: here we rely on _COMPAT_NSIG_WORS to be >=
+		 * than _NSIG_WORDS << 1
+		 */
 		switch (_NSIG_WORDS) {
 		case 4: new_ka.sa.sa_mask.sig[3] = set32.sig[6]
 				| (((long)set32.sig[7]) << 32);
@@ -312,7 +316,10 @@ sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
 	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
 
 	if (!ret && oact) {
-		/* FIXME: here we rely on _COMPAT_NSIG_WORS to be >= than _NSIG_WORDS << 1 */
+		/*
+		 * FIXME: here we rely on _COMPAT_NSIG_WORS to be >=
+		 * than _NSIG_WORDS << 1
+		 */
 		switch (_NSIG_WORDS) {
 		case 4:
 			set32.sig[7] = (old_ka.sa.sa_mask.sig[3] >> 32);
@@ -328,23 +335,26 @@ sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
 			set32.sig[0] = old_ka.sa.sa_mask.sig[0];
 		}
 		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(ptr_to_compat(old_ka.sa.sa_handler), &oact->sa_handler) ||
-		    __put_user(ptr_to_compat(old_ka.sa.sa_restorer), &oact->sa_restorer) ||
+		    __put_user(ptr_to_compat(old_ka.sa.sa_handler),
+			       &oact->sa_handler) ||
+		    __put_user(ptr_to_compat(old_ka.sa.sa_restorer),
+			       &oact->sa_restorer) ||
 		    __put_user(old_ka.sa.sa_flags, &oact->sa_flags) ||
-		    __copy_to_user(&oact->sa_mask, &set32, sizeof(compat_sigset_t)))
+		    __copy_to_user(&oact->sa_mask, &set32,
+				   sizeof(compat_sigset_t)))
 			return -EFAULT;
 	}
 
 	return ret;
 }
 
-asmlinkage long
-sys32_sigaction (int sig, struct old_sigaction32 __user *act, struct old_sigaction32 __user *oact)
+asmlinkage long sys32_sigaction(int sig, struct old_sigaction32 __user *act,
+				struct old_sigaction32 __user *oact)
 {
-        struct k_sigaction new_ka, old_ka;
-        int ret;
+	struct k_sigaction new_ka, old_ka;
+	int ret;
 
-        if (act) {
+	if (act) {
 		compat_old_sigset_t mask;
 		compat_uptr_t handler, restorer;
 
@@ -359,33 +369,35 @@ sys32_sigaction (int sig, struct old_sigaction32 __user *act, struct old_sigacti
 		new_ka.sa.sa_restorer = compat_ptr(restorer);
 
 		siginitset(&new_ka.sa.sa_mask, mask);
-        }
+	}
 
-        ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
 
 	if (!ret && oact) {
 		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(ptr_to_compat(old_ka.sa.sa_handler), &oact->sa_handler) ||
-		    __put_user(ptr_to_compat(old_ka.sa.sa_restorer), &oact->sa_restorer) ||
+		    __put_user(ptr_to_compat(old_ka.sa.sa_handler),
+			       &oact->sa_handler) ||
+		    __put_user(ptr_to_compat(old_ka.sa.sa_restorer),
+			       &oact->sa_restorer) ||
 		    __put_user(old_ka.sa.sa_flags, &oact->sa_flags) ||
 		    __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask))
 			return -EFAULT;
-        }
+	}
 
 	return ret;
 }
 
-asmlinkage long
-sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
-			compat_sigset_t __user *oset, unsigned int sigsetsize)
+asmlinkage long sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
+				     compat_sigset_t __user *oset,
+				     unsigned int sigsetsize)
 {
 	sigset_t s;
 	compat_sigset_t s32;
 	int ret;
 	mm_segment_t old_fs = get_fs();
-	
+
 	if (set) {
-		if (copy_from_user (&s32, set, sizeof(compat_sigset_t)))
+		if (copy_from_user(&s32, set, sizeof(compat_sigset_t)))
 			return -EFAULT;
 		switch (_NSIG_WORDS) {
 		case 4: s.sig[3] = s32.sig[6] | (((long)s32.sig[7]) << 32);
@@ -394,13 +406,14 @@ sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
 		case 1: s.sig[0] = s32.sig[0] | (((long)s32.sig[1]) << 32);
 		}
 	}
-	set_fs (KERNEL_DS);
+	set_fs(KERNEL_DS);
 	ret = sys_rt_sigprocmask(how,
 				 set ? (sigset_t __user *)&s : NULL,
 				 oset ? (sigset_t __user *)&s : NULL,
-				 sigsetsize); 
-	set_fs (old_fs);
-	if (ret) return ret;
+				 sigsetsize);
+	set_fs(old_fs);
+	if (ret)
+		return ret;
 	if (oset) {
 		switch (_NSIG_WORDS) {
 		case 4: s32.sig[7] = (s.sig[3] >> 32); s32.sig[6] = s.sig[3];
@@ -408,52 +421,49 @@ sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
 		case 2: s32.sig[3] = (s.sig[1] >> 32); s32.sig[2] = s.sig[1];
 		case 1: s32.sig[1] = (s.sig[0] >> 32); s32.sig[0] = s.sig[0];
 		}
-		if (copy_to_user (oset, &s32, sizeof(compat_sigset_t)))
+		if (copy_to_user(oset, &s32, sizeof(compat_sigset_t)))
 			return -EFAULT;
 	}
 	return 0;
 }
 
-static inline long
-get_tv32(struct timeval *o, struct compat_timeval __user *i)
+static inline long get_tv32(struct timeval *o, struct compat_timeval __user *i)
 {
-	int err = -EFAULT; 
-	if (access_ok(VERIFY_READ, i, sizeof(*i))) { 
+	int err = -EFAULT;
+
+	if (access_ok(VERIFY_READ, i, sizeof(*i))) {
 		err = __get_user(o->tv_sec, &i->tv_sec);
 		err |= __get_user(o->tv_usec, &i->tv_usec);
 	}
-	return err; 
+	return err;
 }
 
-static inline long
-put_tv32(struct compat_timeval __user *o, struct timeval *i)
+static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i)
 {
 	int err = -EFAULT;
-	if (access_ok(VERIFY_WRITE, o, sizeof(*o))) { 
+
+	if (access_ok(VERIFY_WRITE, o, sizeof(*o))) {
 		err = __put_user(i->tv_sec, &o->tv_sec);
 		err |= __put_user(i->tv_usec, &o->tv_usec);
-	} 
-	return err; 
+	}
+	return err;
 }
 
-extern unsigned int alarm_setitimer(unsigned int seconds);
-
-asmlinkage long
-sys32_alarm(unsigned int seconds)
+asmlinkage long sys32_alarm(unsigned int seconds)
 {
 	return alarm_setitimer(seconds);
 }
 
-/* Translations due to time_t size differences.  Which affects all
-   sorts of things, like timeval and itimerval.  */
-
-extern struct timezone sys_tz;
-
-asmlinkage long
-sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz)
+/*
+ * Translations due to time_t size differences. Which affects all
+ * sorts of things, like timeval and itimerval.
+ */
+asmlinkage long sys32_gettimeofday(struct compat_timeval __user *tv,
+				   struct timezone __user *tz)
 {
 	if (tv) {
 		struct timeval ktv;
+
 		do_gettimeofday(&ktv);
 		if (put_tv32(tv, &ktv))
 			return -EFAULT;
@@ -465,14 +475,14 @@ sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz)
 	return 0;
 }
 
-asmlinkage long
-sys32_settimeofday(struct compat_timeval __user *tv, struct timezone __user *tz)
+asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv,
+				   struct timezone __user *tz)
 {
 	struct timeval ktv;
 	struct timespec kts;
 	struct timezone ktz;
 
- 	if (tv) {
+	if (tv) {
 		if (get_tv32(&ktv, tv))
 			return -EFAULT;
 		kts.tv_sec = ktv.tv_sec;
@@ -494,8 +504,7 @@ struct sel_arg_struct {
 	unsigned int tvp;
 };
 
-asmlinkage long
-sys32_old_select(struct sel_arg_struct __user *arg)
+asmlinkage long sys32_old_select(struct sel_arg_struct __user *arg)
 {
 	struct sel_arg_struct a;
 
@@ -505,50 +514,45 @@ sys32_old_select(struct sel_arg_struct __user *arg)
 				 compat_ptr(a.exp), compat_ptr(a.tvp));
 }
 
-extern asmlinkage long
-compat_sys_wait4(compat_pid_t pid, compat_uint_t * stat_addr, int options,
-		 struct compat_rusage *ru);
-
-asmlinkage long
-sys32_waitpid(compat_pid_t pid, unsigned int *stat_addr, int options)
+asmlinkage long sys32_waitpid(compat_pid_t pid, unsigned int *stat_addr,
+			      int options)
 {
 	return compat_sys_wait4(pid, stat_addr, options, NULL);
 }
 
 /* 32-bit timeval and related flotsam.  */
 
-asmlinkage long
-sys32_sysfs(int option, u32 arg1, u32 arg2)
+asmlinkage long sys32_sysfs(int option, u32 arg1, u32 arg2)
 {
 	return sys_sysfs(option, arg1, arg2);
 }
 
-asmlinkage long
-sys32_sched_rr_get_interval(compat_pid_t pid, struct compat_timespec __user *interval)
+asmlinkage long sys32_sched_rr_get_interval(compat_pid_t pid,
+				    struct compat_timespec __user *interval)
 {
 	struct timespec t;
 	int ret;
-	mm_segment_t old_fs = get_fs ();
-	
-	set_fs (KERNEL_DS);
+	mm_segment_t old_fs = get_fs();
+
+	set_fs(KERNEL_DS);
 	ret = sys_sched_rr_get_interval(pid, (struct timespec __user *)&t);
-	set_fs (old_fs);
+	set_fs(old_fs);
 	if (put_compat_timespec(&t, interval))
 		return -EFAULT;
 	return ret;
 }
 
-asmlinkage long
-sys32_rt_sigpending(compat_sigset_t __user *set, compat_size_t sigsetsize)
+asmlinkage long sys32_rt_sigpending(compat_sigset_t __user *set,
+				    compat_size_t sigsetsize)
 {
 	sigset_t s;
 	compat_sigset_t s32;
 	int ret;
 	mm_segment_t old_fs = get_fs();
-		
-	set_fs (KERNEL_DS);
+
+	set_fs(KERNEL_DS);
 	ret = sys_rt_sigpending((sigset_t __user *)&s, sigsetsize);
-	set_fs (old_fs);
+	set_fs(old_fs);
 	if (!ret) {
 		switch (_NSIG_WORDS) {
 		case 4: s32.sig[7] = (s.sig[3] >> 32); s32.sig[6] = s.sig[3];
@@ -556,30 +560,29 @@ sys32_rt_sigpending(compat_sigset_t __user *set, compat_size_t sigsetsize)
 		case 2: s32.sig[3] = (s.sig[1] >> 32); s32.sig[2] = s.sig[1];
 		case 1: s32.sig[1] = (s.sig[0] >> 32); s32.sig[0] = s.sig[0];
 		}
-		if (copy_to_user (set, &s32, sizeof(compat_sigset_t)))
+		if (copy_to_user(set, &s32, sizeof(compat_sigset_t)))
 			return -EFAULT;
 	}
 	return ret;
 }
 
-asmlinkage long
-sys32_rt_sigqueueinfo(int pid, int sig, compat_siginfo_t __user *uinfo)
+asmlinkage long sys32_rt_sigqueueinfo(int pid, int sig,
+				      compat_siginfo_t __user *uinfo)
 {
 	siginfo_t info;
 	int ret;
 	mm_segment_t old_fs = get_fs();
-	
+
 	if (copy_siginfo_from_user32(&info, uinfo))
 		return -EFAULT;
-	set_fs (KERNEL_DS);
+	set_fs(KERNEL_DS);
 	ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __user *)&info);
-	set_fs (old_fs);
+	set_fs(old_fs);
 	return ret;
 }
 
 /* These are here just in case some old ia32 binary calls it. */
-asmlinkage long
-sys32_pause(void)
+asmlinkage long sys32_pause(void)
 {
 	current->state = TASK_INTERRUPTIBLE;
 	schedule();
@@ -599,25 +602,25 @@ struct sysctl_ia32 {
 };
 
 
-asmlinkage long
-sys32_sysctl(struct sysctl_ia32 __user *args32)
+asmlinkage long sys32_sysctl(struct sysctl_ia32 __user *args32)
 {
 	struct sysctl_ia32 a32;
-	mm_segment_t old_fs = get_fs ();
+	mm_segment_t old_fs = get_fs();
 	void __user *oldvalp, *newvalp;
 	size_t oldlen;
 	int __user *namep;
 	long ret;
 
-	if (copy_from_user(&a32, args32, sizeof (a32)))
+	if (copy_from_user(&a32, args32, sizeof(a32)))
 		return -EFAULT;
 
 	/*
-	 * We need to pre-validate these because we have to disable address checking
-	 * before calling do_sysctl() because of OLDLEN but we can't run the risk of the
-	 * user specifying bad addresses here.  Well, since we're dealing with 32 bit
-	 * addresses, we KNOW that access_ok() will always succeed, so this is an
-	 * expensive NOP, but so what...
+	 * We need to pre-validate these because we have to disable
+	 * address checking before calling do_sysctl() because of
+	 * OLDLEN but we can't run the risk of the user specifying bad
+	 * addresses here.  Well, since we're dealing with 32 bit
+	 * addresses, we KNOW that access_ok() will always succeed, so
+	 * this is an expensive NOP, but so what...
 	 */
 	namep = compat_ptr(a32.name);
 	oldvalp = compat_ptr(a32.oldval);
@@ -636,34 +639,34 @@ sys32_sysctl(struct sysctl_ia32 __user *args32)
 	unlock_kernel();
 	set_fs(old_fs);
 
-	if (oldvalp && put_user (oldlen, (int __user *)compat_ptr(a32.oldlenp)))
+	if (oldvalp && put_user(oldlen, (int __user *)compat_ptr(a32.oldlenp)))
 		return -EFAULT;
 
 	return ret;
 }
 #endif
 
-/* warning: next two assume little endian */ 
-asmlinkage long
-sys32_pread(unsigned int fd, char __user *ubuf, u32 count, u32 poslo, u32 poshi)
+/* warning: next two assume little endian */
+asmlinkage long sys32_pread(unsigned int fd, char __user *ubuf, u32 count,
+			    u32 poslo, u32 poshi)
 {
 	return sys_pread64(fd, ubuf, count,
 			 ((loff_t)AA(poshi) << 32) | AA(poslo));
 }
 
-asmlinkage long
-sys32_pwrite(unsigned int fd, char __user *ubuf, u32 count, u32 poslo, u32 poshi)
+asmlinkage long sys32_pwrite(unsigned int fd, char __user *ubuf, u32 count,
+			     u32 poslo, u32 poshi)
 {
 	return sys_pwrite64(fd, ubuf, count,
 			  ((loff_t)AA(poshi) << 32) | AA(poslo));
 }
 
 
-asmlinkage long
-sys32_personality(unsigned long personality)
+asmlinkage long sys32_personality(unsigned long personality)
 {
 	int ret;
-	if (personality(current->personality) == PER_LINUX32 && 
+
+	if (personality(current->personality) == PER_LINUX32 &&
 		personality == PER_LINUX)
 		personality = PER_LINUX32;
 	ret = sys_personality(personality);
@@ -672,34 +675,33 @@ sys32_personality(unsigned long personality)
 	return ret;
 }
 
-asmlinkage long
-sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, s32 count)
+asmlinkage long sys32_sendfile(int out_fd, int in_fd,
+			       compat_off_t __user *offset, s32 count)
 {
 	mm_segment_t old_fs = get_fs();
 	int ret;
 	off_t of;
-	
+
 	if (offset && get_user(of, offset))
 		return -EFAULT;
-		
+
 	set_fs(KERNEL_DS);
 	ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __user *)&of : NULL,
 			   count);
 	set_fs(old_fs);
-	
+
 	if (offset && put_user(of, offset))
 		return -EFAULT;
-		
 	return ret;
 }
 
 asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
-	unsigned long prot, unsigned long flags,
-	unsigned long fd, unsigned long pgoff)
+			    unsigned long prot, unsigned long flags,
+			    unsigned long fd, unsigned long pgoff)
 {
 	struct mm_struct *mm = current->mm;
 	unsigned long error;
-	struct file * file = NULL;
+	struct file *file = NULL;
 
 	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
 	if (!(flags & MAP_ANONYMOUS)) {
@@ -717,36 +719,35 @@ asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
 	return error;
 }
 
-asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
+asmlinkage long sys32_olduname(struct oldold_utsname __user *name)
 {
+	char *arch = "x86_64";
 	int err;
 
 	if (!name)
 		return -EFAULT;
 	if (!access_ok(VERIFY_WRITE, name, sizeof(struct oldold_utsname)))
 		return -EFAULT;
-  
-  	down_read(&uts_sem);
-
-	err = __copy_to_user(&name->sysname,&utsname()->sysname,
-				__OLD_UTS_LEN);
-	err |= __put_user(0,name->sysname+__OLD_UTS_LEN);
-	err |= __copy_to_user(&name->nodename,&utsname()->nodename,
-				__OLD_UTS_LEN);
-	err |= __put_user(0,name->nodename+__OLD_UTS_LEN);
-	err |= __copy_to_user(&name->release,&utsname()->release,
-				__OLD_UTS_LEN);
-	err |= __put_user(0,name->release+__OLD_UTS_LEN);
-	err |= __copy_to_user(&name->version,&utsname()->version,
-				__OLD_UTS_LEN);
-	err |= __put_user(0,name->version+__OLD_UTS_LEN);
-	{
-		char *arch = "x86_64";
-		if (personality(current->personality) == PER_LINUX32)
-			arch = "i686";
-		 
-		err |= __copy_to_user(&name->machine, arch, strlen(arch)+1);
-	}
+
+	down_read(&uts_sem);
+
+	err = __copy_to_user(&name->sysname, &utsname()->sysname,
+			     __OLD_UTS_LEN);
+	err |= __put_user(0, name->sysname+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->nodename, &utsname()->nodename,
+			      __OLD_UTS_LEN);
+	err |= __put_user(0, name->nodename+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->release, &utsname()->release,
+			      __OLD_UTS_LEN);
+	err |= __put_user(0, name->release+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->version, &utsname()->version,
+			      __OLD_UTS_LEN);
+	err |= __put_user(0, name->version+__OLD_UTS_LEN);
+
+	if (personality(current->personality) == PER_LINUX32)
+		arch = "i686";
+
+	err |= __copy_to_user(&name->machine, arch, strlen(arch) + 1);
 
 	up_read(&uts_sem);
 
@@ -755,17 +756,19 @@ asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
 	return err;
 }
 
-long sys32_uname(struct old_utsname __user * name)
+long sys32_uname(struct old_utsname __user *name)
 {
 	int err;
+
 	if (!name)
 		return -EFAULT;
 	down_read(&uts_sem);
-	err = copy_to_user(name, utsname(), sizeof (*name));
+	err = copy_to_user(name, utsname(), sizeof(*name));
 	up_read(&uts_sem);
-	if (personality(current->personality) == PER_LINUX32) 
+	if (personality(current->personality) == PER_LINUX32)
 		err |= copy_to_user(&name->machine, "i686", 5);
-	return err?-EFAULT:0;
+
+	return err ? -EFAULT : 0;
 }
 
 long sys32_ustat(unsigned dev, struct ustat32 __user *u32p)
@@ -773,27 +776,28 @@ long sys32_ustat(unsigned dev, struct ustat32 __user *u32p)
 	struct ustat u;
 	mm_segment_t seg;
 	int ret;
-	
-	seg = get_fs(); 
-	set_fs(KERNEL_DS); 
+
+	seg = get_fs();
+	set_fs(KERNEL_DS);
 	ret = sys_ustat(dev, (struct ustat __user *)&u);
 	set_fs(seg);
-	if (ret >= 0) { 
-		if (!access_ok(VERIFY_WRITE,u32p,sizeof(struct ustat32)) || 
-		    __put_user((__u32) u.f_tfree, &u32p->f_tfree) ||
-		    __put_user((__u32) u.f_tinode, &u32p->f_tfree) ||
-		    __copy_to_user(&u32p->f_fname, u.f_fname, sizeof(u.f_fname)) ||
-		    __copy_to_user(&u32p->f_fpack, u.f_fpack, sizeof(u.f_fpack)))
-			ret = -EFAULT;
-	}
+	if (ret < 0)
+		return ret;
+
+	if (!access_ok(VERIFY_WRITE, u32p, sizeof(struct ustat32)) ||
+	    __put_user((__u32) u.f_tfree, &u32p->f_tfree) ||
+	    __put_user((__u32) u.f_tinode, &u32p->f_tfree) ||
+	    __copy_to_user(&u32p->f_fname, u.f_fname, sizeof(u.f_fname)) ||
+	    __copy_to_user(&u32p->f_fpack, u.f_fpack, sizeof(u.f_fpack)))
+		ret = -EFAULT;
 	return ret;
-} 
+}
 
 asmlinkage long sys32_execve(char __user *name, compat_uptr_t __user *argv,
 			     compat_uptr_t __user *envp, struct pt_regs *regs)
 {
 	long error;
-	char * filename;
+	char *filename;
 
 	filename = getname(name);
 	error = PTR_ERR(filename);
@@ -814,16 +818,17 @@ asmlinkage long sys32_clone(unsigned int clone_flags, unsigned int newsp,
 {
 	void __user *parent_tid = (void __user *)regs->rdx;
 	void __user *child_tid = (void __user *)regs->rdi;
+
 	if (!newsp)
 		newsp = regs->rsp;
-        return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
+	return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
 }
 
 /*
- * Some system calls that need sign extended arguments. This could be done by a generic wrapper.
- */ 
-
-long sys32_lseek (unsigned int fd, int offset, unsigned int whence)
+ * Some system calls that need sign extended arguments. This could be
+ * done by a generic wrapper.
+ */
+long sys32_lseek(unsigned int fd, int offset, unsigned int whence)
 {
 	return sys_lseek(fd, offset, whence);
 }
@@ -832,49 +837,52 @@ long sys32_kill(int pid, int sig)
 {
 	return sys_kill(pid, sig);
 }
- 
-long sys32_fadvise64_64(int fd, __u32 offset_low, __u32 offset_high, 
+
+long sys32_fadvise64_64(int fd, __u32 offset_low, __u32 offset_high,
 			__u32 len_low, __u32 len_high, int advice)
-{ 
+{
 	return sys_fadvise64_64(fd,
 			       (((u64)offset_high)<<32) | offset_low,
 			       (((u64)len_high)<<32) | len_low,
-			       advice); 
-} 
+				advice);
+}
 
 long sys32_vm86_warning(void)
-{ 
+{
 	struct task_struct *me = current;
 	static char lastcomm[sizeof(me->comm)];
+
 	if (strncmp(lastcomm, me->comm, sizeof(lastcomm))) {
-		compat_printk(KERN_INFO "%s: vm86 mode not supported on 64 bit kernel\n",
-		       me->comm);
+		compat_printk(KERN_INFO
+			      "%s: vm86 mode not supported on 64 bit kernel\n",
+			      me->comm);
 		strncpy(lastcomm, me->comm, sizeof(lastcomm));
-	} 
+	}
 	return -ENOSYS;
-} 
+}
 
 long sys32_lookup_dcookie(u32 addr_low, u32 addr_high,
-			  char __user * buf, size_t len)
+			  char __user *buf, size_t len)
 {
 	return sys_lookup_dcookie(((u64)addr_high << 32) | addr_low, buf, len);
 }
 
-asmlinkage ssize_t sys32_readahead(int fd, unsigned off_lo, unsigned off_hi, size_t count)
+asmlinkage ssize_t sys32_readahead(int fd, unsigned off_lo, unsigned off_hi,
+				   size_t count)
 {
 	return sys_readahead(fd, ((u64)off_hi << 32) | off_lo, count);
 }
 
 asmlinkage long sys32_sync_file_range(int fd, unsigned off_low, unsigned off_hi,
-			   unsigned n_low, unsigned n_hi,  int flags)
+				      unsigned n_low, unsigned n_hi,  int flags)
 {
 	return sys_sync_file_range(fd,
 				   ((u64)off_hi << 32) | off_low,
 				   ((u64)n_hi << 32) | n_low, flags);
 }
 
-asmlinkage long sys32_fadvise64(int fd, unsigned offset_lo, unsigned offset_hi, size_t len,
-		     int advice)
+asmlinkage long sys32_fadvise64(int fd, unsigned offset_lo, unsigned offset_hi,
+				size_t len, int advice)
 {
 	return sys_fadvise64_64(fd, ((u64)offset_hi << 32) | offset_lo,
 				len, advice);
diff --git a/include/linux/compat.h b/include/linux/compat.h
index 0e69d2cf14aa..ba29d4c59643 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -191,6 +191,10 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp,
 		compat_ulong_t __user *outp, compat_ulong_t __user *exp,
 		struct compat_timeval __user *tvp);
 
+asmlinkage long compat_sys_wait4(compat_pid_t pid,
+				 compat_uint_t *stat_addr, int options,
+				 struct compat_rusage *ru);
+
 #define BITS_PER_COMPAT_LONG    (8*sizeof(compat_long_t))
 
 #define BITS_TO_COMPAT_LONGS(bits) \
-- 
cgit v1.2.3


From 4edc8f5454814201d2fb222354284365b4f0537c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:13 +0100
Subject: x86: clean up include/asm-x86/desc_64.h

White space and coding style clenaup.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/desc_64.h | 97 +++++++++++++++++++++++++----------------------
 1 file changed, 52 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 7d9c938e69fd..bb2009ecbbca 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -1,4 +1,4 @@
-/* Written 2000 by Andi Kleen */ 
+/* Written 2000 by Andi Kleen */
 #ifndef __ARCH_DESC_H
 #define __ARCH_DESC_H
 
@@ -35,7 +35,7 @@ static inline unsigned long __store_tr(void)
  * something other than this.
  */
 extern struct desc_struct default_ldt[];
-extern struct gate_struct idt_table[]; 
+extern struct gate_struct idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
 /* the cpu gdt accessor */
@@ -51,40 +51,45 @@ static inline void store_gdt(struct desc_ptr *ptr)
        asm("sgdt %w0":"=m" (*ptr));
 }
 
-static inline void _set_gate(void *adr, unsigned type, unsigned long func, unsigned dpl, unsigned ist)  
+static inline void _set_gate(void *adr, unsigned type, unsigned long func,
+			     unsigned dpl, unsigned ist)
 {
-	struct gate_struct s; 	
-	s.offset_low = PTR_LOW(func); 
+	struct gate_struct s;
+
+	s.offset_low = PTR_LOW(func);
 	s.segment = __KERNEL_CS;
-	s.ist = ist; 
+	s.ist = ist;
 	s.p = 1;
-	s.dpl = dpl; 
+	s.dpl = dpl;
 	s.zero0 = 0;
-	s.zero1 = 0; 
-	s.type = type; 
-	s.offset_middle = PTR_MIDDLE(func); 
-	s.offset_high = PTR_HIGH(func); 
-	/* does not need to be atomic because it is only done once at setup time */ 
-	memcpy(adr, &s, 16); 
-} 
-
-static inline void set_intr_gate(int nr, void *func) 
-{ 
+	s.zero1 = 0;
+	s.type = type;
+	s.offset_middle = PTR_MIDDLE(func);
+	s.offset_high = PTR_HIGH(func);
+	/*
+	 * does not need to be atomic because it is only done once at
+	 * setup time
+	 */
+	memcpy(adr, &s, 16);
+}
+
+static inline void set_intr_gate(int nr, void *func)
+{
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, 0); 
-} 
+	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, 0);
+}
 
-static inline void set_intr_gate_ist(int nr, void *func, unsigned ist) 
-{ 
+static inline void set_intr_gate_ist(int nr, void *func, unsigned ist)
+{
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, ist); 
-} 
+	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, ist);
+}
 
-static inline void set_system_gate(int nr, void *func) 
-{ 
+static inline void set_system_gate(int nr, void *func)
+{
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, 0); 
-} 
+	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, 0);
+}
 
 static inline void set_system_gate_ist(int nr, void *func, unsigned ist)
 {
@@ -101,24 +106,25 @@ static inline void store_idt(struct desc_ptr *dtr)
        asm("sidt %w0":"=m" (*dtr));
 }
 
-static inline void set_tssldt_descriptor(void *ptr, unsigned long tss, unsigned type, 
-					 unsigned size) 
-{ 
+static inline void set_tssldt_descriptor(void *ptr, unsigned long tss,
+					 unsigned type, unsigned size)
+{
 	struct ldttss_desc d;
-	memset(&d,0,sizeof(d)); 
+
+	memset(&d, 0, sizeof(d));
 	d.limit0 = size & 0xFFFF;
-	d.base0 = PTR_LOW(tss); 
-	d.base1 = PTR_MIDDLE(tss) & 0xFF; 
+	d.base0 = PTR_LOW(tss);
+	d.base1 = PTR_MIDDLE(tss) & 0xFF;
 	d.type = type;
-	d.p = 1; 
+	d.p = 1;
 	d.limit1 = (size >> 16) & 0xF;
-	d.base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF; 
-	d.base3 = PTR_HIGH(tss); 
-	memcpy(ptr, &d, 16); 
+	d.base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
+	d.base3 = PTR_HIGH(tss);
+	memcpy(ptr, &d, 16);
 }
 
 static inline void set_tss_desc(unsigned cpu, void *addr)
-{ 
+{
 	/*
 	 * sizeof(unsigned long) coming from an extra "long" at the end
 	 * of the iobitmap. See tss_struct definition in processor.h
@@ -129,18 +135,18 @@ static inline void set_tss_desc(unsigned cpu, void *addr)
 	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_TSS],
 		(unsigned long)addr, DESC_TSS,
 		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
-} 
+}
 
 static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
-{ 
+{
 	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_LDT], (unsigned long)addr,
 			      DESC_LDT, size * 8 - 1);
 }
 
 #define LDT_entry_a(info) \
 	((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
-/* Don't allow setting of the lm bit. It is useless anyways because 
-   64bit system calls require __USER_CS. */ 
+/* Don't allow setting of the lm bit. It is useless anyways because
+   64bit system calls require __USER_CS. */
 #define LDT_entry_b(info) \
 	(((info)->base_addr & 0xff000000) | \
 	(((info)->base_addr & 0x00ff0000) >> 16) | \
@@ -172,12 +178,12 @@ static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 
 	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
 		gdt[i] = t->tls_array[i];
-} 
+}
 
 /*
  * load one particular LDT into the current CPU
  */
-static inline void load_LDT_nolock (mm_context_t *pc, int cpu)
+static inline void load_LDT_nolock(mm_context_t *pc, int cpu)
 {
 	int count = pc->size;
 
@@ -185,7 +191,7 @@ static inline void load_LDT_nolock (mm_context_t *pc, int cpu)
 		clear_LDT();
 		return;
 	}
-		
+
 	set_ldt_desc(cpu, pc->ldt, count);
 	load_LDT_desc();
 }
@@ -193,6 +199,7 @@ static inline void load_LDT_nolock (mm_context_t *pc, int cpu)
 static inline void load_LDT(mm_context_t *pc)
 {
 	int cpu = get_cpu();
+
 	load_LDT_nolock(pc, cpu);
 	put_cpu();
 }
-- 
cgit v1.2.3


From fc2d625c4fac18e672a3b7c61af5de22d7ab7d87 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:13 +0100
Subject: x86: introduce ldt_write accessor

Create a ldt write accessor like the 32 bit one.

Preparatory patch for merging ldt.c and anyway necessary for
64bit paravirt ops.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/ldt_64.c  | 8 +++-----
 include/asm-x86/desc_64.h | 9 +++++++++
 2 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ldt_64.c b/arch/x86/kernel/ldt_64.c
index 6c41db367de8..d72dc7a0636f 100644
--- a/arch/x86/kernel/ldt_64.c
+++ b/arch/x86/kernel/ldt_64.c
@@ -172,7 +172,7 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 {
 	struct task_struct *me = current;
 	struct mm_struct *mm = me->mm;
-	__u32 entry_1, entry_2, *lp;
+	__u32 entry_1, entry_2;
 	int error;
 	struct user_desc ldt_info;
 
@@ -202,8 +202,6 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 			goto out_unlock;
 	}
 
-	lp = (__u32 *)((ldt_info.entry_number << 3) + (char *)mm->context.ldt);
-
 	/* Allow LDTs to be cleared by the user. */
 	if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
 		if (oldmode || LDT_empty(&ldt_info)) {
@@ -220,8 +218,8 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 
 	/* Install the new entry ...  */
 install:
-	*lp	= entry_1;
-	*(lp+1)	= entry_2;
+	write_ldt_entry(mm->context.ldt, ldt_info.entry_number, entry_1,
+			entry_2);
 	error = 0;
 
 out_unlock:
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index bb2009ecbbca..7d48df72aef2 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -38,6 +38,15 @@ extern struct desc_struct default_ldt[];
 extern struct gate_struct idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
+static inline void write_ldt_entry(struct desc_struct *ldt,
+				   int entry, u32 entry_low, u32 entry_high)
+{
+	__u32 *lp = (__u32 *)((entry << 3) + (char *)ldt);
+
+	lp[0] = entry_low;
+	lp[1] = entry_high;
+}
+
 /* the cpu gdt accessor */
 #define cpu_gdt(_cpu) ((struct desc_struct *)cpu_gdt_descr[_cpu].address)
 
-- 
cgit v1.2.3


From 2d539553c96771bc8f77156f27500d35e1fe114c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:14 +0100
Subject: x86: unify include/asm-x86/apicdef_32/64.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apicdef.h    | 399 ++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/apicdef_32.h | 375 ----------------------------------------
 include/asm-x86/apicdef_64.h | 392 ------------------------------------------
 3 files changed, 397 insertions(+), 769 deletions(-)
 delete mode 100644 include/asm-x86/apicdef_32.h
 delete mode 100644 include/asm-x86/apicdef_64.h

(limited to 'include')

diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 4542c220bf4d..83ac1e6e3625 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -1,5 +1,400 @@
+#ifndef _ASM_X86_APICDEF_H
+#define _ASM_X86_APICDEF_H
+
+/*
+ * Constants for various Intel APICs. (local APIC, IOAPIC, etc.)
+ *
+ * Alan Cox <Alan.Cox@linux.org>, 1995.
+ * Ingo Molnar <mingo@redhat.com>, 1999, 2000
+ */
+
+#define	APIC_DEFAULT_PHYS_BASE	0xfee00000
+
+#define	APIC_ID		0x20
+
+#ifdef CONFIG_X86_64
+# define	APIC_ID_MASK		(0xFFu<<24)
+# define	GET_APIC_ID(x)		(((x)>>24)&0xFFu)
+# define	SET_APIC_ID(x)		(((x)<<24))
+#endif
+
+#define	APIC_LVR	0x30
+#define		APIC_LVR_MASK		0xFF00FF
+#define		GET_APIC_VERSION(x)	((x)&0xFFu)
+#define		GET_APIC_MAXLVT(x)	(((x)>>16)&0xFFu)
+#define		APIC_INTEGRATED(x)	((x)&0xF0u)
+#define		APIC_XAPIC(x)		((x) >= 0x14)
+#define	APIC_TASKPRI	0x80
+#define		APIC_TPRI_MASK		0xFFu
+#define	APIC_ARBPRI	0x90
+#define		APIC_ARBPRI_MASK	0xFFu
+#define	APIC_PROCPRI	0xA0
+#define	APIC_EOI	0xB0
+#define		APIC_EIO_ACK		0x0
+#define	APIC_RRR	0xC0
+#define	APIC_LDR	0xD0
+#define		APIC_LDR_MASK		(0xFFu<<24)
+#define		GET_APIC_LOGICAL_ID(x)	(((x)>>24)&0xFFu)
+#define		SET_APIC_LOGICAL_ID(x)	(((x)<<24))
+#define		APIC_ALL_CPUS		0xFFu
+#define	APIC_DFR	0xE0
+#define		APIC_DFR_CLUSTER		0x0FFFFFFFul
+#define		APIC_DFR_FLAT			0xFFFFFFFFul
+#define	APIC_SPIV	0xF0
+#define		APIC_SPIV_FOCUS_DISABLED	(1<<9)
+#define		APIC_SPIV_APIC_ENABLED		(1<<8)
+#define	APIC_ISR	0x100
+#define	APIC_ISR_NR     0x8     /* Number of 32 bit ISR registers. */
+#define	APIC_TMR	0x180
+#define	APIC_IRR	0x200
+#define	APIC_ESR	0x280
+#define		APIC_ESR_SEND_CS	0x00001
+#define		APIC_ESR_RECV_CS	0x00002
+#define		APIC_ESR_SEND_ACC	0x00004
+#define		APIC_ESR_RECV_ACC	0x00008
+#define		APIC_ESR_SENDILL	0x00020
+#define		APIC_ESR_RECVILL	0x00040
+#define		APIC_ESR_ILLREGA	0x00080
+#define	APIC_ICR	0x300
+#define		APIC_DEST_SELF		0x40000
+#define		APIC_DEST_ALLINC	0x80000
+#define		APIC_DEST_ALLBUT	0xC0000
+#define		APIC_ICR_RR_MASK	0x30000
+#define		APIC_ICR_RR_INVALID	0x00000
+#define		APIC_ICR_RR_INPROG	0x10000
+#define		APIC_ICR_RR_VALID	0x20000
+#define		APIC_INT_LEVELTRIG	0x08000
+#define		APIC_INT_ASSERT		0x04000
+#define		APIC_ICR_BUSY		0x01000
+#define		APIC_DEST_LOGICAL	0x00800
+#define		APIC_DEST_PHYSICAL	0x00000
+#define		APIC_DM_FIXED		0x00000
+#define		APIC_DM_LOWEST		0x00100
+#define		APIC_DM_SMI		0x00200
+#define		APIC_DM_REMRD		0x00300
+#define		APIC_DM_NMI		0x00400
+#define		APIC_DM_INIT		0x00500
+#define		APIC_DM_STARTUP		0x00600
+#define		APIC_DM_EXTINT		0x00700
+#define		APIC_VECTOR_MASK	0x000FF
+#define	APIC_ICR2	0x310
+#define		GET_APIC_DEST_FIELD(x)	(((x)>>24)&0xFF)
+#define		SET_APIC_DEST_FIELD(x)	((x)<<24)
+#define	APIC_LVTT	0x320
+#define	APIC_LVTTHMR	0x330
+#define	APIC_LVTPC	0x340
+#define	APIC_LVT0	0x350
+#define		APIC_LVT_TIMER_BASE_MASK	(0x3<<18)
+#define		GET_APIC_TIMER_BASE(x)		(((x)>>18)&0x3)
+#define		SET_APIC_TIMER_BASE(x)		(((x)<<18))
+#define		APIC_TIMER_BASE_CLKIN		0x0
+#define		APIC_TIMER_BASE_TMBASE		0x1
+#define		APIC_TIMER_BASE_DIV		0x2
+#define		APIC_LVT_TIMER_PERIODIC		(1<<17)
+#define		APIC_LVT_MASKED			(1<<16)
+#define		APIC_LVT_LEVEL_TRIGGER		(1<<15)
+#define		APIC_LVT_REMOTE_IRR		(1<<14)
+#define		APIC_INPUT_POLARITY		(1<<13)
+#define		APIC_SEND_PENDING		(1<<12)
+#define		APIC_MODE_MASK			0x700
+#define		GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
+#define		SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
+#define			APIC_MODE_FIXED		0x0
+#define			APIC_MODE_NMI		0x4
+#define			APIC_MODE_EXTINT	0x7
+#define	APIC_LVT1	0x360
+#define	APIC_LVTERR	0x370
+#define	APIC_TMICT	0x380
+#define	APIC_TMCCT	0x390
+#define	APIC_TDCR	0x3E0
+#define		APIC_TDR_DIV_TMBASE	(1<<2)
+#define		APIC_TDR_DIV_1		0xB
+#define		APIC_TDR_DIV_2		0x0
+#define		APIC_TDR_DIV_4		0x1
+#define		APIC_TDR_DIV_8		0x2
+#define		APIC_TDR_DIV_16		0x3
+#define		APIC_TDR_DIV_32		0x8
+#define		APIC_TDR_DIV_64		0x9
+#define		APIC_TDR_DIV_128	0xA
+#define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
+
 #ifdef CONFIG_X86_32
-# include "apicdef_32.h"
+# define MAX_IO_APICS 64
 #else
-# include "apicdef_64.h"
+# define MAX_IO_APICS 128
+# define MAX_LOCAL_APIC 256
+#endif
+
+/*
+ * All x86-64 systems are xAPIC compatible.
+ * In the following, "apicid" is a physical APIC ID.
+ */
+#define XAPIC_DEST_CPUS_SHIFT	4
+#define XAPIC_DEST_CPUS_MASK	((1u << XAPIC_DEST_CPUS_SHIFT) - 1)
+#define XAPIC_DEST_CLUSTER_MASK	(XAPIC_DEST_CPUS_MASK << XAPIC_DEST_CPUS_SHIFT)
+#define APIC_CLUSTER(apicid)	((apicid) & XAPIC_DEST_CLUSTER_MASK)
+#define APIC_CLUSTERID(apicid)	(APIC_CLUSTER(apicid) >> XAPIC_DEST_CPUS_SHIFT)
+#define APIC_CPUID(apicid)	((apicid) & XAPIC_DEST_CPUS_MASK)
+#define NUM_APIC_CLUSTERS	((BAD_APICID + 1) >> XAPIC_DEST_CPUS_SHIFT)
+
+/*
+ * the local APIC register structure, memory mapped. Not terribly well
+ * tested, but we might eventually use this one in the future - the
+ * problem why we cannot use it right now is the P5 APIC, it has an
+ * errata which cannot take 8-bit reads and writes, only 32-bit ones ...
+ */
+#define u32 unsigned int
+
+struct local_apic {
+
+/*000*/	struct { u32 __reserved[4]; } __reserved_01;
+
+/*010*/	struct { u32 __reserved[4]; } __reserved_02;
+
+/*020*/	struct { /* APIC ID Register */
+		u32   __reserved_1	: 24,
+			phys_apic_id	:  4,
+			__reserved_2	:  4;
+		u32 __reserved[3];
+	} id;
+
+/*030*/	const
+	struct { /* APIC Version Register */
+		u32   version		:  8,
+			__reserved_1	:  8,
+			max_lvt		:  8,
+			__reserved_2	:  8;
+		u32 __reserved[3];
+	} version;
+
+/*040*/	struct { u32 __reserved[4]; } __reserved_03;
+
+/*050*/	struct { u32 __reserved[4]; } __reserved_04;
+
+/*060*/	struct { u32 __reserved[4]; } __reserved_05;
+
+/*070*/	struct { u32 __reserved[4]; } __reserved_06;
+
+/*080*/	struct { /* Task Priority Register */
+		u32   priority	:  8,
+			__reserved_1	: 24;
+		u32 __reserved_2[3];
+	} tpr;
+
+/*090*/	const
+	struct { /* Arbitration Priority Register */
+		u32   priority	:  8,
+			__reserved_1	: 24;
+		u32 __reserved_2[3];
+	} apr;
+
+/*0A0*/	const
+	struct { /* Processor Priority Register */
+		u32   priority	:  8,
+			__reserved_1	: 24;
+		u32 __reserved_2[3];
+	} ppr;
+
+/*0B0*/	struct { /* End Of Interrupt Register */
+		u32   eoi;
+		u32 __reserved[3];
+	} eoi;
+
+/*0C0*/	struct { u32 __reserved[4]; } __reserved_07;
+
+/*0D0*/	struct { /* Logical Destination Register */
+		u32   __reserved_1	: 24,
+			logical_dest	:  8;
+		u32 __reserved_2[3];
+	} ldr;
+
+/*0E0*/	struct { /* Destination Format Register */
+		u32   __reserved_1	: 28,
+			model		:  4;
+		u32 __reserved_2[3];
+	} dfr;
+
+/*0F0*/	struct { /* Spurious Interrupt Vector Register */
+		u32	spurious_vector	:  8,
+			apic_enabled	:  1,
+			focus_cpu	:  1,
+			__reserved_2	: 22;
+		u32 __reserved_3[3];
+	} svr;
+
+/*100*/	struct { /* In Service Register */
+/*170*/		u32 bitfield;
+		u32 __reserved[3];
+	} isr [8];
+
+/*180*/	struct { /* Trigger Mode Register */
+/*1F0*/		u32 bitfield;
+		u32 __reserved[3];
+	} tmr [8];
+
+/*200*/	struct { /* Interrupt Request Register */
+/*270*/		u32 bitfield;
+		u32 __reserved[3];
+	} irr [8];
+
+/*280*/	union { /* Error Status Register */
+		struct {
+			u32   send_cs_error			:  1,
+				receive_cs_error		:  1,
+				send_accept_error		:  1,
+				receive_accept_error		:  1,
+				__reserved_1			:  1,
+				send_illegal_vector		:  1,
+				receive_illegal_vector		:  1,
+				illegal_register_address	:  1,
+				__reserved_2			: 24;
+			u32 __reserved_3[3];
+		} error_bits;
+		struct {
+			u32 errors;
+			u32 __reserved_3[3];
+		} all_errors;
+	} esr;
+
+/*290*/	struct { u32 __reserved[4]; } __reserved_08;
+
+/*2A0*/	struct { u32 __reserved[4]; } __reserved_09;
+
+/*2B0*/	struct { u32 __reserved[4]; } __reserved_10;
+
+/*2C0*/	struct { u32 __reserved[4]; } __reserved_11;
+
+/*2D0*/	struct { u32 __reserved[4]; } __reserved_12;
+
+/*2E0*/	struct { u32 __reserved[4]; } __reserved_13;
+
+/*2F0*/	struct { u32 __reserved[4]; } __reserved_14;
+
+/*300*/	struct { /* Interrupt Command Register 1 */
+		u32   vector			:  8,
+			delivery_mode		:  3,
+			destination_mode	:  1,
+			delivery_status		:  1,
+			__reserved_1		:  1,
+			level			:  1,
+			trigger			:  1,
+			__reserved_2		:  2,
+			shorthand		:  2,
+			__reserved_3		:  12;
+		u32 __reserved_4[3];
+	} icr1;
+
+/*310*/	struct { /* Interrupt Command Register 2 */
+		union {
+			u32   __reserved_1	: 24,
+				phys_dest	:  4,
+				__reserved_2	:  4;
+			u32   __reserved_3	: 24,
+				logical_dest	:  8;
+		} dest;
+		u32 __reserved_4[3];
+	} icr2;
+
+/*320*/	struct { /* LVT - Timer */
+		u32   vector		:  8,
+			__reserved_1	:  4,
+			delivery_status	:  1,
+			__reserved_2	:  3,
+			mask		:  1,
+			timer_mode	:  1,
+			__reserved_3	: 14;
+		u32 __reserved_4[3];
+	} lvt_timer;
+
+/*330*/	struct { /* LVT - Thermal Sensor */
+		u32  vector		:  8,
+			delivery_mode	:  3,
+			__reserved_1	:  1,
+			delivery_status	:  1,
+			__reserved_2	:  3,
+			mask		:  1,
+			__reserved_3	: 15;
+		u32 __reserved_4[3];
+	} lvt_thermal;
+
+/*340*/	struct { /* LVT - Performance Counter */
+		u32   vector		:  8,
+			delivery_mode	:  3,
+			__reserved_1	:  1,
+			delivery_status	:  1,
+			__reserved_2	:  3,
+			mask		:  1,
+			__reserved_3	: 15;
+		u32 __reserved_4[3];
+	} lvt_pc;
+
+/*350*/	struct { /* LVT - LINT0 */
+		u32   vector		:  8,
+			delivery_mode	:  3,
+			__reserved_1	:  1,
+			delivery_status	:  1,
+			polarity	:  1,
+			remote_irr	:  1,
+			trigger		:  1,
+			mask		:  1,
+			__reserved_2	: 15;
+		u32 __reserved_3[3];
+	} lvt_lint0;
+
+/*360*/	struct { /* LVT - LINT1 */
+		u32   vector		:  8,
+			delivery_mode	:  3,
+			__reserved_1	:  1,
+			delivery_status	:  1,
+			polarity	:  1,
+			remote_irr	:  1,
+			trigger		:  1,
+			mask		:  1,
+			__reserved_2	: 15;
+		u32 __reserved_3[3];
+	} lvt_lint1;
+
+/*370*/	struct { /* LVT - Error */
+		u32   vector		:  8,
+			__reserved_1	:  4,
+			delivery_status	:  1,
+			__reserved_2	:  3,
+			mask		:  1,
+			__reserved_3	: 15;
+		u32 __reserved_4[3];
+	} lvt_error;
+
+/*380*/	struct { /* Timer Initial Count Register */
+		u32   initial_count;
+		u32 __reserved_2[3];
+	} timer_icr;
+
+/*390*/	const
+	struct { /* Timer Current Count Register */
+		u32   curr_count;
+		u32 __reserved_2[3];
+	} timer_ccr;
+
+/*3A0*/	struct { u32 __reserved[4]; } __reserved_16;
+
+/*3B0*/	struct { u32 __reserved[4]; } __reserved_17;
+
+/*3C0*/	struct { u32 __reserved[4]; } __reserved_18;
+
+/*3D0*/	struct { u32 __reserved[4]; } __reserved_19;
+
+/*3E0*/	struct { /* Timer Divide Configuration Register */
+		u32   divisor		:  4,
+			__reserved_1	: 28;
+		u32 __reserved_2[3];
+	} timer_dcr;
+
+/*3F0*/	struct { u32 __reserved[4]; } __reserved_20;
+
+} __attribute__ ((packed));
+
+#undef u32
+
+#define BAD_APICID 0xFFu
+
 #endif
diff --git a/include/asm-x86/apicdef_32.h b/include/asm-x86/apicdef_32.h
deleted file mode 100644
index 9f6995341fdc..000000000000
--- a/include/asm-x86/apicdef_32.h
+++ /dev/null
@@ -1,375 +0,0 @@
-#ifndef __ASM_APICDEF_H
-#define __ASM_APICDEF_H
-
-/*
- * Constants for various Intel APICs. (local APIC, IOAPIC, etc.)
- *
- * Alan Cox <Alan.Cox@linux.org>, 1995.
- * Ingo Molnar <mingo@redhat.com>, 1999, 2000
- */
-
-#define		APIC_DEFAULT_PHYS_BASE	0xfee00000
- 
-#define		APIC_ID		0x20
-#define		APIC_LVR	0x30
-#define			APIC_LVR_MASK		0xFF00FF
-#define			GET_APIC_VERSION(x)	((x)&0xFF)
-#define			GET_APIC_MAXLVT(x)	(((x)>>16)&0xFF)
-#define			APIC_INTEGRATED(x)	((x)&0xF0)
-#define			APIC_XAPIC(x)		((x) >= 0x14)
-#define		APIC_TASKPRI	0x80
-#define			APIC_TPRI_MASK		0xFF
-#define		APIC_ARBPRI	0x90
-#define			APIC_ARBPRI_MASK	0xFF
-#define		APIC_PROCPRI	0xA0
-#define		APIC_EOI	0xB0
-#define			APIC_EIO_ACK		0x0		/* Write this to the EOI register */
-#define		APIC_RRR	0xC0
-#define		APIC_LDR	0xD0
-#define			APIC_LDR_MASK		(0xFF<<24)
-#define			GET_APIC_LOGICAL_ID(x)	(((x)>>24)&0xFF)
-#define			SET_APIC_LOGICAL_ID(x)	(((x)<<24))
-#define			APIC_ALL_CPUS		0xFF
-#define		APIC_DFR	0xE0
-#define			APIC_DFR_CLUSTER		0x0FFFFFFFul
-#define			APIC_DFR_FLAT			0xFFFFFFFFul
-#define		APIC_SPIV	0xF0
-#define			APIC_SPIV_FOCUS_DISABLED	(1<<9)
-#define			APIC_SPIV_APIC_ENABLED		(1<<8)
-#define		APIC_ISR	0x100
-#define         APIC_ISR_NR     0x8     /* Number of 32 bit ISR registers. */
-#define		APIC_TMR	0x180
-#define 	APIC_IRR	0x200
-#define 	APIC_ESR	0x280
-#define			APIC_ESR_SEND_CS	0x00001
-#define			APIC_ESR_RECV_CS	0x00002
-#define			APIC_ESR_SEND_ACC	0x00004
-#define			APIC_ESR_RECV_ACC	0x00008
-#define			APIC_ESR_SENDILL	0x00020
-#define			APIC_ESR_RECVILL	0x00040
-#define			APIC_ESR_ILLREGA	0x00080
-#define		APIC_ICR	0x300
-#define			APIC_DEST_SELF		0x40000
-#define			APIC_DEST_ALLINC	0x80000
-#define			APIC_DEST_ALLBUT	0xC0000
-#define			APIC_ICR_RR_MASK	0x30000
-#define			APIC_ICR_RR_INVALID	0x00000
-#define			APIC_ICR_RR_INPROG	0x10000
-#define			APIC_ICR_RR_VALID	0x20000
-#define			APIC_INT_LEVELTRIG	0x08000
-#define			APIC_INT_ASSERT		0x04000
-#define			APIC_ICR_BUSY		0x01000
-#define			APIC_DEST_LOGICAL	0x00800
-#define			APIC_DM_FIXED		0x00000
-#define			APIC_DM_LOWEST		0x00100
-#define			APIC_DM_SMI		0x00200
-#define			APIC_DM_REMRD		0x00300
-#define			APIC_DM_NMI		0x00400
-#define			APIC_DM_INIT		0x00500
-#define			APIC_DM_STARTUP		0x00600
-#define			APIC_DM_EXTINT		0x00700
-#define			APIC_VECTOR_MASK	0x000FF
-#define		APIC_ICR2	0x310
-#define			GET_APIC_DEST_FIELD(x)	(((x)>>24)&0xFF)
-#define			SET_APIC_DEST_FIELD(x)	((x)<<24)
-#define		APIC_LVTT	0x320
-#define		APIC_LVTTHMR	0x330
-#define		APIC_LVTPC	0x340
-#define		APIC_LVT0	0x350
-#define			APIC_LVT_TIMER_BASE_MASK	(0x3<<18)
-#define			GET_APIC_TIMER_BASE(x)		(((x)>>18)&0x3)
-#define			SET_APIC_TIMER_BASE(x)		(((x)<<18))
-#define			APIC_TIMER_BASE_CLKIN		0x0
-#define			APIC_TIMER_BASE_TMBASE		0x1
-#define			APIC_TIMER_BASE_DIV		0x2
-#define			APIC_LVT_TIMER_PERIODIC		(1<<17)
-#define			APIC_LVT_MASKED			(1<<16)
-#define			APIC_LVT_LEVEL_TRIGGER		(1<<15)
-#define			APIC_LVT_REMOTE_IRR		(1<<14)
-#define			APIC_INPUT_POLARITY		(1<<13)
-#define			APIC_SEND_PENDING		(1<<12)
-#define			APIC_MODE_MASK			0x700
-#define			GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
-#define			SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
-#define				APIC_MODE_FIXED		0x0
-#define				APIC_MODE_NMI		0x4
-#define				APIC_MODE_EXTINT	0x7
-#define 	APIC_LVT1	0x360
-#define		APIC_LVTERR	0x370
-#define		APIC_TMICT	0x380
-#define		APIC_TMCCT	0x390
-#define		APIC_TDCR	0x3E0
-#define			APIC_TDR_DIV_TMBASE	(1<<2)
-#define			APIC_TDR_DIV_1		0xB
-#define			APIC_TDR_DIV_2		0x0
-#define			APIC_TDR_DIV_4		0x1
-#define			APIC_TDR_DIV_8		0x2
-#define			APIC_TDR_DIV_16		0x3
-#define			APIC_TDR_DIV_32		0x8
-#define			APIC_TDR_DIV_64		0x9
-#define			APIC_TDR_DIV_128	0xA
-
-#define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
-
-#define MAX_IO_APICS 64
-
-/*
- * the local APIC register structure, memory mapped. Not terribly well
- * tested, but we might eventually use this one in the future - the
- * problem why we cannot use it right now is the P5 APIC, it has an
- * errata which cannot take 8-bit reads and writes, only 32-bit ones ...
- */
-#define u32 unsigned int
-
-
-struct local_apic {
-
-/*000*/	struct { u32 __reserved[4]; } __reserved_01;
-
-/*010*/	struct { u32 __reserved[4]; } __reserved_02;
-
-/*020*/	struct { /* APIC ID Register */
-		u32   __reserved_1	: 24,
-			phys_apic_id	:  4,
-			__reserved_2	:  4;
-		u32 __reserved[3];
-	} id;
-
-/*030*/	const
-	struct { /* APIC Version Register */
-		u32   version		:  8,
-			__reserved_1	:  8,
-			max_lvt		:  8,
-			__reserved_2	:  8;
-		u32 __reserved[3];
-	} version;
-
-/*040*/	struct { u32 __reserved[4]; } __reserved_03;
-
-/*050*/	struct { u32 __reserved[4]; } __reserved_04;
-
-/*060*/	struct { u32 __reserved[4]; } __reserved_05;
-
-/*070*/	struct { u32 __reserved[4]; } __reserved_06;
-
-/*080*/	struct { /* Task Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} tpr;
-
-/*090*/	const
-	struct { /* Arbitration Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} apr;
-
-/*0A0*/	const
-	struct { /* Processor Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} ppr;
-
-/*0B0*/	struct { /* End Of Interrupt Register */
-		u32   eoi;
-		u32 __reserved[3];
-	} eoi;
-
-/*0C0*/	struct { u32 __reserved[4]; } __reserved_07;
-
-/*0D0*/	struct { /* Logical Destination Register */
-		u32   __reserved_1	: 24,
-			logical_dest	:  8;
-		u32 __reserved_2[3];
-	} ldr;
-
-/*0E0*/	struct { /* Destination Format Register */
-		u32   __reserved_1	: 28,
-			model		:  4;
-		u32 __reserved_2[3];
-	} dfr;
-
-/*0F0*/	struct { /* Spurious Interrupt Vector Register */
-		u32	spurious_vector	:  8,
-			apic_enabled	:  1,
-			focus_cpu	:  1,
-			__reserved_2	: 22;
-		u32 __reserved_3[3];
-	} svr;
-
-/*100*/	struct { /* In Service Register */
-/*170*/		u32 bitfield;
-		u32 __reserved[3];
-	} isr [8];
-
-/*180*/	struct { /* Trigger Mode Register */
-/*1F0*/		u32 bitfield;
-		u32 __reserved[3];
-	} tmr [8];
-
-/*200*/	struct { /* Interrupt Request Register */
-/*270*/		u32 bitfield;
-		u32 __reserved[3];
-	} irr [8];
-
-/*280*/	union { /* Error Status Register */
-		struct {
-			u32   send_cs_error			:  1,
-				receive_cs_error		:  1,
-				send_accept_error		:  1,
-				receive_accept_error		:  1,
-				__reserved_1			:  1,
-				send_illegal_vector		:  1,
-				receive_illegal_vector		:  1,
-				illegal_register_address	:  1,
-				__reserved_2			: 24;
-			u32 __reserved_3[3];
-		} error_bits;
-		struct {
-			u32 errors;
-			u32 __reserved_3[3];
-		} all_errors;
-	} esr;
-
-/*290*/	struct { u32 __reserved[4]; } __reserved_08;
-
-/*2A0*/	struct { u32 __reserved[4]; } __reserved_09;
-
-/*2B0*/	struct { u32 __reserved[4]; } __reserved_10;
-
-/*2C0*/	struct { u32 __reserved[4]; } __reserved_11;
-
-/*2D0*/	struct { u32 __reserved[4]; } __reserved_12;
-
-/*2E0*/	struct { u32 __reserved[4]; } __reserved_13;
-
-/*2F0*/	struct { u32 __reserved[4]; } __reserved_14;
-
-/*300*/	struct { /* Interrupt Command Register 1 */
-		u32   vector			:  8,
-			delivery_mode		:  3,
-			destination_mode	:  1,
-			delivery_status		:  1,
-			__reserved_1		:  1,
-			level			:  1,
-			trigger			:  1,
-			__reserved_2		:  2,
-			shorthand		:  2,
-			__reserved_3		:  12;
-		u32 __reserved_4[3];
-	} icr1;
-
-/*310*/	struct { /* Interrupt Command Register 2 */
-		union {
-			u32   __reserved_1	: 24,
-				phys_dest	:  4,
-				__reserved_2	:  4;
-			u32   __reserved_3	: 24,
-				logical_dest	:  8;
-		} dest;
-		u32 __reserved_4[3];
-	} icr2;
-
-/*320*/	struct { /* LVT - Timer */
-		u32   vector		:  8,
-			__reserved_1	:  4,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			timer_mode	:  1,
-			__reserved_3	: 14;
-		u32 __reserved_4[3];
-	} lvt_timer;
-
-/*330*/	struct { /* LVT - Thermal Sensor */
-		u32  vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_thermal;
-
-/*340*/	struct { /* LVT - Performance Counter */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_pc;
-
-/*350*/	struct { /* LVT - LINT0 */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			polarity	:  1,
-			remote_irr	:  1,
-			trigger		:  1,
-			mask		:  1,
-			__reserved_2	: 15;
-		u32 __reserved_3[3];
-	} lvt_lint0;
-
-/*360*/	struct { /* LVT - LINT1 */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			polarity	:  1,
-			remote_irr	:  1,
-			trigger		:  1,
-			mask		:  1,
-			__reserved_2	: 15;
-		u32 __reserved_3[3];
-	} lvt_lint1;
-
-/*370*/	struct { /* LVT - Error */
-		u32   vector		:  8,
-			__reserved_1	:  4,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_error;
-
-/*380*/	struct { /* Timer Initial Count Register */
-		u32   initial_count;
-		u32 __reserved_2[3];
-	} timer_icr;
-
-/*390*/	const
-	struct { /* Timer Current Count Register */
-		u32   curr_count;
-		u32 __reserved_2[3];
-	} timer_ccr;
-
-/*3A0*/	struct { u32 __reserved[4]; } __reserved_16;
-
-/*3B0*/	struct { u32 __reserved[4]; } __reserved_17;
-
-/*3C0*/	struct { u32 __reserved[4]; } __reserved_18;
-
-/*3D0*/	struct { u32 __reserved[4]; } __reserved_19;
-
-/*3E0*/	struct { /* Timer Divide Configuration Register */
-		u32   divisor		:  4,
-			__reserved_1	: 28;
-		u32 __reserved_2[3];
-	} timer_dcr;
-
-/*3F0*/	struct { u32 __reserved[4]; } __reserved_20;
-
-} __attribute__ ((packed));
-
-#undef u32
-
-#endif
diff --git a/include/asm-x86/apicdef_64.h b/include/asm-x86/apicdef_64.h
deleted file mode 100644
index 1dd40067c67c..000000000000
--- a/include/asm-x86/apicdef_64.h
+++ /dev/null
@@ -1,392 +0,0 @@
-#ifndef __ASM_APICDEF_H
-#define __ASM_APICDEF_H
-
-/*
- * Constants for various Intel APICs. (local APIC, IOAPIC, etc.)
- *
- * Alan Cox <Alan.Cox@linux.org>, 1995.
- * Ingo Molnar <mingo@redhat.com>, 1999, 2000
- */
-
-#define		APIC_DEFAULT_PHYS_BASE	0xfee00000
- 
-#define		APIC_ID		0x20
-#define			APIC_ID_MASK		(0xFFu<<24)
-#define			GET_APIC_ID(x)		(((x)>>24)&0xFFu)
-#define			SET_APIC_ID(x)		(((x)<<24))
-#define		APIC_LVR	0x30
-#define			APIC_LVR_MASK		0xFF00FF
-#define			GET_APIC_VERSION(x)	((x)&0xFFu)
-#define			GET_APIC_MAXLVT(x)	(((x)>>16)&0xFFu)
-#define			APIC_INTEGRATED(x)	((x)&0xF0u)
-#define		APIC_TASKPRI	0x80
-#define			APIC_TPRI_MASK		0xFFu
-#define		APIC_ARBPRI	0x90
-#define			APIC_ARBPRI_MASK	0xFFu
-#define		APIC_PROCPRI	0xA0
-#define		APIC_EOI	0xB0
-#define			APIC_EIO_ACK		0x0		/* Write this to the EOI register */
-#define		APIC_RRR	0xC0
-#define		APIC_LDR	0xD0
-#define			APIC_LDR_MASK		(0xFFu<<24)
-#define			GET_APIC_LOGICAL_ID(x)	(((x)>>24)&0xFFu)
-#define			SET_APIC_LOGICAL_ID(x)	(((x)<<24))
-#define			APIC_ALL_CPUS		0xFFu
-#define		APIC_DFR	0xE0
-#define			APIC_DFR_CLUSTER		0x0FFFFFFFul
-#define			APIC_DFR_FLAT			0xFFFFFFFFul
-#define		APIC_SPIV	0xF0
-#define			APIC_SPIV_FOCUS_DISABLED	(1<<9)
-#define			APIC_SPIV_APIC_ENABLED		(1<<8)
-#define		APIC_ISR	0x100
-#define		APIC_ISR_NR	0x8	/* Number of 32 bit ISR registers. */
-#define		APIC_TMR	0x180
-#define 	APIC_IRR	0x200
-#define 	APIC_ESR	0x280
-#define			APIC_ESR_SEND_CS	0x00001
-#define			APIC_ESR_RECV_CS	0x00002
-#define			APIC_ESR_SEND_ACC	0x00004
-#define			APIC_ESR_RECV_ACC	0x00008
-#define			APIC_ESR_SENDILL	0x00020
-#define			APIC_ESR_RECVILL	0x00040
-#define			APIC_ESR_ILLREGA	0x00080
-#define		APIC_ICR	0x300
-#define			APIC_DEST_SELF		0x40000
-#define			APIC_DEST_ALLINC	0x80000
-#define			APIC_DEST_ALLBUT	0xC0000
-#define			APIC_ICR_RR_MASK	0x30000
-#define			APIC_ICR_RR_INVALID	0x00000
-#define			APIC_ICR_RR_INPROG	0x10000
-#define			APIC_ICR_RR_VALID	0x20000
-#define			APIC_INT_LEVELTRIG	0x08000
-#define			APIC_INT_ASSERT		0x04000
-#define			APIC_ICR_BUSY		0x01000
-#define			APIC_DEST_LOGICAL	0x00800
-#define			APIC_DEST_PHYSICAL	0x00000
-#define			APIC_DM_FIXED		0x00000
-#define			APIC_DM_LOWEST		0x00100
-#define			APIC_DM_SMI		0x00200
-#define			APIC_DM_REMRD		0x00300
-#define			APIC_DM_NMI		0x00400
-#define			APIC_DM_INIT		0x00500
-#define			APIC_DM_STARTUP		0x00600
-#define			APIC_DM_EXTINT		0x00700
-#define			APIC_VECTOR_MASK	0x000FF
-#define		APIC_ICR2	0x310
-#define			GET_APIC_DEST_FIELD(x)	(((x)>>24)&0xFF)
-#define			SET_APIC_DEST_FIELD(x)	((x)<<24)
-#define		APIC_LVTT	0x320
-#define		APIC_LVTTHMR	0x330
-#define		APIC_LVTPC	0x340
-#define		APIC_LVT0	0x350
-#define			APIC_LVT_TIMER_BASE_MASK	(0x3<<18)
-#define			GET_APIC_TIMER_BASE(x)		(((x)>>18)&0x3)
-#define			SET_APIC_TIMER_BASE(x)		(((x)<<18))
-#define			APIC_TIMER_BASE_CLKIN		0x0
-#define			APIC_TIMER_BASE_TMBASE		0x1
-#define			APIC_TIMER_BASE_DIV		0x2
-#define			APIC_LVT_TIMER_PERIODIC		(1<<17)
-#define			APIC_LVT_MASKED			(1<<16)
-#define			APIC_LVT_LEVEL_TRIGGER		(1<<15)
-#define			APIC_LVT_REMOTE_IRR		(1<<14)
-#define			APIC_INPUT_POLARITY		(1<<13)
-#define			APIC_SEND_PENDING		(1<<12)
-#define			APIC_MODE_MASK			0x700
-#define			GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
-#define			SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
-#define				APIC_MODE_FIXED		0x0
-#define				APIC_MODE_NMI		0x4
-#define				APIC_MODE_EXTINT	0x7
-#define 	APIC_LVT1	0x360
-#define		APIC_LVTERR	0x370
-#define		APIC_TMICT	0x380
-#define		APIC_TMCCT	0x390
-#define		APIC_TDCR	0x3E0
-#define			APIC_TDR_DIV_TMBASE	(1<<2)
-#define			APIC_TDR_DIV_1		0xB
-#define			APIC_TDR_DIV_2		0x0
-#define			APIC_TDR_DIV_4		0x1
-#define			APIC_TDR_DIV_8		0x2
-#define			APIC_TDR_DIV_16		0x3
-#define			APIC_TDR_DIV_32		0x8
-#define			APIC_TDR_DIV_64		0x9
-#define			APIC_TDR_DIV_128	0xA
-
-#define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
-
-#define MAX_IO_APICS 128
-#define MAX_LOCAL_APIC 256
-
-/*
- * All x86-64 systems are xAPIC compatible.
- * In the following, "apicid" is a physical APIC ID.
- */
-#define XAPIC_DEST_CPUS_SHIFT	4
-#define XAPIC_DEST_CPUS_MASK	((1u << XAPIC_DEST_CPUS_SHIFT) - 1)
-#define XAPIC_DEST_CLUSTER_MASK	(XAPIC_DEST_CPUS_MASK << XAPIC_DEST_CPUS_SHIFT)
-#define APIC_CLUSTER(apicid)	((apicid) & XAPIC_DEST_CLUSTER_MASK)
-#define APIC_CLUSTERID(apicid)	(APIC_CLUSTER(apicid) >> XAPIC_DEST_CPUS_SHIFT)
-#define APIC_CPUID(apicid)	((apicid) & XAPIC_DEST_CPUS_MASK)
-#define NUM_APIC_CLUSTERS	((BAD_APICID + 1) >> XAPIC_DEST_CPUS_SHIFT)
-
-/*
- * the local APIC register structure, memory mapped. Not terribly well
- * tested, but we might eventually use this one in the future - the
- * problem why we cannot use it right now is the P5 APIC, it has an
- * errata which cannot take 8-bit reads and writes, only 32-bit ones ...
- */
-#define u32 unsigned int
-
-struct local_apic {
-
-/*000*/	struct { u32 __reserved[4]; } __reserved_01;
-
-/*010*/	struct { u32 __reserved[4]; } __reserved_02;
-
-/*020*/	struct { /* APIC ID Register */
-		u32   __reserved_1	: 24,
-			phys_apic_id	:  4,
-			__reserved_2	:  4;
-		u32 __reserved[3];
-	} id;
-
-/*030*/	const
-	struct { /* APIC Version Register */
-		u32   version		:  8,
-			__reserved_1	:  8,
-			max_lvt		:  8,
-			__reserved_2	:  8;
-		u32 __reserved[3];
-	} version;
-
-/*040*/	struct { u32 __reserved[4]; } __reserved_03;
-
-/*050*/	struct { u32 __reserved[4]; } __reserved_04;
-
-/*060*/	struct { u32 __reserved[4]; } __reserved_05;
-
-/*070*/	struct { u32 __reserved[4]; } __reserved_06;
-
-/*080*/	struct { /* Task Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} tpr;
-
-/*090*/	const
-	struct { /* Arbitration Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} apr;
-
-/*0A0*/	const
-	struct { /* Processor Priority Register */
-		u32   priority	:  8,
-			__reserved_1	: 24;
-		u32 __reserved_2[3];
-	} ppr;
-
-/*0B0*/	struct { /* End Of Interrupt Register */
-		u32   eoi;
-		u32 __reserved[3];
-	} eoi;
-
-/*0C0*/	struct { u32 __reserved[4]; } __reserved_07;
-
-/*0D0*/	struct { /* Logical Destination Register */
-		u32   __reserved_1	: 24,
-			logical_dest	:  8;
-		u32 __reserved_2[3];
-	} ldr;
-
-/*0E0*/	struct { /* Destination Format Register */
-		u32   __reserved_1	: 28,
-			model		:  4;
-		u32 __reserved_2[3];
-	} dfr;
-
-/*0F0*/	struct { /* Spurious Interrupt Vector Register */
-		u32	spurious_vector	:  8,
-			apic_enabled	:  1,
-			focus_cpu	:  1,
-			__reserved_2	: 22;
-		u32 __reserved_3[3];
-	} svr;
-
-/*100*/	struct { /* In Service Register */
-/*170*/		u32 bitfield;
-		u32 __reserved[3];
-	} isr [8];
-
-/*180*/	struct { /* Trigger Mode Register */
-/*1F0*/		u32 bitfield;
-		u32 __reserved[3];
-	} tmr [8];
-
-/*200*/	struct { /* Interrupt Request Register */
-/*270*/		u32 bitfield;
-		u32 __reserved[3];
-	} irr [8];
-
-/*280*/	union { /* Error Status Register */
-		struct {
-			u32   send_cs_error			:  1,
-				receive_cs_error		:  1,
-				send_accept_error		:  1,
-				receive_accept_error		:  1,
-				__reserved_1			:  1,
-				send_illegal_vector		:  1,
-				receive_illegal_vector		:  1,
-				illegal_register_address	:  1,
-				__reserved_2			: 24;
-			u32 __reserved_3[3];
-		} error_bits;
-		struct {
-			u32 errors;
-			u32 __reserved_3[3];
-		} all_errors;
-	} esr;
-
-/*290*/	struct { u32 __reserved[4]; } __reserved_08;
-
-/*2A0*/	struct { u32 __reserved[4]; } __reserved_09;
-
-/*2B0*/	struct { u32 __reserved[4]; } __reserved_10;
-
-/*2C0*/	struct { u32 __reserved[4]; } __reserved_11;
-
-/*2D0*/	struct { u32 __reserved[4]; } __reserved_12;
-
-/*2E0*/	struct { u32 __reserved[4]; } __reserved_13;
-
-/*2F0*/	struct { u32 __reserved[4]; } __reserved_14;
-
-/*300*/	struct { /* Interrupt Command Register 1 */
-		u32   vector			:  8,
-			delivery_mode		:  3,
-			destination_mode	:  1,
-			delivery_status		:  1,
-			__reserved_1		:  1,
-			level			:  1,
-			trigger			:  1,
-			__reserved_2		:  2,
-			shorthand		:  2,
-			__reserved_3		:  12;
-		u32 __reserved_4[3];
-	} icr1;
-
-/*310*/	struct { /* Interrupt Command Register 2 */
-		union {
-			u32   __reserved_1	: 24,
-				phys_dest	:  4,
-				__reserved_2	:  4;
-			u32   __reserved_3	: 24,
-				logical_dest	:  8;
-		} dest;
-		u32 __reserved_4[3];
-	} icr2;
-
-/*320*/	struct { /* LVT - Timer */
-		u32   vector		:  8,
-			__reserved_1	:  4,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			timer_mode	:  1,
-			__reserved_3	: 14;
-		u32 __reserved_4[3];
-	} lvt_timer;
-
-/*330*/	struct { /* LVT - Thermal Sensor */
-		u32  vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_thermal;
-
-/*340*/	struct { /* LVT - Performance Counter */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_pc;
-
-/*350*/	struct { /* LVT - LINT0 */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			polarity	:  1,
-			remote_irr	:  1,
-			trigger		:  1,
-			mask		:  1,
-			__reserved_2	: 15;
-		u32 __reserved_3[3];
-	} lvt_lint0;
-
-/*360*/	struct { /* LVT - LINT1 */
-		u32   vector		:  8,
-			delivery_mode	:  3,
-			__reserved_1	:  1,
-			delivery_status	:  1,
-			polarity	:  1,
-			remote_irr	:  1,
-			trigger		:  1,
-			mask		:  1,
-			__reserved_2	: 15;
-		u32 __reserved_3[3];
-	} lvt_lint1;
-
-/*370*/	struct { /* LVT - Error */
-		u32   vector		:  8,
-			__reserved_1	:  4,
-			delivery_status	:  1,
-			__reserved_2	:  3,
-			mask		:  1,
-			__reserved_3	: 15;
-		u32 __reserved_4[3];
-	} lvt_error;
-
-/*380*/	struct { /* Timer Initial Count Register */
-		u32   initial_count;
-		u32 __reserved_2[3];
-	} timer_icr;
-
-/*390*/	const
-	struct { /* Timer Current Count Register */
-		u32   curr_count;
-		u32 __reserved_2[3];
-	} timer_ccr;
-
-/*3A0*/	struct { u32 __reserved[4]; } __reserved_16;
-
-/*3B0*/	struct { u32 __reserved[4]; } __reserved_17;
-
-/*3C0*/	struct { u32 __reserved[4]; } __reserved_18;
-
-/*3D0*/	struct { u32 __reserved[4]; } __reserved_19;
-
-/*3E0*/	struct { /* Timer Divide Configuration Register */
-		u32   divisor		:  4,
-			__reserved_1	: 28;
-		u32 __reserved_2[3];
-	} timer_dcr;
-
-/*3F0*/	struct { u32 __reserved[4]; } __reserved_20;
-
-} __attribute__ ((packed));
-
-#undef u32
-
-#define BAD_APICID 0xFFu
-
-#endif
-- 
cgit v1.2.3


From cff90dbfe41f2c8eabe2d0a7e379829f4a9980af Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:14 +0100
Subject: x86: prepare unification of include/asm-x86/apic_32/64.h

White space and coding style clenaup.

Move the K8 local apic defines to apicdef.h, where they belong

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apic_32.h | 49 ++++++++++++++++++++++-------------------------
 include/asm-x86/apic_64.h | 39 ++++++++++++++++---------------------
 include/asm-x86/apicdef.h | 10 +++++++++-
 3 files changed, 48 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/apic_32.h b/include/asm-x86/apic_32.h
index be158b27d54b..f909e2daf226 100644
--- a/include/asm-x86/apic_32.h
+++ b/include/asm-x86/apic_32.h
@@ -17,8 +17,6 @@
 #define APIC_VERBOSE 1
 #define APIC_DEBUG   2
 
-extern int apic_verbosity;
-
 /*
  * Define the default level of output to be very little
  * This can be turned up by using apic=verbose for more
@@ -35,6 +33,11 @@ extern void generic_apic_probe(void);
 
 #ifdef CONFIG_X86_LOCAL_APIC
 
+extern int apic_verbosity;
+extern int timer_over_8254;
+extern int local_apic_timer_c2_ok;
+extern int local_apic_timer_disabled;
+
 /*
  * Basic functions accessing APICs.
  */
@@ -65,9 +68,9 @@ static __inline fastcall unsigned long native_apic_read(unsigned long reg)
 	return *((volatile unsigned long *)(APIC_BASE+reg));
 }
 
-void apic_wait_icr_idle(void);
-unsigned long safe_apic_wait_icr_idle(void);
-int get_physical_broadcast(void);
+extern void apic_wait_icr_idle(void);
+extern unsigned long safe_apic_wait_icr_idle(void);
+extern int get_physical_broadcast(void);
 
 #ifdef CONFIG_X86_GOOD_APIC
 # define FORCE_READ_AROUND_WRITE 0
@@ -94,30 +97,24 @@ static inline void ack_APIC_irq(void)
 
 extern int lapic_get_maxlvt(void);
 extern void clear_local_APIC(void);
-extern void connect_bsp_APIC (void);
-extern void disconnect_bsp_APIC (int virt_wire_setup);
-extern void disable_local_APIC (void);
-extern void lapic_shutdown (void);
-extern int verify_local_APIC (void);
-extern void cache_APIC_registers (void);
-extern void sync_Arb_IDs (void);
-extern void init_bsp_APIC (void);
-extern void setup_local_APIC (void);
-extern void init_apic_mappings (void);
-extern void smp_local_timer_interrupt (void);
-extern void setup_boot_APIC_clock (void);
-extern void setup_secondary_APIC_clock (void);
-extern int APIC_init_uniprocessor (void);
-
-extern void enable_NMI_through_LVT0 (void * dummy);
+extern void connect_bsp_APIC(void);
+extern void disconnect_bsp_APIC(int virt_wire_setup);
+extern void disable_local_APIC(void);
+extern void lapic_shutdown(void);
+extern int verify_local_APIC(void);
+extern void cache_APIC_registers(void);
+extern void sync_Arb_IDs(void);
+extern void init_bsp_APIC(void);
+extern void setup_local_APIC(void);
+extern void init_apic_mappings(void);
+extern void smp_local_timer_interrupt(void);
+extern void setup_boot_APIC_clock(void);
+extern void setup_secondary_APIC_clock(void);
+extern int APIC_init_uniprocessor(void);
+extern void enable_NMI_through_LVT0(void *dummy);
 
 #define ARCH_APICTIMER_STOPS_ON_C3	1
 
-extern int timer_over_8254;
-extern int local_apic_timer_c2_ok;
-
-extern int local_apic_timer_disabled;
-
 #else /* !CONFIG_X86_LOCAL_APIC */
 static inline void lapic_shutdown(void) { }
 #define local_apic_timer_c2_ok		1
diff --git a/include/asm-x86/apic_64.h b/include/asm-x86/apic_64.h
index 2747a11a2b19..9a0ec02a49a1 100644
--- a/include/asm-x86/apic_64.h
+++ b/include/asm-x86/apic_64.h
@@ -64,22 +64,22 @@ static inline void ack_APIC_irq(void)
 	apic_write(APIC_EOI, 0);
 }
 
-extern int get_maxlvt (void);
-extern void clear_local_APIC (void);
-extern void connect_bsp_APIC (void);
-extern void disconnect_bsp_APIC (int virt_wire_setup);
-extern void disable_local_APIC (void);
-extern void lapic_shutdown (void);
-extern int verify_local_APIC (void);
-extern void cache_APIC_registers (void);
-extern void sync_Arb_IDs (void);
-extern void init_bsp_APIC (void);
-extern void setup_local_APIC (void);
-extern void init_apic_mappings (void);
-extern void smp_local_timer_interrupt (void);
-extern void setup_boot_APIC_clock (void);
-extern void setup_secondary_APIC_clock (void);
-extern int APIC_init_uniprocessor (void);
+extern int get_maxlvt(void);
+extern void clear_local_APIC(void);
+extern void connect_bsp_APIC(void);
+extern void disconnect_bsp_APIC(int virt_wire_setup);
+extern void disable_local_APIC(void);
+extern void lapic_shutdown(void);
+extern int verify_local_APIC(void);
+extern void cache_APIC_registers(void);
+extern void sync_Arb_IDs(void);
+extern void init_bsp_APIC(void);
+extern void setup_local_APIC(void);
+extern void init_apic_mappings(void);
+extern void smp_local_timer_interrupt(void);
+extern void setup_boot_APIC_clock(void);
+extern void setup_secondary_APIC_clock(void);
+extern int APIC_init_uniprocessor(void);
 extern void setup_apic_routing(void);
 
 extern void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
@@ -87,13 +87,6 @@ extern void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
 
 extern int apic_is_clustered_box(void);
 
-#define K8_APIC_EXT_LVT_BASE    0x500
-#define K8_APIC_EXT_INT_MSG_FIX 0x0
-#define K8_APIC_EXT_INT_MSG_SMI 0x2
-#define K8_APIC_EXT_INT_MSG_NMI 0x4
-#define K8_APIC_EXT_INT_MSG_EXT 0x7
-#define K8_APIC_EXT_LVT_ENTRY_THRESHOLD    0
-
 #define ARCH_APICTIMER_STOPS_ON_C3	1
 
 extern unsigned boot_cpu_id;
diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 83ac1e6e3625..5f7abe9b5f87 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -98,7 +98,7 @@
 #define		APIC_SEND_PENDING		(1<<12)
 #define		APIC_MODE_MASK			0x700
 #define		GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
-#define		SET_APIC_DELIVERY_MODE(x,y)	(((x)&~0x700)|((y)<<8))
+#define		SET_APIC_DELIVERY_MODE(x, y)	(((x)&~0x700)|((y)<<8))
 #define			APIC_MODE_FIXED		0x0
 #define			APIC_MODE_NMI		0x4
 #define			APIC_MODE_EXTINT	0x7
@@ -116,6 +116,14 @@
 #define		APIC_TDR_DIV_32		0x8
 #define		APIC_TDR_DIV_64		0x9
 #define		APIC_TDR_DIV_128	0xA
+
+#define K8_APIC_EXT_LVT_BASE		0x500
+#define K8_APIC_EXT_INT_MSG_FIX		0x0
+#define K8_APIC_EXT_INT_MSG_SMI		0x2
+#define K8_APIC_EXT_INT_MSG_NMI		0x4
+#define K8_APIC_EXT_INT_MSG_EXT		0x7
+#define K8_APIC_EXT_LVT_ENTRY_THRESHOLD	0
+
 #define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
 
 #ifdef CONFIG_X86_32
-- 
cgit v1.2.3


From 37e650c7c8a27de533d409b53c29f4135dcc7af6 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:14 +0100
Subject: x86: rename get_maxlvt to lapic_get_maxlvt

Use the same name for the 32 and 64 bit variant.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    | 12 ++++++------
 arch/x86/kernel/io_apic_64.c |  2 +-
 arch/x86/kernel/smpboot_64.c |  2 +-
 include/asm-x86/apic_64.h    |  2 +-
 4 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index fa6cdee6d303..dfeda91fa80c 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -160,7 +160,7 @@ void enable_NMI_through_LVT0 (void * dummy)
 	apic_write(APIC_LVT0, v);
 }
 
-int get_maxlvt(void)
+int lapic_get_maxlvt(void)
 {
 	unsigned int v, maxlvt;
 
@@ -194,7 +194,7 @@ void clear_local_APIC(void)
 	int maxlvt;
 	unsigned int v;
 
-	maxlvt = get_maxlvt();
+	maxlvt = lapic_get_maxlvt();
 
 	/*
 	 * Masking an LVT entry can trigger a local APIC error
@@ -333,7 +333,7 @@ int __init verify_local_APIC(void)
 	reg1 = GET_APIC_VERSION(reg0);
 	if (reg1 == 0x00 || reg1 == 0xff)
 		return 0;
-	reg1 = get_maxlvt();
+	reg1 = lapic_get_maxlvt();
 	if (reg1 < 0x02 || reg1 == 0xff)
 		return 0;
 
@@ -519,7 +519,7 @@ void __cpuinit setup_local_APIC (void)
 
 	{
 		unsigned oldvalue;
-		maxlvt = get_maxlvt();
+		maxlvt = lapic_get_maxlvt();
 		oldvalue = apic_read(APIC_ESR);
 		value = ERROR_APIC_VECTOR;      // enables sending errors
 		apic_write(APIC_LVTERR, value);
@@ -571,7 +571,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	if (!apic_pm_state.active)
 		return 0;
 
-	maxlvt = get_maxlvt();
+	maxlvt = lapic_get_maxlvt();
 
 	apic_pm_state.apic_id = apic_read(APIC_ID);
 	apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI);
@@ -605,7 +605,7 @@ static int lapic_resume(struct sys_device *dev)
 	if (!apic_pm_state.active)
 		return 0;
 
-	maxlvt = get_maxlvt();
+	maxlvt = lapic_get_maxlvt();
 
 	local_irq_save(flags);
 	rdmsr(MSR_IA32_APICBASE, l, h);
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 23a3ac06a23e..d4f5286101a9 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1069,7 +1069,7 @@ void __apicdebuginit print_local_APIC(void * dummy)
 	v = apic_read(APIC_LVR);
 	printk(KERN_INFO "... APIC VERSION: %08x\n", v);
 	ver = GET_APIC_VERSION(v);
-	maxlvt = get_maxlvt();
+	maxlvt = lapic_get_maxlvt();
 
 	v = apic_read(APIC_TASKPRI);
 	printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index aaf4e1291217..8147b7d4db63 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -466,7 +466,7 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta
 	 */
 	Dprintk("#startup loops: %d.\n", num_starts);
 
-	maxlvt = get_maxlvt();
+	maxlvt = lapic_get_maxlvt();
 
 	for (j = 1; j <= num_starts; j++) {
 		Dprintk("Sending STARTUP #%d.\n",j);
diff --git a/include/asm-x86/apic_64.h b/include/asm-x86/apic_64.h
index 9a0ec02a49a1..b5f850f25114 100644
--- a/include/asm-x86/apic_64.h
+++ b/include/asm-x86/apic_64.h
@@ -64,7 +64,7 @@ static inline void ack_APIC_irq(void)
 	apic_write(APIC_EOI, 0);
 }
 
-extern int get_maxlvt(void);
+extern int lapic_get_maxlvt(void);
 extern void clear_local_APIC(void);
 extern void connect_bsp_APIC(void);
 extern void disconnect_bsp_APIC(int virt_wire_setup);
-- 
cgit v1.2.3


From 3c6bb07ac1b4174318606a26f0de8ceb9f6d8133 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:15 +0100
Subject: x86: use u32 for safe_apic_wait_icr_idle()

Preperatory patch for merging apic headers.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    |  4 ++--
 arch/x86/kernel/smpboot_64.c | 11 ++++++-----
 include/asm-x86/apic_64.h    |  2 +-
 3 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index dfeda91fa80c..3de3764a862c 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -135,9 +135,9 @@ void apic_wait_icr_idle(void)
 		cpu_relax();
 }
 
-unsigned int safe_apic_wait_icr_idle(void)
+u32 safe_apic_wait_icr_idle(void)
 {
-	unsigned int send_status;
+	u32 send_status;
 	int timeout;
 
 	timeout = 0;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 8147b7d4db63..b36d32ff0b39 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -384,19 +384,20 @@ static void inquire_remote_apic(int apicid)
 	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
 	char *names[] = { "ID", "VERSION", "SPIV" };
 	int timeout;
-	unsigned int status;
+	u32 status;
 
 	printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid);
 
 	for (i = 0; i < ARRAY_SIZE(regs); i++) {
-		printk("... APIC #%d %s: ", apicid, names[i]);
+		printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]);
 
 		/*
 		 * Wait for idle.
 		 */
 		status = safe_apic_wait_icr_idle();
 		if (status)
-			printk("a previous APIC delivery may have failed\n");
+			printk(KERN_CONT
+			       "a previous APIC delivery may have failed\n");
 
 		apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
 		apic_write(APIC_ICR, APIC_DM_REMRD | regs[i]);
@@ -410,10 +411,10 @@ static void inquire_remote_apic(int apicid)
 		switch (status) {
 		case APIC_ICR_RR_VALID:
 			status = apic_read(APIC_RRR);
-			printk("%08x\n", status);
+			printk(KERN_CONT "%08x\n", status);
 			break;
 		default:
-			printk("failed\n");
+			printk(KERN_CONT "failed\n");
 		}
 	}
 }
diff --git a/include/asm-x86/apic_64.h b/include/asm-x86/apic_64.h
index b5f850f25114..7bfad0224178 100644
--- a/include/asm-x86/apic_64.h
+++ b/include/asm-x86/apic_64.h
@@ -49,7 +49,7 @@ static __inline unsigned int apic_read(unsigned long reg)
 }
 
 extern void apic_wait_icr_idle(void);
-extern unsigned int safe_apic_wait_icr_idle(void);
+extern u32 safe_apic_wait_icr_idle(void);
 
 static inline void ack_APIC_irq(void)
 {
-- 
cgit v1.2.3


From 42e0a9aa5d467188687c6b705412578e53c14af6 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:15 +0100
Subject: x86: use u32 for some lapic functions

Use u32 so 32 and 64bit have the same interface.

Andrew Morton: xen, lguest build fixes

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c  |  4 ++--
 arch/x86/lguest/boot.c     |  4 ++--
 arch/x86/xen/enlighten.c   |  4 ++--
 include/asm-x86/apic_32.h  | 16 +++++++---------
 include/asm-x86/apic_64.h  |  6 +++---
 include/asm-x86/paravirt.h | 12 ++++++------
 6 files changed, 22 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index a56c782653be..3a069acb270c 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -135,9 +135,9 @@ void apic_wait_icr_idle(void)
 		cpu_relax();
 }
 
-unsigned long safe_apic_wait_icr_idle(void)
+u32 safe_apic_wait_icr_idle(void)
 {
-	unsigned long send_status;
+	u32 send_status;
 	int timeout;
 
 	timeout = 0;
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index 92c56117eae5..df04bf884dd4 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -788,11 +788,11 @@ static void lguest_wbinvd(void)
  * code qualifies for Advanced.  It will also never interrupt anything.  It
  * does, however, allow us to get through the Linux boot code. */
 #ifdef CONFIG_X86_LOCAL_APIC
-static void lguest_apic_write(unsigned long reg, unsigned long v)
+static void lguest_apic_write(unsigned long reg, u32 v)
 {
 }
 
-static unsigned long lguest_apic_read(unsigned long reg)
+static u32 lguest_apic_read(unsigned long reg)
 {
 	return 0;
 }
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 79ad15252150..00829401389e 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -521,12 +521,12 @@ static void xen_io_delay(void)
 }
 
 #ifdef CONFIG_X86_LOCAL_APIC
-static unsigned long xen_apic_read(unsigned long reg)
+static u32 xen_apic_read(unsigned long reg)
 {
 	return 0;
 }
 
-static void xen_apic_write(unsigned long reg, unsigned long val)
+static void xen_apic_write(unsigned long reg, u32 val)
 {
 	/* Warn to see if there's any stray references */
 	WARN_ON(1);
diff --git a/include/asm-x86/apic_32.h b/include/asm-x86/apic_32.h
index f909e2daf226..649e9a6f6683 100644
--- a/include/asm-x86/apic_32.h
+++ b/include/asm-x86/apic_32.h
@@ -51,25 +51,23 @@ extern int local_apic_timer_disabled;
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
-static __inline fastcall void native_apic_write(unsigned long reg,
-						unsigned long v)
+static __inline fastcall void native_apic_write(unsigned long reg, u32 v)
 {
-	*((volatile unsigned long *)(APIC_BASE+reg)) = v;
+	*((volatile u32 *)(APIC_BASE + reg)) = v;
 }
 
-static __inline fastcall void native_apic_write_atomic(unsigned long reg,
-						       unsigned long v)
+static __inline fastcall void native_apic_write_atomic(unsigned long reg, u32 v)
 {
-	xchg((volatile unsigned long *)(APIC_BASE+reg), v);
+	(void) xchg((u32 *)(APIC_BASE + reg), v);
 }
 
-static __inline fastcall unsigned long native_apic_read(unsigned long reg)
+static __inline fastcall u32 native_apic_read(unsigned long reg)
 {
-	return *((volatile unsigned long *)(APIC_BASE+reg));
+	return *((volatile u32 *)(APIC_BASE + reg));
 }
 
 extern void apic_wait_icr_idle(void);
-extern unsigned long safe_apic_wait_icr_idle(void);
+extern u32 safe_apic_wait_icr_idle(void);
 extern int get_physical_broadcast(void);
 
 #ifdef CONFIG_X86_GOOD_APIC
diff --git a/include/asm-x86/apic_64.h b/include/asm-x86/apic_64.h
index 7bfad0224178..9d0c06c4df91 100644
--- a/include/asm-x86/apic_64.h
+++ b/include/asm-x86/apic_64.h
@@ -38,14 +38,14 @@ struct pt_regs;
  * Basic functions accessing APICs.
  */
 
-static __inline void apic_write(unsigned long reg, unsigned int v)
+static __inline void apic_write(unsigned long reg, u32 v)
 {
 	*((volatile unsigned int *)(APIC_BASE+reg)) = v;
 }
 
-static __inline unsigned int apic_read(unsigned long reg)
+static __inline u32 apic_read(unsigned long reg)
 {
-	return *((volatile unsigned int *)(APIC_BASE+reg));
+	return *((volatile u32 *)(APIC_BASE+reg));
 }
 
 extern void apic_wait_icr_idle(void);
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index f59d370c5df4..19fd3e67b08c 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -150,9 +150,9 @@ struct pv_apic_ops {
 	 * Direct APIC operations, principally for VMI.  Ideally
 	 * these shouldn't be in this interface.
 	 */
-	void (*apic_write)(unsigned long reg, unsigned long v);
-	void (*apic_write_atomic)(unsigned long reg, unsigned long v);
-	unsigned long (*apic_read)(unsigned long reg);
+	void (*apic_write)(unsigned long reg, u32 v);
+	void (*apic_write_atomic)(unsigned long reg, u32 v);
+	u32 (*apic_read)(unsigned long reg);
 	void (*setup_boot_clock)(void);
 	void (*setup_secondary_clock)(void);
 
@@ -690,17 +690,17 @@ static inline void slow_down_io(void) {
 /*
  * Basic functions accessing APICs.
  */
-static inline void apic_write(unsigned long reg, unsigned long v)
+static inline void apic_write(unsigned long reg, u32 v)
 {
 	PVOP_VCALL2(pv_apic_ops.apic_write, reg, v);
 }
 
-static inline void apic_write_atomic(unsigned long reg, unsigned long v)
+static inline void apic_write_atomic(unsigned long reg, u32 v)
 {
 	PVOP_VCALL2(pv_apic_ops.apic_write_atomic, reg, v);
 }
 
-static inline unsigned long apic_read(unsigned long reg)
+static inline u32 apic_read(unsigned long reg)
 {
 	return PVOP_CALL1(unsigned long, pv_apic_ops.apic_read, reg);
 }
-- 
cgit v1.2.3


From 67c5fc5c330f63360e26609534b219df1aaa03ca Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:15 +0100
Subject: x86: merge apic_32/64.h

Unify apic.h variants.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apic.h    | 140 +++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/apic_32.h | 122 ----------------------------------------
 include/asm-x86/apic_64.h |  95 -------------------------------
 3 files changed, 137 insertions(+), 220 deletions(-)
 delete mode 100644 include/asm-x86/apic_32.h
 delete mode 100644 include/asm-x86/apic_64.h

(limited to 'include')

diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 9fbcc0bd2ac4..c064c1f84bab 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -1,5 +1,139 @@
-#ifdef CONFIG_X86_32
-# include "apic_32.h"
+#ifndef _ASM_X86_APIC_H
+#define _ASM_X86_APIC_H
+
+#include <linux/pm.h>
+#include <linux/delay.h>
+#include <asm/fixmap.h>
+#include <asm/apicdef.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+
+#define ARCH_APICTIMER_STOPS_ON_C3	1
+
+#define Dprintk(x...)
+
+/*
+ * Debugging macros
+ */
+#define APIC_QUIET   0
+#define APIC_VERBOSE 1
+#define APIC_DEBUG   2
+
+/*
+ * Define the default level of output to be very little
+ * This can be turned up by using apic=verbose for more
+ * information and apic=debug for _lots_ of information.
+ * apic_verbosity is defined in apic.c
+ */
+#define apic_printk(v, s, a...) do {       \
+		if ((v) <= apic_verbosity) \
+			printk(s, ##a);    \
+	} while (0)
+
+
+extern void generic_apic_probe(void);
+
+#ifdef CONFIG_X86_LOCAL_APIC
+
+extern int apic_verbosity;
+extern int timer_over_8254;
+extern int local_apic_timer_c2_ok;
+extern int local_apic_timer_disabled;
+
+extern int apic_runs_main_timer;
+extern int ioapic_force;
+extern int disable_apic_timer;
+extern unsigned boot_cpu_id;
+
+/*
+ * Basic functions accessing APICs.
+ */
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
 #else
-# include "apic_64.h"
+#define apic_write native_apic_write
+#define apic_write_atomic native_apic_write_atomic
+#define apic_read native_apic_read
+#define setup_boot_clock setup_boot_APIC_clock
+#define setup_secondary_clock setup_secondary_APIC_clock
 #endif
+
+static inline fastcall void native_apic_write(unsigned long reg, u32 v)
+{
+	*((volatile u32 *)(APIC_BASE + reg)) = v;
+}
+
+static inline fastcall void native_apic_write_atomic(unsigned long reg, u32 v)
+{
+	(void) xchg((u32*)(APIC_BASE + reg), v);
+}
+
+static inline fastcall u32 native_apic_read(unsigned long reg)
+{
+	return *((volatile u32 *)(APIC_BASE + reg));
+}
+
+extern void apic_wait_icr_idle(void);
+extern u32 safe_apic_wait_icr_idle(void);
+extern int get_physical_broadcast(void);
+
+#ifdef CONFIG_X86_GOOD_APIC
+# define FORCE_READ_AROUND_WRITE 0
+# define apic_read_around(x)
+# define apic_write_around(x, y) apic_write((x), (y))
+#else
+# define FORCE_READ_AROUND_WRITE 1
+# define apic_read_around(x) apic_read(x)
+# define apic_write_around(x, y) apic_write_atomic((x), (y))
+#endif
+
+static inline void ack_APIC_irq(void)
+{
+	/*
+	 * ack_APIC_irq() actually gets compiled as a single instruction:
+	 * - a single rmw on Pentium/82489DX
+	 * - a single write on P6+ cores (CONFIG_X86_GOOD_APIC)
+	 * ... yummie.
+	 */
+
+	/* Docs say use 0 for future compatibility */
+	apic_write_around(APIC_EOI, 0);
+}
+
+extern int lapic_get_maxlvt(void);
+extern void clear_local_APIC(void);
+extern void connect_bsp_APIC(void);
+extern void disconnect_bsp_APIC(int virt_wire_setup);
+extern void disable_local_APIC(void);
+extern void lapic_shutdown(void);
+extern int verify_local_APIC(void);
+extern void cache_APIC_registers(void);
+extern void sync_Arb_IDs(void);
+extern void init_bsp_APIC(void);
+extern void setup_local_APIC(void);
+extern void init_apic_mappings(void);
+extern void smp_local_timer_interrupt(void);
+extern void setup_boot_APIC_clock(void);
+extern void setup_secondary_APIC_clock(void);
+extern int APIC_init_uniprocessor(void);
+extern void enable_NMI_through_LVT0(void *dummy);
+
+/*
+ * On 32bit this is mach-xxx local
+ */
+#ifdef CONFIG_X86_64
+extern void setup_apic_routing(void);
+#endif
+
+extern void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
+				    unsigned char msg_type, unsigned char mask);
+
+extern int apic_is_clustered_box(void);
+
+#else /* !CONFIG_X86_LOCAL_APIC */
+static inline void lapic_shutdown(void) { }
+#define local_apic_timer_c2_ok		1
+
+#endif /* !CONFIG_X86_LOCAL_APIC */
+
+#endif /* __ASM_APIC_H */
diff --git a/include/asm-x86/apic_32.h b/include/asm-x86/apic_32.h
deleted file mode 100644
index 649e9a6f6683..000000000000
--- a/include/asm-x86/apic_32.h
+++ /dev/null
@@ -1,122 +0,0 @@
-#ifndef __ASM_APIC_H
-#define __ASM_APIC_H
-
-#include <linux/pm.h>
-#include <linux/delay.h>
-#include <asm/fixmap.h>
-#include <asm/apicdef.h>
-#include <asm/processor.h>
-#include <asm/system.h>
-
-#define Dprintk(x...)
-
-/*
- * Debugging macros
- */
-#define APIC_QUIET   0
-#define APIC_VERBOSE 1
-#define APIC_DEBUG   2
-
-/*
- * Define the default level of output to be very little
- * This can be turned up by using apic=verbose for more
- * information and apic=debug for _lots_ of information.
- * apic_verbosity is defined in apic.c
- */
-#define apic_printk(v, s, a...) do {       \
-		if ((v) <= apic_verbosity) \
-			printk(s, ##a);    \
-	} while (0)
-
-
-extern void generic_apic_probe(void);
-
-#ifdef CONFIG_X86_LOCAL_APIC
-
-extern int apic_verbosity;
-extern int timer_over_8254;
-extern int local_apic_timer_c2_ok;
-extern int local_apic_timer_disabled;
-
-/*
- * Basic functions accessing APICs.
- */
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define apic_write native_apic_write
-#define apic_write_atomic native_apic_write_atomic
-#define apic_read native_apic_read
-#define setup_boot_clock setup_boot_APIC_clock
-#define setup_secondary_clock setup_secondary_APIC_clock
-#endif
-
-static __inline fastcall void native_apic_write(unsigned long reg, u32 v)
-{
-	*((volatile u32 *)(APIC_BASE + reg)) = v;
-}
-
-static __inline fastcall void native_apic_write_atomic(unsigned long reg, u32 v)
-{
-	(void) xchg((u32 *)(APIC_BASE + reg), v);
-}
-
-static __inline fastcall u32 native_apic_read(unsigned long reg)
-{
-	return *((volatile u32 *)(APIC_BASE + reg));
-}
-
-extern void apic_wait_icr_idle(void);
-extern u32 safe_apic_wait_icr_idle(void);
-extern int get_physical_broadcast(void);
-
-#ifdef CONFIG_X86_GOOD_APIC
-# define FORCE_READ_AROUND_WRITE 0
-# define apic_read_around(x)
-# define apic_write_around(x,y) apic_write((x),(y))
-#else
-# define FORCE_READ_AROUND_WRITE 1
-# define apic_read_around(x) apic_read(x)
-# define apic_write_around(x,y) apic_write_atomic((x),(y))
-#endif
-
-static inline void ack_APIC_irq(void)
-{
-	/*
-	 * ack_APIC_irq() actually gets compiled as a single instruction:
-	 * - a single rmw on Pentium/82489DX
-	 * - a single write on P6+ cores (CONFIG_X86_GOOD_APIC)
-	 * ... yummie.
-	 */
-
-	/* Docs say use 0 for future compatibility */
-	apic_write_around(APIC_EOI, 0);
-}
-
-extern int lapic_get_maxlvt(void);
-extern void clear_local_APIC(void);
-extern void connect_bsp_APIC(void);
-extern void disconnect_bsp_APIC(int virt_wire_setup);
-extern void disable_local_APIC(void);
-extern void lapic_shutdown(void);
-extern int verify_local_APIC(void);
-extern void cache_APIC_registers(void);
-extern void sync_Arb_IDs(void);
-extern void init_bsp_APIC(void);
-extern void setup_local_APIC(void);
-extern void init_apic_mappings(void);
-extern void smp_local_timer_interrupt(void);
-extern void setup_boot_APIC_clock(void);
-extern void setup_secondary_APIC_clock(void);
-extern int APIC_init_uniprocessor(void);
-extern void enable_NMI_through_LVT0(void *dummy);
-
-#define ARCH_APICTIMER_STOPS_ON_C3	1
-
-#else /* !CONFIG_X86_LOCAL_APIC */
-static inline void lapic_shutdown(void) { }
-#define local_apic_timer_c2_ok		1
-
-#endif /* !CONFIG_X86_LOCAL_APIC */
-
-#endif /* __ASM_APIC_H */
diff --git a/include/asm-x86/apic_64.h b/include/asm-x86/apic_64.h
deleted file mode 100644
index 9d0c06c4df91..000000000000
--- a/include/asm-x86/apic_64.h
+++ /dev/null
@@ -1,95 +0,0 @@
-#ifndef __ASM_APIC_H
-#define __ASM_APIC_H
-
-#include <linux/pm.h>
-#include <linux/delay.h>
-#include <asm/fixmap.h>
-#include <asm/apicdef.h>
-#include <asm/system.h>
-
-#define Dprintk(x...)
-
-/*
- * Debugging macros
- */
-#define APIC_QUIET   0
-#define APIC_VERBOSE 1
-#define APIC_DEBUG   2
-
-extern int apic_verbosity;
-extern int apic_runs_main_timer;
-extern int ioapic_force;
-extern int disable_apic_timer;
-
-/*
- * Define the default level of output to be very little
- * This can be turned up by using apic=verbose for more
- * information and apic=debug for _lots_ of information.
- * apic_verbosity is defined in apic.c
- */
-#define apic_printk(v, s, a...) do {       \
-		if ((v) <= apic_verbosity) \
-			printk(s, ##a);    \
-	} while (0)
-
-struct pt_regs;
-
-/*
- * Basic functions accessing APICs.
- */
-
-static __inline void apic_write(unsigned long reg, u32 v)
-{
-	*((volatile unsigned int *)(APIC_BASE+reg)) = v;
-}
-
-static __inline u32 apic_read(unsigned long reg)
-{
-	return *((volatile u32 *)(APIC_BASE+reg));
-}
-
-extern void apic_wait_icr_idle(void);
-extern u32 safe_apic_wait_icr_idle(void);
-
-static inline void ack_APIC_irq(void)
-{
-	/*
-	 * ack_APIC_irq() actually gets compiled as a single instruction:
-	 * - a single rmw on Pentium/82489DX
-	 * - a single write on P6+ cores (CONFIG_X86_GOOD_APIC)
-	 * ... yummie.
-	 */
-
-	/* Docs say use 0 for future compatibility */
-	apic_write(APIC_EOI, 0);
-}
-
-extern int lapic_get_maxlvt(void);
-extern void clear_local_APIC(void);
-extern void connect_bsp_APIC(void);
-extern void disconnect_bsp_APIC(int virt_wire_setup);
-extern void disable_local_APIC(void);
-extern void lapic_shutdown(void);
-extern int verify_local_APIC(void);
-extern void cache_APIC_registers(void);
-extern void sync_Arb_IDs(void);
-extern void init_bsp_APIC(void);
-extern void setup_local_APIC(void);
-extern void init_apic_mappings(void);
-extern void smp_local_timer_interrupt(void);
-extern void setup_boot_APIC_clock(void);
-extern void setup_secondary_APIC_clock(void);
-extern int APIC_init_uniprocessor(void);
-extern void setup_apic_routing(void);
-
-extern void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
-				    unsigned char msg_type, unsigned char mask);
-
-extern int apic_is_clustered_box(void);
-
-#define ARCH_APICTIMER_STOPS_ON_C3	1
-
-extern unsigned boot_cpu_id;
-extern int local_apic_timer_c2_ok;
-
-#endif /* __ASM_APIC_H */
-- 
cgit v1.2.3


From 63fefb7d5bac6e65b5df2b7eb4aa56bb4bbadd48 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:15 +0100
Subject: x86: remove duplicated declarations

Remove declarations which are made already in the appropriate header file.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/io_apic_64.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/io_apic_64.h b/include/asm-x86/io_apic_64.h
index e2c13675ee4e..762a841b939e 100644
--- a/include/asm-x86/io_apic_64.h
+++ b/include/asm-x86/io_apic_64.h
@@ -129,10 +129,4 @@ extern int io_apic_set_pci_routing (int ioapic, int pin, int irq, int, int);
 
 extern int sis_apic_bug; /* dummy */ 
 
-void enable_NMI_through_LVT0 (void * dummy);
-
-extern spinlock_t i8259A_lock;
-
-extern int timer_over_8254;
-
 #endif
-- 
cgit v1.2.3


From 376ff0352c24a5fa47f1250dd60937b5a9077672 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:16 +0100
Subject: x86: move acpi and pci declarations

Move acpi/pci related declarations to the correct headers
and remove the duplicate.

Build fix from: Andrew Morton

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_32.c | 3 +--
 arch/x86/kernel/setup_64.c | 2 --
 include/asm-x86/acpi_32.h  | 2 --
 include/asm-x86/pci.h      | 4 +++-
 include/asm-x86/proto.h    | 3 ---
 5 files changed, 4 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 51bdc0b1b72e..236d30b264d8 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -44,6 +44,7 @@
 #include <linux/crash_dump.h>
 #include <linux/dmi.h>
 #include <linux/pfn.h>
+#include <linux/pci.h>
 
 #include <video/edid.h>
 
@@ -663,9 +664,7 @@ void __init setup_arch(char **cmdline_p)
 	acpi_boot_table_init();
 #endif
 
-#ifdef CONFIG_PCI
 	early_quirks();
-#endif
 
 #ifdef CONFIG_ACPI
 	acpi_boot_init();
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index ec976edf0399..97a497652d20 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -407,9 +407,7 @@ void __init setup_arch(char **cmdline_p)
 	reserve_crashkernel();
 	paging_init();
 
-#ifdef CONFIG_PCI
 	early_quirks();
-#endif
 
 	/*
 	 * set this early, so we dont allocate cpu0
diff --git a/include/asm-x86/acpi_32.h b/include/asm-x86/acpi_32.h
index 723493e6c851..3cce5a0f2d9c 100644
--- a/include/asm-x86/acpi_32.h
+++ b/include/asm-x86/acpi_32.h
@@ -81,8 +81,6 @@ int __acpi_release_global_lock(unsigned int *lock);
         :"=r"(n_hi), "=r"(n_lo)     \
         :"0"(n_hi), "1"(n_lo))
 
-extern void early_quirks(void);
-
 #ifdef CONFIG_ACPI
 extern int acpi_lapic;
 extern int acpi_ioapic;
diff --git a/include/asm-x86/pci.h b/include/asm-x86/pci.h
index e88361966347..6983730d86fd 100644
--- a/include/asm-x86/pci.h
+++ b/include/asm-x86/pci.h
@@ -66,6 +66,7 @@ extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
 
 
 #ifdef CONFIG_PCI
+extern void early_quirks(void);
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
 					unsigned long *strategy_parameter)
@@ -73,9 +74,10 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 	*strat = PCI_DMA_BURST_INFINITY;
 	*strategy_parameter = ~0UL;
 }
+#else
+static inline void early_quirks(void) { }
 #endif
 
-
 #endif  /* __KERNEL__ */
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index dabba55f7ed8..64fe816ea37a 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -58,8 +58,6 @@ extern void show_registers(struct pt_regs *regs);
 
 extern void exception_table_check(void);
 
-extern void acpi_reserve_bootmem(void);
-
 extern void swap_low_mappings(void);
 
 extern void __show_regs(struct pt_regs * regs);
@@ -69,7 +67,6 @@ extern void syscall32_cpu_init(void);
 
 extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end);
 
-extern void early_quirks(void);
 extern void check_efer(void);
 
 extern void select_idle_routine(const struct cpuinfo_x86 *c);
-- 
cgit v1.2.3


From 8c61b900ebb8ec5918ffd776ba1a61a5f022566d Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:16 +0100
Subject: x86: make early_indentify_cpu static

early_indentify_cpu is only used in setup_64.c

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_64.c | 4 +++-
 include/asm-x86/proto.h    | 2 --
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 97a497652d20..0e2bffe2e203 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -141,6 +141,8 @@ struct resource bss_resource = {
 	.flags = IORESOURCE_RAM,
 };
 
+static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c);
+
 #ifdef CONFIG_PROC_VMCORE
 /* elfcorehdr= specifies the location of elf core header
  * stored by the crashed kernel. This option will be passed
@@ -844,7 +846,7 @@ struct cpu_model_info {
 /* Do some early cpuid on the boot CPU to get some parameter that are
    needed before check_bugs. Everything advanced is in identify_cpu
    below. */
-void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
+static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 {
 	u32 tfms;
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 64fe816ea37a..9e3d51fd91a2 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -40,8 +40,6 @@ extern int nohpet;
 
 extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 
-extern void early_identify_cpu(struct cpuinfo_x86 *c);
-
 extern int k8_scan_nodes(unsigned long start, unsigned long end);
 
 extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
-- 
cgit v1.2.3


From fc36367211931cd503f050e4ceff5d9864cec817 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:16 +0100
Subject: x86: move idle related declarations

Move idle related declarations to processor_64.h, where the
the others are as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor_64.h | 4 ++++
 include/asm-x86/proto.h        | 4 ----
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index e4f19970a82b..6b96e1f12a42 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -434,6 +434,10 @@ static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
 
 extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
 
+extern int force_mwait;
+
+extern void select_idle_routine(const struct cpuinfo_x86 *c);
+
 #define stack_current() \
 ({								\
 	struct thread_info *ti;					\
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 9e3d51fd91a2..77b4177d17f5 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -67,8 +67,6 @@ extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 
 extern void check_efer(void);
 
-extern void select_idle_routine(const struct cpuinfo_x86 *c);
-
 extern unsigned long table_start, table_end;
 
 extern int exception_trace;
@@ -80,8 +78,6 @@ extern int notsc_setup(char *);
 
 extern int gsi_irq_sharing(int gsi);
 
-extern int force_mwait;
-
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
 
 #define round_up(x,y) (((x) + (y) - 1) & ~((y)-1))
-- 
cgit v1.2.3


From c9ff03428f24219b927d9d9d3c0c581622967794 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:16 +0100
Subject: x86: move k8 related declarations

Move k8 related declarations to k8.h and fix numa_64.c

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/numa_64.c   | 1 +
 include/asm-x86/k8.h    | 1 +
 include/asm-x86/proto.h | 2 --
 3 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 3d6926ba8995..2d8fdc05f415 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -17,6 +17,7 @@
 #include <asm/dma.h>
 #include <asm/numa.h>
 #include <asm/acpi.h>
+#include <asm/k8.h>
 
 #ifndef Dprintk
 #define Dprintk(x...)
diff --git a/include/asm-x86/k8.h b/include/asm-x86/k8.h
index 699dd6961eda..452e2b696ff4 100644
--- a/include/asm-x86/k8.h
+++ b/include/asm-x86/k8.h
@@ -10,5 +10,6 @@ extern struct pci_dev **k8_northbridges;
 extern int num_k8_northbridges;
 extern int cache_k8_northbridges(void);
 extern void k8_flush_garts(void);
+extern int k8_scan_nodes(unsigned long start, unsigned long end);
 
 #endif
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 77b4177d17f5..9b96dae5b7ac 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -40,8 +40,6 @@ extern int nohpet;
 
 extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 
-extern int k8_scan_nodes(unsigned long start, unsigned long end);
-
 extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
 extern unsigned long numa_free_all_bootmem(void);
 
-- 
cgit v1.2.3


From 718fc13b4675470ea191522ef98b02a55d990fa1 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:17 +0100
Subject: x86: move debug related declarations to kdebug.h

Move them and fixup some users.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820_64.c      | 1 +
 arch/x86/kernel/head64.c       | 1 +
 arch/x86/kernel/process_32.c   | 1 +
 arch/x86/mm/init_64.c          | 1 +
 include/asm-x86/arch_hooks.h   | 5 +----
 include/asm-x86/kdebug.h       | 5 +++++
 include/asm-x86/processor_32.h | 1 -
 include/asm-x86/proto.h        | 8 --------
 include/asm-x86/system_32.h    | 1 -
 9 files changed, 10 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index d41cd2f01733..e510cfd5bb71 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -26,6 +26,7 @@
 #include <asm/proto.h>
 #include <asm/setup.h>
 #include <asm/sections.h>
+#include <asm/kdebug.h>
 
 struct e820map e820;
 
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 6b3469311e42..d156bfbffa9a 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -19,6 +19,7 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
+#include <asm/kdebug.h>
 
 static void __init zap_identity_mappings(void)
 {
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index a63d2d2556ee..a8cdd09ad53f 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -55,6 +55,7 @@
 
 #include <asm/tlbflush.h>
 #include <asm/cpu.h>
+#include <asm/kdebug.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 0f9c8c890658..9677abb6cf8a 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -43,6 +43,7 @@
 #include <asm/proto.h>
 #include <asm/smp.h>
 #include <asm/sections.h>
+#include <asm/kdebug.h>
 
 #ifndef Dprintk
 #define Dprintk(x...)
diff --git a/include/asm-x86/arch_hooks.h b/include/asm-x86/arch_hooks.h
index a8c1fca9726d..768aee8a04ef 100644
--- a/include/asm-x86/arch_hooks.h
+++ b/include/asm-x86/arch_hooks.h
@@ -6,7 +6,7 @@
 /*
  *	linux/include/asm/arch_hooks.h
  *
- *	define the architecture specific hooks 
+ *	define the architecture specific hooks
  */
 
 /* these aren't arch hooks, they are generic routines
@@ -24,7 +24,4 @@ extern void trap_init_hook(void);
 extern void time_init_hook(void);
 extern void mca_nmi_hook(void);
 
-extern int setup_early_printk(char *);
-extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
-
 #endif
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index e2f9b62e535e..49e5c91d490c 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -22,10 +22,15 @@ enum die_val {
 	DIE_PAGE_FAULT,
 };
 
+extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 extern void printk_address(unsigned long address);
 extern void die(const char *,struct pt_regs *,long);
 extern void __die(const char *,struct pt_regs *,long);
 extern void show_registers(struct pt_regs *regs);
+extern void __show_registers(struct pt_regs *, int all);
+extern void show_trace(struct task_struct *, struct pt_regs *, unsigned long *);
+extern void __show_regs(struct pt_regs *regs);
+extern void show_regs(struct pt_regs *regs);
 extern void dump_pagetable(unsigned long);
 extern unsigned long oops_begin(void);
 extern void oops_end(unsigned long);
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 13976b086837..58880a1ddd65 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -423,7 +423,6 @@ extern void prepare_to_copy(struct task_struct *tsk);
 extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
 
 extern unsigned long thread_saved_pc(struct task_struct *tsk);
-void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack);
 
 unsigned long get_wchan(struct task_struct *p);
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 9b96dae5b7ac..b10cd1b82089 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -38,8 +38,6 @@ extern u32 pmtmr_ioport;
 #endif
 extern int nohpet;
 
-extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
-
 extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
 extern unsigned long numa_free_all_bootmem(void);
 
@@ -49,16 +47,10 @@ extern void load_gs_index(unsigned gs);
 
 extern unsigned long end_pfn_map; 
 
-extern void show_trace(struct task_struct *, struct pt_regs *, unsigned long * rsp);
-extern void show_registers(struct pt_regs *regs);
-
 extern void exception_table_check(void);
 
 extern void swap_low_mappings(void);
 
-extern void __show_regs(struct pt_regs * regs);
-extern void show_regs(struct pt_regs * regs);
-
 extern void syscall32_cpu_init(void);
 
 extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end);
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index ef8468883bac..db6283eb5e46 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -315,6 +315,5 @@ extern unsigned long arch_align_stack(unsigned long sp);
 extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
 
 void default_idle(void);
-void __show_registers(struct pt_regs *, int all);
 
 #endif
-- 
cgit v1.2.3


From af7a78e9258ffcca681e080cbc857f854869144f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:17 +0100
Subject: x86: move mce related declarations

Move the mce related declarations where they belong, fix the
users and remove 32bit dependency in mce.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_64.c |  1 +
 include/asm-x86/mce.h      | 14 +++++++-------
 include/asm-x86/proto.h    |  1 -
 3 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 0e2bffe2e203..a84a4efc7fe6 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -59,6 +59,7 @@
 #include <asm/sections.h>
 #include <asm/dmi.h>
 #include <asm/cacheflush.h>
+#include <asm/mce.h>
 
 /*
  * Machine setup..
diff --git a/include/asm-x86/mce.h b/include/asm-x86/mce.h
index df304fd89c27..e6ff507a73b0 100644
--- a/include/asm-x86/mce.h
+++ b/include/asm-x86/mce.h
@@ -85,14 +85,7 @@ struct mce_log {
 #ifdef __KERNEL__
 
 #ifdef CONFIG_X86_32
-#ifdef CONFIG_X86_MCE
-extern void mcheck_init(struct cpuinfo_x86 *c);
-#else
-#define mcheck_init(c) do {} while(0)
-#endif
-
 extern int mce_disabled;
-
 #else /* CONFIG_X86_32 */
 
 #include <asm/atomic.h>
@@ -121,6 +114,13 @@ extern int mce_notify_user(void);
 
 #endif /* !CONFIG_X86_32 */
 
+
+
+#ifdef CONFIG_X86_MCE
+extern void mcheck_init(struct cpuinfo_x86 *c);
+#else
+#define mcheck_init(c) do { } while (0)
+#endif
 extern void stop_mce(void);
 extern void restart_mce(void);
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index b10cd1b82089..02420dea73c6 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -13,7 +13,6 @@ extern void pda_init(int);
 
 extern void early_idt_handler(void);
 
-extern void mcheck_init(struct cpuinfo_x86 *c);
 extern void init_memory_mapping(unsigned long start, unsigned long end);
 
 extern void system_call(void); 
-- 
cgit v1.2.3


From aaa64e04f9af8c05a10ab3d67df44154742d15cf Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:17 +0100
Subject: x86: move numa related declarations

More stuff shuffeled to the correct place

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/init_64.c     | 1 +
 include/asm-x86/numa_64.h | 6 ++++++
 include/asm-x86/proto.h   | 5 -----
 3 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 9677abb6cf8a..0fbb657a8b19 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -44,6 +44,7 @@
 #include <asm/smp.h>
 #include <asm/sections.h>
 #include <asm/kdebug.h>
+#include <asm/numa.h>
 
 #ifndef Dprintk
 #define Dprintk(x...)
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index 0cc5c97a7fc9..e6bc0b5dfe3a 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -21,6 +21,12 @@ extern void srat_reserve_add_area(int nodeid);
 extern int hotadd_percent;
 
 extern unsigned char apicid_to_node[MAX_LOCAL_APIC];
+
+extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
+extern unsigned long numa_free_all_bootmem(void);
+extern void setup_node_bootmem(int nodeid, unsigned long start,
+			       unsigned long end);
+
 #ifdef CONFIG_NUMA
 extern void __init init_cpu_to_node(void);
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 02420dea73c6..71a419c6e041 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -37,9 +37,6 @@ extern u32 pmtmr_ioport;
 #endif
 extern int nohpet;
 
-extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
-extern unsigned long numa_free_all_bootmem(void);
-
 extern void reserve_bootmem_generic(unsigned long phys, unsigned len);
 
 extern void load_gs_index(unsigned gs);
@@ -52,8 +49,6 @@ extern void swap_low_mappings(void);
 
 extern void syscall32_cpu_init(void);
 
-extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end);
-
 extern void check_efer(void);
 
 extern unsigned long table_start, table_end;
-- 
cgit v1.2.3


From a89518b882283fdb3e11fb6d1929dc874b380ca8 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:17 +0100
Subject: x86: move page related declaration

end_pfn is in page.h, so end_pfn_map has a place there as well

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/page_64.h | 1 +
 include/asm-x86/proto.h   | 2 --
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index c3b52bcb171e..6fdc904a5fa5 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -41,6 +41,7 @@
 #ifndef __ASSEMBLY__
 
 extern unsigned long end_pfn;
+extern unsigned long end_pfn_map;
 
 void clear_page(void *);
 void copy_page(void *, void *);
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 71a419c6e041..7ff3d765de05 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -41,8 +41,6 @@ extern void reserve_bootmem_generic(unsigned long phys, unsigned len);
 
 extern void load_gs_index(unsigned gs);
 
-extern unsigned long end_pfn_map; 
-
 extern void exception_table_check(void);
 
 extern void swap_low_mappings(void);
-- 
cgit v1.2.3


From 40fec50ac4dd436b06630f415789e57200c5b06f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:18 +0100
Subject: x86: move pda related declaration

pda has its own header file as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pda.h   | 1 +
 include/asm-x86/proto.h | 1 -
 2 files changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index 35962bbe5e72..3d9dd653628a 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -40,6 +40,7 @@ struct x8664_pda {
 
 extern struct x8664_pda *_cpu_pda[];
 extern struct x8664_pda boot_cpu_pda[];
+extern void pda_init(int);
 
 #define cpu_pda(i) (_cpu_pda[i])
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 7ff3d765de05..8ed1ebf666be 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -9,7 +9,6 @@ struct cpuinfo_x86;
 struct pt_regs;
 
 extern void start_kernel(void);
-extern void pda_init(int); 
 
 extern void early_idt_handler(void);
 
-- 
cgit v1.2.3


From 80ca9c98f50c11b63b87971594d7e38cba0bbcef Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:18 +0100
Subject: x86: move tsc related declarations

tsc has also it's own header file. Nuke the stupid 64 bit ifdef
while at it.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/proto.h | 3 ---
 include/asm-x86/tsc.h   | 3 +--
 2 files changed, 1 insertion(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 8ed1ebf666be..a47e526716f4 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -51,11 +51,8 @@ extern void check_efer(void);
 extern unsigned long table_start, table_end;
 
 extern int exception_trace;
-extern unsigned cpu_khz;
-extern unsigned tsc_khz;
 
 extern int reboot_force;
-extern int notsc_setup(char *);
 
 extern int gsi_irq_sharing(int gsi);
 
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index 6baab30dc2c8..d7b1c4e7a108 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -73,8 +73,7 @@ int check_tsc_unstable(void);
 extern void check_tsc_sync_source(int cpu);
 extern void check_tsc_sync_target(void);
 
-#ifdef CONFIG_X86_64
 extern void tsc_calibrate(void);
-#endif
+extern int notsc_setup(char *);
 
 #endif
-- 
cgit v1.2.3


From 70a20025632ca320316b5068326784d07c8ff351 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:18 +0100
Subject: x86: move pmtmr related declarations

Move more stuff out of proto.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    | 1 +
 arch/x86/kernel/pmtimer_64.c | 4 ++--
 include/asm-x86/proto.h      | 9 ---------
 include/linux/acpi_pmtmr.h   | 2 ++
 4 files changed, 5 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 3de3764a862c..0cb14d4c2c5c 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -26,6 +26,7 @@
 #include <linux/module.h>
 #include <linux/ioport.h>
 #include <linux/clockchips.h>
+#include <linux/acpi_pmtmr.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
diff --git a/arch/x86/kernel/pmtimer_64.c b/arch/x86/kernel/pmtimer_64.c
index ae8f91214f15..b112406f1996 100644
--- a/arch/x86/kernel/pmtimer_64.c
+++ b/arch/x86/kernel/pmtimer_64.c
@@ -19,13 +19,13 @@
 #include <linux/time.h>
 #include <linux/init.h>
 #include <linux/cpumask.h>
+#include <linux/acpi_pmtmr.h>
+
 #include <asm/io.h>
 #include <asm/proto.h>
 #include <asm/msr.h>
 #include <asm/vsyscall.h>
 
-#define ACPI_PM_MASK 0xFFFFFF /* limit it to 24 bits */
-
 static inline u32 cyc2us(u32 cycles)
 {
 	/* The Power Management Timer ticks at 3.579545 ticks per microsecond.
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index a47e526716f4..9074aa7ebc65 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -25,15 +25,6 @@ extern void ia32_sysenter_target(void);
 extern void config_acpi_tables(void);
 extern void ia32_syscall(void);
 
-extern int pmtimer_mark_offset(void);
-extern void pmtimer_resume(void);
-extern void pmtimer_wait(unsigned);
-extern unsigned int do_gettimeoffset_pm(void);
-#ifdef CONFIG_X86_PM_TIMER
-extern u32 pmtmr_ioport;
-#else
-#define pmtmr_ioport 0
-#endif
 extern int nohpet;
 
 extern void reserve_bootmem_generic(unsigned long phys, unsigned len);
diff --git a/include/linux/acpi_pmtmr.h b/include/linux/acpi_pmtmr.h
index 1d0ef1ae8036..7e3d2859be50 100644
--- a/include/linux/acpi_pmtmr.h
+++ b/include/linux/acpi_pmtmr.h
@@ -25,6 +25,8 @@ static inline u32 acpi_pm_read_early(void)
 	return acpi_pm_read_verified() & ACPI_PM_MASK;
 }
 
+extern void pmtimer_wait(unsigned);
+
 #else
 
 static inline u32 acpi_pm_read_early(void)
-- 
cgit v1.2.3


From 6fc30f9e292d2275248b6ee5ca903248b66efca1 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:18 +0100
Subject: x86: remove obsolete nohpet declaration

Lonely user is hpet.c, so no need to declare it elsewhere.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/proto.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 9074aa7ebc65..4eddf2761804 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -25,8 +25,6 @@ extern void ia32_sysenter_target(void);
 extern void config_acpi_tables(void);
 extern void ia32_syscall(void);
 
-extern int nohpet;
-
 extern void reserve_bootmem_generic(unsigned long phys, unsigned len);
 
 extern void load_gs_index(unsigned gs);
-- 
cgit v1.2.3


From eaf76e8b93dd5e7a53e2cd6db53f3ca18ca0fe4c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:19 +0100
Subject: x86: remove duplicate start_kernel declaration

start_kernel is already declared in a generic header file.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/head64.c | 1 +
 include/asm-x86/proto.h  | 2 --
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index d156bfbffa9a..6643f3f994fb 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -10,6 +10,7 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/percpu.h>
+#include <linux/start_kernel.h>
 
 #include <asm/processor.h>
 #include <asm/proto.h>
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 4eddf2761804..5a45478b5d3a 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -8,8 +8,6 @@
 struct cpuinfo_x86; 
 struct pt_regs;
 
-extern void start_kernel(void);
-
 extern void early_idt_handler(void);
 
 extern void init_memory_mapping(unsigned long start, unsigned long end);
-- 
cgit v1.2.3


From 84c873ed2a00eab3e8ac49dc7889d7aad142ce22 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:19 +0100
Subject: x86: remove obsolte declarations from proto.h

Nuke duplicate and obsolete crap from this ugly dump bin.
There are still some entries left which need to be sorted out,
but I'm tired of that puzzle game right now.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/proto.h | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 5a45478b5d3a..3c6fb89f1120 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -5,44 +5,27 @@
 
 /* misc architecture specific prototypes */
 
-struct cpuinfo_x86; 
-struct pt_regs;
-
 extern void early_idt_handler(void);
 
 extern void init_memory_mapping(unsigned long start, unsigned long end);
 
-extern void system_call(void); 
-extern int kernel_syscall(void);
+extern void system_call(void);
 extern void syscall_init(void);
 
 extern void ia32_syscall(void);
-extern void ia32_cstar_target(void); 
-extern void ia32_sysenter_target(void); 
-
-extern void config_acpi_tables(void);
-extern void ia32_syscall(void);
+extern void ia32_cstar_target(void);
+extern void ia32_sysenter_target(void);
 
 extern void reserve_bootmem_generic(unsigned long phys, unsigned len);
 
-extern void load_gs_index(unsigned gs);
-
-extern void exception_table_check(void);
-
-extern void swap_low_mappings(void);
-
 extern void syscall32_cpu_init(void);
 
 extern void check_efer(void);
 
 extern unsigned long table_start, table_end;
 
-extern int exception_trace;
-
 extern int reboot_force;
 
-extern int gsi_irq_sharing(int gsi);
-
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
 
 #define round_up(x,y) (((x) + (y) - 1) & ~((y)-1))
-- 
cgit v1.2.3


From 3e35a0e525253837fc0ea4d0e060de3302bd9537 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:19 +0100
Subject: x86: move ioapic code where it belongs

The commit 399287229c775a8962a852a761d65dc9475dec7c hacked the
ioapic resource mapping into apic.c for no good reason.
Move the code into io_apic_64.c where it belongs.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    | 87 ------------------------------------------
 arch/x86/kernel/io_apic_64.c | 90 ++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/setup_64.c   |  1 +
 include/asm-x86/io_apic_64.h |  2 +
 4 files changed, 93 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 0cb14d4c2c5c..e0191cdd6f7a 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -49,7 +49,6 @@ static int apic_calibrate_pmtmr __initdata;
 int local_apic_timer_c2_ok;
 EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
 
-static struct resource *ioapic_resources;
 static struct resource lapic_resource = {
 	.name = "Local APIC",
 	.flags = IORESOURCE_MEM | IORESOURCE_BUSY,
@@ -714,64 +713,6 @@ static int __init detect_init_APIC (void)
 	return 0;
 }
 
-#ifdef CONFIG_X86_IO_APIC
-static struct resource * __init ioapic_setup_resources(void)
-{
-#define IOAPIC_RESOURCE_NAME_SIZE 11
-	unsigned long n;
-	struct resource *res;
-	char *mem;
-	int i;
-
-	if (nr_ioapics <= 0)
-		return NULL;
-
-	n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource);
-	n *= nr_ioapics;
-
-	mem = alloc_bootmem(n);
-	res = (void *)mem;
-
-	if (mem != NULL) {
-		memset(mem, 0, n);
-		mem += sizeof(struct resource) * nr_ioapics;
-
-		for (i = 0; i < nr_ioapics; i++) {
-			res[i].name = mem;
-			res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-			sprintf(mem,  "IOAPIC %u", i);
-			mem += IOAPIC_RESOURCE_NAME_SIZE;
-		}
-	}
-
-	ioapic_resources = res;
-
-	return res;
-}
-
-static int __init ioapic_insert_resources(void)
-{
-	int i;
-	struct resource *r = ioapic_resources;
-
-	if (!r) {
-		printk("IO APIC resources could be not be allocated.\n");
-		return -1;
-	}
-
-	for (i = 0; i < nr_ioapics; i++) {
-		insert_resource(&iomem_resource, r);
-		r++;
-	}
-
-	return 0;
-}
-
-/* Insert the IO APIC resources after PCI initialization has occured to handle
- * IO APICS that are mapped in on a BAR in PCI space. */
-late_initcall(ioapic_insert_resources);
-#endif
-
 void __init init_apic_mappings(void)
 {
 	unsigned long apic_phys;
@@ -801,34 +742,6 @@ void __init init_apic_mappings(void)
 	 * default configuration (or the MP table is broken).
 	 */
 	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
-
-	{
-		unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
-		int i;
-		struct resource *ioapic_res;
-
-		ioapic_res = ioapic_setup_resources();
-		for (i = 0; i < nr_ioapics; i++) {
-			if (smp_found_config) {
-				ioapic_phys = mp_ioapics[i].mpc_apicaddr;
-			} else {
-				ioapic_phys = (unsigned long)
-					alloc_bootmem_pages(PAGE_SIZE);
-				ioapic_phys = __pa(ioapic_phys);
-			}
-			set_fixmap_nocache(idx, ioapic_phys);
-			apic_printk(APIC_VERBOSE,
-				    "mapped IOAPIC to %016lx (%016lx)\n",
-				    __fix_to_virt(idx), ioapic_phys);
-			idx++;
-
-			if (ioapic_res != NULL) {
-				ioapic_res->start = ioapic_phys;
-				ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
-				ioapic_res++;
-			}
-		}
-	}
 }
 
 /*
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index d4f5286101a9..c6de7854ac63 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -35,6 +35,7 @@
 #ifdef CONFIG_ACPI
 #include <acpi/acpi_bus.h>
 #endif
+#include <linux/bootmem.h>
 
 #include <asm/idle.h>
 #include <asm/io.h>
@@ -2288,3 +2289,92 @@ void __init setup_ioapic_dest(void)
 }
 #endif
 
+#define IOAPIC_RESOURCE_NAME_SIZE 11
+
+static struct resource *ioapic_resources;
+
+static struct resource * __init ioapic_setup_resources(void)
+{
+	unsigned long n;
+	struct resource *res;
+	char *mem;
+	int i;
+
+	if (nr_ioapics <= 0)
+		return NULL;
+
+	n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource);
+	n *= nr_ioapics;
+
+	mem = alloc_bootmem(n);
+	res = (void *)mem;
+
+	if (mem != NULL) {
+		memset(mem, 0, n);
+		mem += sizeof(struct resource) * nr_ioapics;
+
+		for (i = 0; i < nr_ioapics; i++) {
+			res[i].name = mem;
+			res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+			sprintf(mem,  "IOAPIC %u", i);
+			mem += IOAPIC_RESOURCE_NAME_SIZE;
+		}
+	}
+
+	ioapic_resources = res;
+
+	return res;
+}
+
+void __init ioapic_init_mappings(void)
+{
+	unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
+	struct resource *ioapic_res;
+	int i;
+
+	ioapic_res = ioapic_setup_resources();
+	for (i = 0; i < nr_ioapics; i++) {
+		if (smp_found_config) {
+			ioapic_phys = mp_ioapics[i].mpc_apicaddr;
+		} else {
+			ioapic_phys = (unsigned long)
+				alloc_bootmem_pages(PAGE_SIZE);
+			ioapic_phys = __pa(ioapic_phys);
+		}
+		set_fixmap_nocache(idx, ioapic_phys);
+		apic_printk(APIC_VERBOSE,
+			    "mapped IOAPIC to %016lx (%016lx)\n",
+			    __fix_to_virt(idx), ioapic_phys);
+		idx++;
+
+		if (ioapic_res != NULL) {
+			ioapic_res->start = ioapic_phys;
+			ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
+			ioapic_res++;
+		}
+	}
+}
+
+static int __init ioapic_insert_resources(void)
+{
+	int i;
+	struct resource *r = ioapic_resources;
+
+	if (!r) {
+		printk(KERN_ERR
+		       "IO APIC resources could be not be allocated.\n");
+		return -1;
+	}
+
+	for (i = 0; i < nr_ioapics; i++) {
+		insert_resource(&iomem_resource, r);
+		r++;
+	}
+
+	return 0;
+}
+
+/* Insert the IO APIC resources after PCI initialization has occured to handle
+ * IO APICS that are mapped in on a BAR in PCI space. */
+late_initcall(ioapic_insert_resources);
+
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index a84a4efc7fe6..bcb5f3aaa097 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -433,6 +433,7 @@ void __init setup_arch(char **cmdline_p)
 	if (smp_found_config)
 		get_smp_config();
 	init_apic_mappings();
+	ioapic_init_mappings();
 
 	/*
 	 * We trust e820 completely. No explicit ROM probing in memory.
diff --git a/include/asm-x86/io_apic_64.h b/include/asm-x86/io_apic_64.h
index 762a841b939e..3e3c2114d93a 100644
--- a/include/asm-x86/io_apic_64.h
+++ b/include/asm-x86/io_apic_64.h
@@ -127,6 +127,8 @@ extern int io_apic_get_redir_entries (int ioapic);
 extern int io_apic_set_pci_routing (int ioapic, int pin, int irq, int, int);
 #endif
 
+extern void ioapic_init_mappings(void);
+
 extern int sis_apic_bug; /* dummy */ 
 
 #endif
-- 
cgit v1.2.3


From 3a12d93dc0b46bc710317272bf91640d7a8b6f18 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:20 +0100
Subject: x86: make smp_local_timer_interrupt() static

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c | 2 +-
 include/asm-x86/apic.h    | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 032bf1e765df..5bf329ed9d78 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -923,7 +923,7 @@ void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
  * value into /proc/profile.
  */
 
-void smp_local_timer_interrupt(void)
+static void smp_local_timer_interrupt(void)
 {
 	int cpu = smp_processor_id();
 	struct clock_event_device *evt = &per_cpu(lapic_events, cpu);
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index c064c1f84bab..d0a221fa1fc3 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -112,7 +112,6 @@ extern void sync_Arb_IDs(void);
 extern void init_bsp_APIC(void);
 extern void setup_local_APIC(void);
 extern void init_apic_mappings(void);
-extern void smp_local_timer_interrupt(void);
 extern void setup_boot_APIC_clock(void);
 extern void setup_secondary_APIC_clock(void);
 extern int APIC_init_uniprocessor(void);
-- 
cgit v1.2.3


From 2f2239d1d53b06dd475634f79e1905f3775fcedc Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:20 +0100
Subject: x86: prepare merging futex_32/64.h

Replace .quad/.long with a define and use the same asm syntax
for i386 and x86.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/futex_32.h | 64 ++++++++++++++++++++++------------------------
 include/asm-x86/futex_64.h | 24 +++++++++--------
 2 files changed, 44 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/futex_32.h b/include/asm-x86/futex_32.h
index 438ef0ec7101..c80013e19c21 100644
--- a/include/asm-x86/futex_32.h
+++ b/include/asm-x86/futex_32.h
@@ -4,6 +4,8 @@
 #ifdef __KERNEL__
 
 #include <linux/futex.h>
+
+#include <asm/asm.h>
 #include <asm/errno.h>
 #include <asm/system.h>
 #include <asm/processor.h>
@@ -17,13 +19,13 @@
 	jmp	2b\n\
 	.previous\n\
 	.section __ex_table,\"a\"\n\
-	.align	8\n\
-	.long	1b,3b\n\
+	.align	8\n"						\
+	_ASM_PTR "1b,3b\n					\
 	.previous"						\
 	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
 	: "i" (-EFAULT), "0" (oparg), "1" (0))
 
-#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \
+#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
   __asm__ __volatile (						\
 "1:	movl	%2, %0\n\
 	movl	%0, %3\n"					\
@@ -35,8 +37,8 @@
 	jmp	3b\n\
 	.previous\n\
 	.section __ex_table,\"a\"\n\
-	.align	8\n\
-	.long	1b,4b,2b,4b\n\
+	.align	8\n"						\
+	_ASM_PTR "1b,4b,2b,4b\n					\
 	.previous"						\
 	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
 	  "=&r" (tem)						\
@@ -56,36 +58,32 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
 	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
 		return -EFAULT;
 
+#ifndef CONFIG_X86_BSWAP
+	if (op == FUTEX_OP_SET && boot_cpu_data.x86 == 3)
+		return -ENOSYS;
+#endif
+
 	pagefault_disable();
 
-	if (op == FUTEX_OP_SET)
+	switch (op) {
+	case FUTEX_OP_SET:
 		__futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
-	else {
-#ifndef CONFIG_X86_BSWAP
-		if (boot_cpu_data.x86 == 3)
-			ret = -ENOSYS;
-		else
-#endif
-		switch (op) {
-		case FUTEX_OP_ADD:
-			__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret,
-					   oldval, uaddr, oparg);
-			break;
-		case FUTEX_OP_OR:
-			__futex_atomic_op2("orl %4, %3", ret, oldval, uaddr,
-					   oparg);
-			break;
-		case FUTEX_OP_ANDN:
-			__futex_atomic_op2("andl %4, %3", ret, oldval, uaddr,
-					   ~oparg);
-			break;
-		case FUTEX_OP_XOR:
-			__futex_atomic_op2("xorl %4, %3", ret, oldval, uaddr,
-					   oparg);
-			break;
-		default:
-			ret = -ENOSYS;
-		}
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval,
+				   uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op2("orl %4, %3", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op2("andl %4, %3", ret, oldval, uaddr, ~oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op2("xorl %4, %3", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
 	}
 
 	pagefault_enable();
@@ -120,7 +118,7 @@ futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
 
 		"	.section __ex_table, \"a\"		\n"
 		"	.align  8				\n"
-		"	.long   1b,3b				\n"
+			_ASM_PTR " 1b,3b			\n"
 		"	.previous				\n"
 
 		: "=a" (oldval), "+m" (*uaddr)
diff --git a/include/asm-x86/futex_64.h b/include/asm-x86/futex_64.h
index 5cdfb08013c3..02964d225d18 100644
--- a/include/asm-x86/futex_64.h
+++ b/include/asm-x86/futex_64.h
@@ -4,6 +4,8 @@
 #ifdef __KERNEL__
 
 #include <linux/futex.h>
+
+#include <asm/asm.h>
 #include <asm/errno.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -16,13 +18,13 @@
 	jmp	2b\n\
 	.previous\n\
 	.section __ex_table,\"a\"\n\
-	.align	8\n\
-	.quad	1b,3b\n\
+	.align	8\n"						\
+	_ASM_PTR "1b,3b\n					\
 	.previous"						\
-	: "=r" (oldval), "=r" (ret), "=m" (*uaddr)		\
-	: "i" (-EFAULT), "m" (*uaddr), "0" (oparg), "1" (0))
+	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
+	: "i" (-EFAULT), "0" (oparg), "1" (0))
 
-#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \
+#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
   __asm__ __volatile (						\
 "1:	movl	%2, %0\n\
 	movl	%0, %3\n"					\
@@ -34,12 +36,12 @@
 	jmp	3b\n\
 	.previous\n\
 	.section __ex_table,\"a\"\n\
-	.align	8\n\
-	.quad	1b,4b,2b,4b\n\
+	.align	8\n"						\
+	_ASM_PTR "1b,4b,2b,4b\n					\
 	.previous"						\
-	: "=&a" (oldval), "=&r" (ret), "=m" (*uaddr),		\
+	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
 	  "=&r" (tem)						\
-	: "r" (oparg), "i" (-EFAULT), "m" (*uaddr), "1" (0))
+	: "r" (oparg), "i" (-EFAULT), "1" (0))
 
 static inline int
 futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
@@ -110,10 +112,10 @@ futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
 
 		"	.section __ex_table, \"a\"		\n"
 		"	.align  8				\n"
-		"	.quad   1b,3b				\n"
+			_ASM_PTR " 1b,3b			\n"
 		"	.previous				\n"
 
-		: "=a" (oldval), "=m" (*uaddr)
+		: "=a" (oldval), "+m" (*uaddr)
 		: "i" (-EFAULT), "r" (newval), "0" (oldval)
 		: "memory"
 	);
-- 
cgit v1.2.3


From 2f18e47c89f1a29cc815cd225b72dd7fd86acffe Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:20 +0100
Subject: x86: merge futex_32/64.h

Finally merge them together.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/futex.h    | 138 +++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/futex_32.h | 133 -------------------------------------------
 include/asm-x86/futex_64.h | 127 -----------------------------------------
 3 files changed, 134 insertions(+), 264 deletions(-)
 delete mode 100644 include/asm-x86/futex_32.h
 delete mode 100644 include/asm-x86/futex_64.h

(limited to 'include')

diff --git a/include/asm-x86/futex.h b/include/asm-x86/futex.h
index 1f4610e0c613..62828d63f1b1 100644
--- a/include/asm-x86/futex.h
+++ b/include/asm-x86/futex.h
@@ -1,5 +1,135 @@
-#ifdef CONFIG_X86_32
-# include "futex_32.h"
-#else
-# include "futex_64.h"
+#ifndef _ASM_X86_FUTEX_H
+#define _ASM_X86_FUTEX_H
+
+#ifdef __KERNEL__
+
+#include <linux/futex.h>
+
+#include <asm/asm.h>
+#include <asm/errno.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+
+#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg)	\
+  __asm__ __volatile(						\
+"1:	" insn "\n"						\
+"2:	.section .fixup,\"ax\"\n				\
+3:	mov	%3, %1\n					\
+	jmp	2b\n						\
+	.previous\n						\
+	.section __ex_table,\"a\"\n				\
+	.align	8\n"						\
+	_ASM_PTR "1b,3b\n					\
+	.previous"						\
+	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
+	: "i" (-EFAULT), "0" (oparg), "1" (0))
+
+#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
+  __asm__ __volatile(						\
+"1:	movl	%2, %0\n					\
+	movl	%0, %3\n"					\
+	insn "\n"						\
+"2:	" LOCK_PREFIX "cmpxchgl %3, %2\n			\
+	jnz	1b\n						\
+3:	.section .fixup,\"ax\"\n				\
+4:	mov	%5, %1\n					\
+	jmp	3b\n						\
+	.previous\n						\
+	.section __ex_table,\"a\"\n				\
+	.align	8\n"						\
+	_ASM_PTR "1b,4b,2b,4b\n					\
+	.previous"						\
+	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
+	  "=&r" (tem)						\
+	: "r" (oparg), "i" (-EFAULT), "1" (0))
+
+static inline int
+futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+{
+	int op = (encoded_op >> 28) & 7;
+	int cmp = (encoded_op >> 24) & 15;
+	int oparg = (encoded_op << 8) >> 20;
+	int cmparg = (encoded_op << 20) >> 20;
+	int oldval = 0, ret, tem;
+
+	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+		oparg = 1 << oparg;
+
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_BSWAP)
+	/* Real i386 machines can only support FUTEX_OP_SET */
+	if (op != FUTEX_OP_SET && boot_cpu_data.x86 == 3)
+		return -ENOSYS;
+#endif
+
+	pagefault_disable();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval,
+				   uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op2("orl %4, %3", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op2("andl %4, %3", ret, oldval, uaddr, ~oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op2("xorl %4, %3", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+
+	pagefault_enable();
+
+	if (!ret) {
+		switch (cmp) {
+		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
+		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
+		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
+		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
+		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
+		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+		default: ret = -ENOSYS;
+		}
+	}
+	return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+{
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+	__asm__ __volatile__(
+		"1:	" LOCK_PREFIX "cmpxchgl %3, %1		\n"
+
+		"2:	.section .fixup, \"ax\"			\n"
+		"3:	mov     %2, %0				\n"
+		"	jmp     2b				\n"
+		"	.previous				\n"
+
+		"	.section __ex_table, \"a\"		\n"
+		"	.align  8				\n"
+			_ASM_PTR " 1b,3b			\n"
+		"	.previous				\n"
+
+		: "=a" (oldval), "+m" (*uaddr)
+		: "i" (-EFAULT), "r" (newval), "0" (oldval)
+		: "memory"
+	);
+
+	return oldval;
+}
+
+#endif
 #endif
diff --git a/include/asm-x86/futex_32.h b/include/asm-x86/futex_32.h
deleted file mode 100644
index c80013e19c21..000000000000
--- a/include/asm-x86/futex_32.h
+++ /dev/null
@@ -1,133 +0,0 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
-
-#ifdef __KERNEL__
-
-#include <linux/futex.h>
-
-#include <asm/asm.h>
-#include <asm/errno.h>
-#include <asm/system.h>
-#include <asm/processor.h>
-#include <asm/uaccess.h>
-
-#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \
-  __asm__ __volatile (						\
-"1:	" insn "\n"						\
-"2:	.section .fixup,\"ax\"\n\
-3:	mov	%3, %1\n\
-	jmp	2b\n\
-	.previous\n\
-	.section __ex_table,\"a\"\n\
-	.align	8\n"						\
-	_ASM_PTR "1b,3b\n					\
-	.previous"						\
-	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
-	: "i" (-EFAULT), "0" (oparg), "1" (0))
-
-#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
-  __asm__ __volatile (						\
-"1:	movl	%2, %0\n\
-	movl	%0, %3\n"					\
-	insn "\n"						\
-"2:	" LOCK_PREFIX "cmpxchgl %3, %2\n\
-	jnz	1b\n\
-3:	.section .fixup,\"ax\"\n\
-4:	mov	%5, %1\n\
-	jmp	3b\n\
-	.previous\n\
-	.section __ex_table,\"a\"\n\
-	.align	8\n"						\
-	_ASM_PTR "1b,4b,2b,4b\n					\
-	.previous"						\
-	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
-	  "=&r" (tem)						\
-	: "r" (oparg), "i" (-EFAULT), "1" (0))
-
-static inline int
-futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
-{
-	int op = (encoded_op >> 28) & 7;
-	int cmp = (encoded_op >> 24) & 15;
-	int oparg = (encoded_op << 8) >> 20;
-	int cmparg = (encoded_op << 20) >> 20;
-	int oldval = 0, ret, tem;
-	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
-		oparg = 1 << oparg;
-
-	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
-		return -EFAULT;
-
-#ifndef CONFIG_X86_BSWAP
-	if (op == FUTEX_OP_SET && boot_cpu_data.x86 == 3)
-		return -ENOSYS;
-#endif
-
-	pagefault_disable();
-
-	switch (op) {
-	case FUTEX_OP_SET:
-		__futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ADD:
-		__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval,
-				   uaddr, oparg);
-		break;
-	case FUTEX_OP_OR:
-		__futex_atomic_op2("orl %4, %3", ret, oldval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ANDN:
-		__futex_atomic_op2("andl %4, %3", ret, oldval, uaddr, ~oparg);
-		break;
-	case FUTEX_OP_XOR:
-		__futex_atomic_op2("xorl %4, %3", ret, oldval, uaddr, oparg);
-		break;
-	default:
-		ret = -ENOSYS;
-	}
-
-	pagefault_enable();
-
-	if (!ret) {
-		switch (cmp) {
-		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
-		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
-		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
-		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
-		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
-		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
-		default: ret = -ENOSYS;
-		}
-	}
-	return ret;
-}
-
-static inline int
-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
-{
-	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
-		return -EFAULT;
-
-	__asm__ __volatile__(
-		"1:	" LOCK_PREFIX "cmpxchgl %3, %1		\n"
-
-		"2:	.section .fixup, \"ax\"			\n"
-		"3:	mov     %2, %0				\n"
-		"	jmp     2b				\n"
-		"	.previous				\n"
-
-		"	.section __ex_table, \"a\"		\n"
-		"	.align  8				\n"
-			_ASM_PTR " 1b,3b			\n"
-		"	.previous				\n"
-
-		: "=a" (oldval), "+m" (*uaddr)
-		: "i" (-EFAULT), "r" (newval), "0" (oldval)
-		: "memory"
-	);
-
-	return oldval;
-}
-
-#endif
-#endif
diff --git a/include/asm-x86/futex_64.h b/include/asm-x86/futex_64.h
deleted file mode 100644
index 02964d225d18..000000000000
--- a/include/asm-x86/futex_64.h
+++ /dev/null
@@ -1,127 +0,0 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
-
-#ifdef __KERNEL__
-
-#include <linux/futex.h>
-
-#include <asm/asm.h>
-#include <asm/errno.h>
-#include <asm/system.h>
-#include <asm/uaccess.h>
-
-#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \
-  __asm__ __volatile (						\
-"1:	" insn "\n"						\
-"2:	.section .fixup,\"ax\"\n\
-3:	mov	%3, %1\n\
-	jmp	2b\n\
-	.previous\n\
-	.section __ex_table,\"a\"\n\
-	.align	8\n"						\
-	_ASM_PTR "1b,3b\n					\
-	.previous"						\
-	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
-	: "i" (-EFAULT), "0" (oparg), "1" (0))
-
-#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
-  __asm__ __volatile (						\
-"1:	movl	%2, %0\n\
-	movl	%0, %3\n"					\
-	insn "\n"						\
-"2:	" LOCK_PREFIX "cmpxchgl %3, %2\n\
-	jnz	1b\n\
-3:	.section .fixup,\"ax\"\n\
-4:	mov	%5, %1\n\
-	jmp	3b\n\
-	.previous\n\
-	.section __ex_table,\"a\"\n\
-	.align	8\n"						\
-	_ASM_PTR "1b,4b,2b,4b\n					\
-	.previous"						\
-	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
-	  "=&r" (tem)						\
-	: "r" (oparg), "i" (-EFAULT), "1" (0))
-
-static inline int
-futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
-{
-	int op = (encoded_op >> 28) & 7;
-	int cmp = (encoded_op >> 24) & 15;
-	int oparg = (encoded_op << 8) >> 20;
-	int cmparg = (encoded_op << 20) >> 20;
-	int oldval = 0, ret, tem;
-	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
-		oparg = 1 << oparg;
-
-	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
-		return -EFAULT;
-
-	pagefault_disable();
-
-	switch (op) {
-	case FUTEX_OP_SET:
-		__futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ADD:
-		__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval,
-				   uaddr, oparg);
-		break;
-	case FUTEX_OP_OR:
-		__futex_atomic_op2("orl %4, %3", ret, oldval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ANDN:
-		__futex_atomic_op2("andl %4, %3", ret, oldval, uaddr, ~oparg);
-		break;
-	case FUTEX_OP_XOR:
-		__futex_atomic_op2("xorl %4, %3", ret, oldval, uaddr, oparg);
-		break;
-	default:
-		ret = -ENOSYS;
-	}
-
-	pagefault_enable();
-
-	if (!ret) {
-		switch (cmp) {
-		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
-		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
-		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
-		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
-		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
-		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
-		default: ret = -ENOSYS;
-		}
-	}
-	return ret;
-}
-
-static inline int
-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
-{
-	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
-		return -EFAULT;
-
-	__asm__ __volatile__(
-		"1:	" LOCK_PREFIX "cmpxchgl %3, %1		\n"
-
-		"2:	.section .fixup, \"ax\"			\n"
-		"3:	mov     %2, %0				\n"
-		"	jmp     2b				\n"
-		"	.previous				\n"
-
-		"	.section __ex_table, \"a\"		\n"
-		"	.align  8				\n"
-			_ASM_PTR " 1b,3b			\n"
-		"	.previous				\n"
-
-		: "=a" (oldval), "+m" (*uaddr)
-		: "i" (-EFAULT), "r" (newval), "0" (oldval)
-		: "memory"
-	);
-
-	return oldval;
-}
-
-#endif
-#endif
-- 
cgit v1.2.3


From f28b8d608829aecd7343015a0df89f6b6e89d391 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:20 +0100
Subject: x86: merge include/asm-x86/dma.h

Almost identical.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/dma.h    | 318 ++++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/dma_32.h | 297 -------------------------------------------
 include/asm-x86/dma_64.h | 304 --------------------------------------------
 3 files changed, 316 insertions(+), 603 deletions(-)
 delete mode 100644 include/asm-x86/dma_32.h
 delete mode 100644 include/asm-x86/dma_64.h

(limited to 'include')

diff --git a/include/asm-x86/dma.h b/include/asm-x86/dma.h
index 9f936c61a4e5..e9733ce89880 100644
--- a/include/asm-x86/dma.h
+++ b/include/asm-x86/dma.h
@@ -1,5 +1,319 @@
+/*
+ * linux/include/asm/dma.h: Defines for using and allocating dma channels.
+ * Written by Hennus Bergman, 1992.
+ * High DMA channel support & info by Hannu Savolainen
+ * and John Boyd, Nov. 1992.
+ */
+
+#ifndef _ASM_X86_DMA_H
+#define _ASM_X86_DMA_H
+
+#include <linux/spinlock.h>	/* And spinlocks */
+#include <asm/io.h>		/* need byte IO */
+#include <linux/delay.h>
+
+
+#ifdef HAVE_REALLY_SLOW_DMA_CONTROLLER
+#define dma_outb	outb_p
+#else
+#define dma_outb	outb
+#endif
+
+#define dma_inb		inb
+
+/*
+ * NOTES about DMA transfers:
+ *
+ *  controller 1: channels 0-3, byte operations, ports 00-1F
+ *  controller 2: channels 4-7, word operations, ports C0-DF
+ *
+ *  - ALL registers are 8 bits only, regardless of transfer size
+ *  - channel 4 is not used - cascades 1 into 2.
+ *  - channels 0-3 are byte - addresses/counts are for physical bytes
+ *  - channels 5-7 are word - addresses/counts are for physical words
+ *  - transfers must not cross physical 64K (0-3) or 128K (5-7) boundaries
+ *  - transfer count loaded to registers is 1 less than actual count
+ *  - controller 2 offsets are all even (2x offsets for controller 1)
+ *  - page registers for 5-7 don't use data bit 0, represent 128K pages
+ *  - page registers for 0-3 use bit 0, represent 64K pages
+ *
+ * DMA transfers are limited to the lower 16MB of _physical_ memory.
+ * Note that addresses loaded into registers must be _physical_ addresses,
+ * not logical addresses (which may differ if paging is active).
+ *
+ *  Address mapping for channels 0-3:
+ *
+ *   A23 ... A16 A15 ... A8  A7 ... A0    (Physical addresses)
+ *    |  ...  |   |  ... |   |  ... |
+ *    |  ...  |   |  ... |   |  ... |
+ *    |  ...  |   |  ... |   |  ... |
+ *   P7  ...  P0  A7 ... A0  A7 ... A0
+ * |    Page    | Addr MSB | Addr LSB |   (DMA registers)
+ *
+ *  Address mapping for channels 5-7:
+ *
+ *   A23 ... A17 A16 A15 ... A9 A8 A7 ... A1 A0    (Physical addresses)
+ *    |  ...  |   \   \   ... \  \  \  ... \  \
+ *    |  ...  |    \   \   ... \  \  \  ... \  (not used)
+ *    |  ...  |     \   \   ... \  \  \  ... \
+ *   P7  ...  P1 (0) A7 A6  ... A0 A7 A6 ... A0
+ * |      Page      |  Addr MSB   |  Addr LSB  |   (DMA registers)
+ *
+ * Again, channels 5-7 transfer _physical_ words (16 bits), so addresses
+ * and counts _must_ be word-aligned (the lowest address bit is _ignored_ at
+ * the hardware level, so odd-byte transfers aren't possible).
+ *
+ * Transfer count (_not # bytes_) is limited to 64K, represented as actual
+ * count - 1 : 64K => 0xFFFF, 1 => 0x0000.  Thus, count is always 1 or more,
+ * and up to 128K bytes may be transferred on channels 5-7 in one operation.
+ *
+ */
+
+#define MAX_DMA_CHANNELS	8
+
 #ifdef CONFIG_X86_32
-# include "dma_32.h"
+
+/* The maximum address that we can perform a DMA transfer to on this platform */
+#define MAX_DMA_ADDRESS      (PAGE_OFFSET+0x1000000)
+
+#else
+
+/* 16MB ISA DMA zone */
+#define MAX_DMA_PFN   ((16*1024*1024) >> PAGE_SHIFT)
+
+/* 4GB broken PCI/AGP hardware bus master zone */
+#define MAX_DMA32_PFN ((4UL*1024*1024*1024) >> PAGE_SHIFT)
+
+/* Compat define for old dma zone */
+#define MAX_DMA_ADDRESS ((unsigned long)__va(MAX_DMA_PFN << PAGE_SHIFT))
+
+#endif
+
+/* 8237 DMA controllers */
+#define IO_DMA1_BASE	0x00	/* 8 bit slave DMA, channels 0..3 */
+#define IO_DMA2_BASE	0xC0	/* 16 bit master DMA, ch 4(=slave input)..7 */
+
+/* DMA controller registers */
+#define DMA1_CMD_REG		0x08	/* command register (w) */
+#define DMA1_STAT_REG		0x08	/* status register (r) */
+#define DMA1_REQ_REG		0x09    /* request register (w) */
+#define DMA1_MASK_REG		0x0A	/* single-channel mask (w) */
+#define DMA1_MODE_REG		0x0B	/* mode register (w) */
+#define DMA1_CLEAR_FF_REG	0x0C	/* clear pointer flip-flop (w) */
+#define DMA1_TEMP_REG		0x0D    /* Temporary Register (r) */
+#define DMA1_RESET_REG		0x0D	/* Master Clear (w) */
+#define DMA1_CLR_MASK_REG       0x0E    /* Clear Mask */
+#define DMA1_MASK_ALL_REG       0x0F    /* all-channels mask (w) */
+
+#define DMA2_CMD_REG		0xD0	/* command register (w) */
+#define DMA2_STAT_REG		0xD0	/* status register (r) */
+#define DMA2_REQ_REG		0xD2    /* request register (w) */
+#define DMA2_MASK_REG		0xD4	/* single-channel mask (w) */
+#define DMA2_MODE_REG		0xD6	/* mode register (w) */
+#define DMA2_CLEAR_FF_REG	0xD8	/* clear pointer flip-flop (w) */
+#define DMA2_TEMP_REG		0xDA    /* Temporary Register (r) */
+#define DMA2_RESET_REG		0xDA	/* Master Clear (w) */
+#define DMA2_CLR_MASK_REG       0xDC    /* Clear Mask */
+#define DMA2_MASK_ALL_REG       0xDE    /* all-channels mask (w) */
+
+#define DMA_ADDR_0		0x00    /* DMA address registers */
+#define DMA_ADDR_1		0x02
+#define DMA_ADDR_2		0x04
+#define DMA_ADDR_3		0x06
+#define DMA_ADDR_4		0xC0
+#define DMA_ADDR_5		0xC4
+#define DMA_ADDR_6		0xC8
+#define DMA_ADDR_7		0xCC
+
+#define DMA_CNT_0		0x01    /* DMA count registers */
+#define DMA_CNT_1		0x03
+#define DMA_CNT_2		0x05
+#define DMA_CNT_3		0x07
+#define DMA_CNT_4		0xC2
+#define DMA_CNT_5		0xC6
+#define DMA_CNT_6		0xCA
+#define DMA_CNT_7		0xCE
+
+#define DMA_PAGE_0		0x87    /* DMA page registers */
+#define DMA_PAGE_1		0x83
+#define DMA_PAGE_2		0x81
+#define DMA_PAGE_3		0x82
+#define DMA_PAGE_5		0x8B
+#define DMA_PAGE_6		0x89
+#define DMA_PAGE_7		0x8A
+
+/* I/O to memory, no autoinit, increment, single mode */
+#define DMA_MODE_READ		0x44
+/* memory to I/O, no autoinit, increment, single mode */
+#define DMA_MODE_WRITE		0x48
+/* pass thru DREQ->HRQ, DACK<-HLDA only */
+#define DMA_MODE_CASCADE	0xC0
+
+#define DMA_AUTOINIT		0x10
+
+
+extern spinlock_t  dma_spin_lock;
+
+static __inline__ unsigned long claim_dma_lock(void)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&dma_spin_lock, flags);
+	return flags;
+}
+
+static __inline__ void release_dma_lock(unsigned long flags)
+{
+	spin_unlock_irqrestore(&dma_spin_lock, flags);
+}
+
+/* enable/disable a specific DMA channel */
+static __inline__ void enable_dma(unsigned int dmanr)
+{
+	if (dmanr <= 3)
+		dma_outb(dmanr, DMA1_MASK_REG);
+	else
+		dma_outb(dmanr & 3, DMA2_MASK_REG);
+}
+
+static __inline__ void disable_dma(unsigned int dmanr)
+{
+	if (dmanr <= 3)
+		dma_outb(dmanr | 4, DMA1_MASK_REG);
+	else
+		dma_outb((dmanr & 3) | 4, DMA2_MASK_REG);
+}
+
+/* Clear the 'DMA Pointer Flip Flop'.
+ * Write 0 for LSB/MSB, 1 for MSB/LSB access.
+ * Use this once to initialize the FF to a known state.
+ * After that, keep track of it. :-)
+ * --- In order to do that, the DMA routines below should ---
+ * --- only be used while holding the DMA lock ! ---
+ */
+static __inline__ void clear_dma_ff(unsigned int dmanr)
+{
+	if (dmanr <= 3)
+		dma_outb(0, DMA1_CLEAR_FF_REG);
+	else
+		dma_outb(0, DMA2_CLEAR_FF_REG);
+}
+
+/* set mode (above) for a specific DMA channel */
+static __inline__ void set_dma_mode(unsigned int dmanr, char mode)
+{
+	if (dmanr <= 3)
+		dma_outb(mode | dmanr, DMA1_MODE_REG);
+	else
+		dma_outb(mode | (dmanr & 3), DMA2_MODE_REG);
+}
+
+/* Set only the page register bits of the transfer address.
+ * This is used for successive transfers when we know the contents of
+ * the lower 16 bits of the DMA current address register, but a 64k boundary
+ * may have been crossed.
+ */
+static __inline__ void set_dma_page(unsigned int dmanr, char pagenr)
+{
+	switch (dmanr) {
+	case 0:
+		dma_outb(pagenr, DMA_PAGE_0);
+		break;
+	case 1:
+		dma_outb(pagenr, DMA_PAGE_1);
+		break;
+	case 2:
+		dma_outb(pagenr, DMA_PAGE_2);
+		break;
+	case 3:
+		dma_outb(pagenr, DMA_PAGE_3);
+		break;
+	case 5:
+		dma_outb(pagenr & 0xfe, DMA_PAGE_5);
+		break;
+	case 6:
+		dma_outb(pagenr & 0xfe, DMA_PAGE_6);
+		break;
+	case 7:
+		dma_outb(pagenr & 0xfe, DMA_PAGE_7);
+		break;
+	}
+}
+
+
+/* Set transfer address & page bits for specific DMA channel.
+ * Assumes dma flipflop is clear.
+ */
+static __inline__ void set_dma_addr(unsigned int dmanr, unsigned int a)
+{
+	set_dma_page(dmanr, a>>16);
+	if (dmanr <= 3)  {
+		dma_outb(a & 0xff, ((dmanr & 3) << 1) + IO_DMA1_BASE);
+		dma_outb((a >> 8) & 0xff, ((dmanr & 3) << 1) + IO_DMA1_BASE);
+	}  else  {
+	    dma_outb((a >> 1) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
+	    dma_outb((a >> 9) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
+	}
+}
+
+
+/* Set transfer size (max 64k for DMA0..3, 128k for DMA5..7) for
+ * a specific DMA channel.
+ * You must ensure the parameters are valid.
+ * NOTE: from a manual: "the number of transfers is one more
+ * than the initial word count"! This is taken into account.
+ * Assumes dma flip-flop is clear.
+ * NOTE 2: "count" represents _bytes_ and must be even for channels 5-7.
+ */
+static __inline__ void set_dma_count(unsigned int dmanr, unsigned int count)
+{
+	count--;
+	if (dmanr <= 3)  {
+	    dma_outb(count & 0xff, ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
+	    dma_outb((count >> 8) & 0xff,
+		     ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
+	} else {
+	    dma_outb((count >> 1) & 0xff,
+		     ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
+	    dma_outb((count >> 9) & 0xff,
+		     ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
+	}
+}
+
+
+/* Get DMA residue count. After a DMA transfer, this
+ * should return zero. Reading this while a DMA transfer is
+ * still in progress will return unpredictable results.
+ * If called before the channel has been used, it may return 1.
+ * Otherwise, it returns the number of _bytes_ left to transfer.
+ *
+ * Assumes DMA flip-flop is clear.
+ */
+static __inline__ int get_dma_residue(unsigned int dmanr)
+{
+	unsigned int io_port;
+	/* using short to get 16-bit wrap around */
+	unsigned short count;
+
+	io_port = (dmanr <= 3) ? ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE
+		: ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE;
+
+	count = 1 + dma_inb(io_port);
+	count += dma_inb(io_port) << 8;
+
+	return (dmanr <= 3) ? count : (count << 1);
+}
+
+
+/* These are in kernel/dma.c: */
+extern int request_dma(unsigned int dmanr, const char *device_id);
+extern void free_dma(unsigned int dmanr);
+
+/* From PCI */
+
+#ifdef CONFIG_PCI
+extern int isa_dma_bridge_buggy;
 #else
-# include "dma_64.h"
+#define isa_dma_bridge_buggy	(0)
 #endif
+
+#endif /* _ASM_X86_DMA_H */
diff --git a/include/asm-x86/dma_32.h b/include/asm-x86/dma_32.h
deleted file mode 100644
index d23aac8e1a50..000000000000
--- a/include/asm-x86/dma_32.h
+++ /dev/null
@@ -1,297 +0,0 @@
-/* $Id: dma.h,v 1.7 1992/12/14 00:29:34 root Exp root $
- * linux/include/asm/dma.h: Defines for using and allocating dma channels.
- * Written by Hennus Bergman, 1992.
- * High DMA channel support & info by Hannu Savolainen
- * and John Boyd, Nov. 1992.
- */
-
-#ifndef _ASM_DMA_H
-#define _ASM_DMA_H
-
-#include <linux/spinlock.h>	/* And spinlocks */
-#include <asm/io.h>		/* need byte IO */
-#include <linux/delay.h>
-
-
-#ifdef HAVE_REALLY_SLOW_DMA_CONTROLLER
-#define dma_outb	outb_p
-#else
-#define dma_outb	outb
-#endif
-
-#define dma_inb		inb
-
-/*
- * NOTES about DMA transfers:
- *
- *  controller 1: channels 0-3, byte operations, ports 00-1F
- *  controller 2: channels 4-7, word operations, ports C0-DF
- *
- *  - ALL registers are 8 bits only, regardless of transfer size
- *  - channel 4 is not used - cascades 1 into 2.
- *  - channels 0-3 are byte - addresses/counts are for physical bytes
- *  - channels 5-7 are word - addresses/counts are for physical words
- *  - transfers must not cross physical 64K (0-3) or 128K (5-7) boundaries
- *  - transfer count loaded to registers is 1 less than actual count
- *  - controller 2 offsets are all even (2x offsets for controller 1)
- *  - page registers for 5-7 don't use data bit 0, represent 128K pages
- *  - page registers for 0-3 use bit 0, represent 64K pages
- *
- * DMA transfers are limited to the lower 16MB of _physical_ memory.  
- * Note that addresses loaded into registers must be _physical_ addresses,
- * not logical addresses (which may differ if paging is active).
- *
- *  Address mapping for channels 0-3:
- *
- *   A23 ... A16 A15 ... A8  A7 ... A0    (Physical addresses)
- *    |  ...  |   |  ... |   |  ... |
- *    |  ...  |   |  ... |   |  ... |
- *    |  ...  |   |  ... |   |  ... |
- *   P7  ...  P0  A7 ... A0  A7 ... A0   
- * |    Page    | Addr MSB | Addr LSB |   (DMA registers)
- *
- *  Address mapping for channels 5-7:
- *
- *   A23 ... A17 A16 A15 ... A9 A8 A7 ... A1 A0    (Physical addresses)
- *    |  ...  |   \   \   ... \  \  \  ... \  \
- *    |  ...  |    \   \   ... \  \  \  ... \  (not used)
- *    |  ...  |     \   \   ... \  \  \  ... \
- *   P7  ...  P1 (0) A7 A6  ... A0 A7 A6 ... A0   
- * |      Page      |  Addr MSB   |  Addr LSB  |   (DMA registers)
- *
- * Again, channels 5-7 transfer _physical_ words (16 bits), so addresses
- * and counts _must_ be word-aligned (the lowest address bit is _ignored_ at
- * the hardware level, so odd-byte transfers aren't possible).
- *
- * Transfer count (_not # bytes_) is limited to 64K, represented as actual
- * count - 1 : 64K => 0xFFFF, 1 => 0x0000.  Thus, count is always 1 or more,
- * and up to 128K bytes may be transferred on channels 5-7 in one operation. 
- *
- */
-
-#define MAX_DMA_CHANNELS	8
-
-/* The maximum address that we can perform a DMA transfer to on this platform */
-#define MAX_DMA_ADDRESS      (PAGE_OFFSET+0x1000000)
-
-/* 8237 DMA controllers */
-#define IO_DMA1_BASE	0x00	/* 8 bit slave DMA, channels 0..3 */
-#define IO_DMA2_BASE	0xC0	/* 16 bit master DMA, ch 4(=slave input)..7 */
-
-/* DMA controller registers */
-#define DMA1_CMD_REG		0x08	/* command register (w) */
-#define DMA1_STAT_REG		0x08	/* status register (r) */
-#define DMA1_REQ_REG            0x09    /* request register (w) */
-#define DMA1_MASK_REG		0x0A	/* single-channel mask (w) */
-#define DMA1_MODE_REG		0x0B	/* mode register (w) */
-#define DMA1_CLEAR_FF_REG	0x0C	/* clear pointer flip-flop (w) */
-#define DMA1_TEMP_REG           0x0D    /* Temporary Register (r) */
-#define DMA1_RESET_REG		0x0D	/* Master Clear (w) */
-#define DMA1_CLR_MASK_REG       0x0E    /* Clear Mask */
-#define DMA1_MASK_ALL_REG       0x0F    /* all-channels mask (w) */
-
-#define DMA2_CMD_REG		0xD0	/* command register (w) */
-#define DMA2_STAT_REG		0xD0	/* status register (r) */
-#define DMA2_REQ_REG            0xD2    /* request register (w) */
-#define DMA2_MASK_REG		0xD4	/* single-channel mask (w) */
-#define DMA2_MODE_REG		0xD6	/* mode register (w) */
-#define DMA2_CLEAR_FF_REG	0xD8	/* clear pointer flip-flop (w) */
-#define DMA2_TEMP_REG           0xDA    /* Temporary Register (r) */
-#define DMA2_RESET_REG		0xDA	/* Master Clear (w) */
-#define DMA2_CLR_MASK_REG       0xDC    /* Clear Mask */
-#define DMA2_MASK_ALL_REG       0xDE    /* all-channels mask (w) */
-
-#define DMA_ADDR_0              0x00    /* DMA address registers */
-#define DMA_ADDR_1              0x02
-#define DMA_ADDR_2              0x04
-#define DMA_ADDR_3              0x06
-#define DMA_ADDR_4              0xC0
-#define DMA_ADDR_5              0xC4
-#define DMA_ADDR_6              0xC8
-#define DMA_ADDR_7              0xCC
-
-#define DMA_CNT_0               0x01    /* DMA count registers */
-#define DMA_CNT_1               0x03
-#define DMA_CNT_2               0x05
-#define DMA_CNT_3               0x07
-#define DMA_CNT_4               0xC2
-#define DMA_CNT_5               0xC6
-#define DMA_CNT_6               0xCA
-#define DMA_CNT_7               0xCE
-
-#define DMA_PAGE_0              0x87    /* DMA page registers */
-#define DMA_PAGE_1              0x83
-#define DMA_PAGE_2              0x81
-#define DMA_PAGE_3              0x82
-#define DMA_PAGE_5              0x8B
-#define DMA_PAGE_6              0x89
-#define DMA_PAGE_7              0x8A
-
-#define DMA_MODE_READ	0x44	/* I/O to memory, no autoinit, increment, single mode */
-#define DMA_MODE_WRITE	0x48	/* memory to I/O, no autoinit, increment, single mode */
-#define DMA_MODE_CASCADE 0xC0   /* pass thru DREQ->HRQ, DACK<-HLDA only */
-
-#define DMA_AUTOINIT	0x10
-
-
-extern spinlock_t  dma_spin_lock;
-
-static __inline__ unsigned long claim_dma_lock(void)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&dma_spin_lock, flags);
-	return flags;
-}
-
-static __inline__ void release_dma_lock(unsigned long flags)
-{
-	spin_unlock_irqrestore(&dma_spin_lock, flags);
-}
-
-/* enable/disable a specific DMA channel */
-static __inline__ void enable_dma(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(dmanr,  DMA1_MASK_REG);
-	else
-		dma_outb(dmanr & 3,  DMA2_MASK_REG);
-}
-
-static __inline__ void disable_dma(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(dmanr | 4,  DMA1_MASK_REG);
-	else
-		dma_outb((dmanr & 3) | 4,  DMA2_MASK_REG);
-}
-
-/* Clear the 'DMA Pointer Flip Flop'.
- * Write 0 for LSB/MSB, 1 for MSB/LSB access.
- * Use this once to initialize the FF to a known state.
- * After that, keep track of it. :-)
- * --- In order to do that, the DMA routines below should ---
- * --- only be used while holding the DMA lock ! ---
- */
-static __inline__ void clear_dma_ff(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(0,  DMA1_CLEAR_FF_REG);
-	else
-		dma_outb(0,  DMA2_CLEAR_FF_REG);
-}
-
-/* set mode (above) for a specific DMA channel */
-static __inline__ void set_dma_mode(unsigned int dmanr, char mode)
-{
-	if (dmanr<=3)
-		dma_outb(mode | dmanr,  DMA1_MODE_REG);
-	else
-		dma_outb(mode | (dmanr&3),  DMA2_MODE_REG);
-}
-
-/* Set only the page register bits of the transfer address.
- * This is used for successive transfers when we know the contents of
- * the lower 16 bits of the DMA current address register, but a 64k boundary
- * may have been crossed.
- */
-static __inline__ void set_dma_page(unsigned int dmanr, char pagenr)
-{
-	switch(dmanr) {
-		case 0:
-			dma_outb(pagenr, DMA_PAGE_0);
-			break;
-		case 1:
-			dma_outb(pagenr, DMA_PAGE_1);
-			break;
-		case 2:
-			dma_outb(pagenr, DMA_PAGE_2);
-			break;
-		case 3:
-			dma_outb(pagenr, DMA_PAGE_3);
-			break;
-		case 5:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_5);
-			break;
-		case 6:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_6);
-			break;
-		case 7:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_7);
-			break;
-	}
-}
-
-
-/* Set transfer address & page bits for specific DMA channel.
- * Assumes dma flipflop is clear.
- */
-static __inline__ void set_dma_addr(unsigned int dmanr, unsigned int a)
-{
-	set_dma_page(dmanr, a>>16);
-	if (dmanr <= 3)  {
-	    dma_outb( a & 0xff, ((dmanr&3)<<1) + IO_DMA1_BASE );
-            dma_outb( (a>>8) & 0xff, ((dmanr&3)<<1) + IO_DMA1_BASE );
-	}  else  {
-	    dma_outb( (a>>1) & 0xff, ((dmanr&3)<<2) + IO_DMA2_BASE );
-	    dma_outb( (a>>9) & 0xff, ((dmanr&3)<<2) + IO_DMA2_BASE );
-	}
-}
-
-
-/* Set transfer size (max 64k for DMA0..3, 128k for DMA5..7) for
- * a specific DMA channel.
- * You must ensure the parameters are valid.
- * NOTE: from a manual: "the number of transfers is one more
- * than the initial word count"! This is taken into account.
- * Assumes dma flip-flop is clear.
- * NOTE 2: "count" represents _bytes_ and must be even for channels 5-7.
- */
-static __inline__ void set_dma_count(unsigned int dmanr, unsigned int count)
-{
-        count--;
-	if (dmanr <= 3)  {
-	    dma_outb( count & 0xff, ((dmanr&3)<<1) + 1 + IO_DMA1_BASE );
-	    dma_outb( (count>>8) & 0xff, ((dmanr&3)<<1) + 1 + IO_DMA1_BASE );
-        } else {
-	    dma_outb( (count>>1) & 0xff, ((dmanr&3)<<2) + 2 + IO_DMA2_BASE );
-	    dma_outb( (count>>9) & 0xff, ((dmanr&3)<<2) + 2 + IO_DMA2_BASE );
-        }
-}
-
-
-/* Get DMA residue count. After a DMA transfer, this
- * should return zero. Reading this while a DMA transfer is
- * still in progress will return unpredictable results.
- * If called before the channel has been used, it may return 1.
- * Otherwise, it returns the number of _bytes_ left to transfer.
- *
- * Assumes DMA flip-flop is clear.
- */
-static __inline__ int get_dma_residue(unsigned int dmanr)
-{
-	unsigned int io_port = (dmanr<=3)? ((dmanr&3)<<1) + 1 + IO_DMA1_BASE
-					 : ((dmanr&3)<<2) + 2 + IO_DMA2_BASE;
-
-	/* using short to get 16-bit wrap around */
-	unsigned short count;
-
-	count = 1 + dma_inb(io_port);
-	count += dma_inb(io_port) << 8;
-	
-	return (dmanr<=3)? count : (count<<1);
-}
-
-
-/* These are in kernel/dma.c: */
-extern int request_dma(unsigned int dmanr, const char * device_id);	/* reserve a DMA channel */
-extern void free_dma(unsigned int dmanr);	/* release it again */
-
-/* From PCI */
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy 	(0)
-#endif
-
-#endif /* _ASM_DMA_H */
diff --git a/include/asm-x86/dma_64.h b/include/asm-x86/dma_64.h
deleted file mode 100644
index a37c16f06289..000000000000
--- a/include/asm-x86/dma_64.h
+++ /dev/null
@@ -1,304 +0,0 @@
-/*
- * linux/include/asm/dma.h: Defines for using and allocating dma channels.
- * Written by Hennus Bergman, 1992.
- * High DMA channel support & info by Hannu Savolainen
- * and John Boyd, Nov. 1992.
- */
-
-#ifndef _ASM_DMA_H
-#define _ASM_DMA_H
-
-#include <linux/spinlock.h>	/* And spinlocks */
-#include <asm/io.h>		/* need byte IO */
-#include <linux/delay.h>
-
-
-#ifdef HAVE_REALLY_SLOW_DMA_CONTROLLER
-#define dma_outb	outb_p
-#else
-#define dma_outb	outb
-#endif
-
-#define dma_inb		inb
-
-/*
- * NOTES about DMA transfers:
- *
- *  controller 1: channels 0-3, byte operations, ports 00-1F
- *  controller 2: channels 4-7, word operations, ports C0-DF
- *
- *  - ALL registers are 8 bits only, regardless of transfer size
- *  - channel 4 is not used - cascades 1 into 2.
- *  - channels 0-3 are byte - addresses/counts are for physical bytes
- *  - channels 5-7 are word - addresses/counts are for physical words
- *  - transfers must not cross physical 64K (0-3) or 128K (5-7) boundaries
- *  - transfer count loaded to registers is 1 less than actual count
- *  - controller 2 offsets are all even (2x offsets for controller 1)
- *  - page registers for 5-7 don't use data bit 0, represent 128K pages
- *  - page registers for 0-3 use bit 0, represent 64K pages
- *
- * DMA transfers are limited to the lower 16MB of _physical_ memory.  
- * Note that addresses loaded into registers must be _physical_ addresses,
- * not logical addresses (which may differ if paging is active).
- *
- *  Address mapping for channels 0-3:
- *
- *   A23 ... A16 A15 ... A8  A7 ... A0    (Physical addresses)
- *    |  ...  |   |  ... |   |  ... |
- *    |  ...  |   |  ... |   |  ... |
- *    |  ...  |   |  ... |   |  ... |
- *   P7  ...  P0  A7 ... A0  A7 ... A0   
- * |    Page    | Addr MSB | Addr LSB |   (DMA registers)
- *
- *  Address mapping for channels 5-7:
- *
- *   A23 ... A17 A16 A15 ... A9 A8 A7 ... A1 A0    (Physical addresses)
- *    |  ...  |   \   \   ... \  \  \  ... \  \
- *    |  ...  |    \   \   ... \  \  \  ... \  (not used)
- *    |  ...  |     \   \   ... \  \  \  ... \
- *   P7  ...  P1 (0) A7 A6  ... A0 A7 A6 ... A0   
- * |      Page      |  Addr MSB   |  Addr LSB  |   (DMA registers)
- *
- * Again, channels 5-7 transfer _physical_ words (16 bits), so addresses
- * and counts _must_ be word-aligned (the lowest address bit is _ignored_ at
- * the hardware level, so odd-byte transfers aren't possible).
- *
- * Transfer count (_not # bytes_) is limited to 64K, represented as actual
- * count - 1 : 64K => 0xFFFF, 1 => 0x0000.  Thus, count is always 1 or more,
- * and up to 128K bytes may be transferred on channels 5-7 in one operation. 
- *
- */
-
-#define MAX_DMA_CHANNELS	8
-
-
-/* 16MB ISA DMA zone */
-#define MAX_DMA_PFN   ((16*1024*1024) >> PAGE_SHIFT)
-
-/* 4GB broken PCI/AGP hardware bus master zone */
-#define MAX_DMA32_PFN ((4UL*1024*1024*1024) >> PAGE_SHIFT)
-
-/* Compat define for old dma zone */
-#define MAX_DMA_ADDRESS ((unsigned long)__va(MAX_DMA_PFN << PAGE_SHIFT))
-
-/* 8237 DMA controllers */
-#define IO_DMA1_BASE	0x00	/* 8 bit slave DMA, channels 0..3 */
-#define IO_DMA2_BASE	0xC0	/* 16 bit master DMA, ch 4(=slave input)..7 */
-
-/* DMA controller registers */
-#define DMA1_CMD_REG		0x08	/* command register (w) */
-#define DMA1_STAT_REG		0x08	/* status register (r) */
-#define DMA1_REQ_REG            0x09    /* request register (w) */
-#define DMA1_MASK_REG		0x0A	/* single-channel mask (w) */
-#define DMA1_MODE_REG		0x0B	/* mode register (w) */
-#define DMA1_CLEAR_FF_REG	0x0C	/* clear pointer flip-flop (w) */
-#define DMA1_TEMP_REG           0x0D    /* Temporary Register (r) */
-#define DMA1_RESET_REG		0x0D	/* Master Clear (w) */
-#define DMA1_CLR_MASK_REG       0x0E    /* Clear Mask */
-#define DMA1_MASK_ALL_REG       0x0F    /* all-channels mask (w) */
-
-#define DMA2_CMD_REG		0xD0	/* command register (w) */
-#define DMA2_STAT_REG		0xD0	/* status register (r) */
-#define DMA2_REQ_REG            0xD2    /* request register (w) */
-#define DMA2_MASK_REG		0xD4	/* single-channel mask (w) */
-#define DMA2_MODE_REG		0xD6	/* mode register (w) */
-#define DMA2_CLEAR_FF_REG	0xD8	/* clear pointer flip-flop (w) */
-#define DMA2_TEMP_REG           0xDA    /* Temporary Register (r) */
-#define DMA2_RESET_REG		0xDA	/* Master Clear (w) */
-#define DMA2_CLR_MASK_REG       0xDC    /* Clear Mask */
-#define DMA2_MASK_ALL_REG       0xDE    /* all-channels mask (w) */
-
-#define DMA_ADDR_0              0x00    /* DMA address registers */
-#define DMA_ADDR_1              0x02
-#define DMA_ADDR_2              0x04
-#define DMA_ADDR_3              0x06
-#define DMA_ADDR_4              0xC0
-#define DMA_ADDR_5              0xC4
-#define DMA_ADDR_6              0xC8
-#define DMA_ADDR_7              0xCC
-
-#define DMA_CNT_0               0x01    /* DMA count registers */
-#define DMA_CNT_1               0x03
-#define DMA_CNT_2               0x05
-#define DMA_CNT_3               0x07
-#define DMA_CNT_4               0xC2
-#define DMA_CNT_5               0xC6
-#define DMA_CNT_6               0xCA
-#define DMA_CNT_7               0xCE
-
-#define DMA_PAGE_0              0x87    /* DMA page registers */
-#define DMA_PAGE_1              0x83
-#define DMA_PAGE_2              0x81
-#define DMA_PAGE_3              0x82
-#define DMA_PAGE_5              0x8B
-#define DMA_PAGE_6              0x89
-#define DMA_PAGE_7              0x8A
-
-#define DMA_MODE_READ	0x44	/* I/O to memory, no autoinit, increment, single mode */
-#define DMA_MODE_WRITE	0x48	/* memory to I/O, no autoinit, increment, single mode */
-#define DMA_MODE_CASCADE 0xC0   /* pass thru DREQ->HRQ, DACK<-HLDA only */
-
-#define DMA_AUTOINIT	0x10
-
-
-extern spinlock_t  dma_spin_lock;
-
-static __inline__ unsigned long claim_dma_lock(void)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&dma_spin_lock, flags);
-	return flags;
-}
-
-static __inline__ void release_dma_lock(unsigned long flags)
-{
-	spin_unlock_irqrestore(&dma_spin_lock, flags);
-}
-
-/* enable/disable a specific DMA channel */
-static __inline__ void enable_dma(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(dmanr,  DMA1_MASK_REG);
-	else
-		dma_outb(dmanr & 3,  DMA2_MASK_REG);
-}
-
-static __inline__ void disable_dma(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(dmanr | 4,  DMA1_MASK_REG);
-	else
-		dma_outb((dmanr & 3) | 4,  DMA2_MASK_REG);
-}
-
-/* Clear the 'DMA Pointer Flip Flop'.
- * Write 0 for LSB/MSB, 1 for MSB/LSB access.
- * Use this once to initialize the FF to a known state.
- * After that, keep track of it. :-)
- * --- In order to do that, the DMA routines below should ---
- * --- only be used while holding the DMA lock ! ---
- */
-static __inline__ void clear_dma_ff(unsigned int dmanr)
-{
-	if (dmanr<=3)
-		dma_outb(0,  DMA1_CLEAR_FF_REG);
-	else
-		dma_outb(0,  DMA2_CLEAR_FF_REG);
-}
-
-/* set mode (above) for a specific DMA channel */
-static __inline__ void set_dma_mode(unsigned int dmanr, char mode)
-{
-	if (dmanr<=3)
-		dma_outb(mode | dmanr,  DMA1_MODE_REG);
-	else
-		dma_outb(mode | (dmanr&3),  DMA2_MODE_REG);
-}
-
-/* Set only the page register bits of the transfer address.
- * This is used for successive transfers when we know the contents of
- * the lower 16 bits of the DMA current address register, but a 64k boundary
- * may have been crossed.
- */
-static __inline__ void set_dma_page(unsigned int dmanr, char pagenr)
-{
-	switch(dmanr) {
-		case 0:
-			dma_outb(pagenr, DMA_PAGE_0);
-			break;
-		case 1:
-			dma_outb(pagenr, DMA_PAGE_1);
-			break;
-		case 2:
-			dma_outb(pagenr, DMA_PAGE_2);
-			break;
-		case 3:
-			dma_outb(pagenr, DMA_PAGE_3);
-			break;
-		case 5:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_5);
-			break;
-		case 6:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_6);
-			break;
-		case 7:
-			dma_outb(pagenr & 0xfe, DMA_PAGE_7);
-			break;
-	}
-}
-
-
-/* Set transfer address & page bits for specific DMA channel.
- * Assumes dma flipflop is clear.
- */
-static __inline__ void set_dma_addr(unsigned int dmanr, unsigned int a)
-{
-	set_dma_page(dmanr, a>>16);
-	if (dmanr <= 3)  {
-	    dma_outb( a & 0xff, ((dmanr&3)<<1) + IO_DMA1_BASE );
-            dma_outb( (a>>8) & 0xff, ((dmanr&3)<<1) + IO_DMA1_BASE );
-	}  else  {
-	    dma_outb( (a>>1) & 0xff, ((dmanr&3)<<2) + IO_DMA2_BASE );
-	    dma_outb( (a>>9) & 0xff, ((dmanr&3)<<2) + IO_DMA2_BASE );
-	}
-}
-
-
-/* Set transfer size (max 64k for DMA1..3, 128k for DMA5..7) for
- * a specific DMA channel.
- * You must ensure the parameters are valid.
- * NOTE: from a manual: "the number of transfers is one more
- * than the initial word count"! This is taken into account.
- * Assumes dma flip-flop is clear.
- * NOTE 2: "count" represents _bytes_ and must be even for channels 5-7.
- */
-static __inline__ void set_dma_count(unsigned int dmanr, unsigned int count)
-{
-        count--;
-	if (dmanr <= 3)  {
-	    dma_outb( count & 0xff, ((dmanr&3)<<1) + 1 + IO_DMA1_BASE );
-	    dma_outb( (count>>8) & 0xff, ((dmanr&3)<<1) + 1 + IO_DMA1_BASE );
-        } else {
-	    dma_outb( (count>>1) & 0xff, ((dmanr&3)<<2) + 2 + IO_DMA2_BASE );
-	    dma_outb( (count>>9) & 0xff, ((dmanr&3)<<2) + 2 + IO_DMA2_BASE );
-        }
-}
-
-
-/* Get DMA residue count. After a DMA transfer, this
- * should return zero. Reading this while a DMA transfer is
- * still in progress will return unpredictable results.
- * If called before the channel has been used, it may return 1.
- * Otherwise, it returns the number of _bytes_ left to transfer.
- *
- * Assumes DMA flip-flop is clear.
- */
-static __inline__ int get_dma_residue(unsigned int dmanr)
-{
-	unsigned int io_port = (dmanr<=3)? ((dmanr&3)<<1) + 1 + IO_DMA1_BASE
-					 : ((dmanr&3)<<2) + 2 + IO_DMA2_BASE;
-
-	/* using short to get 16-bit wrap around */
-	unsigned short count;
-
-	count = 1 + dma_inb(io_port);
-	count += dma_inb(io_port) << 8;
-	
-	return (dmanr<=3)? count : (count<<1);
-}
-
-
-/* These are in kernel/dma.c: */
-extern int request_dma(unsigned int dmanr, const char * device_id);	/* reserve a DMA channel */
-extern void free_dma(unsigned int dmanr);	/* release it again */
-
-/* From PCI */
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy 	(0)
-#endif
-
-#endif /* _ASM_DMA_H */
-- 
cgit v1.2.3


From 3ebc51d7c95425c3b4667fa042576fb1c6e2ea16 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:21 +0100
Subject: x86: merge include/asm-x86/scatterlist.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/scatterlist.h    | 34 ++++++++++++++++++++++++++++++++--
 include/asm-x86/scatterlist_32.h | 28 ----------------------------
 include/asm-x86/scatterlist_64.h | 29 -----------------------------
 3 files changed, 32 insertions(+), 59 deletions(-)
 delete mode 100644 include/asm-x86/scatterlist_32.h
 delete mode 100644 include/asm-x86/scatterlist_64.h

(limited to 'include')

diff --git a/include/asm-x86/scatterlist.h b/include/asm-x86/scatterlist.h
index 3a1e76257a27..d13c197866d6 100644
--- a/include/asm-x86/scatterlist.h
+++ b/include/asm-x86/scatterlist.h
@@ -1,5 +1,35 @@
+#ifndef _ASM_X86_SCATTERLIST_H
+#define _ASM_X86_SCATTERLIST_H
+
+#include <asm/types.h>
+
+struct scatterlist {
+#ifdef CONFIG_DEBUG_SG
+	unsigned long	sg_magic;
+#endif
+	unsigned long	page_link;
+	unsigned int	offset;
+	unsigned int	length;
+	dma_addr_t	dma_address;
+#ifdef CONFIG_X86_64
+	unsigned int	dma_length;
+#endif
+};
+
+#define ARCH_HAS_SG_CHAIN
+#define ISA_DMA_THRESHOLD (0x00ffffff)
+
+/*
+ * These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
 #ifdef CONFIG_X86_32
-# include "scatterlist_32.h"
+# define sg_dma_len(sg)		((sg)->length)
 #else
-# include "scatterlist_64.h"
+# define sg_dma_len(sg)		((sg)->dma_length)
+#endif
+
 #endif
diff --git a/include/asm-x86/scatterlist_32.h b/include/asm-x86/scatterlist_32.h
deleted file mode 100644
index 0e7d997a34be..000000000000
--- a/include/asm-x86/scatterlist_32.h
+++ /dev/null
@@ -1,28 +0,0 @@
-#ifndef _I386_SCATTERLIST_H
-#define _I386_SCATTERLIST_H
-
-#include <asm/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-    unsigned long	sg_magic;
-#endif
-    unsigned long	page_link;
-    unsigned int	offset;
-    dma_addr_t		dma_address;
-    unsigned int	length;
-};
-
-#define ARCH_HAS_SG_CHAIN
-
-/* These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns.
- */
-#define sg_dma_address(sg)	((sg)->dma_address)
-#define sg_dma_len(sg)		((sg)->length)
-
-#define ISA_DMA_THRESHOLD (0x00ffffff)
-
-#endif /* !(_I386_SCATTERLIST_H) */
diff --git a/include/asm-x86/scatterlist_64.h b/include/asm-x86/scatterlist_64.h
deleted file mode 100644
index 1847c72befeb..000000000000
--- a/include/asm-x86/scatterlist_64.h
+++ /dev/null
@@ -1,29 +0,0 @@
-#ifndef _X8664_SCATTERLIST_H
-#define _X8664_SCATTERLIST_H
-
-#include <asm/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-    unsigned long	sg_magic;
-#endif
-    unsigned long	page_link;
-    unsigned int	offset;
-    unsigned int	length;
-    dma_addr_t		dma_address;
-    unsigned int        dma_length;
-};
-
-#define ARCH_HAS_SG_CHAIN
-
-#define ISA_DMA_THRESHOLD (0x00ffffff)
-
-/* These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns.
- */
-#define sg_dma_address(sg)     ((sg)->dma_address)
-#define sg_dma_len(sg)         ((sg)->dma_length)
-
-#endif 
-- 
cgit v1.2.3


From 6ce60b07e670e800c4c5cfe984ed5188e7a64135 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:26 +0100
Subject: x86: unify mc146818rtc.h - prepare for sharing rtc code

Unify mc146818rtc.h by adding the rtc_cmos_read/write functions to
time_64.c. This is a preparatory patch to finaly share the rtc code,
which is unsurprisingly similar.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/time_64.c        | 21 +++++++++
 include/asm-x86/mc146818rtc.h    | 98 ++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/mc146818rtc_32.h | 97 ---------------------------------------
 include/asm-x86/mc146818rtc_64.h | 29 ------------
 4 files changed, 116 insertions(+), 129 deletions(-)
 delete mode 100644 include/asm-x86/mc146818rtc_32.h
 delete mode 100644 include/asm-x86/mc146818rtc_64.h

(limited to 'include')

diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index 368b1942b39a..0a01504586a5 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -69,6 +69,27 @@ unsigned long profile_pc(struct pt_regs *regs)
 }
 EXPORT_SYMBOL(profile_pc);
 
+/* Routines for accessing the CMOS RAM/RTC. */
+unsigned char rtc_cmos_read(unsigned char addr)
+{
+	unsigned char val;
+	lock_cmos_prefix(addr);
+	outb_p(addr, RTC_PORT(0));
+	val = inb_p(RTC_PORT(1));
+	lock_cmos_suffix(addr);
+	return val;
+}
+EXPORT_SYMBOL(rtc_cmos_read);
+
+void rtc_cmos_write(unsigned char val, unsigned char addr)
+{
+	lock_cmos_prefix(addr);
+	outb_p(addr, RTC_PORT(0));
+	outb_p(val, RTC_PORT(1));
+	lock_cmos_suffix(addr);
+}
+EXPORT_SYMBOL(rtc_cmos_write);
+
 /*
  * In order to set the CMOS clock precisely, set_rtc_mmss has to be called 500
  * ms after the second nowtime has started, because when nowtime is written
diff --git a/include/asm-x86/mc146818rtc.h b/include/asm-x86/mc146818rtc.h
index 5c2bb66caf17..9d39436df231 100644
--- a/include/asm-x86/mc146818rtc.h
+++ b/include/asm-x86/mc146818rtc.h
@@ -1,5 +1,97 @@
-#ifdef CONFIG_X86_32
-# include "mc146818rtc_32.h"
+/*
+ * Machine dependent access functions for RTC registers.
+ */
+#ifndef _ASM_MC146818RTC_H
+#define _ASM_MC146818RTC_H
+
+#include <asm/io.h>
+#include <asm/system.h>
+#include <asm/processor.h>
+#include <linux/mc146818rtc.h>
+
+#ifndef RTC_PORT
+#define RTC_PORT(x)	(0x70 + (x))
+#define RTC_ALWAYS_BCD	1	/* RTC operates in binary mode */
+#endif
+
+#if defined(CONFIG_X86_32) && defined(__HAVE_ARCH_CMPXCHG)
+/*
+ * This lock provides nmi access to the CMOS/RTC registers.  It has some
+ * special properties.  It is owned by a CPU and stores the index register
+ * currently being accessed (if owned).  The idea here is that it works
+ * like a normal lock (normally).  However, in an NMI, the NMI code will
+ * first check to see if its CPU owns the lock, meaning that the NMI
+ * interrupted during the read/write of the device.  If it does, it goes ahead
+ * and performs the access and then restores the index register.  If it does
+ * not, it locks normally.
+ *
+ * Note that since we are working with NMIs, we need this lock even in
+ * a non-SMP machine just to mark that the lock is owned.
+ *
+ * This only works with compare-and-swap.  There is no other way to
+ * atomically claim the lock and set the owner.
+ */
+#include <linux/smp.h>
+extern volatile unsigned long cmos_lock;
+
+/*
+ * All of these below must be called with interrupts off, preempt
+ * disabled, etc.
+ */
+
+static inline void lock_cmos(unsigned char reg)
+{
+	unsigned long new;
+	new = ((smp_processor_id()+1) << 8) | reg;
+	for (;;) {
+		if (cmos_lock) {
+			cpu_relax();
+			continue;
+		}
+		if (__cmpxchg(&cmos_lock, 0, new, sizeof(cmos_lock)) == 0)
+			return;
+	}
+}
+
+static inline void unlock_cmos(void)
+{
+	cmos_lock = 0;
+}
+static inline int do_i_have_lock_cmos(void)
+{
+	return (cmos_lock >> 8) == (smp_processor_id()+1);
+}
+static inline unsigned char current_lock_cmos_reg(void)
+{
+	return cmos_lock & 0xff;
+}
+#define lock_cmos_prefix(reg) \
+	do {					\
+		unsigned long cmos_flags;	\
+		local_irq_save(cmos_flags);	\
+		lock_cmos(reg)
+#define lock_cmos_suffix(reg) \
+		unlock_cmos();			\
+		local_irq_restore(cmos_flags);	\
+	} while (0)
 #else
-# include "mc146818rtc_64.h"
+#define lock_cmos_prefix(reg) do {} while (0)
+#define lock_cmos_suffix(reg) do {} while (0)
+#define lock_cmos(reg)
+#define unlock_cmos()
+#define do_i_have_lock_cmos() 0
+#define current_lock_cmos_reg() 0
 #endif
+
+/*
+ * The yet supported machines all access the RTC index register via
+ * an ISA port access but the way to access the date register differs ...
+ */
+#define CMOS_READ(addr) rtc_cmos_read(addr)
+#define CMOS_WRITE(val, addr) rtc_cmos_write(val, addr)
+unsigned char rtc_cmos_read(unsigned char addr);
+void rtc_cmos_write(unsigned char val, unsigned char addr);
+
+#define RTC_IRQ 8
+
+#endif /* _ASM_MC146818RTC_H */
diff --git a/include/asm-x86/mc146818rtc_32.h b/include/asm-x86/mc146818rtc_32.h
deleted file mode 100644
index 1613b42eaf58..000000000000
--- a/include/asm-x86/mc146818rtc_32.h
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
- * Machine dependent access functions for RTC registers.
- */
-#ifndef _ASM_MC146818RTC_H
-#define _ASM_MC146818RTC_H
-
-#include <asm/io.h>
-#include <asm/system.h>
-#include <asm/processor.h>
-#include <linux/mc146818rtc.h>
-
-#ifndef RTC_PORT
-#define RTC_PORT(x)	(0x70 + (x))
-#define RTC_ALWAYS_BCD	1	/* RTC operates in binary mode */
-#endif
-
-#ifdef __HAVE_ARCH_CMPXCHG
-/*
- * This lock provides nmi access to the CMOS/RTC registers.  It has some
- * special properties.  It is owned by a CPU and stores the index register
- * currently being accessed (if owned).  The idea here is that it works
- * like a normal lock (normally).  However, in an NMI, the NMI code will
- * first check to see if its CPU owns the lock, meaning that the NMI
- * interrupted during the read/write of the device.  If it does, it goes ahead
- * and performs the access and then restores the index register.  If it does
- * not, it locks normally.
- *
- * Note that since we are working with NMIs, we need this lock even in
- * a non-SMP machine just to mark that the lock is owned.
- *
- * This only works with compare-and-swap.  There is no other way to
- * atomically claim the lock and set the owner.
- */
-#include <linux/smp.h>
-extern volatile unsigned long cmos_lock;
-
-/*
- * All of these below must be called with interrupts off, preempt
- * disabled, etc.
- */
-
-static inline void lock_cmos(unsigned char reg)
-{
-	unsigned long new;
-	new = ((smp_processor_id()+1) << 8) | reg;
-	for (;;) {
-		if (cmos_lock) {
-			cpu_relax();
-			continue;
-		}
-		if (__cmpxchg(&cmos_lock, 0, new, sizeof(cmos_lock)) == 0)
-			return;
-	}
-}
-
-static inline void unlock_cmos(void)
-{
-	cmos_lock = 0;
-}
-static inline int do_i_have_lock_cmos(void)
-{
-	return (cmos_lock >> 8) == (smp_processor_id()+1);
-}
-static inline unsigned char current_lock_cmos_reg(void)
-{
-	return cmos_lock & 0xff;
-}
-#define lock_cmos_prefix(reg) \
-	do {					\
-		unsigned long cmos_flags;	\
-		local_irq_save(cmos_flags);	\
-		lock_cmos(reg)
-#define lock_cmos_suffix(reg) \
-		unlock_cmos();			\
-		local_irq_restore(cmos_flags);	\
-	} while (0)
-#else
-#define lock_cmos_prefix(reg) do {} while (0)
-#define lock_cmos_suffix(reg) do {} while (0)
-#define lock_cmos(reg)
-#define unlock_cmos()
-#define do_i_have_lock_cmos() 0
-#define current_lock_cmos_reg() 0
-#endif
-
-/*
- * The yet supported machines all access the RTC index register via
- * an ISA port access but the way to access the date register differs ...
- */
-#define CMOS_READ(addr) rtc_cmos_read(addr)
-#define CMOS_WRITE(val, addr) rtc_cmos_write(val, addr)
-unsigned char rtc_cmos_read(unsigned char addr);
-void rtc_cmos_write(unsigned char val, unsigned char addr);
-
-#define RTC_IRQ 8
-
-#endif /* _ASM_MC146818RTC_H */
diff --git a/include/asm-x86/mc146818rtc_64.h b/include/asm-x86/mc146818rtc_64.h
deleted file mode 100644
index d6e3009430c1..000000000000
--- a/include/asm-x86/mc146818rtc_64.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * Machine dependent access functions for RTC registers.
- */
-#ifndef _ASM_MC146818RTC_H
-#define _ASM_MC146818RTC_H
-
-#include <asm/io.h>
-
-#ifndef RTC_PORT
-#define RTC_PORT(x)	(0x70 + (x))
-#define RTC_ALWAYS_BCD	1	/* RTC operates in binary mode */
-#endif
-
-/*
- * The yet supported machines all access the RTC index register via
- * an ISA port access but the way to access the date register differs ...
- */
-#define CMOS_READ(addr) ({ \
-outb_p((addr),RTC_PORT(0)); \
-inb_p(RTC_PORT(1)); \
-})
-#define CMOS_WRITE(val, addr) ({ \
-outb_p((addr),RTC_PORT(0)); \
-outb_p((val),RTC_PORT(1)); \
-})
-
-#define RTC_IRQ 8
-
-#endif /* _ASM_MC146818RTC_H */
-- 
cgit v1.2.3


From fe599f9fbc5d470ec5b55d08f2bbb991ddecbbc8 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:26 +0100
Subject: x86: isolate the rtc code for sharing

The mach-default/mach_time.h code inline is moved to arch/x86/kernel/rtc.c
and the header files are adjusted.

Shrink the 3 dozen includes to the ones we really need.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/Makefile_32              |   2 +-
 arch/x86/kernel/rtc.c                    | 166 +++++++++++++++++++++++++++++++
 arch/x86/kernel/time_32.c                | 105 +------------------
 include/asm-x86/mach-default/mach_time.h | 111 ---------------------
 include/asm-x86/mc146818rtc.h            |   3 +
 include/asm-x86/time.h                   |   2 +-
 6 files changed, 175 insertions(+), 214 deletions(-)
 create mode 100644 arch/x86/kernel/rtc.c
 delete mode 100644 include/asm-x86/mach-default/mach_time.h

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index 31ff982bc26b..eb2da53578d7 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -8,7 +8,7 @@ CPPFLAGS_vmlinux.lds += -Ui386
 obj-y	:= process_32.o signal_32.o entry_32.o traps_32.o irq_32.o \
 		ptrace_32.o time_32.o ioport_32.o ldt.o setup_32.o i8259_32.o sys_i386_32.o \
 		pci-dma_32.o i386_ksyms_32.o i387_32.o bootflag.o e820_32.o\
-		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o
+		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o rtc.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c
new file mode 100644
index 000000000000..45bf54d9f4c5
--- /dev/null
+++ b/arch/x86/kernel/rtc.c
@@ -0,0 +1,166 @@
+/*
+ * RTC related functions
+ */
+#include <linux/bcd.h>
+#include <linux/mc146818rtc.h>
+
+#include <asm/time.h>
+
+/*
+ * In order to set the CMOS clock precisely, set_rtc_mmss has to be
+ * called 500 ms after the second nowtime has started, because when
+ * nowtime is written into the registers of the CMOS clock, it will
+ * jump to the next second precisely 500 ms later. Check the Motorola
+ * MC146818A or Dallas DS12887 data sheet for details.
+ *
+ * BUG: This routine does not handle hour overflow properly; it just
+ *      sets the minutes. Usually you'll only notice that after reboot!
+ */
+int mach_set_rtc_mmss(unsigned long nowtime)
+{
+	int retval = 0;
+	int real_seconds, real_minutes, cmos_minutes;
+	unsigned char save_control, save_freq_select;
+
+	save_control = CMOS_READ(RTC_CONTROL); /* tell the clock it's being set */
+	CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
+
+	save_freq_select = CMOS_READ(RTC_FREQ_SELECT); /* stop and reset prescaler */
+	CMOS_WRITE((save_freq_select|RTC_DIV_RESET2), RTC_FREQ_SELECT);
+
+	cmos_minutes = CMOS_READ(RTC_MINUTES);
+	if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD)
+		BCD_TO_BIN(cmos_minutes);
+
+	/*
+	 * since we're only adjusting minutes and seconds,
+	 * don't interfere with hour overflow. This avoids
+	 * messing with unknown time zones but requires your
+	 * RTC not to be off by more than 15 minutes
+	 */
+	real_seconds = nowtime % 60;
+	real_minutes = nowtime / 60;
+	if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
+		real_minutes += 30;		/* correct for half hour time zone */
+	real_minutes %= 60;
+
+	if (abs(real_minutes - cmos_minutes) < 30) {
+		if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
+			BIN_TO_BCD(real_seconds);
+			BIN_TO_BCD(real_minutes);
+		}
+		CMOS_WRITE(real_seconds,RTC_SECONDS);
+		CMOS_WRITE(real_minutes,RTC_MINUTES);
+	} else {
+		printk(KERN_WARNING
+		       "set_rtc_mmss: can't update from %d to %d\n",
+		       cmos_minutes, real_minutes);
+		retval = -1;
+	}
+
+	/* The following flags have to be released exactly in this order,
+	 * otherwise the DS12887 (popular MC146818A clone with integrated
+	 * battery and quartz) will not reset the oscillator and will not
+	 * update precisely 500 ms later. You won't find this mentioned in
+	 * the Dallas Semiconductor data sheets, but who believes data
+	 * sheets anyway ...                           -- Markus Kuhn
+	 */
+	CMOS_WRITE(save_control, RTC_CONTROL);
+	CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
+
+	return retval;
+}
+
+unsigned long mach_get_cmos_time(void)
+{
+	unsigned int year, mon, day, hour, min, sec;
+
+	do {
+		sec = CMOS_READ(RTC_SECONDS);
+		min = CMOS_READ(RTC_MINUTES);
+		hour = CMOS_READ(RTC_HOURS);
+		day = CMOS_READ(RTC_DAY_OF_MONTH);
+		mon = CMOS_READ(RTC_MONTH);
+		year = CMOS_READ(RTC_YEAR);
+	} while (sec != CMOS_READ(RTC_SECONDS));
+
+	if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
+		BCD_TO_BIN(sec);
+		BCD_TO_BIN(min);
+		BCD_TO_BIN(hour);
+		BCD_TO_BIN(day);
+		BCD_TO_BIN(mon);
+		BCD_TO_BIN(year);
+	}
+
+	year += 1900;
+	if (year < 1970)
+		year += 100;
+
+	return mktime(year, mon, day, hour, min, sec);
+}
+
+DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL(rtc_lock);
+
+/*
+ * This is a special lock that is owned by the CPU and holds the index
+ * register we are working with.  It is required for NMI access to the
+ * CMOS/RTC registers.  See include/asm-i386/mc146818rtc.h for details.
+ */
+volatile unsigned long cmos_lock = 0;
+EXPORT_SYMBOL(cmos_lock);
+
+/* Routines for accessing the CMOS RAM/RTC. */
+unsigned char rtc_cmos_read(unsigned char addr)
+{
+	unsigned char val;
+
+	lock_cmos_prefix(addr);
+	outb_p(addr, RTC_PORT(0));
+	val = inb_p(RTC_PORT(1));
+	lock_cmos_suffix(addr);
+	return val;
+}
+EXPORT_SYMBOL(rtc_cmos_read);
+
+void rtc_cmos_write(unsigned char val, unsigned char addr)
+{
+	lock_cmos_prefix(addr);
+	outb_p(addr, RTC_PORT(0));
+	outb_p(val, RTC_PORT(1));
+	lock_cmos_suffix(addr);
+}
+EXPORT_SYMBOL(rtc_cmos_write);
+
+static int set_rtc_mmss(unsigned long nowtime)
+{
+	int retval;
+	unsigned long flags;
+
+	/* gets recalled with irq locally disabled */
+	/* XXX - does irqsave resolve this? -johnstul */
+	spin_lock_irqsave(&rtc_lock, flags);
+	retval = set_wallclock(nowtime);
+	spin_unlock_irqrestore(&rtc_lock, flags);
+
+	return retval;
+}
+
+/* not static: needed by APM */
+unsigned long read_persistent_clock(void)
+{
+	unsigned long retval;
+	unsigned long flags;
+
+	spin_lock_irqsave(&rtc_lock, flags);
+	retval = get_wallclock();
+	spin_unlock_irqrestore(&rtc_lock, flags);
+
+	return retval;
+}
+
+int update_persistent_clock(struct timespec now)
+{
+	return set_rtc_mmss(now.tv_sec);
+}
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index 8a322c96bc23..e9ead762abe8 100644
--- a/arch/x86/kernel/time_32.c
+++ b/arch/x86/kernel/time_32.c
@@ -28,98 +28,21 @@
  *	serialize accesses to xtime/lost_ticks).
  */
 
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/param.h>
-#include <linux/string.h>
-#include <linux/mm.h>
+#include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/time.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/smp.h>
-#include <linux/module.h>
-#include <linux/sysdev.h>
-#include <linux/bcd.h>
-#include <linux/efi.h>
 #include <linux/mca.h>
 
-#include <asm/io.h>
-#include <asm/smp.h>
-#include <asm/irq.h>
-#include <asm/msr.h>
-#include <asm/delay.h>
-#include <asm/mpspec.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
-#include <asm/timer.h>
-#include <asm/time.h>
-
-#include "mach_time.h"
-
-#include <linux/timex.h>
-
-#include <asm/hpet.h>
-
 #include <asm/arch_hooks.h>
+#include <asm/hpet.h>
+#include <asm/time.h>
 
 #include "io_ports.h"
-
-#include <asm/i8259.h>
-
 #include "do_timer.h"
 
 unsigned int cpu_khz;	/* Detected as we calibrate the TSC */
 EXPORT_SYMBOL(cpu_khz);
 
-DEFINE_SPINLOCK(rtc_lock);
-EXPORT_SYMBOL(rtc_lock);
-
-/*
- * This is a special lock that is owned by the CPU and holds the index
- * register we are working with.  It is required for NMI access to the
- * CMOS/RTC registers.  See include/asm-i386/mc146818rtc.h for details.
- */
-volatile unsigned long cmos_lock = 0;
-EXPORT_SYMBOL(cmos_lock);
-
-/* Routines for accessing the CMOS RAM/RTC. */
-unsigned char rtc_cmos_read(unsigned char addr)
-{
-	unsigned char val;
-	lock_cmos_prefix(addr);
-	outb_p(addr, RTC_PORT(0));
-	val = inb_p(RTC_PORT(1));
-	lock_cmos_suffix(addr);
-	return val;
-}
-EXPORT_SYMBOL(rtc_cmos_read);
-
-void rtc_cmos_write(unsigned char val, unsigned char addr)
-{
-	lock_cmos_prefix(addr);
-	outb_p(addr, RTC_PORT(0));
-	outb_p(val, RTC_PORT(1));
-	lock_cmos_suffix(addr);
-}
-EXPORT_SYMBOL(rtc_cmos_write);
-
-static int set_rtc_mmss(unsigned long nowtime)
-{
-	int retval;
-	unsigned long flags;
-
-	/* gets recalled with irq locally disabled */
-	/* XXX - does irqsave resolve this? -johnstul */
-	spin_lock_irqsave(&rtc_lock, flags);
-	retval = set_wallclock(nowtime);
-	spin_unlock_irqrestore(&rtc_lock, flags);
-
-	return retval;
-}
-
-
 int timer_ack;
 
 unsigned long profile_pc(struct pt_regs *regs)
@@ -137,7 +60,7 @@ unsigned long profile_pc(struct pt_regs *regs)
 		/* Return address is either directly at stack pointer
 		   or above a saved eflags. Eflags has bits 22-31 zero,
 		   kernel addresses don't. */
- 		if (sp[0] >> 22)
+		if (sp[0] >> 22)
 			return sp[0];
 		if (sp[1] >> 22)
 			return sp[1];
@@ -193,26 +116,6 @@ irqreturn_t timer_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-/* not static: needed by APM */
-unsigned long read_persistent_clock(void)
-{
-	unsigned long retval;
-	unsigned long flags;
-
-	spin_lock_irqsave(&rtc_lock, flags);
-
-	retval = get_wallclock();
-
-	spin_unlock_irqrestore(&rtc_lock, flags);
-
-	return retval;
-}
-
-int update_persistent_clock(struct timespec now)
-{
-	return set_rtc_mmss(now.tv_sec);
-}
-
 extern void (*late_time_init)(void);
 /* Duplicate of time_init() below, with hpet_enable part added */
 void __init hpet_time_init(void)
diff --git a/include/asm-x86/mach-default/mach_time.h b/include/asm-x86/mach-default/mach_time.h
deleted file mode 100644
index 31eb5de6f3dc..000000000000
--- a/include/asm-x86/mach-default/mach_time.h
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- *  include/asm-i386/mach-default/mach_time.h
- *
- *  Machine specific set RTC function for generic.
- *  Split out from time.c by Osamu Tomita <tomita@cinet.co.jp>
- */
-#ifndef _MACH_TIME_H
-#define _MACH_TIME_H
-
-#include <linux/mc146818rtc.h>
-
-/* for check timing call set_rtc_mmss() 500ms     */
-/* used in arch/i386/time.c::do_timer_interrupt() */
-#define USEC_AFTER	500000
-#define USEC_BEFORE	500000
-
-/*
- * In order to set the CMOS clock precisely, set_rtc_mmss has to be
- * called 500 ms after the second nowtime has started, because when
- * nowtime is written into the registers of the CMOS clock, it will
- * jump to the next second precisely 500 ms later. Check the Motorola
- * MC146818A or Dallas DS12887 data sheet for details.
- *
- * BUG: This routine does not handle hour overflow properly; it just
- *      sets the minutes. Usually you'll only notice that after reboot!
- */
-static inline int mach_set_rtc_mmss(unsigned long nowtime)
-{
-	int retval = 0;
-	int real_seconds, real_minutes, cmos_minutes;
-	unsigned char save_control, save_freq_select;
-
-	save_control = CMOS_READ(RTC_CONTROL); /* tell the clock it's being set */
-	CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
-
-	save_freq_select = CMOS_READ(RTC_FREQ_SELECT); /* stop and reset prescaler */
-	CMOS_WRITE((save_freq_select|RTC_DIV_RESET2), RTC_FREQ_SELECT);
-
-	cmos_minutes = CMOS_READ(RTC_MINUTES);
-	if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD)
-		BCD_TO_BIN(cmos_minutes);
-
-	/*
-	 * since we're only adjusting minutes and seconds,
-	 * don't interfere with hour overflow. This avoids
-	 * messing with unknown time zones but requires your
-	 * RTC not to be off by more than 15 minutes
-	 */
-	real_seconds = nowtime % 60;
-	real_minutes = nowtime / 60;
-	if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
-		real_minutes += 30;		/* correct for half hour time zone */
-	real_minutes %= 60;
-
-	if (abs(real_minutes - cmos_minutes) < 30) {
-		if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-			BIN_TO_BCD(real_seconds);
-			BIN_TO_BCD(real_minutes);
-		}
-		CMOS_WRITE(real_seconds,RTC_SECONDS);
-		CMOS_WRITE(real_minutes,RTC_MINUTES);
-	} else {
-		printk(KERN_WARNING
-		       "set_rtc_mmss: can't update from %d to %d\n",
-		       cmos_minutes, real_minutes);
-		retval = -1;
-	}
-
-	/* The following flags have to be released exactly in this order,
-	 * otherwise the DS12887 (popular MC146818A clone with integrated
-	 * battery and quartz) will not reset the oscillator and will not
-	 * update precisely 500 ms later. You won't find this mentioned in
-	 * the Dallas Semiconductor data sheets, but who believes data
-	 * sheets anyway ...                           -- Markus Kuhn
-	 */
-	CMOS_WRITE(save_control, RTC_CONTROL);
-	CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
-
-	return retval;
-}
-
-static inline unsigned long mach_get_cmos_time(void)
-{
-	unsigned int year, mon, day, hour, min, sec;
-
-	do {
-		sec = CMOS_READ(RTC_SECONDS);
-		min = CMOS_READ(RTC_MINUTES);
-		hour = CMOS_READ(RTC_HOURS);
-		day = CMOS_READ(RTC_DAY_OF_MONTH);
-		mon = CMOS_READ(RTC_MONTH);
-		year = CMOS_READ(RTC_YEAR);
-	} while (sec != CMOS_READ(RTC_SECONDS));
-
-	if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-		BCD_TO_BIN(sec);
-		BCD_TO_BIN(min);
-		BCD_TO_BIN(hour);
-		BCD_TO_BIN(day);
-		BCD_TO_BIN(mon);
-		BCD_TO_BIN(year);
-	}
-
-	year += 1900;
-	if (year < 1970)
-		year += 100;
-
-	return mktime(year, mon, day, hour, min, sec);
-}
-
-#endif /* !_MACH_TIME_H */
diff --git a/include/asm-x86/mc146818rtc.h b/include/asm-x86/mc146818rtc.h
index 9d39436df231..cdd9f965835a 100644
--- a/include/asm-x86/mc146818rtc.h
+++ b/include/asm-x86/mc146818rtc.h
@@ -92,6 +92,9 @@ static inline unsigned char current_lock_cmos_reg(void)
 unsigned char rtc_cmos_read(unsigned char addr);
 void rtc_cmos_write(unsigned char val, unsigned char addr);
 
+extern int mach_set_rtc_mmss(unsigned long nowtime);
+extern unsigned long mach_get_cmos_time(void);
+
 #define RTC_IRQ 8
 
 #endif /* _ASM_MC146818RTC_H */
diff --git a/include/asm-x86/time.h b/include/asm-x86/time.h
index eac011366dc2..b3f94cd81ac6 100644
--- a/include/asm-x86/time.h
+++ b/include/asm-x86/time.h
@@ -2,7 +2,7 @@
 #define _ASMi386_TIME_H
 
 #include <linux/efi.h>
-#include "mach_time.h"
+#include <asm/mc146818rtc.h>
 
 static inline unsigned long native_get_wallclock(void)
 {
-- 
cgit v1.2.3


From 02456c708eab4515121e5e581422fa3be14368d1 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:27 +0100
Subject: x86: nuke a ton of dead hpet code

No users, just ballast

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 drivers/char/hpet.c  | 75 ----------------------------------------------------
 include/linux/hpet.h |  3 ---
 2 files changed, 78 deletions(-)

(limited to 'include')

diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 22f5fd02ea87..465ad35ed38f 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -600,63 +600,6 @@ static int hpet_is_known(struct hpet_data *hdp)
 	return 0;
 }
 
-EXPORT_SYMBOL(hpet_alloc);
-EXPORT_SYMBOL(hpet_register);
-EXPORT_SYMBOL(hpet_unregister);
-EXPORT_SYMBOL(hpet_control);
-
-int hpet_register(struct hpet_task *tp, int periodic)
-{
-	unsigned int i;
-	u64 mask;
-	struct hpet_timer __iomem *timer;
-	struct hpet_dev *devp;
-	struct hpets *hpetp;
-
-	switch (periodic) {
-	case 1:
-		mask = Tn_PER_INT_CAP_MASK;
-		break;
-	case 0:
-		mask = 0;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	tp->ht_opaque = NULL;
-
-	spin_lock_irq(&hpet_task_lock);
-	spin_lock(&hpet_lock);
-
-	for (devp = NULL, hpetp = hpets; hpetp && !devp; hpetp = hpetp->hp_next)
-		for (timer = hpetp->hp_hpet->hpet_timers, i = 0;
-		     i < hpetp->hp_ntimer; i++, timer++) {
-			if ((readq(&timer->hpet_config) & Tn_PER_INT_CAP_MASK)
-			    != mask)
-				continue;
-
-			devp = &hpetp->hp_dev[i];
-
-			if (devp->hd_flags & HPET_OPEN || devp->hd_task) {
-				devp = NULL;
-				continue;
-			}
-
-			tp->ht_opaque = devp;
-			devp->hd_task = tp;
-			break;
-		}
-
-	spin_unlock(&hpet_lock);
-	spin_unlock_irq(&hpet_task_lock);
-
-	if (tp->ht_opaque)
-		return 0;
-	else
-		return -EBUSY;
-}
-
 static inline int hpet_tpcheck(struct hpet_task *tp)
 {
 	struct hpet_dev *devp;
@@ -706,24 +649,6 @@ int hpet_unregister(struct hpet_task *tp)
 	return 0;
 }
 
-int hpet_control(struct hpet_task *tp, unsigned int cmd, unsigned long arg)
-{
-	struct hpet_dev *devp;
-	int err;
-
-	if ((err = hpet_tpcheck(tp)))
-		return err;
-
-	spin_lock_irq(&hpet_lock);
-	devp = tp->ht_opaque;
-	if (devp->hd_task != tp) {
-		spin_unlock_irq(&hpet_lock);
-		return -ENXIO;
-	}
-	spin_unlock_irq(&hpet_lock);
-	return hpet_ioctl_common(devp, cmd, arg, 1);
-}
-
 static ctl_table hpet_table[] = {
 	{
 	 .ctl_name = CTL_UNNUMBERED,
diff --git a/include/linux/hpet.h b/include/linux/hpet.h
index e3c0b2aa944c..9cd94bfd07e5 100644
--- a/include/linux/hpet.h
+++ b/include/linux/hpet.h
@@ -115,9 +115,6 @@ static inline void hpet_reserve_timer(struct hpet_data *hd, int timer)
 }
 
 int hpet_alloc(struct hpet_data *);
-int hpet_register(struct hpet_task *, int);
-int hpet_unregister(struct hpet_task *);
-int hpet_control(struct hpet_task *, unsigned int, unsigned long);
 
 #endif /* __KERNEL__ */
 
-- 
cgit v1.2.3


From a72368dd37f6ae333fbab03598e46a995d91decc Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:28 +0100
Subject: x86: remove dead code and exports

No users.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/process_64.c | 7 -------
 include/asm-x86/idle.h       | 1 -
 2 files changed, 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 2c9e59448f4c..98d85952f574 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -72,13 +72,6 @@ void idle_notifier_register(struct notifier_block *n)
 {
 	atomic_notifier_chain_register(&idle_notifier, n);
 }
-EXPORT_SYMBOL_GPL(idle_notifier_register);
-
-void idle_notifier_unregister(struct notifier_block *n)
-{
-	atomic_notifier_chain_unregister(&idle_notifier, n);
-}
-EXPORT_SYMBOL(idle_notifier_unregister);
 
 void enter_idle(void)
 {
diff --git a/include/asm-x86/idle.h b/include/asm-x86/idle.h
index 6bd47dcf2067..d240e5b30a45 100644
--- a/include/asm-x86/idle.h
+++ b/include/asm-x86/idle.h
@@ -6,7 +6,6 @@
 
 struct notifier_block;
 void idle_notifier_register(struct notifier_block *n);
-void idle_notifier_unregister(struct notifier_block *n);
 
 void enter_idle(void);
 void exit_idle(void);
-- 
cgit v1.2.3


From 3abf024d2abb79614d8c4cb25a70d5596f77d0ad Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:28 +0100
Subject: x86: nuke a ton of unused exports

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perfctr-watchdog.c | 1 -
 arch/x86/kernel/e820_64.c              | 1 -
 arch/x86/kernel/init_task.c            | 1 -
 arch/x86/kernel/pci-swiotlb_64.c       | 1 -
 arch/x86/kernel/smpboot_32.c           | 1 -
 arch/x86/kernel/smpboot_64.c           | 2 --
 arch/x86/kernel/stacktrace.c           | 1 -
 arch/x86/kernel/traps_64.c             | 8 --------
 arch/x86/mach-voyager/voyager_smp.c    | 1 -
 arch/x86/mm/numa_64.c                  | 1 -
 include/asm-x86/bug.h                  | 3 ---
 include/asm-x86/mmu_context_64.h       | 2 +-
 12 files changed, 1 insertion(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index c02541e6e653..9b838324b818 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -167,7 +167,6 @@ void release_evntsel_nmi(unsigned int msr)
 	clear_bit(counter, evntsel_nmi_owner);
 }
 
-EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
 EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
 EXPORT_SYMBOL(reserve_perfctr_nmi);
 EXPORT_SYMBOL(release_perfctr_nmi);
diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index e510cfd5bb71..0128b0b0e5b5 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -34,7 +34,6 @@ struct e820map e820;
  * PFN of last memory page.
  */
 unsigned long end_pfn;
-EXPORT_SYMBOL(end_pfn);
 
 /*
  * end_pfn only includes RAM, while end_pfn_map includes all e820 entries.
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index 468c9c437842..5b3ce7934363 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -15,7 +15,6 @@ static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
-EXPORT_SYMBOL(init_mm);
 
 /*
  * Initial thread structure.
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 102866d729a5..82a0a674a003 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -10,7 +10,6 @@
 #include <asm/dma.h>
 
 int swiotlb __read_mostly;
-EXPORT_SYMBOL(swiotlb);
 
 const struct dma_mapping_ops swiotlb_dma_ops = {
 	.mapping_error = swiotlb_dma_mapping_error,
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 4ea80cbe52e5..753d7acf4dac 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -83,7 +83,6 @@ EXPORT_SYMBOL(cpu_online_map);
 
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
-EXPORT_SYMBOL(cpu_callout_map);
 cpumask_t cpu_possible_map;
 EXPORT_SYMBOL(cpu_possible_map);
 static cpumask_t smp_commenced_mask;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index b36d32ff0b39..7552db9ee9ff 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -78,8 +78,6 @@ EXPORT_SYMBOL(cpu_online_map);
  */
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
-EXPORT_SYMBOL(cpu_callout_map);
-
 cpumask_t cpu_possible_map;
 EXPORT_SYMBOL(cpu_possible_map);
 
diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index 55771fd7e545..c571edd11878 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -69,7 +69,6 @@ void save_stack_trace(struct stack_trace *trace)
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
-EXPORT_SYMBOL(save_stack_trace);
 
 void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 {
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index cc68b92316cd..aa248d754533 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -461,14 +461,6 @@ int is_valid_bugaddr(unsigned long rip)
 	return ud2 == 0x0b0f;
 }
 
-#ifdef CONFIG_BUG
-void out_of_line_bug(void)
-{ 
-	BUG(); 
-} 
-EXPORT_SYMBOL(out_of_line_bug);
-#endif
-
 static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
 static int die_owner = -1;
 static unsigned int die_nest_count;
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 73c435ce10fd..981def2b4e9b 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -227,7 +227,6 @@ static cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 /* This is for the new dynamic CPU boot code */
 cpumask_t cpu_callin_map = CPU_MASK_NONE;
 cpumask_t cpu_callout_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_callout_map);
 cpumask_t cpu_possible_map = CPU_MASK_NONE;
 EXPORT_SYMBOL(cpu_possible_map);
 
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 2d8fdc05f415..e7f3f4e9ec85 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -624,7 +624,6 @@ void __init init_cpu_to_node(void)
 
 EXPORT_SYMBOL(cpu_to_node);
 EXPORT_SYMBOL(node_to_cpumask);
-EXPORT_SYMBOL(memnode);
 EXPORT_SYMBOL(node_data);
 
 #ifdef CONFIG_DISCONTIGMEM
diff --git a/include/asm-x86/bug.h b/include/asm-x86/bug.h
index fd8bdc639c48..8d477a201392 100644
--- a/include/asm-x86/bug.h
+++ b/include/asm-x86/bug.h
@@ -33,9 +33,6 @@
 	} while(0)
 #endif
 
-void out_of_line_bug(void);
-#else /* CONFIG_BUG */
-static inline void out_of_line_bug(void) { }
 #endif /* !CONFIG_BUG */
 
 #include <asm-generic/bug.h>
diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h
index 0cce83a78378..29f95c3acc28 100644
--- a/include/asm-x86/mmu_context_64.h
+++ b/include/asm-x86/mmu_context_64.h
@@ -49,7 +49,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	else {
 		write_pda(mmu_state, TLBSTATE_OK);
 		if (read_pda(active_mm) != next)
-			out_of_line_bug();
+			BUG();
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
 			/* We were in lazy tlb mode and leave_mm disabled 
 			 * tlb flush IPI delivery. We must reload CR3
-- 
cgit v1.2.3


From 99122a3fe74c4918cdc86a8d6d63b40240fa8881 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Wed, 30 Jan 2008 13:30:28 +0100
Subject: x86: remove more bogus filenames in comments.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/acpi_32.h                 | 4 +---
 include/asm-x86/acpi_64.h                 | 4 +---
 include/asm-x86/checksum_64.h             | 2 +-
 include/asm-x86/i387_32.h                 | 2 --
 include/asm-x86/i387_64.h                 | 2 --
 include/asm-x86/ia32_unistd.h             | 2 +-
 include/asm-x86/ide.h                     | 2 --
 include/asm-x86/irqflags_32.h             | 2 --
 include/asm-x86/irqflags_64.h             | 2 --
 include/asm-x86/mach-default/apm.h        | 2 --
 include/asm-x86/mach-default/mach_timer.h | 2 --
 include/asm-x86/mach-default/mach_traps.h | 2 --
 include/asm-x86/nmi_32.h                  | 3 ---
 include/asm-x86/nmi_64.h                  | 3 ---
 include/asm-x86/processor_32.h            | 2 --
 include/asm-x86/processor_64.h            | 2 --
 include/asm-x86/rio.h                     | 4 ++--
 include/asm-x86/rwsem.h                   | 2 +-
 include/asm-x86/topology_32.h             | 2 --
 include/asm-x86/xor_32.h                  | 2 --
 include/asm-x86/xor_64.h                  | 2 --
 21 files changed, 7 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/acpi_32.h b/include/asm-x86/acpi_32.h
index 3cce5a0f2d9c..c8afeea0b80b 100644
--- a/include/asm-x86/acpi_32.h
+++ b/include/asm-x86/acpi_32.h
@@ -1,9 +1,7 @@
 /*
- *  asm-i386/acpi.h
- *
  *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
  *  Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
-  *
+ *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
  *  This program is free software; you can redistribute it and/or modify
diff --git a/include/asm-x86/acpi_64.h b/include/asm-x86/acpi_64.h
index 98173357dd89..58114ed08024 100644
--- a/include/asm-x86/acpi_64.h
+++ b/include/asm-x86/acpi_64.h
@@ -1,9 +1,7 @@
 /*
- *  asm-x86_64/acpi.h
- *
  *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
  *  Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
-  *
+ *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
  *  This program is free software; you can redistribute it and/or modify
diff --git a/include/asm-x86/checksum_64.h b/include/asm-x86/checksum_64.h
index 419fe88a0342..e5f79997decc 100644
--- a/include/asm-x86/checksum_64.h
+++ b/include/asm-x86/checksum_64.h
@@ -4,7 +4,7 @@
 /* 
  * Checksums for x86-64 
  * Copyright 2002 by Andi Kleen, SuSE Labs 
- * with some code from asm-i386/checksum.h
+ * with some code from asm-x86/checksum.h
  */ 
 
 #include <linux/compiler.h>
diff --git a/include/asm-x86/i387_32.h b/include/asm-x86/i387_32.h
index cdd1e248e3b4..9ac2502cdd3d 100644
--- a/include/asm-x86/i387_32.h
+++ b/include/asm-x86/i387_32.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-i386/i387.h
- *
  * Copyright (C) 1994 Linus Torvalds
  *
  * Pentium III FXSR, SSE support
diff --git a/include/asm-x86/i387_64.h b/include/asm-x86/i387_64.h
index 3a4ffba3d6bc..12edb9f2d755 100644
--- a/include/asm-x86/i387_64.h
+++ b/include/asm-x86/i387_64.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-x86_64/i387.h
- *
  * Copyright (C) 1994 Linus Torvalds
  *
  * Pentium III FXSR, SSE support
diff --git a/include/asm-x86/ia32_unistd.h b/include/asm-x86/ia32_unistd.h
index 5b52ce507338..61cea9e7c5c1 100644
--- a/include/asm-x86/ia32_unistd.h
+++ b/include/asm-x86/ia32_unistd.h
@@ -5,7 +5,7 @@
  * This file contains the system call numbers of the ia32 port,
  * this is for the kernel only.
  * Only add syscalls here where some part of the kernel needs to know
- * the number. This should be otherwise in sync with asm-i386/unistd.h. -AK
+ * the number. This should be otherwise in sync with asm-x86/unistd_32.h. -AK
  */
 
 #define __NR_ia32_restart_syscall 0
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index 42130adf9c7c..c2552d8bebf7 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -1,6 +1,4 @@
 /*
- *  linux/include/asm-i386/ide.h
- *
  *  Copyright (C) 1994-1996  Linus Torvalds & authors
  */
 
diff --git a/include/asm-x86/irqflags_32.h b/include/asm-x86/irqflags_32.h
index 4c7720089cb5..98b21b9bdce8 100644
--- a/include/asm-x86/irqflags_32.h
+++ b/include/asm-x86/irqflags_32.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-i386/irqflags.h
- *
  * IRQ flags handling
  *
  * This file gets included from lowlevel asm headers too, to provide
diff --git a/include/asm-x86/irqflags_64.h b/include/asm-x86/irqflags_64.h
index bb9163bb29d1..38c07db733cf 100644
--- a/include/asm-x86/irqflags_64.h
+++ b/include/asm-x86/irqflags_64.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-x86_64/irqflags.h
- *
  * IRQ flags handling
  *
  * This file gets included from lowlevel asm headers too, to provide
diff --git a/include/asm-x86/mach-default/apm.h b/include/asm-x86/mach-default/apm.h
index 1f730b8bd1fd..989f34c37d32 100644
--- a/include/asm-x86/mach-default/apm.h
+++ b/include/asm-x86/mach-default/apm.h
@@ -1,6 +1,4 @@
 /*
- *  include/asm-i386/mach-default/apm.h
- *
  *  Machine specific APM BIOS functions for generic.
  *  Split out from apm.c by Osamu Tomita <tomita@cinet.co.jp>
  */
diff --git a/include/asm-x86/mach-default/mach_timer.h b/include/asm-x86/mach-default/mach_timer.h
index 807992fd4171..4b76e536cd98 100644
--- a/include/asm-x86/mach-default/mach_timer.h
+++ b/include/asm-x86/mach-default/mach_timer.h
@@ -1,6 +1,4 @@
 /*
- *  include/asm-i386/mach-default/mach_timer.h
- *
  *  Machine specific calibrate_tsc() for generic.
  *  Split out from timer_tsc.c by Osamu Tomita <tomita@cinet.co.jp>
  */
diff --git a/include/asm-x86/mach-default/mach_traps.h b/include/asm-x86/mach-default/mach_traps.h
index 625438b8a6eb..2fe7705c0484 100644
--- a/include/asm-x86/mach-default/mach_traps.h
+++ b/include/asm-x86/mach-default/mach_traps.h
@@ -1,6 +1,4 @@
 /*
- *  include/asm-i386/mach-default/mach_traps.h
- *
  *  Machine specific NMI handling for generic.
  *  Split out from traps.c by Osamu Tomita <tomita@cinet.co.jp>
  */
diff --git a/include/asm-x86/nmi_32.h b/include/asm-x86/nmi_32.h
index 70a958a8e381..7206c7e8a388 100644
--- a/include/asm-x86/nmi_32.h
+++ b/include/asm-x86/nmi_32.h
@@ -1,6 +1,3 @@
-/*
- *  linux/include/asm-i386/nmi.h
- */
 #ifndef ASM_NMI_H
 #define ASM_NMI_H
 
diff --git a/include/asm-x86/nmi_64.h b/include/asm-x86/nmi_64.h
index 65b6acf3bb59..92f36fb1dc62 100644
--- a/include/asm-x86/nmi_64.h
+++ b/include/asm-x86/nmi_64.h
@@ -1,6 +1,3 @@
-/*
- *  linux/include/asm-i386/nmi.h
- */
 #ifndef ASM_NMI_H
 #define ASM_NMI_H
 
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 58880a1ddd65..3b51a18f16d5 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-i386/processor.h
- *
  * Copyright (C) 1994 Linus Torvalds
  */
 
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 6b96e1f12a42..6c0d96a7af63 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-x86_64/processor.h
- *
  * Copyright (C) 1994 Linus Torvalds
  */
 
diff --git a/include/asm-x86/rio.h b/include/asm-x86/rio.h
index c7350f6d2015..97cdcc9887ba 100644
--- a/include/asm-x86/rio.h
+++ b/include/asm-x86/rio.h
@@ -1,6 +1,6 @@
 /*
- * Derived from include/asm-i386/mach-summit/mach_mpparse.h
- *          and include/asm-i386/mach-default/bios_ebda.h
+ * Derived from include/asm-x86/mach-summit/mach_mpparse.h
+ *          and include/asm-x86/mach-default/bios_ebda.h
  *
  * Author: Laurent Vivier <Laurent.Vivier@bull.net>
  */
diff --git a/include/asm-x86/rwsem.h b/include/asm-x86/rwsem.h
index 041906f3c6df..a7e7e14cb430 100644
--- a/include/asm-x86/rwsem.h
+++ b/include/asm-x86/rwsem.h
@@ -2,7 +2,7 @@
  *
  * Written by David Howells (dhowells@redhat.com).
  *
- * Derived from asm-i386/semaphore.h
+ * Derived from asm-x86/semaphore.h
  *
  *
  * The MSW of the count is the negated number of active writers and waiting
diff --git a/include/asm-x86/topology_32.h b/include/asm-x86/topology_32.h
index 9040f5a61278..fdca80a851e2 100644
--- a/include/asm-x86/topology_32.h
+++ b/include/asm-x86/topology_32.h
@@ -1,6 +1,4 @@
 /*
- * linux/include/asm-i386/topology.h
- *
  * Written by: Matthew Dobson, IBM Corporation
  *
  * Copyright (C) 2002, IBM Corp.
diff --git a/include/asm-x86/xor_32.h b/include/asm-x86/xor_32.h
index 23c86cef3b25..a41ef1bdd424 100644
--- a/include/asm-x86/xor_32.h
+++ b/include/asm-x86/xor_32.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-i386/xor.h
- *
  * Optimized RAID-5 checksumming functions for MMX and SSE.
  *
  * This program is free software; you can redistribute it and/or modify
diff --git a/include/asm-x86/xor_64.h b/include/asm-x86/xor_64.h
index f942fcc21831..1eee7fcb2420 100644
--- a/include/asm-x86/xor_64.h
+++ b/include/asm-x86/xor_64.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-x86_64/xor.h
- *
  * Optimized RAID-5 checksumming functions for MMX and SSE.
  *
  * This program is free software; you can redistribute it and/or modify
-- 
cgit v1.2.3


From f20ebee41882d28c965166e56c1331fbd28778bb Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:29 +0100
Subject: x86: move 8259 defines to i8259.h

Move the i8259 defines and remove the now io_ports.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c               |  2 --
 arch/x86/kernel/apm_32.c                |  2 --
 arch/x86/kernel/i8259_32.c              |  2 --
 arch/x86/kernel/io_apic_32.c            |  2 --
 arch/x86/kernel/time_32.c               |  1 -
 arch/x86/kernel/vmiclock_32.c           |  1 -
 include/asm-x86/i8259.h                 | 17 ++++++++++++++++-
 include/asm-x86/mach-default/io_ports.h | 25 -------------------------
 8 files changed, 16 insertions(+), 36 deletions(-)
 delete mode 100644 include/asm-x86/mach-default/io_ports.h

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 420c15842e44..56352c11a896 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -43,8 +43,6 @@
 #include <mach_apicdef.h>
 #include <mach_ipi.h>
 
-#include "io_ports.h"
-
 /*
  * Sanity check
  */
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index af045ca0f653..7496c2e4b6ae 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -235,8 +235,6 @@
 #include <asm/paravirt.h>
 #include <asm/reboot.h>
 
-#include "io_ports.h"
-
 #if defined(CONFIG_APM_DISPLAY_BLANK) && defined(CONFIG_VT)
 extern int (*console_blank_hook)(int);
 #endif
diff --git a/arch/x86/kernel/i8259_32.c b/arch/x86/kernel/i8259_32.c
index 5f3496d01984..3321ce669295 100644
--- a/arch/x86/kernel/i8259_32.c
+++ b/arch/x86/kernel/i8259_32.c
@@ -21,8 +21,6 @@
 #include <asm/arch_hooks.h>
 #include <asm/i8259.h>
 
-#include <io_ports.h>
-
 /*
  * This is the 'legacy' 8259A Programmable Interrupt Controller,
  * present in the majority of PC/AT boxes.
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index ab77f1905469..75bf8dc77650 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -48,8 +48,6 @@
 #include <mach_apic.h>
 #include <mach_apicdef.h>
 
-#include "io_ports.h"
-
 int (*ioapic_renumber_irq)(int ioapic, int irq);
 atomic_t irq_mis_count;
 
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index e9ead762abe8..2dcbb81b4cd3 100644
--- a/arch/x86/kernel/time_32.c
+++ b/arch/x86/kernel/time_32.c
@@ -37,7 +37,6 @@
 #include <asm/hpet.h>
 #include <asm/time.h>
 
-#include "io_ports.h"
 #include "do_timer.h"
 
 unsigned int cpu_khz;	/* Detected as we calibrate the TSC */
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index b1b5ab08b26e..57f9ef5a324c 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -35,7 +35,6 @@
 #include <asm/i8253.h>
 
 #include <irq_vectors.h>
-#include "io_ports.h"
 
 #define VMI_ONESHOT  (VMI_ALARM_IS_ONESHOT  | VMI_CYCLES_REAL | vmi_get_alarm_wiring())
 #define VMI_PERIODIC (VMI_ALARM_IS_PERIODIC | VMI_CYCLES_REAL | vmi_get_alarm_wiring())
diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h
index 29d8f9a6b3fc..cabcc6cf3923 100644
--- a/include/asm-x86/i8259.h
+++ b/include/asm-x86/i8259.h
@@ -3,10 +3,25 @@
 
 extern unsigned int cached_irq_mask;
 
-#define __byte(x,y) 		(((unsigned char *) &(y))[x])
+#define __byte(x,y)		(((unsigned char *) &(y))[x])
 #define cached_master_mask	(__byte(0, cached_irq_mask))
 #define cached_slave_mask	(__byte(1, cached_irq_mask))
 
+/* i8259A PIC registers */
+#define PIC_MASTER_CMD		0x20
+#define PIC_MASTER_IMR		0x21
+#define PIC_MASTER_ISR		PIC_MASTER_CMD
+#define PIC_MASTER_POLL		PIC_MASTER_ISR
+#define PIC_MASTER_OCW3		PIC_MASTER_ISR
+#define PIC_SLAVE_CMD		0xa0
+#define PIC_SLAVE_IMR		0xa1
+
+/* i8259A PIC related value */
+#define PIC_CASCADE_IR		2
+#define MASTER_ICW4_DEFAULT	0x01
+#define SLAVE_ICW4_DEFAULT	0x01
+#define PIC_ICW4_AEOI		2
+
 extern spinlock_t i8259A_lock;
 
 extern void init_8259A(int auto_eoi);
diff --git a/include/asm-x86/mach-default/io_ports.h b/include/asm-x86/mach-default/io_ports.h
deleted file mode 100644
index 48540ba97166..000000000000
--- a/include/asm-x86/mach-default/io_ports.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- *  arch/i386/mach-generic/io_ports.h
- *
- *  Machine specific IO port address definition for generic.
- *  Written by Osamu Tomita <tomita@cinet.co.jp>
- */
-#ifndef _MACH_IO_PORTS_H
-#define _MACH_IO_PORTS_H
-
-/* i8259A PIC registers */
-#define PIC_MASTER_CMD		0x20
-#define PIC_MASTER_IMR		0x21
-#define PIC_MASTER_ISR		PIC_MASTER_CMD
-#define PIC_MASTER_POLL		PIC_MASTER_ISR
-#define PIC_MASTER_OCW3		PIC_MASTER_ISR
-#define PIC_SLAVE_CMD		0xa0
-#define PIC_SLAVE_IMR		0xa1
-
-/* i8259A PIC related value */
-#define PIC_CASCADE_IR		2
-#define MASTER_ICW4_DEFAULT	0x01
-#define SLAVE_ICW4_DEFAULT	0x01
-#define PIC_ICW4_AEOI		2
-
-#endif /* !_MACH_IO_PORTS_H */
-- 
cgit v1.2.3


From 1d8a1f6b51f6b195dfdcf05821be97edede5664a Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:30 +0100
Subject: x86: prepare merger of <asm/alternative_{32,64}.h>

Prepare for merging <asm/alternative_{32,64}.h> by making the 32- and
64-bit versions textually identical.  This involves:

- removing arbitrary header inclusion differences
- reorganizing the 32-bit version slightly to match the 64-bit version
- using <asm/asm.h> to unify the assembly code
- renaming struct paravirt_patch to struct paravirt_patch_site in the
  64-bit version to match the 32-bit version; there are no references
  to struct paravirt_patch elsewhere in the tree.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/alternative_32.h | 131 +++++++++++++++++++++------------------
 include/asm-x86/alternative_64.h |  84 +++++++++++++------------
 2 files changed, 112 insertions(+), 103 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/alternative_32.h b/include/asm-x86/alternative_32.h
index bda6c810c0f4..4919b5ee8b9f 100644
--- a/include/asm-x86/alternative_32.h
+++ b/include/asm-x86/alternative_32.h
@@ -1,23 +1,63 @@
 #ifndef _I386_ALTERNATIVE_H
 #define _I386_ALTERNATIVE_H
 
-#include <asm/types.h>
-#include <linux/stddef.h>
 #include <linux/types.h>
+#include <linux/stddef.h>
+#include <asm/asm.h>
+
+/*
+ * Alternative inline assembly for SMP.
+ *
+ * The LOCK_PREFIX macro defined here replaces the LOCK and
+ * LOCK_PREFIX macros used everywhere in the source tree.
+ *
+ * SMP alternatives use the same data structures as the other
+ * alternatives and the X86_FEATURE_UP flag to indicate the case of a
+ * UP system running a SMP kernel.  The existing apply_alternatives()
+ * works fine for patching a SMP kernel for UP.
+ *
+ * The SMP alternative tables can be kept after boot and contain both
+ * UP and SMP versions of the instructions to allow switching back to
+ * SMP at runtime, when hotplugging in a new CPU, which is especially
+ * useful in virtualized environments.
+ *
+ * The very common lock prefix is handled as special case in a
+ * separate table which is a pure address list without replacement ptr
+ * and size information.  That keeps the table sizes small.
+ */
+
+#ifdef CONFIG_SMP
+#define LOCK_PREFIX \
+		".section .smp_locks,\"a\"\n"	\
+		_ASM_ALIGN "\n"			\
+		_ASM_PTR "661f\n" /* address */	\
+		".previous\n"			\
+		"661:\n\tlock; "
+
+#else /* ! CONFIG_SMP */
+#define LOCK_PREFIX ""
+#endif
+
+/* This must be included *after* the definition of LOCK_PREFIX */
+#include <asm/cpufeature.h>
 
 struct alt_instr {
-	u8 *instr; 		/* original instruction */
+	u8 *instr;		/* original instruction */
 	u8 *replacement;
 	u8  cpuid;		/* cpuid bit set for replacement */
 	u8  instrlen;		/* length of original instruction */
-	u8  replacementlen; 	/* length of new instruction, <= instrlen */
-	u8  pad;
+	u8  replacementlen;	/* length of new instruction, <= instrlen */
+	u8  pad1;
+#ifdef CONFIG_X86_64
+	u32 pad2;
+#endif
 };
 
 extern void alternative_instructions(void);
 extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
 
 struct module;
+
 #ifdef CONFIG_SMP
 extern void alternatives_smp_module_add(struct module *mod, char *name,
 					void *locks, void *locks_end,
@@ -45,17 +85,17 @@ static inline void alternatives_smp_switch(int smp) {}
  * without volatile and memory clobber.
  */
 #define alternative(oldinstr, newinstr, feature)			\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n" 			\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
 		      ".section .altinstructions,\"a\"\n"		\
-		      "  .align 4\n"					\
-		      "  .long 661b\n"            /* label */		\
-		      "  .long 663f\n"		  /* new instruction */	\
-		      "  .byte %c0\n"             /* feature bit */	\
-		      "  .byte 662b-661b\n"       /* sourcelen */	\
-		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */\
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
 		      ".previous" :: "i" (feature) : "memory")
 
 /*
@@ -66,35 +106,35 @@ static inline void alternatives_smp_switch(int smp) {}
  * Argument numbers start with 1.
  * Best is to use constraints that are fixed size (like (%1) ... "r")
  * If you use variable sized constraints like "m" or "g" in the
- * replacement maake sure to pad to the worst case length.
+ * replacement make sure to pad to the worst case length.
  */
 #define alternative_input(oldinstr, newinstr, feature, input...)	\
 	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
 		      ".section .altinstructions,\"a\"\n"		\
-		      "  .align 4\n"					\
-		      "  .long 661b\n"            /* label */		\
-		      "  .long 663f\n"		  /* new instruction */ \
-		      "  .byte %c0\n"             /* feature bit */	\
-		      "  .byte 662b-661b\n"       /* sourcelen */	\
-		      "  .byte 664f-663f\n"       /* replacementlen */ 	\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */\
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
 		      ".previous" :: "i" (feature), ##input)
 
 /* Like alternative_input, but with a single output argument */
-#define alternative_io(oldinstr, newinstr, feature, output, input...) \
+#define alternative_io(oldinstr, newinstr, feature, output, input...)	\
 	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
 		      ".section .altinstructions,\"a\"\n"		\
-		      "  .align 4\n"					\
-		      "  .long 661b\n"            /* label */		\
-		      "  .long 663f\n"		  /* new instruction */	\
-		      "  .byte %c[feat]\n"        /* feature bit */	\
-		      "  .byte 662b-661b\n"       /* sourcelen */	\
-		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c[feat]\n"	/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */ \
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */ \
 		      ".previous" : output : [feat] "i" (feature), ##input)
 
 /*
@@ -103,39 +143,6 @@ static inline void alternatives_smp_switch(int smp) {}
  */
 #define ASM_OUTPUT2(a, b) a, b
 
-/*
- * Alternative inline assembly for SMP.
- *
- * The LOCK_PREFIX macro defined here replaces the LOCK and
- * LOCK_PREFIX macros used everywhere in the source tree.
- *
- * SMP alternatives use the same data structures as the other
- * alternatives and the X86_FEATURE_UP flag to indicate the case of a
- * UP system running a SMP kernel.  The existing apply_alternatives()
- * works fine for patching a SMP kernel for UP.
- *
- * The SMP alternative tables can be kept after boot and contain both
- * UP and SMP versions of the instructions to allow switching back to
- * SMP at runtime, when hotplugging in a new CPU, which is especially
- * useful in virtualized environments.
- *
- * The very common lock prefix is handled as special case in a
- * separate table which is a pure address list without replacement ptr
- * and size information.  That keeps the table sizes small.
- */
-
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX \
-		".section .smp_locks,\"a\"\n"	\
-		"  .align 4\n"			\
-		"  .long 661f\n" /* address */	\
-		".previous\n"			\
-	       	"661:\n\tlock; "
-
-#else /* ! CONFIG_SMP */
-#define LOCK_PREFIX ""
-#endif
-
 struct paravirt_patch_site;
 #ifdef CONFIG_PARAVIRT
 void apply_paravirt(struct paravirt_patch_site *start,
diff --git a/include/asm-x86/alternative_64.h b/include/asm-x86/alternative_64.h
index ab161e810151..50efcebae33f 100644
--- a/include/asm-x86/alternative_64.h
+++ b/include/asm-x86/alternative_64.h
@@ -1,10 +1,9 @@
 #ifndef _X86_64_ALTERNATIVE_H
 #define _X86_64_ALTERNATIVE_H
 
-#ifdef __KERNEL__
-
 #include <linux/types.h>
 #include <linux/stddef.h>
+#include <asm/asm.h>
 
 /*
  * Alternative inline assembly for SMP.
@@ -30,10 +29,10 @@
 #ifdef CONFIG_SMP
 #define LOCK_PREFIX \
 		".section .smp_locks,\"a\"\n"	\
-		"  .align 8\n"			\
-		"  .quad 661f\n" /* address */	\
+		_ASM_ALIGN "\n"			\
+		_ASM_PTR "661f\n" /* address */	\
 		".previous\n"			\
-	       	"661:\n\tlock; "
+		"661:\n\tlock; "
 
 #else /* ! CONFIG_SMP */
 #define LOCK_PREFIX ""
@@ -43,12 +42,15 @@
 #include <asm/cpufeature.h>
 
 struct alt_instr {
-	u8 *instr; 		/* original instruction */
+	u8 *instr;		/* original instruction */
 	u8 *replacement;
 	u8  cpuid;		/* cpuid bit set for replacement */
 	u8  instrlen;		/* length of original instruction */
-	u8  replacementlen; 	/* length of new instruction, <= instrlen */
-	u8  pad[5];
+	u8  replacementlen;	/* length of new instruction, <= instrlen */
+	u8  pad1;
+#ifdef CONFIG_X86_64
+	u32 pad2;
+#endif
 };
 
 extern void alternative_instructions(void);
@@ -68,9 +70,7 @@ static inline void alternatives_smp_module_add(struct module *mod, char *name,
 					void *text, void *text_end) {}
 static inline void alternatives_smp_module_del(struct module *mod) {}
 static inline void alternatives_smp_switch(int smp) {}
-#endif
-
-#endif
+#endif	/* CONFIG_SMP */
 
 /*
  * Alternative instructions for different CPU types or capabilities.
@@ -84,18 +84,18 @@ static inline void alternatives_smp_switch(int smp) {}
  * For non barrier like inlines please define new variants
  * without volatile and memory clobber.
  */
-#define alternative(oldinstr, newinstr, feature) 	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n" 		     \
-		      ".section .altinstructions,\"a\"\n"     	     \
-		      "  .align 8\n"				       \
-		      "  .quad 661b\n"            /* label */          \
-		      "  .quad 663f\n"		  /* new instruction */ \
-		      "  .byte %c0\n"             /* feature bit */    \
-		      "  .byte 662b-661b\n"       /* sourcelen */      \
-		      "  .byte 664f-663f\n"       /* replacementlen */ \
+#define alternative(oldinstr, newinstr, feature)			\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */ \
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
 		      ".previous" :: "i" (feature) : "memory")
 
 /*
@@ -111,30 +111,30 @@ static inline void alternatives_smp_switch(int smp) {}
 #define alternative_input(oldinstr, newinstr, feature, input...)	\
 	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
 		      ".section .altinstructions,\"a\"\n"		\
-		      "  .align 8\n"					\
-		      "  .quad 661b\n"            /* label */		\
-		      "  .quad 663f\n"		  /* new instruction */	\
-		      "  .byte %c0\n"             /* feature bit */	\
-		      "  .byte 662b-661b\n"       /* sourcelen */	\
-		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */ \
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
 		      ".previous" :: "i" (feature), ##input)
 
 /* Like alternative_input, but with a single output argument */
-#define alternative_io(oldinstr, newinstr, feature, output, input...) \
+#define alternative_io(oldinstr, newinstr, feature, output, input...)	\
 	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
 		      ".section .altinstructions,\"a\"\n"		\
-		      "  .align 8\n"					\
-		      "  .quad 661b\n"            /* label */		\
-		      "  .quad 663f\n"		  /* new instruction */	\
-		      "  .byte %c[feat]\n"        /* feature bit */	\
-		      "  .byte 662b-661b\n"       /* sourcelen */	\
-		      "  .byte 664f-663f\n"       /* replacementlen */	\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c[feat]\n"	/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
 		      ".previous\n"					\
 		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"   /* replacement */ \
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */ \
 		      ".previous" : output : [feat] "i" (feature), ##input)
 
 /*
@@ -143,15 +143,17 @@ static inline void alternatives_smp_switch(int smp) {}
  */
 #define ASM_OUTPUT2(a, b) a, b
 
-struct paravirt_patch;
+struct paravirt_patch_site;
 #ifdef CONFIG_PARAVIRT
-void apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end);
+void apply_paravirt(struct paravirt_patch_site *start,
+		    struct paravirt_patch_site *end);
 #else
 static inline void
-apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end)
+apply_paravirt(struct paravirt_patch_site *start,
+	       struct paravirt_patch_site *end)
 {}
-#define __parainstructions NULL
-#define __parainstructions_end NULL
+#define __parainstructions	NULL
+#define __parainstructions_end	NULL
 #endif
 
 extern void text_poke(void *addr, unsigned char *opcode, int len);
-- 
cgit v1.2.3


From 6b59257082e136fbbce1aa017575a451c0df0592 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:30 +0100
Subject: x86: actually merge <asm/alternative.h>

This actually merges <asm-x86/alternative_{32,64}.h> into
<asm-x86/alternative.h>.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/alternative.h    | 162 ++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/alternative_32.h | 161 --------------------------------------
 include/asm-x86/alternative_64.h | 161 --------------------------------------
 3 files changed, 159 insertions(+), 325 deletions(-)
 delete mode 100644 include/asm-x86/alternative_32.h
 delete mode 100644 include/asm-x86/alternative_64.h

(limited to 'include')

diff --git a/include/asm-x86/alternative.h b/include/asm-x86/alternative.h
index 9eef6a32a130..d8bacf3c4b08 100644
--- a/include/asm-x86/alternative.h
+++ b/include/asm-x86/alternative.h
@@ -1,5 +1,161 @@
-#ifdef CONFIG_X86_32
-# include "alternative_32.h"
+#ifndef _ASM_X86_ALTERNATIVE_H
+#define _ASM_X86_ALTERNATIVE_H
+
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <asm/asm.h>
+
+/*
+ * Alternative inline assembly for SMP.
+ *
+ * The LOCK_PREFIX macro defined here replaces the LOCK and
+ * LOCK_PREFIX macros used everywhere in the source tree.
+ *
+ * SMP alternatives use the same data structures as the other
+ * alternatives and the X86_FEATURE_UP flag to indicate the case of a
+ * UP system running a SMP kernel.  The existing apply_alternatives()
+ * works fine for patching a SMP kernel for UP.
+ *
+ * The SMP alternative tables can be kept after boot and contain both
+ * UP and SMP versions of the instructions to allow switching back to
+ * SMP at runtime, when hotplugging in a new CPU, which is especially
+ * useful in virtualized environments.
+ *
+ * The very common lock prefix is handled as special case in a
+ * separate table which is a pure address list without replacement ptr
+ * and size information.  That keeps the table sizes small.
+ */
+
+#ifdef CONFIG_SMP
+#define LOCK_PREFIX \
+		".section .smp_locks,\"a\"\n"	\
+		_ASM_ALIGN "\n"			\
+		_ASM_PTR "661f\n" /* address */	\
+		".previous\n"			\
+		"661:\n\tlock; "
+
+#else /* ! CONFIG_SMP */
+#define LOCK_PREFIX ""
+#endif
+
+/* This must be included *after* the definition of LOCK_PREFIX */
+#include <asm/cpufeature.h>
+
+struct alt_instr {
+	u8 *instr;		/* original instruction */
+	u8 *replacement;
+	u8  cpuid;		/* cpuid bit set for replacement */
+	u8  instrlen;		/* length of original instruction */
+	u8  replacementlen;	/* length of new instruction, <= instrlen */
+	u8  pad1;
+#ifdef CONFIG_X86_64
+	u32 pad2;
+#endif
+};
+
+extern void alternative_instructions(void);
+extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
+
+struct module;
+
+#ifdef CONFIG_SMP
+extern void alternatives_smp_module_add(struct module *mod, char *name,
+					void *locks, void *locks_end,
+					void *text, void *text_end);
+extern void alternatives_smp_module_del(struct module *mod);
+extern void alternatives_smp_switch(int smp);
+#else
+static inline void alternatives_smp_module_add(struct module *mod, char *name,
+					void *locks, void *locks_end,
+					void *text, void *text_end) {}
+static inline void alternatives_smp_module_del(struct module *mod) {}
+static inline void alternatives_smp_switch(int smp) {}
+#endif	/* CONFIG_SMP */
+
+/*
+ * Alternative instructions for different CPU types or capabilities.
+ *
+ * This allows to use optimized instructions even on generic binary
+ * kernels.
+ *
+ * length of oldinstr must be longer or equal the length of newinstr
+ * It can be padded with nops as needed.
+ *
+ * For non barrier like inlines please define new variants
+ * without volatile and memory clobber.
+ */
+#define alternative(oldinstr, newinstr, feature)			\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
+		      ".previous\n"					\
+		      ".section .altinstr_replacement,\"ax\"\n"		\
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
+		      ".previous" :: "i" (feature) : "memory")
+
+/*
+ * Alternative inline assembly with input.
+ *
+ * Pecularities:
+ * No memory clobber here.
+ * Argument numbers start with 1.
+ * Best is to use constraints that are fixed size (like (%1) ... "r")
+ * If you use variable sized constraints like "m" or "g" in the
+ * replacement make sure to pad to the worst case length.
+ */
+#define alternative_input(oldinstr, newinstr, feature, input...)	\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c0\n"		/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
+		      ".previous\n"					\
+		      ".section .altinstr_replacement,\"ax\"\n"		\
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
+		      ".previous" :: "i" (feature), ##input)
+
+/* Like alternative_input, but with a single output argument */
+#define alternative_io(oldinstr, newinstr, feature, output, input...)	\
+	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
+		      ".section .altinstructions,\"a\"\n"		\
+		      _ASM_ALIGN "\n"					\
+		      _ASM_PTR "661b\n"		/* label */		\
+		      _ASM_PTR "663f\n"		/* new instruction */	\
+		      "	 .byte %c[feat]\n"	/* feature bit */	\
+		      "	 .byte 662b-661b\n"	/* sourcelen */		\
+		      "	 .byte 664f-663f\n"	/* replacementlen */	\
+		      ".previous\n"					\
+		      ".section .altinstr_replacement,\"ax\"\n"		\
+		      "663:\n\t" newinstr "\n664:\n"  /* replacement */ \
+		      ".previous" : output : [feat] "i" (feature), ##input)
+
+/*
+ * use this macro(s) if you need more than one output parameter
+ * in alternative_io
+ */
+#define ASM_OUTPUT2(a, b) a, b
+
+struct paravirt_patch_site;
+#ifdef CONFIG_PARAVIRT
+void apply_paravirt(struct paravirt_patch_site *start,
+		    struct paravirt_patch_site *end);
 #else
-# include "alternative_64.h"
+static inline void
+apply_paravirt(struct paravirt_patch_site *start,
+	       struct paravirt_patch_site *end)
+{}
+#define __parainstructions	NULL
+#define __parainstructions_end	NULL
 #endif
+
+extern void text_poke(void *addr, unsigned char *opcode, int len);
+
+#endif /* _ASM_X86_ALTERNATIVE_H */
diff --git a/include/asm-x86/alternative_32.h b/include/asm-x86/alternative_32.h
deleted file mode 100644
index 4919b5ee8b9f..000000000000
--- a/include/asm-x86/alternative_32.h
+++ /dev/null
@@ -1,161 +0,0 @@
-#ifndef _I386_ALTERNATIVE_H
-#define _I386_ALTERNATIVE_H
-
-#include <linux/types.h>
-#include <linux/stddef.h>
-#include <asm/asm.h>
-
-/*
- * Alternative inline assembly for SMP.
- *
- * The LOCK_PREFIX macro defined here replaces the LOCK and
- * LOCK_PREFIX macros used everywhere in the source tree.
- *
- * SMP alternatives use the same data structures as the other
- * alternatives and the X86_FEATURE_UP flag to indicate the case of a
- * UP system running a SMP kernel.  The existing apply_alternatives()
- * works fine for patching a SMP kernel for UP.
- *
- * The SMP alternative tables can be kept after boot and contain both
- * UP and SMP versions of the instructions to allow switching back to
- * SMP at runtime, when hotplugging in a new CPU, which is especially
- * useful in virtualized environments.
- *
- * The very common lock prefix is handled as special case in a
- * separate table which is a pure address list without replacement ptr
- * and size information.  That keeps the table sizes small.
- */
-
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX \
-		".section .smp_locks,\"a\"\n"	\
-		_ASM_ALIGN "\n"			\
-		_ASM_PTR "661f\n" /* address */	\
-		".previous\n"			\
-		"661:\n\tlock; "
-
-#else /* ! CONFIG_SMP */
-#define LOCK_PREFIX ""
-#endif
-
-/* This must be included *after* the definition of LOCK_PREFIX */
-#include <asm/cpufeature.h>
-
-struct alt_instr {
-	u8 *instr;		/* original instruction */
-	u8 *replacement;
-	u8  cpuid;		/* cpuid bit set for replacement */
-	u8  instrlen;		/* length of original instruction */
-	u8  replacementlen;	/* length of new instruction, <= instrlen */
-	u8  pad1;
-#ifdef CONFIG_X86_64
-	u32 pad2;
-#endif
-};
-
-extern void alternative_instructions(void);
-extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
-
-struct module;
-
-#ifdef CONFIG_SMP
-extern void alternatives_smp_module_add(struct module *mod, char *name,
-					void *locks, void *locks_end,
-					void *text, void *text_end);
-extern void alternatives_smp_module_del(struct module *mod);
-extern void alternatives_smp_switch(int smp);
-#else
-static inline void alternatives_smp_module_add(struct module *mod, char *name,
-					void *locks, void *locks_end,
-					void *text, void *text_end) {}
-static inline void alternatives_smp_module_del(struct module *mod) {}
-static inline void alternatives_smp_switch(int smp) {}
-#endif	/* CONFIG_SMP */
-
-/*
- * Alternative instructions for different CPU types or capabilities.
- *
- * This allows to use optimized instructions even on generic binary
- * kernels.
- *
- * length of oldinstr must be longer or equal the length of newinstr
- * It can be padded with nops as needed.
- *
- * For non barrier like inlines please define new variants
- * without volatile and memory clobber.
- */
-#define alternative(oldinstr, newinstr, feature)			\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c0\n"		/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
-		      ".previous" :: "i" (feature) : "memory")
-
-/*
- * Alternative inline assembly with input.
- *
- * Pecularities:
- * No memory clobber here.
- * Argument numbers start with 1.
- * Best is to use constraints that are fixed size (like (%1) ... "r")
- * If you use variable sized constraints like "m" or "g" in the
- * replacement make sure to pad to the worst case length.
- */
-#define alternative_input(oldinstr, newinstr, feature, input...)	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c0\n"		/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
-		      ".previous" :: "i" (feature), ##input)
-
-/* Like alternative_input, but with a single output argument */
-#define alternative_io(oldinstr, newinstr, feature, output, input...)	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c[feat]\n"	/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */ \
-		      ".previous" : output : [feat] "i" (feature), ##input)
-
-/*
- * use this macro(s) if you need more than one output parameter
- * in alternative_io
- */
-#define ASM_OUTPUT2(a, b) a, b
-
-struct paravirt_patch_site;
-#ifdef CONFIG_PARAVIRT
-void apply_paravirt(struct paravirt_patch_site *start,
-		    struct paravirt_patch_site *end);
-#else
-static inline void
-apply_paravirt(struct paravirt_patch_site *start,
-	       struct paravirt_patch_site *end)
-{}
-#define __parainstructions	NULL
-#define __parainstructions_end	NULL
-#endif
-
-extern void text_poke(void *addr, unsigned char *opcode, int len);
-
-#endif /* _I386_ALTERNATIVE_H */
diff --git a/include/asm-x86/alternative_64.h b/include/asm-x86/alternative_64.h
deleted file mode 100644
index 50efcebae33f..000000000000
--- a/include/asm-x86/alternative_64.h
+++ /dev/null
@@ -1,161 +0,0 @@
-#ifndef _X86_64_ALTERNATIVE_H
-#define _X86_64_ALTERNATIVE_H
-
-#include <linux/types.h>
-#include <linux/stddef.h>
-#include <asm/asm.h>
-
-/*
- * Alternative inline assembly for SMP.
- *
- * The LOCK_PREFIX macro defined here replaces the LOCK and
- * LOCK_PREFIX macros used everywhere in the source tree.
- *
- * SMP alternatives use the same data structures as the other
- * alternatives and the X86_FEATURE_UP flag to indicate the case of a
- * UP system running a SMP kernel.  The existing apply_alternatives()
- * works fine for patching a SMP kernel for UP.
- *
- * The SMP alternative tables can be kept after boot and contain both
- * UP and SMP versions of the instructions to allow switching back to
- * SMP at runtime, when hotplugging in a new CPU, which is especially
- * useful in virtualized environments.
- *
- * The very common lock prefix is handled as special case in a
- * separate table which is a pure address list without replacement ptr
- * and size information.  That keeps the table sizes small.
- */
-
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX \
-		".section .smp_locks,\"a\"\n"	\
-		_ASM_ALIGN "\n"			\
-		_ASM_PTR "661f\n" /* address */	\
-		".previous\n"			\
-		"661:\n\tlock; "
-
-#else /* ! CONFIG_SMP */
-#define LOCK_PREFIX ""
-#endif
-
-/* This must be included *after* the definition of LOCK_PREFIX */
-#include <asm/cpufeature.h>
-
-struct alt_instr {
-	u8 *instr;		/* original instruction */
-	u8 *replacement;
-	u8  cpuid;		/* cpuid bit set for replacement */
-	u8  instrlen;		/* length of original instruction */
-	u8  replacementlen;	/* length of new instruction, <= instrlen */
-	u8  pad1;
-#ifdef CONFIG_X86_64
-	u32 pad2;
-#endif
-};
-
-extern void alternative_instructions(void);
-extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
-
-struct module;
-
-#ifdef CONFIG_SMP
-extern void alternatives_smp_module_add(struct module *mod, char *name,
-					void *locks, void *locks_end,
-					void *text, void *text_end);
-extern void alternatives_smp_module_del(struct module *mod);
-extern void alternatives_smp_switch(int smp);
-#else
-static inline void alternatives_smp_module_add(struct module *mod, char *name,
-					void *locks, void *locks_end,
-					void *text, void *text_end) {}
-static inline void alternatives_smp_module_del(struct module *mod) {}
-static inline void alternatives_smp_switch(int smp) {}
-#endif	/* CONFIG_SMP */
-
-/*
- * Alternative instructions for different CPU types or capabilities.
- *
- * This allows to use optimized instructions even on generic binary
- * kernels.
- *
- * length of oldinstr must be longer or equal the length of newinstr
- * It can be padded with nops as needed.
- *
- * For non barrier like inlines please define new variants
- * without volatile and memory clobber.
- */
-#define alternative(oldinstr, newinstr, feature)			\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c0\n"		/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
-		      ".previous" :: "i" (feature) : "memory")
-
-/*
- * Alternative inline assembly with input.
- *
- * Pecularities:
- * No memory clobber here.
- * Argument numbers start with 1.
- * Best is to use constraints that are fixed size (like (%1) ... "r")
- * If you use variable sized constraints like "m" or "g" in the
- * replacement make sure to pad to the worst case length.
- */
-#define alternative_input(oldinstr, newinstr, feature, input...)	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c0\n"		/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */	\
-		      ".previous" :: "i" (feature), ##input)
-
-/* Like alternative_input, but with a single output argument */
-#define alternative_io(oldinstr, newinstr, feature, output, input...)	\
-	asm volatile ("661:\n\t" oldinstr "\n662:\n"			\
-		      ".section .altinstructions,\"a\"\n"		\
-		      _ASM_ALIGN "\n"					\
-		      _ASM_PTR "661b\n"		/* label */		\
-		      _ASM_PTR "663f\n"		/* new instruction */	\
-		      "	 .byte %c[feat]\n"	/* feature bit */	\
-		      "	 .byte 662b-661b\n"	/* sourcelen */		\
-		      "	 .byte 664f-663f\n"	/* replacementlen */	\
-		      ".previous\n"					\
-		      ".section .altinstr_replacement,\"ax\"\n"		\
-		      "663:\n\t" newinstr "\n664:\n"  /* replacement */ \
-		      ".previous" : output : [feat] "i" (feature), ##input)
-
-/*
- * use this macro(s) if you need more than one output parameter
- * in alternative_io
- */
-#define ASM_OUTPUT2(a, b) a, b
-
-struct paravirt_patch_site;
-#ifdef CONFIG_PARAVIRT
-void apply_paravirt(struct paravirt_patch_site *start,
-		    struct paravirt_patch_site *end);
-#else
-static inline void
-apply_paravirt(struct paravirt_patch_site *start,
-	       struct paravirt_patch_site *end)
-{}
-#define __parainstructions	NULL
-#define __parainstructions_end	NULL
-#endif
-
-extern void text_poke(void *addr, unsigned char *opcode, int len);
-
-#endif /* _X86_64_ALTERNATIVE_H */
-- 
cgit v1.2.3


From 013d23e1567c4cebee0a2db5c8fa97b91b34ac2a Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 30 Jan 2008 13:30:30 +0100
Subject: x86 e820_64.c: make 2 functions static

This patch makes the following needlessly global functions static:
- e820_print_map()
- early_panic()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_64.c | 4 ++--
 include/asm-x86/e820_64.h | 1 -
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 0128b0b0e5b5..11a3d65db0c1 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -373,7 +373,7 @@ unsigned long __init e820_hole_size(unsigned long start, unsigned long end)
 	return end - start - (ram << PAGE_SHIFT);
 }
 
-void __init e820_print_map(char *who)
+static void __init e820_print_map(char *who)
 {
 	int i;
 
@@ -633,7 +633,7 @@ static int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
 	return 0;
 }
 
-void early_panic(char *msg)
+static void early_panic(char *msg)
 {
 	early_printk(msg);
 	panic(msg);
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index 0bd4787a5d57..e535e6044e21 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -21,7 +21,6 @@ extern void contig_e820_setup(void);
 extern unsigned long e820_end_of_ram(void);
 extern void e820_reserve_resources(void);
 extern void e820_mark_nosave_regions(void);
-extern void e820_print_map(char *who);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
 extern unsigned long e820_hole_size(unsigned long start, unsigned long end);
-- 
cgit v1.2.3


From 867ab545668385b903f9379019000383675c49b3 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 30 Jan 2008 13:30:31 +0100
Subject: x86: nmi_64.c: make code static

This patch makes the following needlessly global code static:
- panic_on_timeout
- setup_nmi_watchdog()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/nmi_64.c | 4 ++--
 include/asm-x86/nmi_64.h | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index c3d1476b6a11..53faef632fc6 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -39,7 +39,7 @@ static cpumask_t backtrace_mask = CPU_MASK_NONE;
  *  0: the lapic NMI watchdog is disabled, but can be enabled
  */
 atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
-int panic_on_timeout;
+static int panic_on_timeout;
 
 unsigned int nmi_watchdog = NMI_DEFAULT;
 static unsigned int nmi_hz = HZ;
@@ -136,7 +136,7 @@ int __init check_nmi_watchdog (void)
 	return 0;
 }
 
-int __init setup_nmi_watchdog(char *str)
+static int __init setup_nmi_watchdog(char *str)
 {
 	int nmi;
 
diff --git a/include/asm-x86/nmi_64.h b/include/asm-x86/nmi_64.h
index 92f36fb1dc62..2eeb74e5f3ff 100644
--- a/include/asm-x86/nmi_64.h
+++ b/include/asm-x86/nmi_64.h
@@ -38,7 +38,6 @@ extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 
 #define get_nmi_reason() inb(0x61)
 
-extern int panic_on_timeout;
 extern int unknown_nmi_panic;
 extern int nmi_watchdog_enabled;
 
@@ -57,7 +56,6 @@ extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
 extern void nmi_watchdog_default(void);
-extern int setup_nmi_watchdog(char *);
 
 extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
-- 
cgit v1.2.3


From 3e7593966be6f6d29a15138c0c96b961d437f2f5 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 30 Jan 2008 13:30:31 +0100
Subject: x86: pci-dma_64.c: cleanups

This patch contains the following cleanups:
- make the needlessly global iommu_setup() static
- remove the unused EXPORT_SYMBOL(iommu_merge)

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/pci-dma_64.c | 3 +--
 include/asm-x86/pci_64.h     | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/pci-dma_64.c b/arch/x86/kernel/pci-dma_64.c
index 5552d23d23c2..a82473d192a3 100644
--- a/arch/x86/kernel/pci-dma_64.c
+++ b/arch/x86/kernel/pci-dma_64.c
@@ -13,7 +13,6 @@
 #include <asm/calgary.h>
 
 int iommu_merge __read_mostly = 0;
-EXPORT_SYMBOL(iommu_merge);
 
 dma_addr_t bad_dma_address __read_mostly;
 EXPORT_SYMBOL(bad_dma_address);
@@ -230,7 +229,7 @@ EXPORT_SYMBOL(dma_set_mask);
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
  */
-__init int iommu_setup(char *p)
+static __init int iommu_setup(char *p)
 {
 	iommu_merge = 1;
 
diff --git a/include/asm-x86/pci_64.h b/include/asm-x86/pci_64.h
index ef54226a9325..374690314539 100644
--- a/include/asm-x86/pci_64.h
+++ b/include/asm-x86/pci_64.h
@@ -26,7 +26,6 @@ extern int (*pci_config_write)(int seg, int bus, int dev, int fn, int reg, int l
 
 
 extern void pci_iommu_alloc(void);
-extern int iommu_setup(char *opt);
 
 /* The PCI address space does equal the physical memory
  * address space.  The networking and block device layers use
-- 
cgit v1.2.3


From 2d2ee8de5f6d26ef2942e0b449aa68d9236d5777 Mon Sep 17 00:00:00 2001
From: Paul Jimenez <pj@place.org>
Date: Wed, 30 Jan 2008 13:30:31 +0100
Subject: x86: mtrr use type bool [RESEND AGAIN]

This is a janitorish patch to 1) remove private TRUE/FALSE #def's in
favor of using the standard enum from linux/stddef.h and 2) switch the
variables holding those values to type 'bool' (from linux/types.h)
since it both seems more appropriate and allows for potentially better
optimization.

As a truly minor aside, I removed a couple of comments documenting
a 'do_safe' parameter that seems to no longer exist.

Signed-off-by: Paul Jimenez <pj@place.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/mtrr/amd.c     |  2 --
 arch/x86/kernel/cpu/mtrr/generic.c | 16 +++++++---------
 arch/x86/kernel/cpu/mtrr/if.c      | 15 ++++++++-------
 arch/x86/kernel/cpu/mtrr/main.c    |  8 +++++---
 arch/x86/kernel/cpu/mtrr/mtrr.h    |  6 ++----
 include/asm-x86/mtrr.h             |  8 ++++----
 6 files changed, 26 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/mtrr/amd.c b/arch/x86/kernel/cpu/mtrr/amd.c
index 0949cdbf848a..ee2331b0e58f 100644
--- a/arch/x86/kernel/cpu/mtrr/amd.c
+++ b/arch/x86/kernel/cpu/mtrr/amd.c
@@ -53,8 +53,6 @@ static void amd_set_mtrr(unsigned int reg, unsigned long base,
     <base> The base address of the region.
     <size> The size of the region. If this is 0 the region is disabled.
     <type> The type of the region.
-    <do_safe> If TRUE, do the change safely. If FALSE, safety measures should
-    be done externally.
     [RETURNS] Nothing.
 */
 {
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 992f08dfbb6c..1c331c373a43 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -188,7 +188,7 @@ static inline void k8_enable_fixed_iorrs(void)
  * \param changed pointer which indicates whether the MTRR needed to be changed
  * \param msrwords pointer to the MSR values which the MSR should have
  */
-static void set_fixed_range(int msr, int * changed, unsigned int * msrwords)
+static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords)
 {
 	unsigned lo, hi;
 
@@ -200,7 +200,7 @@ static void set_fixed_range(int msr, int * changed, unsigned int * msrwords)
 		    ((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK))
 			k8_enable_fixed_iorrs();
 		mtrr_wrmsr(msr, msrwords[0], msrwords[1]);
-		*changed = TRUE;
+		*changed = true;
 	}
 }
 
@@ -260,7 +260,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base,
 static int set_fixed_ranges(mtrr_type * frs)
 {
 	unsigned long long *saved = (unsigned long long *) frs;
-	int changed = FALSE;
+	bool changed = false;
 	int block=-1, range;
 
 	while (fixed_range_blocks[++block].ranges)
@@ -273,17 +273,17 @@ static int set_fixed_ranges(mtrr_type * frs)
 
 /*  Set the MSR pair relating to a var range. Returns TRUE if
     changes are made  */
-static int set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
+static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
 {
 	unsigned int lo, hi;
-	int changed = FALSE;
+	bool changed = false;
 
 	rdmsr(MTRRphysBase_MSR(index), lo, hi);
 	if ((vr->base_lo & 0xfffff0ffUL) != (lo & 0xfffff0ffUL)
 	    || (vr->base_hi & (size_and_mask >> (32 - PAGE_SHIFT))) !=
 		(hi & (size_and_mask >> (32 - PAGE_SHIFT)))) {
 		mtrr_wrmsr(MTRRphysBase_MSR(index), vr->base_lo, vr->base_hi);
-		changed = TRUE;
+		changed = true;
 	}
 
 	rdmsr(MTRRphysMask_MSR(index), lo, hi);
@@ -292,7 +292,7 @@ static int set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
 	    || (vr->mask_hi & (size_and_mask >> (32 - PAGE_SHIFT))) !=
 		(hi & (size_and_mask >> (32 - PAGE_SHIFT)))) {
 		mtrr_wrmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi);
-		changed = TRUE;
+		changed = true;
 	}
 	return changed;
 }
@@ -417,8 +417,6 @@ static void generic_set_mtrr(unsigned int reg, unsigned long base,
     <base> The base address of the region.
     <size> The size of the region. If this is 0 the region is disabled.
     <type> The type of the region.
-    <do_safe> If TRUE, do the change safely. If FALSE, safety measures should
-    be done externally.
     [RETURNS] Nothing.
 */
 {
diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c
index c7d8f1756745..14535686c099 100644
--- a/arch/x86/kernel/cpu/mtrr/if.c
+++ b/arch/x86/kernel/cpu/mtrr/if.c
@@ -37,7 +37,7 @@ const char *mtrr_attrib_to_str(int x)
 
 static int
 mtrr_file_add(unsigned long base, unsigned long size,
-	      unsigned int type, char increment, struct file *file, int page)
+	      unsigned int type, bool increment, struct file *file, int page)
 {
 	int reg, max;
 	unsigned int *fcount = FILE_FCOUNT(file); 
@@ -55,7 +55,7 @@ mtrr_file_add(unsigned long base, unsigned long size,
 		base >>= PAGE_SHIFT;
 		size >>= PAGE_SHIFT;
 	}
-	reg = mtrr_add_page(base, size, type, 1);
+	reg = mtrr_add_page(base, size, type, true);
 	if (reg >= 0)
 		++fcount[reg];
 	return reg;
@@ -141,7 +141,7 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
 		size >>= PAGE_SHIFT;
 		err =
 		    mtrr_add_page((unsigned long) base, (unsigned long) size, i,
-				  1);
+				  true);
 		if (err < 0)
 			return err;
 		return len;
@@ -217,7 +217,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 		err =
-		    mtrr_file_add(sentry.base, sentry.size, sentry.type, 1,
+		    mtrr_file_add(sentry.base, sentry.size, sentry.type, true,
 				  file, 0);
 		break;
 	case MTRRIOC_SET_ENTRY:
@@ -226,7 +226,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 #endif
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		err = mtrr_add(sentry.base, sentry.size, sentry.type, 0);
+		err = mtrr_add(sentry.base, sentry.size, sentry.type, false);
 		break;
 	case MTRRIOC_DEL_ENTRY:
 #ifdef CONFIG_COMPAT
@@ -270,7 +270,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 		err =
-		    mtrr_file_add(sentry.base, sentry.size, sentry.type, 1,
+		    mtrr_file_add(sentry.base, sentry.size, sentry.type, true,
 				  file, 1);
 		break;
 	case MTRRIOC_SET_PAGE_ENTRY:
@@ -279,7 +279,8 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 #endif
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		err = mtrr_add_page(sentry.base, sentry.size, sentry.type, 0);
+		err =
+		    mtrr_add_page(sentry.base, sentry.size, sentry.type, false);
 		break;
 	case MTRRIOC_DEL_PAGE_ENTRY:
 #ifdef CONFIG_COMPAT
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index beb45c9c0835..60af5ed2b5c0 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -311,7 +311,7 @@ static void set_mtrr(unsigned int reg, unsigned long base,
  */
 
 int mtrr_add_page(unsigned long base, unsigned long size, 
-		  unsigned int type, char increment)
+		  unsigned int type, bool increment)
 {
 	int i, replace, error;
 	mtrr_type ltype;
@@ -394,7 +394,9 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 		if (likely(replace < 0))
 			usage_table[i] = 1;
 		else {
-			usage_table[i] = usage_table[replace] + !!increment;
+			usage_table[i] = usage_table[replace];
+			if (increment)
+				usage_table[i]++;
 			if (unlikely(replace != i)) {
 				set_mtrr(replace, 0, 0, 0);
 				usage_table[replace] = 0;
@@ -460,7 +462,7 @@ static int mtrr_check(unsigned long base, unsigned long size)
 
 int
 mtrr_add(unsigned long base, unsigned long size, unsigned int type,
-	 char increment)
+	 bool increment)
 {
 	if (mtrr_check(base, size))
 		return -EINVAL;
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 289dfe6030e3..54347e9a95c0 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -2,10 +2,8 @@
  * local mtrr defines.
  */
 
-#ifndef TRUE
-#define TRUE  1
-#define FALSE 0
-#endif
+#include <linux/types.h>
+#include <linux/stddef.h>
 
 #define MTRRcap_MSR     0x0fe
 #define MTRRdefType_MSR 0x2ff
diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index e8320e4e6ca2..262670e42078 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -89,9 +89,9 @@ struct mtrr_gentry
 extern void mtrr_save_fixed_ranges(void *);
 extern void mtrr_save_state(void);
 extern int mtrr_add (unsigned long base, unsigned long size,
-		     unsigned int type, char increment);
+		     unsigned int type, bool increment);
 extern int mtrr_add_page (unsigned long base, unsigned long size,
-		     unsigned int type, char increment);
+		     unsigned int type, bool increment);
 extern int mtrr_del (int reg, unsigned long base, unsigned long size);
 extern int mtrr_del_page (int reg, unsigned long base, unsigned long size);
 extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi);
@@ -101,12 +101,12 @@ extern void mtrr_bp_init(void);
 #define mtrr_save_fixed_ranges(arg) do {} while (0)
 #define mtrr_save_state() do {} while (0)
 static __inline__ int mtrr_add (unsigned long base, unsigned long size,
-				unsigned int type, char increment)
+				unsigned int type, bool increment)
 {
     return -ENODEV;
 }
 static __inline__ int mtrr_add_page (unsigned long base, unsigned long size,
-				unsigned int type, char increment)
+				unsigned int type, bool increment)
 {
     return -ENODEV;
 }
-- 
cgit v1.2.3


From c9cce83dd1d59f52e2c8f8c7d265ba4854c40785 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bwalle@suse.de>
Date: Wed, 30 Jan 2008 13:30:32 +0100
Subject: x86: remove extern declarations for code, data, bss resources

This patch removes the extern struct resource declarations for
data_resource, code_resource and bss_resource on x86 and declares that
three structures as static as done on other architectures like IA64.

On i386, these structures are moved to setup_32.c (from e820_32.c) because
that's code that is not specific to e820 and also required on EFI systems.
That makes the "extern" reference superfluous.

On x86_64, data_resource, code_resource and bss_resource are passed to
e820_reserve_resources() as arguments just as done on i386 and IA64.  That
also avoids the "extern" reference and it's possible to make it static.

Signed-off-by: Bernhard Walle <bwalle@suse.de>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_32.c  | 110 ++-------------------------------------------
 arch/x86/kernel/e820_64.c  |  11 +++--
 arch/x86/kernel/setup_32.c | 106 +++++++++++++++++++++++++++++++++++++++++--
 arch/x86/kernel/setup_64.c |   8 ++--
 include/asm-x86/e820_32.h  |   6 +++
 include/asm-x86/e820_64.h  |   5 ++-
 include/linux/ioport.h     |   2 +
 7 files changed, 127 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c
index 18f500d185a2..87cadc86d5ee 100644
--- a/arch/x86/kernel/e820_32.c
+++ b/arch/x86/kernel/e820_32.c
@@ -37,26 +37,6 @@ unsigned long pci_mem_start = 0x10000000;
 EXPORT_SYMBOL(pci_mem_start);
 #endif
 extern int user_defined_memmap;
-struct resource data_resource = {
-	.name	= "Kernel data",
-	.start	= 0,
-	.end	= 0,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-struct resource code_resource = {
-	.name	= "Kernel code",
-	.start	= 0,
-	.end	= 0,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-struct resource bss_resource = {
-	.name	= "Kernel bss",
-	.start	= 0,
-	.end	= 0,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
-};
 
 static struct resource system_rom_resource = {
 	.name	= "System ROM",
@@ -111,60 +91,6 @@ static struct resource video_rom_resource = {
 	.flags	= IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
 };
 
-static struct resource video_ram_resource = {
-	.name	= "Video RAM area",
-	.start	= 0xa0000,
-	.end	= 0xbffff,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-static struct resource standard_io_resources[] = { {
-	.name	= "dma1",
-	.start	= 0x0000,
-	.end	= 0x001f,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "pic1",
-	.start	= 0x0020,
-	.end	= 0x0021,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name   = "timer0",
-	.start	= 0x0040,
-	.end    = 0x0043,
-	.flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name   = "timer1",
-	.start  = 0x0050,
-	.end    = 0x0053,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "keyboard",
-	.start	= 0x0060,
-	.end	= 0x006f,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "dma page reg",
-	.start	= 0x0080,
-	.end	= 0x008f,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "pic2",
-	.start	= 0x00a0,
-	.end	= 0x00a1,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "dma2",
-	.start	= 0x00c0,
-	.end	= 0x00df,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-	.name	= "fpu",
-	.start	= 0x00f0,
-	.end	= 0x00ff,
-	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
-} };
-
 #define ROMSIGNATURE 0xaa55
 
 static int __init romsignature(const unsigned char *rom)
@@ -260,10 +186,9 @@ static void __init probe_roms(void)
  * Request address space for all standard RAM and ROM resources
  * and also for regions reported as reserved by the e820.
  */
-static void __init
-legacy_init_iomem_resources(struct resource *code_resource,
-			    struct resource *data_resource,
-			    struct resource *bss_resource)
+void __init legacy_init_iomem_resources(struct resource *code_resource,
+		struct resource *data_resource,
+		struct resource *bss_resource)
 {
 	int i;
 
@@ -305,35 +230,6 @@ legacy_init_iomem_resources(struct resource *code_resource,
 	}
 }
 
-/*
- * Request address space for all standard resources
- *
- * This is called just before pcibios_init(), which is also a
- * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
- */
-static int __init request_standard_resources(void)
-{
-	int i;
-
-	printk("Setting up standard PCI resources\n");
-	if (efi_enabled)
-		efi_initialize_iomem_resources(&code_resource,
-				&data_resource, &bss_resource);
-	else
-		legacy_init_iomem_resources(&code_resource,
-				&data_resource, &bss_resource);
-
-	/* EFI systems may still have VGA */
-	request_resource(&iomem_resource, &video_ram_resource);
-
-	/* request I/O space for devices used on all i[345]86 PCs */
-	for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
-		request_resource(&ioport_resource, &standard_io_resources[i]);
-	return 0;
-}
-
-subsys_initcall(request_standard_resources);
-
 #if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION)
 /**
  * e820_mark_nosave_regions - Find the ranges of physical addresses that do not
diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 11a3d65db0c1..151236896243 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -47,8 +47,6 @@ unsigned long end_pfn_map;
  */
 static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
 
-extern struct resource code_resource, data_resource, bss_resource;
-
 /* Check for some hardcoded bad areas that early boot is not allowed to touch */
 static inline int bad_addr(unsigned long *addrp, unsigned long size)
 {
@@ -213,7 +211,8 @@ unsigned long __init e820_end_of_ram(void)
 /*
  * Mark e820 reserved areas as busy for the resource manager.
  */
-void __init e820_reserve_resources(void)
+void __init e820_reserve_resources(struct resource *code_resource,
+		struct resource *data_resource, struct resource *bss_resource)
 {
 	int i;
 	for (i = 0; i < e820.nr_map; i++) {
@@ -235,9 +234,9 @@ void __init e820_reserve_resources(void)
 			 * so we try it repeatedly and let the resource manager
 			 * test it.
 			 */
-			request_resource(res, &code_resource);
-			request_resource(res, &data_resource);
-			request_resource(res, &bss_resource);
+			request_resource(res, code_resource);
+			request_resource(res, data_resource);
+			request_resource(res, bss_resource);
 #ifdef CONFIG_KEXEC
 			if (crashk_res.start != crashk_res.end)
 				request_resource(res, &crashk_res);
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 236d30b264d8..32edf70d6b0d 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -73,9 +73,80 @@ int disable_pse __cpuinitdata = 0;
 /*
  * Machine setup..
  */
-extern struct resource code_resource;
-extern struct resource data_resource;
-extern struct resource bss_resource;
+static struct resource data_resource = {
+	.name	= "Kernel data",
+	.start	= 0,
+	.end	= 0,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource code_resource = {
+	.name	= "Kernel code",
+	.start	= 0,
+	.end	= 0,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource bss_resource = {
+	.name	= "Kernel bss",
+	.start	= 0,
+	.end	= 0,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource video_ram_resource = {
+	.name	= "Video RAM area",
+	.start	= 0xa0000,
+	.end	= 0xbffff,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource standard_io_resources[] = { {
+	.name	= "dma1",
+	.start	= 0x0000,
+	.end	= 0x001f,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "pic1",
+	.start	= 0x0020,
+	.end	= 0x0021,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name   = "timer0",
+	.start	= 0x0040,
+	.end    = 0x0043,
+	.flags  = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name   = "timer1",
+	.start  = 0x0050,
+	.end    = 0x0053,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "keyboard",
+	.start	= 0x0060,
+	.end	= 0x006f,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "dma page reg",
+	.start	= 0x0080,
+	.end	= 0x008f,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "pic2",
+	.start	= 0x00a0,
+	.end	= 0x00a1,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "dma2",
+	.start	= 0x00c0,
+	.end	= 0x00df,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+	.name	= "fpu",
+	.start	= 0x00f0,
+	.end	= 0x00ff,
+	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
+} };
 
 /* cpu data as detected by the assembly code in head.S */
 struct cpuinfo_x86 new_cpu_data __cpuinitdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
@@ -693,3 +764,32 @@ void __init setup_arch(char **cmdline_p)
 #endif
 #endif
 }
+
+/*
+ * Request address space for all standard resources
+ *
+ * This is called just before pcibios_init(), which is also a
+ * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
+ */
+static int __init request_standard_resources(void)
+{
+	int i;
+
+	printk(KERN_INFO "Setting up standard PCI resources\n");
+	if (efi_enabled)
+		efi_initialize_iomem_resources(&code_resource,
+				&data_resource, &bss_resource);
+	else
+		legacy_init_iomem_resources(&code_resource,
+				&data_resource, &bss_resource);
+
+	/* EFI systems may still have VGA */
+	request_resource(&iomem_resource, &video_ram_resource);
+
+	/* request I/O space for devices used on all i[345]86 PCs */
+	for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
+		request_resource(&ioport_resource, &standard_io_resources[i]);
+	return 0;
+}
+
+subsys_initcall(request_standard_resources);
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index bcb5f3aaa097..1acb435a0585 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -123,19 +123,19 @@ struct resource standard_io_resources[] = {
 
 #define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)
 
-struct resource data_resource = {
+static struct resource data_resource = {
 	.name = "Kernel data",
 	.start = 0,
 	.end = 0,
 	.flags = IORESOURCE_RAM,
 };
-struct resource code_resource = {
+static struct resource code_resource = {
 	.name = "Kernel code",
 	.start = 0,
 	.end = 0,
 	.flags = IORESOURCE_RAM,
 };
-struct resource bss_resource = {
+static struct resource bss_resource = {
 	.name = "Kernel bss",
 	.start = 0,
 	.end = 0,
@@ -438,7 +438,7 @@ void __init setup_arch(char **cmdline_p)
 	/*
 	 * We trust e820 completely. No explicit ROM probing in memory.
  	 */
-	e820_reserve_resources(); 
+	e820_reserve_resources(&code_resource, &data_resource, &bss_resource);
 	e820_mark_nosave_regions();
 
 	{
diff --git a/include/asm-x86/e820_32.h b/include/asm-x86/e820_32.h
index 03f60c690c8a..ae5ea19623fa 100644
--- a/include/asm-x86/e820_32.h
+++ b/include/asm-x86/e820_32.h
@@ -12,6 +12,8 @@
 #ifndef __E820_HEADER
 #define __E820_HEADER
 
+#include <linux/ioport.h>
+
 #define HIGH_MEMORY	(1024*1024)
 
 #ifndef __ASSEMBLY__
@@ -26,6 +28,9 @@ extern void register_bootmem_low_pages(unsigned long max_low_pfn);
 extern void e820_register_memory(void);
 extern void limit_regions(unsigned long long size);
 extern void print_memory_map(char *who);
+extern void legacy_init_iomem_resources(struct resource *code_resource,
+			    struct resource *data_resource,
+			    struct resource *bss_resource);
 
 #if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION)
 extern void e820_mark_nosave_regions(void);
@@ -35,5 +40,6 @@ static inline void e820_mark_nosave_regions(void)
 }
 #endif
 
+
 #endif/*!__ASSEMBLY__*/
 #endif/*__E820_HEADER*/
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index e535e6044e21..1c7ba8804176 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -11,6 +11,8 @@
 #ifndef __E820_HEADER
 #define __E820_HEADER
 
+#include <linux/ioport.h>
+
 #ifndef __ASSEMBLY__
 extern unsigned long find_e820_area(unsigned long start, unsigned long end, 
 				    unsigned size);
@@ -19,7 +21,8 @@ extern void add_memory_region(unsigned long start, unsigned long size,
 extern void setup_memory_region(void);
 extern void contig_e820_setup(void); 
 extern unsigned long e820_end_of_ram(void);
-extern void e820_reserve_resources(void);
+extern void e820_reserve_resources(struct resource *code_resource,
+		struct resource *data_resource, struct resource *bss_resource);
 extern void e820_mark_nosave_regions(void);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 6187a8567bc7..605d237364d2 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -8,6 +8,7 @@
 #ifndef _LINUX_IOPORT_H
 #define _LINUX_IOPORT_H
 
+#ifndef __ASSEMBLY__
 #include <linux/compiler.h>
 #include <linux/types.h>
 /*
@@ -153,4 +154,5 @@ extern struct resource * __devm_request_region(struct device *dev,
 extern void __devm_release_region(struct device *dev, struct resource *parent,
 				  resource_size_t start, resource_size_t n);
 
+#endif /* __ASSEMBLY__ */
 #endif	/* _LINUX_IOPORT_H */
-- 
cgit v1.2.3


From ec12fa5c80126b252ebcb474d12c832ec4647daa Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Wed, 30 Jan 2008 13:30:32 +0100
Subject: x86: bitops_32.h style cleanups

Coding style cleanups in x86/bitops_32.h:

- drop space in "* addr"
- whitespace & indentation fixes
- spello fixes

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/bitops_32.h | 48 ++++++++++++++++++++++-----------------------
 1 file changed, 24 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops_32.h b/include/asm-x86/bitops_32.h
index 0b40f6d20bea..5a29cce6a91d 100644
--- a/include/asm-x86/bitops_32.h
+++ b/include/asm-x86/bitops_32.h
@@ -37,7 +37,7 @@
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void set_bit(int nr, volatile unsigned long * addr)
+static inline void set_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__ __volatile__( LOCK_PREFIX
 		"btsl %1,%0"
@@ -54,7 +54,7 @@ static inline void set_bit(int nr, volatile unsigned long * addr)
  * If it's called on the same region of memory simultaneously, the effect
  * may be that only one operation succeeds.
  */
-static inline void __set_bit(int nr, volatile unsigned long * addr)
+static inline void __set_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__(
 		"btsl %1,%0"
@@ -72,7 +72,7 @@ static inline void __set_bit(int nr, volatile unsigned long * addr)
  * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
  * in order to ensure changes are visible on other processors.
  */
-static inline void clear_bit(int nr, volatile unsigned long * addr)
+static inline void clear_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__ __volatile__( LOCK_PREFIX
 		"btrl %1,%0"
@@ -94,7 +94,7 @@ static inline void clear_bit_unlock(unsigned long nr, volatile unsigned long *ad
 	clear_bit(nr, addr);
 }
 
-static inline void __clear_bit(int nr, volatile unsigned long * addr)
+static inline void __clear_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__ __volatile__(
 		"btrl %1,%0"
@@ -132,7 +132,7 @@ static inline void __clear_bit_unlock(unsigned long nr, volatile unsigned long *
  * If it's called on the same region of memory simultaneously, the effect
  * may be that only one operation succeeds.
  */
-static inline void __change_bit(int nr, volatile unsigned long * addr)
+static inline void __change_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__ __volatile__(
 		"btcl %1,%0"
@@ -150,7 +150,7 @@ static inline void __change_bit(int nr, volatile unsigned long * addr)
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void change_bit(int nr, volatile unsigned long * addr)
+static inline void change_bit(int nr, volatile unsigned long *addr)
 {
 	__asm__ __volatile__( LOCK_PREFIX
 		"btcl %1,%0"
@@ -167,7 +167,7 @@ static inline void change_bit(int nr, volatile unsigned long * addr)
  * It may be reordered on other architectures than x86.
  * It also implies a memory barrier.
  */
-static inline int test_and_set_bit(int nr, volatile unsigned long * addr)
+static inline int test_and_set_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
@@ -199,7 +199,7 @@ static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr)
  * If two examples of this operation race, one can appear to succeed
  * but actually fail.  You must protect multiple accesses with a lock.
  */
-static inline int __test_and_set_bit(int nr, volatile unsigned long * addr)
+static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
@@ -219,7 +219,7 @@ static inline int __test_and_set_bit(int nr, volatile unsigned long * addr)
  * It can be reorderdered on other architectures other than x86.
  * It also implies a memory barrier.
  */
-static inline int test_and_clear_bit(int nr, volatile unsigned long * addr)
+static inline int test_and_clear_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
@@ -270,7 +270,7 @@ static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
  * This operation is atomic and cannot be reordered.  
  * It also implies a memory barrier.
  */
-static inline int test_and_change_bit(int nr, volatile unsigned long* addr)
+static inline int test_and_change_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
@@ -287,7 +287,7 @@ static inline int test_and_change_bit(int nr, volatile unsigned long* addr)
  * @nr: bit number to test
  * @addr: Address to start counting from
  */
-static int test_bit(int nr, const volatile void * addr);
+static int test_bit(int nr, const volatile void *addr);
 #endif
 
 static __always_inline int constant_test_bit(int nr, const volatile unsigned long *addr)
@@ -295,7 +295,7 @@ static __always_inline int constant_test_bit(int nr, const volatile unsigned lon
 	return ((1UL << (nr & 31)) & (addr[nr >> 5])) != 0;
 }
 
-static inline int variable_test_bit(int nr, const volatile unsigned long * addr)
+static inline int variable_test_bit(int nr, const volatile unsigned long *addr)
 {
 	int oldbit;
 
@@ -306,10 +306,10 @@ static inline int variable_test_bit(int nr, const volatile unsigned long * addr)
 	return oldbit;
 }
 
-#define test_bit(nr,addr) \
-(__builtin_constant_p(nr) ? \
- constant_test_bit((nr),(addr)) : \
- variable_test_bit((nr),(addr)))
+#define test_bit(nr, addr) \
+	(__builtin_constant_p(nr) ? \
+		constant_test_bit((nr), (addr)) : \
+		variable_test_bit((nr), (addr)))
 
 #undef ADDR
 
@@ -318,7 +318,7 @@ static inline int variable_test_bit(int nr, const volatile unsigned long * addr)
  * @addr: The address to start the search at
  * @size: The maximum size to search
  *
- * Returns the bit-number of the first zero bit, not the number of the byte
+ * Returns the bit number of the first zero bit, not the number of the byte
  * containing a bit.
  */
 static inline int find_first_zero_bit(const unsigned long *addr, unsigned size)
@@ -348,7 +348,7 @@ static inline int find_first_zero_bit(const unsigned long *addr, unsigned size)
 /**
  * find_next_zero_bit - find the first zero bit in a memory region
  * @addr: The address to base the search on
- * @offset: The bitnumber to start searching at
+ * @offset: The bit number to start searching at
  * @size: The maximum size to search
  */
 int find_next_zero_bit(const unsigned long *addr, int size, int offset);
@@ -372,7 +372,7 @@ static inline unsigned long __ffs(unsigned long word)
  * @addr: The address to start the search at
  * @size: The maximum size to search
  *
- * Returns the bit-number of the first set bit, not the number of the byte
+ * Returns the bit number of the first set bit, not the number of the byte
  * containing a bit.
  */
 static inline unsigned find_first_bit(const unsigned long *addr, unsigned size)
@@ -391,7 +391,7 @@ static inline unsigned find_first_bit(const unsigned long *addr, unsigned size)
 /**
  * find_next_bit - find the first set bit in a memory region
  * @addr: The address to base the search on
- * @offset: The bitnumber to start searching at
+ * @offset: The bit number to start searching at
  * @size: The maximum size to search
  */
 int find_next_bit(const unsigned long *addr, int size, int offset);
@@ -460,10 +460,10 @@ static inline int fls(int x)
 
 #include <asm-generic/bitops/ext2-non-atomic.h>
 
-#define ext2_set_bit_atomic(lock,nr,addr) \
-        test_and_set_bit((nr),(unsigned long*)addr)
-#define ext2_clear_bit_atomic(lock,nr, addr) \
-	        test_and_clear_bit((nr),(unsigned long*)addr)
+#define ext2_set_bit_atomic(lock, nr, addr) \
+		test_and_set_bit((nr), (unsigned long *)addr)
+#define ext2_clear_bit_atomic(lock, nr, addr) \
+		test_and_clear_bit((nr), (unsigned long *)addr)
 
 #include <asm-generic/bitops/minix.h>
 
-- 
cgit v1.2.3


From 6ea8bad1c06c8ccde381f1c848afe6a245bb33b5 Mon Sep 17 00:00:00 2001
From: "clameter@sgi.com" <clameter@sgi.com>
Date: Wed, 30 Jan 2008 13:30:32 +0100
Subject: x86: clean up stack allocation and free

Clean up the allocation and freeing of stacks a bit by using a __GFP_ZERO flag
instead of memset.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Mike Travis <travis@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/thread_info_64.h | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index 7f6ee68f0002..1ac23c157231 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -74,20 +74,14 @@ static inline struct thread_info *stack_thread_info(void)
 
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
-#define alloc_thread_info(tsk)					\
-    ({								\
-	struct thread_info *ret;				\
-								\
-	ret = ((struct thread_info *) __get_free_pages(GFP_KERNEL,THREAD_ORDER)); \
-	if (ret)						\
-		memset(ret, 0, THREAD_SIZE);			\
-	ret;							\
-    })
+#define THREAD_FLAGS (GFP_KERNEL | __GFP_ZERO)
 #else
-#define alloc_thread_info(tsk) \
-	((struct thread_info *) __get_free_pages(GFP_KERNEL,THREAD_ORDER))
+#define THREAD_FLAGS GFP_KERNEL
 #endif
 
+#define alloc_thread_info(tsk) \
+	((struct thread_info *) __get_free_pages(THREAD_FLAGS, THREAD_ORDER))
+
 #define free_thread_info(ti) free_pages((unsigned long) (ti), THREAD_ORDER)
 
 #else /* !__ASSEMBLY__ */
-- 
cgit v1.2.3


From 6abcd98ffafbff81f0bfd7ee1d129e634af13245 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:30:33 +0100
Subject: x86: irqflags consolidation

This patch consolidates the irqflags include files containing common
paravirt definitions. The native definition for interrupt handling, halt,
and such, are the same for 32 and 64 bit, and they are kept in irqflags.h.
the differences are split in the arch-specific files.

The syscall function, irq_enable_sysexit, has a very specific i386 naming,
and its name is then changed to a more general one.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/asm-offsets_32.c |   2 +-
 arch/x86/kernel/entry_32.S       |   8 +-
 arch/x86/kernel/paravirt_32.c    |  10 +-
 arch/x86/kernel/vmi_32.c         |   4 +-
 arch/x86/xen/enlighten.c         |   2 +-
 include/asm-x86/irqflags.h       | 246 ++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/irqflags_32.h    | 195 -------------------------------
 include/asm-x86/irqflags_64.h    | 174 ---------------------------
 include/asm-x86/paravirt.h       |   9 +-
 9 files changed, 261 insertions(+), 389 deletions(-)
 delete mode 100644 include/asm-x86/irqflags_32.h
 delete mode 100644 include/asm-x86/irqflags_64.h

(limited to 'include')

diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 0e45981b2dd7..c1ccfabb4a9e 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -123,7 +123,7 @@ void foo(void)
 	OFFSET(PV_IRQ_irq_disable, pv_irq_ops, irq_disable);
 	OFFSET(PV_IRQ_irq_enable, pv_irq_ops, irq_enable);
 	OFFSET(PV_CPU_iret, pv_cpu_ops, iret);
-	OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit);
+	OFFSET(PV_CPU_irq_enable_syscall_ret, pv_cpu_ops, irq_enable_syscall_ret);
 	OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0);
 #endif
 
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index dc7f938e5015..d63609dd64b9 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -58,7 +58,7 @@
  * for paravirtualization.  The following will never clobber any registers:
  *   INTERRUPT_RETURN (aka. "iret")
  *   GET_CR0_INTO_EAX (aka. "movl %cr0, %eax")
- *   ENABLE_INTERRUPTS_SYSEXIT (aka "sti; sysexit").
+ *   ENABLE_INTERRUPTS_SYSCALL_RET (aka "sti; sysexit").
  *
  * For DISABLE_INTERRUPTS/ENABLE_INTERRUPTS (aka "cli"/"sti"), you must
  * specify what registers can be overwritten (CLBR_NONE, CLBR_EAX/EDX/ECX/ANY).
@@ -351,7 +351,7 @@ sysenter_past_esp:
 	xorl %ebp,%ebp
 	TRACE_IRQS_ON
 1:	mov  PT_FS(%esp), %fs
-	ENABLE_INTERRUPTS_SYSEXIT
+	ENABLE_INTERRUPTS_SYSCALL_RET
 	CFI_ENDPROC
 .pushsection .fixup,"ax"
 2:	movl $0,PT_FS(%esp)
@@ -882,10 +882,10 @@ ENTRY(native_iret)
 .previous
 END(native_iret)
 
-ENTRY(native_irq_enable_sysexit)
+ENTRY(native_irq_enable_syscall_ret)
 	sti
 	sysexit
-END(native_irq_enable_sysexit)
+END(native_irq_enable_syscall_ret)
 #endif
 
 KPROBE_ENTRY(int3)
diff --git a/arch/x86/kernel/paravirt_32.c b/arch/x86/kernel/paravirt_32.c
index f5000799f8ef..706b0562ea40 100644
--- a/arch/x86/kernel/paravirt_32.c
+++ b/arch/x86/kernel/paravirt_32.c
@@ -60,7 +60,7 @@ DEF_NATIVE(pv_irq_ops, irq_enable, "sti");
 DEF_NATIVE(pv_irq_ops, restore_fl, "push %eax; popf");
 DEF_NATIVE(pv_irq_ops, save_fl, "pushf; pop %eax");
 DEF_NATIVE(pv_cpu_ops, iret, "iret");
-DEF_NATIVE(pv_cpu_ops, irq_enable_sysexit, "sti; sysexit");
+DEF_NATIVE(pv_cpu_ops, irq_enable_syscall_ret, "sti; sysexit");
 DEF_NATIVE(pv_mmu_ops, read_cr2, "mov %cr2, %eax");
 DEF_NATIVE(pv_mmu_ops, write_cr3, "mov %eax, %cr3");
 DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax");
@@ -88,7 +88,7 @@ static unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
 	SITE(pv_irq_ops, restore_fl);
 	SITE(pv_irq_ops, save_fl);
 	SITE(pv_cpu_ops, iret);
-	SITE(pv_cpu_ops, irq_enable_sysexit);
+	SITE(pv_cpu_ops, irq_enable_syscall_ret);
 	SITE(pv_mmu_ops, read_cr2);
 	SITE(pv_mmu_ops, read_cr3);
 	SITE(pv_mmu_ops, write_cr3);
@@ -186,7 +186,7 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
 		/* If the operation is a nop, then nop the callsite */
 		ret = paravirt_patch_nop();
 	else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) ||
-		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit))
+		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_syscall_ret))
 		/* If operation requires a jmp, then jmp */
 		ret = paravirt_patch_jmp(insnbuf, opfunc, addr, len);
 	else
@@ -237,7 +237,7 @@ static void native_flush_tlb_single(unsigned long addr)
 
 /* These are in entry.S */
 extern void native_iret(void);
-extern void native_irq_enable_sysexit(void);
+extern void native_irq_enable_syscall_ret(void);
 
 static int __init print_banner(void)
 {
@@ -384,7 +384,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.write_idt_entry = write_dt_entry,
 	.load_esp0 = native_load_esp0,
 
-	.irq_enable_sysexit = native_irq_enable_sysexit,
+	.irq_enable_syscall_ret = native_irq_enable_syscall_ret,
 	.iret = native_iret,
 
 	.set_iopl_mask = native_set_iopl_mask,
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index f02bad68abaa..aacce426cbd0 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -148,7 +148,7 @@ static unsigned vmi_patch(u8 type, u16 clobbers, void *insns,
 					      insns, eip);
 		case PARAVIRT_PATCH(pv_cpu_ops.iret):
 			return patch_internal(VMI_CALL_IRET, len, insns, eip);
-		case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit):
+		case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_syscall_ret):
 			return patch_internal(VMI_CALL_SYSEXIT, len, insns, eip);
 		default:
 			break;
@@ -870,7 +870,7 @@ static inline int __init activate_vmi(void)
 	 * the backend.  They are performance critical anyway, so requiring
 	 * a patch is not a big problem.
 	 */
-	pv_cpu_ops.irq_enable_sysexit = (void *)0xfeedbab0;
+	pv_cpu_ops.irq_enable_syscall_ret = (void *)0xfeedbab0;
 	pv_cpu_ops.iret = (void *)0xbadbab0;
 
 #ifdef CONFIG_SMP
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 00829401389e..d3574485cb15 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -953,7 +953,7 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.read_pmc = native_read_pmc,
 
 	.iret = (void *)&hypercall_page[__HYPERVISOR_iret],
-	.irq_enable_sysexit = NULL,  /* never called */
+	.irq_enable_syscall_ret = NULL,  /* never called */
 
 	.load_tr_desc = paravirt_nop,
 	.set_ldt = xen_set_ldt,
diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h
index 1b695ff52687..92021c1ffa3a 100644
--- a/include/asm-x86/irqflags.h
+++ b/include/asm-x86/irqflags.h
@@ -1,5 +1,245 @@
-#ifdef CONFIG_X86_32
-# include "irqflags_32.h"
+#ifndef _X86_IRQFLAGS_H_
+#define _X86_IRQFLAGS_H_
+
+#include <asm/processor-flags.h>
+
+#ifndef __ASSEMBLY__
+/*
+ * Interrupt control:
+ */
+
+static inline unsigned long native_save_fl(void)
+{
+	unsigned long flags;
+
+	__asm__ __volatile__(
+		"# __raw_save_flags\n\t"
+		"pushf ; pop %0"
+		: "=g" (flags)
+		: /* no input */
+		: "memory"
+	);
+
+	return flags;
+}
+
+static inline void native_restore_fl(unsigned long flags)
+{
+	__asm__ __volatile__(
+		"push %0 ; popf"
+		: /* no output */
+		:"g" (flags)
+		:"memory", "cc"
+	);
+}
+
+static inline void native_irq_disable(void)
+{
+	asm volatile("cli": : :"memory");
+}
+
+static inline void native_irq_enable(void)
+{
+	asm volatile("sti": : :"memory");
+}
+
+static inline void native_safe_halt(void)
+{
+	asm volatile("sti; hlt": : :"memory");
+}
+
+static inline void native_halt(void)
+{
+	asm volatile("hlt": : :"memory");
+}
+
+#endif
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#ifndef __ASSEMBLY__
+
+static inline unsigned long __raw_local_save_flags(void)
+{
+	return native_save_fl();
+}
+
+static inline void raw_local_irq_restore(unsigned long flags)
+{
+	native_restore_fl(flags);
+}
+
+static inline void raw_local_irq_disable(void)
+{
+	native_irq_disable();
+}
+
+static inline void raw_local_irq_enable(void)
+{
+	native_irq_enable();
+}
+
+/*
+ * Used in the idle loop; sti takes one instruction cycle
+ * to complete:
+ */
+static inline void raw_safe_halt(void)
+{
+	native_safe_halt();
+}
+
+/*
+ * Used when interrupts are already enabled or to
+ * shutdown the processor:
+ */
+static inline void halt(void)
+{
+	native_halt();
+}
+
+/*
+ * For spinlocks, etc:
+ */
+static inline unsigned long __raw_local_irq_save(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	raw_local_irq_disable();
+
+	return flags;
+}
+#else
+
+#define ENABLE_INTERRUPTS(x)	sti
+#define DISABLE_INTERRUPTS(x)	cli
+
+#ifdef CONFIG_X86_64
+#define INTERRUPT_RETURN	iretq
+#define ENABLE_INTERRUPTS_SYSCALL_RET			\
+			movq	%gs:pda_oldrsp, %rsp;	\
+			swapgs;				\
+			sysretq;
+#else
+#define INTERRUPT_RETURN		iret
+#define ENABLE_INTERRUPTS_SYSCALL_RET	sti; sysexit
+#define GET_CR0_INTO_EAX		movl %cr0, %eax
+#endif
+
+
+#endif /* __ASSEMBLY__ */
+#endif /* CONFIG_PARAVIRT */
+
+#ifndef __ASSEMBLY__
+#define raw_local_save_flags(flags) \
+		do { (flags) = __raw_local_save_flags(); } while (0)
+
+#define raw_local_irq_save(flags) \
+		do { (flags) = __raw_local_irq_save(); } while (0)
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+	return !(flags & X86_EFLAGS_IF);
+}
+
+static inline int raw_irqs_disabled(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	return raw_irqs_disabled_flags(flags);
+}
+
+/*
+ * makes the traced hardirq state match with the machine state
+ *
+ * should be a rarely used function, only in places where its
+ * otherwise impossible to know the irq state, like in traps.
+ */
+static inline void trace_hardirqs_fixup_flags(unsigned long flags)
+{
+	if (raw_irqs_disabled_flags(flags))
+		trace_hardirqs_off();
+	else
+		trace_hardirqs_on();
+}
+
+static inline void trace_hardirqs_fixup(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	trace_hardirqs_fixup_flags(flags);
+}
+
 #else
-# include "irqflags_64.h"
+
+#ifdef CONFIG_X86_64
+/*
+ * Currently paravirt can't handle swapgs nicely when we
+ * don't have a stack we can rely on (such as a user space
+ * stack).  So we either find a way around these or just fault
+ * and emulate if a guest tries to call swapgs directly.
+ *
+ * Either way, this is a good way to document that we don't
+ * have a reliable stack. x86_64 only.
+ */
+#define SWAPGS_UNSAFE_STACK	swapgs
+#define ARCH_TRACE_IRQS_ON		call trace_hardirqs_on_thunk
+#define ARCH_TRACE_IRQS_OFF		call trace_hardirqs_off_thunk
+#define ARCH_LOCKDEP_SYS_EXIT		call lockdep_sys_exit_thunk
+#define ARCH_LOCKDEP_SYS_EXIT_IRQ	\
+	TRACE_IRQS_ON; \
+	sti; \
+	SAVE_REST; \
+	LOCKDEP_SYS_EXIT; \
+	RESTORE_REST; \
+	cli; \
+	TRACE_IRQS_OFF;
+
+#else
+#define ARCH_TRACE_IRQS_ON			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call trace_hardirqs_on;			\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+
+#define ARCH_TRACE_IRQS_OFF			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call trace_hardirqs_off;		\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+
+#define ARCH_LOCKDEP_SYS_EXIT			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call lockdep_sys_exit;			\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+
+#define ARCH_LOCKDEP_SYS_EXIT_IRQ
+#endif
+
+#ifdef CONFIG_TRACE_IRQFLAGS
+#  define TRACE_IRQS_ON		ARCH_TRACE_IRQS_ON
+#  define TRACE_IRQS_OFF	ARCH_TRACE_IRQS_OFF
+#else
+#  define TRACE_IRQS_ON
+#  define TRACE_IRQS_OFF
+#endif
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+#  define LOCKDEP_SYS_EXIT	ARCH_LOCKDEP_SYS_EXIT
+#  define LOCKDEP_SYS_EXIT_IRQ	ARCH_LOCKDEP_SYS_EXIT_IRQ
+# else
+#  define LOCKDEP_SYS_EXIT
+#  define LOCKDEP_SYS_EXIT_IRQ
+# endif
+
+#endif /* __ASSEMBLY__ */
 #endif
diff --git a/include/asm-x86/irqflags_32.h b/include/asm-x86/irqflags_32.h
deleted file mode 100644
index 98b21b9bdce8..000000000000
--- a/include/asm-x86/irqflags_32.h
+++ /dev/null
@@ -1,195 +0,0 @@
-/*
- * IRQ flags handling
- *
- * This file gets included from lowlevel asm headers too, to provide
- * wrapped versions of the local_irq_*() APIs, based on the
- * raw_local_irq_*() functions from the lowlevel headers.
- */
-#ifndef _ASM_IRQFLAGS_H
-#define _ASM_IRQFLAGS_H
-#include <asm/processor-flags.h>
-
-#ifndef __ASSEMBLY__
-static inline unsigned long native_save_fl(void)
-{
-	unsigned long f;
-	asm volatile("pushfl ; popl %0":"=g" (f): /* no input */);
-	return f;
-}
-
-static inline void native_restore_fl(unsigned long f)
-{
-	asm volatile("pushl %0 ; popfl": /* no output */
-			     :"g" (f)
-			     :"memory", "cc");
-}
-
-static inline void native_irq_disable(void)
-{
-	asm volatile("cli": : :"memory");
-}
-
-static inline void native_irq_enable(void)
-{
-	asm volatile("sti": : :"memory");
-}
-
-static inline void native_safe_halt(void)
-{
-	asm volatile("sti; hlt": : :"memory");
-}
-
-static inline void native_halt(void)
-{
-	asm volatile("hlt": : :"memory");
-}
-#endif	/* __ASSEMBLY__ */
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#ifndef __ASSEMBLY__
-
-static inline unsigned long __raw_local_save_flags(void)
-{
-	return native_save_fl();
-}
-
-static inline void raw_local_irq_restore(unsigned long flags)
-{
-	native_restore_fl(flags);
-}
-
-static inline void raw_local_irq_disable(void)
-{
-	native_irq_disable();
-}
-
-static inline void raw_local_irq_enable(void)
-{
-	native_irq_enable();
-}
-
-/*
- * Used in the idle loop; sti takes one instruction cycle
- * to complete:
- */
-static inline void raw_safe_halt(void)
-{
-	native_safe_halt();
-}
-
-/*
- * Used when interrupts are already enabled or to
- * shutdown the processor:
- */
-static inline void halt(void)
-{
-	native_halt();
-}
-
-/*
- * For spinlocks, etc:
- */
-static inline unsigned long __raw_local_irq_save(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	raw_local_irq_disable();
-
-	return flags;
-}
-
-#else
-#define DISABLE_INTERRUPTS(clobbers)	cli
-#define ENABLE_INTERRUPTS(clobbers)	sti
-#define ENABLE_INTERRUPTS_SYSEXIT	sti; sysexit
-#define INTERRUPT_RETURN		iret
-#define GET_CR0_INTO_EAX		movl %cr0, %eax
-#endif /* __ASSEMBLY__ */
-#endif /* CONFIG_PARAVIRT */
-
-#ifndef __ASSEMBLY__
-#define raw_local_save_flags(flags) \
-		do { (flags) = __raw_local_save_flags(); } while (0)
-
-#define raw_local_irq_save(flags) \
-		do { (flags) = __raw_local_irq_save(); } while (0)
-
-static inline int raw_irqs_disabled_flags(unsigned long flags)
-{
-	return !(flags & X86_EFLAGS_IF);
-}
-
-static inline int raw_irqs_disabled(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	return raw_irqs_disabled_flags(flags);
-}
-
-/*
- * makes the traced hardirq state match with the machine state
- *
- * should be a rarely used function, only in places where its
- * otherwise impossible to know the irq state, like in traps.
- */
-static inline void trace_hardirqs_fixup_flags(unsigned long flags)
-{
-	if (raw_irqs_disabled_flags(flags))
-		trace_hardirqs_off();
-	else
-		trace_hardirqs_on();
-}
-
-static inline void trace_hardirqs_fixup(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	trace_hardirqs_fixup_flags(flags);
-}
-#endif /* __ASSEMBLY__ */
-
-/*
- * Do the CPU's IRQ-state tracing from assembly code. We call a
- * C function, so save all the C-clobbered registers:
- */
-#ifdef CONFIG_TRACE_IRQFLAGS
-
-# define TRACE_IRQS_ON				\
-	pushl %eax;				\
-	pushl %ecx;				\
-	pushl %edx;				\
-	call trace_hardirqs_on;			\
-	popl %edx;				\
-	popl %ecx;				\
-	popl %eax;
-
-# define TRACE_IRQS_OFF				\
-	pushl %eax;				\
-	pushl %ecx;				\
-	pushl %edx;				\
-	call trace_hardirqs_off;		\
-	popl %edx;				\
-	popl %ecx;				\
-	popl %eax;
-
-#else
-# define TRACE_IRQS_ON
-# define TRACE_IRQS_OFF
-#endif
-
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-# define LOCKDEP_SYS_EXIT			\
-	pushl %eax;				\
-	pushl %ecx;				\
-	pushl %edx;				\
-	call lockdep_sys_exit;			\
-	popl %edx;				\
-	popl %ecx;				\
-	popl %eax;
-#else
-# define LOCKDEP_SYS_EXIT
-#endif
-
-#endif
diff --git a/include/asm-x86/irqflags_64.h b/include/asm-x86/irqflags_64.h
deleted file mode 100644
index 38c07db733cf..000000000000
--- a/include/asm-x86/irqflags_64.h
+++ /dev/null
@@ -1,174 +0,0 @@
-/*
- * IRQ flags handling
- *
- * This file gets included from lowlevel asm headers too, to provide
- * wrapped versions of the local_irq_*() APIs, based on the
- * raw_local_irq_*() functions from the lowlevel headers.
- */
-#ifndef _ASM_IRQFLAGS_H
-#define _ASM_IRQFLAGS_H
-#include <asm/processor-flags.h>
-
-#ifndef __ASSEMBLY__
-/*
- * Interrupt control:
- */
-
-static inline unsigned long __raw_local_save_flags(void)
-{
-	unsigned long flags;
-
-	__asm__ __volatile__(
-		"# __raw_save_flags\n\t"
-		"pushfq ; popq %q0"
-		: "=g" (flags)
-		: /* no input */
-		: "memory"
-	);
-
-	return flags;
-}
-
-#define raw_local_save_flags(flags) \
-		do { (flags) = __raw_local_save_flags(); } while (0)
-
-static inline void raw_local_irq_restore(unsigned long flags)
-{
-	__asm__ __volatile__(
-		"pushq %0 ; popfq"
-		: /* no output */
-		:"g" (flags)
-		:"memory", "cc"
-	);
-}
-
-#ifdef CONFIG_X86_VSMP
-
-/*
- * Interrupt control for the VSMP architecture:
- */
-
-static inline void raw_local_irq_disable(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	raw_local_irq_restore((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC);
-}
-
-static inline void raw_local_irq_enable(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	raw_local_irq_restore((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC));
-}
-
-static inline int raw_irqs_disabled_flags(unsigned long flags)
-{
-	return !(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC);
-}
-
-#else /* CONFIG_X86_VSMP */
-
-static inline void raw_local_irq_disable(void)
-{
-	__asm__ __volatile__("cli" : : : "memory");
-}
-
-static inline void raw_local_irq_enable(void)
-{
-	__asm__ __volatile__("sti" : : : "memory");
-}
-
-static inline int raw_irqs_disabled_flags(unsigned long flags)
-{
-	return !(flags & X86_EFLAGS_IF);
-}
-
-#endif
-
-/*
- * For spinlocks, etc.:
- */
-
-static inline unsigned long __raw_local_irq_save(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	raw_local_irq_disable();
-
-	return flags;
-}
-
-#define raw_local_irq_save(flags) \
-		do { (flags) = __raw_local_irq_save(); } while (0)
-
-static inline int raw_irqs_disabled(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	return raw_irqs_disabled_flags(flags);
-}
-
-/*
- * makes the traced hardirq state match with the machine state
- *
- * should be a rarely used function, only in places where its
- * otherwise impossible to know the irq state, like in traps.
- */
-static inline void trace_hardirqs_fixup_flags(unsigned long flags)
-{
-	if (raw_irqs_disabled_flags(flags))
-		trace_hardirqs_off();
-	else
-		trace_hardirqs_on();
-}
-
-static inline void trace_hardirqs_fixup(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	trace_hardirqs_fixup_flags(flags);
-}
-/*
- * Used in the idle loop; sti takes one instruction cycle
- * to complete:
- */
-static inline void raw_safe_halt(void)
-{
-	__asm__ __volatile__("sti; hlt" : : : "memory");
-}
-
-/*
- * Used when interrupts are already enabled or to
- * shutdown the processor:
- */
-static inline void halt(void)
-{
-	__asm__ __volatile__("hlt": : :"memory");
-}
-
-#else /* __ASSEMBLY__: */
-# ifdef CONFIG_TRACE_IRQFLAGS
-#  define TRACE_IRQS_ON		call trace_hardirqs_on_thunk
-#  define TRACE_IRQS_OFF	call trace_hardirqs_off_thunk
-# else
-#  define TRACE_IRQS_ON
-#  define TRACE_IRQS_OFF
-# endif
-# ifdef CONFIG_DEBUG_LOCK_ALLOC
-#  define LOCKDEP_SYS_EXIT	call lockdep_sys_exit_thunk
-#  define LOCKDEP_SYS_EXIT_IRQ	\
-	TRACE_IRQS_ON; \
-	sti; \
-	SAVE_REST; \
-	LOCKDEP_SYS_EXIT; \
-	RESTORE_REST; \
-	cli; \
-	TRACE_IRQS_OFF;
-# else
-#  define LOCKDEP_SYS_EXIT
-#  define LOCKDEP_SYS_EXIT_IRQ
-# endif
-#endif
-
-#endif
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 19fd3e67b08c..be7b934f6c54 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -121,7 +121,7 @@ struct pv_cpu_ops {
 	u64 (*read_pmc)(void);
 
 	/* These two are jmp to, not actually called. */
-	void (*irq_enable_sysexit)(void);
+	void (*irq_enable_syscall_ret)(void);
 	void (*iret)(void);
 
 	struct pv_lazy_ops lazy_mode;
@@ -1138,9 +1138,10 @@ static inline unsigned long __raw_local_irq_save(void)
 		  call *%cs:pv_irq_ops+PV_IRQ_irq_enable;		\
 		  popl %edx; popl %ecx; popl %eax)
 
-#define ENABLE_INTERRUPTS_SYSEXIT					       \
-	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_sysexit), CLBR_NONE,\
-		  jmp *%cs:pv_cpu_ops+PV_CPU_irq_enable_sysexit)
+#define ENABLE_INTERRUPTS_SYSCALL_RET					\
+	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_syscall_ret),\
+		  CLBR_NONE,						\
+		  jmp *%cs:pv_cpu_ops+PV_CPU_irq_enable_syscall_ret)
 
 #define GET_CR0_INTO_EAX			\
 	push %ecx; push %edx;			\
-- 
cgit v1.2.3


From 2fed0c507cf0101d511366f36e8573f403dbfea5 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:30:33 +0100
Subject: x86: consolidate spinlock.h

The cli and sti instructions need to be replaced by paravirt hooks.
For the i386 architecture, this is already done. The code requirements
aren't much different from x86_64 POV, so this part is consolidated in
the common header

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/spinlock.h    | 14 +++++++++
 include/asm-x86/spinlock_32.h | 71 ++++++++++++++++---------------------------
 include/asm-x86/spinlock_64.h | 37 ++++++++++++++--------
 3 files changed, 64 insertions(+), 58 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index d74d85e71dcb..e1d555a3dfe5 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -1,5 +1,19 @@
+#ifndef _X86_SPINLOCK_H_
+#define _X86_SPINLOCK_H_
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define CLI_STRING	"cli"
+#define STI_STRING	"sti"
+#define CLI_STI_CLOBBERS
+#define CLI_STI_INPUT_ARGS
+#endif /* CONFIG_PARAVIRT */
+
 #ifdef CONFIG_X86_32
 # include "spinlock_32.h"
 #else
 # include "spinlock_64.h"
 #endif
+
+#endif
diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
index d3bcebed60ca..c42c3f12d7ce 100644
--- a/include/asm-x86/spinlock_32.h
+++ b/include/asm-x86/spinlock_32.h
@@ -5,16 +5,6 @@
 #include <asm/rwlock.h>
 #include <asm/page.h>
 #include <asm/processor.h>
-#include <linux/compiler.h>
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define CLI_STRING	"cli"
-#define STI_STRING	"sti"
-#define CLI_STI_CLOBBERS
-#define CLI_STI_INPUT_ARGS
-#endif /* CONFIG_PARAVIRT */
 
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
@@ -27,23 +17,24 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-static inline int __raw_spin_is_locked(raw_spinlock_t *x)
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
 {
-	return *(volatile signed char *)(&(x)->slock) <= 0;
+	return *(volatile signed char *)(&(lock)->slock) <= 0;
 }
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm volatile("\n1:\t"
-		     LOCK_PREFIX " ; decb %0\n\t"
-		     "jns 3f\n"
-		     "2:\t"
-		     "rep;nop\n\t"
-		     "cmpb $0,%0\n\t"
-		     "jle 2b\n\t"
-		     "jmp 1b\n"
-		     "3:\n\t"
-		     : "+m" (lock->slock) : : "memory");
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; decb %0\n\t"
+		"jns 3f\n"
+		"2:\t"
+		"rep;nop\n\t"
+		"cmpb $0,%0\n\t"
+		"jle 2b\n\t"
+		"jmp 1b\n"
+		"3:\n\t"
+		: "+m" (lock->slock) : : "memory");
 }
 
 /*
@@ -55,7 +46,8 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
  * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
  */
 #ifndef CONFIG_PROVE_LOCKING
-static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
+static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
+					 unsigned long flags)
 {
 	asm volatile(
 		"\n1:\t"
@@ -79,18 +71,20 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla
 		"5:\n\t"
 		: [slock] "+m" (lock->slock)
 		: [flags] "r" (flags)
-	 	  CLI_STI_INPUT_ARGS
+		  CLI_STI_INPUT_ARGS
 		: "memory" CLI_STI_CLOBBERS);
 }
 #endif
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
-	char oldval;
+	signed char oldval;
+
 	asm volatile(
 		"xchgb %b0,%1"
 		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
+
 	return oldval > 0;
 }
 
@@ -112,7 +106,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	char oldval = 1;
+	unsigned char oldval = 1;
 
 	asm volatile("xchgb %b0, %1"
 		     : "=q" (oldval), "+m" (lock->slock)
@@ -139,31 +133,16 @@ static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
  *
  * On x86, we implement read-write locks as a 32-bit counter
  * with the high bit (sign) being the "contended" bit.
- *
- * The inline assembly is non-obvious. Think about it.
- *
- * Changed to use the same technique as rw semaphores.  See
- * semaphore.h for details.  -ben
- *
- * the helpers are in arch/i386/kernel/semaphore.c
  */
 
-/**
- * read_can_lock - would read_trylock() succeed?
- * @lock: the rwlock in question.
- */
-static inline int __raw_read_can_lock(raw_rwlock_t *x)
+static inline int __raw_read_can_lock(raw_rwlock_t *lock)
 {
-	return (int)(x)->lock > 0;
+	return (int)(lock)->lock > 0;
 }
 
-/**
- * write_can_lock - would write_trylock() succeed?
- * @lock: the rwlock in question.
- */
-static inline int __raw_write_can_lock(raw_rwlock_t *x)
+static inline int __raw_write_can_lock(raw_rwlock_t *lock)
 {
-	return (x)->lock == RW_LOCK_BIAS;
+	return (lock)->lock == RW_LOCK_BIAS;
 }
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
@@ -187,6 +166,7 @@ static inline void __raw_write_lock(raw_rwlock_t *rw)
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
 {
 	atomic_t *count = (atomic_t *)lock;
+
 	atomic_dec(count);
 	if (atomic_read(count) >= 0)
 		return 1;
@@ -197,6 +177,7 @@ static inline int __raw_read_trylock(raw_rwlock_t *lock)
 static inline int __raw_write_trylock(raw_rwlock_t *lock)
 {
 	atomic_t *count = (atomic_t *)lock;
+
 	if (atomic_sub_and_test(RW_LOCK_BIAS, count))
 		return 1;
 	atomic_add(RW_LOCK_BIAS, count);
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
index 88bf981e73cf..3b5adf92ad08 100644
--- a/include/asm-x86/spinlock_64.h
+++ b/include/asm-x86/spinlock_64.h
@@ -33,14 +33,21 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 		"cmpl $0,%0\n\t"
 		"jle 3b\n\t"
 		"jmp 1b\n"
-		"2:\t" : "=m" (lock->slock) : : "memory");
+		"2:\t"
+		: "=m" (lock->slock) : : "memory");
 }
 
 /*
- * Same as __raw_spin_lock, but reenable interrupts during spinning.
+ * It is easier for the lock validator if interrupts are not re-enabled
+ * in the middle of a lock-acquire. This is a performance feature anyway
+ * so we turn it off:
+ *
+ * NOTE: there's an irqs-on section here, which normally would have to be
+ * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
  */
 #ifndef CONFIG_PROVE_LOCKING
-static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
+static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
+					 unsigned long flags)
 {
 	asm volatile(
 		"\n1:\t"
@@ -48,12 +55,12 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla
 		"jns 5f\n"
 		"testl $0x200, %1\n\t"	/* interrupts were disabled? */
 		"jz 4f\n\t"
-	        "sti\n"
+		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
 		"cmpl $0, %0\n\t"
 		"jle 3b\n\t"
-		"cli\n\t"
+		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
@@ -61,7 +68,9 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla
 		"jg 1b\n\t"
 		"jmp 4b\n"
 		"5:\n\t"
-		: "+m" (lock->slock) : "r" ((unsigned)flags) : "memory");
+		: "+m" (lock->slock)
+		: "r" ((unsigned)flags) CLI_STI_INPUT_ARGS
+		: "memory" CLI_STI_CLOBBERS);
 }
 #endif
 
@@ -79,7 +88,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	asm volatile("movl $1,%0" :"=m" (lock->slock) :: "memory");
+	asm volatile("movl $1,%0" : "=m" (lock->slock) :: "memory");
 }
 
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
@@ -114,18 +123,18 @@ static inline int __raw_write_can_lock(raw_rwlock_t *lock)
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t"
+	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
 		     "jns 1f\n"
-		     "call __read_lock_failed\n"
+		     "call __read_lock_failed\n\t"
 		     "1:\n"
 		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t"
+	asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t"
 		     "jz 1f\n"
-		     "\tcall __write_lock_failed\n\t"
+		     "call __write_lock_failed\n\t"
 		     "1:\n"
 		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
@@ -133,6 +142,7 @@ static inline void __raw_write_lock(raw_rwlock_t *rw)
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
 {
 	atomic_t *count = (atomic_t *)lock;
+
 	atomic_dec(count);
 	if (atomic_read(count) >= 0)
 		return 1;
@@ -143,6 +153,7 @@ static inline int __raw_read_trylock(raw_rwlock_t *lock)
 static inline int __raw_write_trylock(raw_rwlock_t *lock)
 {
 	atomic_t *count = (atomic_t *)lock;
+
 	if (atomic_sub_and_test(RW_LOCK_BIAS, count))
 		return 1;
 	atomic_add(RW_LOCK_BIAS, count);
@@ -151,12 +162,12 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock)
 
 static inline void __raw_read_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX " ; incl %0" :"=m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "incl %0" :"=m" (rw->lock) : : "memory");
 }
 
 static inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX " ; addl $" RW_LOCK_BIAS_STR ",%0"
+	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ",%0"
 				: "=m" (rw->lock) : : "memory");
 }
 
-- 
cgit v1.2.3


From a33fff3a033f2e8a930067ad608c21e1f86bffab Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:34 +0100
Subject: x86: fix asm constraints in spinlock_32/64.h

Use the correct constraints for the spinlock assembler functions.

read (modify) write functions need "+m" instead of "=m"

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock_32.h |  2 +-
 include/asm-x86/spinlock_64.h | 10 +++++-----
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
index c42c3f12d7ce..fca124a1103d 100644
--- a/include/asm-x86/spinlock_32.h
+++ b/include/asm-x86/spinlock_32.h
@@ -99,7 +99,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	asm volatile("movb $1,%0" : "+m" (lock->slock) :: "memory");
+	asm volatile("movb $1,%0" : "=m" (lock->slock) :: "memory");
 }
 
 #else
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
index 3b5adf92ad08..e81f6c18d877 100644
--- a/include/asm-x86/spinlock_64.h
+++ b/include/asm-x86/spinlock_64.h
@@ -34,7 +34,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 		"jle 3b\n\t"
 		"jmp 1b\n"
 		"2:\t"
-		: "=m" (lock->slock) : : "memory");
+		: "+m" (lock->slock) : : "memory");
 }
 
 /*
@@ -80,7 +80,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 	asm volatile(
 		"xchgl %0,%1"
-		:"=q" (oldval), "=m" (lock->slock)
+		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
 
 	return oldval > 0;
@@ -162,13 +162,13 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock)
 
 static inline void __raw_read_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "incl %0" :"=m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory");
 }
 
 static inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ",%0"
-				: "=m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
+				: "+m" (rw->lock) : : "memory");
 }
 
 #define _raw_spin_relax(lock)	cpu_relax()
-- 
cgit v1.2.3


From 6514f93a2ce643ef5914eae7ce49b978e1d356aa Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:34 +0100
Subject: x86: use immediates instead of RW_LOCK_BIAS_STR

Use immediate instead of the RW_LOCK_BIAS_STR.
Makes the code more readable and gets rid of the string constant.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/rwlock.h      | 1 -
 include/asm-x86/spinlock_32.h | 8 ++++----
 include/asm-x86/spinlock_64.h | 6 +++---
 3 files changed, 7 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/rwlock.h b/include/asm-x86/rwlock.h
index f2b64a429e6b..6a8c0d645108 100644
--- a/include/asm-x86/rwlock.h
+++ b/include/asm-x86/rwlock.h
@@ -2,7 +2,6 @@
 #define _ASM_X86_RWLOCK_H
 
 #define RW_LOCK_BIAS		 0x01000000
-#define RW_LOCK_BIAS_STR	"0x01000000"
 
 /* Actual code is in asm/spinlock.h or in arch/x86/lib/rwlock.S */
 
diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
index fca124a1103d..e7a14ab906e9 100644
--- a/include/asm-x86/spinlock_32.h
+++ b/include/asm-x86/spinlock_32.h
@@ -156,11 +156,11 @@ static inline void __raw_read_lock(raw_rwlock_t *rw)
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t"
+	asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t"
 		     "jz 1f\n"
 		     "call __write_lock_failed\n\t"
 		     "1:\n"
-		     ::"a" (rw) : "memory");
+		     ::"a" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
@@ -191,8 +191,8 @@ static inline void __raw_read_unlock(raw_rwlock_t *rw)
 
 static inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
-				 : "+m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "addl %1, %0"
+		     : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory");
 }
 
 #define _raw_spin_relax(lock)	cpu_relax()
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
index e81f6c18d877..ab50e7f51058 100644
--- a/include/asm-x86/spinlock_64.h
+++ b/include/asm-x86/spinlock_64.h
@@ -127,7 +127,7 @@ static inline void __raw_read_lock(raw_rwlock_t *rw)
 		     "jns 1f\n"
 		     "call __read_lock_failed\n\t"
 		     "1:\n"
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
+		     ::"D" (rw) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
@@ -167,8 +167,8 @@ static inline void __raw_read_unlock(raw_rwlock_t *rw)
 
 static inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
-				: "+m" (rw->lock) : : "memory");
+	asm volatile(LOCK_PREFIX "addl %1, %0"
+		     : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory");
 }
 
 #define _raw_spin_relax(lock)	cpu_relax()
-- 
cgit v1.2.3


From 3b23ed84ec34f04f54f7d5b1e35f258d64a7a5fb Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:34 +0100
Subject: x86: spinlock_32/64 match the jump labels and symbols

Match the jump labels in  the 32/64 variants and switch the
64bit version to symbols, so the functions are almost identical
except for the operand size now.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock_32.h |  1 -
 include/asm-x86/spinlock_64.h | 21 +++++++++++----------
 2 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
index e7a14ab906e9..4ef626d53682 100644
--- a/include/asm-x86/spinlock_32.h
+++ b/include/asm-x86/spinlock_32.h
@@ -53,7 +53,6 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 		"\n1:\t"
 		LOCK_PREFIX " ; decb %[slock]\n\t"
 		"jns 5f\n"
-		"2:\t"
 		"testl $0x200, %[flags]\n\t"
 		"jz 4f\n\t"
 		STI_STRING "\n"
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
index ab50e7f51058..19d483c80185 100644
--- a/include/asm-x86/spinlock_64.h
+++ b/include/asm-x86/spinlock_64.h
@@ -27,13 +27,13 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 	asm volatile(
 		"\n1:\t"
 		LOCK_PREFIX " ; decl %0\n\t"
-		"jns 2f\n"
-		"3:\n"
+		"jns 3f\n"
+		"2:\t"
 		"rep;nop\n\t"
 		"cmpl $0,%0\n\t"
-		"jle 3b\n\t"
+		"jle 2b\n\t"
 		"jmp 1b\n"
-		"2:\t"
+		"3:\n\t"
 		: "+m" (lock->slock) : : "memory");
 }
 
@@ -51,25 +51,26 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 {
 	asm volatile(
 		"\n1:\t"
-		LOCK_PREFIX " ; decl %0\n\t"
+		LOCK_PREFIX " ; decl %[slock]\n\t"
 		"jns 5f\n"
-		"testl $0x200, %1\n\t"	/* interrupts were disabled? */
+		"testl $0x200, %[flags]\n\t"
 		"jz 4f\n\t"
 		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
-		"cmpl $0, %0\n\t"
+		"cmpl $0, %[slock]\n\t"
 		"jle 3b\n\t"
 		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
-		"cmpl $0, %0\n\t"
+		"cmpl $0, %[slock]\n\t"
 		"jg 1b\n\t"
 		"jmp 4b\n"
 		"5:\n\t"
-		: "+m" (lock->slock)
-		: "r" ((unsigned)flags) CLI_STI_INPUT_ARGS
+		: [slock] "+m" (lock->slock)
+		: [flags] "r" ((unsigned)flags)
+		  CLI_STI_INPUT_ARGS
 		: "memory" CLI_STI_CLOBBERS);
 }
 #endif
-- 
cgit v1.2.3


From cf244e30f5b50763cbe85f7de30923d12999e38d Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:34 +0100
Subject: x86: spinlock_32/64 substitute types and instructions

Use _slock_t for the spinlock data types and replace the instructions
by string defines, which makes the code of 32/64 bit versions more
or less identical.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock_32.h | 31 +++++++++++++++++++------------
 include/asm-x86/spinlock_64.h | 31 +++++++++++++++++++------------
 2 files changed, 38 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
index 4ef626d53682..2de9b8b89903 100644
--- a/include/asm-x86/spinlock_32.h
+++ b/include/asm-x86/spinlock_32.h
@@ -17,20 +17,27 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
+typedef char _slock_t;
+#define LOCK_INS_DEC "decb"
+#define LOCK_INS_XCH "xchgb"
+#define LOCK_INS_MOV "movb"
+#define LOCK_INS_CMP "cmpb"
+#define LOCK_PTR_REG "a"
+
 static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
 {
-	return *(volatile signed char *)(&(lock)->slock) <= 0;
+	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
 }
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	asm volatile(
 		"\n1:\t"
-		LOCK_PREFIX " ; decb %0\n\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
 		"jns 3f\n"
 		"2:\t"
 		"rep;nop\n\t"
-		"cmpb $0,%0\n\t"
+		LOCK_INS_CMP " $0,%0\n\t"
 		"jle 2b\n\t"
 		"jmp 1b\n"
 		"3:\n\t"
@@ -51,25 +58,25 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 {
 	asm volatile(
 		"\n1:\t"
-		LOCK_PREFIX " ; decb %[slock]\n\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
 		"jns 5f\n"
 		"testl $0x200, %[flags]\n\t"
 		"jz 4f\n\t"
 		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
-		"cmpb $0, %[slock]\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
 		"jle 3b\n\t"
 		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
-		"cmpb $0, %[slock]\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
 		"jg 1b\n\t"
 		"jmp 4b\n"
 		"5:\n\t"
 		: [slock] "+m" (lock->slock)
-		: [flags] "r" (flags)
+		: [flags] "r" ((u32)flags)
 		  CLI_STI_INPUT_ARGS
 		: "memory" CLI_STI_CLOBBERS);
 }
@@ -77,10 +84,10 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
-	signed char oldval;
+	_slock_t oldval;
 
 	asm volatile(
-		"xchgb %b0,%1"
+		LOCK_INS_XCH " %0,%1"
 		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
 
@@ -98,7 +105,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	asm volatile("movb $1,%0" : "=m" (lock->slock) :: "memory");
+	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
 }
 
 #else
@@ -150,7 +157,7 @@ static inline void __raw_read_lock(raw_rwlock_t *rw)
 		     "jns 1f\n"
 		     "call __read_lock_failed\n\t"
 		     "1:\n"
-		     ::"a" (rw) : "memory");
+		     ::LOCK_PTR_REG (rw) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
@@ -159,7 +166,7 @@ static inline void __raw_write_lock(raw_rwlock_t *rw)
 		     "jz 1f\n"
 		     "call __write_lock_failed\n\t"
 		     "1:\n"
-		     ::"a" (rw), "i" (RW_LOCK_BIAS) : "memory");
+		     ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
index 19d483c80185..f5ba90b7335c 100644
--- a/include/asm-x86/spinlock_64.h
+++ b/include/asm-x86/spinlock_64.h
@@ -17,20 +17,27 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
+typedef int _slock_t;
+#define LOCK_INS_DEC "decl"
+#define LOCK_INS_XCH "xchgl"
+#define LOCK_INS_MOV "movl"
+#define LOCK_INS_CMP "cmpl"
+#define LOCK_PTR_REG "D"
+
 static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
 {
-	return *(volatile signed int *)(&(lock)->slock) <= 0;
+	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
 }
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	asm volatile(
 		"\n1:\t"
-		LOCK_PREFIX " ; decl %0\n\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
 		"jns 3f\n"
 		"2:\t"
 		"rep;nop\n\t"
-		"cmpl $0,%0\n\t"
+		LOCK_INS_CMP " $0,%0\n\t"
 		"jle 2b\n\t"
 		"jmp 1b\n"
 		"3:\n\t"
@@ -51,25 +58,25 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 {
 	asm volatile(
 		"\n1:\t"
-		LOCK_PREFIX " ; decl %[slock]\n\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
 		"jns 5f\n"
 		"testl $0x200, %[flags]\n\t"
 		"jz 4f\n\t"
 		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
-		"cmpl $0, %[slock]\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
 		"jle 3b\n\t"
 		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
-		"cmpl $0, %[slock]\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
 		"jg 1b\n\t"
 		"jmp 4b\n"
 		"5:\n\t"
 		: [slock] "+m" (lock->slock)
-		: [flags] "r" ((unsigned)flags)
+		: [flags] "r" ((u32)flags)
 		  CLI_STI_INPUT_ARGS
 		: "memory" CLI_STI_CLOBBERS);
 }
@@ -77,10 +84,10 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
-	int oldval;
+	_slock_t oldval;
 
 	asm volatile(
-		"xchgl %0,%1"
+		LOCK_INS_XCH " %0,%1"
 		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
 
@@ -89,7 +96,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	asm volatile("movl $1,%0" : "=m" (lock->slock) :: "memory");
+	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
 }
 
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
@@ -128,7 +135,7 @@ static inline void __raw_read_lock(raw_rwlock_t *rw)
 		     "jns 1f\n"
 		     "call __read_lock_failed\n\t"
 		     "1:\n"
-		     ::"D" (rw) : "memory");
+		     ::LOCK_PTR_REG (rw) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
@@ -137,7 +144,7 @@ static inline void __raw_write_lock(raw_rwlock_t *rw)
 		     "jz 1f\n"
 		     "call __write_lock_failed\n\t"
 		     "1:\n"
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
+		     ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.3


From 1075cf7a959f72833e54dd2d4f885617e58e3e0a Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:34 +0100
Subject: x86: merge spinlock.h variants

Merge them finally together

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock.h    | 211 +++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/spinlock_32.h | 208 -----------------------------------------
 include/asm-x86/spinlock_64.h | 186 -------------------------------------
 3 files changed, 209 insertions(+), 396 deletions(-)
 delete mode 100644 include/asm-x86/spinlock_32.h
 delete mode 100644 include/asm-x86/spinlock_64.h

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index e1d555a3dfe5..afd4b80ff0ad 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -1,6 +1,22 @@
 #ifndef _X86_SPINLOCK_H_
 #define _X86_SPINLOCK_H_
 
+#include <asm/atomic.h>
+#include <asm/rwlock.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+
+/*
+ * Your basic SMP spinlocks, allowing only a single CPU anywhere
+ *
+ * Simple spin lock operations.  There are two variants, one clears IRQ's
+ * on the local processor, one does not.
+ *
+ * We make no fairness assumptions. They have a cost.
+ *
+ * (the type definitions are in asm/spinlock_types.h)
+ */
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
@@ -11,9 +27,200 @@
 #endif /* CONFIG_PARAVIRT */
 
 #ifdef CONFIG_X86_32
-# include "spinlock_32.h"
+typedef char _slock_t;
+# define LOCK_INS_DEC "decb"
+# define LOCK_INS_XCH "xchgb"
+# define LOCK_INS_MOV "movb"
+# define LOCK_INS_CMP "cmpb"
+# define LOCK_PTR_REG "a"
 #else
-# include "spinlock_64.h"
+typedef int _slock_t;
+# define LOCK_INS_DEC "decl"
+# define LOCK_INS_XCH "xchgl"
+# define LOCK_INS_MOV "movl"
+# define LOCK_INS_CMP "cmpl"
+# define LOCK_PTR_REG "D"
+#endif
+
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+{
+	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
+}
+
+static inline void __raw_spin_lock(raw_spinlock_t *lock)
+{
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
+		"jns 3f\n"
+		"2:\t"
+		"rep;nop\n\t"
+		LOCK_INS_CMP " $0,%0\n\t"
+		"jle 2b\n\t"
+		"jmp 1b\n"
+		"3:\n\t"
+		: "+m" (lock->slock) : : "memory");
+}
+
+/*
+ * It is easier for the lock validator if interrupts are not re-enabled
+ * in the middle of a lock-acquire. This is a performance feature anyway
+ * so we turn it off:
+ *
+ * NOTE: there's an irqs-on section here, which normally would have to be
+ * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
+ */
+#ifndef CONFIG_PROVE_LOCKING
+static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
+					 unsigned long flags)
+{
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
+		"jns 5f\n"
+		"testl $0x200, %[flags]\n\t"
+		"jz 4f\n\t"
+		STI_STRING "\n"
+		"3:\t"
+		"rep;nop\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
+		"jle 3b\n\t"
+		CLI_STRING "\n\t"
+		"jmp 1b\n"
+		"4:\t"
+		"rep;nop\n\t"
+		LOCK_INS_CMP " $0, %[slock]\n\t"
+		"jg 1b\n\t"
+		"jmp 4b\n"
+		"5:\n\t"
+		: [slock] "+m" (lock->slock)
+		: [flags] "r" ((u32)flags)
+		  CLI_STI_INPUT_ARGS
+		: "memory" CLI_STI_CLOBBERS);
+}
+#endif
+
+static inline int __raw_spin_trylock(raw_spinlock_t *lock)
+{
+	_slock_t oldval;
+
+	asm volatile(
+		LOCK_INS_XCH " %0,%1"
+		:"=q" (oldval), "+m" (lock->slock)
+		:"0" (0) : "memory");
+
+	return oldval > 0;
+}
+
+/*
+ * __raw_spin_unlock based on writing $1 to the low byte.
+ * This method works. Despite all the confusion.
+ * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there)
+ * (PPro errata 66, 92)
+ */
+#if defined(X86_64) || \
+	(!defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE))
+
+static inline void __raw_spin_unlock(raw_spinlock_t *lock)
+{
+	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
+}
+
+#else
+
+static inline void __raw_spin_unlock(raw_spinlock_t *lock)
+{
+	unsigned char oldval = 1;
+
+	asm volatile("xchgb %b0, %1"
+		     : "=q" (oldval), "+m" (lock->slock)
+		     : "0" (oldval) : "memory");
+}
+
 #endif
 
+static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
+{
+	while (__raw_spin_is_locked(lock))
+		cpu_relax();
+}
+
+/*
+ * Read-write spinlocks, allowing multiple readers
+ * but only one writer.
+ *
+ * NOTE! it is quite common to have readers in interrupts
+ * but no interrupt writers. For those circumstances we
+ * can "mix" irq-safe locks - any writer needs to get a
+ * irq-safe write-lock, but readers can get non-irqsafe
+ * read-locks.
+ *
+ * On x86, we implement read-write locks as a 32-bit counter
+ * with the high bit (sign) being the "contended" bit.
+ */
+
+static inline int __raw_read_can_lock(raw_rwlock_t *lock)
+{
+	return (int)(lock)->lock > 0;
+}
+
+static inline int __raw_write_can_lock(raw_rwlock_t *lock)
+{
+	return (lock)->lock == RW_LOCK_BIAS;
+}
+
+static inline void __raw_read_lock(raw_rwlock_t *rw)
+{
+	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
+		     "jns 1f\n"
+		     "call __read_lock_failed\n\t"
+		     "1:\n"
+		     ::LOCK_PTR_REG (rw) : "memory");
+}
+
+static inline void __raw_write_lock(raw_rwlock_t *rw)
+{
+	asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t"
+		     "jz 1f\n"
+		     "call __write_lock_failed\n\t"
+		     "1:\n"
+		     ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory");
+}
+
+static inline int __raw_read_trylock(raw_rwlock_t *lock)
+{
+	atomic_t *count = (atomic_t *)lock;
+
+	atomic_dec(count);
+	if (atomic_read(count) >= 0)
+		return 1;
+	atomic_inc(count);
+	return 0;
+}
+
+static inline int __raw_write_trylock(raw_rwlock_t *lock)
+{
+	atomic_t *count = (atomic_t *)lock;
+
+	if (atomic_sub_and_test(RW_LOCK_BIAS, count))
+		return 1;
+	atomic_add(RW_LOCK_BIAS, count);
+	return 0;
+}
+
+static inline void __raw_read_unlock(raw_rwlock_t *rw)
+{
+	asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory");
+}
+
+static inline void __raw_write_unlock(raw_rwlock_t *rw)
+{
+	asm volatile(LOCK_PREFIX "addl %1, %0"
+		     : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory");
+}
+
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif
diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h
deleted file mode 100644
index 2de9b8b89903..000000000000
--- a/include/asm-x86/spinlock_32.h
+++ /dev/null
@@ -1,208 +0,0 @@
-#ifndef __ASM_SPINLOCK_H
-#define __ASM_SPINLOCK_H
-
-#include <asm/atomic.h>
-#include <asm/rwlock.h>
-#include <asm/page.h>
-#include <asm/processor.h>
-
-/*
- * Your basic SMP spinlocks, allowing only a single CPU anywhere
- *
- * Simple spin lock operations.  There are two variants, one clears IRQ's
- * on the local processor, one does not.
- *
- * We make no fairness assumptions. They have a cost.
- *
- * (the type definitions are in asm/spinlock_types.h)
- */
-
-typedef char _slock_t;
-#define LOCK_INS_DEC "decb"
-#define LOCK_INS_XCH "xchgb"
-#define LOCK_INS_MOV "movb"
-#define LOCK_INS_CMP "cmpb"
-#define LOCK_PTR_REG "a"
-
-static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
-{
-	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
-}
-
-static inline void __raw_spin_lock(raw_spinlock_t *lock)
-{
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
-		"jns 3f\n"
-		"2:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0,%0\n\t"
-		"jle 2b\n\t"
-		"jmp 1b\n"
-		"3:\n\t"
-		: "+m" (lock->slock) : : "memory");
-}
-
-/*
- * It is easier for the lock validator if interrupts are not re-enabled
- * in the middle of a lock-acquire. This is a performance feature anyway
- * so we turn it off:
- *
- * NOTE: there's an irqs-on section here, which normally would have to be
- * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
- */
-#ifndef CONFIG_PROVE_LOCKING
-static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
-					 unsigned long flags)
-{
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
-		"jns 5f\n"
-		"testl $0x200, %[flags]\n\t"
-		"jz 4f\n\t"
-		STI_STRING "\n"
-		"3:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jle 3b\n\t"
-		CLI_STRING "\n\t"
-		"jmp 1b\n"
-		"4:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jg 1b\n\t"
-		"jmp 4b\n"
-		"5:\n\t"
-		: [slock] "+m" (lock->slock)
-		: [flags] "r" ((u32)flags)
-		  CLI_STI_INPUT_ARGS
-		: "memory" CLI_STI_CLOBBERS);
-}
-#endif
-
-static inline int __raw_spin_trylock(raw_spinlock_t *lock)
-{
-	_slock_t oldval;
-
-	asm volatile(
-		LOCK_INS_XCH " %0,%1"
-		:"=q" (oldval), "+m" (lock->slock)
-		:"0" (0) : "memory");
-
-	return oldval > 0;
-}
-
-/*
- * __raw_spin_unlock based on writing $1 to the low byte.
- * This method works. Despite all the confusion.
- * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there)
- * (PPro errata 66, 92)
- */
-
-#if !defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE)
-
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
-{
-	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
-}
-
-#else
-
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
-{
-	unsigned char oldval = 1;
-
-	asm volatile("xchgb %b0, %1"
-		     : "=q" (oldval), "+m" (lock->slock)
-		     : "0" (oldval) : "memory");
-}
-
-#endif
-
-static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
-{
-	while (__raw_spin_is_locked(lock))
-		cpu_relax();
-}
-
-/*
- * Read-write spinlocks, allowing multiple readers
- * but only one writer.
- *
- * NOTE! it is quite common to have readers in interrupts
- * but no interrupt writers. For those circumstances we
- * can "mix" irq-safe locks - any writer needs to get a
- * irq-safe write-lock, but readers can get non-irqsafe
- * read-locks.
- *
- * On x86, we implement read-write locks as a 32-bit counter
- * with the high bit (sign) being the "contended" bit.
- */
-
-static inline int __raw_read_can_lock(raw_rwlock_t *lock)
-{
-	return (int)(lock)->lock > 0;
-}
-
-static inline int __raw_write_can_lock(raw_rwlock_t *lock)
-{
-	return (lock)->lock == RW_LOCK_BIAS;
-}
-
-static inline void __raw_read_lock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
-		     "jns 1f\n"
-		     "call __read_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (rw) : "memory");
-}
-
-static inline void __raw_write_lock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t"
-		     "jz 1f\n"
-		     "call __write_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory");
-}
-
-static inline int __raw_read_trylock(raw_rwlock_t *lock)
-{
-	atomic_t *count = (atomic_t *)lock;
-
-	atomic_dec(count);
-	if (atomic_read(count) >= 0)
-		return 1;
-	atomic_inc(count);
-	return 0;
-}
-
-static inline int __raw_write_trylock(raw_rwlock_t *lock)
-{
-	atomic_t *count = (atomic_t *)lock;
-
-	if (atomic_sub_and_test(RW_LOCK_BIAS, count))
-		return 1;
-	atomic_add(RW_LOCK_BIAS, count);
-	return 0;
-}
-
-static inline void __raw_read_unlock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory");
-}
-
-static inline void __raw_write_unlock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX "addl %1, %0"
-		     : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory");
-}
-
-#define _raw_spin_relax(lock)	cpu_relax()
-#define _raw_read_relax(lock)	cpu_relax()
-#define _raw_write_relax(lock)	cpu_relax()
-
-#endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h
deleted file mode 100644
index f5ba90b7335c..000000000000
--- a/include/asm-x86/spinlock_64.h
+++ /dev/null
@@ -1,186 +0,0 @@
-#ifndef __ASM_SPINLOCK_H
-#define __ASM_SPINLOCK_H
-
-#include <asm/atomic.h>
-#include <asm/rwlock.h>
-#include <asm/page.h>
-#include <asm/processor.h>
-
-/*
- * Your basic SMP spinlocks, allowing only a single CPU anywhere
- *
- * Simple spin lock operations.  There are two variants, one clears IRQ's
- * on the local processor, one does not.
- *
- * We make no fairness assumptions. They have a cost.
- *
- * (the type definitions are in asm/spinlock_types.h)
- */
-
-typedef int _slock_t;
-#define LOCK_INS_DEC "decl"
-#define LOCK_INS_XCH "xchgl"
-#define LOCK_INS_MOV "movl"
-#define LOCK_INS_CMP "cmpl"
-#define LOCK_PTR_REG "D"
-
-static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
-{
-	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
-}
-
-static inline void __raw_spin_lock(raw_spinlock_t *lock)
-{
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
-		"jns 3f\n"
-		"2:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0,%0\n\t"
-		"jle 2b\n\t"
-		"jmp 1b\n"
-		"3:\n\t"
-		: "+m" (lock->slock) : : "memory");
-}
-
-/*
- * It is easier for the lock validator if interrupts are not re-enabled
- * in the middle of a lock-acquire. This is a performance feature anyway
- * so we turn it off:
- *
- * NOTE: there's an irqs-on section here, which normally would have to be
- * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
- */
-#ifndef CONFIG_PROVE_LOCKING
-static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
-					 unsigned long flags)
-{
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
-		"jns 5f\n"
-		"testl $0x200, %[flags]\n\t"
-		"jz 4f\n\t"
-		STI_STRING "\n"
-		"3:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jle 3b\n\t"
-		CLI_STRING "\n\t"
-		"jmp 1b\n"
-		"4:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jg 1b\n\t"
-		"jmp 4b\n"
-		"5:\n\t"
-		: [slock] "+m" (lock->slock)
-		: [flags] "r" ((u32)flags)
-		  CLI_STI_INPUT_ARGS
-		: "memory" CLI_STI_CLOBBERS);
-}
-#endif
-
-static inline int __raw_spin_trylock(raw_spinlock_t *lock)
-{
-	_slock_t oldval;
-
-	asm volatile(
-		LOCK_INS_XCH " %0,%1"
-		:"=q" (oldval), "+m" (lock->slock)
-		:"0" (0) : "memory");
-
-	return oldval > 0;
-}
-
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
-{
-	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
-}
-
-static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
-{
-	while (__raw_spin_is_locked(lock))
-		cpu_relax();
-}
-
-/*
- * Read-write spinlocks, allowing multiple readers
- * but only one writer.
- *
- * NOTE! it is quite common to have readers in interrupts
- * but no interrupt writers. For those circumstances we
- * can "mix" irq-safe locks - any writer needs to get a
- * irq-safe write-lock, but readers can get non-irqsafe
- * read-locks.
- *
- * On x86, we implement read-write locks as a 32-bit counter
- * with the high bit (sign) being the "contended" bit.
- */
-
-static inline int __raw_read_can_lock(raw_rwlock_t *lock)
-{
-	return (int)(lock)->lock > 0;
-}
-
-static inline int __raw_write_can_lock(raw_rwlock_t *lock)
-{
-	return (lock)->lock == RW_LOCK_BIAS;
-}
-
-static inline void __raw_read_lock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
-		     "jns 1f\n"
-		     "call __read_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (rw) : "memory");
-}
-
-static inline void __raw_write_lock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t"
-		     "jz 1f\n"
-		     "call __write_lock_failed\n\t"
-		     "1:\n"
-		     ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory");
-}
-
-static inline int __raw_read_trylock(raw_rwlock_t *lock)
-{
-	atomic_t *count = (atomic_t *)lock;
-
-	atomic_dec(count);
-	if (atomic_read(count) >= 0)
-		return 1;
-	atomic_inc(count);
-	return 0;
-}
-
-static inline int __raw_write_trylock(raw_rwlock_t *lock)
-{
-	atomic_t *count = (atomic_t *)lock;
-
-	if (atomic_sub_and_test(RW_LOCK_BIAS, count))
-		return 1;
-	atomic_add(RW_LOCK_BIAS, count);
-	return 0;
-}
-
-static inline void __raw_read_unlock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory");
-}
-
-static inline void __raw_write_unlock(raw_rwlock_t *rw)
-{
-	asm volatile(LOCK_PREFIX "addl %1, %0"
-		     : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory");
-}
-
-#define _raw_spin_relax(lock)	cpu_relax()
-#define _raw_read_relax(lock)	cpu_relax()
-#define _raw_write_relax(lock)	cpu_relax()
-
-#endif /* __ASM_SPINLOCK_H */
-- 
cgit v1.2.3


From 0b9c99b6f21c2e9e00938e9c57942ed71bfe4d21 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:35 +0100
Subject: x86: cleanup tlbflush.h variants

Bring the tlbflush.h variants into sync to prepare merging and
paravirt support.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/smp_64.c            |  12 ++--
 arch/x86/mach-voyager/voyager_smp.c |   7 +--
 arch/x86/mm/boot_ioremap_32.c       |   2 +-
 include/asm-x86/cpufeature.h        |   9 +++
 include/asm-x86/tlbflush_32.h       | 120 +++++++++++++++---------------------
 include/asm-x86/tlbflush_64.h       |  78 +++++++++++++++++------
 6 files changed, 128 insertions(+), 100 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 62b0f2a1b1e8..7142447b5666 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -55,7 +55,6 @@ union smp_flush_state {
 		cpumask_t flush_cpumask;
 		struct mm_struct *flush_mm;
 		unsigned long flush_va;
-#define FLUSH_ALL	-1ULL
 		spinlock_t tlbstate_lock;
 	};
 	char pad[SMP_CACHE_BYTES];
@@ -153,7 +152,7 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs)
 
 	if (f->flush_mm == read_pda(active_mm)) {
 		if (read_pda(mmu_state) == TLBSTATE_OK) {
-			if (f->flush_va == FLUSH_ALL)
+			if (f->flush_va == TLB_FLUSH_ALL)
 				local_flush_tlb();
 			else
 				__flush_tlb_one(f->flush_va);
@@ -166,11 +165,12 @@ out:
 	add_pda(irq_tlb_count, 1);
 }
 
-static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
-						unsigned long va)
+void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
+			     unsigned long va)
 {
 	int sender;
 	union smp_flush_state *f;
+	cpumask_t cpumask = *cpumaskp;
 
 	/* Caller has disabled preemption */
 	sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS;
@@ -223,7 +223,7 @@ void flush_tlb_current_task(void)
 
 	local_flush_tlb();
 	if (!cpus_empty(cpu_mask))
-		flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+		flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL);
 	preempt_enable();
 }
 
@@ -242,7 +242,7 @@ void flush_tlb_mm (struct mm_struct * mm)
 			leave_mm(smp_processor_id());
 	}
 	if (!cpus_empty(cpu_mask))
-		flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+		flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL);
 
 	preempt_enable();
 }
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 981def2b4e9b..b472a2df0b7f 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -800,7 +800,6 @@ static void smp_reschedule_interrupt(void)
 static struct mm_struct *flush_mm;
 static unsigned long flush_va;
 static DEFINE_SPINLOCK(tlbstate_lock);
-#define FLUSH_ALL	0xffffffff
 
 /*
  * We cannot call mmdrop() because we are in interrupt context,
@@ -834,7 +833,7 @@ static void smp_invalidate_interrupt(void)
 
 	if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) {
 		if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) {
-			if (flush_va == FLUSH_ALL)
+			if (flush_va == TLB_FLUSH_ALL)
 				local_flush_tlb();
 			else
 				__flush_tlb_one(flush_va);
@@ -903,7 +902,7 @@ void flush_tlb_current_task(void)
 	cpu_mask = cpus_addr(mm->cpu_vm_mask)[0] & ~(1 << smp_processor_id());
 	local_flush_tlb();
 	if (cpu_mask)
-		voyager_flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+		voyager_flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL);
 
 	preempt_enable();
 }
@@ -923,7 +922,7 @@ void flush_tlb_mm(struct mm_struct *mm)
 			leave_mm(smp_processor_id());
 	}
 	if (cpu_mask)
-		voyager_flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+		voyager_flush_tlb_others(cpu_mask, mm, TLB_FLUSH_ALL);
 
 	preempt_enable();
 }
diff --git a/arch/x86/mm/boot_ioremap_32.c b/arch/x86/mm/boot_ioremap_32.c
index f14da2a53ece..b20f74a2770f 100644
--- a/arch/x86/mm/boot_ioremap_32.c
+++ b/arch/x86/mm/boot_ioremap_32.c
@@ -57,7 +57,7 @@ static void __boot_ioremap(unsigned long phys_addr, unsigned long nrpages,
 	pte = boot_vaddr_to_pte(virtual_source);
 	for (i=0; i < nrpages; i++, phys_addr += PAGE_SIZE, pte++) {
 		set_pte(pte, pfn_pte(phys_addr>>PAGE_SHIFT, PAGE_KERNEL));
-		__flush_tlb_one(&vaddr[i*PAGE_SIZE]);
+		__flush_tlb_one((unsigned long) &vaddr[i*PAGE_SIZE]);
 	}
 }
 
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 4c7875554d01..acbf6681740d 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -163,6 +163,12 @@
 #define cpu_has_clflush		boot_cpu_has(X86_FEATURE_CLFLSH)
 #define cpu_has_bts		boot_cpu_has(X86_FEATURE_BTS)
 
+#if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64)
+# define cpu_has_invlpg		1
+#else
+# define cpu_has_invlpg		(boot_cpu_data.x86 > 3)
+#endif
+
 #ifdef CONFIG_X86_64
 
 #undef  cpu_has_vme
@@ -183,6 +189,9 @@
 #undef  cpu_has_centaur_mcr
 #define cpu_has_centaur_mcr	0
 
+#undef  cpu_has_pge
+#define cpu_has_pge		1
+
 #endif /* CONFIG_X86_64 */
 
 #endif /* _ASM_X86_CPUFEATURE_H */
diff --git a/include/asm-x86/tlbflush_32.h b/include/asm-x86/tlbflush_32.h
index 2bd5b95e2048..9e07cc8f2d94 100644
--- a/include/asm-x86/tlbflush_32.h
+++ b/include/asm-x86/tlbflush_32.h
@@ -1,8 +1,11 @@
-#ifndef _I386_TLBFLUSH_H
-#define _I386_TLBFLUSH_H
+#ifndef _X86_TLBFLUSH_H
+#define _X86_TLBFLUSH_H
 
 #include <linux/mm.h>
+#include <linux/sched.h>
+
 #include <asm/processor.h>
+#include <asm/system.h>
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
@@ -12,62 +15,41 @@
 #define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
 #endif
 
-#define __native_flush_tlb()						\
-	do {								\
-		unsigned int tmpreg;					\
-									\
-		__asm__ __volatile__(					\
-			"movl %%cr3, %0;              \n"		\
-			"movl %0, %%cr3;  # flush TLB \n"		\
-			: "=r" (tmpreg)					\
-			:: "memory");					\
-	} while (0)
+static inline void __native_flush_tlb(void)
+{
+	write_cr3(read_cr3());
+}
 
-/*
- * Global pages have to be flushed a bit differently. Not a real
- * performance problem because this does not happen often.
- */
-#define __native_flush_tlb_global()					\
-	do {								\
-		unsigned int tmpreg, cr4, cr4_orig;			\
-									\
-		__asm__ __volatile__(					\
-			"movl %%cr4, %2;  # turn off PGE     \n"	\
-			"movl %2, %1;                        \n"	\
-			"andl %3, %1;                        \n"	\
-			"movl %1, %%cr4;                     \n"	\
-			"movl %%cr3, %0;                     \n"	\
-			"movl %0, %%cr3;  # flush TLB        \n"	\
-			"movl %2, %%cr4;  # turn PGE back on \n"	\
-			: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig)	\
-			: "i" (~X86_CR4_PGE)				\
-			: "memory");					\
-	} while (0)
-
-#define __native_flush_tlb_single(addr) 				\
-	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory")
-
-# define __flush_tlb_all()						\
-	do {								\
-		if (cpu_has_pge)					\
-			__flush_tlb_global();				\
-		else							\
-			__flush_tlb();					\
-	} while (0)
-
-#define cpu_has_invlpg	(boot_cpu_data.x86 > 3)
-
-#ifdef CONFIG_X86_INVLPG
-# define __flush_tlb_one(addr) __flush_tlb_single(addr)
-#else
-# define __flush_tlb_one(addr)						\
-	do {								\
-		if (cpu_has_invlpg)					\
-			__flush_tlb_single(addr);			\
-		else							\
-			__flush_tlb();					\
-	} while (0)
-#endif
+static inline void __native_flush_tlb_global(void)
+{
+	unsigned long cr4 = read_cr4();
+
+	/* clear PGE */
+	write_cr4(cr4 & ~X86_CR4_PGE);
+	/* write old PGE again and flush TLBs */
+	write_cr4(cr4);
+}
+
+static inline void __native_flush_tlb_single(unsigned long addr)
+{
+	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
+}
+
+static inline void __flush_tlb_all(void)
+{
+	if (cpu_has_pge)
+		__flush_tlb_global();
+	else
+		__flush_tlb();
+}
+
+static inline void __flush_tlb_one(unsigned long addr)
+{
+	if (cpu_has_invlpg)
+		__flush_tlb_single(addr);
+	else
+		__flush_tlb();
+}
 
 /*
  * TLB flushing:
@@ -86,11 +68,8 @@
 
 #define TLB_FLUSH_ALL	0xffffffff
 
-
 #ifndef CONFIG_SMP
 
-#include <linux/sched.h>
-
 #define flush_tlb() __flush_tlb()
 #define flush_tlb_all() __flush_tlb_all()
 #define local_flush_tlb() __flush_tlb()
@@ -102,21 +81,22 @@ static inline void flush_tlb_mm(struct mm_struct *mm)
 }
 
 static inline void flush_tlb_page(struct vm_area_struct *vma,
-	unsigned long addr)
+				  unsigned long addr)
 {
 	if (vma->vm_mm == current->active_mm)
 		__flush_tlb_one(addr);
 }
 
 static inline void flush_tlb_range(struct vm_area_struct *vma,
-	unsigned long start, unsigned long end)
+				   unsigned long start, unsigned long end)
 {
 	if (vma->vm_mm == current->active_mm)
 		__flush_tlb();
 }
 
 static inline void native_flush_tlb_others(const cpumask_t *cpumask,
-					   struct mm_struct *mm, unsigned long va)
+					   struct mm_struct *mm,
+					   unsigned long va)
 {
 }
 
@@ -124,8 +104,7 @@ static inline void native_flush_tlb_others(const cpumask_t *cpumask,
 
 #include <asm/smp.h>
 
-#define local_flush_tlb() \
-	__flush_tlb()
+#define local_flush_tlb() __flush_tlb()
 
 extern void flush_tlb_all(void);
 extern void flush_tlb_current_task(void);
@@ -134,7 +113,8 @@ extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
 
 #define flush_tlb()	flush_tlb_current_task()
 
-static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
 {
 	flush_tlb_mm(vma->vm_mm);
 }
@@ -152,17 +132,17 @@ struct tlb_state
 	char __cacheline_padding[L1_CACHE_BYTES-8];
 };
 DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
+
 #endif	/* SMP */
 
 #ifndef CONFIG_PARAVIRT
-#define flush_tlb_others(mask, mm, va)		\
-	native_flush_tlb_others(&mask, mm, va)
+#define flush_tlb_others(mask, mm, va)	native_flush_tlb_others(&mask, mm, va)
 #endif
 
 static inline void flush_tlb_kernel_range(unsigned long start,
-					unsigned long end)
+					  unsigned long end)
 {
 	flush_tlb_all();
 }
 
-#endif /* _I386_TLBFLUSH_H */
+#endif /* _X86_TLBFLUSH_H */
diff --git a/include/asm-x86/tlbflush_64.h b/include/asm-x86/tlbflush_64.h
index 7731fd23d572..0bed440ba9fe 100644
--- a/include/asm-x86/tlbflush_64.h
+++ b/include/asm-x86/tlbflush_64.h
@@ -1,26 +1,55 @@
-#ifndef _X8664_TLBFLUSH_H
-#define _X8664_TLBFLUSH_H
+#ifndef _X86_TLBFLUSH_H
+#define _X86_TLBFLUSH_H
 
 #include <linux/mm.h>
 #include <linux/sched.h>
+
 #include <asm/processor.h>
 #include <asm/system.h>
 
-static inline void __flush_tlb(void)
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define __flush_tlb() __native_flush_tlb()
+#define __flush_tlb_global() __native_flush_tlb_global()
+#define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
+#endif
+
+static inline void __native_flush_tlb(void)
 {
 	write_cr3(read_cr3());
 }
 
-static inline void __flush_tlb_all(void)
+static inline void __native_flush_tlb_global(void)
 {
 	unsigned long cr4 = read_cr4();
-	write_cr4(cr4 & ~X86_CR4_PGE);	/* clear PGE */
-	write_cr4(cr4);			/* write old PGE again and flush TLBs */
+
+	/* clear PGE */
+	write_cr4(cr4 & ~X86_CR4_PGE);
+	/* write old PGE again and flush TLBs */
+	write_cr4(cr4);
 }
 
-#define __flush_tlb_one(addr) \
-	__asm__ __volatile__("invlpg (%0)" :: "r" (addr) : "memory")
+static inline void __native_flush_tlb_single(unsigned long addr)
+{
+	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
+}
+
+static inline void __flush_tlb_all(void)
+{
+	if (cpu_has_pge)
+		__flush_tlb_global();
+	else
+		__flush_tlb();
+}
 
+static inline void __flush_tlb_one(unsigned long addr)
+{
+	if (cpu_has_invlpg)
+		__flush_tlb_single(addr);
+	else
+		__flush_tlb();
+}
 
 /*
  * TLB flushing:
@@ -37,6 +66,8 @@ static inline void __flush_tlb_all(void)
  * range a few INVLPGs in a row are a win.
  */
 
+#define TLB_FLUSH_ALL	-1ULL
+
 #ifndef CONFIG_SMP
 
 #define flush_tlb() __flush_tlb()
@@ -50,25 +81,30 @@ static inline void flush_tlb_mm(struct mm_struct *mm)
 }
 
 static inline void flush_tlb_page(struct vm_area_struct *vma,
-	unsigned long addr)
+				  unsigned long addr)
 {
 	if (vma->vm_mm == current->active_mm)
 		__flush_tlb_one(addr);
 }
 
 static inline void flush_tlb_range(struct vm_area_struct *vma,
-	unsigned long start, unsigned long end)
+				   unsigned long start, unsigned long end)
 {
 	if (vma->vm_mm == current->active_mm)
 		__flush_tlb();
 }
 
-#else
+static inline void native_flush_tlb_others(const cpumask_t *cpumask,
+					   struct mm_struct *mm,
+					   unsigned long va)
+{
+}
+
+#else  /* SMP */
 
 #include <asm/smp.h>
 
-#define local_flush_tlb() \
-	__flush_tlb()
+#define local_flush_tlb() __flush_tlb()
 
 extern void flush_tlb_all(void);
 extern void flush_tlb_current_task(void);
@@ -77,24 +113,28 @@ extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
 
 #define flush_tlb()	flush_tlb_current_task()
 
-static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
 {
 	flush_tlb_mm(vma->vm_mm);
 }
 
+void native_flush_tlb_others(const cpumask_t *cpumask, struct mm_struct *mm,
+			     unsigned long va);
+
 #define TLBSTATE_OK	1
 #define TLBSTATE_LAZY	2
 
-/* Roughly an IPI every 20MB with 4k pages for freeing page table
-   ranges. Cost is about 42k of memory for each CPU. */
-#define ARCH_FREE_PTE_NR 5350	
+#endif	/* SMP */
 
+#ifndef CONFIG_PARAVIRT
+#define flush_tlb_others(mask, mm, va)	native_flush_tlb_others(&mask, mm, va)
 #endif
 
 static inline void flush_tlb_kernel_range(unsigned long start,
-					unsigned long end)
+					  unsigned long end)
 {
 	flush_tlb_all();
 }
 
-#endif /* _X8664_TLBFLUSH_H */
+#endif /* _X86_TLBFLUSH_H */
-- 
cgit v1.2.3


From d291cf83639a0e0b67ff783b6ed29c0a747d4901 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:35 +0100
Subject: x86: merge tlbflush.h variants

The delta is now minimal. Merge them

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/tlbflush.h    | 157 +++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/tlbflush_32.h | 148 ---------------------------------------
 include/asm-x86/tlbflush_64.h | 140 -------------------------------------
 3 files changed, 155 insertions(+), 290 deletions(-)
 delete mode 100644 include/asm-x86/tlbflush_32.h
 delete mode 100644 include/asm-x86/tlbflush_64.h

(limited to 'include')

diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h
index 9af4cc83a1af..3998709ed637 100644
--- a/include/asm-x86/tlbflush.h
+++ b/include/asm-x86/tlbflush.h
@@ -1,5 +1,158 @@
+#ifndef _ASM_X86_TLBFLUSH_H
+#define _ASM_X86_TLBFLUSH_H
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+
+#include <asm/processor.h>
+#include <asm/system.h>
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define __flush_tlb() __native_flush_tlb()
+#define __flush_tlb_global() __native_flush_tlb_global()
+#define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
+#endif
+
+static inline void __native_flush_tlb(void)
+{
+	write_cr3(read_cr3());
+}
+
+static inline void __native_flush_tlb_global(void)
+{
+	unsigned long cr4 = read_cr4();
+
+	/* clear PGE */
+	write_cr4(cr4 & ~X86_CR4_PGE);
+	/* write old PGE again and flush TLBs */
+	write_cr4(cr4);
+}
+
+static inline void __native_flush_tlb_single(unsigned long addr)
+{
+	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
+}
+
+static inline void __flush_tlb_all(void)
+{
+	if (cpu_has_pge)
+		__flush_tlb_global();
+	else
+		__flush_tlb();
+}
+
+static inline void __flush_tlb_one(unsigned long addr)
+{
+	if (cpu_has_invlpg)
+		__flush_tlb_single(addr);
+	else
+		__flush_tlb();
+}
+
 #ifdef CONFIG_X86_32
-# include "tlbflush_32.h"
+# define TLB_FLUSH_ALL	0xffffffff
 #else
-# include "tlbflush_64.h"
+# define TLB_FLUSH_ALL	-1ULL
+#endif
+
+/*
+ * TLB flushing:
+ *
+ *  - flush_tlb() flushes the current mm struct TLBs
+ *  - flush_tlb_all() flushes all processes TLBs
+ *  - flush_tlb_mm(mm) flushes the specified mm context TLB's
+ *  - flush_tlb_page(vma, vmaddr) flushes one page
+ *  - flush_tlb_range(vma, start, end) flushes a range of pages
+ *  - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
+ *  - flush_tlb_others(cpumask, mm, va) flushes TLBs on other cpus
+ *
+ * ..but the i386 has somewhat limited tlb flushing capabilities,
+ * and page-granular flushes are available only on i486 and up.
+ *
+ * x86-64 can only flush individual pages or full VMs. For a range flush
+ * we always do the full VM. Might be worth trying if for a small
+ * range a few INVLPGs in a row are a win.
+ */
+
+#ifndef CONFIG_SMP
+
+#define flush_tlb() __flush_tlb()
+#define flush_tlb_all() __flush_tlb_all()
+#define local_flush_tlb() __flush_tlb()
+
+static inline void flush_tlb_mm(struct mm_struct *mm)
+{
+	if (mm == current->active_mm)
+		__flush_tlb();
+}
+
+static inline void flush_tlb_page(struct vm_area_struct *vma,
+				  unsigned long addr)
+{
+	if (vma->vm_mm == current->active_mm)
+		__flush_tlb_one(addr);
+}
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
+{
+	if (vma->vm_mm == current->active_mm)
+		__flush_tlb();
+}
+
+static inline void native_flush_tlb_others(const cpumask_t *cpumask,
+					   struct mm_struct *mm,
+					   unsigned long va)
+{
+}
+
+#else  /* SMP */
+
+#include <asm/smp.h>
+
+#define local_flush_tlb() __flush_tlb()
+
+extern void flush_tlb_all(void);
+extern void flush_tlb_current_task(void);
+extern void flush_tlb_mm(struct mm_struct *);
+extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
+
+#define flush_tlb()	flush_tlb_current_task()
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
+{
+	flush_tlb_mm(vma->vm_mm);
+}
+
+void native_flush_tlb_others(const cpumask_t *cpumask, struct mm_struct *mm,
+			     unsigned long va);
+
+#define TLBSTATE_OK	1
+#define TLBSTATE_LAZY	2
+
+#ifdef CONFIG_X86_32
+struct tlb_state
+{
+	struct mm_struct *active_mm;
+	int state;
+	char __cacheline_padding[L1_CACHE_BYTES-8];
+};
+DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
+#endif
+
+#endif	/* SMP */
+
+#ifndef CONFIG_PARAVIRT
+#define flush_tlb_others(mask, mm, va)	native_flush_tlb_others(&mask, mm, va)
 #endif
+
+static inline void flush_tlb_kernel_range(unsigned long start,
+					  unsigned long end)
+{
+	flush_tlb_all();
+}
+
+#endif /* _ASM_X86_TLBFLUSH_H */
diff --git a/include/asm-x86/tlbflush_32.h b/include/asm-x86/tlbflush_32.h
deleted file mode 100644
index 9e07cc8f2d94..000000000000
--- a/include/asm-x86/tlbflush_32.h
+++ /dev/null
@@ -1,148 +0,0 @@
-#ifndef _X86_TLBFLUSH_H
-#define _X86_TLBFLUSH_H
-
-#include <linux/mm.h>
-#include <linux/sched.h>
-
-#include <asm/processor.h>
-#include <asm/system.h>
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define __flush_tlb() __native_flush_tlb()
-#define __flush_tlb_global() __native_flush_tlb_global()
-#define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
-#endif
-
-static inline void __native_flush_tlb(void)
-{
-	write_cr3(read_cr3());
-}
-
-static inline void __native_flush_tlb_global(void)
-{
-	unsigned long cr4 = read_cr4();
-
-	/* clear PGE */
-	write_cr4(cr4 & ~X86_CR4_PGE);
-	/* write old PGE again and flush TLBs */
-	write_cr4(cr4);
-}
-
-static inline void __native_flush_tlb_single(unsigned long addr)
-{
-	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
-}
-
-static inline void __flush_tlb_all(void)
-{
-	if (cpu_has_pge)
-		__flush_tlb_global();
-	else
-		__flush_tlb();
-}
-
-static inline void __flush_tlb_one(unsigned long addr)
-{
-	if (cpu_has_invlpg)
-		__flush_tlb_single(addr);
-	else
-		__flush_tlb();
-}
-
-/*
- * TLB flushing:
- *
- *  - flush_tlb() flushes the current mm struct TLBs
- *  - flush_tlb_all() flushes all processes TLBs
- *  - flush_tlb_mm(mm) flushes the specified mm context TLB's
- *  - flush_tlb_page(vma, vmaddr) flushes one page
- *  - flush_tlb_range(vma, start, end) flushes a range of pages
- *  - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
- *  - flush_tlb_others(cpumask, mm, va) flushes a TLBs on other cpus
- *
- * ..but the i386 has somewhat limited tlb flushing capabilities,
- * and page-granular flushes are available only on i486 and up.
- */
-
-#define TLB_FLUSH_ALL	0xffffffff
-
-#ifndef CONFIG_SMP
-
-#define flush_tlb() __flush_tlb()
-#define flush_tlb_all() __flush_tlb_all()
-#define local_flush_tlb() __flush_tlb()
-
-static inline void flush_tlb_mm(struct mm_struct *mm)
-{
-	if (mm == current->active_mm)
-		__flush_tlb();
-}
-
-static inline void flush_tlb_page(struct vm_area_struct *vma,
-				  unsigned long addr)
-{
-	if (vma->vm_mm == current->active_mm)
-		__flush_tlb_one(addr);
-}
-
-static inline void flush_tlb_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end)
-{
-	if (vma->vm_mm == current->active_mm)
-		__flush_tlb();
-}
-
-static inline void native_flush_tlb_others(const cpumask_t *cpumask,
-					   struct mm_struct *mm,
-					   unsigned long va)
-{
-}
-
-#else  /* SMP */
-
-#include <asm/smp.h>
-
-#define local_flush_tlb() __flush_tlb()
-
-extern void flush_tlb_all(void);
-extern void flush_tlb_current_task(void);
-extern void flush_tlb_mm(struct mm_struct *);
-extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
-
-#define flush_tlb()	flush_tlb_current_task()
-
-static inline void flush_tlb_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end)
-{
-	flush_tlb_mm(vma->vm_mm);
-}
-
-void native_flush_tlb_others(const cpumask_t *cpumask, struct mm_struct *mm,
-			     unsigned long va);
-
-#define TLBSTATE_OK	1
-#define TLBSTATE_LAZY	2
-
-struct tlb_state
-{
-	struct mm_struct *active_mm;
-	int state;
-	char __cacheline_padding[L1_CACHE_BYTES-8];
-};
-DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
-
-#endif	/* SMP */
-
-#ifndef CONFIG_PARAVIRT
-#define flush_tlb_others(mask, mm, va)	native_flush_tlb_others(&mask, mm, va)
-#endif
-
-static inline void flush_tlb_kernel_range(unsigned long start,
-					  unsigned long end)
-{
-	flush_tlb_all();
-}
-
-#endif /* _X86_TLBFLUSH_H */
diff --git a/include/asm-x86/tlbflush_64.h b/include/asm-x86/tlbflush_64.h
deleted file mode 100644
index 0bed440ba9fe..000000000000
--- a/include/asm-x86/tlbflush_64.h
+++ /dev/null
@@ -1,140 +0,0 @@
-#ifndef _X86_TLBFLUSH_H
-#define _X86_TLBFLUSH_H
-
-#include <linux/mm.h>
-#include <linux/sched.h>
-
-#include <asm/processor.h>
-#include <asm/system.h>
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define __flush_tlb() __native_flush_tlb()
-#define __flush_tlb_global() __native_flush_tlb_global()
-#define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
-#endif
-
-static inline void __native_flush_tlb(void)
-{
-	write_cr3(read_cr3());
-}
-
-static inline void __native_flush_tlb_global(void)
-{
-	unsigned long cr4 = read_cr4();
-
-	/* clear PGE */
-	write_cr4(cr4 & ~X86_CR4_PGE);
-	/* write old PGE again and flush TLBs */
-	write_cr4(cr4);
-}
-
-static inline void __native_flush_tlb_single(unsigned long addr)
-{
-	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
-}
-
-static inline void __flush_tlb_all(void)
-{
-	if (cpu_has_pge)
-		__flush_tlb_global();
-	else
-		__flush_tlb();
-}
-
-static inline void __flush_tlb_one(unsigned long addr)
-{
-	if (cpu_has_invlpg)
-		__flush_tlb_single(addr);
-	else
-		__flush_tlb();
-}
-
-/*
- * TLB flushing:
- *
- *  - flush_tlb() flushes the current mm struct TLBs
- *  - flush_tlb_all() flushes all processes TLBs
- *  - flush_tlb_mm(mm) flushes the specified mm context TLB's
- *  - flush_tlb_page(vma, vmaddr) flushes one page
- *  - flush_tlb_range(vma, start, end) flushes a range of pages
- *  - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
- *
- * x86-64 can only flush individual pages or full VMs. For a range flush
- * we always do the full VM. Might be worth trying if for a small
- * range a few INVLPGs in a row are a win.
- */
-
-#define TLB_FLUSH_ALL	-1ULL
-
-#ifndef CONFIG_SMP
-
-#define flush_tlb() __flush_tlb()
-#define flush_tlb_all() __flush_tlb_all()
-#define local_flush_tlb() __flush_tlb()
-
-static inline void flush_tlb_mm(struct mm_struct *mm)
-{
-	if (mm == current->active_mm)
-		__flush_tlb();
-}
-
-static inline void flush_tlb_page(struct vm_area_struct *vma,
-				  unsigned long addr)
-{
-	if (vma->vm_mm == current->active_mm)
-		__flush_tlb_one(addr);
-}
-
-static inline void flush_tlb_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end)
-{
-	if (vma->vm_mm == current->active_mm)
-		__flush_tlb();
-}
-
-static inline void native_flush_tlb_others(const cpumask_t *cpumask,
-					   struct mm_struct *mm,
-					   unsigned long va)
-{
-}
-
-#else  /* SMP */
-
-#include <asm/smp.h>
-
-#define local_flush_tlb() __flush_tlb()
-
-extern void flush_tlb_all(void);
-extern void flush_tlb_current_task(void);
-extern void flush_tlb_mm(struct mm_struct *);
-extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
-
-#define flush_tlb()	flush_tlb_current_task()
-
-static inline void flush_tlb_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end)
-{
-	flush_tlb_mm(vma->vm_mm);
-}
-
-void native_flush_tlb_others(const cpumask_t *cpumask, struct mm_struct *mm,
-			     unsigned long va);
-
-#define TLBSTATE_OK	1
-#define TLBSTATE_LAZY	2
-
-#endif	/* SMP */
-
-#ifndef CONFIG_PARAVIRT
-#define flush_tlb_others(mask, mm, va)	native_flush_tlb_others(&mask, mm, va)
-#endif
-
-static inline void flush_tlb_kernel_range(unsigned long start,
-					  unsigned long end)
-{
-	flush_tlb_all();
-}
-
-#endif /* _X86_TLBFLUSH_H */
-- 
cgit v1.2.3


From 64883ab0e3386d72112a9091d886352a7b4b8bf6 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:35 +0100
Subject: x86: cleanup mpspec variants

Bring the mpspec variants into sync to prepare merging and
paravirt support.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/mpparse_32.c             |  18 +--
 arch/x86/mach-visws/mpparse.c            |  16 +--
 include/asm-x86/mach-bigsmp/mach_apic.h  |  12 +-
 include/asm-x86/mach-default/mach_apic.h |  18 +--
 include/asm-x86/mach-es7000/mach_apic.h  |  10 +-
 include/asm-x86/mach-numaq/mach_apic.h   |  10 +-
 include/asm-x86/mach-summit/mach_apic.h  |  18 +--
 include/asm-x86/mpspec_32.h              |  49 ++++----
 include/asm-x86/mpspec_64.h              | 196 ++++---------------------------
 include/asm-x86/mpspec_def.h             |  87 ++++++++------
 10 files changed, 148 insertions(+), 286 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 7a05a7f6099a..22fc8d7dec11 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -258,7 +258,7 @@ static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
 	if (!(m->mpc_flags & MPC_APIC_USABLE))
 		return;
 
-	printk(KERN_INFO "I/O APIC #%d Version %d at 0x%lX.\n",
+	printk(KERN_INFO "I/O APIC #%d Version %d at 0x%X.\n",
 		m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
 	if (nr_ioapics >= MAX_IO_APICS) {
 		printk(KERN_CRIT "Max # of I/O APICs (%d) exceeded (found %d).\n",
@@ -405,9 +405,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 
 	mps_oem_check(mpc, oem, str);
 
-	printk("APIC at: 0x%lX\n",mpc->mpc_lapic);
+	printk("APIC at: 0x%X\n", mpc->mpc_lapic);
 
-	/* 
+	/*
 	 * Save the local APIC address (it might be non-default) -- but only
 	 * if we're not using ACPI.
 	 */
@@ -918,14 +918,14 @@ void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 	 */
 	mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
 	mp_ioapic_routing[idx].gsi_base = gsi_base;
-	mp_ioapic_routing[idx].gsi_end = gsi_base + 
+	mp_ioapic_routing[idx].gsi_end = gsi_base +
 		io_apic_get_redir_entries(idx);
 
-	printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, "
-		"GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, 
-		mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
-		mp_ioapic_routing[idx].gsi_base,
-		mp_ioapic_routing[idx].gsi_end);
+	printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
+	       "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
+	       mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
+	       mp_ioapic_routing[idx].gsi_base,
+	       mp_ioapic_routing[idx].gsi_end);
 }
 
 void __init
diff --git a/arch/x86/mach-visws/mpparse.c b/arch/x86/mach-visws/mpparse.c
index f3c74fab8b95..2a8456a1f44f 100644
--- a/arch/x86/mach-visws/mpparse.c
+++ b/arch/x86/mach-visws/mpparse.c
@@ -36,19 +36,19 @@ unsigned int __initdata maxcpus = NR_CPUS;
 
 static void __init MP_processor_info (struct mpc_config_processor *m)
 {
- 	int ver, logical_apicid;
+	int ver, logical_apicid;
 	physid_mask_t apic_cpus;
- 	
+
 	if (!(m->mpc_cpuflag & CPU_ENABLED))
 		return;
 
 	logical_apicid = m->mpc_apicid;
-	printk(KERN_INFO "%sCPU #%d %ld:%ld APIC version %d\n",
-		m->mpc_cpuflag & CPU_BOOTPROCESSOR ? "Bootup " : "",
-		m->mpc_apicid,
-		(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-		(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-		m->mpc_apicver);
+	printk(KERN_INFO "%sCPU #%d %u:%u APIC version %d\n",
+	       m->mpc_cpuflag & CPU_BOOTPROCESSOR ? "Bootup " : "",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver);
 
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR)
 		boot_cpu_physical_apicid = m->mpc_apicid;
diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/mach-bigsmp/mach_apic.h
index ebd319f838ab..6df235e8ea91 100644
--- a/include/asm-x86/mach-bigsmp/mach_apic.h
+++ b/include/asm-x86/mach-bigsmp/mach_apic.h
@@ -110,13 +110,13 @@ static inline int cpu_to_logical_apicid(int cpu)
 }
 
 static inline int mpc_apic_id(struct mpc_config_processor *m,
-			struct mpc_config_translation *translation_record)
+			      struct mpc_config_translation *translation_record)
 {
-	printk("Processor #%d %ld:%ld APIC version %d\n",
-	        m->mpc_apicid,
-	        (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	        (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	        m->mpc_apicver);
+	printk("Processor #%d %u:%u APIC version %d\n",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver);
 	return m->mpc_apicid;
 }
 
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index 6db1c3babe9a..e3c2c1012c1c 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -89,15 +89,15 @@ static inline physid_mask_t apicid_to_cpu_present(int phys_apicid)
 	return physid_mask_of_physid(phys_apicid);
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m, 
-			struct mpc_config_translation *translation_record)
-{
-	printk("Processor #%d %ld:%ld APIC version %d\n",
-			m->mpc_apicid,
-			(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-			(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-			m->mpc_apicver);
-	return (m->mpc_apicid);
+static inline int mpc_apic_id(struct mpc_config_processor *m,
+			      struct mpc_config_translation *translation_record)
+{
+	printk("Processor #%d %u:%u APIC version %d\n",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver);
+	return m->mpc_apicid;
 }
 
 static inline void setup_portio_remap(void)
diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h
index caec64be516d..d23011fdf454 100644
--- a/include/asm-x86/mach-es7000/mach_apic.h
+++ b/include/asm-x86/mach-es7000/mach_apic.h
@@ -131,11 +131,11 @@ static inline int cpu_to_logical_apicid(int cpu)
 
 static inline int mpc_apic_id(struct mpc_config_processor *m, struct mpc_config_translation *unused)
 {
-	printk("Processor #%d %ld:%ld APIC version %d\n",
-	        m->mpc_apicid,
-	        (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	        (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	        m->mpc_apicver);
+	printk("Processor #%d %u:%u APIC version %d\n",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver);
 	return (m->mpc_apicid);
 }
 
diff --git a/include/asm-x86/mach-numaq/mach_apic.h b/include/asm-x86/mach-numaq/mach_apic.h
index 5e5e7dd2692e..17e183bd39c1 100644
--- a/include/asm-x86/mach-numaq/mach_apic.h
+++ b/include/asm-x86/mach-numaq/mach_apic.h
@@ -101,11 +101,11 @@ static inline int mpc_apic_id(struct mpc_config_processor *m,
 	int quad = translation_record->trans_quad;
 	int logical_apicid = generate_logical_apicid(quad, m->mpc_apicid);
 
-	printk("Processor #%d %ld:%ld APIC version %d (quad %d, apic %d)\n",
-			m->mpc_apicid,
-			(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-			(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-			m->mpc_apicver, quad, logical_apicid);
+	printk("Processor #%d %u:%u APIC version %d (quad %d, apic %d)\n",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver, quad, logical_apicid);
 	return logical_apicid;
 }
 
diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h
index 732f776aab8e..062c97f6100b 100644
--- a/include/asm-x86/mach-summit/mach_apic.h
+++ b/include/asm-x86/mach-summit/mach_apic.h
@@ -126,15 +126,15 @@ static inline physid_mask_t apicid_to_cpu_present(int apicid)
 	return physid_mask_of_physid(0);
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m, 
-			struct mpc_config_translation *translation_record)
-{
-	printk("Processor #%d %ld:%ld APIC version %d\n",
-			m->mpc_apicid,
-			(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-			(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-			m->mpc_apicver);
-	return (m->mpc_apicid);
+static inline int mpc_apic_id(struct mpc_config_processor *m,
+			      struct mpc_config_translation *translation_record)
+{
+	printk("Processor #%d %u:%u APIC version %d\n",
+	       m->mpc_apicid,
+	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+	       m->mpc_apicver);
+	return m->mpc_apicid;
 }
 
 static inline void setup_portio_remap(void)
diff --git a/include/asm-x86/mpspec_32.h b/include/asm-x86/mpspec_32.h
index f21349399d14..bb73185e6a7d 100644
--- a/include/asm-x86/mpspec_32.h
+++ b/include/asm-x86/mpspec_32.h
@@ -1,34 +1,37 @@
 #ifndef __ASM_MPSPEC_H
 #define __ASM_MPSPEC_H
 
-#include <linux/cpumask.h>
 #include <asm/mpspec_def.h>
 #include <mach_mpspec.h>
 
-extern int mp_bus_id_to_type [MAX_MP_BUSSES];
-extern int mp_bus_id_to_node [MAX_MP_BUSSES];
-extern int mp_bus_id_to_local [MAX_MP_BUSSES];
-extern int quad_local_to_mp_bus_id [NR_CPUS/4][4];
-extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
+extern int mp_bus_id_to_type[MAX_MP_BUSSES];
+extern int mp_bus_id_to_node[MAX_MP_BUSSES];
+extern int mp_bus_id_to_local[MAX_MP_BUSSES];
+extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
 
 extern unsigned int def_to_bigsmp;
+extern int apic_version[MAX_APICS];
+extern int pic_mode;
+
+extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
+
 extern unsigned int boot_cpu_physical_apicid;
 extern int smp_found_config;
-extern void find_smp_config (void);
-extern void get_smp_config (void);
 extern int nr_ioapics;
-extern int apic_version [MAX_APICS];
 extern int mp_irq_entries;
-extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
+extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 extern int mpc_default_type;
 extern unsigned long mp_lapic_addr;
-extern int pic_mode;
+
+extern void find_smp_config (void);
+extern void get_smp_config (void);
 
 #ifdef CONFIG_ACPI
 extern void mp_register_lapic (u8 id, u8 enabled);
 extern void mp_register_lapic_address (u64 address);
 extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
-extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger, u32 gsi);
+extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger,
+				    u32 gsi);
 extern void mp_config_acpi_legacy_irqs (void);
 extern int mp_register_gsi (u32 gsi, int edge_level, int active_high_low);
 #endif /* CONFIG_ACPI */
@@ -50,7 +53,7 @@ typedef struct physid_mask physid_mask_t;
 #define physids_and(dst, src1, src2)		bitmap_and((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
 #define physids_or(dst, src1, src2)		bitmap_or((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
 #define physids_clear(map)			bitmap_zero((map).mask, MAX_APICS)
-#define physids_complement(dst, src)		bitmap_complement((dst).mask,(src).mask, MAX_APICS)
+#define physids_complement(dst, src)		bitmap_complement((dst).mask, (src).mask, MAX_APICS)
 #define physids_empty(map)			bitmap_empty((map).mask, MAX_APICS)
 #define physids_equal(map1, map2)		bitmap_equal((map1).mask, (map2).mask, MAX_APICS)
 #define physids_weight(map)			bitmap_weight((map).mask, MAX_APICS)
@@ -58,18 +61,18 @@ typedef struct physid_mask physid_mask_t;
 #define physids_shift_left(d, s, n)		bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
 #define physids_coerce(map)			((map).mask[0])
 
-#define physids_promote(physids)						\
-	({									\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;			\
-		__physid_mask.mask[0] = physids;				\
-		__physid_mask;							\
+#define physids_promote(physids)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		__physid_mask.mask[0] = physids;			\
+		__physid_mask;						\
 	})
 
-#define physid_mask_of_physid(physid)						\
-	({									\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;			\
-		physid_set(physid, __physid_mask);				\
-		__physid_mask;							\
+#define physid_mask_of_physid(physid)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		physid_set(physid, __physid_mask);			\
+		__physid_mask;						\
 	})
 
 #define PHYSID_MASK_ALL		{ {[0 ... PHYSID_ARRAY_SIZE-1] = ~0UL} }
diff --git a/include/asm-x86/mpspec_64.h b/include/asm-x86/mpspec_64.h
index 017fddb61dc5..16eab20667cd 100644
--- a/include/asm-x86/mpspec_64.h
+++ b/include/asm-x86/mpspec_64.h
@@ -1,189 +1,35 @@
 #ifndef __ASM_MPSPEC_H
 #define __ASM_MPSPEC_H
 
-/*
- * Structure definitions for SMP machines following the
- * Intel Multiprocessing Specification 1.1 and 1.4.
- */
-
-/*
- * This tag identifies where the SMP configuration
- * information is. 
- */
- 
-#define SMP_MAGIC_IDENT	(('_'<<24)|('P'<<16)|('M'<<8)|'_')
-
-/*
- * A maximum of 255 APICs with the current APIC ID architecture.
- */
-#define MAX_APICS 255
-
-struct intel_mp_floating
-{
-	char mpf_signature[4];		/* "_MP_" 			*/
-	unsigned int mpf_physptr;	/* Configuration table address	*/
-	unsigned char mpf_length;	/* Our length (paragraphs)	*/
-	unsigned char mpf_specification;/* Specification version	*/
-	unsigned char mpf_checksum;	/* Checksum (makes sum 0)	*/
-	unsigned char mpf_feature1;	/* Standard or configuration ? 	*/
-	unsigned char mpf_feature2;	/* Bit7 set for IMCR|PIC	*/
-	unsigned char mpf_feature3;	/* Unused (0)			*/
-	unsigned char mpf_feature4;	/* Unused (0)			*/
-	unsigned char mpf_feature5;	/* Unused (0)			*/
-};
-
-struct mp_config_table
-{
-	char mpc_signature[4];
-#define MPC_SIGNATURE "PCMP"
-	unsigned short mpc_length;	/* Size of table */
-	char  mpc_spec;			/* 0x01 */
-	char  mpc_checksum;
-	char  mpc_oem[8];
-	char  mpc_productid[12];
-	unsigned int mpc_oemptr;	/* 0 if not present */
-	unsigned short mpc_oemsize;	/* 0 if not present */
-	unsigned short mpc_oemcount;
-	unsigned int mpc_lapic;	/* APIC address */
-	unsigned int reserved;
-};
-
-/* Followed by entries */
-
-#define	MP_PROCESSOR	0
-#define	MP_BUS		1
-#define	MP_IOAPIC	2
-#define	MP_INTSRC	3
-#define	MP_LINTSRC	4
-
-struct mpc_config_processor
-{
-	unsigned char mpc_type;
-	unsigned char mpc_apicid;	/* Local APIC number */
-	unsigned char mpc_apicver;	/* Its versions */
-	unsigned char mpc_cpuflag;
-#define CPU_ENABLED		1	/* Processor is available */
-#define CPU_BOOTPROCESSOR	2	/* Processor is the BP */
-	unsigned int mpc_cpufeature;		
-#define CPU_STEPPING_MASK 0x0F
-#define CPU_MODEL_MASK	0xF0
-#define CPU_FAMILY_MASK	0xF00
-	unsigned int mpc_featureflag;	/* CPUID feature value */
-	unsigned int mpc_reserved[2];
-};
-
-struct mpc_config_bus
-{
-	unsigned char mpc_type;
-	unsigned char mpc_busid;
-	unsigned char mpc_bustype[6];
-};
-
-/* List of Bus Type string values, Intel MP Spec. */
-#define BUSTYPE_EISA	"EISA"
-#define BUSTYPE_ISA	"ISA"
-#define BUSTYPE_INTERN	"INTERN"	/* Internal BUS */
-#define BUSTYPE_MCA	"MCA"
-#define BUSTYPE_VL	"VL"		/* Local bus */
-#define BUSTYPE_PCI	"PCI"
-#define BUSTYPE_PCMCIA	"PCMCIA"
-#define BUSTYPE_CBUS	"CBUS"
-#define BUSTYPE_CBUSII	"CBUSII"
-#define BUSTYPE_FUTURE	"FUTURE"
-#define BUSTYPE_MBI	"MBI"
-#define BUSTYPE_MBII	"MBII"
-#define BUSTYPE_MPI	"MPI"
-#define BUSTYPE_MPSA	"MPSA"
-#define BUSTYPE_NUBUS	"NUBUS"
-#define BUSTYPE_TC	"TC"
-#define BUSTYPE_VME	"VME"
-#define BUSTYPE_XPRESS	"XPRESS"
-
-struct mpc_config_ioapic
-{
-	unsigned char mpc_type;
-	unsigned char mpc_apicid;
-	unsigned char mpc_apicver;
-	unsigned char mpc_flags;
-#define MPC_APIC_USABLE		0x01
-	unsigned int mpc_apicaddr;
-};
-
-struct mpc_config_intsrc
-{
-	unsigned char mpc_type;
-	unsigned char mpc_irqtype;
-	unsigned short mpc_irqflag;
-	unsigned char mpc_srcbus;
-	unsigned char mpc_srcbusirq;
-	unsigned char mpc_dstapic;
-	unsigned char mpc_dstirq;
-};
-
-enum mp_irq_source_types {
-	mp_INT = 0,
-	mp_NMI = 1,
-	mp_SMI = 2,
-	mp_ExtINT = 3
-};
-
-#define MP_IRQDIR_DEFAULT	0
-#define MP_IRQDIR_HIGH		1
-#define MP_IRQDIR_LOW		3
-
-
-struct mpc_config_lintsrc
-{
-	unsigned char mpc_type;
-	unsigned char mpc_irqtype;
-	unsigned short mpc_irqflag;
-	unsigned char mpc_srcbusid;
-	unsigned char mpc_srcbusirq;
-	unsigned char mpc_destapic;	
-#define MP_APIC_ALL	0xFF
-	unsigned char mpc_destapiclint;
-};
-
-/*
- *	Default configurations
- *
- *	1	2 CPU ISA 82489DX
- *	2	2 CPU EISA 82489DX neither IRQ 0 timer nor IRQ 13 DMA chaining
- *	3	2 CPU EISA 82489DX
- *	4	2 CPU MCA 82489DX
- *	5	2 CPU ISA+PCI
- *	6	2 CPU EISA+PCI
- *	7	2 CPU MCA+PCI
- */
+#include <asm/mpspec_def.h>
 
 #define MAX_MP_BUSSES 256
 /* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
 #define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
 extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
-extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
+
+extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
 extern int smp_found_config;
-extern void find_smp_config (void);
-extern void get_smp_config (void);
 extern int nr_ioapics;
-extern unsigned char apic_version [MAX_APICS];
 extern int mp_irq_entries;
-extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
+extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 extern int mpc_default_type;
 extern unsigned long mp_lapic_addr;
 
+extern void find_smp_config (void);
+extern void get_smp_config (void);
+
 #ifdef CONFIG_ACPI
 extern void mp_register_lapic (u8 id, u8 enabled);
 extern void mp_register_lapic_address (u64 address);
-
 extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
-extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger, u32 gsi);
+extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger,
+				    u32 gsi);
 extern void mp_config_acpi_legacy_irqs (void);
-extern int mp_register_gsi (u32 gsi, int triggering, int polarity);
-#endif
-
-extern int using_apic_timer;
+extern int mp_register_gsi (u32 gsi, int edge_level, int active_high_low);
+#endif /* CONFIG_ACPI */
 
 #define PHYSID_ARRAY_SIZE	BITS_TO_LONGS(MAX_APICS)
 
@@ -210,18 +56,18 @@ typedef struct physid_mask physid_mask_t;
 #define physids_shift_left(d, s, n)		bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
 #define physids_coerce(map)			((map).mask[0])
 
-#define physids_promote(physids)						\
-	({									\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;			\
-		__physid_mask.mask[0] = physids;				\
-		__physid_mask;							\
+#define physids_promote(physids)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		__physid_mask.mask[0] = physids;			\
+		__physid_mask;						\
 	})
 
-#define physid_mask_of_physid(physid)						\
-	({									\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;			\
-		physid_set(physid, __physid_mask);				\
-		__physid_mask;							\
+#define physid_mask_of_physid(physid)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		physid_set(physid, __physid_mask);			\
+		__physid_mask;						\
 	})
 
 #define PHYSID_MASK_ALL		{ {[0 ... PHYSID_ARRAY_SIZE-1] = ~0UL} }
diff --git a/include/asm-x86/mpspec_def.h b/include/asm-x86/mpspec_def.h
index 13bafb16e7af..3504617fe648 100644
--- a/include/asm-x86/mpspec_def.h
+++ b/include/asm-x86/mpspec_def.h
@@ -8,52 +8,68 @@
 
 /*
  * This tag identifies where the SMP configuration
- * information is. 
+ * information is.
  */
- 
+
 #define SMP_MAGIC_IDENT	(('_'<<24)|('P'<<16)|('M'<<8)|'_')
 
-#define MAX_MPC_ENTRY 1024
-#define MAX_APICS      256
+#ifdef CONFIG_X86_32
+# define MAX_MPC_ENTRY 1024
+# define MAX_APICS      256
+#else
+/*
+ * A maximum of 255 APICs with the current APIC ID architecture.
+ */
+# define MAX_APICS 255
+#endif
 
 struct intel_mp_floating
 {
-	char mpf_signature[4];		/* "_MP_" 			*/
-	unsigned long mpf_physptr;	/* Configuration table address	*/
+	char mpf_signature[4];		/* "_MP_"			*/
+	unsigned int mpf_physptr;	/* Configuration table address	*/
 	unsigned char mpf_length;	/* Our length (paragraphs)	*/
 	unsigned char mpf_specification;/* Specification version	*/
 	unsigned char mpf_checksum;	/* Checksum (makes sum 0)	*/
-	unsigned char mpf_feature1;	/* Standard or configuration ? 	*/
+	unsigned char mpf_feature1;	/* Standard or configuration ?	*/
 	unsigned char mpf_feature2;	/* Bit7 set for IMCR|PIC	*/
 	unsigned char mpf_feature3;	/* Unused (0)			*/
 	unsigned char mpf_feature4;	/* Unused (0)			*/
 	unsigned char mpf_feature5;	/* Unused (0)			*/
 };
 
+#define MPC_SIGNATURE "PCMP"
+
 struct mp_config_table
 {
 	char mpc_signature[4];
-#define MPC_SIGNATURE "PCMP"
 	unsigned short mpc_length;	/* Size of table */
 	char  mpc_spec;			/* 0x01 */
 	char  mpc_checksum;
 	char  mpc_oem[8];
 	char  mpc_productid[12];
-	unsigned long mpc_oemptr;	/* 0 if not present */
+	unsigned int mpc_oemptr;	/* 0 if not present */
 	unsigned short mpc_oemsize;	/* 0 if not present */
 	unsigned short mpc_oemcount;
-	unsigned long mpc_lapic;	/* APIC address */
-	unsigned long reserved;
+	unsigned int mpc_lapic;	/* APIC address */
+	unsigned int reserved;
 };
 
 /* Followed by entries */
 
-#define	MP_PROCESSOR	0
-#define	MP_BUS		1
-#define	MP_IOAPIC	2
-#define	MP_INTSRC	3
-#define	MP_LINTSRC	4
-#define	MP_TRANSLATION  192  /* Used by IBM NUMA-Q to describe node locality */
+#define	MP_PROCESSOR		0
+#define	MP_BUS			1
+#define	MP_IOAPIC		2
+#define	MP_INTSRC		3
+#define	MP_LINTSRC		4
+/* Used by IBM NUMA-Q to describe node locality */
+#define	MP_TRANSLATION		192
+
+#define CPU_ENABLED		1	/* Processor is available */
+#define CPU_BOOTPROCESSOR	2	/* Processor is the BP */
+
+#define CPU_STEPPING_MASK	0x000F
+#define CPU_MODEL_MASK		0x00F0
+#define CPU_FAMILY_MASK		0x0F00
 
 struct mpc_config_processor
 {
@@ -61,14 +77,9 @@ struct mpc_config_processor
 	unsigned char mpc_apicid;	/* Local APIC number */
 	unsigned char mpc_apicver;	/* Its versions */
 	unsigned char mpc_cpuflag;
-#define CPU_ENABLED		1	/* Processor is available */
-#define CPU_BOOTPROCESSOR	2	/* Processor is the BP */
-	unsigned long mpc_cpufeature;		
-#define CPU_STEPPING_MASK 0x0F
-#define CPU_MODEL_MASK	0xF0
-#define CPU_FAMILY_MASK	0xF00
-	unsigned long mpc_featureflag;	/* CPUID feature value */
-	unsigned long mpc_reserved[2];
+	unsigned int mpc_cpufeature;
+	unsigned int mpc_featureflag;	/* CPUID feature value */
+	unsigned int mpc_reserved[2];
 };
 
 struct mpc_config_bus
@@ -98,14 +109,15 @@ struct mpc_config_bus
 #define BUSTYPE_VME	"VME"
 #define BUSTYPE_XPRESS	"XPRESS"
 
+#define MPC_APIC_USABLE		0x01
+
 struct mpc_config_ioapic
 {
 	unsigned char mpc_type;
 	unsigned char mpc_apicid;
 	unsigned char mpc_apicver;
 	unsigned char mpc_flags;
-#define MPC_APIC_USABLE		0x01
-	unsigned long mpc_apicaddr;
+	unsigned int mpc_apicaddr;
 };
 
 struct mpc_config_intsrc
@@ -130,6 +142,7 @@ enum mp_irq_source_types {
 #define MP_IRQDIR_HIGH		1
 #define MP_IRQDIR_LOW		3
 
+#define MP_APIC_ALL	0xFF
 
 struct mpc_config_lintsrc
 {
@@ -138,15 +151,15 @@ struct mpc_config_lintsrc
 	unsigned short mpc_irqflag;
 	unsigned char mpc_srcbusid;
 	unsigned char mpc_srcbusirq;
-	unsigned char mpc_destapic;	
-#define MP_APIC_ALL	0xFF
+	unsigned char mpc_destapic;
 	unsigned char mpc_destapiclint;
 };
 
+#define MPC_OEM_SIGNATURE "_OEM"
+
 struct mp_config_oemtable
 {
 	char oem_signature[4];
-#define MPC_OEM_SIGNATURE "_OEM"
 	unsigned short oem_length;	/* Size of table */
 	char  oem_rev;			/* 0x01 */
 	char  oem_checksum;
@@ -155,13 +168,13 @@ struct mp_config_oemtable
 
 struct mpc_config_translation
 {
-        unsigned char mpc_type;
-        unsigned char trans_len;
-        unsigned char trans_type;
-        unsigned char trans_quad;
-        unsigned char trans_global;
-        unsigned char trans_local;
-        unsigned short trans_reserved;
+	unsigned char mpc_type;
+	unsigned char trans_len;
+	unsigned char trans_type;
+	unsigned char trans_quad;
+	unsigned char trans_global;
+	unsigned char trans_local;
+	unsigned short trans_reserved;
 };
 
 /*
-- 
cgit v1.2.3


From c2805aa1d8ae51c7582d2ccbd736afa545cf5cc4 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:35 +0100
Subject: x86: merge mpspec variants

The delta is now minimal. Merge them

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/mpspec.h    | 113 +++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/mpspec_32.h |  84 --------------------------------
 include/asm-x86/mpspec_64.h |  79 -------------------------------
 3 files changed, 111 insertions(+), 165 deletions(-)
 delete mode 100644 include/asm-x86/mpspec_32.h
 delete mode 100644 include/asm-x86/mpspec_64.h

(limited to 'include')

diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 8f268e8fd2e9..a2a6b2ea4259 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -1,5 +1,114 @@
+#ifndef _AM_X86_MPSPEC_H
+#define _AM_X86_MPSPEC_H
+
+#include <asm/mpspec_def.h>
+
 #ifdef CONFIG_X86_32
-# include "mpspec_32.h"
+#include <mach_mpspec.h>
+
+extern int mp_bus_id_to_type[MAX_MP_BUSSES];
+extern int mp_bus_id_to_node[MAX_MP_BUSSES];
+extern int mp_bus_id_to_local[MAX_MP_BUSSES];
+extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
+
+extern unsigned int def_to_bigsmp;
+extern int apic_version[MAX_APICS];
+extern int pic_mode;
+
 #else
-# include "mpspec_64.h"
+
+#define MAX_MP_BUSSES 256
+/* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
+#define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
+
+extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
+
+#endif
+
+extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
+
+extern unsigned int boot_cpu_physical_apicid;
+extern int smp_found_config;
+extern int nr_ioapics;
+extern int mp_irq_entries;
+extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+extern int mpc_default_type;
+extern unsigned long mp_lapic_addr;
+
+extern void find_smp_config(void);
+extern void get_smp_config(void);
+
+#ifdef CONFIG_ACPI
+extern void mp_register_lapic(u8 id, u8 enabled);
+extern void mp_register_lapic_address(u64 address);
+extern void mp_register_ioapic(u8 id, u32 address, u32 gsi_base);
+extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger,
+				   u32 gsi);
+extern void mp_config_acpi_legacy_irqs(void);
+extern int mp_register_gsi(u32 gsi, int edge_level, int active_high_low);
+#endif /* CONFIG_ACPI */
+
+#define PHYSID_ARRAY_SIZE	BITS_TO_LONGS(MAX_APICS)
+
+struct physid_mask
+{
+	unsigned long mask[PHYSID_ARRAY_SIZE];
+};
+
+typedef struct physid_mask physid_mask_t;
+
+#define physid_set(physid, map)			set_bit(physid, (map).mask)
+#define physid_clear(physid, map)		clear_bit(physid, (map).mask)
+#define physid_isset(physid, map)		test_bit(physid, (map).mask)
+#define physid_test_and_set(physid, map) \
+	test_and_set_bit(physid, (map).mask)
+
+#define physids_and(dst, src1, src2) \
+	bitmap_and((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
+
+#define physids_or(dst, src1, src2) \
+	bitmap_or((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
+
+#define physids_clear(map) \
+	bitmap_zero((map).mask, MAX_APICS)
+
+#define physids_complement(dst, src) \
+	bitmap_complement((dst).mask, (src).mask, MAX_APICS)
+
+#define physids_empty(map) \
+	bitmap_empty((map).mask, MAX_APICS)
+
+#define physids_equal(map1, map2) \
+	bitmap_equal((map1).mask, (map2).mask, MAX_APICS)
+
+#define physids_weight(map) \
+	bitmap_weight((map).mask, MAX_APICS)
+
+#define physids_shift_right(d, s, n) \
+	bitmap_shift_right((d).mask, (s).mask, n, MAX_APICS)
+
+#define physids_shift_left(d, s, n) \
+	bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
+
+#define physids_coerce(map)			((map).mask[0])
+
+#define physids_promote(physids)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		__physid_mask.mask[0] = physids;			\
+		__physid_mask;						\
+	})
+
+#define physid_mask_of_physid(physid)					\
+	({								\
+		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
+		physid_set(physid, __physid_mask);			\
+		__physid_mask;						\
+	})
+
+#define PHYSID_MASK_ALL		{ {[0 ... PHYSID_ARRAY_SIZE-1] = ~0UL} }
+#define PHYSID_MASK_NONE	{ {[0 ... PHYSID_ARRAY_SIZE-1] = 0UL} }
+
+extern physid_mask_t phys_cpu_present_map;
+
 #endif
diff --git a/include/asm-x86/mpspec_32.h b/include/asm-x86/mpspec_32.h
deleted file mode 100644
index bb73185e6a7d..000000000000
--- a/include/asm-x86/mpspec_32.h
+++ /dev/null
@@ -1,84 +0,0 @@
-#ifndef __ASM_MPSPEC_H
-#define __ASM_MPSPEC_H
-
-#include <asm/mpspec_def.h>
-#include <mach_mpspec.h>
-
-extern int mp_bus_id_to_type[MAX_MP_BUSSES];
-extern int mp_bus_id_to_node[MAX_MP_BUSSES];
-extern int mp_bus_id_to_local[MAX_MP_BUSSES];
-extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
-
-extern unsigned int def_to_bigsmp;
-extern int apic_version[MAX_APICS];
-extern int pic_mode;
-
-extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
-
-extern unsigned int boot_cpu_physical_apicid;
-extern int smp_found_config;
-extern int nr_ioapics;
-extern int mp_irq_entries;
-extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
-extern int mpc_default_type;
-extern unsigned long mp_lapic_addr;
-
-extern void find_smp_config (void);
-extern void get_smp_config (void);
-
-#ifdef CONFIG_ACPI
-extern void mp_register_lapic (u8 id, u8 enabled);
-extern void mp_register_lapic_address (u64 address);
-extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
-extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger,
-				    u32 gsi);
-extern void mp_config_acpi_legacy_irqs (void);
-extern int mp_register_gsi (u32 gsi, int edge_level, int active_high_low);
-#endif /* CONFIG_ACPI */
-
-#define PHYSID_ARRAY_SIZE	BITS_TO_LONGS(MAX_APICS)
-
-struct physid_mask
-{
-	unsigned long mask[PHYSID_ARRAY_SIZE];
-};
-
-typedef struct physid_mask physid_mask_t;
-
-#define physid_set(physid, map)			set_bit(physid, (map).mask)
-#define physid_clear(physid, map)		clear_bit(physid, (map).mask)
-#define physid_isset(physid, map)		test_bit(physid, (map).mask)
-#define physid_test_and_set(physid, map)	test_and_set_bit(physid, (map).mask)
-
-#define physids_and(dst, src1, src2)		bitmap_and((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
-#define physids_or(dst, src1, src2)		bitmap_or((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
-#define physids_clear(map)			bitmap_zero((map).mask, MAX_APICS)
-#define physids_complement(dst, src)		bitmap_complement((dst).mask, (src).mask, MAX_APICS)
-#define physids_empty(map)			bitmap_empty((map).mask, MAX_APICS)
-#define physids_equal(map1, map2)		bitmap_equal((map1).mask, (map2).mask, MAX_APICS)
-#define physids_weight(map)			bitmap_weight((map).mask, MAX_APICS)
-#define physids_shift_right(d, s, n)		bitmap_shift_right((d).mask, (s).mask, n, MAX_APICS)
-#define physids_shift_left(d, s, n)		bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
-#define physids_coerce(map)			((map).mask[0])
-
-#define physids_promote(physids)					\
-	({								\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
-		__physid_mask.mask[0] = physids;			\
-		__physid_mask;						\
-	})
-
-#define physid_mask_of_physid(physid)					\
-	({								\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
-		physid_set(physid, __physid_mask);			\
-		__physid_mask;						\
-	})
-
-#define PHYSID_MASK_ALL		{ {[0 ... PHYSID_ARRAY_SIZE-1] = ~0UL} }
-#define PHYSID_MASK_NONE	{ {[0 ... PHYSID_ARRAY_SIZE-1] = 0UL} }
-
-extern physid_mask_t phys_cpu_present_map;
-
-#endif
-
diff --git a/include/asm-x86/mpspec_64.h b/include/asm-x86/mpspec_64.h
deleted file mode 100644
index 16eab20667cd..000000000000
--- a/include/asm-x86/mpspec_64.h
+++ /dev/null
@@ -1,79 +0,0 @@
-#ifndef __ASM_MPSPEC_H
-#define __ASM_MPSPEC_H
-
-#include <asm/mpspec_def.h>
-
-#define MAX_MP_BUSSES 256
-/* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
-#define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
-extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
-
-extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
-
-extern unsigned int boot_cpu_physical_apicid;
-extern int smp_found_config;
-extern int nr_ioapics;
-extern int mp_irq_entries;
-extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
-extern int mpc_default_type;
-extern unsigned long mp_lapic_addr;
-
-extern void find_smp_config (void);
-extern void get_smp_config (void);
-
-#ifdef CONFIG_ACPI
-extern void mp_register_lapic (u8 id, u8 enabled);
-extern void mp_register_lapic_address (u64 address);
-extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
-extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger,
-				    u32 gsi);
-extern void mp_config_acpi_legacy_irqs (void);
-extern int mp_register_gsi (u32 gsi, int edge_level, int active_high_low);
-#endif /* CONFIG_ACPI */
-
-#define PHYSID_ARRAY_SIZE	BITS_TO_LONGS(MAX_APICS)
-
-struct physid_mask
-{
-	unsigned long mask[PHYSID_ARRAY_SIZE];
-};
-
-typedef struct physid_mask physid_mask_t;
-
-#define physid_set(physid, map)			set_bit(physid, (map).mask)
-#define physid_clear(physid, map)		clear_bit(physid, (map).mask)
-#define physid_isset(physid, map)		test_bit(physid, (map).mask)
-#define physid_test_and_set(physid, map)	test_and_set_bit(physid, (map).mask)
-
-#define physids_and(dst, src1, src2)		bitmap_and((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
-#define physids_or(dst, src1, src2)		bitmap_or((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
-#define physids_clear(map)			bitmap_zero((map).mask, MAX_APICS)
-#define physids_complement(dst, src)		bitmap_complement((dst).mask, (src).mask, MAX_APICS)
-#define physids_empty(map)			bitmap_empty((map).mask, MAX_APICS)
-#define physids_equal(map1, map2)		bitmap_equal((map1).mask, (map2).mask, MAX_APICS)
-#define physids_weight(map)			bitmap_weight((map).mask, MAX_APICS)
-#define physids_shift_right(d, s, n)		bitmap_shift_right((d).mask, (s).mask, n, MAX_APICS)
-#define physids_shift_left(d, s, n)		bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
-#define physids_coerce(map)			((map).mask[0])
-
-#define physids_promote(physids)					\
-	({								\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
-		__physid_mask.mask[0] = physids;			\
-		__physid_mask;						\
-	})
-
-#define physid_mask_of_physid(physid)					\
-	({								\
-		physid_mask_t __physid_mask = PHYSID_MASK_NONE;		\
-		physid_set(physid, __physid_mask);			\
-		__physid_mask;						\
-	})
-
-#define PHYSID_MASK_ALL		{ {[0 ... PHYSID_ARRAY_SIZE-1] = ~0UL} }
-#define PHYSID_MASK_NONE	{ {[0 ... PHYSID_ARRAY_SIZE-1] = 0UL} }
-
-extern physid_mask_t phys_cpu_present_map;
-
-#endif
-
-- 
cgit v1.2.3


From ae9d983be1eefac4b5efad69a188e7ac89a75797 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:36 +0100
Subject: x86: cleanup smp.h variants

Bring the smp.h variants into sync to prepare merging and
paravirt support.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/apic.h        |   1 +
 include/asm-x86/mpspec.h      |   3 +
 include/asm-x86/smp_32.h      | 117 ++++++++++++++++---------------------
 include/asm-x86/smp_64.h      | 133 ++++++++++++++++--------------------------
 include/asm-x86/topology_64.h |   2 -
 5 files changed, 104 insertions(+), 152 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index d0a221fa1fc3..18d932dff476 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -42,6 +42,7 @@ extern int local_apic_timer_disabled;
 
 extern int apic_runs_main_timer;
 extern int ioapic_force;
+extern int disable_apic;
 extern int disable_apic_timer;
 extern unsigned boot_cpu_id;
 
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index a2a6b2ea4259..781ad74ab9e9 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -13,8 +13,11 @@ extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
 
 extern unsigned int def_to_bigsmp;
 extern int apic_version[MAX_APICS];
+extern u8 apicid_2_node[];
 extern int pic_mode;
 
+#define MAX_APICID 256
+
 #else
 
 #define MAX_MP_BUSSES 256
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index e10b7affdfe5..c69e960429cc 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -1,51 +1,41 @@
 #ifndef __ASM_SMP_H
 #define __ASM_SMP_H
 
+#ifndef __ASSEMBLY__
+#include <linux/cpumask.h>
+#include <linux/init.h>
+
 /*
  * We need the APIC definitions automatically as part of 'smp.h'
  */
-#ifndef __ASSEMBLY__
-#include <linux/kernel.h>
-#include <linux/threads.h>
-#include <linux/cpumask.h>
+#ifdef CONFIG_X86_LOCAL_APIC
+# include <asm/mpspec.h>
+# include <asm/apic.h>
+# ifdef CONFIG_X86_IO_APIC
+#  include <asm/io_apic.h>
+# endif
 #endif
 
-#if defined(CONFIG_X86_LOCAL_APIC) && !defined(__ASSEMBLY__)
-#include <linux/bitops.h>
-#include <asm/mpspec.h>
-#include <asm/apic.h>
-#ifdef CONFIG_X86_IO_APIC
-#include <asm/io_apic.h>
-#endif
-#endif
+extern cpumask_t cpu_callout_map;
+extern cpumask_t cpu_callin_map;
 
-#define BAD_APICID 0xFFu
-#ifdef CONFIG_SMP
-#ifndef __ASSEMBLY__
+extern int smp_num_siblings;
+extern unsigned int num_processors;
 
-/*
- * Private routines/data
- */
- 
 extern void smp_alloc_memory(void);
-extern int pic_mode;
-extern int smp_num_siblings;
-DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
-DECLARE_PER_CPU(cpumask_t, cpu_core_map);
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
 
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
-extern void lock_ipi_call_lock(void);
-extern void unlock_ipi_call_lock(void);
 
-#define MAX_APICID 256
 extern u8 __initdata x86_cpu_to_apicid_init[];
 extern void *x86_cpu_to_apicid_ptr;
-DECLARE_PER_CPU(u8, x86_cpu_to_apicid);
-
-#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
-extern void set_cpu_sibling_map(int cpu);
+DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
+DECLARE_PER_CPU(cpumask_t, cpu_core_map);
+DECLARE_PER_CPU(u8, cpu_llc_id);
+DECLARE_PER_CPU(u8, x86_cpu_to_apicid);
 
 #ifdef CONFIG_HOTPLUG_CPU
 extern void cpu_exit_clear(void);
@@ -53,6 +43,9 @@ extern void cpu_uninit(void);
 extern void remove_siblinginfo(int cpu);
 #endif
 
+/* Globals due to paravirt */
+extern void set_cpu_sibling_map(int cpu);
+
 struct smp_ops
 {
 	void (*smp_prepare_boot_cpu)(void);
@@ -67,6 +60,7 @@ struct smp_ops
 				      int wait);
 };
 
+#ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
 static inline void smp_prepare_boot_cpu(void)
@@ -107,10 +101,12 @@ int native_cpu_up(unsigned int cpunum);
 void native_smp_cpus_done(unsigned int max_cpus);
 
 #ifndef CONFIG_PARAVIRT
-#define startup_ipi_hook(phys_apicid, start_eip, start_esp) 		\
-do { } while (0)
+#define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
 #endif
 
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
+
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
  * from the initial startup. We map APIC_BASE very early in page_setup(),
@@ -119,9 +115,11 @@ do { } while (0)
 DECLARE_PER_CPU(int, cpu_number);
 #define raw_smp_processor_id() (x86_read_percpu(cpu_number))
 
-extern cpumask_t cpu_callout_map;
-extern cpumask_t cpu_callin_map;
-extern cpumask_t cpu_possible_map;
+#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
+
+extern int safe_smp_processor_id(void);
+
+void __cpuinit smp_store_cpu_info(int id);
 
 /* We don't mark CPUs online until __cpu_up(), so we need another measure */
 static inline int num_booting_cpus(void)
@@ -129,56 +127,39 @@ static inline int num_booting_cpus(void)
 	return cpus_weight(cpu_callout_map);
 }
 
-extern int safe_smp_processor_id(void);
-extern int __cpu_disable(void);
-extern void __cpu_die(unsigned int cpu);
-extern unsigned int num_processors;
-
-void __cpuinit smp_store_cpu_info(int id);
-
-#endif /* !__ASSEMBLY__ */
-
 #else /* CONFIG_SMP */
 
 #define safe_smp_processor_id()		0
 #define cpu_physical_id(cpu)		boot_cpu_physical_apicid
 
-#define NO_PROC_ID		0xFF		/* No processor magic marker */
-
-#endif /* CONFIG_SMP */
-
-#ifndef __ASSEMBLY__
+#endif /* !CONFIG_SMP */
 
 #ifdef CONFIG_X86_LOCAL_APIC
 
-#ifdef APIC_DEFINITION
+static __inline int logical_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
+}
+
+# ifdef APIC_DEFINITION
 extern int hard_smp_processor_id(void);
-#else
-#include <mach_apicdef.h>
+# else
+#  include <mach_apicdef.h>
 static inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(*(unsigned long *)(APIC_BASE+APIC_ID));
+	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
 }
-#endif /* APIC_DEFINITION */
+# endif /* APIC_DEFINITION */
 
 #else /* CONFIG_X86_LOCAL_APIC */
 
-#ifndef CONFIG_SMP
-#define hard_smp_processor_id()		0
-#endif
+# ifndef CONFIG_SMP
+#  define hard_smp_processor_id()	0
+# endif
 
 #endif /* CONFIG_X86_LOCAL_APIC */
 
-extern u8 apicid_2_node[];
-
-#ifdef CONFIG_X86_LOCAL_APIC
-static __inline int logical_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
-}
-#endif
-#endif
-
+#endif /* !ASSEMBLY */
 #endif
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index ab612b0ff270..2feddda91e12 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -1,130 +1,99 @@
 #ifndef __ASM_SMP_H
 #define __ASM_SMP_H
 
-/*
- * We need the APIC definitions automatically as part of 'smp.h'
- */
-#include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <linux/bitops.h>
 #include <linux/init.h>
-extern int disable_apic;
 
-#include <asm/mpspec.h>
+/*
+ * We need the APIC definitions automatically as part of 'smp.h'
+ */
 #include <asm/apic.h>
 #include <asm/io_apic.h>
-#include <asm/thread_info.h>
-
-#ifdef CONFIG_SMP
-
+#include <asm/mpspec.h>
 #include <asm/pda.h>
+#include <asm/thread_info.h>
 
-struct pt_regs;
-
-extern cpumask_t cpu_present_mask;
-extern cpumask_t cpu_possible_map;
-extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_callout_map;
 extern cpumask_t cpu_initialized;
 
-/*
- * Private routines/data
- */
- 
+extern int smp_num_siblings;
+extern unsigned int num_processors;
+
 extern void smp_alloc_memory(void);
-extern volatile unsigned long smp_invalidate_needed;
 extern void lock_ipi_call_lock(void);
 extern void unlock_ipi_call_lock(void);
-extern int smp_num_siblings;
-extern void smp_send_reschedule(int cpu);
+
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-/*
- * cpu_sibling_map and cpu_core_map now live
- * in the per cpu area
- *
- * extern cpumask_t cpu_sibling_map[NR_CPUS];
- * extern cpumask_t cpu_core_map[NR_CPUS];
- */
+extern u8 __initdata x86_cpu_to_apicid_init[];
+extern void *x86_cpu_to_apicid_ptr;
+extern u8 bios_cpu_apicid[];
+
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
 DECLARE_PER_CPU(u8, cpu_llc_id);
+DECLARE_PER_CPU(u8, x86_cpu_to_apicid);
 
-#define SMP_TRAMPOLINE_BASE 0x6000
-
-/*
- * On x86 all CPUs are mapped 1:1 to the APIC space.
- * This simplifies scheduling and IPI sending and
- * compresses data structures.
- */
-
-static inline int num_booting_cpus(void)
+static inline int cpu_present_to_apicid(int mps_cpu)
 {
-	return cpus_weight(cpu_callout_map);
+	if (mps_cpu < NR_CPUS)
+		return (int)bios_cpu_apicid[mps_cpu];
+	else
+		return BAD_APICID;
 }
 
-#define raw_smp_processor_id() read_pda(cpunumber)
+#ifdef CONFIG_SMP
+
+#define SMP_TRAMPOLINE_BASE 0x6000
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 extern void prefill_possible_map(void);
-extern unsigned num_processors;
 extern unsigned __cpuinitdata disabled_cpus;
 
-#define NO_PROC_ID		0xFF		/* No processor magic marker */
-
-#endif /* CONFIG_SMP */
-
-#define safe_smp_processor_id()		smp_processor_id()
+#define raw_smp_processor_id()	read_pda(cpunumber)
+#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
-static inline int hard_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
-}
+#define stack_smp_processor_id()					\
+	({								\
+	struct thread_info *ti;						\
+	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
+	ti->cpu;							\
+})
 
 /*
- * Some lowlevel functions might want to know about
- * the real APIC ID <-> CPU # mapping.
+ * On x86 all CPUs are mapped 1:1 to the APIC space. This simplifies
+ * scheduling and IPI sending and compresses data structures.
  */
-extern u8 __initdata x86_cpu_to_apicid_init[];
-extern void *x86_cpu_to_apicid_ptr;
-DECLARE_PER_CPU(u8, x86_cpu_to_apicid);	/* physical ID */
-extern u8 bios_cpu_apicid[];
-
-static inline int cpu_present_to_apicid(int mps_cpu)
+static inline int num_booting_cpus(void)
 {
-	if (mps_cpu < NR_CPUS)
-		return (int)bios_cpu_apicid[mps_cpu];
-	else
-		return BAD_APICID;
+	return cpus_weight(cpu_callout_map);
 }
 
-#ifndef CONFIG_SMP
+extern void smp_send_reschedule(int cpu);
+
+#else /* CONFIG_SMP */
+
+extern unsigned int boot_cpu_id;
+#define cpu_physical_id(cpu)	boot_cpu_id
 #define stack_smp_processor_id() 0
-#define cpu_logical_map(x) (x)
-#else
-#include <asm/thread_info.h>
-#define stack_smp_processor_id() \
-({ 								\
-	struct thread_info *ti;					\
-	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
-	ti->cpu;						\
-})
-#endif
+
+#endif /* !CONFIG_SMP */
+
+#define safe_smp_processor_id()		smp_processor_id()
 
 static __inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
+	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
+}
+
+static inline int hard_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
 }
 
-#ifdef CONFIG_SMP
-#define cpu_physical_id(cpu)		per_cpu(x86_cpu_to_apicid, cpu)
-#else
-extern unsigned int boot_cpu_id;
-#define cpu_physical_id(cpu)		boot_cpu_id
-#endif /* !CONFIG_SMP */
 #endif
 
diff --git a/include/asm-x86/topology_64.h b/include/asm-x86/topology_64.h
index a718dda037e0..407b22d4e3b0 100644
--- a/include/asm-x86/topology_64.h
+++ b/include/asm-x86/topology_64.h
@@ -7,8 +7,6 @@
 #include <asm/mpspec.h>
 #include <linux/bitops.h>
 
-extern cpumask_t cpu_online_map;
-
 extern unsigned char cpu_to_node[];
 extern cpumask_t     node_to_cpumask[];
 
-- 
cgit v1.2.3


From 8a999410135602d98bd007e835dc21c0c279ad72 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:36 +0100
Subject: x86: cleanup acpi_32/64.h

Fix coding style to get a readable diff

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/acpi_32.h | 35 ++++++++++++++---------------------
 include/asm-x86/acpi_64.h | 41 +++++++++++++++++------------------------
 2 files changed, 31 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/acpi_32.h b/include/asm-x86/acpi_32.h
index c8afeea0b80b..03612bcf6422 100644
--- a/include/asm-x86/acpi_32.h
+++ b/include/asm-x86/acpi_32.h
@@ -24,11 +24,8 @@
 #ifndef _ASM_ACPI_H
 #define _ASM_ACPI_H
 
-#ifdef __KERNEL__
-
 #include <acpi/pdc_intel.h>
-
-#include <asm/system.h>		/* defines cmpxchg */
+#include <asm/numa.h>
 
 #define COMPILER_DEPENDENT_INT64   long long
 #define COMPILER_DEPENDENT_UINT64  unsigned long long
@@ -67,17 +64,17 @@ int __acpi_release_global_lock(unsigned int *lock);
  * Math helper asm macros
  */
 #define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
-        asm("divl %2;"        \
-        :"=a"(q32), "=d"(r32) \
-        :"r"(d32),            \
-        "0"(n_lo), "1"(n_hi))
+	asm("divl %2;"				     \
+	    :"=a"(q32), "=d"(r32)		     \
+	    :"r"(d32),				     \
+	     "0"(n_lo), "1"(n_hi))
 
 
 #define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
-    asm("shrl   $1,%2;"             \
-        "rcrl   $1,%3;"             \
-        :"=r"(n_hi), "=r"(n_lo)     \
-        :"0"(n_hi), "1"(n_lo))
+	asm("shrl   $1,%2	;"	\
+	    "rcrl   $1,%3;"		\
+	    :"=r"(n_hi), "=r"(n_lo)	\
+	    :"0"(n_hi), "1"(n_lo))
 
 #ifdef CONFIG_ACPI
 extern int acpi_lapic;
@@ -87,6 +84,9 @@ extern int acpi_strict;
 extern int acpi_disabled;
 extern int acpi_ht;
 extern int acpi_pci_disabled;
+extern int acpi_skip_timer_override;
+extern int acpi_use_timer_override;
+
 static inline void disable_acpi(void)
 {
 	acpi_disabled = 1;
@@ -100,11 +100,6 @@ static inline void disable_acpi(void)
 
 extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
 
-#ifdef CONFIG_X86_IO_APIC
-extern int acpi_skip_timer_override;
-extern int acpi_use_timer_override;
-#endif
-
 static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
 static inline void acpi_disable_pci(void)
 {
@@ -122,7 +117,7 @@ extern unsigned long acpi_wakeup_address;
 /* early initialization routine */
 extern void acpi_reserve_bootmem(void);
 
-#else	/* !CONFIG_ACPI */
+#else /* !CONFIG_ACPI */
 
 #define acpi_lapic 0
 #define acpi_ioapic 0
@@ -130,10 +125,8 @@ static inline void acpi_noirq_set(void) { }
 static inline void acpi_disable_pci(void) { }
 static inline void disable_acpi(void) { }
 
-#endif	/* !CONFIG_ACPI */
+#endif /* !CONFIG_ACPI */
 
 #define ARCH_HAS_POWER_INIT	1
 
-#endif /*__KERNEL__*/
-
 #endif /*_ASM_ACPI_H*/
diff --git a/include/asm-x86/acpi_64.h b/include/asm-x86/acpi_64.h
index 58114ed08024..b2670c6db8ec 100644
--- a/include/asm-x86/acpi_64.h
+++ b/include/asm-x86/acpi_64.h
@@ -24,8 +24,6 @@
 #ifndef _ASM_ACPI_H
 #define _ASM_ACPI_H
 
-#ifdef __KERNEL__
-
 #include <acpi/pdc_intel.h>
 #include <asm/numa.h>
 
@@ -66,17 +64,17 @@ int __acpi_release_global_lock(unsigned int *lock);
  * Math helper asm macros
  */
 #define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
-        asm("divl %2;"        \
-        :"=a"(q32), "=d"(r32) \
-        :"r"(d32),            \
-        "0"(n_lo), "1"(n_hi))
+	asm("divl %2;"				     \
+	    :"=a"(q32), "=d"(r32)		     \
+	    :"r"(d32),				     \
+	     "0"(n_lo), "1"(n_hi))
 
 
 #define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
-    asm("shrl   $1,%2;"             \
-        "rcrl   $1,%3;"             \
-        :"=r"(n_hi), "=r"(n_lo)     \
-        :"0"(n_hi), "1"(n_lo))
+	asm("shrl   $1,%2	;"	\
+	    "rcrl   $1,%3;"		\
+	    :"=r"(n_hi), "=r"(n_lo)	\
+	    :"0"(n_hi), "1"(n_lo))
 
 #ifdef CONFIG_ACPI
 extern int acpi_lapic;
@@ -84,8 +82,11 @@ extern int acpi_ioapic;
 extern int acpi_noirq;
 extern int acpi_strict;
 extern int acpi_disabled;
-extern int acpi_pci_disabled;
 extern int acpi_ht;
+extern int acpi_pci_disabled;
+extern int acpi_skip_timer_override;
+extern int acpi_use_timer_override;
+
 static inline void disable_acpi(void)
 {
 	acpi_disabled = 1;
@@ -98,6 +99,7 @@ static inline void disable_acpi(void)
 #define FIX_ACPI_PAGES 4
 
 extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
+
 static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
 static inline void acpi_disable_pci(void)
 {
@@ -115,7 +117,7 @@ extern unsigned long acpi_wakeup_address;
 /* early initialization routine */
 extern void acpi_reserve_bootmem(void);
 
-#else	/* !CONFIG_ACPI */
+#else /* !CONFIG_ACPI */
 
 #define acpi_lapic 0
 #define acpi_ioapic 0
@@ -124,19 +126,12 @@ static inline void acpi_disable_pci(void) { }
 
 #endif /* !CONFIG_ACPI */
 
+#define ARCH_HAS_POWER_INIT	1
+
+#ifdef CONFIG_ACPI_NUMA
 extern int acpi_numa;
 extern int acpi_scan_nodes(unsigned long start, unsigned long end);
 #define NR_NODE_MEMBLKS (MAX_NUMNODES*2)
-
-extern int acpi_disabled;
-extern int acpi_pci_disabled;
-
-#define ARCH_HAS_POWER_INIT 1
-
-extern int acpi_skip_timer_override;
-extern int acpi_use_timer_override;
-
-#ifdef CONFIG_ACPI_NUMA
 extern void __init acpi_fake_nodes(const struct bootnode *fake_nodes,
 				   int num_nodes);
 #else
@@ -146,6 +141,4 @@ static inline void acpi_fake_nodes(const struct bootnode *fake_nodes,
 }
 #endif
 
-#endif /*__KERNEL__*/
-
 #endif /*_ASM_ACPI_H*/
-- 
cgit v1.2.3


From 0b80fc721bed2635c7f0b198e1c862a4596dc2cd Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:36 +0100
Subject: x86: merge acpi_32/64.h

Merge the files.

[ mingo@elte.hu: build fix ]

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/acpi.h    | 148 ++++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/acpi_32.h | 132 -----------------------------------------
 include/asm-x86/acpi_64.h | 144 --------------------------------------------
 3 files changed, 143 insertions(+), 281 deletions(-)
 delete mode 100644 include/asm-x86/acpi_32.h
 delete mode 100644 include/asm-x86/acpi_64.h

(limited to 'include')

diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h
index f8a89793ac8c..c477e5741751 100644
--- a/include/asm-x86/acpi.h
+++ b/include/asm-x86/acpi.h
@@ -1,14 +1,123 @@
 #ifndef _ASM_X86_ACPI_H
 #define _ASM_X86_ACPI_H
 
-#ifdef CONFIG_X86_32
-# include "acpi_32.h"
-#else
-# include "acpi_64.h"
-#endif
+/*
+ *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#include <acpi/pdc_intel.h>
 
+#include <asm/numa.h>
 #include <asm/processor.h>
 
+#define COMPILER_DEPENDENT_INT64   long long
+#define COMPILER_DEPENDENT_UINT64  unsigned long long
+
+/*
+ * Calling conventions:
+ *
+ * ACPI_SYSTEM_XFACE        - Interfaces to host OS (handlers, threads)
+ * ACPI_EXTERNAL_XFACE      - External ACPI interfaces
+ * ACPI_INTERNAL_XFACE      - Internal ACPI interfaces
+ * ACPI_INTERNAL_VAR_XFACE  - Internal variable-parameter list interfaces
+ */
+#define ACPI_SYSTEM_XFACE
+#define ACPI_EXTERNAL_XFACE
+#define ACPI_INTERNAL_XFACE
+#define ACPI_INTERNAL_VAR_XFACE
+
+/* Asm macros */
+
+#define ACPI_ASM_MACROS
+#define BREAKPOINT3
+#define ACPI_DISABLE_IRQS() local_irq_disable()
+#define ACPI_ENABLE_IRQS()  local_irq_enable()
+#define ACPI_FLUSH_CPU_CACHE()	wbinvd()
+
+int __acpi_acquire_global_lock(unsigned int *lock);
+int __acpi_release_global_lock(unsigned int *lock);
+
+#define ACPI_ACQUIRE_GLOBAL_LOCK(facs, Acq) \
+	((Acq) = __acpi_acquire_global_lock(&facs->global_lock))
+
+#define ACPI_RELEASE_GLOBAL_LOCK(facs, Acq) \
+	((Acq) = __acpi_release_global_lock(&facs->global_lock))
+
+/*
+ * Math helper asm macros
+ */
+#define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
+	asm("divl %2;"				     \
+	    :"=a"(q32), "=d"(r32)		     \
+	    :"r"(d32),				     \
+	     "0"(n_lo), "1"(n_hi))
+
+
+#define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
+	asm("shrl   $1,%2	;"	\
+	    "rcrl   $1,%3;"		\
+	    :"=r"(n_hi), "=r"(n_lo)	\
+	    :"0"(n_hi), "1"(n_lo))
+
+#ifdef CONFIG_ACPI
+extern int acpi_lapic;
+extern int acpi_ioapic;
+extern int acpi_noirq;
+extern int acpi_strict;
+extern int acpi_disabled;
+extern int acpi_ht;
+extern int acpi_pci_disabled;
+extern int acpi_skip_timer_override;
+extern int acpi_use_timer_override;
+
+static inline void disable_acpi(void)
+{
+	acpi_disabled = 1;
+	acpi_ht = 0;
+	acpi_pci_disabled = 1;
+	acpi_noirq = 1;
+}
+
+/* Fixmap pages to reserve for ACPI boot-time tables (see fixmap.h) */
+#define FIX_ACPI_PAGES 4
+
+extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
+
+static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
+static inline void acpi_disable_pci(void)
+{
+	acpi_pci_disabled = 1;
+	acpi_noirq_set();
+}
+extern int acpi_irq_balance_set(char *str);
+
+/* routines for saving/restoring kernel state */
+extern int acpi_save_state_mem(void);
+extern void acpi_restore_state_mem(void);
+
+extern unsigned long acpi_wakeup_address;
+
+/* early initialization routine */
+extern void acpi_reserve_bootmem(void);
+
 /*
  * Check if the CPU can handle C2 and deeper
  */
@@ -29,4 +138,33 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate)
 		return max_cstate;
 }
 
+#else /* !CONFIG_ACPI */
+
+#define acpi_lapic 0
+#define acpi_ioapic 0
+static inline void acpi_noirq_set(void) { }
+static inline void acpi_disable_pci(void) { }
+static inline void disable_acpi(void) { }
+
+#endif /* !CONFIG_ACPI */
+
+#define ARCH_HAS_POWER_INIT	1
+
+struct bootnode;
+
+#ifdef CONFIG_ACPI_NUMA
+extern int acpi_numa;
+extern int acpi_scan_nodes(unsigned long start, unsigned long end);
+#ifdef CONFIG_X86_64
+# define NR_NODE_MEMBLKS (MAX_NUMNODES*2)
 #endif
+extern void __init acpi_fake_nodes(const struct bootnode *fake_nodes,
+				   int num_nodes);
+#else
+static inline void acpi_fake_nodes(const struct bootnode *fake_nodes,
+				   int num_nodes)
+{
+}
+#endif
+
+#endif /*__X86_ASM_ACPI_H*/
diff --git a/include/asm-x86/acpi_32.h b/include/asm-x86/acpi_32.h
deleted file mode 100644
index 03612bcf6422..000000000000
--- a/include/asm-x86/acpi_32.h
+++ /dev/null
@@ -1,132 +0,0 @@
-/*
- *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
- *  Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
- *
- * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License as published by
- *  the Free Software Foundation; either version 2 of the License, or
- *  (at your option) any later version.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to the Free Software
- *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
- *
- * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- */
-
-#ifndef _ASM_ACPI_H
-#define _ASM_ACPI_H
-
-#include <acpi/pdc_intel.h>
-#include <asm/numa.h>
-
-#define COMPILER_DEPENDENT_INT64   long long
-#define COMPILER_DEPENDENT_UINT64  unsigned long long
-
-/*
- * Calling conventions:
- *
- * ACPI_SYSTEM_XFACE        - Interfaces to host OS (handlers, threads)
- * ACPI_EXTERNAL_XFACE      - External ACPI interfaces
- * ACPI_INTERNAL_XFACE      - Internal ACPI interfaces
- * ACPI_INTERNAL_VAR_XFACE  - Internal variable-parameter list interfaces
- */
-#define ACPI_SYSTEM_XFACE
-#define ACPI_EXTERNAL_XFACE
-#define ACPI_INTERNAL_XFACE
-#define ACPI_INTERNAL_VAR_XFACE
-
-/* Asm macros */
-
-#define ACPI_ASM_MACROS
-#define BREAKPOINT3
-#define ACPI_DISABLE_IRQS() local_irq_disable()
-#define ACPI_ENABLE_IRQS()  local_irq_enable()
-#define ACPI_FLUSH_CPU_CACHE()	wbinvd()
-
-int __acpi_acquire_global_lock(unsigned int *lock);
-int __acpi_release_global_lock(unsigned int *lock);
-
-#define ACPI_ACQUIRE_GLOBAL_LOCK(facs, Acq) \
-	((Acq) = __acpi_acquire_global_lock(&facs->global_lock))
-
-#define ACPI_RELEASE_GLOBAL_LOCK(facs, Acq) \
-	((Acq) = __acpi_release_global_lock(&facs->global_lock))
-
-/*
- * Math helper asm macros
- */
-#define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
-	asm("divl %2;"				     \
-	    :"=a"(q32), "=d"(r32)		     \
-	    :"r"(d32),				     \
-	     "0"(n_lo), "1"(n_hi))
-
-
-#define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
-	asm("shrl   $1,%2	;"	\
-	    "rcrl   $1,%3;"		\
-	    :"=r"(n_hi), "=r"(n_lo)	\
-	    :"0"(n_hi), "1"(n_lo))
-
-#ifdef CONFIG_ACPI
-extern int acpi_lapic;
-extern int acpi_ioapic;
-extern int acpi_noirq;
-extern int acpi_strict;
-extern int acpi_disabled;
-extern int acpi_ht;
-extern int acpi_pci_disabled;
-extern int acpi_skip_timer_override;
-extern int acpi_use_timer_override;
-
-static inline void disable_acpi(void)
-{
-	acpi_disabled = 1;
-	acpi_ht = 0;
-	acpi_pci_disabled = 1;
-	acpi_noirq = 1;
-}
-
-/* Fixmap pages to reserve for ACPI boot-time tables (see fixmap.h) */
-#define FIX_ACPI_PAGES 4
-
-extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
-
-static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
-static inline void acpi_disable_pci(void)
-{
-	acpi_pci_disabled = 1;
-	acpi_noirq_set();
-}
-extern int acpi_irq_balance_set(char *str);
-
-/* routines for saving/restoring kernel state */
-extern int acpi_save_state_mem(void);
-extern void acpi_restore_state_mem(void);
-
-extern unsigned long acpi_wakeup_address;
-
-/* early initialization routine */
-extern void acpi_reserve_bootmem(void);
-
-#else /* !CONFIG_ACPI */
-
-#define acpi_lapic 0
-#define acpi_ioapic 0
-static inline void acpi_noirq_set(void) { }
-static inline void acpi_disable_pci(void) { }
-static inline void disable_acpi(void) { }
-
-#endif /* !CONFIG_ACPI */
-
-#define ARCH_HAS_POWER_INIT	1
-
-#endif /*_ASM_ACPI_H*/
diff --git a/include/asm-x86/acpi_64.h b/include/asm-x86/acpi_64.h
deleted file mode 100644
index b2670c6db8ec..000000000000
--- a/include/asm-x86/acpi_64.h
+++ /dev/null
@@ -1,144 +0,0 @@
-/*
- *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
- *  Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
- *
- * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License as published by
- *  the Free Software Foundation; either version 2 of the License, or
- *  (at your option) any later version.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to the Free Software
- *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
- *
- * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- */
-
-#ifndef _ASM_ACPI_H
-#define _ASM_ACPI_H
-
-#include <acpi/pdc_intel.h>
-#include <asm/numa.h>
-
-#define COMPILER_DEPENDENT_INT64   long long
-#define COMPILER_DEPENDENT_UINT64  unsigned long long
-
-/*
- * Calling conventions:
- *
- * ACPI_SYSTEM_XFACE        - Interfaces to host OS (handlers, threads)
- * ACPI_EXTERNAL_XFACE      - External ACPI interfaces
- * ACPI_INTERNAL_XFACE      - Internal ACPI interfaces
- * ACPI_INTERNAL_VAR_XFACE  - Internal variable-parameter list interfaces
- */
-#define ACPI_SYSTEM_XFACE
-#define ACPI_EXTERNAL_XFACE
-#define ACPI_INTERNAL_XFACE
-#define ACPI_INTERNAL_VAR_XFACE
-
-/* Asm macros */
-
-#define ACPI_ASM_MACROS
-#define BREAKPOINT3
-#define ACPI_DISABLE_IRQS() local_irq_disable()
-#define ACPI_ENABLE_IRQS()  local_irq_enable()
-#define ACPI_FLUSH_CPU_CACHE()	wbinvd()
-
-int __acpi_acquire_global_lock(unsigned int *lock);
-int __acpi_release_global_lock(unsigned int *lock);
-
-#define ACPI_ACQUIRE_GLOBAL_LOCK(facs, Acq) \
-	((Acq) = __acpi_acquire_global_lock(&facs->global_lock))
-
-#define ACPI_RELEASE_GLOBAL_LOCK(facs, Acq) \
-	((Acq) = __acpi_release_global_lock(&facs->global_lock))
-
-/*
- * Math helper asm macros
- */
-#define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
-	asm("divl %2;"				     \
-	    :"=a"(q32), "=d"(r32)		     \
-	    :"r"(d32),				     \
-	     "0"(n_lo), "1"(n_hi))
-
-
-#define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
-	asm("shrl   $1,%2	;"	\
-	    "rcrl   $1,%3;"		\
-	    :"=r"(n_hi), "=r"(n_lo)	\
-	    :"0"(n_hi), "1"(n_lo))
-
-#ifdef CONFIG_ACPI
-extern int acpi_lapic;
-extern int acpi_ioapic;
-extern int acpi_noirq;
-extern int acpi_strict;
-extern int acpi_disabled;
-extern int acpi_ht;
-extern int acpi_pci_disabled;
-extern int acpi_skip_timer_override;
-extern int acpi_use_timer_override;
-
-static inline void disable_acpi(void)
-{
-	acpi_disabled = 1;
-	acpi_ht = 0;
-	acpi_pci_disabled = 1;
-	acpi_noirq = 1;
-}
-
-/* Fixmap pages to reserve for ACPI boot-time tables (see fixmap.h) */
-#define FIX_ACPI_PAGES 4
-
-extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
-
-static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
-static inline void acpi_disable_pci(void)
-{
-	acpi_pci_disabled = 1;
-	acpi_noirq_set();
-}
-extern int acpi_irq_balance_set(char *str);
-
-/* routines for saving/restoring kernel state */
-extern int acpi_save_state_mem(void);
-extern void acpi_restore_state_mem(void);
-
-extern unsigned long acpi_wakeup_address;
-
-/* early initialization routine */
-extern void acpi_reserve_bootmem(void);
-
-#else /* !CONFIG_ACPI */
-
-#define acpi_lapic 0
-#define acpi_ioapic 0
-static inline void acpi_noirq_set(void) { }
-static inline void acpi_disable_pci(void) { }
-
-#endif /* !CONFIG_ACPI */
-
-#define ARCH_HAS_POWER_INIT	1
-
-#ifdef CONFIG_ACPI_NUMA
-extern int acpi_numa;
-extern int acpi_scan_nodes(unsigned long start, unsigned long end);
-#define NR_NODE_MEMBLKS (MAX_NUMNODES*2)
-extern void __init acpi_fake_nodes(const struct bootnode *fake_nodes,
-				   int num_nodes);
-#else
-static inline void acpi_fake_nodes(const struct bootnode *fake_nodes,
-				   int num_nodes)
-{
-}
-#endif
-
-#endif /*_ASM_ACPI_H*/
-- 
cgit v1.2.3


From 4757d7d8d00c5f8b2c176ca03b0a8eabbc206664 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:37 +0100
Subject: x86: put all kern_addr_valid() incarnations to pgtable.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmzone_32.h    | 3 ---
 include/asm-x86/pgtable_32.h   | 8 +++++++-
 include/asm-x86/sparsemem_32.h | 3 ---
 3 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmzone_32.h b/include/asm-x86/mmzone_32.h
index 118e9812778f..5d6f4ce6e6d6 100644
--- a/include/asm-x86/mmzone_32.h
+++ b/include/asm-x86/mmzone_32.h
@@ -87,9 +87,6 @@ static inline int pfn_to_nid(unsigned long pfn)
 	__pgdat->node_start_pfn + __pgdat->node_spanned_pages;		\
 })
 
-/* XXX: FIXME -- wli */
-#define kern_addr_valid(kaddr)	(0)
-
 #ifdef CONFIG_X86_NUMAQ            /* we have contiguous memory on NUMA-Q */
 #define pfn_valid(pfn)          ((pfn) < num_physpages)
 #else
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index ed3e70d8d04b..a8be1ee15a99 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -497,9 +497,15 @@ static inline void paravirt_pagetable_setup_done(pgd_t *base)
 
 #endif /* !__ASSEMBLY__ */
 
+/*
+ * kern_addr_valid() is (1) for FLATMEM and (0) for
+ * SPARSEMEM and DISCONTIGMEM
+ */
 #ifdef CONFIG_FLATMEM
 #define kern_addr_valid(addr)	(1)
-#endif /* CONFIG_FLATMEM */
+#else
+#define kern_addr_valid(kaddr)	(0)
+#endif
 
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
diff --git a/include/asm-x86/sparsemem_32.h b/include/asm-x86/sparsemem_32.h
index cfeed990585f..d89c32100591 100644
--- a/include/asm-x86/sparsemem_32.h
+++ b/include/asm-x86/sparsemem_32.h
@@ -24,8 +24,5 @@
 #define MAX_PHYSMEM_BITS	32
 #endif
 
-/* XXX: FIXME -- wli */
-#define kern_addr_valid(kaddr)  (0)
-
 #endif /* CONFIG_SPARSEMEM */
 #endif /* _I386_SPARSEMEM_H */
-- 
cgit v1.2.3


From 8a423ff0c4a0472607bbed6790fdaeec54af2ebb Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:37 +0100
Subject: x86: merge include/asm-x86/sparsemem.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmzone_64.h    |  2 ++
 include/asm-x86/sparsemem.h    | 35 ++++++++++++++++++++++++++++++++---
 include/asm-x86/sparsemem_32.h | 28 ----------------------------
 include/asm-x86/sparsemem_64.h | 26 --------------------------
 4 files changed, 34 insertions(+), 57 deletions(-)
 delete mode 100644 include/asm-x86/sparsemem_32.h
 delete mode 100644 include/asm-x86/sparsemem_64.h

(limited to 'include')

diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h
index 19a89377b123..1e0ed34a6adc 100644
--- a/include/asm-x86/mmzone_64.h
+++ b/include/asm-x86/mmzone_64.h
@@ -41,6 +41,8 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 #define node_end_pfn(nid)       (NODE_DATA(nid)->node_start_pfn + \
 				 NODE_DATA(nid)->node_spanned_pages)
 
+extern int early_pfn_to_nid(unsigned long pfn);
+
 #ifdef CONFIG_DISCONTIGMEM
 #define pfn_to_nid(pfn) phys_to_nid((unsigned long)(pfn) << PAGE_SHIFT)
 
diff --git a/include/asm-x86/sparsemem.h b/include/asm-x86/sparsemem.h
index 3f203b1d9ee8..fa58cd55411a 100644
--- a/include/asm-x86/sparsemem.h
+++ b/include/asm-x86/sparsemem.h
@@ -1,5 +1,34 @@
+#ifndef _ASM_X86_SPARSEMEM_H
+#define _ASM_X86_SPARSEMEM_H
+
+#ifdef CONFIG_SPARSEMEM
+/*
+ * generic non-linear memory support:
+ *
+ * 1) we will not split memory into more chunks than will fit into the flags
+ *    field of the struct page
+ *
+ * SECTION_SIZE_BITS		2^n: size of each section
+ * MAX_PHYSADDR_BITS		2^n: max size of physical address space
+ * MAX_PHYSMEM_BITS		2^n: how much memory we can have in that space
+ *
+ */
+
 #ifdef CONFIG_X86_32
-# include "sparsemem_32.h"
-#else
-# include "sparsemem_64.h"
+# ifdef CONFIG_X86_PAE
+#  define SECTION_SIZE_BITS	30
+#  define MAX_PHYSADDR_BITS	36
+#  define MAX_PHYSMEM_BITS	36
+# else
+#  define SECTION_SIZE_BITS	26
+#  define MAX_PHYSADDR_BITS	32
+#  define MAX_PHYSMEM_BITS	32
+# endif
+#else /* CONFIG_X86_32 */
+# define SECTION_SIZE_BITS	27 /* matt - 128 is convenient right now */
+# define MAX_PHYSADDR_BITS	40
+# define MAX_PHYSMEM_BITS	40
+#endif
+
+#endif /* CONFIG_SPARSEMEM */
 #endif
diff --git a/include/asm-x86/sparsemem_32.h b/include/asm-x86/sparsemem_32.h
deleted file mode 100644
index d89c32100591..000000000000
--- a/include/asm-x86/sparsemem_32.h
+++ /dev/null
@@ -1,28 +0,0 @@
-#ifndef _I386_SPARSEMEM_H
-#define _I386_SPARSEMEM_H
-#ifdef CONFIG_SPARSEMEM
-
-/*
- * generic non-linear memory support:
- *
- * 1) we will not split memory into more chunks than will fit into the
- *    flags field of the struct page
- */
-
-/*
- * SECTION_SIZE_BITS		2^N: how big each section will be
- * MAX_PHYSADDR_BITS		2^N: how much physical address space we have
- * MAX_PHYSMEM_BITS		2^N: how much memory we can have in that space
- */
-#ifdef CONFIG_X86_PAE
-#define SECTION_SIZE_BITS       30
-#define MAX_PHYSADDR_BITS       36
-#define MAX_PHYSMEM_BITS	36
-#else
-#define SECTION_SIZE_BITS       26
-#define MAX_PHYSADDR_BITS       32
-#define MAX_PHYSMEM_BITS	32
-#endif
-
-#endif /* CONFIG_SPARSEMEM */
-#endif /* _I386_SPARSEMEM_H */
diff --git a/include/asm-x86/sparsemem_64.h b/include/asm-x86/sparsemem_64.h
deleted file mode 100644
index dabb16714a71..000000000000
--- a/include/asm-x86/sparsemem_64.h
+++ /dev/null
@@ -1,26 +0,0 @@
-#ifndef _ASM_X86_64_SPARSEMEM_H
-#define _ASM_X86_64_SPARSEMEM_H 1
-
-#ifdef CONFIG_SPARSEMEM
-
-/*
- * generic non-linear memory support:
- *
- * 1) we will not split memory into more chunks than will fit into the flags
- *    field of the struct page
- *
- * SECTION_SIZE_BITS		2^n: size of each section
- * MAX_PHYSADDR_BITS		2^n: max size of physical address space
- * MAX_PHYSMEM_BITS		2^n: how much memory we can have in that space
- *
- */
-
-#define SECTION_SIZE_BITS	27 /* matt - 128 is convenient right now */
-#define MAX_PHYSADDR_BITS	40
-#define MAX_PHYSMEM_BITS	40
-
-extern int early_pfn_to_nid(unsigned long pfn);
-
-#endif /* CONFIG_SPARSEMEM */
-
-#endif /* _ASM_X86_64_SPARSEMEM_H */
-- 
cgit v1.2.3


From e1d9197862ff4d950dab45669b7f37b5ec3219d8 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:37 +0100
Subject: x86: merge include/asm-x86/sparsemem.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/io_apic.h    | 158 ++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/io_apic_32.h | 155 ------------------------------------------
 include/asm-x86/io_apic_64.h | 134 ------------------------------------
 3 files changed, 156 insertions(+), 291 deletions(-)
 delete mode 100644 include/asm-x86/io_apic_32.h
 delete mode 100644 include/asm-x86/io_apic_64.h

(limited to 'include')

diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h
index 88494966beeb..0f5b3fef0b08 100644
--- a/include/asm-x86/io_apic.h
+++ b/include/asm-x86/io_apic.h
@@ -1,5 +1,159 @@
+#ifndef __ASM_IO_APIC_H
+#define __ASM_IO_APIC_H
+
+#include <asm/types.h>
+#include <asm/mpspec.h>
+#include <asm/apicdef.h>
+
+/*
+ * Intel IO-APIC support for SMP and UP systems.
+ *
+ * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar
+ */
+
+/*
+ * The structure of the IO-APIC:
+ */
+union IO_APIC_reg_00 {
+	u32	raw;
+	struct {
+		u32	__reserved_2	: 14,
+			LTS		:  1,
+			delivery_type	:  1,
+			__reserved_1	:  8,
+			ID		:  8;
+	} __attribute__ ((packed)) bits;
+};
+
+union IO_APIC_reg_01 {
+	u32	raw;
+	struct {
+		u32	version		:  8,
+			__reserved_2	:  7,
+			PRQ		:  1,
+			entries		:  8,
+			__reserved_1	:  8;
+	} __attribute__ ((packed)) bits;
+};
+
+union IO_APIC_reg_02 {
+	u32	raw;
+	struct {
+		u32	__reserved_2	: 24,
+			arbitration	:  4,
+			__reserved_1	:  4;
+	} __attribute__ ((packed)) bits;
+};
+
+union IO_APIC_reg_03 {
+	u32	raw;
+	struct {
+		u32	boot_DT		:  1,
+			__reserved_1	: 31;
+	} __attribute__ ((packed)) bits;
+};
+
+enum ioapic_irq_destination_types {
+	dest_Fixed = 0,
+	dest_LowestPrio = 1,
+	dest_SMI = 2,
+	dest__reserved_1 = 3,
+	dest_NMI = 4,
+	dest_INIT = 5,
+	dest__reserved_2 = 6,
+	dest_ExtINT = 7
+};
+
+struct IO_APIC_route_entry {
+	__u32	vector		:  8,
+		delivery_mode	:  3,	/* 000: FIXED
+					 * 001: lowest prio
+					 * 111: ExtINT
+					 */
+		dest_mode	:  1,	/* 0: physical, 1: logical */
+		delivery_status	:  1,
+		polarity	:  1,
+		irr		:  1,
+		trigger		:  1,	/* 0: edge, 1: level */
+		mask		:  1,	/* 0: enabled, 1: disabled */
+		__reserved_2	: 15;
+
 #ifdef CONFIG_X86_32
-# include "io_apic_32.h"
+	union {
+		struct {
+			__u32	__reserved_1	: 24,
+				physical_dest	:  4,
+				__reserved_2	:  4;
+		} physical;
+
+		struct {
+			__u32	__reserved_1	: 24,
+				logical_dest	:  8;
+		} logical;
+	} dest;
 #else
-# include "io_apic_64.h"
+	__u32	__reserved_3	: 24,
+		dest		:  8;
+#endif
+
+} __attribute__ ((packed));
+
+#ifdef CONFIG_X86_IO_APIC
+
+/*
+ * # of IO-APICs and # of IRQ routing registers
+ */
+extern int nr_ioapics;
+extern int nr_ioapic_registers[MAX_IO_APICS];
+
+/*
+ * MP-BIOS irq configuration table structures:
+ */
+
+/* I/O APIC entries */
+extern struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
+
+/* # of MP IRQ source entries */
+extern int mp_irq_entries;
+
+/* MP IRQ source entries */
+extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+
+/* non-0 if default (table-less) MP configuration */
+extern int mpc_default_type;
+
+/* Older SiS APIC requires we rewrite the index register */
+extern int sis_apic_bug;
+
+/* 1 if "noapic" boot option passed */
+extern int skip_ioapic_setup;
+
+static inline void disable_ioapic_setup(void)
+{
+	skip_ioapic_setup = 1;
+}
+
+/*
+ * If we use the IO-APIC for IRQ routing, disable automatic
+ * assignment of PCI IRQ's.
+ */
+#define io_apic_assign_pci_irqs \
+	(mp_irq_entries && !skip_ioapic_setup && io_apic_irqs)
+
+#ifdef CONFIG_ACPI
+extern int io_apic_get_unique_id(int ioapic, int apic_id);
+extern int io_apic_get_version(int ioapic);
+extern int io_apic_get_redir_entries(int ioapic);
+extern int io_apic_set_pci_routing(int ioapic, int pin, int irq,
+				   int edge_level, int active_high_low);
+extern int timer_uses_ioapic_pin_0;
+#endif /* CONFIG_ACPI */
+
+extern int (*ioapic_renumber_irq)(int ioapic, int irq);
+extern void ioapic_init_mappings(void);
+
+#else  /* !CONFIG_X86_IO_APIC */
+#define io_apic_assign_pci_irqs 0
+#endif
+
 #endif
diff --git a/include/asm-x86/io_apic_32.h b/include/asm-x86/io_apic_32.h
deleted file mode 100644
index 3f087883ea48..000000000000
--- a/include/asm-x86/io_apic_32.h
+++ /dev/null
@@ -1,155 +0,0 @@
-#ifndef __ASM_IO_APIC_H
-#define __ASM_IO_APIC_H
-
-#include <asm/types.h>
-#include <asm/mpspec.h>
-#include <asm/apicdef.h>
-
-/*
- * Intel IO-APIC support for SMP and UP systems.
- *
- * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar
- */
-
-/*
- * The structure of the IO-APIC:
- */
-union IO_APIC_reg_00 {
-	u32	raw;
-	struct {
-		u32	__reserved_2	: 14,
-			LTS		:  1,
-			delivery_type	:  1,
-			__reserved_1	:  8,
-			ID		:  8;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_01 {
-	u32	raw;
-	struct {
-		u32	version		:  8,
-			__reserved_2	:  7,
-			PRQ		:  1,
-			entries		:  8,
-			__reserved_1	:  8;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_02 {
-	u32	raw;
-	struct {
-		u32	__reserved_2	: 24,
-			arbitration	:  4,
-			__reserved_1	:  4;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_03 {
-	u32	raw;
-	struct {
-		u32	boot_DT		:  1,
-			__reserved_1	: 31;
-	} __attribute__ ((packed)) bits;
-};
-
-enum ioapic_irq_destination_types {
-	dest_Fixed = 0,
-	dest_LowestPrio = 1,
-	dest_SMI = 2,
-	dest__reserved_1 = 3,
-	dest_NMI = 4,
-	dest_INIT = 5,
-	dest__reserved_2 = 6,
-	dest_ExtINT = 7
-};
-
-struct IO_APIC_route_entry {
-	__u32	vector		:  8,
-		delivery_mode	:  3,	/* 000: FIXED
-					 * 001: lowest prio
-					 * 111: ExtINT
-					 */
-		dest_mode	:  1,	/* 0: physical, 1: logical */
-		delivery_status	:  1,
-		polarity	:  1,
-		irr		:  1,
-		trigger		:  1,	/* 0: edge, 1: level */
-		mask		:  1,	/* 0: enabled, 1: disabled */
-		__reserved_2	: 15;
-
-	union {		struct { __u32
-					__reserved_1	: 24,
-					physical_dest	:  4,
-					__reserved_2	:  4;
-			} physical;
-
-			struct { __u32
-					__reserved_1	: 24,
-					logical_dest	:  8;
-			} logical;
-	} dest;
-
-} __attribute__ ((packed));
-
-#ifdef CONFIG_X86_IO_APIC
-
-/*
- * # of IO-APICs and # of IRQ routing registers
- */
-extern int nr_ioapics;
-extern int nr_ioapic_registers[MAX_IO_APICS];
-
-/*
- * MP-BIOS irq configuration table structures:
- */
-
-/* I/O APIC entries */
-extern struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
-
-/* # of MP IRQ source entries */
-extern int mp_irq_entries;
-
-/* MP IRQ source entries */
-extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
-
-/* non-0 if default (table-less) MP configuration */
-extern int mpc_default_type;
-
-/* Older SiS APIC requires we rewrite the index register */
-extern int sis_apic_bug;
-
-/* 1 if "noapic" boot option passed */
-extern int skip_ioapic_setup;
-
-static inline void disable_ioapic_setup(void)
-{
-	skip_ioapic_setup = 1;
-}
-
-static inline int ioapic_setup_disabled(void)
-{
-	return skip_ioapic_setup;
-}
-
-/*
- * If we use the IO-APIC for IRQ routing, disable automatic
- * assignment of PCI IRQ's.
- */
-#define io_apic_assign_pci_irqs (mp_irq_entries && !skip_ioapic_setup && io_apic_irqs)
-
-#ifdef CONFIG_ACPI
-extern int io_apic_get_unique_id (int ioapic, int apic_id);
-extern int io_apic_get_version (int ioapic);
-extern int io_apic_get_redir_entries (int ioapic);
-extern int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low);
-extern int timer_uses_ioapic_pin_0;
-#endif /* CONFIG_ACPI */
-
-extern int (*ioapic_renumber_irq)(int ioapic, int irq);
-
-#else  /* !CONFIG_X86_IO_APIC */
-#define io_apic_assign_pci_irqs 0
-#endif
-
-#endif
diff --git a/include/asm-x86/io_apic_64.h b/include/asm-x86/io_apic_64.h
deleted file mode 100644
index 3e3c2114d93a..000000000000
--- a/include/asm-x86/io_apic_64.h
+++ /dev/null
@@ -1,134 +0,0 @@
-#ifndef __ASM_IO_APIC_H
-#define __ASM_IO_APIC_H
-
-#include <asm/types.h>
-#include <asm/mpspec.h>
-#include <asm/apicdef.h>
-
-/*
- * Intel IO-APIC support for SMP and UP systems.
- *
- * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar
- */
-
-#define APIC_MISMATCH_DEBUG
-
-/*
- * The structure of the IO-APIC:
- */
-union IO_APIC_reg_00 {
-	u32	raw;
-	struct {
-		u32	__reserved_2	: 14,
-			LTS		:  1,
-			delivery_type	:  1,
-			__reserved_1	:  8,
-			ID		:  8;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_01 {
-	u32	raw;
-	struct {
-		u32	version		:  8,
-		__reserved_2	:  7,
-		PRQ		:  1,
-		entries		:  8,
-		__reserved_1	:  8;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_02 {
-	u32	raw;
-	struct {
-		u32	__reserved_2	: 24,
-		arbitration	:  4,
-		__reserved_1	:  4;
-	} __attribute__ ((packed)) bits;
-};
-
-union IO_APIC_reg_03 {
-	u32	raw;
-	struct {
-		u32	boot_DT		:  1,
-			__reserved_1	: 31;
-	} __attribute__ ((packed)) bits;
-};
-
-/*
- * # of IO-APICs and # of IRQ routing registers
- */
-extern int nr_ioapics;
-extern int nr_ioapic_registers[MAX_IO_APICS];
-
-enum ioapic_irq_destination_types {
-	dest_Fixed = 0,
-	dest_LowestPrio = 1,
-	dest_SMI = 2,
-	dest__reserved_1 = 3,
-	dest_NMI = 4,
-	dest_INIT = 5,
-	dest__reserved_2 = 6,
-	dest_ExtINT = 7
-};
-
-struct IO_APIC_route_entry {
-	__u32	vector		:  8,
-		delivery_mode	:  3,	/* 000: FIXED
-					 * 001: lowest prio
-					 * 111: ExtINT
-					 */
-		dest_mode	:  1,	/* 0: physical, 1: logical */
-		delivery_status	:  1,
-		polarity	:  1,
-		irr		:  1,
-		trigger		:  1,	/* 0: edge, 1: level */
-		mask		:  1,	/* 0: enabled, 1: disabled */
-		__reserved_2	: 15;
-
-	__u32	__reserved_3	: 24,
-		dest		:  8;
-} __attribute__ ((packed));
-
-/*
- * MP-BIOS irq configuration table structures:
- */
-
-/* I/O APIC entries */
-extern struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
-
-/* # of MP IRQ source entries */
-extern int mp_irq_entries;
-
-/* MP IRQ source entries */
-extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
-
-/* non-0 if default (table-less) MP configuration */
-extern int mpc_default_type;
-
-/* 1 if "noapic" boot option passed */
-extern int skip_ioapic_setup;
-
-static inline void disable_ioapic_setup(void)
-{
-	skip_ioapic_setup = 1;
-}
-
-
-/*
- * If we use the IO-APIC for IRQ routing, disable automatic
- * assignment of PCI IRQ's.
- */
-#define io_apic_assign_pci_irqs (mp_irq_entries && !skip_ioapic_setup && io_apic_irqs)
-
-#ifdef CONFIG_ACPI
-extern int io_apic_get_version (int ioapic);
-extern int io_apic_get_redir_entries (int ioapic);
-extern int io_apic_set_pci_routing (int ioapic, int pin, int irq, int, int);
-#endif
-
-extern void ioapic_init_mappings(void);
-
-extern int sis_apic_bug; /* dummy */ 
-
-#endif
-- 
cgit v1.2.3


From 7462894a7cb03b54b9139f31fab5928366752a78 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:38 +0100
Subject: x86: fixup numa 64 namespace

Using a variable name, which is the same as a macro name is not
really smart. Change the variable names and fixup all users.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/numa_64.c         | 14 +++++++-------
 include/asm-x86/numa_64.h     |  2 +-
 include/asm-x86/topology_64.h | 10 +++++-----
 3 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 4faed6a8f3ae..29b69300aee2 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -30,17 +30,17 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
-unsigned char cpu_to_node[NR_CPUS] __read_mostly = {
+int cpu_to_node_map[NR_CPUS] __read_mostly = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
-EXPORT_SYMBOL(cpu_to_node);
+EXPORT_SYMBOL(cpu_to_node_map);
 
 unsigned char apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
 };
 
-cpumask_t node_to_cpumask[MAX_NUMNODES] __read_mostly;
-EXPORT_SYMBOL(node_to_cpumask);
+cpumask_t node_to_cpumask_map[MAX_NUMNODES] __read_mostly;
+EXPORT_SYMBOL(node_to_cpumask_map);
 
 int numa_off __initdata;
 unsigned long __initdata nodemap_addr;
@@ -542,20 +542,20 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	node_set(0, node_possible_map);
 	for (i = 0; i < NR_CPUS; i++)
 		numa_set_node(i, 0);
-	node_to_cpumask[0] = cpumask_of_cpu(0);
+	node_to_cpumask_map[0] = cpumask_of_cpu(0);
 	e820_register_active_regions(0, start_pfn, end_pfn);
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
 }
 
 __cpuinit void numa_add_cpu(int cpu)
 {
-	set_bit(cpu, &node_to_cpumask[cpu_to_node(cpu)]);
+	set_bit(cpu, &node_to_cpumask_map[cpu_to_node(cpu)]);
 }
 
 void __cpuinit numa_set_node(int cpu, int node)
 {
 	cpu_pda(cpu)->nodenumber = node;
-	cpu_to_node(cpu) = node;
+	cpu_to_node_map[cpu] = node;
 }
 
 unsigned long __init numa_free_all_bootmem(void)
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index e6bc0b5dfe3a..c3c20db1fba3 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -32,7 +32,7 @@ extern void __init init_cpu_to_node(void);
 
 static inline void clear_node_cpumask(int cpu)
 {
-	clear_bit(cpu, &node_to_cpumask[cpu_to_node(cpu)]);
+	clear_bit(cpu, &node_to_cpumask_map[cpu_to_node(cpu)]);
 }
 
 #else
diff --git a/include/asm-x86/topology_64.h b/include/asm-x86/topology_64.h
index 407b22d4e3b0..cf529341b16c 100644
--- a/include/asm-x86/topology_64.h
+++ b/include/asm-x86/topology_64.h
@@ -7,8 +7,8 @@
 #include <asm/mpspec.h>
 #include <linux/bitops.h>
 
-extern unsigned char cpu_to_node[];
-extern cpumask_t     node_to_cpumask[];
+extern int cpu_to_node_map[];
+extern cpumask_t node_to_cpumask_map[];
 
 #ifdef CONFIG_ACPI_NUMA
 extern int __node_distance(int, int);
@@ -16,10 +16,10 @@ extern int __node_distance(int, int);
 /* #else fallback version */
 #endif
 
-#define cpu_to_node(cpu)		(cpu_to_node[cpu])
+#define cpu_to_node(cpu)		(cpu_to_node_map[cpu])
 #define parent_node(node)		(node)
-#define node_to_first_cpu(node) 	(first_cpu(node_to_cpumask[node]))
-#define node_to_cpumask(node)		(node_to_cpumask[node])
+#define node_to_first_cpu(node)		(first_cpu(node_to_cpumask_map[node]))
+#define node_to_cpumask(node)		(node_to_cpumask_map[node])
 #define pcibus_to_node(bus)	((struct pci_sysdata *)((bus)->sysdata))->node
 #define pcibus_to_cpumask(bus)		node_to_cpumask(pcibus_to_node(bus));
 
-- 
cgit v1.2.3


From fe21a445b98c9d52f02f3412d7a2fd39784f3b22 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:38 +0100
Subject: x86: adjust numa 32 namespace

Use the 64bit numa variable names for numa32 as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c  | 16 ++++++++--------
 include/asm-x86/topology_32.h |  8 ++++----
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 753d7acf4dac..239ada1c499c 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -466,19 +466,19 @@ extern struct {
 #ifdef CONFIG_NUMA
 
 /* which logical CPUs are on which nodes */
-cpumask_t node_2_cpu_mask[MAX_NUMNODES] __read_mostly =
+cpumask_t node_to_cpumask_map[MAX_NUMNODES] __read_mostly =
 				{ [0 ... MAX_NUMNODES-1] = CPU_MASK_NONE };
-EXPORT_SYMBOL(node_2_cpu_mask);
+EXPORT_SYMBOL(node_to_cpumask_map);
 /* which node each logical CPU is on */
-int cpu_2_node[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 };
-EXPORT_SYMBOL(cpu_2_node);
+int cpu_to_node_map[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 };
+EXPORT_SYMBOL(cpu_to_node_map);
 
 /* set up a mapping between cpu and node. */
 static inline void map_cpu_to_node(int cpu, int node)
 {
 	printk("Mapping cpu %d to node %d\n", cpu, node);
-	cpu_set(cpu, node_2_cpu_mask[node]);
-	cpu_2_node[cpu] = node;
+	cpu_set(cpu, node_to_cpumask_map[node]);
+	cpu_to_node_map[cpu] = node;
 }
 
 /* undo a mapping between cpu and node. */
@@ -488,8 +488,8 @@ static inline void unmap_cpu_to_node(int cpu)
 
 	printk("Unmapping cpu %d from all nodes\n", cpu);
 	for (node = 0; node < MAX_NUMNODES; node ++)
-		cpu_clear(cpu, node_2_cpu_mask[node]);
-	cpu_2_node[cpu] = 0;
+		cpu_clear(cpu, node_to_cpumask_map[node]);
+	cpu_to_node_map[cpu] = 0;
 }
 #else /* !CONFIG_NUMA */
 
diff --git a/include/asm-x86/topology_32.h b/include/asm-x86/topology_32.h
index fdca80a851e2..e28dbdcc5670 100644
--- a/include/asm-x86/topology_32.h
+++ b/include/asm-x86/topology_32.h
@@ -39,13 +39,13 @@
 #include <linux/cpumask.h>
 
 /* Mappings between logical cpu number and node number */
-extern cpumask_t node_2_cpu_mask[];
-extern int cpu_2_node[];
+extern cpumask_t node_to_cpumask_map[];
+extern int cpu_to_node_map[];
 
 /* Returns the number of the node containing CPU 'cpu' */
 static inline int cpu_to_node(int cpu)
 { 
-	return cpu_2_node[cpu];
+	return cpu_to_node_map[cpu];
 }
 
 /* Returns the number of the node containing Node 'node'.  This architecture is flat, 
@@ -55,7 +55,7 @@ static inline int cpu_to_node(int cpu)
 /* Returns a bitmask of CPUs on Node 'node'. */
 static inline cpumask_t node_to_cpumask(int node)
 {
-	return node_2_cpu_mask[node];
+	return node_to_cpumask_map[node];
 }
 
 /* Returns the number of the first CPU on Node 'node'. */
-- 
cgit v1.2.3


From b4ea9299df1fa04dbd51840f60918a63ff6a8a44 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:38 +0100
Subject: x86: consolidate toloplogy_32/64.h

Reorder defines and do white space / coding style cleanups
to get a readable diff.

Also convert the macros to inline functions. Move the pci
related inlines to pci.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pci.h         | 13 ++++++++++
 include/asm-x86/topology_32.h | 55 +++++++++++++++++++--------------------
 include/asm-x86/topology_64.h | 60 +++++++++++++++++++++++++++++--------------
 3 files changed, 80 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pci.h b/include/asm-x86/pci.h
index 6983730d86fd..c61190cb9e12 100644
--- a/include/asm-x86/pci.h
+++ b/include/asm-x86/pci.h
@@ -92,6 +92,19 @@ static inline void early_quirks(void) { }
 /* generic pci stuff */
 #include <asm-generic/pci.h>
 
+#ifdef CONFIG_NUMA
+/* Returns the node based on pci bus */
+static inline int __pcibus_to_node(struct pci_bus *bus)
+{
+	struct pci_sysdata *sd = bus->sysdata;
+
+	return sd->node;
+}
 
+static inline cpumask_t __pcibus_to_cpumask(struct pci_bus *bus)
+{
+	return node_to_cpumask(__pcibus_to_node(bus));
+}
+#endif
 
 #endif
diff --git a/include/asm-x86/topology_32.h b/include/asm-x86/topology_32.h
index e28dbdcc5670..35b00067587d 100644
--- a/include/asm-x86/topology_32.h
+++ b/include/asm-x86/topology_32.h
@@ -3,7 +3,7 @@
  *
  * Copyright (C) 2002, IBM Corp.
  *
- * All rights reserved.          
+ * All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -22,34 +22,27 @@
  *
  * Send feedback to <colpatch@us.ibm.com>
  */
-#ifndef _ASM_I386_TOPOLOGY_H
-#define _ASM_I386_TOPOLOGY_H
-
-#ifdef CONFIG_X86_HT
-#define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
-#define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
-#define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
-#define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
-#endif
+#ifndef _ASM_X86_TOPOLOGY_H
+#define _ASM_X86_TOPOLOGY_H
 
 #ifdef CONFIG_NUMA
-
-#include <asm/mpspec.h>
-
 #include <linux/cpumask.h>
+#include <asm/mpspec.h>
 
 /* Mappings between logical cpu number and node number */
-extern cpumask_t node_to_cpumask_map[];
 extern int cpu_to_node_map[];
+extern cpumask_t node_to_cpumask_map[];
 
 /* Returns the number of the node containing CPU 'cpu' */
 static inline int cpu_to_node(int cpu)
-{ 
+{
 	return cpu_to_node_map[cpu];
 }
 
-/* Returns the number of the node containing Node 'node'.  This architecture is flat, 
-   so it is a pretty simple function! */
+/*
+ * Returns the number of the node containing Node 'node'. This
+ * architecture is flat, so it is a pretty simple function!
+ */
 #define parent_node(node) (node)
 
 /* Returns a bitmask of CPUs on Node 'node'. */
@@ -60,13 +53,14 @@ static inline cpumask_t node_to_cpumask(int node)
 
 /* Returns the number of the first CPU on Node 'node'. */
 static inline int node_to_first_cpu(int node)
-{ 
+{
 	cpumask_t mask = node_to_cpumask(node);
+
 	return first_cpu(mask);
 }
 
-#define pcibus_to_node(bus) ((struct pci_sysdata *)((bus)->sysdata))->node
-#define pcibus_to_cpumask(bus) node_to_cpumask(pcibus_to_node(bus))
+#define pcibus_to_node(bus) __pcibus_to_node(bus)
+#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus)
 
 /* sched_domains SD_NODE_INIT for NUMAQ machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
@@ -99,21 +93,24 @@ extern unsigned long node_remap_size[];
 
 #define node_has_online_mem(nid) (node_start_pfn[nid] != node_end_pfn[nid])
 
-#else /* !CONFIG_NUMA */
-/*
- * Other i386 platforms should define their own version of the 
- * above macros here.
- */
+#else /* CONFIG_NUMA */
 
 #include <asm-generic/topology.h>
 
-#endif /* CONFIG_NUMA */
+#endif
 
 extern cpumask_t cpu_coregroup_map(int cpu);
 
+#ifdef CONFIG_X86_HT
+#define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
+#define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
+#define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
+#define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
+#endif
+
 #ifdef CONFIG_SMP
-#define mc_capable()	(boot_cpu_data.x86_max_cores > 1)
-#define smt_capable()	(smp_num_siblings > 1)
+#define mc_capable()			(boot_cpu_data.x86_max_cores > 1)
+#define smt_capable()			(smp_num_siblings > 1)
 #endif
 
-#endif /* _ASM_I386_TOPOLOGY_H */
+#endif
diff --git a/include/asm-x86/topology_64.h b/include/asm-x86/topology_64.h
index cf529341b16c..93de948bd337 100644
--- a/include/asm-x86/topology_64.h
+++ b/include/asm-x86/topology_64.h
@@ -1,27 +1,47 @@
-#ifndef _ASM_X86_64_TOPOLOGY_H
-#define _ASM_X86_64_TOPOLOGY_H
-
+#ifndef _ASM_X86_TOPOLOGY_H
+#define _ASM_X86_TOPOLOGY_H
 
 #ifdef CONFIG_NUMA
-
+#include <linux/cpumask.h>
 #include <asm/mpspec.h>
-#include <linux/bitops.h>
 
+/* Mappings between logical cpu number and node number */
 extern int cpu_to_node_map[];
 extern cpumask_t node_to_cpumask_map[];
 
 #ifdef CONFIG_ACPI_NUMA
 extern int __node_distance(int, int);
 #define node_distance(a,b) __node_distance(a,b)
-/* #else fallback version */
 #endif
 
-#define cpu_to_node(cpu)		(cpu_to_node_map[cpu])
-#define parent_node(node)		(node)
-#define node_to_first_cpu(node)		(first_cpu(node_to_cpumask_map[node]))
-#define node_to_cpumask(node)		(node_to_cpumask_map[node])
-#define pcibus_to_node(bus)	((struct pci_sysdata *)((bus)->sysdata))->node
-#define pcibus_to_cpumask(bus)		node_to_cpumask(pcibus_to_node(bus));
+/* Returns the number of the node containing CPU 'cpu' */
+static inline int cpu_to_node(int cpu)
+{
+	return cpu_to_node_map[cpu];
+}
+
+/*
+ * Returns the number of the node containing Node 'node'. This
+ * architecture is flat, so it is a pretty simple function!
+ */
+#define parent_node(node) (node)
+
+/* Returns a bitmask of CPUs on Node 'node'. */
+static inline cpumask_t node_to_cpumask(int node)
+{
+	return node_to_cpumask_map[node];
+}
+
+/* Returns the number of the first CPU on Node 'node'. */
+static inline int node_to_first_cpu(int node)
+{
+	cpumask_t mask = node_to_cpumask(node);
+
+	return first_cpu(mask);
+}
+
+#define pcibus_to_node(bus) __pcibus_to_node(bus)
+#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus)
 
 #define numa_node_id()			read_pda(nodenumber)
 
@@ -38,12 +58,12 @@ extern int __node_distance(int, int);
 	.cache_nice_tries	= 2,			\
 	.busy_idx		= 3,			\
 	.idle_idx		= 2,			\
-	.newidle_idx		= 0, 			\
+	.newidle_idx		= 0,			\
 	.wake_idx		= 1,			\
 	.forkexec_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_FORK	\
 				| SD_BALANCE_EXEC	\
+				| SD_BALANCE_FORK	\
 				| SD_SERIALIZE		\
 				| SD_WAKE_BALANCE,	\
 	.last_balance		= jiffies,		\
@@ -51,19 +71,21 @@ extern int __node_distance(int, int);
 	.nr_balance_failed	= 0,			\
 }
 
+#else /* CONFIG_NUMA */
+
+#include <asm-generic/topology.h>
+
 #endif
 
+extern cpumask_t cpu_coregroup_map(int cpu);
+
 #ifdef CONFIG_SMP
 #define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
 #define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
 #define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
 #define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
 #define mc_capable()			(boot_cpu_data.x86_max_cores > 1)
-#define smt_capable() 			(smp_num_siblings > 1)
+#define smt_capable()			(smp_num_siblings > 1)
 #endif
 
-#include <asm-generic/topology.h>
-
-extern cpumask_t cpu_coregroup_map(int cpu);
-
 #endif
-- 
cgit v1.2.3


From 3367e56f26a3a8647784328ffa846fe8021a5b79 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:38 +0100
Subject: x86: merge topology.h variants

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/topology.h    | 143 +++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/topology_32.h | 116 ----------------------------------
 include/asm-x86/topology_64.h |  91 ---------------------------
 3 files changed, 141 insertions(+), 209 deletions(-)
 delete mode 100644 include/asm-x86/topology_32.h
 delete mode 100644 include/asm-x86/topology_64.h

(limited to 'include')

diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index b10fde9798ea..9c251604ecc4 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -1,5 +1,144 @@
+/*
+ * Written by: Matthew Dobson, IBM Corporation
+ *
+ * Copyright (C) 2002, IBM Corp.
+ *
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT.  See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * Send feedback to <colpatch@us.ibm.com>
+ */
+#ifndef _ASM_X86_TOPOLOGY_H
+#define _ASM_X86_TOPOLOGY_H
+
+#ifdef CONFIG_NUMA
+#include <linux/cpumask.h>
+#include <asm/mpspec.h>
+
+/* Mappings between logical cpu number and node number */
+extern int cpu_to_node_map[];
+extern cpumask_t node_to_cpumask_map[];
+
+/* Returns the number of the node containing CPU 'cpu' */
+static inline int cpu_to_node(int cpu)
+{
+	return cpu_to_node_map[cpu];
+}
+
+/*
+ * Returns the number of the node containing Node 'node'. This
+ * architecture is flat, so it is a pretty simple function!
+ */
+#define parent_node(node) (node)
+
+/* Returns a bitmask of CPUs on Node 'node'. */
+static inline cpumask_t node_to_cpumask(int node)
+{
+	return node_to_cpumask_map[node];
+}
+
+/* Returns the number of the first CPU on Node 'node'. */
+static inline int node_to_first_cpu(int node)
+{
+	cpumask_t mask = node_to_cpumask(node);
+
+	return first_cpu(mask);
+}
+
+#define pcibus_to_node(bus) __pcibus_to_node(bus)
+#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus)
+
 #ifdef CONFIG_X86_32
-# include "topology_32.h"
+extern unsigned long node_start_pfn[];
+extern unsigned long node_end_pfn[];
+extern unsigned long node_remap_size[];
+#define node_has_online_mem(nid) (node_start_pfn[nid] != node_end_pfn[nid])
+
+# ifdef CONFIG_X86_HT
+#  define ENABLE_TOPO_DEFINES
+# endif
+
+# define SD_CACHE_NICE_TRIES	1
+# define SD_IDLE_IDX		1
+# define SD_NEWIDLE_IDX		2
+# define SD_FORKEXEC_IDX	0
+
 #else
-# include "topology_64.h"
+
+# ifdef CONFIG_SMP
+#  define ENABLE_TOPO_DEFINES
+# endif
+
+# define SD_CACHE_NICE_TRIES	2
+# define SD_IDLE_IDX		2
+# define SD_NEWIDLE_IDX		0
+# define SD_FORKEXEC_IDX	1
+
+#endif
+
+/* sched_domains SD_NODE_INIT for NUMAQ machines */
+#define SD_NODE_INIT (struct sched_domain) {		\
+	.span			= CPU_MASK_NONE,	\
+	.parent			= NULL,			\
+	.child			= NULL,			\
+	.groups			= NULL,			\
+	.min_interval		= 8,			\
+	.max_interval		= 32,			\
+	.busy_factor		= 32,			\
+	.imbalance_pct		= 125,			\
+	.cache_nice_tries	= SD_CACHE_NICE_TRIES,	\
+	.busy_idx		= 3,			\
+	.idle_idx		= SD_IDLE_IDX,		\
+	.newidle_idx		= SD_NEWIDLE_IDX,	\
+	.wake_idx		= 1,			\
+	.forkexec_idx		= SD_FORKEXEC_IDX,	\
+	.flags			= SD_LOAD_BALANCE	\
+				| SD_BALANCE_EXEC	\
+				| SD_BALANCE_FORK	\
+				| SD_SERIALIZE		\
+				| SD_WAKE_BALANCE,	\
+	.last_balance		= jiffies,		\
+	.balance_interval	= 1,			\
+	.nr_balance_failed	= 0,			\
+}
+
+#ifdef CONFIG_X86_64_ACPI_NUMA
+extern int __node_distance(int, int);
+#define node_distance(a, b) __node_distance(a, b)
+#endif
+
+#else /* CONFIG_NUMA */
+
+#include <asm-generic/topology.h>
+
+#endif
+
+extern cpumask_t cpu_coregroup_map(int cpu);
+
+#ifdef ENABLE_TOPO_DEFINES
+#define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
+#define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
+#define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
+#define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
+#endif
+
+#ifdef CONFIG_SMP
+#define mc_capable()			(boot_cpu_data.x86_max_cores > 1)
+#define smt_capable()			(smp_num_siblings > 1)
+#endif
+
 #endif
diff --git a/include/asm-x86/topology_32.h b/include/asm-x86/topology_32.h
deleted file mode 100644
index 35b00067587d..000000000000
--- a/include/asm-x86/topology_32.h
+++ /dev/null
@@ -1,116 +0,0 @@
-/*
- * Written by: Matthew Dobson, IBM Corporation
- *
- * Copyright (C) 2002, IBM Corp.
- *
- * All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * Send feedback to <colpatch@us.ibm.com>
- */
-#ifndef _ASM_X86_TOPOLOGY_H
-#define _ASM_X86_TOPOLOGY_H
-
-#ifdef CONFIG_NUMA
-#include <linux/cpumask.h>
-#include <asm/mpspec.h>
-
-/* Mappings between logical cpu number and node number */
-extern int cpu_to_node_map[];
-extern cpumask_t node_to_cpumask_map[];
-
-/* Returns the number of the node containing CPU 'cpu' */
-static inline int cpu_to_node(int cpu)
-{
-	return cpu_to_node_map[cpu];
-}
-
-/*
- * Returns the number of the node containing Node 'node'. This
- * architecture is flat, so it is a pretty simple function!
- */
-#define parent_node(node) (node)
-
-/* Returns a bitmask of CPUs on Node 'node'. */
-static inline cpumask_t node_to_cpumask(int node)
-{
-	return node_to_cpumask_map[node];
-}
-
-/* Returns the number of the first CPU on Node 'node'. */
-static inline int node_to_first_cpu(int node)
-{
-	cpumask_t mask = node_to_cpumask(node);
-
-	return first_cpu(mask);
-}
-
-#define pcibus_to_node(bus) __pcibus_to_node(bus)
-#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus)
-
-/* sched_domains SD_NODE_INIT for NUMAQ machines */
-#define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
-	.parent			= NULL,			\
-	.child			= NULL,			\
-	.groups			= NULL,			\
-	.min_interval		= 8,			\
-	.max_interval		= 32,			\
-	.busy_factor		= 32,			\
-	.imbalance_pct		= 125,			\
-	.cache_nice_tries	= 1,			\
-	.busy_idx		= 3,			\
-	.idle_idx		= 1,			\
-	.newidle_idx		= 2,			\
-	.wake_idx		= 1,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_FORK	\
-				| SD_SERIALIZE		\
-				| SD_WAKE_BALANCE,	\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
-	.nr_balance_failed	= 0,			\
-}
-
-extern unsigned long node_start_pfn[];
-extern unsigned long node_end_pfn[];
-extern unsigned long node_remap_size[];
-
-#define node_has_online_mem(nid) (node_start_pfn[nid] != node_end_pfn[nid])
-
-#else /* CONFIG_NUMA */
-
-#include <asm-generic/topology.h>
-
-#endif
-
-extern cpumask_t cpu_coregroup_map(int cpu);
-
-#ifdef CONFIG_X86_HT
-#define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
-#define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
-#define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
-#define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
-#endif
-
-#ifdef CONFIG_SMP
-#define mc_capable()			(boot_cpu_data.x86_max_cores > 1)
-#define smt_capable()			(smp_num_siblings > 1)
-#endif
-
-#endif
diff --git a/include/asm-x86/topology_64.h b/include/asm-x86/topology_64.h
deleted file mode 100644
index 93de948bd337..000000000000
--- a/include/asm-x86/topology_64.h
+++ /dev/null
@@ -1,91 +0,0 @@
-#ifndef _ASM_X86_TOPOLOGY_H
-#define _ASM_X86_TOPOLOGY_H
-
-#ifdef CONFIG_NUMA
-#include <linux/cpumask.h>
-#include <asm/mpspec.h>
-
-/* Mappings between logical cpu number and node number */
-extern int cpu_to_node_map[];
-extern cpumask_t node_to_cpumask_map[];
-
-#ifdef CONFIG_ACPI_NUMA
-extern int __node_distance(int, int);
-#define node_distance(a,b) __node_distance(a,b)
-#endif
-
-/* Returns the number of the node containing CPU 'cpu' */
-static inline int cpu_to_node(int cpu)
-{
-	return cpu_to_node_map[cpu];
-}
-
-/*
- * Returns the number of the node containing Node 'node'. This
- * architecture is flat, so it is a pretty simple function!
- */
-#define parent_node(node) (node)
-
-/* Returns a bitmask of CPUs on Node 'node'. */
-static inline cpumask_t node_to_cpumask(int node)
-{
-	return node_to_cpumask_map[node];
-}
-
-/* Returns the number of the first CPU on Node 'node'. */
-static inline int node_to_first_cpu(int node)
-{
-	cpumask_t mask = node_to_cpumask(node);
-
-	return first_cpu(mask);
-}
-
-#define pcibus_to_node(bus) __pcibus_to_node(bus)
-#define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus)
-
-#define numa_node_id()			read_pda(nodenumber)
-
-/* sched_domains SD_NODE_INIT for x86_64 machines */
-#define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
-	.parent			= NULL,			\
-	.child			= NULL,			\
-	.groups			= NULL,			\
-	.min_interval		= 8,			\
-	.max_interval		= 32,			\
-	.busy_factor		= 32,			\
-	.imbalance_pct		= 125,			\
-	.cache_nice_tries	= 2,			\
-	.busy_idx		= 3,			\
-	.idle_idx		= 2,			\
-	.newidle_idx		= 0,			\
-	.wake_idx		= 1,			\
-	.forkexec_idx		= 1,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_FORK	\
-				| SD_SERIALIZE		\
-				| SD_WAKE_BALANCE,	\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
-	.nr_balance_failed	= 0,			\
-}
-
-#else /* CONFIG_NUMA */
-
-#include <asm-generic/topology.h>
-
-#endif
-
-extern cpumask_t cpu_coregroup_map(int cpu);
-
-#ifdef CONFIG_SMP
-#define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
-#define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
-#define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
-#define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
-#define mc_capable()			(boot_cpu_data.x86_max_cores > 1)
-#define smt_capable()			(smp_num_siblings > 1)
-#endif
-
-#endif
-- 
cgit v1.2.3


From 3578facf737dc9a60ed19a83458bd9cfdc2af233 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:30:38 +0100
Subject: x86: merge resume-trace.h variants

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/asm.h             |  2 ++
 include/asm-x86/resume-trace.h    | 23 +++++++++++++++++++----
 include/asm-x86/resume-trace_32.h | 13 -------------
 include/asm-x86/resume-trace_64.h | 13 -------------
 4 files changed, 21 insertions(+), 30 deletions(-)
 delete mode 100644 include/asm-x86/resume-trace_32.h
 delete mode 100644 include/asm-x86/resume-trace_64.h

(limited to 'include')

diff --git a/include/asm-x86/asm.h b/include/asm-x86/asm.h
index b5006eb9acba..8661ae75488a 100644
--- a/include/asm-x86/asm.h
+++ b/include/asm-x86/asm.h
@@ -6,12 +6,14 @@
 
 # define _ASM_PTR	" .long "
 # define _ASM_ALIGN	" .balign 4 "
+# define _ASM_MOV_UL	" movl "
 
 #else
 /* 64 bits */
 
 # define _ASM_PTR	" .quad "
 # define _ASM_ALIGN	" .balign 8 "
+# define _ASM_MOV_UL	" movq "
 
 #endif /* CONFIG_X86_32 */
 
diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h
index 9b6dd093a9f7..46f725b0bc82 100644
--- a/include/asm-x86/resume-trace.h
+++ b/include/asm-x86/resume-trace.h
@@ -1,5 +1,20 @@
-#ifdef CONFIG_X86_32
-# include "resume-trace_32.h"
-#else
-# include "resume-trace_64.h"
+#ifndef _ASM_X86_RESUME_TRACE_H
+#define _ASM_X86_RESUME_TRACE_H
+
+#include <asm/asm.h>
+
+#define TRACE_RESUME(user) do {					\
+	if (pm_trace_enabled) {					\
+		void *tracedata;				\
+		asm volatile(_ASM_MOV_UL " $1f,%0\n"		\
+			".section .tracedata,\"a\"\n"		\
+			"1:\t.word %c1\n\t"			\
+			_ASM_PTR " %c2\n"			\
+			".previous"				\
+			:"=r" (tracedata)			\
+			: "i" (__LINE__), "i" (__FILE__));	\
+		generate_resume_trace(tracedata, user);		\
+	}							\
+} while (0)
+
 #endif
diff --git a/include/asm-x86/resume-trace_32.h b/include/asm-x86/resume-trace_32.h
deleted file mode 100644
index ec9cfd656230..000000000000
--- a/include/asm-x86/resume-trace_32.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#define TRACE_RESUME(user) do {					\
-	if (pm_trace_enabled) {					\
-		void *tracedata;				\
-		asm volatile("movl $1f,%0\n"			\
-			".section .tracedata,\"a\"\n"		\
-			"1:\t.word %c1\n"			\
-			"\t.long %c2\n"				\
-			".previous"				\
-			:"=r" (tracedata)			\
-			: "i" (__LINE__), "i" (__FILE__));	\
-		generate_resume_trace(tracedata, user);		\
-	}							\
-} while (0)
diff --git a/include/asm-x86/resume-trace_64.h b/include/asm-x86/resume-trace_64.h
deleted file mode 100644
index 34bf998fdf62..000000000000
--- a/include/asm-x86/resume-trace_64.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#define TRACE_RESUME(user) do {					\
-	if (pm_trace_enabled) {					\
-		void *tracedata;				\
-		asm volatile("movq $1f,%0\n"			\
-			".section .tracedata,\"a\"\n"		\
-			"1:\t.word %c1\n"			\
-			"\t.quad %c2\n"				\
-			".previous"				\
-			:"=r" (tracedata)			\
-			: "i" (__LINE__), "i" (__FILE__));	\
-		generate_resume_trace(tracedata, user);		\
-	}							\
-} while (0)
-- 
cgit v1.2.3


From 1c69524c2e5b83e52a098ebdeb4a8b52169f6a03 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Wed, 30 Jan 2008 13:30:39 +0100
Subject: x86: clear IO_APIC before enabing apic error vector.

4 socket quad core, 8 socket quad core will do apic ID lifting for BSP.

But io-apic regs for ExtINT still use 0 as dest.

so when we enable apic error vector in BSP, we will get one APIC error.

CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 512K (64 bytes/line)
CPU 0/4 -> Node 0
CPU: Physical Processor ID: 1
CPU: Processor Core ID: 0
SMP alternatives: switching to UP code
ACPI: Core revision 20070126
enabled ExtINT on CPU#0
ESR value after enabling vector: 00000000, after 0000000c
APIC error on CPU0: 0c(08)
ENABLING IO-APIC IRQs
Synchronizing Arb IDs.

So move enable_IO_APIC from setup_IO_APIC into setup_local_APIC and call it
before enabling the ACPI error vector.

[ tglx: arch/x86 adaptation ]

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_64.c    | 7 +++++++
 arch/x86/kernel/io_apic_64.c | 7 +++++--
 include/asm-x86/hw_irq_64.h  | 1 +
 3 files changed, 13 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 915808bd8a2a..994298bf4921 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -774,6 +774,13 @@ void __cpuinit setup_local_APIC(void)
 		value = APIC_DM_NMI | APIC_LVT_MASKED;
 	apic_write(APIC_LVT1, value);
 
+	/*
+	 * Now enable IO-APICs, actually call clear_IO_APIC
+	 * We need clear_IO_APIC before enabling vector on BP
+	 */
+	if (!smp_processor_id() && !skip_ioapic_setup && nr_ioapics)
+		enable_IO_APIC();
+
 	{
 		unsigned oldvalue;
 		maxlvt = lapic_get_maxlvt();
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 3e471d0fb150..4ef85a3b3f9f 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1172,7 +1172,7 @@ void __apicdebuginit print_PIC(void)
 
 #endif  /*  0  */
 
-static void __init enable_IO_APIC(void)
+void __init enable_IO_APIC(void)
 {
 	union IO_APIC_reg_01 reg_01;
 	int i8259_apic, i8259_pin;
@@ -1789,7 +1789,10 @@ __setup("no_timer_check", notimercheck);
 
 void __init setup_IO_APIC(void)
 {
-	enable_IO_APIC();
+
+	/*
+	 * calling enable_IO_APIC() is moved to setup_local_APIC for BP
+	 */
 
 	if (acpi_ioapic)
 		io_apic_irqs = ~0;	/* all IRQs go through IOAPIC */
diff --git a/include/asm-x86/hw_irq_64.h b/include/asm-x86/hw_irq_64.h
index a470d59da678..a346159b6ac3 100644
--- a/include/asm-x86/hw_irq_64.h
+++ b/include/asm-x86/hw_irq_64.h
@@ -135,6 +135,7 @@ extern void init_8259A(int aeoi);
 extern void send_IPI_self(int vector);
 extern void init_VISWS_APIC_irqs(void);
 extern void setup_IO_APIC(void);
+extern void enable_IO_APIC(void);
 extern void disable_IO_APIC(void);
 extern void print_IO_APIC(void);
 extern int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn);
-- 
cgit v1.2.3


From a860b63c41f241c685245127a3d32f352cb04c12 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Wed, 30 Jan 2008 13:30:39 +0100
Subject: x86: store core id bits in cpuinfo_x8

We need to store core id bits to cpuinfo_x86 in early_identify_cpu. So we
use it to create acpiid_to_node array in k8topolgy.c

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Len Brown <lenb@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_64.c     | 72 +++++++++++++++++++++++++++---------------
 include/asm-x86/processor_64.h |  1 +
 2 files changed, 48 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 62db062ba138..84f66b7b4d2e 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -542,18 +542,7 @@ static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
 	int node = 0;
 	unsigned apicid = hard_smp_processor_id();
 #endif
-	unsigned ecx = cpuid_ecx(0x80000008);
-
-	c->x86_max_cores = (ecx & 0xff) + 1;
-
-	/* CPU telling us the core id bits shift? */
-	bits = (ecx >> 12) & 0xF;
-
-	/* Otherwise recompute */
-	if (bits == 0) {
-		while ((1 << bits) < c->x86_max_cores)
-			bits++;
-	}
+	bits = c->x86_coreid_bits;
 
 	/* Low order bits define the core id (index of core in socket) */
 	c->cpu_core_id = c->phys_proc_id & ((1 << bits)-1);
@@ -591,6 +580,33 @@ static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
 #endif
 }
 
+static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_SMP
+	unsigned bits, ecx;
+
+	/* Multi core CPU? */
+	if (c->extended_cpuid_level < 0x80000008)
+		return;
+
+	ecx = cpuid_ecx(0x80000008);
+
+	c->x86_max_cores = (ecx & 0xff) + 1;
+
+	/* CPU telling us the core id bits shift? */
+	bits = (ecx >> 12) & 0xF;
+
+	/* Otherwise recompute */
+	if (bits == 0) {
+		while ((1 << bits) < c->x86_max_cores)
+			bits++;
+	}
+
+	c->x86_coreid_bits = bits;
+
+#endif
+}
+
 #define ENABLE_C1E_MASK		0x18000000
 #define CPUID_PROCESSOR_SIGNATURE	1
 #define CPUID_XFAM		0x0ff00000
@@ -858,7 +874,7 @@ struct cpu_model_info {
    below. */
 static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 {
-	u32 tfms;
+	u32 tfms, xlvl;
 
 	c->loops_per_jiffy = loops_per_jiffy;
 	c->x86_cache_size = -1;
@@ -869,6 +885,7 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 	c->x86_clflush_size = 64;
 	c->x86_cache_alignment = c->x86_clflush_size;
 	c->x86_max_cores = 1;
+	c->x86_coreid_bits = 0;
 	c->extended_cpuid_level = 0;
 	memset(&c->x86_capability, 0, sizeof c->x86_capability);
 
@@ -905,18 +922,6 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 #ifdef CONFIG_SMP
 	c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
 #endif
-}
-
-/*
- * This does the hard work of actually picking apart the CPU stuff...
- */
-void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
-{
-	int i;
-	u32 xlvl;
-
-	early_identify_cpu(c);
-
 	/* AMD-defined flags: level 0x80000001 */
 	xlvl = cpuid_eax(0x80000000);
 	c->extended_cpuid_level = xlvl;
@@ -937,6 +942,23 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 			c->x86_capability[2] = cpuid_edx(0x80860001);
 	}
 
+	switch (c->x86_vendor) {
+	case X86_VENDOR_AMD:
+		early_init_amd(c);
+		break;
+	}
+
+}
+
+/*
+ * This does the hard work of actually picking apart the CPU stuff...
+ */
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+{
+	int i;
+
+	early_identify_cpu(c);
+
 	init_scattered_cpuid_features(c);
 
 	c->apicid = phys_pkg_id(0);
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 6c0d96a7af63..2ae8ceb8a74e 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -61,6 +61,7 @@ struct cpuinfo_x86 {
 	int	x86_tlbsize;	/* number of 4K pages in DTLB/ITLB combined(in pages)*/
         __u8    x86_virt_bits, x86_phys_bits;
 	__u8	x86_max_cores;	/* cpuid returned max cores value */
+	__u8	x86_coreid_bits; /* cpuid returned core id bits */
         __u32   x86_power; 	
 	__u32   extended_cpuid_level;	/* Max extended CPUID function supported */
 	unsigned long loops_per_jiffy;
-- 
cgit v1.2.3


From 739f33b38bf88312447e38ae8b7ac3acdbb72a6b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:30:40 +0100
Subject: x86: untable __init references between IO data

Earlier patch added IO APIC setup into local APIC setup. This caused
modpost warnings. Fix them by untangling setup_local_APIC() and splitting
it into smaller functions. The IO APIC initialization is only called
for the BP init.

Also removed some outdated debugging code and minor cleanup.

[ tglx: arch/x86 adaptation ]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_64.c    | 46 ++++++++++++++++++++++----------------------
 arch/x86/kernel/smpboot_64.c |  8 ++++++++
 include/asm-x86/apic.h       |  1 +
 3 files changed, 32 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 994298bf4921..d341f798255c 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -677,7 +677,7 @@ void __init init_bsp_APIC(void)
  */
 void __cpuinit setup_local_APIC(void)
 {
-	unsigned int value, maxlvt;
+	unsigned int value;
 	int i, j;
 
 	value = apic_read(APIC_LVR);
@@ -773,32 +773,23 @@ void __cpuinit setup_local_APIC(void)
 	else
 		value = APIC_DM_NMI | APIC_LVT_MASKED;
 	apic_write(APIC_LVT1, value);
+}
+
+void __cpuinit lapic_setup_esr(void)
+{
+	unsigned maxlvt = lapic_get_maxlvt();
 
+	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR);
 	/*
-	 * Now enable IO-APICs, actually call clear_IO_APIC
-	 * We need clear_IO_APIC before enabling vector on BP
+	 * spec says clear errors after enabling vector.
 	 */
-	if (!smp_processor_id() && !skip_ioapic_setup && nr_ioapics)
-		enable_IO_APIC();
-
-	{
-		unsigned oldvalue;
-		maxlvt = lapic_get_maxlvt();
-		oldvalue = apic_read(APIC_ESR);
-		value = ERROR_APIC_VECTOR;      // enables sending errors
-		apic_write(APIC_LVTERR, value);
-		/*
-		 * spec says clear errors after enabling vector.
-		 */
-		if (maxlvt > 3)
-			apic_write(APIC_ESR, 0);
-		value = apic_read(APIC_ESR);
-		if (value != oldvalue)
-			apic_printk(APIC_VERBOSE,
-			"ESR value after enabling vector: %08x, after %08x\n",
-			oldvalue, value);
-	}
+	if (maxlvt > 3)
+		apic_write(APIC_ESR, 0);
+}
 
+void __cpuinit end_local_APIC_setup(void)
+{
+	lapic_setup_esr();
 	nmi_watchdog_default();
 	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
@@ -879,6 +870,15 @@ int __init APIC_init_uniprocessor(void)
 
 	setup_local_APIC();
 
+	/*
+	 * Now enable IO-APICs, actually call clear_IO_APIC
+	 * We need clear_IO_APIC before enabling vector on BP
+	 */
+	if (!skip_ioapic_setup && nr_ioapics)
+		enable_IO_APIC();
+
+	end_local_APIC_setup();
+
 	if (smp_found_config && !skip_ioapic_setup && nr_ioapics)
 		setup_IO_APIC();
 	else
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 7552db9ee9ff..ddefb38c53fb 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -210,6 +210,7 @@ void __cpuinit smp_callin(void)
 
 	Dprintk("CALLIN, before setup_local_APIC().\n");
 	setup_local_APIC();
+	end_local_APIC_setup();
 
 	/*
 	 * Get our bogomips.
@@ -884,6 +885,13 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	 */
 	setup_local_APIC();
 
+	/*
+	 * Enable IO APIC before setting up error vector
+	 */
+	if (!skip_ioapic_setup && nr_ioapics)
+		enable_IO_APIC();
+	end_local_APIC_setup();
+
 	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_id) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
 		      GET_APIC_ID(apic_read(APIC_ID)), boot_cpu_id);
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 18d932dff476..5e8192d36e5a 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -112,6 +112,7 @@ extern void cache_APIC_registers(void);
 extern void sync_Arb_IDs(void);
 extern void init_bsp_APIC(void);
 extern void setup_local_APIC(void);
+extern void end_local_APIC_setup(void);
 extern void init_apic_mappings(void);
 extern void setup_boot_APIC_clock(void);
 extern void setup_secondary_APIC_clock(void);
-- 
cgit v1.2.3


From 7b83dae7aa31db4f6d6e78c3c6d490a7ac58699c Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 30 Jan 2008 13:30:40 +0100
Subject: x86: extended interrupt LVT support for AMD Barcelona

Also macro definitions in apicdef.h has been updated.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_64.c               | 26 ++++++++++++++++++++++----
 arch/x86/kernel/cpu/mcheck/mce_amd_64.c | 10 +++++-----
 include/asm-x86/apic.h                  |  4 ++--
 include/asm-x86/apicdef.h               | 19 ++++++++++++-------
 4 files changed, 41 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index d341f798255c..027004262105 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -187,17 +187,35 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 }
 
 /*
- * Setup extended LVT (K8 specific)
+ * Setup extended LVT, AMD specific (K8, family 10h)
+ *
+ * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
+ * MCE interrupts are supported. Thus MCE offset must be set to 0.
  */
-void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
-			     unsigned char msg_type, unsigned char mask)
+
+#define APIC_EILVT_LVTOFF_MCE 0
+#define APIC_EILVT_LVTOFF_IBS 1
+
+static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
 {
-	unsigned long reg = (lvt_off << 4) + K8_APIC_EXT_LVT_BASE;
+	unsigned long reg = (lvt_off << 4) + APIC_EILVT0;
 	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
 
 	apic_write(reg, v);
 }
 
+u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_MCE;
+}
+
+u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_IBS;
+}
+
 /*
  * Program the next event, relative to now
  */
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 073afa7dd89a..550502596ca3 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -118,6 +118,7 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 {
 	unsigned int bank, block;
 	unsigned int cpu = smp_processor_id();
+	u8 lvt_off;
 	u32 low = 0, high = 0, address = 0;
 
 	for (bank = 0; bank < NR_BANKS; ++bank) {
@@ -153,14 +154,13 @@ void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
 			if (shared_bank[bank] && c->cpu_core_id)
 				break;
 #endif
+			lvt_off = setup_APIC_eilvt_mce(THRESHOLD_APIC_VECTOR,
+						       APIC_EILVT_MSG_FIX, 0);
+
 			high &= ~MASK_LVTOFF_HI;
-			high |= K8_APIC_EXT_LVT_ENTRY_THRESHOLD << 20;
+			high |= lvt_off << 20;
 			wrmsr(address, low, high);
 
-			setup_APIC_extended_lvt(K8_APIC_EXT_LVT_ENTRY_THRESHOLD,
-						THRESHOLD_APIC_VECTOR,
-						K8_APIC_EXT_INT_MSG_FIX, 0);
-
 			threshold_defaults.address = address;
 			threshold_restart_bank(&threshold_defaults, 0, 0);
 		}
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 5e8192d36e5a..423022759cb2 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -126,8 +126,8 @@ extern void enable_NMI_through_LVT0(void *dummy);
 extern void setup_apic_routing(void);
 #endif
 
-extern void setup_APIC_extended_lvt(unsigned char lvt_off, unsigned char vector,
-				    unsigned char msg_type, unsigned char mask);
+extern u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask);
+extern u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask);
 
 extern int apic_is_clustered_box(void);
 
diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 5f7abe9b5f87..550af7a6f88e 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -116,13 +116,18 @@
 #define		APIC_TDR_DIV_32		0x8
 #define		APIC_TDR_DIV_64		0x9
 #define		APIC_TDR_DIV_128	0xA
-
-#define K8_APIC_EXT_LVT_BASE		0x500
-#define K8_APIC_EXT_INT_MSG_FIX		0x0
-#define K8_APIC_EXT_INT_MSG_SMI		0x2
-#define K8_APIC_EXT_INT_MSG_NMI		0x4
-#define K8_APIC_EXT_INT_MSG_EXT		0x7
-#define K8_APIC_EXT_LVT_ENTRY_THRESHOLD	0
+#define	APIC_EILVT0     0x500
+#define		APIC_EILVT_NR_AMD_K8	1	/* Number of extended interrupts */
+#define		APIC_EILVT_NR_AMD_10H	4
+#define		APIC_EILVT_LVTOFF(x)	(((x)>>4)&0xF)
+#define		APIC_EILVT_MSG_FIX	0x0
+#define		APIC_EILVT_MSG_SMI	0x2
+#define		APIC_EILVT_MSG_NMI	0x4
+#define		APIC_EILVT_MSG_EXT	0x7
+#define		APIC_EILVT_MASKED	(1<<16)
+#define	APIC_EILVT1     0x510
+#define	APIC_EILVT2     0x520
+#define	APIC_EILVT3     0x530
 
 #define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
 
-- 
cgit v1.2.3


From c1d171a002942ea2d93b4fbd0c9583c56fce0772 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Wed, 30 Jan 2008 13:30:40 +0100
Subject: x86: randomize brk

Randomize the location of the heap (brk) for i386 and x86_64.  The range is
randomized in the range starting at current brk location up to 0x02000000
offset for both architectures.  This, together with
pie-executable-randomization.patch and
pie-executable-randomization-fix.patch, should make the address space
randomization on i386 and x86_64 complete.

Arjan says:

This is known to break older versions of some emacs variants, whose dumper
code assumed that the last variable declared in the program is equal to the
start of the dynamically allocated memory region.

(The dumper is the code where emacs effectively dumps core at the end of it's
compilation stage; this coredump is then loaded as the main program during
normal use)

iirc this was 5 years or so; we found this way back when I was at RH and we
first did the security stuff there (including this brk randomization).  It
wasn't all variants of emacs, and it got fixed as a result (I vaguely remember
that emacs already had code to deal with it for other archs/oses, just
ifdeffed wrongly).

It's a rare and wrong assumption as a general thing, just on x86 it mostly
happened to be true (but to be honest, it'll break too if gcc does
something fancy or if the linker does a non-standard order).  Still its
something we should at least document.

Note 2: afaik it only broke the emacs *build*.  I'm not 100% sure about that
(it IS 5 years ago) though.

[ akpm@linux-foundation.org: deuglification ]

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c | 7 +++++++
 arch/x86/kernel/process_64.c | 7 +++++++
 fs/binfmt_elf.c              | 6 ++++++
 include/asm-x86/elf.h        | 3 +++
 mm/mmap.c                    | 3 ++-
 5 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index a8cdd09ad53f..631af167bc51 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -992,3 +992,10 @@ unsigned long arch_align_stack(unsigned long sp)
 		sp -= get_random_int() % 8192;
 	return sp & ~0xf;
 }
+
+unsigned long arch_randomize_brk(struct mm_struct *mm)
+{
+	unsigned long range_end = mm->brk + 0x02000000;
+	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
+}
+
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 98d85952f574..aa9414ed74c7 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -914,3 +914,10 @@ unsigned long arch_align_stack(unsigned long sp)
 		sp -= get_random_int() % 8192;
 	return sp & ~0xf;
 }
+
+unsigned long arch_randomize_brk(struct mm_struct *mm)
+{
+	unsigned long range_end = mm->brk + 0x02000000;
+	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
+}
+
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index f0b3171842f2..043a800c8f71 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1021,6 +1021,12 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	current->mm->end_data = end_data;
 	current->mm->start_stack = bprm->p;
 
+#ifdef arch_randomize_brk
+	if (current->flags & PF_RANDOMIZE)
+		current->mm->brk = current->mm->start_brk =
+			arch_randomize_brk(current->mm);
+#endif
+
 	if (current->personality & MMAP_PAGE_ZERO) {
 		/* Why this, you ask???  Well SVr4 maps page 0 as read-only,
 		   and some applications "depend" upon this behavior.
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index ec42a4d2e83b..cd3204ebbbdd 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -285,6 +285,9 @@ struct linux_binprm;
 extern int arch_setup_additional_pages(struct linux_binprm *bprm,
 				       int executable_stack);
 
+extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+#define arch_randomize_brk arch_randomize_brk
+
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/mm/mmap.c b/mm/mmap.c
index bfa389fc6ded..d2b6d44962b7 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -251,7 +251,8 @@ asmlinkage unsigned long sys_brk(unsigned long brk)
 	 * not page aligned -Ram Gupta
 	 */
 	rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur;
-	if (rlim < RLIM_INFINITY && brk - mm->start_data > rlim)
+	if (rlim < RLIM_INFINITY && (brk - mm->start_brk) +
+			(mm->end_data - mm->start_data) > rlim)
 		goto out;
 
 	newbrk = PAGE_ALIGN(brk);
-- 
cgit v1.2.3


From 7f3646aa16f496201e045183402c6614fa27b0c7 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:41 +0100
Subject: x86 vDSO: use vdso-syms.lds

This patch changes the kernel's references to addresses in the vDSO image
to be based on the symbols defined by vdso-syms.lds instead of the old
vdso-syms.o symbols.  This is all wrapped up in a macro defined by the new
asm-x86/vdso.h header; that's the only place in the kernel source that has
to know the details of the scheme for getting vDSO symbol values.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/vdso/Makefile     |  3 +--
 arch/x86/vdso/vdso-start.S |  2 --
 arch/x86/vdso/vma.c        | 18 +++++++-----------
 include/asm-x86/vdso.h     | 14 ++++++++++++++
 4 files changed, 22 insertions(+), 15 deletions(-)
 delete mode 100644 arch/x86/vdso/vdso-start.S
 create mode 100644 include/asm-x86/vdso.h

(limited to 'include')

diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile
index d8200adae9fa..d2ec17044adc 100644
--- a/arch/x86/vdso/Makefile
+++ b/arch/x86/vdso/Makefile
@@ -3,8 +3,7 @@
 #
 
 # files to link into the vdso
-# vdso-start.o has to be first
-vobjs-y := vdso-start.o vdso-note.o vclock_gettime.o vgetcpu.o vvar.o
+vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o vvar.o
 
 # files to link into kernel
 obj-y := vma.o vdso.o vdso-syms.o
diff --git a/arch/x86/vdso/vdso-start.S b/arch/x86/vdso/vdso-start.S
deleted file mode 100644
index 2dc2cdb84d67..000000000000
--- a/arch/x86/vdso/vdso-start.S
+++ /dev/null
@@ -1,2 +0,0 @@
-	.globl vdso_kernel_start
-vdso_kernel_start:
diff --git a/arch/x86/vdso/vma.c b/arch/x86/vdso/vma.c
index ff9333e5fb08..3fdd51497a83 100644
--- a/arch/x86/vdso/vma.c
+++ b/arch/x86/vdso/vma.c
@@ -11,23 +11,20 @@
 #include <asm/vsyscall.h>
 #include <asm/vgtod.h>
 #include <asm/proto.h>
-#include "voffset.h"
+#include <asm/vdso.h>
 
-int vdso_enabled = 1;
-
-#define VEXTERN(x) extern typeof(__ ## x) *vdso_ ## x;
-#include "vextern.h"
+#include "vextern.h"		/* Just for VMAGIC.  */
 #undef VEXTERN
 
-extern char vdso_kernel_start[], vdso_start[], vdso_end[];
+int vdso_enabled = 1;
+
+extern char vdso_start[], vdso_end[];
 extern unsigned short vdso_sync_cpuid;
 
 struct page **vdso_pages;
 
-static inline void *var_ref(void *vbase, char *var, char *name)
+static inline void *var_ref(void *p, char *name)
 {
-	unsigned offset = var - &vdso_kernel_start[0] + VDSO_TEXT_OFFSET;
-	void *p = vbase + offset;
 	if (*(void **)p != (void *)VMAGIC) {
 		printk("VDSO: variable %s broken\n", name);
 		vdso_enabled = 0;
@@ -62,9 +59,8 @@ static int __init init_vdso_vars(void)
 		vdso_enabled = 0;
 	}
 
-#define V(x) *(typeof(x) *) var_ref(vbase, (char *)RELOC_HIDE(&x, 0), #x)
 #define VEXTERN(x) \
-	V(vdso_ ## x) = &__ ## x;
+	*(typeof(__ ## x) **) var_ref(VDSO64_SYMBOL(vbase, x), #x) = &__ ## x;
 #include "vextern.h"
 #undef VEXTERN
 	return 0;
diff --git a/include/asm-x86/vdso.h b/include/asm-x86/vdso.h
new file mode 100644
index 000000000000..9379ec06fd1f
--- /dev/null
+++ b/include/asm-x86/vdso.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_X86_VDSO_H
+#define _ASM_X86_VDSO_H	1
+
+extern const char VDSO64_PRELINK[];
+
+/*
+ * Given a pointer to the vDSO image, find the pointer to VDSO64_name
+ * as that symbol is defined in the vDSO sources or linker script.
+ */
+#define VDSO64_SYMBOL(base, name) ({		\
+	extern const char VDSO64_##name[];	\
+	(void *) (VDSO64_##name - VDSO64_PRELINK + (unsigned long) (base)); })
+
+#endif	/* asm-x86/vdso.h */
-- 
cgit v1.2.3


From 6c3652efcafa6a6d795093362cb4290c84994b5c Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:42 +0100
Subject: x86 vDSO: i386 vdso32

This makes the i386 kernel use the new vDSO build in arch/x86/vdso/vdso32/
to replace the old one from arch/x86/kernel/.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Makefile_32              |  3 +-
 arch/x86/kernel/Makefile_32       | 43 +------------------------
 arch/x86/kernel/asm-offsets_32.c  |  2 --
 arch/x86/kernel/signal_32.c       |  5 +--
 arch/x86/kernel/sysenter_32.c     |  6 ++--
 arch/x86/kernel/vsyscall_32.S     | 15 ---------
 arch/x86/kernel/vsyscall_32.lds.S | 67 ---------------------------------------
 arch/x86/vdso/Makefile            |  8 +++--
 arch/x86/vdso/vdso32.S            | 15 +++++++++
 arch/x86/vdso/vdso32/note.S       |  3 +-
 arch/x86/vdso/vdso32/sysenter.S   |  3 +-
 arch/x86/xen/setup.c              |  5 ++-
 include/asm-x86/elf.h             | 13 ++------
 include/asm-x86/vdso.h            | 14 ++++++++
 14 files changed, 50 insertions(+), 152 deletions(-)
 delete mode 100644 arch/x86/kernel/vsyscall_32.S
 delete mode 100644 arch/x86/kernel/vsyscall_32.lds.S
 create mode 100644 arch/x86/vdso/vdso32.S

(limited to 'include')

diff --git a/arch/x86/Makefile_32 b/arch/x86/Makefile_32
index b152db71d5b5..9214224773ed 100644
--- a/arch/x86/Makefile_32
+++ b/arch/x86/Makefile_32
@@ -110,7 +110,8 @@ libs-y 					+= arch/x86/lib/
 core-y					+= arch/x86/kernel/ \
 					   arch/x86/mm/ \
 					   $(mcore-y)/ \
-					   arch/x86/crypto/
+					   arch/x86/crypto/ \
+					   arch/x86/vdso/
 drivers-$(CONFIG_MATH_EMULATION)	+= arch/x86/math-emu/
 drivers-$(CONFIG_PCI)			+= arch/x86/pci/
 # must be linked after kernel/
diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index f7c1c1c88dda..0eef8226753b 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -33,7 +33,7 @@ obj-$(CONFIG_X86_NUMAQ)		+= numaq_32.o
 obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit_32.o
 obj-$(CONFIG_KPROBES)		+= kprobes_32.o
 obj-$(CONFIG_MODULES)		+= module_32.o
-obj-y				+= sysenter_32.o vsyscall_32.o
+obj-y				+= sysenter_32.o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat_32.o
 obj-$(CONFIG_EFI) 		+= efi_32.o efi_stub_32.o
 obj-$(CONFIG_DOUBLEFAULT) 	+= doublefault_32.o
@@ -48,44 +48,3 @@ obj-$(CONFIG_PARAVIRT)		+= paravirt_32.o
 obj-y				+= pcspeaker.o
 
 obj-$(CONFIG_SCx200)		+= scx200_32.o
-
-# vsyscall_32.o contains the vsyscall DSO images as __initdata.
-# We must build both images before we can assemble it.
-# Note: kbuild does not track this dependency due to usage of .incbin
-$(obj)/vsyscall_32.o: $(obj)/vsyscall-int80_32.so $(obj)/vsyscall-sysenter_32.so
-targets += $(foreach F,int80 sysenter,vsyscall-$F_32.o vsyscall-$F_32.so)
-targets += vsyscall-note_32.o vsyscall_32.lds
-
-$(obj)/vsyscall-%_32.o: $(src)/../vdso/vdso32/%.S
-	$(call if_changed_dep,as_o_S)
-
-# The DSO images are built using a special linker script.
-quiet_cmd_syscall = SYSCALL $@
-      cmd_syscall = $(CC) -m elf_i386 -nostdlib $(SYSCFLAGS_$(@F)) \
-		          -Wl,-T,$(filter-out FORCE,$^) -o $@
-
-export CPPFLAGS_vsyscall_32.lds += -P -C -Ui386
-
-vsyscall-flags = -shared -s -Wl,-soname=linux-gate.so.1 \
-		 $(call ld-option, -Wl$(comma)--hash-style=sysv)
-SYSCFLAGS_vsyscall-sysenter_32.so	= $(vsyscall-flags)
-SYSCFLAGS_vsyscall-int80_32.so	= $(vsyscall-flags)
-
-$(obj)/vsyscall-int80_32.so $(obj)/vsyscall-sysenter_32.so: \
-$(obj)/vsyscall-%.so: $(src)/vsyscall_32.lds \
-		      $(obj)/vsyscall-%.o $(obj)/vsyscall-note_32.o FORCE
-	$(call if_changed,syscall)
-
-# We also create a special relocatable object that should mirror the symbol
-# table and layout of the linked DSO.  With ld -R we can then refer to
-# these symbols in the kernel code rather than hand-coded addresses.
-extra-y += vsyscall-syms.o
-$(obj)/built-in.o: $(obj)/vsyscall-syms.o
-$(obj)/built-in.o: ld_flags += -R $(obj)/vsyscall-syms.o
-
-SYSCFLAGS_vsyscall-syms.o = -r
-$(obj)/vsyscall-syms.o: $(src)/vsyscall_32.lds \
-			$(obj)/vsyscall-sysenter_32.o $(obj)/vsyscall-note_32.o FORCE
-	$(call if_changed,syscall)
-
-
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 4f750ec810f6..fd7464d23339 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -111,8 +111,6 @@ void foo(void)
 	DEFINE(PTRS_PER_PMD, PTRS_PER_PMD);
 	DEFINE(PTRS_PER_PGD, PTRS_PER_PGD);
 
-	DEFINE(VDSO_PRELINK_asm, VDSO_PRELINK);
-
 	OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx);
 
 #ifdef CONFIG_PARAVIRT
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 5c6170c44b00..1ac53e9a0859 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -23,6 +23,7 @@
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
 #include <asm/i387.h>
+#include <asm/vdso.h>
 #include "sigframe_32.h"
 
 #define DEBUG_SIG 0
@@ -362,7 +363,7 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 	}
 
 	if (current->binfmt->hasvdso)
-		restorer = (void *)VDSO_SYM(&__kernel_sigreturn);
+		restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn);
 	else
 		restorer = (void *)&frame->retcode;
 	if (ka->sa.sa_flags & SA_RESTORER)
@@ -459,7 +460,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		goto give_sigsegv;
 
 	/* Set up to return from userspace.  */
-	restorer = (void *)VDSO_SYM(&__kernel_rt_sigreturn);
+	restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
 	if (ka->sa.sa_flags & SA_RESTORER)
 		restorer = ka->sa.sa_restorer;
 	err |= __put_user(restorer, &frame->pretcode);
diff --git a/arch/x86/kernel/sysenter_32.c b/arch/x86/kernel/sysenter_32.c
index 5a2d951e2608..85c52d23ee40 100644
--- a/arch/x86/kernel/sysenter_32.c
+++ b/arch/x86/kernel/sysenter_32.c
@@ -23,6 +23,7 @@
 #include <asm/unistd.h>
 #include <asm/elf.h>
 #include <asm/tlbflush.h>
+#include <asm/vdso.h>
 
 enum {
 	VDSO_DISABLED = 0,
@@ -259,9 +260,6 @@ int __init sysenter_setup(void)
 	return 0;
 }
 
-/* Defined in vsyscall-sysenter.S */
-extern void SYSENTER_RETURN;
-
 /* Setup a VMA at program startup for the vsyscall page */
 int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 {
@@ -308,7 +306,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 
 	current->mm->context.vdso = (void *)addr;
 	current_thread_info()->sysenter_return =
-		(void *)VDSO_SYM(&SYSENTER_RETURN);
+		VDSO32_SYMBOL(addr, SYSENTER_RETURN);
 
   up_fail:
 	up_write(&mm->mmap_sem);
diff --git a/arch/x86/kernel/vsyscall_32.S b/arch/x86/kernel/vsyscall_32.S
deleted file mode 100644
index a5ab3dc4fd25..000000000000
--- a/arch/x86/kernel/vsyscall_32.S
+++ /dev/null
@@ -1,15 +0,0 @@
-#include <linux/init.h>
-
-__INITDATA
-
-	.globl vsyscall_int80_start, vsyscall_int80_end
-vsyscall_int80_start:
-	.incbin "arch/x86/kernel/vsyscall-int80_32.so"
-vsyscall_int80_end:
-
-	.globl vsyscall_sysenter_start, vsyscall_sysenter_end
-vsyscall_sysenter_start:
-	.incbin "arch/x86/kernel/vsyscall-sysenter_32.so"
-vsyscall_sysenter_end:
-
-__FINIT
diff --git a/arch/x86/kernel/vsyscall_32.lds.S b/arch/x86/kernel/vsyscall_32.lds.S
deleted file mode 100644
index 4a8b0ed9b8fb..000000000000
--- a/arch/x86/kernel/vsyscall_32.lds.S
+++ /dev/null
@@ -1,67 +0,0 @@
-/*
- * Linker script for vsyscall DSO.  The vsyscall page is an ELF shared
- * object prelinked to its virtual address, and with only one read-only
- * segment (that fits in one page).  This script controls its layout.
- */
-#include <asm/asm-offsets.h>
-
-SECTIONS
-{
-  . = VDSO_PRELINK_asm + SIZEOF_HEADERS;
-
-  .hash           : { *(.hash) }		:text
-  .gnu.hash       : { *(.gnu.hash) }
-  .dynsym         : { *(.dynsym) }
-  .dynstr         : { *(.dynstr) }
-  .gnu.version    : { *(.gnu.version) }
-  .gnu.version_d  : { *(.gnu.version_d) }
-  .gnu.version_r  : { *(.gnu.version_r) }
-
-  /* This linker script is used both with -r and with -shared.
-     For the layouts to match, we need to skip more than enough
-     space for the dynamic symbol table et al.  If this amount
-     is insufficient, ld -shared will barf.  Just increase it here.  */
-  . = VDSO_PRELINK_asm + 0x400;
-
-  .text           : { *(.text) }		:text =0x90909090
-  .note		  : { *(.note.*) }		:text :note
-  .eh_frame_hdr   : { *(.eh_frame_hdr) }	:text :eh_frame_hdr
-  .eh_frame       : { KEEP (*(.eh_frame)) }	:text
-  .dynamic        : { *(.dynamic) }		:text :dynamic
-  .useless        : {
-  	*(.got.plt) *(.got)
-	*(.data .data.* .gnu.linkonce.d.*)
-	*(.dynbss)
-	*(.bss .bss.* .gnu.linkonce.b.*)
-  }						:text
-}
-
-/*
- * We must supply the ELF program headers explicitly to get just one
- * PT_LOAD segment, and set the flags explicitly to make segments read-only.
- */
-PHDRS
-{
-  text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
-  dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
-  note PT_NOTE FLAGS(4); /* PF_R */
-  eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */
-}
-
-/*
- * This controls what symbols we export from the DSO.
- */
-VERSION
-{
-  LINUX_2.5 {
-    global:
-    	__kernel_vsyscall;
-    	__kernel_sigreturn;
-    	__kernel_rt_sigreturn;
-
-    local: *;
-  };
-}
-
-/* The ELF entry point can be used to set the AT_SYSINFO value.  */
-ENTRY(__kernel_vsyscall);
diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile
index a02e1ca2a1bc..ca2aabf8ed39 100644
--- a/arch/x86/vdso/Makefile
+++ b/arch/x86/vdso/Makefile
@@ -14,7 +14,8 @@ vdso-install-$(VDSO32-y)	+= $(vdso32-y:=.so)
 vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o vvar.o
 
 # files to link into kernel
-obj-y := vma.o vdso.o
+obj-$(VDSO64-y)			+= vma.o vdso.o
+obj-$(CONFIG_X86_32)		+= vdso32.o
 
 vobjs := $(foreach F,$(vobjs-y),$(obj)/$F)
 
@@ -52,7 +53,7 @@ $(obj)/vclock_gettime.o: KBUILD_CFLAGS = $(CFL)
 $(obj)/vgetcpu.o: KBUILD_CFLAGS = $(CFL)
 
 targets += vdso-syms.lds
-obj-y += vdso-syms.lds
+obj-$(VDSO64-y)			+= vdso-syms.lds
 
 #
 # Match symbols in the DSO that look like VDSO*; produce a file of constants.
@@ -68,6 +69,7 @@ $(obj)/%-syms.lds: $(obj)/%.so.dbg FORCE
 #
 # Build multiple 32-bit vDSO images to choose from at boot time.
 #
+obj-$(VDSO32-y)			+= vdso32-syms.lds
 vdso32.so-$(CONFIG_X86_32)	+= int80
 vdso32.so-$(VDSO32-y)		+= sysenter
 
@@ -84,6 +86,8 @@ targets += vdso32/note.o $(vdso32.so-y:%=vdso32/%.o)
 
 extra-y	+= $(vdso32.so-y:%=vdso32-%.so)
 
+$(obj)/vdso32.o: $(vdso32.so-y:%=$(obj)/vdso32-%.so)
+
 KBUILD_AFLAGS_32 := $(filter-out -m64,$(KBUILD_AFLAGS))
 $(vdso32.so-y:%=$(obj)/vdso32-%.so.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_32)
 $(vdso32.so-y:%=$(obj)/vdso32-%.so.dbg): asflags-$(CONFIG_X86_64) += -m32
diff --git a/arch/x86/vdso/vdso32.S b/arch/x86/vdso/vdso32.S
new file mode 100644
index 000000000000..cab020c99c3d
--- /dev/null
+++ b/arch/x86/vdso/vdso32.S
@@ -0,0 +1,15 @@
+#include <linux/init.h>
+
+__INITDATA
+
+	.globl vsyscall_int80_start, vsyscall_int80_end
+vsyscall_int80_start:
+	.incbin "arch/x86/vdso/vdso32-int80.so"
+vsyscall_int80_end:
+
+	.globl vsyscall_sysenter_start, vsyscall_sysenter_end
+vsyscall_sysenter_start:
+	.incbin "arch/x86/vdso/vdso32-sysenter.so"
+vsyscall_sysenter_end:
+
+__FINIT
diff --git a/arch/x86/vdso/vdso32/note.S b/arch/x86/vdso/vdso32/note.S
index 0cf934f2633e..c83f25734696 100644
--- a/arch/x86/vdso/vdso32/note.S
+++ b/arch/x86/vdso/vdso32/note.S
@@ -35,10 +35,9 @@ ELFNOTE_END
 
 #include "../../xen/vdso.h"	/* Defines VDSO_NOTE_NONEGSEG_BIT.  */
 
-	.globl VDSO_NOTE_MASK
 ELFNOTE_START(GNU, 2, "a")
 	.long 1			/* ncaps */
-VDSO_NOTE_MASK:
+VDSO32_NOTE_MASK:		/* Symbol used by arch/x86/xen/setup.c */
 	.long 0			/* mask */
 	.byte VDSO_NOTE_NONEGSEG_BIT; .asciz "nosegneg"	/* bit, name */
 ELFNOTE_END
diff --git a/arch/x86/vdso/vdso32/sysenter.S b/arch/x86/vdso/vdso32/sysenter.S
index 4b98fc41d3e0..109bfa394eaa 100644
--- a/arch/x86/vdso/vdso32/sysenter.S
+++ b/arch/x86/vdso/vdso32/sysenter.S
@@ -45,8 +45,7 @@ __kernel_vsyscall:
 	/* 14: System call restart point is here! (SYSENTER_RETURN-2) */
 	jmp .Lenter_kernel
 	/* 16: System call normal return point is here! */
-	.globl SYSENTER_RETURN	/* Symbol used by sysenter.c  */
-SYSENTER_RETURN:
+VDSO32_SYSENTER_RETURN:	/* Symbol used by sysenter.c via vdso32-syms.h */
 	pop %ebp
 .Lpop_ebp:
 	pop %edx
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index f84e77226646..fd91568090f4 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -10,6 +10,7 @@
 #include <linux/pm.h>
 
 #include <asm/elf.h>
+#include <asm/vdso.h>
 #include <asm/e820.h>
 #include <asm/setup.h>
 #include <asm/xen/hypervisor.h>
@@ -61,10 +62,8 @@ static void xen_idle(void)
  */
 static void fiddle_vdso(void)
 {
-	extern u32 VDSO_NOTE_MASK; /* See ../kernel/vsyscall-note.S.  */
 	extern char vsyscall_int80_start;
-	u32 *mask = (u32 *) ((unsigned long) &VDSO_NOTE_MASK - VDSO_PRELINK +
-			     &vsyscall_int80_start);
+	u32 *mask = VDSO32_SYMBOL(&vsyscall_int80_start, NOTE_MASK);
 	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
 }
 
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index cd3204ebbbdd..70edff2d5671 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -78,6 +78,7 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 #include <asm/processor.h>
 #include <asm/system.h>		/* for savesegment */
 #include <asm/desc.h>
+#include <asm/vdso.h>
 
 /*
  * This is used to ensure we don't load something for the wrong architecture.
@@ -247,17 +248,9 @@ extern int dump_task_extended_fpu (struct task_struct *,
 
 #define VDSO_HIGH_BASE		(__fix_to_virt(FIX_VDSO))
 #define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
-#define VDSO_PRELINK		0
 
-#define VDSO_SYM(x) \
-		(VDSO_CURRENT_BASE + (unsigned long)(x) - VDSO_PRELINK)
-
-#define VDSO_HIGH_EHDR		((const struct elfhdr *) VDSO_HIGH_BASE)
-#define VDSO_EHDR		((const struct elfhdr *) VDSO_CURRENT_BASE)
-
-extern void __kernel_vsyscall;
-
-#define VDSO_ENTRY		VDSO_SYM(&__kernel_vsyscall)
+#define VDSO_ENTRY \
+	((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
 
 /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
 
diff --git a/include/asm-x86/vdso.h b/include/asm-x86/vdso.h
index 9379ec06fd1f..629bcb6e8e45 100644
--- a/include/asm-x86/vdso.h
+++ b/include/asm-x86/vdso.h
@@ -1,6 +1,7 @@
 #ifndef _ASM_X86_VDSO_H
 #define _ASM_X86_VDSO_H	1
 
+#ifdef CONFIG_X86_64
 extern const char VDSO64_PRELINK[];
 
 /*
@@ -10,5 +11,18 @@ extern const char VDSO64_PRELINK[];
 #define VDSO64_SYMBOL(base, name) ({		\
 	extern const char VDSO64_##name[];	\
 	(void *) (VDSO64_##name - VDSO64_PRELINK + (unsigned long) (base)); })
+#endif
+
+#if defined CONFIG_X86_32 || defined CONFIG_COMPAT
+extern const char VDSO32_PRELINK[];
+
+/*
+ * Given a pointer to the vDSO image, find the pointer to VDSO32_name
+ * as that symbol is defined in the vDSO sources or linker script.
+ */
+#define VDSO32_SYMBOL(base, name) ({		\
+	extern const char VDSO32_##name[];	\
+	(void *) (VDSO32_##name - VDSO32_PRELINK + (unsigned long) (base)); })
+#endif
 
 #endif	/* asm-x86/vdso.h */
-- 
cgit v1.2.3


From 36197c92a20c142fc2a068e0366053d770fa0096 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:43 +0100
Subject: x86 vDSO: ia32 sysenter_return

This changes the 64-bit kernel's support for the 32-bit sysenter
instruction to use stored fields rather than constants for the
user-mode return address, as the 32-bit kernel does.  This adds a
sysenter_return field to struct thread_info, as 32-bit has.  There
is no observable effect from this yet.  It makes the assembly code
independent of the 32-bit vDSO mapping address, paving the way for
making the vDSO address vary as it does on the 32-bit kernel.

[ akpm@linux-foundation.org: build fix on !CONFIG_IA32_EMULATION ]

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ia32entry.S        | 7 +++----
 arch/x86/ia32/syscall32.c        | 4 ++++
 arch/x86/kernel/asm-offsets_64.c | 3 +++
 include/asm-x86/thread_info_64.h | 3 +++
 4 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index df588f0f76e1..2499a324feaa 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -12,7 +12,6 @@
 #include <asm/ia32_unistd.h>	
 #include <asm/thread_info.h>	
 #include <asm/segment.h>
-#include <asm/vsyscall32.h>
 #include <asm/irqflags.h>
 #include <linux/linkage.h>
 
@@ -104,7 +103,7 @@ ENTRY(ia32_sysenter_target)
 	pushfq
 	CFI_ADJUST_CFA_OFFSET 8
 	/*CFI_REL_OFFSET rflags,0*/
-	movl	$VSYSCALL32_SYSEXIT, %r10d
+	movl	8*3-THREAD_SIZE+threadinfo_sysenter_return(%rsp), %r10d
 	CFI_REGISTER rip,r10
 	pushq	$__USER32_CS
 	CFI_ADJUST_CFA_OFFSET 8
@@ -142,6 +141,8 @@ sysenter_do_call:
 	andl    $~TS_COMPAT,threadinfo_status(%r10)
 	/* clear IF, that popfq doesn't enable interrupts early */
 	andl  $~0x200,EFLAGS-R11(%rsp) 
+	movl	RIP-R11(%rsp),%edx		/* User %eip */
+	CFI_REGISTER rip,rdx
 	RESTORE_ARGS 1,24,1,1,1,1
 	popfq
 	CFI_ADJUST_CFA_OFFSET -8
@@ -149,8 +150,6 @@ sysenter_do_call:
 	popq	%rcx				/* User %esp */
 	CFI_ADJUST_CFA_OFFSET -8
 	CFI_REGISTER rsp,rcx
-	movl	$VSYSCALL32_SYSEXIT,%edx	/* User %eip */
-	CFI_REGISTER rip,rdx
 	TRACE_IRQS_ON
 	swapgs
 	sti		/* sti only takes effect after the next instruction */
diff --git a/arch/x86/ia32/syscall32.c b/arch/x86/ia32/syscall32.c
index d751d96c2ef2..98ff99f5b59a 100644
--- a/arch/x86/ia32/syscall32.c
+++ b/arch/x86/ia32/syscall32.c
@@ -46,6 +46,10 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack)
 				      VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC|
 				      VM_ALWAYSDUMP,
 				      syscall32_pages);
+	if (ret == 0) {
+		current->mm->context.vdso = (void __user *)VSYSCALL32_BASE;
+		current_thread_info()->sysenter_return = VSYSCALL32_SYSEXIT;
+	}
 	up_write(&mm->mmap_sem);
 	return ret;
 }
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index 40f41752c1df..c27c646214f4 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -46,6 +46,9 @@ int main(void)
 	ENTRY(addr_limit);
 	ENTRY(preempt_count);
 	ENTRY(status);
+#ifdef CONFIG_IA32_EMULATION
+	ENTRY(sysenter_return);
+#endif
 	BLANK();
 #undef ENTRY
 #define ENTRY(entry) DEFINE(pda_ ## entry, offsetof(struct x8664_pda, entry))
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index 1ac23c157231..c8e7736fc792 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -33,6 +33,9 @@ struct thread_info {
 
 	mm_segment_t		addr_limit;	
 	struct restart_block    restart_block;
+#ifdef CONFIG_IA32_EMULATION
+	void __user		*sysenter_return;
+#endif
 };
 #endif
 
-- 
cgit v1.2.3


From af65d64845a90c8f2fc90b97e2148ff74672e979 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:43 +0100
Subject: x86 vDSO: consolidate vdso32

This makes x86_64's ia32 emulation support share the sources used in the
32-bit kernel for the 32-bit vDSO and much of its setup code.

The 32-bit vDSO mapping now behaves the same on x86_64 as on native 32-bit.
The abi.syscall32 sysctl on x86_64 now takes the same values that
vm.vdso_enabled takes on the 32-bit kernel.  That is, 1 means a randomized
vDSO location, 2 means the fixed old address.  The CONFIG_COMPAT_VDSO
option is now available to make this the default setting, the same meaning
it has for the 32-bit kernel.  (This does not affect the 64-bit vDSO.)

The argument vdso32=[012] can be used on both 32-bit and 64-bit kernels to
set this paramter at boot time.  The vdso=[012] argument still does this
same thing on the 32-bit kernel.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |   5 ++
 arch/x86/Kconfig                    |   4 +-
 arch/x86/ia32/Makefile              |   2 +-
 arch/x86/ia32/ia32_binfmt.c         |   9 ++-
 arch/x86/ia32/ia32_signal.c         |  22 ++++---
 arch/x86/vdso/Makefile              |   2 +-
 arch/x86/vdso/vdso32-setup.c        | 119 ++++++++++++++++++++++++++++--------
 arch/x86/vdso/vdso32.S              |  16 +++--
 arch/x86/xen/setup.c                |   4 +-
 include/asm-x86/elf.h               |  23 ++++---
 10 files changed, 146 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index b427b7c0e5d0..b3f20beea411 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1988,6 +1988,11 @@ and is between 256 and 4096 characters. It is defined in the file
 			vdso=1: enable VDSO (default)
 			vdso=0: disable VDSO mapping
 
+	vdso32=		[X86-32,X86-64]
+			vdso32=2: enable compat VDSO (default with COMPAT_VDSO)
+			vdso32=1: enable 32-bit VDSO (default)
+			vdso32=0: disable 32-bit VDSO mapping
+
 	vector=		[IA-64,SMP]
 			vector=percpu: enable percpu vector domain
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 80b7ba4056db..2f4d88babd36 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1191,9 +1191,9 @@ config HOTPLUG_CPU
 config COMPAT_VDSO
 	bool "Compat VDSO support"
 	default y
-	depends on X86_32
+	depends on X86_32 || IA32_EMULATION
 	help
-	  Map the VDSO to the predictable old-style address too.
+	  Map the 32-bit VDSO to the predictable old-style address too.
 	---help---
 	  Say N here if you are running a sufficiently recent glibc
 	  version (2.3.3 or later), to remove the high-mapped
diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile
index a3c997e9f39a..1f58a21a41dc 100644
--- a/arch/x86/ia32/Makefile
+++ b/arch/x86/ia32/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o tls32.o \
-	ia32_binfmt.o fpu32.o ptrace32.o syscall32.o syscall32_syscall.o
+	ia32_binfmt.o fpu32.o ptrace32.o
 
 sysv-$(CONFIG_SYSVIPC) := ipc32.o
 obj-$(CONFIG_IA32_EMULATION) += $(sysv-y)
diff --git a/arch/x86/ia32/ia32_binfmt.c b/arch/x86/ia32/ia32_binfmt.c
index 55822d2cf053..e32974c3dd3b 100644
--- a/arch/x86/ia32/ia32_binfmt.c
+++ b/arch/x86/ia32/ia32_binfmt.c
@@ -26,7 +26,7 @@
 #include <asm/i387.h>
 #include <asm/uaccess.h>
 #include <asm/ia32.h>
-#include <asm/vsyscall32.h>
+#include <asm/vdso.h>
 
 #undef	ELF_ARCH
 #undef	ELF_CLASS
@@ -47,14 +47,13 @@
 #define AT_SYSINFO 32
 #define AT_SYSINFO_EHDR		33
 
-int sysctl_vsyscall32 = 1;
+extern int sysctl_vsyscall32;
 
 #undef ARCH_DLINFO
 #define ARCH_DLINFO do {  \
 	if (sysctl_vsyscall32) { \
-		current->mm->context.vdso = (void *)VSYSCALL32_BASE;	\
-		NEW_AUX_ENT(AT_SYSINFO, (u32)(u64)VSYSCALL32_VSYSCALL); \
-		NEW_AUX_ENT(AT_SYSINFO_EHDR, VSYSCALL32_BASE);    \
+		NEW_AUX_ENT(AT_SYSINFO, (u32)VDSO_ENTRY);		\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR, (u32)VDSO_CURRENT_BASE);	\
 	}	\
 } while(0)
 
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 0fc5d8563e19..39356a756b28 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -31,7 +31,7 @@
 #include <asm/sigcontext32.h>
 #include <asm/fpu32.h>
 #include <asm/proto.h>
-#include <asm/vsyscall32.h>
+#include <asm/vdso.h>
 
 #define DEBUG_SIG 0
 
@@ -465,13 +465,16 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 			goto give_sigsegv;
 	}
 
-	/* Return stub is in 32bit vsyscall page */
-	if (current->binfmt->hasvdso)
-		restorer = VSYSCALL32_SIGRETURN;
-	else
-		restorer = (void *)&frame->retcode;
-	if (ka->sa.sa_flags & SA_RESTORER)
+	if (ka->sa.sa_flags & SA_RESTORER) {
 		restorer = ka->sa.sa_restorer;
+	} else {
+		/* Return stub is in 32bit vsyscall page */
+		if (current->binfmt->hasvdso)
+			restorer = VDSO32_SYMBOL(current->mm->context.vdso,
+						 sigreturn);
+		else
+			restorer = (void *)&frame->retcode;
+	}
 	err |= __put_user(ptr_to_compat(restorer), &frame->pretcode);
 
 	/*
@@ -519,7 +522,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 {
 	struct rt_sigframe __user *frame;
 	struct exec_domain *ed = current_thread_info()->exec_domain;
-	void __user *restorer = VSYSCALL32_RTSIGRETURN;
+	void __user *restorer;
 	int err = 0;
 
 	/* __copy_to_user optimizes that into a single 8 byte store */
@@ -564,6 +567,9 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	if (ka->sa.sa_flags & SA_RESTORER)
 		restorer = ka->sa.sa_restorer;
+	else
+		restorer = VDSO32_SYMBOL(current->mm->context.vdso,
+					 rt_sigreturn);
 	err |= __put_user(ptr_to_compat(restorer), &frame->pretcode);
 
 	/*
diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile
index 1127c716df02..47bc2760e6a8 100644
--- a/arch/x86/vdso/Makefile
+++ b/arch/x86/vdso/Makefile
@@ -15,7 +15,7 @@ vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o vvar.o
 
 # files to link into kernel
 obj-$(VDSO64-y)			+= vma.o vdso.o
-obj-$(CONFIG_X86_32)		+= vdso32.o vdso32-setup.o
+obj-$(VDSO32-y)			+= vdso32.o vdso32-setup.o
 
 vobjs := $(foreach F,$(vobjs-y),$(obj)/$F)
 
diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c
index fb71a93c5dce..d97a6d7d062b 100644
--- a/arch/x86/vdso/vdso32-setup.c
+++ b/arch/x86/vdso/vdso32-setup.c
@@ -24,6 +24,7 @@
 #include <asm/elf.h>
 #include <asm/tlbflush.h>
 #include <asm/vdso.h>
+#include <asm/proto.h>
 
 enum {
 	VDSO_DISABLED = 0,
@@ -37,14 +38,24 @@ enum {
 #define VDSO_DEFAULT	VDSO_ENABLED
 #endif
 
+#ifdef CONFIG_X86_64
+#define vdso_enabled			sysctl_vsyscall32
+#define arch_setup_additional_pages	syscall32_setup_pages
+#endif
+
+/*
+ * This is the difference between the prelinked addresses in the vDSO images
+ * and the VDSO_HIGH_BASE address where CONFIG_COMPAT_VDSO places the vDSO
+ * in the user address space.
+ */
+#define VDSO_ADDR_ADJUST	(VDSO_HIGH_BASE - (unsigned long)VDSO32_PRELINK)
+
 /*
  * Should the kernel map a VDSO page into processes and pass its
  * address down to glibc upon exec()?
  */
 unsigned int __read_mostly vdso_enabled = VDSO_DEFAULT;
 
-EXPORT_SYMBOL_GPL(vdso_enabled);
-
 static int __init vdso_setup(char *s)
 {
 	vdso_enabled = simple_strtoul(s, NULL, 0);
@@ -52,9 +63,18 @@ static int __init vdso_setup(char *s)
 	return 1;
 }
 
-__setup("vdso=", vdso_setup);
+/*
+ * For consistency, the argument vdso32=[012] affects the 32-bit vDSO
+ * behavior on both 64-bit and 32-bit kernels.
+ * On 32-bit kernels, vdso=[012] means the same thing.
+ */
+__setup("vdso32=", vdso_setup);
 
-extern asmlinkage void ia32_sysenter_target(void);
+#ifdef CONFIG_X86_32
+__setup_param("vdso=", vdso32_setup, vdso_setup, 0);
+
+EXPORT_SYMBOL_GPL(vdso_enabled);
+#endif
 
 static __init void reloc_symtab(Elf32_Ehdr *ehdr,
 				unsigned offset, unsigned size)
@@ -79,7 +99,7 @@ static __init void reloc_symtab(Elf32_Ehdr *ehdr,
 		case STT_FUNC:
 		case STT_SECTION:
 		case STT_FILE:
-			sym->st_value += VDSO_HIGH_BASE;
+			sym->st_value += VDSO_ADDR_ADJUST;
 		}
 	}
 }
@@ -105,7 +125,7 @@ static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset)
 		case DT_VERNEED:
 		case DT_ADDRRNGLO ... DT_ADDRRNGHI:
 			/* definitely pointers needing relocation */
-			dyn->d_un.d_ptr += VDSO_HIGH_BASE;
+			dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
 			break;
 
 		case DT_ENCODING ... OLD_DT_LOOS-1:
@@ -114,7 +134,7 @@ static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset)
 			   they're even */
 			if (dyn->d_tag >= DT_ENCODING &&
 			    (dyn->d_tag & 1) == 0)
-				dyn->d_un.d_ptr += VDSO_HIGH_BASE;
+				dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
 			break;
 
 		case DT_VERDEFNUM:
@@ -143,15 +163,15 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr)
 	int i;
 
 	BUG_ON(memcmp(ehdr->e_ident, ELFMAG, 4) != 0 ||
-	       !elf_check_arch(ehdr) ||
+	       !elf_check_arch_ia32(ehdr) ||
 	       ehdr->e_type != ET_DYN);
 
-	ehdr->e_entry += VDSO_HIGH_BASE;
+	ehdr->e_entry += VDSO_ADDR_ADJUST;
 
 	/* rebase phdrs */
 	phdr = (void *)ehdr + ehdr->e_phoff;
 	for (i = 0; i < ehdr->e_phnum; i++) {
-		phdr[i].p_vaddr += VDSO_HIGH_BASE;
+		phdr[i].p_vaddr += VDSO_ADDR_ADJUST;
 
 		/* relocate dynamic stuff */
 		if (phdr[i].p_type == PT_DYNAMIC)
@@ -164,7 +184,7 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr)
 		if (!(shdr[i].sh_flags & SHF_ALLOC))
 			continue;
 
-		shdr[i].sh_addr += VDSO_HIGH_BASE;
+		shdr[i].sh_addr += VDSO_ADDR_ADJUST;
 
 		if (shdr[i].sh_type == SHT_SYMTAB ||
 		    shdr[i].sh_type == SHT_DYNSYM)
@@ -173,6 +193,45 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr)
 	}
 }
 
+/*
+ * These symbols are defined by vdso32.S to mark the bounds
+ * of the ELF DSO images included therein.
+ */
+extern const char vdso32_default_start, vdso32_default_end;
+extern const char vdso32_sysenter_start, vdso32_sysenter_end;
+static struct page *vdso32_pages[1];
+
+#ifdef CONFIG_X86_64
+
+static int use_sysenter __read_mostly = -1;
+
+#define	vdso32_sysenter()	(use_sysenter > 0)
+
+/* May not be __init: called during resume */
+void syscall32_cpu_init(void)
+{
+	if (use_sysenter < 0)
+		use_sysenter = (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL);
+
+	/* Load these always in case some future AMD CPU supports
+	   SYSENTER from compat mode too. */
+	checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
+	checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
+	checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
+
+	wrmsrl(MSR_CSTAR, ia32_cstar_target);
+}
+
+#define compat_uses_vma		1
+
+static inline void map_compat_vdso(int map)
+{
+}
+
+#else  /* CONFIG_X86_32 */
+
+#define vdso32_sysenter()	(boot_cpu_has(X86_FEATURE_SEP))
+
 void enable_sep_cpu(void)
 {
 	int cpu = get_cpu();
@@ -210,13 +269,7 @@ static int __init gate_vma_init(void)
 	return 0;
 }
 
-/*
- * These symbols are defined by vsyscall.o to mark the bounds
- * of the ELF DSO images included therein.
- */
-extern const char vsyscall_int80_start, vsyscall_int80_end;
-extern const char vsyscall_sysenter_start, vsyscall_sysenter_end;
-static struct page *syscall_pages[1];
+#define compat_uses_vma		0
 
 static void map_compat_vdso(int map)
 {
@@ -227,31 +280,35 @@ static void map_compat_vdso(int map)
 
 	vdso_mapped = map;
 
-	__set_fixmap(FIX_VDSO, page_to_pfn(syscall_pages[0]) << PAGE_SHIFT,
+	__set_fixmap(FIX_VDSO, page_to_pfn(vdso32_pages[0]) << PAGE_SHIFT,
 		     map ? PAGE_READONLY_EXEC : PAGE_NONE);
 
 	/* flush stray tlbs */
 	flush_tlb_all();
 }
 
+#endif	/* CONFIG_X86_64 */
+
 int __init sysenter_setup(void)
 {
 	void *syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
 	const void *vsyscall;
 	size_t vsyscall_len;
 
-	syscall_pages[0] = virt_to_page(syscall_page);
+	vdso32_pages[0] = virt_to_page(syscall_page);
 
+#ifdef CONFIG_X86_32
 	gate_vma_init();
 
 	printk("Compat vDSO mapped to %08lx.\n", __fix_to_virt(FIX_VDSO));
+#endif
 
-	if (!boot_cpu_has(X86_FEATURE_SEP)) {
-		vsyscall = &vsyscall_int80_start;
-		vsyscall_len = &vsyscall_int80_end - &vsyscall_int80_start;
+	if (!vdso32_sysenter()) {
+		vsyscall = &vdso32_default_start;
+		vsyscall_len = &vdso32_default_end - &vdso32_default_start;
 	} else {
-		vsyscall = &vsyscall_sysenter_start;
-		vsyscall_len = &vsyscall_sysenter_end - &vsyscall_sysenter_start;
+		vsyscall = &vdso32_sysenter_start;
+		vsyscall_len = &vdso32_sysenter_end - &vdso32_sysenter_start;
 	}
 
 	memcpy(syscall_page, vsyscall, vsyscall_len);
@@ -284,7 +341,9 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 			ret = addr;
 			goto up_fail;
 		}
+	}
 
+	if (compat_uses_vma || !compat) {
 		/*
 		 * MAYWRITE to allow gdb to COW and set breakpoints
 		 *
@@ -298,7 +357,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 					      VM_READ|VM_EXEC|
 					      VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC|
 					      VM_ALWAYSDUMP,
-					      syscall_pages);
+					      vdso32_pages);
 
 		if (ret)
 			goto up_fail;
@@ -314,6 +373,12 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 	return ret;
 }
 
+#ifdef CONFIG_X86_64
+
+__initcall(sysenter_setup);
+
+#else  /* CONFIG_X86_32 */
+
 const char *arch_vma_name(struct vm_area_struct *vma)
 {
 	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
@@ -342,3 +407,5 @@ int in_gate_area_no_task(unsigned long addr)
 {
 	return 0;
 }
+
+#endif	/* CONFIG_X86_64 */
diff --git a/arch/x86/vdso/vdso32.S b/arch/x86/vdso/vdso32.S
index cab020c99c3d..1e36f72cab86 100644
--- a/arch/x86/vdso/vdso32.S
+++ b/arch/x86/vdso/vdso32.S
@@ -2,14 +2,18 @@
 
 __INITDATA
 
-	.globl vsyscall_int80_start, vsyscall_int80_end
-vsyscall_int80_start:
+	.globl vdso32_default_start, vdso32_default_end
+vdso32_default_start:
+#ifdef CONFIG_X86_32
 	.incbin "arch/x86/vdso/vdso32-int80.so"
-vsyscall_int80_end:
+#else
+	.incbin "arch/x86/vdso/vdso32-syscall.so"
+#endif
+vdso32_default_end:
 
-	.globl vsyscall_sysenter_start, vsyscall_sysenter_end
-vsyscall_sysenter_start:
+	.globl vdso32_sysenter_start, vdso32_sysenter_end
+vdso32_sysenter_start:
 	.incbin "arch/x86/vdso/vdso32-sysenter.so"
-vsyscall_sysenter_end:
+vdso32_sysenter_end:
 
 __FINIT
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index fd91568090f4..7d6d0ef55890 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -62,8 +62,8 @@ static void xen_idle(void)
  */
 static void fiddle_vdso(void)
 {
-	extern char vsyscall_int80_start;
-	u32 *mask = VDSO32_SYMBOL(&vsyscall_int80_start, NOTE_MASK);
+	extern const char vdso32_default_start;
+	u32 *mask = VDSO32_SYMBOL(&vdso32_default_start, NOTE_MASK);
 	*mask |= 1 << VDSO_NOTE_NONEGSEG_BIT;
 }
 
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index 70edff2d5671..60f5101d9483 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -74,17 +74,19 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 
 #ifdef __KERNEL__
 
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch_ia32(x) \
+	(((x)->e_machine == EM_386) || ((x)->e_machine == EM_486))
+
 #ifdef CONFIG_X86_32
 #include <asm/processor.h>
 #include <asm/system.h>		/* for savesegment */
 #include <asm/desc.h>
 #include <asm/vdso.h>
 
-/*
- * This is used to ensure we don't load something for the wrong architecture.
- */
-#define elf_check_arch(x) \
-	(((x)->e_machine == EM_386) || ((x)->e_machine == EM_486))
+#define elf_check_arch(x)	elf_check_arch_ia32(x)
 
 /* SVR4/i386 ABI (pages 3-31, 3-32) says that when the program starts %edx
    contains a pointer to a function which might be registered using `atexit'.
@@ -247,10 +249,6 @@ extern int dump_task_extended_fpu (struct task_struct *,
 #define ELF_CORE_XFPREG_TYPE NT_PRXFPREG
 
 #define VDSO_HIGH_BASE		(__fix_to_virt(FIX_VDSO))
-#define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
-
-#define VDSO_ENTRY \
-	((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
 
 /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
 
@@ -262,6 +260,8 @@ do if (vdso_enabled) {							\
 
 #else /* CONFIG_X86_32 */
 
+#define VDSO_HIGH_BASE		0xffffe000U /* CONFIG_COMPAT_VDSO address */
+
 /* 1GB for 64bit, 8MB for 32bit */
 #define STACK_RND_MASK (test_thread_flag(TIF_IA32) ? 0x7ff : 0x3fffff)
 
@@ -272,6 +272,11 @@ do if (vdso_enabled) {						\
 
 #endif /* !CONFIG_X86_32 */
 
+#define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
+
+#define VDSO_ENTRY \
+	((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
+
 struct linux_binprm;
 
 #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
-- 
cgit v1.2.3


From 16f4bc738d616962a844e80f7b1fcb52c626542a Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:44 +0100
Subject: x86 vDSO: ia32 vsyscall removal

This removes all the old vsyscall code from arch/x86/ia32/ that is
no longer used because arch/x86/vdso/ code has replaced it.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Makefile_64               |   3 -
 arch/x86/ia32/Makefile             |  37 ----------
 arch/x86/ia32/syscall32.c          |  90 -----------------------
 arch/x86/ia32/syscall32_syscall.S  |  17 -----
 arch/x86/ia32/vsyscall-sigreturn.S | 142 -------------------------------------
 arch/x86/ia32/vsyscall-syscall.S   |  69 ------------------
 arch/x86/ia32/vsyscall-sysenter.S  |  95 -------------------------
 arch/x86/ia32/vsyscall.lds         |  80 ---------------------
 include/asm-x86/Kbuild             |   1 -
 include/asm-x86/vsyscall32.h       |  20 ------
 10 files changed, 554 deletions(-)
 delete mode 100644 arch/x86/ia32/syscall32.c
 delete mode 100644 arch/x86/ia32/syscall32_syscall.S
 delete mode 100644 arch/x86/ia32/vsyscall-sigreturn.S
 delete mode 100644 arch/x86/ia32/vsyscall-syscall.S
 delete mode 100644 arch/x86/ia32/vsyscall-sysenter.S
 delete mode 100644 arch/x86/ia32/vsyscall.lds
 delete mode 100644 include/asm-x86/vsyscall32.h

(limited to 'include')

diff --git a/arch/x86/Makefile_64 b/arch/x86/Makefile_64
index 93cf17769ff5..d35a0c50b67b 100644
--- a/arch/x86/Makefile_64
+++ b/arch/x86/Makefile_64
@@ -117,9 +117,6 @@ install: vdso_install
 	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(BOOTIMAGE) $@ 
 
 vdso_install:
-ifeq ($(CONFIG_IA32_EMULATION),y)
-	$(Q)$(MAKE) $(build)=arch/x86/ia32 $@
-endif
 	$(Q)$(MAKE) $(build)=arch/x86/vdso $@
 
 archclean:
diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile
index 1f58a21a41dc..8c19b763b2fb 100644
--- a/arch/x86/ia32/Makefile
+++ b/arch/x86/ia32/Makefile
@@ -12,40 +12,3 @@ obj-$(CONFIG_IA32_AOUT) += ia32_aout.o
 
 audit-class-$(CONFIG_AUDIT) := audit.o
 obj-$(CONFIG_IA32_EMULATION) += $(audit-class-y)
-
-$(obj)/syscall32_syscall.o: \
-	$(foreach F,sysenter syscall,$(obj)/vsyscall-$F.so)
-
-# Teach kbuild about targets
-targets := $(foreach F,$(addprefix vsyscall-,sysenter syscall),\
-		     $F.o $F.so $F.so.dbg)
-
-# The DSO images are built using a special linker script
-quiet_cmd_syscall = SYSCALL $@
-      cmd_syscall = $(CC) -m32 -nostdlib -shared \
-			  $(call ld-option, -Wl$(comma)--hash-style=sysv) \
-			   -Wl,-soname=linux-gate.so.1 -o $@ \
-			   -Wl,-T,$(filter-out FORCE,$^)
-
-$(obj)/%.so: OBJCOPYFLAGS := -S
-$(obj)/%.so: $(obj)/%.so.dbg FORCE
-	$(call if_changed,objcopy)
-
-$(obj)/vsyscall-sysenter.so.dbg $(obj)/vsyscall-syscall.so.dbg: \
-$(obj)/vsyscall-%.so.dbg: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
-	$(call if_changed,syscall)
-
-AFLAGS_vsyscall-sysenter.o = -m32 -Wa,-32
-AFLAGS_vsyscall-syscall.o = -m32 -Wa,-32
-
-vdsos := vdso32-sysenter.so vdso32-syscall.so
-
-quiet_cmd_vdso_install = INSTALL $@
-      cmd_vdso_install = cp $(@:vdso32-%.so=$(obj)/vsyscall-%.so.dbg) \
-			    $(MODLIB)/vdso/$@
-
-$(vdsos):
-	@mkdir -p $(MODLIB)/vdso
-	$(call cmd,vdso_install)
-
-vdso_install: $(vdsos)
diff --git a/arch/x86/ia32/syscall32.c b/arch/x86/ia32/syscall32.c
deleted file mode 100644
index 98ff99f5b59a..000000000000
--- a/arch/x86/ia32/syscall32.c
+++ /dev/null
@@ -1,90 +0,0 @@
-/*
- * Copyright 2002,2003 Andi Kleen, SuSE Labs
- *
- * vsyscall handling for 32bit processes. Map a stub page into it on
- * demand because 32bit cannot reach the kernel's fixmaps
- */
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/kernel.h>
-#include <linux/gfp.h>
-#include <linux/init.h>
-#include <linux/stringify.h>
-#include <linux/security.h>
-#include <asm/proto.h>
-#include <asm/tlbflush.h>
-#include <asm/ia32_unistd.h>
-#include <asm/vsyscall32.h>
-
-extern unsigned char syscall32_syscall[], syscall32_syscall_end[];
-extern unsigned char syscall32_sysenter[], syscall32_sysenter_end[];
-extern int sysctl_vsyscall32;
-
-static struct page *syscall32_pages[1];
-static int use_sysenter = -1;
-
-struct linux_binprm;
-
-/* Setup a VMA at program startup for the vsyscall page */
-int syscall32_setup_pages(struct linux_binprm *bprm, int exstack)
-{
-	struct mm_struct *mm = current->mm;
-	int ret;
-
-	down_write(&mm->mmap_sem);
-	/*
-	 * MAYWRITE to allow gdb to COW and set breakpoints
-	 *
-	 * Make sure the vDSO gets into every core dump.
-	 * Dumping its contents makes post-mortem fully interpretable later
-	 * without matching up the same kernel and hardware config to see
-	 * what PC values meant.
-	 */
-	/* Could randomize here */
-	ret = install_special_mapping(mm, VSYSCALL32_BASE, PAGE_SIZE,
-				      VM_READ|VM_EXEC|
-				      VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC|
-				      VM_ALWAYSDUMP,
-				      syscall32_pages);
-	if (ret == 0) {
-		current->mm->context.vdso = (void __user *)VSYSCALL32_BASE;
-		current_thread_info()->sysenter_return = VSYSCALL32_SYSEXIT;
-	}
-	up_write(&mm->mmap_sem);
-	return ret;
-}
-
-static int __init init_syscall32(void)
-{
-	char *syscall32_page = (void *)get_zeroed_page(GFP_KERNEL);
-
-	if (!syscall32_page)
-		panic("Cannot allocate syscall32 page");
-	syscall32_pages[0] = virt_to_page(syscall32_page);
-	if (use_sysenter > 0) {
-		memcpy(syscall32_page, syscall32_sysenter,
-		       syscall32_sysenter_end - syscall32_sysenter);
-	} else {
-		memcpy(syscall32_page, syscall32_syscall,
-		       syscall32_syscall_end - syscall32_syscall);
-	}
-	return 0;
-}
-__initcall(init_syscall32);
-
-/* May not be __init: called during resume */
-void syscall32_cpu_init(void)
-{
-	if (use_sysenter < 0)
-		use_sysenter = (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL);
-
-	/*
-	 * Load these always in case some future AMD CPU supports
-	 * SYSENTER from compat mode too.
-	 */
-	checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
-	checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
-	checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
-
-	wrmsrl(MSR_CSTAR, ia32_cstar_target);
-}
diff --git a/arch/x86/ia32/syscall32_syscall.S b/arch/x86/ia32/syscall32_syscall.S
deleted file mode 100644
index 933f0f08b1cf..000000000000
--- a/arch/x86/ia32/syscall32_syscall.S
+++ /dev/null
@@ -1,17 +0,0 @@
-/* 32bit VDSOs mapped into user space. */
-
-	.section ".init.data","aw"
-
-	.globl syscall32_syscall
-	.globl syscall32_syscall_end
-
-syscall32_syscall:
-	.incbin "arch/x86/ia32/vsyscall-syscall.so"
-syscall32_syscall_end:
-
-	.globl syscall32_sysenter
-	.globl syscall32_sysenter_end
-
-syscall32_sysenter:
-	.incbin "arch/x86/ia32/vsyscall-sysenter.so"
-syscall32_sysenter_end:
diff --git a/arch/x86/ia32/vsyscall-sigreturn.S b/arch/x86/ia32/vsyscall-sigreturn.S
deleted file mode 100644
index 295eecf91f17..000000000000
--- a/arch/x86/ia32/vsyscall-sigreturn.S
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- * Common code for the sigreturn entry points on the vsyscall page.
- * This code uses SYSCALL_ENTER_KERNEL (either syscall or int $0x80)
- * to enter the kernel.
- * This file is #include'd by vsyscall-*.S to define them after the
- * vsyscall entry point.  The addresses we get for these entry points
- * by doing ".balign 32" must match in both versions of the page.
- */
-
-	.code32
-	.section .text.sigreturn,"ax"
-	.balign 32
-	.globl __kernel_sigreturn
-	.type __kernel_sigreturn,@function
-__kernel_sigreturn:
-.LSTART_sigreturn:
-	popl %eax
-	movl $__NR_ia32_sigreturn, %eax
-	SYSCALL_ENTER_KERNEL
-.LEND_sigreturn:
-	.size __kernel_sigreturn,.-.LSTART_sigreturn
-
-	.section .text.rtsigreturn,"ax"
-	.balign 32
-	.globl __kernel_rt_sigreturn
-	.type __kernel_rt_sigreturn,@function
-__kernel_rt_sigreturn:
-.LSTART_rt_sigreturn:
-	movl $__NR_ia32_rt_sigreturn, %eax
-	SYSCALL_ENTER_KERNEL
-.LEND_rt_sigreturn:
-	.size __kernel_rt_sigreturn,.-.LSTART_rt_sigreturn
-
-	.section .eh_frame,"a",@progbits
-.LSTARTFRAMES:
-        .long .LENDCIES-.LSTARTCIES
-.LSTARTCIES:
-	.long 0			/* CIE ID */
-	.byte 1			/* Version number */
-	.string "zRS"		/* NUL-terminated augmentation string */
-	.uleb128 1		/* Code alignment factor */
-	.sleb128 -4		/* Data alignment factor */
-	.byte 8			/* Return address register column */
-	.uleb128 1		/* Augmentation value length */
-	.byte 0x1b		/* DW_EH_PE_pcrel|DW_EH_PE_sdata4. */
-	.byte 0x0c		/* DW_CFA_def_cfa */
-	.uleb128 4
-	.uleb128 4
-	.byte 0x88		/* DW_CFA_offset, column 0x8 */
-	.uleb128 1
-	.align 4
-.LENDCIES:
-
-	.long .LENDFDE2-.LSTARTFDE2	/* Length FDE */
-.LSTARTFDE2:
-	.long .LSTARTFDE2-.LSTARTFRAMES	/* CIE pointer */
-	/* HACK: The dwarf2 unwind routines will subtract 1 from the
-	   return address to get an address in the middle of the
-	   presumed call instruction.  Since we didn't get here via
-	   a call, we need to include the nop before the real start
-	   to make up for it.  */
-	.long .LSTART_sigreturn-1-.	/* PC-relative start address */
-	.long .LEND_sigreturn-.LSTART_sigreturn+1
-	.uleb128 0			/* Augmentation length */
-	/* What follows are the instructions for the table generation.
-	   We record the locations of each register saved.  This is
-	   complicated by the fact that the "CFA" is always assumed to
-	   be the value of the stack pointer in the caller.  This means
-	   that we must define the CFA of this body of code to be the
-	   saved value of the stack pointer in the sigcontext.  Which
-	   also means that there is no fixed relation to the other 
-	   saved registers, which means that we must use DW_CFA_expression
-	   to compute their addresses.  It also means that when we 
-	   adjust the stack with the popl, we have to do it all over again.  */
-
-#define do_cfa_expr(offset)						\
-	.byte 0x0f;			/* DW_CFA_def_cfa_expression */	\
-	.uleb128 1f-0f;			/*   length */			\
-0:	.byte 0x74;			/*     DW_OP_breg4 */		\
-	.sleb128 offset;		/*      offset */		\
-	.byte 0x06;			/*     DW_OP_deref */		\
-1:
-
-#define do_expr(regno, offset)						\
-	.byte 0x10;			/* DW_CFA_expression */		\
-	.uleb128 regno;			/*   regno */			\
-	.uleb128 1f-0f;			/*   length */			\
-0:	.byte 0x74;			/*     DW_OP_breg4 */		\
-	.sleb128 offset;		/*       offset */		\
-1:
-
-	do_cfa_expr(IA32_SIGCONTEXT_esp+4)
-	do_expr(0, IA32_SIGCONTEXT_eax+4)
-	do_expr(1, IA32_SIGCONTEXT_ecx+4)
-	do_expr(2, IA32_SIGCONTEXT_edx+4)
-	do_expr(3, IA32_SIGCONTEXT_ebx+4)
-	do_expr(5, IA32_SIGCONTEXT_ebp+4)
-	do_expr(6, IA32_SIGCONTEXT_esi+4)
-	do_expr(7, IA32_SIGCONTEXT_edi+4)
-	do_expr(8, IA32_SIGCONTEXT_eip+4)
-
-	.byte 0x42	/* DW_CFA_advance_loc 2 -- nop; popl eax. */
-
-	do_cfa_expr(IA32_SIGCONTEXT_esp)
-	do_expr(0, IA32_SIGCONTEXT_eax)
-	do_expr(1, IA32_SIGCONTEXT_ecx)
-	do_expr(2, IA32_SIGCONTEXT_edx)
-	do_expr(3, IA32_SIGCONTEXT_ebx)
-	do_expr(5, IA32_SIGCONTEXT_ebp)
-	do_expr(6, IA32_SIGCONTEXT_esi)
-	do_expr(7, IA32_SIGCONTEXT_edi)
-	do_expr(8, IA32_SIGCONTEXT_eip)
-
-	.align 4
-.LENDFDE2:
-
-	.long .LENDFDE3-.LSTARTFDE3	/* Length FDE */
-.LSTARTFDE3:
-	.long .LSTARTFDE3-.LSTARTFRAMES	/* CIE pointer */
-	/* HACK: See above wrt unwind library assumptions.  */
-	.long .LSTART_rt_sigreturn-1-.	/* PC-relative start address */
-	.long .LEND_rt_sigreturn-.LSTART_rt_sigreturn+1
-	.uleb128 0			/* Augmentation */
-	/* What follows are the instructions for the table generation.
-	   We record the locations of each register saved.  This is
-	   slightly less complicated than the above, since we don't
-	   modify the stack pointer in the process.  */
-
-	do_cfa_expr(IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_esp)
-	do_expr(0, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_eax)
-	do_expr(1, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ecx)
-	do_expr(2, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_edx)
-	do_expr(3, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ebx)
-	do_expr(5, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ebp)
-	do_expr(6, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_esi)
-	do_expr(7, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_edi)
-	do_expr(8, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_eip)
-
-	.align 4
-.LENDFDE3:
-
-#include "../vdso/vdso32/note.S"
diff --git a/arch/x86/ia32/vsyscall-syscall.S b/arch/x86/ia32/vsyscall-syscall.S
deleted file mode 100644
index cf9ef678de3e..000000000000
--- a/arch/x86/ia32/vsyscall-syscall.S
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * Code for the vsyscall page.  This version uses the syscall instruction.
- */
-
-#include <asm/ia32_unistd.h>
-#include <asm/asm-offsets.h>
-#include <asm/segment.h>
-
-	.code32
-	.text
-	.section .text.vsyscall,"ax"
-	.globl __kernel_vsyscall
-	.type __kernel_vsyscall,@function
-__kernel_vsyscall:
-.LSTART_vsyscall:
-	push	%ebp
-.Lpush_ebp:
-	movl	%ecx, %ebp
-	syscall
-	movl	$__USER32_DS, %ecx
-	movl	%ecx, %ss
-	movl	%ebp, %ecx
-	popl	%ebp
-.Lpop_ebp:
-	ret
-.LEND_vsyscall:
-	.size __kernel_vsyscall,.-.LSTART_vsyscall
-
-	.section .eh_frame,"a",@progbits
-.LSTARTFRAME:
-	.long .LENDCIE-.LSTARTCIE
-.LSTARTCIE:
-	.long 0			/* CIE ID */
-	.byte 1			/* Version number */
-	.string "zR"		/* NUL-terminated augmentation string */
-	.uleb128 1		/* Code alignment factor */
-	.sleb128 -4		/* Data alignment factor */
-	.byte 8			/* Return address register column */
-	.uleb128 1		/* Augmentation value length */
-	.byte 0x1b		/* DW_EH_PE_pcrel|DW_EH_PE_sdata4. */
-	.byte 0x0c		/* DW_CFA_def_cfa */
-	.uleb128 4
-	.uleb128 4
-	.byte 0x88		/* DW_CFA_offset, column 0x8 */
-	.uleb128 1
-	.align 4
-.LENDCIE:
-
-	.long .LENDFDE1-.LSTARTFDE1	/* Length FDE */
-.LSTARTFDE1:
-	.long .LSTARTFDE1-.LSTARTFRAME	/* CIE pointer */
-	.long .LSTART_vsyscall-.	/* PC-relative start address */
-	.long .LEND_vsyscall-.LSTART_vsyscall
-	.uleb128 0			/* Augmentation length */
-	/* What follows are the instructions for the table generation.
-	   We have to record all changes of the stack pointer.  */
-	.byte 0x40 + .Lpush_ebp-.LSTART_vsyscall /* DW_CFA_advance_loc */
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.uleb128 8
-	.byte 0x85, 0x02	/* DW_CFA_offset %ebp -8 */
-	.byte 0x40 + .Lpop_ebp-.Lpush_ebp /* DW_CFA_advance_loc */
-	.byte 0xc5		/* DW_CFA_restore %ebp */
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.uleb128 4
-	.align 4
-.LENDFDE1:
-
-#define SYSCALL_ENTER_KERNEL	syscall
-#include "vsyscall-sigreturn.S"
diff --git a/arch/x86/ia32/vsyscall-sysenter.S b/arch/x86/ia32/vsyscall-sysenter.S
deleted file mode 100644
index ae056e553d13..000000000000
--- a/arch/x86/ia32/vsyscall-sysenter.S
+++ /dev/null
@@ -1,95 +0,0 @@
-/*
- * Code for the vsyscall page.  This version uses the sysenter instruction.
- */
-
-#include <asm/ia32_unistd.h>
-#include <asm/asm-offsets.h>
-
-	.code32
-	.text
-	.section .text.vsyscall,"ax"
-	.globl __kernel_vsyscall
-	.type __kernel_vsyscall,@function
-__kernel_vsyscall:
-.LSTART_vsyscall:
-	push	%ecx
-.Lpush_ecx:
-	push	%edx
-.Lpush_edx:
-	push	%ebp
-.Lenter_kernel:
-	movl	%esp,%ebp
-	sysenter
-	.space 7,0x90
-	jmp	.Lenter_kernel
-	/* 16: System call normal return point is here! */
-	pop	%ebp
-.Lpop_ebp:
-	pop	%edx
-.Lpop_edx:
-	pop	%ecx
-.Lpop_ecx:
-	ret
-.LEND_vsyscall:
-	.size __kernel_vsyscall,.-.LSTART_vsyscall
-
-	.section .eh_frame,"a",@progbits
-.LSTARTFRAME:
-	.long .LENDCIE-.LSTARTCIE
-.LSTARTCIE:
-	.long 0			/* CIE ID */
-	.byte 1			/* Version number */
-	.string "zR"		/* NUL-terminated augmentation string */
-	.uleb128 1		/* Code alignment factor */
-	.sleb128 -4		/* Data alignment factor */
-	.byte 8			/* Return address register column */
-	.uleb128 1		/* Augmentation value length */
-	.byte 0x1b		/* DW_EH_PE_pcrel|DW_EH_PE_sdata4. */
-	.byte 0x0c		/* DW_CFA_def_cfa */
-	.uleb128 4
-	.uleb128 4
-	.byte 0x88		/* DW_CFA_offset, column 0x8 */
-	.uleb128 1
-	.align 4
-.LENDCIE:
-
-	.long .LENDFDE1-.LSTARTFDE1	/* Length FDE */
-.LSTARTFDE1:
-	.long .LSTARTFDE1-.LSTARTFRAME	/* CIE pointer */
-	.long .LSTART_vsyscall-.	/* PC-relative start address */
-	.long .LEND_vsyscall-.LSTART_vsyscall
-	.uleb128 0			/* Augmentation length */
-	/* What follows are the instructions for the table generation.
-	   We have to record all changes of the stack pointer.  */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lpush_ecx-.LSTART_vsyscall
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x08		/* RA at offset 8 now */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lpush_edx-.Lpush_ecx
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x0c		/* RA at offset 12 now */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lenter_kernel-.Lpush_edx
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x10		/* RA at offset 16 now */
-	.byte 0x85, 0x04	/* DW_CFA_offset %ebp -16 */
-	/* Finally the epilogue.  */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lpop_ebp-.Lenter_kernel
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x12		/* RA at offset 12 now */
-	.byte 0xc5		/* DW_CFA_restore %ebp */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lpop_edx-.Lpop_ebp
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x08		/* RA at offset 8 now */
-	.byte 0x04		/* DW_CFA_advance_loc4 */
-	.long .Lpop_ecx-.Lpop_edx
-	.byte 0x0e		/* DW_CFA_def_cfa_offset */
-	.byte 0x04		/* RA at offset 4 now */
-	.align 4
-.LENDFDE1:
-
-#define SYSCALL_ENTER_KERNEL	int $0x80
-#include "vsyscall-sigreturn.S"
diff --git a/arch/x86/ia32/vsyscall.lds b/arch/x86/ia32/vsyscall.lds
deleted file mode 100644
index 1dc86ff5bcb9..000000000000
--- a/arch/x86/ia32/vsyscall.lds
+++ /dev/null
@@ -1,80 +0,0 @@
-/*
- * Linker script for vsyscall DSO.  The vsyscall page is an ELF shared
- * object prelinked to its virtual address. This script controls its layout.
- */
-
-/* This must match <asm/fixmap.h>.  */
-VSYSCALL_BASE = 0xffffe000;
-
-SECTIONS
-{
-  . = VSYSCALL_BASE + SIZEOF_HEADERS;
-
-  .hash           : { *(.hash) }		:text
-  .gnu.hash       : { *(.gnu.hash) }
-  .dynsym         : { *(.dynsym) }
-  .dynstr         : { *(.dynstr) }
-  .gnu.version    : { *(.gnu.version) }
-  .gnu.version_d  : { *(.gnu.version_d) }
-  .gnu.version_r  : { *(.gnu.version_r) }
-
-  /* This linker script is used both with -r and with -shared.
-     For the layouts to match, we need to skip more than enough
-     space for the dynamic symbol table et al.  If this amount
-     is insufficient, ld -shared will barf.  Just increase it here.  */
-  . = VSYSCALL_BASE + 0x400;
-  
-  .text.vsyscall   : { *(.text.vsyscall) } 	:text =0x90909090
-
-  /* This is an 32bit object and we cannot easily get the offsets
-     into the 64bit kernel. Just hardcode them here. This assumes
-     that all the stubs don't need more than 0x100 bytes. */
-  . = VSYSCALL_BASE + 0x500;
-
-  .text.sigreturn  : { *(.text.sigreturn) }	:text =0x90909090
-
-  . = VSYSCALL_BASE + 0x600;
-
-  .text.rtsigreturn : { *(.text.rtsigreturn) }   :text =0x90909090
-	
-  .note		  : { *(.note.*) }		:text :note
-  .eh_frame_hdr   : { *(.eh_frame_hdr) }	:text :eh_frame_hdr
-  .eh_frame       : { KEEP (*(.eh_frame)) }	:text
-  .dynamic        : { *(.dynamic) }		:text :dynamic
-  .useless        : {
-  	*(.got.plt) *(.got)
-	*(.data .data.* .gnu.linkonce.d.*)
-	*(.dynbss)
-	*(.bss .bss.* .gnu.linkonce.b.*)
-  }						:text
-}
-
-/*
- * We must supply the ELF program headers explicitly to get just one
- * PT_LOAD segment, and set the flags explicitly to make segments read-only.
- */
-PHDRS
-{
-  text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
-  dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
-  note PT_NOTE FLAGS(4); /* PF_R */
-  eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */
-}
-
-/*
- * This controls what symbols we export from the DSO.
- */
-VERSION
-{
-  LINUX_2.5 {
-    global:
-    	__kernel_vsyscall;
-    	__kernel_sigreturn;
-    	__kernel_rt_sigreturn;
-
-    local: *;
-  };
-}
-
-/* The ELF entry point can be used to set the AT_SYSINFO value.  */
-ENTRY(__kernel_vsyscall);
diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index 12db5a1cdd74..22e3a284c1ed 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -9,7 +9,6 @@ header-y += prctl.h
 header-y += ptrace-abi.h
 header-y += sigcontext32.h
 header-y += ucontext.h
-header-y += vsyscall32.h
 
 unifdef-y += e820.h
 unifdef-y += ist.h
diff --git a/include/asm-x86/vsyscall32.h b/include/asm-x86/vsyscall32.h
deleted file mode 100644
index c631c082f8f7..000000000000
--- a/include/asm-x86/vsyscall32.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_VSYSCALL32_H
-#define _ASM_VSYSCALL32_H 1
-
-/* Values need to match arch/x86_64/ia32/vsyscall.lds */
-
-#ifdef __ASSEMBLY__
-#define VSYSCALL32_BASE 0xffffe000
-#define VSYSCALL32_SYSEXIT (VSYSCALL32_BASE + 0x410)
-#else
-#define VSYSCALL32_BASE 0xffffe000UL
-#define VSYSCALL32_END (VSYSCALL32_BASE + PAGE_SIZE)
-#define VSYSCALL32_EHDR ((const struct elf32_hdr *) VSYSCALL32_BASE)
-
-#define VSYSCALL32_VSYSCALL ((void *)VSYSCALL32_BASE + 0x400) 
-#define VSYSCALL32_SYSEXIT ((void *)VSYSCALL32_BASE + 0x410)
-#define VSYSCALL32_SIGRETURN ((void __user *)VSYSCALL32_BASE + 0x500) 
-#define VSYSCALL32_RTSIGRETURN ((void __user *)VSYSCALL32_BASE + 0x600) 
-#endif
-
-#endif
-- 
cgit v1.2.3


From 859c0a5b9ca8eba2d65fa42f02505d76f8a712c7 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:44 +0100
Subject: x86: get_desc_base

This defines the get_desc_base function in asm-x86/desc_64.h to match the
one in desc_32.h.  If these two files ever get merged together, this
function could be the same in both.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_64.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 7d48df72aef2..230ac6e50a0f 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -215,6 +215,16 @@ static inline void load_LDT(mm_context_t *pc)
 
 extern struct desc_ptr idt_descr;
 
+static inline unsigned long get_desc_base(const void *ptr)
+{
+	const u32 *desc = ptr;
+	unsigned long base;
+	base = ((desc[0] >> 16)  & 0x0000ffff) |
+		((desc[1] << 16) & 0x00ff0000) |
+		(desc[1] & 0xff000000);
+	return base;
+}
+
 #endif /* !__ASSEMBLY__ */
 
 #endif
-- 
cgit v1.2.3


From 12c3cbbb7e3218b1c3ed4c97b3496fc3f94ea967 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:45 +0100
Subject: x86: desc_empty

This replaces the desc_empty macro with an inline.  It now handles
easily any of the four different types used between 32/64 code to
refer to these 8 bytes.  It's identical in both asm-x86/processor_64.h
and asm-x86/processor_32.h, so if these files ever get merged this
function can be in the common code.

This also removes the desc_equal macro because nothing uses it.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor_32.h |  9 +++++----
 include/asm-x86/processor_64.h | 10 +++++-----
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 3b51a18f16d5..2540bf8d5724 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -28,11 +28,12 @@ struct desc_struct {
 	unsigned long a,b;
 };
 
-#define desc_empty(desc) \
-		(!((desc)->a | (desc)->b))
+static inline int desc_empty(const void *ptr)
+{
+	const u32 *desc = ptr;
+	return !(desc[0] | desc[1]);
+}
 
-#define desc_equal(desc1, desc2) \
-		(((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
 /*
  * Default implementation of macro that returns current
  * instruction pointer ("program counter").
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 2ae8ceb8a74e..20d8935d141a 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -30,11 +30,11 @@
 #define VIP_MASK	0x00100000	/* virtual interrupt pending */
 #define ID_MASK		0x00200000
 
-#define desc_empty(desc) \
-               (!((desc)->a | (desc)->b))
-
-#define desc_equal(desc1, desc2) \
-               (((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
+static inline int desc_empty(const void *ptr)
+{
+	const u32 *desc = ptr;
+	return !(desc[0] | desc[1]);
+}
 
 /*
  * Default implementation of macro that returns current
-- 
cgit v1.2.3


From efd1ca52d04d2f6df337a3332cee56cd60e6d4c4 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:46 +0100
Subject: x86: TLS cleanup

This consolidates the four different places that implemented the same
encoding magic for the GDT-slot 32-bit TLS support.  The old tls32.c was
renamed and is now only slightly modified to be the shared implementation.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Zachary Amsden <zach@vmware.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ia32entry.S    |   4 +-
 arch/x86/kernel/Makefile_32  |   1 +
 arch/x86/kernel/process_32.c | 141 ++-----------------------------------------
 arch/x86/kernel/process_64.c |   3 +-
 arch/x86/kernel/ptrace_32.c  |  91 +++-------------------------
 arch/x86/kernel/ptrace_64.c  |  26 ++++----
 arch/x86/kernel/tls.c        |  96 ++++++++++++-----------------
 include/asm-x86/ia32.h       |   6 --
 include/asm-x86/ptrace.h     |  11 ++++
 9 files changed, 77 insertions(+), 302 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 2499a324feaa..0db0a6291bbd 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -643,8 +643,8 @@ ia32_sys_call_table:
 	.quad compat_sys_futex		/* 240 */
 	.quad compat_sys_sched_setaffinity
 	.quad compat_sys_sched_getaffinity
-	.quad sys32_set_thread_area
-	.quad sys32_get_thread_area
+	.quad sys_set_thread_area
+	.quad sys_get_thread_area
 	.quad compat_sys_io_setup	/* 245 */
 	.quad sys_io_destroy
 	.quad compat_sys_io_getevents
diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index 2c9596b9349c..9a6577a746ba 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -10,6 +10,7 @@ obj-y	:= process_32.o signal_32.o entry_32.o traps_32.o irq_32.o \
 		pci-dma_32.o i386_ksyms_32.o i387_32.o bootflag.o e820_32.o\
 		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o rtc.o
 
+obj-y				+= tls.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
 obj-y				+= acpi/
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 631af167bc51..4d66a56280d3 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -501,32 +501,15 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
 		set_tsk_thread_flag(p, TIF_IO_BITMAP);
 	}
 
+	err = 0;
+
 	/*
 	 * Set a new TLS for the child thread?
 	 */
-	if (clone_flags & CLONE_SETTLS) {
-		struct desc_struct *desc;
-		struct user_desc info;
-		int idx;
-
-		err = -EFAULT;
-		if (copy_from_user(&info, (void __user *)childregs->esi, sizeof(info)))
-			goto out;
-		err = -EINVAL;
-		if (LDT_empty(&info))
-			goto out;
-
-		idx = info.entry_number;
-		if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-			goto out;
-
-		desc = p->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
-		desc->a = LDT_entry_a(&info);
-		desc->b = LDT_entry_b(&info);
-	}
+	if (clone_flags & CLONE_SETTLS)
+		err = do_set_thread_area(p, -1,
+			(struct user_desc __user *)childregs->esi, 0);
 
-	err = 0;
- out:
 	if (err && p->thread.io_bitmap_ptr) {
 		kfree(p->thread.io_bitmap_ptr);
 		p->thread.io_bitmap_max = 0;
@@ -872,120 +855,6 @@ unsigned long get_wchan(struct task_struct *p)
 	return 0;
 }
 
-/*
- * sys_alloc_thread_area: get a yet unused TLS descriptor index.
- */
-static int get_free_idx(void)
-{
-	struct thread_struct *t = &current->thread;
-	int idx;
-
-	for (idx = 0; idx < GDT_ENTRY_TLS_ENTRIES; idx++)
-		if (desc_empty(t->tls_array + idx))
-			return idx + GDT_ENTRY_TLS_MIN;
-	return -ESRCH;
-}
-
-/*
- * Set a given TLS descriptor:
- */
-asmlinkage int sys_set_thread_area(struct user_desc __user *u_info)
-{
-	struct thread_struct *t = &current->thread;
-	struct user_desc info;
-	struct desc_struct *desc;
-	int cpu, idx;
-
-	if (copy_from_user(&info, u_info, sizeof(info)))
-		return -EFAULT;
-	idx = info.entry_number;
-
-	/*
-	 * index -1 means the kernel should try to find and
-	 * allocate an empty descriptor:
-	 */
-	if (idx == -1) {
-		idx = get_free_idx();
-		if (idx < 0)
-			return idx;
-		if (put_user(idx, &u_info->entry_number))
-			return -EFAULT;
-	}
-
-	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-		return -EINVAL;
-
-	desc = t->tls_array + idx - GDT_ENTRY_TLS_MIN;
-
-	/*
-	 * We must not get preempted while modifying the TLS.
-	 */
-	cpu = get_cpu();
-
-	if (LDT_empty(&info)) {
-		desc->a = 0;
-		desc->b = 0;
-	} else {
-		desc->a = LDT_entry_a(&info);
-		desc->b = LDT_entry_b(&info);
-	}
-	load_TLS(t, cpu);
-
-	put_cpu();
-
-	return 0;
-}
-
-/*
- * Get the current Thread-Local Storage area:
- */
-
-#define GET_BASE(desc) ( \
-	(((desc)->a >> 16) & 0x0000ffff) | \
-	(((desc)->b << 16) & 0x00ff0000) | \
-	( (desc)->b        & 0xff000000)   )
-
-#define GET_LIMIT(desc) ( \
-	((desc)->a & 0x0ffff) | \
-	 ((desc)->b & 0xf0000) )
-	
-#define GET_32BIT(desc)		(((desc)->b >> 22) & 1)
-#define GET_CONTENTS(desc)	(((desc)->b >> 10) & 3)
-#define GET_WRITABLE(desc)	(((desc)->b >>  9) & 1)
-#define GET_LIMIT_PAGES(desc)	(((desc)->b >> 23) & 1)
-#define GET_PRESENT(desc)	(((desc)->b >> 15) & 1)
-#define GET_USEABLE(desc)	(((desc)->b >> 20) & 1)
-
-asmlinkage int sys_get_thread_area(struct user_desc __user *u_info)
-{
-	struct user_desc info;
-	struct desc_struct *desc;
-	int idx;
-
-	if (get_user(idx, &u_info->entry_number))
-		return -EFAULT;
-	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-		return -EINVAL;
-
-	memset(&info, 0, sizeof(info));
-
-	desc = current->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
-
-	info.entry_number = idx;
-	info.base_addr = GET_BASE(desc);
-	info.limit = GET_LIMIT(desc);
-	info.seg_32bit = GET_32BIT(desc);
-	info.contents = GET_CONTENTS(desc);
-	info.read_exec_only = !GET_WRITABLE(desc);
-	info.limit_in_pages = GET_LIMIT_PAGES(desc);
-	info.seg_not_present = !GET_PRESENT(desc);
-	info.useable = GET_USEABLE(desc);
-
-	if (copy_to_user(u_info, &info, sizeof(info)))
-		return -EFAULT;
-	return 0;
-}
-
 unsigned long arch_align_stack(unsigned long sp)
 {
 	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 9ea1d7546f80..ccc9d68d5a58 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -524,7 +524,8 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 	if (clone_flags & CLONE_SETTLS) {
 #ifdef CONFIG_IA32_EMULATION
 		if (test_thread_flag(TIF_IA32))
-			err = ia32_child_tls(p, childregs); 
+			err = do_set_thread_area(p, -1,
+				(struct user_desc __user *)childregs->rsi, 0);
 		else 			
 #endif	 
 			err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8); 
diff --git a/arch/x86/kernel/ptrace_32.c b/arch/x86/kernel/ptrace_32.c
index ff5431cc03ee..09227cfb7c4c 100644
--- a/arch/x86/kernel/ptrace_32.c
+++ b/arch/x86/kernel/ptrace_32.c
@@ -276,85 +276,6 @@ void ptrace_disable(struct task_struct *child)
 	clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 }
 
-/*
- * Perform get_thread_area on behalf of the traced child.
- */
-static int
-ptrace_get_thread_area(struct task_struct *child,
-		       int idx, struct user_desc __user *user_desc)
-{
-	struct user_desc info;
-	struct desc_struct *desc;
-
-/*
- * Get the current Thread-Local Storage area:
- */
-
-#define GET_BASE(desc) ( \
-	(((desc)->a >> 16) & 0x0000ffff) | \
-	(((desc)->b << 16) & 0x00ff0000) | \
-	( (desc)->b        & 0xff000000)   )
-
-#define GET_LIMIT(desc) ( \
-	((desc)->a & 0x0ffff) | \
-	 ((desc)->b & 0xf0000) )
-
-#define GET_32BIT(desc)		(((desc)->b >> 22) & 1)
-#define GET_CONTENTS(desc)	(((desc)->b >> 10) & 3)
-#define GET_WRITABLE(desc)	(((desc)->b >>  9) & 1)
-#define GET_LIMIT_PAGES(desc)	(((desc)->b >> 23) & 1)
-#define GET_PRESENT(desc)	(((desc)->b >> 15) & 1)
-#define GET_USEABLE(desc)	(((desc)->b >> 20) & 1)
-
-	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-		return -EINVAL;
-
-	desc = child->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
-
-	info.entry_number = idx;
-	info.base_addr = GET_BASE(desc);
-	info.limit = GET_LIMIT(desc);
-	info.seg_32bit = GET_32BIT(desc);
-	info.contents = GET_CONTENTS(desc);
-	info.read_exec_only = !GET_WRITABLE(desc);
-	info.limit_in_pages = GET_LIMIT_PAGES(desc);
-	info.seg_not_present = !GET_PRESENT(desc);
-	info.useable = GET_USEABLE(desc);
-
-	if (copy_to_user(user_desc, &info, sizeof(info)))
-		return -EFAULT;
-
-	return 0;
-}
-
-/*
- * Perform set_thread_area on behalf of the traced child.
- */
-static int
-ptrace_set_thread_area(struct task_struct *child,
-		       int idx, struct user_desc __user *user_desc)
-{
-	struct user_desc info;
-	struct desc_struct *desc;
-
-	if (copy_from_user(&info, user_desc, sizeof(info)))
-		return -EFAULT;
-
-	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-		return -EINVAL;
-
-	desc = child->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
-	if (LDT_empty(&info)) {
-		desc->a = 0;
-		desc->b = 0;
-	} else {
-		desc->a = LDT_entry_a(&info);
-		desc->b = LDT_entry_b(&info);
-	}
-
-	return 0;
-}
-
 long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 {
 	struct user * dummy = NULL;
@@ -601,13 +522,17 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	}
 
 	case PTRACE_GET_THREAD_AREA:
-		ret = ptrace_get_thread_area(child, addr,
-					(struct user_desc __user *) data);
+		if (addr < 0)
+			return -EIO;
+		ret = do_get_thread_area(child, addr,
+					 (struct user_desc __user *) data);
 		break;
 
 	case PTRACE_SET_THREAD_AREA:
-		ret = ptrace_set_thread_area(child, addr,
-					(struct user_desc __user *) data);
+		if (addr < 0)
+			return -EIO;
+		ret = do_set_thread_area(child, addr,
+					 (struct user_desc __user *) data, 0);
 		break;
 
 	default:
diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index 1edece36044c..375fadc23a25 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -474,23 +474,19 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		   64bit debugger to fully examine them too. Better
 		   don't use it against 64bit processes, use
 		   PTRACE_ARCH_PRCTL instead. */
-	case PTRACE_SET_THREAD_AREA: {
-		struct user_desc __user *p;
-		int old; 
-		p = (struct user_desc __user *)data;
-		get_user(old,  &p->entry_number); 
-		put_user(addr, &p->entry_number);
-		ret = do_set_thread_area(&child->thread, p);
-		put_user(old,  &p->entry_number); 
-		break;
 	case PTRACE_GET_THREAD_AREA:
-		p = (struct user_desc __user *)data;
-		get_user(old,  &p->entry_number); 
-		put_user(addr, &p->entry_number);
-		ret = do_get_thread_area(&child->thread, p);
-		put_user(old,  &p->entry_number); 
+		if (addr < 0)
+			return -EIO;
+		ret = do_get_thread_area(child, addr,
+					 (struct user_desc __user *) data);
+
+		break;
+	case PTRACE_SET_THREAD_AREA:
+		if (addr < 0)
+			return -EIO;
+		ret = do_set_thread_area(child, addr,
+					 (struct user_desc __user *) data, 0);
 		break;
-	} 
 #endif
 		/* normal 64bit interface to access TLS data. 
 		   Works just like arch_prctl, except that the arguments
diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c
index 5291596f19b0..67a377621b12 100644
--- a/arch/x86/kernel/tls.c
+++ b/arch/x86/kernel/tls.c
@@ -19,31 +19,34 @@ static int get_free_idx(void)
 	int idx;
 
 	for (idx = 0; idx < GDT_ENTRY_TLS_ENTRIES; idx++)
-		if (desc_empty((struct n_desc_struct *)(t->tls_array) + idx))
+		if (desc_empty(&t->tls_array[idx]))
 			return idx + GDT_ENTRY_TLS_MIN;
 	return -ESRCH;
 }
 
 /*
  * Set a given TLS descriptor:
- * When you want addresses > 32bit use arch_prctl()
  */
-int do_set_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
+int do_set_thread_area(struct task_struct *p, int idx,
+		       struct user_desc __user *u_info,
+		       int can_allocate)
 {
+	struct thread_struct *t = &p->thread;
 	struct user_desc info;
-	struct n_desc_struct *desc;
-	int cpu, idx;
+	u32 *desc;
+	int cpu;
 
 	if (copy_from_user(&info, u_info, sizeof(info)))
 		return -EFAULT;
 
-	idx = info.entry_number;
+	if (idx == -1)
+		idx = info.entry_number;
 
 	/*
 	 * index -1 means the kernel should try to find and
 	 * allocate an empty descriptor:
 	 */
-	if (idx == -1) {
+	if (idx == -1 && can_allocate) {
 		idx = get_free_idx();
 		if (idx < 0)
 			return idx;
@@ -54,7 +57,7 @@ int do_set_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
 	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
 		return -EINVAL;
 
-	desc = ((struct n_desc_struct *)t->tls_array) + idx - GDT_ENTRY_TLS_MIN;
+	desc = (u32 *) &t->tls_array[idx - GDT_ENTRY_TLS_MIN];
 
 	/*
 	 * We must not get preempted while modifying the TLS.
@@ -62,11 +65,11 @@ int do_set_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
 	cpu = get_cpu();
 
 	if (LDT_empty(&info)) {
-		desc->a = 0;
-		desc->b = 0;
+		desc[0] = 0;
+		desc[1] = 0;
 	} else {
-		desc->a = LDT_entry_a(&info);
-		desc->b = LDT_entry_b(&info);
+		desc[0] = LDT_entry_a(&info);
+		desc[1] = LDT_entry_b(&info);
 	}
 	if (t == &current->thread)
 		load_TLS(t, cpu);
@@ -75,9 +78,9 @@ int do_set_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
 	return 0;
 }
 
-asmlinkage long sys32_set_thread_area(struct user_desc __user *u_info)
+asmlinkage int sys_set_thread_area(struct user_desc __user *u_info)
 {
-	return do_set_thread_area(&current->thread, u_info);
+	return do_set_thread_area(current, -1, u_info, 1);
 }
 
 
@@ -85,34 +88,32 @@ asmlinkage long sys32_set_thread_area(struct user_desc __user *u_info)
  * Get the current Thread-Local Storage area:
  */
 
-#define GET_LIMIT(desc) ( \
-	((desc)->a & 0x0ffff) | \
-	 ((desc)->b & 0xf0000) )
-
-#define GET_32BIT(desc)		(((desc)->b >> 22) & 1)
-#define GET_CONTENTS(desc)	(((desc)->b >> 10) & 3)
-#define GET_WRITABLE(desc)	(((desc)->b >>  9) & 1)
-#define GET_LIMIT_PAGES(desc)	(((desc)->b >> 23) & 1)
-#define GET_PRESENT(desc)	(((desc)->b >> 15) & 1)
-#define GET_USEABLE(desc)	(((desc)->b >> 20) & 1)
-#define GET_LONGMODE(desc)	(((desc)->b >> 21) & 1)
-
-int do_get_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
+#define GET_LIMIT(desc)		(((desc)[0] & 0x0ffff) | ((desc)[1] & 0xf0000))
+#define GET_32BIT(desc)		(((desc)[1] >> 22) & 1)
+#define GET_CONTENTS(desc)	(((desc)[1] >> 10) & 3)
+#define GET_WRITABLE(desc)	(((desc)[1] >>  9) & 1)
+#define GET_LIMIT_PAGES(desc)	(((desc)[1] >> 23) & 1)
+#define GET_PRESENT(desc)	(((desc)[1] >> 15) & 1)
+#define GET_USEABLE(desc)	(((desc)[1] >> 20) & 1)
+#define GET_LONGMODE(desc)	(((desc)[1] >> 21) & 1)
+
+int do_get_thread_area(struct task_struct *p, int idx,
+		       struct user_desc __user *u_info)
 {
+	struct thread_struct *t = &p->thread;
 	struct user_desc info;
-	struct n_desc_struct *desc;
-	int idx;
+	u32 *desc;
 
-	if (get_user(idx, &u_info->entry_number))
+	if (idx == -1 && get_user(idx, &u_info->entry_number))
 		return -EFAULT;
 	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
 		return -EINVAL;
 
-	desc = ((struct n_desc_struct *)t->tls_array) + idx - GDT_ENTRY_TLS_MIN;
+	desc = (u32 *) &t->tls_array[idx - GDT_ENTRY_TLS_MIN];
 
 	memset(&info, 0, sizeof(struct user_desc));
 	info.entry_number = idx;
-	info.base_addr = get_desc_base(desc);
+	info.base_addr = get_desc_base((void *)desc);
 	info.limit = GET_LIMIT(desc);
 	info.seg_32bit = GET_32BIT(desc);
 	info.contents = GET_CONTENTS(desc);
@@ -120,39 +121,16 @@ int do_get_thread_area(struct thread_struct *t, struct user_desc __user *u_info)
 	info.limit_in_pages = GET_LIMIT_PAGES(desc);
 	info.seg_not_present = !GET_PRESENT(desc);
 	info.useable = GET_USEABLE(desc);
+#ifdef CONFIG_X86_64
 	info.lm = GET_LONGMODE(desc);
+#endif
 
 	if (copy_to_user(u_info, &info, sizeof(info)))
 		return -EFAULT;
 	return 0;
 }
 
-asmlinkage long sys32_get_thread_area(struct user_desc __user *u_info)
+asmlinkage int sys_get_thread_area(struct user_desc __user *u_info)
 {
-	return do_get_thread_area(&current->thread, u_info);
-}
-
-
-int ia32_child_tls(struct task_struct *p, struct pt_regs *childregs)
-{
-	struct n_desc_struct *desc;
-	struct user_desc info;
-	struct user_desc __user *cp;
-	int idx;
-
-	cp = (void __user *)childregs->rsi;
-	if (copy_from_user(&info, cp, sizeof(info)))
-		return -EFAULT;
-	if (LDT_empty(&info))
-		return -EINVAL;
-
-	idx = info.entry_number;
-	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-		return -EINVAL;
-
-	desc = (struct n_desc_struct *)(p->thread.tls_array) + idx - GDT_ENTRY_TLS_MIN;
-	desc->a = LDT_entry_a(&info);
-	desc->b = LDT_entry_b(&info);
-
-	return 0;
+	return do_get_thread_area(current, -1, u_info);
 }
diff --git a/include/asm-x86/ia32.h b/include/asm-x86/ia32.h
index 0190b7c4e319..aa9733206e29 100644
--- a/include/asm-x86/ia32.h
+++ b/include/asm-x86/ia32.h
@@ -159,12 +159,6 @@ struct ustat32 {
 #define IA32_STACK_TOP IA32_PAGE_OFFSET
 
 #ifdef __KERNEL__
-struct user_desc;
-struct siginfo_t;
-int do_get_thread_area(struct thread_struct *t, struct user_desc __user *info);
-int do_set_thread_area(struct thread_struct *t, struct user_desc __user *info);
-int ia32_child_tls(struct task_struct *p, struct pt_regs *childregs);
-
 struct linux_binprm;
 extern int ia32_setup_arg_pages(struct linux_binprm *bprm,
 				unsigned long stack_top, int exec_stack);
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 51ddb2590870..105d1534eaf4 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -137,6 +137,17 @@ enum {
 };
 #endif /* __KERNEL__ */
 #endif /* !__i386__ */
+
+#ifdef __KERNEL__
+
+struct user_desc;
+extern int do_get_thread_area(struct task_struct *p, int idx,
+			      struct user_desc __user *info);
+extern int do_set_thread_area(struct task_struct *p, int idx,
+			      struct user_desc __user *info, int can_allocate);
+
+#endif /* __KERNEL__ */
+
 #endif /* !__ASSEMBLY__ */
 
 #endif
-- 
cgit v1.2.3


From b263295dbffd33b0fbff670720fa178c30e3392a Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 30 Jan 2008 13:30:47 +0100
Subject: x86: 64-bit, make sparsemem vmemmap the only memory model

Use sparsemem as the only memory model for UP, SMP and NUMA.  Measurements
indicate that DISCONTIGMEM has a higher overhead than sparsemem.  And
FLATMEMs benefits are minimal.  So I think its best to simply standardize
on sparsemem.

Results of page allocator tests (test can be had via git from slab git
tree branch tests)

Measurements in cycle counts. 1000 allocations were performed and then the
average cycle count was calculated.

Order	FlatMem	Discontig	SparseMem
0	  639	  665		  641
1	  567	  647		  593
2	  679	  774		  692
3	  763	  967		  781
4	  961	 1501		  962
5	 1356	 2344		 1392
6	 2224	 3982		 2336
7	 4869	 7225		 5074
8	12500	14048		12732
9	27926	28223		28165
10	58578	58714		58682

(Note that FlatMem is an SMP config and the rest NUMA configurations)

Memory use:

SMP Sparsemem
-------------

Kernel size:

   text    data     bss     dec     hex filename
3849268  397739 1264856 5511863  541ab7 vmlinux

             total       used       free     shared    buffers     cached
Mem:       8242252      41164    8201088          0        352      11512
-/+ buffers/cache:      29300    8212952
Swap:      9775512          0    9775512

SMP Flatmem
-----------

Kernel size:

   text    data     bss     dec     hex filename
3844612  397739 1264536 5506887  540747 vmlinux

So 4.5k growth in text size vs. FLATMEM.

             total       used       free     shared    buffers     cached
Mem:       8244052      40544    8203508          0        352      11484
-/+ buffers/cache:      28708    8215344

2k growth in overall memory use after boot.

NUMA discontig:

   text    data     bss     dec     hex filename
3888124  470659 1276504 5635287  55fcd7 vmlinux

             total       used       free     shared    buffers     cached
Mem:       8256256      56908    8199348          0        352      11496
-/+ buffers/cache:      45060    8211196
Swap:      9775512          0    9775512

NUMA sparse:

   text    data     bss     dec     hex filename
3896428  470659 1276824 5643911  561e87 vmlinux

8k text growth. Given that we fully inline virt_to_page and friends now
that is rather good.

             total       used       free     shared    buffers     cached
Mem:       8264720      57240    8207480          0        352      11516
-/+ buffers/cache:      45372    8219348
Swap:      9775512          0    9775512

The total available memory is increased by 8k.

This patch makes sparsemem the default and removes discontig and
flatmem support from x86.

[ akpm@linux-foundation.org: allnoconfig build fix ]

Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig                   | 22 ++++++---------
 arch/x86/configs/x86_64_defconfig  |  9 ------
 arch/x86/kernel/machine_kexec_64.c |  5 ----
 arch/x86/mm/init_64.c              | 28 -------------------
 arch/x86/mm/ioremap_64.c           | 17 ------------
 arch/x86/mm/numa_64.c              | 21 --------------
 arch/x86/mm/srat_64.c              | 57 --------------------------------------
 include/asm-x86/mmzone_64.h        |  6 ----
 include/asm-x86/page_64.h          |  3 --
 9 files changed, 9 insertions(+), 159 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 2f4d88babd36..da98368f66af 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -891,25 +891,29 @@ config HAVE_ARCH_ALLOC_REMAP
 
 config ARCH_FLATMEM_ENABLE
 	def_bool y
-	depends on (X86_32 && ARCH_SELECT_MEMORY_MODEL && X86_PC) || (X86_64 && !NUMA)
+	depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && X86_PC
 
 config ARCH_DISCONTIGMEM_ENABLE
 	def_bool y
-	depends on NUMA
+	depends on NUMA && X86_32
 
 config ARCH_DISCONTIGMEM_DEFAULT
 	def_bool y
-	depends on NUMA
+	depends on NUMA && X86_32
+
+config ARCH_SPARSEMEM_DEFAULT
+	def_bool y
+	depends on X86_64
 
 config ARCH_SPARSEMEM_ENABLE
 	def_bool y
-	depends on NUMA || (EXPERIMENTAL && (X86_PC || X86_64))
+	depends on X86_64 || NUMA || (EXPERIMENTAL && X86_PC)
 	select SPARSEMEM_STATIC if X86_32
 	select SPARSEMEM_VMEMMAP_ENABLE if X86_64
 
 config ARCH_SELECT_MEMORY_MODEL
 	def_bool y
-	depends on X86_32 && ARCH_SPARSEMEM_ENABLE
+	depends on ARCH_SPARSEMEM_ENABLE
 
 config ARCH_MEMORY_PROBE
 	def_bool X86_64
@@ -1207,18 +1211,10 @@ config ARCH_ENABLE_MEMORY_HOTPLUG
 	def_bool y
 	depends on X86_64 || (X86_32 && HIGHMEM)
 
-config MEMORY_HOTPLUG_RESERVE
-	def_bool X86_64
-	depends on (MEMORY_HOTPLUG && DISCONTIGMEM)
-
 config HAVE_ARCH_EARLY_PFN_TO_NID
 	def_bool X86_64
 	depends on NUMA
 
-config OUT_OF_LINE_PFN_TO_PAGE
-	def_bool X86_64
-	depends on DISCONTIGMEM
-
 menu "Power management options"
 	depends on !X86_VOYAGER
 
diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig
index 38a83f9c966f..9e2b0ef851de 100644
--- a/arch/x86/configs/x86_64_defconfig
+++ b/arch/x86/configs/x86_64_defconfig
@@ -145,15 +145,6 @@ CONFIG_K8_NUMA=y
 CONFIG_NODES_SHIFT=6
 CONFIG_X86_64_ACPI_NUMA=y
 CONFIG_NUMA_EMU=y
-CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
-CONFIG_ARCH_DISCONTIGMEM_DEFAULT=y
-CONFIG_ARCH_SPARSEMEM_ENABLE=y
-CONFIG_SELECT_MEMORY_MODEL=y
-# CONFIG_FLATMEM_MANUAL is not set
-CONFIG_DISCONTIGMEM_MANUAL=y
-# CONFIG_SPARSEMEM_MANUAL is not set
-CONFIG_DISCONTIGMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_NEED_MULTIPLE_NODES=y
 # CONFIG_SPARSEMEM_STATIC is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index aa3d2c8f7737..a1fef42f8cdb 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -234,10 +234,5 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 void arch_crash_save_vmcoreinfo(void)
 {
 	VMCOREINFO_SYMBOL(init_level4_pgt);
-
-#ifdef CONFIG_ARCH_DISCONTIGMEM_ENABLE
-	VMCOREINFO_SYMBOL(node_data);
-	VMCOREINFO_LENGTH(node_data, MAX_NUMNODES);
-#endif
 }
 
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 0fbb657a8b19..251eeb325ae3 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -486,34 +486,6 @@ EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
 
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
-#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
-/*
- * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
- * just online the pages.
- */
-int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
-{
-	int err = -EIO;
-	unsigned long pfn;
-	unsigned long total = 0, mem = 0;
-	for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
-		if (pfn_valid(pfn)) {
-			online_page(pfn_to_page(pfn));
-			err = 0;
-			mem++;
-		}
-		total++;
-	}
-	if (!err) {
-		z->spanned_pages += total;
-		z->present_pages += mem;
-		z->zone_pgdat->node_spanned_pages += total;
-		z->zone_pgdat->node_present_pages += mem;
-	}
-	return err;
-}
-#endif
-
 static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
 			 kcore_vsyscall;
 
diff --git a/arch/x86/mm/ioremap_64.c b/arch/x86/mm/ioremap_64.c
index 6cac90aa5032..b03db4ca9cad 100644
--- a/arch/x86/mm/ioremap_64.c
+++ b/arch/x86/mm/ioremap_64.c
@@ -86,23 +86,6 @@ void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned l
 	if (phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
 		return (__force void __iomem *)phys_to_virt(phys_addr);
 
-#ifdef CONFIG_FLATMEM
-	/*
-	 * Don't allow anybody to remap normal RAM that we're using..
-	 */
-	if (last_addr < virt_to_phys(high_memory)) {
-		char *t_addr, *t_end;
- 		struct page *page;
-
-		t_addr = __va(phys_addr);
-		t_end = t_addr + (size - 1);
-	   
-		for(page = virt_to_page(t_addr); page <= virt_to_page(t_end); page++)
-			if(!PageReserved(page))
-				return NULL;
-	}
-#endif
-
 	pgprot = __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_GLOBAL
 			  | _PAGE_DIRTY | _PAGE_ACCESSED | flags);
 	/*
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 29b69300aee2..46b4b5e1a02a 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -153,12 +153,10 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes)
 	return shift;
 }
 
-#ifdef CONFIG_SPARSEMEM
 int early_pfn_to_nid(unsigned long pfn)
 {
 	return phys_to_nid(pfn << PAGE_SHIFT);
 }
-#endif
 
 static void * __init early_node_mem(int nodeid, unsigned long start,
 				    unsigned long end, unsigned long size)
@@ -635,23 +633,4 @@ void __init init_cpu_to_node(void)
 	}
 }
 
-#ifdef CONFIG_DISCONTIGMEM
-/*
- * Functions to convert PFNs from/to per node page addresses.
- * These are out of line because they are quite big.
- * They could be all tuned by pre caching more state.
- * Should do that.
- */
 
-int pfn_valid(unsigned long pfn)
-{
-	unsigned nid;
-	if (pfn >= num_physpages)
-		return 0;
-	nid = pfn_to_nid(pfn);
-	if (nid == 0xff)
-		return 0;
-	return pfn >= node_start_pfn(nid) && (pfn) < node_end_pfn(nid);
-}
-EXPORT_SYMBOL(pfn_valid);
-#endif
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index ea85172fc0cc..9be14171144b 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -151,62 +151,6 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 	       pxm, pa->apic_id, node);
 }
 
-#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
-/*
- * Protect against too large hotadd areas that would fill up memory.
- */
-static int hotadd_enough_memory(struct bootnode *nd)
-{
-	static unsigned long allocated;
-	static unsigned long last_area_end;
-	unsigned long pages = (nd->end - nd->start) >> PAGE_SHIFT;
-	long mem = pages * sizeof(struct page);
-	unsigned long addr;
-	unsigned long allowed;
-	unsigned long oldpages = pages;
-
-	if (mem < 0)
-		return 0;
-	allowed = (end_pfn - absent_pages_in_range(0, end_pfn)) * PAGE_SIZE;
-	allowed = (allowed / 100) * hotadd_percent;
-	if (allocated + mem > allowed) {
-		unsigned long range;
-		/* Give them at least part of their hotadd memory upto hotadd_percent
-		   It would be better to spread the limit out
-		   over multiple hotplug areas, but that is too complicated
-		   right now */
-		if (allocated >= allowed)
-			return 0;
-		range = allowed - allocated;
-		pages = (range / PAGE_SIZE);
-		mem = pages * sizeof(struct page);
-		nd->end = nd->start + range;
-	}
-	/* Not completely fool proof, but a good sanity check */
-	addr = find_e820_area(last_area_end, end_pfn<<PAGE_SHIFT, mem);
-	if (addr == -1UL)
-		return 0;
-	if (pages != oldpages)
-		printk(KERN_NOTICE "SRAT: Hotadd area limited to %lu bytes\n",
-			pages << PAGE_SHIFT);
-	last_area_end = addr + mem;
-	allocated += mem;
-	return 1;
-}
-
-static int update_end_of_memory(unsigned long end)
-{
-	found_add_area = 1;
-	if ((end >> PAGE_SHIFT) > end_pfn)
-		end_pfn = end >> PAGE_SHIFT;
-	return 1;
-}
-
-static inline int save_add_info(void)
-{
-	return hotadd_percent > 0;
-}
-#else
 int update_end_of_memory(unsigned long end) {return -1;}
 static int hotadd_enough_memory(struct bootnode *nd) {return 1;}
 #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
@@ -214,7 +158,6 @@ static inline int save_add_info(void) {return 1;}
 #else
 static inline int save_add_info(void) {return 0;}
 #endif
-#endif
 /*
  * Update nodes_add and decide if to include add are in the zone.
  * Both SPARSE and RESERVE need nodes_add infomation.
diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h
index 1e0ed34a6adc..b0c25ae111d9 100644
--- a/include/asm-x86/mmzone_64.h
+++ b/include/asm-x86/mmzone_64.h
@@ -43,12 +43,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 
 extern int early_pfn_to_nid(unsigned long pfn);
 
-#ifdef CONFIG_DISCONTIGMEM
-#define pfn_to_nid(pfn) phys_to_nid((unsigned long)(pfn) << PAGE_SHIFT)
-
-extern int pfn_valid(unsigned long pfn);
-#endif
-
 #ifdef CONFIG_NUMA_EMU
 #define FAKE_NODE_MIN_SIZE	(64*1024*1024)
 #define FAKE_NODE_MIN_HASH_MASK	(~(FAKE_NODE_MIN_SIZE - 1uL))
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 6fdc904a5fa5..d400167c5509 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -122,9 +122,6 @@ extern unsigned long __phys_addr(unsigned long);
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
 #define __boot_va(x)		__va(x)
 #define __boot_pa(x)		__pa(x)
-#ifdef CONFIG_FLATMEM
-#define pfn_valid(pfn)		((pfn) < end_pfn)
-#endif
 
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
 #define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
-- 
cgit v1.2.3


From 2c0b8a7578f7653e1e5312a5232e8ead563cf477 Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Wed, 30 Jan 2008 13:30:47 +0100
Subject: x86: fall back on interrupt disable in cmpxchg8b on 80386 and 80486

Actually, on 386, cmpxchg and cmpxchg_local fall back on
cmpxchg_386_u8/16/32: it disables interruptions around non atomic
updates to mimic the cmpxchg behavior.

The comment:
/* Poor man's cmpxchg for 386. Unsuitable for SMP */

already present in cmpxchg_386_u32 tells much about how this cmpxchg
implementation should not be used in a SMP context. However, the cmpxchg_local
can perfectly use this fallback, since it only needs to be atomic wrt the local
cpu.

This patch adds a cmpxchg_486_u64 and uses it as a fallback for cmpxchg64
and cmpxchg64_local on 80386 and 80486.

Q:
but why is it called cmpxchg_486 when the other functions are called

A:
Because the standard cmpxchg is missing only on 386, but cmpxchg8b is
missing both on 386 and 486.

Citing Intel's Instruction set reference:

cmpxchg:
This instruction is not supported on Intel processors earlier than the
Intel486 processors.

cmpxchg8b:
This instruction encoding is not supported on Intel processors earlier
than the Pentium processors.

Q:
What's the reason to have cmpxchg64_local on 32 bit architectures?
Without that need all this would just be a few simple defines.

A:
cmpxchg64_local on 32 bits architectures takes unsigned long long
parameters, but cmpxchg_local only takes longs. Since we have cmpxchg8b
to execute a 8 byte cmpxchg atomically on pentium and +, it makes sense
to provide a flavor of cmpxchg and cmpxchg_local using this instruction.

Also, for 32 bits architectures lacking the 64 bits atomic cmpxchg, it
makes sense _not_ to define cmpxchg64 while cmpxchg could still be
available.

Moreover, the fallback for cmpxchg8b on i386 for 386 and 486 is a

However, cmpxchg64_local will be emulated by disabling interrupts on all
architectures where it is not supported atomically.

Therefore, we *could* turn cmpxchg64_local into a cmpxchg_local, but it
would make the 386/486 fallbacks ugly, make its design different from
cmpxchg/cmpxchg64 (which really depends on atomic operations and cannot
be emulated) and require the __cmpxchg_local to be expressed as a macro
rather than an inline function so the parameters would not be fixed to
unsigned long long in every case.

So I think cmpxchg64_local makes sense there, but I am open to
suggestions.

Q:
Are there any callers?

A:
I am actually using it in LTTng in my timestamping code. I use it to
work around CPUs with asynchronous TSCs. I need to update 64 bits
values atomically on this 32 bits architecture.

Changelog:
- Ran though checkpatch.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/intel.c  |  17 ++++++
 include/asm-x86/cmpxchg_32.h | 122 ++++++++++++++++++++++++++++---------------
 2 files changed, 96 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index cc8c501b9f39..867ff94579be 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -342,5 +342,22 @@ unsigned long cmpxchg_386_u32(volatile void *ptr, u32 old, u32 new)
 EXPORT_SYMBOL(cmpxchg_386_u32);
 #endif
 
+#ifndef CONFIG_X86_CMPXCHG64
+unsigned long long cmpxchg_486_u64(volatile void *ptr, u64 old, u64 new)
+{
+	u64 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg8b for 386 and 486. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u64 *)ptr;
+	if (prev == old)
+		*(u64 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_486_u64);
+#endif
+
 // arch_initcall(intel_cpu_init);
 
diff --git a/include/asm-x86/cmpxchg_32.h b/include/asm-x86/cmpxchg_32.h
index f86ede28f6dc..cea1dae288a7 100644
--- a/include/asm-x86/cmpxchg_32.h
+++ b/include/asm-x86/cmpxchg_32.h
@@ -105,15 +105,24 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 
 #ifdef CONFIG_X86_CMPXCHG
 #define __HAVE_ARCH_CMPXCHG 1
-#define cmpxchg(ptr,o,n)\
-	((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
-					(unsigned long)(n),sizeof(*(ptr))))
-#define sync_cmpxchg(ptr,o,n)\
-	((__typeof__(*(ptr)))__sync_cmpxchg((ptr),(unsigned long)(o),\
-					(unsigned long)(n),sizeof(*(ptr))))
-#define cmpxchg_local(ptr,o,n)\
-	((__typeof__(*(ptr)))__cmpxchg_local((ptr),(unsigned long)(o),\
-					(unsigned long)(n),sizeof(*(ptr))))
+#define cmpxchg(ptr, o, n)						     \
+	((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(o),	     \
+					(unsigned long)(n), sizeof(*(ptr))))
+#define sync_cmpxchg(ptr, o, n)						     \
+	((__typeof__(*(ptr)))__sync_cmpxchg((ptr), (unsigned long)(o),	     \
+					(unsigned long)(n), sizeof(*(ptr))))
+#define cmpxchg_local(ptr, o, n)					     \
+	((__typeof__(*(ptr)))__cmpxchg_local((ptr), (unsigned long)(o),	     \
+					(unsigned long)(n), sizeof(*(ptr))))
+#endif
+
+#ifdef CONFIG_X86_CMPXCHG64
+#define cmpxchg64(ptr, o, n)						      \
+	((__typeof__(*(ptr)))__cmpxchg64((ptr), (unsigned long long)(o),      \
+					(unsigned long long)(n)))
+#define cmpxchg64_local(ptr, o, n)					      \
+	((__typeof__(*(ptr)))__cmpxchg64_local((ptr), (unsigned long long)(o),\
+					(unsigned long long)(n)))
 #endif
 
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
@@ -203,6 +212,34 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr,
 	return old;
 }
 
+static inline unsigned long long __cmpxchg64(volatile void *ptr,
+			unsigned long long old, unsigned long long new)
+{
+	unsigned long long prev;
+	__asm__ __volatile__(LOCK_PREFIX "cmpxchg8b %3"
+			     : "=A"(prev)
+			     : "b"((unsigned long)new),
+			       "c"((unsigned long)(new >> 32)),
+			       "m"(*__xg(ptr)),
+			       "0"(old)
+			     : "memory");
+	return prev;
+}
+
+static inline unsigned long long __cmpxchg64_local(volatile void *ptr,
+			unsigned long long old, unsigned long long new)
+{
+	unsigned long long prev;
+	__asm__ __volatile__("cmpxchg8b %3"
+			     : "=A"(prev)
+			     : "b"((unsigned long)new),
+			       "c"((unsigned long)(new >> 32)),
+			       "m"(*__xg(ptr)),
+			       "0"(old)
+			     : "memory");
+	return prev;
+}
+
 #ifndef CONFIG_X86_CMPXCHG
 /*
  * Building a kernel capable running on 80386. It may be necessary to
@@ -228,7 +265,7 @@ static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
 	return old;
 }
 
-#define cmpxchg(ptr,o,n)						\
+#define cmpxchg(ptr, o, n)						\
 ({									\
 	__typeof__(*(ptr)) __ret;					\
 	if (likely(boot_cpu_data.x86 > 3))				\
@@ -239,7 +276,7 @@ static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
 					(unsigned long)(n), sizeof(*(ptr))); \
 	__ret;								\
 })
-#define cmpxchg_local(ptr,o,n)						\
+#define cmpxchg_local(ptr, o, n)					\
 ({									\
 	__typeof__(*(ptr)) __ret;					\
 	if (likely(boot_cpu_data.x86 > 3))				\
@@ -252,38 +289,37 @@ static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
 })
 #endif
 
-static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long long old,
-				      unsigned long long new)
-{
-	unsigned long long prev;
-	__asm__ __volatile__(LOCK_PREFIX "cmpxchg8b %3"
-			     : "=A"(prev)
-			     : "b"((unsigned long)new),
-			       "c"((unsigned long)(new >> 32)),
-			       "m"(*__xg(ptr)),
-			       "0"(old)
-			     : "memory");
-	return prev;
-}
+#ifndef CONFIG_X86_CMPXCHG64
+/*
+ * Building a kernel capable running on 80386 and 80486. It may be necessary
+ * to simulate the cmpxchg8b on the 80386 and 80486 CPU.
+ */
 
-static inline unsigned long long __cmpxchg64_local(volatile void *ptr,
-			unsigned long long old, unsigned long long new)
-{
-	unsigned long long prev;
-	__asm__ __volatile__("cmpxchg8b %3"
-			     : "=A"(prev)
-			     : "b"((unsigned long)new),
-			       "c"((unsigned long)(new >> 32)),
-			       "m"(*__xg(ptr)),
-			       "0"(old)
-			     : "memory");
-	return prev;
-}
+extern unsigned long long cmpxchg_486_u64(volatile void *, u64, u64);
+
+#define cmpxchg64(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) __ret;					\
+	if (likely(boot_cpu_data.x86 > 4))				\
+		__ret = __cmpxchg64((ptr), (unsigned long long)(o),	\
+				(unsigned long long)(n));		\
+	else								\
+		__ret = cmpxchg_486_u64((ptr), (unsigned long long)(o),	\
+				(unsigned long long)(n));		\
+	__ret;								\
+})
+#define cmpxchg64_local(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) __ret;					\
+	if (likely(boot_cpu_data.x86 > 4))				\
+		__ret = __cmpxchg64_local((ptr), (unsigned long long)(o), \
+				(unsigned long long)(n));		\
+	else								\
+		__ret = cmpxchg_486_u64((ptr), (unsigned long long)(o),	\
+				(unsigned long long)(n));		\
+	__ret;								\
+})
+
+#endif
 
-#define cmpxchg64(ptr,o,n)\
-	((__typeof__(*(ptr)))__cmpxchg64((ptr),(unsigned long long)(o),\
-					(unsigned long long)(n)))
-#define cmpxchg64_local(ptr,o,n)\
-	((__typeof__(*(ptr)))__cmpxchg64_local((ptr),(unsigned long long)(o),\
-					(unsigned long long)(n)))
 #endif
-- 
cgit v1.2.3


From fb7fa8f1741c91f6c6e958762155abe9339476ca Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:47 +0100
Subject: ptrace: arch_has_single_step

This defines the new macro arch_has_single_step() in linux/ptrace.h, a
default for when asm/ptrace.h does not define it.  It declares the new
user_enable_single_step and user_disable_single_step functions.
This is not used yet, but paves the way to harmonize on this interface
for the arch-specific calls on all machines.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ptrace.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

(limited to 'include')

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index 3ea5750a0f7e..8f06c6fb22a5 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -129,6 +129,52 @@ int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data);
 #define force_successful_syscall_return() do { } while (0)
 #endif
 
+/*
+ * <asm/ptrace.h> should define the following things inside #ifdef __KERNEL__.
+ *
+ * These do-nothing inlines are used when the arch does not
+ * implement single-step.  The kerneldoc comments are here
+ * to document the interface for all arch definitions.
+ */
+
+#ifndef arch_has_single_step
+/**
+ * arch_has_single_step - does this CPU support user-mode single-step?
+ *
+ * If this is defined, then there must be function declarations or
+ * inlines for user_enable_single_step() and user_disable_single_step().
+ * arch_has_single_step() should evaluate to nonzero iff the machine
+ * supports instruction single-step for user mode.
+ * It can be a constant or it can test a CPU feature bit.
+ */
+#define arch_has_single_step()		(0)
+
+/**
+ * user_enable_single_step - single-step in user-mode task
+ * @task: either current or a task stopped in %TASK_TRACED
+ *
+ * This can only be called when arch_has_single_step() has returned nonzero.
+ * Set @task so that when it returns to user mode, it will trap after the
+ * next single instruction executes.
+ */
+static inline void user_enable_single_step(struct task_struct *task)
+{
+	BUG();			/* This can never be called.  */
+}
+
+/**
+ * user_disable_single_step - cancel user-mode single-step
+ * @task: either current or a task stopped in %TASK_TRACED
+ *
+ * Clear @task of the effects of user_enable_single_step().  This can
+ * be called whether or not user_enable_single_step() was ever called
+ * on @task, and even if arch_has_single_step() returned zero.
+ */
+static inline void user_disable_single_step(struct task_struct *task)
+{
+}
+#endif	/* arch_has_single_step */
+
 #endif
 
 #endif
-- 
cgit v1.2.3


From 2608a6584a0a32d3251dfafad31d9f8b2c784466 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:48 +0100
Subject: x86: segment selector macros

This copies into asm-x86/segment_64.h some macros from asm-x86/segment_32.h
for dissecting segment selectors.  This lets other code use these macros
uniformly on 32/64-bit rather than duplicating the constants elsewhere.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/segment_64.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/segment_64.h b/include/asm-x86/segment_64.h
index 04b8ab21328f..dce742101348 100644
--- a/include/asm-x86/segment_64.h
+++ b/include/asm-x86/segment_64.h
@@ -50,4 +50,15 @@
 #define GDT_SIZE (GDT_ENTRIES * 8)
 #define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8) 
 
+/* Bottom two bits of selector give the ring privilege level */
+#define SEGMENT_RPL_MASK	0x3
+/* Bit 2 is table indicator (LDT/GDT) */
+#define SEGMENT_TI_MASK		0x4
+
+/* User mode is privilege level 3 */
+#define USER_RPL		0x3
+/* LDT segment has TI set, GDT has it cleared */
+#define SEGMENT_LDT		0x4
+#define SEGMENT_GDT		0x0
+
 #endif
-- 
cgit v1.2.3


From 7f232343e0ea37ffc0a552cdbd4825482c949281 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:48 +0100
Subject: x86: arch_has_single_step

This defines the new standard arch_has_single_step macro.  It makes the
existing set_singlestep and clear_singlestep entry points global, and
renames them to the new standard names user_enable_single_step and
user_disable_single_step, respectively.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ptrace_32.c | 12 ++++++------
 arch/x86/kernel/ptrace_64.c | 12 ++++++------
 include/asm-x86/ptrace.h    |  7 +++++++
 3 files changed, 19 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace_32.c b/arch/x86/kernel/ptrace_32.c
index 4619bda470b6..1402a54ef61f 100644
--- a/arch/x86/kernel/ptrace_32.c
+++ b/arch/x86/kernel/ptrace_32.c
@@ -218,7 +218,7 @@ static inline int is_setting_trap_flag(struct task_struct *child, struct pt_regs
 	return 0;
 }
 
-static void set_singlestep(struct task_struct *child)
+void user_enable_single_step(struct task_struct *child)
 {
 	struct pt_regs *regs = get_child_regs(child);
 
@@ -249,7 +249,7 @@ static void set_singlestep(struct task_struct *child)
 	child->ptrace |= PT_DTRACE;
 }
 
-static void clear_singlestep(struct task_struct *child)
+void user_disable_single_step(struct task_struct *child)
 {
 	/* Always clear TIF_SINGLESTEP... */
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
@@ -269,7 +269,7 @@ static void clear_singlestep(struct task_struct *child)
  */
 void ptrace_disable(struct task_struct *child)
 { 
-	clear_singlestep(child);
+	user_disable_single_step(child);
 	clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 }
 
@@ -403,7 +403,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		}
 		child->exit_code = data;
 		/* make sure the single step bit is not set. */
-		clear_singlestep(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		ret = 0;
 		break;
@@ -419,7 +419,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 			break;
 		child->exit_code = SIGKILL;
 		/* make sure the single step bit is not set. */
-		clear_singlestep(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		break;
 
@@ -435,7 +435,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 			clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 
 		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-		set_singlestep(child);
+		user_enable_single_step(child);
 		child->exit_code = data;
 		/* give it a chance to run. */
 		wake_up_process(child);
diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index 8e433b3773d2..7373a99facf3 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -170,7 +170,7 @@ static int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
 	return 0;
 }
 
-static void set_singlestep(struct task_struct *child)
+void user_enable_single_step(struct task_struct *child)
 {
 	struct pt_regs *regs = task_pt_regs(child);
 
@@ -201,7 +201,7 @@ static void set_singlestep(struct task_struct *child)
 	child->ptrace |= PT_DTRACE;
 }
 
-static void clear_singlestep(struct task_struct *child)
+void user_disable_single_step(struct task_struct *child)
 {
 	/* Always clear TIF_SINGLESTEP... */
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
@@ -221,7 +221,7 @@ static void clear_singlestep(struct task_struct *child)
  */
 void ptrace_disable(struct task_struct *child)
 { 
-	clear_singlestep(child);
+	user_disable_single_step(child);
 }
 
 static int putreg(struct task_struct *child,
@@ -461,7 +461,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 		child->exit_code = data;
 		/* make sure the single step bit is not set. */
-		clear_singlestep(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		ret = 0;
 		break;
@@ -504,7 +504,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 		child->exit_code = SIGKILL;
 		/* make sure the single step bit is not set. */
-		clear_singlestep(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		break;
 
@@ -513,7 +513,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		if (!valid_signal(data))
 			break;
 		clear_tsk_thread_flag(child,TIF_SYSCALL_TRACE);
-		set_singlestep(child);
+		user_enable_single_step(child);
 		child->exit_code = data;
 		/* give it a chance to run. */
 		wake_up_process(child);
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 105d1534eaf4..fe75422f034b 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -140,6 +140,13 @@ enum {
 
 #ifdef __KERNEL__
 
+/*
+ * These are defined as per linux/ptrace.h, which see.
+ */
+#define arch_has_single_step()	(1)
+extern void user_enable_single_step(struct task_struct *);
+extern void user_disable_single_step(struct task_struct *);
+
 struct user_desc;
 extern int do_get_thread_area(struct task_struct *p, int idx,
 			      struct user_desc __user *info);
-- 
cgit v1.2.3


From e1f287735c1e58c653b516931b5d3dd899edcb77 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:50 +0100
Subject: x86 single_step: TIF_FORCED_TF

This changes the single-step support to use a new thread_info flag
TIF_FORCED_TF instead of the PT_DTRACE flag in task_struct.ptrace.
This keeps arch implementation uses out of this non-arch field.

This changes the ptrace access to eflags to mask TF and maintain
the TIF_FORCED_TF flag directly if userland sets TF, instead of
relying on ptrace_signal_deliver.  The 64-bit and 32-bit kernels
are harmonized on this same behavior.  The ptrace_signal_deliver
approach works now, but this change makes the low-level register
access code reliable when called from different contexts than a
ptrace stop, which will be possible in the future.

The 64-bit do_debug exception handler is also changed not to clear TF
from user-mode registers.  This matches the 32-bit kernel's behavior.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ptrace32.c         | 20 ++++++++++++++++++--
 arch/x86/kernel/process_32.c     |  3 ---
 arch/x86/kernel/process_64.c     |  5 -----
 arch/x86/kernel/ptrace_32.c      | 17 +++++++++++++++++
 arch/x86/kernel/ptrace_64.c      | 20 ++++++++++++++++++++
 arch/x86/kernel/signal_32.c      | 12 +++++-------
 arch/x86/kernel/signal_64.c      | 14 +++++---------
 arch/x86/kernel/step.c           |  9 +++------
 arch/x86/kernel/traps_64.c       | 23 +++++------------------
 include/asm-x86/signal.h         | 11 ++---------
 include/asm-x86/thread_info_32.h |  2 ++
 include/asm-x86/thread_info_64.h |  2 ++
 12 files changed, 79 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ptrace32.c b/arch/x86/ia32/ptrace32.c
index 9d754b640205..5dee33417313 100644
--- a/arch/x86/ia32/ptrace32.c
+++ b/arch/x86/ia32/ptrace32.c
@@ -89,6 +89,15 @@ static int putreg32(struct task_struct *child, unsigned regno, u32 val)
 		__u64 *flags = &stack[offsetof(struct pt_regs, eflags)/8];
 
 		val &= FLAG_MASK;
+		/*
+		 * If the user value contains TF, mark that
+		 * it was not "us" (the debugger) that set it.
+		 * If not, make sure it stays set if we had.
+		 */
+		if (val & X86_EFLAGS_TF)
+			clear_tsk_thread_flag(child, TIF_FORCED_TF);
+		else if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+			val |= X86_EFLAGS_TF;
 		*flags = val | (*flags & ~FLAG_MASK);
 		break;
 	}
@@ -179,9 +188,17 @@ static int getreg32(struct task_struct *child, unsigned regno, u32 *val)
 	R32(eax, rax);
 	R32(orig_eax, orig_rax);
 	R32(eip, rip);
-	R32(eflags, eflags);
 	R32(esp, rsp);
 
+	case offsetof(struct user32, regs.eflags):
+		/*
+		 * If the debugger set TF, hide it from the readout.
+		 */
+		*val = stack[offsetof(struct pt_regs, eflags)/8];
+		if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+			*val &= ~X86_EFLAGS_TF;
+		break;
+
 	case offsetof(struct user32, u_debugreg[0]):
 		*val = child->thread.debugreg0;
 		break;
@@ -425,4 +442,3 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 	put_task_struct(child);
 	return ret;
 }
-
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 4d66a56280d3..d9905c9d0fd5 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -817,9 +817,6 @@ asmlinkage int sys_execve(struct pt_regs regs)
 			(char __user * __user *) regs.edx,
 			&regs);
 	if (error == 0) {
-		task_lock(current);
-		current->ptrace &= ~PT_DTRACE;
-		task_unlock(current);
 		/* Make sure we don't return using sysenter.. */
 		set_thread_flag(TIF_IRET);
 	}
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ccc9d68d5a58..f7356e5517f6 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -709,11 +709,6 @@ long sys_execve(char __user *name, char __user * __user *argv,
 	if (IS_ERR(filename)) 
 		return error;
 	error = do_execve(filename, argv, envp, &regs); 
-	if (error == 0) {
-		task_lock(current);
-		current->ptrace &= ~PT_DTRACE;
-		task_unlock(current);
-	}
 	putname(filename);
 	return error;
 }
diff --git a/arch/x86/kernel/ptrace_32.c b/arch/x86/kernel/ptrace_32.c
index b73960885c3f..bc7fd802dcc7 100644
--- a/arch/x86/kernel/ptrace_32.c
+++ b/arch/x86/kernel/ptrace_32.c
@@ -104,6 +104,15 @@ static int putreg(struct task_struct *child,
 			break;
 		case EFL:
 			value &= FLAG_MASK;
+			/*
+			 * If the user value contains TF, mark that
+			 * it was not "us" (the debugger) that set it.
+			 * If not, make sure it stays set if we had.
+			 */
+			if (value & X86_EFLAGS_TF)
+				clear_tsk_thread_flag(child, TIF_FORCED_TF);
+			else if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+				value |= X86_EFLAGS_TF;
 			value |= get_stack_long(child, EFL_OFFSET) & ~FLAG_MASK;
 			break;
 	}
@@ -119,6 +128,14 @@ static unsigned long getreg(struct task_struct *child,
 	unsigned long retval = ~0UL;
 
 	switch (regno >> 2) {
+		case EFL:
+			/*
+			 * If the debugger set TF, hide it from the readout.
+			 */
+			retval = get_stack_long(child, EFL_OFFSET);
+			if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+				retval &= ~X86_EFLAGS_TF;
+			break;
 		case GS:
 			retval = child->thread.gs;
 			break;
diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index 4abfbced9b26..035d53e99c57 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -143,6 +143,15 @@ static int putreg(struct task_struct *child,
 			return 0;
 		case offsetof(struct user_regs_struct, eflags):
 			value &= FLAG_MASK;
+			/*
+			 * If the user value contains TF, mark that
+			 * it was not "us" (the debugger) that set it.
+			 * If not, make sure it stays set if we had.
+			 */
+			if (value & X86_EFLAGS_TF)
+				clear_tsk_thread_flag(child, TIF_FORCED_TF);
+			else if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+				value |= X86_EFLAGS_TF;
 			tmp = get_stack_long(child, EFL_OFFSET); 
 			tmp &= ~FLAG_MASK; 
 			value |= tmp;
@@ -189,6 +198,17 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
 			if (child->thread.gsindex != GS_TLS_SEL)
 				return 0;
 			return get_desc_base(&child->thread.tls_array[GS_TLS]);
+		case offsetof(struct user_regs_struct, eflags):
+			/*
+			 * If the debugger set TF, hide it from the readout.
+			 */
+			regno = regno - sizeof(struct pt_regs);
+			val = get_stack_long(child, regno);
+			if (test_tsk_thread_flag(child, TIF_IA32))
+				val &= 0xffffffff;
+			if (test_tsk_thread_flag(child, TIF_FORCED_TF))
+				val &= ~X86_EFLAGS_TF;
+			return val;
 		default:
 			regno = regno - sizeof(struct pt_regs);
 			val = get_stack_long(child, regno);
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 1ac53e9a0859..0a7c812212c9 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -545,14 +545,12 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	}
 
 	/*
-	 * If TF is set due to a debugger (PT_DTRACE), clear the TF flag so
-	 * that register information in the sigcontext is correct.
+	 * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF
+	 * flag so that register information in the sigcontext is correct.
 	 */
-	if (unlikely(regs->eflags & TF_MASK)
-	    && likely(current->ptrace & PT_DTRACE)) {
-		current->ptrace &= ~PT_DTRACE;
-		regs->eflags &= ~TF_MASK;
-	}
+	if (unlikely(regs->eflags & X86_EFLAGS_TF) &&
+	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
+		regs->eflags &= ~X86_EFLAGS_TF;
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 38d806467c0f..ab0178ebe00a 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -349,16 +349,12 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	}
 
 	/*
-	 * If TF is set due to a debugger (PT_DTRACE), clear the TF
-	 * flag so that register information in the sigcontext is
-	 * correct.
+	 * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF
+	 * flag so that register information in the sigcontext is correct.
 	 */
-	if (unlikely(regs->eflags & TF_MASK)) {
-		if (likely(current->ptrace & PT_DTRACE)) {
-			current->ptrace &= ~PT_DTRACE;
-			regs->eflags &= ~TF_MASK;
-		}
-	}
+	if (unlikely(regs->eflags & X86_EFLAGS_TF) &&
+	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
+		regs->eflags &= ~X86_EFLAGS_TF;
 
 #ifdef CONFIG_IA32_EMULATION
 	if (test_thread_flag(TIF_IA32)) {
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 6732272e3479..243bff650ca5 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -135,7 +135,7 @@ void user_enable_single_step(struct task_struct *child)
 	if (is_setting_trap_flag(child, regs))
 		return;
 
-	child->ptrace |= PT_DTRACE;
+	set_tsk_thread_flag(child, TIF_FORCED_TF);
 }
 
 void user_disable_single_step(struct task_struct *child)
@@ -144,9 +144,6 @@ void user_disable_single_step(struct task_struct *child)
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 
 	/* But touch TF only if it was set by us.. */
-	if (child->ptrace & PT_DTRACE) {
-		struct pt_regs *regs = task_pt_regs(child);
-		regs->eflags &= ~X86_EFLAGS_TF;
-		child->ptrace &= ~PT_DTRACE;
-	}
+	if (test_and_clear_tsk_thread_flag(child, TIF_FORCED_TF))
+		task_pt_regs(child)->eflags &= ~X86_EFLAGS_TF;
 }
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index aa248d754533..874aca397b02 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -865,27 +865,14 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs,
 
 	tsk->thread.debugreg6 = condition;
 
-	/* Mask out spurious TF errors due to lazy TF clearing */
+
+	/*
+	 * Single-stepping through TF: make sure we ignore any events in
+	 * kernel space (but re-enable TF when returning to user mode).
+	 */
 	if (condition & DR_STEP) {
-		/*
-		 * The TF error should be masked out only if the current
-		 * process is not traced and if the TRAP flag has been set
-		 * previously by a tracing process (condition detected by
-		 * the PT_DTRACE flag); remember that the i386 TRAP flag
-		 * can be modified by the process itself in user mode,
-		 * allowing programs to debug themselves without the ptrace()
-		 * interface.
-		 */
                 if (!user_mode(regs))
                        goto clear_TF_reenable;
-		/*
-		 * Was the TF flag set by a debugger? If so, clear it now,
-		 * so that register information is correct.
-		 */
-		if (tsk->ptrace & PT_DTRACE) {
-			regs->eflags &= ~TF_MASK;
-			tsk->ptrace &= ~PT_DTRACE;
-		}
 	}
 
 	/* Ok, finally something we can handle */
diff --git a/include/asm-x86/signal.h b/include/asm-x86/signal.h
index 987a422a2c78..aee7eca585ab 100644
--- a/include/asm-x86/signal.h
+++ b/include/asm-x86/signal.h
@@ -245,21 +245,14 @@ static __inline__ int sigfindinword(unsigned long word)
 
 struct pt_regs;
 
-#define ptrace_signal_deliver(regs, cookie)		\
-	do {						\
-		if (current->ptrace & PT_DTRACE) {	\
-			current->ptrace &= ~PT_DTRACE;	\
-			(regs)->eflags &= ~TF_MASK;	\
-		}					\
-	} while (0)
-
 #else /* __i386__ */
 
 #undef __HAVE_ARCH_SIG_BITOPS
 
+#endif /* !__i386__ */
+
 #define ptrace_signal_deliver(regs, cookie) do { } while (0)
 
-#endif /* !__i386__ */
 #endif /* __KERNEL__ */
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index a516e9192f11..009ecc6ad38b 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -138,6 +138,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_IO_BITMAP		18	/* uses I/O bitmap */
 #define TIF_FREEZE		19	/* is freezing for suspend */
 #define TIF_NOTSC		20	/* TSC is not accessible in userland */
+#define TIF_FORCED_TF		21	/* true if TF in eflags artificially */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -153,6 +154,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
 #define _TIF_NOTSC		(1<<TIF_NOTSC)
+#define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index c8e7736fc792..e0f41b3deced 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -121,6 +121,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_DEBUG		21	/* uses debug registers */
 #define TIF_IO_BITMAP		22	/* uses I/O bitmap */
 #define TIF_FREEZE		23	/* is freezing for suspend */
+#define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -138,6 +139,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_DEBUG		(1<<TIF_DEBUG)
 #define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
+#define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
-- 
cgit v1.2.3


From 2a84b0d71973692b291f03a870c4d0d13f722d79 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:51 +0100
Subject: powerpc: arch_has_single_step

This defines the new standard arch_has_single_step macro.  It makes the
existing set_single_step and clear_single_step entry points global, and
renames them to the new standard names user_enable_single_step and
user_disable_single_step, respectively.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/powerpc/kernel/ptrace.c | 12 ++++++------
 include/asm-powerpc/ptrace.h |  7 +++++++
 2 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c
index 3e17d154d0d4..b970d7971779 100644
--- a/arch/powerpc/kernel/ptrace.c
+++ b/arch/powerpc/kernel/ptrace.c
@@ -256,7 +256,7 @@ static int set_evrregs(struct task_struct *task, unsigned long *data)
 #endif /* CONFIG_SPE */
 
 
-static void set_single_step(struct task_struct *task)
+void user_enable_single_step(struct task_struct *task)
 {
 	struct pt_regs *regs = task->thread.regs;
 
@@ -271,7 +271,7 @@ static void set_single_step(struct task_struct *task)
 	set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 
-static void clear_single_step(struct task_struct *task)
+void user_disable_single_step(struct task_struct *task)
 {
 	struct pt_regs *regs = task->thread.regs;
 
@@ -313,7 +313,7 @@ static int ptrace_set_debugreg(struct task_struct *task, unsigned long addr,
 void ptrace_disable(struct task_struct *child)
 {
 	/* make sure the single step bit is not set. */
-	clear_single_step(child);
+	user_disable_single_step(child);
 }
 
 /*
@@ -456,7 +456,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 		child->exit_code = data;
 		/* make sure the single step bit is not set. */
-		clear_single_step(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		ret = 0;
 		break;
@@ -473,7 +473,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 			break;
 		child->exit_code = SIGKILL;
 		/* make sure the single step bit is not set. */
-		clear_single_step(child);
+		user_disable_single_step(child);
 		wake_up_process(child);
 		break;
 	}
@@ -483,7 +483,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		if (!valid_signal(data))
 			break;
 		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-		set_single_step(child);
+		user_enable_single_step(child);
 		child->exit_code = data;
 		/* give it a chance to run. */
 		wake_up_process(child);
diff --git a/include/asm-powerpc/ptrace.h b/include/asm-powerpc/ptrace.h
index 13fccc5a4119..3063363f6799 100644
--- a/include/asm-powerpc/ptrace.h
+++ b/include/asm-powerpc/ptrace.h
@@ -119,6 +119,13 @@ do {									      \
 } while (0)
 #endif /* __powerpc64__ */
 
+/*
+ * These are defined as per linux/ptrace.h, which see.
+ */
+#define arch_has_single_step()	(1)
+extern void user_enable_single_step(struct task_struct *);
+extern void user_disable_single_step(struct task_struct *);
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 962ff3804d31a4d090bbcbd3d06a4b63e3a5b5fd Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:52 +0100
Subject: x86: x86-64 ptrace debugreg cleanup

This cleans up the 64-bit ptrace code to separate the guts of the
debug register access from the implementation of PTRACE_PEEKUSR and
PTRACE_POKEUSR.  The new functions ptrace_[gs]et_debugreg are made
global so that the ia32 code can later be changed to call them too.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ptrace_64.c | 140 +++++++++++++++++++++-----------------------
 include/asm-x86/ptrace.h    |   3 +
 2 files changed, 69 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index d0a0aeaaa0c2..4ba66d8af717 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -183,9 +183,63 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
 
 }
 
+unsigned long ptrace_get_debugreg(struct task_struct *child, int n)
+{
+	switch (n) {
+	case 0:		return child->thread.debugreg0;
+	case 1:		return child->thread.debugreg1;
+	case 2:		return child->thread.debugreg2;
+	case 3:		return child->thread.debugreg3;
+	case 6:		return child->thread.debugreg6;
+	case 7:		return child->thread.debugreg7;
+	}
+	return 0;
+}
+
+int ptrace_set_debugreg(struct task_struct *child, int n, unsigned long data)
+{
+	int i;
+
+	if (n < 4) {
+		int dsize = test_tsk_thread_flag(child, TIF_IA32) ? 3 : 7;
+		if (unlikely(data >= TASK_SIZE_OF(child) - dsize))
+			return -EIO;
+	}
+
+	switch (n) {
+	case 0:		child->thread.debugreg0 = data; break;
+	case 1:		child->thread.debugreg1 = data; break;
+	case 2:		child->thread.debugreg2 = data; break;
+	case 3:		child->thread.debugreg3 = data; break;
+
+	case 6:
+		if (data >> 32)
+			return -EIO;
+		child->thread.debugreg6 = data;
+		break;
+
+	case 7:
+		/*
+		 * See ptrace_32.c for an explanation of this awkward check.
+		 */
+		data &= ~DR_CONTROL_RESERVED;
+		for (i = 0; i < 4; i++)
+			if ((0x5554 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
+				return -EIO;
+		child->thread.debugreg7 = data;
+		if (data)
+			set_tsk_thread_flag(child, TIF_DEBUG);
+		else
+			clear_tsk_thread_flag(child, TIF_DEBUG);
+		break;
+	}
+
+	return 0;
+}
+
 long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 {
-	long i, ret;
+	long ret;
 	unsigned ui;
 
 	switch (request) {
@@ -204,32 +258,14 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		    addr > sizeof(struct user) - 7)
 			break;
 
-		switch (addr) { 
-		case 0 ... sizeof(struct user_regs_struct) - sizeof(long):
+		tmp = 0;
+		if (addr < sizeof(struct user_regs_struct))
 			tmp = getreg(child, addr);
-			break;
-		case offsetof(struct user, u_debugreg[0]):
-			tmp = child->thread.debugreg0;
-			break;
-		case offsetof(struct user, u_debugreg[1]):
-			tmp = child->thread.debugreg1;
-			break;
-		case offsetof(struct user, u_debugreg[2]):
-			tmp = child->thread.debugreg2;
-			break;
-		case offsetof(struct user, u_debugreg[3]):
-			tmp = child->thread.debugreg3;
-			break;
-		case offsetof(struct user, u_debugreg[6]):
-			tmp = child->thread.debugreg6;
-			break;
-		case offsetof(struct user, u_debugreg[7]):
-			tmp = child->thread.debugreg7;
-			break;
-		default:
-			tmp = 0;
-			break;
+		else if (addr >= offsetof(struct user, u_debugreg[0])) {
+			addr -= offsetof(struct user, u_debugreg[0]);
+			tmp = ptrace_get_debugreg(child, addr / sizeof(long));
 		}
+
 		ret = put_user(tmp,(unsigned long __user *) data);
 		break;
 	}
@@ -241,63 +277,19 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 
 	case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
-	{
-		int dsize = test_tsk_thread_flag(child, TIF_IA32) ? 3 : 7;
 		ret = -EIO;
 		if ((addr & 7) ||
 		    addr > sizeof(struct user) - 7)
 			break;
 
-		switch (addr) { 
-		case 0 ... sizeof(struct user_regs_struct) - sizeof(long):
+		if (addr < sizeof(struct user_regs_struct))
 			ret = putreg(child, addr, data);
-			break;
-		/* Disallows to set a breakpoint into the vsyscall */
-		case offsetof(struct user, u_debugreg[0]):
-			if (data >= TASK_SIZE_OF(child) - dsize) break;
-			child->thread.debugreg0 = data;
-			ret = 0;
-			break;
-		case offsetof(struct user, u_debugreg[1]):
-			if (data >= TASK_SIZE_OF(child) - dsize) break;
-			child->thread.debugreg1 = data;
-			ret = 0;
-			break;
-		case offsetof(struct user, u_debugreg[2]):
-			if (data >= TASK_SIZE_OF(child) - dsize) break;
-			child->thread.debugreg2 = data;
-			ret = 0;
-			break;
-		case offsetof(struct user, u_debugreg[3]):
-			if (data >= TASK_SIZE_OF(child) - dsize) break;
-			child->thread.debugreg3 = data;
-			ret = 0;
-			break;
-		case offsetof(struct user, u_debugreg[6]):
-				  if (data >> 32)
-				break; 
-			child->thread.debugreg6 = data;
-			ret = 0;
-			break;
-		case offsetof(struct user, u_debugreg[7]):
-			/* See arch/i386/kernel/ptrace.c for an explanation of
-			 * this awkward check.*/
-			data &= ~DR_CONTROL_RESERVED;
-			for(i=0; i<4; i++)
-				if ((0x5554 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
-					break;
-			if (i == 4) {
-			  child->thread.debugreg7 = data;
-			  if (data)
-			  	set_tsk_thread_flag(child, TIF_DEBUG);
-			  else
-			  	clear_tsk_thread_flag(child, TIF_DEBUG);
-			  ret = 0;
-		  	}
-		  break;
+		else if (addr >= offsetof(struct user, u_debugreg[0])) {
+			addr -= offsetof(struct user, u_debugreg[0]);
+			ret = ptrace_set_debugreg(child,
+						  addr / sizeof(long), data);
 		}
 		break;
-	}
 
 #ifdef CONFIG_IA32_EMULATION
 		/* This makes only sense with 32bit programs. Allow a
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index fe75422f034b..d223decd7b01 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -110,6 +110,9 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
 
 struct task_struct;
 
+extern unsigned long ptrace_get_debugreg(struct task_struct *child, int n);
+extern int ptrace_set_debugreg(struct task_struct *child, int n, unsigned long);
+
 extern unsigned long
 convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
-- 
cgit v1.2.3


From dc802c2d2e66e2d1544e023bfd4be6cdee48d57b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:53 +0100
Subject: ptrace: arch_has_block_step

This defines the new macro arch_has_block_step() in linux/ptrace.h, a
default for when asm/ptrace.h does not define it.  This is the analog
of arch_has_single_step() for step-until-branch on machines that have
it.  It declares the new user_enable_block_step function, which goes
with the existing user_enable_single_step and user_disable_single_step.
This is not used yet, but paves the way to harmonize on this interface
for the arch-specific calls on all machines.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ptrace.h | 37 +++++++++++++++++++++++++++++++++----
 1 file changed, 33 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index 8f06c6fb22a5..1febc541dda5 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -155,7 +155,8 @@ int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data);
  *
  * This can only be called when arch_has_single_step() has returned nonzero.
  * Set @task so that when it returns to user mode, it will trap after the
- * next single instruction executes.
+ * next single instruction executes.  If arch_has_block_step() is defined,
+ * this must clear the effects of user_enable_block_step() too.
  */
 static inline void user_enable_single_step(struct task_struct *task)
 {
@@ -166,15 +167,43 @@ static inline void user_enable_single_step(struct task_struct *task)
  * user_disable_single_step - cancel user-mode single-step
  * @task: either current or a task stopped in %TASK_TRACED
  *
- * Clear @task of the effects of user_enable_single_step().  This can
- * be called whether or not user_enable_single_step() was ever called
- * on @task, and even if arch_has_single_step() returned zero.
+ * Clear @task of the effects of user_enable_single_step() and
+ * user_enable_block_step().  This can be called whether or not either
+ * of those was ever called on @task, and even if arch_has_single_step()
+ * returned zero.
  */
 static inline void user_disable_single_step(struct task_struct *task)
 {
 }
 #endif	/* arch_has_single_step */
 
+#ifndef arch_has_block_step
+/**
+ * arch_has_block_step - does this CPU support user-mode block-step?
+ *
+ * If this is defined, then there must be a function declaration or inline
+ * for user_enable_block_step(), and arch_has_single_step() must be defined
+ * too.  arch_has_block_step() should evaluate to nonzero iff the machine
+ * supports step-until-branch for user mode.  It can be a constant or it
+ * can test a CPU feature bit.
+ */
+#define arch_has_single_step()		(0)
+
+/**
+ * user_enable_block_step - step until branch in user-mode task
+ * @task: either current or a task stopped in %TASK_TRACED
+ *
+ * This can only be called when arch_has_block_step() has returned nonzero,
+ * and will never be called when single-instruction stepping is being used.
+ * Set @task so that when it returns to user mode, it will trap after the
+ * next branch or trap taken.
+ */
+static inline void user_enable_block_step(struct task_struct *task)
+{
+	BUG();			/* This can never be called.  */
+}
+#endif	/* arch_has_block_step */
+
 #endif
 
 #endif
-- 
cgit v1.2.3


From 5b88abbf770a0e1975c668743100f42934f385e8 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:53 +0100
Subject: ptrace: generic PTRACE_SINGLEBLOCK

This makes ptrace_request handle PTRACE_SINGLEBLOCK along with
PTRACE_CONT et al.  The new generic code makes use of the
arch_has_block_step macro and generic entry points on machines
that define them.

[ mingo@elte.hu: bugfix ]

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ptrace.h |  2 +-
 kernel/ptrace.c        | 15 ++++++++++++++-
 2 files changed, 15 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index 1febc541dda5..515bff053de8 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -187,7 +187,7 @@ static inline void user_disable_single_step(struct task_struct *task)
  * supports step-until-branch for user mode.  It can be a constant or it
  * can test a CPU feature bit.
  */
-#define arch_has_single_step()		(0)
+#define arch_has_block_step()		(0)
 
 /**
  * user_enable_block_step - step until branch in user-mode task
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index fad5f1e8511f..973d727f5e84 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -373,6 +373,12 @@ static int ptrace_setsiginfo(struct task_struct *child, siginfo_t __user * data)
 #define is_singlestep(request)		0
 #endif
 
+#ifdef PTRACE_SINGLEBLOCK
+#define is_singleblock(request)		((request) == PTRACE_SINGLEBLOCK)
+#else
+#define is_singleblock(request)		0
+#endif
+
 #ifdef PTRACE_SYSEMU
 #define is_sysemu_singlestep(request)	((request) == PTRACE_SYSEMU_SINGLESTEP)
 #else
@@ -396,7 +402,11 @@ static int ptrace_resume(struct task_struct *child, long request, long data)
 		clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 #endif
 
-	if (is_singlestep(request) || is_sysemu_singlestep(request)) {
+	if (is_singleblock(request)) {
+		if (unlikely(!arch_has_block_step()))
+			return -EIO;
+		user_enable_block_step(child);
+	} else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
 		if (unlikely(!arch_has_single_step()))
 			return -EIO;
 		user_enable_single_step(child);
@@ -438,6 +448,9 @@ int ptrace_request(struct task_struct *child, long request,
 #ifdef PTRACE_SINGLESTEP
 	case PTRACE_SINGLESTEP:
 #endif
+#ifdef PTRACE_SINGLEBLOCK
+	case PTRACE_SINGLEBLOCK:
+#endif
 #ifdef PTRACE_SYSEMU
 	case PTRACE_SYSEMU:
 	case PTRACE_SYSEMU_SINGLESTEP:
-- 
cgit v1.2.3


From d2499d8b6c94df02629631807dd96deae7aa7637 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:54 +0100
Subject: x86: debugctlmsr constants

This adds constant macros for a few of the bits in MSR_IA32_DEBUGCTLMSR.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr-index.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index a4944732be04..4045bbe47379 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -63,6 +63,13 @@
 #define MSR_IA32_LASTINTFROMIP		0x000001dd
 #define MSR_IA32_LASTINTTOIP		0x000001de
 
+/* DEBUGCTLMSR bits (others vary by model): */
+#define _DEBUGCTLMSR_LBR	0 /* last branch recording */
+#define _DEBUGCTLMSR_BTF	1 /* single-step on branches */
+
+#define DEBUGCTLMSR_LBR		(1UL << _DEBUGCTLMSR_LBR)
+#define DEBUGCTLMSR_BTF		(1UL << _DEBUGCTLMSR_BTF)
+
 #define MSR_IA32_MC0_CTL		0x00000400
 #define MSR_IA32_MC0_STATUS		0x00000401
 #define MSR_IA32_MC0_ADDR		0x00000402
-- 
cgit v1.2.3


From 7e9916040b3020d0f36d68bb7512e3b80b623097 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:54 +0100
Subject: x86: debugctlmsr context switch

This adds low-level support for a per-thread value of MSR_IA32_DEBUGCTLMSR.
The per-thread value is switched in when TIF_DEBUGCTLMSR is set.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c     | 6 +++++-
 arch/x86/kernel/process_64.c     | 3 +++
 include/asm-x86/processor_32.h   | 2 ++
 include/asm-x86/processor_64.h   | 2 ++
 include/asm-x86/thread_info_32.h | 6 ++++--
 include/asm-x86/thread_info_64.h | 4 +++-
 6 files changed, 19 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index d9905c9d0fd5..d5462f228daf 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -602,10 +602,14 @@ static noinline void
 __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		 struct tss_struct *tss)
 {
-	struct thread_struct *next;
+	struct thread_struct *prev, *next;
 
+	prev = &prev_p->thread;
 	next = &next_p->thread;
 
+	if (next->debugctlmsr != prev->debugctlmsr)
+		wrmsr(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr, 0);
+
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
 		set_debugreg(next->debugreg[0], 0);
 		set_debugreg(next->debugreg[1], 1);
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index f7356e5517f6..ae5eca17aa3c 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -555,6 +555,9 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 	prev = &prev_p->thread,
 	next = &next_p->thread;
 
+	if (next->debugctlmsr != prev->debugctlmsr)
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr);
+
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
 		loaddebug(next, 0);
 		loaddebug(next, 1);
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 2540bf8d5724..3c67eacb3168 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -368,6 +368,8 @@ struct thread_struct {
  	unsigned long	iopl;
 /* max allowed port in the bitmap, in bytes: */
 	unsigned long	io_bitmap_max;
+/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
+	unsigned long	debugctlmsr;
 };
 
 #define INIT_THREAD  {							\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 20d8935d141a..e7bea4fed642 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -238,6 +238,8 @@ struct thread_struct {
 	int		ioperm;
 	unsigned long	*io_bitmap_ptr;
 	unsigned io_bitmap_max;
+/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
+	unsigned long	debugctlmsr;
 /* cached TLS descriptors. */
 	u64 tls_array[GDT_ENTRY_TLS_ENTRIES];
 } __attribute__((aligned(16)));
diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index 009ecc6ad38b..306fc80800e1 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -139,6 +139,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_FREEZE		19	/* is freezing for suspend */
 #define TIF_NOTSC		20	/* TSC is not accessible in userland */
 #define TIF_FORCED_TF		21	/* true if TF in eflags artificially */
+#define TIF_DEBUGCTLMSR		22	/* uses thread_struct.debugctlmsr */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -155,6 +156,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
 #define _TIF_NOTSC		(1<<TIF_NOTSC)
 #define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
+#define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -164,8 +166,8 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_ALLWORK_MASK	(0x0000FFFF & ~_TIF_SECCOMP)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW_NEXT (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUG)
-#define _TIF_WORK_CTXSW_PREV (_TIF_IO_BITMAP | _TIF_NOTSC)
+#define _TIF_WORK_CTXSW_NEXT (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUG | _TIF_DEBUGCTLMSR)
+#define _TIF_WORK_CTXSW_PREV (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUGCTLMSR)
 
 /*
  * Thread-synchronous status.
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index e0f41b3deced..ee35fd12b541 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -122,6 +122,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_IO_BITMAP		22	/* uses I/O bitmap */
 #define TIF_FREEZE		23	/* is freezing for suspend */
 #define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
+#define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -140,6 +141,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
 #define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
+#define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -151,7 +153,7 @@ static inline struct thread_info *stack_thread_info(void)
 	(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY|_TIF_HRTICK_RESCHED)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP)
+#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR)
 
 #define PREEMPT_ACTIVE     0x10000000
 
-- 
cgit v1.2.3


From 10faa81e102e2b7695f386812055cd2ef9e44b4c Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:54 +0100
Subject: x86: debugctlmsr arch_has_block_step

This implements user-mode step-until-branch on x86 using the BTF bit
in MSR_IA32_DEBUGCTLMSR.  It's just like single-step, only less so.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/step.c     | 64 +++++++++++++++++++++++++++++++++++++++++++---
 arch/x86/kernel/traps_32.c |  6 +++++
 arch/x86/kernel/traps_64.c |  6 +++++
 include/asm-x86/ptrace.h   |  7 +++++
 4 files changed, 80 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 243bff650ca5..cf4b9dac4a05 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -107,7 +107,10 @@ static int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
 	return 0;
 }
 
-void user_enable_single_step(struct task_struct *child)
+/*
+ * Enable single-stepping.  Return nonzero if user mode is not using TF itself.
+ */
+static int enable_single_step(struct task_struct *child)
 {
 	struct pt_regs *regs = task_pt_regs(child);
 
@@ -122,7 +125,7 @@ void user_enable_single_step(struct task_struct *child)
 	 * If TF was already set, don't do anything else
 	 */
 	if (regs->eflags & X86_EFLAGS_TF)
-		return;
+		return 0;
 
 	/* Set TF on the kernel stack.. */
 	regs->eflags |= X86_EFLAGS_TF;
@@ -133,13 +136,68 @@ void user_enable_single_step(struct task_struct *child)
 	 * won't clear it by hand later.
 	 */
 	if (is_setting_trap_flag(child, regs))
-		return;
+		return 0;
 
 	set_tsk_thread_flag(child, TIF_FORCED_TF);
+
+	return 1;
+}
+
+/*
+ * Install this value in MSR_IA32_DEBUGCTLMSR whenever child is running.
+ */
+static void write_debugctlmsr(struct task_struct *child, unsigned long val)
+{
+	child->thread.debugctlmsr = val;
+
+	if (child != current)
+		return;
+
+#ifdef CONFIG_X86_64
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, val);
+#else
+	wrmsr(MSR_IA32_DEBUGCTLMSR, val, 0);
+#endif
+}
+
+/*
+ * Enable single or block step.
+ */
+static void enable_step(struct task_struct *child, bool block)
+{
+	/*
+	 * Make sure block stepping (BTF) is not enabled unless it should be.
+	 * Note that we don't try to worry about any is_setting_trap_flag()
+	 * instructions after the first when using block stepping.
+	 * So noone should try to use debugger block stepping in a program
+	 * that uses user-mode single stepping itself.
+	 */
+	if (enable_single_step(child) && block) {
+		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+		write_debugctlmsr(child, DEBUGCTLMSR_BTF);
+	} else if (test_and_clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR)) {
+		write_debugctlmsr(child, 0);
+	}
+}
+
+void user_enable_single_step(struct task_struct *child)
+{
+	enable_step(child, 0);
+}
+
+void user_enable_block_step(struct task_struct *child)
+{
+	enable_step(child, 1);
 }
 
 void user_disable_single_step(struct task_struct *child)
 {
+	/*
+	 * Make sure block stepping (BTF) is disabled.
+	 */
+	if (test_and_clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR))
+		write_debugctlmsr(child, 0);
+
 	/* Always clear TIF_SINGLESTEP... */
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 02d1e1e58e81..9b0bbd508cd5 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -837,6 +837,12 @@ fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 
 	get_debugreg(condition, 6);
 
+	/*
+	 * The processor cleared BTF, so don't mark that we need it set.
+	 */
+	clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR);
+	tsk->thread.debugctlmsr = 0;
+
 	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
 					SIGTRAP) == NOTIFY_STOP)
 		return;
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 874aca397b02..610a64d6bdf0 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -850,6 +850,12 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs,
 
 	get_debugreg(condition, 6);
 
+	/*
+	 * The processor cleared BTF, so don't mark that we need it set.
+	 */
+	clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR);
+	tsk->thread.debugctlmsr = 0;
+
 	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
 						SIGTRAP) == NOTIFY_STOP)
 		return;
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index d223decd7b01..04204f359298 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -150,6 +150,13 @@ enum {
 extern void user_enable_single_step(struct task_struct *);
 extern void user_disable_single_step(struct task_struct *);
 
+extern void user_enable_block_step(struct task_struct *);
+#ifdef CONFIG_X86_DEBUGCTLMSR
+#define arch_has_block_step()	(1)
+#else
+#define arch_has_block_step()	(boot_cpu_data.x86 >= 6)
+#endif
+
 struct user_desc;
 extern int do_get_thread_area(struct task_struct *p, int idx,
 			      struct user_desc __user *info);
-- 
cgit v1.2.3


From 0fa376e027f0bd71368e1485f863f0dff8f7c897 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:55 +0100
Subject: x86: PTRACE_SINGLEBLOCK

This adds the PTRACE_SINGLEBLOCK request on x86, matching the ia64 feature.
The implementation comes from the generic ptrace code and relies on the
low-level machine support provided by arch_has_block_step() et al.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ptrace32.c     | 1 +
 include/asm-x86/ptrace-abi.h | 2 ++
 2 files changed, 3 insertions(+)

(limited to 'include')

diff --git a/arch/x86/ia32/ptrace32.c b/arch/x86/ia32/ptrace32.c
index 5ababea1307a..8c6fb9d8352b 100644
--- a/arch/x86/ia32/ptrace32.c
+++ b/arch/x86/ia32/ptrace32.c
@@ -227,6 +227,7 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 	case PTRACE_KILL:
 	case PTRACE_CONT:
 	case PTRACE_SINGLESTEP:
+	case PTRACE_SINGLEBLOCK:
 	case PTRACE_DETACH:
 	case PTRACE_SYSCALL:
 	case PTRACE_OLDSETOPTIONS:
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index 7524e1233833..adce6b51df2e 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -78,4 +78,6 @@
 # define PTRACE_SYSEMU_SINGLESTEP 32
 #endif
 
+#define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
+
 #endif
-- 
cgit v1.2.3


From 1c54d77078056cde0f195b1a982cb681850efc08 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:30:55 +0100
Subject: x86: partial unification of asm-x86/bitops.h

This unifies the set/clear/test bit functions of asm/bitops.h.

I have not attempted to merge the bit-finding functions, since they
rely on the machine word size and can't be easily restructured to work
generically without a lot of #ifdefs.  In particular, the 64-bit code
can assume the presence of conditional move instructions, whereas
32-bit needs to be more careful.

The inline assembly for the bit operations has been changed to remove
explicit sizing hints on the instructions, so the assembler will pick
the appropriate instruction forms depending on the architecture and
the context.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/bitops.h    | 315 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/bitops_32.h | 308 -------------------------------------------
 include/asm-x86/bitops_64.h | 297 -----------------------------------------
 3 files changed, 315 insertions(+), 605 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h
index 07e3f6d4fe47..c6dd7e259b46 100644
--- a/include/asm-x86/bitops.h
+++ b/include/asm-x86/bitops.h
@@ -1,5 +1,320 @@
+#ifndef _ASM_X86_BITOPS_H
+#define _ASM_X86_BITOPS_H
+
+/*
+ * Copyright 1992, Linus Torvalds.
+ */
+
+#ifndef _LINUX_BITOPS_H
+#error only <linux/bitops.h> can be included directly
+#endif
+
+#include <linux/compiler.h>
+#include <asm/alternative.h>
+
+/*
+ * These have to be done with inline assembly: that way the bit-setting
+ * is guaranteed to be atomic. All bit operations return 0 if the bit
+ * was cleared before the operation and != 0 if it was not.
+ *
+ * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
+ */
+
+#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 1)
+/* Technically wrong, but this avoids compilation errors on some gcc
+   versions. */
+#define ADDR "=m" (*(volatile long *) addr)
+#else
+#define ADDR "+m" (*(volatile long *) addr)
+#endif
+
+/**
+ * set_bit - Atomically set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * This function is atomic and may not be reordered.  See __set_bit()
+ * if you do not require the atomic guarantees.
+ *
+ * Note: there are no guarantees that this function will not be reordered
+ * on non x86 architectures, so if you are writing portable code,
+ * make sure not to rely on its reordering guarantees.
+ *
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void set_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile(LOCK_PREFIX "bts %1,%0"
+		     : ADDR
+		     : "Ir" (nr) : "memory");
+}
+
+/**
+ * __set_bit - Set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * Unlike set_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __set_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile("bts %1,%0"
+		     : ADDR
+		     : "Ir" (nr) : "memory");
+}
+
+
+/**
+ * clear_bit - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * clear_bit() is atomic and may not be reordered.  However, it does
+ * not contain a memory barrier, so if it is used for locking purposes,
+ * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
+ * in order to ensure changes are visible on other processors.
+ */
+static inline void clear_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile(LOCK_PREFIX "btr %1,%0"
+		     : ADDR
+		     : "Ir" (nr));
+}
+
+/*
+ * clear_bit_unlock - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * clear_bit() is atomic and implies release semantics before the memory
+ * operation. It can be used for an unlock.
+ */
+static inline void clear_bit_unlock(unsigned nr, volatile unsigned long *addr)
+{
+	barrier();
+	clear_bit(nr, addr);
+}
+
+static inline void __clear_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile("btr %1,%0" : ADDR : "Ir" (nr));
+}
+
+/*
+ * __clear_bit_unlock - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * __clear_bit() is non-atomic and implies release semantics before the memory
+ * operation. It can be used for an unlock if no other CPUs can concurrently
+ * modify other bits in the word.
+ *
+ * No memory barrier is required here, because x86 cannot reorder stores past
+ * older loads. Same principle as spin_unlock.
+ */
+static inline void __clear_bit_unlock(unsigned nr, volatile unsigned long *addr)
+{
+	barrier();
+	__clear_bit(nr, addr);
+}
+
+#define smp_mb__before_clear_bit()	barrier()
+#define smp_mb__after_clear_bit()	barrier()
+
+/**
+ * __change_bit - Toggle a bit in memory
+ * @nr: the bit to change
+ * @addr: the address to start counting from
+ *
+ * Unlike change_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __change_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile("btc %1,%0" : ADDR : "Ir" (nr));
+}
+
+/**
+ * change_bit - Toggle a bit in memory
+ * @nr: Bit to change
+ * @addr: Address to start counting from
+ *
+ * change_bit() is atomic and may not be reordered.
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void change_bit(int nr, volatile unsigned long *addr)
+{
+	asm volatile(LOCK_PREFIX "btc %1,%0"
+		     : ADDR : "Ir" (nr));
+}
+
+/**
+ * test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_set_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile(LOCK_PREFIX "bts %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), ADDR
+		     : "Ir" (nr) : "memory");
+
+	return oldbit;
+}
+
+/**
+ * test_and_set_bit_lock - Set a bit and return its old value for lock
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This is the same as test_and_set_bit on x86.
+ */
+static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr)
+{
+	return test_and_set_bit(nr, addr);
+}
+
+/**
+ * __test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm("bts %2,%1\n\t"
+	    "sbb %0,%0"
+	    : "=r" (oldbit), ADDR
+	    : "Ir" (nr));
+	return oldbit;
+}
+
+/**
+ * test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_clear_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile(LOCK_PREFIX "btr %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), ADDR
+		     : "Ir" (nr) : "memory");
+
+	return oldbit;
+}
+
+/**
+ * __test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile("btr %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), ADDR
+		     : "Ir" (nr));
+	return oldbit;
+}
+
+/* WARNING: non atomic and it can be reordered! */
+static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile("btc %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), ADDR
+		     : "Ir" (nr) : "memory");
+
+	return oldbit;
+}
+
+/**
+ * test_and_change_bit - Change a bit and return its old value
+ * @nr: Bit to change
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_change_bit(int nr, volatile unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile(LOCK_PREFIX "btc %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), ADDR
+		     : "Ir" (nr) : "memory");
+
+	return oldbit;
+}
+
+static inline int constant_test_bit(int nr, const volatile unsigned long *addr)
+{
+	return ((1UL << (nr % BITS_PER_LONG)) & (addr[nr / BITS_PER_LONG])) != 0;
+}
+
+static inline int variable_test_bit(int nr, volatile const unsigned long *addr)
+{
+	int oldbit;
+
+	asm volatile("bt %2,%1\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit)
+		     : "m" (*addr), "Ir" (nr));
+
+	return oldbit;
+}
+
+#if 0 /* Fool kernel-doc since it doesn't do macros yet */
+/**
+ * test_bit - Determine whether a bit is set
+ * @nr: bit number to test
+ * @addr: Address to start counting from
+ */
+static int test_bit(int nr, const volatile unsigned long *addr);
+#endif
+
+#define test_bit(nr,addr)			\
+	(__builtin_constant_p(nr) ?		\
+	 constant_test_bit((nr),(addr)) :	\
+	 variable_test_bit((nr),(addr)))
+
+#undef ADDR
+
 #ifdef CONFIG_X86_32
 # include "bitops_32.h"
 #else
 # include "bitops_64.h"
 #endif
+
+#endif	/* _ASM_X86_BITOPS_H */
diff --git a/include/asm-x86/bitops_32.h b/include/asm-x86/bitops_32.h
index 5a29cce6a91d..e4d75fcf9c03 100644
--- a/include/asm-x86/bitops_32.h
+++ b/include/asm-x86/bitops_32.h
@@ -5,314 +5,6 @@
  * Copyright 1992, Linus Torvalds.
  */
 
-#ifndef _LINUX_BITOPS_H
-#error only <linux/bitops.h> can be included directly
-#endif
-
-#include <linux/compiler.h>
-#include <asm/alternative.h>
-
-/*
- * These have to be done with inline assembly: that way the bit-setting
- * is guaranteed to be atomic. All bit operations return 0 if the bit
- * was cleared before the operation and != 0 if it was not.
- *
- * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
- */
-
-#define ADDR (*(volatile long *) addr)
-
-/**
- * set_bit - Atomically set a bit in memory
- * @nr: the bit to set
- * @addr: the address to start counting from
- *
- * This function is atomic and may not be reordered.  See __set_bit()
- * if you do not require the atomic guarantees.
- *
- * Note: there are no guarantees that this function will not be reordered
- * on non x86 architectures, so if you are writing portable code,
- * make sure not to rely on its reordering guarantees.
- *
- * Note that @nr may be almost arbitrarily large; this function is not
- * restricted to acting on a single-word quantity.
- */
-static inline void set_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btsl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/**
- * __set_bit - Set a bit in memory
- * @nr: the bit to set
- * @addr: the address to start counting from
- *
- * Unlike set_bit(), this function is non-atomic and may be reordered.
- * If it's called on the same region of memory simultaneously, the effect
- * may be that only one operation succeeds.
- */
-static inline void __set_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__(
-		"btsl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/**
- * clear_bit - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * clear_bit() is atomic and may not be reordered.  However, it does
- * not contain a memory barrier, so if it is used for locking purposes,
- * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
- * in order to ensure changes are visible on other processors.
- */
-static inline void clear_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btrl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/*
- * clear_bit_unlock - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * clear_bit() is atomic and implies release semantics before the memory
- * operation. It can be used for an unlock.
- */
-static inline void clear_bit_unlock(unsigned long nr, volatile unsigned long *addr)
-{
-	barrier();
-	clear_bit(nr, addr);
-}
-
-static inline void __clear_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__ __volatile__(
-		"btrl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/*
- * __clear_bit_unlock - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * __clear_bit() is non-atomic and implies release semantics before the memory
- * operation. It can be used for an unlock if no other CPUs can concurrently
- * modify other bits in the word.
- *
- * No memory barrier is required here, because x86 cannot reorder stores past
- * older loads. Same principle as spin_unlock.
- */
-static inline void __clear_bit_unlock(unsigned long nr, volatile unsigned long *addr)
-{
-	barrier();
-	__clear_bit(nr, addr);
-}
-
-#define smp_mb__before_clear_bit()	barrier()
-#define smp_mb__after_clear_bit()	barrier()
-
-/**
- * __change_bit - Toggle a bit in memory
- * @nr: the bit to change
- * @addr: the address to start counting from
- *
- * Unlike change_bit(), this function is non-atomic and may be reordered.
- * If it's called on the same region of memory simultaneously, the effect
- * may be that only one operation succeeds.
- */
-static inline void __change_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__ __volatile__(
-		"btcl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/**
- * change_bit - Toggle a bit in memory
- * @nr: Bit to change
- * @addr: Address to start counting from
- *
- * change_bit() is atomic and may not be reordered. It may be
- * reordered on other architectures than x86.
- * Note that @nr may be almost arbitrarily large; this function is not
- * restricted to acting on a single-word quantity.
- */
-static inline void change_bit(int nr, volatile unsigned long *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btcl %1,%0"
-		:"+m" (ADDR)
-		:"Ir" (nr));
-}
-
-/**
- * test_and_set_bit - Set a bit and return its old value
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.  
- * It may be reordered on other architectures than x86.
- * It also implies a memory barrier.
- */
-static inline int test_and_set_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btsl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * test_and_set_bit_lock - Set a bit and return its old value for lock
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This is the same as test_and_set_bit on x86.
- */
-static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr)
-{
-	return test_and_set_bit(nr, addr);
-}
-
-/**
- * __test_and_set_bit - Set a bit and return its old value
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This operation is non-atomic and can be reordered.  
- * If two examples of this operation race, one can appear to succeed
- * but actually fail.  You must protect multiple accesses with a lock.
- */
-static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__(
-		"btsl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr));
-	return oldbit;
-}
-
-/**
- * test_and_clear_bit - Clear a bit and return its old value
- * @nr: Bit to clear
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.
- * It can be reorderdered on other architectures other than x86.
- * It also implies a memory barrier.
- */
-static inline int test_and_clear_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btrl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * __test_and_clear_bit - Clear a bit and return its old value
- * @nr: Bit to clear
- * @addr: Address to count from
- *
- * This operation is non-atomic and can be reordered.  
- * If two examples of this operation race, one can appear to succeed
- * but actually fail.  You must protect multiple accesses with a lock.
- */
-static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__(
-		"btrl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr));
-	return oldbit;
-}
-
-/* WARNING: non atomic and it can be reordered! */
-static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__(
-		"btcl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * test_and_change_bit - Change a bit and return its old value
- * @nr: Bit to change
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.  
- * It also implies a memory barrier.
- */
-static inline int test_and_change_bit(int nr, volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btcl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),"+m" (ADDR)
-		:"Ir" (nr) : "memory");
-	return oldbit;
-}
-
-#if 0 /* Fool kernel-doc since it doesn't do macros yet */
-/**
- * test_bit - Determine whether a bit is set
- * @nr: bit number to test
- * @addr: Address to start counting from
- */
-static int test_bit(int nr, const volatile void *addr);
-#endif
-
-static __always_inline int constant_test_bit(int nr, const volatile unsigned long *addr)
-{
-	return ((1UL << (nr & 31)) & (addr[nr >> 5])) != 0;
-}
-
-static inline int variable_test_bit(int nr, const volatile unsigned long *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__(
-		"btl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit)
-		:"m" (ADDR),"Ir" (nr));
-	return oldbit;
-}
-
-#define test_bit(nr, addr) \
-	(__builtin_constant_p(nr) ? \
-		constant_test_bit((nr), (addr)) : \
-		variable_test_bit((nr), (addr)))
-
-#undef ADDR
-
 /**
  * find_first_zero_bit - find the first zero bit in a memory region
  * @addr: The address to start the search at
diff --git a/include/asm-x86/bitops_64.h b/include/asm-x86/bitops_64.h
index 766bcc0470a6..48adbf56ca60 100644
--- a/include/asm-x86/bitops_64.h
+++ b/include/asm-x86/bitops_64.h
@@ -5,303 +5,6 @@
  * Copyright 1992, Linus Torvalds.
  */
 
-#ifndef _LINUX_BITOPS_H
-#error only <linux/bitops.h> can be included directly
-#endif
-
-#include <asm/alternative.h>
-
-#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 1)
-/* Technically wrong, but this avoids compilation errors on some gcc
-   versions. */
-#define ADDR "=m" (*(volatile long *) addr)
-#else
-#define ADDR "+m" (*(volatile long *) addr)
-#endif
-
-/**
- * set_bit - Atomically set a bit in memory
- * @nr: the bit to set
- * @addr: the address to start counting from
- *
- * This function is atomic and may not be reordered.  See __set_bit()
- * if you do not require the atomic guarantees.
- * Note that @nr may be almost arbitrarily large; this function is not
- * restricted to acting on a single-word quantity.
- */
-static inline void set_bit(int nr, volatile void *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btsl %1,%0"
-		:ADDR
-		:"dIr" (nr) : "memory");
-}
-
-/**
- * __set_bit - Set a bit in memory
- * @nr: the bit to set
- * @addr: the address to start counting from
- *
- * Unlike set_bit(), this function is non-atomic and may be reordered.
- * If it's called on the same region of memory simultaneously, the effect
- * may be that only one operation succeeds.
- */
-static inline void __set_bit(int nr, volatile void *addr)
-{
-	__asm__ volatile(
-		"btsl %1,%0"
-		:ADDR
-		:"dIr" (nr) : "memory");
-}
-
-/**
- * clear_bit - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * clear_bit() is atomic and may not be reordered.  However, it does
- * not contain a memory barrier, so if it is used for locking purposes,
- * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
- * in order to ensure changes are visible on other processors.
- */
-static inline void clear_bit(int nr, volatile void *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btrl %1,%0"
-		:ADDR
-		:"dIr" (nr));
-}
-
-/*
- * clear_bit_unlock - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * clear_bit() is atomic and implies release semantics before the memory
- * operation. It can be used for an unlock.
- */
-static inline void clear_bit_unlock(unsigned long nr, volatile unsigned long *addr)
-{
-	barrier();
-	clear_bit(nr, addr);
-}
-
-static inline void __clear_bit(int nr, volatile void *addr)
-{
-	__asm__ __volatile__(
-		"btrl %1,%0"
-		:ADDR
-		:"dIr" (nr));
-}
-
-/*
- * __clear_bit_unlock - Clears a bit in memory
- * @nr: Bit to clear
- * @addr: Address to start counting from
- *
- * __clear_bit() is non-atomic and implies release semantics before the memory
- * operation. It can be used for an unlock if no other CPUs can concurrently
- * modify other bits in the word.
- *
- * No memory barrier is required here, because x86 cannot reorder stores past
- * older loads. Same principle as spin_unlock.
- */
-static inline void __clear_bit_unlock(unsigned long nr, volatile unsigned long *addr)
-{
-	barrier();
-	__clear_bit(nr, addr);
-}
-
-#define smp_mb__before_clear_bit()	barrier()
-#define smp_mb__after_clear_bit()	barrier()
-
-/**
- * __change_bit - Toggle a bit in memory
- * @nr: the bit to change
- * @addr: the address to start counting from
- *
- * Unlike change_bit(), this function is non-atomic and may be reordered.
- * If it's called on the same region of memory simultaneously, the effect
- * may be that only one operation succeeds.
- */
-static inline void __change_bit(int nr, volatile void *addr)
-{
-	__asm__ __volatile__(
-		"btcl %1,%0"
-		:ADDR
-		:"dIr" (nr));
-}
-
-/**
- * change_bit - Toggle a bit in memory
- * @nr: Bit to change
- * @addr: Address to start counting from
- *
- * change_bit() is atomic and may not be reordered.
- * Note that @nr may be almost arbitrarily large; this function is not
- * restricted to acting on a single-word quantity.
- */
-static inline void change_bit(int nr, volatile void *addr)
-{
-	__asm__ __volatile__( LOCK_PREFIX
-		"btcl %1,%0"
-		:ADDR
-		:"dIr" (nr));
-}
-
-/**
- * test_and_set_bit - Set a bit and return its old value
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.  
- * It also implies a memory barrier.
- */
-static inline int test_and_set_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btsl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * test_and_set_bit_lock - Set a bit and return its old value for lock
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This is the same as test_and_set_bit on x86.
- */
-static inline int test_and_set_bit_lock(int nr, volatile void *addr)
-{
-	return test_and_set_bit(nr, addr);
-}
-
-/**
- * __test_and_set_bit - Set a bit and return its old value
- * @nr: Bit to set
- * @addr: Address to count from
- *
- * This operation is non-atomic and can be reordered.  
- * If two examples of this operation race, one can appear to succeed
- * but actually fail.  You must protect multiple accesses with a lock.
- */
-static inline int __test_and_set_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__(
-		"btsl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr));
-	return oldbit;
-}
-
-/**
- * test_and_clear_bit - Clear a bit and return its old value
- * @nr: Bit to clear
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.  
- * It also implies a memory barrier.
- */
-static inline int test_and_clear_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btrl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * __test_and_clear_bit - Clear a bit and return its old value
- * @nr: Bit to clear
- * @addr: Address to count from
- *
- * This operation is non-atomic and can be reordered.  
- * If two examples of this operation race, one can appear to succeed
- * but actually fail.  You must protect multiple accesses with a lock.
- */
-static inline int __test_and_clear_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__(
-		"btrl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr));
-	return oldbit;
-}
-
-/* WARNING: non atomic and it can be reordered! */
-static inline int __test_and_change_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__(
-		"btcl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr) : "memory");
-	return oldbit;
-}
-
-/**
- * test_and_change_bit - Change a bit and return its old value
- * @nr: Bit to change
- * @addr: Address to count from
- *
- * This operation is atomic and cannot be reordered.  
- * It also implies a memory barrier.
- */
-static inline int test_and_change_bit(int nr, volatile void *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__( LOCK_PREFIX
-		"btcl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit),ADDR
-		:"dIr" (nr) : "memory");
-	return oldbit;
-}
-
-#if 0 /* Fool kernel-doc since it doesn't do macros yet */
-/**
- * test_bit - Determine whether a bit is set
- * @nr: bit number to test
- * @addr: Address to start counting from
- */
-static int test_bit(int nr, const volatile void *addr);
-#endif
-
-static inline int constant_test_bit(int nr, const volatile void *addr)
-{
-	return ((1UL << (nr & 31)) & (((const volatile unsigned int *) addr)[nr >> 5])) != 0;
-}
-
-static inline int variable_test_bit(int nr, volatile const void *addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__(
-		"btl %2,%1\n\tsbbl %0,%0"
-		:"=r" (oldbit)
-		:"m" (*(volatile long *)addr),"dIr" (nr));
-	return oldbit;
-}
-
-#define test_bit(nr,addr) \
-(__builtin_constant_p(nr) ? \
- constant_test_bit((nr),(addr)) : \
- variable_test_bit((nr),(addr)))
-
-#undef ADDR
-
 extern long find_first_zero_bit(const unsigned long *addr, unsigned long size);
 extern long find_next_zero_bit(const unsigned long *addr, long size, long offset);
 extern long find_first_bit(const unsigned long *addr, unsigned long size);
-- 
cgit v1.2.3


From 5548fecdff5617ba3a2f09f0e585e1ac6e1bd25c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:30:55 +0100
Subject: x86: clean up bitops-related warnings

Add casts to appropriate places to silence spurious bitops warnings.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_64.c   | 31 ++++++++++++++++---------------
 arch/x86/kernel/smpboot_64.c |  4 ++--
 arch/x86/mm/numa_64.c        |  2 +-
 include/asm-x86/cpufeature.h |  2 +-
 include/asm-x86/numa_64.h    |  2 +-
 include/linux/thread_info.h  | 10 +++++-----
 6 files changed, 26 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 84f66b7b4d2e..63dd39b843b5 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -661,19 +661,19 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 
 	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
 	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
-	clear_bit(0*32+31, &c->x86_capability);
+	clear_bit(0*32+31, (unsigned long *)&c->x86_capability);
 
 	/* On C+ stepping K8 rep microcode works well for copy/memset */
 	level = cpuid_eax(1);
 	if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) ||
 			     level >= 0x0f58))
-		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
 	if (c->x86 == 0x10 || c->x86 == 0x11)
-		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
 
 	/* Enable workaround for FXSAVE leak */
 	if (c->x86 >= 6)
-		set_bit(X86_FEATURE_FXSAVE_LEAK, &c->x86_capability);
+		set_bit(X86_FEATURE_FXSAVE_LEAK, (unsigned long *)&c->x86_capability);
 
 	level = get_model_name(c);
 	if (!level) {
@@ -689,7 +689,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 
 	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
 	if (c->x86_power & (1<<8))
-		set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
+		set_bit(X86_FEATURE_CONSTANT_TSC, (unsigned long *)&c->x86_capability);
 
 	/* Multi core CPU? */
 	if (c->extended_cpuid_level >= 0x80000008)
@@ -702,14 +702,14 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		num_cache_leaves = 3;
 
 	if (c->x86 == 0xf || c->x86 == 0x10 || c->x86 == 0x11)
-		set_bit(X86_FEATURE_K8, &c->x86_capability);
+		set_bit(X86_FEATURE_K8, (unsigned long *)&c->x86_capability);
 
 	/* RDTSC can be speculated around */
-	clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+	clear_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
 
 	/* Family 10 doesn't support C states in MWAIT so don't use it */
 	if (c->x86 == 0x10 && !force_mwait)
-		clear_bit(X86_FEATURE_MWAIT, &c->x86_capability);
+		clear_bit(X86_FEATURE_MWAIT, (unsigned long *)&c->x86_capability);
 
 	if (amd_apic_timer_broken())
 		disable_apic_timer = 1;
@@ -811,16 +811,17 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		unsigned eax = cpuid_eax(10);
 		/* Check for version and the number of counters */
 		if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
-			set_bit(X86_FEATURE_ARCH_PERFMON, &c->x86_capability);
+			set_bit(X86_FEATURE_ARCH_PERFMON,
+				(unsigned long *)&c->x86_capability);
 	}
 
 	if (cpu_has_ds) {
 		unsigned int l1, l2;
 		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
 		if (!(l1 & (1<<11)))
-			set_bit(X86_FEATURE_BTS, c->x86_capability);
+			set_bit(X86_FEATURE_BTS, (unsigned long *)c->x86_capability);
 		if (!(l1 & (1<<12)))
-			set_bit(X86_FEATURE_PEBS, c->x86_capability);
+			set_bit(X86_FEATURE_PEBS, (unsigned long *)c->x86_capability);
 	}
 
 	n = c->extended_cpuid_level;
@@ -839,13 +840,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		c->x86_cache_alignment = c->x86_clflush_size * 2;
 	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
 	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
-		set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
+		set_bit(X86_FEATURE_CONSTANT_TSC, (unsigned long *)&c->x86_capability);
 	if (c->x86 == 6)
-		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
 	if (c->x86 == 15)
-		set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+		set_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
 	else
-		clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+		clear_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
 	c->x86_max_cores = intel_num_cpu_cores(c);
 
 	srat_detect_node();
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 8ac8eb620428..ac1089f2b917 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -691,7 +691,7 @@ do_rest:
 	}
 	if (boot_error) {
 		cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */
-		clear_bit(cpu, &cpu_initialized); /* was set by cpu_init() */
+		clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */
 		clear_node_cpumask(cpu); /* was set by numa_add_cpu */
 		cpu_clear(cpu, cpu_present_map);
 		cpu_clear(cpu, cpu_possible_map);
@@ -1036,7 +1036,7 @@ void remove_cpu_from_maps(void)
 
 	cpu_clear(cpu, cpu_callout_map);
 	cpu_clear(cpu, cpu_callin_map);
-	clear_bit(cpu, &cpu_initialized); /* was set by cpu_init() */
+	clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */
 	clear_node_cpumask(cpu);
 }
 
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 46b4b5e1a02a..848231481619 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -547,7 +547,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 
 __cpuinit void numa_add_cpu(int cpu)
 {
-	set_bit(cpu, &node_to_cpumask_map[cpu_to_node(cpu)]);
+	set_bit(cpu, (unsigned long *)&node_to_cpumask_map[cpu_to_node(cpu)]);
 }
 
 void __cpuinit numa_set_node(int cpu, int node)
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index acbf6681740d..761922972f6f 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -124,7 +124,7 @@
 	   (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) ||	\
 	   (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) )	\
 	  ? 1 :								\
-	  test_bit(bit, (c)->x86_capability))
+	 test_bit(bit, (unsigned long *)(c)->x86_capability))
 #define boot_cpu_has(bit)	cpu_has(&boot_cpu_data, bit)
 
 #define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index c3c20db1fba3..4449889bb0c0 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -32,7 +32,7 @@ extern void __init init_cpu_to_node(void);
 
 static inline void clear_node_cpumask(int cpu)
 {
-	clear_bit(cpu, &node_to_cpumask_map[cpu_to_node(cpu)]);
+	clear_bit(cpu, (unsigned long *)&node_to_cpumask_map[cpu_to_node(cpu)]);
 }
 
 #else
diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index 9c4ad755d7e5..dfbdfb9836f4 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -42,27 +42,27 @@ extern long do_no_restart_syscall(struct restart_block *parm);
 
 static inline void set_ti_thread_flag(struct thread_info *ti, int flag)
 {
-	set_bit(flag,&ti->flags);
+	set_bit(flag, (unsigned long *)&ti->flags);
 }
 
 static inline void clear_ti_thread_flag(struct thread_info *ti, int flag)
 {
-	clear_bit(flag,&ti->flags);
+	clear_bit(flag, (unsigned long *)&ti->flags);
 }
 
 static inline int test_and_set_ti_thread_flag(struct thread_info *ti, int flag)
 {
-	return test_and_set_bit(flag,&ti->flags);
+	return test_and_set_bit(flag, (unsigned long *)&ti->flags);
 }
 
 static inline int test_and_clear_ti_thread_flag(struct thread_info *ti, int flag)
 {
-	return test_and_clear_bit(flag,&ti->flags);
+	return test_and_clear_bit(flag, (unsigned long *)&ti->flags);
 }
 
 static inline int test_ti_thread_flag(struct thread_info *ti, int flag)
 {
-	return test_bit(flag,&ti->flags);
+	return test_bit(flag, (unsigned long *)&ti->flags);
 }
 
 #define set_thread_flag(flag) \
-- 
cgit v1.2.3


From 53756d3722172815f52272b28c6d5d5e9639adde Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:30:55 +0100
Subject: x86: add set/clear_cpu_cap operations

The patch to suppress bitops-related warnings added a pile of ugly
casts.  Many of these were related to the management of x86 CPU
capabilities.  Clean these up by adding specific set/clear_cpu_cap
macros, and use them consistently.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/alternative.c              | 13 +++++++------
 arch/x86/kernel/apic_32.c                  |  8 ++++----
 arch/x86/kernel/apic_64.c                  |  2 +-
 arch/x86/kernel/cpu/addon_cpuid_features.c |  2 +-
 arch/x86/kernel/cpu/mcheck/mce_64.c        |  2 +-
 arch/x86/kernel/setup_32.c                 |  2 +-
 arch/x86/kernel/setup_64.c                 | 29 ++++++++++++++---------------
 arch/x86/kernel/vmi_32.c                   | 10 +++++-----
 include/asm-x86/cpufeature.h               |  5 ++++-
 9 files changed, 38 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index d6405e0842b5..cdc43242da92 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -356,15 +356,15 @@ void alternatives_smp_switch(int smp)
 	spin_lock_irqsave(&smp_alt, flags);
 	if (smp) {
 		printk(KERN_INFO "SMP alternatives: switching to SMP code\n");
-		clear_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
-		clear_bit(X86_FEATURE_UP, cpu_data(0).x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
+		clear_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
 		list_for_each_entry(mod, &smp_alt_modules, next)
 			alternatives_smp_lock(mod->locks, mod->locks_end,
 					      mod->text, mod->text_end);
 	} else {
 		printk(KERN_INFO "SMP alternatives: switching to UP code\n");
-		set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
-		set_bit(X86_FEATURE_UP, cpu_data(0).x86_capability);
+		set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
+		set_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
 		list_for_each_entry(mod, &smp_alt_modules, next)
 			alternatives_smp_unlock(mod->locks, mod->locks_end,
 						mod->text, mod->text_end);
@@ -431,8 +431,9 @@ void __init alternative_instructions(void)
 	if (smp_alt_once) {
 		if (1 == num_possible_cpus()) {
 			printk(KERN_INFO "SMP alternatives: switching to UP code\n");
-			set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
-			set_bit(X86_FEATURE_UP, cpu_data(0).x86_capability);
+			set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
+			set_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
+
 			alternatives_smp_unlock(__smp_locks, __smp_locks_end,
 						_text, _etext);
 		}
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 56352c11a896..1ee443a8e61b 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1078,7 +1078,7 @@ static int __init detect_init_APIC (void)
 		printk(KERN_WARNING "Could not enable APIC!\n");
 		return -1;
 	}
-	set_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+	set_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 	mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
 
 	/* The BIOS may have set up the APIC at some other address */
@@ -1168,7 +1168,7 @@ fake_ioapic_page:
 int __init APIC_init_uniprocessor (void)
 {
 	if (enable_local_apic < 0)
-		clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 
 	if (!smp_found_config && !cpu_has_apic)
 		return -1;
@@ -1180,7 +1180,7 @@ int __init APIC_init_uniprocessor (void)
 	    APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) {
 		printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n",
 		       boot_cpu_physical_apicid);
-		clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 		return -1;
 	}
 
@@ -1536,7 +1536,7 @@ early_param("lapic", parse_lapic);
 static int __init parse_nolapic(char *arg)
 {
 	enable_local_apic = -1;
-	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+	clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 	return 0;
 }
 early_param("nolapic", parse_nolapic);
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 027004262105..ab4ae50399fd 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -1211,7 +1211,7 @@ early_param("apic", apic_set_verbosity);
 static __init int setup_disableapic(char *str)
 {
 	disable_apic = 1;
-	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+	clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 	return 0;
 }
 early_param("disableapic", setup_disableapic);
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index 3e91d3ee26ec..238468ae1993 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -45,6 +45,6 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 			&regs[CR_ECX], &regs[CR_EDX]);
 
 		if (regs[cb->reg] & (1 << cb->bit))
-			set_bit(cb->feature, c->x86_capability);
+			set_cpu_cap(c, cb->feature);
 	}
 }
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 8ef6a6bfd112..3c7672c40cf4 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -80,7 +80,7 @@ void mce_log(struct mce *mce)
 			/* When the buffer fills up discard new entries. Assume
 			   that the earlier errors are the more interesting. */
 			if (entry >= MCE_LOG_LEN) {
-				set_bit(MCE_OVERFLOW, &mcelog.flags);
+				set_bit(MCE_OVERFLOW, (unsigned long *)&mcelog.flags);
 				return;
 			}
 			/* Old left over entry. Skip. */
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 32edf70d6b0d..e9ede0fc585a 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -238,7 +238,7 @@ static int __init parse_mem(char *arg)
 		return -EINVAL;
 
 	if (strcmp(arg, "nopentium") == 0) {
-		clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PSE);
 		disable_pse = 1;
 	} else {
 		/* If the user specifies memory size, we
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 63dd39b843b5..ce4d6b52ce36 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -667,13 +667,13 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	level = cpuid_eax(1);
 	if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) ||
 			     level >= 0x0f58))
-		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	if (c->x86 == 0x10 || c->x86 == 0x11)
-		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 
 	/* Enable workaround for FXSAVE leak */
 	if (c->x86 >= 6)
-		set_bit(X86_FEATURE_FXSAVE_LEAK, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
 
 	level = get_model_name(c);
 	if (!level) {
@@ -689,7 +689,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 
 	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
 	if (c->x86_power & (1<<8))
-		set_bit(X86_FEATURE_CONSTANT_TSC, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 
 	/* Multi core CPU? */
 	if (c->extended_cpuid_level >= 0x80000008)
@@ -702,14 +702,14 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		num_cache_leaves = 3;
 
 	if (c->x86 == 0xf || c->x86 == 0x10 || c->x86 == 0x11)
-		set_bit(X86_FEATURE_K8, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_K8);
 
 	/* RDTSC can be speculated around */
-	clear_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
+	clear_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
 
 	/* Family 10 doesn't support C states in MWAIT so don't use it */
 	if (c->x86 == 0x10 && !force_mwait)
-		clear_bit(X86_FEATURE_MWAIT, (unsigned long *)&c->x86_capability);
+		clear_cpu_cap(c, X86_FEATURE_MWAIT);
 
 	if (amd_apic_timer_broken())
 		disable_apic_timer = 1;
@@ -811,17 +811,16 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		unsigned eax = cpuid_eax(10);
 		/* Check for version and the number of counters */
 		if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
-			set_bit(X86_FEATURE_ARCH_PERFMON,
-				(unsigned long *)&c->x86_capability);
+			set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
 	}
 
 	if (cpu_has_ds) {
 		unsigned int l1, l2;
 		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
 		if (!(l1 & (1<<11)))
-			set_bit(X86_FEATURE_BTS, (unsigned long *)c->x86_capability);
+			set_cpu_cap(c, X86_FEATURE_BTS);
 		if (!(l1 & (1<<12)))
-			set_bit(X86_FEATURE_PEBS, (unsigned long *)c->x86_capability);
+			set_cpu_cap(c, X86_FEATURE_PEBS);
 	}
 
 	n = c->extended_cpuid_level;
@@ -840,13 +839,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		c->x86_cache_alignment = c->x86_clflush_size * 2;
 	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
 	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
-		set_bit(X86_FEATURE_CONSTANT_TSC, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 	if (c->x86 == 6)
-		set_bit(X86_FEATURE_REP_GOOD, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	if (c->x86 == 15)
-		set_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
+		set_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
 	else
-		clear_bit(X86_FEATURE_SYNC_RDTSC, (unsigned long *)&c->x86_capability);
+		clear_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
 	c->x86_max_cores = intel_num_cpu_cores(c);
 
 	srat_detect_node();
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index aacce426cbd0..87e5633805a9 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -963,19 +963,19 @@ static int __init parse_vmi(char *arg)
 		return -EINVAL;
 
 	if (!strcmp(arg, "disable_pge")) {
-		clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PGE);
 		disable_pge = 1;
 	} else if (!strcmp(arg, "disable_pse")) {
-		clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PSE);
 		disable_pse = 1;
 	} else if (!strcmp(arg, "disable_sep")) {
-		clear_bit(X86_FEATURE_SEP, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_SEP);
 		disable_sep = 1;
 	} else if (!strcmp(arg, "disable_tsc")) {
-		clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC);
 		disable_tsc = 1;
 	} else if (!strcmp(arg, "disable_mtrr")) {
-		clear_bit(X86_FEATURE_MTRR, boot_cpu_data.x86_capability);
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_MTRR);
 		disable_mtrr = 1;
 	} else if (!strcmp(arg, "disable_timer")) {
 		disable_vmi_timer = 1;
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 761922972f6f..87dd900154d1 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -124,9 +124,12 @@
 	   (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) ||	\
 	   (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) )	\
 	  ? 1 :								\
-	 test_bit(bit, (unsigned long *)(c)->x86_capability))
+	 test_bit(bit, (unsigned long *)((c)->x86_capability)))
 #define boot_cpu_has(bit)	cpu_has(&boot_cpu_data, bit)
 
+#define set_cpu_cap(c, bit)	set_bit(bit, (unsigned long *)((c)->x86_capability))
+#define clear_cpu_cap(c, bit)	clear_bit(bit, (unsigned long *)((c)->x86_capability))
+
 #define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
 #define cpu_has_vme		boot_cpu_has(X86_FEATURE_VME)
 #define cpu_has_de		boot_cpu_has(X86_FEATURE_DE)
-- 
cgit v1.2.3


From 65ea5b0349903585bfed9720fa06f5edb4f1cd25 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:56 +0100
Subject: x86: rename the struct pt_regs members for 32/64-bit consistency

We have a lot of code which differs only by the naming of specific
members of structures that contain registers.  In order to enable
additional unifications, this patch drops the e- or r- size prefix
from the register names in struct pt_regs, and drops the x- prefixes
for segment registers on the 32-bit side.

This patch also performs the equivalent renames in some additional
places that might be candidates for unification in the future.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ia32_aout.c           |  30 +++---
 arch/x86/ia32/ia32_binfmt.c         |  42 ++++----
 arch/x86/ia32/ia32_signal.c         | 100 +++++++++----------
 arch/x86/ia32/ptrace32.c            |  44 ++++-----
 arch/x86/ia32/sys_ia32.c            |   6 +-
 arch/x86/kernel/acpi/wakeup_64.S    |  32 +++---
 arch/x86/kernel/asm-offsets_32.c    |  32 +++---
 arch/x86/kernel/asm-offsets_64.c    |  18 ++--
 arch/x86/kernel/cpu/common.c        |   2 +-
 arch/x86/kernel/cpu/mcheck/mce_64.c |  14 +--
 arch/x86/kernel/i8259_32.c          |   2 +-
 arch/x86/kernel/io_apic_64.c        |   2 +-
 arch/x86/kernel/ioport_32.c         |   8 +-
 arch/x86/kernel/ioport_64.c         |   6 +-
 arch/x86/kernel/irq_32.c            |  20 ++--
 arch/x86/kernel/irq_64.c            |  10 +-
 arch/x86/kernel/kprobes_32.c        |  92 ++++++++---------
 arch/x86/kernel/kprobes_64.c        |  69 ++++++-------
 arch/x86/kernel/process_32.c        | 140 +++++++++++++-------------
 arch/x86/kernel/process_64.c        |  44 ++++-----
 arch/x86/kernel/ptrace_32.c         |  26 ++---
 arch/x86/kernel/ptrace_64.c         |  24 ++---
 arch/x86/kernel/signal_32.c         | 192 ++++++++++++++++++------------------
 arch/x86/kernel/signal_64.c         | 115 ++++++++++-----------
 arch/x86/kernel/smp_64.c            |   2 +-
 arch/x86/kernel/smpboot_32.c        |  10 +-
 arch/x86/kernel/step.c              |  15 +--
 arch/x86/kernel/suspend_asm_64.S    |  32 +++---
 arch/x86/kernel/time_32.c           |   8 +-
 arch/x86/kernel/time_64.c           |   4 +-
 arch/x86/kernel/traps_32.c          | 114 ++++++++++-----------
 arch/x86/kernel/traps_64.c          |  84 ++++++++--------
 arch/x86/kernel/vm86_32.c           |  94 +++++++++---------
 arch/x86/kernel/vmi_32.c            |  50 +++++-----
 arch/x86/kernel/vsyscall_64.c       |   2 +-
 arch/x86/lguest/boot.c              |  22 ++---
 arch/x86/mach-voyager/voyager_smp.c |  12 +--
 arch/x86/mm/extable_32.c            |   6 +-
 arch/x86/mm/fault_32.c              |  38 +++----
 arch/x86/mm/fault_64.c              |  22 ++---
 arch/x86/oprofile/backtrace.c       |   6 +-
 arch/x86/xen/enlighten.c            |  18 ++--
 arch/x86/xen/events.c               |   2 +-
 include/asm-x86/compat.h            |   2 +-
 include/asm-x86/elf.h               |  66 ++++++-------
 include/asm-x86/kexec_32.h          |  36 +++----
 include/asm-x86/kexec_64.h          |  20 ++--
 include/asm-x86/kprobes_32.h        |   2 +-
 include/asm-x86/kprobes_64.h        |   2 +-
 include/asm-x86/mce.h               |   4 +-
 include/asm-x86/processor_32.h      |  20 ++--
 include/asm-x86/processor_64.h      |   8 +-
 include/asm-x86/ptrace.h            |  80 ++++++++++++---
 kernel/signal.c                     |   4 +-
 54 files changed, 953 insertions(+), 902 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index a764e4e95314..f1a0f83676dc 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -53,7 +53,7 @@ static void dump_thread32(struct pt_regs *regs, struct user32 *dump)
 /* changed the size calculations - should hopefully work better. lbt */
 	dump->magic = CMAGIC;
 	dump->start_code = 0;
-	dump->start_stack = regs->rsp & ~(PAGE_SIZE - 1);
+	dump->start_stack = regs->sp & ~(PAGE_SIZE - 1);
 	dump->u_tsize = ((unsigned long) current->mm->end_code) >> PAGE_SHIFT;
 	dump->u_dsize = ((unsigned long)
 			 (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
@@ -75,22 +75,22 @@ static void dump_thread32(struct pt_regs *regs, struct user32 *dump)
 		dump->u_ssize = tmp >> PAGE_SHIFT;
 	}
 
-	dump->regs.ebx = regs->rbx;
-	dump->regs.ecx = regs->rcx;
-	dump->regs.edx = regs->rdx;
-	dump->regs.esi = regs->rsi;
-	dump->regs.edi = regs->rdi;
-	dump->regs.ebp = regs->rbp;
-	dump->regs.eax = regs->rax;
+	dump->regs.bx = regs->bx;
+	dump->regs.cx = regs->cx;
+	dump->regs.dx = regs->dx;
+	dump->regs.si = regs->si;
+	dump->regs.di = regs->di;
+	dump->regs.bp = regs->bp;
+	dump->regs.ax = regs->ax;
 	dump->regs.ds = current->thread.ds;
 	dump->regs.es = current->thread.es;
 	asm("movl %%fs,%0" : "=r" (fs)); dump->regs.fs = fs;
 	asm("movl %%gs,%0" : "=r" (gs)); dump->regs.gs = gs;
-	dump->regs.orig_eax = regs->orig_rax;
-	dump->regs.eip = regs->rip;
+	dump->regs.orig_ax = regs->orig_ax;
+	dump->regs.ip = regs->ip;
 	dump->regs.cs = regs->cs;
-	dump->regs.eflags = regs->eflags;
-	dump->regs.esp = regs->rsp;
+	dump->regs.flags = regs->flags;
+	dump->regs.sp = regs->sp;
 	dump->regs.ss = regs->ss;
 
 #if 1 /* FIXME */
@@ -432,9 +432,9 @@ beyond_if:
 	asm volatile("movl %0,%%fs" :: "r" (0)); \
 	asm volatile("movl %0,%%es; movl %0,%%ds": :"r" (__USER32_DS));
 	load_gs_index(0);
-	(regs)->rip = ex.a_entry;
-	(regs)->rsp = current->mm->start_stack;
-	(regs)->eflags = 0x200;
+	(regs)->ip = ex.a_entry;
+	(regs)->sp = current->mm->start_stack;
+	(regs)->flags = 0x200;
 	(regs)->cs = __USER32_CS;
 	(regs)->ss = __USER32_DS;
 	regs->r8 = regs->r9 = regs->r10 = regs->r11 =
diff --git a/arch/x86/ia32/ia32_binfmt.c b/arch/x86/ia32/ia32_binfmt.c
index e32974c3dd3b..806135c3f975 100644
--- a/arch/x86/ia32/ia32_binfmt.c
+++ b/arch/x86/ia32/ia32_binfmt.c
@@ -73,22 +73,22 @@ struct file;
 /* Assumes current==process to be dumped */
 #undef	ELF_CORE_COPY_REGS
 #define ELF_CORE_COPY_REGS(pr_reg, regs)       		\
-	pr_reg[0] = regs->rbx;				\
-	pr_reg[1] = regs->rcx;				\
-	pr_reg[2] = regs->rdx;				\
-	pr_reg[3] = regs->rsi;				\
-	pr_reg[4] = regs->rdi;				\
-	pr_reg[5] = regs->rbp;				\
-	pr_reg[6] = regs->rax;				\
+	pr_reg[0] = regs->bx;				\
+	pr_reg[1] = regs->cx;				\
+	pr_reg[2] = regs->dx;				\
+	pr_reg[3] = regs->si;				\
+	pr_reg[4] = regs->di;				\
+	pr_reg[5] = regs->bp;				\
+	pr_reg[6] = regs->ax;				\
 	pr_reg[7] = _GET_SEG(ds);   			\
 	pr_reg[8] = _GET_SEG(es);			\
 	pr_reg[9] = _GET_SEG(fs);			\
 	pr_reg[10] = _GET_SEG(gs);			\
-	pr_reg[11] = regs->orig_rax;			\
-	pr_reg[12] = regs->rip;				\
+	pr_reg[11] = regs->orig_ax;			\
+	pr_reg[12] = regs->ip;				\
 	pr_reg[13] = regs->cs;				\
-	pr_reg[14] = regs->eflags;			\
-	pr_reg[15] = regs->rsp;				\
+	pr_reg[14] = regs->flags;			\
+	pr_reg[15] = regs->sp;				\
 	pr_reg[16] = regs->ss;
 
 
@@ -205,9 +205,9 @@ do {							\
 	asm volatile("movl %0,%%fs" :: "r" (0)); \
 	asm volatile("movl %0,%%es; movl %0,%%ds": :"r" (__USER32_DS)); \
 	load_gs_index(0); \
-	(regs)->rip = (new_rip); \
-	(regs)->rsp = (new_rsp); \
-	(regs)->eflags = 0x200; \
+	(regs)->ip = (new_rip); \
+	(regs)->sp = (new_rsp); \
+	(regs)->flags = X86_EFLAGS_IF; \
 	(regs)->cs = __USER32_CS; \
 	(regs)->ss = __USER32_DS; \
 	set_fs(USER_DS); \
@@ -233,13 +233,13 @@ extern int syscall32_setup_pages(struct linux_binprm *, int exstack);
 static void elf32_init(struct pt_regs *regs)
 {
 	struct task_struct *me = current; 
-	regs->rdi = 0;
-	regs->rsi = 0;
-	regs->rdx = 0;
-	regs->rcx = 0;
-	regs->rax = 0;
-	regs->rbx = 0; 
-	regs->rbp = 0; 
+	regs->di = 0;
+	regs->si = 0;
+	regs->dx = 0;
+	regs->cx = 0;
+	regs->ax = 0;
+	regs->bx = 0;
+	regs->bp = 0;
 	regs->r8 = regs->r9 = regs->r10 = regs->r11 = regs->r12 =
 		regs->r13 = regs->r14 = regs->r15 = 0; 
     me->thread.fs = 0; 
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 39356a756b28..f2da443f8c7b 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -154,7 +154,7 @@ asmlinkage long sys32_sigaltstack(const stack_ia32_t __user *uss_ptr,
 	}
 	seg = get_fs();
 	set_fs(KERNEL_DS);
-	ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss, regs->rsp);
+	ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss, regs->sp);
 	set_fs(seg);
 	if (ret >= 0 && uoss_ptr)  {
 		if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(stack_ia32_t)) ||
@@ -195,7 +195,7 @@ struct rt_sigframe
 #define COPY(x)		{ \
 	unsigned int reg;			\
 	err |= __get_user(reg, &sc->e ##x);	\
-	regs->r ## x = reg;			\
+	regs->x = reg;			\
 }
 
 #define RELOAD_SEG(seg,mask)						\
@@ -220,7 +220,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 #if DEBUG_SIG
 	printk(KERN_DEBUG "SIG restore_sigcontext: "
 	       "sc=%p err(%x) eip(%x) cs(%x) flg(%x)\n",
-	       sc, sc->err, sc->eip, sc->cs, sc->eflags);
+	       sc, sc->err, sc->ip, sc->cs, sc->flags);
 #endif
 
 	/*
@@ -249,9 +249,9 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 	regs->ss |= 3;
 
 	err |= __get_user(tmpflags, &sc->eflags);
-	regs->eflags = (regs->eflags & ~0x40DD5) | (tmpflags & 0x40DD5);
+	regs->flags = (regs->flags & ~0x40DD5) | (tmpflags & 0x40DD5);
 	/* disable syscall checks */
-	regs->orig_rax = -1;
+	regs->orig_ax = -1;
 
 	err |= __get_user(tmp, &sc->fpstate);
 	buf = compat_ptr(tmp);
@@ -279,9 +279,9 @@ badframe:
 
 asmlinkage long sys32_sigreturn(struct pt_regs *regs)
 {
-	struct sigframe __user *frame = (struct sigframe __user *)(regs->rsp-8);
+	struct sigframe __user *frame = (struct sigframe __user *)(regs->sp-8);
 	sigset_t set;
-	unsigned int eax;
+	unsigned int ax;
 
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
 		goto badframe;
@@ -298,9 +298,9 @@ asmlinkage long sys32_sigreturn(struct pt_regs *regs)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	if (ia32_restore_sigcontext(regs, &frame->sc, &eax))
+	if (ia32_restore_sigcontext(regs, &frame->sc, &ax))
 		goto badframe;
-	return eax;
+	return ax;
 
 badframe:
 	signal_fault(regs, frame, "32bit sigreturn");
@@ -311,10 +311,10 @@ asmlinkage long sys32_rt_sigreturn(struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
 	sigset_t set;
-	unsigned int eax;
+	unsigned int ax;
 	struct pt_regs tregs;
 
-	frame = (struct rt_sigframe __user *)(regs->rsp - 4);
+	frame = (struct rt_sigframe __user *)(regs->sp - 4);
 
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
 		goto badframe;
@@ -327,14 +327,14 @@ asmlinkage long sys32_rt_sigreturn(struct pt_regs *regs)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	if (ia32_restore_sigcontext(regs, &frame->uc.uc_mcontext, &eax))
+	if (ia32_restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
 		goto badframe;
 
 	tregs = *regs;
 	if (sys32_sigaltstack(&frame->uc.uc_stack, NULL, &tregs) == -EFAULT)
 		goto badframe;
 
-	return eax;
+	return ax;
 
 badframe:
 	signal_fault(regs, frame, "32bit rt sigreturn");
@@ -361,21 +361,21 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
 	__asm__("movl %%es,%0" : "=r"(tmp): "0"(tmp));
 	err |= __put_user(tmp, (unsigned int __user *)&sc->es);
 
-	err |= __put_user((u32)regs->rdi, &sc->edi);
-	err |= __put_user((u32)regs->rsi, &sc->esi);
-	err |= __put_user((u32)regs->rbp, &sc->ebp);
-	err |= __put_user((u32)regs->rsp, &sc->esp);
-	err |= __put_user((u32)regs->rbx, &sc->ebx);
-	err |= __put_user((u32)regs->rdx, &sc->edx);
-	err |= __put_user((u32)regs->rcx, &sc->ecx);
-	err |= __put_user((u32)regs->rax, &sc->eax);
+	err |= __put_user((u32)regs->di, &sc->edi);
+	err |= __put_user((u32)regs->si, &sc->esi);
+	err |= __put_user((u32)regs->bp, &sc->ebp);
+	err |= __put_user((u32)regs->sp, &sc->esp);
+	err |= __put_user((u32)regs->bx, &sc->ebx);
+	err |= __put_user((u32)regs->dx, &sc->edx);
+	err |= __put_user((u32)regs->cx, &sc->ecx);
+	err |= __put_user((u32)regs->ax, &sc->eax);
 	err |= __put_user((u32)regs->cs, &sc->cs);
 	err |= __put_user((u32)regs->ss, &sc->ss);
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
-	err |= __put_user((u32)regs->rip, &sc->eip);
-	err |= __put_user((u32)regs->eflags, &sc->eflags);
-	err |= __put_user((u32)regs->rsp, &sc->esp_at_signal);
+	err |= __put_user((u32)regs->ip, &sc->eip);
+	err |= __put_user((u32)regs->flags, &sc->eflags);
+	err |= __put_user((u32)regs->sp, &sc->esp_at_signal);
 
 	tmp = save_i387_ia32(current, fpstate, regs, 0);
 	if (tmp < 0)
@@ -400,28 +400,28 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
 static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
 				 size_t frame_size)
 {
-	unsigned long rsp;
+	unsigned long sp;
 
 	/* Default to using normal stack */
-	rsp = regs->rsp;
+	sp = regs->sp;
 
 	/* This is the X/Open sanctioned signal stack switching.  */
 	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(rsp) == 0)
-			rsp = current->sas_ss_sp + current->sas_ss_size;
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
 	}
 
 	/* This is the legacy signal stack switching. */
 	else if ((regs->ss & 0xffff) != __USER_DS &&
 		!(ka->sa.sa_flags & SA_RESTORER) &&
 		 ka->sa.sa_restorer)
-		rsp = (unsigned long) ka->sa.sa_restorer;
+		sp = (unsigned long) ka->sa.sa_restorer;
 
-	rsp -= frame_size;
+	sp -= frame_size;
 	/* Align the stack pointer according to the i386 ABI,
 	 * i.e. so that on function entry ((sp + 4) & 15) == 0. */
-	rsp = ((rsp + 4) & -16ul) - 4;
-	return (void __user *) rsp;
+	sp = ((sp + 4) & -16ul) - 4;
+	return (void __user *) sp;
 }
 
 int ia32_setup_frame(int sig, struct k_sigaction *ka,
@@ -486,13 +486,13 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->rsp = (unsigned long) frame;
-	regs->rip = (unsigned long) ka->sa.sa_handler;
+	regs->sp = (unsigned long) frame;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
 
 	/* Make -mregparm=3 work */
-	regs->rax = sig;
-	regs->rdx = 0;
-	regs->rcx = 0;
+	regs->ax = sig;
+	regs->dx = 0;
+	regs->cx = 0;
 
 	asm volatile("movl %0,%%ds" :: "r" (__USER32_DS));
 	asm volatile("movl %0,%%es" :: "r" (__USER32_DS));
@@ -501,13 +501,13 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	regs->ss = __USER32_DS;
 
 	set_fs(USER_DS);
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk(KERN_DEBUG "SIG deliver (%s:%d): sp=%p pc=%lx ra=%u\n",
-	       current->comm, current->pid, frame, regs->rip, frame->pretcode);
+	       current->comm, current->pid, frame, regs->ip, frame->pretcode);
 #endif
 
 	return 0;
@@ -556,7 +556,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->rsp),
+	err |= __put_user(sas_ss_flags(regs->sp),
 			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
 	err |= ia32_setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate,
@@ -581,18 +581,18 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->rsp = (unsigned long) frame;
-	regs->rip = (unsigned long) ka->sa.sa_handler;
+	regs->sp = (unsigned long) frame;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
 
 	/* Make -mregparm=3 work */
-	regs->rax = sig;
-	regs->rdx = (unsigned long) &frame->info;
-	regs->rcx = (unsigned long) &frame->uc;
+	regs->ax = sig;
+	regs->dx = (unsigned long) &frame->info;
+	regs->cx = (unsigned long) &frame->uc;
 
 	/* Make -mregparm=3 work */
-	regs->rax = sig;
-	regs->rdx = (unsigned long) &frame->info;
-	regs->rcx = (unsigned long) &frame->uc;
+	regs->ax = sig;
+	regs->dx = (unsigned long) &frame->info;
+	regs->cx = (unsigned long) &frame->uc;
 
 	asm volatile("movl %0,%%ds" :: "r" (__USER32_DS));
 	asm volatile("movl %0,%%es" :: "r" (__USER32_DS));
@@ -601,13 +601,13 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->ss = __USER32_DS;
 
 	set_fs(USER_DS);
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk(KERN_DEBUG "SIG deliver (%s:%d): sp=%p pc=%lx ra=%u\n",
-	       current->comm, current->pid, frame, regs->rip, frame->pretcode);
+	       current->comm, current->pid, frame, regs->ip, frame->pretcode);
 #endif
 
 	return 0;
diff --git a/arch/x86/ia32/ptrace32.c b/arch/x86/ia32/ptrace32.c
index 8c6fb9d8352b..1e382e3bd882 100644
--- a/arch/x86/ia32/ptrace32.c
+++ b/arch/x86/ia32/ptrace32.c
@@ -73,19 +73,19 @@ static int putreg32(struct task_struct *child, unsigned regno, u32 val)
 		stack[offsetof(struct pt_regs, cs)/8] = val & 0xffff;
 		break;
 
-	R32(ebx, rbx);
-	R32(ecx, rcx);
-	R32(edx, rdx);
-	R32(edi, rdi);
-	R32(esi, rsi);
-	R32(ebp, rbp);
-	R32(eax, rax);
-	R32(orig_eax, orig_rax);
-	R32(eip, rip);
-	R32(esp, rsp);
+	R32(ebx, bx);
+	R32(ecx, cx);
+	R32(edx, dx);
+	R32(edi, di);
+	R32(esi, si);
+	R32(ebp, bp);
+	R32(eax, ax);
+	R32(orig_eax, orig_ax);
+	R32(eip, ip);
+	R32(esp, sp);
 
 	case offsetof(struct user32, regs.eflags): {
-		__u64 *flags = &stack[offsetof(struct pt_regs, eflags)/8];
+		__u64 *flags = &stack[offsetof(struct pt_regs, flags)/8];
 
 		val &= FLAG_MASK;
 		/*
@@ -145,22 +145,22 @@ static int getreg32(struct task_struct *child, unsigned regno, u32 *val)
 
 	R32(cs, cs);
 	R32(ss, ss);
-	R32(ebx, rbx);
-	R32(ecx, rcx);
-	R32(edx, rdx);
-	R32(edi, rdi);
-	R32(esi, rsi);
-	R32(ebp, rbp);
-	R32(eax, rax);
-	R32(orig_eax, orig_rax);
-	R32(eip, rip);
-	R32(esp, rsp);
+	R32(ebx, bx);
+	R32(ecx, cx);
+	R32(edx, dx);
+	R32(edi, di);
+	R32(esi, si);
+	R32(ebp, bp);
+	R32(eax, ax);
+	R32(orig_eax, orig_ax);
+	R32(eip, ip);
+	R32(esp, sp);
 
 	case offsetof(struct user32, regs.eflags):
 		/*
 		 * If the debugger set TF, hide it from the readout.
 		 */
-		*val = stack[offsetof(struct pt_regs, eflags)/8];
+		*val = stack[offsetof(struct pt_regs, flags)/8];
 		if (test_tsk_thread_flag(child, TIF_FORCED_TF))
 			*val &= ~X86_EFLAGS_TF;
 		break;
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index 58991abc5b59..abf71d26fc2a 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -816,11 +816,11 @@ asmlinkage long sys32_execve(char __user *name, compat_uptr_t __user *argv,
 asmlinkage long sys32_clone(unsigned int clone_flags, unsigned int newsp,
 			    struct pt_regs *regs)
 {
-	void __user *parent_tid = (void __user *)regs->rdx;
-	void __user *child_tid = (void __user *)regs->rdi;
+	void __user *parent_tid = (void __user *)regs->dx;
+	void __user *child_tid = (void __user *)regs->di;
 
 	if (!newsp)
-		newsp = regs->rsp;
+		newsp = regs->sp;
 	return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
 }
 
diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
index 5ed3bc5c61d7..2e1b9e0d0767 100644
--- a/arch/x86/kernel/acpi/wakeup_64.S
+++ b/arch/x86/kernel/acpi/wakeup_64.S
@@ -344,13 +344,13 @@ do_suspend_lowlevel:
 	call	save_processor_state
 
 	movq	$saved_context, %rax
-	movq	%rsp, pt_regs_rsp(%rax)
-	movq	%rbp, pt_regs_rbp(%rax)
-	movq	%rsi, pt_regs_rsi(%rax)
-	movq	%rdi, pt_regs_rdi(%rax)
-	movq	%rbx, pt_regs_rbx(%rax)
-	movq	%rcx, pt_regs_rcx(%rax)
-	movq	%rdx, pt_regs_rdx(%rax)
+	movq	%rsp, pt_regs_sp(%rax)
+	movq	%rbp, pt_regs_bp(%rax)
+	movq	%rsi, pt_regs_si(%rax)
+	movq	%rdi, pt_regs_di(%rax)
+	movq	%rbx, pt_regs_bx(%rax)
+	movq	%rcx, pt_regs_cx(%rax)
+	movq	%rdx, pt_regs_dx(%rax)
 	movq	%r8, pt_regs_r8(%rax)
 	movq	%r9, pt_regs_r9(%rax)
 	movq	%r10, pt_regs_r10(%rax)
@@ -360,7 +360,7 @@ do_suspend_lowlevel:
 	movq	%r14, pt_regs_r14(%rax)
 	movq	%r15, pt_regs_r15(%rax)
 	pushfq
-	popq	pt_regs_eflags(%rax)
+	popq	pt_regs_flags(%rax)
 
 	movq	$.L97, saved_rip(%rip)
 
@@ -391,15 +391,15 @@ do_suspend_lowlevel:
 	movq	%rbx, %cr2
 	movq	saved_context_cr0(%rax), %rbx
 	movq	%rbx, %cr0
-	pushq	pt_regs_eflags(%rax)
+	pushq	pt_regs_flags(%rax)
 	popfq
-	movq	pt_regs_rsp(%rax), %rsp
-	movq	pt_regs_rbp(%rax), %rbp
-	movq	pt_regs_rsi(%rax), %rsi
-	movq	pt_regs_rdi(%rax), %rdi
-	movq	pt_regs_rbx(%rax), %rbx
-	movq	pt_regs_rcx(%rax), %rcx
-	movq	pt_regs_rdx(%rax), %rdx
+	movq	pt_regs_sp(%rax), %rsp
+	movq	pt_regs_bp(%rax), %rbp
+	movq	pt_regs_si(%rax), %rsi
+	movq	pt_regs_di(%rax), %rdi
+	movq	pt_regs_bx(%rax), %rbx
+	movq	pt_regs_cx(%rax), %rcx
+	movq	pt_regs_dx(%rax), %rdx
 	movq	pt_regs_r8(%rax), %r8
 	movq	pt_regs_r9(%rax), %r9
 	movq	pt_regs_r10(%rax), %r10
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index fd7464d23339..a3a8be7618d1 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -75,22 +75,22 @@ void foo(void)
 	OFFSET(GDS_pad, Xgt_desc_struct, pad);
 	BLANK();
 
-	OFFSET(PT_EBX, pt_regs, ebx);
-	OFFSET(PT_ECX, pt_regs, ecx);
-	OFFSET(PT_EDX, pt_regs, edx);
-	OFFSET(PT_ESI, pt_regs, esi);
-	OFFSET(PT_EDI, pt_regs, edi);
-	OFFSET(PT_EBP, pt_regs, ebp);
-	OFFSET(PT_EAX, pt_regs, eax);
-	OFFSET(PT_DS,  pt_regs, xds);
-	OFFSET(PT_ES,  pt_regs, xes);
-	OFFSET(PT_FS,  pt_regs, xfs);
-	OFFSET(PT_ORIG_EAX, pt_regs, orig_eax);
-	OFFSET(PT_EIP, pt_regs, eip);
-	OFFSET(PT_CS,  pt_regs, xcs);
-	OFFSET(PT_EFLAGS, pt_regs, eflags);
-	OFFSET(PT_OLDESP, pt_regs, esp);
-	OFFSET(PT_OLDSS,  pt_regs, xss);
+	OFFSET(PT_EBX, pt_regs, bx);
+	OFFSET(PT_ECX, pt_regs, cx);
+	OFFSET(PT_EDX, pt_regs, dx);
+	OFFSET(PT_ESI, pt_regs, si);
+	OFFSET(PT_EDI, pt_regs, di);
+	OFFSET(PT_EBP, pt_regs, bp);
+	OFFSET(PT_EAX, pt_regs, ax);
+	OFFSET(PT_DS,  pt_regs, ds);
+	OFFSET(PT_ES,  pt_regs, es);
+	OFFSET(PT_FS,  pt_regs, fs);
+	OFFSET(PT_ORIG_EAX, pt_regs, orig_ax);
+	OFFSET(PT_EIP, pt_regs, ip);
+	OFFSET(PT_CS,  pt_regs, cs);
+	OFFSET(PT_EFLAGS, pt_regs, flags);
+	OFFSET(PT_OLDESP, pt_regs, sp);
+	OFFSET(PT_OLDSS,  pt_regs, ss);
 	BLANK();
 
 	OFFSET(EXEC_DOMAIN_handler, exec_domain, handler);
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index c27c646214f4..2e918ebf21d3 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -83,14 +83,14 @@ int main(void)
 	DEFINE(pbe_next, offsetof(struct pbe, next));
 	BLANK();
 #define ENTRY(entry) DEFINE(pt_regs_ ## entry, offsetof(struct pt_regs, entry))
-	ENTRY(rbx);
-	ENTRY(rbx);
-	ENTRY(rcx);
-	ENTRY(rdx);
-	ENTRY(rsp);
-	ENTRY(rbp);
-	ENTRY(rsi);
-	ENTRY(rdi);
+	ENTRY(bx);
+	ENTRY(bx);
+	ENTRY(cx);
+	ENTRY(dx);
+	ENTRY(sp);
+	ENTRY(bp);
+	ENTRY(si);
+	ENTRY(di);
 	ENTRY(r8);
 	ENTRY(r9);
 	ENTRY(r10);
@@ -99,7 +99,7 @@ int main(void)
 	ENTRY(r13);
 	ENTRY(r14);
 	ENTRY(r15);
-	ENTRY(eflags);
+	ENTRY(flags);
 	BLANK();
 #undef ENTRY
 #define ENTRY(entry) DEFINE(saved_context_ ## entry, offsetof(struct saved_context, entry))
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index e2fcf2051bdb..5db2a163bf4b 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -634,7 +634,7 @@ void __init early_cpu_init(void)
 struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
 {
 	memset(regs, 0, sizeof(struct pt_regs));
-	regs->xfs = __KERNEL_PERCPU;
+	regs->fs = __KERNEL_PERCPU;
 	return regs;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 3c7672c40cf4..0adad772d0da 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -110,12 +110,12 @@ static void print_mce(struct mce *m)
 	       KERN_EMERG
 	       "CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n",
 	       m->cpu, m->mcgstatus, m->bank, m->status);
-	if (m->rip) {
+	if (m->ip) {
 		printk(KERN_EMERG "RIP%s %02x:<%016Lx> ",
 		       !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
-		       m->cs, m->rip);
+		       m->cs, m->ip);
 		if (m->cs == __KERNEL_CS)
-			print_symbol("{%s}", m->rip);
+			print_symbol("{%s}", m->ip);
 		printk("\n");
 	}
 	printk(KERN_EMERG "TSC %Lx ", m->tsc);
@@ -156,16 +156,16 @@ static int mce_available(struct cpuinfo_x86 *c)
 static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
 {
 	if (regs && (m->mcgstatus & MCG_STATUS_RIPV)) {
-		m->rip = regs->rip;
+		m->ip = regs->ip;
 		m->cs = regs->cs;
 	} else {
-		m->rip = 0;
+		m->ip = 0;
 		m->cs = 0;
 	}
 	if (rip_msr) {
 		/* Assume the RIP in the MSR is exact. Is this true? */
 		m->mcgstatus |= MCG_STATUS_EIPV;
-		rdmsrl(rip_msr, m->rip);
+		rdmsrl(rip_msr, m->ip);
 		m->cs = 0;
 	}
 }
@@ -288,7 +288,7 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 		 * instruction which caused the MCE.
 		 */
 		if (m.mcgstatus & MCG_STATUS_EIPV)
-			user_space = panicm.rip && (panicm.cs & 3);
+			user_space = panicm.ip && (panicm.cs & 3);
 
 		/*
 		 * If we know that the error was in user space, send a
diff --git a/arch/x86/kernel/i8259_32.c b/arch/x86/kernel/i8259_32.c
index 3321ce669295..f201e7da1bbc 100644
--- a/arch/x86/kernel/i8259_32.c
+++ b/arch/x86/kernel/i8259_32.c
@@ -339,7 +339,7 @@ static irqreturn_t math_error_irq(int cpl, void *dev_id)
 	outb(0,0xF0);
 	if (ignore_fpu_irq || !boot_cpu_data.hard_math)
 		return IRQ_NONE;
-	math_error((void __user *)get_irq_regs()->eip);
+	math_error((void __user *)get_irq_regs()->ip);
 	return IRQ_HANDLED;
 }
 
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 4ef85a3b3f9f..fa70005be5e8 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1412,7 +1412,7 @@ static void irq_complete_move(unsigned int irq)
 	if (likely(!cfg->move_in_progress))
 		return;
 
-	vector = ~get_irq_regs()->orig_rax;
+	vector = ~get_irq_regs()->orig_ax;
 	me = smp_processor_id();
 	if ((vector == cfg->vector) && cpu_isset(me, cfg->domain)) {
 		cpumask_t cleanup_mask;
diff --git a/arch/x86/kernel/ioport_32.c b/arch/x86/kernel/ioport_32.c
index c281ffa18259..9295e01ff49c 100644
--- a/arch/x86/kernel/ioport_32.c
+++ b/arch/x86/kernel/ioport_32.c
@@ -100,7 +100,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
  * beyond the 0x3ff range: to get the full 65536 ports bitmapped
  * you'd need 8kB of bitmaps/process, which is a bit excessive.
  *
- * Here we just change the eflags value on the stack: we allow
+ * Here we just change the flags value on the stack: we allow
  * only the super-user to do it. This depends on the stack-layout
  * on system-call entry - see also fork() and the signal handling
  * code.
@@ -109,8 +109,8 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
 asmlinkage long sys_iopl(unsigned long regsp)
 {
 	volatile struct pt_regs *regs = (struct pt_regs *)&regsp;
-	unsigned int level = regs->ebx;
-	unsigned int old = (regs->eflags >> 12) & 3;
+	unsigned int level = regs->bx;
+	unsigned int old = (regs->flags >> 12) & 3;
 	struct thread_struct *t = &current->thread;
 
 	if (level > 3)
@@ -122,7 +122,7 @@ asmlinkage long sys_iopl(unsigned long regsp)
 	}
 
 	t->iopl = level << 12;
-	regs->eflags = (regs->eflags & ~X86_EFLAGS_IOPL) | t->iopl;
+	regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) | t->iopl;
 	set_iopl_mask(t->iopl);
 
 	return 0;
diff --git a/arch/x86/kernel/ioport_64.c b/arch/x86/kernel/ioport_64.c
index 5f62fad64dab..ff7514b757e5 100644
--- a/arch/x86/kernel/ioport_64.c
+++ b/arch/x86/kernel/ioport_64.c
@@ -95,7 +95,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
  * beyond the 0x3ff range: to get the full 65536 ports bitmapped
  * you'd need 8kB of bitmaps/process, which is a bit excessive.
  *
- * Here we just change the eflags value on the stack: we allow
+ * Here we just change the flags value on the stack: we allow
  * only the super-user to do it. This depends on the stack-layout
  * on system-call entry - see also fork() and the signal handling
  * code.
@@ -103,7 +103,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
 
 asmlinkage long sys_iopl(unsigned int level, struct pt_regs *regs)
 {
-	unsigned int old = (regs->eflags >> 12) & 3;
+	unsigned int old = (regs->flags >> 12) & 3;
 
 	if (level > 3)
 		return -EINVAL;
@@ -112,6 +112,6 @@ asmlinkage long sys_iopl(unsigned int level, struct pt_regs *regs)
 		if (!capable(CAP_SYS_RAWIO))
 			return -EPERM;
 	}
-	regs->eflags = (regs->eflags &~ X86_EFLAGS_IOPL) | (level << 12);
+	regs->flags = (regs->flags &~ X86_EFLAGS_IOPL) | (level << 12);
 	return 0;
 }
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index d3fde94f7345..b49616bcc16b 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -70,7 +70,7 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs)
 {	
 	struct pt_regs *old_regs;
 	/* high bit used in ret_from_ code */
-	int irq = ~regs->orig_eax;
+	int irq = ~regs->orig_ax;
 	struct irq_desc *desc = irq_desc + irq;
 #ifdef CONFIG_4KSTACKS
 	union irq_ctx *curctx, *irqctx;
@@ -88,13 +88,13 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs)
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 	/* Debugging check for stack overflow: is there less than 1KB free? */
 	{
-		long esp;
+		long sp;
 
 		__asm__ __volatile__("andl %%esp,%0" :
-					"=r" (esp) : "0" (THREAD_SIZE - 1));
-		if (unlikely(esp < (sizeof(struct thread_info) + STACK_WARN))) {
+					"=r" (sp) : "0" (THREAD_SIZE - 1));
+		if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
 			printk("do_IRQ: stack overflow: %ld\n",
-				esp - sizeof(struct thread_info));
+				sp - sizeof(struct thread_info));
 			dump_stack();
 		}
 	}
@@ -112,7 +112,7 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs)
 	 * current stack (which is the irq stack already after all)
 	 */
 	if (curctx != irqctx) {
-		int arg1, arg2, ebx;
+		int arg1, arg2, bx;
 
 		/* build the stack frame on the IRQ stack */
 		isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
@@ -128,10 +128,10 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs)
 			(curctx->tinfo.preempt_count & SOFTIRQ_MASK);
 
 		asm volatile(
-			"       xchgl  %%ebx,%%esp      \n"
-			"       call   *%%edi           \n"
-			"       movl   %%ebx,%%esp      \n"
-			: "=a" (arg1), "=d" (arg2), "=b" (ebx)
+			"       xchgl  %%ebx,%%esp    \n"
+			"       call   *%%edi         \n"
+			"       movl   %%ebx,%%esp    \n"
+			: "=a" (arg1), "=d" (arg2), "=b" (bx)
 			:  "0" (irq),   "1" (desc),  "2" (isp),
 			   "D" (desc->handle_irq)
 			: "memory", "cc"
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index 6c3a3b6e5cf4..3aac15466a91 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -53,11 +53,11 @@ static inline void stack_overflow_check(struct pt_regs *regs)
 	u64 curbase = (u64)task_stack_page(current);
 	static unsigned long warned = -60*HZ;
 
-	if (regs->rsp >= curbase && regs->rsp <= curbase + THREAD_SIZE &&
-	    regs->rsp <  curbase + sizeof(struct thread_info) + 128 &&
+	if (regs->sp >= curbase && regs->sp <= curbase + THREAD_SIZE &&
+	    regs->sp <  curbase + sizeof(struct thread_info) + 128 &&
 	    time_after(jiffies, warned + 60*HZ)) {
-		printk("do_IRQ: %s near stack overflow (cur:%Lx,rsp:%lx)\n",
-		       current->comm, curbase, regs->rsp);
+		printk("do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n",
+		       current->comm, curbase, regs->sp);
 		show_stack(NULL,NULL);
 		warned = jiffies;
 	}
@@ -162,7 +162,7 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
 	/* high bit used in ret_from_ code  */
-	unsigned vector = ~regs->orig_rax;
+	unsigned vector = ~regs->orig_ax;
 	unsigned irq;
 
 	exit_idle();
diff --git a/arch/x86/kernel/kprobes_32.c b/arch/x86/kernel/kprobes_32.c
index bc4a68367cd0..d708cd4f956f 100644
--- a/arch/x86/kernel/kprobes_32.c
+++ b/arch/x86/kernel/kprobes_32.c
@@ -212,7 +212,7 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 {
 	__get_cpu_var(current_kprobe) = p;
 	kcb->kprobe_saved_eflags = kcb->kprobe_old_eflags
-		= (regs->eflags & (TF_MASK | IF_MASK));
+		= (regs->flags & (TF_MASK | IF_MASK));
 	if (is_IF_modifier(p->opcode))
 		kcb->kprobe_saved_eflags &= ~IF_MASK;
 }
@@ -232,20 +232,20 @@ static __always_inline void restore_btf(void)
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
 	clear_btf();
-	regs->eflags |= TF_MASK;
-	regs->eflags &= ~IF_MASK;
+	regs->flags |= TF_MASK;
+	regs->flags &= ~IF_MASK;
 	/*single step inline if the instruction is an int3*/
 	if (p->opcode == BREAKPOINT_INSTRUCTION)
-		regs->eip = (unsigned long)p->addr;
+		regs->ip = (unsigned long)p->addr;
 	else
-		regs->eip = (unsigned long)p->ainsn.insn;
+		regs->ip = (unsigned long)p->ainsn.insn;
 }
 
 /* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
-	unsigned long *sara = (unsigned long *)&regs->esp;
+	unsigned long *sara = (unsigned long *)&regs->sp;
 
 	ri->ret_addr = (kprobe_opcode_t *) *sara;
 
@@ -264,7 +264,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 	kprobe_opcode_t *addr;
 	struct kprobe_ctlblk *kcb;
 
-	addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t));
+	addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
 
 	/*
 	 * We don't want to be preempted for the entire
@@ -279,8 +279,8 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 		if (p) {
 			if (kcb->kprobe_status == KPROBE_HIT_SS &&
 				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
-				regs->eflags &= ~TF_MASK;
-				regs->eflags |= kcb->kprobe_saved_eflags;
+				regs->flags &= ~TF_MASK;
+				regs->flags |= kcb->kprobe_saved_eflags;
 				goto no_kprobe;
 			}
 			/* We have reentered the kprobe_handler(), since
@@ -301,7 +301,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * another cpu right after we hit, no further
 			 * handling of this interrupt is appropriate
 			 */
-				regs->eip -= sizeof(kprobe_opcode_t);
+				regs->ip -= sizeof(kprobe_opcode_t);
 				ret = 1;
 				goto no_kprobe;
 			}
@@ -325,7 +325,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * Back up over the (now missing) int3 and run
 			 * the original instruction.
 			 */
-			regs->eip -= sizeof(kprobe_opcode_t);
+			regs->ip -= sizeof(kprobe_opcode_t);
 			ret = 1;
 		}
 		/* Not one of ours: let kernel handle it */
@@ -344,7 +344,7 @@ ss_probe:
 	if (p->ainsn.boostable == 1 && !p->post_handler){
 		/* Boost up -- we can execute copied instructions directly */
 		reset_current_kprobe();
-		regs->eip = (unsigned long)p->ainsn.insn;
+		regs->ip = (unsigned long)p->ainsn.insn;
 		preempt_enable_no_resched();
 		return 1;
 	}
@@ -368,7 +368,7 @@ no_kprobe:
 	asm volatile ( ".global kretprobe_trampoline\n"
 			"kretprobe_trampoline: \n"
 			"	pushf\n"
-			/* skip cs, eip, orig_eax */
+			/* skip cs, ip, orig_ax */
 			"	subl $12, %esp\n"
 			"	pushl %fs\n"
 			"	pushl %ds\n"
@@ -382,10 +382,10 @@ no_kprobe:
 			"	pushl %ebx\n"
 			"	movl %esp, %eax\n"
 			"	call trampoline_handler\n"
-			/* move eflags to cs */
+			/* move flags to cs */
 			"	movl 52(%esp), %edx\n"
 			"	movl %edx, 48(%esp)\n"
-			/* save true return address on eflags */
+			/* save true return address on flags */
 			"	movl %eax, 52(%esp)\n"
 			"	popl %ebx\n"
 			"	popl %ecx\n"
@@ -394,7 +394,7 @@ no_kprobe:
 			"	popl %edi\n"
 			"	popl %ebp\n"
 			"	popl %eax\n"
-			/* skip eip, orig_eax, es, ds, fs */
+			/* skip ip, orig_ax, es, ds, fs */
 			"	addl $20, %esp\n"
 			"	popf\n"
 			"	ret\n");
@@ -415,9 +415,9 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs)
 	spin_lock_irqsave(&kretprobe_lock, flags);
 	head = kretprobe_inst_table_head(current);
 	/* fixup registers */
-	regs->xcs = __KERNEL_CS | get_kernel_rpl();
-	regs->eip = trampoline_address;
-	regs->orig_eax = 0xffffffff;
+	regs->cs = __KERNEL_CS | get_kernel_rpl();
+	regs->ip = trampoline_address;
+	regs->orig_ax = 0xffffffff;
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -478,11 +478,11 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs)
  * interrupt.  We have to fix up the stack as follows:
  *
  * 0) Except in the case of absolute or indirect jump or call instructions,
- * the new eip is relative to the copied instruction.  We need to make
+ * the new ip is relative to the copied instruction.  We need to make
  * it relative to the original instruction.
  *
  * 1) If the single-stepped instruction was pushfl, then the TF and IF
- * flags are set in the just-pushed eflags, and may need to be cleared.
+ * flags are set in the just-pushed flags, and may need to be cleared.
  *
  * 2) If the single-stepped instruction was a call, the return address
  * that is atop the stack is the address following the copied instruction.
@@ -493,11 +493,11 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs)
 static void __kprobes resume_execution(struct kprobe *p,
 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
 {
-	unsigned long *tos = (unsigned long *)&regs->esp;
+	unsigned long *tos = (unsigned long *)&regs->sp;
 	unsigned long copy_eip = (unsigned long)p->ainsn.insn;
 	unsigned long orig_eip = (unsigned long)p->addr;
 
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	switch (p->ainsn.insn[0]) {
 	case 0x9c:		/* pushfl */
 		*tos &= ~(TF_MASK | IF_MASK);
@@ -508,8 +508,8 @@ static void __kprobes resume_execution(struct kprobe *p,
 	case 0xca:
 	case 0xcb:
 	case 0xcf:
-	case 0xea:		/* jmp absolute -- eip is correct */
-		/* eip is already adjusted, no more changes required */
+	case 0xea:		/* jmp absolute -- ip is correct */
+		/* ip is already adjusted, no more changes required */
 		p->ainsn.boostable = 1;
 		goto no_change;
 	case 0xe8:		/* call relative - Fix return addr */
@@ -522,14 +522,14 @@ static void __kprobes resume_execution(struct kprobe *p,
 		if ((p->ainsn.insn[1] & 0x30) == 0x10) {
 			/*
 			 * call absolute, indirect
-			 * Fix return addr; eip is correct.
+			 * Fix return addr; ip is correct.
 			 * But this is not boostable
 			 */
 			*tos = orig_eip + (*tos - copy_eip);
 			goto no_change;
 		} else if (((p->ainsn.insn[1] & 0x31) == 0x20) ||	/* jmp near, absolute indirect */
 			   ((p->ainsn.insn[1] & 0x31) == 0x21)) {	/* jmp far, absolute indirect */
-			/* eip is correct. And this is boostable */
+			/* ip is correct. And this is boostable */
 			p->ainsn.boostable = 1;
 			goto no_change;
 		}
@@ -538,21 +538,21 @@ static void __kprobes resume_execution(struct kprobe *p,
 	}
 
 	if (p->ainsn.boostable == 0) {
-		if ((regs->eip > copy_eip) &&
-		    (regs->eip - copy_eip) + 5 < MAX_INSN_SIZE) {
+		if ((regs->ip > copy_eip) &&
+		    (regs->ip - copy_eip) + 5 < MAX_INSN_SIZE) {
 			/*
 			 * These instructions can be executed directly if it
 			 * jumps back to correct address.
 			 */
-			set_jmp_op((void *)regs->eip,
-				   (void *)orig_eip + (regs->eip - copy_eip));
+			set_jmp_op((void *)regs->ip,
+				   (void *)orig_eip + (regs->ip - copy_eip));
 			p->ainsn.boostable = 1;
 		} else {
 			p->ainsn.boostable = -1;
 		}
 	}
 
-	regs->eip = orig_eip + (regs->eip - copy_eip);
+	regs->ip = orig_eip + (regs->ip - copy_eip);
 
 no_change:
 	restore_btf();
@@ -578,8 +578,8 @@ static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 	}
 
 	resume_execution(cur, regs, kcb);
-	regs->eflags |= kcb->kprobe_saved_eflags;
-	trace_hardirqs_fixup_flags(regs->eflags);
+	regs->flags |= kcb->kprobe_saved_eflags;
+	trace_hardirqs_fixup_flags(regs->flags);
 
 	/*Restore back the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
@@ -591,11 +591,11 @@ out:
 	preempt_enable_no_resched();
 
 	/*
-	 * if somebody else is singlestepping across a probe point, eflags
+	 * if somebody else is singlestepping across a probe point, flags
 	 * will have TF set, in which case, continue the remaining processing
 	 * of do_debug, as if this is not a probe hit.
 	 */
-	if (regs->eflags & TF_MASK)
+	if (regs->flags & TF_MASK)
 		return 0;
 
 	return 1;
@@ -612,12 +612,12 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		/*
 		 * We are here because the instruction being single
 		 * stepped caused a page fault. We reset the current
-		 * kprobe and the eip points back to the probe address
+		 * kprobe and the ip points back to the probe address
 		 * and allow the page fault handler to continue as a
 		 * normal page fault.
 		 */
-		regs->eip = (unsigned long)cur->addr;
-		regs->eflags |= kcb->kprobe_old_eflags;
+		regs->ip = (unsigned long)cur->addr;
+		regs->flags |= kcb->kprobe_old_eflags;
 		if (kcb->kprobe_status == KPROBE_REENTER)
 			restore_previous_kprobe(kcb);
 		else
@@ -703,7 +703,7 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 
 	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_esp = &regs->esp;
+	kcb->jprobe_saved_esp = &regs->sp;
 	addr = (unsigned long)(kcb->jprobe_saved_esp);
 
 	/*
@@ -715,9 +715,9 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	 */
 	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
 			MIN_STACK_SIZE(addr));
-	regs->eflags &= ~IF_MASK;
+	regs->flags &= ~IF_MASK;
 	trace_hardirqs_off();
-	regs->eip = (unsigned long)(jp->entry);
+	regs->ip = (unsigned long)(jp->entry);
 	return 1;
 }
 
@@ -736,15 +736,15 @@ void __kprobes jprobe_return(void)
 int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	u8 *addr = (u8 *) (regs->eip - 1);
+	u8 *addr = (u8 *) (regs->ip - 1);
 	unsigned long stack_addr = (unsigned long)(kcb->jprobe_saved_esp);
 	struct jprobe *jp = container_of(p, struct jprobe, kp);
 
 	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if (&regs->esp != kcb->jprobe_saved_esp) {
+		if (&regs->sp != kcb->jprobe_saved_esp) {
 			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
-			printk("current esp %p does not match saved esp %p\n",
-			       &regs->esp, kcb->jprobe_saved_esp);
+			printk("current sp %p does not match saved sp %p\n",
+			       &regs->sp, kcb->jprobe_saved_esp);
 			printk("Saved registers for jprobe %p\n", jp);
 			show_registers(saved_regs);
 			printk("Current registers\n");
diff --git a/arch/x86/kernel/kprobes_64.c b/arch/x86/kernel/kprobes_64.c
index 10d66e323c7d..f6837cd3bed5 100644
--- a/arch/x86/kernel/kprobes_64.c
+++ b/arch/x86/kernel/kprobes_64.c
@@ -251,7 +251,7 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 {
 	__get_cpu_var(current_kprobe) = p;
 	kcb->kprobe_saved_rflags = kcb->kprobe_old_rflags
-		= (regs->eflags & (TF_MASK | IF_MASK));
+		= (regs->flags & (TF_MASK | IF_MASK));
 	if (is_IF_modifier(p->ainsn.insn))
 		kcb->kprobe_saved_rflags &= ~IF_MASK;
 }
@@ -271,20 +271,20 @@ static __always_inline void restore_btf(void)
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
 	clear_btf();
-	regs->eflags |= TF_MASK;
-	regs->eflags &= ~IF_MASK;
+	regs->flags |= TF_MASK;
+	regs->flags &= ~IF_MASK;
 	/*single step inline if the instruction is an int3*/
 	if (p->opcode == BREAKPOINT_INSTRUCTION)
-		regs->rip = (unsigned long)p->addr;
+		regs->ip = (unsigned long)p->addr;
 	else
-		regs->rip = (unsigned long)p->ainsn.insn;
+		regs->ip = (unsigned long)p->ainsn.insn;
 }
 
 /* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
-	unsigned long *sara = (unsigned long *)regs->rsp;
+	unsigned long *sara = (unsigned long *)regs->sp;
 
 	ri->ret_addr = (kprobe_opcode_t *) *sara;
 	/* Replace the return addr with trampoline addr */
@@ -295,7 +295,7 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
 	int ret = 0;
-	kprobe_opcode_t *addr = (kprobe_opcode_t *)(regs->rip - sizeof(kprobe_opcode_t));
+	kprobe_opcode_t *addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
 	struct kprobe_ctlblk *kcb;
 
 	/*
@@ -311,8 +311,8 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 		if (p) {
 			if (kcb->kprobe_status == KPROBE_HIT_SS &&
 				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
-				regs->eflags &= ~TF_MASK;
-				regs->eflags |= kcb->kprobe_saved_rflags;
+				regs->flags &= ~TF_MASK;
+				regs->flags |= kcb->kprobe_saved_rflags;
 				goto no_kprobe;
 			} else if (kcb->kprobe_status == KPROBE_HIT_SSDONE) {
 				/* TODO: Provide re-entrancy from
@@ -321,7 +321,7 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 				 * the instruction of the new probe.
 				 */
 				arch_disarm_kprobe(p);
-				regs->rip = (unsigned long)p->addr;
+				regs->ip = (unsigned long)p->addr;
 				reset_current_kprobe();
 				ret = 1;
 			} else {
@@ -345,7 +345,7 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * another cpu right after we hit, no further
 			 * handling of this interrupt is appropriate
 			 */
-				regs->rip = (unsigned long)addr;
+				regs->ip = (unsigned long)addr;
 				ret = 1;
 				goto no_kprobe;
 			}
@@ -369,7 +369,7 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * Back up over the (now missing) int3 and run
 			 * the original instruction.
 			 */
-			regs->rip = (unsigned long)addr;
+			regs->ip = (unsigned long)addr;
 			ret = 1;
 		}
 		/* Not one of ours: let kernel handle it */
@@ -454,7 +454,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 
 	kretprobe_assert(ri, orig_ret_address, trampoline_address);
-	regs->rip = orig_ret_address;
+	regs->ip = orig_ret_address;
 
 	reset_current_kprobe();
 	spin_unlock_irqrestore(&kretprobe_lock, flags);
@@ -484,11 +484,11 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
  * interrupt.  We have to fix up the stack as follows:
  *
  * 0) Except in the case of absolute or indirect jump or call instructions,
- * the new rip is relative to the copied instruction.  We need to make
+ * the new ip is relative to the copied instruction.  We need to make
  * it relative to the original instruction.
  *
  * 1) If the single-stepped instruction was pushfl, then the TF and IF
- * flags are set in the just-pushed eflags, and may need to be cleared.
+ * flags are set in the just-pushed flags, and may need to be cleared.
  *
  * 2) If the single-stepped instruction was a call, the return address
  * that is atop the stack is the address following the copied instruction.
@@ -497,7 +497,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 static void __kprobes resume_execution(struct kprobe *p,
 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
 {
-	unsigned long *tos = (unsigned long *)regs->rsp;
+	unsigned long *tos = (unsigned long *)regs->sp;
 	unsigned long copy_rip = (unsigned long)p->ainsn.insn;
 	unsigned long orig_rip = (unsigned long)p->addr;
 	kprobe_opcode_t *insn = p->ainsn.insn;
@@ -506,7 +506,7 @@ static void __kprobes resume_execution(struct kprobe *p,
 	if (*insn >= 0x40 && *insn <= 0x4f)
 		insn++;
 
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	switch (*insn) {
 	case 0x9c:	/* pushfl */
 		*tos &= ~(TF_MASK | IF_MASK);
@@ -538,7 +538,8 @@ static void __kprobes resume_execution(struct kprobe *p,
 		break;
 	}
 
-	regs->rip = orig_rip + (regs->rip - copy_rip);
+	regs->ip = orig_rip + (regs->ip - copy_rip);
+
 no_change:
 	restore_btf();
 
@@ -559,8 +560,8 @@ int __kprobes post_kprobe_handler(struct pt_regs *regs)
 	}
 
 	resume_execution(cur, regs, kcb);
-	regs->eflags |= kcb->kprobe_saved_rflags;
-	trace_hardirqs_fixup_flags(regs->eflags);
+	regs->flags |= kcb->kprobe_saved_rflags;
+	trace_hardirqs_fixup_flags(regs->flags);
 
 	/* Restore the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
@@ -572,11 +573,11 @@ out:
 	preempt_enable_no_resched();
 
 	/*
-	 * if somebody else is singlestepping across a probe point, eflags
+	 * if somebody else is singlestepping across a probe point, flags
 	 * will have TF set, in which case, continue the remaining processing
 	 * of do_debug, as if this is not a probe hit.
 	 */
-	if (regs->eflags & TF_MASK)
+	if (regs->flags & TF_MASK)
 		return 0;
 
 	return 1;
@@ -594,12 +595,12 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		/*
 		 * We are here because the instruction being single
 		 * stepped caused a page fault. We reset the current
-		 * kprobe and the rip points back to the probe address
+		 * kprobe and the ip points back to the probe address
 		 * and allow the page fault handler to continue as a
 		 * normal page fault.
 		 */
-		regs->rip = (unsigned long)cur->addr;
-		regs->eflags |= kcb->kprobe_old_rflags;
+		regs->ip = (unsigned long)cur->addr;
+		regs->flags |= kcb->kprobe_old_rflags;
 		if (kcb->kprobe_status == KPROBE_REENTER)
 			restore_previous_kprobe(kcb);
 		else
@@ -629,9 +630,9 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		 * In case the user-specified fault handler returned
 		 * zero, try to fix up.
 		 */
-		fixup = search_exception_tables(regs->rip);
+		fixup = search_exception_tables(regs->ip);
 		if (fixup) {
-			regs->rip = fixup->fixup;
+			regs->ip = fixup->fixup;
 			return 1;
 		}
 
@@ -688,7 +689,7 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 
 	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_rsp = (long *) regs->rsp;
+	kcb->jprobe_saved_rsp = (long *) regs->sp;
 	addr = (unsigned long)(kcb->jprobe_saved_rsp);
 	/*
 	 * As Linus pointed out, gcc assumes that the callee
@@ -699,9 +700,9 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	 */
 	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
 			MIN_STACK_SIZE(addr));
-	regs->eflags &= ~IF_MASK;
+	regs->flags &= ~IF_MASK;
 	trace_hardirqs_off();
-	regs->rip = (unsigned long)(jp->entry);
+	regs->ip = (unsigned long)(jp->entry);
 	return 1;
 }
 
@@ -720,15 +721,15 @@ void __kprobes jprobe_return(void)
 int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	u8 *addr = (u8 *) (regs->rip - 1);
+	u8 *addr = (u8 *) (regs->ip - 1);
 	unsigned long stack_addr = (unsigned long)(kcb->jprobe_saved_rsp);
 	struct jprobe *jp = container_of(p, struct jprobe, kp);
 
 	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if ((unsigned long *)regs->rsp != kcb->jprobe_saved_rsp) {
+		if ((unsigned long *)regs->sp != kcb->jprobe_saved_rsp) {
 			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
-			printk("current rsp %p does not match saved rsp %p\n",
-			       (long *)regs->rsp, kcb->jprobe_saved_rsp);
+			printk("current sp %p does not match saved sp %p\n",
+			       (long *)regs->sp, kcb->jprobe_saved_rsp);
 			printk("Saved registers for jprobe %p\n", jp);
 			show_registers(saved_regs);
 			printk("Current registers\n");
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index d5462f228daf..c9f28e02e86d 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -265,13 +265,13 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait);
  * New with Core Duo processors, MWAIT can take some hints based on CPU
  * capability.
  */
-void mwait_idle_with_hints(unsigned long eax, unsigned long ecx)
+void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 {
 	if (!need_resched()) {
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
 		smp_mb();
 		if (!need_resched())
-			__mwait(eax, ecx);
+			__mwait(ax, cx);
 	}
 }
 
@@ -320,15 +320,15 @@ void __show_registers(struct pt_regs *regs, int all)
 {
 	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
 	unsigned long d0, d1, d2, d3, d6, d7;
-	unsigned long esp;
+	unsigned long sp;
 	unsigned short ss, gs;
 
 	if (user_mode_vm(regs)) {
-		esp = regs->esp;
-		ss = regs->xss & 0xffff;
+		sp = regs->sp;
+		ss = regs->ss & 0xffff;
 		savesegment(gs, gs);
 	} else {
-		esp = (unsigned long) (&regs->esp);
+		sp = (unsigned long) (&regs->sp);
 		savesegment(ss, ss);
 		savesegment(gs, gs);
 	}
@@ -341,17 +341,17 @@ void __show_registers(struct pt_regs *regs, int all)
 			init_utsname()->version);
 
 	printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
-			0xffff & regs->xcs, regs->eip, regs->eflags,
+			0xffff & regs->cs, regs->ip, regs->flags,
 			smp_processor_id());
-	print_symbol("EIP is at %s\n", regs->eip);
+	print_symbol("EIP is at %s\n", regs->ip);
 
 	printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
-		regs->eax, regs->ebx, regs->ecx, regs->edx);
+		regs->ax, regs->bx, regs->cx, regs->dx);
 	printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n",
-		regs->esi, regs->edi, regs->ebp, esp);
+		regs->si, regs->di, regs->bp, sp);
 	printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n",
-	       regs->xds & 0xffff, regs->xes & 0xffff,
-	       regs->xfs & 0xffff, gs, ss);
+	       regs->ds & 0xffff, regs->es & 0xffff,
+	       regs->fs & 0xffff, gs, ss);
 
 	if (!all)
 		return;
@@ -379,12 +379,12 @@ void __show_registers(struct pt_regs *regs, int all)
 void show_regs(struct pt_regs *regs)
 {
 	__show_registers(regs, 1);
-	show_trace(NULL, regs, &regs->esp);
+	show_trace(NULL, regs, &regs->sp);
 }
 
 /*
- * This gets run with %ebx containing the
- * function to call, and %edx containing
+ * This gets run with %bx containing the
+ * function to call, and %dx containing
  * the "args".
  */
 extern void kernel_thread_helper(void);
@@ -398,16 +398,16 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 
 	memset(&regs, 0, sizeof(regs));
 
-	regs.ebx = (unsigned long) fn;
-	regs.edx = (unsigned long) arg;
+	regs.bx = (unsigned long) fn;
+	regs.dx = (unsigned long) arg;
 
-	regs.xds = __USER_DS;
-	regs.xes = __USER_DS;
-	regs.xfs = __KERNEL_PERCPU;
-	regs.orig_eax = -1;
-	regs.eip = (unsigned long) kernel_thread_helper;
-	regs.xcs = __KERNEL_CS | get_kernel_rpl();
-	regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
+	regs.ds = __USER_DS;
+	regs.es = __USER_DS;
+	regs.fs = __KERNEL_PERCPU;
+	regs.orig_ax = -1;
+	regs.ip = (unsigned long) kernel_thread_helper;
+	regs.cs = __KERNEL_CS | get_kernel_rpl();
+	regs.flags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
 
 	/* Ok, create the new process.. */
 	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
@@ -470,7 +470,7 @@ void prepare_to_copy(struct task_struct *tsk)
 	unlazy_fpu(tsk);
 }
 
-int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
+int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 	unsigned long unused,
 	struct task_struct * p, struct pt_regs * regs)
 {
@@ -480,8 +480,8 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
 
 	childregs = task_pt_regs(p);
 	*childregs = *regs;
-	childregs->eax = 0;
-	childregs->esp = esp;
+	childregs->ax = 0;
+	childregs->sp = sp;
 
 	p->thread.esp = (unsigned long) childregs;
 	p->thread.esp0 = (unsigned long) (childregs+1);
@@ -508,7 +508,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
 	 */
 	if (clone_flags & CLONE_SETTLS)
 		err = do_set_thread_area(p, -1,
-			(struct user_desc __user *)childregs->esi, 0);
+			(struct user_desc __user *)childregs->si, 0);
 
 	if (err && p->thread.io_bitmap_ptr) {
 		kfree(p->thread.io_bitmap_ptr);
@@ -527,7 +527,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 /* changed the size calculations - should hopefully work better. lbt */
 	dump->magic = CMAGIC;
 	dump->start_code = 0;
-	dump->start_stack = regs->esp & ~(PAGE_SIZE - 1);
+	dump->start_stack = regs->sp & ~(PAGE_SIZE - 1);
 	dump->u_tsize = ((unsigned long) current->mm->end_code) >> PAGE_SHIFT;
 	dump->u_dsize = ((unsigned long) (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
 	dump->u_dsize -= dump->u_tsize;
@@ -538,23 +538,23 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 	if (dump->start_stack < TASK_SIZE)
 		dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
 
-	dump->regs.ebx = regs->ebx;
-	dump->regs.ecx = regs->ecx;
-	dump->regs.edx = regs->edx;
-	dump->regs.esi = regs->esi;
-	dump->regs.edi = regs->edi;
-	dump->regs.ebp = regs->ebp;
-	dump->regs.eax = regs->eax;
-	dump->regs.ds = regs->xds;
-	dump->regs.es = regs->xes;
-	dump->regs.fs = regs->xfs;
+	dump->regs.ebx = regs->bx;
+	dump->regs.ecx = regs->cx;
+	dump->regs.edx = regs->dx;
+	dump->regs.esi = regs->si;
+	dump->regs.edi = regs->di;
+	dump->regs.ebp = regs->bp;
+	dump->regs.eax = regs->ax;
+	dump->regs.ds = regs->ds;
+	dump->regs.es = regs->es;
+	dump->regs.fs = regs->fs;
 	savesegment(gs,dump->regs.gs);
-	dump->regs.orig_eax = regs->orig_eax;
-	dump->regs.eip = regs->eip;
-	dump->regs.cs = regs->xcs;
-	dump->regs.eflags = regs->eflags;
-	dump->regs.esp = regs->esp;
-	dump->regs.ss = regs->xss;
+	dump->regs.orig_eax = regs->orig_ax;
+	dump->regs.eip = regs->ip;
+	dump->regs.cs = regs->cs;
+	dump->regs.eflags = regs->flags;
+	dump->regs.esp = regs->sp;
+	dump->regs.ss = regs->ss;
 
 	dump->u_fpvalid = dump_fpu (regs, &dump->i387);
 }
@@ -566,10 +566,10 @@ EXPORT_SYMBOL(dump_thread);
 int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
 {
 	struct pt_regs ptregs = *task_pt_regs(tsk);
-	ptregs.xcs &= 0xffff;
-	ptregs.xds &= 0xffff;
-	ptregs.xes &= 0xffff;
-	ptregs.xss &= 0xffff;
+	ptregs.cs &= 0xffff;
+	ptregs.ds &= 0xffff;
+	ptregs.es &= 0xffff;
+	ptregs.ss &= 0xffff;
 
 	elf_core_copy_regs(regs, &ptregs);
 
@@ -684,7 +684,7 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
  * More important, however, is the fact that this allows us much
  * more flexibility.
  *
- * The return value (in %eax) will be the "prev" task after
+ * The return value (in %ax) will be the "prev" task after
  * the task-switch, and shows up in ret_from_fork in entry.S,
  * for example.
  */
@@ -771,7 +771,7 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 
 asmlinkage int sys_fork(struct pt_regs regs)
 {
-	return do_fork(SIGCHLD, regs.esp, &regs, 0, NULL, NULL);
+	return do_fork(SIGCHLD, regs.sp, &regs, 0, NULL, NULL);
 }
 
 asmlinkage int sys_clone(struct pt_regs regs)
@@ -780,12 +780,12 @@ asmlinkage int sys_clone(struct pt_regs regs)
 	unsigned long newsp;
 	int __user *parent_tidptr, *child_tidptr;
 
-	clone_flags = regs.ebx;
-	newsp = regs.ecx;
-	parent_tidptr = (int __user *)regs.edx;
-	child_tidptr = (int __user *)regs.edi;
+	clone_flags = regs.bx;
+	newsp = regs.cx;
+	parent_tidptr = (int __user *)regs.dx;
+	child_tidptr = (int __user *)regs.di;
 	if (!newsp)
-		newsp = regs.esp;
+		newsp = regs.sp;
 	return do_fork(clone_flags, newsp, &regs, 0, parent_tidptr, child_tidptr);
 }
 
@@ -801,7 +801,7 @@ asmlinkage int sys_clone(struct pt_regs regs)
  */
 asmlinkage int sys_vfork(struct pt_regs regs)
 {
-	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.esp, &regs, 0, NULL, NULL);
+	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.sp, &regs, 0, NULL, NULL);
 }
 
 /*
@@ -812,13 +812,13 @@ asmlinkage int sys_execve(struct pt_regs regs)
 	int error;
 	char * filename;
 
-	filename = getname((char __user *) regs.ebx);
+	filename = getname((char __user *) regs.bx);
 	error = PTR_ERR(filename);
 	if (IS_ERR(filename))
 		goto out;
 	error = do_execve(filename,
-			(char __user * __user *) regs.ecx,
-			(char __user * __user *) regs.edx,
+			(char __user * __user *) regs.cx,
+			(char __user * __user *) regs.dx,
 			&regs);
 	if (error == 0) {
 		/* Make sure we don't return using sysenter.. */
@@ -834,24 +834,24 @@ out:
 
 unsigned long get_wchan(struct task_struct *p)
 {
-	unsigned long ebp, esp, eip;
+	unsigned long bp, sp, ip;
 	unsigned long stack_page;
 	int count = 0;
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 	stack_page = (unsigned long)task_stack_page(p);
-	esp = p->thread.esp;
-	if (!stack_page || esp < stack_page || esp > top_esp+stack_page)
+	sp = p->thread.esp;
+	if (!stack_page || sp < stack_page || sp > top_esp+stack_page)
 		return 0;
-	/* include/asm-i386/system.h:switch_to() pushes ebp last. */
-	ebp = *(unsigned long *) esp;
+	/* include/asm-i386/system.h:switch_to() pushes bp last. */
+	bp = *(unsigned long *) sp;
 	do {
-		if (ebp < stack_page || ebp > top_ebp+stack_page)
+		if (bp < stack_page || bp > top_ebp+stack_page)
 			return 0;
-		eip = *(unsigned long *) (ebp+4);
-		if (!in_sched_functions(eip))
-			return eip;
-		ebp = *(unsigned long *) ebp;
+		ip = *(unsigned long *) (bp+4);
+		if (!in_sched_functions(ip))
+			return ip;
+		bp = *(unsigned long *) bp;
 	} while (count++ < 16);
 	return 0;
 }
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ae5eca17aa3c..efbb1a2eab97 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -257,13 +257,13 @@ void cpu_idle(void)
  * New with Core Duo processors, MWAIT can take some hints based on CPU
  * capability.
  */
-void mwait_idle_with_hints(unsigned long eax, unsigned long ecx)
+void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 {
 	if (!need_resched()) {
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
 		smp_mb();
 		if (!need_resched())
-			__mwait(eax, ecx);
+			__mwait(ax, cx);
 	}
 }
 
@@ -330,16 +330,16 @@ void __show_regs(struct pt_regs * regs)
 		init_utsname()->release,
 		(int)strcspn(init_utsname()->version, " "),
 		init_utsname()->version);
-	printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->rip);
-	printk_address(regs->rip); 
-	printk("RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss, regs->rsp,
-		regs->eflags);
+	printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip);
+	printk_address(regs->ip);
+	printk("RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss, regs->sp,
+		regs->flags);
 	printk("RAX: %016lx RBX: %016lx RCX: %016lx\n",
-	       regs->rax, regs->rbx, regs->rcx);
+	       regs->ax, regs->bx, regs->cx);
 	printk("RDX: %016lx RSI: %016lx RDI: %016lx\n",
-	       regs->rdx, regs->rsi, regs->rdi); 
+	       regs->dx, regs->si, regs->di);
 	printk("RBP: %016lx R08: %016lx R09: %016lx\n",
-	       regs->rbp, regs->r8, regs->r9); 
+	       regs->bp, regs->r8, regs->r9);
 	printk("R10: %016lx R11: %016lx R12: %016lx\n",
 	       regs->r10, regs->r11, regs->r12); 
 	printk("R13: %016lx R14: %016lx R15: %016lx\n",
@@ -476,7 +476,7 @@ void prepare_to_copy(struct task_struct *tsk)
 	unlazy_fpu(tsk);
 }
 
-int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp, 
+int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 		unsigned long unused,
 	struct task_struct * p, struct pt_regs * regs)
 {
@@ -488,10 +488,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 			(THREAD_SIZE + task_stack_page(p))) - 1;
 	*childregs = *regs;
 
-	childregs->rax = 0;
-	childregs->rsp = rsp;
-	if (rsp == ~0UL)
-		childregs->rsp = (unsigned long)childregs;
+	childregs->ax = 0;
+	childregs->sp = sp;
+	if (sp == ~0UL)
+		childregs->sp = (unsigned long)childregs;
 
 	p->thread.rsp = (unsigned long) childregs;
 	p->thread.rsp0 = (unsigned long) (childregs+1);
@@ -525,7 +525,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 #ifdef CONFIG_IA32_EMULATION
 		if (test_thread_flag(TIF_IA32))
 			err = do_set_thread_area(p, -1,
-				(struct user_desc __user *)childregs->rsi, 0);
+				(struct user_desc __user *)childregs->si, 0);
 		else 			
 #endif	 
 			err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8); 
@@ -732,7 +732,7 @@ void set_personality_64bit(void)
 
 asmlinkage long sys_fork(struct pt_regs *regs)
 {
-	return do_fork(SIGCHLD, regs->rsp, regs, 0, NULL, NULL);
+	return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
 }
 
 asmlinkage long
@@ -740,7 +740,7 @@ sys_clone(unsigned long clone_flags, unsigned long newsp,
 	  void __user *parent_tid, void __user *child_tid, struct pt_regs *regs)
 {
 	if (!newsp)
-		newsp = regs->rsp;
+		newsp = regs->sp;
 	return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
 }
 
@@ -756,14 +756,14 @@ sys_clone(unsigned long clone_flags, unsigned long newsp,
  */
 asmlinkage long sys_vfork(struct pt_regs *regs)
 {
-	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->rsp, regs, 0,
+	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0,
 		    NULL, NULL);
 }
 
 unsigned long get_wchan(struct task_struct *p)
 {
 	unsigned long stack;
-	u64 fp,rip;
+	u64 fp,ip;
 	int count = 0;
 
 	if (!p || p == current || p->state==TASK_RUNNING)
@@ -776,9 +776,9 @@ unsigned long get_wchan(struct task_struct *p)
 		if (fp < (unsigned long)stack ||
 		    fp > (unsigned long)stack+THREAD_SIZE)
 			return 0; 
-		rip = *(u64 *)(fp+8); 
-		if (!in_sched_functions(rip))
-			return rip; 
+		ip = *(u64 *)(fp+8);
+		if (!in_sched_functions(ip))
+			return ip;
 		fp = *(u64 *)fp; 
 	} while (count++ < 16); 
 	return 0;
diff --git a/arch/x86/kernel/ptrace_32.c b/arch/x86/kernel/ptrace_32.c
index 512f8412b799..f81e2f1827d4 100644
--- a/arch/x86/kernel/ptrace_32.c
+++ b/arch/x86/kernel/ptrace_32.c
@@ -39,10 +39,10 @@
 
 static long *pt_regs_access(struct pt_regs *regs, unsigned long regno)
 {
-	BUILD_BUG_ON(offsetof(struct pt_regs, ebx) != 0);
+	BUILD_BUG_ON(offsetof(struct pt_regs, bx) != 0);
 	if (regno > FS)
 		--regno;
-	return &regs->ebx + regno;
+	return &regs->bx + regno;
 }
 
 static int putreg(struct task_struct *child,
@@ -80,7 +80,7 @@ static int putreg(struct task_struct *child,
 				clear_tsk_thread_flag(child, TIF_FORCED_TF);
 			else if (test_tsk_thread_flag(child, TIF_FORCED_TF))
 				value |= X86_EFLAGS_TF;
-			value |= regs->eflags & ~FLAG_MASK;
+			value |= regs->flags & ~FLAG_MASK;
 			break;
 	}
 	*pt_regs_access(regs, regno) = value;
@@ -98,7 +98,7 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
 			/*
 			 * If the debugger set TF, hide it from the readout.
 			 */
-			retval = regs->eflags;
+			retval = regs->flags;
 			if (test_tsk_thread_flag(child, TIF_FORCED_TF))
 				retval &= ~X86_EFLAGS_TF;
 			break;
@@ -369,8 +369,8 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
 	info.si_signo = SIGTRAP;
 	info.si_code = TRAP_BRKPT;
 
-	/* User-mode eip? */
-	info.si_addr = user_mode_vm(regs) ? (void __user *) regs->eip : NULL;
+	/* User-mode ip? */
+	info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL;
 
 	/* Send us the fake SIGTRAP */
 	force_sig_info(SIGTRAP, &info, tsk);
@@ -392,12 +392,12 @@ int do_syscall_trace(struct pt_regs *regs, int entryexit)
 
 	/* do the secure computing check first */
 	if (!entryexit)
-		secure_computing(regs->orig_eax);
+		secure_computing(regs->orig_ax);
 
 	if (unlikely(current->audit_context)) {
 		if (entryexit)
-			audit_syscall_exit(AUDITSC_RESULT(regs->eax),
-						regs->eax);
+			audit_syscall_exit(AUDITSC_RESULT(regs->ax),
+						regs->ax);
 		/* Debug traps, when using PTRACE_SINGLESTEP, must be sent only
 		 * on the syscall exit path. Normally, when TIF_SYSCALL_AUDIT is
 		 * not used, entry.S will call us only on syscall exit, not
@@ -445,13 +445,13 @@ int do_syscall_trace(struct pt_regs *regs, int entryexit)
 	ret = is_sysemu;
 out:
 	if (unlikely(current->audit_context) && !entryexit)
-		audit_syscall_entry(AUDIT_ARCH_I386, regs->orig_eax,
-				    regs->ebx, regs->ecx, regs->edx, regs->esi);
+		audit_syscall_entry(AUDIT_ARCH_I386, regs->orig_ax,
+				    regs->bx, regs->cx, regs->dx, regs->si);
 	if (ret == 0)
 		return 0;
 
-	regs->orig_eax = -1; /* force skip of syscall restarting */
+	regs->orig_ax = -1; /* force skip of syscall restarting */
 	if (unlikely(current->audit_context))
-		audit_syscall_exit(AUDITSC_RESULT(regs->eax), regs->eax);
+		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 	return 1;
 }
diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index 4ba66d8af717..bee20bb1a6c0 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -119,7 +119,7 @@ static int putreg(struct task_struct *child,
 				clear_tsk_thread_flag(child, TIF_FORCED_TF);
 			else if (test_tsk_thread_flag(child, TIF_FORCED_TF))
 				value |= X86_EFLAGS_TF;
-			value |= regs->eflags & ~FLAG_MASK;
+			value |= regs->flags & ~FLAG_MASK;
 			break;
 		case offsetof(struct user_regs_struct,cs): 
 			if ((value & 3) != 3)
@@ -168,7 +168,7 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
 			/*
 			 * If the debugger set TF, hide it from the readout.
 			 */
-			val = regs->eflags;
+			val = regs->flags;
 			if (test_tsk_thread_flag(child, TIF_IA32))
 				val &= 0xffffffff;
 			if (test_tsk_thread_flag(child, TIF_FORCED_TF))
@@ -383,9 +383,9 @@ static void syscall_trace(struct pt_regs *regs)
 {
 
 #if 0
-	printk("trace %s rip %lx rsp %lx rax %d origrax %d caller %lx tiflags %x ptrace %x\n",
+	printk("trace %s ip %lx sp %lx ax %d origrax %d caller %lx tiflags %x ptrace %x\n",
 	       current->comm,
-	       regs->rip, regs->rsp, regs->rax, regs->orig_rax, __builtin_return_address(0),
+	       regs->ip, regs->sp, regs->ax, regs->orig_ax, __builtin_return_address(0),
 	       current_thread_info()->flags, current->ptrace); 
 #endif
 
@@ -405,7 +405,7 @@ static void syscall_trace(struct pt_regs *regs)
 asmlinkage void syscall_trace_enter(struct pt_regs *regs)
 {
 	/* do the secure computing check first */
-	secure_computing(regs->orig_rax);
+	secure_computing(regs->orig_ax);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE)
 	    && (current->ptrace & PT_PTRACED))
@@ -414,14 +414,14 @@ asmlinkage void syscall_trace_enter(struct pt_regs *regs)
 	if (unlikely(current->audit_context)) {
 		if (test_thread_flag(TIF_IA32)) {
 			audit_syscall_entry(AUDIT_ARCH_I386,
-					    regs->orig_rax,
-					    regs->rbx, regs->rcx,
-					    regs->rdx, regs->rsi);
+					    regs->orig_ax,
+					    regs->bx, regs->cx,
+					    regs->dx, regs->si);
 		} else {
 			audit_syscall_entry(AUDIT_ARCH_X86_64,
-					    regs->orig_rax,
-					    regs->rdi, regs->rsi,
-					    regs->rdx, regs->r10);
+					    regs->orig_ax,
+					    regs->di, regs->si,
+					    regs->dx, regs->r10);
 		}
 	}
 }
@@ -429,7 +429,7 @@ asmlinkage void syscall_trace_enter(struct pt_regs *regs)
 asmlinkage void syscall_trace_leave(struct pt_regs *regs)
 {
 	if (unlikely(current->audit_context))
-		audit_syscall_exit(AUDITSC_RESULT(regs->rax), regs->rax);
+		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
 	if ((test_thread_flag(TIF_SYSCALL_TRACE)
 	     || test_thread_flag(TIF_SINGLESTEP))
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 0a7c812212c9..40fd3515ccf1 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -82,14 +82,14 @@ sys_sigaction(int sig, const struct old_sigaction __user *act,
 }
 
 asmlinkage int
-sys_sigaltstack(unsigned long ebx)
+sys_sigaltstack(unsigned long bx)
 {
 	/* This is needed to make gcc realize it doesn't own the "struct pt_regs" */
-	struct pt_regs *regs = (struct pt_regs *)&ebx;
-	const stack_t __user *uss = (const stack_t __user *)ebx;
-	stack_t __user *uoss = (stack_t __user *)regs->ecx;
+	struct pt_regs *regs = (struct pt_regs *)&bx;
+	const stack_t __user *uss = (const stack_t __user *)bx;
+	stack_t __user *uoss = (stack_t __user *)regs->cx;
 
-	return do_sigaltstack(uss, uoss, regs->esp);
+	return do_sigaltstack(uss, uoss, regs->sp);
 }
 
 
@@ -105,17 +105,17 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPY(x)		err |= __get_user(regs->x, &sc->x)
+#define COPY(x)		err |= __get_user(regs->x, &sc->e ## x)
 
 #define COPY_SEG(seg)							\
 	{ unsigned short tmp;						\
 	  err |= __get_user(tmp, &sc->seg);				\
-	  regs->x##seg = tmp; }
+	  regs->seg = tmp; }
 
 #define COPY_SEG_STRICT(seg)						\
 	{ unsigned short tmp;						\
 	  err |= __get_user(tmp, &sc->seg);				\
-	  regs->x##seg = tmp|3; }
+	  regs->seg = tmp|3; }
 
 #define GET_SEG(seg)							\
 	{ unsigned short tmp;						\
@@ -131,22 +131,22 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 	COPY_SEG(fs);
 	COPY_SEG(es);
 	COPY_SEG(ds);
-	COPY(edi);
-	COPY(esi);
-	COPY(ebp);
-	COPY(esp);
-	COPY(ebx);
-	COPY(edx);
-	COPY(ecx);
-	COPY(eip);
+	COPY(di);
+	COPY(si);
+	COPY(bp);
+	COPY(sp);
+	COPY(bx);
+	COPY(dx);
+	COPY(cx);
+	COPY(ip);
 	COPY_SEG_STRICT(cs);
 	COPY_SEG_STRICT(ss);
 	
 	{
 		unsigned int tmpflags;
 		err |= __get_user(tmpflags, &sc->eflags);
-		regs->eflags = (regs->eflags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
-		regs->orig_eax = -1;		/* disable syscall checks */
+		regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
+		regs->orig_ax = -1;		/* disable syscall checks */
 	}
 
 	{
@@ -175,9 +175,9 @@ badframe:
 asmlinkage int sys_sigreturn(unsigned long __unused)
 {
 	struct pt_regs *regs = (struct pt_regs *) &__unused;
-	struct sigframe __user *frame = (struct sigframe __user *)(regs->esp - 8);
+	struct sigframe __user *frame = (struct sigframe __user *)(regs->sp - 8);
 	sigset_t set;
-	int eax;
+	int ax;
 
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
 		goto badframe;
@@ -193,17 +193,17 @@ asmlinkage int sys_sigreturn(unsigned long __unused)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 	
-	if (restore_sigcontext(regs, &frame->sc, &eax))
+	if (restore_sigcontext(regs, &frame->sc, &ax))
 		goto badframe;
-	return eax;
+	return ax;
 
 badframe:
 	if (show_unhandled_signals && printk_ratelimit())
-		printk("%s%s[%d] bad frame in sigreturn frame:%p eip:%lx"
-		       " esp:%lx oeax:%lx\n",
+		printk("%s%s[%d] bad frame in sigreturn frame:%p ip:%lx"
+		       " sp:%lx oeax:%lx\n",
 		    task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG,
-		    current->comm, task_pid_nr(current), frame, regs->eip,
-		    regs->esp, regs->orig_eax);
+		    current->comm, task_pid_nr(current), frame, regs->ip,
+		    regs->sp, regs->orig_ax);
 
 	force_sig(SIGSEGV, current);
 	return 0;
@@ -212,9 +212,9 @@ badframe:
 asmlinkage int sys_rt_sigreturn(unsigned long __unused)
 {
 	struct pt_regs *regs = (struct pt_regs *) &__unused;
-	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)(regs->esp - 4);
+	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)(regs->sp - 4);
 	sigset_t set;
-	int eax;
+	int ax;
 
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
 		goto badframe;
@@ -227,13 +227,13 @@ asmlinkage int sys_rt_sigreturn(unsigned long __unused)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 	
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &eax))
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
 		goto badframe;
 
-	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->esp) == -EFAULT)
+	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT)
 		goto badframe;
 
-	return eax;
+	return ax;
 
 badframe:
 	force_sig(SIGSEGV, current);
@@ -250,27 +250,27 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
 {
 	int tmp, err = 0;
 
-	err |= __put_user(regs->xfs, (unsigned int __user *)&sc->fs);
+	err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs);
 	savesegment(gs, tmp);
 	err |= __put_user(tmp, (unsigned int __user *)&sc->gs);
 
-	err |= __put_user(regs->xes, (unsigned int __user *)&sc->es);
-	err |= __put_user(regs->xds, (unsigned int __user *)&sc->ds);
-	err |= __put_user(regs->edi, &sc->edi);
-	err |= __put_user(regs->esi, &sc->esi);
-	err |= __put_user(regs->ebp, &sc->ebp);
-	err |= __put_user(regs->esp, &sc->esp);
-	err |= __put_user(regs->ebx, &sc->ebx);
-	err |= __put_user(regs->edx, &sc->edx);
-	err |= __put_user(regs->ecx, &sc->ecx);
-	err |= __put_user(regs->eax, &sc->eax);
+	err |= __put_user(regs->es, (unsigned int __user *)&sc->es);
+	err |= __put_user(regs->ds, (unsigned int __user *)&sc->ds);
+	err |= __put_user(regs->di, &sc->edi);
+	err |= __put_user(regs->si, &sc->esi);
+	err |= __put_user(regs->bp, &sc->ebp);
+	err |= __put_user(regs->sp, &sc->esp);
+	err |= __put_user(regs->bx, &sc->ebx);
+	err |= __put_user(regs->dx, &sc->edx);
+	err |= __put_user(regs->cx, &sc->ecx);
+	err |= __put_user(regs->ax, &sc->eax);
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
-	err |= __put_user(regs->eip, &sc->eip);
-	err |= __put_user(regs->xcs, (unsigned int __user *)&sc->cs);
-	err |= __put_user(regs->eflags, &sc->eflags);
-	err |= __put_user(regs->esp, &sc->esp_at_signal);
-	err |= __put_user(regs->xss, (unsigned int __user *)&sc->ss);
+	err |= __put_user(regs->ip, &sc->eip);
+	err |= __put_user(regs->cs, (unsigned int __user *)&sc->cs);
+	err |= __put_user(regs->flags, &sc->eflags);
+	err |= __put_user(regs->sp, &sc->esp_at_signal);
+	err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss);
 
 	tmp = save_i387(fpstate);
 	if (tmp < 0)
@@ -291,36 +291,36 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
 static inline void __user *
 get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size)
 {
-	unsigned long esp;
+	unsigned long sp;
 
 	/* Default to using normal stack */
-	esp = regs->esp;
+	sp = regs->sp;
 
 	/*
 	 * If we are on the alternate signal stack and would overflow it, don't.
 	 * Return an always-bogus address instead so we will die with SIGSEGV.
 	 */
-	if (on_sig_stack(esp) && !likely(on_sig_stack(esp - frame_size)))
+	if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
 		return (void __user *) -1L;
 
 	/* This is the X/Open sanctioned signal stack switching.  */
 	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(esp) == 0)
-			esp = current->sas_ss_sp + current->sas_ss_size;
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
 	}
 
 	/* This is the legacy signal stack switching. */
-	else if ((regs->xss & 0xffff) != __USER_DS &&
+	else if ((regs->ss & 0xffff) != __USER_DS &&
 		 !(ka->sa.sa_flags & SA_RESTORER) &&
 		 ka->sa.sa_restorer) {
-		esp = (unsigned long) ka->sa.sa_restorer;
+		sp = (unsigned long) ka->sa.sa_restorer;
 	}
 
-	esp -= frame_size;
+	sp -= frame_size;
 	/* Align the stack pointer according to the i386 ABI,
 	 * i.e. so that on function entry ((sp + 4) & 15) == 0. */
-	esp = ((esp + 4) & -16ul) - 4;
-	return (void __user *) esp;
+	sp = ((sp + 4) & -16ul) - 4;
+	return (void __user *) sp;
 }
 
 /* These symbols are defined with the addresses in the vsyscall page.
@@ -387,16 +387,16 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->esp = (unsigned long) frame;
-	regs->eip = (unsigned long) ka->sa.sa_handler;
-	regs->eax = (unsigned long) sig;
-	regs->edx = (unsigned long) 0;
-	regs->ecx = (unsigned long) 0;
+	regs->sp = (unsigned long) frame;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
+	regs->ax = (unsigned long) sig;
+	regs->dx = (unsigned long) 0;
+	regs->cx = (unsigned long) 0;
 
-	regs->xds = __USER_DS;
-	regs->xes = __USER_DS;
-	regs->xss = __USER_DS;
-	regs->xcs = __USER_CS;
+	regs->ds = __USER_DS;
+	regs->es = __USER_DS;
+	regs->ss = __USER_DS;
+	regs->cs = __USER_CS;
 
 	/*
 	 * Clear TF when entering the signal handler, but
@@ -404,13 +404,13 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 	 * The tracer may want to single-step inside the
 	 * handler too.
 	 */
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
-		current->comm, current->pid, frame, regs->eip, frame->pretcode);
+		current->comm, current->pid, frame, regs->ip, frame->pretcode);
 #endif
 
 	return 0;
@@ -450,7 +450,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->esp),
+	err |= __put_user(sas_ss_flags(regs->sp),
 			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
 	err |= setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate,
@@ -466,7 +466,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(restorer, &frame->pretcode);
 	 
 	/*
-	 * This is movl $,%eax ; int $0x80
+	 * This is movl $,%ax ; int $0x80
 	 *
 	 * WE DO NOT USE IT ANY MORE! It's only left here for historical
 	 * reasons and because gdb uses it as a signature to notice
@@ -480,16 +480,16 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->esp = (unsigned long) frame;
-	regs->eip = (unsigned long) ka->sa.sa_handler;
-	regs->eax = (unsigned long) usig;
-	regs->edx = (unsigned long) &frame->info;
-	regs->ecx = (unsigned long) &frame->uc;
+	regs->sp = (unsigned long) frame;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
+	regs->ax = (unsigned long) usig;
+	regs->dx = (unsigned long) &frame->info;
+	regs->cx = (unsigned long) &frame->uc;
 
-	regs->xds = __USER_DS;
-	regs->xes = __USER_DS;
-	regs->xss = __USER_DS;
-	regs->xcs = __USER_CS;
+	regs->ds = __USER_DS;
+	regs->es = __USER_DS;
+	regs->ss = __USER_DS;
+	regs->cs = __USER_CS;
 
 	/*
 	 * Clear TF when entering the signal handler, but
@@ -497,13 +497,13 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	 * The tracer may want to single-step inside the
 	 * handler too.
 	 */
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
-		current->comm, current->pid, frame, regs->eip, frame->pretcode);
+		current->comm, current->pid, frame, regs->ip, frame->pretcode);
 #endif
 
 	return 0;
@@ -524,23 +524,23 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	int ret;
 
 	/* Are we from a system call? */
-	if (regs->orig_eax >= 0) {
+	if (regs->orig_ax >= 0) {
 		/* If so, check system call restarting.. */
-		switch (regs->eax) {
+		switch (regs->ax) {
 		        case -ERESTART_RESTARTBLOCK:
 			case -ERESTARTNOHAND:
-				regs->eax = -EINTR;
+				regs->ax = -EINTR;
 				break;
 
 			case -ERESTARTSYS:
 				if (!(ka->sa.sa_flags & SA_RESTART)) {
-					regs->eax = -EINTR;
+					regs->ax = -EINTR;
 					break;
 				}
 			/* fallthrough */
 			case -ERESTARTNOINTR:
-				regs->eax = regs->orig_eax;
-				regs->eip -= 2;
+				regs->ax = regs->orig_ax;
+				regs->ip -= 2;
 		}
 	}
 
@@ -548,9 +548,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	 * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF
 	 * flag so that register information in the sigcontext is correct.
 	 */
-	if (unlikely(regs->eflags & X86_EFLAGS_TF) &&
+	if (unlikely(regs->flags & X86_EFLAGS_TF) &&
 	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
-		regs->eflags &= ~X86_EFLAGS_TF;
+		regs->flags &= ~X86_EFLAGS_TF;
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
@@ -622,19 +622,19 @@ static void fastcall do_signal(struct pt_regs *regs)
 	}
 
 	/* Did we come from a system call? */
-	if (regs->orig_eax >= 0) {
+	if (regs->orig_ax >= 0) {
 		/* Restart the system call - no handlers present */
-		switch (regs->eax) {
+		switch (regs->ax) {
 		case -ERESTARTNOHAND:
 		case -ERESTARTSYS:
 		case -ERESTARTNOINTR:
-			regs->eax = regs->orig_eax;
-			regs->eip -= 2;
+			regs->ax = regs->orig_ax;
+			regs->ip -= 2;
 			break;
 
 		case -ERESTART_RESTARTBLOCK:
-			regs->eax = __NR_restart_syscall;
-			regs->eip -= 2;
+			regs->ax = __NR_restart_syscall;
+			regs->ip -= 2;
 			break;
 		}
 	}
@@ -657,7 +657,7 @@ void do_notify_resume(struct pt_regs *regs, void *_unused,
 {
 	/* Pending single-step? */
 	if (thread_info_flags & _TIF_SINGLESTEP) {
-		regs->eflags |= TF_MASK;
+		regs->flags |= TF_MASK;
 		clear_thread_flag(TIF_SINGLESTEP);
 	}
 
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index ab0178ebe00a..4b228fd83b31 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -39,7 +39,7 @@ asmlinkage long
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		struct pt_regs *regs)
 {
-	return do_sigaltstack(uss, uoss, regs->rsp);
+	return do_sigaltstack(uss, uoss, regs->sp);
 }
 
 
@@ -62,10 +62,11 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
+#define COPYR(x)	err |= __get_user(regs->x, &sc->r ## x)
 #define COPY(x)		err |= __get_user(regs->x, &sc->x)
 
-	COPY(rdi); COPY(rsi); COPY(rbp); COPY(rsp); COPY(rbx);
-	COPY(rdx); COPY(rcx); COPY(rip);
+	COPYR(di); COPYR(si); COPYR(bp); COPYR(sp); COPYR(bx);
+	COPYR(dx); COPYR(cx); COPYR(ip);
 	COPY(r8);
 	COPY(r9);
 	COPY(r10);
@@ -87,8 +88,8 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned
 	{
 		unsigned int tmpflags;
 		err |= __get_user(tmpflags, &sc->eflags);
-		regs->eflags = (regs->eflags & ~0x40DD5) | (tmpflags & 0x40DD5);
-		regs->orig_rax = -1;		/* disable syscall checks */
+		regs->flags = (regs->flags & ~0x40DD5) | (tmpflags & 0x40DD5);
+		regs->orig_ax = -1;		/* disable syscall checks */
 	}
 
 	{
@@ -119,9 +120,9 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
 	sigset_t set;
-	unsigned long eax;
+	unsigned long ax;
 
-	frame = (struct rt_sigframe __user *)(regs->rsp - 8);
+	frame = (struct rt_sigframe __user *)(regs->sp - 8);
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) {
 		goto badframe;
 	} 
@@ -135,17 +136,17 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 	
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &eax))
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
 		goto badframe;
 
 #ifdef DEBUG_SIG
-	printk("%d sigreturn rip:%lx rsp:%lx frame:%p rax:%lx\n",current->pid,regs->rip,regs->rsp,frame,eax);
+	printk("%d sigreturn ip:%lx sp:%lx frame:%p ax:%lx\n",current->pid,regs->ip,regs->sp,frame,ax);
 #endif
 
-	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->rsp) == -EFAULT)
+	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT)
 		goto badframe;
 
-	return eax;
+	return ax;
 
 badframe:
 	signal_fault(regs,frame,"sigreturn");
@@ -165,14 +166,14 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned lo
 	err |= __put_user(0, &sc->gs);
 	err |= __put_user(0, &sc->fs);
 
-	err |= __put_user(regs->rdi, &sc->rdi);
-	err |= __put_user(regs->rsi, &sc->rsi);
-	err |= __put_user(regs->rbp, &sc->rbp);
-	err |= __put_user(regs->rsp, &sc->rsp);
-	err |= __put_user(regs->rbx, &sc->rbx);
-	err |= __put_user(regs->rdx, &sc->rdx);
-	err |= __put_user(regs->rcx, &sc->rcx);
-	err |= __put_user(regs->rax, &sc->rax);
+	err |= __put_user(regs->di, &sc->rdi);
+	err |= __put_user(regs->si, &sc->rsi);
+	err |= __put_user(regs->bp, &sc->rbp);
+	err |= __put_user(regs->sp, &sc->rsp);
+	err |= __put_user(regs->bx, &sc->rbx);
+	err |= __put_user(regs->dx, &sc->rdx);
+	err |= __put_user(regs->cx, &sc->rcx);
+	err |= __put_user(regs->ax, &sc->rax);
 	err |= __put_user(regs->r8, &sc->r8);
 	err |= __put_user(regs->r9, &sc->r9);
 	err |= __put_user(regs->r10, &sc->r10);
@@ -183,8 +184,8 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned lo
 	err |= __put_user(regs->r15, &sc->r15);
 	err |= __put_user(me->thread.trap_no, &sc->trapno);
 	err |= __put_user(me->thread.error_code, &sc->err);
-	err |= __put_user(regs->rip, &sc->rip);
-	err |= __put_user(regs->eflags, &sc->eflags);
+	err |= __put_user(regs->ip, &sc->rip);
+	err |= __put_user(regs->flags, &sc->eflags);
 	err |= __put_user(mask, &sc->oldmask);
 	err |= __put_user(me->thread.cr2, &sc->cr2);
 
@@ -198,18 +199,18 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned lo
 static void __user *
 get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size)
 {
-	unsigned long rsp;
+	unsigned long sp;
 
 	/* Default to using normal stack - redzone*/
-	rsp = regs->rsp - 128;
+	sp = regs->sp - 128;
 
 	/* This is the X/Open sanctioned signal stack switching.  */
 	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(rsp) == 0)
-			rsp = current->sas_ss_sp + current->sas_ss_size;
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
 	}
 
-	return (void __user *)round_down(rsp - size, 16); 
+	return (void __user *)round_down(sp - size, 16);
 }
 
 static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
@@ -246,7 +247,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->rsp),
+	err |= __put_user(sas_ss_flags(regs->sp),
 			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size);
 	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0], me);
@@ -271,21 +272,21 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		goto give_sigsegv;
 
 #ifdef DEBUG_SIG
-	printk("%d old rip %lx old rsp %lx old rax %lx\n", current->pid,regs->rip,regs->rsp,regs->rax);
+	printk("%d old ip %lx old sp %lx old ax %lx\n", current->pid,regs->ip,regs->sp,regs->ax);
 #endif
 
 	/* Set up registers for signal handler */
-	regs->rdi = sig;
+	regs->di = sig;
 	/* In case the signal handler was declared without prototypes */ 
-	regs->rax = 0;	
+	regs->ax = 0;
 
 	/* This also works for non SA_SIGINFO handlers because they expect the
 	   next argument after the signal number on the stack. */
-	regs->rsi = (unsigned long)&frame->info; 
-	regs->rdx = (unsigned long)&frame->uc; 
-	regs->rip = (unsigned long) ka->sa.sa_handler;
+	regs->si = (unsigned long)&frame->info;
+	regs->dx = (unsigned long)&frame->uc;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
 
-	regs->rsp = (unsigned long)frame;
+	regs->sp = (unsigned long)frame;
 
 	/* Set up the CS register to run signal handlers in 64-bit mode,
 	   even if the handler happens to be interrupting 32-bit code. */
@@ -295,12 +296,12 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	   see include/asm-x86_64/uaccess.h for details. */
 	set_fs(USER_DS);
 
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 #ifdef DEBUG_SIG
 	printk("SIG deliver (%s:%d): sp=%p pc=%lx ra=%p\n",
-		current->comm, current->pid, frame, regs->rip, frame->pretcode);
+		current->comm, current->pid, frame, regs->ip, frame->pretcode);
 #endif
 
 	return 0;
@@ -321,29 +322,29 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	int ret;
 
 #ifdef DEBUG_SIG
-	printk("handle_signal pid:%d sig:%lu rip:%lx rsp:%lx regs=%p\n",
+	printk("handle_signal pid:%d sig:%lu ip:%lx sp:%lx regs=%p\n",
 		current->pid, sig,
-		regs->rip, regs->rsp, regs);
+		regs->ip, regs->sp, regs);
 #endif
 
 	/* Are we from a system call? */
-	if ((long)regs->orig_rax >= 0) {
+	if ((long)regs->orig_ax >= 0) {
 		/* If so, check system call restarting.. */
-		switch (regs->rax) {
+		switch (regs->ax) {
 		        case -ERESTART_RESTARTBLOCK:
 			case -ERESTARTNOHAND:
-				regs->rax = -EINTR;
+				regs->ax = -EINTR;
 				break;
 
 			case -ERESTARTSYS:
 				if (!(ka->sa.sa_flags & SA_RESTART)) {
-					regs->rax = -EINTR;
+					regs->ax = -EINTR;
 					break;
 				}
 				/* fallthrough */
 			case -ERESTARTNOINTR:
-				regs->rax = regs->orig_rax;
-				regs->rip -= 2;
+				regs->ax = regs->orig_ax;
+				regs->ip -= 2;
 				break;
 		}
 	}
@@ -352,9 +353,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	 * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF
 	 * flag so that register information in the sigcontext is correct.
 	 */
-	if (unlikely(regs->eflags & X86_EFLAGS_TF) &&
+	if (unlikely(regs->flags & X86_EFLAGS_TF) &&
 	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
-		regs->eflags &= ~X86_EFLAGS_TF;
+		regs->flags &= ~X86_EFLAGS_TF;
 
 #ifdef CONFIG_IA32_EMULATION
 	if (test_thread_flag(TIF_IA32)) {
@@ -426,21 +427,21 @@ static void do_signal(struct pt_regs *regs)
 	}
 
 	/* Did we come from a system call? */
-	if ((long)regs->orig_rax >= 0) {
+	if ((long)regs->orig_ax >= 0) {
 		/* Restart the system call - no handlers present */
-		long res = regs->rax;
+		long res = regs->ax;
 		switch (res) {
 		case -ERESTARTNOHAND:
 		case -ERESTARTSYS:
 		case -ERESTARTNOINTR:
-			regs->rax = regs->orig_rax;
-			regs->rip -= 2;
+			regs->ax = regs->orig_ax;
+			regs->ip -= 2;
 			break;
 		case -ERESTART_RESTARTBLOCK:
-			regs->rax = test_thread_flag(TIF_IA32) ?
+			regs->ax = test_thread_flag(TIF_IA32) ?
 					__NR_ia32_restart_syscall :
 					__NR_restart_syscall;
-			regs->rip -= 2;
+			regs->ip -= 2;
 			break;
 		}
 	}
@@ -457,13 +458,13 @@ void
 do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
 #ifdef DEBUG_SIG
-	printk("do_notify_resume flags:%x rip:%lx rsp:%lx caller:%p pending:%x\n",
-	       thread_info_flags, regs->rip, regs->rsp, __builtin_return_address(0),signal_pending(current)); 
+	printk("do_notify_resume flags:%x ip:%lx sp:%lx caller:%p pending:%x\n",
+	       thread_info_flags, regs->ip, regs->sp, __builtin_return_address(0),signal_pending(current));
 #endif
 	       
 	/* Pending single-step? */
 	if (thread_info_flags & _TIF_SINGLESTEP) {
-		regs->eflags |= TF_MASK;
+		regs->flags |= TF_MASK;
 		clear_thread_flag(TIF_SINGLESTEP);
 	}
 
@@ -485,8 +486,8 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
 { 
 	struct task_struct *me = current; 
 	if (show_unhandled_signals && printk_ratelimit())
-		printk("%s[%d] bad frame in %s frame:%p rip:%lx rsp:%lx orax:%lx\n",
-	       me->comm,me->pid,where,frame,regs->rip,regs->rsp,regs->orig_rax); 
+		printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx\n",
+	       me->comm,me->pid,where,frame,regs->ip,regs->sp,regs->orig_ax);
 
 	force_sig(SIGSEGV, me); 
 } 
diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 7142447b5666..02a6533e8909 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -136,7 +136,7 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs)
 	 * orig_rax contains the negated interrupt vector.
 	 * Use that to determine where the sender put the data.
 	 */
-	sender = ~regs->orig_rax - INVALIDATE_TLB_VECTOR_START;
+	sender = ~regs->orig_ax - INVALIDATE_TLB_VECTOR_START;
 	f = &per_cpu(flush_state, sender);
 
 	if (!cpu_isset(cpu, f->flush_cpumask))
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 0bf7f20baba0..3566191832b3 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -447,7 +447,7 @@ void __devinit initialize_secondary(void)
 {
 	/*
 	 * We don't actually need to load the full TSS,
-	 * basically just the stack pointer and the eip.
+	 * basically just the stack pointer and the ip.
 	 */
 
 	asm volatile(
@@ -459,7 +459,7 @@ void __devinit initialize_secondary(void)
 
 /* Static state in head.S used to set up a CPU */
 extern struct {
-	void * esp;
+	void * sp;
 	unsigned short ss;
 } stack_start;
 
@@ -667,7 +667,7 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 	 * target processor state.
 	 */
 	startup_ipi_hook(phys_apicid, (unsigned long) start_secondary,
-		         (unsigned long) stack_start.esp);
+		         (unsigned long) stack_start.sp);
 
 	/*
 	 * Run STARTUP IPI loop.
@@ -806,9 +806,9 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
 	alternatives_smp_switch(1);
 
 	/* So we see what's up   */
-	printk("Booting processor %d/%d eip %lx\n", cpu, apicid, start_eip);
+	printk("Booting processor %d/%d ip %lx\n", cpu, apicid, start_eip);
 	/* Stack for startup_32 can be just as for start_secondary onwards */
-	stack_start.esp = (void *) idle->thread.esp;
+	stack_start.sp = (void *) idle->thread.esp;
 
 	irq_ctx_init(cpu);
 
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index cf4b9dac4a05..f55c003f5b63 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -12,17 +12,12 @@ unsigned long convert_rip_to_linear(struct task_struct *child, struct pt_regs *r
 {
 	unsigned long addr, seg;
 
-#ifdef CONFIG_X86_64
-	addr = regs->rip;
+	addr = regs->ip;
 	seg = regs->cs & 0xffff;
-#else
-	addr = regs->eip;
-	seg = regs->xcs & 0xffff;
-	if (regs->eflags & X86_EFLAGS_VM) {
+	if (v8086_mode(regs)) {
 		addr = (addr & 0xffff) + (seg << 4);
 		return addr;
 	}
-#endif
 
 	/*
 	 * We'll assume that the code segments in the GDT
@@ -124,11 +119,11 @@ static int enable_single_step(struct task_struct *child)
 	/*
 	 * If TF was already set, don't do anything else
 	 */
-	if (regs->eflags & X86_EFLAGS_TF)
+	if (regs->flags & X86_EFLAGS_TF)
 		return 0;
 
 	/* Set TF on the kernel stack.. */
-	regs->eflags |= X86_EFLAGS_TF;
+	regs->flags |= X86_EFLAGS_TF;
 
 	/*
 	 * ..but if TF is changed by the instruction we will trace,
@@ -203,5 +198,5 @@ void user_disable_single_step(struct task_struct *child)
 
 	/* But touch TF only if it was set by us.. */
 	if (test_and_clear_tsk_thread_flag(child, TIF_FORCED_TF))
-		task_pt_regs(child)->eflags &= ~X86_EFLAGS_TF;
+		task_pt_regs(child)->flags &= ~X86_EFLAGS_TF;
 }
diff --git a/arch/x86/kernel/suspend_asm_64.S b/arch/x86/kernel/suspend_asm_64.S
index 72f952103e50..aeb9a4d7681e 100644
--- a/arch/x86/kernel/suspend_asm_64.S
+++ b/arch/x86/kernel/suspend_asm_64.S
@@ -18,13 +18,13 @@
 
 ENTRY(swsusp_arch_suspend)
 	movq	$saved_context, %rax
-	movq	%rsp, pt_regs_rsp(%rax)
-	movq	%rbp, pt_regs_rbp(%rax)
-	movq	%rsi, pt_regs_rsi(%rax)
-	movq	%rdi, pt_regs_rdi(%rax)
-	movq	%rbx, pt_regs_rbx(%rax)
-	movq	%rcx, pt_regs_rcx(%rax)
-	movq	%rdx, pt_regs_rdx(%rax)
+	movq	%rsp, pt_regs_sp(%rax)
+	movq	%rbp, pt_regs_bp(%rax)
+	movq	%rsi, pt_regs_si(%rax)
+	movq	%rdi, pt_regs_di(%rax)
+	movq	%rbx, pt_regs_bx(%rax)
+	movq	%rcx, pt_regs_cx(%rax)
+	movq	%rdx, pt_regs_dx(%rax)
 	movq	%r8, pt_regs_r8(%rax)
 	movq	%r9, pt_regs_r9(%rax)
 	movq	%r10, pt_regs_r10(%rax)
@@ -34,7 +34,7 @@ ENTRY(swsusp_arch_suspend)
 	movq	%r14, pt_regs_r14(%rax)
 	movq	%r15, pt_regs_r15(%rax)
 	pushfq
-	popq	pt_regs_eflags(%rax)
+	popq	pt_regs_flags(%rax)
 
 	/* save the address of restore_registers */
 	movq	$restore_registers, %rax
@@ -115,13 +115,13 @@ ENTRY(restore_registers)
 
 	/* We don't restore %rax, it must be 0 anyway */
 	movq	$saved_context, %rax
-	movq	pt_regs_rsp(%rax), %rsp
-	movq	pt_regs_rbp(%rax), %rbp
-	movq	pt_regs_rsi(%rax), %rsi
-	movq	pt_regs_rdi(%rax), %rdi
-	movq	pt_regs_rbx(%rax), %rbx
-	movq	pt_regs_rcx(%rax), %rcx
-	movq	pt_regs_rdx(%rax), %rdx
+	movq	pt_regs_sp(%rax), %rsp
+	movq	pt_regs_bp(%rax), %rbp
+	movq	pt_regs_si(%rax), %rsi
+	movq	pt_regs_di(%rax), %rdi
+	movq	pt_regs_bx(%rax), %rbx
+	movq	pt_regs_cx(%rax), %rcx
+	movq	pt_regs_dx(%rax), %rdx
 	movq	pt_regs_r8(%rax), %r8
 	movq	pt_regs_r9(%rax), %r9
 	movq	pt_regs_r10(%rax), %r10
@@ -130,7 +130,7 @@ ENTRY(restore_registers)
 	movq	pt_regs_r13(%rax), %r13
 	movq	pt_regs_r14(%rax), %r14
 	movq	pt_regs_r15(%rax), %r15
-	pushq	pt_regs_eflags(%rax)
+	pushq	pt_regs_flags(%rax)
 	popfq
 
 	xorq	%rax, %rax
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index 2dcbb81b4cd3..1a89e93f3f1c 100644
--- a/arch/x86/kernel/time_32.c
+++ b/arch/x86/kernel/time_32.c
@@ -49,15 +49,15 @@ unsigned long profile_pc(struct pt_regs *regs)
 	unsigned long pc = instruction_pointer(regs);
 
 #ifdef CONFIG_SMP
-	if (!v8086_mode(regs) && SEGMENT_IS_KERNEL_CODE(regs->xcs) &&
+	if (!v8086_mode(regs) && SEGMENT_IS_KERNEL_CODE(regs->cs) &&
 	    in_lock_functions(pc)) {
 #ifdef CONFIG_FRAME_POINTER
-		return *(unsigned long *)(regs->ebp + 4);
+		return *(unsigned long *)(regs->bp + 4);
 #else
-		unsigned long *sp = (unsigned long *)&regs->esp;
+		unsigned long *sp = (unsigned long *)&regs->sp;
 
 		/* Return address is either directly at stack pointer
-		   or above a saved eflags. Eflags has bits 22-31 zero,
+		   or above a saved flags. Eflags has bits 22-31 zero,
 		   kernel addresses don't. */
 		if (sp[0] >> 22)
 			return sp[0];
diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index f88bf6b802e3..bf0bcc9bb001 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -29,10 +29,10 @@ unsigned long profile_pc(struct pt_regs *regs)
 	unsigned long pc = instruction_pointer(regs);
 
 	/* Assume the lock function has either no stack frame or a copy
-	   of eflags from PUSHF
+	   of flags from PUSHF
 	   Eflags always has bits 22 and up cleared unlike kernel addresses. */
 	if (!user_mode(regs) && in_lock_functions(pc)) {
-		unsigned long *sp = (unsigned long *)regs->rsp;
+		unsigned long *sp = (unsigned long *)regs->sp;
 		if (sp[0] >> 22)
 			return sp[0];
 		if (sp[1] >> 22)
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 9b0bbd508cd5..931ef10960ee 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -114,11 +114,11 @@ struct stack_frame {
 };
 
 static inline unsigned long print_context_stack(struct thread_info *tinfo,
-				unsigned long *stack, unsigned long ebp,
+				unsigned long *stack, unsigned long bp,
 				const struct stacktrace_ops *ops, void *data)
 {
 #ifdef	CONFIG_FRAME_POINTER
-	struct stack_frame *frame = (struct stack_frame *)ebp;
+	struct stack_frame *frame = (struct stack_frame *)bp;
 	while (valid_stack_ptr(tinfo, frame, sizeof(*frame))) {
 		struct stack_frame *next;
 		unsigned long addr;
@@ -145,7 +145,7 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 			ops->address(data, addr);
 	}
 #endif
-	return ebp;
+	return bp;
 }
 
 #define MSG(msg) ops->warning(data, msg)
@@ -154,7 +154,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	        unsigned long *stack,
 		const struct stacktrace_ops *ops, void *data)
 {
-	unsigned long ebp = 0;
+	unsigned long bp = 0;
 
 	if (!task)
 		task = current;
@@ -167,13 +167,13 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	}
 
 #ifdef CONFIG_FRAME_POINTER
-	if (!ebp) {
+	if (!bp) {
 		if (task == current) {
-			/* Grab ebp right from our regs */
-			asm ("movl %%ebp, %0" : "=r" (ebp) : );
+			/* Grab bp right from our regs */
+			asm ("movl %%ebp, %0" : "=r" (bp) : );
 		} else {
-			/* ebp is the last reg pushed by switch_to */
-			ebp = *(unsigned long *) task->thread.esp;
+			/* bp is the last reg pushed by switch_to */
+			bp = *(unsigned long *) task->thread.esp;
 		}
 	}
 #endif
@@ -182,7 +182,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		struct thread_info *context;
 		context = (struct thread_info *)
 			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		ebp = print_context_stack(context, stack, ebp, ops, data);
+		bp = print_context_stack(context, stack, bp, ops, data);
 		/* Should be after the line below, but somewhere
 		   in early boot context comes out corrupted and we
 		   can't reference it -AK */
@@ -246,19 +246,19 @@ void show_trace(struct task_struct *task, struct pt_regs *regs,
 }
 
 static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
-			       unsigned long *esp, char *log_lvl)
+			       unsigned long *sp, char *log_lvl)
 {
 	unsigned long *stack;
 	int i;
 
-	if (esp == NULL) {
+	if (sp == NULL) {
 		if (task)
-			esp = (unsigned long*)task->thread.esp;
+			sp = (unsigned long*)task->thread.esp;
 		else
-			esp = (unsigned long *)&esp;
+			sp = (unsigned long *)&sp;
 	}
 
-	stack = esp;
+	stack = sp;
 	for(i = 0; i < kstack_depth_to_print; i++) {
 		if (kstack_end(stack))
 			break;
@@ -267,13 +267,13 @@ static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 		printk("%08lx ", *stack++);
 	}
 	printk("\n%sCall Trace:\n", log_lvl);
-	show_trace_log_lvl(task, regs, esp, log_lvl);
+	show_trace_log_lvl(task, regs, sp, log_lvl);
 }
 
-void show_stack(struct task_struct *task, unsigned long *esp)
+void show_stack(struct task_struct *task, unsigned long *sp)
 {
 	printk("       ");
-	show_stack_log_lvl(task, NULL, esp, "");
+	show_stack_log_lvl(task, NULL, sp, "");
 }
 
 /*
@@ -307,30 +307,30 @@ void show_registers(struct pt_regs *regs)
 	 * time of the fault..
 	 */
 	if (!user_mode_vm(regs)) {
-		u8 *eip;
+		u8 *ip;
 		unsigned int code_prologue = code_bytes * 43 / 64;
 		unsigned int code_len = code_bytes;
 		unsigned char c;
 
 		printk("\n" KERN_EMERG "Stack: ");
-		show_stack_log_lvl(NULL, regs, &regs->esp, KERN_EMERG);
+		show_stack_log_lvl(NULL, regs, &regs->sp, KERN_EMERG);
 
 		printk(KERN_EMERG "Code: ");
 
-		eip = (u8 *)regs->eip - code_prologue;
-		if (eip < (u8 *)PAGE_OFFSET ||
-			probe_kernel_address(eip, c)) {
+		ip = (u8 *)regs->ip - code_prologue;
+		if (ip < (u8 *)PAGE_OFFSET ||
+			probe_kernel_address(ip, c)) {
 			/* try starting at EIP */
-			eip = (u8 *)regs->eip;
+			ip = (u8 *)regs->ip;
 			code_len = code_len - code_prologue + 1;
 		}
-		for (i = 0; i < code_len; i++, eip++) {
-			if (eip < (u8 *)PAGE_OFFSET ||
-				probe_kernel_address(eip, c)) {
+		for (i = 0; i < code_len; i++, ip++) {
+			if (ip < (u8 *)PAGE_OFFSET ||
+				probe_kernel_address(ip, c)) {
 				printk(" Bad EIP value.");
 				break;
 			}
-			if (eip == (u8 *)regs->eip)
+			if (ip == (u8 *)regs->ip)
 				printk("<%02x> ", c);
 			else
 				printk("%02x ", c);
@@ -339,13 +339,13 @@ void show_registers(struct pt_regs *regs)
 	printk("\n");
 }	
 
-int is_valid_bugaddr(unsigned long eip)
+int is_valid_bugaddr(unsigned long ip)
 {
 	unsigned short ud2;
 
-	if (eip < PAGE_OFFSET)
+	if (ip < PAGE_OFFSET)
 		return 0;
-	if (probe_kernel_address((unsigned short *)eip, ud2))
+	if (probe_kernel_address((unsigned short *)ip, ud2))
 		return 0;
 
 	return ud2 == 0x0b0f;
@@ -382,10 +382,10 @@ void die(const char * str, struct pt_regs * regs, long err)
 		raw_local_irq_save(flags);
 
 	if (++die.lock_owner_depth < 3) {
-		unsigned long esp;
+		unsigned long sp;
 		unsigned short ss;
 
-		report_bug(regs->eip, regs);
+		report_bug(regs->ip, regs);
 
 		printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff,
 		       ++die_counter);
@@ -405,15 +405,15 @@ void die(const char * str, struct pt_regs * regs, long err)
 				NOTIFY_STOP) {
 			show_registers(regs);
 			/* Executive summary in case the oops scrolled away */
-			esp = (unsigned long) (&regs->esp);
+			sp = (unsigned long) (&regs->sp);
 			savesegment(ss, ss);
 			if (user_mode(regs)) {
-				esp = regs->esp;
-				ss = regs->xss & 0xffff;
+				sp = regs->sp;
+				ss = regs->ss & 0xffff;
 			}
-			printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip);
-			print_symbol("%s", regs->eip);
-			printk(" SS:ESP %04x:%08lx\n", ss, esp);
+			printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
+			print_symbol("%s", regs->ip);
+			printk(" SS:ESP %04x:%08lx\n", ss, sp);
 		}
 		else
 			regs = NULL;
@@ -454,7 +454,7 @@ static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86,
 {
 	struct task_struct *tsk = current;
 
-	if (regs->eflags & VM_MASK) {
+	if (regs->flags & VM_MASK) {
 		if (vm86)
 			goto vm86_trap;
 		goto trap_signal;
@@ -548,13 +548,13 @@ fastcall void do_##name(struct pt_regs * regs, long error_code) \
 	do_trap(trapnr, signr, str, 1, regs, error_code, &info); \
 }
 
-DO_VM86_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->eip)
+DO_VM86_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->ip)
 #ifndef CONFIG_KPROBES
 DO_VM86_ERROR( 3, SIGTRAP, "int3", int3)
 #endif
 DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow)
 DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip, 0)
+DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0)
 DO_ERROR( 9, SIGFPE,  "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
 DO_ERROR(11, SIGBUS,  "segment not present", segment_not_present)
@@ -596,7 +596,7 @@ fastcall void __kprobes do_general_protection(struct pt_regs * regs,
 	}
 	put_cpu();
 
-	if (regs->eflags & VM_MASK)
+	if (regs->flags & VM_MASK)
 		goto gp_in_vm86;
 
 	if (!user_mode(regs))
@@ -607,9 +607,9 @@ fastcall void __kprobes do_general_protection(struct pt_regs * regs,
 	if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) &&
 	    printk_ratelimit())
 		printk(KERN_INFO
-		    "%s[%d] general protection eip:%lx esp:%lx error:%lx\n",
+		    "%s[%d] general protection ip:%lx sp:%lx error:%lx\n",
 		    current->comm, task_pid_nr(current),
-		    regs->eip, regs->esp, error_code);
+		    regs->ip, regs->sp, error_code);
 
 	force_sig(SIGSEGV, current);
 	return;
@@ -705,8 +705,8 @@ void __kprobes die_nmi(struct pt_regs *regs, const char *msg)
 	*/
 	bust_spinlocks(1);
 	printk(KERN_EMERG "%s", msg);
-	printk(" on CPU%d, eip %08lx, registers:\n",
-		smp_processor_id(), regs->eip);
+	printk(" on CPU%d, ip %08lx, registers:\n",
+		smp_processor_id(), regs->ip);
 	show_registers(regs);
 	console_silent();
 	spin_unlock(&nmi_print_lock);
@@ -847,7 +847,7 @@ fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 					SIGTRAP) == NOTIFY_STOP)
 		return;
 	/* It's safe to allow irq's after DR6 has been saved */
-	if (regs->eflags & X86_EFLAGS_IF)
+	if (regs->flags & X86_EFLAGS_IF)
 		local_irq_enable();
 
 	/* Mask out spurious debug traps due to lazy DR7 setting */
@@ -856,7 +856,7 @@ fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 			goto clear_dr7;
 	}
 
-	if (regs->eflags & VM_MASK)
+	if (regs->flags & VM_MASK)
 		goto debug_vm86;
 
 	/* Save debug status register where ptrace can see it */
@@ -892,7 +892,7 @@ debug_vm86:
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	return;
 }
 
@@ -901,7 +901,7 @@ clear_TF_reenable:
  * the correct behaviour even in the presence of the asynchronous
  * IRQ13 behaviour
  */
-void math_error(void __user *eip)
+void math_error(void __user *ip)
 {
 	struct task_struct * task;
 	siginfo_t info;
@@ -917,7 +917,7 @@ void math_error(void __user *eip)
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;
 	info.si_code = __SI_FAULT;
-	info.si_addr = eip;
+	info.si_addr = ip;
 	/*
 	 * (~cwd & swd) will mask out exceptions that are not set to unmasked
 	 * status.  0x3f is the exception bits in these regs, 0x200 is the
@@ -963,10 +963,10 @@ void math_error(void __user *eip)
 fastcall void do_coprocessor_error(struct pt_regs * regs, long error_code)
 {
 	ignore_fpu_irq = 1;
-	math_error((void __user *)regs->eip);
+	math_error((void __user *)regs->ip);
 }
 
-static void simd_math_error(void __user *eip)
+static void simd_math_error(void __user *ip)
 {
 	struct task_struct * task;
 	siginfo_t info;
@@ -982,7 +982,7 @@ static void simd_math_error(void __user *eip)
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;
 	info.si_code = __SI_FAULT;
-	info.si_addr = eip;
+	info.si_addr = ip;
 	/*
 	 * The SIMD FPU exceptions are handled a little differently, as there
 	 * is only a single status/control register.  Thus, to determine which
@@ -1020,13 +1020,13 @@ fastcall void do_simd_coprocessor_error(struct pt_regs * regs,
 	if (cpu_has_xmm) {
 		/* Handle SIMD FPU exceptions on PIII+ processors. */
 		ignore_fpu_irq = 1;
-		simd_math_error((void __user *)regs->eip);
+		simd_math_error((void __user *)regs->ip);
 	} else {
 		/*
 		 * Handle strange cache flush from user space exception
 		 * in all other cases.  This is undocumented behaviour.
 		 */
-		if (regs->eflags & VM_MASK) {
+		if (regs->flags & VM_MASK) {
 			handle_vm86_fault((struct kernel_vm86_regs *)regs,
 					  error_code);
 			return;
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 610a64d6bdf0..f7fecf9d47c3 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -76,20 +76,20 @@ asmlinkage void spurious_interrupt_bug(void);
 
 static inline void conditional_sti(struct pt_regs *regs)
 {
-	if (regs->eflags & X86_EFLAGS_IF)
+	if (regs->flags & X86_EFLAGS_IF)
 		local_irq_enable();
 }
 
 static inline void preempt_conditional_sti(struct pt_regs *regs)
 {
 	preempt_disable();
-	if (regs->eflags & X86_EFLAGS_IF)
+	if (regs->flags & X86_EFLAGS_IF)
 		local_irq_enable();
 }
 
 static inline void preempt_conditional_cli(struct pt_regs *regs)
 {
-	if (regs->eflags & X86_EFLAGS_IF)
+	if (regs->flags & X86_EFLAGS_IF)
 		local_irq_disable();
 	/* Make sure to not schedule here because we could be running
 	   on an exception stack. */
@@ -353,7 +353,7 @@ show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack)
 }
 
 static void
-_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
+_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp)
 {
 	unsigned long *stack;
 	int i;
@@ -364,14 +364,14 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 	// debugging aid: "show_stack(NULL, NULL);" prints the
 	// back trace for this cpu.
 
-	if (rsp == NULL) {
+	if (sp == NULL) {
 		if (tsk)
-			rsp = (unsigned long *)tsk->thread.rsp;
+			sp = (unsigned long *)tsk->thread.rsp;
 		else
-			rsp = (unsigned long *)&rsp;
+			sp = (unsigned long *)&sp;
 	}
 
-	stack = rsp;
+	stack = sp;
 	for(i=0; i < kstack_depth_to_print; i++) {
 		if (stack >= irqstack && stack <= irqstack_end) {
 			if (stack == irqstack_end) {
@@ -387,12 +387,12 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 		printk(" %016lx", *stack++);
 		touch_nmi_watchdog();
 	}
-	show_trace(tsk, regs, rsp);
+	show_trace(tsk, regs, sp);
 }
 
-void show_stack(struct task_struct *tsk, unsigned long * rsp)
+void show_stack(struct task_struct *tsk, unsigned long * sp)
 {
-	_show_stack(tsk, NULL, rsp);
+	_show_stack(tsk, NULL, sp);
 }
 
 /*
@@ -416,11 +416,11 @@ void show_registers(struct pt_regs *regs)
 {
 	int i;
 	int in_kernel = !user_mode(regs);
-	unsigned long rsp;
+	unsigned long sp;
 	const int cpu = smp_processor_id();
 	struct task_struct *cur = cpu_pda(cpu)->pcurrent;
 
-	rsp = regs->rsp;
+	sp = regs->sp;
 	printk("CPU %d ", cpu);
 	__show_regs(regs);
 	printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
@@ -432,15 +432,15 @@ void show_registers(struct pt_regs *regs)
 	 */
 	if (in_kernel) {
 		printk("Stack: ");
-		_show_stack(NULL, regs, (unsigned long*)rsp);
+		_show_stack(NULL, regs, (unsigned long*)sp);
 
 		printk("\nCode: ");
-		if (regs->rip < PAGE_OFFSET)
+		if (regs->ip < PAGE_OFFSET)
 			goto bad;
 
 		for (i=0; i<20; i++) {
 			unsigned char c;
-			if (__get_user(c, &((unsigned char*)regs->rip)[i])) {
+			if (__get_user(c, &((unsigned char*)regs->ip)[i])) {
 bad:
 				printk(" Bad RIP value.");
 				break;
@@ -451,11 +451,11 @@ bad:
 	printk("\n");
 }	
 
-int is_valid_bugaddr(unsigned long rip)
+int is_valid_bugaddr(unsigned long ip)
 {
 	unsigned short ud2;
 
-	if (__copy_from_user(&ud2, (const void __user *) rip, sizeof(ud2)))
+	if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2)))
 		return 0;
 
 	return ud2 == 0x0b0f;
@@ -521,8 +521,8 @@ void __kprobes __die(const char * str, struct pt_regs * regs, long err)
 	add_taint(TAINT_DIE);
 	/* Executive summary in case the oops scrolled away */
 	printk(KERN_ALERT "RIP ");
-	printk_address(regs->rip); 
-	printk(" RSP <%016lx>\n", regs->rsp); 
+	printk_address(regs->ip);
+	printk(" RSP <%016lx>\n", regs->sp);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
 }
@@ -532,7 +532,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 	unsigned long flags = oops_begin();
 
 	if (!user_mode(regs))
-		report_bug(regs->rip, regs);
+		report_bug(regs->ip, regs);
 
 	__die(str, regs, err);
 	oops_end(flags);
@@ -582,9 +582,9 @@ static void __kprobes do_trap(int trapnr, int signr, char *str,
 		if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
 		    printk_ratelimit())
 			printk(KERN_INFO
-			       "%s[%d] trap %s rip:%lx rsp:%lx error:%lx\n",
+			       "%s[%d] trap %s ip:%lx sp:%lx error:%lx\n",
 			       tsk->comm, tsk->pid, str,
-			       regs->rip, regs->rsp, error_code); 
+			       regs->ip, regs->sp, error_code);
 
 		if (info)
 			force_sig_info(signr, info, tsk);
@@ -597,9 +597,9 @@ static void __kprobes do_trap(int trapnr, int signr, char *str,
 	/* kernel trap */ 
 	{	     
 		const struct exception_table_entry *fixup;
-		fixup = search_exception_tables(regs->rip);
+		fixup = search_exception_tables(regs->ip);
 		if (fixup)
-			regs->rip = fixup->fixup;
+			regs->ip = fixup->fixup;
 		else {
 			tsk->thread.error_code = error_code;
 			tsk->thread.trap_no = trapnr;
@@ -635,10 +635,10 @@ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
 	do_trap(trapnr, signr, str, regs, error_code, &info); \
 }
 
-DO_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->rip)
+DO_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->ip)
 DO_ERROR( 4, SIGSEGV, "overflow", overflow)
 DO_ERROR( 5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->rip)
+DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
 DO_ERROR( 7, SIGSEGV, "device not available", device_not_available)
 DO_ERROR( 9, SIGFPE,  "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
@@ -688,9 +688,9 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 		if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
 		    printk_ratelimit())
 			printk(KERN_INFO
-		       "%s[%d] general protection rip:%lx rsp:%lx error:%lx\n",
+		       "%s[%d] general protection ip:%lx sp:%lx error:%lx\n",
 			       tsk->comm, tsk->pid,
-			       regs->rip, regs->rsp, error_code); 
+			       regs->ip, regs->sp, error_code);
 
 		force_sig(SIGSEGV, tsk);
 		return;
@@ -699,9 +699,9 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 	/* kernel gp */
 	{
 		const struct exception_table_entry *fixup;
-		fixup = search_exception_tables(regs->rip);
+		fixup = search_exception_tables(regs->ip);
 		if (fixup) {
-			regs->rip = fixup->fixup;
+			regs->ip = fixup->fixup;
 			return;
 		}
 
@@ -824,15 +824,15 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
 	struct pt_regs *regs = eregs;
 	/* Did already sync */
-	if (eregs == (struct pt_regs *)eregs->rsp)
+	if (eregs == (struct pt_regs *)eregs->sp)
 		;
 	/* Exception from user space */
 	else if (user_mode(eregs))
 		regs = task_pt_regs(current);
 	/* Exception from kernel and interrupts are enabled. Move to
  	   kernel process stack. */
-	else if (eregs->eflags & X86_EFLAGS_IF)
-		regs = (struct pt_regs *)(eregs->rsp -= sizeof(struct pt_regs));
+	else if (eregs->flags & X86_EFLAGS_IF)
+		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
 	if (eregs != regs)
 		*regs = *eregs;
 	return regs;
@@ -887,7 +887,7 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs,
 	info.si_signo = SIGTRAP;
 	info.si_errno = 0;
 	info.si_code = TRAP_BRKPT;
-	info.si_addr = user_mode(regs) ? (void __user *)regs->rip : NULL;
+	info.si_addr = user_mode(regs) ? (void __user *)regs->ip : NULL;
 	force_sig_info(SIGTRAP, &info, tsk);
 
 clear_dr7:
@@ -897,16 +897,16 @@ clear_dr7:
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-	regs->eflags &= ~TF_MASK;
+	regs->flags &= ~TF_MASK;
 	preempt_conditional_cli(regs);
 }
 
 static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
 {
 	const struct exception_table_entry *fixup;
-	fixup = search_exception_tables(regs->rip);
+	fixup = search_exception_tables(regs->ip);
 	if (fixup) {
-		regs->rip = fixup->fixup;
+		regs->ip = fixup->fixup;
 		return 1;
 	}
 	notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE);
@@ -923,7 +923,7 @@ static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
  */
 asmlinkage void do_coprocessor_error(struct pt_regs *regs)
 {
-	void __user *rip = (void __user *)(regs->rip);
+	void __user *ip = (void __user *)(regs->ip);
 	struct task_struct * task;
 	siginfo_t info;
 	unsigned short cwd, swd;
@@ -943,7 +943,7 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs)
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;
 	info.si_code = __SI_FAULT;
-	info.si_addr = rip;
+	info.si_addr = ip;
 	/*
 	 * (~cwd & swd) will mask out exceptions that are not set to unmasked
 	 * status.  0x3f is the exception bits in these regs, 0x200 is the
@@ -992,7 +992,7 @@ asmlinkage void bad_intr(void)
 
 asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 {
-	void __user *rip = (void __user *)(regs->rip);
+	void __user *ip = (void __user *)(regs->ip);
 	struct task_struct * task;
 	siginfo_t info;
 	unsigned short mxcsr;
@@ -1012,7 +1012,7 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;
 	info.si_code = __SI_FAULT;
-	info.si_addr = rip;
+	info.si_addr = ip;
 	/*
 	 * The SIMD FPU exceptions are handled a little differently, as there
 	 * is only a single status/control register.  Thus, to determine which
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index 157e4bedd3c5..980e85b90091 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -70,10 +70,10 @@
 /*
  * 8- and 16-bit register defines..
  */
-#define AL(regs)	(((unsigned char *)&((regs)->pt.eax))[0])
-#define AH(regs)	(((unsigned char *)&((regs)->pt.eax))[1])
-#define IP(regs)	(*(unsigned short *)&((regs)->pt.eip))
-#define SP(regs)	(*(unsigned short *)&((regs)->pt.esp))
+#define AL(regs)	(((unsigned char *)&((regs)->pt.ax))[0])
+#define AH(regs)	(((unsigned char *)&((regs)->pt.ax))[1])
+#define IP(regs)	(*(unsigned short *)&((regs)->pt.ip))
+#define SP(regs)	(*(unsigned short *)&((regs)->pt.sp))
 
 /*
  * virtual flags (16 and 32-bit versions)
@@ -93,12 +93,12 @@ static int copy_vm86_regs_to_user(struct vm86_regs __user *user,
 {
 	int ret = 0;
 
-	/* kernel_vm86_regs is missing xgs, so copy everything up to
+	/* kernel_vm86_regs is missing gs, so copy everything up to
 	   (but not including) orig_eax, and then rest including orig_eax. */
-	ret += copy_to_user(user, regs, offsetof(struct kernel_vm86_regs, pt.orig_eax));
-	ret += copy_to_user(&user->orig_eax, &regs->pt.orig_eax,
+	ret += copy_to_user(user, regs, offsetof(struct kernel_vm86_regs, pt.orig_ax));
+	ret += copy_to_user(&user->orig_eax, &regs->pt.orig_ax,
 			    sizeof(struct kernel_vm86_regs) -
-			    offsetof(struct kernel_vm86_regs, pt.orig_eax));
+			    offsetof(struct kernel_vm86_regs, pt.orig_ax));
 
 	return ret;
 }
@@ -110,12 +110,12 @@ static int copy_vm86_regs_from_user(struct kernel_vm86_regs *regs,
 {
 	int ret = 0;
 
-	/* copy eax-xfs inclusive */
-	ret += copy_from_user(regs, user, offsetof(struct kernel_vm86_regs, pt.orig_eax));
-	/* copy orig_eax-__gsh+extra */
-	ret += copy_from_user(&regs->pt.orig_eax, &user->orig_eax,
+	/* copy ax-fs inclusive */
+	ret += copy_from_user(regs, user, offsetof(struct kernel_vm86_regs, pt.orig_ax));
+	/* copy orig_ax-__gsh+extra */
+	ret += copy_from_user(&regs->pt.orig_ax, &user->orig_eax,
 			      sizeof(struct kernel_vm86_regs) -
-			      offsetof(struct kernel_vm86_regs, pt.orig_eax) +
+			      offsetof(struct kernel_vm86_regs, pt.orig_ax) +
 			      extra);
 	return ret;
 }
@@ -138,7 +138,7 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs)
 		printk("no vm86_info: BAD\n");
 		do_exit(SIGSEGV);
 	}
-	set_flags(regs->pt.eflags, VEFLAGS, VIF_MASK | current->thread.v86mask);
+	set_flags(regs->pt.flags, VEFLAGS, VIF_MASK | current->thread.v86mask);
 	tmp = copy_vm86_regs_to_user(&current->thread.vm86_info->regs,regs);
 	tmp += put_user(current->thread.screen_bitmap,&current->thread.vm86_info->screen_bitmap);
 	if (tmp) {
@@ -155,7 +155,7 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs)
 
 	ret = KVM86->regs32;
 
-	ret->xfs = current->thread.saved_fs;
+	ret->fs = current->thread.saved_fs;
 	loadsegment(gs, current->thread.saved_gs);
 
 	return ret;
@@ -197,7 +197,7 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
 
 asmlinkage int sys_vm86old(struct pt_regs regs)
 {
-	struct vm86_struct __user *v86 = (struct vm86_struct __user *)regs.ebx;
+	struct vm86_struct __user *v86 = (struct vm86_struct __user *)regs.bx;
 	struct kernel_vm86_struct info; /* declare this _on top_,
 					 * this avoids wasting of stack space.
 					 * This remains on the stack until we
@@ -237,12 +237,12 @@ asmlinkage int sys_vm86(struct pt_regs regs)
 	struct vm86plus_struct __user *v86;
 
 	tsk = current;
-	switch (regs.ebx) {
+	switch (regs.bx) {
 		case VM86_REQUEST_IRQ:
 		case VM86_FREE_IRQ:
 		case VM86_GET_IRQ_BITS:
 		case VM86_GET_AND_RESET_IRQ:
-			ret = do_vm86_irq_handling(regs.ebx, (int)regs.ecx);
+			ret = do_vm86_irq_handling(regs.bx, (int)regs.cx);
 			goto out;
 		case VM86_PLUS_INSTALL_CHECK:
 			/* NOTE: on old vm86 stuff this will return the error
@@ -258,7 +258,7 @@ asmlinkage int sys_vm86(struct pt_regs regs)
 	ret = -EPERM;
 	if (tsk->thread.saved_esp0)
 		goto out;
-	v86 = (struct vm86plus_struct __user *)regs.ecx;
+	v86 = (struct vm86plus_struct __user *)regs.cx;
 	tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs,
 				       offsetof(struct kernel_vm86_struct, regs32) -
 				       sizeof(info.regs));
@@ -281,23 +281,23 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
 /*
  * make sure the vm86() system call doesn't try to do anything silly
  */
-	info->regs.pt.xds = 0;
-	info->regs.pt.xes = 0;
-	info->regs.pt.xfs = 0;
+	info->regs.pt.ds = 0;
+	info->regs.pt.es = 0;
+	info->regs.pt.fs = 0;
 
 /* we are clearing gs later just before "jmp resume_userspace",
  * because it is not saved/restored.
  */
 
 /*
- * The eflags register is also special: we cannot trust that the user
+ * The flags register is also special: we cannot trust that the user
  * has set it up safely, so this makes sure interrupt etc flags are
  * inherited from protected mode.
  */
- 	VEFLAGS = info->regs.pt.eflags;
-	info->regs.pt.eflags &= SAFE_MASK;
-	info->regs.pt.eflags |= info->regs32->eflags & ~SAFE_MASK;
-	info->regs.pt.eflags |= VM_MASK;
+	VEFLAGS = info->regs.pt.flags;
+	info->regs.pt.flags &= SAFE_MASK;
+	info->regs.pt.flags |= info->regs32->flags & ~SAFE_MASK;
+	info->regs.pt.flags |= VM_MASK;
 
 	switch (info->cpu_type) {
 		case CPU_286:
@@ -315,11 +315,11 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
 	}
 
 /*
- * Save old state, set default return value (%eax) to 0
+ * Save old state, set default return value (%ax) to 0
  */
-	info->regs32->eax = 0;
+	info->regs32->ax = 0;
 	tsk->thread.saved_esp0 = tsk->thread.esp0;
-	tsk->thread.saved_fs = info->regs32->xfs;
+	tsk->thread.saved_fs = info->regs32->fs;
 	savesegment(gs, tsk->thread.saved_gs);
 
 	tss = &per_cpu(init_tss, get_cpu());
@@ -352,7 +352,7 @@ static inline void return_to_32bit(struct kernel_vm86_regs * regs16, int retval)
 	struct pt_regs * regs32;
 
 	regs32 = save_v86_state(regs16);
-	regs32->eax = retval;
+	regs32->ax = retval;
 	__asm__ __volatile__("movl %0,%%esp\n\t"
 		"movl %1,%%ebp\n\t"
 		"jmp resume_userspace"
@@ -373,12 +373,12 @@ static inline void clear_IF(struct kernel_vm86_regs * regs)
 
 static inline void clear_TF(struct kernel_vm86_regs * regs)
 {
-	regs->pt.eflags &= ~TF_MASK;
+	regs->pt.flags &= ~TF_MASK;
 }
 
 static inline void clear_AC(struct kernel_vm86_regs * regs)
 {
-	regs->pt.eflags &= ~AC_MASK;
+	regs->pt.flags &= ~AC_MASK;
 }
 
 /* It is correct to call set_IF(regs) from the set_vflags_*
@@ -392,11 +392,11 @@ static inline void clear_AC(struct kernel_vm86_regs * regs)
  * [KD]
  */
 
-static inline void set_vflags_long(unsigned long eflags, struct kernel_vm86_regs * regs)
+static inline void set_vflags_long(unsigned long flags, struct kernel_vm86_regs * regs)
 {
-	set_flags(VEFLAGS, eflags, current->thread.v86mask);
-	set_flags(regs->pt.eflags, eflags, SAFE_MASK);
-	if (eflags & IF_MASK)
+	set_flags(VEFLAGS, flags, current->thread.v86mask);
+	set_flags(regs->pt.flags, flags, SAFE_MASK);
+	if (flags & IF_MASK)
 		set_IF(regs);
 	else
 		clear_IF(regs);
@@ -405,7 +405,7 @@ static inline void set_vflags_long(unsigned long eflags, struct kernel_vm86_regs
 static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_regs * regs)
 {
 	set_flags(VFLAGS, flags, current->thread.v86mask);
-	set_flags(regs->pt.eflags, flags, SAFE_MASK);
+	set_flags(regs->pt.flags, flags, SAFE_MASK);
 	if (flags & IF_MASK)
 		set_IF(regs);
 	else
@@ -414,7 +414,7 @@ static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_reg
 
 static inline unsigned long get_vflags(struct kernel_vm86_regs * regs)
 {
-	unsigned long flags = regs->pt.eflags & RETURN_MASK;
+	unsigned long flags = regs->pt.flags & RETURN_MASK;
 
 	if (VEFLAGS & VIF_MASK)
 		flags |= IF_MASK;
@@ -518,7 +518,7 @@ static void do_int(struct kernel_vm86_regs *regs, int i,
 	unsigned long __user *intr_ptr;
 	unsigned long segoffs;
 
-	if (regs->pt.xcs == BIOSSEG)
+	if (regs->pt.cs == BIOSSEG)
 		goto cannot_handle;
 	if (is_revectored(i, &KVM86->int_revectored))
 		goto cannot_handle;
@@ -530,9 +530,9 @@ static void do_int(struct kernel_vm86_regs *regs, int i,
 	if ((segoffs >> 16) == BIOSSEG)
 		goto cannot_handle;
 	pushw(ssp, sp, get_vflags(regs), cannot_handle);
-	pushw(ssp, sp, regs->pt.xcs, cannot_handle);
+	pushw(ssp, sp, regs->pt.cs, cannot_handle);
 	pushw(ssp, sp, IP(regs), cannot_handle);
-	regs->pt.xcs = segoffs >> 16;
+	regs->pt.cs = segoffs >> 16;
 	SP(regs) -= 6;
 	IP(regs) = segoffs & 0xffff;
 	clear_TF(regs);
@@ -549,7 +549,7 @@ int handle_vm86_trap(struct kernel_vm86_regs * regs, long error_code, int trapno
 	if (VMPI.is_vm86pus) {
 		if ( (trapno==3) || (trapno==1) )
 			return_to_32bit(regs, VM86_TRAP + (trapno << 8));
-		do_int(regs, trapno, (unsigned char __user *) (regs->pt.xss << 4), SP(regs));
+		do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs));
 		return 0;
 	}
 	if (trapno !=1)
@@ -585,10 +585,10 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code)
 		handle_vm86_trap(regs, 0, 1); \
 	return; } while (0)
 
-	orig_flags = *(unsigned short *)&regs->pt.eflags;
+	orig_flags = *(unsigned short *)&regs->pt.flags;
 
-	csp = (unsigned char __user *) (regs->pt.xcs << 4);
-	ssp = (unsigned char __user *) (regs->pt.xss << 4);
+	csp = (unsigned char __user *) (regs->pt.cs << 4);
+	ssp = (unsigned char __user *) (regs->pt.ss << 4);
 	sp = SP(regs);
 	ip = IP(regs);
 
@@ -675,7 +675,7 @@ void handle_vm86_fault(struct kernel_vm86_regs * regs, long error_code)
 			SP(regs) += 6;
 		}
 		IP(regs) = newip;
-		regs->pt.xcs = newcs;
+		regs->pt.cs = newcs;
 		CHECK_IF_IN_TRAP;
 		if (data32) {
 			set_vflags_long(newflags, regs);
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 87e5633805a9..599b6f2ed562 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -88,13 +88,13 @@ struct vmi_timer_ops vmi_timer_ops;
 #define IRQ_PATCH_DISABLE  5
 
 static inline void patch_offset(void *insnbuf,
-				unsigned long eip, unsigned long dest)
+				unsigned long ip, unsigned long dest)
 {
-        *(unsigned long *)(insnbuf+1) = dest-eip-5;
+        *(unsigned long *)(insnbuf+1) = dest-ip-5;
 }
 
 static unsigned patch_internal(int call, unsigned len, void *insnbuf,
-			       unsigned long eip)
+			       unsigned long ip)
 {
 	u64 reloc;
 	struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc;
@@ -103,13 +103,13 @@ static unsigned patch_internal(int call, unsigned len, void *insnbuf,
 		case VMI_RELOCATION_CALL_REL:
 			BUG_ON(len < 5);
 			*(char *)insnbuf = MNEM_CALL;
-			patch_offset(insnbuf, eip, (unsigned long)rel->eip);
+			patch_offset(insnbuf, ip, (unsigned long)rel->eip);
 			return 5;
 
 		case VMI_RELOCATION_JUMP_REL:
 			BUG_ON(len < 5);
 			*(char *)insnbuf = MNEM_JMP;
-			patch_offset(insnbuf, eip, (unsigned long)rel->eip);
+			patch_offset(insnbuf, ip, (unsigned long)rel->eip);
 			return 5;
 
 		case VMI_RELOCATION_NOP:
@@ -131,25 +131,25 @@ static unsigned patch_internal(int call, unsigned len, void *insnbuf,
  * sequence.  The callee does nop padding for us.
  */
 static unsigned vmi_patch(u8 type, u16 clobbers, void *insns,
-			  unsigned long eip, unsigned len)
+			  unsigned long ip, unsigned len)
 {
 	switch (type) {
 		case PARAVIRT_PATCH(pv_irq_ops.irq_disable):
 			return patch_internal(VMI_CALL_DisableInterrupts, len,
-					      insns, eip);
+					      insns, ip);
 		case PARAVIRT_PATCH(pv_irq_ops.irq_enable):
 			return patch_internal(VMI_CALL_EnableInterrupts, len,
-					      insns, eip);
+					      insns, ip);
 		case PARAVIRT_PATCH(pv_irq_ops.restore_fl):
 			return patch_internal(VMI_CALL_SetInterruptMask, len,
-					      insns, eip);
+					      insns, ip);
 		case PARAVIRT_PATCH(pv_irq_ops.save_fl):
 			return patch_internal(VMI_CALL_GetInterruptMask, len,
-					      insns, eip);
+					      insns, ip);
 		case PARAVIRT_PATCH(pv_cpu_ops.iret):
-			return patch_internal(VMI_CALL_IRET, len, insns, eip);
+			return patch_internal(VMI_CALL_IRET, len, insns, ip);
 		case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_syscall_ret):
-			return patch_internal(VMI_CALL_SYSEXIT, len, insns, eip);
+			return patch_internal(VMI_CALL_SYSEXIT, len, insns, ip);
 		default:
 			break;
 	}
@@ -157,29 +157,29 @@ static unsigned vmi_patch(u8 type, u16 clobbers, void *insns,
 }
 
 /* CPUID has non-C semantics, and paravirt-ops API doesn't match hardware ISA */
-static void vmi_cpuid(unsigned int *eax, unsigned int *ebx,
-                               unsigned int *ecx, unsigned int *edx)
+static void vmi_cpuid(unsigned int *ax, unsigned int *bx,
+                               unsigned int *cx, unsigned int *dx)
 {
 	int override = 0;
-	if (*eax == 1)
+	if (*ax == 1)
 		override = 1;
         asm volatile ("call *%6"
-                      : "=a" (*eax),
-                        "=b" (*ebx),
-                        "=c" (*ecx),
-                        "=d" (*edx)
-                      : "0" (*eax), "2" (*ecx), "r" (vmi_ops.cpuid));
+                      : "=a" (*ax),
+                        "=b" (*bx),
+                        "=c" (*cx),
+                        "=d" (*dx)
+                      : "0" (*ax), "2" (*cx), "r" (vmi_ops.cpuid));
 	if (override) {
 		if (disable_pse)
-			*edx &= ~X86_FEATURE_PSE;
+			*dx &= ~X86_FEATURE_PSE;
 		if (disable_pge)
-			*edx &= ~X86_FEATURE_PGE;
+			*dx &= ~X86_FEATURE_PGE;
 		if (disable_sep)
-			*edx &= ~X86_FEATURE_SEP;
+			*dx &= ~X86_FEATURE_SEP;
 		if (disable_tsc)
-			*edx &= ~X86_FEATURE_TSC;
+			*dx &= ~X86_FEATURE_TSC;
 		if (disable_mtrr)
-			*edx &= ~X86_FEATURE_MTRR;
+			*dx &= ~X86_FEATURE_MTRR;
 	}
 }
 
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index ad4005c6d4a1..018f7cf33790 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -43,7 +43,7 @@
 #include <asm/vgtod.h>
 
 #define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
-#define __syscall_clobber "r11","rcx","memory"
+#define __syscall_clobber "r11","cx","memory"
 #define __pa_vsymbol(x)			\
 	({unsigned long v;  		\
 	extern char __vsyscall_0; 	\
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index df04bf884dd4..ea46d05853bb 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -175,8 +175,8 @@ static void lguest_leave_lazy_mode(void)
  * check there when it wants to deliver an interrupt.
  */
 
-/* save_flags() is expected to return the processor state (ie. "eflags").  The
- * eflags word contains all kind of stuff, but in practice Linux only cares
+/* save_flags() is expected to return the processor state (ie. "flags").  The
+ * flags word contains all kind of stuff, but in practice Linux only cares
  * about the interrupt flag.  Our "save_flags()" just returns that. */
 static unsigned long save_fl(void)
 {
@@ -323,30 +323,30 @@ static void lguest_load_tr_desc(void)
  * anyone (including userspace) can just use the raw "cpuid" instruction and
  * the Host won't even notice since it isn't privileged.  So we try not to get
  * too worked up about it. */
-static void lguest_cpuid(unsigned int *eax, unsigned int *ebx,
-			 unsigned int *ecx, unsigned int *edx)
+static void lguest_cpuid(unsigned int *ax, unsigned int *bx,
+			 unsigned int *cx, unsigned int *dx)
 {
-	int function = *eax;
+	int function = *ax;
 
-	native_cpuid(eax, ebx, ecx, edx);
+	native_cpuid(ax, bx, cx, dx);
 	switch (function) {
 	case 1:	/* Basic feature request. */
 		/* We only allow kernel to see SSE3, CMPXCHG16B and SSSE3 */
-		*ecx &= 0x00002201;
+		*cx &= 0x00002201;
 		/* SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, FPU. */
-		*edx &= 0x07808101;
+		*dx &= 0x07808101;
 		/* The Host can do a nice optimization if it knows that the
 		 * kernel mappings (addresses above 0xC0000000 or whatever
 		 * PAGE_OFFSET is set to) haven't changed.  But Linux calls
 		 * flush_tlb_user() for both user and kernel mappings unless
 		 * the Page Global Enable (PGE) feature bit is set. */
-		*edx |= 0x00002000;
+		*dx |= 0x00002000;
 		break;
 	case 0x80000000:
 		/* Futureproof this a little: if they ask how much extended
 		 * processor information there is, limit it to known fields. */
-		if (*eax > 0x80000008)
-			*eax = 0x80000008;
+		if (*ax > 0x80000008)
+			*ax = 0x80000008;
 		break;
 	}
 }
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index b472a2df0b7f..f2c13482acc0 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -526,7 +526,7 @@ static void __init do_boot_cpu(__u8 cpu)
 	 * initial kernel stack.  We need to alter this to give the
 	 * booting CPU a new stack (taken from its idle process) */
 	extern struct {
-		__u8 *esp;
+		__u8 *sp;
 		unsigned short ss;
 	} stack_start;
 	/* This is the format of the CPI IDT gate (in real mode) which
@@ -555,9 +555,9 @@ static void __init do_boot_cpu(__u8 cpu)
 	idle = fork_idle(cpu);
 	if (IS_ERR(idle))
 		panic("failed fork for CPU%d", cpu);
-	idle->thread.eip = (unsigned long)start_secondary;
+	idle->thread.ip = (unsigned long)start_secondary;
 	/* init_tasks (in sched.c) is indexed logically */
-	stack_start.esp = (void *)idle->thread.esp;
+	stack_start.sp = (void *)idle->thread.sp;
 
 	init_gdt(cpu);
 	per_cpu(current_task, cpu) = idle;
@@ -567,7 +567,7 @@ static void __init do_boot_cpu(__u8 cpu)
 	/* Note: Don't modify initial ss override */
 	VDEBUG(("VOYAGER SMP: Booting CPU%d at 0x%lx[%x:%x], stack %p\n", cpu,
 		(unsigned long)hijack_source.val, hijack_source.idt.Segment,
-		hijack_source.idt.Offset, stack_start.esp));
+		hijack_source.idt.Offset, stack_start.sp));
 
 	/* init lowmem identity mapping */
 	clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
@@ -745,8 +745,8 @@ void __init initialize_secondary(void)
 	 */
 
 	asm volatile ("movl %0,%%esp\n\t"
-		      "jmp *%1"::"r" (current->thread.esp),
-		      "r"(current->thread.eip));
+		      "jmp *%1"::"r" (current->thread.sp),
+		      "r"(current->thread.ip));
 }
 
 /* handle a Voyager SYS_INT -- If we don't, the base board will
diff --git a/arch/x86/mm/extable_32.c b/arch/x86/mm/extable_32.c
index 0ce4f22a2635..41685461f8b2 100644
--- a/arch/x86/mm/extable_32.c
+++ b/arch/x86/mm/extable_32.c
@@ -11,7 +11,7 @@ int fixup_exception(struct pt_regs *regs)
 	const struct exception_table_entry *fixup;
 
 #ifdef CONFIG_PNPBIOS
-	if (unlikely(SEGMENT_IS_PNP_CODE(regs->xcs)))
+	if (unlikely(SEGMENT_IS_PNP_CODE(regs->cs)))
 	{
 		extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp;
 		extern u32 pnp_bios_is_utter_crap;
@@ -25,9 +25,9 @@ int fixup_exception(struct pt_regs *regs)
 	}
 #endif
 
-	fixup = search_exception_tables(regs->eip);
+	fixup = search_exception_tables(regs->ip);
 	if (fixup) {
-		regs->eip = fixup->fixup;
+		regs->ip = fixup->fixup;
 		return 1;
 	}
 
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index a2273d44aa27..6056c6d71835 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -72,15 +72,15 @@ static inline int notify_page_fault(struct pt_regs *regs)
 static inline unsigned long get_segment_eip(struct pt_regs *regs,
 					    unsigned long *eip_limit)
 {
-	unsigned long eip = regs->eip;
-	unsigned seg = regs->xcs & 0xffff;
+	unsigned long ip = regs->ip;
+	unsigned seg = regs->cs & 0xffff;
 	u32 seg_ar, seg_limit, base, *desc;
 
 	/* Unlikely, but must come before segment checks. */
-	if (unlikely(regs->eflags & VM_MASK)) {
+	if (unlikely(regs->flags & VM_MASK)) {
 		base = seg << 4;
 		*eip_limit = base + 0xffff;
-		return base + (eip & 0xffff);
+		return base + (ip & 0xffff);
 	}
 
 	/* The standard kernel/user address space limit. */
@@ -88,16 +88,16 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs,
 	
 	/* By far the most common cases. */
 	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
-		return eip;
+		return ip;
 
 	/* Check the segment exists, is within the current LDT/GDT size,
 	   that kernel/user (ring 0..3) has the appropriate privilege,
 	   that it's a code segment, and get the limit. */
 	__asm__ ("larl %3,%0; lsll %3,%1"
 		 : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
-	if ((~seg_ar & 0x9800) || eip > seg_limit) {
+	if ((~seg_ar & 0x9800) || ip > seg_limit) {
 		*eip_limit = 0;
-		return 1;	 /* So that returned eip > *eip_limit. */
+		return 1;	 /* So that returned ip > *eip_limit. */
 	}
 
 	/* Get the GDT/LDT descriptor base. 
@@ -127,7 +127,7 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs,
 	seg_limit += base;
 	if (seg_limit < *eip_limit && seg_limit >= base)
 		*eip_limit = seg_limit;
-	return eip + base;
+	return ip + base;
 }
 
 /* 
@@ -345,7 +345,7 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 
 	/* It's safe to allow irq's after cr2 has been saved and the vmalloc
 	   fault has been handled. */
-	if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
+	if (regs->flags & (X86_EFLAGS_IF|VM_MASK))
 		local_irq_enable();
 
 	mm = tsk->mm;
@@ -374,7 +374,7 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 	 */
 	if (!down_read_trylock(&mm->mmap_sem)) {
 		if ((error_code & 4) == 0 &&
-		    !search_exception_tables(regs->eip))
+		    !search_exception_tables(regs->ip))
 			goto bad_area_nosemaphore;
 		down_read(&mm->mmap_sem);
 	}
@@ -388,12 +388,12 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
 		goto bad_area;
 	if (error_code & 4) {
 		/*
-		 * Accessing the stack below %esp is always a bug.
+		 * Accessing the stack below %sp is always a bug.
 		 * The large cushion allows instructions like enter
 		 * and pusha to work.  ("enter $65535,$31" pushes
-		 * 32 pointers and then decrements %esp by 65535.)
+		 * 32 pointers and then decrements %sp by 65535.)
 		 */
-		if (address + 65536 + 32 * sizeof(unsigned long) < regs->esp)
+		if (address + 65536 + 32 * sizeof(unsigned long) < regs->sp)
 			goto bad_area;
 	}
 	if (expand_stack(vma, address))
@@ -442,7 +442,7 @@ good_area:
 	/*
 	 * Did it hit the DOS screen memory VA from vm86 mode?
 	 */
-	if (regs->eflags & VM_MASK) {
+	if (regs->flags & VM_MASK) {
 		unsigned long bit = (address - 0xA0000) >> PAGE_SHIFT;
 		if (bit < 32)
 			tsk->thread.screen_bitmap |= 1 << bit;
@@ -474,11 +474,11 @@ bad_area_nosemaphore:
 
 		if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
 		    printk_ratelimit()) {
-			printk("%s%s[%d]: segfault at %08lx eip %08lx "
-			    "esp %08lx error %lx\n",
+			printk("%s%s[%d]: segfault at %08lx ip %08lx "
+			    "sp %08lx error %lx\n",
 			    task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
-			    tsk->comm, task_pid_nr(tsk), address, regs->eip,
-			    regs->esp, error_code);
+			    tsk->comm, task_pid_nr(tsk), address, regs->ip,
+			    regs->sp, error_code);
 		}
 		tsk->thread.cr2 = address;
 		/* Kernel addresses are always protection faults */
@@ -544,7 +544,7 @@ no_context:
 			printk(KERN_ALERT "BUG: unable to handle kernel paging"
 					" request");
 		printk(" at virtual address %08lx\n",address);
-		printk(KERN_ALERT "printing eip: %08lx ", regs->eip);
+		printk(KERN_ALERT "printing ip: %08lx ", regs->ip);
 
 		page = read_cr3();
 		page = ((__typeof__(page) *) __va(page))[address >> PGDIR_SHIFT];
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 0e26230669ca..88a7abda29ce 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -198,7 +198,7 @@ KERN_ERR "******* Disabling USB legacy in the BIOS may also help.\n";
 static int is_errata93(struct pt_regs *regs, unsigned long address) 
 {
 	static int warned;
-	if (address != regs->rip)
+	if (address != regs->ip)
 		return 0;
 	if ((address >> 32) != 0) 
 		return 0;
@@ -209,7 +209,7 @@ static int is_errata93(struct pt_regs *regs, unsigned long address)
 			printk(errata93_warning); 		
 			warned = 1;
 		}
-		regs->rip = address;
+		regs->ip = address;
 		return 1;
 	}
 	return 0;
@@ -355,7 +355,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 	if (notify_page_fault(regs))
 		return;
 
-	if (likely(regs->eflags & X86_EFLAGS_IF))
+	if (likely(regs->flags & X86_EFLAGS_IF))
 		local_irq_enable();
 
 	if (unlikely(error_code & PF_RSVD))
@@ -393,7 +393,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 	 */
 	if (!down_read_trylock(&mm->mmap_sem)) {
 		if ((error_code & PF_USER) == 0 &&
-		    !search_exception_tables(regs->rip))
+		    !search_exception_tables(regs->ip))
 			goto bad_area_nosemaphore;
 		down_read(&mm->mmap_sem);
 	}
@@ -409,7 +409,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 		/* Allow userspace just enough access below the stack pointer
 		 * to let the 'enter' instruction work.
 		 */
-		if (address + 65536 + 32 * sizeof(unsigned long) < regs->rsp)
+		if (address + 65536 + 32 * sizeof(unsigned long) < regs->sp)
 			goto bad_area;
 	}
 	if (expand_stack(vma, address))
@@ -488,10 +488,10 @@ bad_area_nosemaphore:
 		if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
 		    printk_ratelimit()) {
 			printk(
-		       "%s%s[%d]: segfault at %lx rip %lx rsp %lx error %lx\n",
+		       "%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx\n",
 					tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
-					tsk->comm, tsk->pid, address, regs->rip,
-					regs->rsp, error_code);
+					tsk->comm, tsk->pid, address, regs->ip,
+					regs->sp, error_code);
 		}
        
 		tsk->thread.cr2 = address;
@@ -509,9 +509,9 @@ bad_area_nosemaphore:
 no_context:
 	
 	/* Are we prepared to handle this kernel fault?  */
-	fixup = search_exception_tables(regs->rip);
+	fixup = search_exception_tables(regs->ip);
 	if (fixup) {
-		regs->rip = fixup->fixup;
+		regs->ip = fixup->fixup;
 		return;
 	}
 
@@ -537,7 +537,7 @@ no_context:
 	else
 		printk(KERN_ALERT "Unable to handle kernel paging request");
 	printk(" at %016lx RIP: \n" KERN_ALERT,address);
-	printk_address(regs->rip);
+	printk_address(regs->ip);
 	dump_pagetable(address);
 	tsk->thread.cr2 = address;
 	tsk->thread.trap_no = 14;
diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c
index 0ed046a187f7..cc353a0b183e 100644
--- a/arch/x86/oprofile/backtrace.c
+++ b/arch/x86/oprofile/backtrace.c
@@ -48,7 +48,7 @@ static struct stacktrace_ops backtrace_ops = {
 };
 
 struct frame_head {
-	struct frame_head *ebp;
+	struct frame_head *bp;
 	unsigned long ret;
 } __attribute__((packed));
 
@@ -67,10 +67,10 @@ dump_user_backtrace(struct frame_head * head)
 
 	/* frame pointers should strictly progress back up the stack
 	 * (towards higher addresses) */
-	if (head >= bufhead[0].ebp)
+	if (head >= bufhead[0].bp)
 		return NULL;
 
-	return bufhead[0].ebp;
+	return bufhead[0].bp;
 }
 
 void
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index d3574485cb15..29517faaa735 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -141,8 +141,8 @@ static void __init xen_banner(void)
 	printk(KERN_INFO "Hypervisor signature: %s\n", xen_start_info->magic);
 }
 
-static void xen_cpuid(unsigned int *eax, unsigned int *ebx,
-		      unsigned int *ecx, unsigned int *edx)
+static void xen_cpuid(unsigned int *ax, unsigned int *bx,
+		      unsigned int *cx, unsigned int *dx)
 {
 	unsigned maskedx = ~0;
 
@@ -150,18 +150,18 @@ static void xen_cpuid(unsigned int *eax, unsigned int *ebx,
 	 * Mask out inconvenient features, to try and disable as many
 	 * unsupported kernel subsystems as possible.
 	 */
-	if (*eax == 1)
+	if (*ax == 1)
 		maskedx = ~((1 << X86_FEATURE_APIC) |  /* disable APIC */
 			    (1 << X86_FEATURE_ACPI) |  /* disable ACPI */
 			    (1 << X86_FEATURE_ACC));   /* thermal monitoring */
 
 	asm(XEN_EMULATE_PREFIX "cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (*eax), "2" (*ecx));
-	*edx &= maskedx;
+		: "=a" (*ax),
+		  "=b" (*bx),
+		  "=c" (*cx),
+		  "=d" (*dx)
+		: "0" (*ax), "2" (*cx));
+	*dx &= maskedx;
 }
 
 static void xen_set_debugreg(int reg, unsigned long val)
diff --git a/arch/x86/xen/events.c b/arch/x86/xen/events.c
index 6d1da5809e6f..aebab9704dd7 100644
--- a/arch/x86/xen/events.c
+++ b/arch/x86/xen/events.c
@@ -487,7 +487,7 @@ fastcall void xen_evtchn_do_upcall(struct pt_regs *regs)
 			int irq = evtchn_to_irq[port];
 
 			if (irq != -1) {
-				regs->orig_eax = ~irq;
+				regs->orig_ax = ~irq;
 				do_IRQ(regs);
 			}
 		}
diff --git a/include/asm-x86/compat.h b/include/asm-x86/compat.h
index 66ba7987184a..b270ee04959e 100644
--- a/include/asm-x86/compat.h
+++ b/include/asm-x86/compat.h
@@ -207,7 +207,7 @@ static inline compat_uptr_t ptr_to_compat(void __user *uptr)
 static __inline__ void __user *compat_alloc_user_space(long len)
 {
 	struct pt_regs *regs = task_pt_regs(current);
-	return (void __user *)regs->rsp - len; 
+	return (void __user *)regs->sp - len;
 }
 
 static inline int is_compat_task(void)
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index 60f5101d9483..5e5705bf082a 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -99,32 +99,32 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
    just to make things more deterministic.
  */
 #define ELF_PLAT_INIT(_r, load_addr)	do { \
-	_r->ebx = 0; _r->ecx = 0; _r->edx = 0; \
-	_r->esi = 0; _r->edi = 0; _r->ebp = 0; \
-	_r->eax = 0; \
+	_r->bx = 0; _r->cx = 0; _r->dx = 0; \
+	_r->si = 0; _r->di = 0; _r->bp = 0; \
+	_r->ax = 0; \
 } while (0)
 
 /* regs is struct pt_regs, pr_reg is elf_gregset_t (which is
    now struct_user_regs, they are different) */
 
 #define ELF_CORE_COPY_REGS(pr_reg, regs)		\
-	pr_reg[0] = regs->ebx;				\
-	pr_reg[1] = regs->ecx;				\
-	pr_reg[2] = regs->edx;				\
-	pr_reg[3] = regs->esi;				\
-	pr_reg[4] = regs->edi;				\
-	pr_reg[5] = regs->ebp;				\
-	pr_reg[6] = regs->eax;				\
-	pr_reg[7] = regs->xds & 0xffff;			\
-	pr_reg[8] = regs->xes & 0xffff;			\
-	pr_reg[9] = regs->xfs & 0xffff;			\
+	pr_reg[0] = regs->bx;				\
+	pr_reg[1] = regs->cx;				\
+	pr_reg[2] = regs->dx;				\
+	pr_reg[3] = regs->si;				\
+	pr_reg[4] = regs->di;				\
+	pr_reg[5] = regs->bp;				\
+	pr_reg[6] = regs->ax;				\
+	pr_reg[7] = regs->ds & 0xffff;			\
+	pr_reg[8] = regs->es & 0xffff;			\
+	pr_reg[9] = regs->fs & 0xffff;			\
 	savesegment(gs,pr_reg[10]);			\
-	pr_reg[11] = regs->orig_eax;			\
-	pr_reg[12] = regs->eip;				\
-	pr_reg[13] = regs->xcs & 0xffff;		\
-	pr_reg[14] = regs->eflags;			\
-	pr_reg[15] = regs->esp;				\
-	pr_reg[16] = regs->xss & 0xffff;
+	pr_reg[11] = regs->orig_ax;			\
+	pr_reg[12] = regs->ip;				\
+	pr_reg[13] = regs->cs & 0xffff;			\
+	pr_reg[14] = regs->flags;			\
+	pr_reg[15] = regs->sp;				\
+	pr_reg[16] = regs->ss & 0xffff;
 
 #define ELF_PLATFORM	(utsname()->machine)
 #define set_personality_64bit()	do { } while (0)
@@ -142,9 +142,9 @@ extern unsigned int vdso_enabled;
 
 #define ELF_PLAT_INIT(_r, load_addr)	do {		  \
 	struct task_struct *cur = current;		  \
-	(_r)->rbx = 0; (_r)->rcx = 0; (_r)->rdx = 0;	  \
-	(_r)->rsi = 0; (_r)->rdi = 0; (_r)->rbp = 0;	  \
-	(_r)->rax = 0;					  \
+	(_r)->bx = 0; (_r)->cx = 0; (_r)->dx = 0;	  \
+	(_r)->si = 0; (_r)->di = 0; (_r)->bp = 0;	  \
+	(_r)->ax = 0;					  \
 	(_r)->r8 = 0;					  \
 	(_r)->r9 = 0;					  \
 	(_r)->r10 = 0;					  \
@@ -169,22 +169,22 @@ extern unsigned int vdso_enabled;
 	(pr_reg)[1] = (regs)->r14;				\
 	(pr_reg)[2] = (regs)->r13;				\
 	(pr_reg)[3] = (regs)->r12;				\
-	(pr_reg)[4] = (regs)->rbp;				\
-	(pr_reg)[5] = (regs)->rbx;				\
+	(pr_reg)[4] = (regs)->bp;				\
+	(pr_reg)[5] = (regs)->bx;				\
 	(pr_reg)[6] = (regs)->r11;				\
 	(pr_reg)[7] = (regs)->r10;				\
 	(pr_reg)[8] = (regs)->r9;				\
 	(pr_reg)[9] = (regs)->r8;				\
-	(pr_reg)[10] = (regs)->rax;				\
-	(pr_reg)[11] = (regs)->rcx;				\
-	(pr_reg)[12] = (regs)->rdx;				\
-	(pr_reg)[13] = (regs)->rsi;				\
-	(pr_reg)[14] = (regs)->rdi;				\
-	(pr_reg)[15] = (regs)->orig_rax;			\
-	(pr_reg)[16] = (regs)->rip;				\
+	(pr_reg)[10] = (regs)->ax;				\
+	(pr_reg)[11] = (regs)->cx;				\
+	(pr_reg)[12] = (regs)->dx;				\
+	(pr_reg)[13] = (regs)->si;				\
+	(pr_reg)[14] = (regs)->di;				\
+	(pr_reg)[15] = (regs)->orig_ax;			\
+	(pr_reg)[16] = (regs)->ip;				\
 	(pr_reg)[17] = (regs)->cs;				\
-	(pr_reg)[18] = (regs)->eflags;				\
-	(pr_reg)[19] = (regs)->rsp;				\
+	(pr_reg)[18] = (regs)->flags;				\
+	(pr_reg)[19] = (regs)->sp;				\
 	(pr_reg)[20] = (regs)->ss;				\
 	(pr_reg)[21] = current->thread.fs;			\
 	(pr_reg)[22] = current->thread.gs;			\
diff --git a/include/asm-x86/kexec_32.h b/include/asm-x86/kexec_32.h
index 4b9dc9e6b701..ff39d2f88022 100644
--- a/include/asm-x86/kexec_32.h
+++ b/include/asm-x86/kexec_32.h
@@ -45,7 +45,7 @@
 /* We can also handle crash dumps from 64 bit kernel. */
 #define vmcore_elf_check_arch_cross(x) ((x)->e_machine == EM_X86_64)
 
-/* CPU does not save ss and esp on stack if execution is already
+/* CPU does not save ss and sp on stack if execution is already
  * running in kernel mode at the time of NMI occurrence. This code
  * fixes it.
  */
@@ -53,16 +53,16 @@ static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
 					struct pt_regs *oldregs)
 {
 	memcpy(newregs, oldregs, sizeof(*newregs));
-	newregs->esp = (unsigned long)&(oldregs->esp);
+	newregs->sp = (unsigned long)&(oldregs->sp);
 	__asm__ __volatile__(
 			"xorl %%eax, %%eax\n\t"
 			"movw %%ss, %%ax\n\t"
-			:"=a"(newregs->xss));
+			:"=a"(newregs->ss));
 }
 
 /*
  * This function is responsible for capturing register states if coming
- * via panic otherwise just fix up the ss and esp if coming via kernel
+ * via panic otherwise just fix up the ss and sp if coming via kernel
  * mode exception.
  */
 static inline void crash_setup_regs(struct pt_regs *newregs,
@@ -71,21 +71,21 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
        if (oldregs)
                crash_fixup_ss_esp(newregs, oldregs);
        else {
-               __asm__ __volatile__("movl %%ebx,%0" : "=m"(newregs->ebx));
-               __asm__ __volatile__("movl %%ecx,%0" : "=m"(newregs->ecx));
-               __asm__ __volatile__("movl %%edx,%0" : "=m"(newregs->edx));
-               __asm__ __volatile__("movl %%esi,%0" : "=m"(newregs->esi));
-               __asm__ __volatile__("movl %%edi,%0" : "=m"(newregs->edi));
-               __asm__ __volatile__("movl %%ebp,%0" : "=m"(newregs->ebp));
-               __asm__ __volatile__("movl %%eax,%0" : "=m"(newregs->eax));
-               __asm__ __volatile__("movl %%esp,%0" : "=m"(newregs->esp));
-               __asm__ __volatile__("movw %%ss, %%ax;" :"=a"(newregs->xss));
-               __asm__ __volatile__("movw %%cs, %%ax;" :"=a"(newregs->xcs));
-               __asm__ __volatile__("movw %%ds, %%ax;" :"=a"(newregs->xds));
-               __asm__ __volatile__("movw %%es, %%ax;" :"=a"(newregs->xes));
-               __asm__ __volatile__("pushfl; popl %0" :"=m"(newregs->eflags));
+               __asm__ __volatile__("movl %%ebx,%0" : "=m"(newregs->bx));
+               __asm__ __volatile__("movl %%ecx,%0" : "=m"(newregs->cx));
+               __asm__ __volatile__("movl %%edx,%0" : "=m"(newregs->dx));
+               __asm__ __volatile__("movl %%esi,%0" : "=m"(newregs->si));
+               __asm__ __volatile__("movl %%edi,%0" : "=m"(newregs->di));
+               __asm__ __volatile__("movl %%ebp,%0" : "=m"(newregs->bp));
+               __asm__ __volatile__("movl %%eax,%0" : "=m"(newregs->ax));
+               __asm__ __volatile__("movl %%esp,%0" : "=m"(newregs->sp));
+               __asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
+               __asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
+               __asm__ __volatile__("movl %%ds, %%eax;" :"=a"(newregs->ds));
+               __asm__ __volatile__("movl %%es, %%eax;" :"=a"(newregs->es));
+               __asm__ __volatile__("pushfl; popl %0" :"=m"(newregs->flags));
 
-               newregs->eip = (unsigned long)current_text_addr();
+               newregs->ip = (unsigned long)current_text_addr();
        }
 }
 asmlinkage NORET_TYPE void
diff --git a/include/asm-x86/kexec_64.h b/include/asm-x86/kexec_64.h
index 738e581b67f8..b5f989b15c0b 100644
--- a/include/asm-x86/kexec_64.h
+++ b/include/asm-x86/kexec_64.h
@@ -60,14 +60,14 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 	if (oldregs)
 		memcpy(newregs, oldregs, sizeof(*newregs));
 	else {
-		__asm__ __volatile__("movq %%rbx,%0" : "=m"(newregs->rbx));
-		__asm__ __volatile__("movq %%rcx,%0" : "=m"(newregs->rcx));
-		__asm__ __volatile__("movq %%rdx,%0" : "=m"(newregs->rdx));
-		__asm__ __volatile__("movq %%rsi,%0" : "=m"(newregs->rsi));
-		__asm__ __volatile__("movq %%rdi,%0" : "=m"(newregs->rdi));
-		__asm__ __volatile__("movq %%rbp,%0" : "=m"(newregs->rbp));
-		__asm__ __volatile__("movq %%rax,%0" : "=m"(newregs->rax));
-		__asm__ __volatile__("movq %%rsp,%0" : "=m"(newregs->rsp));
+		__asm__ __volatile__("movq %%rbx,%0" : "=m"(newregs->bx));
+		__asm__ __volatile__("movq %%rcx,%0" : "=m"(newregs->cx));
+		__asm__ __volatile__("movq %%rdx,%0" : "=m"(newregs->dx));
+		__asm__ __volatile__("movq %%rsi,%0" : "=m"(newregs->si));
+		__asm__ __volatile__("movq %%rdi,%0" : "=m"(newregs->di));
+		__asm__ __volatile__("movq %%rbp,%0" : "=m"(newregs->bp));
+		__asm__ __volatile__("movq %%rax,%0" : "=m"(newregs->ax));
+		__asm__ __volatile__("movq %%rsp,%0" : "=m"(newregs->sp));
 		__asm__ __volatile__("movq %%r8,%0" : "=m"(newregs->r8));
 		__asm__ __volatile__("movq %%r9,%0" : "=m"(newregs->r9));
 		__asm__ __volatile__("movq %%r10,%0" : "=m"(newregs->r10));
@@ -78,9 +78,9 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 		__asm__ __volatile__("movq %%r15,%0" : "=m"(newregs->r15));
 		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
 		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
-		__asm__ __volatile__("pushfq; popq %0" :"=m"(newregs->eflags));
+		__asm__ __volatile__("pushfq; popq %0" :"=m"(newregs->flags));
 
-		newregs->rip = (unsigned long)current_text_addr();
+		newregs->ip = (unsigned long)current_text_addr();
 	}
 }
 
diff --git a/include/asm-x86/kprobes_32.h b/include/asm-x86/kprobes_32.h
index 9fe8f3bddfd5..2f38315bc39f 100644
--- a/include/asm-x86/kprobes_32.h
+++ b/include/asm-x86/kprobes_32.h
@@ -84,7 +84,7 @@ struct kprobe_ctlblk {
  */
 static inline void restore_interrupts(struct pt_regs *regs)
 {
-	if (regs->eflags & IF_MASK)
+	if (regs->flags & IF_MASK)
 		local_irq_enable();
 }
 
diff --git a/include/asm-x86/kprobes_64.h b/include/asm-x86/kprobes_64.h
index 743d76218fc9..8c919d35cdd3 100644
--- a/include/asm-x86/kprobes_64.h
+++ b/include/asm-x86/kprobes_64.h
@@ -77,7 +77,7 @@ struct kprobe_ctlblk {
  */
 static inline void restore_interrupts(struct pt_regs *regs)
 {
-	if (regs->eflags & IF_MASK)
+	if (regs->flags & IF_MASK)
 		local_irq_enable();
 }
 
diff --git a/include/asm-x86/mce.h b/include/asm-x86/mce.h
index e6ff507a73b0..94f1fd79e22a 100644
--- a/include/asm-x86/mce.h
+++ b/include/asm-x86/mce.h
@@ -13,7 +13,7 @@
 #define MCG_CTL_P	 (1UL<<8)   /* MCG_CAP register available */
 
 #define MCG_STATUS_RIPV  (1UL<<0)   /* restart ip valid */
-#define MCG_STATUS_EIPV  (1UL<<1)   /* eip points to correct instruction */
+#define MCG_STATUS_EIPV  (1UL<<1)   /* ip points to correct instruction */
 #define MCG_STATUS_MCIP  (1UL<<2)   /* machine check in progress */
 
 #define MCI_STATUS_VAL   (1UL<<63)  /* valid error */
@@ -30,7 +30,7 @@ struct mce {
 	__u64 misc;
 	__u64 addr;
 	__u64 mcgstatus;
-	__u64 rip;
+	__u64 ip;
 	__u64 tsc;	/* cpu time stamp counter */
 	__u64 res1;	/* for future extension */
 	__u64 res2;	/* dito. */
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 3c67eacb3168..c85400fe58c4 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -398,14 +398,14 @@ struct thread_struct {
 
 #define start_thread(regs, new_eip, new_esp) do {		\
 	__asm__("movl %0,%%gs": :"r" (0));			\
-	regs->xfs = 0;						\
+	regs->fs = 0;						\
 	set_fs(USER_DS);					\
-	regs->xds = __USER_DS;					\
-	regs->xes = __USER_DS;					\
-	regs->xss = __USER_DS;					\
-	regs->xcs = __USER_CS;					\
-	regs->eip = new_eip;					\
-	regs->esp = new_esp;					\
+	regs->ds = __USER_DS;					\
+	regs->es = __USER_DS;					\
+	regs->ss = __USER_DS;					\
+	regs->cs = __USER_CS;					\
+	regs->ip = new_eip;					\
+	regs->sp = new_esp;					\
 } while (0)
 
 /* Forward declaration, a strange C thing */
@@ -440,7 +440,7 @@ unsigned long get_wchan(struct task_struct *p);
  * is accessable even if the CPU haven't stored the SS/ESP registers
  * on the stack (interrupt gate does not save these registers
  * when switching to the same priv ring).
- * Therefore beware: accessing the xss/esp fields of the
+ * Therefore beware: accessing the ss/esp fields of the
  * "struct pt_regs" is possible, but they may contain the
  * completely wrong values.
  */
@@ -451,8 +451,8 @@ unsigned long get_wchan(struct task_struct *p);
        __regs__ - 1;                                                   \
 })
 
-#define KSTK_EIP(task) (task_pt_regs(task)->eip)
-#define KSTK_ESP(task) (task_pt_regs(task)->esp)
+#define KSTK_EIP(task) (task_pt_regs(task)->ip)
+#define KSTK_ESP(task) (task_pt_regs(task)->sp)
 
 
 struct microcode_header {
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index e7bea4fed642..797770113e6d 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -258,12 +258,12 @@ struct thread_struct {
 #define start_thread(regs,new_rip,new_rsp) do { \
 	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));	 \
 	load_gs_index(0);							\
-	(regs)->rip = (new_rip);						 \
-	(regs)->rsp = (new_rsp);						 \
+	(regs)->ip = (new_rip);						 \
+	(regs)->sp = (new_rsp);						 \
 	write_pda(oldrsp, (new_rsp));						 \
 	(regs)->cs = __USER_CS;							 \
 	(regs)->ss = __USER_DS;							 \
-	(regs)->eflags = 0x200;							 \
+	(regs)->flags = 0x200;							 \
 	set_fs(USER_DS);							 \
 } while(0) 
 
@@ -297,7 +297,7 @@ extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
 
 extern unsigned long get_wchan(struct task_struct *p);
 #define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.rsp0 - 1)
-#define KSTK_EIP(tsk) (task_pt_regs(tsk)->rip)
+#define KSTK_EIP(tsk) (task_pt_regs(tsk)->ip)
 #define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
 
 
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 04204f359298..9187b2fab754 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -10,6 +10,8 @@
 /* this struct defines the way the registers are stored on the
    stack during a system call. */
 
+#ifndef __KERNEL__
+
 struct pt_regs {
 	long ebx;
 	long ecx;
@@ -21,7 +23,7 @@ struct pt_regs {
 	int  xds;
 	int  xes;
 	int  xfs;
-	/* int  xgs; */
+	/* int  gs; */
 	long orig_eax;
 	long eip;
 	int  xcs;
@@ -30,7 +32,27 @@ struct pt_regs {
 	int  xss;
 };
 
-#ifdef __KERNEL__
+#else /* __KERNEL__ */
+
+struct pt_regs {
+	long bx;
+	long cx;
+	long dx;
+	long si;
+	long di;
+	long bp;
+	long ax;
+	int  ds;
+	int  es;
+	int  fs;
+	/* int  gs; */
+	long orig_ax;
+	long ip;
+	int  cs;
+	long flags;
+	long sp;
+	int  ss;
+};
 
 #include <asm/vm86.h>
 #include <asm/segment.h>
@@ -47,27 +69,30 @@ extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int erro
  */
 static inline int user_mode(struct pt_regs *regs)
 {
-	return (regs->xcs & SEGMENT_RPL_MASK) == USER_RPL;
+	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
 }
 static inline int user_mode_vm(struct pt_regs *regs)
 {
-	return ((regs->xcs & SEGMENT_RPL_MASK) | (regs->eflags & VM_MASK)) >= USER_RPL;
+	return ((regs->cs & SEGMENT_RPL_MASK) |
+		(regs->flags & VM_MASK)) >= USER_RPL;
 }
 static inline int v8086_mode(struct pt_regs *regs)
 {
-	return (regs->eflags & VM_MASK);
+	return (regs->flags & VM_MASK);
 }
 
-#define instruction_pointer(regs) ((regs)->eip)
-#define frame_pointer(regs) ((regs)->ebp)
+#define instruction_pointer(regs) ((regs)->ip)
+#define frame_pointer(regs) ((regs)->bp)
 #define stack_pointer(regs) ((unsigned long)(regs))
-#define regs_return_value(regs) ((regs)->eax)
+#define regs_return_value(regs) ((regs)->ax)
 
 extern unsigned long profile_pc(struct pt_regs *regs);
 #endif /* __KERNEL__ */
 
 #else /* __i386__ */
 
+#ifndef __KERNEL__
+
 struct pt_regs {
 	unsigned long r15;
 	unsigned long r14;
@@ -96,14 +121,43 @@ struct pt_regs {
 /* top of stack page */
 };
 
-#ifdef __KERNEL__
+#else /* __KERNEL__ */
+
+struct pt_regs {
+	unsigned long r15;
+	unsigned long r14;
+	unsigned long r13;
+	unsigned long r12;
+	unsigned long bp;
+	unsigned long bx;
+/* arguments: non interrupts/non tracing syscalls only save upto here*/
+	unsigned long r11;
+	unsigned long r10;
+	unsigned long r9;
+	unsigned long r8;
+	unsigned long ax;
+	unsigned long cx;
+	unsigned long dx;
+	unsigned long si;
+	unsigned long di;
+	unsigned long orig_ax;
+/* end of arguments */
+/* cpu exception frame or undefined */
+	unsigned long ip;
+	unsigned long cs;
+	unsigned long flags;
+	unsigned long sp;
+	unsigned long ss;
+/* top of stack page */
+};
 
 #define user_mode(regs) (!!((regs)->cs & 3))
 #define user_mode_vm(regs) user_mode(regs)
-#define instruction_pointer(regs) ((regs)->rip)
-#define frame_pointer(regs) ((regs)->rbp)
-#define stack_pointer(regs) ((regs)->rsp)
-#define regs_return_value(regs) ((regs)->rax)
+#define v8086_mode(regs) 0	/* No V86 mode support in long mode */
+#define instruction_pointer(regs) ((regs)->ip)
+#define frame_pointer(regs) ((regs)->bp)
+#define stack_pointer(regs) ((regs)->sp)
+#define regs_return_value(regs) ((regs)->ax)
 
 extern unsigned long profile_pc(struct pt_regs *regs);
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
diff --git a/kernel/signal.c b/kernel/signal.c
index afa4f781f924..bf49ce6f016b 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -733,13 +733,13 @@ static void print_fatal_signal(struct pt_regs *regs, int signr)
 		current->comm, task_pid_nr(current), signr);
 
 #if defined(__i386__) && !defined(__arch_um__)
-	printk("code at %08lx: ", regs->eip);
+	printk("code at %08lx: ", regs->ip);
 	{
 		int i;
 		for (i = 0; i < 16; i++) {
 			unsigned char insn;
 
-			__get_user(insn, (unsigned char *)(regs->eip + i));
+			__get_user(insn, (unsigned char *)(regs->ip + i));
 			printk("%02x ", insn);
 		}
 	}
-- 
cgit v1.2.3


From 153d5f2e5787c74e9cbb6b6687c9b04be1b59893 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:56 +0100
Subject: x86: use generic register names in struct user_regs_struct

Switch struct user_regs_struct (defined in <asm/user.h>, which is no
longer exported to userspace) to using register names without e- or
r-prefixes for both 32 and 64 bit x86.  This is intended as a
preliminary step in unifying this code between architectures.

Also, be a bit more strict in truncating 32-bit "extended" segment
register values to 16 bits.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c | 35 ++++++++++++++++++-----------------
 arch/x86/kernel/ptrace_64.c  |  4 ++--
 include/asm-x86/user_32.h    | 24 +++++++++++++++++-------
 include/asm-x86/user_64.h    | 41 +++++++++++++++++++++++++++++++----------
 4 files changed, 68 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index c9f28e02e86d..53406461074f 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -523,6 +523,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 void dump_thread(struct pt_regs * regs, struct user * dump)
 {
 	int i;
+	u16 gs;
 
 /* changed the size calculations - should hopefully work better. lbt */
 	dump->magic = CMAGIC;
@@ -538,23 +539,23 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 	if (dump->start_stack < TASK_SIZE)
 		dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
 
-	dump->regs.ebx = regs->bx;
-	dump->regs.ecx = regs->cx;
-	dump->regs.edx = regs->dx;
-	dump->regs.esi = regs->si;
-	dump->regs.edi = regs->di;
-	dump->regs.ebp = regs->bp;
-	dump->regs.eax = regs->ax;
-	dump->regs.ds = regs->ds;
-	dump->regs.es = regs->es;
-	dump->regs.fs = regs->fs;
-	savesegment(gs,dump->regs.gs);
-	dump->regs.orig_eax = regs->orig_ax;
-	dump->regs.eip = regs->ip;
-	dump->regs.cs = regs->cs;
-	dump->regs.eflags = regs->flags;
-	dump->regs.esp = regs->sp;
-	dump->regs.ss = regs->ss;
+	dump->regs.bx = regs->bx;
+	dump->regs.cx = regs->cx;
+	dump->regs.dx = regs->dx;
+	dump->regs.si = regs->si;
+	dump->regs.di = regs->di;
+	dump->regs.bp = regs->bp;
+	dump->regs.ax = regs->ax;
+	dump->regs.ds = (u16)regs->ds;
+	dump->regs.es = (u16)regs->es;
+	dump->regs.fs = (u16)regs->fs;
+	savesegment(gs,gs);
+	dump->regs.orig_ax = regs->orig_ax;
+	dump->regs.ip = regs->ip;
+	dump->regs.cs = (u16)regs->cs;
+	dump->regs.flags = regs->flags;
+	dump->regs.sp = regs->sp;
+	dump->regs.ss = (u16)regs->ss;
 
 	dump->u_fpvalid = dump_fpu (regs, &dump->i387);
 }
diff --git a/arch/x86/kernel/ptrace_64.c b/arch/x86/kernel/ptrace_64.c
index bee20bb1a6c0..56b31cd3b865 100644
--- a/arch/x86/kernel/ptrace_64.c
+++ b/arch/x86/kernel/ptrace_64.c
@@ -108,7 +108,7 @@ static int putreg(struct task_struct *child,
 			if (child->thread.gs != value)
 				return do_arch_prctl(child, ARCH_SET_GS, value);
 			return 0;
-		case offsetof(struct user_regs_struct, eflags):
+		case offsetof(struct user_regs_struct,flags):
 			value &= FLAG_MASK;
 			/*
 			 * If the user value contains TF, mark that
@@ -164,7 +164,7 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
 			if (child->thread.gsindex != GS_TLS_SEL)
 				return 0;
 			return get_desc_base(&child->thread.tls_array[GS_TLS]);
-		case offsetof(struct user_regs_struct, eflags):
+		case offsetof(struct user_regs_struct, flags):
 			/*
 			 * If the debugger set TF, hide it from the readout.
 			 */
diff --git a/include/asm-x86/user_32.h b/include/asm-x86/user_32.h
index 0e85d2a5e33a..ed8b8fc6906c 100644
--- a/include/asm-x86/user_32.h
+++ b/include/asm-x86/user_32.h
@@ -75,13 +75,23 @@ struct user_fxsr_struct {
  * doesn't use the extra segment registers)
  */
 struct user_regs_struct {
-	long ebx, ecx, edx, esi, edi, ebp, eax;
-	unsigned short ds, __ds, es, __es;
-	unsigned short fs, __fs, gs, __gs;
-	long orig_eax, eip;
-	unsigned short cs, __cs;
-	long eflags, esp;
-	unsigned short ss, __ss;
+	unsigned long	bx;
+	unsigned long	cx;
+	unsigned long	dx;
+	unsigned long	si;
+	unsigned long	di;
+	unsigned long	bp;
+	unsigned long	ax;
+	unsigned long	ds;
+	unsigned long	es;
+	unsigned long	fs;
+	unsigned long	gs;
+	unsigned long	orig_ax;
+	unsigned long	ip;
+	unsigned long	cs;
+	unsigned long	flags;
+	unsigned long	sp;
+	unsigned long	ss;
 };
 
 /* When the kernel dumps core, it starts by dumping the user struct -
diff --git a/include/asm-x86/user_64.h b/include/asm-x86/user_64.h
index 12785c649ac5..a5449d456cc0 100644
--- a/include/asm-x86/user_64.h
+++ b/include/asm-x86/user_64.h
@@ -40,13 +40,13 @@
  * and both the standard and SIMD floating point data can be accessed via
  * the new ptrace requests.  In either case, changes to the FPU environment
  * will be reflected in the task's state as expected.
- * 
+ *
  * x86-64 support by Andi Kleen.
  */
 
 /* This matches the 64bit FXSAVE format as defined by AMD. It is the same
    as the 32bit format defined by Intel, except that the selector:offset pairs for
-   data and eip are replaced with flat 64bit pointers. */ 
+   data and eip are replaced with flat 64bit pointers. */
 struct user_i387_struct {
 	unsigned short	cwd;
 	unsigned short	swd;
@@ -65,13 +65,34 @@ struct user_i387_struct {
  * Segment register layout in coredumps.
  */
 struct user_regs_struct {
-	unsigned long r15,r14,r13,r12,rbp,rbx,r11,r10;
-	unsigned long r9,r8,rax,rcx,rdx,rsi,rdi,orig_rax;
-	unsigned long rip,cs,eflags;
-	unsigned long rsp,ss;
-  	unsigned long fs_base, gs_base;
-	unsigned long ds,es,fs,gs; 
-}; 
+	unsigned long	r15;
+	unsigned long	r14;
+	unsigned long	r13;
+	unsigned long	r12;
+	unsigned long	bp;
+	unsigned long	bx;
+	unsigned long	r11;
+	unsigned long	r10;
+	unsigned long	r9;
+	unsigned long	r8;
+	unsigned long	ax;
+	unsigned long	cx;
+	unsigned long	dx;
+	unsigned long	si;
+	unsigned long	di;
+	unsigned long	orig_ax;
+	unsigned long	ip;
+	unsigned long	cs;
+	unsigned long	flags;
+	unsigned long	sp;
+	unsigned long	ss;
+	unsigned long	fs_base;
+	unsigned long	gs_base;
+	unsigned long	ds;
+	unsigned long	es;
+	unsigned long	fs;
+	unsigned long	gs;
+};
 
 /* When the kernel dumps core, it starts by dumping the user struct -
    this will be used by gdb to figure out where the data and stack segments
@@ -94,7 +115,7 @@ struct user{
 				   This is actually the bottom of the stack,
 				   the top of the stack is always found in the
 				   esp register.  */
-  long int signal;     		/* Signal that caused the core dump. */
+  long int signal;		/* Signal that caused the core dump. */
   int reserved;			/* No longer used */
   int pad1;
   struct user_pt_regs * u_ar0;	/* Used by gdb to help find the values for */
-- 
cgit v1.2.3


From 742fa54a62be6a263df14a553bf832724471dfbe Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:30:56 +0100
Subject: x86: use generic register names in struct sigcontext

Switch struct sigcontext (defined in <asm/sigcontext*.h>) to using
register names withut e- or r-prefixes for both 32- and 64-bit x86.
This is intended as a preliminary step in unifying this code between
architectures.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ia32_signal.c      | 32 ++++++++++++------------
 arch/x86/kernel/asm-offsets_32.c | 18 +++++++-------
 arch/x86/kernel/asm-offsets_64.c | 18 +++++++-------
 arch/x86/kernel/signal_32.c      | 28 ++++++++++-----------
 arch/x86/kernel/signal_64.c      | 29 +++++++++++----------
 arch/x86/vdso/vdso32/sigreturn.S | 54 ++++++++++++++++++++--------------------
 include/asm-x86/sigcontext.h     | 42 +++++++++++++++----------------
 include/asm-x86/sigcontext32.h   | 22 ++++++++--------
 8 files changed, 121 insertions(+), 122 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index f2da443f8c7b..d03d43f32f4c 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -192,9 +192,9 @@ struct rt_sigframe
 	char retcode[8];
 };
 
-#define COPY(x)		{ \
-	unsigned int reg;			\
-	err |= __get_user(reg, &sc->e ##x);	\
+#define COPY(x)		{ 		\
+	unsigned int reg;		\
+	err |= __get_user(reg, &sc->x);	\
 	regs->x = reg;			\
 }
 
@@ -248,7 +248,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 	err |= __get_user(regs->ss, &sc->ss);
 	regs->ss |= 3;
 
-	err |= __get_user(tmpflags, &sc->eflags);
+	err |= __get_user(tmpflags, &sc->flags);
 	regs->flags = (regs->flags & ~0x40DD5) | (tmpflags & 0x40DD5);
 	/* disable syscall checks */
 	regs->orig_ax = -1;
@@ -268,7 +268,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 		}
 	}
 
-	err |= __get_user(tmp, &sc->eax);
+	err |= __get_user(tmp, &sc->ax);
 	*peax = tmp;
 
 	return err;
@@ -361,21 +361,21 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
 	__asm__("movl %%es,%0" : "=r"(tmp): "0"(tmp));
 	err |= __put_user(tmp, (unsigned int __user *)&sc->es);
 
-	err |= __put_user((u32)regs->di, &sc->edi);
-	err |= __put_user((u32)regs->si, &sc->esi);
-	err |= __put_user((u32)regs->bp, &sc->ebp);
-	err |= __put_user((u32)regs->sp, &sc->esp);
-	err |= __put_user((u32)regs->bx, &sc->ebx);
-	err |= __put_user((u32)regs->dx, &sc->edx);
-	err |= __put_user((u32)regs->cx, &sc->ecx);
-	err |= __put_user((u32)regs->ax, &sc->eax);
+	err |= __put_user((u32)regs->di, &sc->di);
+	err |= __put_user((u32)regs->si, &sc->si);
+	err |= __put_user((u32)regs->bp, &sc->bp);
+	err |= __put_user((u32)regs->sp, &sc->sp);
+	err |= __put_user((u32)regs->bx, &sc->bx);
+	err |= __put_user((u32)regs->dx, &sc->dx);
+	err |= __put_user((u32)regs->cx, &sc->cx);
+	err |= __put_user((u32)regs->ax, &sc->ax);
 	err |= __put_user((u32)regs->cs, &sc->cs);
 	err |= __put_user((u32)regs->ss, &sc->ss);
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
-	err |= __put_user((u32)regs->ip, &sc->eip);
-	err |= __put_user((u32)regs->flags, &sc->eflags);
-	err |= __put_user((u32)regs->sp, &sc->esp_at_signal);
+	err |= __put_user((u32)regs->ip, &sc->ip);
+	err |= __put_user((u32)regs->flags, &sc->flags);
+	err |= __put_user((u32)regs->sp, &sc->sp_at_signal);
 
 	tmp = save_i387_ia32(current, fpstate, regs, 0);
 	if (tmp < 0)
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index a3a8be7618d1..4fc24a61f431 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -38,15 +38,15 @@ void foo(void);
 
 void foo(void)
 {
-	OFFSET(IA32_SIGCONTEXT_eax, sigcontext, eax);
-	OFFSET(IA32_SIGCONTEXT_ebx, sigcontext, ebx);
-	OFFSET(IA32_SIGCONTEXT_ecx, sigcontext, ecx);
-	OFFSET(IA32_SIGCONTEXT_edx, sigcontext, edx);
-	OFFSET(IA32_SIGCONTEXT_esi, sigcontext, esi);
-	OFFSET(IA32_SIGCONTEXT_edi, sigcontext, edi);
-	OFFSET(IA32_SIGCONTEXT_ebp, sigcontext, ebp);
-	OFFSET(IA32_SIGCONTEXT_esp, sigcontext, esp);
-	OFFSET(IA32_SIGCONTEXT_eip, sigcontext, eip);
+	OFFSET(IA32_SIGCONTEXT_ax, sigcontext, ax);
+	OFFSET(IA32_SIGCONTEXT_bx, sigcontext, bx);
+	OFFSET(IA32_SIGCONTEXT_cx, sigcontext, cx);
+	OFFSET(IA32_SIGCONTEXT_dx, sigcontext, dx);
+	OFFSET(IA32_SIGCONTEXT_si, sigcontext, si);
+	OFFSET(IA32_SIGCONTEXT_di, sigcontext, di);
+	OFFSET(IA32_SIGCONTEXT_bp, sigcontext, bp);
+	OFFSET(IA32_SIGCONTEXT_sp, sigcontext, sp);
+	OFFSET(IA32_SIGCONTEXT_ip, sigcontext, ip);
 	BLANK();
 
 	OFFSET(CPUINFO_x86, cpuinfo_x86, x86);
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index 2e918ebf21d3..a05428764314 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -63,15 +63,15 @@ int main(void)
 #undef ENTRY
 #ifdef CONFIG_IA32_EMULATION
 #define ENTRY(entry) DEFINE(IA32_SIGCONTEXT_ ## entry, offsetof(struct sigcontext_ia32, entry))
-	ENTRY(eax);
-	ENTRY(ebx);
-	ENTRY(ecx);
-	ENTRY(edx);
-	ENTRY(esi);
-	ENTRY(edi);
-	ENTRY(ebp);
-	ENTRY(esp);
-	ENTRY(eip);
+	ENTRY(ax);
+	ENTRY(bx);
+	ENTRY(cx);
+	ENTRY(dx);
+	ENTRY(si);
+	ENTRY(di);
+	ENTRY(bp);
+	ENTRY(sp);
+	ENTRY(ip);
 	BLANK();
 #undef ENTRY
 	DEFINE(IA32_RT_SIGFRAME_sigcontext,
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 40fd3515ccf1..64cb3c05de69 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -105,7 +105,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPY(x)		err |= __get_user(regs->x, &sc->e ## x)
+#define COPY(x)		err |= __get_user(regs->x, &sc->x)
 
 #define COPY_SEG(seg)							\
 	{ unsigned short tmp;						\
@@ -144,7 +144,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 	
 	{
 		unsigned int tmpflags;
-		err |= __get_user(tmpflags, &sc->eflags);
+		err |= __get_user(tmpflags, &sc->flags);
 		regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
 		regs->orig_ax = -1;		/* disable syscall checks */
 	}
@@ -165,7 +165,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
 		}
 	}
 
-	err |= __get_user(*peax, &sc->eax);
+	err |= __get_user(*peax, &sc->ax);
 	return err;
 
 badframe:
@@ -256,20 +256,20 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
 
 	err |= __put_user(regs->es, (unsigned int __user *)&sc->es);
 	err |= __put_user(regs->ds, (unsigned int __user *)&sc->ds);
-	err |= __put_user(regs->di, &sc->edi);
-	err |= __put_user(regs->si, &sc->esi);
-	err |= __put_user(regs->bp, &sc->ebp);
-	err |= __put_user(regs->sp, &sc->esp);
-	err |= __put_user(regs->bx, &sc->ebx);
-	err |= __put_user(regs->dx, &sc->edx);
-	err |= __put_user(regs->cx, &sc->ecx);
-	err |= __put_user(regs->ax, &sc->eax);
+	err |= __put_user(regs->di, &sc->di);
+	err |= __put_user(regs->si, &sc->si);
+	err |= __put_user(regs->bp, &sc->bp);
+	err |= __put_user(regs->sp, &sc->sp);
+	err |= __put_user(regs->bx, &sc->bx);
+	err |= __put_user(regs->dx, &sc->dx);
+	err |= __put_user(regs->cx, &sc->cx);
+	err |= __put_user(regs->ax, &sc->ax);
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
-	err |= __put_user(regs->ip, &sc->eip);
+	err |= __put_user(regs->ip, &sc->ip);
 	err |= __put_user(regs->cs, (unsigned int __user *)&sc->cs);
-	err |= __put_user(regs->flags, &sc->eflags);
-	err |= __put_user(regs->sp, &sc->esp_at_signal);
+	err |= __put_user(regs->flags, &sc->flags);
+	err |= __put_user(regs->sp, &sc->sp_at_signal);
 	err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss);
 
 	tmp = save_i387(fpstate);
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 4b228fd83b31..1c9bca56fb55 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -62,11 +62,10 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPYR(x)	err |= __get_user(regs->x, &sc->r ## x)
 #define COPY(x)		err |= __get_user(regs->x, &sc->x)
 
-	COPYR(di); COPYR(si); COPYR(bp); COPYR(sp); COPYR(bx);
-	COPYR(dx); COPYR(cx); COPYR(ip);
+	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
+	COPY(dx); COPY(cx); COPY(ip);
 	COPY(r8);
 	COPY(r9);
 	COPY(r10);
@@ -87,7 +86,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned
 
 	{
 		unsigned int tmpflags;
-		err |= __get_user(tmpflags, &sc->eflags);
+		err |= __get_user(tmpflags, &sc->flags);
 		regs->flags = (regs->flags & ~0x40DD5) | (tmpflags & 0x40DD5);
 		regs->orig_ax = -1;		/* disable syscall checks */
 	}
@@ -109,7 +108,7 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, unsigned
 		}
 	}
 
-	err |= __get_user(*prax, &sc->rax);
+	err |= __get_user(*prax, &sc->ax);
 	return err;
 
 badframe:
@@ -166,14 +165,14 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned lo
 	err |= __put_user(0, &sc->gs);
 	err |= __put_user(0, &sc->fs);
 
-	err |= __put_user(regs->di, &sc->rdi);
-	err |= __put_user(regs->si, &sc->rsi);
-	err |= __put_user(regs->bp, &sc->rbp);
-	err |= __put_user(regs->sp, &sc->rsp);
-	err |= __put_user(regs->bx, &sc->rbx);
-	err |= __put_user(regs->dx, &sc->rdx);
-	err |= __put_user(regs->cx, &sc->rcx);
-	err |= __put_user(regs->ax, &sc->rax);
+	err |= __put_user(regs->di, &sc->di);
+	err |= __put_user(regs->si, &sc->si);
+	err |= __put_user(regs->bp, &sc->bp);
+	err |= __put_user(regs->sp, &sc->sp);
+	err |= __put_user(regs->bx, &sc->bx);
+	err |= __put_user(regs->dx, &sc->dx);
+	err |= __put_user(regs->cx, &sc->cx);
+	err |= __put_user(regs->ax, &sc->ax);
 	err |= __put_user(regs->r8, &sc->r8);
 	err |= __put_user(regs->r9, &sc->r9);
 	err |= __put_user(regs->r10, &sc->r10);
@@ -184,8 +183,8 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, unsigned lo
 	err |= __put_user(regs->r15, &sc->r15);
 	err |= __put_user(me->thread.trap_no, &sc->trapno);
 	err |= __put_user(me->thread.error_code, &sc->err);
-	err |= __put_user(regs->ip, &sc->rip);
-	err |= __put_user(regs->flags, &sc->eflags);
+	err |= __put_user(regs->ip, &sc->ip);
+	err |= __put_user(regs->flags, &sc->flags);
 	err |= __put_user(mask, &sc->oldmask);
 	err |= __put_user(me->thread.cr2, &sc->cr2);
 
diff --git a/arch/x86/vdso/vdso32/sigreturn.S b/arch/x86/vdso/vdso32/sigreturn.S
index cade2752928b..31776d0efc8c 100644
--- a/arch/x86/vdso/vdso32/sigreturn.S
+++ b/arch/x86/vdso/vdso32/sigreturn.S
@@ -92,27 +92,27 @@ __kernel_rt_sigreturn:
 	.sleb128 offset;		/*       offset */		\
 1:
 
-	do_cfa_expr(IA32_SIGCONTEXT_esp+4)
-	do_expr(0, IA32_SIGCONTEXT_eax+4)
-	do_expr(1, IA32_SIGCONTEXT_ecx+4)
-	do_expr(2, IA32_SIGCONTEXT_edx+4)
-	do_expr(3, IA32_SIGCONTEXT_ebx+4)
-	do_expr(5, IA32_SIGCONTEXT_ebp+4)
-	do_expr(6, IA32_SIGCONTEXT_esi+4)
-	do_expr(7, IA32_SIGCONTEXT_edi+4)
-	do_expr(8, IA32_SIGCONTEXT_eip+4)
+	do_cfa_expr(IA32_SIGCONTEXT_sp+4)
+	do_expr(0, IA32_SIGCONTEXT_ax+4)
+	do_expr(1, IA32_SIGCONTEXT_cx+4)
+	do_expr(2, IA32_SIGCONTEXT_dx+4)
+	do_expr(3, IA32_SIGCONTEXT_bx+4)
+	do_expr(5, IA32_SIGCONTEXT_bp+4)
+	do_expr(6, IA32_SIGCONTEXT_si+4)
+	do_expr(7, IA32_SIGCONTEXT_di+4)
+	do_expr(8, IA32_SIGCONTEXT_ip+4)
 
 	.byte 0x42	/* DW_CFA_advance_loc 2 -- nop; popl eax. */
 
-	do_cfa_expr(IA32_SIGCONTEXT_esp)
-	do_expr(0, IA32_SIGCONTEXT_eax)
-	do_expr(1, IA32_SIGCONTEXT_ecx)
-	do_expr(2, IA32_SIGCONTEXT_edx)
-	do_expr(3, IA32_SIGCONTEXT_ebx)
-	do_expr(5, IA32_SIGCONTEXT_ebp)
-	do_expr(6, IA32_SIGCONTEXT_esi)
-	do_expr(7, IA32_SIGCONTEXT_edi)
-	do_expr(8, IA32_SIGCONTEXT_eip)
+	do_cfa_expr(IA32_SIGCONTEXT_sp)
+	do_expr(0, IA32_SIGCONTEXT_ax)
+	do_expr(1, IA32_SIGCONTEXT_cx)
+	do_expr(2, IA32_SIGCONTEXT_dx)
+	do_expr(3, IA32_SIGCONTEXT_bx)
+	do_expr(5, IA32_SIGCONTEXT_bp)
+	do_expr(6, IA32_SIGCONTEXT_si)
+	do_expr(7, IA32_SIGCONTEXT_di)
+	do_expr(8, IA32_SIGCONTEXT_ip)
 
 	.align 4
 .LENDFDEDLSI1:
@@ -129,15 +129,15 @@ __kernel_rt_sigreturn:
 	   slightly less complicated than the above, since we don't
 	   modify the stack pointer in the process.  */
 
-	do_cfa_expr(IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_esp)
-	do_expr(0, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_eax)
-	do_expr(1, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ecx)
-	do_expr(2, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_edx)
-	do_expr(3, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ebx)
-	do_expr(5, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ebp)
-	do_expr(6, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_esi)
-	do_expr(7, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_edi)
-	do_expr(8, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_eip)
+	do_cfa_expr(IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_sp)
+	do_expr(0, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ax)
+	do_expr(1, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_cx)
+	do_expr(2, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_dx)
+	do_expr(3, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_bx)
+	do_expr(5, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_bp)
+	do_expr(6, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_si)
+	do_expr(7, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_di)
+	do_expr(8, IA32_RT_SIGFRAME_sigcontext-4 + IA32_SIGCONTEXT_ip)
 
 	.align 4
 .LENDFDEDLSI2:
diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h
index c047f9dc3423..681deade5f00 100644
--- a/include/asm-x86/sigcontext.h
+++ b/include/asm-x86/sigcontext.h
@@ -63,20 +63,20 @@ struct sigcontext {
 	unsigned short fs, __fsh;
 	unsigned short es, __esh;
 	unsigned short ds, __dsh;
-	unsigned long edi;
-	unsigned long esi;
-	unsigned long ebp;
-	unsigned long esp;
-	unsigned long ebx;
-	unsigned long edx;
-	unsigned long ecx;
-	unsigned long eax;
+	unsigned long di;
+	unsigned long si;
+	unsigned long bp;
+	unsigned long sp;
+	unsigned long bx;
+	unsigned long dx;
+	unsigned long cx;
+	unsigned long ax;
 	unsigned long trapno;
 	unsigned long err;
-	unsigned long eip;
+	unsigned long ip;
 	unsigned short cs, __csh;
-	unsigned long eflags;
-	unsigned long esp_at_signal;
+	unsigned long flags;
+	unsigned long sp_at_signal;
 	unsigned short ss, __ssh;
 	struct _fpstate __user * fpstate;
 	unsigned long oldmask;
@@ -111,16 +111,16 @@ struct sigcontext {
 	unsigned long r13;
 	unsigned long r14;
 	unsigned long r15;
-	unsigned long rdi;
-	unsigned long rsi;
-	unsigned long rbp;
-	unsigned long rbx;
-	unsigned long rdx;
-	unsigned long rax;
-	unsigned long rcx;
-	unsigned long rsp;
-	unsigned long rip;
-	unsigned long eflags;		/* RFLAGS */
+	unsigned long di;
+	unsigned long si;
+	unsigned long bp;
+	unsigned long bx;
+	unsigned long dx;
+	unsigned long ax;
+	unsigned long cx;
+	unsigned long sp;
+	unsigned long ip;
+	unsigned long flags;
 	unsigned short cs;
 	unsigned short gs;
 	unsigned short fs;
diff --git a/include/asm-x86/sigcontext32.h b/include/asm-x86/sigcontext32.h
index 3d657038ab7c..6ffab4fd593a 100644
--- a/include/asm-x86/sigcontext32.h
+++ b/include/asm-x86/sigcontext32.h
@@ -48,20 +48,20 @@ struct sigcontext_ia32 {
        unsigned short fs, __fsh;
        unsigned short es, __esh;
        unsigned short ds, __dsh;
-       unsigned int edi;
-       unsigned int esi;
-       unsigned int ebp;
-       unsigned int esp;
-       unsigned int ebx;
-       unsigned int edx;
-       unsigned int ecx;
-       unsigned int eax;
+       unsigned int di;
+       unsigned int si;
+       unsigned int bp;
+       unsigned int sp;
+       unsigned int bx;
+       unsigned int dx;
+       unsigned int cx;
+       unsigned int ax;
        unsigned int trapno;
        unsigned int err;
-       unsigned int eip;
+       unsigned int ip;
        unsigned short cs, __csh;
-       unsigned int eflags;
-       unsigned int esp_at_signal;
+       unsigned int flags;
+       unsigned int sp_at_signal;
        unsigned short ss, __ssh;
        unsigned int fpstate;		/* really (struct _fpstate_ia32 *) */
        unsigned int oldmask;
-- 
cgit v1.2.3


From 35d0991ffa544945d2e1992169c097286b7c0bfb Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:57 +0100
Subject: x86: eflags enum

This removes the EF_* enum from <asm/ptrace.h>.  It is no longer used,
and duplicates the X86_EFLAGS_* constants from <asm/processor-flags.h>.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/ptrace.h | 22 ----------------------
 1 file changed, 22 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 9187b2fab754..1b7a8b8d6f60 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -170,28 +170,6 @@ extern int ptrace_set_debugreg(struct task_struct *child, int n, unsigned long);
 extern unsigned long
 convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
-enum {
-	EF_CF   = 0x00000001,
-	EF_PF   = 0x00000004,
-	EF_AF   = 0x00000010,
-	EF_ZF   = 0x00000040,
-	EF_SF   = 0x00000080,
-	EF_TF   = 0x00000100,
-	EF_IE   = 0x00000200,
-	EF_DF   = 0x00000400,
-	EF_OF   = 0x00000800,
-	EF_IOPL = 0x00003000,
-	EF_IOPL_RING0 = 0x00000000,
-	EF_IOPL_RING1 = 0x00001000,
-	EF_IOPL_RING2 = 0x00002000,
-	EF_NT   = 0x00004000,   /* nested task */
-	EF_RF   = 0x00010000,   /* resume */
-	EF_VM   = 0x00020000,   /* virtual mode */
-	EF_AC   = 0x00040000,   /* alignment */
-	EF_VIF  = 0x00080000,   /* virtual interrupt */
-	EF_VIP  = 0x00100000,   /* virtual interrupt pending */
-	EF_ID   = 0x00200000,   /* id */
-};
 #endif /* __KERNEL__ */
 #endif /* !__i386__ */
 
-- 
cgit v1.2.3


From 0f5340933f9bacb403f49baaf8073320e3984841 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:30:59 +0100
Subject: x86: x86-32 thread_struct.debugreg

This replaces the debugreg[7] member of thread_struct with individual
members debugreg0, etc.  This saves two words for the dummies 4 and 5,
and harmonizes the code between 32 and 64.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c   | 31 ++++++++++++++++++++-----------
 arch/x86/kernel/ptrace_32.c    | 29 ++++++++++++++++++++++++-----
 arch/x86/kernel/signal_32.c    |  4 ++--
 arch/x86/kernel/traps_32.c     |  4 ++--
 arch/x86/power/cpu.c           | 14 +++++++-------
 include/asm-x86/processor_32.h |  7 ++++++-
 6 files changed, 61 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 53406461074f..3744cf63682c 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -445,7 +445,12 @@ void flush_thread(void)
 {
 	struct task_struct *tsk = current;
 
-	memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8);
+	tsk->thread.debugreg0 = 0;
+	tsk->thread.debugreg1 = 0;
+	tsk->thread.debugreg2 = 0;
+	tsk->thread.debugreg3 = 0;
+	tsk->thread.debugreg6 = 0;
+	tsk->thread.debugreg7 = 0;
 	memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));	
 	clear_tsk_thread_flag(tsk, TIF_DEBUG);
 	/*
@@ -522,7 +527,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
  */
 void dump_thread(struct pt_regs * regs, struct user * dump)
 {
-	int i;
 	u16 gs;
 
 /* changed the size calculations - should hopefully work better. lbt */
@@ -533,8 +537,14 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 	dump->u_dsize = ((unsigned long) (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
 	dump->u_dsize -= dump->u_tsize;
 	dump->u_ssize = 0;
-	for (i = 0; i < 8; i++)
-		dump->u_debugreg[i] = current->thread.debugreg[i];  
+	dump->u_debugreg[0] = current->thread.debugreg0;
+	dump->u_debugreg[1] = current->thread.debugreg1;
+	dump->u_debugreg[2] = current->thread.debugreg2;
+	dump->u_debugreg[3] = current->thread.debugreg3;
+	dump->u_debugreg[4] = 0;
+	dump->u_debugreg[5] = 0;
+	dump->u_debugreg[6] = current->thread.debugreg6;
+	dump->u_debugreg[7] = current->thread.debugreg7;
 
 	if (dump->start_stack < TASK_SIZE)
 		dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
@@ -612,13 +622,13 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		wrmsr(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr, 0);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
-		set_debugreg(next->debugreg[0], 0);
-		set_debugreg(next->debugreg[1], 1);
-		set_debugreg(next->debugreg[2], 2);
-		set_debugreg(next->debugreg[3], 3);
+		set_debugreg(next->debugreg0, 0);
+		set_debugreg(next->debugreg1, 1);
+		set_debugreg(next->debugreg2, 2);
+		set_debugreg(next->debugreg3, 3);
 		/* no 4 and 5 */
-		set_debugreg(next->debugreg[6], 6);
-		set_debugreg(next->debugreg[7], 7);
+		set_debugreg(next->debugreg6, 6);
+		set_debugreg(next->debugreg7, 7);
 	}
 
 #ifdef CONFIG_SECCOMP
@@ -869,4 +879,3 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
 	unsigned long range_end = mm->brk + 0x02000000;
 	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
 }
-
diff --git a/arch/x86/kernel/ptrace_32.c b/arch/x86/kernel/ptrace_32.c
index 26071305de2c..fed83d066135 100644
--- a/arch/x86/kernel/ptrace_32.c
+++ b/arch/x86/kernel/ptrace_32.c
@@ -133,19 +133,39 @@ static unsigned long getreg(struct task_struct *child, unsigned long regno)
  */
 static unsigned long ptrace_get_debugreg(struct task_struct *child, int n)
 {
-	return child->thread.debugreg[n];
+	switch (n) {
+	case 0:		return child->thread.debugreg0;
+	case 1:		return child->thread.debugreg1;
+	case 2:		return child->thread.debugreg2;
+	case 3:		return child->thread.debugreg3;
+	case 6:		return child->thread.debugreg6;
+	case 7:		return child->thread.debugreg7;
+	}
+	return 0;
 }
 
 static int ptrace_set_debugreg(struct task_struct *child,
 			       int n, unsigned long data)
 {
+	int i;
+
 	if (unlikely(n == 4 || n == 5))
 		return -EIO;
 
 	if (n < 4 && unlikely(data >= TASK_SIZE - 3))
 		return -EIO;
 
-	if (n == 7) {
+	switch (n) {
+	case 0:		child->thread.debugreg0 = data; break;
+	case 1:		child->thread.debugreg1 = data; break;
+	case 2:		child->thread.debugreg2 = data; break;
+	case 3:		child->thread.debugreg3 = data; break;
+
+	case 6:
+		child->thread.debugreg6 = data;
+		break;
+
+	case 7:
 		/*
 		 * Sanity-check data. Take one half-byte at once with
 		 * check = (val >> (16 + 4*i)) & 0xf. It contains the
@@ -176,19 +196,18 @@ static int ptrace_set_debugreg(struct task_struct *child,
 		 * 64-bit kernel), so the x86_64 mask value is 0x5454.
 		 * See the AMD manual no. 24593 (AMD64 System Programming)
 		 */
-		int i;
 		data &= ~DR_CONTROL_RESERVED;
 		for (i = 0; i < 4; i++)
 			if ((0x5f54 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
 				return -EIO;
+		child->thread.debugreg7 = data;
 		if (data)
 			set_tsk_thread_flag(child, TIF_DEBUG);
 		else
 			clear_tsk_thread_flag(child, TIF_DEBUG);
+		break;
 	}
 
-	child->thread.debugreg[n] = data;
-
 	return 0;
 }
 
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 64cb3c05de69..2bf5c9aed106 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -605,8 +605,8 @@ static void fastcall do_signal(struct pt_regs *regs)
 		 * have been cleared if the watchpoint triggered
 		 * inside the kernel.
 		 */
-		if (unlikely(current->thread.debugreg[7]))
-			set_debugreg(current->thread.debugreg[7], 7);
+		if (unlikely(current->thread.debugreg7))
+			set_debugreg(current->thread.debugreg7, 7);
 
 		/* Whee!  Actually deliver the signal.  */
 		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 931ef10960ee..27713553cc59 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -852,7 +852,7 @@ fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 
 	/* Mask out spurious debug traps due to lazy DR7 setting */
 	if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
-		if (!tsk->thread.debugreg[7])
+		if (!tsk->thread.debugreg7)
 			goto clear_dr7;
 	}
 
@@ -860,7 +860,7 @@ fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 		goto debug_vm86;
 
 	/* Save debug status register where ptrace can see it */
-	tsk->thread.debugreg[6] = condition;
+	tsk->thread.debugreg6 = condition;
 
 	/*
 	 * Single-stepping through TF: make sure we ignore any events in
diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c
index 998fd3ec0d68..5a98dc35addf 100644
--- a/arch/x86/power/cpu.c
+++ b/arch/x86/power/cpu.c
@@ -74,14 +74,14 @@ static void fix_processor_context(void)
 	/*
 	 * Now maybe reload the debug registers
 	 */
-	if (current->thread.debugreg[7]){
-		set_debugreg(current->thread.debugreg[0], 0);
-		set_debugreg(current->thread.debugreg[1], 1);
-		set_debugreg(current->thread.debugreg[2], 2);
-		set_debugreg(current->thread.debugreg[3], 3);
+	if (current->thread.debugreg7) {
+		set_debugreg(current->thread.debugreg0, 0);
+		set_debugreg(current->thread.debugreg1, 1);
+		set_debugreg(current->thread.debugreg2, 2);
+		set_debugreg(current->thread.debugreg3, 3);
 		/* no 4 and 5 */
-		set_debugreg(current->thread.debugreg[6], 6);
-		set_debugreg(current->thread.debugreg[7], 7);
+		set_debugreg(current->thread.debugreg6, 6);
+		set_debugreg(current->thread.debugreg7, 7);
 	}
 
 }
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index c85400fe58c4..d50a4b48d441 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -353,7 +353,12 @@ struct thread_struct {
 	unsigned long	fs;
 	unsigned long	gs;
 /* Hardware debugging registers */
-	unsigned long	debugreg[8];  /* %%db0-7 debug registers */
+	unsigned long	debugreg0;
+	unsigned long	debugreg1;
+	unsigned long	debugreg2;
+	unsigned long	debugreg3;
+	unsigned long	debugreg6;
+	unsigned long	debugreg7;
 /* fault info */
 	unsigned long	cr2, trap_no, error_code;
 /* floating point info */
-- 
cgit v1.2.3


From cbc9d9d98215f08ed998228e7bce88502d1ce360 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:02 +0100
Subject: x86: x86 ptrace merge complete

This switches over the 64-bit build to use the shared ptrace code,
instead of the old ptrace_64.c and arch/x86/ia32/ptrace32.c code.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/Makefile      | 2 +-
 arch/x86/kernel/Makefile_64 | 3 ++-
 include/asm-x86/ptrace.h    | 3 ---
 3 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile
index ea6088640847..ec71cfeac87e 100644
--- a/arch/x86/ia32/Makefile
+++ b/arch/x86/ia32/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o \
-	ia32_binfmt.o fpu32.o ptrace32.o
+	ia32_binfmt.o fpu32.o
 
 sysv-$(CONFIG_SYSVIPC) := ipc32.o
 obj-$(CONFIG_IA32_EMULATION) += $(sysv-y)
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index d908f0175e76..b41e838dc788 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -7,12 +7,13 @@ CPPFLAGS_vmlinux.lds += -Ux86_64
 EXTRA_AFLAGS	:= -traditional
 
 obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
-		ptrace_64.o time_64.o ioport_64.o ldt.o setup_64.o i8259_64.o sys_x86_64.o \
+		time_64.o ioport_64.o ldt.o setup_64.o i8259_64.o sys_x86_64.o \
 		x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \
 		setup64.o bootflag.o e820_64.o reboot_64.o quirks.o i8237.o \
 		pci-dma_64.o pci-nommu_64.o alternative.o hpet.o tsc_64.o bugs_64.o \
 		i8253.o io_delay.o rtc.o
 
+obj-y				+= ptrace.o
 obj-y				+= step.o
 
 obj-$(CONFIG_IA32_EMULATION)	+= tls.o
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 1b7a8b8d6f60..9228870f6157 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -164,9 +164,6 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
 
 struct task_struct;
 
-extern unsigned long ptrace_get_debugreg(struct task_struct *child, int n);
-extern int ptrace_set_debugreg(struct task_struct *child, int n, unsigned long);
-
 extern unsigned long
 convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
-- 
cgit v1.2.3


From faca62273b602ab482fb7d3d940dbf41ef08b00e Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:31:02 +0100
Subject: x86: use generic register name in the thread and tss structures

This changes size-specific register names (eip/rip, esp/rsp, etc.) to
generic names in the thread and tss structures.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/asm-offsets_32.c |  4 ++--
 arch/x86/kernel/cpu/common.c     |  2 +-
 arch/x86/kernel/doublefault_32.c | 15 ++++++++-------
 arch/x86/kernel/entry_32.S       |  6 +++---
 arch/x86/kernel/paravirt_32.c    |  2 +-
 arch/x86/kernel/process_32.c     | 12 ++++++------
 arch/x86/kernel/process_64.c     | 16 ++++++++--------
 arch/x86/kernel/smpboot_32.c     |  8 ++++----
 arch/x86/kernel/smpboot_64.c     |  6 +++---
 arch/x86/kernel/traps_32.c       |  6 +++---
 arch/x86/kernel/traps_64.c       |  4 ++--
 arch/x86/kernel/vm86_32.c        | 16 ++++++++--------
 arch/x86/kernel/vmi_32.c         | 10 +++++-----
 arch/x86/lguest/boot.c           |  6 +++---
 arch/x86/vdso/vdso32-setup.c     |  4 ++--
 arch/x86/xen/enlighten.c         |  6 +++---
 arch/x86/xen/smp.c               |  4 ++--
 drivers/lguest/x86/core.c        |  4 ++--
 include/asm-x86/paravirt.h       |  6 +++---
 include/asm-x86/processor_32.h   | 37 +++++++++++++++++--------------------
 include/asm-x86/processor_64.h   | 20 ++++++++++----------
 include/asm-x86/system_32.h      |  4 ++--
 include/asm-x86/system_64.h      |  2 +-
 23 files changed, 99 insertions(+), 101 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 4fc24a61f431..415313556708 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -101,8 +101,8 @@ void foo(void)
 	OFFSET(pbe_orig_address, pbe, orig_address);
 	OFFSET(pbe_next, pbe, next);
 
-	/* Offset from the sysenter stack to tss.esp0 */
-	DEFINE(TSS_sysenter_esp0, offsetof(struct tss_struct, x86_tss.esp0) -
+	/* Offset from the sysenter stack to tss.sp0 */
+	DEFINE(TSS_sysenter_sp0, offsetof(struct tss_struct, x86_tss.sp0) -
 		 sizeof(struct tss_struct));
 
 	DEFINE(PAGE_SIZE_asm, PAGE_SIZE);
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 5db2a163bf4b..235cd615b89d 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -691,7 +691,7 @@ void __cpuinit cpu_init(void)
 		BUG();
 	enter_lazy_tlb(&init_mm, curr);
 
-	load_esp0(t, thread);
+	load_sp0(t, thread);
 	set_tss_desc(cpu,t);
 	load_TR_desc();
 	load_LDT(&init_mm.context);
diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
index 40978af630e7..cc19a3ea403a 100644
--- a/arch/x86/kernel/doublefault_32.c
+++ b/arch/x86/kernel/doublefault_32.c
@@ -35,12 +35,13 @@ static void doublefault_fn(void)
 		if (ptr_ok(tss)) {
 			struct i386_hw_tss *t = (struct i386_hw_tss *)tss;
 
-			printk(KERN_EMERG "eip = %08lx, esp = %08lx\n", t->eip, t->esp);
+			printk(KERN_EMERG "eip = %08lx, esp = %08lx\n",
+			       t->ip, t->sp);
 
 			printk(KERN_EMERG "eax = %08lx, ebx = %08lx, ecx = %08lx, edx = %08lx\n",
-				t->eax, t->ebx, t->ecx, t->edx);
+				t->ax, t->bx, t->cx, t->dx);
 			printk(KERN_EMERG "esi = %08lx, edi = %08lx\n",
-				t->esi, t->edi);
+				t->si, t->di);
 		}
 	}
 
@@ -50,15 +51,15 @@ static void doublefault_fn(void)
 
 struct tss_struct doublefault_tss __cacheline_aligned = {
 	.x86_tss = {
-		.esp0		= STACK_START,
+		.sp0		= STACK_START,
 		.ss0		= __KERNEL_DS,
 		.ldt		= 0,
 		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,
 
-		.eip		= (unsigned long) doublefault_fn,
+		.ip		= (unsigned long) doublefault_fn,
 		/* 0x2 bit is always set */
-		.eflags		= X86_EFLAGS_SF | 0x2,
-		.esp		= STACK_START,
+		.flags		= X86_EFLAGS_SF | 0x2,
+		.sp		= STACK_START,
 		.es		= __USER_DS,
 		.cs		= __KERNEL_CS,
 		.ss		= __KERNEL_DS,
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 153bb87a4eea..6a474e1028c7 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -288,7 +288,7 @@ ENTRY(ia32_sysenter_target)
 	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA esp, 0
 	CFI_REGISTER esp, ebp
-	movl TSS_sysenter_esp0(%esp),%esp
+	movl TSS_sysenter_sp0(%esp),%esp
 sysenter_past_esp:
 	/*
 	 * No need to follow this irqs on/off section: the syscall
@@ -743,7 +743,7 @@ END(device_not_available)
  * that sets up the real kernel stack. Check here, since we can't
  * allow the wrong stack to be used.
  *
- * "TSS_sysenter_esp0+12" is because the NMI/debug handler will have
+ * "TSS_sysenter_sp0+12" is because the NMI/debug handler will have
  * already pushed 3 words if it hits on the sysenter instruction:
  * eflags, cs and eip.
  *
@@ -755,7 +755,7 @@ END(device_not_available)
 	cmpw $__KERNEL_CS,4(%esp);		\
 	jne ok;					\
 label:						\
-	movl TSS_sysenter_esp0+offset(%esp),%esp;	\
+	movl TSS_sysenter_sp0+offset(%esp),%esp;	\
 	CFI_DEF_CFA esp, 0;			\
 	CFI_UNDEFINED eip;			\
 	pushfl;					\
diff --git a/arch/x86/kernel/paravirt_32.c b/arch/x86/kernel/paravirt_32.c
index 706b0562ea40..f4e3a8e01cf2 100644
--- a/arch/x86/kernel/paravirt_32.c
+++ b/arch/x86/kernel/paravirt_32.c
@@ -382,7 +382,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.write_ldt_entry = write_dt_entry,
 	.write_gdt_entry = write_dt_entry,
 	.write_idt_entry = write_dt_entry,
-	.load_esp0 = native_load_esp0,
+	.load_sp0 = native_load_sp0,
 
 	.irq_enable_syscall_ret = native_irq_enable_syscall_ret,
 	.iret = native_iret,
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 3744cf63682c..add3bf34e205 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -75,7 +75,7 @@ EXPORT_PER_CPU_SYMBOL(cpu_number);
  */
 unsigned long thread_saved_pc(struct task_struct *tsk)
 {
-	return ((unsigned long *)tsk->thread.esp)[3];
+	return ((unsigned long *)tsk->thread.sp)[3];
 }
 
 /*
@@ -488,10 +488,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 	childregs->ax = 0;
 	childregs->sp = sp;
 
-	p->thread.esp = (unsigned long) childregs;
-	p->thread.esp0 = (unsigned long) (childregs+1);
+	p->thread.sp = (unsigned long) childregs;
+	p->thread.sp0 = (unsigned long) (childregs+1);
 
-	p->thread.eip = (unsigned long) ret_from_fork;
+	p->thread.ip = (unsigned long) ret_from_fork;
 
 	savesegment(gs,p->thread.gs);
 
@@ -718,7 +718,7 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 	/*
 	 * Reload esp0.
 	 */
-	load_esp0(tss, next);
+	load_sp0(tss, next);
 
 	/*
 	 * Save away %gs. No need to save %fs, as it was saved on the
@@ -851,7 +851,7 @@ unsigned long get_wchan(struct task_struct *p)
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 	stack_page = (unsigned long)task_stack_page(p);
-	sp = p->thread.esp;
+	sp = p->thread.sp;
 	if (!stack_page || sp < stack_page || sp > top_esp+stack_page)
 		return 0;
 	/* include/asm-i386/system.h:switch_to() pushes bp last. */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index efbb1a2eab97..238193822e23 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -493,9 +493,9 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 	if (sp == ~0UL)
 		childregs->sp = (unsigned long)childregs;
 
-	p->thread.rsp = (unsigned long) childregs;
-	p->thread.rsp0 = (unsigned long) (childregs+1);
-	p->thread.userrsp = me->thread.userrsp; 
+	p->thread.sp = (unsigned long) childregs;
+	p->thread.sp0 = (unsigned long) (childregs+1);
+	p->thread.usersp = me->thread.usersp;
 
 	set_tsk_thread_flag(p, TIF_FORK);
 
@@ -607,7 +607,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/*
 	 * Reload esp0, LDT and the page table pointer:
 	 */
-	tss->rsp0 = next->rsp0;
+	tss->sp0 = next->sp0;
 
 	/* 
 	 * Switch DS and ES.
@@ -666,8 +666,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/* 
 	 * Switch the PDA and FPU contexts.
 	 */
-	prev->userrsp = read_pda(oldrsp); 
-	write_pda(oldrsp, next->userrsp); 
+	prev->usersp = read_pda(oldrsp);
+	write_pda(oldrsp, next->usersp);
 	write_pda(pcurrent, next_p); 
 
 	write_pda(kernelstack,
@@ -769,9 +769,9 @@ unsigned long get_wchan(struct task_struct *p)
 	if (!p || p == current || p->state==TASK_RUNNING)
 		return 0; 
 	stack = (unsigned long)task_stack_page(p);
-	if (p->thread.rsp < stack || p->thread.rsp > stack+THREAD_SIZE)
+	if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE)
 		return 0;
-	fp = *(u64 *)(p->thread.rsp);
+	fp = *(u64 *)(p->thread.sp);
 	do { 
 		if (fp < (unsigned long)stack ||
 		    fp > (unsigned long)stack+THREAD_SIZE)
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 3566191832b3..0f294d6e22cf 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -454,7 +454,7 @@ void __devinit initialize_secondary(void)
 		"movl %0,%%esp\n\t"
 		"jmp *%1"
 		:
-		:"m" (current->thread.esp),"m" (current->thread.eip));
+		:"m" (current->thread.sp),"m" (current->thread.ip));
 }
 
 /* Static state in head.S used to set up a CPU */
@@ -753,7 +753,7 @@ static inline struct task_struct * __cpuinit alloc_idle_task(int cpu)
 		/* initialize thread_struct.  we really want to avoid destroy
 		 * idle tread
 		 */
-		idle->thread.esp = (unsigned long)task_pt_regs(idle);
+		idle->thread.sp = (unsigned long)task_pt_regs(idle);
 		init_idle(idle, cpu);
 		return idle;
 	}
@@ -798,7 +798,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
  	per_cpu(current_task, cpu) = idle;
 	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
 
-	idle->thread.eip = (unsigned long) start_secondary;
+	idle->thread.ip = (unsigned long) start_secondary;
 	/* start_eip had better be page-aligned! */
 	start_eip = setup_trampoline();
 
@@ -808,7 +808,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
 	/* So we see what's up   */
 	printk("Booting processor %d/%d ip %lx\n", cpu, apicid, start_eip);
 	/* Stack for startup_32 can be just as for start_secondary onwards */
-	stack_start.sp = (void *) idle->thread.esp;
+	stack_start.sp = (void *) idle->thread.sp;
 
 	irq_ctx_init(cpu);
 
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index ac1089f2b917..c3f2736ba530 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -577,7 +577,7 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid)
 	c_idle.idle = get_idle_for_cpu(cpu);
 
 	if (c_idle.idle) {
-		c_idle.idle->thread.rsp = (unsigned long) (((struct pt_regs *)
+		c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *)
 			(THREAD_SIZE +  task_stack_page(c_idle.idle))) - 1);
 		init_idle(c_idle.idle, cpu);
 		goto do_rest;
@@ -613,8 +613,8 @@ do_rest:
 
 	start_rip = setup_trampoline();
 
-	init_rsp = c_idle.idle->thread.rsp;
-	per_cpu(init_tss,cpu).rsp0 = init_rsp;
+	init_rsp = c_idle.idle->thread.sp;
+	per_cpu(init_tss,cpu).sp0 = init_rsp;
 	initial_code = start_secondary;
 	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
 
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 27713553cc59..57491942cc4e 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -163,7 +163,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		unsigned long dummy;
 		stack = &dummy;
 		if (task != current)
-			stack = (unsigned long *)task->thread.esp;
+			stack = (unsigned long *)task->thread.sp;
 	}
 
 #ifdef CONFIG_FRAME_POINTER
@@ -173,7 +173,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 			asm ("movl %%ebp, %0" : "=r" (bp) : );
 		} else {
 			/* bp is the last reg pushed by switch_to */
-			bp = *(unsigned long *) task->thread.esp;
+			bp = *(unsigned long *) task->thread.sp;
 		}
 	}
 #endif
@@ -253,7 +253,7 @@ static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 
 	if (sp == NULL) {
 		if (task)
-			sp = (unsigned long*)task->thread.esp;
+			sp = (unsigned long*)task->thread.sp;
 		else
 			sp = (unsigned long *)&sp;
 	}
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index f7fecf9d47c3..965f2cc3a013 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -230,7 +230,7 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
 		unsigned long dummy;
 		stack = &dummy;
 		if (tsk && tsk != current)
-			stack = (unsigned long *)tsk->thread.rsp;
+			stack = (unsigned long *)tsk->thread.sp;
 	}
 
 	/*
@@ -366,7 +366,7 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp)
 
 	if (sp == NULL) {
 		if (tsk)
-			sp = (unsigned long *)tsk->thread.rsp;
+			sp = (unsigned long *)tsk->thread.sp;
 		else
 			sp = (unsigned long *)&sp;
 	}
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index 980e85b90091..e85bb44265cb 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -147,10 +147,10 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs)
 	}
 
 	tss = &per_cpu(init_tss, get_cpu());
-	current->thread.esp0 = current->thread.saved_esp0;
+	current->thread.sp0 = current->thread.saved_sp0;
 	current->thread.sysenter_cs = __KERNEL_CS;
-	load_esp0(tss, &current->thread);
-	current->thread.saved_esp0 = 0;
+	load_sp0(tss, &current->thread);
+	current->thread.saved_sp0 = 0;
 	put_cpu();
 
 	ret = KVM86->regs32;
@@ -207,7 +207,7 @@ asmlinkage int sys_vm86old(struct pt_regs regs)
 	int tmp, ret = -EPERM;
 
 	tsk = current;
-	if (tsk->thread.saved_esp0)
+	if (tsk->thread.saved_sp0)
 		goto out;
 	tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs,
 				       offsetof(struct kernel_vm86_struct, vm86plus) -
@@ -256,7 +256,7 @@ asmlinkage int sys_vm86(struct pt_regs regs)
 
 	/* we come here only for functions VM86_ENTER, VM86_ENTER_NO_BYPASS */
 	ret = -EPERM;
-	if (tsk->thread.saved_esp0)
+	if (tsk->thread.saved_sp0)
 		goto out;
 	v86 = (struct vm86plus_struct __user *)regs.cx;
 	tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs,
@@ -318,15 +318,15 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
  * Save old state, set default return value (%ax) to 0
  */
 	info->regs32->ax = 0;
-	tsk->thread.saved_esp0 = tsk->thread.esp0;
+	tsk->thread.saved_sp0 = tsk->thread.sp0;
 	tsk->thread.saved_fs = info->regs32->fs;
 	savesegment(gs, tsk->thread.saved_gs);
 
 	tss = &per_cpu(init_tss, get_cpu());
-	tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0;
+	tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0;
 	if (cpu_has_sep)
 		tsk->thread.sysenter_cs = 0;
-	load_esp0(tss, &tsk->thread);
+	load_sp0(tss, &tsk->thread);
 	put_cpu();
 
 	tsk->thread.screen_bitmap = info->screen_bitmap;
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 599b6f2ed562..4cfda7dbe90f 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -62,7 +62,7 @@ static struct {
 	void (*cpuid)(void /* non-c */);
 	void (*_set_ldt)(u32 selector);
 	void (*set_tr)(u32 selector);
-	void (*set_kernel_stack)(u32 selector, u32 esp0);
+	void (*set_kernel_stack)(u32 selector, u32 sp0);
 	void (*allocate_page)(u32, u32, u32, u32, u32);
 	void (*release_page)(u32, u32);
 	void (*set_pte)(pte_t, pte_t *, unsigned);
@@ -214,17 +214,17 @@ static void vmi_set_tr(void)
 	vmi_ops.set_tr(GDT_ENTRY_TSS*sizeof(struct desc_struct));
 }
 
-static void vmi_load_esp0(struct tss_struct *tss,
+static void vmi_load_sp0(struct tss_struct *tss,
 				   struct thread_struct *thread)
 {
-	tss->x86_tss.esp0 = thread->esp0;
+	tss->x86_tss.sp0 = thread->sp0;
 
 	/* This can only happen when SEP is enabled, no need to test "SEP"arately */
 	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
 		tss->x86_tss.ss1 = thread->sysenter_cs;
 		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
 	}
-	vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.esp0);
+	vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.sp0);
 }
 
 static void vmi_flush_tlb_user(void)
@@ -793,7 +793,7 @@ static inline int __init activate_vmi(void)
 	para_fill(pv_cpu_ops.write_ldt_entry, WriteLDTEntry);
 	para_fill(pv_cpu_ops.write_gdt_entry, WriteGDTEntry);
 	para_fill(pv_cpu_ops.write_idt_entry, WriteIDTEntry);
-	para_wrap(pv_cpu_ops.load_esp0, vmi_load_esp0, set_kernel_stack, UpdateKernelStack);
+	para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack);
 	para_fill(pv_cpu_ops.set_iopl_mask, SetIOPLMask);
 	para_fill(pv_cpu_ops.io_delay, IODelay);
 
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index ea46d05853bb..c751e3c03e85 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -755,10 +755,10 @@ static void lguest_time_init(void)
  * segment), the privilege level (we're privilege level 1, the Host is 0 and
  * will not tolerate us trying to use that), the stack pointer, and the number
  * of pages in the stack. */
-static void lguest_load_esp0(struct tss_struct *tss,
+static void lguest_load_sp0(struct tss_struct *tss,
 				     struct thread_struct *thread)
 {
-	lazy_hcall(LHCALL_SET_STACK, __KERNEL_DS|0x1, thread->esp0,
+	lazy_hcall(LHCALL_SET_STACK, __KERNEL_DS|0x1, thread->sp0,
 		   THREAD_SIZE/PAGE_SIZE);
 }
 
@@ -957,7 +957,7 @@ __init void lguest_init(void)
 	pv_cpu_ops.cpuid = lguest_cpuid;
 	pv_cpu_ops.load_idt = lguest_load_idt;
 	pv_cpu_ops.iret = lguest_iret;
-	pv_cpu_ops.load_esp0 = lguest_load_esp0;
+	pv_cpu_ops.load_sp0 = lguest_load_sp0;
 	pv_cpu_ops.load_tr_desc = lguest_load_tr_desc;
 	pv_cpu_ops.set_ldt = lguest_set_ldt;
 	pv_cpu_ops.load_tls = lguest_load_tls;
diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c
index d97a6d7d062b..e0feb66a2408 100644
--- a/arch/x86/vdso/vdso32-setup.c
+++ b/arch/x86/vdso/vdso32-setup.c
@@ -243,9 +243,9 @@ void enable_sep_cpu(void)
 	}
 
 	tss->x86_tss.ss1 = __KERNEL_CS;
-	tss->x86_tss.esp1 = sizeof(struct tss_struct) + (unsigned long) tss;
+	tss->x86_tss.sp1 = sizeof(struct tss_struct) + (unsigned long) tss;
 	wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
-	wrmsr(MSR_IA32_SYSENTER_ESP, tss->x86_tss.esp1, 0);
+	wrmsr(MSR_IA32_SYSENTER_ESP, tss->x86_tss.sp1, 0);
 	wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long) ia32_sysenter_target, 0);
 	put_cpu();	
 }
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 29517faaa735..d81e8d709102 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -499,11 +499,11 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
 	preempt_enable();
 }
 
-static void xen_load_esp0(struct tss_struct *tss,
+static void xen_load_sp0(struct tss_struct *tss,
 			  struct thread_struct *thread)
 {
 	struct multicall_space mcs = xen_mc_entry(0);
-	MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->esp0);
+	MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0);
 	xen_mc_issue(PARAVIRT_LAZY_CPU);
 }
 
@@ -968,7 +968,7 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.write_ldt_entry = xen_write_ldt_entry,
 	.write_gdt_entry = xen_write_gdt_entry,
 	.write_idt_entry = xen_write_idt_entry,
-	.load_esp0 = xen_load_esp0,
+	.load_sp0 = xen_load_sp0,
 
 	.set_iopl_mask = xen_set_iopl_mask,
 	.io_delay = xen_io_delay,
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 8e1234e14559..aafc54437403 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -239,10 +239,10 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	ctxt->gdt_ents      = ARRAY_SIZE(gdt->gdt);
 
 	ctxt->user_regs.cs = __KERNEL_CS;
-	ctxt->user_regs.esp = idle->thread.esp0 - sizeof(struct pt_regs);
+	ctxt->user_regs.esp = idle->thread.sp0 - sizeof(struct pt_regs);
 
 	ctxt->kernel_ss = __KERNEL_DS;
-	ctxt->kernel_sp = idle->thread.esp0;
+	ctxt->kernel_sp = idle->thread.sp0;
 
 	ctxt->event_callback_cs     = __KERNEL_CS;
 	ctxt->event_callback_eip    = (unsigned long)xen_hypervisor_callback;
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 96d0fd07c57d..44adb00e1490 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -94,7 +94,7 @@ static void copy_in_guest_info(struct lguest *lg, struct lguest_pages *pages)
 	/* Set up the two "TSS" members which tell the CPU what stack to use
 	 * for traps which do directly into the Guest (ie. traps at privilege
 	 * level 1). */
-	pages->state.guest_tss.esp1 = lg->esp1;
+	pages->state.guest_tss.sp1 = lg->esp1;
 	pages->state.guest_tss.ss1 = lg->ss1;
 
 	/* Copy direct-to-Guest trap entries. */
@@ -416,7 +416,7 @@ void __init lguest_arch_host_init(void)
 		/* We know where we want the stack to be when the Guest enters
 		 * the switcher: in pages->regs.  The stack grows upwards, so
 		 * we start it at the end of that structure. */
-		state->guest_tss.esp0 = (long)(&pages->regs + 1);
+		state->guest_tss.sp0 = (long)(&pages->regs + 1);
 		/* And this is the GDT entry to use for the stack: we keep a
 		 * couple of special LGUEST entries. */
 		state->guest_tss.ss0 = LGUEST_DS;
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index be7b934f6c54..d1780e32722e 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -101,7 +101,7 @@ struct pv_cpu_ops {
 				int entrynum, u32 low, u32 high);
 	void (*write_idt_entry)(struct desc_struct *,
 				int entrynum, u32 low, u32 high);
-	void (*load_esp0)(struct tss_struct *tss, struct thread_struct *t);
+	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
 
 	void (*set_iopl_mask)(unsigned mask);
 
@@ -449,10 +449,10 @@ static inline int paravirt_enabled(void)
 	return pv_info.paravirt_enabled;
 }
 
-static inline void load_esp0(struct tss_struct *tss,
+static inline void load_sp0(struct tss_struct *tss,
 			     struct thread_struct *thread)
 {
-	PVOP_VCALL2(pv_cpu_ops.load_esp0, tss, thread);
+	PVOP_VCALL2(pv_cpu_ops.load_sp0, tss, thread);
 }
 
 #define ARCH_SETUP			pv_init_ops.arch_setup();
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index d50a4b48d441..6846cc346400 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -292,20 +292,17 @@ struct thread_struct;
 /* This is the TSS defined by the hardware. */
 struct i386_hw_tss {
 	unsigned short	back_link,__blh;
-	unsigned long	esp0;
+	unsigned long	sp0;
 	unsigned short	ss0,__ss0h;
-	unsigned long	esp1;
+	unsigned long	sp1;
 	unsigned short	ss1,__ss1h;	/* ss1 is used to cache MSR_IA32_SYSENTER_CS */
-	unsigned long	esp2;
+	unsigned long	sp2;
 	unsigned short	ss2,__ss2h;
 	unsigned long	__cr3;
-	unsigned long	eip;
-	unsigned long	eflags;
-	unsigned long	eax,ecx,edx,ebx;
-	unsigned long	esp;
-	unsigned long	ebp;
-	unsigned long	esi;
-	unsigned long	edi;
+	unsigned long	ip;
+	unsigned long	flags;
+	unsigned long	ax, cx, dx, bx;
+	unsigned long	sp, bp, si, di;
 	unsigned short	es, __esh;
 	unsigned short	cs, __csh;
 	unsigned short	ss, __ssh;
@@ -346,10 +343,10 @@ struct tss_struct {
 struct thread_struct {
 /* cached TLS descriptors. */
 	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
-	unsigned long	esp0;
+	unsigned long	sp0;
 	unsigned long	sysenter_cs;
-	unsigned long	eip;
-	unsigned long	esp;
+	unsigned long	ip;
+	unsigned long	sp;
 	unsigned long	fs;
 	unsigned long	gs;
 /* Hardware debugging registers */
@@ -366,7 +363,7 @@ struct thread_struct {
 /* virtual 86 mode info */
 	struct vm86_struct __user * vm86_info;
 	unsigned long		screen_bitmap;
-	unsigned long		v86flags, v86mask, saved_esp0;
+	unsigned long		v86flags, v86mask, saved_sp0;
 	unsigned int		saved_fs, saved_gs;
 /* IO permissions */
 	unsigned long	*io_bitmap_ptr;
@@ -378,7 +375,7 @@ struct thread_struct {
 };
 
 #define INIT_THREAD  {							\
-	.esp0 = sizeof(init_stack) + (long)&init_stack,			\
+	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
 	.vm86_info = NULL,						\
 	.sysenter_cs = __KERNEL_CS,					\
 	.io_bitmap_ptr = NULL,						\
@@ -393,7 +390,7 @@ struct thread_struct {
  */
 #define INIT_TSS  {							\
 	.x86_tss = {							\
-		.esp0		= sizeof(init_stack) + (long)&init_stack, \
+		.sp0		= sizeof(init_stack) + (long)&init_stack, \
 		.ss0		= __KERNEL_DS,				\
 		.ss1		= __KERNEL_CS,				\
 		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		\
@@ -503,9 +500,9 @@ static inline void rep_nop(void)
 
 #define cpu_relax()	rep_nop()
 
-static inline void native_load_esp0(struct tss_struct *tss, struct thread_struct *thread)
+static inline void native_load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
-	tss->x86_tss.esp0 = thread->esp0;
+	tss->x86_tss.sp0 = thread->sp0;
 	/* This can only happen when SEP is enabled, no need to test "SEP"arately */
 	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
 		tss->x86_tss.ss1 = thread->sysenter_cs;
@@ -585,9 +582,9 @@ static inline void native_set_iopl_mask(unsigned mask)
 #define paravirt_enabled() 0
 #define __cpuid native_cpuid
 
-static inline void load_esp0(struct tss_struct *tss, struct thread_struct *thread)
+static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
-	native_load_esp0(tss, thread);
+	native_load_sp0(tss, thread);
 }
 
 /*
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 797770113e6d..0780f3e3fdfe 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -177,9 +177,9 @@ union i387_union {
 
 struct tss_struct {
 	u32 reserved1;
-	u64 rsp0;	
-	u64 rsp1;
-	u64 rsp2;
+	u64 sp0;
+	u64 sp1;
+	u64 sp2;
 	u64 reserved2;
 	u64 ist[7];
 	u32 reserved3;
@@ -216,9 +216,9 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #endif
 
 struct thread_struct {
-	unsigned long	rsp0;
-	unsigned long	rsp;
-	unsigned long 	userrsp;	/* Copy from PDA */ 
+	unsigned long	sp0;
+	unsigned long	sp;
+	unsigned long 	usersp;	/* Copy from PDA */
 	unsigned long	fs;
 	unsigned long	gs;
 	unsigned short	es, ds, fsindex, gsindex;	
@@ -245,11 +245,11 @@ struct thread_struct {
 } __attribute__((aligned(16)));
 
 #define INIT_THREAD  { \
-	.rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
 
 #define INIT_TSS  { \
-	.rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
 
 #define INIT_MMAP \
@@ -293,10 +293,10 @@ extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
  * Return saved PC of a blocked thread.
  * What is this good for? it will be always the scheduler or ret_from_fork.
  */
-#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.rsp - 8))
+#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8))
 
 extern unsigned long get_wchan(struct task_struct *p);
-#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.rsp0 - 1)
+#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
 #define KSTK_EIP(tsk) (task_pt_regs(tsk)->ip)
 #define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
 
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index db6283eb5e46..f5b3f77f5310 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -28,9 +28,9 @@ extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struc
 		     "1:\t"						\
 		     "popl %%ebp\n\t"					\
 		     "popfl"						\
-		     :"=m" (prev->thread.esp),"=m" (prev->thread.eip),	\
+		     :"=m" (prev->thread.sp),"=m" (prev->thread.ip),	\
 		      "=a" (last),"=S" (esi),"=D" (edi)			\
-		     :"m" (next->thread.esp),"m" (next->thread.eip),	\
+		     :"m" (next->thread.sp),"m" (next->thread.ip),	\
 		      "2" (prev), "d" (next));				\
 } while (0)
 
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 6e9e4841a2da..3dcb217a7202 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -40,7 +40,7 @@
 		     RESTORE_CONTEXT						    \
 		     : "=a" (last)					  	  \
 		     : [next] "S" (next), [prev] "D" (prev),			  \
-		       [threadrsp] "i" (offsetof(struct task_struct, thread.rsp)), \
+		       [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \
 		       [ti_flags] "i" (offsetof(struct thread_info, flags)),\
 		       [tif_fork] "i" (TIF_FORK),			  \
 		       [thread_info] "i" (offsetof(struct task_struct, stack)), \
-- 
cgit v1.2.3


From 1c6f70309900c47eeaef12c97b206fcccde1ebe2 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 30 Jan 2008 13:31:02 +0100
Subject: x86: define all _PAGE_* in terms of _PAGE_BIT_* on 64-bit

This patch defines the _PAGE_* paging attributes in pgtable_64.h in terms of
the former defined _PAGE_BIT_* values.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_64.h | 27 +++++++++++++++------------
 1 file changed, 15 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 9b0ff477b39e..9d4f11dd566f 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -150,19 +150,23 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define _PAGE_BIT_ACCESSED	5
 #define _PAGE_BIT_DIRTY		6
 #define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page */
+#define _PAGE_BIT_FILE		6
 #define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
-#define _PAGE_PRESENT	0x001
-#define _PAGE_RW	0x002
-#define _PAGE_USER	0x004
-#define _PAGE_PWT	0x008
-#define _PAGE_PCD	0x010
-#define _PAGE_ACCESSED	0x020
-#define _PAGE_DIRTY	0x040
-#define _PAGE_PSE	0x080	/* 2MB page */
-#define _PAGE_FILE	0x040	/* nonlinear file mapping, saved PTE; unset:swap */
-#define _PAGE_GLOBAL	0x100	/* Global TLB entry */
+#define _PAGE_PRESENT	(_AC(1,UL)<<_PAGE_BIT_PRESENT)
+#define _PAGE_RW	(_AC(1,UL)<<_PAGE_BIT_RW)
+#define _PAGE_USER	(_AC(1,UL)<<_PAGE_BIT_USER)
+#define _PAGE_PWT	(_AC(1,UL)<<_PAGE_BIT_PWT)
+#define _PAGE_PCD	(_AC(1,UL)<<_PAGE_BIT_PCD)
+#define _PAGE_ACCESSED	(_AC(1,UL)<<_PAGE_BIT_ACCESSED)
+#define _PAGE_DIRTY	(_AC(1,UL)<<_PAGE_BIT_DIRTY)
+/* 2MB page */
+#define _PAGE_PSE	(_AC(1,UL)<<_PAGE_BIT_PSE)
+/* nonlinear file mapping, saved PTE; unset:swap */
+#define _PAGE_FILE	(_AC(1,UL)<<_PAGE_BIT_FILE)
+/* Global TLB entry */
+#define _PAGE_GLOBAL	(_AC(1,UL)<<_PAGE_BIT_GLOBAL)
 
 #define _PAGE_PROTNONE	0x080	/* If not present */
 #define _PAGE_NX        (_AC(1,UL)<<_PAGE_BIT_NX)
@@ -248,8 +252,7 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 
-#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this
-						   right? */
+#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this right? */
 #define pte_page(x)	pfn_to_page(pte_pfn(x))
 #define pte_pfn(x)  ((pte_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
-- 
cgit v1.2.3


From 751de83c0c94a5235f14cff8549d3b39e745eb2b Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:03 +0100
Subject: x86: unify msr smp funcs

The functions under #ifdef CONFIG_SMP in msr.h are the same
for both x86_64 and i386, and this patches removes one of them,
putting them in a single location

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h | 33 +++++++--------------------------
 1 file changed, 7 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 80b027081b3c..5c95d0be1132 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -139,29 +139,6 @@ static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
 	} while(0)
 #endif	/* !CONFIG_PARAVIRT */
 
-#ifdef CONFIG_SMP
-void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
-void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
-int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
-int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
-#else  /*  CONFIG_SMP  */
-static inline void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h)
-{
-	rdmsr(msr_no, *l, *h);
-}
-static inline void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
-{
-	wrmsr(msr_no, l, h);
-}
-static inline int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h)
-{
-	return rdmsr_safe(msr_no, l, h);
-}
-static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
-{
-	return wrmsr_safe(msr_no, l, h);
-}
-#endif  /*  CONFIG_SMP  */
 #endif  /* ! __ASSEMBLY__ */
 #endif  /* __KERNEL__ */
 
@@ -327,6 +304,12 @@ static inline unsigned int cpuid_edx(unsigned int op)
 			:"c"(msr), "i"(-EIO), "0"(0));			\
 	  ret__; })
 
+#endif  /* __ASSEMBLY__ */
+
+#endif  /* !__i386__ */
+
+#ifndef __ASSEMBLY__
+
 #ifdef CONFIG_SMP
 void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
@@ -351,8 +334,6 @@ static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
 }
 #endif  /* CONFIG_SMP */
 #endif  /* __KERNEL__ */
-#endif  /* __ASSEMBLY__ */
-
-#endif  /* !__i386__ */
+#endif /* __ASSEMBLY__ */
 
 #endif
-- 
cgit v1.2.3


From 4e87173eacfd0d798aeeba14026893797826bc93 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:03 +0100
Subject: x86: split get_cycles_sync

This patch splits get_cycles_sync() into  __get_cycles_sync(),
and the rdtscll part. Paravirt guests cannot issue rdtscl directly,
as it involves a function call in vdso area.

So, using the __get_cycles_sync() base, we introduce vget_cycles_sync,
which then calls the native version of rdtscll. Ideally, however, a guest
should define its own clocksource, together with a vread function

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/tsc_64.c |  2 +-
 include/asm-x86/tsc.h    | 37 +++++++++++++++++++++++++++++++++----
 2 files changed, 34 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c
index 204a9080af04..3723401c4593 100644
--- a/arch/x86/kernel/tsc_64.c
+++ b/arch/x86/kernel/tsc_64.c
@@ -306,7 +306,7 @@ static cycle_t read_tsc(void)
 
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
-	cycle_t ret = (cycle_t)get_cycles_sync();
+	cycle_t ret = (cycle_t)vget_cycles_sync();
 	return ret;
 }
 
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index d7b1c4e7a108..9b7d264897fa 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -33,14 +33,14 @@ static inline cycles_t get_cycles(void)
 }
 
 /* Like get_cycles, but make sure the CPU is synchronized. */
-static __always_inline cycles_t get_cycles_sync(void)
+static __always_inline cycles_t __get_cycles_sync(void)
 {
 	unsigned long long ret;
 	unsigned eax, edx;
 
 	/*
-  	 * Use RDTSCP if possible; it is guaranteed to be synchronous
- 	 * and doesn't cause a VMEXIT on Hypervisors
+	 * Use RDTSCP if possible; it is guaranteed to be synchronous
+	 * and doesn't cause a VMEXIT on Hypervisors
 	 */
 	alternative_io(ASM_NOP3, ".byte 0x0f,0x01,0xf9", X86_FEATURE_RDTSCP,
 		       ASM_OUTPUT2("=a" (eax), "=d" (edx)),
@@ -55,11 +55,40 @@ static __always_inline cycles_t get_cycles_sync(void)
 	 */
 	alternative_io("cpuid", ASM_NOP2, X86_FEATURE_SYNC_RDTSC,
 			  "=a" (eax), "0" (1) : "ebx","ecx","edx","memory");
-	rdtscll(ret);
 
+	return 0;
+}
+
+static __always_inline cycles_t get_cycles_sync(void)
+{
+	unsigned long long ret;
+	ret = __get_cycles_sync();
+	if (!ret)
+		rdtscll(ret);
 	return ret;
 }
 
+#ifdef CONFIG_PARAVIRT
+/*
+ * For paravirt guests, some functionalities are executed through function
+ * pointers in the various pvops structures.
+ * These function pointers exist inside the kernel and can not
+ * be accessed by user space. To avoid this, we make a copy of the
+ * get_cycles_sync (called in kernel) but force the use of native_read_tsc.
+ * Ideally, the guest should set up it's own clock and vread
+ */
+static __always_inline long long vget_cycles_sync(void)
+{
+	unsigned long long ret;
+	ret = __get_cycles_sync();
+	if (!ret)
+		ret = native_read_tsc();
+	return ret;
+}
+#else
+# define vget_cycles_sync() get_cycles_sync()
+#endif
+
 extern void tsc_init(void);
 extern void mark_tsc_unstable(char *reason);
 extern int unsynchronized_tsc(void);
-- 
cgit v1.2.3


From c758ecf62ad94ddfeb4e7d8a5498bdcb2e3c85db Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:03 +0100
Subject: x86: unify cpuid functions

cpuid is not very different between i386 and x86_64.
We move away the x86_64 version from msr.h, and
unify them at processor.h, where they belong.

cpuid() paravirt then comes for free.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h          | 67 ------------------------------------
 include/asm-x86/processor.h    | 78 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 69 -------------------------------------
 3 files changed, 78 insertions(+), 136 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 5c95d0be1132..0648b2d57a38 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -203,73 +203,6 @@ static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
 			  : "c" (counter))
 
 
-static inline void cpuid(int op, unsigned int *eax, unsigned int *ebx,
-			 unsigned int *ecx, unsigned int *edx)
-{
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op));
-}
-
-/* Some CPUID calls want 'count' to be placed in ecx */
-static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
-			       int *edx)
-{
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op), "c" (count));
-}
-
-/*
- * CPUID functions returning a single datum
- */
-static inline unsigned int cpuid_eax(unsigned int op)
-{
-	unsigned int eax;
-
-	__asm__("cpuid"
-		: "=a" (eax)
-		: "0" (op)
-		: "bx", "cx", "dx");
-	return eax;
-}
-static inline unsigned int cpuid_ebx(unsigned int op)
-{
-	unsigned int eax, ebx;
-
-	__asm__("cpuid"
-		: "=a" (eax), "=b" (ebx)
-		: "0" (op)
-		: "cx", "dx" );
-	return ebx;
-}
-static inline unsigned int cpuid_ecx(unsigned int op)
-{
-	unsigned int eax, ecx;
-
-	__asm__("cpuid"
-		: "=a" (eax), "=c" (ecx)
-		: "0" (op)
-		: "bx", "dx" );
-	return ecx;
-}
-static inline unsigned int cpuid_edx(unsigned int op)
-{
-	unsigned int eax, edx;
-
-	__asm__("cpuid"
-		: "=a" (eax), "=d" (edx)
-		: "0" (op)
-		: "bx", "cx");
-	return edx;
-}
-
 #ifdef __KERNEL__
 
 /* wrmsr with exception handling */
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 46e1c04e309c..dea81b70895d 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -1,5 +1,83 @@
+#ifndef __ASM_X86_PROCESSOR_H
+#define __ASM_X86_PROCESSOR_H
+
+static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
+					 unsigned int *ecx, unsigned int *edx)
+{
+	/* ecx is often an input as well as an output. */
+	__asm__("cpuid"
+		: "=a" (*eax),
+		  "=b" (*ebx),
+		  "=c" (*ecx),
+		  "=d" (*edx)
+		: "0" (*eax), "2" (*ecx));
+}
+
+
 #ifdef CONFIG_X86_32
 # include "processor_32.h"
 #else
 # include "processor_64.h"
 #endif
+
+#ifndef CONFIG_PARAVIRT
+#define __cpuid native_cpuid
+#endif
+
+/*
+ * Generic CPUID function
+ * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
+ * resulting in stale register contents being returned.
+ */
+static inline void cpuid(unsigned int op,
+			 unsigned int *eax, unsigned int *ebx,
+			 unsigned int *ecx, unsigned int *edx)
+{
+	*eax = op;
+	*ecx = 0;
+	__cpuid(eax, ebx, ecx, edx);
+}
+
+/* Some CPUID calls want 'count' to be placed in ecx */
+static inline void cpuid_count(unsigned int op, int count,
+			       unsigned int *eax, unsigned int *ebx,
+			       unsigned int *ecx, unsigned int *edx)
+{
+	*eax = op;
+	*ecx = count;
+	__cpuid(eax, ebx, ecx, edx);
+}
+
+/*
+ * CPUID functions returning a single datum
+ */
+static inline unsigned int cpuid_eax(unsigned int op)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	cpuid(op, &eax, &ebx, &ecx, &edx);
+	return eax;
+}
+static inline unsigned int cpuid_ebx(unsigned int op)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	cpuid(op, &eax, &ebx, &ecx, &edx);
+	return ebx;
+}
+static inline unsigned int cpuid_ecx(unsigned int op)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	cpuid(op, &eax, &ebx, &ecx, &edx);
+	return ecx;
+}
+static inline unsigned int cpuid_edx(unsigned int op)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	cpuid(op, &eax, &ebx, &ecx, &edx);
+	return edx;
+}
+
+#endif
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 6846cc346400..0d83da198127 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -133,18 +133,6 @@ extern void detect_ht(struct cpuinfo_x86 *c);
 static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #endif
 
-static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
-					 unsigned int *ecx, unsigned int *edx)
-{
-	/* ecx is often an input as well as an output. */
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (*eax), "2" (*ecx));
-}
-
 #define load_cr3(pgdir) write_cr3(__pa(pgdir))
 
 /*
@@ -580,7 +568,6 @@ static inline void native_set_iopl_mask(unsigned mask)
 #include <asm/paravirt.h>
 #else
 #define paravirt_enabled() 0
-#define __cpuid native_cpuid
 
 static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
@@ -598,62 +585,6 @@ static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread
 #define set_iopl_mask native_set_iopl_mask
 #endif /* CONFIG_PARAVIRT */
 
-/*
- * Generic CPUID function
- * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
- * resulting in stale register contents being returned.
- */
-static inline void cpuid(unsigned int op,
-			 unsigned int *eax, unsigned int *ebx,
-			 unsigned int *ecx, unsigned int *edx)
-{
-	*eax = op;
-	*ecx = 0;
-	__cpuid(eax, ebx, ecx, edx);
-}
-
-/* Some CPUID calls want 'count' to be placed in ecx */
-static inline void cpuid_count(unsigned int op, int count,
-			       unsigned int *eax, unsigned int *ebx,
-			       unsigned int *ecx, unsigned int *edx)
-{
-	*eax = op;
-	*ecx = count;
-	__cpuid(eax, ebx, ecx, edx);
-}
-
-/*
- * CPUID functions returning a single datum
- */
-static inline unsigned int cpuid_eax(unsigned int op)
-{
-	unsigned int eax, ebx, ecx, edx;
-
-	cpuid(op, &eax, &ebx, &ecx, &edx);
-	return eax;
-}
-static inline unsigned int cpuid_ebx(unsigned int op)
-{
-	unsigned int eax, ebx, ecx, edx;
-
-	cpuid(op, &eax, &ebx, &ecx, &edx);
-	return ebx;
-}
-static inline unsigned int cpuid_ecx(unsigned int op)
-{
-	unsigned int eax, ebx, ecx, edx;
-
-	cpuid(op, &eax, &ebx, &ecx, &edx);
-	return ecx;
-}
-static inline unsigned int cpuid_edx(unsigned int op)
-{
-	unsigned int eax, ebx, ecx, edx;
-
-	cpuid(op, &eax, &ebx, &ecx, &edx);
-	return edx;
-}
-
 /* generic versions from gas */
 #define GENERIC_NOP1	".byte 0x90\n"
 #define GENERIC_NOP2    	".byte 0x89,0xf6\n"
-- 
cgit v1.2.3


From 8f12dea6135d0a55b151dcb4c6bbe211f5f8d35d Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:06 +0100
Subject: x86: introduce native_read_tscp

Targetting paravirt, this patch introduces native_read_tscp, in
place of rdtscp() macro. When in a paravirt guest, this will
involve a function call, and thus, cannot be done in the vdso area.
These users then have to call the native version directly

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/vsyscall_64.c |  4 ++--
 arch/x86/vdso/vgetcpu.c       |  4 ++--
 include/asm-x86/msr.h         | 29 +++++++++++++++++++++--------
 3 files changed, 25 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 018f7cf33790..c4c5e765cd2c 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -190,7 +190,7 @@ time_t __vsyscall(1) vtime(time_t *t)
 long __vsyscall(2)
 vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 {
-	unsigned int dummy, p;
+	unsigned int p;
 	unsigned long j = 0;
 
 	/* Fast cache - only recompute value once per jiffies and avoid
@@ -205,7 +205,7 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 		p = tcache->blob[1];
 	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
 		/* Load per CPU data from RDTSCP */
-		rdtscp(dummy, dummy, p);
+		native_read_tscp(&p);
 	} else {
 		/* Load per CPU data from GDT */
 		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
diff --git a/arch/x86/vdso/vgetcpu.c b/arch/x86/vdso/vgetcpu.c
index 3b1ae1abfba9..c8097f17f8a9 100644
--- a/arch/x86/vdso/vgetcpu.c
+++ b/arch/x86/vdso/vgetcpu.c
@@ -15,11 +15,11 @@
 
 long __vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused)
 {
-	unsigned int dummy, p;
+	unsigned int p;
 
 	if (*vdso_vgetcpu_mode == VGETCPU_RDTSCP) {
 		/* Load per CPU data from RDTSCP */
-		rdtscp(dummy, dummy, p);
+		native_read_tscp(&p);
 	} else {
 		/* Load per CPU data from GDT */
 		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 0648b2d57a38..b6262e99fc8e 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -7,6 +7,27 @@
 # include <linux/types.h>
 #endif
 
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+static inline unsigned long long native_read_tscp(int *aux)
+{
+	unsigned long low, high;
+	asm volatile (".byte 0x0f,0x01,0xf9"
+		      : "=a" (low), "=d" (high), "=c" (*aux));
+	return low | ((u64)high >> 32);
+}
+
+#define rdtscp(low, high, aux)						\
+       do {                                                            \
+		unsigned long long _val = native_read_tscp(&(aux));     \
+		(low) = (u32)_val;                                      \
+		(high) = (u32)(_val >> 32);                             \
+       } while (0)
+
+#define rdtscpll(val, aux) (val) = native_read_tscp(&(aux))
+#endif
+#endif
+
 #ifdef __i386__
 
 #ifdef __KERNEL__
@@ -178,8 +199,6 @@ static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
 #define rdtscl(low) \
      __asm__ __volatile__ ("rdtsc" : "=a" (low) : : "edx")
 
-#define rdtscp(low,high,aux) \
-     __asm__ __volatile__ (".byte 0x0f,0x01,0xf9" : "=a" (low), "=d" (high), "=c" (aux))
 
 #define rdtscll(val) do { \
      unsigned int __a,__d; \
@@ -187,12 +206,6 @@ static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
      (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
 } while(0)
 
-#define rdtscpll(val, aux) do { \
-     unsigned long __a, __d; \
-     __asm__ __volatile__ (".byte 0x0f,0x01,0xf9" : "=a" (__a), "=d" (__d), "=c" (aux)); \
-     (val) = (__d << 32) | __a; \
-} while (0)
-
 #define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
 
 #define write_rdtscp_aux(val) wrmsr(0xc0000103, val, 0)
-- 
cgit v1.2.3


From b8d1fae7dbde6a1227fa142acecb48dc3dd63817 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: change rdpmc interface

the rdpmc instruction gets a counter argument in rcx. However,
the i386 version was ignoring it. To make both x86_64 and i386 versions
the same, as well as to comply with the instruction semantics, this
parameter is added in the i386 version

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h      | 6 +++---
 include/asm-x86/paravirt.h | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index b6262e99fc8e..effb7319c0a5 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -94,10 +94,10 @@ static inline unsigned long long native_read_tsc(void)
 	return val;
 }
 
-static inline unsigned long long native_read_pmc(void)
+static inline unsigned long long native_read_pmc(int counter)
 {
 	unsigned long long val;
-	asm volatile("rdpmc" : "=A" (val));
+	asm volatile("rdpmc" : "=A" (val) : "c" (counter));
 	return val;
 }
 
@@ -154,7 +154,7 @@ static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
 
 #define rdpmc(counter,low,high)					\
 	do {							\
-		u64 _l = native_read_pmc();			\
+		u64 _l = native_read_pmc(counter);		\
 		(low)  = (u32)_l;				\
 		(high) = (u32)(_l >> 32);			\
 	} while(0)
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index d1780e32722e..4782be68d0fa 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -118,7 +118,7 @@ struct pv_cpu_ops {
 	int (*write_msr)(unsigned int msr, u64 val);
 
 	u64 (*read_tsc)(void);
-	u64 (*read_pmc)(void);
+	u64 (*read_pmc)(int counter);
 
 	/* These two are jmp to, not actually called. */
 	void (*irq_enable_syscall_ret)(void);
-- 
cgit v1.2.3


From c9dcda5ce46c395c5c99003e259e1973dface640 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: change write msr functions interface

This patche changes the native_write_msr() and friends interface
to explicitly take 2 32-bit registers instead of a 64-bit value.
The change will ease the merge with 64-bit code. As the 64-bit
value will be passed as two registers anyway in i386,
the PVOP_CALL interface has to account for that and use low/high parameters
It would force the x86_64 version to be different.

The change does not make i386 generated code less efficient. As said above,
it would get the values from two registers anyway.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h      | 19 ++++++++++---------
 include/asm-x86/paravirt.h |  2 +-
 2 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index effb7319c0a5..cb7222358897 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -63,13 +63,14 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 	return val;
 }
 
-static inline void native_write_msr(unsigned int msr, unsigned long long val)
+static inline void native_write_msr(unsigned int msr,
+				    unsigned low, unsigned high)
 {
-	asm volatile("wrmsr" : : "c" (msr), "A"(val));
+	asm volatile("wrmsr" : : "c" (msr), "a"(low), "d" (high));
 }
 
 static inline int native_write_msr_safe(unsigned int msr,
-					unsigned long long val)
+					unsigned low, unsigned high)
 {
 	int err;
 	asm volatile("2: wrmsr ; xorl %0,%0\n"
@@ -82,7 +83,7 @@ static inline int native_write_msr_safe(unsigned int msr,
 		     "   .long	2b,3b\n\t"
 		     ".previous"
 		     : "=a" (err)
-		     : "c" (msr), "0" ((u32)val), "d" ((u32)(val>>32)),
+		     : "c" (msr), "0" (low), "d" (high),
 		       "i" (-EFAULT));
 	return err;
 }
@@ -118,20 +119,20 @@ static inline unsigned long long native_read_pmc(int counter)
 		(val2) = (u32)(__val >> 32);				\
 	} while(0)
 
-static inline void wrmsr(u32 __msr, u32 __low, u32 __high)
+static inline void wrmsr(unsigned msr, unsigned low, unsigned high)
 {
-	native_write_msr(__msr, ((u64)__high << 32) | __low);
+	native_write_msr(msr, low, high);
 }
 
 #define rdmsrl(msr,val)							\
 	((val) = native_read_msr(msr))
 
-#define wrmsrl(msr,val)	native_write_msr(msr, val)
+#define wrmsrl(msr, val) native_write_msr(msr, (u32)val, (u32)(val >> 32))
 
 /* wrmsr with exception handling */
-static inline int wrmsr_safe(u32 __msr, u32 __low, u32 __high)
+static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 {
-	return native_write_msr_safe(__msr, ((u64)__high << 32) | __low);
+	return native_write_msr_safe(msr, low, high);
 }
 
 /* rdmsr with exception handling */
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 4782be68d0fa..e95c2a655165 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -115,7 +115,7 @@ struct pv_cpu_ops {
 	/* MSR, PMC and TSR operations.
 	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
 	u64 (*read_msr)(unsigned int msr, int *err);
-	int (*write_msr)(unsigned int msr, u64 val);
+	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
 
 	u64 (*read_tsc)(void);
 	u64 (*read_pmc)(int counter);
-- 
cgit v1.2.3


From 56ec1ddcff967e51d98427e4efcbfc90de67efe3 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: make fixups wordsize agnostic

This patch uses the _ASM_ALIGN and _ASM_PTR macros
to make the fixups in native_read/write_msr_safe look the same
for x86_64 and i386. Besides using this macros, we also have to
take the explicit instruction suffixes out. It's okay
because all this instructions uses registers, and can be sized by
them.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index cb7222358897..792fde2e8908 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -33,6 +33,7 @@ static inline unsigned long long native_read_tscp(int *aux)
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
+#include <asm/asm.h>
 #include <asm/errno.h>
 
 static inline unsigned long long native_read_msr(unsigned int msr)
@@ -48,14 +49,14 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 {
 	unsigned long long val;
 
-	asm volatile("2: rdmsr ; xorl %0,%0\n"
+	asm volatile("2: rdmsr ; xor %0,%0\n"
 		     "1:\n\t"
 		     ".section .fixup,\"ax\"\n\t"
-		     "3:  movl %3,%0 ; jmp 1b\n\t"
+		     "3:  mov %3,%0 ; jmp 1b\n\t"
 		     ".previous\n\t"
 		     ".section __ex_table,\"a\"\n"
-		     "   .align 4\n\t"
-		     "   .long	2b,3b\n\t"
+		     _ASM_ALIGN "\n\t"
+		     _ASM_PTR " 2b,3b\n\t"
 		     ".previous"
 		     : "=r" (*err), "=A" (val)
 		     : "c" (msr), "i" (-EFAULT));
@@ -73,14 +74,14 @@ static inline int native_write_msr_safe(unsigned int msr,
 					unsigned low, unsigned high)
 {
 	int err;
-	asm volatile("2: wrmsr ; xorl %0,%0\n"
+	asm volatile("2: wrmsr ; xor %0,%0\n"
 		     "1:\n\t"
 		     ".section .fixup,\"ax\"\n\t"
-		     "3:  movl %4,%0 ; jmp 1b\n\t"
+		     "3:  mov %4,%0 ; jmp 1b\n\t"
 		     ".previous\n\t"
 		     ".section __ex_table,\"a\"\n"
-		     "   .align 4\n\t"
-		     "   .long	2b,3b\n\t"
+		     _ASM_ALIGN "\n\t"
+		     _ASM_PTR " 2b,3b\n\t"
 		     ".previous"
 		     : "=a" (err)
 		     : "c" (msr), "0" (low), "d" (high),
-- 
cgit v1.2.3


From c210d24986dc19e387c10508c0bc2faadadc1a2e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: integrate 32-bit and 64-bit code in msr.h

This patches proceeds with the integration of msr.h, making
the code unified, instead of having a version for each architecture.
We stick with the native_* functions, and then paravirt comes for free.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h | 167 ++++++++++++++------------------------------------
 1 file changed, 47 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 792fde2e8908..040d3910d891 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -9,6 +9,10 @@
 
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
+
+#include <asm/asm.h>
+#include <asm/errno.h>
+
 static inline unsigned long long native_read_tscp(int *aux)
 {
 	unsigned long low, high;
@@ -17,37 +21,36 @@ static inline unsigned long long native_read_tscp(int *aux)
 	return low | ((u64)high >> 32);
 }
 
-#define rdtscp(low, high, aux)						\
-       do {                                                            \
-		unsigned long long _val = native_read_tscp(&(aux));     \
-		(low) = (u32)_val;                                      \
-		(high) = (u32)(_val >> 32);                             \
-       } while (0)
-
-#define rdtscpll(val, aux) (val) = native_read_tscp(&(aux))
-#endif
+/*
+ * i386 calling convention returns 64-bit value in edx:eax, while
+ * x86_64 returns at rax. Also, the "A" constraint does not really
+ * mean rdx:rax in x86_64, so we need specialized behaviour for each
+ * architecture
+ */
+#ifdef CONFIG_X86_64
+#define DECLARE_ARGS(val, low, high)	unsigned low, high
+#define EAX_EDX_VAL(val, low, high)	(low | ((u64)(high) << 32))
+#define EAX_EDX_ARGS(val, low, high)	"a" (low), "d" (high)
+#define EAX_EDX_RET(val, low, high)	"=a" (low), "=d" (high)
+#else
+#define DECLARE_ARGS(val, low, high)	unsigned long long val
+#define EAX_EDX_VAL(val, low, high)	(val)
+#define EAX_EDX_ARGS(val, low, high)	"A" (val)
+#define EAX_EDX_RET(val, low, high)	"=A" (val)
 #endif
 
-#ifdef __i386__
-
-#ifdef __KERNEL__
-#ifndef __ASSEMBLY__
-
-#include <asm/asm.h>
-#include <asm/errno.h>
-
 static inline unsigned long long native_read_msr(unsigned int msr)
 {
-	unsigned long long val;
+	DECLARE_ARGS(val, low, high);
 
-	asm volatile("rdmsr" : "=A" (val) : "c" (msr));
-	return val;
+	asm volatile("rdmsr" : EAX_EDX_RET(val, low, high) : "c" (msr));
+	return EAX_EDX_VAL(val, low, high);
 }
 
 static inline unsigned long long native_read_msr_safe(unsigned int msr,
 						      int *err)
 {
-	unsigned long long val;
+	DECLARE_ARGS(val, low, high);
 
 	asm volatile("2: rdmsr ; xor %0,%0\n"
 		     "1:\n\t"
@@ -58,10 +61,9 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 		     _ASM_ALIGN "\n\t"
 		     _ASM_PTR " 2b,3b\n\t"
 		     ".previous"
-		     : "=r" (*err), "=A" (val)
+		     : "=r" (*err), EAX_EDX_RET(val, low, high)
 		     : "c" (msr), "i" (-EFAULT));
-
-	return val;
+	return EAX_EDX_VAL(val, low, high);
 }
 
 static inline void native_write_msr(unsigned int msr,
@@ -91,16 +93,18 @@ static inline int native_write_msr_safe(unsigned int msr,
 
 static inline unsigned long long native_read_tsc(void)
 {
-	unsigned long long val;
-	asm volatile("rdtsc" : "=A" (val));
-	return val;
+	DECLARE_ARGS(val, low, high);
+
+	asm volatile("rdtsc" : EAX_EDX_RET(val, low, high));
+	return EAX_EDX_VAL(val, low, high);
 }
 
 static inline unsigned long long native_read_pmc(int counter)
 {
-	unsigned long long val;
-	asm volatile("rdpmc" : "=A" (val) : "c" (counter));
-	return val;
+	DECLARE_ARGS(val, low, high);
+
+	asm volatile("rdpmc" : EAX_EDX_RET(val, low, high) : "c" (counter));
+	return EAX_EDX_VAL(val, low, high);
 }
 
 #ifdef CONFIG_PARAVIRT
@@ -128,7 +132,8 @@ static inline void wrmsr(unsigned msr, unsigned low, unsigned high)
 #define rdmsrl(msr,val)							\
 	((val) = native_read_msr(msr))
 
-#define wrmsrl(msr, val) native_write_msr(msr, (u32)val, (u32)(val >> 32))
+#define wrmsrl(msr, val)						\
+	native_write_msr(msr, (u32)((u64)(val)), (u32)((u64)(val) >> 32))
 
 /* wrmsr with exception handling */
 static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
@@ -160,104 +165,25 @@ static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 		(low)  = (u32)_l;				\
 		(high) = (u32)(_l >> 32);			\
 	} while(0)
-#endif	/* !CONFIG_PARAVIRT */
-
-#endif  /* ! __ASSEMBLY__ */
-#endif  /* __KERNEL__ */
-
-#else   /* __i386__ */
-
-#ifndef __ASSEMBLY__
-#include <linux/errno.h>
-/*
- * Access to machine-specific registers (available on 586 and better only)
- * Note: the rd* operations modify the parameters directly (without using
- * pointer indirection), this allows gcc to optimize better
- */
-
-#define rdmsr(msr,val1,val2) \
-       __asm__ __volatile__("rdmsr" \
-			    : "=a" (val1), "=d" (val2) \
-			    : "c" (msr))
-
 
-#define rdmsrl(msr,val) do { unsigned long a__,b__; \
-       __asm__ __volatile__("rdmsr" \
-			    : "=a" (a__), "=d" (b__) \
-			    : "c" (msr)); \
-       val = a__ | (b__<<32); \
-} while(0)
-
-#define wrmsr(msr,val1,val2) \
-     __asm__ __volatile__("wrmsr" \
-			  : /* no outputs */ \
-			  : "c" (msr), "a" (val1), "d" (val2))
-
-#define wrmsrl(msr,val) wrmsr(msr,(__u32)((__u64)(val)),((__u64)(val))>>32)
+#define rdtscp(low, high, aux)						\
+       do {                                                            \
+		unsigned long long _val = native_read_tscp(&(aux));     \
+		(low) = (u32)_val;                                      \
+		(high) = (u32)(_val >> 32);                             \
+       } while (0)
 
-#define rdtsc(low,high) \
-     __asm__ __volatile__("rdtsc" : "=a" (low), "=d" (high))
+#define rdtscpll(val, aux) (val) = native_read_tscp(&(aux))
 
-#define rdtscl(low) \
-     __asm__ __volatile__ ("rdtsc" : "=a" (low) : : "edx")
+#endif	/* !CONFIG_PARAVIRT */
 
 
-#define rdtscll(val) do { \
-     unsigned int __a,__d; \
-     __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \
-     (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
-} while(0)
+#define checking_wrmsrl(msr,val) wrmsr_safe(msr,(u32)(val),(u32)((val)>>32))
 
 #define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
 
 #define write_rdtscp_aux(val) wrmsr(0xc0000103, val, 0)
 
-#define rdpmc(counter,low,high) \
-     __asm__ __volatile__("rdpmc" \
-			  : "=a" (low), "=d" (high) \
-			  : "c" (counter))
-
-
-#ifdef __KERNEL__
-
-/* wrmsr with exception handling */
-#define wrmsr_safe(msr,a,b) ({ int ret__;			\
-	asm volatile("2: wrmsr ; xorl %0,%0\n"			\
-		     "1:\n\t"					\
-		     ".section .fixup,\"ax\"\n\t"		\
-		     "3:  movl %4,%0 ; jmp 1b\n\t"		\
-		     ".previous\n\t"				\
-		     ".section __ex_table,\"a\"\n"		\
-		     "   .align 8\n\t"				\
-		     "   .quad	2b,3b\n\t"			\
-		     ".previous"				\
-		     : "=a" (ret__)				\
-		     : "c" (msr), "0" (a), "d" (b), "i" (-EFAULT)); \
-	ret__; })
-
-#define checking_wrmsrl(msr,val) wrmsr_safe(msr,(u32)(val),(u32)((val)>>32))
-
-#define rdmsr_safe(msr,a,b) \
-	({ int ret__;						\
-	  asm volatile ("1:       rdmsr\n"			\
-			"2:\n"					\
-			".section .fixup,\"ax\"\n"		\
-			"3:       movl %4,%0\n"			\
-			" jmp 2b\n"				\
-			".previous\n"				\
-			".section __ex_table,\"a\"\n"		\
-			" .align 8\n"				\
-			" .quad 1b,3b\n"				\
-			".previous":"=&bDS" (ret__), "=a"(*(a)), "=d"(*(b)) \
-			:"c"(msr), "i"(-EIO), "0"(0));			\
-	  ret__; })
-
-#endif  /* __ASSEMBLY__ */
-
-#endif  /* !__i386__ */
-
-#ifndef __ASSEMBLY__
-
 #ifdef CONFIG_SMP
 void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
@@ -281,7 +207,8 @@ static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
 	return wrmsr_safe(msr_no, l, h);
 }
 #endif  /* CONFIG_SMP */
-#endif  /* __KERNEL__ */
 #endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+
 
 #endif
-- 
cgit v1.2.3


From 82f74e7159749cc511ebf5954a7b9ea6ad634949 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: unify include/asm-x86/linkage_[32|64].h

Remove definitions of FASTCALL/fastcall from linkage_32 as compiled with
-regparm=3 by default since 2.6.20 and should no longer be needed.

CONFIG X86_64 and CONFIG_X86_ALIGNMENT_16 are mutually exclusive as found
in Kconfig.cpu so it should be fine to test them separately.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/linkage.h    | 21 ++++++++++++++++++---
 include/asm-x86/linkage_32.h | 15 ---------------
 include/asm-x86/linkage_64.h |  6 ------
 3 files changed, 18 insertions(+), 24 deletions(-)
 delete mode 100644 include/asm-x86/linkage_32.h
 delete mode 100644 include/asm-x86/linkage_64.h

(limited to 'include')

diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h
index 94b257fa8701..5a4c95905420 100644
--- a/include/asm-x86/linkage.h
+++ b/include/asm-x86/linkage.h
@@ -1,5 +1,20 @@
+#ifndef __ASM_LINKAGE_H
+#define __ASM_LINKAGE_H
+
+#ifdef CONFIG_X86_64
+#define __ALIGN .p2align 4,,15
+#define __ALIGN_STR ".p2align 4,,15"
+#endif
+
 #ifdef CONFIG_X86_32
-# include "linkage_32.h"
-#else
-# include "linkage_64.h"
+#define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
+#define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret))
+#endif
+
+#ifdef CONFIG_X86_ALIGNMENT_16
+#define __ALIGN .align 16,0x90
+#define __ALIGN_STR ".align 16,0x90"
+#endif
+
 #endif
+
diff --git a/include/asm-x86/linkage_32.h b/include/asm-x86/linkage_32.h
deleted file mode 100644
index f4a6ebac0247..000000000000
--- a/include/asm-x86/linkage_32.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef __ASM_LINKAGE_H
-#define __ASM_LINKAGE_H
-
-#define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
-#define FASTCALL(x)	x __attribute__((regparm(3)))
-#define fastcall	__attribute__((regparm(3)))
-
-#define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret))
-
-#ifdef CONFIG_X86_ALIGNMENT_16
-#define __ALIGN .align 16,0x90
-#define __ALIGN_STR ".align 16,0x90"
-#endif
-
-#endif
diff --git a/include/asm-x86/linkage_64.h b/include/asm-x86/linkage_64.h
deleted file mode 100644
index b5f39d0189ce..000000000000
--- a/include/asm-x86/linkage_64.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __ASM_LINKAGE_H
-#define __ASM_LINKAGE_H
-
-#define __ALIGN .p2align 4,,15
-
-#endif
-- 
cgit v1.2.3


From cc503c1b43e002e3f1fed70f46d947e2bf349bb6 Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Wed, 30 Jan 2008 13:31:07 +0100
Subject: x86: PIE executable randomization

main executable of (specially compiled/linked -pie/-fpie) ET_DYN binaries
onto a random address (in cases in which mmap() is allowed to perform a
randomization).

The code has been extraced from Ingo's exec-shield patch
http://people.redhat.com/mingo/exec-shield/

[akpm@linux-foundation.org: fix used-uninitialsied warning]
[kamezawa.hiroyu@jp.fujitsu.com: fixed ia32 ELF on x86_64 handling]

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/ia64/ia32/binfmt_elf32.c |   2 +-
 arch/x86/kernel/sys_x86_64.c  |  98 ++++++++++++++++++++++++++++++++++
 arch/x86/mm/mmap_64.c         | 119 ++++++++++++++++++++++++++++++++++++------
 fs/binfmt_elf.c               | 107 +++++++++++++++++++++++++++++--------
 include/asm-x86/pgtable_64.h  |   1 +
 5 files changed, 287 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c
index 3e35987af458..2a662215359c 100644
--- a/arch/ia64/ia32/binfmt_elf32.c
+++ b/arch/ia64/ia32/binfmt_elf32.c
@@ -222,7 +222,7 @@ elf32_set_personality (void)
 }
 
 static unsigned long
-elf32_map (struct file *filep, unsigned long addr, struct elf_phdr *eppnt, int prot, int type)
+elf32_map (struct file *filep, unsigned long addr, struct elf_phdr *eppnt, int prot, int type, unsigned long unused)
 {
 	unsigned long pgoff = (eppnt->p_vaddr) & ~IA32_PAGE_MASK;
 
diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
index 907942ee6e76..95485e63fd2f 100644
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -12,6 +12,7 @@
 #include <linux/file.h>
 #include <linux/utsname.h>
 #include <linux/personality.h>
+#include <linux/random.h>
 
 #include <asm/uaccess.h>
 #include <asm/ia32.h>
@@ -65,6 +66,7 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 			   unsigned long *end)
 {
 	if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) {
+		unsigned long new_begin;
 		/* This is usually used needed to map code in small
 		   model, so it needs to be in the first 31bit. Limit
 		   it to that.  This means we need to move the
@@ -74,6 +76,11 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 		   of playground for now. -AK */ 
 		*begin = 0x40000000; 
 		*end = 0x80000000;		
+		if (current->flags & PF_RANDOMIZE) {
+			new_begin = randomize_range(*begin, *begin + 0x02000000, 0);
+			if (new_begin)
+				*begin = new_begin;
+		}
 	} else {
 		*begin = TASK_UNMAPPED_BASE;
 		*end = TASK_SIZE; 
@@ -143,6 +150,97 @@ full_search:
 	}
 }
 
+
+unsigned long
+arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+			  const unsigned long len, const unsigned long pgoff,
+			  const unsigned long flags)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	unsigned long addr = addr0;
+
+	/* requested length too big for entire address space */
+	if (len > TASK_SIZE)
+		return -ENOMEM;
+
+	if (flags & MAP_FIXED)
+		return addr;
+
+	/* for MAP_32BIT mappings we force the legact mmap base */
+	if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT))
+		goto bottomup;
+
+	/* requesting a specific address */
+	if (addr) {
+		addr = PAGE_ALIGN(addr);
+		vma = find_vma(mm, addr);
+		if (TASK_SIZE - len >= addr &&
+				(!vma || addr + len <= vma->vm_start))
+			return addr;
+	}
+
+	/* check if free_area_cache is useful for us */
+	if (len <= mm->cached_hole_size) {
+ 	        mm->cached_hole_size = 0;
+ 		mm->free_area_cache = mm->mmap_base;
+ 	}
+
+	/* either no address requested or can't fit in requested address hole */
+	addr = mm->free_area_cache;
+
+	/* make sure it can fit in the remaining address space */
+	if (addr > len) {
+		vma = find_vma(mm, addr-len);
+		if (!vma || addr <= vma->vm_start)
+			/* remember the address as a hint for next time */
+			return (mm->free_area_cache = addr-len);
+	}
+
+	if (mm->mmap_base < len)
+		goto bottomup;
+
+	addr = mm->mmap_base-len;
+
+	do {
+		/*
+		 * Lookup failure means no vma is above this address,
+		 * else if new region fits below vma->vm_start,
+		 * return with success:
+		 */
+		vma = find_vma(mm, addr);
+		if (!vma || addr+len <= vma->vm_start)
+			/* remember the address as a hint for next time */
+			return (mm->free_area_cache = addr);
+
+ 		/* remember the largest hole we saw so far */
+ 		if (addr + mm->cached_hole_size < vma->vm_start)
+ 		        mm->cached_hole_size = vma->vm_start - addr;
+
+		/* try just below the current vma->vm_start */
+		addr = vma->vm_start-len;
+	} while (len < vma->vm_start);
+
+bottomup:
+	/*
+	 * A failed mmap() very likely causes application failure,
+	 * so fall back to the bottom-up function here. This scenario
+	 * can happen with large stack limits and large mmap()
+	 * allocations.
+	 */
+	mm->cached_hole_size = ~0UL;
+  	mm->free_area_cache = TASK_UNMAPPED_BASE;
+	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
+	/*
+	 * Restore the topdown base:
+	 */
+	mm->free_area_cache = mm->mmap_base;
+	mm->cached_hole_size = ~0UL;
+
+	return addr;
+}
+
+
 asmlinkage long sys_uname(struct new_utsname __user * name)
 {
 	int err;
diff --git a/arch/x86/mm/mmap_64.c b/arch/x86/mm/mmap_64.c
index ffb71a31bb6e..8cf03ea651f8 100644
--- a/arch/x86/mm/mmap_64.c
+++ b/arch/x86/mm/mmap_64.c
@@ -1,32 +1,117 @@
-/* Copyright 2005 Andi Kleen, SuSE Labs.
- * Licensed under GPL, v.2
+/*
+ *  linux/arch/x86-64/mm/mmap.c
+ *
+ *  flexible mmap layout support
+ *
+ * Based on code by Ingo Molnar and Andi Kleen, copyrighted
+ * as follows:
+ *
+ * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina.
+ * All Rights Reserved.
+ * Copyright 2005 Andi Kleen, SUSE Labs.
+ * Copyright 2007 Jiri Kosina, SUSE Labs.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
  */
+
+#include <linux/personality.h>
 #include <linux/mm.h>
-#include <linux/sched.h>
 #include <linux/random.h>
+#include <linux/limits.h>
+#include <linux/sched.h>
 #include <asm/ia32.h>
 
-/* Notebook: move the mmap code from sys_x86_64.c over here. */
+/*
+ * Top of mmap area (just below the process stack).
+ *
+ * Leave an at least ~128 MB hole.
+ */
+#define MIN_GAP (128*1024*1024)
+#define MAX_GAP (TASK_SIZE/6*5)
 
-void arch_pick_mmap_layout(struct mm_struct *mm)
+static inline unsigned long mmap_base(void)
+{
+	unsigned long gap = current->signal->rlim[RLIMIT_STACK].rlim_cur;
+
+	if (gap < MIN_GAP)
+		gap = MIN_GAP;
+	else if (gap > MAX_GAP)
+		gap = MAX_GAP;
+
+	return TASK_SIZE - (gap & PAGE_MASK);
+}
+
+static inline int mmap_is_32(void)
 {
 #ifdef CONFIG_IA32_EMULATION
-	if (current_thread_info()->flags & _TIF_IA32)
-		return ia32_pick_mmap_layout(mm);
+	if (test_thread_flag(TIF_IA32))
+		return 1;
 #endif
-	mm->mmap_base = TASK_UNMAPPED_BASE;
+	return 0;
+}
+
+static inline int mmap_is_legacy(void)
+{
+	if (current->personality & ADDR_COMPAT_LAYOUT)
+		return 1;
+
+	if (current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY)
+		return 1;
+
+	return sysctl_legacy_va_layout;
+}
+
+/*
+ * This function, called very early during the creation of a new
+ * process VM image, sets up which VM layout function to use:
+ */
+void arch_pick_mmap_layout(struct mm_struct *mm)
+{
+	int rnd = 0;
 	if (current->flags & PF_RANDOMIZE) {
 		/*
-		 * Add 28bit randomness which is about 40bits of
-		 * address space because mmap base has to be page
-		 * aligned.  or ~1/128 of the total user VM (total
-		 * user address space is 47bits)
+		 * Add 28bit randomness which is about 40bits of address space
+		 * because mmap base has to be page aligned.
+		 * or ~1/128 of the total user VM
+		 * (total user address space is 47bits)
 		 */
-		unsigned rnd = get_random_int() & 0xfffffff;
+		rnd = get_random_int() & 0xfffffff;
+	}
 
-		mm->mmap_base += ((unsigned long)rnd) << PAGE_SHIFT;
+	/*
+	 * Fall back to the standard layout if the personality
+	 * bit is set, or if the expected stack growth is unlimited:
+	 */
+	if (mmap_is_32()) {
+#ifdef CONFIG_IA32_EMULATION
+		/* ia32_pick_mmap_layout has its own. */
+		return ia32_pick_mmap_layout(mm);
+#endif
+	} else if(mmap_is_legacy()) {
+		mm->mmap_base = TASK_UNMAPPED_BASE;
+		mm->get_unmapped_area = arch_get_unmapped_area;
+		mm->unmap_area = arch_unmap_area;
+	} else {
+		mm->mmap_base = mmap_base();
+		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
+		mm->unmap_area = arch_unmap_area_topdown;
+		if (current->flags & PF_RANDOMIZE)
+			rnd = -rnd;
+	}
+	if (current->flags & PF_RANDOMIZE) {
+		mm->mmap_base += ((long)rnd) << PAGE_SHIFT;
 	}
-	mm->get_unmapped_area = arch_get_unmapped_area;
-	mm->unmap_area = arch_unmap_area;
 }
-
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 043a800c8f71..8193d24be159 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -45,7 +45,7 @@
 
 static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs);
 static int load_elf_library(struct file *);
-static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int);
+static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int, unsigned long);
 
 /*
  * If we don't support core dumping, then supply a NULL so we
@@ -298,33 +298,70 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 #ifndef elf_map
 
 static unsigned long elf_map(struct file *filep, unsigned long addr,
-		struct elf_phdr *eppnt, int prot, int type)
+		struct elf_phdr *eppnt, int prot, int type,
+		unsigned long total_size)
 {
 	unsigned long map_addr;
-	unsigned long pageoffset = ELF_PAGEOFFSET(eppnt->p_vaddr);
+	unsigned long size = eppnt->p_filesz + ELF_PAGEOFFSET(eppnt->p_vaddr);
+	unsigned long off = eppnt->p_offset - ELF_PAGEOFFSET(eppnt->p_vaddr);
+	addr = ELF_PAGESTART(addr);
+	size = ELF_PAGEALIGN(size);
 
-	down_write(&current->mm->mmap_sem);
 	/* mmap() will return -EINVAL if given a zero size, but a
 	 * segment with zero filesize is perfectly valid */
-	if (eppnt->p_filesz + pageoffset)
-		map_addr = do_mmap(filep, ELF_PAGESTART(addr),
-				   eppnt->p_filesz + pageoffset, prot, type,
-				   eppnt->p_offset - pageoffset);
-	else
-		map_addr = ELF_PAGESTART(addr);
+	if (!size)
+		return addr;
+
+	down_write(&current->mm->mmap_sem);
+	/*
+	* total_size is the size of the ELF (interpreter) image.
+	* The _first_ mmap needs to know the full size, otherwise
+	* randomization might put this image into an overlapping
+	* position with the ELF binary image. (since size < total_size)
+	* So we first map the 'big' image - and unmap the remainder at
+	* the end. (which unmap is needed for ELF images with holes.)
+	*/
+	if (total_size) {
+		total_size = ELF_PAGEALIGN(total_size);
+		map_addr = do_mmap(filep, addr, total_size, prot, type, off);
+		if (!BAD_ADDR(map_addr))
+			do_munmap(current->mm, map_addr+size, total_size-size);
+	} else
+		map_addr = do_mmap(filep, addr, size, prot, type, off);
+
 	up_write(&current->mm->mmap_sem);
 	return(map_addr);
 }
 
 #endif /* !elf_map */
 
+static unsigned long total_mapping_size(struct elf_phdr *cmds, int nr)
+{
+	int i, first_idx = -1, last_idx = -1;
+
+	for (i = 0; i < nr; i++) {
+		if (cmds[i].p_type == PT_LOAD) {
+			last_idx = i;
+			if (first_idx == -1)
+				first_idx = i;
+		}
+	}
+	if (first_idx == -1)
+		return 0;
+
+	return cmds[last_idx].p_vaddr + cmds[last_idx].p_memsz -
+				ELF_PAGESTART(cmds[first_idx].p_vaddr);
+}
+
+
 /* This is much more generalized than the library routine read function,
    so we keep this separate.  Technically the library read function
    is only provided so that we can read a.out libraries that have
    an ELF header */
 
 static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
-		struct file *interpreter, unsigned long *interp_load_addr)
+		struct file *interpreter, unsigned long *interp_map_addr,
+		unsigned long no_base)
 {
 	struct elf_phdr *elf_phdata;
 	struct elf_phdr *eppnt;
@@ -332,6 +369,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 	int load_addr_set = 0;
 	unsigned long last_bss = 0, elf_bss = 0;
 	unsigned long error = ~0UL;
+	unsigned long total_size;
 	int retval, i, size;
 
 	/* First of all, some simple consistency checks */
@@ -370,6 +408,12 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 		goto out_close;
 	}
 
+	total_size = total_mapping_size(elf_phdata, interp_elf_ex->e_phnum);
+	if (!total_size) {
+		error = -EINVAL;
+		goto out_close;
+	}
+
 	eppnt = elf_phdata;
 	for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) {
 		if (eppnt->p_type == PT_LOAD) {
@@ -387,9 +431,14 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 			vaddr = eppnt->p_vaddr;
 			if (interp_elf_ex->e_type == ET_EXEC || load_addr_set)
 				elf_type |= MAP_FIXED;
+			else if (no_base && interp_elf_ex->e_type == ET_DYN)
+				load_addr = -vaddr;
 
 			map_addr = elf_map(interpreter, load_addr + vaddr,
-					   eppnt, elf_prot, elf_type);
+					   eppnt, elf_prot, elf_type, total_size);
+			total_size = 0;
+			if (!*interp_map_addr)
+				*interp_map_addr = map_addr;
 			error = map_addr;
 			if (BAD_ADDR(map_addr))
 				goto out_close;
@@ -455,8 +504,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 			goto out_close;
 	}
 
-	*interp_load_addr = load_addr;
-	error = ((unsigned long)interp_elf_ex->e_entry) + load_addr;
+	error = load_addr;
 
 out_close:
 	kfree(elf_phdata);
@@ -553,7 +601,8 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	int elf_exec_fileno;
 	int retval, i;
 	unsigned int size;
-	unsigned long elf_entry, interp_load_addr = 0;
+	unsigned long elf_entry;
+	unsigned long interp_load_addr = 0;
 	unsigned long start_code, end_code, start_data, end_data;
 	unsigned long reloc_func_desc = 0;
 	char passed_fileno[6];
@@ -825,9 +874,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	current->mm->start_stack = bprm->p;
 
 	/* Now we do a little grungy work by mmaping the ELF image into
-	   the correct location in memory.  At this point, we assume that
-	   the image should be loaded at fixed address, not at a variable
-	   address. */
+	   the correct location in memory. */
 	for(i = 0, elf_ppnt = elf_phdata;
 	    i < loc->elf_ex.e_phnum; i++, elf_ppnt++) {
 		int elf_prot = 0, elf_flags;
@@ -881,11 +928,15 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 			 * default mmap base, as well as whatever program they
 			 * might try to exec.  This is because the brk will
 			 * follow the loader, and is not movable.  */
+#ifdef CONFIG_X86
+			load_bias = 0;
+#else
 			load_bias = ELF_PAGESTART(ELF_ET_DYN_BASE - vaddr);
+#endif
 		}
 
 		error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt,
-				elf_prot, elf_flags);
+				elf_prot, elf_flags,0);
 		if (BAD_ADDR(error)) {
 			send_sig(SIGKILL, current, 0);
 			retval = IS_ERR((void *)error) ?
@@ -961,13 +1012,25 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	}
 
 	if (elf_interpreter) {
-		if (interpreter_type == INTERPRETER_AOUT)
+		if (interpreter_type == INTERPRETER_AOUT) {
 			elf_entry = load_aout_interp(&loc->interp_ex,
 						     interpreter);
-		else
+		} else {
+			unsigned long uninitialized_var(interp_map_addr);
+
 			elf_entry = load_elf_interp(&loc->interp_elf_ex,
 						    interpreter,
-						    &interp_load_addr);
+						    &interp_map_addr,
+						    load_bias);
+			if (!IS_ERR((void *)elf_entry)) {
+				/*
+				 * load_elf_interp() returns relocation
+				 * adjustment
+				 */
+				interp_load_addr = elf_entry;
+				elf_entry += loc->interp_elf_ex.e_entry;
+			}
+		}
 		if (BAD_ADDR(elf_entry)) {
 			force_sig(SIGSEGV, current);
 			retval = IS_ERR((void *)elf_entry) ?
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 9d4f11dd566f..6cf40dec0932 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -413,6 +413,7 @@ pte_t *lookup_address(unsigned long addr);
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
 #define HAVE_ARCH_UNMAPPED_AREA
+#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
 
 #define pgtable_cache_init()   do { } while (0)
 #define check_pgt_cache()      do { } while (0)
-- 
cgit v1.2.3


From 929fd589135ed417663f1b75b8031c9cf6687700 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: some whitespace cleanups in paging code

This patch does some whitespace cleanups in the paging code to fix some
checkpatch.pl warnings of my formerly merged cleanup patches.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_64.c        |  2 +-
 include/asm-x86/pgtable_64.h | 22 +++++++++++-----------
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 3f8bf298dbb8..5cadbb40da3e 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -226,7 +226,7 @@ __meminit void *early_ioremap(unsigned long addr, unsigned long size)
 		vaddr += addr & ~PMD_MASK;
 		addr &= PMD_MASK;
 		for (i = 0; i < pmds; i++, addr += PMD_SIZE)
-			set_pmd(pmd + i,__pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
+			set_pmd(pmd+i, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
 		__flush_tlb();
 		return (void *)vaddr;
 	next:
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 6cf40dec0932..3072619365a8 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -154,22 +154,22 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
-#define _PAGE_PRESENT	(_AC(1,UL)<<_PAGE_BIT_PRESENT)
-#define _PAGE_RW	(_AC(1,UL)<<_PAGE_BIT_RW)
-#define _PAGE_USER	(_AC(1,UL)<<_PAGE_BIT_USER)
-#define _PAGE_PWT	(_AC(1,UL)<<_PAGE_BIT_PWT)
-#define _PAGE_PCD	(_AC(1,UL)<<_PAGE_BIT_PCD)
-#define _PAGE_ACCESSED	(_AC(1,UL)<<_PAGE_BIT_ACCESSED)
-#define _PAGE_DIRTY	(_AC(1,UL)<<_PAGE_BIT_DIRTY)
+#define _PAGE_PRESENT	(_AC(1, UL)<<_PAGE_BIT_PRESENT)
+#define _PAGE_RW	(_AC(1, UL)<<_PAGE_BIT_RW)
+#define _PAGE_USER	(_AC(1, UL)<<_PAGE_BIT_USER)
+#define _PAGE_PWT	(_AC(1, UL)<<_PAGE_BIT_PWT)
+#define _PAGE_PCD	(_AC(1, UL)<<_PAGE_BIT_PCD)
+#define _PAGE_ACCESSED	(_AC(1, UL)<<_PAGE_BIT_ACCESSED)
+#define _PAGE_DIRTY	(_AC(1, UL)<<_PAGE_BIT_DIRTY)
 /* 2MB page */
-#define _PAGE_PSE	(_AC(1,UL)<<_PAGE_BIT_PSE)
+#define _PAGE_PSE	(_AC(1, UL)<<_PAGE_BIT_PSE)
 /* nonlinear file mapping, saved PTE; unset:swap */
-#define _PAGE_FILE	(_AC(1,UL)<<_PAGE_BIT_FILE)
+#define _PAGE_FILE	(_AC(1, UL)<<_PAGE_BIT_FILE)
 /* Global TLB entry */
-#define _PAGE_GLOBAL	(_AC(1,UL)<<_PAGE_BIT_GLOBAL)
+#define _PAGE_GLOBAL	(_AC(1, UL)<<_PAGE_BIT_GLOBAL)
 
 #define _PAGE_PROTNONE	0x080	/* If not present */
-#define _PAGE_NX        (_AC(1,UL)<<_PAGE_BIT_NX)
+#define _PAGE_NX        (_AC(1, UL)<<_PAGE_BIT_NX)
 
 #define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-- 
cgit v1.2.3


From e34907ae180f4fe6c28bb4516c679c2f81b0c9ed Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: remove volatile keyword from clflush.

the p parameter is an explicit memory reference, and is
enough to prevent gcc to being nasty here. The volatile
seems completely not needed.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index f5b3f77f5310..28978b17b07a 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -161,7 +161,7 @@ static inline void native_wbinvd(void)
 	asm volatile("wbinvd": : :"memory");
 }
 
-static inline void clflush(volatile void *__p)
+static inline void clflush(void *__p)
 {
 	asm volatile("clflush %0" : "+m" (*(char __force *)__p));
 }
-- 
cgit v1.2.3


From d89542229b657bdcce6a6f76168f9098ee3e9344 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: put together equal pieces of system.h

This patch puts together pieces of system_{32,64}.h that
looks like the same. It's the first step towards integration
of this file.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_64.c |  2 +-
 include/asm-x86/system.h     | 69 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/system_32.h  | 58 -------------------------------------
 include/asm-x86/system_64.h  | 12 --------
 4 files changed, 70 insertions(+), 71 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 4c4d8b3f046e..057b5442ffda 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -99,7 +99,7 @@ void exit_idle(void)
  * We use this if we don't have any better
  * idle routine..
  */
-static void default_idle(void)
+void default_idle(void)
 {
 	current_thread_info()->status &= ~TS_POLLING;
 	/*
diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 692562b48f2a..d0803f8c70c4 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -1,5 +1,74 @@
+#ifndef _ASM_X86_SYSTEM_H_
+#define _ASM_X86_SYSTEM_H_
+
+#include <asm/asm.h>
+
 #ifdef CONFIG_X86_32
 # include "system_32.h"
 #else
 # include "system_64.h"
 #endif
+
+#ifdef __KERNEL__
+#define _set_base(addr, base) do { unsigned long __pr; \
+__asm__ __volatile__ ("movw %%dx,%1\n\t" \
+	"rorl $16,%%edx\n\t" \
+	"movb %%dl,%2\n\t" \
+	"movb %%dh,%3" \
+	:"=&d" (__pr) \
+	:"m" (*((addr)+2)), \
+	 "m" (*((addr)+4)), \
+	 "m" (*((addr)+7)), \
+	 "0" (base) \
+	); } while (0)
+
+#define _set_limit(addr, limit) do { unsigned long __lr; \
+__asm__ __volatile__ ("movw %%dx,%1\n\t" \
+	"rorl $16,%%edx\n\t" \
+	"movb %2,%%dh\n\t" \
+	"andb $0xf0,%%dh\n\t" \
+	"orb %%dh,%%dl\n\t" \
+	"movb %%dl,%2" \
+	:"=&d" (__lr) \
+	:"m" (*(addr)), \
+	 "m" (*((addr)+6)), \
+	 "0" (limit) \
+	); } while (0)
+
+#define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base))
+#define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1))
+
+/*
+ * Save a segment register away
+ */
+#define savesegment(seg, value) \
+	asm volatile("mov %%" #seg ",%0":"=rm" (value))
+
+static inline unsigned long get_limit(unsigned long segment)
+{
+	unsigned long __limit;
+	__asm__("lsll %1,%0"
+		:"=r" (__limit):"r" (segment));
+	return __limit+1;
+}
+#endif /* __KERNEL__ */
+
+static inline void clflush(void *__p)
+{
+	asm volatile("clflush %0" : "+m" (*(char __force *)__p));
+}
+
+#define nop() __asm__ __volatile__ ("nop")
+
+void disable_hlt(void);
+void enable_hlt(void);
+
+extern int es7000_plat;
+void cpu_idle_wait(void);
+
+extern unsigned long arch_align_stack(unsigned long sp);
+extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
+
+void default_idle(void);
+
+#endif
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index 28978b17b07a..fb457642ac58 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -34,34 +34,6 @@ extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struc
 		      "2" (prev), "d" (next));				\
 } while (0)
 
-#define _set_base(addr,base) do { unsigned long __pr; \
-__asm__ __volatile__ ("movw %%dx,%1\n\t" \
-	"rorl $16,%%edx\n\t" \
-	"movb %%dl,%2\n\t" \
-	"movb %%dh,%3" \
-	:"=&d" (__pr) \
-	:"m" (*((addr)+2)), \
-	 "m" (*((addr)+4)), \
-	 "m" (*((addr)+7)), \
-         "0" (base) \
-        ); } while(0)
-
-#define _set_limit(addr,limit) do { unsigned long __lr; \
-__asm__ __volatile__ ("movw %%dx,%1\n\t" \
-	"rorl $16,%%edx\n\t" \
-	"movb %2,%%dh\n\t" \
-	"andb $0xf0,%%dh\n\t" \
-	"orb %%dh,%%dl\n\t" \
-	"movb %%dl,%2" \
-	:"=&d" (__lr) \
-	:"m" (*(addr)), \
-	 "m" (*((addr)+6)), \
-	 "0" (limit) \
-        ); } while(0)
-
-#define set_base(ldt,base) _set_base( ((char *)&(ldt)) , (base) )
-#define set_limit(ldt,limit) _set_limit( ((char *)&(ldt)) , ((limit)-1) )
-
 /*
  * Load a segment. Fall back on loading the zero
  * segment if something goes wrong..
@@ -83,12 +55,6 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t" \
 		".previous"			\
 		: :"rm" (value))
 
-/*
- * Save a segment register away
- */
-#define savesegment(seg, value) \
-	asm volatile("mov %%" #seg ",%0":"=rm" (value))
-
 
 static inline void native_clts(void)
 {
@@ -161,11 +127,6 @@ static inline void native_wbinvd(void)
 	asm volatile("wbinvd": : :"memory");
 }
 
-static inline void clflush(void *__p)
-{
-	asm volatile("clflush %0" : "+m" (*(char __force *)__p));
-}
-
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
@@ -190,15 +151,6 @@ static inline void clflush(void *__p)
 
 #endif	/* __KERNEL__ */
 
-static inline unsigned long get_limit(unsigned long segment)
-{
-	unsigned long __limit;
-	__asm__("lsll %1,%0"
-		:"=r" (__limit):"r" (segment));
-	return __limit+1;
-}
-
-#define nop() __asm__ __volatile__ ("nop")
 
 /*
  * Force strict CPU ordering.
@@ -305,15 +257,5 @@ static inline unsigned long get_limit(unsigned long segment)
  * disable hlt during certain critical i/o operations
  */
 #define HAVE_DISABLE_HLT
-void disable_hlt(void);
-void enable_hlt(void);
-
-extern int es7000_plat;
-void cpu_idle_wait(void);
-
-extern unsigned long arch_align_stack(unsigned long sp);
-extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
-
-void default_idle(void);
 
 #endif
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 3dcb217a7202..cc5b2666a044 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -141,13 +141,6 @@ static inline void write_cr8(unsigned long val)
 
 #endif	/* __KERNEL__ */
 
-static inline void clflush(volatile void *__p)
-{
-	asm volatile("clflush %0" : "+m" (*(char __force *)__p));
-}
-
-#define nop() __asm__ __volatile__ ("nop")
-
 #ifdef CONFIG_SMP
 #define smp_mb()	mb()
 #define smp_rmb()	barrier()
@@ -177,9 +170,4 @@ static inline void clflush(volatile void *__p)
 
 #include <linux/irqflags.h>
 
-void cpu_idle_wait(void);
-
-extern unsigned long arch_align_stack(unsigned long sp);
-extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
-
 #endif
-- 
cgit v1.2.3


From a6b4655258efd39b590e519815ed43bb74cd7188 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: unify load_segment macro

This patch unifies the load_segment() macro, making them equal in both
x86_64 and i386 architectures. The common version goes to system.h,
and the old are deleted.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h    | 21 +++++++++++++++++++++
 include/asm-x86/system_32.h | 22 ----------------------
 include/asm-x86/system_64.h | 20 --------------------
 3 files changed, 21 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index d0803f8c70c4..3740bada097c 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -38,6 +38,27 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t" \
 #define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base))
 #define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1))
 
+/*
+ * Load a segment. Fall back on loading the zero
+ * segment if something goes wrong..
+ */
+#define loadsegment(seg, value)			\
+	asm volatile("\n"			\
+		"1:\t"				\
+		"movl %k0,%%" #seg "\n"		\
+		"2:\n"				\
+		".section .fixup,\"ax\"\n"	\
+		"3:\t"				\
+		"movl %k1, %%" #seg "\n\t"	\
+		"jmp 2b\n"			\
+		".previous\n"			\
+		".section __ex_table,\"a\"\n\t"	\
+		_ASM_ALIGN "\n\t"		\
+		_ASM_PTR " 1b,3b\n"		\
+		".previous"			\
+		: :"r" (value), "r" (0))
+
+
 /*
  * Save a segment register away
  */
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index fb457642ac58..8db478984ed1 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -34,28 +34,6 @@ extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struc
 		      "2" (prev), "d" (next));				\
 } while (0)
 
-/*
- * Load a segment. Fall back on loading the zero
- * segment if something goes wrong..
- */
-#define loadsegment(seg,value)			\
-	asm volatile("\n"			\
-		"1:\t"				\
-		"mov %0,%%" #seg "\n"		\
-		"2:\n"				\
-		".section .fixup,\"ax\"\n"	\
-		"3:\t"				\
-		"pushl $0\n\t"			\
-		"popl %%" #seg "\n\t"		\
-		"jmp 2b\n"			\
-		".previous\n"			\
-		".section __ex_table,\"a\"\n\t"	\
-		".align 4\n\t"			\
-		".long 1b,3b\n"			\
-		".previous"			\
-		: :"rm" (value))
-
-
 static inline void native_clts(void)
 {
 	asm volatile ("clts");
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index cc5b2666a044..0885caace5d4 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -49,26 +49,6 @@
     
 extern void load_gs_index(unsigned); 
 
-/*
- * Load a segment. Fall back on loading the zero
- * segment if something goes wrong..
- */
-#define loadsegment(seg,value)	\
-	asm volatile("\n"			\
-		"1:\t"				\
-		"movl %k0,%%" #seg "\n"		\
-		"2:\n"				\
-		".section .fixup,\"ax\"\n"	\
-		"3:\t"				\
-		"movl %1,%%" #seg "\n\t" 	\
-		"jmp 2b\n"			\
-		".previous\n"			\
-		".section __ex_table,\"a\"\n\t"	\
-		".align 8\n\t"			\
-		".quad 1b,3b\n"			\
-		".previous"			\
-		: :"r" (value), "r" (0))
-
 /*
  * Clear and set 'TS' bit respectively
  */
-- 
cgit v1.2.3


From d3ca901f94b3299dfe3a814770d751844608419f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: unify paravirt parts of system.h

This patch moves the i386 control registers manipulation functions,
wbinvd, and clts functions to system.h. They are essentially the same
as in x86_64.

With this, system.h paravirt comes for free in x86_64.

[ mingo@elte.hu: reintroduced the cr8 bits - needed for resume images ]

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h    | 110 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/system_32.h |  94 -------------------------------------
 include/asm-x86/system_64.h |  85 +++++-----------------------------
 3 files changed, 122 insertions(+), 167 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 3740bada097c..01ba1f8e64d1 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -3,6 +3,8 @@
 
 #include <asm/asm.h>
 
+#include <linux/kernel.h>
+
 #ifdef CONFIG_X86_32
 # include "system_32.h"
 #else
@@ -38,6 +40,8 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t" \
 #define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base))
 #define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1))
 
+extern void load_gs_index(unsigned);
+
 /*
  * Load a segment. Fall back on loading the zero
  * segment if something goes wrong..
@@ -72,6 +76,112 @@ static inline unsigned long get_limit(unsigned long segment)
 		:"=r" (__limit):"r" (segment));
 	return __limit+1;
 }
+
+static inline void native_clts(void)
+{
+	asm volatile ("clts");
+}
+
+/*
+ * Volatile isn't enough to prevent the compiler from reordering the
+ * read/write functions for the control registers and messing everything up.
+ * A memory clobber would solve the problem, but would prevent reordering of
+ * all loads stores around it, which can hurt performance. Solution is to
+ * use a variable and mimic reads and writes to it to enforce serialization
+ */
+static unsigned long __force_order;
+
+static inline unsigned long native_read_cr0(void)
+{
+	unsigned long val;
+	asm volatile("mov %%cr0,%0\n\t" :"=r" (val), "=m" (__force_order));
+	return val;
+}
+
+static inline void native_write_cr0(unsigned long val)
+{
+	asm volatile("mov %0,%%cr0": :"r" (val), "m" (__force_order));
+}
+
+static inline unsigned long native_read_cr2(void)
+{
+	unsigned long val;
+	asm volatile("mov %%cr2,%0\n\t" :"=r" (val), "=m" (__force_order));
+	return val;
+}
+
+static inline void native_write_cr2(unsigned long val)
+{
+	asm volatile("mov %0,%%cr2": :"r" (val), "m" (__force_order));
+}
+
+static inline unsigned long native_read_cr3(void)
+{
+	unsigned long val;
+	asm volatile("mov %%cr3,%0\n\t" :"=r" (val), "=m" (__force_order));
+	return val;
+}
+
+static inline void native_write_cr3(unsigned long val)
+{
+	asm volatile("mov %0,%%cr3": :"r" (val), "m" (__force_order));
+}
+
+static inline unsigned long native_read_cr4(void)
+{
+	unsigned long val;
+	asm volatile("mov %%cr4,%0\n\t" :"=r" (val), "=m" (__force_order));
+	return val;
+}
+
+static inline unsigned long native_read_cr4_safe(void)
+{
+	unsigned long val;
+	/* This could fault if %cr4 does not exist. In x86_64, a cr4 always
+	 * exists, so it will never fail. */
+#ifdef CONFIG_X86_32
+	asm volatile("1: mov %%cr4, %0		\n"
+		"2:				\n"
+		".section __ex_table,\"a\"	\n"
+		".long 1b,2b			\n"
+		".previous			\n"
+		: "=r" (val), "=m" (__force_order) : "0" (0));
+#else
+	val = native_read_cr4();
+#endif
+	return val;
+}
+
+static inline void native_write_cr4(unsigned long val)
+{
+	asm volatile("mov %0,%%cr4": :"r" (val), "m" (__force_order));
+}
+
+static inline void native_wbinvd(void)
+{
+	asm volatile("wbinvd": : :"memory");
+}
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define read_cr0()	(native_read_cr0())
+#define write_cr0(x)	(native_write_cr0(x))
+#define read_cr2()	(native_read_cr2())
+#define write_cr2(x)	(native_write_cr2(x))
+#define read_cr3()	(native_read_cr3())
+#define write_cr3(x)	(native_write_cr3(x))
+#define read_cr4()	(native_read_cr4())
+#define read_cr4_safe()	(native_read_cr4_safe())
+#define write_cr4(x)	(native_write_cr4(x))
+#define wbinvd()	(native_wbinvd())
+
+/* Clear the 'TS' bit */
+#define clts()		(native_clts())
+
+#endif/* CONFIG_PARAVIRT */
+
+#define stts() write_cr0(8 | read_cr0())
+
 #endif /* __KERNEL__ */
 
 static inline void clflush(void *__p)
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index 8db478984ed1..c05568290add 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_SYSTEM_H
 #define __ASM_SYSTEM_H
 
-#include <linux/kernel.h>
 #include <asm/segment.h>
 #include <asm/cpufeature.h>
 #include <asm/cmpxchg.h>
@@ -34,99 +33,6 @@ extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struc
 		      "2" (prev), "d" (next));				\
 } while (0)
 
-static inline void native_clts(void)
-{
-	asm volatile ("clts");
-}
-
-static inline unsigned long native_read_cr0(void)
-{
-	unsigned long val;
-	asm volatile("movl %%cr0,%0\n\t" :"=r" (val));
-	return val;
-}
-
-static inline void native_write_cr0(unsigned long val)
-{
-	asm volatile("movl %0,%%cr0": :"r" (val));
-}
-
-static inline unsigned long native_read_cr2(void)
-{
-	unsigned long val;
-	asm volatile("movl %%cr2,%0\n\t" :"=r" (val));
-	return val;
-}
-
-static inline void native_write_cr2(unsigned long val)
-{
-	asm volatile("movl %0,%%cr2": :"r" (val));
-}
-
-static inline unsigned long native_read_cr3(void)
-{
-	unsigned long val;
-	asm volatile("movl %%cr3,%0\n\t" :"=r" (val));
-	return val;
-}
-
-static inline void native_write_cr3(unsigned long val)
-{
-	asm volatile("movl %0,%%cr3": :"r" (val));
-}
-
-static inline unsigned long native_read_cr4(void)
-{
-	unsigned long val;
-	asm volatile("movl %%cr4,%0\n\t" :"=r" (val));
-	return val;
-}
-
-static inline unsigned long native_read_cr4_safe(void)
-{
-	unsigned long val;
-	/* This could fault if %cr4 does not exist */
-	asm volatile("1: movl %%cr4, %0		\n"
-		"2:				\n"
-		".section __ex_table,\"a\"	\n"
-		".long 1b,2b			\n"
-		".previous			\n"
-		: "=r" (val): "0" (0));
-	return val;
-}
-
-static inline void native_write_cr4(unsigned long val)
-{
-	asm volatile("movl %0,%%cr4": :"r" (val));
-}
-
-static inline void native_wbinvd(void)
-{
-	asm volatile("wbinvd": : :"memory");
-}
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define read_cr0()	(native_read_cr0())
-#define write_cr0(x)	(native_write_cr0(x))
-#define read_cr2()	(native_read_cr2())
-#define write_cr2(x)	(native_write_cr2(x))
-#define read_cr3()	(native_read_cr3())
-#define write_cr3(x)	(native_write_cr3(x))
-#define read_cr4()	(native_read_cr4())
-#define read_cr4_safe()	(native_read_cr4_safe())
-#define write_cr4(x)	(native_write_cr4(x))
-#define wbinvd()	(native_wbinvd())
-
-/* Clear the 'TS' bit */
-#define clts()		(native_clts())
-
-#endif/* CONFIG_PARAVIRT */
-
-/* Set the 'TS' bit */
-#define stts() write_cr0(8 | read_cr0())
-
 #endif	/* __KERNEL__ */
 
 
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 0885caace5d4..14ad6138439b 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_SYSTEM_H
 #define __ASM_SYSTEM_H
 
-#include <linux/kernel.h>
 #include <asm/segment.h>
 #include <asm/cmpxchg.h>
 
@@ -47,78 +46,6 @@
 		       [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent))   \
 		     : "memory", "cc" __EXTRA_CLOBBER)
     
-extern void load_gs_index(unsigned); 
-
-/*
- * Clear and set 'TS' bit respectively
- */
-#define clts() __asm__ __volatile__ ("clts")
-
-static inline unsigned long read_cr0(void)
-{ 
-	unsigned long cr0;
-	asm volatile("movq %%cr0,%0" : "=r" (cr0));
-	return cr0;
-}
-
-static inline void write_cr0(unsigned long val) 
-{ 
-	asm volatile("movq %0,%%cr0" :: "r" (val));
-}
-
-static inline unsigned long read_cr2(void)
-{
-	unsigned long cr2;
-	asm volatile("movq %%cr2,%0" : "=r" (cr2));
-	return cr2;
-}
-
-static inline void write_cr2(unsigned long val)
-{
-	asm volatile("movq %0,%%cr2" :: "r" (val));
-}
-
-static inline unsigned long read_cr3(void)
-{ 
-	unsigned long cr3;
-	asm volatile("movq %%cr3,%0" : "=r" (cr3));
-	return cr3;
-}
-
-static inline void write_cr3(unsigned long val)
-{
-	asm volatile("movq %0,%%cr3" :: "r" (val) : "memory");
-}
-
-static inline unsigned long read_cr4(void)
-{ 
-	unsigned long cr4;
-	asm volatile("movq %%cr4,%0" : "=r" (cr4));
-	return cr4;
-}
-
-static inline void write_cr4(unsigned long val)
-{ 
-	asm volatile("movq %0,%%cr4" :: "r" (val) : "memory");
-}
-
-static inline unsigned long read_cr8(void)
-{
-	unsigned long cr8;
-	asm volatile("movq %%cr8,%0" : "=r" (cr8));
-	return cr8;
-}
-
-static inline void write_cr8(unsigned long val)
-{
-	asm volatile("movq %0,%%cr8" :: "r" (val) : "memory");
-}
-
-#define stts() write_cr0(8 | read_cr0())
-
-#define wbinvd() \
-	__asm__ __volatile__ ("wbinvd": : :"memory")
-
 #endif	/* __KERNEL__ */
 
 #ifdef CONFIG_SMP
@@ -148,6 +75,18 @@ static inline void write_cr8(unsigned long val)
 
 #define warn_if_not_ulong(x) do { unsigned long foo; (void) (&(x) == &foo); } while (0)
 
+static inline unsigned long read_cr8(void)
+{
+	unsigned long cr8;
+	asm volatile("movq %%cr8,%0" : "=r" (cr8));
+	return cr8;
+}
+
+static inline void write_cr8(unsigned long val)
+{
+	asm volatile("movq %0,%%cr8" :: "r" (val) : "memory");
+}
+
 #include <linux/irqflags.h>
 
 #endif
-- 
cgit v1.2.3


From 62fe164c5b036f4bdb19fbfb8f18a75631e67eee Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: remove unused macro

Mr. Grep says warn_if_not_ulong() is not used anymore anywhere
in the code. So, we remove it.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system_64.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 14ad6138439b..560470ea27c8 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -73,8 +73,6 @@
 #define read_barrier_depends()	do {} while(0)
 #define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
 
-#define warn_if_not_ulong(x) do { unsigned long foo; (void) (&(x) == &foo); } while (0)
-
 static inline unsigned long read_cr8(void)
 {
 	unsigned long cr8;
-- 
cgit v1.2.3


From 833d8469b102365f427f7791e79ec1843ff5f164 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: unify smp parts of system.h

The memory barrier parts of system.h are not very different between
i386 and x86_64, the main difference being the availability of
instructions, which we handle with the use of ifdefs.

They are consolidated in system.h file, and then removed from
the arch-specific headers.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h    | 105 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/system_32.h |  99 -----------------------------------------
 include/asm-x86/system_64.h |  25 -----------
 3 files changed, 105 insertions(+), 124 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 01ba1f8e64d1..4c15eb11a917 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -202,4 +202,109 @@ extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
 
 void default_idle(void);
 
+/*
+ * Force strict CPU ordering.
+ * And yes, this is required on UP too when we're talking
+ * to devices.
+ */
+#ifdef CONFIG_X86_32
+/*
+ * For now, "wmb()" doesn't actually do anything, as all
+ * Intel CPU's follow what Intel calls a *Processor Order*,
+ * in which all writes are seen in the program order even
+ * outside the CPU.
+ *
+ * I expect future Intel CPU's to have a weaker ordering,
+ * but I'd also expect them to finally get their act together
+ * and add some real memory barriers if so.
+ *
+ * Some non intel clones support out of order store. wmb() ceases to be a
+ * nop for these.
+ */
+#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
+#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
+#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
+#else
+#define mb() 	asm volatile("mfence":::"memory")
+#define rmb()	asm volatile("lfence":::"memory")
+#define wmb()	asm volatile("sfence" ::: "memory")
+#endif
+
+/**
+ * read_barrier_depends - Flush all pending reads that subsequents reads
+ * depend on.
+ *
+ * No data-dependent reads from memory-like regions are ever reordered
+ * over this barrier.  All reads preceding this primitive are guaranteed
+ * to access memory (but not necessarily other CPUs' caches) before any
+ * reads following this primitive that depend on the data return by
+ * any of the preceding reads.  This primitive is much lighter weight than
+ * rmb() on most CPUs, and is never heavier weight than is
+ * rmb().
+ *
+ * These ordering constraints are respected by both the local CPU
+ * and the compiler.
+ *
+ * Ordering is not guaranteed by anything other than these primitives,
+ * not even by data dependencies.  See the documentation for
+ * memory_barrier() for examples and URLs to more information.
+ *
+ * For example, the following code would force ordering (the initial
+ * value of "a" is zero, "b" is one, and "p" is "&a"):
+ *
+ * <programlisting>
+ *	CPU 0				CPU 1
+ *
+ *	b = 2;
+ *	memory_barrier();
+ *	p = &b;				q = p;
+ *					read_barrier_depends();
+ *					d = *q;
+ * </programlisting>
+ *
+ * because the read of "*q" depends on the read of "p" and these
+ * two reads are separated by a read_barrier_depends().  However,
+ * the following code, with the same initial values for "a" and "b":
+ *
+ * <programlisting>
+ *	CPU 0				CPU 1
+ *
+ *	a = 2;
+ *	memory_barrier();
+ *	b = 3;				y = b;
+ *					read_barrier_depends();
+ *					x = a;
+ * </programlisting>
+ *
+ * does not enforce ordering, since there is no data dependency between
+ * the read of "a" and the read of "b".  Therefore, on some CPUs, such
+ * as Alpha, "y" could be set to 3 and "x" to 0.  Use rmb()
+ * in cases like this where there are no data dependencies.
+ **/
+
+#define read_barrier_depends()	do { } while (0)
+
+#ifdef CONFIG_SMP
+#define smp_mb()	mb()
+#ifdef CONFIG_X86_PPRO_FENCE
+# define smp_rmb()	rmb()
+#else
+# define smp_rmb()	barrier()
+#endif
+#ifdef CONFIG_X86_OOSTORE
+# define smp_wmb() 	wmb()
+#else
+# define smp_wmb()	barrier()
+#endif
+#define smp_read_barrier_depends()	read_barrier_depends()
+#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
+#else
+#define smp_mb()	barrier()
+#define smp_rmb()	barrier()
+#define smp_wmb()	barrier()
+#define smp_read_barrier_depends()	do { } while (0)
+#define set_mb(var, value) do { var = value; barrier(); } while (0)
+#endif
+
+
 #endif
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index c05568290add..7da0716fb317 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -36,105 +36,6 @@ extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struc
 #endif	/* __KERNEL__ */
 
 
-/*
- * Force strict CPU ordering.
- * And yes, this is required on UP too when we're talking
- * to devices.
- *
- * For now, "wmb()" doesn't actually do anything, as all
- * Intel CPU's follow what Intel calls a *Processor Order*,
- * in which all writes are seen in the program order even
- * outside the CPU.
- *
- * I expect future Intel CPU's to have a weaker ordering,
- * but I'd also expect them to finally get their act together
- * and add some real memory barriers if so.
- *
- * Some non intel clones support out of order store. wmb() ceases to be a
- * nop for these.
- */
- 
-
-#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
-#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
-#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
-
-/**
- * read_barrier_depends - Flush all pending reads that subsequents reads
- * depend on.
- *
- * No data-dependent reads from memory-like regions are ever reordered
- * over this barrier.  All reads preceding this primitive are guaranteed
- * to access memory (but not necessarily other CPUs' caches) before any
- * reads following this primitive that depend on the data return by
- * any of the preceding reads.  This primitive is much lighter weight than
- * rmb() on most CPUs, and is never heavier weight than is
- * rmb().
- *
- * These ordering constraints are respected by both the local CPU
- * and the compiler.
- *
- * Ordering is not guaranteed by anything other than these primitives,
- * not even by data dependencies.  See the documentation for
- * memory_barrier() for examples and URLs to more information.
- *
- * For example, the following code would force ordering (the initial
- * value of "a" is zero, "b" is one, and "p" is "&a"):
- *
- * <programlisting>
- *	CPU 0				CPU 1
- *
- *	b = 2;
- *	memory_barrier();
- *	p = &b;				q = p;
- *					read_barrier_depends();
- *					d = *q;
- * </programlisting>
- *
- * because the read of "*q" depends on the read of "p" and these
- * two reads are separated by a read_barrier_depends().  However,
- * the following code, with the same initial values for "a" and "b":
- *
- * <programlisting>
- *	CPU 0				CPU 1
- *
- *	a = 2;
- *	memory_barrier();
- *	b = 3;				y = b;
- *					read_barrier_depends();
- *					x = a;
- * </programlisting>
- *
- * does not enforce ordering, since there is no data dependency between
- * the read of "a" and the read of "b".  Therefore, on some CPUs, such
- * as Alpha, "y" could be set to 3 and "x" to 0.  Use rmb()
- * in cases like this where there are no data dependencies.
- **/
-
-#define read_barrier_depends()	do { } while(0)
-
-#ifdef CONFIG_SMP
-#define smp_mb()	mb()
-#ifdef CONFIG_X86_PPRO_FENCE
-# define smp_rmb()	rmb()
-#else
-# define smp_rmb()	barrier()
-#endif
-#ifdef CONFIG_X86_OOSTORE
-# define smp_wmb() 	wmb()
-#else
-# define smp_wmb()	barrier()
-#endif
-#define smp_read_barrier_depends()	read_barrier_depends()
-#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
-#else
-#define smp_mb()	barrier()
-#define smp_rmb()	barrier()
-#define smp_wmb()	barrier()
-#define smp_read_barrier_depends()	do { } while(0)
-#define set_mb(var, value) do { var = value; barrier(); } while (0)
-#endif
-
 #include <linux/irqflags.h>
 
 /*
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 560470ea27c8..9def35eb75e3 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -48,31 +48,6 @@
     
 #endif	/* __KERNEL__ */
 
-#ifdef CONFIG_SMP
-#define smp_mb()	mb()
-#define smp_rmb()	barrier()
-#define smp_wmb()	barrier()
-#define smp_read_barrier_depends()	do {} while(0)
-#else
-#define smp_mb()	barrier()
-#define smp_rmb()	barrier()
-#define smp_wmb()	barrier()
-#define smp_read_barrier_depends()	do {} while(0)
-#endif
-
-    
-/*
- * Force strict CPU ordering.
- * And yes, this is required on UP too when we're talking
- * to devices.
- */
-#define mb() 	asm volatile("mfence":::"memory")
-#define rmb()	asm volatile("lfence":::"memory")
-#define wmb()	asm volatile("sfence" ::: "memory")
-
-#define read_barrier_depends()	do {} while(0)
-#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
-
 static inline unsigned long read_cr8(void)
 {
 	unsigned long cr8;
-- 
cgit v1.2.3


From 0a3b4d151a8e57f50ce5a12c7ea5ec9965cf0b5a Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: move switch_to macro to system.h

This patch moves the switch_to() macro to system.h

As those macros are fundamentally different between i386 and x86_64,
they are enclosed around an ifdef.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h    | 61 +++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/system_32.h | 31 -----------------------
 include/asm-x86/system_64.h | 43 --------------------------------
 3 files changed, 61 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 4c15eb11a917..ba3403f1d020 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -6,8 +6,69 @@
 #include <linux/kernel.h>
 
 #ifdef CONFIG_X86_32
+#define AT_VECTOR_SIZE_ARCH 2 /* entries in ARCH_DLINFO */
+
+struct task_struct; /* one of the stranger aspects of C forward declarations */
+extern struct task_struct *FASTCALL(__switch_to(struct task_struct *prev,
+						struct task_struct *next));
+
+/*
+ * Saving eflags is important. It switches not only IOPL between tasks,
+ * it also protects other tasks from NT leaking through sysenter etc.
+ */
+#define switch_to(prev, next, last) do {				\
+	unsigned long esi, edi;						\
+	asm volatile("pushfl\n\t"		/* Save flags */	\
+		     "pushl %%ebp\n\t"					\
+		     "movl %%esp,%0\n\t"	/* save ESP */		\
+		     "movl %5,%%esp\n\t"	/* restore ESP */	\
+		     "movl $1f,%1\n\t"		/* save EIP */		\
+		     "pushl %6\n\t"		/* restore EIP */	\
+		     "jmp __switch_to\n"				\
+		     "1:\t"						\
+		     "popl %%ebp\n\t"					\
+		     "popfl"						\
+		     :"=m" (prev->thread.sp), "=m" (prev->thread.ip),	\
+		      "=a" (last), "=S" (esi), "=D" (edi)		\
+		     :"m" (next->thread.sp), "m" (next->thread.ip),	\
+		      "2" (prev), "d" (next));				\
+} while (0)
+
 # include "system_32.h"
 #else
+#define __SAVE(reg, offset) "movq %%" #reg ",(14-" #offset ")*8(%%rsp)\n\t"
+#define __RESTORE(reg, offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
+
+/* frame pointer must be last for get_wchan */
+#define SAVE_CONTEXT    "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
+#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\t"
+
+#define __EXTRA_CLOBBER  \
+	, "rcx", "rbx", "rdx", "r8", "r9", "r10", "r11", \
+	  "r12", "r13", "r14", "r15"
+
+/* Save restore flags to clear handle leaking NT */
+#define switch_to(prev, next, last) \
+	asm volatile(SAVE_CONTEXT					  \
+	     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
+	     "movq %P[threadrsp](%[next]),%%rsp\n\t" /* restore RSP */	  \
+	     "call __switch_to\n\t"					  \
+	     ".globl thread_return\n"					  \
+	     "thread_return:\n\t"					  \
+	     "movq %%gs:%P[pda_pcurrent],%%rsi\n\t"			  \
+	     "movq %P[thread_info](%%rsi),%%r8\n\t"			  \
+	     LOCK_PREFIX "btr  %[tif_fork],%P[ti_flags](%%r8)\n\t"	  \
+	     "movq %%rax,%%rdi\n\t" 					  \
+	     "jc   ret_from_fork\n\t"					  \
+	     RESTORE_CONTEXT						  \
+	     : "=a" (last)					  	  \
+	     : [next] "S" (next), [prev] "D" (prev),			  \
+	       [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \
+	       [ti_flags] "i" (offsetof(struct thread_info, flags)),	  \
+	       [tif_fork] "i" (TIF_FORK),			  	  \
+	       [thread_info] "i" (offsetof(struct task_struct, stack)),   \
+	       [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent))  \
+	     : "memory", "cc" __EXTRA_CLOBBER)
 # include "system_64.h"
 #endif
 
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
index 7da0716fb317..83af46443bd0 100644
--- a/include/asm-x86/system_32.h
+++ b/include/asm-x86/system_32.h
@@ -5,37 +5,6 @@
 #include <asm/cpufeature.h>
 #include <asm/cmpxchg.h>
 
-#ifdef __KERNEL__
-#define AT_VECTOR_SIZE_ARCH 2 /* entries in ARCH_DLINFO */
-
-struct task_struct;	/* one of the stranger aspects of C forward declarations.. */
-extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struct task_struct *next));
-
-/*
- * Saving eflags is important. It switches not only IOPL between tasks,
- * it also protects other tasks from NT leaking through sysenter etc.
- */
-#define switch_to(prev,next,last) do {					\
-	unsigned long esi,edi;						\
-	asm volatile("pushfl\n\t"		/* Save flags */	\
-		     "pushl %%ebp\n\t"					\
-		     "movl %%esp,%0\n\t"	/* save ESP */		\
-		     "movl %5,%%esp\n\t"	/* restore ESP */	\
-		     "movl $1f,%1\n\t"		/* save EIP */		\
-		     "pushl %6\n\t"		/* restore EIP */	\
-		     "jmp __switch_to\n"				\
-		     "1:\t"						\
-		     "popl %%ebp\n\t"					\
-		     "popfl"						\
-		     :"=m" (prev->thread.sp),"=m" (prev->thread.ip),	\
-		      "=a" (last),"=S" (esi),"=D" (edi)			\
-		     :"m" (next->thread.sp),"m" (next->thread.ip),	\
-		      "2" (prev), "d" (next));				\
-} while (0)
-
-#endif	/* __KERNEL__ */
-
-
 #include <linux/irqflags.h>
 
 /*
diff --git a/include/asm-x86/system_64.h b/include/asm-x86/system_64.h
index 9def35eb75e3..97fa251ccb2b 100644
--- a/include/asm-x86/system_64.h
+++ b/include/asm-x86/system_64.h
@@ -4,49 +4,6 @@
 #include <asm/segment.h>
 #include <asm/cmpxchg.h>
 
-#ifdef __KERNEL__
-
-/* entries in ARCH_DLINFO: */
-#ifdef CONFIG_IA32_EMULATION
-# define AT_VECTOR_SIZE_ARCH 2
-#else
-# define AT_VECTOR_SIZE_ARCH 1
-#endif
-
-#define __SAVE(reg,offset) "movq %%" #reg ",(14-" #offset ")*8(%%rsp)\n\t"
-#define __RESTORE(reg,offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
-
-/* frame pointer must be last for get_wchan */
-#define SAVE_CONTEXT    "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
-#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\t"
-
-#define __EXTRA_CLOBBER  \
-	,"rcx","rbx","rdx","r8","r9","r10","r11","r12","r13","r14","r15"
-
-/* Save restore flags to clear handle leaking NT */
-#define switch_to(prev,next,last) \
-	asm volatile(SAVE_CONTEXT						    \
-		     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
-		     "movq %P[threadrsp](%[next]),%%rsp\n\t" /* restore RSP */	  \
-		     "call __switch_to\n\t"					  \
-		     ".globl thread_return\n"					\
-		     "thread_return:\n\t"					    \
-		     "movq %%gs:%P[pda_pcurrent],%%rsi\n\t"			  \
-		     "movq %P[thread_info](%%rsi),%%r8\n\t"			  \
-		     LOCK_PREFIX "btr  %[tif_fork],%P[ti_flags](%%r8)\n\t"	  \
-		     "movq %%rax,%%rdi\n\t" 					  \
-		     "jc   ret_from_fork\n\t"					  \
-		     RESTORE_CONTEXT						    \
-		     : "=a" (last)					  	  \
-		     : [next] "S" (next), [prev] "D" (prev),			  \
-		       [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \
-		       [ti_flags] "i" (offsetof(struct thread_info, flags)),\
-		       [tif_fork] "i" (TIF_FORK),			  \
-		       [thread_info] "i" (offsetof(struct task_struct, stack)), \
-		       [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent))   \
-		     : "memory", "cc" __EXTRA_CLOBBER)
-    
-#endif	/* __KERNEL__ */
 
 static inline unsigned long read_cr8(void)
 {
-- 
cgit v1.2.3


From d46d7d754014a299fa9b3400d080e09bfe4d629f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:08 +0100
Subject: x86: unify system.h

This patch finishes the unification of system.h file.
i386 needs a constant to be defined, and it is defined inside an ifdef.

Other than that, pretty much nothing but includes are left in the arch
specific headers, and they are deleted.

[ mingo@elte.hu: 64-bit needs the cr8 access inlines. ]

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h    | 26 ++++++++++++++++++++++++--
 include/asm-x86/system_32.h | 15 ---------------
 2 files changed, 24 insertions(+), 17 deletions(-)
 delete mode 100644 include/asm-x86/system_32.h

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index ba3403f1d020..8a37dad38bc0 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -2,8 +2,12 @@
 #define _ASM_X86_SYSTEM_H_
 
 #include <asm/asm.h>
+#include <asm/segment.h>
+#include <asm/cpufeature.h>
+#include <asm/cmpxchg.h>
 
 #include <linux/kernel.h>
+#include <linux/irqflags.h>
 
 #ifdef CONFIG_X86_32
 #define AT_VECTOR_SIZE_ARCH 2 /* entries in ARCH_DLINFO */
@@ -34,7 +38,10 @@ extern struct task_struct *FASTCALL(__switch_to(struct task_struct *prev,
 		      "2" (prev), "d" (next));				\
 } while (0)
 
-# include "system_32.h"
+/*
+ * disable hlt during certain critical i/o operations
+ */
+#define HAVE_DISABLE_HLT
 #else
 #define __SAVE(reg, offset) "movq %%" #reg ",(14-" #offset ")*8(%%rsp)\n\t"
 #define __RESTORE(reg, offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
@@ -69,7 +76,6 @@ extern struct task_struct *FASTCALL(__switch_to(struct task_struct *prev,
 	       [thread_info] "i" (offsetof(struct task_struct, stack)),   \
 	       [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent))  \
 	     : "memory", "cc" __EXTRA_CLOBBER)
-# include "system_64.h"
 #endif
 
 #ifdef __KERNEL__
@@ -236,6 +242,22 @@ static inline void native_wbinvd(void)
 #define write_cr4(x)	(native_write_cr4(x))
 #define wbinvd()	(native_wbinvd())
 
+#ifdef CONFIG_X86_64
+
+static inline unsigned long read_cr8(void)
+{
+	unsigned long cr8;
+	asm volatile("movq %%cr8,%0" : "=r" (cr8));
+	return cr8;
+}
+
+static inline void write_cr8(unsigned long val)
+{
+	asm volatile("movq %0,%%cr8" :: "r" (val) : "memory");
+}
+
+#endif
+
 /* Clear the 'TS' bit */
 #define clts()		(native_clts())
 
diff --git a/include/asm-x86/system_32.h b/include/asm-x86/system_32.h
deleted file mode 100644
index 83af46443bd0..000000000000
--- a/include/asm-x86/system_32.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef __ASM_SYSTEM_H
-#define __ASM_SYSTEM_H
-
-#include <asm/segment.h>
-#include <asm/cpufeature.h>
-#include <asm/cmpxchg.h>
-
-#include <linux/irqflags.h>
-
-/*
- * disable hlt during certain critical i/o operations
- */
-#define HAVE_DISABLE_HLT
-
-#endif
-- 
cgit v1.2.3


From eee3af4a2c83a97fff107ddc445d9df6fded9ce4 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x86, ptrace: support for branch trace store(BTS)

Resend using different mail client

Changes to the last version:
- split implementation into two layers: ds/bts and ptrace
- renamed TIF's
- save/restore ds save area msr in __switch_to_xtra()
- make block-stepping only look at BTF bit

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile_32      |   1 +
 arch/x86/kernel/Makefile_64      |   1 +
 arch/x86/kernel/cpu/intel.c      |   5 +
 arch/x86/kernel/ds.c             | 429 +++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/process_32.c     |  19 +-
 arch/x86/kernel/process_64.c     |  26 ++-
 arch/x86/kernel/ptrace.c         | 212 +++++++++++++++++++
 arch/x86/kernel/setup_64.c       |   5 +
 arch/x86/kernel/step.c           |  18 +-
 include/asm-x86/ds.h             |  65 ++++++
 include/asm-x86/processor_32.h   |   3 +
 include/asm-x86/processor_64.h   |   3 +
 include/asm-x86/ptrace-abi.h     |  52 +++++
 include/asm-x86/ptrace.h         |  11 +
 include/asm-x86/thread_info_32.h |  12 +-
 include/asm-x86/thread_info_64.h |   9 +-
 16 files changed, 859 insertions(+), 12 deletions(-)
 create mode 100644 arch/x86/kernel/ds.c
 create mode 100644 include/asm-x86/ds.h

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index b2d7aea4c82d..cc2651bcc07f 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -11,6 +11,7 @@ obj-y	:= process_32.o signal_32.o entry_32.o traps_32.o irq_32.o \
 		quirks.o i8237.o topology.o alternative.o i8253.o tsc_32.o io_delay.o rtc.o
 
 obj-y				+= ptrace.o
+obj-y				+= ds.o
 obj-y				+= tls.o
 obj-y				+= step.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index 19af64e1a3fc..2ec96acf6486 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -13,6 +13,7 @@ obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
 		i8253.o io_delay.o rtc.o
 
 obj-y				+= ptrace.o
+obj-y				+= ds.o
 obj-y				+= step.o
 
 obj-$(CONFIG_IA32_EMULATION)	+= tls.o
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index 867ff94579be..e4b7e73e9024 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -11,6 +11,8 @@
 #include <asm/pgtable.h>
 #include <asm/msr.h>
 #include <asm/uaccess.h>
+#include <asm/ptrace.h>
+#include <asm/ds.h>
 
 #include "cpu.h"
 
@@ -219,6 +221,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		if (!(l1 & (1<<12)))
 			set_bit(X86_FEATURE_PEBS, c->x86_capability);
 	}
+
+	if (cpu_has_bts)
+		ds_init_intel(c);
 }
 
 static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
new file mode 100644
index 000000000000..996a7c4f5963
--- /dev/null
+++ b/arch/x86/kernel/ds.c
@@ -0,0 +1,429 @@
+/*
+ * Debug Store support
+ *
+ * This provides a low-level interface to the hardware's Debug Store
+ * feature that is used for last branch recording (LBR) and
+ * precise-event based sampling (PEBS).
+ *
+ * Different architectures use a different DS layout/pointer size.
+ * The below functions therefore work on a void*.
+ *
+ *
+ * Since there is no user for PEBS, yet, only LBR (or branch
+ * trace store, BTS) is supported.
+ *
+ *
+ * Copyright (C) 2007 Intel Corporation.
+ * Markus Metzger <markus.t.metzger@intel.com>, Dec 2007
+ */
+
+#include <asm/ds.h>
+
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+
+
+/*
+ * Debug Store (DS) save area configuration (see Intel64 and IA32
+ * Architectures Software Developer's Manual, section 18.5)
+ *
+ * The DS configuration consists of the following fields; different
+ * architetures vary in the size of those fields.
+ * - double-word aligned base linear address of the BTS buffer
+ * - write pointer into the BTS buffer
+ * - end linear address of the BTS buffer (one byte beyond the end of
+ *   the buffer)
+ * - interrupt pointer into BTS buffer
+ *   (interrupt occurs when write pointer passes interrupt pointer)
+ * - double-word aligned base linear address of the PEBS buffer
+ * - write pointer into the PEBS buffer
+ * - end linear address of the PEBS buffer (one byte beyond the end of
+ *   the buffer)
+ * - interrupt pointer into PEBS buffer
+ *   (interrupt occurs when write pointer passes interrupt pointer)
+ * - value to which counter is reset following counter overflow
+ *
+ * On later architectures, the last branch recording hardware uses
+ * 64bit pointers even in 32bit mode.
+ *
+ *
+ * Branch Trace Store (BTS) records store information about control
+ * flow changes. They at least provide the following information:
+ * - source linear address
+ * - destination linear address
+ *
+ * Netburst supported a predicated bit that had been dropped in later
+ * architectures. We do not suppor it.
+ *
+ *
+ * In order to abstract from the actual DS and BTS layout, we describe
+ * the access to the relevant fields.
+ * Thanks to Andi Kleen for proposing this design.
+ *
+ * The implementation, however, is not as general as it might seem. In
+ * order to stay somewhat simple and efficient, we assume an
+ * underlying unsigned type (mostly a pointer type) and we expect the
+ * field to be at least as big as that type.
+ */
+
+/*
+ * A special from_ip address to indicate that the BTS record is an
+ * info record that needs to be interpreted or skipped.
+ */
+#define BTS_ESCAPE_ADDRESS (-1)
+
+/*
+ * A field access descriptor
+ */
+struct access_desc {
+	unsigned char offset;
+	unsigned char size;
+};
+
+/*
+ * The configuration for a particular DS/BTS hardware implementation.
+ */
+struct ds_configuration {
+	/* the DS configuration */
+	unsigned char  sizeof_ds;
+	struct access_desc bts_buffer_base;
+	struct access_desc bts_index;
+	struct access_desc bts_absolute_maximum;
+	struct access_desc bts_interrupt_threshold;
+	/* the BTS configuration */
+	unsigned char  sizeof_bts;
+	struct access_desc from_ip;
+	struct access_desc to_ip;
+	/* BTS variants used to store additional information like
+	   timestamps */
+	struct access_desc info_type;
+	struct access_desc info_data;
+	unsigned long debugctl_mask;
+};
+
+/*
+ * The global configuration used by the below accessor functions
+ */
+static struct ds_configuration ds_cfg;
+
+/*
+ * Accessor functions for some DS and BTS fields using the above
+ * global ptrace_bts_cfg.
+ */
+static inline void *get_bts_buffer_base(char *base)
+{
+	return *(void **)(base + ds_cfg.bts_buffer_base.offset);
+}
+static inline void set_bts_buffer_base(char *base, void *value)
+{
+	(*(void **)(base + ds_cfg.bts_buffer_base.offset)) = value;
+}
+static inline void *get_bts_index(char *base)
+{
+	return *(void **)(base + ds_cfg.bts_index.offset);
+}
+static inline void set_bts_index(char *base, void *value)
+{
+	(*(void **)(base + ds_cfg.bts_index.offset)) = value;
+}
+static inline void *get_bts_absolute_maximum(char *base)
+{
+	return *(void **)(base + ds_cfg.bts_absolute_maximum.offset);
+}
+static inline void set_bts_absolute_maximum(char *base, void *value)
+{
+	(*(void **)(base + ds_cfg.bts_absolute_maximum.offset)) = value;
+}
+static inline void *get_bts_interrupt_threshold(char *base)
+{
+	return *(void **)(base + ds_cfg.bts_interrupt_threshold.offset);
+}
+static inline void set_bts_interrupt_threshold(char *base, void *value)
+{
+	(*(void **)(base + ds_cfg.bts_interrupt_threshold.offset)) = value;
+}
+static inline long get_from_ip(char *base)
+{
+	return *(long *)(base + ds_cfg.from_ip.offset);
+}
+static inline void set_from_ip(char *base, long value)
+{
+	(*(long *)(base + ds_cfg.from_ip.offset)) = value;
+}
+static inline long get_to_ip(char *base)
+{
+	return *(long *)(base + ds_cfg.to_ip.offset);
+}
+static inline void set_to_ip(char *base, long value)
+{
+	(*(long *)(base + ds_cfg.to_ip.offset)) = value;
+}
+static inline unsigned char get_info_type(char *base)
+{
+	return *(unsigned char *)(base + ds_cfg.info_type.offset);
+}
+static inline void set_info_type(char *base, unsigned char value)
+{
+	(*(unsigned char *)(base + ds_cfg.info_type.offset)) = value;
+}
+/*
+ * The info data might overlap with the info type on some architectures.
+ * We therefore read and write the exact number of bytes.
+ */
+static inline unsigned long long get_info_data(char *base)
+{
+	unsigned long long value = 0;
+	memcpy(&value,
+	       base + ds_cfg.info_data.offset,
+	       ds_cfg.info_data.size);
+	return value;
+}
+static inline void set_info_data(char *base, unsigned long long value)
+{
+	memcpy(base + ds_cfg.info_data.offset,
+	       &value,
+	       ds_cfg.info_data.size);
+}
+
+
+int ds_allocate(void **dsp, size_t bts_size_in_records)
+{
+	size_t bts_size_in_bytes = 0;
+	void *bts = 0;
+	void *ds = 0;
+
+	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
+		return -EOPNOTSUPP;
+
+	if (bts_size_in_records < 0)
+		return -EINVAL;
+
+	bts_size_in_bytes =
+		bts_size_in_records * ds_cfg.sizeof_bts;
+
+	if (bts_size_in_bytes <= 0)
+		return -EINVAL;
+
+	bts = kzalloc(bts_size_in_bytes, GFP_KERNEL);
+
+	if (!bts)
+		return -ENOMEM;
+
+	ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL);
+
+	if (!ds) {
+		kfree(bts);
+		return -ENOMEM;
+	}
+
+	set_bts_buffer_base(ds, bts);
+	set_bts_index(ds, bts);
+	set_bts_absolute_maximum(ds, bts + bts_size_in_bytes);
+	set_bts_interrupt_threshold(ds, bts + bts_size_in_bytes + 1);
+
+	*dsp = ds;
+	return 0;
+}
+
+int ds_free(void **dsp)
+{
+	if (*dsp)
+		kfree(get_bts_buffer_base(*dsp));
+	kfree(*dsp);
+	*dsp = 0;
+
+	return 0;
+}
+
+int ds_get_bts_size(void *ds)
+{
+	size_t size_in_bytes;
+
+	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
+		return -EOPNOTSUPP;
+
+	size_in_bytes =
+		get_bts_absolute_maximum(ds) -
+		get_bts_buffer_base(ds);
+
+	return size_in_bytes / ds_cfg.sizeof_bts;
+}
+
+int ds_get_bts_index(void *ds)
+{
+	size_t index_offset_in_bytes;
+
+	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
+		return -EOPNOTSUPP;
+
+	index_offset_in_bytes =
+		get_bts_index(ds) -
+		get_bts_buffer_base(ds);
+
+	return index_offset_in_bytes / ds_cfg.sizeof_bts;
+}
+
+int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
+{
+	void *bts;
+
+	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
+		return -EOPNOTSUPP;
+
+	if (index < 0)
+		return -EINVAL;
+
+	if (index >= ds_get_bts_size(ds))
+		return -EINVAL;
+
+	bts = get_bts_buffer_base(ds);
+	bts = (char *)bts + (index * ds_cfg.sizeof_bts);
+
+	memset(out, 0, sizeof(*out));
+	if (get_from_ip(bts) == BTS_ESCAPE_ADDRESS) {
+		out->qualifier         = get_info_type(bts);
+		out->variant.timestamp = get_info_data(bts);
+	} else {
+		out->qualifier = BTS_BRANCH;
+		out->variant.lbr.from_ip = get_from_ip(bts);
+		out->variant.lbr.to_ip   = get_to_ip(bts);
+	}
+
+	return 0;
+}
+
+int ds_write_bts(void *ds, const struct bts_struct *in)
+{
+	void *bts;
+
+	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
+		return -EOPNOTSUPP;
+
+	if (ds_get_bts_size(ds) <= 0)
+		return -ENXIO;
+
+	bts = get_bts_index(ds);
+
+	memset(bts, 0, ds_cfg.sizeof_bts);
+	switch (in->qualifier) {
+	case BTS_INVALID:
+		break;
+
+	case BTS_BRANCH:
+		set_from_ip(bts, in->variant.lbr.from_ip);
+		set_to_ip(bts, in->variant.lbr.to_ip);
+		break;
+
+	case BTS_TASK_ARRIVES:
+	case BTS_TASK_DEPARTS:
+		set_from_ip(bts, BTS_ESCAPE_ADDRESS);
+		set_info_type(bts, in->qualifier);
+		set_info_data(bts, in->variant.timestamp);
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	bts = (char *)bts + ds_cfg.sizeof_bts;
+	if (bts >= get_bts_absolute_maximum(ds))
+		bts = get_bts_buffer_base(ds);
+	set_bts_index(ds, bts);
+
+	return 0;
+}
+
+unsigned long ds_debugctl_mask(void)
+{
+	return ds_cfg.debugctl_mask;
+}
+
+#ifdef __i386__
+static const struct ds_configuration ds_cfg_netburst = {
+	.sizeof_ds = 9 * 4,
+	.bts_buffer_base = { 0, 4 },
+	.bts_index = { 4, 4 },
+	.bts_absolute_maximum = { 8, 4 },
+	.bts_interrupt_threshold = { 12, 4 },
+	.sizeof_bts = 3 * 4,
+	.from_ip = { 0, 4 },
+	.to_ip = { 4, 4 },
+	.info_type = { 4, 1 },
+	.info_data = { 5, 7 },
+	.debugctl_mask = (1<<2)|(1<<3)
+};
+
+static const struct ds_configuration ds_cfg_pentium_m = {
+	.sizeof_ds = 9 * 4,
+	.bts_buffer_base = { 0, 4 },
+	.bts_index = { 4, 4 },
+	.bts_absolute_maximum = { 8, 4 },
+	.bts_interrupt_threshold = { 12, 4 },
+	.sizeof_bts = 3 * 4,
+	.from_ip = { 0, 4 },
+	.to_ip = { 4, 4 },
+	.info_type = { 4, 1 },
+	.info_data = { 5, 7 },
+	.debugctl_mask = (1<<6)|(1<<7)
+};
+#endif /* _i386_ */
+
+static const struct ds_configuration ds_cfg_core2 = {
+	.sizeof_ds = 9 * 8,
+	.bts_buffer_base = { 0, 8 },
+	.bts_index = { 8, 8 },
+	.bts_absolute_maximum = { 16, 8 },
+	.bts_interrupt_threshold = { 24, 8 },
+	.sizeof_bts = 3 * 8,
+	.from_ip = { 0, 8 },
+	.to_ip = { 8, 8 },
+	.info_type = { 8, 1 },
+	.info_data = { 9, 7 },
+	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
+};
+
+static inline void
+ds_configure(const struct ds_configuration *cfg)
+{
+	ds_cfg = *cfg;
+}
+
+void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
+{
+	switch (c->x86) {
+	case 0x6:
+		switch (c->x86_model) {
+#ifdef __i386__
+		case 0xD:
+		case 0xE: /* Pentium M */
+			ds_configure(&ds_cfg_pentium_m);
+			break;
+#endif /* _i386_ */
+		case 0xF: /* Core2 */
+			ds_configure(&ds_cfg_core2);
+			break;
+		default:
+			/* sorry, don't know about them */
+			break;
+		}
+		break;
+	case 0xF:
+		switch (c->x86_model) {
+#ifdef __i386__
+		case 0x0:
+		case 0x1:
+		case 0x2: /* Netburst */
+			ds_configure(&ds_cfg_netburst);
+			break;
+#endif /* _i386_ */
+		default:
+			/* sorry, don't know about them */
+			break;
+		}
+		break;
+	default:
+		/* sorry, don't know about them */
+		break;
+	}
+}
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 5350763a2d03..2b9db9371060 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -614,11 +614,21 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		 struct tss_struct *tss)
 {
 	struct thread_struct *prev, *next;
+	unsigned long debugctl;
 
 	prev = &prev_p->thread;
 	next = &next_p->thread;
 
-	if (next->debugctlmsr != prev->debugctlmsr)
+	debugctl = prev->debugctlmsr;
+	if (next->ds_area_msr != prev->ds_area_msr) {
+		/* we clear debugctl to make sure DS
+		 * is not in use when we change it */
+		debugctl = 0;
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
+		wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0);
+	}
+
+	if (next->debugctlmsr != debugctl)
 		wrmsr(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr, 0);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
@@ -642,6 +652,13 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 	}
 #endif
 
+	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
+		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
+
+	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
+		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
+
+
 	if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
 		/*
 		 * Disable the bitmap via an invalid offset. We still cache
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 057b5442ffda..843bf0c978a4 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -568,11 +568,21 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 				    struct tss_struct *tss)
 {
 	struct thread_struct *prev, *next;
+	unsigned long debugctl;
 
 	prev = &prev_p->thread,
 	next = &next_p->thread;
 
-	if (next->debugctlmsr != prev->debugctlmsr)
+	debugctl = prev->debugctlmsr;
+	if (next->ds_area_msr != prev->ds_area_msr) {
+		/* we clear debugctl to make sure DS
+		 * is not in use when we change it */
+		debugctl = 0;
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
+		wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr);
+	}
+
+	if (next->debugctlmsr != debugctl)
 		wrmsrl(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
@@ -598,6 +608,16 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 		 */
 		memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
 	}
+
+	/*
+	 * Last branch recording recofiguration of trace hardware and
+	 * disentangling of trace data per task.
+	 */
+	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
+		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
+
+	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
+		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
 }
 
 /*
@@ -701,8 +721,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/*
 	 * Now maybe reload the debug registers and handle I/O bitmaps
 	 */
-	if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
-	    || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
+	if (unlikely(task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT ||
+		     task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV))
 		__switch_to_xtra(prev_p, next_p, tss);
 
 	/* If the task has used fpu the last 5 timeslices, just do a full
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 3399c1be79b8..8d0dd8b5effe 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -2,6 +2,9 @@
 /*
  * Pentium III FXSR, SSE support
  *	Gareth Hughes <gareth@valinux.com>, May 2000
+ *
+ * BTS tracing
+ *	Markus Metzger <markus.t.metzger@intel.com>, Dec 2007
  */
 
 #include <linux/kernel.h>
@@ -26,6 +29,14 @@
 #include <asm/desc.h>
 #include <asm/prctl.h>
 #include <asm/proto.h>
+#include <asm/ds.h>
+
+
+/*
+ * The maximal size of a BTS buffer per traced task in number of BTS
+ * records.
+ */
+#define PTRACE_BTS_BUFFER_MAX 4000
 
 /*
  * does not yet catch signals sent when the child dies.
@@ -455,6 +466,165 @@ static int ptrace_set_debugreg(struct task_struct *child,
 	return 0;
 }
 
+static int ptrace_bts_max_buffer_size(void)
+{
+	return PTRACE_BTS_BUFFER_MAX;
+}
+
+static int ptrace_bts_get_buffer_size(struct task_struct *child)
+{
+	if (!child->thread.ds_area_msr)
+		return -ENXIO;
+
+	return ds_get_bts_size((void *)child->thread.ds_area_msr);
+}
+
+static int ptrace_bts_get_index(struct task_struct *child)
+{
+	if (!child->thread.ds_area_msr)
+		return -ENXIO;
+
+	return ds_get_bts_index((void *)child->thread.ds_area_msr);
+}
+
+static int ptrace_bts_read_record(struct task_struct *child,
+				  long index,
+				  struct bts_struct __user *out)
+{
+	struct bts_struct ret;
+	int retval;
+
+	if (!child->thread.ds_area_msr)
+		return -ENXIO;
+
+	retval = ds_read_bts((void *)child->thread.ds_area_msr,
+			     index, &ret);
+	if (retval)
+		return retval;
+
+	if (copy_to_user(out, &ret, sizeof(ret)))
+		return -EFAULT;
+
+	return sizeof(ret);
+}
+
+static int ptrace_bts_write_record(struct task_struct *child,
+				   const struct bts_struct *in)
+{
+	int retval;
+
+	if (!child->thread.ds_area_msr)
+		return -ENXIO;
+
+	retval = ds_write_bts((void *)child->thread.ds_area_msr, in);
+	if (retval)
+		return retval;
+
+	return sizeof(*in);
+}
+
+static int ptrace_bts_config(struct task_struct *child,
+			     unsigned long options)
+{
+	unsigned long debugctl_mask = ds_debugctl_mask();
+	int retval;
+
+	retval = ptrace_bts_get_buffer_size(child);
+	if (retval < 0)
+		return retval;
+	if (retval == 0)
+		return -ENXIO;
+
+	if (options & PTRACE_BTS_O_TRACE_TASK) {
+		child->thread.debugctlmsr |= debugctl_mask;
+		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+	} else {
+		/* there is no way for us to check whether we 'own'
+		 * the respective bits in the DEBUGCTL MSR, we're
+		 * about to clear */
+		child->thread.debugctlmsr &= ~debugctl_mask;
+
+		if (!child->thread.debugctlmsr)
+			clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+	}
+
+	if (options & PTRACE_BTS_O_TIMESTAMPS)
+		set_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
+	else
+		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
+
+	return 0;
+}
+
+static int ptrace_bts_status(struct task_struct *child)
+{
+	unsigned long debugctl_mask = ds_debugctl_mask();
+	int retval, status = 0;
+
+	retval = ptrace_bts_get_buffer_size(child);
+	if (retval < 0)
+		return retval;
+	if (retval == 0)
+		return -ENXIO;
+
+	if (ptrace_bts_get_buffer_size(child) <= 0)
+		return -ENXIO;
+
+	if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
+	    child->thread.debugctlmsr & debugctl_mask)
+		status |= PTRACE_BTS_O_TRACE_TASK;
+	if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
+		status |= PTRACE_BTS_O_TIMESTAMPS;
+
+	return status;
+}
+
+static int ptrace_bts_allocate_bts(struct task_struct *child,
+				   int size_in_records)
+{
+	int retval = 0;
+	void *ds;
+
+	if (size_in_records < 0)
+		return -EINVAL;
+
+	if (size_in_records > ptrace_bts_max_buffer_size())
+		return -EINVAL;
+
+	if (size_in_records == 0) {
+		ptrace_bts_config(child, /* options = */ 0);
+	} else {
+		retval = ds_allocate(&ds, size_in_records);
+		if (retval)
+			return retval;
+	}
+
+	if (child->thread.ds_area_msr)
+		ds_free((void **)&child->thread.ds_area_msr);
+
+	child->thread.ds_area_msr = (unsigned long)ds;
+	if (child->thread.ds_area_msr)
+		set_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+	else
+		clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+
+	return retval;
+}
+
+void ptrace_bts_take_timestamp(struct task_struct *tsk,
+			       enum bts_qualifier qualifier)
+{
+	struct bts_struct rec = {
+		.qualifier = qualifier,
+		.variant.timestamp = sched_clock()
+	};
+
+	if (ptrace_bts_get_buffer_size(tsk) <= 0)
+		return;
+
+	ptrace_bts_write_record(tsk, &rec);
+}
+
 /*
  * Called by kernel/ptrace.c when detaching..
  *
@@ -466,6 +636,11 @@ void ptrace_disable(struct task_struct *child)
 #ifdef TIF_SYSCALL_EMU
 	clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 #endif
+	ptrace_bts_config(child, /* options = */ 0);
+	if (child->thread.ds_area_msr) {
+	    ds_free((void **)&child->thread.ds_area_msr);
+	    clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+	}
 }
 
 long arch_ptrace(struct task_struct *child, long request, long addr, long data)
@@ -626,6 +801,36 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 #endif
 
+	case PTRACE_BTS_MAX_BUFFER_SIZE:
+		ret = ptrace_bts_max_buffer_size();
+		break;
+
+	case PTRACE_BTS_ALLOCATE_BUFFER:
+		ret = ptrace_bts_allocate_bts(child, data);
+		break;
+
+	case PTRACE_BTS_GET_BUFFER_SIZE:
+		ret = ptrace_bts_get_buffer_size(child);
+		break;
+
+	case PTRACE_BTS_GET_INDEX:
+		ret = ptrace_bts_get_index(child);
+		break;
+
+	case PTRACE_BTS_READ_RECORD:
+		ret = ptrace_bts_read_record
+			(child, data,
+			 (struct bts_struct __user *) addr);
+		break;
+
+	case PTRACE_BTS_CONFIG:
+		ret = ptrace_bts_config(child, data);
+		break;
+
+	case PTRACE_BTS_STATUS:
+		ret = ptrace_bts_status(child);
+		break;
+
 	default:
 		ret = ptrace_request(child, request, addr, data);
 		break;
@@ -809,6 +1014,13 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 	case PTRACE_SETOPTIONS:
 	case PTRACE_SET_THREAD_AREA:
 	case PTRACE_GET_THREAD_AREA:
+	case PTRACE_BTS_MAX_BUFFER_SIZE:
+	case PTRACE_BTS_ALLOCATE_BUFFER:
+	case PTRACE_BTS_GET_BUFFER_SIZE:
+	case PTRACE_BTS_GET_INDEX:
+	case PTRACE_BTS_READ_RECORD:
+	case PTRACE_BTS_CONFIG:
+	case PTRACE_BTS_STATUS:
 		return sys_ptrace(request, pid, addr, data);
 
 	default:
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index ce4d6b52ce36..f2b131ef844e 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -60,6 +60,7 @@
 #include <asm/dmi.h>
 #include <asm/cacheflush.h>
 #include <asm/mce.h>
+#include <asm/ds.h>
 
 /*
  * Machine setup..
@@ -823,6 +824,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 			set_cpu_cap(c, X86_FEATURE_PEBS);
 	}
 
+
+	if (cpu_has_bts)
+		ds_init_intel(c);
+
 	n = c->extended_cpuid_level;
 	if (n >= 0x80000008) {
 		unsigned eax = cpuid_eax(0x80000008);
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index f55c003f5b63..21ea22fda5fc 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -169,9 +169,14 @@ static void enable_step(struct task_struct *child, bool block)
 	 */
 	if (enable_single_step(child) && block) {
 		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
-		write_debugctlmsr(child, DEBUGCTLMSR_BTF);
-	} else if (test_and_clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR)) {
-		write_debugctlmsr(child, 0);
+		write_debugctlmsr(child,
+				  child->thread.debugctlmsr | DEBUGCTLMSR_BTF);
+	} else {
+	    write_debugctlmsr(child,
+			      child->thread.debugctlmsr & ~TIF_DEBUGCTLMSR);
+
+	    if (!child->thread.debugctlmsr)
+		    clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
 	}
 }
 
@@ -190,8 +195,11 @@ void user_disable_single_step(struct task_struct *child)
 	/*
 	 * Make sure block stepping (BTF) is disabled.
 	 */
-	if (test_and_clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR))
-		write_debugctlmsr(child, 0);
+	write_debugctlmsr(child,
+			  child->thread.debugctlmsr & ~TIF_DEBUGCTLMSR);
+
+	if (!child->thread.debugctlmsr)
+		clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
 
 	/* Always clear TIF_SINGLESTEP... */
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h
new file mode 100644
index 000000000000..edd8467740a6
--- /dev/null
+++ b/include/asm-x86/ds.h
@@ -0,0 +1,65 @@
+/*
+ * Debug Store (DS) support
+ *
+ * This provides a low-level interface to the hardware's Debug Store
+ * feature that is used for last branch recording (LBR) and
+ * precise-event based sampling (PEBS).
+ *
+ * Different architectures use a different DS layout/pointer size.
+ * The below functions therefore work on a void*.
+ *
+ *
+ * Since there is no user for PEBS, yet, only LBR (or branch
+ * trace store, BTS) is supported.
+ *
+ *
+ * Copyright (C) 2007 Intel Corporation.
+ * Markus Metzger <markus.t.metzger@intel.com>, Dec 2007
+ */
+
+#ifndef _ASM_X86_DS_H
+#define _ASM_X86_DS_H
+
+#include <linux/types.h>
+#include <linux/init.h>
+
+struct cpuinfo_x86;
+
+
+/* a branch trace record entry
+ *
+ * In order to unify the interface between various processor versions,
+ * we use the below data structure for all processors.
+ */
+enum bts_qualifier {
+	BTS_INVALID = 0,
+	BTS_BRANCH,
+	BTS_TASK_ARRIVES,
+	BTS_TASK_DEPARTS
+};
+
+struct bts_struct {
+	enum bts_qualifier qualifier;
+	union {
+		/* BTS_BRANCH */
+		struct {
+			long from_ip;
+			long to_ip;
+		} lbr;
+		/* BTS_TASK_ARRIVES or
+		   BTS_TASK_DEPARTS */
+		unsigned long long timestamp;
+	} variant;
+};
+
+
+extern int ds_allocate(void **, size_t);
+extern int ds_free(void **);
+extern int ds_get_bts_size(void *);
+extern int ds_get_bts_index(void *);
+extern int ds_read_bts(void *, size_t, struct bts_struct *);
+extern int ds_write_bts(void *, const struct bts_struct *);
+extern unsigned long ds_debugctl_mask(void);
+extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *c);
+
+#endif /* _ASM_X86_DS_H */
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 0d83da198127..9c0ab7f26bd9 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -360,6 +360,9 @@ struct thread_struct {
 	unsigned long	io_bitmap_max;
 /* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
 	unsigned long	debugctlmsr;
+/* Debug Store - if not 0 points to a DS Save Area configuration;
+ *               goes into MSR_IA32_DS_AREA */
+	unsigned long	ds_area_msr;
 };
 
 #define INIT_THREAD  {							\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 0780f3e3fdfe..7b7f8a142e20 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -240,6 +240,9 @@ struct thread_struct {
 	unsigned io_bitmap_max;
 /* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
 	unsigned long	debugctlmsr;
+/* Debug Store - if not 0 points to a DS Save Area configuration;
+ *               goes into MSR_IA32_DS_AREA */
+	unsigned long	ds_area_msr;
 /* cached TLS descriptors. */
 	u64 tls_array[GDT_ENTRY_TLS_ENTRIES];
 } __attribute__((aligned(16)));
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index adce6b51df2e..6fadc5214e14 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -80,4 +80,56 @@
 
 #define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
 
+/* Return maximal BTS buffer size in number of records,
+   if successuf; -1, otherwise.
+   EOPNOTSUPP...processor does not support bts tracing */
+#define PTRACE_BTS_MAX_BUFFER_SIZE 40
+
+/* Allocate new bts buffer (free old one, if exists) of size DATA bts records;
+   parameter ADDR is ignored.
+   Return 0, if successful; -1, otherwise.
+   EOPNOTSUPP...processor does not support bts tracing
+   EINVAL.......invalid size in records
+   ENOMEM.......out of memory */
+#define PTRACE_BTS_ALLOCATE_BUFFER 41
+
+/* Return the size of the bts buffer in number of bts records,
+   if successful; -1, otherwise.
+   EOPNOTSUPP...processor does not support bts tracing
+   ENXIO........no buffer allocated */
+#define PTRACE_BTS_GET_BUFFER_SIZE 42
+
+/* Return the index of the next bts record to be written,
+   if successful; -1, otherwise.
+   EOPNOTSUPP...processor does not support bts tracing
+   ENXIO........no buffer allocated
+   After the first warp-around, this is the start of the circular bts buffer. */
+#define PTRACE_BTS_GET_INDEX 43
+
+/* Read the DATA'th bts record into a ptrace_bts_record buffer provided in ADDR.
+   Return 0, if successful; -1, otherwise
+   EOPNOTSUPP...processor does not support bts tracing
+   ENXIO........no buffer allocated
+   EINVAL.......invalid index */
+#define PTRACE_BTS_READ_RECORD 44
+
+/* Configure last branch trace; the configuration is given as a bit-mask of
+   PTRACE_BTS_O_* options in DATA; parameter ADDR is ignored.
+   Return 0, if successful; -1, otherwise
+   EOPNOTSUPP...processor does not support bts tracing
+   ENXIO........no buffer allocated */
+#define PTRACE_BTS_CONFIG 45
+
+/* Return the configuration as bit-mask of PTRACE_BTS_O_* options
+   if successful; -1, otherwise.
+   EOPNOTSUPP...processor does not support bts tracing
+   ENXIO........no buffer allocated */
+#define PTRACE_BTS_STATUS 46
+
+/* Trace configuration options */
+/* Collect last branch trace */
+#define PTRACE_BTS_O_TRACE_TASK 0x1
+/* Take timestamps when the task arrives and departs */
+#define PTRACE_BTS_O_TIMESTAMPS 0x2
+
 #endif
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 9228870f6157..a9a1bab1451a 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -4,8 +4,19 @@
 #include <linux/compiler.h>	/* For __user */
 #include <asm/ptrace-abi.h>
 
+
 #ifndef __ASSEMBLY__
 
+#ifdef __KERNEL__
+
+#include <asm/ds.h>
+
+struct task_struct;
+extern void ptrace_bts_take_timestamp(struct task_struct *, enum bts_qualifier);
+
+#endif /* __KERNEL__ */
+
+
 #ifdef __i386__
 /* this struct defines the way the registers are stored on the
    stack during a system call. */
diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index 306fc80800e1..5bd508260ffb 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -140,6 +140,8 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_NOTSC		20	/* TSC is not accessible in userland */
 #define TIF_FORCED_TF		21	/* true if TF in eflags artificially */
 #define TIF_DEBUGCTLMSR		22	/* uses thread_struct.debugctlmsr */
+#define TIF_DS_AREA_MSR 	23      /* uses thread_struct.ds_area_msr */
+#define TIF_BTS_TRACE_TS        24      /* record scheduling event timestamps */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -157,6 +159,8 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NOTSC		(1<<TIF_NOTSC)
 #define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
 #define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
+#define _TIF_DS_AREA_MSR	(1<<TIF_DS_AREA_MSR)
+#define _TIF_BTS_TRACE_TS	(1<<TIF_BTS_TRACE_TS)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -166,8 +170,12 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_ALLWORK_MASK	(0x0000FFFF & ~_TIF_SECCOMP)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW_NEXT (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUG | _TIF_DEBUGCTLMSR)
-#define _TIF_WORK_CTXSW_PREV (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUGCTLMSR)
+#define _TIF_WORK_CTXSW \
+    (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUGCTLMSR | \
+     _TIF_DS_AREA_MSR | _TIF_BTS_TRACE_TS)
+#define _TIF_WORK_CTXSW_PREV _TIF_WORK_CTXSW
+#define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW | _TIF_DEBUG)
+
 
 /*
  * Thread-synchronous status.
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index ee35fd12b541..c2911a99cc32 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -123,6 +123,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_FREEZE		23	/* is freezing for suspend */
 #define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
 #define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
+#define TIF_DS_AREA_MSR	25      /* uses thread_struct.ds_area_msr */
+#define TIF_BTS_TRACE_TS	26      /* record scheduling event timestamps */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -142,6 +144,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_FREEZE		(1<<TIF_FREEZE)
 #define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
 #define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
+#define _TIF_DS_AREA_MSR	(1<<TIF_DS_AREA_MSR)
+#define _TIF_BTS_TRACE_TS	(1<<TIF_BTS_TRACE_TS)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -153,7 +157,10 @@ static inline struct thread_info *stack_thread_info(void)
 	(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY|_TIF_HRTICK_RESCHED)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR)
+#define _TIF_WORK_CTXSW \
+    (_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_BTS_TRACE_TS)
+#define _TIF_WORK_CTXSW_PREV _TIF_WORK_CTXSW
+#define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG)
 
 #define PREEMPT_ACTIVE     0x10000000
 
-- 
cgit v1.2.3


From cef189508b2f043fdc2e354645803b6ce004aac3 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x86: put get_kernel_rpl in a common location

This macro is useful for both i386 and x86_64, so put it in a common
location, where both arches can grab it.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/segment.h    | 9 +++++++++
 include/asm-x86/segment_32.h | 3 ---
 2 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index 605068280e28..b3a7a3e9fb19 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -1,5 +1,14 @@
+#ifndef _ASM_X86_SEGMENT_H_
+#define _ASM_X86_SEGMENT_H_
+
 #ifdef CONFIG_X86_32
 # include "segment_32.h"
 #else
 # include "segment_64.h"
 #endif
+
+#ifndef CONFIG_PARAVIRT
+#define get_kernel_rpl()  0
+#endif
+
+#endif
diff --git a/include/asm-x86/segment_32.h b/include/asm-x86/segment_32.h
index 597a47c2515f..ff186e34798a 100644
--- a/include/asm-x86/segment_32.h
+++ b/include/asm-x86/segment_32.h
@@ -129,9 +129,6 @@
 #define SEGMENT_LDT		0x4
 #define SEGMENT_GDT		0x0
 
-#ifndef CONFIG_PARAVIRT
-#define get_kernel_rpl()  0
-#endif
 /*
  * Matching rules for certain types of segments.
  */
-- 
cgit v1.2.3


From 15ee09b40831ae88eb8a804a647b6935eb8b7114 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x86: unify common parts of segment.h

Although segment handling in i386 and x86_64 are very different,
there's a common part. Put them in segment.h instead of arch specific
headers

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/segment.h    | 16 ++++++++++++++++
 include/asm-x86/segment_32.h | 22 ----------------------
 include/asm-x86/segment_64.h | 17 -----------------
 3 files changed, 16 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index b3a7a3e9fb19..3c9b306bc4bf 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -11,4 +11,20 @@
 #define get_kernel_rpl()  0
 #endif
 
+/* User mode is privilege level 3 */
+#define USER_RPL		0x3
+/* LDT segment has TI set, GDT has it cleared */
+#define SEGMENT_LDT		0x4
+#define SEGMENT_GDT		0x0
+
+/* Bottom two bits of selector give the ring privilege level */
+#define SEGMENT_RPL_MASK	0x3
+/* Bit 2 is table indicator (LDT/GDT) */
+#define SEGMENT_TI_MASK		0x4
+
+#define IDT_ENTRIES 256
+#define GDT_SIZE (GDT_ENTRIES * 8)
+#define GDT_ENTRY_TLS_ENTRIES 3
+#define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8)
+
 #endif
diff --git a/include/asm-x86/segment_32.h b/include/asm-x86/segment_32.h
index ff186e34798a..3e133ad4252d 100644
--- a/include/asm-x86/segment_32.h
+++ b/include/asm-x86/segment_32.h
@@ -45,12 +45,9 @@
  *  30 - unused
  *  31 - TSS for double fault handler
  */
-#define GDT_ENTRY_TLS_ENTRIES	3
 #define GDT_ENTRY_TLS_MIN	6
 #define GDT_ENTRY_TLS_MAX 	(GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
 
-#define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8)
-
 #define GDT_ENTRY_DEFAULT_USER_CS	14
 #define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS * 8 + 3)
 
@@ -87,7 +84,6 @@
  * The GDT has 32 entries
  */
 #define GDT_ENTRIES 32
-#define GDT_SIZE (GDT_ENTRIES * 8)
 
 /* Simple and small GDT entries for booting only */
 
@@ -111,24 +107,6 @@
 #define PNP_TS1    (GDT_ENTRY_PNPBIOS_TS1 * 8)	/* transfer data segment */
 #define PNP_TS2    (GDT_ENTRY_PNPBIOS_TS2 * 8)	/* another data segment */
 
-/*
- * The interrupt descriptor table has room for 256 idt's,
- * the global descriptor table is dependent on the number
- * of tasks we can have..
- */
-#define IDT_ENTRIES 256
-
-/* Bottom two bits of selector give the ring privilege level */
-#define SEGMENT_RPL_MASK	0x3
-/* Bit 2 is table indicator (LDT/GDT) */
-#define SEGMENT_TI_MASK		0x4
-
-/* User mode is privilege level 3 */
-#define USER_RPL		0x3
-/* LDT segment has TI set, GDT has it cleared */
-#define SEGMENT_LDT		0x4
-#define SEGMENT_GDT		0x0
-
 /*
  * Matching rules for certain types of segments.
  */
diff --git a/include/asm-x86/segment_64.h b/include/asm-x86/segment_64.h
index dce742101348..5cc7deb8b034 100644
--- a/include/asm-x86/segment_64.h
+++ b/include/asm-x86/segment_64.h
@@ -33,8 +33,6 @@
 #define GDT_ENTRY_TLS_MIN 12
 #define GDT_ENTRY_TLS_MAX 14
 
-#define GDT_ENTRY_TLS_ENTRIES 3
-
 #define GDT_ENTRY_PER_CPU 15	/* Abused to load per CPU data from limit */
 #define __PER_CPU_SEG	(GDT_ENTRY_PER_CPU * 8 + 3)
 
@@ -45,20 +43,5 @@
 #define GS_TLS_SEL ((GDT_ENTRY_TLS_MIN+GS_TLS)*8 + 3)
 #define FS_TLS_SEL ((GDT_ENTRY_TLS_MIN+FS_TLS)*8 + 3)
 
-#define IDT_ENTRIES 256
 #define GDT_ENTRIES 16
-#define GDT_SIZE (GDT_ENTRIES * 8)
-#define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8) 
-
-/* Bottom two bits of selector give the ring privilege level */
-#define SEGMENT_RPL_MASK	0x3
-/* Bit 2 is table indicator (LDT/GDT) */
-#define SEGMENT_TI_MASK		0x4
-
-/* User mode is privilege level 3 */
-#define USER_RPL		0x3
-/* LDT segment has TI set, GDT has it cleared */
-#define SEGMENT_LDT		0x4
-#define SEGMENT_GDT		0x0
-
 #endif
-- 
cgit v1.2.3


From b23be399da88a5008b1db4db06f03146b25cdc52 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x86: remove arch specific segment headers

This file puts the remainder of the arch specificic segment
headers in segment.h.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/segment.h    | 177 ++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/segment_32.h | 122 -----------------------------
 include/asm-x86/segment_64.h |  47 ------------
 3 files changed, 175 insertions(+), 171 deletions(-)
 delete mode 100644 include/asm-x86/segment_64.h

(limited to 'include')

diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index 3c9b306bc4bf..c7509df79141 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -2,9 +2,182 @@
 #define _ASM_X86_SEGMENT_H_
 
 #ifdef CONFIG_X86_32
-# include "segment_32.h"
+/*
+ * The layout of the per-CPU GDT under Linux:
+ *
+ *   0 - null
+ *   1 - reserved
+ *   2 - reserved
+ *   3 - reserved
+ *
+ *   4 - unused			<==== new cacheline
+ *   5 - unused
+ *
+ *  ------- start of TLS (Thread-Local Storage) segments:
+ *
+ *   6 - TLS segment #1			[ glibc's TLS segment ]
+ *   7 - TLS segment #2			[ Wine's %fs Win32 segment ]
+ *   8 - TLS segment #3
+ *   9 - reserved
+ *  10 - reserved
+ *  11 - reserved
+ *
+ *  ------- start of kernel segments:
+ *
+ *  12 - kernel code segment		<==== new cacheline
+ *  13 - kernel data segment
+ *  14 - default user CS
+ *  15 - default user DS
+ *  16 - TSS
+ *  17 - LDT
+ *  18 - PNPBIOS support (16->32 gate)
+ *  19 - PNPBIOS support
+ *  20 - PNPBIOS support
+ *  21 - PNPBIOS support
+ *  22 - PNPBIOS support
+ *  23 - APM BIOS support
+ *  24 - APM BIOS support
+ *  25 - APM BIOS support
+ *
+ *  26 - ESPFIX small SS
+ *  27 - per-cpu			[ offset to per-cpu data area ]
+ *  28 - unused
+ *  29 - unused
+ *  30 - unused
+ *  31 - TSS for double fault handler
+ */
+#define GDT_ENTRY_TLS_MIN	6
+#define GDT_ENTRY_TLS_MAX 	(GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
+
+#define GDT_ENTRY_DEFAULT_USER_CS	14
+#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS * 8 + 3)
+
+#define GDT_ENTRY_DEFAULT_USER_DS	15
+#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS * 8 + 3)
+
+#define GDT_ENTRY_KERNEL_BASE	12
+
+#define GDT_ENTRY_KERNEL_CS		(GDT_ENTRY_KERNEL_BASE + 0)
+#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8)
+
+#define GDT_ENTRY_KERNEL_DS		(GDT_ENTRY_KERNEL_BASE + 1)
+#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8)
+
+#define GDT_ENTRY_TSS			(GDT_ENTRY_KERNEL_BASE + 4)
+#define GDT_ENTRY_LDT			(GDT_ENTRY_KERNEL_BASE + 5)
+
+#define GDT_ENTRY_PNPBIOS_BASE		(GDT_ENTRY_KERNEL_BASE + 6)
+#define GDT_ENTRY_APMBIOS_BASE		(GDT_ENTRY_KERNEL_BASE + 11)
+
+#define GDT_ENTRY_ESPFIX_SS		(GDT_ENTRY_KERNEL_BASE + 14)
+#define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)
+
+#define GDT_ENTRY_PERCPU			(GDT_ENTRY_KERNEL_BASE + 15)
+#ifdef CONFIG_SMP
+#define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8)
+#else
+#define __KERNEL_PERCPU 0
+#endif
+
+#define GDT_ENTRY_DOUBLEFAULT_TSS	31
+
+/*
+ * The GDT has 32 entries
+ */
+#define GDT_ENTRIES 32
+
+/* Simple and small GDT entries for booting only */
+
+#define GDT_ENTRY_BOOT_CS		2
+#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
+
+#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
+#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
+
+/* The PnP BIOS entries in the GDT */
+#define GDT_ENTRY_PNPBIOS_CS32		(GDT_ENTRY_PNPBIOS_BASE + 0)
+#define GDT_ENTRY_PNPBIOS_CS16		(GDT_ENTRY_PNPBIOS_BASE + 1)
+#define GDT_ENTRY_PNPBIOS_DS		(GDT_ENTRY_PNPBIOS_BASE + 2)
+#define GDT_ENTRY_PNPBIOS_TS1		(GDT_ENTRY_PNPBIOS_BASE + 3)
+#define GDT_ENTRY_PNPBIOS_TS2		(GDT_ENTRY_PNPBIOS_BASE + 4)
+
+/* The PnP BIOS selectors */
+#define PNP_CS32   (GDT_ENTRY_PNPBIOS_CS32 * 8)	/* segment for calling fn */
+#define PNP_CS16   (GDT_ENTRY_PNPBIOS_CS16 * 8)	/* code segment for BIOS */
+#define PNP_DS     (GDT_ENTRY_PNPBIOS_DS * 8)	/* data segment for BIOS */
+#define PNP_TS1    (GDT_ENTRY_PNPBIOS_TS1 * 8)	/* transfer data segment */
+#define PNP_TS2    (GDT_ENTRY_PNPBIOS_TS2 * 8)	/* another data segment */
+
+/* Bottom two bits of selector give the ring privilege level */
+#define SEGMENT_RPL_MASK	0x3
+/* Bit 2 is table indicator (LDT/GDT) */
+#define SEGMENT_TI_MASK		0x4
+
+/* User mode is privilege level 3 */
+#define USER_RPL		0x3
+/* LDT segment has TI set, GDT has it cleared */
+#define SEGMENT_LDT		0x4
+#define SEGMENT_GDT		0x0
+
+/*
+ * Matching rules for certain types of segments.
+ */
+
+/* Matches only __KERNEL_CS, ignoring PnP / USER / APM segments */
+#define SEGMENT_IS_KERNEL_CODE(x) (((x) & 0xfc) == GDT_ENTRY_KERNEL_CS * 8)
+
+/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */
+#define SEGMENT_IS_FLAT_CODE(x)  (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8)
+
+/* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
+#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
+
+
 #else
-# include "segment_64.h"
+#include <asm/cache.h>
+
+/* Simple and small GDT entries for booting only */
+
+#define GDT_ENTRY_BOOT_CS		2
+#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
+
+#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
+#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
+
+#define __KERNEL_CS	0x10
+#define __KERNEL_DS	0x18
+
+#define __KERNEL32_CS   0x08
+
+/*
+ * we cannot use the same code segment descriptor for user and kernel
+ * -- not even in the long flat mode, because of different DPL /kkeil
+ * The segment offset needs to contain a RPL. Grr. -AK
+ * GDT layout to get 64bit syscall right (sysret hardcodes gdt offsets)
+ */
+
+#define __USER32_CS   0x23   /* 4*8+3 */
+#define __USER_DS     0x2b   /* 5*8+3 */
+#define __USER_CS     0x33   /* 6*8+3 */
+#define __USER32_DS	__USER_DS
+
+#define GDT_ENTRY_TSS 8	/* needs two entries */
+#define GDT_ENTRY_LDT 10 /* needs two entries */
+#define GDT_ENTRY_TLS_MIN 12
+#define GDT_ENTRY_TLS_MAX 14
+
+#define GDT_ENTRY_PER_CPU 15	/* Abused to load per CPU data from limit */
+#define __PER_CPU_SEG	(GDT_ENTRY_PER_CPU * 8 + 3)
+
+/* TLS indexes for 64bit - hardcoded in arch_prctl */
+#define FS_TLS 0
+#define GS_TLS 1
+
+#define GS_TLS_SEL ((GDT_ENTRY_TLS_MIN+GS_TLS)*8 + 3)
+#define FS_TLS_SEL ((GDT_ENTRY_TLS_MIN+FS_TLS)*8 + 3)
+
+#define GDT_ENTRIES 16
+
 #endif
 
 #ifndef CONFIG_PARAVIRT
diff --git a/include/asm-x86/segment_32.h b/include/asm-x86/segment_32.h
index 3e133ad4252d..8b137891791f 100644
--- a/include/asm-x86/segment_32.h
+++ b/include/asm-x86/segment_32.h
@@ -1,123 +1 @@
-#ifndef _ASM_SEGMENT_H
-#define _ASM_SEGMENT_H
 
-/*
- * The layout of the per-CPU GDT under Linux:
- *
- *   0 - null
- *   1 - reserved
- *   2 - reserved
- *   3 - reserved
- *
- *   4 - unused			<==== new cacheline
- *   5 - unused
- *
- *  ------- start of TLS (Thread-Local Storage) segments:
- *
- *   6 - TLS segment #1			[ glibc's TLS segment ]
- *   7 - TLS segment #2			[ Wine's %fs Win32 segment ]
- *   8 - TLS segment #3
- *   9 - reserved
- *  10 - reserved
- *  11 - reserved
- *
- *  ------- start of kernel segments:
- *
- *  12 - kernel code segment		<==== new cacheline
- *  13 - kernel data segment
- *  14 - default user CS
- *  15 - default user DS
- *  16 - TSS
- *  17 - LDT
- *  18 - PNPBIOS support (16->32 gate)
- *  19 - PNPBIOS support
- *  20 - PNPBIOS support
- *  21 - PNPBIOS support
- *  22 - PNPBIOS support
- *  23 - APM BIOS support
- *  24 - APM BIOS support
- *  25 - APM BIOS support 
- *
- *  26 - ESPFIX small SS
- *  27 - per-cpu			[ offset to per-cpu data area ]
- *  28 - unused
- *  29 - unused
- *  30 - unused
- *  31 - TSS for double fault handler
- */
-#define GDT_ENTRY_TLS_MIN	6
-#define GDT_ENTRY_TLS_MAX 	(GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
-
-#define GDT_ENTRY_DEFAULT_USER_CS	14
-#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS * 8 + 3)
-
-#define GDT_ENTRY_DEFAULT_USER_DS	15
-#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS * 8 + 3)
-
-#define GDT_ENTRY_KERNEL_BASE	12
-
-#define GDT_ENTRY_KERNEL_CS		(GDT_ENTRY_KERNEL_BASE + 0)
-#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8)
-
-#define GDT_ENTRY_KERNEL_DS		(GDT_ENTRY_KERNEL_BASE + 1)
-#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8)
-
-#define GDT_ENTRY_TSS			(GDT_ENTRY_KERNEL_BASE + 4)
-#define GDT_ENTRY_LDT			(GDT_ENTRY_KERNEL_BASE + 5)
-
-#define GDT_ENTRY_PNPBIOS_BASE		(GDT_ENTRY_KERNEL_BASE + 6)
-#define GDT_ENTRY_APMBIOS_BASE		(GDT_ENTRY_KERNEL_BASE + 11)
-
-#define GDT_ENTRY_ESPFIX_SS		(GDT_ENTRY_KERNEL_BASE + 14)
-#define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)
-
-#define GDT_ENTRY_PERCPU			(GDT_ENTRY_KERNEL_BASE + 15)
-#ifdef CONFIG_SMP
-#define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8)
-#else
-#define __KERNEL_PERCPU 0
-#endif
-
-#define GDT_ENTRY_DOUBLEFAULT_TSS	31
-
-/*
- * The GDT has 32 entries
- */
-#define GDT_ENTRIES 32
-
-/* Simple and small GDT entries for booting only */
-
-#define GDT_ENTRY_BOOT_CS		2
-#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
-
-#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
-#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
-
-/* The PnP BIOS entries in the GDT */
-#define GDT_ENTRY_PNPBIOS_CS32		(GDT_ENTRY_PNPBIOS_BASE + 0)
-#define GDT_ENTRY_PNPBIOS_CS16		(GDT_ENTRY_PNPBIOS_BASE + 1)
-#define GDT_ENTRY_PNPBIOS_DS		(GDT_ENTRY_PNPBIOS_BASE + 2)
-#define GDT_ENTRY_PNPBIOS_TS1		(GDT_ENTRY_PNPBIOS_BASE + 3)
-#define GDT_ENTRY_PNPBIOS_TS2		(GDT_ENTRY_PNPBIOS_BASE + 4)
-
-/* The PnP BIOS selectors */
-#define PNP_CS32   (GDT_ENTRY_PNPBIOS_CS32 * 8)	/* segment for calling fn */
-#define PNP_CS16   (GDT_ENTRY_PNPBIOS_CS16 * 8)	/* code segment for BIOS */
-#define PNP_DS     (GDT_ENTRY_PNPBIOS_DS * 8)	/* data segment for BIOS */
-#define PNP_TS1    (GDT_ENTRY_PNPBIOS_TS1 * 8)	/* transfer data segment */
-#define PNP_TS2    (GDT_ENTRY_PNPBIOS_TS2 * 8)	/* another data segment */
-
-/*
- * Matching rules for certain types of segments.
- */
-
-/* Matches only __KERNEL_CS, ignoring PnP / USER / APM segments */
-#define SEGMENT_IS_KERNEL_CODE(x) (((x) & 0xfc) == GDT_ENTRY_KERNEL_CS * 8)
-
-/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */
-#define SEGMENT_IS_FLAT_CODE(x)  (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8)
-
-/* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
-#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
-
-#endif
diff --git a/include/asm-x86/segment_64.h b/include/asm-x86/segment_64.h
deleted file mode 100644
index 5cc7deb8b034..000000000000
--- a/include/asm-x86/segment_64.h
+++ /dev/null
@@ -1,47 +0,0 @@
-#ifndef _ASM_SEGMENT_H
-#define _ASM_SEGMENT_H
-
-#include <asm/cache.h>
-
-/* Simple and small GDT entries for booting only */
-
-#define GDT_ENTRY_BOOT_CS		2
-#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
-
-#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
-#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
-
-#define __KERNEL_CS	0x10
-#define __KERNEL_DS	0x18
-
-#define __KERNEL32_CS   0x08
-
-/* 
- * we cannot use the same code segment descriptor for user and kernel
- * -- not even in the long flat mode, because of different DPL /kkeil 
- * The segment offset needs to contain a RPL. Grr. -AK
- * GDT layout to get 64bit syscall right (sysret hardcodes gdt offsets) 
- */
-
-#define __USER32_CS   0x23   /* 4*8+3 */ 
-#define __USER_DS     0x2b   /* 5*8+3 */ 
-#define __USER_CS     0x33   /* 6*8+3 */ 
-#define __USER32_DS	__USER_DS 
-
-#define GDT_ENTRY_TSS 8	/* needs two entries */
-#define GDT_ENTRY_LDT 10 /* needs two entries */
-#define GDT_ENTRY_TLS_MIN 12
-#define GDT_ENTRY_TLS_MAX 14
-
-#define GDT_ENTRY_PER_CPU 15	/* Abused to load per CPU data from limit */
-#define __PER_CPU_SEG	(GDT_ENTRY_PER_CPU * 8 + 3)
-
-/* TLS indexes for 64bit - hardcoded in arch_prctl */
-#define FS_TLS 0	
-#define GS_TLS 1	
-
-#define GS_TLS_SEL ((GDT_ENTRY_TLS_MIN+GS_TLS)*8 + 3)
-#define FS_TLS_SEL ((GDT_ENTRY_TLS_MIN+FS_TLS)*8 + 3)
-
-#define GDT_ENTRIES 16
-#endif
-- 
cgit v1.2.3


From 6b4b05bd790389962e6fcfc862562e7fa239c9d2 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x64/page.h: convert some macros to inlines

Convert clear_page/copy_page macros to inline functions for type-checking.
Andrew wants to extirpate these ugly macros. (Ingo too. Thomas as well.
Please send us more "kill ugly macros" patches! :-)

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page_32.h | 39 +++++++++++++++++++++++++++++++++------
 include/asm-x86/page_64.h | 21 ++++++++++++++++-----
 2 files changed, 49 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 80ecc66b6d86..c620c934e557 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -12,12 +12,21 @@
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
+#include <linux/string.h>
+
 #ifdef CONFIG_X86_USE_3DNOW
 
 #include <asm/mmx.h>
 
-#define clear_page(page)	mmx_clear_page((void *)(page))
-#define copy_page(to,from)	mmx_copy_page(to,from)
+static inline void clear_page(void *page)
+{
+	mmx_clear_page(page);
+}
+
+static inline void copy_page(void *to, void *from)
+{
+	mmx_copy_page(to, from);
+}
 
 #else
 
@@ -26,13 +35,31 @@
  *	Maybe the K6-III ?
  */
  
-#define clear_page(page)	memset((void *)(page), 0, PAGE_SIZE)
-#define copy_page(to,from)	memcpy((void *)(to), (void *)(from), PAGE_SIZE)
+static inline void clear_page(void *page)
+{
+	memset(page, 0, PAGE_SIZE);
+}
+
+static inline void copy_page(void *to, void *from)
+{
+	memcpy(to, from, PAGE_SIZE);
+}
 
 #endif
 
-#define clear_user_page(page, vaddr, pg)	clear_page(page)
-#define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+struct page;
+
+static void inline clear_user_page(void *page, unsigned long vaddr,
+				struct page *pg)
+{
+	clear_page(page);
+}
+
+static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
+				struct page *topage)
+{
+	copy_page(to, from);
+}
 
 #define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \
 	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index d400167c5509..068b5e4cea4c 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -43,11 +43,22 @@
 extern unsigned long end_pfn;
 extern unsigned long end_pfn_map;
 
-void clear_page(void *);
-void copy_page(void *, void *);
-
-#define clear_user_page(page, vaddr, pg)	clear_page(page)
-#define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+void clear_page(void *page);
+void copy_page(void *to, void *from);
+
+struct page;
+
+static void inline clear_user_page(void *page, unsigned long vaddr,
+				struct page *pg)
+{
+	clear_page(page);
+}
+
+static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
+				struct page *topage)
+{
+	copy_page(to, from);
+}
 
 #define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \
 	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
-- 
cgit v1.2.3


From 27ec161f93b8f018e0270a5383ea1d1c36f00c8a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:31:09 +0100
Subject: x86: kill mk_pte_huge

It only has a single use, which can be trivially replaced.

Signed-off-by: Jeremy Fitzhardinge <Jeremy.Fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_64.c        | 3 +--
 include/asm-x86/pgtable_64.h | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 5cadbb40da3e..15e05a004fcf 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -742,8 +742,7 @@ int __meminit vmemmap_populate(struct page *start_page,
 			if (!p)
 				return -ENOMEM;
 
-			entry = pfn_pte(__pa(p) >> PAGE_SHIFT, PAGE_KERNEL);
-			mk_pte_huge(entry);
+			entry = pfn_pte(__pa(p) >> PAGE_SHIFT, PAGE_KERNEL_LARGE);
 			set_pmd(pmd, __pmd(pte_val(entry)));
 
 			printk(KERN_DEBUG " [%lx-%lx] PMD ->%p on node %d\n",
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 3072619365a8..5c252d9fee1b 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -355,7 +355,6 @@ static inline int pmd_large(pmd_t pte) {
 
 /* page, protection -> pte */
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
-#define mk_pte_huge(entry) (pte_val(entry) |= _PAGE_PRESENT | _PAGE_PSE)
  
 /* Change flags of a PTE */
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
-- 
cgit v1.2.3


From 2355188570790930718fb72444cddc2959039d9d Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:31:10 +0100
Subject: x86: avoid build warning

fix this build warning:

 include/asm/topology_32.h: In function 'node_to_first_cpu':
 include/asm/topology_32.h:66: warning: unused variable 'mask'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/cpumask.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 85bd790c201e..7047f58306a7 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -218,8 +218,8 @@ int __first_cpu(const cpumask_t *srcp);
 int __next_cpu(int n, const cpumask_t *srcp);
 #define next_cpu(n, src) __next_cpu((n), &(src))
 #else
-#define first_cpu(src)		0
-#define next_cpu(n, src)	1
+#define first_cpu(src)		({ (void)(src); 0; })
+#define next_cpu(n, src)	({ (void)(src); 1; })
 #endif
 
 #define cpumask_of_cpu(cpu)						\
-- 
cgit v1.2.3


From ee238e5ca66858f80170f87724f84d67183b069a Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:10 +0100
Subject: x86: prepare time related functions for paravirt

This patch add provisions for time related functions so they
can be later replaced by paravirt versions.

it basically encloses {g,s}et_wallclock inside the
already existent functions update_persistent_clock and
read_persistent_clock, and defines {s,g}et_wallclock
to the core of such functions.

it also allow for a later-on-game time initialization, as done
by i386. Paravirt guests can set a function to do their own
initialization this way.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/time_64.c | 12 +++++++++---
 include/asm-x86/time.h    | 26 +++++++++++++++++++++-----
 2 files changed, 30 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index bf0bcc9bb001..91d4d495904e 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -21,6 +21,8 @@
 #include <asm/hpet.h>
 #include <asm/nmi.h>
 #include <asm/vgtod.h>
+#include <asm/time.h>
+#include <asm/timer.h>
 
 volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
 
@@ -54,7 +56,7 @@ static irqreturn_t timer_event_interrupt(int irq, void *dev_id)
 /* calibrate_cpu is used on systems with fixed rate TSCs to determine
  * processor frequency */
 #define TICK_COUNT 100000000
-static unsigned int __init tsc_calibrate_cpu_khz(void)
+unsigned long __init native_calculate_cpu_khz(void)
 {
 	int tsc_start, tsc_now;
 	int i, no_ctr_free;
@@ -104,20 +106,23 @@ static struct irqaction irq0 = {
 	.name		= "timer"
 };
 
-void __init time_init(void)
+void __init hpet_time_init(void)
 {
 	if (!hpet_enable())
 		setup_pit_timer();
 
 	setup_irq(0, &irq0);
+}
 
+void __init time_init(void)
+{
 	tsc_calibrate();
 
 	cpu_khz = tsc_khz;
 	if (cpu_has(&boot_cpu_data, X86_FEATURE_CONSTANT_TSC) &&
 		boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
 		boot_cpu_data.x86 == 16)
-		cpu_khz = tsc_calibrate_cpu_khz();
+		cpu_khz = calculate_cpu_khz();
 
 	if (unsynchronized_tsc())
 		mark_tsc_unstable("TSCs unsynchronized");
@@ -130,4 +135,5 @@ void __init time_init(void)
 	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
 		cpu_khz / 1000, cpu_khz % 1000);
 	init_tsc_clocksource();
+	late_time_init = choose_time_init();
 }
diff --git a/include/asm-x86/time.h b/include/asm-x86/time.h
index b3f94cd81ac6..68779b048a3e 100644
--- a/include/asm-x86/time.h
+++ b/include/asm-x86/time.h
@@ -1,8 +1,12 @@
-#ifndef _ASMi386_TIME_H
-#define _ASMi386_TIME_H
+#ifndef _ASMX86_TIME_H
+#define _ASMX86_TIME_H
+
+extern void (*late_time_init)(void);
+extern void hpet_time_init(void);
 
-#include <linux/efi.h>
 #include <asm/mc146818rtc.h>
+#ifdef CONFIG_X86_32
+#include <linux/efi.h>
 
 static inline unsigned long native_get_wallclock(void)
 {
@@ -28,8 +32,20 @@ static inline int native_set_wallclock(unsigned long nowtime)
 	return retval;
 }
 
-extern void (*late_time_init)(void);
-extern void hpet_time_init(void);
+#else
+extern void native_time_init_hook(void);
+
+static inline unsigned long native_get_wallclock(void)
+{
+	return mach_get_cmos_time();
+}
+
+static inline int native_set_wallclock(unsigned long nowtime)
+{
+	return mach_set_rtc_mmss(nowtime);
+}
+
+#endif
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
-- 
cgit v1.2.3


From ba082427ae6ffbf8e48a26ae4f72f4501a6b80c1 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:10 +0100
Subject: x86: tweak io_64.h for paravirt.

We need something here because we can't call in and out instructions
directly. However, we have to be careful, because no indirections are
allowed in misc_64.c , and paravirt_ops is a kind of one. So just
call it directly there

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/boot/compressed/misc_64.c |  6 ++++++
 include/asm-x86/io_64.h            | 30 +++++++++++++++++++++++-------
 2 files changed, 29 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/boot/compressed/misc_64.c b/arch/x86/boot/compressed/misc_64.c
index 43e5fcc37be9..7d8338e21b00 100644
--- a/arch/x86/boot/compressed/misc_64.c
+++ b/arch/x86/boot/compressed/misc_64.c
@@ -9,6 +9,12 @@
  * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996
  */
 
+/*
+ * we have to be careful, because no indirections are allowed here, and
+ * paravirt_ops is a kind of one. As it will only run in baremetal anyway,
+ * we just keep it from happening
+ */
+#undef CONFIG_PARAVIRT
 #define _LINUX_STRING_H_ 1
 #define __LINUX_BITMAP_H 1
 
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index dbcc03aa1c6a..fef0ce2ced81 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -40,6 +40,10 @@ extern void native_io_delay(void);
 extern int io_delay_type;
 extern void io_delay_init(void);
 
+#if defined(CONFIG_PARAVIRT)
+#include <asm/paravirt.h>
+#else
+
 static inline void slow_down_io(void)
 {
 	native_io_delay();
@@ -49,6 +53,7 @@ static inline void slow_down_io(void)
 	native_io_delay();
 #endif
 }
+#endif
 
 /*
  * Talk about misusing macros..
@@ -57,21 +62,32 @@ static inline void slow_down_io(void)
 static inline void out##s(unsigned x value, unsigned short port) {
 
 #define __OUT2(s,s1,s2) \
-__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1" : : "a" (value), "Nd" (port))
+__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1"
+
+#ifndef REALLY_SLOW_IO
+#define REALLY_SLOW_IO
+#define UNSET_REALLY_SLOW_IO
+#endif
 
 #define __OUT(s,s1,x) \
-__OUT1(s,x) __OUT2(s,s1,"w"); } \
-__OUT1(s##_p,x) __OUT2(s,s1,"w"); slow_down_io(); }
+__OUT1(s,x) __OUT2(s,s1,"w") : : "a" (value), "Nd" (port)); } \
+__OUT1(s##_p, x) __OUT2(s, s1, "w") : : "a" (value), "Nd" (port)); \
+		slow_down_io(); }
 
 #define __IN1(s) \
 static inline RETURN_TYPE in##s(unsigned short port) { RETURN_TYPE _v;
 
 #define __IN2(s,s1,s2) \
-__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0" : "=a" (_v) : "Nd" (port))
+__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0"
+
+#define __IN(s,s1,i...) \
+__IN1(s) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i); return _v; } \
+__IN1(s##_p) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i);	  \
+				slow_down_io(); return _v; }
 
-#define __IN(s,s1) \
-__IN1(s) __IN2(s,s1,"w"); return _v; } \
-__IN1(s##_p) __IN2(s,s1,"w"); slow_down_io(); return _v; }
+#ifdef UNSET_REALLY_SLOW_IO
+#undef REALLY_SLOW_IO
+#endif
 
 #define __INS(s) \
 static inline void ins##s(unsigned short port, void * addr, unsigned long count) \
-- 
cgit v1.2.3


From 746ef0cd0c7190d570c65b8e39a4ac67550ae43a Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:11 +0100
Subject: x86: prepare 64-bit architecture initialization for paravirt

This patch prepares the x86_64 architecture initialization for
paravirt. It requires a memory initialization step, which is done
by implementing 64-bit version for machine_specific_memory_setup,
and putting an ARCH_SETUP hook, for guest-dependent initialization.
This last step is done akin to i386

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_64.c    |  9 +++++++--
 arch/x86/kernel/setup_64.c   | 28 +++++++++++++++++++++++++++-
 arch/x86/kernel/smpboot_64.c |  4 ++--
 include/asm-x86/setup.h      | 11 ++++++++---
 4 files changed, 44 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 8e7321101a0a..abc473bcabe8 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -638,8 +638,10 @@ static void early_panic(char *msg)
 	panic(msg);
 }
 
-void __init setup_memory_region(void)
+/* We're not void only for x86 32-bit compat */
+char * __init machine_specific_memory_setup(void)
 {
+	char *who = "BIOS-e820";
 	/*
 	 * Try to copy the BIOS-supplied E820-map.
 	 *
@@ -650,7 +652,10 @@ void __init setup_memory_region(void)
 	if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0)
 		early_panic("Cannot find a valid memory map");
 	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-	e820_print_map("BIOS-e820");
+	e820_print_map(who);
+
+	/* In case someone cares... */
+	return who;
 }
 
 static int __init parse_memopt(char *p)
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index f2b131ef844e..8dd110d93e73 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -39,6 +39,7 @@
 #include <linux/dmi.h>
 #include <linux/dma-mapping.h>
 #include <linux/ctype.h>
+#include <linux/uaccess.h>
 
 #include <asm/mtrr.h>
 #include <asm/uaccess.h>
@@ -62,6 +63,12 @@
 #include <asm/mce.h>
 #include <asm/ds.h>
 
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define ARCH_SETUP
+#endif
+
 /*
  * Machine setup..
  */
@@ -246,6 +253,16 @@ static void discover_ebda(void)
 	 * 4K EBDA area at 0x40E
 	 */
 	ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER);
+	/*
+	 * There can be some situations, like paravirtualized guests,
+	 * in which there is no available ebda information. In such
+	 * case, just skip it
+	 */
+	if (!ebda_addr) {
+		ebda_size = 0;
+		return;
+	}
+
 	ebda_addr <<= 4;
 
 	ebda_size = *(unsigned short *)__va(ebda_addr);
@@ -259,6 +276,12 @@ static void discover_ebda(void)
 		ebda_size = 64*1024;
 }
 
+/* Overridden in paravirt.c if CONFIG_PARAVIRT */
+void __attribute__((weak)) memory_setup(void)
+{
+       machine_specific_memory_setup();
+}
+
 void __init setup_arch(char **cmdline_p)
 {
 	unsigned i;
@@ -276,7 +299,10 @@ void __init setup_arch(char **cmdline_p)
 	rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
 	rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
 #endif
-	setup_memory_region();
+
+	ARCH_SETUP
+
+	memory_setup();
 	copy_edd();
 
 	if (!boot_params.hdr.root_flags)
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index c3f2736ba530..cb73c4da87fc 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -369,7 +369,7 @@ void __cpuinit start_secondary(void)
 
 	unlock_ipi_call_lock();
 
-	setup_secondary_APIC_clock();
+	setup_secondary_clock();
 
 	cpu_idle();
 }
@@ -923,7 +923,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	 * Set up local APIC timer on boot CPU.
 	 */
 
-	setup_boot_APIC_clock();
+	setup_boot_clock();
 }
 
 /*
diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h
index 24d786e07b49..071e054abd82 100644
--- a/include/asm-x86/setup.h
+++ b/include/asm-x86/setup.h
@@ -3,6 +3,13 @@
 
 #define COMMAND_LINE_SIZE 2048
 
+#ifndef __ASSEMBLY__
+char *machine_specific_memory_setup(void);
+#ifndef CONFIG_PARAVIRT
+#define paravirt_post_allocator_init()	do {} while (0)
+#endif
+#endif /* __ASSEMBLY__ */
+
 #ifdef __KERNEL__
 
 #ifdef __i386__
@@ -51,9 +58,7 @@ void __init add_memory_region(unsigned long long start,
 
 extern unsigned long init_pg_tables_end;
 
-#ifndef CONFIG_PARAVIRT
-#define paravirt_post_allocator_init()	do {} while (0)
-#endif
+
 
 #endif /* __i386__ */
 #endif /* _SETUP */
-- 
cgit v1.2.3


From 6842ef0e85a9cc1295f3ef933a230f863b01eb0f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:11 +0100
Subject: x86: unify desc_struct

This patch aims to make the access of struct desc_struct variables
equal across architectures. In this patch, I unify the i386 and x86_64
versions under an anonymous union, keeping the way they are accessed
untouched (a and b for 32-bit code, individual bit-fields for 64-bit).

This solution is not beautiful, but will allow us to integrate common
code that differed by the way descriptors were used. This is to be viewed
incrementally. There's simply too much code to be fixed at once.

In the future, goal is to set up in a single way of acessing
the desc_struct fields.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apm_32.c       |  2 +-
 arch/x86/kernel/cpu/common.c   | 35 +++++++++++++++++++++--------------
 arch/x86/kernel/process_64.c   |  2 +-
 arch/x86/kernel/traps_32.c     |  3 ++-
 include/asm-x86/desc_defs.h    | 25 +++++++++++++++++--------
 include/asm-x86/lguest.h       |  4 ++--
 include/asm-x86/processor_32.h |  5 +----
 7 files changed, 45 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 7496c2e4b6ae..e32f6c37db9b 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -405,7 +405,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
 static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
 static struct apm_user *	user_list;
 static DEFINE_SPINLOCK(user_list_lock);
-static const struct desc_struct	bad_bios_desc = { 0, 0x00409200 };
+static const struct desc_struct	bad_bios_desc = { { { 0, 0x00409200 } } };
 
 static const char		driver_version[] = "1.16ac";	/* no spaces */
 
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 235cd615b89d..4f9e31912a25 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -22,31 +22,38 @@
 #include "cpu.h"
 
 DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
-	[GDT_ENTRY_KERNEL_CS] = { 0x0000ffff, 0x00cf9a00 },
-	[GDT_ENTRY_KERNEL_DS] = { 0x0000ffff, 0x00cf9200 },
-	[GDT_ENTRY_DEFAULT_USER_CS] = { 0x0000ffff, 0x00cffa00 },
-	[GDT_ENTRY_DEFAULT_USER_DS] = { 0x0000ffff, 0x00cff200 },
+	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } },
+	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } },
+	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } },
+	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } },
 	/*
 	 * Segments used for calling PnP BIOS have byte granularity.
 	 * They code segments and data segments have fixed 64k limits,
 	 * the transfer segment sizes are set at run time.
 	 */
-	[GDT_ENTRY_PNPBIOS_CS32] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
-	[GDT_ENTRY_PNPBIOS_CS16] = { 0x0000ffff, 0x00009a00 },/* 16-bit code */
-	[GDT_ENTRY_PNPBIOS_DS] = { 0x0000ffff, 0x00009200 }, /* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS1] = { 0x00000000, 0x00009200 },/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS2] = { 0x00000000, 0x00009200 },/* 16-bit data */
+	/* 32-bit code */
+	[GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } },
+	/* 16-bit code */
+	[GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } },
+	/* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } },
+	/* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } },
+	/* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } },
 	/*
 	 * The APM segments have byte granularity and their bases
 	 * are set at run time.  All have 64k limits.
 	 */
-	[GDT_ENTRY_APMBIOS_BASE] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
+	/* 32-bit code */
+	[GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } },
 	/* 16-bit code */
-	[GDT_ENTRY_APMBIOS_BASE+1] = { 0x0000ffff, 0x00009a00 },
-	[GDT_ENTRY_APMBIOS_BASE+2] = { 0x0000ffff, 0x00409200 }, /* data */
+	[GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } },
+	/* data */
+	[GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } },
 
-	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
-	[GDT_ENTRY_PERCPU] = { 0x00000000, 0x00000000 },
+	[GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } },
+	[GDT_ENTRY_PERCPU] = { { { 0x00000000, 0x00000000 } } },
 } };
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 843bf0c978a4..86c310acc989 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -455,7 +455,7 @@ static inline void set_32bit_tls(struct task_struct *t, int tls, u32 addr)
 		.limit_in_pages = 1,
 		.useable = 1,
 	};
-	struct n_desc_struct *desc = (void *)t->thread.tls_array;
+	struct desc_struct *desc = (void *)t->thread.tls_array;
 	desc += tls;
 	desc->a = LDT_entry_a(&ud);
 	desc->b = LDT_entry_b(&ud);
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 57491942cc4e..0d45017ed824 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -76,7 +76,8 @@ char ignore_fpu_irq = 0;
  * F0 0F bug workaround.. We have a special link segment
  * for this.
  */
-struct desc_struct idt_table[256] __attribute__((__section__(".data.idt"))) = { {0, 0}, };
+struct desc_struct idt_table[256]
+	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 
 asmlinkage void divide_error(void);
 asmlinkage void debug(void);
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index 089004070099..de47eb0a23aa 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -11,17 +11,26 @@
 
 #include <linux/types.h>
 
+/*
+ * FIXME: Acessing the desc_struct through its fields is more elegant,
+ * and should be the one valid thing to do. However, a lot of open code
+ * still touches the a and b acessors, and doing this allow us to do it
+ * incrementally. We keep the signature as a struct, rather than an union,
+ * so we can get rid of it transparently in the future -- glommer
+ */
 // 8 byte segment descriptor
 struct desc_struct {
-	u16 limit0;
-	u16 base0;
-	unsigned base1 : 8, type : 4, s : 1, dpl : 2, p : 1;
-	unsigned limit : 4, avl : 1, l : 1, d : 1, g : 1, base2 : 8;
-} __attribute__((packed));
+	union {
+		struct { unsigned int a, b; };
+		struct {
+			u16 limit0;
+			u16 base0;
+			unsigned base1: 8, type: 4, s: 1, dpl: 2, p: 1;
+			unsigned limit: 4, avl: 1, l: 1, d: 1, g: 1, base2: 8;
+		};
 
-struct n_desc_struct {
-	unsigned int a,b;
-};
+	};
+} __attribute__((packed));
 
 enum {
 	GATE_INTERRUPT = 0xE,
diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h
index ccd338460811..17c908c0ef1c 100644
--- a/include/asm-x86/lguest.h
+++ b/include/asm-x86/lguest.h
@@ -78,8 +78,8 @@ static inline void lguest_set_ts(void)
 }
 
 /* Full 4G segment descriptors, suitable for CS and DS. */
-#define FULL_EXEC_SEGMENT ((struct desc_struct){0x0000ffff, 0x00cf9b00})
-#define FULL_SEGMENT ((struct desc_struct){0x0000ffff, 0x00cf9300})
+#define FULL_EXEC_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9b00} } })
+#define FULL_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9300} } })
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 9c0ab7f26bd9..bc48ad64de47 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -20,14 +20,11 @@
 #include <linux/cpumask.h>
 #include <linux/init.h>
 #include <asm/processor-flags.h>
+#include <asm/desc_defs.h>
 
 /* flag for disabling the tsc */
 extern int tsc_disable;
 
-struct desc_struct {
-	unsigned long a,b;
-};
-
 static inline int desc_empty(const void *ptr)
 {
 	const u32 *desc = ptr;
-- 
cgit v1.2.3


From 6b68f01baa810e9f63fbf39e9d5c3ef1d94a966f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:12 +0100
Subject: x86: unify struct desc_ptr

This patch unifies struct desc_ptr between i386 and x86_64.
They can be expressed in the exact same way in C code, only
having to change the name of one of them. As Xgt_desc_struct
is ugly and big, this is the one that goes away.

There's also a padding field in i386, but it is not really
needed in the C structure definition.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/asm-offsets_32.c   |  5 ++---
 arch/x86/kernel/cpu/common.c       |  2 +-
 arch/x86/kernel/doublefault_32.c   |  2 +-
 arch/x86/kernel/efi_32.c           |  4 ++--
 arch/x86/kernel/machine_kexec_32.c |  4 ++--
 arch/x86/kernel/reboot_32.c        |  2 +-
 arch/x86/lguest/boot.c             |  4 ++--
 arch/x86/xen/enlighten.c           | 10 +++++-----
 drivers/kvm/svm.c                  |  2 +-
 include/asm-x86/desc_32.h          | 16 +++++-----------
 include/asm-x86/lguest.h           |  8 ++++----
 include/asm-x86/paravirt.h         | 18 +++++++++---------
 include/asm-x86/processor_32.h     |  2 +-
 include/asm-x86/suspend_32.h       |  4 ++--
 14 files changed, 38 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 415313556708..afd84463b712 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -70,9 +70,8 @@ void foo(void)
 	OFFSET(TI_cpu, thread_info, cpu);
 	BLANK();
 
-	OFFSET(GDS_size, Xgt_desc_struct, size);
-	OFFSET(GDS_address, Xgt_desc_struct, address);
-	OFFSET(GDS_pad, Xgt_desc_struct, pad);
+	OFFSET(GDS_size, desc_ptr, size);
+	OFFSET(GDS_address, desc_ptr, address);
 	BLANK();
 
 	OFFSET(PT_EBX, pt_regs, bx);
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 4f9e31912a25..69507ae8a65b 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -649,7 +649,7 @@ struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
  * it's on the real one. */
 void switch_to_new_gdt(void)
 {
-	struct Xgt_desc_struct gdt_descr;
+	struct desc_ptr gdt_descr;
 
 	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
 	gdt_descr.size = GDT_SIZE - 1;
diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
index cc19a3ea403a..d16122a8e4eb 100644
--- a/arch/x86/kernel/doublefault_32.c
+++ b/arch/x86/kernel/doublefault_32.c
@@ -17,7 +17,7 @@ static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE];
 
 static void doublefault_fn(void)
 {
-	struct Xgt_desc_struct gdt_desc = {0, 0};
+	struct desc_ptr gdt_desc = {0, 0};
 	unsigned long gdt, tss;
 
 	store_gdt(&gdt_desc);
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index e2be78f49399..863e8926f2bb 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -69,7 +69,7 @@ static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 {
 	unsigned long cr4;
 	unsigned long temp;
-	struct Xgt_desc_struct gdt_descr;
+	struct desc_ptr gdt_descr;
 
 	spin_lock(&efi_rt_lock);
 	local_irq_save(efi_rt_eflags);
@@ -111,7 +111,7 @@ static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 static void efi_call_phys_epilog(void) __releases(efi_rt_lock)
 {
 	unsigned long cr4;
-	struct Xgt_desc_struct gdt_descr;
+	struct desc_ptr gdt_descr;
 
 	gdt_descr.address = (unsigned long)get_cpu_gdt_table(0);
 	gdt_descr.size = GDT_SIZE - 1;
diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 11b935f4f886..c1cfd60639d4 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -32,7 +32,7 @@ static u32 kexec_pte1[1024] PAGE_ALIGNED;
 
 static void set_idt(void *newidt, __u16 limit)
 {
-	struct Xgt_desc_struct curidt;
+	struct desc_ptr curidt;
 
 	/* ia32 supports unaliged loads & stores */
 	curidt.size    = limit;
@@ -44,7 +44,7 @@ static void set_idt(void *newidt, __u16 limit)
 
 static void set_gdt(void *newgdt, __u16 limit)
 {
-	struct Xgt_desc_struct curgdt;
+	struct desc_ptr curgdt;
 
 	/* ia32 supports unaligned loads & stores */
 	curgdt.size    = limit;
diff --git a/arch/x86/kernel/reboot_32.c b/arch/x86/kernel/reboot_32.c
index bb1a0f889c5e..c3376fae639d 100644
--- a/arch/x86/kernel/reboot_32.c
+++ b/arch/x86/kernel/reboot_32.c
@@ -161,7 +161,7 @@ real_mode_gdt_entries [3] =
 	0x000092000100ffffULL	/* 16-bit real-mode 64k data at 0x00000100 */
 };
 
-static struct Xgt_desc_struct
+static struct desc_ptr
 real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, (long)real_mode_gdt_entries },
 real_mode_idt = { 0x3ff, 0 },
 no_idt = { 0, 0 };
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index c751e3c03e85..aa0bdd5fc4bb 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -229,7 +229,7 @@ static void lguest_write_idt_entry(struct desc_struct *dt,
 /* Changing to a different IDT is very rare: we keep the IDT up-to-date every
  * time it is written, so we can simply loop through all entries and tell the
  * Host about them. */
-static void lguest_load_idt(const struct Xgt_desc_struct *desc)
+static void lguest_load_idt(const struct desc_ptr *desc)
 {
 	unsigned int i;
 	struct desc_struct *idt = (void *)desc->address;
@@ -252,7 +252,7 @@ static void lguest_load_idt(const struct Xgt_desc_struct *desc)
  * hypercall and use that repeatedly to load a new IDT.  I don't think it
  * really matters, but wouldn't it be nice if they were the same?
  */
-static void lguest_load_gdt(const struct Xgt_desc_struct *desc)
+static void lguest_load_gdt(const struct desc_ptr *desc)
 {
 	BUG_ON((desc->size+1)/8 != GDT_ENTRIES);
 	hcall(LHCALL_LOAD_GDT, __pa(desc->address), GDT_ENTRIES, 0);
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index d81e8d709102..c32e0fd0f838 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -295,7 +295,7 @@ static void xen_set_ldt(const void *addr, unsigned entries)
 	xen_mc_issue(PARAVIRT_LAZY_CPU);
 }
 
-static void xen_load_gdt(const struct Xgt_desc_struct *dtr)
+static void xen_load_gdt(const struct desc_ptr *dtr)
 {
 	unsigned long *frames;
 	unsigned long va = dtr->address;
@@ -395,7 +395,7 @@ static int cvt_gate_to_trap(int vector, u32 low, u32 high,
 }
 
 /* Locations of each CPU's IDT */
-static DEFINE_PER_CPU(struct Xgt_desc_struct, idt_desc);
+static DEFINE_PER_CPU(struct desc_ptr, idt_desc);
 
 /* Set an IDT entry.  If the entry is part of the current IDT, then
    also update Xen. */
@@ -427,7 +427,7 @@ static void xen_write_idt_entry(struct desc_struct *dt, int entrynum,
 	preempt_enable();
 }
 
-static void xen_convert_trap_info(const struct Xgt_desc_struct *desc,
+static void xen_convert_trap_info(const struct desc_ptr *desc,
 				  struct trap_info *traps)
 {
 	unsigned in, out, count;
@@ -446,7 +446,7 @@ static void xen_convert_trap_info(const struct Xgt_desc_struct *desc,
 
 void xen_copy_trap_info(struct trap_info *traps)
 {
-	const struct Xgt_desc_struct *desc = &__get_cpu_var(idt_desc);
+	const struct desc_ptr *desc = &__get_cpu_var(idt_desc);
 
 	xen_convert_trap_info(desc, traps);
 }
@@ -454,7 +454,7 @@ void xen_copy_trap_info(struct trap_info *traps)
 /* Load a new IDT into Xen.  In principle this can be per-CPU, so we
    hold a spinlock to protect the static traps[] array (static because
    it avoids allocation, and saves stack space). */
-static void xen_load_idt(const struct Xgt_desc_struct *desc)
+static void xen_load_idt(const struct desc_ptr *desc)
 {
 	static DEFINE_SPINLOCK(lock);
 	static struct trap_info traps[257];
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index 4e04e49a2f1c..ced4ac1955db 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -290,7 +290,7 @@ static void svm_hardware_enable(void *garbage)
 #ifdef CONFIG_X86_64
 	struct desc_ptr gdt_descr;
 #else
-	struct Xgt_desc_struct gdt_descr;
+	struct desc_ptr gdt_descr;
 #endif
 	struct desc_struct *gdt;
 	int me = raw_smp_processor_id();
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index c547403f341d..bc5ca3453466 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -12,12 +12,6 @@
 
 #include <asm/mmu.h>
 
-struct Xgt_desc_struct {
-	unsigned short size;
-	unsigned long address __attribute__((packed));
-	unsigned short pad;
-} __attribute__ ((packed));
-
 struct gdt_page
 {
 	struct desc_struct gdt[GDT_ENTRIES];
@@ -29,7 +23,7 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 	return per_cpu(gdt_page, cpu).gdt;
 }
 
-extern struct Xgt_desc_struct idt_descr;
+extern struct desc_ptr idt_descr;
 extern struct desc_struct idt_table[];
 extern void set_intr_gate(unsigned int irq, void * addr);
 
@@ -107,22 +101,22 @@ static inline void native_load_tr_desc(void)
 	asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8));
 }
 
-static inline void native_load_gdt(const struct Xgt_desc_struct *dtr)
+static inline void native_load_gdt(const struct desc_ptr *dtr)
 {
 	asm volatile("lgdt %0"::"m" (*dtr));
 }
 
-static inline void native_load_idt(const struct Xgt_desc_struct *dtr)
+static inline void native_load_idt(const struct desc_ptr *dtr)
 {
 	asm volatile("lidt %0"::"m" (*dtr));
 }
 
-static inline void native_store_gdt(struct Xgt_desc_struct *dtr)
+static inline void native_store_gdt(struct desc_ptr *dtr)
 {
 	asm ("sgdt %0":"=m" (*dtr));
 }
 
-static inline void native_store_idt(struct Xgt_desc_struct *dtr)
+static inline void native_store_idt(struct desc_ptr *dtr)
 {
 	asm ("sidt %0":"=m" (*dtr));
 }
diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h
index 17c908c0ef1c..3585a1628b59 100644
--- a/include/asm-x86/lguest.h
+++ b/include/asm-x86/lguest.h
@@ -44,13 +44,13 @@ struct lguest_ro_state
 {
 	/* Host information we need to restore when we switch back. */
 	u32 host_cr3;
-	struct Xgt_desc_struct host_idt_desc;
-	struct Xgt_desc_struct host_gdt_desc;
+	struct desc_ptr host_idt_desc;
+	struct desc_ptr host_gdt_desc;
 	u32 host_sp;
 
 	/* Fields which are used when guest is running. */
-	struct Xgt_desc_struct guest_idt_desc;
-	struct Xgt_desc_struct guest_gdt_desc;
+	struct desc_ptr guest_idt_desc;
+	struct desc_ptr guest_gdt_desc;
 	struct i386_hw_tss guest_tss;
 	struct desc_struct guest_idt[IDT_ENTRIES];
 	struct desc_struct guest_gdt[GDT_ENTRIES];
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index e95c2a655165..0333fb6988b5 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -20,7 +20,7 @@
 
 struct page;
 struct thread_struct;
-struct Xgt_desc_struct;
+struct desc_ptr;
 struct tss_struct;
 struct mm_struct;
 struct desc_struct;
@@ -88,10 +88,10 @@ struct pv_cpu_ops {
 
 	/* Segment descriptor handling */
 	void (*load_tr_desc)(void);
-	void (*load_gdt)(const struct Xgt_desc_struct *);
-	void (*load_idt)(const struct Xgt_desc_struct *);
-	void (*store_gdt)(struct Xgt_desc_struct *);
-	void (*store_idt)(struct Xgt_desc_struct *);
+	void (*load_gdt)(const struct desc_ptr *);
+	void (*load_idt)(const struct desc_ptr *);
+	void (*store_gdt)(struct desc_ptr *);
+	void (*store_idt)(struct desc_ptr *);
 	void (*set_ldt)(const void *desc, unsigned entries);
 	unsigned long (*store_tr)(void);
 	void (*load_tls)(struct thread_struct *t, unsigned int cpu);
@@ -630,11 +630,11 @@ static inline void load_TR_desc(void)
 {
 	PVOP_VCALL0(pv_cpu_ops.load_tr_desc);
 }
-static inline void load_gdt(const struct Xgt_desc_struct *dtr)
+static inline void load_gdt(const struct desc_ptr *dtr)
 {
 	PVOP_VCALL1(pv_cpu_ops.load_gdt, dtr);
 }
-static inline void load_idt(const struct Xgt_desc_struct *dtr)
+static inline void load_idt(const struct desc_ptr *dtr)
 {
 	PVOP_VCALL1(pv_cpu_ops.load_idt, dtr);
 }
@@ -642,11 +642,11 @@ static inline void set_ldt(const void *addr, unsigned entries)
 {
 	PVOP_VCALL2(pv_cpu_ops.set_ldt, addr, entries);
 }
-static inline void store_gdt(struct Xgt_desc_struct *dtr)
+static inline void store_gdt(struct desc_ptr *dtr)
 {
 	PVOP_VCALL1(pv_cpu_ops.store_gdt, dtr);
 }
-static inline void store_idt(struct Xgt_desc_struct *dtr)
+static inline void store_idt(struct desc_ptr *dtr)
 {
 	PVOP_VCALL1(pv_cpu_ops.store_idt, dtr);
 }
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index bc48ad64de47..e5056ab9dd9f 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -707,7 +707,7 @@ extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
 /* Defined in head.S */
-extern struct Xgt_desc_struct early_gdt_descr;
+extern struct desc_ptr early_gdt_descr;
 
 extern void cpu_set_gdt(int);
 extern void switch_to_new_gdt(void);
diff --git a/include/asm-x86/suspend_32.h b/include/asm-x86/suspend_32.h
index a2520732ffd6..1bbda3ad7796 100644
--- a/include/asm-x86/suspend_32.h
+++ b/include/asm-x86/suspend_32.h
@@ -12,8 +12,8 @@ static inline int arch_prepare_suspend(void) { return 0; }
 struct saved_context {
   	u16 es, fs, gs, ss;
 	unsigned long cr0, cr2, cr3, cr4;
-	struct Xgt_desc_struct gdt;
-	struct Xgt_desc_struct idt;
+	struct desc_ptr gdt;
+	struct desc_ptr idt;
 	u16 ldt;
 	u16 tss;
 	unsigned long tr;
-- 
cgit v1.2.3


From f6dc247cac1a8ee64e59019c3b948d23f3b1a36f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:12 +0100
Subject: x86: change gdt acessor macro name

This patch changes the name of x86_64 macro used to access the per-cpu
gdt. It is now equal to the i386 version, which will allow code to be shared.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup64.c     |  2 +-
 arch/x86/kernel/suspend_64.c  |  2 +-
 arch/x86/kernel/vsyscall_64.c |  2 +-
 include/asm-x86/desc_64.h     | 10 +++++-----
 4 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index 51297cca6b42..05cafcb94109 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -228,7 +228,7 @@ void __cpuinit cpu_init (void)
 	 * and set up the GDT descriptor:
 	 */
 	if (cpu)
- 		memcpy(cpu_gdt(cpu), cpu_gdt_table, GDT_SIZE);
+		memcpy(get_cpu_gdt_table(cpu), cpu_gdt_table, GDT_SIZE);
 
 	cpu_gdt_descr[cpu].size = GDT_SIZE;
 	load_gdt((const struct desc_ptr *)&cpu_gdt_descr[cpu]);
diff --git a/arch/x86/kernel/suspend_64.c b/arch/x86/kernel/suspend_64.c
index 569f1b540e36..279c25775d19 100644
--- a/arch/x86/kernel/suspend_64.c
+++ b/arch/x86/kernel/suspend_64.c
@@ -140,7 +140,7 @@ void fix_processor_context(void)
 
 	set_tss_desc(cpu,t);	/* This just modifies memory; should not be necessary. But... This is necessary, because 386 hardware has concept of busy TSS or some similar stupidity. */
 
-	cpu_gdt(cpu)[GDT_ENTRY_TSS].type = 9;
+	get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9;
 
 	syscall_init();                         /* This sets MSR_*STAR and related */
 	load_TR_desc();				/* This does ltr */
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index c4c5e765cd2c..e5c1118a8098 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -297,7 +297,7 @@ static void __cpuinit vsyscall_set_cpu(int cpu)
 	/* Store cpu number in limit so that it can be loaded quickly
 	   in user space in vgetcpu.
 	   12 bits for the CPU and 8 bits for the node. */
-	d = (unsigned long *)(cpu_gdt(cpu) + GDT_ENTRY_PER_CPU);
+	d = (unsigned long *)(get_cpu_gdt_table(cpu) + GDT_ENTRY_PER_CPU);
 	*d = 0x0f40000000000ULL;
 	*d |= cpu;
 	*d |= (node & 0xf) << 12;
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 230ac6e50a0f..4d67f61b4c23 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -48,7 +48,7 @@ static inline void write_ldt_entry(struct desc_struct *ldt,
 }
 
 /* the cpu gdt accessor */
-#define cpu_gdt(_cpu) ((struct desc_struct *)cpu_gdt_descr[_cpu].address)
+#define get_cpu_gdt_table(x) ((struct desc_struct *)cpu_gdt_descr[x].address)
 
 static inline void load_gdt(const struct desc_ptr *ptr)
 {
@@ -141,15 +141,15 @@ static inline void set_tss_desc(unsigned cpu, void *addr)
 	 * -1? seg base+limit should be pointing to the address of the
 	 * last valid byte
 	 */
-	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_TSS],
+	set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS],
 		(unsigned long)addr, DESC_TSS,
 		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
 }
 
 static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
 {
-	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_LDT], (unsigned long)addr,
-			      DESC_LDT, size * 8 - 1);
+	set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT],
+			     (unsigned long)addr, DESC_LDT, size * 8 - 1);
 }
 
 #define LDT_entry_a(info) \
@@ -183,7 +183,7 @@ static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
 static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
 	unsigned int i;
-	u64 *gdt = (u64 *)(cpu_gdt(cpu) + GDT_ENTRY_TLS_MIN);
+	u64 *gdt = (u64 *)(get_cpu_gdt_table(cpu) + GDT_ENTRY_TLS_MIN);
 
 	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
 		gdt[i] = t->tls_array[i];
-- 
cgit v1.2.3


From 7e6ebe1432b04c2980cd030c9f0d1ed480e1fe4d Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:12 +0100
Subject: x86: removed unused variable

This variable is not used anywere, and is then removed

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_64.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 4d67f61b4c23..8bd84a4ef2d5 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -30,11 +30,6 @@ static inline unsigned long __store_tr(void)
 
 #define store_tr(tr) (tr) = __store_tr()
 
-/*
- * This is the ldt that every process will get unless we need
- * something other than this.
- */
-extern struct desc_struct default_ldt[];
 extern struct gate_struct idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
-- 
cgit v1.2.3


From 010d4f8221cf51a2ab8b037d0149506b397d073f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:12 +0100
Subject: x86: introduce gate_desc type.

To account for the differences in gate descriptor in i386 and x86_64
a gate_desc type is introduced.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/traps_32.c  |  2 +-
 include/asm-x86/desc_32.h   | 15 ++++++++-------
 include/asm-x86/desc_64.h   |  4 ++--
 include/asm-x86/desc_defs.h |  8 +++++++-
 4 files changed, 18 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 0d45017ed824..c70c41fd710b 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -76,7 +76,7 @@ char ignore_fpu_irq = 0;
  * F0 0F bug workaround.. We have a special link segment
  * for this.
  */
-struct desc_struct idt_table[256]
+gate_desc idt_table[256]
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 
 asmlinkage void divide_error(void);
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index bc5ca3453466..77f1e5a4ad7c 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -3,6 +3,7 @@
 
 #include <asm/ldt.h>
 #include <asm/segment.h>
+#include <asm/desc_defs.h>
 
 #ifndef __ASSEMBLY__
 
@@ -24,7 +25,7 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 }
 
 extern struct desc_ptr idt_descr;
-extern struct desc_struct idt_table[];
+extern gate_desc idt_table[];
 extern void set_intr_gate(unsigned int irq, void * addr);
 
 static inline void pack_descriptor(__u32 *a, __u32 *b,
@@ -35,11 +36,11 @@ static inline void pack_descriptor(__u32 *a, __u32 *b,
 		(limit & 0x000f0000) | ((type & 0xff) << 8) | ((flags & 0xf) << 20);
 }
 
-static inline void pack_gate(__u32 *a, __u32 *b,
+static inline void pack_gate(gate_desc *gate,
 	unsigned long base, unsigned short seg, unsigned char type, unsigned char flags)
 {
-	*a = (seg << 16) | (base & 0xffff);
-	*b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
+	gate->a = (seg << 16) | (base & 0xffff);
+	gate->b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
 }
 
 #define DESCTYPE_LDT 	0x82	/* present, system, DPL-0, LDT */
@@ -139,9 +140,9 @@ static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
 
 static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg)
 {
-	__u32 a, b;
-	pack_gate(&a, &b, (unsigned long)addr, seg, type, 0);
-	write_idt_entry(idt_table, gate, a, b);
+	gate_desc g;
+	pack_gate(&g, (unsigned long)addr, seg, type, 0);
+	write_idt_entry(idt_table, gate, g.a, g.b);
 }
 
 static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 8bd84a4ef2d5..887393f044b1 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -30,7 +30,7 @@ static inline unsigned long __store_tr(void)
 
 #define store_tr(tr) (tr) = __store_tr()
 
-extern struct gate_struct idt_table[];
+extern gate_desc idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
 static inline void write_ldt_entry(struct desc_struct *ldt,
@@ -58,7 +58,7 @@ static inline void store_gdt(struct desc_ptr *ptr)
 static inline void _set_gate(void *adr, unsigned type, unsigned long func,
 			     unsigned dpl, unsigned ist)
 {
-	struct gate_struct s;
+	gate_desc s;
 
 	s.offset_low = PTR_LOW(func);
 	s.segment = __KERNEL_CS;
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index de47eb0a23aa..9732285116af 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -39,7 +39,7 @@ enum {
 };
 
 // 16byte gate
-struct gate_struct {
+struct gate_struct64 {
 	u16 offset_low;
 	u16 segment;
 	unsigned ist : 3, zero0 : 5, type : 5, dpl : 2, p : 1;
@@ -67,6 +67,12 @@ struct ldttss_desc {
 	u32 zero1;
 } __attribute__((packed));
 
+#ifdef CONFIG_X86_64
+typedef struct gate_struct64 gate_desc;
+#else
+typedef struct desc_struct gate_desc;
+#endif
+
 struct desc_ptr {
 	unsigned short size;
 	unsigned long address;
-- 
cgit v1.2.3


From 8d947344c47a40626730bb80d136d8daac9f2060 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:12 +0100
Subject: x86: change write_idt_entry signature

this patch changes write_idt_entry signature. It now takes a gate_desc
instead of the a and b parameters. It will allow it to be later unified
between i386 and x86_64.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
CC: Zachary Amsden <zach@vmware.com>
CC: Jeremy Fitzhardinge <Jeremy.Fitzhardinge.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/paravirt_32.c |  2 +-
 arch/x86/kernel/vmi_32.c      | 10 +++++++++-
 arch/x86/lguest/boot.c        |  9 +++++----
 arch/x86/xen/enlighten.c      |  8 ++++----
 include/asm-x86/desc_32.h     | 10 ++++++++--
 include/asm-x86/paravirt.h    |  9 +++++----
 6 files changed, 32 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/paravirt_32.c b/arch/x86/kernel/paravirt_32.c
index f4e3a8e01cf2..13bbc99b639b 100644
--- a/arch/x86/kernel/paravirt_32.c
+++ b/arch/x86/kernel/paravirt_32.c
@@ -381,7 +381,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.load_tls = native_load_tls,
 	.write_ldt_entry = write_dt_entry,
 	.write_gdt_entry = write_dt_entry,
-	.write_idt_entry = write_dt_entry,
+	.write_idt_entry = native_write_idt_entry,
 	.load_sp0 = native_load_sp0,
 
 	.irq_enable_syscall_ret = native_irq_enable_syscall_ret,
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 4cfda7dbe90f..a635b22de25f 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -62,6 +62,7 @@ static struct {
 	void (*cpuid)(void /* non-c */);
 	void (*_set_ldt)(u32 selector);
 	void (*set_tr)(u32 selector);
+	void (*write_idt_entry)(struct desc_struct *, int, u32, u32);
 	void (*set_kernel_stack)(u32 selector, u32 sp0);
 	void (*allocate_page)(u32, u32, u32, u32, u32);
 	void (*release_page)(u32, u32);
@@ -214,6 +215,12 @@ static void vmi_set_tr(void)
 	vmi_ops.set_tr(GDT_ENTRY_TSS*sizeof(struct desc_struct));
 }
 
+static void vmi_write_idt_entry(gate_desc *dt, int entry, const gate_desc *g)
+{
+	u32 *idt_entry = (u32 *)g;
+	vmi_ops.write_idt_entry(dt, entry, idt_entry[0], idt_entry[2]);
+}
+
 static void vmi_load_sp0(struct tss_struct *tss,
 				   struct thread_struct *thread)
 {
@@ -792,7 +799,8 @@ static inline int __init activate_vmi(void)
 	pv_cpu_ops.load_tls = vmi_load_tls;
 	para_fill(pv_cpu_ops.write_ldt_entry, WriteLDTEntry);
 	para_fill(pv_cpu_ops.write_gdt_entry, WriteGDTEntry);
-	para_fill(pv_cpu_ops.write_idt_entry, WriteIDTEntry);
+	para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry,
+		  write_idt_entry, WriteIDTEntry);
 	para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack);
 	para_fill(pv_cpu_ops.set_iopl_mask, SetIOPLMask);
 	para_fill(pv_cpu_ops.io_delay, IODelay);
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index aa0bdd5fc4bb..b50c8ad25ab4 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -217,13 +217,14 @@ static void irq_enable(void)
  * address of the handler, and... well, who cares?  The Guest just asks the
  * Host to make the change anyway, because the Host controls the real IDT.
  */
-static void lguest_write_idt_entry(struct desc_struct *dt,
-				   int entrynum, u32 low, u32 high)
+static void lguest_write_idt_entry(gate_desc *dt,
+				   int entrynum, const gate_desc *g)
 {
+	u32 *desc = (u32 *)g;
 	/* Keep the local copy up to date. */
-	write_dt_entry(dt, entrynum, low, high);
+	native_write_idt_entry(dt, entrynum, g);
 	/* Tell Host about this new entry. */
-	hcall(LHCALL_LOAD_IDT_ENTRY, entrynum, low, high);
+	hcall(LHCALL_LOAD_IDT_ENTRY, entrynum, desc[0], desc[1]);
 }
 
 /* Changing to a different IDT is very rare: we keep the IDT up-to-date every
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index c32e0fd0f838..b7b7346d8cdc 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -399,8 +399,7 @@ static DEFINE_PER_CPU(struct desc_ptr, idt_desc);
 
 /* Set an IDT entry.  If the entry is part of the current IDT, then
    also update Xen. */
-static void xen_write_idt_entry(struct desc_struct *dt, int entrynum,
-				u32 low, u32 high)
+static void xen_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g)
 {
 	unsigned long p = (unsigned long)&dt[entrynum];
 	unsigned long start, end;
@@ -412,14 +411,15 @@ static void xen_write_idt_entry(struct desc_struct *dt, int entrynum,
 
 	xen_mc_flush();
 
-	write_dt_entry(dt, entrynum, low, high);
+	native_write_idt_entry(dt, entrynum, g);
 
 	if (p >= start && (p + 8) <= end) {
 		struct trap_info info[2];
+		u32 *desc = (u32 *)g;
 
 		info[1].address = 0;
 
-		if (cvt_gate_to_trap(entrynum, low, high, &info[0]))
+		if (cvt_gate_to_trap(entrynum, desc[0], desc[1], &info[0]))
 			if (HYPERVISOR_set_trap_table(info))
 				BUG();
 	}
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 77f1e5a4ad7c..54b2314f2ddf 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -70,9 +70,15 @@ static inline void pack_gate(gate_desc *gate,
 
 #define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
 #define write_gdt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
-#define write_idt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
 #endif
 
+static inline void native_write_idt_entry(gate_desc *idt, int entry,
+					  const gate_desc *gate)
+{
+	memcpy(&idt[entry], gate, sizeof(*gate));
+}
+
 static inline void write_dt_entry(struct desc_struct *dt,
 				  int entry, u32 entry_low, u32 entry_high)
 {
@@ -142,7 +148,7 @@ static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned s
 {
 	gate_desc g;
 	pack_gate(&g, (unsigned long)addr, seg, type, 0);
-	write_idt_entry(idt_table, gate, g.a, g.b);
+	write_idt_entry(idt_table, gate, &g);
 }
 
 static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 0333fb6988b5..86a9d7b0920f 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -17,6 +17,7 @@
 #include <linux/types.h>
 #include <linux/cpumask.h>
 #include <asm/kmap_types.h>
+#include <asm/desc_defs.h>
 
 struct page;
 struct thread_struct;
@@ -99,8 +100,8 @@ struct pv_cpu_ops {
 				int entrynum, u32 low, u32 high);
 	void (*write_gdt_entry)(struct desc_struct *,
 				int entrynum, u32 low, u32 high);
-	void (*write_idt_entry)(struct desc_struct *,
-				int entrynum, u32 low, u32 high);
+	void (*write_idt_entry)(gate_desc *,
+				int entrynum, const gate_desc *gate);
 	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
 
 	void (*set_iopl_mask)(unsigned mask);
@@ -667,9 +668,9 @@ static inline void write_gdt_entry(void *dt, int entry, u32 low, u32 high)
 {
 	PVOP_VCALL4(pv_cpu_ops.write_gdt_entry, dt, entry, low, high);
 }
-static inline void write_idt_entry(void *dt, int entry, u32 low, u32 high)
+static inline void write_idt_entry(gate_desc *dt, int entry, const gate_desc *g)
 {
-	PVOP_VCALL4(pv_cpu_ops.write_idt_entry, dt, entry, low, high);
+	PVOP_VCALL3(pv_cpu_ops.write_idt_entry, dt, entry, g);
 }
 static inline void set_iopl_mask(unsigned mask)
 {
-- 
cgit v1.2.3


From 25deca53e9b74a899efd3ae1f4493f4f2ce36846 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:13 +0100
Subject: x86: introduce ldt_desc type.

this patch introduces ldt_desc type to account for the differences
in the ldt descriptor in x86_64 and i386

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_64.h   | 2 +-
 include/asm-x86/desc_defs.h | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 887393f044b1..c49f928ed8b6 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -113,7 +113,7 @@ static inline void store_idt(struct desc_ptr *dtr)
 static inline void set_tssldt_descriptor(void *ptr, unsigned long tss,
 					 unsigned type, unsigned size)
 {
-	struct ldttss_desc d;
+	struct ldttss_desc64 d;
 
 	memset(&d, 0, sizeof(d));
 	d.limit0 = size & 0xFFFF;
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index 9732285116af..09ea05243497 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -58,7 +58,7 @@ enum {
 };
 
 // LDT or TSS descriptor in the GDT. 16 bytes.
-struct ldttss_desc {
+struct ldttss_desc64 {
 	u16 limit0;
 	u16 base0;
 	unsigned base1 : 8, type : 5, dpl : 2, p : 1;
@@ -69,8 +69,10 @@ struct ldttss_desc {
 
 #ifdef CONFIG_X86_64
 typedef struct gate_struct64 gate_desc;
+typedef struct ldttss_desc64 ldt_desc;
 #else
 typedef struct desc_struct gate_desc;
+typedef struct desc_struct ldt_desc;
 #endif
 
 struct desc_ptr {
-- 
cgit v1.2.3


From 80fbb69a8d1268ef48dfe21da80e68cb01922f31 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:13 +0100
Subject: x86: introduce fill_ldt

This patch introduces fill_ldt(), which populates a ldt descriptor
from a user_desc in once, instead of relying in the LDT_entry_a and
LDT_entry_b macros

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ldt.c        |  3 +--
 arch/x86/kernel/process_64.c |  3 +--
 arch/x86/kernel/tls.c        |  7 +++----
 include/asm-x86/desc.h       | 29 +++++++++++++++++++++++++++++
 include/asm-x86/desc_32.h    | 15 ---------------
 include/asm-x86/desc_64.h    | 17 -----------------
 6 files changed, 34 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 7eb0c8a45734..3e872b468533 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -223,8 +223,7 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 		}
 	}
 
-	ldt.a = LDT_entry_a(&ldt_info);
-	ldt.b = LDT_entry_b(&ldt_info);
+	fill_ldt(&ldt, &ldt_info);
 	if (oldmode)
 		ldt.avl = 0;
 
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 86c310acc989..f91521e26335 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -457,8 +457,7 @@ static inline void set_32bit_tls(struct task_struct *t, int tls, u32 addr)
 	};
 	struct desc_struct *desc = (void *)t->thread.tls_array;
 	desc += tls;
-	desc->a = LDT_entry_a(&ud);
-	desc->b = LDT_entry_b(&ud);
+	fill_ldt(desc, &ud);
 }
 
 static inline u32 read_32bit_tls(struct task_struct *t, int tls)
diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c
index 67a377621b12..74d2b65a82eb 100644
--- a/arch/x86/kernel/tls.c
+++ b/arch/x86/kernel/tls.c
@@ -67,10 +67,9 @@ int do_set_thread_area(struct task_struct *p, int idx,
 	if (LDT_empty(&info)) {
 		desc[0] = 0;
 		desc[1] = 0;
-	} else {
-		desc[0] = LDT_entry_a(&info);
-		desc[1] = LDT_entry_b(&info);
-	}
+	} else
+		fill_ldt((struct desc_struct *)desc, &info);
+
 	if (t == &current->thread)
 		load_TLS(t, cpu);
 
diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index 6065c5092265..47086d2d9298 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -1,5 +1,34 @@
+#ifndef _ASM_DESC_H_
+#define _ASM_DESC_H_
+
+#ifndef __ASSEMBLY__
+#include <asm/desc_defs.h>
+#include <asm/ldt.h>
+
+static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
+{
+	desc->limit0 = info->limit & 0x0ffff;
+	desc->base0 = info->base_addr & 0x0000ffff;
+
+	desc->base1 = (info->base_addr & 0x00ff0000) >> 16;
+	desc->type = (info->read_exec_only ^ 1) << 1;
+	desc->type |= info->contents << 2;
+	desc->s = 1;
+	desc->dpl = 0x3;
+	desc->p = info->seg_not_present ^ 1;
+	desc->limit = (info->limit & 0xf0000) >> 16;
+	desc->avl = info->useable;
+	desc->d = info->seg_32bit;
+	desc->g = info->limit_in_pages;
+	desc->base2 = (info->base_addr & 0xff000000) >> 24;
+}
+
+#endif
+
 #ifdef CONFIG_X86_32
 # include "desc_32.h"
 #else
 # include "desc_64.h"
 #endif
+
+#endif
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 54b2314f2ddf..03700991c5db 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -163,21 +163,6 @@ static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const vo
 
 #define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
 
-#define LDT_entry_a(info) \
-	((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
-
-#define LDT_entry_b(info) \
-	(((info)->base_addr & 0xff000000) | \
-	(((info)->base_addr & 0x00ff0000) >> 16) | \
-	((info)->limit & 0xf0000) | \
-	(((info)->read_exec_only ^ 1) << 9) | \
-	((info)->contents << 10) | \
-	(((info)->seg_not_present ^ 1) << 15) | \
-	((info)->seg_32bit << 22) | \
-	((info)->limit_in_pages << 23) | \
-	((info)->useable << 20) | \
-	0x7000)
-
 #define LDT_empty(info) (\
 	(info)->base_addr	== 0	&& \
 	(info)->limit		== 0	&& \
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index c49f928ed8b6..ba7fb87d10f3 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -147,23 +147,6 @@ static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
 			     (unsigned long)addr, DESC_LDT, size * 8 - 1);
 }
 
-#define LDT_entry_a(info) \
-	((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
-/* Don't allow setting of the lm bit. It is useless anyways because
-   64bit system calls require __USER_CS. */
-#define LDT_entry_b(info) \
-	(((info)->base_addr & 0xff000000) | \
-	(((info)->base_addr & 0x00ff0000) >> 16) | \
-	((info)->limit & 0xf0000) | \
-	(((info)->read_exec_only ^ 1) << 9) | \
-	((info)->contents << 10) | \
-	(((info)->seg_not_present ^ 1) << 15) | \
-	((info)->seg_32bit << 22) | \
-	((info)->limit_in_pages << 23) | \
-	((info)->useable << 20) | \
-	/* ((info)->lm << 21) | */ \
-	0x7000)
-
 #define LDT_empty(info) (\
 	(info)->base_addr	== 0	&& \
 	(info)->limit		== 0	&& \
-- 
cgit v1.2.3


From 18245d5b6e54ca76b60b687e43eb893f9e9dd611 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:13 +0100
Subject: x86: provide tss_desc

Provide a new type, tss_desc, to represent the tss descriptor
in a unified way accross x86_64 and i386

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_defs.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index 09ea05243497..c811f1e0829f 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -70,9 +70,11 @@ struct ldttss_desc64 {
 #ifdef CONFIG_X86_64
 typedef struct gate_struct64 gate_desc;
 typedef struct ldttss_desc64 ldt_desc;
+typedef struct ldttss_desc64 tss_desc;
 #else
 typedef struct desc_struct gate_desc;
 typedef struct desc_struct ldt_desc;
+typedef struct desc_struct tss_desc;
 #endif
 
 struct desc_ptr {
-- 
cgit v1.2.3


From 014b15be30c04622d130946ab7c0a9101b523a8a Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:13 +0100
Subject: x86: change write_gdt_entry signature.

This patch changes the write_gdt_entry function signature.
Instead of the old "a" and "b" parameters, it now receives
a pointer to a desc_struct, and the size of the entry being
handled. This is because x86_64 can have some 16-byte entries
as well as 8-byte ones.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
CC: Zachary Amsden <zach@vmware.com>
CC: Jeremy Fitzhardinge <Jeremy.Fitzhardinge.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/paravirt_32.c  |  2 +-
 arch/x86/kernel/smpcommon_32.c |  7 ++++---
 arch/x86/kernel/vmi_32.c       | 21 +++++++++++++++------
 arch/x86/lguest/boot.c         |  6 +++---
 arch/x86/xen/enlighten.c       | 11 +++++------
 include/asm-x86/desc_32.h      | 34 ++++++++++++++++++++++------------
 include/asm-x86/paravirt.h     |  9 ++++++---
 7 files changed, 56 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/paravirt_32.c b/arch/x86/kernel/paravirt_32.c
index 13bbc99b639b..77602c1252d7 100644
--- a/arch/x86/kernel/paravirt_32.c
+++ b/arch/x86/kernel/paravirt_32.c
@@ -380,7 +380,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.store_tr = native_store_tr,
 	.load_tls = native_load_tls,
 	.write_ldt_entry = write_dt_entry,
-	.write_gdt_entry = write_dt_entry,
+	.write_gdt_entry = native_write_gdt_entry,
 	.write_idt_entry = native_write_idt_entry,
 	.load_sp0 = native_load_sp0,
 
diff --git a/arch/x86/kernel/smpcommon_32.c b/arch/x86/kernel/smpcommon_32.c
index bbfe85a0f699..8bc38af29aef 100644
--- a/arch/x86/kernel/smpcommon_32.c
+++ b/arch/x86/kernel/smpcommon_32.c
@@ -14,10 +14,11 @@ __cpuinit void init_gdt(int cpu)
 {
 	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
 
-	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PERCPU].a,
-			(u32 *)&gdt[GDT_ENTRY_PERCPU].b,
+	pack_descriptor(&gdt[GDT_ENTRY_PERCPU],
 			__per_cpu_offset[cpu], 0xFFFFF,
-			0x80 | DESCTYPE_S | 0x2, 0x8);
+			0x2 | DESCTYPE_S, 0x8);
+
+	gdt[GDT_ENTRY_PERCPU].s = 1;
 
 	per_cpu(this_cpu_off, cpu) = __per_cpu_offset[cpu];
 	per_cpu(cpu_number, cpu) = cpu;
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index a635b22de25f..21edd0d6eae5 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -63,6 +63,7 @@ static struct {
 	void (*_set_ldt)(u32 selector);
 	void (*set_tr)(u32 selector);
 	void (*write_idt_entry)(struct desc_struct *, int, u32, u32);
+	void (*write_gdt_entry)(struct desc_struct *, int, u32, u32);
 	void (*set_kernel_stack)(u32 selector, u32 sp0);
 	void (*allocate_page)(u32, u32, u32, u32, u32);
 	void (*release_page)(u32, u32);
@@ -187,7 +188,7 @@ static void vmi_cpuid(unsigned int *ax, unsigned int *bx,
 static inline void vmi_maybe_load_tls(struct desc_struct *gdt, int nr, struct desc_struct *new)
 {
 	if (gdt[nr].a != new->a || gdt[nr].b != new->b)
-		write_gdt_entry(gdt, nr, new->a, new->b);
+		write_gdt_entry(gdt, nr, new, 0);
 }
 
 static void vmi_load_tls(struct thread_struct *t, unsigned int cpu)
@@ -201,12 +202,12 @@ static void vmi_load_tls(struct thread_struct *t, unsigned int cpu)
 static void vmi_set_ldt(const void *addr, unsigned entries)
 {
 	unsigned cpu = smp_processor_id();
-	u32 low, high;
+	struct desc_struct desc;
 
-	pack_descriptor(&low, &high, (unsigned long)addr,
+	pack_descriptor(&desc, (unsigned long)addr,
 			entries * sizeof(struct desc_struct) - 1,
-			DESCTYPE_LDT, 0);
-	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, low, high);
+			DESC_LDT, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, &desc, DESC_LDT);
 	vmi_ops._set_ldt(entries ? GDT_ENTRY_LDT*sizeof(struct desc_struct) : 0);
 }
 
@@ -221,6 +222,13 @@ static void vmi_write_idt_entry(gate_desc *dt, int entry, const gate_desc *g)
 	vmi_ops.write_idt_entry(dt, entry, idt_entry[0], idt_entry[2]);
 }
 
+static void vmi_write_gdt_entry(struct desc_struct *dt, int entry,
+				const void *desc, int type)
+{
+	u32 *gdt_entry = (u32 *)desc;
+	vmi_ops.write_gdt_entry(dt, entry, gdt_entry[0], gdt_entry[2]);
+}
+
 static void vmi_load_sp0(struct tss_struct *tss,
 				   struct thread_struct *thread)
 {
@@ -798,7 +806,8 @@ static inline int __init activate_vmi(void)
 	para_fill(pv_cpu_ops.store_tr, GetTR);
 	pv_cpu_ops.load_tls = vmi_load_tls;
 	para_fill(pv_cpu_ops.write_ldt_entry, WriteLDTEntry);
-	para_fill(pv_cpu_ops.write_gdt_entry, WriteGDTEntry);
+	para_wrap(pv_cpu_ops.write_gdt_entry, vmi_write_gdt_entry,
+		  write_gdt_entry, WriteGDTEntry);
 	para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry,
 		  write_idt_entry, WriteIDTEntry);
 	para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack);
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index b50c8ad25ab4..a63373759f08 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -262,10 +262,10 @@ static void lguest_load_gdt(const struct desc_ptr *desc)
 /* For a single GDT entry which changes, we do the lazy thing: alter our GDT,
  * then tell the Host to reload the entire thing.  This operation is so rare
  * that this naive implementation is reasonable. */
-static void lguest_write_gdt_entry(struct desc_struct *dt,
-				   int entrynum, u32 low, u32 high)
+static void lguest_write_gdt_entry(struct desc_struct *dt, int entrynum,
+				   const void *desc, int type)
 {
-	write_dt_entry(dt, entrynum, low, high);
+	native_write_gdt_entry(dt, entrynum, desc, type);
 	hcall(LHCALL_LOAD_GDT, __pa(dt), GDT_ENTRIES, 0);
 }
 
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index b7b7346d8cdc..7f98c63f6381 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -475,22 +475,21 @@ static void xen_load_idt(const struct desc_ptr *desc)
 /* Write a GDT descriptor entry.  Ignore LDT descriptors, since
    they're handled differently. */
 static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
-				u32 low, u32 high)
+				const void *desc, int type)
 {
 	preempt_disable();
 
-	switch ((high >> 8) & 0xff) {
-	case DESCTYPE_LDT:
-	case DESCTYPE_TSS:
+	switch (type) {
+	case DESC_LDT:
+	case DESC_TSS:
 		/* ignore */
 		break;
 
 	default: {
 		xmaddr_t maddr = virt_to_machine(&dt[entry]);
-		u64 desc = (u64)high << 32 | low;
 
 		xen_mc_flush();
-		if (HYPERVISOR_update_descriptor(maddr.maddr, desc))
+		if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc))
 			BUG();
 	}
 
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 03700991c5db..41d8214c6173 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -28,12 +28,13 @@ extern struct desc_ptr idt_descr;
 extern gate_desc idt_table[];
 extern void set_intr_gate(unsigned int irq, void * addr);
 
-static inline void pack_descriptor(__u32 *a, __u32 *b,
+static inline void pack_descriptor(struct desc_struct *desc,
 	unsigned long base, unsigned long limit, unsigned char type, unsigned char flags)
 {
-	*a = ((base & 0xffff) << 16) | (limit & 0xffff);
-	*b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
+	desc->a = ((base & 0xffff) << 16) | (limit & 0xffff);
+	desc->b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
 		(limit & 0x000f0000) | ((type & 0xff) << 8) | ((flags & 0xf) << 20);
+	desc->p = 1;
 }
 
 static inline void pack_gate(gate_desc *gate,
@@ -69,7 +70,8 @@ static inline void pack_gate(gate_desc *gate,
 #define set_ldt native_set_ldt
 
 #define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
-#define write_gdt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_gdt_entry(dt, entry, desc, type) \
+				native_write_gdt_entry(dt, entry, desc, type)
 #define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
 #endif
 
@@ -79,6 +81,12 @@ static inline void native_write_idt_entry(gate_desc *idt, int entry,
 	memcpy(&idt[entry], gate, sizeof(*gate));
 }
 
+static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry,
+					  const void *desc, int type)
+{
+	memcpy(&gdt[entry], desc, sizeof(struct desc_struct));
+}
+
 static inline void write_dt_entry(struct desc_struct *dt,
 				  int entry, u32 entry_low, u32 entry_high)
 {
@@ -86,18 +94,20 @@ static inline void write_dt_entry(struct desc_struct *dt,
 	dt[entry].b = entry_high;
 }
 
+
 static inline void native_set_ldt(const void *addr, unsigned int entries)
 {
 	if (likely(entries == 0))
 		__asm__ __volatile__("lldt %w0"::"q" (0));
 	else {
 		unsigned cpu = smp_processor_id();
-		__u32 a, b;
+		ldt_desc ldt;
 
-		pack_descriptor(&a, &b, (unsigned long)addr,
+		pack_descriptor(&ldt, (unsigned long)addr,
 				entries * sizeof(struct desc_struct) - 1,
-				DESCTYPE_LDT, 0);
-		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, a, b);
+				DESC_LDT, 0);
+		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT,
+				&ldt, DESC_LDT);
 		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
 	}
 }
@@ -153,11 +163,11 @@ static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned s
 
 static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
 {
-	__u32 a, b;
-	pack_descriptor(&a, &b, (unsigned long)addr,
+	tss_desc tss;
+	pack_descriptor(&tss, (unsigned long)addr,
 			offsetof(struct tss_struct, __cacheline_filler) - 1,
-			DESCTYPE_TSS, 0);
-	write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b);
+			DESC_TSS, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), entry, &tss, DESC_TSS);
 }
 
 
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 86a9d7b0920f..3f2abf295e2a 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -99,7 +99,7 @@ struct pv_cpu_ops {
 	void (*write_ldt_entry)(struct desc_struct *,
 				int entrynum, u32 low, u32 high);
 	void (*write_gdt_entry)(struct desc_struct *,
-				int entrynum, u32 low, u32 high);
+				int entrynum, const void *desc, int size);
 	void (*write_idt_entry)(gate_desc *,
 				int entrynum, const gate_desc *gate);
 	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
@@ -664,10 +664,13 @@ static inline void write_ldt_entry(void *dt, int entry, u32 low, u32 high)
 {
 	PVOP_VCALL4(pv_cpu_ops.write_ldt_entry, dt, entry, low, high);
 }
-static inline void write_gdt_entry(void *dt, int entry, u32 low, u32 high)
+
+static inline void write_gdt_entry(struct desc_struct *dt, int entry,
+				   void *desc, int type)
 {
-	PVOP_VCALL4(pv_cpu_ops.write_gdt_entry, dt, entry, low, high);
+	PVOP_VCALL4(pv_cpu_ops.write_gdt_entry, dt, entry, desc, type);
 }
+
 static inline void write_idt_entry(gate_desc *dt, int entry, const gate_desc *g)
 {
 	PVOP_VCALL3(pv_cpu_ops.write_idt_entry, dt, entry, g);
-- 
cgit v1.2.3


From 75b8bb3e56ca09a467fbbe5229bc68627f7445be Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:13 +0100
Subject: x86: change write_ldt_entry signature

this patch changes the signature of write_ldt_entry.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
CC: Zachary Amsden <zach@vmware.com>
CC: Jeremy Fitzhardinge <Jeremy.Fitzhardinge.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ldt.c         |  3 +--
 arch/x86/kernel/paravirt_32.c |  2 +-
 arch/x86/kernel/vmi_32.c      | 11 ++++++++++-
 arch/x86/xen/enlighten.c      |  4 ++--
 include/asm-x86/desc_32.h     |  9 ++++++++-
 include/asm-x86/desc_64.h     |  7 ++-----
 include/asm-x86/paravirt.h    | 10 ++++++----
 7 files changed, 30 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 3e872b468533..b8ef46270e24 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -229,8 +229,7 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
 
 	/* Install the new entry ...  */
 install:
-	write_ldt_entry(mm->context.ldt, ldt_info.entry_number,
-			ldt.a, ldt.b);
+	write_ldt_entry(mm->context.ldt, ldt_info.entry_number, &ldt);
 	error = 0;
 
 out_unlock:
diff --git a/arch/x86/kernel/paravirt_32.c b/arch/x86/kernel/paravirt_32.c
index 77602c1252d7..dd063fba2b21 100644
--- a/arch/x86/kernel/paravirt_32.c
+++ b/arch/x86/kernel/paravirt_32.c
@@ -379,7 +379,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.store_idt = native_store_idt,
 	.store_tr = native_store_tr,
 	.load_tls = native_load_tls,
-	.write_ldt_entry = write_dt_entry,
+	.write_ldt_entry = native_write_ldt_entry,
 	.write_gdt_entry = native_write_gdt_entry,
 	.write_idt_entry = native_write_idt_entry,
 	.load_sp0 = native_load_sp0,
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 21edd0d6eae5..10c46419d35d 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -64,6 +64,7 @@ static struct {
 	void (*set_tr)(u32 selector);
 	void (*write_idt_entry)(struct desc_struct *, int, u32, u32);
 	void (*write_gdt_entry)(struct desc_struct *, int, u32, u32);
+	void (*write_ldt_entry)(struct desc_struct *, int, u32, u32);
 	void (*set_kernel_stack)(u32 selector, u32 sp0);
 	void (*allocate_page)(u32, u32, u32, u32, u32);
 	void (*release_page)(u32, u32);
@@ -229,6 +230,13 @@ static void vmi_write_gdt_entry(struct desc_struct *dt, int entry,
 	vmi_ops.write_gdt_entry(dt, entry, gdt_entry[0], gdt_entry[2]);
 }
 
+static void vmi_write_ldt_entry(struct desc_struct *dt, int entry,
+				const void *desc)
+{
+	u32 *ldt_entry = (u32 *)desc;
+	vmi_ops.write_idt_entry(dt, entry, ldt_entry[0], ldt_entry[2]);
+}
+
 static void vmi_load_sp0(struct tss_struct *tss,
 				   struct thread_struct *thread)
 {
@@ -805,7 +813,8 @@ static inline int __init activate_vmi(void)
 	para_fill(pv_cpu_ops.store_idt, GetIDT);
 	para_fill(pv_cpu_ops.store_tr, GetTR);
 	pv_cpu_ops.load_tls = vmi_load_tls;
-	para_fill(pv_cpu_ops.write_ldt_entry, WriteLDTEntry);
+	para_wrap(pv_cpu_ops.write_ldt_entry, vmi_write_ldt_entry,
+		  write_ldt_entry, WriteLDTEntry);
 	para_wrap(pv_cpu_ops.write_gdt_entry, vmi_write_gdt_entry,
 		  write_gdt_entry, WriteGDTEntry);
 	para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry,
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 7f98c63f6381..72dd14d0685c 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -357,11 +357,11 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu)
 }
 
 static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
-				u32 low, u32 high)
+				const void *ptr)
 {
 	unsigned long lp = (unsigned long)&dt[entrynum];
 	xmaddr_t mach_lp = virt_to_machine(lp);
-	u64 entry = (u64)high << 32 | low;
+	u64 entry = *(u64 *)ptr;
 
 	preempt_disable();
 
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 41d8214c6173..92a72b0381e2 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -69,12 +69,19 @@ static inline void pack_gate(gate_desc *gate,
 #define load_TLS(t, cpu) native_load_tls(t, cpu)
 #define set_ldt native_set_ldt
 
-#define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_ldt_entry(dt, entry, desc) \
+				native_write_ldt_entry(dt, entry, desc)
 #define write_gdt_entry(dt, entry, desc, type) \
 				native_write_gdt_entry(dt, entry, desc, type)
 #define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
 #endif
 
+static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry,
+					  const void *desc)
+{
+	memcpy(&ldt[entry], desc, sizeof(struct desc_struct));
+}
+
 static inline void native_write_idt_entry(gate_desc *idt, int entry,
 					  const gate_desc *gate)
 {
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index ba7fb87d10f3..7cdd3f0420fd 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -34,12 +34,9 @@ extern gate_desc idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
 static inline void write_ldt_entry(struct desc_struct *ldt,
-				   int entry, u32 entry_low, u32 entry_high)
+				   int entry, void *ptr)
 {
-	__u32 *lp = (__u32 *)((entry << 3) + (char *)ldt);
-
-	lp[0] = entry_low;
-	lp[1] = entry_high;
+	memcpy(&ldt[entry], ptr, 8);
 }
 
 /* the cpu gdt accessor */
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 3f2abf295e2a..4f23f434a1f3 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -96,8 +96,8 @@ struct pv_cpu_ops {
 	void (*set_ldt)(const void *desc, unsigned entries);
 	unsigned long (*store_tr)(void);
 	void (*load_tls)(struct thread_struct *t, unsigned int cpu);
-	void (*write_ldt_entry)(struct desc_struct *,
-				int entrynum, u32 low, u32 high);
+	void (*write_ldt_entry)(struct desc_struct *ldt, int entrynum,
+				const void *desc);
 	void (*write_gdt_entry)(struct desc_struct *,
 				int entrynum, const void *desc, int size);
 	void (*write_idt_entry)(gate_desc *,
@@ -660,9 +660,11 @@ static inline void load_TLS(struct thread_struct *t, unsigned cpu)
 {
 	PVOP_VCALL2(pv_cpu_ops.load_tls, t, cpu);
 }
-static inline void write_ldt_entry(void *dt, int entry, u32 low, u32 high)
+
+static inline void write_ldt_entry(struct desc_struct *dt, int entry,
+				   const void *desc)
 {
-	PVOP_VCALL4(pv_cpu_ops.write_ldt_entry, dt, entry, low, high);
+	PVOP_VCALL3(pv_cpu_ops.write_ldt_entry, dt, entry, desc);
 }
 
 static inline void write_gdt_entry(struct desc_struct *dt, int entry,
-- 
cgit v1.2.3


From 59fbed775aa5a6f352bfbde6b40508b541086b7e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: move constants to desc_defs.h

this patch moves constant definitions regarding descriptor types
from desc_32.h to desc_defs.h. The change from defines to enum
to comply with previous versions in desc_defs.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_32.h   | 8 --------
 include/asm-x86/desc_defs.h | 5 +++++
 2 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 92a72b0381e2..14238df87c7d 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -44,14 +44,6 @@ static inline void pack_gate(gate_desc *gate,
 	gate->b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
 }
 
-#define DESCTYPE_LDT 	0x82	/* present, system, DPL-0, LDT */
-#define DESCTYPE_TSS 	0x89	/* present, system, DPL-0, 32-bit TSS */
-#define DESCTYPE_TASK	0x85	/* present, system, DPL-0, task gate */
-#define DESCTYPE_INT	0x8e	/* present, system, DPL-0, interrupt gate */
-#define DESCTYPE_TRAP	0x8f	/* present, system, DPL-0, trap gate */
-#define DESCTYPE_DPL3	0x60	/* DPL-3 */
-#define DESCTYPE_S	0x10	/* !system */
-
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index c811f1e0829f..79fe59fc50ec 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -55,6 +55,11 @@ struct gate_struct64 {
 enum {
 	DESC_TSS = 0x9,
 	DESC_LDT = 0x2,
+	DESCTYPE_TASK = 0x85,   /* present, system, DPL-0, task gate */
+	DESCTYPE_INT =  0x8e,   /* present, system, DPL-0, interrupt gate */
+	DESCTYPE_TRAP = 0x8f,   /* present, system, DPL-0, trap gate */
+	DESCTYPE_DPL3 = 0x60,   /* DPL-3 */
+	DESCTYPE_S =	0x10,	/* !system */
 };
 
 // LDT or TSS descriptor in the GDT. 16 bytes.
-- 
cgit v1.2.3


From 881c2975d02b6575c417a6eee40868d60a9ecba9 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: unify non-paravirt parts of desc.h

This patch unifies the non-paravirt part of desc_{32,64}.h into
desc.h. Most of it, is simply common code, that is moved to
the shared header. The only exception is the set_ldt_desc in desc_64.h,
which is changed - included its name - to accomodate for the way
the ldt is set up in i386.

Also, constant definitions used in desc_32.h are moved to desc_defs.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc.h           | 65 +++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/desc_32.h        | 58 -----------------------------------
 include/asm-x86/desc_64.h        | 55 +++++++---------------------------
 include/asm-x86/mmu_context_64.h |  4 +--
 4 files changed, 76 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index 47086d2d9298..99c4adc851e2 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -4,6 +4,7 @@
 #ifndef __ASSEMBLY__
 #include <asm/desc_defs.h>
 #include <asm/ldt.h>
+#include <asm/mmu.h>
 
 static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
 {
@@ -23,7 +24,8 @@ static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
 	desc->base2 = (info->base_addr & 0xff000000) >> 24;
 }
 
-#endif
+extern struct desc_ptr idt_descr;
+extern gate_desc idt_table[];
 
 #ifdef CONFIG_X86_32
 # include "desc_32.h"
@@ -31,4 +33,65 @@ static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
 # include "desc_64.h"
 #endif
 
+#define _LDT_empty(info) (\
+	(info)->base_addr	== 0	&& \
+	(info)->limit		== 0	&& \
+	(info)->contents	== 0	&& \
+	(info)->read_exec_only	== 1	&& \
+	(info)->seg_32bit	== 0	&& \
+	(info)->limit_in_pages	== 0	&& \
+	(info)->seg_not_present	== 1	&& \
+	(info)->useable		== 0)
+
+#ifdef CONFIG_X86_64
+#define LDT_empty(info) (_LDT_empty(info) && ((info)->lm == 0))
+#else
+#define LDT_empty(info) (_LDT_empty(info))
+#endif
+
+static inline void clear_LDT(void)
+{
+	set_ldt(NULL, 0);
+}
+
+/*
+ * load one particular LDT into the current CPU
+ */
+static inline void load_LDT_nolock(mm_context_t *pc)
+{
+	set_ldt(pc->ldt, pc->size);
+}
+
+static inline void load_LDT(mm_context_t *pc)
+{
+	preempt_disable();
+	load_LDT_nolock(pc);
+	preempt_enable();
+}
+
+#else
+/*
+ * GET_DESC_BASE reads the descriptor base of the specified segment.
+ *
+ * Args:
+ *    idx - descriptor index
+ *    gdt - GDT pointer
+ *    base - 32bit register to which the base will be written
+ *    lo_w - lo word of the "base" register
+ *    lo_b - lo byte of the "base" register
+ *    hi_b - hi byte of the low word of the "base" register
+ *
+ * Example:
+ *    GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah)
+ *    Will read the base address of GDT_ENTRY_ESPFIX_SS and put it into %eax.
+ */
+#define GET_DESC_BASE(idx, gdt, base, lo_w, lo_b, hi_b) \
+	movb idx*8+4(gdt), lo_b; \
+	movb idx*8+7(gdt), hi_b; \
+	shll $16, base; \
+	movw idx*8+2(gdt), lo_w;
+
+
+#endif /* __ASSEMBLY__ */
+
 #endif
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 14238df87c7d..01415ad35fdf 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -11,8 +11,6 @@
 #include <linux/smp.h>
 #include <linux/percpu.h>
 
-#include <asm/mmu.h>
-
 struct gdt_page
 {
 	struct desc_struct gdt[GDT_ENTRIES];
@@ -24,8 +22,6 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 	return per_cpu(gdt_page, cpu).gdt;
 }
 
-extern struct desc_ptr idt_descr;
-extern gate_desc idt_table[];
 extern void set_intr_gate(unsigned int irq, void * addr);
 
 static inline void pack_descriptor(struct desc_struct *desc,
@@ -172,36 +168,6 @@ static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const vo
 
 #define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
 
-#define LDT_empty(info) (\
-	(info)->base_addr	== 0	&& \
-	(info)->limit		== 0	&& \
-	(info)->contents	== 0	&& \
-	(info)->read_exec_only	== 1	&& \
-	(info)->seg_32bit	== 0	&& \
-	(info)->limit_in_pages	== 0	&& \
-	(info)->seg_not_present	== 1	&& \
-	(info)->useable		== 0	)
-
-static inline void clear_LDT(void)
-{
-	set_ldt(NULL, 0);
-}
-
-/*
- * load one particular LDT into the current CPU
- */
-static inline void load_LDT_nolock(mm_context_t *pc)
-{
-	set_ldt(pc->ldt, pc->size);
-}
-
-static inline void load_LDT(mm_context_t *pc)
-{
-	preempt_disable();
-	load_LDT_nolock(pc);
-	preempt_enable();
-}
-
 static inline unsigned long get_desc_base(unsigned long *desc)
 {
 	unsigned long base;
@@ -210,30 +176,6 @@ static inline unsigned long get_desc_base(unsigned long *desc)
 		(desc[1] & 0xff000000);
 	return base;
 }
-
-#else /* __ASSEMBLY__ */
-
-/*
- * GET_DESC_BASE reads the descriptor base of the specified segment.
- *
- * Args:
- *    idx - descriptor index
- *    gdt - GDT pointer
- *    base - 32bit register to which the base will be written
- *    lo_w - lo word of the "base" register
- *    lo_b - lo byte of the "base" register
- *    hi_b - hi byte of the low word of the "base" register
- *
- * Example:
- *    GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah)
- *    Will read the base address of GDT_ENTRY_ESPFIX_SS and put it into %eax.
- */
-#define GET_DESC_BASE(idx, gdt, base, lo_w, lo_b, hi_b) \
-	movb idx*8+4(gdt), lo_b; \
-	movb idx*8+7(gdt), hi_b; \
-	shll $16, base; \
-	movw idx*8+2(gdt), lo_w;
-
 #endif /* !__ASSEMBLY__ */
 
 #endif
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 7cdd3f0420fd..c037c7d55b92 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -9,16 +9,13 @@
 
 #include <linux/string.h>
 #include <linux/smp.h>
-#include <asm/desc_defs.h>
 
 #include <asm/segment.h>
-#include <asm/mmu.h>
 
 extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
 
 #define load_TR_desc() asm volatile("ltr %w0"::"r" (GDT_ENTRY_TSS*8))
 #define load_LDT_desc() asm volatile("lldt %w0"::"r" (GDT_ENTRY_LDT*8))
-#define clear_LDT()  asm volatile("lldt %w0"::"r" (0))
 
 static inline unsigned long __store_tr(void)
 {
@@ -30,7 +27,6 @@ static inline unsigned long __store_tr(void)
 
 #define store_tr(tr) (tr) = __store_tr()
 
-extern gate_desc idt_table[];
 extern struct desc_ptr cpu_gdt_descr[];
 
 static inline void write_ldt_entry(struct desc_struct *ldt,
@@ -138,22 +134,18 @@ static inline void set_tss_desc(unsigned cpu, void *addr)
 		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
 }
 
-static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
+static inline void set_ldt(void *addr, int entries)
 {
-	set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT],
-			     (unsigned long)addr, DESC_LDT, size * 8 - 1);
-}
+	if (likely(entries == 0))
+		__asm__ __volatile__("lldt %w0"::"q" (0));
+	else {
+		unsigned cpu = smp_processor_id();
 
-#define LDT_empty(info) (\
-	(info)->base_addr	== 0	&& \
-	(info)->limit		== 0	&& \
-	(info)->contents	== 0	&& \
-	(info)->read_exec_only	== 1	&& \
-	(info)->seg_32bit	== 0	&& \
-	(info)->limit_in_pages	== 0	&& \
-	(info)->seg_not_present	== 1	&& \
-	(info)->useable		== 0	&& \
-	(info)->lm		== 0)
+		set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT],
+			     (unsigned long)addr, DESC_LDT, entries * 8 - 1);
+		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
+	}
+}
 
 static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
@@ -164,32 +156,6 @@ static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 		gdt[i] = t->tls_array[i];
 }
 
-/*
- * load one particular LDT into the current CPU
- */
-static inline void load_LDT_nolock(mm_context_t *pc, int cpu)
-{
-	int count = pc->size;
-
-	if (likely(!count)) {
-		clear_LDT();
-		return;
-	}
-
-	set_ldt_desc(cpu, pc->ldt, count);
-	load_LDT_desc();
-}
-
-static inline void load_LDT(mm_context_t *pc)
-{
-	int cpu = get_cpu();
-
-	load_LDT_nolock(pc, cpu);
-	put_cpu();
-}
-
-extern struct desc_ptr idt_descr;
-
 static inline unsigned long get_desc_base(const void *ptr)
 {
 	const u32 *desc = ptr;
@@ -199,7 +165,6 @@ static inline unsigned long get_desc_base(const void *ptr)
 		(desc[1] & 0xff000000);
 	return base;
 }
-
 #endif /* !__ASSEMBLY__ */
 
 #endif
diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h
index 29f95c3acc28..98bfe43dd806 100644
--- a/include/asm-x86/mmu_context_64.h
+++ b/include/asm-x86/mmu_context_64.h
@@ -43,7 +43,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 		load_cr3(next->pgd);
 
 		if (unlikely(next->context.ldt != prev->context.ldt)) 
-			load_LDT_nolock(&next->context, cpu);
+			load_LDT_nolock(&next->context);
 	}
 #ifdef CONFIG_SMP
 	else {
@@ -56,7 +56,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 			 * to make sure to use no freed page tables.
 			 */
 			load_cr3(next->pgd);
-			load_LDT_nolock(&next->context, cpu);
+			load_LDT_nolock(&next->context);
 		}
 	}
 #endif
-- 
cgit v1.2.3


From 26048d75e8d6c840742468667f4a7ab8c2df74c9 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: use the same data type for tls_array.

This patch changes the type of tls_array in x86_64 to
a desc_struct. Now, both i386 and x86_64 tls_array have
the same type, and code accessing it can be shared.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc_64.h      | 2 +-
 include/asm-x86/processor_64.h | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index c037c7d55b92..448f96ed973f 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -150,7 +150,7 @@ static inline void set_ldt(void *addr, int entries)
 static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
 	unsigned int i;
-	u64 *gdt = (u64 *)(get_cpu_gdt_table(cpu) + GDT_ENTRY_TLS_MIN);
+	struct desc_struct *gdt = (get_cpu_gdt_table(cpu) + GDT_ENTRY_TLS_MIN);
 
 	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
 		gdt[i] = t->tls_array[i];
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 7b7f8a142e20..a0a9e5515097 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -19,6 +19,7 @@
 #include <linux/personality.h>
 #include <linux/cpumask.h>
 #include <asm/processor-flags.h>
+#include <asm/desc_defs.h>
 
 #define TF_MASK		0x00000100
 #define IF_MASK		0x00000200
@@ -244,7 +245,7 @@ struct thread_struct {
  *               goes into MSR_IA32_DS_AREA */
 	unsigned long	ds_area_msr;
 /* cached TLS descriptors. */
-	u64 tls_array[GDT_ENTRY_TLS_ENTRIES];
+	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
 } __attribute__((aligned(16)));
 
 #define INIT_THREAD  { \
-- 
cgit v1.2.3


From cc6978528cbd475d952e0eb5073375839dfb600e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: modify get_desc_base

This patch makes get_desc_base() receive a struct desc_struct,
and then uses its internal fields to compute the base address.
This is done at both i386 and x86_64, and then it is moved
to common header

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/tls.c     | 2 +-
 arch/x86/mm/fault_32.c    | 2 +-
 include/asm-x86/desc.h    | 5 +++++
 include/asm-x86/desc_32.h | 8 --------
 include/asm-x86/desc_64.h | 9 ---------
 5 files changed, 7 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c
index 74d2b65a82eb..98f428be8e8c 100644
--- a/arch/x86/kernel/tls.c
+++ b/arch/x86/kernel/tls.c
@@ -112,7 +112,7 @@ int do_get_thread_area(struct task_struct *p, int idx,
 
 	memset(&info, 0, sizeof(struct user_desc));
 	info.entry_number = idx;
-	info.base_addr = get_desc_base((void *)desc);
+	info.base_addr = get_desc_base((struct desc_struct *)desc);
 	info.limit = GET_LIMIT(desc);
 	info.seg_32bit = GET_32BIT(desc);
 	info.contents = GET_CONTENTS(desc);
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 6056c6d71835..ef5ab2b925c4 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -115,7 +115,7 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs,
 	}
 
 	/* Decode the code segment base from the descriptor */
-	base = get_desc_base((unsigned long *)desc);
+	base = get_desc_base((struct desc_struct *)desc);
 
 	if (seg & (1<<2)) { 
 		mutex_unlock(&current->mm->context.lock);
diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index 99c4adc851e2..a6fdd7c7b6b2 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -69,6 +69,11 @@ static inline void load_LDT(mm_context_t *pc)
 	preempt_enable();
 }
 
+static inline unsigned long get_desc_base(struct desc_struct *desc)
+{
+	return desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24);
+}
+
 #else
 /*
  * GET_DESC_BASE reads the descriptor base of the specified segment.
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 01415ad35fdf..8450c2a99c3a 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -168,14 +168,6 @@ static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const vo
 
 #define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
 
-static inline unsigned long get_desc_base(unsigned long *desc)
-{
-	unsigned long base;
-	base = ((desc[0] >> 16)  & 0x0000ffff) |
-		((desc[1] << 16) & 0x00ff0000) |
-		(desc[1] & 0xff000000);
-	return base;
-}
 #endif /* !__ASSEMBLY__ */
 
 #endif
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 448f96ed973f..a7a6c301c6bc 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -156,15 +156,6 @@ static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 		gdt[i] = t->tls_array[i];
 }
 
-static inline unsigned long get_desc_base(const void *ptr)
-{
-	const u32 *desc = ptr;
-	unsigned long base;
-	base = ((desc[0] >> 16)  & 0x0000ffff) |
-		((desc[1] << 16) & 0x00ff0000) |
-		(desc[1] & 0xff000000);
-	return base;
-}
 #endif /* !__ASSEMBLY__ */
 
 #endif
-- 
cgit v1.2.3


From 54cd0eac7286b83ef1a657d2dddd74e0556209e7 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: unify paravirt pieces of descriptor handling

With the types used to access descriptors in x86_64 and i386
now being the same, the code that effectively handles them can
now be easily shared. This patch moves the paravirt part of
desc_32.h into desc.h, and then, we get paravirt support in x86_64
for free.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc.h      | 169 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/desc_32.h   | 130 ----------------------------------
 include/asm-x86/desc_64.h   | 104 +++------------------------
 include/asm-x86/desc_defs.h |   6 +-
 4 files changed, 183 insertions(+), 226 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index a6fdd7c7b6b2..e61a5a38caba 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -5,6 +5,7 @@
 #include <asm/desc_defs.h>
 #include <asm/ldt.h>
 #include <asm/mmu.h>
+#include <linux/smp.h>
 
 static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
 {
@@ -27,6 +28,174 @@ static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
 extern struct desc_ptr idt_descr;
 extern gate_desc idt_table[];
 
+#ifdef CONFIG_X86_64
+extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
+extern struct desc_ptr cpu_gdt_descr[];
+/* the cpu gdt accessor */
+#define get_cpu_gdt_table(x) ((struct desc_struct *)cpu_gdt_descr[x].address)
+#else
+struct gdt_page {
+	struct desc_struct gdt[GDT_ENTRIES];
+} __attribute__((aligned(PAGE_SIZE)));
+DECLARE_PER_CPU(struct gdt_page, gdt_page);
+
+static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
+{
+	return per_cpu(gdt_page, cpu).gdt;
+}
+#endif
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define load_TR_desc() native_load_tr_desc()
+#define load_gdt(dtr) native_load_gdt(dtr)
+#define load_idt(dtr) native_load_idt(dtr)
+#define load_tr(tr) __asm__ __volatile("ltr %0"::"m" (tr))
+#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"m" (ldt))
+
+#define store_gdt(dtr) native_store_gdt(dtr)
+#define store_idt(dtr) native_store_idt(dtr)
+#define store_tr(tr) (tr = native_store_tr())
+#define store_ldt(ldt) __asm__ ("sldt %0":"=m" (ldt))
+
+#define load_TLS(t, cpu) native_load_tls(t, cpu)
+#define set_ldt native_set_ldt
+
+#define write_ldt_entry(dt, entry, desc) \
+				native_write_ldt_entry(dt, entry, desc)
+#define write_gdt_entry(dt, entry, desc, type) \
+				native_write_gdt_entry(dt, entry, desc, type)
+#define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
+#endif
+
+static inline void native_write_idt_entry(gate_desc *idt, int entry,
+					  const gate_desc *gate)
+{
+	memcpy(&idt[entry], gate, sizeof(*gate));
+}
+
+static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry,
+					  const void *desc)
+{
+	memcpy(&ldt[entry], desc, 8);
+}
+
+static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry,
+					  const void *desc, int type)
+{
+	unsigned int size;
+	switch (type) {
+	case DESC_TSS:
+		size = sizeof(tss_desc);
+		break;
+	case DESC_LDT:
+		size = sizeof(ldt_desc);
+		break;
+	default:
+		size = sizeof(struct desc_struct);
+		break;
+	}
+	memcpy(&gdt[entry], desc, size);
+}
+
+static inline void set_tssldt_descriptor(struct ldttss_desc64 *d,
+					 unsigned long tss, unsigned type,
+					 unsigned size)
+{
+	memset(d, 0, sizeof(*d));
+	d->limit0 = size & 0xFFFF;
+	d->base0 = PTR_LOW(tss);
+	d->base1 = PTR_MIDDLE(tss) & 0xFF;
+	d->type = type;
+	d->p = 1;
+	d->limit1 = (size >> 16) & 0xF;
+	d->base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
+	d->base3 = PTR_HIGH(tss);
+}
+
+static inline void pack_descriptor(struct desc_struct *desc, unsigned long base,
+				   unsigned long limit, unsigned char type,
+				   unsigned char flags)
+{
+	desc->a = ((base & 0xffff) << 16) | (limit & 0xffff);
+	desc->b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
+		  (limit & 0x000f0000) | ((type & 0xff) << 8) |
+		  ((flags & 0xf) << 20);
+	desc->p = 1;
+}
+
+static inline void pack_ldt(ldt_desc *ldt, unsigned long addr,
+			   unsigned size)
+{
+
+#ifdef CONFIG_X86_64
+		set_tssldt_descriptor(ldt,
+			     addr, DESC_LDT, size);
+#else
+		pack_descriptor(ldt, (unsigned long)addr,
+				size,
+				0x80 | DESC_LDT, 0);
+#endif
+}
+
+static inline void native_set_ldt(const void *addr, unsigned int entries)
+{
+	if (likely(entries == 0))
+		__asm__ __volatile__("lldt %w0"::"q" (0));
+	else {
+		unsigned cpu = smp_processor_id();
+		ldt_desc ldt;
+
+		pack_ldt(&ldt, (unsigned long)addr,
+				entries * sizeof(ldt) - 1);
+		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT,
+				&ldt, DESC_LDT);
+		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
+	}
+}
+
+static inline void native_load_tr_desc(void)
+{
+	asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8));
+}
+
+static inline void native_load_gdt(const struct desc_ptr *dtr)
+{
+	asm volatile("lgdt %0"::"m" (*dtr));
+}
+
+static inline void native_load_idt(const struct desc_ptr *dtr)
+{
+	asm volatile("lidt %0"::"m" (*dtr));
+}
+
+static inline void native_store_gdt(struct desc_ptr *dtr)
+{
+	asm volatile("sgdt %0":"=m" (*dtr));
+}
+
+static inline void native_store_idt(struct desc_ptr *dtr)
+{
+	asm volatile("sidt %0":"=m" (*dtr));
+}
+
+static inline unsigned long native_store_tr(void)
+{
+	unsigned long tr;
+	asm volatile("str %0":"=r" (tr));
+	return tr;
+}
+
+static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
+{
+	unsigned int i;
+	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
+
+	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
+		gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i];
+}
+
 #ifdef CONFIG_X86_32
 # include "desc_32.h"
 #else
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 8450c2a99c3a..4bf20b7dd741 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -8,31 +8,10 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/preempt.h>
-#include <linux/smp.h>
 #include <linux/percpu.h>
 
-struct gdt_page
-{
-	struct desc_struct gdt[GDT_ENTRIES];
-} __attribute__((aligned(PAGE_SIZE)));
-DECLARE_PER_CPU(struct gdt_page, gdt_page);
-
-static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
-{
-	return per_cpu(gdt_page, cpu).gdt;
-}
-
 extern void set_intr_gate(unsigned int irq, void * addr);
 
-static inline void pack_descriptor(struct desc_struct *desc,
-	unsigned long base, unsigned long limit, unsigned char type, unsigned char flags)
-{
-	desc->a = ((base & 0xffff) << 16) | (limit & 0xffff);
-	desc->b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
-		(limit & 0x000f0000) | ((type & 0xff) << 8) | ((flags & 0xf) << 20);
-	desc->p = 1;
-}
-
 static inline void pack_gate(gate_desc *gate,
 	unsigned long base, unsigned short seg, unsigned char type, unsigned char flags)
 {
@@ -40,115 +19,6 @@ static inline void pack_gate(gate_desc *gate,
 	gate->b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
 }
 
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define load_TR_desc() native_load_tr_desc()
-#define load_gdt(dtr) native_load_gdt(dtr)
-#define load_idt(dtr) native_load_idt(dtr)
-#define load_tr(tr) __asm__ __volatile("ltr %0"::"m" (tr))
-#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"m" (ldt))
-
-#define store_gdt(dtr) native_store_gdt(dtr)
-#define store_idt(dtr) native_store_idt(dtr)
-#define store_tr(tr) (tr = native_store_tr())
-#define store_ldt(ldt) __asm__ ("sldt %0":"=m" (ldt))
-
-#define load_TLS(t, cpu) native_load_tls(t, cpu)
-#define set_ldt native_set_ldt
-
-#define write_ldt_entry(dt, entry, desc) \
-				native_write_ldt_entry(dt, entry, desc)
-#define write_gdt_entry(dt, entry, desc, type) \
-				native_write_gdt_entry(dt, entry, desc, type)
-#define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
-#endif
-
-static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry,
-					  const void *desc)
-{
-	memcpy(&ldt[entry], desc, sizeof(struct desc_struct));
-}
-
-static inline void native_write_idt_entry(gate_desc *idt, int entry,
-					  const gate_desc *gate)
-{
-	memcpy(&idt[entry], gate, sizeof(*gate));
-}
-
-static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry,
-					  const void *desc, int type)
-{
-	memcpy(&gdt[entry], desc, sizeof(struct desc_struct));
-}
-
-static inline void write_dt_entry(struct desc_struct *dt,
-				  int entry, u32 entry_low, u32 entry_high)
-{
-	dt[entry].a = entry_low;
-	dt[entry].b = entry_high;
-}
-
-
-static inline void native_set_ldt(const void *addr, unsigned int entries)
-{
-	if (likely(entries == 0))
-		__asm__ __volatile__("lldt %w0"::"q" (0));
-	else {
-		unsigned cpu = smp_processor_id();
-		ldt_desc ldt;
-
-		pack_descriptor(&ldt, (unsigned long)addr,
-				entries * sizeof(struct desc_struct) - 1,
-				DESC_LDT, 0);
-		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT,
-				&ldt, DESC_LDT);
-		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
-	}
-}
-
-
-static inline void native_load_tr_desc(void)
-{
-	asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8));
-}
-
-static inline void native_load_gdt(const struct desc_ptr *dtr)
-{
-	asm volatile("lgdt %0"::"m" (*dtr));
-}
-
-static inline void native_load_idt(const struct desc_ptr *dtr)
-{
-	asm volatile("lidt %0"::"m" (*dtr));
-}
-
-static inline void native_store_gdt(struct desc_ptr *dtr)
-{
-	asm ("sgdt %0":"=m" (*dtr));
-}
-
-static inline void native_store_idt(struct desc_ptr *dtr)
-{
-	asm ("sidt %0":"=m" (*dtr));
-}
-
-static inline unsigned long native_store_tr(void)
-{
-	unsigned long tr;
-	asm ("str %0":"=r" (tr));
-	return tr;
-}
-
-static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
-{
-	unsigned int i;
-	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
-
-	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
-		gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i];
-}
-
 static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg)
 {
 	gate_desc g;
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index a7a6c301c6bc..b0290c45319c 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -8,47 +8,10 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/string.h>
-#include <linux/smp.h>
 
 #include <asm/segment.h>
 
-extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
-
-#define load_TR_desc() asm volatile("ltr %w0"::"r" (GDT_ENTRY_TSS*8))
-#define load_LDT_desc() asm volatile("lldt %w0"::"r" (GDT_ENTRY_LDT*8))
-
-static inline unsigned long __store_tr(void)
-{
-       unsigned long tr;
-
-       asm volatile ("str %w0":"=r" (tr));
-       return tr;
-}
-
-#define store_tr(tr) (tr) = __store_tr()
-
-extern struct desc_ptr cpu_gdt_descr[];
-
-static inline void write_ldt_entry(struct desc_struct *ldt,
-				   int entry, void *ptr)
-{
-	memcpy(&ldt[entry], ptr, 8);
-}
-
-/* the cpu gdt accessor */
-#define get_cpu_gdt_table(x) ((struct desc_struct *)cpu_gdt_descr[x].address)
-
-static inline void load_gdt(const struct desc_ptr *ptr)
-{
-	asm volatile("lgdt %w0"::"m" (*ptr));
-}
-
-static inline void store_gdt(struct desc_ptr *ptr)
-{
-       asm("sgdt %w0":"=m" (*ptr));
-}
-
-static inline void _set_gate(void *adr, unsigned type, unsigned long func,
+static inline void _set_gate(int gate, unsigned type, unsigned long func,
 			     unsigned dpl, unsigned ist)
 {
 	gate_desc s;
@@ -67,61 +30,37 @@ static inline void _set_gate(void *adr, unsigned type, unsigned long func,
 	 * does not need to be atomic because it is only done once at
 	 * setup time
 	 */
-	memcpy(adr, &s, 16);
+	write_idt_entry(idt_table, gate, &s);
 }
 
 static inline void set_intr_gate(int nr, void *func)
 {
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, 0);
+	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 0, 0);
 }
 
 static inline void set_intr_gate_ist(int nr, void *func, unsigned ist)
 {
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, ist);
+	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 0, ist);
 }
 
 static inline void set_system_gate(int nr, void *func)
 {
 	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, 0);
+	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 3, 0);
 }
 
 static inline void set_system_gate_ist(int nr, void *func, unsigned ist)
 {
-	_set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, ist);
-}
-
-static inline void load_idt(const struct desc_ptr *ptr)
-{
-	asm volatile("lidt %w0"::"m" (*ptr));
-}
-
-static inline void store_idt(struct desc_ptr *dtr)
-{
-       asm("sidt %w0":"=m" (*dtr));
-}
-
-static inline void set_tssldt_descriptor(void *ptr, unsigned long tss,
-					 unsigned type, unsigned size)
-{
-	struct ldttss_desc64 d;
-
-	memset(&d, 0, sizeof(d));
-	d.limit0 = size & 0xFFFF;
-	d.base0 = PTR_LOW(tss);
-	d.base1 = PTR_MIDDLE(tss) & 0xFF;
-	d.type = type;
-	d.p = 1;
-	d.limit1 = (size >> 16) & 0xF;
-	d.base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
-	d.base3 = PTR_HIGH(tss);
-	memcpy(ptr, &d, 16);
+	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 3, ist);
 }
 
 static inline void set_tss_desc(unsigned cpu, void *addr)
 {
+	struct desc_struct *d = get_cpu_gdt_table(cpu);
+	tss_desc tss;
+
 	/*
 	 * sizeof(unsigned long) coming from an extra "long" at the end
 	 * of the iobitmap. See tss_struct definition in processor.h
@@ -129,31 +68,10 @@ static inline void set_tss_desc(unsigned cpu, void *addr)
 	 * -1? seg base+limit should be pointing to the address of the
 	 * last valid byte
 	 */
-	set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS],
+	set_tssldt_descriptor(&tss,
 		(unsigned long)addr, DESC_TSS,
 		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
-}
-
-static inline void set_ldt(void *addr, int entries)
-{
-	if (likely(entries == 0))
-		__asm__ __volatile__("lldt %w0"::"q" (0));
-	else {
-		unsigned cpu = smp_processor_id();
-
-		set_tssldt_descriptor(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT],
-			     (unsigned long)addr, DESC_LDT, entries * 8 - 1);
-		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
-	}
-}
-
-static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
-{
-	unsigned int i;
-	struct desc_struct *gdt = (get_cpu_gdt_table(cpu) + GDT_ENTRY_TLS_MIN);
-
-	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
-		gdt[i] = t->tls_array[i];
+	write_gdt_entry(d, GDT_ENTRY_TSS, &tss, DESC_TSS);
 }
 
 #endif /* !__ASSEMBLY__ */
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index 79fe59fc50ec..ebb64fe3a450 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -48,9 +48,9 @@ struct gate_struct64 {
 	u32 zero1;
 } __attribute__((packed));
 
-#define PTR_LOW(x) ((unsigned long)(x) & 0xFFFF)
-#define PTR_MIDDLE(x) (((unsigned long)(x) >> 16) & 0xFFFF)
-#define PTR_HIGH(x) ((unsigned long)(x) >> 32)
+#define PTR_LOW(x) ((unsigned long long)(x) & 0xFFFF)
+#define PTR_MIDDLE(x) (((unsigned long long)(x) >> 16) & 0xFFFF)
+#define PTR_HIGH(x) ((unsigned long long)(x) >> 32)
 
 enum {
 	DESC_TSS = 0x9,
-- 
cgit v1.2.3


From 507f90c9f92592e7630b1c1e87bf92d2c9858cc6 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: move _set_gate and its users to a common location

This patch moves _set_gate and its users to desc.h. We can now
use common code for x86_64 and i386.

[ mingo@elte.hu: set_system_gate() fixes for nasty crashes. ]

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/traps_32.c  | 34 -----------------
 include/asm-x86/desc.h      | 92 +++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/desc_32.h   | 16 --------
 include/asm-x86/desc_64.h   | 45 ----------------------
 include/asm-x86/desc_defs.h |  5 +--
 5 files changed, 93 insertions(+), 99 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index c70c41fd710b..3065b3f41928 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -1102,40 +1102,6 @@ asmlinkage void math_emulate(long arg)
 
 #endif /* CONFIG_MATH_EMULATION */
 
-/*
- * This needs to use 'idt_table' rather than 'idt', and
- * thus use the _nonmapped_ version of the IDT, as the
- * Pentium F0 0F bugfix can have resulted in the mapped
- * IDT being write-protected.
- */
-void set_intr_gate(unsigned int n, void *addr)
-{
-	_set_gate(n, DESCTYPE_INT, addr, __KERNEL_CS);
-}
-
-/*
- * This routine sets up an interrupt gate at directory privilege level 3.
- */
-static inline void set_system_intr_gate(unsigned int n, void *addr)
-{
-	_set_gate(n, DESCTYPE_INT | DESCTYPE_DPL3, addr, __KERNEL_CS);
-}
-
-static void __init set_trap_gate(unsigned int n, void *addr)
-{
-	_set_gate(n, DESCTYPE_TRAP, addr, __KERNEL_CS);
-}
-
-static void __init set_system_gate(unsigned int n, void *addr)
-{
-	_set_gate(n, DESCTYPE_TRAP | DESCTYPE_DPL3, addr, __KERNEL_CS);
-}
-
-static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
-{
-	_set_gate(n, DESCTYPE_TASK, (void *)0, (gdt_entry<<3));
-}
-
 
 void __init trap_init(void)
 {
diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index e61a5a38caba..d75bc0634313 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -33,6 +33,22 @@ extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
 extern struct desc_ptr cpu_gdt_descr[];
 /* the cpu gdt accessor */
 #define get_cpu_gdt_table(x) ((struct desc_struct *)cpu_gdt_descr[x].address)
+
+static inline void pack_gate(gate_desc *gate, unsigned type, unsigned long func,
+			     unsigned dpl, unsigned ist, unsigned seg)
+{
+	gate->offset_low = PTR_LOW(func);
+	gate->segment = __KERNEL_CS;
+	gate->ist = ist;
+	gate->p = 1;
+	gate->dpl = dpl;
+	gate->zero0 = 0;
+	gate->zero1 = 0;
+	gate->type = type;
+	gate->offset_middle = PTR_MIDDLE(func);
+	gate->offset_high = PTR_HIGH(func);
+}
+
 #else
 struct gdt_page {
 	struct desc_struct gdt[GDT_ENTRIES];
@@ -43,6 +59,16 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 {
 	return per_cpu(gdt_page, cpu).gdt;
 }
+
+static inline void pack_gate(gate_desc *gate, unsigned char type,
+       unsigned long base, unsigned dpl, unsigned flags, unsigned short seg)
+
+{
+	gate->a = (seg << 16) | (base & 0xffff);
+	gate->b = (base & 0xffff0000) |
+		  (((0x80 | type | (dpl << 5)) & 0xff) << 8);
+}
+
 #endif
 
 #ifdef CONFIG_PARAVIRT
@@ -242,6 +268,72 @@ static inline unsigned long get_desc_base(struct desc_struct *desc)
 {
 	return desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24);
 }
+static inline void _set_gate(int gate, unsigned type, void *addr,
+			      unsigned dpl, unsigned ist, unsigned seg)
+{
+	gate_desc s;
+	pack_gate(&s, type, (unsigned long)addr, dpl, ist, seg);
+	/*
+	 * does not need to be atomic because it is only done once at
+	 * setup time
+	 */
+	write_idt_entry(idt_table, gate, &s);
+}
+
+/*
+ * This needs to use 'idt_table' rather than 'idt', and
+ * thus use the _nonmapped_ version of the IDT, as the
+ * Pentium F0 0F bugfix can have resulted in the mapped
+ * IDT being write-protected.
+ */
+static inline void set_intr_gate(unsigned int n, void *addr)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS);
+}
+
+/*
+ * This routine sets up an interrupt gate at directory privilege level 3.
+ */
+static inline void set_system_intr_gate(unsigned int n, void *addr)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS);
+}
+
+static inline void set_trap_gate(unsigned int n, void *addr)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS);
+}
+
+static inline void set_system_gate(unsigned int n, void *addr)
+{
+	BUG_ON((unsigned)n > 0xFF);
+#ifdef CONFIG_X86_32
+	_set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS);
+#else
+	_set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS);
+#endif
+}
+
+static inline void set_task_gate(unsigned int n, unsigned int gdt_entry)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_TASK, (void *)0, 0, 0, (gdt_entry<<3));
+}
+
+static inline void set_intr_gate_ist(int n, void *addr, unsigned ist)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS);
+}
+
+static inline void set_system_gate_ist(int n, void *addr, unsigned ist)
+{
+	BUG_ON((unsigned)n > 0xFF);
+	_set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS);
+}
 
 #else
 /*
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
index 4bf20b7dd741..3b112ec186a0 100644
--- a/include/asm-x86/desc_32.h
+++ b/include/asm-x86/desc_32.h
@@ -10,22 +10,6 @@
 #include <linux/preempt.h>
 #include <linux/percpu.h>
 
-extern void set_intr_gate(unsigned int irq, void * addr);
-
-static inline void pack_gate(gate_desc *gate,
-	unsigned long base, unsigned short seg, unsigned char type, unsigned char flags)
-{
-	gate->a = (seg << 16) | (base & 0xffff);
-	gate->b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
-}
-
-static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg)
-{
-	gate_desc g;
-	pack_gate(&g, (unsigned long)addr, seg, type, 0);
-	write_idt_entry(idt_table, gate, &g);
-}
-
 static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
 {
 	tss_desc tss;
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index b0290c45319c..6bc92e6e5cc3 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -11,51 +11,6 @@
 
 #include <asm/segment.h>
 
-static inline void _set_gate(int gate, unsigned type, unsigned long func,
-			     unsigned dpl, unsigned ist)
-{
-	gate_desc s;
-
-	s.offset_low = PTR_LOW(func);
-	s.segment = __KERNEL_CS;
-	s.ist = ist;
-	s.p = 1;
-	s.dpl = dpl;
-	s.zero0 = 0;
-	s.zero1 = 0;
-	s.type = type;
-	s.offset_middle = PTR_MIDDLE(func);
-	s.offset_high = PTR_HIGH(func);
-	/*
-	 * does not need to be atomic because it is only done once at
-	 * setup time
-	 */
-	write_idt_entry(idt_table, gate, &s);
-}
-
-static inline void set_intr_gate(int nr, void *func)
-{
-	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 0, 0);
-}
-
-static inline void set_intr_gate_ist(int nr, void *func, unsigned ist)
-{
-	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 0, ist);
-}
-
-static inline void set_system_gate(int nr, void *func)
-{
-	BUG_ON((unsigned)nr > 0xFF);
-	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 3, 0);
-}
-
-static inline void set_system_gate_ist(int nr, void *func, unsigned ist)
-{
-	_set_gate(nr, GATE_INTERRUPT, (unsigned long) func, 3, ist);
-}
-
 static inline void set_tss_desc(unsigned cpu, void *addr)
 {
 	struct desc_struct *d = get_cpu_gdt_table(cpu);
diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index ebb64fe3a450..e33f078b3e54 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -36,6 +36,7 @@ enum {
 	GATE_INTERRUPT = 0xE,
 	GATE_TRAP = 0xF,
 	GATE_CALL = 0xC,
+	GATE_TASK = 0x5,
 };
 
 // 16byte gate
@@ -55,10 +56,6 @@ struct gate_struct64 {
 enum {
 	DESC_TSS = 0x9,
 	DESC_LDT = 0x2,
-	DESCTYPE_TASK = 0x85,   /* present, system, DPL-0, task gate */
-	DESCTYPE_INT =  0x8e,   /* present, system, DPL-0, interrupt gate */
-	DESCTYPE_TRAP = 0x8f,   /* present, system, DPL-0, trap gate */
-	DESCTYPE_DPL3 = 0x60,   /* DPL-3 */
 	DESCTYPE_S =	0x10,	/* !system */
 };
 
-- 
cgit v1.2.3


From c81c6ca45a69478c7877b729af1942d2b80ef582 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: x86: unify set_tss_desc

This patch unifies the set_tss_desc between i386 and x86_64,
which can now have a common implementation. After the old
functions are removed from desc_{32,64}.h, nothing important is
left, and the files can be removed.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc.h    | 39 +++++++++++++++++++++++++++++++++------
 include/asm-x86/desc_32.h | 27 ---------------------------
 include/asm-x86/desc_64.h | 33 ---------------------------------
 3 files changed, 33 insertions(+), 66 deletions(-)
 delete mode 100644 include/asm-x86/desc_32.h

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index d75bc0634313..c31715cf2218 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -165,6 +165,39 @@ static inline void pack_ldt(ldt_desc *ldt, unsigned long addr,
 #endif
 }
 
+static inline void pack_tss(tss_desc *tss, unsigned long addr,
+			   unsigned size, unsigned entry)
+{
+#ifdef CONFIG_X86_64
+		set_tssldt_descriptor(tss,
+			     addr, entry, size);
+#else
+		pack_descriptor(tss, (unsigned long)addr,
+				size,
+				0x80 | entry, 0);
+#endif
+}
+
+static inline void __set_tss_desc(unsigned cpu, unsigned int entry, void *addr)
+{
+	struct desc_struct *d = get_cpu_gdt_table(cpu);
+	tss_desc tss;
+
+	/*
+	 * sizeof(unsigned long) coming from an extra "long" at the end
+	 * of the iobitmap. See tss_struct definition in processor.h
+	 *
+	 * -1? seg base+limit should be pointing to the address of the
+	 * last valid byte
+	 */
+	pack_tss(&tss, (unsigned long)addr,
+		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1,
+		DESC_TSS);
+	write_gdt_entry(d, entry, &tss, DESC_TSS);
+}
+
+#define set_tss_desc(cpu, addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+
 static inline void native_set_ldt(const void *addr, unsigned int entries)
 {
 	if (likely(entries == 0))
@@ -222,12 +255,6 @@ static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
 		gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i];
 }
 
-#ifdef CONFIG_X86_32
-# include "desc_32.h"
-#else
-# include "desc_64.h"
-#endif
-
 #define _LDT_empty(info) (\
 	(info)->base_addr	== 0	&& \
 	(info)->limit		== 0	&& \
diff --git a/include/asm-x86/desc_32.h b/include/asm-x86/desc_32.h
deleted file mode 100644
index 3b112ec186a0..000000000000
--- a/include/asm-x86/desc_32.h
+++ /dev/null
@@ -1,27 +0,0 @@
-#ifndef __ARCH_DESC_H
-#define __ARCH_DESC_H
-
-#include <asm/ldt.h>
-#include <asm/segment.h>
-#include <asm/desc_defs.h>
-
-#ifndef __ASSEMBLY__
-
-#include <linux/preempt.h>
-#include <linux/percpu.h>
-
-static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
-{
-	tss_desc tss;
-	pack_descriptor(&tss, (unsigned long)addr,
-			offsetof(struct tss_struct, __cacheline_filler) - 1,
-			DESC_TSS, 0);
-	write_gdt_entry(get_cpu_gdt_table(cpu), entry, &tss, DESC_TSS);
-}
-
-
-#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
-
-#endif /* !__ASSEMBLY__ */
-
-#endif
diff --git a/include/asm-x86/desc_64.h b/include/asm-x86/desc_64.h
index 6bc92e6e5cc3..8b137891791f 100644
--- a/include/asm-x86/desc_64.h
+++ b/include/asm-x86/desc_64.h
@@ -1,34 +1 @@
-/* Written 2000 by Andi Kleen */
-#ifndef __ARCH_DESC_H
-#define __ARCH_DESC_H
 
-#include <linux/threads.h>
-#include <asm/ldt.h>
-
-#ifndef __ASSEMBLY__
-
-#include <linux/string.h>
-
-#include <asm/segment.h>
-
-static inline void set_tss_desc(unsigned cpu, void *addr)
-{
-	struct desc_struct *d = get_cpu_gdt_table(cpu);
-	tss_desc tss;
-
-	/*
-	 * sizeof(unsigned long) coming from an extra "long" at the end
-	 * of the iobitmap. See tss_struct definition in processor.h
-	 *
-	 * -1? seg base+limit should be pointing to the address of the
-	 * last valid byte
-	 */
-	set_tssldt_descriptor(&tss,
-		(unsigned long)addr, DESC_TSS,
-		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
-	write_gdt_entry(d, GDT_ENTRY_TSS, &tss, DESC_TSS);
-}
-
-#endif /* !__ASSEMBLY__ */
-
-#endif
-- 
cgit v1.2.3


From 332540d5fe7f3370bb860090e4b906340a2e2afb Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:14 +0100
Subject: UML: remove remaining FASTCALL uses

With the x86 removal, FASTCALL is always empty now.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/um/kernel/ksyms.c   | 4 ++--
 include/asm-um/linkage.h | 1 -
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c
index 1b388b41d95d..7c7142ba3bd7 100644
--- a/arch/um/kernel/ksyms.c
+++ b/arch/um/kernel/ksyms.c
@@ -71,10 +71,10 @@ EXPORT_SYMBOL(dump_thread);
 
 /* required for SMP */
 
-extern void FASTCALL( __write_lock_failed(rwlock_t *rw));
+extern void __write_lock_failed(rwlock_t *rw);
 EXPORT_SYMBOL(__write_lock_failed);
 
-extern void FASTCALL( __read_lock_failed(rwlock_t *rw));
+extern void __read_lock_failed(rwlock_t *rw);
 EXPORT_SYMBOL(__read_lock_failed);
 
 #endif
diff --git a/include/asm-um/linkage.h b/include/asm-um/linkage.h
index 78b862472b36..cdb3024a699a 100644
--- a/include/asm-um/linkage.h
+++ b/include/asm-um/linkage.h
@@ -6,7 +6,6 @@
 
 /* <linux/linkage.h> will pick sane defaults */
 #ifdef CONFIG_GPROF
-#undef FASTCALL
 #undef fastcall
 #endif
 
-- 
cgit v1.2.3


From 56c4da454de1264e381256f658f61b9ef690dd21 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:17 +0100
Subject: core: remove last users of empty FASTCALL macro

FASTCALL is always empty after the x86 removal.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 fs/aio.c                  | 2 +-
 include/asm-m32r/signal.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/aio.c b/fs/aio.c
index 9dec7d2d546e..8a37dbbf3437 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -397,7 +397,7 @@ void fastcall __put_ioctx(struct kioctx *ctx)
  * This prevents races between the aio code path referencing the
  * req (after submitting it) and aio_complete() freeing the req.
  */
-static struct kiocb *FASTCALL(__aio_get_req(struct kioctx *ctx));
+static struct kiocb *__aio_get_req(struct kioctx *ctx);
 static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx)
 {
 	struct kiocb *req = NULL;
diff --git a/include/asm-m32r/signal.h b/include/asm-m32r/signal.h
index 937258686ba5..1a607066bc64 100644
--- a/include/asm-m32r/signal.h
+++ b/include/asm-m32r/signal.h
@@ -157,7 +157,7 @@ typedef struct sigaltstack {
 #undef __HAVE_ARCH_SIG_BITOPS
 
 struct pt_regs;
-extern int FASTCALL(do_signal(struct pt_regs *regs, sigset_t *oldset));
+extern int do_signal(struct pt_regs *regs, sigset_t *oldset);
 
 #define ptrace_signal_deliver(regs, cookie)	do { } while (0)
 
-- 
cgit v1.2.3


From 341d8854d38fe6e7655f7748babae5c25a51aa70 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:17 +0100
Subject: x86: remove fastcall from include/asm-x86

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/apic.h         |  6 +++---
 include/asm-x86/hw_irq_32.h    | 14 +++++++-------
 include/asm-x86/mutex_32.h     |  7 +++----
 include/asm-x86/semaphore_32.h |  8 ++++----
 4 files changed, 17 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 423022759cb2..72bf09cf13ab 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -59,17 +59,17 @@ extern unsigned boot_cpu_id;
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
-static inline fastcall void native_apic_write(unsigned long reg, u32 v)
+static inline void native_apic_write(unsigned long reg, u32 v)
 {
 	*((volatile u32 *)(APIC_BASE + reg)) = v;
 }
 
-static inline fastcall void native_apic_write_atomic(unsigned long reg, u32 v)
+static inline void native_apic_write_atomic(unsigned long reg, u32 v)
 {
 	(void) xchg((u32*)(APIC_BASE + reg), v);
 }
 
-static inline fastcall u32 native_apic_read(unsigned long reg)
+static inline u32 native_apic_read(unsigned long reg)
 {
 	return *((volatile u32 *)(APIC_BASE + reg));
 }
diff --git a/include/asm-x86/hw_irq_32.h b/include/asm-x86/hw_irq_32.h
index 0bedbdf5e907..b93e35a708ac 100644
--- a/include/asm-x86/hw_irq_32.h
+++ b/include/asm-x86/hw_irq_32.h
@@ -29,16 +29,16 @@
 extern void (*interrupt[NR_IRQS])(void);
 
 #ifdef CONFIG_SMP
-fastcall void reschedule_interrupt(void);
-fastcall void invalidate_interrupt(void);
-fastcall void call_function_interrupt(void);
+void reschedule_interrupt(void);
+void invalidate_interrupt(void);
+void call_function_interrupt(void);
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
-fastcall void apic_timer_interrupt(void);
-fastcall void error_interrupt(void);
-fastcall void spurious_interrupt(void);
-fastcall void thermal_interrupt(void);
+void apic_timer_interrupt(void);
+void error_interrupt(void);
+void spurious_interrupt(void);
+void thermal_interrupt(void);
 #define platform_legacy_irq(irq)	((irq) < 16)
 #endif
 
diff --git a/include/asm-x86/mutex_32.h b/include/asm-x86/mutex_32.h
index 7a17d9e58ad6..bbeefb96ddfd 100644
--- a/include/asm-x86/mutex_32.h
+++ b/include/asm-x86/mutex_32.h
@@ -26,7 +26,7 @@ do {									\
 	unsigned int dummy;						\
 									\
 	typecheck(atomic_t *, count);					\
-	typecheck_fn(fastcall void (*)(atomic_t *), fail_fn);		\
+	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   decl (%%eax)	\n"			\
@@ -51,8 +51,7 @@ do {									\
  * or anything the slow path function returns
  */
 static inline int
-__mutex_fastpath_lock_retval(atomic_t *count,
-			     int fastcall (*fail_fn)(atomic_t *))
+__mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
 {
 	if (unlikely(atomic_dec_return(count) < 0))
 		return fail_fn(count);
@@ -78,7 +77,7 @@ do {									\
 	unsigned int dummy;						\
 									\
 	typecheck(atomic_t *, count);					\
-	typecheck_fn(fastcall void (*)(atomic_t *), fail_fn);		\
+	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   incl (%%eax)	\n"			\
diff --git a/include/asm-x86/semaphore_32.h b/include/asm-x86/semaphore_32.h
index 835c1d751a9f..cc826e85323f 100644
--- a/include/asm-x86/semaphore_32.h
+++ b/include/asm-x86/semaphore_32.h
@@ -83,10 +83,10 @@ static inline void init_MUTEX_LOCKED (struct semaphore *sem)
 	sema_init(sem, 0);
 }
 
-fastcall void __down_failed(void /* special register calling convention */);
-fastcall int  __down_failed_interruptible(void  /* params in registers */);
-fastcall int  __down_failed_trylock(void  /* params in registers */);
-fastcall void __up_wakeup(void /* special register calling convention */);
+void __down_failed(void /* special register calling convention */);
+int  __down_failed_interruptible(void  /* params in registers */);
+int  __down_failed_trylock(void  /* params in registers */);
+void __up_wakeup(void /* special register calling convention */);
 
 /*
  * This is ugly, but we want the default case to fall through.
-- 
cgit v1.2.3


From 1e160cc3f3b5c38bba79c58172e82c1e24934546 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 30 Jan 2008 13:31:17 +0100
Subject: x86: __vdso_getcpu() warning fix

arch/x86/vdso/vgetcpu.c: In function '__vdso_getcpu':
arch/x86/vdso/vgetcpu.c:22: warning: pointer targets in passing argument 1 of 'native_read_tscp' differ in signedness

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 040d3910d891..088652a62d58 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -13,7 +13,7 @@
 #include <asm/asm.h>
 #include <asm/errno.h>
 
-static inline unsigned long long native_read_tscp(int *aux)
+static inline unsigned long long native_read_tscp(unsigned int *aux)
 {
 	unsigned long low, high;
 	asm volatile (".byte 0x0f,0x01,0xf9"
-- 
cgit v1.2.3


From 634e49723b08d15f4cb9860ea86547fbd2eeae18 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 30 Jan 2008 13:31:17 +0100
Subject: uml: add asm-um/asm.h

git-x86.patch introduces asm-x86/asm.h.  UML needs to follow this with
an asm/asm.h of its own.

Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-um/asm.h | 6 ++++++
 1 file changed, 6 insertions(+)
 create mode 100644 include/asm-um/asm.h

(limited to 'include')

diff --git a/include/asm-um/asm.h b/include/asm-um/asm.h
new file mode 100644
index 000000000000..af1269a1e9eb
--- /dev/null
+++ b/include/asm-um/asm.h
@@ -0,0 +1,6 @@
+#ifndef __UM_ASM_H
+#define __UM_ASM_H
+
+#include "asm/arch/asm.h"
+
+#endif
-- 
cgit v1.2.3


From 5f561d3be8f0db54f9b4fc5cb5db05343f372431 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:31:17 +0100
Subject: x86: dummy placeholder for acpi/reboot.h

dummy placeholder for acpi/reboot.h

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/acpi/reboot.h | 9 +++++++++
 1 file changed, 9 insertions(+)
 create mode 100644 include/acpi/reboot.h

(limited to 'include')

diff --git a/include/acpi/reboot.h b/include/acpi/reboot.h
new file mode 100644
index 000000000000..8857f57e0b78
--- /dev/null
+++ b/include/acpi/reboot.h
@@ -0,0 +1,9 @@
+
+/*
+ * Dummy placeholder to make the EFI patches apply to the x86 tree.
+ * Andrew/Len, please just kill this file if you encounter it.
+ */
+#ifndef acpi_reboot
+# define acpi_reboot() do { } while (0)
+#endif
+
-- 
cgit v1.2.3


From 5b83683f32b113d07edfb67a33ce389fc624423d Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:31:19 +0100
Subject: x86: EFI runtime service support

This patch adds basic runtime services support for EFI x86_64 system.  The
main file of the patch is the addition of efi_64.c for x86_64.  This file is
modeled after the EFI IA32 avatar.  EFI runtime services initialization are
implemented in efi_64.c.  Some x86_64 specifics are worth noting here.  On
x86_64, parameters passed to EFI firmware services need to follow the EFI
calling convention.  For this purpose, a set of functions named efi_call<x>
(<x> is the number of parameters) are implemented.  EFI function calls are
wrapped before calling the firmware service.  The duplicated code between
efi_32.c and efi_64.c is placed in efi.c to remove them from efi_32.c.

Signed-off-by: Chandramouli Narayanan <mouli@linux.intel.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig              |   2 +-
 arch/x86/kernel/Makefile_64   |   1 +
 arch/x86/kernel/efi.c         | 480 ++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/efi_64.c      | 164 +++++++++++++++
 arch/x86/kernel/efi_stub_64.S | 109 ++++++++++
 arch/x86/kernel/setup_64.c    |  17 +-
 include/asm-x86/bootparam.h   |   5 +-
 include/asm-x86/efi.h         |  70 ++++++
 include/asm-x86/fixmap_64.h   |   3 +
 9 files changed, 847 insertions(+), 4 deletions(-)
 create mode 100644 arch/x86/kernel/efi.c
 create mode 100644 arch/x86/kernel/efi_64.c
 create mode 100644 arch/x86/kernel/efi_stub_64.S
 create mode 100644 include/asm-x86/efi.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index fef944bb920e..23936301db56 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -959,7 +959,7 @@ config MTRR
 config EFI
 	def_bool n
 	prompt "Boot from EFI support"
-	depends on X86_32 && ACPI
+	depends on ACPI
 	---help---
 	This enables the kernel to boot on EFI platforms using
 	system configuration information passed to it from the firmware.
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index 2ec96acf6486..e5093dd8cf01 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -39,6 +39,7 @@ obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
 obj-$(CONFIG_X86_VSMP)		+= vsmp_64.o
 obj-$(CONFIG_K8_NB)		+= k8.o
 obj-$(CONFIG_AUDIT)		+= audit_64.o
+obj-$(CONFIG_EFI)		+= efi.o efi_64.o efi_stub_64.o
 
 obj-$(CONFIG_MODULES)		+= module_64.o
 obj-$(CONFIG_PCI)		+= early-quirks.o
diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
new file mode 100644
index 000000000000..0a61522e85c7
--- /dev/null
+++ b/arch/x86/kernel/efi.c
@@ -0,0 +1,480 @@
+/*
+ * Common EFI (Extensible Firmware Interface) support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 1999 VA Linux Systems
+ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
+ * Copyright (C) 1999-2002 Hewlett-Packard Co.
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
+ *	Stephane Eranian <eranian@hpl.hp.com>
+ * Copyright (C) 2005-2008 Intel Co.
+ *	Fenghua Yu <fenghua.yu@intel.com>
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Chandramouli Narayanan <mouli@linux.intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ *
+ * Copied from efi_32.c to eliminate the duplicated code between EFI
+ * 32/64 support code. --ying 2007-10-26
+ *
+ * All EFI Runtime Services are not implemented yet as EFI only
+ * supports physical mode addressing on SoftSDV. This is to be fixed
+ * in a future version.  --drummond 1999-07-20
+ *
+ * Implemented EFI runtime services and virtual mode calls.  --davidm
+ *
+ * Goutham Rao: <goutham.rao@intel.com>
+ *	Skip non-WB memory and ignore empty memory ranges.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/efi.h>
+#include <linux/bootmem.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/time.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/bcd.h>
+
+#include <asm/setup.h>
+#include <asm/efi.h>
+#include <asm/time.h>
+
+#define EFI_DEBUG	1
+#define PFX 		"EFI: "
+
+int efi_enabled;
+EXPORT_SYMBOL(efi_enabled);
+
+struct efi efi;
+EXPORT_SYMBOL(efi);
+
+struct efi_memory_map memmap;
+
+struct efi efi_phys __initdata;
+static efi_system_table_t efi_systab __initdata;
+
+static efi_status_t virt_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc)
+{
+	return efi_call_virt2(get_time, tm, tc);
+}
+
+static efi_status_t virt_efi_set_time(efi_time_t *tm)
+{
+	return efi_call_virt1(set_time, tm);
+}
+
+static efi_status_t virt_efi_get_wakeup_time(efi_bool_t *enabled,
+					     efi_bool_t *pending,
+					     efi_time_t *tm)
+{
+	return efi_call_virt3(get_wakeup_time,
+			      enabled, pending, tm);
+}
+
+static efi_status_t virt_efi_set_wakeup_time(efi_bool_t enabled, efi_time_t *tm)
+{
+	return efi_call_virt2(set_wakeup_time,
+			      enabled, tm);
+}
+
+static efi_status_t virt_efi_get_variable(efi_char16_t *name,
+					  efi_guid_t *vendor,
+					  u32 *attr,
+					  unsigned long *data_size,
+					  void *data)
+{
+	return efi_call_virt5(get_variable,
+			      name, vendor, attr,
+			      data_size, data);
+}
+
+static efi_status_t virt_efi_get_next_variable(unsigned long *name_size,
+					       efi_char16_t *name,
+					       efi_guid_t *vendor)
+{
+	return efi_call_virt3(get_next_variable,
+			      name_size, name, vendor);
+}
+
+static efi_status_t virt_efi_set_variable(efi_char16_t *name,
+					  efi_guid_t *vendor,
+					  unsigned long attr,
+					  unsigned long data_size,
+					  void *data)
+{
+	return efi_call_virt5(set_variable,
+			      name, vendor, attr,
+			      data_size, data);
+}
+
+static efi_status_t virt_efi_get_next_high_mono_count(u32 *count)
+{
+	return efi_call_virt1(get_next_high_mono_count, count);
+}
+
+static void virt_efi_reset_system(int reset_type,
+				  efi_status_t status,
+				  unsigned long data_size,
+				  efi_char16_t *data)
+{
+	efi_call_virt4(reset_system, reset_type, status,
+		       data_size, data);
+}
+
+static efi_status_t virt_efi_set_virtual_address_map(
+	unsigned long memory_map_size,
+	unsigned long descriptor_size,
+	u32 descriptor_version,
+	efi_memory_desc_t *virtual_map)
+{
+	return efi_call_virt4(set_virtual_address_map,
+			      memory_map_size, descriptor_size,
+			      descriptor_version, virtual_map);
+}
+
+static efi_status_t __init phys_efi_set_virtual_address_map(
+	unsigned long memory_map_size,
+	unsigned long descriptor_size,
+	u32 descriptor_version,
+	efi_memory_desc_t *virtual_map)
+{
+	efi_status_t status;
+
+	efi_call_phys_prelog();
+	status = efi_call_phys4(efi_phys.set_virtual_address_map,
+				memory_map_size, descriptor_size,
+				descriptor_version, virtual_map);
+	efi_call_phys_epilog();
+	return status;
+}
+
+static efi_status_t __init phys_efi_get_time(efi_time_t *tm,
+					     efi_time_cap_t *tc)
+{
+	efi_status_t status;
+
+	efi_call_phys_prelog();
+	status = efi_call_phys2(efi_phys.get_time, tm, tc);
+	efi_call_phys_epilog();
+	return status;
+}
+
+int efi_set_rtc_mmss(unsigned long nowtime)
+{
+	int real_seconds, real_minutes;
+	efi_status_t 	status;
+	efi_time_t 	eft;
+	efi_time_cap_t 	cap;
+
+	status = efi.get_time(&eft, &cap);
+	if (status != EFI_SUCCESS) {
+		printk(KERN_ERR "Oops: efitime: can't read time!\n");
+		return -1;
+	}
+
+	real_seconds = nowtime % 60;
+	real_minutes = nowtime / 60;
+	if (((abs(real_minutes - eft.minute) + 15)/30) & 1)
+		real_minutes += 30;
+	real_minutes %= 60;
+	eft.minute = real_minutes;
+	eft.second = real_seconds;
+
+	status = efi.set_time(&eft);
+	if (status != EFI_SUCCESS) {
+		printk(KERN_ERR "Oops: efitime: can't write time!\n");
+		return -1;
+	}
+	return 0;
+}
+
+unsigned long efi_get_time(void)
+{
+	efi_status_t status;
+	efi_time_t eft;
+	efi_time_cap_t cap;
+
+	status = efi.get_time(&eft, &cap);
+	if (status != EFI_SUCCESS)
+		printk(KERN_ERR "Oops: efitime: can't read time!\n");
+
+	return mktime(eft.year, eft.month, eft.day, eft.hour,
+		      eft.minute, eft.second);
+}
+
+#if EFI_DEBUG
+static void __init print_efi_memmap(void)
+{
+	efi_memory_desc_t *md;
+	void *p;
+	int i;
+
+	for (p = memmap.map, i = 0;
+	     p < memmap.map_end;
+	     p += memmap.desc_size, i++) {
+		md = p;
+		printk(KERN_INFO PFX "mem%02u: type=%u, attr=0x%llx, "
+			"range=[0x%016llx-0x%016llx) (%lluMB)\n",
+			i, md->type, md->attribute, md->phys_addr,
+			md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT),
+			(md->num_pages >> (20 - EFI_PAGE_SHIFT)));
+	}
+}
+#endif  /*  EFI_DEBUG  */
+
+void __init efi_init(void)
+{
+	efi_config_table_t *config_tables;
+	efi_runtime_services_t *runtime;
+	efi_char16_t *c16;
+	char vendor[100] = "unknown";
+	int i = 0;
+	void *tmp;
+
+#ifdef CONFIG_X86_32
+	efi_phys.systab = (efi_system_table_t *)boot_params.efi_info.efi_systab;
+	memmap.phys_map = (void *)boot_params.efi_info.efi_memmap;
+#else
+	efi_phys.systab = (efi_system_table_t *)
+		(boot_params.efi_info.efi_systab |
+		 ((__u64)boot_params.efi_info.efi_systab_hi<<32));
+	memmap.phys_map = (void *)
+		(boot_params.efi_info.efi_memmap |
+		 ((__u64)boot_params.efi_info.efi_memmap_hi<<32));
+#endif
+	memmap.nr_map = boot_params.efi_info.efi_memmap_size /
+		boot_params.efi_info.efi_memdesc_size;
+	memmap.desc_version = boot_params.efi_info.efi_memdesc_version;
+	memmap.desc_size = boot_params.efi_info.efi_memdesc_size;
+
+	efi.systab = efi_early_ioremap((unsigned long)efi_phys.systab,
+				       sizeof(efi_system_table_t));
+	if (efi.systab == NULL)
+		printk(KERN_ERR "Couldn't map the EFI system table!\n");
+	memcpy(&efi_systab, efi.systab, sizeof(efi_system_table_t));
+	efi_early_iounmap(efi.systab, sizeof(efi_system_table_t));
+	efi.systab = &efi_systab;
+
+	/*
+	 * Verify the EFI Table
+	 */
+	if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
+		printk(KERN_ERR "EFI system table signature incorrect!\n");
+	if ((efi.systab->hdr.revision >> 16) == 0)
+		printk(KERN_ERR "Warning: EFI system table version "
+		       "%d.%02d, expected 1.00 or greater!\n",
+		       efi.systab->hdr.revision >> 16,
+		       efi.systab->hdr.revision & 0xffff);
+
+	/*
+	 * Show what we know for posterity
+	 */
+	c16 = tmp = efi_early_ioremap(efi.systab->fw_vendor, 2);
+	if (c16) {
+		for (i = 0; i < sizeof(vendor) && *c16; ++i)
+			vendor[i] = *c16++;
+		vendor[i] = '\0';
+	} else
+		printk(KERN_ERR PFX "Could not map the firmware vendor!\n");
+	efi_early_iounmap(tmp, 2);
+
+	printk(KERN_INFO "EFI v%u.%.02u by %s \n",
+	       efi.systab->hdr.revision >> 16,
+	       efi.systab->hdr.revision & 0xffff, vendor);
+
+	/*
+	 * Let's see what config tables the firmware passed to us.
+	 */
+	config_tables = efi_early_ioremap(
+		efi.systab->tables,
+		efi.systab->nr_tables * sizeof(efi_config_table_t));
+	if (config_tables == NULL)
+		printk(KERN_ERR "Could not map EFI Configuration Table!\n");
+
+	printk(KERN_INFO);
+	for (i = 0; i < efi.systab->nr_tables; i++) {
+		if (!efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID)) {
+			efi.mps = config_tables[i].table;
+			printk(" MPS=0x%lx ", config_tables[i].table);
+		} else if (!efi_guidcmp(config_tables[i].guid,
+					ACPI_20_TABLE_GUID)) {
+			efi.acpi20 = config_tables[i].table;
+			printk(" ACPI 2.0=0x%lx ", config_tables[i].table);
+		} else if (!efi_guidcmp(config_tables[i].guid,
+					ACPI_TABLE_GUID)) {
+			efi.acpi = config_tables[i].table;
+			printk(" ACPI=0x%lx ", config_tables[i].table);
+		} else if (!efi_guidcmp(config_tables[i].guid,
+					SMBIOS_TABLE_GUID)) {
+			efi.smbios = config_tables[i].table;
+			printk(" SMBIOS=0x%lx ", config_tables[i].table);
+		} else if (!efi_guidcmp(config_tables[i].guid,
+					HCDP_TABLE_GUID)) {
+			efi.hcdp = config_tables[i].table;
+			printk(" HCDP=0x%lx ", config_tables[i].table);
+		} else if (!efi_guidcmp(config_tables[i].guid,
+					UGA_IO_PROTOCOL_GUID)) {
+			efi.uga = config_tables[i].table;
+			printk(" UGA=0x%lx ", config_tables[i].table);
+		}
+	}
+	printk("\n");
+	efi_early_iounmap(config_tables,
+			  efi.systab->nr_tables * sizeof(efi_config_table_t));
+
+	/*
+	 * Check out the runtime services table. We need to map
+	 * the runtime services table so that we can grab the physical
+	 * address of several of the EFI runtime functions, needed to
+	 * set the firmware into virtual mode.
+	 */
+	runtime = efi_early_ioremap((unsigned long)efi.systab->runtime,
+				    sizeof(efi_runtime_services_t));
+	if (runtime != NULL) {
+		/*
+		 * We will only need *early* access to the following
+		 * two EFI runtime services before set_virtual_address_map
+		 * is invoked.
+		 */
+		efi_phys.get_time = (efi_get_time_t *)runtime->get_time;
+		efi_phys.set_virtual_address_map =
+			(efi_set_virtual_address_map_t *)
+			runtime->set_virtual_address_map;
+		/*
+		 * Make efi_get_time can be called before entering
+		 * virtual mode.
+		 */
+		efi.get_time = phys_efi_get_time;
+	} else
+		printk(KERN_ERR "Could not map the EFI runtime service "
+		       "table!\n");
+	efi_early_iounmap(runtime, sizeof(efi_runtime_services_t));
+
+	/* Map the EFI memory map */
+	memmap.map = efi_early_ioremap((unsigned long)memmap.phys_map,
+				       memmap.nr_map * memmap.desc_size);
+	if (memmap.map == NULL)
+		printk(KERN_ERR "Could not map the EFI memory map!\n");
+	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
+	if (memmap.desc_size != sizeof(efi_memory_desc_t))
+		printk(KERN_WARNING "Kernel-defined memdesc"
+		       "doesn't match the one from EFI!\n");
+
+#ifdef CONFIG_X86_64
+	/* Setup for EFI runtime service */
+	reboot_type = BOOT_EFI;
+
+#endif
+#if EFI_DEBUG
+	print_efi_memmap();
+#endif
+}
+
+/*
+ * This function will switch the EFI runtime services to virtual mode.
+ * Essentially, look through the EFI memmap and map every region that
+ * has the runtime attribute bit set in its memory descriptor and update
+ * that memory descriptor with the virtual address obtained from ioremap().
+ * This enables the runtime services to be called without having to
+ * thunk back into physical mode for every invocation.
+ */
+void __init efi_enter_virtual_mode(void)
+{
+	efi_memory_desc_t *md;
+	efi_status_t status;
+	unsigned long end;
+	void *p;
+
+	efi.systab = NULL;
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		if (!(md->attribute & EFI_MEMORY_RUNTIME))
+			continue;
+		if ((md->attribute & EFI_MEMORY_WB) &&
+		    (((md->phys_addr + (md->num_pages<<EFI_PAGE_SHIFT)) >>
+		      PAGE_SHIFT) < end_pfn_map))
+			md->virt_addr = (unsigned long)__va(md->phys_addr);
+		else
+			md->virt_addr = (unsigned long)
+				efi_ioremap(md->phys_addr,
+					    md->num_pages << EFI_PAGE_SHIFT);
+		if (!md->virt_addr)
+			printk(KERN_ERR PFX "ioremap of 0x%llX failed!\n",
+			       (unsigned long long)md->phys_addr);
+		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
+		if ((md->phys_addr <= (unsigned long)efi_phys.systab) &&
+		    ((unsigned long)efi_phys.systab < end))
+			efi.systab = (efi_system_table_t *)(unsigned long)
+				(md->virt_addr - md->phys_addr +
+				 (unsigned long)efi_phys.systab);
+	}
+
+	BUG_ON(!efi.systab);
+
+	status = phys_efi_set_virtual_address_map(
+		memmap.desc_size * memmap.nr_map,
+		memmap.desc_size,
+		memmap.desc_version,
+		memmap.phys_map);
+
+	if (status != EFI_SUCCESS) {
+		printk(KERN_ALERT "Unable to switch EFI into virtual mode "
+		       "(status=%lx)!\n", status);
+		panic("EFI call to SetVirtualAddressMap() failed!");
+	}
+
+	/*
+	 * Now that EFI is in virtual mode, update the function
+	 * pointers in the runtime service table to the new virtual addresses.
+	 *
+	 * Call EFI services through wrapper functions.
+	 */
+	efi.get_time = virt_efi_get_time;
+	efi.set_time = virt_efi_set_time;
+	efi.get_wakeup_time = virt_efi_get_wakeup_time;
+	efi.set_wakeup_time = virt_efi_set_wakeup_time;
+	efi.get_variable = virt_efi_get_variable;
+	efi.get_next_variable = virt_efi_get_next_variable;
+	efi.set_variable = virt_efi_set_variable;
+	efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count;
+	efi.reset_system = virt_efi_reset_system;
+	efi.set_virtual_address_map = virt_efi_set_virtual_address_map;
+#ifdef CONFIG_X86_64
+	runtime_code_page_mkexec();
+#endif
+}
+
+/*
+ * Convenience functions to obtain memory types and attributes
+ */
+u32 efi_mem_type(unsigned long phys_addr)
+{
+	efi_memory_desc_t *md;
+	void *p;
+
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		if ((md->phys_addr <= phys_addr) &&
+		    (phys_addr < (md->phys_addr +
+				  (md->num_pages << EFI_PAGE_SHIFT))))
+			return md->type;
+	}
+	return 0;
+}
+
+u64 efi_mem_attributes(unsigned long phys_addr)
+{
+	efi_memory_desc_t *md;
+	void *p;
+
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		if ((md->phys_addr <= phys_addr) &&
+		    (phys_addr < (md->phys_addr +
+				  (md->num_pages << EFI_PAGE_SHIFT))))
+			return md->attribute;
+	}
+	return 0;
+}
diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c
new file mode 100644
index 000000000000..f2000dbc7195
--- /dev/null
+++ b/arch/x86/kernel/efi_64.c
@@ -0,0 +1,164 @@
+/*
+ * x86_64 specific EFI support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 2005-2008 Intel Co.
+ *	Fenghua Yu <fenghua.yu@intel.com>
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Chandramouli Narayanan <mouli@linux.intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ *
+ * Code to convert EFI to E820 map has been implemented in elilo bootloader
+ * based on a EFI patch by Edgar Hucek. Based on the E820 map, the page table
+ * is setup appropriately for EFI runtime code.
+ * - mouli 06/14/2007.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/bootmem.h>
+#include <linux/ioport.h>
+#include <linux/module.h>
+#include <linux/efi.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+
+#include <asm/setup.h>
+#include <asm/page.h>
+#include <asm/e820.h>
+#include <asm/pgtable.h>
+#include <asm/tlbflush.h>
+#include <asm/cacheflush.h>
+#include <asm/proto.h>
+#include <asm/efi.h>
+
+static pgd_t save_pgd __initdata;
+static unsigned long efi_flags __initdata;
+
+static int __init setup_noefi(char *arg)
+{
+	efi_enabled = 0;
+	return 0;
+}
+early_param("noefi", setup_noefi);
+
+static void __init early_mapping_set_exec(unsigned long start,
+					  unsigned long end,
+					  int executable)
+{
+	pte_t *kpte;
+	int level;
+
+	while (start < end) {
+		kpte = lookup_address((unsigned long)__va(start), &level);
+		BUG_ON(!kpte);
+		if (executable)
+			set_pte(kpte, pte_mkexec(*kpte));
+		else
+			set_pte(kpte, __pte((pte_val(*kpte) | _PAGE_NX) & \
+					    __supported_pte_mask));
+		if (pte_huge(*kpte))
+			start = (start + PMD_SIZE) & PMD_MASK;
+		else
+			start = (start + PAGE_SIZE) & PAGE_MASK;
+	}
+}
+
+static void __init early_runtime_code_mapping_set_exec(int executable)
+{
+	efi_memory_desc_t *md;
+	void *p;
+
+	/* Make EFI runtime service code area executable */
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		if (md->type == EFI_RUNTIME_SERVICES_CODE) {
+			unsigned long end;
+			end = md->phys_addr + (md->num_pages << PAGE_SHIFT);
+			early_mapping_set_exec(md->phys_addr, end, executable);
+		}
+	}
+}
+
+void __init efi_call_phys_prelog(void)
+{
+	unsigned long vaddress;
+
+	local_irq_save(efi_flags);
+	early_runtime_code_mapping_set_exec(1);
+	vaddress = (unsigned long)__va(0x0UL);
+	pgd_val(save_pgd) = pgd_val(*pgd_offset_k(0x0UL));
+	set_pgd(pgd_offset_k(0x0UL), *pgd_offset_k(vaddress));
+	__flush_tlb_all();
+}
+
+void __init efi_call_phys_epilog(void)
+{
+	/*
+	 * After the lock is released, the original page table is restored.
+	 */
+	set_pgd(pgd_offset_k(0x0UL), save_pgd);
+	early_runtime_code_mapping_set_exec(0);
+	__flush_tlb_all();
+	local_irq_restore(efi_flags);
+}
+
+/*
+ * We need to map the EFI memory map again after init_memory_mapping().
+ */
+void __init efi_map_memmap(void)
+{
+	memmap.map = __va(memmap.phys_map);
+	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
+}
+
+void __init efi_reserve_bootmem(void)
+{
+	reserve_bootmem_generic((unsigned long)memmap.phys_map,
+				memmap.nr_map * memmap.desc_size);
+}
+
+void __init runtime_code_page_mkexec(void)
+{
+	efi_memory_desc_t *md;
+	void *p;
+
+	/* Make EFI runtime service code area executable */
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		if (md->type == EFI_RUNTIME_SERVICES_CODE)
+			change_page_attr_addr(md->virt_addr,
+					      md->num_pages,
+					      PAGE_KERNEL_EXEC);
+	}
+	__flush_tlb_all();
+}
+
+void __iomem * __init efi_ioremap(unsigned long offset,
+				  unsigned long size)
+{
+	static unsigned pages_mapped;
+	unsigned long last_addr;
+	unsigned i, pages;
+
+	last_addr = offset + size - 1;
+	offset &= PAGE_MASK;
+	pages = (PAGE_ALIGN(last_addr) - offset) >> PAGE_SHIFT;
+	if (pages_mapped + pages > MAX_EFI_IO_PAGES)
+		return NULL;
+
+	for (i = 0; i < pages; i++) {
+		set_fixmap_nocache(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped,
+				   offset);
+		offset += PAGE_SIZE;
+		pages_mapped++;
+	}
+
+	return (void __iomem *)__fix_to_virt(FIX_EFI_IO_MAP_FIRST_PAGE - \
+					     (pages_mapped - pages));
+}
diff --git a/arch/x86/kernel/efi_stub_64.S b/arch/x86/kernel/efi_stub_64.S
new file mode 100644
index 000000000000..99b47d48c9f4
--- /dev/null
+++ b/arch/x86/kernel/efi_stub_64.S
@@ -0,0 +1,109 @@
+/*
+ * Function calling ABI conversion from Linux to EFI for x86_64
+ *
+ * Copyright (C) 2007 Intel Corp
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ */
+
+#include <linux/linkage.h>
+
+#define SAVE_XMM			\
+	mov %rsp, %rax;			\
+	subq $0x70, %rsp;		\
+	and $~0xf, %rsp;		\
+	mov %rax, (%rsp);		\
+	mov %cr0, %rax;			\
+	clts;				\
+	mov %rax, 0x8(%rsp);		\
+	movaps %xmm0, 0x60(%rsp);	\
+	movaps %xmm1, 0x50(%rsp);	\
+	movaps %xmm2, 0x40(%rsp);	\
+	movaps %xmm3, 0x30(%rsp);	\
+	movaps %xmm4, 0x20(%rsp);	\
+	movaps %xmm5, 0x10(%rsp)
+
+#define RESTORE_XMM			\
+	movaps 0x60(%rsp), %xmm0;	\
+	movaps 0x50(%rsp), %xmm1;	\
+	movaps 0x40(%rsp), %xmm2;	\
+	movaps 0x30(%rsp), %xmm3;	\
+	movaps 0x20(%rsp), %xmm4;	\
+	movaps 0x10(%rsp), %xmm5;	\
+	mov 0x8(%rsp), %rsi;		\
+	mov %rsi, %cr0;			\
+	mov (%rsp), %rsp
+
+ENTRY(efi_call0)
+	SAVE_XMM
+	subq $32, %rsp
+	call *%rdi
+	addq $32, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call1)
+	SAVE_XMM
+	subq $32, %rsp
+	mov  %rsi, %rcx
+	call *%rdi
+	addq $32, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call2)
+	SAVE_XMM
+	subq $32, %rsp
+	mov  %rsi, %rcx
+	call *%rdi
+	addq $32, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call3)
+	SAVE_XMM
+	subq $32, %rsp
+	mov  %rcx, %r8
+	mov  %rsi, %rcx
+	call *%rdi
+	addq $32, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call4)
+	SAVE_XMM
+	subq $32, %rsp
+	mov %r8, %r9
+	mov %rcx, %r8
+	mov %rsi, %rcx
+	call *%rdi
+	addq $32, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call5)
+	SAVE_XMM
+	subq $48, %rsp
+	mov %r9, 32(%rsp)
+	mov %r8, %r9
+	mov %rcx, %r8
+	mov %rsi, %rcx
+	call *%rdi
+	addq $48, %rsp
+	RESTORE_XMM
+	ret
+
+ENTRY(efi_call6)
+	SAVE_XMM
+	mov (%rsp), %rax
+	mov 8(%rax), %rax
+	subq $48, %rsp
+	mov %r9, 32(%rsp)
+	mov %rax, 40(%rsp)
+	mov %r8, %r9
+	mov %rcx, %r8
+	mov %rsi, %rcx
+	call *%rdi
+	addq $48, %rsp
+	RESTORE_XMM
+	ret
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 8dd110d93e73..90b8bb4748b9 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -30,6 +30,7 @@
 #include <linux/crash_dump.h>
 #include <linux/root_dev.h>
 #include <linux/pci.h>
+#include <linux/efi.h>
 #include <linux/acpi.h>
 #include <linux/kallsyms.h>
 #include <linux/edd.h>
@@ -299,6 +300,11 @@ void __init setup_arch(char **cmdline_p)
 	rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
 	rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
 #endif
+#ifdef CONFIG_EFI
+	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     "EL64", 4))
+		efi_enabled = 1;
+#endif
 
 	ARCH_SETUP
 
@@ -341,6 +347,8 @@ void __init setup_arch(char **cmdline_p)
 	discover_ebda();
 
 	init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
+	if (efi_enabled)
+		efi_init();
 
 	dmi_scan_machine();
 
@@ -414,6 +422,12 @@ void __init setup_arch(char **cmdline_p)
 	 */
        acpi_reserve_bootmem();
 #endif
+
+	if (efi_enabled) {
+		efi_map_memmap();
+		efi_reserve_bootmem();
+	}
+
        /*
 	* Find and reserve possible boot-time SMP configuration:
 	*/
@@ -479,7 +493,8 @@ void __init setup_arch(char **cmdline_p)
 
 #ifdef CONFIG_VT
 #if defined(CONFIG_VGA_CONSOLE)
-	conswitchp = &vga_con;
+	if (!efi_enabled || (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
+		conswitchp = &vga_con;
 #elif defined(CONFIG_DUMMY_CONSOLE)
 	conswitchp = &dummy_con;
 #endif
diff --git a/include/asm-x86/bootparam.h b/include/asm-x86/bootparam.h
index 19f3ddf2df4b..51151356840f 100644
--- a/include/asm-x86/bootparam.h
+++ b/include/asm-x86/bootparam.h
@@ -54,13 +54,14 @@ struct sys_desc_table {
 };
 
 struct efi_info {
-	__u32 _pad1;
+	__u32 efi_loader_signature;
 	__u32 efi_systab;
 	__u32 efi_memdesc_size;
 	__u32 efi_memdesc_version;
 	__u32 efi_memmap;
 	__u32 efi_memmap_size;
-	__u32 _pad2[2];
+	__u32 efi_systab_hi;
+	__u32 efi_memmap_hi;
 };
 
 /* The so-called "zeropage" */
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
new file mode 100644
index 000000000000..1459d3d58653
--- /dev/null
+++ b/include/asm-x86/efi.h
@@ -0,0 +1,70 @@
+#ifndef _ASM_X86_EFI_H
+#define _ASM_X86_EFI_H
+
+#ifdef CONFIG_X86_32
+#else /* !CONFIG_X86_32 */
+
+#define MAX_EFI_IO_PAGES	100
+
+extern u64 efi_call0(void *fp);
+extern u64 efi_call1(void *fp, u64 arg1);
+extern u64 efi_call2(void *fp, u64 arg1, u64 arg2);
+extern u64 efi_call3(void *fp, u64 arg1, u64 arg2, u64 arg3);
+extern u64 efi_call4(void *fp, u64 arg1, u64 arg2, u64 arg3, u64 arg4);
+extern u64 efi_call5(void *fp, u64 arg1, u64 arg2, u64 arg3,
+		     u64 arg4, u64 arg5);
+extern u64 efi_call6(void *fp, u64 arg1, u64 arg2, u64 arg3,
+		     u64 arg4, u64 arg5, u64 arg6);
+
+#define efi_call_phys0(f)			\
+	efi_call0((void *)(f))
+#define efi_call_phys1(f, a1)			\
+	efi_call1((void *)(f), (u64)(a1))
+#define efi_call_phys2(f, a1, a2)			\
+	efi_call2((void *)(f), (u64)(a1), (u64)(a2))
+#define efi_call_phys3(f, a1, a2, a3)				\
+	efi_call3((void *)(f), (u64)(a1), (u64)(a2), (u64)(a3))
+#define efi_call_phys4(f, a1, a2, a3, a4)				\
+	efi_call4((void *)(f), (u64)(a1), (u64)(a2), (u64)(a3),		\
+		  (u64)(a4))
+#define efi_call_phys5(f, a1, a2, a3, a4, a5)				\
+	efi_call5((void *)(f), (u64)(a1), (u64)(a2), (u64)(a3),		\
+		  (u64)(a4), (u64)(a5))
+#define efi_call_phys6(f, a1, a2, a3, a4, a5, a6)			\
+	efi_call6((void *)(f), (u64)(a1), (u64)(a2), (u64)(a3),		\
+		  (u64)(a4), (u64)(a5), (u64)(a6))
+
+#define efi_call_virt0(f)				\
+	efi_call0((void *)(efi.systab->runtime->f))
+#define efi_call_virt1(f, a1)					\
+	efi_call1((void *)(efi.systab->runtime->f), (u64)(a1))
+#define efi_call_virt2(f, a1, a2)					\
+	efi_call2((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2))
+#define efi_call_virt3(f, a1, a2, a3)					\
+	efi_call3((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \
+		  (u64)(a3))
+#define efi_call_virt4(f, a1, a2, a3, a4)				\
+	efi_call4((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \
+		  (u64)(a3), (u64)(a4))
+#define efi_call_virt5(f, a1, a2, a3, a4, a5)				\
+	efi_call5((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \
+		  (u64)(a3), (u64)(a4), (u64)(a5))
+#define efi_call_virt6(f, a1, a2, a3, a4, a5, a6)			\
+	efi_call6((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \
+		  (u64)(a3), (u64)(a4), (u64)(a5), (u64)(a6))
+
+#define efi_early_ioremap(addr, size)		early_ioremap(addr, size)
+#define efi_early_iounmap(vaddr, size)		early_iounmap(vaddr, size)
+
+extern void *efi_ioremap(unsigned long offset, unsigned long size);
+
+extern int efi_time;
+
+#endif /* CONFIG_X86_32 */
+
+extern void efi_reserve_bootmem(void);
+extern void efi_call_phys_prelog(void);
+extern void efi_call_phys_epilog(void);
+extern void runtime_code_page_mkexec(void);
+
+#endif
diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h
index cdfbe4a6ae6f..8f44782e5fe5 100644
--- a/include/asm-x86/fixmap_64.h
+++ b/include/asm-x86/fixmap_64.h
@@ -15,6 +15,7 @@
 #include <asm/apicdef.h>
 #include <asm/page.h>
 #include <asm/vsyscall.h>
+#include <asm/efi.h>
 
 /*
  * Here we define all the compile-time 'special' virtual
@@ -41,6 +42,8 @@ enum fixed_addresses {
 	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
 	FIX_IO_APIC_BASE_0,
 	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
+	FIX_EFI_IO_MAP_LAST_PAGE,
+	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE+MAX_EFI_IO_PAGES-1,
 	__end_of_fixed_addresses
 };
 
-- 
cgit v1.2.3


From de18c850af701ac9512b7239e88fa45e4c168771 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:31:19 +0100
Subject: x86: EFI runtime service support: EFI runtime services

This patch adds support for several EFI runtime services for EFI x86_64
system.

The EFI support for emergency_restart is added.

Signed-off-by: Chandramouli Narayanan <mouli@linux.intel.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/reboot_64.c         | 20 +++++++++++++-------
 include/asm-x86/emergency-restart.h |  9 +++++++++
 2 files changed, 22 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/reboot_64.c b/arch/x86/kernel/reboot_64.c
index 307f996a3933..d6bdf93ffca9 100644
--- a/arch/x86/kernel/reboot_64.c
+++ b/arch/x86/kernel/reboot_64.c
@@ -9,6 +9,7 @@
 #include <linux/pm.h>
 #include <linux/kdebug.h>
 #include <linux/sched.h>
+#include <linux/efi.h>
 #include <acpi/reboot.h>
 #include <asm/io.h>
 #include <asm/delay.h>
@@ -28,20 +29,17 @@ void (*pm_power_off)(void);
 EXPORT_SYMBOL(pm_power_off);
 
 static long no_idt[3];
-static enum { 
-	BOOT_TRIPLE = 't',
-	BOOT_KBD = 'k',
-	BOOT_ACPI = 'a'
-} reboot_type = BOOT_KBD;
+enum reboot_type reboot_type = BOOT_KBD;
 static int reboot_mode = 0;
 int reboot_force;
 
-/* reboot=t[riple] | k[bd] [, [w]arm | [c]old]
+/* reboot=t[riple] | k[bd] | e[fi] [, [w]arm | [c]old]
    warm   Don't set the cold reboot flag
    cold   Set the cold reboot flag
    triple Force a triple fault (init)
    kbd    Use the keyboard controller. cold reset (default)
    acpi   Use the RESET_REG in the FADT
+   efi    Use efi reset_system runtime service
    force  Avoid anything that could hang.
  */ 
 static int __init reboot_setup(char *str)
@@ -60,6 +58,7 @@ static int __init reboot_setup(char *str)
 		case 'a':
 		case 'b':
 		case 'k':
+		case 'e':
 			reboot_type = *str;
 			break;
 		case 'f':
@@ -155,7 +154,14 @@ void machine_emergency_restart(void)
 			acpi_reboot();
 			reboot_type = BOOT_KBD;
 			break;
-		}      
+
+		case BOOT_EFI:
+			if (efi_enabled)
+				efi.reset_system(reboot_mode ? EFI_RESET_WARM : EFI_RESET_COLD,
+						 EFI_SUCCESS, 0, NULL);
+			reboot_type = BOOT_KBD;
+			break;
+		}
 	}      
 }
 
diff --git a/include/asm-x86/emergency-restart.h b/include/asm-x86/emergency-restart.h
index 680c39563345..54189084462a 100644
--- a/include/asm-x86/emergency-restart.h
+++ b/include/asm-x86/emergency-restart.h
@@ -1,6 +1,15 @@
 #ifndef _ASM_EMERGENCY_RESTART_H
 #define _ASM_EMERGENCY_RESTART_H
 
+enum reboot_type {
+	BOOT_TRIPLE = 't',
+	BOOT_KBD = 'k',
+	BOOT_ACPI = 'a',
+	BOOT_EFI = 'e'
+};
+
+extern enum reboot_type reboot_type;
+
 extern void machine_emergency_restart(void);
 
 #endif /* _ASM_EMERGENCY_RESTART_H */
-- 
cgit v1.2.3


From e429795c68d3001ecae74f6465420c9f043b0ece Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:31:19 +0100
Subject: x86: EFI runtime service support: remove duplicated code from
 efi_32.c

This patch removes the duplicated code between efi_32.c and efi.c.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile_32 |   2 +-
 arch/x86/kernel/e820_32.c   |   5 -
 arch/x86/kernel/efi_32.c    | 430 +-------------------------------------------
 arch/x86/kernel/setup_32.c  |  11 +-
 include/asm-x86/efi.h       |  42 +++++
 5 files changed, 47 insertions(+), 443 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index cc2651bcc07f..a854e23eac0b 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -38,7 +38,7 @@ obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit_32.o
 obj-$(CONFIG_KPROBES)		+= kprobes_32.o
 obj-$(CONFIG_MODULES)		+= module_32.o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat_32.o
-obj-$(CONFIG_EFI) 		+= efi_32.o efi_stub_32.o
+obj-$(CONFIG_EFI) 		+= efi.o efi_32.o efi_stub_32.o
 obj-$(CONFIG_DOUBLEFAULT) 	+= doublefault_32.o
 obj-$(CONFIG_VM86)		+= vm86_32.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c
index 87cadc86d5ee..56335a85a15a 100644
--- a/arch/x86/kernel/e820_32.c
+++ b/arch/x86/kernel/e820_32.c
@@ -17,11 +17,6 @@
 #include <asm/e820.h>
 #include <asm/setup.h>
 
-#ifdef CONFIG_EFI
-int efi_enabled = 0;
-EXPORT_SYMBOL(efi_enabled);
-#endif
-
 struct e820map e820;
 struct change_member {
 	struct e820entry *pbios; /* pointer to original bios entry */
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index 863e8926f2bb..1df13725e519 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -39,21 +39,8 @@
 #include <asm/desc.h>
 #include <asm/tlbflush.h>
 
-#define EFI_DEBUG	0
 #define PFX 		"EFI: "
 
-extern efi_status_t asmlinkage efi_call_phys(void *, ...);
-
-struct efi efi;
-EXPORT_SYMBOL(efi);
-static struct efi efi_phys;
-struct efi_memory_map memmap;
-
-/*
- * We require an early boot_ioremap mapping mechanism initially
- */
-extern void * boot_ioremap(unsigned long, unsigned long);
-
 /*
  * To make EFI call EFI runtime service in physical addressing mode we need
  * prelog/epilog before/after the invocation to disable interrupt, to
@@ -65,7 +52,7 @@ static unsigned long efi_rt_eflags;
 static DEFINE_SPINLOCK(efi_rt_lock);
 static pgd_t efi_bak_pg_dir_pointer[2];
 
-static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
+void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 {
 	unsigned long cr4;
 	unsigned long temp;
@@ -108,7 +95,7 @@ static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 	load_gdt(&gdt_descr);
 }
 
-static void efi_call_phys_epilog(void) __releases(efi_rt_lock)
+void efi_call_phys_epilog(void) __releases(efi_rt_lock)
 {
 	unsigned long cr4;
 	struct desc_ptr gdt_descr;
@@ -138,87 +125,6 @@ static void efi_call_phys_epilog(void) __releases(efi_rt_lock)
 	spin_unlock(&efi_rt_lock);
 }
 
-static efi_status_t
-phys_efi_set_virtual_address_map(unsigned long memory_map_size,
-				 unsigned long descriptor_size,
-				 u32 descriptor_version,
-				 efi_memory_desc_t *virtual_map)
-{
-	efi_status_t status;
-
-	efi_call_phys_prelog();
-	status = efi_call_phys(efi_phys.set_virtual_address_map,
-				     memory_map_size, descriptor_size,
-				     descriptor_version, virtual_map);
-	efi_call_phys_epilog();
-	return status;
-}
-
-static efi_status_t
-phys_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc)
-{
-	efi_status_t status;
-
-	efi_call_phys_prelog();
-	status = efi_call_phys(efi_phys.get_time, tm, tc);
-	efi_call_phys_epilog();
-	return status;
-}
-
-inline int efi_set_rtc_mmss(unsigned long nowtime)
-{
-	int real_seconds, real_minutes;
-	efi_status_t 	status;
-	efi_time_t 	eft;
-	efi_time_cap_t 	cap;
-
-	spin_lock(&efi_rt_lock);
-	status = efi.get_time(&eft, &cap);
-	spin_unlock(&efi_rt_lock);
-	if (status != EFI_SUCCESS)
-		panic("Ooops, efitime: can't read time!\n");
-	real_seconds = nowtime % 60;
-	real_minutes = nowtime / 60;
-
-	if (((abs(real_minutes - eft.minute) + 15)/30) & 1)
-		real_minutes += 30;
-	real_minutes %= 60;
-
-	eft.minute = real_minutes;
-	eft.second = real_seconds;
-
-	if (status != EFI_SUCCESS) {
-		printk("Ooops: efitime: can't read time!\n");
-		return -1;
-	}
-	return 0;
-}
-/*
- * This is used during kernel init before runtime
- * services have been remapped and also during suspend, therefore,
- * we'll need to call both in physical and virtual modes.
- */
-inline unsigned long efi_get_time(void)
-{
-	efi_status_t status;
-	efi_time_t eft;
-	efi_time_cap_t cap;
-
-	if (efi.get_time) {
-		/* if we are in virtual mode use remapped function */
- 		status = efi.get_time(&eft, &cap);
-	} else {
-		/* we are in physical mode */
-		status = phys_efi_get_time(&eft, &cap);
-	}
-
-	if (status != EFI_SUCCESS)
-		printk("Oops: efitime: can't read time status: 0x%lx\n",status);
-
-	return mktime(eft.year, eft.month, eft.day, eft.hour,
-			eft.minute, eft.second);
-}
-
 int is_available_memory(efi_memory_desc_t * md)
 {
 	if (!(md->attribute & EFI_MEMORY_WB))
@@ -250,24 +156,6 @@ void __init efi_map_memmap(void)
 	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
 }
 
-#if EFI_DEBUG
-static void __init print_efi_memmap(void)
-{
-	efi_memory_desc_t *md;
-	void *p;
-	int i;
-
-	for (p = memmap.map, i = 0; p < memmap.map_end; p += memmap.desc_size, i++) {
-		md = p;
-		printk(KERN_INFO "mem%02u: type=%u, attr=0x%llx, "
-			"range=[0x%016llx-0x%016llx) (%lluMB)\n",
-			i, md->type, md->attribute, md->phys_addr,
-			md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT),
-			(md->num_pages >> (20 - EFI_PAGE_SHIFT)));
-	}
-}
-#endif  /*  EFI_DEBUG  */
-
 /*
  * Walks the EFI memory map and calls CALLBACK once for each EFI
  * memory descriptor that has memory that is available for kernel use.
@@ -319,288 +207,6 @@ void efi_memmap_walk(efi_freemem_callback_t callback, void *arg)
 	}
 }
 
-void __init efi_init(void)
-{
-	efi_config_table_t *config_tables;
-	efi_runtime_services_t *runtime;
-	efi_char16_t *c16;
-	char vendor[100] = "unknown";
-	unsigned long num_config_tables;
-	int i = 0;
-
-	memset(&efi, 0, sizeof(efi) );
-	memset(&efi_phys, 0, sizeof(efi_phys));
-
-	efi_phys.systab =
-		(efi_system_table_t *)boot_params.efi_info.efi_systab;
-	memmap.phys_map = (void *)boot_params.efi_info.efi_memmap;
-	memmap.nr_map = boot_params.efi_info.efi_memmap_size/
-		boot_params.efi_info.efi_memdesc_size;
-	memmap.desc_version = boot_params.efi_info.efi_memdesc_version;
-	memmap.desc_size = boot_params.efi_info.efi_memdesc_size;
-
-	efi.systab = (efi_system_table_t *)
-		boot_ioremap((unsigned long) efi_phys.systab,
-			sizeof(efi_system_table_t));
-	/*
-	 * Verify the EFI Table
-	 */
-	if (efi.systab == NULL)
-		printk(KERN_ERR PFX "Woah! Couldn't map the EFI system table.\n");
-	if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
-		printk(KERN_ERR PFX "Woah! EFI system table signature incorrect\n");
-	if ((efi.systab->hdr.revision >> 16) == 0)
-		printk(KERN_ERR PFX "Warning: EFI system table version "
-		       "%d.%02d, expected 1.00 or greater\n",
-		       efi.systab->hdr.revision >> 16,
-		       efi.systab->hdr.revision & 0xffff);
-
-	/*
-	 * Grab some details from the system table
-	 */
-	num_config_tables = efi.systab->nr_tables;
-	config_tables = (efi_config_table_t *)efi.systab->tables;
-	runtime = efi.systab->runtime;
-
-	/*
-	 * Show what we know for posterity
-	 */
-	c16 = (efi_char16_t *) boot_ioremap(efi.systab->fw_vendor, 2);
-	if (c16) {
-		for (i = 0; i < (sizeof(vendor) - 1) && *c16; ++i)
-			vendor[i] = *c16++;
-		vendor[i] = '\0';
-	} else
-		printk(KERN_ERR PFX "Could not map the firmware vendor!\n");
-
-	printk(KERN_INFO PFX "EFI v%u.%.02u by %s \n",
-	       efi.systab->hdr.revision >> 16,
-	       efi.systab->hdr.revision & 0xffff, vendor);
-
-	/*
-	 * Let's see what config tables the firmware passed to us.
-	 */
-	config_tables = (efi_config_table_t *)
-				boot_ioremap((unsigned long) config_tables,
-			        num_config_tables * sizeof(efi_config_table_t));
-
-	if (config_tables == NULL)
-		printk(KERN_ERR PFX "Could not map EFI Configuration Table!\n");
-
-	efi.mps        = EFI_INVALID_TABLE_ADDR;
-	efi.acpi       = EFI_INVALID_TABLE_ADDR;
-	efi.acpi20     = EFI_INVALID_TABLE_ADDR;
-	efi.smbios     = EFI_INVALID_TABLE_ADDR;
-	efi.sal_systab = EFI_INVALID_TABLE_ADDR;
-	efi.boot_info  = EFI_INVALID_TABLE_ADDR;
-	efi.hcdp       = EFI_INVALID_TABLE_ADDR;
-	efi.uga        = EFI_INVALID_TABLE_ADDR;
-
-	for (i = 0; i < num_config_tables; i++) {
-		if (efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID) == 0) {
-			efi.mps = config_tables[i].table;
-			printk(KERN_INFO " MPS=0x%lx ", config_tables[i].table);
-		} else
-		    if (efi_guidcmp(config_tables[i].guid, ACPI_20_TABLE_GUID) == 0) {
-			efi.acpi20 = config_tables[i].table;
-			printk(KERN_INFO " ACPI 2.0=0x%lx ", config_tables[i].table);
-		} else
-		    if (efi_guidcmp(config_tables[i].guid, ACPI_TABLE_GUID) == 0) {
-			efi.acpi = config_tables[i].table;
-			printk(KERN_INFO " ACPI=0x%lx ", config_tables[i].table);
-		} else
-		    if (efi_guidcmp(config_tables[i].guid, SMBIOS_TABLE_GUID) == 0) {
-			efi.smbios = config_tables[i].table;
-			printk(KERN_INFO " SMBIOS=0x%lx ", config_tables[i].table);
-		} else
-		    if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) {
-			efi.hcdp = config_tables[i].table;
-			printk(KERN_INFO " HCDP=0x%lx ", config_tables[i].table);
-		} else
-		    if (efi_guidcmp(config_tables[i].guid, UGA_IO_PROTOCOL_GUID) == 0) {
-			efi.uga = config_tables[i].table;
-			printk(KERN_INFO " UGA=0x%lx ", config_tables[i].table);
-		}
-	}
-	printk("\n");
-
-	/*
-	 * Check out the runtime services table. We need to map
-	 * the runtime services table so that we can grab the physical
-	 * address of several of the EFI runtime functions, needed to
-	 * set the firmware into virtual mode.
-	 */
-
-	runtime = (efi_runtime_services_t *) boot_ioremap((unsigned long)
-						runtime,
-				      		sizeof(efi_runtime_services_t));
-	if (runtime != NULL) {
-		/*
-	 	 * We will only need *early* access to the following
-		 * two EFI runtime services before set_virtual_address_map
-		 * is invoked.
- 	 	 */
-		efi_phys.get_time = (efi_get_time_t *) runtime->get_time;
-		efi_phys.set_virtual_address_map =
-			(efi_set_virtual_address_map_t *)
-				runtime->set_virtual_address_map;
-	} else
-		printk(KERN_ERR PFX "Could not map the runtime service table!\n");
-
-	/* Map the EFI memory map for use until paging_init() */
-	memmap.map = boot_ioremap(boot_params.efi_info.efi_memmap,
-				  boot_params.efi_info.efi_memmap_size);
-	if (memmap.map == NULL)
-		printk(KERN_ERR PFX "Could not map the EFI memory map!\n");
-
-	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
-
-#if EFI_DEBUG
-	print_efi_memmap();
-#endif
-}
-
-static inline void __init check_range_for_systab(efi_memory_desc_t *md)
-{
-	if (((unsigned long)md->phys_addr <= (unsigned long)efi_phys.systab) &&
-		((unsigned long)efi_phys.systab < md->phys_addr +
-		((unsigned long)md->num_pages << EFI_PAGE_SHIFT))) {
-		unsigned long addr;
-
-		addr = md->virt_addr - md->phys_addr +
-			(unsigned long)efi_phys.systab;
-		efi.systab = (efi_system_table_t *)addr;
-	}
-}
-
-/*
- * Wrap all the virtual calls in a way that forces the parameters on the stack.
- */
-
-#define efi_call_virt(f, args...) \
-     ((efi_##f##_t __attribute__((regparm(0)))*)efi.systab->runtime->f)(args)
-
-static efi_status_t virt_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc)
-{
-	return efi_call_virt(get_time, tm, tc);
-}
-
-static efi_status_t virt_efi_set_time (efi_time_t *tm)
-{
-	return efi_call_virt(set_time, tm);
-}
-
-static efi_status_t virt_efi_get_wakeup_time (efi_bool_t *enabled,
-					      efi_bool_t *pending,
-					      efi_time_t *tm)
-{
-	return efi_call_virt(get_wakeup_time, enabled, pending, tm);
-}
-
-static efi_status_t virt_efi_set_wakeup_time (efi_bool_t enabled,
-					      efi_time_t *tm)
-{
-	return efi_call_virt(set_wakeup_time, enabled, tm);
-}
-
-static efi_status_t virt_efi_get_variable (efi_char16_t *name,
-					   efi_guid_t *vendor, u32 *attr,
-					   unsigned long *data_size, void *data)
-{
-	return efi_call_virt(get_variable, name, vendor, attr, data_size, data);
-}
-
-static efi_status_t virt_efi_get_next_variable (unsigned long *name_size,
-						efi_char16_t *name,
-						efi_guid_t *vendor)
-{
-	return efi_call_virt(get_next_variable, name_size, name, vendor);
-}
-
-static efi_status_t virt_efi_set_variable (efi_char16_t *name,
-					   efi_guid_t *vendor,
-					   unsigned long attr,
-					   unsigned long data_size, void *data)
-{
-	return efi_call_virt(set_variable, name, vendor, attr, data_size, data);
-}
-
-static efi_status_t virt_efi_get_next_high_mono_count (u32 *count)
-{
-	return efi_call_virt(get_next_high_mono_count, count);
-}
-
-static void virt_efi_reset_system (int reset_type, efi_status_t status,
-				   unsigned long data_size,
-				   efi_char16_t *data)
-{
-	efi_call_virt(reset_system, reset_type, status, data_size, data);
-}
-
-/*
- * This function will switch the EFI runtime services to virtual mode.
- * Essentially, look through the EFI memmap and map every region that
- * has the runtime attribute bit set in its memory descriptor and update
- * that memory descriptor with the virtual address obtained from ioremap().
- * This enables the runtime services to be called without having to
- * thunk back into physical mode for every invocation.
- */
-
-void __init efi_enter_virtual_mode(void)
-{
-	efi_memory_desc_t *md;
-	efi_status_t status;
-	void *p;
-
-	efi.systab = NULL;
-
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-
-		if (!(md->attribute & EFI_MEMORY_RUNTIME))
-			continue;
-
-		md->virt_addr = (unsigned long)ioremap(md->phys_addr,
-			md->num_pages << EFI_PAGE_SHIFT);
-		if (!(unsigned long)md->virt_addr) {
-			printk(KERN_ERR PFX "ioremap of 0x%lX failed\n",
-				(unsigned long)md->phys_addr);
-		}
-		/* update the virtual address of the EFI system table */
-		check_range_for_systab(md);
-	}
-
-	BUG_ON(!efi.systab);
-
-	status = phys_efi_set_virtual_address_map(
-			memmap.desc_size * memmap.nr_map,
-			memmap.desc_size,
-			memmap.desc_version,
-		       	memmap.phys_map);
-
-	if (status != EFI_SUCCESS) {
-		printk (KERN_ALERT "You are screwed! "
-			"Unable to switch EFI into virtual mode "
-			"(status=%lx)\n", status);
-		panic("EFI call to SetVirtualAddressMap() failed!");
-	}
-
-	/*
-	 * Now that EFI is in virtual mode, update the function
-	 * pointers in the runtime service table to the new virtual addresses.
-	 */
-
-	efi.get_time = virt_efi_get_time;
-	efi.set_time = virt_efi_set_time;
-	efi.get_wakeup_time = virt_efi_get_wakeup_time;
-	efi.set_wakeup_time = virt_efi_set_wakeup_time;
-	efi.get_variable = virt_efi_get_variable;
-	efi.get_next_variable = virt_efi_get_next_variable;
-	efi.set_variable = virt_efi_set_variable;
-	efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count;
-	efi.reset_system = virt_efi_reset_system;
-}
-
 void __init
 efi_initialize_iomem_resources(struct resource *code_resource,
 			       struct resource *data_resource,
@@ -683,35 +289,3 @@ efi_initialize_iomem_resources(struct resource *code_resource,
 		}
 	}
 }
-
-/*
- * Convenience functions to obtain memory types and attributes
- */
-
-u32 efi_mem_type(unsigned long phys_addr)
-{
-	efi_memory_desc_t *md;
-	void *p;
-
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-		if ((md->phys_addr <= phys_addr) && (phys_addr <
-			(md->phys_addr + (md-> num_pages << EFI_PAGE_SHIFT)) ))
-			return md->type;
-	}
-	return 0;
-}
-
-u64 efi_mem_attributes(unsigned long phys_addr)
-{
-	efi_memory_desc_t *md;
-	void *p;
-
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-		if ((md->phys_addr <= phys_addr) && (phys_addr <
-			(md->phys_addr + (md-> num_pages << EFI_PAGE_SHIFT)) ))
-			return md->attribute;
-	}
-	return 0;
-}
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index e9ede0fc585a..32fc87adc4a3 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -618,16 +618,9 @@ void __init setup_arch(char **cmdline_p)
 	pre_setup_arch_hook();
 	early_cpu_init();
 
-	/*
-	 * FIXME: This isn't an official loader_type right
-	 * now but does currently work with elilo.
-	 * If we were configured as an EFI kernel, check to make
-	 * sure that we were loaded correctly from elilo and that
-	 * the system table is valid.  If not, then initialize normally.
-	 */
 #ifdef CONFIG_EFI
-	if ((boot_params.hdr.type_of_loader == 0x50) &&
-	    boot_params.efi_info.efi_systab)
+	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     "EL32", 4))
 		efi_enabled = 1;
 #endif
 
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index 1459d3d58653..6d54502755aa 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -2,6 +2,48 @@
 #define _ASM_X86_EFI_H
 
 #ifdef CONFIG_X86_32
+
+extern unsigned long asmlinkage efi_call_phys(void *, ...);
+
+#define efi_call_phys0(f)		efi_call_phys(f)
+#define efi_call_phys1(f, a1)		efi_call_phys(f, a1)
+#define efi_call_phys2(f, a1, a2)	efi_call_phys(f, a1, a2)
+#define efi_call_phys3(f, a1, a2, a3)	efi_call_phys(f, a1, a2, a3)
+#define efi_call_phys4(f, a1, a2, a3, a4)	\
+	efi_call_phys(f, a1, a2, a3, a4)
+#define efi_call_phys5(f, a1, a2, a3, a4, a5)	\
+	efi_call_phys(f, a1, a2, a3, a4, a5)
+#define efi_call_phys6(f, a1, a2, a3, a4, a5, a6)	\
+	efi_call_phys(f, a1, a2, a3, a4, a5, a6)
+/*
+ * Wrap all the virtual calls in a way that forces the parameters on the stack.
+ */
+
+#define efi_call_virt(f, args...) \
+     ((efi_##f##_t __attribute__((regparm(0)))*)efi.systab->runtime->f)(args)
+
+#define efi_call_virt0(f)		efi_call_virt(f)
+#define efi_call_virt1(f, a1)		efi_call_virt(f, a1)
+#define efi_call_virt2(f, a1, a2)	efi_call_virt(f, a1, a2)
+#define efi_call_virt3(f, a1, a2, a3)	efi_call_virt(f, a1, a2, a3)
+#define efi_call_virt4(f, a1, a2, a3, a4)	\
+	efi_call_virt(f, a1, a2, a3, a4)
+#define efi_call_virt5(f, a1, a2, a3, a4, a5)	\
+	efi_call_virt(f, a1, a2, a3, a4, a5)
+#define efi_call_virt6(f, a1, a2, a3, a4, a5, a6)	\
+	efi_call_virt(f, a1, a2, a3, a4, a5, a6)
+/*
+ * We require an early boot_ioremap mapping mechanism initially
+ */
+extern void *boot_ioremap(unsigned long, unsigned long);
+
+#define efi_early_ioremap(addr, size)		boot_ioremap(addr, size)
+#define efi_early_iounmap(vaddr, size)
+
+#define efi_ioremap(addr, size)			ioremap(addr, size)
+
+#define end_pfn_map				max_low_pfn
+
 #else /* !CONFIG_X86_32 */
 
 #define MAX_EFI_IO_PAGES	100
-- 
cgit v1.2.3


From 2215e69d2cf5024647f9a034807990590d25dd4e Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:31:19 +0100
Subject: x86 boot: use E820 memory map on EFI 32 platform

Because the EFI memory map are converted to e820 memory map in bootloader, the
EFI memory map handling code is removed to clean up.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_32.c  | 117 +++--------------------------------
 arch/x86/kernel/efi_32.c   | 150 ---------------------------------------------
 arch/x86/kernel/setup_32.c |  16 ++---
 arch/x86/mm/init_32.c      |  18 ------
 include/asm-x86/e820_32.h  |   2 +-
 5 files changed, 16 insertions(+), 287 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c
index 56335a85a15a..931934a7b353 100644
--- a/arch/x86/kernel/e820_32.c
+++ b/arch/x86/kernel/e820_32.c
@@ -7,7 +7,6 @@
 #include <linux/kexec.h>
 #include <linux/module.h>
 #include <linux/mm.h>
-#include <linux/efi.h>
 #include <linux/pfn.h>
 #include <linux/uaccess.h>
 #include <linux/suspend.h>
@@ -181,7 +180,7 @@ static void __init probe_roms(void)
  * Request address space for all standard RAM and ROM resources
  * and also for regions reported as reserved by the e820.
  */
-void __init legacy_init_iomem_resources(struct resource *code_resource,
+void __init init_iomem_resources(struct resource *code_resource,
 		struct resource *data_resource,
 		struct resource *bss_resource)
 {
@@ -261,19 +260,17 @@ void __init add_memory_region(unsigned long long start,
 {
 	int x;
 
-	if (!efi_enabled) {
-       		x = e820.nr_map;
+	x = e820.nr_map;
 
-		if (x == E820MAX) {
-		    printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
-		    return;
-		}
-
-		e820.map[x].addr = start;
-		e820.map[x].size = size;
-		e820.map[x].type = type;
-		e820.nr_map++;
+	if (x == E820MAX) {
+		printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
+		return;
 	}
+
+	e820.map[x].addr = start;
+	e820.map[x].size = size;
+	e820.map[x].type = type;
+	e820.nr_map++;
 } /* add_memory_region */
 
 /*
@@ -488,29 +485,6 @@ int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
 	return 0;
 }
 
-/*
- * Callback for efi_memory_walk.
- */
-static int __init
-efi_find_max_pfn(unsigned long start, unsigned long end, void *arg)
-{
-	unsigned long *max_pfn = arg, pfn;
-
-	if (start < end) {
-		pfn = PFN_UP(end -1);
-		if (pfn > *max_pfn)
-			*max_pfn = pfn;
-	}
-	return 0;
-}
-
-static int __init
-efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg)
-{
-	memory_present(0, PFN_UP(start), PFN_DOWN(end));
-	return 0;
-}
-
 /*
  * Find the highest page frame number we have available
  */
@@ -519,11 +493,6 @@ void __init find_max_pfn(void)
 	int i;
 
 	max_pfn = 0;
-	if (efi_enabled) {
-		efi_memmap_walk(efi_find_max_pfn, &max_pfn);
-		efi_memmap_walk(efi_memory_present_wrapper, NULL);
-		return;
-	}
 
 	for (i = 0; i < e820.nr_map; i++) {
 		unsigned long start, end;
@@ -540,24 +509,6 @@ void __init find_max_pfn(void)
 	}
 }
 
-/*
- * Free all available memory for boot time allocation.  Used
- * as a callback function by efi_memory_walk()
- */
-
-static int __init
-free_available_memory(unsigned long start, unsigned long end, void *arg)
-{
-	/* check max_low_pfn */
-	if (start >= (max_low_pfn << PAGE_SHIFT))
-		return 0;
-	if (end >= (max_low_pfn << PAGE_SHIFT))
-		end = max_low_pfn << PAGE_SHIFT;
-	if (start < end)
-		free_bootmem(start, end - start);
-
-	return 0;
-}
 /*
  * Register fully available low RAM pages with the bootmem allocator.
  */
@@ -565,10 +516,6 @@ void __init register_bootmem_low_pages(unsigned long max_low_pfn)
 {
 	int i;
 
-	if (efi_enabled) {
-		efi_memmap_walk(free_available_memory, NULL);
-		return;
-	}
 	for (i = 0; i < e820.nr_map; i++) {
 		unsigned long curr_pfn, last_pfn, size;
 		/*
@@ -676,56 +623,12 @@ void __init print_memory_map(char *who)
 	}
 }
 
-static __init __always_inline void efi_limit_regions(unsigned long long size)
-{
-	unsigned long long current_addr = 0;
-	efi_memory_desc_t *md, *next_md;
-	void *p, *p1;
-	int i, j;
-
-	j = 0;
-	p1 = memmap.map;
-	for (p = p1, i = 0; p < memmap.map_end; p += memmap.desc_size, i++) {
-		md = p;
-		next_md = p1;
-		current_addr = md->phys_addr +
-			PFN_PHYS(md->num_pages);
-		if (is_available_memory(md)) {
-			if (md->phys_addr >= size) continue;
-			memcpy(next_md, md, memmap.desc_size);
-			if (current_addr >= size) {
-				next_md->num_pages -=
-					PFN_UP(current_addr-size);
-			}
-			p1 += memmap.desc_size;
-			next_md = p1;
-			j++;
-		} else if ((md->attribute & EFI_MEMORY_RUNTIME) ==
-			   EFI_MEMORY_RUNTIME) {
-			/* In order to make runtime services
-			 * available we have to include runtime
-			 * memory regions in memory map */
-			memcpy(next_md, md, memmap.desc_size);
-			p1 += memmap.desc_size;
-			next_md = p1;
-			j++;
-		}
-	}
-	memmap.nr_map = j;
-	memmap.map_end = memmap.map +
-		(memmap.nr_map * memmap.desc_size);
-}
-
 void __init limit_regions(unsigned long long size)
 {
 	unsigned long long current_addr;
 	int i;
 
 	print_memory_map("limit_regions start");
-	if (efi_enabled) {
-		efi_limit_regions(size);
-		return;
-	}
 	for (i = 0; i < e820.nr_map; i++) {
 		current_addr = e820.map[i].addr + e820.map[i].size;
 		if (current_addr < size)
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index 1df13725e519..30f937116288 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -125,22 +125,6 @@ void efi_call_phys_epilog(void) __releases(efi_rt_lock)
 	spin_unlock(&efi_rt_lock);
 }
 
-int is_available_memory(efi_memory_desc_t * md)
-{
-	if (!(md->attribute & EFI_MEMORY_WB))
-		return 0;
-
-	switch (md->type) {
-		case EFI_LOADER_CODE:
-		case EFI_LOADER_DATA:
-		case EFI_BOOT_SERVICES_CODE:
-		case EFI_BOOT_SERVICES_DATA:
-		case EFI_CONVENTIONAL_MEMORY:
-			return 1;
-	}
-	return 0;
-}
-
 /*
  * We need to map the EFI memory map again after paging_init().
  */
@@ -155,137 +139,3 @@ void __init efi_map_memmap(void)
 
 	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
 }
-
-/*
- * Walks the EFI memory map and calls CALLBACK once for each EFI
- * memory descriptor that has memory that is available for kernel use.
- */
-void efi_memmap_walk(efi_freemem_callback_t callback, void *arg)
-{
-	int prev_valid = 0;
-	struct range {
-		unsigned long start;
-		unsigned long end;
-	} uninitialized_var(prev), curr;
-	efi_memory_desc_t *md;
-	unsigned long start, end;
-	void *p;
-
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-
-		if ((md->num_pages == 0) || (!is_available_memory(md)))
-			continue;
-
-		curr.start = md->phys_addr;
-		curr.end = curr.start + (md->num_pages << EFI_PAGE_SHIFT);
-
-		if (!prev_valid) {
-			prev = curr;
-			prev_valid = 1;
-		} else {
-			if (curr.start < prev.start)
-				printk(KERN_INFO PFX "Unordered memory map\n");
-			if (prev.end == curr.start)
-				prev.end = curr.end;
-			else {
-				start =
-				    (unsigned long) (PAGE_ALIGN(prev.start));
-				end = (unsigned long) (prev.end & PAGE_MASK);
-				if ((end > start)
-				    && (*callback) (start, end, arg) < 0)
-					return;
-				prev = curr;
-			}
-		}
-	}
-	if (prev_valid) {
-		start = (unsigned long) PAGE_ALIGN(prev.start);
-		end = (unsigned long) (prev.end & PAGE_MASK);
-		if (end > start)
-			(*callback) (start, end, arg);
-	}
-}
-
-void __init
-efi_initialize_iomem_resources(struct resource *code_resource,
-			       struct resource *data_resource,
-			       struct resource *bss_resource)
-{
-	struct resource *res;
-	efi_memory_desc_t *md;
-	void *p;
-
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-
-		if ((md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT)) >
-		    0x100000000ULL)
-			continue;
-		res = kzalloc(sizeof(struct resource), GFP_ATOMIC);
-		switch (md->type) {
-		case EFI_RESERVED_TYPE:
-			res->name = "Reserved Memory";
-			break;
-		case EFI_LOADER_CODE:
-			res->name = "Loader Code";
-			break;
-		case EFI_LOADER_DATA:
-			res->name = "Loader Data";
-			break;
-		case EFI_BOOT_SERVICES_DATA:
-			res->name = "BootServices Data";
-			break;
-		case EFI_BOOT_SERVICES_CODE:
-			res->name = "BootServices Code";
-			break;
-		case EFI_RUNTIME_SERVICES_CODE:
-			res->name = "Runtime Service Code";
-			break;
-		case EFI_RUNTIME_SERVICES_DATA:
-			res->name = "Runtime Service Data";
-			break;
-		case EFI_CONVENTIONAL_MEMORY:
-			res->name = "Conventional Memory";
-			break;
-		case EFI_UNUSABLE_MEMORY:
-			res->name = "Unusable Memory";
-			break;
-		case EFI_ACPI_RECLAIM_MEMORY:
-			res->name = "ACPI Reclaim";
-			break;
-		case EFI_ACPI_MEMORY_NVS:
-			res->name = "ACPI NVS";
-			break;
-		case EFI_MEMORY_MAPPED_IO:
-			res->name = "Memory Mapped IO";
-			break;
-		case EFI_MEMORY_MAPPED_IO_PORT_SPACE:
-			res->name = "Memory Mapped IO Port Space";
-			break;
-		default:
-			res->name = "Reserved";
-			break;
-		}
-		res->start = md->phys_addr;
-		res->end = res->start + ((md->num_pages << EFI_PAGE_SHIFT) - 1);
-		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-		if (request_resource(&iomem_resource, res) < 0)
-			printk(KERN_ERR PFX "Failed to allocate res %s : "
-				"0x%llx-0x%llx\n", res->name,
-				(unsigned long long)res->start,
-				(unsigned long long)res->end);
-		/*
-		 * We don't know which region contains kernel data so we try
-		 * it repeatedly and let the resource manager test it.
-		 */
-		if (md->type == EFI_CONVENTIONAL_MEMORY) {
-			request_resource(res, code_resource);
-			request_resource(res, data_resource);
-			request_resource(res, bss_resource);
-#ifdef CONFIG_KEXEC
-			request_resource(res, &crashk_res);
-#endif
-		}
-	}
-}
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 32fc87adc4a3..2e805da337a2 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -644,12 +644,12 @@ void __init setup_arch(char **cmdline_p)
 	rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
 #endif
 	ARCH_SETUP
+
+	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
+	print_memory_map(memory_setup());
+
 	if (efi_enabled)
 		efi_init();
-	else {
-		printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-		print_memory_map(memory_setup());
-	}
 
 	copy_edd();
 
@@ -769,14 +769,8 @@ static int __init request_standard_resources(void)
 	int i;
 
 	printk(KERN_INFO "Setting up standard PCI resources\n");
-	if (efi_enabled)
-		efi_initialize_iomem_resources(&code_resource,
-				&data_resource, &bss_resource);
-	else
-		legacy_init_iomem_resources(&code_resource,
-				&data_resource, &bss_resource);
+	init_iomem_resources(&code_resource, &data_resource, &bss_resource);
 
-	/* EFI systems may still have VGA */
 	request_resource(&iomem_resource, &video_ram_resource);
 
 	/* request I/O space for devices used on all i[345]86 PCs */
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index e6e34c7dcabf..29130970c193 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -27,7 +27,6 @@
 #include <linux/bootmem.h>
 #include <linux/slab.h>
 #include <linux/proc_fs.h>
-#include <linux/efi.h>
 #include <linux/memory_hotplug.h>
 #include <linux/initrd.h>
 #include <linux/cpumask.h>
@@ -216,23 +215,6 @@ int page_is_ram(unsigned long pagenr)
 	int i;
 	unsigned long addr, end;
 
-	if (efi_enabled) {
-		efi_memory_desc_t *md;
-		void *p;
-
-		for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-			md = p;
-			if (!is_available_memory(md))
-				continue;
-			addr = (md->phys_addr+PAGE_SIZE-1) >> PAGE_SHIFT;
-			end = (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT)) >> PAGE_SHIFT;
-
-			if ((pagenr >= addr) && (pagenr < end))
-				return 1;
-		}
-		return 0;
-	}
-
 	for (i = 0; i < e820.nr_map; i++) {
 
 		if (e820.map[i].type != E820_RAM)	/* not usable memory */
diff --git a/include/asm-x86/e820_32.h b/include/asm-x86/e820_32.h
index ae5ea19623fa..e2faf5f3a0bb 100644
--- a/include/asm-x86/e820_32.h
+++ b/include/asm-x86/e820_32.h
@@ -28,7 +28,7 @@ extern void register_bootmem_low_pages(unsigned long max_low_pfn);
 extern void e820_register_memory(void);
 extern void limit_regions(unsigned long long size);
 extern void print_memory_map(char *who);
-extern void legacy_init_iomem_resources(struct resource *code_resource,
+extern void init_iomem_resources(struct resource *code_resource,
 			    struct resource *data_resource,
 			    struct resource *bss_resource);
 
-- 
cgit v1.2.3


From f6e0eba1b133192e298052d616e9e17356d32517 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:20 +0100
Subject: x86: avoid ifdefs in desc.h, getting rid of pack_ldt and pack_tss

By Andi Kleen's suggestion, this patch removes pack_ldt() and pack_tss()
wrappers in favour of a general wrapper. It saves us an ifdef and some lines
of code, but more importantly, it's more elegant.

No functional change is made, although the object code is expected to be
different.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc.h | 58 ++++++++++++++++----------------------------------
 1 file changed, 18 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index c31715cf2218..e9356a852020 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -125,21 +125,6 @@ static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry,
 	memcpy(&gdt[entry], desc, size);
 }
 
-static inline void set_tssldt_descriptor(struct ldttss_desc64 *d,
-					 unsigned long tss, unsigned type,
-					 unsigned size)
-{
-	memset(d, 0, sizeof(*d));
-	d->limit0 = size & 0xFFFF;
-	d->base0 = PTR_LOW(tss);
-	d->base1 = PTR_MIDDLE(tss) & 0xFF;
-	d->type = type;
-	d->p = 1;
-	d->limit1 = (size >> 16) & 0xF;
-	d->base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
-	d->base3 = PTR_HIGH(tss);
-}
-
 static inline void pack_descriptor(struct desc_struct *desc, unsigned long base,
 				   unsigned long limit, unsigned char type,
 				   unsigned char flags)
@@ -151,30 +136,24 @@ static inline void pack_descriptor(struct desc_struct *desc, unsigned long base,
 	desc->p = 1;
 }
 
-static inline void pack_ldt(ldt_desc *ldt, unsigned long addr,
-			   unsigned size)
-{
 
-#ifdef CONFIG_X86_64
-		set_tssldt_descriptor(ldt,
-			     addr, DESC_LDT, size);
-#else
-		pack_descriptor(ldt, (unsigned long)addr,
-				size,
-				0x80 | DESC_LDT, 0);
-#endif
-}
-
-static inline void pack_tss(tss_desc *tss, unsigned long addr,
-			   unsigned size, unsigned entry)
+static inline void set_tssldt_descriptor(void *d, unsigned long addr,
+					 unsigned type, unsigned size)
 {
 #ifdef CONFIG_X86_64
-		set_tssldt_descriptor(tss,
-			     addr, entry, size);
+	struct ldttss_desc64 *desc = d;
+	memset(desc, 0, sizeof(*desc));
+	desc->limit0 = size & 0xFFFF;
+	desc->base0 = PTR_LOW(addr);
+	desc->base1 = PTR_MIDDLE(addr) & 0xFF;
+	desc->type = type;
+	desc->p = 1;
+	desc->limit1 = (size >> 16) & 0xF;
+	desc->base2 = (PTR_MIDDLE(addr) >> 8) & 0xFF;
+	desc->base3 = PTR_HIGH(addr);
 #else
-		pack_descriptor(tss, (unsigned long)addr,
-				size,
-				0x80 | entry, 0);
+
+	pack_descriptor((struct desc_struct *)d, addr, size, 0x80 | type, 0);
 #endif
 }
 
@@ -190,9 +169,8 @@ static inline void __set_tss_desc(unsigned cpu, unsigned int entry, void *addr)
 	 * -1? seg base+limit should be pointing to the address of the
 	 * last valid byte
 	 */
-	pack_tss(&tss, (unsigned long)addr,
-		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1,
-		DESC_TSS);
+	set_tssldt_descriptor(&tss, (unsigned long)addr, DESC_TSS,
+		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
 	write_gdt_entry(d, entry, &tss, DESC_TSS);
 }
 
@@ -206,8 +184,8 @@ static inline void native_set_ldt(const void *addr, unsigned int entries)
 		unsigned cpu = smp_processor_id();
 		ldt_desc ldt;
 
-		pack_ldt(&ldt, (unsigned long)addr,
-				entries * sizeof(ldt) - 1);
+		set_tssldt_descriptor(&ldt, (unsigned long)addr,
+				      DESC_LDT, entries * sizeof(ldt) - 1);
 		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT,
 				&ldt, DESC_LDT);
 		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
-- 
cgit v1.2.3


From 3c68904fee1459b6d51040864e15d19098eedef7 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:31:20 +0100
Subject: x86, ptrace: use jiffies for BTS timestamps

Replace sched_clock() with jiffies for BTS timestamps.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ds.c     | 30 ++++++++++--------------------
 arch/x86/kernel/ptrace.c |  2 +-
 include/asm-x86/ds.h     |  2 +-
 3 files changed, 12 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index 996a7c4f5963..e7855def97c3 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -167,23 +167,13 @@ static inline void set_info_type(char *base, unsigned char value)
 {
 	(*(unsigned char *)(base + ds_cfg.info_type.offset)) = value;
 }
-/*
- * The info data might overlap with the info type on some architectures.
- * We therefore read and write the exact number of bytes.
- */
-static inline unsigned long long get_info_data(char *base)
+static inline unsigned long get_info_data(char *base)
 {
-	unsigned long long value = 0;
-	memcpy(&value,
-	       base + ds_cfg.info_data.offset,
-	       ds_cfg.info_data.size);
-	return value;
+	return *(unsigned long *)(base + ds_cfg.info_data.offset);
 }
-static inline void set_info_data(char *base, unsigned long long value)
+static inline void set_info_data(char *base, unsigned long value)
 {
-	memcpy(base + ds_cfg.info_data.offset,
-	       &value,
-	       ds_cfg.info_data.size);
+	(*(unsigned long *)(base + ds_cfg.info_data.offset)) = value;
 }
 
 
@@ -282,8 +272,8 @@ int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
 
 	memset(out, 0, sizeof(*out));
 	if (get_from_ip(bts) == BTS_ESCAPE_ADDRESS) {
-		out->qualifier         = get_info_type(bts);
-		out->variant.timestamp = get_info_data(bts);
+		out->qualifier       = get_info_type(bts);
+		out->variant.jiffies = get_info_data(bts);
 	} else {
 		out->qualifier = BTS_BRANCH;
 		out->variant.lbr.from_ip = get_from_ip(bts);
@@ -319,7 +309,7 @@ int ds_write_bts(void *ds, const struct bts_struct *in)
 	case BTS_TASK_DEPARTS:
 		set_from_ip(bts, BTS_ESCAPE_ADDRESS);
 		set_info_type(bts, in->qualifier);
-		set_info_data(bts, in->variant.timestamp);
+		set_info_data(bts, in->variant.jiffies);
 		break;
 
 	default:
@@ -350,7 +340,7 @@ static const struct ds_configuration ds_cfg_netburst = {
 	.from_ip = { 0, 4 },
 	.to_ip = { 4, 4 },
 	.info_type = { 4, 1 },
-	.info_data = { 5, 7 },
+	.info_data = { 8, 4 },
 	.debugctl_mask = (1<<2)|(1<<3)
 };
 
@@ -364,7 +354,7 @@ static const struct ds_configuration ds_cfg_pentium_m = {
 	.from_ip = { 0, 4 },
 	.to_ip = { 4, 4 },
 	.info_type = { 4, 1 },
-	.info_data = { 5, 7 },
+	.info_data = { 8, 4 },
 	.debugctl_mask = (1<<6)|(1<<7)
 };
 #endif /* _i386_ */
@@ -379,7 +369,7 @@ static const struct ds_configuration ds_cfg_core2 = {
 	.from_ip = { 0, 8 },
 	.to_ip = { 8, 8 },
 	.info_type = { 8, 1 },
-	.info_data = { 9, 7 },
+	.info_data = { 16, 8 },
 	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
 };
 
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 8d0dd8b5effe..ec86abaab530 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -616,7 +616,7 @@ void ptrace_bts_take_timestamp(struct task_struct *tsk,
 {
 	struct bts_struct rec = {
 		.qualifier = qualifier,
-		.variant.timestamp = sched_clock()
+		.variant.jiffies = jiffies
 	};
 
 	if (ptrace_bts_get_buffer_size(tsk) <= 0)
diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h
index edd8467740a6..c9e15381dc7f 100644
--- a/include/asm-x86/ds.h
+++ b/include/asm-x86/ds.h
@@ -48,7 +48,7 @@ struct bts_struct {
 		} lbr;
 		/* BTS_TASK_ARRIVES or
 		   BTS_TASK_DEPARTS */
-		unsigned long long timestamp;
+		unsigned long jiffies;
 	} variant;
 };
 
-- 
cgit v1.2.3


From e4811f2568c55e595a7bf15a3b9aba863b31fb94 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:31:20 +0100
Subject: x86, ptrace: change BTS GET ptrace interface

Change the ptrace interface to mimick an array from newst to oldest.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ptrace.c     | 30 ++++++++++++++++--------------
 include/asm-x86/ptrace-abi.h | 17 ++++++-----------
 2 files changed, 22 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index ec86abaab530..3e78c124e2d2 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -479,26 +479,33 @@ static int ptrace_bts_get_buffer_size(struct task_struct *child)
 	return ds_get_bts_size((void *)child->thread.ds_area_msr);
 }
 
-static int ptrace_bts_get_index(struct task_struct *child)
-{
-	if (!child->thread.ds_area_msr)
-		return -ENXIO;
-
-	return ds_get_bts_index((void *)child->thread.ds_area_msr);
-}
-
 static int ptrace_bts_read_record(struct task_struct *child,
 				  long index,
 				  struct bts_struct __user *out)
 {
 	struct bts_struct ret;
 	int retval;
+	int bts_size;
+	int bts_index;
 
 	if (!child->thread.ds_area_msr)
 		return -ENXIO;
 
+	if (index < 0)
+		return -EINVAL;
+
+	bts_size = ds_get_bts_size((void *)child->thread.ds_area_msr);
+	if (bts_size <= index)
+		return -EINVAL;
+
+	/* translate the ptrace bts index into the ds bts index */
+	bts_index = ds_get_bts_index((void *)child->thread.ds_area_msr);
+	bts_index -= (index + 1);
+	if (bts_index < 0)
+		bts_index += bts_size;
+
 	retval = ds_read_bts((void *)child->thread.ds_area_msr,
-			     index, &ret);
+			     bts_index, &ret);
 	if (retval)
 		return retval;
 
@@ -813,10 +820,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		ret = ptrace_bts_get_buffer_size(child);
 		break;
 
-	case PTRACE_BTS_GET_INDEX:
-		ret = ptrace_bts_get_index(child);
-		break;
-
 	case PTRACE_BTS_READ_RECORD:
 		ret = ptrace_bts_read_record
 			(child, data,
@@ -1017,7 +1020,6 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 	case PTRACE_BTS_MAX_BUFFER_SIZE:
 	case PTRACE_BTS_ALLOCATE_BUFFER:
 	case PTRACE_BTS_GET_BUFFER_SIZE:
-	case PTRACE_BTS_GET_INDEX:
 	case PTRACE_BTS_READ_RECORD:
 	case PTRACE_BTS_CONFIG:
 	case PTRACE_BTS_STATUS:
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index 6fadc5214e14..b473ad45e9ca 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -99,32 +99,27 @@
    ENXIO........no buffer allocated */
 #define PTRACE_BTS_GET_BUFFER_SIZE 42
 
-/* Return the index of the next bts record to be written,
-   if successful; -1, otherwise.
-   EOPNOTSUPP...processor does not support bts tracing
-   ENXIO........no buffer allocated
-   After the first warp-around, this is the start of the circular bts buffer. */
-#define PTRACE_BTS_GET_INDEX 43
-
-/* Read the DATA'th bts record into a ptrace_bts_record buffer provided in ADDR.
+/* Read the DATA'th bts record into a ptrace_bts_record buffer
+   provided in ADDR.
+   Records are ordered from newest to oldest.
    Return 0, if successful; -1, otherwise
    EOPNOTSUPP...processor does not support bts tracing
    ENXIO........no buffer allocated
    EINVAL.......invalid index */
-#define PTRACE_BTS_READ_RECORD 44
+#define PTRACE_BTS_READ_RECORD 43
 
 /* Configure last branch trace; the configuration is given as a bit-mask of
    PTRACE_BTS_O_* options in DATA; parameter ADDR is ignored.
    Return 0, if successful; -1, otherwise
    EOPNOTSUPP...processor does not support bts tracing
    ENXIO........no buffer allocated */
-#define PTRACE_BTS_CONFIG 45
+#define PTRACE_BTS_CONFIG 44
 
 /* Return the configuration as bit-mask of PTRACE_BTS_O_* options
    if successful; -1, otherwise.
    EOPNOTSUPP...processor does not support bts tracing
    ENXIO........no buffer allocated */
-#define PTRACE_BTS_STATUS 46
+#define PTRACE_BTS_STATUS 45
 
 /* Trace configuration options */
 /* Collect last branch trace */
-- 
cgit v1.2.3


From a95d67f87e1a5f1b4429be3ba3bf7b4051657908 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:31:20 +0100
Subject: x86, ptrace: new ptrace BTS API

Here's the new ptrace BTS API that supports two different overflow handling mechanisms (wrap-around and buffer-full-signal) to support two different use cases (debugging and profiling).

It further combines buffer allocation and configuration.

Opens:
- memory rlimit
- overflow signal

What would be the right signal to use?

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ds.c         |  56 ++++++++++-
 arch/x86/kernel/ptrace.c     | 221 +++++++++++++++++++++++++------------------
 include/asm-x86/ds.h         |   7 ++
 include/asm-x86/ptrace-abi.h |  94 +++++++++---------
 include/asm-x86/ptrace.h     |   1 +
 5 files changed, 237 insertions(+), 142 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index e7855def97c3..6eb5d49a36bb 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -177,18 +177,20 @@ static inline void set_info_data(char *base, unsigned long value)
 }
 
 
-int ds_allocate(void **dsp, size_t bts_size_in_records)
+int ds_allocate(void **dsp, size_t bts_size_in_bytes)
 {
-	size_t bts_size_in_bytes = 0;
-	void *bts = 0;
-	void *ds = 0;
+	size_t bts_size_in_records;
+	void *bts;
+	void *ds;
 
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
 		return -EOPNOTSUPP;
 
-	if (bts_size_in_records < 0)
+	if (bts_size_in_bytes < 0)
 		return -EINVAL;
 
+	bts_size_in_records =
+		bts_size_in_bytes / ds_cfg.sizeof_bts;
 	bts_size_in_bytes =
 		bts_size_in_records * ds_cfg.sizeof_bts;
 
@@ -233,9 +235,21 @@ int ds_get_bts_size(void *ds)
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
 		return -EOPNOTSUPP;
 
+	if (!ds)
+		return 0;
+
 	size_in_bytes =
 		get_bts_absolute_maximum(ds) -
 		get_bts_buffer_base(ds);
+	return size_in_bytes;
+}
+
+int ds_get_bts_end(void *ds)
+{
+	size_t size_in_bytes = ds_get_bts_size(ds);
+
+	if (size_in_bytes <= 0)
+		return size_in_bytes;
 
 	return size_in_bytes / ds_cfg.sizeof_bts;
 }
@@ -254,6 +268,38 @@ int ds_get_bts_index(void *ds)
 	return index_offset_in_bytes / ds_cfg.sizeof_bts;
 }
 
+int ds_set_overflow(void *ds, int method)
+{
+	switch (method) {
+	case DS_O_SIGNAL:
+		return -EOPNOTSUPP;
+	case DS_O_WRAP:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+int ds_get_overflow(void *ds)
+{
+	return DS_O_WRAP;
+}
+
+int ds_clear(void *ds)
+{
+	int bts_size = ds_get_bts_size(ds);
+	void *bts_base;
+
+	if (bts_size <= 0)
+		return bts_size;
+
+	bts_base = get_bts_buffer_base(ds);
+	memset(bts_base, 0, bts_size);
+
+	set_bts_index(ds, bts_base);
+	return 0;
+}
+
 int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
 {
 	void *bts;
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 3e78c124e2d2..18972a305890 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -32,12 +32,6 @@
 #include <asm/ds.h>
 
 
-/*
- * The maximal size of a BTS buffer per traced task in number of BTS
- * records.
- */
-#define PTRACE_BTS_BUFFER_MAX 4000
-
 /*
  * does not yet catch signals sent when the child dies.
  * in exit.c or in signal.c.
@@ -466,17 +460,12 @@ static int ptrace_set_debugreg(struct task_struct *child,
 	return 0;
 }
 
-static int ptrace_bts_max_buffer_size(void)
-{
-	return PTRACE_BTS_BUFFER_MAX;
-}
-
-static int ptrace_bts_get_buffer_size(struct task_struct *child)
+static int ptrace_bts_get_size(struct task_struct *child)
 {
 	if (!child->thread.ds_area_msr)
 		return -ENXIO;
 
-	return ds_get_bts_size((void *)child->thread.ds_area_msr);
+	return ds_get_bts_index((void *)child->thread.ds_area_msr);
 }
 
 static int ptrace_bts_read_record(struct task_struct *child,
@@ -485,7 +474,7 @@ static int ptrace_bts_read_record(struct task_struct *child,
 {
 	struct bts_struct ret;
 	int retval;
-	int bts_size;
+	int bts_end;
 	int bts_index;
 
 	if (!child->thread.ds_area_msr)
@@ -494,15 +483,15 @@ static int ptrace_bts_read_record(struct task_struct *child,
 	if (index < 0)
 		return -EINVAL;
 
-	bts_size = ds_get_bts_size((void *)child->thread.ds_area_msr);
-	if (bts_size <= index)
+	bts_end = ds_get_bts_end((void *)child->thread.ds_area_msr);
+	if (bts_end <= index)
 		return -EINVAL;
 
 	/* translate the ptrace bts index into the ds bts index */
 	bts_index = ds_get_bts_index((void *)child->thread.ds_area_msr);
 	bts_index -= (index + 1);
 	if (bts_index < 0)
-		bts_index += bts_size;
+		bts_index += bts_end;
 
 	retval = ds_read_bts((void *)child->thread.ds_area_msr,
 			     bts_index, &ret);
@@ -530,19 +519,97 @@ static int ptrace_bts_write_record(struct task_struct *child,
 	return sizeof(*in);
 }
 
-static int ptrace_bts_config(struct task_struct *child,
-			     unsigned long options)
+static int ptrace_bts_clear(struct task_struct *child)
 {
-	unsigned long debugctl_mask = ds_debugctl_mask();
-	int retval;
+	if (!child->thread.ds_area_msr)
+		return -ENXIO;
 
-	retval = ptrace_bts_get_buffer_size(child);
-	if (retval < 0)
-		return retval;
-	if (retval == 0)
+	return ds_clear((void *)child->thread.ds_area_msr);
+}
+
+static int ptrace_bts_drain(struct task_struct *child,
+			    struct bts_struct __user *out)
+{
+	int end, i;
+	void *ds = (void *)child->thread.ds_area_msr;
+
+	if (!ds)
 		return -ENXIO;
 
-	if (options & PTRACE_BTS_O_TRACE_TASK) {
+	end = ds_get_bts_index(ds);
+	if (end <= 0)
+		return end;
+
+	for (i = 0; i < end; i++, out++) {
+		struct bts_struct ret;
+		int retval;
+
+		retval = ds_read_bts(ds, i, &ret);
+		if (retval < 0)
+			return retval;
+
+		if (copy_to_user(out, &ret, sizeof(ret)))
+			return -EFAULT;
+	}
+
+	ds_clear(ds);
+
+	return i;
+}
+
+static int ptrace_bts_config(struct task_struct *child,
+			     const struct ptrace_bts_config __user *ucfg)
+{
+	struct ptrace_bts_config cfg;
+	unsigned long debugctl_mask;
+	int bts_size, ret;
+	void *ds;
+
+	if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
+		return -EFAULT;
+
+	bts_size = 0;
+	ds = (void *)child->thread.ds_area_msr;
+	if (ds) {
+		bts_size = ds_get_bts_size(ds);
+		if (bts_size < 0)
+			return bts_size;
+	}
+
+	if (bts_size != cfg.size) {
+		ret = ds_free((void **)&child->thread.ds_area_msr);
+		if (ret < 0)
+			return ret;
+
+		if (cfg.size > 0)
+			ret = ds_allocate((void **)&child->thread.ds_area_msr,
+					  cfg.size);
+		ds = (void *)child->thread.ds_area_msr;
+		if (ds)
+			set_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+		else
+			clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+
+		if (ret < 0)
+			return ret;
+
+		bts_size = ds_get_bts_size(ds);
+		if (bts_size <= 0)
+			return bts_size;
+	}
+
+	if (ds) {
+		if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
+			ret = ds_set_overflow(ds, DS_O_SIGNAL);
+		} else {
+			ret = ds_set_overflow(ds, DS_O_WRAP);
+		}
+		if (ret < 0)
+			return ret;
+	}
+
+	debugctl_mask = ds_debugctl_mask();
+	if (ds && (cfg.flags & PTRACE_BTS_O_TRACE)) {
 		child->thread.debugctlmsr |= debugctl_mask;
 		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
 	} else {
@@ -555,7 +622,7 @@ static int ptrace_bts_config(struct task_struct *child,
 			clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
 	}
 
-	if (options & PTRACE_BTS_O_TIMESTAMPS)
+	if (ds && (cfg.flags & PTRACE_BTS_O_SCHED))
 		set_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
 	else
 		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
@@ -563,59 +630,32 @@ static int ptrace_bts_config(struct task_struct *child,
 	return 0;
 }
 
-static int ptrace_bts_status(struct task_struct *child)
+static int ptrace_bts_status(struct task_struct *child,
+			     struct ptrace_bts_config __user *ucfg)
 {
-	unsigned long debugctl_mask = ds_debugctl_mask();
-	int retval, status = 0;
-
-	retval = ptrace_bts_get_buffer_size(child);
-	if (retval < 0)
-		return retval;
-	if (retval == 0)
-		return -ENXIO;
-
-	if (ptrace_bts_get_buffer_size(child) <= 0)
-		return -ENXIO;
+	void *ds = (void *)child->thread.ds_area_msr;
+	struct ptrace_bts_config cfg;
 
-	if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
-	    child->thread.debugctlmsr & debugctl_mask)
-		status |= PTRACE_BTS_O_TRACE_TASK;
-	if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
-		status |= PTRACE_BTS_O_TIMESTAMPS;
+	memset(&cfg, 0, sizeof(cfg));
 
-	return status;
-}
+	if (ds) {
+		cfg.size = ds_get_bts_size(ds);
 
-static int ptrace_bts_allocate_bts(struct task_struct *child,
-				   int size_in_records)
-{
-	int retval = 0;
-	void *ds;
+		if (ds_get_overflow(ds) == DS_O_SIGNAL)
+			cfg.flags |= PTRACE_BTS_O_SIGNAL;
 
-	if (size_in_records < 0)
-		return -EINVAL;
+		if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
+		    child->thread.debugctlmsr & ds_debugctl_mask())
+			cfg.flags |= PTRACE_BTS_O_TRACE;
 
-	if (size_in_records > ptrace_bts_max_buffer_size())
-		return -EINVAL;
-
-	if (size_in_records == 0) {
-		ptrace_bts_config(child, /* options = */ 0);
-	} else {
-		retval = ds_allocate(&ds, size_in_records);
-		if (retval)
-			return retval;
+		if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
+			cfg.flags |= PTRACE_BTS_O_SCHED;
 	}
 
-	if (child->thread.ds_area_msr)
-		ds_free((void **)&child->thread.ds_area_msr);
-
-	child->thread.ds_area_msr = (unsigned long)ds;
-	if (child->thread.ds_area_msr)
-		set_tsk_thread_flag(child, TIF_DS_AREA_MSR);
-	else
-		clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
+	if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
+		return -EFAULT;
 
-	return retval;
+	return sizeof(cfg);
 }
 
 void ptrace_bts_take_timestamp(struct task_struct *tsk,
@@ -626,9 +666,6 @@ void ptrace_bts_take_timestamp(struct task_struct *tsk,
 		.variant.jiffies = jiffies
 	};
 
-	if (ptrace_bts_get_buffer_size(tsk) <= 0)
-		return;
-
 	ptrace_bts_write_record(tsk, &rec);
 }
 
@@ -808,30 +845,32 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 #endif
 
-	case PTRACE_BTS_MAX_BUFFER_SIZE:
-		ret = ptrace_bts_max_buffer_size();
+	case PTRACE_BTS_CONFIG:
+		ret = ptrace_bts_config
+			(child, (struct ptrace_bts_config __user *)addr);
 		break;
 
-	case PTRACE_BTS_ALLOCATE_BUFFER:
-		ret = ptrace_bts_allocate_bts(child, data);
+	case PTRACE_BTS_STATUS:
+		ret = ptrace_bts_status
+			(child, (struct ptrace_bts_config __user *)addr);
 		break;
 
-	case PTRACE_BTS_GET_BUFFER_SIZE:
-		ret = ptrace_bts_get_buffer_size(child);
+	case PTRACE_BTS_SIZE:
+		ret = ptrace_bts_get_size(child);
 		break;
 
-	case PTRACE_BTS_READ_RECORD:
+	case PTRACE_BTS_GET:
 		ret = ptrace_bts_read_record
-			(child, data,
-			 (struct bts_struct __user *) addr);
+			(child, data, (struct bts_struct __user *) addr);
 		break;
 
-	case PTRACE_BTS_CONFIG:
-		ret = ptrace_bts_config(child, data);
+	case PTRACE_BTS_CLEAR:
+		ret = ptrace_bts_clear(child);
 		break;
 
-	case PTRACE_BTS_STATUS:
-		ret = ptrace_bts_status(child);
+	case PTRACE_BTS_DRAIN:
+		ret = ptrace_bts_drain
+			(child, (struct bts_struct __user *) addr);
 		break;
 
 	default:
@@ -1017,12 +1056,12 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 	case PTRACE_SETOPTIONS:
 	case PTRACE_SET_THREAD_AREA:
 	case PTRACE_GET_THREAD_AREA:
-	case PTRACE_BTS_MAX_BUFFER_SIZE:
-	case PTRACE_BTS_ALLOCATE_BUFFER:
-	case PTRACE_BTS_GET_BUFFER_SIZE:
-	case PTRACE_BTS_READ_RECORD:
 	case PTRACE_BTS_CONFIG:
 	case PTRACE_BTS_STATUS:
+	case PTRACE_BTS_SIZE:
+	case PTRACE_BTS_GET:
+	case PTRACE_BTS_CLEAR:
+	case PTRACE_BTS_DRAIN:
 		return sys_ptrace(request, pid, addr, data);
 
 	default:
diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h
index c9e15381dc7f..b84040abee68 100644
--- a/include/asm-x86/ds.h
+++ b/include/asm-x86/ds.h
@@ -52,11 +52,18 @@ struct bts_struct {
 	} variant;
 };
 
+/* Overflow handling mechanisms */
+#define DS_O_SIGNAL	1 /* send overflow signal */
+#define DS_O_WRAP	2 /* wrap around */
 
 extern int ds_allocate(void **, size_t);
 extern int ds_free(void **);
 extern int ds_get_bts_size(void *);
+extern int ds_get_bts_end(void *);
 extern int ds_get_bts_index(void *);
+extern int ds_set_overflow(void *, int);
+extern int ds_get_overflow(void *);
+extern int ds_clear(void *);
 extern int ds_read_bts(void *, size_t, struct bts_struct *);
 extern int ds_write_bts(void *, const struct bts_struct *);
 extern unsigned long ds_debugctl_mask(void);
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index b473ad45e9ca..cf2fe4633ee5 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -80,51 +80,53 @@
 
 #define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
 
-/* Return maximal BTS buffer size in number of records,
-   if successuf; -1, otherwise.
-   EOPNOTSUPP...processor does not support bts tracing */
-#define PTRACE_BTS_MAX_BUFFER_SIZE 40
-
-/* Allocate new bts buffer (free old one, if exists) of size DATA bts records;
-   parameter ADDR is ignored.
-   Return 0, if successful; -1, otherwise.
-   EOPNOTSUPP...processor does not support bts tracing
-   EINVAL.......invalid size in records
-   ENOMEM.......out of memory */
-#define PTRACE_BTS_ALLOCATE_BUFFER 41
-
-/* Return the size of the bts buffer in number of bts records,
-   if successful; -1, otherwise.
-   EOPNOTSUPP...processor does not support bts tracing
-   ENXIO........no buffer allocated */
-#define PTRACE_BTS_GET_BUFFER_SIZE 42
-
-/* Read the DATA'th bts record into a ptrace_bts_record buffer
-   provided in ADDR.
-   Records are ordered from newest to oldest.
-   Return 0, if successful; -1, otherwise
-   EOPNOTSUPP...processor does not support bts tracing
-   ENXIO........no buffer allocated
-   EINVAL.......invalid index */
-#define PTRACE_BTS_READ_RECORD 43
-
-/* Configure last branch trace; the configuration is given as a bit-mask of
-   PTRACE_BTS_O_* options in DATA; parameter ADDR is ignored.
-   Return 0, if successful; -1, otherwise
-   EOPNOTSUPP...processor does not support bts tracing
-   ENXIO........no buffer allocated */
-#define PTRACE_BTS_CONFIG 44
-
-/* Return the configuration as bit-mask of PTRACE_BTS_O_* options
-   if successful; -1, otherwise.
-   EOPNOTSUPP...processor does not support bts tracing
-   ENXIO........no buffer allocated */
-#define PTRACE_BTS_STATUS 45
-
-/* Trace configuration options */
-/* Collect last branch trace */
-#define PTRACE_BTS_O_TRACE_TASK 0x1
-/* Take timestamps when the task arrives and departs */
-#define PTRACE_BTS_O_TIMESTAMPS 0x2
+/* configuration/status structure used in PTRACE_BTS_CONFIG and
+   PTRACE_BTS_STATUS commands.
+*/
+struct ptrace_bts_config {
+	/* requested or actual size of BTS buffer in bytes */
+	unsigned long size;
+	/* bitmask of below flags */
+	unsigned long flags;
+};
+
+#define PTRACE_BTS_O_TRACE	0x1 /* branch trace */
+#define PTRACE_BTS_O_SCHED	0x2 /* scheduling events w/ jiffies */
+#define PTRACE_BTS_O_SIGNAL     0x4 /* send SIG? on buffer overflow
+				       instead of wrapping around */
+#define PTRACE_BTS_O_CUT_SIZE	0x8 /* cut requested size to max available
+				       instead of failing */
+
+#define PTRACE_BTS_CONFIG	40
+/* Configure branch trace recording.
+   DATA is ignored, ADDR points to a struct ptrace_bts_config.
+   A new buffer is allocated, iff the size changes.
+*/
+#define PTRACE_BTS_STATUS	41
+/* Return the current configuration.
+   DATA is ignored, ADDR points to a struct ptrace_bts_config
+   that will contain the result.
+*/
+#define PTRACE_BTS_SIZE		42
+/* Return the number of available BTS records.
+   DATA and ADDR are ignored.
+*/
+#define PTRACE_BTS_GET		43
+/* Get a single BTS record.
+   DATA defines the index into the BTS array, where 0 is the newest
+   entry, and higher indices refer to older entries.
+   ADDR is pointing to struct bts_struct (see asm/ds.h).
+*/
+#define PTRACE_BTS_CLEAR	44
+/* Clear the BTS buffer.
+   DATA and ADDR are ignored.
+*/
+#define PTRACE_BTS_DRAIN	45
+/* Read all available BTS records and clear the buffer.
+   DATA is ignored. ADDR points to an array of struct bts_struct of
+   suitable size.
+   BTS records are read from oldest to newest.
+   Returns number of BTS records drained.
+*/
 
 #endif
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index a9a1bab1451a..61946fe8c085 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -9,6 +9,7 @@
 
 #ifdef __KERNEL__
 
+/* the DS BTS struct is used for ptrace as well */
 #include <asm/ds.h>
 
 struct task_struct;
-- 
cgit v1.2.3


From 95c354fe9f7d6decc08a92aa26eb233ecc2155bf Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 30 Jan 2008 13:31:20 +0100
Subject: spinlock: lockbreak cleanup

The break_lock data structure and code for spinlocks is quite nasty.
Not only does it double the size of a spinlock but it changes locking to
a potentially less optimal trylock.

Put all of that under CONFIG_GENERIC_LOCKBREAK, and introduce a
__raw_spin_is_contended that uses the lock data itself to determine whether
there are waiters on the lock, to be used if CONFIG_GENERIC_LOCKBREAK is
not set.

Rename need_lockbreak to spin_needbreak, make it use spin_is_contended to
decouple it from the spinlock implementation, and make it typesafe (rwlocks
do not have any need_lockbreak sites -- why do they even get bloated up
with that break_lock then?).

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/arm/Kconfig               |  5 +++++
 arch/ia64/Kconfig              |  5 +++++
 arch/m32r/Kconfig              |  5 +++++
 arch/mips/Kconfig              |  5 +++++
 arch/parisc/Kconfig            |  5 +++++
 arch/powerpc/Kconfig           |  5 +++++
 arch/sparc64/Kconfig           |  5 +++++
 arch/x86/Kconfig               |  4 ++++
 fs/jbd/checkpoint.c            |  3 ++-
 fs/jbd/commit.c                |  2 +-
 fs/jbd2/checkpoint.c           |  3 ++-
 fs/jbd2/commit.c               |  2 +-
 include/linux/sched.h          | 21 +++++++--------------
 include/linux/spinlock.h       |  6 ++++++
 include/linux/spinlock_types.h |  4 ++--
 include/linux/spinlock_up.h    |  2 ++
 kernel/sched.c                 | 16 ++++++----------
 kernel/spinlock.c              |  3 +--
 mm/memory.c                    |  8 +++-----
 19 files changed, 72 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index de211ac3853e..77201d3f7479 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -91,6 +91,11 @@ config GENERIC_IRQ_PROBE
 	bool
 	default y
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_GENERIC_SPINLOCK
 	bool
 	default y
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index bef47725d4ad..4a81b7fb191a 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -42,6 +42,11 @@ config MMU
 config SWIOTLB
        bool
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_XCHGADD_ALGORITHM
 	bool
 	default y
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index ab9a264cb194..f7237c5f531e 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -235,6 +235,11 @@ config IRAM_SIZE
 # Define implied options from the CPU selection here
 #
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_GENERIC_SPINLOCK
 	bool
 	depends on M32R
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 6b0f85f02c79..4fad0a34b997 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -694,6 +694,11 @@ source "arch/mips/vr41xx/Kconfig"
 
 endmenu
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_GENERIC_SPINLOCK
 	bool
 	default y
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index b8ef1787a191..2b649c46631c 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -19,6 +19,11 @@ config MMU
 config STACK_GROWSUP
 	def_bool y
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y
 
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 232c298c933f..c17a194beb0e 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -53,6 +53,11 @@ config RWSEM_XCHGADD_ALGORITHM
 	bool
 	default y
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config ARCH_HAS_ILOG2_U32
 	bool
 	default y
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index 10b212a1f9f5..1e25bce0366d 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -200,6 +200,11 @@ config US2E_FREQ
 	  If in doubt, say N.
 
 # Global things across all Sun machines.
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 config RWSEM_GENERIC_SPINLOCK
 	bool
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 23936301db56..db434f8171d3 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -19,6 +19,10 @@ config X86_64
 config X86
 	def_bool y
 
+config GENERIC_LOCKBREAK
+	def_bool y
+	depends on SMP && PREEMPT
+
 config GENERIC_TIME
 	def_bool y
 
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 0f69c416eebc..a5432bbbfb88 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -347,7 +347,8 @@ restart:
 				break;
 			}
 			retry = __process_buffer(journal, jh, bhs,&batch_count);
-			if (!retry && lock_need_resched(&journal->j_list_lock)){
+			if (!retry && (need_resched() ||
+				spin_needbreak(&journal->j_list_lock))) {
 				spin_unlock(&journal->j_list_lock);
 				retry = 1;
 				break;
diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 610264b99a8e..31853eb65b4c 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -265,7 +265,7 @@ write_out_data:
 			put_bh(bh);
 		}
 
-		if (lock_need_resched(&journal->j_list_lock)) {
+		if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
 			spin_unlock(&journal->j_list_lock);
 			goto write_out_data;
 		}
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 1b7f282c1ae9..6914598022ce 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -353,7 +353,8 @@ restart:
 			}
 			retry = __process_buffer(journal, jh, bhs, &batch_count,
 						 transaction);
-			if (!retry && lock_need_resched(&journal->j_list_lock)){
+			if (!retry && (need_resched() ||
+				spin_needbreak(&journal->j_list_lock))) {
 				spin_unlock(&journal->j_list_lock);
 				retry = 1;
 				break;
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index da8d0eb3b7b9..4f302d279279 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -341,7 +341,7 @@ write_out_data:
 			put_bh(bh);
 		}
 
-		if (lock_need_resched(&journal->j_list_lock)) {
+		if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
 			spin_unlock(&journal->j_list_lock);
 			goto write_out_data;
 		}
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2d0546e884ea..9d4797609aa5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1922,23 +1922,16 @@ extern int cond_resched_softirq(void);
 
 /*
  * Does a critical section need to be broken due to another
- * task waiting?:
+ * task waiting?: (technically does not depend on CONFIG_PREEMPT,
+ * but a general need for low latency)
  */
-#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP)
-# define need_lockbreak(lock) ((lock)->break_lock)
-#else
-# define need_lockbreak(lock) 0
-#endif
-
-/*
- * Does a critical section need to be broken due to another
- * task waiting or preemption being signalled:
- */
-static inline int lock_need_resched(spinlock_t *lock)
+static inline int spin_needbreak(spinlock_t *lock)
 {
-	if (need_lockbreak(lock) || need_resched())
-		return 1;
+#ifdef CONFIG_PREEMPT
+	return spin_is_contended(lock);
+#else
 	return 0;
+#endif
 }
 
 /*
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index c376f3b36c89..124449733c55 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -120,6 +120,12 @@ do {								\
 
 #define spin_is_locked(lock)	__raw_spin_is_locked(&(lock)->raw_lock)
 
+#ifdef CONFIG_GENERIC_LOCKBREAK
+#define spin_is_contended(lock) ((lock)->break_lock)
+#else
+#define spin_is_contended(lock)	__raw_spin_is_contended(&(lock)->raw_lock)
+#endif
+
 /**
  * spin_unlock_wait - wait until the spinlock gets unlocked
  * @lock: the spinlock in question.
diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h
index f6a3a951b79e..68d88f71f1a2 100644
--- a/include/linux/spinlock_types.h
+++ b/include/linux/spinlock_types.h
@@ -19,7 +19,7 @@
 
 typedef struct {
 	raw_spinlock_t raw_lock;
-#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP)
+#ifdef CONFIG_GENERIC_LOCKBREAK
 	unsigned int break_lock;
 #endif
 #ifdef CONFIG_DEBUG_SPINLOCK
@@ -35,7 +35,7 @@ typedef struct {
 
 typedef struct {
 	raw_rwlock_t raw_lock;
-#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP)
+#ifdef CONFIG_GENERIC_LOCKBREAK
 	unsigned int break_lock;
 #endif
 #ifdef CONFIG_DEBUG_SPINLOCK
diff --git a/include/linux/spinlock_up.h b/include/linux/spinlock_up.h
index ea54c4c9a4ec..938234c4a996 100644
--- a/include/linux/spinlock_up.h
+++ b/include/linux/spinlock_up.h
@@ -64,6 +64,8 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 # define __raw_spin_trylock(lock)	({ (void)(lock); 1; })
 #endif /* DEBUG_SPINLOCK */
 
+#define __raw_spin_is_contended(lock)	(((void)(lock), 0))
+
 #define __raw_read_can_lock(lock)	(((void)(lock), 1))
 #define __raw_write_can_lock(lock)	(((void)(lock), 1))
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 524285e46fa7..ba4c88088f62 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4945,19 +4945,15 @@ EXPORT_SYMBOL(_cond_resched);
  */
 int cond_resched_lock(spinlock_t *lock)
 {
+	int resched = need_resched() && system_state == SYSTEM_RUNNING;
 	int ret = 0;
 
-	if (need_lockbreak(lock)) {
+	if (spin_needbreak(lock) || resched) {
 		spin_unlock(lock);
-		cpu_relax();
-		ret = 1;
-		spin_lock(lock);
-	}
-	if (need_resched() && system_state == SYSTEM_RUNNING) {
-		spin_release(&lock->dep_map, 1, _THIS_IP_);
-		_raw_spin_unlock(lock);
-		preempt_enable_no_resched();
-		__cond_resched();
+		if (resched && need_resched())
+			__cond_resched();
+		else
+			cpu_relax();
 		ret = 1;
 		spin_lock(lock);
 	}
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index cd72424c2662..ae28c8245123 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -65,8 +65,7 @@ EXPORT_SYMBOL(_write_trylock);
  * even on CONFIG_PREEMPT, because lockdep assumes that interrupts are
  * not re-enabled during lock-acquire (which the preempt-spin-ops do):
  */
-#if !defined(CONFIG_PREEMPT) || !defined(CONFIG_SMP) || \
-	defined(CONFIG_DEBUG_LOCK_ALLOC)
+#if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC)
 
 void __lockfunc _read_lock(rwlock_t *lock)
 {
diff --git a/mm/memory.c b/mm/memory.c
index 4b0144b24c12..673ebbf499c7 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -513,8 +513,7 @@ again:
 		if (progress >= 32) {
 			progress = 0;
 			if (need_resched() ||
-			    need_lockbreak(src_ptl) ||
-			    need_lockbreak(dst_ptl))
+			    spin_needbreak(src_ptl) || spin_needbreak(dst_ptl))
 				break;
 		}
 		if (pte_none(*src_pte)) {
@@ -853,7 +852,7 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp,
 			tlb_finish_mmu(*tlbp, tlb_start, start);
 
 			if (need_resched() ||
-				(i_mmap_lock && need_lockbreak(i_mmap_lock))) {
+				(i_mmap_lock && spin_needbreak(i_mmap_lock))) {
 				if (i_mmap_lock) {
 					*tlbp = NULL;
 					goto out;
@@ -1768,8 +1767,7 @@ again:
 
 	restart_addr = zap_page_range(vma, start_addr,
 					end_addr - start_addr, details);
-	need_break = need_resched() ||
-			need_lockbreak(details->i_mmap_lock);
+	need_break = need_resched() || spin_needbreak(details->i_mmap_lock);
 
 	if (restart_addr >= end_addr) {
 		/* We have now completed this vma: mark it so */
-- 
cgit v1.2.3


From 314cdbefd1fd0a7acf3780e9628465b77ea6a836 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: x86: FIFO ticket spinlocks

Introduce ticket lock spinlocks for x86 which are FIFO. The implementation
is described in the comments. The straight-line lock/unlock instruction
sequence is slightly slower than the dec based locks on modern x86 CPUs,
however the difference is quite small on Core2 and Opteron when working out of
cache, and becomes almost insignificant even on P4 when the lock misses cache.
trylock is more significantly slower, but they are relatively rare.

On an 8 core (2 socket) Opteron, spinlock unfairness is extremely noticable,
with a userspace test having a difference of up to 2x runtime per thread, and
some threads are starved or "unfairly" granted the lock up to 1 000 000 (!)
times. After this patch, all threads appear to finish at exactly the same
time.

The memory ordering of the lock does conform to x86 standards, and the
implementation has been reviewed by Intel and AMD engineers.

The algorithm also tells us how many CPUs are contending the lock, so
lockbreak becomes trivial and we no longer have to waste 4 bytes per
spinlock for it.

After this, we can no longer spin on any locks with preempt enabled
and cannot reenable interrupts when spinning on an irq safe lock, because
at that point we have already taken a ticket and the would deadlock if
the same CPU tries to take the lock again.  These are questionable anyway:
if the lock happens to be called under a preempt or interrupt disabled section,
then it will just have the same latency problems. The real fix is to keep
critical sections short, and ensure locks are reasonably fair (which this
patch does).

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Kconfig                 |   3 +-
 include/asm-x86/paravirt.h       |  21 ------
 include/asm-x86/spinlock.h       | 153 +++++++++++++++++++++------------------
 include/asm-x86/spinlock_types.h |   2 +-
 4 files changed, 83 insertions(+), 96 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index db434f8171d3..1992b8fe6a2f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -20,8 +20,7 @@ config X86
 	def_bool y
 
 config GENERIC_LOCKBREAK
-	def_bool y
-	depends on SMP && PREEMPT
+	def_bool n
 
 config GENERIC_TIME
 	def_bool y
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 4f23f434a1f3..24406703007f 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1077,27 +1077,6 @@ static inline unsigned long __raw_local_irq_save(void)
 	return f;
 }
 
-#define CLI_STRING							\
-	_paravirt_alt("pushl %%ecx; pushl %%edx;"			\
-		      "call *%[paravirt_cli_opptr];"			\
-		      "popl %%edx; popl %%ecx",				\
-		      "%c[paravirt_cli_type]", "%c[paravirt_clobber]")
-
-#define STI_STRING							\
-	_paravirt_alt("pushl %%ecx; pushl %%edx;"			\
-		      "call *%[paravirt_sti_opptr];"			\
-		      "popl %%edx; popl %%ecx",				\
-		      "%c[paravirt_sti_type]", "%c[paravirt_clobber]")
-
-#define CLI_STI_CLOBBERS , "%eax"
-#define CLI_STI_INPUT_ARGS						\
-	,								\
-	[paravirt_cli_type] "i" (PARAVIRT_PATCH(pv_irq_ops.irq_disable)),		\
-	[paravirt_cli_opptr] "m" (pv_irq_ops.irq_disable),		\
-	[paravirt_sti_type] "i" (PARAVIRT_PATCH(pv_irq_ops.irq_enable)),		\
-	[paravirt_sti_opptr] "m" (pv_irq_ops.irq_enable),		\
-	paravirt_clobber(CLBR_EAX)
-
 /* Make sure as little as possible of this mess escapes. */
 #undef PARAVIRT_CALL
 #undef __PVOP_CALL
diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index afd4b80ff0ad..97d52b506af8 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -5,6 +5,7 @@
 #include <asm/rwlock.h>
 #include <asm/page.h>
 #include <asm/processor.h>
+#include <linux/compiler.h>
 
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
@@ -12,7 +13,8 @@
  * Simple spin lock operations.  There are two variants, one clears IRQ's
  * on the local processor, one does not.
  *
- * We make no fairness assumptions. They have a cost.
+ * These are fair FIFO ticket locks, which are currently limited to 256
+ * CPUs.
  *
  * (the type definitions are in asm/spinlock_types.h)
  */
@@ -42,103 +44,102 @@ typedef int _slock_t;
 # define LOCK_PTR_REG "D"
 #endif
 
+#if (NR_CPUS > 256)
+#error spinlock supports a maximum of 256 CPUs
+#endif
+
 static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
 {
-	return *(volatile _slock_t *)(&(lock)->slock) <= 0;
+	int tmp = *(volatile signed int *)(&(lock)->slock);
+
+	return (((tmp >> 8) & 0xff) != (tmp & 0xff));
 }
 
-static inline void __raw_spin_lock(raw_spinlock_t *lock)
+static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
 {
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t"
-		"jns 3f\n"
-		"2:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0,%0\n\t"
-		"jle 2b\n\t"
-		"jmp 1b\n"
-		"3:\n\t"
-		: "+m" (lock->slock) : : "memory");
+	int tmp = *(volatile signed int *)(&(lock)->slock);
+
+	return (((tmp >> 8) & 0xff) - (tmp & 0xff)) > 1;
 }
 
-/*
- * It is easier for the lock validator if interrupts are not re-enabled
- * in the middle of a lock-acquire. This is a performance feature anyway
- * so we turn it off:
- *
- * NOTE: there's an irqs-on section here, which normally would have to be
- * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
- */
-#ifndef CONFIG_PROVE_LOCKING
-static inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
-					 unsigned long flags)
+static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm volatile(
-		"\n1:\t"
-		LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t"
-		"jns 5f\n"
-		"testl $0x200, %[flags]\n\t"
-		"jz 4f\n\t"
-		STI_STRING "\n"
-		"3:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jle 3b\n\t"
-		CLI_STRING "\n\t"
+	short inc = 0x0100;
+
+	/*
+	 * Ticket locks are conceptually two bytes, one indicating the current
+	 * head of the queue, and the other indicating the current tail. The
+	 * lock is acquired by atomically noting the tail and incrementing it
+	 * by one (thus adding ourself to the queue and noting our position),
+	 * then waiting until the head becomes equal to the the initial value
+	 * of the tail.
+	 *
+	 * This uses a 16-bit xadd to increment the tail and also load the
+	 * position of the head, which takes care of memory ordering issues
+	 * and should be optimal for the uncontended case. Note the tail must
+	 * be in the high byte, otherwise the 16-bit wide increment of the low
+	 * byte would carry up and contaminate the high byte.
+	 */
+
+	__asm__ __volatile__ (
+		LOCK_PREFIX "xaddw %w0, %1\n"
+		"1:\t"
+		"cmpb %h0, %b0\n\t"
+		"je 2f\n\t"
+		"rep ; nop\n\t"
+		"movb %1, %b0\n\t"
+		/* don't need lfence here, because loads are in-order */
 		"jmp 1b\n"
-		"4:\t"
-		"rep;nop\n\t"
-		LOCK_INS_CMP " $0, %[slock]\n\t"
-		"jg 1b\n\t"
-		"jmp 4b\n"
-		"5:\n\t"
-		: [slock] "+m" (lock->slock)
-		: [flags] "r" ((u32)flags)
-		  CLI_STI_INPUT_ARGS
-		: "memory" CLI_STI_CLOBBERS);
+		"2:"
+		:"+Q" (inc), "+m" (lock->slock)
+		:
+		:"memory", "cc");
 }
-#endif
+
+#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
-	_slock_t oldval;
+	int tmp;
+	short new;
 
 	asm volatile(
-		LOCK_INS_XCH " %0,%1"
-		:"=q" (oldval), "+m" (lock->slock)
-		:"0" (0) : "memory");
-
-	return oldval > 0;
+		"movw %2,%w0\n\t"
+		"cmpb %h0,%b0\n\t"
+		"jne 1f\n\t"
+		"movw %w0,%w1\n\t"
+		"incb %h1\n\t"
+		"lock ; cmpxchgw %w1,%2\n\t"
+		"1:"
+		"sete %b1\n\t"
+		"movzbl %b1,%0\n\t"
+		:"=&a" (tmp), "=Q" (new), "+m" (lock->slock)
+		:
+		: "memory", "cc");
+
+	return tmp;
 }
 
+#if defined(CONFIG_X86_32) && \
+	(defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE))
 /*
- * __raw_spin_unlock based on writing $1 to the low byte.
- * This method works. Despite all the confusion.
- * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there)
+ * On PPro SMP or if we are using OOSTORE, we use a locked operation to unlock
  * (PPro errata 66, 92)
  */
-#if defined(X86_64) || \
-	(!defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE))
-
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
-{
-	asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory");
-}
-
+# define UNLOCK_LOCK_PREFIX LOCK_PREFIX
 #else
+# define UNLOCK_LOCK_PREFIX
+#endif
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	unsigned char oldval = 1;
-
-	asm volatile("xchgb %b0, %1"
-		     : "=q" (oldval), "+m" (lock->slock)
-		     : "0" (oldval) : "memory");
+	__asm__ __volatile__(
+		UNLOCK_LOCK_PREFIX "incb %0"
+		:"+m" (lock->slock)
+		:
+		:"memory", "cc");
 }
 
-#endif
-
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
 {
 	while (__raw_spin_is_locked(lock))
@@ -159,11 +160,19 @@ static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
  * with the high bit (sign) being the "contended" bit.
  */
 
+/**
+ * read_can_lock - would read_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
 static inline int __raw_read_can_lock(raw_rwlock_t *lock)
 {
 	return (int)(lock)->lock > 0;
 }
 
+/**
+ * write_can_lock - would write_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
 static inline int __raw_write_can_lock(raw_rwlock_t *lock)
 {
 	return (lock)->lock == RW_LOCK_BIAS;
diff --git a/include/asm-x86/spinlock_types.h b/include/asm-x86/spinlock_types.h
index 4da9345c1500..9029cf78cf5d 100644
--- a/include/asm-x86/spinlock_types.h
+++ b/include/asm-x86/spinlock_types.h
@@ -9,7 +9,7 @@ typedef struct {
 	unsigned int slock;
 } raw_spinlock_t;
 
-#define __RAW_SPIN_LOCK_UNLOCKED	{ 1 }
+#define __RAW_SPIN_LOCK_UNLOCKED	{ 0 }
 
 typedef struct {
 	unsigned int lock;
-- 
cgit v1.2.3


From aa470140e86e45723cf8387292edbce9106ddc1f Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: x86: kprobe-booster for x86-64

This patch adds kprobe-booster to kprobes_64.c.

- Changes are based on x86-32.
- Add REX prefix checking code.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/kprobes_64.c | 142 ++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/kprobes_64.h |  10 +++
 2 files changed, 149 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/kprobes_64.c b/arch/x86/kernel/kprobes_64.c
index f6837cd3bed5..bf0e18473677 100644
--- a/arch/x86/kernel/kprobes_64.c
+++ b/arch/x86/kernel/kprobes_64.c
@@ -55,6 +55,105 @@ struct kretprobe_blackpoint kretprobe_blacklist[] = {
 };
 const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist);
 
+/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
+static __always_inline void set_jmp_op(void *from, void *to)
+{
+	struct __arch_jmp_op {
+		char op;
+		s32 raddr;
+	} __attribute__((packed)) * jop;
+	jop = (struct __arch_jmp_op *)from;
+	jop->raddr = (s32)((long)(to) - ((long)(from) + 5));
+	jop->op = RELATIVEJUMP_INSTRUCTION;
+}
+
+/*
+ * returns non-zero if opcode is boostable
+ * RIP relative instructions are adjusted at copying time
+ */
+static __always_inline int can_boost(kprobe_opcode_t *opcodes)
+{
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
+	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
+	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
+	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
+	 << (row % 64))
+	/*
+	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
+	 * Groups, and some special opcodes can not boost.
+	 */
+	static const unsigned long twobyte_is_boostable[256 / 64] = {
+		/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f        */
+		/*      ----------------------------------------------        */
+		W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0)|/* 00 */
+		W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 10 */
+		W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 20 */
+		W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0),/* 30 */
+		W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)|/* 40 */
+		W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 50 */
+		W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1)|/* 60 */
+		W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1),/* 70 */
+		W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 80 */
+		W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)|/* 90 */
+		W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* a0 */
+		W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1),/* b0 */
+		W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1)|/* c0 */
+		W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* d0 */
+		W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* e0 */
+		W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0) /* f0 */
+		/*      -----------------------------------------------       */
+		/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f        */
+	};
+#undef W
+	kprobe_opcode_t opcode;
+	kprobe_opcode_t *orig_opcodes = opcodes;
+
+retry:
+	if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+		return 0;
+	opcode = *(opcodes++);
+
+	/* 2nd-byte opcode */
+	if (opcode == 0x0f) {
+		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+			return 0;
+		return test_bit(*opcodes, twobyte_is_boostable);
+	}
+
+	switch (opcode & 0xf0) {
+	case 0x40:
+		goto retry; /* REX prefix is boostable */
+	case 0x60:
+		if (0x63 < opcode && opcode < 0x67)
+			goto retry; /* prefixes */
+		/* can't boost Address-size override and bound */
+		return (opcode != 0x62 && opcode != 0x67);
+	case 0x70:
+		return 0; /* can't boost conditional jump */
+	case 0xc0:
+		/* can't boost software-interruptions */
+		return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
+	case 0xd0:
+		/* can boost AA* and XLAT */
+		return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7);
+	case 0xe0:
+		/* can boost in/out and absolute jmps */
+		return ((opcode & 0x04) || opcode == 0xea);
+	case 0xf0:
+		if ((opcode & 0x0c) == 0 && opcode != 0xf1)
+			goto retry; /* lock/rep(ne) prefix */
+		/* clear and set flags are boostable */
+		return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
+	default:
+		/* segment override prefixes are boostable */
+		if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
+			goto retry; /* prefixes */
+		/* CS override prefix and call are not boostable */
+		return (opcode != 0x2e && opcode != 0x9a);
+	}
+}
+
 /*
  * returns non-zero if opcode modifies the interrupt flag.
  */
@@ -86,7 +185,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 
 /*
  * Determine if the instruction uses the %rip-relative addressing mode.
- * If it does, return the address of the 32-bit displacement word.
+ * If it does, Return the address of the 32-bit displacement word.
  * If not, return null.
  */
 static s32 __kprobes *is_riprel(u8 *insn)
@@ -210,6 +309,11 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)
 		BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
 		*ripdisp = disp;
 	}
+	if (can_boost(p->addr)) {
+		p->ainsn.boostable = 0;
+	} else {
+		p->ainsn.boostable = -1;
+	}
 	p->opcode = *p->addr;
 }
 
@@ -226,7 +330,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
 	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, 0);
+	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
 	mutex_unlock(&kprobe_mutex);
 }
 
@@ -384,6 +488,15 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 		return 1;
 
 ss_probe:
+#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
+	if (p->ainsn.boostable == 1 && !p->post_handler) {
+		/* Boost up -- we can execute copied instructions directly */
+		reset_current_kprobe();
+		regs->ip = (unsigned long)p->ainsn.insn;
+		preempt_enable_no_resched();
+		return 1;
+	}
+#endif
 	prepare_singlestep(p, regs);
 	kcb->kprobe_status = KPROBE_HIT_SS;
 	return 1;
@@ -493,6 +606,11 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
  * 2) If the single-stepped instruction was a call, the return address
  * that is atop the stack is the address following the copied instruction.
  * We need to make it the address following the original instruction.
+ *
+ * If this is the first time we've single-stepped the instruction at
+ * this probepoint, and the instruction is boostable, boost it: add a
+ * jump instruction after the copied instruction, that jumps to the next
+ * instruction after the probepoint.
  */
 static void __kprobes resume_execution(struct kprobe *p,
 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
@@ -519,6 +637,7 @@ static void __kprobes resume_execution(struct kprobe *p,
 	case 0xcf:
 	case 0xea:	/* jmp absolute -- ip is correct */
 		/* ip is already adjusted, no more changes required */
+		p->ainsn.boostable = 1;
 		goto no_change;
 	case 0xe8:	/* call relative - Fix return addr */
 		*tos = orig_rip + (*tos - copy_rip);
@@ -527,17 +646,34 @@ static void __kprobes resume_execution(struct kprobe *p,
 		if ((insn[1] & 0x30) == 0x10) {
 			/* call absolute, indirect */
 			/* Fix return addr; ip is correct. */
+			/* not boostable */
 			*tos = orig_rip + (*tos - copy_rip);
 			goto no_change;
 		} else if (((insn[1] & 0x31) == 0x20) ||	/* jmp near, absolute indirect */
 			   ((insn[1] & 0x31) == 0x21)) {	/* jmp far, absolute indirect */
-			/* ip is correct. */
+			/* ip is correct. And this is boostable */
+			p->ainsn.boostable = 1;
 			goto no_change;
 		}
 	default:
 		break;
 	}
 
+	if (p->ainsn.boostable == 0) {
+		if ((regs->ip > copy_rip) &&
+		    (regs->ip - copy_rip) + 5 < MAX_INSN_SIZE) {
+			/*
+			 * These instructions can be executed directly if it
+			 * jumps back to correct address.
+			 */
+			set_jmp_op((void *)regs->ip,
+				   (void *)orig_rip + (regs->ip - copy_rip));
+			p->ainsn.boostable = 1;
+		} else {
+			p->ainsn.boostable = -1;
+		}
+	}
+
 	regs->ip = orig_rip + (regs->ip - copy_rip);
 
 no_change:
diff --git a/include/asm-x86/kprobes_64.h b/include/asm-x86/kprobes_64.h
index 8c919d35cdd3..e7e921dac7c2 100644
--- a/include/asm-x86/kprobes_64.h
+++ b/include/asm-x86/kprobes_64.h
@@ -34,6 +34,7 @@ struct kprobe;
 
 typedef u8 kprobe_opcode_t;
 #define BREAKPOINT_INSTRUCTION	0xcc
+#define RELATIVEJUMP_INSTRUCTION 0xe9
 #define MAX_INSN_SIZE 15
 #define MAX_STACK_SIZE 64
 #define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
@@ -52,6 +53,15 @@ extern void arch_remove_kprobe(struct kprobe *p);
 struct arch_specific_insn {
 	/* copy of the original instruction */
 	kprobe_opcode_t *insn;
+	/*
+	 * boostable = -1: This instruction type is not boostable.
+	 * boostable = 0: This instruction type is boostable.
+	 * boostable = 1: This instruction has been boosted: we have
+	 * added a relative jump after the instruction copy in insn,
+	 * so no single-step and fixup are needed (unless there's
+	 * a post_handler or break_handler).
+	 */
+	int boostable;
 };
 
 struct prev_kprobe {
-- 
cgit v1.2.3


From 8533bbe9f87b01f49ff951f665ea1988252fa3c2 Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: x86: prepare kprobes code for x86 unification

This patch cleanup kprobes code on x86 for unification.
This patch is based on Arjan's previous work.

- Remove spurious whitespace changes
- Add harmless includes
- Make the 32/64 files more identical
 - Generalize structure fields' and local variable name.
 - Wrap accessing to stack address by macros.
 - Modify bitmap making macro.
 - Merge fixup code into is_riprel() and change its name to fix_riprel().
 - Set MAX_INSN_SIZE to 16 on both arch.
 - Use u32 for bitmaps on both architectures.
 - Clarify some comments.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/kprobes_32.c | 321 ++++++++++++++++++++-------------
 arch/x86/kernel/kprobes_64.c | 409 ++++++++++++++++++++++---------------------
 arch/x86/mm/fault_32.c       |   1 -
 arch/x86/mm/fault_64.c       |   1 -
 include/asm-x86/kprobes_32.h |  32 ++--
 include/asm-x86/kprobes_64.h |  30 ++--
 6 files changed, 447 insertions(+), 347 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/kprobes_32.c b/arch/x86/kernel/kprobes_32.c
index 8eccd2d04709..8e06431d8b03 100644
--- a/arch/x86/kernel/kprobes_32.c
+++ b/arch/x86/kernel/kprobes_32.c
@@ -29,10 +29,15 @@
 
 #include <linux/kprobes.h>
 #include <linux/ptrace.h>
+#include <linux/string.h>
+#include <linux/slab.h>
 #include <linux/preempt.h>
+#include <linux/module.h>
 #include <linux/kdebug.h>
+
 #include <asm/cacheflush.h>
 #include <asm/desc.h>
+#include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/alternative.h>
 
@@ -41,65 +46,121 @@ void jprobe_return_end(void);
 DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
 DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
 
+/*
+ * "&regs->sp" looks wrong, but it's correct for x86_32.  x86_32 CPUs
+ * don't save the ss and esp registers if the CPU is already in kernel
+ * mode when it traps.  So for kprobes, regs->sp and regs->ss are not
+ * the [nonexistent] saved stack pointer and ss register, but rather
+ * the top 8 bytes of the pre-int3 stack.  So &regs->sp happens to
+ * point to the top of the pre-int3 stack.
+ */
+#define stack_addr(regs) ((unsigned long *)&regs->sp)
+
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
+	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
+	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
+	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
+	 << (row % 32))
+	/*
+	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
+	 * Groups, and some special opcodes can not boost.
+	 */
+static const u32 twobyte_is_boostable[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      ----------------------------------------------          */
+	W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) , /* 70 */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 80 */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 90 */
+	W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* a0 */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* c0 */
+	W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) , /* d0 */
+	W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* e0 */
+	W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 onebyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
+	W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
+	W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
+	W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
+	W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
+	W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
+	W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
+	W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 twobyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
+	W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
+	W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
+	W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
+	W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
+	W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
+	W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
+	W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+	W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+	W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
+	W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0)   /* ff */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+#undef W
+
 struct kretprobe_blackpoint kretprobe_blacklist[] = {
 	{"__switch_to", }, /* This function switches only current task, but
-			     doesn't switch kernel stack.*/
+			      doesn't switch kernel stack.*/
 	{NULL, NULL}	/* Terminator */
 };
 const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist);
 
-/* insert a jmp code */
+/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
 static __always_inline void set_jmp_op(void *from, void *to)
 {
 	struct __arch_jmp_op {
 		char op;
-		long raddr;
-	} __attribute__((packed)) *jop;
+		s32 raddr;
+	} __attribute__((packed)) * jop;
 	jop = (struct __arch_jmp_op *)from;
-	jop->raddr = (long)(to) - ((long)(from) + 5);
+	jop->raddr = (s32)((long)(to) - ((long)(from) + 5));
 	jop->op = RELATIVEJUMP_INSTRUCTION;
 }
 
 /*
- * returns non-zero if opcodes can be boosted.
+ * returns non-zero if opcode is boostable.
  */
 static __always_inline int can_boost(kprobe_opcode_t *opcodes)
 {
-#define W(row,b0,b1,b2,b3,b4,b5,b6,b7,b8,b9,ba,bb,bc,bd,be,bf)		      \
-	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
-	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
-	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
-	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
-	 << (row % 32))
-	/*
-	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
-	 * Groups, and some special opcodes can not be boost.
-	 */
-	static const unsigned long twobyte_is_boostable[256 / 32] = {
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-		/*      -------------------------------         */
-		W(0x00, 0,0,1,1,0,0,1,0,1,1,0,0,0,0,0,0)| /* 00 */
-		W(0x10, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 10 */
-		W(0x20, 1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0)| /* 20 */
-		W(0x30, 0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 30 */
-		W(0x40, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 40 */
-		W(0x50, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 50 */
-		W(0x60, 1,1,1,1,1,1,1,1,1,1,1,1,0,0,1,1)| /* 60 */
-		W(0x70, 0,0,0,0,1,1,1,1,0,0,0,0,0,0,1,1), /* 70 */
-		W(0x80, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 80 */
-		W(0x90, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1), /* 90 */
-		W(0xa0, 1,1,0,1,1,1,0,0,1,1,0,1,1,1,0,1)| /* a0 */
-		W(0xb0, 1,1,1,1,1,1,1,1,0,0,0,1,1,1,1,1), /* b0 */
-		W(0xc0, 1,1,0,0,0,0,0,0,1,1,1,1,1,1,1,1)| /* c0 */
-		W(0xd0, 0,1,1,1,0,1,0,0,1,1,0,1,1,1,0,1), /* d0 */
-		W(0xe0, 0,1,1,0,0,1,0,0,1,1,0,1,1,1,0,1)| /* e0 */
-		W(0xf0, 0,1,1,1,0,1,0,0,1,1,1,0,1,1,1,0)  /* f0 */
-		/*      -------------------------------         */
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-	};
-#undef W
 	kprobe_opcode_t opcode;
 	kprobe_opcode_t *orig_opcodes = opcodes;
+
 retry:
 	if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
 		return 0;
@@ -109,7 +170,8 @@ retry:
 	if (opcode == 0x0f) {
 		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
 			return 0;
-		return test_bit(*opcodes, twobyte_is_boostable);
+		return test_bit(*opcodes,
+				(unsigned long *)twobyte_is_boostable);
 	}
 
 	switch (opcode & 0xf0) {
@@ -132,12 +194,13 @@ retry:
 	case 0xf0:
 		if ((opcode & 0x0c) == 0 && opcode != 0xf1)
 			goto retry; /* lock/rep(ne) prefix */
-		/* clear and set flags can be boost */
+		/* clear and set flags are boostable */
 		return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
 	default:
+		/* segment override prefixes are boostable */
 		if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
 			goto retry; /* prefixes */
-		/* can't boost CS override and call */
+		/* CS override prefix and call are not boostable */
 		return (opcode != 0x2e && opcode != 0x9a);
 	}
 }
@@ -145,9 +208,9 @@ retry:
 /*
  * returns non-zero if opcode modifies the interrupt flag.
  */
-static int __kprobes is_IF_modifier(kprobe_opcode_t opcode)
+static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
 {
-	switch (opcode) {
+	switch (*insn) {
 	case 0xfa:		/* cli */
 	case 0xfb:		/* sti */
 	case 0xcf:		/* iret/iretd */
@@ -157,20 +220,24 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t opcode)
 	return 0;
 }
 
+static void __kprobes arch_copy_kprobe(struct kprobe *p)
+{
+	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	if (can_boost(p->addr))
+		p->ainsn.boostable = 0;
+	else
+		p->ainsn.boostable = -1;
+
+	p->opcode = *p->addr;
+}
+
 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
-	/* insn: must be on special executable page on i386. */
+	/* insn: must be on special executable page on x86. */
 	p->ainsn.insn = get_insn_slot();
 	if (!p->ainsn.insn)
 		return -ENOMEM;
-
-	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
-	p->opcode = *p->addr;
-	if (can_boost(p->addr)) {
-		p->ainsn.boostable = 0;
-	} else {
-		p->ainsn.boostable = -1;
-	}
+	arch_copy_kprobe(p);
 	return 0;
 }
 
@@ -195,26 +262,26 @@ static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
 	kcb->prev_kprobe.kp = kprobe_running();
 	kcb->prev_kprobe.status = kcb->kprobe_status;
-	kcb->prev_kprobe.old_eflags = kcb->kprobe_old_eflags;
-	kcb->prev_kprobe.saved_eflags = kcb->kprobe_saved_eflags;
+	kcb->prev_kprobe.old_flags = kcb->kprobe_old_flags;
+	kcb->prev_kprobe.saved_flags = kcb->kprobe_saved_flags;
 }
 
 static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
 	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
 	kcb->kprobe_status = kcb->prev_kprobe.status;
-	kcb->kprobe_old_eflags = kcb->prev_kprobe.old_eflags;
-	kcb->kprobe_saved_eflags = kcb->prev_kprobe.saved_eflags;
+	kcb->kprobe_old_flags = kcb->prev_kprobe.old_flags;
+	kcb->kprobe_saved_flags = kcb->prev_kprobe.saved_flags;
 }
 
 static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 				struct kprobe_ctlblk *kcb)
 {
 	__get_cpu_var(current_kprobe) = p;
-	kcb->kprobe_saved_eflags = kcb->kprobe_old_eflags
+	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
 		= (regs->flags & (TF_MASK | IF_MASK));
-	if (is_IF_modifier(p->opcode))
-		kcb->kprobe_saved_eflags &= ~IF_MASK;
+	if (is_IF_modifier(p->ainsn.insn))
+		kcb->kprobe_saved_flags &= ~IF_MASK;
 }
 
 static __always_inline void clear_btf(void)
@@ -245,7 +312,7 @@ static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
-	unsigned long *sara = (unsigned long *)&regs->sp;
+	unsigned long *sara = stack_addr(regs);
 
 	ri->ret_addr = (kprobe_opcode_t *) *sara;
 
@@ -280,7 +347,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 			if (kcb->kprobe_status == KPROBE_HIT_SS &&
 				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
 				regs->flags &= ~TF_MASK;
-				regs->flags |= kcb->kprobe_saved_eflags;
+				regs->flags |= kcb->kprobe_saved_flags;
 				goto no_kprobe;
 			}
 			/* We have reentered the kprobe_handler(), since
@@ -301,7 +368,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * another cpu right after we hit, no further
 			 * handling of this interrupt is appropriate
 			 */
-				regs->ip -= sizeof(kprobe_opcode_t);
+				regs->ip = (unsigned long)addr;
 				ret = 1;
 				goto no_kprobe;
 			}
@@ -325,7 +392,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 			 * Back up over the (now missing) int3 and run
 			 * the original instruction.
 			 */
-			regs->ip -= sizeof(kprobe_opcode_t);
+			regs->ip = (unsigned long)addr;
 			ret = 1;
 		}
 		/* Not one of ours: let kernel handle it */
@@ -341,7 +408,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 
 ss_probe:
 #if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
-	if (p->ainsn.boostable == 1 && !p->post_handler){
+	if (p->ainsn.boostable == 1 && !p->post_handler) {
 		/* Boost up -- we can execute copied instructions directly */
 		reset_current_kprobe();
 		regs->ip = (unsigned long)p->ainsn.insn;
@@ -359,16 +426,18 @@ no_kprobe:
 }
 
 /*
- * For function-return probes, init_kprobes() establishes a probepoint
- * here. When a retprobed function returns, this probe is hit and
- * trampoline_probe_handler() runs, calling the kretprobe's handler.
+ * When a retprobed function returns, this code saves registers and
+ * calls trampoline_handler() runs, which calls the kretprobe's handler.
  */
  void __kprobes kretprobe_trampoline_holder(void)
  {
 	asm volatile ( ".global kretprobe_trampoline\n"
 			"kretprobe_trampoline: \n"
 			"	pushf\n"
-			/* skip cs, ip, orig_ax */
+			/*
+			 * Skip cs, ip, orig_ax.
+			 * trampoline_handler() will plug in these values
+			 */
 			"	subl $12, %esp\n"
 			"	pushl %fs\n"
 			"	pushl %ds\n"
@@ -382,10 +451,10 @@ no_kprobe:
 			"	pushl %ebx\n"
 			"	movl %esp, %eax\n"
 			"	call trampoline_handler\n"
-			/* move flags to cs */
+			/* Move flags to cs */
 			"	movl 52(%esp), %edx\n"
 			"	movl %edx, 48(%esp)\n"
-			/* save true return address on flags */
+			/* Replace saved flags with true return address. */
 			"	movl %eax, 52(%esp)\n"
 			"	popl %ebx\n"
 			"	popl %ecx\n"
@@ -394,16 +463,16 @@ no_kprobe:
 			"	popl %edi\n"
 			"	popl %ebp\n"
 			"	popl %eax\n"
-			/* skip ip, orig_ax, es, ds, fs */
+			/* Skip ip, orig_ax, es, ds, fs */
 			"	addl $20, %esp\n"
 			"	popf\n"
 			"	ret\n");
-}
+ }
 
 /*
  * Called from kretprobe_trampoline
  */
-void *__kprobes trampoline_handler(struct pt_regs *regs)
+void * __kprobes trampoline_handler(struct pt_regs *regs)
 {
 	struct kretprobe_instance *ri = NULL;
 	struct hlist_head *head, empty_rp;
@@ -417,27 +486,27 @@ void *__kprobes trampoline_handler(struct pt_regs *regs)
 	/* fixup registers */
 	regs->cs = __KERNEL_CS | get_kernel_rpl();
 	regs->ip = trampoline_address;
-	regs->orig_ax = 0xffffffff;
+	regs->orig_ax = ~0UL;
 
 	/*
 	 * It is possible to have multiple instances associated with a given
-	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * task either because multiple functions in the call path have
+	 * return probes installed on them, and/or more then one
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
-	 *     - instances are always inserted at the head of the list
+	 *     - instances are always pushed into the head of the list
 	 *     - when multiple return probes are registered for the same
-	 *       function, the first instance's ret_addr will point to the
-	 *       real return address, and all the rest will point to
-	 *       kretprobe_trampoline
+	 *	 function, the (chronologically) first instance's ret_addr
+	 *	 will be the real return address, and all the rest will
+	 *	 point to kretprobe_trampoline.
 	 */
 	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
 		if (ri->task != current)
 			/* another task is sharing our hash bucket */
 			continue;
 
-		if (ri->rp && ri->rp->handler){
+		if (ri->rp && ri->rp->handler) {
 			__get_cpu_var(current_kprobe) = &ri->rp->kp;
 			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
 			ri->rp->handler(ri, regs);
@@ -457,13 +526,14 @@ void *__kprobes trampoline_handler(struct pt_regs *regs)
 	}
 
 	kretprobe_assert(ri, orig_ret_address, trampoline_address);
+
 	spin_unlock_irqrestore(&kretprobe_lock, flags);
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
 		hlist_del(&ri->hlist);
 		kfree(ri);
 	}
-	return (void*)orig_ret_address;
+	return (void *)orig_ret_address;
 }
 
 /*
@@ -488,48 +558,55 @@ void *__kprobes trampoline_handler(struct pt_regs *regs)
  * that is atop the stack is the address following the copied instruction.
  * We need to make it the address following the original instruction.
  *
- * This function also checks instruction size for preparing direct execution.
+ * If this is the first time we've single-stepped the instruction at
+ * this probepoint, and the instruction is boostable, boost it: add a
+ * jump instruction after the copied instruction, that jumps to the next
+ * instruction after the probepoint.
  */
 static void __kprobes resume_execution(struct kprobe *p,
 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
 {
-	unsigned long *tos = (unsigned long *)&regs->sp;
-	unsigned long copy_eip = (unsigned long)p->ainsn.insn;
-	unsigned long orig_eip = (unsigned long)p->addr;
+	unsigned long *tos = stack_addr(regs);
+	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
+	unsigned long orig_ip = (unsigned long)p->addr;
+	kprobe_opcode_t *insn = p->ainsn.insn;
 
 	regs->flags &= ~TF_MASK;
-	switch (p->ainsn.insn[0]) {
-	case 0x9c:		/* pushfl */
+	switch (*insn) {
+	case 0x9c:	/* pushfl */
 		*tos &= ~(TF_MASK | IF_MASK);
-		*tos |= kcb->kprobe_old_eflags;
+		*tos |= kcb->kprobe_old_flags;
 		break;
-	case 0xc2:		/* iret/ret/lret */
+	case 0xc2:	/* iret/ret/lret */
 	case 0xc3:
 	case 0xca:
 	case 0xcb:
 	case 0xcf:
-	case 0xea:		/* jmp absolute -- ip is correct */
+	case 0xea:	/* jmp absolute -- ip is correct */
 		/* ip is already adjusted, no more changes required */
 		p->ainsn.boostable = 1;
 		goto no_change;
-	case 0xe8:		/* call relative - Fix return addr */
-		*tos = orig_eip + (*tos - copy_eip);
+	case 0xe8:	/* call relative - Fix return addr */
+		*tos = orig_ip + (*tos - copy_ip);
 		break;
-	case 0x9a:		/* call absolute -- same as call absolute, indirect */
-		*tos = orig_eip + (*tos - copy_eip);
+	case 0x9a:	/* call absolute -- same as call absolute, indirect */
+		*tos = orig_ip + (*tos - copy_ip);
 		goto no_change;
 	case 0xff:
-		if ((p->ainsn.insn[1] & 0x30) == 0x10) {
+		if ((insn[1] & 0x30) == 0x10) {
 			/*
 			 * call absolute, indirect
 			 * Fix return addr; ip is correct.
 			 * But this is not boostable
 			 */
-			*tos = orig_eip + (*tos - copy_eip);
+			*tos = orig_ip + (*tos - copy_ip);
 			goto no_change;
-		} else if (((p->ainsn.insn[1] & 0x31) == 0x20) ||	/* jmp near, absolute indirect */
-			   ((p->ainsn.insn[1] & 0x31) == 0x21)) {	/* jmp far, absolute indirect */
-			/* ip is correct. And this is boostable */
+		} else if (((insn[1] & 0x31) == 0x20) ||
+			   ((insn[1] & 0x31) == 0x21)) {
+			/*
+			 * jmp near and far, absolute indirect
+			 * ip is correct. And this is boostable
+			 */
 			p->ainsn.boostable = 1;
 			goto no_change;
 		}
@@ -538,21 +615,21 @@ static void __kprobes resume_execution(struct kprobe *p,
 	}
 
 	if (p->ainsn.boostable == 0) {
-		if ((regs->ip > copy_eip) &&
-		    (regs->ip - copy_eip) + 5 < MAX_INSN_SIZE) {
+		if ((regs->ip > copy_ip) &&
+		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
 			/*
 			 * These instructions can be executed directly if it
 			 * jumps back to correct address.
 			 */
 			set_jmp_op((void *)regs->ip,
-				   (void *)orig_eip + (regs->ip - copy_eip));
+				   (void *)orig_ip + (regs->ip - copy_ip));
 			p->ainsn.boostable = 1;
 		} else {
 			p->ainsn.boostable = -1;
 		}
 	}
 
-	regs->ip = orig_eip + (regs->ip - copy_eip);
+	regs->ip += orig_ip - copy_ip;
 
 no_change:
 	restore_btf();
@@ -578,10 +655,10 @@ static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 	}
 
 	resume_execution(cur, regs, kcb);
-	regs->flags |= kcb->kprobe_saved_eflags;
+	regs->flags |= kcb->kprobe_saved_flags;
 	trace_hardirqs_fixup_flags(regs->flags);
 
-	/*Restore back the original saved kprobes variables and continue. */
+	/* Restore back the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
 		restore_previous_kprobe(kcb);
 		goto out;
@@ -617,7 +694,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		 * normal page fault.
 		 */
 		regs->ip = (unsigned long)cur->addr;
-		regs->flags |= kcb->kprobe_old_eflags;
+		regs->flags |= kcb->kprobe_old_flags;
 		if (kcb->kprobe_status == KPROBE_REENTER)
 			restore_previous_kprobe(kcb);
 		else
@@ -628,7 +705,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
@@ -651,7 +728,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 			return 1;
 
 		/*
-		 * fixup_exception() could not handle it,
+		 * fixup routine could not handle it,
 		 * Let do_page_fault() fix it.
 		 */
 		break;
@@ -662,7 +739,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 }
 
 /*
- * Wrapper routine to for handling exceptions.
+ * Wrapper routine for handling exceptions.
  */
 int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
 				       unsigned long val, void *data)
@@ -703,11 +780,11 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 
 	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_esp = &regs->sp;
-	addr = (unsigned long)(kcb->jprobe_saved_esp);
+	kcb->jprobe_saved_sp = stack_addr(regs);
+	addr = (unsigned long)(kcb->jprobe_saved_sp);
 
 	/*
-	 * TBD: As Linus pointed out, gcc assumes that the callee
+	 * As Linus pointed out, gcc assumes that the callee
 	 * owns the argument space and could overwrite it, e.g.
 	 * tailcall optimization. So, to be absolutely safe
 	 * we also save and restore enough stack bytes to cover
@@ -730,21 +807,20 @@ void __kprobes jprobe_return(void)
 		      "       .globl jprobe_return_end	\n"
 		      "       jprobe_return_end:	\n"
 		      "       nop			\n"::"b"
-		      (kcb->jprobe_saved_esp):"memory");
+		      (kcb->jprobe_saved_sp):"memory");
 }
 
 int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	u8 *addr = (u8 *) (regs->ip - 1);
-	unsigned long stack_addr = (unsigned long)(kcb->jprobe_saved_esp);
 	struct jprobe *jp = container_of(p, struct jprobe, kp);
 
 	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if (&regs->sp != kcb->jprobe_saved_esp) {
+		if (stack_addr(regs) != kcb->jprobe_saved_sp) {
 			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
 			printk("current sp %p does not match saved sp %p\n",
-			       &regs->sp, kcb->jprobe_saved_esp);
+			       stack_addr(regs), kcb->jprobe_saved_sp);
 			printk("Saved registers for jprobe %p\n", jp);
 			show_registers(saved_regs);
 			printk("Current registers\n");
@@ -752,20 +828,21 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 			BUG();
 		}
 		*regs = kcb->jprobe_saved_regs;
-		memcpy((kprobe_opcode_t *) stack_addr, kcb->jprobes_stack,
-		       MIN_STACK_SIZE(stack_addr));
+		memcpy((kprobe_opcode_t *)(kcb->jprobe_saved_sp),
+		       kcb->jprobes_stack,
+		       MIN_STACK_SIZE(kcb->jprobe_saved_sp));
 		preempt_enable_no_resched();
 		return 1;
 	}
 	return 0;
 }
 
-int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+int __init arch_init_kprobes(void)
 {
 	return 0;
 }
 
-int __init arch_init_kprobes(void)
+int __kprobes arch_trampoline_kprobe(struct kprobe *p)
 {
 	return 0;
 }
diff --git a/arch/x86/kernel/kprobes_64.c b/arch/x86/kernel/kprobes_64.c
index bc93b1dd9a01..2d7763749b1b 100644
--- a/arch/x86/kernel/kprobes_64.c
+++ b/arch/x86/kernel/kprobes_64.c
@@ -40,16 +40,97 @@
 #include <linux/module.h>
 #include <linux/kdebug.h>
 
+#include <asm/cacheflush.h>
+#include <asm/desc.h>
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/alternative.h>
 
 void jprobe_return_end(void);
-static void __kprobes arch_copy_kprobe(struct kprobe *p);
 
 DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
 DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
 
+#define stack_addr(regs) ((unsigned long *)regs->sp)
+
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
+	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
+	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
+	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
+	 << (row % 32))
+	/*
+	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
+	 * Groups, and some special opcodes can not boost.
+	 */
+static const u32 twobyte_is_boostable[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      ----------------------------------------------          */
+	W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) , /* 70 */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 80 */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 90 */
+	W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* a0 */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* c0 */
+	W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) , /* d0 */
+	W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* e0 */
+	W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 onebyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
+	W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
+	W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
+	W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
+	W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
+	W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
+	W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
+	W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 twobyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
+	W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
+	W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
+	W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
+	W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
+	W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
+	W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
+	W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+	W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+	W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
+	W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0)   /* ff */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+#undef W
+
 struct kretprobe_blackpoint kretprobe_blacklist[] = {
 	{"__switch_to", }, /* This function switches only current task, but
 			      doesn't switch kernel stack.*/
@@ -70,44 +151,11 @@ static __always_inline void set_jmp_op(void *from, void *to)
 }
 
 /*
- * returns non-zero if opcode is boostable
+ * returns non-zero if opcode is boostable.
  * RIP relative instructions are adjusted at copying time
  */
 static __always_inline int can_boost(kprobe_opcode_t *opcodes)
 {
-#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
-	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
-	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
-	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
-	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
-	 << (row % 64))
-	/*
-	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
-	 * Groups, and some special opcodes can not boost.
-	 */
-	static const unsigned long twobyte_is_boostable[256 / 64] = {
-		/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f        */
-		/*      ----------------------------------------------        */
-		W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0)|/* 00 */
-		W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 10 */
-		W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 20 */
-		W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0),/* 30 */
-		W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)|/* 40 */
-		W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 50 */
-		W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1)|/* 60 */
-		W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1),/* 70 */
-		W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)|/* 80 */
-		W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)|/* 90 */
-		W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* a0 */
-		W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1),/* b0 */
-		W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1)|/* c0 */
-		W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* d0 */
-		W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1)|/* e0 */
-		W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0) /* f0 */
-		/*      -----------------------------------------------       */
-		/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f        */
-	};
-#undef W
 	kprobe_opcode_t opcode;
 	kprobe_opcode_t *orig_opcodes = opcodes;
 
@@ -120,7 +168,8 @@ retry:
 	if (opcode == 0x0f) {
 		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
 			return 0;
-		return test_bit(*opcodes, twobyte_is_boostable);
+		return test_bit(*opcodes,
+				(unsigned long *)twobyte_is_boostable);
 	}
 
 	switch (opcode & 0xf0) {
@@ -169,80 +218,25 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
 		return 1;
 	}
 
-	if (*insn  >= 0x40 && *insn <= 0x4f && *++insn == 0xcf)
-		return 1;
-	return 0;
-}
-
-int __kprobes arch_prepare_kprobe(struct kprobe *p)
-{
-	/* insn: must be on special executable page on x86_64. */
-	p->ainsn.insn = get_insn_slot();
-	if (!p->ainsn.insn) {
-		return -ENOMEM;
-	}
-	arch_copy_kprobe(p);
+	/*
+	 * on 64 bit x86, 0x40-0x4f are prefixes so we need to look
+	 * at the next byte instead.. but of course not recurse infinitely
+	 */
+	if (*insn  >= 0x40 && *insn <= 0x4f)
+		return is_IF_modifier(++insn);
 	return 0;
 }
 
 /*
- * Determine if the instruction uses the %rip-relative addressing mode.
+ * Adjust the displacement if the instruction uses the %rip-relative
+ * addressing mode.
  * If it does, Return the address of the 32-bit displacement word.
  * If not, return null.
  */
-static s32 __kprobes *is_riprel(u8 *insn)
+static void __kprobes fix_riprel(struct kprobe *p)
 {
-#define W(row,b0,b1,b2,b3,b4,b5,b6,b7,b8,b9,ba,bb,bc,bd,be,bf)		      \
-	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
-	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
-	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
-	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
-	 << (row % 64))
-	static const u64 onebyte_has_modrm[256 / 64] = {
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-		/*      -------------------------------         */
-		W(0x00, 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,0)| /* 00 */
-		W(0x10, 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,0)| /* 10 */
-		W(0x20, 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,0)| /* 20 */
-		W(0x30, 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,0), /* 30 */
-		W(0x40, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 40 */
-		W(0x50, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 50 */
-		W(0x60, 0,0,1,1,0,0,0,0,0,1,0,1,0,0,0,0)| /* 60 */
-		W(0x70, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 70 */
-		W(0x80, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 80 */
-		W(0x90, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 90 */
-		W(0xa0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* a0 */
-		W(0xb0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* b0 */
-		W(0xc0, 1,1,0,0,1,1,1,1,0,0,0,0,0,0,0,0)| /* c0 */
-		W(0xd0, 1,1,1,1,0,0,0,0,1,1,1,1,1,1,1,1)| /* d0 */
-		W(0xe0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* e0 */
-		W(0xf0, 0,0,0,0,0,0,1,1,0,0,0,0,0,0,1,1)  /* f0 */
-		/*      -------------------------------         */
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-	};
-	static const u64 twobyte_has_modrm[256 / 64] = {
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-		/*      -------------------------------         */
-		W(0x00, 1,1,1,1,0,0,0,0,0,0,0,0,0,1,0,1)| /* 0f */
-		W(0x10, 1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0)| /* 1f */
-		W(0x20, 1,1,1,1,1,0,1,0,1,1,1,1,1,1,1,1)| /* 2f */
-		W(0x30, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), /* 3f */
-		W(0x40, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 4f */
-		W(0x50, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 5f */
-		W(0x60, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 6f */
-		W(0x70, 1,1,1,1,1,1,1,0,0,0,0,0,1,1,1,1), /* 7f */
-		W(0x80, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)| /* 8f */
-		W(0x90, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* 9f */
-		W(0xa0, 0,0,0,1,1,1,1,1,0,0,0,1,1,1,1,1)| /* af */
-		W(0xb0, 1,1,1,1,1,1,1,1,0,0,1,1,1,1,1,1), /* bf */
-		W(0xc0, 1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0)| /* cf */
-		W(0xd0, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* df */
-		W(0xe0, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)| /* ef */
-		W(0xf0, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0)  /* ff */
-		/*      -------------------------------         */
-		/*      0 1 2 3 4 5 6 7 8 9 a b c d e f         */
-	};
-#undef	W
+	u8 *insn = p->ainsn.insn;
+	s64 disp;
 	int need_modrm;
 
 	/* Skip legacy instruction prefixes.  */
@@ -271,54 +265,60 @@ static s32 __kprobes *is_riprel(u8 *insn)
 
 	if (*insn == 0x0f) {	/* Two-byte opcode.  */
 		++insn;
-		need_modrm = test_bit(*insn, twobyte_has_modrm);
-	} else {		/* One-byte opcode.  */
-		need_modrm = test_bit(*insn, onebyte_has_modrm);
-	}
+		need_modrm = test_bit(*insn,
+				      (unsigned long *)twobyte_has_modrm);
+	} else			/* One-byte opcode.  */
+		need_modrm = test_bit(*insn,
+				      (unsigned long *)onebyte_has_modrm);
 
 	if (need_modrm) {
 		u8 modrm = *++insn;
 		if ((modrm & 0xc7) == 0x05) { /* %rip+disp32 addressing mode */
 			/* Displacement follows ModRM byte.  */
-			return (s32 *) ++insn;
+			++insn;
+			/*
+			 * The copied instruction uses the %rip-relative
+			 * addressing mode.  Adjust the displacement for the
+			 * difference between the original location of this
+			 * instruction and the location of the copy that will
+			 * actually be run.  The tricky bit here is making sure
+			 * that the sign extension happens correctly in this
+			 * calculation, since we need a signed 32-bit result to
+			 * be sign-extended to 64 bits when it's added to the
+			 * %rip value and yield the same 64-bit result that the
+			 * sign-extension of the original signed 32-bit
+			 * displacement would have given.
+			 */
+			disp = (u8 *) p->addr + *((s32 *) insn) -
+			       (u8 *) p->ainsn.insn;
+			BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
+			*(s32 *)insn = (s32) disp;
 		}
 	}
-
-	/* No %rip-relative addressing mode here.  */
-	return NULL;
 }
 
 static void __kprobes arch_copy_kprobe(struct kprobe *p)
 {
-	s32 *ripdisp;
-	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE);
-	ripdisp = is_riprel(p->ainsn.insn);
-	if (ripdisp) {
-		/*
-		 * The copied instruction uses the %rip-relative
-		 * addressing mode.  Adjust the displacement for the
-		 * difference between the original location of this
-		 * instruction and the location of the copy that will
-		 * actually be run.  The tricky bit here is making sure
-		 * that the sign extension happens correctly in this
-		 * calculation, since we need a signed 32-bit result to
-		 * be sign-extended to 64 bits when it's added to the
-		 * %rip value and yield the same 64-bit result that the
-		 * sign-extension of the original signed 32-bit
-		 * displacement would have given.
-		 */
-		s64 disp = (u8 *) p->addr + *ripdisp - (u8 *) p->ainsn.insn;
-		BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
-		*ripdisp = disp;
-	}
-	if (can_boost(p->addr)) {
+	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	fix_riprel(p);
+	if (can_boost(p->addr))
 		p->ainsn.boostable = 0;
-	} else {
+	else
 		p->ainsn.boostable = -1;
-	}
+
 	p->opcode = *p->addr;
 }
 
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	/* insn: must be on special executable page on x86. */
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn)
+		return -ENOMEM;
+	arch_copy_kprobe(p);
+	return 0;
+}
+
 void __kprobes arch_arm_kprobe(struct kprobe *p)
 {
 	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
@@ -340,26 +340,26 @@ static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
 	kcb->prev_kprobe.kp = kprobe_running();
 	kcb->prev_kprobe.status = kcb->kprobe_status;
-	kcb->prev_kprobe.old_rflags = kcb->kprobe_old_rflags;
-	kcb->prev_kprobe.saved_rflags = kcb->kprobe_saved_rflags;
+	kcb->prev_kprobe.old_flags = kcb->kprobe_old_flags;
+	kcb->prev_kprobe.saved_flags = kcb->kprobe_saved_flags;
 }
 
 static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
 	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
 	kcb->kprobe_status = kcb->prev_kprobe.status;
-	kcb->kprobe_old_rflags = kcb->prev_kprobe.old_rflags;
-	kcb->kprobe_saved_rflags = kcb->prev_kprobe.saved_rflags;
+	kcb->kprobe_old_flags = kcb->prev_kprobe.old_flags;
+	kcb->kprobe_saved_flags = kcb->prev_kprobe.saved_flags;
 }
 
 static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 				struct kprobe_ctlblk *kcb)
 {
 	__get_cpu_var(current_kprobe) = p;
-	kcb->kprobe_saved_rflags = kcb->kprobe_old_rflags
+	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
 		= (regs->flags & (TF_MASK | IF_MASK));
 	if (is_IF_modifier(p->ainsn.insn))
-		kcb->kprobe_saved_rflags &= ~IF_MASK;
+		kcb->kprobe_saved_flags &= ~IF_MASK;
 }
 
 static __always_inline void clear_btf(void)
@@ -390,20 +390,27 @@ static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
-	unsigned long *sara = (unsigned long *)regs->sp;
+	unsigned long *sara = stack_addr(regs);
 
 	ri->ret_addr = (kprobe_opcode_t *) *sara;
+
 	/* Replace the return addr with trampoline addr */
 	*sara = (unsigned long) &kretprobe_trampoline;
 }
 
-int __kprobes kprobe_handler(struct pt_regs *regs)
+/*
+ * Interrupts are disabled on entry as trap3 is an interrupt gate and they
+ * remain disabled thorough out this function.
+ */
+static int __kprobes kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
 	int ret = 0;
-	kprobe_opcode_t *addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
+	kprobe_opcode_t *addr;
 	struct kprobe_ctlblk *kcb;
 
+	addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
+
 	/*
 	 * We don't want to be preempted for the entire
 	 * duration of kprobe processing
@@ -418,7 +425,7 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 			if (kcb->kprobe_status == KPROBE_HIT_SS &&
 				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
 				regs->flags &= ~TF_MASK;
-				regs->flags |= kcb->kprobe_saved_rflags;
+				regs->flags |= kcb->kprobe_saved_flags;
 				goto no_kprobe;
 			} else if (kcb->kprobe_status == KPROBE_HIT_SSDONE) {
 				/* TODO: Provide re-entrancy from
@@ -429,22 +436,20 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
 				arch_disarm_kprobe(p);
 				regs->ip = (unsigned long)p->addr;
 				reset_current_kprobe();
-				ret = 1;
-			} else {
-				/* We have reentered the kprobe_handler(), since
-				 * another probe was hit while within the
-				 * handler. We here save the original kprobe
-				 * variables and just single step on instruction
-				 * of the new probe without calling any user
-				 * handlers.
-				 */
-				save_previous_kprobe(kcb);
-				set_current_kprobe(p, regs, kcb);
-				kprobes_inc_nmissed_count(p);
-				prepare_singlestep(p, regs);
-				kcb->kprobe_status = KPROBE_REENTER;
 				return 1;
 			}
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe(kcb);
+			set_current_kprobe(p, regs, kcb);
+			kprobes_inc_nmissed_count(p);
+			prepare_singlestep(p, regs);
+			kcb->kprobe_status = KPROBE_REENTER;
+			return 1;
 		} else {
 			if (*addr != BREAKPOINT_INSTRUCTION) {
 			/* The breakpoint instruction was removed by
@@ -578,23 +583,23 @@ fastcall void * __kprobes trampoline_handler(struct pt_regs *regs)
 	INIT_HLIST_HEAD(&empty_rp);
 	spin_lock_irqsave(&kretprobe_lock, flags);
 	head = kretprobe_inst_table_head(current);
-	/* fixup rt_regs */
+	/* fixup registers */
 	regs->cs = __KERNEL_CS;
 	regs->ip = trampoline_address;
-	regs->orig_ax = 0xffffffffffffffff;
+	regs->orig_ax = ~0UL;
 
 	/*
 	 * It is possible to have multiple instances associated with a given
-	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * task either because multiple functions in the call path have
+	 * return probes installed on them, and/or more then one
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
-	 *     - instances are always inserted at the head of the list
+	 *     - instances are always pushed into the head of the list
 	 *     - when multiple return probes are registered for the same
-	 *       function, the first instance's ret_addr will point to the
-	 *       real return address, and all the rest will point to
-	 *       kretprobe_trampoline
+	 *	 function, the (chronologically) first instance's ret_addr
+	 *	 will be the real return address, and all the rest will
+	 *	 point to kretprobe_trampoline.
 	 */
 	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
 		if (ri->task != current)
@@ -661,9 +666,9 @@ fastcall void * __kprobes trampoline_handler(struct pt_regs *regs)
 static void __kprobes resume_execution(struct kprobe *p,
 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
 {
-	unsigned long *tos = (unsigned long *)regs->sp;
-	unsigned long copy_rip = (unsigned long)p->ainsn.insn;
-	unsigned long orig_rip = (unsigned long)p->addr;
+	unsigned long *tos = stack_addr(regs);
+	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
+	unsigned long orig_ip = (unsigned long)p->addr;
 	kprobe_opcode_t *insn = p->ainsn.insn;
 
 	/*skip the REX prefix*/
@@ -674,7 +679,7 @@ static void __kprobes resume_execution(struct kprobe *p,
 	switch (*insn) {
 	case 0x9c:	/* pushfl */
 		*tos &= ~(TF_MASK | IF_MASK);
-		*tos |= kcb->kprobe_old_rflags;
+		*tos |= kcb->kprobe_old_flags;
 		break;
 	case 0xc2:	/* iret/ret/lret */
 	case 0xc3:
@@ -686,18 +691,23 @@ static void __kprobes resume_execution(struct kprobe *p,
 		p->ainsn.boostable = 1;
 		goto no_change;
 	case 0xe8:	/* call relative - Fix return addr */
-		*tos = orig_rip + (*tos - copy_rip);
+		*tos = orig_ip + (*tos - copy_ip);
 		break;
 	case 0xff:
 		if ((insn[1] & 0x30) == 0x10) {
-			/* call absolute, indirect */
-			/* Fix return addr; ip is correct. */
-			/* not boostable */
-			*tos = orig_rip + (*tos - copy_rip);
+			/*
+			 * call absolute, indirect
+			 * Fix return addr; ip is correct.
+			 * But this is not boostable
+			 */
+			*tos = orig_ip + (*tos - copy_ip);
 			goto no_change;
-		} else if (((insn[1] & 0x31) == 0x20) ||	/* jmp near, absolute indirect */
-			   ((insn[1] & 0x31) == 0x21)) {	/* jmp far, absolute indirect */
-			/* ip is correct. And this is boostable */
+		} else if (((insn[1] & 0x31) == 0x20) ||
+			   ((insn[1] & 0x31) == 0x21)) {
+			/*
+			 * jmp near and far, absolute indirect
+			 * ip is correct. And this is boostable
+			 */
 			p->ainsn.boostable = 1;
 			goto no_change;
 		}
@@ -706,21 +716,21 @@ static void __kprobes resume_execution(struct kprobe *p,
 	}
 
 	if (p->ainsn.boostable == 0) {
-		if ((regs->ip > copy_rip) &&
-		    (regs->ip - copy_rip) + 5 < MAX_INSN_SIZE) {
+		if ((regs->ip > copy_ip) &&
+		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
 			/*
 			 * These instructions can be executed directly if it
 			 * jumps back to correct address.
 			 */
 			set_jmp_op((void *)regs->ip,
-				   (void *)orig_rip + (regs->ip - copy_rip));
+				   (void *)orig_ip + (regs->ip - copy_ip));
 			p->ainsn.boostable = 1;
 		} else {
 			p->ainsn.boostable = -1;
 		}
 	}
 
-	regs->ip = orig_rip + (regs->ip - copy_rip);
+	regs->ip += orig_ip - copy_ip;
 
 no_change:
 	restore_btf();
@@ -728,7 +738,11 @@ no_change:
 	return;
 }
 
-int __kprobes post_kprobe_handler(struct pt_regs *regs)
+/*
+ * Interrupts are disabled on entry as trap1 is an interrupt gate and they
+ * remain disabled thoroughout this function.
+ */
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *cur = kprobe_running();
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
@@ -742,10 +756,10 @@ int __kprobes post_kprobe_handler(struct pt_regs *regs)
 	}
 
 	resume_execution(cur, regs, kcb);
-	regs->flags |= kcb->kprobe_saved_rflags;
+	regs->flags |= kcb->kprobe_saved_flags;
 	trace_hardirqs_fixup_flags(regs->flags);
 
-	/* Restore the original saved kprobes variables and continue. */
+	/* Restore back the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
 		restore_previous_kprobe(kcb);
 		goto out;
@@ -782,7 +796,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		 * normal page fault.
 		 */
 		regs->ip = (unsigned long)cur->addr;
-		regs->flags |= kcb->kprobe_old_rflags;
+		regs->flags |= kcb->kprobe_old_flags;
 		if (kcb->kprobe_status == KPROBE_REENTER)
 			restore_previous_kprobe(kcb);
 		else
@@ -793,7 +807,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
@@ -819,7 +833,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		}
 
 		/*
-		 * fixup() could not handle it,
+		 * fixup routine could not handle it,
 		 * Let do_page_fault() fix it.
 		 */
 		break;
@@ -838,7 +852,7 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
 	struct die_args *args = (struct die_args *)data;
 	int ret = NOTIFY_DONE;
 
-	if (args->regs && user_mode(args->regs))
+	if (args->regs && user_mode_vm(args->regs))
 		return ret;
 
 	switch (val) {
@@ -871,8 +885,9 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 
 	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_rsp = (long *) regs->sp;
-	addr = (unsigned long)(kcb->jprobe_saved_rsp);
+	kcb->jprobe_saved_sp = stack_addr(regs);
+	addr = (unsigned long)(kcb->jprobe_saved_sp);
+
 	/*
 	 * As Linus pointed out, gcc assumes that the callee
 	 * owns the argument space and could overwrite it, e.g.
@@ -897,21 +912,20 @@ void __kprobes jprobe_return(void)
 		      "       .globl jprobe_return_end	\n"
 		      "       jprobe_return_end:	\n"
 		      "       nop			\n"::"b"
-		      (kcb->jprobe_saved_rsp):"memory");
+		      (kcb->jprobe_saved_sp):"memory");
 }
 
 int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	u8 *addr = (u8 *) (regs->ip - 1);
-	unsigned long stack_addr = (unsigned long)(kcb->jprobe_saved_rsp);
 	struct jprobe *jp = container_of(p, struct jprobe, kp);
 
 	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if ((unsigned long *)regs->sp != kcb->jprobe_saved_rsp) {
+		if (stack_addr(regs) != kcb->jprobe_saved_sp) {
 			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
 			printk("current sp %p does not match saved sp %p\n",
-			       (long *)regs->sp, kcb->jprobe_saved_rsp);
+			       stack_addr(regs), kcb->jprobe_saved_sp);
 			printk("Saved registers for jprobe %p\n", jp);
 			show_registers(saved_regs);
 			printk("Current registers\n");
@@ -919,8 +933,9 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 			BUG();
 		}
 		*regs = kcb->jprobe_saved_regs;
-		memcpy((kprobe_opcode_t *) stack_addr, kcb->jprobes_stack,
-		       MIN_STACK_SIZE(stack_addr));
+		memcpy((kprobe_opcode_t *)(kcb->jprobe_saved_sp),
+		       kcb->jprobes_stack,
+		       MIN_STACK_SIZE(kcb->jprobe_saved_sp));
 		preempt_enable_no_resched();
 		return 1;
 	}
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 8aed912b04ec..db8d748814e4 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -25,7 +25,6 @@
 #include <linux/kprobes.h>
 #include <linux/uaccess.h>
 #include <linux/kdebug.h>
-#include <linux/kprobes.h>
 
 #include <asm/system.h>
 #include <asm/desc.h>
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 88a7abda29ce..162050d4e5a3 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -25,7 +25,6 @@
 #include <linux/kprobes.h>
 #include <linux/uaccess.h>
 #include <linux/kdebug.h>
-#include <linux/kprobes.h>
 
 #include <asm/system.h>
 #include <asm/pgalloc.h>
diff --git a/include/asm-x86/kprobes_32.h b/include/asm-x86/kprobes_32.h
index 2f38315bc39f..94dff09b2ebe 100644
--- a/include/asm-x86/kprobes_32.h
+++ b/include/asm-x86/kprobes_32.h
@@ -2,7 +2,6 @@
 #define _ASM_KPROBES_H
 /*
  *  Kernel Probes (KProbes)
- *  include/asm-i386/kprobes.h
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -23,14 +22,17 @@
  * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
  *		Probes initial implementation ( includes suggestions from
  *		Rusty Russell).
+ * 2004-Oct	Prasanna S Panchamukhi <prasanna@in.ibm.com> and Jim Keniston
+ *		kenistoj@us.ibm.com adopted from i386.
  */
 #include <linux/types.h>
 #include <linux/ptrace.h>
+#include <linux/percpu.h>
 
 #define  __ARCH_WANT_KPROBES_INSN_SLOT
 
-struct kprobe;
 struct pt_regs;
+struct kprobe;
 
 typedef u8 kprobe_opcode_t;
 #define BREAKPOINT_INSTRUCTION	0xcc
@@ -38,9 +40,11 @@ typedef u8 kprobe_opcode_t;
 #define MAX_INSN_SIZE 16
 #define MAX_STACK_SIZE 64
 #define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) \
+	(((unsigned long)current_thread_info()) + THREAD_SIZE \
+	 - (unsigned long)(ADDR))) \
 	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
+	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
+	   - (unsigned long)(ADDR)))
 
 #define ARCH_SUPPORTS_KRETPROBES
 #define flush_insn_slot(p)	do { } while (0)
@@ -55,8 +59,12 @@ struct arch_specific_insn {
 	/* copy of the original instruction */
 	kprobe_opcode_t *insn;
 	/*
-	 * If this flag is not 0, this kprobe can be boost when its
-	 * post_handler and break_handler is not set.
+	 * boostable = -1: This instruction type is not boostable.
+	 * boostable = 0: This instruction type is boostable.
+	 * boostable = 1: This instruction has been boosted: we have
+	 * added a relative jump after the instruction copy in insn,
+	 * so no single-step and fixup are needed (unless there's
+	 * a post_handler or break_handler).
 	 */
 	int boostable;
 };
@@ -64,16 +72,16 @@ struct arch_specific_insn {
 struct prev_kprobe {
 	struct kprobe *kp;
 	unsigned long status;
-	unsigned long old_eflags;
-	unsigned long saved_eflags;
+	unsigned long old_flags;
+	unsigned long saved_flags;
 };
 
 /* per-cpu kprobe control block */
 struct kprobe_ctlblk {
 	unsigned long kprobe_status;
-	unsigned long kprobe_old_eflags;
-	unsigned long kprobe_saved_eflags;
-	unsigned long *jprobe_saved_esp;
+	unsigned long kprobe_old_flags;
+	unsigned long kprobe_saved_flags;
+	unsigned long *jprobe_saved_sp;
 	struct pt_regs jprobe_saved_regs;
 	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
 	struct prev_kprobe prev_kprobe;
@@ -88,7 +96,7 @@ static inline void restore_interrupts(struct pt_regs *regs)
 		local_irq_enable();
 }
 
+extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
-extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
 #endif				/* _ASM_KPROBES_H */
diff --git a/include/asm-x86/kprobes_64.h b/include/asm-x86/kprobes_64.h
index e7e921dac7c2..94dff09b2ebe 100644
--- a/include/asm-x86/kprobes_64.h
+++ b/include/asm-x86/kprobes_64.h
@@ -2,7 +2,6 @@
 #define _ASM_KPROBES_H
 /*
  *  Kernel Probes (KProbes)
- *  include/asm-x86_64/kprobes.h
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -20,6 +19,9 @@
  *
  * Copyright (C) IBM Corporation, 2002, 2004
  *
+ * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
+ *		Probes initial implementation ( includes suggestions from
+ *		Rusty Russell).
  * 2004-Oct	Prasanna S Panchamukhi <prasanna@in.ibm.com> and Jim Keniston
  *		kenistoj@us.ibm.com adopted from i386.
  */
@@ -35,19 +37,22 @@ struct kprobe;
 typedef u8 kprobe_opcode_t;
 #define BREAKPOINT_INSTRUCTION	0xcc
 #define RELATIVEJUMP_INSTRUCTION 0xe9
-#define MAX_INSN_SIZE 15
+#define MAX_INSN_SIZE 16
 #define MAX_STACK_SIZE 64
 #define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) \
+	(((unsigned long)current_thread_info()) + THREAD_SIZE \
+	 - (unsigned long)(ADDR))) \
 	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
+	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
+	   - (unsigned long)(ADDR)))
 
 #define ARCH_SUPPORTS_KRETPROBES
+#define flush_insn_slot(p)	do { } while (0)
+
 extern const int kretprobe_blacklist_size;
 
+void arch_remove_kprobe(struct kprobe *p);
 void kretprobe_trampoline(void);
-extern void arch_remove_kprobe(struct kprobe *p);
-#define flush_insn_slot(p)	do { } while (0)
 
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
@@ -67,16 +72,16 @@ struct arch_specific_insn {
 struct prev_kprobe {
 	struct kprobe *kp;
 	unsigned long status;
-	unsigned long old_rflags;
-	unsigned long saved_rflags;
+	unsigned long old_flags;
+	unsigned long saved_flags;
 };
 
 /* per-cpu kprobe control block */
 struct kprobe_ctlblk {
 	unsigned long kprobe_status;
-	unsigned long kprobe_old_rflags;
-	unsigned long kprobe_saved_rflags;
-	unsigned long *jprobe_saved_rsp;
+	unsigned long kprobe_old_flags;
+	unsigned long kprobe_saved_flags;
+	unsigned long *jprobe_saved_sp;
 	struct pt_regs jprobe_saved_regs;
 	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
 	struct prev_kprobe prev_kprobe;
@@ -91,10 +96,7 @@ static inline void restore_interrupts(struct pt_regs *regs)
 		local_irq_enable();
 }
 
-extern int post_kprobe_handler(struct pt_regs *regs);
 extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
-extern int kprobe_handler(struct pt_regs *regs);
-
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
 #endif				/* _ASM_KPROBES_H */
-- 
cgit v1.2.3


From d6be29b871e285d33be0e3025929e2d6bcabb0c0 Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: x86: kprobes code for x86 unification

This patch unifies kprobes code.

- Unify kprobes_*.h to kprobes.h
- Unify kprobes_*.c to kprobes.c
  (Differences are separated by ifdefs)
 - Most differences are related to REX prefix and rip relatives.
 - Two inline assembly code are different.
 - One difference in kprobe_handlre()
 - One fixup exception code is different, but it will be unified
   if mm/extable_*.c are unified.
- Merge history logs into arch/x86/kernel/kprobes.c.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile_32  |    2 +-
 arch/x86/kernel/Makefile_64  |    2 +-
 arch/x86/kernel/kprobes.c    | 1045 ++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/kprobes_32.c |  848 ----------------------------------
 arch/x86/kernel/kprobes_64.c |  953 --------------------------------------
 include/asm-x86/kprobes.h    |  103 ++++-
 include/asm-x86/kprobes_32.h |  102 -----
 include/asm-x86/kprobes_64.h |  102 -----
 8 files changed, 1145 insertions(+), 2012 deletions(-)
 create mode 100644 arch/x86/kernel/kprobes.c
 delete mode 100644 arch/x86/kernel/kprobes_32.c
 delete mode 100644 arch/x86/kernel/kprobes_64.c
 delete mode 100644 include/asm-x86/kprobes_32.h
 delete mode 100644 include/asm-x86/kprobes_64.h

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index a854e23eac0b..c7a959da363a 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -35,7 +35,7 @@ obj-$(CONFIG_KEXEC)		+= machine_kexec_32.o relocate_kernel_32.o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_32.o
 obj-$(CONFIG_X86_NUMAQ)		+= numaq_32.o
 obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit_32.o
-obj-$(CONFIG_KPROBES)		+= kprobes_32.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_MODULES)		+= module_32.o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat_32.o
 obj-$(CONFIG_EFI) 		+= efi.o efi_32.o efi_stub_32.o
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index e5093dd8cf01..fbb370071239 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -34,7 +34,7 @@ obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
 obj-$(CONFIG_CALGARY_IOMMU)	+= pci-calgary_64.o tce_64.o
 obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb_64.o
-obj-$(CONFIG_KPROBES)		+= kprobes_64.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
 obj-$(CONFIG_X86_VSMP)		+= vsmp_64.o
 obj-$(CONFIG_K8_NB)		+= k8.o
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
new file mode 100644
index 000000000000..9aadd4d4a229
--- /dev/null
+++ b/arch/x86/kernel/kprobes.c
@@ -0,0 +1,1045 @@
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
+ *		Probes initial implementation ( includes contributions from
+ *		Rusty Russell).
+ * 2004-July	Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
+ *		interface to access function arguments.
+ * 2004-Oct	Jim Keniston <jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ *		<prasanna@in.ibm.com> adapted for x86_64 from i386.
+ * 2005-Mar	Roland McGrath <roland@redhat.com>
+ *		Fixed to handle %rip-relative addressing mode correctly.
+ * 2005-May	Hien Nguyen <hien@us.ibm.com>, Jim Keniston
+ *		<jkenisto@us.ibm.com> and Prasanna S Panchamukhi
+ *		<prasanna@in.ibm.com> added function-return probes.
+ * 2005-May	Rusty Lynch <rusty.lynch@intel.com>
+ * 		Added function return probes functionality
+ * 2006-Feb	Masami Hiramatsu <hiramatu@sdl.hitachi.co.jp> added
+ * 		kprobe-booster and kretprobe-booster for i386.
+ * 2007-Dec	Masami Hiramatsu <mhiramat@redhat.com> added kprobe-booster
+ * 		and kretprobe-booster for x86-64
+ * 2007-Dec	Masami Hiramatsu <mhiramat@redhat.com>, Arjan van de Ven
+ * 		<arjan@infradead.org> and Jim Keniston <jkenisto@us.ibm.com>
+ * 		unified x86 kprobes code.
+ */
+
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/preempt.h>
+#include <linux/module.h>
+#include <linux/kdebug.h>
+
+#include <asm/cacheflush.h>
+#include <asm/desc.h>
+#include <asm/pgtable.h>
+#include <asm/uaccess.h>
+#include <asm/alternative.h>
+
+void jprobe_return_end(void);
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+#ifdef CONFIG_X86_64
+#define stack_addr(regs) ((unsigned long *)regs->sp)
+#else
+/*
+ * "&regs->sp" looks wrong, but it's correct for x86_32.  x86_32 CPUs
+ * don't save the ss and esp registers if the CPU is already in kernel
+ * mode when it traps.  So for kprobes, regs->sp and regs->ss are not
+ * the [nonexistent] saved stack pointer and ss register, but rather
+ * the top 8 bytes of the pre-int3 stack.  So &regs->sp happens to
+ * point to the top of the pre-int3 stack.
+ */
+#define stack_addr(regs) ((unsigned long *)&regs->sp)
+#endif
+
+#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
+	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
+	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
+	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
+	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
+	 << (row % 32))
+	/*
+	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
+	 * Groups, and some special opcodes can not boost.
+	 */
+static const u32 twobyte_is_boostable[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      ----------------------------------------------          */
+	W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) , /* 70 */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 80 */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 90 */
+	W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* a0 */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* c0 */
+	W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) , /* d0 */
+	W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* e0 */
+	W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 onebyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
+	W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
+	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
+	W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
+	W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
+	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
+	W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
+	W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
+	W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
+	W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
+	W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
+	W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
+	W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
+	W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
+	W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
+	W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1)   /* f0 */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+static const u32 twobyte_has_modrm[256 / 32] = {
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+	/*      -----------------------------------------------         */
+	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
+	W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
+	W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
+	W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
+	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
+	W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
+	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
+	W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
+	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
+	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
+	W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
+	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
+	W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
+	W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
+	W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
+	W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0)   /* ff */
+	/*      -----------------------------------------------         */
+	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
+};
+#undef W
+
+struct kretprobe_blackpoint kretprobe_blacklist[] = {
+	{"__switch_to", }, /* This function switches only current task, but
+			      doesn't switch kernel stack.*/
+	{NULL, NULL}	/* Terminator */
+};
+const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist);
+
+/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
+static __always_inline void set_jmp_op(void *from, void *to)
+{
+	struct __arch_jmp_op {
+		char op;
+		s32 raddr;
+	} __attribute__((packed)) * jop;
+	jop = (struct __arch_jmp_op *)from;
+	jop->raddr = (s32)((long)(to) - ((long)(from) + 5));
+	jop->op = RELATIVEJUMP_INSTRUCTION;
+}
+
+/*
+ * Returns non-zero if opcode is boostable.
+ * RIP relative instructions are adjusted at copying time in 64 bits mode
+ */
+static __always_inline int can_boost(kprobe_opcode_t *opcodes)
+{
+	kprobe_opcode_t opcode;
+	kprobe_opcode_t *orig_opcodes = opcodes;
+
+retry:
+	if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+		return 0;
+	opcode = *(opcodes++);
+
+	/* 2nd-byte opcode */
+	if (opcode == 0x0f) {
+		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
+			return 0;
+		return test_bit(*opcodes,
+				(unsigned long *)twobyte_is_boostable);
+	}
+
+	switch (opcode & 0xf0) {
+#ifdef CONFIG_X86_64
+	case 0x40:
+		goto retry; /* REX prefix is boostable */
+#endif
+	case 0x60:
+		if (0x63 < opcode && opcode < 0x67)
+			goto retry; /* prefixes */
+		/* can't boost Address-size override and bound */
+		return (opcode != 0x62 && opcode != 0x67);
+	case 0x70:
+		return 0; /* can't boost conditional jump */
+	case 0xc0:
+		/* can't boost software-interruptions */
+		return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
+	case 0xd0:
+		/* can boost AA* and XLAT */
+		return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7);
+	case 0xe0:
+		/* can boost in/out and absolute jmps */
+		return ((opcode & 0x04) || opcode == 0xea);
+	case 0xf0:
+		if ((opcode & 0x0c) == 0 && opcode != 0xf1)
+			goto retry; /* lock/rep(ne) prefix */
+		/* clear and set flags are boostable */
+		return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
+	default:
+		/* segment override prefixes are boostable */
+		if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
+			goto retry; /* prefixes */
+		/* CS override prefix and call are not boostable */
+		return (opcode != 0x2e && opcode != 0x9a);
+	}
+}
+
+/*
+ * Returns non-zero if opcode modifies the interrupt flag.
+ */
+static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
+{
+	switch (*insn) {
+	case 0xfa:		/* cli */
+	case 0xfb:		/* sti */
+	case 0xcf:		/* iret/iretd */
+	case 0x9d:		/* popf/popfd */
+		return 1;
+	}
+#ifdef CONFIG_X86_64
+	/*
+	 * on 64 bit x86, 0x40-0x4f are prefixes so we need to look
+	 * at the next byte instead.. but of course not recurse infinitely
+	 */
+	if (*insn  >= 0x40 && *insn <= 0x4f)
+		return is_IF_modifier(++insn);
+#endif
+	return 0;
+}
+
+#ifdef CONFIG_X86_64
+/*
+ * Adjust the displacement if the instruction uses the %rip-relative
+ * addressing mode.
+ * If it does, Return the address of the 32-bit displacement word.
+ * If not, return null.
+ */
+static void __kprobes fix_riprel(struct kprobe *p)
+{
+	u8 *insn = p->ainsn.insn;
+	s64 disp;
+	int need_modrm;
+
+	/* Skip legacy instruction prefixes.  */
+	while (1) {
+		switch (*insn) {
+		case 0x66:
+		case 0x67:
+		case 0x2e:
+		case 0x3e:
+		case 0x26:
+		case 0x64:
+		case 0x65:
+		case 0x36:
+		case 0xf0:
+		case 0xf3:
+		case 0xf2:
+			++insn;
+			continue;
+		}
+		break;
+	}
+
+	/* Skip REX instruction prefix.  */
+	if ((*insn & 0xf0) == 0x40)
+		++insn;
+
+	if (*insn == 0x0f) {
+		/* Two-byte opcode.  */
+		++insn;
+		need_modrm = test_bit(*insn,
+				      (unsigned long *)twobyte_has_modrm);
+	} else
+		/* One-byte opcode.  */
+		need_modrm = test_bit(*insn,
+				      (unsigned long *)onebyte_has_modrm);
+
+	if (need_modrm) {
+		u8 modrm = *++insn;
+		if ((modrm & 0xc7) == 0x05) {
+			/* %rip+disp32 addressing mode */
+			/* Displacement follows ModRM byte.  */
+			++insn;
+			/*
+			 * The copied instruction uses the %rip-relative
+			 * addressing mode.  Adjust the displacement for the
+			 * difference between the original location of this
+			 * instruction and the location of the copy that will
+			 * actually be run.  The tricky bit here is making sure
+			 * that the sign extension happens correctly in this
+			 * calculation, since we need a signed 32-bit result to
+			 * be sign-extended to 64 bits when it's added to the
+			 * %rip value and yield the same 64-bit result that the
+			 * sign-extension of the original signed 32-bit
+			 * displacement would have given.
+			 */
+			disp = (u8 *) p->addr + *((s32 *) insn) -
+			       (u8 *) p->ainsn.insn;
+			BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
+			*(s32 *)insn = (s32) disp;
+		}
+	}
+}
+#endif
+
+static void __kprobes arch_copy_kprobe(struct kprobe *p)
+{
+	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+#ifdef CONFIG_X86_64
+	fix_riprel(p);
+#endif
+	if (can_boost(p->addr))
+		p->ainsn.boostable = 0;
+	else
+		p->ainsn.boostable = -1;
+
+	p->opcode = *p->addr;
+}
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	/* insn: must be on special executable page on x86. */
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn)
+		return -ENOMEM;
+	arch_copy_kprobe(p);
+	return 0;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	text_poke(p->addr, &p->opcode, 1);
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	mutex_lock(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
+	mutex_unlock(&kprobe_mutex);
+}
+
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	kcb->prev_kprobe.kp = kprobe_running();
+	kcb->prev_kprobe.status = kcb->kprobe_status;
+	kcb->prev_kprobe.old_flags = kcb->kprobe_old_flags;
+	kcb->prev_kprobe.saved_flags = kcb->kprobe_saved_flags;
+}
+
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
+	kcb->kprobe_status = kcb->prev_kprobe.status;
+	kcb->kprobe_old_flags = kcb->prev_kprobe.old_flags;
+	kcb->kprobe_saved_flags = kcb->prev_kprobe.saved_flags;
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+				struct kprobe_ctlblk *kcb)
+{
+	__get_cpu_var(current_kprobe) = p;
+	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
+		= (regs->flags & (TF_MASK | IF_MASK));
+	if (is_IF_modifier(p->ainsn.insn))
+		kcb->kprobe_saved_flags &= ~IF_MASK;
+}
+
+static __always_inline void clear_btf(void)
+{
+	if (test_thread_flag(TIF_DEBUGCTLMSR))
+		wrmsr(MSR_IA32_DEBUGCTLMSR, 0, 0);
+}
+
+static __always_inline void restore_btf(void)
+{
+	if (test_thread_flag(TIF_DEBUGCTLMSR))
+		wrmsr(MSR_IA32_DEBUGCTLMSR, current->thread.debugctlmsr, 0);
+}
+
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
+	clear_btf();
+	regs->flags |= TF_MASK;
+	regs->flags &= ~IF_MASK;
+	/*single step inline if the instruction is an int3*/
+	if (p->opcode == BREAKPOINT_INSTRUCTION)
+		regs->ip = (unsigned long)p->addr;
+	else
+		regs->ip = (unsigned long)p->ainsn.insn;
+}
+
+/* Called with kretprobe_lock held */
+void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
+				      struct pt_regs *regs)
+{
+	unsigned long *sara = stack_addr(regs);
+
+	ri->ret_addr = (kprobe_opcode_t *) *sara;
+
+	/* Replace the return addr with trampoline addr */
+	*sara = (unsigned long) &kretprobe_trampoline;
+}
+
+/*
+ * Interrupts are disabled on entry as trap3 is an interrupt gate and they
+ * remain disabled thorough out this function.
+ */
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	int ret = 0;
+	kprobe_opcode_t *addr;
+	struct kprobe_ctlblk *kcb;
+
+	addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
+
+	/*
+	 * We don't want to be preempted for the entire
+	 * duration of kprobe processing
+	 */
+	preempt_disable();
+	kcb = get_kprobe_ctlblk();
+
+	/* Check we're not actually recursing */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kcb->kprobe_status == KPROBE_HIT_SS &&
+				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
+				regs->flags &= ~TF_MASK;
+				regs->flags |= kcb->kprobe_saved_flags;
+				goto no_kprobe;
+#ifdef CONFIG_X86_64
+			} else if (kcb->kprobe_status == KPROBE_HIT_SSDONE) {
+				/* TODO: Provide re-entrancy from
+				 * post_kprobes_handler() and avoid exception
+				 * stack corruption while single-stepping on
+				 * the instruction of the new probe.
+				 */
+				arch_disarm_kprobe(p);
+				regs->ip = (unsigned long)p->addr;
+				reset_current_kprobe();
+				return 1;
+#endif
+			}
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe(kcb);
+			set_current_kprobe(p, regs, kcb);
+			kprobes_inc_nmissed_count(p);
+			prepare_singlestep(p, regs);
+			kcb->kprobe_status = KPROBE_REENTER;
+			return 1;
+		} else {
+			if (*addr != BREAKPOINT_INSTRUCTION) {
+			/* The breakpoint instruction was removed by
+			 * another cpu right after we hit, no further
+			 * handling of this interrupt is appropriate
+			 */
+				regs->ip = (unsigned long)addr;
+				ret = 1;
+				goto no_kprobe;
+			}
+			p = __get_cpu_var(current_kprobe);
+			if (p->break_handler && p->break_handler(p, regs))
+				goto ss_probe;
+		}
+		goto no_kprobe;
+	}
+
+	p = get_kprobe(addr);
+	if (!p) {
+		if (*addr != BREAKPOINT_INSTRUCTION) {
+			/*
+			 * The breakpoint instruction was removed right
+			 * after we hit it.  Another cpu has removed
+			 * either a probepoint or a debugger breakpoint
+			 * at this address.  In either case, no further
+			 * handling of this interrupt is appropriate.
+			 * Back up over the (now missing) int3 and run
+			 * the original instruction.
+			 */
+			regs->ip = (unsigned long)addr;
+			ret = 1;
+		}
+		/* Not one of ours: let kernel handle it */
+		goto no_kprobe;
+	}
+
+	set_current_kprobe(p, regs, kcb);
+	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+
+	if (p->pre_handler && p->pre_handler(p, regs))
+		/* handler has already set things up, so skip ss setup */
+		return 1;
+
+ss_probe:
+#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
+	if (p->ainsn.boostable == 1 && !p->post_handler) {
+		/* Boost up -- we can execute copied instructions directly */
+		reset_current_kprobe();
+		regs->ip = (unsigned long)p->ainsn.insn;
+		preempt_enable_no_resched();
+		return 1;
+	}
+#endif
+	prepare_singlestep(p, regs);
+	kcb->kprobe_status = KPROBE_HIT_SS;
+	return 1;
+
+no_kprobe:
+	preempt_enable_no_resched();
+	return ret;
+}
+
+/*
+ * When a retprobed function returns, this code saves registers and
+ * calls trampoline_handler() runs, which calls the kretprobe's handler.
+ */
+ void __kprobes kretprobe_trampoline_holder(void)
+ {
+	asm volatile (
+			".global kretprobe_trampoline\n"
+			"kretprobe_trampoline: \n"
+#ifdef CONFIG_X86_64
+			/* We don't bother saving the ss register */
+			"	pushq %rsp\n"
+			"	pushfq\n"
+			/*
+			 * Skip cs, ip, orig_ax.
+			 * trampoline_handler() will plug in these values
+			 */
+			"	subq $24, %rsp\n"
+			"	pushq %rdi\n"
+			"	pushq %rsi\n"
+			"	pushq %rdx\n"
+			"	pushq %rcx\n"
+			"	pushq %rax\n"
+			"	pushq %r8\n"
+			"	pushq %r9\n"
+			"	pushq %r10\n"
+			"	pushq %r11\n"
+			"	pushq %rbx\n"
+			"	pushq %rbp\n"
+			"	pushq %r12\n"
+			"	pushq %r13\n"
+			"	pushq %r14\n"
+			"	pushq %r15\n"
+			"	movq %rsp, %rdi\n"
+			"	call trampoline_handler\n"
+			/* Replace saved sp with true return address. */
+			"	movq %rax, 152(%rsp)\n"
+			"	popq %r15\n"
+			"	popq %r14\n"
+			"	popq %r13\n"
+			"	popq %r12\n"
+			"	popq %rbp\n"
+			"	popq %rbx\n"
+			"	popq %r11\n"
+			"	popq %r10\n"
+			"	popq %r9\n"
+			"	popq %r8\n"
+			"	popq %rax\n"
+			"	popq %rcx\n"
+			"	popq %rdx\n"
+			"	popq %rsi\n"
+			"	popq %rdi\n"
+			/* Skip orig_ax, ip, cs */
+			"	addq $24, %rsp\n"
+			"	popfq\n"
+#else
+			"	pushf\n"
+			/*
+			 * Skip cs, ip, orig_ax.
+			 * trampoline_handler() will plug in these values
+			 */
+			"	subl $12, %esp\n"
+			"	pushl %fs\n"
+			"	pushl %ds\n"
+			"	pushl %es\n"
+			"	pushl %eax\n"
+			"	pushl %ebp\n"
+			"	pushl %edi\n"
+			"	pushl %esi\n"
+			"	pushl %edx\n"
+			"	pushl %ecx\n"
+			"	pushl %ebx\n"
+			"	movl %esp, %eax\n"
+			"	call trampoline_handler\n"
+			/* Move flags to cs */
+			"	movl 52(%esp), %edx\n"
+			"	movl %edx, 48(%esp)\n"
+			/* Replace saved flags with true return address. */
+			"	movl %eax, 52(%esp)\n"
+			"	popl %ebx\n"
+			"	popl %ecx\n"
+			"	popl %edx\n"
+			"	popl %esi\n"
+			"	popl %edi\n"
+			"	popl %ebp\n"
+			"	popl %eax\n"
+			/* Skip ip, orig_ax, es, ds, fs */
+			"	addl $20, %esp\n"
+			"	popf\n"
+#endif
+			"	ret\n");
+ }
+
+/*
+ * Called from kretprobe_trampoline
+ */
+void * __kprobes trampoline_handler(struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head, empty_rp;
+	struct hlist_node *node, *tmp;
+	unsigned long flags, orig_ret_address = 0;
+	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
+
+	INIT_HLIST_HEAD(&empty_rp);
+	spin_lock_irqsave(&kretprobe_lock, flags);
+	head = kretprobe_inst_table_head(current);
+	/* fixup registers */
+#ifdef CONFIG_X86_64
+	regs->cs = __KERNEL_CS;
+#else
+	regs->cs = __KERNEL_CS | get_kernel_rpl();
+#endif
+	regs->ip = trampoline_address;
+	regs->orig_ax = ~0UL;
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because multiple functions in the call path have
+	 * return probes installed on them, and/or more then one
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always pushed into the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *	 function, the (chronologically) first instance's ret_addr
+	 *	 will be the real return address, and all the rest will
+	 *	 point to kretprobe_trampoline.
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		if (ri->rp && ri->rp->handler) {
+			__get_cpu_var(current_kprobe) = &ri->rp->kp;
+			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
+			ri->rp->handler(ri, regs);
+			__get_cpu_var(current_kprobe) = NULL;
+		}
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri, &empty_rp);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	kretprobe_assert(ri, orig_ret_address, trampoline_address);
+
+	spin_unlock_irqrestore(&kretprobe_lock, flags);
+
+	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
+		hlist_del(&ri->hlist);
+		kfree(ri);
+	}
+	return (void *)orig_ret_address;
+}
+
+/*
+ * Called after single-stepping.  p->addr is the address of the
+ * instruction whose first byte has been replaced by the "int 3"
+ * instruction.  To avoid the SMP problems that can occur when we
+ * temporarily put back the original opcode to single-step, we
+ * single-stepped a copy of the instruction.  The address of this
+ * copy is p->ainsn.insn.
+ *
+ * This function prepares to return from the post-single-step
+ * interrupt.  We have to fix up the stack as follows:
+ *
+ * 0) Except in the case of absolute or indirect jump or call instructions,
+ * the new ip is relative to the copied instruction.  We need to make
+ * it relative to the original instruction.
+ *
+ * 1) If the single-stepped instruction was pushfl, then the TF and IF
+ * flags are set in the just-pushed flags, and may need to be cleared.
+ *
+ * 2) If the single-stepped instruction was a call, the return address
+ * that is atop the stack is the address following the copied instruction.
+ * We need to make it the address following the original instruction.
+ *
+ * If this is the first time we've single-stepped the instruction at
+ * this probepoint, and the instruction is boostable, boost it: add a
+ * jump instruction after the copied instruction, that jumps to the next
+ * instruction after the probepoint.
+ */
+static void __kprobes resume_execution(struct kprobe *p,
+		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
+{
+	unsigned long *tos = stack_addr(regs);
+	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
+	unsigned long orig_ip = (unsigned long)p->addr;
+	kprobe_opcode_t *insn = p->ainsn.insn;
+
+#ifdef CONFIG_X86_64
+	/*skip the REX prefix*/
+	if (*insn >= 0x40 && *insn <= 0x4f)
+		insn++;
+#endif
+
+	regs->flags &= ~TF_MASK;
+	switch (*insn) {
+	case 0x9c:	/* pushfl */
+		*tos &= ~(TF_MASK | IF_MASK);
+		*tos |= kcb->kprobe_old_flags;
+		break;
+	case 0xc2:	/* iret/ret/lret */
+	case 0xc3:
+	case 0xca:
+	case 0xcb:
+	case 0xcf:
+	case 0xea:	/* jmp absolute -- ip is correct */
+		/* ip is already adjusted, no more changes required */
+		p->ainsn.boostable = 1;
+		goto no_change;
+	case 0xe8:	/* call relative - Fix return addr */
+		*tos = orig_ip + (*tos - copy_ip);
+		break;
+#ifndef CONFIG_X86_64
+	case 0x9a:	/* call absolute -- same as call absolute, indirect */
+		*tos = orig_ip + (*tos - copy_ip);
+		goto no_change;
+#endif
+	case 0xff:
+		if ((insn[1] & 0x30) == 0x10) {
+			/*
+			 * call absolute, indirect
+			 * Fix return addr; ip is correct.
+			 * But this is not boostable
+			 */
+			*tos = orig_ip + (*tos - copy_ip);
+			goto no_change;
+		} else if (((insn[1] & 0x31) == 0x20) ||
+			   ((insn[1] & 0x31) == 0x21)) {
+			/*
+			 * jmp near and far, absolute indirect
+			 * ip is correct. And this is boostable
+			 */
+			p->ainsn.boostable = 1;
+			goto no_change;
+		}
+	default:
+		break;
+	}
+
+	if (p->ainsn.boostable == 0) {
+		if ((regs->ip > copy_ip) &&
+		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
+			/*
+			 * These instructions can be executed directly if it
+			 * jumps back to correct address.
+			 */
+			set_jmp_op((void *)regs->ip,
+				   (void *)orig_ip + (regs->ip - copy_ip));
+			p->ainsn.boostable = 1;
+		} else {
+			p->ainsn.boostable = -1;
+		}
+	}
+
+	regs->ip += orig_ip - copy_ip;
+
+no_change:
+	restore_btf();
+
+	return;
+}
+
+/*
+ * Interrupts are disabled on entry as trap1 is an interrupt gate and they
+ * remain disabled thoroughout this function.
+ */
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	if (!cur)
+		return 0;
+
+	if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
+		kcb->kprobe_status = KPROBE_HIT_SSDONE;
+		cur->post_handler(cur, regs, 0);
+	}
+
+	resume_execution(cur, regs, kcb);
+	regs->flags |= kcb->kprobe_saved_flags;
+	trace_hardirqs_fixup_flags(regs->flags);
+
+	/* Restore back the original saved kprobes variables and continue. */
+	if (kcb->kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe(kcb);
+		goto out;
+	}
+	reset_current_kprobe();
+out:
+	preempt_enable_no_resched();
+
+	/*
+	 * if somebody else is singlestepping across a probe point, flags
+	 * will have TF set, in which case, continue the remaining processing
+	 * of do_debug, as if this is not a probe hit.
+	 */
+	if (regs->flags & TF_MASK)
+		return 0;
+
+	return 1;
+}
+
+int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	switch (kcb->kprobe_status) {
+	case KPROBE_HIT_SS:
+	case KPROBE_REENTER:
+		/*
+		 * We are here because the instruction being single
+		 * stepped caused a page fault. We reset the current
+		 * kprobe and the ip points back to the probe address
+		 * and allow the page fault handler to continue as a
+		 * normal page fault.
+		 */
+		regs->ip = (unsigned long)cur->addr;
+		regs->flags |= kcb->kprobe_old_flags;
+		if (kcb->kprobe_status == KPROBE_REENTER)
+			restore_previous_kprobe(kcb);
+		else
+			reset_current_kprobe();
+		preempt_enable_no_resched();
+		break;
+	case KPROBE_HIT_ACTIVE:
+	case KPROBE_HIT_SSDONE:
+		/*
+		 * We increment the nmissed count for accounting,
+		 * we can also use npre/npostfault count for accounting
+		 * these specific fault cases.
+		 */
+		kprobes_inc_nmissed_count(cur);
+
+		/*
+		 * We come here because instructions in the pre/post
+		 * handler caused the page_fault, this could happen
+		 * if handler tries to access user space by
+		 * copy_from_user(), get_user() etc. Let the
+		 * user-specified handler try to fix it first.
+		 */
+		if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+			return 1;
+
+		/*
+		 * In case the user-specified fault handler returned
+		 * zero, try to fix up.
+		 */
+#ifdef CONFIG_X86_64
+		{
+			const struct exception_table_entry *fixup;
+			fixup = search_exception_tables(regs->ip);
+			if (fixup) {
+				regs->ip = fixup->fixup;
+				return 1;
+			}
+		}
+#else
+		if (fixup_exception(regs))
+			return 1;
+#endif
+		/*
+		 * fixup routine could not handle it,
+		 * Let do_page_fault() fix it.
+		 */
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+/*
+ * Wrapper routine for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	if (args->regs && user_mode_vm(args->regs))
+		return ret;
+
+	switch (val) {
+	case DIE_INT3:
+		if (kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_DEBUG:
+		if (post_kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_GPF:
+		/* kprobe_running() needs smp_processor_id() */
+		preempt_disable();
+		if (kprobe_running() &&
+		    kprobe_fault_handler(args->regs, args->trapnr))
+			ret = NOTIFY_STOP;
+		preempt_enable();
+		break;
+	default:
+		break;
+	}
+	return ret;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+	unsigned long addr;
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	kcb->jprobe_saved_regs = *regs;
+	kcb->jprobe_saved_sp = stack_addr(regs);
+	addr = (unsigned long)(kcb->jprobe_saved_sp);
+
+	/*
+	 * As Linus pointed out, gcc assumes that the callee
+	 * owns the argument space and could overwrite it, e.g.
+	 * tailcall optimization. So, to be absolutely safe
+	 * we also save and restore enough stack bytes to cover
+	 * the argument area.
+	 */
+	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
+	       MIN_STACK_SIZE(addr));
+	regs->flags &= ~IF_MASK;
+	trace_hardirqs_off();
+	regs->ip = (unsigned long)(jp->entry);
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	asm volatile (
+#ifdef CONFIG_X86_64
+			"       xchg   %%rbx,%%rsp	\n"
+#else
+			"       xchgl   %%ebx,%%esp	\n"
+#endif
+			"       int3			\n"
+			"       .globl jprobe_return_end\n"
+			"       jprobe_return_end:	\n"
+			"       nop			\n"::"b"
+			(kcb->jprobe_saved_sp):"memory");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	u8 *addr = (u8 *) (regs->ip - 1);
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+
+	if ((addr > (u8 *) jprobe_return) &&
+	    (addr < (u8 *) jprobe_return_end)) {
+		if (stack_addr(regs) != kcb->jprobe_saved_sp) {
+			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
+			printk(KERN_ERR
+			       "current sp %p does not match saved sp %p\n",
+			       stack_addr(regs), kcb->jprobe_saved_sp);
+			printk(KERN_ERR "Saved registers for jprobe %p\n", jp);
+			show_registers(saved_regs);
+			printk(KERN_ERR "Current registers\n");
+			show_registers(regs);
+			BUG();
+		}
+		*regs = kcb->jprobe_saved_regs;
+		memcpy((kprobe_opcode_t *)(kcb->jprobe_saved_sp),
+		       kcb->jprobes_stack,
+		       MIN_STACK_SIZE(kcb->jprobe_saved_sp));
+		preempt_enable_no_resched();
+		return 1;
+	}
+	return 0;
+}
+
+int __init arch_init_kprobes(void)
+{
+	return 0;
+}
+
+int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+{
+	return 0;
+}
diff --git a/arch/x86/kernel/kprobes_32.c b/arch/x86/kernel/kprobes_32.c
deleted file mode 100644
index 8e06431d8b03..000000000000
--- a/arch/x86/kernel/kprobes_32.c
+++ /dev/null
@@ -1,848 +0,0 @@
-/*
- *  Kernel Probes (KProbes)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
- *
- * Copyright (C) IBM Corporation, 2002, 2004
- *
- * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
- *		Probes initial implementation ( includes contributions from
- *		Rusty Russell).
- * 2004-July	Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
- *		interface to access function arguments.
- * 2005-May	Hien Nguyen <hien@us.ibm.com>, Jim Keniston
- *		<jkenisto@us.ibm.com> and Prasanna S Panchamukhi
- *		<prasanna@in.ibm.com> added function-return probes.
- */
-
-#include <linux/kprobes.h>
-#include <linux/ptrace.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/preempt.h>
-#include <linux/module.h>
-#include <linux/kdebug.h>
-
-#include <asm/cacheflush.h>
-#include <asm/desc.h>
-#include <asm/pgtable.h>
-#include <asm/uaccess.h>
-#include <asm/alternative.h>
-
-void jprobe_return_end(void);
-
-DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
-DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
-
-/*
- * "&regs->sp" looks wrong, but it's correct for x86_32.  x86_32 CPUs
- * don't save the ss and esp registers if the CPU is already in kernel
- * mode when it traps.  So for kprobes, regs->sp and regs->ss are not
- * the [nonexistent] saved stack pointer and ss register, but rather
- * the top 8 bytes of the pre-int3 stack.  So &regs->sp happens to
- * point to the top of the pre-int3 stack.
- */
-#define stack_addr(regs) ((unsigned long *)&regs->sp)
-
-#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
-	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
-	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
-	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
-	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
-	 << (row % 32))
-	/*
-	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
-	 * Groups, and some special opcodes can not boost.
-	 */
-static const u32 twobyte_is_boostable[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      ----------------------------------------------          */
-	W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
-	W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 10 */
-	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 20 */
-	W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 30 */
-	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 40 */
-	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
-	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1) | /* 60 */
-	W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) , /* 70 */
-	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 80 */
-	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 90 */
-	W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* a0 */
-	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) , /* b0 */
-	W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* c0 */
-	W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) , /* d0 */
-	W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* e0 */
-	W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0)   /* f0 */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-static const u32 onebyte_has_modrm[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      -----------------------------------------------         */
-	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
-	W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
-	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
-	W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
-	W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
-	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
-	W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
-	W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
-	W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
-	W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
-	W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
-	W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
-	W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
-	W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
-	W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
-	W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1)   /* f0 */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-static const u32 twobyte_has_modrm[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      -----------------------------------------------         */
-	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
-	W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
-	W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
-	W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
-	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
-	W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
-	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
-	W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
-	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
-	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
-	W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
-	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
-	W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
-	W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
-	W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
-	W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0)   /* ff */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-#undef W
-
-struct kretprobe_blackpoint kretprobe_blacklist[] = {
-	{"__switch_to", }, /* This function switches only current task, but
-			      doesn't switch kernel stack.*/
-	{NULL, NULL}	/* Terminator */
-};
-const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist);
-
-/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
-static __always_inline void set_jmp_op(void *from, void *to)
-{
-	struct __arch_jmp_op {
-		char op;
-		s32 raddr;
-	} __attribute__((packed)) * jop;
-	jop = (struct __arch_jmp_op *)from;
-	jop->raddr = (s32)((long)(to) - ((long)(from) + 5));
-	jop->op = RELATIVEJUMP_INSTRUCTION;
-}
-
-/*
- * returns non-zero if opcode is boostable.
- */
-static __always_inline int can_boost(kprobe_opcode_t *opcodes)
-{
-	kprobe_opcode_t opcode;
-	kprobe_opcode_t *orig_opcodes = opcodes;
-
-retry:
-	if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
-		return 0;
-	opcode = *(opcodes++);
-
-	/* 2nd-byte opcode */
-	if (opcode == 0x0f) {
-		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
-			return 0;
-		return test_bit(*opcodes,
-				(unsigned long *)twobyte_is_boostable);
-	}
-
-	switch (opcode & 0xf0) {
-	case 0x60:
-		if (0x63 < opcode && opcode < 0x67)
-			goto retry; /* prefixes */
-		/* can't boost Address-size override and bound */
-		return (opcode != 0x62 && opcode != 0x67);
-	case 0x70:
-		return 0; /* can't boost conditional jump */
-	case 0xc0:
-		/* can't boost software-interruptions */
-		return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
-	case 0xd0:
-		/* can boost AA* and XLAT */
-		return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7);
-	case 0xe0:
-		/* can boost in/out and absolute jmps */
-		return ((opcode & 0x04) || opcode == 0xea);
-	case 0xf0:
-		if ((opcode & 0x0c) == 0 && opcode != 0xf1)
-			goto retry; /* lock/rep(ne) prefix */
-		/* clear and set flags are boostable */
-		return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
-	default:
-		/* segment override prefixes are boostable */
-		if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
-			goto retry; /* prefixes */
-		/* CS override prefix and call are not boostable */
-		return (opcode != 0x2e && opcode != 0x9a);
-	}
-}
-
-/*
- * returns non-zero if opcode modifies the interrupt flag.
- */
-static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
-{
-	switch (*insn) {
-	case 0xfa:		/* cli */
-	case 0xfb:		/* sti */
-	case 0xcf:		/* iret/iretd */
-	case 0x9d:		/* popf/popfd */
-		return 1;
-	}
-	return 0;
-}
-
-static void __kprobes arch_copy_kprobe(struct kprobe *p)
-{
-	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
-	if (can_boost(p->addr))
-		p->ainsn.boostable = 0;
-	else
-		p->ainsn.boostable = -1;
-
-	p->opcode = *p->addr;
-}
-
-int __kprobes arch_prepare_kprobe(struct kprobe *p)
-{
-	/* insn: must be on special executable page on x86. */
-	p->ainsn.insn = get_insn_slot();
-	if (!p->ainsn.insn)
-		return -ENOMEM;
-	arch_copy_kprobe(p);
-	return 0;
-}
-
-void __kprobes arch_arm_kprobe(struct kprobe *p)
-{
-	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
-}
-
-void __kprobes arch_disarm_kprobe(struct kprobe *p)
-{
-	text_poke(p->addr, &p->opcode, 1);
-}
-
-void __kprobes arch_remove_kprobe(struct kprobe *p)
-{
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
-	mutex_unlock(&kprobe_mutex);
-}
-
-static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
-{
-	kcb->prev_kprobe.kp = kprobe_running();
-	kcb->prev_kprobe.status = kcb->kprobe_status;
-	kcb->prev_kprobe.old_flags = kcb->kprobe_old_flags;
-	kcb->prev_kprobe.saved_flags = kcb->kprobe_saved_flags;
-}
-
-static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
-{
-	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
-	kcb->kprobe_status = kcb->prev_kprobe.status;
-	kcb->kprobe_old_flags = kcb->prev_kprobe.old_flags;
-	kcb->kprobe_saved_flags = kcb->prev_kprobe.saved_flags;
-}
-
-static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
-				struct kprobe_ctlblk *kcb)
-{
-	__get_cpu_var(current_kprobe) = p;
-	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
-		= (regs->flags & (TF_MASK | IF_MASK));
-	if (is_IF_modifier(p->ainsn.insn))
-		kcb->kprobe_saved_flags &= ~IF_MASK;
-}
-
-static __always_inline void clear_btf(void)
-{
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsr(MSR_IA32_DEBUGCTLMSR, 0, 0);
-}
-
-static __always_inline void restore_btf(void)
-{
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsr(MSR_IA32_DEBUGCTLMSR, current->thread.debugctlmsr, 0);
-}
-
-static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
-{
-	clear_btf();
-	regs->flags |= TF_MASK;
-	regs->flags &= ~IF_MASK;
-	/*single step inline if the instruction is an int3*/
-	if (p->opcode == BREAKPOINT_INSTRUCTION)
-		regs->ip = (unsigned long)p->addr;
-	else
-		regs->ip = (unsigned long)p->ainsn.insn;
-}
-
-/* Called with kretprobe_lock held */
-void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
-				      struct pt_regs *regs)
-{
-	unsigned long *sara = stack_addr(regs);
-
-	ri->ret_addr = (kprobe_opcode_t *) *sara;
-
-	/* Replace the return addr with trampoline addr */
-	*sara = (unsigned long) &kretprobe_trampoline;
-}
-
-/*
- * Interrupts are disabled on entry as trap3 is an interrupt gate and they
- * remain disabled thorough out this function.
- */
-static int __kprobes kprobe_handler(struct pt_regs *regs)
-{
-	struct kprobe *p;
-	int ret = 0;
-	kprobe_opcode_t *addr;
-	struct kprobe_ctlblk *kcb;
-
-	addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
-
-	/*
-	 * We don't want to be preempted for the entire
-	 * duration of kprobe processing
-	 */
-	preempt_disable();
-	kcb = get_kprobe_ctlblk();
-
-	/* Check we're not actually recursing */
-	if (kprobe_running()) {
-		p = get_kprobe(addr);
-		if (p) {
-			if (kcb->kprobe_status == KPROBE_HIT_SS &&
-				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
-				regs->flags &= ~TF_MASK;
-				regs->flags |= kcb->kprobe_saved_flags;
-				goto no_kprobe;
-			}
-			/* We have reentered the kprobe_handler(), since
-			 * another probe was hit while within the handler.
-			 * We here save the original kprobes variables and
-			 * just single step on the instruction of the new probe
-			 * without calling any user handlers.
-			 */
-			save_previous_kprobe(kcb);
-			set_current_kprobe(p, regs, kcb);
-			kprobes_inc_nmissed_count(p);
-			prepare_singlestep(p, regs);
-			kcb->kprobe_status = KPROBE_REENTER;
-			return 1;
-		} else {
-			if (*addr != BREAKPOINT_INSTRUCTION) {
-			/* The breakpoint instruction was removed by
-			 * another cpu right after we hit, no further
-			 * handling of this interrupt is appropriate
-			 */
-				regs->ip = (unsigned long)addr;
-				ret = 1;
-				goto no_kprobe;
-			}
-			p = __get_cpu_var(current_kprobe);
-			if (p->break_handler && p->break_handler(p, regs)) {
-				goto ss_probe;
-			}
-		}
-		goto no_kprobe;
-	}
-
-	p = get_kprobe(addr);
-	if (!p) {
-		if (*addr != BREAKPOINT_INSTRUCTION) {
-			/*
-			 * The breakpoint instruction was removed right
-			 * after we hit it.  Another cpu has removed
-			 * either a probepoint or a debugger breakpoint
-			 * at this address.  In either case, no further
-			 * handling of this interrupt is appropriate.
-			 * Back up over the (now missing) int3 and run
-			 * the original instruction.
-			 */
-			regs->ip = (unsigned long)addr;
-			ret = 1;
-		}
-		/* Not one of ours: let kernel handle it */
-		goto no_kprobe;
-	}
-
-	set_current_kprobe(p, regs, kcb);
-	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
-
-	if (p->pre_handler && p->pre_handler(p, regs))
-		/* handler has already set things up, so skip ss setup */
-		return 1;
-
-ss_probe:
-#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
-	if (p->ainsn.boostable == 1 && !p->post_handler) {
-		/* Boost up -- we can execute copied instructions directly */
-		reset_current_kprobe();
-		regs->ip = (unsigned long)p->ainsn.insn;
-		preempt_enable_no_resched();
-		return 1;
-	}
-#endif
-	prepare_singlestep(p, regs);
-	kcb->kprobe_status = KPROBE_HIT_SS;
-	return 1;
-
-no_kprobe:
-	preempt_enable_no_resched();
-	return ret;
-}
-
-/*
- * When a retprobed function returns, this code saves registers and
- * calls trampoline_handler() runs, which calls the kretprobe's handler.
- */
- void __kprobes kretprobe_trampoline_holder(void)
- {
-	asm volatile ( ".global kretprobe_trampoline\n"
-			"kretprobe_trampoline: \n"
-			"	pushf\n"
-			/*
-			 * Skip cs, ip, orig_ax.
-			 * trampoline_handler() will plug in these values
-			 */
-			"	subl $12, %esp\n"
-			"	pushl %fs\n"
-			"	pushl %ds\n"
-			"	pushl %es\n"
-			"	pushl %eax\n"
-			"	pushl %ebp\n"
-			"	pushl %edi\n"
-			"	pushl %esi\n"
-			"	pushl %edx\n"
-			"	pushl %ecx\n"
-			"	pushl %ebx\n"
-			"	movl %esp, %eax\n"
-			"	call trampoline_handler\n"
-			/* Move flags to cs */
-			"	movl 52(%esp), %edx\n"
-			"	movl %edx, 48(%esp)\n"
-			/* Replace saved flags with true return address. */
-			"	movl %eax, 52(%esp)\n"
-			"	popl %ebx\n"
-			"	popl %ecx\n"
-			"	popl %edx\n"
-			"	popl %esi\n"
-			"	popl %edi\n"
-			"	popl %ebp\n"
-			"	popl %eax\n"
-			/* Skip ip, orig_ax, es, ds, fs */
-			"	addl $20, %esp\n"
-			"	popf\n"
-			"	ret\n");
- }
-
-/*
- * Called from kretprobe_trampoline
- */
-void * __kprobes trampoline_handler(struct pt_regs *regs)
-{
-	struct kretprobe_instance *ri = NULL;
-	struct hlist_head *head, empty_rp;
-	struct hlist_node *node, *tmp;
-	unsigned long flags, orig_ret_address = 0;
-	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
-
-	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
-	/* fixup registers */
-	regs->cs = __KERNEL_CS | get_kernel_rpl();
-	regs->ip = trampoline_address;
-	regs->orig_ax = ~0UL;
-
-	/*
-	 * It is possible to have multiple instances associated with a given
-	 * task either because multiple functions in the call path have
-	 * return probes installed on them, and/or more then one
-	 * return probe was registered for a target function.
-	 *
-	 * We can handle this because:
-	 *     - instances are always pushed into the head of the list
-	 *     - when multiple return probes are registered for the same
-	 *	 function, the (chronologically) first instance's ret_addr
-	 *	 will be the real return address, and all the rest will
-	 *	 point to kretprobe_trampoline.
-	 */
-	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
-		if (ri->task != current)
-			/* another task is sharing our hash bucket */
-			continue;
-
-		if (ri->rp && ri->rp->handler) {
-			__get_cpu_var(current_kprobe) = &ri->rp->kp;
-			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
-			ri->rp->handler(ri, regs);
-			__get_cpu_var(current_kprobe) = NULL;
-		}
-
-		orig_ret_address = (unsigned long)ri->ret_addr;
-		recycle_rp_inst(ri, &empty_rp);
-
-		if (orig_ret_address != trampoline_address)
-			/*
-			 * This is the real return address. Any other
-			 * instances associated with this task are for
-			 * other calls deeper on the call stack
-			 */
-			break;
-	}
-
-	kretprobe_assert(ri, orig_ret_address, trampoline_address);
-
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
-
-	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
-		hlist_del(&ri->hlist);
-		kfree(ri);
-	}
-	return (void *)orig_ret_address;
-}
-
-/*
- * Called after single-stepping.  p->addr is the address of the
- * instruction whose first byte has been replaced by the "int 3"
- * instruction.  To avoid the SMP problems that can occur when we
- * temporarily put back the original opcode to single-step, we
- * single-stepped a copy of the instruction.  The address of this
- * copy is p->ainsn.insn.
- *
- * This function prepares to return from the post-single-step
- * interrupt.  We have to fix up the stack as follows:
- *
- * 0) Except in the case of absolute or indirect jump or call instructions,
- * the new ip is relative to the copied instruction.  We need to make
- * it relative to the original instruction.
- *
- * 1) If the single-stepped instruction was pushfl, then the TF and IF
- * flags are set in the just-pushed flags, and may need to be cleared.
- *
- * 2) If the single-stepped instruction was a call, the return address
- * that is atop the stack is the address following the copied instruction.
- * We need to make it the address following the original instruction.
- *
- * If this is the first time we've single-stepped the instruction at
- * this probepoint, and the instruction is boostable, boost it: add a
- * jump instruction after the copied instruction, that jumps to the next
- * instruction after the probepoint.
- */
-static void __kprobes resume_execution(struct kprobe *p,
-		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
-{
-	unsigned long *tos = stack_addr(regs);
-	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
-	unsigned long orig_ip = (unsigned long)p->addr;
-	kprobe_opcode_t *insn = p->ainsn.insn;
-
-	regs->flags &= ~TF_MASK;
-	switch (*insn) {
-	case 0x9c:	/* pushfl */
-		*tos &= ~(TF_MASK | IF_MASK);
-		*tos |= kcb->kprobe_old_flags;
-		break;
-	case 0xc2:	/* iret/ret/lret */
-	case 0xc3:
-	case 0xca:
-	case 0xcb:
-	case 0xcf:
-	case 0xea:	/* jmp absolute -- ip is correct */
-		/* ip is already adjusted, no more changes required */
-		p->ainsn.boostable = 1;
-		goto no_change;
-	case 0xe8:	/* call relative - Fix return addr */
-		*tos = orig_ip + (*tos - copy_ip);
-		break;
-	case 0x9a:	/* call absolute -- same as call absolute, indirect */
-		*tos = orig_ip + (*tos - copy_ip);
-		goto no_change;
-	case 0xff:
-		if ((insn[1] & 0x30) == 0x10) {
-			/*
-			 * call absolute, indirect
-			 * Fix return addr; ip is correct.
-			 * But this is not boostable
-			 */
-			*tos = orig_ip + (*tos - copy_ip);
-			goto no_change;
-		} else if (((insn[1] & 0x31) == 0x20) ||
-			   ((insn[1] & 0x31) == 0x21)) {
-			/*
-			 * jmp near and far, absolute indirect
-			 * ip is correct. And this is boostable
-			 */
-			p->ainsn.boostable = 1;
-			goto no_change;
-		}
-	default:
-		break;
-	}
-
-	if (p->ainsn.boostable == 0) {
-		if ((regs->ip > copy_ip) &&
-		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
-			/*
-			 * These instructions can be executed directly if it
-			 * jumps back to correct address.
-			 */
-			set_jmp_op((void *)regs->ip,
-				   (void *)orig_ip + (regs->ip - copy_ip));
-			p->ainsn.boostable = 1;
-		} else {
-			p->ainsn.boostable = -1;
-		}
-	}
-
-	regs->ip += orig_ip - copy_ip;
-
-no_change:
-	restore_btf();
-
-	return;
-}
-
-/*
- * Interrupts are disabled on entry as trap1 is an interrupt gate and they
- * remain disabled thoroughout this function.
- */
-static int __kprobes post_kprobe_handler(struct pt_regs *regs)
-{
-	struct kprobe *cur = kprobe_running();
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	if (!cur)
-		return 0;
-
-	if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
-		kcb->kprobe_status = KPROBE_HIT_SSDONE;
-		cur->post_handler(cur, regs, 0);
-	}
-
-	resume_execution(cur, regs, kcb);
-	regs->flags |= kcb->kprobe_saved_flags;
-	trace_hardirqs_fixup_flags(regs->flags);
-
-	/* Restore back the original saved kprobes variables and continue. */
-	if (kcb->kprobe_status == KPROBE_REENTER) {
-		restore_previous_kprobe(kcb);
-		goto out;
-	}
-	reset_current_kprobe();
-out:
-	preempt_enable_no_resched();
-
-	/*
-	 * if somebody else is singlestepping across a probe point, flags
-	 * will have TF set, in which case, continue the remaining processing
-	 * of do_debug, as if this is not a probe hit.
-	 */
-	if (regs->flags & TF_MASK)
-		return 0;
-
-	return 1;
-}
-
-int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
-{
-	struct kprobe *cur = kprobe_running();
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	switch(kcb->kprobe_status) {
-	case KPROBE_HIT_SS:
-	case KPROBE_REENTER:
-		/*
-		 * We are here because the instruction being single
-		 * stepped caused a page fault. We reset the current
-		 * kprobe and the ip points back to the probe address
-		 * and allow the page fault handler to continue as a
-		 * normal page fault.
-		 */
-		regs->ip = (unsigned long)cur->addr;
-		regs->flags |= kcb->kprobe_old_flags;
-		if (kcb->kprobe_status == KPROBE_REENTER)
-			restore_previous_kprobe(kcb);
-		else
-			reset_current_kprobe();
-		preempt_enable_no_resched();
-		break;
-	case KPROBE_HIT_ACTIVE:
-	case KPROBE_HIT_SSDONE:
-		/*
-		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accounting
-		 * these specific fault cases.
-		 */
-		kprobes_inc_nmissed_count(cur);
-
-		/*
-		 * We come here because instructions in the pre/post
-		 * handler caused the page_fault, this could happen
-		 * if handler tries to access user space by
-		 * copy_from_user(), get_user() etc. Let the
-		 * user-specified handler try to fix it first.
-		 */
-		if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
-			return 1;
-
-		/*
-		 * In case the user-specified fault handler returned
-		 * zero, try to fix up.
-		 */
-		if (fixup_exception(regs))
-			return 1;
-
-		/*
-		 * fixup routine could not handle it,
-		 * Let do_page_fault() fix it.
-		 */
-		break;
-	default:
-		break;
-	}
-	return 0;
-}
-
-/*
- * Wrapper routine for handling exceptions.
- */
-int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
-				       unsigned long val, void *data)
-{
-	struct die_args *args = (struct die_args *)data;
-	int ret = NOTIFY_DONE;
-
-	if (args->regs && user_mode_vm(args->regs))
-		return ret;
-
-	switch (val) {
-	case DIE_INT3:
-		if (kprobe_handler(args->regs))
-			ret = NOTIFY_STOP;
-		break;
-	case DIE_DEBUG:
-		if (post_kprobe_handler(args->regs))
-			ret = NOTIFY_STOP;
-		break;
-	case DIE_GPF:
-		/* kprobe_running() needs smp_processor_id() */
-		preempt_disable();
-		if (kprobe_running() &&
-		    kprobe_fault_handler(args->regs, args->trapnr))
-			ret = NOTIFY_STOP;
-		preempt_enable();
-		break;
-	default:
-		break;
-	}
-	return ret;
-}
-
-int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
-{
-	struct jprobe *jp = container_of(p, struct jprobe, kp);
-	unsigned long addr;
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_sp = stack_addr(regs);
-	addr = (unsigned long)(kcb->jprobe_saved_sp);
-
-	/*
-	 * As Linus pointed out, gcc assumes that the callee
-	 * owns the argument space and could overwrite it, e.g.
-	 * tailcall optimization. So, to be absolutely safe
-	 * we also save and restore enough stack bytes to cover
-	 * the argument area.
-	 */
-	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
-			MIN_STACK_SIZE(addr));
-	regs->flags &= ~IF_MASK;
-	trace_hardirqs_off();
-	regs->ip = (unsigned long)(jp->entry);
-	return 1;
-}
-
-void __kprobes jprobe_return(void)
-{
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	asm volatile ("       xchgl   %%ebx,%%esp     \n"
-		      "       int3			\n"
-		      "       .globl jprobe_return_end	\n"
-		      "       jprobe_return_end:	\n"
-		      "       nop			\n"::"b"
-		      (kcb->jprobe_saved_sp):"memory");
-}
-
-int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
-{
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	u8 *addr = (u8 *) (regs->ip - 1);
-	struct jprobe *jp = container_of(p, struct jprobe, kp);
-
-	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if (stack_addr(regs) != kcb->jprobe_saved_sp) {
-			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
-			printk("current sp %p does not match saved sp %p\n",
-			       stack_addr(regs), kcb->jprobe_saved_sp);
-			printk("Saved registers for jprobe %p\n", jp);
-			show_registers(saved_regs);
-			printk("Current registers\n");
-			show_registers(regs);
-			BUG();
-		}
-		*regs = kcb->jprobe_saved_regs;
-		memcpy((kprobe_opcode_t *)(kcb->jprobe_saved_sp),
-		       kcb->jprobes_stack,
-		       MIN_STACK_SIZE(kcb->jprobe_saved_sp));
-		preempt_enable_no_resched();
-		return 1;
-	}
-	return 0;
-}
-
-int __init arch_init_kprobes(void)
-{
-	return 0;
-}
-
-int __kprobes arch_trampoline_kprobe(struct kprobe *p)
-{
-	return 0;
-}
diff --git a/arch/x86/kernel/kprobes_64.c b/arch/x86/kernel/kprobes_64.c
deleted file mode 100644
index 2d7763749b1b..000000000000
--- a/arch/x86/kernel/kprobes_64.c
+++ /dev/null
@@ -1,953 +0,0 @@
-/*
- *  Kernel Probes (KProbes)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
- *
- * Copyright (C) IBM Corporation, 2002, 2004
- *
- * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
- *		Probes initial implementation ( includes contributions from
- *		Rusty Russell).
- * 2004-July	Suparna Bhattacharya <suparna@in.ibm.com> added jumper probes
- *		interface to access function arguments.
- * 2004-Oct	Jim Keniston <kenistoj@us.ibm.com> and Prasanna S Panchamukhi
- *		<prasanna@in.ibm.com> adapted for x86_64
- * 2005-Mar	Roland McGrath <roland@redhat.com>
- *		Fixed to handle %rip-relative addressing mode correctly.
- * 2005-May     Rusty Lynch <rusty.lynch@intel.com>
- *              Added function return probes functionality
- * 2007-Dec	Masami Hiramatsu <mhiramat@redhat.com> added kprobe-booster
- * 		and kretprobe-booster for x86-64
- */
-
-#include <linux/kprobes.h>
-#include <linux/ptrace.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/preempt.h>
-#include <linux/module.h>
-#include <linux/kdebug.h>
-
-#include <asm/cacheflush.h>
-#include <asm/desc.h>
-#include <asm/pgtable.h>
-#include <asm/uaccess.h>
-#include <asm/alternative.h>
-
-void jprobe_return_end(void);
-
-DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
-DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
-
-#define stack_addr(regs) ((unsigned long *)regs->sp)
-
-#define W(row, b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, ba, bb, bc, bd, be, bf)\
-	(((b0##UL << 0x0)|(b1##UL << 0x1)|(b2##UL << 0x2)|(b3##UL << 0x3) |   \
-	  (b4##UL << 0x4)|(b5##UL << 0x5)|(b6##UL << 0x6)|(b7##UL << 0x7) |   \
-	  (b8##UL << 0x8)|(b9##UL << 0x9)|(ba##UL << 0xa)|(bb##UL << 0xb) |   \
-	  (bc##UL << 0xc)|(bd##UL << 0xd)|(be##UL << 0xe)|(bf##UL << 0xf))    \
-	 << (row % 32))
-	/*
-	 * Undefined/reserved opcodes, conditional jump, Opcode Extension
-	 * Groups, and some special opcodes can not boost.
-	 */
-static const u32 twobyte_is_boostable[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      ----------------------------------------------          */
-	W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
-	W(0x10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 10 */
-	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 20 */
-	W(0x30, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 30 */
-	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 40 */
-	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
-	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1) | /* 60 */
-	W(0x70, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) , /* 70 */
-	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 80 */
-	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 90 */
-	W(0xa0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* a0 */
-	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) , /* b0 */
-	W(0xc0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* c0 */
-	W(0xd0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) , /* d0 */
-	W(0xe0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1) | /* e0 */
-	W(0xf0, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0)   /* f0 */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-static const u32 onebyte_has_modrm[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      -----------------------------------------------         */
-	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
-	W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
-	W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
-	W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
-	W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
-	W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
-	W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
-	W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
-	W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
-	W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
-	W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
-	W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
-	W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
-	W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
-	W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
-	W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1)   /* f0 */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-static const u32 twobyte_has_modrm[256 / 32] = {
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-	/*      -----------------------------------------------         */
-	W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
-	W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
-	W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
-	W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
-	W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
-	W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
-	W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
-	W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
-	W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
-	W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
-	W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
-	W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
-	W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
-	W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
-	W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
-	W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0)   /* ff */
-	/*      -----------------------------------------------         */
-	/*      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f          */
-};
-#undef W
-
-struct kretprobe_blackpoint kretprobe_blacklist[] = {
-	{"__switch_to", }, /* This function switches only current task, but
-			      doesn't switch kernel stack.*/
-	{NULL, NULL}	/* Terminator */
-};
-const int kretprobe_blacklist_size = ARRAY_SIZE(kretprobe_blacklist);
-
-/* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
-static __always_inline void set_jmp_op(void *from, void *to)
-{
-	struct __arch_jmp_op {
-		char op;
-		s32 raddr;
-	} __attribute__((packed)) * jop;
-	jop = (struct __arch_jmp_op *)from;
-	jop->raddr = (s32)((long)(to) - ((long)(from) + 5));
-	jop->op = RELATIVEJUMP_INSTRUCTION;
-}
-
-/*
- * returns non-zero if opcode is boostable.
- * RIP relative instructions are adjusted at copying time
- */
-static __always_inline int can_boost(kprobe_opcode_t *opcodes)
-{
-	kprobe_opcode_t opcode;
-	kprobe_opcode_t *orig_opcodes = opcodes;
-
-retry:
-	if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
-		return 0;
-	opcode = *(opcodes++);
-
-	/* 2nd-byte opcode */
-	if (opcode == 0x0f) {
-		if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1)
-			return 0;
-		return test_bit(*opcodes,
-				(unsigned long *)twobyte_is_boostable);
-	}
-
-	switch (opcode & 0xf0) {
-	case 0x40:
-		goto retry; /* REX prefix is boostable */
-	case 0x60:
-		if (0x63 < opcode && opcode < 0x67)
-			goto retry; /* prefixes */
-		/* can't boost Address-size override and bound */
-		return (opcode != 0x62 && opcode != 0x67);
-	case 0x70:
-		return 0; /* can't boost conditional jump */
-	case 0xc0:
-		/* can't boost software-interruptions */
-		return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
-	case 0xd0:
-		/* can boost AA* and XLAT */
-		return (opcode == 0xd4 || opcode == 0xd5 || opcode == 0xd7);
-	case 0xe0:
-		/* can boost in/out and absolute jmps */
-		return ((opcode & 0x04) || opcode == 0xea);
-	case 0xf0:
-		if ((opcode & 0x0c) == 0 && opcode != 0xf1)
-			goto retry; /* lock/rep(ne) prefix */
-		/* clear and set flags are boostable */
-		return (opcode == 0xf5 || (0xf7 < opcode && opcode < 0xfe));
-	default:
-		/* segment override prefixes are boostable */
-		if (opcode == 0x26 || opcode == 0x36 || opcode == 0x3e)
-			goto retry; /* prefixes */
-		/* CS override prefix and call are not boostable */
-		return (opcode != 0x2e && opcode != 0x9a);
-	}
-}
-
-/*
- * returns non-zero if opcode modifies the interrupt flag.
- */
-static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
-{
-	switch (*insn) {
-	case 0xfa:		/* cli */
-	case 0xfb:		/* sti */
-	case 0xcf:		/* iret/iretd */
-	case 0x9d:		/* popf/popfd */
-		return 1;
-	}
-
-	/*
-	 * on 64 bit x86, 0x40-0x4f are prefixes so we need to look
-	 * at the next byte instead.. but of course not recurse infinitely
-	 */
-	if (*insn  >= 0x40 && *insn <= 0x4f)
-		return is_IF_modifier(++insn);
-	return 0;
-}
-
-/*
- * Adjust the displacement if the instruction uses the %rip-relative
- * addressing mode.
- * If it does, Return the address of the 32-bit displacement word.
- * If not, return null.
- */
-static void __kprobes fix_riprel(struct kprobe *p)
-{
-	u8 *insn = p->ainsn.insn;
-	s64 disp;
-	int need_modrm;
-
-	/* Skip legacy instruction prefixes.  */
-	while (1) {
-		switch (*insn) {
-		case 0x66:
-		case 0x67:
-		case 0x2e:
-		case 0x3e:
-		case 0x26:
-		case 0x64:
-		case 0x65:
-		case 0x36:
-		case 0xf0:
-		case 0xf3:
-		case 0xf2:
-			++insn;
-			continue;
-		}
-		break;
-	}
-
-	/* Skip REX instruction prefix.  */
-	if ((*insn & 0xf0) == 0x40)
-		++insn;
-
-	if (*insn == 0x0f) {	/* Two-byte opcode.  */
-		++insn;
-		need_modrm = test_bit(*insn,
-				      (unsigned long *)twobyte_has_modrm);
-	} else			/* One-byte opcode.  */
-		need_modrm = test_bit(*insn,
-				      (unsigned long *)onebyte_has_modrm);
-
-	if (need_modrm) {
-		u8 modrm = *++insn;
-		if ((modrm & 0xc7) == 0x05) { /* %rip+disp32 addressing mode */
-			/* Displacement follows ModRM byte.  */
-			++insn;
-			/*
-			 * The copied instruction uses the %rip-relative
-			 * addressing mode.  Adjust the displacement for the
-			 * difference between the original location of this
-			 * instruction and the location of the copy that will
-			 * actually be run.  The tricky bit here is making sure
-			 * that the sign extension happens correctly in this
-			 * calculation, since we need a signed 32-bit result to
-			 * be sign-extended to 64 bits when it's added to the
-			 * %rip value and yield the same 64-bit result that the
-			 * sign-extension of the original signed 32-bit
-			 * displacement would have given.
-			 */
-			disp = (u8 *) p->addr + *((s32 *) insn) -
-			       (u8 *) p->ainsn.insn;
-			BUG_ON((s64) (s32) disp != disp); /* Sanity check.  */
-			*(s32 *)insn = (s32) disp;
-		}
-	}
-}
-
-static void __kprobes arch_copy_kprobe(struct kprobe *p)
-{
-	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
-	fix_riprel(p);
-	if (can_boost(p->addr))
-		p->ainsn.boostable = 0;
-	else
-		p->ainsn.boostable = -1;
-
-	p->opcode = *p->addr;
-}
-
-int __kprobes arch_prepare_kprobe(struct kprobe *p)
-{
-	/* insn: must be on special executable page on x86. */
-	p->ainsn.insn = get_insn_slot();
-	if (!p->ainsn.insn)
-		return -ENOMEM;
-	arch_copy_kprobe(p);
-	return 0;
-}
-
-void __kprobes arch_arm_kprobe(struct kprobe *p)
-{
-	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
-}
-
-void __kprobes arch_disarm_kprobe(struct kprobe *p)
-{
-	text_poke(p->addr, &p->opcode, 1);
-}
-
-void __kprobes arch_remove_kprobe(struct kprobe *p)
-{
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
-	mutex_unlock(&kprobe_mutex);
-}
-
-static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
-{
-	kcb->prev_kprobe.kp = kprobe_running();
-	kcb->prev_kprobe.status = kcb->kprobe_status;
-	kcb->prev_kprobe.old_flags = kcb->kprobe_old_flags;
-	kcb->prev_kprobe.saved_flags = kcb->kprobe_saved_flags;
-}
-
-static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
-{
-	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
-	kcb->kprobe_status = kcb->prev_kprobe.status;
-	kcb->kprobe_old_flags = kcb->prev_kprobe.old_flags;
-	kcb->kprobe_saved_flags = kcb->prev_kprobe.saved_flags;
-}
-
-static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
-				struct kprobe_ctlblk *kcb)
-{
-	__get_cpu_var(current_kprobe) = p;
-	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
-		= (regs->flags & (TF_MASK | IF_MASK));
-	if (is_IF_modifier(p->ainsn.insn))
-		kcb->kprobe_saved_flags &= ~IF_MASK;
-}
-
-static __always_inline void clear_btf(void)
-{
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
-}
-
-static __always_inline void restore_btf(void)
-{
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, current->thread.debugctlmsr);
-}
-
-static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
-{
-	clear_btf();
-	regs->flags |= TF_MASK;
-	regs->flags &= ~IF_MASK;
-	/*single step inline if the instruction is an int3*/
-	if (p->opcode == BREAKPOINT_INSTRUCTION)
-		regs->ip = (unsigned long)p->addr;
-	else
-		regs->ip = (unsigned long)p->ainsn.insn;
-}
-
-/* Called with kretprobe_lock held */
-void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
-				      struct pt_regs *regs)
-{
-	unsigned long *sara = stack_addr(regs);
-
-	ri->ret_addr = (kprobe_opcode_t *) *sara;
-
-	/* Replace the return addr with trampoline addr */
-	*sara = (unsigned long) &kretprobe_trampoline;
-}
-
-/*
- * Interrupts are disabled on entry as trap3 is an interrupt gate and they
- * remain disabled thorough out this function.
- */
-static int __kprobes kprobe_handler(struct pt_regs *regs)
-{
-	struct kprobe *p;
-	int ret = 0;
-	kprobe_opcode_t *addr;
-	struct kprobe_ctlblk *kcb;
-
-	addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t));
-
-	/*
-	 * We don't want to be preempted for the entire
-	 * duration of kprobe processing
-	 */
-	preempt_disable();
-	kcb = get_kprobe_ctlblk();
-
-	/* Check we're not actually recursing */
-	if (kprobe_running()) {
-		p = get_kprobe(addr);
-		if (p) {
-			if (kcb->kprobe_status == KPROBE_HIT_SS &&
-				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
-				regs->flags &= ~TF_MASK;
-				regs->flags |= kcb->kprobe_saved_flags;
-				goto no_kprobe;
-			} else if (kcb->kprobe_status == KPROBE_HIT_SSDONE) {
-				/* TODO: Provide re-entrancy from
-				 * post_kprobes_handler() and avoid exception
-				 * stack corruption while single-stepping on
-				 * the instruction of the new probe.
-				 */
-				arch_disarm_kprobe(p);
-				regs->ip = (unsigned long)p->addr;
-				reset_current_kprobe();
-				return 1;
-			}
-			/* We have reentered the kprobe_handler(), since
-			 * another probe was hit while within the handler.
-			 * We here save the original kprobes variables and
-			 * just single step on the instruction of the new probe
-			 * without calling any user handlers.
-			 */
-			save_previous_kprobe(kcb);
-			set_current_kprobe(p, regs, kcb);
-			kprobes_inc_nmissed_count(p);
-			prepare_singlestep(p, regs);
-			kcb->kprobe_status = KPROBE_REENTER;
-			return 1;
-		} else {
-			if (*addr != BREAKPOINT_INSTRUCTION) {
-			/* The breakpoint instruction was removed by
-			 * another cpu right after we hit, no further
-			 * handling of this interrupt is appropriate
-			 */
-				regs->ip = (unsigned long)addr;
-				ret = 1;
-				goto no_kprobe;
-			}
-			p = __get_cpu_var(current_kprobe);
-			if (p->break_handler && p->break_handler(p, regs)) {
-				goto ss_probe;
-			}
-		}
-		goto no_kprobe;
-	}
-
-	p = get_kprobe(addr);
-	if (!p) {
-		if (*addr != BREAKPOINT_INSTRUCTION) {
-			/*
-			 * The breakpoint instruction was removed right
-			 * after we hit it.  Another cpu has removed
-			 * either a probepoint or a debugger breakpoint
-			 * at this address.  In either case, no further
-			 * handling of this interrupt is appropriate.
-			 * Back up over the (now missing) int3 and run
-			 * the original instruction.
-			 */
-			regs->ip = (unsigned long)addr;
-			ret = 1;
-		}
-		/* Not one of ours: let kernel handle it */
-		goto no_kprobe;
-	}
-
-	set_current_kprobe(p, regs, kcb);
-	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
-
-	if (p->pre_handler && p->pre_handler(p, regs))
-		/* handler has already set things up, so skip ss setup */
-		return 1;
-
-ss_probe:
-#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
-	if (p->ainsn.boostable == 1 && !p->post_handler) {
-		/* Boost up -- we can execute copied instructions directly */
-		reset_current_kprobe();
-		regs->ip = (unsigned long)p->ainsn.insn;
-		preempt_enable_no_resched();
-		return 1;
-	}
-#endif
-	prepare_singlestep(p, regs);
-	kcb->kprobe_status = KPROBE_HIT_SS;
-	return 1;
-
-no_kprobe:
-	preempt_enable_no_resched();
-	return ret;
-}
-
-/*
- * When a retprobed function returns, this code saves registers and
- * calls trampoline_handler() runs, which calls the kretprobe's handler.
- */
- void __kprobes kretprobe_trampoline_holder(void)
- {
- 	asm volatile (  ".global kretprobe_trampoline\n"
-			"kretprobe_trampoline: \n"
-			/* We don't bother saving the ss register */
-			"	pushq %rsp\n"
-			"	pushfq\n"
-			/*
-			 * Skip cs, ip, orig_ax.
-			 * trampoline_handler() will plug in these values
-			 */
-			"	subq $24, %rsp\n"
-			"	pushq %rdi\n"
-			"	pushq %rsi\n"
-			"	pushq %rdx\n"
-			"	pushq %rcx\n"
-			"	pushq %rax\n"
-			"	pushq %r8\n"
-			"	pushq %r9\n"
-			"	pushq %r10\n"
-			"	pushq %r11\n"
-			"	pushq %rbx\n"
-			"	pushq %rbp\n"
-			"	pushq %r12\n"
-			"	pushq %r13\n"
-			"	pushq %r14\n"
-			"	pushq %r15\n"
-			"	movq %rsp, %rdi\n"
-			"	call trampoline_handler\n"
-			/* Replace saved sp with true return address. */
-			"	movq %rax, 152(%rsp)\n"
-			"	popq %r15\n"
-			"	popq %r14\n"
-			"	popq %r13\n"
-			"	popq %r12\n"
-			"	popq %rbp\n"
-			"	popq %rbx\n"
-			"	popq %r11\n"
-			"	popq %r10\n"
-			"	popq %r9\n"
-			"	popq %r8\n"
-			"	popq %rax\n"
-			"	popq %rcx\n"
-			"	popq %rdx\n"
-			"	popq %rsi\n"
-			"	popq %rdi\n"
-			/* Skip orig_ax, ip, cs */
-			"	addq $24, %rsp\n"
-			"	popfq\n"
-			"	ret\n");
- }
-
-/*
- * Called from kretprobe_trampoline
- */
-fastcall void * __kprobes trampoline_handler(struct pt_regs *regs)
-{
-	struct kretprobe_instance *ri = NULL;
-	struct hlist_head *head, empty_rp;
-	struct hlist_node *node, *tmp;
-	unsigned long flags, orig_ret_address = 0;
-	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
-
-	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
-	/* fixup registers */
-	regs->cs = __KERNEL_CS;
-	regs->ip = trampoline_address;
-	regs->orig_ax = ~0UL;
-
-	/*
-	 * It is possible to have multiple instances associated with a given
-	 * task either because multiple functions in the call path have
-	 * return probes installed on them, and/or more then one
-	 * return probe was registered for a target function.
-	 *
-	 * We can handle this because:
-	 *     - instances are always pushed into the head of the list
-	 *     - when multiple return probes are registered for the same
-	 *	 function, the (chronologically) first instance's ret_addr
-	 *	 will be the real return address, and all the rest will
-	 *	 point to kretprobe_trampoline.
-	 */
-	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
-		if (ri->task != current)
-			/* another task is sharing our hash bucket */
-			continue;
-
-		if (ri->rp && ri->rp->handler) {
-			__get_cpu_var(current_kprobe) = &ri->rp->kp;
-			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
-			ri->rp->handler(ri, regs);
-			__get_cpu_var(current_kprobe) = NULL;
-		}
-
-		orig_ret_address = (unsigned long)ri->ret_addr;
-		recycle_rp_inst(ri, &empty_rp);
-
-		if (orig_ret_address != trampoline_address)
-			/*
-			 * This is the real return address. Any other
-			 * instances associated with this task are for
-			 * other calls deeper on the call stack
-			 */
-			break;
-	}
-
-	kretprobe_assert(ri, orig_ret_address, trampoline_address);
-
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
-
-	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
-		hlist_del(&ri->hlist);
-		kfree(ri);
-	}
-	return (void *)orig_ret_address;
-}
-
-/*
- * Called after single-stepping.  p->addr is the address of the
- * instruction whose first byte has been replaced by the "int 3"
- * instruction.  To avoid the SMP problems that can occur when we
- * temporarily put back the original opcode to single-step, we
- * single-stepped a copy of the instruction.  The address of this
- * copy is p->ainsn.insn.
- *
- * This function prepares to return from the post-single-step
- * interrupt.  We have to fix up the stack as follows:
- *
- * 0) Except in the case of absolute or indirect jump or call instructions,
- * the new ip is relative to the copied instruction.  We need to make
- * it relative to the original instruction.
- *
- * 1) If the single-stepped instruction was pushfl, then the TF and IF
- * flags are set in the just-pushed flags, and may need to be cleared.
- *
- * 2) If the single-stepped instruction was a call, the return address
- * that is atop the stack is the address following the copied instruction.
- * We need to make it the address following the original instruction.
- *
- * If this is the first time we've single-stepped the instruction at
- * this probepoint, and the instruction is boostable, boost it: add a
- * jump instruction after the copied instruction, that jumps to the next
- * instruction after the probepoint.
- */
-static void __kprobes resume_execution(struct kprobe *p,
-		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
-{
-	unsigned long *tos = stack_addr(regs);
-	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
-	unsigned long orig_ip = (unsigned long)p->addr;
-	kprobe_opcode_t *insn = p->ainsn.insn;
-
-	/*skip the REX prefix*/
-	if (*insn >= 0x40 && *insn <= 0x4f)
-		insn++;
-
-	regs->flags &= ~TF_MASK;
-	switch (*insn) {
-	case 0x9c:	/* pushfl */
-		*tos &= ~(TF_MASK | IF_MASK);
-		*tos |= kcb->kprobe_old_flags;
-		break;
-	case 0xc2:	/* iret/ret/lret */
-	case 0xc3:
-	case 0xca:
-	case 0xcb:
-	case 0xcf:
-	case 0xea:	/* jmp absolute -- ip is correct */
-		/* ip is already adjusted, no more changes required */
-		p->ainsn.boostable = 1;
-		goto no_change;
-	case 0xe8:	/* call relative - Fix return addr */
-		*tos = orig_ip + (*tos - copy_ip);
-		break;
-	case 0xff:
-		if ((insn[1] & 0x30) == 0x10) {
-			/*
-			 * call absolute, indirect
-			 * Fix return addr; ip is correct.
-			 * But this is not boostable
-			 */
-			*tos = orig_ip + (*tos - copy_ip);
-			goto no_change;
-		} else if (((insn[1] & 0x31) == 0x20) ||
-			   ((insn[1] & 0x31) == 0x21)) {
-			/*
-			 * jmp near and far, absolute indirect
-			 * ip is correct. And this is boostable
-			 */
-			p->ainsn.boostable = 1;
-			goto no_change;
-		}
-	default:
-		break;
-	}
-
-	if (p->ainsn.boostable == 0) {
-		if ((regs->ip > copy_ip) &&
-		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
-			/*
-			 * These instructions can be executed directly if it
-			 * jumps back to correct address.
-			 */
-			set_jmp_op((void *)regs->ip,
-				   (void *)orig_ip + (regs->ip - copy_ip));
-			p->ainsn.boostable = 1;
-		} else {
-			p->ainsn.boostable = -1;
-		}
-	}
-
-	regs->ip += orig_ip - copy_ip;
-
-no_change:
-	restore_btf();
-
-	return;
-}
-
-/*
- * Interrupts are disabled on entry as trap1 is an interrupt gate and they
- * remain disabled thoroughout this function.
- */
-static int __kprobes post_kprobe_handler(struct pt_regs *regs)
-{
-	struct kprobe *cur = kprobe_running();
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	if (!cur)
-		return 0;
-
-	if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
-		kcb->kprobe_status = KPROBE_HIT_SSDONE;
-		cur->post_handler(cur, regs, 0);
-	}
-
-	resume_execution(cur, regs, kcb);
-	regs->flags |= kcb->kprobe_saved_flags;
-	trace_hardirqs_fixup_flags(regs->flags);
-
-	/* Restore back the original saved kprobes variables and continue. */
-	if (kcb->kprobe_status == KPROBE_REENTER) {
-		restore_previous_kprobe(kcb);
-		goto out;
-	}
-	reset_current_kprobe();
-out:
-	preempt_enable_no_resched();
-
-	/*
-	 * if somebody else is singlestepping across a probe point, flags
-	 * will have TF set, in which case, continue the remaining processing
-	 * of do_debug, as if this is not a probe hit.
-	 */
-	if (regs->flags & TF_MASK)
-		return 0;
-
-	return 1;
-}
-
-int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
-{
-	struct kprobe *cur = kprobe_running();
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	const struct exception_table_entry *fixup;
-
-	switch(kcb->kprobe_status) {
-	case KPROBE_HIT_SS:
-	case KPROBE_REENTER:
-		/*
-		 * We are here because the instruction being single
-		 * stepped caused a page fault. We reset the current
-		 * kprobe and the ip points back to the probe address
-		 * and allow the page fault handler to continue as a
-		 * normal page fault.
-		 */
-		regs->ip = (unsigned long)cur->addr;
-		regs->flags |= kcb->kprobe_old_flags;
-		if (kcb->kprobe_status == KPROBE_REENTER)
-			restore_previous_kprobe(kcb);
-		else
-			reset_current_kprobe();
-		preempt_enable_no_resched();
-		break;
-	case KPROBE_HIT_ACTIVE:
-	case KPROBE_HIT_SSDONE:
-		/*
-		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accounting
-		 * these specific fault cases.
-		 */
-		kprobes_inc_nmissed_count(cur);
-
-		/*
-		 * We come here because instructions in the pre/post
-		 * handler caused the page_fault, this could happen
-		 * if handler tries to access user space by
-		 * copy_from_user(), get_user() etc. Let the
-		 * user-specified handler try to fix it first.
-		 */
-		if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
-			return 1;
-
-		/*
-		 * In case the user-specified fault handler returned
-		 * zero, try to fix up.
-		 */
-		fixup = search_exception_tables(regs->ip);
-		if (fixup) {
-			regs->ip = fixup->fixup;
-			return 1;
-		}
-
-		/*
-		 * fixup routine could not handle it,
-		 * Let do_page_fault() fix it.
-		 */
-		break;
-	default:
-		break;
-	}
-	return 0;
-}
-
-/*
- * Wrapper routine for handling exceptions.
- */
-int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
-				       unsigned long val, void *data)
-{
-	struct die_args *args = (struct die_args *)data;
-	int ret = NOTIFY_DONE;
-
-	if (args->regs && user_mode_vm(args->regs))
-		return ret;
-
-	switch (val) {
-	case DIE_INT3:
-		if (kprobe_handler(args->regs))
-			ret = NOTIFY_STOP;
-		break;
-	case DIE_DEBUG:
-		if (post_kprobe_handler(args->regs))
-			ret = NOTIFY_STOP;
-		break;
-	case DIE_GPF:
-		/* kprobe_running() needs smp_processor_id() */
-		preempt_disable();
-		if (kprobe_running() &&
-		    kprobe_fault_handler(args->regs, args->trapnr))
-			ret = NOTIFY_STOP;
-		preempt_enable();
-		break;
-	default:
-		break;
-	}
-	return ret;
-}
-
-int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
-{
-	struct jprobe *jp = container_of(p, struct jprobe, kp);
-	unsigned long addr;
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	kcb->jprobe_saved_regs = *regs;
-	kcb->jprobe_saved_sp = stack_addr(regs);
-	addr = (unsigned long)(kcb->jprobe_saved_sp);
-
-	/*
-	 * As Linus pointed out, gcc assumes that the callee
-	 * owns the argument space and could overwrite it, e.g.
-	 * tailcall optimization. So, to be absolutely safe
-	 * we also save and restore enough stack bytes to cover
-	 * the argument area.
-	 */
-	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
-			MIN_STACK_SIZE(addr));
-	regs->flags &= ~IF_MASK;
-	trace_hardirqs_off();
-	regs->ip = (unsigned long)(jp->entry);
-	return 1;
-}
-
-void __kprobes jprobe_return(void)
-{
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-
-	asm volatile ("       xchg   %%rbx,%%rsp     \n"
-		      "       int3			\n"
-		      "       .globl jprobe_return_end	\n"
-		      "       jprobe_return_end:	\n"
-		      "       nop			\n"::"b"
-		      (kcb->jprobe_saved_sp):"memory");
-}
-
-int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
-{
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	u8 *addr = (u8 *) (regs->ip - 1);
-	struct jprobe *jp = container_of(p, struct jprobe, kp);
-
-	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
-		if (stack_addr(regs) != kcb->jprobe_saved_sp) {
-			struct pt_regs *saved_regs = &kcb->jprobe_saved_regs;
-			printk("current sp %p does not match saved sp %p\n",
-			       stack_addr(regs), kcb->jprobe_saved_sp);
-			printk("Saved registers for jprobe %p\n", jp);
-			show_registers(saved_regs);
-			printk("Current registers\n");
-			show_registers(regs);
-			BUG();
-		}
-		*regs = kcb->jprobe_saved_regs;
-		memcpy((kprobe_opcode_t *)(kcb->jprobe_saved_sp),
-		       kcb->jprobes_stack,
-		       MIN_STACK_SIZE(kcb->jprobe_saved_sp));
-		preempt_enable_no_resched();
-		return 1;
-	}
-	return 0;
-}
-
-int __init arch_init_kprobes(void)
-{
-	return 0;
-}
-
-int __kprobes arch_trampoline_kprobe(struct kprobe *p)
-{
-	return 0;
-}
diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h
index b7bbd25ba2a6..6e6371a856e7 100644
--- a/include/asm-x86/kprobes.h
+++ b/include/asm-x86/kprobes.h
@@ -1,5 +1,98 @@
-#ifdef CONFIG_X86_32
-# include "kprobes_32.h"
-#else
-# include "kprobes_64.h"
-#endif
+#ifndef _ASM_KPROBES_H
+#define _ASM_KPROBES_H
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * See arch/x86/kernel/kprobes.c for x86 kprobes history.
+ */
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/percpu.h>
+
+#define  __ARCH_WANT_KPROBES_INSN_SLOT
+
+struct pt_regs;
+struct kprobe;
+
+typedef u8 kprobe_opcode_t;
+#define BREAKPOINT_INSTRUCTION	0xcc
+#define RELATIVEJUMP_INSTRUCTION 0xe9
+#define MAX_INSN_SIZE 16
+#define MAX_STACK_SIZE 64
+#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
+	(((unsigned long)current_thread_info()) + THREAD_SIZE \
+	 - (unsigned long)(ADDR))) \
+	? (MAX_STACK_SIZE) \
+	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
+	   - (unsigned long)(ADDR)))
+
+#define ARCH_SUPPORTS_KRETPROBES
+#define flush_insn_slot(p)	do { } while (0)
+
+extern const int kretprobe_blacklist_size;
+
+void arch_remove_kprobe(struct kprobe *p);
+void kretprobe_trampoline(void);
+
+/* Architecture specific copy of original instruction*/
+struct arch_specific_insn {
+	/* copy of the original instruction */
+	kprobe_opcode_t *insn;
+	/*
+	 * boostable = -1: This instruction type is not boostable.
+	 * boostable = 0: This instruction type is boostable.
+	 * boostable = 1: This instruction has been boosted: we have
+	 * added a relative jump after the instruction copy in insn,
+	 * so no single-step and fixup are needed (unless there's
+	 * a post_handler or break_handler).
+	 */
+	int boostable;
+};
+
+struct prev_kprobe {
+	struct kprobe *kp;
+	unsigned long status;
+	unsigned long old_flags;
+	unsigned long saved_flags;
+};
+
+/* per-cpu kprobe control block */
+struct kprobe_ctlblk {
+	unsigned long kprobe_status;
+	unsigned long kprobe_old_flags;
+	unsigned long kprobe_saved_flags;
+	unsigned long *jprobe_saved_sp;
+	struct pt_regs jprobe_saved_regs;
+	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
+	struct prev_kprobe prev_kprobe;
+};
+
+/* trap3/1 are intr gates for kprobes.  So, restore the status of IF,
+ * if necessary, before executing the original int3/1 (trap) handler.
+ */
+static inline void restore_interrupts(struct pt_regs *regs)
+{
+	if (regs->flags & IF_MASK)
+		local_irq_enable();
+}
+
+extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
+extern int kprobe_exceptions_notify(struct notifier_block *self,
+				    unsigned long val, void *data);
+#endif				/* _ASM_KPROBES_H */
diff --git a/include/asm-x86/kprobes_32.h b/include/asm-x86/kprobes_32.h
deleted file mode 100644
index 94dff09b2ebe..000000000000
--- a/include/asm-x86/kprobes_32.h
+++ /dev/null
@@ -1,102 +0,0 @@
-#ifndef _ASM_KPROBES_H
-#define _ASM_KPROBES_H
-/*
- *  Kernel Probes (KProbes)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
- *
- * Copyright (C) IBM Corporation, 2002, 2004
- *
- * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
- *		Probes initial implementation ( includes suggestions from
- *		Rusty Russell).
- * 2004-Oct	Prasanna S Panchamukhi <prasanna@in.ibm.com> and Jim Keniston
- *		kenistoj@us.ibm.com adopted from i386.
- */
-#include <linux/types.h>
-#include <linux/ptrace.h>
-#include <linux/percpu.h>
-
-#define  __ARCH_WANT_KPROBES_INSN_SLOT
-
-struct pt_regs;
-struct kprobe;
-
-typedef u8 kprobe_opcode_t;
-#define BREAKPOINT_INSTRUCTION	0xcc
-#define RELATIVEJUMP_INSTRUCTION 0xe9
-#define MAX_INSN_SIZE 16
-#define MAX_STACK_SIZE 64
-#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE \
-	 - (unsigned long)(ADDR))) \
-	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
-	   - (unsigned long)(ADDR)))
-
-#define ARCH_SUPPORTS_KRETPROBES
-#define flush_insn_slot(p)	do { } while (0)
-
-extern const int kretprobe_blacklist_size;
-
-void arch_remove_kprobe(struct kprobe *p);
-void kretprobe_trampoline(void);
-
-/* Architecture specific copy of original instruction*/
-struct arch_specific_insn {
-	/* copy of the original instruction */
-	kprobe_opcode_t *insn;
-	/*
-	 * boostable = -1: This instruction type is not boostable.
-	 * boostable = 0: This instruction type is boostable.
-	 * boostable = 1: This instruction has been boosted: we have
-	 * added a relative jump after the instruction copy in insn,
-	 * so no single-step and fixup are needed (unless there's
-	 * a post_handler or break_handler).
-	 */
-	int boostable;
-};
-
-struct prev_kprobe {
-	struct kprobe *kp;
-	unsigned long status;
-	unsigned long old_flags;
-	unsigned long saved_flags;
-};
-
-/* per-cpu kprobe control block */
-struct kprobe_ctlblk {
-	unsigned long kprobe_status;
-	unsigned long kprobe_old_flags;
-	unsigned long kprobe_saved_flags;
-	unsigned long *jprobe_saved_sp;
-	struct pt_regs jprobe_saved_regs;
-	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
-	struct prev_kprobe prev_kprobe;
-};
-
-/* trap3/1 are intr gates for kprobes.  So, restore the status of IF,
- * if necessary, before executing the original int3/1 (trap) handler.
- */
-static inline void restore_interrupts(struct pt_regs *regs)
-{
-	if (regs->flags & IF_MASK)
-		local_irq_enable();
-}
-
-extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
-extern int kprobe_exceptions_notify(struct notifier_block *self,
-				    unsigned long val, void *data);
-#endif				/* _ASM_KPROBES_H */
diff --git a/include/asm-x86/kprobes_64.h b/include/asm-x86/kprobes_64.h
deleted file mode 100644
index 94dff09b2ebe..000000000000
--- a/include/asm-x86/kprobes_64.h
+++ /dev/null
@@ -1,102 +0,0 @@
-#ifndef _ASM_KPROBES_H
-#define _ASM_KPROBES_H
-/*
- *  Kernel Probes (KProbes)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
- *
- * Copyright (C) IBM Corporation, 2002, 2004
- *
- * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
- *		Probes initial implementation ( includes suggestions from
- *		Rusty Russell).
- * 2004-Oct	Prasanna S Panchamukhi <prasanna@in.ibm.com> and Jim Keniston
- *		kenistoj@us.ibm.com adopted from i386.
- */
-#include <linux/types.h>
-#include <linux/ptrace.h>
-#include <linux/percpu.h>
-
-#define  __ARCH_WANT_KPROBES_INSN_SLOT
-
-struct pt_regs;
-struct kprobe;
-
-typedef u8 kprobe_opcode_t;
-#define BREAKPOINT_INSTRUCTION	0xcc
-#define RELATIVEJUMP_INSTRUCTION 0xe9
-#define MAX_INSN_SIZE 16
-#define MAX_STACK_SIZE 64
-#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE \
-	 - (unsigned long)(ADDR))) \
-	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
-	   - (unsigned long)(ADDR)))
-
-#define ARCH_SUPPORTS_KRETPROBES
-#define flush_insn_slot(p)	do { } while (0)
-
-extern const int kretprobe_blacklist_size;
-
-void arch_remove_kprobe(struct kprobe *p);
-void kretprobe_trampoline(void);
-
-/* Architecture specific copy of original instruction*/
-struct arch_specific_insn {
-	/* copy of the original instruction */
-	kprobe_opcode_t *insn;
-	/*
-	 * boostable = -1: This instruction type is not boostable.
-	 * boostable = 0: This instruction type is boostable.
-	 * boostable = 1: This instruction has been boosted: we have
-	 * added a relative jump after the instruction copy in insn,
-	 * so no single-step and fixup are needed (unless there's
-	 * a post_handler or break_handler).
-	 */
-	int boostable;
-};
-
-struct prev_kprobe {
-	struct kprobe *kp;
-	unsigned long status;
-	unsigned long old_flags;
-	unsigned long saved_flags;
-};
-
-/* per-cpu kprobe control block */
-struct kprobe_ctlblk {
-	unsigned long kprobe_status;
-	unsigned long kprobe_old_flags;
-	unsigned long kprobe_saved_flags;
-	unsigned long *jprobe_saved_sp;
-	struct pt_regs jprobe_saved_regs;
-	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
-	struct prev_kprobe prev_kprobe;
-};
-
-/* trap3/1 are intr gates for kprobes.  So, restore the status of IF,
- * if necessary, before executing the original int3/1 (trap) handler.
- */
-static inline void restore_interrupts(struct pt_regs *regs)
-{
-	if (regs->flags & IF_MASK)
-		local_irq_enable();
-}
-
-extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
-extern int kprobe_exceptions_notify(struct notifier_block *self,
-				    unsigned long val, void *data);
-#endif				/* _ASM_KPROBES_H */
-- 
cgit v1.2.3


From bdb4f156064e5f627213af82292eb8b5cf2dc5aa Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: i386: hard_{en,dis}able_TSC can be static

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c   | 4 ++--
 include/asm-x86/processor_32.h | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 48e92e3758c2..40cc29695eba 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -588,7 +588,7 @@ int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
 }
 
 #ifdef CONFIG_SECCOMP
-void hard_disable_TSC(void)
+static void hard_disable_TSC(void)
 {
 	write_cr4(read_cr4() | X86_CR4_TSD);
 }
@@ -603,7 +603,7 @@ void disable_TSC(void)
 		hard_disable_TSC();
 	preempt_enable();
 }
-void hard_enable_TSC(void)
+static void hard_enable_TSC(void)
 {
 	write_cr4(read_cr4() & ~X86_CR4_TSD);
 }
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index e5056ab9dd9f..a5560a3b57ef 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -206,9 +206,7 @@ extern int bootloader_type;
 
 #define HAVE_ARCH_PICK_MMAP_LAYOUT
 
-extern void hard_disable_TSC(void);
 extern void disable_TSC(void);
-extern void hard_enable_TSC(void);
 
 /*
  * Size of io_bitmap.
-- 
cgit v1.2.3


From d88879b2d0225da3ba460bbdb8361bb049653671 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:21 +0100
Subject: x86-64: make pda's cpunumber and nodenumber unsigned

This generally allows better code to be generated, since the zero-
extension during 32-bit operations comes for free (needed when the
result is used as array index or similar), whereas sign extension must
be done explicitly and frequently requires a one byte larger
instruction due to the necessary rex64 prefix.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pda.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index 3d9dd653628a..b620d0c39a93 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -15,14 +15,14 @@ struct x8664_pda {
 	unsigned long kernelstack;  /* 16 top of kernel stack for current */
 	unsigned long oldrsp; 	    /* 24 user rsp for system call */
         int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
-	int cpunumber;		    /* 36 Logical CPU number */
+	unsigned int cpunumber;	    /* 36 Logical CPU number */
 #ifdef CONFIG_CC_STACKPROTECTOR
 	unsigned long stack_canary;	/* 40 stack canary value */
 					/* gcc-ABI: this canary MUST be at
 					   offset 40!!! */
 #endif
 	char *irqstackptr;
-	int nodenumber;		    /* number of current node */
+	unsigned int nodenumber;	/* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
 	short mmu_state;
-- 
cgit v1.2.3


From 22f5991c85dec1281cce5c8df9ee92b43b1738c8 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:23 +0100
Subject: x86-64: honor notify_die() returning NOTIFY_STOP

This requires making die() return a value, making its callers honor
this (and be prepared that it may return), and making oops_end() have
two additional parameters.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/mcheck/mce_64.c |  8 ++++----
 arch/x86/kernel/traps_64.c          | 23 +++++++++++++++--------
 arch/x86/mm/fault_64.c              | 12 ++++++------
 include/asm-x86/kdebug.h            |  4 ++--
 4 files changed, 27 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 8cd47fe2ef2c..bc6e35153d83 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -192,10 +192,10 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 
 	atomic_inc(&mce_entry);
 
-	if (regs)
-		notify_die(DIE_NMI, "machine check", regs, error_code, 18,
-			   SIGKILL);
-	if (!banks)
+	if ((regs
+	     && notify_die(DIE_NMI, "machine check", regs, error_code,
+			   18, SIGKILL) == NOTIFY_STOP)
+	    || !banks)
 		goto out2;
 
 	memset(&m, 0, sizeof(struct mce));
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 1a12a81fdb1b..cf90ceb48f5f 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -488,7 +488,7 @@ unsigned __kprobes long oops_begin(void)
 	return flags;
 }
 
-void __kprobes oops_end(unsigned long flags)
+void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
 { 
 	die_owner = -1;
 	bust_spinlocks(0);
@@ -497,12 +497,17 @@ void __kprobes oops_end(unsigned long flags)
 		/* Nest count reaches zero, release the lock. */
 		__raw_spin_unlock(&die_lock);
 	raw_local_irq_restore(flags);
+	if (!regs) {
+		oops_exit();
+		return;
+	}
 	if (panic_on_oops)
 		panic("Fatal exception");
 	oops_exit();
+	do_exit(signr);
 }
 
-void __kprobes __die(const char * str, struct pt_regs * regs, long err)
+int __kprobes __die(const char * str, struct pt_regs * regs, long err)
 {
 	static int die_counter;
 	printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff,++die_counter);
@@ -516,7 +521,8 @@ void __kprobes __die(const char * str, struct pt_regs * regs, long err)
 	printk("DEBUG_PAGEALLOC");
 #endif
 	printk("\n");
-	notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV);
+	if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+		return 1;
 	show_registers(regs);
 	add_taint(TAINT_DIE);
 	/* Executive summary in case the oops scrolled away */
@@ -525,6 +531,7 @@ void __kprobes __die(const char * str, struct pt_regs * regs, long err)
 	printk(" RSP <%016lx>\n", regs->sp);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
+	return 0;
 }
 
 void die(const char * str, struct pt_regs * regs, long err)
@@ -534,9 +541,9 @@ void die(const char * str, struct pt_regs * regs, long err)
 	if (!user_mode(regs))
 		report_bug(regs->ip, regs);
 
-	__die(str, regs, err);
-	oops_end(flags);
-	do_exit(SIGSEGV); 
+	if (__die(str, regs, err))
+		regs = NULL;
+	oops_end(flags, regs, SIGSEGV);
 }
 
 void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
@@ -553,10 +560,10 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 		crash_kexec(regs);
 	if (do_panic || panic_on_oops)
 		panic("Non maskable interrupt");
-	oops_end(flags);
+	oops_end(flags, NULL, SIGBUS);
 	nmi_exit();
 	local_irq_enable();
-	do_exit(SIGSEGV);
+	do_exit(SIGBUS);
 }
 
 static void __kprobes do_trap(int trapnr, int signr, char *str,
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 162050d4e5a3..121c7bda6297 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -227,9 +227,9 @@ static noinline void pgtable_bad(unsigned long address, struct pt_regs *regs,
 	tsk->thread.cr2 = address;
 	tsk->thread.trap_no = 14;
 	tsk->thread.error_code = error_code;
-	__die("Bad pagetable", regs, error_code);
-	oops_end(flags);
-	do_exit(SIGKILL);
+	if (__die("Bad pagetable", regs, error_code))
+		regs = NULL;
+	oops_end(flags, regs, SIGKILL);
 }
 
 /*
@@ -541,11 +541,11 @@ no_context:
 	tsk->thread.cr2 = address;
 	tsk->thread.trap_no = 14;
 	tsk->thread.error_code = error_code;
-	__die("Oops", regs, error_code);
+	if (__die("Oops", regs, error_code))
+		regs = NULL;
 	/* Executive summary in case the body of the oops scrolled away */
 	printk(KERN_EMERG "CR2: %016lx\n", address);
-	oops_end(flags);
-	do_exit(SIGKILL);
+	oops_end(flags, regs, SIGKILL);
 
 /*
  * We ran out of memory, or some other thing happened to us that made
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index 49e5c91d490c..a5e5e3b7eb23 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -25,7 +25,7 @@ enum die_val {
 extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 extern void printk_address(unsigned long address);
 extern void die(const char *,struct pt_regs *,long);
-extern void __die(const char *,struct pt_regs *,long);
+extern int __must_check __die(const char *, struct pt_regs *, long);
 extern void show_registers(struct pt_regs *regs);
 extern void __show_registers(struct pt_regs *, int all);
 extern void show_trace(struct task_struct *, struct pt_regs *, unsigned long *);
@@ -33,6 +33,6 @@ extern void __show_regs(struct pt_regs *regs);
 extern void show_regs(struct pt_regs *regs);
 extern void dump_pagetable(unsigned long);
 extern unsigned long oops_begin(void);
-extern void oops_end(unsigned long);
+extern void oops_end(unsigned long, struct pt_regs *, int signr);
 
 #endif
-- 
cgit v1.2.3


From 3e7622f9d7807a0a826d042cafc211cd1a29448c Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:23 +0100
Subject: x86: move to .rodata/.init.data

The array is never written, and on 64-bits it's not even being used
past initial boot.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/entry_32.S  | 2 +-
 arch/x86/kernel/i8259_64.c  | 2 +-
 include/asm-x86/hw_irq_32.h | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 6a474e1028c7..be5c31d04884 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -583,7 +583,7 @@ END(syscall_badsys)
  * Build the entry stubs and pointer table with
  * some assembler magic.
  */
-.data
+.section .rodata,"a"
 ENTRY(interrupt)
 .text
 
diff --git a/arch/x86/kernel/i8259_64.c b/arch/x86/kernel/i8259_64.c
index be82b1217691..99c8406ae253 100644
--- a/arch/x86/kernel/i8259_64.c
+++ b/arch/x86/kernel/i8259_64.c
@@ -77,7 +77,7 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd) BUILD_16_IRQS(0xe) BUILD_16_IRQS(0xf)
 	IRQ(x,c), IRQ(x,d), IRQ(x,e), IRQ(x,f)
 
 /* for the irq vectors */
-static void (*interrupt[NR_VECTORS - FIRST_EXTERNAL_VECTOR])(void) = {
+static void (*__initdata interrupt[NR_VECTORS - FIRST_EXTERNAL_VECTOR])(void) = {
 					  IRQLIST_16(0x2), IRQLIST_16(0x3),
 	IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7),
 	IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb),
diff --git a/include/asm-x86/hw_irq_32.h b/include/asm-x86/hw_irq_32.h
index b93e35a708ac..6d65fbb6358b 100644
--- a/include/asm-x86/hw_irq_32.h
+++ b/include/asm-x86/hw_irq_32.h
@@ -26,7 +26,7 @@
  * Interrupt entry/exit code at both C and assembly level
  */
 
-extern void (*interrupt[NR_IRQS])(void);
+extern void (*const interrupt[NR_IRQS])(void);
 
 #ifdef CONFIG_SMP
 void reschedule_interrupt(void);
-- 
cgit v1.2.3


From cae4595764cb3b08f6517e99bac1e3862854b1a1 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:23 +0100
Subject: x86: make __{save,restore}_processor_state static

.. allowing to remove their declarations from a global include file
(the symbols don't exist for anything but x86).

Likewise for 64-bits' fix_processor_context(), just that that one was
properly declared in an arch-specific header.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/suspend_64.c | 8 +++++---
 arch/x86/power/cpu.c         | 4 ++--
 include/asm-x86/suspend_64.h | 2 --
 include/linux/suspend.h      | 3 ---
 4 files changed, 7 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/suspend_64.c b/arch/x86/kernel/suspend_64.c
index 279c25775d19..09199511c256 100644
--- a/arch/x86/kernel/suspend_64.c
+++ b/arch/x86/kernel/suspend_64.c
@@ -17,6 +17,8 @@
 /* References to section boundaries */
 extern const void __nosave_begin, __nosave_end;
 
+static void fix_processor_context(void);
+
 struct saved_context saved_context;
 
 /**
@@ -34,7 +36,7 @@ struct saved_context saved_context;
  *	needed by kernel A, so that it can operate correctly after the resume
  *	regardless of what kernel B does in the meantime.
  */
-void __save_processor_state(struct saved_context *ctxt)
+static void __save_processor_state(struct saved_context *ctxt)
 {
 	kernel_fpu_begin();
 
@@ -89,7 +91,7 @@ static void do_fpu_end(void)
  *		by __save_processor_state()
  *	@ctxt - structure to load the registers contents from
  */
-void __restore_processor_state(struct saved_context *ctxt)
+static void __restore_processor_state(struct saved_context *ctxt)
 {
 	/*
 	 * control registers
@@ -133,7 +135,7 @@ void restore_processor_state(void)
 	__restore_processor_state(&saved_context);
 }
 
-void fix_processor_context(void)
+static void fix_processor_context(void)
 {
 	int cpu = smp_processor_id();
 	struct tss_struct *t = &per_cpu(init_tss, cpu);
diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c
index 5a98dc35addf..efcf620d1439 100644
--- a/arch/x86/power/cpu.c
+++ b/arch/x86/power/cpu.c
@@ -19,7 +19,7 @@ unsigned long saved_context_esp, saved_context_ebp;
 unsigned long saved_context_esi, saved_context_edi;
 unsigned long saved_context_eflags;
 
-void __save_processor_state(struct saved_context *ctxt)
+static void __save_processor_state(struct saved_context *ctxt)
 {
 	mtrr_save_fixed_ranges(NULL);
 	kernel_fpu_begin();
@@ -86,7 +86,7 @@ static void fix_processor_context(void)
 
 }
 
-void __restore_processor_state(struct saved_context *ctxt)
+static void __restore_processor_state(struct saved_context *ctxt)
 {
 	/*
 	 * control registers
diff --git a/include/asm-x86/suspend_64.h b/include/asm-x86/suspend_64.h
index 4404668f9aa4..2eb92cb81a0d 100644
--- a/include/asm-x86/suspend_64.h
+++ b/include/asm-x86/suspend_64.h
@@ -45,8 +45,6 @@ struct saved_context {
 #define loaddebug(thread,register) \
 	set_debugreg((thread)->debugreg##register, register)
 
-extern void fix_processor_context(void);
-
 /* routines for saving/restoring kernel state */
 extern int acpi_save_state_mem(void);
 extern char core_restore_code;
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 4360e0816956..40280df2a3db 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -211,9 +211,6 @@ static inline int hibernate(void) { return -ENOSYS; }
 #ifdef CONFIG_PM_SLEEP
 void save_processor_state(void);
 void restore_processor_state(void);
-struct saved_context;
-void __save_processor_state(struct saved_context *ctxt);
-void __restore_processor_state(struct saved_context *ctxt);
 
 /* kernel/power/main.c */
 extern struct blocking_notifier_head pm_chain_head;
-- 
cgit v1.2.3


From e94271017f0933b29362a3c9dea5a6b9d04d98e1 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:24 +0100
Subject: x86: adjust enable_NMI_through_LVT0()

Its previous use in a call to on_each_cpu() was pointless, as at the
time that code gets executed only one CPU is online. Further, the
function can be __cpuinit, and for this to work without
CONFIG_HOTPLUG_CPU setup_nmi() must also get an attribute (this one
can even be __init; on 64-bits check_timer() also was lacking that
attribute).

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_32.c    | 2 +-
 arch/x86/kernel/apic_64.c    | 2 +-
 arch/x86/kernel/io_apic_32.c | 4 ++--
 arch/x86/kernel/io_apic_64.c | 6 +++---
 arch/x86/kernel/smpboot_32.c | 2 +-
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/apic.h       | 2 +-
 7 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 69a13d127da3..4c014fca2057 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -152,7 +152,7 @@ u32 safe_apic_wait_icr_idle(void)
 /**
  * enable_NMI_through_LVT0 - enable NMI through local vector table 0
  */
-void enable_NMI_through_LVT0 (void * dummy)
+void __cpuinit enable_NMI_through_LVT0(void)
 {
 	unsigned int v = APIC_DM_NMI;
 
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index ab4ae50399fd..47b8ef51dde0 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -130,7 +130,7 @@ u32 safe_apic_wait_icr_idle(void)
 /**
  * enable_NMI_through_LVT0 - enable NMI through local vector table 0
  */
-void enable_NMI_through_LVT0(void *dummy)
+void __cpuinit enable_NMI_through_LVT0(void)
 {
 	unsigned int v;
 
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 76f11c3e3906..0d204237489e 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -2078,7 +2078,7 @@ static struct irq_chip lapic_chip __read_mostly = {
 	.eoi		= ack_apic,
 };
 
-static void setup_nmi (void)
+static void __init setup_nmi(void)
 {
 	/*
  	 * Dirty trick to enable the NMI watchdog ...
@@ -2091,7 +2091,7 @@ static void setup_nmi (void)
 	 */ 
 	apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ...");
 
-	on_each_cpu(enable_NMI_through_LVT0, NULL, 1, 1);
+	enable_NMI_through_LVT0();
 
 	apic_printk(APIC_VERBOSE, " done.\n");
 }
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index fa70005be5e8..f914d84a21da 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1566,7 +1566,7 @@ static struct hw_interrupt_type lapic_irq_type __read_mostly = {
 	.end = end_lapic_irq,
 };
 
-static void setup_nmi (void)
+static void __init setup_nmi(void)
 {
 	/*
  	 * Dirty trick to enable the NMI watchdog ...
@@ -1579,7 +1579,7 @@ static void setup_nmi (void)
 	 */ 
 	printk(KERN_INFO "activating NMI Watchdog ...");
 
-	enable_NMI_through_LVT0(NULL);
+	enable_NMI_through_LVT0();
 
 	printk(" done.\n");
 }
@@ -1655,7 +1655,7 @@ static inline void unlock_ExtINT_logic(void)
  *
  * FIXME: really need to revamp this for modern platforms only.
  */
-static inline void check_timer(void)
+static inline void __init check_timer(void)
 {
 	struct irq_cfg *cfg = irq_cfg + 0;
 	int apic1, pin1, apic2, pin2;
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 0f294d6e22cf..2034332ad080 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -404,7 +404,7 @@ static void __cpuinit start_secondary(void *unused)
 	setup_secondary_clock();
 	if (nmi_watchdog == NMI_IO_APIC) {
 		disable_8259A_irq(0);
-		enable_NMI_through_LVT0(NULL);
+		enable_NMI_through_LVT0();
 		enable_8259A_irq(0);
 	}
 	/*
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index cb73c4da87fc..4c03ddccd681 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -337,7 +337,7 @@ void __cpuinit start_secondary(void)
 
 	if (nmi_watchdog == NMI_IO_APIC) {
 		disable_8259A_irq(0);
-		enable_NMI_through_LVT0(NULL);
+		enable_NMI_through_LVT0();
 		enable_8259A_irq(0);
 	}
 
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 72bf09cf13ab..bcfc07fd3661 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -117,7 +117,7 @@ extern void init_apic_mappings(void);
 extern void setup_boot_APIC_clock(void);
 extern void setup_secondary_APIC_clock(void);
 extern int APIC_init_uniprocessor(void);
-extern void enable_NMI_through_LVT0(void *dummy);
+extern void enable_NMI_through_LVT0(void);
 
 /*
  * On 32bit this is mach-xxx local
-- 
cgit v1.2.3


From ded9aa0db815b4e1271001561e177755cb8b9468 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 30 Jan 2008 13:31:24 +0100
Subject: x86: also define AT_VECTOR_SIZE_ARCH

The patch introducing this left out x86-64, despite it also having
extra entries.

[ mingo@elte.hu: re-merged this to after the unification patches. ]

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 8a37dad38bc0..6c7d1fda4995 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -9,8 +9,14 @@
 #include <linux/kernel.h>
 #include <linux/irqflags.h>
 
+/* entries in ARCH_DLINFO: */
+#ifdef CONFIG_IA32_EMULATION
+# define AT_VECTOR_SIZE_ARCH 2
+#else
+# define AT_VECTOR_SIZE_ARCH 1
+#endif
+
 #ifdef CONFIG_X86_32
-#define AT_VECTOR_SIZE_ARCH 2 /* entries in ARCH_DLINFO */
 
 struct task_struct; /* one of the stranger aspects of C forward declarations */
 extern struct task_struct *FASTCALL(__switch_to(struct task_struct *prev,
@@ -56,7 +62,7 @@ extern struct task_struct *FASTCALL(__switch_to(struct task_struct *prev,
 
 /* Save restore flags to clear handle leaking NT */
 #define switch_to(prev, next, last) \
-	asm volatile(SAVE_CONTEXT					  \
+	asm volatile(SAVE_CONTEXT						    \
 	     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
 	     "movq %P[threadrsp](%[next]),%%rsp\n\t" /* restore RSP */	  \
 	     "call __switch_to\n\t"					  \
-- 
cgit v1.2.3


From fe758fb1192790f8e465a7f59efe47ca73717d3e Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@gmail.com>
Date: Wed, 30 Jan 2008 13:31:25 +0100
Subject: x86: clean up include/asm-x86/pda.h

clean up include/asm-x86/pda.h, as suggested by checkpatch.pl.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pda.h | 70 +++++++++++++++++++++++++--------------------------
 1 file changed, 35 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index b620d0c39a93..c0305bff0f19 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -7,15 +7,15 @@
 #include <linux/cache.h>
 #include <asm/page.h>
 
-/* Per processor datastructure. %gs points to it while the kernel runs */ 
+/* Per processor datastructure. %gs points to it while the kernel runs */
 struct x8664_pda {
 	struct task_struct *pcurrent;	/* 0  Current process */
 	unsigned long data_offset;	/* 8 Per cpu data offset from linker
 					   address */
-	unsigned long kernelstack;  /* 16 top of kernel stack for current */
-	unsigned long oldrsp; 	    /* 24 user rsp for system call */
-        int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
-	unsigned int cpunumber;	    /* 36 Logical CPU number */
+	unsigned long kernelstack;	/* 16 top of kernel stack for current */
+	unsigned long oldrsp;		/* 24 user rsp for system call */
+	int irqcount;			/* 32 Irq nesting counter. Starts -1 */
+	unsigned int cpunumber;		/* 36 Logical CPU number */
 #ifdef CONFIG_CC_STACKPROTECTOR
 	unsigned long stack_canary;	/* 40 stack canary value */
 					/* gcc-ABI: this canary MUST be at
@@ -44,10 +44,10 @@ extern void pda_init(int);
 
 #define cpu_pda(i) (_cpu_pda[i])
 
-/* 
+/*
  * There is no fast way to get the base address of the PDA, all the accesses
  * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
- */ 
+ */
 extern void __bad_pda_field(void) __attribute__((noreturn));
 
 /*
@@ -58,70 +58,70 @@ extern struct x8664_pda _proxy_pda;
 
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
-#define pda_to_op(op,field,val) do {		\
+#define pda_to_op(op, field, val) do {		\
 	typedef typeof(_proxy_pda.field) T__;	\
 	if (0) { T__ tmp__; tmp__ = (val); }	/* type checking */ \
 	switch (sizeof(_proxy_pda.field)) {	\
 	case 2:					\
-		asm(op "w %1,%%gs:%c2" : 	\
+		asm(op "w %1,%%gs:%c2" :	\
 		    "+m" (_proxy_pda.field) :	\
 		    "ri" ((T__)val),		\
-		    "i"(pda_offset(field))); 	\
- 		break;				\
+		    "i"(pda_offset(field)));	\
+		break;				\
 	case 4:					\
-		asm(op "l %1,%%gs:%c2" : 	\
+		asm(op "l %1,%%gs:%c2" :	\
 		    "+m" (_proxy_pda.field) :	\
 		    "ri" ((T__)val),		\
-		    "i" (pda_offset(field))); 	\
+		    "i" (pda_offset(field)));	\
 		break;				\
 	case 8:					\
-		asm(op "q %1,%%gs:%c2": 	\
+		asm(op "q %1,%%gs:%c2":		\
 		    "+m" (_proxy_pda.field) :	\
 		    "ri" ((T__)val),		\
-		    "i"(pda_offset(field))); 	\
+		    "i"(pda_offset(field)));	\
 		break;				\
-       default: 				\
+	default:				\
 		__bad_pda_field();		\
-       }					\
-       } while (0)
+	}					\
+	} while (0)
 
 #define pda_from_op(op,field) ({		\
 	typeof(_proxy_pda.field) ret__;		\
 	switch (sizeof(_proxy_pda.field)) {	\
-       	case 2:					\
-		asm(op "w %%gs:%c1,%0" : 	\
+	case 2:					\
+		asm(op "w %%gs:%c1,%0" :	\
 		    "=r" (ret__) :		\
-		    "i" (pda_offset(field)), 	\
-		    "m" (_proxy_pda.field)); 	\
+		    "i" (pda_offset(field)),	\
+		    "m" (_proxy_pda.field));	\
 		 break;				\
 	case 4:					\
 		asm(op "l %%gs:%c1,%0":		\
 		    "=r" (ret__):		\
-		    "i" (pda_offset(field)), 	\
-		    "m" (_proxy_pda.field)); 	\
+		    "i" (pda_offset(field)),	\
+		    "m" (_proxy_pda.field));	\
 		 break;				\
-       case 8:					\
+	case 8:					\
 		asm(op "q %%gs:%c1,%0":		\
 		    "=r" (ret__) :		\
-		    "i" (pda_offset(field)), 	\
-		    "m" (_proxy_pda.field)); 	\
+		    "i" (pda_offset(field)),	\
+		    "m" (_proxy_pda.field));	\
 		 break;				\
-       default: 				\
+	default:				\
 		__bad_pda_field();		\
        }					\
        ret__; })
 
-#define read_pda(field) pda_from_op("mov",field)
-#define write_pda(field,val) pda_to_op("mov",field,val)
-#define add_pda(field,val) pda_to_op("add",field,val)
-#define sub_pda(field,val) pda_to_op("sub",field,val)
-#define or_pda(field,val) pda_to_op("or",field,val)
+#define read_pda(field)		pda_from_op("mov", field)
+#define write_pda(field, val)	pda_to_op("mov", field, val)
+#define add_pda(field, val)	pda_to_op("add", field, val)
+#define sub_pda(field, val)	pda_to_op("sub", field, val)
+#define or_pda(field, val)	pda_to_op("or", field, val)
 
 /* This is not atomic against other CPUs -- CPU preemption needs to be off */
-#define test_and_clear_bit_pda(bit,field) ({		\
+#define test_and_clear_bit_pda(bit, field) ({		\
 	int old__;						\
 	asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0"		\
-	    : "=r" (old__), "+m" (_proxy_pda.field) 		\
+	    : "=r" (old__), "+m" (_proxy_pda.field)		\
 	    : "dIr" (bit), "i" (pda_offset(field)) : "memory");	\
 	old__;							\
 })
-- 
cgit v1.2.3


From 4ad02718345783a3b84bae1895917666b779850d Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: clean up local_{32|64}.h

Common prefix from both files moved to local.h

Change __inline__ to inline

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/local_32.h | 34 ++++++++++------------------------
 include/asm-x86/local_64.h | 25 ++++++-------------------
 2 files changed, 16 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/local_32.h b/include/asm-x86/local_32.h
index 6e85975b9ed2..33d9c7bf463c 100644
--- a/include/asm-x86/local_32.h
+++ b/include/asm-x86/local_32.h
@@ -1,35 +1,21 @@
 #ifndef _ARCH_I386_LOCAL_H
 #define _ARCH_I386_LOCAL_H
 
-#include <linux/percpu.h>
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-typedef struct
-{
-	atomic_long_t a;
-} local_t;
-
-#define LOCAL_INIT(i)	{ ATOMIC_LONG_INIT(i) }
-
-#define local_read(l)	atomic_long_read(&(l)->a)
-#define local_set(l,i)	atomic_long_set(&(l)->a, (i))
-
-static __inline__ void local_inc(local_t *l)
+static inline void local_inc(local_t *l)
 {
 	__asm__ __volatile__(
 		"incl %0"
 		:"+m" (l->a.counter));
 }
 
-static __inline__ void local_dec(local_t *l)
+static inline void local_dec(local_t *l)
 {
 	__asm__ __volatile__(
 		"decl %0"
 		:"+m" (l->a.counter));
 }
 
-static __inline__ void local_add(long i, local_t *l)
+static inline void local_add(long i, local_t *l)
 {
 	__asm__ __volatile__(
 		"addl %1,%0"
@@ -37,7 +23,7 @@ static __inline__ void local_add(long i, local_t *l)
 		:"ir" (i));
 }
 
-static __inline__ void local_sub(long i, local_t *l)
+static inline void local_sub(long i, local_t *l)
 {
 	__asm__ __volatile__(
 		"subl %1,%0"
@@ -54,7 +40,7 @@ static __inline__ void local_sub(long i, local_t *l)
  * true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int local_sub_and_test(long i, local_t *l)
+static inline int local_sub_and_test(long i, local_t *l)
 {
 	unsigned char c;
 
@@ -73,7 +59,7 @@ static __inline__ int local_sub_and_test(long i, local_t *l)
  * returns true if the result is 0, or false for all other
  * cases.
  */
-static __inline__ int local_dec_and_test(local_t *l)
+static inline int local_dec_and_test(local_t *l)
 {
 	unsigned char c;
 
@@ -92,7 +78,7 @@ static __inline__ int local_dec_and_test(local_t *l)
  * and returns true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int local_inc_and_test(local_t *l)
+static inline int local_inc_and_test(local_t *l)
 {
 	unsigned char c;
 
@@ -112,7 +98,7 @@ static __inline__ int local_inc_and_test(local_t *l)
  * if the result is negative, or false when
  * result is greater than or equal to zero.
  */
-static __inline__ int local_add_negative(long i, local_t *l)
+static inline int local_add_negative(long i, local_t *l)
 {
 	unsigned char c;
 
@@ -130,7 +116,7 @@ static __inline__ int local_add_negative(long i, local_t *l)
  *
  * Atomically adds @i to @l and returns @i + @l
  */
-static __inline__ long local_add_return(long i, local_t *l)
+static inline long local_add_return(long i, local_t *l)
 {
 	long __i;
 #ifdef CONFIG_M386
@@ -156,7 +142,7 @@ no_xadd: /* Legacy 386 processor */
 #endif
 }
 
-static __inline__ long local_sub_return(long i, local_t *l)
+static inline long local_sub_return(long i, local_t *l)
 {
 	return local_add_return(-i,l);
 }
diff --git a/include/asm-x86/local_64.h b/include/asm-x86/local_64.h
index e87492bb0693..92330f8135f8 100644
--- a/include/asm-x86/local_64.h
+++ b/include/asm-x86/local_64.h
@@ -1,19 +1,6 @@
 #ifndef _ARCH_X8664_LOCAL_H
 #define _ARCH_X8664_LOCAL_H
 
-#include <linux/percpu.h>
-#include <asm/atomic.h>
-
-typedef struct
-{
-	atomic_long_t a;
-} local_t;
-
-#define LOCAL_INIT(i)	{ ATOMIC_LONG_INIT(i) }
-
-#define local_read(l)	atomic_long_read(&(l)->a)
-#define local_set(l,i)	atomic_long_set(&(l)->a, (i))
-
 static inline void local_inc(local_t *l)
 {
 	__asm__ __volatile__(
@@ -55,7 +42,7 @@ static inline void local_sub(long i, local_t *l)
  * true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int local_sub_and_test(long i, local_t *l)
+static inline int local_sub_and_test(long i, local_t *l)
 {
 	unsigned char c;
 
@@ -74,7 +61,7 @@ static __inline__ int local_sub_and_test(long i, local_t *l)
  * returns true if the result is 0, or false for all other
  * cases.
  */
-static __inline__ int local_dec_and_test(local_t *l)
+static inline int local_dec_and_test(local_t *l)
 {
 	unsigned char c;
 
@@ -93,7 +80,7 @@ static __inline__ int local_dec_and_test(local_t *l)
  * and returns true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int local_inc_and_test(local_t *l)
+static inline int local_inc_and_test(local_t *l)
 {
 	unsigned char c;
 
@@ -113,7 +100,7 @@ static __inline__ int local_inc_and_test(local_t *l)
  * if the result is negative, or false when
  * result is greater than or equal to zero.
  */
-static __inline__ int local_add_negative(long i, local_t *l)
+static inline int local_add_negative(long i, local_t *l)
 {
 	unsigned char c;
 
@@ -131,7 +118,7 @@ static __inline__ int local_add_negative(long i, local_t *l)
  *
  * Atomically adds @i to @l and returns @i + @l
  */
-static __inline__ long local_add_return(long i, local_t *l)
+static inline long local_add_return(long i, local_t *l)
 {
 	long __i = i;
 	__asm__ __volatile__(
@@ -141,7 +128,7 @@ static __inline__ long local_add_return(long i, local_t *l)
 	return i + __i;
 }
 
-static __inline__ long local_sub_return(long i, local_t *l)
+static inline long local_sub_return(long i, local_t *l)
 {
 	return local_add_return(-i,l);
 }
-- 
cgit v1.2.3


From 992b95920a311db3267659ea17160e4812a05830 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: fix asm memory constraints in local_64.h

Use the shorter +m form rather than =m and m.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/local_64.h | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/local_64.h b/include/asm-x86/local_64.h
index 92330f8135f8..50e99eddd628 100644
--- a/include/asm-x86/local_64.h
+++ b/include/asm-x86/local_64.h
@@ -5,32 +5,30 @@ static inline void local_inc(local_t *l)
 {
 	__asm__ __volatile__(
 		"incq %0"
-		:"=m" (l->a.counter)
-		:"m" (l->a.counter));
+		:"+m" (l->a.counter));
 }
 
 static inline void local_dec(local_t *l)
 {
 	__asm__ __volatile__(
 		"decq %0"
-		:"=m" (l->a.counter)
-		:"m" (l->a.counter));
+		:"+m" (l->a.counter));
 }
 
 static inline void local_add(long i, local_t *l)
 {
 	__asm__ __volatile__(
 		"addq %1,%0"
-		:"=m" (l->a.counter)
-		:"ir" (i), "m" (l->a.counter));
+		:"+m" (l->a.counter)
+		:"ir" (i));
 }
 
 static inline void local_sub(long i, local_t *l)
 {
 	__asm__ __volatile__(
 		"subq %1,%0"
-		:"=m" (l->a.counter)
-		:"ir" (i), "m" (l->a.counter));
+		:"+m" (l->a.counter)
+		:"ir" (i));
 }
 
 /**
@@ -48,8 +46,8 @@ static inline int local_sub_and_test(long i, local_t *l)
 
 	__asm__ __volatile__(
 		"subq %2,%0; sete %1"
-		:"=m" (l->a.counter), "=qm" (c)
-		:"ir" (i), "m" (l->a.counter) : "memory");
+		:"+m" (l->a.counter), "=qm" (c)
+		:"ir" (i) : "memory");
 	return c;
 }
 
@@ -67,8 +65,8 @@ static inline int local_dec_and_test(local_t *l)
 
 	__asm__ __volatile__(
 		"decq %0; sete %1"
-		:"=m" (l->a.counter), "=qm" (c)
-		:"m" (l->a.counter) : "memory");
+		:"+m" (l->a.counter), "=qm" (c)
+		: : "memory");
 	return c != 0;
 }
 
@@ -86,8 +84,8 @@ static inline int local_inc_and_test(local_t *l)
 
 	__asm__ __volatile__(
 		"incq %0; sete %1"
-		:"=m" (l->a.counter), "=qm" (c)
-		:"m" (l->a.counter) : "memory");
+		:"+m" (l->a.counter), "=qm" (c)
+		: : "memory");
 	return c != 0;
 }
 
@@ -106,8 +104,8 @@ static inline int local_add_negative(long i, local_t *l)
 
 	__asm__ __volatile__(
 		"addq %2,%0; sets %1"
-		:"=m" (l->a.counter), "=qm" (c)
-		:"ir" (i), "m" (l->a.counter) : "memory");
+		:"+m" (l->a.counter), "=qm" (c)
+		:"ir" (i) : "memory");
 	return c;
 }
 
-- 
cgit v1.2.3


From 8ee5797a91bdb713b4031741b33bd035f9c43870 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: introduce asm helpers in local_{32|64}.h

Handle the use of long on X86_32 and quad on X86_64

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/asm.h      | 12 ++++++++++++
 include/asm-x86/local_32.h | 18 +++++++++---------
 include/asm-x86/local_64.h | 18 +++++++++---------
 3 files changed, 30 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/asm.h b/include/asm-x86/asm.h
index 8661ae75488a..1a6980a60fc6 100644
--- a/include/asm-x86/asm.h
+++ b/include/asm-x86/asm.h
@@ -8,6 +8,12 @@
 # define _ASM_ALIGN	" .balign 4 "
 # define _ASM_MOV_UL	" movl "
 
+# define _ASM_INC	" incl "
+# define _ASM_DEC	" decl "
+# define _ASM_ADD	" addl "
+# define _ASM_SUB	" subl "
+# define _ASM_XADD	" xaddl "
+
 #else
 /* 64 bits */
 
@@ -15,6 +21,12 @@
 # define _ASM_ALIGN	" .balign 8 "
 # define _ASM_MOV_UL	" movq "
 
+# define _ASM_INC	" incq "
+# define _ASM_DEC	" decq "
+# define _ASM_ADD	" addq "
+# define _ASM_SUB	" subq "
+# define _ASM_XADD	" xaddq "
+
 #endif /* CONFIG_X86_32 */
 
 #endif /* _ASM_X86_ASM_H */
diff --git a/include/asm-x86/local_32.h b/include/asm-x86/local_32.h
index 33d9c7bf463c..c219fe56b3df 100644
--- a/include/asm-x86/local_32.h
+++ b/include/asm-x86/local_32.h
@@ -4,21 +4,21 @@
 static inline void local_inc(local_t *l)
 {
 	__asm__ __volatile__(
-		"incl %0"
+		_ASM_INC "%0"
 		:"+m" (l->a.counter));
 }
 
 static inline void local_dec(local_t *l)
 {
 	__asm__ __volatile__(
-		"decl %0"
+		_ASM_DEC "%0"
 		:"+m" (l->a.counter));
 }
 
 static inline void local_add(long i, local_t *l)
 {
 	__asm__ __volatile__(
-		"addl %1,%0"
+		_ASM_ADD "%1,%0"
 		:"+m" (l->a.counter)
 		:"ir" (i));
 }
@@ -26,7 +26,7 @@ static inline void local_add(long i, local_t *l)
 static inline void local_sub(long i, local_t *l)
 {
 	__asm__ __volatile__(
-		"subl %1,%0"
+		_ASM_SUB "%1,%0"
 		:"+m" (l->a.counter)
 		:"ir" (i));
 }
@@ -45,7 +45,7 @@ static inline int local_sub_and_test(long i, local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"subl %2,%0; sete %1"
+		_ASM_SUB "%2,%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		:"ir" (i) : "memory");
 	return c;
@@ -64,7 +64,7 @@ static inline int local_dec_and_test(local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"decl %0; sete %1"
+		_ASM_DEC "%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		: : "memory");
 	return c != 0;
@@ -83,7 +83,7 @@ static inline int local_inc_and_test(local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"incl %0; sete %1"
+		_ASM_INC "%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		: : "memory");
 	return c != 0;
@@ -103,7 +103,7 @@ static inline int local_add_negative(long i, local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"addl %2,%0; sets %1"
+		_ASM_ADD "%2,%0; sets %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		:"ir" (i) : "memory");
 	return c;
@@ -127,7 +127,7 @@ static inline long local_add_return(long i, local_t *l)
 	/* Modern 486+ processor */
 	__i = i;
 	__asm__ __volatile__(
-		"xaddl %0, %1;"
+		_ASM_XADD "%0, %1;"
 		:"+r" (i), "+m" (l->a.counter)
 		: : "memory");
 	return i + __i;
diff --git a/include/asm-x86/local_64.h b/include/asm-x86/local_64.h
index 50e99eddd628..d685cd7e014f 100644
--- a/include/asm-x86/local_64.h
+++ b/include/asm-x86/local_64.h
@@ -4,21 +4,21 @@
 static inline void local_inc(local_t *l)
 {
 	__asm__ __volatile__(
-		"incq %0"
+		_ASM_INC "%0"
 		:"+m" (l->a.counter));
 }
 
 static inline void local_dec(local_t *l)
 {
 	__asm__ __volatile__(
-		"decq %0"
+		_ASM_DEC "%0"
 		:"+m" (l->a.counter));
 }
 
 static inline void local_add(long i, local_t *l)
 {
 	__asm__ __volatile__(
-		"addq %1,%0"
+		_ASM_ADD "%1,%0"
 		:"+m" (l->a.counter)
 		:"ir" (i));
 }
@@ -26,7 +26,7 @@ static inline void local_add(long i, local_t *l)
 static inline void local_sub(long i, local_t *l)
 {
 	__asm__ __volatile__(
-		"subq %1,%0"
+		_ASM_SUB "%1,%0"
 		:"+m" (l->a.counter)
 		:"ir" (i));
 }
@@ -45,7 +45,7 @@ static inline int local_sub_and_test(long i, local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"subq %2,%0; sete %1"
+		_ASM_SUB "%2,%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		:"ir" (i) : "memory");
 	return c;
@@ -64,7 +64,7 @@ static inline int local_dec_and_test(local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"decq %0; sete %1"
+		_ASM_DEC "%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		: : "memory");
 	return c != 0;
@@ -83,7 +83,7 @@ static inline int local_inc_and_test(local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"incq %0; sete %1"
+		_ASM_INC "%0; sete %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		: : "memory");
 	return c != 0;
@@ -103,7 +103,7 @@ static inline int local_add_negative(long i, local_t *l)
 	unsigned char c;
 
 	__asm__ __volatile__(
-		"addq %2,%0; sets %1"
+		_ASM_ADD "%2,%0; sets %1"
 		:"+m" (l->a.counter), "=qm" (c)
 		:"ir" (i) : "memory");
 	return c;
@@ -120,7 +120,7 @@ static inline long local_add_return(long i, local_t *l)
 {
 	long __i = i;
 	__asm__ __volatile__(
-		"xaddq %0, %1;"
+		_ASM_XADD "%0, %1;"
 		:"+r" (i), "+m" (l->a.counter)
 		: : "memory");
 	return i + __i;
-- 
cgit v1.2.3


From 5638f99394d478bcba6384ac7c51e763d1a3a448 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: unify local_{32|64}.h

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/local.h    | 244 ++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/local_32.h | 219 ----------------------------------------
 include/asm-x86/local_64.h | 207 --------------------------------------
 3 files changed, 240 insertions(+), 430 deletions(-)
 delete mode 100644 include/asm-x86/local_32.h
 delete mode 100644 include/asm-x86/local_64.h

(limited to 'include')

diff --git a/include/asm-x86/local.h b/include/asm-x86/local.h
index c7a1b1c66c96..f5677e208601 100644
--- a/include/asm-x86/local.h
+++ b/include/asm-x86/local.h
@@ -1,5 +1,241 @@
-#ifdef CONFIG_X86_32
-# include "local_32.h"
-#else
-# include "local_64.h"
+#ifndef _ARCH_LOCAL_H
+#define _ARCH_LOCAL_H
+
+#include <linux/percpu.h>
+
+#include <asm/system.h>
+#include <asm/atomic.h>
+#include <asm/asm.h>
+
+typedef struct
+{
+	atomic_long_t a;
+} local_t;
+
+#define LOCAL_INIT(i)	{ ATOMIC_LONG_INIT(i) }
+
+#define local_read(l)	atomic_long_read(&(l)->a)
+#define local_set(l,i)	atomic_long_set(&(l)->a, (i))
+
+static inline void local_inc(local_t *l)
+{
+	__asm__ __volatile__(
+		_ASM_INC "%0"
+		:"+m" (l->a.counter));
+}
+
+static inline void local_dec(local_t *l)
+{
+	__asm__ __volatile__(
+		_ASM_DEC "%0"
+		:"+m" (l->a.counter));
+}
+
+static inline void local_add(long i, local_t *l)
+{
+	__asm__ __volatile__(
+		_ASM_ADD "%1,%0"
+		:"+m" (l->a.counter)
+		:"ir" (i));
+}
+
+static inline void local_sub(long i, local_t *l)
+{
+	__asm__ __volatile__(
+		_ASM_SUB "%1,%0"
+		:"+m" (l->a.counter)
+		:"ir" (i));
+}
+
+/**
+ * local_sub_and_test - subtract value from variable and test result
+ * @i: integer value to subtract
+ * @l: pointer to type local_t
+ *
+ * Atomically subtracts @i from @l and returns
+ * true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int local_sub_and_test(long i, local_t *l)
+{
+	unsigned char c;
+
+	__asm__ __volatile__(
+		_ASM_SUB "%2,%0; sete %1"
+		:"+m" (l->a.counter), "=qm" (c)
+		:"ir" (i) : "memory");
+	return c;
+}
+
+/**
+ * local_dec_and_test - decrement and test
+ * @l: pointer to type local_t
+ *
+ * Atomically decrements @l by 1 and
+ * returns true if the result is 0, or false for all other
+ * cases.
+ */
+static inline int local_dec_and_test(local_t *l)
+{
+	unsigned char c;
+
+	__asm__ __volatile__(
+		_ASM_DEC "%0; sete %1"
+		:"+m" (l->a.counter), "=qm" (c)
+		: : "memory");
+	return c != 0;
+}
+
+/**
+ * local_inc_and_test - increment and test
+ * @l: pointer to type local_t
+ *
+ * Atomically increments @l by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+static inline int local_inc_and_test(local_t *l)
+{
+	unsigned char c;
+
+	__asm__ __volatile__(
+		_ASM_INC "%0; sete %1"
+		:"+m" (l->a.counter), "=qm" (c)
+		: : "memory");
+	return c != 0;
+}
+
+/**
+ * local_add_negative - add and test if negative
+ * @i: integer value to add
+ * @l: pointer to type local_t
+ *
+ * Atomically adds @i to @l and returns true
+ * if the result is negative, or false when
+ * result is greater than or equal to zero.
+ */
+static inline int local_add_negative(long i, local_t *l)
+{
+	unsigned char c;
+
+	__asm__ __volatile__(
+		_ASM_ADD "%2,%0; sets %1"
+		:"+m" (l->a.counter), "=qm" (c)
+		:"ir" (i) : "memory");
+	return c;
+}
+
+/**
+ * local_add_return - add and return
+ * @i: integer value to add
+ * @l: pointer to type local_t
+ *
+ * Atomically adds @i to @l and returns @i + @l
+ */
+static inline long local_add_return(long i, local_t *l)
+{
+	long __i;
+#ifdef CONFIG_M386
+	unsigned long flags;
+	if(unlikely(boot_cpu_data.x86 <= 3))
+		goto no_xadd;
 #endif
+	/* Modern 486+ processor */
+	__i = i;
+	__asm__ __volatile__(
+		_ASM_XADD "%0, %1;"
+		:"+r" (i), "+m" (l->a.counter)
+		: : "memory");
+	return i + __i;
+
+#ifdef CONFIG_M386
+no_xadd: /* Legacy 386 processor */
+	local_irq_save(flags);
+	__i = local_read(l);
+	local_set(l, i + __i);
+	local_irq_restore(flags);
+	return i + __i;
+#endif
+}
+
+static inline long local_sub_return(long i, local_t *l)
+{
+	return local_add_return(-i,l);
+}
+
+#define local_inc_return(l)  (local_add_return(1,l))
+#define local_dec_return(l)  (local_sub_return(1,l))
+
+#define local_cmpxchg(l, o, n) \
+	(cmpxchg_local(&((l)->a.counter), (o), (n)))
+/* Always has a lock prefix */
+#define local_xchg(l, n) (xchg(&((l)->a.counter), (n)))
+
+/**
+ * local_add_unless - add unless the number is a given value
+ * @l: pointer of type local_t
+ * @a: the amount to add to l...
+ * @u: ...unless l is equal to u.
+ *
+ * Atomically adds @a to @l, so long as it was not @u.
+ * Returns non-zero if @l was not @u, and zero otherwise.
+ */
+#define local_add_unless(l, a, u)				\
+({								\
+	long c, old;						\
+	c = local_read(l);					\
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = local_cmpxchg((l), c, c + (a));	\
+		if (likely(old == c))				\
+			break;					\
+		c = old;					\
+	}							\
+	c != (u);						\
+})
+#define local_inc_not_zero(l) local_add_unless((l), 1, 0)
+
+/* On x86_32, these are no better than the atomic variants.
+ * On x86-64 these are better than the atomic variants on SMP kernels
+ * because they dont use a lock prefix.
+ */
+#define __local_inc(l)		local_inc(l)
+#define __local_dec(l)		local_dec(l)
+#define __local_add(i,l)	local_add((i),(l))
+#define __local_sub(i,l)	local_sub((i),(l))
+
+/* Use these for per-cpu local_t variables: on some archs they are
+ * much more efficient than these naive implementations.  Note they take
+ * a variable, not an address.
+ *
+ * X86_64: This could be done better if we moved the per cpu data directly
+ * after GS.
+ */
+
+/* Need to disable preemption for the cpu local counters otherwise we could
+   still access a variable of a previous CPU in a non atomic way. */
+#define cpu_local_wrap_v(l)	 	\
+	({ local_t res__;		\
+	   preempt_disable(); 		\
+	   res__ = (l);			\
+	   preempt_enable();		\
+	   res__; })
+#define cpu_local_wrap(l)		\
+	({ preempt_disable();		\
+	   l;				\
+	   preempt_enable(); })		\
+
+#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
+#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
+#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
+#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
+#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
+#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
+
+#define __cpu_local_inc(l)	cpu_local_inc(l)
+#define __cpu_local_dec(l)	cpu_local_dec(l)
+#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
+#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
+
+#endif /* _ARCH_LOCAL_H */
diff --git a/include/asm-x86/local_32.h b/include/asm-x86/local_32.h
deleted file mode 100644
index c219fe56b3df..000000000000
--- a/include/asm-x86/local_32.h
+++ /dev/null
@@ -1,219 +0,0 @@
-#ifndef _ARCH_I386_LOCAL_H
-#define _ARCH_I386_LOCAL_H
-
-static inline void local_inc(local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_INC "%0"
-		:"+m" (l->a.counter));
-}
-
-static inline void local_dec(local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_DEC "%0"
-		:"+m" (l->a.counter));
-}
-
-static inline void local_add(long i, local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_ADD "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
-}
-
-static inline void local_sub(long i, local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_SUB "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
-}
-
-/**
- * local_sub_and_test - subtract value from variable and test result
- * @i: integer value to subtract
- * @l: pointer of type local_t
- *
- * Atomically subtracts @i from @l and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-static inline int local_sub_and_test(long i, local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_SUB "%2,%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
-	return c;
-}
-
-/**
- * local_dec_and_test - decrement and test
- * @l: pointer of type local_t
- *
- * Atomically decrements @l by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-static inline int local_dec_and_test(local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_DEC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
-	return c != 0;
-}
-
-/**
- * local_inc_and_test - increment and test
- * @l: pointer of type local_t
- *
- * Atomically increments @l by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-static inline int local_inc_and_test(local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_INC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
-	return c != 0;
-}
-
-/**
- * local_add_negative - add and test if negative
- * @l: pointer of type local_t
- * @i: integer value to add
- *
- * Atomically adds @i to @l and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-static inline int local_add_negative(long i, local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_ADD "%2,%0; sets %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
-	return c;
-}
-
-/**
- * local_add_return - add and return
- * @l: pointer of type local_t
- * @i: integer value to add
- *
- * Atomically adds @i to @l and returns @i + @l
- */
-static inline long local_add_return(long i, local_t *l)
-{
-	long __i;
-#ifdef CONFIG_M386
-	unsigned long flags;
-	if(unlikely(boot_cpu_data.x86 <= 3))
-		goto no_xadd;
-#endif
-	/* Modern 486+ processor */
-	__i = i;
-	__asm__ __volatile__(
-		_ASM_XADD "%0, %1;"
-		:"+r" (i), "+m" (l->a.counter)
-		: : "memory");
-	return i + __i;
-
-#ifdef CONFIG_M386
-no_xadd: /* Legacy 386 processor */
-	local_irq_save(flags);
-	__i = local_read(l);
-	local_set(l, i + __i);
-	local_irq_restore(flags);
-	return i + __i;
-#endif
-}
-
-static inline long local_sub_return(long i, local_t *l)
-{
-	return local_add_return(-i,l);
-}
-
-#define local_inc_return(l)  (local_add_return(1,l))
-#define local_dec_return(l)  (local_sub_return(1,l))
-
-#define local_cmpxchg(l, o, n) \
-	(cmpxchg_local(&((l)->a.counter), (o), (n)))
-/* Always has a lock prefix */
-#define local_xchg(l, n) (xchg(&((l)->a.counter), (n)))
-
-/**
- * local_add_unless - add unless the number is a given value
- * @l: pointer of type local_t
- * @a: the amount to add to l...
- * @u: ...unless l is equal to u.
- *
- * Atomically adds @a to @l, so long as it was not @u.
- * Returns non-zero if @l was not @u, and zero otherwise.
- */
-#define local_add_unless(l, a, u)				\
-({								\
-	long c, old;						\
-	c = local_read(l);					\
-	for (;;) {						\
-		if (unlikely(c == (u)))				\
-			break;					\
-		old = local_cmpxchg((l), c, c + (a));	\
-		if (likely(old == c))				\
-			break;					\
-		c = old;					\
-	}							\
-	c != (u);						\
-})
-#define local_inc_not_zero(l) local_add_unless((l), 1, 0)
-
-/* On x86, these are no better than the atomic variants. */
-#define __local_inc(l)		local_inc(l)
-#define __local_dec(l)		local_dec(l)
-#define __local_add(i,l)	local_add((i),(l))
-#define __local_sub(i,l)	local_sub((i),(l))
-
-/* Use these for per-cpu local_t variables: on some archs they are
- * much more efficient than these naive implementations.  Note they take
- * a variable, not an address.
- */
-
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
-#define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
-#endif /* _ARCH_I386_LOCAL_H */
diff --git a/include/asm-x86/local_64.h b/include/asm-x86/local_64.h
deleted file mode 100644
index d685cd7e014f..000000000000
--- a/include/asm-x86/local_64.h
+++ /dev/null
@@ -1,207 +0,0 @@
-#ifndef _ARCH_X8664_LOCAL_H
-#define _ARCH_X8664_LOCAL_H
-
-static inline void local_inc(local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_INC "%0"
-		:"+m" (l->a.counter));
-}
-
-static inline void local_dec(local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_DEC "%0"
-		:"+m" (l->a.counter));
-}
-
-static inline void local_add(long i, local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_ADD "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
-}
-
-static inline void local_sub(long i, local_t *l)
-{
-	__asm__ __volatile__(
-		_ASM_SUB "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
-}
-
-/**
- * local_sub_and_test - subtract value from variable and test result
- * @i: integer value to subtract
- * @l: pointer to type local_t
- *
- * Atomically subtracts @i from @l and returns
- * true if the result is zero, or false for all
- * other cases.
- */
-static inline int local_sub_and_test(long i, local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_SUB "%2,%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
-	return c;
-}
-
-/**
- * local_dec_and_test - decrement and test
- * @l: pointer to type local_t
- *
- * Atomically decrements @l by 1 and
- * returns true if the result is 0, or false for all other
- * cases.
- */
-static inline int local_dec_and_test(local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_DEC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
-	return c != 0;
-}
-
-/**
- * local_inc_and_test - increment and test
- * @l: pointer to type local_t
- *
- * Atomically increments @l by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-static inline int local_inc_and_test(local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_INC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
-	return c != 0;
-}
-
-/**
- * local_add_negative - add and test if negative
- * @i: integer value to add
- * @l: pointer to type local_t
- *
- * Atomically adds @i to @l and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
- */
-static inline int local_add_negative(long i, local_t *l)
-{
-	unsigned char c;
-
-	__asm__ __volatile__(
-		_ASM_ADD "%2,%0; sets %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
-	return c;
-}
-
-/**
- * local_add_return - add and return
- * @i: integer value to add
- * @l: pointer to type local_t
- *
- * Atomically adds @i to @l and returns @i + @l
- */
-static inline long local_add_return(long i, local_t *l)
-{
-	long __i = i;
-	__asm__ __volatile__(
-		_ASM_XADD "%0, %1;"
-		:"+r" (i), "+m" (l->a.counter)
-		: : "memory");
-	return i + __i;
-}
-
-static inline long local_sub_return(long i, local_t *l)
-{
-	return local_add_return(-i,l);
-}
-
-#define local_inc_return(l)  (local_add_return(1,l))
-#define local_dec_return(l)  (local_sub_return(1,l))
-
-#define local_cmpxchg(l, o, n) \
-	(cmpxchg_local(&((l)->a.counter), (o), (n)))
-/* Always has a lock prefix */
-#define local_xchg(l, n) (xchg(&((l)->a.counter), (n)))
-
-/**
- * atomic_up_add_unless - add unless the number is a given value
- * @l: pointer of type local_t
- * @a: the amount to add to l...
- * @u: ...unless l is equal to u.
- *
- * Atomically adds @a to @l, so long as it was not @u.
- * Returns non-zero if @l was not @u, and zero otherwise.
- */
-#define local_add_unless(l, a, u)				\
-({								\
-	long c, old;						\
-	c = local_read(l);					\
-	for (;;) {						\
-		if (unlikely(c == (u)))				\
-			break;					\
-		old = local_cmpxchg((l), c, c + (a));	\
-		if (likely(old == c))				\
-			break;					\
-		c = old;					\
-	}							\
-	c != (u);						\
-})
-#define local_inc_not_zero(l) local_add_unless((l), 1, 0)
-
-/* On x86-64 these are better than the atomic variants on SMP kernels
-   because they dont use a lock prefix. */
-#define __local_inc(l)		local_inc(l)
-#define __local_dec(l)		local_dec(l)
-#define __local_add(i,l)	local_add((i),(l))
-#define __local_sub(i,l)	local_sub((i),(l))
-
-/* Use these for per-cpu local_t variables: on some archs they are
- * much more efficient than these naive implementations.  Note they take
- * a variable, not an address.
- *
- * This could be done better if we moved the per cpu data directly
- * after GS.
- */
-
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
-#define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
-#endif /* _ARCH_X8664_LOCAL_H */
-- 
cgit v1.2.3


From 01c57fb6a3eefc34da17be1399453c0800bcff68 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: local.h fix checkpatch warnings

Mostly space after comma, one space after if.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/local.h | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/local.h b/include/asm-x86/local.h
index f5677e208601..f852c62b3319 100644
--- a/include/asm-x86/local.h
+++ b/include/asm-x86/local.h
@@ -7,15 +7,14 @@
 #include <asm/atomic.h>
 #include <asm/asm.h>
 
-typedef struct
-{
+typedef struct {
 	atomic_long_t a;
 } local_t;
 
 #define LOCAL_INIT(i)	{ ATOMIC_LONG_INIT(i) }
 
 #define local_read(l)	atomic_long_read(&(l)->a)
-#define local_set(l,i)	atomic_long_set(&(l)->a, (i))
+#define local_set(l, i)	atomic_long_set(&(l)->a, (i))
 
 static inline void local_inc(local_t *l)
 {
@@ -137,7 +136,7 @@ static inline long local_add_return(long i, local_t *l)
 	long __i;
 #ifdef CONFIG_M386
 	unsigned long flags;
-	if(unlikely(boot_cpu_data.x86 <= 3))
+	if (unlikely(boot_cpu_data.x86 <= 3))
 		goto no_xadd;
 #endif
 	/* Modern 486+ processor */
@@ -160,11 +159,11 @@ no_xadd: /* Legacy 386 processor */
 
 static inline long local_sub_return(long i, local_t *l)
 {
-	return local_add_return(-i,l);
+	return local_add_return(-i, l);
 }
 
-#define local_inc_return(l)  (local_add_return(1,l))
-#define local_dec_return(l)  (local_sub_return(1,l))
+#define local_inc_return(l)  (local_add_return(1, l))
+#define local_dec_return(l)  (local_sub_return(1, l))
 
 #define local_cmpxchg(l, o, n) \
 	(cmpxchg_local(&((l)->a.counter), (o), (n)))
@@ -202,8 +201,8 @@ static inline long local_sub_return(long i, local_t *l)
  */
 #define __local_inc(l)		local_inc(l)
 #define __local_dec(l)		local_dec(l)
-#define __local_add(i,l)	local_add((i),(l))
-#define __local_sub(i,l)	local_sub((i),(l))
+#define __local_add(i, l)	local_add((i), (l))
+#define __local_sub(i, l)	local_sub((i), (l))
 
 /* Use these for per-cpu local_t variables: on some archs they are
  * much more efficient than these naive implementations.  Note they take
-- 
cgit v1.2.3


From 3c233d1334ffc8de63a4b6a6a86c40961aed335e Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: unify kexec_{32|64}.h

One section collecting all constant defines.  Ifdef the asm
blocks for X86_32/64.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vivek Goyal <vgoyal@redhat.com>
---
 include/asm-x86/kexec.h    | 169 ++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/kexec_32.h |  99 --------------------------
 include/asm-x86/kexec_64.h |  94 -------------------------
 3 files changed, 167 insertions(+), 195 deletions(-)
 delete mode 100644 include/asm-x86/kexec_32.h
 delete mode 100644 include/asm-x86/kexec_64.h

(limited to 'include')

diff --git a/include/asm-x86/kexec.h b/include/asm-x86/kexec.h
index 718ddbfb9516..c90d3c77afc2 100644
--- a/include/asm-x86/kexec.h
+++ b/include/asm-x86/kexec.h
@@ -1,5 +1,170 @@
+#ifndef _KEXEC_H
+#define _KEXEC_H
+
 #ifdef CONFIG_X86_32
-# include "kexec_32.h"
+# define PA_CONTROL_PAGE	0
+# define VA_CONTROL_PAGE	1
+# define PA_PGD			2
+# define VA_PGD			3
+# define PA_PTE_0		4
+# define VA_PTE_0		5
+# define PA_PTE_1		6
+# define VA_PTE_1		7
+# ifdef CONFIG_X86_PAE
+#  define PA_PMD_0		8
+#  define VA_PMD_0		9
+#  define PA_PMD_1		10
+#  define VA_PMD_1		11
+#  define PAGES_NR		12
+# else
+#  define PAGES_NR		8
+# endif
 #else
-# include "kexec_64.h"
+# define PA_CONTROL_PAGE	0
+# define VA_CONTROL_PAGE	1
+# define PA_PGD			2
+# define VA_PGD			3
+# define PA_PUD_0		4
+# define VA_PUD_0		5
+# define PA_PMD_0		6
+# define VA_PMD_0		7
+# define PA_PTE_0		8
+# define VA_PTE_0		9
+# define PA_PUD_1		10
+# define VA_PUD_1		11
+# define PA_PMD_1		12
+# define VA_PMD_1		13
+# define PA_PTE_1		14
+# define VA_PTE_1		15
+# define PA_TABLE_PAGE		16
+# define PAGES_NR		17
 #endif
+
+#ifndef __ASSEMBLY__
+
+#include <linux/string.h>
+
+#include <asm/page.h>
+#include <asm/ptrace.h>
+
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ *
+ * So far x86_64 is limited to 40 physical address bits.
+ */
+#ifdef CONFIG_X86_32
+/* Maximum physical address we can use pages from */
+# define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+/* Maximum address we can reach in physical address mode */
+# define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+/* Maximum address we can use for the control code buffer */
+# define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE
+
+# define KEXEC_CONTROL_CODE_SIZE	4096
+
+/* The native architecture */
+# define KEXEC_ARCH KEXEC_ARCH_386
+
+/* We can also handle crash dumps from 64 bit kernel. */
+# define vmcore_elf_check_arch_cross(x) ((x)->e_machine == EM_X86_64)
+#else
+/* Maximum physical address we can use pages from */
+# define KEXEC_SOURCE_MEMORY_LIMIT      (0xFFFFFFFFFFUL)
+/* Maximum address we can reach in physical address mode */
+# define KEXEC_DESTINATION_MEMORY_LIMIT (0xFFFFFFFFFFUL)
+/* Maximum address we can use for the control pages */
+# define KEXEC_CONTROL_MEMORY_LIMIT     (0xFFFFFFFFFFUL)
+
+/* Allocate one page for the pdp and the second for the code */
+# define KEXEC_CONTROL_CODE_SIZE  (4096UL + 4096UL)
+
+/* The native architecture */
+# define KEXEC_ARCH KEXEC_ARCH_X86_64
+#endif
+
+/*
+ * CPU does not save ss and sp on stack if execution is already
+ * running in kernel mode at the time of NMI occurrence. This code
+ * fixes it.
+ */
+static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
+				      struct pt_regs *oldregs)
+{
+#ifdef CONFIG_X86_32
+	newregs->sp = (unsigned long)&(oldregs->sp);
+	__asm__ __volatile__(
+			"xorl %%eax, %%eax\n\t"
+			"movw %%ss, %%ax\n\t"
+			:"=a"(newregs->ss));
+#endif
+}
+
+/*
+ * This function is responsible for capturing register states if coming
+ * via panic otherwise just fix up the ss and sp if coming via kernel
+ * mode exception.
+ */
+static inline void crash_setup_regs(struct pt_regs *newregs,
+				    struct pt_regs *oldregs)
+{
+	if (oldregs) {
+		memcpy(newregs, oldregs, sizeof(*newregs));
+		crash_fixup_ss_esp(newregs, oldregs);
+	} else {
+#ifdef CONFIG_X86_32
+		__asm__ __volatile__("movl %%ebx,%0" : "=m"(newregs->bx));
+		__asm__ __volatile__("movl %%ecx,%0" : "=m"(newregs->cx));
+		__asm__ __volatile__("movl %%edx,%0" : "=m"(newregs->dx));
+		__asm__ __volatile__("movl %%esi,%0" : "=m"(newregs->si));
+		__asm__ __volatile__("movl %%edi,%0" : "=m"(newregs->di));
+		__asm__ __volatile__("movl %%ebp,%0" : "=m"(newregs->bp));
+		__asm__ __volatile__("movl %%eax,%0" : "=m"(newregs->ax));
+		__asm__ __volatile__("movl %%esp,%0" : "=m"(newregs->sp));
+		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
+		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
+		__asm__ __volatile__("movl %%ds, %%eax;" :"=a"(newregs->ds));
+		__asm__ __volatile__("movl %%es, %%eax;" :"=a"(newregs->es));
+		__asm__ __volatile__("pushfl; popl %0" :"=m"(newregs->flags));
+#else
+		__asm__ __volatile__("movq %%rbx,%0" : "=m"(newregs->bx));
+		__asm__ __volatile__("movq %%rcx,%0" : "=m"(newregs->cx));
+		__asm__ __volatile__("movq %%rdx,%0" : "=m"(newregs->dx));
+		__asm__ __volatile__("movq %%rsi,%0" : "=m"(newregs->si));
+		__asm__ __volatile__("movq %%rdi,%0" : "=m"(newregs->di));
+		__asm__ __volatile__("movq %%rbp,%0" : "=m"(newregs->bp));
+		__asm__ __volatile__("movq %%rax,%0" : "=m"(newregs->ax));
+		__asm__ __volatile__("movq %%rsp,%0" : "=m"(newregs->sp));
+		__asm__ __volatile__("movq %%r8,%0" : "=m"(newregs->r8));
+		__asm__ __volatile__("movq %%r9,%0" : "=m"(newregs->r9));
+		__asm__ __volatile__("movq %%r10,%0" : "=m"(newregs->r10));
+		__asm__ __volatile__("movq %%r11,%0" : "=m"(newregs->r11));
+		__asm__ __volatile__("movq %%r12,%0" : "=m"(newregs->r12));
+		__asm__ __volatile__("movq %%r13,%0" : "=m"(newregs->r13));
+		__asm__ __volatile__("movq %%r14,%0" : "=m"(newregs->r14));
+		__asm__ __volatile__("movq %%r15,%0" : "=m"(newregs->r15));
+		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
+		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
+		__asm__ __volatile__("pushfq; popq %0" :"=m"(newregs->flags));
+#endif
+		newregs->ip = (unsigned long)current_text_addr();
+	}
+}
+
+#ifdef CONFIG_X86_32
+asmlinkage NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long control_page,
+		unsigned long start_address,
+		unsigned int has_pae) ATTRIB_NORET;
+#else
+NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long page_list,
+		unsigned long start_address) ATTRIB_NORET;
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* _KEXEC_H */
diff --git a/include/asm-x86/kexec_32.h b/include/asm-x86/kexec_32.h
deleted file mode 100644
index ff39d2f88022..000000000000
--- a/include/asm-x86/kexec_32.h
+++ /dev/null
@@ -1,99 +0,0 @@
-#ifndef _I386_KEXEC_H
-#define _I386_KEXEC_H
-
-#define PA_CONTROL_PAGE  0
-#define VA_CONTROL_PAGE  1
-#define PA_PGD           2
-#define VA_PGD           3
-#define PA_PTE_0         4
-#define VA_PTE_0         5
-#define PA_PTE_1         6
-#define VA_PTE_1         7
-#ifdef CONFIG_X86_PAE
-#define PA_PMD_0         8
-#define VA_PMD_0         9
-#define PA_PMD_1         10
-#define VA_PMD_1         11
-#define PAGES_NR         12
-#else
-#define PAGES_NR         8
-#endif
-
-#ifndef __ASSEMBLY__
-
-#include <asm/ptrace.h>
-#include <asm/string.h>
-
-/*
- * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
- * I.e. Maximum page that is mapped directly into kernel memory,
- * and kmap is not required.
- */
-
-/* Maximum physical address we can use pages from */
-#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
-/* Maximum address we can reach in physical address mode */
-#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
-/* Maximum address we can use for the control code buffer */
-#define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE
-
-#define KEXEC_CONTROL_CODE_SIZE	4096
-
-/* The native architecture */
-#define KEXEC_ARCH KEXEC_ARCH_386
-
-/* We can also handle crash dumps from 64 bit kernel. */
-#define vmcore_elf_check_arch_cross(x) ((x)->e_machine == EM_X86_64)
-
-/* CPU does not save ss and sp on stack if execution is already
- * running in kernel mode at the time of NMI occurrence. This code
- * fixes it.
- */
-static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
-					struct pt_regs *oldregs)
-{
-	memcpy(newregs, oldregs, sizeof(*newregs));
-	newregs->sp = (unsigned long)&(oldregs->sp);
-	__asm__ __volatile__(
-			"xorl %%eax, %%eax\n\t"
-			"movw %%ss, %%ax\n\t"
-			:"=a"(newregs->ss));
-}
-
-/*
- * This function is responsible for capturing register states if coming
- * via panic otherwise just fix up the ss and sp if coming via kernel
- * mode exception.
- */
-static inline void crash_setup_regs(struct pt_regs *newregs,
-                                       struct pt_regs *oldregs)
-{
-       if (oldregs)
-               crash_fixup_ss_esp(newregs, oldregs);
-       else {
-               __asm__ __volatile__("movl %%ebx,%0" : "=m"(newregs->bx));
-               __asm__ __volatile__("movl %%ecx,%0" : "=m"(newregs->cx));
-               __asm__ __volatile__("movl %%edx,%0" : "=m"(newregs->dx));
-               __asm__ __volatile__("movl %%esi,%0" : "=m"(newregs->si));
-               __asm__ __volatile__("movl %%edi,%0" : "=m"(newregs->di));
-               __asm__ __volatile__("movl %%ebp,%0" : "=m"(newregs->bp));
-               __asm__ __volatile__("movl %%eax,%0" : "=m"(newregs->ax));
-               __asm__ __volatile__("movl %%esp,%0" : "=m"(newregs->sp));
-               __asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
-               __asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
-               __asm__ __volatile__("movl %%ds, %%eax;" :"=a"(newregs->ds));
-               __asm__ __volatile__("movl %%es, %%eax;" :"=a"(newregs->es));
-               __asm__ __volatile__("pushfl; popl %0" :"=m"(newregs->flags));
-
-               newregs->ip = (unsigned long)current_text_addr();
-       }
-}
-asmlinkage NORET_TYPE void
-relocate_kernel(unsigned long indirection_page,
-		unsigned long control_page,
-		unsigned long start_address,
-		unsigned int has_pae) ATTRIB_NORET;
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* _I386_KEXEC_H */
diff --git a/include/asm-x86/kexec_64.h b/include/asm-x86/kexec_64.h
deleted file mode 100644
index b5f989b15c0b..000000000000
--- a/include/asm-x86/kexec_64.h
+++ /dev/null
@@ -1,94 +0,0 @@
-#ifndef _X86_64_KEXEC_H
-#define _X86_64_KEXEC_H
-
-#define PA_CONTROL_PAGE  0
-#define VA_CONTROL_PAGE  1
-#define PA_PGD           2
-#define VA_PGD           3
-#define PA_PUD_0         4
-#define VA_PUD_0         5
-#define PA_PMD_0         6
-#define VA_PMD_0         7
-#define PA_PTE_0         8
-#define VA_PTE_0         9
-#define PA_PUD_1         10
-#define VA_PUD_1         11
-#define PA_PMD_1         12
-#define VA_PMD_1         13
-#define PA_PTE_1         14
-#define VA_PTE_1         15
-#define PA_TABLE_PAGE    16
-#define PAGES_NR         17
-
-#ifndef __ASSEMBLY__
-
-#include <linux/string.h>
-
-#include <asm/page.h>
-#include <asm/ptrace.h>
-
-/*
- * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
- * I.e. Maximum page that is mapped directly into kernel memory,
- * and kmap is not required.
- *
- * So far x86_64 is limited to 40 physical address bits.
- */
-
-/* Maximum physical address we can use pages from */
-#define KEXEC_SOURCE_MEMORY_LIMIT      (0xFFFFFFFFFFUL)
-/* Maximum address we can reach in physical address mode */
-#define KEXEC_DESTINATION_MEMORY_LIMIT (0xFFFFFFFFFFUL)
-/* Maximum address we can use for the control pages */
-#define KEXEC_CONTROL_MEMORY_LIMIT     (0xFFFFFFFFFFUL)
-
-/* Allocate one page for the pdp and the second for the code */
-#define KEXEC_CONTROL_CODE_SIZE  (4096UL + 4096UL)
-
-/* The native architecture */
-#define KEXEC_ARCH KEXEC_ARCH_X86_64
-
-/*
- * Saving the registers of the cpu on which panic occured in
- * crash_kexec to save a valid sp. The registers of other cpus
- * will be saved in machine_crash_shutdown while shooting down them.
- */
-
-static inline void crash_setup_regs(struct pt_regs *newregs,
-						struct pt_regs *oldregs)
-{
-	if (oldregs)
-		memcpy(newregs, oldregs, sizeof(*newregs));
-	else {
-		__asm__ __volatile__("movq %%rbx,%0" : "=m"(newregs->bx));
-		__asm__ __volatile__("movq %%rcx,%0" : "=m"(newregs->cx));
-		__asm__ __volatile__("movq %%rdx,%0" : "=m"(newregs->dx));
-		__asm__ __volatile__("movq %%rsi,%0" : "=m"(newregs->si));
-		__asm__ __volatile__("movq %%rdi,%0" : "=m"(newregs->di));
-		__asm__ __volatile__("movq %%rbp,%0" : "=m"(newregs->bp));
-		__asm__ __volatile__("movq %%rax,%0" : "=m"(newregs->ax));
-		__asm__ __volatile__("movq %%rsp,%0" : "=m"(newregs->sp));
-		__asm__ __volatile__("movq %%r8,%0" : "=m"(newregs->r8));
-		__asm__ __volatile__("movq %%r9,%0" : "=m"(newregs->r9));
-		__asm__ __volatile__("movq %%r10,%0" : "=m"(newregs->r10));
-		__asm__ __volatile__("movq %%r11,%0" : "=m"(newregs->r11));
-		__asm__ __volatile__("movq %%r12,%0" : "=m"(newregs->r12));
-		__asm__ __volatile__("movq %%r13,%0" : "=m"(newregs->r13));
-		__asm__ __volatile__("movq %%r14,%0" : "=m"(newregs->r14));
-		__asm__ __volatile__("movq %%r15,%0" : "=m"(newregs->r15));
-		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
-		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
-		__asm__ __volatile__("pushfq; popq %0" :"=m"(newregs->flags));
-
-		newregs->ip = (unsigned long)current_text_addr();
-	}
-}
-
-NORET_TYPE void
-relocate_kernel(unsigned long indirection_page,
-		unsigned long page_list,
-		unsigned long start_address) ATTRIB_NORET;
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* _X86_64_KEXEC_H */
-- 
cgit v1.2.3


From 73018a66e70fa631935192e5ed6ef8fa2d20b22d Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:26 +0100
Subject: x86: move tsc definitions to were they belong

This patch wipes out the definitions of tsc_disable from processor_32.h
and move it to tsc.h, were it belongs

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor_32.h | 5 -----
 include/asm-x86/tsc.h          | 4 ++++
 2 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index a5560a3b57ef..76cada930d22 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -22,9 +22,6 @@
 #include <asm/processor-flags.h>
 #include <asm/desc_defs.h>
 
-/* flag for disabling the tsc */
-extern int tsc_disable;
-
 static inline int desc_empty(const void *ptr)
 {
 	const u32 *desc = ptr;
@@ -206,8 +203,6 @@ extern int bootloader_type;
 
 #define HAVE_ARCH_PICK_MMAP_LAYOUT
 
-extern void disable_TSC(void);
-
 /*
  * Size of io_bitmap.
  */
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index 9b7d264897fa..401303724130 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -16,6 +16,10 @@ typedef unsigned long long cycles_t;
 
 extern unsigned int cpu_khz;
 extern unsigned int tsc_khz;
+/* flag for disabling the tsc */
+extern int tsc_disable;
+
+extern void disable_TSC(void);
 
 static inline cycles_t get_cycles(void)
 {
-- 
cgit v1.2.3


From 053de044411111da00272d1b4e174e7dd743f499 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: get rid of _MASK flags

There's no need for the *_MASK flags (TF_MASK, IF_MASK, etc), found in
processor.h (both _32 and _64). They have a one-to-one mapping with the
EFLAGS value. This patch removes the definitions, and use the already
existent X86_EFLAGS_ version when applicable.

[ roland@redhat.com: KVM build fixes. ]

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/ia32_signal.c    |  4 ++--
 arch/x86/kernel/kprobes.c      | 18 +++++++++---------
 arch/x86/kernel/signal_64.c    |  4 ++--
 arch/x86/kernel/traps_64.c     |  2 +-
 drivers/kvm/vmx.c              |  8 ++++----
 include/asm-x86/kprobes.h      |  2 +-
 include/asm-x86/processor.h    |  2 ++
 include/asm-x86/processor_32.h |  1 -
 include/asm-x86/processor_64.h | 11 -----------
 9 files changed, 21 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index d03d43f32f4c..0e24e3fda3d7 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -501,7 +501,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	regs->ss = __USER32_DS;
 
 	set_fs(USER_DS);
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
@@ -601,7 +601,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->ss = __USER32_DS;
 
 	set_fs(USER_DS);
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 9aadd4d4a229..8de82c8cedd6 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -387,9 +387,9 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 {
 	__get_cpu_var(current_kprobe) = p;
 	kcb->kprobe_saved_flags = kcb->kprobe_old_flags
-		= (regs->flags & (TF_MASK | IF_MASK));
+		= (regs->flags & (X86_EFLAGS_TF | X86_EFLAGS_IF));
 	if (is_IF_modifier(p->ainsn.insn))
-		kcb->kprobe_saved_flags &= ~IF_MASK;
+		kcb->kprobe_saved_flags &= ~X86_EFLAGS_IF;
 }
 
 static __always_inline void clear_btf(void)
@@ -407,8 +407,8 @@ static __always_inline void restore_btf(void)
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
 	clear_btf();
-	regs->flags |= TF_MASK;
-	regs->flags &= ~IF_MASK;
+	regs->flags |= X86_EFLAGS_TF;
+	regs->flags &= ~X86_EFLAGS_IF;
 	/*single step inline if the instruction is an int3*/
 	if (p->opcode == BREAKPOINT_INSTRUCTION)
 		regs->ip = (unsigned long)p->addr;
@@ -454,7 +454,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 		if (p) {
 			if (kcb->kprobe_status == KPROBE_HIT_SS &&
 				*p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
-				regs->flags &= ~TF_MASK;
+				regs->flags &= ~X86_EFLAGS_TF;
 				regs->flags |= kcb->kprobe_saved_flags;
 				goto no_kprobe;
 #ifdef CONFIG_X86_64
@@ -749,10 +749,10 @@ static void __kprobes resume_execution(struct kprobe *p,
 		insn++;
 #endif
 
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	switch (*insn) {
 	case 0x9c:	/* pushfl */
-		*tos &= ~(TF_MASK | IF_MASK);
+		*tos &= ~(X86_EFLAGS_TF | X86_EFLAGS_IF);
 		*tos |= kcb->kprobe_old_flags;
 		break;
 	case 0xc2:	/* iret/ret/lret */
@@ -852,7 +852,7 @@ out:
 	 * will have TF set, in which case, continue the remaining processing
 	 * of do_debug, as if this is not a probe hit.
 	 */
-	if (regs->flags & TF_MASK)
+	if (regs->flags & X86_EFLAGS_TF)
 		return 0;
 
 	return 1;
@@ -982,7 +982,7 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	 */
 	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
 	       MIN_STACK_SIZE(addr));
-	regs->flags &= ~IF_MASK;
+	regs->flags &= ~X86_EFLAGS_IF;
 	trace_hardirqs_off();
 	regs->ip = (unsigned long)(jp->entry);
 	return 1;
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 1c9bca56fb55..4eb751c60390 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -295,7 +295,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	   see include/asm-x86_64/uaccess.h for details. */
 	set_fs(USER_DS);
 
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 #ifdef DEBUG_SIG
@@ -463,7 +463,7 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 	       
 	/* Pending single-step? */
 	if (thread_info_flags & _TIF_SINGLESTEP) {
-		regs->flags |= TF_MASK;
+		regs->flags |= X86_EFLAGS_TF;
 		clear_thread_flag(TIF_SINGLESTEP);
 	}
 
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index cf90ceb48f5f..7118fa0320ae 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -904,7 +904,7 @@ clear_dr7:
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	preempt_conditional_cli(regs);
 }
 
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index bb56ae3f89b6..5b397b6c9f93 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -524,7 +524,7 @@ static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu)
 static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
 {
 	if (vcpu->rmode.active)
-		rflags |= IOPL_MASK | X86_EFLAGS_VM;
+		rflags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
 	vmcs_writel(GUEST_RFLAGS, rflags);
 }
 
@@ -1050,7 +1050,7 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
 	vmcs_write32(GUEST_TR_AR_BYTES, vcpu->rmode.tr.ar);
 
 	flags = vmcs_readl(GUEST_RFLAGS);
-	flags &= ~(IOPL_MASK | X86_EFLAGS_VM);
+	flags &= ~(X86_EFLAGS_IOPL | X86_EFLAGS_VM);
 	flags |= (vcpu->rmode.save_iopl << IOPL_SHIFT);
 	vmcs_writel(GUEST_RFLAGS, flags);
 
@@ -1107,9 +1107,9 @@ static void enter_rmode(struct kvm_vcpu *vcpu)
 	vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
 
 	flags = vmcs_readl(GUEST_RFLAGS);
-	vcpu->rmode.save_iopl = (flags & IOPL_MASK) >> IOPL_SHIFT;
+	vcpu->rmode.save_iopl = (flags & X86_EFLAGS_IOPL) >> IOPL_SHIFT;
 
-	flags |= IOPL_MASK | X86_EFLAGS_VM;
+	flags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
 
 	vmcs_writel(GUEST_RFLAGS, flags);
 	vmcs_writel(GUEST_CR4, vmcs_readl(GUEST_CR4) | X86_CR4_VME);
diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h
index 6e6371a856e7..143476a3cb52 100644
--- a/include/asm-x86/kprobes.h
+++ b/include/asm-x86/kprobes.h
@@ -88,7 +88,7 @@ struct kprobe_ctlblk {
  */
 static inline void restore_interrupts(struct pt_regs *regs)
 {
-	if (regs->flags & IF_MASK)
+	if (regs->flags & X86_EFLAGS_IF)
 		local_irq_enable();
 }
 
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index dea81b70895d..e8dd394c9f46 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -1,6 +1,8 @@
 #ifndef __ASM_X86_PROCESSOR_H
 #define __ASM_X86_PROCESSOR_H
 
+#include <asm/processor-flags.h>
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 76cada930d22..b9dbe4668e75 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -19,7 +19,6 @@
 #include <asm/percpu.h>
 #include <linux/cpumask.h>
 #include <linux/init.h>
-#include <asm/processor-flags.h>
 #include <asm/desc_defs.h>
 
 static inline int desc_empty(const void *ptr)
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index a0a9e5515097..c49716a76644 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -18,19 +18,8 @@
 #include <asm/percpu.h>
 #include <linux/personality.h>
 #include <linux/cpumask.h>
-#include <asm/processor-flags.h>
 #include <asm/desc_defs.h>
 
-#define TF_MASK		0x00000100
-#define IF_MASK		0x00000200
-#define IOPL_MASK	0x00003000
-#define NT_MASK		0x00004000
-#define VM_MASK		0x00020000
-#define AC_MASK		0x00040000
-#define VIF_MASK	0x00080000	/* virtual interrupt flag */
-#define VIP_MASK	0x00100000	/* virtual interrupt pending */
-#define ID_MASK		0x00200000
-
 static inline int desc_empty(const void *ptr)
 {
 	const u32 *desc = ptr;
-- 
cgit v1.2.3


From 746ff60f2627626fca0d8ddbf6b1f04d505782f4 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: move desc_empty to where they belong

This patch moves the (duplicated) desc_empty implementation to desc.h,
where the descriptor things belong.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/desc.h         | 6 ++++++
 include/asm-x86/processor_32.h | 6 ------
 include/asm-x86/processor_64.h | 6 ------
 3 files changed, 6 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index e9356a852020..e17581ad8824 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -71,6 +71,12 @@ static inline void pack_gate(gate_desc *gate, unsigned char type,
 
 #endif
 
+static inline int desc_empty(const void *ptr)
+{
+	const u32 *desc = ptr;
+	return !(desc[0] | desc[1]);
+}
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index b9dbe4668e75..cadeffb54a72 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -21,12 +21,6 @@
 #include <linux/init.h>
 #include <asm/desc_defs.h>
 
-static inline int desc_empty(const void *ptr)
-{
-	const u32 *desc = ptr;
-	return !(desc[0] | desc[1]);
-}
-
 /*
  * Default implementation of macro that returns current
  * instruction pointer ("program counter").
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index c49716a76644..acecef80a510 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -20,12 +20,6 @@
 #include <linux/cpumask.h>
 #include <asm/desc_defs.h>
 
-static inline int desc_empty(const void *ptr)
-{
-	const u32 *desc = ptr;
-	return !(desc[0] | desc[1]);
-}
-
 /*
  * Default implementation of macro that returns current
  * instruction pointer ("program counter").
-- 
cgit v1.2.3


From c72dcf83ff8585c95739abffa3be7c87ca63d66b Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: move load_cr3 to a common place.

There are currently two definitions of load_cr3, that essentially do the
same thing. This patch moves them all to processor.h.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/mmu_context_64.h | 5 -----
 include/asm-x86/processor.h      | 7 +++++++
 include/asm-x86/processor_32.h   | 1 -
 3 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h
index 98bfe43dd806..7e2aa23fccbf 100644
--- a/include/asm-x86/mmu_context_64.h
+++ b/include/asm-x86/mmu_context_64.h
@@ -23,11 +23,6 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 #endif
 }
 
-static inline void load_cr3(pgd_t *pgd)
-{
-	asm volatile("movq %0,%%cr3" :: "r" (__pa(pgd)) : "memory");
-}
-
 static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, 
 			     struct task_struct *tsk)
 {
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index e8dd394c9f46..36ee9881b74f 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -3,6 +3,9 @@
 
 #include <asm/processor-flags.h>
 
+#include <asm/page.h>
+#include <asm/system.h>
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
@@ -15,6 +18,10 @@ static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 		: "0" (*eax), "2" (*ecx));
 }
 
+static inline void load_cr3(pgd_t *pgdir)
+{
+	write_cr3(__pa(pgdir));
+}
 
 #ifdef CONFIG_X86_32
 # include "processor_32.h"
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index cadeffb54a72..b586902acd85 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -120,7 +120,6 @@ extern void detect_ht(struct cpuinfo_x86 *c);
 static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #endif
 
-#define load_cr3(pgdir) write_cr3(__pa(pgdir))
 
 /*
  * Save the cr4 feature set we're using (ie
-- 
cgit v1.2.3


From 1b46cbe0ccaad25786526601bc54426f2e2abb20 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: unify paravirt pieces of processor.h

This patch unifies the paravirt pieces of processor.h
The functionality present in 32 bit, but not (yet) in 64-bit,
like load_sp0 is _not_ done here, and let to a different patch.

With this unification, we get paravirt for free in x86_64 processor.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 90 +++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/processor_32.h | 87 ----------------------------------------
 include/asm-x86/processor_64.h | 36 -----------------
 3 files changed, 89 insertions(+), 124 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 36ee9881b74f..8d0af7bf090a 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -29,9 +29,97 @@ static inline void load_cr3(pgd_t *pgdir)
 # include "processor_64.h"
 #endif
 
+static inline unsigned long native_get_debugreg(int regno)
+{
+	unsigned long val = 0; 	/* Damn you, gcc! */
+
+	switch (regno) {
+	case 0:
+		asm("mov %%db0, %0" :"=r" (val)); break;
+	case 1:
+		asm("mov %%db1, %0" :"=r" (val)); break;
+	case 2:
+		asm("mov %%db2, %0" :"=r" (val)); break;
+	case 3:
+		asm("mov %%db3, %0" :"=r" (val)); break;
+	case 6:
+		asm("mov %%db6, %0" :"=r" (val)); break;
+	case 7:
+		asm("mov %%db7, %0" :"=r" (val)); break;
+	default:
+		BUG();
+	}
+	return val;
+}
+
+static inline void native_set_debugreg(int regno, unsigned long value)
+{
+	switch (regno) {
+	case 0:
+		asm("mov %0,%%db0"	: /* no output */ :"r" (value));
+		break;
+	case 1:
+		asm("mov %0,%%db1"	: /* no output */ :"r" (value));
+		break;
+	case 2:
+		asm("mov %0,%%db2"	: /* no output */ :"r" (value));
+		break;
+	case 3:
+		asm("mov %0,%%db3"	: /* no output */ :"r" (value));
+		break;
+	case 6:
+		asm("mov %0,%%db6"	: /* no output */ :"r" (value));
+		break;
+	case 7:
+		asm("mov %0,%%db7"	: /* no output */ :"r" (value));
+		break;
+	default:
+		BUG();
+	}
+}
+
+
 #ifndef CONFIG_PARAVIRT
 #define __cpuid native_cpuid
-#endif
+#define paravirt_enabled() 0
+
+/*
+ * These special macros can be used to get or set a debugging register
+ */
+#define get_debugreg(var, register)				\
+	(var) = native_get_debugreg(register)
+#define set_debugreg(value, register)				\
+	native_set_debugreg(register, value)
+
+#endif /* CONFIG_PARAVIRT */
+
+/*
+ * Save the cr4 feature set we're using (ie
+ * Pentium 4MB enable and PPro Global page
+ * enable), so that any CPU's that boot up
+ * after us can get the correct flags.
+ */
+extern unsigned long mmu_cr4_features;
+
+static inline void set_in_cr4(unsigned long mask)
+{
+	unsigned cr4;
+	mmu_cr4_features |= mask;
+	cr4 = read_cr4();
+	cr4 |= mask;
+	write_cr4(cr4);
+}
+
+static inline void clear_in_cr4(unsigned long mask)
+{
+	unsigned cr4;
+	mmu_cr4_features &= ~mask;
+	cr4 = read_cr4();
+	cr4 &= ~mask;
+	write_cr4(cr4);
+}
+
+
 
 /*
  * Generic CPUID function
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index b586902acd85..e7fc023cd810 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -120,33 +120,6 @@ extern void detect_ht(struct cpuinfo_x86 *c);
 static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #endif
 
-
-/*
- * Save the cr4 feature set we're using (ie
- * Pentium 4MB enable and PPro Global page
- * enable), so that any CPU's that boot up
- * after us can get the correct flags.
- */
-extern unsigned long mmu_cr4_features;
-
-static inline void set_in_cr4 (unsigned long mask)
-{
-	unsigned cr4;
-	mmu_cr4_features |= mask;
-	cr4 = read_cr4();
-	cr4 |= mask;
-	write_cr4(cr4);
-}
-
-static inline void clear_in_cr4 (unsigned long mask)
-{
-	unsigned cr4;
-	mmu_cr4_features &= ~mask;
-	cr4 = read_cr4();
-	cr4 &= ~mask;
-	write_cr4(cr4);
-}
-
 /* Stop speculative execution */
 static inline void sync_core(void)
 {
@@ -482,57 +455,6 @@ static inline void native_load_sp0(struct tss_struct *tss, struct thread_struct
 		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
 	}
 }
-
-
-static inline unsigned long native_get_debugreg(int regno)
-{
-	unsigned long val = 0; 	/* Damn you, gcc! */
-
-	switch (regno) {
-	case 0:
-		asm("movl %%db0, %0" :"=r" (val)); break;
-	case 1:
-		asm("movl %%db1, %0" :"=r" (val)); break;
-	case 2:
-		asm("movl %%db2, %0" :"=r" (val)); break;
-	case 3:
-		asm("movl %%db3, %0" :"=r" (val)); break;
-	case 6:
-		asm("movl %%db6, %0" :"=r" (val)); break;
-	case 7:
-		asm("movl %%db7, %0" :"=r" (val)); break;
-	default:
-		BUG();
-	}
-	return val;
-}
-
-static inline void native_set_debugreg(int regno, unsigned long value)
-{
-	switch (regno) {
-	case 0:
-		asm("movl %0,%%db0"	: /* no output */ :"r" (value));
-		break;
-	case 1:
-		asm("movl %0,%%db1"	: /* no output */ :"r" (value));
-		break;
-	case 2:
-		asm("movl %0,%%db2"	: /* no output */ :"r" (value));
-		break;
-	case 3:
-		asm("movl %0,%%db3"	: /* no output */ :"r" (value));
-		break;
-	case 6:
-		asm("movl %0,%%db6"	: /* no output */ :"r" (value));
-		break;
-	case 7:
-		asm("movl %0,%%db7"	: /* no output */ :"r" (value));
-		break;
-	default:
-		BUG();
-	}
-}
-
 /*
  * Set IOPL bits in EFLAGS from given mask
  */
@@ -552,21 +474,12 @@ static inline void native_set_iopl_mask(unsigned mask)
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
-#define paravirt_enabled() 0
 
 static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
 	native_load_sp0(tss, thread);
 }
 
-/*
- * These special macros can be used to get or set a debugging register
- */
-#define get_debugreg(var, register)				\
-	(var) = native_get_debugreg(register)
-#define set_debugreg(value, register)				\
-	native_set_debugreg(register, value)
-
 #define set_iopl_mask native_set_iopl_mask
 #endif /* CONFIG_PARAVIRT */
 
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index acecef80a510..6abe1ba30fe0 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -88,34 +88,6 @@ extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
 
-/*
- * Save the cr4 feature set we're using (ie
- * Pentium 4MB enable and PPro Global page
- * enable), so that any CPU's that boot up
- * after us can get the correct flags.
- */
-extern unsigned long mmu_cr4_features;
-
-static inline void set_in_cr4 (unsigned long mask)
-{
-	mmu_cr4_features |= mask;
-	__asm__("movq %%cr4,%%rax\n\t"
-		"orq %0,%%rax\n\t"
-		"movq %%rax,%%cr4\n"
-		: : "irg" (mask)
-		:"ax");
-}
-
-static inline void clear_in_cr4 (unsigned long mask)
-{
-	mmu_cr4_features &= ~mask;
-	__asm__("movq %%cr4,%%rax\n\t"
-		"andq %0,%%rax\n\t"
-		"movq %%rax,%%cr4\n"
-		: : "irg" (~mask)
-		:"ax");
-}
-
 
 /*
  * User space process size. 47bits minus one guard page.
@@ -254,14 +226,6 @@ struct thread_struct {
 	set_fs(USER_DS);							 \
 } while(0) 
 
-#define get_debugreg(var, register)				\
-		__asm__("movq %%db" #register ", %0"		\
-			:"=r" (var))
-#define set_debugreg(value, register)			\
-		__asm__("movq %0,%%db" #register		\
-			: /* no output */			\
-			:"r" (value))
-
 struct task_struct;
 struct mm_struct;
 
-- 
cgit v1.2.3


From 62d7d7ed11760a0fea40e4fc6f0553e721d00443 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: move the definition of set_iopl_mask to common header

This patch moves the definition of set_iopl_mask to processor.h,
instead of letting it at processor_32.h.
For x86_64, nothing is done, as we don't really need such a function.
However, having it on both arches saves us from putting an ifdef in the
pv_cpu_ops struct.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 19 +++++++++++++++++++
 include/asm-x86/processor_32.h | 17 -----------------
 2 files changed, 19 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 8d0af7bf090a..8b7794766884 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -78,6 +78,24 @@ static inline void native_set_debugreg(int regno, unsigned long value)
 	}
 }
 
+/*
+ * Set IOPL bits in EFLAGS from given mask
+ */
+static inline void native_set_iopl_mask(unsigned mask)
+{
+#ifdef CONFIG_X86_32
+	unsigned int reg;
+	__asm__ __volatile__ ("pushfl;"
+			      "popl %0;"
+			      "andl %1, %0;"
+			      "orl %2, %0;"
+			      "pushl %0;"
+			      "popfl"
+				: "=&r" (reg)
+				: "i" (~X86_EFLAGS_IOPL), "r" (mask));
+#endif
+}
+
 
 #ifndef CONFIG_PARAVIRT
 #define __cpuid native_cpuid
@@ -91,6 +109,7 @@ static inline void native_set_debugreg(int regno, unsigned long value)
 #define set_debugreg(value, register)				\
 	native_set_debugreg(register, value)
 
+#define set_iopl_mask native_set_iopl_mask
 #endif /* CONFIG_PARAVIRT */
 
 /*
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index e7fc023cd810..516a38ad2a7a 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -455,21 +455,6 @@ static inline void native_load_sp0(struct tss_struct *tss, struct thread_struct
 		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
 	}
 }
-/*
- * Set IOPL bits in EFLAGS from given mask
- */
-static inline void native_set_iopl_mask(unsigned mask)
-{
-	unsigned int reg;
-	__asm__ __volatile__ ("pushfl;"
-			      "popl %0;"
-			      "andl %1, %0;"
-			      "orl %2, %0;"
-			      "pushl %0;"
-			      "popfl"
-				: "=&r" (reg)
-				: "i" (~X86_EFLAGS_IOPL), "r" (mask));
-}
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
@@ -479,8 +464,6 @@ static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread
 {
 	native_load_sp0(tss, thread);
 }
-
-#define set_iopl_mask native_set_iopl_mask
 #endif /* CONFIG_PARAVIRT */
 
 /* generic versions from gas */
-- 
cgit v1.2.3


From 683e0253dbd12554b2ee969b15e68105252bff57 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: unify common parts of processor.h

This patch moves the pieces of processor_32.h and processor_64 that are
equal to processor.h. Only what's exactly the same is moved around, the rest
not being touched.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 120 +++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 111 --------------------------------------
 include/asm-x86/processor_64.h | 116 ---------------------------------------
 3 files changed, 120 insertions(+), 227 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 8b7794766884..52e3637ef59e 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -3,6 +3,10 @@
 
 #include <asm/processor-flags.h>
 
+/* Forward declaration, a strange C thing */
+struct task_struct;
+struct mm_struct;
+
 #include <asm/page.h>
 #include <asm/system.h>
 
@@ -29,6 +33,11 @@ static inline void load_cr3(pgd_t *pgdir)
 # include "processor_64.h"
 #endif
 
+extern void print_cpu_info(struct cpuinfo_x86 *);
+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
+extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
+extern unsigned short num_cache_leaves;
+
 static inline unsigned long native_get_debugreg(int regno)
 {
 	unsigned long val = 0; 	/* Damn you, gcc! */
@@ -138,7 +147,53 @@ static inline void clear_in_cr4(unsigned long mask)
 	write_cr4(cr4);
 }
 
+struct microcode_header {
+	unsigned int hdrver;
+	unsigned int rev;
+	unsigned int date;
+	unsigned int sig;
+	unsigned int cksum;
+	unsigned int ldrver;
+	unsigned int pf;
+	unsigned int datasize;
+	unsigned int totalsize;
+	unsigned int reserved[3];
+};
+
+struct microcode {
+	struct microcode_header hdr;
+	unsigned int bits[0];
+};
+
+typedef struct microcode microcode_t;
+typedef struct microcode_header microcode_header_t;
+
+/* microcode format is extended from prescott processors */
+struct extended_signature {
+	unsigned int sig;
+	unsigned int pf;
+	unsigned int cksum;
+};
+
+struct extended_sigtable {
+	unsigned int count;
+	unsigned int cksum;
+	unsigned int reserved[3];
+	struct extended_signature sigs[0];
+};
+
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
+
+/* Free all resources held by a thread. */
+extern void release_thread(struct task_struct *);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+extern void prepare_to_copy(struct task_struct *tsk);
 
+unsigned long get_wchan(struct task_struct *p);
 
 /*
  * Generic CPUID function
@@ -196,4 +251,69 @@ static inline unsigned int cpuid_edx(unsigned int op)
 	return edx;
 }
 
+/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
+static inline void rep_nop(void)
+{
+	__asm__ __volatile__("rep;nop": : :"memory");
+}
+
+/* Stop speculative execution */
+static inline void sync_core(void)
+{
+	int tmp;
+	asm volatile("cpuid" : "=a" (tmp) : "0" (1)
+					  : "ebx", "ecx", "edx", "memory");
+}
+
+#define cpu_relax()   rep_nop()
+
+static inline void __monitor(const void *eax, unsigned long ecx,
+		unsigned long edx)
+{
+	/* "monitor %eax,%ecx,%edx;" */
+	asm volatile(
+		".byte 0x0f,0x01,0xc8;"
+		: :"a" (eax), "c" (ecx), "d"(edx));
+}
+
+static inline void __mwait(unsigned long eax, unsigned long ecx)
+{
+	/* "mwait %eax,%ecx;" */
+	asm volatile(
+		".byte 0x0f,0x01,0xc9;"
+		: :"a" (eax), "c" (ecx));
+}
+
+static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
+{
+	/* "mwait %eax,%ecx;" */
+	asm volatile(
+		"sti; .byte 0x0f,0x01,0xc9;"
+		: :"a" (eax), "c" (ecx));
+}
+
+extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
+
+extern int force_mwait;
+
+extern void select_idle_routine(const struct cpuinfo_x86 *c);
+
+extern unsigned long boot_option_idle_override;
+
+/* Boot loader type from the setup header */
+extern int bootloader_type;
+#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
+
+#define HAVE_ARCH_PICK_MMAP_LAYOUT 1
+#define ARCH_HAS_PREFETCHW
+#define ARCH_HAS_SPINLOCK_PREFETCH
+
+#define spin_lock_prefetch(x)	prefetchw(x)
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
+
+#define KSTK_EIP(task) (task_pt_regs(task)->ip)
+
 #endif
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 516a38ad2a7a..9e119d3789b4 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -109,10 +109,6 @@ void __init cpu_detect(struct cpuinfo_x86 *c);
 
 extern void identify_boot_cpu(void);
 extern void identify_secondary_cpu(struct cpuinfo_x86 *);
-extern void print_cpu_info(struct cpuinfo_x86 *);
-extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
-extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
-extern unsigned short num_cache_leaves;
 
 #ifdef CONFIG_X86_HT
 extern void detect_ht(struct cpuinfo_x86 *c);
@@ -120,32 +116,6 @@ extern void detect_ht(struct cpuinfo_x86 *c);
 static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #endif
 
-/* Stop speculative execution */
-static inline void sync_core(void)
-{
-	int tmp;
-	asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
-}
-
-static inline void __monitor(const void *eax, unsigned long ecx,
-		unsigned long edx)
-{
-	/* "monitor %eax,%ecx,%edx;" */
-	asm volatile(
-		".byte 0x0f,0x01,0xc8;"
-		: :"a" (eax), "c" (ecx), "d"(edx));
-}
-
-static inline void __mwait(unsigned long eax, unsigned long ecx)
-{
-	/* "mwait %eax,%ecx;" */
-	asm volatile(
-		".byte 0x0f,0x01,0xc9;"
-		: :"a" (eax), "c" (ecx));
-}
-
-extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
-
 /* from system description table in BIOS.  Mostly for MCA use, but
 others may find it useful. */
 extern unsigned int machine_id;
@@ -153,20 +123,11 @@ extern unsigned int machine_submodel_id;
 extern unsigned int BIOS_revision;
 extern unsigned int mca_pentium_flag;
 
-/* Boot loader type from the setup header */
-extern int bootloader_type;
-
 /*
  * User space process size: 3GB (default).
  */
 #define TASK_SIZE	(PAGE_OFFSET)
 
-/* This decides where the kernel will search for a free chunk of vm
- * space during mmap's.
- */
-#define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
-
-#define HAVE_ARCH_PICK_MMAP_LAYOUT
 
 /*
  * Size of io_bitmap.
@@ -356,25 +317,9 @@ struct thread_struct {
 	regs->sp = new_esp;					\
 } while (0)
 
-/* Forward declaration, a strange C thing */
-struct task_struct;
-struct mm_struct;
-
-/* Free all resources held by a thread. */
-extern void release_thread(struct task_struct *);
-
-/* Prepare to copy thread state - unlazy all lazy status */
-extern void prepare_to_copy(struct task_struct *tsk);
-
-/*
- * create a kernel thread without removing it from tasklists
- */
-extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
 
 extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
-unsigned long get_wchan(struct task_struct *p);
-
 #define THREAD_SIZE_LONGS      (THREAD_SIZE/sizeof(unsigned long))
 #define KSTK_TOP(info)                                                 \
 ({                                                                     \
@@ -399,53 +344,8 @@ unsigned long get_wchan(struct task_struct *p);
        __regs__ - 1;                                                   \
 })
 
-#define KSTK_EIP(task) (task_pt_regs(task)->ip)
 #define KSTK_ESP(task) (task_pt_regs(task)->sp)
 
-
-struct microcode_header {
-	unsigned int hdrver;
-	unsigned int rev;
-	unsigned int date;
-	unsigned int sig;
-	unsigned int cksum;
-	unsigned int ldrver;
-	unsigned int pf;
-	unsigned int datasize;
-	unsigned int totalsize;
-	unsigned int reserved[3];
-};
-
-struct microcode {
-	struct microcode_header hdr;
-	unsigned int bits[0];
-};
-
-typedef struct microcode microcode_t;
-typedef struct microcode_header microcode_header_t;
-
-/* microcode format is extended from prescott processors */
-struct extended_signature {
-	unsigned int sig;
-	unsigned int pf;
-	unsigned int cksum;
-};
-
-struct extended_sigtable {
-	unsigned int count;
-	unsigned int cksum;
-	unsigned int reserved[3];
-	struct extended_signature sigs[0];
-};
-
-/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
-static inline void rep_nop(void)
-{
-	__asm__ __volatile__("rep;nop": : :"memory");
-}
-
-#define cpu_relax()	rep_nop()
-
 static inline void native_load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
 	tss->x86_tss.sp0 = thread->sp0;
@@ -555,7 +455,6 @@ static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread
    because they are microcoded there and very slow.
    However we don't do prefetches for pre XP Athlons currently
    That should be fixed. */
-#define ARCH_HAS_PREFETCH
 static inline void prefetch(const void *x)
 {
 	alternative_input(ASM_NOP4,
@@ -565,8 +464,6 @@ static inline void prefetch(const void *x)
 }
 
 #define ARCH_HAS_PREFETCH
-#define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
 
 /* 3dnow! prefetch to get an exclusive cache line. Useful for 
    spinlocks to avoid one state transition in the cache coherency protocol. */
@@ -577,13 +474,7 @@ static inline void prefetchw(const void *x)
 			  X86_FEATURE_3DNOW,
 			  "r" (x));
 }
-#define spin_lock_prefetch(x)	prefetchw(x)
-
-extern void select_idle_routine(const struct cpuinfo_x86 *c);
 
-#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
-
-extern unsigned long boot_option_idle_override;
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
@@ -595,6 +486,4 @@ extern void switch_to_new_gdt(void);
 extern void cpu_init(void);
 extern void init_gdt(int cpu);
 
-extern int force_mwait;
-
 #endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 6abe1ba30fe0..5f5c7fc63797 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -83,11 +83,6 @@ DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 extern char ignore_irq13;
 
 extern void identify_cpu(struct cpuinfo_x86 *);
-extern void print_cpu_info(struct cpuinfo_x86 *);
-extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
-extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
-extern unsigned short num_cache_leaves;
-
 
 /*
  * User space process size. 47bits minus one guard page.
@@ -102,8 +97,6 @@ extern unsigned short num_cache_leaves;
 #define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
 #define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
 
-#define TASK_UNMAPPED_BASE	PAGE_ALIGN(TASK_SIZE/3)
-
 /*
  * Size of io_bitmap.
  */
@@ -226,68 +219,16 @@ struct thread_struct {
 	set_fs(USER_DS);							 \
 } while(0) 
 
-struct task_struct;
-struct mm_struct;
-
-/* Free all resources held by a thread. */
-extern void release_thread(struct task_struct *);
-
-/* Prepare to copy thread state - unlazy all lazy status */
-extern void prepare_to_copy(struct task_struct *tsk);
-
-/*
- * create a kernel thread without removing it from tasklists
- */
-extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
-
 /*
  * Return saved PC of a blocked thread.
  * What is this good for? it will be always the scheduler or ret_from_fork.
  */
 #define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8))
 
-extern unsigned long get_wchan(struct task_struct *p);
 #define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
-#define KSTK_EIP(tsk) (task_pt_regs(tsk)->ip)
 #define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
 
 
-struct microcode_header {
-	unsigned int hdrver;
-	unsigned int rev;
-	unsigned int date;
-	unsigned int sig;
-	unsigned int cksum;
-	unsigned int ldrver;
-	unsigned int pf;
-	unsigned int datasize;
-	unsigned int totalsize;
-	unsigned int reserved[3];
-};
-
-struct microcode {
-	struct microcode_header hdr;
-	unsigned int bits[0];
-};
-
-typedef struct microcode microcode_t;
-typedef struct microcode_header microcode_header_t;
-
-/* microcode format is extended from prescott processors */
-struct extended_signature {
-	unsigned int sig;
-	unsigned int pf;
-	unsigned int cksum;
-};
-
-struct extended_sigtable {
-	unsigned int count;
-	unsigned int cksum;
-	unsigned int reserved[3];
-	struct extended_signature sigs[0];
-};
-
-
 #if defined(CONFIG_MPSC) || defined(CONFIG_MCORE2)
 #define ASM_NOP1 P6_NOP1
 #define ASM_NOP2 P6_NOP2
@@ -331,20 +272,6 @@ struct extended_sigtable {
 
 #define ASM_NOP_MAX 8
 
-/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
-static inline void rep_nop(void)
-{
-	__asm__ __volatile__("rep;nop": : :"memory");
-}
-
-/* Stop speculative execution */
-static inline void sync_core(void)
-{ 
-	int tmp;
-	asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
-} 
-
-#define ARCH_HAS_PREFETCHW 1
 static inline void prefetchw(void *x) 
 { 
 	alternative_input("prefetcht0 (%1)",
@@ -353,42 +280,6 @@ static inline void prefetchw(void *x)
 			  "r" (x));
 } 
 
-#define ARCH_HAS_SPINLOCK_PREFETCH 1
-
-#define spin_lock_prefetch(x)  prefetchw(x)
-
-#define cpu_relax()   rep_nop()
-
-static inline void __monitor(const void *eax, unsigned long ecx,
-		unsigned long edx)
-{
-	/* "monitor %eax,%ecx,%edx;" */
-	asm volatile(
-		".byte 0x0f,0x01,0xc8;"
-		: :"a" (eax), "c" (ecx), "d"(edx));
-}
-
-static inline void __mwait(unsigned long eax, unsigned long ecx)
-{
-	/* "mwait %eax,%ecx;" */
-	asm volatile(
-		".byte 0x0f,0x01,0xc9;"
-		: :"a" (eax), "c" (ecx));
-}
-
-static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
-{
-	/* "mwait %eax,%ecx;" */
-	asm volatile(
-		"sti; .byte 0x0f,0x01,0xc9;"
-		: :"a" (eax), "c" (ecx));
-}
-
-extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
-
-extern int force_mwait;
-
-extern void select_idle_routine(const struct cpuinfo_x86 *c);
 
 #define stack_current() \
 ({								\
@@ -397,12 +288,5 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 	ti->task;					\
 })
 
-#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
-
-extern unsigned long boot_option_idle_override;
-/* Boot loader type from the setup header */
-extern int bootloader_type;
-
-#define HAVE_ARCH_PICK_MMAP_LAYOUT 1
 
 #endif /* __ASM_X86_64_PROCESSOR_H */
-- 
cgit v1.2.3


From 0ccb8acc51693a2aef0f38024943808046d81251 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:27 +0100
Subject: x86: unify current_text_addr

current_text_addr() has a different implementation in x86_64 and
i386, but it is not fundamentally different. I stick to the i386
implementation, that seem to be a common base, and move it to processor.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 11 +++++++++++
 include/asm-x86/processor_32.h |  6 ------
 include/asm-x86/processor_64.h |  6 ------
 3 files changed, 11 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 52e3637ef59e..3deb5ba55f55 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -10,6 +10,17 @@ struct mm_struct;
 #include <asm/page.h>
 #include <asm/system.h>
 
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+static inline void *current_text_addr(void)
+{
+	void *pc;
+	asm volatile("mov $1f,%0\n1:":"=r" (pc));
+	return pc;
+}
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 9e119d3789b4..6cd2149dcbad 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -21,12 +21,6 @@
 #include <linux/init.h>
 #include <asm/desc_defs.h>
 
-/*
- * Default implementation of macro that returns current
- * instruction pointer ("program counter").
- */
-#define current_text_addr() ({ void *pc; __asm__("movl $1f,%0\n1:":"=g" (pc)); pc; })
-
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
  *  Members of this structure are referenced in head.S, so think twice
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 5f5c7fc63797..1984a4a38b74 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -20,12 +20,6 @@
 #include <linux/cpumask.h>
 #include <asm/desc_defs.h>
 
-/*
- * Default implementation of macro that returns current
- * instruction pointer ("program counter").
- */
-#define current_text_addr() ({ void *pc; asm volatile("leaq 1f(%%rip),%0\n1:":"=r"(pc)); pc; })
-
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
  */
-- 
cgit v1.2.3


From ca241c75037b32e0216a68e39ad2801d04fa1f87 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:31 +0100
Subject: x86: unify tss_struct

Although slighly different, the tss_struct is very similar in x86_64 and
i386. The really different part, which matchs the hardware vision of it, is
now called x86_hw_tss, and each of the architectures provides yours.
It's then used as a field in the outter tss_struct.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/asm-offsets_64.c |  2 +-
 arch/x86/kernel/doublefault_32.c |  2 +-
 arch/x86/kernel/process_64.c     |  2 +-
 arch/x86/kernel/setup64.c        |  4 +--
 arch/x86/kernel/smpboot_64.c     |  2 +-
 include/asm-x86/lguest.h         |  2 +-
 include/asm-x86/processor.h      | 77 ++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h   | 62 --------------------------------
 include/asm-x86/processor_64.h   | 35 +-----------------
 9 files changed, 85 insertions(+), 103 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index a05428764314..2b32719a3fea 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -110,7 +110,7 @@ int main(void)
 	ENTRY(cr8);
 	BLANK();
 #undef ENTRY
-	DEFINE(TSS_ist, offsetof(struct tss_struct, ist));
+	DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist));
 	BLANK();
 	DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx));
 	BLANK();
diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
index d16122a8e4eb..a47798b59f07 100644
--- a/arch/x86/kernel/doublefault_32.c
+++ b/arch/x86/kernel/doublefault_32.c
@@ -33,7 +33,7 @@ static void doublefault_fn(void)
 		printk(KERN_EMERG "double fault, tss at %08lx\n", tss);
 
 		if (ptr_ok(tss)) {
-			struct i386_hw_tss *t = (struct i386_hw_tss *)tss;
+			struct x86_hw_tss *t = (struct x86_hw_tss *)tss;
 
 			printk(KERN_EMERG "eip = %08lx, esp = %08lx\n",
 			       t->ip, t->sp);
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 836a71adfa11..af56104b73ff 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -639,7 +639,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/*
 	 * Reload esp0, LDT and the page table pointer:
 	 */
-	tss->sp0 = next->sp0;
+	tss->x86_tss.sp0 = next->sp0;
 
 	/* 
 	 * Switch DS and ES.
diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index 05cafcb94109..3b0ffa31f3c0 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -258,10 +258,10 @@ void __cpuinit cpu_init (void)
 				      v, cpu); 
 		}
 		estacks += PAGE_SIZE << order[v];
-		orig_ist->ist[v] = t->ist[v] = (unsigned long)estacks;
+		orig_ist->ist[v] = t->x86_tss.ist[v] = (unsigned long)estacks;
 	}
 
-	t->io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
 	/*
 	 * <= is required because the CPU will access up to
 	 * 8 bits beyond the end of the IO permission bitmap.
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 4c03ddccd681..2ea02a71b644 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -614,7 +614,7 @@ do_rest:
 	start_rip = setup_trampoline();
 
 	init_rsp = c_idle.idle->thread.sp;
-	per_cpu(init_tss,cpu).sp0 = init_rsp;
+	per_cpu(init_tss, cpu).x86_tss.sp0 = init_rsp;
 	initial_code = start_secondary;
 	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
 
diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h
index 3585a1628b59..1c8367a692f6 100644
--- a/include/asm-x86/lguest.h
+++ b/include/asm-x86/lguest.h
@@ -51,7 +51,7 @@ struct lguest_ro_state
 	/* Fields which are used when guest is running. */
 	struct desc_ptr guest_idt_desc;
 	struct desc_ptr guest_gdt_desc;
-	struct i386_hw_tss guest_tss;
+	struct x86_hw_tss guest_tss;
 	struct desc_struct guest_idt[IDT_ENTRIES];
 	struct desc_struct guest_gdt[GDT_ENTRIES];
 };
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 3deb5ba55f55..cede9ad3dc6e 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -8,6 +8,7 @@ struct task_struct;
 struct mm_struct;
 
 #include <asm/page.h>
+#include <asm/percpu.h>
 #include <asm/system.h>
 
 /*
@@ -38,6 +39,82 @@ static inline void load_cr3(pgd_t *pgdir)
 	write_cr3(__pa(pgdir));
 }
 
+#ifdef CONFIG_X86_32
+/* This is the TSS defined by the hardware. */
+struct x86_hw_tss {
+	unsigned short	back_link, __blh;
+	unsigned long	sp0;
+	unsigned short	ss0, __ss0h;
+	unsigned long	sp1;
+	unsigned short	ss1, __ss1h;	/* ss1 caches MSR_IA32_SYSENTER_CS */
+	unsigned long	sp2;
+	unsigned short	ss2, __ss2h;
+	unsigned long	__cr3;
+	unsigned long	ip;
+	unsigned long	flags;
+	unsigned long	ax, cx, dx, bx;
+	unsigned long	sp, bp, si, di;
+	unsigned short	es, __esh;
+	unsigned short	cs, __csh;
+	unsigned short	ss, __ssh;
+	unsigned short	ds, __dsh;
+	unsigned short	fs, __fsh;
+	unsigned short	gs, __gsh;
+	unsigned short	ldt, __ldth;
+	unsigned short	trace, io_bitmap_base;
+} __attribute__((packed));
+#else
+struct x86_hw_tss {
+	u32 reserved1;
+	u64 sp0;
+	u64 sp1;
+	u64 sp2;
+	u64 reserved2;
+	u64 ist[7];
+	u32 reserved3;
+	u32 reserved4;
+	u16 reserved5;
+	u16 io_bitmap_base;
+} __attribute__((packed)) ____cacheline_aligned;
+#endif
+
+/*
+ * Size of io_bitmap.
+ */
+#define IO_BITMAP_BITS  65536
+#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
+#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
+#define IO_BITMAP_OFFSET offsetof(struct tss_struct, io_bitmap)
+#define INVALID_IO_BITMAP_OFFSET 0x8000
+#define INVALID_IO_BITMAP_OFFSET_LAZY 0x9000
+
+struct tss_struct {
+	struct x86_hw_tss x86_tss;
+
+	/*
+	 * The extra 1 is there because the CPU will access an
+	 * additional byte beyond the end of the IO permission
+	 * bitmap. The extra byte must be all 1 bits, and must
+	 * be within the limit.
+	 */
+	unsigned long	io_bitmap[IO_BITMAP_LONGS + 1];
+	/*
+	 * Cache the current maximum and the last task that used the bitmap:
+	 */
+	unsigned long io_bitmap_max;
+	struct thread_struct *io_bitmap_owner;
+	/*
+	 * pads the TSS to be cacheline-aligned (size is 0x100)
+	 */
+	unsigned long __cacheline_filler[35];
+	/*
+	 * .. and then another 0x100 bytes for emergency kernel stack
+	 */
+	unsigned long stack[64];
+} __attribute__((packed));
+
+DECLARE_PER_CPU(struct tss_struct, init_tss);
+
 #ifdef CONFIG_X86_32
 # include "processor_32.h"
 #else
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 6cd2149dcbad..57b345bc3c74 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -81,7 +81,6 @@ struct cpuinfo_x86 {
 extern struct cpuinfo_x86 boot_cpu_data;
 extern struct cpuinfo_x86 new_cpu_data;
 extern struct tss_struct doublefault_tss;
-DECLARE_PER_CPU(struct tss_struct, init_tss);
 
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
@@ -123,16 +122,6 @@ extern unsigned int mca_pentium_flag;
 #define TASK_SIZE	(PAGE_OFFSET)
 
 
-/*
- * Size of io_bitmap.
- */
-#define IO_BITMAP_BITS  65536
-#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
-#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
-#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
-#define INVALID_IO_BITMAP_OFFSET 0x8000
-#define INVALID_IO_BITMAP_OFFSET_LAZY 0x9000
-
 struct i387_fsave_struct {
 	long	cwd;
 	long	swd;
@@ -185,57 +174,6 @@ typedef struct {
 	unsigned long seg;
 } mm_segment_t;
 
-struct thread_struct;
-
-/* This is the TSS defined by the hardware. */
-struct i386_hw_tss {
-	unsigned short	back_link,__blh;
-	unsigned long	sp0;
-	unsigned short	ss0,__ss0h;
-	unsigned long	sp1;
-	unsigned short	ss1,__ss1h;	/* ss1 is used to cache MSR_IA32_SYSENTER_CS */
-	unsigned long	sp2;
-	unsigned short	ss2,__ss2h;
-	unsigned long	__cr3;
-	unsigned long	ip;
-	unsigned long	flags;
-	unsigned long	ax, cx, dx, bx;
-	unsigned long	sp, bp, si, di;
-	unsigned short	es, __esh;
-	unsigned short	cs, __csh;
-	unsigned short	ss, __ssh;
-	unsigned short	ds, __dsh;
-	unsigned short	fs, __fsh;
-	unsigned short	gs, __gsh;
-	unsigned short	ldt, __ldth;
-	unsigned short	trace, io_bitmap_base;
-} __attribute__((packed));
-
-struct tss_struct {
-	struct i386_hw_tss x86_tss;
-
-	/*
-	 * The extra 1 is there because the CPU will access an
-	 * additional byte beyond the end of the IO permission
-	 * bitmap. The extra byte must be all 1 bits, and must
-	 * be within the limit.
-	 */
-	unsigned long	io_bitmap[IO_BITMAP_LONGS + 1];
-	/*
-	 * Cache the current maximum and the last task that used the bitmap:
-	 */
-	unsigned long io_bitmap_max;
-	struct thread_struct *io_bitmap_owner;
-	/*
-	 * pads the TSS to be cacheline-aligned (size is 0x100)
-	 */
-	unsigned long __cacheline_filler[35];
-	/*
-	 * .. and then another 0x100 bytes for emergency kernel stack
-	 */
-	unsigned long stack[64];
-} __attribute__((packed));
-
 #define ARCH_MIN_TASKALIGN	16
 
 struct thread_struct {
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 1984a4a38b74..8d342c23ad14 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -91,14 +91,6 @@ extern void identify_cpu(struct cpuinfo_x86 *);
 #define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
 #define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
 
-/*
- * Size of io_bitmap.
- */
-#define IO_BITMAP_BITS  65536
-#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
-#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
-#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
-#define INVALID_IO_BITMAP_OFFSET 0x8000
 
 struct i387_fxsave_struct {
 	u16	cwd;
@@ -118,32 +110,7 @@ union i387_union {
 	struct i387_fxsave_struct	fxsave;
 };
 
-struct tss_struct {
-	u32 reserved1;
-	u64 sp0;
-	u64 sp1;
-	u64 sp2;
-	u64 reserved2;
-	u64 ist[7];
-	u32 reserved3;
-	u32 reserved4;
-	u16 reserved5;
-	u16 io_bitmap_base;
-	/*
-	 * The extra 1 is there because the CPU will access an
-	 * additional byte beyond the end of the IO permission
-	 * bitmap. The extra byte must be all 1 bits, and must
-	 * be within the limit. Thus we have:
-	 *
-	 * 128 bytes, the bitmap itself, for ports 0..0x3ff
-	 * 8 bytes, for an extra "long" of ~0UL
-	 */
-	unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
-} __attribute__((packed)) ____cacheline_aligned;
-
-
 extern struct cpuinfo_x86 boot_cpu_data;
-DECLARE_PER_CPU(struct tss_struct,init_tss);
 /* Save the original ist values for checking stack pointers during debugging */
 struct orig_ist {
 	unsigned long ist[7];
@@ -195,7 +162,7 @@ struct thread_struct {
 }
 
 #define INIT_TSS  { \
-	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
 
 #define INIT_MMAP \
-- 
cgit v1.2.3


From 7818a1e0294debee02d5135e17b89f28b8871887 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:31 +0100
Subject: x86: provide 64-bit with a load_sp0 function.

Paravirt guests need to inform the underlying hypervisor whenever the sp0
tss field changes. i386 already has such a function, and we use it for
x86_64 too. There's an unnecessary (for 64-bit) msr handling part in the original
version, and it is placed around an ifdef. Making no more sense in
processor_32.h, it is moved to the common header

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_64.c   |  2 +-
 arch/x86/kernel/smpboot_64.c   |  2 +-
 include/asm-x86/processor.h    | 22 +++++++++++++++++++++-
 include/asm-x86/processor_32.h | 20 --------------------
 4 files changed, 23 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index af56104b73ff..e3a3610ade10 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -639,7 +639,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/*
 	 * Reload esp0, LDT and the page table pointer:
 	 */
-	tss->x86_tss.sp0 = next->sp0;
+	load_sp0(tss, next);
 
 	/* 
 	 * Switch DS and ES.
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 2ea02a71b644..5bd42ce144da 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -614,7 +614,7 @@ do_rest:
 	start_rip = setup_trampoline();
 
 	init_rsp = c_idle.idle->thread.sp;
-	per_cpu(init_tss, cpu).x86_tss.sp0 = init_rsp;
+	load_sp0(&per_cpu(init_tss, cpu), &c_idle.idle->thread);
 	initial_code = start_secondary;
 	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
 
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index cede9ad3dc6e..b1ea52156362 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -193,8 +193,22 @@ static inline void native_set_iopl_mask(unsigned mask)
 #endif
 }
 
+static inline void native_load_sp0(struct tss_struct *tss,
+				   struct thread_struct *thread)
+{
+	tss->x86_tss.sp0 = thread->sp0;
+#ifdef CONFIG_X86_32
+	/* Only happens when SEP is enabled, no need to test "SEP"arately */
+	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
+		tss->x86_tss.ss1 = thread->sysenter_cs;
+		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
+	}
+#endif
+}
 
-#ifndef CONFIG_PARAVIRT
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
 #define __cpuid native_cpuid
 #define paravirt_enabled() 0
 
@@ -206,6 +220,12 @@ static inline void native_set_iopl_mask(unsigned mask)
 #define set_debugreg(value, register)				\
 	native_set_debugreg(register, value)
 
+static inline void load_sp0(struct tss_struct *tss,
+			    struct thread_struct *thread)
+{
+	native_load_sp0(tss, thread);
+}
+
 #define set_iopl_mask native_set_iopl_mask
 #endif /* CONFIG_PARAVIRT */
 
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 57b345bc3c74..53037d1a6ae6 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -278,26 +278,6 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
 #define KSTK_ESP(task) (task_pt_regs(task)->sp)
 
-static inline void native_load_sp0(struct tss_struct *tss, struct thread_struct *thread)
-{
-	tss->x86_tss.sp0 = thread->sp0;
-	/* This can only happen when SEP is enabled, no need to test "SEP"arately */
-	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
-		tss->x86_tss.ss1 = thread->sysenter_cs;
-		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
-	}
-}
-
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-
-static inline void load_sp0(struct tss_struct *tss, struct thread_struct *thread)
-{
-	native_load_sp0(tss, thread);
-}
-#endif /* CONFIG_PARAVIRT */
-
 /* generic versions from gas */
 #define GENERIC_NOP1	".byte 0x90\n"
 #define GENERIC_NOP2    	".byte 0x89,0xf6\n"
-- 
cgit v1.2.3


From cb38d377337cadd73fe4c582f77c5273685aeb71 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:31 +0100
Subject: x86: unify thread struct.

The thread_struct is not fundamentally different between architectures, and
this patch puts it in the common header. What's really unique for each of
them is enclosed in ifdefs.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 44 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 37 -----------------------------------
 include/asm-x86/processor_64.h | 32 ------------------------------
 3 files changed, 44 insertions(+), 69 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index b1ea52156362..0e82ad5a22a4 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -126,6 +126,50 @@ extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
 
+struct thread_struct {
+/* cached TLS descriptors. */
+	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
+	unsigned long	sp0;
+	unsigned long	sp;
+#ifdef CONFIG_X86_32
+	unsigned long	sysenter_cs;
+#else
+	unsigned long 	usersp;	/* Copy from PDA */
+	unsigned short	es, ds, fsindex, gsindex;
+#endif
+	unsigned long	ip;
+	unsigned long	fs;
+	unsigned long	gs;
+/* Hardware debugging registers */
+	unsigned long	debugreg0;
+	unsigned long	debugreg1;
+	unsigned long	debugreg2;
+	unsigned long	debugreg3;
+	unsigned long	debugreg6;
+	unsigned long	debugreg7;
+/* fault info */
+	unsigned long	cr2, trap_no, error_code;
+/* floating point info */
+	union i387_union	i387 __attribute__((aligned(16)));;
+#ifdef CONFIG_X86_32
+/* virtual 86 mode info */
+	struct vm86_struct __user *vm86_info;
+	unsigned long		screen_bitmap;
+	unsigned long		v86flags, v86mask, saved_sp0;
+	unsigned int		saved_fs, saved_gs;
+#endif
+/* IO permissions */
+	unsigned long	*io_bitmap_ptr;
+	unsigned long	iopl;
+/* max allowed port in the bitmap, in bytes: */
+	unsigned io_bitmap_max;
+/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
+	unsigned long	debugctlmsr;
+/* Debug Store - if not 0 points to a DS Save Area configuration;
+ *               goes into MSR_IA32_DS_AREA */
+	unsigned long	ds_area_msr;
+};
+
 static inline unsigned long native_get_debugreg(int regno)
 {
 	unsigned long val = 0; 	/* Damn you, gcc! */
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 53037d1a6ae6..8cb05cec3cb9 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -176,43 +176,6 @@ typedef struct {
 
 #define ARCH_MIN_TASKALIGN	16
 
-struct thread_struct {
-/* cached TLS descriptors. */
-	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
-	unsigned long	sp0;
-	unsigned long	sysenter_cs;
-	unsigned long	ip;
-	unsigned long	sp;
-	unsigned long	fs;
-	unsigned long	gs;
-/* Hardware debugging registers */
-	unsigned long	debugreg0;
-	unsigned long	debugreg1;
-	unsigned long	debugreg2;
-	unsigned long	debugreg3;
-	unsigned long	debugreg6;
-	unsigned long	debugreg7;
-/* fault info */
-	unsigned long	cr2, trap_no, error_code;
-/* floating point info */
-	union i387_union	i387;
-/* virtual 86 mode info */
-	struct vm86_struct __user * vm86_info;
-	unsigned long		screen_bitmap;
-	unsigned long		v86flags, v86mask, saved_sp0;
-	unsigned int		saved_fs, saved_gs;
-/* IO permissions */
-	unsigned long	*io_bitmap_ptr;
- 	unsigned long	iopl;
-/* max allowed port in the bitmap, in bytes: */
-	unsigned long	io_bitmap_max;
-/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
-	unsigned long	debugctlmsr;
-/* Debug Store - if not 0 points to a DS Save Area configuration;
- *               goes into MSR_IA32_DS_AREA */
-	unsigned long	ds_area_msr;
-};
-
 #define INIT_THREAD  {							\
 	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
 	.vm86_info = NULL,						\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 8d342c23ad14..4b4ec3d79910 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -125,38 +125,6 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #define ARCH_MIN_MMSTRUCT_ALIGN	0
 #endif
 
-struct thread_struct {
-	unsigned long	sp0;
-	unsigned long	sp;
-	unsigned long 	usersp;	/* Copy from PDA */
-	unsigned long	fs;
-	unsigned long	gs;
-	unsigned short	es, ds, fsindex, gsindex;	
-/* Hardware debugging registers */
-	unsigned long	debugreg0;  
-	unsigned long	debugreg1;  
-	unsigned long	debugreg2;  
-	unsigned long	debugreg3;  
-	unsigned long	debugreg6;  
-	unsigned long	debugreg7;  
-/* fault info */
-	unsigned long	cr2, trap_no, error_code;
-/* floating point info */
-	union i387_union	i387  __attribute__((aligned(16)));
-/* IO permissions. the bitmap could be moved into the GDT, that would make
-   switch faster for a limited number of ioperm using tasks. -AK */
-	int		ioperm;
-	unsigned long	*io_bitmap_ptr;
-	unsigned io_bitmap_max;
-/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
-	unsigned long	debugctlmsr;
-/* Debug Store - if not 0 points to a DS Save Area configuration;
- *               goes into MSR_IA32_DS_AREA */
-	unsigned long	ds_area_msr;
-/* cached TLS descriptors. */
-	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
-} __attribute__((aligned(16)));
-
 #define INIT_THREAD  { \
 	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
-- 
cgit v1.2.3


From dbcb4660246c240a159b4037067fdedb563a63cb Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:31 +0100
Subject: x86: unify TASK_ALIGN definitions

This patch moves the TASK_ALIGN constraints to common header.
The base of it is the same for x86_64 and i386. The only difference
is the presence of vSMP in x86_64. As it's not a worry in i386, we can
safely use the same code for both.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 8 ++++++++
 include/asm-x86/processor_32.h | 1 -
 include/asm-x86/processor_64.h | 8 --------
 3 files changed, 8 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 0e82ad5a22a4..8b56c23f043d 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -22,6 +22,14 @@ static inline void *current_text_addr(void)
 	return pc;
 }
 
+#ifdef CONFIG_X86_VSMP
+#define ARCH_MIN_TASKALIGN	(1 << INTERNODE_CACHE_SHIFT)
+#define ARCH_MIN_MMSTRUCT_ALIGN	(1 << INTERNODE_CACHE_SHIFT)
+#else
+#define ARCH_MIN_TASKALIGN	16
+#define ARCH_MIN_MMSTRUCT_ALIGN	0
+#endif
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 8cb05cec3cb9..8380243a19d3 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -174,7 +174,6 @@ typedef struct {
 	unsigned long seg;
 } mm_segment_t;
 
-#define ARCH_MIN_TASKALIGN	16
 
 #define INIT_THREAD  {							\
 	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 4b4ec3d79910..7fe5c1461c83 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -117,14 +117,6 @@ struct orig_ist {
 };
 DECLARE_PER_CPU(struct orig_ist, orig_ist);
 
-#ifdef CONFIG_X86_VSMP
-#define ARCH_MIN_TASKALIGN	(1 << INTERNODE_CACHE_SHIFT)
-#define ARCH_MIN_MMSTRUCT_ALIGN	(1 << INTERNODE_CACHE_SHIFT)
-#else
-#define ARCH_MIN_TASKALIGN	16
-#define ARCH_MIN_MMSTRUCT_ALIGN	0
-#endif
-
 #define INIT_THREAD  { \
 	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
-- 
cgit v1.2.3


From 26996dd22b3cbc9dbe18cf908d2f844a116b6ec1 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:31 +0100
Subject: x86: change bitwise operations to get a void parameter.

This patch changes the bitwise operations in bitops.h to get
a void pointers as a parameter. Before this patch, a lot of warnings
can be seen. They're gone after it.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/bitops.h | 39 ++++++++++++++++++++-------------------
 1 file changed, 20 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h
index c6dd7e259b46..1a23ce1a5697 100644
--- a/include/asm-x86/bitops.h
+++ b/include/asm-x86/bitops.h
@@ -43,7 +43,7 @@
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void set_bit(int nr, volatile unsigned long *addr)
+static inline void set_bit(int nr, volatile void *addr)
 {
 	asm volatile(LOCK_PREFIX "bts %1,%0"
 		     : ADDR
@@ -59,7 +59,7 @@ static inline void set_bit(int nr, volatile unsigned long *addr)
  * If it's called on the same region of memory simultaneously, the effect
  * may be that only one operation succeeds.
  */
-static inline void __set_bit(int nr, volatile unsigned long *addr)
+static inline void __set_bit(int nr, volatile void *addr)
 {
 	asm volatile("bts %1,%0"
 		     : ADDR
@@ -77,7 +77,7 @@ static inline void __set_bit(int nr, volatile unsigned long *addr)
  * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
  * in order to ensure changes are visible on other processors.
  */
-static inline void clear_bit(int nr, volatile unsigned long *addr)
+static inline void clear_bit(int nr, volatile void *addr)
 {
 	asm volatile(LOCK_PREFIX "btr %1,%0"
 		     : ADDR
@@ -92,13 +92,13 @@ static inline void clear_bit(int nr, volatile unsigned long *addr)
  * clear_bit() is atomic and implies release semantics before the memory
  * operation. It can be used for an unlock.
  */
-static inline void clear_bit_unlock(unsigned nr, volatile unsigned long *addr)
+static inline void clear_bit_unlock(unsigned nr, volatile void *addr)
 {
 	barrier();
 	clear_bit(nr, addr);
 }
 
-static inline void __clear_bit(int nr, volatile unsigned long *addr)
+static inline void __clear_bit(int nr, volatile void *addr)
 {
 	asm volatile("btr %1,%0" : ADDR : "Ir" (nr));
 }
@@ -115,7 +115,7 @@ static inline void __clear_bit(int nr, volatile unsigned long *addr)
  * No memory barrier is required here, because x86 cannot reorder stores past
  * older loads. Same principle as spin_unlock.
  */
-static inline void __clear_bit_unlock(unsigned nr, volatile unsigned long *addr)
+static inline void __clear_bit_unlock(unsigned nr, volatile void *addr)
 {
 	barrier();
 	__clear_bit(nr, addr);
@@ -133,7 +133,7 @@ static inline void __clear_bit_unlock(unsigned nr, volatile unsigned long *addr)
  * If it's called on the same region of memory simultaneously, the effect
  * may be that only one operation succeeds.
  */
-static inline void __change_bit(int nr, volatile unsigned long *addr)
+static inline void __change_bit(int nr, volatile void *addr)
 {
 	asm volatile("btc %1,%0" : ADDR : "Ir" (nr));
 }
@@ -147,7 +147,7 @@ static inline void __change_bit(int nr, volatile unsigned long *addr)
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void change_bit(int nr, volatile unsigned long *addr)
+static inline void change_bit(int nr, volatile void *addr)
 {
 	asm volatile(LOCK_PREFIX "btc %1,%0"
 		     : ADDR : "Ir" (nr));
@@ -161,7 +161,7 @@ static inline void change_bit(int nr, volatile unsigned long *addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int test_and_set_bit(int nr, volatile unsigned long *addr)
+static inline int test_and_set_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -180,7 +180,7 @@ static inline int test_and_set_bit(int nr, volatile unsigned long *addr)
  *
  * This is the same as test_and_set_bit on x86.
  */
-static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr)
+static inline int test_and_set_bit_lock(int nr, volatile void *addr)
 {
 	return test_and_set_bit(nr, addr);
 }
@@ -194,7 +194,7 @@ static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr)
  * If two examples of this operation race, one can appear to succeed
  * but actually fail.  You must protect multiple accesses with a lock.
  */
-static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
+static inline int __test_and_set_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -213,7 +213,7 @@ static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int test_and_clear_bit(int nr, volatile unsigned long *addr)
+static inline int test_and_clear_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -234,7 +234,7 @@ static inline int test_and_clear_bit(int nr, volatile unsigned long *addr)
  * If two examples of this operation race, one can appear to succeed
  * but actually fail.  You must protect multiple accesses with a lock.
  */
-static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
+static inline int __test_and_clear_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -246,7 +246,7 @@ static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
 }
 
 /* WARNING: non atomic and it can be reordered! */
-static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
+static inline int __test_and_change_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -266,7 +266,7 @@ static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int test_and_change_bit(int nr, volatile unsigned long *addr)
+static inline int test_and_change_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
@@ -278,19 +278,20 @@ static inline int test_and_change_bit(int nr, volatile unsigned long *addr)
 	return oldbit;
 }
 
-static inline int constant_test_bit(int nr, const volatile unsigned long *addr)
+static inline int constant_test_bit(int nr, const volatile void *addr)
 {
-	return ((1UL << (nr % BITS_PER_LONG)) & (addr[nr / BITS_PER_LONG])) != 0;
+	return ((1UL << (nr % BITS_PER_LONG)) &
+		(((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0;
 }
 
-static inline int variable_test_bit(int nr, volatile const unsigned long *addr)
+static inline int variable_test_bit(int nr, volatile const void *addr)
 {
 	int oldbit;
 
 	asm volatile("bt %2,%1\n\t"
 		     "sbb %0,%0"
 		     : "=r" (oldbit)
-		     : "m" (*addr), "Ir" (nr));
+		     : "m" (*(unsigned long *)addr), "Ir" (nr));
 
 	return oldbit;
 }
-- 
cgit v1.2.3


From 5300db887e251276eaab2801af297acf4b53b9eb Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:33 +0100
Subject: x86: unify x86_cpuinfo struct.

x86_cpuinfo is one more to the family of "not fundamentally different"
structs. It's unified in processor.h, with very specific fields enclosed
around ifdefs.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/common.c   |  6 ++--
 include/asm-x86/processor.h    | 82 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 67 ----------------------------------
 include/asm-x86/processor_64.h | 57 -----------------------------
 4 files changed, 85 insertions(+), 127 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 69507ae8a65b..ecd13c0e8542 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -427,7 +427,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 
 	printk(KERN_DEBUG "CPU: After generic identify, caps:");
 	for (i = 0; i < NCAPINTS; i++)
-		printk(" %08lx", c->x86_capability[i]);
+		printk(" %08x", c->x86_capability[i]);
 	printk("\n");
 
 	if (this_cpu->c_identify) {
@@ -435,7 +435,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 
 		printk(KERN_DEBUG "CPU: After vendor identify, caps:");
 		for (i = 0; i < NCAPINTS; i++)
-			printk(" %08lx", c->x86_capability[i]);
+			printk(" %08x", c->x86_capability[i]);
 		printk("\n");
 	}
 
@@ -493,7 +493,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 
 	printk(KERN_DEBUG "CPU: After all inits, caps:");
 	for (i = 0; i < NCAPINTS; i++)
-		printk(" %08lx", c->x86_capability[i]);
+		printk(" %08x", c->x86_capability[i]);
 	printk("\n");
 
 	/*
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 8b56c23f043d..2b896b0aa3fa 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -10,6 +10,9 @@ struct mm_struct;
 #include <asm/page.h>
 #include <asm/percpu.h>
 #include <asm/system.h>
+#include <asm/percpu.h>
+#include <linux/cpumask.h>
+#include <linux/cache.h>
 
 /*
  * Default implementation of macro that returns current
@@ -30,6 +33,85 @@ static inline void *current_text_addr(void)
 #define ARCH_MIN_MMSTRUCT_ALIGN	0
 #endif
 
+/*
+ *  CPU type and hardware bug flags. Kept separately for each CPU.
+ *  Members of this structure are referenced in head.S, so think twice
+ *  before touching them. [mj]
+ */
+
+struct cpuinfo_x86 {
+	__u8	x86;		/* CPU family */
+	__u8	x86_vendor;	/* CPU vendor */
+	__u8	x86_model;
+	__u8	x86_mask;
+#ifdef CONFIG_X86_32
+	char	wp_works_ok;	/* It doesn't on 386's */
+	char	hlt_works_ok;	/* Problems on some 486Dx4's and old 386's */
+	char	hard_math;
+	char	rfu;
+	char	fdiv_bug;
+	char	f00f_bug;
+	char	coma_bug;
+	char	pad0;
+#else
+	/* number of 4K pages in DTLB/ITLB combined(in pages)*/
+	int     x86_tlbsize;
+	__u8    x86_virt_bits, x86_phys_bits;
+	/* cpuid returned core id bits */
+	__u8    x86_coreid_bits;
+	/* Max extended CPUID function supported */
+	__u32   extended_cpuid_level;
+#endif
+	int	cpuid_level;	/* Maximum supported CPUID level, -1=no CPUID */
+	__u32	x86_capability[NCAPINTS];
+	char	x86_vendor_id[16];
+	char	x86_model_id[64];
+	int 	x86_cache_size;  /* in KB - valid for CPUS which support this
+				    call  */
+	int 	x86_cache_alignment;	/* In bytes */
+	int	x86_power;
+	unsigned long loops_per_jiffy;
+#ifdef CONFIG_SMP
+	cpumask_t llc_shared_map;	/* cpus sharing the last level cache */
+#endif
+	unsigned char x86_max_cores;	/* cpuid returned max cores value */
+	unsigned char apicid;
+	unsigned short x86_clflush_size;
+#ifdef CONFIG_SMP
+	unsigned char booted_cores;	/* number of cores as seen by OS */
+	__u8 phys_proc_id; 		/* Physical processor id. */
+	__u8 cpu_core_id;  		/* Core id */
+	__u8 cpu_index;			/* index into per_cpu list */
+#endif
+} __attribute__((__aligned__(SMP_CACHE_BYTES)));
+
+#define X86_VENDOR_INTEL 0
+#define X86_VENDOR_CYRIX 1
+#define X86_VENDOR_AMD 2
+#define X86_VENDOR_UMC 3
+#define X86_VENDOR_NEXGEN 4
+#define X86_VENDOR_CENTAUR 5
+#define X86_VENDOR_TRANSMETA 7
+#define X86_VENDOR_NSC 8
+#define X86_VENDOR_NUM 9
+#define X86_VENDOR_UNKNOWN 0xff
+
+extern struct cpuinfo_x86 boot_cpu_data;
+
+#ifdef CONFIG_SMP
+DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
+#define cpu_data(cpu)		per_cpu(cpu_info, cpu)
+#define current_cpu_data	cpu_data(smp_processor_id())
+#else
+#define cpu_data(cpu)		boot_cpu_data
+#define current_cpu_data	boot_cpu_data
+#endif
+
+extern void print_cpu_info(struct cpuinfo_x86 *);
+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
+extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
+extern unsigned short num_cache_leaves;
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 8380243a19d3..f0f7ca48444e 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -14,83 +14,16 @@
 #include <asm/cpufeature.h>
 #include <asm/msr.h>
 #include <asm/system.h>
-#include <linux/cache.h>
 #include <linux/threads.h>
-#include <asm/percpu.h>
-#include <linux/cpumask.h>
 #include <linux/init.h>
 #include <asm/desc_defs.h>
 
-/*
- *  CPU type and hardware bug flags. Kept separately for each CPU.
- *  Members of this structure are referenced in head.S, so think twice
- *  before touching them. [mj]
- */
-
-struct cpuinfo_x86 {
-	__u8	x86;		/* CPU family */
-	__u8	x86_vendor;	/* CPU vendor */
-	__u8	x86_model;
-	__u8	x86_mask;
-	char	wp_works_ok;	/* It doesn't on 386's */
-	char	hlt_works_ok;	/* Problems on some 486Dx4's and old 386's */
-	char	hard_math;
-	char	rfu;
-       	int	cpuid_level;	/* Maximum supported CPUID level, -1=no CPUID */
-	unsigned long	x86_capability[NCAPINTS];
-	char	x86_vendor_id[16];
-	char	x86_model_id[64];
-	int 	x86_cache_size;  /* in KB - valid for CPUS which support this
-				    call  */
-	int 	x86_cache_alignment;	/* In bytes */
-	char	fdiv_bug;
-	char	f00f_bug;
-	char	coma_bug;
-	char	pad0;
-	int	x86_power;
-	unsigned long loops_per_jiffy;
-#ifdef CONFIG_SMP
-	cpumask_t llc_shared_map;	/* cpus sharing the last level cache */
-#endif
-	unsigned char x86_max_cores;	/* cpuid returned max cores value */
-	unsigned char apicid;
-	unsigned short x86_clflush_size;
-#ifdef CONFIG_SMP
-	unsigned char booted_cores;	/* number of cores as seen by OS */
-	__u8 phys_proc_id; 		/* Physical processor id. */
-	__u8 cpu_core_id;  		/* Core id */
-	__u8 cpu_index;			/* index into per_cpu list */
-#endif
-} __attribute__((__aligned__(SMP_CACHE_BYTES)));
-
-#define X86_VENDOR_INTEL 0
-#define X86_VENDOR_CYRIX 1
-#define X86_VENDOR_AMD 2
-#define X86_VENDOR_UMC 3
-#define X86_VENDOR_NEXGEN 4
-#define X86_VENDOR_CENTAUR 5
-#define X86_VENDOR_TRANSMETA 7
-#define X86_VENDOR_NSC 8
-#define X86_VENDOR_NUM 9
-#define X86_VENDOR_UNKNOWN 0xff
-
 /*
  * capabilities of CPUs
  */
-
-extern struct cpuinfo_x86 boot_cpu_data;
 extern struct cpuinfo_x86 new_cpu_data;
 extern struct tss_struct doublefault_tss;
 
-#ifdef CONFIG_SMP
-DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
-#define cpu_data(cpu)		per_cpu(cpu_info, cpu)
-#define current_cpu_data	cpu_data(smp_processor_id())
-#else
-#define cpu_data(cpu)		boot_cpu_data
-#define current_cpu_data	boot_cpu_data
-#endif
-
 /*
  * the following now lives in the per cpu area:
  * extern	int cpu_llc_id[NR_CPUS];
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 7fe5c1461c83..5045abe91e37 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -15,65 +15,9 @@
 #include <asm/current.h>
 #include <asm/system.h>
 #include <asm/mmsegment.h>
-#include <asm/percpu.h>
 #include <linux/personality.h>
-#include <linux/cpumask.h>
 #include <asm/desc_defs.h>
 
-/*
- *  CPU type and hardware bug flags. Kept separately for each CPU.
- */
-
-struct cpuinfo_x86 {
-	__u8	x86;		/* CPU family */
-	__u8	x86_vendor;	/* CPU vendor */
-	__u8	x86_model;
-	__u8	x86_mask;
-	int	cpuid_level;	/* Maximum supported CPUID level, -1=no CPUID */
-	__u32	x86_capability[NCAPINTS];
-	char	x86_vendor_id[16];
-	char	x86_model_id[64];
-	int 	x86_cache_size;  /* in KB */
-	int	x86_clflush_size;
-	int	x86_cache_alignment;
-	int	x86_tlbsize;	/* number of 4K pages in DTLB/ITLB combined(in pages)*/
-        __u8    x86_virt_bits, x86_phys_bits;
-	__u8	x86_max_cores;	/* cpuid returned max cores value */
-	__u8	x86_coreid_bits; /* cpuid returned core id bits */
-        __u32   x86_power; 	
-	__u32   extended_cpuid_level;	/* Max extended CPUID function supported */
-	unsigned long loops_per_jiffy;
-#ifdef CONFIG_SMP
-	cpumask_t llc_shared_map;	/* cpus sharing the last level cache */
-#endif
-	__u8	apicid;
-#ifdef CONFIG_SMP
-	__u8	booted_cores;	/* number of cores as seen by OS */
-	__u8	phys_proc_id;	/* Physical Processor id. */
-	__u8	cpu_core_id;	/* Core id. */
-	__u8	cpu_index;	/* index into per_cpu list */
-#endif
-} ____cacheline_aligned;
-
-#define X86_VENDOR_INTEL 0
-#define X86_VENDOR_CYRIX 1
-#define X86_VENDOR_AMD 2
-#define X86_VENDOR_UMC 3
-#define X86_VENDOR_NEXGEN 4
-#define X86_VENDOR_CENTAUR 5
-#define X86_VENDOR_TRANSMETA 7
-#define X86_VENDOR_NUM 8
-#define X86_VENDOR_UNKNOWN 0xff
-
-#ifdef CONFIG_SMP
-DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
-#define cpu_data(cpu)		per_cpu(cpu_info, cpu)
-#define current_cpu_data	cpu_data(smp_processor_id())
-#else
-#define cpu_data(cpu)		boot_cpu_data
-#define current_cpu_data	boot_cpu_data
-#endif
-
 extern char ignore_irq13;
 
 extern void identify_cpu(struct cpuinfo_x86 *);
@@ -110,7 +54,6 @@ union i387_union {
 	struct i387_fxsave_struct	fxsave;
 };
 
-extern struct cpuinfo_x86 boot_cpu_data;
 /* Save the original ist values for checking stack pointers during debugging */
 struct orig_ist {
 	unsigned long ist[7];
-- 
cgit v1.2.3


From 2c1c468a382e64afccacdfe03fa4c0615cd4264f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:37 +0100
Subject: x86: remove legacy stuff from processor_64.h

This patch removes definitions and macros that are not used anymore
from processor_64.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor_64.h | 14 --------------
 1 file changed, 14 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 5045abe91e37..bfc4ce6e2f00 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -18,8 +18,6 @@
 #include <linux/personality.h>
 #include <asm/desc_defs.h>
 
-extern char ignore_irq13;
-
 extern void identify_cpu(struct cpuinfo_x86 *);
 
 /*
@@ -68,9 +66,6 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
 
-#define INIT_MMAP \
-{ &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, NULL, NULL }
-
 #define start_thread(regs,new_rip,new_rsp) do { \
 	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));	 \
 	load_gs_index(0);							\
@@ -144,13 +139,4 @@ static inline void prefetchw(void *x)
 			  "r" (x));
 } 
 
-
-#define stack_current() \
-({								\
-	struct thread_info *ti;					\
-	asm("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
-	ti->task;					\
-})
-
-
 #endif /* __ASM_X86_64_PROCESSOR_H */
-- 
cgit v1.2.3


From fc87e9061a5635fe1b367dad95498846dd5eda31 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:38 +0100
Subject: x86: unify mm_segment_t definition

This patch moves the mm_segment_t structure definition to processor.h
This makes mmsegment.h file useless, and it is deleted.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/mmsegment.h      | 8 --------
 include/asm-x86/processor.h      | 5 +++++
 include/asm-x86/processor_32.h   | 5 -----
 include/asm-x86/processor_64.h   | 1 -
 include/asm-x86/thread_info_64.h | 2 +-
 5 files changed, 6 insertions(+), 15 deletions(-)
 delete mode 100644 include/asm-x86/mmsegment.h

(limited to 'include')

diff --git a/include/asm-x86/mmsegment.h b/include/asm-x86/mmsegment.h
deleted file mode 100644
index d3f80c996330..000000000000
--- a/include/asm-x86/mmsegment.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef _ASM_MMSEGMENT_H
-#define _ASM_MMSEGMENT_H 1
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-#endif
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 2b896b0aa3fa..87466b6d3b92 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -424,6 +424,11 @@ struct extended_sigtable {
 	struct extended_signature sigs[0];
 };
 
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
+
 /*
  * create a kernel thread without removing it from tasklists
  */
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index f0f7ca48444e..fa921aa1a93f 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -103,11 +103,6 @@ union i387_union {
 	struct i387_soft_struct soft;
 };
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-
 #define INIT_THREAD  {							\
 	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
 	.vm86_info = NULL,						\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index bfc4ce6e2f00..1a258749b7ca 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -14,7 +14,6 @@
 #include <asm/msr.h>
 #include <asm/current.h>
 #include <asm/system.h>
-#include <asm/mmsegment.h>
 #include <linux/personality.h>
 #include <asm/desc_defs.h>
 
diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index c2911a99cc32..9b531ea015a8 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -21,7 +21,7 @@
 #ifndef __ASSEMBLY__
 struct task_struct;
 struct exec_domain;
-#include <asm/mmsegment.h>
+#include <asm/processor.h>
 
 struct thread_info {
 	struct task_struct	*task;		/* main task structure */
-- 
cgit v1.2.3


From 1a53905adddf6cc6d795bd7e988c60a19773f72e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:39 +0100
Subject: x86: move definitions to processor.h

This patch moves definitions that are present in only one of the files
(between processor_32.h and processor_64.h), to processor.h. They're mostly
structures and function definitions.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/common.c   |  2 +-
 arch/x86/kernel/setup_64.c     |  2 +-
 include/asm-x86/processor.h    | 41 +++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 35 -----------------------------------
 include/asm-x86/processor_64.h |  6 ------
 5 files changed, 43 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index ecd13c0e8542..e48832a6c2a9 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -399,7 +399,7 @@ __setup("serialnumber", x86_serial_nr_setup);
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 {
 	int i;
 
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 90b8bb4748b9..02409100f456 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -757,7 +757,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		disable_apic_timer = 1;
 }
 
-static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_SMP
 	u32 eax, ebx, ecx, edx;
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 87466b6d3b92..c6b749a018a7 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -96,7 +96,12 @@ struct cpuinfo_x86 {
 #define X86_VENDOR_NUM 9
 #define X86_VENDOR_UNKNOWN 0xff
 
+/*
+ * capabilities of CPUs
+ */
 extern struct cpuinfo_x86 boot_cpu_data;
+extern struct cpuinfo_x86 new_cpu_data;
+extern struct tss_struct doublefault_tss;
 
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
@@ -107,11 +112,22 @@ DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 #define current_cpu_data	boot_cpu_data
 #endif
 
+void cpu_detect(struct cpuinfo_x86 *c);
+
+extern void identify_cpu(struct cpuinfo_x86 *);
+extern void identify_boot_cpu(void);
+extern void identify_secondary_cpu(struct cpuinfo_x86 *);
 extern void print_cpu_info(struct cpuinfo_x86 *);
 extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
 
+#if defined(CONFIG_X86_HT) || defined(CONFIG_X86_64)
+extern void detect_ht(struct cpuinfo_x86 *c);
+#else
+static inline void detect_ht(struct cpuinfo_x86 *c) {}
+#endif
+
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 					 unsigned int *ecx, unsigned int *edx)
 {
@@ -205,6 +221,11 @@ struct tss_struct {
 
 DECLARE_PER_CPU(struct tss_struct, init_tss);
 
+/* Save the original ist values for checking stack pointers during debugging */
+struct orig_ist {
+	unsigned long ist[7];
+};
+
 #ifdef CONFIG_X86_32
 # include "processor_32.h"
 #else
@@ -547,8 +568,28 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 
 extern unsigned long boot_option_idle_override;
 
+extern void enable_sep_cpu(void);
+extern int sysenter_setup(void);
+
+/* Defined in head.S */
+extern struct desc_ptr early_gdt_descr;
+
+extern void cpu_set_gdt(int);
+extern void switch_to_new_gdt(void);
+extern void cpu_init(void);
+extern void init_gdt(int cpu);
+
+/* from system description table in BIOS.  Mostly for MCA use, but
+ * others may find it useful. */
+extern unsigned int machine_id;
+extern unsigned int machine_submodel_id;
+extern unsigned int BIOS_revision;
+extern unsigned int mca_pentium_flag;
+
 /* Boot loader type from the setup header */
 extern int bootloader_type;
+
+extern char ignore_fpu_irq;
 #define cache_line_size() (boot_cpu_data.x86_cache_alignment)
 
 #define HAVE_ARCH_PICK_MMAP_LAYOUT 1
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index fa921aa1a93f..84a4c5e47d57 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -18,36 +18,12 @@
 #include <linux/init.h>
 #include <asm/desc_defs.h>
 
-/*
- * capabilities of CPUs
- */
-extern struct cpuinfo_x86 new_cpu_data;
-extern struct tss_struct doublefault_tss;
 
 /*
  * the following now lives in the per cpu area:
  * extern	int cpu_llc_id[NR_CPUS];
  */
 DECLARE_PER_CPU(u8, cpu_llc_id);
-extern char ignore_fpu_irq;
-
-void __init cpu_detect(struct cpuinfo_x86 *c);
-
-extern void identify_boot_cpu(void);
-extern void identify_secondary_cpu(struct cpuinfo_x86 *);
-
-#ifdef CONFIG_X86_HT
-extern void detect_ht(struct cpuinfo_x86 *c);
-#else
-static inline void detect_ht(struct cpuinfo_x86 *c) {}
-#endif
-
-/* from system description table in BIOS.  Mostly for MCA use, but
-others may find it useful. */
-extern unsigned int machine_id;
-extern unsigned int machine_submodel_id;
-extern unsigned int BIOS_revision;
-extern unsigned int mca_pentium_flag;
 
 /*
  * User space process size: 3GB (default).
@@ -277,15 +253,4 @@ static inline void prefetchw(const void *x)
 			  "r" (x));
 }
 
-extern void enable_sep_cpu(void);
-extern int sysenter_setup(void);
-
-/* Defined in head.S */
-extern struct desc_ptr early_gdt_descr;
-
-extern void cpu_set_gdt(int);
-extern void switch_to_new_gdt(void);
-extern void cpu_init(void);
-extern void init_gdt(int cpu);
-
 #endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 1a258749b7ca..45e382989b33 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -17,8 +17,6 @@
 #include <linux/personality.h>
 #include <asm/desc_defs.h>
 
-extern void identify_cpu(struct cpuinfo_x86 *);
-
 /*
  * User space process size. 47bits minus one guard page.
  */
@@ -51,10 +49,6 @@ union i387_union {
 	struct i387_fxsave_struct	fxsave;
 };
 
-/* Save the original ist values for checking stack pointers during debugging */
-struct orig_ist {
-	unsigned long ist[7];
-};
 DECLARE_PER_CPU(struct orig_ist, orig_ist);
 
 #define INIT_THREAD  { \
-- 
cgit v1.2.3


From ae2e15eb3b6c2a011bee615470bf52d2beb99a4b Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:40 +0100
Subject: x86: unify prefetch operations

This patch moves the prefetch[w]? functions to processor.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 30 ++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 25 -------------------------
 include/asm-x86/processor_64.h |  8 --------
 3 files changed, 30 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index c6b749a018a7..bfac9739f57e 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -596,6 +596,36 @@ extern char ignore_fpu_irq;
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
 
+#ifdef CONFIG_X86_32
+#define BASE_PREFETCH	ASM_NOP4
+#define ARCH_HAS_PREFETCH
+#else
+#define BASE_PREFETCH	"prefetcht0 (%1)"
+#endif
+
+/* Prefetch instructions for Pentium III and AMD Athlon */
+/* It's not worth to care about 3dnow! prefetches for the K6
+   because they are microcoded there and very slow.
+   However we don't do prefetches for pre XP Athlons currently
+   That should be fixed. */
+static inline void prefetch(const void *x)
+{
+	alternative_input(BASE_PREFETCH,
+			  "prefetchnta (%1)",
+			  X86_FEATURE_XMM,
+			  "r" (x));
+}
+
+/* 3dnow! prefetch to get an exclusive cache line. Useful for
+   spinlocks to avoid one state transition in the cache coherency protocol. */
+static inline void prefetchw(const void *x)
+{
+	alternative_input(BASE_PREFETCH,
+			  "prefetchw (%1)",
+			  X86_FEATURE_3DNOW,
+			  "r" (x));
+}
+
 #define spin_lock_prefetch(x)	prefetchw(x)
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 84a4c5e47d57..61a9cae2364b 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -228,29 +228,4 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
 #define ASM_NOP_MAX 8
 
-/* Prefetch instructions for Pentium III and AMD Athlon */
-/* It's not worth to care about 3dnow! prefetches for the K6
-   because they are microcoded there and very slow.
-   However we don't do prefetches for pre XP Athlons currently
-   That should be fixed. */
-static inline void prefetch(const void *x)
-{
-	alternative_input(ASM_NOP4,
-			  "prefetchnta (%1)",
-			  X86_FEATURE_XMM,
-			  "r" (x));
-}
-
-#define ARCH_HAS_PREFETCH
-
-/* 3dnow! prefetch to get an exclusive cache line. Useful for 
-   spinlocks to avoid one state transition in the cache coherency protocol. */
-static inline void prefetchw(const void *x)
-{
-	alternative_input(ASM_NOP4,
-			  "prefetchw (%1)",
-			  X86_FEATURE_3DNOW,
-			  "r" (x));
-}
-
 #endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 45e382989b33..08b965124b97 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -124,12 +124,4 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 
 #define ASM_NOP_MAX 8
 
-static inline void prefetchw(void *x) 
-{ 
-	alternative_input("prefetcht0 (%1)",
-			  "prefetchw (%1)",
-			  X86_FEATURE_3DNOW,
-			  "r" (x));
-} 
-
 #endif /* __ASM_X86_64_PROCESSOR_H */
-- 
cgit v1.2.3


From ea5e3593a493298c69b8c4a958628feace930478 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:40 +0100
Subject: x86: unify asm nops

There's only one difference between the NOPs used in asm code for i386 and x86_64:
i386 has a lot more variants. The code is moved to processor.h, and adjusted
accordingly.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 85 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 85 ------------------------------------------
 include/asm-x86/processor_64.h | 44 ----------------------
 3 files changed, 85 insertions(+), 129 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index bfac9739f57e..a8f1702ec3ba 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -592,6 +592,91 @@ extern int bootloader_type;
 extern char ignore_fpu_irq;
 #define cache_line_size() (boot_cpu_data.x86_cache_alignment)
 
+/* generic versions from gas */
+#define GENERIC_NOP1	".byte 0x90\n"
+#define GENERIC_NOP2    	".byte 0x89,0xf6\n"
+#define GENERIC_NOP3        ".byte 0x8d,0x76,0x00\n"
+#define GENERIC_NOP4        ".byte 0x8d,0x74,0x26,0x00\n"
+#define GENERIC_NOP5        GENERIC_NOP1 GENERIC_NOP4
+#define GENERIC_NOP6	".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP7	".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP8	GENERIC_NOP1 GENERIC_NOP7
+
+/* Opteron nops */
+#define K8_NOP1 GENERIC_NOP1
+#define K8_NOP2	".byte 0x66,0x90\n"
+#define K8_NOP3	".byte 0x66,0x66,0x90\n"
+#define K8_NOP4	".byte 0x66,0x66,0x66,0x90\n"
+#define K8_NOP5	K8_NOP3 K8_NOP2
+#define K8_NOP6	K8_NOP3 K8_NOP3
+#define K8_NOP7	K8_NOP4 K8_NOP3
+#define K8_NOP8	K8_NOP4 K8_NOP4
+
+/* K7 nops */
+/* uses eax dependencies (arbitary choice) */
+#define K7_NOP1  GENERIC_NOP1
+#define K7_NOP2	".byte 0x8b,0xc0\n"
+#define K7_NOP3	".byte 0x8d,0x04,0x20\n"
+#define K7_NOP4	".byte 0x8d,0x44,0x20,0x00\n"
+#define K7_NOP5	K7_NOP4 ASM_NOP1
+#define K7_NOP6	".byte 0x8d,0x80,0,0,0,0\n"
+#define K7_NOP7        ".byte 0x8D,0x04,0x05,0,0,0,0\n"
+#define K7_NOP8        K7_NOP7 ASM_NOP1
+
+/* P6 nops */
+/* uses eax dependencies (Intel-recommended choice) */
+#define P6_NOP1	GENERIC_NOP1
+#define P6_NOP2	".byte 0x66,0x90\n"
+#define P6_NOP3	".byte 0x0f,0x1f,0x00\n"
+#define P6_NOP4	".byte 0x0f,0x1f,0x40,0\n"
+#define P6_NOP5	".byte 0x0f,0x1f,0x44,0x00,0\n"
+#define P6_NOP6	".byte 0x66,0x0f,0x1f,0x44,0x00,0\n"
+#define P6_NOP7	".byte 0x0f,0x1f,0x80,0,0,0,0\n"
+#define P6_NOP8	".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n"
+
+#ifdef CONFIG_MK7
+#define ASM_NOP1 K7_NOP1
+#define ASM_NOP2 K7_NOP2
+#define ASM_NOP3 K7_NOP3
+#define ASM_NOP4 K7_NOP4
+#define ASM_NOP5 K7_NOP5
+#define ASM_NOP6 K7_NOP6
+#define ASM_NOP7 K7_NOP7
+#define ASM_NOP8 K7_NOP8
+#elif defined(CONFIG_M686) || defined(CONFIG_MPENTIUMII) || \
+      defined(CONFIG_MPENTIUMIII) || defined(CONFIG_MPENTIUMM) || \
+      defined(CONFIG_MCORE2) || defined(CONFIG_PENTIUM4) || \
+      defined(CONFIG_MPSC)
+#define ASM_NOP1 P6_NOP1
+#define ASM_NOP2 P6_NOP2
+#define ASM_NOP3 P6_NOP3
+#define ASM_NOP4 P6_NOP4
+#define ASM_NOP5 P6_NOP5
+#define ASM_NOP6 P6_NOP6
+#define ASM_NOP7 P6_NOP7
+#define ASM_NOP8 P6_NOP8
+#elif defined(CONFIG_MK8) || defined(CONFIG_X86_64)
+#define ASM_NOP1 K8_NOP1
+#define ASM_NOP2 K8_NOP2
+#define ASM_NOP3 K8_NOP3
+#define ASM_NOP4 K8_NOP4
+#define ASM_NOP5 K8_NOP5
+#define ASM_NOP6 K8_NOP6
+#define ASM_NOP7 K8_NOP7
+#define ASM_NOP8 K8_NOP8
+#else
+#define ASM_NOP1 GENERIC_NOP1
+#define ASM_NOP2 GENERIC_NOP2
+#define ASM_NOP3 GENERIC_NOP3
+#define ASM_NOP4 GENERIC_NOP4
+#define ASM_NOP5 GENERIC_NOP5
+#define ASM_NOP6 GENERIC_NOP6
+#define ASM_NOP7 GENERIC_NOP7
+#define ASM_NOP8 GENERIC_NOP8
+#endif
+
+#define ASM_NOP_MAX 8
+
 #define HAVE_ARCH_PICK_MMAP_LAYOUT 1
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 61a9cae2364b..82fafabbfe7a 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -18,7 +18,6 @@
 #include <linux/init.h>
 #include <asm/desc_defs.h>
 
-
 /*
  * the following now lives in the per cpu area:
  * extern	int cpu_llc_id[NR_CPUS];
@@ -144,88 +143,4 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
 #define KSTK_ESP(task) (task_pt_regs(task)->sp)
 
-/* generic versions from gas */
-#define GENERIC_NOP1	".byte 0x90\n"
-#define GENERIC_NOP2    	".byte 0x89,0xf6\n"
-#define GENERIC_NOP3        ".byte 0x8d,0x76,0x00\n"
-#define GENERIC_NOP4        ".byte 0x8d,0x74,0x26,0x00\n"
-#define GENERIC_NOP5        GENERIC_NOP1 GENERIC_NOP4
-#define GENERIC_NOP6	".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
-#define GENERIC_NOP7	".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
-#define GENERIC_NOP8	GENERIC_NOP1 GENERIC_NOP7
-
-/* Opteron nops */
-#define K8_NOP1 GENERIC_NOP1
-#define K8_NOP2	".byte 0x66,0x90\n" 
-#define K8_NOP3	".byte 0x66,0x66,0x90\n" 
-#define K8_NOP4	".byte 0x66,0x66,0x66,0x90\n" 
-#define K8_NOP5	K8_NOP3 K8_NOP2 
-#define K8_NOP6	K8_NOP3 K8_NOP3
-#define K8_NOP7	K8_NOP4 K8_NOP3
-#define K8_NOP8	K8_NOP4 K8_NOP4
-
-/* K7 nops */
-/* uses eax dependencies (arbitary choice) */
-#define K7_NOP1  GENERIC_NOP1
-#define K7_NOP2	".byte 0x8b,0xc0\n" 
-#define K7_NOP3	".byte 0x8d,0x04,0x20\n"
-#define K7_NOP4	".byte 0x8d,0x44,0x20,0x00\n"
-#define K7_NOP5	K7_NOP4 ASM_NOP1
-#define K7_NOP6	".byte 0x8d,0x80,0,0,0,0\n"
-#define K7_NOP7        ".byte 0x8D,0x04,0x05,0,0,0,0\n"
-#define K7_NOP8        K7_NOP7 ASM_NOP1
-
-/* P6 nops */
-/* uses eax dependencies (Intel-recommended choice) */
-#define P6_NOP1	GENERIC_NOP1
-#define P6_NOP2	".byte 0x66,0x90\n"
-#define P6_NOP3	".byte 0x0f,0x1f,0x00\n"
-#define P6_NOP4	".byte 0x0f,0x1f,0x40,0\n"
-#define P6_NOP5	".byte 0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP6	".byte 0x66,0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP7	".byte 0x0f,0x1f,0x80,0,0,0,0\n"
-#define P6_NOP8	".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n"
-
-#ifdef CONFIG_MK8
-#define ASM_NOP1 K8_NOP1
-#define ASM_NOP2 K8_NOP2
-#define ASM_NOP3 K8_NOP3
-#define ASM_NOP4 K8_NOP4
-#define ASM_NOP5 K8_NOP5
-#define ASM_NOP6 K8_NOP6
-#define ASM_NOP7 K8_NOP7
-#define ASM_NOP8 K8_NOP8
-#elif defined(CONFIG_MK7)
-#define ASM_NOP1 K7_NOP1
-#define ASM_NOP2 K7_NOP2
-#define ASM_NOP3 K7_NOP3
-#define ASM_NOP4 K7_NOP4
-#define ASM_NOP5 K7_NOP5
-#define ASM_NOP6 K7_NOP6
-#define ASM_NOP7 K7_NOP7
-#define ASM_NOP8 K7_NOP8
-#elif defined(CONFIG_M686) || defined(CONFIG_MPENTIUMII) || \
-      defined(CONFIG_MPENTIUMIII) || defined(CONFIG_MPENTIUMM) || \
-      defined(CONFIG_MCORE2) || defined(CONFIG_PENTIUM4)
-#define ASM_NOP1 P6_NOP1
-#define ASM_NOP2 P6_NOP2
-#define ASM_NOP3 P6_NOP3
-#define ASM_NOP4 P6_NOP4
-#define ASM_NOP5 P6_NOP5
-#define ASM_NOP6 P6_NOP6
-#define ASM_NOP7 P6_NOP7
-#define ASM_NOP8 P6_NOP8
-#else
-#define ASM_NOP1 GENERIC_NOP1
-#define ASM_NOP2 GENERIC_NOP2
-#define ASM_NOP3 GENERIC_NOP3
-#define ASM_NOP4 GENERIC_NOP4
-#define ASM_NOP5 GENERIC_NOP5
-#define ASM_NOP6 GENERIC_NOP6
-#define ASM_NOP7 GENERIC_NOP7
-#define ASM_NOP8 GENERIC_NOP8
-#endif
-
-#define ASM_NOP_MAX 8
-
 #endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 08b965124b97..86fc2dcc8687 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -80,48 +80,4 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
 #define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
 
-
-#if defined(CONFIG_MPSC) || defined(CONFIG_MCORE2)
-#define ASM_NOP1 P6_NOP1
-#define ASM_NOP2 P6_NOP2
-#define ASM_NOP3 P6_NOP3
-#define ASM_NOP4 P6_NOP4
-#define ASM_NOP5 P6_NOP5
-#define ASM_NOP6 P6_NOP6
-#define ASM_NOP7 P6_NOP7
-#define ASM_NOP8 P6_NOP8
-#else
-#define ASM_NOP1 K8_NOP1
-#define ASM_NOP2 K8_NOP2
-#define ASM_NOP3 K8_NOP3
-#define ASM_NOP4 K8_NOP4
-#define ASM_NOP5 K8_NOP5
-#define ASM_NOP6 K8_NOP6
-#define ASM_NOP7 K8_NOP7
-#define ASM_NOP8 K8_NOP8
-#endif
-
-/* Opteron nops */
-#define K8_NOP1 ".byte 0x90\n"
-#define K8_NOP2	".byte 0x66,0x90\n" 
-#define K8_NOP3	".byte 0x66,0x66,0x90\n" 
-#define K8_NOP4	".byte 0x66,0x66,0x66,0x90\n" 
-#define K8_NOP5	K8_NOP3 K8_NOP2 
-#define K8_NOP6	K8_NOP3 K8_NOP3
-#define K8_NOP7	K8_NOP4 K8_NOP3
-#define K8_NOP8	K8_NOP4 K8_NOP4
-
-/* P6 nops */
-/* uses eax dependencies (Intel-recommended choice) */
-#define P6_NOP1	".byte 0x90\n"
-#define P6_NOP2	".byte 0x66,0x90\n"
-#define P6_NOP3	".byte 0x0f,0x1f,0x00\n"
-#define P6_NOP4	".byte 0x0f,0x1f,0x40,0\n"
-#define P6_NOP5	".byte 0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP6	".byte 0x66,0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP7	".byte 0x0f,0x1f,0x80,0,0,0,0\n"
-#define P6_NOP8	".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n"
-
-#define ASM_NOP_MAX 8
-
 #endif /* __ASM_X86_64_PROCESSOR_H */
-- 
cgit v1.2.3


From 46265df040533f57c191bb2b019d6b25c3bf1f34 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:41 +0100
Subject: x86: move i387 definitions to processor.h

This patch moves i387 definitions from processor_32.h and processor_64.h
to processor.h. They are different. Very different. And there's appearently
nothing we can do about it, so they're enclosed inside ifdefs.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h    | 66 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor_32.h | 49 -------------------------------
 include/asm-x86/processor_64.h | 17 -----------
 3 files changed, 66 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index a8f1702ec3ba..80c50020ef60 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -227,8 +227,74 @@ struct orig_ist {
 };
 
 #ifdef CONFIG_X86_32
+struct i387_fsave_struct {
+	long	cwd;
+	long	swd;
+	long	twd;
+	long	fip;
+	long	fcs;
+	long	foo;
+	long	fos;
+	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
+	long	status;		/* software status information */
+};
+
+struct i387_fxsave_struct {
+	unsigned short	cwd;
+	unsigned short	swd;
+	unsigned short	twd;
+	unsigned short	fop;
+	long	fip;
+	long	fcs;
+	long	foo;
+	long	fos;
+	long	mxcsr;
+	long	mxcsr_mask;
+	long	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
+	long	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
+	long	padding[56];
+} __attribute__((aligned(16)));
+
+struct i387_soft_struct {
+	long	cwd;
+	long	swd;
+	long	twd;
+	long	fip;
+	long	fcs;
+	long	foo;
+	long	fos;
+	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
+	unsigned char	ftop, changed, lookahead, no_update, rm, alimit;
+	struct info	*info;
+	unsigned long	entry_eip;
+};
+
+union i387_union {
+	struct i387_fsave_struct	fsave;
+	struct i387_fxsave_struct	fxsave;
+	struct i387_soft_struct soft;
+};
+
 # include "processor_32.h"
 #else
+struct i387_fxsave_struct {
+	u16	cwd;
+	u16	swd;
+	u16	twd;
+	u16	fop;
+	u64	rip;
+	u64	rdp;
+	u32	mxcsr;
+	u32	mxcsr_mask;
+	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
+	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
+	u32	padding[24];
+} __attribute__((aligned(16)));
+
+union i387_union {
+	struct i387_fxsave_struct	fxsave;
+};
+
 # include "processor_64.h"
 #endif
 
diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
index 82fafabbfe7a..0d6a430b2bc3 100644
--- a/include/asm-x86/processor_32.h
+++ b/include/asm-x86/processor_32.h
@@ -29,55 +29,6 @@ DECLARE_PER_CPU(u8, cpu_llc_id);
  */
 #define TASK_SIZE	(PAGE_OFFSET)
 
-
-struct i387_fsave_struct {
-	long	cwd;
-	long	swd;
-	long	twd;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	long	status;		/* software status information */
-};
-
-struct i387_fxsave_struct {
-	unsigned short	cwd;
-	unsigned short	swd;
-	unsigned short	twd;
-	unsigned short	fop;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	mxcsr;
-	long	mxcsr_mask;
-	long	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
-	long	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
-	long	padding[56];
-} __attribute__ ((aligned (16)));
-
-struct i387_soft_struct {
-	long	cwd;
-	long	swd;
-	long	twd;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	unsigned char	ftop, changed, lookahead, no_update, rm, alimit;
-	struct info	*info;
-	unsigned long	entry_eip;
-};
-
-union i387_union {
-	struct i387_fsave_struct	fsave;
-	struct i387_fxsave_struct	fxsave;
-	struct i387_soft_struct soft;
-};
-
 #define INIT_THREAD  {							\
 	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
 	.vm86_info = NULL,						\
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
index 86fc2dcc8687..04ce823ea7e6 100644
--- a/include/asm-x86/processor_64.h
+++ b/include/asm-x86/processor_64.h
@@ -31,23 +31,6 @@
 #define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
 
 
-struct i387_fxsave_struct {
-	u16	cwd;
-	u16	swd;
-	u16	twd;
-	u16	fop;
-	u64	rip;
-	u64	rdp; 
-	u32	mxcsr;
-	u32	mxcsr_mask;
-	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
-	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
-	u32	padding[24];
-} __attribute__ ((aligned (16)));
-
-union i387_union {
-	struct i387_fxsave_struct	fxsave;
-};
 
 DECLARE_PER_CPU(struct orig_ist, orig_ist);
 
-- 
cgit v1.2.3


From 85f2adf16955460c98131360f3d76aeb51aba073 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:31:42 +0100
Subject: x86: use helper in fault_64.c

Use the fixup_exception() helper in fault_64.c

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/fault_64.c       | 5 +----
 include/asm-x86/uaccess_64.h | 2 ++
 2 files changed, 3 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 121c7bda6297..3a94941578fa 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -298,7 +298,6 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 	struct mm_struct *mm;
 	struct vm_area_struct * vma;
 	unsigned long address;
-	const struct exception_table_entry *fixup;
 	int write, fault;
 	unsigned long flags;
 	siginfo_t info;
@@ -508,9 +507,7 @@ bad_area_nosemaphore:
 no_context:
 	
 	/* Are we prepared to handle this kernel fault?  */
-	fixup = search_exception_tables(regs->ip);
-	if (fixup) {
-		regs->ip = fixup->fixup;
+	if (fixup_exception(regs)) {
 		return;
 	}
 
diff --git a/include/asm-x86/uaccess_64.h b/include/asm-x86/uaccess_64.h
index f4ce8768ad44..31d794702719 100644
--- a/include/asm-x86/uaccess_64.h
+++ b/include/asm-x86/uaccess_64.h
@@ -65,6 +65,8 @@ struct exception_table_entry
 	unsigned long insn, fixup;
 };
 
+extern int fixup_exception(struct pt_regs *regs);
+
 #define ARCH_HAS_SEARCH_EXTABLE
 
 /*
-- 
cgit v1.2.3


From 0ddc9cc8fdfe3df7a90557e66069e3da2c584725 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Tue, 18 Dec 2007 03:58:10 -0800
Subject: x86: unify module_{32|64}.h

This adds one case to the MODULE_PROC_FAMILY block testing
for X86_64.  There are no new things defined on X86_64 than
there were before.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/module.h    | 81 +++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/module_32.h | 75 -----------------------------------------
 include/asm-x86/module_64.h | 10 ------
 3 files changed, 79 insertions(+), 87 deletions(-)
 delete mode 100644 include/asm-x86/module_32.h
 delete mode 100644 include/asm-x86/module_64.h

(limited to 'include')

diff --git a/include/asm-x86/module.h b/include/asm-x86/module.h
index 2b2f18d8a531..bfedb247871c 100644
--- a/include/asm-x86/module.h
+++ b/include/asm-x86/module.h
@@ -1,5 +1,82 @@
+#ifndef _ASM_MODULE_H
+#define _ASM_MODULE_H
+
+/* x86_32/64 are simple */
+struct mod_arch_specific {};
+
 #ifdef CONFIG_X86_32
-# include "module_32.h"
+# define Elf_Shdr Elf32_Shdr
+# define Elf_Sym Elf32_Sym
+# define Elf_Ehdr Elf32_Ehdr
 #else
-# include "module_64.h"
+# define Elf_Shdr Elf64_Shdr
+# define Elf_Sym Elf64_Sym
+# define Elf_Ehdr Elf64_Ehdr
 #endif
+
+#ifdef CONFIG_X86_64
+/* X86_64 does not define MODULE_PROC_FAMILY */
+#elif defined CONFIG_M386
+#define MODULE_PROC_FAMILY "386 "
+#elif defined CONFIG_M486
+#define MODULE_PROC_FAMILY "486 "
+#elif defined CONFIG_M586
+#define MODULE_PROC_FAMILY "586 "
+#elif defined CONFIG_M586TSC
+#define MODULE_PROC_FAMILY "586TSC "
+#elif defined CONFIG_M586MMX
+#define MODULE_PROC_FAMILY "586MMX "
+#elif defined CONFIG_MCORE2
+#define MODULE_PROC_FAMILY "CORE2 "
+#elif defined CONFIG_M686
+#define MODULE_PROC_FAMILY "686 "
+#elif defined CONFIG_MPENTIUMII
+#define MODULE_PROC_FAMILY "PENTIUMII "
+#elif defined CONFIG_MPENTIUMIII
+#define MODULE_PROC_FAMILY "PENTIUMIII "
+#elif defined CONFIG_MPENTIUMM
+#define MODULE_PROC_FAMILY "PENTIUMM "
+#elif defined CONFIG_MPENTIUM4
+#define MODULE_PROC_FAMILY "PENTIUM4 "
+#elif defined CONFIG_MK6
+#define MODULE_PROC_FAMILY "K6 "
+#elif defined CONFIG_MK7
+#define MODULE_PROC_FAMILY "K7 "
+#elif defined CONFIG_MK8
+#define MODULE_PROC_FAMILY "K8 "
+#elif defined CONFIG_X86_ELAN
+#define MODULE_PROC_FAMILY "ELAN "
+#elif defined CONFIG_MCRUSOE
+#define MODULE_PROC_FAMILY "CRUSOE "
+#elif defined CONFIG_MEFFICEON
+#define MODULE_PROC_FAMILY "EFFICEON "
+#elif defined CONFIG_MWINCHIPC6
+#define MODULE_PROC_FAMILY "WINCHIPC6 "
+#elif defined CONFIG_MWINCHIP2
+#define MODULE_PROC_FAMILY "WINCHIP2 "
+#elif defined CONFIG_MWINCHIP3D
+#define MODULE_PROC_FAMILY "WINCHIP3D "
+#elif defined CONFIG_MCYRIXIII
+#define MODULE_PROC_FAMILY "CYRIXIII "
+#elif defined CONFIG_MVIAC3_2
+#define MODULE_PROC_FAMILY "VIAC3-2 "
+#elif defined CONFIG_MVIAC7
+#define MODULE_PROC_FAMILY "VIAC7 "
+#elif defined CONFIG_MGEODEGX1
+#define MODULE_PROC_FAMILY "GEODEGX1 "
+#elif defined CONFIG_MGEODE_LX
+#define MODULE_PROC_FAMILY "GEODE "
+#else
+#error unknown processor family
+#endif
+
+#ifdef CONFIG_X86_32
+# ifdef CONFIG_4KSTACKS
+#  define MODULE_STACKSIZE "4KSTACKS "
+# else
+#  define MODULE_STACKSIZE ""
+# endif
+# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE
+#endif
+
+#endif /* _ASM_MODULE_H */
diff --git a/include/asm-x86/module_32.h b/include/asm-x86/module_32.h
deleted file mode 100644
index 7e5fda6c3976..000000000000
--- a/include/asm-x86/module_32.h
+++ /dev/null
@@ -1,75 +0,0 @@
-#ifndef _ASM_I386_MODULE_H
-#define _ASM_I386_MODULE_H
-
-/* x86 is simple */
-struct mod_arch_specific
-{
-};
-
-#define Elf_Shdr Elf32_Shdr
-#define Elf_Sym Elf32_Sym
-#define Elf_Ehdr Elf32_Ehdr
-
-#ifdef CONFIG_M386
-#define MODULE_PROC_FAMILY "386 "
-#elif defined CONFIG_M486
-#define MODULE_PROC_FAMILY "486 "
-#elif defined CONFIG_M586
-#define MODULE_PROC_FAMILY "586 "
-#elif defined CONFIG_M586TSC
-#define MODULE_PROC_FAMILY "586TSC "
-#elif defined CONFIG_M586MMX
-#define MODULE_PROC_FAMILY "586MMX "
-#elif defined CONFIG_MCORE2
-#define MODULE_PROC_FAMILY "CORE2 "
-#elif defined CONFIG_M686
-#define MODULE_PROC_FAMILY "686 "
-#elif defined CONFIG_MPENTIUMII
-#define MODULE_PROC_FAMILY "PENTIUMII "
-#elif defined CONFIG_MPENTIUMIII
-#define MODULE_PROC_FAMILY "PENTIUMIII "
-#elif defined CONFIG_MPENTIUMM
-#define MODULE_PROC_FAMILY "PENTIUMM "
-#elif defined CONFIG_MPENTIUM4
-#define MODULE_PROC_FAMILY "PENTIUM4 "
-#elif defined CONFIG_MK6
-#define MODULE_PROC_FAMILY "K6 "
-#elif defined CONFIG_MK7
-#define MODULE_PROC_FAMILY "K7 "
-#elif defined CONFIG_MK8
-#define MODULE_PROC_FAMILY "K8 "
-#elif defined CONFIG_X86_ELAN
-#define MODULE_PROC_FAMILY "ELAN "
-#elif defined CONFIG_MCRUSOE
-#define MODULE_PROC_FAMILY "CRUSOE "
-#elif defined CONFIG_MEFFICEON
-#define MODULE_PROC_FAMILY "EFFICEON "
-#elif defined CONFIG_MWINCHIPC6
-#define MODULE_PROC_FAMILY "WINCHIPC6 "
-#elif defined CONFIG_MWINCHIP2
-#define MODULE_PROC_FAMILY "WINCHIP2 "
-#elif defined CONFIG_MWINCHIP3D
-#define MODULE_PROC_FAMILY "WINCHIP3D "
-#elif defined CONFIG_MCYRIXIII
-#define MODULE_PROC_FAMILY "CYRIXIII "
-#elif defined CONFIG_MVIAC3_2
-#define MODULE_PROC_FAMILY "VIAC3-2 "
-#elif defined CONFIG_MVIAC7
-#define MODULE_PROC_FAMILY "VIAC7 "
-#elif defined CONFIG_MGEODEGX1
-#define MODULE_PROC_FAMILY "GEODEGX1 "
-#elif defined CONFIG_MGEODE_LX
-#define MODULE_PROC_FAMILY "GEODE "
-#else
-#error unknown processor family
-#endif
-
-#ifdef CONFIG_4KSTACKS
-#define MODULE_STACKSIZE "4KSTACKS "
-#else
-#define MODULE_STACKSIZE ""
-#endif
-
-#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE
-
-#endif /* _ASM_I386_MODULE_H */
diff --git a/include/asm-x86/module_64.h b/include/asm-x86/module_64.h
deleted file mode 100644
index 67f8f69fa7b1..000000000000
--- a/include/asm-x86/module_64.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef _ASM_X8664_MODULE_H
-#define _ASM_X8664_MODULE_H
-
-struct mod_arch_specific {}; 
-
-#define Elf_Shdr Elf64_Shdr
-#define Elf_Sym Elf64_Sym
-#define Elf_Ehdr Elf64_Ehdr
-
-#endif 
-- 
cgit v1.2.3


From bdf88217b70dbb18c4ee27a6c497286e040a6705 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:44 +0100
Subject: x86: user_regset header

The new header <linux/regset.h> defines the types struct user_regset and
struct user_regset_view, with some associated declarations.  This new set
of interfaces will become the standard way for arch code to expose
user-mode machine-specific state.  A single set of entry points into arch
code can do all the low-level work in one place to fill the needs of core
dumps, ptrace, and any other user-mode debugging facilities that might come
along in the future.

For existing arch code to adapt to the user_regset interfaces, each arch
can work from the code it already has to support core files and ptrace.
The formats you want for user_regset are the core file formats.  The only
wrinkle in adapting old ptrace implementation code as user_regset get and
set functions is that these functions can be called on current as well as
on another task_struct that is stopped and switched out as for ptrace.
For some kinds of machine state, you may have to load it directly from CPU
registers or otherwise differently for current than for another thread.
(Your core dump support already handles this in elf_core_copy_regs for
current and elf_core_copy_task_regs for other tasks, so just check there.)
The set function should also be made to work on current in case that
entails some special cases, though this was never required before for
ptrace.  Adding this flexibility covers the arch needs to open the door to
more sophisticated new debugging facilities that don't always need to
context-switch to do every little thing.

The copyin/copyout helper functions (in a later patch) relieve the arch
code of most of the cumbersome details of the flexible get/set interfaces.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/regset.h | 206 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 206 insertions(+)
 create mode 100644 include/linux/regset.h

(limited to 'include')

diff --git a/include/linux/regset.h b/include/linux/regset.h
new file mode 100644
index 000000000000..85d0fb0a014d
--- /dev/null
+++ b/include/linux/regset.h
@@ -0,0 +1,206 @@
+/*
+ * User-mode machine state access
+ *
+ * Copyright (C) 2007 Red Hat, Inc.  All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ *
+ * Red Hat Author: Roland McGrath.
+ */
+
+#ifndef _LINUX_REGSET_H
+#define _LINUX_REGSET_H	1
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+struct task_struct;
+struct user_regset;
+
+
+/**
+ * user_regset_active_fn - type of @active function in &struct user_regset
+ * @target:	thread being examined
+ * @regset:	regset being examined
+ *
+ * Return -%ENODEV if not available on the hardware found.
+ * Return %0 if no interesting state in this thread.
+ * Return >%0 number of @size units of interesting state.
+ * Any get call fetching state beyond that number will
+ * see the default initialization state for this data,
+ * so a caller that knows what the default state is need
+ * not copy it all out.
+ * This call is optional; the pointer is %NULL if there
+ * is no inexpensive check to yield a value < @n.
+ */
+typedef int user_regset_active_fn(struct task_struct *target,
+				  const struct user_regset *regset);
+
+/**
+ * user_regset_get_fn - type of @get function in &struct user_regset
+ * @target:	thread being examined
+ * @regset:	regset being examined
+ * @pos:	offset into the regset data to access, in bytes
+ * @count:	amount of data to copy, in bytes
+ * @kbuf:	if not %NULL, a kernel-space pointer to copy into
+ * @ubuf:	if @kbuf is %NULL, a user-space pointer to copy into
+ *
+ * Fetch register values.  Return %0 on success; -%EIO or -%ENODEV
+ * are usual failure returns.  The @pos and @count values are in
+ * bytes, but must be properly aligned.  If @kbuf is non-null, that
+ * buffer is used and @ubuf is ignored.  If @kbuf is %NULL, then
+ * ubuf gives a userland pointer to access directly, and an -%EFAULT
+ * return value is possible.
+ */
+typedef int user_regset_get_fn(struct task_struct *target,
+			       const struct user_regset *regset,
+			       unsigned int pos, unsigned int count,
+			       void *kbuf, void __user *ubuf);
+
+/**
+ * user_regset_set_fn - type of @set function in &struct user_regset
+ * @target:	thread being examined
+ * @regset:	regset being examined
+ * @pos:	offset into the regset data to access, in bytes
+ * @count:	amount of data to copy, in bytes
+ * @kbuf:	if not %NULL, a kernel-space pointer to copy from
+ * @ubuf:	if @kbuf is %NULL, a user-space pointer to copy from
+ *
+ * Store register values.  Return %0 on success; -%EIO or -%ENODEV
+ * are usual failure returns.  The @pos and @count values are in
+ * bytes, but must be properly aligned.  If @kbuf is non-null, that
+ * buffer is used and @ubuf is ignored.  If @kbuf is %NULL, then
+ * ubuf gives a userland pointer to access directly, and an -%EFAULT
+ * return value is possible.
+ */
+typedef int user_regset_set_fn(struct task_struct *target,
+			       const struct user_regset *regset,
+			       unsigned int pos, unsigned int count,
+			       const void *kbuf, const void __user *ubuf);
+
+/**
+ * user_regset_writeback_fn - type of @writeback function in &struct user_regset
+ * @target:	thread being examined
+ * @regset:	regset being examined
+ * @immediate:	zero if writeback at completion of next context switch is OK
+ *
+ * This call is optional; usually the pointer is %NULL.  When
+ * provided, there is some user memory associated with this regset's
+ * hardware, such as memory backing cached register data on register
+ * window machines; the regset's data controls what user memory is
+ * used (e.g. via the stack pointer value).
+ *
+ * Write register data back to user memory.  If the @immediate flag
+ * is nonzero, it must be written to the user memory so uaccess or
+ * access_process_vm() can see it when this call returns; if zero,
+ * then it must be written back by the time the task completes a
+ * context switch (as synchronized with wait_task_inactive()).
+ * Return %0 on success or if there was nothing to do, -%EFAULT for
+ * a memory problem (bad stack pointer or whatever), or -%EIO for a
+ * hardware problem.
+ */
+typedef int user_regset_writeback_fn(struct task_struct *target,
+				     const struct user_regset *regset,
+				     int immediate);
+
+/**
+ * struct user_regset - accessible thread CPU state
+ * @n:			Number of slots (registers).
+ * @size:		Size in bytes of a slot (register).
+ * @align:		Required alignment, in bytes.
+ * @bias:		Bias from natural indexing.
+ * @core_note_type:	ELF note @n_type value used in core dumps.
+ * @get:		Function to fetch values.
+ * @set:		Function to store values.
+ * @active:		Function to report if regset is active, or %NULL.
+ * @writeback:		Function to write data back to user memory, or %NULL.
+ *
+ * This data structure describes a machine resource we call a register set.
+ * This is part of the state of an individual thread, not necessarily
+ * actual CPU registers per se.  A register set consists of a number of
+ * similar slots, given by @n.  Each slot is @size bytes, and aligned to
+ * @align bytes (which is at least @size).
+ *
+ * These functions must be called only on the current thread or on a
+ * thread that is in %TASK_STOPPED or %TASK_TRACED state, that we are
+ * guaranteed will not be woken up and return to user mode, and that we
+ * have called wait_task_inactive() on.  (The target thread always might
+ * wake up for SIGKILL while these functions are working, in which case
+ * that thread's user_regset state might be scrambled.)
+ *
+ * The @pos argument must be aligned according to @align; the @count
+ * argument must be a multiple of @size.  These functions are not
+ * responsible for checking for invalid arguments.
+ *
+ * When there is a natural value to use as an index, @bias gives the
+ * difference between the natural index and the slot index for the
+ * register set.  For example, x86 GDT segment descriptors form a regset;
+ * the segment selector produces a natural index, but only a subset of
+ * that index space is available as a regset (the TLS slots); subtracting
+ * @bias from a segment selector index value computes the regset slot.
+ *
+ * If nonzero, @core_note_type gives the n_type field (NT_* value)
+ * of the core file note in which this regset's data appears.
+ * NT_PRSTATUS is a special case in that the regset data starts at
+ * offsetof(struct elf_prstatus, pr_reg) into the note data; that is
+ * part of the per-machine ELF formats userland knows about.  In
+ * other cases, the core file note contains exactly the whole regset
+ * (@n * @size) and nothing else.  The core file note is normally
+ * omitted when there is an @active function and it returns zero.
+ */
+struct user_regset {
+	user_regset_get_fn		*get;
+	user_regset_set_fn		*set;
+	user_regset_active_fn		*active;
+	user_regset_writeback_fn	*writeback;
+	unsigned int			n;
+	unsigned int 			size;
+	unsigned int 			align;
+	unsigned int 			bias;
+	unsigned int 			core_note_type;
+};
+
+/**
+ * struct user_regset_view - available regsets
+ * @name:	Identifier, e.g. UTS_MACHINE string.
+ * @regsets:	Array of @n regsets available in this view.
+ * @n:		Number of elements in @regsets.
+ * @e_machine:	ELF header @e_machine %EM_* value written in core dumps.
+ * @e_flags:	ELF header @e_flags value written in core dumps.
+ * @ei_osabi:	ELF header @e_ident[%EI_OSABI] value written in core dumps.
+ *
+ * A regset view is a collection of regsets (&struct user_regset,
+ * above).  This describes all the state of a thread that can be seen
+ * from a given architecture/ABI environment.  More than one view might
+ * refer to the same &struct user_regset, or more than one regset
+ * might refer to the same machine-specific state in the thread.  For
+ * example, a 32-bit thread's state could be examined from the 32-bit
+ * view or from the 64-bit view.  Either method reaches the same thread
+ * register state, doing appropriate widening or truncation.
+ */
+struct user_regset_view {
+	const char *name;
+	const struct user_regset *regsets;
+	unsigned int n;
+	u32 e_flags;
+	u16 e_machine;
+	u8 ei_osabi;
+};
+
+/*
+ * This is documented here rather than at the definition sites because its
+ * implementation is machine-dependent but its interface is universal.
+ */
+/**
+ * task_user_regset_view - Return the process's native regset view.
+ * @tsk: a thread of the process in question
+ *
+ * Return the &struct user_regset_view that is native for the given process.
+ * For example, what it would access when it called ptrace().
+ * Throughout the life of the process, this only changes at exec.
+ */
+const struct user_regset_view *task_user_regset_view(struct task_struct *tsk);
+
+
+#endif	/* <linux/regset.h> */
-- 
cgit v1.2.3


From bae3f7c39dee5951bcbedeaedb6744f882a00173 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:45 +0100
Subject: x86: user_regset helpers

This adds some inlines to linux/regset.h intended for arch code to use in
its user_regset get and set functions.  These make it pretty easy to deal
with the interface's optional kernel-space or user-space pointers and its
generalized access to a part of the register data at a time.

In simple cases where the internal data structure matches the exported
layout (core dump format), a get function can be nothing but a call to
user_regset_copyout, and a set function a call to user_regset_copyin.

In other cases the exported layout is usually made up of a few pieces each
stored contiguously in a different internal data structure.  These helpers
make it straightforward to write a get or set function by processing each
contiguous chunk of the data in order.  The start_pos and end_pos arguments
are always constants, so these inlines collapse to a small amount of code.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/regset.h | 116 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 116 insertions(+)

(limited to 'include')

diff --git a/include/linux/regset.h b/include/linux/regset.h
index 85d0fb0a014d..761c931af975 100644
--- a/include/linux/regset.h
+++ b/include/linux/regset.h
@@ -15,6 +15,7 @@
 
 #include <linux/compiler.h>
 #include <linux/types.h>
+#include <linux/uaccess.h>
 struct task_struct;
 struct user_regset;
 
@@ -203,4 +204,119 @@ struct user_regset_view {
 const struct user_regset_view *task_user_regset_view(struct task_struct *tsk);
 
 
+/*
+ * These are helpers for writing regset get/set functions in arch code.
+ * Because @start_pos and @end_pos are always compile-time constants,
+ * these are inlined into very little code though they look large.
+ *
+ * Use one or more calls sequentially for each chunk of regset data stored
+ * contiguously in memory.  Call with constants for @start_pos and @end_pos,
+ * giving the range of byte positions in the regset that data corresponds
+ * to; @end_pos can be -1 if this chunk is at the end of the regset layout.
+ * Each call updates the arguments to point past its chunk.
+ */
+
+static inline int user_regset_copyout(unsigned int *pos, unsigned int *count,
+				      void **kbuf,
+				      void __user **ubuf, const void *data,
+				      const int start_pos, const int end_pos)
+{
+	if (*count == 0)
+		return 0;
+	BUG_ON(*pos < start_pos);
+	if (end_pos < 0 || *pos < end_pos) {
+		unsigned int copy = (end_pos < 0 ? *count
+				     : min(*count, end_pos - *pos));
+		data += *pos - start_pos;
+		if (*kbuf) {
+			memcpy(*kbuf, data, copy);
+			*kbuf += copy;
+		} else if (__copy_to_user(*ubuf, data, copy))
+			return -EFAULT;
+		else
+			*ubuf += copy;
+		*pos += copy;
+		*count -= copy;
+	}
+	return 0;
+}
+
+static inline int user_regset_copyin(unsigned int *pos, unsigned int *count,
+				     const void **kbuf,
+				     const void __user **ubuf, void *data,
+				     const int start_pos, const int end_pos)
+{
+	if (*count == 0)
+		return 0;
+	BUG_ON(*pos < start_pos);
+	if (end_pos < 0 || *pos < end_pos) {
+		unsigned int copy = (end_pos < 0 ? *count
+				     : min(*count, end_pos - *pos));
+		data += *pos - start_pos;
+		if (*kbuf) {
+			memcpy(data, *kbuf, copy);
+			*kbuf += copy;
+		} else if (__copy_from_user(data, *ubuf, copy))
+			return -EFAULT;
+		else
+			*ubuf += copy;
+		*pos += copy;
+		*count -= copy;
+	}
+	return 0;
+}
+
+/*
+ * These two parallel the two above, but for portions of a regset layout
+ * that always read as all-zero or for which writes are ignored.
+ */
+static inline int user_regset_copyout_zero(unsigned int *pos,
+					   unsigned int *count,
+					   void **kbuf, void __user **ubuf,
+					   const int start_pos,
+					   const int end_pos)
+{
+	if (*count == 0)
+		return 0;
+	BUG_ON(*pos < start_pos);
+	if (end_pos < 0 || *pos < end_pos) {
+		unsigned int copy = (end_pos < 0 ? *count
+				     : min(*count, end_pos - *pos));
+		if (*kbuf) {
+			memset(*kbuf, 0, copy);
+			*kbuf += copy;
+		} else if (__clear_user(*ubuf, copy))
+			return -EFAULT;
+		else
+			*ubuf += copy;
+		*pos += copy;
+		*count -= copy;
+	}
+	return 0;
+}
+
+static inline int user_regset_copyin_ignore(unsigned int *pos,
+					    unsigned int *count,
+					    const void **kbuf,
+					    const void __user **ubuf,
+					    const int start_pos,
+					    const int end_pos)
+{
+	if (*count == 0)
+		return 0;
+	BUG_ON(*pos < start_pos);
+	if (end_pos < 0 || *pos < end_pos) {
+		unsigned int copy = (end_pos < 0 ? *count
+				     : min(*count, end_pos - *pos));
+		if (*kbuf)
+			*kbuf += copy;
+		else
+			*ubuf += copy;
+		*pos += copy;
+		*count -= copy;
+	}
+	return 0;
+}
+
+
 #endif	/* <linux/regset.h> */
-- 
cgit v1.2.3


From 5bde4d181793be84351bc21c256d8c71cfcd313a Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:47 +0100
Subject: x86: user_regset user-copy helpers

This defines two new inlines in linux/regset.h, for use in arch_ptrace
implementations and the like.  These provide simplified wrappers for using
the user_regset interfaces to copy thread regset data into the caller's
user-space memory.  The inlines are trivial, but make the common uses in
places such as ptrace implementation much more concise, easier to read, and
less prone to code-copying errors.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/regset.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

(limited to 'include')

diff --git a/include/linux/regset.h b/include/linux/regset.h
index 761c931af975..8abee6556223 100644
--- a/include/linux/regset.h
+++ b/include/linux/regset.h
@@ -318,5 +318,51 @@ static inline int user_regset_copyin_ignore(unsigned int *pos,
 	return 0;
 }
 
+/**
+ * copy_regset_to_user - fetch a thread's user_regset data into user memory
+ * @target:	thread to be examined
+ * @view:	&struct user_regset_view describing user thread machine state
+ * @setno:	index in @view->regsets
+ * @offset:	offset into the regset data, in bytes
+ * @size:	amount of data to copy, in bytes
+ * @data:	user-mode pointer to copy into
+ */
+static inline int copy_regset_to_user(struct task_struct *target,
+				      const struct user_regset_view *view,
+				      unsigned int setno,
+				      unsigned int offset, unsigned int size,
+				      void __user *data)
+{
+	const struct user_regset *regset = &view->regsets[setno];
+
+	if (!access_ok(VERIFY_WRITE, data, size))
+		return -EIO;
+
+	return regset->get(target, regset, offset, size, NULL, data);
+}
+
+/**
+ * copy_regset_from_user - store into thread's user_regset data from user memory
+ * @target:	thread to be examined
+ * @view:	&struct user_regset_view describing user thread machine state
+ * @setno:	index in @view->regsets
+ * @offset:	offset into the regset data, in bytes
+ * @size:	amount of data to copy, in bytes
+ * @data:	user-mode pointer to copy from
+ */
+static inline int copy_regset_from_user(struct task_struct *target,
+					const struct user_regset_view *view,
+					unsigned int setno,
+					unsigned int offset, unsigned int size,
+					const void __user *data)
+{
+	const struct user_regset *regset = &view->regsets[setno];
+
+	if (!access_ok(VERIFY_READ, data, size))
+		return -EIO;
+
+	return regset->set(target, regset, offset, size, NULL, data);
+}
+
 
 #endif	/* <linux/regset.h> */
-- 
cgit v1.2.3


From 032d82d9065dec0e26718eca376c2029e4bd0595 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:47 +0100
Subject: x86: compat_ptrace_request

This adds a compat_ptrace_request that is the analogue of ptrace_request
for the things that 32-on-64 ptrace implementations can share in common.
So far there are just a couple of requests handled generically.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/compat.h |  4 ++++
 kernel/ptrace.c        | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 42 insertions(+)

(limited to 'include')

diff --git a/include/linux/compat.h b/include/linux/compat.h
index ba29d4c59643..a907fbede6c3 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -243,6 +243,10 @@ asmlinkage long compat_sys_migrate_pages(compat_pid_t pid,
 		compat_ulong_t maxnode, const compat_ulong_t __user *old_nodes,
 		const compat_ulong_t __user *new_nodes);
 
+extern int compat_ptrace_request(struct task_struct *child,
+				 compat_long_t request,
+				 compat_ulong_t addr, compat_ulong_t data);
+
 /*
  * epoll (fs/eventpoll.c) compat bits follow ...
  */
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index e6a99d2793b3..ed1c3d56c2cd 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -607,3 +607,41 @@ int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data)
 	copied = access_process_vm(tsk, addr, &data, sizeof(data), 1);
 	return (copied == sizeof(data)) ? 0 : -EIO;
 }
+
+#ifdef CONFIG_COMPAT
+#include <linux/compat.h>
+
+int compat_ptrace_request(struct task_struct *child, compat_long_t request,
+			  compat_ulong_t addr, compat_ulong_t data)
+{
+	compat_ulong_t __user *datap = compat_ptr(data);
+	compat_ulong_t word;
+	int ret;
+
+	switch (request) {
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKDATA:
+		ret = access_process_vm(child, addr, &word, sizeof(word), 0);
+		if (ret != sizeof(word))
+			ret = -EIO;
+		else
+			ret = put_user(word, datap);
+		break;
+
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEDATA:
+		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
+		ret = (ret != sizeof(data) ? -EIO : 0);
+		break;
+
+	case PTRACE_GETEVENTMSG:
+		ret = put_user((compat_ulong_t) child->ptrace_message, datap);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+	}
+
+	return ret;
+}
+#endif	/* CONFIG_COMPAT */
-- 
cgit v1.2.3


From c269f19617f508cc5c29c0b064c1a437d7011a46 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:48 +0100
Subject: x86: compat_sys_ptrace

This adds a generic definition of compat_sys_ptrace that calls
compat_arch_ptrace, parallel to sys_ptrace/arch_ptrace.  Some
machines needing this already define a function by that name.
The new generic function is defined only on machines that
put #define __ARCH_WANT_COMPAT_SYS_PTRACE into asm/ptrace.h.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/compat.h |  7 +++++++
 kernel/ptrace.c        | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 53 insertions(+)

(limited to 'include')

diff --git a/include/linux/compat.h b/include/linux/compat.h
index a907fbede6c3..d38655f2be70 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -247,6 +247,13 @@ extern int compat_ptrace_request(struct task_struct *child,
 				 compat_long_t request,
 				 compat_ulong_t addr, compat_ulong_t data);
 
+#ifdef __ARCH_WANT_COMPAT_SYS_PTRACE
+extern long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
+			       compat_ulong_t addr, compat_ulong_t data);
+asmlinkage long compat_sys_ptrace(compat_long_t request, compat_long_t pid,
+				  compat_long_t addr, compat_long_t data);
+#endif	/* __ARCH_WANT_COMPAT_SYS_PTRACE */
+
 /*
  * epoll (fs/eventpoll.c) compat bits follow ...
  */
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index ed1c3d56c2cd..e6e9b8be4b05 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -644,4 +644,50 @@ int compat_ptrace_request(struct task_struct *child, compat_long_t request,
 
 	return ret;
 }
+
+#ifdef __ARCH_WANT_COMPAT_SYS_PTRACE
+asmlinkage long compat_sys_ptrace(compat_long_t request, compat_long_t pid,
+				  compat_long_t addr, compat_long_t data)
+{
+	struct task_struct *child;
+	long ret;
+
+	/*
+	 * This lock_kernel fixes a subtle race with suid exec
+	 */
+	lock_kernel();
+	if (request == PTRACE_TRACEME) {
+		ret = ptrace_traceme();
+		goto out;
+	}
+
+	child = ptrace_get_task_struct(pid);
+	if (IS_ERR(child)) {
+		ret = PTR_ERR(child);
+		goto out;
+	}
+
+	if (request == PTRACE_ATTACH) {
+		ret = ptrace_attach(child);
+		/*
+		 * Some architectures need to do book-keeping after
+		 * a ptrace attach.
+		 */
+		if (!ret)
+			arch_ptrace_attach(child);
+		goto out_put_task_struct;
+	}
+
+	ret = ptrace_check_attach(child, request == PTRACE_KILL);
+	if (!ret)
+		ret = compat_arch_ptrace(child, request, addr, data);
+
+ out_put_task_struct:
+	put_task_struct(child);
+ out:
+	unlock_kernel();
+	return ret;
+}
+#endif /* __ARCH_WANT_COMPAT_SYS_PTRACE */
+
 #endif	/* CONFIG_COMPAT */
-- 
cgit v1.2.3


From 863aec860259f287797f886c3f373389cdafc8e3 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:48 +0100
Subject: x86: x86-64 wired cpu_has macros

This adds hard-wired definitions for the remaining cpu_has_* macros
that correspond to flags required-features.h demands are set for
64-bit.  Using these can efficiently avoid some #ifdef's when
merging 32-bit and 64-bit code together.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/cpufeature.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 87dd900154d1..75e2f78a7fda 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -192,9 +192,21 @@
 #undef  cpu_has_centaur_mcr
 #define cpu_has_centaur_mcr	0
 
+#undef  cpu_has_pse
+#define cpu_has_pse		1
+
 #undef  cpu_has_pge
 #define cpu_has_pge		1
 
+#undef  cpu_has_xmm
+#define cpu_has_xmm		1
+
+#undef  cpu_has_xmm2
+#define cpu_has_xmm2		1
+
+#undef  cpu_has_fxsr
+#define cpu_has_fxsr		1
+
 #endif /* CONFIG_X86_64 */
 
 #endif /* _ASM_X86_CPUFEATURE_H */
-- 
cgit v1.2.3


From 99f8ecdf4506010feda26ffe4ac0d23962947106 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:48 +0100
Subject: x86: x86 i387 unify structs

The i387_fxsave_struct formats really have the same layout
on 32 and 64, with only some slightly different use of a few
fields.  The i387_fsave_struct and i387_soft_struct formats
are never used by 64-bit kernels, but it doesn't hurt to
have the unused types in the union and cuts down on the
amount of #ifdef hair required throughout the i387 code.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h | 92 +++++++++++++++++++++------------------------
 1 file changed, 42 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 80c50020ef60..ea222cfe7b00 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -226,64 +226,37 @@ struct orig_ist {
 	unsigned long ist[7];
 };
 
-#ifdef CONFIG_X86_32
-struct i387_fsave_struct {
-	long	cwd;
-	long	swd;
-	long	twd;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	long	status;		/* software status information */
-};
-
-struct i387_fxsave_struct {
-	unsigned short	cwd;
-	unsigned short	swd;
-	unsigned short	twd;
-	unsigned short	fop;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	mxcsr;
-	long	mxcsr_mask;
-	long	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
-	long	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
-	long	padding[56];
-} __attribute__((aligned(16)));
+#define	MXCSR_DEFAULT		0x1f80
 
-struct i387_soft_struct {
-	long	cwd;
-	long	swd;
-	long	twd;
-	long	fip;
-	long	fcs;
-	long	foo;
-	long	fos;
-	long	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	unsigned char	ftop, changed, lookahead, no_update, rm, alimit;
-	struct info	*info;
-	unsigned long	entry_eip;
-};
-
-union i387_union {
-	struct i387_fsave_struct	fsave;
-	struct i387_fxsave_struct	fxsave;
-	struct i387_soft_struct soft;
+struct i387_fsave_struct {
+	u32	cwd;
+	u32	swd;
+	u32	twd;
+	u32	fip;
+	u32	fcs;
+	u32	foo;
+	u32	fos;
+	u32	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
+	u32	status;		/* software status information */
 };
 
-# include "processor_32.h"
-#else
 struct i387_fxsave_struct {
 	u16	cwd;
 	u16	swd;
 	u16	twd;
 	u16	fop;
-	u64	rip;
-	u64	rdp;
+	union {
+		struct {
+			u64	rip;
+			u64	rdp;
+		};
+		struct {
+			u32	fip;
+			u32	fcs;
+			u32	foo;
+			u32	fos;
+		};
+	};
 	u32	mxcsr;
 	u32	mxcsr_mask;
 	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
@@ -291,10 +264,29 @@ struct i387_fxsave_struct {
 	u32	padding[24];
 } __attribute__((aligned(16)));
 
+struct i387_soft_struct {
+	u32	cwd;
+	u32	swd;
+	u32	twd;
+	u32	fip;
+	u32	fcs;
+	u32	foo;
+	u32	fos;
+	u32	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
+	u8	ftop, changed, lookahead, no_update, rm, alimit;
+	struct info	*info;
+	u32	entry_eip;
+};
+
 union i387_union {
+	struct i387_fsave_struct	fsave;
 	struct i387_fxsave_struct	fxsave;
+	struct i387_soft_struct 	soft;
 };
 
+#ifdef CONFIG_X86_32
+# include "processor_32.h"
+#else
 # include "processor_64.h"
 #endif
 
-- 
cgit v1.2.3


From cc927a25bd704448b18b095c658cbfdd79dab865 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:49 +0100
Subject: x86: x86 i387 header cleanup

This moves some code into asm-x86/i387_64.h in preparation for
unifying this code between 32 and 64.  The 32-bit versions of
some things are copied in some existing names changed to match
32-bit names and share code.  For 64, save_i387 is moved into
an inline from i387_64.c; this matches restore_i387, which is
already an inline, and makes sense since there is exactly one
caller (in signal_64.c).  The save_i387 function could use more
cosmetic cleanup, but it is just moved verbatim in this patch.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/i387_64.c |  31 -----
 include/asm-x86/i387_64.h | 315 ++++++++++++++++++++++++++++++++++------------
 2 files changed, 233 insertions(+), 113 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/i387_64.c b/arch/x86/kernel/i387_64.c
index bfaff28fb134..f335a76d7ea7 100644
--- a/arch/x86/kernel/i387_64.c
+++ b/arch/x86/kernel/i387_64.c
@@ -71,37 +71,6 @@ void init_fpu(struct task_struct *child)
 	set_stopped_child_used_math(child);
 }
 
-/*
- * Signal frame handlers.
- */
-
-int save_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err = 0;
-
-	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
-			sizeof(tsk->thread.i387.fxsave));
-
-	if ((unsigned long)buf % 16) 
-		printk("save_i387: bad fpstate %p\n",buf); 
-
-	if (!used_math())
-		return 0;
-	clear_used_math(); /* trigger finit */
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);
-		if (err) return err;
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;
-		stts();
-	} else {
-		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
-				   sizeof(struct i387_fxsave_struct)))
-			return -1;
-	}
-	return 1;
-}
-
 /*
  * ptrace request handlers.
  */
diff --git a/include/asm-x86/i387_64.h b/include/asm-x86/i387_64.h
index 12edb9f2d755..4cea7373ac78 100644
--- a/include/asm-x86/i387_64.h
+++ b/include/asm-x86/i387_64.h
@@ -7,33 +7,23 @@
  * x86-64 work by Andi Kleen 2002
  */
 
-#ifndef __ASM_X86_64_I387_H
-#define __ASM_X86_64_I387_H
+#ifndef _ASM_X86_I387_H
+#define _ASM_X86_I387_H
 
 #include <linux/sched.h>
+#include <linux/kernel_stat.h>
 #include <asm/processor.h>
 #include <asm/sigcontext.h>
 #include <asm/user.h>
-#include <asm/thread_info.h>
 #include <asm/uaccess.h>
 
 extern void fpu_init(void);
 extern unsigned int mxcsr_feature_mask;
 extern void mxcsr_feature_mask_init(void);
 extern void init_fpu(struct task_struct *child);
-extern int save_i387(struct _fpstate __user *buf);
 extern asmlinkage void math_state_restore(void);
 
-/*
- * FPU lazy state save handling...
- */
-
-#define unlazy_fpu(tsk) do { \
-	if (task_thread_info(tsk)->status & TS_USEDFPU) \
-		save_init_fpu(tsk); 			\
-	else						\
-		tsk->fpu_counter = 0;			\
-} while (0)
+#ifdef CONFIG_X86_64
 
 /* Ignore delayed exceptions from user space */
 static inline void tolerant_fwait(void)
@@ -46,52 +36,8 @@ static inline void tolerant_fwait(void)
 		     "	.previous\n");
 }
 
-#define clear_fpu(tsk) do { \
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {	\
-		tolerant_fwait();				\
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;	\
-		stts();						\
-	}							\
-} while (0)
-
-/*
- * ptrace request handers...
- */
-extern int get_fpregs(struct user_i387_struct __user *buf,
-		      struct task_struct *tsk);
-extern int set_fpregs(struct task_struct *tsk,
-		      struct user_i387_struct __user *buf);
-
-/*
- * i387 state interaction
- */
-#define get_fpu_mxcsr(t) ((t)->thread.i387.fxsave.mxcsr)
-#define get_fpu_cwd(t) ((t)->thread.i387.fxsave.cwd)
-#define get_fpu_fxsr_twd(t) ((t)->thread.i387.fxsave.twd)
-#define get_fpu_swd(t) ((t)->thread.i387.fxsave.swd)
-#define set_fpu_cwd(t,val) ((t)->thread.i387.fxsave.cwd = (val))
-#define set_fpu_swd(t,val) ((t)->thread.i387.fxsave.swd = (val))
-#define set_fpu_fxsr_twd(t,val) ((t)->thread.i387.fxsave.twd = (val))
-
-#define X87_FSW_ES (1 << 7)	/* Exception Summary */
-
-/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
-   is pending. Clear the x87 state here by setting it to fixed
-   values. The kernel data segment can be sometimes 0 and sometimes
-   new user value. Both should be ok.
-   Use the PDA as safe address because it should be already in L1. */
-static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
+static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 {
-	if (unlikely(fx->swd & X87_FSW_ES))
-		 asm volatile("fnclex");
-	alternative_input(ASM_NOP8 ASM_NOP2,
-	     	     "    emms\n"		/* clear stack tags */
-	     	     "    fildl %%gs:0",	/* load to clear state */
-		     X86_FEATURE_FXSAVE_LEAK);
-}
-
-static inline int restore_fpu_checking(struct i387_fxsave_struct *fx) 
-{ 
 	int err;
 
 	asm volatile("1:  rex64/fxrstor (%[fx])\n\t"
@@ -105,7 +51,7 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 		     "   .quad  1b,3b\n"
 		     ".previous"
 		     : [err] "=r" (err)
-#if 0 /* See comment in __fxsave_clear() below. */
+#if 0 /* See comment in __save_init_fpu() below. */
 		     : [fx] "r" (fx), "m" (*fx), "0" (0));
 #else
 		     : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0));
@@ -113,10 +59,27 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 	if (unlikely(err))
 		init_fpu(current);
 	return err;
-} 
+}
 
-static inline int save_i387_checking(struct i387_fxsave_struct __user *fx) 
-{ 
+#define X87_FSW_ES (1 << 7)	/* Exception Summary */
+
+/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
+   is pending. Clear the x87 state here by setting it to fixed
+   values. The kernel data segment can be sometimes 0 and sometimes
+   new user value. Both should be ok.
+   Use the PDA as safe address because it should be already in L1. */
+static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
+{
+	if (unlikely(fx->swd & X87_FSW_ES))
+		 asm volatile("fnclex");
+	alternative_input(ASM_NOP8 ASM_NOP2,
+		     "    emms\n"		/* clear stack tags */
+		     "    fildl %%gs:0",	/* load to clear state */
+		     X86_FEATURE_FXSAVE_LEAK);
+}
+
+static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
+{
 	int err;
 
 	asm volatile("1:  rex64/fxsave (%[fx])\n\t"
@@ -139,9 +102,9 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
 		err = -EFAULT;
 	/* No need to clear here because the caller clears USED_MATH */
 	return err;
-} 
+}
 
-static inline void __fxsave_clear(struct task_struct *tsk)
+static inline void __save_init_fpu(struct task_struct *tsk)
 {
 	/* Using "rex64; fxsave %0" is broken because, if the memory operand
 	   uses any extended registers for addressing, a second REX prefix
@@ -169,18 +132,142 @@ static inline void __fxsave_clear(struct task_struct *tsk)
 					      thread.i387.fxsave)));
 #endif
 	clear_fpu_state(&tsk->thread.i387.fxsave);
+	task_thread_info(tsk)->status &= ~TS_USEDFPU;
+}
+
+/*
+ * Signal frame handlers.
+ */
+
+static inline int save_i387(struct _fpstate __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err = 0;
+
+	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
+			sizeof(tsk->thread.i387.fxsave));
+
+	if ((unsigned long)buf % 16)
+		printk("save_i387: bad fpstate %p\n", buf);
+
+	if (!used_math())
+		return 0;
+	clear_used_math(); /* trigger finit */
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);
+		if (err) return err;
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	} else {
+		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
+				   sizeof(struct i387_fxsave_struct)))
+			return -1;
+	}
+	return 1;
+}
+
+/*
+ * This restores directly out of user space. Exceptions are handled.
+ */
+static inline int restore_i387(struct _fpstate __user *buf)
+{
+	set_used_math();
+	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
+		clts();
+		task_thread_info(current)->status |= TS_USEDFPU;
+	}
+	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
+}
+
+#else  /* CONFIG_X86_32 */
+
+static inline void tolerant_fwait(void)
+{
+	asm volatile("fnclex ; fwait");
+}
+
+static inline void restore_fpu(struct task_struct *tsk)
+{
+	/*
+	 * The "nop" is needed to make the instructions the same
+	 * length.
+	 */
+	alternative_input(
+		"nop ; frstor %1",
+		"fxrstor %1",
+		X86_FEATURE_FXSR,
+		"m" ((tsk)->thread.i387.fxsave));
+}
+
+/* We need a safe address that is cheap to find and that is already
+   in L1 during context switch. The best choices are unfortunately
+   different for UP and SMP */
+#ifdef CONFIG_SMP
+#define safe_address (__per_cpu_offset[0])
+#else
+#define safe_address (kstat_cpu(0).cpustat.user)
+#endif
+
+/*
+ * These must be called with preempt disabled
+ */
+static inline void __save_init_fpu(struct task_struct *tsk)
+{
+	/* Use more nops than strictly needed in case the compiler
+	   varies code */
+	alternative_input(
+		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
+		"fxsave %[fx]\n"
+		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
+		X86_FEATURE_FXSR,
+		[fx] "m" (tsk->thread.i387.fxsave),
+		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");
+	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
+	   is pending.  Clear the x87 state here by setting it to fixed
+	   values. safe_address is a random variable that should be in L1 */
+	alternative_input(
+		GENERIC_NOP8 GENERIC_NOP2,
+		"emms\n\t"	  	/* clear stack tags */
+		"fildl %[addr]", 	/* set F?P to defined value */
+		X86_FEATURE_FXSAVE_LEAK,
+		[addr] "m" (safe_address));
+	task_thread_info(tsk)->status &= ~TS_USEDFPU;
+}
+
+/*
+ * Signal frame handlers...
+ */
+extern int save_i387(struct _fpstate __user *buf);
+extern int restore_i387(struct _fpstate __user *buf);
+
+#endif	/* CONFIG_X86_64 */
+
+static inline void __unlazy_fpu(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		__save_init_fpu(tsk);
+		stts();
+	} else
+		tsk->fpu_counter = 0;
+}
+
+static inline void __clear_fpu(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		tolerant_fwait();
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	}
 }
 
 static inline void kernel_fpu_begin(void)
 {
 	struct thread_info *me = current_thread_info();
 	preempt_disable();
-	if (me->status & TS_USEDFPU) {
-		__fxsave_clear(me->task);
-		me->status &= ~TS_USEDFPU;
-		return;
-	}
-	clts();
+	if (me->status & TS_USEDFPU)
+		__save_init_fpu(me->task);
+	else
+		clts();
 }
 
 static inline void kernel_fpu_end(void)
@@ -189,24 +276,88 @@ static inline void kernel_fpu_end(void)
 	preempt_enable();
 }
 
+#ifdef CONFIG_X86_64
+
 static inline void save_init_fpu(struct task_struct *tsk)
 {
- 	__fxsave_clear(tsk);
-	task_thread_info(tsk)->status &= ~TS_USEDFPU;
+	__save_init_fpu(tsk);
 	stts();
 }
 
-/* 
- * This restores directly out of user space. Exceptions are handled.
+#define unlazy_fpu	__unlazy_fpu
+#define clear_fpu	__clear_fpu
+
+#else  /* CONFIG_X86_32 */
+
+/*
+ * These disable preemption on their own and are safe
  */
-static inline int restore_i387(struct _fpstate __user *buf)
+static inline void save_init_fpu(struct task_struct *tsk)
 {
-	set_used_math();
-	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
-		clts();
-		task_thread_info(current)->status |= TS_USEDFPU;
+	preempt_disable();
+	__save_init_fpu(tsk);
+	stts();
+	preempt_enable();
+}
+
+static inline void unlazy_fpu(struct task_struct *tsk)
+{
+	preempt_disable();
+	__unlazy_fpu(tsk);
+	preempt_enable();
+}
+
+static inline void clear_fpu(struct task_struct *tsk)
+{
+	preempt_disable();
+	__clear_fpu(tsk);
+	preempt_enable();
+}
+
+#endif	/* CONFIG_X86_64 */
+
+/*
+ * ptrace request handlers...
+ */
+extern int get_fpregs(struct user_i387_struct __user *buf,
+		      struct task_struct *tsk);
+extern int set_fpregs(struct task_struct *tsk,
+		      struct user_i387_struct __user *buf);
+
+struct user_fxsr_struct;
+extern int get_fpxregs(struct user_fxsr_struct __user *buf,
+		       struct task_struct *tsk);
+extern int set_fpxregs(struct task_struct *tsk,
+		       struct user_fxsr_struct __user *buf);
+
+/*
+ * i387 state interaction
+ */
+static inline unsigned short get_fpu_cwd(struct task_struct *tsk)
+{
+	if (cpu_has_fxsr) {
+		return tsk->thread.i387.fxsave.cwd;
+	} else {
+		return (unsigned short)tsk->thread.i387.fsave.cwd;
+	}
+}
+
+static inline unsigned short get_fpu_swd(struct task_struct *tsk)
+{
+	if (cpu_has_fxsr) {
+		return tsk->thread.i387.fxsave.swd;
+	} else {
+		return (unsigned short)tsk->thread.i387.fsave.swd;
+	}
+}
+
+static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)
+{
+	if (cpu_has_xmm) {
+		return tsk->thread.i387.fxsave.mxcsr;
+	} else {
+		return MXCSR_DEFAULT;
 	}
-	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
 }
 
-#endif /* __ASM_X86_64_I387_H */
+#endif	/* _ASM_X86_I387_H */
-- 
cgit v1.2.3


From 4421011120b2304e5c248ae4165a2704588aedf1 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:50 +0100
Subject: x86: x86 i387 user_regset

This revamps the i387 code to be shared across 32-bit, 64-bit,
and 32-on-64.  It does so by consolidating the code in one place
based on the user_regset accessor interfaces.  This switches
32-bit to using the i387_64.h header and 64-bit to using the
i387.c that was previously i387_32.c, but that's what took the
least cleanup in each file.  Here i387.h is stubbed to always
include i387_64.h rather than renaming the file, to keep this
diff smaller and easier to read.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/Makefile      |   2 +-
 arch/x86/ia32/ia32_signal.c |   5 +-
 arch/x86/kernel/Makefile_64 |   3 +-
 arch/x86/kernel/i387.c      | 481 +++++++++++++++++++++++---------------------
 arch/x86/kernel/ptrace.c    |   1 -
 include/asm-x86/fpu32.h     |  10 -
 include/asm-x86/i387.h      |   6 +-
 include/asm-x86/i387_64.h   |  11 +
 8 files changed, 265 insertions(+), 254 deletions(-)
 delete mode 100644 include/asm-x86/fpu32.h

(limited to 'include')

diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile
index ec71cfeac87e..93a6fda65f49 100644
--- a/arch/x86/ia32/Makefile
+++ b/arch/x86/ia32/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o \
-	ia32_binfmt.o fpu32.o
+	ia32_binfmt.o
 
 sysv-$(CONFIG_SYSVIPC) := ipc32.o
 obj-$(CONFIG_IA32_EMULATION) += $(sysv-y)
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 0e24e3fda3d7..0a34c24f19e5 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -29,7 +29,6 @@
 #include <asm/ia32_unistd.h>
 #include <asm/user32.h>
 #include <asm/sigcontext32.h>
-#include <asm/fpu32.h>
 #include <asm/proto.h>
 #include <asm/vdso.h>
 
@@ -258,7 +257,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 	if (buf) {
 		if (!access_ok(VERIFY_READ, buf, sizeof(*buf)))
 			goto badframe;
-		err |= restore_i387_ia32(current, buf, 0);
+		err |= restore_i387_ia32(buf);
 	} else {
 		struct task_struct *me = current;
 
@@ -377,7 +376,7 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
 	err |= __put_user((u32)regs->flags, &sc->flags);
 	err |= __put_user((u32)regs->sp, &sc->sp_at_signal);
 
-	tmp = save_i387_ia32(current, fpstate, regs, 0);
+	tmp = save_i387_ia32(fpstate);
 	if (tmp < 0)
 		err = -EFAULT;
 	else {
diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index fbb370071239..7fcf972aa5d6 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -7,7 +7,7 @@ CPPFLAGS_vmlinux.lds += -Ux86_64
 
 obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
 		time_64.o ioport_64.o ldt.o setup_64.o i8259_64.o sys_x86_64.o \
-		x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \
+		x8664_ksyms_64.o syscall_64.o vsyscall_64.o \
 		setup64.o bootflag.o e820_64.o reboot_64.o quirks.o i8237.o \
 		pci-dma_64.o pci-nommu_64.o alternative.o hpet.o tsc_64.o bugs_64.o \
 		i8253.o io_delay.o rtc.o
@@ -16,6 +16,7 @@ obj-y				+= ptrace.o
 obj-y				+= ds.o
 obj-y				+= step.o
 
+obj-y				+= i387.o
 obj-$(CONFIG_IA32_EMULATION)	+= tls.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index bebe03463461..f7f7568dd7bc 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -8,6 +8,7 @@
 
 #include <linux/sched.h>
 #include <linux/module.h>
+#include <linux/regset.h>
 #include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/math_emu.h>
@@ -16,13 +17,29 @@
 #include <asm/ptrace.h>
 #include <asm/uaccess.h>
 
+#ifdef CONFIG_X86_64
+
+#include <asm/sigcontext32.h>
+#include <asm/user32.h>
+
+#else
+
+#define	save_i387_ia32		save_i387
+#define	restore_i387_ia32	restore_i387
+
+#define _fpstate_ia32 		_fpstate
+#define user_i387_ia32_struct	user_i387_struct
+#define user32_fxsr_struct	user_fxsr_struct
+
+#endif
+
 #ifdef CONFIG_MATH_EMULATION
 #define HAVE_HWFP (boot_cpu_data.hard_math)
 #else
 #define HAVE_HWFP 1
 #endif
 
-static unsigned long mxcsr_feature_mask __read_mostly = 0xffffffff;
+unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu;
 
 void mxcsr_feature_mask_init(void)
 {
@@ -40,6 +57,30 @@ void mxcsr_feature_mask_init(void)
 	stts();
 }
 
+#ifdef CONFIG_X86_64
+/*
+ * Called at bootup to set up the initial FPU state that is later cloned
+ * into all processes.
+ */
+void __cpuinit fpu_init(void)
+{
+	unsigned long oldcr0 = read_cr0();
+	extern void __bad_fxsave_alignment(void);
+
+	if (offsetof(struct task_struct, thread.i387.fxsave) & 15)
+		__bad_fxsave_alignment();
+	set_in_cr4(X86_CR4_OSFXSR);
+	set_in_cr4(X86_CR4_OSXMMEXCPT);
+
+	write_cr0(oldcr0 & ~((1UL<<3)|(1UL<<2))); /* clear TS and EM */
+
+	mxcsr_feature_mask_init();
+	/* clean state in init */
+	current_thread_info()->status = 0;
+	clear_used_math();
+}
+#endif	/* CONFIG_X86_64 */
+
 /*
  * The _current_ task is using the FPU for the first time
  * so initialize it and set the mxcsr to its default
@@ -48,12 +89,18 @@ void mxcsr_feature_mask_init(void)
  */
 void init_fpu(struct task_struct *tsk)
 {
+	if (tsk_used_math(tsk)) {
+		if (tsk == current)
+			unlazy_fpu(tsk);
+		return;
+	}
+
 	if (cpu_has_fxsr) {
 		memset(&tsk->thread.i387.fxsave, 0,
 		       sizeof(struct i387_fxsave_struct));
 		tsk->thread.i387.fxsave.cwd = 0x37f;
 		if (cpu_has_xmm)
-			tsk->thread.i387.fxsave.mxcsr = 0x1f80;
+			tsk->thread.i387.fxsave.mxcsr = MXCSR_DEFAULT;
 	} else {
 		memset(&tsk->thread.i387.fsave, 0,
 		       sizeof(struct i387_fsave_struct));
@@ -62,27 +109,59 @@ void init_fpu(struct task_struct *tsk)
 		tsk->thread.i387.fsave.twd = 0xffffffffu;
 		tsk->thread.i387.fsave.fos = 0xffff0000u;
 	}
-	/* only the device not available exception
-	 * or ptrace can call init_fpu */
+	/*
+	 * Only the device not available exception or ptrace can call init_fpu.
+	 */
 	set_stopped_child_used_math(tsk);
 }
 
-/*
- * FPU lazy state save handling.
- */
+int fpregs_active(struct task_struct *target, const struct user_regset *regset)
+{
+	return tsk_used_math(target) ? regset->n : 0;
+}
 
-void kernel_fpu_begin(void)
+int xfpregs_active(struct task_struct *target, const struct user_regset *regset)
 {
-	struct thread_info *thread = current_thread_info();
+	return (cpu_has_fxsr && tsk_used_math(target)) ? regset->n : 0;
+}
 
-	preempt_disable();
-	if (thread->status & TS_USEDFPU) {
-		__save_init_fpu(thread->task);
-		return;
-	}
-	clts();
+int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		void *kbuf, void __user *ubuf)
+{
+	if (!cpu_has_fxsr)
+		return -ENODEV;
+
+	unlazy_fpu(target);
+
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   &target->thread.i387.fxsave, 0, -1);
 }
-EXPORT_SYMBOL_GPL(kernel_fpu_begin);
+
+int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		const void *kbuf, const void __user *ubuf)
+{
+	int ret;
+
+	if (!cpu_has_fxsr)
+		return -ENODEV;
+
+	unlazy_fpu(target);
+	set_stopped_child_used_math(target);
+
+	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+				 &target->thread.i387.fxsave, 0, -1);
+
+	/*
+	 * mxcsr reserved bits must be masked to zero for security reasons.
+	 */
+	target->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
+
+	return ret;
+}
+
+#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
 
 /*
  * FPU tag word conversions.
@@ -94,210 +173,187 @@ static inline unsigned short twd_i387_to_fxsr(unsigned short twd)
 
 	/* Transform each pair of bits into 01 (valid) or 00 (empty) */
 	tmp = ~twd;
-	tmp = (tmp | (tmp >> 1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */
+	tmp = (tmp | (tmp>>1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */
 	/* and move the valid bits to the lower byte. */
 	tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */
 	tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */
 	tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */
-
 	return tmp;
 }
 
-static inline unsigned long twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave)
-{
-	struct _fpxreg *st = NULL;
-	unsigned long tos = (fxsave->swd >> 11) & 7;
-	unsigned long twd = (unsigned long) fxsave->twd;
-	unsigned long tag;
-	unsigned long ret = 0xffff0000u;
-	int i;
-
 #define FPREG_ADDR(f, n)	((void *)&(f)->st_space + (n) * 16);
+#define FP_EXP_TAG_VALID	0
+#define FP_EXP_TAG_ZERO		1
+#define FP_EXP_TAG_SPECIAL	2
+#define FP_EXP_TAG_EMPTY	3
+
+static inline u32 twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave)
+{
+	struct _fpxreg *st;
+	u32 tos = (fxsave->swd >> 11) & 7;
+	u32 twd = (unsigned long) fxsave->twd;
+	u32 tag;
+	u32 ret = 0xffff0000u;
+	int i;
 
-	for (i = 0; i < 8; i++) {
+	for (i = 0; i < 8; i++, twd >>= 1) {
 		if (twd & 0x1) {
 			st = FPREG_ADDR(fxsave, (i - tos) & 7);
 
 			switch (st->exponent & 0x7fff) {
 			case 0x7fff:
-				tag = 2;		/* Special */
+				tag = FP_EXP_TAG_SPECIAL;
 				break;
 			case 0x0000:
 				if (!st->significand[0] &&
 				    !st->significand[1] &&
 				    !st->significand[2] &&
-				    !st->significand[3]) {
-					tag = 1;	/* Zero */
-				} else {
-					tag = 2;	/* Special */
-				}
+				    !st->significand[3])
+					tag = FP_EXP_TAG_ZERO;
+				else
+					tag = FP_EXP_TAG_SPECIAL;
 				break;
 			default:
-				if (st->significand[3] & 0x8000) {
-					tag = 0;	/* Valid */
-				} else {
-					tag = 2;	/* Special */
-				}
+				if (st->significand[3] & 0x8000)
+					tag = FP_EXP_TAG_VALID;
+				else
+					tag = FP_EXP_TAG_SPECIAL;
 				break;
 			}
 		} else {
-			tag = 3;			/* Empty */
+			tag = FP_EXP_TAG_EMPTY;
 		}
-		ret |= (tag << (2 * i));
-		twd = twd >> 1;
+		ret |= tag << (2 * i);
 	}
 	return ret;
 }
 
 /*
- * FPU state interaction.
+ * FXSR floating point environment conversions.
  */
 
-unsigned short get_fpu_cwd(struct task_struct *tsk)
+static void convert_from_fxsr(struct user_i387_ia32_struct *env,
+			      struct task_struct *tsk)
 {
-	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.cwd;
-	} else {
-		return (unsigned short)tsk->thread.i387.fsave.cwd;
-	}
-}
+	struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave;
+	struct _fpreg *to = (struct _fpreg *) &env->st_space[0];
+	struct _fpxreg *from = (struct _fpxreg *) &fxsave->st_space[0];
+	int i;
 
-unsigned short get_fpu_swd(struct task_struct *tsk)
-{
-	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.swd;
+	env->cwd = fxsave->cwd | 0xffff0000u;
+	env->swd = fxsave->swd | 0xffff0000u;
+	env->twd = twd_fxsr_to_i387(fxsave);
+
+#ifdef CONFIG_X86_64
+	env->fip = fxsave->rip;
+	env->foo = fxsave->rdp;
+	if (tsk == current) {
+		/*
+		 * should be actually ds/cs at fpu exception time, but
+		 * that information is not available in 64bit mode.
+		 */
+		asm("mov %%ds,%0" : "=r" (env->fos));
+		asm("mov %%cs,%0" : "=r" (env->fcs));
 	} else {
-		return (unsigned short)tsk->thread.i387.fsave.swd;
+		struct pt_regs *regs = task_pt_regs(tsk);
+		env->fos = 0xffff0000 | tsk->thread.ds;
+		env->fcs = regs->cs;
 	}
-}
-
-#if 0
-unsigned short get_fpu_twd(struct task_struct *tsk)
-{
-	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.twd;
-	} else {
-		return (unsigned short)tsk->thread.i387.fsave.twd;
-	}
-}
-#endif  /*  0  */
+#else
+	env->fip = fxsave->fip;
+	env->fcs = fxsave->fcs;
+	env->foo = fxsave->foo;
+	env->fos = fxsave->fos;
+#endif
 
-unsigned short get_fpu_mxcsr(struct task_struct *tsk)
-{
-	if (cpu_has_xmm) {
-		return tsk->thread.i387.fxsave.mxcsr;
-	} else {
-		return 0x1f80;
-	}
+	for (i = 0; i < 8; ++i)
+		memcpy(&to[i], &from[i], sizeof(to[0]));
 }
 
-#if 0
+static void convert_to_fxsr(struct task_struct *tsk,
+			    const struct user_i387_ia32_struct *env)
 
-void set_fpu_cwd(struct task_struct *tsk, unsigned short cwd)
 {
-	if (cpu_has_fxsr) {
-		tsk->thread.i387.fxsave.cwd = cwd;
-	} else {
-		tsk->thread.i387.fsave.cwd = ((long)cwd | 0xffff0000u);
-	}
-}
+	struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave;
+	struct _fpreg *from = (struct _fpreg *) &env->st_space[0];
+	struct _fpxreg *to = (struct _fpxreg *) &fxsave->st_space[0];
+	int i;
 
-void set_fpu_swd(struct task_struct *tsk, unsigned short swd)
-{
-	if (cpu_has_fxsr) {
-		tsk->thread.i387.fxsave.swd = swd;
-	} else {
-		tsk->thread.i387.fsave.swd = ((long)swd | 0xffff0000u);
-	}
-}
+	fxsave->cwd = env->cwd;
+	fxsave->swd = env->swd;
+	fxsave->twd = twd_i387_to_fxsr(env->twd);
+	fxsave->fop = (u16) ((u32) env->fcs >> 16);
+#ifdef CONFIG_X86_64
+	fxsave->rip = env->fip;
+	fxsave->rdp = env->foo;
+	/* cs and ds ignored */
+#else
+	fxsave->fip = env->fip;
+	fxsave->fcs = (env->fcs & 0xffff);
+	fxsave->foo = env->foo;
+	fxsave->fos = env->fos;
+#endif
 
-void set_fpu_twd(struct task_struct *tsk, unsigned short twd)
-{
-	if (cpu_has_fxsr) {
-		tsk->thread.i387.fxsave.twd = twd_i387_to_fxsr(twd);
-	} else {
-		tsk->thread.i387.fsave.twd = ((long)twd | 0xffff0000u);
-	}
+	for (i = 0; i < 8; ++i)
+		memcpy(&to[i], &from[i], sizeof(from[0]));
 }
 
-#endif  /*  0  */
-
-/*
- * FXSR floating point environment conversions.
- */
-
-static int convert_fxsr_to_user(struct _fpstate __user *buf,
-				struct i387_fxsave_struct *fxsave)
+int fpregs_get(struct task_struct *target, const struct user_regset *regset,
+	       unsigned int pos, unsigned int count,
+	       void *kbuf, void __user *ubuf)
 {
-	unsigned long env[7];
-	struct _fpreg __user *to;
-	struct _fpxreg *from;
-	int i;
+	struct user_i387_ia32_struct env;
 
-	env[0] = (unsigned long)fxsave->cwd | 0xffff0000ul;
-	env[1] = (unsigned long)fxsave->swd | 0xffff0000ul;
-	env[2] = twd_fxsr_to_i387(fxsave);
-	env[3] = fxsave->fip;
-	env[4] = fxsave->fcs | ((unsigned long)fxsave->fop << 16);
-	env[5] = fxsave->foo;
-	env[6] = fxsave->fos;
+	if (!HAVE_HWFP)
+		return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
 
-	if (__copy_to_user(buf, env, 7 * sizeof(unsigned long)))
-		return 1;
+	unlazy_fpu(target);
 
-	to = &buf->_st[0];
-	from = (struct _fpxreg *) &fxsave->st_space[0];
-	for (i = 0; i < 8; i++, to++, from++) {
-		unsigned long __user *t = (unsigned long __user *)to;
-		unsigned long *f = (unsigned long *)from;
+	if (!cpu_has_fxsr)
+		return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+					   &target->thread.i387.fsave, 0, -1);
 
-		if (__put_user(*f, t) ||
-		    __put_user(*(f + 1), t + 1) ||
-		    __put_user(from->exponent, &to->exponent))
-			return 1;
+	if (kbuf && pos == 0 && count == sizeof(env)) {
+		convert_from_fxsr(kbuf, target);
+		return 0;
 	}
-	return 0;
+
+	convert_from_fxsr(&env, target);
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, &env, 0, -1);
 }
 
-static int convert_fxsr_from_user(struct i387_fxsave_struct *fxsave,
-				  struct _fpstate __user *buf)
+int fpregs_set(struct task_struct *target, const struct user_regset *regset,
+	       unsigned int pos, unsigned int count,
+	       const void *kbuf, const void __user *ubuf)
 {
-	unsigned long env[7];
-	struct _fpxreg *to;
-	struct _fpreg __user *from;
-	int i;
+	struct user_i387_ia32_struct env;
+	int ret;
 
-	if (__copy_from_user(env, buf, 7 * sizeof(long)))
-		return 1;
+	if (!HAVE_HWFP)
+		return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
 
-	fxsave->cwd = (unsigned short)(env[0] & 0xffff);
-	fxsave->swd = (unsigned short)(env[1] & 0xffff);
-	fxsave->twd = twd_i387_to_fxsr((unsigned short)(env[2] & 0xffff));
-	fxsave->fip = env[3];
-	fxsave->fop = (unsigned short)((env[4] & 0xffff0000ul) >> 16);
-	fxsave->fcs = (env[4] & 0xffff);
-	fxsave->foo = env[5];
-	fxsave->fos = env[6];
-
-	to = (struct _fpxreg *) &fxsave->st_space[0];
-	from = &buf->_st[0];
-	for (i = 0; i < 8; i++, to++, from++) {
-		unsigned long *t = (unsigned long *)to;
-		unsigned long __user *f = (unsigned long __user *)from;
-
-		if (__get_user(*t, f) ||
-		    __get_user(*(t + 1), f + 1) ||
-		    __get_user(to->exponent, &from->exponent))
-			return 1;
-	}
-	return 0;
+	unlazy_fpu(target);
+	set_stopped_child_used_math(target);
+
+	if (!cpu_has_fxsr)
+		return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+					  &target->thread.i387.fsave, 0, -1);
+
+	if (pos > 0 || count < sizeof(env))
+		convert_from_fxsr(&env, target);
+
+	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &env, 0, -1);
+	if (!ret)
+		convert_to_fxsr(target, &env);
+
+	return ret;
 }
 
 /*
  * Signal frame handlers.
  */
 
-static inline int save_i387_fsave(struct _fpstate __user *buf)
+static inline int save_i387_fsave(struct _fpstate_ia32 __user *buf)
 {
 	struct task_struct *tsk = current;
 
@@ -309,14 +365,16 @@ static inline int save_i387_fsave(struct _fpstate __user *buf)
 	return 1;
 }
 
-static int save_i387_fxsave(struct _fpstate __user *buf)
+static int save_i387_fxsave(struct _fpstate_ia32 __user *buf)
 {
 	struct task_struct *tsk = current;
+	struct user_i387_ia32_struct env;
 	int err = 0;
 
 	unlazy_fpu(tsk);
 
-	if (convert_fxsr_to_user(buf, &tsk->thread.i387.fxsave))
+	convert_from_fxsr(&env, tsk);
+	if (__copy_to_user(buf, &env, sizeof(env)))
 		return -1;
 
 	err |= __put_user(tsk->thread.i387.fxsave.swd, &buf->status);
@@ -330,7 +388,7 @@ static int save_i387_fxsave(struct _fpstate __user *buf)
 	return 1;
 }
 
-int save_i387(struct _fpstate __user *buf)
+int save_i387_ia32(struct _fpstate_ia32 __user *buf)
 {
 	if (!used_math())
 		return 0;
@@ -347,11 +405,13 @@ int save_i387(struct _fpstate __user *buf)
 			return save_i387_fsave(buf);
 		}
 	} else {
-		return save_i387_soft(&current->thread.i387.soft, buf);
+		return fpregs_soft_get(current, NULL,
+				       0, sizeof(struct user_i387_ia32_struct),
+				       NULL, buf) ? -1 : 1;
 	}
 }
 
-static inline int restore_i387_fsave(struct _fpstate __user *buf)
+static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
 {
 	struct task_struct *tsk = current;
 	clear_fpu(tsk);
@@ -359,19 +419,23 @@ static inline int restore_i387_fsave(struct _fpstate __user *buf)
 				sizeof(struct i387_fsave_struct));
 }
 
-static int restore_i387_fxsave(struct _fpstate __user *buf)
+static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf)
 {
 	int err;
 	struct task_struct *tsk = current;
+	struct user_i387_ia32_struct env;
 	clear_fpu(tsk);
 	err = __copy_from_user(&tsk->thread.i387.fxsave, &buf->_fxsr_env[0],
 			       sizeof(struct i387_fxsave_struct));
 	/* mxcsr reserved bits must be masked to zero for security reasons */
 	tsk->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
-	return err ? 1 : convert_fxsr_from_user(&tsk->thread.i387.fxsave, buf);
+	if (err || __copy_from_user(&env, buf, sizeof(env)))
+		return 1;
+	convert_to_fxsr(tsk, &env);
+	return 0;
 }
 
-int restore_i387(struct _fpstate __user *buf)
+int restore_i387_ia32(struct _fpstate_ia32 __user *buf)
 {
 	int err;
 
@@ -382,101 +446,52 @@ int restore_i387(struct _fpstate __user *buf)
 			err = restore_i387_fsave(buf);
 		}
 	} else {
-		err = restore_i387_soft(&current->thread.i387.soft, buf);
+		err = fpregs_soft_set(current, NULL,
+				      0, sizeof(struct user_i387_ia32_struct),
+				      NULL, buf) != 0;
 	}
 	set_used_math();
 	return err;
 }
 
-/*
- * ptrace request handlers.
- */
+#endif	/* CONFIG_X86_32 || CONFIG_IA32_EMULATION */
 
-static inline int get_fpregs_fsave(struct user_i387_struct __user *buf,
-				   struct task_struct *tsk)
-{
-	return __copy_to_user(buf, &tsk->thread.i387.fsave,
-			      sizeof(struct user_i387_struct));
-}
-
-static inline int get_fpregs_fxsave(struct user_i387_struct __user *buf,
-				    struct task_struct *tsk)
-{
-	return convert_fxsr_to_user((struct _fpstate __user *)buf,
-				    &tsk->thread.i387.fxsave);
-}
+#ifdef CONFIG_X86_64
 
 int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
 {
-	if (HAVE_HWFP) {
-		if (cpu_has_fxsr) {
-			return get_fpregs_fxsave(buf, tsk);
-		} else {
-			return get_fpregs_fsave(buf, tsk);
-		}
-	} else {
-		return save_i387_soft(&tsk->thread.i387.soft,
-				      (struct _fpstate __user *)buf);
-	}
+	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
-static inline int set_fpregs_fsave(struct task_struct *tsk,
-				   struct user_i387_struct __user *buf)
+int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
 {
-	return __copy_from_user(&tsk->thread.i387.fsave, buf,
-				sizeof(struct user_i387_struct));
+	return xfpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
-static inline int set_fpregs_fxsave(struct task_struct *tsk,
-				    struct user_i387_struct __user *buf)
+#else
+
+int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
 {
-	return convert_fxsr_from_user(&tsk->thread.i387.fxsave,
-				      (struct _fpstate __user *)buf);
+	return fpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
 int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
 {
-	if (HAVE_HWFP) {
-		if (cpu_has_fxsr) {
-			return set_fpregs_fxsave(tsk, buf);
-		} else {
-			return set_fpregs_fsave(tsk, buf);
-		}
-	} else {
-		return restore_i387_soft(&tsk->thread.i387.soft,
-					 (struct _fpstate __user *)buf);
-	}
+	return fpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
 int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *tsk)
 {
-	if (cpu_has_fxsr) {
-		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
-				   sizeof(struct user_fxsr_struct)))
-			return -EFAULT;
-		return 0;
-	} else {
-		return -EIO;
-	}
+	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
 int set_fpxregs(struct task_struct *tsk, struct user_fxsr_struct __user *buf)
 {
-	int ret = 0;
-
-	if (cpu_has_fxsr) {
-		if (__copy_from_user(&tsk->thread.i387.fxsave, buf,
-				     sizeof(struct user_fxsr_struct)))
-			ret = -EFAULT;
-		/* mxcsr reserved bits must be masked to zero
-		 * for security reasons */
-		tsk->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
-	} else {
-		ret = -EIO;
-	}
-	return ret;
+	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
 }
 
+#endif
+
 /*
  * FPU state for core dumps.
  */
@@ -538,7 +553,7 @@ int dump_task_fpu(struct task_struct *tsk, struct user_i387_struct *fpu)
 }
 
 int dump_task_extended_fpu(struct task_struct *tsk,
-			   struct user_fxsr_struct *fpu)
+			   struct user32_fxsr_struct *fpu)
 {
 	int fpvalid = tsk_used_math(tsk) && cpu_has_fxsr;
 
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 18972a305890..084805ab7323 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -886,7 +886,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 #include <linux/compat.h>
 #include <linux/syscalls.h>
 #include <asm/ia32.h>
-#include <asm/fpu32.h>
 #include <asm/user32.h>
 
 #define R32(l,q)							\
diff --git a/include/asm-x86/fpu32.h b/include/asm-x86/fpu32.h
deleted file mode 100644
index 4153db5c0c31..000000000000
--- a/include/asm-x86/fpu32.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef _FPU32_H
-#define _FPU32_H 1
-
-struct _fpstate_ia32;
-
-int restore_i387_ia32(struct task_struct *tsk, struct _fpstate_ia32 __user *buf, int fsave);
-int save_i387_ia32(struct task_struct *tsk, struct _fpstate_ia32 __user *buf, 
-		   struct pt_regs *regs, int fsave);
-
-#endif
diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index a8bbed349664..b2bc0050ce99 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -1,5 +1 @@
-#ifdef CONFIG_X86_32
-# include "i387_32.h"
-#else
-# include "i387_64.h"
-#endif
+#include "i387_64.h"
diff --git a/include/asm-x86/i387_64.h b/include/asm-x86/i387_64.h
index 4cea7373ac78..de435b9114df 100644
--- a/include/asm-x86/i387_64.h
+++ b/include/asm-x86/i387_64.h
@@ -12,6 +12,7 @@
 
 #include <linux/sched.h>
 #include <linux/kernel_stat.h>
+#include <linux/regset.h>
 #include <asm/processor.h>
 #include <asm/sigcontext.h>
 #include <asm/user.h>
@@ -23,6 +24,16 @@ extern void mxcsr_feature_mask_init(void);
 extern void init_fpu(struct task_struct *child);
 extern asmlinkage void math_state_restore(void);
 
+extern user_regset_active_fn fpregs_active, xfpregs_active;
+extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
+extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
+
+#ifdef CONFIG_IA32_EMULATION
+struct _fpstate_ia32;
+extern int save_i387_ia32(struct _fpstate_ia32 __user *buf);
+extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf);
+#endif
+
 #ifdef CONFIG_X86_64
 
 /* Ignore delayed exceptions from user space */
-- 
cgit v1.2.3


From 1eeaed7679eab3666d2d8c964d060c2169b3813b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:51 +0100
Subject: x86: x86 i387 cleanup

This removes all the old code that is no longer used after
the i387 unification and cleanup.  The i387_64.h is renamed
to i387.h with no changes, but since it replaces the nonempty
one-line stub i387.h it looks like a big diff and not a rename.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/ia32/fpu32.c     | 181 ----------------------
 arch/x86/kernel/i387_64.c | 119 ---------------
 include/asm-x86/i387.h    | 375 +++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/i387_32.h | 149 ------------------
 include/asm-x86/i387_64.h | 374 ---------------------------------------------
 5 files changed, 374 insertions(+), 824 deletions(-)
 delete mode 100644 arch/x86/ia32/fpu32.c
 delete mode 100644 arch/x86/kernel/i387_64.c
 delete mode 100644 include/asm-x86/i387_32.h
 delete mode 100644 include/asm-x86/i387_64.h

(limited to 'include')

diff --git a/arch/x86/ia32/fpu32.c b/arch/x86/ia32/fpu32.c
deleted file mode 100644
index ae80745f668f..000000000000
--- a/arch/x86/ia32/fpu32.c
+++ /dev/null
@@ -1,181 +0,0 @@
-/*
- * Copyright 2002 Andi Kleen, SuSE Labs.
- * FXSAVE<->i387 conversion support. Based on code by Gareth Hughes.
- * This is used for ptrace, signals and coredumps in 32bit emulation.
- */
-
-#include <linux/sched.h>
-#include <asm/sigcontext32.h>
-#include <asm/processor.h>
-#include <asm/uaccess.h>
-#include <asm/i387.h>
-
-static inline unsigned short twd_i387_to_fxsr(unsigned short twd)
-{
-	unsigned int tmp; /* to avoid 16 bit prefixes in the code */
-
-	/* Transform each pair of bits into 01 (valid) or 00 (empty) */
-	tmp = ~twd;
-	tmp = (tmp | (tmp>>1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */
-	/* and move the valid bits to the lower byte. */
-	tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */
-	tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */
-	tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */
-	return tmp;
-}
-
-#define FPREG_ADDR(f, n)	((void *)&(f)->st_space + (n) * 16);
-#define FP_EXP_TAG_VALID	0
-#define FP_EXP_TAG_ZERO		1
-#define FP_EXP_TAG_SPECIAL	2
-#define FP_EXP_TAG_EMPTY	3
-
-static inline unsigned long twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave)
-{
-	struct _fpxreg *st;
-	unsigned long tos = (fxsave->swd >> 11) & 7;
-	unsigned long twd = (unsigned long) fxsave->twd;
-	unsigned long tag;
-	unsigned long ret = 0xffff0000;
-	int i;
-
-	for (i = 0; i < 8; i++, twd >>= 1) {
-		if (twd & 0x1) {
-			st = FPREG_ADDR(fxsave, (i - tos) & 7);
-
-			switch (st->exponent & 0x7fff) {
-			case 0x7fff:
-				tag = FP_EXP_TAG_SPECIAL;
-				break;
-			case 0x0000:
-				if (!st->significand[0] &&
-				    !st->significand[1] &&
-				    !st->significand[2] &&
-				    !st->significand[3])
-					tag = FP_EXP_TAG_ZERO;
-				else
-					tag = FP_EXP_TAG_SPECIAL;
-				break;
-			default:
-				if (st->significand[3] & 0x8000)
-					tag = FP_EXP_TAG_VALID;
-				else
-					tag = FP_EXP_TAG_SPECIAL;
-				break;
-			}
-		} else {
-			tag = FP_EXP_TAG_EMPTY;
-		}
-		ret |= tag << (2 * i);
-	}
-	return ret;
-}
-
-#define G(num, val) err |= __get_user(val, num + (u32 __user *)buf)
-
-static inline int convert_fxsr_from_user(struct i387_fxsave_struct *fxsave,
-					 struct _fpstate_ia32 __user *buf)
-{
-	struct _fpxreg *to;
-	struct _fpreg __user *from;
-	int i, err = 0;
-	u32 v;
-
-	G(0, fxsave->cwd);
-	G(1, fxsave->swd);
-	G(2, fxsave->twd);
-	fxsave->twd = twd_i387_to_fxsr(fxsave->twd);
-	G(3, fxsave->rip);
-	G(4, v);
-	/* cs ignored */
-	fxsave->fop = v>>16;
-	G(5, fxsave->rdp);
-	/* 6: ds ignored */
-	if (err)
-		return -1;
-
-	to = (struct _fpxreg *)&fxsave->st_space[0];
-	from = &buf->_st[0];
-	for (i = 0; i < 8; i++, to++, from++) {
-		if (__copy_from_user(to, from, sizeof(*from)))
-			return -1;
-	}
-	return 0;
-}
-
-#define P(num, val) err |= __put_user(val, num + (u32 __user *)buf)
-
-static inline int convert_fxsr_to_user(struct _fpstate_ia32 __user *buf,
-				       struct i387_fxsave_struct *fxsave,
-				       struct pt_regs *regs,
-				       struct task_struct *tsk)
-{
-	struct _fpreg __user *to;
-	struct _fpxreg *from;
-	int i, err = 0;
-	u16 cs, ds;
-
-	if (tsk == current) {
-		/*
-		 * should be actually ds/cs at fpu exception time, but
-		 * that information is not available in 64bit mode.
-		 */
-		asm("movw %%ds,%0 " : "=r" (ds));
-		asm("movw %%cs,%0 " : "=r" (cs));
-	} else {
-		 /* ptrace. task has stopped. */
-		ds = tsk->thread.ds;
-		cs = regs->cs;
-	}
-
-	P(0, (u32)fxsave->cwd | 0xffff0000);
-	P(1, (u32)fxsave->swd | 0xffff0000);
-	P(2, twd_fxsr_to_i387(fxsave));
-	P(3, (u32)fxsave->rip);
-	P(4,  cs | ((u32)fxsave->fop) << 16);
-	P(5, fxsave->rdp);
-	P(6, 0xffff0000 | ds);
-
-	if (err)
-		return -1;
-
-	to = &buf->_st[0];
-	from = (struct _fpxreg *) &fxsave->st_space[0];
-	for (i = 0; i < 8; i++, to++, from++) {
-		if (__copy_to_user(to, from, sizeof(*to)))
-			return -1;
-	}
-	return 0;
-}
-
-int restore_i387_ia32(struct task_struct *tsk,
-		      struct _fpstate_ia32 __user *buf, int fsave)
-{
-	clear_fpu(tsk);
-	if (!fsave) {
-		if (__copy_from_user(&tsk->thread.i387.fxsave,
-				     &buf->_fxsr_env[0],
-				     sizeof(struct i387_fxsave_struct)))
-			return -1;
-		tsk->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
-		set_stopped_child_used_math(tsk);
-	}
-	return convert_fxsr_from_user(&tsk->thread.i387.fxsave, buf);
-}
-
-int save_i387_ia32(struct task_struct *tsk, struct _fpstate_ia32 __user *buf,
-		   struct pt_regs *regs, int fsave)
-{
-	int err = 0;
-
-	init_fpu(tsk);
-	if (convert_fxsr_to_user(buf, &tsk->thread.i387.fxsave, regs, tsk))
-		return -1;
-	if (fsave)
-		return 0;
-	err |= __put_user(tsk->thread.i387.fxsave.swd, &buf->status);
-	err |= __put_user(X86_FXSR_MAGIC, &buf->magic);
-	err |= __copy_to_user(&buf->_fxsr_env[0], &tsk->thread.i387.fxsave,
-			      sizeof(struct i387_fxsave_struct));
-	return err ? -1 : 1;
-}
diff --git a/arch/x86/kernel/i387_64.c b/arch/x86/kernel/i387_64.c
deleted file mode 100644
index f335a76d7ea7..000000000000
--- a/arch/x86/kernel/i387_64.c
+++ /dev/null
@@ -1,119 +0,0 @@
-/*
- *  Copyright (C) 1994 Linus Torvalds
- *  Copyright (C) 2002 Andi Kleen, SuSE Labs
- *
- *  Pentium III FXSR, SSE support
- *  General FPU state handling cleanups
- *	Gareth Hughes <gareth@valinux.com>, May 2000
- * 
- *  x86-64 rework 2002 Andi Kleen. 
- *  Does direct fxsave in and out of user space now for signal handlers.
- *  All the FSAVE<->FXSAVE conversion code has been moved to the 32bit emulation,
- *  the 64bit user space sees a FXSAVE frame directly. 
- */
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <asm/processor.h>
-#include <asm/i387.h>
-#include <asm/sigcontext.h>
-#include <asm/user.h>
-#include <asm/ptrace.h>
-#include <asm/uaccess.h>
-
-unsigned int mxcsr_feature_mask __read_mostly = 0xffffffff;
-
-void mxcsr_feature_mask_init(void)
-{
-	unsigned int mask;
-	clts();
-	memset(&current->thread.i387.fxsave, 0, sizeof(struct i387_fxsave_struct));
-	asm volatile("fxsave %0" : : "m" (current->thread.i387.fxsave));
-	mask = current->thread.i387.fxsave.mxcsr_mask;
-	if (mask == 0) mask = 0x0000ffbf;
-	mxcsr_feature_mask &= mask;
-	stts();
-}
-
-/*
- * Called at bootup to set up the initial FPU state that is later cloned
- * into all processes.
- */
-void __cpuinit fpu_init(void)
-{
-	unsigned long oldcr0 = read_cr0();
-	extern void __bad_fxsave_alignment(void);
-		
-	if (offsetof(struct task_struct, thread.i387.fxsave) & 15)
-		__bad_fxsave_alignment();
-	set_in_cr4(X86_CR4_OSFXSR);
-	set_in_cr4(X86_CR4_OSXMMEXCPT);
-
-	write_cr0(oldcr0 & ~((1UL<<3)|(1UL<<2))); /* clear TS and EM */
-
-	mxcsr_feature_mask_init();
-	/* clean state in init */
-	current_thread_info()->status = 0;
-	clear_used_math();
-}
-
-void init_fpu(struct task_struct *child)
-{
-	if (tsk_used_math(child)) {
-		if (child == current)
-			unlazy_fpu(child);
-		return;
-	}	
-	memset(&child->thread.i387.fxsave, 0, sizeof(struct i387_fxsave_struct));
-	child->thread.i387.fxsave.cwd = 0x37f;
-	child->thread.i387.fxsave.mxcsr = 0x1f80;
-	/* only the device not available exception or ptrace can call init_fpu */
-	set_stopped_child_used_math(child);
-}
-
-/*
- * ptrace request handlers.
- */
-
-int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
-{
-	init_fpu(tsk);
-	return __copy_to_user(buf, &tsk->thread.i387.fxsave,
-			       sizeof(struct user_i387_struct)) ? -EFAULT : 0;
-}
-
-int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
-{
-	if (__copy_from_user(&tsk->thread.i387.fxsave, buf, 
-			     sizeof(struct user_i387_struct)))
-		return -EFAULT;
-		return 0;
-}
-
-/*
- * FPU state for core dumps.
- */
-
-int dump_fpu( struct pt_regs *regs, struct user_i387_struct *fpu )
-{
-	struct task_struct *tsk = current;
-
-	if (!used_math())
-		return 0;
-
-	unlazy_fpu(tsk);
-	memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(struct user_i387_struct)); 
-	return 1; 
-}
-
-int dump_task_fpu(struct task_struct *tsk, struct user_i387_struct *fpu)
-{
-	int fpvalid = !!tsk_used_math(tsk);
-
-	if (fpvalid) {
-		if (tsk == current)
-			unlazy_fpu(tsk);
-		memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(struct user_i387_struct)); 	
-}
-	return fpvalid;
-}
diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index b2bc0050ce99..de435b9114df 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -1 +1,374 @@
-#include "i387_64.h"
+/*
+ * Copyright (C) 1994 Linus Torvalds
+ *
+ * Pentium III FXSR, SSE support
+ * General FPU state handling cleanups
+ *	Gareth Hughes <gareth@valinux.com>, May 2000
+ * x86-64 work by Andi Kleen 2002
+ */
+
+#ifndef _ASM_X86_I387_H
+#define _ASM_X86_I387_H
+
+#include <linux/sched.h>
+#include <linux/kernel_stat.h>
+#include <linux/regset.h>
+#include <asm/processor.h>
+#include <asm/sigcontext.h>
+#include <asm/user.h>
+#include <asm/uaccess.h>
+
+extern void fpu_init(void);
+extern unsigned int mxcsr_feature_mask;
+extern void mxcsr_feature_mask_init(void);
+extern void init_fpu(struct task_struct *child);
+extern asmlinkage void math_state_restore(void);
+
+extern user_regset_active_fn fpregs_active, xfpregs_active;
+extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
+extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
+
+#ifdef CONFIG_IA32_EMULATION
+struct _fpstate_ia32;
+extern int save_i387_ia32(struct _fpstate_ia32 __user *buf);
+extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf);
+#endif
+
+#ifdef CONFIG_X86_64
+
+/* Ignore delayed exceptions from user space */
+static inline void tolerant_fwait(void)
+{
+	asm volatile("1: fwait\n"
+		     "2:\n"
+		     "   .section __ex_table,\"a\"\n"
+		     "	.align 8\n"
+		     "	.quad 1b,2b\n"
+		     "	.previous\n");
+}
+
+static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
+{
+	int err;
+
+	asm volatile("1:  rex64/fxrstor (%[fx])\n\t"
+		     "2:\n"
+		     ".section .fixup,\"ax\"\n"
+		     "3:  movl $-1,%[err]\n"
+		     "    jmp  2b\n"
+		     ".previous\n"
+		     ".section __ex_table,\"a\"\n"
+		     "   .align 8\n"
+		     "   .quad  1b,3b\n"
+		     ".previous"
+		     : [err] "=r" (err)
+#if 0 /* See comment in __save_init_fpu() below. */
+		     : [fx] "r" (fx), "m" (*fx), "0" (0));
+#else
+		     : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0));
+#endif
+	if (unlikely(err))
+		init_fpu(current);
+	return err;
+}
+
+#define X87_FSW_ES (1 << 7)	/* Exception Summary */
+
+/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
+   is pending. Clear the x87 state here by setting it to fixed
+   values. The kernel data segment can be sometimes 0 and sometimes
+   new user value. Both should be ok.
+   Use the PDA as safe address because it should be already in L1. */
+static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
+{
+	if (unlikely(fx->swd & X87_FSW_ES))
+		 asm volatile("fnclex");
+	alternative_input(ASM_NOP8 ASM_NOP2,
+		     "    emms\n"		/* clear stack tags */
+		     "    fildl %%gs:0",	/* load to clear state */
+		     X86_FEATURE_FXSAVE_LEAK);
+}
+
+static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
+{
+	int err;
+
+	asm volatile("1:  rex64/fxsave (%[fx])\n\t"
+		     "2:\n"
+		     ".section .fixup,\"ax\"\n"
+		     "3:  movl $-1,%[err]\n"
+		     "    jmp  2b\n"
+		     ".previous\n"
+		     ".section __ex_table,\"a\"\n"
+		     "   .align 8\n"
+		     "   .quad  1b,3b\n"
+		     ".previous"
+		     : [err] "=r" (err), "=m" (*fx)
+#if 0 /* See comment in __fxsave_clear() below. */
+		     : [fx] "r" (fx), "0" (0));
+#else
+		     : [fx] "cdaSDb" (fx), "0" (0));
+#endif
+	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))
+		err = -EFAULT;
+	/* No need to clear here because the caller clears USED_MATH */
+	return err;
+}
+
+static inline void __save_init_fpu(struct task_struct *tsk)
+{
+	/* Using "rex64; fxsave %0" is broken because, if the memory operand
+	   uses any extended registers for addressing, a second REX prefix
+	   will be generated (to the assembler, rex64 followed by semicolon
+	   is a separate instruction), and hence the 64-bitness is lost. */
+#if 0
+	/* Using "fxsaveq %0" would be the ideal choice, but is only supported
+	   starting with gas 2.16. */
+	__asm__ __volatile__("fxsaveq %0"
+			     : "=m" (tsk->thread.i387.fxsave));
+#elif 0
+	/* Using, as a workaround, the properly prefixed form below isn't
+	   accepted by any binutils version so far released, complaining that
+	   the same type of prefix is used twice if an extended register is
+	   needed for addressing (fix submitted to mainline 2005-11-21). */
+	__asm__ __volatile__("rex64/fxsave %0"
+			     : "=m" (tsk->thread.i387.fxsave));
+#else
+	/* This, however, we can work around by forcing the compiler to select
+	   an addressing mode that doesn't require extended registers. */
+	__asm__ __volatile__("rex64/fxsave %P2(%1)"
+			     : "=m" (tsk->thread.i387.fxsave)
+			     : "cdaSDb" (tsk),
+				"i" (offsetof(__typeof__(*tsk),
+					      thread.i387.fxsave)));
+#endif
+	clear_fpu_state(&tsk->thread.i387.fxsave);
+	task_thread_info(tsk)->status &= ~TS_USEDFPU;
+}
+
+/*
+ * Signal frame handlers.
+ */
+
+static inline int save_i387(struct _fpstate __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err = 0;
+
+	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
+			sizeof(tsk->thread.i387.fxsave));
+
+	if ((unsigned long)buf % 16)
+		printk("save_i387: bad fpstate %p\n", buf);
+
+	if (!used_math())
+		return 0;
+	clear_used_math(); /* trigger finit */
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);
+		if (err) return err;
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	} else {
+		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
+				   sizeof(struct i387_fxsave_struct)))
+			return -1;
+	}
+	return 1;
+}
+
+/*
+ * This restores directly out of user space. Exceptions are handled.
+ */
+static inline int restore_i387(struct _fpstate __user *buf)
+{
+	set_used_math();
+	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
+		clts();
+		task_thread_info(current)->status |= TS_USEDFPU;
+	}
+	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
+}
+
+#else  /* CONFIG_X86_32 */
+
+static inline void tolerant_fwait(void)
+{
+	asm volatile("fnclex ; fwait");
+}
+
+static inline void restore_fpu(struct task_struct *tsk)
+{
+	/*
+	 * The "nop" is needed to make the instructions the same
+	 * length.
+	 */
+	alternative_input(
+		"nop ; frstor %1",
+		"fxrstor %1",
+		X86_FEATURE_FXSR,
+		"m" ((tsk)->thread.i387.fxsave));
+}
+
+/* We need a safe address that is cheap to find and that is already
+   in L1 during context switch. The best choices are unfortunately
+   different for UP and SMP */
+#ifdef CONFIG_SMP
+#define safe_address (__per_cpu_offset[0])
+#else
+#define safe_address (kstat_cpu(0).cpustat.user)
+#endif
+
+/*
+ * These must be called with preempt disabled
+ */
+static inline void __save_init_fpu(struct task_struct *tsk)
+{
+	/* Use more nops than strictly needed in case the compiler
+	   varies code */
+	alternative_input(
+		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
+		"fxsave %[fx]\n"
+		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
+		X86_FEATURE_FXSR,
+		[fx] "m" (tsk->thread.i387.fxsave),
+		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");
+	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
+	   is pending.  Clear the x87 state here by setting it to fixed
+	   values. safe_address is a random variable that should be in L1 */
+	alternative_input(
+		GENERIC_NOP8 GENERIC_NOP2,
+		"emms\n\t"	  	/* clear stack tags */
+		"fildl %[addr]", 	/* set F?P to defined value */
+		X86_FEATURE_FXSAVE_LEAK,
+		[addr] "m" (safe_address));
+	task_thread_info(tsk)->status &= ~TS_USEDFPU;
+}
+
+/*
+ * Signal frame handlers...
+ */
+extern int save_i387(struct _fpstate __user *buf);
+extern int restore_i387(struct _fpstate __user *buf);
+
+#endif	/* CONFIG_X86_64 */
+
+static inline void __unlazy_fpu(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		__save_init_fpu(tsk);
+		stts();
+	} else
+		tsk->fpu_counter = 0;
+}
+
+static inline void __clear_fpu(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		tolerant_fwait();
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	}
+}
+
+static inline void kernel_fpu_begin(void)
+{
+	struct thread_info *me = current_thread_info();
+	preempt_disable();
+	if (me->status & TS_USEDFPU)
+		__save_init_fpu(me->task);
+	else
+		clts();
+}
+
+static inline void kernel_fpu_end(void)
+{
+	stts();
+	preempt_enable();
+}
+
+#ifdef CONFIG_X86_64
+
+static inline void save_init_fpu(struct task_struct *tsk)
+{
+	__save_init_fpu(tsk);
+	stts();
+}
+
+#define unlazy_fpu	__unlazy_fpu
+#define clear_fpu	__clear_fpu
+
+#else  /* CONFIG_X86_32 */
+
+/*
+ * These disable preemption on their own and are safe
+ */
+static inline void save_init_fpu(struct task_struct *tsk)
+{
+	preempt_disable();
+	__save_init_fpu(tsk);
+	stts();
+	preempt_enable();
+}
+
+static inline void unlazy_fpu(struct task_struct *tsk)
+{
+	preempt_disable();
+	__unlazy_fpu(tsk);
+	preempt_enable();
+}
+
+static inline void clear_fpu(struct task_struct *tsk)
+{
+	preempt_disable();
+	__clear_fpu(tsk);
+	preempt_enable();
+}
+
+#endif	/* CONFIG_X86_64 */
+
+/*
+ * ptrace request handlers...
+ */
+extern int get_fpregs(struct user_i387_struct __user *buf,
+		      struct task_struct *tsk);
+extern int set_fpregs(struct task_struct *tsk,
+		      struct user_i387_struct __user *buf);
+
+struct user_fxsr_struct;
+extern int get_fpxregs(struct user_fxsr_struct __user *buf,
+		       struct task_struct *tsk);
+extern int set_fpxregs(struct task_struct *tsk,
+		       struct user_fxsr_struct __user *buf);
+
+/*
+ * i387 state interaction
+ */
+static inline unsigned short get_fpu_cwd(struct task_struct *tsk)
+{
+	if (cpu_has_fxsr) {
+		return tsk->thread.i387.fxsave.cwd;
+	} else {
+		return (unsigned short)tsk->thread.i387.fsave.cwd;
+	}
+}
+
+static inline unsigned short get_fpu_swd(struct task_struct *tsk)
+{
+	if (cpu_has_fxsr) {
+		return tsk->thread.i387.fxsave.swd;
+	} else {
+		return (unsigned short)tsk->thread.i387.fsave.swd;
+	}
+}
+
+static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)
+{
+	if (cpu_has_xmm) {
+		return tsk->thread.i387.fxsave.mxcsr;
+	} else {
+		return MXCSR_DEFAULT;
+	}
+}
+
+#endif	/* _ASM_X86_I387_H */
diff --git a/include/asm-x86/i387_32.h b/include/asm-x86/i387_32.h
deleted file mode 100644
index 9ac2502cdd3d..000000000000
--- a/include/asm-x86/i387_32.h
+++ /dev/null
@@ -1,149 +0,0 @@
-/*
- * Copyright (C) 1994 Linus Torvalds
- *
- * Pentium III FXSR, SSE support
- * General FPU state handling cleanups
- *	Gareth Hughes <gareth@valinux.com>, May 2000
- */
-
-#ifndef __ASM_I386_I387_H
-#define __ASM_I386_I387_H
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <linux/kernel_stat.h>
-#include <asm/processor.h>
-#include <asm/sigcontext.h>
-#include <asm/user.h>
-
-extern void mxcsr_feature_mask_init(void);
-extern void init_fpu(struct task_struct *);
-
-/*
- * FPU lazy state save handling...
- */
-
-/*
- * The "nop" is needed to make the instructions the same
- * length.
- */
-#define restore_fpu(tsk)			\
-	alternative_input(			\
-		"nop ; frstor %1",		\
-		"fxrstor %1",			\
-		X86_FEATURE_FXSR,		\
-		"m" ((tsk)->thread.i387.fxsave))
-
-extern void kernel_fpu_begin(void);
-#define kernel_fpu_end() do { stts(); preempt_enable(); } while(0)
-
-/* We need a safe address that is cheap to find and that is already
-   in L1 during context switch. The best choices are unfortunately
-   different for UP and SMP */
-#ifdef CONFIG_SMP
-#define safe_address (__per_cpu_offset[0])
-#else
-#define safe_address (kstat_cpu(0).cpustat.user)
-#endif
-
-/*
- * These must be called with preempt disabled
- */
-static inline void __save_init_fpu( struct task_struct *tsk )
-{
-	/* Use more nops than strictly needed in case the compiler
-	   varies code */
-	alternative_input(
-		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
-		"fxsave %[fx]\n"
-		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
-		X86_FEATURE_FXSR,
-		[fx] "m" (tsk->thread.i387.fxsave),
-		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");
-	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
-	   is pending.  Clear the x87 state here by setting it to fixed
-   	   values. safe_address is a random variable that should be in L1 */
-	alternative_input(
-		GENERIC_NOP8 GENERIC_NOP2,
-		"emms\n\t"	  	/* clear stack tags */
-		"fildl %[addr]", 	/* set F?P to defined value */
-		X86_FEATURE_FXSAVE_LEAK,
-		[addr] "m" (safe_address));
-	task_thread_info(tsk)->status &= ~TS_USEDFPU;
-}
-
-#define __unlazy_fpu( tsk ) do {				\
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {	\
-		__save_init_fpu(tsk);				\
-		stts();						\
-	} else							\
-		tsk->fpu_counter = 0;				\
-} while (0)
-
-#define __clear_fpu( tsk )					\
-do {								\
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {	\
-		asm volatile("fnclex ; fwait");			\
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;	\
-		stts();						\
-	}							\
-} while (0)
-
-
-/*
- * These disable preemption on their own and are safe
- */
-static inline void save_init_fpu( struct task_struct *tsk )
-{
-	preempt_disable();
-	__save_init_fpu(tsk);
-	stts();
-	preempt_enable();
-}
-
-#define unlazy_fpu( tsk ) do {	\
-	preempt_disable();	\
-	__unlazy_fpu(tsk);	\
-	preempt_enable();	\
-} while (0)
-
-#define clear_fpu( tsk ) do {	\
-	preempt_disable();	\
-	__clear_fpu( tsk );	\
-	preempt_enable();	\
-} while (0)
-
-/*
- * FPU state interaction...
- */
-extern unsigned short get_fpu_cwd( struct task_struct *tsk );
-extern unsigned short get_fpu_swd( struct task_struct *tsk );
-extern unsigned short get_fpu_mxcsr( struct task_struct *tsk );
-extern asmlinkage void math_state_restore(void);
-
-/*
- * Signal frame handlers...
- */
-extern int save_i387( struct _fpstate __user *buf );
-extern int restore_i387( struct _fpstate __user *buf );
-
-/*
- * ptrace request handers...
- */
-extern int get_fpregs( struct user_i387_struct __user *buf,
-		       struct task_struct *tsk );
-extern int set_fpregs( struct task_struct *tsk,
-		       struct user_i387_struct __user *buf );
-
-extern int get_fpxregs( struct user_fxsr_struct __user *buf,
-			struct task_struct *tsk );
-extern int set_fpxregs( struct task_struct *tsk,
-			struct user_fxsr_struct __user *buf );
-
-/*
- * FPU state for core dumps...
- */
-extern int dump_fpu( struct pt_regs *regs,
-		     struct user_i387_struct *fpu );
-
-#endif /* __ASM_I386_I387_H */
diff --git a/include/asm-x86/i387_64.h b/include/asm-x86/i387_64.h
deleted file mode 100644
index de435b9114df..000000000000
--- a/include/asm-x86/i387_64.h
+++ /dev/null
@@ -1,374 +0,0 @@
-/*
- * Copyright (C) 1994 Linus Torvalds
- *
- * Pentium III FXSR, SSE support
- * General FPU state handling cleanups
- *	Gareth Hughes <gareth@valinux.com>, May 2000
- * x86-64 work by Andi Kleen 2002
- */
-
-#ifndef _ASM_X86_I387_H
-#define _ASM_X86_I387_H
-
-#include <linux/sched.h>
-#include <linux/kernel_stat.h>
-#include <linux/regset.h>
-#include <asm/processor.h>
-#include <asm/sigcontext.h>
-#include <asm/user.h>
-#include <asm/uaccess.h>
-
-extern void fpu_init(void);
-extern unsigned int mxcsr_feature_mask;
-extern void mxcsr_feature_mask_init(void);
-extern void init_fpu(struct task_struct *child);
-extern asmlinkage void math_state_restore(void);
-
-extern user_regset_active_fn fpregs_active, xfpregs_active;
-extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
-extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
-
-#ifdef CONFIG_IA32_EMULATION
-struct _fpstate_ia32;
-extern int save_i387_ia32(struct _fpstate_ia32 __user *buf);
-extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf);
-#endif
-
-#ifdef CONFIG_X86_64
-
-/* Ignore delayed exceptions from user space */
-static inline void tolerant_fwait(void)
-{
-	asm volatile("1: fwait\n"
-		     "2:\n"
-		     "   .section __ex_table,\"a\"\n"
-		     "	.align 8\n"
-		     "	.quad 1b,2b\n"
-		     "	.previous\n");
-}
-
-static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
-{
-	int err;
-
-	asm volatile("1:  rex64/fxrstor (%[fx])\n\t"
-		     "2:\n"
-		     ".section .fixup,\"ax\"\n"
-		     "3:  movl $-1,%[err]\n"
-		     "    jmp  2b\n"
-		     ".previous\n"
-		     ".section __ex_table,\"a\"\n"
-		     "   .align 8\n"
-		     "   .quad  1b,3b\n"
-		     ".previous"
-		     : [err] "=r" (err)
-#if 0 /* See comment in __save_init_fpu() below. */
-		     : [fx] "r" (fx), "m" (*fx), "0" (0));
-#else
-		     : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0));
-#endif
-	if (unlikely(err))
-		init_fpu(current);
-	return err;
-}
-
-#define X87_FSW_ES (1 << 7)	/* Exception Summary */
-
-/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
-   is pending. Clear the x87 state here by setting it to fixed
-   values. The kernel data segment can be sometimes 0 and sometimes
-   new user value. Both should be ok.
-   Use the PDA as safe address because it should be already in L1. */
-static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
-{
-	if (unlikely(fx->swd & X87_FSW_ES))
-		 asm volatile("fnclex");
-	alternative_input(ASM_NOP8 ASM_NOP2,
-		     "    emms\n"		/* clear stack tags */
-		     "    fildl %%gs:0",	/* load to clear state */
-		     X86_FEATURE_FXSAVE_LEAK);
-}
-
-static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
-{
-	int err;
-
-	asm volatile("1:  rex64/fxsave (%[fx])\n\t"
-		     "2:\n"
-		     ".section .fixup,\"ax\"\n"
-		     "3:  movl $-1,%[err]\n"
-		     "    jmp  2b\n"
-		     ".previous\n"
-		     ".section __ex_table,\"a\"\n"
-		     "   .align 8\n"
-		     "   .quad  1b,3b\n"
-		     ".previous"
-		     : [err] "=r" (err), "=m" (*fx)
-#if 0 /* See comment in __fxsave_clear() below. */
-		     : [fx] "r" (fx), "0" (0));
-#else
-		     : [fx] "cdaSDb" (fx), "0" (0));
-#endif
-	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))
-		err = -EFAULT;
-	/* No need to clear here because the caller clears USED_MATH */
-	return err;
-}
-
-static inline void __save_init_fpu(struct task_struct *tsk)
-{
-	/* Using "rex64; fxsave %0" is broken because, if the memory operand
-	   uses any extended registers for addressing, a second REX prefix
-	   will be generated (to the assembler, rex64 followed by semicolon
-	   is a separate instruction), and hence the 64-bitness is lost. */
-#if 0
-	/* Using "fxsaveq %0" would be the ideal choice, but is only supported
-	   starting with gas 2.16. */
-	__asm__ __volatile__("fxsaveq %0"
-			     : "=m" (tsk->thread.i387.fxsave));
-#elif 0
-	/* Using, as a workaround, the properly prefixed form below isn't
-	   accepted by any binutils version so far released, complaining that
-	   the same type of prefix is used twice if an extended register is
-	   needed for addressing (fix submitted to mainline 2005-11-21). */
-	__asm__ __volatile__("rex64/fxsave %0"
-			     : "=m" (tsk->thread.i387.fxsave));
-#else
-	/* This, however, we can work around by forcing the compiler to select
-	   an addressing mode that doesn't require extended registers. */
-	__asm__ __volatile__("rex64/fxsave %P2(%1)"
-			     : "=m" (tsk->thread.i387.fxsave)
-			     : "cdaSDb" (tsk),
-				"i" (offsetof(__typeof__(*tsk),
-					      thread.i387.fxsave)));
-#endif
-	clear_fpu_state(&tsk->thread.i387.fxsave);
-	task_thread_info(tsk)->status &= ~TS_USEDFPU;
-}
-
-/*
- * Signal frame handlers.
- */
-
-static inline int save_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err = 0;
-
-	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
-			sizeof(tsk->thread.i387.fxsave));
-
-	if ((unsigned long)buf % 16)
-		printk("save_i387: bad fpstate %p\n", buf);
-
-	if (!used_math())
-		return 0;
-	clear_used_math(); /* trigger finit */
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);
-		if (err) return err;
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;
-		stts();
-	} else {
-		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
-				   sizeof(struct i387_fxsave_struct)))
-			return -1;
-	}
-	return 1;
-}
-
-/*
- * This restores directly out of user space. Exceptions are handled.
- */
-static inline int restore_i387(struct _fpstate __user *buf)
-{
-	set_used_math();
-	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
-		clts();
-		task_thread_info(current)->status |= TS_USEDFPU;
-	}
-	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
-}
-
-#else  /* CONFIG_X86_32 */
-
-static inline void tolerant_fwait(void)
-{
-	asm volatile("fnclex ; fwait");
-}
-
-static inline void restore_fpu(struct task_struct *tsk)
-{
-	/*
-	 * The "nop" is needed to make the instructions the same
-	 * length.
-	 */
-	alternative_input(
-		"nop ; frstor %1",
-		"fxrstor %1",
-		X86_FEATURE_FXSR,
-		"m" ((tsk)->thread.i387.fxsave));
-}
-
-/* We need a safe address that is cheap to find and that is already
-   in L1 during context switch. The best choices are unfortunately
-   different for UP and SMP */
-#ifdef CONFIG_SMP
-#define safe_address (__per_cpu_offset[0])
-#else
-#define safe_address (kstat_cpu(0).cpustat.user)
-#endif
-
-/*
- * These must be called with preempt disabled
- */
-static inline void __save_init_fpu(struct task_struct *tsk)
-{
-	/* Use more nops than strictly needed in case the compiler
-	   varies code */
-	alternative_input(
-		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
-		"fxsave %[fx]\n"
-		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
-		X86_FEATURE_FXSR,
-		[fx] "m" (tsk->thread.i387.fxsave),
-		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");
-	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
-	   is pending.  Clear the x87 state here by setting it to fixed
-	   values. safe_address is a random variable that should be in L1 */
-	alternative_input(
-		GENERIC_NOP8 GENERIC_NOP2,
-		"emms\n\t"	  	/* clear stack tags */
-		"fildl %[addr]", 	/* set F?P to defined value */
-		X86_FEATURE_FXSAVE_LEAK,
-		[addr] "m" (safe_address));
-	task_thread_info(tsk)->status &= ~TS_USEDFPU;
-}
-
-/*
- * Signal frame handlers...
- */
-extern int save_i387(struct _fpstate __user *buf);
-extern int restore_i387(struct _fpstate __user *buf);
-
-#endif	/* CONFIG_X86_64 */
-
-static inline void __unlazy_fpu(struct task_struct *tsk)
-{
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		__save_init_fpu(tsk);
-		stts();
-	} else
-		tsk->fpu_counter = 0;
-}
-
-static inline void __clear_fpu(struct task_struct *tsk)
-{
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		tolerant_fwait();
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;
-		stts();
-	}
-}
-
-static inline void kernel_fpu_begin(void)
-{
-	struct thread_info *me = current_thread_info();
-	preempt_disable();
-	if (me->status & TS_USEDFPU)
-		__save_init_fpu(me->task);
-	else
-		clts();
-}
-
-static inline void kernel_fpu_end(void)
-{
-	stts();
-	preempt_enable();
-}
-
-#ifdef CONFIG_X86_64
-
-static inline void save_init_fpu(struct task_struct *tsk)
-{
-	__save_init_fpu(tsk);
-	stts();
-}
-
-#define unlazy_fpu	__unlazy_fpu
-#define clear_fpu	__clear_fpu
-
-#else  /* CONFIG_X86_32 */
-
-/*
- * These disable preemption on their own and are safe
- */
-static inline void save_init_fpu(struct task_struct *tsk)
-{
-	preempt_disable();
-	__save_init_fpu(tsk);
-	stts();
-	preempt_enable();
-}
-
-static inline void unlazy_fpu(struct task_struct *tsk)
-{
-	preempt_disable();
-	__unlazy_fpu(tsk);
-	preempt_enable();
-}
-
-static inline void clear_fpu(struct task_struct *tsk)
-{
-	preempt_disable();
-	__clear_fpu(tsk);
-	preempt_enable();
-}
-
-#endif	/* CONFIG_X86_64 */
-
-/*
- * ptrace request handlers...
- */
-extern int get_fpregs(struct user_i387_struct __user *buf,
-		      struct task_struct *tsk);
-extern int set_fpregs(struct task_struct *tsk,
-		      struct user_i387_struct __user *buf);
-
-struct user_fxsr_struct;
-extern int get_fpxregs(struct user_fxsr_struct __user *buf,
-		       struct task_struct *tsk);
-extern int set_fpxregs(struct task_struct *tsk,
-		       struct user_fxsr_struct __user *buf);
-
-/*
- * i387 state interaction
- */
-static inline unsigned short get_fpu_cwd(struct task_struct *tsk)
-{
-	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.cwd;
-	} else {
-		return (unsigned short)tsk->thread.i387.fsave.cwd;
-	}
-}
-
-static inline unsigned short get_fpu_swd(struct task_struct *tsk)
-{
-	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.swd;
-	} else {
-		return (unsigned short)tsk->thread.i387.fsave.swd;
-	}
-}
-
-static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)
-{
-	if (cpu_has_xmm) {
-		return tsk->thread.i387.fxsave.mxcsr;
-	} else {
-		return MXCSR_DEFAULT;
-	}
-}
-
-#endif	/* _ASM_X86_I387_H */
-- 
cgit v1.2.3


From 1bd5718ce58fb49ac158653380fa200f4759daad Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:51 +0100
Subject: x86: x86 TLS desc_struct cleanup

This cleans up the TLS code to use struct desc_struct and to separate the
encoding and installation magic from the interface wrappers.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/tls.c  | 89 +++++++++++++++++++++++++-------------------------
 include/asm-x86/desc.h | 11 +++++--
 2 files changed, 54 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c
index 98f428be8e8c..f11c92a3faac 100644
--- a/arch/x86/kernel/tls.c
+++ b/arch/x86/kernel/tls.c
@@ -24,6 +24,29 @@ static int get_free_idx(void)
 	return -ESRCH;
 }
 
+static void set_tls_desc(struct task_struct *p, int idx,
+			 const struct user_desc *info)
+{
+	struct thread_struct *t = &p->thread;
+	struct desc_struct *desc = &t->tls_array[idx - GDT_ENTRY_TLS_MIN];
+	int cpu;
+
+	/*
+	 * We must not get preempted while modifying the TLS.
+	 */
+	cpu = get_cpu();
+
+	if (LDT_empty(info))
+		desc->a = desc->b = 0;
+	else
+		fill_ldt(desc, info);
+
+	if (t == &current->thread)
+		load_TLS(t, cpu);
+
+	put_cpu();
+}
+
 /*
  * Set a given TLS descriptor:
  */
@@ -31,10 +54,7 @@ int do_set_thread_area(struct task_struct *p, int idx,
 		       struct user_desc __user *u_info,
 		       int can_allocate)
 {
-	struct thread_struct *t = &p->thread;
 	struct user_desc info;
-	u32 *desc;
-	int cpu;
 
 	if (copy_from_user(&info, u_info, sizeof(info)))
 		return -EFAULT;
@@ -57,23 +77,8 @@ int do_set_thread_area(struct task_struct *p, int idx,
 	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
 		return -EINVAL;
 
-	desc = (u32 *) &t->tls_array[idx - GDT_ENTRY_TLS_MIN];
+	set_tls_desc(p, idx, &info);
 
-	/*
-	 * We must not get preempted while modifying the TLS.
-	 */
-	cpu = get_cpu();
-
-	if (LDT_empty(&info)) {
-		desc[0] = 0;
-		desc[1] = 0;
-	} else
-		fill_ldt((struct desc_struct *)desc, &info);
-
-	if (t == &current->thread)
-		load_TLS(t, cpu);
-
-	put_cpu();
 	return 0;
 }
 
@@ -87,42 +92,38 @@ asmlinkage int sys_set_thread_area(struct user_desc __user *u_info)
  * Get the current Thread-Local Storage area:
  */
 
-#define GET_LIMIT(desc)		(((desc)[0] & 0x0ffff) | ((desc)[1] & 0xf0000))
-#define GET_32BIT(desc)		(((desc)[1] >> 22) & 1)
-#define GET_CONTENTS(desc)	(((desc)[1] >> 10) & 3)
-#define GET_WRITABLE(desc)	(((desc)[1] >>  9) & 1)
-#define GET_LIMIT_PAGES(desc)	(((desc)[1] >> 23) & 1)
-#define GET_PRESENT(desc)	(((desc)[1] >> 15) & 1)
-#define GET_USEABLE(desc)	(((desc)[1] >> 20) & 1)
-#define GET_LONGMODE(desc)	(((desc)[1] >> 21) & 1)
+static void fill_user_desc(struct user_desc *info, int idx,
+			   const struct desc_struct *desc)
+
+{
+	memset(info, 0, sizeof(*info));
+	info->entry_number = idx;
+	info->base_addr = get_desc_base(desc);
+	info->limit = get_desc_limit(desc);
+	info->seg_32bit = desc->d;
+	info->contents = desc->type >> 2;
+	info->read_exec_only = !(desc->type & 2);
+	info->limit_in_pages = desc->g;
+	info->seg_not_present = !desc->p;
+	info->useable = desc->avl;
+#ifdef CONFIG_X86_64
+	info->lm = desc->l;
+#endif
+}
 
 int do_get_thread_area(struct task_struct *p, int idx,
 		       struct user_desc __user *u_info)
 {
-	struct thread_struct *t = &p->thread;
 	struct user_desc info;
-	u32 *desc;
 
 	if (idx == -1 && get_user(idx, &u_info->entry_number))
 		return -EFAULT;
+
 	if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
 		return -EINVAL;
 
-	desc = (u32 *) &t->tls_array[idx - GDT_ENTRY_TLS_MIN];
-
-	memset(&info, 0, sizeof(struct user_desc));
-	info.entry_number = idx;
-	info.base_addr = get_desc_base((struct desc_struct *)desc);
-	info.limit = GET_LIMIT(desc);
-	info.seg_32bit = GET_32BIT(desc);
-	info.contents = GET_CONTENTS(desc);
-	info.read_exec_only = !GET_WRITABLE(desc);
-	info.limit_in_pages = GET_LIMIT_PAGES(desc);
-	info.seg_not_present = !GET_PRESENT(desc);
-	info.useable = GET_USEABLE(desc);
-#ifdef CONFIG_X86_64
-	info.lm = GET_LONGMODE(desc);
-#endif
+	fill_user_desc(&info, idx,
+		       &p->thread.tls_array[idx - GDT_ENTRY_TLS_MIN]);
 
 	if (copy_to_user(u_info, &info, sizeof(info)))
 		return -EFAULT;
diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index e17581ad8824..5b6a05d3a771 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -7,7 +7,8 @@
 #include <asm/mmu.h>
 #include <linux/smp.h>
 
-static inline void fill_ldt(struct desc_struct *desc, struct user_desc *info)
+static inline void fill_ldt(struct desc_struct *desc,
+			    const struct user_desc *info)
 {
 	desc->limit0 = info->limit & 0x0ffff;
 	desc->base0 = info->base_addr & 0x0000ffff;
@@ -275,10 +276,16 @@ static inline void load_LDT(mm_context_t *pc)
 	preempt_enable();
 }
 
-static inline unsigned long get_desc_base(struct desc_struct *desc)
+static inline unsigned long get_desc_base(const struct desc_struct *desc)
 {
 	return desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24);
 }
+
+static inline unsigned long get_desc_limit(const struct desc_struct *desc)
+{
+	return desc->limit0 | (desc->limit << 16);
+}
+
 static inline void _set_gate(int gate, unsigned type, void *addr,
 			      unsigned dpl, unsigned ist, unsigned seg)
 {
-- 
cgit v1.2.3


From 975511be74d1b82bf041aaed2861a242abfe3886 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:54 +0100
Subject: x86: x86 CORE_DUMP_USE_REGSET

This switches x86 to the user_regset-based code for ELF core dumps.
The core dumps come out exactly the same as before.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/elf.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index 5e5705bf082a..123e2d599c6d 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -201,6 +201,7 @@ extern int vdso_enabled;
 
 #endif /* !CONFIG_X86_32 */
 
+#define CORE_DUMP_USE_REGSET
 #define USE_ELF_CORE_DUMP
 #define ELF_EXEC_PAGESIZE	4096
 
-- 
cgit v1.2.3


From 60b3b9af35aad66345e395be911e46fb8443f0c5 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:55 +0100
Subject: x86: x86 user_regset cleanup

This removes a bunch of dead code that is no longer needed now
that the user_regset interfaces are being used for all these jobs.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/i387.c        | 105 ++++--------------------------------------
 arch/x86/kernel/process_32.c  |  16 -------
 arch/x86/kernel/process_64.c  |  19 --------
 arch/x86/math-emu/fpu_entry.c |  14 ------
 include/asm-x86/elf.h         |  68 ---------------------------
 include/asm-x86/i387.h        |  14 ------
 include/asm-x86/math_emu.h    |   5 --
 7 files changed, 9 insertions(+), 232 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index f7f7568dd7bc..26719bd2c77c 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -454,113 +454,26 @@ int restore_i387_ia32(struct _fpstate_ia32 __user *buf)
 	return err;
 }
 
-#endif	/* CONFIG_X86_32 || CONFIG_IA32_EMULATION */
-
-#ifdef CONFIG_X86_64
-
-int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
-{
-	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
-{
-	return xfpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-#else
-
-int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
-{
-	return fpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
-{
-	return fpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *tsk)
-{
-	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int set_fpxregs(struct task_struct *tsk, struct user_fxsr_struct __user *buf)
-{
-	return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-#endif
-
 /*
  * FPU state for core dumps.
+ * This is only used for a.out dumps now.
+ * It is declared generically using elf_fpregset_t (which is
+ * struct user_i387_struct) but is in fact only used for 32-bit
+ * dumps, so on 64-bit it is really struct user_i387_ia32_struct.
  */
-
-static inline void copy_fpu_fsave(struct task_struct *tsk,
-				  struct user_i387_struct *fpu)
-{
-	memcpy(fpu, &tsk->thread.i387.fsave,
-	       sizeof(struct user_i387_struct));
-}
-
-static inline void copy_fpu_fxsave(struct task_struct *tsk,
-				   struct user_i387_struct *fpu)
-{
-	unsigned short *to;
-	unsigned short *from;
-	int i;
-
-	memcpy(fpu, &tsk->thread.i387.fxsave, 7 * sizeof(long));
-
-	to = (unsigned short *)&fpu->st_space[0];
-	from = (unsigned short *)&tsk->thread.i387.fxsave.st_space[0];
-	for (i = 0; i < 8; i++, to += 5, from += 8)
-		memcpy(to, from, 5 * sizeof(unsigned short));
-}
-
 int dump_fpu(struct pt_regs *regs, struct user_i387_struct *fpu)
 {
 	int fpvalid;
 	struct task_struct *tsk = current;
 
 	fpvalid = !!used_math();
-	if (fpvalid) {
-		unlazy_fpu(tsk);
-		if (cpu_has_fxsr) {
-			copy_fpu_fxsave(tsk, fpu);
-		} else {
-			copy_fpu_fsave(tsk, fpu);
-		}
-	}
+	if (fpvalid)
+		fpvalid = !fpregs_get(tsk, NULL,
+				      0, sizeof(struct user_i387_ia32_struct),
+				      fpu, NULL);
 
 	return fpvalid;
 }
 EXPORT_SYMBOL(dump_fpu);
 
-int dump_task_fpu(struct task_struct *tsk, struct user_i387_struct *fpu)
-{
-	int fpvalid = !!tsk_used_math(tsk);
-
-	if (fpvalid) {
-		if (tsk == current)
-			unlazy_fpu(tsk);
-		if (cpu_has_fxsr)
-			copy_fpu_fxsave(tsk, fpu);
-		else
-			copy_fpu_fsave(tsk, fpu);
-	}
-	return fpvalid;
-}
-
-int dump_task_extended_fpu(struct task_struct *tsk,
-			   struct user32_fxsr_struct *fpu)
-{
-	int fpvalid = tsk_used_math(tsk) && cpu_has_fxsr;
-
-	if (fpvalid) {
-		if (tsk == current)
-		       unlazy_fpu(tsk);
-		memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(*fpu));
-	}
-	return fpvalid;
-}
+#endif	/* CONFIG_X86_32 || CONFIG_IA32_EMULATION */
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 40cc29695eba..35a6f318c541 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -571,22 +571,6 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 }
 EXPORT_SYMBOL(dump_thread);
 
-/*
- * Capture the user space registers if the task is not running (in user space)
- */
-int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
-{
-	struct pt_regs ptregs = *task_pt_regs(tsk);
-	ptregs.cs &= 0xffff;
-	ptregs.ds &= 0xffff;
-	ptregs.es &= 0xffff;
-	ptregs.ss &= 0xffff;
-
-	elf_core_copy_regs(regs, &ptregs);
-
-	return 1;
-}
-
 #ifdef CONFIG_SECCOMP
 static void hard_disable_TSC(void)
 {
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index e3a3610ade10..78d80067b7f9 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -544,24 +544,6 @@ out:
  */
 #define loaddebug(thread, r) set_debugreg(thread->debugreg ## r, r)
 
-/*
- * Capture the user space registers if the task is not running (in user space)
- */
-int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
-{
-	struct pt_regs *pp, ptregs;
-
-	pp = task_pt_regs(tsk);
-
-	ptregs = *pp;
-	ptregs.cs &= 0xffff;
-	ptregs.ss &= 0xffff;
-
-	elf_core_copy_regs(regs, &ptregs);
-
-	return 1;
-}
-
 static inline void __switch_to_xtra(struct task_struct *prev_p,
 				    struct task_struct *next_p,
 				    struct tss_struct *tss)
@@ -929,4 +911,3 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
 	unsigned long range_end = mm->brk + 0x02000000;
 	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
 }
-
diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c
index cfbdaa1532ce..760baeea5f07 100644
--- a/arch/x86/math-emu/fpu_entry.c
+++ b/arch/x86/math-emu/fpu_entry.c
@@ -761,17 +761,3 @@ int fpregs_soft_get(struct task_struct *target,
 
 	return ret;
 }
-
-int save_i387_soft(void *s387, struct _fpstate __user *buf)
-{
-	return fpregs_soft_get(current, NULL,
-			       0, sizeof(struct user_i387_struct),
-			       NULL, buf) ? -1 : 1;
-}
-
-int restore_i387_soft(void *s387, struct _fpstate __user *buf)
-{
-	return fpregs_soft_set(current, NULL,
-			       0, sizeof(struct user_i387_struct),
-			       NULL, buf) ? -1 : 1;
-}
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index 123e2d599c6d..d6bf7421d7b0 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -104,28 +104,6 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 	_r->ax = 0; \
 } while (0)
 
-/* regs is struct pt_regs, pr_reg is elf_gregset_t (which is
-   now struct_user_regs, they are different) */
-
-#define ELF_CORE_COPY_REGS(pr_reg, regs)		\
-	pr_reg[0] = regs->bx;				\
-	pr_reg[1] = regs->cx;				\
-	pr_reg[2] = regs->dx;				\
-	pr_reg[3] = regs->si;				\
-	pr_reg[4] = regs->di;				\
-	pr_reg[5] = regs->bp;				\
-	pr_reg[6] = regs->ax;				\
-	pr_reg[7] = regs->ds & 0xffff;			\
-	pr_reg[8] = regs->es & 0xffff;			\
-	pr_reg[9] = regs->fs & 0xffff;			\
-	savesegment(gs,pr_reg[10]);			\
-	pr_reg[11] = regs->orig_ax;			\
-	pr_reg[12] = regs->ip;				\
-	pr_reg[13] = regs->cs & 0xffff;			\
-	pr_reg[14] = regs->flags;			\
-	pr_reg[15] = regs->sp;				\
-	pr_reg[16] = regs->ss & 0xffff;
-
 #define ELF_PLATFORM	(utsname()->machine)
 #define set_personality_64bit()	do { } while (0)
 extern unsigned int vdso_enabled;
@@ -159,41 +137,6 @@ extern unsigned int vdso_enabled;
 	clear_thread_flag(TIF_IA32);			  \
 } while (0)
 
-/* regs is struct pt_regs, pr_reg is elf_gregset_t (which is
-   now struct_user_regs, they are different). Assumes current is the process
-   getting dumped. */
-
-#define ELF_CORE_COPY_REGS(pr_reg, regs)  do {			\
-	unsigned v;						\
-	(pr_reg)[0] = (regs)->r15;				\
-	(pr_reg)[1] = (regs)->r14;				\
-	(pr_reg)[2] = (regs)->r13;				\
-	(pr_reg)[3] = (regs)->r12;				\
-	(pr_reg)[4] = (regs)->bp;				\
-	(pr_reg)[5] = (regs)->bx;				\
-	(pr_reg)[6] = (regs)->r11;				\
-	(pr_reg)[7] = (regs)->r10;				\
-	(pr_reg)[8] = (regs)->r9;				\
-	(pr_reg)[9] = (regs)->r8;				\
-	(pr_reg)[10] = (regs)->ax;				\
-	(pr_reg)[11] = (regs)->cx;				\
-	(pr_reg)[12] = (regs)->dx;				\
-	(pr_reg)[13] = (regs)->si;				\
-	(pr_reg)[14] = (regs)->di;				\
-	(pr_reg)[15] = (regs)->orig_ax;			\
-	(pr_reg)[16] = (regs)->ip;				\
-	(pr_reg)[17] = (regs)->cs;				\
-	(pr_reg)[18] = (regs)->flags;				\
-	(pr_reg)[19] = (regs)->sp;				\
-	(pr_reg)[20] = (regs)->ss;				\
-	(pr_reg)[21] = current->thread.fs;			\
-	(pr_reg)[22] = current->thread.gs;			\
-	asm("movl %%ds,%0" : "=r" (v)); (pr_reg)[23] = v;	\
-	asm("movl %%es,%0" : "=r" (v)); (pr_reg)[24] = v;	\
-	asm("movl %%fs,%0" : "=r" (v)); (pr_reg)[25] = v;	\
-	asm("movl %%gs,%0" : "=r" (v)); (pr_reg)[26] = v;	\
-} while(0);
-
 /* I'm not sure if we can use '-' here */
 #define ELF_PLATFORM       ("x86_64")
 extern void set_personality_64bit(void);
@@ -236,18 +179,7 @@ extern int vdso_enabled;
 
 struct task_struct;
 
-extern int dump_task_regs (struct task_struct *, elf_gregset_t *);
-extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *);
-
-#define ELF_CORE_COPY_TASK_REGS(tsk, elf_regs) dump_task_regs(tsk, elf_regs)
-#define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
-
 #ifdef CONFIG_X86_32
-extern int dump_task_extended_fpu (struct task_struct *,
-				   struct user_fxsr_struct *);
-#define ELF_CORE_COPY_XFPREGS(tsk, elf_xfpregs) \
-	dump_task_extended_fpu(tsk, elf_xfpregs)
-#define ELF_CORE_XFPREG_TYPE NT_PRXFPREG
 
 #define VDSO_HIGH_BASE		(__fix_to_virt(FIX_VDSO))
 
diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index de435b9114df..ba8105ca822b 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -327,20 +327,6 @@ static inline void clear_fpu(struct task_struct *tsk)
 
 #endif	/* CONFIG_X86_64 */
 
-/*
- * ptrace request handlers...
- */
-extern int get_fpregs(struct user_i387_struct __user *buf,
-		      struct task_struct *tsk);
-extern int set_fpregs(struct task_struct *tsk,
-		      struct user_i387_struct __user *buf);
-
-struct user_fxsr_struct;
-extern int get_fpxregs(struct user_fxsr_struct __user *buf,
-		       struct task_struct *tsk);
-extern int set_fpxregs(struct task_struct *tsk,
-		       struct user_fxsr_struct __user *buf);
-
 /*
  * i387 state interaction
  */
diff --git a/include/asm-x86/math_emu.h b/include/asm-x86/math_emu.h
index a4b0aa3320e6..9bf4ae93ab10 100644
--- a/include/asm-x86/math_emu.h
+++ b/include/asm-x86/math_emu.h
@@ -1,11 +1,6 @@
 #ifndef _I386_MATH_EMU_H
 #define _I386_MATH_EMU_H
 
-#include <asm/sigcontext.h>
-
-int restore_i387_soft(void *s387, struct _fpstate __user *buf);
-int save_i387_soft(void *s387, struct _fpstate __user *buf);
-
 /* This structure matches the layout of the data saved to the stack
    following a device-not-present interrupt, part of it saved
    automatically by the 80386/80486.
-- 
cgit v1.2.3


From a97f52e67890fda6b373c1c1895ff1c1c69b36c8 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:55 +0100
Subject: x86: compat_binfmt_elf

This switches x86-64's 32-bit ELF support to use the shared
fs/compat_binfmt_elf.c code instead of our own ia32_binfmt.c.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig             |  1 +
 arch/x86/ia32/Makefile       |  3 +-
 arch/x86/vdso/vdso32-setup.c | 33 +++++++++++++++++
 include/asm-x86/elf.h        | 88 ++++++++++++++++++++++++++++++++------------
 4 files changed, 99 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 1992b8fe6a2f..d89b94528153 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1532,6 +1532,7 @@ source "fs/Kconfig.binfmt"
 config IA32_EMULATION
 	bool "IA32 Emulation"
 	depends on X86_64
+	select COMPAT_BINFMT_ELF
 	help
 	  Include code to run 32-bit programs under a 64-bit kernel. You should
 	  likely turn this on, unless you're 100% sure that you don't have any
diff --git a/arch/x86/ia32/Makefile b/arch/x86/ia32/Makefile
index 93a6fda65f49..52d0ccfcf6ea 100644
--- a/arch/x86/ia32/Makefile
+++ b/arch/x86/ia32/Makefile
@@ -2,8 +2,7 @@
 # Makefile for the ia32 kernel emulation subsystem.
 #
 
-obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o \
-	ia32_binfmt.o
+obj-$(CONFIG_IA32_EMULATION) := ia32entry.o sys_ia32.o ia32_signal.o
 
 sysv-$(CONFIG_SYSVIPC) := ipc32.o
 obj-$(CONFIG_IA32_EMULATION) += $(sysv-y)
diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c
index e0feb66a2408..348f1341e1c8 100644
--- a/arch/x86/vdso/vdso32-setup.c
+++ b/arch/x86/vdso/vdso32-setup.c
@@ -377,6 +377,39 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
 
 __initcall(sysenter_setup);
 
+#ifdef CONFIG_SYSCTL
+/* Register vsyscall32 into the ABI table */
+#include <linux/sysctl.h>
+
+static ctl_table abi_table2[] = {
+	{
+		.procname	= "vsyscall32",
+		.data		= &sysctl_vsyscall32,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{}
+};
+
+static ctl_table abi_root_table2[] = {
+	{
+		.ctl_name = CTL_ABI,
+		.procname = "abi",
+		.mode = 0555,
+		.child = abi_table2
+	},
+	{}
+};
+
+static __init int ia32_binfmt_init(void)
+{
+	register_sysctl_table(abi_root_table2);
+	return 0;
+}
+__initcall(ia32_binfmt_init);
+#endif
+
 #else  /* CONFIG_X86_32 */
 
 const char *arch_vma_name(struct vm_area_struct *vma)
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index d6bf7421d7b0..dab4744f9586 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -73,6 +73,9 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 #endif
 
 #ifdef __KERNEL__
+#include <asm/vdso.h>
+
+extern unsigned int vdso_enabled;
 
 /*
  * This is used to ensure we don't load something for the wrong architecture.
@@ -84,7 +87,6 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 #include <asm/processor.h>
 #include <asm/system.h>		/* for savesegment */
 #include <asm/desc.h>
-#include <asm/vdso.h>
 
 #define elf_check_arch(x)	elf_check_arch_ia32(x)
 
@@ -106,7 +108,6 @@ typedef struct user_fxsr_struct elf_fpxregset_t;
 
 #define ELF_PLATFORM	(utsname()->machine)
 #define set_personality_64bit()	do { } while (0)
-extern unsigned int vdso_enabled;
 
 #else /* CONFIG_X86_32 */
 
@@ -118,29 +119,57 @@ extern unsigned int vdso_enabled;
 #define elf_check_arch(x) \
 	((x)->e_machine == EM_X86_64)
 
+#define compat_elf_check_arch(x)	elf_check_arch_ia32(x)
+
+static inline void start_ia32_thread(struct pt_regs *regs, u32 ip, u32 sp)
+{
+	asm volatile("movl %0,%%fs" :: "r" (0));
+	asm volatile("movl %0,%%es; movl %0,%%ds" : : "r" (__USER32_DS));
+	load_gs_index(0);
+	regs->ip = ip;
+	regs->sp = sp;
+	regs->flags = X86_EFLAGS_IF;
+	regs->cs = __USER32_CS;
+	regs->ss = __USER32_DS;
+}
+
+static inline void elf_common_init(struct thread_struct *t,
+				   struct pt_regs *regs, const u16 ds)
+{
+	regs->ax = regs->bx = regs->cx = regs->dx = 0;
+	regs->si = regs->di = regs->bp = 0;
+	regs->r8 = regs->r9 = regs->r10 = regs->r11 = 0;
+	regs->r12 = regs->r13 = regs->r14 = regs->r15 = 0;
+	t->fs = t->gs = 0;
+	t->fsindex = t->gsindex = 0;
+	t->ds = t->es = ds;
+}
+
 #define ELF_PLAT_INIT(_r, load_addr)	do {		  \
-	struct task_struct *cur = current;		  \
-	(_r)->bx = 0; (_r)->cx = 0; (_r)->dx = 0;	  \
-	(_r)->si = 0; (_r)->di = 0; (_r)->bp = 0;	  \
-	(_r)->ax = 0;					  \
-	(_r)->r8 = 0;					  \
-	(_r)->r9 = 0;					  \
-	(_r)->r10 = 0;					  \
-	(_r)->r11 = 0;					  \
-	(_r)->r12 = 0;					  \
-	(_r)->r13 = 0;					  \
-	(_r)->r14 = 0;					  \
-	(_r)->r15 = 0;					  \
-	cur->thread.fs = 0; cur->thread.gs = 0;		  \
-	cur->thread.fsindex = 0; cur->thread.gsindex = 0; \
-	cur->thread.ds = 0; cur->thread.es = 0;		  \
+	elf_common_init(&current->thread, _r, 0);	  \
 	clear_thread_flag(TIF_IA32);			  \
 } while (0)
 
+#define	COMPAT_ELF_PLAT_INIT(regs, load_addr)	\
+	elf_common_init(&current->thread, regs, __USER_DS)
+#define	compat_start_thread(regs, ip, sp)	do {		\
+		start_ia32_thread(regs, ip, sp);		\
+		set_fs(USER_DS);				\
+	} while (0)
+#define COMPAT_SET_PERSONALITY(ex, ibcs2)	do {		\
+		if (test_thread_flag(TIF_IA32))			\
+			clear_thread_flag(TIF_ABI_PENDING);	\
+		else						\
+			set_thread_flag(TIF_ABI_PENDING);	\
+		current->personality |= force_personality32;	\
+	} while (0)
+#define COMPAT_ELF_PLATFORM			("i686")
+
 /* I'm not sure if we can use '-' here */
 #define ELF_PLATFORM       ("x86_64")
 extern void set_personality_64bit(void);
-extern int vdso_enabled;
+extern unsigned int sysctl_vsyscall32;
+extern int force_personality32;
 
 #endif /* !CONFIG_X86_32 */
 
@@ -179,17 +208,19 @@ extern int vdso_enabled;
 
 struct task_struct;
 
+#define	ARCH_DLINFO_IA32(vdso_enabled) \
+do if (vdso_enabled) {							\
+		NEW_AUX_ENT(AT_SYSINFO,	VDSO_ENTRY);			\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_CURRENT_BASE);	\
+} while (0)
+
 #ifdef CONFIG_X86_32
 
 #define VDSO_HIGH_BASE		(__fix_to_virt(FIX_VDSO))
 
-/* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
+#define ARCH_DLINFO		ARCH_DLINFO_IA32(vdso_enabled)
 
-#define ARCH_DLINFO \
-do if (vdso_enabled) {							\
-		NEW_AUX_ENT(AT_SYSINFO,	VDSO_ENTRY);			\
-		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_CURRENT_BASE);	\
-} while (0)
+/* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
 
 #else /* CONFIG_X86_32 */
 
@@ -203,6 +234,12 @@ do if (vdso_enabled) {						\
 	NEW_AUX_ENT(AT_SYSINFO_EHDR,(unsigned long)current->mm->context.vdso);\
 } while (0)
 
+#define AT_SYSINFO		32
+
+#define COMPAT_ARCH_DLINFO	ARCH_DLINFO_IA32(sysctl_vsyscall32)
+
+#define COMPAT_ELF_ET_DYN_BASE	(TASK_UNMAPPED_BASE + 0x1000000)
+
 #endif /* !CONFIG_X86_32 */
 
 #define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
@@ -216,6 +253,9 @@ struct linux_binprm;
 extern int arch_setup_additional_pages(struct linux_binprm *bprm,
 				       int executable_stack);
 
+extern int syscall32_setup_pages(struct linux_binprm *, int exstack);
+#define compat_arch_setup_additional_pages	syscall32_setup_pages
+
 extern unsigned long arch_randomize_brk(struct mm_struct *mm);
 #define arch_randomize_brk arch_randomize_brk
 
-- 
cgit v1.2.3


From bb61682b3f31dec7d058cae2f6edd2275248a704 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:31:56 +0100
Subject: x86: x86 core dump TLS

This makes ELF core dumps of 32-bit processes include a new
note type NT_386_TLS (0x200) giving the contents of the TLS
slots in struct user_desc format.  This lets post mortem
examination figure out what the segment registers mean like
the debugger does with get_thread_area on a live process.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ptrace.c | 1 +
 include/linux/elf.h      | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index f8b89059e6ed..e6a680c7daf7 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -1313,6 +1313,7 @@ static const struct user_regset x86_32_regsets[] = {
 		.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
 	},
 	[REGSET_TLS] = {
+		.core_note_type = NT_386_TLS,
 		.n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN,
 		.size = sizeof(struct user_desc),
 		.align = sizeof(struct user_desc),
diff --git a/include/linux/elf.h b/include/linux/elf.h
index 576e83bd6d88..7ceb24d87c1a 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -355,6 +355,7 @@ typedef struct elf64_shdr {
 #define NT_AUXV		6
 #define NT_PRXFPREG     0x46e62b7f      /* copied from gdb5.1/include/elf/common.h */
 #define NT_PPC_VMX	0x100		/* PowerPC Altivec/VMX registers */
+#define NT_386_TLS	0x200		/* i386 TLS slots (struct user_desc) */
 
 
 /* Note header in a PT_NOTE section */
-- 
cgit v1.2.3


From 7ba65c7e17a292fe1e6e48fd38d1b1ecb626b586 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:57 +0100
Subject: x86: remove __init modifier from header declaration

This patch removes the __init modifier from an extern function
declaration in acpi.h.

Besides not being strictly needed, it requires the inclusion of
linux/init.h, which is usually not even included directly, increasing
header mess by a lot.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/acpi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h
index c477e5741751..2feb0c494be7 100644
--- a/include/asm-x86/acpi.h
+++ b/include/asm-x86/acpi.h
@@ -158,7 +158,7 @@ extern int acpi_scan_nodes(unsigned long start, unsigned long end);
 #ifdef CONFIG_X86_64
 # define NR_NODE_MEMBLKS (MAX_NUMNODES*2)
 #endif
-extern void __init acpi_fake_nodes(const struct bootnode *fake_nodes,
+extern void acpi_fake_nodes(const struct bootnode *fake_nodes,
 				   int num_nodes);
 #else
 static inline void acpi_fake_nodes(const struct bootnode *fake_nodes,
-- 
cgit v1.2.3


From 2f66dcc933f012cd487c5ebf5d400e50b1d5c488 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:31:57 +0100
Subject: x86: finish processor.h integration

What's left in processor_32.h and processor_64.h cannot be cleanly
integrated. However, it's just a couple of definitions. They are moved
to processor.h around ifdefs, and the original files are deleted. Note that
there's much less headers included in the final version.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor.h | 140 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 137 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index ea222cfe7b00..e6fa06fee72a 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -7,12 +7,24 @@
 struct task_struct;
 struct mm_struct;
 
+#include <asm/vm86.h>
+#include <asm/math_emu.h>
+#include <asm/segment.h>
 #include <asm/page.h>
-#include <asm/percpu.h>
+#include <asm/types.h>
+#include <asm/sigcontext.h>
+#include <asm/current.h>
+#include <asm/cpufeature.h>
 #include <asm/system.h>
+#include <asm/page.h>
 #include <asm/percpu.h>
+#include <asm/msr.h>
+#include <asm/desc_defs.h>
+#include <linux/personality.h>
 #include <linux/cpumask.h>
 #include <linux/cache.h>
+#include <linux/threads.h>
+#include <linux/init.h>
 
 /*
  * Default implementation of macro that returns current
@@ -285,9 +297,13 @@ union i387_union {
 };
 
 #ifdef CONFIG_X86_32
-# include "processor_32.h"
+/*
+ * the following now lives in the per cpu area:
+ * extern	int cpu_llc_id[NR_CPUS];
+ */
+DECLARE_PER_CPU(u8, cpu_llc_id);
 #else
-# include "processor_64.h"
+DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #endif
 
 extern void print_cpu_info(struct cpuinfo_x86 *);
@@ -770,6 +786,124 @@ static inline void prefetchw(const void *x)
 }
 
 #define spin_lock_prefetch(x)	prefetchw(x)
+#ifdef CONFIG_X86_32
+/*
+ * User space process size: 3GB (default).
+ */
+#define TASK_SIZE	(PAGE_OFFSET)
+
+#define INIT_THREAD  {							\
+	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
+	.vm86_info = NULL,						\
+	.sysenter_cs = __KERNEL_CS,					\
+	.io_bitmap_ptr = NULL,						\
+	.fs = __KERNEL_PERCPU,						\
+}
+
+/*
+ * Note that the .io_bitmap member must be extra-big. This is because
+ * the CPU will access an additional byte beyond the end of the IO
+ * permission bitmap. The extra byte must be all 1 bits, and must
+ * be within the limit.
+ */
+#define INIT_TSS  {							\
+	.x86_tss = {							\
+		.sp0		= sizeof(init_stack) + (long)&init_stack, \
+		.ss0		= __KERNEL_DS,				\
+		.ss1		= __KERNEL_CS,				\
+		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		\
+	 },								\
+	.io_bitmap	= { [0 ... IO_BITMAP_LONGS] = ~0 },		\
+}
+
+#define start_thread(regs, new_eip, new_esp) do {		\
+	__asm__("movl %0,%%gs": :"r" (0));			\
+	regs->fs = 0;						\
+	set_fs(USER_DS);					\
+	regs->ds = __USER_DS;					\
+	regs->es = __USER_DS;					\
+	regs->ss = __USER_DS;					\
+	regs->cs = __USER_CS;					\
+	regs->ip = new_eip;					\
+	regs->sp = new_esp;					\
+} while (0)
+
+
+extern unsigned long thread_saved_pc(struct task_struct *tsk);
+
+#define THREAD_SIZE_LONGS      (THREAD_SIZE/sizeof(unsigned long))
+#define KSTK_TOP(info)                                                 \
+({                                                                     \
+       unsigned long *__ptr = (unsigned long *)(info);                 \
+       (unsigned long)(&__ptr[THREAD_SIZE_LONGS]);                     \
+})
+
+/*
+ * The below -8 is to reserve 8 bytes on top of the ring0 stack.
+ * This is necessary to guarantee that the entire "struct pt_regs"
+ * is accessable even if the CPU haven't stored the SS/ESP registers
+ * on the stack (interrupt gate does not save these registers
+ * when switching to the same priv ring).
+ * Therefore beware: accessing the ss/esp fields of the
+ * "struct pt_regs" is possible, but they may contain the
+ * completely wrong values.
+ */
+#define task_pt_regs(task)                                             \
+({                                                                     \
+       struct pt_regs *__regs__;                                       \
+       __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \
+       __regs__ - 1;                                                   \
+})
+
+#define KSTK_ESP(task) (task_pt_regs(task)->sp)
+
+#else
+/*
+ * User space process size. 47bits minus one guard page.
+ */
+#define TASK_SIZE64	(0x800000000000UL - 4096)
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \
+			   0xc0000000 : 0xFFFFe000)
+
+#define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? \
+				 IA32_PAGE_OFFSET : TASK_SIZE64)
+#define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? \
+				  IA32_PAGE_OFFSET : TASK_SIZE64)
+
+#define INIT_THREAD  { \
+	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+}
+
+#define INIT_TSS  { \
+	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+}
+
+#define start_thread(regs, new_rip, new_rsp) do { 			     \
+	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));  \
+	load_gs_index(0);						     \
+	(regs)->ip = (new_rip);						     \
+	(regs)->sp = (new_rsp);						     \
+	write_pda(oldrsp, (new_rsp));					     \
+	(regs)->cs = __USER_CS;						     \
+	(regs)->ss = __USER_DS;						     \
+	(regs)->flags = 0x200;						     \
+	set_fs(USER_DS);						     \
+} while (0)
+
+/*
+ * Return saved PC of a blocked thread.
+ * What is this good for? it will be always the scheduler or ret_from_fork.
+ */
+#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8))
+
+#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
+#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
+#endif /* CONFIG_X86_64 */
+
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
  */
-- 
cgit v1.2.3


From aca46ba29298810b329518b96f97ace985027b59 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:31:58 +0100
Subject: x86: remove unused include/asm-x86/processor_32/64.h

remove unused include/asm-x86/processor_32/64.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/processor_32.h | 97 ------------------------------------------
 include/asm-x86/processor_64.h | 66 ----------------------------
 2 files changed, 163 deletions(-)
 delete mode 100644 include/asm-x86/processor_32.h
 delete mode 100644 include/asm-x86/processor_64.h

(limited to 'include')

diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h
deleted file mode 100644
index 0d6a430b2bc3..000000000000
--- a/include/asm-x86/processor_32.h
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
- * Copyright (C) 1994 Linus Torvalds
- */
-
-#ifndef __ASM_I386_PROCESSOR_H
-#define __ASM_I386_PROCESSOR_H
-
-#include <asm/vm86.h>
-#include <asm/math_emu.h>
-#include <asm/segment.h>
-#include <asm/page.h>
-#include <asm/types.h>
-#include <asm/sigcontext.h>
-#include <asm/cpufeature.h>
-#include <asm/msr.h>
-#include <asm/system.h>
-#include <linux/threads.h>
-#include <linux/init.h>
-#include <asm/desc_defs.h>
-
-/*
- * the following now lives in the per cpu area:
- * extern	int cpu_llc_id[NR_CPUS];
- */
-DECLARE_PER_CPU(u8, cpu_llc_id);
-
-/*
- * User space process size: 3GB (default).
- */
-#define TASK_SIZE	(PAGE_OFFSET)
-
-#define INIT_THREAD  {							\
-	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
-	.vm86_info = NULL,						\
-	.sysenter_cs = __KERNEL_CS,					\
-	.io_bitmap_ptr = NULL,						\
-	.fs = __KERNEL_PERCPU,						\
-}
-
-/*
- * Note that the .io_bitmap member must be extra-big. This is because
- * the CPU will access an additional byte beyond the end of the IO
- * permission bitmap. The extra byte must be all 1 bits, and must
- * be within the limit.
- */
-#define INIT_TSS  {							\
-	.x86_tss = {							\
-		.sp0		= sizeof(init_stack) + (long)&init_stack, \
-		.ss0		= __KERNEL_DS,				\
-		.ss1		= __KERNEL_CS,				\
-		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		\
-	 },								\
-	.io_bitmap	= { [ 0 ... IO_BITMAP_LONGS] = ~0 },		\
-}
-
-#define start_thread(regs, new_eip, new_esp) do {		\
-	__asm__("movl %0,%%gs": :"r" (0));			\
-	regs->fs = 0;						\
-	set_fs(USER_DS);					\
-	regs->ds = __USER_DS;					\
-	regs->es = __USER_DS;					\
-	regs->ss = __USER_DS;					\
-	regs->cs = __USER_CS;					\
-	regs->ip = new_eip;					\
-	regs->sp = new_esp;					\
-} while (0)
-
-
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
-#define THREAD_SIZE_LONGS      (THREAD_SIZE/sizeof(unsigned long))
-#define KSTK_TOP(info)                                                 \
-({                                                                     \
-       unsigned long *__ptr = (unsigned long *)(info);                 \
-       (unsigned long)(&__ptr[THREAD_SIZE_LONGS]);                     \
-})
-
-/*
- * The below -8 is to reserve 8 bytes on top of the ring0 stack.
- * This is necessary to guarantee that the entire "struct pt_regs"
- * is accessable even if the CPU haven't stored the SS/ESP registers
- * on the stack (interrupt gate does not save these registers
- * when switching to the same priv ring).
- * Therefore beware: accessing the ss/esp fields of the
- * "struct pt_regs" is possible, but they may contain the
- * completely wrong values.
- */
-#define task_pt_regs(task)                                             \
-({                                                                     \
-       struct pt_regs *__regs__;                                       \
-       __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \
-       __regs__ - 1;                                                   \
-})
-
-#define KSTK_ESP(task) (task_pt_regs(task)->sp)
-
-#endif /* __ASM_I386_PROCESSOR_H */
diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h
deleted file mode 100644
index 04ce823ea7e6..000000000000
--- a/include/asm-x86/processor_64.h
+++ /dev/null
@@ -1,66 +0,0 @@
-/*
- * Copyright (C) 1994 Linus Torvalds
- */
-
-#ifndef __ASM_X86_64_PROCESSOR_H
-#define __ASM_X86_64_PROCESSOR_H
-
-#include <asm/segment.h>
-#include <asm/page.h>
-#include <asm/types.h>
-#include <asm/sigcontext.h>
-#include <asm/cpufeature.h>
-#include <linux/threads.h>
-#include <asm/msr.h>
-#include <asm/current.h>
-#include <asm/system.h>
-#include <linux/personality.h>
-#include <asm/desc_defs.h>
-
-/*
- * User space process size. 47bits minus one guard page.
- */
-#define TASK_SIZE64	(0x800000000000UL - 4096)
-
-/* This decides where the kernel will search for a free chunk of vm
- * space during mmap's.
- */
-#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000)
-
-#define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
-#define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
-
-
-
-DECLARE_PER_CPU(struct orig_ist, orig_ist);
-
-#define INIT_THREAD  { \
-	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
-}
-
-#define INIT_TSS  { \
-	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
-}
-
-#define start_thread(regs,new_rip,new_rsp) do { \
-	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));	 \
-	load_gs_index(0);							\
-	(regs)->ip = (new_rip);						 \
-	(regs)->sp = (new_rsp);						 \
-	write_pda(oldrsp, (new_rsp));						 \
-	(regs)->cs = __USER_CS;							 \
-	(regs)->ss = __USER_DS;							 \
-	(regs)->flags = 0x200;							 \
-	set_fs(USER_DS);							 \
-} while(0) 
-
-/*
- * Return saved PC of a blocked thread.
- * What is this good for? it will be always the scheduler or ret_from_fork.
- */
-#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8))
-
-#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
-#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
-
-#endif /* __ASM_X86_64_PROCESSOR_H */
-- 
cgit v1.2.3


From bca25bafbb390eeec376ac994954b99489d198ec Mon Sep 17 00:00:00 2001
From: Parag Warudkar <parag.warudkar@gmail.com>
Date: Wed, 30 Jan 2008 13:31:59 +0100
Subject: x86: fix dmi_alloc() to not advance alloc index in case of

dmi_alloc() for CONFIG_X86_64 is defined to allocate from a static array
and it maintains a allocation index which is advanced each time allocation
is attempted - it gets incremented even if an allocation fails thereby
depriving any future request that may be small enough to be satisfied from
the array.

Fix this by first testing if allocation is going to be possible and
incrementing alloc index only then.

Signed-off-by: Parag Warudkar <parag.warudkar@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/dmi.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/dmi.h b/include/asm-x86/dmi.h
index 8e2b0e6aa8e7..5008c365e6e4 100644
--- a/include/asm-x86/dmi.h
+++ b/include/asm-x86/dmi.h
@@ -22,8 +22,9 @@ extern char dmi_alloc_data[DMI_MAX_DATA];
 static inline void *dmi_alloc(unsigned len)
 {
 	int idx = dmi_alloc_index;
-	if ((dmi_alloc_index += len) > DMI_MAX_DATA)
+	if ((dmi_alloc_index + len) > DMI_MAX_DATA)
 		return NULL;
+	dmi_alloc_index += len;
 	return dmi_alloc_data + idx;
 }
 
-- 
cgit v1.2.3


From bde6f5f59c2b2b48a7a849c129d5b48838fe77ee Mon Sep 17 00:00:00 2001
From: Venki Pallipadi <venkatesh.pallipadi@intel.com>
Date: Wed, 30 Jan 2008 13:32:01 +0100
Subject: x86: voluntary leave_mm before entering ACPI C3

Aviod TLB flush IPIs during C3 states by voluntary leave_mm()
before entering C3.

The performance impact of TLB flush on C3 should not be significant with
respect to C3 wakeup latency. Also, CPUs tend to flush TLB in hardware while in
C3 anyways.

On a 8 logical CPU system, running make -j2, the number of tlbflush IPIs goes
down from 40 per second to ~ 0. Total number of interrupts during the run
of this workload was ~1200 per second, which makes it ~3% savings in wakeups.

There was no measurable performance or power impact however.

[ akpm@linux-foundation.org: symbol export fixes. ]

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/smp_32.c         | 3 ++-
 arch/x86/kernel/smp_64.c         | 3 ++-
 drivers/acpi/processor_idle.c    | 2 ++
 include/asm-ia64/acpi.h          | 2 ++
 include/asm-x86/acpi.h           | 3 +++
 include/asm-x86/mmu.h            | 8 ++++++++
 include/asm-x86/mmu_context_32.h | 2 --
 7 files changed, 19 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_32.c b/arch/x86/kernel/smp_32.c
index 070816ac79e1..dc0cde9d16fb 100644
--- a/arch/x86/kernel/smp_32.c
+++ b/arch/x86/kernel/smp_32.c
@@ -256,13 +256,14 @@ static DEFINE_SPINLOCK(tlbstate_lock);
  * We need to reload %cr3 since the page tables may be going
  * away from under us..
  */
-void leave_mm(unsigned long cpu)
+void leave_mm(int cpu)
 {
 	if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
 		BUG();
 	cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask);
 	load_cr3(swapper_pg_dir);
 }
+EXPORT_SYMBOL_GPL(leave_mm);
 
 /*
  *
diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 02a6533e8909..2fd74b06db67 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -69,13 +69,14 @@ static DEFINE_PER_CPU(union smp_flush_state, flush_state);
  * We cannot call mmdrop() because we are in interrupt context,
  * instead update mm->cpu_vm_mask.
  */
-static inline void leave_mm(int cpu)
+void leave_mm(int cpu)
 {
 	if (read_pda(mmu_state) == TLBSTATE_OK)
 		BUG();
 	cpu_clear(cpu, read_pda(active_mm)->cpu_vm_mask);
 	load_cr3(swapper_pg_dir);
 }
+EXPORT_SYMBOL_GPL(leave_mm);
 
 /*
  *
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 2235f4e02d26..0721a8183c89 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -534,6 +534,7 @@ static void acpi_processor_idle(void)
 		break;
 
 	case ACPI_STATE_C3:
+		acpi_unlazy_tlb(smp_processor_id());
 		/*
 		 * Must be done before busmaster disable as we might
 		 * need to access HPET !
@@ -1423,6 +1424,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 		return 0;
 	}
 
+	acpi_unlazy_tlb(smp_processor_id());
 	/*
 	 * Must be done before busmaster disable as we might need to
 	 * access HPET !
diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h
index 81bcd5e51789..cd1cc39b5599 100644
--- a/include/asm-ia64/acpi.h
+++ b/include/asm-ia64/acpi.h
@@ -127,6 +127,8 @@ extern int __devinitdata pxm_to_nid_map[MAX_PXM_DOMAINS];
 extern int __initdata nid_to_pxm_map[MAX_NUMNODES];
 #endif
 
+#define acpi_unlazy_tlb(x)
+
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/
diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h
index 2feb0c494be7..98a9ca266531 100644
--- a/include/asm-x86/acpi.h
+++ b/include/asm-x86/acpi.h
@@ -27,6 +27,7 @@
 
 #include <asm/numa.h>
 #include <asm/processor.h>
+#include <asm/mmu.h>
 
 #define COMPILER_DEPENDENT_INT64   long long
 #define COMPILER_DEPENDENT_UINT64  unsigned long long
@@ -167,4 +168,6 @@ static inline void acpi_fake_nodes(const struct bootnode *fake_nodes,
 }
 #endif
 
+#define acpi_unlazy_tlb(x)	leave_mm(x)
+
 #endif /*__X86_ASM_ACPI_H*/
diff --git a/include/asm-x86/mmu.h b/include/asm-x86/mmu.h
index 3f922c8e1c88..efa962c38897 100644
--- a/include/asm-x86/mmu.h
+++ b/include/asm-x86/mmu.h
@@ -20,4 +20,12 @@ typedef struct {
 	void *vdso;
 } mm_context_t;
 
+#ifdef CONFIG_SMP
+void leave_mm(int cpu);
+#else
+static inline void leave_mm(int cpu)
+{
+}
+#endif
+
 #endif /* _ASM_X86_MMU_H */
diff --git a/include/asm-x86/mmu_context_32.h b/include/asm-x86/mmu_context_32.h
index 7eb0b0b1fb3c..8198d1cca1f3 100644
--- a/include/asm-x86/mmu_context_32.h
+++ b/include/asm-x86/mmu_context_32.h
@@ -32,8 +32,6 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 #endif
 }
 
-void leave_mm(unsigned long cpu);
-
 static inline void switch_mm(struct mm_struct *prev,
 			     struct mm_struct *next,
 			     struct task_struct *tsk)
-- 
cgit v1.2.3


From e6ae5d9540727b0e2e5e2fbeb683c84671ed0a31 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:32:03 +0100
Subject: x86, ptrace: support 32bit-cross-64bit BTS recording

Support BTS recording of 32bit and 64bit tasks from 32bit or 64bit tasks.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ds.c         | 91 ++++++++++++++++++++++----------------------
 arch/x86/kernel/ptrace.c     |  2 +-
 include/asm-x86/ds.h         | 10 ++---
 include/asm-x86/ptrace-abi.h |  4 +-
 4 files changed, 53 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index 6eb5d49a36bb..1c5ca4d18787 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -111,53 +111,53 @@ static struct ds_configuration ds_cfg;
  * Accessor functions for some DS and BTS fields using the above
  * global ptrace_bts_cfg.
  */
-static inline void *get_bts_buffer_base(char *base)
+static inline unsigned long get_bts_buffer_base(char *base)
 {
-	return *(void **)(base + ds_cfg.bts_buffer_base.offset);
+	return *(unsigned long *)(base + ds_cfg.bts_buffer_base.offset);
 }
-static inline void set_bts_buffer_base(char *base, void *value)
+static inline void set_bts_buffer_base(char *base, unsigned long value)
 {
-	(*(void **)(base + ds_cfg.bts_buffer_base.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.bts_buffer_base.offset)) = value;
 }
-static inline void *get_bts_index(char *base)
+static inline unsigned long get_bts_index(char *base)
 {
-	return *(void **)(base + ds_cfg.bts_index.offset);
+	return *(unsigned long *)(base + ds_cfg.bts_index.offset);
 }
-static inline void set_bts_index(char *base, void *value)
+static inline void set_bts_index(char *base, unsigned long value)
 {
-	(*(void **)(base + ds_cfg.bts_index.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.bts_index.offset)) = value;
 }
-static inline void *get_bts_absolute_maximum(char *base)
+static inline unsigned long get_bts_absolute_maximum(char *base)
 {
-	return *(void **)(base + ds_cfg.bts_absolute_maximum.offset);
+	return *(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset);
 }
-static inline void set_bts_absolute_maximum(char *base, void *value)
+static inline void set_bts_absolute_maximum(char *base, unsigned long value)
 {
-	(*(void **)(base + ds_cfg.bts_absolute_maximum.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.bts_absolute_maximum.offset)) = value;
 }
-static inline void *get_bts_interrupt_threshold(char *base)
+static inline unsigned long get_bts_interrupt_threshold(char *base)
 {
-	return *(void **)(base + ds_cfg.bts_interrupt_threshold.offset);
+	return *(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset);
 }
-static inline void set_bts_interrupt_threshold(char *base, void *value)
+static inline void set_bts_interrupt_threshold(char *base, unsigned long value)
 {
-	(*(void **)(base + ds_cfg.bts_interrupt_threshold.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.bts_interrupt_threshold.offset)) = value;
 }
-static inline long get_from_ip(char *base)
+static inline unsigned long get_from_ip(char *base)
 {
-	return *(long *)(base + ds_cfg.from_ip.offset);
+	return *(unsigned long *)(base + ds_cfg.from_ip.offset);
 }
-static inline void set_from_ip(char *base, long value)
+static inline void set_from_ip(char *base, unsigned long value)
 {
-	(*(long *)(base + ds_cfg.from_ip.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.from_ip.offset)) = value;
 }
-static inline long get_to_ip(char *base)
+static inline unsigned long get_to_ip(char *base)
 {
-	return *(long *)(base + ds_cfg.to_ip.offset);
+	return *(unsigned long *)(base + ds_cfg.to_ip.offset);
 }
-static inline void set_to_ip(char *base, long value)
+static inline void set_to_ip(char *base, unsigned long value)
 {
-	(*(long *)(base + ds_cfg.to_ip.offset)) = value;
+	(*(unsigned long *)(base + ds_cfg.to_ip.offset)) = value;
 }
 static inline unsigned char get_info_type(char *base)
 {
@@ -180,7 +180,7 @@ static inline void set_info_data(char *base, unsigned long value)
 int ds_allocate(void **dsp, size_t bts_size_in_bytes)
 {
 	size_t bts_size_in_records;
-	void *bts;
+	unsigned long bts;
 	void *ds;
 
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
@@ -197,7 +197,7 @@ int ds_allocate(void **dsp, size_t bts_size_in_bytes)
 	if (bts_size_in_bytes <= 0)
 		return -EINVAL;
 
-	bts = kzalloc(bts_size_in_bytes, GFP_KERNEL);
+	bts = (unsigned long)kzalloc(bts_size_in_bytes, GFP_KERNEL);
 
 	if (!bts)
 		return -ENOMEM;
@@ -205,7 +205,7 @@ int ds_allocate(void **dsp, size_t bts_size_in_bytes)
 	ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL);
 
 	if (!ds) {
-		kfree(bts);
+		kfree((void *)bts);
 		return -ENOMEM;
 	}
 
@@ -221,7 +221,7 @@ int ds_allocate(void **dsp, size_t bts_size_in_bytes)
 int ds_free(void **dsp)
 {
 	if (*dsp)
-		kfree(get_bts_buffer_base(*dsp));
+		kfree((void *)get_bts_buffer_base(*dsp));
 	kfree(*dsp);
 	*dsp = 0;
 
@@ -230,7 +230,7 @@ int ds_free(void **dsp)
 
 int ds_get_bts_size(void *ds)
 {
-	size_t size_in_bytes;
+	int size_in_bytes;
 
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
 		return -EOPNOTSUPP;
@@ -246,7 +246,7 @@ int ds_get_bts_size(void *ds)
 
 int ds_get_bts_end(void *ds)
 {
-	size_t size_in_bytes = ds_get_bts_size(ds);
+	int size_in_bytes = ds_get_bts_size(ds);
 
 	if (size_in_bytes <= 0)
 		return size_in_bytes;
@@ -256,7 +256,7 @@ int ds_get_bts_end(void *ds)
 
 int ds_get_bts_index(void *ds)
 {
-	size_t index_offset_in_bytes;
+	int index_offset_in_bytes;
 
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
 		return -EOPNOTSUPP;
@@ -288,19 +288,19 @@ int ds_get_overflow(void *ds)
 int ds_clear(void *ds)
 {
 	int bts_size = ds_get_bts_size(ds);
-	void *bts_base;
+	unsigned long bts_base;
 
 	if (bts_size <= 0)
 		return bts_size;
 
 	bts_base = get_bts_buffer_base(ds);
-	memset(bts_base, 0, bts_size);
+	memset((void *)bts_base, 0, bts_size);
 
 	set_bts_index(ds, bts_base);
 	return 0;
 }
 
-int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
+int ds_read_bts(void *ds, int index, struct bts_struct *out)
 {
 	void *bts;
 
@@ -313,8 +313,7 @@ int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
 	if (index >= ds_get_bts_size(ds))
 		return -EINVAL;
 
-	bts = get_bts_buffer_base(ds);
-	bts = (char *)bts + (index * ds_cfg.sizeof_bts);
+	bts = (void *)(get_bts_buffer_base(ds) + (index * ds_cfg.sizeof_bts));
 
 	memset(out, 0, sizeof(*out));
 	if (get_from_ip(bts) == BTS_ESCAPE_ADDRESS) {
@@ -326,12 +325,12 @@ int ds_read_bts(void *ds, size_t index, struct bts_struct *out)
 		out->variant.lbr.to_ip   = get_to_ip(bts);
 	}
 
-	return 0;
+	return sizeof(*out);;
 }
 
 int ds_write_bts(void *ds, const struct bts_struct *in)
 {
-	void *bts;
+	unsigned long bts;
 
 	if (!ds_cfg.sizeof_ds || !ds_cfg.sizeof_bts)
 		return -EOPNOTSUPP;
@@ -341,33 +340,33 @@ int ds_write_bts(void *ds, const struct bts_struct *in)
 
 	bts = get_bts_index(ds);
 
-	memset(bts, 0, ds_cfg.sizeof_bts);
+	memset((void *)bts, 0, ds_cfg.sizeof_bts);
 	switch (in->qualifier) {
 	case BTS_INVALID:
 		break;
 
 	case BTS_BRANCH:
-		set_from_ip(bts, in->variant.lbr.from_ip);
-		set_to_ip(bts, in->variant.lbr.to_ip);
+		set_from_ip((void *)bts, in->variant.lbr.from_ip);
+		set_to_ip((void *)bts, in->variant.lbr.to_ip);
 		break;
 
 	case BTS_TASK_ARRIVES:
 	case BTS_TASK_DEPARTS:
-		set_from_ip(bts, BTS_ESCAPE_ADDRESS);
-		set_info_type(bts, in->qualifier);
-		set_info_data(bts, in->variant.jiffies);
+		set_from_ip((void *)bts, BTS_ESCAPE_ADDRESS);
+		set_info_type((void *)bts, in->qualifier);
+		set_info_data((void *)bts, in->variant.jiffies);
 		break;
 
 	default:
 		return -EINVAL;
 	}
 
-	bts = (char *)bts + ds_cfg.sizeof_bts;
+	bts = bts + ds_cfg.sizeof_bts;
 	if (bts >= get_bts_absolute_maximum(ds))
 		bts = get_bts_buffer_base(ds);
 	set_bts_index(ds, bts);
 
-	return 0;
+	return ds_cfg.sizeof_bts;
 }
 
 unsigned long ds_debugctl_mask(void)
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 88ed1e74cee9..236528bec6eb 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -558,7 +558,7 @@ static int ptrace_bts_read_record(struct task_struct *child,
 
 	retval = ds_read_bts((void *)child->thread.ds_area_msr,
 			     bts_index, &ret);
-	if (retval)
+	if (retval < 0)
 		return retval;
 
 	if (copy_to_user(out, &ret, sizeof(ret)))
diff --git a/include/asm-x86/ds.h b/include/asm-x86/ds.h
index b84040abee68..7881368142fa 100644
--- a/include/asm-x86/ds.h
+++ b/include/asm-x86/ds.h
@@ -39,16 +39,16 @@ enum bts_qualifier {
 };
 
 struct bts_struct {
-	enum bts_qualifier qualifier;
+	u64 qualifier;
 	union {
 		/* BTS_BRANCH */
 		struct {
-			long from_ip;
-			long to_ip;
+			u64 from_ip;
+			u64 to_ip;
 		} lbr;
 		/* BTS_TASK_ARRIVES or
 		   BTS_TASK_DEPARTS */
-		unsigned long jiffies;
+		u64 jiffies;
 	} variant;
 };
 
@@ -64,7 +64,7 @@ extern int ds_get_bts_index(void *);
 extern int ds_set_overflow(void *, int);
 extern int ds_get_overflow(void *);
 extern int ds_clear(void *);
-extern int ds_read_bts(void *, size_t, struct bts_struct *);
+extern int ds_read_bts(void *, int, struct bts_struct *);
 extern int ds_write_bts(void *, const struct bts_struct *);
 extern unsigned long ds_debugctl_mask(void);
 extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *c);
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index cf2fe4633ee5..32fe137822bf 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -85,9 +85,9 @@
 */
 struct ptrace_bts_config {
 	/* requested or actual size of BTS buffer in bytes */
-	unsigned long size;
+	unsigned int size;
 	/* bitmask of below flags */
-	unsigned long flags;
+	unsigned int flags;
 };
 
 #define PTRACE_BTS_O_TRACE	0x1 /* branch trace */
-- 
cgit v1.2.3


From cba4b65d359268c40679ca75ac92c0b93cecf6de Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:32:03 +0100
Subject: x86, ptrace: add buffer size checks

Pass the buffer size for (most) ptrace commands that pass user-allocated buffers and check that size before accessing the buffer. Unfortunately, PTRACE_BTS_GET already uses all 4 parameters.
Commands that access user buffers return the number of bytes or records read or written.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ptrace.c     | 25 +++++++++++++++++++++----
 include/asm-x86/ptrace-abi.h | 14 ++++++++------
 2 files changed, 29 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 236528bec6eb..e19a91db9b35 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -591,6 +591,7 @@ static int ptrace_bts_clear(struct task_struct *child)
 }
 
 static int ptrace_bts_drain(struct task_struct *child,
+			    long size,
 			    struct bts_struct __user *out)
 {
 	int end, i;
@@ -603,6 +604,9 @@ static int ptrace_bts_drain(struct task_struct *child,
 	if (end <= 0)
 		return end;
 
+	if (size < (end * sizeof(struct bts_struct)))
+		return -EIO;
+
 	for (i = 0; i < end; i++, out++) {
 		struct bts_struct ret;
 		int retval;
@@ -617,7 +621,7 @@ static int ptrace_bts_drain(struct task_struct *child,
 
 	ds_clear(ds);
 
-	return i;
+	return end;
 }
 
 static int ptrace_bts_realloc(struct task_struct *child,
@@ -690,15 +694,22 @@ out:
 }
 
 static int ptrace_bts_config(struct task_struct *child,
+			     long cfg_size,
 			     const struct ptrace_bts_config __user *ucfg)
 {
 	struct ptrace_bts_config cfg;
 	int bts_size, ret = 0;
 	void *ds;
 
+	if (cfg_size < sizeof(cfg))
+		return -EIO;
+
 	if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
 		return -EFAULT;
 
+	if ((int)cfg.size < 0)
+		return -EINVAL;
+
 	bts_size = 0;
 	ds = (void *)child->thread.ds_area_msr;
 	if (ds) {
@@ -734,6 +745,8 @@ static int ptrace_bts_config(struct task_struct *child,
 	else
 		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
 
+	ret = sizeof(cfg);
+
 out:
 	if (child->thread.debugctlmsr)
 		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
@@ -749,11 +762,15 @@ errout:
 }
 
 static int ptrace_bts_status(struct task_struct *child,
+			     long cfg_size,
 			     struct ptrace_bts_config __user *ucfg)
 {
 	void *ds = (void *)child->thread.ds_area_msr;
 	struct ptrace_bts_config cfg;
 
+	if (cfg_size < sizeof(cfg))
+		return -EIO;
+
 	memset(&cfg, 0, sizeof(cfg));
 
 	if (ds) {
@@ -923,12 +940,12 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_BTS_CONFIG:
 		ret = ptrace_bts_config
-			(child, (struct ptrace_bts_config __user *)addr);
+			(child, data, (struct ptrace_bts_config __user *)addr);
 		break;
 
 	case PTRACE_BTS_STATUS:
 		ret = ptrace_bts_status
-			(child, (struct ptrace_bts_config __user *)addr);
+			(child, data, (struct ptrace_bts_config __user *)addr);
 		break;
 
 	case PTRACE_BTS_SIZE:
@@ -946,7 +963,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_BTS_DRAIN:
 		ret = ptrace_bts_drain
-			(child, (struct bts_struct __user *) addr);
+			(child, data, (struct bts_struct __user *) addr);
 		break;
 
 	default:
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index 32fe137822bf..bcf67044754c 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -99,13 +99,15 @@ struct ptrace_bts_config {
 
 #define PTRACE_BTS_CONFIG	40
 /* Configure branch trace recording.
-   DATA is ignored, ADDR points to a struct ptrace_bts_config.
+   ADDR points to a struct ptrace_bts_config.
+   DATA gives the size of that buffer.
    A new buffer is allocated, iff the size changes.
+   Returns the number of bytes read.
 */
 #define PTRACE_BTS_STATUS	41
-/* Return the current configuration.
-   DATA is ignored, ADDR points to a struct ptrace_bts_config
-   that will contain the result.
+/* Return the current configuration in a struct ptrace_bts_config
+   pointed to by ADDR; DATA gives the size of that buffer.
+   Returns the number of bytes written.
 */
 #define PTRACE_BTS_SIZE		42
 /* Return the number of available BTS records.
@@ -123,8 +125,8 @@ struct ptrace_bts_config {
 */
 #define PTRACE_BTS_DRAIN	45
 /* Read all available BTS records and clear the buffer.
-   DATA is ignored. ADDR points to an array of struct bts_struct of
-   suitable size.
+   ADDR points to an array of struct bts_struct.
+   DATA gives the size of that buffer.
    BTS records are read from oldest to newest.
    Returns number of BTS records drained.
 */
-- 
cgit v1.2.3


From c6334593c61c71ab2e666c015eef13995736f49a Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:32:04 +0100
Subject: x86, ptrace: overflow signal API

Establish the user API for sending a user-defined signal to the traced task on a BTS buffer overflow.

This should complete the user API for the BTS ptrace extension.
The patches so far implement wrap-around overflow handling as is needed for debugging.

The remaining open is another overflow handling mechanism that sends a signal to the traced task on a buffer overflow.
This will take some more time from my side.

Since, from a user perspective, this occurs behind the scenes, the patch set should already be useful. More features may/will be added on top of it (overflow signal, pageable back-up buffers, kernel tracing, core file support, profiling, ...).

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/ptrace-abi.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index bcf67044754c..b3b9e023afce 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -88,11 +88,13 @@ struct ptrace_bts_config {
 	unsigned int size;
 	/* bitmask of below flags */
 	unsigned int flags;
+	/* buffer overflow signal */
+	unsigned int signal;
 };
 
 #define PTRACE_BTS_O_TRACE	0x1 /* branch trace */
 #define PTRACE_BTS_O_SCHED	0x2 /* scheduling events w/ jiffies */
-#define PTRACE_BTS_O_SIGNAL     0x4 /* send SIG? on buffer overflow
+#define PTRACE_BTS_O_SIGNAL     0x4 /* send SIG<signal> on buffer overflow
 				       instead of wrapping around */
 #define PTRACE_BTS_O_CUT_SIZE	0x8 /* cut requested size to max available
 				       instead of failing */
-- 
cgit v1.2.3


From a4746364da0c7caa155cc945a5c10312e7925b46 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:05 +0100
Subject: x86: adjust PVOP_CALL/VCALL macros for 64-bit

This patch adjust the PVOP_VCALL and PVOP_CALL macros to
work with x86_64. It has a different calling convention, and
we use auxiliary macros to account for both calling conventions
as cleanly as possible

Comments are adjusted accordingly.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 87 ++++++++++++++++++++++++++++++++++------------
 1 file changed, 65 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 24406703007f..bef932772661 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -320,7 +320,7 @@ int paravirt_disable_iospace(void);
  * runtime.
  *
  * Normally, a call to a pv_op function is a simple indirect call:
- * (paravirt_ops.operations)(args...).
+ * (pv_op_struct.operations)(args...).
  *
  * Unfortunately, this is a relatively slow operation for modern CPUs,
  * because it cannot necessarily determine what the destination
@@ -330,11 +330,17 @@ int paravirt_disable_iospace(void);
  * calls are essentially free, because the call and return addresses
  * are completely predictable.)
  *
- * These macros rely on the standard gcc "regparm(3)" calling
+ * For i386, these macros rely on the standard gcc "regparm(3)" calling
  * convention, in which the first three arguments are placed in %eax,
  * %edx, %ecx (in that order), and the remaining arguments are placed
  * on the stack.  All caller-save registers (eax,edx,ecx) are expected
  * to be modified (either clobbered or used for return values).
+ * X86_64, on the other hand, already specifies a register-based calling
+ * conventions, returning at %rax, with parameteres going on %rdi, %rsi,
+ * %rdx, and %rcx. Note that for this reason, x86_64 does not need any
+ * special handling for dealing with 4 arguments, unlike i386.
+ * However, x86_64 also have to clobber all caller saved registers, which
+ * unfortunately, are quite a bit (r8 - r11)
  *
  * The call instruction itself is marked by placing its start address
  * and size into the .parainstructions section, so that
@@ -357,10 +363,12 @@ int paravirt_disable_iospace(void);
  * the return type.  The macro then uses sizeof() on that type to
  * determine whether its a 32 or 64 bit value, and places the return
  * in the right register(s) (just %eax for 32-bit, and %edx:%eax for
- * 64-bit).
+ * 64-bit). For x86_64 machines, it just returns at %rax regardless of
+ * the return value size.
  *
  * 64-bit arguments are passed as a pair of adjacent 32-bit arguments
- * in low,high order.
+ * i386 also passes 64-bit arguments as a pair of adjacent 32-bit arguments
+ * in low,high order
  *
  * Small structures are passed and returned in registers.  The macro
  * calling convention can't directly deal with this, so the wrapper
@@ -370,46 +378,67 @@ int paravirt_disable_iospace(void);
  * means that all uses must be wrapped in inline functions.  This also
  * makes sure the incoming and outgoing types are always correct.
  */
+#ifdef CONFIG_X86_32
+#define PVOP_VCALL_ARGS			unsigned long __eax, __edx, __ecx
+#define PVOP_CALL_ARGS			PVOP_VCALL_ARGS
+#define PVOP_VCALL_CLOBBERS		"=a" (__eax), "=d" (__edx),	\
+					"=c" (__ecx)
+#define PVOP_CALL_CLOBBERS		PVOP_VCALL_CLOBBERS
+#define EXTRA_CLOBBERS
+#define VEXTRA_CLOBBERS
+#else
+#define PVOP_VCALL_ARGS		unsigned long __edi, __esi, __edx, __ecx
+#define PVOP_CALL_ARGS		PVOP_VCALL_ARGS, __eax
+#define PVOP_VCALL_CLOBBERS	"=D" (__edi),				\
+				"=S" (__esi), "=d" (__edx),		\
+				"=c" (__ecx)
+
+#define PVOP_CALL_CLOBBERS	PVOP_VCALL_CLOBBERS, "=a" (__eax)
+
+#define EXTRA_CLOBBERS	 , "r8", "r9", "r10", "r11"
+#define VEXTRA_CLOBBERS	 , "rax", "r8", "r9", "r10", "r11"
+#endif
+
 #define __PVOP_CALL(rettype, op, pre, post, ...)			\
 	({								\
 		rettype __ret;						\
-		unsigned long __eax, __edx, __ecx;			\
+		PVOP_CALL_ARGS;					\
+		/* This is 32-bit specific, but is okay in 64-bit */	\
+		/* since this condition will never hold */		\
 		if (sizeof(rettype) > sizeof(unsigned long)) {		\
 			asm volatile(pre				\
 				     paravirt_alt(PARAVIRT_CALL)	\
 				     post				\
-				     : "=a" (__eax), "=d" (__edx),	\
-				       "=c" (__ecx)			\
+				     : PVOP_CALL_CLOBBERS		\
 				     : paravirt_type(op),		\
 				       paravirt_clobber(CLBR_ANY),	\
 				       ##__VA_ARGS__			\
-				     : "memory", "cc");			\
+				     : "memory", "cc" EXTRA_CLOBBERS);	\
 			__ret = (rettype)((((u64)__edx) << 32) | __eax); \
 		} else {						\
 			asm volatile(pre				\
 				     paravirt_alt(PARAVIRT_CALL)	\
 				     post				\
-				     : "=a" (__eax), "=d" (__edx),	\
-				       "=c" (__ecx)			\
+				     : PVOP_CALL_CLOBBERS		\
 				     : paravirt_type(op),		\
 				       paravirt_clobber(CLBR_ANY),	\
 				       ##__VA_ARGS__			\
-				     : "memory", "cc");			\
+				     : "memory", "cc" EXTRA_CLOBBERS);	\
 			__ret = (rettype)__eax;				\
 		}							\
 		__ret;							\
 	})
 #define __PVOP_VCALL(op, pre, post, ...)				\
 	({								\
-		unsigned long __eax, __edx, __ecx;			\
+		PVOP_VCALL_ARGS;					\
 		asm volatile(pre					\
 			     paravirt_alt(PARAVIRT_CALL)		\
 			     post					\
-			     : "=a" (__eax), "=d" (__edx), "=c" (__ecx) \
+			     : PVOP_VCALL_CLOBBERS			\
 			     : paravirt_type(op),			\
 			       paravirt_clobber(CLBR_ANY),		\
 			       ##__VA_ARGS__				\
-			     : "memory", "cc");				\
+			     : "memory", "cc" VEXTRA_CLOBBERS);		\
 	})
 
 #define PVOP_CALL0(rettype, op)						\
@@ -418,22 +447,26 @@ int paravirt_disable_iospace(void);
 	__PVOP_VCALL(op, "", "")
 
 #define PVOP_CALL1(rettype, op, arg1)					\
-	__PVOP_CALL(rettype, op, "", "", "0" ((u32)(arg1)))
+	__PVOP_CALL(rettype, op, "", "", "0" ((unsigned long)(arg1)))
 #define PVOP_VCALL1(op, arg1)						\
-	__PVOP_VCALL(op, "", "", "0" ((u32)(arg1)))
+	__PVOP_VCALL(op, "", "", "0" ((unsigned long)(arg1)))
 
 #define PVOP_CALL2(rettype, op, arg1, arg2)				\
-	__PVOP_CALL(rettype, op, "", "", "0" ((u32)(arg1)), "1" ((u32)(arg2)))
+	__PVOP_CALL(rettype, op, "", "", "0" ((unsigned long)(arg1)), 	\
+	"1" ((unsigned long)(arg2)))
 #define PVOP_VCALL2(op, arg1, arg2)					\
-	__PVOP_VCALL(op, "", "", "0" ((u32)(arg1)), "1" ((u32)(arg2)))
+	__PVOP_VCALL(op, "", "", "0" ((unsigned long)(arg1)), 		\
+	"1" ((unsigned long)(arg2)))
 
 #define PVOP_CALL3(rettype, op, arg1, arg2, arg3)			\
-	__PVOP_CALL(rettype, op, "", "", "0" ((u32)(arg1)),		\
-		    "1"((u32)(arg2)), "2"((u32)(arg3)))
+	__PVOP_CALL(rettype, op, "", "", "0" ((unsigned long)(arg1)),	\
+	"1"((unsigned long)(arg2)), "2"((unsigned long)(arg3)))
 #define PVOP_VCALL3(op, arg1, arg2, arg3)				\
-	__PVOP_VCALL(op, "", "", "0" ((u32)(arg1)), "1"((u32)(arg2)),	\
-		     "2"((u32)(arg3)))
+	__PVOP_VCALL(op, "", "", "0" ((unsigned long)(arg1)),		\
+	"1"((unsigned long)(arg2)), "2"((unsigned long)(arg3)))
 
+/* This is the only difference in x86_64. We can make it much simpler */
+#ifdef CONFIG_X86_32
 #define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
 	__PVOP_CALL(rettype, op,					\
 		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
@@ -444,6 +477,16 @@ int paravirt_disable_iospace(void);
 		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
 		    "0" ((u32)(arg1)), "1" ((u32)(arg2)),		\
 		    "2" ((u32)(arg3)), [_arg4] "mr" ((u32)(arg4)))
+#else
+#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
+	__PVOP_CALL(rettype, op, "", "", "0" ((unsigned long)(arg1)),	\
+	"1"((unsigned long)(arg2)), "2"((unsigned long)(arg3)),		\
+	"3"((unsigned long)(arg4)))
+#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4)				\
+	__PVOP_VCALL(op, "", "", "0" ((unsigned long)(arg1)),		\
+	"1"((unsigned long)(arg2)), "2"((unsigned long)(arg3)),		\
+	"3"((unsigned long)(arg4)))
+#endif
 
 static inline int paravirt_enabled(void)
 {
-- 
cgit v1.2.3


From f72a9ef979c5a828c64deb88ebba743f7d899907 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:05 +0100
Subject: x86: cleanup write_tsc

write_tsc() does not need to be enclosed in any paravirt closure,
as it uses wrmsr(). So we rip off the duplicate in msr.h
and the definition from paravirt.h

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr.h      | 2 --
 include/asm-x86/paravirt.h | 2 --
 2 files changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 088652a62d58..bca8c3950132 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -157,8 +157,6 @@ static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 #define rdtscll(val)						\
 	((val) = native_read_tsc())
 
-#define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
-
 #define rdpmc(counter,low,high)					\
 	do {							\
 		u64 _l = native_read_pmc(counter);		\
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index bef932772661..66290f58e939 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -657,8 +657,6 @@ static inline unsigned long long paravirt_sched_clock(void)
 }
 #define calculate_cpu_khz() (pv_time_ops.get_cpu_khz())
 
-#define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
-
 static inline unsigned long long paravirt_read_pmc(int counter)
 {
 	return PVOP_CALL1(u64, pv_cpu_ops.read_pmc, counter);
-- 
cgit v1.2.3


From e5aaac443635c7c6f842f0bf8169f71f3236d574 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:05 +0100
Subject: x86: provide paravirtualized hook for rdtscp

This patch adds a field in pv_cpu_ops for a paravirtualized hook
for rdtscp, needed for x86_64.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/paravirt.c |  1 +
 include/asm-x86/paravirt.h | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 1a170877f46c..072c1a08efe6 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -374,6 +374,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.write_msr = native_write_msr_safe,
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,
+	.read_tscp = native_read_tscp,
 	.load_tr_desc = native_load_tr_desc,
 	.set_ldt = native_set_ldt,
 	.load_gdt = native_load_gdt,
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 66290f58e939..8f7984319c30 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -120,6 +120,7 @@ struct pv_cpu_ops {
 
 	u64 (*read_tsc)(void);
 	u64 (*read_pmc)(int counter);
+	unsigned long long (*read_tscp)(unsigned int *aux);
 
 	/* These two are jmp to, not actually called. */
 	void (*irq_enable_syscall_ret)(void);
@@ -668,6 +669,27 @@ static inline unsigned long long paravirt_read_pmc(int counter)
 	high = _l >> 32;			\
 } while(0)
 
+static inline unsigned long long paravirt_rdtscp(unsigned int *aux)
+{
+	return PVOP_CALL1(u64, pv_cpu_ops.read_tscp, aux);
+}
+
+#define rdtscp(low, high, aux)				\
+do {							\
+	int __aux;					\
+	unsigned long __val = paravirt_rdtscp(&__aux);	\
+	(low) = (u32)__val;				\
+	(high) = (u32)(__val >> 32);			\
+	(aux) = __aux;					\
+} while (0)
+
+#define rdtscpll(val, aux)				\
+do {							\
+	unsigned long __aux; 				\
+	val = paravirt_rdtscp(&__aux);			\
+	(aux) = __aux;					\
+} while (0)
+
 static inline void load_TR_desc(void)
 {
 	PVOP_VCALL0(pv_cpu_ops.load_tr_desc);
-- 
cgit v1.2.3


From 658be9d395630d30a501c850bde90ac791a678c9 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:06 +0100
Subject: x86: change assembly definition of paravirt_patch_site

To account for differences in x86_64, we change the macros that
create raw instances of the paravirt_patch_site struct.
We need to align 64-pointers to 64-bit boundaries, so we add an alignment
directive. Also, we need to make room for a word-sized pointer,
instead of a fixed 32-bit one

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 8f7984319c30..56389954d100 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -5,6 +5,7 @@
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/page.h>
+#include <asm/asm.h>
 
 /* Bitmask of what can be clobbered: usually at least eax. */
 #define CLBR_NONE 0x0
@@ -281,7 +282,8 @@ extern struct pv_mmu_ops pv_mmu_ops;
 #define _paravirt_alt(insn_string, type, clobber)	\
 	"771:\n\t" insn_string "\n" "772:\n"		\
 	".pushsection .parainstructions,\"a\"\n"	\
-	"  .long 771b\n"				\
+	_ASM_ALIGN "\n"					\
+	_ASM_PTR " 771b\n"				\
 	"  .byte " type "\n"				\
 	"  .byte 772b-771b\n"				\
 	"  .short " clobber "\n"			\
@@ -1159,17 +1161,25 @@ static inline unsigned long __raw_local_irq_save(void)
 
 #define PARA_PATCH(struct, off)	((PARAVIRT_PATCH_##struct + (off)) / 4)
 
-#define PARA_SITE(ptype, clobbers, ops)		\
+#define _PVSITE(ptype, clobbers, ops, word, algn)	\
 771:;						\
 	ops;					\
 772:;						\
 	.pushsection .parainstructions,"a";	\
-	 .long 771b;				\
+	 .align	algn;				\
+	 word 771b;				\
 	 .byte ptype;				\
 	 .byte 772b-771b;			\
 	 .short clobbers;			\
 	.popsection
 
+
+#ifdef CONFIG_X86_64
+#define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .quad, 8)
+#else
+#define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4)
+#endif
+
 #define INTERRUPT_RETURN						\
 	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_iret), CLBR_NONE,	\
 		  jmp *%cs:pv_cpu_ops+PV_CPU_iret)
-- 
cgit v1.2.3


From 6057fc827b3a9f2fe5db18b882ebf96500500b64 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:06 +0100
Subject: x86: adjust assembly macros on 64-bit as well.

This patch adjust the paravirt macros used in assembly code
to accomodate for x86_64 as well.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 56389954d100..854c19364da3 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1159,8 +1159,6 @@ static inline unsigned long __raw_local_irq_save(void)
 
 #else  /* __ASSEMBLY__ */
 
-#define PARA_PATCH(struct, off)	((PARAVIRT_PATCH_##struct + (off)) / 4)
-
 #define _PVSITE(ptype, clobbers, ops, word, algn)	\
 771:;						\
 	ops;					\
@@ -1175,8 +1173,14 @@ static inline unsigned long __raw_local_irq_save(void)
 
 
 #ifdef CONFIG_X86_64
+#define PV_SAVE_REGS   pushq %rax; pushq %rdi; pushq %rcx; pushq %rdx
+#define PV_RESTORE_REGS popq %rdx; popq %rcx; popq %rdi; popq %rax
+#define PARA_PATCH(struct, off)        ((PARAVIRT_PATCH_##struct + (off)) / 8)
 #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .quad, 8)
 #else
+#define PV_SAVE_REGS   pushl %eax; pushl %edi; pushl %ecx; pushl %edx
+#define PV_RESTORE_REGS popl %edx; popl %ecx; popl %edi; popl %eax
+#define PARA_PATCH(struct, off)        ((PARAVIRT_PATCH_##struct + (off)) / 4)
 #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4)
 #endif
 
@@ -1186,25 +1190,27 @@ static inline unsigned long __raw_local_irq_save(void)
 
 #define DISABLE_INTERRUPTS(clobbers)					\
 	PARA_SITE(PARA_PATCH(pv_irq_ops, PV_IRQ_irq_disable), clobbers, \
-		  pushl %eax; pushl %ecx; pushl %edx;			\
+		  PV_SAVE_REGS;			\
 		  call *%cs:pv_irq_ops+PV_IRQ_irq_disable;		\
-		  popl %edx; popl %ecx; popl %eax)			\
+		  PV_RESTORE_REGS;)			\
 
 #define ENABLE_INTERRUPTS(clobbers)					\
 	PARA_SITE(PARA_PATCH(pv_irq_ops, PV_IRQ_irq_enable), clobbers,	\
-		  pushl %eax; pushl %ecx; pushl %edx;			\
+		  PV_SAVE_REGS;			\
 		  call *%cs:pv_irq_ops+PV_IRQ_irq_enable;		\
-		  popl %edx; popl %ecx; popl %eax)
+		  PV_RESTORE_REGS;)
 
 #define ENABLE_INTERRUPTS_SYSCALL_RET					\
 	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_syscall_ret),\
 		  CLBR_NONE,						\
 		  jmp *%cs:pv_cpu_ops+PV_CPU_irq_enable_syscall_ret)
 
+#ifdef CONFIG_X86_32
 #define GET_CR0_INTO_EAX			\
 	push %ecx; push %edx;			\
 	call *pv_cpu_ops+PV_CPU_read_cr0;	\
 	pop %edx; pop %ecx
+#endif
 
 #endif /* __ASSEMBLY__ */
 #endif /* CONFIG_PARAVIRT */
-- 
cgit v1.2.3


From 2e47d3e6c35bb5b78fea2b2584c7eeaf782f138d Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:07 +0100
Subject: x86: change irq functions to accomodate 64-bit

This patch changes the irq handling function definitions
in paravirt.h (like raw_local_irq_disable) to accomodate for x86_64.
The differences are in the calling convention.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 43 ++++++++++++++++++++++++++++++-------------
 1 file changed, 30 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 854c19364da3..0735a90f531b 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1085,52 +1085,68 @@ struct paravirt_patch_site {
 extern struct paravirt_patch_site __parainstructions[],
 	__parainstructions_end[];
 
+#ifdef CONFIG_X86_32
+#define PV_SAVE_REGS "pushl %%ecx; pushl %%edx;"
+#define PV_RESTORE_REGS "popl %%edx; popl %%ecx"
+#define PV_FLAGS_ARG "0"
+#define PV_EXTRA_CLOBBERS
+#define PV_VEXTRA_CLOBBERS
+#else
+/* We save some registers, but all of them, that's too much. We clobber all
+ * caller saved registers but the argument parameter */
+#define PV_SAVE_REGS "pushq %%rdi;"
+#define PV_RESTORE_REGS "popq %%rdi;"
+#define PV_EXTRA_CLOBBERS EXTRA_CLOBBERS, "rcx" , "rdx"
+#define PV_VEXTRA_CLOBBERS EXTRA_CLOBBERS, "rdi", "rcx" , "rdx"
+#define PV_FLAGS_ARG "D"
+#endif
+
 static inline unsigned long __raw_local_save_flags(void)
 {
 	unsigned long f;
 
-	asm volatile(paravirt_alt("pushl %%ecx; pushl %%edx;"
+	asm volatile(paravirt_alt(PV_SAVE_REGS
 				  PARAVIRT_CALL
-				  "popl %%edx; popl %%ecx")
+				  PV_RESTORE_REGS)
 		     : "=a"(f)
 		     : paravirt_type(pv_irq_ops.save_fl),
 		       paravirt_clobber(CLBR_EAX)
-		     : "memory", "cc");
+		     : "memory", "cc" PV_VEXTRA_CLOBBERS);
 	return f;
 }
 
 static inline void raw_local_irq_restore(unsigned long f)
 {
-	asm volatile(paravirt_alt("pushl %%ecx; pushl %%edx;"
+	asm volatile(paravirt_alt(PV_SAVE_REGS
 				  PARAVIRT_CALL
-				  "popl %%edx; popl %%ecx")
+				  PV_RESTORE_REGS)
 		     : "=a"(f)
-		     : "0"(f),
+		     : PV_FLAGS_ARG(f),
 		       paravirt_type(pv_irq_ops.restore_fl),
 		       paravirt_clobber(CLBR_EAX)
-		     : "memory", "cc");
+		     : "memory", "cc" PV_EXTRA_CLOBBERS);
 }
 
 static inline void raw_local_irq_disable(void)
 {
-	asm volatile(paravirt_alt("pushl %%ecx; pushl %%edx;"
+	asm volatile(paravirt_alt(PV_SAVE_REGS
 				  PARAVIRT_CALL
-				  "popl %%edx; popl %%ecx")
+				  PV_RESTORE_REGS)
 		     :
 		     : paravirt_type(pv_irq_ops.irq_disable),
 		       paravirt_clobber(CLBR_EAX)
-		     : "memory", "eax", "cc");
+		     : "memory", "eax", "cc" PV_EXTRA_CLOBBERS);
 }
 
 static inline void raw_local_irq_enable(void)
 {
-	asm volatile(paravirt_alt("pushl %%ecx; pushl %%edx;"
+	asm volatile(paravirt_alt(PV_SAVE_REGS
 				  PARAVIRT_CALL
-				  "popl %%edx; popl %%ecx")
+				  PV_RESTORE_REGS)
 		     :
 		     : paravirt_type(pv_irq_ops.irq_enable),
 		       paravirt_clobber(CLBR_EAX)
-		     : "memory", "eax", "cc");
+		     : "memory", "eax", "cc" PV_EXTRA_CLOBBERS);
 }
 
 static inline unsigned long __raw_local_irq_save(void)
@@ -1205,6 +1221,7 @@ static inline unsigned long __raw_local_irq_save(void)
 		  CLBR_NONE,						\
 		  jmp *%cs:pv_cpu_ops+PV_CPU_irq_enable_syscall_ret)
 
+
 #ifdef CONFIG_X86_32
 #define GET_CR0_INTO_EAX			\
 	push %ecx; push %edx;			\
-- 
cgit v1.2.3


From 4a8c4c4e10d1bc2c3bd80caabb6a76a66849f7e8 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:07 +0100
Subject: x86: add macro for privileged 64-bit operation

i386 has a macro GET_CR0_INTO_EAX, used in early trap handling code.
x86_64 has similar needs, only it needs to put cr2 into rcx. We provide
a macro for such task, in the same way

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 0735a90f531b..7d1c126e2249 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1227,6 +1227,12 @@ static inline unsigned long __raw_local_irq_save(void)
 	push %ecx; push %edx;			\
 	call *pv_cpu_ops+PV_CPU_read_cr0;	\
 	pop %edx; pop %ecx
+#else
+#define GET_CR2_INTO_RCX			\
+	call *pv_mmu_ops+PV_MMU_read_cr2;	\
+	movq %rax, %rcx;			\
+	xorq %rax, %rax;
+
 #endif
 
 #endif /* __ASSEMBLY__ */
-- 
cgit v1.2.3


From e801f864ec7e5b149bd05337800e419f408523bb Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:08 +0100
Subject: x86: adds paravirt hook for swapgs

This patch adds paravirt hook for swapgs operation, which is a privileged
operation in x86_64.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/paravirt.c  | 1 +
 include/asm-x86/paravirt.h  | 9 +++++++++
 include/asm-x86/processor.h | 8 ++++++++
 3 files changed, 18 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 072c1a08efe6..e7c17cc4a99e 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -390,6 +390,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 
 	.irq_enable_syscall_ret = native_irq_enable_syscall_ret,
 	.iret = native_iret,
+	.swapgs = native_swapgs,
 
 	.set_iopl_mask = native_set_iopl_mask,
 	.io_delay = native_io_delay,
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 7d1c126e2249..e8fbf742d425 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -127,6 +127,8 @@ struct pv_cpu_ops {
 	void (*irq_enable_syscall_ret)(void);
 	void (*iret)(void);
 
+	void (*swapgs)(void);
+
 	struct pv_lazy_ops lazy_mode;
 };
 
@@ -1228,6 +1230,13 @@ static inline unsigned long __raw_local_irq_save(void)
 	call *pv_cpu_ops+PV_CPU_read_cr0;	\
 	pop %edx; pop %ecx
 #else
+#define SWAPGS								\
+	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_swapgs), CLBR_NONE,	\
+		  PV_SAVE_REGS;						\
+		  call *pv_cpu_ops+PV_CPU_swapgs;			\
+		  PV_RESTORE_REGS					\
+		 )
+
 #define GET_CR2_INTO_RCX			\
 	call *pv_mmu_ops+PV_MMU_read_cr2;	\
 	movq %rax, %rcx;			\
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index e6fa06fee72a..72740c6f1109 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -435,6 +435,13 @@ static inline void native_load_sp0(struct tss_struct *tss,
 #endif
 }
 
+static inline void native_swapgs(void)
+{
+#ifdef CONFIG_X86_64
+	asm volatile("swapgs" ::: "memory");
+#endif
+}
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
@@ -456,6 +463,7 @@ static inline void load_sp0(struct tss_struct *tss,
 }
 
 #define set_iopl_mask native_set_iopl_mask
+#define SWAPGS	swapgs
 #endif /* CONFIG_PARAVIRT */
 
 /*
-- 
cgit v1.2.3


From 1954448fb0c03dd5e301d7bef2dda71cd9b2fcf2 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:09 +0100
Subject: x86: cleanup CLI_STRING, STI_STRING and friends

Since the advent of ticket locking, CLI_STRING, STI_STRING, and friends
are not used anymore. They can now be safely deleted.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/spinlock.h | 9 ---------
 1 file changed, 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index 97d52b506af8..2076d5d62d83 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -19,15 +19,6 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#ifdef CONFIG_PARAVIRT
-#include <asm/paravirt.h>
-#else
-#define CLI_STRING	"cli"
-#define STI_STRING	"sti"
-#define CLI_STI_CLOBBERS
-#define CLI_STI_INPUT_ARGS
-#endif /* CONFIG_PARAVIRT */
-
 #ifdef CONFIG_X86_32
 typedef char _slock_t;
 # define LOCK_INS_DEC "decb"
-- 
cgit v1.2.3


From 21438f7c138f0b893a32df3cc77434e39a2145f8 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:09 +0100
Subject: x86: add CLBR_ defines for 64-bit

x86_64 needs a potentially larger clobber list than i386, due to its calling
convention. So we add more CLBR_ defines for it.
Note that CLBR_ANY is different for each of the architectures, since it comprises
the notion of "All call clobbers in this architecture"

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index e8fbf742d425..5935c273af1f 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -8,11 +8,24 @@
 #include <asm/asm.h>
 
 /* Bitmask of what can be clobbered: usually at least eax. */
-#define CLBR_NONE 0x0
-#define CLBR_EAX 0x1
-#define CLBR_ECX 0x2
-#define CLBR_EDX 0x4
-#define CLBR_ANY 0x7
+#define CLBR_NONE 0
+#define CLBR_EAX  (1 << 0)
+#define CLBR_ECX  (1 << 1)
+#define CLBR_EDX  (1 << 2)
+
+#ifdef CONFIG_X86_64
+#define CLBR_RSI  (1 << 3)
+#define CLBR_RDI  (1 << 4)
+#define CLBR_R8   (1 << 5)
+#define CLBR_R9   (1 << 6)
+#define CLBR_R10  (1 << 7)
+#define CLBR_R11  (1 << 8)
+#define CLBR_ANY  ((1 << 9) - 1)
+#include <asm/desc_defs.h>
+#else
+/* CLBR_ANY should match all regs platform has. For i386, that's just it */
+#define CLBR_ANY  ((1 << 3) - 1)
+#endif /* X86_64 */
 
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
-- 
cgit v1.2.3


From 2f485ef568372af4680c4e2f8490efb9f2523b05 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:32:10 +0100
Subject: x86: move patching code to arch-specific file.

The core patching code for paravirt is sufficiently different
among i386 and x86_64, and we move them to specific files.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile_32         |  2 +-
 arch/x86/kernel/paravirt.c          | 50 -------------------------------------
 arch/x86/kernel/paravirt_patch_32.c | 49 ++++++++++++++++++++++++++++++++++++
 include/asm-x86/paravirt.h          |  8 ++++++
 4 files changed, 58 insertions(+), 51 deletions(-)
 create mode 100644 arch/x86/kernel/paravirt_patch_32.c

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile_32 b/arch/x86/kernel/Makefile_32
index cfb71a5deb19..86c6327798b2 100644
--- a/arch/x86/kernel/Makefile_32
+++ b/arch/x86/kernel/Makefile_32
@@ -48,7 +48,7 @@ obj-$(CONFIG_K8_NB)		+= k8.o
 obj-$(CONFIG_MGEODE_LX)		+= geode_32.o mfgpt_32.o
 
 obj-$(CONFIG_VMI)		+= vmi_32.o vmiclock_32.o
-obj-$(CONFIG_PARAVIRT)		+= paravirt.o
+obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirt_patch_32.o
 obj-y				+= pcspeaker.o
 
 obj-$(CONFIG_SCx200)		+= scx200_32.o
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index e7c17cc4a99e..864be0498a32 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -58,59 +58,9 @@ char *memory_setup(void)
 	extern const char start_##ops##_##name[], end_##ops##_##name[];	\
 	asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":")
 
-DEF_NATIVE(pv_irq_ops, irq_disable, "cli");
-DEF_NATIVE(pv_irq_ops, irq_enable, "sti");
-DEF_NATIVE(pv_irq_ops, restore_fl, "push %eax; popf");
-DEF_NATIVE(pv_irq_ops, save_fl, "pushf; pop %eax");
-DEF_NATIVE(pv_cpu_ops, iret, "iret");
-DEF_NATIVE(pv_cpu_ops, irq_enable_syscall_ret, "sti; sysexit");
-DEF_NATIVE(pv_mmu_ops, read_cr2, "mov %cr2, %eax");
-DEF_NATIVE(pv_mmu_ops, write_cr3, "mov %eax, %cr3");
-DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax");
-DEF_NATIVE(pv_cpu_ops, clts, "clts");
-DEF_NATIVE(pv_cpu_ops, read_tsc, "rdtsc");
-
 /* Undefined instruction for dealing with missing ops pointers. */
 static const unsigned char ud2a[] = { 0x0f, 0x0b };
 
-static unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
-			     unsigned long addr, unsigned len)
-{
-	const unsigned char *start, *end;
-	unsigned ret;
-
-	switch(type) {
-#define SITE(ops, x)						\
-	case PARAVIRT_PATCH(ops.x):				\
-		start = start_##ops##_##x;			\
-		end = end_##ops##_##x;				\
-		goto patch_site
-
-	SITE(pv_irq_ops, irq_disable);
-	SITE(pv_irq_ops, irq_enable);
-	SITE(pv_irq_ops, restore_fl);
-	SITE(pv_irq_ops, save_fl);
-	SITE(pv_cpu_ops, iret);
-	SITE(pv_cpu_ops, irq_enable_syscall_ret);
-	SITE(pv_mmu_ops, read_cr2);
-	SITE(pv_mmu_ops, read_cr3);
-	SITE(pv_mmu_ops, write_cr3);
-	SITE(pv_cpu_ops, clts);
-	SITE(pv_cpu_ops, read_tsc);
-#undef SITE
-
-	patch_site:
-		ret = paravirt_patch_insns(ibuf, len, start, end);
-		break;
-
-	default:
-		ret = paravirt_patch_default(type, clobbers, ibuf, addr, len);
-		break;
-	}
-
-	return ret;
-}
-
 unsigned paravirt_patch_nop(void)
 {
 	return 0;
diff --git a/arch/x86/kernel/paravirt_patch_32.c b/arch/x86/kernel/paravirt_patch_32.c
new file mode 100644
index 000000000000..82fc5fcab4f4
--- /dev/null
+++ b/arch/x86/kernel/paravirt_patch_32.c
@@ -0,0 +1,49 @@
+#include <asm/paravirt.h>
+
+DEF_NATIVE(pv_irq_ops, irq_disable, "cli");
+DEF_NATIVE(pv_irq_ops, irq_enable, "sti");
+DEF_NATIVE(pv_irq_ops, restore_fl, "push %eax; popf");
+DEF_NATIVE(pv_irq_ops, save_fl, "pushf; pop %eax");
+DEF_NATIVE(pv_cpu_ops, iret, "iret");
+DEF_NATIVE(pv_cpu_ops, irq_enable_syscall_ret, "sti; sysexit");
+DEF_NATIVE(pv_mmu_ops, read_cr2, "mov %cr2, %eax");
+DEF_NATIVE(pv_mmu_ops, write_cr3, "mov %eax, %cr3");
+DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax");
+DEF_NATIVE(pv_cpu_ops, clts, "clts");
+DEF_NATIVE(pv_cpu_ops, read_tsc, "rdtsc");
+
+unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
+		      unsigned long addr, unsigned len)
+{
+	const unsigned char *start, *end;
+	unsigned ret;
+
+#define PATCH_SITE(ops, x)					\
+		case PARAVIRT_PATCH(ops.x):			\
+			start = start_##ops##_##x;		\
+			end = end_##ops##_##x;			\
+			goto patch_site
+	switch(type) {
+		PATCH_SITE(pv_irq_ops, irq_disable);
+		PATCH_SITE(pv_irq_ops, irq_enable);
+		PATCH_SITE(pv_irq_ops, restore_fl);
+		PATCH_SITE(pv_irq_ops, save_fl);
+		PATCH_SITE(pv_cpu_ops, iret);
+		PATCH_SITE(pv_cpu_ops, irq_enable_syscall_ret);
+		PATCH_SITE(pv_mmu_ops, read_cr2);
+		PATCH_SITE(pv_mmu_ops, read_cr3);
+		PATCH_SITE(pv_mmu_ops, write_cr3);
+		PATCH_SITE(pv_cpu_ops, clts);
+		PATCH_SITE(pv_cpu_ops, read_tsc);
+
+	patch_site:
+		ret = paravirt_patch_insns(ibuf, len, start, end);
+		break;
+
+	default:
+		ret = paravirt_patch_default(type, clobbers, ibuf, addr, len);
+		break;
+	}
+#undef PATCH_SITE
+	return ret;
+}
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 5935c273af1f..b35ed95166e4 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -308,6 +308,11 @@ extern struct pv_mmu_ops pv_mmu_ops;
 #define paravirt_alt(insn_string)					\
 	_paravirt_alt(insn_string, "%c[paravirt_typenum]", "%c[paravirt_clobber]")
 
+/* Simple instruction patching code. */
+#define DEF_NATIVE(ops, name, code) 					\
+	extern const char start_##ops##_##name[], end_##ops##_##name[];	\
+	asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":")
+
 unsigned paravirt_patch_nop(void);
 unsigned paravirt_patch_ignore(unsigned len);
 unsigned paravirt_patch_call(void *insnbuf,
@@ -322,6 +327,9 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
 unsigned paravirt_patch_insns(void *insnbuf, unsigned len,
 			      const char *start, const char *end);
 
+unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
+		      unsigned long addr, unsigned len);
+
 int paravirt_disable_iospace(void);
 
 /*
-- 
cgit v1.2.3


From 27efeb67714608b28c0b213cceb6080749435c6b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:36 +0100
Subject: x86: make ptrace.h safe to include from assembler code

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/ptrace-abi.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index b3b9e023afce..08a12b790a77 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -80,6 +80,7 @@
 
 #define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
 
+#ifndef __ASSEMBLY__
 /* configuration/status structure used in PTRACE_BTS_CONFIG and
    PTRACE_BTS_STATUS commands.
 */
@@ -91,6 +92,7 @@ struct ptrace_bts_config {
 	/* buffer overflow signal */
 	unsigned int signal;
 };
+#endif
 
 #define PTRACE_BTS_O_TRACE	0x1 /* branch trace */
 #define PTRACE_BTS_O_SCHED	0x2 /* scheduling events w/ jiffies */
-- 
cgit v1.2.3


From de4218634e3df6d73a3e6cdfdf3a17fa3bc7e013 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:37 +0100
Subject: x86: implement support to synchronize RDTSC through MFENCE on AMD
 CPUs

According to AMD RDTSC can be synchronized through MFENCE.
Implement the necessary CPUID bit for that.

Cc: andreas.herrmann3@amd.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/amd.c    | 3 +++
 arch/x86/kernel/setup_64.c   | 4 ++--
 include/asm-x86/cpufeature.h | 1 +
 3 files changed, 6 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 1ff88c7f45cf..aaa8101d3d80 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -301,6 +301,9 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	/* K6s reports MCEs but don't actually have all the MSRs */
 	if (c->x86 < 6)
 		clear_bit(X86_FEATURE_MCE, c->x86_capability);
+
+	if (cpu_has_xmm)
+		set_bit(X86_FEATURE_MFENCE_RDTSC, c->x86_capability);
 }
 
 static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 02409100f456..2139aa6ac469 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -746,8 +746,8 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	if (c->x86 == 0xf || c->x86 == 0x10 || c->x86 == 0x11)
 		set_cpu_cap(c, X86_FEATURE_K8);
 
-	/* RDTSC can be speculated around */
-	clear_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
+	/* MFENCE stops RDTSC speculation */
+	set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
 
 	/* Family 10 doesn't support C states in MWAIT so don't use it */
 	if (c->x86 == 0x10 && !force_mwait)
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 75e2f78a7fda..7d53eea8b946 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -79,6 +79,7 @@
 /* 14 free */
 #define X86_FEATURE_SYNC_RDTSC	(3*32+15)  /* RDTSC synchronizes the CPU */
 #define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well on this CPU */
+#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
 #define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
-- 
cgit v1.2.3


From 707fa8ed923b1b6a3d7af0d386b0b3abad28ed19 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:37 +0100
Subject: x86: Implement support to synchronize RDTSC with LFENCE on Intel CPUs

According to Intel RDTSC can be always synchronized with LFENCE
on all current CPUs. Implement the necessary CPUID bit for that.

It is unclear yet if that is true for all future CPUs too,
but if there's another way the kernel can be always updated.

Cc: asit.k.mallick@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/intel.c  | 3 ++-
 arch/x86/kernel/setup_64.c   | 5 +----
 include/asm-x86/cpufeature.h | 1 +
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index e4b7e73e9024..0a4abdb61ae4 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -203,9 +203,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 	}
 #endif
 
+	if (cpu_has_xmm)
+		set_bit(X86_FEATURE_LFENCE_RDTSC, c->x86_capability);
 	if (c->x86 == 15) {
 		set_bit(X86_FEATURE_P4, c->x86_capability);
-		set_bit(X86_FEATURE_SYNC_RDTSC, c->x86_capability);
 	}
 	if (c->x86 == 6) 
 		set_bit(X86_FEATURE_P3, c->x86_capability);
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 2139aa6ac469..bc7758ea06af 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -888,10 +888,7 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 	if (c->x86 == 6)
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	if (c->x86 == 15)
-		set_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
-	else
-		clear_cpu_cap(c, X86_FEATURE_SYNC_RDTSC);
+	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
 	c->x86_max_cores = intel_num_cpu_cores(c);
 
 	srat_detect_node();
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 7d53eea8b946..c1a7e07859c8 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -80,6 +80,7 @@
 #define X86_FEATURE_SYNC_RDTSC	(3*32+15)  /* RDTSC synchronizes the CPU */
 #define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well on this CPU */
 #define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */
+#define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
 #define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
-- 
cgit v1.2.3


From bd61643ef6139ba44142f2ba6e79e03267e6b54a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:38 +0100
Subject: x86: move nop declarations into separate include file

Moving things out of processor.h is always a good thing.

Also needed to avoid include loop in later patch.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/nops.h      | 90 +++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/processor.h | 86 +------------------------------------------
 2 files changed, 91 insertions(+), 85 deletions(-)
 create mode 100644 include/asm-x86/nops.h

(limited to 'include')

diff --git a/include/asm-x86/nops.h b/include/asm-x86/nops.h
new file mode 100644
index 000000000000..fec025c7f58c
--- /dev/null
+++ b/include/asm-x86/nops.h
@@ -0,0 +1,90 @@
+#ifndef _ASM_NOPS_H
+#define _ASM_NOPS_H 1
+
+/* Define nops for use with alternative() */
+
+/* generic versions from gas */
+#define GENERIC_NOP1	".byte 0x90\n"
+#define GENERIC_NOP2    	".byte 0x89,0xf6\n"
+#define GENERIC_NOP3        ".byte 0x8d,0x76,0x00\n"
+#define GENERIC_NOP4        ".byte 0x8d,0x74,0x26,0x00\n"
+#define GENERIC_NOP5        GENERIC_NOP1 GENERIC_NOP4
+#define GENERIC_NOP6	".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP7	".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP8	GENERIC_NOP1 GENERIC_NOP7
+
+/* Opteron 64bit nops */
+#define K8_NOP1 GENERIC_NOP1
+#define K8_NOP2	".byte 0x66,0x90\n"
+#define K8_NOP3	".byte 0x66,0x66,0x90\n"
+#define K8_NOP4	".byte 0x66,0x66,0x66,0x90\n"
+#define K8_NOP5	K8_NOP3 K8_NOP2
+#define K8_NOP6	K8_NOP3 K8_NOP3
+#define K8_NOP7	K8_NOP4 K8_NOP3
+#define K8_NOP8	K8_NOP4 K8_NOP4
+
+/* K7 nops */
+/* uses eax dependencies (arbitary choice) */
+#define K7_NOP1  GENERIC_NOP1
+#define K7_NOP2	".byte 0x8b,0xc0\n"
+#define K7_NOP3	".byte 0x8d,0x04,0x20\n"
+#define K7_NOP4	".byte 0x8d,0x44,0x20,0x00\n"
+#define K7_NOP5	K7_NOP4 ASM_NOP1
+#define K7_NOP6	".byte 0x8d,0x80,0,0,0,0\n"
+#define K7_NOP7        ".byte 0x8D,0x04,0x05,0,0,0,0\n"
+#define K7_NOP8        K7_NOP7 ASM_NOP1
+
+/* P6 nops */
+/* uses eax dependencies (Intel-recommended choice) */
+#define P6_NOP1	GENERIC_NOP1
+#define P6_NOP2	".byte 0x66,0x90\n"
+#define P6_NOP3	".byte 0x0f,0x1f,0x00\n"
+#define P6_NOP4	".byte 0x0f,0x1f,0x40,0\n"
+#define P6_NOP5	".byte 0x0f,0x1f,0x44,0x00,0\n"
+#define P6_NOP6	".byte 0x66,0x0f,0x1f,0x44,0x00,0\n"
+#define P6_NOP7	".byte 0x0f,0x1f,0x80,0,0,0,0\n"
+#define P6_NOP8	".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n"
+
+#if defined(CONFIG_MK8)
+#define ASM_NOP1 K8_NOP1
+#define ASM_NOP2 K8_NOP2
+#define ASM_NOP3 K8_NOP3
+#define ASM_NOP4 K8_NOP4
+#define ASM_NOP5 K8_NOP5
+#define ASM_NOP6 K8_NOP6
+#define ASM_NOP7 K8_NOP7
+#define ASM_NOP8 K8_NOP8
+#elif defined(CONFIG_MK7)
+#define ASM_NOP1 K7_NOP1
+#define ASM_NOP2 K7_NOP2
+#define ASM_NOP3 K7_NOP3
+#define ASM_NOP4 K7_NOP4
+#define ASM_NOP5 K7_NOP5
+#define ASM_NOP6 K7_NOP6
+#define ASM_NOP7 K7_NOP7
+#define ASM_NOP8 K7_NOP8
+#elif defined(CONFIG_M686) || defined(CONFIG_MPENTIUMII) || \
+      defined(CONFIG_MPENTIUMIII) || defined(CONFIG_MPENTIUMM) || \
+      defined(CONFIG_MCORE2) || defined(CONFIG_PENTIUM4)
+#define ASM_NOP1 P6_NOP1
+#define ASM_NOP2 P6_NOP2
+#define ASM_NOP3 P6_NOP3
+#define ASM_NOP4 P6_NOP4
+#define ASM_NOP5 P6_NOP5
+#define ASM_NOP6 P6_NOP6
+#define ASM_NOP7 P6_NOP7
+#define ASM_NOP8 P6_NOP8
+#else
+#define ASM_NOP1 GENERIC_NOP1
+#define ASM_NOP2 GENERIC_NOP2
+#define ASM_NOP3 GENERIC_NOP3
+#define ASM_NOP4 GENERIC_NOP4
+#define ASM_NOP5 GENERIC_NOP5
+#define ASM_NOP6 GENERIC_NOP6
+#define ASM_NOP7 GENERIC_NOP7
+#define ASM_NOP8 GENERIC_NOP8
+#endif
+
+#define ASM_NOP_MAX 8
+
+#endif
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 72740c6f1109..41c002440de7 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -20,6 +20,7 @@ struct mm_struct;
 #include <asm/percpu.h>
 #include <asm/msr.h>
 #include <asm/desc_defs.h>
+#include <asm/nops.h>
 #include <linux/personality.h>
 #include <linux/cpumask.h>
 #include <linux/cache.h>
@@ -674,91 +675,6 @@ extern int bootloader_type;
 extern char ignore_fpu_irq;
 #define cache_line_size() (boot_cpu_data.x86_cache_alignment)
 
-/* generic versions from gas */
-#define GENERIC_NOP1	".byte 0x90\n"
-#define GENERIC_NOP2    	".byte 0x89,0xf6\n"
-#define GENERIC_NOP3        ".byte 0x8d,0x76,0x00\n"
-#define GENERIC_NOP4        ".byte 0x8d,0x74,0x26,0x00\n"
-#define GENERIC_NOP5        GENERIC_NOP1 GENERIC_NOP4
-#define GENERIC_NOP6	".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
-#define GENERIC_NOP7	".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
-#define GENERIC_NOP8	GENERIC_NOP1 GENERIC_NOP7
-
-/* Opteron nops */
-#define K8_NOP1 GENERIC_NOP1
-#define K8_NOP2	".byte 0x66,0x90\n"
-#define K8_NOP3	".byte 0x66,0x66,0x90\n"
-#define K8_NOP4	".byte 0x66,0x66,0x66,0x90\n"
-#define K8_NOP5	K8_NOP3 K8_NOP2
-#define K8_NOP6	K8_NOP3 K8_NOP3
-#define K8_NOP7	K8_NOP4 K8_NOP3
-#define K8_NOP8	K8_NOP4 K8_NOP4
-
-/* K7 nops */
-/* uses eax dependencies (arbitary choice) */
-#define K7_NOP1  GENERIC_NOP1
-#define K7_NOP2	".byte 0x8b,0xc0\n"
-#define K7_NOP3	".byte 0x8d,0x04,0x20\n"
-#define K7_NOP4	".byte 0x8d,0x44,0x20,0x00\n"
-#define K7_NOP5	K7_NOP4 ASM_NOP1
-#define K7_NOP6	".byte 0x8d,0x80,0,0,0,0\n"
-#define K7_NOP7        ".byte 0x8D,0x04,0x05,0,0,0,0\n"
-#define K7_NOP8        K7_NOP7 ASM_NOP1
-
-/* P6 nops */
-/* uses eax dependencies (Intel-recommended choice) */
-#define P6_NOP1	GENERIC_NOP1
-#define P6_NOP2	".byte 0x66,0x90\n"
-#define P6_NOP3	".byte 0x0f,0x1f,0x00\n"
-#define P6_NOP4	".byte 0x0f,0x1f,0x40,0\n"
-#define P6_NOP5	".byte 0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP6	".byte 0x66,0x0f,0x1f,0x44,0x00,0\n"
-#define P6_NOP7	".byte 0x0f,0x1f,0x80,0,0,0,0\n"
-#define P6_NOP8	".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n"
-
-#ifdef CONFIG_MK7
-#define ASM_NOP1 K7_NOP1
-#define ASM_NOP2 K7_NOP2
-#define ASM_NOP3 K7_NOP3
-#define ASM_NOP4 K7_NOP4
-#define ASM_NOP5 K7_NOP5
-#define ASM_NOP6 K7_NOP6
-#define ASM_NOP7 K7_NOP7
-#define ASM_NOP8 K7_NOP8
-#elif defined(CONFIG_M686) || defined(CONFIG_MPENTIUMII) || \
-      defined(CONFIG_MPENTIUMIII) || defined(CONFIG_MPENTIUMM) || \
-      defined(CONFIG_MCORE2) || defined(CONFIG_PENTIUM4) || \
-      defined(CONFIG_MPSC)
-#define ASM_NOP1 P6_NOP1
-#define ASM_NOP2 P6_NOP2
-#define ASM_NOP3 P6_NOP3
-#define ASM_NOP4 P6_NOP4
-#define ASM_NOP5 P6_NOP5
-#define ASM_NOP6 P6_NOP6
-#define ASM_NOP7 P6_NOP7
-#define ASM_NOP8 P6_NOP8
-#elif defined(CONFIG_MK8) || defined(CONFIG_X86_64)
-#define ASM_NOP1 K8_NOP1
-#define ASM_NOP2 K8_NOP2
-#define ASM_NOP3 K8_NOP3
-#define ASM_NOP4 K8_NOP4
-#define ASM_NOP5 K8_NOP5
-#define ASM_NOP6 K8_NOP6
-#define ASM_NOP7 K8_NOP7
-#define ASM_NOP8 K8_NOP8
-#else
-#define ASM_NOP1 GENERIC_NOP1
-#define ASM_NOP2 GENERIC_NOP2
-#define ASM_NOP3 GENERIC_NOP3
-#define ASM_NOP4 GENERIC_NOP4
-#define ASM_NOP5 GENERIC_NOP5
-#define ASM_NOP6 GENERIC_NOP6
-#define ASM_NOP7 GENERIC_NOP7
-#define ASM_NOP8 GENERIC_NOP8
-#endif
-
-#define ASM_NOP_MAX 8
-
 #define HAVE_ARCH_PICK_MMAP_LAYOUT 1
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
-- 
cgit v1.2.3


From 2a10e7c41254941cac87be1eccdcb6379ce097f5 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 30 Jan 2008 13:32:38 +0100
Subject: git-x86: unbreak UML

Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-um/nops.h | 6 ++++++
 1 file changed, 6 insertions(+)
 create mode 100644 include/asm-um/nops.h

(limited to 'include')

diff --git a/include/asm-um/nops.h b/include/asm-um/nops.h
new file mode 100644
index 000000000000..814e9bf5dea6
--- /dev/null
+++ b/include/asm-um/nops.h
@@ -0,0 +1,6 @@
+#ifndef __UM_NOPS_H
+#define __UM_NOPS_H
+
+#include "asm/arch/nops.h"
+
+#endif
-- 
cgit v1.2.3


From fde1b3fa947c2512e3715962ebb1d3a6a9b9bb7d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:38 +0100
Subject: x86: introduce rdtsc_barrier()

rdtsc_barrier() is a new barrier primitive that stops RDTSC speculation
to avoid races with timer interrupts on other CPUs.

It expands either to LFENCE (for Intel CPUs) or MFENCE (for
AMD CPUs) which stops RDTSC on all currently known microarchitectures
that implement SSE. On CPUs without SSE there is generally no RDTSC
speculation.

[ mingo@elte.hu: renamed it to rdtsc_barrier() and made it x86-only ]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/system.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 6c7d1fda4995..39474f2957a3 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -5,6 +5,7 @@
 #include <asm/segment.h>
 #include <asm/cpufeature.h>
 #include <asm/cmpxchg.h>
+#include <asm/nops.h>
 
 #include <linux/kernel.h>
 #include <linux/irqflags.h>
@@ -395,5 +396,17 @@ void default_idle(void);
 #define set_mb(var, value) do { var = value; barrier(); } while (0)
 #endif
 
+/*
+ * Stop RDTSC speculation. This is needed when you need to use RDTSC
+ * (or get_cycles or vread that possibly accesses the TSC) in a defined
+ * code region.
+ *
+ * (Could use an alternative three way for this if there was one.)
+ */
+static inline void rdtsc_barrier(void)
+{
+	alternative(ASM_NOP3, "mfence", X86_FEATURE_MFENCE_RDTSC);
+	alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC);
+}
 
 #endif
-- 
cgit v1.2.3


From cdc7957d1954908a39a6964e9c6f643916e76c4b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:39 +0100
Subject: x86: move native_read_tsc() offline

move native_read_tsc() offline.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/rtc.c | 10 ++++++++++
 include/asm-x86/msr.h |  8 +-------
 2 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c
index d040840ff1b6..c42cf1263882 100644
--- a/arch/x86/kernel/rtc.c
+++ b/arch/x86/kernel/rtc.c
@@ -6,6 +6,7 @@
 #include <linux/mc146818rtc.h>
 
 #include <asm/time.h>
+#include <asm/vsyscall.h>
 
 #ifdef CONFIG_X86_32
 # define CMOS_YEARS_OFFS 1900
@@ -194,3 +195,12 @@ int update_persistent_clock(struct timespec now)
 {
 	return set_rtc_mmss(now.tv_sec);
 }
+
+unsigned long long __vsyscall_fn native_read_tsc(void)
+{
+	DECLARE_ARGS(val, low, high);
+
+	asm volatile("rdtsc" : EAX_EDX_RET(val, low, high));
+	return EAX_EDX_VAL(val, low, high);
+}
+EXPORT_SYMBOL_GPL(native_read_tsc);
diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index bca8c3950132..decfec4ab17c 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -91,13 +91,7 @@ static inline int native_write_msr_safe(unsigned int msr,
 	return err;
 }
 
-static inline unsigned long long native_read_tsc(void)
-{
-	DECLARE_ARGS(val, low, high);
-
-	asm volatile("rdtsc" : EAX_EDX_RET(val, low, high));
-	return EAX_EDX_VAL(val, low, high);
-}
+extern unsigned long long native_read_tsc(void);
 
 static inline unsigned long long native_read_pmc(int counter)
 {
-- 
cgit v1.2.3


From e4026440130b84101f2da7f5a0c7a3b046173d3c Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:39 +0100
Subject: x86: map vsyscalls early enough

map vsyscalls early enough. This is important if a __vsyscall_fn
function is used by other kernel code too.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_64.c    | 2 ++
 arch/x86/kernel/vsyscall_64.c | 3 +--
 include/asm-x86/vsyscall.h    | 2 ++
 3 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index bc7758ea06af..8618178db842 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -45,6 +45,7 @@
 #include <asm/mtrr.h>
 #include <asm/uaccess.h>
 #include <asm/system.h>
+#include <asm/vsyscall.h>
 #include <asm/io.h>
 #include <asm/smp.h>
 #include <asm/msr.h>
@@ -453,6 +454,7 @@ void __init setup_arch(char **cmdline_p)
 #endif
 	reserve_crashkernel();
 	paging_init();
+	map_vsyscall();
 
 	early_quirks();
 
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index e5c1118a8098..3f8242774580 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -319,7 +319,7 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
 	return NOTIFY_DONE;
 }
 
-static void __init map_vsyscall(void)
+void __init map_vsyscall(void)
 {
 	extern char __vsyscall_0;
 	unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0);
@@ -335,7 +335,6 @@ static int __init vsyscall_init(void)
 	BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
 	BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
 	BUG_ON((unsigned long) &vgetcpu != VSYSCALL_ADDR(__NR_vgetcpu));
-	map_vsyscall();
 #ifdef CONFIG_SYSCTL
 	register_sysctl_table(kernel_root_table2);
 #endif
diff --git a/include/asm-x86/vsyscall.h b/include/asm-x86/vsyscall.h
index f01c49f5d108..17b3700949bf 100644
--- a/include/asm-x86/vsyscall.h
+++ b/include/asm-x86/vsyscall.h
@@ -36,6 +36,8 @@ extern volatile unsigned long __jiffies;
 extern int vgetcpu_mode;
 extern struct timezone sys_tz;
 
+extern void map_vsyscall(void);
+
 #endif /* __KERNEL__ */
 
 #endif /* _ASM_X86_64_VSYSCALL_H_ */
-- 
cgit v1.2.3


From 6d63de8dbcda98511206897562ecfcdacf18f523 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:39 +0100
Subject: x86: remove get_cycles_sync

rdtsc is now speculation-safe, so no need for the sync variants of
the APIs.

[ mingo@elte.hu: removed the nsec_barrier() complication. ]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/time_64.c  |  2 +-
 arch/x86/kernel/tsc_64.c   | 12 +++++-----
 arch/x86/kernel/tsc_sync.c |  4 ++--
 include/asm-x86/tsc.h      | 60 +++++++---------------------------------------
 4 files changed, 17 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index 91d4d495904e..61b17f5ec867 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -83,7 +83,7 @@ unsigned long __init native_calculate_cpu_khz(void)
 	rdtscl(tsc_start);
 	do {
 		rdmsrl(MSR_K7_PERFCTR0 + i, pmc_now);
-		tsc_now = get_cycles_sync();
+		tsc_now = get_cycles();
 	} while ((tsc_now - tsc_start) < TICK_COUNT);
 
 	local_irq_restore(flags);
diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c
index 3723401c4593..2cc55b726c22 100644
--- a/arch/x86/kernel/tsc_64.c
+++ b/arch/x86/kernel/tsc_64.c
@@ -181,12 +181,12 @@ static unsigned long __init tsc_read_refs(unsigned long *pm,
 	int i;
 
 	for (i = 0; i < MAX_RETRIES; i++) {
-		t1 = get_cycles_sync();
+		t1 = get_cycles();
 		if (hpet)
 			*hpet = hpet_readl(HPET_COUNTER) & 0xFFFFFFFF;
 		else
 			*pm = acpi_pm_read_early();
-		t2 = get_cycles_sync();
+		t2 = get_cycles();
 		if ((t2 - t1) < SMI_TRESHOLD)
 			return t2;
 	}
@@ -210,9 +210,9 @@ void __init tsc_calibrate(void)
 	outb(0xb0, 0x43);
 	outb((CLOCK_TICK_RATE / (1000 / 50)) & 0xff, 0x42);
 	outb((CLOCK_TICK_RATE / (1000 / 50)) >> 8, 0x42);
-	tr1 = get_cycles_sync();
+	tr1 = get_cycles();
 	while ((inb(0x61) & 0x20) == 0);
-	tr2 = get_cycles_sync();
+	tr2 = get_cycles();
 
 	tsc2 = tsc_read_refs(&pm2, hpet ? &hpet2 : NULL);
 
@@ -300,13 +300,13 @@ __setup("notsc", notsc_setup);
 /* clock source code: */
 static cycle_t read_tsc(void)
 {
-	cycle_t ret = (cycle_t)get_cycles_sync();
+	cycle_t ret = (cycle_t)get_cycles();
 	return ret;
 }
 
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
-	cycle_t ret = (cycle_t)vget_cycles_sync();
+	cycle_t ret = (cycle_t)vget_cycles();
 	return ret;
 }
 
diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c
index 05d8f25de6ae..ace340524c01 100644
--- a/arch/x86/kernel/tsc_sync.c
+++ b/arch/x86/kernel/tsc_sync.c
@@ -46,7 +46,7 @@ static __cpuinit void check_tsc_warp(void)
 	cycles_t start, now, prev, end;
 	int i;
 
-	start = get_cycles_sync();
+	start = get_cycles();
 	/*
 	 * The measurement runs for 20 msecs:
 	 */
@@ -61,7 +61,7 @@ static __cpuinit void check_tsc_warp(void)
 		 */
 		__raw_spin_lock(&sync_lock);
 		prev = last_tsc;
-		now = get_cycles_sync();
+		now = get_cycles();
 		last_tsc = now;
 		__raw_spin_unlock(&sync_lock);
 
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index 401303724130..f51a50da35aa 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -36,62 +36,18 @@ static inline cycles_t get_cycles(void)
 	return ret;
 }
 
-/* Like get_cycles, but make sure the CPU is synchronized. */
-static __always_inline cycles_t __get_cycles_sync(void)
+static inline cycles_t vget_cycles(void)
 {
-	unsigned long long ret;
-	unsigned eax, edx;
-
 	/*
-	 * Use RDTSCP if possible; it is guaranteed to be synchronous
-	 * and doesn't cause a VMEXIT on Hypervisors
+	 * We only do VDSOs on TSC capable CPUs, so this shouldnt
+	 * access boot_cpu_data (which is not VDSO-safe):
 	 */
-	alternative_io(ASM_NOP3, ".byte 0x0f,0x01,0xf9", X86_FEATURE_RDTSCP,
-		       ASM_OUTPUT2("=a" (eax), "=d" (edx)),
-		       "a" (0U), "d" (0U) : "ecx", "memory");
-	ret = (((unsigned long long)edx) << 32) | ((unsigned long long)eax);
-	if (ret)
-		return ret;
-
-	/*
-	 * Don't do an additional sync on CPUs where we know
-	 * RDTSC is already synchronous:
-	 */
-	alternative_io("cpuid", ASM_NOP2, X86_FEATURE_SYNC_RDTSC,
-			  "=a" (eax), "0" (1) : "ebx","ecx","edx","memory");
-
-	return 0;
-}
-
-static __always_inline cycles_t get_cycles_sync(void)
-{
-	unsigned long long ret;
-	ret = __get_cycles_sync();
-	if (!ret)
-		rdtscll(ret);
-	return ret;
-}
-
-#ifdef CONFIG_PARAVIRT
-/*
- * For paravirt guests, some functionalities are executed through function
- * pointers in the various pvops structures.
- * These function pointers exist inside the kernel and can not
- * be accessed by user space. To avoid this, we make a copy of the
- * get_cycles_sync (called in kernel) but force the use of native_read_tsc.
- * Ideally, the guest should set up it's own clock and vread
- */
-static __always_inline long long vget_cycles_sync(void)
-{
-	unsigned long long ret;
-	ret = __get_cycles_sync();
-	if (!ret)
-		ret = native_read_tsc();
-	return ret;
-}
-#else
-# define vget_cycles_sync() get_cycles_sync()
+#ifndef CONFIG_X86_TSC
+	if (!cpu_has_tsc)
+		return 0;
 #endif
+	return (cycles_t) native_read_tsc();
+}
 
 extern void tsc_init(void);
 extern void mark_tsc_unstable(char *reason);
-- 
cgit v1.2.3


From 92767af0e3904b4d35ed547fb514ff6cb227e678 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:40 +0100
Subject: x86: fix sched_clock()

[ andi@firstfloor.org: build fix ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/rtc.c | 13 ++++---------
 include/asm-x86/msr.h | 11 +++++++++++
 include/asm-x86/tsc.h |  2 +-
 3 files changed, 16 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c
index 276cb7073ab1..eb9b1a198f5e 100644
--- a/arch/x86/kernel/rtc.c
+++ b/arch/x86/kernel/rtc.c
@@ -196,14 +196,9 @@ int update_persistent_clock(struct timespec now)
 	return set_rtc_mmss(now.tv_sec);
 }
 
-unsigned long long __vsyscall_fn native_read_tsc(void)
+unsigned long long native_read_tsc(void)
 {
-	DECLARE_ARGS(val, low, high);
-
-	rdtsc_barrier();
-	asm volatile("rdtsc" : EAX_EDX_RET(val, low, high));
-	rdtsc_barrier();
-
-	return EAX_EDX_VAL(val, low, high);
+	return __native_read_tsc();
 }
-EXPORT_SYMBOL_GPL(native_read_tsc);
+EXPORT_SYMBOL(native_read_tsc);
+
diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index decfec4ab17c..204a8a30fecf 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -93,6 +93,17 @@ static inline int native_write_msr_safe(unsigned int msr,
 
 extern unsigned long long native_read_tsc(void);
 
+static __always_inline unsigned long long __native_read_tsc(void)
+{
+	DECLARE_ARGS(val, low, high);
+
+	rdtsc_barrier();
+	asm volatile("rdtsc" : EAX_EDX_RET(val, low, high));
+	rdtsc_barrier();
+
+	return EAX_EDX_VAL(val, low, high);
+}
+
 static inline unsigned long long native_read_pmc(int counter)
 {
 	DECLARE_ARGS(val, low, high);
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index f51a50da35aa..071e0ce5b664 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -46,7 +46,7 @@ static inline cycles_t vget_cycles(void)
 	if (!cpu_has_tsc)
 		return 0;
 #endif
-	return (cycles_t) native_read_tsc();
+	return (cycles_t) __native_read_tsc();
 }
 
 extern void tsc_init(void);
-- 
cgit v1.2.3


From 68071a96655c883b316da9ef497f6dec8953529f Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:32:40 +0100
Subject: x86: remove the now unused X86_FEATURE_SYNC_RDTSC

we need to know whether RDTSC is synchronous or not.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/cpufeature.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index c1a7e07859c8..93c143fb8a51 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -77,7 +77,7 @@
 #define X86_FEATURE_PEBS	(3*32+12)  /* Precise-Event Based Sampling */
 #define X86_FEATURE_BTS		(3*32+13)  /* Branch Trace Store */
 /* 14 free */
-#define X86_FEATURE_SYNC_RDTSC	(3*32+15)  /* RDTSC synchronizes the CPU */
+/* 15 free */
 #define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well on this CPU */
 #define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */
 #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */
-- 
cgit v1.2.3


From 83a5101bf2fa7dcf09ffd436078a021d32c97f85 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:41 +0100
Subject: x86: page.h: unify constants

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199317360 28800
# Node ID ba0ec40a50a7aef1a3153cea124c35e261f5a2df
# Parent  c45c263179cb78284b6b869c574457df088027d1
x86: page.h: unify constants

There are many constants which are shared by 32 and 64-bit.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 125 ++++++++++++++++++++++++++++++++++++++++++----
 include/asm-x86/page_32.h |  46 -----------------
 include/asm-x86/page_64.h |  74 ++-------------------------
 3 files changed, 119 insertions(+), 126 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index a757eb26141d..e0fa4a032ea7 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -1,13 +1,116 @@
-#ifdef __KERNEL__
-# ifdef CONFIG_X86_32
-#  include "page_32.h"
-# else
-#  include "page_64.h"
-# endif
+#ifndef _ASM_X86_PAGE_H
+#define _ASM_X86_PAGE_H
+
+#include <linux/const.h>
+
+/* PAGE_SHIFT determines the page size */
+#define PAGE_SHIFT	12
+#define PAGE_SIZE	(_AC(1,UL) << PAGE_SHIFT)
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+
+#define PHYSICAL_PAGE_MASK	(PAGE_MASK & __PHYSICAL_MASK)
+
+#define LARGE_PAGE_SIZE		(_AC(1,UL) << PMD_SHIFT)
+#define LARGE_PAGE_MASK		(~(LARGE_PAGE_SIZE-1))
+
+#define HPAGE_SHIFT		PMD_SHIFT
+#define HPAGE_SIZE		(_AC(1,UL) << HPAGE_SHIFT)
+#define HPAGE_MASK		(~(HPAGE_SIZE - 1))
+#define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
+
+/* to align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
+
+#define __PHYSICAL_MASK		((_AC(1,UL) << __PHYSICAL_MASK_SHIFT) - 1)
+#define __VIRTUAL_MASK		((_AC(1,UL) << __VIRTUAL_MASK_SHIFT) - 1)
+
+
+#ifdef CONFIG_X86_64
+#define THREAD_ORDER	1
+#define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
+#define CURRENT_MASK (~(THREAD_SIZE-1))
+
+#define EXCEPTION_STACK_ORDER 0
+#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
+
+#define DEBUG_STACK_ORDER (EXCEPTION_STACK_ORDER + 1)
+#define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER)
+
+#define IRQSTACK_ORDER 2
+#define IRQSTACKSIZE (PAGE_SIZE << IRQSTACK_ORDER)
+
+#define STACKFAULT_STACK 1
+#define DOUBLEFAULT_STACK 2
+#define NMI_STACK 3
+#define DEBUG_STACK 4
+#define MCE_STACK 5
+#define N_EXCEPTION_STACKS 5  /* hw limit: 7 */
+
+#define __PAGE_OFFSET           _AC(0xffff810000000000, UL)
+
+#define __PHYSICAL_START	CONFIG_PHYSICAL_START
+#define __KERNEL_ALIGN		0x200000
+
+/*
+ * Make sure kernel is aligned to 2MB address. Catching it at compile
+ * time is better. Change your config file and compile the kernel
+ * for a 2MB aligned address (CONFIG_PHYSICAL_START)
+ */
+#if (CONFIG_PHYSICAL_START % __KERNEL_ALIGN) != 0
+#error "CONFIG_PHYSICAL_START must be a multiple of 2MB"
+#endif
+
+#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
+#define __START_KERNEL_map	_AC(0xffffffff80000000, UL)
+
+/* See Documentation/x86_64/mm.txt for a description of the memory map. */
+#define __PHYSICAL_MASK_SHIFT	46
+#define __VIRTUAL_MASK_SHIFT	48
+
+#define KERNEL_TEXT_SIZE  (40*1024*1024)
+#define KERNEL_TEXT_START _AC(0xffffffff80000000, UL)
+
+#endif	/* CONFIG_X86_64 */
+
+#ifdef CONFIG_X86_32
+
+/*
+ * This handles the memory map.
+ *
+ * A __PAGE_OFFSET of 0xC0000000 means that the kernel has
+ * a virtual address space of one gigabyte, which limits the
+ * amount of physical memory you can use to about 950MB.
+ *
+ * If you want more physical memory than this then see the CONFIG_HIGHMEM4G
+ * and CONFIG_HIGHMEM64G options in the kernel configuration.
+ */
+#define __PAGE_OFFSET		_AC(CONFIG_PAGE_OFFSET, UL)
+
+#ifdef CONFIG_X86_PAE
+#define __PHYSICAL_MASK_SHIFT	36
+#define __VIRTUAL_MASK_SHIFT	32
+#else  /* !CONFIG_X86_PAE */
+#define __PHYSICAL_MASK_SHIFT	32
+#define __VIRTUAL_MASK_SHIFT	32
+#endif	/* CONFIG_X86_PAE */
+
+#ifdef CONFIG_HUGETLB_PAGE
+#define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
+#endif
+
+#endif	/* CONFIG_X86_32 */
+
+#define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
+
+#define VM_DATA_DEFAULT_FLAGS \
+	(((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
+	 VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+
+#ifdef CONFIG_X86_32
+# include "page_32.h"
 #else
-# ifdef __i386__
-#  include "page_32.h"
-# else
-#  include "page_64.h"
-# endif
+# include "page_64.h"
 #endif
+
+#endif	/* _ASM_X86_PAGE_H */
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index c620c934e557..69e520059b26 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -1,14 +1,6 @@
 #ifndef _I386_PAGE_H
 #define _I386_PAGE_H
 
-/* PAGE_SHIFT determines the page size */
-#define PAGE_SHIFT	12
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
-#define PAGE_MASK	(~(PAGE_SIZE-1))
-
-#define LARGE_PAGE_MASK (~(LARGE_PAGE_SIZE-1))
-#define LARGE_PAGE_SIZE (1UL << PMD_SHIFT)
-
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
@@ -111,7 +103,6 @@ static inline pte_t native_make_pte(unsigned long long val)
 #define __pmd(x)	native_make_pmd(x)
 #endif
 
-#define HPAGE_SHIFT	21
 #include <asm-generic/pgtable-nopud.h>
 #else  /* !CONFIG_X86_PAE */
 typedef struct { unsigned long pte_low; } pte_t;
@@ -139,19 +130,11 @@ static inline pte_t native_make_pte(unsigned long val)
 	return (pte_t) { .pte_low = val };
 }
 
-#define HPAGE_SHIFT	22
 #include <asm-generic/pgtable-nopmd.h>
 #endif	/* CONFIG_X86_PAE */
 
 #define PTE_MASK	PAGE_MASK
 
-#ifdef CONFIG_HUGETLB_PAGE
-#define HPAGE_SIZE	((1UL) << HPAGE_SHIFT)
-#define HPAGE_MASK	(~(HPAGE_SIZE - 1))
-#define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
-#define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
-#endif
-
 #define pgprot_val(x)	((x).pgprot)
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
@@ -164,22 +147,6 @@ static inline pte_t native_make_pte(unsigned long val)
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
-/*
- * This handles the memory map.. We could make this a config
- * option, but too many people screw it up, and too few need
- * it.
- *
- * A __PAGE_OFFSET of 0xC0000000 means that the kernel has
- * a virtual address space of one gigabyte, which limits the
- * amount of physical memory you can use to about 950MB. 
- *
- * If you want more physical memory than this then see the CONFIG_HIGHMEM4G
- * and CONFIG_HIGHMEM64G options in the kernel configuration.
- */
-
 #ifndef __ASSEMBLY__
 
 struct vm_area_struct;
@@ -196,14 +163,6 @@ extern int page_is_ram(unsigned long pagenr);
 
 #endif /* __ASSEMBLY__ */
 
-#ifdef __ASSEMBLY__
-#define __PAGE_OFFSET		CONFIG_PAGE_OFFSET
-#else
-#define __PAGE_OFFSET		((unsigned long)CONFIG_PAGE_OFFSET)
-#endif
-
-
-#define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
 #define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
 #define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
 #define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
@@ -219,11 +178,6 @@ extern int page_is_ram(unsigned long pagenr);
 
 #define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
 
-#define VM_DATA_DEFAULT_FLAGS \
-	(VM_READ | VM_WRITE | \
-	((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
-		 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 068b5e4cea4c..69ef7cf9dadd 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -1,42 +1,6 @@
 #ifndef _X86_64_PAGE_H
 #define _X86_64_PAGE_H
 
-#include <linux/const.h>
-
-/* PAGE_SHIFT determines the page size */
-#define PAGE_SHIFT	12
-#define PAGE_SIZE	(_AC(1,UL) << PAGE_SHIFT)
-#define PAGE_MASK	(~(PAGE_SIZE-1))
-#define PHYSICAL_PAGE_MASK	(~(PAGE_SIZE-1) & __PHYSICAL_MASK)
-
-#define THREAD_ORDER 1 
-#define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
-#define CURRENT_MASK (~(THREAD_SIZE-1))
-
-#define EXCEPTION_STACK_ORDER 0
-#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
-
-#define DEBUG_STACK_ORDER (EXCEPTION_STACK_ORDER + 1)
-#define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER)
-
-#define IRQSTACK_ORDER 2
-#define IRQSTACKSIZE (PAGE_SIZE << IRQSTACK_ORDER)
-
-#define STACKFAULT_STACK 1
-#define DOUBLEFAULT_STACK 2
-#define NMI_STACK 3
-#define DEBUG_STACK 4
-#define MCE_STACK 5
-#define N_EXCEPTION_STACKS 5  /* hw limit: 7 */
-
-#define LARGE_PAGE_MASK (~(LARGE_PAGE_SIZE-1))
-#define LARGE_PAGE_SIZE (_AC(1,UL) << PMD_SHIFT)
-
-#define HPAGE_SHIFT PMD_SHIFT
-#define HPAGE_SIZE	(_AC(1,UL) << HPAGE_SHIFT)
-#define HPAGE_MASK	(~(HPAGE_SIZE - 1))
-#define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
-
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
@@ -90,35 +54,6 @@ extern unsigned long phys_base;
 
 #endif /* !__ASSEMBLY__ */
 
-#define __PHYSICAL_START	CONFIG_PHYSICAL_START
-#define __KERNEL_ALIGN		0x200000
-
-/*
- * Make sure kernel is aligned to 2MB address. Catching it at compile
- * time is better. Change your config file and compile the kernel
- * for a 2MB aligned address (CONFIG_PHYSICAL_START)
- */
-#if (CONFIG_PHYSICAL_START % __KERNEL_ALIGN) != 0
-#error "CONFIG_PHYSICAL_START must be a multiple of 2MB"
-#endif
-
-#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
-#define __START_KERNEL_map	_AC(0xffffffff80000000, UL)
-#define __PAGE_OFFSET           _AC(0xffff810000000000, UL)
-
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
-/* See Documentation/x86_64/mm.txt for a description of the memory map. */
-#define __PHYSICAL_MASK_SHIFT	46
-#define __PHYSICAL_MASK		((_AC(1,UL) << __PHYSICAL_MASK_SHIFT) - 1)
-#define __VIRTUAL_MASK_SHIFT	48
-#define __VIRTUAL_MASK		((_AC(1,UL) << __VIRTUAL_MASK_SHIFT) - 1)
-
-#define KERNEL_TEXT_SIZE  (40*1024*1024)
-#define KERNEL_TEXT_START _AC(0xffffffff80000000, UL)
-#define PAGE_OFFSET		__PAGE_OFFSET
-
 #ifndef __ASSEMBLY__
 
 #include <asm/bug.h>
@@ -138,10 +73,6 @@ extern unsigned long __phys_addr(unsigned long);
 #define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
 #define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
 
-#define VM_DATA_DEFAULT_FLAGS \
-	(((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
-	 VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
-
 #define __HAVE_ARCH_GATE_AREA 1	
 #define vmemmap ((struct page *)VMEMMAP_START)
 
@@ -150,4 +81,9 @@ extern unsigned long __phys_addr(unsigned long);
 
 #endif /* __KERNEL__ */
 
+#ifdef CONFIG_FLATMEM
+#define pfn_valid(pfn)          ((pfn) < end_pfn)
+#endif
+
+
 #endif /* _X86_64_PAGE_H */
-- 
cgit v1.2.3


From 345b904c3f7c24fbfadfee7cddd5896d13b176d9 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:42 +0100
Subject: x86: page.h: unify page copying and clearing

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199317362 28800
# Node ID 4d9a413a0f4c1d98dbea704f0366457b5117045d
# Parent  ba0ec40a50a7aef1a3153cea124c35e261f5a2df
x86: page.h: unify page copying and clearing

Move, and to some extent unify, the various page copying and clearing
functions.  The only unification here is that both architectures use
the same function for copying/clearing user and kernel pages.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 55 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/page_32.h | 53 ---------------------------------------------
 include/asm-x86/page_64.h | 20 -----------------
 3 files changed, 55 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index e0fa4a032ea7..2ebb6977e00c 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -70,6 +70,11 @@
 #define KERNEL_TEXT_SIZE  (40*1024*1024)
 #define KERNEL_TEXT_START _AC(0xffffffff80000000, UL)
 
+#ifndef __ASSEMBLY__
+void clear_page(void *page);
+void copy_page(void *to, void *from);
+#endif	/* !__ASSEMBLY__ */
+
 #endif	/* CONFIG_X86_64 */
 
 #ifdef CONFIG_X86_32
@@ -98,6 +103,34 @@
 #define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
 #endif
 
+#ifndef __ASSEMBLY__
+#ifdef CONFIG_X86_USE_3DNOW
+#include <asm/mmx.h>
+
+static inline void clear_page(void *page)
+{
+	mmx_clear_page(page);
+}
+
+static inline void copy_page(void *to, void *from)
+{
+	mmx_copy_page(to, from);
+}
+#else  /* !CONFIG_X86_USE_3DNOW */
+#include <linux/string.h>
+
+static inline void clear_page(void *page)
+{
+	memset(page, 0, PAGE_SIZE);
+}
+
+static inline void copy_page(void *to, void *from)
+{
+	memcpy(to, from, PAGE_SIZE);
+}
+#endif	/* CONFIG_X86_3DNOW */
+#endif	/* !__ASSEMBLY__ */
+
 #endif	/* CONFIG_X86_32 */
 
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
@@ -107,6 +140,28 @@
 	 VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
 
+#ifndef __ASSEMBLY__
+struct page;
+
+static void inline clear_user_page(void *page, unsigned long vaddr,
+				struct page *pg)
+{
+	clear_page(page);
+}
+
+static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
+				struct page *topage)
+{
+	copy_page(to, from);
+}
+
+#define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \
+	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
+#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
+
+#endif	/* __ASSEMBLY__ */
+
+
 #ifdef CONFIG_X86_32
 # include "page_32.h"
 #else
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 69e520059b26..b4750ef44920 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -4,59 +4,6 @@
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
-#include <linux/string.h>
-
-#ifdef CONFIG_X86_USE_3DNOW
-
-#include <asm/mmx.h>
-
-static inline void clear_page(void *page)
-{
-	mmx_clear_page(page);
-}
-
-static inline void copy_page(void *to, void *from)
-{
-	mmx_copy_page(to, from);
-}
-
-#else
-
-/*
- *	On older X86 processors it's not a win to use MMX here it seems.
- *	Maybe the K6-III ?
- */
- 
-static inline void clear_page(void *page)
-{
-	memset(page, 0, PAGE_SIZE);
-}
-
-static inline void copy_page(void *to, void *from)
-{
-	memcpy(to, from, PAGE_SIZE);
-}
-
-#endif
-
-struct page;
-
-static void inline clear_user_page(void *page, unsigned long vaddr,
-				struct page *pg)
-{
-	clear_page(page);
-}
-
-static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
-				struct page *topage)
-{
-	copy_page(to, from);
-}
-
-#define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \
-	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
-#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
-
 /*
  * These are used to make use of C type-checking..
  */
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 69ef7cf9dadd..f5e0543b8afa 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -7,26 +7,6 @@
 extern unsigned long end_pfn;
 extern unsigned long end_pfn_map;
 
-void clear_page(void *page);
-void copy_page(void *to, void *from);
-
-struct page;
-
-static void inline clear_user_page(void *page, unsigned long vaddr,
-				struct page *pg)
-{
-	clear_page(page);
-}
-
-static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
-				struct page *topage)
-{
-	copy_page(to, from);
-}
-
-#define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \
-	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
-#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
 /*
  * These are used to make use of C type-checking..
  */
-- 
cgit v1.2.3


From 74ef649fe847fdfbd3e1732d21b923f59ca04e8c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:42 +0100
Subject: x86: add _AT() macro to conditionally cast

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199317452 28800
# Node ID f7e7db3facd9406545103164f9be8f9ba1a2b549
# Parent  4d9a413a0f4c1d98dbea704f0366457b5117045d
x86: add _AT() macro to conditionally cast

Define _AT(type, value) to conditionally cast a value when compiling C
code, but not when used in assembler.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/const.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/linux/const.h b/include/linux/const.h
index 07b300bfe34b..c22c707c455d 100644
--- a/include/linux/const.h
+++ b/include/linux/const.h
@@ -7,13 +7,18 @@
  * C code.  Therefore we cannot annotate them always with
  * 'UL' and other type specifiers unilaterally.  We
  * use the following macros to deal with this.
+ *
+ * Similarly, _AT() will cast an expression with a type in C, but
+ * leave it unchanged in asm.
  */
 
 #ifdef __ASSEMBLY__
 #define _AC(X,Y)	X
+#define _AT(T,X)	X
 #else
 #define __AC(X,Y)	(X##Y)
 #define _AC(X,Y)	__AC(X,Y)
+#define _AT(T,X)	((T)(X))
 #endif
 
 #endif /* !(_LINUX_CONST_H) */
-- 
cgit v1.2.3


From 3da1bcc2659de27094592e5a037d0b1d59351e03 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:42 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #1

based on:

 Subject: x86: page.h: move and unify types for pagetable entry
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 2ebb6977e00c..911d7e0b3cd9 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -9,6 +9,7 @@
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
 #define PHYSICAL_PAGE_MASK	(PAGE_MASK & __PHYSICAL_MASK)
+#define PTE_MASK		PHYSICAL_PAGE_MASK
 
 #define LARGE_PAGE_SIZE		(_AC(1,UL) << PMD_SHIFT)
 #define LARGE_PAGE_MASK		(~(LARGE_PAGE_SIZE-1))
@@ -24,8 +25,13 @@
 #define __PHYSICAL_MASK		((_AC(1,UL) << __PHYSICAL_MASK_SHIFT) - 1)
 #define __VIRTUAL_MASK		((_AC(1,UL) << __VIRTUAL_MASK_SHIFT) - 1)
 
+#ifndef __ASSEMBLY__
+#include <linux/types.h>
+#endif
 
 #ifdef CONFIG_X86_64
+#define PAGETABLE_LEVELS	4
+
 #define THREAD_ORDER	1
 #define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
 #define CURRENT_MASK (~(THREAD_SIZE-1))
-- 
cgit v1.2.3


From 7a2389b457a03f226b441b5c53f9808d4eefdcea Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:42 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #2

based on:

 Subject: x86: page.h: move and unify types for pagetable entry
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_64.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 5c252d9fee1b..3f280105d7b3 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -79,12 +79,12 @@ static inline void set_pte(pte_t *dst, pte_t val)
 
 static inline void set_pmd(pmd_t *dst, pmd_t val)
 {
-        pmd_val(*dst) = pmd_val(val); 
+	*dst = val;
 } 
 
 static inline void set_pud(pud_t *dst, pud_t val)
 {
-	pud_val(*dst) = pud_val(val);
+	*dst = val;
 }
 
 static inline void pud_clear (pud_t *pud)
@@ -94,7 +94,7 @@ static inline void pud_clear (pud_t *pud)
 
 static inline void set_pgd(pgd_t *dst, pgd_t val)
 {
-	pgd_val(*dst) = pgd_val(val); 
+	*dst = val;
 } 
 
 static inline void pgd_clear (pgd_t * pgd)
-- 
cgit v1.2.3


From ba2b6c525eec459e1c6397f0f3eaeeafad153eab Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:42 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #3

based on:

 Subject: x86: page.h: move and unify types for pagetable entry
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 13 +++++++++++++
 include/asm-x86/page_64.h |  1 -
 2 files changed, 13 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 911d7e0b3cd9..a522fbada97e 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -79,6 +79,19 @@
 #ifndef __ASSEMBLY__
 void clear_page(void *page);
 void copy_page(void *to, void *from);
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef unsigned long	pteval_t;
+typedef unsigned long	pmdval_t;
+typedef unsigned long	pudval_t;
+typedef unsigned long	pgdval_t;
+typedef unsigned long	pgprotval_t;
+typedef unsigned long	phys_addr_t;
+
+typedef struct { pteval_t pte; } pte_t;
+
 #endif	/* !__ASSEMBLY__ */
 
 #endif	/* CONFIG_X86_64 */
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index f5e0543b8afa..f5e82c695d26 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -10,7 +10,6 @@ extern unsigned long end_pfn_map;
 /*
  * These are used to make use of C type-checking..
  */
-typedef struct { unsigned long pte; } pte_t;
 typedef struct { unsigned long pmd; } pmd_t;
 typedef struct { unsigned long pud; } pud_t;
 typedef struct { unsigned long pgd; } pgd_t;
-- 
cgit v1.2.3


From d6e3cf63e927310f65e21132f1d1c50e3aa4c2b0 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:43 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #4

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index a522fbada97e..530439e0d340 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -92,6 +92,9 @@ typedef unsigned long	phys_addr_t;
 
 typedef struct { pteval_t pte; } pte_t;
 
+#define native_pte_val(x)	((x).pte)
+#define native_make_pte(x) ((pte_t) { (x) } )
+
 #endif	/* !__ASSEMBLY__ */
 
 #endif	/* CONFIG_X86_64 */
-- 
cgit v1.2.3


From 38f0f12793a490ac633dbba2418172b7abfa077e Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:43 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #5

based on:

 Subject: x86: page.h: move and unify types for pagetable entry
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 72 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/page_64.h | 21 --------------
 2 files changed, 72 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 530439e0d340..014f087d7c27 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -181,6 +181,78 @@ static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
 	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
 #define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
 
+#ifdef CONFIG_X86_64
+typedef struct { pgdval_t pgd; } pgd_t;
+typedef struct { pgprotval_t pgprot; } pgprot_t;
+
+static inline pgd_t native_make_pgd(pgdval_t val)
+{
+	return (pgd_t) { val };
+}
+
+static inline pgdval_t native_pgd_val(pgd_t pgd)
+{
+	return pgd.pgd;
+}
+
+#if PAGETABLE_LEVELS >= 3
+#if PAGETABLE_LEVELS == 4
+typedef struct { pudval_t pud; } pud_t;
+
+static inline pud_t native_make_pud(pmdval_t val)
+{
+	return (pud_t) { val };
+}
+
+static inline pudval_t native_pud_val(pud_t pud)
+{
+	return pud.pud;
+}
+#else	/* PAGETABLE_LEVELS == 3 */
+#include <asm-generic/pgtable-nopud.h>
+#endif	/* PAGETABLE_LEVELS == 4 */
+
+typedef struct { pmdval_t pmd; } pmd_t;
+
+static inline pmd_t native_make_pmd(pmdval_t val)
+{
+	return (pmd_t) { val };
+}
+
+static inline pmdval_t native_pmd_val(pmd_t pmd)
+{
+	return pmd.pmd;
+}
+#else  /* PAGETABLE_LEVELS == 2 */
+#include <asm-generic/pgtable-nopmd.h>
+#endif	/* PAGETABLE_LEVELS >= 3 */
+
+#define pgprot_val(x)	((x).pgprot)
+#define __pgprot(x)	((pgprot_t) { (x) } )
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else  /* !CONFIG_PARAVIRT */
+
+#define pgd_val(x)	native_pgd_val(x)
+#define __pgd(x)	native_make_pgd(x)
+
+#ifndef __PAGETABLE_PUD_FOLDED
+#define pud_val(x)	native_pud_val(x)
+#define __pud(x)	native_make_pud(x)
+#endif
+
+#ifndef __PAGETABLE_PMD_FOLDED
+#define pmd_val(x)	native_pmd_val(x)
+#define __pmd(x)	native_make_pmd(x)
+#endif
+
+#define pte_val(x)	native_pte_val(x)
+#define __pte(x)	native_make_pte(x)
+
+#endif	/* CONFIG_PARAVIRT */
+
+#endif /* CONFIG_X86_64 */
 #endif	/* __ASSEMBLY__ */
 
 
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index f5e82c695d26..f8a07bf54c3c 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -7,30 +7,9 @@
 extern unsigned long end_pfn;
 extern unsigned long end_pfn_map;
 
-/*
- * These are used to make use of C type-checking..
- */
-typedef struct { unsigned long pmd; } pmd_t;
-typedef struct { unsigned long pud; } pud_t;
-typedef struct { unsigned long pgd; } pgd_t;
-#define PTE_MASK	PHYSICAL_PAGE_MASK
-
-typedef struct { unsigned long pgprot; } pgprot_t;
 
 extern unsigned long phys_base;
 
-#define pte_val(x)	((x).pte)
-#define pmd_val(x)	((x).pmd)
-#define pud_val(x)	((x).pud)
-#define pgd_val(x)	((x).pgd)
-#define pgprot_val(x)	((x).pgprot)
-
-#define __pte(x) ((pte_t) { (x) } )
-#define __pmd(x) ((pmd_t) { (x) } )
-#define __pud(x) ((pud_t) { (x) } )
-#define __pgd(x) ((pgd_t) { (x) } )
-#define __pgprot(x)	((pgprot_t) { (x) } )
-
 #endif /* !__ASSEMBLY__ */
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From 881d90d0daaeac018b0d5beb739dd825ccee0143 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:43 +0100
Subject: x86: page.h: move and unify types for pagetable entry

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199319654 28800
# Node ID 3bd7db6e85e66e7f3362874802df26a82fcb2d92
# Parent  f7e7db3facd9406545103164f9be8f9ba1a2b549
x86: page.h: move and unify types for pagetable entry definitions

This patch:

1. Defines arch-specific types for the contents of a pagetable entry.
That is, 32-bit entries for 32-bit non-PAE, and 64-bit entries for
32-bit PAE and 64-bit.  However, even though the latter two are the
same size, they're defined with different types in order to retain
compatibility with printk format strings, etc.

2. Defines arch-specific pte_t.  This is different because 32-bit PAE
defines it in two halves, whereas 32-bit PAE and 64-bit define it as a
single entry.  All the other pagetable levels can be defined in a
common way.  This also defines arch-specific pte_val/make_pte functions.

3. Define PAGETABLE_LEVELS for each architecture variation, for later use.

4. Define common pagetable entry accessors in a paravirt-compatible
way. (64-bit does not yet use paravirt-ops in any way).

5. Convert a few instances of using a *_val() as an lvalue where it is
no longer a macro.  There are still places in the 64-bit code which
use pte_val() as an lvalue.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 50 ++++++++++++++++++++++++++--
 include/asm-x86/page_32.h | 83 -----------------------------------------------
 2 files changed, 48 insertions(+), 85 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 014f087d7c27..00353eb5666d 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -116,9 +116,57 @@ typedef struct { pteval_t pte; } pte_t;
 #ifdef CONFIG_X86_PAE
 #define __PHYSICAL_MASK_SHIFT	36
 #define __VIRTUAL_MASK_SHIFT	32
+#define PAGETABLE_LEVELS	3
+
+#ifndef __ASSEMBLY__
+typedef u64	pteval_t;
+typedef u64	pmdval_t;
+typedef u64	pudval_t;
+typedef u64	pgdval_t;
+typedef u64	pgprotval_t;
+typedef u64	phys_addr_t;
+
+typedef struct { unsigned long pte_low, pte_high; } pte_t;
+
+static inline unsigned long long native_pte_val(pte_t pte)
+{
+	return pte.pte_low | ((unsigned long long)pte.pte_high << 32);
+}
+
+static inline pte_t native_make_pte(unsigned long long val)
+{
+	return (pte_t) { .pte_low = val, .pte_high = (val >> 32) } ;
+}
+
+#endif	/* __ASSEMBLY__
+ */
 #else  /* !CONFIG_X86_PAE */
 #define __PHYSICAL_MASK_SHIFT	32
 #define __VIRTUAL_MASK_SHIFT	32
+#define PAGETABLE_LEVELS	2
+
+#ifndef __ASSEMBLY__
+typedef unsigned long	pteval_t;
+typedef unsigned long	pmdval_t;
+typedef unsigned long	pudval_t;
+typedef unsigned long	pgdval_t;
+typedef unsigned long	pgprotval_t;
+typedef unsigned long	phys_addr_t;
+
+typedef struct { pteval_t pte_low; } pte_t;
+typedef pte_t boot_pte_t;
+
+static inline unsigned long native_pte_val(pte_t pte)
+{
+	return pte.pte_low;
+}
+
+static inline pte_t native_make_pte(unsigned long val)
+{
+	return (pte_t) { .pte_low = val };
+}
+
+#endif	/* __ASSEMBLY__ */
 #endif	/* CONFIG_X86_PAE */
 
 #ifdef CONFIG_HUGETLB_PAGE
@@ -181,7 +229,6 @@ static void inline copy_user_page(void *to, void *from, unsigned long vaddr,
 	alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
 #define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
 
-#ifdef CONFIG_X86_64
 typedef struct { pgdval_t pgd; } pgd_t;
 typedef struct { pgprotval_t pgprot; } pgprot_t;
 
@@ -252,7 +299,6 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 
 #endif	/* CONFIG_PARAVIRT */
 
-#endif /* CONFIG_X86_64 */
 #endif	/* __ASSEMBLY__ */
 
 
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index b4750ef44920..2d60661ab340 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -9,89 +9,6 @@
  */
 extern int nx_enabled;
 
-#ifdef CONFIG_X86_PAE
-typedef struct { unsigned long pte_low, pte_high; } pte_t;
-typedef struct { unsigned long long pmd; } pmd_t;
-typedef struct { unsigned long long pgd; } pgd_t;
-typedef struct { unsigned long long pgprot; } pgprot_t;
-
-static inline unsigned long long native_pgd_val(pgd_t pgd)
-{
-	return pgd.pgd;
-}
-
-static inline unsigned long long native_pmd_val(pmd_t pmd)
-{
-	return pmd.pmd;
-}
-
-static inline unsigned long long native_pte_val(pte_t pte)
-{
-	return pte.pte_low | ((unsigned long long)pte.pte_high << 32);
-}
-
-static inline pgd_t native_make_pgd(unsigned long long val)
-{
-	return (pgd_t) { val };
-}
-
-static inline pmd_t native_make_pmd(unsigned long long val)
-{
-	return (pmd_t) { val };
-}
-
-static inline pte_t native_make_pte(unsigned long long val)
-{
-	return (pte_t) { .pte_low = val, .pte_high = (val >> 32) } ;
-}
-
-#ifndef CONFIG_PARAVIRT
-#define pmd_val(x)	native_pmd_val(x)
-#define __pmd(x)	native_make_pmd(x)
-#endif
-
-#include <asm-generic/pgtable-nopud.h>
-#else  /* !CONFIG_X86_PAE */
-typedef struct { unsigned long pte_low; } pte_t;
-typedef struct { unsigned long pgd; } pgd_t;
-typedef struct { unsigned long pgprot; } pgprot_t;
-#define boot_pte_t pte_t /* or would you rather have a typedef */
-
-static inline unsigned long native_pgd_val(pgd_t pgd)
-{
-	return pgd.pgd;
-}
-
-static inline unsigned long native_pte_val(pte_t pte)
-{
-	return pte.pte_low;
-}
-
-static inline pgd_t native_make_pgd(unsigned long val)
-{
-	return (pgd_t) { val };
-}
-
-static inline pte_t native_make_pte(unsigned long val)
-{
-	return (pte_t) { .pte_low = val };
-}
-
-#include <asm-generic/pgtable-nopmd.h>
-#endif	/* CONFIG_X86_PAE */
-
-#define PTE_MASK	PAGE_MASK
-
-#define pgprot_val(x)	((x).pgprot)
-#define __pgprot(x)	((pgprot_t) { (x) } )
-
-#ifndef CONFIG_PARAVIRT
-#define pgd_val(x)	native_pgd_val(x)
-#define __pgd(x)	native_make_pgd(x)
-#define pte_val(x)	native_pte_val(x)
-#define __pte(x)	native_make_pte(x)
-#endif
-
 #endif /* !__ASSEMBLY__ */
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From 6724a1d2fc6b40d55f232b8dacc12a915878bdc7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:43 +0100
Subject: x86: page.h: move and unify types for pagetable entry, #6

based on:

 Subject: x86: page.h: move and unify types for pagetable entry
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 00353eb5666d..ef393c3df902 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -22,7 +22,7 @@
 /* to align the pointer to the (next) page boundary */
 #define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
 
-#define __PHYSICAL_MASK		((_AC(1,UL) << __PHYSICAL_MASK_SHIFT) - 1)
+#define __PHYSICAL_MASK		_AT(phys_addr_t, (_AC(1,ULL) << __PHYSICAL_MASK_SHIFT) - 1)
 #define __VIRTUAL_MASK		((_AC(1,UL) << __VIRTUAL_MASK_SHIFT) - 1)
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From 98fd5aee348f0420afd1c636790d50aaaec6ceec Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:43 +0100
Subject: x86: page.h: move pa and va related things

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199319656 28800
# Node ID d617b72a0cc9d14bde2087d065c36d4ed3265761
# Parent  3bd7db6e85e66e7f3362874802df26a82fcb2d92
x86: page.h: move pa and va related things

Move and unify the virtual<->physical address space conversion
functions.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 24 ++++++++++++++++++++++++
 include/asm-x86/page_32.h | 13 +------------
 include/asm-x86/page_64.h | 13 -------------
 3 files changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index ef393c3df902..9830eeb57173 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -80,6 +80,9 @@
 void clear_page(void *page);
 void copy_page(void *to, void *from);
 
+extern unsigned long __phys_addr(unsigned long);
+#define __phys_reloc_hide(x)	(x)
+
 /*
  * These are used to make use of C type-checking..
  */
@@ -174,6 +177,13 @@ static inline pte_t native_make_pte(unsigned long val)
 #endif
 
 #ifndef __ASSEMBLY__
+#define __phys_addr(x)		((x)-PAGE_OFFSET)
+#define __phys_reloc_hide(x)	RELOC_HIDE((x), 0)
+
+#ifdef CONFIG_FLATMEM
+#define pfn_valid(pfn)		((pfn) < max_mapnr)
+#endif /* CONFIG_FLATMEM */
+
 #ifdef CONFIG_X86_USE_3DNOW
 #include <asm/mmx.h>
 
@@ -299,6 +309,20 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 
 #endif	/* CONFIG_PARAVIRT */
 
+#define __pa(x)		__phys_addr((unsigned long)(x))
+/* __pa_symbol should be used for C visible symbols.
+   This seems to be the official gcc blessed way to do such arithmetic. */
+#define __pa_symbol(x)	__pa(__phys_reloc_hide((unsigned long)(x)))
+
+#define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
+
+#define __boot_va(x)		__va(x)
+#define __boot_pa(x)		__pa(x)
+
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+#define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+
 #endif	/* __ASSEMBLY__ */
 
 
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 2d60661ab340..a27424a2ee0d 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -29,18 +29,7 @@ extern int page_is_ram(unsigned long pagenr);
 
 #define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
 #define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
-#define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
-/* __pa_symbol should be used for C visible symbols.
-   This seems to be the official gcc blessed way to do such arithmetic. */
-#define __pa_symbol(x)          __pa(RELOC_HIDE((unsigned long)(x),0))
-#define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
-#define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
-#ifdef CONFIG_FLATMEM
-#define pfn_valid(pfn)		((pfn) < max_mapnr)
-#endif /* CONFIG_FLATMEM */
-#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
-
-#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+
 
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index f8a07bf54c3c..7094684270a6 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -16,21 +16,8 @@ extern unsigned long phys_base;
 
 #include <asm/bug.h>
 
-extern unsigned long __phys_addr(unsigned long);
-
 #endif /* __ASSEMBLY__ */
 
-#define __pa(x)		__phys_addr((unsigned long)(x))
-#define __pa_symbol(x)	__phys_addr((unsigned long)(x))
-
-#define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
-#define __boot_va(x)		__va(x)
-#define __boot_pa(x)		__pa(x)
-
-#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
-#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
-#define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
-
 #define __HAVE_ARCH_GATE_AREA 1	
 #define vmemmap ((struct page *)VMEMMAP_START)
 
-- 
cgit v1.2.3


From e62f4473f32a882a537b32cb7202da8d5d7c4f1f Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:44 +0100
Subject: x86: page.h: move remaining bits and pieces

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199319657 28800
# Node ID bba9287641ff90e836d090d80b5c0a846aab7162
# Parent  d617b72a0cc9d14bde2087d065c36d4ed3265761
x86: page.h: move remaining bits and pieces

Move the remaining odds and ends into page.h.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h    | 23 +++++++++++++++++++++++
 include/asm-x86/page_32.h | 17 -----------------
 include/asm-x86/page_64.h | 13 -------------
 3 files changed, 23 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 9830eeb57173..f65a2ae6e323 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -80,6 +80,10 @@
 void clear_page(void *page);
 void copy_page(void *to, void *from);
 
+extern unsigned long end_pfn;
+extern unsigned long end_pfn_map;
+extern unsigned long phys_base;
+
 extern unsigned long __phys_addr(unsigned long);
 #define __phys_reloc_hide(x)	(x)
 
@@ -98,6 +102,8 @@ typedef struct { pteval_t pte; } pte_t;
 #define native_pte_val(x)	((x).pte)
 #define native_make_pte(x) ((pte_t) { (x) } )
 
+#define vmemmap ((struct page *)VMEMMAP_START)
+
 #endif	/* !__ASSEMBLY__ */
 
 #endif	/* CONFIG_X86_64 */
@@ -184,6 +190,19 @@ static inline pte_t native_make_pte(unsigned long val)
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
 #endif /* CONFIG_FLATMEM */
 
+extern int nx_enabled;
+
+/*
+ * This much address space is reserved for vmalloc() and iomap()
+ * as well as fixmap mappings.
+ */
+extern unsigned int __VMALLOC_RESERVE;
+extern int sysctl_legacy_va_layout;
+extern int page_is_ram(unsigned long pagenr);
+
+#define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
+#define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
+
 #ifdef CONFIG_X86_USE_3DNOW
 #include <asm/mmx.h>
 
@@ -325,6 +344,10 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 
 #endif	/* __ASSEMBLY__ */
 
+#include <asm-generic/memory_model.h>
+#include <asm-generic/page.h>
+
+#define __HAVE_ARCH_GATE_AREA 1
 
 #ifdef CONFIG_X86_32
 # include "page_32.h"
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index a27424a2ee0d..c27b41fd0254 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -7,7 +7,6 @@
 /*
  * These are used to make use of C type-checking..
  */
-extern int nx_enabled;
 
 #endif /* !__ASSEMBLY__ */
 
@@ -15,26 +14,10 @@ extern int nx_enabled;
 
 struct vm_area_struct;
 
-/*
- * This much address space is reserved for vmalloc() and iomap()
- * as well as fixmap mappings.
- */
-extern unsigned int __VMALLOC_RESERVE;
-
-extern int sysctl_legacy_va_layout;
-
-extern int page_is_ram(unsigned long pagenr);
-
 #endif /* __ASSEMBLY__ */
 
-#define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
-#define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
-
 
-#include <asm-generic/memory_model.h>
-#include <asm-generic/page.h>
 
-#define __HAVE_ARCH_GATE_AREA 1
 #endif /* __KERNEL__ */
 
 #endif /* _I386_PAGE_H */
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 7094684270a6..94d4803aab2d 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -4,26 +4,13 @@
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
-extern unsigned long end_pfn;
-extern unsigned long end_pfn_map;
-
-
-extern unsigned long phys_base;
-
 #endif /* !__ASSEMBLY__ */
 
 #ifndef __ASSEMBLY__
 
-#include <asm/bug.h>
 
 #endif /* __ASSEMBLY__ */
 
-#define __HAVE_ARCH_GATE_AREA 1	
-#define vmemmap ((struct page *)VMEMMAP_START)
-
-#include <asm-generic/memory_model.h>
-#include <asm-generic/page.h>
-
 #endif /* __KERNEL__ */
 
 #ifdef CONFIG_FLATMEM
-- 
cgit v1.2.3


From 11b7c7dc945b8802c459baa3304c1b501d2bf872 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:44 +0100
Subject: x86: page.h: move things back to their own files

# HG changeset patch
# User Jeremy Fitzhardinge <jeremy@xensource.com>
# Date 1199321648 28800
# Node ID 22f6a5902285b58bfc1fbbd9e183498c9017bd78
# Parent  bba9287641ff90e836d090d80b5c0a846aab7162
x86: page.h: move things back to their own files

Oops, asm/page.h has turned into an #ifdef hellhole.  Move
32/64-specific things back to their own headers to make it somewhat
comprehensible...

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/Kbuild    |   3 +-
 include/asm-x86/page.h    | 212 ++--------------------------------------------
 include/asm-x86/page_32.h | 121 +++++++++++++++++++++++---
 include/asm-x86/page_64.h |  74 ++++++++++++++--
 4 files changed, 187 insertions(+), 223 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index 22e3a284c1ed..e6189b229143 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -15,8 +15,7 @@ unifdef-y += ist.h
 unifdef-y += mce.h
 unifdef-y += msr.h
 unifdef-y += mtrr.h
-unifdef-y += page_32.h
-unifdef-y += page_64.h
+unifdef-y += page.h
 unifdef-y += posix_types_32.h
 unifdef-y += posix_types_64.h
 unifdef-y += ptrace.h
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index f65a2ae6e323..3a00d0d9dfd3 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -8,6 +8,8 @@
 #define PAGE_SIZE	(_AC(1,UL) << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
+#ifdef __KERNEL__
+
 #define PHYSICAL_PAGE_MASK	(PAGE_MASK & __PHYSICAL_MASK)
 #define PTE_MASK		PHYSICAL_PAGE_MASK
 
@@ -30,208 +32,11 @@
 #endif
 
 #ifdef CONFIG_X86_64
-#define PAGETABLE_LEVELS	4
-
-#define THREAD_ORDER	1
-#define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
-#define CURRENT_MASK (~(THREAD_SIZE-1))
-
-#define EXCEPTION_STACK_ORDER 0
-#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
-
-#define DEBUG_STACK_ORDER (EXCEPTION_STACK_ORDER + 1)
-#define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER)
-
-#define IRQSTACK_ORDER 2
-#define IRQSTACKSIZE (PAGE_SIZE << IRQSTACK_ORDER)
-
-#define STACKFAULT_STACK 1
-#define DOUBLEFAULT_STACK 2
-#define NMI_STACK 3
-#define DEBUG_STACK 4
-#define MCE_STACK 5
-#define N_EXCEPTION_STACKS 5  /* hw limit: 7 */
-
-#define __PAGE_OFFSET           _AC(0xffff810000000000, UL)
-
-#define __PHYSICAL_START	CONFIG_PHYSICAL_START
-#define __KERNEL_ALIGN		0x200000
-
-/*
- * Make sure kernel is aligned to 2MB address. Catching it at compile
- * time is better. Change your config file and compile the kernel
- * for a 2MB aligned address (CONFIG_PHYSICAL_START)
- */
-#if (CONFIG_PHYSICAL_START % __KERNEL_ALIGN) != 0
-#error "CONFIG_PHYSICAL_START must be a multiple of 2MB"
-#endif
-
-#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
-#define __START_KERNEL_map	_AC(0xffffffff80000000, UL)
-
-/* See Documentation/x86_64/mm.txt for a description of the memory map. */
-#define __PHYSICAL_MASK_SHIFT	46
-#define __VIRTUAL_MASK_SHIFT	48
-
-#define KERNEL_TEXT_SIZE  (40*1024*1024)
-#define KERNEL_TEXT_START _AC(0xffffffff80000000, UL)
-
-#ifndef __ASSEMBLY__
-void clear_page(void *page);
-void copy_page(void *to, void *from);
-
-extern unsigned long end_pfn;
-extern unsigned long end_pfn_map;
-extern unsigned long phys_base;
-
-extern unsigned long __phys_addr(unsigned long);
-#define __phys_reloc_hide(x)	(x)
-
-/*
- * These are used to make use of C type-checking..
- */
-typedef unsigned long	pteval_t;
-typedef unsigned long	pmdval_t;
-typedef unsigned long	pudval_t;
-typedef unsigned long	pgdval_t;
-typedef unsigned long	pgprotval_t;
-typedef unsigned long	phys_addr_t;
-
-typedef struct { pteval_t pte; } pte_t;
-
-#define native_pte_val(x)	((x).pte)
-#define native_make_pte(x) ((pte_t) { (x) } )
-
-#define vmemmap ((struct page *)VMEMMAP_START)
-
-#endif	/* !__ASSEMBLY__ */
-
+#include <asm/page_64.h>
+#else
+#include <asm/page_32.h>
 #endif	/* CONFIG_X86_64 */
 
-#ifdef CONFIG_X86_32
-
-/*
- * This handles the memory map.
- *
- * A __PAGE_OFFSET of 0xC0000000 means that the kernel has
- * a virtual address space of one gigabyte, which limits the
- * amount of physical memory you can use to about 950MB.
- *
- * If you want more physical memory than this then see the CONFIG_HIGHMEM4G
- * and CONFIG_HIGHMEM64G options in the kernel configuration.
- */
-#define __PAGE_OFFSET		_AC(CONFIG_PAGE_OFFSET, UL)
-
-#ifdef CONFIG_X86_PAE
-#define __PHYSICAL_MASK_SHIFT	36
-#define __VIRTUAL_MASK_SHIFT	32
-#define PAGETABLE_LEVELS	3
-
-#ifndef __ASSEMBLY__
-typedef u64	pteval_t;
-typedef u64	pmdval_t;
-typedef u64	pudval_t;
-typedef u64	pgdval_t;
-typedef u64	pgprotval_t;
-typedef u64	phys_addr_t;
-
-typedef struct { unsigned long pte_low, pte_high; } pte_t;
-
-static inline unsigned long long native_pte_val(pte_t pte)
-{
-	return pte.pte_low | ((unsigned long long)pte.pte_high << 32);
-}
-
-static inline pte_t native_make_pte(unsigned long long val)
-{
-	return (pte_t) { .pte_low = val, .pte_high = (val >> 32) } ;
-}
-
-#endif	/* __ASSEMBLY__
- */
-#else  /* !CONFIG_X86_PAE */
-#define __PHYSICAL_MASK_SHIFT	32
-#define __VIRTUAL_MASK_SHIFT	32
-#define PAGETABLE_LEVELS	2
-
-#ifndef __ASSEMBLY__
-typedef unsigned long	pteval_t;
-typedef unsigned long	pmdval_t;
-typedef unsigned long	pudval_t;
-typedef unsigned long	pgdval_t;
-typedef unsigned long	pgprotval_t;
-typedef unsigned long	phys_addr_t;
-
-typedef struct { pteval_t pte_low; } pte_t;
-typedef pte_t boot_pte_t;
-
-static inline unsigned long native_pte_val(pte_t pte)
-{
-	return pte.pte_low;
-}
-
-static inline pte_t native_make_pte(unsigned long val)
-{
-	return (pte_t) { .pte_low = val };
-}
-
-#endif	/* __ASSEMBLY__ */
-#endif	/* CONFIG_X86_PAE */
-
-#ifdef CONFIG_HUGETLB_PAGE
-#define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
-#endif
-
-#ifndef __ASSEMBLY__
-#define __phys_addr(x)		((x)-PAGE_OFFSET)
-#define __phys_reloc_hide(x)	RELOC_HIDE((x), 0)
-
-#ifdef CONFIG_FLATMEM
-#define pfn_valid(pfn)		((pfn) < max_mapnr)
-#endif /* CONFIG_FLATMEM */
-
-extern int nx_enabled;
-
-/*
- * This much address space is reserved for vmalloc() and iomap()
- * as well as fixmap mappings.
- */
-extern unsigned int __VMALLOC_RESERVE;
-extern int sysctl_legacy_va_layout;
-extern int page_is_ram(unsigned long pagenr);
-
-#define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
-#define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
-
-#ifdef CONFIG_X86_USE_3DNOW
-#include <asm/mmx.h>
-
-static inline void clear_page(void *page)
-{
-	mmx_clear_page(page);
-}
-
-static inline void copy_page(void *to, void *from)
-{
-	mmx_copy_page(to, from);
-}
-#else  /* !CONFIG_X86_USE_3DNOW */
-#include <linux/string.h>
-
-static inline void clear_page(void *page)
-{
-	memset(page, 0, PAGE_SIZE);
-}
-
-static inline void copy_page(void *to, void *from)
-{
-	memcpy(to, from, PAGE_SIZE);
-}
-#endif	/* CONFIG_X86_3DNOW */
-#endif	/* !__ASSEMBLY__ */
-
-#endif	/* CONFIG_X86_32 */
-
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
 
 #define VM_DATA_DEFAULT_FLAGS \
@@ -349,10 +154,5 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 
 #define __HAVE_ARCH_GATE_AREA 1
 
-#ifdef CONFIG_X86_32
-# include "page_32.h"
-#else
-# include "page_64.h"
-#endif
-
+#endif	/* __KERNEL__ */
 #endif	/* _ASM_X86_PAGE_H */
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index c27b41fd0254..e424bef4395a 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -1,23 +1,124 @@
-#ifndef _I386_PAGE_H
-#define _I386_PAGE_H
+#ifndef _ASM_X86_PAGE_32_H
+#define _ASM_X86_PAGE_32_H
+
+/*
+ * This handles the memory map.
+ *
+ * A __PAGE_OFFSET of 0xC0000000 means that the kernel has
+ * a virtual address space of one gigabyte, which limits the
+ * amount of physical memory you can use to about 950MB.
+ *
+ * If you want more physical memory than this then see the CONFIG_HIGHMEM4G
+ * and CONFIG_HIGHMEM64G options in the kernel configuration.
+ */
+#define __PAGE_OFFSET		_AC(CONFIG_PAGE_OFFSET, UL)
+
+#ifdef CONFIG_X86_PAE
+#define __PHYSICAL_MASK_SHIFT	36
+#define __VIRTUAL_MASK_SHIFT	32
+#define PAGETABLE_LEVELS	3
 
-#ifdef __KERNEL__
 #ifndef __ASSEMBLY__
+typedef u64	pteval_t;
+typedef u64	pmdval_t;
+typedef u64	pudval_t;
+typedef u64	pgdval_t;
+typedef u64	pgprotval_t;
+typedef u64	phys_addr_t;
 
-/*
- * These are used to make use of C type-checking..
+typedef struct { unsigned long pte_low, pte_high; } pte_t;
+
+static inline unsigned long long native_pte_val(pte_t pte)
+{
+	return pte.pte_low | ((unsigned long long)pte.pte_high << 32);
+}
+
+static inline pte_t native_make_pte(unsigned long long val)
+{
+	return (pte_t) { .pte_low = val, .pte_high = (val >> 32) } ;
+}
+
+#endif	/* __ASSEMBLY__
  */
+#else  /* !CONFIG_X86_PAE */
+#define __PHYSICAL_MASK_SHIFT	32
+#define __VIRTUAL_MASK_SHIFT	32
+#define PAGETABLE_LEVELS	2
+
+#ifndef __ASSEMBLY__
+typedef unsigned long	pteval_t;
+typedef unsigned long	pmdval_t;
+typedef unsigned long	pudval_t;
+typedef unsigned long	pgdval_t;
+typedef unsigned long	pgprotval_t;
+typedef unsigned long	phys_addr_t;
 
-#endif /* !__ASSEMBLY__ */
+typedef struct { pteval_t pte_low; } pte_t;
+typedef pte_t boot_pte_t;
+
+static inline unsigned long native_pte_val(pte_t pte)
+{
+	return pte.pte_low;
+}
+
+static inline pte_t native_make_pte(unsigned long val)
+{
+	return (pte_t) { .pte_low = val };
+}
+
+#endif	/* __ASSEMBLY__ */
+#endif	/* CONFIG_X86_PAE */
+
+#ifdef CONFIG_HUGETLB_PAGE
+#define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
+#endif
 
 #ifndef __ASSEMBLY__
+#define __phys_addr(x)		((x)-PAGE_OFFSET)
+#define __phys_reloc_hide(x)	RELOC_HIDE((x), 0)
+
+#ifdef CONFIG_FLATMEM
+#define pfn_valid(pfn)		((pfn) < max_mapnr)
+#endif /* CONFIG_FLATMEM */
+
+extern int nx_enabled;
+
+/*
+ * This much address space is reserved for vmalloc() and iomap()
+ * as well as fixmap mappings.
+ */
+extern unsigned int __VMALLOC_RESERVE;
+extern int sysctl_legacy_va_layout;
+extern int page_is_ram(unsigned long pagenr);
+
+#define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
+#define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
 
-struct vm_area_struct;
+#ifdef CONFIG_X86_USE_3DNOW
+#include <asm/mmx.h>
 
-#endif /* __ASSEMBLY__ */
+static inline void clear_page(void *page)
+{
+	mmx_clear_page(page);
+}
 
+static inline void copy_page(void *to, void *from)
+{
+	mmx_copy_page(to, from);
+}
+#else  /* !CONFIG_X86_USE_3DNOW */
+#include <linux/string.h>
 
+static inline void clear_page(void *page)
+{
+	memset(page, 0, PAGE_SIZE);
+}
 
-#endif /* __KERNEL__ */
+static inline void copy_page(void *to, void *from)
+{
+	memcpy(to, from, PAGE_SIZE);
+}
+#endif	/* CONFIG_X86_3DNOW */
+#endif	/* !__ASSEMBLY__ */
 
-#endif /* _I386_PAGE_H */
+#endif /* _ASM_X86_PAGE_32_H */
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 94d4803aab2d..ebf977a809a8 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -1,17 +1,81 @@
 #ifndef _X86_64_PAGE_H
 #define _X86_64_PAGE_H
 
-#ifdef __KERNEL__
-#ifndef __ASSEMBLY__
+#define PAGETABLE_LEVELS	4
+
+#define THREAD_ORDER	1
+#define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
+#define CURRENT_MASK (~(THREAD_SIZE-1))
+
+#define EXCEPTION_STACK_ORDER 0
+#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
+
+#define DEBUG_STACK_ORDER (EXCEPTION_STACK_ORDER + 1)
+#define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER)
+
+#define IRQSTACK_ORDER 2
+#define IRQSTACKSIZE (PAGE_SIZE << IRQSTACK_ORDER)
+
+#define STACKFAULT_STACK 1
+#define DOUBLEFAULT_STACK 2
+#define NMI_STACK 3
+#define DEBUG_STACK 4
+#define MCE_STACK 5
+#define N_EXCEPTION_STACKS 5  /* hw limit: 7 */
+
+#define __PAGE_OFFSET           _AC(0xffff810000000000, UL)
+
+#define __PHYSICAL_START	CONFIG_PHYSICAL_START
+#define __KERNEL_ALIGN		0x200000
 
-#endif /* !__ASSEMBLY__ */
+/*
+ * Make sure kernel is aligned to 2MB address. Catching it at compile
+ * time is better. Change your config file and compile the kernel
+ * for a 2MB aligned address (CONFIG_PHYSICAL_START)
+ */
+#if (CONFIG_PHYSICAL_START % __KERNEL_ALIGN) != 0
+#error "CONFIG_PHYSICAL_START must be a multiple of 2MB"
+#endif
+
+#define __START_KERNEL		(__START_KERNEL_map + __PHYSICAL_START)
+#define __START_KERNEL_map	_AC(0xffffffff80000000, UL)
+
+/* See Documentation/x86_64/mm.txt for a description of the memory map. */
+#define __PHYSICAL_MASK_SHIFT	46
+#define __VIRTUAL_MASK_SHIFT	48
+
+#define KERNEL_TEXT_SIZE  (40*1024*1024)
+#define KERNEL_TEXT_START _AC(0xffffffff80000000, UL)
 
 #ifndef __ASSEMBLY__
+void clear_page(void *page);
+void copy_page(void *to, void *from);
+
+extern unsigned long end_pfn;
+extern unsigned long end_pfn_map;
+extern unsigned long phys_base;
+
+extern unsigned long __phys_addr(unsigned long);
+#define __phys_reloc_hide(x)	(x)
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef unsigned long	pteval_t;
+typedef unsigned long	pmdval_t;
+typedef unsigned long	pudval_t;
+typedef unsigned long	pgdval_t;
+typedef unsigned long	pgprotval_t;
+typedef unsigned long	phys_addr_t;
+
+typedef struct { pteval_t pte; } pte_t;
 
+#define native_pte_val(x)	((x).pte)
+#define native_make_pte(x) ((pte_t) { (x) } )
 
-#endif /* __ASSEMBLY__ */
+#define vmemmap ((struct page *)VMEMMAP_START)
 
-#endif /* __KERNEL__ */
+#endif	/* !__ASSEMBLY__ */
 
 #ifdef CONFIG_FLATMEM
 #define pfn_valid(pfn)          ((pfn) < end_pfn)
-- 
cgit v1.2.3


From 0c2bd5a5e3b9469c6ab2b96c403980f192db0bf7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:49 +0100
Subject: x86: clean up include/asm-x86/calling.h

clean up include/asm-x86/calling.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/calling.h | 194 ++++++++++++++++++++++++----------------------
 1 file changed, 100 insertions(+), 94 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/calling.h b/include/asm-x86/calling.h
index 6f4f63af96e1..f13e62e2cb3e 100644
--- a/include/asm-x86/calling.h
+++ b/include/asm-x86/calling.h
@@ -1,162 +1,168 @@
-/* 
+/*
  * Some macros to handle stack frames in assembly.
- */ 
+ */
 
+#define R15		  0
+#define R14		  8
+#define R13		 16
+#define R12		 24
+#define RBP		 32
+#define RBX		 40
 
-#define R15 0
-#define R14 8
-#define R13 16
-#define R12 24
-#define RBP 32
-#define RBX 40
 /* arguments: interrupts/non tracing syscalls only save upto here*/
-#define R11 48
-#define R10 56	
-#define R9 64
-#define R8 72
-#define RAX 80
-#define RCX 88
-#define RDX 96
-#define RSI 104
-#define RDI 112
-#define ORIG_RAX 120       /* + error_code */ 
-/* end of arguments */ 	
+#define R11		 48
+#define R10		 56
+#define R9		 64
+#define R8		 72
+#define RAX		 80
+#define RCX		 88
+#define RDX		 96
+#define RSI		104
+#define RDI		112
+#define ORIG_RAX	120       /* + error_code */
+/* end of arguments */
+
 /* cpu exception frame or undefined in case of fast syscall. */
-#define RIP 128
-#define CS 136
-#define EFLAGS 144
-#define RSP 152
-#define SS 160
-#define ARGOFFSET R11
-#define SWFRAME ORIG_RAX
+#define RIP		128
+#define CS		136
+#define EFLAGS		144
+#define RSP		152
+#define SS		160
+
+#define ARGOFFSET	R11
+#define SWFRAME		ORIG_RAX
 
-	.macro SAVE_ARGS addskip=0,norcx=0,nor891011=0
-	subq  $9*8+\addskip,%rsp
+	.macro SAVE_ARGS addskip=0, norcx=0, nor891011=0
+	subq  $9*8+\addskip, %rsp
 	CFI_ADJUST_CFA_OFFSET	9*8+\addskip
-	movq  %rdi,8*8(%rsp) 
-	CFI_REL_OFFSET	rdi,8*8
-	movq  %rsi,7*8(%rsp) 
-	CFI_REL_OFFSET	rsi,7*8
-	movq  %rdx,6*8(%rsp)
-	CFI_REL_OFFSET	rdx,6*8
+	movq  %rdi, 8*8(%rsp)
+	CFI_REL_OFFSET	rdi, 8*8
+	movq  %rsi, 7*8(%rsp)
+	CFI_REL_OFFSET	rsi, 7*8
+	movq  %rdx, 6*8(%rsp)
+	CFI_REL_OFFSET	rdx, 6*8
 	.if \norcx
 	.else
-	movq  %rcx,5*8(%rsp)
-	CFI_REL_OFFSET	rcx,5*8
+	movq  %rcx, 5*8(%rsp)
+	CFI_REL_OFFSET	rcx, 5*8
 	.endif
-	movq  %rax,4*8(%rsp) 
-	CFI_REL_OFFSET	rax,4*8
+	movq  %rax, 4*8(%rsp)
+	CFI_REL_OFFSET	rax, 4*8
 	.if \nor891011
 	.else
-	movq  %r8,3*8(%rsp) 
-	CFI_REL_OFFSET	r8,3*8
-	movq  %r9,2*8(%rsp) 
-	CFI_REL_OFFSET	r9,2*8
-	movq  %r10,1*8(%rsp) 
-	CFI_REL_OFFSET	r10,1*8
-	movq  %r11,(%rsp) 
-	CFI_REL_OFFSET	r11,0*8
+	movq  %r8, 3*8(%rsp)
+	CFI_REL_OFFSET	r8,  3*8
+	movq  %r9, 2*8(%rsp)
+	CFI_REL_OFFSET	r9,  2*8
+	movq  %r10, 1*8(%rsp)
+	CFI_REL_OFFSET	r10, 1*8
+	movq  %r11, (%rsp)
+	CFI_REL_OFFSET	r11, 0*8
 	.endif
 	.endm
 
-#define ARG_SKIP 9*8
-	.macro RESTORE_ARGS skiprax=0,addskip=0,skiprcx=0,skipr11=0,skipr8910=0,skiprdx=0
+#define ARG_SKIP	9*8
+
+	.macro RESTORE_ARGS skiprax=0, addskip=0, skiprcx=0, skipr11=0, \
+			    skipr8910=0, skiprdx=0
 	.if \skipr11
 	.else
-	movq (%rsp),%r11
+	movq (%rsp), %r11
 	CFI_RESTORE r11
 	.endif
 	.if \skipr8910
 	.else
-	movq 1*8(%rsp),%r10
+	movq 1*8(%rsp), %r10
 	CFI_RESTORE r10
-	movq 2*8(%rsp),%r9
+	movq 2*8(%rsp), %r9
 	CFI_RESTORE r9
-	movq 3*8(%rsp),%r8
+	movq 3*8(%rsp), %r8
 	CFI_RESTORE r8
 	.endif
 	.if \skiprax
 	.else
-	movq 4*8(%rsp),%rax
+	movq 4*8(%rsp), %rax
 	CFI_RESTORE rax
 	.endif
 	.if \skiprcx
 	.else
-	movq 5*8(%rsp),%rcx
+	movq 5*8(%rsp), %rcx
 	CFI_RESTORE rcx
 	.endif
 	.if \skiprdx
 	.else
-	movq 6*8(%rsp),%rdx
+	movq 6*8(%rsp), %rdx
 	CFI_RESTORE rdx
 	.endif
-	movq 7*8(%rsp),%rsi
+	movq 7*8(%rsp), %rsi
 	CFI_RESTORE rsi
-	movq 8*8(%rsp),%rdi
+	movq 8*8(%rsp), %rdi
 	CFI_RESTORE rdi
 	.if ARG_SKIP+\addskip > 0
-	addq $ARG_SKIP+\addskip,%rsp
+	addq $ARG_SKIP+\addskip, %rsp
 	CFI_ADJUST_CFA_OFFSET	-(ARG_SKIP+\addskip)
 	.endif
-	.endm	
+	.endm
 
 	.macro LOAD_ARGS offset
-	movq \offset(%rsp),%r11
-	movq \offset+8(%rsp),%r10
-	movq \offset+16(%rsp),%r9
-	movq \offset+24(%rsp),%r8
-	movq \offset+40(%rsp),%rcx
-	movq \offset+48(%rsp),%rdx
-	movq \offset+56(%rsp),%rsi
-	movq \offset+64(%rsp),%rdi
-	movq \offset+72(%rsp),%rax
+	movq \offset(%rsp),    %r11
+	movq \offset+8(%rsp),  %r10
+	movq \offset+16(%rsp), %r9
+	movq \offset+24(%rsp), %r8
+	movq \offset+40(%rsp), %rcx
+	movq \offset+48(%rsp), %rdx
+	movq \offset+56(%rsp), %rsi
+	movq \offset+64(%rsp), %rdi
+	movq \offset+72(%rsp), %rax
 	.endm
-			
-#define REST_SKIP 6*8			
+
+#define REST_SKIP	6*8
+
 	.macro SAVE_REST
-	subq $REST_SKIP,%rsp
+	subq $REST_SKIP, %rsp
 	CFI_ADJUST_CFA_OFFSET	REST_SKIP
-	movq %rbx,5*8(%rsp) 
-	CFI_REL_OFFSET	rbx,5*8
-	movq %rbp,4*8(%rsp) 
-	CFI_REL_OFFSET	rbp,4*8
-	movq %r12,3*8(%rsp) 
-	CFI_REL_OFFSET	r12,3*8
-	movq %r13,2*8(%rsp) 
-	CFI_REL_OFFSET	r13,2*8
-	movq %r14,1*8(%rsp) 
-	CFI_REL_OFFSET	r14,1*8
-	movq %r15,(%rsp) 
-	CFI_REL_OFFSET	r15,0*8
-	.endm		
+	movq %rbx, 5*8(%rsp)
+	CFI_REL_OFFSET	rbx, 5*8
+	movq %rbp, 4*8(%rsp)
+	CFI_REL_OFFSET	rbp, 4*8
+	movq %r12, 3*8(%rsp)
+	CFI_REL_OFFSET	r12, 3*8
+	movq %r13, 2*8(%rsp)
+	CFI_REL_OFFSET	r13, 2*8
+	movq %r14, 1*8(%rsp)
+	CFI_REL_OFFSET	r14, 1*8
+	movq %r15, (%rsp)
+	CFI_REL_OFFSET	r15, 0*8
+	.endm
 
 	.macro RESTORE_REST
-	movq (%rsp),%r15
+	movq (%rsp),     %r15
 	CFI_RESTORE r15
-	movq 1*8(%rsp),%r14
+	movq 1*8(%rsp),  %r14
 	CFI_RESTORE r14
-	movq 2*8(%rsp),%r13
+	movq 2*8(%rsp),  %r13
 	CFI_RESTORE r13
-	movq 3*8(%rsp),%r12
+	movq 3*8(%rsp),  %r12
 	CFI_RESTORE r12
-	movq 4*8(%rsp),%rbp
+	movq 4*8(%rsp),  %rbp
 	CFI_RESTORE rbp
-	movq 5*8(%rsp),%rbx
+	movq 5*8(%rsp),  %rbx
 	CFI_RESTORE rbx
-	addq $REST_SKIP,%rsp
+	addq $REST_SKIP, %rsp
 	CFI_ADJUST_CFA_OFFSET	-(REST_SKIP)
 	.endm
-		
+
 	.macro SAVE_ALL
 	SAVE_ARGS
 	SAVE_REST
 	.endm
-		
+
 	.macro RESTORE_ALL addskip=0
 	RESTORE_REST
-	RESTORE_ARGS 0,\addskip
+	RESTORE_ARGS 0, \addskip
 	.endm
 
 	.macro icebp
 	.byte 0xf1
 	.endm
+
-- 
cgit v1.2.3


From 3a6a62f96f168d192fb0cc9c0b5ee2584740b32d Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 30 Jan 2008 13:32:50 +0100
Subject: debug: introduce __WARN()

Introduce __WARN() in the generic case, so the generic WARN_ON()
can use arch-specific code for when the condition is true.

Signed-off-by: Olof Johansson <olof@lixom.net>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-generic/bug.h | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index d56fedbb457a..1a0e1a7684bd 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -31,14 +31,19 @@ struct bug_entry {
 #define BUG_ON(condition) do { if (unlikely(condition)) BUG(); } while(0)
 #endif
 
-#ifndef HAVE_ARCH_WARN_ON
+#ifndef __WARN
+#define __WARN() do {							\
+	printk("WARNING: at %s:%d %s()\n", __FILE__,			\
+		__LINE__, __FUNCTION__);				\
+	dump_stack();							\
+} while (0)
+#endif
+
+#ifndef WARN_ON
 #define WARN_ON(condition) ({						\
 	int __ret_warn_on = !!(condition);				\
-	if (unlikely(__ret_warn_on)) {					\
-		printk("WARNING: at %s:%d %s()\n", __FILE__,		\
-			__LINE__, __FUNCTION__);			\
-		dump_stack();						\
-	}								\
+	if (unlikely(__ret_warn_on))					\
+		__WARN();						\
 	unlikely(__ret_warn_on);					\
 })
 #endif
-- 
cgit v1.2.3


From 79b4cc5ee7a8086ac2c9c0afa52e6d687ce1ffef Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Wed, 30 Jan 2008 13:32:50 +0100
Subject: debug: move WARN_ON() out of line

A quick grep shows that there are currently 1145 instances of WARN_ON
in the kernel. Currently, WARN_ON is pretty much entirely inlined,
which makes it hard to enhance it without growing the size of the kernel
(and getting Andrew unhappy).

This patch build on top of Olof's patch that introduces __WARN,
and places the slowpath out of line. It also uses Ingo's suggestion
to not use __FUNCTION__ but to use kallsyms to do the lookup;
this saves a ton of extra space since gcc doesn't need to store the function
string twice now:

3936367  833603  624736 5394706  525112 vmlinux.before
3917508  833603  624736 5375847  520767 vmlinux-slowpath

15Kb savings...

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Olof Johansson <olof@lixom.net>
Acked-by: Matt Meckall <mpm@selenic.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-generic/bug.h | 10 +++++-----
 kernel/panic.c            | 15 +++++++++++++++
 2 files changed, 20 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 1a0e1a7684bd..2632328d8646 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -32,11 +32,11 @@ struct bug_entry {
 #endif
 
 #ifndef __WARN
-#define __WARN() do {							\
-	printk("WARNING: at %s:%d %s()\n", __FILE__,			\
-		__LINE__, __FUNCTION__);				\
-	dump_stack();							\
-} while (0)
+#ifndef __ASSEMBLY__
+extern void warn_on_slowpath(const char *file, const int line);
+#define WANT_WARN_ON_SLOWPATH
+#endif
+#define __WARN() warn_on_slowpath(__FILE__, __LINE__)
 #endif
 
 #ifndef WARN_ON
diff --git a/kernel/panic.c b/kernel/panic.c
index da4d6bac270e..0ebea438278a 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -20,6 +20,7 @@
 #include <linux/kexec.h>
 #include <linux/debug_locks.h>
 #include <linux/random.h>
+#include <linux/kallsyms.h>
 
 int panic_on_oops;
 int tainted;
@@ -292,6 +293,20 @@ void oops_exit(void)
 		(unsigned long long)oops_id);
 }
 
+#ifdef WANT_WARN_ON_SLOWPATH
+void warn_on_slowpath(const char *file, int line)
+{
+	char function[KSYM_SYMBOL_LEN];
+	unsigned long caller = (unsigned long) __builtin_return_address(0);
+
+	sprint_symbol(function, caller);
+	printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file,
+		line, function);
+	dump_stack();
+}
+EXPORT_SYMBOL(warn_on_slowpath);
+#endif
+
 #ifdef CONFIG_CC_STACKPROTECTOR
 /*
  * Called when gcc's -fstack-protector feature is used, and
-- 
cgit v1.2.3


From 4d022e35fd7e07c522c7863fee6f07e53cf3fc14 Mon Sep 17 00:00:00 2001
From: Miguel Boton <mboton.lkml@gmail.com>
Date: Wed, 30 Jan 2008 13:32:51 +0100
Subject: x86: reboot_{32|64}.c unification

reboot_{32|64}.c unification patch.

This patch unifies the code from the reboot_32.c and reboot_64.c files.

It has been tested in computers with X86_32 and X86_64 kernels and it
looks like all reboot modes work fine (EFI restart system hasn't been
tested yet).

Probably I made some mistakes (like I usually do) so I hope
we can identify and fix them soon.

Signed-off-by: Miguel Boton <mboton@gmail.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile            |   4 +-
 arch/x86/kernel/reboot.c            | 449 ++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/reboot_32.c         | 413 ---------------------------------
 arch/x86/kernel/reboot_64.c         | 191 ---------------
 include/asm-x86/emergency-restart.h |   3 +
 5 files changed, 454 insertions(+), 606 deletions(-)
 create mode 100644 arch/x86/kernel/reboot.c
 delete mode 100644 arch/x86/kernel/reboot_32.c
 delete mode 100644 arch/x86/kernel/reboot_64.c

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 0903bbf0ca4d..b40bed4baa77 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -30,8 +30,8 @@ obj-y				+= step.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-y				+= cpu/
 obj-y				+= acpi/
-obj-$(CONFIG_X86_BIOS_REBOOT)	+= reboot_32.o
-obj-$(CONFIG_X86_64)            += reboot_64.o
+obj-$(CONFIG_X86_BIOS_REBOOT)	+= reboot.o
+obj-$(CONFIG_X86_64)		+= reboot.o
 obj-$(CONFIG_MCA)		+= mca_32.o
 obj-$(CONFIG_X86_MSR)		+= msr.o
 obj-$(CONFIG_X86_CPUID)		+= cpuid.o
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
new file mode 100644
index 000000000000..5b32f0b4d133
--- /dev/null
+++ b/arch/x86/kernel/reboot.c
@@ -0,0 +1,449 @@
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <linux/pm.h>
+#include <linux/efi.h>
+#include <acpi/reboot.h>
+#include <asm/io.h>
+#include <asm/apic.h>
+#include <asm/desc.h>
+#include <asm/hpet.h>
+#include <asm/reboot_fixups.h>
+#include <asm/reboot.h>
+
+#ifdef CONFIG_X86_32
+# include <linux/dmi.h>
+# include <linux/ctype.h>
+# include <linux/mc146818rtc.h>
+# include <asm/pgtable.h>
+#else
+# include <asm/iommu.h>
+#endif
+
+/*
+ * Power off function, if any
+ */
+void (*pm_power_off)(void);
+EXPORT_SYMBOL(pm_power_off);
+
+static long no_idt[3];
+static int reboot_mode;
+enum reboot_type reboot_type = BOOT_KBD;
+int reboot_force;
+
+#if defined(CONFIG_X86_32) && defined(CONFIG_SMP)
+static int reboot_cpu = -1;
+#endif
+
+/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old]
+   warm   Don't set the cold reboot flag
+   cold   Set the cold reboot flag
+   bios   Reboot by jumping through the BIOS (only for X86_32)
+   smp    Reboot by executing reset on BSP or other CPU (only for X86_32)
+   triple Force a triple fault (init)
+   kbd    Use the keyboard controller. cold reset (default)
+   acpi   Use the RESET_REG in the FADT
+   efi    Use efi reset_system runtime service
+   force  Avoid anything that could hang.
+ */
+static int __init reboot_setup(char *str)
+{
+	for (;;) {
+		switch (*str) {
+		case 'w':
+			reboot_mode = 0x1234;
+			break;
+
+		case 'c':
+			reboot_mode = 0;
+			break;
+
+#ifdef CONFIG_X86_32
+#ifdef CONFIG_SMP
+		case 's':
+			if (isdigit(*(str+1))) {
+				reboot_cpu = (int) (*(str+1) - '0');
+				if (isdigit(*(str+2)))
+					reboot_cpu = reboot_cpu*10 + (int)(*(str+2) - '0');
+			}
+				/* we will leave sorting out the final value
+				   when we are ready to reboot, since we might not
+				   have set up boot_cpu_id or smp_num_cpu */
+			break;
+#endif /* CONFIG_SMP */
+
+		case 'b':
+#endif
+		case 'a':
+		case 'k':
+		case 't':
+		case 'e':
+			reboot_type = *str;
+			break;
+
+		case 'f':
+			reboot_force = 1;
+			break;
+		}
+
+		str = strchr(str, ',');
+		if (str)
+			str++;
+		else
+			break;
+	}
+	return 1;
+}
+
+__setup("reboot=", reboot_setup);
+
+
+#ifdef CONFIG_X86_32
+/*
+ * Reboot options and system auto-detection code provided by
+ * Dell Inc. so their systems "just work". :-)
+ */
+
+/*
+ * Some machines require the "reboot=b"  commandline option,
+ * this quirk makes that automatic.
+ */
+static int __init set_bios_reboot(const struct dmi_system_id *d)
+{
+	if (reboot_type != BOOT_BIOS) {
+		reboot_type = BOOT_BIOS;
+		printk(KERN_INFO "%s series board detected. Selecting BIOS-method for reboots.\n", d->ident);
+	}
+	return 0;
+}
+
+static struct dmi_system_id __initdata reboot_dmi_table[] = {
+	{	/* Handle problems with rebooting on Dell E520's */
+		.callback = set_bios_reboot,
+		.ident = "Dell E520",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Dell DM061"),
+		},
+	},
+	{	/* Handle problems with rebooting on Dell 1300's */
+		.callback = set_bios_reboot,
+		.ident = "Dell PowerEdge 1300",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1300/"),
+		},
+	},
+	{	/* Handle problems with rebooting on Dell 300's */
+		.callback = set_bios_reboot,
+		.ident = "Dell PowerEdge 300",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 300/"),
+		},
+	},
+	{       /* Handle problems with rebooting on Dell Optiplex 745's SFF*/
+		.callback = set_bios_reboot,
+		.ident = "Dell OptiPlex 745",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"),
+			DMI_MATCH(DMI_BOARD_NAME, "0WF810"),
+		},
+	},
+	{	/* Handle problems with rebooting on Dell 2400's */
+		.callback = set_bios_reboot,
+		.ident = "Dell PowerEdge 2400",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2400"),
+		},
+	},
+	{	/* Handle problems with rebooting on HP laptops */
+		.callback = set_bios_reboot,
+		.ident = "HP Compaq Laptop",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"),
+		},
+	},
+	{ }
+};
+
+static int __init reboot_init(void)
+{
+	dmi_check_system(reboot_dmi_table);
+	return 0;
+}
+core_initcall(reboot_init);
+
+/* The following code and data reboots the machine by switching to real
+   mode and jumping to the BIOS reset entry point, as if the CPU has
+   really been reset.  The previous version asked the keyboard
+   controller to pulse the CPU reset line, which is more thorough, but
+   doesn't work with at least one type of 486 motherboard.  It is easy
+   to stop this code working; hence the copious comments. */
+static unsigned long long
+real_mode_gdt_entries [3] =
+{
+	0x0000000000000000ULL,	/* Null descriptor */
+	0x00009a000000ffffULL,	/* 16-bit real-mode 64k code at 0x00000000 */
+	0x000092000100ffffULL	/* 16-bit real-mode 64k data at 0x00000100 */
+};
+
+static struct desc_ptr
+real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, (long)real_mode_gdt_entries },
+real_mode_idt = { 0x3ff, 0 };
+
+/* This is 16-bit protected mode code to disable paging and the cache,
+   switch to real mode and jump to the BIOS reset code.
+
+   The instruction that switches to real mode by writing to CR0 must be
+   followed immediately by a far jump instruction, which set CS to a
+   valid value for real mode, and flushes the prefetch queue to avoid
+   running instructions that have already been decoded in protected
+   mode.
+
+   Clears all the flags except ET, especially PG (paging), PE
+   (protected-mode enable) and TS (task switch for coprocessor state
+   save).  Flushes the TLB after paging has been disabled.  Sets CD and
+   NW, to disable the cache on a 486, and invalidates the cache.  This
+   is more like the state of a 486 after reset.  I don't know if
+   something else should be done for other chips.
+
+   More could be done here to set up the registers as if a CPU reset had
+   occurred; hopefully real BIOSs don't assume much. */
+static unsigned char real_mode_switch [] =
+{
+	0x66, 0x0f, 0x20, 0xc0,			/*    movl  %cr0,%eax        */
+	0x66, 0x83, 0xe0, 0x11,			/*    andl  $0x00000011,%eax */
+	0x66, 0x0d, 0x00, 0x00, 0x00, 0x60,	/*    orl   $0x60000000,%eax */
+	0x66, 0x0f, 0x22, 0xc0,			/*    movl  %eax,%cr0        */
+	0x66, 0x0f, 0x22, 0xd8,			/*    movl  %eax,%cr3        */
+	0x66, 0x0f, 0x20, 0xc3,			/*    movl  %cr0,%ebx        */
+	0x66, 0x81, 0xe3, 0x00, 0x00, 0x00, 0x60,	/*    andl  $0x60000000,%ebx */
+	0x74, 0x02,				/*    jz    f                */
+	0x0f, 0x09,				/*    wbinvd                 */
+	0x24, 0x10,				/* f: andb  $0x10,al         */
+	0x66, 0x0f, 0x22, 0xc0			/*    movl  %eax,%cr0        */
+};
+static unsigned char jump_to_bios [] =
+{
+	0xea, 0x00, 0x00, 0xff, 0xff		/*    ljmp  $0xffff,$0x0000  */
+};
+
+/*
+ * Switch to real mode and then execute the code
+ * specified by the code and length parameters.
+ * We assume that length will aways be less that 100!
+ */
+void machine_real_restart(unsigned char *code, int length)
+{
+	local_irq_disable();
+
+	/* Write zero to CMOS register number 0x0f, which the BIOS POST
+	   routine will recognize as telling it to do a proper reboot.  (Well
+	   that's what this book in front of me says -- it may only apply to
+	   the Phoenix BIOS though, it's not clear).  At the same time,
+	   disable NMIs by setting the top bit in the CMOS address register,
+	   as we're about to do peculiar things to the CPU.  I'm not sure if
+	   `outb_p' is needed instead of just `outb'.  Use it to be on the
+	   safe side.  (Yes, CMOS_WRITE does outb_p's. -  Paul G.)
+	 */
+	spin_lock(&rtc_lock);
+	CMOS_WRITE(0x00, 0x8f);
+	spin_unlock(&rtc_lock);
+
+	/* Remap the kernel at virtual address zero, as well as offset zero
+	   from the kernel segment.  This assumes the kernel segment starts at
+	   virtual address PAGE_OFFSET. */
+	memcpy(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
+		sizeof(swapper_pg_dir [0]) * KERNEL_PGD_PTRS);
+
+	/*
+	 * Use `swapper_pg_dir' as our page directory.
+	 */
+	load_cr3(swapper_pg_dir);
+
+	/* Write 0x1234 to absolute memory location 0x472.  The BIOS reads
+	   this on booting to tell it to "Bypass memory test (also warm
+	   boot)".  This seems like a fairly standard thing that gets set by
+	   REBOOT.COM programs, and the previous reset routine did this
+	   too. */
+	*((unsigned short *)0x472) = reboot_mode;
+
+	/* For the switch to real mode, copy some code to low memory.  It has
+	   to be in the first 64k because it is running in 16-bit mode, and it
+	   has to have the same physical and virtual address, because it turns
+	   off paging.  Copy it near the end of the first page, out of the way
+	   of BIOS variables. */
+	memcpy((void *)(0x1000 - sizeof(real_mode_switch) - 100),
+		real_mode_switch, sizeof (real_mode_switch));
+	memcpy((void *)(0x1000 - 100), code, length);
+
+	/* Set up the IDT for real mode. */
+	load_idt(&real_mode_idt);
+
+	/* Set up a GDT from which we can load segment descriptors for real
+	   mode.  The GDT is not used in real mode; it is just needed here to
+	   prepare the descriptors. */
+	load_gdt(&real_mode_gdt);
+
+	/* Load the data segment registers, and thus the descriptors ready for
+	   real mode.  The base address of each segment is 0x100, 16 times the
+	   selector value being loaded here.  This is so that the segment
+	   registers don't have to be reloaded after switching to real mode:
+	   the values are consistent for real mode operation already. */
+	__asm__ __volatile__ ("movl $0x0010,%%eax\n"
+				"\tmovl %%eax,%%ds\n"
+				"\tmovl %%eax,%%es\n"
+				"\tmovl %%eax,%%fs\n"
+				"\tmovl %%eax,%%gs\n"
+				"\tmovl %%eax,%%ss" : : : "eax");
+
+	/* Jump to the 16-bit code that we copied earlier.  It disables paging
+	   and the cache, switches to real mode, and jumps to the BIOS reset
+	   entry point. */
+	__asm__ __volatile__ ("ljmp $0x0008,%0"
+				:
+				: "i" ((void *)(0x1000 - sizeof (real_mode_switch) - 100)));
+}
+#ifdef CONFIG_APM_MODULE
+EXPORT_SYMBOL(machine_real_restart);
+#endif
+
+#endif /* CONFIG_X86_32 */
+
+static inline void kb_wait(void)
+{
+	int i;
+
+	for (i = 0; i < 0x10000; i++)
+		if ((inb_p(0x64) & 0x02) == 0)
+			break;
+}
+
+void machine_emergency_restart(void)
+{
+	int i;
+
+	/* Tell the BIOS if we want cold or warm reboot */
+	*((unsigned short *)__va(0x472)) = reboot_mode;
+
+	for (;;) {
+		/* Could also try the reset bit in the Hammer NB */
+		switch (reboot_type) {
+		case BOOT_KBD:
+			for (i = 0; i < 10; i++) {
+				kb_wait();
+				udelay(50);
+				outb(0xfe, 0x64); /* pulse reset low */
+				udelay(50);
+			}
+
+		case BOOT_TRIPLE:
+			load_idt((const struct desc_ptr *)&no_idt);
+			__asm__ __volatile__("int3");
+
+			reboot_type = BOOT_KBD;
+			break;
+
+#ifdef CONFIG_X86_32
+		case BOOT_BIOS:
+			machine_real_restart(jump_to_bios, sizeof(jump_to_bios));
+
+			reboot_type = BOOT_KBD;
+			break;
+#endif
+
+		case BOOT_ACPI:
+			acpi_reboot();
+			reboot_type = BOOT_KBD;
+			break;
+
+
+		case BOOT_EFI:
+			if (efi_enabled)
+				efi.reset_system(reboot_mode ? EFI_RESET_WARM : EFI_RESET_COLD,
+						 EFI_SUCCESS, 0, NULL);
+
+			reboot_type = BOOT_KBD;
+			break;
+		}
+	}
+}
+
+void machine_shutdown(void)
+{
+	/* Stop the cpus and apics */
+#ifdef CONFIG_SMP
+	int reboot_cpu_id;
+
+	/* The boot cpu is always logical cpu 0 */
+	reboot_cpu_id = 0;
+
+#ifdef CONFIG_X86_32
+	/* See if there has been given a command line override */
+	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
+		cpu_isset(reboot_cpu, cpu_online_map))
+		reboot_cpu_id = reboot_cpu;
+#endif
+
+	/* Make certain the cpu I'm about to reboot on is online */
+	if (!cpu_isset(reboot_cpu_id, cpu_online_map))
+		reboot_cpu_id = smp_processor_id();
+
+	/* Make certain I only run on the appropriate processor */
+	set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id));
+
+	/* O.K Now that I'm on the appropriate processor,
+	 * stop all of the others.
+	 */
+	smp_send_stop();
+#endif
+
+	lapic_shutdown();
+
+#ifdef CONFIG_X86_IO_APIC
+	disable_IO_APIC();
+#endif
+
+#ifdef CONFIG_HPET_TIMER
+	hpet_disable();
+#endif
+
+#ifdef CONFIG_X86_64
+	pci_iommu_shutdown();
+#endif
+}
+
+void machine_restart(char *__unused)
+{
+	printk("machine restart\n");
+
+	if (!reboot_force)
+		machine_shutdown();
+	machine_emergency_restart();
+}
+
+void machine_halt(void)
+{
+}
+
+void machine_power_off(void)
+{
+	if (pm_power_off) {
+		if (!reboot_force)
+			machine_shutdown();
+		pm_power_off();
+	}
+}
+
+struct machine_ops machine_ops = {
+	.power_off = machine_power_off,
+	.shutdown = machine_shutdown,
+	.emergency_restart = machine_emergency_restart,
+	.restart = machine_restart,
+	.halt = machine_halt
+};
diff --git a/arch/x86/kernel/reboot_32.c b/arch/x86/kernel/reboot_32.c
deleted file mode 100644
index c3376fae639d..000000000000
--- a/arch/x86/kernel/reboot_32.c
+++ /dev/null
@@ -1,413 +0,0 @@
-#include <linux/mm.h>
-#include <linux/module.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/mc146818rtc.h>
-#include <linux/efi.h>
-#include <linux/dmi.h>
-#include <linux/ctype.h>
-#include <linux/pm.h>
-#include <linux/reboot.h>
-#include <asm/uaccess.h>
-#include <asm/apic.h>
-#include <asm/hpet.h>
-#include <asm/desc.h>
-#include "mach_reboot.h"
-#include <asm/reboot_fixups.h>
-#include <asm/reboot.h>
-
-/*
- * Power off function, if any
- */
-void (*pm_power_off)(void);
-EXPORT_SYMBOL(pm_power_off);
-
-static int reboot_mode;
-static int reboot_thru_bios;
-
-#ifdef CONFIG_SMP
-static int reboot_cpu = -1;
-#endif
-static int __init reboot_setup(char *str)
-{
-	while(1) {
-		switch (*str) {
-		case 'w': /* "warm" reboot (no memory testing etc) */
-			reboot_mode = 0x1234;
-			break;
-		case 'c': /* "cold" reboot (with memory testing etc) */
-			reboot_mode = 0x0;
-			break;
-		case 'b': /* "bios" reboot by jumping through the BIOS */
-			reboot_thru_bios = 1;
-			break;
-		case 'h': /* "hard" reboot by toggling RESET and/or crashing the CPU */
-			reboot_thru_bios = 0;
-			break;
-#ifdef CONFIG_SMP
-		case 's': /* "smp" reboot by executing reset on BSP or other CPU*/
-			if (isdigit(*(str+1))) {
-				reboot_cpu = (int) (*(str+1) - '0');
-				if (isdigit(*(str+2)))
-					reboot_cpu = reboot_cpu*10 + (int)(*(str+2) - '0');
-			}
-				/* we will leave sorting out the final value 
-				when we are ready to reboot, since we might not
- 				have set up boot_cpu_id or smp_num_cpu */
-			break;
-#endif
-		}
-		if((str = strchr(str,',')) != NULL)
-			str++;
-		else
-			break;
-	}
-	return 1;
-}
-
-__setup("reboot=", reboot_setup);
-
-/*
- * Reboot options and system auto-detection code provided by
- * Dell Inc. so their systems "just work". :-)
- */
-
-/*
- * Some machines require the "reboot=b"  commandline option, this quirk makes that automatic.
- */
-static int __init set_bios_reboot(const struct dmi_system_id *d)
-{
-	if (!reboot_thru_bios) {
-		reboot_thru_bios = 1;
-		printk(KERN_INFO "%s series board detected. Selecting BIOS-method for reboots.\n", d->ident);
-	}
-	return 0;
-}
-
-static struct dmi_system_id __initdata reboot_dmi_table[] = {
-	{	/* Handle problems with rebooting on Dell E520's */
-		.callback = set_bios_reboot,
-		.ident = "Dell E520",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "Dell DM061"),
-		},
-	},
-	{	/* Handle problems with rebooting on Dell 1300's */
-		.callback = set_bios_reboot,
-		.ident = "Dell PowerEdge 1300",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1300/"),
-		},
-	},
-	{	/* Handle problems with rebooting on Dell 300's */
-		.callback = set_bios_reboot,
-		.ident = "Dell PowerEdge 300",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 300/"),
-		},
-	},
-	{       /* Handle problems with rebooting on Dell Optiplex 745's SFF*/
-		.callback = set_bios_reboot,
-		.ident = "Dell OptiPlex 745",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"),
-			DMI_MATCH(DMI_BOARD_NAME, "0WF810"),
-		},
-	},
-	{	/* Handle problems with rebooting on Dell 2400's */
-		.callback = set_bios_reboot,
-		.ident = "Dell PowerEdge 2400",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2400"),
-		},
-	},
-	{	/* Handle problems with rebooting on HP laptops */
-		.callback = set_bios_reboot,
-		.ident = "HP Compaq Laptop",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"),
-		},
-	},
-	{ }
-};
-
-static int __init reboot_init(void)
-{
-	dmi_check_system(reboot_dmi_table);
-	return 0;
-}
-
-core_initcall(reboot_init);
-
-/* The following code and data reboots the machine by switching to real
-   mode and jumping to the BIOS reset entry point, as if the CPU has
-   really been reset.  The previous version asked the keyboard
-   controller to pulse the CPU reset line, which is more thorough, but
-   doesn't work with at least one type of 486 motherboard.  It is easy
-   to stop this code working; hence the copious comments. */
-
-static unsigned long long
-real_mode_gdt_entries [3] =
-{
-	0x0000000000000000ULL,	/* Null descriptor */
-	0x00009a000000ffffULL,	/* 16-bit real-mode 64k code at 0x00000000 */
-	0x000092000100ffffULL	/* 16-bit real-mode 64k data at 0x00000100 */
-};
-
-static struct desc_ptr
-real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, (long)real_mode_gdt_entries },
-real_mode_idt = { 0x3ff, 0 },
-no_idt = { 0, 0 };
-
-
-/* This is 16-bit protected mode code to disable paging and the cache,
-   switch to real mode and jump to the BIOS reset code.
-
-   The instruction that switches to real mode by writing to CR0 must be
-   followed immediately by a far jump instruction, which set CS to a
-   valid value for real mode, and flushes the prefetch queue to avoid
-   running instructions that have already been decoded in protected
-   mode.
-
-   Clears all the flags except ET, especially PG (paging), PE
-   (protected-mode enable) and TS (task switch for coprocessor state
-   save).  Flushes the TLB after paging has been disabled.  Sets CD and
-   NW, to disable the cache on a 486, and invalidates the cache.  This
-   is more like the state of a 486 after reset.  I don't know if
-   something else should be done for other chips.
-
-   More could be done here to set up the registers as if a CPU reset had
-   occurred; hopefully real BIOSs don't assume much. */
-
-static unsigned char real_mode_switch [] =
-{
-	0x66, 0x0f, 0x20, 0xc0,			/*    movl  %cr0,%eax        */
-	0x66, 0x83, 0xe0, 0x11,			/*    andl  $0x00000011,%eax */
-	0x66, 0x0d, 0x00, 0x00, 0x00, 0x60,	/*    orl   $0x60000000,%eax */
-	0x66, 0x0f, 0x22, 0xc0,			/*    movl  %eax,%cr0        */
-	0x66, 0x0f, 0x22, 0xd8,			/*    movl  %eax,%cr3        */
-	0x66, 0x0f, 0x20, 0xc3,			/*    movl  %cr0,%ebx        */
-	0x66, 0x81, 0xe3, 0x00, 0x00, 0x00, 0x60,	/*    andl  $0x60000000,%ebx */
-	0x74, 0x02,				/*    jz    f                */
-	0x0f, 0x09,				/*    wbinvd                 */
-	0x24, 0x10,				/* f: andb  $0x10,al         */
-	0x66, 0x0f, 0x22, 0xc0			/*    movl  %eax,%cr0        */
-};
-static unsigned char jump_to_bios [] =
-{
-	0xea, 0x00, 0x00, 0xff, 0xff		/*    ljmp  $0xffff,$0x0000  */
-};
-
-/*
- * Switch to real mode and then execute the code
- * specified by the code and length parameters.
- * We assume that length will aways be less that 100!
- */
-void machine_real_restart(unsigned char *code, int length)
-{
-	local_irq_disable();
-
-	/* Write zero to CMOS register number 0x0f, which the BIOS POST
-	   routine will recognize as telling it to do a proper reboot.  (Well
-	   that's what this book in front of me says -- it may only apply to
-	   the Phoenix BIOS though, it's not clear).  At the same time,
-	   disable NMIs by setting the top bit in the CMOS address register,
-	   as we're about to do peculiar things to the CPU.  I'm not sure if
-	   `outb_p' is needed instead of just `outb'.  Use it to be on the
-	   safe side.  (Yes, CMOS_WRITE does outb_p's. -  Paul G.)
-	 */
-
-	spin_lock(&rtc_lock);
-	CMOS_WRITE(0x00, 0x8f);
-	spin_unlock(&rtc_lock);
-
-	/* Remap the kernel at virtual address zero, as well as offset zero
-	   from the kernel segment.  This assumes the kernel segment starts at
-	   virtual address PAGE_OFFSET. */
-
-	memcpy (swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
-		sizeof (swapper_pg_dir [0]) * KERNEL_PGD_PTRS);
-
-	/*
-	 * Use `swapper_pg_dir' as our page directory.
-	 */
-	load_cr3(swapper_pg_dir);
-
-	/* Write 0x1234 to absolute memory location 0x472.  The BIOS reads
-	   this on booting to tell it to "Bypass memory test (also warm
-	   boot)".  This seems like a fairly standard thing that gets set by
-	   REBOOT.COM programs, and the previous reset routine did this
-	   too. */
-
-	*((unsigned short *)0x472) = reboot_mode;
-
-	/* For the switch to real mode, copy some code to low memory.  It has
-	   to be in the first 64k because it is running in 16-bit mode, and it
-	   has to have the same physical and virtual address, because it turns
-	   off paging.  Copy it near the end of the first page, out of the way
-	   of BIOS variables. */
-
-	memcpy ((void *) (0x1000 - sizeof (real_mode_switch) - 100),
-		real_mode_switch, sizeof (real_mode_switch));
-	memcpy ((void *) (0x1000 - 100), code, length);
-
-	/* Set up the IDT for real mode. */
-
-	load_idt(&real_mode_idt);
-
-	/* Set up a GDT from which we can load segment descriptors for real
-	   mode.  The GDT is not used in real mode; it is just needed here to
-	   prepare the descriptors. */
-
-	load_gdt(&real_mode_gdt);
-
-	/* Load the data segment registers, and thus the descriptors ready for
-	   real mode.  The base address of each segment is 0x100, 16 times the
-	   selector value being loaded here.  This is so that the segment
-	   registers don't have to be reloaded after switching to real mode:
-	   the values are consistent for real mode operation already. */
-
-	__asm__ __volatile__ ("movl $0x0010,%%eax\n"
-				"\tmovl %%eax,%%ds\n"
-				"\tmovl %%eax,%%es\n"
-				"\tmovl %%eax,%%fs\n"
-				"\tmovl %%eax,%%gs\n"
-				"\tmovl %%eax,%%ss" : : : "eax");
-
-	/* Jump to the 16-bit code that we copied earlier.  It disables paging
-	   and the cache, switches to real mode, and jumps to the BIOS reset
-	   entry point. */
-
-	__asm__ __volatile__ ("ljmp $0x0008,%0"
-				:
-				: "i" ((void *) (0x1000 - sizeof (real_mode_switch) - 100)));
-}
-#ifdef CONFIG_APM_MODULE
-EXPORT_SYMBOL(machine_real_restart);
-#endif
-
-static void native_machine_shutdown(void)
-{
-#ifdef CONFIG_SMP
-	int reboot_cpu_id;
-
-	/* The boot cpu is always logical cpu 0 */
-	reboot_cpu_id = 0;
-
-	/* See if there has been given a command line override */
-	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
-		cpu_isset(reboot_cpu, cpu_online_map)) {
-		reboot_cpu_id = reboot_cpu;
-	}
-
-	/* Make certain the cpu I'm rebooting on is online */
-	if (!cpu_isset(reboot_cpu_id, cpu_online_map)) {
-		reboot_cpu_id = smp_processor_id();
-	}
-
-	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id));
-
-	/* O.K. Now that I'm on the appropriate processor, stop
-	 * all of the others, and disable their local APICs.
-	 */
-
-	smp_send_stop();
-#endif /* CONFIG_SMP */
-
-	lapic_shutdown();
-
-#ifdef CONFIG_X86_IO_APIC
-	disable_IO_APIC();
-#endif
-#ifdef CONFIG_HPET_TIMER
-	hpet_disable();
-#endif
-}
-
-void __attribute__((weak)) mach_reboot_fixups(void)
-{
-}
-
-static void native_machine_emergency_restart(void)
-{
-	if (!reboot_thru_bios) {
-		if (efi_enabled) {
-			efi.reset_system(EFI_RESET_COLD, EFI_SUCCESS, 0, NULL);
-			load_idt(&no_idt);
-			__asm__ __volatile__("int3");
-		}
-		/* rebooting needs to touch the page at absolute addr 0 */
-		*((unsigned short *)__va(0x472)) = reboot_mode;
-		for (;;) {
-			mach_reboot_fixups(); /* for board specific fixups */
-			mach_reboot();
-			/* That didn't work - force a triple fault.. */
-			load_idt(&no_idt);
-			__asm__ __volatile__("int3");
-		}
-	}
-	if (efi_enabled)
-		efi.reset_system(EFI_RESET_WARM, EFI_SUCCESS, 0, NULL);
-
-	machine_real_restart(jump_to_bios, sizeof(jump_to_bios));
-}
-
-static void native_machine_restart(char * __unused)
-{
-	machine_shutdown();
-	machine_emergency_restart();
-}
-
-static void native_machine_halt(void)
-{
-}
-
-static void native_machine_power_off(void)
-{
-	if (pm_power_off) {
-		machine_shutdown();
-		pm_power_off();
-	}
-}
-
-
-struct machine_ops machine_ops = {
-	.power_off = native_machine_power_off,
-	.shutdown = native_machine_shutdown,
-	.emergency_restart = native_machine_emergency_restart,
-	.restart = native_machine_restart,
-	.halt = native_machine_halt,
-};
-
-void machine_power_off(void)
-{
-	machine_ops.power_off();
-}
-
-void machine_shutdown(void)
-{
-	machine_ops.shutdown();
-}
-
-void machine_emergency_restart(void)
-{
-	machine_ops.emergency_restart();
-}
-
-void machine_restart(char *cmd)
-{
-	machine_ops.restart(cmd);
-}
-
-void machine_halt(void)
-{
-	machine_ops.halt();
-}
diff --git a/arch/x86/kernel/reboot_64.c b/arch/x86/kernel/reboot_64.c
deleted file mode 100644
index d6bdf93ffca9..000000000000
--- a/arch/x86/kernel/reboot_64.c
+++ /dev/null
@@ -1,191 +0,0 @@
-/* Various gunk just to reboot the machine. */ 
-#include <linux/module.h>
-#include <linux/reboot.h>
-#include <linux/init.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
-#include <linux/ctype.h>
-#include <linux/string.h>
-#include <linux/pm.h>
-#include <linux/kdebug.h>
-#include <linux/sched.h>
-#include <linux/efi.h>
-#include <acpi/reboot.h>
-#include <asm/io.h>
-#include <asm/delay.h>
-#include <asm/desc.h>
-#include <asm/hw_irq.h>
-#include <asm/system.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <asm/apic.h>
-#include <asm/hpet.h>
-#include <asm/gart.h>
-
-/*
- * Power off function, if any
- */
-void (*pm_power_off)(void);
-EXPORT_SYMBOL(pm_power_off);
-
-static long no_idt[3];
-enum reboot_type reboot_type = BOOT_KBD;
-static int reboot_mode = 0;
-int reboot_force;
-
-/* reboot=t[riple] | k[bd] | e[fi] [, [w]arm | [c]old]
-   warm   Don't set the cold reboot flag
-   cold   Set the cold reboot flag
-   triple Force a triple fault (init)
-   kbd    Use the keyboard controller. cold reset (default)
-   acpi   Use the RESET_REG in the FADT
-   efi    Use efi reset_system runtime service
-   force  Avoid anything that could hang.
- */ 
-static int __init reboot_setup(char *str)
-{
-	for (;;) {
-		switch (*str) {
-		case 'w': 
-			reboot_mode = 0x1234;
-			break;
-
-		case 'c':
-			reboot_mode = 0;
-			break;
-
-		case 't':
-		case 'a':
-		case 'b':
-		case 'k':
-		case 'e':
-			reboot_type = *str;
-			break;
-		case 'f':
-			reboot_force = 1;
-			break;
-		}
-		if((str = strchr(str,',')) != NULL)
-			str++;
-		else
-			break;
-	}
-	return 1;
-}
-
-__setup("reboot=", reboot_setup);
-
-static inline void kb_wait(void)
-{
-	int i;
-
-	for (i=0; i<0x10000; i++)
-		if ((inb_p(0x64) & 0x02) == 0)
-			break;
-}
-
-void machine_shutdown(void)
-{
-	unsigned long flags;
-
-	/* Stop the cpus and apics */
-#ifdef CONFIG_SMP
-	int reboot_cpu_id;
-
-	/* The boot cpu is always logical cpu 0 */
-	reboot_cpu_id = 0;
-
-	/* Make certain the cpu I'm about to reboot on is online */
-	if (!cpu_isset(reboot_cpu_id, cpu_online_map)) {
-		reboot_cpu_id = smp_processor_id();
-	}
-
-	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed(current, cpumask_of_cpu(reboot_cpu_id));
-
-	/* O.K Now that I'm on the appropriate processor,
-	 * stop all of the others.
-	 */
-	smp_send_stop();
-#endif
-
-	local_irq_save(flags);
-
-#ifndef CONFIG_SMP
-	disable_local_APIC();
-#endif
-
-	disable_IO_APIC();
-
-#ifdef CONFIG_HPET_TIMER
-	hpet_disable();
-#endif
-	local_irq_restore(flags);
-
-	pci_iommu_shutdown();
-}
-
-void machine_emergency_restart(void)
-{
-	int i;
-
-	/* Tell the BIOS if we want cold or warm reboot */
-	*((unsigned short *)__va(0x472)) = reboot_mode;
-       
-	for (;;) {
-		/* Could also try the reset bit in the Hammer NB */
-		switch (reboot_type) { 
-		case BOOT_KBD:
-		for (i=0; i<10; i++) {
-			kb_wait();
-			udelay(50);
-			outb(0xfe,0x64);         /* pulse reset low */
-			udelay(50);
-		}
-
-		case BOOT_TRIPLE: 
-			load_idt((const struct desc_ptr *)&no_idt);
-			__asm__ __volatile__("int3");
-
-			reboot_type = BOOT_KBD;
-			break;
-
-		case BOOT_ACPI:
-			acpi_reboot();
-			reboot_type = BOOT_KBD;
-			break;
-
-		case BOOT_EFI:
-			if (efi_enabled)
-				efi.reset_system(reboot_mode ? EFI_RESET_WARM : EFI_RESET_COLD,
-						 EFI_SUCCESS, 0, NULL);
-			reboot_type = BOOT_KBD;
-			break;
-		}
-	}      
-}
-
-void machine_restart(char * __unused)
-{
-	printk("machine restart\n");
-
-	if (!reboot_force) {
-		machine_shutdown();
-	}
-	machine_emergency_restart();
-}
-
-void machine_halt(void)
-{
-}
-
-void machine_power_off(void)
-{
-	if (pm_power_off) {
-		if (!reboot_force) {
-			machine_shutdown();
-		}
-		pm_power_off();
-	}
-}
-
diff --git a/include/asm-x86/emergency-restart.h b/include/asm-x86/emergency-restart.h
index 54189084462a..8e6aef19f8f0 100644
--- a/include/asm-x86/emergency-restart.h
+++ b/include/asm-x86/emergency-restart.h
@@ -4,6 +4,9 @@
 enum reboot_type {
 	BOOT_TRIPLE = 't',
 	BOOT_KBD = 'k',
+#ifdef CONFIG_X86_32
+	BOOT_BIOS = 'b',
+#endif
 	BOOT_ACPI = 'a',
 	BOOT_EFI = 'e'
 };
-- 
cgit v1.2.3


From b32ef636a59aad12f9f9b5dc34c93222842c58ba Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:51 +0100
Subject: percpu: use a kconfig variable to signal arch specific percpu setup

The use of the __GENERIC_PERCPU is a bit problematic since arches
may want to run their own percpu setup while using the generic
percpu definitions. Replace it through a kconfig variable.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/ia64/Kconfig            | 3 +++
 arch/powerpc/Kconfig         | 3 +++
 arch/sparc64/Kconfig         | 3 +++
 arch/x86/Kconfig             | 3 +++
 include/asm-generic/percpu.h | 1 -
 include/asm-s390/percpu.h    | 2 --
 include/asm-x86/percpu_32.h  | 2 --
 init/main.c                  | 4 ++--
 8 files changed, 14 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 4a81b7fb191a..5a41e75ae1fe 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -80,6 +80,9 @@ config GENERIC_TIME_VSYSCALL
 	bool
 	default y
 
+config ARCH_SETS_UP_PER_CPU_AREA
+	def_bool y
+
 config DMI
 	bool
 	default y
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index c17a194beb0e..fb85f6b72fcf 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -42,6 +42,9 @@ config GENERIC_HARDIRQS
 	bool
 	default y
 
+config ARCH_SETS_UP_PER_CPU_AREA
+	def_bool PPC64
+
 config IRQ_PER_CPU
 	bool
 	default y
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index 1e25bce0366d..26f5791baa33 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -66,6 +66,9 @@ config AUDIT_ARCH
 	bool
 	default y
 
+config ARCH_SETS_UP_PER_CPU_AREA
+	def_bool y
+
 config ARCH_NO_VIRT_TO_BUS
 	def_bool y
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 1de28e850aa4..dc80a9f1f303 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -97,6 +97,9 @@ config GENERIC_TIME_VSYSCALL
 	bool
 	default X86_64
 
+config ARCH_SETS_UP_PER_CPU_AREA
+	def_bool X86_64
+
 config ARCH_SUPPORTS_OPROFILE
 	bool
 	default y
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index d85172e9ed45..b5e53b9ab1f7 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -3,7 +3,6 @@
 #include <linux/compiler.h>
 #include <linux/threads.h>
 
-#define __GENERIC_PER_CPU
 #ifdef CONFIG_SMP
 
 extern unsigned long __per_cpu_offset[NR_CPUS];
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 545857e64443..f94d0d3cdb2f 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -4,8 +4,6 @@
 #include <linux/compiler.h>
 #include <asm/lowcore.h>
 
-#define __GENERIC_PER_CPU
-
 /*
  * s390 uses its own implementation for per cpu data, the offset of
  * the cpu local data area is cached in the cpu's lowcore memory.
diff --git a/include/asm-x86/percpu_32.h b/include/asm-x86/percpu_32.h
index a7ebd436f3cc..3949586bf94e 100644
--- a/include/asm-x86/percpu_32.h
+++ b/include/asm-x86/percpu_32.h
@@ -41,8 +41,6 @@
  *    PER_CPU(cpu_gdt_descr, %ebx)
  */
 #ifdef CONFIG_SMP
-/* Same as generic implementation except for optimized local access. */
-#define __GENERIC_PER_CPU
 
 /* This is used for other cpus to find our section. */
 extern unsigned long __per_cpu_offset[];
diff --git a/init/main.c b/init/main.c
index f287ca5862b9..3f8aba291ed3 100644
--- a/init/main.c
+++ b/init/main.c
@@ -363,7 +363,7 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 
 #else
 
-#ifdef __GENERIC_PER_CPU
+#ifndef CONFIG_ARCH_SETS_UP_PER_CPU_AREA
 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 
 EXPORT_SYMBOL(__per_cpu_offset);
@@ -384,7 +384,7 @@ static void __init setup_per_cpu_areas(void)
 		ptr += size;
 	}
 }
-#endif /* !__GENERIC_PER_CPU */
+#endif /* CONFIG_ARCH_SETS_UP_CPU_AREA */
 
 /* Called by boot processor to activate the rest. */
 static void __init smp_init(void)
-- 
cgit v1.2.3


From 5280e004fc22314122c84978c0b6a741cf96dc0f Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:52 +0100
Subject: percpu: move arch XX_PER_CPU_XX definitions into linux/percpu.h

- Special consideration for IA64: Add the ability to specify
  arch specific per cpu flags

- remove .data.percpu attribute from DEFINE_PER_CPU for non-smp case.

The arch definitions are all the same. So move them into linux/percpu.h.

We cannot move DECLARE_PER_CPU since some include files just include
asm/percpu.h to avoid include recursion problems.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-generic/percpu.h | 18 ------------------
 include/asm-ia64/percpu.h    | 24 ++----------------------
 include/asm-powerpc/percpu.h | 17 -----------------
 include/asm-s390/percpu.h    | 18 ------------------
 include/asm-sparc64/percpu.h | 16 ----------------
 include/asm-x86/percpu_32.h  | 12 ------------
 include/asm-x86/percpu_64.h  | 17 -----------------
 include/linux/percpu.h       | 24 ++++++++++++++++++++++++
 8 files changed, 26 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index b5e53b9ab1f7..e038f13594e5 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -9,15 +9,6 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 
 #define per_cpu_offset(x) (__per_cpu_offset[x])
 
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
 /* var is in discarded region: offset to particular copy we want */
 #define per_cpu(var, cpu) (*({				\
 	extern int simple_identifier_##var(void);	\
@@ -35,12 +26,6 @@ do {								\
 } while (0)
 #else /* ! SMP */
 
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
-
 #define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 #define __raw_get_cpu_var(var)			per_cpu__##var
@@ -49,7 +34,4 @@ do {								\
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 #endif /* _ASM_GENERIC_PERCPU_H_ */
diff --git a/include/asm-ia64/percpu.h b/include/asm-ia64/percpu.h
index c4f1e328a5ba..0095bcf79848 100644
--- a/include/asm-ia64/percpu.h
+++ b/include/asm-ia64/percpu.h
@@ -16,28 +16,11 @@
 #include <linux/threads.h>
 
 #ifdef HAVE_MODEL_SMALL_ATTRIBUTE
-# define __SMALL_ADDR_AREA	__attribute__((__model__ (__small__)))
-#else
-# define __SMALL_ADDR_AREA
+# define PER_CPU_ATTRIBUTES	__attribute__((__model__ (__small__)))
 #endif
 
 #define DECLARE_PER_CPU(type, name)				\
-	extern __SMALL_ADDR_AREA __typeof__(type) per_cpu__##name
-
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name)				\
-	__attribute__((__section__(".data.percpu")))		\
-	__SMALL_ADDR_AREA __typeof__(type) per_cpu__##name
-
-#ifdef CONFIG_SMP
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)			\
-	__attribute__((__section__(".data.percpu.shared_aligned")))	\
-	__SMALL_ADDR_AREA __typeof__(type) per_cpu__##name		\
-	____cacheline_aligned_in_smp
-#else
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-	DEFINE_PER_CPU(type, name)
-#endif
+	extern PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
 
 /*
  * Pretty much a literal copy of asm-generic/percpu.h, except that percpu_modcopy() is an
@@ -68,9 +51,6 @@ extern void *per_cpu_init(void);
 
 #endif	/* SMP */
 
-#define EXPORT_PER_CPU_SYMBOL(var)		EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var)		EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 /*
  * Be extremely careful when taking the address of this variable!  Due to virtual
  * remapping, it is different from the canonical address returned by __get_cpu_var(var)!
diff --git a/include/asm-powerpc/percpu.h b/include/asm-powerpc/percpu.h
index 6b229626d3ff..cc1cbf656b02 100644
--- a/include/asm-powerpc/percpu.h
+++ b/include/asm-powerpc/percpu.h
@@ -16,15 +16,6 @@
 #define __my_cpu_offset() get_paca()->data_offset
 #define per_cpu_offset(x) (__per_cpu_offset(x))
 
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
 /* var is in discarded region: offset to particular copy we want */
 #define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)))
 #define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()))
@@ -43,11 +34,6 @@ extern void setup_per_cpu_areas(void);
 
 #else /* ! SMP */
 
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
-
 #define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 #define __raw_get_cpu_var(var)			per_cpu__##var
@@ -56,9 +42,6 @@ extern void setup_per_cpu_areas(void);
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 #else
 #include <asm-generic/percpu.h>
 #endif
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index f94d0d3cdb2f..2d676a873858 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -34,16 +34,6 @@
 
 extern unsigned long __per_cpu_offset[NR_CPUS];
 
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) \
-    __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
 #define __get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset)
 #define __raw_get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset)
 #define per_cpu(var,cpu) __reloc_hide(var,__per_cpu_offset[cpu])
@@ -60,11 +50,6 @@ do {								\
 
 #else /* ! SMP */
 
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
-
 #define __get_cpu_var(var) __reloc_hide(var,0)
 #define __raw_get_cpu_var(var) __reloc_hide(var,0)
 #define per_cpu(var,cpu) __reloc_hide(var,0)
@@ -73,7 +58,4 @@ do {								\
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 #endif /* __ARCH_S390_PERCPU__ */
diff --git a/include/asm-sparc64/percpu.h b/include/asm-sparc64/percpu.h
index a1f53a4da405..c7e52decba98 100644
--- a/include/asm-sparc64/percpu.h
+++ b/include/asm-sparc64/percpu.h
@@ -16,15 +16,6 @@ extern unsigned long __per_cpu_shift;
 	(__per_cpu_base + ((unsigned long)(__cpu) << __per_cpu_shift))
 #define per_cpu_offset(x) (__per_cpu_offset(x))
 
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
 /* var is in discarded region: offset to particular copy we want */
 #define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)))
 #define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __local_per_cpu_offset))
@@ -41,10 +32,6 @@ do {								\
 #else /* ! SMP */
 
 #define real_setup_per_cpu_areas()		do { } while (0)
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
 
 #define per_cpu(var, cpu)			(*((void)cpu, &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
@@ -54,7 +41,4 @@ do {								\
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 #endif /* __ARCH_SPARC64_PERCPU__ */
diff --git a/include/asm-x86/percpu_32.h b/include/asm-x86/percpu_32.h
index 3949586bf94e..77bd0045f331 100644
--- a/include/asm-x86/percpu_32.h
+++ b/include/asm-x86/percpu_32.h
@@ -47,16 +47,7 @@ extern unsigned long __per_cpu_offset[];
 
 #define per_cpu_offset(x) (__per_cpu_offset[x])
 
-/* Separate out the type, so (int[3], foo) works. */
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
 /* We can use this directly for local CPU (faster). */
 DECLARE_PER_CPU(unsigned long, this_cpu_off);
 
@@ -81,9 +72,6 @@ do {								\
 		       (src), (size));				\
 } while (0)
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 /* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
 #define __percpu_seg "%%fs:"
 #else  /* !SMP */
diff --git a/include/asm-x86/percpu_64.h b/include/asm-x86/percpu_64.h
index 5abd48270101..24fe7075248d 100644
--- a/include/asm-x86/percpu_64.h
+++ b/include/asm-x86/percpu_64.h
@@ -16,15 +16,6 @@
 
 #define per_cpu_offset(x) (__per_cpu_offset(x))
 
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_internodealigned_in_smp
-
 /* var is in discarded region: offset to particular copy we want */
 #define per_cpu(var, cpu) (*({				\
 	extern int simple_identifier_##var(void);	\
@@ -49,11 +40,6 @@ extern void setup_per_cpu_areas(void);
 
 #else /* ! SMP */
 
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
-
 #define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
 #define __get_cpu_var(var)			per_cpu__##var
 #define __raw_get_cpu_var(var)			per_cpu__##var
@@ -62,7 +48,4 @@ extern void setup_per_cpu_areas(void);
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
 #endif /* _ASM_X8664_PERCPU_H_ */
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 926adaae0f96..00412bb494c4 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -9,6 +9,30 @@
 
 #include <asm/percpu.h>
 
+#ifndef PER_CPU_ATTRIBUTES
+#define PER_CPU_ATTRIBUTES
+#endif
+
+#ifdef CONFIG_SMP
+#define DEFINE_PER_CPU(type, name)					\
+	__attribute__((__section__(".data.percpu")))			\
+	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
+
+#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)			\
+	__attribute__((__section__(".data.percpu.shared_aligned")))	\
+	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name		\
+	____cacheline_aligned_in_smp
+#else
+#define DEFINE_PER_CPU(type, name)					\
+	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
+
+#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		      \
+	DEFINE_PER_CPU(type, name)
+#endif
+
+#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
+#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
+
 /* Enough to cover all DEFINE_PER_CPUs in kernel, including modules. */
 #ifndef PERCPU_ENOUGH_ROOM
 #ifdef CONFIG_MODULES
-- 
cgit v1.2.3


From acdac87202a408133ee8f7985076de9d2e0dc5ab Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:52 +0100
Subject: percpu: make the asm-generic/percpu.h more "generic"

- add support for PER_CPU_ATTRIBUTES

- fix generic smp percpu_modcopy to use per_cpu_offset() macro.

Add the ability to use generic/percpu even if the arch needs to override
several aspects of its operations. This will enable the use of generic
percpu.h for all arches.

An arch may define:

__per_cpu_offset	Do not use the generic pointer array. Arch must
			define per_cpu_offset(cpu) (used by x86_64, s390).

__my_cpu_offset		Can be defined to provide an optimized way to determine
			the offset for variables of the currently executing
			processor. Used by ia64, x86_64, x86_32, sparc64, s/390.

SHIFT_PTR(ptr, offset)	If an arch defines it then special handling
			of pointer arithmentic may be implemented. Used
			by s/390.

(Some of these special percpu arch implementations may be later consolidated
so that there are less cases to deal with.)

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/ia64/kernel/module.c    |  2 +-
 include/asm-generic/percpu.h | 74 +++++++++++++++++++++++++++++++++++++-------
 2 files changed, 64 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c
index 196287928bae..e699eb6c44be 100644
--- a/arch/ia64/kernel/module.c
+++ b/arch/ia64/kernel/module.c
@@ -947,7 +947,7 @@ percpu_modcopy (void *pcpudst, const void *src, unsigned long size)
 {
 	unsigned int i;
 	for_each_possible_cpu(i) {
-		memcpy(pcpudst + __per_cpu_offset[i], src, size);
+		memcpy(pcpudst + per_cpu_offset(i), src, size);
 	}
 }
 #endif /* CONFIG_SMP */
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index e038f13594e5..c41b1a731129 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -3,35 +3,87 @@
 #include <linux/compiler.h>
 #include <linux/threads.h>
 
+/*
+ * Determine the real variable name from the name visible in the
+ * kernel sources.
+ */
+#define per_cpu_var(var) per_cpu__##var
+
 #ifdef CONFIG_SMP
 
+/*
+ * per_cpu_offset() is the offset that has to be added to a
+ * percpu variable to get to the instance for a certain processor.
+ *
+ * Most arches use the __per_cpu_offset array for those offsets but
+ * some arches have their own ways of determining the offset (x86_64, s390).
+ */
+#ifndef __per_cpu_offset
 extern unsigned long __per_cpu_offset[NR_CPUS];
 
 #define per_cpu_offset(x) (__per_cpu_offset[x])
+#endif
+
+/*
+ * Determine the offset for the currently active processor.
+ * An arch may define __my_cpu_offset to provide a more effective
+ * means of obtaining the offset to the per cpu variables of the
+ * current processor.
+ */
+#ifndef __my_cpu_offset
+#define __my_cpu_offset per_cpu_offset(raw_smp_processor_id())
+#define my_cpu_offset per_cpu_offset(smp_processor_id())
+#else
+#define my_cpu_offset __my_cpu_offset
+#endif
+
+/*
+ * Add a offset to a pointer but keep the pointer as is.
+ *
+ * Only S390 provides its own means of moving the pointer.
+ */
+#ifndef SHIFT_PERCPU_PTR
+#define SHIFT_PERCPU_PTR(__p, __offset)	RELOC_HIDE((__p), (__offset))
+#endif
 
-/* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*({				\
-	extern int simple_identifier_##var(void);	\
-	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]); }))
-#define __get_cpu_var(var) per_cpu(var, smp_processor_id())
-#define __raw_get_cpu_var(var) per_cpu(var, raw_smp_processor_id())
+/*
+ * A percpu variable may point to a discarded reghions. The following are
+ * established ways to produce a usable pointer from the percpu variable
+ * offset.
+ */
+#define per_cpu(var, cpu) \
+	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), per_cpu_offset(cpu)))
+#define __get_cpu_var(var) \
+	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), my_cpu_offset))
+#define __raw_get_cpu_var(var) \
+	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), __my_cpu_offset))
+
+
+#ifdef CONFIG_ARCH_SETS_UP_PER_CPU_AREA
+extern void setup_per_cpu_areas(void);
+#endif
 
 /* A macro to avoid #include hell... */
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
 	unsigned int __i;					\
 	for_each_possible_cpu(__i)				\
-		memcpy((pcpudst)+__per_cpu_offset[__i],		\
+		memcpy((pcpudst)+per_cpu_offset(__i),		\
 		       (src), (size));				\
 } while (0)
 #else /* ! SMP */
 
-#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
-#define __get_cpu_var(var)			per_cpu__##var
-#define __raw_get_cpu_var(var)			per_cpu__##var
+#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu_var(var)))
+#define __get_cpu_var(var)			per_cpu_var(var)
+#define __raw_get_cpu_var(var)			per_cpu_var(var)
 
 #endif	/* SMP */
 
-#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
+#ifndef PER_CPU_ATTRIBUTES
+#define PER_CPU_ATTRIBUTES
+#endif
+
+#define DECLARE_PER_CPU(type, name) extern PER_CPU_ATTRIBUTES \
+					__typeof__(type) per_cpu_var(name)
 
 #endif /* _ASM_GENERIC_PERCPU_H_ */
-- 
cgit v1.2.3


From 0af8a5ccc51ee2269712c90ab09c91b0150f4224 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:52 +0100
Subject: x86_32: use generic percpu.h

x86_32 only provides a special way to obtain the local per cpu area offset
via x86_read_percpu. Otherwise it can fully use the generic handling.

Cc: ak@suse.de
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/percpu_32.h | 30 +++++++++---------------------
 1 file changed, 9 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/percpu_32.h b/include/asm-x86/percpu_32.h
index 77bd0045f331..e62ce2fe2c9c 100644
--- a/include/asm-x86/percpu_32.h
+++ b/include/asm-x86/percpu_32.h
@@ -42,26 +42,7 @@
  */
 #ifdef CONFIG_SMP
 
-/* This is used for other cpus to find our section. */
-extern unsigned long __per_cpu_offset[];
-
-#define per_cpu_offset(x) (__per_cpu_offset[x])
-
-#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
-/* We can use this directly for local CPU (faster). */
-DECLARE_PER_CPU(unsigned long, this_cpu_off);
-
-/* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*({				\
-	extern int simple_indentifier_##var(void);	\
-	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]); }))
-
-#define __raw_get_cpu_var(var) (*({					\
-	extern int simple_indentifier_##var(void);			\
-	RELOC_HIDE(&per_cpu__##var, x86_read_percpu(this_cpu_off));	\
-}))
-
-#define __get_cpu_var(var) __raw_get_cpu_var(var)
+#define __my_cpu_offset x86_read_percpu(this_cpu_off)
 
 /* A macro to avoid #include hell... */
 #define percpu_modcopy(pcpudst, src, size)			\
@@ -74,11 +55,18 @@ do {								\
 
 /* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
 #define __percpu_seg "%%fs:"
+
 #else  /* !SMP */
-#include <asm-generic/percpu.h>
+
 #define __percpu_seg ""
+
 #endif	/* SMP */
 
+#include <asm-generic/percpu.h>
+
+/* We can use this directly for local CPU (faster). */
+DECLARE_PER_CPU(unsigned long, this_cpu_off);
+
 /* For arch-specific code, we can use direct single-insn ops (they
  * don't give an lvalue though). */
 extern void __bad_percpu_size(void);
-- 
cgit v1.2.3


From 072a1e6953d2caffbad7d217ae51cdc61125960f Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:52 +0100
Subject: x86: use generic percpu on 64-bit

x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/percpu_64.h | 23 ++---------------------
 1 file changed, 2 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/percpu_64.h b/include/asm-x86/percpu_64.h
index 24fe7075248d..e7f3f6d21759 100644
--- a/include/asm-x86/percpu_64.h
+++ b/include/asm-x86/percpu_64.h
@@ -12,21 +12,10 @@
 #include <asm/pda.h>
 
 #define __per_cpu_offset(cpu) (cpu_pda(cpu)->data_offset)
-#define __my_cpu_offset() read_pda(data_offset)
+#define __my_cpu_offset read_pda(data_offset)
 
 #define per_cpu_offset(x) (__per_cpu_offset(x))
 
-/* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*({				\
-	extern int simple_identifier_##var(void);	\
-	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)); }))
-#define __get_cpu_var(var) (*({				\
-	extern int simple_identifier_##var(void);	\
-	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
-#define __raw_get_cpu_var(var) (*({			\
-	extern int simple_identifier_##var(void);	\
-	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
-
 /* A macro to avoid #include hell... */
 #define percpu_modcopy(pcpudst, src, size)			\
 do {								\
@@ -36,16 +25,8 @@ do {								\
 		       (src), (size));				\
 } while (0)
 
-extern void setup_per_cpu_areas(void);
-
-#else /* ! SMP */
-
-#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu__##var))
-#define __get_cpu_var(var)			per_cpu__##var
-#define __raw_get_cpu_var(var)			per_cpu__##var
-
 #endif	/* SMP */
 
-#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
+#include <asm-generic/percpu.h>
 
 #endif /* _ASM_X8664_PERCPU_H_ */
-- 
cgit v1.2.3


From 3334052a321aca0ffecb54244d666311f98f5487 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:32:53 +0100
Subject: x86: unify percpu.h

Form a single percpu.h from percpu_32.h and percpu_64.h. Both are now pretty
small so this is simply adding them together.

Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/percpu.h    | 145 ++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/percpu_32.h | 128 --------------------------------------
 include/asm-x86/percpu_64.h |  32 ----------
 3 files changed, 141 insertions(+), 164 deletions(-)
 delete mode 100644 include/asm-x86/percpu_32.h
 delete mode 100644 include/asm-x86/percpu_64.h

(limited to 'include')

diff --git a/include/asm-x86/percpu.h b/include/asm-x86/percpu.h
index a1aaad274cca..0dec00f27eb4 100644
--- a/include/asm-x86/percpu.h
+++ b/include/asm-x86/percpu.h
@@ -1,5 +1,142 @@
-#ifdef CONFIG_X86_32
-# include "percpu_32.h"
-#else
-# include "percpu_64.h"
+#ifndef _ASM_X86_PERCPU_H_
+#define _ASM_X86_PERCPU_H_
+
+#ifdef CONFIG_X86_64
+#include <linux/compiler.h>
+
+/* Same as asm-generic/percpu.h, except that we store the per cpu offset
+   in the PDA. Longer term the PDA and every per cpu variable
+   should be just put into a single section and referenced directly
+   from %gs */
+
+#ifdef CONFIG_SMP
+#include <asm/pda.h>
+
+#define __per_cpu_offset(cpu) (cpu_pda(cpu)->data_offset)
+#define __my_cpu_offset read_pda(data_offset)
+
+#define per_cpu_offset(x) (__per_cpu_offset(x))
+
 #endif
+#include <asm-generic/percpu.h>
+
+DECLARE_PER_CPU(struct x8664_pda, pda);
+
+#else /* CONFIG_X86_64 */
+
+#ifdef __ASSEMBLY__
+
+/*
+ * PER_CPU finds an address of a per-cpu variable.
+ *
+ * Args:
+ *    var - variable name
+ *    reg - 32bit register
+ *
+ * The resulting address is stored in the "reg" argument.
+ *
+ * Example:
+ *    PER_CPU(cpu_gdt_descr, %ebx)
+ */
+#ifdef CONFIG_SMP
+#define PER_CPU(var, reg)				\
+	movl %fs:per_cpu__##this_cpu_off, reg;		\
+	lea per_cpu__##var(reg), reg
+#define PER_CPU_VAR(var)	%fs:per_cpu__##var
+#else /* ! SMP */
+#define PER_CPU(var, reg)			\
+	movl $per_cpu__##var, reg
+#define PER_CPU_VAR(var)	per_cpu__##var
+#endif	/* SMP */
+
+#else /* ...!ASSEMBLY */
+
+/*
+ * PER_CPU finds an address of a per-cpu variable.
+ *
+ * Args:
+ *    var - variable name
+ *    cpu - 32bit register containing the current CPU number
+ *
+ * The resulting address is stored in the "cpu" argument.
+ *
+ * Example:
+ *    PER_CPU(cpu_gdt_descr, %ebx)
+ */
+#ifdef CONFIG_SMP
+
+#define __my_cpu_offset x86_read_percpu(this_cpu_off)
+
+/* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
+#define __percpu_seg "%%fs:"
+
+#else  /* !SMP */
+
+#define __percpu_seg ""
+
+#endif	/* SMP */
+
+#include <asm-generic/percpu.h>
+
+/* We can use this directly for local CPU (faster). */
+DECLARE_PER_CPU(unsigned long, this_cpu_off);
+
+/* For arch-specific code, we can use direct single-insn ops (they
+ * don't give an lvalue though). */
+extern void __bad_percpu_size(void);
+
+#define percpu_to_op(op,var,val)				\
+	do {							\
+		typedef typeof(var) T__;			\
+		if (0) { T__ tmp__; tmp__ = (val); }		\
+		switch (sizeof(var)) {				\
+		case 1:						\
+			asm(op "b %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		case 2:						\
+			asm(op "w %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		case 4:						\
+			asm(op "l %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		default: __bad_percpu_size();			\
+		}						\
+	} while (0)
+
+#define percpu_from_op(op,var)					\
+	({							\
+		typeof(var) ret__;				\
+		switch (sizeof(var)) {				\
+		case 1:						\
+			asm(op "b "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		case 2:						\
+			asm(op "w "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		case 4:						\
+			asm(op "l "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		default: __bad_percpu_size();			\
+		}						\
+		ret__; })
+
+#define x86_read_percpu(var) percpu_from_op("mov", per_cpu__##var)
+#define x86_write_percpu(var,val) percpu_to_op("mov", per_cpu__##var, val)
+#define x86_add_percpu(var,val) percpu_to_op("add", per_cpu__##var, val)
+#define x86_sub_percpu(var,val) percpu_to_op("sub", per_cpu__##var, val)
+#define x86_or_percpu(var,val) percpu_to_op("or", per_cpu__##var, val)
+#endif /* !__ASSEMBLY__ */
+#endif /* !CONFIG_X86_64 */
+#endif /* _ASM_X86_PERCPU_H_ */
diff --git a/include/asm-x86/percpu_32.h b/include/asm-x86/percpu_32.h
deleted file mode 100644
index e62ce2fe2c9c..000000000000
--- a/include/asm-x86/percpu_32.h
+++ /dev/null
@@ -1,128 +0,0 @@
-#ifndef __ARCH_I386_PERCPU__
-#define __ARCH_I386_PERCPU__
-
-#ifdef __ASSEMBLY__
-
-/*
- * PER_CPU finds an address of a per-cpu variable.
- *
- * Args:
- *    var - variable name
- *    reg - 32bit register
- *
- * The resulting address is stored in the "reg" argument.
- *
- * Example:
- *    PER_CPU(cpu_gdt_descr, %ebx)
- */
-#ifdef CONFIG_SMP
-#define PER_CPU(var, reg)				\
-	movl %fs:per_cpu__##this_cpu_off, reg;		\
-	lea per_cpu__##var(reg), reg
-#define PER_CPU_VAR(var)	%fs:per_cpu__##var
-#else /* ! SMP */
-#define PER_CPU(var, reg)			\
-	movl $per_cpu__##var, reg
-#define PER_CPU_VAR(var)	per_cpu__##var
-#endif	/* SMP */
-
-#else /* ...!ASSEMBLY */
-
-/*
- * PER_CPU finds an address of a per-cpu variable.
- *
- * Args:
- *    var - variable name
- *    cpu - 32bit register containing the current CPU number
- *
- * The resulting address is stored in the "cpu" argument.
- *
- * Example:
- *    PER_CPU(cpu_gdt_descr, %ebx)
- */
-#ifdef CONFIG_SMP
-
-#define __my_cpu_offset x86_read_percpu(this_cpu_off)
-
-/* A macro to avoid #include hell... */
-#define percpu_modcopy(pcpudst, src, size)			\
-do {								\
-	unsigned int __i;					\
-	for_each_possible_cpu(__i)				\
-		memcpy((pcpudst)+__per_cpu_offset[__i],		\
-		       (src), (size));				\
-} while (0)
-
-/* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
-#define __percpu_seg "%%fs:"
-
-#else  /* !SMP */
-
-#define __percpu_seg ""
-
-#endif	/* SMP */
-
-#include <asm-generic/percpu.h>
-
-/* We can use this directly for local CPU (faster). */
-DECLARE_PER_CPU(unsigned long, this_cpu_off);
-
-/* For arch-specific code, we can use direct single-insn ops (they
- * don't give an lvalue though). */
-extern void __bad_percpu_size(void);
-
-#define percpu_to_op(op,var,val)				\
-	do {							\
-		typedef typeof(var) T__;			\
-		if (0) { T__ tmp__; tmp__ = (val); }		\
-		switch (sizeof(var)) {				\
-		case 1:						\
-			asm(op "b %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		case 2:						\
-			asm(op "w %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		case 4:						\
-			asm(op "l %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		default: __bad_percpu_size();			\
-		}						\
-	} while (0)
-
-#define percpu_from_op(op,var)					\
-	({							\
-		typeof(var) ret__;				\
-		switch (sizeof(var)) {				\
-		case 1:						\
-			asm(op "b "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		case 2:						\
-			asm(op "w "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		case 4:						\
-			asm(op "l "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		default: __bad_percpu_size();			\
-		}						\
-		ret__; })
-
-#define x86_read_percpu(var) percpu_from_op("mov", per_cpu__##var)
-#define x86_write_percpu(var,val) percpu_to_op("mov", per_cpu__##var, val)
-#define x86_add_percpu(var,val) percpu_to_op("add", per_cpu__##var, val)
-#define x86_sub_percpu(var,val) percpu_to_op("sub", per_cpu__##var, val)
-#define x86_or_percpu(var,val) percpu_to_op("or", per_cpu__##var, val)
-#endif /* !__ASSEMBLY__ */
-
-#endif /* __ARCH_I386_PERCPU__ */
diff --git a/include/asm-x86/percpu_64.h b/include/asm-x86/percpu_64.h
deleted file mode 100644
index e7f3f6d21759..000000000000
--- a/include/asm-x86/percpu_64.h
+++ /dev/null
@@ -1,32 +0,0 @@
-#ifndef _ASM_X8664_PERCPU_H_
-#define _ASM_X8664_PERCPU_H_
-#include <linux/compiler.h>
-
-/* Same as asm-generic/percpu.h, except that we store the per cpu offset
-   in the PDA. Longer term the PDA and every per cpu variable
-   should be just put into a single section and referenced directly
-   from %gs */
-
-#ifdef CONFIG_SMP
-
-#include <asm/pda.h>
-
-#define __per_cpu_offset(cpu) (cpu_pda(cpu)->data_offset)
-#define __my_cpu_offset read_pda(data_offset)
-
-#define per_cpu_offset(x) (__per_cpu_offset(x))
-
-/* A macro to avoid #include hell... */
-#define percpu_modcopy(pcpudst, src, size)			\
-do {								\
-	unsigned int __i;					\
-	for_each_possible_cpu(__i)				\
-		memcpy((pcpudst)+__per_cpu_offset(__i),		\
-		       (src), (size));				\
-} while (0)
-
-#endif	/* SMP */
-
-#include <asm-generic/percpu.h>
-
-#endif /* _ASM_X8664_PERCPU_H_ */
-- 
cgit v1.2.3


From 8c1c9356429741a82ff176d0f3400fb9e06b2a30 Mon Sep 17 00:00:00 2001
From: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Date: Wed, 30 Jan 2008 13:32:53 +0100
Subject: x86: kprobes: add kprobes smoke tests that run on boot

Here is a quick and naive smoke test for kprobes. This is intended to
just verify if some unrelated change broke the *probes subsystem. It is
self contained, architecture agnostic and isn't of any great use by itself.

This needs to be built in the kernel and runs a basic set of tests to
verify if kprobes, jprobes and kretprobes run fine on the kernel. In case
of an error, it'll print out a message with a "BUG" prefix.

This is a start; we intend to add more tests to this bucket over time.

Thanks to Jim Keniston and Masami Hiramatsu for comments and suggestions.

Tested on x86 (32/64) and powerpc.

Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/kprobes.h |  10 +++
 kernel/Makefile         |   1 +
 kernel/kprobes.c        |   2 +
 kernel/test_kprobes.c   | 216 ++++++++++++++++++++++++++++++++++++++++++++++++
 lib/Kconfig.debug       |  12 +++
 5 files changed, 241 insertions(+)
 create mode 100644 kernel/test_kprobes.c

(limited to 'include')

diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 81891581e89b..6168c0a44172 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -182,6 +182,15 @@ static inline void kretprobe_assert(struct kretprobe_instance *ri,
 	}
 }
 
+#ifdef CONFIG_KPROBES_SANITY_TEST
+extern int init_test_probes(void);
+#else
+static inline int init_test_probes(void)
+{
+	return 0;
+}
+#endif /* CONFIG_KPROBES_SANITY_TEST */
+
 extern spinlock_t kretprobe_lock;
 extern struct mutex kprobe_mutex;
 extern int arch_prepare_kprobe(struct kprobe *p);
@@ -227,6 +236,7 @@ void unregister_kretprobe(struct kretprobe *rp);
 
 void kprobe_flush_task(struct task_struct *tk);
 void recycle_rp_inst(struct kretprobe_instance *ri, struct hlist_head *head);
+
 #else /* CONFIG_KPROBES */
 
 #define __kprobes	/**/
diff --git a/kernel/Makefile b/kernel/Makefile
index 390d42146267..62015c3d8d91 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_CPUSETS) += cpuset.o
 obj-$(CONFIG_CGROUP_NS) += ns_cgroup.o
 obj-$(CONFIG_IKCONFIG) += configs.o
 obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
+obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
 obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
 obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
 obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index e3a5d817ac9b..d0493eafea3e 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -824,6 +824,8 @@ static int __init init_kprobes(void)
 	if (!err)
 		err = register_die_notifier(&kprobe_exceptions_nb);
 
+	if (!err)
+		init_test_probes();
 	return err;
 }
 
diff --git a/kernel/test_kprobes.c b/kernel/test_kprobes.c
new file mode 100644
index 000000000000..88cdb109e13c
--- /dev/null
+++ b/kernel/test_kprobes.c
@@ -0,0 +1,216 @@
+/*
+ * test_kprobes.c - simple sanity test for *probes
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/random.h>
+
+#define div_factor 3
+
+static u32 rand1, preh_val, posth_val, jph_val;
+static int errors, handler_errors, num_tests;
+
+static noinline u32 kprobe_target(u32 value)
+{
+	/*
+	 * gcc ignores noinline on some architectures unless we stuff
+	 * sufficient lard into the function. The get_kprobe() here is
+	 * just for that.
+	 *
+	 * NOTE: We aren't concerned about the correctness of get_kprobe()
+	 * here; hence, this call is neither under !preempt nor with the
+	 * kprobe_mutex held. This is fine(tm)
+	 */
+	if (get_kprobe((void *)0xdeadbeef))
+		printk(KERN_INFO "Kprobe smoke test: probe on 0xdeadbeef!\n");
+
+	return (value / div_factor);
+}
+
+static int kp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	preh_val = (rand1 / div_factor);
+	return 0;
+}
+
+static void kp_post_handler(struct kprobe *p, struct pt_regs *regs,
+		unsigned long flags)
+{
+	if (preh_val != (rand1 / div_factor)) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in post_handler\n");
+	}
+	posth_val = preh_val + div_factor;
+}
+
+static struct kprobe kp = {
+	.symbol_name = "kprobe_target",
+	.pre_handler = kp_pre_handler,
+	.post_handler = kp_post_handler
+};
+
+static int test_kprobe(void)
+{
+	int ret;
+
+	ret = register_kprobe(&kp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_kprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_kprobe(&kp);
+
+	if (preh_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kprobe pre_handler not called\n");
+		handler_errors++;
+	}
+
+	if (posth_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kprobe post_handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+
+static u32 j_kprobe_target(u32 value)
+{
+	if (value != rand1) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in jprobe handler\n");
+	}
+
+	jph_val = rand1;
+	jprobe_return();
+	return 0;
+}
+
+static struct jprobe jp = {
+	.entry		= j_kprobe_target,
+	.kp.symbol_name = "kprobe_target"
+};
+
+static int test_jprobe(void)
+{
+	int ret;
+
+	ret = register_jprobe(&jp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_jprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_jprobe(&jp);
+	if (jph_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"jprobe handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_KRETPROBES
+static u32 krph_val;
+
+static int return_handler(struct kretprobe_instance *ri, struct pt_regs *regs)
+{
+	unsigned long ret = regs_return_value(regs);
+
+	if (ret != (rand1 / div_factor)) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in kretprobe handler\n");
+	}
+
+	krph_val = (rand1 / div_factor);
+	return 0;
+}
+
+static struct kretprobe rp = {
+	.handler	= return_handler,
+	.kp.symbol_name = "kprobe_target"
+};
+
+static int test_kretprobe(void)
+{
+	int ret;
+
+	ret = register_kretprobe(&rp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_kretprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_kretprobe(&rp);
+	if (krph_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kretprobe handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+#endif /* CONFIG_KRETPROBES */
+
+int init_test_probes(void)
+{
+	int ret;
+
+	do {
+		rand1 = random32();
+	} while (rand1 <= div_factor);
+
+	printk(KERN_INFO "Kprobe smoke test started\n");
+	num_tests++;
+	ret = test_kprobe();
+	if (ret < 0)
+		errors++;
+
+	num_tests++;
+	ret = test_jprobe();
+	if (ret < 0)
+		errors++;
+
+#ifdef CONFIG_KRETPROBES
+	num_tests++;
+	ret = test_kretprobe();
+	if (ret < 0)
+		errors++;
+#endif /* CONFIG_KRETPROBES */
+
+	if (errors)
+		printk(KERN_ERR "BUG: Kprobe smoke test: %d out of "
+				"%d tests failed\n", errors, num_tests);
+	else if (handler_errors)
+		printk(KERN_ERR "BUG: Kprobe smoke test: %d error(s) "
+				"running handlers\n", handler_errors);
+	else
+		printk(KERN_INFO "Kprobe smoke test passed successfully\n");
+
+	return 0;
+}
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index c4ecb2994ba3..f535b9b5eb00 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -494,6 +494,18 @@ config RCU_TORTURE_TEST
 	  Say M if you want the RCU torture tests to build as a module.
 	  Say N if you are unsure.
 
+config KPROBES_SANITY_TEST
+	bool "Kprobes sanity tests"
+	depends on DEBUG_KERNEL
+	depends on KPROBES
+	default n
+	help
+	  This option provides for testing basic kprobes functionality on
+	  boot. A sample kprobe, jprobe and kretprobe are inserted and
+	  verified for functionality.
+
+	  Say N if you are unsure.
+
 config LKDTM
 	tristate "Linux Kernel Dump Test Tool Module"
 	depends on DEBUG_KERNEL
-- 
cgit v1.2.3


From 58f6f6ea1d5031f998b5e7d640d7bc45ed86d0c8 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:54 +0100
Subject: x86: migration helpers for KVM

migration helpers for KVM.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 41c002440de7..cc549bf819a4 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -3,6 +3,10 @@
 
 #include <asm/processor-flags.h>
 
+/* migration helpers, for KVM - will be removed in 2.6.25: */
+#include <asm/vm86.h>
+#define Xgt_desc_struct	desc_ptr
+
 /* Forward declaration, a strange C thing */
 struct task_struct;
 struct mm_struct;
-- 
cgit v1.2.3


From 87e8407f9ad2a2df901c4b690ab0a2bf0fb168c5 Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Wed, 30 Jan 2008 13:32:54 +0100
Subject: x86, ptrace: add bts_struct size to status command

Return the size of bts_struct in the PTRACE_BTS_STATUS command.
Change types to u32.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/ptrace.c     |  2 ++
 include/asm-x86/ptrace-abi.h | 11 ++++++++---
 2 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index e19a91db9b35..96286df1bb81 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -787,6 +787,8 @@ static int ptrace_bts_status(struct task_struct *child,
 			cfg.flags |= PTRACE_BTS_O_SCHED;
 	}
 
+	cfg.bts_size = sizeof(struct bts_struct);
+
 	if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
 		return -EFAULT;
 
diff --git a/include/asm-x86/ptrace-abi.h b/include/asm-x86/ptrace-abi.h
index 08a12b790a77..81a8ee4c55fc 100644
--- a/include/asm-x86/ptrace-abi.h
+++ b/include/asm-x86/ptrace-abi.h
@@ -81,16 +81,21 @@
 #define PTRACE_SINGLEBLOCK	33	/* resume execution until next branch */
 
 #ifndef __ASSEMBLY__
+
+#include <asm/types.h>
+
 /* configuration/status structure used in PTRACE_BTS_CONFIG and
    PTRACE_BTS_STATUS commands.
 */
 struct ptrace_bts_config {
 	/* requested or actual size of BTS buffer in bytes */
-	unsigned int size;
+	u32 size;
 	/* bitmask of below flags */
-	unsigned int flags;
+	u32 flags;
 	/* buffer overflow signal */
-	unsigned int signal;
+	u32 signal;
+	/* actual size of bts_struct in bytes */
+	u32 bts_size;
 };
 #endif
 
-- 
cgit v1.2.3


From 82bc03fc158e28c90d7ed9919410776039cb4e14 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:54 +0100
Subject: x86: add PWT to NOCACHE flags

add PWT bit to NOCACHE flags. No real difference to CPUs, but needed
later for PAT.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_32.h | 2 +-
 include/asm-x86/pgtable_64.h | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index a8be1ee15a99..16da5d5cce40 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -156,7 +156,7 @@ void paging_init(void);
 extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define __PAGE_KERNEL_RO		(__PAGE_KERNEL & ~_PAGE_RW)
 #define __PAGE_KERNEL_RX		(__PAGE_KERNEL_EXEC & ~_PAGE_RW)
-#define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD)
+#define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_LARGE		(__PAGE_KERNEL | _PAGE_PSE)
 #define __PAGE_KERNEL_LARGE_EXEC	(__PAGE_KERNEL_EXEC | _PAGE_PSE)
 
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 3f280105d7b3..9c9cddf5138b 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -189,13 +189,13 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define __PAGE_KERNEL_EXEC \
 	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
 #define __PAGE_KERNEL_NOCACHE \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_PCD | _PAGE_ACCESSED | _PAGE_NX)
+	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_PCD | _PAGE_PWT | _PAGE_ACCESSED | _PAGE_NX)
 #define __PAGE_KERNEL_RO \
 	(_PAGE_PRESENT | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
 #define __PAGE_KERNEL_VSYSCALL \
 	(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
 #define __PAGE_KERNEL_VSYSCALL_NOCACHE \
-	(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_PCD)
+	(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_LARGE \
 	(__PAGE_KERNEL | _PAGE_PSE)
 #define __PAGE_KERNEL_LARGE_EXEC \
-- 
cgit v1.2.3


From 6c3866558213ff706d8331053386915371ad63ec Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:55 +0100
Subject: x86: move all asm/pgtable constants into one place

32 and 64-bit use the same flags for pagetable entries, so make them all common.

[ mingo@elte.hu: fixes ]

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 119 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/pgtable_32.h | 110 ---------------------------------------
 include/asm-x86/pgtable_64.h |  91 ---------------------------------
 3 files changed, 119 insertions(+), 201 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 1039140652af..5b858f0285a8 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -1,5 +1,124 @@
+#ifndef _ASM_X86_PGTABLE_H
+#define _ASM_X86_PGTABLE_H
+
+#define USER_PTRS_PER_PGD	((TASK_SIZE-1)/PGDIR_SIZE+1)
+#define FIRST_USER_ADDRESS	0
+
+#define _PAGE_BIT_PRESENT	0
+#define _PAGE_BIT_RW		1
+#define _PAGE_BIT_USER		2
+#define _PAGE_BIT_PWT		3
+#define _PAGE_BIT_PCD		4
+#define _PAGE_BIT_ACCESSED	5
+#define _PAGE_BIT_DIRTY		6
+#define _PAGE_BIT_FILE		6
+#define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page */
+#define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
+#define _PAGE_BIT_UNUSED1	9	/* available for programmer */
+#define _PAGE_BIT_UNUSED2	10
+#define _PAGE_BIT_UNUSED3	11
+#define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
+
+#define _PAGE_PRESENT	(_AC(1, UL)<<_PAGE_BIT_PRESENT)
+#define _PAGE_RW	(_AC(1, UL)<<_PAGE_BIT_RW)
+#define _PAGE_USER	(_AC(1, UL)<<_PAGE_BIT_USER)
+#define _PAGE_PWT	(_AC(1, UL)<<_PAGE_BIT_PWT)
+#define _PAGE_PCD	(_AC(1, UL)<<_PAGE_BIT_PCD)
+#define _PAGE_ACCESSED	(_AC(1, UL)<<_PAGE_BIT_ACCESSED)
+#define _PAGE_DIRTY	(_AC(1, UL)<<_PAGE_BIT_DIRTY)
+#define _PAGE_PSE	(_AC(1, UL)<<_PAGE_BIT_PSE)	/* 2MB page */
+#define _PAGE_GLOBAL	(_AC(1, UL)<<_PAGE_BIT_GLOBAL)	/* Global TLB entry */
+#define _PAGE_UNUSED1	(_AC(1, UL)<<_PAGE_BIT_UNUSED1)
+#define _PAGE_UNUSED2	(_AC(1, UL)<<_PAGE_BIT_UNUSED2)
+#define _PAGE_UNUSED3	(_AC(1, UL)<<_PAGE_BIT_UNUSED3)
+
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+#define _PAGE_NX	(_AC(1, ULL) << _PAGE_BIT_NX)
+#else
+#define _PAGE_NX	0
+#endif
+
+/* If _PAGE_PRESENT is clear, we use these: */
+#define _PAGE_FILE	_PAGE_DIRTY	/* nonlinear file mapping, saved PTE; unset:swap */
+#define _PAGE_PROTNONE	_PAGE_PSE	/* if the user mapped it with PROT_NONE;
+					   pte_present gives true */
+
+#define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+
+#define PAGE_SHARED_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_COPY_EXEC		__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY		PAGE_COPY_NOEXEC
+#define PAGE_READONLY		__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_READONLY_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+
+#ifdef CONFIG_X86_32
+#define _PAGE_KERNEL_EXEC \
+	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
+#define _PAGE_KERNEL (_PAGE_KERNEL_EXEC | _PAGE_NX)
+
+#ifndef __ASSEMBLY__
+extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
+#endif	/* __ASSEMBLY__ */
+#else
+#define __PAGE_KERNEL_EXEC						\
+	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
+#define __PAGE_KERNEL		(__PAGE_KERNEL_EXEC | _PAGE_NX)
+#endif
+
+#define __PAGE_KERNEL_RO		(__PAGE_KERNEL & ~_PAGE_RW)
+#define __PAGE_KERNEL_RX		(__PAGE_KERNEL_EXEC & ~_PAGE_RW)
+#define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
+#define __PAGE_KERNEL_VSYSCALL		(__PAGE_KERNEL_RX | _PAGE_USER)
+#define __PAGE_KERNEL_VSYSCALL_NOCACHE	(__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT)
+#define __PAGE_KERNEL_LARGE		(__PAGE_KERNEL | _PAGE_PSE)
+#define __PAGE_KERNEL_LARGE_EXEC	(__PAGE_KERNEL_EXEC | _PAGE_PSE)
+
+#ifdef CONFIG_X86_32
+# define MAKE_GLOBAL(x)			__pgprot((x))
+#else
+# define MAKE_GLOBAL(x)			__pgprot((x) | _PAGE_GLOBAL)
+#endif
+
+#define PAGE_KERNEL			MAKE_GLOBAL(__PAGE_KERNEL)
+#define PAGE_KERNEL_RO			MAKE_GLOBAL(__PAGE_KERNEL_RO)
+#define PAGE_KERNEL_EXEC		MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
+#define PAGE_KERNEL_RX			MAKE_GLOBAL(__PAGE_KERNEL_RX)
+#define PAGE_KERNEL_NOCACHE		MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
+#define PAGE_KERNEL_LARGE		MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
+#define PAGE_KERNEL_LARGE_EXEC		MAKE_GLOBAL(__PAGE_KERNEL_LARGE_EXEC)
+#define PAGE_KERNEL_VSYSCALL		MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
+#define PAGE_KERNEL_VSYSCALL_NOCACHE	MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL_NOCACHE)
+
+/*         xwr */
+#define __P000	PAGE_NONE
+#define __P001	PAGE_READONLY
+#define __P010	PAGE_COPY
+#define __P011	PAGE_COPY
+#define __P100	PAGE_READONLY_EXEC
+#define __P101	PAGE_READONLY_EXEC
+#define __P110	PAGE_COPY_EXEC
+#define __P111	PAGE_COPY_EXEC
+
+#define __S000	PAGE_NONE
+#define __S001	PAGE_READONLY
+#define __S010	PAGE_SHARED
+#define __S011	PAGE_SHARED
+#define __S100	PAGE_READONLY_EXEC
+#define __S101	PAGE_READONLY_EXEC
+#define __S110	PAGE_SHARED_EXEC
+#define __S111	PAGE_SHARED_EXEC
+
 #ifdef CONFIG_X86_32
 # include "pgtable_32.h"
 #else
 # include "pgtable_64.h"
 #endif
+
+#endif	/* _ASM_X86_PGTABLE_H */
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 16da5d5cce40..3b40cb6dea18 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -58,9 +58,6 @@ void paging_init(void);
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
 #define PGDIR_MASK	(~(PGDIR_SIZE-1))
 
-#define USER_PTRS_PER_PGD	(TASK_SIZE/PGDIR_SIZE)
-#define FIRST_USER_ADDRESS	0
-
 #define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
 #define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
 
@@ -84,113 +81,6 @@ void paging_init(void);
 # define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
 #endif
 
-/*
- * _PAGE_PSE set in the page directory entry just means that
- * the page directory entry points directly to a 4MB-aligned block of
- * memory. 
- */
-#define _PAGE_BIT_PRESENT	0
-#define _PAGE_BIT_RW		1
-#define _PAGE_BIT_USER		2
-#define _PAGE_BIT_PWT		3
-#define _PAGE_BIT_PCD		4
-#define _PAGE_BIT_ACCESSED	5
-#define _PAGE_BIT_DIRTY		6
-#define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page, Pentium+, if present.. */
-#define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
-#define _PAGE_BIT_UNUSED1	9	/* available for programmer */
-#define _PAGE_BIT_UNUSED2	10
-#define _PAGE_BIT_UNUSED3	11
-#define _PAGE_BIT_NX		63
-
-#define _PAGE_PRESENT	0x001
-#define _PAGE_RW	0x002
-#define _PAGE_USER	0x004
-#define _PAGE_PWT	0x008
-#define _PAGE_PCD	0x010
-#define _PAGE_ACCESSED	0x020
-#define _PAGE_DIRTY	0x040
-#define _PAGE_PSE	0x080	/* 4 MB (or 2MB) page, Pentium+, if present.. */
-#define _PAGE_GLOBAL	0x100	/* Global TLB entry PPro+ */
-#define _PAGE_UNUSED1	0x200	/* available for programmer */
-#define _PAGE_UNUSED2	0x400
-#define _PAGE_UNUSED3	0x800
-
-/* If _PAGE_PRESENT is clear, we use these: */
-#define _PAGE_FILE	0x040	/* nonlinear file mapping, saved PTE; unset:swap */
-#define _PAGE_PROTNONE	0x080	/* if the user mapped it with PROT_NONE;
-				   pte_present gives true */
-#ifdef CONFIG_X86_PAE
-#define _PAGE_NX	(1ULL<<_PAGE_BIT_NX)
-#else
-#define _PAGE_NX	0
-#endif
-
-#define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
-
-#define PAGE_NONE \
-	__pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
-#define PAGE_SHARED \
-	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
-
-#define PAGE_SHARED_EXEC \
-	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
-#define PAGE_COPY_NOEXEC \
-	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_COPY_EXEC \
-	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
-#define PAGE_COPY \
-	PAGE_COPY_NOEXEC
-#define PAGE_READONLY \
-	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_READONLY_EXEC \
-	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
-
-#define _PAGE_KERNEL \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
-#define _PAGE_KERNEL_EXEC \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
-
-extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
-#define __PAGE_KERNEL_RO		(__PAGE_KERNEL & ~_PAGE_RW)
-#define __PAGE_KERNEL_RX		(__PAGE_KERNEL_EXEC & ~_PAGE_RW)
-#define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
-#define __PAGE_KERNEL_LARGE		(__PAGE_KERNEL | _PAGE_PSE)
-#define __PAGE_KERNEL_LARGE_EXEC	(__PAGE_KERNEL_EXEC | _PAGE_PSE)
-
-#define PAGE_KERNEL		__pgprot(__PAGE_KERNEL)
-#define PAGE_KERNEL_RO		__pgprot(__PAGE_KERNEL_RO)
-#define PAGE_KERNEL_EXEC	__pgprot(__PAGE_KERNEL_EXEC)
-#define PAGE_KERNEL_RX		__pgprot(__PAGE_KERNEL_RX)
-#define PAGE_KERNEL_NOCACHE	__pgprot(__PAGE_KERNEL_NOCACHE)
-#define PAGE_KERNEL_LARGE	__pgprot(__PAGE_KERNEL_LARGE)
-#define PAGE_KERNEL_LARGE_EXEC	__pgprot(__PAGE_KERNEL_LARGE_EXEC)
-
-/*
- * The i386 can't do page protection for execute, and considers that
- * the same are read. Also, write permissions imply read permissions.
- * This is the closest we can get..
- */
-#define __P000	PAGE_NONE
-#define __P001	PAGE_READONLY
-#define __P010	PAGE_COPY
-#define __P011	PAGE_COPY
-#define __P100	PAGE_READONLY_EXEC
-#define __P101	PAGE_READONLY_EXEC
-#define __P110	PAGE_COPY_EXEC
-#define __P111	PAGE_COPY_EXEC
-
-#define __S000	PAGE_NONE
-#define __S001	PAGE_READONLY
-#define __S010	PAGE_SHARED
-#define __S011	PAGE_SHARED
-#define __S100	PAGE_READONLY_EXEC
-#define __S101	PAGE_READONLY_EXEC
-#define __S110	PAGE_SHARED_EXEC
-#define __S111	PAGE_SHARED_EXEC
-
 /*
  * Define this if things work differently on an i386 and an i486:
  * it will (on an i486) warn about kernel memory accesses that are
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 9c9cddf5138b..93eb4df2ec11 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -131,8 +131,6 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define PGDIR_SIZE	(_AC(1,UL) << PGDIR_SHIFT)
 #define PGDIR_MASK	(~(PGDIR_SIZE-1))
 
-#define USER_PTRS_PER_PGD	((TASK_SIZE-1)/PGDIR_SIZE+1)
-#define FIRST_USER_ADDRESS	0
 
 #define MAXMEM		 _AC(0x3fffffffffff, UL)
 #define VMALLOC_START    _AC(0xffffc20000000000, UL)
@@ -142,95 +140,6 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define MODULES_END      _AC(0xfffffffffff00000, UL)
 #define MODULES_LEN   (MODULES_END - MODULES_VADDR)
 
-#define _PAGE_BIT_PRESENT	0
-#define _PAGE_BIT_RW		1
-#define _PAGE_BIT_USER		2
-#define _PAGE_BIT_PWT		3
-#define _PAGE_BIT_PCD		4
-#define _PAGE_BIT_ACCESSED	5
-#define _PAGE_BIT_DIRTY		6
-#define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page */
-#define _PAGE_BIT_FILE		6
-#define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
-#define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
-
-#define _PAGE_PRESENT	(_AC(1, UL)<<_PAGE_BIT_PRESENT)
-#define _PAGE_RW	(_AC(1, UL)<<_PAGE_BIT_RW)
-#define _PAGE_USER	(_AC(1, UL)<<_PAGE_BIT_USER)
-#define _PAGE_PWT	(_AC(1, UL)<<_PAGE_BIT_PWT)
-#define _PAGE_PCD	(_AC(1, UL)<<_PAGE_BIT_PCD)
-#define _PAGE_ACCESSED	(_AC(1, UL)<<_PAGE_BIT_ACCESSED)
-#define _PAGE_DIRTY	(_AC(1, UL)<<_PAGE_BIT_DIRTY)
-/* 2MB page */
-#define _PAGE_PSE	(_AC(1, UL)<<_PAGE_BIT_PSE)
-/* nonlinear file mapping, saved PTE; unset:swap */
-#define _PAGE_FILE	(_AC(1, UL)<<_PAGE_BIT_FILE)
-/* Global TLB entry */
-#define _PAGE_GLOBAL	(_AC(1, UL)<<_PAGE_BIT_GLOBAL)
-
-#define _PAGE_PROTNONE	0x080	/* If not present */
-#define _PAGE_NX        (_AC(1, UL)<<_PAGE_BIT_NX)
-
-#define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-
-#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
-
-#define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_SHARED_EXEC __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
-#define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_COPY PAGE_COPY_NOEXEC
-#define PAGE_COPY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
-#define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
-#define __PAGE_KERNEL \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
-#define __PAGE_KERNEL_EXEC \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
-#define __PAGE_KERNEL_NOCACHE \
-	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_PCD | _PAGE_PWT | _PAGE_ACCESSED | _PAGE_NX)
-#define __PAGE_KERNEL_RO \
-	(_PAGE_PRESENT | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
-#define __PAGE_KERNEL_VSYSCALL \
-	(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
-#define __PAGE_KERNEL_VSYSCALL_NOCACHE \
-	(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_PCD | _PAGE_PWT)
-#define __PAGE_KERNEL_LARGE \
-	(__PAGE_KERNEL | _PAGE_PSE)
-#define __PAGE_KERNEL_LARGE_EXEC \
-	(__PAGE_KERNEL_EXEC | _PAGE_PSE)
-
-#define MAKE_GLOBAL(x) __pgprot((x) | _PAGE_GLOBAL)
-
-#define PAGE_KERNEL MAKE_GLOBAL(__PAGE_KERNEL)
-#define PAGE_KERNEL_EXEC MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
-#define PAGE_KERNEL_RO MAKE_GLOBAL(__PAGE_KERNEL_RO)
-#define PAGE_KERNEL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
-#define PAGE_KERNEL_VSYSCALL32 __pgprot(__PAGE_KERNEL_VSYSCALL)
-#define PAGE_KERNEL_VSYSCALL MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
-#define PAGE_KERNEL_LARGE MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
-#define PAGE_KERNEL_VSYSCALL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL_NOCACHE)
-
-/*         xwr */
-#define __P000	PAGE_NONE
-#define __P001	PAGE_READONLY
-#define __P010	PAGE_COPY
-#define __P011	PAGE_COPY
-#define __P100	PAGE_READONLY_EXEC
-#define __P101	PAGE_READONLY_EXEC
-#define __P110	PAGE_COPY_EXEC
-#define __P111	PAGE_COPY_EXEC
-
-#define __S000	PAGE_NONE
-#define __S001	PAGE_READONLY
-#define __S010	PAGE_SHARED
-#define __S011	PAGE_SHARED
-#define __S100	PAGE_READONLY_EXEC
-#define __S101	PAGE_READONLY_EXEC
-#define __S110	PAGE_SHARED_EXEC
-#define __S111	PAGE_SHARED_EXEC
-
 #ifndef __ASSEMBLY__
 
 static inline unsigned long pgd_bad(pgd_t pgd)
-- 
cgit v1.2.3


From 61f38226def55d972cfd0e789971e952525ff8e5 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:55 +0100
Subject: x86/pgtable: fix constant sign extension problem

based on:

 Subject: x86/pgtable: fix constant sign extension problem
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 5b858f0285a8..73e2b47c055e 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -19,18 +19,18 @@
 #define _PAGE_BIT_UNUSED3	11
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
-#define _PAGE_PRESENT	(_AC(1, UL)<<_PAGE_BIT_PRESENT)
-#define _PAGE_RW	(_AC(1, UL)<<_PAGE_BIT_RW)
-#define _PAGE_USER	(_AC(1, UL)<<_PAGE_BIT_USER)
-#define _PAGE_PWT	(_AC(1, UL)<<_PAGE_BIT_PWT)
-#define _PAGE_PCD	(_AC(1, UL)<<_PAGE_BIT_PCD)
-#define _PAGE_ACCESSED	(_AC(1, UL)<<_PAGE_BIT_ACCESSED)
-#define _PAGE_DIRTY	(_AC(1, UL)<<_PAGE_BIT_DIRTY)
-#define _PAGE_PSE	(_AC(1, UL)<<_PAGE_BIT_PSE)	/* 2MB page */
-#define _PAGE_GLOBAL	(_AC(1, UL)<<_PAGE_BIT_GLOBAL)	/* Global TLB entry */
-#define _PAGE_UNUSED1	(_AC(1, UL)<<_PAGE_BIT_UNUSED1)
-#define _PAGE_UNUSED2	(_AC(1, UL)<<_PAGE_BIT_UNUSED2)
-#define _PAGE_UNUSED3	(_AC(1, UL)<<_PAGE_BIT_UNUSED3)
+#define _PAGE_PRESENT	(_AC(1, L)<<_PAGE_BIT_PRESENT)
+#define _PAGE_RW	(_AC(1, L)<<_PAGE_BIT_RW)
+#define _PAGE_USER	(_AC(1, L)<<_PAGE_BIT_USER)
+#define _PAGE_PWT	(_AC(1, L)<<_PAGE_BIT_PWT)
+#define _PAGE_PCD	(_AC(1, L)<<_PAGE_BIT_PCD)
+#define _PAGE_ACCESSED	(_AC(1, L)<<_PAGE_BIT_ACCESSED)
+#define _PAGE_DIRTY	(_AC(1, L)<<_PAGE_BIT_DIRTY)
+#define _PAGE_PSE	(_AC(1, L)<<_PAGE_BIT_PSE)	/* 2MB page */
+#define _PAGE_GLOBAL	(_AC(1, L)<<_PAGE_BIT_GLOBAL)	/* Global TLB entry */
+#define _PAGE_UNUSED1	(_AC(1, L)<<_PAGE_BIT_UNUSED1)
+#define _PAGE_UNUSED2	(_AC(1, L)<<_PAGE_BIT_UNUSED2)
+#define _PAGE_UNUSED3	(_AC(1, L)<<_PAGE_BIT_UNUSED3)
 
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
 #define _PAGE_NX	(_AC(1, ULL) << _PAGE_BIT_NX)
-- 
cgit v1.2.3


From dcbae6b377d78190954055ef2d8909ae83ff57de Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:55 +0100
Subject: x86/pgtable: unify pagetable accessors, #1

based on:

 Subject: x86/pgtable: unify pagetable accessors
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_64.h | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 93eb4df2ec11..3a0588088480 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -179,21 +179,27 @@ static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
  * Undefined behaviour if not..
  */
 #define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
+
 static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
 static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
 static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
 
-static inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY)); return pte; }
-static inline pte_t pte_mkold(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED)); return pte; }
-static inline pte_t pte_wrprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_RW)); return pte; }
-static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_NX)); return pte; }
-static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
-static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
-static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
-static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_PSE)); return pte; }
-static inline pte_t pte_clrhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_PSE)); return pte; }
+static inline int pmd_large(pmd_t pte) {
+	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
+		(_PAGE_PSE|_PAGE_PRESENT);
+}
+
+static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
+static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
+static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
+static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
+static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
+static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
+static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
+static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
+static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
 
 struct vm_area_struct;
 
-- 
cgit v1.2.3


From 92ac166f3a776afc51766395076ffcfde112740a Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:55 +0100
Subject: x86/pgtable: unify pagetable accessors, #2

change the pte_dirty/* inlines to the unified format. Non-NOP!

based on:

 Subject: x86/pgtable: unify pagetable accessors
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_32.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 3b40cb6dea18..57d4aab5594b 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -105,15 +105,15 @@ extern unsigned long pg0[];
  * The following only work if pte_present() is true.
  * Undefined behaviour if not..
  */
-static inline int pte_dirty(pte_t pte)		{ return (pte).pte_low & _PAGE_DIRTY; }
-static inline int pte_young(pte_t pte)		{ return (pte).pte_low & _PAGE_ACCESSED; }
-static inline int pte_write(pte_t pte)		{ return (pte).pte_low & _PAGE_RW; }
-static inline int pte_huge(pte_t pte)		{ return (pte).pte_low & _PAGE_PSE; }
+static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
+static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
 
 /*
  * The following only works if pte_present() is not true.
  */
-static inline int pte_file(pte_t pte)		{ return (pte).pte_low & _PAGE_FILE; }
+static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 
 static inline pte_t pte_mkclean(pte_t pte)	{ (pte).pte_low &= ~_PAGE_DIRTY; return pte; }
 static inline pte_t pte_mkold(pte_t pte)	{ (pte).pte_low &= ~_PAGE_ACCESSED; return pte; }
-- 
cgit v1.2.3


From 1444d2da4468f30daff47cacfd98acbd7baf43a7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:56 +0100
Subject: x86/pgtable: unify pagetable accessors, #3

change the pte_mk inlines to the unified format. Non-NOP!

based on:

 Subject: x86/pgtable: unify pagetable accessors
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_32.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 57d4aab5594b..c1e855fdf60e 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -115,13 +115,13 @@ static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
  */
 static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 
-static inline pte_t pte_mkclean(pte_t pte)	{ (pte).pte_low &= ~_PAGE_DIRTY; return pte; }
-static inline pte_t pte_mkold(pte_t pte)	{ (pte).pte_low &= ~_PAGE_ACCESSED; return pte; }
-static inline pte_t pte_wrprotect(pte_t pte)	{ (pte).pte_low &= ~_PAGE_RW; return pte; }
-static inline pte_t pte_mkdirty(pte_t pte)	{ (pte).pte_low |= _PAGE_DIRTY; return pte; }
-static inline pte_t pte_mkyoung(pte_t pte)	{ (pte).pte_low |= _PAGE_ACCESSED; return pte; }
-static inline pte_t pte_mkwrite(pte_t pte)	{ (pte).pte_low |= _PAGE_RW; return pte; }
-static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return pte; }
+static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
+static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
+static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
+static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
+static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
+static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
+static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
 
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
-- 
cgit v1.2.3


From 7d00a1ae54779db1bec03fa575cf53d99cb6cc48 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:56 +0100
Subject: x86/pgtable: unify pagetable accessors, #4

add new ops to 32-bit.

based on:

 Subject: x86/pgtable: unify pagetable accessors
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_32.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index c1e855fdf60e..b3e92124fc3f 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -123,6 +123,14 @@ static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_A
 static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
 static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
 
+static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
+static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
+
+static inline int pmd_large(pmd_t pte) {
+	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
+		(_PAGE_PSE|_PAGE_PRESENT);
+}
+
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
 #else
-- 
cgit v1.2.3


From e8a4852453c5af27d2a9c606e76809afa3dedb64 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:56 +0100
Subject: x86/pgtable: unify pagetable accessors, #5

reorder. NOP.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_32.h | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index b3e92124fc3f..1c43c302609c 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -108,28 +108,23 @@ extern unsigned long pg0[];
 static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
 static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
+static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
 
-/*
- * The following only works if pte_present() is not true.
- */
-static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
+static inline int pmd_large(pmd_t pte) {
+	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
+		(_PAGE_PSE|_PAGE_PRESENT);
+}
 
 static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
 static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
 static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
+static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
 static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
 static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
 static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
 static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
-
 static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
-static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
-
-static inline int pmd_large(pmd_t pte) {
-	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-		(_PAGE_PSE|_PAGE_PRESENT);
-}
 
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
-- 
cgit v1.2.3


From 4614139c6a74fe02c85f702ba9c0e57f8e38647e Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:56 +0100
Subject: x86/pgtable: unify pagetable accessors, #6

Unify functions to test and set bits in pagetable entries.

NOP: only moves existing code around, without any change to it.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 28 ++++++++++++++++++++++++++++
 include/asm-x86/pgtable_32.h | 28 ----------------------------
 include/asm-x86/pgtable_64.h | 32 --------------------------------
 3 files changed, 28 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 73e2b47c055e..4c03ff2246fd 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -115,6 +115,34 @@ extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define __S110	PAGE_SHARED_EXEC
 #define __S111	PAGE_SHARED_EXEC
 
+#ifndef __ASSEMBLY__
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
+static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
+static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
+
+static inline int pmd_large(pmd_t pte) {
+	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
+		(_PAGE_PSE|_PAGE_PRESENT);
+}
+
+static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
+static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
+static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
+static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
+static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
+static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
+static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
+static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
+static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
+
+#endif	/* __ASSEMBLY__ */
+
 #ifdef CONFIG_X86_32
 # include "pgtable_32.h"
 #else
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 1c43c302609c..0aaefdda5158 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -101,31 +101,6 @@ extern unsigned long pg0[];
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
 
-/*
- * The following only work if pte_present() is true.
- * Undefined behaviour if not..
- */
-static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
-static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
-static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
-static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
-static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
-
-static inline int pmd_large(pmd_t pte) {
-	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-		(_PAGE_PSE|_PAGE_PRESENT);
-}
-
-static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
-static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
-static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
-static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
-static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
-static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
-static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
-static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
-static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
-
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
 #else
@@ -276,9 +251,6 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return pte;
 }
 
-#define pmd_large(pmd) \
-((pmd_val(pmd) & (_PAGE_PSE|_PAGE_PRESENT)) == (_PAGE_PSE|_PAGE_PRESENT))
-
 /*
  * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
  *
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 3a0588088480..84f47f9d7896 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -173,34 +173,6 @@ static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
 	pte_val(pte) &= __supported_pte_mask;
 	return pte;
 }
-
-/*
- * The following only work if pte_present() is true.
- * Undefined behaviour if not..
- */
-#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
-
-static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
-static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
-static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
-static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
-static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
-
-static inline int pmd_large(pmd_t pte) {
-	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-		(_PAGE_PSE|_PAGE_PRESENT);
-}
-
-static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
-static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
-static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
-static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
-static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
-static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
-static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
-static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
-static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
-
 struct vm_area_struct;
 
 static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
@@ -220,10 +192,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
  */
 #define pgprot_noncached(prot)	(__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
 
-static inline int pmd_large(pmd_t pte) { 
-	return (pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE; 
-} 	
-
 
 /*
  * Conversion functions: convert a page and protection to a page entry,
-- 
cgit v1.2.3


From c3bcfb57e1e64b9b2f8b2d90564826637e21c5ea Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:57 +0100
Subject: x86: mask NX from pte_pfn

In 32-bit PAE, mask NX from pte_pfn, since it isn't part of the PFN.
This code is due for unification anyway, but this fixes a latent bug.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable-3level.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 948a33414118..9816346f7df6 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -155,7 +155,7 @@ static inline int pte_none(pte_t pte)
 
 static inline unsigned long pte_pfn(pte_t pte)
 {
-	return pte_val(pte) >> PAGE_SHIFT;
+	return (pte_val(pte) & ~_PAGE_NX) >> PAGE_SHIFT;
 }
 
 extern unsigned long long __supported_pte_mask;
-- 
cgit v1.2.3


From 6fdc05d4794056e0b98901646c4b68c60a01d5eb Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:57 +0100
Subject: x86: unify pgtable accessors which use

Make users of supported_pte_mask common.  This has the side-effect of
introducing the variable for 32-bit non-PAE, but I think its a pretty
small cost to simplify the code.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_32.c            |  5 +++--
 include/asm-x86/pgtable-2level.h |  2 --
 include/asm-x86/pgtable-3level.h | 14 --------------
 include/asm-x86/pgtable.h        | 14 ++++++++++++++
 include/asm-x86/pgtable_64.h     | 18 ------------------
 5 files changed, 17 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 29130970c193..04332c09ad1d 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -478,11 +478,12 @@ void zap_low_mappings (void)
 
 int nx_enabled = 0;
 
+pteval_t __supported_pte_mask __read_mostly = ~_PAGE_NX;
+EXPORT_SYMBOL_GPL(__supported_pte_mask);
+
 #ifdef CONFIG_X86_PAE
 
 static int disable_nx __initdata = 0;
-u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
-EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
 /*
  * noexec = on|off
diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index 84b03cf56a79..d3171c0cb157 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -53,8 +53,6 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 #define pte_page(x)		pfn_to_page(pte_pfn(x))
 #define pte_none(x)		(!(x).pte_low)
 #define pte_pfn(x)		(pte_val(x) >> PAGE_SHIFT)
-#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
-#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
 
 /*
  * All present pages are kernel-executable:
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 9816346f7df6..61a7d0029f60 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -158,20 +158,6 @@ static inline unsigned long pte_pfn(pte_t pte)
 	return (pte_val(pte) & ~_PAGE_NX) >> PAGE_SHIFT;
 }
 
-extern unsigned long long __supported_pte_mask;
-
-static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
-{
-	return __pte((((unsigned long long)page_nr << PAGE_SHIFT) |
-		      pgprot_val(pgprot)) & __supported_pte_mask);
-}
-
-static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
-{
-	return __pmd((((unsigned long long)page_nr << PAGE_SHIFT) |
-		      pgprot_val(pgprot)) & __supported_pte_mask);
-}
-
 /*
  * Bits 0, 6 and 7 are taken in the low part of the pte,
  * put the 32 bits of offset into the high part.
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 4c03ff2246fd..b29a122fc1ff 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -141,6 +141,20 @@ static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_R
 static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
 static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
 
+extern pteval_t __supported_pte_mask;
+
+static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
+{
+	return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) |
+		      pgprot_val(pgprot)) & __supported_pte_mask);
+}
+
+static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
+{
+	return __pmd((((phys_addr_t)page_nr << PAGE_SHIFT) |
+		      pgprot_val(pgprot)) & __supported_pte_mask);
+}
+
 #endif	/* __ASSEMBLY__ */
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 84f47f9d7896..57e4ebe0f7fc 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -17,7 +17,6 @@ extern pud_t level3_kernel_pgt[512];
 extern pud_t level3_ident_pgt[512];
 extern pmd_t level2_kernel_pgt[512];
 extern pgd_t init_level4_pgt[];
-extern unsigned long __supported_pte_mask;
 
 #define swapper_pg_dir init_level4_pgt
 
@@ -165,14 +164,6 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 #define pte_page(x)	pfn_to_page(pte_pfn(x))
 #define pte_pfn(x)  ((pte_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
-static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
-{
-	pte_t pte;
-	pte_val(pte) = (page_nr << PAGE_SHIFT);
-	pte_val(pte) |= pgprot_val(pgprot);
-	pte_val(pte) &= __supported_pte_mask;
-	return pte;
-}
 struct vm_area_struct;
 
 static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
@@ -239,15 +230,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 /* page, protection -> pte */
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
  
-/* Change flags of a PTE */
-static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
-{ 
-	pte_val(pte) &= _PAGE_CHG_MASK;
-	pte_val(pte) |= pgprot_val(newprot);
-	pte_val(pte) &= __supported_pte_mask;
-       return pte; 
-}
-
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
-- 
cgit v1.2.3


From 38472311679d8784c2934d071ddda23d9b3fe264 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:57 +0100
Subject: x86: unify pgtable accessors which use, #2

based on:

 Subject: x86: unify pgtable accessors which use supported_pte_mask
 From: Jeremy Fitzhardinge <jeremy@goop.org>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 15 +++++++++++++++
 include/asm-x86/pgtable_32.h | 16 ----------------
 2 files changed, 15 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index b29a122fc1ff..6452286e71a1 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -155,6 +155,21 @@ static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
 		      pgprot_val(pgprot)) & __supported_pte_mask);
 }
 
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	pteval_t val = pte_val(pte);
+
+	/*
+	 * Chop off the NX bit (if present), and add the NX portion of
+	 * the newprot (if present):
+	 */
+	val &= _PAGE_CHG_MASK & ~_PAGE_NX;
+	val |= pgprot_val(newprot) & __supported_pte_mask;
+
+	return __pte(val);
+}
+
+
 #endif	/* __ASSEMBLY__ */
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 0aaefdda5158..109dad5e16ee 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -235,22 +235,6 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
 
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
 
-static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
-{
-	pte.pte_low &= _PAGE_CHG_MASK;
-	pte.pte_low |= pgprot_val(newprot);
-#ifdef CONFIG_X86_PAE
-	/*
-	 * Chop off the NX bit (if present), and add the NX portion of
-	 * the newprot (if present):
-	 */
-	pte.pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
-	pte.pte_high |= (pgprot_val(newprot) >> 32) & \
-					(__supported_pte_mask >> 32);
-#endif
-	return pte;
-}
-
 /*
  * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
  *
-- 
cgit v1.2.3


From b7fff536d0ad45c4810f9b99845c707ceadc3afc Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:32:57 +0100
Subject: x86: fix pte_modify() bug

fix sign extension bug in PTE_MASK / _PTE_CHG_MASK.

this resolves the following bootup crash on PAE systems:

[   94.710726] init[1]: segfault at 00000004 ip 49471cbb sp bff0c6c0 error 4
[   94.717764] init[1]: segfault at 00000004 ip 49471cbb sp bff0c6c0 error 4
[   94.724772] init[1]: segfault at 00000004 ip 49471cbb sp bff0c6c0 error 4
[   94.731777] init[1]: segfault at 00000004 ip 49471cbb sp bff0c6c0 error 4

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 3a00d0d9dfd3..91920565a575 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -11,7 +11,7 @@
 #ifdef __KERNEL__
 
 #define PHYSICAL_PAGE_MASK	(PAGE_MASK & __PHYSICAL_MASK)
-#define PTE_MASK		PHYSICAL_PAGE_MASK
+#define PTE_MASK		(_AT(long, PHYSICAL_PAGE_MASK))
 
 #define LARGE_PAGE_SIZE		(_AC(1,UL) << PMD_SHIFT)
 #define LARGE_PAGE_MASK		(~(LARGE_PAGE_SIZE-1))
-- 
cgit v1.2.3


From c8e5393ab38564d2f45b560a2f95bc8f9ff6f823 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:57 +0100
Subject: x86: page.h: make pte_t a union to always include

Make sure pte_t, whatever its definition, has a pte element with type
pteval_t.  This allows common code to access it without needing to be
specifically parameterised on what pagetable mode we're compiling for.
For 32-bit, this means that pte_t becomes a union with "pte" and "{
pte_low, pte_high }" (PAE) or just "pte_low" (non-PAE).

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/xen/mmu.c               |  4 +---
 include/asm-x86/page.h           | 10 ++++++++++
 include/asm-x86/page_32.h        | 30 +++++++-----------------------
 include/asm-x86/page_64.h        |  3 ---
 include/asm-x86/paravirt.h       |  2 +-
 include/asm-x86/pgtable-2level.h |  4 ++--
 include/asm-x86/pgtable-3level.h |  4 ++--
 include/asm-x86/pgtable_64.h     |  6 +++---
 include/asm-x86/processor.h      |  1 -
 include/xen/page.h               |  6 +++---
 10 files changed, 29 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 0ac6c5dc49ba..52f392893008 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -244,9 +244,7 @@ pte_t xen_make_pte(unsigned long long pte)
 	if (pte & 1)
 		pte = phys_to_machine(XPADDR(pte)).maddr;
 
-	pte &= ~_PAGE_PCD;
-
-	return (pte_t){ pte, pte >> 32 };
+	return (pte_t){ .pte = pte };
 }
 
 pmd_t xen_make_pmd(unsigned long long pmd)
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 91920565a575..a6495eb5c605 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -108,6 +108,16 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 #include <asm-generic/pgtable-nopmd.h>
 #endif	/* PAGETABLE_LEVELS >= 3 */
 
+static inline pte_t native_make_pte(pteval_t val)
+{
+	return (pte_t) { .pte = val };
+}
+
+static inline pteval_t native_pte_val(pte_t pte)
+{
+	return pte.pte;
+}
+
 #define pgprot_val(x)	((x).pgprot)
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index e424bef4395a..11c4b39cada1 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -26,18 +26,12 @@ typedef u64	pgdval_t;
 typedef u64	pgprotval_t;
 typedef u64	phys_addr_t;
 
-typedef struct { unsigned long pte_low, pte_high; } pte_t;
-
-static inline unsigned long long native_pte_val(pte_t pte)
-{
-	return pte.pte_low | ((unsigned long long)pte.pte_high << 32);
-}
-
-static inline pte_t native_make_pte(unsigned long long val)
-{
-	return (pte_t) { .pte_low = val, .pte_high = (val >> 32) } ;
-}
-
+typedef union {
+	struct {
+		unsigned long pte_low, pte_high;
+	};
+	pteval_t pte;
+} pte_t;
 #endif	/* __ASSEMBLY__
  */
 #else  /* !CONFIG_X86_PAE */
@@ -53,19 +47,9 @@ typedef unsigned long	pgdval_t;
 typedef unsigned long	pgprotval_t;
 typedef unsigned long	phys_addr_t;
 
-typedef struct { pteval_t pte_low; } pte_t;
+typedef union { pteval_t pte, pte_low; } pte_t;
 typedef pte_t boot_pte_t;
 
-static inline unsigned long native_pte_val(pte_t pte)
-{
-	return pte.pte_low;
-}
-
-static inline pte_t native_make_pte(unsigned long val)
-{
-	return (pte_t) { .pte_low = val };
-}
-
 #endif	/* __ASSEMBLY__ */
 #endif	/* CONFIG_X86_PAE */
 
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index ebf977a809a8..c1ac42d8707f 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -70,9 +70,6 @@ typedef unsigned long	phys_addr_t;
 
 typedef struct { pteval_t pte; } pte_t;
 
-#define native_pte_val(x)	((x).pte)
-#define native_make_pte(x) ((pte_t) { (x) } )
-
 #define vmemmap ((struct page *)VMEMMAP_START)
 
 #endif	/* !__ASSEMBLY__ */
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index b35ed95166e4..24385decbd47 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -922,7 +922,7 @@ static inline pte_t __pte(unsigned long long val)
 	unsigned long long ret = PVOP_CALL2(unsigned long long,
 					    pv_mmu_ops.make_pte,
 					    val, val >> 32);
-	return (pte_t) { ret, ret >> 32 };
+	return (pte_t) { .pte = ret };
 }
 
 static inline pmd_t __pmd(unsigned long long val)
diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index d3171c0cb157..f949bb083089 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -72,13 +72,13 @@ static inline int pte_exec_kernel(pte_t pte)
 	((((pte).pte_low >> 1) & 0x1f ) + (((pte).pte_low >> 8) << 5 ))
 
 #define pgoff_to_pte(off) \
-	((pte_t) { (((off) & 0x1f) << 1) + (((off) >> 5) << 8) + _PAGE_FILE })
+	((pte_t) { .pte_low = (((off) & 0x1f) << 1) + (((off) >> 5) << 8) + _PAGE_FILE })
 
 /* Encode and de-code a swap entry */
 #define __swp_type(x)			(((x).val >> 1) & 0x1f)
 #define __swp_offset(x)			((x).val >> 8)
 #define __swp_entry(type, offset)	((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
 #define __pte_to_swp_entry(pte)		((swp_entry_t) { (pte).pte_low })
-#define __swp_entry_to_pte(x)		((pte_t) { (x).val })
+#define __swp_entry_to_pte(x)		((pte_t) { .pte = (x).val })
 
 #endif /* _I386_PGTABLE_2LEVEL_H */
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 61a7d0029f60..3da96f792dd0 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -163,7 +163,7 @@ static inline unsigned long pte_pfn(pte_t pte)
  * put the 32 bits of offset into the high part.
  */
 #define pte_to_pgoff(pte) ((pte).pte_high)
-#define pgoff_to_pte(off) ((pte_t) { _PAGE_FILE, (off) })
+#define pgoff_to_pte(off) ((pte_t) { { .pte_low = _PAGE_FILE, .pte_high = (off) } })
 #define PTE_FILE_MAX_BITS       32
 
 /* Encode and de-code a swap entry */
@@ -171,7 +171,7 @@ static inline unsigned long pte_pfn(pte_t pte)
 #define __swp_offset(x)			((x).val >> 5)
 #define __swp_entry(type, offset)	((swp_entry_t){(type) | (offset) << 5})
 #define __pte_to_swp_entry(pte)		((swp_entry_t){ (pte).pte_high })
-#define __swp_entry_to_pte(x)		((pte_t){ 0, (x).val })
+#define __swp_entry_to_pte(x)		((pte_t){ { .pte_high = (x).val } })
 
 #define __pmd_free_tlb(tlb, x)		do { } while (0)
 
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 57e4ebe0f7fc..77038d8e9bfd 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -72,7 +72,7 @@ extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
 
 static inline void set_pte(pte_t *dst, pte_t val)
 {
-	pte_val(*dst) = pte_val(val);
+	*dst = val;
 } 
 #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
 
@@ -222,7 +222,7 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 #define pmd_pfn(x)  ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
 #define pte_to_pgoff(pte) ((pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
-#define pgoff_to_pte(off) ((pte_t) { ((off) << PAGE_SHIFT) | _PAGE_FILE })
+#define pgoff_to_pte(off) ((pte_t) { .pte = ((off) << PAGE_SHIFT) | _PAGE_FILE })
 #define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
 
 /* PTE - Level 1 access. */
@@ -264,7 +264,7 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 #define __swp_offset(x)			((x).val >> 8)
 #define __swp_entry(type, offset)	((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
 #define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) })
-#define __swp_entry_to_pte(x)		((pte_t) { (x).val })
+#define __swp_entry_to_pte(x)		((pte_t) { .pte = (x).val })
 
 extern spinlock_t pgd_lock;
 extern struct list_head pgd_list;
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index cc549bf819a4..e701ac5487e5 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -14,7 +14,6 @@ struct mm_struct;
 #include <asm/vm86.h>
 #include <asm/math_emu.h>
 #include <asm/segment.h>
-#include <asm/page.h>
 #include <asm/types.h>
 #include <asm/sigcontext.h>
 #include <asm/current.h>
diff --git a/include/xen/page.h b/include/xen/page.h
index c0c8fcb27899..031ef22a971e 100644
--- a/include/xen/page.h
+++ b/include/xen/page.h
@@ -156,16 +156,16 @@ static inline pte_t mfn_pte(unsigned long page_nr, pgprot_t pgprot)
 
 static inline unsigned long long pte_val_ma(pte_t x)
 {
-	return ((unsigned long long)x.pte_high << 32) | x.pte_low;
+	return x.pte;
 }
 #define pmd_val_ma(v) ((v).pmd)
 #define pud_val_ma(v) ((v).pgd.pgd)
-#define __pte_ma(x)	((pte_t) { .pte_low = (x), .pte_high = (x)>>32 } )
+#define __pte_ma(x)	((pte_t) { .pte = (x) })
 #define __pmd_ma(x)	((pmd_t) { (x) } )
 #else  /* !X86_PAE */
 #define pte_mfn(_pte) ((_pte).pte_low >> PAGE_SHIFT)
 #define mfn_pte(pfn, prot)	__pte_ma(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
-#define pte_val_ma(x)	((x).pte_low)
+#define pte_val_ma(x)	((x).pte)
 #define pmd_val_ma(v)	((v).pud.pgd.pgd)
 #define __pte_ma(x)	((pte_t) { (x) } )
 #endif	/* CONFIG_X86_PAE */
-- 
cgit v1.2.3


From 195466dc4b9b8a4cc89d37ea1211746f3afbc941 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:58 +0100
Subject: x86: pgtable: unify pte accessors

Make various pte accessors common.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable-2level.h |   1 -
 include/asm-x86/pgtable-3level.h |   1 -
 include/asm-x86/pgtable.h        | 110 ++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/pgtable_32.h     | 101 -----------------------------------
 include/asm-x86/pgtable_64.h     |  43 ++-------------
 5 files changed, 113 insertions(+), 143 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index f949bb083089..65cdc8f9e3b3 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -33,7 +33,6 @@ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
 #define set_pte_atomic(pteptr, pteval) set_pte(pteptr,pteval)
 #define set_pte_present(mm,addr,ptep,pteval) set_pte_at(mm,addr,ptep,pteval)
 
-#define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 #define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 
 static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *xp)
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 3da96f792dd0..a123e687112f 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -101,7 +101,6 @@ static inline void native_pmd_clear(pmd_t *pmd)
 #define set_pte_atomic(ptep, pte)		native_set_pte_atomic(ptep, pte)
 #define set_pmd(pmdp, pmd)			native_set_pmd(pmdp, pmd)
 #define set_pud(pudp, pud)			native_set_pud(pudp, pud)
-#define pte_clear(mm, addr, ptep)		native_pte_clear(mm, addr, ptep)
 #define pmd_clear(pmd)				native_pmd_clear(pmd)
 #endif
 
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 6452286e71a1..62fa856c2491 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -116,6 +116,7 @@ extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define __S111	PAGE_SHARED_EXEC
 
 #ifndef __ASSEMBLY__
+
 /*
  * The following only work if pte_present() is true.
  * Undefined behaviour if not..
@@ -169,7 +170,6 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return __pte(val);
 }
 
-
 #endif	/* __ASSEMBLY__ */
 
 #ifdef CONFIG_X86_32
@@ -178,4 +178,112 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 # include "pgtable_64.h"
 #endif
 
+#ifndef __ASSEMBLY__
+
+#ifndef CONFIG_PARAVIRT
+/*
+ * Rules for using pte_update - it must be called after any PTE update which
+ * has not been done using the set_pte / clear_pte interfaces.  It is used by
+ * shadow mode hypervisors to resynchronize the shadow page tables.  Kernel PTE
+ * updates should either be sets, clears, or set_pte_atomic for P->P
+ * transitions, which means this hook should only be called for user PTEs.
+ * This hook implies a P->P protection or access change has taken place, which
+ * requires a subsequent TLB flush.  The notification can optionally be delayed
+ * until the TLB flush event by using the pte_update_defer form of the
+ * interface, but care must be taken to assure that the flush happens while
+ * still holding the same page table lock so that the shadow and primary pages
+ * do not become out of sync on SMP.
+ */
+#define pte_update(mm, addr, ptep)		do { } while (0)
+#define pte_update_defer(mm, addr, ptep)	do { } while (0)
+#endif
+
+/* local pte updates need not use xchg for locking */
+static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
+{
+	pte_t res = *ptep;
+
+	/* Pure native function needs no input for mm, addr */
+	native_pte_clear(NULL, 0, ptep);
+	return res;
+}
+
+/*
+ * We only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time.
+ */
+#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(vma, address, ptep, entry, dirty)		\
+({									\
+	int __changed = !pte_same(*(ptep), entry);			\
+	if (__changed && dirty) {					\
+		*ptep = entry;						\
+		pte_update_defer((vma)->vm_mm, (address), (ptep));	\
+		flush_tlb_page(vma, address);				\
+	}								\
+	__changed;							\
+})
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+#define ptep_test_and_clear_young(vma, addr, ptep) ({			\
+	int __ret = 0;							\
+	if (pte_young(*(ptep)))						\
+		__ret = test_and_clear_bit(_PAGE_BIT_ACCESSED,		\
+					   &(ptep)->pte);		\
+	if (__ret)							\
+		pte_update((vma)->vm_mm, addr, ptep);			\
+	__ret;								\
+})
+
+#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
+#define ptep_clear_flush_young(vma, address, ptep)			\
+({									\
+	int __young;							\
+	__young = ptep_test_and_clear_young((vma), (address), (ptep));	\
+	if (__young)							\
+		flush_tlb_page(vma, address);				\
+	__young;							\
+})
+
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+	pte_t pte = native_ptep_get_and_clear(ptep);
+	pte_update(mm, addr, ptep);
+	return pte;
+}
+
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
+static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+{
+	pte_t pte;
+	if (full) {
+		/*
+		 * Full address destruction in progress; paravirt does not
+		 * care about updates and native needs no locking
+		 */
+		pte = native_local_ptep_get_and_clear(ptep);
+	} else {
+		pte = ptep_get_and_clear(mm, addr, ptep);
+	}
+	return pte;
+}
+
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
+static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+	clear_bit(_PAGE_BIT_RW, &ptep->pte);
+	pte_update(mm, addr, ptep);
+}
+
+#ifndef CONFIG_PARAVIRT
+#define pte_clear(mm, addr, ptep)		native_pte_clear(mm, addr, ptep)
+#endif	/* !CONFIG_PARAVIRT */
+
+#include <asm-generic/pgtable.h>
+#endif	/* __ASSEMBLY__ */
+
 #endif	/* _ASM_X86_PGTABLE_H */
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 109dad5e16ee..d4d238c10293 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -107,105 +107,6 @@ extern unsigned long pg0[];
 # include <asm/pgtable-2level.h>
 #endif
 
-#ifndef CONFIG_PARAVIRT
-/*
- * Rules for using pte_update - it must be called after any PTE update which
- * has not been done using the set_pte / clear_pte interfaces.  It is used by
- * shadow mode hypervisors to resynchronize the shadow page tables.  Kernel PTE
- * updates should either be sets, clears, or set_pte_atomic for P->P
- * transitions, which means this hook should only be called for user PTEs.
- * This hook implies a P->P protection or access change has taken place, which
- * requires a subsequent TLB flush.  The notification can optionally be delayed
- * until the TLB flush event by using the pte_update_defer form of the
- * interface, but care must be taken to assure that the flush happens while
- * still holding the same page table lock so that the shadow and primary pages
- * do not become out of sync on SMP.
- */
-#define pte_update(mm, addr, ptep)		do { } while (0)
-#define pte_update_defer(mm, addr, ptep)	do { } while (0)
-#endif
-
-/* local pte updates need not use xchg for locking */
-static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
-{
-	pte_t res = *ptep;
-
-	/* Pure native function needs no input for mm, addr */
-	native_pte_clear(NULL, 0, ptep);
-	return res;
-}
-
-/*
- * We only update the dirty/accessed state if we set
- * the dirty bit by hand in the kernel, since the hardware
- * will do the accessed bit for us, and we don't want to
- * race with other CPU's that might be updating the dirty
- * bit at the same time.
- */
-#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-#define ptep_set_access_flags(vma, address, ptep, entry, dirty)		\
-({									\
-	int __changed = !pte_same(*(ptep), entry);			\
-	if (__changed && dirty) {					\
-		(ptep)->pte_low = (entry).pte_low;			\
-		pte_update_defer((vma)->vm_mm, (address), (ptep));	\
-		flush_tlb_page(vma, address);				\
-	}								\
-	__changed;							\
-})
-
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define ptep_test_and_clear_young(vma, addr, ptep) ({			\
-	int __ret = 0;							\
-	if (pte_young(*(ptep)))						\
-		__ret = test_and_clear_bit(_PAGE_BIT_ACCESSED,		\
-						&(ptep)->pte_low);	\
-	if (__ret)							\
-		pte_update((vma)->vm_mm, addr, ptep);			\
-	__ret;								\
-})
-
-#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
-#define ptep_clear_flush_young(vma, address, ptep)			\
-({									\
-	int __young;							\
-	__young = ptep_test_and_clear_young((vma), (address), (ptep));	\
-	if (__young)							\
-		flush_tlb_page(vma, address);				\
-	__young;							\
-})
-
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	pte_t pte = native_ptep_get_and_clear(ptep);
-	pte_update(mm, addr, ptep);
-	return pte;
-}
-
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
-{
-	pte_t pte;
-	if (full) {
-		/*
-		 * Full address destruction in progress; paravirt does not
-		 * care about updates and native needs no locking
-		 */
-		pte = native_local_ptep_get_and_clear(ptep);
-	} else {
-		pte = ptep_get_and_clear(mm, addr, ptep);
-	}
-	return pte;
-}
-
-#define __HAVE_ARCH_PTEP_SET_WRPROTECT
-static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
-	pte_update(mm, addr, ptep);
-}
-
 /*
  * clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
  *
@@ -359,6 +260,4 @@ static inline void paravirt_pagetable_setup_done(pgd_t *base)
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
-#include <asm-generic/pgtable.h>
-
 #endif /* _I386_PGTABLE_H */
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 77038d8e9bfd..987f51f684a5 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -101,18 +101,18 @@ static inline void pgd_clear (pgd_t * pgd)
 	set_pgd(pgd, __pgd(0));
 }
 
-#define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte, 0))
+#define native_ptep_get_and_clear(xp) __pte(xchg(&(xp)->pte, 0))
 
 struct mm_struct;
 
-static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+static inline pte_t native_ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
 {
 	pte_t pte;
 	if (full) {
 		pte = *ptep;
 		*ptep = __pte(0);
 	} else {
-		pte = ptep_get_and_clear(mm, addr, ptep);
+		pte = native_ptep_get_and_clear(ptep);
 	}
 	return pte;
 }
@@ -158,26 +158,12 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 
 #define pte_none(x)	(!pte_val(x))
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
-#define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
+#define native_pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this right? */
 #define pte_page(x)	pfn_to_page(pte_pfn(x))
 #define pte_pfn(x)  ((pte_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
-struct vm_area_struct;
-
-static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
-	if (!pte_young(*ptep))
-		return 0;
-	return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte);
-}
-
-static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	clear_bit(_PAGE_BIT_RW, &ptep->pte);
-}
-
 /*
  * Macro to mark a page protection value as "uncacheable".
  */
@@ -243,22 +229,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 
 #define update_mmu_cache(vma,address,pte) do { } while (0)
 
-/* We only update the dirty/accessed state if we set
- * the dirty bit by hand in the kernel, since the hardware
- * will do the accessed bit for us, and we don't want to
- * race with other CPU's that might be updating the dirty
- * bit at the same time. */
-#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
-({									  \
-	int __changed = !pte_same(*(__ptep), __entry);			  \
-	if (__changed && __dirty) {					  \
-		set_pte(__ptep, __entry);			  	  \
-		flush_tlb_page(__vma, __address);		  	  \
-	}								  \
-	__changed;							  \
-})
-
 /* Encode and de-code a swap entry */
 #define __swp_type(x)			(((x).val >> 1) & 0x3f)
 #define __swp_offset(x)			((x).val >> 8)
@@ -290,12 +260,7 @@ pte_t *lookup_address(unsigned long addr);
 #define	kc_offset_to_vaddr(o) \
    (((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
 
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-#define __HAVE_ARCH_PTEP_SET_WRPROTECT
 #define __HAVE_ARCH_PTE_SAME
-#include <asm-generic/pgtable.h>
 #endif /* !__ASSEMBLY__ */
 
 #endif /* _X86_64_PGTABLE_H */
-- 
cgit v1.2.3


From d8d89827fc0c9c2ea6ac4c22e14e30586a856f58 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:58 +0100
Subject: x86: fix warning

&ptep->pte isn't always an unsigned long *, so cast it to avoid a warning.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 62fa856c2491..820db41dbe4c 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -275,7 +275,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define __HAVE_ARCH_PTEP_SET_WRPROTECT
 static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	clear_bit(_PAGE_BIT_RW, &ptep->pte);
+	clear_bit(_PAGE_BIT_RW, (unsigned long *)&ptep->pte);
 	pte_update(mm, addr, ptep);
 }
 
-- 
cgit v1.2.3


From 8405b122ad0dd75354b3bfed4de9a96514fd40cb Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:58 +0100
Subject: x86: unify zero_page definition

Move ZERO_PAGE/empty_zero_page to common place.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 8 ++++++++
 include/asm-x86/pgtable_32.h | 6 ------
 include/asm-x86/pgtable_64.h | 7 -------
 3 files changed, 8 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 820db41dbe4c..eb14a70a4728 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -117,6 +117,14 @@ extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 
 #ifndef __ASSEMBLY__
 
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used
+ * for zero-mapped memory areas etc..
+ */
+extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+
+
 /*
  * The following only work if pte_present() is true.
  * Undefined behaviour if not..
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index d4d238c10293..eb8cccfa6a49 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -25,12 +25,6 @@
 struct mm_struct;
 struct vm_area_struct;
 
-/*
- * ZERO_PAGE is a global shared page that is always zero: used
- * for zero-mapped memory areas etc..
- */
-#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
-extern unsigned long empty_zero_page[1024];
 extern pgd_t swapper_pg_dir[1024];
 extern struct kmem_cache *pmd_cache;
 extern spinlock_t pgd_lock;
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 987f51f684a5..435e17187eb6 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -23,13 +23,6 @@ extern pgd_t init_level4_pgt[];
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
-/*
- * ZERO_PAGE is a global shared page that is always zero: used
- * for zero-mapped memory areas etc..
- */
-extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
-#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
-
 #endif /* !__ASSEMBLY__ */
 
 /*
-- 
cgit v1.2.3


From 4891645e764d2e181b834509a689fcd12e890c10 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:58 +0100
Subject: x86: unify paravirt pagetable accessors

Put all the defines for mapping pagetable operations to their native
versions (for the non-paravirt case) into one place.  Make the
corresponding changes to paravirt.h.

The tricky part here is that when a pagetable entry can't be updated
atomically (ie, 32-bit PAE), we need special handlers for pte_clear,
set_pte_atomic and set_pte_present.  However, the other two modes
don't need special handling for these, and can use a common
set_pte(_at) path.

[ mingo@elte.hu: fixes ]

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h       | 21 +++++++++++++
 include/asm-x86/pgtable-2level.h | 30 ++++++++++---------
 include/asm-x86/pgtable-3level.h | 15 ----------
 include/asm-x86/pgtable.h        | 63 ++++++++++++++++++++++++++++++---------
 include/asm-x86/pgtable_64.h     | 64 ++++++++++++++++++++++------------------
 5 files changed, 121 insertions(+), 72 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 24385decbd47..4aa06b929e48 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1039,6 +1039,27 @@ static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	PVOP_VCALL2(pv_mmu_ops.set_pmd, pmdp, pmdval.pud.pgd.pgd);
 }
+
+static inline void pmd_clear(pmd_t *pmdp)
+{
+	set_pmd(pmdp, __pmd(0));
+}
+
+static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+	set_pte_at(mm, addr, ptep, __pte(0));
+}
+
+static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+	set_pte(ptep, pte);
+}
+
+static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
+				   pte_t *ptep, pte_t pte)
+{
+	set_pte(ptep, pte);
+}
 #endif	/* CONFIG_X86_PAE */
 
 /* Lazy mode for batching updates / context switch */
diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index 65cdc8f9e3b3..ce486bdbbbb7 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -15,29 +15,31 @@ static inline void native_set_pte(pte_t *ptep , pte_t pte)
 {
 	*ptep = pte;
 }
-static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
-				     pte_t *ptep , pte_t pte)
-{
-	native_set_pte(ptep, pte);
-}
+
 static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
 {
 	*pmdp = pmd;
 }
-#ifndef CONFIG_PARAVIRT
-#define set_pte(pteptr, pteval)		native_set_pte(pteptr, pteval)
-#define set_pte_at(mm,addr,ptep,pteval) native_set_pte_at(mm, addr, ptep, pteval)
-#define set_pmd(pmdptr, pmdval)		native_set_pmd(pmdptr, pmdval)
-#endif
 
-#define set_pte_atomic(pteptr, pteval) set_pte(pteptr,pteval)
-#define set_pte_present(mm,addr,ptep,pteval) set_pte_at(mm,addr,ptep,pteval)
+static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+	native_set_pte(ptep, pte);
+}
+
+static inline void native_set_pte_present(struct mm_struct *mm, unsigned long addr,
+					  pte_t *ptep, pte_t pte)
+{
+	native_set_pte(ptep, pte);
+}
 
-#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
+static inline void native_pmd_clear(pmd_t *pmdp)
+{
+	native_set_pmd(pmdp, __pmd(0));
+}
 
 static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *xp)
 {
-	*xp = __pte(0);
+	*xp = native_make_pte(0);
 }
 
 #ifdef CONFIG_SMP
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index a123e687112f..5af7a44ed902 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -39,11 +39,6 @@ static inline void native_set_pte(pte_t *ptep, pte_t pte)
 	smp_wmb();
 	ptep->pte_low = pte.pte_low;
 }
-static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
-				     pte_t *ptep , pte_t pte)
-{
-	native_set_pte(ptep, pte);
-}
 
 /*
  * Since this is only called on user PTEs, and the page fault handler
@@ -94,16 +89,6 @@ static inline void native_pmd_clear(pmd_t *pmd)
 	*(tmp + 1) = 0;
 }
 
-#ifndef CONFIG_PARAVIRT
-#define set_pte(ptep, pte)			native_set_pte(ptep, pte)
-#define set_pte_at(mm, addr, ptep, pte)		native_set_pte_at(mm, addr, ptep, pte)
-#define set_pte_present(mm, addr, ptep, pte)	native_set_pte_present(mm, addr, ptep, pte)
-#define set_pte_atomic(ptep, pte)		native_set_pte_atomic(ptep, pte)
-#define set_pmd(pmdp, pmd)			native_set_pmd(pmdp, pmd)
-#define set_pud(pudp, pud)			native_set_pud(pudp, pud)
-#define pmd_clear(pmd)				native_pmd_clear(pmd)
-#endif
-
 /*
  * Pentium-II erratum A13: in PAE mode we explicitly have to flush
  * the TLB via cr3 if the top-level pgd is changed...
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index eb14a70a4728..a1eeacdf9d8c 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -178,6 +178,39 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return __pte(val);
 }
 
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else  /* !CONFIG_PARAVIRT */
+#define set_pte(ptep, pte)		native_set_pte(ptep, pte)
+#define set_pte_at(mm, addr, ptep, pte)	native_set_pte_at(mm, addr, ptep, pte)
+
+#define set_pte_present(mm, addr, ptep, pte)				\
+	native_set_pte_present(mm, addr, ptep, pte)
+#define set_pte_atomic(ptep, pte)					\
+	native_set_pte_atomic(ptep, pte)
+
+#define set_pmd(pmdp, pmd)		native_set_pmd(pmdp, pmd)
+
+#ifndef __PAGETABLE_PUD_FOLDED
+#define set_pgd(pgdp, pgd)		native_set_pgd(pgdp, pgd)
+#define pgd_clear(pgd)			native_pgd_clear(pgd)
+#endif
+
+#ifndef set_pud
+# define set_pud(pudp, pud)		native_set_pud(pudp, pud)
+#endif
+
+#ifndef __PAGETABLE_PMD_FOLDED
+#define pud_clear(pud)			native_pud_clear(pud)
+#endif
+
+#define pte_clear(mm, addr, ptep)	native_pte_clear(mm, addr, ptep)
+#define pmd_clear(pmd)			native_pmd_clear(pmd)
+
+#define pte_update(mm, addr, ptep)              do { } while (0)
+#define pte_update_defer(mm, addr, ptep)        do { } while (0)
+#endif	/* CONFIG_PARAVIRT */
+
 #endif	/* __ASSEMBLY__ */
 
 #ifdef CONFIG_X86_32
@@ -188,6 +221,22 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #ifndef __ASSEMBLY__
 
+/* local pte updates need not use xchg for locking */
+static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
+{
+	pte_t res = *ptep;
+
+	/* Pure native function needs no input for mm, addr */
+	native_pte_clear(NULL, 0, ptep);
+	return res;
+}
+
+static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
+				     pte_t *ptep , pte_t pte)
+{
+	native_set_pte(ptep, pte);
+}
+
 #ifndef CONFIG_PARAVIRT
 /*
  * Rules for using pte_update - it must be called after any PTE update which
@@ -206,16 +255,6 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_update_defer(mm, addr, ptep)	do { } while (0)
 #endif
 
-/* local pte updates need not use xchg for locking */
-static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
-{
-	pte_t res = *ptep;
-
-	/* Pure native function needs no input for mm, addr */
-	native_pte_clear(NULL, 0, ptep);
-	return res;
-}
-
 /*
  * We only update the dirty/accessed state if we set
  * the dirty bit by hand in the kernel, since the hardware
@@ -287,10 +326,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 	pte_update(mm, addr, ptep);
 }
 
-#ifndef CONFIG_PARAVIRT
-#define pte_clear(mm, addr, ptep)		native_pte_clear(mm, addr, ptep)
-#endif	/* !CONFIG_PARAVIRT */
-
 #include <asm-generic/pgtable.h>
 #endif	/* __ASSEMBLY__ */
 
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 435e17187eb6..29fdeb8111bb 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -63,51 +63,59 @@ extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 #define pgd_none(x)	(!pgd_val(x))
 #define pud_none(x)	(!pud_val(x))
 
-static inline void set_pte(pte_t *dst, pte_t val)
+struct mm_struct;
+
+static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr,
+				    pte_t *ptep)
 {
-	*dst = val;
-} 
-#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+	*ptep = native_make_pte(0);
+}
 
-static inline void set_pmd(pmd_t *dst, pmd_t val)
+static inline void native_set_pte(pte_t *ptep, pte_t pte)
 {
-	*dst = val;
-} 
+	*ptep = pte;
+}
 
-static inline void set_pud(pud_t *dst, pud_t val)
+static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 {
-	*dst = val;
+#ifdef CONFIG_SMP
+	return native_make_pte(xchg(&xp->pte, 0));
+#else
+	/* native_local_ptep_get_and_clear, but duplicated because of cyclic dependency */
+	pte_t ret = *xp;
+	native_pte_clear(NULL, 0, xp);
+	return ret;
+#endif
 }
 
-static inline void pud_clear (pud_t *pud)
+static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
 {
-	set_pud(pud, __pud(0));
+	*pmdp = pmd;
 }
 
-static inline void set_pgd(pgd_t *dst, pgd_t val)
+static inline void native_pmd_clear(pmd_t *pmd)
 {
-	*dst = val;
-} 
+	native_set_pmd(pmd, native_make_pmd(0));
+}
 
-static inline void pgd_clear (pgd_t * pgd)
+static inline void native_set_pud(pud_t *pudp, pud_t pud)
 {
-	set_pgd(pgd, __pgd(0));
+	*pudp = pud;
 }
 
-#define native_ptep_get_and_clear(xp) __pte(xchg(&(xp)->pte, 0))
+static inline void native_pud_clear(pud_t *pud)
+{
+	native_set_pud(pud, native_make_pud(0));
+}
 
-struct mm_struct;
+static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd)
+{
+	*pgdp = pgd;
+}
 
-static inline pte_t native_ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+static inline void native_pgd_clear(pgd_t * pgd)
 {
-	pte_t pte;
-	if (full) {
-		pte = *ptep;
-		*ptep = __pte(0);
-	} else {
-		pte = native_ptep_get_and_clear(ptep);
-	}
-	return pte;
+	native_set_pgd(pgd, native_make_pgd(0));
 }
 
 #define pte_same(a, b)		((a).pte == (b).pte)
@@ -151,7 +159,6 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 
 #define pte_none(x)	(!pte_val(x))
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
-#define native_pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this right? */
 #define pte_page(x)	pfn_to_page(pte_pfn(x))
@@ -196,7 +203,6 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 			pmd_index(address))
 #define pmd_none(x)	(!pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
-#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 #define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
 #define pmd_pfn(x)  ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
-- 
cgit v1.2.3


From f29192320f22f1617d50a9c790862603eeef64e6 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:32:59 +0100
Subject: x86/pgtable: explain constant sign extension problem

When the _PAGE_FOO constants are defined as (1ul << _PAGE_BIT_FOO), they
become unsigned longs.  In 32-bit PAE mode, these end up being
implicitly cast to 64-bit types when used to manipulate a pte, and
because they're unsigned the top 32-bits are 0, destroying the upper
bits of the pte.

When _PAGE_FOO constants are given a signed integer type, the cast to
64-bits will sign-extend so that the upper bits are all ones,
preserving the upper pte bits in manipulations.

Explain this in a prominent place.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index a1eeacdf9d8c..75a656e6b3f8 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -19,6 +19,11 @@
 #define _PAGE_BIT_UNUSED3	11
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
+/*
+ * Note: we use _AC(1, L) instead of _AC(1, UL) so that we get a
+ * sign-extended value on 32-bit with all 1's in the upper word,
+ * which preserves the upper pte values on 64-bit ptes:
+ */
 #define _PAGE_PRESENT	(_AC(1, L)<<_PAGE_BIT_PRESENT)
 #define _PAGE_RW	(_AC(1, L)<<_PAGE_BIT_RW)
 #define _PAGE_USER	(_AC(1, L)<<_PAGE_BIT_USER)
-- 
cgit v1.2.3


From 3a556b26a2718e48aa2b6ce06ea4875ddcd0778e Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 30 Jan 2008 13:33:00 +0100
Subject: x86: big ticket locks

This implements ticket lock support for more than 255 CPUs on x86. The
code gets switched according to the configured NR_CPUS.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/spinlock.h | 124 +++++++++++++++++++++++++++++++++++----------
 1 file changed, 97 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index 2076d5d62d83..23804c1890ff 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -35,10 +35,35 @@ typedef int _slock_t;
 # define LOCK_PTR_REG "D"
 #endif
 
-#if (NR_CPUS > 256)
-#error spinlock supports a maximum of 256 CPUs
+#if defined(CONFIG_X86_32) && \
+	(defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE))
+/*
+ * On PPro SMP or if we are using OOSTORE, we use a locked operation to unlock
+ * (PPro errata 66, 92)
+ */
+# define UNLOCK_LOCK_PREFIX LOCK_PREFIX
+#else
+# define UNLOCK_LOCK_PREFIX
 #endif
 
+/*
+ * Ticket locks are conceptually two parts, one indicating the current head of
+ * the queue, and the other indicating the current tail. The lock is acquired
+ * by atomically noting the tail and incrementing it by one (thus adding
+ * ourself to the queue and noting our position), then waiting until the head
+ * becomes equal to the the initial value of the tail.
+ *
+ * We use an xadd covering *both* parts of the lock, to increment the tail and
+ * also load the position of the head, which takes care of memory ordering
+ * issues and should be optimal for the uncontended case. Note the tail must be
+ * in the high part, because a wide xadd increment of the low part would carry
+ * up and contaminate the high part.
+ *
+ * With fewer than 2^8 possible CPUs, we can use x86's partial registers to
+ * save some instructions and make the code more elegant. There really isn't
+ * much between them in performance though, especially as locks are out of line.
+ */
+#if (NR_CPUS < 256)
 static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
 {
 	int tmp = *(volatile signed int *)(&(lock)->slock);
@@ -57,21 +82,6 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	short inc = 0x0100;
 
-	/*
-	 * Ticket locks are conceptually two bytes, one indicating the current
-	 * head of the queue, and the other indicating the current tail. The
-	 * lock is acquired by atomically noting the tail and incrementing it
-	 * by one (thus adding ourself to the queue and noting our position),
-	 * then waiting until the head becomes equal to the the initial value
-	 * of the tail.
-	 *
-	 * This uses a 16-bit xadd to increment the tail and also load the
-	 * position of the head, which takes care of memory ordering issues
-	 * and should be optimal for the uncontended case. Note the tail must
-	 * be in the high byte, otherwise the 16-bit wide increment of the low
-	 * byte would carry up and contaminate the high byte.
-	 */
-
 	__asm__ __volatile__ (
 		LOCK_PREFIX "xaddw %w0, %1\n"
 		"1:\t"
@@ -111,25 +121,85 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 	return tmp;
 }
 
-#if defined(CONFIG_X86_32) && \
-	(defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE))
-/*
- * On PPro SMP or if we are using OOSTORE, we use a locked operation to unlock
- * (PPro errata 66, 92)
- */
-# define UNLOCK_LOCK_PREFIX LOCK_PREFIX
+static inline void __raw_spin_unlock(raw_spinlock_t *lock)
+{
+	__asm__ __volatile__(
+		UNLOCK_LOCK_PREFIX "incb %0"
+		:"+m" (lock->slock)
+		:
+		:"memory", "cc");
+}
 #else
-# define UNLOCK_LOCK_PREFIX
-#endif
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+{
+	int tmp = *(volatile signed int *)(&(lock)->slock);
+
+	return (((tmp >> 16) & 0xffff) != (tmp & 0xffff));
+}
+
+static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
+{
+	int tmp = *(volatile signed int *)(&(lock)->slock);
+
+	return (((tmp >> 16) & 0xffff) - (tmp & 0xffff)) > 1;
+}
+
+static inline void __raw_spin_lock(raw_spinlock_t *lock)
+{
+	int inc = 0x00010000;
+	int tmp;
+
+	__asm__ __volatile__ (
+		"lock ; xaddl %0, %1\n"
+		"movzwl %w0, %2\n\t"
+		"shrl $16, %0\n\t"
+		"1:\t"
+		"cmpl %0, %2\n\t"
+		"je 2f\n\t"
+		"rep ; nop\n\t"
+		"movzwl %1, %2\n\t"
+		/* don't need lfence here, because loads are in-order */
+		"jmp 1b\n"
+		"2:"
+		:"+Q" (inc), "+m" (lock->slock), "=r" (tmp)
+		:
+		:"memory", "cc");
+}
+
+#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
+
+static inline int __raw_spin_trylock(raw_spinlock_t *lock)
+{
+	int tmp;
+	int new;
+
+	asm volatile(
+		"movl %2,%0\n\t"
+		"movl %0,%1\n\t"
+		"roll $16, %0\n\t"
+		"cmpl %0,%1\n\t"
+		"jne 1f\n\t"
+		"addl $0x00010000, %1\n\t"
+		"lock ; cmpxchgl %1,%2\n\t"
+		"1:"
+		"sete %b1\n\t"
+		"movzbl %b1,%0\n\t"
+		:"=&a" (tmp), "=r" (new), "+m" (lock->slock)
+		:
+		: "memory", "cc");
+
+	return tmp;
+}
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	__asm__ __volatile__(
-		UNLOCK_LOCK_PREFIX "incb %0"
+		UNLOCK_LOCK_PREFIX "incw %0"
 		:"+m" (lock->slock)
 		:
 		:"memory", "cc");
 }
+#endif
 
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
 {
-- 
cgit v1.2.3


From d50efc6c40620b2e11648cac64ebf4a824e40382 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:00 +0100
Subject: x86: fix UML and -regparm=3

introduce the "asmregparm" calling convention: for functions
implemented in assembly with a fixed regparm input parameters
calling convention.

mark the semaphore and rwsem slowpath functions with that.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/linkage.h      |  5 +++++
 include/asm-x86/rwsem.h        | 12 ++++++++----
 include/asm-x86/semaphore_32.h |  8 ++++----
 include/linux/linkage.h        |  4 ++++
 lib/rwsem.c                    |  8 ++++----
 5 files changed, 25 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h
index 5a4c95905420..31739c7d66a9 100644
--- a/include/asm-x86/linkage.h
+++ b/include/asm-x86/linkage.h
@@ -9,6 +9,11 @@
 #ifdef CONFIG_X86_32
 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
 #define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret))
+/*
+ * For 32-bit UML - mark functions implemented in assembly that use
+ * regparm input parameters:
+ */
+#define asmregparm __attribute__((regparm(3)))
 #endif
 
 #ifdef CONFIG_X86_ALIGNMENT_16
diff --git a/include/asm-x86/rwsem.h b/include/asm-x86/rwsem.h
index a7e7e14cb430..520a379f4b80 100644
--- a/include/asm-x86/rwsem.h
+++ b/include/asm-x86/rwsem.h
@@ -44,10 +44,14 @@
 
 struct rwsem_waiter;
 
-extern struct rw_semaphore *FASTCALL(rwsem_down_read_failed(struct rw_semaphore *sem));
-extern struct rw_semaphore *FASTCALL(rwsem_down_write_failed(struct rw_semaphore *sem));
-extern struct rw_semaphore *FASTCALL(rwsem_wake(struct rw_semaphore *));
-extern struct rw_semaphore *FASTCALL(rwsem_downgrade_wake(struct rw_semaphore *sem));
+extern asmregparm struct rw_semaphore *
+ rwsem_down_read_failed(struct rw_semaphore *sem);
+extern asmregparm struct rw_semaphore *
+ rwsem_down_write_failed(struct rw_semaphore *sem);
+extern asmregparm struct rw_semaphore *
+ rwsem_wake(struct rw_semaphore *);
+extern asmregparm struct rw_semaphore *
+ rwsem_downgrade_wake(struct rw_semaphore *sem);
 
 /*
  * the semaphore definition
diff --git a/include/asm-x86/semaphore_32.h b/include/asm-x86/semaphore_32.h
index cc826e85323f..ac96d3804d0c 100644
--- a/include/asm-x86/semaphore_32.h
+++ b/include/asm-x86/semaphore_32.h
@@ -83,10 +83,10 @@ static inline void init_MUTEX_LOCKED (struct semaphore *sem)
 	sema_init(sem, 0);
 }
 
-void __down_failed(void /* special register calling convention */);
-int  __down_failed_interruptible(void  /* params in registers */);
-int  __down_failed_trylock(void  /* params in registers */);
-void __up_wakeup(void /* special register calling convention */);
+extern asmregparm void __down_failed(atomic_t *count_ptr);
+extern asmregparm int  __down_failed_interruptible(atomic_t *count_ptr);
+extern asmregparm int  __down_failed_trylock(atomic_t *count_ptr);
+extern asmregparm void __up_wakeup(atomic_t *count_ptr);
 
 /*
  * This is ugly, but we want the default case to fall through.
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index ff203dd02919..ceeeb5ea5b15 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -13,6 +13,10 @@
 #define asmlinkage CPP_ASMLINKAGE
 #endif
 
+#ifndef asmregparm
+# define asmregparm
+#endif
+
 #ifndef prevent_tail_call
 # define prevent_tail_call(ret) do { } while (0)
 #endif
diff --git a/lib/rwsem.c b/lib/rwsem.c
index 7d02700a4b0e..3e3365e5665e 100644
--- a/lib/rwsem.c
+++ b/lib/rwsem.c
@@ -187,7 +187,7 @@ rwsem_down_failed_common(struct rw_semaphore *sem,
 /*
  * wait for the read lock to be granted
  */
-struct rw_semaphore fastcall __sched *
+asmregparm struct rw_semaphore __sched *
 rwsem_down_read_failed(struct rw_semaphore *sem)
 {
 	struct rwsem_waiter waiter;
@@ -201,7 +201,7 @@ rwsem_down_read_failed(struct rw_semaphore *sem)
 /*
  * wait for the write lock to be granted
  */
-struct rw_semaphore fastcall __sched *
+asmregparm struct rw_semaphore __sched *
 rwsem_down_write_failed(struct rw_semaphore *sem)
 {
 	struct rwsem_waiter waiter;
@@ -216,7 +216,7 @@ rwsem_down_write_failed(struct rw_semaphore *sem)
  * handle waking up a waiter on the semaphore
  * - up_read/up_write has decremented the active part of count if we come here
  */
-struct rw_semaphore fastcall *rwsem_wake(struct rw_semaphore *sem)
+asmregparm struct rw_semaphore *rwsem_wake(struct rw_semaphore *sem)
 {
 	unsigned long flags;
 
@@ -236,7 +236,7 @@ struct rw_semaphore fastcall *rwsem_wake(struct rw_semaphore *sem)
  * - caller incremented waiting part of count and discovered it still negative
  * - just wake up any readers at the front of the queue
  */
-struct rw_semaphore fastcall *rwsem_downgrade_wake(struct rw_semaphore *sem)
+asmregparm struct rw_semaphore *rwsem_downgrade_wake(struct rw_semaphore *sem)
 {
 	unsigned long flags;
 
-- 
cgit v1.2.3


From 1379a5ce3ffc549a7ff3daffc49c5e1c372717a3 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:00 +0100
Subject: x86: move get_segment_eip() to step.c

get_segment_eip has similarities to convert_rip_to_linear(),
and is used in a similar context.  Move get_segment_eip to
step.c to allow easier consolidation.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/step.c   | 81 ++++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/mm/fault_32.c   | 77 ---------------------------------------------
 arch/x86/mm/fault_64.c   | 77 ---------------------------------------------
 include/asm-x86/ptrace.h |  2 ++
 4 files changed, 83 insertions(+), 154 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 21ea22fda5fc..5884dd485db8 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -5,6 +5,87 @@
 #include <linux/mm.h>
 #include <linux/ptrace.h>
 
+#ifdef CONFIG_X86_32
+#include <linux/uaccess.h>
+
+#include <asm/desc.h>
+
+/*
+ * Return EIP plus the CS segment base.  The segment limit is also
+ * adjusted, clamped to the kernel/user address space (whichever is
+ * appropriate), and returned in *eip_limit.
+ *
+ * The segment is checked, because it might have been changed by another
+ * task between the original faulting instruction and here.
+ *
+ * If CS is no longer a valid code segment, or if EIP is beyond the
+ * limit, or if it is a kernel address when CS is not a kernel segment,
+ * then the returned value will be greater than *eip_limit.
+ *
+ * This is slow, but is very rarely executed.
+ */
+unsigned long get_segment_eip(struct pt_regs *regs,
+					    unsigned long *eip_limit)
+{
+	unsigned long ip = regs->ip;
+	unsigned seg = regs->cs & 0xffff;
+	u32 seg_ar, seg_limit, base, *desc;
+
+	/* Unlikely, but must come before segment checks. */
+	if (unlikely(regs->flags & VM_MASK)) {
+		base = seg << 4;
+		*eip_limit = base + 0xffff;
+		return base + (ip & 0xffff);
+	}
+
+	/* The standard kernel/user address space limit. */
+	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
+
+	/* By far the most common cases. */
+	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
+		return ip;
+
+	/* Check the segment exists, is within the current LDT/GDT size,
+	   that kernel/user (ring 0..3) has the appropriate privilege,
+	   that it's a code segment, and get the limit. */
+	__asm__("larl %3,%0; lsll %3,%1"
+		 : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
+	if ((~seg_ar & 0x9800) || ip > seg_limit) {
+		*eip_limit = 0;
+		return 1;	 /* So that returned ip > *eip_limit. */
+	}
+
+	/* Get the GDT/LDT descriptor base.
+	   When you look for races in this code remember that
+	   LDT and other horrors are only used in user space. */
+	if (seg & (1<<2)) {
+		/* Must lock the LDT while reading it. */
+		mutex_lock(&current->mm->context.lock);
+		desc = current->mm->context.ldt;
+		desc = (void *)desc + (seg & ~7);
+	} else {
+		/* Must disable preemption while reading the GDT. */
+		desc = (u32 *)get_cpu_gdt_table(get_cpu());
+		desc = (void *)desc + (seg & ~7);
+	}
+
+	/* Decode the code segment base from the descriptor */
+	base = get_desc_base((struct desc_struct *)desc);
+
+	if (seg & (1<<2))
+		mutex_unlock(&current->mm->context.lock);
+	else
+		put_cpu();
+
+	/* Adjust EIP and segment limit, and clamp at the kernel limit.
+	   It's legitimate for segments to wrap at 0xffffffff. */
+	seg_limit += base;
+	if (seg_limit < *eip_limit && seg_limit >= base)
+		*eip_limit = seg_limit;
+	return ip + base;
+}
+#endif
+
 #ifdef CONFIG_X86_32
 static
 #endif
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 421e276770ad..b92922a1d65f 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -61,83 +61,6 @@ static inline int notify_page_fault(struct pt_regs *regs)
 #endif
 }
 
-#ifdef CONFIG_X86_32
-/*
- * Return EIP plus the CS segment base.  The segment limit is also
- * adjusted, clamped to the kernel/user address space (whichever is
- * appropriate), and returned in *eip_limit.
- *
- * The segment is checked, because it might have been changed by another
- * task between the original faulting instruction and here.
- *
- * If CS is no longer a valid code segment, or if EIP is beyond the
- * limit, or if it is a kernel address when CS is not a kernel segment,
- * then the returned value will be greater than *eip_limit.
- *
- * This is slow, but is very rarely executed.
- */
-static inline unsigned long get_segment_eip(struct pt_regs *regs,
-					    unsigned long *eip_limit)
-{
-	unsigned long ip = regs->ip;
-	unsigned seg = regs->cs & 0xffff;
-	u32 seg_ar, seg_limit, base, *desc;
-
-	/* Unlikely, but must come before segment checks. */
-	if (unlikely(regs->flags & VM_MASK)) {
-		base = seg << 4;
-		*eip_limit = base + 0xffff;
-		return base + (ip & 0xffff);
-	}
-
-	/* The standard kernel/user address space limit. */
-	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
-
-	/* By far the most common cases. */
-	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
-		return ip;
-
-	/* Check the segment exists, is within the current LDT/GDT size,
-	   that kernel/user (ring 0..3) has the appropriate privilege,
-	   that it's a code segment, and get the limit. */
-	__asm__ ("larl %3,%0; lsll %3,%1"
-		 : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
-	if ((~seg_ar & 0x9800) || ip > seg_limit) {
-		*eip_limit = 0;
-		return 1;	 /* So that returned ip > *eip_limit. */
-	}
-
-	/* Get the GDT/LDT descriptor base.
-	   When you look for races in this code remember that
-	   LDT and other horrors are only used in user space. */
-	if (seg & (1<<2)) {
-		/* Must lock the LDT while reading it. */
-		mutex_lock(&current->mm->context.lock);
-		desc = current->mm->context.ldt;
-		desc = (void *)desc + (seg & ~7);
-	} else {
-		/* Must disable preemption while reading the GDT. */
-		desc = (u32 *)get_cpu_gdt_table(get_cpu());
-		desc = (void *)desc + (seg & ~7);
-	}
-
-	/* Decode the code segment base from the descriptor */
-	base = get_desc_base((struct desc_struct *)desc);
-
-	if (seg & (1<<2))
-		mutex_unlock(&current->mm->context.lock);
-	else
-		put_cpu();
-
-	/* Adjust EIP and segment limit, and clamp at the kernel limit.
-	   It's legitimate for segments to wrap at 0xffffffff. */
-	seg_limit += base;
-	if (seg_limit < *eip_limit && seg_limit >= base)
-		*eip_limit = seg_limit;
-	return ip + base;
-}
-#endif
-
 /*
  * X86_32
  * Sometimes AMD Athlon/Opteron CPUs report invalid exceptions on prefetch.
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index 95f142f5b5cc..e82832961d72 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -64,83 +64,6 @@ static inline int notify_page_fault(struct pt_regs *regs)
 #endif
 }
 
-#ifdef CONFIG_X86_32
-/*
- * Return EIP plus the CS segment base.  The segment limit is also
- * adjusted, clamped to the kernel/user address space (whichever is
- * appropriate), and returned in *eip_limit.
- *
- * The segment is checked, because it might have been changed by another
- * task between the original faulting instruction and here.
- *
- * If CS is no longer a valid code segment, or if EIP is beyond the
- * limit, or if it is a kernel address when CS is not a kernel segment,
- * then the returned value will be greater than *eip_limit.
- *
- * This is slow, but is very rarely executed.
- */
-static inline unsigned long get_segment_eip(struct pt_regs *regs,
-					    unsigned long *eip_limit)
-{
-	unsigned long ip = regs->ip;
-	unsigned seg = regs->cs & 0xffff;
-	u32 seg_ar, seg_limit, base, *desc;
-
-	/* Unlikely, but must come before segment checks. */
-	if (unlikely(regs->flags & VM_MASK)) {
-		base = seg << 4;
-		*eip_limit = base + 0xffff;
-		return base + (ip & 0xffff);
-	}
-
-	/* The standard kernel/user address space limit. */
-	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
-
-	/* By far the most common cases. */
-	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
-		return ip;
-
-	/* Check the segment exists, is within the current LDT/GDT size,
-	   that kernel/user (ring 0..3) has the appropriate privilege,
-	   that it's a code segment, and get the limit. */
-	__asm__("larl %3,%0; lsll %3,%1"
-		 : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
-	if ((~seg_ar & 0x9800) || ip > seg_limit) {
-		*eip_limit = 0;
-		return 1;	 /* So that returned ip > *eip_limit. */
-	}
-
-	/* Get the GDT/LDT descriptor base.
-	   When you look for races in this code remember that
-	   LDT and other horrors are only used in user space. */
-	if (seg & (1<<2)) {
-		/* Must lock the LDT while reading it. */
-		mutex_lock(&current->mm->context.lock);
-		desc = current->mm->context.ldt;
-		desc = (void *)desc + (seg & ~7);
-	} else {
-		/* Must disable preemption while reading the GDT. */
-		desc = (u32 *)get_cpu_gdt_table(get_cpu());
-		desc = (void *)desc + (seg & ~7);
-	}
-
-	/* Decode the code segment base from the descriptor */
-	base = get_desc_base((struct desc_struct *)desc);
-
-	if (seg & (1<<2))
-		mutex_unlock(&current->mm->context.lock);
-	else
-		put_cpu();
-
-	/* Adjust EIP and segment limit, and clamp at the kernel limit.
-	   It's legitimate for segments to wrap at 0xffffffff. */
-	seg_limit += base;
-	if (seg_limit < *eip_limit && seg_limit >= base)
-		*eip_limit = seg_limit;
-	return ip + base;
-}
-#endif
-
 /*
  * X86_32
  * Sometimes AMD Athlon/Opteron CPUs report invalid exceptions on prefetch.
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 61946fe8c085..cc4456667d89 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -184,6 +184,8 @@ convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 #ifdef __KERNEL__
 
+unsigned long get_segment_eip(struct pt_regs *regs, unsigned long *eip_limit);
+
 /*
  * These are defined as per linux/ptrace.h, which see.
  */
-- 
cgit v1.2.3


From 2a6648e65a2939b80c44262975176a15bac3a75e Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:33:01 +0100
Subject: x86: <asm/segment.h>: boot GDT entries are 32/64-independent

The boot GDT entries are common between 32- and 64-bit mode, so move
them to common code instead of having two identical copies.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/segment.h    | 24 ++++++++----------------
 include/asm-x86/segment_32.h |  1 -
 2 files changed, 8 insertions(+), 17 deletions(-)
 delete mode 100644 include/asm-x86/segment_32.h

(limited to 'include')

diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index c7509df79141..1ff05a9f5513 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -1,6 +1,14 @@
 #ifndef _ASM_X86_SEGMENT_H_
 #define _ASM_X86_SEGMENT_H_
 
+/* Simple and small GDT entries for booting only */
+
+#define GDT_ENTRY_BOOT_CS	2
+#define __BOOT_CS		(GDT_ENTRY_BOOT_CS * 8)
+
+#define GDT_ENTRY_BOOT_DS	(GDT_ENTRY_BOOT_CS + 1)
+#define __BOOT_DS		(GDT_ENTRY_BOOT_DS * 8)
+
 #ifdef CONFIG_X86_32
 /*
  * The layout of the per-CPU GDT under Linux:
@@ -86,14 +94,6 @@
  */
 #define GDT_ENTRIES 32
 
-/* Simple and small GDT entries for booting only */
-
-#define GDT_ENTRY_BOOT_CS		2
-#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
-
-#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
-#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
-
 /* The PnP BIOS entries in the GDT */
 #define GDT_ENTRY_PNPBIOS_CS32		(GDT_ENTRY_PNPBIOS_BASE + 0)
 #define GDT_ENTRY_PNPBIOS_CS16		(GDT_ENTRY_PNPBIOS_BASE + 1)
@@ -136,14 +136,6 @@
 #else
 #include <asm/cache.h>
 
-/* Simple and small GDT entries for booting only */
-
-#define GDT_ENTRY_BOOT_CS		2
-#define __BOOT_CS	(GDT_ENTRY_BOOT_CS * 8)
-
-#define GDT_ENTRY_BOOT_DS		(GDT_ENTRY_BOOT_CS + 1)
-#define __BOOT_DS	(GDT_ENTRY_BOOT_DS * 8)
-
 #define __KERNEL_CS	0x10
 #define __KERNEL_DS	0x18
 
diff --git a/include/asm-x86/segment_32.h b/include/asm-x86/segment_32.h
deleted file mode 100644
index 8b137891791f..000000000000
--- a/include/asm-x86/segment_32.h
+++ /dev/null
@@ -1 +0,0 @@
-
-- 
cgit v1.2.3


From 88089519f302f1296b4739be45699f06f728ec31 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 30 Jan 2008 13:33:02 +0100
Subject: x86 setup: initialize LDTR and TR to make life easier to Intel VT

Intel VT doesn't like to engage when the protected-mode state isn't
fully initialized.  Make life easier for it by initializing LDTR (to
null) and TR (to a dummy hunk of low memory which will never actually
be touched.)

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/boot/pm.c        | 4 ++++
 arch/x86/boot/pmjump.S    | 7 +++++++
 include/asm-x86/segment.h | 3 +++
 3 files changed, 14 insertions(+)

(limited to 'include')

diff --git a/arch/x86/boot/pm.c b/arch/x86/boot/pm.c
index 09fb342cc62e..b23cbdc7d547 100644
--- a/arch/x86/boot/pm.c
+++ b/arch/x86/boot/pm.c
@@ -121,6 +121,10 @@ static void setup_gdt(void)
 		[GDT_ENTRY_BOOT_CS] = GDT_ENTRY(0xc09b, 0, 0xfffff),
 		/* DS: data, read/write, 4 GB, base 0 */
 		[GDT_ENTRY_BOOT_DS] = GDT_ENTRY(0xc093, 0, 0xfffff),
+		/* TSS: 32-bit tss, 104 bytes, base 4096 */
+		/* We only have a TSS here to keep Intel VT happy;
+		   we don't actually use it for anything. */
+		[GDT_ENTRY_BOOT_TSS] = GDT_ENTRY(0x0089, 4096, 103),
 	};
 	/* Xen HVM incorrectly stores a pointer to the gdt_ptr, instead
 	   of the gdt_ptr contents.  Thus, make it static so it will
diff --git a/arch/x86/boot/pmjump.S b/arch/x86/boot/pmjump.S
index ef0da1f2c7fd..f7153d0d476e 100644
--- a/arch/x86/boot/pmjump.S
+++ b/arch/x86/boot/pmjump.S
@@ -36,6 +36,7 @@ protected_mode_jump:
 	addl	%ebx, 2f
 
 	movw	$__BOOT_DS, %cx
+	movw	$__BOOT_TSS, %di
 
 	movl	%cr0, %edx
 	orb	$1, %dl			# Protected mode (PE) bit
@@ -63,6 +64,9 @@ in_pm32:
 	# a valid stack if some debugging hack wants to use it.
 	addl	%ebx, %esp
 
+	# Set up TR to make Intel VT happy
+	ltr	%di
+
 	# Clear registers to allow for future extensions to the
 	# 32-bit boot protocol
 	xorl	%ecx, %ecx
@@ -71,6 +75,9 @@ in_pm32:
 	xorl	%ebp, %ebp
 	xorl	%edi, %edi
 
+	# Set up LDTR to make Intel VT happy
+	lldt	%cx
+
 	jmpl	*%eax			# Jump to the 32-bit entrypoint
 
 	.size	in_pm32, .-in_pm32
diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index 1ff05a9f5513..57c8d3723836 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -9,6 +9,9 @@
 #define GDT_ENTRY_BOOT_DS	(GDT_ENTRY_BOOT_CS + 1)
 #define __BOOT_DS		(GDT_ENTRY_BOOT_DS * 8)
 
+#define GDT_ENTRY_BOOT_TSS	(GDT_ENTRY_BOOT_CS + 2)
+#define __BOOT_TSS		(GDT_ENTRY_BOOT_TSS * 8)
+
 #ifdef CONFIG_X86_32
 /*
  * The layout of the per-CPU GDT under Linux:
-- 
cgit v1.2.3


From 8866cd9dc9d0bbadcf361a14e0cdfecb66473087 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 30 Jan 2008 13:33:06 +0100
Subject: x86: early_idt_handler improvements, 64-bit

It's not too pretty, but I found this made the "PANIC: early exception"
messages become much more reliably useful: 1. print the vector number,
2. print the %cs value, 3. handle error-code-pushing vs non-pushing vectors.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/head64.c  |  2 +-
 arch/x86/kernel/head_64.S | 34 ++++++++++++++++++++++++++++++----
 include/asm-x86/segment.h |  6 ++++++
 3 files changed, 37 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 4a1c1356c41a..85c1e6bf8022 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -59,7 +59,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	zap_identity_mappings();
 
 	for (i = 0; i < IDT_ENTRIES; i++)
-		set_intr_gate(i, early_idt_handler);
+		set_intr_gate(i, &early_idt_handlers[i]);
 	load_idt((const struct desc_ptr *)&idt_descr);
 
 	early_printk("Kernel alive\n");
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index c31b1c96a9d3..8b4c35cb519a 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -267,14 +267,40 @@ init_rsp:
 bad_address:
 	jmp bad_address
 
+.macro early_idt_tramp first, last
+	.ifgt \last-\first
+	early_idt_tramp \first, \last-1
+	.endif
+	movl $\last,%esi
+	jmp early_idt_handler
+.endm
+
+	.globl early_idt_handlers
+early_idt_handlers:
+	early_idt_tramp 0, 63
+	early_idt_tramp 64, 127
+	early_idt_tramp 128, 191
+	early_idt_tramp 192, 255
+
 ENTRY(early_idt_handler)
 	cmpl $2,early_recursion_flag(%rip)
 	jz  1f
 	incl early_recursion_flag(%rip)
-	xorl %eax,%eax
-	movq 8(%rsp),%rsi	# get rip
-	movq (%rsp),%rdx
 	GET_CR2_INTO_RCX
+	movq %rcx,%r9
+	xorl %r8d,%r8d		# zero for error code
+	movl %esi,%ecx		# get vector number
+	# Test %ecx against mask of vectors that push error code.
+	cmpl $31,%ecx
+	ja 0f
+	movl $1,%eax
+	salq %cl,%rax
+	testl $0x27d00,%eax
+	je 0f
+	popq %r8		# get error code
+0:	movq 0(%rsp),%rcx	# get ip
+	movq 8(%rsp),%rdx	# get cs
+	xorl %eax,%eax
 	leaq early_idt_msg(%rip),%rdi
 	call early_printk
 	cmpl $2,early_recursion_flag(%rip)
@@ -291,7 +317,7 @@ early_recursion_flag:
 	.long 0
 
 early_idt_msg:
-	.asciz "PANIC: early exception rip %lx error %lx cr2 %lx\n"
+	.asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n"
 early_idt_ripmsg:
 	.asciz "RIP %s\n"
 
diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index 57c8d3723836..23f0535fec61 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -195,4 +195,10 @@
 #define GDT_ENTRY_TLS_ENTRIES 3
 #define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8)
 
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+extern const char early_idt_handlers[IDT_ENTRIES][10];
+#endif
+#endif
+
 #endif
-- 
cgit v1.2.3


From 076f9776f5d8d131b36955db8641aba3893c2c1b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:06 +0100
Subject: x86: make early printk selectable on 64-bit as well

Enable CONFIG_EMBEDDED to select CONFIG_EARLY_PRINTK on 64-bit as well.

saves ~2K:

   text    data     bss     dec     hex filename
   7290283 3672091 1907848 12870222         c4624e vmlinux.before
   7288373 3671795 1907848 12868016         c459b0 vmlinux.after

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug    | 2 +-
 arch/x86/kernel/head64.c  | 7 ++++++-
 arch/x86/kernel/head_64.S | 7 +++++++
 include/asm-x86/kdebug.h  | 1 -
 include/linux/kernel.h    | 3 +++
 kernel/printk.c           | 7 +++++++
 6 files changed, 24 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 15854b53badc..88420af98140 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -6,7 +6,7 @@ config TRACE_IRQFLAGS_SUPPORT
 source "lib/Kconfig.debug"
 
 config EARLY_PRINTK
-	bool "Early printk" if EMBEDDED && DEBUG_KERNEL && X86_32
+	bool "Early printk" if EMBEDDED
 	default y
 	help
 	  Write kernel log output directly into the VGA buffer or to a serial
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 85c1e6bf8022..87e031d4abf1 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -58,8 +58,13 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	/* Make NULL pointers segfault */
 	zap_identity_mappings();
 
-	for (i = 0; i < IDT_ENTRIES; i++)
+	for (i = 0; i < IDT_ENTRIES; i++) {
+#ifdef CONFIG_EARLY_PRINTK
 		set_intr_gate(i, &early_idt_handlers[i]);
+#else
+		set_intr_gate(i, early_idt_handler);
+#endif
+	}
 	load_idt((const struct desc_ptr *)&idt_descr);
 
 	early_printk("Kernel alive\n");
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 8b4c35cb519a..1d5a7a361200 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -267,6 +267,7 @@ init_rsp:
 bad_address:
 	jmp bad_address
 
+#ifdef CONFIG_EARLY_PRINTK
 .macro early_idt_tramp first, last
 	.ifgt \last-\first
 	early_idt_tramp \first, \last-1
@@ -281,8 +282,10 @@ early_idt_handlers:
 	early_idt_tramp 64, 127
 	early_idt_tramp 128, 191
 	early_idt_tramp 192, 255
+#endif
 
 ENTRY(early_idt_handler)
+#ifdef CONFIG_EARLY_PRINTK
 	cmpl $2,early_recursion_flag(%rip)
 	jz  1f
 	incl early_recursion_flag(%rip)
@@ -311,8 +314,11 @@ ENTRY(early_idt_handler)
 	movq 8(%rsp),%rsi	# get rip again
 	call __print_symbol
 #endif
+#endif /* EARLY_PRINTK */
 1:	hlt
 	jmp 1b
+
+#ifdef CONFIG_EARLY_PRINTK
 early_recursion_flag:
 	.long 0
 
@@ -320,6 +326,7 @@ early_idt_msg:
 	.asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n"
 early_idt_ripmsg:
 	.asciz "RIP %s\n"
+#endif /* CONFIG_EARLY_PRINTK */
 
 .balign PAGE_SIZE
 
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index a5e5e3b7eb23..e9f42d1ac38f 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -22,7 +22,6 @@ enum die_val {
 	DIE_PAGE_FAULT,
 };
 
-extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 extern void printk_address(unsigned long address);
 extern void die(const char *,struct pt_regs *,long);
 extern int __must_check __die(const char *, struct pt_regs *, long);
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index a7283c9beadf..ff356b2ee478 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -194,6 +194,9 @@ static inline int log_buf_read(int idx) { return 0; }
 static inline int log_buf_copy(char *dest, int idx, int len) { return 0; }
 #endif
 
+extern void __attribute__((format(printf, 1, 2)))
+	early_printk(const char *fmt, ...);
+
 unsigned long int_sqrt(unsigned long);
 
 extern int printk_ratelimit(void);
diff --git a/kernel/printk.c b/kernel/printk.c
index 3b7c968d0ef9..58bbec684119 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -36,6 +36,13 @@
 
 #include <asm/uaccess.h>
 
+/*
+ * Architectures can override it:
+ */
+void __attribute__((weak)) early_printk(const char *fmt, ...)
+{
+}
+
 #define __LOG_BUF_LEN	(1 << CONFIG_LOG_BUF_SHIFT)
 
 /* printk's without a loglevel use this.. */
-- 
cgit v1.2.3


From bc850d6b374fffd08336996f4b4d3bbd6bf427f6 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Wed, 30 Jan 2008 13:33:07 +0100
Subject: x86: add the capability to print fuzzy backtraces

For enhancing the 32 bit EBP based backtracer, I need the capability
for the backtracer to tell it's customer that an entry is either
reliable or unreliable, and the backtrace printing code then needs to
print the unreliable ones slightly different.

This patch adds the basic capability, the next patch will add a user
of this capability.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/pci-gart_64.c |  2 +-
 arch/x86/kernel/process_64.c  |  4 ++--
 arch/x86/kernel/stacktrace.c  |  2 +-
 arch/x86/kernel/traps_32.c    |  8 +++++---
 arch/x86/kernel/traps_64.c    | 39 +++++++++++++++++++++++----------------
 arch/x86/mm/fault_64.c        |  2 +-
 arch/x86/oprofile/backtrace.c |  2 +-
 include/asm-x86/kdebug.h      |  5 +++--
 include/asm-x86/stacktrace.h  |  5 +++--
 9 files changed, 40 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index d2b46b489412..04ca5c5221d7 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -167,7 +167,7 @@ static void dump_leak(void)
 	       iommu_leak_pages);
 	for (i = 0; i < iommu_leak_pages; i += 2) {
 		printk(KERN_DEBUG "%lu: ", iommu_pages-i);
-		printk_address((unsigned long) iommu_leak_tab[iommu_pages-i]);
+		printk_address((unsigned long) iommu_leak_tab[iommu_pages-i], 0);
 		printk(KERN_CONT "%c", (i+1)%2 == 0 ? '\n' : ' ');
 	}
 	printk(KERN_DEBUG "\n");
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index a0130eb2fa50..383760bfd283 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -329,7 +329,7 @@ void __show_regs(struct pt_regs * regs)
 		(int)strcspn(init_utsname()->version, " "),
 		init_utsname()->version);
 	printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip);
-	printk_address(regs->ip);
+	printk_address(regs->ip, regs->bp);
 	printk("RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss, regs->sp,
 		regs->flags);
 	printk("RAX: %016lx RBX: %016lx RCX: %016lx\n",
@@ -377,7 +377,7 @@ void show_regs(struct pt_regs *regs)
 {
 	printk("CPU %d:", smp_processor_id());
 	__show_regs(regs);
-	show_trace(NULL, regs, (void *)(regs + 1));
+	show_trace(NULL, regs, (void *)(regs + 1), regs->bp);
 }
 
 /*
diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index c571edd11878..8c4e4f5bf040 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -22,7 +22,7 @@ static int save_stack_stack(void *data, char *name)
 	return -1;
 }
 
-static void save_stack_address(void *data, unsigned long addr)
+static void save_stack_address(void *data, unsigned long addr, int reliable)
 {
 	struct stack_trace *trace = (struct stack_trace *)data;
 	if (trace->skip > 0) {
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index acc9af260fac..8ef8a9ddfec6 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -126,7 +126,7 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 
 		addr = frame->return_address;
 		if (__kernel_text_address(addr))
-			ops->address(data, addr);
+			ops->address(data, addr, 1);
 		/*
 		 * break out of recursive entries (such as
 		 * end_of_stack_stop_unwind_function). Also,
@@ -145,7 +145,7 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 
 		addr = *stack++;
 		if (__kernel_text_address(addr))
-			ops->address(data, addr);
+			ops->address(data, addr, 1);
 	}
 #endif
 	return bp;
@@ -220,9 +220,11 @@ static int print_trace_stack(void *data, char *name)
 /*
  * Print one address/symbol entries per line.
  */
-static void print_trace_address(void *data, unsigned long addr)
+static void print_trace_address(void *data, unsigned long addr, int reliable)
 {
 	printk("%s [<%08lx>] ", (char *)data, addr);
+	if (!reliable)
+		printk("? ");
 	print_symbol("%s\n", addr);
 	touch_nmi_watchdog();
 }
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 37b07d08704b..62c4d8f46ee9 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -99,13 +99,14 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
 int kstack_depth_to_print = 12;
 
 #ifdef CONFIG_KALLSYMS
-void printk_address(unsigned long address)
+void printk_address(unsigned long address, int reliable)
 {
 	unsigned long offset = 0, symsize;
 	const char *symname;
 	char *modname;
 	char *delim = ":";
 	char namebuf[128];
+	char reliab[4] = "";;
 
 	symname = kallsyms_lookup(address, &symsize, &offset,
 					&modname, namebuf);
@@ -113,13 +114,16 @@ void printk_address(unsigned long address)
 		printk(" [<%016lx>]\n", address);
 		return;
 	}
+	if (!reliable)
+		strcpy(reliab, "? ");
+
 	if (!modname)
 		modname = delim = ""; 		
-	printk(" [<%016lx>] %s%s%s%s+0x%lx/0x%lx\n",
-		address, delim, modname, delim, symname, offset, symsize);
+	printk(" [<%016lx>] %s%s%s%s%s+0x%lx/0x%lx\n",
+		address, reliab, delim, modname, delim, symname, offset, symsize);
 }
 #else
-void printk_address(unsigned long address)
+void printk_address(unsigned long address, int reliable)
 {
 	printk(" [<%016lx>]\n", address);
 }
@@ -215,7 +219,7 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
 }
 
 void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
-		unsigned long *stack,
+		unsigned long *stack, unsigned long bp,
 		const struct stacktrace_ops *ops, void *data)
 {
 	const unsigned cpu = get_cpu();
@@ -252,7 +256,7 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
 			 * down the cause of the crash will be able to figure \
 			 * out the call path that was taken. \
 			 */ \
-			ops->address(data, addr);   \
+			ops->address(data, addr, 1);   \
 		} \
 	} while (0)
 
@@ -331,10 +335,10 @@ static int print_trace_stack(void *data, char *name)
 	return 0;
 }
 
-static void print_trace_address(void *data, unsigned long addr)
+static void print_trace_address(void *data, unsigned long addr, int reliable)
 {
 	touch_nmi_watchdog();
-	printk_address(addr);
+	printk_address(addr, reliable);
 }
 
 static const struct stacktrace_ops print_trace_ops = {
@@ -345,15 +349,17 @@ static const struct stacktrace_ops print_trace_ops = {
 };
 
 void
-show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack)
+show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack,
+		unsigned long bp)
 {
 	printk("\nCall Trace:\n");
-	dump_trace(tsk, regs, stack, &print_trace_ops, NULL);
+	dump_trace(tsk, regs, stack, bp, &print_trace_ops, NULL);
 	printk("\n");
 }
 
 static void
-_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp)
+_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp,
+							unsigned long bp)
 {
 	unsigned long *stack;
 	int i;
@@ -387,12 +393,12 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp)
 		printk(" %016lx", *stack++);
 		touch_nmi_watchdog();
 	}
-	show_trace(tsk, regs, sp);
+	show_trace(tsk, regs, sp, bp);
 }
 
 void show_stack(struct task_struct *tsk, unsigned long * sp)
 {
-	_show_stack(tsk, NULL, sp);
+	_show_stack(tsk, NULL, sp, 0);
 }
 
 /*
@@ -401,13 +407,14 @@ void show_stack(struct task_struct *tsk, unsigned long * sp)
 void dump_stack(void)
 {
 	unsigned long dummy;
+	unsigned long bp = 0;
 
 	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
 		current->pid, current->comm, print_tainted(),
 		init_utsname()->release,
 		(int)strcspn(init_utsname()->version, " "),
 		init_utsname()->version);
-	show_trace(NULL, NULL, &dummy);
+	show_trace(NULL, NULL, &dummy, bp);
 }
 
 EXPORT_SYMBOL(dump_stack);
@@ -432,7 +439,7 @@ void show_registers(struct pt_regs *regs)
 	 */
 	if (in_kernel) {
 		printk("Stack: ");
-		_show_stack(NULL, regs, (unsigned long*)sp);
+		_show_stack(NULL, regs, (unsigned long *)sp, regs->bp);
 
 		printk("\nCode: ");
 		if (regs->ip < PAGE_OFFSET)
@@ -527,7 +534,7 @@ int __kprobes __die(const char * str, struct pt_regs * regs, long err)
 	add_taint(TAINT_DIE);
 	/* Executive summary in case the oops scrolled away */
 	printk(KERN_ALERT "RIP ");
-	printk_address(regs->ip);
+	printk_address(regs->ip, regs->bp);
 	printk(" RSP <%016lx>\n", regs->sp);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index e82832961d72..cf7e99895b91 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -578,7 +578,7 @@ no_context:
 	else
 		printk(KERN_ALERT "Unable to handle kernel paging request");
 	printk(" at %016lx RIP: \n" KERN_ALERT, address);
-	printk_address(regs->ip);
+	printk_address(regs->ip, regs->bp);
 	dump_pagetable(address);
 	tsk->thread.cr2 = address;
 	tsk->thread.trap_no = 14;
diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c
index cc353a0b183e..671a7ecf11aa 100644
--- a/arch/x86/oprofile/backtrace.c
+++ b/arch/x86/oprofile/backtrace.c
@@ -32,7 +32,7 @@ static int backtrace_stack(void *data, char *name)
 	return 0;
 }
 
-static void backtrace_address(void *data, unsigned long addr)
+static void backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	unsigned int *depth = data;
 
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index e9f42d1ac38f..dd442a1632c0 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -22,12 +22,13 @@ enum die_val {
 	DIE_PAGE_FAULT,
 };
 
-extern void printk_address(unsigned long address);
+extern void printk_address(unsigned long address, int reliable);
 extern void die(const char *,struct pt_regs *,long);
 extern int __must_check __die(const char *, struct pt_regs *, long);
 extern void show_registers(struct pt_regs *regs);
 extern void __show_registers(struct pt_regs *, int all);
-extern void show_trace(struct task_struct *, struct pt_regs *, unsigned long *);
+extern void show_trace(struct task_struct *t, struct pt_regs *regs,
+			unsigned long *sp, unsigned long bp);
 extern void __show_regs(struct pt_regs *regs);
 extern void show_regs(struct pt_regs *regs);
 extern void dump_pagetable(unsigned long);
diff --git a/include/asm-x86/stacktrace.h b/include/asm-x86/stacktrace.h
index 70dd5bae3235..30f82526a8e2 100644
--- a/include/asm-x86/stacktrace.h
+++ b/include/asm-x86/stacktrace.h
@@ -9,12 +9,13 @@ struct stacktrace_ops {
 	void (*warning)(void *data, char *msg);
 	/* msg must contain %s for the symbol */
 	void (*warning_symbol)(void *data, char *msg, unsigned long symbol);
-	void (*address)(void *data, unsigned long address);
+	void (*address)(void *data, unsigned long address, int reliable);
 	/* On negative return stop dumping */
 	int (*stack)(void *data, char *name);
 };
 
-void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack,
+void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp,
 		const struct stacktrace_ops *ops, void *data);
 
 #endif
-- 
cgit v1.2.3


From 306c142f99caef5a30a0d61a39a956c069941b63 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:08 +0100
Subject: x86: pat: e820 cleanup

NOP change.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/ioremap_64.c  | 4 +---
 include/asm-x86/e820.h    | 3 +++
 include/asm-x86/e820_64.h | 5 +++++
 3 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap_64.c b/arch/x86/mm/ioremap_64.c
index b03db4ca9cad..6fbd3c7932e0 100644
--- a/arch/x86/mm/ioremap_64.c
+++ b/arch/x86/mm/ioremap_64.c
@@ -19,6 +19,7 @@
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 #include <asm/proto.h>
+#include <asm/e820.h>
 
 unsigned long __phys_addr(unsigned long x)
 {
@@ -28,9 +29,6 @@ unsigned long __phys_addr(unsigned long x)
 }
 EXPORT_SYMBOL(__phys_addr);
 
-#define ISA_START_ADDRESS      0xa0000
-#define ISA_END_ADDRESS                0x100000
-
 /*
  * Fix up the linear direct mapping of the kernel to avoid cache attribute
  * conflicts.
diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h
index 3e214f39fad3..f96f1853bc41 100644
--- a/include/asm-x86/e820.h
+++ b/include/asm-x86/e820.h
@@ -22,6 +22,9 @@ struct e820map {
 };
 #endif /* __ASSEMBLY__ */
 
+#define ISA_START_ADDRESS	0xa0000
+#define ISA_END_ADDRESS		0x100000
+
 #ifdef __KERNEL__
 #ifdef CONFIG_X86_32
 # include "e820_32.h"
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index 1c7ba8804176..8cba49da4795 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -26,6 +26,10 @@ extern void e820_reserve_resources(struct resource *code_resource,
 extern void e820_mark_nosave_regions(void);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
+extern int e820_any_non_reserved(unsigned long start, unsigned long end);
+extern int is_memory_any_valid(unsigned long start, unsigned long end);
+extern int e820_all_non_reserved(unsigned long start, unsigned long end);
+extern int is_memory_all_valid(unsigned long start, unsigned long end);
 extern unsigned long e820_hole_size(unsigned long start, unsigned long end);
 
 extern void e820_setup_gap(void);
@@ -38,6 +42,7 @@ extern struct e820map e820;
 
 extern unsigned ebda_addr, ebda_size;
 extern unsigned long nodemap_addr, nodemap_size;
+
 #endif/*!__ASSEMBLY__*/
 
 #endif/*__E820_HEADER*/
-- 
cgit v1.2.3


From aaf230424204864e2833dcc1da23e2cb0b9f39cd Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Wed, 30 Jan 2008 13:33:09 +0100
Subject: x86: disable the GART early, 64-bit

For K8 system: 4G RAM with memory hole remapping enabled, or more than
4G RAM installed.

when try to use kexec second kernel, and the first doesn't include
gart_shutdown. the second kernel could have different aper position than
the first kernel. and second kernel could use that hole as RAM that is
still used by GART set by the first kernel. esp. when try to kexec
2.6.24 with sparse mem enable from previous kernel (from RHEL 5 or SLES
10). the new kernel will use aper by GART (set by first kernel) for
vmemmap. and after new kernel setting one new GART. the position will be
real RAM. the _mapcount set is lost.

Bad page state in process 'swapper'
page:ffffe2000e600020 flags:0x0000000000000000 mapping:0000000000000000 mapcount:1 count:0
Trying to fix it up, but a reboot is needed
Backtrace:
Pid: 0, comm: swapper Not tainted 2.6.24-rc7-smp-gcdf71a10-dirty #13

Call Trace:
 [<ffffffff8026401f>] bad_page+0x63/0x8d
 [<ffffffff80264169>] __free_pages_ok+0x7c/0x2a5
 [<ffffffff80ba75d1>] free_all_bootmem_core+0xd0/0x198
 [<ffffffff80ba3a42>] numa_free_all_bootmem+0x3b/0x76
 [<ffffffff80ba3461>] mem_init+0x3b/0x152
 [<ffffffff80b959d3>] start_kernel+0x236/0x2c2
 [<ffffffff80b9511a>] _sinittext+0x11a/0x121

and
 [ffffe2000e600000-ffffe2000e7fffff] PMD ->ffff81001c200000 on node 0
phys addr is : 0x1c200000

RHEL 5.1 kernel -53 said:
PCI-DMA: aperture base @ 1c000000 size 65536 KB

new kernel said:
Mapping aperture over 65536 KB of RAM @ 3c000000

So could try to disable that GART if possible.

According to Ingo

> hm, i'm wondering, instead of modifying the GART, why dont we simply
> _detect_ whatever GART settings we have inherited, and propagate that
> into our e820 maps? I.e. if there's inconsistency, then punch that out
> from the memory maps and just dont use that memory.
>
> that way it would not matter whether the GART settings came from a [old
> or crashing] Linux kernel that has not called gart_iommu_shutdown(), or
> whether it's a BIOS that has set up an aperture hole inconsistent with
> the memory map it passed. (or the memory map we _think_ i tried to pass
> us)
>
> it would also be more robust to only read and do a memory map quirk
> based on that, than actively trying to change the GART so early in the
> bootup. Later on we have to re-enable the GART _anyway_ and have to
> punch a hole for it.
>
> and as a bonus, we would have shored up our defenses against crappy
> BIOSes as well.

add e820 modification for gart inconsistent setting.

gart_fix_e820=off could be used to disable e820 fix.

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |  4 ++
 arch/x86/kernel/aperture_64.c       | 89 +++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/e820_64.c           | 12 +++++
 arch/x86/kernel/setup_64.c          |  3 ++
 include/asm-x86/e820_64.h           |  1 +
 include/asm-x86/gart.h              |  5 +++
 6 files changed, 114 insertions(+)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 40db7dd1d92a..860a90875491 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -660,6 +660,10 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	gamma=		[HW,DRM]
 
+	gart_fix_e820=  [X86_64] disable the fix e820 for K8 GART
+			Format: off | on
+			default: on
+
 	gdth=		[HW,SCSI]
 			See header of drivers/scsi/gdth.c.
 
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index 52d2beac4556..bf1b469d5847 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -218,6 +218,95 @@ static __u32 __init search_agp_bridge(u32 *order, int *valid_agp)
 	return 0;
 }
 
+static int gart_fix_e820 __initdata = 1;
+
+static int __init parse_gart_mem(char *p)
+{
+	if (!p)
+		return -EINVAL;
+
+	if (!strncmp(p, "off", 3))
+		gart_fix_e820 = 0;
+	else if (!strncmp(p, "on", 2))
+		gart_fix_e820 = 1;
+
+	return 0;
+}
+early_param("gart_fix_e820", parse_gart_mem);
+
+void __init early_gart_iommu_check(void)
+{
+	/*
+	 * in case it is enabled before, esp for kexec/kdump,
+	 * previous kernel already enable that. memset called
+	 * by allocate_aperture/__alloc_bootmem_nopanic cause restart.
+	 * or second kernel have different position for GART hole. and new
+	 * kernel could use hole as RAM that is still used by GART set by
+	 * first kernel
+	 * or BIOS forget to put that in reserved.
+	 * try to update e820 to make that region as reserved.
+	 */
+	int fix, num;
+	u32 ctl;
+	u32 aper_size = 0, aper_order = 0, last_aper_order = 0;
+	u64 aper_base = 0, last_aper_base = 0;
+	int aper_enabled = 0, last_aper_enabled = 0;
+
+	if (!early_pci_allowed())
+		return;
+
+	fix = 0;
+	for (num = 24; num < 32; num++) {
+		if (!early_is_k8_nb(read_pci_config(0, num, 3, 0x00)))
+			continue;
+
+		ctl = read_pci_config(0, num, 3, 0x90);
+		aper_enabled = ctl & 1;
+		aper_order = (ctl >> 1) & 7;
+		aper_size = (32 * 1024 * 1024) << aper_order;
+		aper_base = read_pci_config(0, num, 3, 0x94) & 0x7fff;
+		aper_base <<= 25;
+
+		if ((last_aper_order && aper_order != last_aper_order) ||
+		    (last_aper_base && aper_base != last_aper_base) ||
+		    (last_aper_enabled && aper_enabled != last_aper_enabled)) {
+			fix = 1;
+			break;
+		}
+		last_aper_order = aper_order;
+		last_aper_base = aper_base;
+		last_aper_enabled = aper_enabled;
+	}
+
+	if (!fix && !aper_enabled)
+		return;
+
+	if (!aper_base || !aper_size || aper_base + aper_size > 0x100000000UL)
+		fix = 1;
+
+	if (gart_fix_e820 && !fix && aper_enabled) {
+		if (e820_any_mapped(aper_base, aper_base + aper_size,
+				    E820_RAM)) {
+			/* reserved it, so we can resuse it in second kernel */
+			printk(KERN_INFO "update e820 for GART\n");
+			add_memory_region(aper_base, aper_size, E820_RESERVED);
+			update_e820();
+		}
+		return;
+	}
+
+	/* different nodes have different setting, disable them all at first*/
+	for (num = 24; num < 32; num++) {
+		if (!early_is_k8_nb(read_pci_config(0, num, 3, 0x00)))
+			continue;
+
+		ctl = read_pci_config(0, num, 3, 0x90);
+		ctl &= ~1;
+		write_pci_config(0, num, 3, 0x90, ctl);
+	}
+
+}
+
 void __init gart_iommu_hole_init(void)
 {
 	u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0;
diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index abc473bcabe8..07cfaae7ab07 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -728,6 +728,18 @@ void __init finish_e820_parsing(void)
 	}
 }
 
+void __init update_e820(void)
+{
+	u8 nr_map;
+
+	nr_map = e820.nr_map;
+	if (sanitize_e820_map(e820.map, &nr_map))
+		return;
+	e820.nr_map = nr_map;
+	printk(KERN_INFO "modified physical RAM map:\n");
+	e820_print_map("modified");
+}
+
 unsigned long pci_mem_start = 0xaeedbabe;
 EXPORT_SYMBOL(pci_mem_start);
 
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 07547febac7a..12bad27d66f8 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -53,6 +53,7 @@
 #include <video/edid.h>
 #include <asm/e820.h>
 #include <asm/dma.h>
+#include <asm/gart.h>
 #include <asm/mpspec.h>
 #include <asm/mmu_context.h>
 #include <asm/proto.h>
@@ -335,6 +336,8 @@ void __init setup_arch(char **cmdline_p)
 
 	finish_e820_parsing();
 
+	early_gart_iommu_check();
+
 	e820_register_active_regions(0, 0, -1UL);
 	/*
 	 * partially used pages are not usable - thus
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index 8cba49da4795..ff36f434f00b 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -39,6 +39,7 @@ extern void e820_register_active_regions(int nid,
 extern void finish_e820_parsing(void);
 
 extern struct e820map e820;
+extern void update_e820(void);
 
 extern unsigned ebda_addr, ebda_size;
 extern unsigned long nodemap_addr, nodemap_size;
diff --git a/include/asm-x86/gart.h b/include/asm-x86/gart.h
index f704c50519b8..90958ed993fa 100644
--- a/include/asm-x86/gart.h
+++ b/include/asm-x86/gart.h
@@ -9,6 +9,7 @@ extern int iommu_detected;
 extern void gart_iommu_init(void);
 extern void gart_iommu_shutdown(void);
 extern void __init gart_parse_options(char *);
+extern void early_gart_iommu_check(void);
 extern void gart_iommu_hole_init(void);
 extern int fallback_aper_order;
 extern int fallback_aper_force;
@@ -20,6 +21,10 @@ extern int fix_aperture;
 #define gart_iommu_aperture 0
 #define gart_iommu_aperture_allowed 0
 
+static inline void early_gart_iommu_check(void)
+{
+}
+
 static inline void gart_iommu_shutdown(void)
 {
 }
-- 
cgit v1.2.3


From ef97001f3d869d7cc1956e0cc0d89e514e3f7db0 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:10 +0100
Subject: x86: change size of APICIDs from u8 to u16

Change the size of APICIDs from u8 to u16.  This partially
supports the new x2apic mode that will be present on future
processor chips. (Chips actually support 32-bit APICIDs, but that
change is more intrusive. Supporting 16-bit is sufficient for now).

Signed-off-by: Jack Steiner <steiner@sgi.com>

I've included just the partial change from u8 to u16 apicids.  The
remaining x2apic changes will be in a separate patch.

In addition, the fake_node_to_pxm_map[] and fake_apicid_to_node[]
tables have been moved from local data to the __initdata section
reducing stack pressure when MAX_NUMNODES and MAX_LOCAL_APIC are
increased in size.

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/genapic_64.c |  4 ++--
 arch/x86/kernel/mpparse_64.c |  4 ++--
 arch/x86/kernel/smpboot_64.c |  2 +-
 arch/x86/mm/numa_64.c        |  2 +-
 arch/x86/mm/srat_64.c        | 22 +++++++++++++---------
 include/asm-x86/processor.h  | 14 +++++++-------
 include/asm-x86/smp_64.h     |  8 ++++----
 7 files changed, 30 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index ce703e21c912..ac2b78f24074 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -32,10 +32,10 @@
  * array during this time.  Is it zeroed when the per_cpu
  * data area is removed.
  */
-u8 x86_cpu_to_apicid_init[NR_CPUS] __initdata
+u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata
 					= { [0 ... NR_CPUS-1] = BAD_APICID };
 void *x86_cpu_to_apicid_ptr;
-DEFINE_PER_CPU(u8, x86_cpu_to_apicid) = BAD_APICID;
+DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
 struct genapic __read_mostly *genapic = &apic_flat;
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index ef4aab123581..17d21e5b22d6 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -67,7 +67,7 @@ unsigned disabled_cpus __cpuinitdata;
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
 
-u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
+u16 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
 
 
 /*
@@ -132,7 +132,7 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 	 * area is created.
 	 */
 	if (x86_cpu_to_apicid_ptr) {
-		u8 *x86_cpu_to_apicid = (u8 *)x86_cpu_to_apicid_ptr;
+		u16 *x86_cpu_to_apicid = (u16 *)x86_cpu_to_apicid_ptr;
 		x86_cpu_to_apicid[cpu] = m->mpc_apicid;
 	} else {
 		per_cpu(x86_cpu_to_apicid, cpu) = m->mpc_apicid;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 5bd42ce144da..1fea185c9dca 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -65,7 +65,7 @@ int smp_num_siblings = 1;
 EXPORT_SYMBOL(smp_num_siblings);
 
 /* Last level cache ID of each logical CPU */
-DEFINE_PER_CPU(u8, cpu_llc_id) = BAD_APICID;
+DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
 
 /* Bitmask of currently online CPUs */
 cpumask_t cpu_online_map __read_mostly;
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 551e3590e5c5..650001a87c8f 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -626,7 +626,7 @@ void __init init_cpu_to_node(void)
 	int i;
 
 	for (i = 0; i < NR_CPUS; i++) {
-		u8 apicid = x86_cpu_to_apicid_init[i];
+		u16 apicid = x86_cpu_to_apicid_init[i];
 
 		if (apicid == BAD_APICID)
 			continue;
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index 5c0637e4c2f4..b367bc342fe0 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -130,6 +130,9 @@ void __init
 acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 {
 	int pxm, node;
+	int apic_id;
+
+	apic_id = pa->apic_id;
 	if (srat_disabled())
 		return;
 	if (pa->header.length != sizeof(struct acpi_srat_cpu_affinity)) {
@@ -145,10 +148,10 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 		bad_srat();
 		return;
 	}
-	apicid_to_node[pa->apic_id] = node;
+	apicid_to_node[apic_id] = node;
 	acpi_numa = 1;
 	printk(KERN_INFO "SRAT: PXM %u -> APIC %u -> Node %u\n",
-	       pxm, pa->apic_id, node);
+	       pxm, apic_id, node);
 }
 
 int update_end_of_memory(unsigned long end) {return -1;}
@@ -343,7 +346,8 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 	/* First clean up the node list */
 	for (i = 0; i < MAX_NUMNODES; i++) {
 		cutoff_node(i, start, end);
-		if ((nodes[i].end - nodes[i].start) < NODE_MIN_SIZE) {
+		/* ZZZ why was this needed. At least add a comment */
+		if (nodes[i].end && (nodes[i].end - nodes[i].start) < NODE_MIN_SIZE) {
 			unparse_node(i);
 			node_set_offline(i);
 		}
@@ -384,6 +388,12 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 }
 
 #ifdef CONFIG_NUMA_EMU
+static int fake_node_to_pxm_map[MAX_NUMNODES] __initdata = {
+	[0 ... MAX_NUMNODES-1] = PXM_INVAL
+};
+static unsigned char fake_apicid_to_node[MAX_LOCAL_APIC] __initdata = {
+	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
+};
 static int __init find_node_by_addr(unsigned long addr)
 {
 	int ret = NUMA_NO_NODE;
@@ -414,12 +424,6 @@ static int __init find_node_by_addr(unsigned long addr)
 void __init acpi_fake_nodes(const struct bootnode *fake_nodes, int num_nodes)
 {
 	int i, j;
-	int fake_node_to_pxm_map[MAX_NUMNODES] = {
-		[0 ... MAX_NUMNODES-1] = PXM_INVAL
-	};
-	unsigned char fake_apicid_to_node[MAX_LOCAL_APIC] = {
-		[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
-	};
 
 	printk(KERN_INFO "Faking PXM affinity for fake nodes on real "
 			 "topology.\n");
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index e701ac5487e5..81ecfed83e47 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -90,14 +90,14 @@ struct cpuinfo_x86 {
 #ifdef CONFIG_SMP
 	cpumask_t llc_shared_map;	/* cpus sharing the last level cache */
 #endif
-	unsigned char x86_max_cores;	/* cpuid returned max cores value */
-	unsigned char apicid;
-	unsigned short x86_clflush_size;
+	u16 x86_max_cores;		/* cpuid returned max cores value */
+	u16 apicid;
+	u16 x86_clflush_size;
 #ifdef CONFIG_SMP
-	unsigned char booted_cores;	/* number of cores as seen by OS */
-	__u8 phys_proc_id; 		/* Physical processor id. */
-	__u8 cpu_core_id;  		/* Core id */
-	__u8 cpu_index;			/* index into per_cpu list */
+	u16 booted_cores;		/* number of cores as seen by OS */
+	u16 phys_proc_id; 		/* Physical processor id. */
+	u16 cpu_core_id;  		/* Core id */
+	u16 cpu_index;			/* index into per_cpu list */
 #endif
 } __attribute__((__aligned__(SMP_CACHE_BYTES)));
 
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 2feddda91e12..b1d5381aa760 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -26,14 +26,14 @@ extern void unlock_ipi_call_lock(void);
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-extern u8 __initdata x86_cpu_to_apicid_init[];
+extern u16 __initdata x86_cpu_to_apicid_init[];
 extern void *x86_cpu_to_apicid_ptr;
-extern u8 bios_cpu_apicid[];
+extern u16 bios_cpu_apicid[];
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
-DECLARE_PER_CPU(u8, cpu_llc_id);
-DECLARE_PER_CPU(u8, x86_cpu_to_apicid);
+DECLARE_PER_CPU(u16, cpu_llc_id);
+DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
-- 
cgit v1.2.3


From 3cc87e3f4042a099fb649c5df33d3ccfae36173f Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:11 +0100
Subject: x86: change size of node ids from u8 to u16

Change the size of node ids from 8 bits to 16 bits to
accomodate more than 256 nodes.

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/numa_64.c      | 9 ++++++---
 arch/x86/mm/srat_64.c      | 2 +-
 include/asm-x86/numa_64.h  | 4 ++--
 include/asm-x86/topology.h | 2 +-
 4 files changed, 10 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 650001a87c8f..bc6dd5051d60 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -11,6 +11,7 @@
 #include <linux/ctype.h>
 #include <linux/module.h>
 #include <linux/nodemask.h>
+#include <linux/sched.h>
 
 #include <asm/e820.h>
 #include <asm/proto.h>
@@ -30,12 +31,12 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
-int cpu_to_node_map[NR_CPUS] __read_mostly = {
+u16 cpu_to_node_map[NR_CPUS] __read_mostly = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
 EXPORT_SYMBOL(cpu_to_node_map);
 
-unsigned char apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
+u16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
 };
 
@@ -543,7 +544,9 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	node_set(0, node_possible_map);
 	for (i = 0; i < NR_CPUS; i++)
 		numa_set_node(i, 0);
-	node_to_cpumask_map[0] = cpumask_of_cpu(0);
+	/* we can't use cpumask_of_cpu() yet */
+	memset(&node_to_cpumask_map[0], 0, sizeof(node_to_cpumask_map[0]));
+	cpu_set(0, node_to_cpumask_map[0]);
 	e820_register_active_regions(0, start_pfn, end_pfn);
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
 }
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index b367bc342fe0..bccbdc7be434 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -391,7 +391,7 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 static int fake_node_to_pxm_map[MAX_NUMNODES] __initdata = {
 	[0 ... MAX_NUMNODES-1] = PXM_INVAL
 };
-static unsigned char fake_apicid_to_node[MAX_LOCAL_APIC] __initdata = {
+static u16 fake_apicid_to_node[MAX_LOCAL_APIC] __initdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
 };
 static int __init find_node_by_addr(unsigned long addr)
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index 4449889bb0c0..941889471b52 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -20,7 +20,7 @@ extern void numa_set_node(int cpu, int node);
 extern void srat_reserve_add_area(int nodeid);
 extern int hotadd_percent;
 
-extern unsigned char apicid_to_node[MAX_LOCAL_APIC];
+extern u16 apicid_to_node[MAX_LOCAL_APIC];
 
 extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
 extern unsigned long numa_free_all_bootmem(void);
@@ -40,6 +40,6 @@ static inline void clear_node_cpumask(int cpu)
 #define clear_node_cpumask(cpu) do {} while (0)
 #endif
 
-#define NUMA_NO_NODE 0xff
+#define NUMA_NO_NODE 0xffff
 
 #endif
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 9c251604ecc4..f8706b2a3d88 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -30,7 +30,7 @@
 #include <asm/mpspec.h>
 
 /* Mappings between logical cpu number and node number */
-extern int cpu_to_node_map[];
+extern u16 cpu_to_node_map[];
 extern cpumask_t node_to_cpumask_map[];
 
 /* Returns the number of the node containing CPU 'cpu' */
-- 
cgit v1.2.3


From 30964d54e94229f567a7312a0e6666f9deb6a488 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:11 +0100
Subject: x86: change NR_CPUS arrays in topology

Change the following static arrays sized by NR_CPUS to
per_cpu data variables:

	i386_cpu cpu_devices[NR_CPUS];

(And change the struct name to x86_cpu.)

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/topology.c | 8 ++++----
 include/asm-x86/cpu.h      | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/topology.c b/arch/x86/kernel/topology.c
index 7e16d675eb85..a0d1719bda79 100644
--- a/arch/x86/kernel/topology.c
+++ b/arch/x86/kernel/topology.c
@@ -31,7 +31,7 @@
 #include <linux/mmzone.h>
 #include <asm/cpu.h>
 
-static struct i386_cpu cpu_devices[NR_CPUS];
+static DEFINE_PER_CPU(struct x86_cpu, cpu_devices);
 
 int __cpuinit arch_register_cpu(int num)
 {
@@ -46,16 +46,16 @@ int __cpuinit arch_register_cpu(int num)
 	 */
 #ifdef CONFIG_HOTPLUG_CPU
 	if (num)
-		cpu_devices[num].cpu.hotpluggable = 1;
+		per_cpu(cpu_devices, num).cpu.hotpluggable = 1;
 #endif
 
-	return register_cpu(&cpu_devices[num].cpu, num);
+	return register_cpu(&per_cpu(cpu_devices, num).cpu, num);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
 void arch_unregister_cpu(int num)
 {
-	return unregister_cpu(&cpu_devices[num].cpu);
+	return unregister_cpu(&per_cpu(cpu_devices, num).cpu);
 }
 EXPORT_SYMBOL(arch_register_cpu);
 EXPORT_SYMBOL(arch_unregister_cpu);
diff --git a/include/asm-x86/cpu.h b/include/asm-x86/cpu.h
index b1bc7b1b64b0..85ece5f10e9e 100644
--- a/include/asm-x86/cpu.h
+++ b/include/asm-x86/cpu.h
@@ -7,7 +7,7 @@
 #include <linux/nodemask.h>
 #include <linux/percpu.h>
 
-struct i386_cpu {
+struct x86_cpu {
 	struct cpu cpu;
 };
 extern int arch_register_cpu(int num);
-- 
cgit v1.2.3


From 3b41908902df1dba141cd7de1a727bb03718a654 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:11 +0100
Subject: x86: cleanup x86_cpu_to_apicid references

Clean up references to x86_cpu_to_apicid.  Removes extraneous
comments and standardizes on "x86_*_early_ptr" for the early
kernel init references.

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/genapic_64.c | 11 ++---------
 arch/x86/kernel/mpparse_64.c | 11 +++--------
 arch/x86/kernel/setup_64.c   |  2 +-
 arch/x86/kernel/smpboot_32.c |  9 ++-------
 arch/x86/kernel/smpboot_64.c | 16 +++++++++-------
 include/asm-x86/smp_32.h     |  2 +-
 include/asm-x86/smp_64.h     |  2 +-
 7 files changed, 19 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index ac2b78f24074..4ae7b6440260 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -24,17 +24,10 @@
 #include <acpi/acpi_bus.h>
 #endif
 
-/*
- * which logical CPU number maps to which CPU (physical APIC ID)
- *
- * The following static array is used during kernel startup
- * and the x86_cpu_to_apicid_ptr contains the address of the
- * array during this time.  Is it zeroed when the per_cpu
- * data area is removed.
- */
+/* which logical CPU number maps to which CPU (physical APIC ID) */
 u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata
 					= { [0 ... NR_CPUS-1] = BAD_APICID };
-void *x86_cpu_to_apicid_ptr;
+void *x86_cpu_to_apicid_early_ptr;
 DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 17d21e5b22d6..528ad9696d96 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -125,14 +125,9 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 		cpu = 0;
  	}
 	bios_cpu_apicid[cpu] = m->mpc_apicid;
-	/*
-	 * We get called early in the the start_kernel initialization
-	 * process when the per_cpu data area is not yet setup, so we
-	 * use a static array that is removed after the per_cpu data
-	 * area is created.
-	 */
-	if (x86_cpu_to_apicid_ptr) {
-		u16 *x86_cpu_to_apicid = (u16 *)x86_cpu_to_apicid_ptr;
+	/* are we being called early in kernel startup? */
+	if (x86_cpu_to_apicid_early_ptr) {
+		u16 *x86_cpu_to_apicid = (u16 *)x86_cpu_to_apicid_early_ptr;
 		x86_cpu_to_apicid[cpu] = m->mpc_apicid;
 	} else {
 		per_cpu(x86_cpu_to_apicid, cpu) = m->mpc_apicid;
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 12bad27d66f8..e2beb4cba15f 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -360,7 +360,7 @@ void __init setup_arch(char **cmdline_p)
 
 #ifdef CONFIG_SMP
 	/* setup to use the static apicid table during kernel startup */
-	x86_cpu_to_apicid_ptr = (void *)&x86_cpu_to_apicid_init;
+	x86_cpu_to_apicid_early_ptr = (void *)&x86_cpu_to_apicid_init;
 #endif
 
 #ifdef CONFIG_ACPI
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 2034332ad080..915ec6267326 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -91,15 +91,10 @@ static cpumask_t smp_commenced_mask;
 DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
-/*
- * The following static array is used during kernel startup
- * and the x86_cpu_to_apicid_ptr contains the address of the
- * array during this time.  Is it zeroed when the per_cpu
- * data area is removed.
- */
+/* which logical CPU number maps to which CPU (physical APIC ID) */
 u8 x86_cpu_to_apicid_init[NR_CPUS] __initdata =
 			{ [0 ... NR_CPUS-1] = BAD_APICID };
-void *x86_cpu_to_apicid_ptr;
+void *x86_cpu_to_apicid_early_ptr;
 DEFINE_PER_CPU(u8, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 50e207a8261f..a2076b5f12af 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -852,23 +852,25 @@ static int __init smp_sanity_check(unsigned max_cpus)
 }
 
 /*
- * Copy apicid's found by MP_processor_info from initial array to the per cpu
- * data area.  The x86_cpu_to_apicid_init array is then expendable and the
- * x86_cpu_to_apicid_ptr is zeroed indicating that the static array is no
- * longer available.
+ * Copy data used in early init routines from the initial arrays to the
+ * per cpu data areas.  These arrays then become expendable and the
+ * *_ptrs are zeroed indicating that the static arrays are gone.
  */
 void __init smp_set_apicids(void)
 {
 	int cpu;
 
-	for_each_cpu_mask(cpu, cpu_possible_map) {
+	for_each_possible_cpu(cpu) {
 		if (per_cpu_offset(cpu))
 			per_cpu(x86_cpu_to_apicid, cpu) =
 						x86_cpu_to_apicid_init[cpu];
+		else
+			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
+									cpu);
 	}
 
-	/* indicate the static array will be going away soon */
-	x86_cpu_to_apicid_ptr = NULL;
+	/* indicate the early static arrays are gone */
+	x86_cpu_to_apicid_early_ptr = NULL;
 }
 
 static void __init smp_cpu_index_default(void)
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index c69e960429cc..56152e312287 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -30,7 +30,7 @@ extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
 
 extern u8 __initdata x86_cpu_to_apicid_init[];
-extern void *x86_cpu_to_apicid_ptr;
+extern void *x86_cpu_to_apicid_early_ptr;
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index b1d5381aa760..6fa332db29cc 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -27,7 +27,7 @@ extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
 extern u16 __initdata x86_cpu_to_apicid_init[];
-extern void *x86_cpu_to_apicid_ptr;
+extern void *x86_cpu_to_apicid_early_ptr;
 extern u16 bios_cpu_apicid[];
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
-- 
cgit v1.2.3


From df3825c56dd70a4d7796041388f3cfe51c1db832 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:11 +0100
Subject: x86: change NR_CPUS arrays in numa_64

Change the following static arrays sized by NR_CPUS to
per_cpu data variables:

	char cpu_to_node_map[NR_CPUS];

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_64.c   |  4 +++-
 arch/x86/kernel/smpboot_64.c |  6 +++++-
 arch/x86/mm/numa_64.c        | 20 ++++++++++++++++----
 include/asm-x86/numa_64.h    |  2 --
 include/asm-x86/topology.h   | 15 +++++++++++++--
 net/sunrpc/svc.c             |  1 +
 6 files changed, 38 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index e2beb4cba15f..529e45c37b1c 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -65,6 +65,7 @@
 #include <asm/cacheflush.h>
 #include <asm/mce.h>
 #include <asm/ds.h>
+#include <asm/topology.h>
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
@@ -359,8 +360,9 @@ void __init setup_arch(char **cmdline_p)
 	io_delay_init();
 
 #ifdef CONFIG_SMP
-	/* setup to use the static apicid table during kernel startup */
+	/* setup to use the early static init tables during kernel startup */
 	x86_cpu_to_apicid_early_ptr = (void *)&x86_cpu_to_apicid_init;
+	x86_cpu_to_node_map_early_ptr = (void *)&x86_cpu_to_node_map_init;
 #endif
 
 #ifdef CONFIG_ACPI
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index a2076b5f12af..a8bc2bcdb74a 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -861,9 +861,12 @@ void __init smp_set_apicids(void)
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
-		if (per_cpu_offset(cpu))
+		if (per_cpu_offset(cpu)) {
 			per_cpu(x86_cpu_to_apicid, cpu) =
 						x86_cpu_to_apicid_init[cpu];
+			per_cpu(x86_cpu_to_node_map, cpu) =
+						x86_cpu_to_node_map_init[cpu];
+		}
 		else
 			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
 									cpu);
@@ -871,6 +874,7 @@ void __init smp_set_apicids(void)
 
 	/* indicate the early static arrays are gone */
 	x86_cpu_to_apicid_early_ptr = NULL;
+	x86_cpu_to_node_map_early_ptr = NULL;
 }
 
 static void __init smp_cpu_index_default(void)
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index bc6dd5051d60..382377d6421d 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -31,10 +31,14 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
-u16 cpu_to_node_map[NR_CPUS] __read_mostly = {
+u16 x86_cpu_to_node_map_init[NR_CPUS] __initdata = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
-EXPORT_SYMBOL(cpu_to_node_map);
+void *x86_cpu_to_node_map_early_ptr;
+EXPORT_SYMBOL(x86_cpu_to_node_map_init);
+EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
+DEFINE_PER_CPU(u16, x86_cpu_to_node_map) = NUMA_NO_NODE;
+EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map);
 
 u16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
@@ -544,7 +548,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	node_set(0, node_possible_map);
 	for (i = 0; i < NR_CPUS; i++)
 		numa_set_node(i, 0);
-	/* we can't use cpumask_of_cpu() yet */
+	/* cpumask_of_cpu() may not be available during early startup */
 	memset(&node_to_cpumask_map[0], 0, sizeof(node_to_cpumask_map[0]));
 	cpu_set(0, node_to_cpumask_map[0]);
 	e820_register_active_regions(0, start_pfn, end_pfn);
@@ -558,8 +562,16 @@ __cpuinit void numa_add_cpu(int cpu)
 
 void __cpuinit numa_set_node(int cpu, int node)
 {
+	u16 *cpu_to_node_map = (u16 *)x86_cpu_to_node_map_early_ptr;
+
 	cpu_pda(cpu)->nodenumber = node;
-	cpu_to_node_map[cpu] = node;
+
+	if(cpu_to_node_map)
+		cpu_to_node_map[cpu] = node;
+	else if(per_cpu_offset(cpu))
+		per_cpu(x86_cpu_to_node_map, cpu) = node;
+	else
+		Dprintk(KERN_INFO "Setting node for non-present cpu %d\n", cpu);
 }
 
 unsigned long __init numa_free_all_bootmem(void)
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index 941889471b52..c797cd523d2b 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -40,6 +40,4 @@ static inline void clear_node_cpumask(int cpu)
 #define clear_node_cpumask(cpu) do {} while (0)
 #endif
 
-#define NUMA_NO_NODE 0xffff
-
 #endif
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index f8706b2a3d88..e612ed8d2bc7 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -30,13 +30,24 @@
 #include <asm/mpspec.h>
 
 /* Mappings between logical cpu number and node number */
-extern u16 cpu_to_node_map[];
+DECLARE_PER_CPU(u16, x86_cpu_to_node_map);
+extern u16 __initdata x86_cpu_to_node_map_init[];
+extern void *x86_cpu_to_node_map_early_ptr;
 extern cpumask_t node_to_cpumask_map[];
 
+#define NUMA_NO_NODE	((u16)(~0))
+
 /* Returns the number of the node containing CPU 'cpu' */
 static inline int cpu_to_node(int cpu)
 {
-	return cpu_to_node_map[cpu];
+	u16 *cpu_to_node_map = (u16 *)x86_cpu_to_node_map_early_ptr;
+
+	if (cpu_to_node_map)
+		return cpu_to_node_map[cpu];
+	else if (per_cpu_offset(cpu))
+		return per_cpu(x86_cpu_to_node_map, cpu);
+	else
+		return NUMA_NO_NODE;
 }
 
 /*
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index a4a6bf7deaa4..4ad5fbbb18b4 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -18,6 +18,7 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
+#include <linux/sched.h>
 
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/xdr.h>
-- 
cgit v1.2.3


From e8c10ef9dde3ab7b7d7db6804859d9daf38f01c4 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:12 +0100
Subject: x86: change bios_cpu_apicid to percpu data variable

Change static bios_cpu_apicid array to a per_cpu data variable.
This includes using a static array used during initialization
similar to the way x86_cpu_to_apicid[] is handled.

There is one early use of bios_cpu_apicid in apic_is_clustered_box().
The other reference in cpu_present_to_apicid() is called after
smp_set_apicids() has setup the percpu version of bios_cpu_apicid.

[ mingo@elte.hu: build fix ]

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_64.c    | 16 ++++++++++++++--
 arch/x86/kernel/mpparse_64.c | 17 ++++++++++++-----
 arch/x86/kernel/setup_64.c   |  3 +++
 arch/x86/kernel/smpboot_64.c |  7 +++++++
 include/asm-x86/smp_64.h     |  8 +++++---
 5 files changed, 41 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 01d4ca27ecf0..f9919c492699 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -1180,14 +1180,26 @@ __cpuinit int apic_is_clustered_box(void)
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
 	for (i = 0; i < NR_CPUS; i++) {
-		id = bios_cpu_apicid[i];
+		/* are we being called early in kernel startup? */
+		if (x86_bios_cpu_apicid_early_ptr) {
+			id = ((u16 *)x86_bios_cpu_apicid_early_ptr)[i];
+		}
+		else if (i < nr_cpu_ids) {
+			if (cpu_present(i))
+				id = per_cpu(x86_bios_cpu_apicid, i);
+			else
+				continue;
+		}
+		else
+			break;
+
 		if (id != BAD_APICID)
 			__set_bit(APIC_CLUSTERID(id), clustermap);
 	}
 
 	/* Problem:  Partially populated chassis may not have CPUs in some of
 	 * the APIC clusters they have been allocated.  Only present CPUs have
-	 * bios_cpu_apicid entries, thus causing zeroes in the bitmap.  Since
+	 * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap.  Since
 	 * clusters are allocated sequentially, count zeros only if they are
 	 * bounded by ones.
 	 */
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 528ad9696d96..fd671754dcb5 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -67,7 +67,11 @@ unsigned disabled_cpus __cpuinitdata;
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
 
-u16 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
+u16 x86_bios_cpu_apicid_init[NR_CPUS] __initdata
+				= { [0 ... NR_CPUS-1] = BAD_APICID };
+void *x86_bios_cpu_apicid_early_ptr;
+DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID;
+EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid);
 
 
 /*
@@ -118,19 +122,22 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 	physid_set(m->mpc_apicid, phys_cpu_present_map);
  	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
  		/*
- 		 * bios_cpu_apicid is required to have processors listed
+ 		 * x86_bios_cpu_apicid is required to have processors listed
  		 * in same order as logical cpu numbers. Hence the first
  		 * entry is BSP, and so on.
  		 */
 		cpu = 0;
  	}
-	bios_cpu_apicid[cpu] = m->mpc_apicid;
 	/* are we being called early in kernel startup? */
 	if (x86_cpu_to_apicid_early_ptr) {
-		u16 *x86_cpu_to_apicid = (u16 *)x86_cpu_to_apicid_early_ptr;
-		x86_cpu_to_apicid[cpu] = m->mpc_apicid;
+		u16 *cpu_to_apicid = (u16 *)x86_cpu_to_apicid_early_ptr;
+		u16 *bios_cpu_apicid = (u16 *)x86_bios_cpu_apicid_early_ptr;
+
+		cpu_to_apicid[cpu] = m->mpc_apicid;
+		bios_cpu_apicid[cpu] = m->mpc_apicid;
 	} else {
 		per_cpu(x86_cpu_to_apicid, cpu) = m->mpc_apicid;
+		per_cpu(x86_bios_cpu_apicid, cpu) = m->mpc_apicid;
 	}
 
 	cpu_set(cpu, cpu_possible_map);
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 529e45c37b1c..71a420c7fee7 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -362,8 +362,11 @@ void __init setup_arch(char **cmdline_p)
 #ifdef CONFIG_SMP
 	/* setup to use the early static init tables during kernel startup */
 	x86_cpu_to_apicid_early_ptr = (void *)&x86_cpu_to_apicid_init;
+#ifdef CONFIG_NUMA
 	x86_cpu_to_node_map_early_ptr = (void *)&x86_cpu_to_node_map_init;
 #endif
+	x86_bios_cpu_apicid_early_ptr = (void *)&x86_bios_cpu_apicid_init;
+#endif
 
 #ifdef CONFIG_ACPI
 	/*
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index a8bc2bcdb74a..93071cdf0849 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -864,8 +864,12 @@ void __init smp_set_apicids(void)
 		if (per_cpu_offset(cpu)) {
 			per_cpu(x86_cpu_to_apicid, cpu) =
 						x86_cpu_to_apicid_init[cpu];
+#ifdef CONFIG_NUMA
 			per_cpu(x86_cpu_to_node_map, cpu) =
 						x86_cpu_to_node_map_init[cpu];
+#endif
+			per_cpu(x86_bios_cpu_apicid, cpu) =
+						x86_bios_cpu_apicid_init[cpu];
 		}
 		else
 			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
@@ -874,7 +878,10 @@ void __init smp_set_apicids(void)
 
 	/* indicate the early static arrays are gone */
 	x86_cpu_to_apicid_early_ptr = NULL;
+#ifdef CONFIG_NUMA
 	x86_cpu_to_node_map_early_ptr = NULL;
+#endif
+	x86_bios_cpu_apicid_early_ptr = NULL;
 }
 
 static void __init smp_cpu_index_default(void)
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 6fa332db29cc..e0a75519ad21 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -27,18 +27,20 @@ extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
 extern u16 __initdata x86_cpu_to_apicid_init[];
+extern u16 __initdata x86_bios_cpu_apicid_init[];
 extern void *x86_cpu_to_apicid_early_ptr;
-extern u16 bios_cpu_apicid[];
+extern void *x86_bios_cpu_apicid_early_ptr;
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
 DECLARE_PER_CPU(u16, cpu_llc_id);
 DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
+DECLARE_PER_CPU(u16, x86_bios_cpu_apicid);
 
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
-	if (mps_cpu < NR_CPUS)
-		return (int)bios_cpu_apicid[mps_cpu];
+	if (cpu_present(mps_cpu))
+		return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
 	else
 		return BAD_APICID;
 }
-- 
cgit v1.2.3


From 37cd9cf3dafed82f7cf905785883300f6ff7c818 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:12 +0100
Subject: x86: common x86_32|64 naming

Rename convert_rip_to_linear to convert_ip_to_linear for shared
X86_32|64 use.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/step.c   | 4 ++--
 arch/x86/mm/fault_32.c   | 2 +-
 arch/x86/mm/fault_64.c   | 2 +-
 include/asm-x86/ptrace.h | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index b801e76cebf7..80b37181a42b 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -89,7 +89,7 @@ unsigned long get_segment_eip(struct pt_regs *regs,
 #ifdef CONFIG_X86_32
 static
 #endif
-unsigned long convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs)
+unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs)
 {
 	unsigned long addr, seg;
 
@@ -136,7 +136,7 @@ static int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
 {
 	int i, copied;
 	unsigned char opcode[15];
-	unsigned long addr = convert_rip_to_linear(child, regs);
+	unsigned long addr = convert_ip_to_linear(child, regs);
 
 	copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0);
 	for (i = 0; i < copied; i++) {
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index b92922a1d65f..b4d19c2d4f05 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -95,7 +95,7 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	/* If it was a exec fault ignore */
 	if (error_code & PF_INSTR)
 		return 0;
-	instr = (unsigned char __user *)convert_rip_to_linear(current, regs);
+	instr = (unsigned char __user *)convert_ip_to_linear(current, regs);
 #endif
 
 	max_instr = instr + 15;
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index cf7e99895b91..d519b41f1962 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -98,7 +98,7 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	/* If it was a exec fault ignore */
 	if (error_code & PF_INSTR)
 		return 0;
-	instr = (unsigned char __user *)convert_rip_to_linear(current, regs);
+	instr = (unsigned char __user *)convert_ip_to_linear(current, regs);
 #endif
 
 	max_instr = instr + 15;
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index cc4456667d89..35c103714906 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -177,7 +177,7 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
 struct task_struct;
 
 extern unsigned long
-convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
+convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 #endif /* __KERNEL__ */
 #endif /* !__i386__ */
-- 
cgit v1.2.3


From f2857ce92023409df1544737d5b3499b4630a183 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:12 +0100
Subject: x86: remove last user of get_segment_eip

is_prefetch was the last user of get_segment_eip and only on
X86_32.  This function returned the faulting instruction's
address and set the upper segment limit.

Instead, use the convert_ip_to_linear helper and rely on
probe_kernel_address to do the segment checks which was
already done everywhere the segment limit was being checked
on X86_32.

Remove get_segment_eip as well.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/step.c   | 84 ------------------------------------------------
 arch/x86/mm/fault_32.c   | 15 ++-------
 arch/x86/mm/fault_64.c   | 15 ++-------
 include/asm-x86/ptrace.h |  6 ++--
 4 files changed, 8 insertions(+), 112 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 80b37181a42b..2ef1a5f8d675 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -5,90 +5,6 @@
 #include <linux/mm.h>
 #include <linux/ptrace.h>
 
-#ifdef CONFIG_X86_32
-#include <linux/uaccess.h>
-
-#include <asm/desc.h>
-
-/*
- * Return EIP plus the CS segment base.  The segment limit is also
- * adjusted, clamped to the kernel/user address space (whichever is
- * appropriate), and returned in *eip_limit.
- *
- * The segment is checked, because it might have been changed by another
- * task between the original faulting instruction and here.
- *
- * If CS is no longer a valid code segment, or if EIP is beyond the
- * limit, or if it is a kernel address when CS is not a kernel segment,
- * then the returned value will be greater than *eip_limit.
- *
- * This is slow, but is very rarely executed.
- */
-unsigned long get_segment_eip(struct pt_regs *regs,
-					    unsigned long *eip_limit)
-{
-	unsigned long ip = regs->ip;
-	unsigned seg = regs->cs & 0xffff;
-	u32 seg_ar, seg_limit, base, *desc;
-
-	/* Unlikely, but must come before segment checks. */
-	if (unlikely(regs->flags & VM_MASK)) {
-		base = seg << 4;
-		*eip_limit = base + 0xffff;
-		return base + (ip & 0xffff);
-	}
-
-	/* The standard kernel/user address space limit. */
-	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
-
-	/* By far the most common cases. */
-	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
-		return ip;
-
-	/* Check the segment exists, is within the current LDT/GDT size,
-	   that kernel/user (ring 0..3) has the appropriate privilege,
-	   that it's a code segment, and get the limit. */
-	__asm__("larl %3,%0; lsll %3,%1"
-		 : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
-	if ((~seg_ar & 0x9800) || ip > seg_limit) {
-		*eip_limit = 0;
-		return 1;	 /* So that returned ip > *eip_limit. */
-	}
-
-	/* Get the GDT/LDT descriptor base.
-	   When you look for races in this code remember that
-	   LDT and other horrors are only used in user space. */
-	if (seg & (1<<2)) {
-		/* Must lock the LDT while reading it. */
-		mutex_lock(&current->mm->context.lock);
-		desc = current->mm->context.ldt;
-		desc = (void *)desc + (seg & ~7);
-	} else {
-		/* Must disable preemption while reading the GDT. */
-		desc = (u32 *)get_cpu_gdt_table(get_cpu());
-		desc = (void *)desc + (seg & ~7);
-	}
-
-	/* Decode the code segment base from the descriptor */
-	base = get_desc_base((struct desc_struct *)desc);
-
-	if (seg & (1<<2))
-		mutex_unlock(&current->mm->context.lock);
-	else
-		put_cpu();
-
-	/* Adjust EIP and segment limit, and clamp at the kernel limit.
-	   It's legitimate for segments to wrap at 0xffffffff. */
-	seg_limit += base;
-	if (seg_limit < *eip_limit && seg_limit >= base)
-		*eip_limit = seg_limit;
-	return ip + base;
-}
-#endif
-
-#ifdef CONFIG_X86_32
-static
-#endif
 unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs)
 {
 	unsigned long addr, seg;
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index b4d19c2d4f05..36cb67e02b04 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -81,7 +81,6 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	unsigned char *max_instr;
 
 #ifdef CONFIG_X86_32
-	unsigned long limit;
 	if (unlikely(boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
 		     boot_cpu_data.x86 >= 6)) {
 		/* Catch an obscure case of prefetch inside an NX page. */
@@ -90,30 +89,23 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	} else {
 		return 0;
 	}
-	instr = (unsigned char *)get_segment_eip(regs, &limit);
 #else
 	/* If it was a exec fault ignore */
 	if (error_code & PF_INSTR)
 		return 0;
-	instr = (unsigned char __user *)convert_ip_to_linear(current, regs);
 #endif
 
+	instr = (unsigned char *)convert_ip_to_linear(current, regs);
 	max_instr = instr + 15;
 
-#ifdef CONFIG_X86_64
 	if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
 		return 0;
-#endif
 
 	while (scan_more && instr < max_instr) {
 		unsigned char opcode;
 		unsigned char instr_hi;
 		unsigned char instr_lo;
 
-#ifdef CONFIG_X86_32
-		if (instr > (unsigned char *)limit)
-			break;
-#endif
 		if (probe_kernel_address(instr, opcode))
 			break;
 
@@ -155,10 +147,7 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 		case 0x00:
 			/* Prefetch instruction is 0x0F0D or 0x0F18 */
 			scan_more = 0;
-#ifdef CONFIG_X86_32
-			if (instr > (unsigned char *)limit)
-				break;
-#endif
+
 			if (probe_kernel_address(instr, opcode))
 				break;
 			prefetch = (instr_lo == 0xF) &&
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index d519b41f1962..80f8436ac8b2 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -84,7 +84,6 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	unsigned char *max_instr;
 
 #ifdef CONFIG_X86_32
-	unsigned long limit;
 	if (unlikely(boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
 		     boot_cpu_data.x86 >= 6)) {
 		/* Catch an obscure case of prefetch inside an NX page. */
@@ -93,30 +92,23 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	} else {
 		return 0;
 	}
-	instr = (unsigned char *)get_segment_eip(regs, &limit);
 #else
 	/* If it was a exec fault ignore */
 	if (error_code & PF_INSTR)
 		return 0;
-	instr = (unsigned char __user *)convert_ip_to_linear(current, regs);
 #endif
 
+	instr = (unsigned char *)convert_ip_to_linear(current, regs);
 	max_instr = instr + 15;
 
-#ifdef CONFIG_X86_64
 	if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
 		return 0;
-#endif
 
 	while (scan_more && instr < max_instr) {
 		unsigned char opcode;
 		unsigned char instr_hi;
 		unsigned char instr_lo;
 
-#ifdef CONFIG_X86_32
-		if (instr > (unsigned char *)limit)
-			break;
-#endif
 		if (probe_kernel_address(instr, opcode))
 			break;
 
@@ -158,10 +150,7 @@ static int is_prefetch(struct pt_regs *regs, unsigned long addr,
 		case 0x00:
 			/* Prefetch instruction is 0x0F0D or 0x0F18 */
 			scan_more = 0;
-#ifdef CONFIG_X86_32
-			if (instr > (unsigned char *)limit)
-				break;
-#endif
+
 			if (probe_kernel_address(instr, opcode))
 				break;
 			prefetch = (instr_lo == 0xF) &&
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 35c103714906..ee4b595e1ccc 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -70,6 +70,10 @@ struct pt_regs {
 #include <asm/segment.h>
 
 struct task_struct;
+
+extern unsigned long
+convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
+
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
 
 /*
@@ -184,8 +188,6 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 #ifdef __KERNEL__
 
-unsigned long get_segment_eip(struct pt_regs *regs, unsigned long *eip_limit);
-
 /*
  * These are defined as per linux/ptrace.h, which see.
  */
-- 
cgit v1.2.3


From 6b8be6df7f971919622d152d144c8798ad7fd160 Mon Sep 17 00:00:00 2001
From: John Reiser <jreiser@BitWagon.com>
Date: Wed, 30 Jan 2008 13:33:13 +0100
Subject: x86: add ENDPROC() markers

The ENDPROCs() were not used everywhere.  Some code used just END() instead,
while other code used nothing.  um/sys-i386/checksum.S didn't #include
<linux/linkage.h> .  I also got confused because gcc puts the
.type near the ENTRY, while ENDPROC puts it on the opposite end.

Signed off by: John Reiser <jreiser@BitWagon.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/lib/semaphore_32.S | 20 ++++++++++----------
 include/linux/linkage.h     |  4 ++++
 2 files changed, 14 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/lib/semaphore_32.S b/arch/x86/lib/semaphore_32.S
index 444fba400983..e2c6e0d5e125 100644
--- a/arch/x86/lib/semaphore_32.S
+++ b/arch/x86/lib/semaphore_32.S
@@ -49,7 +49,7 @@ ENTRY(__down_failed)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__down_failed)
+	ENDPROC(__down_failed)
 
 ENTRY(__down_failed_interruptible)
 	CFI_STARTPROC
@@ -70,7 +70,7 @@ ENTRY(__down_failed_interruptible)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__down_failed_interruptible)
+	ENDPROC(__down_failed_interruptible)
 
 ENTRY(__down_failed_trylock)
 	CFI_STARTPROC
@@ -91,7 +91,7 @@ ENTRY(__down_failed_trylock)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__down_failed_trylock)
+	ENDPROC(__down_failed_trylock)
 
 ENTRY(__up_wakeup)
 	CFI_STARTPROC
@@ -112,7 +112,7 @@ ENTRY(__up_wakeup)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__up_wakeup)
+	ENDPROC(__up_wakeup)
 
 /*
  * rw spinlock fallbacks
@@ -132,7 +132,7 @@ ENTRY(__write_lock_failed)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__write_lock_failed)
+	ENDPROC(__write_lock_failed)
 
 ENTRY(__read_lock_failed)
 	CFI_STARTPROC
@@ -148,7 +148,7 @@ ENTRY(__read_lock_failed)
 	ENDFRAME
 	ret
 	CFI_ENDPROC
-	END(__read_lock_failed)
+	ENDPROC(__read_lock_failed)
 
 #endif
 
@@ -170,7 +170,7 @@ ENTRY(call_rwsem_down_read_failed)
 	CFI_ADJUST_CFA_OFFSET -4
 	ret
 	CFI_ENDPROC
-	END(call_rwsem_down_read_failed)
+	ENDPROC(call_rwsem_down_read_failed)
 
 ENTRY(call_rwsem_down_write_failed)
 	CFI_STARTPROC
@@ -182,7 +182,7 @@ ENTRY(call_rwsem_down_write_failed)
 	CFI_ADJUST_CFA_OFFSET -4
 	ret
 	CFI_ENDPROC
-	END(call_rwsem_down_write_failed)
+	ENDPROC(call_rwsem_down_write_failed)
 
 ENTRY(call_rwsem_wake)
 	CFI_STARTPROC
@@ -196,7 +196,7 @@ ENTRY(call_rwsem_wake)
 	CFI_ADJUST_CFA_OFFSET -4
 1:	ret
 	CFI_ENDPROC
-	END(call_rwsem_wake)
+	ENDPROC(call_rwsem_wake)
 
 /* Fix up special calling conventions */
 ENTRY(call_rwsem_downgrade_wake)
@@ -214,6 +214,6 @@ ENTRY(call_rwsem_downgrade_wake)
 	CFI_ADJUST_CFA_OFFSET -4
 	ret
 	CFI_ENDPROC
-	END(call_rwsem_downgrade_wake)
+	ENDPROC(call_rwsem_downgrade_wake)
 
 #endif
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index ceeeb5ea5b15..3faf599ea58e 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -57,6 +57,10 @@
   .size name, .-name
 #endif
 
+/* If symbol 'name' is treated as a subroutine (gets called, and returns)
+ * then please use ENDPROC to mark 'name' as STT_FUNC for the benefit of
+ * static analysis tools such as stack depth analyzer.
+ */
 #ifndef ENDPROC
 #define ENDPROC(name) \
   .type name, @function; \
-- 
cgit v1.2.3


From 466eed22d127a1f16e1251cdc54a9f8f944140c0 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Wed, 30 Jan 2008 13:33:14 +0100
Subject: x86: isolate PIC/PIT in/out calls

Rather than remove and/or mangle inb_p/outb_p we want to remove the use
of them from inappropriate places. For the PIC/PIT this may eventually
depend on 32/64bitism or similar so start by adding inb/outb_pit and
inb/outb_pic so that we can make them use any scheme we settle on without
disturbing the existing, correct (for ISA), port 0x80 usage. (eg we can
make inb_pit use udelay without messing up inb_p).

Floppy already does this for the fdc. That really only leaves the CMOS as
a core logic item to tackle, and bits of parallel port handling in the
chipset layers.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apm_32.c      |  6 +++---
 arch/x86/kernel/i8253.c       | 30 +++++++++++++++---------------
 arch/x86/kernel/i8259_32.c    | 20 ++++++++++----------
 arch/x86/kernel/i8259_64.c    | 20 ++++++++++----------
 arch/x86/kernel/vmiclock_32.c |  2 +-
 include/asm-x86/i8253.h       |  3 +++
 include/asm-x86/i8259.h       |  3 +++
 7 files changed, 45 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 955dd43b1676..d4438ef296d8 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -1174,11 +1174,11 @@ static void reinit_timer(void)
 
 	spin_lock_irqsave(&i8253_lock, flags);
 	/* set the clock to HZ */
-	outb_p(0x34, PIT_MODE);		/* binary, mode 2, LSB/MSB, ch 0 */
+	outb_pit(0x34, PIT_MODE);		/* binary, mode 2, LSB/MSB, ch 0 */
 	udelay(10);
-	outb_p(LATCH & 0xff, PIT_CH0);	/* LSB */
+	outb_pit(LATCH & 0xff, PIT_CH0);	/* LSB */
 	udelay(10);
-	outb(LATCH >> 8, PIT_CH0);	/* MSB */
+	outb_pit(LATCH >> 8, PIT_CH0);	/* MSB */
 	udelay(10);
 	spin_unlock_irqrestore(&i8253_lock, flags);
 #endif
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index c76fef1ce355..ef62b07b2b48 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -43,26 +43,26 @@ static void init_pit_timer(enum clock_event_mode mode,
 	switch(mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 		/* binary, mode 2, LSB/MSB, ch 0 */
-		outb_p(0x34, PIT_MODE);
-		outb_p(LATCH & 0xff , PIT_CH0);	/* LSB */
-		outb(LATCH >> 8 , PIT_CH0);	/* MSB */
+		outb_pit(0x34, PIT_MODE);
+		outb_pit(LATCH & 0xff , PIT_CH0);	/* LSB */
+		outb_pit(LATCH >> 8 , PIT_CH0);		/* MSB */
 		break;
 
 	case CLOCK_EVT_MODE_SHUTDOWN:
 	case CLOCK_EVT_MODE_UNUSED:
 		if (evt->mode == CLOCK_EVT_MODE_PERIODIC ||
 		    evt->mode == CLOCK_EVT_MODE_ONESHOT) {
-			outb_p(0x30, PIT_MODE);
-			outb_p(0, PIT_CH0);
-			outb_p(0, PIT_CH0);
+			outb_pit(0x30, PIT_MODE);
+			outb_pit(0, PIT_CH0);
+			outb_pit(0, PIT_CH0);
 		}
 		pit_disable_clocksource();
 		break;
 
 	case CLOCK_EVT_MODE_ONESHOT:
 		/* One shot setup */
-		outb_p(0x38, PIT_MODE);
 		pit_disable_clocksource();
+		outb_pit(0x38, PIT_MODE);
 		break;
 
 	case CLOCK_EVT_MODE_RESUME:
@@ -80,8 +80,8 @@ static void init_pit_timer(enum clock_event_mode mode,
 static int pit_next_event(unsigned long delta, struct clock_event_device *evt)
 {
 	spin_lock(&i8253_lock);
-	outb_p(delta & 0xff , PIT_CH0);	/* LSB */
-	outb(delta >> 8 , PIT_CH0);	/* MSB */
+	outb_pit(delta & 0xff , PIT_CH0);	/* LSB */
+	outb_pit(delta >> 8 , PIT_CH0);		/* MSB */
 	spin_unlock(&i8253_lock);
 
 	return 0;
@@ -153,15 +153,15 @@ static cycle_t pit_read(void)
 	 * count), it cannot be newer.
 	 */
 	jifs = jiffies;
-	outb_p(0x00, PIT_MODE);	/* latch the count ASAP */
-	count = inb_p(PIT_CH0);	/* read the latched count */
-	count |= inb_p(PIT_CH0) << 8;
+	outb_pit(0x00, PIT_MODE);	/* latch the count ASAP */
+	count = inb_pit(PIT_CH0);	/* read the latched count */
+	count |= inb_pit(PIT_CH0) << 8;
 
 	/* VIA686a test code... reset the latch if count > max + 1 */
 	if (count > LATCH) {
-		outb_p(0x34, PIT_MODE);
-		outb_p(LATCH & 0xff, PIT_CH0);
-		outb(LATCH >> 8, PIT_CH0);
+		outb_pit(0x34, PIT_MODE);
+		outb_pit(LATCH & 0xff, PIT_CH0);
+		outb_pit(LATCH >> 8, PIT_CH0);
 		count = LATCH - 1;
 	}
 
diff --git a/arch/x86/kernel/i8259_32.c b/arch/x86/kernel/i8259_32.c
index f201e7da1bbc..2d25b77102fe 100644
--- a/arch/x86/kernel/i8259_32.c
+++ b/arch/x86/kernel/i8259_32.c
@@ -289,20 +289,20 @@ void init_8259A(int auto_eoi)
 	outb(0xff, PIC_SLAVE_IMR);	/* mask all of 8259A-2 */
 
 	/*
-	 * outb_p - this has to work on a wide range of PC hardware.
+	 * outb_pic - this has to work on a wide range of PC hardware.
 	 */
-	outb_p(0x11, PIC_MASTER_CMD);	/* ICW1: select 8259A-1 init */
-	outb_p(0x20 + 0, PIC_MASTER_IMR);	/* ICW2: 8259A-1 IR0-7 mapped to 0x20-0x27 */
-	outb_p(1U << PIC_CASCADE_IR, PIC_MASTER_IMR);	/* 8259A-1 (the master) has a slave on IR2 */
+	outb_pic(0x11, PIC_MASTER_CMD);	/* ICW1: select 8259A-1 init */
+	outb_pic(0x20 + 0, PIC_MASTER_IMR);	/* ICW2: 8259A-1 IR0-7 mapped to 0x20-0x27 */
+	outb_pic(1U << PIC_CASCADE_IR, PIC_MASTER_IMR);	/* 8259A-1 (the master) has a slave on IR2 */
 	if (auto_eoi)	/* master does Auto EOI */
-		outb_p(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR);
+		outb_pic(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR);
 	else		/* master expects normal EOI */
-		outb_p(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR);
+		outb_pic(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR);
 
-	outb_p(0x11, PIC_SLAVE_CMD);	/* ICW1: select 8259A-2 init */
-	outb_p(0x20 + 8, PIC_SLAVE_IMR);	/* ICW2: 8259A-2 IR0-7 mapped to 0x28-0x2f */
-	outb_p(PIC_CASCADE_IR, PIC_SLAVE_IMR);	/* 8259A-2 is a slave on master's IR2 */
-	outb_p(SLAVE_ICW4_DEFAULT, PIC_SLAVE_IMR); /* (slave's support for AEOI in flat mode is to be investigated) */
+	outb_pic(0x11, PIC_SLAVE_CMD);	/* ICW1: select 8259A-2 init */
+	outb_pic(0x20 + 8, PIC_SLAVE_IMR);	/* ICW2: 8259A-2 IR0-7 mapped to 0x28-0x2f */
+	outb_pic(PIC_CASCADE_IR, PIC_SLAVE_IMR);	/* 8259A-2 is a slave on master's IR2 */
+	outb_pic(SLAVE_ICW4_DEFAULT, PIC_SLAVE_IMR); /* (slave's support for AEOI in flat mode is to be investigated) */
 	if (auto_eoi)
 		/*
 		 * In AEOI mode we just have to mask the interrupt
diff --git a/arch/x86/kernel/i8259_64.c b/arch/x86/kernel/i8259_64.c
index 99c8406ae253..d3edb9f23f2c 100644
--- a/arch/x86/kernel/i8259_64.c
+++ b/arch/x86/kernel/i8259_64.c
@@ -359,25 +359,25 @@ void init_8259A(int auto_eoi)
 	outb(0xff, PIC_SLAVE_IMR);	/* mask all of 8259A-2 */
 
 	/*
-	 * outb_p - this has to work on a wide range of PC hardware.
+	 * outb_pic - this has to work on a wide range of PC hardware.
 	 */
-	outb_p(0x11, PIC_MASTER_CMD);	/* ICW1: select 8259A-1 init */
+	outb_pic(0x11, PIC_MASTER_CMD);	/* ICW1: select 8259A-1 init */
 	/* ICW2: 8259A-1 IR0-7 mapped to 0x30-0x37 */
-	outb_p(IRQ0_VECTOR, PIC_MASTER_IMR);
+	outb_pic(IRQ0_VECTOR, PIC_MASTER_IMR);
 	/* 8259A-1 (the master) has a slave on IR2 */
-	outb_p(0x04, PIC_MASTER_IMR);
+	outb_pic(0x04, PIC_MASTER_IMR);
 	if (auto_eoi)	/* master does Auto EOI */
-		outb_p(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR);
+		outb_pic(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR);
 	else		/* master expects normal EOI */
-		outb_p(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR);
+		outb_pic(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR);
 
-	outb_p(0x11, PIC_SLAVE_CMD);	/* ICW1: select 8259A-2 init */
+	outb_pic(0x11, PIC_SLAVE_CMD);	/* ICW1: select 8259A-2 init */
 	/* ICW2: 8259A-2 IR0-7 mapped to 0x38-0x3f */
-	outb_p(IRQ8_VECTOR, PIC_SLAVE_IMR);
+	outb_pic(IRQ8_VECTOR, PIC_SLAVE_IMR);
 	/* 8259A-2 is a slave on master's IR2 */
-	outb_p(PIC_CASCADE_IR, PIC_SLAVE_IMR);
+	outb_pic(PIC_CASCADE_IR, PIC_SLAVE_IMR);
 	/* (slave's support for AEOI in flat mode is to be investigated) */
-	outb_p(SLAVE_ICW4_DEFAULT, PIC_SLAVE_IMR);
+	outb_pic(SLAVE_ICW4_DEFAULT, PIC_SLAVE_IMR);
 
 	if (auto_eoi)
 		/*
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 57f9ef5a324c..a2b030780aa9 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -237,7 +237,7 @@ static void __devinit vmi_time_init_clockevent(void)
 void __init vmi_time_init(void)
 {
 	/* Disable PIT: BIOSes start PIT CH0 with 18.2hz peridic. */
-	outb_p(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */
+	outb_pit(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */
 
 	vmi_time_init_clockevent();
 	setup_irq(0, &vmi_clock_action);
diff --git a/include/asm-x86/i8253.h b/include/asm-x86/i8253.h
index 747548ec5d1d..b51c0487fc41 100644
--- a/include/asm-x86/i8253.h
+++ b/include/asm-x86/i8253.h
@@ -12,4 +12,7 @@ extern struct clock_event_device *global_clock_event;
 
 extern void setup_pit_timer(void);
 
+#define inb_pit		inb_p
+#define outb_pit	outb_p
+
 #endif	/* __ASM_I8253_H__ */
diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h
index cabcc6cf3923..67c319e0efc7 100644
--- a/include/asm-x86/i8259.h
+++ b/include/asm-x86/i8259.h
@@ -29,4 +29,7 @@ extern void enable_8259A_irq(unsigned int irq);
 extern void disable_8259A_irq(unsigned int irq);
 extern unsigned int startup_8259A_irq(unsigned int irq);
 
+#define inb_pic		inb_p
+#define outb_pic	outb_p
+
 #endif	/* __ASM_I8259_H__ */
-- 
cgit v1.2.3


From 693e3c560317577a29c625d89f6745d5c7cfd918 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:14 +0100
Subject: x86: reduce memory and intra-node effects

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic_64.c    | 5 +++--
 arch/x86/kernel/mpparse_64.c | 6 +++---
 arch/x86/mm/numa_64.c        | 2 +-
 arch/x86/mm/srat_64.c        | 8 ++++++--
 include/asm-x86/topology.h   | 2 +-
 5 files changed, 14 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index f9919c492699..85bd3d463cdf 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -1175,14 +1175,15 @@ __cpuinit int apic_is_clustered_box(void)
 {
 	int i, clusters, zeros;
 	unsigned id;
+	u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
 	DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS);
 
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
 	for (i = 0; i < NR_CPUS; i++) {
 		/* are we being called early in kernel startup? */
-		if (x86_bios_cpu_apicid_early_ptr) {
-			id = ((u16 *)x86_bios_cpu_apicid_early_ptr)[i];
+		if (bios_cpu_apicid) {
+			id = bios_cpu_apicid[i];
 		}
 		else if (i < nr_cpu_ids) {
 			if (cpu_present(i))
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index fd671754dcb5..d3260f8f17dc 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -122,7 +122,7 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 	physid_set(m->mpc_apicid, phys_cpu_present_map);
  	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
  		/*
- 		 * x86_bios_cpu_apicid is required to have processors listed
+		 * x86_bios_cpu_apicid is required to have processors listed
  		 * in same order as logical cpu numbers. Hence the first
  		 * entry is BSP, and so on.
  		 */
@@ -130,8 +130,8 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
  	}
 	/* are we being called early in kernel startup? */
 	if (x86_cpu_to_apicid_early_ptr) {
-		u16 *cpu_to_apicid = (u16 *)x86_cpu_to_apicid_early_ptr;
-		u16 *bios_cpu_apicid = (u16 *)x86_bios_cpu_apicid_early_ptr;
+		u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr;
+		u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
 
 		cpu_to_apicid[cpu] = m->mpc_apicid;
 		bios_cpu_apicid[cpu] = m->mpc_apicid;
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 382377d6421d..dca58fb39b08 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -562,7 +562,7 @@ __cpuinit void numa_add_cpu(int cpu)
 
 void __cpuinit numa_set_node(int cpu, int node)
 {
-	u16 *cpu_to_node_map = (u16 *)x86_cpu_to_node_map_early_ptr;
+	u16 *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
 	cpu_pda(cpu)->nodenumber = node;
 
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index bccbdc7be434..e5a1ec8342dc 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -346,8 +346,12 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 	/* First clean up the node list */
 	for (i = 0; i < MAX_NUMNODES; i++) {
 		cutoff_node(i, start, end);
-		/* ZZZ why was this needed. At least add a comment */
-		if (nodes[i].end && (nodes[i].end - nodes[i].start) < NODE_MIN_SIZE) {
+		/*
+		 * don't confuse VM with a node that doesn't have the
+		 * minimum memory.
+		 */
+		if (nodes[i].end &&
+			(nodes[i].end - nodes[i].start) < NODE_MIN_SIZE) {
 			unparse_node(i);
 			node_set_offline(i);
 		}
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index e612ed8d2bc7..a89b46eb2526 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -40,7 +40,7 @@ extern cpumask_t node_to_cpumask_map[];
 /* Returns the number of the node containing CPU 'cpu' */
 static inline int cpu_to_node(int cpu)
 {
-	u16 *cpu_to_node_map = (u16 *)x86_cpu_to_node_map_early_ptr;
+	u16 *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
 	if (cpu_to_node_map)
 		return cpu_to_node_map[cpu];
-- 
cgit v1.2.3


From 118c890961a2ca456947d72ebe0ce43de07907f0 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 30 Jan 2008 13:33:14 +0100
Subject: arch/x86/mm/numa_64.c: section fix

WARNING: vmlinux.o(__ksymtab+0x670): Section mismatch: reference to .init.data:x86_cpu_to_node_map_init (between '__ksymtab_x86_cpu_to_node_map_init' and '__ksymtab_node_data')

Cc: Matthew Dobson <colpatch@us.ibm.com>
Cc: Mike Travis <travis@sgi.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/numa_64.c      | 2 +-
 include/asm-x86/topology.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index dca58fb39b08..7b03d2090c28 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -31,7 +31,7 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
-u16 x86_cpu_to_node_map_init[NR_CPUS] __initdata = {
+u16 x86_cpu_to_node_map_init[NR_CPUS] = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
 void *x86_cpu_to_node_map_early_ptr;
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index a89b46eb2526..2da1464ecbef 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -31,7 +31,7 @@
 
 /* Mappings between logical cpu number and node number */
 DECLARE_PER_CPU(u16, x86_cpu_to_node_map);
-extern u16 __initdata x86_cpu_to_node_map_init[];
+extern u16 x86_cpu_to_node_map_init[];
 extern void *x86_cpu_to_node_map_early_ptr;
 extern cpumask_t node_to_cpumask_map[];
 
-- 
cgit v1.2.3


From 2f98b2faac1b5fec327edbde945cdf7f7a53f351 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:14 +0100
Subject: add native_pud_val and _pmd_val for 2 and 3

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/page.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index a6495eb5c605..eba88d940092 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -91,6 +91,11 @@ static inline pudval_t native_pud_val(pud_t pud)
 }
 #else	/* PAGETABLE_LEVELS == 3 */
 #include <asm-generic/pgtable-nopud.h>
+
+static inline pudval_t native_pud_val(pud_t pud)
+{
+	return native_pgd_val(pud.pgd);
+}
 #endif	/* PAGETABLE_LEVELS == 4 */
 
 typedef struct { pmdval_t pmd; } pmd_t;
@@ -106,6 +111,11 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
 }
 #else  /* PAGETABLE_LEVELS == 2 */
 #include <asm-generic/pgtable-nopmd.h>
+
+static inline pmdval_t native_pmd_val(pmd_t pmd)
+{
+	return native_pgd_val(pmd.pud.pgd);
+}
 #endif	/* PAGETABLE_LEVELS >= 3 */
 
 static inline pte_t native_make_pte(pteval_t val)
-- 
cgit v1.2.3


From 5b8dd1e95b5f2b7b80cfc7e6bb709603aef8bbc1 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: rearrange common mmu_ops

Rearrange the various pagetable mmu_ops to remove duplication.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 30 +++++++++++++++++-------------
 1 file changed, 17 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 4aa06b929e48..c38cf1ac4e48 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -230,28 +230,32 @@ struct pv_mmu_ops {
 	void (*pte_update_defer)(struct mm_struct *mm,
 				 unsigned long addr, pte_t *ptep);
 
+	pteval_t (*pte_val)(pte_t);
+	pte_t (*make_pte)(pteval_t pte);
+
+	pgdval_t (*pgd_val)(pgd_t);
+	pgd_t (*make_pgd)(pgdval_t pgd);
+
+#if PAGETABLE_LEVELS >= 3
 #ifdef CONFIG_X86_PAE
 	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
 	void (*set_pte_present)(struct mm_struct *mm, unsigned long addr,
 				pte_t *ptep, pte_t pte);
-	void (*set_pud)(pud_t *pudp, pud_t pudval);
 	void (*pte_clear)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 	void (*pmd_clear)(pmd_t *pmdp);
 
-	unsigned long long (*pte_val)(pte_t);
-	unsigned long long (*pmd_val)(pmd_t);
-	unsigned long long (*pgd_val)(pgd_t);
+#endif	/* CONFIG_X86_PAE */
+
+	void (*set_pud)(pud_t *pudp, pud_t pudval);
 
-	pte_t (*make_pte)(unsigned long long pte);
-	pmd_t (*make_pmd)(unsigned long long pmd);
-	pgd_t (*make_pgd)(unsigned long long pgd);
-#else
-	unsigned long (*pte_val)(pte_t);
-	unsigned long (*pgd_val)(pgd_t);
+	pmdval_t (*pmd_val)(pmd_t);
+	pmd_t (*make_pmd)(pmdval_t pmd);
 
-	pte_t (*make_pte)(unsigned long pte);
-	pgd_t (*make_pgd)(unsigned long pgd);
-#endif
+#if PAGETABLE_LEVELS == 4
+	pudval_t (*pud_val)(pud_t);
+	pud_t (*make_pud)(pudval_t pud);
+#endif	/* PAGETABLE_LEVELS == 4 */
+#endif	/* PAGETABLE_LEVELS >= 3 */
 
 #ifdef CONFIG_HIGHPTE
 	void *(*kmap_atomic_pte)(struct page *page, enum km_type type);
-- 
cgit v1.2.3


From 773221f46f82dc2f277dacc331d9d2ef2c690cb6 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: common implementation for pte value ops

Remove duplicate __pte/pte_val functions.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 48 ++++++++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index c38cf1ac4e48..b7342efab834 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -920,15 +920,37 @@ static inline void pte_update_defer(struct mm_struct *mm, unsigned long addr,
 	PVOP_VCALL3(pv_mmu_ops.pte_update_defer, mm, addr, ptep);
 }
 
-#ifdef CONFIG_X86_PAE
-static inline pte_t __pte(unsigned long long val)
+static inline pte_t __pte(pteval_t val)
 {
-	unsigned long long ret = PVOP_CALL2(unsigned long long,
-					    pv_mmu_ops.make_pte,
-					    val, val >> 32);
+	pteval_t ret;
+
+	if (sizeof(pteval_t) > sizeof(long))
+		ret = PVOP_CALL2(pteval_t,
+				 pv_mmu_ops.make_pte,
+				 val, (u64)val >> 32);
+	else
+		ret = PVOP_CALL1(pteval_t,
+				 pv_mmu_ops.make_pte,
+				 val);
+
 	return (pte_t) { .pte = ret };
 }
 
+static inline pteval_t pte_val(pte_t pte)
+{
+	pteval_t ret;
+
+	if (sizeof(pteval_t) > sizeof(long))
+		ret = PVOP_CALL2(pteval_t, pv_mmu_ops.pte_val,
+				 pte.pte, (u64)pte.pte >> 32);
+	else
+		ret = PVOP_CALL1(pteval_t, pv_mmu_ops.pte_val,
+				 pte.pte);
+
+	return ret;
+}
+
+#ifdef CONFIG_X86_PAE
 static inline pmd_t __pmd(unsigned long long val)
 {
 	return (pmd_t) { PVOP_CALL2(unsigned long long, pv_mmu_ops.make_pmd,
@@ -941,12 +963,6 @@ static inline pgd_t __pgd(unsigned long long val)
 				    val, val >> 32) };
 }
 
-static inline unsigned long long pte_val(pte_t x)
-{
-	return PVOP_CALL2(unsigned long long, pv_mmu_ops.pte_val,
-			  x.pte_low, x.pte_high);
-}
-
 static inline unsigned long long pmd_val(pmd_t x)
 {
 	return PVOP_CALL2(unsigned long long, pv_mmu_ops.pmd_val,
@@ -1008,21 +1024,11 @@ static inline void pmd_clear(pmd_t *pmdp)
 
 #else  /* !CONFIG_X86_PAE */
 
-static inline pte_t __pte(unsigned long val)
-{
-	return (pte_t) { PVOP_CALL1(unsigned long, pv_mmu_ops.make_pte, val) };
-}
-
 static inline pgd_t __pgd(unsigned long val)
 {
 	return (pgd_t) { PVOP_CALL1(unsigned long, pv_mmu_ops.make_pgd, val) };
 }
 
-static inline unsigned long pte_val(pte_t x)
-{
-	return PVOP_CALL1(unsigned long, pv_mmu_ops.pte_val, x.pte_low);
-}
-
 static inline unsigned long pgd_val(pgd_t x)
 {
 	return PVOP_CALL1(unsigned long, pv_mmu_ops.pgd_val, x.pgd);
-- 
cgit v1.2.3


From ef38503e034f82ad16a75a9dfea40ed7adaf00a8 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: common implementation for pgd value ops

Remove duplicate __pgd/pgd_val functions.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 50 ++++++++++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index b7342efab834..eac25b5323a2 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -950,16 +950,38 @@ static inline pteval_t pte_val(pte_t pte)
 	return ret;
 }
 
-#ifdef CONFIG_X86_PAE
-static inline pmd_t __pmd(unsigned long long val)
+static inline pgd_t __pgd(pgdval_t val)
 {
-	return (pmd_t) { PVOP_CALL2(unsigned long long, pv_mmu_ops.make_pmd,
-				    val, val >> 32) };
+	pgdval_t ret;
+
+	if (sizeof(pgdval_t) > sizeof(long))
+		ret = PVOP_CALL2(pgdval_t, pv_mmu_ops.make_pgd,
+				 val, (u64)val >> 32);
+	else
+		ret = PVOP_CALL1(pgdval_t, pv_mmu_ops.make_pgd,
+				 val);
+
+	return (pgd_t) { ret };
+}
+
+static inline pgdval_t pgd_val(pgd_t pgd)
+{
+	pgdval_t ret;
+
+	if (sizeof(pgdval_t) > sizeof(long))
+		ret =  PVOP_CALL2(pgdval_t, pv_mmu_ops.pgd_val,
+				  pgd.pgd, (u64)pgd.pgd >> 32);
+	else
+		ret =  PVOP_CALL1(pgdval_t, pv_mmu_ops.pgd_val,
+				  pgd.pgd);
+
+	return ret;
 }
 
-static inline pgd_t __pgd(unsigned long long val)
+#ifdef CONFIG_X86_PAE
+static inline pmd_t __pmd(unsigned long long val)
 {
-	return (pgd_t) { PVOP_CALL2(unsigned long long, pv_mmu_ops.make_pgd,
+	return (pmd_t) { PVOP_CALL2(unsigned long long, pv_mmu_ops.make_pmd,
 				    val, val >> 32) };
 }
 
@@ -969,12 +991,6 @@ static inline unsigned long long pmd_val(pmd_t x)
 			  x.pmd, x.pmd >> 32);
 }
 
-static inline unsigned long long pgd_val(pgd_t x)
-{
-	return PVOP_CALL2(unsigned long long, pv_mmu_ops.pgd_val,
-			  x.pgd, x.pgd >> 32);
-}
-
 static inline void set_pte(pte_t *ptep, pte_t pteval)
 {
 	PVOP_VCALL3(pv_mmu_ops.set_pte, ptep, pteval.pte_low, pteval.pte_high);
@@ -1024,16 +1040,6 @@ static inline void pmd_clear(pmd_t *pmdp)
 
 #else  /* !CONFIG_X86_PAE */
 
-static inline pgd_t __pgd(unsigned long val)
-{
-	return (pgd_t) { PVOP_CALL1(unsigned long, pv_mmu_ops.make_pgd, val) };
-}
-
-static inline unsigned long pgd_val(pgd_t x)
-{
-	return PVOP_CALL1(unsigned long, pv_mmu_ops.pgd_val, x.pgd);
-}
-
 static inline void set_pte(pte_t *ptep, pte_t pteval)
 {
 	PVOP_VCALL2(pv_mmu_ops.set_pte, ptep, pteval.pte_low);
-- 
cgit v1.2.3


From a632da2fc9c1e4e8021c5665fc34201fea485dcf Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: common implementation for pmd value ops

Remove duplicate __pmd/pmd_val functions.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 33 ++++++++++++++++++++++++++-------
 1 file changed, 26 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index eac25b5323a2..3593211e345b 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -978,18 +978,37 @@ static inline pgdval_t pgd_val(pgd_t pgd)
 	return ret;
 }
 
-#ifdef CONFIG_X86_PAE
-static inline pmd_t __pmd(unsigned long long val)
+#if PAGETABLE_LEVELS >= 3
+static inline pmd_t __pmd(pmdval_t val)
 {
-	return (pmd_t) { PVOP_CALL2(unsigned long long, pv_mmu_ops.make_pmd,
-				    val, val >> 32) };
+	pmdval_t ret;
+
+	if (sizeof(pmdval_t) > sizeof(long))
+		ret = PVOP_CALL2(pmdval_t, pv_mmu_ops.make_pmd,
+				 val, (u64)val >> 32);
+	else
+		ret = PVOP_CALL1(pmdval_t, pv_mmu_ops.make_pmd,
+				 val);
+
+	return (pmd_t) { ret };
 }
 
-static inline unsigned long long pmd_val(pmd_t x)
+static inline pmdval_t pmd_val(pmd_t pmd)
 {
-	return PVOP_CALL2(unsigned long long, pv_mmu_ops.pmd_val,
-			  x.pmd, x.pmd >> 32);
+	pmdval_t ret;
+
+	if (sizeof(pmdval_t) > sizeof(long))
+		ret =  PVOP_CALL2(pmdval_t, pv_mmu_ops.pmd_val,
+				  pmd.pmd, (u64)pmd.pmd >> 32);
+	else
+		ret =  PVOP_CALL1(pmdval_t, pv_mmu_ops.pmd_val,
+				  pmd.pmd);
+
+	return ret;
 }
+#endif	/* PAGETABLE_LEVELS >= 3 */
+
+#ifdef CONFIG_X86_PAE
 
 static inline void set_pte(pte_t *ptep, pte_t pteval)
 {
-- 
cgit v1.2.3


From 4eed80cd7a1668df6ddd884f9474beccc81ef3f7 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: make set_pte operations common

Remove duplicate set_pte* operations.  PAE still needs to have special
variants of some of these because it can't atomically update a 64-bit
pte, so there's still some duplication.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 116 +++++++++++++++++++++++----------------------
 1 file changed, 60 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 3593211e345b..ba1b94334144 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -978,6 +978,66 @@ static inline pgdval_t pgd_val(pgd_t pgd)
 	return ret;
 }
 
+static inline void set_pte(pte_t *ptep, pte_t pte)
+{
+	if (sizeof(pteval_t) > sizeof(long))
+		PVOP_VCALL3(pv_mmu_ops.set_pte, ptep,
+			    pte.pte, (u64)pte.pte >> 32);
+	else
+		PVOP_VCALL2(pv_mmu_ops.set_pte, ptep,
+			    pte.pte);
+}
+
+static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
+			      pte_t *ptep, pte_t pte)
+{
+	if (sizeof(pteval_t) > sizeof(long))
+		/* 5 arg words */
+		pv_mmu_ops.set_pte_at(mm, addr, ptep, pte);
+	else
+		PVOP_VCALL4(pv_mmu_ops.set_pte_at, mm, addr, ptep, pte.pte);
+}
+
+#ifdef CONFIG_X86_PAE
+/* Special-case pte-setting operations for PAE, which can't update a
+   64-bit pte atomically */
+static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+	PVOP_VCALL3(pv_mmu_ops.set_pte_atomic, ptep,
+		    pte.pte, pte.pte >> 32);
+}
+
+static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
+				   pte_t *ptep, pte_t pte)
+{
+	/* 5 arg words */
+	pv_mmu_ops.set_pte_present(mm, addr, ptep, pte);
+}
+
+static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
+			     pte_t *ptep)
+{
+	PVOP_VCALL3(pv_mmu_ops.pte_clear, mm, addr, ptep);
+}
+#else  /* !CONFIG_X86_PAE */
+static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+	set_pte(ptep, pte);
+}
+
+static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
+				   pte_t *ptep, pte_t pte)
+{
+	set_pte(ptep, pte);
+}
+
+static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
+			     pte_t *ptep)
+{
+	set_pte_at(mm, addr, ptep, __pte(0));
+}
+#endif	/* CONFIG_X86_PAE */
+
 #if PAGETABLE_LEVELS >= 3
 static inline pmd_t __pmd(pmdval_t val)
 {
@@ -1010,31 +1070,6 @@ static inline pmdval_t pmd_val(pmd_t pmd)
 
 #ifdef CONFIG_X86_PAE
 
-static inline void set_pte(pte_t *ptep, pte_t pteval)
-{
-	PVOP_VCALL3(pv_mmu_ops.set_pte, ptep, pteval.pte_low, pteval.pte_high);
-}
-
-static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
-			      pte_t *ptep, pte_t pteval)
-{
-	/* 5 arg words */
-	pv_mmu_ops.set_pte_at(mm, addr, ptep, pteval);
-}
-
-static inline void set_pte_atomic(pte_t *ptep, pte_t pteval)
-{
-	PVOP_VCALL3(pv_mmu_ops.set_pte_atomic, ptep,
-		    pteval.pte_low, pteval.pte_high);
-}
-
-static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
-				   pte_t *ptep, pte_t pte)
-{
-	/* 5 arg words */
-	pv_mmu_ops.set_pte_present(mm, addr, ptep, pte);
-}
-
 static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	PVOP_VCALL3(pv_mmu_ops.set_pmd, pmdp,
@@ -1047,11 +1082,6 @@ static inline void set_pud(pud_t *pudp, pud_t pudval)
 		    pudval.pgd.pgd, pudval.pgd.pgd >> 32);
 }
 
-static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	PVOP_VCALL3(pv_mmu_ops.pte_clear, mm, addr, ptep);
-}
-
 static inline void pmd_clear(pmd_t *pmdp)
 {
 	PVOP_VCALL1(pv_mmu_ops.pmd_clear, pmdp);
@@ -1059,17 +1089,6 @@ static inline void pmd_clear(pmd_t *pmdp)
 
 #else  /* !CONFIG_X86_PAE */
 
-static inline void set_pte(pte_t *ptep, pte_t pteval)
-{
-	PVOP_VCALL2(pv_mmu_ops.set_pte, ptep, pteval.pte_low);
-}
-
-static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
-			      pte_t *ptep, pte_t pteval)
-{
-	PVOP_VCALL4(pv_mmu_ops.set_pte_at, mm, addr, ptep, pteval.pte_low);
-}
-
 static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	PVOP_VCALL2(pv_mmu_ops.set_pmd, pmdp, pmdval.pud.pgd.pgd);
@@ -1080,21 +1099,6 @@ static inline void pmd_clear(pmd_t *pmdp)
 	set_pmd(pmdp, __pmd(0));
 }
 
-static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	set_pte_at(mm, addr, ptep, __pte(0));
-}
-
-static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
-{
-	set_pte(ptep, pte);
-}
-
-static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
-				   pte_t *ptep, pte_t pte)
-{
-	set_pte(ptep, pte);
-}
 #endif	/* CONFIG_X86_PAE */
 
 /* Lazy mode for batching updates / context switch */
-- 
cgit v1.2.3


From 60b3f626dee9f0b1b656918e40e211822c18200d Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: make set_pmd operation common

Remove duplicate set_pmd()s.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 43 ++++++++++++++++++++-----------------------
 1 file changed, 20 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index ba1b94334144..d9957ece49b5 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -998,6 +998,16 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 		PVOP_VCALL4(pv_mmu_ops.set_pte_at, mm, addr, ptep, pte.pte);
 }
 
+static inline void set_pmd(pmd_t *pmdp, pmd_t pmd)
+{
+	pmdval_t val = native_pmd_val(pmd);
+
+	if (sizeof(pmdval_t) > sizeof(long))
+		PVOP_VCALL3(pv_mmu_ops.set_pmd, pmdp, val, (u64)val >> 32);
+	else
+		PVOP_VCALL2(pv_mmu_ops.set_pmd, pmdp, val);
+}
+
 #ifdef CONFIG_X86_PAE
 /* Special-case pte-setting operations for PAE, which can't update a
    64-bit pte atomically */
@@ -1019,6 +1029,11 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
 {
 	PVOP_VCALL3(pv_mmu_ops.pte_clear, mm, addr, ptep);
 }
+
+static inline void pmd_clear(pmd_t *pmdp)
+{
+	PVOP_VCALL1(pv_mmu_ops.pmd_clear, pmdp);
+}
 #else  /* !CONFIG_X86_PAE */
 static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
 {
@@ -1036,6 +1051,11 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
 {
 	set_pte_at(mm, addr, ptep, __pte(0));
 }
+
+static inline void pmd_clear(pmd_t *pmdp)
+{
+	set_pmd(pmdp, __pmd(0));
+}
 #endif	/* CONFIG_X86_PAE */
 
 #if PAGETABLE_LEVELS >= 3
@@ -1070,35 +1090,12 @@ static inline pmdval_t pmd_val(pmd_t pmd)
 
 #ifdef CONFIG_X86_PAE
 
-static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
-{
-	PVOP_VCALL3(pv_mmu_ops.set_pmd, pmdp,
-		    pmdval.pmd, pmdval.pmd >> 32);
-}
-
 static inline void set_pud(pud_t *pudp, pud_t pudval)
 {
 	PVOP_VCALL3(pv_mmu_ops.set_pud, pudp,
 		    pudval.pgd.pgd, pudval.pgd.pgd >> 32);
 }
 
-static inline void pmd_clear(pmd_t *pmdp)
-{
-	PVOP_VCALL1(pv_mmu_ops.pmd_clear, pmdp);
-}
-
-#else  /* !CONFIG_X86_PAE */
-
-static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
-{
-	PVOP_VCALL2(pv_mmu_ops.set_pmd, pmdp, pmdval.pud.pgd.pgd);
-}
-
-static inline void pmd_clear(pmd_t *pmdp)
-{
-	set_pmd(pmdp, __pmd(0));
-}
-
 #endif	/* CONFIG_X86_PAE */
 
 /* Lazy mode for batching updates / context switch */
-- 
cgit v1.2.3


From 28c6075c72a312986f6d7802005251bdb91ca80c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86/paravirt: make set_pud operation common

Remove duplicate set_pud()s.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index d9957ece49b5..c56b17a5eba0 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1086,17 +1086,19 @@ static inline pmdval_t pmd_val(pmd_t pmd)
 
 	return ret;
 }
-#endif	/* PAGETABLE_LEVELS >= 3 */
-
-#ifdef CONFIG_X86_PAE
 
-static inline void set_pud(pud_t *pudp, pud_t pudval)
+static inline void set_pud(pud_t *pudp, pud_t pud)
 {
-	PVOP_VCALL3(pv_mmu_ops.set_pud, pudp,
-		    pudval.pgd.pgd, pudval.pgd.pgd >> 32);
-}
+	pudval_t val = native_pud_val(pud);
 
-#endif	/* CONFIG_X86_PAE */
+	if (sizeof(pudval_t) > sizeof(long))
+		PVOP_VCALL3(pv_mmu_ops.set_pud, pudp,
+			    val, (u64)val >> 32);
+	else
+		PVOP_VCALL2(pv_mmu_ops.set_pud, pudp,
+			    val);
+}
+#endif	/* PAGETABLE_LEVELS >= 3 */
 
 /* Lazy mode for batching updates / context switch */
 enum paravirt_lazy_mode {
-- 
cgit v1.2.3


From e34cda04d4ec0114185459eb101a2245563bbeba Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@mcmartin.ca>
Date: Wed, 30 Jan 2008 13:33:15 +0100
Subject: x86: 64-bit, remove redundant cpu_has_ definitions

PSE, PGE, XMM, XMM2, and FXSR are defined as required features, and
will be optimized to a constant at compile time. Remove their redundant
definitions.

Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/cpufeature.h | 15 ---------------
 1 file changed, 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 93c143fb8a51..29727bf0e177 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -194,21 +194,6 @@
 #undef  cpu_has_centaur_mcr
 #define cpu_has_centaur_mcr	0
 
-#undef  cpu_has_pse
-#define cpu_has_pse		1
-
-#undef  cpu_has_pge
-#define cpu_has_pge		1
-
-#undef  cpu_has_xmm
-#define cpu_has_xmm		1
-
-#undef  cpu_has_xmm2
-#define cpu_has_xmm2		1
-
-#undef  cpu_has_fxsr
-#define cpu_has_fxsr		1
-
 #endif /* CONFIG_X86_64 */
 
 #endif /* _ASM_X86_CPUFEATURE_H */
-- 
cgit v1.2.3


From 770181d99098bc141031fe0a4b1d6221258c1182 Mon Sep 17 00:00:00 2001
From: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Date: Wed, 30 Jan 2008 13:33:16 +0100
Subject: x86_32: remove the useless NR_syscalls macro

This is against current x86.git.

The size of the system call table for 32-bit x86 kernels is obtained by
compile-time calculation of the sys_call_table array, not from the value,
which the NR_syscalls macro expands to. This trivial patch removes the
fossil macro.

Manually tested by grepping the x86 files for the "NR_syscalls" string.
No relevant use cases found.

Build-tested using allyesconfig, allnoconfig and a couple of randconfig
instances. All builds successfully finished.

Runtime test performed using a stripped-down Debian-ish config. The system
booted successfully.

Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/unistd_32.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index 9b15545eb9b5..8d8f9b5adbb9 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -333,8 +333,6 @@
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 325
-
 #define __ARCH_WANT_IPC_PARSE_VERSION
 #define __ARCH_WANT_OLD_READDIR
 #define __ARCH_WANT_OLD_STAT
-- 
cgit v1.2.3


From 064954761254ef17fae2b84fb5a034d48a769143 Mon Sep 17 00:00:00 2001
From: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Date: Wed, 30 Jan 2008 13:33:16 +0100
Subject: x86: kdump failure

kdump needs ELF_CORE_COPY_REGS in crash_save_cpu().
This lack of the macro causes the following BUG.

 SysRq : Trigger a crashdump
 ------------[ cut here ]------------
 kernel BUG at include/linux/elfcore.h:105!
 invalid opcode: 0000 [1] PREEMPT SMP

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/elf.h | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index dab4744f9586..d9c94e707289 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -106,6 +106,31 @@ extern unsigned int vdso_enabled;
 	_r->ax = 0; \
 } while (0)
 
+/*
+ * regs is struct pt_regs, pr_reg is elf_gregset_t (which is
+ * now struct_user_regs, they are different)
+ */
+
+#define ELF_CORE_COPY_REGS(pr_reg, regs) do {		\
+	pr_reg[0] = regs->bx;				\
+	pr_reg[1] = regs->cx;				\
+	pr_reg[2] = regs->dx;				\
+	pr_reg[3] = regs->si;				\
+	pr_reg[4] = regs->di;				\
+	pr_reg[5] = regs->bp;				\
+	pr_reg[6] = regs->ax;				\
+	pr_reg[7] = regs->ds & 0xffff;			\
+	pr_reg[8] = regs->es & 0xffff;			\
+	pr_reg[9] = regs->fs & 0xffff;			\
+	savesegment(gs, pr_reg[10]);			\
+	pr_reg[11] = regs->orig_ax;			\
+	pr_reg[12] = regs->ip;				\
+	pr_reg[13] = regs->cs & 0xffff;			\
+	pr_reg[14] = regs->flags;			\
+	pr_reg[15] = regs->sp;				\
+	pr_reg[16] = regs->ss & 0xffff;			\
+} while (0);
+
 #define ELF_PLATFORM	(utsname()->machine)
 #define set_personality_64bit()	do { } while (0)
 
@@ -165,6 +190,43 @@ static inline void elf_common_init(struct thread_struct *t,
 	} while (0)
 #define COMPAT_ELF_PLATFORM			("i686")
 
+/*
+ * regs is struct pt_regs, pr_reg is elf_gregset_t (which is
+ * now struct_user_regs, they are different). Assumes current is the process
+ * getting dumped.
+ */
+
+#define ELF_CORE_COPY_REGS(pr_reg, regs)  do {			\
+	unsigned v;						\
+	(pr_reg)[0] = (regs)->r15;				\
+	(pr_reg)[1] = (regs)->r14;				\
+	(pr_reg)[2] = (regs)->r13;				\
+	(pr_reg)[3] = (regs)->r12;				\
+	(pr_reg)[4] = (regs)->bp;				\
+	(pr_reg)[5] = (regs)->bx;				\
+	(pr_reg)[6] = (regs)->r11;				\
+	(pr_reg)[7] = (regs)->r10;				\
+	(pr_reg)[8] = (regs)->r9;				\
+	(pr_reg)[9] = (regs)->r8;				\
+	(pr_reg)[10] = (regs)->ax;				\
+	(pr_reg)[11] = (regs)->cx;				\
+	(pr_reg)[12] = (regs)->dx;				\
+	(pr_reg)[13] = (regs)->si;				\
+	(pr_reg)[14] = (regs)->di;				\
+	(pr_reg)[15] = (regs)->orig_ax;				\
+	(pr_reg)[16] = (regs)->ip;				\
+	(pr_reg)[17] = (regs)->cs;				\
+	(pr_reg)[18] = (regs)->flags;				\
+	(pr_reg)[19] = (regs)->sp;				\
+	(pr_reg)[20] = (regs)->ss;				\
+	(pr_reg)[21] = current->thread.fs;			\
+	(pr_reg)[22] = current->thread.gs;			\
+	asm("movl %%ds,%0" : "=r" (v)); (pr_reg)[23] = v;	\
+	asm("movl %%es,%0" : "=r" (v)); (pr_reg)[24] = v;	\
+	asm("movl %%fs,%0" : "=r" (v)); (pr_reg)[25] = v;	\
+	asm("movl %%gs,%0" : "=r" (v)); (pr_reg)[26] = v;	\
+} while (0);
+
 /* I'm not sure if we can use '-' here */
 #define ELF_PLATFORM       ("x86_64")
 extern void set_personality_64bit(void);
-- 
cgit v1.2.3


From 90d43d728d22a5e20c63c8db1cfb497210768f19 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:16 +0100
Subject: x86: unify pt_regs accessors ptrace.h

Unify the definiton of:
v8086_mode
user_mode
user_mode_vm
stack_pointer
instruction_pointer
frame_pointer

in ptrace.h to make it clear where the differences are between
32 and 64 bit.  Changes macros to static inlines as well.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/ptrace.h | 84 +++++++++++++++++++++++++++++++-----------------
 1 file changed, 54 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index ee4b595e1ccc..0ad9a2b7f8c0 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -76,30 +76,6 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
 
-/*
- * user_mode_vm(regs) determines whether a register set came from user mode.
- * This is true if V8086 mode was enabled OR if the register set was from
- * protected mode with RPL-3 CS value.  This tricky test checks that with
- * one comparison.  Many places in the kernel can bypass this full check
- * if they have already ruled out V8086 mode, so user_mode(regs) can be used.
- */
-static inline int user_mode(struct pt_regs *regs)
-{
-	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
-}
-static inline int user_mode_vm(struct pt_regs *regs)
-{
-	return ((regs->cs & SEGMENT_RPL_MASK) |
-		(regs->flags & VM_MASK)) >= USER_RPL;
-}
-static inline int v8086_mode(struct pt_regs *regs)
-{
-	return (regs->flags & VM_MASK);
-}
-
-#define instruction_pointer(regs) ((regs)->ip)
-#define frame_pointer(regs) ((regs)->bp)
-#define stack_pointer(regs) ((unsigned long)(regs))
 #define regs_return_value(regs) ((regs)->ax)
 
 extern unsigned long profile_pc(struct pt_regs *regs);
@@ -167,12 +143,6 @@ struct pt_regs {
 /* top of stack page */
 };
 
-#define user_mode(regs) (!!((regs)->cs & 3))
-#define user_mode_vm(regs) user_mode(regs)
-#define v8086_mode(regs) 0	/* No V86 mode support in long mode */
-#define instruction_pointer(regs) ((regs)->ip)
-#define frame_pointer(regs) ((regs)->bp)
-#define stack_pointer(regs) ((regs)->sp)
 #define regs_return_value(regs) ((regs)->ax)
 
 extern unsigned long profile_pc(struct pt_regs *regs);
@@ -188,6 +158,60 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 #ifdef __KERNEL__
 
+/*
+ * user_mode_vm(regs) determines whether a register set came from user mode.
+ * This is true if V8086 mode was enabled OR if the register set was from
+ * protected mode with RPL-3 CS value.  This tricky test checks that with
+ * one comparison.  Many places in the kernel can bypass this full check
+ * if they have already ruled out V8086 mode, so user_mode(regs) can be used.
+ */
+static inline int user_mode(struct pt_regs *regs)
+{
+#ifdef CONFIG_X86_32
+	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
+#else
+	return !!(regs->cs & 3);
+#endif
+}
+
+static inline int user_mode_vm(struct pt_regs *regs)
+{
+#ifdef CONFIG_X86_32
+	return ((regs->cs & SEGMENT_RPL_MASK) |
+		(regs->flags & VM_MASK)) >= USER_RPL;
+#else
+	return user_mode(regs);
+#endif
+}
+
+static inline int v8086_mode(struct pt_regs *regs)
+{
+#ifdef CONFIG_X86_32
+	return (regs->flags & VM_MASK);
+#else
+	return 0;	/* No V86 mode support in long mode */
+#endif
+}
+
+static inline unsigned long stack_pointer(struct pt_regs *regs)
+{
+#ifdef CONFIG_X86_32
+	return (unsigned long)regs;
+#else
+	return regs->sp;
+#endif
+}
+
+static inline unsigned long instruction_pointer(struct pt_regs *regs)
+{
+	return regs->ip;
+}
+
+static inline unsigned long frame_pointer(struct pt_regs *regs)
+{
+	return regs->bp;
+}
+
 /*
  * These are defined as per linux/ptrace.h, which see.
  */
-- 
cgit v1.2.3


From dbe3533b7f687402651e893d5b3a7b66f22d6487 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:16 +0100
Subject: x86: clean up ptrace.h

Leave definition of pt_regs in its own section, move all kernel
code to section afterwards, unify prototype definitions, has some
conditional prototypes to make it clear what was only defined in
32 and 64 bit.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/ptrace.h | 43 ++++++++++++++++---------------------------
 1 file changed, 16 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 0ad9a2b7f8c0..79d5b8fcd7b1 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -7,17 +7,6 @@
 
 #ifndef __ASSEMBLY__
 
-#ifdef __KERNEL__
-
-/* the DS BTS struct is used for ptrace as well */
-#include <asm/ds.h>
-
-struct task_struct;
-extern void ptrace_bts_take_timestamp(struct task_struct *, enum bts_qualifier);
-
-#endif /* __KERNEL__ */
-
-
 #ifdef __i386__
 /* this struct defines the way the registers are stored on the
    stack during a system call. */
@@ -69,16 +58,6 @@ struct pt_regs {
 #include <asm/vm86.h>
 #include <asm/segment.h>
 
-struct task_struct;
-
-extern unsigned long
-convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
-
-extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
-
-#define regs_return_value(regs) ((regs)->ax)
-
-extern unsigned long profile_pc(struct pt_regs *regs);
 #endif /* __KERNEL__ */
 
 #else /* __i386__ */
@@ -143,20 +122,30 @@ struct pt_regs {
 /* top of stack page */
 };
 
-#define regs_return_value(regs) ((regs)->ax)
+#endif /* __KERNEL__ */
+#endif /* !__i386__ */
 
-extern unsigned long profile_pc(struct pt_regs *regs);
-void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
+#ifdef __KERNEL__
+
+/* the DS BTS struct is used for ptrace as well */
+#include <asm/ds.h>
 
 struct task_struct;
 
+extern void ptrace_bts_take_timestamp(struct task_struct *, enum bts_qualifier);
+
+extern unsigned long profile_pc(struct pt_regs *regs);
+
 extern unsigned long
 convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
-#endif /* __KERNEL__ */
-#endif /* !__i386__ */
+#ifdef CONFIG_X86_32
+extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
+#else
+void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
+#endif
 
-#ifdef __KERNEL__
+#define regs_return_value(regs) ((regs)->ax)
 
 /*
  * user_mode_vm(regs) determines whether a register set came from user mode.
-- 
cgit v1.2.3


From f6e8e28410c017adb273983057023a17a8791ad3 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 30 Jan 2008 13:33:16 +0100
Subject: x86: rename stack_pointer to kernel_trap_sp

Choose a less generic name for such a special case.  Add
a comment explaining the odd use in X86_32.

Change the one user of stack_pointer.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/oprofile/backtrace.c | 2 +-
 include/asm-x86/ptrace.h      | 8 +++++++-
 2 files changed, 8 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c
index 0ca4815a2938..e2095cba409f 100644
--- a/arch/x86/oprofile/backtrace.c
+++ b/arch/x86/oprofile/backtrace.c
@@ -77,7 +77,7 @@ void
 x86_backtrace(struct pt_regs * const regs, unsigned int depth)
 {
 	struct frame_head *head = (struct frame_head *)frame_pointer(regs);
-	unsigned long stack = stack_pointer(regs);
+	unsigned long stack = kernel_trap_sp(regs);
 
 	if (!user_mode_vm(regs)) {
 		if (depth)
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 79d5b8fcd7b1..d9e04b46a440 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -182,7 +182,13 @@ static inline int v8086_mode(struct pt_regs *regs)
 #endif
 }
 
-static inline unsigned long stack_pointer(struct pt_regs *regs)
+/*
+ * X86_32 CPUs don't save ss and esp if the CPU is already in kernel mode
+ * when it traps.  So regs will be the current sp.
+ *
+ * This is valid only for kernel mode traps.
+ */
+static inline unsigned long kernel_trap_sp(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
 	return (unsigned long)regs;
-- 
cgit v1.2.3


From 751752789162fde69474edfa15935d0a77c0bc17 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:17 +0100
Subject: x86: replace hard coded reservations in 64-bit early boot code with
 dynamic table

On x86-64 there are several memory allocations before bootmem. To avoid
them stomping on each other they used to be all hard coded in bad_area().
Replace this with an array that is filled as needed.

This cleans up the code considerably and allows to expand its use.

Cc: peterz@infradead.org
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_64.c  | 95 +++++++++++++++++++++++++---------------------
 arch/x86/kernel/head64.c   | 48 +++++++++++++++++++++++
 arch/x86/kernel/setup_64.c | 67 ++------------------------------
 arch/x86/mm/init_64.c      |  5 ++-
 arch/x86/mm/numa_64.c      |  1 +
 include/asm-x86/e820_64.h  |  4 +-
 include/asm-x86/proto.h    |  2 -
 7 files changed, 110 insertions(+), 112 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 07cfaae7ab07..f8b7bebb4344 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -47,56 +47,65 @@ unsigned long end_pfn_map;
  */
 static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
 
-/* Check for some hardcoded bad areas that early boot is not allowed to touch */
-static inline int bad_addr(unsigned long *addrp, unsigned long size)
-{
-	unsigned long addr = *addrp, last = addr + size;
+/*
+ * Early reserved memory areas.
+ */
+#define MAX_EARLY_RES 20
+
+struct early_res {
+	unsigned long start, end;
+};
+static struct early_res early_res[MAX_EARLY_RES] __initdata = {
+	{ 0, PAGE_SIZE },			/* BIOS data page */
+#ifdef CONFIG_SMP
+	{ SMP_TRAMPOLINE_BASE, SMP_TRAMPOLINE_BASE + 2*PAGE_SIZE },
+#endif
+	{}
+};
 
-	/* various gunk below that needed for SMP startup */
-	if (addr < 0x8000) {
-		*addrp = PAGE_ALIGN(0x8000);
-		return 1;
+void __init reserve_early(unsigned long start, unsigned long end)
+{
+	int i;
+	struct early_res *r;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		r = &early_res[i];
+		if (end > r->start && start < r->end)
+			panic("Duplicated early reservation %lx-%lx\n",
+			      start, end);
 	}
+	if (i >= MAX_EARLY_RES)
+		panic("Too many early reservations");
+	r = &early_res[i];
+	r->start = start;
+	r->end = end;
+}
 
-	/* direct mapping tables of the kernel */
-	if (last >= table_start<<PAGE_SHIFT && addr < table_end<<PAGE_SHIFT) {
-		*addrp = PAGE_ALIGN(table_end << PAGE_SHIFT);
-		return 1;
+void __init early_res_to_bootmem(void)
+{
+	int i;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		struct early_res *r = &early_res[i];
+		reserve_bootmem_generic(r->start, r->end - r->start);
 	}
+}
 
-	/* initrd */
-#ifdef CONFIG_BLK_DEV_INITRD
-	if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
-		unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
-		unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
-		unsigned long ramdisk_end   = ramdisk_image+ramdisk_size;
-
-		if (last >= ramdisk_image && addr < ramdisk_end) {
-			*addrp = PAGE_ALIGN(ramdisk_end);
-			return 1;
+/* Check for already reserved areas */
+static inline int bad_addr(unsigned long *addrp, unsigned long size)
+{
+	int i;
+	unsigned long addr = *addrp, last;
+	int changed = 0;
+again:
+	last = addr + size;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		struct early_res *r = &early_res[i];
+		if (last >= r->start && addr < r->end) {
+			*addrp = addr = r->end;
+			changed = 1;
+			goto again;
 		}
 	}
-#endif
-	/* kernel code */
-	if (last >= __pa_symbol(&_text) && addr < __pa_symbol(&_end)) {
-		*addrp = PAGE_ALIGN(__pa_symbol(&_end));
-		return 1;
-	}
-
-	if (last >= ebda_addr && addr < ebda_addr + ebda_size) {
-		*addrp = PAGE_ALIGN(ebda_addr + ebda_size);
-		return 1;
-	}
-
-#ifdef CONFIG_NUMA
-	/* NUMA memory to node map */
-	if (last >= nodemap_addr && addr < nodemap_addr + nodemap_size) {
-		*addrp = nodemap_addr + nodemap_size;
-		return 1;
-	}
-#endif
-	/* XXX ramdisk image here? */
-	return 0;
+	return changed;
 }
 
 /*
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 87e031d4abf1..58438bafedca 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -21,6 +21,7 @@
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
 #include <asm/kdebug.h>
+#include <asm/e820.h>
 
 static void __init zap_identity_mappings(void)
 {
@@ -48,6 +49,35 @@ static void __init copy_bootdata(char *real_mode_data)
 	}
 }
 
+#define EBDA_ADDR_POINTER 0x40E
+
+static __init void reserve_ebda(void)
+{
+	unsigned ebda_addr, ebda_size;
+
+	/*
+	 * there is a real-mode segmented pointer pointing to the
+	 * 4K EBDA area at 0x40E
+	 */
+	ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER);
+	ebda_addr <<= 4;
+
+	if (!ebda_addr)
+		return;
+
+	ebda_size = *(unsigned short *)__va(ebda_addr);
+
+	/* Round EBDA up to pages */
+	if (ebda_size == 0)
+		ebda_size = 1;
+	ebda_size <<= 10;
+	ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
+	if (ebda_size > 64*1024)
+		ebda_size = 64*1024;
+
+	reserve_early(ebda_addr, ebda_addr + ebda_size);
+}
+
 void __init x86_64_start_kernel(char * real_mode_data)
 {
 	int i;
@@ -75,5 +105,23 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	pda_init(0);
 	copy_bootdata(__va(real_mode_data));
 
+	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end));
+
+	/* Reserve INITRD */
+	if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
+		unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
+		unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
+		unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
+		reserve_early(ramdisk_image, ramdisk_end);
+	}
+
+	reserve_ebda();
+
+	/*
+	 * At this point everything still needed from the boot loader
+	 * or BIOS or kernel text should be early reserved or marked not
+	 * RAM in e820. All other memory is free game.
+	 */
+
 	start_kernel();
 }
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 4a3f00b49236..6cbd15625dce 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -245,41 +245,6 @@ static inline void __init reserve_crashkernel(void)
 {}
 #endif
 
-#define EBDA_ADDR_POINTER 0x40E
-
-unsigned __initdata ebda_addr;
-unsigned __initdata ebda_size;
-
-static void __init discover_ebda(void)
-{
-	/*
-	 * there is a real-mode segmented pointer pointing to the
-	 * 4K EBDA area at 0x40E
-	 */
-	ebda_addr = *(unsigned short *)__va(EBDA_ADDR_POINTER);
-	/*
-	 * There can be some situations, like paravirtualized guests,
-	 * in which there is no available ebda information. In such
-	 * case, just skip it
-	 */
-	if (!ebda_addr) {
-		ebda_size = 0;
-		return;
-	}
-
-	ebda_addr <<= 4;
-
-	ebda_size = *(unsigned short *)__va(ebda_addr);
-
-	/* Round EBDA up to pages */
-	if (ebda_size == 0)
-		ebda_size = 1;
-	ebda_size <<= 10;
-	ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
-	if (ebda_size > 64*1024)
-		ebda_size = 64*1024;
-}
-
 /* Overridden in paravirt.c if CONFIG_PARAVIRT */
 void __attribute__((weak)) __init memory_setup(void)
 {
@@ -349,8 +314,6 @@ void __init setup_arch(char **cmdline_p)
 
 	check_efer();
 
-	discover_ebda();
-
 	init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
 	if (efi_enabled)
 		efi_init();
@@ -397,33 +360,7 @@ void __init setup_arch(char **cmdline_p)
 	contig_initmem_init(0, end_pfn);
 #endif
 
-	/* Reserve direct mapping */
-	reserve_bootmem_generic(table_start << PAGE_SHIFT,
-				(table_end - table_start) << PAGE_SHIFT);
-
-	/* reserve kernel */
-	reserve_bootmem_generic(__pa_symbol(&_text),
-				__pa_symbol(&_end) - __pa_symbol(&_text));
-
-	/*
-	 * reserve physical page 0 - it's a special BIOS page on many boxes,
-	 * enabling clean reboots, SMP operation, laptop functions.
-	 */
-	reserve_bootmem_generic(0, PAGE_SIZE);
-
-	/* reserve ebda region */
-	if (ebda_addr)
-		reserve_bootmem_generic(ebda_addr, ebda_size);
-#ifdef CONFIG_NUMA
-	/* reserve nodemap region */
-	if (nodemap_addr)
-		reserve_bootmem_generic(nodemap_addr, nodemap_size);
-#endif
-
-#ifdef CONFIG_SMP
-	/* Reserve SMP trampoline */
-	reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, 2*PAGE_SIZE);
-#endif
+	early_res_to_bootmem();
 
 #ifdef CONFIG_ACPI_SLEEP
 	/*
@@ -453,6 +390,8 @@ void __init setup_arch(char **cmdline_p)
 			initrd_start = ramdisk_image + PAGE_OFFSET;
 			initrd_end = initrd_start+ramdisk_size;
 		} else {
+			/* Assumes everything on node 0 */
+			free_bootmem(ramdisk_image, ramdisk_size);
 			printk(KERN_ERR "initrd extends beyond end of memory "
 			       "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
 			       ramdisk_end, end_of_mem);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 05f12c527b02..8198840c3dcb 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -176,7 +176,8 @@ __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
 	set_pte_phys(address, phys, prot);
 }
 
-unsigned long __meminitdata table_start, table_end;
+static unsigned long __initdata table_start;
+static unsigned long __meminitdata table_end;
 
 static __meminit void *alloc_low_page(unsigned long *phys)
 { 
@@ -387,6 +388,8 @@ void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 	if (!after_bootmem)
 		mmu_cr4_features = read_cr4();
 	__flush_tlb_all();
+
+	reserve_early(table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
 }
 
 #ifndef CONFIG_NUMA
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 37d429beba96..5d24dc1ec237 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -102,6 +102,7 @@ static int __init allocate_cachealigned_memnodemap(void)
 	}
 	pad_addr = (nodemap_addr + pad) & ~pad;
 	memnodemap = phys_to_virt(pad_addr);
+	reserve_early(nodemap_addr, nodemap_addr + nodemap_size);
 
 	printk(KERN_DEBUG "NUMA: Allocated memnodemap from %lx - %lx\n",
 	       nodemap_addr, nodemap_addr + nodemap_size);
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index ff36f434f00b..51e4170f9ca5 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -41,8 +41,8 @@ extern void finish_e820_parsing(void);
 extern struct e820map e820;
 extern void update_e820(void);
 
-extern unsigned ebda_addr, ebda_size;
-extern unsigned long nodemap_addr, nodemap_size;
+extern void reserve_early(unsigned long start, unsigned long end);
+extern void early_res_to_bootmem(void);
 
 #endif/*!__ASSEMBLY__*/
 
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 3c6fb89f1120..68563c0709ac 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -22,8 +22,6 @@ extern void syscall32_cpu_init(void);
 
 extern void check_efer(void);
 
-extern unsigned long table_start, table_end;
-
 extern int reboot_force;
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
-- 
cgit v1.2.3


From ca74a6f84e68b44867022f4a4f3ec17c087c864e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:17 +0100
Subject: x86: optimize lock prefix switching to run less frequently

On VMs implemented using JITs that cache translated code changing the lock
prefixes is a quite costly operation that forces the JIT to throw away and
retranslate a lot of code.

Previously a SMP kernel would rewrite the locks once for each CPU which
is quite unnecessary. This patch changes the code to never switch at boot in
 the normal case (SMP kernel booting with >1 CPU) or only once for SMP kernel
on UP.

This makes a significant difference in boot up performance on AMD SimNow!
Also I expect it to be a little faster on native systems too because a smp
switch does a lot of text_poke()s which each synchronize the pipeline.

v1->v2: Rename max_cpus
v1->v2: Fix off by one in UP check (Thomas Gleixner)

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/alternative.c | 16 ++++++++++++++--
 include/linux/smp.h           |  2 ++
 init/main.c                   | 16 ++++++++--------
 3 files changed, 24 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index cdc43242da92..318a4f9b7ece 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -273,6 +273,7 @@ struct smp_alt_module {
 };
 static LIST_HEAD(smp_alt_modules);
 static DEFINE_SPINLOCK(smp_alt);
+static int smp_mode = 1;	/* protected by smp_alt */
 
 void alternatives_smp_module_add(struct module *mod, char *name,
 				 void *locks, void *locks_end,
@@ -354,7 +355,14 @@ void alternatives_smp_switch(int smp)
 	BUG_ON(!smp && (num_online_cpus() > 1));
 
 	spin_lock_irqsave(&smp_alt, flags);
-	if (smp) {
+
+	/*
+	 * Avoid unnecessary switches because it forces JIT based VMs to
+	 * throw away all cached translations, which can be quite costly.
+	 */
+	if (smp == smp_mode) {
+		/* nothing */
+	} else if (smp) {
 		printk(KERN_INFO "SMP alternatives: switching to SMP code\n");
 		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
 		clear_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
@@ -369,6 +377,7 @@ void alternatives_smp_switch(int smp)
 			alternatives_smp_unlock(mod->locks, mod->locks_end,
 						mod->text, mod->text_end);
 	}
+	smp_mode = smp;
 	spin_unlock_irqrestore(&smp_alt, flags);
 }
 
@@ -441,7 +450,10 @@ void __init alternative_instructions(void)
 		alternatives_smp_module_add(NULL, "core kernel",
 					    __smp_locks, __smp_locks_end,
 					    _text, _etext);
-		alternatives_smp_switch(0);
+
+		/* Only switch to UP mode if we don't immediately boot others */
+		if (num_possible_cpus() == 1 || setup_max_cpus <= 1)
+			alternatives_smp_switch(0);
 	}
 #endif
  	apply_paravirt(__parainstructions, __parainstructions_end);
diff --git a/include/linux/smp.h b/include/linux/smp.h
index c25e66bcecf3..55232ccf9cfd 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -78,6 +78,8 @@ int on_each_cpu(void (*func) (void *info), void *info, int retry, int wait);
  */
 void smp_prepare_boot_cpu(void);
 
+extern unsigned int setup_max_cpus;
+
 #else /* !SMP */
 
 /*
diff --git a/init/main.c b/init/main.c
index 3f8aba291ed3..5843fe996703 100644
--- a/init/main.c
+++ b/init/main.c
@@ -128,7 +128,7 @@ static char *ramdisk_execute_command;
 
 #ifdef CONFIG_SMP
 /* Setup configured maximum number of CPUs to activate */
-static unsigned int __initdata max_cpus = NR_CPUS;
+unsigned int __initdata setup_max_cpus = NR_CPUS;
 
 /*
  * Setup routine for controlling SMP activation
@@ -146,7 +146,7 @@ static inline void disable_ioapic_setup(void) {};
 
 static int __init nosmp(char *str)
 {
-	max_cpus = 0;
+	setup_max_cpus = 0;
 	disable_ioapic_setup();
 	return 0;
 }
@@ -155,8 +155,8 @@ early_param("nosmp", nosmp);
 
 static int __init maxcpus(char *str)
 {
-	get_option(&str, &max_cpus);
-	if (max_cpus == 0)
+	get_option(&str, &setup_max_cpus);
+	if (setup_max_cpus == 0)
 		disable_ioapic_setup();
 
 	return 0;
@@ -164,7 +164,7 @@ static int __init maxcpus(char *str)
 
 early_param("maxcpus", maxcpus);
 #else
-#define max_cpus NR_CPUS
+#define setup_max_cpus NR_CPUS
 #endif
 
 /*
@@ -393,7 +393,7 @@ static void __init smp_init(void)
 
 	/* FIXME: This should be done in userspace --RR */
 	for_each_present_cpu(cpu) {
-		if (num_online_cpus() >= max_cpus)
+		if (num_online_cpus() >= setup_max_cpus)
 			break;
 		if (!cpu_online(cpu))
 			cpu_up(cpu);
@@ -401,7 +401,7 @@ static void __init smp_init(void)
 
 	/* Any cleanup work */
 	printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
-	smp_cpus_done(max_cpus);
+	smp_cpus_done(setup_max_cpus);
 }
 
 #endif
@@ -824,7 +824,7 @@ static int __init kernel_init(void * unused)
 	__set_special_pids(1, 1);
 	cad_pid = task_pid(current);
 
-	smp_prepare_cpus(max_cpus);
+	smp_prepare_cpus(setup_max_cpus);
 
 	do_pre_smp_initcalls();
 
-- 
cgit v1.2.3


From 03252919b79891063cf99145612360efbdf9500b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:18 +0100
Subject: x86: print which shared library/executable faulted in segfault etc.
 messages v3

They now look like:

hal-resmgr[13791]: segfault at 3c rip 2b9c8caec182 rsp 7fff1e825d30 error 4 in libacl.so.1.1.0[2b9c8caea000+6000]

This makes it easier to pinpoint bugs to specific libraries.

And printing the offset into a mapping also always allows to find the
correct fault point in a library even with randomized mappings. Previously
there was no way to actually find the correct code address inside
the randomized mapping.

Relies on earlier patch to shorten the printk formats.

They are often now longer than 80 characters, but I think that's worth it.

[includes fix from Eric Dumazet to check d_path error value]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/signal_32.c |  7 +++++--
 arch/x86/kernel/signal_64.c |  7 +++++--
 arch/x86/kernel/traps_32.c  |  7 +++++--
 arch/x86/kernel/traps_64.c  | 14 ++++++++++----
 arch/x86/mm/fault_32.c      |  4 +++-
 arch/x86/mm/fault_64.c      |  4 +++-
 include/linux/mm.h          |  1 +
 mm/memory.c                 | 31 +++++++++++++++++++++++++++++++
 8 files changed, 63 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 74df55895c8c..89a690edf999 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -198,12 +198,15 @@ asmlinkage int sys_sigreturn(unsigned long __unused)
 	return ax;
 
 badframe:
-	if (show_unhandled_signals && printk_ratelimit())
+	if (show_unhandled_signals && printk_ratelimit()) {
 		printk("%s%s[%d] bad frame in sigreturn frame:%p ip:%lx"
-		       " sp:%lx oeax:%lx\n",
+		       " sp:%lx oeax:%lx",
 		    task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG,
 		    current->comm, task_pid_nr(current), frame, regs->ip,
 		    regs->sp, regs->orig_ax);
+		print_vma_addr(" in ", regs->ip);
+		printk("\n");
+	}
 
 	force_sig(SIGSEGV, current);
 	return 0;
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 4eb751c60390..7347bb14e306 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -484,9 +484,12 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
 { 
 	struct task_struct *me = current; 
-	if (show_unhandled_signals && printk_ratelimit())
-		printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx\n",
+	if (show_unhandled_signals && printk_ratelimit()) {
+		printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
 	       me->comm,me->pid,where,frame,regs->ip,regs->sp,regs->orig_ax);
+		print_vma_addr(" in ", regs->ip);
+		printk("\n");
+	}
 
 	force_sig(SIGSEGV, me); 
 } 
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 6f3bb287c702..270cfd483160 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -609,11 +609,14 @@ void __kprobes do_general_protection(struct pt_regs * regs,
 	current->thread.error_code = error_code;
 	current->thread.trap_no = 13;
 	if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) &&
-	    printk_ratelimit())
+	    printk_ratelimit()) {
 		printk(KERN_INFO
-		    "%s[%d] general protection ip:%lx sp:%lx error:%lx\n",
+		    "%s[%d] general protection ip:%lx sp:%lx error:%lx",
 		    current->comm, task_pid_nr(current),
 		    regs->ip, regs->sp, error_code);
+		print_vma_addr(" in ", regs->ip);
+		printk("\n");
+	}
 
 	force_sig(SIGSEGV, current);
 	return;
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 814801f4eb9e..911ed28afff8 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -642,11 +642,14 @@ static void __kprobes do_trap(int trapnr, int signr, char *str,
 		tsk->thread.trap_no = trapnr;
 
 		if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
-		    printk_ratelimit())
+		    printk_ratelimit()) {
 			printk(KERN_INFO
-			       "%s[%d] trap %s ip:%lx sp:%lx error:%lx\n",
+			       "%s[%d] trap %s ip:%lx sp:%lx error:%lx",
 			       tsk->comm, tsk->pid, str,
 			       regs->ip, regs->sp, error_code);
+			print_vma_addr(" in ", regs->ip);
+			printk("\n");
+		}
 
 		if (info)
 			force_sig_info(signr, info, tsk);
@@ -741,11 +744,14 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 		tsk->thread.trap_no = 13;
 
 		if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
-		    printk_ratelimit())
+		    printk_ratelimit()) {
 			printk(KERN_INFO
-		       "%s[%d] general protection ip:%lx sp:%lx error:%lx\n",
+		       "%s[%d] general protection ip:%lx sp:%lx error:%lx",
 			       tsk->comm, tsk->pid,
 			       regs->ip, regs->sp, error_code);
+			print_vma_addr(" in ", regs->ip);
+			printk("\n");
+		}
 
 		force_sig(SIGSEGV, tsk);
 		return;
diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 13d295506a17..276863dc4bdd 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -514,11 +514,13 @@ bad_area_nosemaphore:
 #ifdef CONFIG_X86_32
 			"%s%s[%d]: segfault at %lx ip %08lx sp %08lx error %lx",
 #else
-			"%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx\n",
+			"%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx",
 #endif
 			task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
 			tsk->comm, task_pid_nr(tsk), address, regs->ip,
 			regs->sp, error_code);
+			print_vma_addr(" in ", regs->ip);
+			printk("\n");
 		}
 		tsk->thread.cr2 = address;
 		/* Kernel addresses are always protection faults */
diff --git a/arch/x86/mm/fault_64.c b/arch/x86/mm/fault_64.c
index b606bdefbb72..9ef0306efe9e 100644
--- a/arch/x86/mm/fault_64.c
+++ b/arch/x86/mm/fault_64.c
@@ -552,11 +552,13 @@ bad_area_nosemaphore:
 #ifdef CONFIG_X86_32
 			"%s%s[%d]: segfault at %lx ip %08lx sp %08lx error %lx",
 #else
-			"%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx\n",
+			"%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx",
 #endif
 			task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
 			tsk->comm, task_pid_nr(tsk), address, regs->ip,
 			regs->sp, error_code);
+			print_vma_addr(" in ", regs->ip);
+			printk("\n");
 		}
 
 		tsk->thread.cr2 = address;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 1897ca223eca..3c22d971afa7 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1146,6 +1146,7 @@ extern int randomize_va_space;
 #endif
 
 const char * arch_vma_name(struct vm_area_struct *vma);
+void print_vma_addr(char *prefix, unsigned long rip);
 
 struct page *sparse_mem_map_populate(unsigned long pnum, int nid);
 pgd_t *vmemmap_pgd_populate(unsigned long addr, int node);
diff --git a/mm/memory.c b/mm/memory.c
index 673ebbf499c7..d902d0e25edc 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2754,3 +2754,34 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 
 	return buf - old_buf;
 }
+
+/*
+ * Print the name of a VMA.
+ */
+void print_vma_addr(char *prefix, unsigned long ip)
+{
+	struct mm_struct *mm = current->mm;
+	struct vm_area_struct *vma;
+
+	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, ip);
+	if (vma && vma->vm_file) {
+		struct file *f = vma->vm_file;
+		char *buf = (char *)__get_free_page(GFP_KERNEL);
+		if (buf) {
+			char *p, *s;
+
+			p = d_path(f->f_dentry, f->f_vfsmnt, buf, PAGE_SIZE);
+			if (IS_ERR(p))
+				p = "?";
+			s = strrchr(p, '/');
+			if (s)
+				p = s+1;
+			printk("%s%s[%lx+%lx]", prefix, p,
+					vma->vm_start,
+					vma->vm_end - vma->vm_start);
+			free_page((unsigned long)buf);
+		}
+	}
+	up_read(&current->mm->mmap_sem);
+}
-- 
cgit v1.2.3


From 99fc8d424bc5d803fe92cad56c068fe64e73747a Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jesse.barnes@intel.com>
Date: Wed, 30 Jan 2008 13:33:18 +0100
Subject: x86, 32-bit: trim memory not covered by wb mtrrs

On some machines, buggy BIOSes don't properly setup WB MTRRs to cover all
available RAM, meaning the last few megs (or even gigs) of memory will be
marked uncached.  Since Linux tends to allocate from high memory addresses
first, this causes the machine to be unusably slow as soon as the kernel
starts really using memory (i.e.  right around init time).

This patch works around the problem by scanning the MTRRs at boot and
figuring out whether the current end_pfn value (setup by early e820 code)
goes beyond the highest WB MTRR range, and if so, trimming it to match.  A
fairly obnoxious KERN_WARNING is printed too, letting the user know that
not all of their memory is available due to a likely BIOS bug.

Something similar could be done on i386 if needed, but the boot ordering
would be slightly different, since the MTRR code on i386 depends on the
boot_cpu_data structure being setup.

This patch fixes a bug in the last patch that caused the code to run on
non-Intel machines (AMD machines apparently don't need it and it's untested
on other non-Intel machines, so best keep it off).

Further enhancements and fixes from:

  Yinghai Lu <Yinghai.Lu@Sun.COM>
  Andi Kleen <ak@suse.de>

Signed-off-by: Jesse Barnes <jesse.barnes@intel.com>
Tested-by: Justin Piszcz <jpiszcz@lucidpixels.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |   6 ++
 arch/x86/kernel/bugs_64.c           |   1 -
 arch/x86/kernel/cpu/mtrr/generic.c  |   8 +--
 arch/x86/kernel/cpu/mtrr/if.c       |   8 +--
 arch/x86/kernel/cpu/mtrr/main.c     | 140 +++++++++++++++++++++++++++++-------
 arch/x86/kernel/cpu/mtrr/mtrr.h     |   3 +
 arch/x86/kernel/setup_64.c          |   7 ++
 include/asm-x86/mtrr.h              |   6 +-
 8 files changed, 140 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 860a90875491..b8fadf5f75a3 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -570,6 +570,12 @@ and is between 256 and 4096 characters. It is defined in the file
 			See drivers/char/README.epca and
 			Documentation/digiepca.txt.
 
+	disable_mtrr_trim [X86-64, Intel only]
+			By default the kernel will trim any uncacheable
+			memory out of your available memory pool based on
+			MTRR settings.  This parameter disables that behavior,
+			possibly causing your machine to run very slowly.
+
 	dmasound=	[HW,OSS] Sound subsystem buffers
 
 	dscc4.setup=	[NET]
diff --git a/arch/x86/kernel/bugs_64.c b/arch/x86/kernel/bugs_64.c
index 9a189cef6404..8f520f93ffd4 100644
--- a/arch/x86/kernel/bugs_64.c
+++ b/arch/x86/kernel/bugs_64.c
@@ -13,7 +13,6 @@
 void __init check_bugs(void)
 {
 	identify_cpu(&boot_cpu_data);
-	mtrr_bp_init();
 #if !defined(CONFIG_SMP)
 	printk("CPU: ");
 	print_cpu_info(&boot_cpu_data);
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 55d31ff118fb..103d61a59b19 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -14,7 +14,7 @@
 #include "mtrr.h"
 
 struct mtrr_state {
-	struct mtrr_var_range *var_ranges;
+	struct mtrr_var_range var_ranges[MAX_VAR_RANGES];
 	mtrr_type fixed_ranges[NUM_FIXED_RANGES];
 	unsigned char enabled;
 	unsigned char have_fixed;
@@ -86,12 +86,6 @@ void __init get_mtrr_state(void)
 	struct mtrr_var_range *vrs;
 	unsigned lo, dummy;
 
-	if (!mtrr_state.var_ranges) {
-		mtrr_state.var_ranges = kmalloc(num_var_ranges * sizeof (struct mtrr_var_range), 
-						GFP_KERNEL);
-		if (!mtrr_state.var_ranges)
-			return;
-	} 
 	vrs = mtrr_state.var_ranges;
 
 	rdmsr(MTRRcap_MSR, lo, dummy);
diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c
index 14535686c099..91e150acb46c 100644
--- a/arch/x86/kernel/cpu/mtrr/if.c
+++ b/arch/x86/kernel/cpu/mtrr/if.c
@@ -11,10 +11,6 @@
 #include <asm/mtrr.h>
 #include "mtrr.h"
 
-/* RED-PEN: this is accessed without any locking */
-extern unsigned int *usage_table;
-
-
 #define FILE_FCOUNT(f) (((struct seq_file *)((f)->private_data))->private)
 
 static const char *const mtrr_strings[MTRR_NUM_TYPES] =
@@ -397,7 +393,7 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset)
 	for (i = 0; i < max; i++) {
 		mtrr_if->get(i, &base, &size, &type);
 		if (size == 0)
-			usage_table[i] = 0;
+			mtrr_usage_table[i] = 0;
 		else {
 			if (size < (0x100000 >> PAGE_SHIFT)) {
 				/* less than 1MB */
@@ -411,7 +407,7 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset)
 			len += seq_printf(seq, 
 				   "reg%02i: base=0x%05lx000 (%4luMB), size=%4lu%cB: %s, count=%d\n",
 			     i, base, base >> (20 - PAGE_SHIFT), size, factor,
-			     mtrr_attrib_to_str(type), usage_table[i]);
+			     mtrr_attrib_to_str(type), mtrr_usage_table[i]);
 		}
 	}
 	return 0;
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 60af5ed2b5c0..ccd36ed2187b 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -38,8 +38,8 @@
 #include <linux/cpu.h>
 #include <linux/mutex.h>
 
+#include <asm/e820.h>
 #include <asm/mtrr.h>
-
 #include <asm/uaccess.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
@@ -47,7 +47,7 @@
 
 u32 num_var_ranges = 0;
 
-unsigned int *usage_table;
+unsigned int mtrr_usage_table[MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
 
 u64 size_or_mask, size_and_mask;
@@ -121,13 +121,8 @@ static void __init init_table(void)
 	int i, max;
 
 	max = num_var_ranges;
-	if ((usage_table = kmalloc(max * sizeof *usage_table, GFP_KERNEL))
-	    == NULL) {
-		printk(KERN_ERR "mtrr: could not allocate\n");
-		return;
-	}
 	for (i = 0; i < max; i++)
-		usage_table[i] = 1;
+		mtrr_usage_table[i] = 1;
 }
 
 struct set_mtrr_data {
@@ -383,7 +378,7 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 			goto out;
 		}
 		if (increment)
-			++usage_table[i];
+			++mtrr_usage_table[i];
 		error = i;
 		goto out;
 	}
@@ -391,15 +386,15 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 	i = mtrr_if->get_free_region(base, size, replace);
 	if (i >= 0) {
 		set_mtrr(i, base, size, type);
-		if (likely(replace < 0))
-			usage_table[i] = 1;
-		else {
-			usage_table[i] = usage_table[replace];
+		if (likely(replace < 0)) {
+			mtrr_usage_table[i] = 1;
+		} else {
+			mtrr_usage_table[i] = mtrr_usage_table[replace];
 			if (increment)
-				usage_table[i]++;
+				mtrr_usage_table[i]++;
 			if (unlikely(replace != i)) {
 				set_mtrr(replace, 0, 0, 0);
-				usage_table[replace] = 0;
+				mtrr_usage_table[replace] = 0;
 			}
 		}
 	} else
@@ -529,11 +524,11 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 		printk(KERN_WARNING "mtrr: MTRR %d not used\n", reg);
 		goto out;
 	}
-	if (usage_table[reg] < 1) {
+	if (mtrr_usage_table[reg] < 1) {
 		printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
 		goto out;
 	}
-	if (--usage_table[reg] < 1)
+	if (--mtrr_usage_table[reg] < 1)
 		set_mtrr(reg, 0, 0, 0);
 	error = reg;
  out:
@@ -593,16 +588,11 @@ struct mtrr_value {
 	unsigned long	lsize;
 };
 
-static struct mtrr_value * mtrr_state;
+static struct mtrr_value mtrr_state[MAX_VAR_RANGES];
 
 static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
 {
 	int i;
-	int size = num_var_ranges * sizeof(struct mtrr_value);
-
-	mtrr_state = kzalloc(size,GFP_ATOMIC);
-	if (!mtrr_state)
-		return -ENOMEM;
 
 	for (i = 0; i < num_var_ranges; i++) {
 		mtrr_if->get(i,
@@ -624,7 +614,6 @@ static int mtrr_restore(struct sys_device * sysdev)
 				 mtrr_state[i].lsize,
 				 mtrr_state[i].ltype);
 	}
-	kfree(mtrr_state);
 	return 0;
 }
 
@@ -635,6 +624,109 @@ static struct sysdev_driver mtrr_sysdev_driver = {
 	.resume		= mtrr_restore,
 };
 
+#ifdef CONFIG_X86_64
+static int disable_mtrr_trim;
+
+static int __init disable_mtrr_trim_setup(char *str)
+{
+	disable_mtrr_trim = 1;
+	return 0;
+}
+early_param("disable_mtrr_trim", disable_mtrr_trim_setup);
+
+/*
+ * Newer AMD K8s and later CPUs have a special magic MSR way to force WB
+ * for memory >4GB. Check for that here.
+ * Note this won't check if the MTRRs < 4GB where the magic bit doesn't
+ * apply to are wrong, but so far we don't know of any such case in the wild.
+ */
+#define Tom2Enabled (1U << 21)
+#define Tom2ForceMemTypeWB (1U << 22)
+
+static __init int amd_special_default_mtrr(unsigned long end_pfn)
+{
+	u32 l, h;
+
+	/* Doesn't apply to memory < 4GB */
+	if (end_pfn <= (0xffffffff >> PAGE_SHIFT))
+		return 0;
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
+		return 0;
+	if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
+		return 0;
+	/* In case some hypervisor doesn't pass SYSCFG through */
+	if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0)
+		return 0;
+	/*
+	 * Memory between 4GB and top of mem is forced WB by this magic bit.
+	 * Reserved before K8RevF, but should be zero there.
+	 */
+	if ((l & (Tom2Enabled | Tom2ForceMemTypeWB)) ==
+		 (Tom2Enabled | Tom2ForceMemTypeWB))
+		return 1;
+	return 0;
+}
+
+/**
+ * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs
+ *
+ * Some buggy BIOSes don't setup the MTRRs properly for systems with certain
+ * memory configurations.  This routine checks that the highest MTRR matches
+ * the end of memory, to make sure the MTRRs having a write back type cover
+ * all of the memory the kernel is intending to use. If not, it'll trim any
+ * memory off the end by adjusting end_pfn, removing it from the kernel's
+ * allocation pools, warning the user with an obnoxious message.
+ */
+int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
+{
+	unsigned long i, base, size, highest_addr = 0, def, dummy;
+	mtrr_type type;
+	u64 trim_start, trim_size;
+
+	/*
+	 * Make sure we only trim uncachable memory on machines that
+	 * support the Intel MTRR architecture:
+	 */
+	rdmsr(MTRRdefType_MSR, def, dummy);
+	def &= 0xff;
+	if (!is_cpu(INTEL) || disable_mtrr_trim || def != MTRR_TYPE_UNCACHABLE)
+		return 0;
+
+	/* Find highest cached pfn */
+	for (i = 0; i < num_var_ranges; i++) {
+		mtrr_if->get(i, &base, &size, &type);
+		if (type != MTRR_TYPE_WRBACK)
+			continue;
+		base <<= PAGE_SHIFT;
+		size <<= PAGE_SHIFT;
+		if (highest_addr < base + size)
+			highest_addr = base + size;
+	}
+
+	if (amd_special_default_mtrr(end_pfn))
+		return 0;
+
+	if ((highest_addr >> PAGE_SHIFT) < end_pfn) {
+		printk(KERN_WARNING "***************\n");
+		printk(KERN_WARNING "**** WARNING: likely BIOS bug\n");
+		printk(KERN_WARNING "**** MTRRs don't cover all of "
+		       "memory, trimmed %ld pages\n", end_pfn -
+		       (highest_addr >> PAGE_SHIFT));
+		printk(KERN_WARNING "***************\n");
+
+		printk(KERN_INFO "update e820 for mtrr\n");
+		trim_start = highest_addr;
+		trim_size = end_pfn;
+		trim_size <<= PAGE_SHIFT;
+		trim_size -= trim_start;
+		add_memory_region(trim_start, trim_size, E820_RESERVED);
+		update_e820();
+		return 1;
+	}
+
+	return 0;
+}
+#endif
 
 /**
  * mtrr_bp_init - initialize mtrrs on the boot CPU
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 54347e9a95c0..fb74a2c20814 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -12,6 +12,7 @@
 #define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
 
 #define NUM_FIXED_RANGES 88
+#define MAX_VAR_RANGES 256
 #define MTRRfix64K_00000_MSR 0x250
 #define MTRRfix16K_80000_MSR 0x258
 #define MTRRfix16K_A0000_MSR 0x259
@@ -32,6 +33,8 @@
    an 8 bit field: */
 typedef u8 mtrr_type;
 
+extern unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+
 struct mtrr_ops {
 	u32	vendor;
 	u32	use_intel_if;
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 6cbd15625dce..12948316e6a6 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -310,6 +310,13 @@ void __init setup_arch(char **cmdline_p)
 	 * we are rounding upwards:
 	 */
 	end_pfn = e820_end_of_ram();
+	/* update e820 for memory not covered by WB MTRRs */
+	mtrr_bp_init();
+	if (mtrr_trim_uncached_memory(end_pfn)) {
+		e820_register_active_regions(0, 0, -1UL);
+		end_pfn = e820_end_of_ram();
+	}
+
 	num_physpages = end_pfn;
 
 	check_efer();
diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index 262670e42078..319d065800be 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -97,6 +97,7 @@ extern int mtrr_del_page (int reg, unsigned long base, unsigned long size);
 extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi);
 extern void mtrr_ap_init(void);
 extern void mtrr_bp_init(void);
+extern int mtrr_trim_uncached_memory(unsigned long end_pfn);
 #  else
 #define mtrr_save_fixed_ranges(arg) do {} while (0)
 #define mtrr_save_state() do {} while (0)
@@ -120,7 +121,10 @@ static __inline__ int mtrr_del_page (int reg, unsigned long base,
 {
     return -ENODEV;
 }
-
+static inline int mtrr_trim_uncached_memory(unsigned long end_pfn)
+{
+	return 0;
+}
 static __inline__ void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) {;}
 
 #define mtrr_ap_init() do {} while (0)
-- 
cgit v1.2.3


From 2274c33ebdfbb848e7291849e662d36519b52bf4 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Wed, 30 Jan 2008 13:33:18 +0100
Subject: x86: msr for AMD Fam 10h mmio

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/msr-index.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index 4045bbe47379..fae118a25278 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -95,6 +95,14 @@
 #define MSR_AMD64_IBSDCPHYSAD		0xc0011039
 #define MSR_AMD64_IBSCTL		0xc001103a
 
+/* Fam 10h MSRs */
+#define MSR_FAM10H_MMIO_CONF_BASE	0xc0010058
+#define FAM10H_MMIO_CONF_ENABLE		(1<<0)
+#define FAM10H_MMIO_CONF_BUSRANGE_MASK	0xf
+#define FAM10H_MMIO_CONF_BUSRANGE_SHIFT 2
+#define FAM10H_MMIO_CONF_BASE_MASK	0xfffffff
+#define FAM10H_MMIO_CONF_BASE_SHIFT	20
+
 /* K8 MSRs */
 #define MSR_K8_TOP_MEM1			0xc001001a
 #define MSR_K8_TOP_MEM2			0xc001001d
-- 
cgit v1.2.3


From b03878307a78ba0248f6f8d24279c846f1a26f6e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:33:19 +0100
Subject: x86: provide a native_init_IRQ function on 64-bit

x86_64 lacks a native_init_IRQ() function, so we turn the arch's
init_IRQ() function into a native construct

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/i8259_64.c  | 4 +++-
 include/asm-x86/hw_irq_64.h | 1 +
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/i8259_64.c b/arch/x86/kernel/i8259_64.c
index d3edb9f23f2c..fa57a1568508 100644
--- a/arch/x86/kernel/i8259_64.c
+++ b/arch/x86/kernel/i8259_64.c
@@ -456,7 +456,9 @@ void __init init_ISA_irqs (void)
 	}
 }
 
-void __init init_IRQ(void)
+void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
+
+void __init native_init_IRQ(void)
 {
 	int i;
 
diff --git a/include/asm-x86/hw_irq_64.h b/include/asm-x86/hw_irq_64.h
index a346159b6ac3..312a58d6dac6 100644
--- a/include/asm-x86/hw_irq_64.h
+++ b/include/asm-x86/hw_irq_64.h
@@ -141,6 +141,7 @@ extern void print_IO_APIC(void);
 extern int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn);
 extern void send_IPI(int dest, int vector);
 extern void setup_ioapic_dest(void);
+extern void native_init_IRQ(void);
 
 extern unsigned long io_apic_irqs;
 
-- 
cgit v1.2.3


From 046627130591993fd04acef9aebccf70fdad3702 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:33:19 +0100
Subject: x86: put generic mm_hooks include into PARAVIRT

With PARAVIRT, we actually have arch_{dup,exit}_mmap functions,
so we can't include the generic header

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/mmu_context_64.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h
index 7e2aa23fccbf..ad6dc821ef9e 100644
--- a/include/asm-x86/mmu_context_64.h
+++ b/include/asm-x86/mmu_context_64.h
@@ -7,7 +7,9 @@
 #include <asm/pda.h>
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
+#ifndef CONFIG_PARAVIRT
 #include <asm-generic/mm_hooks.h>
+#endif
 
 /*
  * possibly do the LDT unload here?
-- 
cgit v1.2.3


From 4c9890c246121d070deb8cf5cf53e80caffc4dde Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:33:19 +0100
Subject: x86: puts read and write cr8 into pv_cpu_ops

This patch adds room for read and write_cr8 functions back in
pv_cpu_ops struct

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index c56b17a5eba0..73547acbbbf5 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -101,6 +101,11 @@ struct pv_cpu_ops {
 	unsigned long (*read_cr4)(void);
 	void (*write_cr4)(unsigned long);
 
+#ifdef CONFIG_X86_64
+	unsigned long (*read_cr8)(void);
+	void (*write_cr8)(unsigned long);
+#endif
+
 	/* Segment descriptor handling */
 	void (*load_tr_desc)(void);
 	void (*load_gdt)(const struct desc_ptr *);
@@ -614,6 +619,16 @@ static inline void write_cr4(unsigned long x)
 	PVOP_VCALL1(pv_cpu_ops.write_cr4, x);
 }
 
+static inline unsigned long read_cr8(void)
+{
+	return PVOP_CALL0(unsigned long, pv_cpu_ops.read_cr8);
+}
+
+static inline void write_cr8(unsigned long x)
+{
+	PVOP_VCALL1(pv_cpu_ops.write_cr8, x);
+}
+
 static inline void raw_safe_halt(void)
 {
 	PVOP_VCALL0(pv_irq_ops.safe_halt);
-- 
cgit v1.2.3


From 94ea03cdda520f0c0dc80cbb4674ab9a33749ee2 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:33:19 +0100
Subject: x86: provide read and write cr8 paravirt hooks

Since the cr8 manipulation functions ended up staying in the tree,
they can't be defined just when PARAVIRT is off: In this patch,
those functions are defined for the PARAVIRT case too.

[ mingo@elte.hu: fixes ]

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h |  2 ++
 include/asm-x86/system.h   | 30 ++++++++++++++++--------------
 2 files changed, 18 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 73547acbbbf5..a55f95e921ee 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -619,6 +619,7 @@ static inline void write_cr4(unsigned long x)
 	PVOP_VCALL1(pv_cpu_ops.write_cr4, x);
 }
 
+#ifdef CONFIG_X86_64
 static inline unsigned long read_cr8(void)
 {
 	return PVOP_CALL0(unsigned long, pv_cpu_ops.read_cr8);
@@ -628,6 +629,7 @@ static inline void write_cr8(unsigned long x)
 {
 	PVOP_VCALL1(pv_cpu_ops.write_cr8, x);
 }
+#endif
 
 static inline void raw_safe_halt(void)
 {
diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 39474f2957a3..ee32ef9367f4 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -231,6 +231,20 @@ static inline void native_write_cr4(unsigned long val)
 	asm volatile("mov %0,%%cr4": :"r" (val), "m" (__force_order));
 }
 
+#ifdef CONFIG_X86_64
+static inline unsigned long native_read_cr8(void)
+{
+	unsigned long cr8;
+	asm volatile("movq %%cr8,%0" : "=r" (cr8));
+	return cr8;
+}
+
+static inline void native_write_cr8(unsigned long val)
+{
+	asm volatile("movq %0,%%cr8" :: "r" (val) : "memory");
+}
+#endif
+
 static inline void native_wbinvd(void)
 {
 	asm volatile("wbinvd": : :"memory");
@@ -248,21 +262,9 @@ static inline void native_wbinvd(void)
 #define read_cr4_safe()	(native_read_cr4_safe())
 #define write_cr4(x)	(native_write_cr4(x))
 #define wbinvd()	(native_wbinvd())
-
 #ifdef CONFIG_X86_64
-
-static inline unsigned long read_cr8(void)
-{
-	unsigned long cr8;
-	asm volatile("movq %%cr8,%0" : "=r" (cr8));
-	return cr8;
-}
-
-static inline void write_cr8(unsigned long val)
-{
-	asm volatile("movq %0,%%cr8" :: "r" (val) : "memory");
-}
-
+#define read_cr8()	(native_read_cr8())
+#define write_cr8(x)	(native_write_cr8(x))
 #endif
 
 /* Clear the 'TS' bit */
-- 
cgit v1.2.3


From 1fe91514a3dc07ab540bc891589c46aaa47e92f6 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 30 Jan 2008 13:33:19 +0100
Subject: x86: change function orders in paravirt.h

__pmd, pmd_val and set_pud are used before they are defined (as static)
We move them a little up in the file, so it doesn't happen.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 84 +++++++++++++++++++++++-----------------------
 1 file changed, 42 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index a55f95e921ee..62cecb7778ee 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -1025,6 +1025,48 @@ static inline void set_pmd(pmd_t *pmdp, pmd_t pmd)
 		PVOP_VCALL2(pv_mmu_ops.set_pmd, pmdp, val);
 }
 
+#if PAGETABLE_LEVELS >= 3
+static inline pmd_t __pmd(pmdval_t val)
+{
+	pmdval_t ret;
+
+	if (sizeof(pmdval_t) > sizeof(long))
+		ret = PVOP_CALL2(pmdval_t, pv_mmu_ops.make_pmd,
+				 val, (u64)val >> 32);
+	else
+		ret = PVOP_CALL1(pmdval_t, pv_mmu_ops.make_pmd,
+				 val);
+
+	return (pmd_t) { ret };
+}
+
+static inline pmdval_t pmd_val(pmd_t pmd)
+{
+	pmdval_t ret;
+
+	if (sizeof(pmdval_t) > sizeof(long))
+		ret =  PVOP_CALL2(pmdval_t, pv_mmu_ops.pmd_val,
+				  pmd.pmd, (u64)pmd.pmd >> 32);
+	else
+		ret =  PVOP_CALL1(pmdval_t, pv_mmu_ops.pmd_val,
+				  pmd.pmd);
+
+	return ret;
+}
+
+static inline void set_pud(pud_t *pudp, pud_t pud)
+{
+	pudval_t val = native_pud_val(pud);
+
+	if (sizeof(pudval_t) > sizeof(long))
+		PVOP_VCALL3(pv_mmu_ops.set_pud, pudp,
+			    val, (u64)val >> 32);
+	else
+		PVOP_VCALL2(pv_mmu_ops.set_pud, pudp,
+			    val);
+}
+#endif	/* PAGETABLE_LEVELS >= 3 */
+
 #ifdef CONFIG_X86_PAE
 /* Special-case pte-setting operations for PAE, which can't update a
    64-bit pte atomically */
@@ -1075,48 +1117,6 @@ static inline void pmd_clear(pmd_t *pmdp)
 }
 #endif	/* CONFIG_X86_PAE */
 
-#if PAGETABLE_LEVELS >= 3
-static inline pmd_t __pmd(pmdval_t val)
-{
-	pmdval_t ret;
-
-	if (sizeof(pmdval_t) > sizeof(long))
-		ret = PVOP_CALL2(pmdval_t, pv_mmu_ops.make_pmd,
-				 val, (u64)val >> 32);
-	else
-		ret = PVOP_CALL1(pmdval_t, pv_mmu_ops.make_pmd,
-				 val);
-
-	return (pmd_t) { ret };
-}
-
-static inline pmdval_t pmd_val(pmd_t pmd)
-{
-	pmdval_t ret;
-
-	if (sizeof(pmdval_t) > sizeof(long))
-		ret =  PVOP_CALL2(pmdval_t, pv_mmu_ops.pmd_val,
-				  pmd.pmd, (u64)pmd.pmd >> 32);
-	else
-		ret =  PVOP_CALL1(pmdval_t, pv_mmu_ops.pmd_val,
-				  pmd.pmd);
-
-	return ret;
-}
-
-static inline void set_pud(pud_t *pudp, pud_t pud)
-{
-	pudval_t val = native_pud_val(pud);
-
-	if (sizeof(pudval_t) > sizeof(long))
-		PVOP_VCALL3(pv_mmu_ops.set_pud, pudp,
-			    val, (u64)val >> 32);
-	else
-		PVOP_VCALL2(pv_mmu_ops.set_pud, pudp,
-			    val);
-}
-#endif	/* PAGETABLE_LEVELS >= 3 */
-
 /* Lazy mode for batching updates / context switch */
 enum paravirt_lazy_mode {
 	PARAVIRT_LAZY_NONE,
-- 
cgit v1.2.3


From 9042219cd8d43b81322b826d463dd6e52acae6cf Mon Sep 17 00:00:00 2001
From: Eduardo Habkost <ehabkost@redhat.com>
Date: Wed, 30 Jan 2008 13:33:20 +0100
Subject: x86: include/asm-x86/paravirt.h: x86_64 mmu operations

Add .set_pgd field to pv_mmu_ops.

Implement pud_val(), __pud(), set_pgd(), pud_clear(), pgd_clear().

pud_clear() and pgd_clear() are implemented simply using set_pud()
and set_pmd(). They don't have a field at pv_mmu_ops.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/paravirt.h | 55 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 55 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 62cecb7778ee..9f9ff57b3efb 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -259,6 +259,8 @@ struct pv_mmu_ops {
 #if PAGETABLE_LEVELS == 4
 	pudval_t (*pud_val)(pud_t);
 	pud_t (*make_pud)(pudval_t pud);
+
+	void (*set_pgd)(pgd_t *pudp, pgd_t pgdval);
 #endif	/* PAGETABLE_LEVELS == 4 */
 #endif	/* PAGETABLE_LEVELS >= 3 */
 
@@ -1065,6 +1067,59 @@ static inline void set_pud(pud_t *pudp, pud_t pud)
 		PVOP_VCALL2(pv_mmu_ops.set_pud, pudp,
 			    val);
 }
+#if PAGETABLE_LEVELS == 4
+static inline pud_t __pud(pudval_t val)
+{
+	pudval_t ret;
+
+	if (sizeof(pudval_t) > sizeof(long))
+		ret = PVOP_CALL2(pudval_t, pv_mmu_ops.make_pud,
+				 val, (u64)val >> 32);
+	else
+		ret = PVOP_CALL1(pudval_t, pv_mmu_ops.make_pud,
+				 val);
+
+	return (pud_t) { ret };
+}
+
+static inline pudval_t pud_val(pud_t pud)
+{
+	pudval_t ret;
+
+	if (sizeof(pudval_t) > sizeof(long))
+		ret =  PVOP_CALL2(pudval_t, pv_mmu_ops.pud_val,
+				  pud.pud, (u64)pud.pud >> 32);
+	else
+		ret =  PVOP_CALL1(pudval_t, pv_mmu_ops.pud_val,
+				  pud.pud);
+
+	return ret;
+}
+
+static inline void set_pgd(pgd_t *pgdp, pgd_t pgd)
+{
+	pgdval_t val = native_pgd_val(pgd);
+
+	if (sizeof(pgdval_t) > sizeof(long))
+		PVOP_VCALL3(pv_mmu_ops.set_pgd, pgdp,
+			    val, (u64)val >> 32);
+	else
+		PVOP_VCALL2(pv_mmu_ops.set_pgd, pgdp,
+			    val);
+}
+
+static inline void pgd_clear(pgd_t *pgdp)
+{
+	set_pgd(pgdp, __pgd(0));
+}
+
+static inline void pud_clear(pud_t *pudp)
+{
+	set_pud(pudp, __pud(0));
+}
+
+#endif	/* PAGETABLE_LEVELS == 4 */
+
 #endif	/* PAGETABLE_LEVELS >= 3 */
 
 #ifdef CONFIG_X86_PAE
-- 
cgit v1.2.3


From 7d851c8d3db0f79b92c8b14361779ede8acd2488 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:20 +0100
Subject: x86: add framework to disable CPUID bits on the command line

There are already various options to disable specific cpuid bits
on the command line. They all use their own variable. Add a generic
mask to make this easier in the future.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/common.c | 6 ++++++
 arch/x86/kernel/setup_64.c   | 6 ++++++
 include/asm-x86/cpufeature.h | 4 ++++
 include/asm-x86/processor.h  | 1 +
 4 files changed, 17 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 4bd326d0322c..f0f29ddf33a2 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -57,6 +57,8 @@ DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
 } };
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
+__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
+
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_fxsr __cpuinitdata;
 static int disable_x86_serial_nr __cpuinitdata = 1;
@@ -497,6 +499,10 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
 	}
 
+	/* Clear all flags overriden by options */
+	for (i = 0; i < NCAPINTS; i++)
+		c->x86_capability[i] ^= cleared_cpu_caps[i];
+
 	/* Init Machine Check Exception if available. */
 	mcheck_init(c);
 
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 7edb43f0b279..df159520bbd2 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -80,6 +80,8 @@
 struct cpuinfo_x86 boot_cpu_data __read_mostly;
 EXPORT_SYMBOL(boot_cpu_data);
 
+__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
+
 unsigned long mmu_cr4_features;
 
 /* Boot loader ID as an integer, for the benefit of proc_dointvec */
@@ -1013,6 +1015,10 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
 	}
 
+	/* Clear all flags overriden by options */
+	for (i = 0; i < NCAPINTS; i++)
+		c->x86_capability[i] ^= cleared_cpu_caps[i];
+
 #ifdef CONFIG_X86_MCE
 	mcheck_init(c);
 #endif
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 29727bf0e177..b8f53f869e1f 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -131,6 +131,10 @@
 
 #define set_cpu_cap(c, bit)	set_bit(bit, (unsigned long *)((c)->x86_capability))
 #define clear_cpu_cap(c, bit)	clear_bit(bit, (unsigned long *)((c)->x86_capability))
+#define setup_clear_cpu_cap(bit) do { \
+	clear_cpu_cap(&boot_cpu_data, bit);	\
+	set_bit(bit, cleared_cpu_caps); 	\
+} while (0)
 
 #define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
 #define cpu_has_vme		boot_cpu_has(X86_FEATURE_VME)
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 81ecfed83e47..ab4d0c2a3f8f 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -118,6 +118,7 @@ struct cpuinfo_x86 {
 extern struct cpuinfo_x86 boot_cpu_data;
 extern struct cpuinfo_x86 new_cpu_data;
 extern struct tss_struct doublefault_tss;
+extern __u32 cleared_cpu_caps[NCAPINTS];
 
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
-- 
cgit v1.2.3


From 404ee5b14b68d3cba287c2596588b83790c49f7b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:20 +0100
Subject: x86: convert TSC disabling to generic cpuid disable bitmap

Fix from: Ian Campbell <ijc@hellion.org.uk>

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/bugs.c    |  2 +-
 arch/x86/kernel/cpu/common.c  |  9 ---------
 arch/x86/kernel/numaq_32.c    |  2 +-
 arch/x86/kernel/tsc_32.c      | 14 +++-----------
 arch/x86/mach-voyager/setup.c |  2 +-
 arch/x86/xen/time.c           |  2 +-
 include/asm-x86/cpufeature.h  |  4 ++++
 include/asm-x86/tsc.h         |  2 --
 8 files changed, 11 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index a96abd453e0d..9b95edcfc6ae 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -154,7 +154,7 @@ static void __init check_config(void)
  * If we configured ourselves for a TSC, we'd better have one!
  */
 #ifdef CONFIG_X86_TSC
-	if (!cpu_has_tsc && !tsc_disable)
+	if (!cpu_has_tsc)
 		panic("Kernel compiled for Pentium+, requires TSC feature!");
 #endif
 
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index c66991a04a8a..dfc9563fc4f0 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -446,10 +446,6 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	 * we do "generic changes."
 	 */
 
-	/* TSC disabled? */
-	if ( tsc_disable )
-		clear_bit(X86_FEATURE_TSC, c->x86_capability);
-
 	/* If the model name is still unset, do table lookup. */
 	if ( !c->x86_model_id[0] ) {
 		char *p;
@@ -650,11 +646,6 @@ void __cpuinit cpu_init(void)
 
 	if (cpu_has_vme || cpu_has_tsc || cpu_has_de)
 		clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
-	if (tsc_disable && cpu_has_tsc) {
-		printk(KERN_NOTICE "Disabling TSC...\n");
-		/**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/
-		clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability);
-	}
 
 	load_idt(&idt_descr);
 	switch_to_new_gdt();
diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c
index 9000d82c6dc0..e65281b1634b 100644
--- a/arch/x86/kernel/numaq_32.c
+++ b/arch/x86/kernel/numaq_32.c
@@ -82,7 +82,7 @@ static int __init numaq_tsc_disable(void)
 {
 	if (num_online_nodes() > 1) {
 		printk(KERN_DEBUG "NUMAQ: disabling TSC\n");
-		tsc_disable = 1;
+		setup_clear_cpu_cap(X86_FEATURE_TSC);
 	}
 	return 0;
 }
diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index 2a7b95bd8509..43517e324be8 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -24,8 +24,6 @@ static int tsc_enabled;
 unsigned int tsc_khz;
 EXPORT_SYMBOL_GPL(tsc_khz);
 
-int tsc_disable;
-
 #ifdef CONFIG_X86_TSC
 static int __init tsc_setup(char *str)
 {
@@ -40,8 +38,7 @@ static int __init tsc_setup(char *str)
  */
 static int __init tsc_setup(char *str)
 {
-	tsc_disable = 1;
-
+	setup_clear_cpu_cap(X86_FEATURE_TSC);
 	return 1;
 }
 #endif
@@ -395,7 +392,7 @@ void __init tsc_init(void)
 {
 	int cpu;
 
-	if (!cpu_has_tsc || tsc_disable)
+	if (!cpu_has_tsc)
 		goto out_no_tsc;
 
 	cpu_khz = calculate_cpu_khz();
@@ -439,10 +436,5 @@ void __init tsc_init(void)
 	return;
 
 out_no_tsc:
-	/*
-	 * Set the tsc_disable flag if there's no TSC support, this
-	 * makes it a fast flag for the kernel to see whether it
-	 * should be using the TSC.
-	 */
-	tsc_disable = 1;
+	setup_clear_cpu_cap(X86_FEATURE_TSC);
 }
diff --git a/arch/x86/mach-voyager/setup.c b/arch/x86/mach-voyager/setup.c
index 81257a861984..5ae5466b9eb9 100644
--- a/arch/x86/mach-voyager/setup.c
+++ b/arch/x86/mach-voyager/setup.c
@@ -37,7 +37,7 @@ void __init pre_setup_arch_hook(void)
 {
 	/* Voyagers run their CPUs from independent clocks, so disable
 	 * the TSC code because we can't sync them */
-	tsc_disable = 1;
+	setup_clear_cpu_cap(X86_FEATURE_TSC);
 }
 
 void __init trap_init_hook(void)
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index d083ff5ef088..b3721fd6877b 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -592,7 +592,7 @@ __init void xen_time_init(void)
 	set_normalized_timespec(&wall_to_monotonic,
 				-xtime.tv_sec, -xtime.tv_nsec);
 
-	tsc_disable = 0;
+	setup_force_cpu_cap(X86_FEATURE_TSC);
 
 	xen_setup_timer(cpu);
 	xen_setup_cpu_clockevents();
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index b8f53f869e1f..3fb7dfa7fc91 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -135,6 +135,10 @@
 	clear_cpu_cap(&boot_cpu_data, bit);	\
 	set_bit(bit, cleared_cpu_caps); 	\
 } while (0)
+#define setup_force_cpu_cap(bit) do { \
+	set_cpu_cap(&boot_cpu_data, bit);	\
+	clear_bit(bit, cleared_cpu_caps); 	\
+} while (0)
 
 #define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
 #define cpu_has_vme		boot_cpu_has(X86_FEATURE_VME)
diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index 071e0ce5b664..a6e8d35c3f86 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -16,8 +16,6 @@ typedef unsigned long long cycles_t;
 
 extern unsigned int cpu_khz;
 extern unsigned int tsc_khz;
-/* flag for disabling the tsc */
-extern int tsc_disable;
 
 extern void disable_TSC(void);
 
-- 
cgit v1.2.3


From 834beda15ecc43c110c0a6ac39ec1aa79f891716 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:21 +0100
Subject: x86: change NR_CPUS arrays in numa_64 fixup

Change the following static arrays sized by NR_CPUS to
per_cpu data variables:

	char cpu_to_node_map[NR_CPUS];

fixup:

  - Split cpu_to_node function into "early" and "late" versions
    so that x86_cpu_to_node_map_early_ptr is not EXPORT'ed and
    the cpu_to_node inline function is more streamlined.

  - This also involves setting up the percpu maps as early as possible.

  - Fix X86_32 NUMA build errors that previous version of this
    patch caused.

V2->V3:
    - add early_cpu_to_node function to keep cpu_to_node efficient
    - move and rename smp_set_apicids() to setup_percpu_maps()
    - call setup_percpu_maps() as early as possible

V1->V2:
    - Removed extraneous casts
    - Fix !NUMA builds with '#ifdef CONFIG_NUMA"

Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup64.c    | 41 +++++++++++++++++++++++++++++++++++++++--
 arch/x86/kernel/smpboot_32.c |  2 +-
 arch/x86/kernel/smpboot_64.c | 34 ----------------------------------
 arch/x86/mm/srat_64.c        |  5 +++--
 include/asm-x86/topology.h   | 23 +++++++++++++++++++++++
 5 files changed, 66 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index 8fa0de810d0b..855ec82e4f76 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -83,6 +83,40 @@ static int __init nonx32_setup(char *str)
 }
 __setup("noexec32=", nonx32_setup);
 
+/*
+ * Copy data used in early init routines from the initial arrays to the
+ * per cpu data areas.  These arrays then become expendable and the *_ptrs
+ * are zeroed indicating that the static arrays are gone.
+ */
+void __init setup_percpu_maps(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+#ifdef CONFIG_SMP
+		if (per_cpu_offset(cpu)) {
+#endif
+			per_cpu(x86_cpu_to_apicid, cpu) =
+						x86_cpu_to_apicid_init[cpu];
+#ifdef CONFIG_NUMA
+			per_cpu(x86_cpu_to_node_map, cpu) =
+						x86_cpu_to_node_map_init[cpu];
+#endif
+#ifdef CONFIG_SMP
+		}
+		else
+			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
+									cpu);
+#endif
+	}
+
+	/* indicate the early static arrays are gone */
+	x86_cpu_to_apicid_early_ptr = NULL;
+#ifdef CONFIG_NUMA
+	x86_cpu_to_node_map_early_ptr = NULL;
+#endif
+}
+
 /*
  * Great future plan:
  * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
@@ -104,18 +138,21 @@ void __init setup_per_cpu_areas(void)
 	for_each_cpu_mask (i, cpu_possible_map) {
 		char *ptr;
 
-		if (!NODE_DATA(cpu_to_node(i))) {
+		if (!NODE_DATA(early_cpu_to_node(i))) {
 			printk("cpu with no node %d, num_online_nodes %d\n",
 			       i, num_online_nodes());
 			ptr = alloc_bootmem_pages(size);
 		} else { 
-			ptr = alloc_bootmem_pages_node(NODE_DATA(cpu_to_node(i)), size);
+			ptr = alloc_bootmem_pages_node(NODE_DATA(early_cpu_to_node(i)), size);
 		}
 		if (!ptr)
 			panic("Cannot allocate cpu data for CPU %d\n", i);
 		cpu_pda(i)->data_offset = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 	}
+
+	/* setup percpu data maps early */
+	setup_percpu_maps();
 } 
 
 void pda_init(int cpu)
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 915ec6267326..50232d476a27 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -465,7 +465,7 @@ cpumask_t node_to_cpumask_map[MAX_NUMNODES] __read_mostly =
 				{ [0 ... MAX_NUMNODES-1] = CPU_MASK_NONE };
 EXPORT_SYMBOL(node_to_cpumask_map);
 /* which node each logical CPU is on */
-int cpu_to_node_map[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 };
+u8 cpu_to_node_map[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = 0 };
 EXPORT_SYMBOL(cpu_to_node_map);
 
 /* set up a mapping between cpu and node. */
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 93071cdf0849..4e14ecb90764 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -851,39 +851,6 @@ static int __init smp_sanity_check(unsigned max_cpus)
 	return 0;
 }
 
-/*
- * Copy data used in early init routines from the initial arrays to the
- * per cpu data areas.  These arrays then become expendable and the
- * *_ptrs are zeroed indicating that the static arrays are gone.
- */
-void __init smp_set_apicids(void)
-{
-	int cpu;
-
-	for_each_possible_cpu(cpu) {
-		if (per_cpu_offset(cpu)) {
-			per_cpu(x86_cpu_to_apicid, cpu) =
-						x86_cpu_to_apicid_init[cpu];
-#ifdef CONFIG_NUMA
-			per_cpu(x86_cpu_to_node_map, cpu) =
-						x86_cpu_to_node_map_init[cpu];
-#endif
-			per_cpu(x86_bios_cpu_apicid, cpu) =
-						x86_bios_cpu_apicid_init[cpu];
-		}
-		else
-			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
-									cpu);
-	}
-
-	/* indicate the early static arrays are gone */
-	x86_cpu_to_apicid_early_ptr = NULL;
-#ifdef CONFIG_NUMA
-	x86_cpu_to_node_map_early_ptr = NULL;
-#endif
-	x86_bios_cpu_apicid_early_ptr = NULL;
-}
-
 static void __init smp_cpu_index_default(void)
 {
 	int i;
@@ -906,7 +873,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	smp_cpu_index_default();
 	current_cpu_data = boot_cpu_data;
 	current_thread_info()->cpu = 0;  /* needed? */
-	smp_set_apicids();
 	set_cpu_sibling_map(0);
 
 	if (smp_sanity_check(max_cpus) < 0) {
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index e5a1ec8342dc..04cbecaeca81 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -382,9 +382,10 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 			setup_node_bootmem(i, nodes[i].start, nodes[i].end);
 
 	for (i = 0; i < NR_CPUS; i++) {
-		if (cpu_to_node(i) == NUMA_NO_NODE)
+		int node = cpu_to_node(i);
+		if (node == NUMA_NO_NODE)
 			continue;
-		if (!node_isset(cpu_to_node(i), node_possible_map))
+		if (!node_isset(node, node_possible_map))
 			numa_set_node(i, NUMA_NO_NODE);
 	}
 	numa_init_array();
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 2da1464ecbef..040374f030cf 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -30,15 +30,29 @@
 #include <asm/mpspec.h>
 
 /* Mappings between logical cpu number and node number */
+#ifdef CONFIG_X86_32
+extern u8 cpu_to_node_map[];
+
+#else
 DECLARE_PER_CPU(u16, x86_cpu_to_node_map);
 extern u16 x86_cpu_to_node_map_init[];
 extern void *x86_cpu_to_node_map_early_ptr;
+#endif
+
 extern cpumask_t node_to_cpumask_map[];
 
 #define NUMA_NO_NODE	((u16)(~0))
 
 /* Returns the number of the node containing CPU 'cpu' */
+#ifdef CONFIG_X86_32
+#define early_cpu_to_node(cpu)	cpu_to_node(cpu)
 static inline int cpu_to_node(int cpu)
+{
+	return cpu_to_node_map[cpu];
+}
+
+#else /* CONFIG_X86_64 */
+static inline int early_cpu_to_node(int cpu)
 {
 	u16 *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
@@ -50,6 +64,15 @@ static inline int cpu_to_node(int cpu)
 		return NUMA_NO_NODE;
 }
 
+static inline int cpu_to_node(int cpu)
+{
+	if(per_cpu_offset(cpu))
+		return per_cpu(x86_cpu_to_node_map, cpu);
+	else
+		return NUMA_NO_NODE;
+}
+#endif /* CONFIG_X86_64 */
+
 /*
  * Returns the number of the node containing Node 'node'. This
  * architecture is flat, so it is a pretty simple function!
-- 
cgit v1.2.3


From c49a4955ea504c82f6b690491639bba5b8c1dc47 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:22 +0100
Subject: x86: add debug of invalid per_cpu map accesses

Provide a means to trap usages of per_cpu map variables before
they are setup.  Define CONFIG_DEBUG_PER_CPU_MAPS to activate.

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug     | 12 ++++++++++++
 arch/x86/mm/numa_64.c      |  3 +++
 include/asm-x86/topology.h |  7 +++++++
 3 files changed, 22 insertions(+)

(limited to 'include')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 88420af98140..0a82b889d76e 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -47,6 +47,18 @@ config DEBUG_PAGEALLOC
 	  This results in a large slowdown, but helps to find certain types
 	  of memory corruptions.
 
+config DEBUG_PER_CPU_MAPS
+	bool "Debug access to per_cpu maps"
+	depends on DEBUG_KERNEL
+	depends on X86_64_SMP
+	default n
+	help
+	  Say Y to verify that the per_cpu map being accessed has
+	  been setup.  Adds a fair amount of code to kernel memory
+	  and decreases performance.
+
+	  Say N if unsure.
+
 config DEBUG_RODATA
 	bool "Write protect kernel read-only data structures"
 	depends on DEBUG_KERNEL
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 5d24dc1ec237..e157cb274b25 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -37,6 +37,9 @@ u16 x86_cpu_to_node_map_init[NR_CPUS] = {
 void *x86_cpu_to_node_map_early_ptr;
 DEFINE_PER_CPU(u16, x86_cpu_to_node_map) = NUMA_NO_NODE;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map);
+#ifdef	CONFIG_DEBUG_PER_CPU_MAPS
+EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
+#endif
 
 u16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 040374f030cf..f1e12329078e 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -66,6 +66,13 @@ static inline int early_cpu_to_node(int cpu)
 
 static inline int cpu_to_node(int cpu)
 {
+#ifdef	CONFIG_DEBUG_PER_CPU_MAPS
+	if(x86_cpu_to_node_map_early_ptr) {
+		printk("KERN_NOTICE cpu_to_node(%d): usage too early!\n",
+			(int)cpu);
+		BUG();
+	}
+#endif
 	if(per_cpu_offset(cpu))
 		return per_cpu(x86_cpu_to_node_map, cpu);
 	else
-- 
cgit v1.2.3


From 5f5cd8fd60c71ce47d2ce4e60e7ccbc306e91c64 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:23 +0100
Subject: x86: add debug of invalid per_cpu map accesses

dont crash survivable situations.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/topology.h | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index f1e12329078e..9a8228a177c7 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -66,14 +66,15 @@ static inline int early_cpu_to_node(int cpu)
 
 static inline int cpu_to_node(int cpu)
 {
-#ifdef	CONFIG_DEBUG_PER_CPU_MAPS
-	if(x86_cpu_to_node_map_early_ptr) {
+#ifdef CONFIG_DEBUG_PER_CPU_MAPS
+	if (x86_cpu_to_node_map_early_ptr) {
 		printk("KERN_NOTICE cpu_to_node(%d): usage too early!\n",
 			(int)cpu);
-		BUG();
+		dump_stack();
+		return ((int *)x86_cpu_to_node_map_early_ptr)[cpu];
 	}
 #endif
-	if(per_cpu_offset(cpu))
+	if (per_cpu_offset(cpu))
 		return per_cpu(x86_cpu_to_node_map, cpu);
 	else
 		return NUMA_NO_NODE;
-- 
cgit v1.2.3


From 75f2ce033168ff435e72bf5bb615176d9930e77f Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:24 +0100
Subject: x86: get_cycles() fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/tsc.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index a6e8d35c3f86..7d3e27f7d484 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -27,10 +27,8 @@ static inline cycles_t get_cycles(void)
 	if (!cpu_has_tsc)
 		return 0;
 #endif
-
-#if defined(CONFIG_X86_GENERIC) || defined(CONFIG_X86_TSC)
 	rdtscll(ret);
-#endif
+
 	return ret;
 }
 
-- 
cgit v1.2.3


From 625d6cffcac1c96faa18d629f1271d63af0e05f2 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:25 +0100
Subject: x86: fix early cpu_to_node panic from nr_free_zone_pages

call early_cpu_to_node() since per_cpu(cpu_to_node_map) might not be setup
yet.

I also had to export x86_cpu_to_node_map_early_ptr because of some calls
from the network code to numa_node_id():

	net/ipv4/netfilter/arp_tables.c:
	net/ipv4/netfilter/ip_tables.c:
	net/ipv4/netfilter/ip_tables.c:

Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/numa_64.c      | 2 --
 include/asm-x86/topology.h | 2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 70b463abf6f4..441af4edea59 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -37,9 +37,7 @@ u16 x86_cpu_to_node_map_init[NR_CPUS] = {
 void *x86_cpu_to_node_map_early_ptr;
 DEFINE_PER_CPU(u16, x86_cpu_to_node_map) = NUMA_NO_NODE;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map);
-#ifdef	CONFIG_DEBUG_PER_CPU_MAPS
 EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
-#endif
 
 u16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 9a8228a177c7..7b0cce2f370d 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -37,6 +37,8 @@ extern u8 cpu_to_node_map[];
 DECLARE_PER_CPU(u16, x86_cpu_to_node_map);
 extern u16 x86_cpu_to_node_map_init[];
 extern void *x86_cpu_to_node_map_early_ptr;
+/* Returns the number of the current Node. */
+#define numa_node_id()		(early_cpu_to_node(raw_smp_processor_id()))
 #endif
 
 extern cpumask_t node_to_cpumask_map[];
-- 
cgit v1.2.3


From 4323838215184f5a2f081e0d17b8d60731b03164 Mon Sep 17 00:00:00 2001
From: "travis@sgi.com" <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:25 +0100
Subject: x86: change size of node ids from u8 to s16

Change the size of node ids for X86_64 from u8 to s16 to
accomodate more than 32k nodes and allow for NUMA_NO_NODE
(-1) to be sign extended to int.

Cc: David Rientjes <rientjes@google.com>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig            |  1 +
 arch/x86/mm/numa_64.c       | 12 ++++++------
 include/asm-x86/mmzone_64.h |  6 +++---
 include/asm-x86/numa_64.h   |  2 +-
 include/asm-x86/topology.h  | 10 +++++-----
 5 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 2114b4c71fa0..81af31e5a9f1 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -848,6 +848,7 @@ config NUMA_EMU
 
 config NODES_SHIFT
 	int
+	range 1 15  if X86_64
 	default "6" if X86_64
 	default "4" if X86_NUMAQ
 	default "3"
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 441af4edea59..0ca96d86cb71 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -31,15 +31,15 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
-u16 x86_cpu_to_node_map_init[NR_CPUS] = {
+int x86_cpu_to_node_map_init[NR_CPUS] = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
 void *x86_cpu_to_node_map_early_ptr;
-DEFINE_PER_CPU(u16, x86_cpu_to_node_map) = NUMA_NO_NODE;
+DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map);
 EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
 
-u16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
+s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
 };
 
@@ -63,7 +63,7 @@ static int __init populate_memnodemap(const struct bootnode *nodes,
 	unsigned long addr, end;
 	int i, res = -1;
 
-	memset(memnodemap, 0xff, memnodemapsize);
+	memset(memnodemap, 0xff, sizeof(s16)*memnodemapsize);
 	for (i = 0; i < numnodes; i++) {
 		addr = nodes[i].start;
 		end = nodes[i].end;
@@ -72,7 +72,7 @@ static int __init populate_memnodemap(const struct bootnode *nodes,
 		if ((end >> shift) >= memnodemapsize)
 			return 0;
 		do {
-			if (memnodemap[addr >> shift] != 0xff)
+			if (memnodemap[addr >> shift] != NUMA_NO_NODE)
 				return -1;
 			memnodemap[addr >> shift] = i;
 			addr += (1UL << shift);
@@ -533,7 +533,7 @@ __cpuinit void numa_add_cpu(int cpu)
 
 void __cpuinit numa_set_node(int cpu, int node)
 {
-	u16 *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
+	int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
 	cpu_pda(cpu)->nodenumber = node;
 
diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h
index b0c25ae111d9..ebaf9663aa8a 100644
--- a/include/asm-x86/mmzone_64.h
+++ b/include/asm-x86/mmzone_64.h
@@ -15,9 +15,9 @@
 struct memnode {
 	int shift;
 	unsigned int mapsize;
-	u8 *map;
-	u8 embedded_map[64-16];
-} ____cacheline_aligned; /* total size = 64 bytes */
+	s16 *map;
+	s16 embedded_map[64-8];
+} ____cacheline_aligned; /* total size = 128 bytes */
 extern struct memnode memnode;
 #define memnode_shift memnode.shift
 #define memnodemap memnode.map
diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index c797cd523d2b..15fe07cde586 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -20,7 +20,7 @@ extern void numa_set_node(int cpu, int node);
 extern void srat_reserve_add_area(int nodeid);
 extern int hotadd_percent;
 
-extern u16 apicid_to_node[MAX_LOCAL_APIC];
+extern s16 apicid_to_node[MAX_LOCAL_APIC];
 
 extern void numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn);
 extern unsigned long numa_free_all_bootmem(void);
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 7b0cce2f370d..8af05a93f097 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -31,11 +31,11 @@
 
 /* Mappings between logical cpu number and node number */
 #ifdef CONFIG_X86_32
-extern u8 cpu_to_node_map[];
+extern int cpu_to_node_map[];
 
 #else
-DECLARE_PER_CPU(u16, x86_cpu_to_node_map);
-extern u16 x86_cpu_to_node_map_init[];
+DECLARE_PER_CPU(int, x86_cpu_to_node_map);
+extern int x86_cpu_to_node_map_init[];
 extern void *x86_cpu_to_node_map_early_ptr;
 /* Returns the number of the current Node. */
 #define numa_node_id()		(early_cpu_to_node(raw_smp_processor_id()))
@@ -43,7 +43,7 @@ extern void *x86_cpu_to_node_map_early_ptr;
 
 extern cpumask_t node_to_cpumask_map[];
 
-#define NUMA_NO_NODE	((u16)(~0))
+#define NUMA_NO_NODE	(-1)
 
 /* Returns the number of the node containing CPU 'cpu' */
 #ifdef CONFIG_X86_32
@@ -56,7 +56,7 @@ static inline int cpu_to_node(int cpu)
 #else /* CONFIG_X86_64 */
 static inline int early_cpu_to_node(int cpu)
 {
-	u16 *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
+	int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
 	if (cpu_to_node_map)
 		return cpu_to_node_map[cpu];
-- 
cgit v1.2.3


From 1bdbdaacf774f2979ed4cb0c4a4316c9e578c897 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bwalle@suse.de>
Date: Wed, 30 Jan 2008 13:33:28 +0100
Subject: x86, rtc: make CONFIG_HPET_EMULATE_RTC usable from modules

enabled, then interrupts don't work for the rtc-cmos driver which results in
RTC_AIE*, RTC_PIE* and RTC_ALM being unusable.  This affects hwclock from
util-linux-ng at least on i386 since that uses RTC_PIE_ON.  (For x86-64, a
polling method is used for unknown reasons.)

This patch series now

  1. export the functions from arch/x86/kernel/hpet.c that the old char/rtc
     driver uses to work around that problem,

  2. makes it possible to compile the old rtc driver as module, while still
     having CONFIG_HPET_EMULATE_RTC enabled and

  3. makes use of the exported functions in (1) in the new rtc-cmos driver.

This patch:

This patch makes the RTC emulation functions in arch/x86/kernel/hpet.c usable
for kernel modules. It

  - exports the functions (EXPORT_SYMBOL_GPL()),
  - adds an interface to register the interrupt callback function
    instead of using only a fixed callback function and
  - replaces the rtc_get_rtc_time() function which depends on
    CONFIG_RTC with a call to get_rtc_time() which is defined in
    include/asm-generic/rtc.h.

The only dependency to CONFIG_RTC is the call to rtc_interrupt() which is
removed by the next patch. After this, there's no (code) dependency of
this functions to CONFIG_RTC=y any more.

Signed-off-by: Bernhard Walle <bwalle@suse.de>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Cc: Andi Kleen <ak@suse.de>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Robert Picco <Robert.Picco@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/hpet.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/hpet.h |  3 +++
 2 files changed, 49 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 5c1702789be4..d65ced59a18f 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -107,6 +107,7 @@ int is_hpet_enabled(void)
 {
 	return is_hpet_capable() && hpet_legacy_int_enabled;
 }
+EXPORT_SYMBOL_GPL(is_hpet_enabled);
 
 /*
  * When the hpet driver (/dev/hpet) is enabled, we need to reserve
@@ -475,6 +476,7 @@ void hpet_disable(void)
  */
 #include <linux/mc146818rtc.h>
 #include <linux/rtc.h>
+#include <asm/rtc.h>
 
 #define DEFAULT_RTC_INT_FREQ	64
 #define DEFAULT_RTC_SHIFT	6
@@ -489,6 +491,38 @@ static unsigned long hpet_default_delta;
 static unsigned long hpet_pie_delta;
 static unsigned long hpet_pie_limit;
 
+static rtc_irq_handler irq_handler;
+
+/*
+ * Registers a IRQ handler.
+ */
+int hpet_register_irq_handler(rtc_irq_handler handler)
+{
+	if (!is_hpet_enabled())
+		return -ENODEV;
+	if (irq_handler)
+		return -EBUSY;
+
+	irq_handler = handler;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hpet_register_irq_handler);
+
+/*
+ * Deregisters the IRQ handler registered with hpet_register_irq_handler()
+ * and does cleanup.
+ */
+void hpet_unregister_irq_handler(rtc_irq_handler handler)
+{
+	if (!is_hpet_enabled())
+		return;
+
+	irq_handler = NULL;
+	hpet_rtc_flags = 0;
+}
+EXPORT_SYMBOL_GPL(hpet_unregister_irq_handler);
+
 /*
  * Timer 1 for RTC emulation. We use one shot mode, as periodic mode
  * is not supported by all HPET implementations for timer 1.
@@ -530,6 +564,7 @@ int hpet_rtc_timer_init(void)
 
 	return 1;
 }
+EXPORT_SYMBOL_GPL(hpet_rtc_timer_init);
 
 /*
  * The functions below are called from rtc driver.
@@ -544,6 +579,7 @@ int hpet_mask_rtc_irq_bit(unsigned long bit_mask)
 	hpet_rtc_flags &= ~bit_mask;
 	return 1;
 }
+EXPORT_SYMBOL_GPL(hpet_mask_rtc_irq_bit);
 
 int hpet_set_rtc_irq_bit(unsigned long bit_mask)
 {
@@ -559,6 +595,7 @@ int hpet_set_rtc_irq_bit(unsigned long bit_mask)
 
 	return 1;
 }
+EXPORT_SYMBOL_GPL(hpet_set_rtc_irq_bit);
 
 int hpet_set_alarm_time(unsigned char hrs, unsigned char min,
 			unsigned char sec)
@@ -572,6 +609,7 @@ int hpet_set_alarm_time(unsigned char hrs, unsigned char min,
 
 	return 1;
 }
+EXPORT_SYMBOL_GPL(hpet_set_alarm_time);
 
 int hpet_set_periodic_freq(unsigned long freq)
 {
@@ -590,11 +628,13 @@ int hpet_set_periodic_freq(unsigned long freq)
 	}
 	return 1;
 }
+EXPORT_SYMBOL_GPL(hpet_set_periodic_freq);
 
 int hpet_rtc_dropped_irq(void)
 {
 	return is_hpet_enabled();
 }
+EXPORT_SYMBOL_GPL(hpet_rtc_dropped_irq);
 
 static void hpet_rtc_timer_reinit(void)
 {
@@ -638,9 +678,10 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
 	unsigned long rtc_int_flag = 0;
 
 	hpet_rtc_timer_reinit();
+	memset(&curr_time, 0, sizeof(struct rtc_time));
 
 	if (hpet_rtc_flags & (RTC_UIE | RTC_AIE))
-		rtc_get_rtc_time(&curr_time);
+		get_rtc_time(&curr_time);
 
 	if (hpet_rtc_flags & RTC_UIE &&
 	    curr_time.tm_sec != hpet_prev_update_sec) {
@@ -662,8 +703,12 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
 
 	if (rtc_int_flag) {
 		rtc_int_flag |= (RTC_IRQF | (RTC_NUM_INTS << 8));
+		if (irq_handler)
+			irq_handler(rtc_int_flag, dev_id);
+
 		rtc_interrupt(rtc_int_flag, dev_id);
 	}
 	return IRQ_HANDLED;
 }
+EXPORT_SYMBOL_GPL(hpet_rtc_interrupt);
 #endif
diff --git a/include/asm-x86/hpet.h b/include/asm-x86/hpet.h
index 624f600f7161..6a9b4ac59bf7 100644
--- a/include/asm-x86/hpet.h
+++ b/include/asm-x86/hpet.h
@@ -69,6 +69,7 @@ extern void force_hpet_resume(void);
 
 #include <linux/interrupt.h>
 
+typedef irqreturn_t (*rtc_irq_handler)(int interrupt, void *cookie);
 extern int hpet_mask_rtc_irq_bit(unsigned long bit_mask);
 extern int hpet_set_rtc_irq_bit(unsigned long bit_mask);
 extern int hpet_set_alarm_time(unsigned char hrs, unsigned char min,
@@ -77,6 +78,8 @@ extern int hpet_set_periodic_freq(unsigned long freq);
 extern int hpet_rtc_dropped_irq(void);
 extern int hpet_rtc_timer_init(void);
 extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id);
+extern int hpet_register_irq_handler(rtc_irq_handler handler);
+extern void hpet_unregister_irq_handler(rtc_irq_handler handler);
 
 #endif /* CONFIG_HPET_EMULATE_RTC */
 
-- 
cgit v1.2.3


From 093af8d7f0ba3c6be1485973508584ef081e9f93 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Wed, 30 Jan 2008 13:33:32 +0100
Subject: x86_32: trim memory by updating e820

when MTRRs are not covering the whole e820 table, we need to trim the
RAM and need to update e820.

reuse some code on 64-bit as well.

here need to add early_get_cap and use it in early_cpu_detect, and move
mtrr_bp_init early.

The code successfully trimmed the memory map on Justin's system:

from:

 [    0.000000]  BIOS-e820: 0000000100000000 - 000000022c000000 (usable)

to:

 [    0.000000]   modified: 0000000100000000 - 0000000228000000 (usable)
 [    0.000000]   modified: 0000000228000000 - 000000022c000000 (reserved)

According to Justin it makes quite a difference:

|  When I boot the box without any trimming it acts like a 286 or 386,
|  takes about 10 minutes to boot (using raptor disks).

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Tested-by: Justin Piszcz <jpiszcz@lucidpixels.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |  2 +-
 arch/x86/kernel/cpu/common.c        | 30 +++++++++++++++++++++++++++++-
 arch/x86/kernel/cpu/mtrr/main.c     | 22 ++++++++++++++--------
 arch/x86/kernel/e820_32.c           | 11 +++++++++++
 arch/x86/kernel/setup_32.c          |  6 ++++++
 include/asm-x86/e820_32.h           |  3 +++
 6 files changed, 64 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 50d564dabb13..fe3031d56431 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -583,7 +583,7 @@ and is between 256 and 4096 characters. It is defined in the file
 			See drivers/char/README.epca and
 			Documentation/digiepca.txt.
 
-	disable_mtrr_trim [X86-64, Intel only]
+	disable_mtrr_trim [X86, Intel and AMD only]
 			By default the kernel will trim any uncacheable
 			memory out of your available memory pool based on
 			MTRR settings.  This parameter disables that behavior,
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 56cc341cc586..bba850b05d0e 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -278,6 +278,33 @@ void __init cpu_detect(struct cpuinfo_x86 *c)
 			c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
 	}
 }
+static void __cpuinit early_get_cap(struct cpuinfo_x86 *c)
+{
+	u32 tfms, xlvl;
+	int ebx;
+
+	memset(&c->x86_capability, 0, sizeof c->x86_capability);
+	if (have_cpuid_p()) {
+		/* Intel-defined flags: level 0x00000001 */
+		if (c->cpuid_level >= 0x00000001) {
+			u32 capability, excap;
+			cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
+			c->x86_capability[0] = capability;
+			c->x86_capability[4] = excap;
+		}
+
+		/* AMD-defined flags: level 0x80000001 */
+		xlvl = cpuid_eax(0x80000000);
+		if ((xlvl & 0xffff0000) == 0x80000000) {
+			if (xlvl >= 0x80000001) {
+				c->x86_capability[1] = cpuid_edx(0x80000001);
+				c->x86_capability[6] = cpuid_ecx(0x80000001);
+			}
+		}
+
+	}
+
+}
 
 /* Do minimum CPU detection early.
    Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment.
@@ -306,6 +333,8 @@ static void __init early_cpu_detect(void)
 		early_init_intel(c);
 		break;
 	}
+
+	early_get_cap(c);
 }
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 * c)
@@ -485,7 +514,6 @@ void __init identify_boot_cpu(void)
 	identify_cpu(&boot_cpu_data);
 	sysenter_setup();
 	enable_sep_cpu();
-	mtrr_bp_init();
 }
 
 void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index ccd36ed2187b..ac4b6338f3f4 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -624,7 +624,6 @@ static struct sysdev_driver mtrr_sysdev_driver = {
 	.resume		= mtrr_restore,
 };
 
-#ifdef CONFIG_X86_64
 static int disable_mtrr_trim;
 
 static int __init disable_mtrr_trim_setup(char *str)
@@ -643,13 +642,10 @@ early_param("disable_mtrr_trim", disable_mtrr_trim_setup);
 #define Tom2Enabled (1U << 21)
 #define Tom2ForceMemTypeWB (1U << 22)
 
-static __init int amd_special_default_mtrr(unsigned long end_pfn)
+static __init int amd_special_default_mtrr(void)
 {
 	u32 l, h;
 
-	/* Doesn't apply to memory < 4GB */
-	if (end_pfn <= (0xffffffff >> PAGE_SHIFT))
-		return 0;
 	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
 		return 0;
 	if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
@@ -687,9 +683,14 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 	 * Make sure we only trim uncachable memory on machines that
 	 * support the Intel MTRR architecture:
 	 */
+	if (!is_cpu(INTEL) || disable_mtrr_trim)
+		return 0;
 	rdmsr(MTRRdefType_MSR, def, dummy);
 	def &= 0xff;
-	if (!is_cpu(INTEL) || disable_mtrr_trim || def != MTRR_TYPE_UNCACHABLE)
+	if (def != MTRR_TYPE_UNCACHABLE)
+		return 0;
+
+	if (amd_special_default_mtrr())
 		return 0;
 
 	/* Find highest cached pfn */
@@ -703,8 +704,14 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 			highest_addr = base + size;
 	}
 
-	if (amd_special_default_mtrr(end_pfn))
+	/* kvm/qemu doesn't have mtrr set right, don't trim them all */
+	if (!highest_addr) {
+		printk(KERN_WARNING "***************\n");
+		printk(KERN_WARNING "**** WARNING: likely strange cpu\n");
+		printk(KERN_WARNING "**** MTRRs all blank, cpu in qemu?\n");
+		printk(KERN_WARNING "***************\n");
 		return 0;
+	}
 
 	if ((highest_addr >> PAGE_SHIFT) < end_pfn) {
 		printk(KERN_WARNING "***************\n");
@@ -726,7 +733,6 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 
 	return 0;
 }
-#endif
 
 /**
  * mtrr_bp_init - initialize mtrrs on the boot CPU
diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c
index 931934a7b353..4e16ef4a2659 100644
--- a/arch/x86/kernel/e820_32.c
+++ b/arch/x86/kernel/e820_32.c
@@ -749,3 +749,14 @@ static int __init parse_memmap(char *arg)
 	return 0;
 }
 early_param("memmap", parse_memmap);
+void __init update_e820(void)
+{
+	u8 nr_map;
+
+	nr_map = e820.nr_map;
+	if (sanitize_e820_map(e820.map, &nr_map))
+		return;
+	e820.nr_map = nr_map;
+	printk(KERN_INFO "modified physical RAM map:\n");
+	print_memory_map("modified");
+}
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 26a56f714d34..83ba3ca5f431 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -48,6 +48,7 @@
 
 #include <video/edid.h>
 
+#include <asm/mtrr.h>
 #include <asm/apic.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
@@ -758,6 +759,11 @@ void __init setup_arch(char **cmdline_p)
 
 	max_low_pfn = setup_memory();
 
+	/* update e820 for memory not covered by WB MTRRs */
+	mtrr_bp_init();
+	if (mtrr_trim_uncached_memory(max_pfn))
+		max_low_pfn = setup_memory();
+
 #ifdef CONFIG_VMI
 	/*
 	 * Must be after max_low_pfn is determined, and before kernel
diff --git a/include/asm-x86/e820_32.h b/include/asm-x86/e820_32.h
index e2faf5f3a0bb..f1da7ebd1905 100644
--- a/include/asm-x86/e820_32.h
+++ b/include/asm-x86/e820_32.h
@@ -19,12 +19,15 @@
 #ifndef __ASSEMBLY__
 
 extern struct e820map e820;
+extern void update_e820(void);
 
 extern int e820_all_mapped(unsigned long start, unsigned long end,
 			   unsigned type);
 extern int e820_any_mapped(u64 start, u64 end, unsigned type);
 extern void find_max_pfn(void);
 extern void register_bootmem_low_pages(unsigned long max_low_pfn);
+extern void add_memory_region(unsigned long long start,
+			      unsigned long long size, int type);
 extern void e820_register_memory(void);
 extern void limit_regions(unsigned long long size);
 extern void print_memory_map(char *who);
-- 
cgit v1.2.3


From dd5af90a7f3d79e04b7eace9a98644dbf2038f4d Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Wed, 30 Jan 2008 13:33:32 +0100
Subject: x86/non-x86: percpu, node ids, apic ids x86.git fixup

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig             |  2 +-
 include/asm-generic/percpu.h | 12 ++----------
 init/main.c                  |  4 ++--
 kernel/module.c              |  8 ++++++++
 4 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index f0887d12a5bb..8e1b33c5405f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -97,7 +97,7 @@ config GENERIC_TIME_VSYSCALL
 	bool
 	default X86_64
 
-config ARCH_SETS_UP_PER_CPU_AREA
+config HAVE_SETUP_PER_CPU_AREA
 	def_bool X86_64
 
 config ARCH_SUPPORTS_OPROFILE
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index c41b1a731129..4b8d31cda1a0 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -47,7 +47,7 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #endif
 
 /*
- * A percpu variable may point to a discarded reghions. The following are
+ * A percpu variable may point to a discarded regions. The following are
  * established ways to produce a usable pointer from the percpu variable
  * offset.
  */
@@ -59,18 +59,10 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), __my_cpu_offset))
 
 
-#ifdef CONFIG_ARCH_SETS_UP_PER_CPU_AREA
+#ifdef CONFIG_HAVE_SETUP_PER_CPU_AREA
 extern void setup_per_cpu_areas(void);
 #endif
 
-/* A macro to avoid #include hell... */
-#define percpu_modcopy(pcpudst, src, size)			\
-do {								\
-	unsigned int __i;					\
-	for_each_possible_cpu(__i)				\
-		memcpy((pcpudst)+per_cpu_offset(__i),		\
-		       (src), (size));				\
-} while (0)
 #else /* ! SMP */
 
 #define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu_var(var)))
diff --git a/init/main.c b/init/main.c
index 5843fe996703..3316dffe3e57 100644
--- a/init/main.c
+++ b/init/main.c
@@ -363,7 +363,7 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 
 #else
 
-#ifndef CONFIG_ARCH_SETS_UP_PER_CPU_AREA
+#ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 
 EXPORT_SYMBOL(__per_cpu_offset);
@@ -384,7 +384,7 @@ static void __init setup_per_cpu_areas(void)
 		ptr += size;
 	}
 }
-#endif /* CONFIG_ARCH_SETS_UP_CPU_AREA */
+#endif /* CONFIG_HAVE_SETUP_PER_CPU_AREA */
 
 /* Called by boot processor to activate the rest. */
 static void __init smp_init(void)
diff --git a/kernel/module.c b/kernel/module.c
index f6a4e721fd49..bd60278ee703 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -430,6 +430,14 @@ static unsigned int find_pcpusec(Elf_Ehdr *hdr,
 	return find_sec(hdr, sechdrs, secstrings, ".data.percpu");
 }
 
+static void percpu_modcopy(void *pcpudest, const void *from, unsigned long size)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		memcpy(pcpudest + per_cpu_offset(cpu), from, size);
+}
+
 static int percpu_modinit(void)
 {
 	pcpu_num_used = 2;
-- 
cgit v1.2.3


From ade761496dcd0aea1c41da21d5a6ced4897f02e7 Mon Sep 17 00:00:00 2001
From: Andres Salomon <dilinger@queued.net>
Date: Wed, 30 Jan 2008 13:33:35 +0100
Subject: x86: GEODE: update GPIO API to support setting multiple GPIOs at once

The existing Geode GPIO API only allows for updating one GPIO at once.  There
are instances where users want to update multiple GPIOs at once.  With the
current API, they are given two choices; either ignore the GPIO API:

      outl(0xc000, gpio_base + GPIO_OUTPUT_VAL);
      outl(0xc000, gpio_base + GPIO_OUTPUT_ENABLE);

Alternatively, call each GPIO update separately:

      geode_gpio_set(14, GPIO_OUTPUT_VAL);
      geode_gpio_set(15, GPIO_OUTPUT_VAL);
      geode_gpio_set(14, GPIO_OUTPUT_ENABLE);
      geode_gpio_set(15, GPIO_OUTPUT_ENABLE);

Neither are desirable.  This patch changes the GPIO API to allow for setting
of multiple GPIOs at once; rather than being passed an integer, we pass
a bitmask and provide a translation function.  The above code would now
look like this:

      geode_gpio_set(geode_gpio(14)|geode_gpio(15), GPIO_OUTPUT_VAL);
      geode_gpio_set(geode_gpio(14)|geode_gpio(15), GPIO_OUTPUT_ENABLE);

Since there are no upstream users of the GPIO API yet (afaik), best to
change this now.  This also adds a bit of sanity checking; it is no
longer possible to use a GPIO above 28.

Note the semantics of geode_gpio_isset() have changed:
geode_gpio_isset(geode_gpio(3)|geode_gpio(4), ...)
will only return true iff both GPIOs are set.

Signed-off-by: Andres Salomon <dilinger@debian.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/geode_32.c | 48 +++++++++++++++++++++++++++++++---------------
 include/asm-x86/geode.h    | 12 +++++++++---
 2 files changed, 42 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/geode_32.c b/arch/x86/kernel/geode_32.c
index f12d8c5d9809..9c7f7d395968 100644
--- a/arch/x86/kernel/geode_32.c
+++ b/arch/x86/kernel/geode_32.c
@@ -1,6 +1,7 @@
 /*
  * AMD Geode southbridge support code
  * Copyright (C) 2006, Advanced Micro Devices, Inc.
+ * Copyright (C) 2007, Andres Salomon <dilinger@debian.org>
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of version 2 of the GNU General Public License
@@ -51,45 +52,62 @@ EXPORT_SYMBOL_GPL(geode_get_dev_base);
 
 /* === GPIO API === */
 
-void geode_gpio_set(unsigned int gpio, unsigned int reg)
+void geode_gpio_set(u32 gpio, unsigned int reg)
 {
 	u32 base = geode_get_dev_base(GEODE_DEV_GPIO);
 
 	if (!base)
 		return;
 
-	if (gpio < 16)
-		outl(1 << gpio, base + reg);
-	else
-		outl(1 << (gpio - 16), base + 0x80 + reg);
+	/* low bank register */
+	if (gpio & 0xFFFF)
+		outl(gpio & 0xFFFF, base + reg);
+	/* high bank register */
+	gpio >>= 16;
+	if (gpio)
+		outl(gpio, base + 0x80 + reg);
 }
 EXPORT_SYMBOL_GPL(geode_gpio_set);
 
-void geode_gpio_clear(unsigned int gpio, unsigned int reg)
+void geode_gpio_clear(u32 gpio, unsigned int reg)
 {
 	u32 base = geode_get_dev_base(GEODE_DEV_GPIO);
 
 	if (!base)
 		return;
 
-	if (gpio < 16)
-		outl(1 << (gpio + 16), base + reg);
-	else
-		outl(1 << gpio, base + 0x80 + reg);
+	/* low bank register */
+	if (gpio & 0xFFFF)
+		outl((gpio & 0xFFFF) << 16, base + reg);
+	/* high bank register */
+	gpio &= (0xFFFF << 16);
+	if (gpio)
+		outl(gpio, base + 0x80 + reg);
 }
 EXPORT_SYMBOL_GPL(geode_gpio_clear);
 
-int geode_gpio_isset(unsigned int gpio, unsigned int reg)
+int geode_gpio_isset(u32 gpio, unsigned int reg)
 {
 	u32 base = geode_get_dev_base(GEODE_DEV_GPIO);
+	u32 val;
 
 	if (!base)
 		return 0;
 
-	if (gpio < 16)
-		return (inl(base + reg) & (1 << gpio)) ? 1 : 0;
-	else
-		return (inl(base + 0x80 + reg) & (1 << (gpio - 16))) ? 1 : 0;
+	/* low bank register */
+	if (gpio & 0xFFFF) {
+		val = inl(base + reg) & (gpio & 0xFFFF);
+		if ((gpio & 0xFFFF) == val)
+			return 1;
+	}
+	/* high bank register */
+	gpio >>= 16;
+	if (gpio) {
+		val = inl(base + 0x80 + reg) & gpio;
+		if (gpio == val)
+			return 1;
+	}
+	return 0;
 }
 EXPORT_SYMBOL_GPL(geode_gpio_isset);
 
diff --git a/include/asm-x86/geode.h b/include/asm-x86/geode.h
index 771af336734f..811fe14f70b2 100644
--- a/include/asm-x86/geode.h
+++ b/include/asm-x86/geode.h
@@ -121,9 +121,15 @@ extern int geode_get_dev_base(unsigned int dev);
 #define GPIO_MAP_Z		0xE8
 #define GPIO_MAP_W		0xEC
 
-extern void geode_gpio_set(unsigned int, unsigned int);
-extern void geode_gpio_clear(unsigned int, unsigned int);
-extern int geode_gpio_isset(unsigned int, unsigned int);
+static inline u32 geode_gpio(unsigned int nr)
+{
+	BUG_ON(nr > 28);
+	return 1 << nr;
+}
+
+extern void geode_gpio_set(u32, unsigned int);
+extern void geode_gpio_clear(u32, unsigned int);
+extern int geode_gpio_isset(u32, unsigned int);
 extern void geode_gpio_setup_event(unsigned int, int, int);
 extern void geode_gpio_set_irq(unsigned int, unsigned int);
 
-- 
cgit v1.2.3


From a608295935d237bdbe95eefdba1e3fa40676df31 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <florian.fainelli@telecomint.eu>
Date: Wed, 30 Jan 2008 13:33:35 +0100
Subject: x86: add generic GPIO support to x86

This patch adds the generic GPIO support to the x86
architecture. We do the same as for MIPS, we let
the machine override the gpio callbacks and provide
defaults one in mach-generic.

Signed-off-by: Florian Fainelli <florian.fainelli@telecomint.eu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig                    |  3 +++
 include/asm-x86/gpio.h              |  6 ++++++
 include/asm-x86/mach-generic/gpio.h | 15 +++++++++++++++
 3 files changed, 24 insertions(+)
 create mode 100644 include/asm-x86/gpio.h
 create mode 100644 include/asm-x86/mach-generic/gpio.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 8e1b33c5405f..32e267186035 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -72,6 +72,9 @@ config GENERIC_BUG
 config GENERIC_HWEIGHT
 	def_bool y
 
+config GENERIC_GPIO
+	def_bool n
+
 config ARCH_MAY_HAVE_PC_FDC
 	def_bool y
 
diff --git a/include/asm-x86/gpio.h b/include/asm-x86/gpio.h
new file mode 100644
index 000000000000..ff87fca0caf9
--- /dev/null
+++ b/include/asm-x86/gpio.h
@@ -0,0 +1,6 @@
+#ifndef _ASM_I386_GPIO_H
+#define _ASM_I386_GPIO_H
+
+#include <gpio.h>
+
+#endif /* _ASM_I386_GPIO_H */
diff --git a/include/asm-x86/mach-generic/gpio.h b/include/asm-x86/mach-generic/gpio.h
new file mode 100644
index 000000000000..5305dcb96df2
--- /dev/null
+++ b/include/asm-x86/mach-generic/gpio.h
@@ -0,0 +1,15 @@
+#ifndef __ASM_MACH_GENERIC_GPIO_H
+#define __ASM_MACH_GENERIC_GPIO_H
+
+int gpio_request(unsigned gpio, const char *label);
+void gpio_free(unsigned gpio);
+int gpio_direction_input(unsigned gpio);
+int gpio_direction_output(unsigned gpio, int value);
+int gpio_get_value(unsigned gpio);
+void gpio_set_value(unsigned gpio, int value);
+int gpio_to_irq(unsigned gpio);
+int irq_to_gpio(unsigned irq);
+
+#include <asm-generic/gpio.h>           /* cansleep wrappers */
+
+#endif /* __ASM_MACH_GENERIC_GPIO_H */
-- 
cgit v1.2.3


From 0acf8e3447b893ff921863c2a4258e210d584452 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <florian.fainelli@telecomint.eu>
Date: Wed, 30 Jan 2008 13:33:36 +0100
Subject: pci: add PCI identifiers for the RDC devices

This patch defines the PCI identifiers found in
the RDC R-321x System-on-Chip.

Signed-off-by: Florian Fainelli <florian.fainelli@telecomint.eu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/pci_ids.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index c69531348363..41f6f28690f6 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2085,6 +2085,13 @@
 #define PCI_VENDOR_ID_BELKIN		0x1799
 #define PCI_DEVICE_ID_BELKIN_F5D7010V7	0x701f
 
+#define PCI_VENDOR_ID_RDC		0x17f3
+#define PCI_DEVICE_ID_RDC_R6020		0x6020
+#define PCI_DEVICE_ID_RDC_R6030		0x6030
+#define PCI_DEVICE_ID_RDC_R6040		0x6040
+#define PCI_DEVICE_ID_RDC_R6060		0x6060
+#define PCI_DEVICE_ID_RDC_R6061		0x6061
+
 #define PCI_VENDOR_ID_LENOVO		0x17aa
 
 #define PCI_VENDOR_ID_ARECA		0x17d3
-- 
cgit v1.2.3


From 5e3a77e9a9b7bfc1f69f51fe6d34aa649887980c Mon Sep 17 00:00:00 2001
From: Florian Fainelli <florian.fainelli@telecomint.eu>
Date: Wed, 30 Jan 2008 13:33:36 +0100
Subject: x86: add support for the RDC R-321x SoC

This patch adds support for the RDC R-321x system-on-chip,
also known as R-861x-(G). It uses the generic GPIO API and
has support for the on-chip hardware watchdog.

Build-fix from: Randy Dunlap <randy.dunlap@oracle.com>

Signed-off-by: Florian Fainelli <florian.fainelli@telecomint.eu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig                            |  14 +-
 arch/x86/Makefile                           |   5 +
 arch/x86/mach-rdc321x/Makefile              |   5 +
 arch/x86/mach-rdc321x/gpio.c                |  91 +++++++++
 arch/x86/mach-rdc321x/platform.c            |  68 +++++++
 arch/x86/mach-rdc321x/wdt.c                 | 275 ++++++++++++++++++++++++++++
 include/asm-x86/mach-rdc321x/gpio.h         |  56 ++++++
 include/asm-x86/mach-rdc321x/rdc321x_defs.h |   6 +
 include/asm-x86/timex.h                     |   2 +
 9 files changed, 521 insertions(+), 1 deletion(-)
 create mode 100644 arch/x86/mach-rdc321x/Makefile
 create mode 100644 arch/x86/mach-rdc321x/gpio.c
 create mode 100644 arch/x86/mach-rdc321x/platform.c
 create mode 100644 arch/x86/mach-rdc321x/wdt.c
 create mode 100644 include/asm-x86/mach-rdc321x/gpio.h
 create mode 100644 include/asm-x86/mach-rdc321x/rdc321x_defs.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 32e267186035..92dc919c7640 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -291,6 +291,18 @@ config X86_ES7000
 	  Only choose this option if you have such a system, otherwise you
 	  should say N here.
 
+config X86_RDC321X
+	bool "RDC R-321x SoC"
+	depends on X86_32
+	select M486
+	select X86_REBOOTFIXUPS
+	select GENERIC_GPIO
+	select LEDS_GPIO
+	help
+	  This option is needed for RDC R-321x system-on-chip, also known
+	  as R-8610-(G).
+	  If you don't have one of these chips, you should say N here.
+
 config X86_VSMP
 	bool "Support for ScaleMP vSMP"
 	depends on X86_64 && PCI
@@ -637,7 +649,7 @@ config X86_REBOOTFIXUPS
 	  system.
 
 	  Currently, the only fixup is for the Geode machines using
-	  CS5530A and CS5536 chipsets.
+	  CS5530A and CS5536 chipsets and the RDC R-321x SoC.
 
 	  Say Y if you want to enable the fixup. Currently, it's safe to
 	  enable this option even if you don't need it.
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index e2ffccfd6af6..b08f18261df6 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -139,6 +139,11 @@ mflags-$(CONFIG_X86_ES7000)	:= -Iinclude/asm-x86/mach-es7000
 fcore-$(CONFIG_X86_ES7000)	:= arch/x86/mach-es7000/
 mcore-$(CONFIG_X86_ES7000)	:= arch/x86/mach-default/
 
+# RDC R-321x subarch support
+mflags-$(CONFIG_X86_RDC321X)	:= -Iinclude/asm-x86/mach-rdc321x
+mcore-$(CONFIG_X86_RDC321X)	:= arch/x86/mach-default
+core-$(CONFIG_X86_RDC321X)	+= arch/x86/mach-rdc321x/
+
 # default subarch .h files
 mflags-y += -Iinclude/asm-x86/mach-default
 
diff --git a/arch/x86/mach-rdc321x/Makefile b/arch/x86/mach-rdc321x/Makefile
new file mode 100644
index 000000000000..1faac8125e3d
--- /dev/null
+++ b/arch/x86/mach-rdc321x/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the RDC321x specific parts of the kernel
+#
+obj-$(CONFIG_X86_RDC321X)        := gpio.o platform.o wdt.o
+
diff --git a/arch/x86/mach-rdc321x/gpio.c b/arch/x86/mach-rdc321x/gpio.c
new file mode 100644
index 000000000000..031269163bd6
--- /dev/null
+++ b/arch/x86/mach-rdc321x/gpio.c
@@ -0,0 +1,91 @@
+/*
+ *  Copyright (C) 2007, OpenWrt.org, Florian Fainelli <florian@openwrt.org>
+ *  	RDC321x architecture specific GPIO support
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ */
+
+#include <linux/autoconf.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+
+#include <asm/mach-rdc321x/rdc321x_defs.h>
+
+static inline int rdc_gpio_is_valid(unsigned gpio)
+{
+	return (gpio <= RDC_MAX_GPIO);
+}
+
+static unsigned int rdc_gpio_read(unsigned gpio)
+{
+	unsigned int val;
+
+	val = 0x80000000 | (7 << 11) | ((gpio&0x20?0x84:0x48));
+	outl(val, RDC3210_CFGREG_ADDR);
+	udelay(10);
+	val = inl(RDC3210_CFGREG_DATA);
+	val |= (0x1 << (gpio & 0x1F));
+	outl(val, RDC3210_CFGREG_DATA);
+	udelay(10);
+	val = 0x80000000 | (7 << 11) | ((gpio&0x20?0x88:0x4C));
+	outl(val, RDC3210_CFGREG_ADDR);
+	udelay(10);
+	val = inl(RDC3210_CFGREG_DATA);
+
+	return val;
+}
+
+static void rdc_gpio_write(unsigned int val)
+{
+	if (val) {
+		outl(val, RDC3210_CFGREG_DATA);
+		udelay(10);
+	}
+}
+
+int rdc_gpio_get_value(unsigned gpio)
+{
+	if (rdc_gpio_is_valid(gpio))
+		return (int)rdc_gpio_read(gpio);
+	else
+		return -EINVAL;
+}
+EXPORT_SYMBOL(rdc_gpio_get_value);
+
+void rdc_gpio_set_value(unsigned gpio, int value)
+{
+	unsigned int val;
+
+	if (!rdc_gpio_is_valid(gpio))
+		return;
+
+	val = rdc_gpio_read(gpio);
+
+	if (value)
+		val &= ~(0x1 << (gpio & 0x1F));
+	else
+		val |= (0x1 << (gpio & 0x1F));
+
+	rdc_gpio_write(val);
+}
+EXPORT_SYMBOL(rdc_gpio_set_value);
+
+int rdc_gpio_direction_input(unsigned gpio)
+{
+	return 0;
+}
+EXPORT_SYMBOL(rdc_gpio_direction_input);
+
+int rdc_gpio_direction_output(unsigned gpio, int value)
+{
+	return 0;
+}
+EXPORT_SYMBOL(rdc_gpio_direction_output);
+
+
diff --git a/arch/x86/mach-rdc321x/platform.c b/arch/x86/mach-rdc321x/platform.c
new file mode 100644
index 000000000000..dda6024a5862
--- /dev/null
+++ b/arch/x86/mach-rdc321x/platform.c
@@ -0,0 +1,68 @@
+/*
+ *  Generic RDC321x platform devices
+ *
+ *  Copyright (C) 2007 Florian Fainelli <florian@openwrt.org>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version 2
+ *  of the License, or (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the
+ *  Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+ *  Boston, MA  02110-1301, USA.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/version.h>
+#include <linux/leds.h>
+
+#include <asm/gpio.h>
+
+/* LEDS */
+static struct gpio_led default_leds[] = {
+	{ .name = "rdc:dmz", .gpio = 1, },
+};
+
+static struct gpio_led_platform_data rdc321x_led_data = {
+	.num_leds = ARRAY_SIZE(default_leds),
+	.leds = default_leds,
+};
+
+static struct platform_device rdc321x_leds = {
+	.name = "leds-gpio",
+	.id = -1,
+	.dev = {
+		.platform_data = &rdc321x_led_data,
+	}
+};
+
+/* Watchdog */
+static struct platform_device rdc321x_wdt = {
+	.name = "rdc321x-wdt",
+	.id = -1,
+	.num_resources = 0,
+};
+
+static struct platform_device *rdc321x_devs[] = {
+	&rdc321x_leds,
+	&rdc321x_wdt
+};
+
+static int __init rdc_board_setup(void)
+{
+	return platform_add_devices(rdc321x_devs, ARRAY_SIZE(rdc321x_devs));
+}
+
+arch_initcall(rdc_board_setup);
diff --git a/arch/x86/mach-rdc321x/wdt.c b/arch/x86/mach-rdc321x/wdt.c
new file mode 100644
index 000000000000..ec5625ae7061
--- /dev/null
+++ b/arch/x86/mach-rdc321x/wdt.c
@@ -0,0 +1,275 @@
+/*
+ * RDC321x watchdog driver
+ *
+ * Copyright (C) 2007 Florian Fainelli <florian@openwrt.org>
+ *
+ * This driver is highly inspired from the cpu5_wdt driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/miscdevice.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/timer.h>
+#include <linux/completion.h>
+#include <linux/jiffies.h>
+#include <linux/platform_device.h>
+#include <linux/watchdog.h>
+#include <linux/io.h>
+#include <linux/uaccess.h>
+
+#include <asm/mach-rdc321x/rdc321x_defs.h>
+
+#define RDC_WDT_MASK	0x80000000 /* Mask */
+#define RDC_WDT_EN	0x00800000 /* Enable bit */
+#define RDC_WDT_WTI	0x00200000 /* Generate CPU reset/NMI/WDT on timeout */
+#define RDC_WDT_RST	0x00100000 /* Reset bit */
+#define RDC_WDT_WIF	0x00040000 /* WDT IRQ Flag */
+#define RDC_WDT_IRT	0x00000100 /* IRQ Routing table */
+#define RDC_WDT_CNT	0x00000001 /* WDT count */
+
+#define RDC_CLS_TMR	0x80003844 /* Clear timer */
+
+#define RDC_WDT_INTERVAL	(HZ/10+1)
+
+int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static int ticks = 1000;
+
+/* some device data */
+
+static struct {
+	struct completion stop;
+	volatile int running;
+	struct timer_list timer;
+	volatile int queue;
+	int default_ticks;
+	unsigned long inuse;
+} rdc321x_wdt_device;
+
+/* generic helper functions */
+
+static void rdc321x_wdt_trigger(unsigned long unused)
+{
+	if (rdc321x_wdt_device.running)
+		ticks--;
+
+	/* keep watchdog alive */
+	outl(RDC_WDT_EN|inl(RDC3210_CFGREG_DATA), RDC3210_CFGREG_DATA);
+
+	/* requeue?? */
+	if (rdc321x_wdt_device.queue && ticks)
+		mod_timer(&rdc321x_wdt_device.timer,
+				jiffies + RDC_WDT_INTERVAL);
+	else {
+		/* ticks doesn't matter anyway */
+		complete(&rdc321x_wdt_device.stop);
+	}
+
+}
+
+static void rdc321x_wdt_reset(void)
+{
+	ticks = rdc321x_wdt_device.default_ticks;
+}
+
+static void rdc321x_wdt_start(void)
+{
+	if (!rdc321x_wdt_device.queue) {
+		rdc321x_wdt_device.queue = 1;
+
+		/* Clear the timer */
+		outl(RDC_CLS_TMR, RDC3210_CFGREG_ADDR);
+
+		/* Enable watchdog and set the timeout to 81.92 us */
+		outl(RDC_WDT_EN|RDC_WDT_CNT, RDC3210_CFGREG_DATA);
+
+		mod_timer(&rdc321x_wdt_device.timer,
+				jiffies + RDC_WDT_INTERVAL);
+	}
+
+	/* if process dies, counter is not decremented */
+	rdc321x_wdt_device.running++;
+}
+
+static int rdc321x_wdt_stop(void)
+{
+	if (rdc321x_wdt_device.running)
+		rdc321x_wdt_device.running = 0;
+
+	ticks = rdc321x_wdt_device.default_ticks;
+
+	return -EIO;
+}
+
+/* filesystem operations */
+
+static int rdc321x_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &rdc321x_wdt_device.inuse))
+		return -EBUSY;
+
+	return nonseekable_open(inode, file);
+}
+
+static int rdc321x_wdt_release(struct inode *inode, struct file *file)
+{
+	clear_bit(0, &rdc321x_wdt_device.inuse);
+	return 0;
+}
+
+static int rdc321x_wdt_ioctl(struct inode *inode, struct file *file,
+				unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	unsigned int value;
+	static struct watchdog_info ident = {
+		.options = WDIOF_CARDRESET,
+		.identity = "RDC321x WDT",
+	};
+
+	switch (cmd) {
+	case WDIOC_KEEPALIVE:
+		rdc321x_wdt_reset();
+		break;
+	case WDIOC_GETSTATUS:
+		/* Read the value from the DATA register */
+		value = inl(RDC3210_CFGREG_DATA);
+		if (copy_to_user(argp, &value, sizeof(int)))
+			return -EFAULT;
+		break;
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user(argp, &ident, sizeof(ident)))
+			return -EFAULT;
+		break;
+	case WDIOC_SETOPTIONS:
+		if (copy_from_user(&value, argp, sizeof(int)))
+			return -EFAULT;
+		switch (value) {
+		case WDIOS_ENABLECARD:
+			rdc321x_wdt_start();
+			break;
+		case WDIOS_DISABLECARD:
+			return rdc321x_wdt_stop();
+		default:
+			return -EINVAL;
+		}
+		break;
+	default:
+		return -ENOTTY;
+	}
+	return 0;
+}
+
+static ssize_t rdc321x_wdt_write(struct file *file, const char __user *buf,
+				size_t count, loff_t *ppos)
+{
+	if (!count)
+		return -EIO;
+
+	rdc321x_wdt_reset();
+
+	return count;
+}
+
+static const struct file_operations rdc321x_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.ioctl		= rdc321x_wdt_ioctl,
+	.open		= rdc321x_wdt_open,
+	.write		= rdc321x_wdt_write,
+	.release	= rdc321x_wdt_release,
+};
+
+static struct miscdevice rdc321x_wdt_misc = {
+	.minor	= WATCHDOG_MINOR,
+	.name	= "watchdog",
+	.fops	= &rdc321x_wdt_fops,
+};
+
+static int __devinit rdc321x_wdt_probe(struct platform_device *pdev)
+{
+	int err;
+
+	err = misc_register(&rdc321x_wdt_misc);
+	if (err < 0) {
+		printk(KERN_ERR PFX "watchdog misc_register failed\n");
+		return err;
+	}
+
+	/* Reset the watchdog */
+	outl(RDC_WDT_RST, RDC3210_CFGREG_DATA);
+
+	init_completion(&rdc321x_wdt_device.stop);
+	rdc321x_wdt_device.queue = 0;
+
+	clear_bit(0, &rdc321x_wdt_device.inuse);
+
+	setup_timer(&rdc321x_wdt_device.timer, rdc321x_wdt_trigger, 0);
+
+	rdc321x_wdt_device.default_ticks = ticks;
+
+	printk(KERN_INFO PFX "watchdog init success\n");
+
+	return 0;
+}
+
+static int rdc321x_wdt_remove(struct platform_device *pdev)
+{
+	if (rdc321x_wdt_device.queue) {
+		rdc321x_wdt_device.queue = 0;
+		wait_for_completion(&rdc321x_wdt_device.stop);
+	}
+
+	misc_deregister(&rdc321x_wdt_misc);
+
+	return 0;
+}
+
+static struct platform_driver rdc321x_wdt_driver = {
+	.probe = rdc321x_wdt_probe,
+	.remove = rdc321x_wdt_remove,
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = "rdc321x-wdt",
+	},
+};
+
+static int __init rdc321x_wdt_init(void)
+{
+	return platform_driver_register(&rdc321x_wdt_driver);
+}
+
+static void __exit rdc321x_wdt_exit(void)
+{
+	platform_driver_unregister(&rdc321x_wdt_driver);
+}
+
+module_init(rdc321x_wdt_init);
+module_exit(rdc321x_wdt_exit);
+
+MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
+MODULE_DESCRIPTION("RDC321x watchdog driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/include/asm-x86/mach-rdc321x/gpio.h b/include/asm-x86/mach-rdc321x/gpio.h
new file mode 100644
index 000000000000..db31b929b990
--- /dev/null
+++ b/include/asm-x86/mach-rdc321x/gpio.h
@@ -0,0 +1,56 @@
+#ifndef _RDC321X_GPIO_H
+#define _RDC321X_GPIO_H
+
+extern int rdc_gpio_get_value(unsigned gpio);
+extern void rdc_gpio_set_value(unsigned gpio, int value);
+extern int rdc_gpio_direction_input(unsigned gpio);
+extern int rdc_gpio_direction_output(unsigned gpio, int value);
+
+
+/* Wrappers for the arch-neutral GPIO API */
+
+static inline int gpio_request(unsigned gpio, const char *label)
+{
+	/* Not yet implemented */
+	return 0;
+}
+
+static inline void gpio_free(unsigned gpio)
+{
+	/* Not yet implemented */
+}
+
+static inline int gpio_direction_input(unsigned gpio)
+{
+	return rdc_gpio_direction_input(gpio);
+}
+
+static inline int gpio_direction_output(unsigned gpio, int value)
+{
+	return rdc_gpio_direction_output(gpio, value);
+}
+
+static inline int gpio_get_value(unsigned gpio)
+{
+	return rdc_gpio_get_value(gpio);
+}
+
+static inline void gpio_set_value(unsigned gpio, int value)
+{
+	rdc_gpio_set_value(gpio, value);
+}
+
+static inline int gpio_to_irq(unsigned gpio)
+{
+	return gpio;
+}
+
+static inline int irq_to_gpio(unsigned irq)
+{
+	return irq;
+}
+
+/* For cansleep */
+#include <asm-generic/gpio.h>
+
+#endif /* _RDC321X_GPIO_H_ */
diff --git a/include/asm-x86/mach-rdc321x/rdc321x_defs.h b/include/asm-x86/mach-rdc321x/rdc321x_defs.h
new file mode 100644
index 000000000000..838ba8f64fd3
--- /dev/null
+++ b/include/asm-x86/mach-rdc321x/rdc321x_defs.h
@@ -0,0 +1,6 @@
+#define PFX	"rdc321x: "
+
+/* General purpose configuration and data registers */
+#define RDC3210_CFGREG_ADDR     0x0CF8
+#define RDC3210_CFGREG_DATA     0x0CFC
+#define RDC_MAX_GPIO		0x3A
diff --git a/include/asm-x86/timex.h b/include/asm-x86/timex.h
index 39a21ab030f0..27cfd6c599ba 100644
--- a/include/asm-x86/timex.h
+++ b/include/asm-x86/timex.h
@@ -7,6 +7,8 @@
 
 #ifdef CONFIG_X86_ELAN
 #  define PIT_TICK_RATE 1189200 /* AMD Elan has different frequency! */
+#elif defined(CONFIG_X86_RDC321X)
+#  define PIT_TICK_RATE 1041667 /* Underlying HZ for R8610 */
 #else
 #  define PIT_TICK_RATE 1193182 /* Underlying HZ */
 #endif
-- 
cgit v1.2.3


From 6c435456dc91ace468b4e9d72ad0e13dafa22a45 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:39 +0100
Subject: x86: add mm parameter to paravirt_alloc_pd

Add mm to paravirt_alloc_pd, partly to make it consistent with
paravirt_alloc_pt, and because later changes will make use of it.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/vmi_32.c     | 2 +-
 arch/x86/mm/init_32.c        | 4 ++--
 arch/x86/mm/pgtable_32.c     | 4 +++-
 include/asm-x86/paravirt.h   | 6 +++---
 include/asm-x86/pgalloc_32.h | 3 +--
 5 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 2ee5d8e0ada5..4525bc2c2e19 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -398,7 +398,7 @@ static void vmi_allocate_pt(struct mm_struct *mm, u32 pfn)
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0);
 }
 
-static void vmi_allocate_pd(u32 pfn)
+static void vmi_allocate_pd(struct mm_struct *mm, u32 pfn)
 {
  	/*
 	 * This call comes in very early, before mem_map is setup.
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 04332c09ad1d..98d2acae4f64 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -65,7 +65,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
 	if (!(pgd_val(*pgd) & _PAGE_PRESENT)) {
 		pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
 
-		paravirt_alloc_pd(__pa(pmd_table) >> PAGE_SHIFT);
+		paravirt_alloc_pd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT);
 		set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
 		pud = pud_offset(pgd, 0);
 		if (pmd_table != pmd_offset(pud, 0))
@@ -365,7 +365,7 @@ void __init native_pagetable_setup_start(pgd_t *base)
 	memset(&base[USER_PTRS_PER_PGD], 0,
 	       KERNEL_PGD_PTRS * sizeof(pgd_t));
 #else
-	paravirt_alloc_pd(__pa(swapper_pg_dir) >> PAGE_SHIFT);
+	paravirt_alloc_pd(&init_mm, __pa(base) >> PAGE_SHIFT);
 #endif
 }
 
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index be61a1d845a4..f85ee44720d2 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -330,13 +330,15 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
 	if (PTRS_PER_PMD == 1 || !pgd)
 		return pgd;
 
+	mm->pgd = pgd;		/* so that alloc_pd can use it */
+
  	for (i = 0; i < UNSHARED_PTRS_PER_PGD; ++i) {
 		pmd_t *pmd = pmd_cache_alloc(i);
 
 		if (!pmd)
 			goto out_oom;
 
-		paravirt_alloc_pd(__pa(pmd) >> PAGE_SHIFT);
+		paravirt_alloc_pd(mm, __pa(pmd) >> PAGE_SHIFT);
 		set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
 	}
 	return pgd;
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 9f9ff57b3efb..d6236eb46466 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -221,7 +221,7 @@ struct pv_mmu_ops {
 
 	/* Hooks for allocating/releasing pagetable pages */
 	void (*alloc_pt)(struct mm_struct *mm, u32 pfn);
-	void (*alloc_pd)(u32 pfn);
+	void (*alloc_pd)(struct mm_struct *mm, u32 pfn);
 	void (*alloc_pd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count);
 	void (*release_pt)(u32 pfn);
 	void (*release_pd)(u32 pfn);
@@ -903,9 +903,9 @@ static inline void paravirt_release_pt(unsigned pfn)
 	PVOP_VCALL1(pv_mmu_ops.release_pt, pfn);
 }
 
-static inline void paravirt_alloc_pd(unsigned pfn)
+static inline void paravirt_alloc_pd(struct mm_struct *mm, unsigned pfn)
 {
-	PVOP_VCALL1(pv_mmu_ops.alloc_pd, pfn);
+	PVOP_VCALL2(pv_mmu_ops.alloc_pd, mm, pfn);
 }
 
 static inline void paravirt_alloc_pd_clone(unsigned pfn, unsigned clonepfn,
diff --git a/include/asm-x86/pgalloc_32.h b/include/asm-x86/pgalloc_32.h
index f2fc33ceb9f2..fbc6357f5eba 100644
--- a/include/asm-x86/pgalloc_32.h
+++ b/include/asm-x86/pgalloc_32.h
@@ -8,8 +8,7 @@
 #include <asm/paravirt.h>
 #else
 #define paravirt_alloc_pt(mm, pfn) do { } while (0)
-#define paravirt_alloc_pd(pfn) do { } while (0)
-#define paravirt_alloc_pd(pfn) do { } while (0)
+#define paravirt_alloc_pd(mm, pfn) do { } while (0)
 #define paravirt_alloc_pd_clone(pfn, clonepfn, start, count) do { } while (0)
 #define paravirt_release_pt(pfn) do { } while (0)
 #define paravirt_release_pd(pfn) do { } while (0)
-- 
cgit v1.2.3


From a5a19c63f4e55e32dc0bc3d936d7f94793d8b380 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:33:39 +0100
Subject: x86: demacro asm-x86/pgalloc_32.h

Convert macros into inline functions, for better type-checking.

This patch required a little bit of fiddling with headers in order to
make __(pte|pmd)_free_tlb inline rather than macros.
asm-generic/tlb.h includes asm/pgalloc.h, though it doesn't directly
use any pgalloc definitions.  I removed this include to avoid an
include cycle, but it may cause secondary compile failures by things
depending on the indirect inclusion; arch/x86/mm/hugetlbpage.c was one
such place; there may be others.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/hugetlbpage.c        |  3 +-
 arch/x86/mm/init_32.c            |  1 +
 include/asm-generic/tlb.h        |  1 -
 include/asm-x86/pgalloc_32.h     | 61 +++++++++++++++++++++++++---------------
 include/asm-x86/pgtable-3level.h |  2 --
 include/linux/swap.h             |  1 +
 6 files changed, 43 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index 6c06d9c0488e..4fbafb4bc2f0 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -15,6 +15,7 @@
 #include <asm/mman.h>
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
+#include <asm/pgalloc.h>
 
 static unsigned long page_table_shareable(struct vm_area_struct *svma,
 				struct vm_area_struct *vma,
@@ -88,7 +89,7 @@ static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud)
 
 	spin_lock(&mm->page_table_lock);
 	if (pud_none(*pud))
-		pud_populate(mm, pud, (unsigned long) spte & PAGE_MASK);
+		pud_populate(mm, pud, (pmd_t *)((unsigned long)spte & PAGE_MASK));
 	else
 		put_page(virt_to_page(spte));
 	spin_unlock(&mm->page_table_lock);
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 98d2acae4f64..459b384acda9 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -41,6 +41,7 @@
 #include <asm/apic.h>
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
+#include <asm/pgalloc.h>
 #include <asm/sections.h>
 #include <asm/paravirt.h>
 
diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h
index 75f2bfab614f..6ce9f3ab928d 100644
--- a/include/asm-generic/tlb.h
+++ b/include/asm-generic/tlb.h
@@ -15,7 +15,6 @@
 
 #include <linux/swap.h>
 #include <linux/quicklist.h>
-#include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
 
 /*
diff --git a/include/asm-x86/pgalloc_32.h b/include/asm-x86/pgalloc_32.h
index fbc6357f5eba..3482c3427897 100644
--- a/include/asm-x86/pgalloc_32.h
+++ b/include/asm-x86/pgalloc_32.h
@@ -3,6 +3,8 @@
 
 #include <linux/threads.h>
 #include <linux/mm.h>		/* for struct page */
+#include <asm/tlb.h>
+#include <asm-generic/tlb.h>
 
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
@@ -14,19 +16,20 @@
 #define paravirt_release_pd(pfn) do { } while (0)
 #endif
 
-#define pmd_populate_kernel(mm, pmd, pte)			\
-do {								\
-	paravirt_alloc_pt(mm, __pa(pte) >> PAGE_SHIFT);		\
-	set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)));		\
-} while (0)
+static inline void pmd_populate_kernel(struct mm_struct *mm,
+				       pmd_t *pmd, pte_t *pte)
+{
+	paravirt_alloc_pt(mm, __pa(pte) >> PAGE_SHIFT);
+	set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE));
+}
 
-#define pmd_populate(mm, pmd, pte) 				\
-do {								\
-	paravirt_alloc_pt(mm, page_to_pfn(pte));		\
-	set_pmd(pmd, __pmd(_PAGE_TABLE +			\
-		((unsigned long long)page_to_pfn(pte) <<	\
-			(unsigned long long) PAGE_SHIFT)));	\
-} while (0)
+static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *pte)
+{
+	unsigned long pfn = page_to_pfn(pte);
+
+	paravirt_alloc_pt(mm, pfn);
+	set_pmd(pmd, __pmd(((pteval_t)pfn << PAGE_SHIFT) | _PAGE_TABLE));
+}
 
 /*
  * Allocate and free page tables.
@@ -48,20 +51,34 @@ static inline void pte_free(struct page *pte)
 }
 
 
-#define __pte_free_tlb(tlb,pte) 					\
-do {									\
-	paravirt_release_pt(page_to_pfn(pte));				\
-	tlb_remove_page((tlb),(pte));					\
-} while (0)
+static inline void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte)
+{
+	paravirt_release_pt(page_to_pfn(pte));
+	tlb_remove_page(tlb, pte);
+}
 
 #ifdef CONFIG_X86_PAE
 /*
  * In the PAE case we free the pmds as part of the pgd.
  */
-#define pmd_alloc_one(mm, addr)		({ BUG(); ((pmd_t *)2); })
-#define pmd_free(x)			do { } while (0)
-#define __pmd_free_tlb(tlb,x)		do { } while (0)
-#define pud_populate(mm, pmd, pte)	BUG()
-#endif
+static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
+{
+	BUG();
+	return (pmd_t *)2;
+}
+
+static inline void pmd_free(pmd_t *pmd)
+{
+}
+
+static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
+{
+}
+
+static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
+{
+	BUG();
+}
+#endif	/* CONFIG_X86_PAE */
 
 #endif /* _I386_PGALLOC_H */
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 5af7a44ed902..62bb06575d5a 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -157,6 +157,4 @@ static inline unsigned long pte_pfn(pte_t pte)
 #define __pte_to_swp_entry(pte)		((swp_entry_t){ (pte).pte_high })
 #define __swp_entry_to_pte(x)		((pte_t){ { .pte_high = (x).val } })
 
-#define __pmd_free_tlb(tlb, x)		do { } while (0)
-
 #endif /* _I386_PGTABLE_3LEVEL_H */
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 4f3838adbb30..2c3ce4c69b25 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -6,6 +6,7 @@
 #include <linux/mmzone.h>
 #include <linux/list.h>
 #include <linux/sched.h>
+#include <linux/pagemap.h>
 
 #include <asm/atomic.h>
 #include <asm/page.h>
-- 
cgit v1.2.3


From 6371b495991debfd1417b17c2bc4f7d7bae05739 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:40 +0100
Subject: x86: change ioremap() to default to uncached

Prepare ioremap() to default to uncached. This will be the
safest - but first we have to fix CPA.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/io_32.h | 15 ++++++++++++---
 include/asm-x86/io_64.h | 23 +++++++++++++++++------
 2 files changed, 29 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index 2a04bd17eac5..db3978846379 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -111,18 +111,27 @@ extern void __iomem * __ioremap(unsigned long offset, unsigned long size, unsign
  * make bus memory CPU accessible via the readb/readw/readl/writeb/
  * writew/writel functions and the other mmio helpers. The returned
  * address is not guaranteed to be usable directly as a virtual
- * address. 
+ * address.
  *
  * If the area you are trying to map is a PCI BAR you should have a
  * look at pci_iomap().
  */
+extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
 
-static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
+static inline void __iomem *
+ioremap_cache(unsigned long offset, unsigned long size)
 {
 	return __ioremap(offset, size, 0);
 }
 
-extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
+/*
+ * The default ioremap() behavior is non-cached:
+ */
+static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
+{
+	return ioremap_nocache(offset, size);
+}
+
 extern void iounmap(volatile void __iomem *addr);
 
 /*
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index fef0ce2ced81..7dee3c6e9c39 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -152,11 +152,6 @@ static inline void * phys_to_virt(unsigned long address)
 
 extern void __iomem *__ioremap(unsigned long offset, unsigned long size, unsigned long flags);
 
-static inline void __iomem * ioremap (unsigned long offset, unsigned long size)
-{
-	return __ioremap(offset, size, 0);
-}
-
 extern void *early_ioremap(unsigned long addr, unsigned long size);
 extern void early_iounmap(void *addr, unsigned long size);
 
@@ -165,8 +160,24 @@ extern void early_iounmap(void *addr, unsigned long size);
  * it's useful if some control registers are in such an area and write combining
  * or read caching is not desirable:
  */
-extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size);
+extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
+
+static inline void __iomem *
+ioremap_cache(unsigned long offset, unsigned long size)
+{
+	return __ioremap(offset, size, 0);
+}
+
+/*
+ * The default ioremap() behavior is non-cached:
+ */
+static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
+{
+	return ioremap_cache(offset, size);
+}
+
 extern void iounmap(volatile void __iomem *addr);
+
 extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 
 /*
-- 
cgit v1.2.3


From 4c3c4b4513a361cc6ac5ee8677695260c4f0f25f Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:42 +0100
Subject: x86: clean up pte_exec

- Rename it to pte_exec() from pte_exec_kernel(). There is nothing
kernel specific in there.
- Move it into the common file because _PAGE_NX is 0 on !PAE and then
pte_exec() will be always evaluate to true.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/fault_32.c           | 2 +-
 arch/x86/mm/init_32.c            | 2 +-
 include/asm-x86/pgtable-2level.h | 8 --------
 include/asm-x86/pgtable-3level.h | 8 --------
 include/asm-x86/pgtable.h        | 1 +
 5 files changed, 3 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index 41c31968a74c..f7972ae7da07 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -615,7 +615,7 @@ no_context:
 		if (error_code & PF_INSTR) {
 			pte_t *pte = lookup_address(address);
 
-			if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
+			if (pte && pte_present(*pte) && !pte_exec(*pte))
 				printk(KERN_CRIT "kernel tried to execute "
 					"NX-protected page - exploit attempt? "
 					"(uid: %d)\n", current->uid);
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 724a5eff6f34..5080646da771 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -542,7 +542,7 @@ int __init set_kernel_exec(unsigned long vaddr, int enable)
 	pte = lookup_address(vaddr);
 	BUG_ON(!pte);
 
-	if (!pte_exec_kernel(*pte))
+	if (!pte_exec(*pte))
 		ret = 0;
 
 	if (enable)
diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index ce486bdbbbb7..701404fab308 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -55,14 +55,6 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 #define pte_none(x)		(!(x).pte_low)
 #define pte_pfn(x)		(pte_val(x) >> PAGE_SHIFT)
 
-/*
- * All present pages are kernel-executable:
- */
-static inline int pte_exec_kernel(pte_t pte)
-{
-	return 1;
-}
-
 /*
  * Bits 0, 6 and 7 are taken, split up the 29 bits of offset
  * into this range:
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 62bb06575d5a..62a1ffbc8784 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -19,14 +19,6 @@
 #define pud_bad(pud)				0
 #define pud_present(pud)			1
 
-/*
- * All present pages with !NX bit are kernel-executable:
- */
-static inline int pte_exec_kernel(pte_t pte)
-{
-	return !(pte_val(pte) & _PAGE_NX);
-}
-
 /* Rules for using set_pte: the pte being assigned *must* be
  * either not present or in a state where the hardware will
  * not attempt to update the pte.  In places where this is
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 75a656e6b3f8..2f3d90807efb 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -139,6 +139,7 @@ static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED;
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
 static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
+static inline int pte_exec(pte_t pte)		{ return !(pte_val(pte) & _PAGE_NX); }
 
 static inline int pmd_large(pmd_t pte) {
 	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-- 
cgit v1.2.3


From a5a5dc31794c3271c066835ad2c90c58a3805569 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:42 +0100
Subject: x86: add pte accessors for the global bit

Needed for some test code.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 2f3d90807efb..18a5f9e32834 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -139,6 +139,7 @@ static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED;
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
 static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
 static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
+static inline int pte_global(pte_t pte) 	{ return pte_val(pte) & _PAGE_GLOBAL; }
 static inline int pte_exec(pte_t pte)		{ return !(pte_val(pte) & _PAGE_NX); }
 
 static inline int pmd_large(pmd_t pte) {
@@ -155,6 +156,8 @@ static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_A
 static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
 static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
 static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
+static inline pte_t pte_mkglobal(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_GLOBAL); }
+static inline pte_t pte_clrglobal(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_GLOBAL); }
 
 extern pteval_t __supported_pte_mask;
 
-- 
cgit v1.2.3


From f0646e43acb18f0e00b00085dc88bc3f403e7930 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:43 +0100
Subject: x86: return the page table level in lookup_address()

based on this patch from Andi Kleen:

|  Subject: CPA: Return the page table level in lookup_address()
|  From: Andi Kleen <ak@suse.de>
|
|  Needed for the next change.
|
|  And change all the callers.

and ported it to x86.git.

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/fault_32.c       | 3 ++-
 arch/x86/mm/init_32.c        | 3 ++-
 arch/x86/mm/pageattr_32.c    | 7 +++++--
 arch/x86/mm/pageattr_64.c    | 7 +++++--
 arch/x86/xen/mmu.c           | 9 ++++++---
 include/asm-x86/pgtable_32.h | 2 +-
 include/asm-x86/pgtable_64.h | 2 +-
 7 files changed, 22 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/fault_32.c b/arch/x86/mm/fault_32.c
index f7972ae7da07..f4f8c324715f 100644
--- a/arch/x86/mm/fault_32.c
+++ b/arch/x86/mm/fault_32.c
@@ -613,7 +613,8 @@ no_context:
 
 #ifdef CONFIG_X86_PAE
 		if (error_code & PF_INSTR) {
-			pte_t *pte = lookup_address(address);
+			int level;
+			pte_t *pte = lookup_address(address, &level);
 
 			if (pte && pte_present(*pte) && !pte_exec(*pte))
 				printk(KERN_CRIT "kernel tried to execute "
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 5080646da771..206e3f6800b9 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -535,11 +535,12 @@ int __init set_kernel_exec(unsigned long vaddr, int enable)
 {
 	pte_t *pte;
 	int ret = 1;
+	int level;
 
 	if (!nx_enabled)
 		goto out;
 
-	pte = lookup_address(vaddr);
+	pte = lookup_address(vaddr, &level);
 	BUG_ON(!pte);
 
 	if (!pte_exec(*pte))
diff --git a/arch/x86/mm/pageattr_32.c b/arch/x86/mm/pageattr_32.c
index be4656403d77..523fd5b37df9 100644
--- a/arch/x86/mm/pageattr_32.c
+++ b/arch/x86/mm/pageattr_32.c
@@ -18,7 +18,7 @@
 static DEFINE_SPINLOCK(cpa_lock);
 static struct list_head df_list = LIST_HEAD_INIT(df_list);
 
-pte_t *lookup_address(unsigned long address)
+pte_t *lookup_address(unsigned long address, int *level)
 {
 	pgd_t *pgd = pgd_offset_k(address);
 	pud_t *pud;
@@ -32,8 +32,10 @@ pte_t *lookup_address(unsigned long address)
 	pmd = pmd_offset(pud, address);
 	if (pmd_none(*pmd))
 		return NULL;
+	*level = 2;
 	if (pmd_large(*pmd))
 		return (pte_t *)pmd;
+	*level = 3;
 
 	return pte_offset_kernel(pmd, address);
 }
@@ -156,11 +158,12 @@ static int __change_page_attr(struct page *page, pgprot_t prot)
 	struct page *kpte_page;
 	unsigned long address;
 	pte_t *kpte;
+	int level;
 
 	BUG_ON(PageHighMem(page));
 	address = (unsigned long)page_address(page);
 
-	kpte = lookup_address(address);
+	kpte = lookup_address(address, &level);
 	if (!kpte)
 		return -EINVAL;
 
diff --git a/arch/x86/mm/pageattr_64.c b/arch/x86/mm/pageattr_64.c
index 14ab327cde0c..59cd066f6741 100644
--- a/arch/x86/mm/pageattr_64.c
+++ b/arch/x86/mm/pageattr_64.c
@@ -14,7 +14,7 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
-pte_t *lookup_address(unsigned long address)
+pte_t *lookup_address(unsigned long address, int *level)
 {
 	pgd_t *pgd = pgd_offset_k(address);
 	pud_t *pud;
@@ -29,8 +29,10 @@ pte_t *lookup_address(unsigned long address)
 	pmd = pmd_offset(pud, address);
 	if (!pmd_present(*pmd))
 		return NULL;
+	*level = 3;
 	if (pmd_large(*pmd))
 		return (pte_t *)pmd;
+	*level = 4;
 
 	pte = pte_offset_kernel(pmd, address);
 	if (pte && !pte_present(*pte))
@@ -140,8 +142,9 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
 	struct page *kpte_page;
 	pgprot_t ref_prot2;
 	pte_t *kpte;
+	int level;
 
-	kpte = lookup_address(address);
+	kpte = lookup_address(address, &level);
 	if (!kpte)
 		return 0;
 
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index e6184735545f..45aa771e73a9 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -58,7 +58,8 @@
 
 xmaddr_t arbitrary_virt_to_machine(unsigned long address)
 {
-	pte_t *pte = lookup_address(address);
+	int level;
+	pte_t *pte = lookup_address(address, &level);
 	unsigned offset = address & PAGE_MASK;
 
 	BUG_ON(pte == NULL);
@@ -70,8 +71,9 @@ void make_lowmem_page_readonly(void *vaddr)
 {
 	pte_t *pte, ptev;
 	unsigned long address = (unsigned long)vaddr;
+	int level;
 
-	pte = lookup_address(address);
+	pte = lookup_address(address, &level);
 	BUG_ON(pte == NULL);
 
 	ptev = pte_wrprotect(*pte);
@@ -84,8 +86,9 @@ void make_lowmem_page_readwrite(void *vaddr)
 {
 	pte_t *pte, ptev;
 	unsigned long address = (unsigned long)vaddr;
+	int level;
 
-	pte = lookup_address(address);
+	pte = lookup_address(address, &level);
 	BUG_ON(pte == NULL);
 
 	ptev = pte_mkwrite(*pte);
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index eb8cccfa6a49..9381bd37b9b1 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -182,7 +182,7 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
  * NOTE: the return type is pte_t but if the pmd is PSE then we return it
  * as a pte too.
  */
-extern pte_t *lookup_address(unsigned long address);
+extern pte_t *lookup_address(unsigned long address, int *level);
 
 /*
  * Make a given kernel text page executable/non-executable.
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 29fdeb8111bb..139da50cd510 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -240,7 +240,7 @@ extern struct list_head pgd_list;
 
 extern int kern_addr_valid(unsigned long addr); 
 
-pte_t *lookup_address(unsigned long addr);
+pte_t *lookup_address(unsigned long addr, int *level);
 
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
-- 
cgit v1.2.3


From 0947b2f31ca1ea1211d3cde2dbd8fcec579ef395 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:33:44 +0100
Subject: i386 boot: replace boot_ioremap with enhanced bt_ioremap - enhance
 bt_ioremap

This patch makes it possible for bt_ioremap() to be used before
paging_init(), via providing an early implementation of set_fixmap()
that can be used before paging_init().

This way boot_ioremap() can be replaced by bt_ioremap().

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_32.c |  1 +
 arch/x86/mm/init_32.c      |  2 ++
 arch/x86/mm/ioremap_32.c   | 87 ++++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/io_32.h    |  3 ++
 4 files changed, 91 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 83ba3ca5f431..35db426de300 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -697,6 +697,7 @@ void __init setup_arch(char **cmdline_p)
 	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
 	pre_setup_arch_hook();
 	early_cpu_init();
+	bt_ioremap_init();
 
 #ifdef CONFIG_EFI
 	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 206e3f6800b9..f4e1894367a6 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -429,9 +429,11 @@ static void __init pagetable_init (void)
 	 * Fixed mappings, only the page table structure has to be
 	 * created - mappings will be set by set_fixmap():
 	 */
+	bt_ioremap_clear();
 	vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
 	end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK;
 	page_table_range_init(vaddr, end, pgd_base);
+	bt_ioremap_reset();
 
 	permanent_kmaps_init(pgd_base);
 
diff --git a/arch/x86/mm/ioremap_32.c b/arch/x86/mm/ioremap_32.c
index ef0f6a452ee1..fd1f5b6cfa20 100644
--- a/arch/x86/mm/ioremap_32.c
+++ b/arch/x86/mm/ioremap_32.c
@@ -208,6 +208,89 @@ void iounmap(volatile void __iomem *addr)
 }
 EXPORT_SYMBOL(iounmap);
 
+static __initdata int after_paging_init;
+static __initdata unsigned long bm_pte[1024]
+				__attribute__((aligned(PAGE_SIZE)));
+
+static inline unsigned long * __init bt_ioremap_pgd(unsigned long addr)
+{
+	return (unsigned long *)swapper_pg_dir + ((addr >> 22) & 1023);
+}
+
+static inline unsigned long * __init bt_ioremap_pte(unsigned long addr)
+{
+	return bm_pte + ((addr >> PAGE_SHIFT) & 1023);
+}
+
+void __init bt_ioremap_init(void)
+{
+	unsigned long *pgd;
+
+	pgd = bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
+	*pgd = __pa(bm_pte) | _PAGE_TABLE;
+	memset(bm_pte, 0, sizeof(bm_pte));
+	BUG_ON(pgd != bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_END)));
+}
+
+void __init bt_ioremap_clear(void)
+{
+	unsigned long *pgd;
+
+	pgd = bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
+	*pgd = 0;
+	__flush_tlb_all();
+}
+
+void __init bt_ioremap_reset(void)
+{
+	enum fixed_addresses idx;
+	unsigned long *pte, phys, addr;
+
+	after_paging_init = 1;
+	for (idx = FIX_BTMAP_BEGIN; idx <= FIX_BTMAP_END; idx--) {
+		addr = fix_to_virt(idx);
+		pte = bt_ioremap_pte(addr);
+		if (!*pte & _PAGE_PRESENT) {
+			phys = *pte & PAGE_MASK;
+			set_fixmap(idx, phys);
+		}
+	}
+}
+
+static void __init __bt_set_fixmap(enum fixed_addresses idx,
+				   unsigned long phys, pgprot_t flags)
+{
+	unsigned long *pte, addr = __fix_to_virt(idx);
+
+	if (idx >= __end_of_fixed_addresses) {
+		BUG();
+		return;
+	}
+	pte = bt_ioremap_pte(addr);
+	if (pgprot_val(flags))
+		*pte = (phys & PAGE_MASK) | pgprot_val(flags);
+	else
+		*pte = 0;
+	__flush_tlb_one(addr);
+}
+
+static inline void __init bt_set_fixmap(enum fixed_addresses idx,
+					unsigned long phys)
+{
+	if (after_paging_init)
+		set_fixmap(idx, phys);
+	else
+		__bt_set_fixmap(idx, phys, PAGE_KERNEL);
+}
+
+static inline void __init bt_clear_fixmap(enum fixed_addresses idx)
+{
+	if (after_paging_init)
+		clear_fixmap(idx);
+	else
+		__bt_set_fixmap(idx, 0, __pgprot(0));
+}
+
 void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
 {
 	unsigned long offset, last_addr;
@@ -244,7 +327,7 @@ void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
 	 */
 	idx = FIX_BTMAP_BEGIN;
 	while (nrpages > 0) {
-		set_fixmap(idx, phys_addr);
+		bt_set_fixmap(idx, phys_addr);
 		phys_addr += PAGE_SIZE;
 		--idx;
 		--nrpages;
@@ -267,7 +350,7 @@ void __init bt_iounmap(void *addr, unsigned long size)
 
 	idx = FIX_BTMAP_BEGIN;
 	while (nrpages > 0) {
-		clear_fixmap(idx);
+		bt_clear_fixmap(idx);
 		--idx;
 		--nrpages;
 	}
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index db3978846379..f7b733d9ace9 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -139,6 +139,9 @@ extern void iounmap(volatile void __iomem *addr);
  * mappings, before the real ioremap() is functional.
  * A boot-time mapping is currently limited to at most 16 pages.
  */
+extern void bt_ioremap_init(void);
+extern void bt_ioremap_clear(void);
+extern void bt_ioremap_reset(void);
 extern void *bt_ioremap(unsigned long offset, unsigned long size);
 extern void bt_iounmap(void *addr, unsigned long size);
 extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
-- 
cgit v1.2.3


From 4716e79c9946044a53a65418cfba04836f6a5c36 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:33:44 +0100
Subject: x86: replace boot_ioremap() with enhanced bt_ioremap() - remove
 boot_ioremap()

This patch replaces boot_ioremap invokation with bt_ioremap and
removes the boot_ioremap implementation.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig              |   6 ---
 arch/x86/kernel/srat_32.c     |   8 ++--
 arch/x86/mm/Makefile_32       |   1 -
 arch/x86/mm/boot_ioremap_32.c | 100 ------------------------------------------
 include/asm-x86/efi.h         |   8 +---
 5 files changed, 5 insertions(+), 118 deletions(-)
 delete mode 100644 arch/x86/mm/boot_ioremap_32.c

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 92dc919c7640..fb3eea3e38ee 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1009,12 +1009,6 @@ config IRQBALANCE
 	  The default yes will allow the kernel to do irq load balancing.
 	  Saying no will keep the kernel from doing irq load balancing.
 
-# turning this on wastes a bunch of space.
-# Summit needs it only when NUMA is on
-config BOOT_IOREMAP
-	def_bool y
-	depends on X86_32 && (((X86_SUMMIT || X86_GENERICARCH) && NUMA) || (X86 && EFI))
-
 config SECCOMP
 	def_bool y
 	prompt "Enable seccomp to safely compute untrusted bytecode"
diff --git a/arch/x86/kernel/srat_32.c b/arch/x86/kernel/srat_32.c
index 2a8713ec0f9a..24bfd4a9e62a 100644
--- a/arch/x86/kernel/srat_32.c
+++ b/arch/x86/kernel/srat_32.c
@@ -57,8 +57,6 @@ static struct node_memory_chunk_s node_memory_chunk[MAXCHUNKS];
 static int num_memory_chunks;		/* total number of memory chunks */
 static u8 __initdata apicid_to_pxm[MAX_APICID];
 
-extern void * boot_ioremap(unsigned long, unsigned long);
-
 /* Identify CPU proximity domains */
 static void __init parse_cpu_affinity_structure(char *p)
 {
@@ -299,7 +297,7 @@ int __init get_memcfg_from_srat(void)
 	}
 
 	rsdt = (struct acpi_table_rsdt *)
-	    boot_ioremap(rsdp->rsdt_physical_address, sizeof(struct acpi_table_rsdt));
+	    bt_ioremap(rsdp->rsdt_physical_address, sizeof(struct acpi_table_rsdt));
 
 	if (!rsdt) {
 		printk(KERN_WARNING
@@ -339,11 +337,11 @@ int __init get_memcfg_from_srat(void)
 	for (i = 0; i < tables; i++) {
 		/* Map in header, then map in full table length. */
 		header = (struct acpi_table_header *)
-			boot_ioremap(saved_rsdt.table.table_offset_entry[i], sizeof(struct acpi_table_header));
+			bt_ioremap(saved_rsdt.table.table_offset_entry[i], sizeof(struct acpi_table_header));
 		if (!header)
 			break;
 		header = (struct acpi_table_header *)
-			boot_ioremap(saved_rsdt.table.table_offset_entry[i], header->length);
+			bt_ioremap(saved_rsdt.table.table_offset_entry[i], header->length);
 		if (!header)
 			break;
 
diff --git a/arch/x86/mm/Makefile_32 b/arch/x86/mm/Makefile_32
index be2f55160bf8..1aeba3bf34bd 100644
--- a/arch/x86/mm/Makefile_32
+++ b/arch/x86/mm/Makefile_32
@@ -8,4 +8,3 @@ obj-$(CONFIG_CPA_DEBUG) += pageattr-test.o
 obj-$(CONFIG_NUMA) += discontig_32.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_HIGHMEM) += highmem_32.o
-obj-$(CONFIG_BOOT_IOREMAP) += boot_ioremap_32.o
diff --git a/arch/x86/mm/boot_ioremap_32.c b/arch/x86/mm/boot_ioremap_32.c
deleted file mode 100644
index 7ede1e3e1f42..000000000000
--- a/arch/x86/mm/boot_ioremap_32.c
+++ /dev/null
@@ -1,100 +0,0 @@
-/*
- * arch/i386/mm/boot_ioremap.c
- *
- * Re-map functions for early boot-time before paging_init() when the
- * boot-time pagetables are still in use
- *
- * Written by Dave Hansen <haveblue@us.ibm.com>
- */
-
-
-/*
- * We need to use the 2-level pagetable functions, but CONFIG_X86_PAE
- * keeps that from happening.  If anyone has a better way, I'm listening.
- *
- * boot_pte_t is defined only if this all works correctly
- */
-
-#undef CONFIG_X86_PAE
-#undef CONFIG_PARAVIRT
-#include <asm/page.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <linux/init.h>
-#include <linux/stddef.h>
-
-/*
- * I'm cheating here.  It is known that the two boot PTE pages are
- * allocated next to each other.  I'm pretending that they're just
- * one big array.
- */
-
-#define BOOT_PTE_PTRS (PTRS_PER_PTE*2)
-
-static unsigned long boot_pte_index(unsigned long vaddr)
-{
-	return __pa(vaddr) >> PAGE_SHIFT;
-}
-
-static inline boot_pte_t* boot_vaddr_to_pte(void *address)
-{
-	boot_pte_t* boot_pg = (boot_pte_t*)pg0;
-	return &boot_pg[boot_pte_index((unsigned long)address)];
-}
-
-/*
- * This is only for a caller who is clever enough to page-align
- * phys_addr and virtual_source, and who also has a preference
- * about which virtual address from which to steal ptes
- */
-static void __boot_ioremap(unsigned long phys_addr, unsigned long nrpages,
-		    void* virtual_source)
-{
-	boot_pte_t* pte;
-	int i;
-	char *vaddr = virtual_source;
-
-	pte = boot_vaddr_to_pte(virtual_source);
-	for (i=0; i < nrpages; i++, phys_addr += PAGE_SIZE, pte++) {
-		set_pte(pte, pfn_pte(phys_addr>>PAGE_SHIFT, PAGE_KERNEL));
-		__flush_tlb_one((unsigned long) &vaddr[i*PAGE_SIZE]);
-	}
-}
-
-/* the virtual space we're going to remap comes from this array */
-#define BOOT_IOREMAP_PAGES 4
-#define BOOT_IOREMAP_SIZE (BOOT_IOREMAP_PAGES*PAGE_SIZE)
-static __initdata char boot_ioremap_space[BOOT_IOREMAP_SIZE]
-		       __attribute__ ((aligned (PAGE_SIZE)));
-
-/*
- * This only applies to things which need to ioremap before paging_init()
- * bt_ioremap() and plain ioremap() are both useless at this point.
- *
- * When used, we're still using the boot-time pagetables, which only
- * have 2 PTE pages mapping the first 8MB
- *
- * There is no unmap.  The boot-time PTE pages aren't used after boot.
- * If you really want the space back, just remap it yourself.
- * boot_ioremap(&ioremap_space-PAGE_OFFSET, BOOT_IOREMAP_SIZE)
- */
-__init void* boot_ioremap(unsigned long phys_addr, unsigned long size)
-{
-	unsigned long last_addr, offset;
-	unsigned int nrpages;
-
-	last_addr = phys_addr + size - 1;
-
-	/* page align the requested address */
-	offset = phys_addr & ~PAGE_MASK;
-	phys_addr &= PAGE_MASK;
-	size = PAGE_ALIGN(last_addr) - phys_addr;
-
-	nrpages = size >> PAGE_SHIFT;
-	if (nrpages > BOOT_IOREMAP_PAGES)
-		return NULL;
-
-	__boot_ioremap(phys_addr, nrpages, boot_ioremap_space);
-
-	return &boot_ioremap_space[offset];
-}
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index 6d54502755aa..10fcf20bdc73 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -32,13 +32,9 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...);
 	efi_call_virt(f, a1, a2, a3, a4, a5)
 #define efi_call_virt6(f, a1, a2, a3, a4, a5, a6)	\
 	efi_call_virt(f, a1, a2, a3, a4, a5, a6)
-/*
- * We require an early boot_ioremap mapping mechanism initially
- */
-extern void *boot_ioremap(unsigned long, unsigned long);
 
-#define efi_early_ioremap(addr, size)		boot_ioremap(addr, size)
-#define efi_early_iounmap(vaddr, size)
+#define efi_early_ioremap(addr, size)		bt_ioremap(addr, size)
+#define efi_early_iounmap(vaddr, size)		bt_iounmap(vaddr, size)
 
 #define efi_ioremap(addr, size)			ioremap(addr, size)
 
-- 
cgit v1.2.3


From beacfaac3f23b30814aafee37a055257c7062ef3 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:33:44 +0100
Subject: x86 32-bit boot: rename bt_ioremap() to early_ioremap()

This patch renames bt_ioremap to early_ioremap, which is used in
x86_64. This makes it easier to merge i386 and x86_64 usage.

[ mingo@elte.hu: fix ]

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/efi.c      | 24 ++++++++++++------------
 arch/x86/kernel/efi_32.c   |  2 +-
 arch/x86/kernel/setup_32.c |  6 +++---
 arch/x86/kernel/srat_32.c  |  6 +++---
 arch/x86/mm/init_32.c      |  4 ++--
 arch/x86/mm/ioremap_32.c   | 38 +++++++++++++++++++-------------------
 include/asm-x86/dmi.h      |  7 ++-----
 include/asm-x86/efi.h      |  8 --------
 include/asm-x86/io_32.h    | 16 ++++++++--------
 9 files changed, 50 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 2939b015c2ed..5d492f99d967 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -256,12 +256,12 @@ void __init efi_init(void)
 	memmap.desc_version = boot_params.efi_info.efi_memdesc_version;
 	memmap.desc_size = boot_params.efi_info.efi_memdesc_size;
 
-	efi.systab = efi_early_ioremap((unsigned long)efi_phys.systab,
-				       sizeof(efi_system_table_t));
+	efi.systab = early_ioremap((unsigned long)efi_phys.systab,
+				   sizeof(efi_system_table_t));
 	if (efi.systab == NULL)
 		printk(KERN_ERR "Couldn't map the EFI system table!\n");
 	memcpy(&efi_systab, efi.systab, sizeof(efi_system_table_t));
-	efi_early_iounmap(efi.systab, sizeof(efi_system_table_t));
+	early_iounmap(efi.systab, sizeof(efi_system_table_t));
 	efi.systab = &efi_systab;
 
 	/*
@@ -278,14 +278,14 @@ void __init efi_init(void)
 	/*
 	 * Show what we know for posterity
 	 */
-	c16 = tmp = efi_early_ioremap(efi.systab->fw_vendor, 2);
+	c16 = tmp = early_ioremap(efi.systab->fw_vendor, 2);
 	if (c16) {
 		for (i = 0; i < sizeof(vendor) && *c16; ++i)
 			vendor[i] = *c16++;
 		vendor[i] = '\0';
 	} else
 		printk(KERN_ERR PFX "Could not map the firmware vendor!\n");
-	efi_early_iounmap(tmp, 2);
+	early_iounmap(tmp, 2);
 
 	printk(KERN_INFO "EFI v%u.%.02u by %s \n",
 	       efi.systab->hdr.revision >> 16,
@@ -294,7 +294,7 @@ void __init efi_init(void)
 	/*
 	 * Let's see what config tables the firmware passed to us.
 	 */
-	config_tables = efi_early_ioremap(
+	config_tables = early_ioremap(
 		efi.systab->tables,
 		efi.systab->nr_tables * sizeof(efi_config_table_t));
 	if (config_tables == NULL)
@@ -328,7 +328,7 @@ void __init efi_init(void)
 		}
 	}
 	printk("\n");
-	efi_early_iounmap(config_tables,
+	early_iounmap(config_tables,
 			  efi.systab->nr_tables * sizeof(efi_config_table_t));
 
 	/*
@@ -337,8 +337,8 @@ void __init efi_init(void)
 	 * address of several of the EFI runtime functions, needed to
 	 * set the firmware into virtual mode.
 	 */
-	runtime = efi_early_ioremap((unsigned long)efi.systab->runtime,
-				    sizeof(efi_runtime_services_t));
+	runtime = early_ioremap((unsigned long)efi.systab->runtime,
+				sizeof(efi_runtime_services_t));
 	if (runtime != NULL) {
 		/*
 		 * We will only need *early* access to the following
@@ -357,11 +357,11 @@ void __init efi_init(void)
 	} else
 		printk(KERN_ERR "Could not map the EFI runtime service "
 		       "table!\n");
-	efi_early_iounmap(runtime, sizeof(efi_runtime_services_t));
+	early_iounmap(runtime, sizeof(efi_runtime_services_t));
 
 	/* Map the EFI memory map */
-	memmap.map = efi_early_ioremap((unsigned long)memmap.phys_map,
-				       memmap.nr_map * memmap.desc_size);
+	memmap.map = early_ioremap((unsigned long)memmap.phys_map,
+				   memmap.nr_map * memmap.desc_size);
 	if (memmap.map == NULL)
 		printk(KERN_ERR "Could not map the EFI memory map!\n");
 	memmap.map_end = memmap.map + (memmap.nr_map * memmap.desc_size);
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index afd2c3b039d6..114b896d7573 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -117,7 +117,7 @@ void __init efi_map_memmap(void)
 {
 	memmap.map = NULL;
 
-	memmap.map = bt_ioremap((unsigned long) memmap.phys_map,
+	memmap.map = early_ioremap((unsigned long) memmap.phys_map,
 			(memmap.nr_map * memmap.desc_size));
 	if (memmap.map == NULL)
 		printk(KERN_ERR "Could not remap the EFI memmap!\n");
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 35db426de300..c038b09b1723 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -577,9 +577,9 @@ static void __init relocate_initrd(void)
 		if (clen > MAX_MAP_CHUNK-slop)
 			clen = MAX_MAP_CHUNK-slop;
 		mapaddr = ramdisk_image & PAGE_MASK;
-		p = bt_ioremap(mapaddr, clen+slop);
+		p = early_ioremap(mapaddr, clen+slop);
 		memcpy(q, p+slop, clen);
-		bt_iounmap(p, clen+slop);
+		early_iounmap(p, clen+slop);
 		q += clen;
 		ramdisk_image += clen;
 		ramdisk_size  -= clen;
@@ -697,7 +697,7 @@ void __init setup_arch(char **cmdline_p)
 	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
 	pre_setup_arch_hook();
 	early_cpu_init();
-	bt_ioremap_init();
+	early_ioremap_init();
 
 #ifdef CONFIG_EFI
 	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
diff --git a/arch/x86/kernel/srat_32.c b/arch/x86/kernel/srat_32.c
index 24bfd4a9e62a..2bf6903cb444 100644
--- a/arch/x86/kernel/srat_32.c
+++ b/arch/x86/kernel/srat_32.c
@@ -297,7 +297,7 @@ int __init get_memcfg_from_srat(void)
 	}
 
 	rsdt = (struct acpi_table_rsdt *)
-	    bt_ioremap(rsdp->rsdt_physical_address, sizeof(struct acpi_table_rsdt));
+	    early_ioremap(rsdp->rsdt_physical_address, sizeof(struct acpi_table_rsdt));
 
 	if (!rsdt) {
 		printk(KERN_WARNING
@@ -337,11 +337,11 @@ int __init get_memcfg_from_srat(void)
 	for (i = 0; i < tables; i++) {
 		/* Map in header, then map in full table length. */
 		header = (struct acpi_table_header *)
-			bt_ioremap(saved_rsdt.table.table_offset_entry[i], sizeof(struct acpi_table_header));
+			early_ioremap(saved_rsdt.table.table_offset_entry[i], sizeof(struct acpi_table_header));
 		if (!header)
 			break;
 		header = (struct acpi_table_header *)
-			bt_ioremap(saved_rsdt.table.table_offset_entry[i], header->length);
+			early_ioremap(saved_rsdt.table.table_offset_entry[i], header->length);
 		if (!header)
 			break;
 
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index f4e1894367a6..e00c1d7128b1 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -429,11 +429,11 @@ static void __init pagetable_init (void)
 	 * Fixed mappings, only the page table structure has to be
 	 * created - mappings will be set by set_fixmap():
 	 */
-	bt_ioremap_clear();
+	early_ioremap_clear();
 	vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
 	end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK;
 	page_table_range_init(vaddr, end, pgd_base);
-	bt_ioremap_reset();
+	early_ioremap_reset();
 
 	permanent_kmaps_init(pgd_base);
 
diff --git a/arch/x86/mm/ioremap_32.c b/arch/x86/mm/ioremap_32.c
index fd1f5b6cfa20..43b5e9fd6370 100644
--- a/arch/x86/mm/ioremap_32.c
+++ b/arch/x86/mm/ioremap_32.c
@@ -212,36 +212,36 @@ static __initdata int after_paging_init;
 static __initdata unsigned long bm_pte[1024]
 				__attribute__((aligned(PAGE_SIZE)));
 
-static inline unsigned long * __init bt_ioremap_pgd(unsigned long addr)
+static inline unsigned long * __init early_ioremap_pgd(unsigned long addr)
 {
 	return (unsigned long *)swapper_pg_dir + ((addr >> 22) & 1023);
 }
 
-static inline unsigned long * __init bt_ioremap_pte(unsigned long addr)
+static inline unsigned long * __init early_ioremap_pte(unsigned long addr)
 {
 	return bm_pte + ((addr >> PAGE_SHIFT) & 1023);
 }
 
-void __init bt_ioremap_init(void)
+void __init early_ioremap_init(void)
 {
 	unsigned long *pgd;
 
-	pgd = bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
+	pgd = early_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
 	*pgd = __pa(bm_pte) | _PAGE_TABLE;
 	memset(bm_pte, 0, sizeof(bm_pte));
-	BUG_ON(pgd != bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_END)));
+	BUG_ON(pgd != early_ioremap_pgd(fix_to_virt(FIX_BTMAP_END)));
 }
 
-void __init bt_ioremap_clear(void)
+void __init early_ioremap_clear(void)
 {
 	unsigned long *pgd;
 
-	pgd = bt_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
+	pgd = early_ioremap_pgd(fix_to_virt(FIX_BTMAP_BEGIN));
 	*pgd = 0;
 	__flush_tlb_all();
 }
 
-void __init bt_ioremap_reset(void)
+void __init early_ioremap_reset(void)
 {
 	enum fixed_addresses idx;
 	unsigned long *pte, phys, addr;
@@ -249,7 +249,7 @@ void __init bt_ioremap_reset(void)
 	after_paging_init = 1;
 	for (idx = FIX_BTMAP_BEGIN; idx <= FIX_BTMAP_END; idx--) {
 		addr = fix_to_virt(idx);
-		pte = bt_ioremap_pte(addr);
+		pte = early_ioremap_pte(addr);
 		if (!*pte & _PAGE_PRESENT) {
 			phys = *pte & PAGE_MASK;
 			set_fixmap(idx, phys);
@@ -257,7 +257,7 @@ void __init bt_ioremap_reset(void)
 	}
 }
 
-static void __init __bt_set_fixmap(enum fixed_addresses idx,
+static void __init __early_set_fixmap(enum fixed_addresses idx,
 				   unsigned long phys, pgprot_t flags)
 {
 	unsigned long *pte, addr = __fix_to_virt(idx);
@@ -266,7 +266,7 @@ static void __init __bt_set_fixmap(enum fixed_addresses idx,
 		BUG();
 		return;
 	}
-	pte = bt_ioremap_pte(addr);
+	pte = early_ioremap_pte(addr);
 	if (pgprot_val(flags))
 		*pte = (phys & PAGE_MASK) | pgprot_val(flags);
 	else
@@ -274,24 +274,24 @@ static void __init __bt_set_fixmap(enum fixed_addresses idx,
 	__flush_tlb_one(addr);
 }
 
-static inline void __init bt_set_fixmap(enum fixed_addresses idx,
+static inline void __init early_set_fixmap(enum fixed_addresses idx,
 					unsigned long phys)
 {
 	if (after_paging_init)
 		set_fixmap(idx, phys);
 	else
-		__bt_set_fixmap(idx, phys, PAGE_KERNEL);
+		__early_set_fixmap(idx, phys, PAGE_KERNEL);
 }
 
-static inline void __init bt_clear_fixmap(enum fixed_addresses idx)
+static inline void __init early_clear_fixmap(enum fixed_addresses idx)
 {
 	if (after_paging_init)
 		clear_fixmap(idx);
 	else
-		__bt_set_fixmap(idx, 0, __pgprot(0));
+		__early_set_fixmap(idx, 0, __pgprot(0));
 }
 
-void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
+void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 {
 	unsigned long offset, last_addr;
 	unsigned int nrpages;
@@ -327,7 +327,7 @@ void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
 	 */
 	idx = FIX_BTMAP_BEGIN;
 	while (nrpages > 0) {
-		bt_set_fixmap(idx, phys_addr);
+		early_set_fixmap(idx, phys_addr);
 		phys_addr += PAGE_SIZE;
 		--idx;
 		--nrpages;
@@ -335,7 +335,7 @@ void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
 	return (void*) (offset + fix_to_virt(FIX_BTMAP_BEGIN));
 }
 
-void __init bt_iounmap(void *addr, unsigned long size)
+void __init early_iounmap(void *addr, unsigned long size)
 {
 	unsigned long virt_addr;
 	unsigned long offset;
@@ -350,7 +350,7 @@ void __init bt_iounmap(void *addr, unsigned long size)
 
 	idx = FIX_BTMAP_BEGIN;
 	while (nrpages > 0) {
-		bt_clear_fixmap(idx);
+		early_clear_fixmap(idx);
 		--idx;
 		--nrpages;
 	}
diff --git a/include/asm-x86/dmi.h b/include/asm-x86/dmi.h
index 5008c365e6e4..1241e6ad1935 100644
--- a/include/asm-x86/dmi.h
+++ b/include/asm-x86/dmi.h
@@ -5,9 +5,6 @@
 
 #ifdef CONFIG_X86_32
 
-/* Use early IO mappings for DMI because it's initialized early */
-#define dmi_ioremap bt_ioremap
-#define dmi_iounmap bt_iounmap
 #define dmi_alloc alloc_bootmem
 
 #else /* CONFIG_X86_32 */
@@ -28,9 +25,9 @@ static inline void *dmi_alloc(unsigned len)
 	return dmi_alloc_data + idx;
 }
 
+#endif
+
 #define dmi_ioremap early_ioremap
 #define dmi_iounmap early_iounmap
 
 #endif
-
-#endif
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index 10fcf20bdc73..8380131ca542 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -33,9 +33,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...);
 #define efi_call_virt6(f, a1, a2, a3, a4, a5, a6)	\
 	efi_call_virt(f, a1, a2, a3, a4, a5, a6)
 
-#define efi_early_ioremap(addr, size)		bt_ioremap(addr, size)
-#define efi_early_iounmap(vaddr, size)		bt_iounmap(vaddr, size)
-
 #define efi_ioremap(addr, size)			ioremap(addr, size)
 
 #define end_pfn_map				max_low_pfn
@@ -91,13 +88,8 @@ extern u64 efi_call6(void *fp, u64 arg1, u64 arg2, u64 arg3,
 	efi_call6((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \
 		  (u64)(a3), (u64)(a4), (u64)(a5), (u64)(a6))
 
-#define efi_early_ioremap(addr, size)		early_ioremap(addr, size)
-#define efi_early_iounmap(vaddr, size)		early_iounmap(vaddr, size)
-
 extern void *efi_ioremap(unsigned long offset, unsigned long size);
 
-extern int efi_time;
-
 #endif /* CONFIG_X86_32 */
 
 extern void efi_reserve_bootmem(void);
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index f7b733d9ace9..059a1fee4de3 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -135,20 +135,20 @@ static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
 extern void iounmap(volatile void __iomem *addr);
 
 /*
- * bt_ioremap() and bt_iounmap() are for temporary early boot-time
+ * early_ioremap() and early_iounmap() are for temporary early boot-time
  * mappings, before the real ioremap() is functional.
  * A boot-time mapping is currently limited to at most 16 pages.
  */
-extern void bt_ioremap_init(void);
-extern void bt_ioremap_clear(void);
-extern void bt_ioremap_reset(void);
-extern void *bt_ioremap(unsigned long offset, unsigned long size);
-extern void bt_iounmap(void *addr, unsigned long size);
+extern void early_ioremap_init(void);
+extern void early_ioremap_clear(void);
+extern void early_ioremap_reset(void);
+extern void *early_ioremap(unsigned long offset, unsigned long size);
+extern void early_iounmap(void *addr, unsigned long size);
 extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 
 /* Use early IO mappings for DMI because it's initialized early */
-#define dmi_ioremap bt_ioremap
-#define dmi_iounmap bt_iounmap
+#define dmi_ioremap early_ioremap
+#define dmi_iounmap early_iounmap
 #define dmi_alloc alloc_bootmem
 
 /*
-- 
cgit v1.2.3


From 1b42f51630c7eebce6fb780b480731eb81afd325 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:45 +0100
Subject: x86: enhance early_ioremap()

 - allow nesting of up to 4 levels

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/ioremap_32.c    | 36 ++++++++++++++++++++++++++----------
 include/asm-x86/fixmap_32.h |  9 +++++----
 2 files changed, 31 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap_32.c b/arch/x86/mm/ioremap_32.c
index bfd7b8b2fe60..63f76ecae44c 100644
--- a/arch/x86/mm/ioremap_32.c
+++ b/arch/x86/mm/ioremap_32.c
@@ -291,23 +291,28 @@ static inline void __init early_clear_fixmap(enum fixed_addresses idx)
 		__early_set_fixmap(idx, 0, __pgprot(0));
 }
 
+
+int __initdata early_ioremap_nested;
+
 void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 {
 	unsigned long offset, last_addr;
-	unsigned int nrpages;
-	enum fixed_addresses idx;
+	unsigned int nrpages, nesting;
+	enum fixed_addresses idx0, idx;
+
+	WARN_ON(system_state != SYSTEM_BOOTING);
+
+	nesting = early_ioremap_nested;
 
 	/* Don't allow wraparound or zero size */
 	last_addr = phys_addr + size - 1;
 	if (!size || last_addr < phys_addr)
 		return NULL;
 
-	/*
-	 * Don't remap the low PCI/ISA area, it's always mapped..
-	 */
-	if (phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
-		return phys_to_virt(phys_addr);
+	if (nesting >= FIX_BTMAPS_NESTING)
+		return NULL;
 
+	early_ioremap_nested++;
 	/*
 	 * Mappings have to be page-aligned
 	 */
@@ -325,14 +330,16 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 	/*
 	 * Ok, go for it..
 	 */
-	idx = FIX_BTMAP_BEGIN;
+	idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting;
+	idx = idx0;
 	while (nrpages > 0) {
 		early_set_fixmap(idx, phys_addr);
 		phys_addr += PAGE_SIZE;
 		--idx;
 		--nrpages;
 	}
-	return (void*) (offset + fix_to_virt(FIX_BTMAP_BEGIN));
+
+	return (void*) (offset + fix_to_virt(idx0));
 }
 
 void __init early_iounmap(void *addr, unsigned long size)
@@ -341,17 +348,26 @@ void __init early_iounmap(void *addr, unsigned long size)
 	unsigned long offset;
 	unsigned int nrpages;
 	enum fixed_addresses idx;
+	unsigned int nesting;
+
+	nesting = --early_ioremap_nested;
 
 	virt_addr = (unsigned long)addr;
 	if (virt_addr < fix_to_virt(FIX_BTMAP_BEGIN))
 		return;
+
 	offset = virt_addr & ~PAGE_MASK;
 	nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT;
 
-	idx = FIX_BTMAP_BEGIN;
+	idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting;
 	while (nrpages > 0) {
 		early_clear_fixmap(idx);
 		--idx;
 		--nrpages;
 	}
 }
+
+void __this_fixmap_does_not_exist(void)
+{
+	WARN_ON(1);
+}
diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index 249e753ac805..0e990218a09c 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -65,7 +65,7 @@ enum fixed_addresses {
 #endif
 #ifdef CONFIG_X86_VISWS_APIC
 	FIX_CO_CPU,	/* Cobalt timer */
-	FIX_CO_APIC,	/* Cobalt APIC Redirection Table */ 
+	FIX_CO_APIC,	/* Cobalt APIC Redirection Table */
 	FIX_LI_PCIA,	/* Lithium PCI Bridge A */
 	FIX_LI_PCIB,	/* Lithium PCI Bridge B */
 #endif
@@ -74,7 +74,7 @@ enum fixed_addresses {
 #endif
 #ifdef CONFIG_X86_CYCLONE_TIMER
 	FIX_CYCLONE_TIMER, /*cyclone timer register*/
-#endif 
+#endif
 #ifdef CONFIG_HIGHMEM
 	FIX_KMAP_BEGIN,	/* reserved pte's for temporary kernel mappings */
 	FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
@@ -91,9 +91,10 @@ enum fixed_addresses {
 #endif
 	__end_of_permanent_fixed_addresses,
 	/* temporary boot-time mappings, used before ioremap() is functional */
-#define NR_FIX_BTMAPS	16
+#define NR_FIX_BTMAPS		16
+#define FIX_BTMAPS_NESTING	4
 	FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS - 1,
+	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
 	FIX_WP_TEST,
 	__end_of_fixed_addresses
 };
-- 
cgit v1.2.3


From a8efa1cd51d6f6407df5f42d8f1a7e9fc7178d1a Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:45 +0100
Subject: x86: increase the number of boot-mappings

increase max early_ioremap() remapping size from 64K to 256K.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/fixmap_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index 0e990218a09c..46bedb7b5426 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -91,7 +91,7 @@ enum fixed_addresses {
 #endif
 	__end_of_permanent_fixed_addresses,
 	/* temporary boot-time mappings, used before ioremap() is functional */
-#define NR_FIX_BTMAPS		16
+#define NR_FIX_BTMAPS		64
 #define FIX_BTMAPS_NESTING	4
 	FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
 	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
-- 
cgit v1.2.3


From d701fda8601fe267fbd3648f108f0e751305101b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:48 +0100
Subject: x86: fix early_ioremap()/btmap

fix a long-standing weakness of the early-ioremap allocator: it
uses a single pgd entry for the boot mappings, and was not properly
protecting itself against crossing a 2MB (4MB) boundary.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/fixmap_32.h | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index 46bedb7b5426..fde140fd6d95 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -90,10 +90,18 @@ enum fixed_addresses {
 	FIX_PARAVIRT_BOOTMAP,
 #endif
 	__end_of_permanent_fixed_addresses,
-	/* temporary boot-time mappings, used before ioremap() is functional */
+	/*
+	 * 256 temporary boot-time mappings, used by early_ioremap(),
+	 * before ioremap() is functional.
+	 *
+	 * We round it up to the next 512 pages boundary so that we
+	 * can have a single pgd entry and a single pte table:
+	 */
 #define NR_FIX_BTMAPS		64
 #define FIX_BTMAPS_NESTING	4
-	FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
+	FIX_BTMAP_END =
+		__end_of_permanent_fixed_addresses + 512 -
+			(__end_of_permanent_fixed_addresses & 511),
 	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
 	FIX_WP_TEST,
 	__end_of_fixed_addresses
-- 
cgit v1.2.3


From c93c82bbea20bef627dbe8e34a4bc8e526217104 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:50 +0100
Subject: x86: shrink __PAGE_KERNEL/__PAGE_KERNEL_EXEC on non PAE kernels

No need to make it 64bit there.

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_32.c     | 4 ++--
 include/asm-x86/pgtable.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index e00c1d7128b1..08ba3a14c72c 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -332,9 +332,9 @@ static void __init set_highmem_pages_init(int bad_ppro)
 #define set_highmem_pages_init(bad_ppro) do { } while (0)
 #endif /* CONFIG_HIGHMEM */
 
-unsigned long long __PAGE_KERNEL = _PAGE_KERNEL;
+pteval_t __PAGE_KERNEL = _PAGE_KERNEL;
 EXPORT_SYMBOL(__PAGE_KERNEL);
-unsigned long long __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
+pteval_t __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
 
 #ifdef CONFIG_NUMA
 extern void __init remap_numa_kva(void);
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 18a5f9e32834..0c3ecca0f79c 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -69,7 +69,7 @@
 #define _PAGE_KERNEL (_PAGE_KERNEL_EXEC | _PAGE_NX)
 
 #ifndef __ASSEMBLY__
-extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
+extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #endif	/* __ASSEMBLY__ */
 #else
 #define __PAGE_KERNEL_EXEC						\
-- 
cgit v1.2.3


From c6ca18eb759c43ae20d953ef90c3de4557f9342a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:51 +0100
Subject: x86: add pte_pgprot to 32-bit

64bit already had it.

Needed for later patches.

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 2 ++
 include/asm-x86/pgtable_64.h | 2 --
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 0c3ecca0f79c..b4ee5939afe5 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -187,6 +187,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return __pte(val);
 }
 
+#define pte_pgprot(x) __pgprot(pte_val(x) & (0xfff | _PAGE_NX))
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else  /* !CONFIG_PARAVIRT */
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 139da50cd510..df1fcbd12e43 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -120,8 +120,6 @@ static inline void native_pgd_clear(pgd_t * pgd)
 
 #define pte_same(a, b)		((a).pte == (b).pte)
 
-#define pte_pgprot(a)	(__pgprot((a).pte & ~PHYSICAL_PAGE_MASK))
-
 #endif /* !__ASSEMBLY__ */
 
 #define PMD_SIZE	(_AC(1,UL) << PMD_SHIFT)
-- 
cgit v1.2.3


From aaa0e89023656c078c67cbd0460fc2ccbf3323c6 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:51 +0100
Subject: x86: don't drop NX bit in pte modifier functions on 32-bit

The pte_* modifier functions that cleared bits dropped the NX bit on 32bit
PAE because they only worked in int, but NX is in bit 63. Fix that
by adding appropiate casts so that the arithmetic happens as long long
on PAE kernels.

I decided to just use 64bit arithmetic instead of open coding like
pte_modify() because gcc should generate good enough code for that now.

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index b4ee5939afe5..2f93723a3e18 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -147,17 +147,17 @@ static inline int pmd_large(pmd_t pte) {
 		(_PAGE_PSE|_PAGE_PRESENT);
 }
 
-static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_DIRTY); }
-static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_ACCESSED); }
-static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_RW); }
-static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_NX); }
+static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_DIRTY); }
+static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_ACCESSED); }
+static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_RW); }
+static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_NX); }
 static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
 static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
 static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
 static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
-static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_PSE); }
+static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_PSE); }
 static inline pte_t pte_mkglobal(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_GLOBAL); }
-static inline pte_t pte_clrglobal(pte_t pte)	{ return __pte(pte_val(pte) & ~_PAGE_GLOBAL); }
+static inline pte_t pte_clrglobal(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_GLOBAL); }
 
 extern pteval_t __supported_pte_mask;
 
-- 
cgit v1.2.3


From 1e8e23bc469a11fc190a6bace857075d83dc2555 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:53 +0100
Subject: x86: introduce canon_pgprot()

Introduce canon_pgprot()

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 2f93723a3e18..2726ddc457fd 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -189,6 +189,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #define pte_pgprot(x) __pgprot(pte_val(x) & (0xfff | _PAGE_NX))
 
+#define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask)
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else  /* !CONFIG_PARAVIRT */
-- 
cgit v1.2.3


From 934d15854d04e8ca2495d8f5698164df990d5d66 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 30 Jan 2008 13:33:53 +0100
Subject: x86: remove set_kernel_exec()

The SMP trampoline always runs in real mode, so making it executable
in the page tables doesn't make much sense because it executes
before page tables are set up. That was the only user of
set_kernel_exec(). Remove set_kernel_exec().

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/smpboot_32.c | 11 -----------
 arch/x86/mm/init_32.c        | 29 -----------------------------
 include/asm-x86/pgtable_32.h | 12 ------------
 3 files changed, 52 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 7a62dced61ca..5787a0c3e296 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -107,7 +107,6 @@ u8 apicid_2_node[MAX_APICID];
 extern const unsigned char trampoline_data [];
 extern const unsigned char trampoline_end  [];
 static unsigned char *trampoline_base;
-static int trampoline_exec;
 
 static void map_cpu_to_logical_apicid(void);
 
@@ -139,10 +138,6 @@ void __init smp_alloc_memory(void)
 	 */
 	if (__pa(trampoline_base) >= 0x9F000)
 		BUG();
-	/*
-	 * Make the SMP trampoline executable:
-	 */
-	trampoline_exec = set_kernel_exec((unsigned long)trampoline_base, 1);
 }
 
 /*
@@ -1290,12 +1285,6 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
 	setup_ioapic_dest();
 #endif
 	zap_low_mappings();
-#ifndef CONFIG_HOTPLUG_CPU
-	/*
-	 * Disable executability of the SMP trampoline:
-	 */
-	set_kernel_exec((unsigned long)trampoline_base, trampoline_exec);
-#endif
 }
 
 void __init smp_intr_init(void)
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 08ba3a14c72c..016c8ccd1d8d 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -529,35 +529,6 @@ static void __init set_nx(void)
 	}
 }
 
-/*
- * Enables/disables executability of a given kernel page and
- * returns the previous setting.
- */
-int __init set_kernel_exec(unsigned long vaddr, int enable)
-{
-	pte_t *pte;
-	int ret = 1;
-	int level;
-
-	if (!nx_enabled)
-		goto out;
-
-	pte = lookup_address(vaddr, &level);
-	BUG_ON(!pte);
-
-	if (!pte_exec(*pte))
-		ret = 0;
-
-	if (enable)
-		pte->pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
-	else
-		pte->pte_high |= 1 << (_PAGE_BIT_NX - 32);
-	pte_update_defer(&init_mm, vaddr, pte);
-	__flush_tlb_all();
-out:
-	return ret;
-}
-
 #endif
 
 /*
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 9381bd37b9b1..72eb06c2bac9 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -184,18 +184,6 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
  */
 extern pte_t *lookup_address(unsigned long address, int *level);
 
-/*
- * Make a given kernel text page executable/non-executable.
- * Returns the previous executability setting of that page (which
- * is used to restore the previous state). Used by the SMP bootup code.
- * NOTE: this is an __init function for security reasons.
- */
-#ifdef CONFIG_X86_PAE
- extern int set_kernel_exec(unsigned long vaddr, int enable);
-#else
- static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
-#endif
-
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
 	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + pte_index(address))
-- 
cgit v1.2.3


From a2172e2586f6662af996e47f417bb718c37cf8d2 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Wed, 30 Jan 2008 13:33:55 +0100
Subject: x86: fix some bugs about EFI runtime code mapping

This patch fixes some bugs of making EFI runtime code executable.

- Use change_page_attr in i386 too. Because the runtime code may be
  mapped not through ioremap.

- If there is no _PAGE_NX in __supported_pte_mask, the change_page_attr
  is not called.

- Make efi_ioremap map pages as PAGE_KERNEL_EXEC_NOCACHE, because EFI runtime
  code may be mapped through efi_ioremap.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/efi.c    | 35 ++++++++++++++++++++++++++++++-----
 arch/x86/kernel/efi_64.c | 26 ++++++--------------------
 include/asm-x86/efi.h    |  1 -
 3 files changed, 36 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 5d492f99d967..834ecfb41e97 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -40,6 +40,8 @@
 #include <asm/setup.h>
 #include <asm/efi.h>
 #include <asm/time.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
 
 #define EFI_DEBUG	1
 #define PFX 		"EFI: "
@@ -379,6 +381,32 @@ void __init efi_init(void)
 #endif
 }
 
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+static void __init runtime_code_page_mkexec(void)
+{
+	efi_memory_desc_t *md;
+	unsigned long end;
+	void *p;
+
+	if (!(__supported_pte_mask & _PAGE_NX))
+		return;
+
+	/* Make EFI runtime service code area executable */
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
+		if (md->type == EFI_RUNTIME_SERVICES_CODE &&
+		    (end >> PAGE_SHIFT) <= end_pfn_map)
+			change_page_attr_addr(md->virt_addr,
+					      md->num_pages,
+					      PAGE_KERNEL_EXEC_NOCACHE);
+	}
+	__flush_tlb_all();
+}
+#else
+static inline void __init runtime_code_page_mkexec(void) { }
+#endif
+
 /*
  * This function will switch the EFI runtime services to virtual mode.
  * Essentially, look through the EFI memmap and map every region that
@@ -399,9 +427,9 @@ void __init efi_enter_virtual_mode(void)
 		md = p;
 		if (!(md->attribute & EFI_MEMORY_RUNTIME))
 			continue;
+		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
 		if ((md->attribute & EFI_MEMORY_WB) &&
-		    (((md->phys_addr + (md->num_pages<<EFI_PAGE_SHIFT)) >>
-		      PAGE_SHIFT) < end_pfn_map))
+		    ((end >> PAGE_SHIFT) <= end_pfn_map))
 			md->virt_addr = (unsigned long)__va(md->phys_addr);
 		else
 			md->virt_addr = (unsigned long)
@@ -410,7 +438,6 @@ void __init efi_enter_virtual_mode(void)
 		if (!md->virt_addr)
 			printk(KERN_ERR PFX "ioremap of 0x%llX failed!\n",
 			       (unsigned long long)md->phys_addr);
-		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
 		if ((md->phys_addr <= (unsigned long)efi_phys.systab) &&
 		    ((unsigned long)efi_phys.systab < end))
 			efi.systab = (efi_system_table_t *)(unsigned long)
@@ -448,9 +475,7 @@ void __init efi_enter_virtual_mode(void)
 	efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count;
 	efi.reset_system = virt_efi_reset_system;
 	efi.set_virtual_address_map = virt_efi_set_virtual_address_map;
-#ifdef CONFIG_X86_64
 	runtime_code_page_mkexec();
-#endif
 }
 
 /*
diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c
index 1f8bbd9644d7..9f8a75594398 100644
--- a/arch/x86/kernel/efi_64.c
+++ b/arch/x86/kernel/efi_64.c
@@ -33,7 +33,6 @@
 #include <asm/e820.h>
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
-#include <asm/cacheflush.h>
 #include <asm/proto.h>
 #include <asm/efi.h>
 
@@ -55,7 +54,7 @@ static void __init early_mapping_set_exec(unsigned long start,
 		else
 			set_pte(kpte, __pte((pte_val(*kpte) | _PAGE_NX) & \
 					    __supported_pte_mask));
-		if (pte_huge(*kpte))
+		if (level == 4)
 			start = (start + PMD_SIZE) & PMD_MASK;
 		else
 			start = (start + PAGE_SIZE) & PAGE_MASK;
@@ -67,6 +66,9 @@ static void __init early_runtime_code_mapping_set_exec(int executable)
 	efi_memory_desc_t *md;
 	void *p;
 
+	if (!(__supported_pte_mask & _PAGE_NX))
+		return;
+
 	/* Make EFI runtime service code area executable */
 	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
 		md = p;
@@ -116,22 +118,6 @@ void __init efi_reserve_bootmem(void)
 				memmap.nr_map * memmap.desc_size);
 }
 
-void __init runtime_code_page_mkexec(void)
-{
-	efi_memory_desc_t *md;
-	void *p;
-
-	/* Make EFI runtime service code area executable */
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-		if (md->type == EFI_RUNTIME_SERVICES_CODE)
-			change_page_attr_addr(md->virt_addr,
-					      md->num_pages,
-					      PAGE_KERNEL_EXEC);
-	}
-	__flush_tlb_all();
-}
-
 void __iomem * __init efi_ioremap(unsigned long offset,
 				  unsigned long size)
 {
@@ -146,8 +132,8 @@ void __iomem * __init efi_ioremap(unsigned long offset,
 		return NULL;
 
 	for (i = 0; i < pages; i++) {
-		set_fixmap_nocache(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped,
-				   offset);
+		__set_fixmap(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped,
+			     offset, PAGE_KERNEL_EXEC_NOCACHE);
 		offset += PAGE_SIZE;
 		pages_mapped++;
 	}
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index 8380131ca542..9e08ae84795d 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -95,6 +95,5 @@ extern void *efi_ioremap(unsigned long offset, unsigned long size);
 extern void efi_reserve_bootmem(void);
 extern void efi_call_phys_prelog(void);
 extern void efi_call_phys_epilog(void);
-extern void runtime_code_page_mkexec(void);
 
 #endif
-- 
cgit v1.2.3


From 12d6f21eacc21d84a809829543f2fe45c7e37319 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:33:58 +0100
Subject: x86: do not PSE on CONFIG_DEBUG_PAGEALLOC=y

get more testing of the c_p_a() code done by not turning off
PSE on DEBUG_PAGEALLOC.

this simplifies the early pagetable setup code, and tests
the largepage-splitup code quite heavily.

In the end, all the largepages will be split up pretty quickly,
so there's no difference to how DEBUG_PAGEALLOC worked before.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/common.c |  7 -------
 arch/x86/mm/pageattr_32.c    | 12 +++++++++++-
 include/asm-x86/cacheflush.h |  5 -----
 include/linux/mm.h           | 14 +++++++++++++-
 init/main.c                  |  5 +++++
 5 files changed, 29 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index bba850b05d0e..db28aa9e2f69 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -641,13 +641,6 @@ void __init early_cpu_init(void)
 	nexgen_init_cpu();
 	umc_init_cpu();
 	early_cpu_detect();
-
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	/* pse is not compatible with on-the-fly unmapping,
-	 * disable it even if the cpus claim to support it.
-	 */
-	setup_clear_cpu_cap(X86_FEATURE_PSE);
-#endif
 }
 
 /* Make sure %fs is initialized properly in idle threads */
diff --git a/arch/x86/mm/pageattr_32.c b/arch/x86/mm/pageattr_32.c
index 9cf2fea54eb5..dd49b16b3a0e 100644
--- a/arch/x86/mm/pageattr_32.c
+++ b/arch/x86/mm/pageattr_32.c
@@ -61,13 +61,17 @@ static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte)
 static int split_large_page(pte_t *kpte, unsigned long address)
 {
 	pgprot_t ref_prot = pte_pgprot(pte_clrhuge(*kpte));
+	gfp_t gfp_flags = GFP_KERNEL;
 	unsigned long flags;
 	unsigned long addr;
 	pte_t *pbase, *tmp;
 	struct page *base;
 	int i, level;
 
-	base = alloc_pages(GFP_KERNEL, 0);
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	gfp_flags = GFP_ATOMIC;
+#endif
+	base = alloc_pages(gfp_flags, 0);
 	if (!base)
 		return -ENOMEM;
 
@@ -218,6 +222,12 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 					   numpages * PAGE_SIZE);
 	}
 
+	/*
+	 * If page allocator is not up yet then do not call c_p_a():
+	 */
+	if (!debug_pagealloc_enabled)
+		return;
+
 	/*
 	 * the return value is ignored - the calls cannot fail,
 	 * large pages are disabled at boot time.
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 9411a2d3f19c..fccb563e2305 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -29,11 +29,6 @@ int change_page_attr(struct page *page, int numpages, pgprot_t prot);
 int change_page_attr_addr(unsigned long addr, int numpages, pgprot_t prot);
 void clflush_cache_range(void *addr, int size);
 
-#ifdef CONFIG_DEBUG_PAGEALLOC
-/* internal debugging function */
-void kernel_map_pages(struct page *page, int numpages, int enable);
-#endif
-
 #ifdef CONFIG_DEBUG_RODATA
 void mark_rodata_ro(void);
 #endif
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 3c22d971afa7..1bba6789a50a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1118,9 +1118,21 @@ static inline void vm_stat_account(struct mm_struct *mm,
 }
 #endif /* CONFIG_PROC_FS */
 
-#ifndef CONFIG_DEBUG_PAGEALLOC
+#ifdef CONFIG_DEBUG_PAGEALLOC
+extern int debug_pagealloc_enabled;
+
+extern void kernel_map_pages(struct page *page, int numpages, int enable);
+
+static inline void enable_debug_pagealloc(void)
+{
+	debug_pagealloc_enabled = 1;
+}
+#else
 static inline void
 kernel_map_pages(struct page *page, int numpages, int enable) {}
+static inline void enable_debug_pagealloc(void)
+{
+}
 #endif
 
 extern struct vm_area_struct *get_gate_vma(struct task_struct *tsk);
diff --git a/init/main.c b/init/main.c
index 3316dffe3e57..cb81ed116f62 100644
--- a/init/main.c
+++ b/init/main.c
@@ -318,6 +318,10 @@ static int __init unknown_bootoption(char *param, char *val)
 	return 0;
 }
 
+#ifdef CONFIG_DEBUG_PAGEALLOC
+int __read_mostly debug_pagealloc_enabled = 0;
+#endif
+
 static int __init init_setup(char *str)
 {
 	unsigned int i;
@@ -552,6 +556,7 @@ asmlinkage void __init start_kernel(void)
 	preempt_disable();
 	build_all_zonelists();
 	page_alloc_init();
+	enable_debug_pagealloc();
 	printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line);
 	parse_early_param();
 	parse_args("Booting kernel", static_command_line, __start___param,
-- 
cgit v1.2.3


From b65e6390fb99ca93877ffc1c84e93c857f32dafc Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:01 +0100
Subject: x86: introduce native_set_pte_atomic() on 64-bit too

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable_64.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index df1fcbd12e43..223b0032bab5 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -76,6 +76,11 @@ static inline void native_set_pte(pte_t *ptep, pte_t pte)
 	*ptep = pte;
 }
 
+static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+	native_set_pte(ptep, pte);
+}
+
 static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 {
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3


From e4b71dcf54fa90fc30fb901bbce7e38a46467af7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:04 +0100
Subject: x86: clean up arch/x86/mm/pageattr.c

do some leftover cleanups in the now unified arch/x86/mm/pageattr.c
file.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/pageattr.c       | 13 +++++--------
 include/asm-x86/pgtable_64.h |  2 ++
 2 files changed, 7 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 510ff4091667..a270f9ccebfb 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -2,7 +2,6 @@
  * Copyright 2002 Andi Kleen, SuSE Labs.
  * Thanks to Ben LaHaise for precious feedback.
  */
-
 #include <linux/highmem.h>
 #include <linux/module.h>
 #include <linux/sched.h>
@@ -50,9 +49,7 @@ static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte)
 	/* change init_mm */
 	set_pte_atomic(kpte, pte);
 #ifdef CONFIG_X86_32
-	if (SHARED_KERNEL_PMD)
-		return;
-	{
+	if (!SHARED_KERNEL_PMD) {
 		struct page *page;
 
 		for (page = pgd_list; page; page = (struct page *)page->index) {
@@ -277,14 +274,14 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 		return;
 
 	/*
-	 * the return value is ignored - the calls cannot fail,
-	 * large pages are disabled at boot time.
+	 * The return value is ignored - the calls cannot fail,
+	 * large pages are disabled at boot time:
 	 */
 	change_page_attr(page, numpages, enable ? PAGE_KERNEL : __pgprot(0));
 
 	/*
-	 * we should perform an IPI and flush all tlbs,
-	 * but that can deadlock->flush only current cpu.
+	 * We should perform an IPI and flush all tlbs,
+	 * but that can deadlock->flush only current cpu:
 	 */
 	__flush_tlb_all();
 }
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 223b0032bab5..faa6a96c2a5c 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -25,6 +25,8 @@ extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
 #endif /* !__ASSEMBLY__ */
 
+#define SHARED_KERNEL_PMD	1
+
 /*
  * PGDIR_SHIFT determines what a top-level page table entry can map
  */
-- 
cgit v1.2.3


From 0a663088cd6ff9e89f285ae7689e6eee46cfb54c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:04 +0100
Subject: x86: clean up lookup_address() declarations

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h    | 8 ++++++++
 include/asm-x86/pgtable_32.h | 8 --------
 include/asm-x86/pgtable_64.h | 2 --
 3 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 2726ddc457fd..4409dabe31c6 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -234,6 +234,14 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #ifndef __ASSEMBLY__
 
+/*
+ * Helper function that returns the kernel pagetable entry controlling
+ * the virtual address 'address'. NULL means no pagetable entry present.
+ * NOTE: the return type is pte_t but if the pmd is PSE then we return it
+ * as a pte too.
+ */
+extern pte_t *lookup_address(unsigned long address, int *level);
+
 /* local pte updates need not use xchg for locking */
 static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
 {
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 72eb06c2bac9..ca7b150ca8b7 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -176,14 +176,6 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
 #define pmd_page_vaddr(pmd) \
 		((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
-/*
- * Helper function that returns the kernel pagetable entry controlling
- * the virtual address 'address'. NULL means no pagetable entry present.
- * NOTE: the return type is pte_t but if the pmd is PSE then we return it
- * as a pte too.
- */
-extern pte_t *lookup_address(unsigned long address, int *level);
-
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
 	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + pte_index(address))
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index faa6a96c2a5c..ce1dbab75e39 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -245,8 +245,6 @@ extern struct list_head pgd_list;
 
 extern int kern_addr_valid(unsigned long addr); 
 
-pte_t *lookup_address(unsigned long addr, int *level);
-
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
-- 
cgit v1.2.3


From 30551bb3ce9257a2352b3cb4e45010d415cc0ad5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:04 +0100
Subject: x86: add PG_LEVEL enum

this way PG_LEVEL_1GB will be an easy change.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/pageattr-test.c | 4 +---
 arch/x86/mm/pageattr.c      | 9 ++++++---
 include/asm-x86/pgtable.h   | 6 ++++++
 3 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr-test.c b/arch/x86/mm/pageattr-test.c
index 6cc106b388a5..d7a93008cc12 100644
--- a/arch/x86/mm/pageattr-test.c
+++ b/arch/x86/mm/pageattr-test.c
@@ -16,14 +16,12 @@
 
 enum {
 	NTEST			= 400,
+	LOWEST_LEVEL		= PG_LEVEL_4K,
 #ifdef CONFIG_X86_64
-	LOWEST_LEVEL		= 4,
 	LPS			= (1 << PMD_SHIFT),
 #elif defined(CONFIG_X86_PAE)
-	LOWEST_LEVEL		= 4,
 	LPS			= (1 << PMD_SHIFT),
 #else
-	LOWEST_LEVEL		= 4, /* lookup_address lies here */
 	LPS			= (1 << 22),
 #endif
 	GPS			= (1<<30)
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index a270f9ccebfb..4589a1382fa1 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -28,6 +28,8 @@ pte_t *lookup_address(unsigned long address, int *level)
 	pud_t *pud;
 	pmd_t *pmd;
 
+	*level = PG_LEVEL_NONE;
+
 	if (pgd_none(*pgd))
 		return NULL;
 	pud = pud_offset(pgd, address);
@@ -36,11 +38,12 @@ pte_t *lookup_address(unsigned long address, int *level)
 	pmd = pmd_offset(pud, address);
 	if (pmd_none(*pmd))
 		return NULL;
-	*level = 3;
+
+	*level = PG_LEVEL_2M;
 	if (pmd_large(*pmd))
 		return (pte_t *)pmd;
-	*level = 4;
 
+	*level = PG_LEVEL_4K;
 	return pte_offset_kernel(pmd, address);
 }
 
@@ -145,7 +148,7 @@ repeat:
 		address < (unsigned long)&_etext &&
 	       (pgprot_val(prot) & _PAGE_NX));
 
-	if (level == 4) {
+	if (level == PG_LEVEL_4K) {
 		set_pte_atomic(kpte, mk_pte(page, canon_pgprot(prot)));
 	} else {
 		err = split_large_page(kpte, address);
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 4409dabe31c6..7aa34c8eb220 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -234,6 +234,12 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #ifndef __ASSEMBLY__
 
+enum {
+	PG_LEVEL_NONE,
+	PG_LEVEL_4K,
+	PG_LEVEL_2M,
+};
+
 /*
  * Helper function that returns the kernel pagetable entry controlling
  * the virtual address 'address'. NULL means no pagetable entry present.
-- 
cgit v1.2.3


From d2e626f45cc450c00f5f98a89b8b4c4ac3c9bf5f Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:04 +0100
Subject: x86: add PAGE_KERNEL_EXEC_NOCACHE

add PAGE_KERNEL_EXEC_NOCACHE.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgtable.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 7aa34c8eb220..ee40a88882f6 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -79,6 +79,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 
 #define __PAGE_KERNEL_RO		(__PAGE_KERNEL & ~_PAGE_RW)
 #define __PAGE_KERNEL_RX		(__PAGE_KERNEL_EXEC & ~_PAGE_RW)
+#define __PAGE_KERNEL_EXEC_NOCACHE	(__PAGE_KERNEL_EXEC | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_VSYSCALL		(__PAGE_KERNEL_RX | _PAGE_USER)
 #define __PAGE_KERNEL_VSYSCALL_NOCACHE	(__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT)
@@ -96,6 +97,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define PAGE_KERNEL_EXEC		MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
 #define PAGE_KERNEL_RX			MAKE_GLOBAL(__PAGE_KERNEL_RX)
 #define PAGE_KERNEL_NOCACHE		MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
+#define PAGE_KERNEL_EXEC_NOCACHE	MAKE_GLOBAL(__PAGE_KERNEL_EXEC_NOCACHE)
 #define PAGE_KERNEL_LARGE		MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
 #define PAGE_KERNEL_LARGE_EXEC		MAKE_GLOBAL(__PAGE_KERNEL_LARGE_EXEC)
 #define PAGE_KERNEL_VSYSCALL		MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
-- 
cgit v1.2.3


From f87519e8f4f1de9b39a40e56479a7ad2443169dd Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:05 +0100
Subject: x86: introduce max_pfn_mapped

64bit uses end_pfn_map and 32bit uses max_low_pfn. There are several
files which have #ifdef'ed defines which map either to end_pfn_map or
max_low_pfn. Replace this by a universal define and clean up all the
other instances.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/efi.c       |  4 ++--
 arch/x86/mm/pageattr-test.c | 21 +++++++--------------
 include/asm-x86/efi.h       |  2 --
 include/asm-x86/page.h      |  2 ++
 4 files changed, 11 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 834ecfb41e97..57b57778bf60 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -396,7 +396,7 @@ static void __init runtime_code_page_mkexec(void)
 		md = p;
 		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
 		if (md->type == EFI_RUNTIME_SERVICES_CODE &&
-		    (end >> PAGE_SHIFT) <= end_pfn_map)
+		    (end >> PAGE_SHIFT) <= max_pfn_mapped)
 			change_page_attr_addr(md->virt_addr,
 					      md->num_pages,
 					      PAGE_KERNEL_EXEC_NOCACHE);
@@ -429,7 +429,7 @@ void __init efi_enter_virtual_mode(void)
 			continue;
 		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
 		if ((md->attribute & EFI_MEMORY_WB) &&
-		    ((end >> PAGE_SHIFT) <= end_pfn_map))
+		    ((end >> PAGE_SHIFT) <= max_pfn_mapped))
 			md->virt_addr = (unsigned long)__va(md->phys_addr);
 		else
 			md->virt_addr = (unsigned long)
diff --git a/arch/x86/mm/pageattr-test.c b/arch/x86/mm/pageattr-test.c
index d7a93008cc12..6a41a0f0c149 100644
--- a/arch/x86/mm/pageattr-test.c
+++ b/arch/x86/mm/pageattr-test.c
@@ -27,13 +27,6 @@ enum {
 	GPS			= (1<<30)
 };
 
-#ifdef CONFIG_X86_64
-# include <asm/proto.h>
-# define max_mapped		end_pfn_map
-#else
-# define max_mapped		max_low_pfn
-#endif
-
 struct split_state {
 	long lpg, gpg, spg, exec;
 	long min_exec, max_exec;
@@ -48,7 +41,7 @@ static __init int print_split(struct split_state *s)
 	s->lpg = s->gpg = s->spg = s->exec = 0;
 	s->min_exec = ~0UL;
 	s->max_exec = 0;
-	for (i = 0; i < max_mapped; ) {
+	for (i = 0; i < max_pfn_mapped; ) {
 		unsigned long addr = (unsigned long)__va(i << PAGE_SHIFT);
 		int level;
 		pte_t *pte;
@@ -97,8 +90,8 @@ static __init int print_split(struct split_state *s)
 
 	expected = (s->gpg*GPS + s->lpg*LPS)/PAGE_SIZE + s->spg + missed;
 	if (expected != i) {
-		printk(KERN_ERR "CPA max_mapped %lu but expected %lu\n",
-			max_mapped, expected);
+		printk(KERN_ERR "CPA max_pfn_mapped %lu but expected %lu\n",
+			max_pfn_mapped, expected);
 		return 1;
 	}
 	return err;
@@ -120,22 +113,22 @@ static __init int exercise_pageattr(void)
 
 	printk(KERN_INFO "CPA exercising pageattr\n");
 
-	bm = vmalloc((max_mapped + 7) / 8);
+	bm = vmalloc((max_pfn_mapped + 7) / 8);
 	if (!bm) {
 		printk(KERN_ERR "CPA Cannot vmalloc bitmap\n");
 		return -ENOMEM;
 	}
-	memset(bm, 0, (max_mapped + 7) / 8);
+	memset(bm, 0, (max_pfn_mapped + 7) / 8);
 
 	failed += print_split(&sa);
 	srandom32(100);
 
 	for (i = 0; i < NTEST; i++) {
-		unsigned long pfn = random32() % max_mapped;
+		unsigned long pfn = random32() % max_pfn_mapped;
 
 		addr[i] = (unsigned long)__va(pfn << PAGE_SHIFT);
 		len[i] = random32() % 100;
-		len[i] = min_t(unsigned long, len[i], max_mapped - pfn - 1);
+		len[i] = min_t(unsigned long, len[i], max_pfn_mapped - pfn - 1);
 
 		if (len[i] == 0)
 			len[i] = 1;
diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index 9e08ae84795d..9c68a1f098d8 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -35,8 +35,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...);
 
 #define efi_ioremap(addr, size)			ioremap(addr, size)
 
-#define end_pfn_map				max_low_pfn
-
 #else /* !CONFIG_X86_32 */
 
 #define MAX_EFI_IO_PAGES	100
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index eba88d940092..e2c79d8dcdcf 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -33,8 +33,10 @@
 
 #ifdef CONFIG_X86_64
 #include <asm/page_64.h>
+#define max_pfn_mapped		end_pfn_map
 #else
 #include <asm/page_32.h>
+#define max_pfn_mapped		max_low_pfn
 #endif	/* CONFIG_X86_64 */
 
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
-- 
cgit v1.2.3


From 75cbade8ea3127a84e3da7c2c15808e54f0df7e8 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: a new API for drivers/etc to control cache and other page
 attributes

Right now, if drivers or other code want to change, say, a cache attribute of a
page, the only API they have is change_page_attr(). c-p-a is a really bad API
for this, because it forces the caller to know *ALL* the attributes he wants
for the page, not just the 1 thing he wants to change. So code that wants to
set a page uncachable, needs to be aware of the NX status as well etc etc etc.

This patch introduces a set of new APIs for this, set_pages_<attr> and
set_memory_<attr>, that offer a logical change to the user, and leave all
attributes not implied by the requested logical change alone.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pageattr.c       | 197 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/cacheflush.h |  15 ++++
 2 files changed, 212 insertions(+)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 4208571334db..f3c9510b8d92 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -211,6 +211,8 @@ repeat:
  * mem_map entry (pfn_valid() is false).
  *
  * See change_page_attr() documentation for more details.
+ *
+ * Modules and drivers should use the set_memory_* APIs instead.
  */
 
 int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
@@ -273,6 +275,8 @@ int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
  * (e.g. in user space) * This function only deals with the kernel linear map.
  *
  * For MMIO areas without mem_map use change_page_attr_addr() instead.
+ *
+ * Modules and drivers should use the set_pages_* APIs instead.
  */
 int change_page_attr(struct page *page, int numpages, pgprot_t prot)
 {
@@ -282,6 +286,199 @@ int change_page_attr(struct page *page, int numpages, pgprot_t prot)
 }
 EXPORT_SYMBOL(change_page_attr);
 
+/**
+ * change_page_attr_set - Change page table attributes in the linear mapping.
+ * @addr: Virtual address in linear mapping.
+ * @numpages: Number of pages to change
+ * @prot: Protection/caching type bits to set (PAGE_*)
+ *
+ * Returns 0 on success, otherwise a negated errno.
+ *
+ * This should be used when a page is mapped with a different caching policy
+ * than write-back somewhere - some CPUs do not like it when mappings with
+ * different caching policies exist. This changes the page attributes of the
+ * in kernel linear mapping too.
+ *
+ * Caller must call global_flush_tlb() later to make the changes active.
+ *
+ * The caller needs to ensure that there are no conflicting mappings elsewhere
+ * (e.g. in user space) * This function only deals with the kernel linear map.
+ *
+ * This function is different from change_page_attr() in that only selected bits
+ * are impacted, all other bits remain as is.
+ */
+int change_page_attr_set(unsigned long addr, int numpages, pgprot_t prot)
+{
+	pgprot_t current_prot;
+	int level;
+	pte_t *pte;
+
+	pte = lookup_address(addr, &level);
+	if (pte)
+		current_prot = pte_pgprot(*pte);
+	else
+		pgprot_val(current_prot) = 0;
+
+	pgprot_val(prot) = pgprot_val(current_prot) | pgprot_val(prot);
+
+	return change_page_attr_addr(addr, numpages, prot);
+}
+
+/**
+ * change_page_attr_clear - Change page table attributes in the linear mapping.
+ * @addr: Virtual address in linear mapping.
+ * @numpages: Number of pages to change
+ * @prot: Protection/caching type bits to clear (PAGE_*)
+ *
+ * Returns 0 on success, otherwise a negated errno.
+ *
+ * This should be used when a page is mapped with a different caching policy
+ * than write-back somewhere - some CPUs do not like it when mappings with
+ * different caching policies exist. This changes the page attributes of the
+ * in kernel linear mapping too.
+ *
+ * Caller must call global_flush_tlb() later to make the changes active.
+ *
+ * The caller needs to ensure that there are no conflicting mappings elsewhere
+ * (e.g. in user space) * This function only deals with the kernel linear map.
+ *
+ * This function is different from change_page_attr() in that only selected bits
+ * are impacted, all other bits remain as is.
+ */
+int change_page_attr_clear(unsigned long addr, int numpages, pgprot_t prot)
+{
+	pgprot_t current_prot;
+	int level;
+	pte_t *pte;
+
+	pte = lookup_address(addr, &level);
+	if (pte)
+		current_prot = pte_pgprot(*pte);
+	else
+		pgprot_val(current_prot) = 0;
+
+	pgprot_val(prot) = pgprot_val(current_prot) & ~pgprot_val(prot);
+
+	return change_page_attr_addr(addr, numpages, prot);
+}
+
+
+
+int set_memory_uc(unsigned long addr, int numpages)
+{
+	pgprot_t uncached;
+
+	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
+	return change_page_attr_set(addr, numpages, uncached);
+}
+EXPORT_SYMBOL(set_memory_uc);
+
+int set_memory_wb(unsigned long addr, int numpages)
+{
+	pgprot_t uncached;
+
+	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
+	return change_page_attr_clear(addr, numpages, uncached);
+}
+EXPORT_SYMBOL(set_memory_wb);
+
+int set_memory_x(unsigned long addr, int numpages)
+{
+	pgprot_t nx;
+
+	pgprot_val(nx) = _PAGE_NX;
+	return change_page_attr_clear(addr, numpages, nx);
+}
+EXPORT_SYMBOL(set_memory_x);
+
+int set_memory_nx(unsigned long addr, int numpages)
+{
+	pgprot_t nx;
+
+	pgprot_val(nx) = _PAGE_NX;
+	return change_page_attr_set(addr, numpages, nx);
+}
+EXPORT_SYMBOL(set_memory_nx);
+
+int set_memory_ro(unsigned long addr, int numpages)
+{
+	pgprot_t rw;
+
+	pgprot_val(rw) = _PAGE_RW;
+	return change_page_attr_clear(addr, numpages, rw);
+}
+EXPORT_SYMBOL(set_memory_ro);
+
+int set_memory_rw(unsigned long addr, int numpages)
+{
+	pgprot_t rw;
+
+	pgprot_val(rw) = _PAGE_RW;
+	return change_page_attr_set(addr, numpages, rw);
+}
+EXPORT_SYMBOL(set_memory_rw);
+
+int set_pages_uc(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t uncached;
+
+	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
+	return change_page_attr_set(addr, numpages, uncached);
+}
+EXPORT_SYMBOL(set_pages_uc);
+
+int set_pages_wb(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t uncached;
+
+	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
+	return change_page_attr_clear(addr, numpages, uncached);
+}
+EXPORT_SYMBOL(set_pages_wb);
+
+int set_pages_x(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t nx;
+
+	pgprot_val(nx) = _PAGE_NX;
+	return change_page_attr_clear(addr, numpages, nx);
+}
+EXPORT_SYMBOL(set_pages_x);
+
+int set_pages_nx(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t nx;
+
+	pgprot_val(nx) = _PAGE_NX;
+	return change_page_attr_set(addr, numpages, nx);
+}
+EXPORT_SYMBOL(set_pages_nx);
+
+int set_pages_ro(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t rw;
+
+	pgprot_val(rw) = _PAGE_RW;
+	return change_page_attr_clear(addr, numpages, rw);
+}
+EXPORT_SYMBOL(set_pages_ro);
+
+int set_pages_rw(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	pgprot_t rw;
+
+	pgprot_val(rw) = _PAGE_RW;
+	return change_page_attr_set(addr, numpages, rw);
+}
+EXPORT_SYMBOL(set_pages_rw);
+
+
 void clflush_cache_range(void *addr, int size)
 {
 	int i;
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index fccb563e2305..7b928d55c190 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -27,6 +27,21 @@
 void global_flush_tlb(void);
 int change_page_attr(struct page *page, int numpages, pgprot_t prot);
 int change_page_attr_addr(unsigned long addr, int numpages, pgprot_t prot);
+
+int set_pages_uc(struct page *page, int numpages);
+int set_pages_wb(struct page *page, int numpages);
+int set_pages_x(struct page *page, int numpages);
+int set_pages_nx(struct page *page, int numpages);
+int set_pages_ro(struct page *page, int numpages);
+int set_pages_rw(struct page *page, int numpages);
+
+int set_memory_uc(unsigned long addr, int numpages);
+int set_memory_wb(unsigned long addr, int numpages);
+int set_memory_x(unsigned long addr, int numpages);
+int set_memory_nx(unsigned long addr, int numpages);
+int set_memory_ro(unsigned long addr, int numpages);
+int set_memory_rw(unsigned long addr, int numpages);
+
 void clflush_cache_range(void *addr, int size);
 
 #ifdef CONFIG_DEBUG_RODATA
-- 
cgit v1.2.3


From 6d238cc4dc8a36a3915c26202fe49f58a0683fb9 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: convert CPA users to the new set_page_ API

This patch converts various users of change_page_attr() to the new,
more intent driven set_page_*/set_memory_* API set.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/efi.c               |  8 ++++----
 arch/x86/kernel/pci-gart_64.c       |  3 +--
 arch/x86/mm/init_32.c               | 20 +++++++-------------
 arch/x86/mm/init_64.c               | 10 ++++------
 drivers/char/agp/intel-agp.c        |  6 +++---
 drivers/video/vermilion/vermilion.c |  9 +++------
 include/asm-x86/agp.h               |  4 ++--
 sound/pci/intel8x0.c                |  7 +++++--
 8 files changed, 29 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 57b57778bf60..a70fe77354b8 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -396,10 +396,10 @@ static void __init runtime_code_page_mkexec(void)
 		md = p;
 		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
 		if (md->type == EFI_RUNTIME_SERVICES_CODE &&
-		    (end >> PAGE_SHIFT) <= max_pfn_mapped)
-			change_page_attr_addr(md->virt_addr,
-					      md->num_pages,
-					      PAGE_KERNEL_EXEC_NOCACHE);
+		    (end >> PAGE_SHIFT) <= max_pfn_mapped) {
+			set_memory_x(md->virt_addr, md->num_pages);
+			set_memory_uc(md->virt_addr, md->num_pages);
+		}
 	}
 	__flush_tlb_all();
 }
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index 04ca5c5221d7..8860c6eba8ab 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -570,8 +570,7 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 	gatt = (void *)__get_free_pages(GFP_KERNEL, get_order(gatt_size));
 	if (!gatt)
 		panic("Cannot allocate GATT table");
-	if (change_page_attr_addr((unsigned long)gatt, gatt_size >> PAGE_SHIFT,
-				  PAGE_KERNEL_NOCACHE))
+	if (set_memory_uc((unsigned long)gatt, gatt_size >> PAGE_SHIFT))
 		panic("Could not set GART PTEs to uncacheable pages");
 	global_flush_tlb();
 
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 016c8ccd1d8d..7c9bb3076b8a 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -770,34 +770,30 @@ void mark_rodata_ro(void)
 	if (num_possible_cpus() <= 1)
 #endif
 	{
-		change_page_attr(virt_to_page(start),
-		                 size >> PAGE_SHIFT, PAGE_KERNEL_RX);
+		set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
 		printk("Write protecting the kernel text: %luk\n", size >> 10);
 
 #ifdef CONFIG_CPA_DEBUG
 		global_flush_tlb();
 
 		printk("Testing CPA: Reverting %lx-%lx\n", start, start+size);
-		change_page_attr(virt_to_page(start), size>>PAGE_SHIFT,
-				 PAGE_KERNEL_EXEC);
+		set_pages_rw(virt_to_page(start), size>>PAGE_SHIFT);
 		global_flush_tlb();
 
 		printk("Testing CPA: write protecting again\n");
-		change_page_attr(virt_to_page(start), size>>PAGE_SHIFT,
-				PAGE_KERNEL_RX);
+		set_pages_ro(virt_to_page(start), size>>PAGE_SHIFT);
 		global_flush_tlb();
 #endif
 	}
 #endif
 	start += size;
 	size = (unsigned long)__end_rodata - start;
-	change_page_attr(virt_to_page(start),
-	                 size >> PAGE_SHIFT, PAGE_KERNEL_RO);
+	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
 	printk("Write protecting the kernel read-only data: %luk\n",
 	       size >> 10);
 
 	/*
-	 * change_page_attr() requires a global_flush_tlb() call after it.
+	 * set_pages_*() requires a global_flush_tlb() call after it.
 	 * We do this after the printk so that if something went wrong in the
 	 * change, the printk gets out at least to give a better debug hint
 	 * of who is the culprit.
@@ -806,13 +802,11 @@ void mark_rodata_ro(void)
 
 #ifdef CONFIG_CPA_DEBUG
 	printk("Testing CPA: undo %lx-%lx\n", start, start + size);
-	change_page_attr(virt_to_page(start), size >> PAGE_SHIFT,
-				PAGE_KERNEL);
+	set_pages_rw(virt_to_page(start), size >> PAGE_SHIFT);
 	global_flush_tlb();
 
 	printk("Testing CPA: write protecting again\n");
-	change_page_attr(virt_to_page(start), size >> PAGE_SHIFT,
-				PAGE_KERNEL_RO);
+	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
 	global_flush_tlb();
 #endif
 }
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index c250580a9432..05bb12db0b09 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -556,8 +556,6 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		init_page_count(virt_to_page(addr));
 		memset((void *)(addr & ~(PAGE_SIZE-1)),
 			POISON_FREE_INITMEM, PAGE_SIZE);
-		if (addr >= __START_KERNEL_map)
-			change_page_attr_addr(addr, 1, __pgprot(0));
 		free_page(addr);
 		totalram_pages++;
 	}
@@ -594,13 +592,13 @@ void mark_rodata_ro(void)
 	if (end <= start)
 		return;
 
-	change_page_attr_addr(start, (end - start) >> PAGE_SHIFT, PAGE_KERNEL_RO);
+	set_memory_ro(start, (end - start) >> PAGE_SHIFT);
 
 	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
 	       (end - start) >> 10);
 
 	/*
-	 * change_page_attr_addr() requires a global_flush_tlb() call after it.
+	 * set_memory_*() requires a global_flush_tlb() call after it.
 	 * We do this after the printk so that if something went wrong in the
 	 * change, the printk gets out at least to give a better debug hint
 	 * of who is the culprit.
@@ -609,11 +607,11 @@ void mark_rodata_ro(void)
 
 #ifdef CONFIG_CPA_DEBUG
 	printk("Testing CPA: undo %lx-%lx\n", start, end);
-	change_page_attr_addr(start, (end - start) >> PAGE_SHIFT, PAGE_KERNEL);
+	set_memory_rw(start, (end-start) >> PAGE_SHIFT);
 	global_flush_tlb();
 
 	printk("Testing CPA: again\n");
-	change_page_attr_addr(start, (end - start) >> PAGE_SHIFT, PAGE_KERNEL_RO);
+	set_memory_ro(start, (end-start) >> PAGE_SHIFT);
 	global_flush_tlb();
 #endif
 }
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index 03eac1eb8e0f..c03a7143928f 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -210,8 +210,8 @@ static void *i8xx_alloc_pages(void)
 	if (page == NULL)
 		return NULL;
 
-	if (change_page_attr(page, 4, PAGE_KERNEL_NOCACHE) < 0) {
-		change_page_attr(page, 4, PAGE_KERNEL);
+	if (set_pages_uc(page, 4) < 0) {
+		set_pages_wb(page, 4);
 		global_flush_tlb();
 		__free_pages(page, 2);
 		return NULL;
@@ -230,7 +230,7 @@ static void i8xx_destroy_pages(void *addr)
 		return;
 
 	page = virt_to_page(addr);
-	change_page_attr(page, 4, PAGE_KERNEL);
+	set_pages_wb(page, 4);
 	global_flush_tlb();
 	put_page(page);
 	__free_pages(page, 2);
diff --git a/drivers/video/vermilion/vermilion.c b/drivers/video/vermilion/vermilion.c
index c31f549ebea0..fb72778dee48 100644
--- a/drivers/video/vermilion/vermilion.c
+++ b/drivers/video/vermilion/vermilion.c
@@ -88,9 +88,7 @@ static int vmlfb_alloc_vram_area(struct vram_area *va, unsigned max_order,
 {
 	gfp_t flags;
 	unsigned long i;
-	pgprot_t wc_pageprot;
 
-	wc_pageprot = PAGE_KERNEL_NOCACHE;
 	max_order++;
 	do {
 		/*
@@ -131,8 +129,7 @@ static int vmlfb_alloc_vram_area(struct vram_area *va, unsigned max_order,
 	 */
 
 	global_flush_tlb();
-	change_page_attr(virt_to_page(va->logical), va->size >> PAGE_SHIFT,
-			 wc_pageprot);
+	set_pages_uc(virt_to_page(va->logical), va->size >> PAGE_SHIFT);
 	global_flush_tlb();
 
 	printk(KERN_DEBUG MODULE_NAME
@@ -157,8 +154,8 @@ static void vmlfb_free_vram_area(struct vram_area *va)
 		 * Reset the linear kernel map caching policy.
 		 */
 
-		change_page_attr(virt_to_page(va->logical),
-				 va->size >> PAGE_SHIFT, PAGE_KERNEL);
+		set_pages_wb(virt_to_page(va->logical),
+				 va->size >> PAGE_SHIFT);
 		global_flush_tlb();
 
 		/*
diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h
index 62df2a9e7130..f6df72561832 100644
--- a/include/asm-x86/agp.h
+++ b/include/asm-x86/agp.h
@@ -16,8 +16,8 @@
  * Caller's responsibility to call global_flush_tlb() for performance
  * reasons
  */
-#define map_page_into_agp(page) change_page_attr(page, 1, PAGE_KERNEL_NOCACHE)
-#define unmap_page_from_agp(page) change_page_attr(page, 1, PAGE_KERNEL)
+#define map_page_into_agp(page) set_pages_uc(page, 1)
+#define unmap_page_from_agp(page) set_pages_wb(page, 1)
 #define flush_agp_mappings() global_flush_tlb()
 
 /*
diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index b4a38a3d855b..e5650905296e 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -711,11 +711,14 @@ static void snd_intel8x0_setup_periods(struct intel8x0 *chip, struct ichdev *ich
 static void fill_nocache(void *buf, int size, int nocache)
 {
 	size = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	change_page_attr(virt_to_page(buf), size, nocache ? PAGE_KERNEL_NOCACHE : PAGE_KERNEL);
+	if (nocache)
+		set_pages_uc(virt_to_page(buf), size);
+	else
+		set_pages_wb(virt_to_page(buf), size);
 	global_flush_tlb();
 }
 #else
-#define fill_nocache(buf,size,nocache)
+#define fill_nocache(buf, size, nocache) do { ; } while (0)
 #endif
 
 /*
-- 
cgit v1.2.3


From e1271f686a0e376aa6ee97984c16f91a787e4480 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: deprecate change_page_attr() for drivers

With the introduction of the new API, no driver or non-archcore code needs
to use c-p-a anymore, so this patch also deprecates the EXPORT_SYMBOL of CPA
(it's a horrible API after all).

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pageattr.c       | 2 +-
 include/asm-x86/cacheflush.h | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index f3c9510b8d92..19f7f7a0b36f 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -284,7 +284,7 @@ int change_page_attr(struct page *page, int numpages, pgprot_t prot)
 
 	return change_page_attr_addr(addr, numpages, prot);
 }
-EXPORT_SYMBOL(change_page_attr);
+EXPORT_UNUSED_SYMBOL(change_page_attr); /* to be removed in 2.6.27 */
 
 /**
  * change_page_attr_set - Change page table attributes in the linear mapping.
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 7b928d55c190..e79159bc0987 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -25,7 +25,8 @@
 	memcpy(dst, src, len)
 
 void global_flush_tlb(void);
-int change_page_attr(struct page *page, int numpages, pgprot_t prot);
+int __deprecated_for_modules change_page_attr(struct page *page, int numpages,
+								pgprot_t prot);
 int change_page_attr_addr(unsigned long addr, int numpages, pgprot_t prot);
 
 int set_pages_uc(struct page *page, int numpages);
-- 
cgit v1.2.3


From 5f5192b9feeff6a96c97c143c3ca558fdbe2dc8e Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: move page_is_ram() function

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_32.c     | 22 ----------------------
 arch/x86/mm/ioremap.c     | 24 ++++++++++++++++++++++++
 include/asm-x86/page.h    |  3 +++
 include/asm-x86/page_32.h |  1 -
 4 files changed, 27 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 7c9bb3076b8a..f7b941c3b2c3 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -211,28 +211,6 @@ static inline int page_kills_ppro(unsigned long pagenr)
 	return 0;
 }
 
-int page_is_ram(unsigned long pagenr)
-{
-	int i;
-	unsigned long addr, end;
-
-	for (i = 0; i < e820.nr_map; i++) {
-
-		if (e820.map[i].type != E820_RAM)	/* not usable memory */
-			continue;
-		/*
-		 *	!!!FIXME!!! Some BIOSen report areas as RAM that
-		 *	are not. Notably the 640->1Mb area. We need a sanity
-		 *	check here.
-		 */
-		addr = (e820.map[i].addr+PAGE_SIZE-1) >> PAGE_SHIFT;
-		end = (e820.map[i].addr+e820.map[i].size) >> PAGE_SHIFT;
-		if  ((pagenr >= addr) && (pagenr < end))
-			return 1;
-	}
-	return 0;
-}
-
 #ifdef CONFIG_HIGHMEM
 pte_t *kmap_pte;
 pgprot_t kmap_prot;
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index f4a2082568c8..d3026e1906f9 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -31,6 +31,30 @@ EXPORT_SYMBOL(__phys_addr);
 
 #endif
 
+int page_is_ram(unsigned long pagenr)
+{
+	unsigned long addr, end;
+	int i;
+
+	for (i = 0; i < e820.nr_map; i++) {
+		/*
+		 * Not usable memory:
+		 */
+		if (e820.map[i].type != E820_RAM)
+			continue;
+		/*
+		 *	!!!FIXME!!! Some BIOSen report areas as RAM that
+		 *	are not. Notably the 640->1Mb area. We need a sanity
+		 *	check here.
+		 */
+		addr = (e820.map[i].addr + PAGE_SIZE-1) >> PAGE_SHIFT;
+		end = (e820.map[i].addr + e820.map[i].size) >> PAGE_SHIFT;
+		if ((pagenr >= addr) && (pagenr < end))
+			return 1;
+	}
+	return 0;
+}
+
 /*
  * Fix up the linear direct mapping of the kernel to avoid cache attribute
  * conflicts.
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index e2c79d8dcdcf..c8b30efeed85 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -47,6 +47,9 @@
 
 
 #ifndef __ASSEMBLY__
+
+extern int page_is_ram(unsigned long pagenr);
+
 struct page;
 
 static void inline clear_user_page(void *page, unsigned long vaddr,
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 11c4b39cada1..a6fd10f230d2 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -73,7 +73,6 @@ extern int nx_enabled;
  */
 extern unsigned int __VMALLOC_RESERVE;
 extern int sysctl_legacy_va_layout;
-extern int page_is_ram(unsigned long pagenr);
 
 #define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
 #define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
-- 
cgit v1.2.3


From 950f9d95bed1a366434d3597ea75f5b9d772d74f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: fix the missing BIOS area check in page_is_ram

page_is_ram has a FIXME since ages, which reminds to sanity check the
BIOS area between 640k and 1M, which is sometimes falsely reported as
RAM in the e820 tables.

Implement the sanity check. Move the BIOS range defines from
pageattr.c into e820.h to avoid duplicate defines.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/ioremap.c  | 15 ++++++++++-----
 arch/x86/mm/pageattr.c |  7 +------
 include/asm-x86/e820.h |  3 +++
 3 files changed, 14 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index d3026e1906f9..24e42cab8d58 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -42,13 +42,18 @@ int page_is_ram(unsigned long pagenr)
 		 */
 		if (e820.map[i].type != E820_RAM)
 			continue;
-		/*
-		 *	!!!FIXME!!! Some BIOSen report areas as RAM that
-		 *	are not. Notably the 640->1Mb area. We need a sanity
-		 *	check here.
-		 */
 		addr = (e820.map[i].addr + PAGE_SIZE-1) >> PAGE_SHIFT;
 		end = (e820.map[i].addr + e820.map[i].size) >> PAGE_SHIFT;
+
+		/*
+		 * Sanity check: Some BIOSen report areas as RAM that
+		 * are not. Notably the 640->1Mb area, which is the
+		 * PCI BIOS area.
+		 */
+		if (addr >= (BIOS_BEGIN >> PAGE_SHIFT) &&
+		    end < (BIOS_END >> PAGE_SHIFT))
+			continue;
+
 		if ((pagenr >= addr) && (pagenr < end))
 			return 1;
 	}
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 19f7f7a0b36f..fcd96125c5ae 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -9,18 +9,13 @@
 #include <linux/slab.h>
 #include <linux/mm.h>
 
+#include <asm/e820.h>
 #include <asm/processor.h>
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
 #include <asm/uaccess.h>
 #include <asm/pgalloc.h>
 
-/*
- * We must allow the BIOS range to be executable:
- */
-#define BIOS_BEGIN		0x000a0000
-#define BIOS_END		0x00100000
-
 static inline int
 within(unsigned long addr, unsigned long start, unsigned long end)
 {
diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h
index f96f1853bc41..7004251fc66b 100644
--- a/include/asm-x86/e820.h
+++ b/include/asm-x86/e820.h
@@ -25,6 +25,9 @@ struct e820map {
 #define ISA_START_ADDRESS	0xa0000
 #define ISA_END_ADDRESS		0x100000
 
+#define BIOS_BEGIN		0x000a0000
+#define BIOS_END		0x00100000
+
 #ifdef __KERNEL__
 #ifdef CONFIG_X86_32
 # include "e820_32.h"
-- 
cgit v1.2.3


From 5f8681529cb243b3a492e55f2da9d632ad0d5e32 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:06 +0100
Subject: x86: fix ioremap API

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/ioremap.c   | 11 ++++++++---
 include/asm-x86/io_32.h | 13 +++----------
 include/asm-x86/io_64.h | 13 +++----------
 3 files changed, 14 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index e84c09e7d2c1..8777bb7688f4 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -106,8 +106,8 @@ static int ioremap_change_attr(unsigned long phys_addr, unsigned long size,
  * have to convert them into an offset in a page-aligned mapping, but the
  * caller shouldn't need to know that small detail.
  */
-void __iomem *__ioremap(unsigned long phys_addr, unsigned long size,
-			unsigned long flags)
+static void __iomem *__ioremap(unsigned long phys_addr, unsigned long size,
+			       unsigned long flags)
 {
 	void __iomem *addr;
 	struct vm_struct *area;
@@ -164,7 +164,6 @@ void __iomem *__ioremap(unsigned long phys_addr, unsigned long size,
 
 	return (void __iomem *) (offset + (char __iomem *)addr);
 }
-EXPORT_SYMBOL(__ioremap);
 
 /**
  * ioremap_nocache     -   map bus memory into CPU space
@@ -193,6 +192,12 @@ void __iomem *ioremap_nocache(unsigned long phys_addr, unsigned long size)
 }
 EXPORT_SYMBOL(ioremap_nocache);
 
+void __iomem *ioremap_cache(unsigned long phys_addr, unsigned long size)
+{
+	return __ioremap(phys_addr, size, 0);
+}
+EXPORT_SYMBOL(ioremap_cache);
+
 /**
  * iounmap - Free a IO remapping
  * @addr: virtual address from ioremap_*
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index 059a1fee4de3..586d7aa54ceb 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -100,8 +100,6 @@ static inline void * phys_to_virt(unsigned long address)
  */
 #define page_to_phys(page)    ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
 
-extern void __iomem * __ioremap(unsigned long offset, unsigned long size, unsigned long flags);
-
 /**
  * ioremap     -   map bus memory into CPU space
  * @offset:    bus address of the memory
@@ -116,18 +114,13 @@ extern void __iomem * __ioremap(unsigned long offset, unsigned long size, unsign
  * If the area you are trying to map is a PCI BAR you should have a
  * look at pci_iomap().
  */
-extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
-
-static inline void __iomem *
-ioremap_cache(unsigned long offset, unsigned long size)
-{
-	return __ioremap(offset, size, 0);
-}
+extern void __iomem *ioremap_nocache(unsigned long offset, unsigned long size);
+extern void __iomem *ioremap_cache(unsigned long offset, unsigned long size);
 
 /*
  * The default ioremap() behavior is non-cached:
  */
-static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
+static inline void __iomem *ioremap(unsigned long offset, unsigned long size)
 {
 	return ioremap_nocache(offset, size);
 }
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index 7dee3c6e9c39..ee7a5c955962 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -150,8 +150,6 @@ static inline void * phys_to_virt(unsigned long address)
 
 #include <asm-generic/iomap.h>
 
-extern void __iomem *__ioremap(unsigned long offset, unsigned long size, unsigned long flags);
-
 extern void *early_ioremap(unsigned long addr, unsigned long size);
 extern void early_iounmap(void *addr, unsigned long size);
 
@@ -160,18 +158,13 @@ extern void early_iounmap(void *addr, unsigned long size);
  * it's useful if some control registers are in such an area and write combining
  * or read caching is not desirable:
  */
-extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
-
-static inline void __iomem *
-ioremap_cache(unsigned long offset, unsigned long size)
-{
-	return __ioremap(offset, size, 0);
-}
+extern void __iomem *ioremap_nocache(unsigned long offset, unsigned long size);
+extern void __iomem *ioremap_cache(unsigned long offset, unsigned long size);
 
 /*
  * The default ioremap() behavior is non-cached:
  */
-static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
+static inline void __iomem *ioremap(unsigned long offset, unsigned long size)
 {
 	return ioremap_cache(offset, size);
 }
-- 
cgit v1.2.3


From f62d0f008e889915c93631c04d4c7d871f05bea7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:07 +0100
Subject: x86: cpa: set_memory_notpresent()

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_64.c        | 17 +++++++++++++++--
 arch/x86/mm/pageattr.c       | 34 ++++++++++++++++++++++++++--------
 include/asm-x86/cacheflush.h |  1 +
 3 files changed, 42 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 05bb12db0b09..4757be7b5e55 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -559,8 +559,21 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		free_page(addr);
 		totalram_pages++;
 	}
-	if (addr > __START_KERNEL_map)
-		global_flush_tlb();
+#ifdef CONFIG_DEBUG_RODATA
+	/*
+	 * This will make the __init pages not present and
+	 * not executable, so that any attempt to use a
+	 * __init function from now on will fault immediately
+	 * rather than supriously later when memory gets reused.
+	 *
+	 * We only do this for DEBUG_RODATA to not break up the
+	 * 2Mb kernel mapping just for this debug feature.
+	 */
+	if (begin >= __START_KERNEL_map) {
+		set_memory_np(begin, (end - begin)/PAGE_SIZE);
+		set_memory_nx(begin, (end - begin)/PAGE_SIZE);
+	}
+#endif
 }
 
 void free_initmem(void)
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index fcd96125c5ae..e5910ac37e59 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -357,8 +357,6 @@ int change_page_attr_clear(unsigned long addr, int numpages, pgprot_t prot)
 	return change_page_attr_addr(addr, numpages, prot);
 }
 
-
-
 int set_memory_uc(unsigned long addr, int numpages)
 {
 	pgprot_t uncached;
@@ -402,7 +400,6 @@ int set_memory_ro(unsigned long addr, int numpages)
 	pgprot_val(rw) = _PAGE_RW;
 	return change_page_attr_clear(addr, numpages, rw);
 }
-EXPORT_SYMBOL(set_memory_ro);
 
 int set_memory_rw(unsigned long addr, int numpages)
 {
@@ -411,7 +408,14 @@ int set_memory_rw(unsigned long addr, int numpages)
 	pgprot_val(rw) = _PAGE_RW;
 	return change_page_attr_set(addr, numpages, rw);
 }
-EXPORT_SYMBOL(set_memory_rw);
+
+int set_memory_np(unsigned long addr, int numpages)
+{
+	pgprot_t present;
+
+	pgprot_val(present) = _PAGE_PRESENT;
+	return change_page_attr_clear(addr, numpages, present);
+}
 
 int set_pages_uc(struct page *page, int numpages)
 {
@@ -461,7 +465,6 @@ int set_pages_ro(struct page *page, int numpages)
 	pgprot_val(rw) = _PAGE_RW;
 	return change_page_attr_clear(addr, numpages, rw);
 }
-EXPORT_SYMBOL(set_pages_ro);
 
 int set_pages_rw(struct page *page, int numpages)
 {
@@ -471,8 +474,6 @@ int set_pages_rw(struct page *page, int numpages)
 	pgprot_val(rw) = _PAGE_RW;
 	return change_page_attr_set(addr, numpages, rw);
 }
-EXPORT_SYMBOL(set_pages_rw);
-
 
 void clflush_cache_range(void *addr, int size)
 {
@@ -503,6 +504,20 @@ void global_flush_tlb(void)
 EXPORT_SYMBOL(global_flush_tlb);
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
+
+static int __set_pages_p(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	return change_page_attr_set(addr, numpages,
+				__pgprot(_PAGE_PRESENT | _PAGE_RW));
+}
+
+static int __set_pages_np(struct page *page, int numpages)
+{
+	unsigned long addr = (unsigned long)page_address(page);
+	return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_PRESENT));
+}
+
 void kernel_map_pages(struct page *page, int numpages, int enable)
 {
 	if (PageHighMem(page))
@@ -522,7 +537,10 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 	 * The return value is ignored - the calls cannot fail,
 	 * large pages are disabled at boot time:
 	 */
-	change_page_attr(page, numpages, enable ? PAGE_KERNEL : __pgprot(0));
+	if (enable)
+		__set_pages_p(page, numpages);
+	else
+		__set_pages_np(page, numpages);
 
 	/*
 	 * We should perform an IPI and flush all tlbs,
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index e79159bc0987..a95afaf1240d 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -42,6 +42,7 @@ int set_memory_x(unsigned long addr, int numpages);
 int set_memory_nx(unsigned long addr, int numpages);
 int set_memory_ro(unsigned long addr, int numpages);
 int set_memory_rw(unsigned long addr, int numpages);
+int set_memory_np(unsigned long addr, int numpages);
 
 void clflush_cache_range(void *addr, int size);
 
-- 
cgit v1.2.3


From d1028a154c65d7fadd1b2d0276c077014d401ec7 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Wed, 30 Jan 2008 13:34:07 +0100
Subject: x86: make various pageattr.c functions static

change_page_attr_add is only used in pageattr.c now, so we can
make this function static.
change_page_attr() isn't used anywere at all anymore; this function
is a really bad API anyway so just remove the bloat entirely.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/Makefile_32      |  1 -
 arch/x86/mm/Makefile_64      |  1 -
 arch/x86/mm/pageattr.c       | 51 ++++++++++++--------------------------------
 include/asm-x86/cacheflush.h |  1 -
 4 files changed, 14 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/Makefile_32 b/arch/x86/mm/Makefile_32
index af0d39bea6c5..ffa6d46a1e73 100644
--- a/arch/x86/mm/Makefile_32
+++ b/arch/x86/mm/Makefile_32
@@ -4,7 +4,6 @@
 
 obj-y	:= init_32.o pgtable_32.o fault_32.o ioremap.o extable.o pageattr.o mmap.o
 
-obj-$(CONFIG_CPA_DEBUG) += pageattr-test.o
 obj-$(CONFIG_NUMA) += discontig_32.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_HIGHMEM) += highmem_32.o
diff --git a/arch/x86/mm/Makefile_64 b/arch/x86/mm/Makefile_64
index b564b5a760da..27a090c86e9b 100644
--- a/arch/x86/mm/Makefile_64
+++ b/arch/x86/mm/Makefile_64
@@ -7,4 +7,3 @@ obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_NUMA) += numa_64.o
 obj-$(CONFIG_K8_NUMA) += k8topology_64.o
 obj-$(CONFIG_ACPI_NUMA) += srat_64.o
-obj-$(CONFIG_CPA_DEBUG) += pageattr-test.o
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index e5910ac37e59..e4d2b6930e61 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -210,7 +210,8 @@ repeat:
  * Modules and drivers should use the set_memory_* APIs instead.
  */
 
-int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
+static int change_page_attr_addr(unsigned long address, int numpages,
+								pgprot_t prot)
 {
 	int err = 0, kernel_map = 0, i;
 
@@ -251,36 +252,6 @@ int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
 	return err;
 }
 
-/**
- * change_page_attr - Change page table attributes in the linear mapping.
- * @page: First page to change
- * @numpages: Number of pages to change
- * @prot: New protection/caching type (PAGE_*)
- *
- * Returns 0 on success, otherwise a negated errno.
- *
- * This should be used when a page is mapped with a different caching policy
- * than write-back somewhere - some CPUs do not like it when mappings with
- * different caching policies exist. This changes the page attributes of the
- * in kernel linear mapping too.
- *
- * Caller must call global_flush_tlb() later to make the changes active.
- *
- * The caller needs to ensure that there are no conflicting mappings elsewhere
- * (e.g. in user space) * This function only deals with the kernel linear map.
- *
- * For MMIO areas without mem_map use change_page_attr_addr() instead.
- *
- * Modules and drivers should use the set_pages_* APIs instead.
- */
-int change_page_attr(struct page *page, int numpages, pgprot_t prot)
-{
-	unsigned long addr = (unsigned long)page_address(page);
-
-	return change_page_attr_addr(addr, numpages, prot);
-}
-EXPORT_UNUSED_SYMBOL(change_page_attr); /* to be removed in 2.6.27 */
-
 /**
  * change_page_attr_set - Change page table attributes in the linear mapping.
  * @addr: Virtual address in linear mapping.
@@ -294,15 +265,14 @@ EXPORT_UNUSED_SYMBOL(change_page_attr); /* to be removed in 2.6.27 */
  * different caching policies exist. This changes the page attributes of the
  * in kernel linear mapping too.
  *
- * Caller must call global_flush_tlb() later to make the changes active.
- *
  * The caller needs to ensure that there are no conflicting mappings elsewhere
  * (e.g. in user space) * This function only deals with the kernel linear map.
  *
  * This function is different from change_page_attr() in that only selected bits
  * are impacted, all other bits remain as is.
  */
-int change_page_attr_set(unsigned long addr, int numpages, pgprot_t prot)
+static int change_page_attr_set(unsigned long addr, int numpages,
+								pgprot_t prot)
 {
 	pgprot_t current_prot;
 	int level;
@@ -332,15 +302,14 @@ int change_page_attr_set(unsigned long addr, int numpages, pgprot_t prot)
  * different caching policies exist. This changes the page attributes of the
  * in kernel linear mapping too.
  *
- * Caller must call global_flush_tlb() later to make the changes active.
- *
  * The caller needs to ensure that there are no conflicting mappings elsewhere
  * (e.g. in user space) * This function only deals with the kernel linear map.
  *
  * This function is different from change_page_attr() in that only selected bits
  * are impacted, all other bits remain as is.
  */
-int change_page_attr_clear(unsigned long addr, int numpages, pgprot_t prot)
+static int change_page_attr_clear(unsigned long addr, int numpages,
+								pgprot_t prot)
 {
 	pgprot_t current_prot;
 	int level;
@@ -549,3 +518,11 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 	__flush_tlb_all();
 }
 #endif
+
+/*
+ * The testcases use internal knowledge of the implementation that shouldn't
+ * be exposed to the rest of the kernel. Include these directly here.
+ */
+#ifdef CONFIG_CPA_DEBUG
+#include "pageattr-test.c"
+#endif
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index a95afaf1240d..d15ff359d3e3 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -27,7 +27,6 @@
 void global_flush_tlb(void);
 int __deprecated_for_modules change_page_attr(struct page *page, int numpages,
 								pgprot_t prot);
-int change_page_attr_addr(unsigned long addr, int numpages, pgprot_t prot);
 
 int set_pages_uc(struct page *page, int numpages);
 int set_pages_wb(struct page *page, int numpages);
-- 
cgit v1.2.3


From d7c8f21a8cad0228c7c5ce2bb6dbd95d1ee49d13 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 30 Jan 2008 13:34:07 +0100
Subject: x86: cpa: move flush to cpa

The set_memory_* and set_pages_* family of API's currently requires the
callers to do a global tlb flush after the function call; forgetting this is
a very nasty deathtrap. This patch moves the global tlb flush into
each of the callers

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/pci-gart_64.c       |   1 -
 arch/x86/mm/init_32.c               |  14 ----
 arch/x86/mm/init_64.c               |  10 ---
 arch/x86/mm/ioremap.c               |   2 -
 arch/x86/mm/pageattr.c              | 137 +++++++++++++++++++-----------------
 drivers/char/agp/ali-agp.c          |   2 -
 drivers/char/agp/i460-agp.c         |   2 -
 drivers/char/agp/intel-agp.c        |   5 --
 drivers/video/vermilion/vermilion.c |   6 --
 include/asm-x86/agp.h               |   6 +-
 include/asm-x86/cacheflush.h        |   1 -
 sound/pci/intel8x0.c                |   1 -
 12 files changed, 72 insertions(+), 115 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index 8860c6eba8ab..4d5cc7181982 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -572,7 +572,6 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 		panic("Cannot allocate GATT table");
 	if (set_memory_uc((unsigned long)gatt, gatt_size >> PAGE_SHIFT))
 		panic("Could not set GART PTEs to uncacheable pages");
-	global_flush_tlb();
 
 	memset(gatt, 0, gatt_size);
 	agp_gatt_table = gatt;
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index f7b941c3b2c3..0d3369b900e9 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -752,15 +752,11 @@ void mark_rodata_ro(void)
 		printk("Write protecting the kernel text: %luk\n", size >> 10);
 
 #ifdef CONFIG_CPA_DEBUG
-		global_flush_tlb();
-
 		printk("Testing CPA: Reverting %lx-%lx\n", start, start+size);
 		set_pages_rw(virt_to_page(start), size>>PAGE_SHIFT);
-		global_flush_tlb();
 
 		printk("Testing CPA: write protecting again\n");
 		set_pages_ro(virt_to_page(start), size>>PAGE_SHIFT);
-		global_flush_tlb();
 #endif
 	}
 #endif
@@ -770,22 +766,12 @@ void mark_rodata_ro(void)
 	printk("Write protecting the kernel read-only data: %luk\n",
 	       size >> 10);
 
-	/*
-	 * set_pages_*() requires a global_flush_tlb() call after it.
-	 * We do this after the printk so that if something went wrong in the
-	 * change, the printk gets out at least to give a better debug hint
-	 * of who is the culprit.
-	 */
-	global_flush_tlb();
-
 #ifdef CONFIG_CPA_DEBUG
 	printk("Testing CPA: undo %lx-%lx\n", start, start + size);
 	set_pages_rw(virt_to_page(start), size >> PAGE_SHIFT);
-	global_flush_tlb();
 
 	printk("Testing CPA: write protecting again\n");
 	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
-	global_flush_tlb();
 #endif
 }
 #endif
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 4757be7b5e55..9b69fa54a831 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -610,22 +610,12 @@ void mark_rodata_ro(void)
 	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
 	       (end - start) >> 10);
 
-	/*
-	 * set_memory_*() requires a global_flush_tlb() call after it.
-	 * We do this after the printk so that if something went wrong in the
-	 * change, the printk gets out at least to give a better debug hint
-	 * of who is the culprit.
-	 */
-	global_flush_tlb();
-
 #ifdef CONFIG_CPA_DEBUG
 	printk("Testing CPA: undo %lx-%lx\n", start, end);
 	set_memory_rw(start, (end-start) >> PAGE_SHIFT);
-	global_flush_tlb();
 
 	printk("Testing CPA: again\n");
 	set_memory_ro(start, (end-start) >> PAGE_SHIFT);
-	global_flush_tlb();
 #endif
 }
 #endif
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index b86f66fa5185..6a9a1418bc98 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -96,8 +96,6 @@ static int ioremap_change_attr(unsigned long paddr, unsigned long size,
 		err = set_memory_wb(vaddr, nrpages);
 		break;
 	}
-	if (!err)
-		global_flush_tlb();
 
 	return err;
 }
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index e4d2b6930e61..a2d747c06147 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -22,6 +22,36 @@ within(unsigned long addr, unsigned long start, unsigned long end)
 	return addr >= start && addr < end;
 }
 
+/*
+ * Flushing functions
+ */
+void clflush_cache_range(void *addr, int size)
+{
+	int i;
+
+	for (i = 0; i < size; i += boot_cpu_data.x86_clflush_size)
+		clflush(addr+i);
+}
+
+static void flush_kernel_map(void *arg)
+{
+	/*
+	 * Flush all to work around Errata in early athlons regarding
+	 * large page flushing.
+	 */
+	__flush_tlb_all();
+
+	if (boot_cpu_data.x86_model >= 4)
+		wbinvd();
+}
+
+static void global_flush_tlb(void)
+{
+	BUG_ON(irqs_disabled());
+
+	on_each_cpu(flush_kernel_map, NULL, 1, 1);
+}
+
 /*
  * Certain areas of memory on x86 require very specific protection flags,
  * for example the BIOS area or kernel text. Callers don't always get this
@@ -328,149 +358,124 @@ static int change_page_attr_clear(unsigned long addr, int numpages,
 
 int set_memory_uc(unsigned long addr, int numpages)
 {
-	pgprot_t uncached;
+	int err;
 
-	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
-	return change_page_attr_set(addr, numpages, uncached);
+	err = change_page_attr_set(addr, numpages,
+				__pgprot(_PAGE_PCD | _PAGE_PWT));
+	global_flush_tlb();
+	return err;
 }
 EXPORT_SYMBOL(set_memory_uc);
 
 int set_memory_wb(unsigned long addr, int numpages)
 {
-	pgprot_t uncached;
+	int err;
 
-	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
-	return change_page_attr_clear(addr, numpages, uncached);
+	err = change_page_attr_clear(addr, numpages,
+				__pgprot(_PAGE_PCD | _PAGE_PWT));
+	global_flush_tlb();
+	return err;
 }
 EXPORT_SYMBOL(set_memory_wb);
 
 int set_memory_x(unsigned long addr, int numpages)
 {
-	pgprot_t nx;
+	int err;
 
-	pgprot_val(nx) = _PAGE_NX;
-	return change_page_attr_clear(addr, numpages, nx);
+	err = change_page_attr_clear(addr, numpages,
+				__pgprot(_PAGE_NX));
+	global_flush_tlb();
+	return err;
 }
 EXPORT_SYMBOL(set_memory_x);
 
 int set_memory_nx(unsigned long addr, int numpages)
 {
-	pgprot_t nx;
+	int err;
 
-	pgprot_val(nx) = _PAGE_NX;
-	return change_page_attr_set(addr, numpages, nx);
+	err = change_page_attr_set(addr, numpages,
+				__pgprot(_PAGE_NX));
+	global_flush_tlb();
+	return err;
 }
 EXPORT_SYMBOL(set_memory_nx);
 
 int set_memory_ro(unsigned long addr, int numpages)
 {
-	pgprot_t rw;
+	int err;
 
-	pgprot_val(rw) = _PAGE_RW;
-	return change_page_attr_clear(addr, numpages, rw);
+	err = change_page_attr_clear(addr, numpages,
+				__pgprot(_PAGE_RW));
+	global_flush_tlb();
+	return err;
 }
 
 int set_memory_rw(unsigned long addr, int numpages)
 {
-	pgprot_t rw;
+	int err;
 
-	pgprot_val(rw) = _PAGE_RW;
-	return change_page_attr_set(addr, numpages, rw);
+	err = change_page_attr_set(addr, numpages,
+				__pgprot(_PAGE_RW));
+	global_flush_tlb();
+	return err;
 }
 
 int set_memory_np(unsigned long addr, int numpages)
 {
-	pgprot_t present;
+	int err;
 
-	pgprot_val(present) = _PAGE_PRESENT;
-	return change_page_attr_clear(addr, numpages, present);
+	err = change_page_attr_clear(addr, numpages,
+				__pgprot(_PAGE_PRESENT));
+	global_flush_tlb();
+	return err;
 }
 
 int set_pages_uc(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t uncached;
 
-	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
-	return change_page_attr_set(addr, numpages, uncached);
+	return set_memory_uc(addr, numpages);
 }
 EXPORT_SYMBOL(set_pages_uc);
 
 int set_pages_wb(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t uncached;
 
-	pgprot_val(uncached) = _PAGE_PCD | _PAGE_PWT;
-	return change_page_attr_clear(addr, numpages, uncached);
+	return set_memory_wb(addr, numpages);
 }
 EXPORT_SYMBOL(set_pages_wb);
 
 int set_pages_x(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t nx;
 
-	pgprot_val(nx) = _PAGE_NX;
-	return change_page_attr_clear(addr, numpages, nx);
+	return set_memory_x(addr, numpages);
 }
 EXPORT_SYMBOL(set_pages_x);
 
 int set_pages_nx(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t nx;
 
-	pgprot_val(nx) = _PAGE_NX;
-	return change_page_attr_set(addr, numpages, nx);
+	return set_memory_nx(addr, numpages);
 }
 EXPORT_SYMBOL(set_pages_nx);
 
 int set_pages_ro(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t rw;
 
-	pgprot_val(rw) = _PAGE_RW;
-	return change_page_attr_clear(addr, numpages, rw);
+	return set_memory_ro(addr, numpages);
 }
 
 int set_pages_rw(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
-	pgprot_t rw;
-
-	pgprot_val(rw) = _PAGE_RW;
-	return change_page_attr_set(addr, numpages, rw);
-}
-
-void clflush_cache_range(void *addr, int size)
-{
-	int i;
-
-	for (i = 0; i < size; i += boot_cpu_data.x86_clflush_size)
-		clflush(addr+i);
-}
 
-static void flush_kernel_map(void *arg)
-{
-	/*
-	 * Flush all to work around Errata in early athlons regarding
-	 * large page flushing.
-	 */
-	__flush_tlb_all();
-
-	if (boot_cpu_data.x86_model >= 4)
-		wbinvd();
+	return set_memory_rw(addr, numpages);
 }
 
-void global_flush_tlb(void)
-{
-	BUG_ON(irqs_disabled());
-
-	on_each_cpu(flush_kernel_map, NULL, 1, 1);
-}
-EXPORT_SYMBOL(global_flush_tlb);
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 
diff --git a/drivers/char/agp/ali-agp.c b/drivers/char/agp/ali-agp.c
index aa5ddb716ffb..1ffb381130c3 100644
--- a/drivers/char/agp/ali-agp.c
+++ b/drivers/char/agp/ali-agp.c
@@ -145,7 +145,6 @@ static void *m1541_alloc_page(struct agp_bridge_data *bridge)
 	void *addr = agp_generic_alloc_page(agp_bridge);
 	u32 temp;
 
-	global_flush_tlb();
 	if (!addr)
 		return NULL;
 
@@ -162,7 +161,6 @@ static void ali_destroy_page(void * addr, int flags)
 		if (flags & AGP_PAGE_DESTROY_UNMAP) {
 			global_cache_flush();	/* is this really needed?  --hch */
 			agp_generic_destroy_page(addr, flags);
-			global_flush_tlb();
 		} else
 			agp_generic_destroy_page(addr, flags);
 	}
diff --git a/drivers/char/agp/i460-agp.c b/drivers/char/agp/i460-agp.c
index e72a83e2bad5..76f581c85a7d 100644
--- a/drivers/char/agp/i460-agp.c
+++ b/drivers/char/agp/i460-agp.c
@@ -527,7 +527,6 @@ static void *i460_alloc_page (struct agp_bridge_data *bridge)
 
 	if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) {
 		page = agp_generic_alloc_page(agp_bridge);
-		global_flush_tlb();
 	} else
 		/* Returning NULL would cause problems */
 		/* AK: really dubious code. */
@@ -539,7 +538,6 @@ static void i460_destroy_page (void *page, int flags)
 {
 	if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) {
 		agp_generic_destroy_page(page, flags);
-		global_flush_tlb();
 	}
 }
 
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index c03a7143928f..189efb6ef970 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -212,11 +212,9 @@ static void *i8xx_alloc_pages(void)
 
 	if (set_pages_uc(page, 4) < 0) {
 		set_pages_wb(page, 4);
-		global_flush_tlb();
 		__free_pages(page, 2);
 		return NULL;
 	}
-	global_flush_tlb();
 	get_page(page);
 	atomic_inc(&agp_bridge->current_memory_agp);
 	return page_address(page);
@@ -231,7 +229,6 @@ static void i8xx_destroy_pages(void *addr)
 
 	page = virt_to_page(addr);
 	set_pages_wb(page, 4);
-	global_flush_tlb();
 	put_page(page);
 	__free_pages(page, 2);
 	atomic_dec(&agp_bridge->current_memory_agp);
@@ -341,7 +338,6 @@ static struct agp_memory *alloc_agpphysmem_i8xx(size_t pg_count, int type)
 
 	switch (pg_count) {
 	case 1: addr = agp_bridge->driver->agp_alloc_page(agp_bridge);
-		global_flush_tlb();
 		break;
 	case 4:
 		/* kludge to get 4 physical pages for ARGB cursor */
@@ -404,7 +400,6 @@ static void intel_i810_free_by_type(struct agp_memory *curr)
 		else {
 			agp_bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[0]),
 							     AGP_PAGE_DESTROY_UNMAP);
-			global_flush_tlb();
 			agp_bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[0]),
 							     AGP_PAGE_DESTROY_FREE);
 		}
diff --git a/drivers/video/vermilion/vermilion.c b/drivers/video/vermilion/vermilion.c
index fb72778dee48..1c656667b937 100644
--- a/drivers/video/vermilion/vermilion.c
+++ b/drivers/video/vermilion/vermilion.c
@@ -124,13 +124,8 @@ static int vmlfb_alloc_vram_area(struct vram_area *va, unsigned max_order,
 	/*
 	 * Change caching policy of the linear kernel map to avoid
 	 * mapping type conflicts with user-space mappings.
-	 * The first global_flush_tlb() is really only there to do a global
-	 * wbinvd().
 	 */
-
-	global_flush_tlb();
 	set_pages_uc(virt_to_page(va->logical), va->size >> PAGE_SHIFT);
-	global_flush_tlb();
 
 	printk(KERN_DEBUG MODULE_NAME
 	       ": Allocated %ld bytes vram area at 0x%08lx\n",
@@ -156,7 +151,6 @@ static void vmlfb_free_vram_area(struct vram_area *va)
 
 		set_pages_wb(virt_to_page(va->logical),
 				 va->size >> PAGE_SHIFT);
-		global_flush_tlb();
 
 		/*
 		 * Decrease the usage count on the pages we've used
diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h
index f6df72561832..0c309b9a5217 100644
--- a/include/asm-x86/agp.h
+++ b/include/asm-x86/agp.h
@@ -12,13 +12,9 @@
  * page. This avoids data corruption on some CPUs.
  */
 
-/*
- * Caller's responsibility to call global_flush_tlb() for performance
- * reasons
- */
 #define map_page_into_agp(page) set_pages_uc(page, 1)
 #define unmap_page_from_agp(page) set_pages_wb(page, 1)
-#define flush_agp_mappings() global_flush_tlb()
+#define flush_agp_mappings() do { } while (0)
 
 /*
  * Could use CLFLUSH here if the cpu supports it. But then it would
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index d15ff359d3e3..157da0206ccc 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -24,7 +24,6 @@
 #define copy_from_user_page(vma, page, vaddr, dst, src, len) \
 	memcpy(dst, src, len)
 
-void global_flush_tlb(void);
 int __deprecated_for_modules change_page_attr(struct page *page, int numpages,
 								pgprot_t prot);
 
diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index e5650905296e..4bb97646a67a 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -715,7 +715,6 @@ static void fill_nocache(void *buf, int size, int nocache)
 		set_pages_uc(virt_to_page(buf), size);
 	else
 		set_pages_wb(virt_to_page(buf), size);
-	global_flush_tlb();
 }
 #else
 #define fill_nocache(buf, size, nocache) do { ; } while (0)
-- 
cgit v1.2.3


From 5398f9854f60d670e8ef1ea08c0e0310f253eeb1 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:07 +0100
Subject: x86: remove flush_agp_mappings()

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/char/agp/backend.c | 3 ---
 drivers/char/agp/generic.c | 3 ---
 include/asm-alpha/agp.h    | 1 -
 include/asm-ia64/agp.h     | 1 -
 include/asm-parisc/agp.h   | 1 -
 include/asm-powerpc/agp.h  | 1 -
 include/asm-sparc64/agp.h  | 1 -
 include/asm-x86/agp.h      | 1 -
 8 files changed, 12 deletions(-)

(limited to 'include')

diff --git a/drivers/char/agp/backend.c b/drivers/char/agp/backend.c
index 832ded20fe70..2720882e66fe 100644
--- a/drivers/char/agp/backend.c
+++ b/drivers/char/agp/backend.c
@@ -147,7 +147,6 @@ static int agp_backend_initialize(struct agp_bridge_data *bridge)
 			printk(KERN_ERR PFX "unable to get memory for scratch page.\n");
 			return -ENOMEM;
 		}
-		flush_agp_mappings();
 
 		bridge->scratch_page_real = virt_to_gart(addr);
 		bridge->scratch_page =
@@ -191,7 +190,6 @@ err_out:
 	if (bridge->driver->needs_scratch_page) {
 		bridge->driver->agp_destroy_page(gart_to_virt(bridge->scratch_page_real),
 						 AGP_PAGE_DESTROY_UNMAP);
-		flush_agp_mappings();
 		bridge->driver->agp_destroy_page(gart_to_virt(bridge->scratch_page_real),
 						 AGP_PAGE_DESTROY_FREE);
 	}
@@ -219,7 +217,6 @@ static void agp_backend_cleanup(struct agp_bridge_data *bridge)
 	    bridge->driver->needs_scratch_page) {
 		bridge->driver->agp_destroy_page(gart_to_virt(bridge->scratch_page_real),
 						 AGP_PAGE_DESTROY_UNMAP);
-		flush_agp_mappings();
 		bridge->driver->agp_destroy_page(gart_to_virt(bridge->scratch_page_real),
 						 AGP_PAGE_DESTROY_FREE);
 	}
diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c
index 64b2f6d7059d..1a4674ce0c71 100644
--- a/drivers/char/agp/generic.c
+++ b/drivers/char/agp/generic.c
@@ -197,7 +197,6 @@ void agp_free_memory(struct agp_memory *curr)
 		for (i = 0; i < curr->page_count; i++) {
 			curr->bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[i]), AGP_PAGE_DESTROY_UNMAP);
 		}
-		flush_agp_mappings();
 		for (i = 0; i < curr->page_count; i++) {
 			curr->bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[i]), AGP_PAGE_DESTROY_FREE);
 		}
@@ -267,8 +266,6 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge,
 	}
 	new->bridge = bridge;
 
-	flush_agp_mappings();
-
 	return new;
 }
 EXPORT_SYMBOL(agp_allocate_memory);
diff --git a/include/asm-alpha/agp.h b/include/asm-alpha/agp.h
index ef855a3bc0f5..26c179135293 100644
--- a/include/asm-alpha/agp.h
+++ b/include/asm-alpha/agp.h
@@ -7,7 +7,6 @@
 
 #define map_page_into_agp(page) 
 #define unmap_page_from_agp(page) 
-#define flush_agp_mappings() 
 #define flush_agp_cache() mb()
 
 /* Convert a physical address to an address suitable for the GART. */
diff --git a/include/asm-ia64/agp.h b/include/asm-ia64/agp.h
index 4e517f0e6afa..c11fdd8ab4d7 100644
--- a/include/asm-ia64/agp.h
+++ b/include/asm-ia64/agp.h
@@ -15,7 +15,6 @@
  */
 #define map_page_into_agp(page)		/* nothing */
 #define unmap_page_from_agp(page)	/* nothing */
-#define flush_agp_mappings()		/* nothing */
 #define flush_agp_cache()		mb()
 
 /* Convert a physical address to an address suitable for the GART. */
diff --git a/include/asm-parisc/agp.h b/include/asm-parisc/agp.h
index 9f61d4eb6c01..9651660da639 100644
--- a/include/asm-parisc/agp.h
+++ b/include/asm-parisc/agp.h
@@ -9,7 +9,6 @@
 
 #define map_page_into_agp(page)		/* nothing */
 #define unmap_page_from_agp(page)	/* nothing */
-#define flush_agp_mappings()		/* nothing */
 #define flush_agp_cache()		mb()
 
 /* Convert a physical address to an address suitable for the GART. */
diff --git a/include/asm-powerpc/agp.h b/include/asm-powerpc/agp.h
index e5ccaca2f5a4..86455c4c31ee 100644
--- a/include/asm-powerpc/agp.h
+++ b/include/asm-powerpc/agp.h
@@ -6,7 +6,6 @@
 
 #define map_page_into_agp(page)
 #define unmap_page_from_agp(page)
-#define flush_agp_mappings()
 #define flush_agp_cache() mb()
 
 /* Convert a physical address to an address suitable for the GART. */
diff --git a/include/asm-sparc64/agp.h b/include/asm-sparc64/agp.h
index 58f8cb6ae767..e9fcf0e781ea 100644
--- a/include/asm-sparc64/agp.h
+++ b/include/asm-sparc64/agp.h
@@ -5,7 +5,6 @@
 
 #define map_page_into_agp(page) 
 #define unmap_page_from_agp(page) 
-#define flush_agp_mappings() 
 #define flush_agp_cache() mb()
 
 /* Convert a physical address to an address suitable for the GART. */
diff --git a/include/asm-x86/agp.h b/include/asm-x86/agp.h
index 0c309b9a5217..e4004a9f6a9a 100644
--- a/include/asm-x86/agp.h
+++ b/include/asm-x86/agp.h
@@ -14,7 +14,6 @@
 
 #define map_page_into_agp(page) set_pages_uc(page, 1)
 #define unmap_page_from_agp(page) set_pages_wb(page, 1)
-#define flush_agp_mappings() do { } while (0)
 
 /*
  * Could use CLFLUSH here if the cpu supports it. But then it would
-- 
cgit v1.2.3


From edeed30589f5defe63ce6aaae56f2b7c855e4520 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Wed, 30 Jan 2008 13:34:08 +0100
Subject: x86: add testcases for RODATA and NX protections/attributes

Latest update; I now have 4 NX tests, but 2 fail so they're #if 0'd.
I also cleaned up the NX test code quite a bit, and got rid of the ugly
exception table sorting stuff.

From: Arjan van de Ven <arjan@linux.intel.com>

This patch adds testcases for the CONFIG_DEBUG_RODATA configuration option
as well as the NX CPU feature/mappings. Both testcases can move to tests/
once that patch gets merged into mainline.
(I'm half considering moving the rodata test into mm/init.c but I'll
wait with that until init.c is unified)

As part of this I had to fix a not-quite-right alignment in the vmlinux.lds.h
for the RODATA sections, which lead to 1 page less being marked read only.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/test_nx.c         | 176 ++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/test_rodata.c     |  86 +++++++++++++++++++
 arch/x86/mm/init_32.c             |   3 +
 arch/x86/mm/init_64.c             |   3 +
 include/asm-generic/vmlinux.lds.h |   1 +
 include/asm-x86/cacheflush.h      |   7 ++
 6 files changed, 276 insertions(+)
 create mode 100644 arch/x86/kernel/test_nx.c
 create mode 100644 arch/x86/kernel/test_rodata.c

(limited to 'include')

diff --git a/arch/x86/kernel/test_nx.c b/arch/x86/kernel/test_nx.c
new file mode 100644
index 000000000000..6d7ef11e7975
--- /dev/null
+++ b/arch/x86/kernel/test_nx.c
@@ -0,0 +1,176 @@
+/*
+ * test_nx.c: functional test for NX functionality
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#include <linux/module.h>
+#include <linux/sort.h>
+#include <asm/uaccess.h>
+
+extern int rodata_test_data;
+
+/*
+ * This file checks 4 things:
+ * 1) Check if the stack is not executable
+ * 2) Check if kmalloc memory is not executable
+ * 3) Check if the .rodata section is not executable
+ * 4) Check if the .data section of a module is not executable
+ *
+ * To do this, the test code tries to execute memory in stack/kmalloc/etc,
+ * and then checks if the expected trap happens.
+ *
+ * Sadly, this implies having a dynamic exception handling table entry.
+ * ... which can be done (and will make Rusty cry)... but it can only
+ * be done in a stand-alone module with only 1 entry total.
+ * (otherwise we'd have to sort and that's just too messy)
+ */
+
+
+
+/*
+ * We want to set up an exception handling point on our stack,
+ * which means a variable value. This function is rather dirty
+ * and walks the exception table of the module, looking for a magic
+ * marker and replaces it with a specific function.
+ */
+static void fudze_exception_table(void *marker, void *new)
+{
+	struct module *mod = THIS_MODULE;
+	struct exception_table_entry *extable;
+
+	/*
+	 * Note: This module has only 1 exception table entry,
+	 * so searching and sorting is not needed. If that changes,
+	 * this would be the place to search and re-sort the exception
+	 * table.
+	 */
+	if (mod->num_exentries > 1) {
+		printk(KERN_ERR "test_nx: too many exception table entries!\n");
+		printk(KERN_ERR "test_nx: test results are not reliable.\n");
+		return;
+	}
+	extable = (struct exception_table_entry *)mod->extable;
+	extable[0].insn = (unsigned long)new;
+}
+
+
+/*
+ * exception tables get their symbols translated so we need
+ * to use a fake function to put in there, which we can then
+ * replace at runtime.
+ */
+void foo_label(void);
+
+/*
+ * returns 0 for not-executable, negative for executable
+ *
+ * Note: we cannot allow this function to be inlined, because
+ * that would give us more than 1 exception table entry.
+ * This in turn would break the assumptions above.
+ */
+static noinline int test_address(void *address)
+{
+	unsigned long result;
+
+	/* Set up an exception table entry for our address */
+	fudze_exception_table(&foo_label, address);
+	result = 1;
+	asm volatile(
+		"foo_label:\n"
+		"0:	call *%[fake_code]\n"
+		"1:\n"
+		".section .fixup,\"ax\"\n"
+		"2:	mov %[zero], %[rslt]\n"
+		"	ret\n"
+		".previous\n"
+		".section __ex_table,\"a\"\n"
+		"       .align 8\n"
+		"	.quad 0b\n"
+		"	.quad 2b\n"
+		".previous\n"
+		: [rslt] "=r" (result)
+		: [fake_code] "r" (address), [zero] "r" (0UL), "0" (result)
+	);
+	/* change the exception table back for the next round */
+	fudze_exception_table(address, &foo_label);
+
+	if (result)
+		return -ENODEV;
+	return 0;
+}
+
+static unsigned char test_data = 0xC3; /* 0xC3 is the opcode for "ret" */
+
+static int test_NX(void)
+{
+	int ret = 0;
+	/* 0xC3 is the opcode for "ret" */
+	char stackcode[] = {0xC3, 0x90, 0 };
+	char *heap;
+
+	test_data = 0xC3;
+
+	printk(KERN_INFO "Testing NX protection\n");
+
+	/* Test 1: check if the stack is not executable */
+	if (test_address(&stackcode)) {
+		printk(KERN_ERR "test_nx: stack was executable\n");
+		ret = -ENODEV;
+	}
+
+
+	/* Test 2: Check if the heap is executable */
+	heap = kmalloc(64, GFP_KERNEL);
+	if (!heap)
+		return -ENOMEM;
+	heap[0] = 0xC3; /* opcode for "ret" */
+
+	if (test_address(heap)) {
+		printk(KERN_ERR "test_nx: heap was executable\n");
+		ret = -ENODEV;
+	}
+	kfree(heap);
+
+	/*
+	 * The following 2 tests currently fail, this needs to get fixed
+	 * Until then, don't run them to avoid too many people getting scared
+	 * by the error message
+	 */
+#if 0
+
+#ifdef CONFIG_DEBUG_RODATA
+	/* Test 3: Check if the .rodata section is executable */
+	if (rodata_test_data != 0xC3) {
+		printk(KERN_ERR "test_nx: .rodata marker has invalid value\n");
+		ret = -ENODEV;
+	} else if (test_address(&rodata_test_data)) {
+		printk(KERN_ERR "test_nx: .rodata section is executable\n");
+		ret = -ENODEV;
+	}
+#endif
+
+	/* Test 4: Check if the .data section of a module is executable */
+	if (test_address(&test_data)) {
+		printk(KERN_ERR "test_nx: .data section is executable\n");
+		ret = -ENODEV;
+	}
+
+#endif
+	return 0;
+}
+
+static void test_exit(void)
+{
+}
+
+module_init(test_NX);
+module_exit(test_exit);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Testcase for the NX infrastructure");
+MODULE_AUTHOR("Arjan van de Ven <arjan@linux.intel.com>");
diff --git a/arch/x86/kernel/test_rodata.c b/arch/x86/kernel/test_rodata.c
new file mode 100644
index 000000000000..4c163772000e
--- /dev/null
+++ b/arch/x86/kernel/test_rodata.c
@@ -0,0 +1,86 @@
+/*
+ * test_rodata.c: functional test for mark_rodata_ro function
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#include <linux/module.h>
+#include <asm/sections.h>
+extern int rodata_test_data;
+
+int rodata_test(void)
+{
+	unsigned long result;
+	unsigned long start, end;
+
+	/* test 1: read the value */
+	/* If this test fails, some previous testrun has clobbered the state */
+	if (!rodata_test_data) {
+		printk(KERN_ERR "rodata_test: test 1 fails (start data)\n");
+		return -ENODEV;
+	}
+
+	/* test 2: write to the variable; this should fault */
+	/*
+	 * If this test fails, we managed to overwrite the data
+	 *
+	 * This is written in assembly to be able to catch the
+	 * exception that is supposed to happen in the correct
+	 * case
+	 */
+
+	result = 1;
+	asm volatile(
+		"0:	mov %[zero],(%[rodata_test])\n"
+		"	mov %[zero], %[rslt]\n"
+		"1:\n"
+		".section .fixup,\"ax\"\n"
+		"2:	jmp 1b\n"
+		".previous\n"
+		".section __ex_table,\"a\"\n"
+		"       .align 16\n"
+#ifdef CONFIG_X86_32
+		"	.long 0b,2b\n"
+#else
+		"	.quad 0b,2b\n"
+#endif
+		".previous"
+		: [rslt] "=r" (result)
+		: [rodata_test] "r" (&rodata_test_data), [zero] "r" (0UL)
+	);
+
+
+	if (!result) {
+		printk(KERN_ERR "rodata_test: test data was not read only\n");
+		return -ENODEV;
+	}
+
+	/* test 3: check the value hasn't changed */
+	/* If this test fails, we managed to overwrite the data */
+	if (!rodata_test_data) {
+		printk(KERN_ERR "rodata_test: Test 3 failes (end data)\n");
+		return -ENODEV;
+	}
+	/* test 4: check if the rodata section is 4Kb aligned */
+	start = (unsigned long)__start_rodata;
+	end = (unsigned long)__end_rodata;
+	if (start & (PAGE_SIZE - 1)) {
+		printk(KERN_ERR "rodata_test: .rodata is not 4k aligned\n");
+		return -ENODEV;
+	}
+	if (end & (PAGE_SIZE - 1)) {
+		printk(KERN_ERR "rodata_test: .rodata end is not 4k aligned\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Testcase for the DEBUG_RODATA infrastructure");
+MODULE_AUTHOR("Arjan van de Ven <arjan@linux.intel.com>");
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 4d1156545194..a72737c05747 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -736,6 +736,8 @@ static int noinline do_test_wp_bit(void)
 }
 
 #ifdef CONFIG_DEBUG_RODATA
+const int rodata_test_data = 0xC3;
+EXPORT_SYMBOL_GPL(rodata_test_data);
 
 void mark_rodata_ro(void)
 {
@@ -765,6 +767,7 @@ void mark_rodata_ro(void)
 	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
 	printk("Write protecting the kernel read-only data: %luk\n",
 	       size >> 10);
+	rodata_test();
 
 #ifdef CONFIG_CPA_DEBUG
 	printk("Testing CPA: undo %lx-%lx\n", start, start + size);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index f97ace7a55e5..50d29f5da02b 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -573,6 +573,7 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		set_memory_rw(begin, (end - begin)/PAGE_SIZE);
 		set_memory_np(begin, (end - begin)/PAGE_SIZE);
 		set_memory_nx(begin, (end - begin)/PAGE_SIZE);
+		rodata_test();
 	}
 #endif
 }
@@ -585,6 +586,8 @@ void free_initmem(void)
 }
 
 #ifdef CONFIG_DEBUG_RODATA
+const int rodata_test_data = 0xC3;
+EXPORT_SYMBOL_GPL(rodata_test_data);
 
 void mark_rodata_ro(void)
 {
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 76df771be585..f784d2f34149 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -184,6 +184,7 @@
 		VMLINUX_SYMBOL(__start___param) = .;			\
 		*(__param)						\
 		VMLINUX_SYMBOL(__stop___param) = .;			\
+		. = ALIGN((align));					\
 		VMLINUX_SYMBOL(__end_rodata) = .;			\
 	}								\
 	. = ALIGN((align));
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 157da0206ccc..3e74aff90809 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -47,5 +47,12 @@ void clflush_cache_range(void *addr, int size);
 #ifdef CONFIG_DEBUG_RODATA
 void mark_rodata_ro(void);
 #endif
+#ifdef CONFIG_DEBUG_RODATA_TEST
+void rodata_test(void);
+#else
+static inline void rodata_test(void)
+{
+}
+#endif
 
 #endif
-- 
cgit v1.2.3


From 4c61afcdb2cd4be299c1442b33adf312b695e2d7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:09 +0100
Subject: x86: fix clflush_page_range logic

only present ptes must be flushed.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pageattr.c       | 31 ++++++++++++++++++++++++-------
 include/asm-x86/cacheflush.h |  2 +-
 2 files changed, 25 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index bbfc8e2466ab..97ec9e7d29d9 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -26,7 +26,6 @@ within(unsigned long addr, unsigned long start, unsigned long end)
  * Flushing functions
  */
 
-
 /**
  * clflush_cache_range - flush a cache range with clflush
  * @addr:	virtual start address
@@ -35,13 +34,19 @@ within(unsigned long addr, unsigned long start, unsigned long end)
  * clflush is an unordered instruction which needs fencing with mfence
  * to avoid ordering issues.
  */
-void clflush_cache_range(void *addr, int size)
+void clflush_cache_range(void *vaddr, unsigned int size)
 {
-	int i;
+	void *vend = vaddr + size - 1;
 
 	mb();
-	for (i = 0; i < size; i += boot_cpu_data.x86_clflush_size)
-		clflush(addr+i);
+
+	for (; vaddr < vend; vaddr += boot_cpu_data.x86_clflush_size)
+		clflush(vaddr);
+	/*
+	 * Flush any possible final partial cacheline:
+	 */
+	clflush(vend);
+
 	mb();
 }
 
@@ -74,9 +79,13 @@ static void __cpa_flush_range(void *arg)
 	__flush_tlb_all();
 }
 
-static void cpa_flush_range(unsigned long addr, int numpages)
+static void cpa_flush_range(unsigned long start, int numpages)
 {
+	unsigned int i, level;
+	unsigned long addr;
+
 	BUG_ON(irqs_disabled());
+	WARN_ON(PAGE_ALIGN(start) != start);
 
 	on_each_cpu(__cpa_flush_range, NULL, 1, 1);
 
@@ -86,7 +95,15 @@ static void cpa_flush_range(unsigned long addr, int numpages)
 	 * will cause all other CPUs to flush the same
 	 * cachelines:
 	 */
-	clflush_cache_range((void *) addr, numpages * PAGE_SIZE);
+	for (i = 0, addr = start; i < numpages; i++, addr += PAGE_SIZE) {
+		pte_t *pte = lookup_address(addr, &level);
+
+		/*
+		 * Only flush present addresses:
+		 */
+		if (pte && pte_present(*pte))
+			clflush_cache_range((void *) addr, PAGE_SIZE);
+	}
 }
 
 /*
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 3e74aff90809..8dd8c5e3cc7f 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -42,7 +42,7 @@ int set_memory_ro(unsigned long addr, int numpages);
 int set_memory_rw(unsigned long addr, int numpages);
 int set_memory_np(unsigned long addr, int numpages);
 
-void clflush_cache_range(void *addr, int size);
+void clflush_cache_range(void *addr, unsigned int size);
 
 #ifdef CONFIG_DEBUG_RODATA
 void mark_rodata_ro(void);
-- 
cgit v1.2.3


From 86f03989d99cfa2e1216cdd7aa996852236909cf Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:09 +0100
Subject: x86: cpa: fix the self-test

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug      |  2 +-
 arch/x86/mm/init_32.c       |  4 +--
 arch/x86/mm/init_64.c       | 16 ----------
 arch/x86/mm/pageattr-test.c | 15 ++++-----
 arch/x86/mm/pageattr.c      | 74 +++++++++++++++++++++++----------------------
 include/asm-x86/pgtable.h   |  1 +
 6 files changed, 48 insertions(+), 64 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 2d0bd33b73aa..2e1e3af28c3a 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -40,7 +40,7 @@ comment "Page alloc debug is incompatible with Software Suspend on i386"
 
 config DEBUG_PAGEALLOC
 	bool "Debug page memory allocations"
-	depends on DEBUG_KERNEL
+	depends on DEBUG_KERNEL && X86_32
 	help
 	  Unmap pages from the kernel linear mapping after free_pages().
 	  This results in a large slowdown, but helps to find certain types
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 8d7f723cfc28..8ed5c189d7aa 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -781,8 +781,6 @@ void mark_rodata_ro(void)
 
 void free_init_pages(char *what, unsigned long begin, unsigned long end)
 {
-	unsigned long addr;
-
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	/*
 	 * If debugging page accesses then do not free this memory but
@@ -793,6 +791,8 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		begin, PAGE_ALIGN(end));
 	set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
 #else
+	unsigned long addr;
+
 	/*
 	 * We just marked the kernel text read only above, now that
 	 * we are going to free part of that, we need to make that
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index e0c1e98ad1bf..8a7b725ce3c7 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -569,22 +569,6 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
 		free_page(addr);
 		totalram_pages++;
 	}
-#ifdef CONFIG_DEBUG_RODATA
-	/*
-	 * This will make the __init pages not present and
-	 * not executable, so that any attempt to use a
-	 * __init function from now on will fault immediately
-	 * rather than supriously later when memory gets reused.
-	 *
-	 * We only do this for DEBUG_RODATA to not break up the
-	 * 2Mb kernel mapping just for this debug feature.
-	 */
-	if (begin >= __START_KERNEL_map) {
-		set_memory_rw(begin, (end - begin)/PAGE_SIZE);
-		set_memory_np(begin, (end - begin)/PAGE_SIZE);
-		set_memory_nx(begin, (end - begin)/PAGE_SIZE);
-	}
-#endif
 #endif
 }
 
diff --git a/arch/x86/mm/pageattr-test.c b/arch/x86/mm/pageattr-test.c
index 554820265b95..06353d43f72e 100644
--- a/arch/x86/mm/pageattr-test.c
+++ b/arch/x86/mm/pageattr-test.c
@@ -15,8 +15,7 @@
 #include <asm/kdebug.h>
 
 enum {
-	NTEST			= 400,
-	LOWEST_LEVEL		= PG_LEVEL_4K,
+	NTEST			= 4000,
 #ifdef CONFIG_X86_64
 	LPS			= (1 << PMD_SHIFT),
 #elif defined(CONFIG_X86_PAE)
@@ -59,10 +58,10 @@ static __init int print_split(struct split_state *s)
 			continue;
 		}
 
-		if (level == 2 && sizeof(long) == 8) {
+		if (level == PG_LEVEL_1G && sizeof(long) == 8) {
 			s->gpg++;
 			i += GPS/PAGE_SIZE;
-		} else if (level != LOWEST_LEVEL) {
+		} else if (level == PG_LEVEL_2M) {
 			if (!(pte_val(*pte) & _PAGE_PSE)) {
 				printk(KERN_ERR
 					"%lx level %d but not PSE %Lx\n",
@@ -162,7 +161,7 @@ static __init int exercise_pageattr(void)
 			continue;
 		}
 
-		err = __change_page_attr_clear(addr[i], len[i],
+		err = change_page_attr_clear(addr[i], len[i],
 					       __pgprot(_PAGE_GLOBAL));
 		if (err < 0) {
 			printk(KERN_ERR "CPA %d failed %d\n", i, err);
@@ -175,7 +174,7 @@ static __init int exercise_pageattr(void)
 				pte ? (u64)pte_val(*pte) : 0ULL);
 			failed++;
 		}
-		if (level != LOWEST_LEVEL) {
+		if (level != PG_LEVEL_4K) {
 			printk(KERN_ERR "CPA %lx: unexpected level %d\n",
 				addr[i], level);
 			failed++;
@@ -183,7 +182,6 @@ static __init int exercise_pageattr(void)
 
 	}
 	vfree(bm);
-	cpa_flush_all();
 
 	failed += print_split(&sb);
 
@@ -197,7 +195,7 @@ static __init int exercise_pageattr(void)
 			failed++;
 			continue;
 		}
-		err = __change_page_attr_set(addr[i], len[i],
+		err = change_page_attr_set(addr[i], len[i],
 					     __pgprot(_PAGE_GLOBAL));
 		if (err < 0) {
 			printk(KERN_ERR "CPA reverting failed: %d\n", err);
@@ -211,7 +209,6 @@ static __init int exercise_pageattr(void)
 		}
 
 	}
-	cpa_flush_all();
 
 	failed += print_split(&sc);
 
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 97ec9e7d29d9..532a40bc0e7e 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -197,10 +197,11 @@ static int split_large_page(pte_t *kpte, unsigned long address)
 	unsigned long addr;
 	pte_t *pbase, *tmp;
 	struct page *base;
-	int i, level;
+	unsigned int i, level;
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
-	gfp_flags = GFP_ATOMIC;
+	gfp_flags = __GFP_HIGH | __GFP_NOFAIL | __GFP_NOWARN;
+	gfp_flags = GFP_ATOMIC | __GFP_NOWARN;
 #endif
 	base = alloc_pages(gfp_flags, 0);
 	if (!base)
@@ -224,6 +225,7 @@ static int split_large_page(pte_t *kpte, unsigned long address)
 	paravirt_alloc_pt(&init_mm, page_to_pfn(base));
 #endif
 
+	pgprot_val(ref_prot) &= ~_PAGE_NX;
 	for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE)
 		set_pte(&pbase[i], pfn_pte(addr >> PAGE_SHIFT, ref_prot));
 
@@ -248,7 +250,8 @@ out_unlock:
 }
 
 static int
-__change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot)
+__change_page_attr(unsigned long address, unsigned long pfn,
+		   pgprot_t mask_set, pgprot_t mask_clr)
 {
 	struct page *kpte_page;
 	int level, err = 0;
@@ -267,15 +270,20 @@ repeat:
 	BUG_ON(PageLRU(kpte_page));
 	BUG_ON(PageCompound(kpte_page));
 
-	prot = static_protections(prot, address);
-
 	if (level == PG_LEVEL_4K) {
-		WARN_ON_ONCE(pgprot_val(prot) & _PAGE_PSE);
-		set_pte_atomic(kpte, pfn_pte(pfn, canon_pgprot(prot)));
-	} else {
-		/* Clear the PSE bit for the 4k level pages ! */
-		pgprot_val(prot) = pgprot_val(prot) & ~_PAGE_PSE;
+		pgprot_t new_prot = pte_pgprot(*kpte);
+		pte_t new_pte, old_pte = *kpte;
+
+		pgprot_val(new_prot) &= ~pgprot_val(mask_clr);
+		pgprot_val(new_prot) |= pgprot_val(mask_set);
+
+		new_prot = static_protections(new_prot, address);
+
+		new_pte = pfn_pte(pfn, canon_pgprot(new_prot));
+		BUG_ON(pte_pfn(new_pte) != pte_pfn(old_pte));
 
+		set_pte_atomic(kpte, new_pte);
+	} else {
 		err = split_large_page(kpte, address);
 		if (!err)
 			goto repeat;
@@ -297,22 +305,26 @@ repeat:
  * Modules and drivers should use the set_memory_* APIs instead.
  */
 
-static int change_page_attr_addr(unsigned long address, pgprot_t prot)
+static int
+change_page_attr_addr(unsigned long address, pgprot_t mask_set,
+							pgprot_t mask_clr)
 {
 	int err = 0, kernel_map = 0;
-	unsigned long pfn = __pa(address) >> PAGE_SHIFT;
+	unsigned long pfn;
 
 #ifdef CONFIG_X86_64
 	if (address >= __START_KERNEL_map &&
 			address < __START_KERNEL_map + KERNEL_TEXT_SIZE) {
 
-		address = (unsigned long)__va(__pa(address));
+		address = (unsigned long)__va(__pa((void *)address));
 		kernel_map = 1;
 	}
 #endif
 
-	if (!kernel_map || pte_present(pfn_pte(0, prot))) {
-		err = __change_page_attr(address, pfn, prot);
+	pfn = __pa(address) >> PAGE_SHIFT;
+
+	if (!kernel_map || 1) {
+		err = __change_page_attr(address, pfn, mask_set, mask_clr);
 		if (err)
 			return err;
 	}
@@ -324,12 +336,15 @@ static int change_page_attr_addr(unsigned long address, pgprot_t prot)
 	 */
 	if (__pa(address) < KERNEL_TEXT_SIZE) {
 		unsigned long addr2;
-		pgprot_t prot2;
 
-		addr2 = __START_KERNEL_map + __pa(address);
+		addr2 = __pa(address) + __START_KERNEL_map - phys_base;
 		/* Make sure the kernel mappings stay executable */
-		prot2 = pte_pgprot(pte_mkexec(pfn_pte(0, prot)));
-		err = __change_page_attr(addr2, pfn, prot2);
+		pgprot_val(mask_clr) |= _PAGE_NX;
+		/*
+		 * Our high aliases are imprecise, so do not propagate
+		 * failures back to users:
+		 */
+		__change_page_attr(addr2, pfn, mask_set, mask_clr);
 	}
 #endif
 
@@ -339,26 +354,13 @@ static int change_page_attr_addr(unsigned long address, pgprot_t prot)
 static int __change_page_attr_set_clr(unsigned long addr, int numpages,
 				      pgprot_t mask_set, pgprot_t mask_clr)
 {
-	pgprot_t new_prot;
-	int level;
-	pte_t *pte;
-	int i, ret;
-
-	for (i = 0; i < numpages ; i++) {
-
-		pte = lookup_address(addr, &level);
-		if (!pte)
-			return -EINVAL;
-
-		new_prot = pte_pgprot(*pte);
-
-		pgprot_val(new_prot) &= ~pgprot_val(mask_clr);
-		pgprot_val(new_prot) |= pgprot_val(mask_set);
+	unsigned int i;
+	int ret;
 
-		ret = change_page_attr_addr(addr, new_prot);
+	for (i = 0; i < numpages ; i++, addr += PAGE_SIZE) {
+		ret = change_page_attr_addr(addr, mask_set, mask_clr);
 		if (ret)
 			return ret;
-		addr += PAGE_SIZE;
 	}
 
 	return 0;
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index ee40a88882f6..269e7e29ea8e 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -240,6 +240,7 @@ enum {
 	PG_LEVEL_NONE,
 	PG_LEVEL_4K,
 	PG_LEVEL_2M,
+	PG_LEVEL_1G,
 };
 
 /*
-- 
cgit v1.2.3


From 9af993a92623e022c176459fa6607a564b9a7eaf Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:09 +0100
Subject: x86: make ioremap() UC by default

Yes! A mere 120 c_p_a() fixing and rewriting patches later,
we are now confident that we can enable UC by default for
ioremap(), on x86 too.

Every other architectures was doing this already. Doing so
makes Linux more robust against MTRR mixups (which might go
unnoticed if BIOS writers test other OSs only - where PAT
might override bad MTRRs defaults).

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/io_64.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index ee7a5c955962..f64a59cc396d 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -166,7 +166,7 @@ extern void __iomem *ioremap_cache(unsigned long offset, unsigned long size);
  */
 static inline void __iomem *ioremap(unsigned long offset, unsigned long size)
 {
-	return ioremap_cache(offset, size);
+	return ioremap_nocache(offset, size);
 }
 
 extern void iounmap(volatile void __iomem *addr);
-- 
cgit v1.2.3


From 8550eb99821b3f78cddfd19964f30e8bc4e429e0 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 30 Jan 2008 13:34:10 +0100
Subject: x86: arch/x86/mm/init_32.c cleanup

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_32.c     | 240 +++++++++++++++++++++++-----------------------
 include/asm-x86/bugs.h    |   3 +-
 include/asm-x86/numa_32.h |  14 ++-
 3 files changed, 137 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 8ed5c189d7aa..c6975fc6944a 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -39,6 +39,7 @@
 #include <asm/fixmap.h>
 #include <asm/e820.h>
 #include <asm/apic.h>
+#include <asm/bugs.h>
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 #include <asm/pgalloc.h>
@@ -50,7 +51,7 @@ unsigned int __VMALLOC_RESERVE = 128 << 20;
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long highstart_pfn, highend_pfn;
 
-static int noinline do_test_wp_bit(void);
+static noinline int do_test_wp_bit(void);
 
 /*
  * Creates a middle page table and puts a pointer to it in the
@@ -61,7 +62,7 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
 {
 	pud_t *pud;
 	pmd_t *pmd_table;
-		
+
 #ifdef CONFIG_X86_PAE
 	if (!(pgd_val(*pgd) & _PAGE_PRESENT)) {
 		pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
@@ -69,18 +70,18 @@ static pmd_t * __init one_md_table_init(pgd_t *pgd)
 		paravirt_alloc_pd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT);
 		set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
 		pud = pud_offset(pgd, 0);
-		if (pmd_table != pmd_offset(pud, 0))
-			BUG();
+		BUG_ON(pmd_table != pmd_offset(pud, 0));
 	}
 #endif
 	pud = pud_offset(pgd, 0);
 	pmd_table = pmd_offset(pud, 0);
+
 	return pmd_table;
 }
 
 /*
  * Create a page table and place a pointer to it in a middle page
- * directory entry.
+ * directory entry:
  */
 static pte_t * __init one_page_table_init(pmd_t *pmd)
 {
@@ -90,9 +91,10 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 #ifdef CONFIG_DEBUG_PAGEALLOC
 		page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
 #endif
-		if (!page_table)
+		if (!page_table) {
 			page_table =
 				(pte_t *)alloc_bootmem_low_pages(PAGE_SIZE);
+		}
 
 		paravirt_alloc_pt(&init_mm, __pa(page_table) >> PAGE_SHIFT);
 		set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
@@ -103,22 +105,21 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
 }
 
 /*
- * This function initializes a certain range of kernel virtual memory 
+ * This function initializes a certain range of kernel virtual memory
  * with new bootmem page tables, everywhere page tables are missing in
  * the given range.
- */
-
-/*
- * NOTE: The pagetables are allocated contiguous on the physical space 
- * so we can cache the place of the first one and move around without 
+ *
+ * NOTE: The pagetables are allocated contiguous on the physical space
+ * so we can cache the place of the first one and move around without
  * checking the pgd every time.
  */
-static void __init page_table_range_init (unsigned long start, unsigned long end, pgd_t *pgd_base)
+static void __init
+page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base)
 {
-	pgd_t *pgd;
-	pmd_t *pmd;
 	int pgd_idx, pmd_idx;
 	unsigned long vaddr;
+	pgd_t *pgd;
+	pmd_t *pmd;
 
 	vaddr = start;
 	pgd_idx = pgd_index(vaddr);
@@ -128,7 +129,8 @@ static void __init page_table_range_init (unsigned long start, unsigned long end
 	for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) {
 		pmd = one_md_table_init(pgd);
 		pmd = pmd + pmd_index(vaddr);
-		for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) {
+		for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end);
+							pmd++, pmd_idx++) {
 			one_page_table_init(pmd);
 
 			vaddr += PMD_SIZE;
@@ -145,17 +147,17 @@ static inline int is_kernel_text(unsigned long addr)
 }
 
 /*
- * This maps the physical memory to kernel virtual address space, a total 
- * of max_low_pfn pages, by creating page tables starting from address 
- * PAGE_OFFSET.
+ * This maps the physical memory to kernel virtual address space, a total
+ * of max_low_pfn pages, by creating page tables starting from address
+ * PAGE_OFFSET:
  */
 static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 {
+	int pgd_idx, pmd_idx, pte_ofs;
 	unsigned long pfn;
 	pgd_t *pgd;
 	pmd_t *pmd;
 	pte_t *pte;
-	int pgd_idx, pmd_idx, pte_ofs;
 
 	pgd_idx = pgd_index(PAGE_OFFSET);
 	pgd = pgd_base + pgd_idx;
@@ -165,40 +167,43 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 		pmd = one_md_table_init(pgd);
 		if (pfn >= max_low_pfn)
 			continue;
+
 		for (pmd_idx = 0;
 		     pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn;
 		     pmd++, pmd_idx++) {
-			unsigned int address = pfn * PAGE_SIZE + PAGE_OFFSET;
+			unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET;
 
-			/* Map with big pages if possible, otherwise
-			   create normal page tables. */
+			/*
+			 * Map with big pages if possible, otherwise
+			 * create normal page tables:
+			 */
 			if (cpu_has_pse) {
-				unsigned int address2;
+				unsigned int addr2;
 				pgprot_t prot = PAGE_KERNEL_LARGE;
 
-				address2 = (pfn + PTRS_PER_PTE - 1) * PAGE_SIZE +
+				addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE +
 					PAGE_OFFSET + PAGE_SIZE-1;
 
-				if (is_kernel_text(address) ||
-				    is_kernel_text(address2))
+				if (is_kernel_text(addr) ||
+				    is_kernel_text(addr2))
 					prot = PAGE_KERNEL_LARGE_EXEC;
 
 				set_pmd(pmd, pfn_pmd(pfn, prot));
 
 				pfn += PTRS_PER_PTE;
-			} else {
-				pte = one_page_table_init(pmd);
+				continue;
+			}
+			pte = one_page_table_init(pmd);
 
-				for (pte_ofs = 0;
-				     pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn;
-				     pte++, pfn++, pte_ofs++, address += PAGE_SIZE) {
-					pgprot_t prot = PAGE_KERNEL;
+			for (pte_ofs = 0;
+			     pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn;
+			     pte++, pfn++, pte_ofs++, addr += PAGE_SIZE) {
+				pgprot_t prot = PAGE_KERNEL;
 
-					if (is_kernel_text(address))
-						prot = PAGE_KERNEL_EXEC;
+				if (is_kernel_text(addr))
+					prot = PAGE_KERNEL_EXEC;
 
-					set_pte(pte, pfn_pte(pfn, prot));
-				}
+				set_pte(pte, pfn_pte(pfn, prot));
 			}
 		}
 	}
@@ -215,14 +220,19 @@ static inline int page_kills_ppro(unsigned long pagenr)
 pte_t *kmap_pte;
 pgprot_t kmap_prot;
 
-#define kmap_get_fixmap_pte(vaddr)					\
-	pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), vaddr), (vaddr)), (vaddr))
+static inline pte_t *kmap_get_fixmap_pte(unsigned long vaddr)
+{
+	return pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr),
+			vaddr), vaddr), vaddr);
+}
 
 static void __init kmap_init(void)
 {
 	unsigned long kmap_vstart;
 
-	/* cache the first kmap pte */
+	/*
+	 * Cache the first kmap pte:
+	 */
 	kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
 	kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
 
@@ -231,11 +241,11 @@ static void __init kmap_init(void)
 
 static void __init permanent_kmaps_init(pgd_t *pgd_base)
 {
+	unsigned long vaddr;
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
-	unsigned long vaddr;
 
 	vaddr = PKMAP_BASE;
 	page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);
@@ -244,7 +254,7 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base)
 	pud = pud_offset(pgd, vaddr);
 	pmd = pmd_offset(pud, vaddr);
 	pte = pte_offset_kernel(pmd, vaddr);
-	pkmap_page_table = pte;	
+	pkmap_page_table = pte;
 }
 
 static void __meminit free_new_highpage(struct page *page)
@@ -263,7 +273,8 @@ void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
 		SetPageReserved(page);
 }
 
-static int __meminit add_one_highpage_hotplug(struct page *page, unsigned long pfn)
+static int __meminit
+add_one_highpage_hotplug(struct page *page, unsigned long pfn)
 {
 	free_new_highpage(page);
 	totalram_pages++;
@@ -271,6 +282,7 @@ static int __meminit add_one_highpage_hotplug(struct page *page, unsigned long p
 	max_mapnr = max(pfn, max_mapnr);
 #endif
 	num_physpages++;
+
 	return 0;
 }
 
@@ -278,7 +290,7 @@ static int __meminit add_one_highpage_hotplug(struct page *page, unsigned long p
  * Not currently handling the NUMA case.
  * Assuming single node and all memory that
  * has been added dynamically that would be
- * onlined here is in HIGHMEM
+ * onlined here is in HIGHMEM.
  */
 void __meminit online_page(struct page *page)
 {
@@ -286,13 +298,11 @@ void __meminit online_page(struct page *page)
 	add_one_highpage_hotplug(page, page_to_pfn(page));
 }
 
-
-#ifdef CONFIG_NUMA
-extern void set_highmem_pages_init(int);
-#else
+#ifndef CONFIG_NUMA
 static void __init set_highmem_pages_init(int bad_ppro)
 {
 	int pfn;
+
 	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) {
 		/*
 		 * Holes under sparsemem might not have no mem_map[]:
@@ -302,23 +312,18 @@ static void __init set_highmem_pages_init(int bad_ppro)
 	}
 	totalram_pages += totalhigh_pages;
 }
-#endif /* CONFIG_FLATMEM */
+#endif /* !CONFIG_NUMA */
 
 #else
-#define kmap_init() do { } while (0)
-#define permanent_kmaps_init(pgd_base) do { } while (0)
-#define set_highmem_pages_init(bad_ppro) do { } while (0)
+# define kmap_init()				do { } while (0)
+# define permanent_kmaps_init(pgd_base)		do { } while (0)
+# define set_highmem_pages_init(bad_ppro)	do { } while (0)
 #endif /* CONFIG_HIGHMEM */
 
 pteval_t __PAGE_KERNEL = _PAGE_KERNEL;
 EXPORT_SYMBOL(__PAGE_KERNEL);
-pteval_t __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
 
-#ifdef CONFIG_NUMA
-extern void __init remap_numa_kva(void);
-#else
-#define remap_numa_kva() do {} while (0)
-#endif
+pteval_t __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
 
 void __init native_pagetable_setup_start(pgd_t *base)
 {
@@ -382,10 +387,10 @@ void __init native_pagetable_setup_done(pgd_t *base)
  * be partially populated, and so it avoids stomping on any existing
  * mappings.
  */
-static void __init pagetable_init (void)
+static void __init pagetable_init(void)
 {
-	unsigned long vaddr, end;
 	pgd_t *pgd_base = swapper_pg_dir;
+	unsigned long vaddr, end;
 
 	paravirt_pagetable_setup_start(pgd_base);
 
@@ -424,7 +429,7 @@ static void __init pagetable_init (void)
  * driver might have split up a kernel 4MB mapping.
  */
 char __nosavedata swsusp_pg_dir[PAGE_SIZE]
-	__attribute__ ((aligned (PAGE_SIZE)));
+	__attribute__ ((aligned(PAGE_SIZE)));
 
 static inline void save_pg_dir(void)
 {
@@ -436,7 +441,7 @@ static inline void save_pg_dir(void)
 }
 #endif
 
-void zap_low_mappings (void)
+void zap_low_mappings(void)
 {
 	int i;
 
@@ -448,23 +453,24 @@ void zap_low_mappings (void)
 	 * Note that "pgd_clear()" doesn't do it for
 	 * us, because pgd_clear() is a no-op on i386.
 	 */
-	for (i = 0; i < USER_PTRS_PER_PGD; i++)
+	for (i = 0; i < USER_PTRS_PER_PGD; i++) {
 #ifdef CONFIG_X86_PAE
 		set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
 #else
 		set_pgd(swapper_pg_dir+i, __pgd(0));
 #endif
+	}
 	flush_tlb_all();
 }
 
-int nx_enabled = 0;
+int nx_enabled;
 
 pteval_t __supported_pte_mask __read_mostly = ~_PAGE_NX;
 EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
 #ifdef CONFIG_X86_PAE
 
-static int disable_nx __initdata = 0;
+static int disable_nx __initdata;
 
 /*
  * noexec = on|off
@@ -481,11 +487,14 @@ static int __init noexec_setup(char *str)
 			__supported_pte_mask |= _PAGE_NX;
 			disable_nx = 0;
 		}
-	} else if (!strcmp(str,"off")) {
-		disable_nx = 1;
-		__supported_pte_mask &= ~_PAGE_NX;
-	} else
-		return -EINVAL;
+	} else {
+		if (!strcmp(str, "off")) {
+			disable_nx = 1;
+			__supported_pte_mask &= ~_PAGE_NX;
+		} else {
+			return -EINVAL;
+		}
+	}
 
 	return 0;
 }
@@ -497,6 +506,7 @@ static void __init set_nx(void)
 
 	if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) {
 		cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]);
+
 		if ((v[3] & (1 << 20)) && !disable_nx) {
 			rdmsr(MSR_EFER, l, h);
 			l |= EFER_NX;
@@ -506,7 +516,6 @@ static void __init set_nx(void)
 		}
 	}
 }
-
 #endif
 
 /*
@@ -523,7 +532,6 @@ void __init paging_init(void)
 	if (nx_enabled)
 		printk("NX (Execute Disable) protection: active\n");
 #endif
-
 	pagetable_init();
 
 	load_cr3(swapper_pg_dir);
@@ -547,7 +555,6 @@ void __init paging_init(void)
  * used to involve black magic jumps to work around some nasty CPU bugs,
  * but fortunately the switch to using exceptions got rid of all that.
  */
-
 static void __init test_wp_bit(void)
 {
 	printk("Checking if this processor honours the WP bit even in supervisor mode... ");
@@ -567,19 +574,16 @@ static void __init test_wp_bit(void)
 	}
 }
 
-static struct kcore_list kcore_mem, kcore_vmalloc; 
+static struct kcore_list kcore_mem, kcore_vmalloc;
 
 void __init mem_init(void)
 {
-	extern int ppro_with_ram_bug(void);
 	int codesize, reservedpages, datasize, initsize;
-	int tmp;
-	int bad_ppro;
+	int tmp, bad_ppro;
 
 #ifdef CONFIG_FLATMEM
 	BUG_ON(!mem_map);
 #endif
-	
 	bad_ppro = ppro_with_ram_bug();
 
 #ifdef CONFIG_HIGHMEM
@@ -591,14 +595,13 @@ void __init mem_init(void)
 		BUG();
 	}
 #endif
- 
 	/* this will put all low memory onto the freelists */
 	totalram_pages += free_all_bootmem();
 
 	reservedpages = 0;
 	for (tmp = 0; tmp < max_low_pfn; tmp++)
 		/*
-		 * Only count reserved RAM pages
+		 * Only count reserved RAM pages:
 		 */
 		if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
 			reservedpages++;
@@ -609,11 +612,12 @@ void __init mem_init(void)
 	datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
 	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
 
-	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT); 
-	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START, 
+	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
+	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
 		   VMALLOC_END-VMALLOC_START);
 
-	printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n",
+	printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, "
+			"%dk reserved, %dk data, %dk init, %ldk highmem)\n",
 		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
 		num_physpages << (PAGE_SHIFT-10),
 		codesize >> 10,
@@ -625,44 +629,45 @@ void __init mem_init(void)
 
 #if 1 /* double-sanity-check paranoia */
 	printk("virtual kernel memory layout:\n"
-	       "    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+		"    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 #ifdef CONFIG_HIGHMEM
-	       "    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+		"    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 #endif
-	       "    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
-	       "    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
-	       "      .init : 0x%08lx - 0x%08lx   (%4ld kB)\n"
-	       "      .data : 0x%08lx - 0x%08lx   (%4ld kB)\n"
-	       "      .text : 0x%08lx - 0x%08lx   (%4ld kB)\n",
-	       FIXADDR_START, FIXADDR_TOP,
-	       (FIXADDR_TOP - FIXADDR_START) >> 10,
+		"    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+		"    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+		"      .init : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+		"      .data : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+		"      .text : 0x%08lx - 0x%08lx   (%4ld kB)\n",
+		FIXADDR_START, FIXADDR_TOP,
+		(FIXADDR_TOP - FIXADDR_START) >> 10,
 
 #ifdef CONFIG_HIGHMEM
-	       PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
-	       (LAST_PKMAP*PAGE_SIZE) >> 10,
+		PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
+		(LAST_PKMAP*PAGE_SIZE) >> 10,
 #endif
 
-	       VMALLOC_START, VMALLOC_END,
-	       (VMALLOC_END - VMALLOC_START) >> 20,
+		VMALLOC_START, VMALLOC_END,
+		(VMALLOC_END - VMALLOC_START) >> 20,
 
-	       (unsigned long)__va(0), (unsigned long)high_memory,
-	       ((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,
+		(unsigned long)__va(0), (unsigned long)high_memory,
+		((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,
 
-	       (unsigned long)&__init_begin, (unsigned long)&__init_end,
-	       ((unsigned long)&__init_end - (unsigned long)&__init_begin) >> 10,
+		(unsigned long)&__init_begin, (unsigned long)&__init_end,
+		((unsigned long)&__init_end -
+		 (unsigned long)&__init_begin) >> 10,
 
-	       (unsigned long)&_etext, (unsigned long)&_edata,
-	       ((unsigned long)&_edata - (unsigned long)&_etext) >> 10,
+		(unsigned long)&_etext, (unsigned long)&_edata,
+		((unsigned long)&_edata - (unsigned long)&_etext) >> 10,
 
-	       (unsigned long)&_text, (unsigned long)&_etext,
-	       ((unsigned long)&_etext - (unsigned long)&_text) >> 10);
+		(unsigned long)&_text, (unsigned long)&_etext,
+		((unsigned long)&_etext - (unsigned long)&_text) >> 10);
 
 #ifdef CONFIG_HIGHMEM
-	BUG_ON(PKMAP_BASE+LAST_PKMAP*PAGE_SIZE > FIXADDR_START);
-	BUG_ON(VMALLOC_END                     > PKMAP_BASE);
+	BUG_ON(PKMAP_BASE + LAST_PKMAP*PAGE_SIZE	> FIXADDR_START);
+	BUG_ON(VMALLOC_END				> PKMAP_BASE);
 #endif
-	BUG_ON(VMALLOC_START                   > VMALLOC_END);
-	BUG_ON((unsigned long)high_memory      > VMALLOC_START);
+	BUG_ON(VMALLOC_START				> VMALLOC_END);
+	BUG_ON((unsigned long)high_memory		> VMALLOC_START);
 #endif /* double-sanity-check paranoia */
 
 #ifdef CONFIG_X86_PAE
@@ -693,45 +698,45 @@ int arch_add_memory(int nid, u64 start, u64 size)
 
 	return __add_pages(zone, start_pfn, nr_pages);
 }
-
 #endif
 
 struct kmem_cache *pmd_cache;
 
 void __init pgtable_cache_init(void)
 {
-	if (PTRS_PER_PMD > 1)
+	if (PTRS_PER_PMD > 1) {
 		pmd_cache = kmem_cache_create("pmd",
 					      PTRS_PER_PMD*sizeof(pmd_t),
 					      PTRS_PER_PMD*sizeof(pmd_t),
 					      SLAB_PANIC,
 					      pmd_ctor);
+	}
 }
 
 /*
  * This function cannot be __init, since exceptions don't work in that
  * section.  Put this after the callers, so that it cannot be inlined.
  */
-static int noinline do_test_wp_bit(void)
+static noinline int do_test_wp_bit(void)
 {
 	char tmp_reg;
 	int flag;
 
 	__asm__ __volatile__(
-		"	movb %0,%1	\n"
-		"1:	movb %1,%0	\n"
-		"	xorl %2,%2	\n"
+		"	movb %0, %1	\n"
+		"1:	movb %1, %0	\n"
+		"	xorl %2, %2	\n"
 		"2:			\n"
-		".section __ex_table,\"a\"\n"
+		".section __ex_table, \"a\"\n"
 		"	.align 4	\n"
-		"	.long 1b,2b	\n"
+		"	.long 1b, 2b	\n"
 		".previous		\n"
 		:"=m" (*(char *)fix_to_virt(FIX_WP_TEST)),
 		 "=q" (tmp_reg),
 		 "=r" (flag)
 		:"2" (1)
 		:"memory");
-	
+
 	return flag;
 }
 
@@ -824,4 +829,3 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 	free_init_pages("initrd memory", start, end);
 }
 #endif
-
diff --git a/include/asm-x86/bugs.h b/include/asm-x86/bugs.h
index aac8317420af..3fcc30dc0731 100644
--- a/include/asm-x86/bugs.h
+++ b/include/asm-x86/bugs.h
@@ -1,6 +1,7 @@
 #ifndef _ASM_X86_BUGS_H
 #define _ASM_X86_BUGS_H
 
-void check_bugs(void);
+extern void check_bugs(void);
+extern int ppro_with_ram_bug(void);
 
 #endif /* _ASM_X86_BUGS_H */
diff --git a/include/asm-x86/numa_32.h b/include/asm-x86/numa_32.h
index 96fcb157db1d..03d0f7a9bf02 100644
--- a/include/asm-x86/numa_32.h
+++ b/include/asm-x86/numa_32.h
@@ -1,3 +1,15 @@
+#ifndef _ASM_X86_32_NUMA_H
+#define _ASM_X86_32_NUMA_H 1
 
-int pxm_to_nid(int pxm);
+extern int pxm_to_nid(int pxm);
 
+#ifdef CONFIG_NUMA
+extern void __init remap_numa_kva(void);
+extern void set_highmem_pages_init(int);
+#else
+static inline void remap_numa_kva(void)
+{
+}
+#endif
+
+#endif /* _ASM_X86_32_NUMA_H */
-- 
cgit v1.2.3


From 6194ba6ff6ccf8d5c54c857600843c67aa82c407 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:34:11 +0100
Subject: x86: don't special-case pmd allocations as much

In x86 PAE mode, stop treating pmds as a special case.  Previously
they were always allocated and freed with the pgd.  The modifies the
code to be the same as 64-bit mode, where they are allocated on
demand.

This is a step on the way to unifying 32/64-bit pagetable allocation
as much as possible.

There is a complicating wart, however.  When you install a new
reference to a pmd in the pgd, the processor isn't guaranteed to see
it unless you reload cr3.  Since reloading cr3 also has the
side-effect of flushing the tlb, this is an expense that we want to
avoid whereever possible.

This patch simply avoids reloading cr3 unless the update is to the
current pagetable.  Later patches will optimise this further.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: William Irwin <wli@holomorphy.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/init_32.c            | 13 --------
 arch/x86/mm/pgtable_32.c         | 68 ----------------------------------------
 include/asm-x86/pgalloc_32.h     | 22 ++++++++++---
 include/asm-x86/pgtable-3level.h | 39 ++++++++++++++++-------
 include/asm-x86/pgtable_32.h     |  3 +-
 5 files changed, 47 insertions(+), 98 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 02d269c07b96..da524fb22422 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -704,19 +704,6 @@ int arch_add_memory(int nid, u64 start, u64 size)
 }
 #endif
 
-struct kmem_cache *pmd_cache;
-
-void __init pgtable_cache_init(void)
-{
-	if (PTRS_PER_PMD > 1) {
-		pmd_cache = kmem_cache_create("pmd",
-					      PTRS_PER_PMD*sizeof(pmd_t),
-					      PTRS_PER_PMD*sizeof(pmd_t),
-					      SLAB_PANIC,
-					      pmd_ctor);
-	}
-}
-
 /*
  * This function cannot be __init, since exceptions don't work in that
  * section.  Put this after the callers, so that it cannot be inlined.
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 3a6c9200058d..5ca3552474ae 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -195,11 +195,6 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	return pte;
 }
 
-void pmd_ctor(struct kmem_cache *cache, void *pmd)
-{
-	memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
-}
-
 /*
  * List of all pgd's needed for non-PAE so it can invalidate entries
  * in both cached and uncached pgd's; not needed for PAE since the
@@ -285,7 +280,6 @@ static void pgd_dtor(void *pgd)
 	if (SHARED_KERNEL_PMD)
 		return;
 
-	paravirt_release_pd(__pa(pgd) >> PAGE_SHIFT);
 	spin_lock_irqsave(&pgd_lock, flags);
 	pgd_list_del(pgd);
 	spin_unlock_irqrestore(&pgd_lock, flags);
@@ -367,84 +361,22 @@ static void pgd_mop_up_pmds(pgd_t *pgd)
 }
 #endif	/* CONFIG_X86_PAE */
 
-/* If we allocate a pmd for part of the kernel address space, then
-   make sure its initialized with the appropriate kernel mappings.
-   Otherwise use a cached zeroed pmd.  */
-static pmd_t *pmd_cache_alloc(int idx)
-{
-	pmd_t *pmd;
-
-	if (idx >= USER_PTRS_PER_PGD) {
-		pmd = (pmd_t *)__get_free_page(GFP_KERNEL);
-
-		if (pmd)
-			memcpy(pmd,
-			       (void *)pgd_page_vaddr(swapper_pg_dir[idx]),
-			       sizeof(pmd_t) * PTRS_PER_PMD);
-	} else
-		pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
-
-	return pmd;
-}
-
-static void pmd_cache_free(pmd_t *pmd, int idx)
-{
-	if (idx >= USER_PTRS_PER_PGD)
-		free_page((unsigned long)pmd);
-	else
-		kmem_cache_free(pmd_cache, pmd);
-}
-
 pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-	int i;
 	pgd_t *pgd = quicklist_alloc(0, GFP_KERNEL, pgd_ctor);
 
-	if (PTRS_PER_PMD == 1 || !pgd)
-		return pgd;
-
 	mm->pgd = pgd;		/* so that alloc_pd can use it */
 
- 	for (i = 0; i < UNSHARED_PTRS_PER_PGD; ++i) {
-		pmd_t *pmd = pmd_cache_alloc(i);
-
-		if (!pmd)
-			goto out_oom;
-
-		paravirt_alloc_pd(mm, __pa(pmd) >> PAGE_SHIFT);
-		set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
-	}
 	if (pgd && !pgd_prepopulate_pmd(mm, pgd)) {
 		quicklist_free(0, pgd_dtor, pgd);
 		pgd = NULL;
 	}
 
 	return pgd;
-
-out_oom:
-	for (i--; i >= 0; i--) {
-		pgd_t pgdent = pgd[i];
-		void* pmd = (void *)__va(pgd_val(pgdent)-1);
-		paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
-		pmd_cache_free(pmd, i);
-	}
-	quicklist_free(0, pgd_dtor, pgd);
-	return NULL;
 }
 
 void pgd_free(pgd_t *pgd)
 {
-	int i;
-
-	/* in the PAE case user pgd entries are overwritten before usage */
-	if (PTRS_PER_PMD > 1)
-		for (i = 0; i < UNSHARED_PTRS_PER_PGD; ++i) {
-			pgd_t pgdent = pgd[i];
-			void* pmd = (void *)__va(pgd_val(pgdent)-1);
-			paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
-			pmd_cache_free(pmd, i);
-		}
-	/* in the non-PAE case, free_pgtables() clears user pgd entries */
 	pgd_mop_up_pmds(pgd);
 	quicklist_free(0, pgd_dtor, pgd);
 }
diff --git a/include/asm-x86/pgalloc_32.h b/include/asm-x86/pgalloc_32.h
index 3482c3427897..0caa37a9a25f 100644
--- a/include/asm-x86/pgalloc_32.h
+++ b/include/asm-x86/pgalloc_32.h
@@ -63,21 +63,35 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte)
  */
 static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
 {
-	BUG();
-	return (pmd_t *)2;
+	return (pmd_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
 }
 
 static inline void pmd_free(pmd_t *pmd)
 {
+	BUG_ON((unsigned long)pmd & (PAGE_SIZE-1));
+	free_page((unsigned long)pmd);
 }
 
 static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 {
+	paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
+	tlb_remove_page(tlb, virt_to_page(pmd));
 }
 
-static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
+static inline void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
 {
-	BUG();
+	paravirt_alloc_pd(mm, __pa(pmd) >> PAGE_SHIFT);
+
+	/* Note: almost everything apart from _PAGE_PRESENT is
+	   reserved at the pmd (PDPT) level. */
+	set_pud(pudp, __pud(__pa(pmd) | _PAGE_PRESENT));
+
+	/*
+	 * Pentium-II erratum A13: in PAE mode we explicitly have to flush
+	 * the TLB via cr3 if the top-level pgd is changed...
+	 */
+	if (mm == current->active_mm)
+		write_cr3(read_cr3());
 }
 #endif	/* CONFIG_X86_PAE */
 
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 62a1ffbc8784..ed4c6f0e57ec 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -15,9 +15,19 @@
 #define pgd_ERROR(e) \
 	printk("%s:%d: bad pgd %p(%016Lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
 
-#define pud_none(pud)				0
-#define pud_bad(pud)				0
-#define pud_present(pud)			1
+
+static inline int pud_none(pud_t pud)
+{
+	return pud_val(pud) == 0;
+}
+static inline int pud_bad(pud_t pud)
+{
+	return (pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER)) != 0;
+}
+static inline int pud_present(pud_t pud)
+{
+	return pud_val(pud) & _PAGE_PRESENT;
+}
 
 /* Rules for using set_pte: the pte being assigned *must* be
  * either not present or in a state where the hardware will
@@ -58,7 +68,7 @@ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
 }
 static inline void native_set_pud(pud_t *pudp, pud_t pud)
 {
-	*pudp = pud;
+	set_64bit((unsigned long long *)(pudp),native_pud_val(pud));
 }
 
 /*
@@ -81,13 +91,20 @@ static inline void native_pmd_clear(pmd_t *pmd)
 	*(tmp + 1) = 0;
 }
 
-/*
- * Pentium-II erratum A13: in PAE mode we explicitly have to flush
- * the TLB via cr3 if the top-level pgd is changed...
- * We do not let the generic code free and clear pgd entries due to
- * this erratum.
- */
-static inline void pud_clear (pud_t * pud) { }
+static inline void pud_clear(pud_t *pudp)
+{
+	set_pud(pudp, __pud(0));
+
+	/*
+	 * Pentium-II erratum A13: in PAE mode we explicitly have to flush
+	 * the TLB via cr3 if the top-level pgd is changed...
+	 *
+	 * XXX I don't think we need to worry about this here, since
+	 * when clearing the pud, the calling code needs to flush the
+	 * tlb anyway.  But do it now for safety's sake. - jsgf
+	 */
+	write_cr3(read_cr3());
+}
 
 #define pud_page(pud) \
 ((struct page *) __va(pud_val(pud) & PAGE_MASK))
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index ca7b150ca8b7..7b61cb5989b0 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -31,8 +31,7 @@ extern spinlock_t pgd_lock;
 extern struct page *pgd_list;
 void check_pgt_cache(void);
 
-void pmd_ctor(struct kmem_cache *, void *);
-void pgtable_cache_init(void);
+static inline void pgtable_cache_init(void) {}
 void paging_init(void);
 
 
-- 
cgit v1.2.3


From f212ec4b7b4d84290f12c9c0416cdea283bf5f40 Mon Sep 17 00:00:00 2001
From: Bernhard Kaindl <bk@suse.de>
Date: Wed, 30 Jan 2008 13:34:11 +0100
Subject: x86: early boot debugging via FireWire (ohci1394_dma=early)

This patch adds a new configuration option, which adds support for a new
early_param which gets checked in arch/x86/kernel/setup_{32,64}.c:setup_arch()
to decide wether OHCI-1394 FireWire controllers should be initialized and
enabled for physical DMA access to allow remote debugging of early problems
like issues ACPI or other subsystems which are executed very early.

If the config option is not enabled, no code is changed, and if the boot
paramenter is not given, no new code is executed, and independent of that,
all new code is freed after boot, so the config option can be even enabled
in standard, non-debug kernels.

With specialized tools, it is then possible to get debugging information
from machines which have no serial ports (notebooks) such as the printk
buffer contents, or any data which can be referenced from global pointers,
if it is stored below the 4GB limit and even memory dumps of of the physical
RAM region below the 4GB limit can be taken without any cooperation from the
CPU of the host, so the machine can be crashed early, it does not matter.

In the extreme, even kernel debuggers can be accessed in this way. I wrote
a small kgdb module and an accompanying gdb stub for FireWire which allows
to gdb to talk to kgdb using remote remory reads and writes over FireWire.

An version of the gdb stub fore FireWire is able to read all global data
from a system which is running a a normal kernel without any kernel debugger,
without any interruption or support of the system's CPU. That way, e.g. the
task struct and so on can be read and even manipulated when the physical DMA
access is granted.

A HOWTO is included in this patch, in Documentation/debugging-via-ohci1394.txt
and I've put a copy online at
ftp://ftp.suse.de/private/bk/firewire/docs/debugging-via-ohci1394.txt

It also has links to all the tools which are available to make use of it
another copy of it is online at:
ftp://ftp.suse.de/private/bk/firewire/kernel/ohci1394_dma_early-v2.diff

Signed-Off-By: Bernhard Kaindl <bk@suse.de>
Tested-By: Thomas Renninger <trenn@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/debugging-via-ohci1394.txt | 179 +++++++++++++++++++
 arch/x86/kernel/setup_32.c               |  11 ++
 arch/x86/kernel/setup_64.c               |  11 ++
 drivers/Makefile                         |   2 +-
 drivers/ieee1394/Makefile                |   1 +
 drivers/ieee1394/init_ohci1394_dma.c     | 285 +++++++++++++++++++++++++++++++
 include/asm-x86/fixmap_32.h              |   3 +
 include/asm-x86/fixmap_64.h              |   3 +
 include/linux/init_ohci1394_dma.h        |   4 +
 lib/Kconfig.debug                        |  28 +++
 10 files changed, 526 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/debugging-via-ohci1394.txt
 create mode 100644 drivers/ieee1394/init_ohci1394_dma.c
 create mode 100644 include/linux/init_ohci1394_dma.h

(limited to 'include')

diff --git a/Documentation/debugging-via-ohci1394.txt b/Documentation/debugging-via-ohci1394.txt
new file mode 100644
index 000000000000..de4804e8b396
--- /dev/null
+++ b/Documentation/debugging-via-ohci1394.txt
@@ -0,0 +1,179 @@
+
+  Using physical DMA provided by OHCI-1394 FireWire controllers for debugging
+  ---------------------------------------------------------------------------
+
+Introduction
+------------
+
+Basically all FireWire controllers which are in use today are compliant
+to the OHCI-1394 specification which defines the controller to be a PCI
+bus master which uses DMA to offload data transfers from the CPU and has
+a "Physical Response Unit" which executes specific requests by employing
+PCI-Bus master DMA after applying filters defined by the OHCI-1394 driver.
+
+Once properly configured, remote machines can send these requests to
+ask the OHCI-1394 controller to perform read and write requests on
+physical system memory and, for read requests, send the result of
+the physical memory read back to the requester.
+
+With that, it is possible to debug issues by reading interesting memory
+locations such as buffers like the printk buffer or the process table.
+
+Retrieving a full system memory dump is also possible over the FireWire,
+using data transfer rates in the order of 10MB/s or more.
+
+Memory access is currently limited to the low 4G of physical address
+space which can be a problem on IA64 machines where memory is located
+mostly above that limit, but it is rarely a problem on more common
+hardware such as hardware based on x86, x86-64 and PowerPC.
+
+Together with a early initialization of the OHCI-1394 controller for debugging,
+this facility proved most useful for examining long debugs logs in the printk
+buffer on to debug early boot problems in areas like ACPI where the system
+fails to boot and other means for debugging (serial port) are either not
+available (notebooks) or too slow for extensive debug information (like ACPI).
+
+Drivers
+-------
+
+The OHCI-1394 drivers in drivers/firewire and drivers/ieee1394 initialize
+the OHCI-1394 controllers to a working state and can be used to enable
+physical DMA. By default you only have to load the driver, and physical
+DMA access will be granted to all remote nodes, but it can be turned off
+when using the ohci1394 driver.
+
+Because these drivers depend on the PCI enumeration to be completed, an
+initialization routine which can runs pretty early (long before console_init(),
+which makes the printk buffer appear on the console can be called) was written.
+
+To activate it, enable CONFIG_PROVIDE_OHCI1394_DMA_INIT (Kernel hacking menu:
+Provide code for enabling DMA over FireWire early on boot) and pass the
+parameter "ohci1394_dma=early" to the recompiled kernel on boot.
+
+Tools
+-----
+
+firescope - Originally developed by Benjamin Herrenschmidt, Andi Kleen ported
+it from PowerPC to x86 and x86_64 and added functionality, firescope can now
+be used to view the printk buffer of a remote machine, even with live update.
+
+Bernhard Kaindl enhanced firescope to support accessing 64-bit machines
+from 32-bit firescope and vice versa:
+- ftp://ftp.suse.de/private/bk/firewire/tools/firescope-0.2.2.tar.bz2
+
+and he implemented fast system dump (alpha version - read README.txt):
+- ftp://ftp.suse.de/private/bk/firewire/tools/firedump-0.1.tar.bz2
+
+There is also a gdb proxy for firewire which allows to use gdb to access
+data which can be referenced from symbols found by gdb in vmlinux:
+- ftp://ftp.suse.de/private/bk/firewire/tools/fireproxy-0.33.tar.bz2
+
+The latest version of this gdb proxy (fireproxy-0.34) can communicate (not
+yet stable) with kgdb over an memory-based communication module (kgdbom).
+
+Getting Started
+---------------
+
+The OHCI-1394 specification regulates that the OHCI-1394 controller must
+disable all physical DMA on each bus reset.
+
+This means that if you want to debug an issue in a system state where
+interrupts are disabled and where no polling of the OHCI-1394 controller
+for bus resets takes place, you have to establish any FireWire cable
+connections and fully initialize all FireWire hardware __before__ the
+system enters such state.
+
+Step-by-step instructions for using firescope with early OHCI initialization:
+
+1) Verify that your hardware is supported:
+
+   Load the ohci1394 or the fw-ohci module and check your kernel logs.
+   You should see a line similar to
+
+   ohci1394: fw-host0: OHCI-1394 1.1 (PCI): IRQ=[18]  MMIO=[fe9ff800-fe9fffff]
+   ... Max Packet=[2048]  IR/IT contexts=[4/8]
+
+   when loading the driver. If you have no supported controller, many PCI,
+   CardBus and even some Express cards which are fully compliant to OHCI-1394
+   specification are available. If it requires no driver for Windows operating
+   systems, it most likely is. Only specialized shops have cards which are not
+   compliant, they are based on TI PCILynx chips and require drivers for Win-
+   dows operating systems.
+
+2) Establish a working FireWire cable connection:
+
+   Any FireWire cable, as long at it provides electrically and mechanically
+   stable connection and has matching connectors (there are small 4-pin and
+   large 6-pin FireWire ports) will do.
+
+   If an driver is running on both machines you should see a line like
+
+   ieee1394: Node added: ID:BUS[0-01:1023]  GUID[0090270001b84bba]
+
+   on both machines in the kernel log when the cable is plugged in
+   and connects the two machines.
+
+3) Test physical DMA using firescope:
+
+   On the debug host,
+	- load the raw1394 module,
+	- make sure that /dev/raw1394 is accessible,
+   then start firescope:
+
+	$ firescope
+	Port 0 (ohci1394) opened, 2 nodes detected
+
+	FireScope
+	---------
+	Target : <unspecified>
+	Gen    : 1
+	[Ctrl-T] choose target
+	[Ctrl-H] this menu
+	[Ctrl-Q] quit
+
+    ------> Press Ctrl-T now, the output should be similar to:
+
+	2 nodes available, local node is: 0
+	 0: ffc0, uuid: 00000000 00000000 [LOCAL]
+	 1: ffc1, uuid: 00279000 ba4bb801
+
+   Besides the [LOCAL] node, it must show another node without error message.
+
+4) Prepare for debugging with early OHCI-1394 initialization:
+
+   4.1) Kernel compilation and installation on debug target
+
+   Compile the kernel to be debugged with CONFIG_PROVIDE_OHCI1394_DMA_INIT
+   (Kernel hacking: Provide code for enabling DMA over FireWire early on boot)
+   enabled and install it on the machine to be debugged (debug target).
+
+   4.2) Transfer the System.map of the debugged kernel to the debug host
+
+   Copy the System.map of the kernel be debugged to the debug host (the host
+   which is connected to the debugged machine over the FireWire cable).
+
+5) Retrieving the printk buffer contents:
+
+   With the FireWire cable connected, the OHCI-1394 driver on the debugging
+   host loaded, reboot the debugged machine, booting the kernel which has
+   CONFIG_PROVIDE_OHCI1394_DMA_INIT enabled, with the option ohci1394_dma=early.
+
+   Then, on the debugging host, run firescope, for example by using -A:
+
+	firescope -A System.map-of-debug-target-kernel
+
+   Note: -A automatically attaches to the first non-local node. It only works
+   reliably if only connected two machines are connected using FireWire.
+
+   After having attached to the debug target, press Ctrl-D to view the
+   complete printk buffer or Ctrl-U to enter auto update mode and get an
+   updated live view of recent kernel messages logged on the debug target.
+
+   Call "firescope -h" to get more information on firescope's options.
+
+Notes
+-----
+Documentation and specifications: ftp://ftp.suse.de/private/bk/firewire/docs
+
+FireWire is a trademark of Apple Inc. - for more information please refer to:
+http://en.wikipedia.org/wiki/FireWire
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 9c0ef4945a58..62adc5f20be5 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -45,6 +45,7 @@
 #include <linux/dmi.h>
 #include <linux/pfn.h>
 #include <linux/pci.h>
+#include <linux/init_ohci1394_dma.h>
 
 #include <video/edid.h>
 
@@ -787,6 +788,16 @@ void __init setup_arch(char **cmdline_p)
 	smp_alloc_memory(); /* AP processor realmode stacks in low memory*/
 #endif
 	paging_init();
+
+	/*
+	 * NOTE: On x86-32, only from this point on, fixmaps are ready for use.
+	 */
+
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+	if (init_ohci1394_dma_early)
+		init_ohci1394_dma_on_all_controllers();
+#endif
+
 	remapped_pgdat_init();
 	sparse_init();
 	zone_sizes_init();
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 697533e86822..77fb87bf6e5a 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -41,6 +41,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/ctype.h>
 #include <linux/uaccess.h>
+#include <linux/init_ohci1394_dma.h>
 
 #include <asm/mtrr.h>
 #include <asm/uaccess.h>
@@ -253,6 +254,11 @@ void __attribute__((weak)) __init memory_setup(void)
        machine_specific_memory_setup();
 }
 
+/*
+ * setup_arch - architecture-specific boot-time initializations
+ *
+ * Note: On x86_64, fixmaps are ready for use even before this is called.
+ */
 void __init setup_arch(char **cmdline_p)
 {
 	unsigned i;
@@ -302,6 +308,11 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+	if (init_ohci1394_dma_early)
+		init_ohci1394_dma_on_all_controllers();
+#endif
+
 	finish_e820_parsing();
 
 	early_gart_iommu_check();
diff --git a/drivers/Makefile b/drivers/Makefile
index 8cb37e3557d4..d92d4d82d001 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -38,7 +38,7 @@ obj-$(CONFIG_SCSI)		+= scsi/
 obj-$(CONFIG_ATA)		+= ata/
 obj-$(CONFIG_FUSION)		+= message/
 obj-$(CONFIG_FIREWIRE)		+= firewire/
-obj-$(CONFIG_IEEE1394)		+= ieee1394/
+obj-y				+= ieee1394/
 obj-$(CONFIG_UIO)		+= uio/
 obj-y				+= cdrom/
 obj-y				+= auxdisplay/
diff --git a/drivers/ieee1394/Makefile b/drivers/ieee1394/Makefile
index 489c133664d5..1f8153b57503 100644
--- a/drivers/ieee1394/Makefile
+++ b/drivers/ieee1394/Makefile
@@ -15,3 +15,4 @@ obj-$(CONFIG_IEEE1394_SBP2) += sbp2.o
 obj-$(CONFIG_IEEE1394_DV1394) += dv1394.o
 obj-$(CONFIG_IEEE1394_ETH1394) += eth1394.o
 
+obj-$(CONFIG_PROVIDE_OHCI1394_DMA_INIT) += init_ohci1394_dma.o
diff --git a/drivers/ieee1394/init_ohci1394_dma.c b/drivers/ieee1394/init_ohci1394_dma.c
new file mode 100644
index 000000000000..ddaab6eb8ace
--- /dev/null
+++ b/drivers/ieee1394/init_ohci1394_dma.c
@@ -0,0 +1,285 @@
+/*
+ * init_ohci1394_dma.c - Initializes physical DMA on all OHCI 1394 controllers
+ *
+ * Copyright (C) 2006-2007      Bernhard Kaindl <bk@suse.de>
+ *
+ * Derived from drivers/ieee1394/ohci1394.c and arch/x86/kernel/early-quirks.c
+ * this file has functions to:
+ * - scan the PCI very early on boot for all OHCI 1394-compliant controllers
+ * - reset and initialize them and make them join the IEEE1394 bus and
+ * - enable physical DMA on them to allow remote debugging
+ *
+ * All code and data is marked as __init and __initdata, respective as
+ * during boot, all OHCI1394 controllers may be claimed by the firewire
+ * stack and at this point, this code should not touch them anymore.
+ *
+ * To use physical DMA after the initialization of the firewire stack,
+ * be sure that the stack enables it and (re-)attach after the bus reset
+ * which may be caused by the firewire stack initialization.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#include <linux/interrupt.h>	/* for ohci1394.h */
+#include <linux/delay.h>
+#include <linux/pci.h>		/* for PCI defines */
+#include <linux/init_ohci1394_dma.h>
+#include <asm/pci-direct.h>	/* for direct PCI config space access */
+#include <asm/fixmap.h>
+
+#include "ieee1394_types.h"
+#include "ohci1394.h"
+
+int __initdata init_ohci1394_dma_early;
+
+/* Reads a PHY register of an OHCI-1394 controller */
+static inline u8 __init get_phy_reg(struct ti_ohci *ohci, u8 addr)
+{
+	int i;
+	quadlet_t r;
+
+	reg_write(ohci, OHCI1394_PhyControl, (addr << 8) | 0x00008000);
+
+	for (i = 0; i < OHCI_LOOP_COUNT; i++) {
+		if (reg_read(ohci, OHCI1394_PhyControl) & 0x80000000)
+			break;
+		mdelay(1);
+	}
+	r = reg_read(ohci, OHCI1394_PhyControl);
+
+	return (r & 0x00ff0000) >> 16;
+}
+
+/* Writes to a PHY register of an OHCI-1394 controller */
+static inline void __init set_phy_reg(struct ti_ohci *ohci, u8 addr, u8 data)
+{
+	int i;
+
+	reg_write(ohci, OHCI1394_PhyControl, (addr << 8) | data | 0x00004000);
+
+	for (i = 0; i < OHCI_LOOP_COUNT; i++) {
+		u32 r = reg_read(ohci, OHCI1394_PhyControl);
+		if (!(r & 0x00004000))
+			break;
+		mdelay(1);
+	}
+}
+
+/* Resets an OHCI-1394 controller (for sane state before initialization) */
+static inline void __init init_ohci1394_soft_reset(struct ti_ohci *ohci) {
+	int i;
+
+	reg_write(ohci, OHCI1394_HCControlSet, OHCI1394_HCControl_softReset);
+
+	for (i = 0; i < OHCI_LOOP_COUNT; i++) {
+		if (!(reg_read(ohci, OHCI1394_HCControlSet)
+				   & OHCI1394_HCControl_softReset))
+			break;
+		mdelay(1);
+	}
+}
+
+/* Basic OHCI-1394 register and port inititalization */
+static inline void __init init_ohci1394_initialize(struct ti_ohci *ohci)
+{
+	quadlet_t bus_options;
+	int num_ports, i;
+
+	/* Put some defaults to these undefined bus options */
+	bus_options = reg_read(ohci, OHCI1394_BusOptions);
+	bus_options |=  0x60000000; /* Enable CMC and ISC */
+	bus_options &= ~0x00ff0000; /* XXX: Set cyc_clk_acc to zero for now */
+	bus_options &= ~0x18000000; /* Disable PMC and BMC */
+	reg_write(ohci, OHCI1394_BusOptions, bus_options);
+
+	/* Set the bus number */
+	reg_write(ohci, OHCI1394_NodeID, 0x0000ffc0);
+
+	/* Enable posted writes */
+	reg_write(ohci, OHCI1394_HCControlSet,
+			OHCI1394_HCControl_postedWriteEnable);
+
+	/* Clear link control register */
+	reg_write(ohci, OHCI1394_LinkControlClear, 0xffffffff);
+
+	/* enable phys */
+	reg_write(ohci, OHCI1394_LinkControlSet,
+			OHCI1394_LinkControl_RcvPhyPkt);
+
+	/* Don't accept phy packets into AR request context */
+	reg_write(ohci, OHCI1394_LinkControlClear, 0x00000400);
+
+	/* Clear the Isochonouys interrupt masks */
+	reg_write(ohci, OHCI1394_IsoRecvIntMaskClear, 0xffffffff);
+	reg_write(ohci, OHCI1394_IsoRecvIntEventClear, 0xffffffff);
+	reg_write(ohci, OHCI1394_IsoXmitIntMaskClear, 0xffffffff);
+	reg_write(ohci, OHCI1394_IsoXmitIntEventClear, 0xffffffff);
+
+	/* Accept asyncronous transfer requests from all nodes for now */
+	reg_write(ohci,OHCI1394_AsReqFilterHiSet, 0x80000000);
+
+	/* Specify asyncronous transfer retries */
+	reg_write(ohci, OHCI1394_ATRetries,
+		  OHCI1394_MAX_AT_REQ_RETRIES |
+		  (OHCI1394_MAX_AT_RESP_RETRIES<<4) |
+		  (OHCI1394_MAX_PHYS_RESP_RETRIES<<8));
+
+	/* We don't want hardware swapping */
+	reg_write(ohci, OHCI1394_HCControlClear, OHCI1394_HCControl_noByteSwap);
+
+	/* Enable link */
+	reg_write(ohci, OHCI1394_HCControlSet, OHCI1394_HCControl_linkEnable);
+
+	/* If anything is connected to a port, make sure it is enabled */
+	num_ports = get_phy_reg(ohci, 2) & 0xf;
+	for (i = 0; i < num_ports; i++) {
+		unsigned int status;
+
+		set_phy_reg(ohci, 7, i);
+		status = get_phy_reg(ohci, 8);
+
+		if (status & 0x20)
+			set_phy_reg(ohci, 8, status & ~1);
+	}
+}
+
+/**
+ * init_ohci1394_wait_for_busresets - wait until bus resets are completed
+ *
+ * OHCI1394 initialization itself and any device going on- or offline
+ * and any cable issue cause a IEEE1394 bus reset. The OHCI1394 spec
+ * specifies that physical DMA is disabled on each bus reset and it
+ * has to be enabled after each bus reset when needed. We resort
+ * to polling here because on early boot, we have no interrupts.
+ */
+static inline void __init init_ohci1394_wait_for_busresets(struct ti_ohci *ohci)
+{
+	int i, events;
+
+	for (i=0; i < 9; i++) {
+		mdelay(200);
+		events = reg_read(ohci, OHCI1394_IntEventSet);
+		if (events & OHCI1394_busReset)
+			reg_write(ohci, OHCI1394_IntEventClear,
+					OHCI1394_busReset);
+	}
+}
+
+/**
+ * init_ohci1394_enable_physical_dma - Enable physical DMA for remote debugging
+ * This enables remote DMA access over IEEE1394 from every host for the low
+ * 4GB of address space. DMA accesses above 4GB are not available currently.
+ */
+static inline void __init init_ohci1394_enable_physical_dma(struct ti_ohci *hci)
+{
+	reg_write(hci, OHCI1394_PhyReqFilterHiSet, 0xffffffff);
+	reg_write(hci, OHCI1394_PhyReqFilterLoSet, 0xffffffff);
+	reg_write(hci, OHCI1394_PhyUpperBound, 0xffff0000);
+}
+
+/**
+ * init_ohci1394_reset_and_init_dma - init controller and enable DMA
+ * This initializes the given controller and enables physical DMA engine in it.
+ */
+static inline void __init init_ohci1394_reset_and_init_dma(struct ti_ohci *ohci)
+{
+	/* Start off with a soft reset, clears everything to a sane state. */
+	init_ohci1394_soft_reset(ohci);
+
+	/* Accessing some registers without LPS enabled may cause lock up */
+	reg_write(ohci, OHCI1394_HCControlSet, OHCI1394_HCControl_LPS);
+
+	/* Disable and clear interrupts */
+	reg_write(ohci, OHCI1394_IntEventClear, 0xffffffff);
+	reg_write(ohci, OHCI1394_IntMaskClear, 0xffffffff);
+
+	mdelay(50); /* Wait 50msec to make sure we have full link enabled */
+
+	init_ohci1394_initialize(ohci);
+	/*
+	 * The initialization causes at least one IEEE1394 bus reset. Enabling
+	 * physical DMA only works *after* *all* bus resets have calmed down:
+	 */
+	init_ohci1394_wait_for_busresets(ohci);
+
+	/* We had to wait and do this now if we want to debug early problems */
+	init_ohci1394_enable_physical_dma(ohci);
+}
+
+/**
+ * init_ohci1394_controller - Map the registers of the controller and init DMA
+ * This maps the registers of the specified controller and initializes it
+ */
+static inline void __init init_ohci1394_controller(int num, int slot, int func)
+{
+	unsigned long ohci_base;
+	struct ti_ohci ohci;
+
+	printk(KERN_INFO "init_ohci1394_dma: initializing OHCI-1394"
+			 " at %02x:%02x.%x\n", num, slot, func);
+
+	ohci_base = read_pci_config(num, slot, func, PCI_BASE_ADDRESS_0+(0<<2))
+						   & PCI_BASE_ADDRESS_MEM_MASK;
+
+	set_fixmap_nocache(FIX_OHCI1394_BASE, ohci_base);
+
+	ohci.registers = (void *)fix_to_virt(FIX_OHCI1394_BASE);
+
+	init_ohci1394_reset_and_init_dma(&ohci);
+}
+
+/**
+ * debug_init_ohci1394_dma - scan for OHCI1394 controllers and init DMA on them
+ * Scans the whole PCI space for OHCI1394 controllers and inits DMA on them
+ */
+void __init init_ohci1394_dma_on_all_controllers(void)
+{
+	int num, slot, func;
+
+	if (!early_pci_allowed())
+		return;
+
+	/* Poor man's PCI discovery, the only thing we can do at early boot */
+	for (num = 0; num < 32; num++) {
+		for (slot = 0; slot < 32; slot++) {
+			for (func = 0; func < 8; func++) {
+				u32 class = read_pci_config(num,slot,func,
+							PCI_CLASS_REVISION);
+				if ((class == 0xffffffff))
+					continue; /* No device at this func */
+
+				if (class>>8 != PCI_CLASS_SERIAL_FIREWIRE_OHCI)
+					continue; /* Not an OHCI-1394 device */
+
+				init_ohci1394_controller(num, slot, func);
+				break; /* Assume one controller per device */
+			}
+		}
+	}
+	printk(KERN_INFO "init_ohci1394_dma: finished initializing OHCI DMA\n");
+}
+
+/**
+ * setup_init_ohci1394_early - enables early OHCI1394 DMA initialization
+ */
+static int __init setup_ohci1394_dma(char *opt)
+{
+	if (!strcmp(opt, "early"))
+		init_ohci1394_dma_early = 1;
+	return 0;
+}
+
+/* passing ohci1394_dma=early on boot causes early OHCI1394 DMA initialization */
+early_param("ohci1394_dma", setup_ohci1394_dma);
diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index fde140fd6d95..a7404d50686b 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -104,6 +104,9 @@ enum fixed_addresses {
 			(__end_of_permanent_fixed_addresses & 511),
 	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
 	FIX_WP_TEST,
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+	FIX_OHCI1394_BASE,
+#endif
 	__end_of_fixed_addresses
 };
 
diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h
index 8f44782e5fe5..70ddb21e6458 100644
--- a/include/asm-x86/fixmap_64.h
+++ b/include/asm-x86/fixmap_64.h
@@ -44,6 +44,9 @@ enum fixed_addresses {
 	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
 	FIX_EFI_IO_MAP_LAST_PAGE,
 	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE+MAX_EFI_IO_PAGES-1,
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+	FIX_OHCI1394_BASE,
+#endif
 	__end_of_fixed_addresses
 };
 
diff --git a/include/linux/init_ohci1394_dma.h b/include/linux/init_ohci1394_dma.h
new file mode 100644
index 000000000000..3c03a4bba5e4
--- /dev/null
+++ b/include/linux/init_ohci1394_dma.h
@@ -0,0 +1,4 @@
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+extern int __initdata init_ohci1394_dma_early;
+extern void __init init_ohci1394_dma_on_all_controllers(void);
+#endif
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index aa56e631580d..89f4035b526c 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -586,5 +586,33 @@ config LATENCYTOP
 	  Enable this option if you want to use the LatencyTOP tool
 	  to find out which userspace is blocking on what kernel operations.
 
+config PROVIDE_OHCI1394_DMA_INIT
+	bool "Provide code for enabling DMA over FireWire early on boot"
+	depends on PCI && X86
+	help
+	  If you want to debug problems which hang or crash the kernel early
+	  on boot and the crashing machine has a FireWire port, you can use
+	  this feature to remotely access the memory of the crashed machine
+	  over FireWire. This employs remote DMA as part of the OHCI1394
+	  specification which is now the standard for FireWire controllers.
+
+	  With remote DMA, you can monitor the printk buffer remotely using
+	  firescope and access all memory below 4GB using fireproxy from gdb.
+	  Even controlling a kernel debugger is possible using remote DMA.
+
+	  Usage:
+
+	  If ohci1394_dma=early is used as boot parameter, it will initialize
+	  all OHCI1394 controllers which are found in the PCI config space.
+
+	  As all changes to the FireWire bus such as enabling and disabling
+	  devices cause a bus reset and thereby disable remote DMA for all
+	  devices, be sure to have the cable plugged and FireWire enabled on
+	  the debugging host before booting the debug target for debugging.
+
+	  This code (~1k) is freed after boot. By then, the firewire stack
+	  in charge of the OHCI-1394 controllers should be used instead.
+
+	  See Documentation/debugging-via-ohci1394.txt for more information.
 
 source "samples/Kconfig"
-- 
cgit v1.2.3


From fa28ba21cec24d3fa1279bcae7e5d5ff6224635a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:34:11 +0100
Subject: x86: defer cr3 reload when doing pud_clear()

PAE mode requires that we reload cr3 in order to guarantee that
changes to the pgd will be noticed by the processor.  This means that
in principle pud_clear needs to reload cr3 every time.  However,
because reloading cr3 implies a tlb flush, we want to avoid it where
possible.

pud_clear() is only used in a couple of places:
 - in free_pmd_range(), when pulling down a range of process address space, and
 - huge_pmd_unshare()

In both cases, the calling code will do a a tlb flush anyway, so
there's no need to do it within pud_clear().

In free_pmd_range(), the pud_clear is immediately followed by
pmd_free_tlb(); we can hook that to make the mmu_gather do an
unconditional full flush to make sure cr3 gets reloaded.

In huge_pmd_unshare, it is followed by flush_tlb_range, which always
results in a full cr3-reload tlb flush.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: William Irwin <wli@holomorphy.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/asm-x86/pgalloc_32.h     |  7 +++++++
 include/asm-x86/pgtable-3level.h | 21 +++++++++++++++------
 2 files changed, 22 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgalloc_32.h b/include/asm-x86/pgalloc_32.h
index 0caa37a9a25f..10c2b452e64c 100644
--- a/include/asm-x86/pgalloc_32.h
+++ b/include/asm-x86/pgalloc_32.h
@@ -74,6 +74,13 @@ static inline void pmd_free(pmd_t *pmd)
 
 static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 {
+	/* This is called just after the pmd has been detached from
+	   the pgd, which requires a full tlb flush to be recognized
+	   by the CPU.  Rather than incurring multiple tlb flushes
+	   while the address space is being pulled down, make the tlb
+	   gathering machinery do a full flush when we're done. */
+	tlb->fullmm = 1;
+
 	paravirt_release_pd(__pa(pmd) >> PAGE_SHIFT);
 	tlb_remove_page(tlb, virt_to_page(pmd));
 }
diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index ed4c6f0e57ec..a195c3e757b9 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -96,14 +96,23 @@ static inline void pud_clear(pud_t *pudp)
 	set_pud(pudp, __pud(0));
 
 	/*
-	 * Pentium-II erratum A13: in PAE mode we explicitly have to flush
-	 * the TLB via cr3 if the top-level pgd is changed...
+	 * In principle we need to do a cr3 reload here to make sure
+	 * the processor recognizes the changed pgd.  In practice, all
+	 * the places where pud_clear() gets called are followed by
+	 * full tlb flushes anyway, so we can defer the cost here.
 	 *
-	 * XXX I don't think we need to worry about this here, since
-	 * when clearing the pud, the calling code needs to flush the
-	 * tlb anyway.  But do it now for safety's sake. - jsgf
+	 * Specifically:
+	 *
+	 * mm/memory.c:free_pmd_range() - immediately after the
+	 * pud_clear() it does a pmd_free_tlb().  We change the
+	 * mmu_gather structure to do a full tlb flush (which has the
+	 * effect of reloading cr3) when the pagetable free is
+	 * complete.
+	 *
+	 * arch/x86/mm/hugetlbpage.c:huge_pmd_unshare() - the call to
+	 * this is followed by a flush_tlb_range, which on x86 does a
+	 * full tlb flush.
 	 */
-	write_cr3(read_cr3());
 }
 
 #define pud_page(pud) \
-- 
cgit v1.2.3


From e3ed910db221768f8fd6192b13373e17d61bcdf0 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 30 Jan 2008 13:34:11 +0100
Subject: x86: use the same pgd_list for PAE and 64-bit

Use a standard list threaded through page->lru for maintaining the pgd
list on PAE.  This is the same as 64-bit, and seems saner than using a
non-standard list via page->index.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/fault.c          | 10 +++-------
 arch/x86/mm/pageattr.c       |  2 +-
 arch/x86/mm/pgtable_32.c     | 19 +++++--------------
 include/asm-x86/pgtable.h    |  2 ++
 include/asm-x86/pgtable_32.h |  2 --
 include/asm-x86/pgtable_64.h |  3 ---
 6 files changed, 11 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 72547a7e32c6..e28cc5277b16 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -907,10 +907,8 @@ do_sigbus:
 	force_sig_info_fault(SIGBUS, BUS_ADRERR, address, tsk);
 }
 
-#ifdef CONFIG_X86_64
 DEFINE_SPINLOCK(pgd_lock);
 LIST_HEAD(pgd_list);
-#endif
 
 void vmalloc_sync_all(void)
 {
@@ -935,13 +933,11 @@ void vmalloc_sync_all(void)
 			struct page *page;
 
 			spin_lock_irqsave(&pgd_lock, flags);
-			for (page = pgd_list; page; page =
-					(struct page *)page->index)
+			list_for_each_entry(page, &pgd_list, lru) {
 				if (!vmalloc_sync_one(page_address(page),
-								address)) {
-					BUG_ON(page != pgd_list);
+						      address))
 					break;
-				}
+			}
 			spin_unlock_irqrestore(&pgd_lock, flags);
 			if (!page)
 				set_bit(pgd_index(address), insync);
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index ec07c1873d65..1cc6607eacb0 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -175,7 +175,7 @@ static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte)
 	if (!SHARED_KERNEL_PMD) {
 		struct page *page;
 
-		for (page = pgd_list; page; page = (struct page *)page->index) {
+		list_for_each_entry(page, &pgd_list, lru) {
 			pgd_t *pgd;
 			pud_t *pud;
 			pmd_t *pmd;
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 5ca3552474ae..2ae5999a795a 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -205,27 +205,18 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
  * vmalloc faults work because attached pagetables are never freed.
  * -- wli
  */
-DEFINE_SPINLOCK(pgd_lock);
-struct page *pgd_list;
-
 static inline void pgd_list_add(pgd_t *pgd)
 {
 	struct page *page = virt_to_page(pgd);
-	page->index = (unsigned long)pgd_list;
-	if (pgd_list)
-		set_page_private(pgd_list, (unsigned long)&page->index);
-	pgd_list = page;
-	set_page_private(page, (unsigned long)&pgd_list);
+
+	list_add(&page->lru, &pgd_list);
 }
 
 static inline void pgd_list_del(pgd_t *pgd)
 {
-	struct page *next, **pprev, *page = virt_to_page(pgd);
-	next = (struct page *)page->index;
-	pprev = (struct page **)page_private(page);
-	*pprev = next;
-	if (next)
-		set_page_private(next, (unsigned long)pprev);
+	struct page *page = virt_to_page(pgd);
+
+	list_del(&page->lru);
 }
 
 
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 269e7e29ea8e..cd2524f07452 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -131,6 +131,8 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
 #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 
+extern spinlock_t pgd_lock;
+extern struct list_head pgd_list;
 
 /*
  * The following only work if pte_present() is true.
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 7b61cb5989b0..21e70fbf1dae 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -27,8 +27,6 @@ struct vm_area_struct;
 
 extern pgd_t swapper_pg_dir[1024];
 extern struct kmem_cache *pmd_cache;
-extern spinlock_t pgd_lock;
-extern struct page *pgd_list;
 void check_pgt_cache(void);
 
 static inline void pgtable_cache_init(void) {}
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index ce1dbab75e39..6e615a103c2f 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -240,9 +240,6 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 #define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) })
 #define __swp_entry_to_pte(x)		((pte_t) { .pte = (x).val })
 
-extern spinlock_t pgd_lock;
-extern struct list_head pgd_list;
-
 extern int kern_addr_valid(unsigned long addr); 
 
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
-- 
cgit v1.2.3


From 7aa81cc04781b5b99a0647ec04533599d78cd219 Mon Sep 17 00:00:00 2001
From: Anthony Liguori <aliguori@us.ibm.com>
Date: Mon, 17 Sep 2007 14:57:50 -0500
Subject: KVM: Refactor hypercall infrastructure (v3)

This patch refactors the current hypercall infrastructure to better
support live migration and SMP.  It eliminates the hypercall page by
trapping the UD exception that would occur if you used the wrong hypercall
instruction for the underlying architecture and replacing it with the right
one lazily.

A fall-out of this patch is that the unhandled hypercalls no longer trap to
userspace.  There is very little reason though to use a hypercall to
communicate with userspace as PIO or MMIO can be used.  There is no code
in tree that uses userspace hypercalls.

[avi: fix #ud injection on vmx]

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/kvm.h         |   8 +--
 drivers/kvm/kvm_main.c    | 156 ++++++++++++++-------------------------------
 drivers/kvm/svm.c         |  19 +++++-
 drivers/kvm/vmx.c         |  29 ++++++++-
 drivers/kvm/x86_emulate.c |  11 +++-
 include/linux/kvm_para.h  | 159 ++++++++++++++++++++++++++++------------------
 6 files changed, 199 insertions(+), 183 deletions(-)

(limited to 'include')

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 3b0bc4bda5f2..da9c3aa1c08c 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -46,6 +46,7 @@
 #define KVM_MAX_CPUID_ENTRIES 40
 
 #define DE_VECTOR 0
+#define UD_VECTOR 6
 #define NM_VECTOR 7
 #define DF_VECTOR 8
 #define TS_VECTOR 10
@@ -317,9 +318,6 @@ struct kvm_vcpu {
 	unsigned long cr0;
 	unsigned long cr2;
 	unsigned long cr3;
-	gpa_t para_state_gpa;
-	struct page *para_state_page;
-	gpa_t hypercall_gpa;
 	unsigned long cr4;
 	unsigned long cr8;
 	u64 pdptrs[4]; /* pae */
@@ -622,7 +620,9 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
 int kvm_mmu_load(struct kvm_vcpu *vcpu);
 void kvm_mmu_unload(struct kvm_vcpu *vcpu);
 
-int kvm_hypercall(struct kvm_vcpu *vcpu, struct kvm_run *run);
+int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);
+
+int kvm_fix_hypercall(struct kvm_vcpu *vcpu);
 
 static inline void kvm_guest_enter(void)
 {
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index c0f372f1d761..1c662f63b7a9 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -39,6 +39,7 @@
 #include <linux/smp.h>
 #include <linux/anon_inodes.h>
 #include <linux/profile.h>
+#include <linux/kvm_para.h>
 
 #include <asm/processor.h>
 #include <asm/msr.h>
@@ -1362,51 +1363,61 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu)
 }
 EXPORT_SYMBOL_GPL(kvm_emulate_halt);
 
-int kvm_hypercall(struct kvm_vcpu *vcpu, struct kvm_run *run)
+int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
 {
-	unsigned long nr, a0, a1, a2, a3, a4, a5, ret;
+	unsigned long nr, a0, a1, a2, a3, ret;
 
 	kvm_x86_ops->cache_regs(vcpu);
-	ret = -KVM_EINVAL;
-#ifdef CONFIG_X86_64
-	if (is_long_mode(vcpu)) {
-		nr = vcpu->regs[VCPU_REGS_RAX];
-		a0 = vcpu->regs[VCPU_REGS_RDI];
-		a1 = vcpu->regs[VCPU_REGS_RSI];
-		a2 = vcpu->regs[VCPU_REGS_RDX];
-		a3 = vcpu->regs[VCPU_REGS_RCX];
-		a4 = vcpu->regs[VCPU_REGS_R8];
-		a5 = vcpu->regs[VCPU_REGS_R9];
-	} else
-#endif
-	{
-		nr = vcpu->regs[VCPU_REGS_RBX] & -1u;
-		a0 = vcpu->regs[VCPU_REGS_RAX] & -1u;
-		a1 = vcpu->regs[VCPU_REGS_RCX] & -1u;
-		a2 = vcpu->regs[VCPU_REGS_RDX] & -1u;
-		a3 = vcpu->regs[VCPU_REGS_RSI] & -1u;
-		a4 = vcpu->regs[VCPU_REGS_RDI] & -1u;
-		a5 = vcpu->regs[VCPU_REGS_RBP] & -1u;
+
+	nr = vcpu->regs[VCPU_REGS_RAX];
+	a0 = vcpu->regs[VCPU_REGS_RBX];
+	a1 = vcpu->regs[VCPU_REGS_RCX];
+	a2 = vcpu->regs[VCPU_REGS_RDX];
+	a3 = vcpu->regs[VCPU_REGS_RSI];
+
+	if (!is_long_mode(vcpu)) {
+		nr &= 0xFFFFFFFF;
+		a0 &= 0xFFFFFFFF;
+		a1 &= 0xFFFFFFFF;
+		a2 &= 0xFFFFFFFF;
+		a3 &= 0xFFFFFFFF;
 	}
+
 	switch (nr) {
 	default:
-		run->hypercall.nr = nr;
-		run->hypercall.args[0] = a0;
-		run->hypercall.args[1] = a1;
-		run->hypercall.args[2] = a2;
-		run->hypercall.args[3] = a3;
-		run->hypercall.args[4] = a4;
-		run->hypercall.args[5] = a5;
-		run->hypercall.ret = ret;
-		run->hypercall.longmode = is_long_mode(vcpu);
-		kvm_x86_ops->decache_regs(vcpu);
-		return 0;
+		ret = -KVM_ENOSYS;
+		break;
 	}
 	vcpu->regs[VCPU_REGS_RAX] = ret;
 	kvm_x86_ops->decache_regs(vcpu);
-	return 1;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_hypercall);
+
+int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
+{
+	char instruction[3];
+	int ret = 0;
+
+	mutex_lock(&vcpu->kvm->lock);
+
+	/*
+	 * Blow out the MMU to ensure that no other VCPU has an active mapping
+	 * to ensure that the updated hypercall appears atomically across all
+	 * VCPUs.
+	 */
+	kvm_mmu_zap_all(vcpu->kvm);
+
+	kvm_x86_ops->cache_regs(vcpu);
+	kvm_x86_ops->patch_hypercall(vcpu, instruction);
+	if (emulator_write_emulated(vcpu->rip, instruction, 3, vcpu)
+	    != X86EMUL_CONTINUE)
+		ret = -EFAULT;
+
+	mutex_unlock(&vcpu->kvm->lock);
+
+	return ret;
 }
-EXPORT_SYMBOL_GPL(kvm_hypercall);
 
 static u64 mk_cr_64(u64 curr_cr, u32 new_val)
 {
@@ -1474,75 +1485,6 @@ void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
 	}
 }
 
-/*
- * Register the para guest with the host:
- */
-static int vcpu_register_para(struct kvm_vcpu *vcpu, gpa_t para_state_gpa)
-{
-	struct kvm_vcpu_para_state *para_state;
-	hpa_t para_state_hpa, hypercall_hpa;
-	struct page *para_state_page;
-	unsigned char *hypercall;
-	gpa_t hypercall_gpa;
-
-	printk(KERN_DEBUG "kvm: guest trying to enter paravirtual mode\n");
-	printk(KERN_DEBUG ".... para_state_gpa: %08Lx\n", para_state_gpa);
-
-	/*
-	 * Needs to be page aligned:
-	 */
-	if (para_state_gpa != PAGE_ALIGN(para_state_gpa))
-		goto err_gp;
-
-	para_state_hpa = gpa_to_hpa(vcpu, para_state_gpa);
-	printk(KERN_DEBUG ".... para_state_hpa: %08Lx\n", para_state_hpa);
-	if (is_error_hpa(para_state_hpa))
-		goto err_gp;
-
-	mark_page_dirty(vcpu->kvm, para_state_gpa >> PAGE_SHIFT);
-	para_state_page = pfn_to_page(para_state_hpa >> PAGE_SHIFT);
-	para_state = kmap(para_state_page);
-
-	printk(KERN_DEBUG "....  guest version: %d\n", para_state->guest_version);
-	printk(KERN_DEBUG "....           size: %d\n", para_state->size);
-
-	para_state->host_version = KVM_PARA_API_VERSION;
-	/*
-	 * We cannot support guests that try to register themselves
-	 * with a newer API version than the host supports:
-	 */
-	if (para_state->guest_version > KVM_PARA_API_VERSION) {
-		para_state->ret = -KVM_EINVAL;
-		goto err_kunmap_skip;
-	}
-
-	hypercall_gpa = para_state->hypercall_gpa;
-	hypercall_hpa = gpa_to_hpa(vcpu, hypercall_gpa);
-	printk(KERN_DEBUG ".... hypercall_hpa: %08Lx\n", hypercall_hpa);
-	if (is_error_hpa(hypercall_hpa)) {
-		para_state->ret = -KVM_EINVAL;
-		goto err_kunmap_skip;
-	}
-
-	printk(KERN_DEBUG "kvm: para guest successfully registered.\n");
-	vcpu->para_state_page = para_state_page;
-	vcpu->para_state_gpa = para_state_gpa;
-	vcpu->hypercall_gpa = hypercall_gpa;
-
-	mark_page_dirty(vcpu->kvm, hypercall_gpa >> PAGE_SHIFT);
-	hypercall = kmap_atomic(pfn_to_page(hypercall_hpa >> PAGE_SHIFT),
-				KM_USER1) + (hypercall_hpa & ~PAGE_MASK);
-	kvm_x86_ops->patch_hypercall(vcpu, hypercall);
-	kunmap_atomic(hypercall, KM_USER1);
-
-	para_state->ret = 0;
-err_kunmap_skip:
-	kunmap(para_state_page);
-	return 0;
-err_gp:
-	return 1;
-}
-
 int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 {
 	u64 data;
@@ -1656,12 +1598,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 	case MSR_IA32_MISC_ENABLE:
 		vcpu->ia32_misc_enable_msr = data;
 		break;
-	/*
-	 * This is the 'probe whether the host is KVM' logic:
-	 */
-	case MSR_KVM_API_MAGIC:
-		return vcpu_register_para(vcpu, data);
-
 	default:
 		pr_unimpl(vcpu, "unhandled wrmsr: 0x%x\n", msr);
 		return 1;
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index ced4ac1955db..794d95416f7b 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -476,7 +476,8 @@ static void init_vmcb(struct vmcb *vmcb)
 					INTERCEPT_DR5_MASK |
 					INTERCEPT_DR7_MASK;
 
-	control->intercept_exceptions = 1 << PF_VECTOR;
+	control->intercept_exceptions = (1 << PF_VECTOR) |
+					(1 << UD_VECTOR);
 
 
 	control->intercept = 	(1ULL << INTERCEPT_INTR) |
@@ -979,6 +980,17 @@ static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 	return 0;
 }
 
+static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	int er;
+
+	er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0);
+	if (er != EMULATE_DONE)
+		inject_ud(&svm->vcpu);
+
+	return 1;
+}
+
 static int nm_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
@@ -1045,7 +1057,8 @@ static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	svm->next_rip = svm->vmcb->save.rip + 3;
 	skip_emulated_instruction(&svm->vcpu);
-	return kvm_hypercall(&svm->vcpu, kvm_run);
+	kvm_emulate_hypercall(&svm->vcpu);
+	return 1;
 }
 
 static int invalid_op_interception(struct vcpu_svm *svm,
@@ -1241,6 +1254,7 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
 	[SVM_EXIT_WRITE_DR3]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR5]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR7]			= emulate_on_interception,
+	[SVM_EXIT_EXCP_BASE + UD_VECTOR]	= ud_interception,
 	[SVM_EXIT_EXCP_BASE + PF_VECTOR] 	= pf_interception,
 	[SVM_EXIT_EXCP_BASE + NM_VECTOR] 	= nm_interception,
 	[SVM_EXIT_INTR] 			= nop_on_interception,
@@ -1675,7 +1689,6 @@ svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
 	hypercall[0] = 0x0f;
 	hypercall[1] = 0x01;
 	hypercall[2] = 0xd9;
-	hypercall[3] = 0xc3;
 }
 
 static void svm_check_processor_compat(void *rtn)
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 5b397b6c9f93..47c827d3007c 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -164,6 +164,13 @@ static inline int is_no_device(u32 intr_info)
 		(INTR_TYPE_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK);
 }
 
+static inline int is_invalid_opcode(u32 intr_info)
+{
+	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
+			     INTR_INFO_VALID_MASK)) ==
+		(INTR_TYPE_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK);
+}
+
 static inline int is_external_interrupt(u32 intr_info)
 {
 	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VALID_MASK))
@@ -315,7 +322,7 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu)
 {
 	u32 eb;
 
-	eb = 1u << PF_VECTOR;
+	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
 	if (!vcpu->fpu_active)
 		eb |= 1u << NM_VECTOR;
 	if (vcpu->guest_debug.enabled)
@@ -560,6 +567,14 @@ static void vmx_inject_gp(struct kvm_vcpu *vcpu, unsigned error_code)
 		     INTR_INFO_VALID_MASK);
 }
 
+static void vmx_inject_ud(struct kvm_vcpu *vcpu)
+{
+	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
+		     UD_VECTOR |
+		     INTR_TYPE_EXCEPTION |
+		     INTR_INFO_VALID_MASK);
+}
+
 /*
  * Swap MSR entry in host/guest MSR entry array.
  */
@@ -1771,6 +1786,14 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		return 1;
 	}
 
+	if (is_invalid_opcode(intr_info)) {
+		er = emulate_instruction(vcpu, kvm_run, 0, 0);
+		if (er != EMULATE_DONE)
+			vmx_inject_ud(vcpu);
+
+		return 1;
+	}
+
 	error_code = 0;
 	rip = vmcs_readl(GUEST_RIP);
 	if (intr_info & INTR_INFO_DELIEVER_CODE_MASK)
@@ -1873,7 +1896,6 @@ vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
 	hypercall[0] = 0x0f;
 	hypercall[1] = 0x01;
 	hypercall[2] = 0xc1;
-	hypercall[3] = 0xc3;
 }
 
 static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
@@ -2059,7 +2081,8 @@ static int handle_halt(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	skip_emulated_instruction(vcpu);
-	return kvm_hypercall(vcpu, kvm_run);
+	kvm_emulate_hypercall(vcpu);
+	return 1;
 }
 
 /*
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index 84af9cc737fa..f12bc2c74040 100644
--- a/drivers/kvm/x86_emulate.c
+++ b/drivers/kvm/x86_emulate.c
@@ -1384,7 +1384,11 @@ twobyte_insn:
 			if (modrm_mod != 3 || modrm_rm != 1)
 				goto cannot_emulate;
 
-			/* nop */
+			rc = kvm_fix_hypercall(ctxt->vcpu);
+			if (rc)
+				goto done;
+
+			kvm_emulate_hypercall(ctxt->vcpu);
 			break;
 		case 2: /* lgdt */
 			rc = read_descriptor(ctxt, ops, src.ptr,
@@ -1395,7 +1399,10 @@ twobyte_insn:
 			break;
 		case 3: /* lidt/vmmcall */
 			if (modrm_mod == 3 && modrm_rm == 1) {
-				/* nop */
+				rc = kvm_fix_hypercall(ctxt->vcpu);
+				if (rc)
+					goto done;
+				kvm_emulate_hypercall(ctxt->vcpu);
 			} else {
 				rc = read_descriptor(ctxt, ops, src.ptr,
 						     &size, &address,
diff --git a/include/linux/kvm_para.h b/include/linux/kvm_para.h
index 3b292565a693..cc5dfb433706 100644
--- a/include/linux/kvm_para.h
+++ b/include/linux/kvm_para.h
@@ -1,73 +1,110 @@
 #ifndef __LINUX_KVM_PARA_H
 #define __LINUX_KVM_PARA_H
 
-/*
- * Guest OS interface for KVM paravirtualization
- *
- * Note: this interface is totally experimental, and is certain to change
- *       as we make progress.
+/* This CPUID returns the signature 'KVMKVMKVM' in ebx, ecx, and edx.  It
+ * should be used to determine that a VM is running under KVM.
  */
+#define KVM_CPUID_SIGNATURE	0x40000000
 
-/*
- * Per-VCPU descriptor area shared between guest and host. Writable to
- * both guest and host. Registered with the host by the guest when
- * a guest acknowledges paravirtual mode.
- *
- * NOTE: all addresses are guest-physical addresses (gpa), to make it
- * easier for the hypervisor to map between the various addresses.
- */
-struct kvm_vcpu_para_state {
-	/*
-	 * API version information for compatibility. If there's any support
-	 * mismatch (too old host trying to execute too new guest) then
-	 * the host will deny entry into paravirtual mode. Any other
-	 * combination (new host + old guest and new host + new guest)
-	 * is supposed to work - new host versions will support all old
-	 * guest API versions.
-	 */
-	u32 guest_version;
-	u32 host_version;
-	u32 size;
-	u32 ret;
-
-	/*
-	 * The address of the vm exit instruction (VMCALL or VMMCALL),
-	 * which the host will patch according to the CPU model the
-	 * VM runs on:
-	 */
-	u64 hypercall_gpa;
-
-} __attribute__ ((aligned(PAGE_SIZE)));
-
-#define KVM_PARA_API_VERSION 1
-
-/*
- * This is used for an RDMSR's ECX parameter to probe for a KVM host.
- * Hopefully no CPU vendor will use up this number. This is placed well
- * out of way of the typical space occupied by CPU vendors' MSR indices,
- * and we think (or at least hope) it wont be occupied in the future
- * either.
+/* This CPUID returns a feature bitmap in eax.  Before enabling a particular
+ * paravirtualization, the appropriate feature bit should be checked.
  */
-#define MSR_KVM_API_MAGIC 0x87655678
+#define KVM_CPUID_FEATURES	0x40000001
 
-#define KVM_EINVAL 1
+/* Return values for hypercalls */
+#define KVM_ENOSYS		1000
 
-/*
- * Hypercall calling convention:
- *
- * Each hypercall may have 0-6 parameters.
- *
- * 64-bit hypercall index is in RAX, goes from 0 to __NR_hypercalls-1
- *
- * 64-bit parameters 1-6 are in the standard gcc x86_64 calling convention
- * order: RDI, RSI, RDX, RCX, R8, R9.
- *
- * 32-bit index is EBX, parameters are: EAX, ECX, EDX, ESI, EDI, EBP.
- * (the first 3 are according to the gcc regparm calling convention)
+#ifdef __KERNEL__
+#include <asm/processor.h>
+
+/* This instruction is vmcall.  On non-VT architectures, it will generate a
+ * trap that we will then rewrite to the appropriate instruction.
+ */
+#define KVM_HYPERCALL ".byte 0x0f,0x01,0xc1"
+
+/* For KVM hypercalls, a three-byte sequence of either the vmrun or the vmmrun
+ * instruction.  The hypervisor may replace it with something else but only the
+ * instructions are guaranteed to be supported.
  *
- * No registers are clobbered by the hypercall, except that the
- * return value is in RAX.
+ * Up to four arguments may be passed in rbx, rcx, rdx, and rsi respectively.
+ * The hypercall number should be placed in rax and the return value will be
+ * placed in rax.  No other registers will be clobbered unless explicited
+ * noted by the particular hypercall.
  */
-#define __NR_hypercalls			0
+
+static inline long kvm_hypercall0(unsigned int nr)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr));
+	return ret;
+}
+
+static inline long kvm_hypercall1(unsigned int nr, unsigned long p1)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1));
+	return ret;
+}
+
+static inline long kvm_hypercall2(unsigned int nr, unsigned long p1,
+				  unsigned long p2)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2));
+	return ret;
+}
+
+static inline long kvm_hypercall3(unsigned int nr, unsigned long p1,
+				  unsigned long p2, unsigned long p3)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3));
+	return ret;
+}
+
+static inline long kvm_hypercall4(unsigned int nr, unsigned long p1,
+				  unsigned long p2, unsigned long p3,
+				  unsigned long p4)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3), "S"(p4));
+	return ret;
+}
+
+static inline int kvm_para_available(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+	char signature[13];
+
+	cpuid(KVM_CPUID_SIGNATURE, &eax, &ebx, &ecx, &edx);
+	memcpy(signature + 0, &ebx, 4);
+	memcpy(signature + 4, &ecx, 4);
+	memcpy(signature + 8, &edx, 4);
+	signature[12] = 0;
+
+	if (strcmp(signature, "KVMKVMKVM") == 0)
+		return 1;
+
+	return 0;
+}
+
+static inline int kvm_para_has_feature(unsigned int feature)
+{
+	if (cpuid_eax(KVM_CPUID_FEATURES) & (1UL << feature))
+		return 1;
+	return 0;
+}
+
+#endif
 
 #endif
-- 
cgit v1.2.3


From 82ce2c96831f049a37118733ced5c8f7c8848102 Mon Sep 17 00:00:00 2001
From: Izik Eidus <izike@qumranet.com>
Date: Tue, 2 Oct 2007 18:52:55 +0200
Subject: KVM: Allow dynamic allocation of the mmu shadow cache size

The user is now able to set how many mmu pages will be allocated to the guest.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/kvm.h      |  7 ++++++-
 drivers/kvm/kvm_main.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/kvm/mmu.c      | 40 ++++++++++++++++++++++++++++++++++++++--
 include/linux/kvm.h    |  3 +++
 4 files changed, 94 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 1965438f18cc..9f10c373b74c 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -40,6 +40,8 @@
 #define KVM_MAX_VCPUS 4
 #define KVM_ALIAS_SLOTS 4
 #define KVM_MEMORY_SLOTS 8
+#define KVM_PERMILLE_MMU_PAGES 20
+#define KVM_MIN_ALLOC_MMU_PAGES 64
 #define KVM_NUM_MMU_PAGES 1024
 #define KVM_MIN_FREE_MMU_PAGES 5
 #define KVM_REFILL_PAGES 25
@@ -418,7 +420,9 @@ struct kvm {
 	 * Hash table of struct kvm_mmu_page.
 	 */
 	struct list_head active_mmu_pages;
-	int n_free_mmu_pages;
+	unsigned int n_free_mmu_pages;
+	unsigned int n_requested_mmu_pages;
+	unsigned int n_alloc_mmu_pages;
 	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
 	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
 	unsigned long rmap_overflow;
@@ -547,6 +551,7 @@ void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
 int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
 void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
 void kvm_mmu_zap_all(struct kvm *kvm);
+void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);
 
 hpa_t gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa);
 #define HPA_MSB ((sizeof(hpa_t) * 8) - 1)
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index cac328b8421c..2bb1f2f66efa 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -743,6 +743,24 @@ static int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
 	if (mem->slot >= kvm->nmemslots)
 		kvm->nmemslots = mem->slot + 1;
 
+	if (!kvm->n_requested_mmu_pages) {
+		unsigned int n_pages;
+
+		if (npages) {
+			n_pages = npages * KVM_PERMILLE_MMU_PAGES / 1000;
+			kvm_mmu_change_mmu_pages(kvm, kvm->n_alloc_mmu_pages +
+						 n_pages);
+		} else {
+			unsigned int nr_mmu_pages;
+
+			n_pages = old.npages * KVM_PERMILLE_MMU_PAGES / 1000;
+			nr_mmu_pages = kvm->n_alloc_mmu_pages - n_pages;
+			nr_mmu_pages = max(nr_mmu_pages,
+				        (unsigned int) KVM_MIN_ALLOC_MMU_PAGES);
+			kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
+		}
+	}
+
 	*memslot = new;
 
 	kvm_mmu_slot_remove_write_access(kvm, mem->slot);
@@ -760,6 +778,26 @@ out:
 	return r;
 }
 
+static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
+					  u32 kvm_nr_mmu_pages)
+{
+	if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+
+	kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages);
+	kvm->n_requested_mmu_pages = kvm_nr_mmu_pages;
+
+	mutex_unlock(&kvm->lock);
+	return 0;
+}
+
+static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm)
+{
+	return kvm->n_alloc_mmu_pages;
+}
+
 /*
  * Get (and clear) the dirty memory log for a memory slot.
  */
@@ -3071,6 +3109,14 @@ static long kvm_vm_ioctl(struct file *filp,
 			goto out;
 		break;
 	}
+	case KVM_SET_NR_MMU_PAGES:
+		r = kvm_vm_ioctl_set_nr_mmu_pages(kvm, arg);
+		if (r)
+			goto out;
+		break;
+	case KVM_GET_NR_MMU_PAGES:
+		r = kvm_vm_ioctl_get_nr_mmu_pages(kvm);
+		break;
 	case KVM_GET_DIRTY_LOG: {
 		struct kvm_dirty_log log;
 
@@ -3278,6 +3324,7 @@ static long kvm_dev_ioctl(struct file *filp,
 		switch (ext) {
 		case KVM_CAP_IRQCHIP:
 		case KVM_CAP_HLT:
+		case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
 			r = 1;
 			break;
 		default:
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 72757db15065..6cda1feb9a95 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
@@ -747,6 +747,40 @@ static void kvm_mmu_zap_page(struct kvm *kvm,
 	kvm_mmu_reset_last_pte_updated(kvm);
 }
 
+/*
+ * Changing the number of mmu pages allocated to the vm
+ * Note: if kvm_nr_mmu_pages is too small, you will get dead lock
+ */
+void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages)
+{
+	/*
+	 * If we set the number of mmu pages to be smaller be than the
+	 * number of actived pages , we must to free some mmu pages before we
+	 * change the value
+	 */
+
+	if ((kvm->n_alloc_mmu_pages - kvm->n_free_mmu_pages) >
+	    kvm_nr_mmu_pages) {
+		int n_used_mmu_pages = kvm->n_alloc_mmu_pages
+				       - kvm->n_free_mmu_pages;
+
+		while (n_used_mmu_pages > kvm_nr_mmu_pages) {
+			struct kvm_mmu_page *page;
+
+			page = container_of(kvm->active_mmu_pages.prev,
+					    struct kvm_mmu_page, link);
+			kvm_mmu_zap_page(kvm, page);
+			n_used_mmu_pages--;
+		}
+		kvm->n_free_mmu_pages = 0;
+	}
+	else
+		kvm->n_free_mmu_pages += kvm_nr_mmu_pages
+					 - kvm->n_alloc_mmu_pages;
+
+	kvm->n_alloc_mmu_pages = kvm_nr_mmu_pages;
+}
+
 static int kvm_mmu_unprotect_page(struct kvm_vcpu *vcpu, gfn_t gfn)
 {
 	unsigned index;
@@ -1297,8 +1331,10 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
 
 	ASSERT(vcpu);
 
-	vcpu->kvm->n_free_mmu_pages = KVM_NUM_MMU_PAGES;
-
+	if (vcpu->kvm->n_requested_mmu_pages)
+		vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_requested_mmu_pages;
+	else
+		vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_alloc_mmu_pages;
 	/*
 	 * When emulating 32-bit mode, cr3 is only 32 bits even on x86_64.
 	 * Therefore we need to allocate shadow page tables in the first
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 057a7f34ee36..d2fd973d81fe 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -347,11 +347,14 @@ struct kvm_signal_mask {
  */
 #define KVM_CAP_IRQCHIP	  0
 #define KVM_CAP_HLT	  1
+#define KVM_CAP_MMU_SHADOW_CACHE_CONTROL 2
 
 /*
  * ioctls for VM fds
  */
 #define KVM_SET_MEMORY_REGION     _IOW(KVMIO, 0x40, struct kvm_memory_region)
+#define KVM_SET_NR_MMU_PAGES      _IO(KVMIO, 0x44)
+#define KVM_GET_NR_MMU_PAGES      _IO(KVMIO, 0x45)
 /*
  * KVM_CREATE_VCPU receives as a parameter the vcpu slot, and returns
  * a vcpu fd.
-- 
cgit v1.2.3


From 6fc138d2278078990f597cb1f62fde9e5b458f96 Mon Sep 17 00:00:00 2001
From: Izik Eidus <izike@qumranet.com>
Date: Tue, 9 Oct 2007 19:20:39 +0200
Subject: KVM: Support assigning userspace memory to the guest

Instead of having the kernel allocate memory to the guest, let userspace
allocate it and pass the address to the kernel.

This is required for s390 support, but also enables features like memory
sharing and using hugetlbfs backed memory.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/kvm.h      |  1 +
 drivers/kvm/kvm_main.c | 81 +++++++++++++++++++++++++++++++++++++++++++-------
 include/linux/kvm.h    | 12 ++++++++
 3 files changed, 83 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index ec5b498945ae..3eaed4dafec7 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -408,6 +408,7 @@ struct kvm_memory_slot {
 	struct page **phys_mem;
 	unsigned long *rmap;
 	unsigned long *dirty_bitmap;
+	int user_alloc; /* user allocated memory */
 };
 
 struct kvm {
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index a1983d2d5b8f..22b143feb66d 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -40,6 +40,7 @@
 #include <linux/anon_inodes.h>
 #include <linux/profile.h>
 #include <linux/kvm_para.h>
+#include <linux/pagemap.h>
 
 #include <asm/processor.h>
 #include <asm/msr.h>
@@ -300,19 +301,40 @@ static struct kvm *kvm_create_vm(void)
 	return kvm;
 }
 
+static void kvm_free_userspace_physmem(struct kvm_memory_slot *free)
+{
+	int i;
+
+	for (i = 0; i < free->npages; ++i) {
+		if (free->phys_mem[i]) {
+			if (!PageReserved(free->phys_mem[i]))
+				SetPageDirty(free->phys_mem[i]);
+			page_cache_release(free->phys_mem[i]);
+		}
+	}
+}
+
+static void kvm_free_kernel_physmem(struct kvm_memory_slot *free)
+{
+	int i;
+
+	for (i = 0; i < free->npages; ++i)
+		if (free->phys_mem[i])
+			__free_page(free->phys_mem[i]);
+}
+
 /*
  * Free any memory in @free but not in @dont.
  */
 static void kvm_free_physmem_slot(struct kvm_memory_slot *free,
 				  struct kvm_memory_slot *dont)
 {
-	int i;
-
 	if (!dont || free->phys_mem != dont->phys_mem)
 		if (free->phys_mem) {
-			for (i = 0; i < free->npages; ++i)
-				if (free->phys_mem[i])
-					__free_page(free->phys_mem[i]);
+			if (free->user_alloc)
+				kvm_free_userspace_physmem(free);
+			else
+				kvm_free_kernel_physmem(free);
 			vfree(free->phys_mem);
 		}
 	if (!dont || free->rmap != dont->rmap)
@@ -652,7 +674,9 @@ EXPORT_SYMBOL_GPL(fx_init);
  * Discontiguous memory is allowed, mostly for framebuffers.
  */
 static int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
-					  struct kvm_memory_region *mem)
+					  struct
+					  kvm_userspace_memory_region *mem,
+					  int user_alloc)
 {
 	int r;
 	gfn_t base_gfn;
@@ -728,11 +752,27 @@ static int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
 
 		memset(new.phys_mem, 0, npages * sizeof(struct page *));
 		memset(new.rmap, 0, npages * sizeof(*new.rmap));
-		for (i = 0; i < npages; ++i) {
-			new.phys_mem[i] = alloc_page(GFP_HIGHUSER
-						     | __GFP_ZERO);
-			if (!new.phys_mem[i])
+		if (user_alloc) {
+			unsigned long pages_num;
+
+			new.user_alloc = 1;
+			down_read(&current->mm->mmap_sem);
+
+			pages_num = get_user_pages(current, current->mm,
+						   mem->userspace_addr,
+						   npages, 1, 1, new.phys_mem,
+						   NULL);
+
+			up_read(&current->mm->mmap_sem);
+			if (pages_num != npages)
 				goto out_unlock;
+		} else {
+			for (i = 0; i < npages; ++i) {
+				new.phys_mem[i] = alloc_page(GFP_HIGHUSER
+							     | __GFP_ZERO);
+				if (!new.phys_mem[i])
+					goto out_unlock;
+			}
 		}
 	}
 
@@ -3108,11 +3148,29 @@ static long kvm_vm_ioctl(struct file *filp,
 		break;
 	case KVM_SET_MEMORY_REGION: {
 		struct kvm_memory_region kvm_mem;
+		struct kvm_userspace_memory_region kvm_userspace_mem;
 
 		r = -EFAULT;
 		if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
 			goto out;
-		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_mem);
+		kvm_userspace_mem.slot = kvm_mem.slot;
+		kvm_userspace_mem.flags = kvm_mem.flags;
+		kvm_userspace_mem.guest_phys_addr = kvm_mem.guest_phys_addr;
+		kvm_userspace_mem.memory_size = kvm_mem.memory_size;
+		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 0);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_SET_USER_MEMORY_REGION: {
+		struct kvm_userspace_memory_region kvm_userspace_mem;
+
+		r = -EFAULT;
+		if (copy_from_user(&kvm_userspace_mem, argp,
+						sizeof kvm_userspace_mem))
+			goto out;
+
+		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 1);
 		if (r)
 			goto out;
 		break;
@@ -3332,6 +3390,7 @@ static long kvm_dev_ioctl(struct file *filp,
 		case KVM_CAP_IRQCHIP:
 		case KVM_CAP_HLT:
 		case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
+		case KVM_CAP_USER_MEMORY:
 			r = 1;
 			break;
 		default:
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index d2fd973d81fe..971f465e26fa 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -23,6 +23,15 @@ struct kvm_memory_region {
 	__u64 memory_size; /* bytes */
 };
 
+/* for KVM_SET_USER_MEMORY_REGION */
+struct kvm_userspace_memory_region {
+	__u32 slot;
+	__u32 flags;
+	__u64 guest_phys_addr;
+	__u64 memory_size; /* bytes */
+	__u64 userspace_addr; /* start of the userspace allocated memory */
+};
+
 /* for kvm_memory_region::flags */
 #define KVM_MEM_LOG_DIRTY_PAGES  1UL
 
@@ -348,6 +357,7 @@ struct kvm_signal_mask {
 #define KVM_CAP_IRQCHIP	  0
 #define KVM_CAP_HLT	  1
 #define KVM_CAP_MMU_SHADOW_CACHE_CONTROL 2
+#define KVM_CAP_USER_MEMORY 3
 
 /*
  * ioctls for VM fds
@@ -355,6 +365,8 @@ struct kvm_signal_mask {
 #define KVM_SET_MEMORY_REGION     _IOW(KVMIO, 0x40, struct kvm_memory_region)
 #define KVM_SET_NR_MMU_PAGES      _IO(KVMIO, 0x44)
 #define KVM_GET_NR_MMU_PAGES      _IO(KVMIO, 0x45)
+#define KVM_SET_USER_MEMORY_REGION _IOW(KVMIO, 0x46,\
+					struct kvm_userspace_memory_region)
 /*
  * KVM_CREATE_VCPU receives as a parameter the vcpu slot, and returns
  * a vcpu fd.
-- 
cgit v1.2.3


From 5f43238d036fb30e73563e81e42d9c6f1de5551a Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Thu, 11 Oct 2007 15:34:17 +0200
Subject: KVM: Per-architecture hypercall definitions

Currently kvm provides hypercalls only for x86* architectures. To
provide hypercall infrastructure for other kvm architectures I split
kvm_para.h into a generic header file and architecture specific
definitions.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm_para.h | 105 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/kvm_para.h   | 105 ++++++---------------------------------------
 2 files changed, 117 insertions(+), 93 deletions(-)
 create mode 100644 include/asm-x86/kvm_para.h

(limited to 'include')

diff --git a/include/asm-x86/kvm_para.h b/include/asm-x86/kvm_para.h
new file mode 100644
index 000000000000..c6f3fd8d8c53
--- /dev/null
+++ b/include/asm-x86/kvm_para.h
@@ -0,0 +1,105 @@
+#ifndef __X86_KVM_PARA_H
+#define __X86_KVM_PARA_H
+
+/* This CPUID returns the signature 'KVMKVMKVM' in ebx, ecx, and edx.  It
+ * should be used to determine that a VM is running under KVM.
+ */
+#define KVM_CPUID_SIGNATURE	0x40000000
+
+/* This CPUID returns a feature bitmap in eax.  Before enabling a particular
+ * paravirtualization, the appropriate feature bit should be checked.
+ */
+#define KVM_CPUID_FEATURES	0x40000001
+
+#ifdef __KERNEL__
+#include <asm/processor.h>
+
+/* This instruction is vmcall.  On non-VT architectures, it will generate a
+ * trap that we will then rewrite to the appropriate instruction.
+ */
+#define KVM_HYPERCALL ".byte 0x0f,0x01,0xc1"
+
+/* For KVM hypercalls, a three-byte sequence of either the vmrun or the vmmrun
+ * instruction.  The hypervisor may replace it with something else but only the
+ * instructions are guaranteed to be supported.
+ *
+ * Up to four arguments may be passed in rbx, rcx, rdx, and rsi respectively.
+ * The hypercall number should be placed in rax and the return value will be
+ * placed in rax.  No other registers will be clobbered unless explicited
+ * noted by the particular hypercall.
+ */
+
+static inline long kvm_hypercall0(unsigned int nr)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr));
+	return ret;
+}
+
+static inline long kvm_hypercall1(unsigned int nr, unsigned long p1)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1));
+	return ret;
+}
+
+static inline long kvm_hypercall2(unsigned int nr, unsigned long p1,
+				  unsigned long p2)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2));
+	return ret;
+}
+
+static inline long kvm_hypercall3(unsigned int nr, unsigned long p1,
+				  unsigned long p2, unsigned long p3)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3));
+	return ret;
+}
+
+static inline long kvm_hypercall4(unsigned int nr, unsigned long p1,
+				  unsigned long p2, unsigned long p3,
+				  unsigned long p4)
+{
+	long ret;
+	asm volatile(KVM_HYPERCALL
+		     : "=a"(ret)
+		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3), "S"(p4));
+	return ret;
+}
+
+static inline int kvm_para_available(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+	char signature[13];
+
+	cpuid(KVM_CPUID_SIGNATURE, &eax, &ebx, &ecx, &edx);
+	memcpy(signature + 0, &ebx, 4);
+	memcpy(signature + 4, &ecx, 4);
+	memcpy(signature + 8, &edx, 4);
+	signature[12] = 0;
+
+	if (strcmp(signature, "KVMKVMKVM") == 0)
+		return 1;
+
+	return 0;
+}
+
+static inline unsigned int kvm_arch_para_features(void)
+{
+	return cpuid_eax(KVM_CPUID_FEATURES);
+}
+
+#endif
+
+#endif
diff --git a/include/linux/kvm_para.h b/include/linux/kvm_para.h
index cc5dfb433706..e4db25ffdb52 100644
--- a/include/linux/kvm_para.h
+++ b/include/linux/kvm_para.h
@@ -1,110 +1,29 @@
 #ifndef __LINUX_KVM_PARA_H
 #define __LINUX_KVM_PARA_H
 
-/* This CPUID returns the signature 'KVMKVMKVM' in ebx, ecx, and edx.  It
- * should be used to determine that a VM is running under KVM.
+/*
+ * This header file provides a method for making a hypercall to the host
+ * Architectures should define:
+ * - kvm_hypercall0, kvm_hypercall1...
+ * - kvm_arch_para_features
+ * - kvm_para_available
  */
-#define KVM_CPUID_SIGNATURE	0x40000000
-
-/* This CPUID returns a feature bitmap in eax.  Before enabling a particular
- * paravirtualization, the appropriate feature bit should be checked.
- */
-#define KVM_CPUID_FEATURES	0x40000001
 
 /* Return values for hypercalls */
 #define KVM_ENOSYS		1000
 
 #ifdef __KERNEL__
-#include <asm/processor.h>
-
-/* This instruction is vmcall.  On non-VT architectures, it will generate a
- * trap that we will then rewrite to the appropriate instruction.
- */
-#define KVM_HYPERCALL ".byte 0x0f,0x01,0xc1"
-
-/* For KVM hypercalls, a three-byte sequence of either the vmrun or the vmmrun
- * instruction.  The hypervisor may replace it with something else but only the
- * instructions are guaranteed to be supported.
- *
- * Up to four arguments may be passed in rbx, rcx, rdx, and rsi respectively.
- * The hypercall number should be placed in rax and the return value will be
- * placed in rax.  No other registers will be clobbered unless explicited
- * noted by the particular hypercall.
+/*
+ * hypercalls use architecture specific
  */
-
-static inline long kvm_hypercall0(unsigned int nr)
-{
-	long ret;
-	asm volatile(KVM_HYPERCALL
-		     : "=a"(ret)
-		     : "a"(nr));
-	return ret;
-}
-
-static inline long kvm_hypercall1(unsigned int nr, unsigned long p1)
-{
-	long ret;
-	asm volatile(KVM_HYPERCALL
-		     : "=a"(ret)
-		     : "a"(nr), "b"(p1));
-	return ret;
-}
-
-static inline long kvm_hypercall2(unsigned int nr, unsigned long p1,
-				  unsigned long p2)
-{
-	long ret;
-	asm volatile(KVM_HYPERCALL
-		     : "=a"(ret)
-		     : "a"(nr), "b"(p1), "c"(p2));
-	return ret;
-}
-
-static inline long kvm_hypercall3(unsigned int nr, unsigned long p1,
-				  unsigned long p2, unsigned long p3)
-{
-	long ret;
-	asm volatile(KVM_HYPERCALL
-		     : "=a"(ret)
-		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3));
-	return ret;
-}
-
-static inline long kvm_hypercall4(unsigned int nr, unsigned long p1,
-				  unsigned long p2, unsigned long p3,
-				  unsigned long p4)
-{
-	long ret;
-	asm volatile(KVM_HYPERCALL
-		     : "=a"(ret)
-		     : "a"(nr), "b"(p1), "c"(p2), "d"(p3), "S"(p4));
-	return ret;
-}
-
-static inline int kvm_para_available(void)
-{
-	unsigned int eax, ebx, ecx, edx;
-	char signature[13];
-
-	cpuid(KVM_CPUID_SIGNATURE, &eax, &ebx, &ecx, &edx);
-	memcpy(signature + 0, &ebx, 4);
-	memcpy(signature + 4, &ecx, 4);
-	memcpy(signature + 8, &edx, 4);
-	signature[12] = 0;
-
-	if (strcmp(signature, "KVMKVMKVM") == 0)
-		return 1;
-
-	return 0;
-}
+#include <asm/kvm_para.h>
 
 static inline int kvm_para_has_feature(unsigned int feature)
 {
-	if (cpuid_eax(KVM_CPUID_FEATURES) & (1UL << feature))
+	if (kvm_arch_para_features() & (1UL << feature))
 		return 1;
 	return 0;
 }
+#endif /* __KERNEL__ */
+#endif /* __LINUX_KVM_PARA_H */
 
-#endif
-
-#endif
-- 
cgit v1.2.3


From cbc9402297b9a233981f74587786364cda21c771 Mon Sep 17 00:00:00 2001
From: Izik Eidus <izike@qumranet.com>
Date: Thu, 25 Oct 2007 00:29:55 +0200
Subject: KVM: Add ioctl to tss address from userspace,

Currently kvm has a wart in that it requires three extra pages for use
as a tss when emulating real mode on Intel.  This patch moves the allocation
internally, only requiring userspace to tell us where in the physical address
space we can place the tss.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/kvm.h      |  3 +++
 drivers/kvm/kvm_main.c | 16 ++++++++++++++++
 drivers/kvm/svm.c      |  7 +++++++
 drivers/kvm/vmx.c      | 27 +++++++++++++++++++++++++--
 include/linux/kvm.h    |  2 ++
 5 files changed, 53 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 82c3b13e5d0a..3d07d9b1b815 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -382,6 +382,7 @@ struct kvm {
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	int round_robin_prev_vcpu;
+	unsigned int tss_addr;
 };
 
 static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
@@ -467,6 +468,8 @@ struct kvm_x86_ops {
 	void (*inject_pending_irq)(struct kvm_vcpu *vcpu);
 	void (*inject_pending_vectors)(struct kvm_vcpu *vcpu,
 				       struct kvm_run *run);
+
+	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
 };
 
 extern struct kvm_x86_ops *kvm_x86_ops;
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index a044856bd02b..50fd040b9ea9 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -2661,6 +2661,16 @@ static int create_vcpu_fd(struct kvm_vcpu *vcpu)
 	return fd;
 }
 
+static int kvm_vm_ioctl_set_tss_addr(struct kvm *kvm, unsigned long addr)
+{
+	int ret;
+
+	if (addr > (unsigned int)(-3 * PAGE_SIZE))
+		return -1;
+	ret = kvm_x86_ops->set_tss_addr(kvm, addr);
+	return ret;
+}
+
 /*
  * Creates some virtual cpus.  Good luck creating more than one.
  */
@@ -2957,6 +2967,11 @@ static long kvm_vm_ioctl(struct file *filp,
 	int r = -EINVAL;
 
 	switch (ioctl) {
+	case KVM_SET_TSS_ADDR:
+		r = kvm_vm_ioctl_set_tss_addr(kvm, arg);
+		if (r < 0)
+			goto out;
+		break;
 	case KVM_CREATE_VCPU:
 		r = kvm_vm_ioctl_create_vcpu(kvm, arg);
 		if (r < 0)
@@ -3183,6 +3198,7 @@ static long kvm_dev_ioctl(struct file *filp,
 		case KVM_CAP_HLT:
 		case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
 		case KVM_CAP_USER_MEMORY:
+		case KVM_CAP_SET_TSS_ADDR:
 			r = 1;
 			break;
 		default:
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index 953c111544e1..beb1e684b68a 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -1439,6 +1439,11 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 		control->intercept &= ~(1ULL << INTERCEPT_VINTR);
 }
 
+static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
+{
+	return 0;
+}
+
 static void save_db_regs(unsigned long *db_regs)
 {
 	asm volatile ("mov %%dr0, %0" : "=r"(db_regs[0]));
@@ -1744,6 +1749,8 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.set_irq = svm_set_irq,
 	.inject_pending_irq = svm_intr_assist,
 	.inject_pending_vectors = do_interrupt_requests,
+
+	.set_tss_addr = svm_set_tss_addr,
 };
 
 static int __init svm_init(void)
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 768ea88beb45..46b29184a4d8 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -1105,8 +1105,12 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
 
 static gva_t rmode_tss_base(struct kvm *kvm)
 {
-	gfn_t base_gfn = kvm->memslots[0].base_gfn + kvm->memslots[0].npages - 3;
-	return base_gfn << PAGE_SHIFT;
+	if (!kvm->tss_addr) {
+		gfn_t base_gfn = kvm->memslots[0].base_gfn +
+				 kvm->memslots[0].npages - 3;
+		return base_gfn << PAGE_SHIFT;
+	}
+	return kvm->tss_addr;
 }
 
 static void fix_rmode_seg(int seg, struct kvm_save_segment *save)
@@ -1735,6 +1739,23 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
 }
 
+static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
+{
+	int ret;
+	struct kvm_userspace_memory_region tss_mem = {
+		.slot = 8,
+		.guest_phys_addr = addr,
+		.memory_size = PAGE_SIZE * 3,
+		.flags = 0,
+	};
+
+	ret = kvm_set_memory_region(kvm, &tss_mem, 0);
+	if (ret)
+		return ret;
+	kvm->tss_addr = addr;
+	return 0;
+}
+
 static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
 {
 	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
@@ -2543,6 +2564,8 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.set_irq = vmx_inject_irq,
 	.inject_pending_irq = vmx_intr_assist,
 	.inject_pending_vectors = do_interrupt_requests,
+
+	.set_tss_addr = vmx_set_tss_addr,
 };
 
 static int __init vmx_init(void)
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 971f465e26fa..71d33d62944f 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -358,6 +358,7 @@ struct kvm_signal_mask {
 #define KVM_CAP_HLT	  1
 #define KVM_CAP_MMU_SHADOW_CACHE_CONTROL 2
 #define KVM_CAP_USER_MEMORY 3
+#define KVM_CAP_SET_TSS_ADDR 4
 
 /*
  * ioctls for VM fds
@@ -367,6 +368,7 @@ struct kvm_signal_mask {
 #define KVM_GET_NR_MMU_PAGES      _IO(KVMIO, 0x45)
 #define KVM_SET_USER_MEMORY_REGION _IOW(KVMIO, 0x46,\
 					struct kvm_userspace_memory_region)
+#define KVM_SET_TSS_ADDR          _IO(KVMIO, 0x47)
 /*
  * KVM_CREATE_VCPU receives as a parameter the vcpu slot, and returns
  * a vcpu fd.
-- 
cgit v1.2.3


From f6a40e3bdf5fe0a7d7d7f2dbc5b10158fbdad968 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:31 -0600
Subject: KVM: Portability: Move kvm_memory_alias to asm/kvm.h

This patch moves sturct kvm_memory_alias from include/linux/kvm.h
to include/asm-x86/kvm.h. Also have include/linux/kvm.h include
include/asm/kvm.h.

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 20 ++++++++++++++++++++
 include/linux/kvm.h   |  8 +-------
 2 files changed, 21 insertions(+), 7 deletions(-)
 create mode 100644 include/asm-x86/kvm.h

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
new file mode 100644
index 000000000000..37cf8e995c83
--- /dev/null
+++ b/include/asm-x86/kvm.h
@@ -0,0 +1,20 @@
+#ifndef __LINUX_KVM_X86_H
+#define __LINUX_KVM_X86_H
+
+/*
+ * KVM x86 specific structures and definitions
+ *
+ */
+
+#include <asm/types.h>
+#include <linux/ioctl.h>
+
+struct kvm_memory_alias {
+	__u32 slot;  /* this has a different namespace than memory slots */
+	__u32 flags;
+	__u64 guest_phys_addr;
+	__u64 memory_size;
+	__u64 target_phys_addr;
+};
+
+#endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 71d33d62944f..d09dd5dd5493 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -9,6 +9,7 @@
 
 #include <asm/types.h>
 #include <linux/ioctl.h>
+#include <asm/kvm.h>
 
 #define KVM_API_VERSION 12
 
@@ -35,13 +36,6 @@ struct kvm_userspace_memory_region {
 /* for kvm_memory_region::flags */
 #define KVM_MEM_LOG_DIRTY_PAGES  1UL
 
-struct kvm_memory_alias {
-	__u32 slot;  /* this has a different namespace than memory slots */
-	__u32 flags;
-	__u64 guest_phys_addr;
-	__u64 memory_size;
-	__u64 target_phys_addr;
-};
 
 /* for KVM_IRQ_LINE */
 struct kvm_irq_level {
-- 
cgit v1.2.3


From da1386a5bc43fa0faf17089a68c2991cf1d526c8 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:32 -0600
Subject: KVM: Portability: Move x86 pic strutctures

This patch moves structures:
	kvm_pic_state
	kvm_ioapic_state

to inclue/asm-x86/kvm.h.

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/kvm.h   | 48 ------------------------------------------------
 2 files changed, 49 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 37cf8e995c83..80752bccc756 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -17,4 +17,53 @@ struct kvm_memory_alias {
 	__u64 target_phys_addr;
 };
 
+/* for KVM_GET_IRQCHIP and KVM_SET_IRQCHIP */
+struct kvm_pic_state {
+	__u8 last_irr;	/* edge detection */
+	__u8 irr;		/* interrupt request register */
+	__u8 imr;		/* interrupt mask register */
+	__u8 isr;		/* interrupt service register */
+	__u8 priority_add;	/* highest irq priority */
+	__u8 irq_base;
+	__u8 read_reg_select;
+	__u8 poll;
+	__u8 special_mask;
+	__u8 init_state;
+	__u8 auto_eoi;
+	__u8 rotate_on_auto_eoi;
+	__u8 special_fully_nested_mode;
+	__u8 init4;		/* true if 4 byte init */
+	__u8 elcr;		/* PIIX edge/trigger selection */
+	__u8 elcr_mask;
+};
+
+#define KVM_IOAPIC_NUM_PINS  24
+struct kvm_ioapic_state {
+	__u64 base_address;
+	__u32 ioregsel;
+	__u32 id;
+	__u32 irr;
+	__u32 pad;
+	union {
+		__u64 bits;
+		struct {
+			__u8 vector;
+			__u8 delivery_mode:3;
+			__u8 dest_mode:1;
+			__u8 delivery_status:1;
+			__u8 polarity:1;
+			__u8 remote_irr:1;
+			__u8 trig_mode:1;
+			__u8 mask:1;
+			__u8 reserve:7;
+			__u8 reserved[4];
+			__u8 dest_id;
+		} fields;
+	} redirtbl[KVM_IOAPIC_NUM_PINS];
+};
+
+#define KVM_IRQCHIP_PIC_MASTER   0
+#define KVM_IRQCHIP_PIC_SLAVE    1
+#define KVM_IRQCHIP_IOAPIC       2
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index d09dd5dd5493..1779c3d3bdc9 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -48,54 +48,6 @@ struct kvm_irq_level {
 	__u32 level;
 };
 
-/* for KVM_GET_IRQCHIP and KVM_SET_IRQCHIP */
-struct kvm_pic_state {
-	__u8 last_irr;	/* edge detection */
-	__u8 irr;		/* interrupt request register */
-	__u8 imr;		/* interrupt mask register */
-	__u8 isr;		/* interrupt service register */
-	__u8 priority_add;	/* highest irq priority */
-	__u8 irq_base;
-	__u8 read_reg_select;
-	__u8 poll;
-	__u8 special_mask;
-	__u8 init_state;
-	__u8 auto_eoi;
-	__u8 rotate_on_auto_eoi;
-	__u8 special_fully_nested_mode;
-	__u8 init4;		/* true if 4 byte init */
-	__u8 elcr;		/* PIIX edge/trigger selection */
-	__u8 elcr_mask;
-};
-
-#define KVM_IOAPIC_NUM_PINS  24
-struct kvm_ioapic_state {
-	__u64 base_address;
-	__u32 ioregsel;
-	__u32 id;
-	__u32 irr;
-	__u32 pad;
-	union {
-		__u64 bits;
-		struct {
-			__u8 vector;
-			__u8 delivery_mode:3;
-			__u8 dest_mode:1;
-			__u8 delivery_status:1;
-			__u8 polarity:1;
-			__u8 remote_irr:1;
-			__u8 trig_mode:1;
-			__u8 mask:1;
-			__u8 reserve:7;
-			__u8 reserved[4];
-			__u8 dest_id;
-		} fields;
-	} redirtbl[KVM_IOAPIC_NUM_PINS];
-};
-
-#define KVM_IRQCHIP_PIC_MASTER   0
-#define KVM_IRQCHIP_PIC_SLAVE    1
-#define KVM_IRQCHIP_IOAPIC       2
 
 struct kvm_irqchip {
 	__u32 chip_id;
-- 
cgit v1.2.3


From 19d30b164417e7580d7f8c495a03137e1a45d4a7 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:33 -0600
Subject: KVM: Portability: Move kvm_regs to <asm/kvm.h>

This patch moves structure kvm_regs to include/asm-x86/kvm.h.
Each architecture will need to create there own version of this
structure.

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 10 ++++++++++
 include/linux/kvm.h   |  9 ---------
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 80752bccc756..c83a2ff35bf1 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -66,4 +66,14 @@ struct kvm_ioapic_state {
 #define KVM_IRQCHIP_PIC_SLAVE    1
 #define KVM_IRQCHIP_IOAPIC       2
 
+/* for KVM_GET_REGS and KVM_SET_REGS */
+struct kvm_regs {
+	/* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
+	__u64 rax, rbx, rcx, rdx;
+	__u64 rsi, rdi, rsp, rbp;
+	__u64 r8,  r9,  r10, r11;
+	__u64 r12, r13, r14, r15;
+	__u64 rip, rflags;
+};
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 1779c3d3bdc9..0d83efc9462e 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -134,15 +134,6 @@ struct kvm_run {
 	};
 };
 
-/* for KVM_GET_REGS and KVM_SET_REGS */
-struct kvm_regs {
-	/* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
-	__u64 rax, rbx, rcx, rdx;
-	__u64 rsi, rdi, rsp, rbp;
-	__u64 r8,  r9,  r10, r11;
-	__u64 r12, r13, r14, r15;
-	__u64 rip, rflags;
-};
 
 /* for KVM_GET_FPU and KVM_SET_FPU */
 struct kvm_fpu {
-- 
cgit v1.2.3


From d9ecf9281069c4c8dd429e1eb61cb7e0f3dda6b4 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:34 -0600
Subject: KVM: Portability: Move structure lapic_state to <asm/kvm.h>

This patch moves structure lapic_state from include/linux/kvm.h
to include/asm-x86/kvm.h

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 6 ++++++
 include/linux/kvm.h   | 5 -----
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index c83a2ff35bf1..a2c65b5c4a85 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -76,4 +76,10 @@ struct kvm_regs {
 	__u64 rip, rflags;
 };
 
+/* for KVM_GET_LAPIC and KVM_SET_LAPIC */
+#define KVM_APIC_REG_SIZE 0x400
+struct kvm_lapic_state {
+	char regs[KVM_APIC_REG_SIZE];
+};
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 0d83efc9462e..280ec0d85b4c 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -150,11 +150,6 @@ struct kvm_fpu {
 	__u32 pad2;
 };
 
-/* for KVM_GET_LAPIC and KVM_SET_LAPIC */
-#define KVM_APIC_REG_SIZE 0x400
-struct kvm_lapic_state {
-	char regs[KVM_APIC_REG_SIZE];
-};
 
 struct kvm_segment {
 	__u64 base;
-- 
cgit v1.2.3


From 3a56b20104f51bd6caf07e49fa1e3c09bee222e5 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:35 -0600
Subject: KVM: Portability: Move kvm_segment & kvm_dtable structure to 
 <asm/kvm.h>

This patch moves structures:
	kvm_segment
	kvm_dtable
from include/linux/kvm.h to include/asm-x86/kvm.h

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 17 +++++++++++++++++
 include/linux/kvm.h   | 15 ---------------
 2 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index a2c65b5c4a85..644a325077f0 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -82,4 +82,21 @@ struct kvm_lapic_state {
 	char regs[KVM_APIC_REG_SIZE];
 };
 
+struct kvm_segment {
+	__u64 base;
+	__u32 limit;
+	__u16 selector;
+	__u8  type;
+	__u8  present, dpl, db, s, l, g, avl;
+	__u8  unusable;
+	__u8  padding;
+};
+
+struct kvm_dtable {
+	__u64 base;
+	__u16 limit;
+	__u16 padding[3];
+};
+
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 280ec0d85b4c..442cb58bb73c 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -151,21 +151,6 @@ struct kvm_fpu {
 };
 
 
-struct kvm_segment {
-	__u64 base;
-	__u32 limit;
-	__u16 selector;
-	__u8  type;
-	__u8  present, dpl, db, s, l, g, avl;
-	__u8  unusable;
-	__u8  padding;
-};
-
-struct kvm_dtable {
-	__u64 base;
-	__u16 limit;
-	__u16 padding[3];
-};
 
 /* for KVM_GET_SREGS and KVM_SET_SREGS */
 struct kvm_sregs {
-- 
cgit v1.2.3


From 244d57ece94b56692c61f2d507d1251b5844c550 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:36 -0600
Subject: KVM: Portability: Move kvm_sregs and msr structures to <asm/kvm.h>

Move structures:
	kvm_sregs
	kvm_msr_entry
	kvm_msrs
	kvm_msr_list

from include/linux/kvm.h to include/asm-x86/kvm.h

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/kvm.h   | 36 ------------------------------------
 2 files changed, 36 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 644a325077f0..32c7dda61474 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -9,6 +9,9 @@
 #include <asm/types.h>
 #include <linux/ioctl.h>
 
+/* Architectural interrupt line count. */
+#define KVM_NR_INTERRUPTS 256
+
 struct kvm_memory_alias {
 	__u32 slot;  /* this has a different namespace than memory slots */
 	__u32 flags;
@@ -99,4 +102,37 @@ struct kvm_dtable {
 };
 
 
+/* for KVM_GET_SREGS and KVM_SET_SREGS */
+struct kvm_sregs {
+	/* out (KVM_GET_SREGS) / in (KVM_SET_SREGS) */
+	struct kvm_segment cs, ds, es, fs, gs, ss;
+	struct kvm_segment tr, ldt;
+	struct kvm_dtable gdt, idt;
+	__u64 cr0, cr2, cr3, cr4, cr8;
+	__u64 efer;
+	__u64 apic_base;
+	__u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
+};
+
+struct kvm_msr_entry {
+	__u32 index;
+	__u32 reserved;
+	__u64 data;
+};
+
+/* for KVM_GET_MSRS and KVM_SET_MSRS */
+struct kvm_msrs {
+	__u32 nmsrs; /* number of msrs in entries */
+	__u32 pad;
+
+	struct kvm_msr_entry entries[0];
+};
+
+/* for KVM_GET_MSR_INDEX_LIST */
+struct kvm_msr_list {
+	__u32 nmsrs; /* number of msrs in entries */
+	__u32 indices[0];
+};
+
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 442cb58bb73c..e6867aa6daf6 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -13,9 +13,6 @@
 
 #define KVM_API_VERSION 12
 
-/* Architectural interrupt line count. */
-#define KVM_NR_INTERRUPTS 256
-
 /* for KVM_CREATE_MEMORY_REGION */
 struct kvm_memory_region {
 	__u32 slot;
@@ -151,39 +148,6 @@ struct kvm_fpu {
 };
 
 
-
-/* for KVM_GET_SREGS and KVM_SET_SREGS */
-struct kvm_sregs {
-	/* out (KVM_GET_SREGS) / in (KVM_SET_SREGS) */
-	struct kvm_segment cs, ds, es, fs, gs, ss;
-	struct kvm_segment tr, ldt;
-	struct kvm_dtable gdt, idt;
-	__u64 cr0, cr2, cr3, cr4, cr8;
-	__u64 efer;
-	__u64 apic_base;
-	__u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
-};
-
-struct kvm_msr_entry {
-	__u32 index;
-	__u32 reserved;
-	__u64 data;
-};
-
-/* for KVM_GET_MSRS and KVM_SET_MSRS */
-struct kvm_msrs {
-	__u32 nmsrs; /* number of msrs in entries */
-	__u32 pad;
-
-	struct kvm_msr_entry entries[0];
-};
-
-/* for KVM_GET_MSR_INDEX_LIST */
-struct kvm_msr_list {
-	__u32 nmsrs; /* number of msrs in entries */
-	__u32 indices[0];
-};
-
 /* for KVM_TRANSLATE */
 struct kvm_translation {
 	/* in */
-- 
cgit v1.2.3


From a162dd58736d446f52ca45b170508c3a5868807b Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 19 Nov 2007 17:06:37 -0600
Subject: KVM: Portability: Move cpuid structures to <asm/kvm.h>

This patch moves structures:
	kvm_cpuid_entry
	kvm_cpuid

from include/linux/kvm.h to include/asm-x86/kvm.h

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 17 +++++++++++++++++
 include/linux/kvm.h   | 16 ----------------
 2 files changed, 17 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 32c7dda61474..4837d759bec0 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -135,4 +135,21 @@ struct kvm_msr_list {
 };
 
 
+struct kvm_cpuid_entry {
+	__u32 function;
+	__u32 eax;
+	__u32 ebx;
+	__u32 ecx;
+	__u32 edx;
+	__u32 padding;
+};
+
+/* for KVM_SET_CPUID */
+struct kvm_cpuid {
+	__u32 nent;
+	__u32 padding;
+	struct kvm_cpuid_entry entries[0];
+};
+
+
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index e6867aa6daf6..fd4f900fcce3 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -192,22 +192,6 @@ struct kvm_dirty_log {
 	};
 };
 
-struct kvm_cpuid_entry {
-	__u32 function;
-	__u32 eax;
-	__u32 ebx;
-	__u32 ecx;
-	__u32 edx;
-	__u32 padding;
-};
-
-/* for KVM_SET_CPUID */
-struct kvm_cpuid {
-	__u32 nent;
-	__u32 padding;
-	struct kvm_cpuid_entry entries[0];
-};
-
 /* for KVM_SET_SIGNAL_MASK */
 struct kvm_signal_mask {
 	__u32 len;
-- 
cgit v1.2.3


From 2d2a7b9cff7f6324e69daa3ab76799641a94a0e8 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Tue, 20 Nov 2007 11:45:14 +0200
Subject: KVM: Export include/asm-x86/kvm.h

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/Kbuild | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index e6189b229143..3c6f0f80e827 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -3,6 +3,7 @@ include include/asm-generic/Kbuild.asm
 header-y += boot.h
 header-y += bootparam.h
 header-y += debugreg.h
+header-y += kvm.h
 header-y += ldt.h
 header-y += msr-index.h
 header-y += prctl.h
-- 
cgit v1.2.3


From 0771671749b59a507b6da4efb931c44d9691e248 Mon Sep 17 00:00:00 2001
From: Dan Kenigsberg <danken@qumranet.com>
Date: Wed, 21 Nov 2007 17:10:04 +0200
Subject: KVM: Enhance guest cpuid management

The current cpuid management suffers from several problems, which inhibit
passing through the host feature set to the guest:

 - No way to tell which features the host supports

  While some features can be supported with no changes to kvm, others
  need explicit support.  That means kvm needs to vet the feature set
  before it is passed to the guest.

 - No support for indexed or stateful cpuid entries

  Some cpuid entries depend on ecx as well as on eax, or on internal
  state in the processor (running cpuid multiple times with the same
  input returns different output).  The current cpuid machinery only
  supports keying on eax.

 - No support for save/restore/migrate

  The internal state above needs to be exposed to userspace so it can
  be saved or migrated.

This patch adds extended cpuid support by means of three new ioctls:

 - KVM_GET_SUPPORTED_CPUID: get all cpuid entries the host (and kvm)
   supports

 - KVM_SET_CPUID2: sets the vcpu's cpuid table

 - KVM_GET_CPUID2: gets the vcpu's cpuid table, including hidden state

[avi: fix original KVM_SET_CPUID not removing nx on non-nx hosts as it did
      before]

Signed-off-by: Dan Kenigsberg <danken@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/x86.c     | 331 ++++++++++++++++++++++++++++++++++++++++++++++++--
 drivers/kvm/x86.h     |   2 +-
 include/asm-x86/kvm.h |  21 ++++
 include/linux/kvm.h   |   4 +
 4 files changed, 348 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 15e1203faef0..7237cb25f77d 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -646,6 +646,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
 	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_SET_TSS_ADDR:
+	case KVM_CAP_EXT_CPUID:
 		r = 1;
 		break;
 	default:
@@ -708,13 +709,19 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	kvm_put_guest_fpu(vcpu);
 }
 
-static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
+static int is_efer_nx(void)
 {
 	u64 efer;
-	int i;
-	struct kvm_cpuid_entry *e, *entry;
 
 	rdmsrl(MSR_EFER, efer);
+	return efer & EFER_NX;
+}
+
+static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_cpuid_entry2 *e, *entry;
+
 	entry = NULL;
 	for (i = 0; i < vcpu->cpuid_nent; ++i) {
 		e = &vcpu->cpuid_entries[i];
@@ -723,15 +730,56 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
 			break;
 		}
 	}
-	if (entry && (entry->edx & (1 << 20)) && !(efer & EFER_NX)) {
+	if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) {
 		entry->edx &= ~(1 << 20);
 		printk(KERN_INFO "kvm: guest NX capability removed\n");
 	}
 }
 
+/* when an old userspace process fills a new kernel module */
 static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
 				    struct kvm_cpuid *cpuid,
 				    struct kvm_cpuid_entry __user *entries)
+{
+	int r, i;
+	struct kvm_cpuid_entry *cpuid_entries;
+
+	r = -E2BIG;
+	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
+		goto out;
+	r = -ENOMEM;
+	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent);
+	if (!cpuid_entries)
+		goto out;
+	r = -EFAULT;
+	if (copy_from_user(cpuid_entries, entries,
+			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
+		goto out_free;
+	for (i = 0; i < cpuid->nent; i++) {
+		vcpu->cpuid_entries[i].function = cpuid_entries[i].function;
+		vcpu->cpuid_entries[i].eax = cpuid_entries[i].eax;
+		vcpu->cpuid_entries[i].ebx = cpuid_entries[i].ebx;
+		vcpu->cpuid_entries[i].ecx = cpuid_entries[i].ecx;
+		vcpu->cpuid_entries[i].edx = cpuid_entries[i].edx;
+		vcpu->cpuid_entries[i].index = 0;
+		vcpu->cpuid_entries[i].flags = 0;
+		vcpu->cpuid_entries[i].padding[0] = 0;
+		vcpu->cpuid_entries[i].padding[1] = 0;
+		vcpu->cpuid_entries[i].padding[2] = 0;
+	}
+	vcpu->cpuid_nent = cpuid->nent;
+	cpuid_fix_nx_cap(vcpu);
+	r = 0;
+
+out_free:
+	vfree(cpuid_entries);
+out:
+	return r;
+}
+
+static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
 {
 	int r;
 
@@ -740,16 +788,198 @@ static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
 		goto out;
 	r = -EFAULT;
 	if (copy_from_user(&vcpu->cpuid_entries, entries,
-			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
+			   cpuid->nent * sizeof(struct kvm_cpuid_entry2)))
 		goto out;
 	vcpu->cpuid_nent = cpuid->nent;
-	cpuid_fix_nx_cap(vcpu);
 	return 0;
 
 out:
 	return r;
 }
 
+static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
+{
+	int r;
+
+	r = -E2BIG;
+	if (cpuid->nent < vcpu->cpuid_nent)
+		goto out;
+	r = -EFAULT;
+	if (copy_to_user(entries, &vcpu->cpuid_entries,
+			   vcpu->cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
+		goto out;
+	return 0;
+
+out:
+	cpuid->nent = vcpu->cpuid_nent;
+	return r;
+}
+
+static inline u32 bit(int bitno)
+{
+	return 1 << (bitno & 31);
+}
+
+static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function,
+			  u32 index)
+{
+	entry->function = function;
+	entry->index = index;
+	cpuid_count(entry->function, entry->index,
+		&entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
+	entry->flags = 0;
+}
+
+static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
+			 u32 index, int *nent, int maxnent)
+{
+	const u32 kvm_supported_word0_x86_features = bit(X86_FEATURE_FPU) |
+		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
+		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
+		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
+		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
+		bit(X86_FEATURE_SEP) | bit(X86_FEATURE_PGE) |
+		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
+		bit(X86_FEATURE_CLFLSH) | bit(X86_FEATURE_MMX) |
+		bit(X86_FEATURE_FXSR) | bit(X86_FEATURE_XMM) |
+		bit(X86_FEATURE_XMM2) | bit(X86_FEATURE_SELFSNOOP);
+	const u32 kvm_supported_word1_x86_features = bit(X86_FEATURE_FPU) |
+		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
+		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
+		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
+		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
+		bit(X86_FEATURE_PGE) |
+		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
+		bit(X86_FEATURE_MMX) | bit(X86_FEATURE_FXSR) |
+		bit(X86_FEATURE_SYSCALL) |
+		(bit(X86_FEATURE_NX) && is_efer_nx()) |
+#ifdef CONFIG_X86_64
+		bit(X86_FEATURE_LM) |
+#endif
+		bit(X86_FEATURE_MMXEXT) |
+		bit(X86_FEATURE_3DNOWEXT) |
+		bit(X86_FEATURE_3DNOW);
+	const u32 kvm_supported_word3_x86_features =
+		bit(X86_FEATURE_XMM3) | bit(X86_FEATURE_CX16);
+	const u32 kvm_supported_word6_x86_features =
+		bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY);
+
+	/* all func 2 cpuid_count() should be called on the same cpu */
+	get_cpu();
+	do_cpuid_1_ent(entry, function, index);
+	++*nent;
+
+	switch (function) {
+	case 0:
+		entry->eax = min(entry->eax, (u32)0xb);
+		break;
+	case 1:
+		entry->edx &= kvm_supported_word0_x86_features;
+		entry->ecx &= kvm_supported_word3_x86_features;
+		break;
+	/* function 2 entries are STATEFUL. That is, repeated cpuid commands
+	 * may return different values. This forces us to get_cpu() before
+	 * issuing the first command, and also to emulate this annoying behavior
+	 * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */
+	case 2: {
+		int t, times = entry->eax & 0xff;
+
+		entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
+		for (t = 1; t < times && *nent < maxnent; ++t) {
+			do_cpuid_1_ent(&entry[t], function, 0);
+			entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
+			++*nent;
+		}
+		break;
+	}
+	/* function 4 and 0xb have additional index. */
+	case 4: {
+		int index, cache_type;
+
+		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+		/* read more entries until cache_type is zero */
+		for (index = 1; *nent < maxnent; ++index) {
+			cache_type = entry[index - 1].eax & 0x1f;
+			if (!cache_type)
+				break;
+			do_cpuid_1_ent(&entry[index], function, index);
+			entry[index].flags |=
+			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+			++*nent;
+		}
+		break;
+	}
+	case 0xb: {
+		int index, level_type;
+
+		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+		/* read more entries until level_type is zero */
+		for (index = 1; *nent < maxnent; ++index) {
+			level_type = entry[index - 1].ecx & 0xff;
+			if (!level_type)
+				break;
+			do_cpuid_1_ent(&entry[index], function, index);
+			entry[index].flags |=
+			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+			++*nent;
+		}
+		break;
+	}
+	case 0x80000000:
+		entry->eax = min(entry->eax, 0x8000001a);
+		break;
+	case 0x80000001:
+		entry->edx &= kvm_supported_word1_x86_features;
+		entry->ecx &= kvm_supported_word6_x86_features;
+		break;
+	}
+	put_cpu();
+}
+
+static int kvm_vm_ioctl_get_supported_cpuid(struct kvm *kvm,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
+{
+	struct kvm_cpuid_entry2 *cpuid_entries;
+	int limit, nent = 0, r = -E2BIG;
+	u32 func;
+
+	if (cpuid->nent < 1)
+		goto out;
+	r = -ENOMEM;
+	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
+	if (!cpuid_entries)
+		goto out;
+
+	do_cpuid_ent(&cpuid_entries[0], 0, 0, &nent, cpuid->nent);
+	limit = cpuid_entries[0].eax;
+	for (func = 1; func <= limit && nent < cpuid->nent; ++func)
+		do_cpuid_ent(&cpuid_entries[nent], func, 0,
+				&nent, cpuid->nent);
+	r = -E2BIG;
+	if (nent >= cpuid->nent)
+		goto out_free;
+
+	do_cpuid_ent(&cpuid_entries[nent], 0x80000000, 0, &nent, cpuid->nent);
+	limit = cpuid_entries[nent - 1].eax;
+	for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func)
+		do_cpuid_ent(&cpuid_entries[nent], func, 0,
+			       &nent, cpuid->nent);
+	r = -EFAULT;
+	if (copy_to_user(entries, cpuid_entries,
+			nent * sizeof(struct kvm_cpuid_entry2)))
+		goto out_free;
+	cpuid->nent = nent;
+	r = 0;
+
+out_free:
+	vfree(cpuid_entries);
+out:
+	return r;
+}
+
 static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu,
 				    struct kvm_lapic_state *s)
 {
@@ -816,6 +1046,36 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 			goto out;
 		break;
 	}
+	case KVM_SET_CPUID2: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid,
+				cpuid_arg->entries);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_GET_CPUID2: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid,
+				cpuid_arg->entries);
+		if (r)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
+			goto out;
+		r = 0;
+		break;
+	}
 	case KVM_GET_MSRS:
 		r = msr_io(vcpu, argp, kvm_get_msr, 1);
 		break;
@@ -1111,6 +1371,24 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
+	case KVM_GET_SUPPORTED_CPUID: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vm_ioctl_get_supported_cpuid(kvm, &cpuid,
+			cpuid_arg->entries);
+		if (r)
+			goto out;
+
+		r = -EFAULT;
+		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
+			goto out;
+		r = 0;
+		break;
+	}
 	default:
 		;
 	}
@@ -1908,14 +2186,47 @@ void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
 	}
 }
 
+static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i)
+{
+	struct kvm_cpuid_entry2 *e = &vcpu->cpuid_entries[i];
+	int j, nent = vcpu->cpuid_nent;
+
+	e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
+	/* when no next entry is found, the current entry[i] is reselected */
+	for (j = i + 1; j == i; j = (j + 1) % nent) {
+		struct kvm_cpuid_entry2 *ej = &vcpu->cpuid_entries[j];
+		if (ej->function == e->function) {
+			ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
+			return j;
+		}
+	}
+	return 0; /* silence gcc, even though control never reaches here */
+}
+
+/* find an entry with matching function, matching index (if needed), and that
+ * should be read next (if it's stateful) */
+static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e,
+	u32 function, u32 index)
+{
+	if (e->function != function)
+		return 0;
+	if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index)
+		return 0;
+	if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) &&
+		!(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
+		return 0;
+	return 1;
+}
+
 void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
 {
 	int i;
-	u32 function;
-	struct kvm_cpuid_entry *e, *best;
+	u32 function, index;
+	struct kvm_cpuid_entry2 *e, *best;
 
 	kvm_x86_ops->cache_regs(vcpu);
 	function = vcpu->regs[VCPU_REGS_RAX];
+	index = vcpu->regs[VCPU_REGS_RCX];
 	vcpu->regs[VCPU_REGS_RAX] = 0;
 	vcpu->regs[VCPU_REGS_RBX] = 0;
 	vcpu->regs[VCPU_REGS_RCX] = 0;
@@ -1923,7 +2234,9 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
 	best = NULL;
 	for (i = 0; i < vcpu->cpuid_nent; ++i) {
 		e = &vcpu->cpuid_entries[i];
-		if (e->function == function) {
+		if (is_matching_cpuid_entry(e, function, index)) {
+			if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
+				move_to_next_stateful_cpuid_entry(vcpu, i);
 			best = e;
 			break;
 		}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index b1528c9f566f..78ab1e108d8b 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -149,7 +149,7 @@ struct kvm_vcpu {
 	int halt_request; /* real mode on Intel only */
 
 	int cpuid_nent;
-	struct kvm_cpuid_entry cpuid_entries[KVM_MAX_CPUID_ENTRIES];
+	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
 
 	/* emulate context */
 
diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 4837d759bec0..17afa81d6d66 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -151,5 +151,26 @@ struct kvm_cpuid {
 	struct kvm_cpuid_entry entries[0];
 };
 
+struct kvm_cpuid_entry2 {
+	__u32 function;
+	__u32 index;
+	__u32 flags;
+	__u32 eax;
+	__u32 ebx;
+	__u32 ecx;
+	__u32 edx;
+	__u32 padding[3];
+};
+
+#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 1
+#define KVM_CPUID_FLAG_STATEFUL_FUNC    2
+#define KVM_CPUID_FLAG_STATE_READ_NEXT  4
+
+/* for KVM_SET_CPUID2 */
+struct kvm_cpuid2 {
+	__u32 nent;
+	__u32 padding;
+	struct kvm_cpuid_entry2 entries[0];
+};
 
 #endif
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index fd4f900fcce3..b751552f2e30 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -224,6 +224,7 @@ struct kvm_signal_mask {
 #define KVM_CAP_MMU_SHADOW_CACHE_CONTROL 2
 #define KVM_CAP_USER_MEMORY 3
 #define KVM_CAP_SET_TSS_ADDR 4
+#define KVM_CAP_EXT_CPUID 5
 
 /*
  * ioctls for VM fds
@@ -241,6 +242,7 @@ struct kvm_signal_mask {
 #define KVM_CREATE_VCPU           _IO(KVMIO,  0x41)
 #define KVM_GET_DIRTY_LOG         _IOW(KVMIO, 0x42, struct kvm_dirty_log)
 #define KVM_SET_MEMORY_ALIAS      _IOW(KVMIO, 0x43, struct kvm_memory_alias)
+#define KVM_GET_SUPPORTED_CPUID   _IOWR(KVMIO, 0x48, struct kvm_cpuid2)
 /* Device model IOC */
 #define KVM_CREATE_IRQCHIP	  _IO(KVMIO,  0x60)
 #define KVM_IRQ_LINE		  _IOW(KVMIO, 0x61, struct kvm_irq_level)
@@ -266,5 +268,7 @@ struct kvm_signal_mask {
 #define KVM_SET_FPU               _IOW(KVMIO,  0x8d, struct kvm_fpu)
 #define KVM_GET_LAPIC             _IOR(KVMIO,  0x8e, struct kvm_lapic_state)
 #define KVM_SET_LAPIC             _IOW(KVMIO,  0x8f, struct kvm_lapic_state)
+#define KVM_SET_CPUID2            _IOW(KVMIO,  0x90, struct kvm_cpuid2)
+#define KVM_GET_CPUID2            _IOWR(KVMIO, 0x91, struct kvm_cpuid2)
 
 #endif
-- 
cgit v1.2.3


From 51e296258cba41759ff0de5da110d8d52675ee67 Mon Sep 17 00:00:00 2001
From: Jerone Young <jyoung5@us.ibm.com>
Date: Mon, 26 Nov 2007 08:33:53 -0600
Subject: KVM: Add ifdef in irqchip struct for x86 only structures

This patch fixes a small issue where sturctures:
	kvm_pic_state
	kvm_ioapic_state

are defined inside x86 specific code and may or may not
be defined in anyway for other architectures. The problem
caused is one cannot compile userspace apps (ex. libkvm)
for other archs since a size cannot be determined for these
structures.

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/linux/Kbuild | 2 +-
 include/linux/kvm.h  | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 27b9350052b4..c263a69b5680 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -100,7 +100,6 @@ header-y += iso_fs.h
 header-y += ixjuser.h
 header-y += jffs2.h
 header-y += keyctl.h
-header-y += kvm.h
 header-y += limits.h
 header-y += lock_dlm_plock.h
 header-y += magic.h
@@ -256,6 +255,7 @@ unifdef-y += kd.h
 unifdef-y += kernelcapi.h
 unifdef-y += kernel.h
 unifdef-y += keyboard.h
+unifdef-y += kvm.h
 unifdef-y += llc.h
 unifdef-y += loop.h
 unifdef-y += lp.h
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index b751552f2e30..f0bebd61ff77 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -51,8 +51,10 @@ struct kvm_irqchip {
 	__u32 pad;
         union {
 		char dummy[512];  /* reserving space */
+#ifdef CONFIG_X86
 		struct kvm_pic_state pic;
 		struct kvm_ioapic_state ioapic;
+#endif
 	} chip;
 };
 
-- 
cgit v1.2.3


From fb56dbb31c4738a3918db81fd24da732ce3b4ae6 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Sun, 2 Dec 2007 10:50:06 +0200
Subject: KVM: Export include/linux/kvm.h only if $ARCH actually supports KVM

Currently, make headers_check barfs due to <asm/kvm.h>, which <linux/kvm.h>
includes, not existing.  Rather than add a zillion <asm/kvm.h>s, export kvm.h
only if the arch actually supports it.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/Kconfig     | 1 +
 drivers/kvm/Kconfig  | 7 +++++--
 include/linux/Kbuild | 2 +-
 3 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index fb3eea3e38ee..d289cfcf92c4 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -107,6 +107,7 @@ config ARCH_SUPPORTS_OPROFILE
 	bool
 	default y
 
+select HAVE_KVM
 
 config ZONE_DMA32
 	bool
diff --git a/drivers/kvm/Kconfig b/drivers/kvm/Kconfig
index 656920636cb2..c83e1c9b5129 100644
--- a/drivers/kvm/Kconfig
+++ b/drivers/kvm/Kconfig
@@ -1,9 +1,12 @@
 #
 # KVM configuration
 #
+config HAVE_KVM
+       bool
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
-	depends on X86
+	depends on HAVE_KVM || X86
 	default y
 	---help---
 	  Say Y here to get to see options for using your Linux host to run other
@@ -16,7 +19,7 @@ if VIRTUALIZATION
 
 config KVM
 	tristate "Kernel-based Virtual Machine (KVM) support"
-	depends on X86 && EXPERIMENTAL
+	depends on HAVE_KVM && EXPERIMENTAL
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
 	---help---
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index c263a69b5680..85b2482cc736 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -255,7 +255,7 @@ unifdef-y += kd.h
 unifdef-y += kernelcapi.h
 unifdef-y += kernel.h
 unifdef-y += keyboard.h
-unifdef-y += kvm.h
+unifdef-$(CONFIG_HAVE_KVM) += kvm.h
 unifdef-y += llc.h
 unifdef-y += loop.h
 unifdef-y += lp.h
-- 
cgit v1.2.3


From edf884172e9828c6234b254208af04655855038d Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Sun, 16 Dec 2007 11:02:48 +0200
Subject: KVM: Move arch dependent files to new directory arch/x86/kvm/

This paves the way for multiple architecture support.  Note that while
ioapic.c could potentially be shared with ia64, it is also moved.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/Kconfig                  |    2 +
 arch/x86/Makefile                 |    2 +
 arch/x86/kvm/Kconfig              |   57 +
 arch/x86/kvm/Makefile             |   15 +
 arch/x86/kvm/i8259.c              |  450 ++++++
 arch/x86/kvm/ioapic.c             |  400 +++++
 arch/x86/kvm/irq.c                |   98 ++
 arch/x86/kvm/irq.h                |  195 +++
 arch/x86/kvm/kvm_svm.h            |   45 +
 arch/x86/kvm/lapic.c              | 1085 +++++++++++++
 arch/x86/kvm/mmu.c                | 1805 +++++++++++++++++++++
 arch/x86/kvm/mmu.h                |   44 +
 arch/x86/kvm/paging_tmpl.h        |  461 ++++++
 arch/x86/kvm/segment_descriptor.h |   29 +
 arch/x86/kvm/svm.c                | 1725 ++++++++++++++++++++
 arch/x86/kvm/svm.h                |  325 ++++
 arch/x86/kvm/vmx.c                | 2671 +++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx.h                |  324 ++++
 arch/x86/kvm/x86.c                | 3146 ++++++++++++++++++++++++++++++++++++
 arch/x86/kvm/x86_emulate.c        | 1912 ++++++++++++++++++++++
 drivers/Kconfig                   |    2 -
 drivers/Makefile                  |    1 -
 drivers/kvm/Kconfig               |   57 -
 drivers/kvm/Makefile              |   10 -
 drivers/kvm/i8259.c               |  449 ------
 drivers/kvm/ioapic.c              |  402 -----
 drivers/kvm/iodev.h               |    2 +-
 drivers/kvm/irq.c                 |   99 --
 drivers/kvm/irq.h                 |  196 ---
 drivers/kvm/kvm.h                 |  289 ----
 drivers/kvm/kvm_main.c            |    2 +-
 drivers/kvm/kvm_svm.h             |   45 -
 drivers/kvm/lapic.c               | 1087 -------------
 drivers/kvm/mmu.c                 | 1806 ---------------------
 drivers/kvm/mmu.h                 |   44 -
 drivers/kvm/paging_tmpl.h         |  461 ------
 drivers/kvm/segment_descriptor.h  |   29 -
 drivers/kvm/svm.c                 | 1725 --------------------
 drivers/kvm/svm.h                 |  325 ----
 drivers/kvm/types.h               |   54 -
 drivers/kvm/vmx.c                 | 2673 -------------------------------
 drivers/kvm/vmx.h                 |  324 ----
 drivers/kvm/x86.c                 | 3148 -------------------------------------
 drivers/kvm/x86.h                 |  602 -------
 drivers/kvm/x86_emulate.c         | 1913 ----------------------
 drivers/kvm/x86_emulate.h         |  186 ---
 include/asm-x86/kvm_host.h        |  601 +++++++
 include/asm-x86/kvm_x86_emulate.h |  186 +++
 include/linux/kvm_host.h          |  289 ++++
 include/linux/kvm_types.h         |   54 +
 50 files changed, 15923 insertions(+), 15929 deletions(-)
 create mode 100644 arch/x86/kvm/Kconfig
 create mode 100644 arch/x86/kvm/Makefile
 create mode 100644 arch/x86/kvm/i8259.c
 create mode 100644 arch/x86/kvm/ioapic.c
 create mode 100644 arch/x86/kvm/irq.c
 create mode 100644 arch/x86/kvm/irq.h
 create mode 100644 arch/x86/kvm/kvm_svm.h
 create mode 100644 arch/x86/kvm/lapic.c
 create mode 100644 arch/x86/kvm/mmu.c
 create mode 100644 arch/x86/kvm/mmu.h
 create mode 100644 arch/x86/kvm/paging_tmpl.h
 create mode 100644 arch/x86/kvm/segment_descriptor.h
 create mode 100644 arch/x86/kvm/svm.c
 create mode 100644 arch/x86/kvm/svm.h
 create mode 100644 arch/x86/kvm/vmx.c
 create mode 100644 arch/x86/kvm/vmx.h
 create mode 100644 arch/x86/kvm/x86.c
 create mode 100644 arch/x86/kvm/x86_emulate.c
 delete mode 100644 drivers/kvm/Kconfig
 delete mode 100644 drivers/kvm/Makefile
 delete mode 100644 drivers/kvm/i8259.c
 delete mode 100644 drivers/kvm/ioapic.c
 delete mode 100644 drivers/kvm/irq.c
 delete mode 100644 drivers/kvm/irq.h
 delete mode 100644 drivers/kvm/kvm.h
 delete mode 100644 drivers/kvm/kvm_svm.h
 delete mode 100644 drivers/kvm/lapic.c
 delete mode 100644 drivers/kvm/mmu.c
 delete mode 100644 drivers/kvm/mmu.h
 delete mode 100644 drivers/kvm/paging_tmpl.h
 delete mode 100644 drivers/kvm/segment_descriptor.h
 delete mode 100644 drivers/kvm/svm.c
 delete mode 100644 drivers/kvm/svm.h
 delete mode 100644 drivers/kvm/types.h
 delete mode 100644 drivers/kvm/vmx.c
 delete mode 100644 drivers/kvm/vmx.h
 delete mode 100644 drivers/kvm/x86.c
 delete mode 100644 drivers/kvm/x86.h
 delete mode 100644 drivers/kvm/x86_emulate.c
 delete mode 100644 drivers/kvm/x86_emulate.h
 create mode 100644 include/asm-x86/kvm_host.h
 create mode 100644 include/asm-x86/kvm_x86_emulate.h
 create mode 100644 include/linux/kvm_host.h
 create mode 100644 include/linux/kvm_types.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index d289cfcf92c4..65b449134cf7 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1599,4 +1599,6 @@ source "security/Kconfig"
 
 source "crypto/Kconfig"
 
+source "arch/x86/kvm/Kconfig"
+
 source "lib/Kconfig"
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index b08f18261df6..da8f4129780b 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -7,6 +7,8 @@ else
         KBUILD_DEFCONFIG := $(ARCH)_defconfig
 endif
 
+core-$(CONFIG_KVM) += arch/x86/kvm/
+
 # BITS is used as extension for files which are available in a 32 bit
 # and a 64 bit version to simplify shared Makefiles.
 # e.g.: obj-y += foo_$(BITS).o
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
new file mode 100644
index 000000000000..c83e1c9b5129
--- /dev/null
+++ b/arch/x86/kvm/Kconfig
@@ -0,0 +1,57 @@
+#
+# KVM configuration
+#
+config HAVE_KVM
+       bool
+
+menuconfig VIRTUALIZATION
+	bool "Virtualization"
+	depends on HAVE_KVM || X86
+	default y
+	---help---
+	  Say Y here to get to see options for using your Linux host to run other
+	  operating systems inside virtual machines (guests).
+	  This option alone does not add any kernel code.
+
+	  If you say N, all options in this submenu will be skipped and disabled.
+
+if VIRTUALIZATION
+
+config KVM
+	tristate "Kernel-based Virtual Machine (KVM) support"
+	depends on HAVE_KVM && EXPERIMENTAL
+	select PREEMPT_NOTIFIERS
+	select ANON_INODES
+	---help---
+	  Support hosting fully virtualized guest machines using hardware
+	  virtualization extensions.  You will need a fairly recent
+	  processor equipped with virtualization extensions. You will also
+	  need to select one or more of the processor modules below.
+
+	  This module provides access to the hardware capabilities through
+	  a character device node named /dev/kvm.
+
+	  To compile this as a module, choose M here: the module
+	  will be called kvm.
+
+	  If unsure, say N.
+
+config KVM_INTEL
+	tristate "KVM for Intel processors support"
+	depends on KVM
+	---help---
+	  Provides support for KVM on Intel processors equipped with the VT
+	  extensions.
+
+config KVM_AMD
+	tristate "KVM for AMD processors support"
+	depends on KVM
+	---help---
+	  Provides support for KVM on AMD processors equipped with the AMD-V
+	  (SVM) extensions.
+
+# OK, it's a little counter-intuitive to do this, but it puts it neatly under
+# the virtualization menu.
+source drivers/lguest/Kconfig
+
+endif # VIRTUALIZATION
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
new file mode 100644
index 000000000000..880ffe403b35
--- /dev/null
+++ b/arch/x86/kvm/Makefile
@@ -0,0 +1,15 @@
+#
+# Makefile for Kernel-based Virtual Machine module
+#
+
+common-objs = $(addprefix ../../../drivers/kvm/, kvm_main.o)
+
+EXTRA_CFLAGS += -I drivers/kvm
+
+kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o \
+	ioapic.o
+obj-$(CONFIG_KVM) += kvm.o
+kvm-intel-objs = vmx.o
+obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
+kvm-amd-objs = svm.o
+obj-$(CONFIG_KVM_AMD) += kvm-amd.o
diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c
new file mode 100644
index 000000000000..ab29cf2def47
--- /dev/null
+++ b/arch/x86/kvm/i8259.c
@@ -0,0 +1,450 @@
+/*
+ * 8259 interrupt controller emulation
+ *
+ * Copyright (c) 2003-2004 Fabrice Bellard
+ * Copyright (c) 2007 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ * Authors:
+ *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
+ *   Port from Qemu.
+ */
+#include <linux/mm.h>
+#include "irq.h"
+
+#include <linux/kvm_host.h>
+
+/*
+ * set irq level. If an edge is detected, then the IRR is set to 1
+ */
+static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level)
+{
+	int mask;
+	mask = 1 << irq;
+	if (s->elcr & mask)	/* level triggered */
+		if (level) {
+			s->irr |= mask;
+			s->last_irr |= mask;
+		} else {
+			s->irr &= ~mask;
+			s->last_irr &= ~mask;
+		}
+	else	/* edge triggered */
+		if (level) {
+			if ((s->last_irr & mask) == 0)
+				s->irr |= mask;
+			s->last_irr |= mask;
+		} else
+			s->last_irr &= ~mask;
+}
+
+/*
+ * return the highest priority found in mask (highest = smallest
+ * number). Return 8 if no irq
+ */
+static inline int get_priority(struct kvm_kpic_state *s, int mask)
+{
+	int priority;
+	if (mask == 0)
+		return 8;
+	priority = 0;
+	while ((mask & (1 << ((priority + s->priority_add) & 7))) == 0)
+		priority++;
+	return priority;
+}
+
+/*
+ * return the pic wanted interrupt. return -1 if none
+ */
+static int pic_get_irq(struct kvm_kpic_state *s)
+{
+	int mask, cur_priority, priority;
+
+	mask = s->irr & ~s->imr;
+	priority = get_priority(s, mask);
+	if (priority == 8)
+		return -1;
+	/*
+	 * compute current priority. If special fully nested mode on the
+	 * master, the IRQ coming from the slave is not taken into account
+	 * for the priority computation.
+	 */
+	mask = s->isr;
+	if (s->special_fully_nested_mode && s == &s->pics_state->pics[0])
+		mask &= ~(1 << 2);
+	cur_priority = get_priority(s, mask);
+	if (priority < cur_priority)
+		/*
+		 * higher priority found: an irq should be generated
+		 */
+		return (priority + s->priority_add) & 7;
+	else
+		return -1;
+}
+
+/*
+ * raise irq to CPU if necessary. must be called every time the active
+ * irq may change
+ */
+static void pic_update_irq(struct kvm_pic *s)
+{
+	int irq2, irq;
+
+	irq2 = pic_get_irq(&s->pics[1]);
+	if (irq2 >= 0) {
+		/*
+		 * if irq request by slave pic, signal master PIC
+		 */
+		pic_set_irq1(&s->pics[0], 2, 1);
+		pic_set_irq1(&s->pics[0], 2, 0);
+	}
+	irq = pic_get_irq(&s->pics[0]);
+	if (irq >= 0)
+		s->irq_request(s->irq_request_opaque, 1);
+	else
+		s->irq_request(s->irq_request_opaque, 0);
+}
+
+void kvm_pic_update_irq(struct kvm_pic *s)
+{
+	pic_update_irq(s);
+}
+
+void kvm_pic_set_irq(void *opaque, int irq, int level)
+{
+	struct kvm_pic *s = opaque;
+
+	pic_set_irq1(&s->pics[irq >> 3], irq & 7, level);
+	pic_update_irq(s);
+}
+
+/*
+ * acknowledge interrupt 'irq'
+ */
+static inline void pic_intack(struct kvm_kpic_state *s, int irq)
+{
+	if (s->auto_eoi) {
+		if (s->rotate_on_auto_eoi)
+			s->priority_add = (irq + 1) & 7;
+	} else
+		s->isr |= (1 << irq);
+	/*
+	 * We don't clear a level sensitive interrupt here
+	 */
+	if (!(s->elcr & (1 << irq)))
+		s->irr &= ~(1 << irq);
+}
+
+int kvm_pic_read_irq(struct kvm_pic *s)
+{
+	int irq, irq2, intno;
+
+	irq = pic_get_irq(&s->pics[0]);
+	if (irq >= 0) {
+		pic_intack(&s->pics[0], irq);
+		if (irq == 2) {
+			irq2 = pic_get_irq(&s->pics[1]);
+			if (irq2 >= 0)
+				pic_intack(&s->pics[1], irq2);
+			else
+				/*
+				 * spurious IRQ on slave controller
+				 */
+				irq2 = 7;
+			intno = s->pics[1].irq_base + irq2;
+			irq = irq2 + 8;
+		} else
+			intno = s->pics[0].irq_base + irq;
+	} else {
+		/*
+		 * spurious IRQ on host controller
+		 */
+		irq = 7;
+		intno = s->pics[0].irq_base + irq;
+	}
+	pic_update_irq(s);
+
+	return intno;
+}
+
+void kvm_pic_reset(struct kvm_kpic_state *s)
+{
+	s->last_irr = 0;
+	s->irr = 0;
+	s->imr = 0;
+	s->isr = 0;
+	s->priority_add = 0;
+	s->irq_base = 0;
+	s->read_reg_select = 0;
+	s->poll = 0;
+	s->special_mask = 0;
+	s->init_state = 0;
+	s->auto_eoi = 0;
+	s->rotate_on_auto_eoi = 0;
+	s->special_fully_nested_mode = 0;
+	s->init4 = 0;
+}
+
+static void pic_ioport_write(void *opaque, u32 addr, u32 val)
+{
+	struct kvm_kpic_state *s = opaque;
+	int priority, cmd, irq;
+
+	addr &= 1;
+	if (addr == 0) {
+		if (val & 0x10) {
+			kvm_pic_reset(s);	/* init */
+			/*
+			 * deassert a pending interrupt
+			 */
+			s->pics_state->irq_request(s->pics_state->
+						   irq_request_opaque, 0);
+			s->init_state = 1;
+			s->init4 = val & 1;
+			if (val & 0x02)
+				printk(KERN_ERR "single mode not supported");
+			if (val & 0x08)
+				printk(KERN_ERR
+				       "level sensitive irq not supported");
+		} else if (val & 0x08) {
+			if (val & 0x04)
+				s->poll = 1;
+			if (val & 0x02)
+				s->read_reg_select = val & 1;
+			if (val & 0x40)
+				s->special_mask = (val >> 5) & 1;
+		} else {
+			cmd = val >> 5;
+			switch (cmd) {
+			case 0:
+			case 4:
+				s->rotate_on_auto_eoi = cmd >> 2;
+				break;
+			case 1:	/* end of interrupt */
+			case 5:
+				priority = get_priority(s, s->isr);
+				if (priority != 8) {
+					irq = (priority + s->priority_add) & 7;
+					s->isr &= ~(1 << irq);
+					if (cmd == 5)
+						s->priority_add = (irq + 1) & 7;
+					pic_update_irq(s->pics_state);
+				}
+				break;
+			case 3:
+				irq = val & 7;
+				s->isr &= ~(1 << irq);
+				pic_update_irq(s->pics_state);
+				break;
+			case 6:
+				s->priority_add = (val + 1) & 7;
+				pic_update_irq(s->pics_state);
+				break;
+			case 7:
+				irq = val & 7;
+				s->isr &= ~(1 << irq);
+				s->priority_add = (irq + 1) & 7;
+				pic_update_irq(s->pics_state);
+				break;
+			default:
+				break;	/* no operation */
+			}
+		}
+	} else
+		switch (s->init_state) {
+		case 0:		/* normal mode */
+			s->imr = val;
+			pic_update_irq(s->pics_state);
+			break;
+		case 1:
+			s->irq_base = val & 0xf8;
+			s->init_state = 2;
+			break;
+		case 2:
+			if (s->init4)
+				s->init_state = 3;
+			else
+				s->init_state = 0;
+			break;
+		case 3:
+			s->special_fully_nested_mode = (val >> 4) & 1;
+			s->auto_eoi = (val >> 1) & 1;
+			s->init_state = 0;
+			break;
+		}
+}
+
+static u32 pic_poll_read(struct kvm_kpic_state *s, u32 addr1)
+{
+	int ret;
+
+	ret = pic_get_irq(s);
+	if (ret >= 0) {
+		if (addr1 >> 7) {
+			s->pics_state->pics[0].isr &= ~(1 << 2);
+			s->pics_state->pics[0].irr &= ~(1 << 2);
+		}
+		s->irr &= ~(1 << ret);
+		s->isr &= ~(1 << ret);
+		if (addr1 >> 7 || ret != 2)
+			pic_update_irq(s->pics_state);
+	} else {
+		ret = 0x07;
+		pic_update_irq(s->pics_state);
+	}
+
+	return ret;
+}
+
+static u32 pic_ioport_read(void *opaque, u32 addr1)
+{
+	struct kvm_kpic_state *s = opaque;
+	unsigned int addr;
+	int ret;
+
+	addr = addr1;
+	addr &= 1;
+	if (s->poll) {
+		ret = pic_poll_read(s, addr1);
+		s->poll = 0;
+	} else
+		if (addr == 0)
+			if (s->read_reg_select)
+				ret = s->isr;
+			else
+				ret = s->irr;
+		else
+			ret = s->imr;
+	return ret;
+}
+
+static void elcr_ioport_write(void *opaque, u32 addr, u32 val)
+{
+	struct kvm_kpic_state *s = opaque;
+	s->elcr = val & s->elcr_mask;
+}
+
+static u32 elcr_ioport_read(void *opaque, u32 addr1)
+{
+	struct kvm_kpic_state *s = opaque;
+	return s->elcr;
+}
+
+static int picdev_in_range(struct kvm_io_device *this, gpa_t addr)
+{
+	switch (addr) {
+	case 0x20:
+	case 0x21:
+	case 0xa0:
+	case 0xa1:
+	case 0x4d0:
+	case 0x4d1:
+		return 1;
+	default:
+		return 0;
+	}
+}
+
+static void picdev_write(struct kvm_io_device *this,
+			 gpa_t addr, int len, const void *val)
+{
+	struct kvm_pic *s = this->private;
+	unsigned char data = *(unsigned char *)val;
+
+	if (len != 1) {
+		if (printk_ratelimit())
+			printk(KERN_ERR "PIC: non byte write\n");
+		return;
+	}
+	switch (addr) {
+	case 0x20:
+	case 0x21:
+	case 0xa0:
+	case 0xa1:
+		pic_ioport_write(&s->pics[addr >> 7], addr, data);
+		break;
+	case 0x4d0:
+	case 0x4d1:
+		elcr_ioport_write(&s->pics[addr & 1], addr, data);
+		break;
+	}
+}
+
+static void picdev_read(struct kvm_io_device *this,
+			gpa_t addr, int len, void *val)
+{
+	struct kvm_pic *s = this->private;
+	unsigned char data = 0;
+
+	if (len != 1) {
+		if (printk_ratelimit())
+			printk(KERN_ERR "PIC: non byte read\n");
+		return;
+	}
+	switch (addr) {
+	case 0x20:
+	case 0x21:
+	case 0xa0:
+	case 0xa1:
+		data = pic_ioport_read(&s->pics[addr >> 7], addr);
+		break;
+	case 0x4d0:
+	case 0x4d1:
+		data = elcr_ioport_read(&s->pics[addr & 1], addr);
+		break;
+	}
+	*(unsigned char *)val = data;
+}
+
+/*
+ * callback when PIC0 irq status changed
+ */
+static void pic_irq_request(void *opaque, int level)
+{
+	struct kvm *kvm = opaque;
+	struct kvm_vcpu *vcpu = kvm->vcpus[0];
+
+	pic_irqchip(kvm)->output = level;
+	if (vcpu)
+		kvm_vcpu_kick(vcpu);
+}
+
+struct kvm_pic *kvm_create_pic(struct kvm *kvm)
+{
+	struct kvm_pic *s;
+	s = kzalloc(sizeof(struct kvm_pic), GFP_KERNEL);
+	if (!s)
+		return NULL;
+	s->pics[0].elcr_mask = 0xf8;
+	s->pics[1].elcr_mask = 0xde;
+	s->irq_request = pic_irq_request;
+	s->irq_request_opaque = kvm;
+	s->pics[0].pics_state = s;
+	s->pics[1].pics_state = s;
+
+	/*
+	 * Initialize PIO device
+	 */
+	s->dev.read = picdev_read;
+	s->dev.write = picdev_write;
+	s->dev.in_range = picdev_in_range;
+	s->dev.private = s;
+	kvm_io_bus_register_dev(&kvm->pio_bus, &s->dev);
+	return s;
+}
diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
new file mode 100644
index 000000000000..72f12f75495d
--- /dev/null
+++ b/arch/x86/kvm/ioapic.c
@@ -0,0 +1,400 @@
+/*
+ *  Copyright (C) 2001  MandrakeSoft S.A.
+ *
+ *    MandrakeSoft S.A.
+ *    43, rue d'Aboukir
+ *    75002 Paris - France
+ *    http://www.linux-mandrake.com/
+ *    http://www.mandrakesoft.com/
+ *
+ *  This library is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU Lesser General Public
+ *  License as published by the Free Software Foundation; either
+ *  version 2 of the License, or (at your option) any later version.
+ *
+ *  This library is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ *  Lesser General Public License for more details.
+ *
+ *  You should have received a copy of the GNU Lesser General Public
+ *  License along with this library; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ *  Yunhong Jiang <yunhong.jiang@intel.com>
+ *  Yaozu (Eddie) Dong <eddie.dong@intel.com>
+ *  Based on Xen 3.1 code.
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/kvm.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/smp.h>
+#include <linux/hrtimer.h>
+#include <linux/io.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+#include <asm/current.h>
+#include "irq.h"
+#if 0
+#define ioapic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg)
+#else
+#define ioapic_debug(fmt, arg...)
+#endif
+static void ioapic_deliver(struct kvm_ioapic *vioapic, int irq);
+
+static unsigned long ioapic_read_indirect(struct kvm_ioapic *ioapic,
+					  unsigned long addr,
+					  unsigned long length)
+{
+	unsigned long result = 0;
+
+	switch (ioapic->ioregsel) {
+	case IOAPIC_REG_VERSION:
+		result = ((((IOAPIC_NUM_PINS - 1) & 0xff) << 16)
+			  | (IOAPIC_VERSION_ID & 0xff));
+		break;
+
+	case IOAPIC_REG_APIC_ID:
+	case IOAPIC_REG_ARB_ID:
+		result = ((ioapic->id & 0xf) << 24);
+		break;
+
+	default:
+		{
+			u32 redir_index = (ioapic->ioregsel - 0x10) >> 1;
+			u64 redir_content;
+
+			ASSERT(redir_index < IOAPIC_NUM_PINS);
+
+			redir_content = ioapic->redirtbl[redir_index].bits;
+			result = (ioapic->ioregsel & 0x1) ?
+			    (redir_content >> 32) & 0xffffffff :
+			    redir_content & 0xffffffff;
+			break;
+		}
+	}
+
+	return result;
+}
+
+static void ioapic_service(struct kvm_ioapic *ioapic, unsigned int idx)
+{
+	union ioapic_redir_entry *pent;
+
+	pent = &ioapic->redirtbl[idx];
+
+	if (!pent->fields.mask) {
+		ioapic_deliver(ioapic, idx);
+		if (pent->fields.trig_mode == IOAPIC_LEVEL_TRIG)
+			pent->fields.remote_irr = 1;
+	}
+	if (!pent->fields.trig_mode)
+		ioapic->irr &= ~(1 << idx);
+}
+
+static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val)
+{
+	unsigned index;
+
+	switch (ioapic->ioregsel) {
+	case IOAPIC_REG_VERSION:
+		/* Writes are ignored. */
+		break;
+
+	case IOAPIC_REG_APIC_ID:
+		ioapic->id = (val >> 24) & 0xf;
+		break;
+
+	case IOAPIC_REG_ARB_ID:
+		break;
+
+	default:
+		index = (ioapic->ioregsel - 0x10) >> 1;
+
+		ioapic_debug("change redir index %x val %x\n", index, val);
+		if (index >= IOAPIC_NUM_PINS)
+			return;
+		if (ioapic->ioregsel & 1) {
+			ioapic->redirtbl[index].bits &= 0xffffffff;
+			ioapic->redirtbl[index].bits |= (u64) val << 32;
+		} else {
+			ioapic->redirtbl[index].bits &= ~0xffffffffULL;
+			ioapic->redirtbl[index].bits |= (u32) val;
+			ioapic->redirtbl[index].fields.remote_irr = 0;
+		}
+		if (ioapic->irr & (1 << index))
+			ioapic_service(ioapic, index);
+		break;
+	}
+}
+
+static void ioapic_inj_irq(struct kvm_ioapic *ioapic,
+			   struct kvm_vcpu *vcpu,
+			   u8 vector, u8 trig_mode, u8 delivery_mode)
+{
+	ioapic_debug("irq %d trig %d deliv %d\n", vector, trig_mode,
+		     delivery_mode);
+
+	ASSERT((delivery_mode == IOAPIC_FIXED) ||
+	       (delivery_mode == IOAPIC_LOWEST_PRIORITY));
+
+	kvm_apic_set_irq(vcpu, vector, trig_mode);
+}
+
+static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
+				       u8 dest_mode)
+{
+	u32 mask = 0;
+	int i;
+	struct kvm *kvm = ioapic->kvm;
+	struct kvm_vcpu *vcpu;
+
+	ioapic_debug("dest %d dest_mode %d\n", dest, dest_mode);
+
+	if (dest_mode == 0) {	/* Physical mode. */
+		if (dest == 0xFF) {	/* Broadcast. */
+			for (i = 0; i < KVM_MAX_VCPUS; ++i)
+				if (kvm->vcpus[i] && kvm->vcpus[i]->arch.apic)
+					mask |= 1 << i;
+			return mask;
+		}
+		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+			vcpu = kvm->vcpus[i];
+			if (!vcpu)
+				continue;
+			if (kvm_apic_match_physical_addr(vcpu->arch.apic, dest)) {
+				if (vcpu->arch.apic)
+					mask = 1 << i;
+				break;
+			}
+		}
+	} else if (dest != 0)	/* Logical mode, MDA non-zero. */
+		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+			vcpu = kvm->vcpus[i];
+			if (!vcpu)
+				continue;
+			if (vcpu->arch.apic &&
+			    kvm_apic_match_logical_addr(vcpu->arch.apic, dest))
+				mask |= 1 << vcpu->vcpu_id;
+		}
+	ioapic_debug("mask %x\n", mask);
+	return mask;
+}
+
+static void ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
+{
+	u8 dest = ioapic->redirtbl[irq].fields.dest_id;
+	u8 dest_mode = ioapic->redirtbl[irq].fields.dest_mode;
+	u8 delivery_mode = ioapic->redirtbl[irq].fields.delivery_mode;
+	u8 vector = ioapic->redirtbl[irq].fields.vector;
+	u8 trig_mode = ioapic->redirtbl[irq].fields.trig_mode;
+	u32 deliver_bitmask;
+	struct kvm_vcpu *vcpu;
+	int vcpu_id;
+
+	ioapic_debug("dest=%x dest_mode=%x delivery_mode=%x "
+		     "vector=%x trig_mode=%x\n",
+		     dest, dest_mode, delivery_mode, vector, trig_mode);
+
+	deliver_bitmask = ioapic_get_delivery_bitmask(ioapic, dest, dest_mode);
+	if (!deliver_bitmask) {
+		ioapic_debug("no target on destination\n");
+		return;
+	}
+
+	switch (delivery_mode) {
+	case IOAPIC_LOWEST_PRIORITY:
+		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm, vector,
+				deliver_bitmask);
+		if (vcpu != NULL)
+			ioapic_inj_irq(ioapic, vcpu, vector,
+				       trig_mode, delivery_mode);
+		else
+			ioapic_debug("null lowest prio vcpu: "
+				     "mask=%x vector=%x delivery_mode=%x\n",
+				     deliver_bitmask, vector, IOAPIC_LOWEST_PRIORITY);
+		break;
+	case IOAPIC_FIXED:
+		for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) {
+			if (!(deliver_bitmask & (1 << vcpu_id)))
+				continue;
+			deliver_bitmask &= ~(1 << vcpu_id);
+			vcpu = ioapic->kvm->vcpus[vcpu_id];
+			if (vcpu) {
+				ioapic_inj_irq(ioapic, vcpu, vector,
+					       trig_mode, delivery_mode);
+			}
+		}
+		break;
+
+		/* TODO: NMI */
+	default:
+		printk(KERN_WARNING "Unsupported delivery mode %d\n",
+		       delivery_mode);
+		break;
+	}
+}
+
+void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
+{
+	u32 old_irr = ioapic->irr;
+	u32 mask = 1 << irq;
+	union ioapic_redir_entry entry;
+
+	if (irq >= 0 && irq < IOAPIC_NUM_PINS) {
+		entry = ioapic->redirtbl[irq];
+		level ^= entry.fields.polarity;
+		if (!level)
+			ioapic->irr &= ~mask;
+		else {
+			ioapic->irr |= mask;
+			if ((!entry.fields.trig_mode && old_irr != ioapic->irr)
+			    || !entry.fields.remote_irr)
+				ioapic_service(ioapic, irq);
+		}
+	}
+}
+
+static int get_eoi_gsi(struct kvm_ioapic *ioapic, int vector)
+{
+	int i;
+
+	for (i = 0; i < IOAPIC_NUM_PINS; i++)
+		if (ioapic->redirtbl[i].fields.vector == vector)
+			return i;
+	return -1;
+}
+
+void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
+{
+	struct kvm_ioapic *ioapic = kvm->arch.vioapic;
+	union ioapic_redir_entry *ent;
+	int gsi;
+
+	gsi = get_eoi_gsi(ioapic, vector);
+	if (gsi == -1) {
+		printk(KERN_WARNING "Can't find redir item for %d EOI\n",
+		       vector);
+		return;
+	}
+
+	ent = &ioapic->redirtbl[gsi];
+	ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
+
+	ent->fields.remote_irr = 0;
+	if (!ent->fields.mask && (ioapic->irr & (1 << gsi)))
+		ioapic_deliver(ioapic, gsi);
+}
+
+static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr)
+{
+	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
+
+	return ((addr >= ioapic->base_address &&
+		 (addr < ioapic->base_address + IOAPIC_MEM_LENGTH)));
+}
+
+static void ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len,
+			     void *val)
+{
+	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
+	u32 result;
+
+	ioapic_debug("addr %lx\n", (unsigned long)addr);
+	ASSERT(!(addr & 0xf));	/* check alignment */
+
+	addr &= 0xff;
+	switch (addr) {
+	case IOAPIC_REG_SELECT:
+		result = ioapic->ioregsel;
+		break;
+
+	case IOAPIC_REG_WINDOW:
+		result = ioapic_read_indirect(ioapic, addr, len);
+		break;
+
+	default:
+		result = 0;
+		break;
+	}
+	switch (len) {
+	case 8:
+		*(u64 *) val = result;
+		break;
+	case 1:
+	case 2:
+	case 4:
+		memcpy(val, (char *)&result, len);
+		break;
+	default:
+		printk(KERN_WARNING "ioapic: wrong length %d\n", len);
+	}
+}
+
+static void ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len,
+			      const void *val)
+{
+	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
+	u32 data;
+
+	ioapic_debug("ioapic_mmio_write addr=%p len=%d val=%p\n",
+		     (void*)addr, len, val);
+	ASSERT(!(addr & 0xf));	/* check alignment */
+	if (len == 4 || len == 8)
+		data = *(u32 *) val;
+	else {
+		printk(KERN_WARNING "ioapic: Unsupported size %d\n", len);
+		return;
+	}
+
+	addr &= 0xff;
+	switch (addr) {
+	case IOAPIC_REG_SELECT:
+		ioapic->ioregsel = data;
+		break;
+
+	case IOAPIC_REG_WINDOW:
+		ioapic_write_indirect(ioapic, data);
+		break;
+#ifdef	CONFIG_IA64
+	case IOAPIC_REG_EOI:
+		kvm_ioapic_update_eoi(ioapic, data);
+		break;
+#endif
+
+	default:
+		break;
+	}
+}
+
+void kvm_ioapic_reset(struct kvm_ioapic *ioapic)
+{
+	int i;
+
+	for (i = 0; i < IOAPIC_NUM_PINS; i++)
+		ioapic->redirtbl[i].fields.mask = 1;
+	ioapic->base_address = IOAPIC_DEFAULT_BASE_ADDRESS;
+	ioapic->ioregsel = 0;
+	ioapic->irr = 0;
+	ioapic->id = 0;
+}
+
+int kvm_ioapic_init(struct kvm *kvm)
+{
+	struct kvm_ioapic *ioapic;
+
+	ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL);
+	if (!ioapic)
+		return -ENOMEM;
+	kvm->arch.vioapic = ioapic;
+	kvm_ioapic_reset(ioapic);
+	ioapic->dev.read = ioapic_mmio_read;
+	ioapic->dev.write = ioapic_mmio_write;
+	ioapic->dev.in_range = ioapic_in_range;
+	ioapic->dev.private = ioapic;
+	ioapic->kvm = kvm;
+	kvm_io_bus_register_dev(&kvm->mmio_bus, &ioapic->dev);
+	return 0;
+}
diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
new file mode 100644
index 000000000000..07a09aad4fd6
--- /dev/null
+++ b/arch/x86/kvm/irq.c
@@ -0,0 +1,98 @@
+/*
+ * irq.c: API for in kernel interrupt controller
+ * Copyright (c) 2007, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ * Authors:
+ *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kvm_host.h>
+
+#include "irq.h"
+
+/*
+ * check if there is pending interrupt without
+ * intack.
+ */
+int kvm_cpu_has_interrupt(struct kvm_vcpu *v)
+{
+	struct kvm_pic *s;
+
+	if (kvm_apic_has_interrupt(v) == -1) {	/* LAPIC */
+		if (kvm_apic_accept_pic_intr(v)) {
+			s = pic_irqchip(v->kvm);	/* PIC */
+			return s->output;
+		} else
+			return 0;
+	}
+	return 1;
+}
+EXPORT_SYMBOL_GPL(kvm_cpu_has_interrupt);
+
+/*
+ * Read pending interrupt vector and intack.
+ */
+int kvm_cpu_get_interrupt(struct kvm_vcpu *v)
+{
+	struct kvm_pic *s;
+	int vector;
+
+	vector = kvm_get_apic_interrupt(v);	/* APIC */
+	if (vector == -1) {
+		if (kvm_apic_accept_pic_intr(v)) {
+			s = pic_irqchip(v->kvm);
+			s->output = 0;		/* PIC */
+			vector = kvm_pic_read_irq(s);
+		}
+	}
+	return vector;
+}
+EXPORT_SYMBOL_GPL(kvm_cpu_get_interrupt);
+
+static void vcpu_kick_intr(void *info)
+{
+#ifdef DEBUG
+	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
+	printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
+#endif
+}
+
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
+{
+	int ipi_pcpu = vcpu->cpu;
+
+	if (waitqueue_active(&vcpu->wq)) {
+		wake_up_interruptible(&vcpu->wq);
+		++vcpu->stat.halt_wakeup;
+	}
+	if (vcpu->guest_mode)
+		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
+}
+
+void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu)
+{
+	kvm_inject_apic_timer_irqs(vcpu);
+	/* TODO: PIT, RTC etc. */
+}
+EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs);
+
+void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
+{
+	kvm_apic_timer_intr_post(vcpu, vec);
+	/* TODO: PIT, RTC etc. */
+}
+EXPORT_SYMBOL_GPL(kvm_timer_intr_post);
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
new file mode 100644
index 000000000000..6316638eec9f
--- /dev/null
+++ b/arch/x86/kvm/irq.h
@@ -0,0 +1,195 @@
+/*
+ * irq.h: in kernel interrupt controller related definitions
+ * Copyright (c) 2007, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ * Authors:
+ *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
+ *
+ */
+
+#ifndef __IRQ_H
+#define __IRQ_H
+
+#include <linux/mm_types.h>
+#include <linux/hrtimer.h>
+#include <linux/kvm_host.h>
+#include "iodev.h"
+
+struct kvm;
+struct kvm_vcpu;
+
+typedef void irq_request_func(void *opaque, int level);
+
+struct kvm_kpic_state {
+	u8 last_irr;	/* edge detection */
+	u8 irr;		/* interrupt request register */
+	u8 imr;		/* interrupt mask register */
+	u8 isr;		/* interrupt service register */
+	u8 priority_add;	/* highest irq priority */
+	u8 irq_base;
+	u8 read_reg_select;
+	u8 poll;
+	u8 special_mask;
+	u8 init_state;
+	u8 auto_eoi;
+	u8 rotate_on_auto_eoi;
+	u8 special_fully_nested_mode;
+	u8 init4;		/* true if 4 byte init */
+	u8 elcr;		/* PIIX edge/trigger selection */
+	u8 elcr_mask;
+	struct kvm_pic *pics_state;
+};
+
+struct kvm_pic {
+	struct kvm_kpic_state pics[2]; /* 0 is master pic, 1 is slave pic */
+	irq_request_func *irq_request;
+	void *irq_request_opaque;
+	int output;		/* intr from master PIC */
+	struct kvm_io_device dev;
+};
+
+struct kvm_pic *kvm_create_pic(struct kvm *kvm);
+void kvm_pic_set_irq(void *opaque, int irq, int level);
+int kvm_pic_read_irq(struct kvm_pic *s);
+void kvm_pic_update_irq(struct kvm_pic *s);
+
+#define IOAPIC_NUM_PINS  KVM_IOAPIC_NUM_PINS
+#define IOAPIC_VERSION_ID 0x11	/* IOAPIC version */
+#define IOAPIC_EDGE_TRIG  0
+#define IOAPIC_LEVEL_TRIG 1
+
+#define IOAPIC_DEFAULT_BASE_ADDRESS  0xfec00000
+#define IOAPIC_MEM_LENGTH            0x100
+
+/* Direct registers. */
+#define IOAPIC_REG_SELECT  0x00
+#define IOAPIC_REG_WINDOW  0x10
+#define IOAPIC_REG_EOI     0x40	/* IA64 IOSAPIC only */
+
+/* Indirect registers. */
+#define IOAPIC_REG_APIC_ID 0x00	/* x86 IOAPIC only */
+#define IOAPIC_REG_VERSION 0x01
+#define IOAPIC_REG_ARB_ID  0x02	/* x86 IOAPIC only */
+
+/*ioapic delivery mode*/
+#define	IOAPIC_FIXED			0x0
+#define	IOAPIC_LOWEST_PRIORITY		0x1
+#define	IOAPIC_PMI			0x2
+#define	IOAPIC_NMI			0x4
+#define	IOAPIC_INIT			0x5
+#define	IOAPIC_EXTINT			0x7
+
+struct kvm_ioapic {
+	u64 base_address;
+	u32 ioregsel;
+	u32 id;
+	u32 irr;
+	u32 pad;
+	union ioapic_redir_entry {
+		u64 bits;
+		struct {
+			u8 vector;
+			u8 delivery_mode:3;
+			u8 dest_mode:1;
+			u8 delivery_status:1;
+			u8 polarity:1;
+			u8 remote_irr:1;
+			u8 trig_mode:1;
+			u8 mask:1;
+			u8 reserve:7;
+			u8 reserved[4];
+			u8 dest_id;
+		} fields;
+	} redirtbl[IOAPIC_NUM_PINS];
+	struct kvm_io_device dev;
+	struct kvm *kvm;
+};
+
+struct kvm_lapic {
+	unsigned long base_address;
+	struct kvm_io_device dev;
+	struct {
+		atomic_t pending;
+		s64 period;	/* unit: ns */
+		u32 divide_count;
+		ktime_t last_update;
+		struct hrtimer dev;
+	} timer;
+	struct kvm_vcpu *vcpu;
+	struct page *regs_page;
+	void *regs;
+};
+
+#ifdef DEBUG
+#define ASSERT(x)  							\
+do {									\
+	if (!(x)) {							\
+		printk(KERN_EMERG "assertion failed %s: %d: %s\n",	\
+		       __FILE__, __LINE__, #x);				\
+		BUG();							\
+	}								\
+} while (0)
+#else
+#define ASSERT(x) do { } while (0)
+#endif
+
+static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
+{
+	return kvm->arch.vpic;
+}
+
+static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
+{
+	return kvm->arch.vioapic;
+}
+
+static inline int irqchip_in_kernel(struct kvm *kvm)
+{
+	return pic_irqchip(kvm) != NULL;
+}
+
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
+int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu);
+int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu);
+int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu);
+int kvm_create_lapic(struct kvm_vcpu *vcpu);
+void kvm_lapic_reset(struct kvm_vcpu *vcpu);
+void kvm_pic_reset(struct kvm_kpic_state *s);
+void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
+void kvm_free_lapic(struct kvm_vcpu *vcpu);
+u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu);
+void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8);
+void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value);
+
+struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
+				       unsigned long bitmap);
+u64 kvm_get_apic_base(struct kvm_vcpu *vcpu);
+void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data);
+int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest);
+void kvm_ioapic_update_eoi(struct kvm *kvm, int vector);
+int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda);
+int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig);
+void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu);
+int kvm_ioapic_init(struct kvm *kvm);
+void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
+int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
+int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
+void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
+void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
+void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
+void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
+void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
+
+#endif
diff --git a/arch/x86/kvm/kvm_svm.h b/arch/x86/kvm/kvm_svm.h
new file mode 100644
index 000000000000..ecdfe97e4635
--- /dev/null
+++ b/arch/x86/kvm/kvm_svm.h
@@ -0,0 +1,45 @@
+#ifndef __KVM_SVM_H
+#define __KVM_SVM_H
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/kvm_host.h>
+#include <asm/msr.h>
+
+#include "svm.h"
+
+static const u32 host_save_user_msrs[] = {
+#ifdef CONFIG_X86_64
+	MSR_STAR, MSR_LSTAR, MSR_CSTAR, MSR_SYSCALL_MASK, MSR_KERNEL_GS_BASE,
+	MSR_FS_BASE,
+#endif
+	MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
+};
+
+#define NR_HOST_SAVE_USER_MSRS ARRAY_SIZE(host_save_user_msrs)
+#define NUM_DB_REGS 4
+
+struct kvm_vcpu;
+
+struct vcpu_svm {
+	struct kvm_vcpu vcpu;
+	struct vmcb *vmcb;
+	unsigned long vmcb_pa;
+	struct svm_cpu_data *svm_data;
+	uint64_t asid_generation;
+
+	unsigned long db_regs[NUM_DB_REGS];
+
+	u64 next_rip;
+
+	u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS];
+	u64 host_gs_base;
+	unsigned long host_cr2;
+	unsigned long host_db_regs[NUM_DB_REGS];
+	unsigned long host_dr6;
+	unsigned long host_dr7;
+};
+
+#endif
+
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
new file mode 100644
index 000000000000..4076331b01ee
--- /dev/null
+++ b/arch/x86/kvm/lapic.c
@@ -0,0 +1,1085 @@
+
+/*
+ * Local APIC virtualization
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ * Copyright (C) 2007 Novell
+ * Copyright (C) 2007 Intel
+ *
+ * Authors:
+ *   Dor Laor <dor.laor@qumranet.com>
+ *   Gregory Haskins <ghaskins@novell.com>
+ *   Yaozu (Eddie) Dong <eddie.dong@intel.com>
+ *
+ * Based on Xen 3.1 code, Copyright (c) 2004, Intel Corporation.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/kvm.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/smp.h>
+#include <linux/hrtimer.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/page.h>
+#include <asm/current.h>
+#include <asm/apicdef.h>
+#include <asm/atomic.h>
+#include <asm/div64.h>
+#include "irq.h"
+
+#define PRId64 "d"
+#define PRIx64 "llx"
+#define PRIu64 "u"
+#define PRIo64 "o"
+
+#define APIC_BUS_CYCLE_NS 1
+
+/* #define apic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg) */
+#define apic_debug(fmt, arg...)
+
+#define APIC_LVT_NUM			6
+/* 14 is the version for Xeon and Pentium 8.4.8*/
+#define APIC_VERSION			(0x14UL | ((APIC_LVT_NUM - 1) << 16))
+#define LAPIC_MMIO_LENGTH		(1 << 12)
+/* followed define is not in apicdef.h */
+#define APIC_SHORT_MASK			0xc0000
+#define APIC_DEST_NOSHORT		0x0
+#define APIC_DEST_MASK			0x800
+#define MAX_APIC_VECTOR			256
+
+#define VEC_POS(v) ((v) & (32 - 1))
+#define REG_POS(v) (((v) >> 5) << 4)
+
+static inline u32 apic_get_reg(struct kvm_lapic *apic, int reg_off)
+{
+	return *((u32 *) (apic->regs + reg_off));
+}
+
+static inline void apic_set_reg(struct kvm_lapic *apic, int reg_off, u32 val)
+{
+	*((u32 *) (apic->regs + reg_off)) = val;
+}
+
+static inline int apic_test_and_set_vector(int vec, void *bitmap)
+{
+	return test_and_set_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
+}
+
+static inline int apic_test_and_clear_vector(int vec, void *bitmap)
+{
+	return test_and_clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
+}
+
+static inline void apic_set_vector(int vec, void *bitmap)
+{
+	set_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
+}
+
+static inline void apic_clear_vector(int vec, void *bitmap)
+{
+	clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
+}
+
+static inline int apic_hw_enabled(struct kvm_lapic *apic)
+{
+	return (apic)->vcpu->arch.apic_base & MSR_IA32_APICBASE_ENABLE;
+}
+
+static inline int  apic_sw_enabled(struct kvm_lapic *apic)
+{
+	return apic_get_reg(apic, APIC_SPIV) & APIC_SPIV_APIC_ENABLED;
+}
+
+static inline int apic_enabled(struct kvm_lapic *apic)
+{
+	return apic_sw_enabled(apic) &&	apic_hw_enabled(apic);
+}
+
+#define LVT_MASK	\
+	(APIC_LVT_MASKED | APIC_SEND_PENDING | APIC_VECTOR_MASK)
+
+#define LINT_MASK	\
+	(LVT_MASK | APIC_MODE_MASK | APIC_INPUT_POLARITY | \
+	 APIC_LVT_REMOTE_IRR | APIC_LVT_LEVEL_TRIGGER)
+
+static inline int kvm_apic_id(struct kvm_lapic *apic)
+{
+	return (apic_get_reg(apic, APIC_ID) >> 24) & 0xff;
+}
+
+static inline int apic_lvt_enabled(struct kvm_lapic *apic, int lvt_type)
+{
+	return !(apic_get_reg(apic, lvt_type) & APIC_LVT_MASKED);
+}
+
+static inline int apic_lvt_vector(struct kvm_lapic *apic, int lvt_type)
+{
+	return apic_get_reg(apic, lvt_type) & APIC_VECTOR_MASK;
+}
+
+static inline int apic_lvtt_period(struct kvm_lapic *apic)
+{
+	return apic_get_reg(apic, APIC_LVTT) & APIC_LVT_TIMER_PERIODIC;
+}
+
+static unsigned int apic_lvt_mask[APIC_LVT_NUM] = {
+	LVT_MASK | APIC_LVT_TIMER_PERIODIC,	/* LVTT */
+	LVT_MASK | APIC_MODE_MASK,	/* LVTTHMR */
+	LVT_MASK | APIC_MODE_MASK,	/* LVTPC */
+	LINT_MASK, LINT_MASK,	/* LVT0-1 */
+	LVT_MASK		/* LVTERR */
+};
+
+static int find_highest_vector(void *bitmap)
+{
+	u32 *word = bitmap;
+	int word_offset = MAX_APIC_VECTOR >> 5;
+
+	while ((word_offset != 0) && (word[(--word_offset) << 2] == 0))
+		continue;
+
+	if (likely(!word_offset && !word[0]))
+		return -1;
+	else
+		return fls(word[word_offset << 2]) - 1 + (word_offset << 5);
+}
+
+static inline int apic_test_and_set_irr(int vec, struct kvm_lapic *apic)
+{
+	return apic_test_and_set_vector(vec, apic->regs + APIC_IRR);
+}
+
+static inline void apic_clear_irr(int vec, struct kvm_lapic *apic)
+{
+	apic_clear_vector(vec, apic->regs + APIC_IRR);
+}
+
+static inline int apic_find_highest_irr(struct kvm_lapic *apic)
+{
+	int result;
+
+	result = find_highest_vector(apic->regs + APIC_IRR);
+	ASSERT(result == -1 || result >= 16);
+
+	return result;
+}
+
+int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	int highest_irr;
+
+	if (!apic)
+		return 0;
+	highest_irr = apic_find_highest_irr(apic);
+
+	return highest_irr;
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_find_highest_irr);
+
+int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (!apic_test_and_set_irr(vec, apic)) {
+		/* a new pending irq is set in IRR */
+		if (trig)
+			apic_set_vector(vec, apic->regs + APIC_TMR);
+		else
+			apic_clear_vector(vec, apic->regs + APIC_TMR);
+		kvm_vcpu_kick(apic->vcpu);
+		return 1;
+	}
+	return 0;
+}
+
+static inline int apic_find_highest_isr(struct kvm_lapic *apic)
+{
+	int result;
+
+	result = find_highest_vector(apic->regs + APIC_ISR);
+	ASSERT(result == -1 || result >= 16);
+
+	return result;
+}
+
+static void apic_update_ppr(struct kvm_lapic *apic)
+{
+	u32 tpr, isrv, ppr;
+	int isr;
+
+	tpr = apic_get_reg(apic, APIC_TASKPRI);
+	isr = apic_find_highest_isr(apic);
+	isrv = (isr != -1) ? isr : 0;
+
+	if ((tpr & 0xf0) >= (isrv & 0xf0))
+		ppr = tpr & 0xff;
+	else
+		ppr = isrv & 0xf0;
+
+	apic_debug("vlapic %p, ppr 0x%x, isr 0x%x, isrv 0x%x",
+		   apic, ppr, isr, isrv);
+
+	apic_set_reg(apic, APIC_PROCPRI, ppr);
+}
+
+static void apic_set_tpr(struct kvm_lapic *apic, u32 tpr)
+{
+	apic_set_reg(apic, APIC_TASKPRI, tpr);
+	apic_update_ppr(apic);
+}
+
+int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest)
+{
+	return kvm_apic_id(apic) == dest;
+}
+
+int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda)
+{
+	int result = 0;
+	u8 logical_id;
+
+	logical_id = GET_APIC_LOGICAL_ID(apic_get_reg(apic, APIC_LDR));
+
+	switch (apic_get_reg(apic, APIC_DFR)) {
+	case APIC_DFR_FLAT:
+		if (logical_id & mda)
+			result = 1;
+		break;
+	case APIC_DFR_CLUSTER:
+		if (((logical_id >> 4) == (mda >> 0x4))
+		    && (logical_id & mda & 0xf))
+			result = 1;
+		break;
+	default:
+		printk(KERN_WARNING "Bad DFR vcpu %d: %08x\n",
+		       apic->vcpu->vcpu_id, apic_get_reg(apic, APIC_DFR));
+		break;
+	}
+
+	return result;
+}
+
+static int apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source,
+			   int short_hand, int dest, int dest_mode)
+{
+	int result = 0;
+	struct kvm_lapic *target = vcpu->arch.apic;
+
+	apic_debug("target %p, source %p, dest 0x%x, "
+		   "dest_mode 0x%x, short_hand 0x%x",
+		   target, source, dest, dest_mode, short_hand);
+
+	ASSERT(!target);
+	switch (short_hand) {
+	case APIC_DEST_NOSHORT:
+		if (dest_mode == 0) {
+			/* Physical mode. */
+			if ((dest == 0xFF) || (dest == kvm_apic_id(target)))
+				result = 1;
+		} else
+			/* Logical mode. */
+			result = kvm_apic_match_logical_addr(target, dest);
+		break;
+	case APIC_DEST_SELF:
+		if (target == source)
+			result = 1;
+		break;
+	case APIC_DEST_ALLINC:
+		result = 1;
+		break;
+	case APIC_DEST_ALLBUT:
+		if (target != source)
+			result = 1;
+		break;
+	default:
+		printk(KERN_WARNING "Bad dest shorthand value %x\n",
+		       short_hand);
+		break;
+	}
+
+	return result;
+}
+
+/*
+ * Add a pending IRQ into lapic.
+ * Return 1 if successfully added and 0 if discarded.
+ */
+static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
+			     int vector, int level, int trig_mode)
+{
+	int orig_irr, result = 0;
+	struct kvm_vcpu *vcpu = apic->vcpu;
+
+	switch (delivery_mode) {
+	case APIC_DM_FIXED:
+	case APIC_DM_LOWEST:
+		/* FIXME add logic for vcpu on reset */
+		if (unlikely(!apic_enabled(apic)))
+			break;
+
+		orig_irr = apic_test_and_set_irr(vector, apic);
+		if (orig_irr && trig_mode) {
+			apic_debug("level trig mode repeatedly for vector %d",
+				   vector);
+			break;
+		}
+
+		if (trig_mode) {
+			apic_debug("level trig mode for vector %d", vector);
+			apic_set_vector(vector, apic->regs + APIC_TMR);
+		} else
+			apic_clear_vector(vector, apic->regs + APIC_TMR);
+
+		if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
+			kvm_vcpu_kick(vcpu);
+		else if (vcpu->arch.mp_state == VCPU_MP_STATE_HALTED) {
+			vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
+			if (waitqueue_active(&vcpu->wq))
+				wake_up_interruptible(&vcpu->wq);
+		}
+
+		result = (orig_irr == 0);
+		break;
+
+	case APIC_DM_REMRD:
+		printk(KERN_DEBUG "Ignoring delivery mode 3\n");
+		break;
+
+	case APIC_DM_SMI:
+		printk(KERN_DEBUG "Ignoring guest SMI\n");
+		break;
+	case APIC_DM_NMI:
+		printk(KERN_DEBUG "Ignoring guest NMI\n");
+		break;
+
+	case APIC_DM_INIT:
+		if (level) {
+			if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
+				printk(KERN_DEBUG
+				       "INIT on a runnable vcpu %d\n",
+				       vcpu->vcpu_id);
+			vcpu->arch.mp_state = VCPU_MP_STATE_INIT_RECEIVED;
+			kvm_vcpu_kick(vcpu);
+		} else {
+			printk(KERN_DEBUG
+			       "Ignoring de-assert INIT to vcpu %d\n",
+			       vcpu->vcpu_id);
+		}
+
+		break;
+
+	case APIC_DM_STARTUP:
+		printk(KERN_DEBUG "SIPI to vcpu %d vector 0x%02x\n",
+		       vcpu->vcpu_id, vector);
+		if (vcpu->arch.mp_state == VCPU_MP_STATE_INIT_RECEIVED) {
+			vcpu->arch.sipi_vector = vector;
+			vcpu->arch.mp_state = VCPU_MP_STATE_SIPI_RECEIVED;
+			if (waitqueue_active(&vcpu->wq))
+				wake_up_interruptible(&vcpu->wq);
+		}
+		break;
+
+	default:
+		printk(KERN_ERR "TODO: unsupported delivery mode %x\n",
+		       delivery_mode);
+		break;
+	}
+	return result;
+}
+
+static struct kvm_lapic *kvm_apic_round_robin(struct kvm *kvm, u8 vector,
+				       unsigned long bitmap)
+{
+	int last;
+	int next;
+	struct kvm_lapic *apic = NULL;
+
+	last = kvm->arch.round_robin_prev_vcpu;
+	next = last;
+
+	do {
+		if (++next == KVM_MAX_VCPUS)
+			next = 0;
+		if (kvm->vcpus[next] == NULL || !test_bit(next, &bitmap))
+			continue;
+		apic = kvm->vcpus[next]->arch.apic;
+		if (apic && apic_enabled(apic))
+			break;
+		apic = NULL;
+	} while (next != last);
+	kvm->arch.round_robin_prev_vcpu = next;
+
+	if (!apic)
+		printk(KERN_DEBUG "vcpu not ready for apic_round_robin\n");
+
+	return apic;
+}
+
+struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
+		unsigned long bitmap)
+{
+	struct kvm_lapic *apic;
+
+	apic = kvm_apic_round_robin(kvm, vector, bitmap);
+	if (apic)
+		return apic->vcpu;
+	return NULL;
+}
+
+static void apic_set_eoi(struct kvm_lapic *apic)
+{
+	int vector = apic_find_highest_isr(apic);
+
+	/*
+	 * Not every write EOI will has corresponding ISR,
+	 * one example is when Kernel check timer on setup_IO_APIC
+	 */
+	if (vector == -1)
+		return;
+
+	apic_clear_vector(vector, apic->regs + APIC_ISR);
+	apic_update_ppr(apic);
+
+	if (apic_test_and_clear_vector(vector, apic->regs + APIC_TMR))
+		kvm_ioapic_update_eoi(apic->vcpu->kvm, vector);
+}
+
+static void apic_send_ipi(struct kvm_lapic *apic)
+{
+	u32 icr_low = apic_get_reg(apic, APIC_ICR);
+	u32 icr_high = apic_get_reg(apic, APIC_ICR2);
+
+	unsigned int dest = GET_APIC_DEST_FIELD(icr_high);
+	unsigned int short_hand = icr_low & APIC_SHORT_MASK;
+	unsigned int trig_mode = icr_low & APIC_INT_LEVELTRIG;
+	unsigned int level = icr_low & APIC_INT_ASSERT;
+	unsigned int dest_mode = icr_low & APIC_DEST_MASK;
+	unsigned int delivery_mode = icr_low & APIC_MODE_MASK;
+	unsigned int vector = icr_low & APIC_VECTOR_MASK;
+
+	struct kvm_vcpu *target;
+	struct kvm_vcpu *vcpu;
+	unsigned long lpr_map = 0;
+	int i;
+
+	apic_debug("icr_high 0x%x, icr_low 0x%x, "
+		   "short_hand 0x%x, dest 0x%x, trig_mode 0x%x, level 0x%x, "
+		   "dest_mode 0x%x, delivery_mode 0x%x, vector 0x%x\n",
+		   icr_high, icr_low, short_hand, dest,
+		   trig_mode, level, dest_mode, delivery_mode, vector);
+
+	for (i = 0; i < KVM_MAX_VCPUS; i++) {
+		vcpu = apic->vcpu->kvm->vcpus[i];
+		if (!vcpu)
+			continue;
+
+		if (vcpu->arch.apic &&
+		    apic_match_dest(vcpu, apic, short_hand, dest, dest_mode)) {
+			if (delivery_mode == APIC_DM_LOWEST)
+				set_bit(vcpu->vcpu_id, &lpr_map);
+			else
+				__apic_accept_irq(vcpu->arch.apic, delivery_mode,
+						  vector, level, trig_mode);
+		}
+	}
+
+	if (delivery_mode == APIC_DM_LOWEST) {
+		target = kvm_get_lowest_prio_vcpu(vcpu->kvm, vector, lpr_map);
+		if (target != NULL)
+			__apic_accept_irq(target->arch.apic, delivery_mode,
+					  vector, level, trig_mode);
+	}
+}
+
+static u32 apic_get_tmcct(struct kvm_lapic *apic)
+{
+	u64 counter_passed;
+	ktime_t passed, now;
+	u32 tmcct;
+
+	ASSERT(apic != NULL);
+
+	now = apic->timer.dev.base->get_time();
+	tmcct = apic_get_reg(apic, APIC_TMICT);
+
+	/* if initial count is 0, current count should also be 0 */
+	if (tmcct == 0)
+		return 0;
+
+	if (unlikely(ktime_to_ns(now) <=
+		ktime_to_ns(apic->timer.last_update))) {
+		/* Wrap around */
+		passed = ktime_add(( {
+				    (ktime_t) {
+				    .tv64 = KTIME_MAX -
+				    (apic->timer.last_update).tv64}; }
+				   ), now);
+		apic_debug("time elapsed\n");
+	} else
+		passed = ktime_sub(now, apic->timer.last_update);
+
+	counter_passed = div64_64(ktime_to_ns(passed),
+				  (APIC_BUS_CYCLE_NS * apic->timer.divide_count));
+
+	if (counter_passed > tmcct) {
+		if (unlikely(!apic_lvtt_period(apic))) {
+			/* one-shot timers stick at 0 until reset */
+			tmcct = 0;
+		} else {
+			/*
+			 * periodic timers reset to APIC_TMICT when they
+			 * hit 0. The while loop simulates this happening N
+			 * times. (counter_passed %= tmcct) would also work,
+			 * but might be slower or not work on 32-bit??
+			 */
+			while (counter_passed > tmcct)
+				counter_passed -= tmcct;
+			tmcct -= counter_passed;
+		}
+	} else {
+		tmcct -= counter_passed;
+	}
+
+	return tmcct;
+}
+
+static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
+{
+	u32 val = 0;
+
+	if (offset >= LAPIC_MMIO_LENGTH)
+		return 0;
+
+	switch (offset) {
+	case APIC_ARBPRI:
+		printk(KERN_WARNING "Access APIC ARBPRI register "
+		       "which is for P6\n");
+		break;
+
+	case APIC_TMCCT:	/* Timer CCR */
+		val = apic_get_tmcct(apic);
+		break;
+
+	default:
+		apic_update_ppr(apic);
+		val = apic_get_reg(apic, offset);
+		break;
+	}
+
+	return val;
+}
+
+static void apic_mmio_read(struct kvm_io_device *this,
+			   gpa_t address, int len, void *data)
+{
+	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
+	unsigned int offset = address - apic->base_address;
+	unsigned char alignment = offset & 0xf;
+	u32 result;
+
+	if ((alignment + len) > 4) {
+		printk(KERN_ERR "KVM_APIC_READ: alignment error %lx %d",
+		       (unsigned long)address, len);
+		return;
+	}
+	result = __apic_read(apic, offset & ~0xf);
+
+	switch (len) {
+	case 1:
+	case 2:
+	case 4:
+		memcpy(data, (char *)&result + alignment, len);
+		break;
+	default:
+		printk(KERN_ERR "Local APIC read with len = %x, "
+		       "should be 1,2, or 4 instead\n", len);
+		break;
+	}
+}
+
+static void update_divide_count(struct kvm_lapic *apic)
+{
+	u32 tmp1, tmp2, tdcr;
+
+	tdcr = apic_get_reg(apic, APIC_TDCR);
+	tmp1 = tdcr & 0xf;
+	tmp2 = ((tmp1 & 0x3) | ((tmp1 & 0x8) >> 1)) + 1;
+	apic->timer.divide_count = 0x1 << (tmp2 & 0x7);
+
+	apic_debug("timer divide count is 0x%x\n",
+				   apic->timer.divide_count);
+}
+
+static void start_apic_timer(struct kvm_lapic *apic)
+{
+	ktime_t now = apic->timer.dev.base->get_time();
+
+	apic->timer.last_update = now;
+
+	apic->timer.period = apic_get_reg(apic, APIC_TMICT) *
+		    APIC_BUS_CYCLE_NS * apic->timer.divide_count;
+	atomic_set(&apic->timer.pending, 0);
+	hrtimer_start(&apic->timer.dev,
+		      ktime_add_ns(now, apic->timer.period),
+		      HRTIMER_MODE_ABS);
+
+	apic_debug("%s: bus cycle is %" PRId64 "ns, now 0x%016"
+			   PRIx64 ", "
+			   "timer initial count 0x%x, period %lldns, "
+			   "expire @ 0x%016" PRIx64 ".\n", __FUNCTION__,
+			   APIC_BUS_CYCLE_NS, ktime_to_ns(now),
+			   apic_get_reg(apic, APIC_TMICT),
+			   apic->timer.period,
+			   ktime_to_ns(ktime_add_ns(now,
+					apic->timer.period)));
+}
+
+static void apic_mmio_write(struct kvm_io_device *this,
+			    gpa_t address, int len, const void *data)
+{
+	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
+	unsigned int offset = address - apic->base_address;
+	unsigned char alignment = offset & 0xf;
+	u32 val;
+
+	/*
+	 * APIC register must be aligned on 128-bits boundary.
+	 * 32/64/128 bits registers must be accessed thru 32 bits.
+	 * Refer SDM 8.4.1
+	 */
+	if (len != 4 || alignment) {
+		if (printk_ratelimit())
+			printk(KERN_ERR "apic write: bad size=%d %lx\n",
+			       len, (long)address);
+		return;
+	}
+
+	val = *(u32 *) data;
+
+	/* too common printing */
+	if (offset != APIC_EOI)
+		apic_debug("%s: offset 0x%x with length 0x%x, and value is "
+			   "0x%x\n", __FUNCTION__, offset, len, val);
+
+	offset &= 0xff0;
+
+	switch (offset) {
+	case APIC_ID:		/* Local APIC ID */
+		apic_set_reg(apic, APIC_ID, val);
+		break;
+
+	case APIC_TASKPRI:
+		apic_set_tpr(apic, val & 0xff);
+		break;
+
+	case APIC_EOI:
+		apic_set_eoi(apic);
+		break;
+
+	case APIC_LDR:
+		apic_set_reg(apic, APIC_LDR, val & APIC_LDR_MASK);
+		break;
+
+	case APIC_DFR:
+		apic_set_reg(apic, APIC_DFR, val | 0x0FFFFFFF);
+		break;
+
+	case APIC_SPIV:
+		apic_set_reg(apic, APIC_SPIV, val & 0x3ff);
+		if (!(val & APIC_SPIV_APIC_ENABLED)) {
+			int i;
+			u32 lvt_val;
+
+			for (i = 0; i < APIC_LVT_NUM; i++) {
+				lvt_val = apic_get_reg(apic,
+						       APIC_LVTT + 0x10 * i);
+				apic_set_reg(apic, APIC_LVTT + 0x10 * i,
+					     lvt_val | APIC_LVT_MASKED);
+			}
+			atomic_set(&apic->timer.pending, 0);
+
+		}
+		break;
+
+	case APIC_ICR:
+		/* No delay here, so we always clear the pending bit */
+		apic_set_reg(apic, APIC_ICR, val & ~(1 << 12));
+		apic_send_ipi(apic);
+		break;
+
+	case APIC_ICR2:
+		apic_set_reg(apic, APIC_ICR2, val & 0xff000000);
+		break;
+
+	case APIC_LVTT:
+	case APIC_LVTTHMR:
+	case APIC_LVTPC:
+	case APIC_LVT0:
+	case APIC_LVT1:
+	case APIC_LVTERR:
+		/* TODO: Check vector */
+		if (!apic_sw_enabled(apic))
+			val |= APIC_LVT_MASKED;
+
+		val &= apic_lvt_mask[(offset - APIC_LVTT) >> 4];
+		apic_set_reg(apic, offset, val);
+
+		break;
+
+	case APIC_TMICT:
+		hrtimer_cancel(&apic->timer.dev);
+		apic_set_reg(apic, APIC_TMICT, val);
+		start_apic_timer(apic);
+		return;
+
+	case APIC_TDCR:
+		if (val & 4)
+			printk(KERN_ERR "KVM_WRITE:TDCR %x\n", val);
+		apic_set_reg(apic, APIC_TDCR, val);
+		update_divide_count(apic);
+		break;
+
+	default:
+		apic_debug("Local APIC Write to read-only register %x\n",
+			   offset);
+		break;
+	}
+
+}
+
+static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr)
+{
+	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
+	int ret = 0;
+
+
+	if (apic_hw_enabled(apic) &&
+	    (addr >= apic->base_address) &&
+	    (addr < (apic->base_address + LAPIC_MMIO_LENGTH)))
+		ret = 1;
+
+	return ret;
+}
+
+void kvm_free_lapic(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu->arch.apic)
+		return;
+
+	hrtimer_cancel(&vcpu->arch.apic->timer.dev);
+
+	if (vcpu->arch.apic->regs_page)
+		__free_page(vcpu->arch.apic->regs_page);
+
+	kfree(vcpu->arch.apic);
+}
+
+/*
+ *----------------------------------------------------------------------
+ * LAPIC interface
+ *----------------------------------------------------------------------
+ */
+
+void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (!apic)
+		return;
+	apic_set_tpr(apic, ((cr8 & 0x0f) << 4));
+}
+
+u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	u64 tpr;
+
+	if (!apic)
+		return 0;
+	tpr = (u64) apic_get_reg(apic, APIC_TASKPRI);
+
+	return (tpr & 0xf0) >> 4;
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_get_cr8);
+
+void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (!apic) {
+		value |= MSR_IA32_APICBASE_BSP;
+		vcpu->arch.apic_base = value;
+		return;
+	}
+	if (apic->vcpu->vcpu_id)
+		value &= ~MSR_IA32_APICBASE_BSP;
+
+	vcpu->arch.apic_base = value;
+	apic->base_address = apic->vcpu->arch.apic_base &
+			     MSR_IA32_APICBASE_BASE;
+
+	/* with FSB delivery interrupt, we can restart APIC functionality */
+	apic_debug("apic base msr is 0x%016" PRIx64 ", and base address is "
+		   "0x%lx.\n", apic->vcpu->arch.apic_base, apic->base_address);
+
+}
+
+u64 kvm_lapic_get_base(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.apic_base;
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_get_base);
+
+void kvm_lapic_reset(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic;
+	int i;
+
+	apic_debug("%s\n", __FUNCTION__);
+
+	ASSERT(vcpu);
+	apic = vcpu->arch.apic;
+	ASSERT(apic != NULL);
+
+	/* Stop the timer in case it's a reset to an active apic */
+	hrtimer_cancel(&apic->timer.dev);
+
+	apic_set_reg(apic, APIC_ID, vcpu->vcpu_id << 24);
+	apic_set_reg(apic, APIC_LVR, APIC_VERSION);
+
+	for (i = 0; i < APIC_LVT_NUM; i++)
+		apic_set_reg(apic, APIC_LVTT + 0x10 * i, APIC_LVT_MASKED);
+	apic_set_reg(apic, APIC_LVT0,
+		     SET_APIC_DELIVERY_MODE(0, APIC_MODE_EXTINT));
+
+	apic_set_reg(apic, APIC_DFR, 0xffffffffU);
+	apic_set_reg(apic, APIC_SPIV, 0xff);
+	apic_set_reg(apic, APIC_TASKPRI, 0);
+	apic_set_reg(apic, APIC_LDR, 0);
+	apic_set_reg(apic, APIC_ESR, 0);
+	apic_set_reg(apic, APIC_ICR, 0);
+	apic_set_reg(apic, APIC_ICR2, 0);
+	apic_set_reg(apic, APIC_TDCR, 0);
+	apic_set_reg(apic, APIC_TMICT, 0);
+	for (i = 0; i < 8; i++) {
+		apic_set_reg(apic, APIC_IRR + 0x10 * i, 0);
+		apic_set_reg(apic, APIC_ISR + 0x10 * i, 0);
+		apic_set_reg(apic, APIC_TMR + 0x10 * i, 0);
+	}
+	update_divide_count(apic);
+	atomic_set(&apic->timer.pending, 0);
+	if (vcpu->vcpu_id == 0)
+		vcpu->arch.apic_base |= MSR_IA32_APICBASE_BSP;
+	apic_update_ppr(apic);
+
+	apic_debug(KERN_INFO "%s: vcpu=%p, id=%d, base_msr="
+		   "0x%016" PRIx64 ", base_address=0x%0lx.\n", __FUNCTION__,
+		   vcpu, kvm_apic_id(apic),
+		   vcpu->arch.apic_base, apic->base_address);
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_reset);
+
+int kvm_lapic_enabled(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	int ret = 0;
+
+	if (!apic)
+		return 0;
+	ret = apic_enabled(apic);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_enabled);
+
+/*
+ *----------------------------------------------------------------------
+ * timer interface
+ *----------------------------------------------------------------------
+ */
+
+/* TODO: make sure __apic_timer_fn runs in current pCPU */
+static int __apic_timer_fn(struct kvm_lapic *apic)
+{
+	int result = 0;
+	wait_queue_head_t *q = &apic->vcpu->wq;
+
+	atomic_inc(&apic->timer.pending);
+	if (waitqueue_active(q)) {
+		apic->vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
+		wake_up_interruptible(q);
+	}
+	if (apic_lvtt_period(apic)) {
+		result = 1;
+		apic->timer.dev.expires = ktime_add_ns(
+					apic->timer.dev.expires,
+					apic->timer.period);
+	}
+	return result;
+}
+
+static int __inject_apic_timer_irq(struct kvm_lapic *apic)
+{
+	int vector;
+
+	vector = apic_lvt_vector(apic, APIC_LVTT);
+	return __apic_accept_irq(apic, APIC_DM_FIXED, vector, 1, 0);
+}
+
+static enum hrtimer_restart apic_timer_fn(struct hrtimer *data)
+{
+	struct kvm_lapic *apic;
+	int restart_timer = 0;
+
+	apic = container_of(data, struct kvm_lapic, timer.dev);
+
+	restart_timer = __apic_timer_fn(apic);
+
+	if (restart_timer)
+		return HRTIMER_RESTART;
+	else
+		return HRTIMER_NORESTART;
+}
+
+int kvm_create_lapic(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic;
+
+	ASSERT(vcpu != NULL);
+	apic_debug("apic_init %d\n", vcpu->vcpu_id);
+
+	apic = kzalloc(sizeof(*apic), GFP_KERNEL);
+	if (!apic)
+		goto nomem;
+
+	vcpu->arch.apic = apic;
+
+	apic->regs_page = alloc_page(GFP_KERNEL);
+	if (apic->regs_page == NULL) {
+		printk(KERN_ERR "malloc apic regs error for vcpu %x\n",
+		       vcpu->vcpu_id);
+		goto nomem_free_apic;
+	}
+	apic->regs = page_address(apic->regs_page);
+	memset(apic->regs, 0, PAGE_SIZE);
+	apic->vcpu = vcpu;
+
+	hrtimer_init(&apic->timer.dev, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	apic->timer.dev.function = apic_timer_fn;
+	apic->base_address = APIC_DEFAULT_PHYS_BASE;
+	vcpu->arch.apic_base = APIC_DEFAULT_PHYS_BASE;
+
+	kvm_lapic_reset(vcpu);
+	apic->dev.read = apic_mmio_read;
+	apic->dev.write = apic_mmio_write;
+	apic->dev.in_range = apic_mmio_range;
+	apic->dev.private = apic;
+
+	return 0;
+nomem_free_apic:
+	kfree(apic);
+nomem:
+	return -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(kvm_create_lapic);
+
+int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	int highest_irr;
+
+	if (!apic || !apic_enabled(apic))
+		return -1;
+
+	apic_update_ppr(apic);
+	highest_irr = apic_find_highest_irr(apic);
+	if ((highest_irr == -1) ||
+	    ((highest_irr & 0xF0) <= apic_get_reg(apic, APIC_PROCPRI)))
+		return -1;
+	return highest_irr;
+}
+
+int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu)
+{
+	u32 lvt0 = apic_get_reg(vcpu->arch.apic, APIC_LVT0);
+	int r = 0;
+
+	if (vcpu->vcpu_id == 0) {
+		if (!apic_hw_enabled(vcpu->arch.apic))
+			r = 1;
+		if ((lvt0 & APIC_LVT_MASKED) == 0 &&
+		    GET_APIC_DELIVERY_MODE(lvt0) == APIC_MODE_EXTINT)
+			r = 1;
+	}
+	return r;
+}
+
+void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (apic && apic_lvt_enabled(apic, APIC_LVTT) &&
+		atomic_read(&apic->timer.pending) > 0) {
+		if (__inject_apic_timer_irq(apic))
+			atomic_dec(&apic->timer.pending);
+	}
+}
+
+void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (apic && apic_lvt_vector(apic, APIC_LVTT) == vec)
+		apic->timer.last_update = ktime_add_ns(
+				apic->timer.last_update,
+				apic->timer.period);
+}
+
+int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu)
+{
+	int vector = kvm_apic_has_interrupt(vcpu);
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (vector == -1)
+		return -1;
+
+	apic_set_vector(vector, apic->regs + APIC_ISR);
+	apic_update_ppr(apic);
+	apic_clear_irr(vector, apic);
+	return vector;
+}
+
+void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	apic->base_address = vcpu->arch.apic_base &
+			     MSR_IA32_APICBASE_BASE;
+	apic_set_reg(apic, APIC_LVR, APIC_VERSION);
+	apic_update_ppr(apic);
+	hrtimer_cancel(&apic->timer.dev);
+	update_divide_count(apic);
+	start_apic_timer(apic);
+}
+
+void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	struct hrtimer *timer;
+
+	if (!apic)
+		return;
+
+	timer = &apic->timer.dev;
+	if (hrtimer_cancel(timer))
+		hrtimer_start(timer, timer->expires, HRTIMER_MODE_ABS);
+}
+EXPORT_SYMBOL_GPL(kvm_migrate_apic_timer);
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
new file mode 100644
index 000000000000..401eb7ce3207
--- /dev/null
+++ b/arch/x86/kvm/mmu.c
@@ -0,0 +1,1805 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * This module enables machines with Intel VT-x extensions to run virtual
+ * machines without emulation or binary translation.
+ *
+ * MMU support
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ *
+ * Authors:
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *   Avi Kivity   <avi@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "vmx.h"
+#include "mmu.h"
+
+#include <linux/kvm_host.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/swap.h>
+
+#include <asm/page.h>
+#include <asm/cmpxchg.h>
+#include <asm/io.h>
+
+#undef MMU_DEBUG
+
+#undef AUDIT
+
+#ifdef AUDIT
+static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg);
+#else
+static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg) {}
+#endif
+
+#ifdef MMU_DEBUG
+
+#define pgprintk(x...) do { if (dbg) printk(x); } while (0)
+#define rmap_printk(x...) do { if (dbg) printk(x); } while (0)
+
+#else
+
+#define pgprintk(x...) do { } while (0)
+#define rmap_printk(x...) do { } while (0)
+
+#endif
+
+#if defined(MMU_DEBUG) || defined(AUDIT)
+static int dbg = 1;
+#endif
+
+#ifndef MMU_DEBUG
+#define ASSERT(x) do { } while (0)
+#else
+#define ASSERT(x)							\
+	if (!(x)) {							\
+		printk(KERN_WARNING "assertion failed %s:%d: %s\n",	\
+		       __FILE__, __LINE__, #x);				\
+	}
+#endif
+
+#define PT64_PT_BITS 9
+#define PT64_ENT_PER_PAGE (1 << PT64_PT_BITS)
+#define PT32_PT_BITS 10
+#define PT32_ENT_PER_PAGE (1 << PT32_PT_BITS)
+
+#define PT_WRITABLE_SHIFT 1
+
+#define PT_PRESENT_MASK (1ULL << 0)
+#define PT_WRITABLE_MASK (1ULL << PT_WRITABLE_SHIFT)
+#define PT_USER_MASK (1ULL << 2)
+#define PT_PWT_MASK (1ULL << 3)
+#define PT_PCD_MASK (1ULL << 4)
+#define PT_ACCESSED_MASK (1ULL << 5)
+#define PT_DIRTY_MASK (1ULL << 6)
+#define PT_PAGE_SIZE_MASK (1ULL << 7)
+#define PT_PAT_MASK (1ULL << 7)
+#define PT_GLOBAL_MASK (1ULL << 8)
+#define PT64_NX_SHIFT 63
+#define PT64_NX_MASK (1ULL << PT64_NX_SHIFT)
+
+#define PT_PAT_SHIFT 7
+#define PT_DIR_PAT_SHIFT 12
+#define PT_DIR_PAT_MASK (1ULL << PT_DIR_PAT_SHIFT)
+
+#define PT32_DIR_PSE36_SIZE 4
+#define PT32_DIR_PSE36_SHIFT 13
+#define PT32_DIR_PSE36_MASK \
+	(((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT)
+
+
+#define PT_FIRST_AVAIL_BITS_SHIFT 9
+#define PT64_SECOND_AVAIL_BITS_SHIFT 52
+
+#define PT_SHADOW_IO_MARK (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
+
+#define VALID_PAGE(x) ((x) != INVALID_PAGE)
+
+#define PT64_LEVEL_BITS 9
+
+#define PT64_LEVEL_SHIFT(level) \
+		(PAGE_SHIFT + (level - 1) * PT64_LEVEL_BITS)
+
+#define PT64_LEVEL_MASK(level) \
+		(((1ULL << PT64_LEVEL_BITS) - 1) << PT64_LEVEL_SHIFT(level))
+
+#define PT64_INDEX(address, level)\
+	(((address) >> PT64_LEVEL_SHIFT(level)) & ((1 << PT64_LEVEL_BITS) - 1))
+
+
+#define PT32_LEVEL_BITS 10
+
+#define PT32_LEVEL_SHIFT(level) \
+		(PAGE_SHIFT + (level - 1) * PT32_LEVEL_BITS)
+
+#define PT32_LEVEL_MASK(level) \
+		(((1ULL << PT32_LEVEL_BITS) - 1) << PT32_LEVEL_SHIFT(level))
+
+#define PT32_INDEX(address, level)\
+	(((address) >> PT32_LEVEL_SHIFT(level)) & ((1 << PT32_LEVEL_BITS) - 1))
+
+
+#define PT64_BASE_ADDR_MASK (((1ULL << 52) - 1) & ~(u64)(PAGE_SIZE-1))
+#define PT64_DIR_BASE_ADDR_MASK \
+	(PT64_BASE_ADDR_MASK & ~((1ULL << (PAGE_SHIFT + PT64_LEVEL_BITS)) - 1))
+
+#define PT32_BASE_ADDR_MASK PAGE_MASK
+#define PT32_DIR_BASE_ADDR_MASK \
+	(PAGE_MASK & ~((1ULL << (PAGE_SHIFT + PT32_LEVEL_BITS)) - 1))
+
+#define PT64_PERM_MASK (PT_PRESENT_MASK | PT_WRITABLE_MASK | PT_USER_MASK \
+			| PT64_NX_MASK)
+
+#define PFERR_PRESENT_MASK (1U << 0)
+#define PFERR_WRITE_MASK (1U << 1)
+#define PFERR_USER_MASK (1U << 2)
+#define PFERR_FETCH_MASK (1U << 4)
+
+#define PT64_ROOT_LEVEL 4
+#define PT32_ROOT_LEVEL 2
+#define PT32E_ROOT_LEVEL 3
+
+#define PT_DIRECTORY_LEVEL 2
+#define PT_PAGE_TABLE_LEVEL 1
+
+#define RMAP_EXT 4
+
+#define ACC_EXEC_MASK    1
+#define ACC_WRITE_MASK   PT_WRITABLE_MASK
+#define ACC_USER_MASK    PT_USER_MASK
+#define ACC_ALL          (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK)
+
+struct kvm_rmap_desc {
+	u64 *shadow_ptes[RMAP_EXT];
+	struct kvm_rmap_desc *more;
+};
+
+static struct kmem_cache *pte_chain_cache;
+static struct kmem_cache *rmap_desc_cache;
+static struct kmem_cache *mmu_page_header_cache;
+
+static u64 __read_mostly shadow_trap_nonpresent_pte;
+static u64 __read_mostly shadow_notrap_nonpresent_pte;
+
+void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte)
+{
+	shadow_trap_nonpresent_pte = trap_pte;
+	shadow_notrap_nonpresent_pte = notrap_pte;
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes);
+
+static int is_write_protection(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.cr0 & X86_CR0_WP;
+}
+
+static int is_cpuid_PSE36(void)
+{
+	return 1;
+}
+
+static int is_nx(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.shadow_efer & EFER_NX;
+}
+
+static int is_present_pte(unsigned long pte)
+{
+	return pte & PT_PRESENT_MASK;
+}
+
+static int is_shadow_present_pte(u64 pte)
+{
+	pte &= ~PT_SHADOW_IO_MARK;
+	return pte != shadow_trap_nonpresent_pte
+		&& pte != shadow_notrap_nonpresent_pte;
+}
+
+static int is_writeble_pte(unsigned long pte)
+{
+	return pte & PT_WRITABLE_MASK;
+}
+
+static int is_dirty_pte(unsigned long pte)
+{
+	return pte & PT_DIRTY_MASK;
+}
+
+static int is_io_pte(unsigned long pte)
+{
+	return pte & PT_SHADOW_IO_MARK;
+}
+
+static int is_rmap_pte(u64 pte)
+{
+	return pte != shadow_trap_nonpresent_pte
+		&& pte != shadow_notrap_nonpresent_pte;
+}
+
+static gfn_t pse36_gfn_delta(u32 gpte)
+{
+	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
+
+	return (gpte & PT32_DIR_PSE36_MASK) << shift;
+}
+
+static void set_shadow_pte(u64 *sptep, u64 spte)
+{
+#ifdef CONFIG_X86_64
+	set_64bit((unsigned long *)sptep, spte);
+#else
+	set_64bit((unsigned long long *)sptep, spte);
+#endif
+}
+
+static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache,
+				  struct kmem_cache *base_cache, int min)
+{
+	void *obj;
+
+	if (cache->nobjs >= min)
+		return 0;
+	while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
+		obj = kmem_cache_zalloc(base_cache, GFP_KERNEL);
+		if (!obj)
+			return -ENOMEM;
+		cache->objects[cache->nobjs++] = obj;
+	}
+	return 0;
+}
+
+static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc)
+{
+	while (mc->nobjs)
+		kfree(mc->objects[--mc->nobjs]);
+}
+
+static int mmu_topup_memory_cache_page(struct kvm_mmu_memory_cache *cache,
+				       int min)
+{
+	struct page *page;
+
+	if (cache->nobjs >= min)
+		return 0;
+	while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
+		page = alloc_page(GFP_KERNEL);
+		if (!page)
+			return -ENOMEM;
+		set_page_private(page, 0);
+		cache->objects[cache->nobjs++] = page_address(page);
+	}
+	return 0;
+}
+
+static void mmu_free_memory_cache_page(struct kvm_mmu_memory_cache *mc)
+{
+	while (mc->nobjs)
+		free_page((unsigned long)mc->objects[--mc->nobjs]);
+}
+
+static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
+{
+	int r;
+
+	kvm_mmu_free_some_pages(vcpu);
+	r = mmu_topup_memory_cache(&vcpu->arch.mmu_pte_chain_cache,
+				   pte_chain_cache, 4);
+	if (r)
+		goto out;
+	r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache,
+				   rmap_desc_cache, 1);
+	if (r)
+		goto out;
+	r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8);
+	if (r)
+		goto out;
+	r = mmu_topup_memory_cache(&vcpu->arch.mmu_page_header_cache,
+				   mmu_page_header_cache, 4);
+out:
+	return r;
+}
+
+static void mmu_free_memory_caches(struct kvm_vcpu *vcpu)
+{
+	mmu_free_memory_cache(&vcpu->arch.mmu_pte_chain_cache);
+	mmu_free_memory_cache(&vcpu->arch.mmu_rmap_desc_cache);
+	mmu_free_memory_cache_page(&vcpu->arch.mmu_page_cache);
+	mmu_free_memory_cache(&vcpu->arch.mmu_page_header_cache);
+}
+
+static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
+				    size_t size)
+{
+	void *p;
+
+	BUG_ON(!mc->nobjs);
+	p = mc->objects[--mc->nobjs];
+	memset(p, 0, size);
+	return p;
+}
+
+static struct kvm_pte_chain *mmu_alloc_pte_chain(struct kvm_vcpu *vcpu)
+{
+	return mmu_memory_cache_alloc(&vcpu->arch.mmu_pte_chain_cache,
+				      sizeof(struct kvm_pte_chain));
+}
+
+static void mmu_free_pte_chain(struct kvm_pte_chain *pc)
+{
+	kfree(pc);
+}
+
+static struct kvm_rmap_desc *mmu_alloc_rmap_desc(struct kvm_vcpu *vcpu)
+{
+	return mmu_memory_cache_alloc(&vcpu->arch.mmu_rmap_desc_cache,
+				      sizeof(struct kvm_rmap_desc));
+}
+
+static void mmu_free_rmap_desc(struct kvm_rmap_desc *rd)
+{
+	kfree(rd);
+}
+
+/*
+ * Take gfn and return the reverse mapping to it.
+ * Note: gfn must be unaliased before this function get called
+ */
+
+static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn)
+{
+	struct kvm_memory_slot *slot;
+
+	slot = gfn_to_memslot(kvm, gfn);
+	return &slot->rmap[gfn - slot->base_gfn];
+}
+
+/*
+ * Reverse mapping data structures:
+ *
+ * If rmapp bit zero is zero, then rmapp point to the shadw page table entry
+ * that points to page_address(page).
+ *
+ * If rmapp bit zero is one, (then rmap & ~1) points to a struct kvm_rmap_desc
+ * containing more mappings.
+ */
+static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
+{
+	struct kvm_mmu_page *sp;
+	struct kvm_rmap_desc *desc;
+	unsigned long *rmapp;
+	int i;
+
+	if (!is_rmap_pte(*spte))
+		return;
+	gfn = unalias_gfn(vcpu->kvm, gfn);
+	sp = page_header(__pa(spte));
+	sp->gfns[spte - sp->spt] = gfn;
+	rmapp = gfn_to_rmap(vcpu->kvm, gfn);
+	if (!*rmapp) {
+		rmap_printk("rmap_add: %p %llx 0->1\n", spte, *spte);
+		*rmapp = (unsigned long)spte;
+	} else if (!(*rmapp & 1)) {
+		rmap_printk("rmap_add: %p %llx 1->many\n", spte, *spte);
+		desc = mmu_alloc_rmap_desc(vcpu);
+		desc->shadow_ptes[0] = (u64 *)*rmapp;
+		desc->shadow_ptes[1] = spte;
+		*rmapp = (unsigned long)desc | 1;
+	} else {
+		rmap_printk("rmap_add: %p %llx many->many\n", spte, *spte);
+		desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
+		while (desc->shadow_ptes[RMAP_EXT-1] && desc->more)
+			desc = desc->more;
+		if (desc->shadow_ptes[RMAP_EXT-1]) {
+			desc->more = mmu_alloc_rmap_desc(vcpu);
+			desc = desc->more;
+		}
+		for (i = 0; desc->shadow_ptes[i]; ++i)
+			;
+		desc->shadow_ptes[i] = spte;
+	}
+}
+
+static void rmap_desc_remove_entry(unsigned long *rmapp,
+				   struct kvm_rmap_desc *desc,
+				   int i,
+				   struct kvm_rmap_desc *prev_desc)
+{
+	int j;
+
+	for (j = RMAP_EXT - 1; !desc->shadow_ptes[j] && j > i; --j)
+		;
+	desc->shadow_ptes[i] = desc->shadow_ptes[j];
+	desc->shadow_ptes[j] = NULL;
+	if (j != 0)
+		return;
+	if (!prev_desc && !desc->more)
+		*rmapp = (unsigned long)desc->shadow_ptes[0];
+	else
+		if (prev_desc)
+			prev_desc->more = desc->more;
+		else
+			*rmapp = (unsigned long)desc->more | 1;
+	mmu_free_rmap_desc(desc);
+}
+
+static void rmap_remove(struct kvm *kvm, u64 *spte)
+{
+	struct kvm_rmap_desc *desc;
+	struct kvm_rmap_desc *prev_desc;
+	struct kvm_mmu_page *sp;
+	struct page *page;
+	unsigned long *rmapp;
+	int i;
+
+	if (!is_rmap_pte(*spte))
+		return;
+	sp = page_header(__pa(spte));
+	page = pfn_to_page((*spte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT);
+	mark_page_accessed(page);
+	if (is_writeble_pte(*spte))
+		kvm_release_page_dirty(page);
+	else
+		kvm_release_page_clean(page);
+	rmapp = gfn_to_rmap(kvm, sp->gfns[spte - sp->spt]);
+	if (!*rmapp) {
+		printk(KERN_ERR "rmap_remove: %p %llx 0->BUG\n", spte, *spte);
+		BUG();
+	} else if (!(*rmapp & 1)) {
+		rmap_printk("rmap_remove:  %p %llx 1->0\n", spte, *spte);
+		if ((u64 *)*rmapp != spte) {
+			printk(KERN_ERR "rmap_remove:  %p %llx 1->BUG\n",
+			       spte, *spte);
+			BUG();
+		}
+		*rmapp = 0;
+	} else {
+		rmap_printk("rmap_remove:  %p %llx many->many\n", spte, *spte);
+		desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
+		prev_desc = NULL;
+		while (desc) {
+			for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i)
+				if (desc->shadow_ptes[i] == spte) {
+					rmap_desc_remove_entry(rmapp,
+							       desc, i,
+							       prev_desc);
+					return;
+				}
+			prev_desc = desc;
+			desc = desc->more;
+		}
+		BUG();
+	}
+}
+
+static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte)
+{
+	struct kvm_rmap_desc *desc;
+	struct kvm_rmap_desc *prev_desc;
+	u64 *prev_spte;
+	int i;
+
+	if (!*rmapp)
+		return NULL;
+	else if (!(*rmapp & 1)) {
+		if (!spte)
+			return (u64 *)*rmapp;
+		return NULL;
+	}
+	desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
+	prev_desc = NULL;
+	prev_spte = NULL;
+	while (desc) {
+		for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i) {
+			if (prev_spte == spte)
+				return desc->shadow_ptes[i];
+			prev_spte = desc->shadow_ptes[i];
+		}
+		desc = desc->more;
+	}
+	return NULL;
+}
+
+static void rmap_write_protect(struct kvm *kvm, u64 gfn)
+{
+	unsigned long *rmapp;
+	u64 *spte;
+
+	gfn = unalias_gfn(kvm, gfn);
+	rmapp = gfn_to_rmap(kvm, gfn);
+
+	spte = rmap_next(kvm, rmapp, NULL);
+	while (spte) {
+		BUG_ON(!spte);
+		BUG_ON(!(*spte & PT_PRESENT_MASK));
+		rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte);
+		if (is_writeble_pte(*spte))
+			set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK);
+		kvm_flush_remote_tlbs(kvm);
+		spte = rmap_next(kvm, rmapp, spte);
+	}
+}
+
+#ifdef MMU_DEBUG
+static int is_empty_shadow_page(u64 *spt)
+{
+	u64 *pos;
+	u64 *end;
+
+	for (pos = spt, end = pos + PAGE_SIZE / sizeof(u64); pos != end; pos++)
+		if ((*pos & ~PT_SHADOW_IO_MARK) != shadow_trap_nonpresent_pte) {
+			printk(KERN_ERR "%s: %p %llx\n", __FUNCTION__,
+			       pos, *pos);
+			return 0;
+		}
+	return 1;
+}
+#endif
+
+static void kvm_mmu_free_page(struct kvm *kvm, struct kvm_mmu_page *sp)
+{
+	ASSERT(is_empty_shadow_page(sp->spt));
+	list_del(&sp->link);
+	__free_page(virt_to_page(sp->spt));
+	__free_page(virt_to_page(sp->gfns));
+	kfree(sp);
+	++kvm->arch.n_free_mmu_pages;
+}
+
+static unsigned kvm_page_table_hashfn(gfn_t gfn)
+{
+	return gfn;
+}
+
+static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
+					       u64 *parent_pte)
+{
+	struct kvm_mmu_page *sp;
+
+	if (!vcpu->kvm->arch.n_free_mmu_pages)
+		return NULL;
+
+	sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
+	sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
+	sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
+	set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
+	list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
+	ASSERT(is_empty_shadow_page(sp->spt));
+	sp->slot_bitmap = 0;
+	sp->multimapped = 0;
+	sp->parent_pte = parent_pte;
+	--vcpu->kvm->arch.n_free_mmu_pages;
+	return sp;
+}
+
+static void mmu_page_add_parent_pte(struct kvm_vcpu *vcpu,
+				    struct kvm_mmu_page *sp, u64 *parent_pte)
+{
+	struct kvm_pte_chain *pte_chain;
+	struct hlist_node *node;
+	int i;
+
+	if (!parent_pte)
+		return;
+	if (!sp->multimapped) {
+		u64 *old = sp->parent_pte;
+
+		if (!old) {
+			sp->parent_pte = parent_pte;
+			return;
+		}
+		sp->multimapped = 1;
+		pte_chain = mmu_alloc_pte_chain(vcpu);
+		INIT_HLIST_HEAD(&sp->parent_ptes);
+		hlist_add_head(&pte_chain->link, &sp->parent_ptes);
+		pte_chain->parent_ptes[0] = old;
+	}
+	hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) {
+		if (pte_chain->parent_ptes[NR_PTE_CHAIN_ENTRIES-1])
+			continue;
+		for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i)
+			if (!pte_chain->parent_ptes[i]) {
+				pte_chain->parent_ptes[i] = parent_pte;
+				return;
+			}
+	}
+	pte_chain = mmu_alloc_pte_chain(vcpu);
+	BUG_ON(!pte_chain);
+	hlist_add_head(&pte_chain->link, &sp->parent_ptes);
+	pte_chain->parent_ptes[0] = parent_pte;
+}
+
+static void mmu_page_remove_parent_pte(struct kvm_mmu_page *sp,
+				       u64 *parent_pte)
+{
+	struct kvm_pte_chain *pte_chain;
+	struct hlist_node *node;
+	int i;
+
+	if (!sp->multimapped) {
+		BUG_ON(sp->parent_pte != parent_pte);
+		sp->parent_pte = NULL;
+		return;
+	}
+	hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link)
+		for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) {
+			if (!pte_chain->parent_ptes[i])
+				break;
+			if (pte_chain->parent_ptes[i] != parent_pte)
+				continue;
+			while (i + 1 < NR_PTE_CHAIN_ENTRIES
+				&& pte_chain->parent_ptes[i + 1]) {
+				pte_chain->parent_ptes[i]
+					= pte_chain->parent_ptes[i + 1];
+				++i;
+			}
+			pte_chain->parent_ptes[i] = NULL;
+			if (i == 0) {
+				hlist_del(&pte_chain->link);
+				mmu_free_pte_chain(pte_chain);
+				if (hlist_empty(&sp->parent_ptes)) {
+					sp->multimapped = 0;
+					sp->parent_pte = NULL;
+				}
+			}
+			return;
+		}
+	BUG();
+}
+
+static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
+{
+	unsigned index;
+	struct hlist_head *bucket;
+	struct kvm_mmu_page *sp;
+	struct hlist_node *node;
+
+	pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
+	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
+	bucket = &kvm->arch.mmu_page_hash[index];
+	hlist_for_each_entry(sp, node, bucket, hash_link)
+		if (sp->gfn == gfn && !sp->role.metaphysical) {
+			pgprintk("%s: found role %x\n",
+				 __FUNCTION__, sp->role.word);
+			return sp;
+		}
+	return NULL;
+}
+
+static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
+					     gfn_t gfn,
+					     gva_t gaddr,
+					     unsigned level,
+					     int metaphysical,
+					     unsigned access,
+					     u64 *parent_pte,
+					     bool *new_page)
+{
+	union kvm_mmu_page_role role;
+	unsigned index;
+	unsigned quadrant;
+	struct hlist_head *bucket;
+	struct kvm_mmu_page *sp;
+	struct hlist_node *node;
+
+	role.word = 0;
+	role.glevels = vcpu->arch.mmu.root_level;
+	role.level = level;
+	role.metaphysical = metaphysical;
+	role.access = access;
+	if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) {
+		quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level));
+		quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1;
+		role.quadrant = quadrant;
+	}
+	pgprintk("%s: looking gfn %lx role %x\n", __FUNCTION__,
+		 gfn, role.word);
+	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
+	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
+	hlist_for_each_entry(sp, node, bucket, hash_link)
+		if (sp->gfn == gfn && sp->role.word == role.word) {
+			mmu_page_add_parent_pte(vcpu, sp, parent_pte);
+			pgprintk("%s: found\n", __FUNCTION__);
+			return sp;
+		}
+	sp = kvm_mmu_alloc_page(vcpu, parent_pte);
+	if (!sp)
+		return sp;
+	pgprintk("%s: adding gfn %lx role %x\n", __FUNCTION__, gfn, role.word);
+	sp->gfn = gfn;
+	sp->role = role;
+	hlist_add_head(&sp->hash_link, bucket);
+	vcpu->arch.mmu.prefetch_page(vcpu, sp);
+	if (!metaphysical)
+		rmap_write_protect(vcpu->kvm, gfn);
+	if (new_page)
+		*new_page = 1;
+	return sp;
+}
+
+static void kvm_mmu_page_unlink_children(struct kvm *kvm,
+					 struct kvm_mmu_page *sp)
+{
+	unsigned i;
+	u64 *pt;
+	u64 ent;
+
+	pt = sp->spt;
+
+	if (sp->role.level == PT_PAGE_TABLE_LEVEL) {
+		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+			if (is_shadow_present_pte(pt[i]))
+				rmap_remove(kvm, &pt[i]);
+			pt[i] = shadow_trap_nonpresent_pte;
+		}
+		kvm_flush_remote_tlbs(kvm);
+		return;
+	}
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+		ent = pt[i];
+
+		pt[i] = shadow_trap_nonpresent_pte;
+		if (!is_shadow_present_pte(ent))
+			continue;
+		ent &= PT64_BASE_ADDR_MASK;
+		mmu_page_remove_parent_pte(page_header(ent), &pt[i]);
+	}
+	kvm_flush_remote_tlbs(kvm);
+}
+
+static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte)
+{
+	mmu_page_remove_parent_pte(sp, parent_pte);
+}
+
+static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm)
+{
+	int i;
+
+	for (i = 0; i < KVM_MAX_VCPUS; ++i)
+		if (kvm->vcpus[i])
+			kvm->vcpus[i]->arch.last_pte_updated = NULL;
+}
+
+static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
+{
+	u64 *parent_pte;
+
+	++kvm->stat.mmu_shadow_zapped;
+	while (sp->multimapped || sp->parent_pte) {
+		if (!sp->multimapped)
+			parent_pte = sp->parent_pte;
+		else {
+			struct kvm_pte_chain *chain;
+
+			chain = container_of(sp->parent_ptes.first,
+					     struct kvm_pte_chain, link);
+			parent_pte = chain->parent_ptes[0];
+		}
+		BUG_ON(!parent_pte);
+		kvm_mmu_put_page(sp, parent_pte);
+		set_shadow_pte(parent_pte, shadow_trap_nonpresent_pte);
+	}
+	kvm_mmu_page_unlink_children(kvm, sp);
+	if (!sp->root_count) {
+		hlist_del(&sp->hash_link);
+		kvm_mmu_free_page(kvm, sp);
+	} else
+		list_move(&sp->link, &kvm->arch.active_mmu_pages);
+	kvm_mmu_reset_last_pte_updated(kvm);
+}
+
+/*
+ * Changing the number of mmu pages allocated to the vm
+ * Note: if kvm_nr_mmu_pages is too small, you will get dead lock
+ */
+void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages)
+{
+	/*
+	 * If we set the number of mmu pages to be smaller be than the
+	 * number of actived pages , we must to free some mmu pages before we
+	 * change the value
+	 */
+
+	if ((kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages) >
+	    kvm_nr_mmu_pages) {
+		int n_used_mmu_pages = kvm->arch.n_alloc_mmu_pages
+				       - kvm->arch.n_free_mmu_pages;
+
+		while (n_used_mmu_pages > kvm_nr_mmu_pages) {
+			struct kvm_mmu_page *page;
+
+			page = container_of(kvm->arch.active_mmu_pages.prev,
+					    struct kvm_mmu_page, link);
+			kvm_mmu_zap_page(kvm, page);
+			n_used_mmu_pages--;
+		}
+		kvm->arch.n_free_mmu_pages = 0;
+	}
+	else
+		kvm->arch.n_free_mmu_pages += kvm_nr_mmu_pages
+					 - kvm->arch.n_alloc_mmu_pages;
+
+	kvm->arch.n_alloc_mmu_pages = kvm_nr_mmu_pages;
+}
+
+static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
+{
+	unsigned index;
+	struct hlist_head *bucket;
+	struct kvm_mmu_page *sp;
+	struct hlist_node *node, *n;
+	int r;
+
+	pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
+	r = 0;
+	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
+	bucket = &kvm->arch.mmu_page_hash[index];
+	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link)
+		if (sp->gfn == gfn && !sp->role.metaphysical) {
+			pgprintk("%s: gfn %lx role %x\n", __FUNCTION__, gfn,
+				 sp->role.word);
+			kvm_mmu_zap_page(kvm, sp);
+			r = 1;
+		}
+	return r;
+}
+
+static void mmu_unshadow(struct kvm *kvm, gfn_t gfn)
+{
+	struct kvm_mmu_page *sp;
+
+	while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) {
+		pgprintk("%s: zap %lx %x\n", __FUNCTION__, gfn, sp->role.word);
+		kvm_mmu_zap_page(kvm, sp);
+	}
+}
+
+static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn)
+{
+	int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn));
+	struct kvm_mmu_page *sp = page_header(__pa(pte));
+
+	__set_bit(slot, &sp->slot_bitmap);
+}
+
+struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
+{
+	gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
+
+	if (gpa == UNMAPPED_GVA)
+		return NULL;
+	return gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+}
+
+static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
+			 unsigned pt_access, unsigned pte_access,
+			 int user_fault, int write_fault, int dirty,
+			 int *ptwrite, gfn_t gfn)
+{
+	u64 spte;
+	int was_rmapped = is_rmap_pte(*shadow_pte);
+	struct page *page;
+
+	pgprintk("%s: spte %llx access %x write_fault %d"
+		 " user_fault %d gfn %lx\n",
+		 __FUNCTION__, *shadow_pte, pt_access,
+		 write_fault, user_fault, gfn);
+
+	/*
+	 * We don't set the accessed bit, since we sometimes want to see
+	 * whether the guest actually used the pte (in order to detect
+	 * demand paging).
+	 */
+	spte = PT_PRESENT_MASK | PT_DIRTY_MASK;
+	if (!dirty)
+		pte_access &= ~ACC_WRITE_MASK;
+	if (!(pte_access & ACC_EXEC_MASK))
+		spte |= PT64_NX_MASK;
+
+	page = gfn_to_page(vcpu->kvm, gfn);
+
+	spte |= PT_PRESENT_MASK;
+	if (pte_access & ACC_USER_MASK)
+		spte |= PT_USER_MASK;
+
+	if (is_error_page(page)) {
+		set_shadow_pte(shadow_pte,
+			       shadow_trap_nonpresent_pte | PT_SHADOW_IO_MARK);
+		kvm_release_page_clean(page);
+		return;
+	}
+
+	spte |= page_to_phys(page);
+
+	if ((pte_access & ACC_WRITE_MASK)
+	    || (write_fault && !is_write_protection(vcpu) && !user_fault)) {
+		struct kvm_mmu_page *shadow;
+
+		spte |= PT_WRITABLE_MASK;
+		if (user_fault) {
+			mmu_unshadow(vcpu->kvm, gfn);
+			goto unshadowed;
+		}
+
+		shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn);
+		if (shadow) {
+			pgprintk("%s: found shadow page for %lx, marking ro\n",
+				 __FUNCTION__, gfn);
+			pte_access &= ~ACC_WRITE_MASK;
+			if (is_writeble_pte(spte)) {
+				spte &= ~PT_WRITABLE_MASK;
+				kvm_x86_ops->tlb_flush(vcpu);
+			}
+			if (write_fault)
+				*ptwrite = 1;
+		}
+	}
+
+unshadowed:
+
+	if (pte_access & ACC_WRITE_MASK)
+		mark_page_dirty(vcpu->kvm, gfn);
+
+	pgprintk("%s: setting spte %llx\n", __FUNCTION__, spte);
+	set_shadow_pte(shadow_pte, spte);
+	page_header_update_slot(vcpu->kvm, shadow_pte, gfn);
+	if (!was_rmapped) {
+		rmap_add(vcpu, shadow_pte, gfn);
+		if (!is_rmap_pte(*shadow_pte))
+			kvm_release_page_clean(page);
+	}
+	else
+		kvm_release_page_clean(page);
+	if (!ptwrite || !*ptwrite)
+		vcpu->arch.last_pte_updated = shadow_pte;
+}
+
+static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
+{
+}
+
+static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
+{
+	int level = PT32E_ROOT_LEVEL;
+	hpa_t table_addr = vcpu->arch.mmu.root_hpa;
+	int pt_write = 0;
+
+	for (; ; level--) {
+		u32 index = PT64_INDEX(v, level);
+		u64 *table;
+
+		ASSERT(VALID_PAGE(table_addr));
+		table = __va(table_addr);
+
+		if (level == 1) {
+			mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL,
+				     0, write, 1, &pt_write, gfn);
+			return pt_write || is_io_pte(table[index]);
+		}
+
+		if (table[index] == shadow_trap_nonpresent_pte) {
+			struct kvm_mmu_page *new_table;
+			gfn_t pseudo_gfn;
+
+			pseudo_gfn = (v & PT64_DIR_BASE_ADDR_MASK)
+				>> PAGE_SHIFT;
+			new_table = kvm_mmu_get_page(vcpu, pseudo_gfn,
+						     v, level - 1,
+						     1, ACC_ALL, &table[index],
+						     NULL);
+			if (!new_table) {
+				pgprintk("nonpaging_map: ENOMEM\n");
+				return -ENOMEM;
+			}
+
+			table[index] = __pa(new_table->spt) | PT_PRESENT_MASK
+				| PT_WRITABLE_MASK | PT_USER_MASK;
+		}
+		table_addr = table[index] & PT64_BASE_ADDR_MASK;
+	}
+}
+
+static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu,
+				    struct kvm_mmu_page *sp)
+{
+	int i;
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
+		sp->spt[i] = shadow_trap_nonpresent_pte;
+}
+
+static void mmu_free_roots(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_mmu_page *sp;
+
+	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
+		return;
+#ifdef CONFIG_X86_64
+	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+		hpa_t root = vcpu->arch.mmu.root_hpa;
+
+		sp = page_header(root);
+		--sp->root_count;
+		vcpu->arch.mmu.root_hpa = INVALID_PAGE;
+		return;
+	}
+#endif
+	for (i = 0; i < 4; ++i) {
+		hpa_t root = vcpu->arch.mmu.pae_root[i];
+
+		if (root) {
+			root &= PT64_BASE_ADDR_MASK;
+			sp = page_header(root);
+			--sp->root_count;
+		}
+		vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
+	}
+	vcpu->arch.mmu.root_hpa = INVALID_PAGE;
+}
+
+static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
+{
+	int i;
+	gfn_t root_gfn;
+	struct kvm_mmu_page *sp;
+
+	root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT;
+
+#ifdef CONFIG_X86_64
+	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+		hpa_t root = vcpu->arch.mmu.root_hpa;
+
+		ASSERT(!VALID_PAGE(root));
+		sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
+				      PT64_ROOT_LEVEL, 0, ACC_ALL, NULL, NULL);
+		root = __pa(sp->spt);
+		++sp->root_count;
+		vcpu->arch.mmu.root_hpa = root;
+		return;
+	}
+#endif
+	for (i = 0; i < 4; ++i) {
+		hpa_t root = vcpu->arch.mmu.pae_root[i];
+
+		ASSERT(!VALID_PAGE(root));
+		if (vcpu->arch.mmu.root_level == PT32E_ROOT_LEVEL) {
+			if (!is_present_pte(vcpu->arch.pdptrs[i])) {
+				vcpu->arch.mmu.pae_root[i] = 0;
+				continue;
+			}
+			root_gfn = vcpu->arch.pdptrs[i] >> PAGE_SHIFT;
+		} else if (vcpu->arch.mmu.root_level == 0)
+			root_gfn = 0;
+		sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
+				      PT32_ROOT_LEVEL, !is_paging(vcpu),
+				      ACC_ALL, NULL, NULL);
+		root = __pa(sp->spt);
+		++sp->root_count;
+		vcpu->arch.mmu.pae_root[i] = root | PT_PRESENT_MASK;
+	}
+	vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root);
+}
+
+static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr)
+{
+	return vaddr;
+}
+
+static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
+				u32 error_code)
+{
+	gfn_t gfn;
+	int r;
+
+	pgprintk("%s: gva %lx error %x\n", __FUNCTION__, gva, error_code);
+	r = mmu_topup_memory_caches(vcpu);
+	if (r)
+		return r;
+
+	ASSERT(vcpu);
+	ASSERT(VALID_PAGE(vcpu->arch.mmu.root_hpa));
+
+	gfn = gva >> PAGE_SHIFT;
+
+	return nonpaging_map(vcpu, gva & PAGE_MASK,
+			     error_code & PFERR_WRITE_MASK, gfn);
+}
+
+static void nonpaging_free(struct kvm_vcpu *vcpu)
+{
+	mmu_free_roots(vcpu);
+}
+
+static int nonpaging_init_context(struct kvm_vcpu *vcpu)
+{
+	struct kvm_mmu *context = &vcpu->arch.mmu;
+
+	context->new_cr3 = nonpaging_new_cr3;
+	context->page_fault = nonpaging_page_fault;
+	context->gva_to_gpa = nonpaging_gva_to_gpa;
+	context->free = nonpaging_free;
+	context->prefetch_page = nonpaging_prefetch_page;
+	context->root_level = 0;
+	context->shadow_root_level = PT32E_ROOT_LEVEL;
+	context->root_hpa = INVALID_PAGE;
+	return 0;
+}
+
+void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
+{
+	++vcpu->stat.tlb_flush;
+	kvm_x86_ops->tlb_flush(vcpu);
+}
+
+static void paging_new_cr3(struct kvm_vcpu *vcpu)
+{
+	pgprintk("%s: cr3 %lx\n", __FUNCTION__, vcpu->cr3);
+	mmu_free_roots(vcpu);
+}
+
+static void inject_page_fault(struct kvm_vcpu *vcpu,
+			      u64 addr,
+			      u32 err_code)
+{
+	kvm_inject_page_fault(vcpu, addr, err_code);
+}
+
+static void paging_free(struct kvm_vcpu *vcpu)
+{
+	nonpaging_free(vcpu);
+}
+
+#define PTTYPE 64
+#include "paging_tmpl.h"
+#undef PTTYPE
+
+#define PTTYPE 32
+#include "paging_tmpl.h"
+#undef PTTYPE
+
+static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level)
+{
+	struct kvm_mmu *context = &vcpu->arch.mmu;
+
+	ASSERT(is_pae(vcpu));
+	context->new_cr3 = paging_new_cr3;
+	context->page_fault = paging64_page_fault;
+	context->gva_to_gpa = paging64_gva_to_gpa;
+	context->prefetch_page = paging64_prefetch_page;
+	context->free = paging_free;
+	context->root_level = level;
+	context->shadow_root_level = level;
+	context->root_hpa = INVALID_PAGE;
+	return 0;
+}
+
+static int paging64_init_context(struct kvm_vcpu *vcpu)
+{
+	return paging64_init_context_common(vcpu, PT64_ROOT_LEVEL);
+}
+
+static int paging32_init_context(struct kvm_vcpu *vcpu)
+{
+	struct kvm_mmu *context = &vcpu->arch.mmu;
+
+	context->new_cr3 = paging_new_cr3;
+	context->page_fault = paging32_page_fault;
+	context->gva_to_gpa = paging32_gva_to_gpa;
+	context->free = paging_free;
+	context->prefetch_page = paging32_prefetch_page;
+	context->root_level = PT32_ROOT_LEVEL;
+	context->shadow_root_level = PT32E_ROOT_LEVEL;
+	context->root_hpa = INVALID_PAGE;
+	return 0;
+}
+
+static int paging32E_init_context(struct kvm_vcpu *vcpu)
+{
+	return paging64_init_context_common(vcpu, PT32E_ROOT_LEVEL);
+}
+
+static int init_kvm_mmu(struct kvm_vcpu *vcpu)
+{
+	ASSERT(vcpu);
+	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
+
+	if (!is_paging(vcpu))
+		return nonpaging_init_context(vcpu);
+	else if (is_long_mode(vcpu))
+		return paging64_init_context(vcpu);
+	else if (is_pae(vcpu))
+		return paging32E_init_context(vcpu);
+	else
+		return paging32_init_context(vcpu);
+}
+
+static void destroy_kvm_mmu(struct kvm_vcpu *vcpu)
+{
+	ASSERT(vcpu);
+	if (VALID_PAGE(vcpu->arch.mmu.root_hpa)) {
+		vcpu->arch.mmu.free(vcpu);
+		vcpu->arch.mmu.root_hpa = INVALID_PAGE;
+	}
+}
+
+int kvm_mmu_reset_context(struct kvm_vcpu *vcpu)
+{
+	destroy_kvm_mmu(vcpu);
+	return init_kvm_mmu(vcpu);
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_reset_context);
+
+int kvm_mmu_load(struct kvm_vcpu *vcpu)
+{
+	int r;
+
+	mutex_lock(&vcpu->kvm->lock);
+	r = mmu_topup_memory_caches(vcpu);
+	if (r)
+		goto out;
+	mmu_alloc_roots(vcpu);
+	kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
+	kvm_mmu_flush_tlb(vcpu);
+out:
+	mutex_unlock(&vcpu->kvm->lock);
+	return r;
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_load);
+
+void kvm_mmu_unload(struct kvm_vcpu *vcpu)
+{
+	mmu_free_roots(vcpu);
+}
+
+static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu,
+				  struct kvm_mmu_page *sp,
+				  u64 *spte)
+{
+	u64 pte;
+	struct kvm_mmu_page *child;
+
+	pte = *spte;
+	if (is_shadow_present_pte(pte)) {
+		if (sp->role.level == PT_PAGE_TABLE_LEVEL)
+			rmap_remove(vcpu->kvm, spte);
+		else {
+			child = page_header(pte & PT64_BASE_ADDR_MASK);
+			mmu_page_remove_parent_pte(child, spte);
+		}
+	}
+	set_shadow_pte(spte, shadow_trap_nonpresent_pte);
+}
+
+static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu,
+				  struct kvm_mmu_page *sp,
+				  u64 *spte,
+				  const void *new, int bytes,
+				  int offset_in_pte)
+{
+	if (sp->role.level != PT_PAGE_TABLE_LEVEL) {
+		++vcpu->kvm->stat.mmu_pde_zapped;
+		return;
+	}
+
+	++vcpu->kvm->stat.mmu_pte_updated;
+	if (sp->role.glevels == PT32_ROOT_LEVEL)
+		paging32_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
+	else
+		paging64_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
+}
+
+static bool need_remote_flush(u64 old, u64 new)
+{
+	if (!is_shadow_present_pte(old))
+		return false;
+	if (!is_shadow_present_pte(new))
+		return true;
+	if ((old ^ new) & PT64_BASE_ADDR_MASK)
+		return true;
+	old ^= PT64_NX_MASK;
+	new ^= PT64_NX_MASK;
+	return (old & ~new & PT64_PERM_MASK) != 0;
+}
+
+static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, u64 old, u64 new)
+{
+	if (need_remote_flush(old, new))
+		kvm_flush_remote_tlbs(vcpu->kvm);
+	else
+		kvm_mmu_flush_tlb(vcpu);
+}
+
+static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
+{
+	u64 *spte = vcpu->arch.last_pte_updated;
+
+	return !!(spte && (*spte & PT_ACCESSED_MASK));
+}
+
+void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
+		       const u8 *new, int bytes)
+{
+	gfn_t gfn = gpa >> PAGE_SHIFT;
+	struct kvm_mmu_page *sp;
+	struct hlist_node *node, *n;
+	struct hlist_head *bucket;
+	unsigned index;
+	u64 entry;
+	u64 *spte;
+	unsigned offset = offset_in_page(gpa);
+	unsigned pte_size;
+	unsigned page_offset;
+	unsigned misaligned;
+	unsigned quadrant;
+	int level;
+	int flooded = 0;
+	int npte;
+
+	pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
+	++vcpu->kvm->stat.mmu_pte_write;
+	kvm_mmu_audit(vcpu, "pre pte write");
+	if (gfn == vcpu->arch.last_pt_write_gfn
+	    && !last_updated_pte_accessed(vcpu)) {
+		++vcpu->arch.last_pt_write_count;
+		if (vcpu->arch.last_pt_write_count >= 3)
+			flooded = 1;
+	} else {
+		vcpu->arch.last_pt_write_gfn = gfn;
+		vcpu->arch.last_pt_write_count = 1;
+		vcpu->arch.last_pte_updated = NULL;
+	}
+	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
+	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
+	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) {
+		if (sp->gfn != gfn || sp->role.metaphysical)
+			continue;
+		pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
+		misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1);
+		misaligned |= bytes < 4;
+		if (misaligned || flooded) {
+			/*
+			 * Misaligned accesses are too much trouble to fix
+			 * up; also, they usually indicate a page is not used
+			 * as a page table.
+			 *
+			 * If we're seeing too many writes to a page,
+			 * it may no longer be a page table, or we may be
+			 * forking, in which case it is better to unmap the
+			 * page.
+			 */
+			pgprintk("misaligned: gpa %llx bytes %d role %x\n",
+				 gpa, bytes, sp->role.word);
+			kvm_mmu_zap_page(vcpu->kvm, sp);
+			++vcpu->kvm->stat.mmu_flooded;
+			continue;
+		}
+		page_offset = offset;
+		level = sp->role.level;
+		npte = 1;
+		if (sp->role.glevels == PT32_ROOT_LEVEL) {
+			page_offset <<= 1;	/* 32->64 */
+			/*
+			 * A 32-bit pde maps 4MB while the shadow pdes map
+			 * only 2MB.  So we need to double the offset again
+			 * and zap two pdes instead of one.
+			 */
+			if (level == PT32_ROOT_LEVEL) {
+				page_offset &= ~7; /* kill rounding error */
+				page_offset <<= 1;
+				npte = 2;
+			}
+			quadrant = page_offset >> PAGE_SHIFT;
+			page_offset &= ~PAGE_MASK;
+			if (quadrant != sp->role.quadrant)
+				continue;
+		}
+		spte = &sp->spt[page_offset / sizeof(*spte)];
+		while (npte--) {
+			entry = *spte;
+			mmu_pte_write_zap_pte(vcpu, sp, spte);
+			mmu_pte_write_new_pte(vcpu, sp, spte, new, bytes,
+					      page_offset & (pte_size - 1));
+			mmu_pte_write_flush_tlb(vcpu, entry, *spte);
+			++spte;
+		}
+	}
+	kvm_mmu_audit(vcpu, "post pte write");
+}
+
+int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
+{
+	gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
+
+	return kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+}
+
+void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
+{
+	while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
+		struct kvm_mmu_page *sp;
+
+		sp = container_of(vcpu->kvm->arch.active_mmu_pages.prev,
+				  struct kvm_mmu_page, link);
+		kvm_mmu_zap_page(vcpu->kvm, sp);
+		++vcpu->kvm->stat.mmu_recycled;
+	}
+}
+
+int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
+{
+	int r;
+	enum emulation_result er;
+
+	mutex_lock(&vcpu->kvm->lock);
+	r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code);
+	if (r < 0)
+		goto out;
+
+	if (!r) {
+		r = 1;
+		goto out;
+	}
+
+	r = mmu_topup_memory_caches(vcpu);
+	if (r)
+		goto out;
+
+	er = emulate_instruction(vcpu, vcpu->run, cr2, error_code, 0);
+	mutex_unlock(&vcpu->kvm->lock);
+
+	switch (er) {
+	case EMULATE_DONE:
+		return 1;
+	case EMULATE_DO_MMIO:
+		++vcpu->stat.mmio_exits;
+		return 0;
+	case EMULATE_FAIL:
+		kvm_report_emulation_failure(vcpu, "pagetable");
+		return 1;
+	default:
+		BUG();
+	}
+out:
+	mutex_unlock(&vcpu->kvm->lock);
+	return r;
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_page_fault);
+
+static void free_mmu_pages(struct kvm_vcpu *vcpu)
+{
+	struct kvm_mmu_page *sp;
+
+	while (!list_empty(&vcpu->kvm->arch.active_mmu_pages)) {
+		sp = container_of(vcpu->kvm->arch.active_mmu_pages.next,
+				  struct kvm_mmu_page, link);
+		kvm_mmu_zap_page(vcpu->kvm, sp);
+	}
+	free_page((unsigned long)vcpu->arch.mmu.pae_root);
+}
+
+static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
+{
+	struct page *page;
+	int i;
+
+	ASSERT(vcpu);
+
+	if (vcpu->kvm->arch.n_requested_mmu_pages)
+		vcpu->kvm->arch.n_free_mmu_pages =
+					vcpu->kvm->arch.n_requested_mmu_pages;
+	else
+		vcpu->kvm->arch.n_free_mmu_pages =
+					vcpu->kvm->arch.n_alloc_mmu_pages;
+	/*
+	 * When emulating 32-bit mode, cr3 is only 32 bits even on x86_64.
+	 * Therefore we need to allocate shadow page tables in the first
+	 * 4GB of memory, which happens to fit the DMA32 zone.
+	 */
+	page = alloc_page(GFP_KERNEL | __GFP_DMA32);
+	if (!page)
+		goto error_1;
+	vcpu->arch.mmu.pae_root = page_address(page);
+	for (i = 0; i < 4; ++i)
+		vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
+
+	return 0;
+
+error_1:
+	free_mmu_pages(vcpu);
+	return -ENOMEM;
+}
+
+int kvm_mmu_create(struct kvm_vcpu *vcpu)
+{
+	ASSERT(vcpu);
+	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
+
+	return alloc_mmu_pages(vcpu);
+}
+
+int kvm_mmu_setup(struct kvm_vcpu *vcpu)
+{
+	ASSERT(vcpu);
+	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
+
+	return init_kvm_mmu(vcpu);
+}
+
+void kvm_mmu_destroy(struct kvm_vcpu *vcpu)
+{
+	ASSERT(vcpu);
+
+	destroy_kvm_mmu(vcpu);
+	free_mmu_pages(vcpu);
+	mmu_free_memory_caches(vcpu);
+}
+
+void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
+{
+	struct kvm_mmu_page *sp;
+
+	list_for_each_entry(sp, &kvm->arch.active_mmu_pages, link) {
+		int i;
+		u64 *pt;
+
+		if (!test_bit(slot, &sp->slot_bitmap))
+			continue;
+
+		pt = sp->spt;
+		for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
+			/* avoid RMW */
+			if (pt[i] & PT_WRITABLE_MASK)
+				pt[i] &= ~PT_WRITABLE_MASK;
+	}
+}
+
+void kvm_mmu_zap_all(struct kvm *kvm)
+{
+	struct kvm_mmu_page *sp, *node;
+
+	list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
+		kvm_mmu_zap_page(kvm, sp);
+
+	kvm_flush_remote_tlbs(kvm);
+}
+
+void kvm_mmu_module_exit(void)
+{
+	if (pte_chain_cache)
+		kmem_cache_destroy(pte_chain_cache);
+	if (rmap_desc_cache)
+		kmem_cache_destroy(rmap_desc_cache);
+	if (mmu_page_header_cache)
+		kmem_cache_destroy(mmu_page_header_cache);
+}
+
+int kvm_mmu_module_init(void)
+{
+	pte_chain_cache = kmem_cache_create("kvm_pte_chain",
+					    sizeof(struct kvm_pte_chain),
+					    0, 0, NULL);
+	if (!pte_chain_cache)
+		goto nomem;
+	rmap_desc_cache = kmem_cache_create("kvm_rmap_desc",
+					    sizeof(struct kvm_rmap_desc),
+					    0, 0, NULL);
+	if (!rmap_desc_cache)
+		goto nomem;
+
+	mmu_page_header_cache = kmem_cache_create("kvm_mmu_page_header",
+						  sizeof(struct kvm_mmu_page),
+						  0, 0, NULL);
+	if (!mmu_page_header_cache)
+		goto nomem;
+
+	return 0;
+
+nomem:
+	kvm_mmu_module_exit();
+	return -ENOMEM;
+}
+
+/*
+ * Caculate mmu pages needed for kvm.
+ */
+unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm)
+{
+	int i;
+	unsigned int nr_mmu_pages;
+	unsigned int  nr_pages = 0;
+
+	for (i = 0; i < kvm->nmemslots; i++)
+		nr_pages += kvm->memslots[i].npages;
+
+	nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000;
+	nr_mmu_pages = max(nr_mmu_pages,
+			(unsigned int) KVM_MIN_ALLOC_MMU_PAGES);
+
+	return nr_mmu_pages;
+}
+
+#ifdef AUDIT
+
+static const char *audit_msg;
+
+static gva_t canonicalize(gva_t gva)
+{
+#ifdef CONFIG_X86_64
+	gva = (long long)(gva << 16) >> 16;
+#endif
+	return gva;
+}
+
+static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
+				gva_t va, int level)
+{
+	u64 *pt = __va(page_pte & PT64_BASE_ADDR_MASK);
+	int i;
+	gva_t va_delta = 1ul << (PAGE_SHIFT + 9 * (level - 1));
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i, va += va_delta) {
+		u64 ent = pt[i];
+
+		if (ent == shadow_trap_nonpresent_pte)
+			continue;
+
+		va = canonicalize(va);
+		if (level > 1) {
+			if (ent == shadow_notrap_nonpresent_pte)
+				printk(KERN_ERR "audit: (%s) nontrapping pte"
+				       " in nonleaf level: levels %d gva %lx"
+				       " level %d pte %llx\n", audit_msg,
+				       vcpu->arch.mmu.root_level, va, level, ent);
+
+			audit_mappings_page(vcpu, ent, va, level - 1);
+		} else {
+			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);
+			struct page *page = gpa_to_page(vcpu, gpa);
+			hpa_t hpa = page_to_phys(page);
+
+			if (is_shadow_present_pte(ent)
+			    && (ent & PT64_BASE_ADDR_MASK) != hpa)
+				printk(KERN_ERR "xx audit error: (%s) levels %d"
+				       " gva %lx gpa %llx hpa %llx ent %llx %d\n",
+				       audit_msg, vcpu->arch.mmu.root_level,
+				       va, gpa, hpa, ent,
+				       is_shadow_present_pte(ent));
+			else if (ent == shadow_notrap_nonpresent_pte
+				 && !is_error_hpa(hpa))
+				printk(KERN_ERR "audit: (%s) notrap shadow,"
+				       " valid guest gva %lx\n", audit_msg, va);
+			kvm_release_page_clean(page);
+
+		}
+	}
+}
+
+static void audit_mappings(struct kvm_vcpu *vcpu)
+{
+	unsigned i;
+
+	if (vcpu->arch.mmu.root_level == 4)
+		audit_mappings_page(vcpu, vcpu->arch.mmu.root_hpa, 0, 4);
+	else
+		for (i = 0; i < 4; ++i)
+			if (vcpu->arch.mmu.pae_root[i] & PT_PRESENT_MASK)
+				audit_mappings_page(vcpu,
+						    vcpu->arch.mmu.pae_root[i],
+						    i << 30,
+						    2);
+}
+
+static int count_rmaps(struct kvm_vcpu *vcpu)
+{
+	int nmaps = 0;
+	int i, j, k;
+
+	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
+		struct kvm_memory_slot *m = &vcpu->kvm->memslots[i];
+		struct kvm_rmap_desc *d;
+
+		for (j = 0; j < m->npages; ++j) {
+			unsigned long *rmapp = &m->rmap[j];
+
+			if (!*rmapp)
+				continue;
+			if (!(*rmapp & 1)) {
+				++nmaps;
+				continue;
+			}
+			d = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
+			while (d) {
+				for (k = 0; k < RMAP_EXT; ++k)
+					if (d->shadow_ptes[k])
+						++nmaps;
+					else
+						break;
+				d = d->more;
+			}
+		}
+	}
+	return nmaps;
+}
+
+static int count_writable_mappings(struct kvm_vcpu *vcpu)
+{
+	int nmaps = 0;
+	struct kvm_mmu_page *sp;
+	int i;
+
+	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
+		u64 *pt = sp->spt;
+
+		if (sp->role.level != PT_PAGE_TABLE_LEVEL)
+			continue;
+
+		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+			u64 ent = pt[i];
+
+			if (!(ent & PT_PRESENT_MASK))
+				continue;
+			if (!(ent & PT_WRITABLE_MASK))
+				continue;
+			++nmaps;
+		}
+	}
+	return nmaps;
+}
+
+static void audit_rmap(struct kvm_vcpu *vcpu)
+{
+	int n_rmap = count_rmaps(vcpu);
+	int n_actual = count_writable_mappings(vcpu);
+
+	if (n_rmap != n_actual)
+		printk(KERN_ERR "%s: (%s) rmap %d actual %d\n",
+		       __FUNCTION__, audit_msg, n_rmap, n_actual);
+}
+
+static void audit_write_protection(struct kvm_vcpu *vcpu)
+{
+	struct kvm_mmu_page *sp;
+	struct kvm_memory_slot *slot;
+	unsigned long *rmapp;
+	gfn_t gfn;
+
+	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
+		if (sp->role.metaphysical)
+			continue;
+
+		slot = gfn_to_memslot(vcpu->kvm, sp->gfn);
+		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
+		rmapp = &slot->rmap[gfn - slot->base_gfn];
+		if (*rmapp)
+			printk(KERN_ERR "%s: (%s) shadow page has writable"
+			       " mappings: gfn %lx role %x\n",
+			       __FUNCTION__, audit_msg, sp->gfn,
+			       sp->role.word);
+	}
+}
+
+static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg)
+{
+	int olddbg = dbg;
+
+	dbg = 0;
+	audit_msg = msg;
+	audit_rmap(vcpu);
+	audit_write_protection(vcpu);
+	audit_mappings(vcpu);
+	dbg = olddbg;
+}
+
+#endif
diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
new file mode 100644
index 000000000000..1fce19ec7a23
--- /dev/null
+++ b/arch/x86/kvm/mmu.h
@@ -0,0 +1,44 @@
+#ifndef __KVM_X86_MMU_H
+#define __KVM_X86_MMU_H
+
+#include <linux/kvm_host.h>
+
+static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
+{
+	if (unlikely(vcpu->kvm->arch.n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
+		__kvm_mmu_free_some_pages(vcpu);
+}
+
+static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
+{
+	if (likely(vcpu->arch.mmu.root_hpa != INVALID_PAGE))
+		return 0;
+
+	return kvm_mmu_load(vcpu);
+}
+
+static inline int is_long_mode(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_X86_64
+	return vcpu->arch.shadow_efer & EFER_LME;
+#else
+	return 0;
+#endif
+}
+
+static inline int is_pae(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.cr4 & X86_CR4_PAE;
+}
+
+static inline int is_pse(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.cr4 & X86_CR4_PSE;
+}
+
+static inline int is_paging(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.cr0 & X86_CR0_PG;
+}
+
+#endif
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
new file mode 100644
index 000000000000..56b88f7e83ef
--- /dev/null
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -0,0 +1,461 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * This module enables machines with Intel VT-x extensions to run virtual
+ * machines without emulation or binary translation.
+ *
+ * MMU support
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ *
+ * Authors:
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *   Avi Kivity   <avi@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+/*
+ * We need the mmu code to access both 32-bit and 64-bit guest ptes,
+ * so the code in this file is compiled twice, once per pte size.
+ */
+
+#if PTTYPE == 64
+	#define pt_element_t u64
+	#define guest_walker guest_walker64
+	#define FNAME(name) paging##64_##name
+	#define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK
+	#define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK
+	#define PT_INDEX(addr, level) PT64_INDEX(addr, level)
+	#define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
+	#define PT_LEVEL_MASK(level) PT64_LEVEL_MASK(level)
+	#define PT_LEVEL_BITS PT64_LEVEL_BITS
+	#ifdef CONFIG_X86_64
+	#define PT_MAX_FULL_LEVELS 4
+	#define CMPXCHG cmpxchg
+	#else
+	#define CMPXCHG cmpxchg64
+	#define PT_MAX_FULL_LEVELS 2
+	#endif
+#elif PTTYPE == 32
+	#define pt_element_t u32
+	#define guest_walker guest_walker32
+	#define FNAME(name) paging##32_##name
+	#define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK
+	#define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK
+	#define PT_INDEX(addr, level) PT32_INDEX(addr, level)
+	#define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
+	#define PT_LEVEL_MASK(level) PT32_LEVEL_MASK(level)
+	#define PT_LEVEL_BITS PT32_LEVEL_BITS
+	#define PT_MAX_FULL_LEVELS 2
+	#define CMPXCHG cmpxchg
+#else
+	#error Invalid PTTYPE value
+#endif
+
+#define gpte_to_gfn FNAME(gpte_to_gfn)
+#define gpte_to_gfn_pde FNAME(gpte_to_gfn_pde)
+
+/*
+ * The guest_walker structure emulates the behavior of the hardware page
+ * table walker.
+ */
+struct guest_walker {
+	int level;
+	gfn_t table_gfn[PT_MAX_FULL_LEVELS];
+	pt_element_t ptes[PT_MAX_FULL_LEVELS];
+	gpa_t pte_gpa[PT_MAX_FULL_LEVELS];
+	unsigned pt_access;
+	unsigned pte_access;
+	gfn_t gfn;
+	u32 error_code;
+};
+
+static gfn_t gpte_to_gfn(pt_element_t gpte)
+{
+	return (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT;
+}
+
+static gfn_t gpte_to_gfn_pde(pt_element_t gpte)
+{
+	return (gpte & PT_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT;
+}
+
+static bool FNAME(cmpxchg_gpte)(struct kvm *kvm,
+			 gfn_t table_gfn, unsigned index,
+			 pt_element_t orig_pte, pt_element_t new_pte)
+{
+	pt_element_t ret;
+	pt_element_t *table;
+	struct page *page;
+
+	page = gfn_to_page(kvm, table_gfn);
+	table = kmap_atomic(page, KM_USER0);
+
+	ret = CMPXCHG(&table[index], orig_pte, new_pte);
+
+	kunmap_atomic(table, KM_USER0);
+
+	kvm_release_page_dirty(page);
+
+	return (ret != orig_pte);
+}
+
+static unsigned FNAME(gpte_access)(struct kvm_vcpu *vcpu, pt_element_t gpte)
+{
+	unsigned access;
+
+	access = (gpte & (PT_WRITABLE_MASK | PT_USER_MASK)) | ACC_EXEC_MASK;
+#if PTTYPE == 64
+	if (is_nx(vcpu))
+		access &= ~(gpte >> PT64_NX_SHIFT);
+#endif
+	return access;
+}
+
+/*
+ * Fetch a guest pte for a guest virtual address
+ */
+static int FNAME(walk_addr)(struct guest_walker *walker,
+			    struct kvm_vcpu *vcpu, gva_t addr,
+			    int write_fault, int user_fault, int fetch_fault)
+{
+	pt_element_t pte;
+	gfn_t table_gfn;
+	unsigned index, pt_access, pte_access;
+	gpa_t pte_gpa;
+
+	pgprintk("%s: addr %lx\n", __FUNCTION__, addr);
+walk:
+	walker->level = vcpu->arch.mmu.root_level;
+	pte = vcpu->arch.cr3;
+#if PTTYPE == 64
+	if (!is_long_mode(vcpu)) {
+		pte = vcpu->arch.pdptrs[(addr >> 30) & 3];
+		if (!is_present_pte(pte))
+			goto not_present;
+		--walker->level;
+	}
+#endif
+	ASSERT((!is_long_mode(vcpu) && is_pae(vcpu)) ||
+	       (vcpu->cr3 & CR3_NONPAE_RESERVED_BITS) == 0);
+
+	pt_access = ACC_ALL;
+
+	for (;;) {
+		index = PT_INDEX(addr, walker->level);
+
+		table_gfn = gpte_to_gfn(pte);
+		pte_gpa = gfn_to_gpa(table_gfn);
+		pte_gpa += index * sizeof(pt_element_t);
+		walker->table_gfn[walker->level - 1] = table_gfn;
+		walker->pte_gpa[walker->level - 1] = pte_gpa;
+		pgprintk("%s: table_gfn[%d] %lx\n", __FUNCTION__,
+			 walker->level - 1, table_gfn);
+
+		kvm_read_guest(vcpu->kvm, pte_gpa, &pte, sizeof(pte));
+
+		if (!is_present_pte(pte))
+			goto not_present;
+
+		if (write_fault && !is_writeble_pte(pte))
+			if (user_fault || is_write_protection(vcpu))
+				goto access_error;
+
+		if (user_fault && !(pte & PT_USER_MASK))
+			goto access_error;
+
+#if PTTYPE == 64
+		if (fetch_fault && is_nx(vcpu) && (pte & PT64_NX_MASK))
+			goto access_error;
+#endif
+
+		if (!(pte & PT_ACCESSED_MASK)) {
+			mark_page_dirty(vcpu->kvm, table_gfn);
+			if (FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn,
+			    index, pte, pte|PT_ACCESSED_MASK))
+				goto walk;
+			pte |= PT_ACCESSED_MASK;
+		}
+
+		pte_access = pt_access & FNAME(gpte_access)(vcpu, pte);
+
+		walker->ptes[walker->level - 1] = pte;
+
+		if (walker->level == PT_PAGE_TABLE_LEVEL) {
+			walker->gfn = gpte_to_gfn(pte);
+			break;
+		}
+
+		if (walker->level == PT_DIRECTORY_LEVEL
+		    && (pte & PT_PAGE_SIZE_MASK)
+		    && (PTTYPE == 64 || is_pse(vcpu))) {
+			walker->gfn = gpte_to_gfn_pde(pte);
+			walker->gfn += PT_INDEX(addr, PT_PAGE_TABLE_LEVEL);
+			if (PTTYPE == 32 && is_cpuid_PSE36())
+				walker->gfn += pse36_gfn_delta(pte);
+			break;
+		}
+
+		pt_access = pte_access;
+		--walker->level;
+	}
+
+	if (write_fault && !is_dirty_pte(pte)) {
+		bool ret;
+
+		mark_page_dirty(vcpu->kvm, table_gfn);
+		ret = FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn, index, pte,
+			    pte|PT_DIRTY_MASK);
+		if (ret)
+			goto walk;
+		pte |= PT_DIRTY_MASK;
+		kvm_mmu_pte_write(vcpu, pte_gpa, (u8 *)&pte, sizeof(pte));
+		walker->ptes[walker->level - 1] = pte;
+	}
+
+	walker->pt_access = pt_access;
+	walker->pte_access = pte_access;
+	pgprintk("%s: pte %llx pte_access %x pt_access %x\n",
+		 __FUNCTION__, (u64)pte, pt_access, pte_access);
+	return 1;
+
+not_present:
+	walker->error_code = 0;
+	goto err;
+
+access_error:
+	walker->error_code = PFERR_PRESENT_MASK;
+
+err:
+	if (write_fault)
+		walker->error_code |= PFERR_WRITE_MASK;
+	if (user_fault)
+		walker->error_code |= PFERR_USER_MASK;
+	if (fetch_fault)
+		walker->error_code |= PFERR_FETCH_MASK;
+	return 0;
+}
+
+static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
+			      u64 *spte, const void *pte, int bytes,
+			      int offset_in_pte)
+{
+	pt_element_t gpte;
+	unsigned pte_access;
+
+	gpte = *(const pt_element_t *)pte;
+	if (~gpte & (PT_PRESENT_MASK | PT_ACCESSED_MASK)) {
+		if (!offset_in_pte && !is_present_pte(gpte))
+			set_shadow_pte(spte, shadow_notrap_nonpresent_pte);
+		return;
+	}
+	if (bytes < sizeof(pt_element_t))
+		return;
+	pgprintk("%s: gpte %llx spte %p\n", __FUNCTION__, (u64)gpte, spte);
+	pte_access = page->role.access & FNAME(gpte_access)(vcpu, gpte);
+	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
+		     gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte));
+}
+
+/*
+ * Fetch a shadow pte for a specific level in the paging hierarchy.
+ */
+static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
+			 struct guest_walker *walker,
+			 int user_fault, int write_fault, int *ptwrite)
+{
+	hpa_t shadow_addr;
+	int level;
+	u64 *shadow_ent;
+	unsigned access = walker->pt_access;
+
+	if (!is_present_pte(walker->ptes[walker->level - 1]))
+		return NULL;
+
+	shadow_addr = vcpu->arch.mmu.root_hpa;
+	level = vcpu->arch.mmu.shadow_root_level;
+	if (level == PT32E_ROOT_LEVEL) {
+		shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
+		shadow_addr &= PT64_BASE_ADDR_MASK;
+		--level;
+	}
+
+	for (; ; level--) {
+		u32 index = SHADOW_PT_INDEX(addr, level);
+		struct kvm_mmu_page *shadow_page;
+		u64 shadow_pte;
+		int metaphysical;
+		gfn_t table_gfn;
+		bool new_page = 0;
+
+		shadow_ent = ((u64 *)__va(shadow_addr)) + index;
+		if (is_shadow_present_pte(*shadow_ent)) {
+			if (level == PT_PAGE_TABLE_LEVEL)
+				break;
+			shadow_addr = *shadow_ent & PT64_BASE_ADDR_MASK;
+			continue;
+		}
+
+		if (level == PT_PAGE_TABLE_LEVEL)
+			break;
+
+		if (level - 1 == PT_PAGE_TABLE_LEVEL
+		    && walker->level == PT_DIRECTORY_LEVEL) {
+			metaphysical = 1;
+			if (!is_dirty_pte(walker->ptes[level - 1]))
+				access &= ~ACC_WRITE_MASK;
+			table_gfn = gpte_to_gfn(walker->ptes[level - 1]);
+		} else {
+			metaphysical = 0;
+			table_gfn = walker->table_gfn[level - 2];
+		}
+		shadow_page = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1,
+					       metaphysical, access,
+					       shadow_ent, &new_page);
+		if (new_page && !metaphysical) {
+			pt_element_t curr_pte;
+			kvm_read_guest(vcpu->kvm, walker->pte_gpa[level - 2],
+				       &curr_pte, sizeof(curr_pte));
+			if (curr_pte != walker->ptes[level - 2])
+				return NULL;
+		}
+		shadow_addr = __pa(shadow_page->spt);
+		shadow_pte = shadow_addr | PT_PRESENT_MASK | PT_ACCESSED_MASK
+			| PT_WRITABLE_MASK | PT_USER_MASK;
+		*shadow_ent = shadow_pte;
+	}
+
+	mmu_set_spte(vcpu, shadow_ent, access, walker->pte_access & access,
+		     user_fault, write_fault,
+		     walker->ptes[walker->level-1] & PT_DIRTY_MASK,
+		     ptwrite, walker->gfn);
+
+	return shadow_ent;
+}
+
+/*
+ * Page fault handler.  There are several causes for a page fault:
+ *   - there is no shadow pte for the guest pte
+ *   - write access through a shadow pte marked read only so that we can set
+ *     the dirty bit
+ *   - write access to a shadow pte marked read only so we can update the page
+ *     dirty bitmap, when userspace requests it
+ *   - mmio access; in this case we will never install a present shadow pte
+ *   - normal guest page fault due to the guest pte marked not present, not
+ *     writable, or not executable
+ *
+ *  Returns: 1 if we need to emulate the instruction, 0 otherwise, or
+ *           a negative value on error.
+ */
+static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
+			       u32 error_code)
+{
+	int write_fault = error_code & PFERR_WRITE_MASK;
+	int user_fault = error_code & PFERR_USER_MASK;
+	int fetch_fault = error_code & PFERR_FETCH_MASK;
+	struct guest_walker walker;
+	u64 *shadow_pte;
+	int write_pt = 0;
+	int r;
+
+	pgprintk("%s: addr %lx err %x\n", __FUNCTION__, addr, error_code);
+	kvm_mmu_audit(vcpu, "pre page fault");
+
+	r = mmu_topup_memory_caches(vcpu);
+	if (r)
+		return r;
+
+	/*
+	 * Look up the shadow pte for the faulting address.
+	 */
+	r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault,
+			     fetch_fault);
+
+	/*
+	 * The page is not mapped by the guest.  Let the guest handle it.
+	 */
+	if (!r) {
+		pgprintk("%s: guest page fault\n", __FUNCTION__);
+		inject_page_fault(vcpu, addr, walker.error_code);
+		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
+		return 0;
+	}
+
+	shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
+				  &write_pt);
+	pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
+		 shadow_pte, *shadow_pte, write_pt);
+
+	if (!write_pt)
+		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
+
+	/*
+	 * mmio: emulate if accessible, otherwise its a guest fault.
+	 */
+	if (shadow_pte && is_io_pte(*shadow_pte))
+		return 1;
+
+	++vcpu->stat.pf_fixed;
+	kvm_mmu_audit(vcpu, "post page fault (fixed)");
+
+	return write_pt;
+}
+
+static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr)
+{
+	struct guest_walker walker;
+	gpa_t gpa = UNMAPPED_GVA;
+	int r;
+
+	r = FNAME(walk_addr)(&walker, vcpu, vaddr, 0, 0, 0);
+
+	if (r) {
+		gpa = gfn_to_gpa(walker.gfn);
+		gpa |= vaddr & ~PAGE_MASK;
+	}
+
+	return gpa;
+}
+
+static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
+				 struct kvm_mmu_page *sp)
+{
+	int i, offset = 0;
+	pt_element_t *gpt;
+	struct page *page;
+
+	if (sp->role.metaphysical
+	    || (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) {
+		nonpaging_prefetch_page(vcpu, sp);
+		return;
+	}
+
+	if (PTTYPE == 32)
+		offset = sp->role.quadrant << PT64_LEVEL_BITS;
+	page = gfn_to_page(vcpu->kvm, sp->gfn);
+	gpt = kmap_atomic(page, KM_USER0);
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
+		if (is_present_pte(gpt[offset + i]))
+			sp->spt[i] = shadow_trap_nonpresent_pte;
+		else
+			sp->spt[i] = shadow_notrap_nonpresent_pte;
+	kunmap_atomic(gpt, KM_USER0);
+	kvm_release_page_clean(page);
+}
+
+#undef pt_element_t
+#undef guest_walker
+#undef FNAME
+#undef PT_BASE_ADDR_MASK
+#undef PT_INDEX
+#undef SHADOW_PT_INDEX
+#undef PT_LEVEL_MASK
+#undef PT_DIR_BASE_ADDR_MASK
+#undef PT_LEVEL_BITS
+#undef PT_MAX_FULL_LEVELS
+#undef gpte_to_gfn
+#undef gpte_to_gfn_pde
+#undef CMPXCHG
diff --git a/arch/x86/kvm/segment_descriptor.h b/arch/x86/kvm/segment_descriptor.h
new file mode 100644
index 000000000000..56fc4c873389
--- /dev/null
+++ b/arch/x86/kvm/segment_descriptor.h
@@ -0,0 +1,29 @@
+#ifndef __SEGMENT_DESCRIPTOR_H
+#define __SEGMENT_DESCRIPTOR_H
+
+struct segment_descriptor {
+	u16 limit_low;
+	u16 base_low;
+	u8  base_mid;
+	u8  type : 4;
+	u8  system : 1;
+	u8  dpl : 2;
+	u8  present : 1;
+	u8  limit_high : 4;
+	u8  avl : 1;
+	u8  long_mode : 1;
+	u8  default_op : 1;
+	u8  granularity : 1;
+	u8  base_high;
+} __attribute__((packed));
+
+#ifdef CONFIG_X86_64
+/* LDT or TSS descriptor in the GDT. 16 bytes. */
+struct segment_descriptor_64 {
+	struct segment_descriptor s;
+	u32 base_higher;
+	u32 pad_zero;
+};
+
+#endif
+#endif
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
new file mode 100644
index 000000000000..3d4b71a94440
--- /dev/null
+++ b/arch/x86/kvm/svm.c
@@ -0,0 +1,1725 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * AMD SVM support
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ *
+ * Authors:
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *   Avi Kivity   <avi@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+#include <linux/kvm_host.h>
+
+#include "kvm_svm.h"
+#include "irq.h"
+#include "mmu.h"
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/vmalloc.h>
+#include <linux/highmem.h>
+#include <linux/sched.h>
+
+#include <asm/desc.h>
+
+MODULE_AUTHOR("Qumranet");
+MODULE_LICENSE("GPL");
+
+#define IOPM_ALLOC_ORDER 2
+#define MSRPM_ALLOC_ORDER 1
+
+#define DB_VECTOR 1
+#define UD_VECTOR 6
+#define GP_VECTOR 13
+
+#define DR7_GD_MASK (1 << 13)
+#define DR6_BD_MASK (1 << 13)
+
+#define SEG_TYPE_LDT 2
+#define SEG_TYPE_BUSY_TSS16 3
+
+#define SVM_FEATURE_NPT  (1 << 0)
+#define SVM_FEATURE_LBRV (1 << 1)
+#define SVM_DEATURE_SVML (1 << 2)
+
+static void kvm_reput_irq(struct vcpu_svm *svm);
+
+static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu)
+{
+	return container_of(vcpu, struct vcpu_svm, vcpu);
+}
+
+unsigned long iopm_base;
+unsigned long msrpm_base;
+
+struct kvm_ldttss_desc {
+	u16 limit0;
+	u16 base0;
+	unsigned base1 : 8, type : 5, dpl : 2, p : 1;
+	unsigned limit1 : 4, zero0 : 3, g : 1, base2 : 8;
+	u32 base3;
+	u32 zero1;
+} __attribute__((packed));
+
+struct svm_cpu_data {
+	int cpu;
+
+	u64 asid_generation;
+	u32 max_asid;
+	u32 next_asid;
+	struct kvm_ldttss_desc *tss_desc;
+
+	struct page *save_area;
+};
+
+static DEFINE_PER_CPU(struct svm_cpu_data *, svm_data);
+static uint32_t svm_features;
+
+struct svm_init_data {
+	int cpu;
+	int r;
+};
+
+static u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000};
+
+#define NUM_MSR_MAPS ARRAY_SIZE(msrpm_ranges)
+#define MSRS_RANGE_SIZE 2048
+#define MSRS_IN_RANGE (MSRS_RANGE_SIZE * 8 / 2)
+
+#define MAX_INST_SIZE 15
+
+static inline u32 svm_has(u32 feat)
+{
+	return svm_features & feat;
+}
+
+static inline u8 pop_irq(struct kvm_vcpu *vcpu)
+{
+	int word_index = __ffs(vcpu->arch.irq_summary);
+	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
+	int irq = word_index * BITS_PER_LONG + bit_index;
+
+	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+	if (!vcpu->arch.irq_pending[word_index])
+		clear_bit(word_index, &vcpu->arch.irq_summary);
+	return irq;
+}
+
+static inline void push_irq(struct kvm_vcpu *vcpu, u8 irq)
+{
+	set_bit(irq, vcpu->arch.irq_pending);
+	set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
+}
+
+static inline void clgi(void)
+{
+	asm volatile (SVM_CLGI);
+}
+
+static inline void stgi(void)
+{
+	asm volatile (SVM_STGI);
+}
+
+static inline void invlpga(unsigned long addr, u32 asid)
+{
+	asm volatile (SVM_INVLPGA :: "a"(addr), "c"(asid));
+}
+
+static inline unsigned long kvm_read_cr2(void)
+{
+	unsigned long cr2;
+
+	asm volatile ("mov %%cr2, %0" : "=r" (cr2));
+	return cr2;
+}
+
+static inline void kvm_write_cr2(unsigned long val)
+{
+	asm volatile ("mov %0, %%cr2" :: "r" (val));
+}
+
+static inline unsigned long read_dr6(void)
+{
+	unsigned long dr6;
+
+	asm volatile ("mov %%dr6, %0" : "=r" (dr6));
+	return dr6;
+}
+
+static inline void write_dr6(unsigned long val)
+{
+	asm volatile ("mov %0, %%dr6" :: "r" (val));
+}
+
+static inline unsigned long read_dr7(void)
+{
+	unsigned long dr7;
+
+	asm volatile ("mov %%dr7, %0" : "=r" (dr7));
+	return dr7;
+}
+
+static inline void write_dr7(unsigned long val)
+{
+	asm volatile ("mov %0, %%dr7" :: "r" (val));
+}
+
+static inline void force_new_asid(struct kvm_vcpu *vcpu)
+{
+	to_svm(vcpu)->asid_generation--;
+}
+
+static inline void flush_guest_tlb(struct kvm_vcpu *vcpu)
+{
+	force_new_asid(vcpu);
+}
+
+static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer)
+{
+	if (!(efer & EFER_LMA))
+		efer &= ~EFER_LME;
+
+	to_svm(vcpu)->vmcb->save.efer = efer | MSR_EFER_SVME_MASK;
+	vcpu->arch.shadow_efer = efer;
+}
+
+static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
+				bool has_error_code, u32 error_code)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->control.event_inj = nr
+		| SVM_EVTINJ_VALID
+		| (has_error_code ? SVM_EVTINJ_VALID_ERR : 0)
+		| SVM_EVTINJ_TYPE_EXEPT;
+	svm->vmcb->control.event_inj_err = error_code;
+}
+
+static bool svm_exception_injected(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	return !(svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_VALID);
+}
+
+static int is_external_interrupt(u32 info)
+{
+	info &= SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID;
+	return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR);
+}
+
+static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	if (!svm->next_rip) {
+		printk(KERN_DEBUG "%s: NOP\n", __FUNCTION__);
+		return;
+	}
+	if (svm->next_rip - svm->vmcb->save.rip > MAX_INST_SIZE)
+		printk(KERN_ERR "%s: ip 0x%llx next 0x%llx\n",
+		       __FUNCTION__,
+		       svm->vmcb->save.rip,
+		       svm->next_rip);
+
+	vcpu->arch.rip = svm->vmcb->save.rip = svm->next_rip;
+	svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
+
+	vcpu->arch.interrupt_window_open = 1;
+}
+
+static int has_svm(void)
+{
+	uint32_t eax, ebx, ecx, edx;
+
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) {
+		printk(KERN_INFO "has_svm: not amd\n");
+		return 0;
+	}
+
+	cpuid(0x80000000, &eax, &ebx, &ecx, &edx);
+	if (eax < SVM_CPUID_FUNC) {
+		printk(KERN_INFO "has_svm: can't execute cpuid_8000000a\n");
+		return 0;
+	}
+
+	cpuid(0x80000001, &eax, &ebx, &ecx, &edx);
+	if (!(ecx & (1 << SVM_CPUID_FEATURE_SHIFT))) {
+		printk(KERN_DEBUG "has_svm: svm not available\n");
+		return 0;
+	}
+	return 1;
+}
+
+static void svm_hardware_disable(void *garbage)
+{
+	struct svm_cpu_data *svm_data
+		= per_cpu(svm_data, raw_smp_processor_id());
+
+	if (svm_data) {
+		uint64_t efer;
+
+		wrmsrl(MSR_VM_HSAVE_PA, 0);
+		rdmsrl(MSR_EFER, efer);
+		wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK);
+		per_cpu(svm_data, raw_smp_processor_id()) = NULL;
+		__free_page(svm_data->save_area);
+		kfree(svm_data);
+	}
+}
+
+static void svm_hardware_enable(void *garbage)
+{
+
+	struct svm_cpu_data *svm_data;
+	uint64_t efer;
+#ifdef CONFIG_X86_64
+	struct desc_ptr gdt_descr;
+#else
+	struct desc_ptr gdt_descr;
+#endif
+	struct desc_struct *gdt;
+	int me = raw_smp_processor_id();
+
+	if (!has_svm()) {
+		printk(KERN_ERR "svm_cpu_init: err EOPNOTSUPP on %d\n", me);
+		return;
+	}
+	svm_data = per_cpu(svm_data, me);
+
+	if (!svm_data) {
+		printk(KERN_ERR "svm_cpu_init: svm_data is NULL on %d\n",
+		       me);
+		return;
+	}
+
+	svm_data->asid_generation = 1;
+	svm_data->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1;
+	svm_data->next_asid = svm_data->max_asid + 1;
+	svm_features = cpuid_edx(SVM_CPUID_FUNC);
+
+	asm volatile ("sgdt %0" : "=m"(gdt_descr));
+	gdt = (struct desc_struct *)gdt_descr.address;
+	svm_data->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS);
+
+	rdmsrl(MSR_EFER, efer);
+	wrmsrl(MSR_EFER, efer | MSR_EFER_SVME_MASK);
+
+	wrmsrl(MSR_VM_HSAVE_PA,
+	       page_to_pfn(svm_data->save_area) << PAGE_SHIFT);
+}
+
+static int svm_cpu_init(int cpu)
+{
+	struct svm_cpu_data *svm_data;
+	int r;
+
+	svm_data = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL);
+	if (!svm_data)
+		return -ENOMEM;
+	svm_data->cpu = cpu;
+	svm_data->save_area = alloc_page(GFP_KERNEL);
+	r = -ENOMEM;
+	if (!svm_data->save_area)
+		goto err_1;
+
+	per_cpu(svm_data, cpu) = svm_data;
+
+	return 0;
+
+err_1:
+	kfree(svm_data);
+	return r;
+
+}
+
+static void set_msr_interception(u32 *msrpm, unsigned msr,
+				 int read, int write)
+{
+	int i;
+
+	for (i = 0; i < NUM_MSR_MAPS; i++) {
+		if (msr >= msrpm_ranges[i] &&
+		    msr < msrpm_ranges[i] + MSRS_IN_RANGE) {
+			u32 msr_offset = (i * MSRS_IN_RANGE + msr -
+					  msrpm_ranges[i]) * 2;
+
+			u32 *base = msrpm + (msr_offset / 32);
+			u32 msr_shift = msr_offset % 32;
+			u32 mask = ((write) ? 0 : 2) | ((read) ? 0 : 1);
+			*base = (*base & ~(0x3 << msr_shift)) |
+				(mask << msr_shift);
+			return;
+		}
+	}
+	BUG();
+}
+
+static __init int svm_hardware_setup(void)
+{
+	int cpu;
+	struct page *iopm_pages;
+	struct page *msrpm_pages;
+	void *iopm_va, *msrpm_va;
+	int r;
+
+	iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER);
+
+	if (!iopm_pages)
+		return -ENOMEM;
+
+	iopm_va = page_address(iopm_pages);
+	memset(iopm_va, 0xff, PAGE_SIZE * (1 << IOPM_ALLOC_ORDER));
+	clear_bit(0x80, iopm_va); /* allow direct access to PC debug port */
+	iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT;
+
+
+	msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
+
+	r = -ENOMEM;
+	if (!msrpm_pages)
+		goto err_1;
+
+	msrpm_va = page_address(msrpm_pages);
+	memset(msrpm_va, 0xff, PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER));
+	msrpm_base = page_to_pfn(msrpm_pages) << PAGE_SHIFT;
+
+#ifdef CONFIG_X86_64
+	set_msr_interception(msrpm_va, MSR_GS_BASE, 1, 1);
+	set_msr_interception(msrpm_va, MSR_FS_BASE, 1, 1);
+	set_msr_interception(msrpm_va, MSR_KERNEL_GS_BASE, 1, 1);
+	set_msr_interception(msrpm_va, MSR_LSTAR, 1, 1);
+	set_msr_interception(msrpm_va, MSR_CSTAR, 1, 1);
+	set_msr_interception(msrpm_va, MSR_SYSCALL_MASK, 1, 1);
+#endif
+	set_msr_interception(msrpm_va, MSR_K6_STAR, 1, 1);
+	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_CS, 1, 1);
+	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_ESP, 1, 1);
+	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_EIP, 1, 1);
+
+	for_each_online_cpu(cpu) {
+		r = svm_cpu_init(cpu);
+		if (r)
+			goto err_2;
+	}
+	return 0;
+
+err_2:
+	__free_pages(msrpm_pages, MSRPM_ALLOC_ORDER);
+	msrpm_base = 0;
+err_1:
+	__free_pages(iopm_pages, IOPM_ALLOC_ORDER);
+	iopm_base = 0;
+	return r;
+}
+
+static __exit void svm_hardware_unsetup(void)
+{
+	__free_pages(pfn_to_page(msrpm_base >> PAGE_SHIFT), MSRPM_ALLOC_ORDER);
+	__free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), IOPM_ALLOC_ORDER);
+	iopm_base = msrpm_base = 0;
+}
+
+static void init_seg(struct vmcb_seg *seg)
+{
+	seg->selector = 0;
+	seg->attrib = SVM_SELECTOR_P_MASK | SVM_SELECTOR_S_MASK |
+		SVM_SELECTOR_WRITE_MASK; /* Read/Write Data Segment */
+	seg->limit = 0xffff;
+	seg->base = 0;
+}
+
+static void init_sys_seg(struct vmcb_seg *seg, uint32_t type)
+{
+	seg->selector = 0;
+	seg->attrib = SVM_SELECTOR_P_MASK | type;
+	seg->limit = 0xffff;
+	seg->base = 0;
+}
+
+static void init_vmcb(struct vmcb *vmcb)
+{
+	struct vmcb_control_area *control = &vmcb->control;
+	struct vmcb_save_area *save = &vmcb->save;
+
+	control->intercept_cr_read = 	INTERCEPT_CR0_MASK |
+					INTERCEPT_CR3_MASK |
+					INTERCEPT_CR4_MASK |
+					INTERCEPT_CR8_MASK;
+
+	control->intercept_cr_write = 	INTERCEPT_CR0_MASK |
+					INTERCEPT_CR3_MASK |
+					INTERCEPT_CR4_MASK |
+					INTERCEPT_CR8_MASK;
+
+	control->intercept_dr_read = 	INTERCEPT_DR0_MASK |
+					INTERCEPT_DR1_MASK |
+					INTERCEPT_DR2_MASK |
+					INTERCEPT_DR3_MASK;
+
+	control->intercept_dr_write = 	INTERCEPT_DR0_MASK |
+					INTERCEPT_DR1_MASK |
+					INTERCEPT_DR2_MASK |
+					INTERCEPT_DR3_MASK |
+					INTERCEPT_DR5_MASK |
+					INTERCEPT_DR7_MASK;
+
+	control->intercept_exceptions = (1 << PF_VECTOR) |
+					(1 << UD_VECTOR);
+
+
+	control->intercept = 	(1ULL << INTERCEPT_INTR) |
+				(1ULL << INTERCEPT_NMI) |
+				(1ULL << INTERCEPT_SMI) |
+		/*
+		 * selective cr0 intercept bug?
+		 *    	0:   0f 22 d8                mov    %eax,%cr3
+		 *	3:   0f 20 c0                mov    %cr0,%eax
+		 *	6:   0d 00 00 00 80          or     $0x80000000,%eax
+		 *	b:   0f 22 c0                mov    %eax,%cr0
+		 * set cr3 ->interception
+		 * get cr0 ->interception
+		 * set cr0 -> no interception
+		 */
+		/*              (1ULL << INTERCEPT_SELECTIVE_CR0) | */
+				(1ULL << INTERCEPT_CPUID) |
+				(1ULL << INTERCEPT_INVD) |
+				(1ULL << INTERCEPT_HLT) |
+				(1ULL << INTERCEPT_INVLPGA) |
+				(1ULL << INTERCEPT_IOIO_PROT) |
+				(1ULL << INTERCEPT_MSR_PROT) |
+				(1ULL << INTERCEPT_TASK_SWITCH) |
+				(1ULL << INTERCEPT_SHUTDOWN) |
+				(1ULL << INTERCEPT_VMRUN) |
+				(1ULL << INTERCEPT_VMMCALL) |
+				(1ULL << INTERCEPT_VMLOAD) |
+				(1ULL << INTERCEPT_VMSAVE) |
+				(1ULL << INTERCEPT_STGI) |
+				(1ULL << INTERCEPT_CLGI) |
+				(1ULL << INTERCEPT_SKINIT) |
+				(1ULL << INTERCEPT_WBINVD) |
+				(1ULL << INTERCEPT_MONITOR) |
+				(1ULL << INTERCEPT_MWAIT);
+
+	control->iopm_base_pa = iopm_base;
+	control->msrpm_base_pa = msrpm_base;
+	control->tsc_offset = 0;
+	control->int_ctl = V_INTR_MASKING_MASK;
+
+	init_seg(&save->es);
+	init_seg(&save->ss);
+	init_seg(&save->ds);
+	init_seg(&save->fs);
+	init_seg(&save->gs);
+
+	save->cs.selector = 0xf000;
+	/* Executable/Readable Code Segment */
+	save->cs.attrib = SVM_SELECTOR_READ_MASK | SVM_SELECTOR_P_MASK |
+		SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK;
+	save->cs.limit = 0xffff;
+	/*
+	 * cs.base should really be 0xffff0000, but vmx can't handle that, so
+	 * be consistent with it.
+	 *
+	 * Replace when we have real mode working for vmx.
+	 */
+	save->cs.base = 0xf0000;
+
+	save->gdtr.limit = 0xffff;
+	save->idtr.limit = 0xffff;
+
+	init_sys_seg(&save->ldtr, SEG_TYPE_LDT);
+	init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16);
+
+	save->efer = MSR_EFER_SVME_MASK;
+	save->dr6 = 0xffff0ff0;
+	save->dr7 = 0x400;
+	save->rflags = 2;
+	save->rip = 0x0000fff0;
+
+	/*
+	 * cr0 val on cpu init should be 0x60000010, we enable cpu
+	 * cache by default. the orderly way is to enable cache in bios.
+	 */
+	save->cr0 = 0x00000010 | X86_CR0_PG | X86_CR0_WP;
+	save->cr4 = X86_CR4_PAE;
+	/* rdx = ?? */
+}
+
+static int svm_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	init_vmcb(svm->vmcb);
+
+	if (vcpu->vcpu_id != 0) {
+		svm->vmcb->save.rip = 0;
+		svm->vmcb->save.cs.base = svm->vcpu.arch.sipi_vector << 12;
+		svm->vmcb->save.cs.selector = svm->vcpu.arch.sipi_vector << 8;
+	}
+
+	return 0;
+}
+
+static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
+{
+	struct vcpu_svm *svm;
+	struct page *page;
+	int err;
+
+	svm = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
+	if (!svm) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = kvm_vcpu_init(&svm->vcpu, kvm, id);
+	if (err)
+		goto free_svm;
+
+	page = alloc_page(GFP_KERNEL);
+	if (!page) {
+		err = -ENOMEM;
+		goto uninit;
+	}
+
+	svm->vmcb = page_address(page);
+	clear_page(svm->vmcb);
+	svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT;
+	svm->asid_generation = 0;
+	memset(svm->db_regs, 0, sizeof(svm->db_regs));
+	init_vmcb(svm->vmcb);
+
+	fx_init(&svm->vcpu);
+	svm->vcpu.fpu_active = 1;
+	svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
+	if (svm->vcpu.vcpu_id == 0)
+		svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP;
+
+	return &svm->vcpu;
+
+uninit:
+	kvm_vcpu_uninit(&svm->vcpu);
+free_svm:
+	kmem_cache_free(kvm_vcpu_cache, svm);
+out:
+	return ERR_PTR(err);
+}
+
+static void svm_free_vcpu(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	__free_page(pfn_to_page(svm->vmcb_pa >> PAGE_SHIFT));
+	kvm_vcpu_uninit(vcpu);
+	kmem_cache_free(kvm_vcpu_cache, svm);
+}
+
+static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	int i;
+
+	if (unlikely(cpu != vcpu->cpu)) {
+		u64 tsc_this, delta;
+
+		/*
+		 * Make sure that the guest sees a monotonically
+		 * increasing TSC.
+		 */
+		rdtscll(tsc_this);
+		delta = vcpu->arch.host_tsc - tsc_this;
+		svm->vmcb->control.tsc_offset += delta;
+		vcpu->cpu = cpu;
+		kvm_migrate_apic_timer(vcpu);
+	}
+
+	for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
+		rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
+}
+
+static void svm_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	int i;
+
+	++vcpu->stat.host_state_reload;
+	for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
+		wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
+
+	rdtscll(vcpu->arch.host_tsc);
+}
+
+static void svm_vcpu_decache(struct kvm_vcpu *vcpu)
+{
+}
+
+static void svm_cache_regs(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
+	vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
+	vcpu->arch.rip = svm->vmcb->save.rip;
+}
+
+static void svm_decache_regs(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX];
+	svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP];
+	svm->vmcb->save.rip = vcpu->arch.rip;
+}
+
+static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu)
+{
+	return to_svm(vcpu)->vmcb->save.rflags;
+}
+
+static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
+{
+	to_svm(vcpu)->vmcb->save.rflags = rflags;
+}
+
+static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg)
+{
+	struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save;
+
+	switch (seg) {
+	case VCPU_SREG_CS: return &save->cs;
+	case VCPU_SREG_DS: return &save->ds;
+	case VCPU_SREG_ES: return &save->es;
+	case VCPU_SREG_FS: return &save->fs;
+	case VCPU_SREG_GS: return &save->gs;
+	case VCPU_SREG_SS: return &save->ss;
+	case VCPU_SREG_TR: return &save->tr;
+	case VCPU_SREG_LDTR: return &save->ldtr;
+	}
+	BUG();
+	return NULL;
+}
+
+static u64 svm_get_segment_base(struct kvm_vcpu *vcpu, int seg)
+{
+	struct vmcb_seg *s = svm_seg(vcpu, seg);
+
+	return s->base;
+}
+
+static void svm_get_segment(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg)
+{
+	struct vmcb_seg *s = svm_seg(vcpu, seg);
+
+	var->base = s->base;
+	var->limit = s->limit;
+	var->selector = s->selector;
+	var->type = s->attrib & SVM_SELECTOR_TYPE_MASK;
+	var->s = (s->attrib >> SVM_SELECTOR_S_SHIFT) & 1;
+	var->dpl = (s->attrib >> SVM_SELECTOR_DPL_SHIFT) & 3;
+	var->present = (s->attrib >> SVM_SELECTOR_P_SHIFT) & 1;
+	var->avl = (s->attrib >> SVM_SELECTOR_AVL_SHIFT) & 1;
+	var->l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1;
+	var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1;
+	var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1;
+	var->unusable = !var->present;
+}
+
+static void svm_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	dt->limit = svm->vmcb->save.idtr.limit;
+	dt->base = svm->vmcb->save.idtr.base;
+}
+
+static void svm_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->save.idtr.limit = dt->limit;
+	svm->vmcb->save.idtr.base = dt->base ;
+}
+
+static void svm_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	dt->limit = svm->vmcb->save.gdtr.limit;
+	dt->base = svm->vmcb->save.gdtr.base;
+}
+
+static void svm_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->save.gdtr.limit = dt->limit;
+	svm->vmcb->save.gdtr.base = dt->base ;
+}
+
+static void svm_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
+{
+}
+
+static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+#ifdef CONFIG_X86_64
+	if (vcpu->arch.shadow_efer & EFER_LME) {
+		if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
+			vcpu->arch.shadow_efer |= EFER_LMA;
+			svm->vmcb->save.efer |= EFER_LMA | EFER_LME;
+		}
+
+		if (is_paging(vcpu) && !(cr0 & X86_CR0_PG)) {
+			vcpu->arch.shadow_efer &= ~EFER_LMA;
+			svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME);
+		}
+	}
+#endif
+	if ((vcpu->arch.cr0 & X86_CR0_TS) && !(cr0 & X86_CR0_TS)) {
+		svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
+		vcpu->fpu_active = 1;
+	}
+
+	vcpu->arch.cr0 = cr0;
+	cr0 |= X86_CR0_PG | X86_CR0_WP;
+	cr0 &= ~(X86_CR0_CD | X86_CR0_NW);
+	svm->vmcb->save.cr0 = cr0;
+}
+
+static void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
+{
+       vcpu->arch.cr4 = cr4;
+       to_svm(vcpu)->vmcb->save.cr4 = cr4 | X86_CR4_PAE;
+}
+
+static void svm_set_segment(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb_seg *s = svm_seg(vcpu, seg);
+
+	s->base = var->base;
+	s->limit = var->limit;
+	s->selector = var->selector;
+	if (var->unusable)
+		s->attrib = 0;
+	else {
+		s->attrib = (var->type & SVM_SELECTOR_TYPE_MASK);
+		s->attrib |= (var->s & 1) << SVM_SELECTOR_S_SHIFT;
+		s->attrib |= (var->dpl & 3) << SVM_SELECTOR_DPL_SHIFT;
+		s->attrib |= (var->present & 1) << SVM_SELECTOR_P_SHIFT;
+		s->attrib |= (var->avl & 1) << SVM_SELECTOR_AVL_SHIFT;
+		s->attrib |= (var->l & 1) << SVM_SELECTOR_L_SHIFT;
+		s->attrib |= (var->db & 1) << SVM_SELECTOR_DB_SHIFT;
+		s->attrib |= (var->g & 1) << SVM_SELECTOR_G_SHIFT;
+	}
+	if (seg == VCPU_SREG_CS)
+		svm->vmcb->save.cpl
+			= (svm->vmcb->save.cs.attrib
+			   >> SVM_SELECTOR_DPL_SHIFT) & 3;
+
+}
+
+/* FIXME:
+
+	svm(vcpu)->vmcb->control.int_ctl &= ~V_TPR_MASK;
+	svm(vcpu)->vmcb->control.int_ctl |= (sregs->cr8 & V_TPR_MASK);
+
+*/
+
+static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+{
+	return -EOPNOTSUPP;
+}
+
+static int svm_get_irq(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	u32 exit_int_info = svm->vmcb->control.exit_int_info;
+
+	if (is_external_interrupt(exit_int_info))
+		return exit_int_info & SVM_EVTINJ_VEC_MASK;
+	return -1;
+}
+
+static void load_host_msrs(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_X86_64
+	wrmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base);
+#endif
+}
+
+static void save_host_msrs(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_X86_64
+	rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base);
+#endif
+}
+
+static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *svm_data)
+{
+	if (svm_data->next_asid > svm_data->max_asid) {
+		++svm_data->asid_generation;
+		svm_data->next_asid = 1;
+		svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ALL_ASID;
+	}
+
+	svm->vcpu.cpu = svm_data->cpu;
+	svm->asid_generation = svm_data->asid_generation;
+	svm->vmcb->control.asid = svm_data->next_asid++;
+}
+
+static unsigned long svm_get_dr(struct kvm_vcpu *vcpu, int dr)
+{
+	return to_svm(vcpu)->db_regs[dr];
+}
+
+static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value,
+		       int *exception)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	*exception = 0;
+
+	if (svm->vmcb->save.dr7 & DR7_GD_MASK) {
+		svm->vmcb->save.dr7 &= ~DR7_GD_MASK;
+		svm->vmcb->save.dr6 |= DR6_BD_MASK;
+		*exception = DB_VECTOR;
+		return;
+	}
+
+	switch (dr) {
+	case 0 ... 3:
+		svm->db_regs[dr] = value;
+		return;
+	case 4 ... 5:
+		if (vcpu->arch.cr4 & X86_CR4_DE) {
+			*exception = UD_VECTOR;
+			return;
+		}
+	case 7: {
+		if (value & ~((1ULL << 32) - 1)) {
+			*exception = GP_VECTOR;
+			return;
+		}
+		svm->vmcb->save.dr7 = value;
+		return;
+	}
+	default:
+		printk(KERN_DEBUG "%s: unexpected dr %u\n",
+		       __FUNCTION__, dr);
+		*exception = UD_VECTOR;
+		return;
+	}
+}
+
+static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	u32 exit_int_info = svm->vmcb->control.exit_int_info;
+	struct kvm *kvm = svm->vcpu.kvm;
+	u64 fault_address;
+	u32 error_code;
+
+	if (!irqchip_in_kernel(kvm) &&
+		is_external_interrupt(exit_int_info))
+		push_irq(&svm->vcpu, exit_int_info & SVM_EVTINJ_VEC_MASK);
+
+	fault_address  = svm->vmcb->control.exit_info_2;
+	error_code = svm->vmcb->control.exit_info_1;
+	return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code);
+}
+
+static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	int er;
+
+	er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0);
+	if (er != EMULATE_DONE)
+		kvm_queue_exception(&svm->vcpu, UD_VECTOR);
+	return 1;
+}
+
+static int nm_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
+	if (!(svm->vcpu.arch.cr0 & X86_CR0_TS))
+		svm->vmcb->save.cr0 &= ~X86_CR0_TS;
+	svm->vcpu.fpu_active = 1;
+
+	return 1;
+}
+
+static int shutdown_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	/*
+	 * VMCB is undefined after a SHUTDOWN intercept
+	 * so reinitialize it.
+	 */
+	clear_page(svm->vmcb);
+	init_vmcb(svm->vmcb);
+
+	kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
+	return 0;
+}
+
+static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */
+	int size, down, in, string, rep;
+	unsigned port;
+
+	++svm->vcpu.stat.io_exits;
+
+	svm->next_rip = svm->vmcb->control.exit_info_2;
+
+	string = (io_info & SVM_IOIO_STR_MASK) != 0;
+
+	if (string) {
+		if (emulate_instruction(&svm->vcpu,
+					kvm_run, 0, 0, 0) == EMULATE_DO_MMIO)
+			return 0;
+		return 1;
+	}
+
+	in = (io_info & SVM_IOIO_TYPE_MASK) != 0;
+	port = io_info >> 16;
+	size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT;
+	rep = (io_info & SVM_IOIO_REP_MASK) != 0;
+	down = (svm->vmcb->save.rflags & X86_EFLAGS_DF) != 0;
+
+	return kvm_emulate_pio(&svm->vcpu, kvm_run, in, size, port);
+}
+
+static int nop_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	return 1;
+}
+
+static int halt_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	svm->next_rip = svm->vmcb->save.rip + 1;
+	skip_emulated_instruction(&svm->vcpu);
+	return kvm_emulate_halt(&svm->vcpu);
+}
+
+static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	svm->next_rip = svm->vmcb->save.rip + 3;
+	skip_emulated_instruction(&svm->vcpu);
+	kvm_emulate_hypercall(&svm->vcpu);
+	return 1;
+}
+
+static int invalid_op_interception(struct vcpu_svm *svm,
+				   struct kvm_run *kvm_run)
+{
+	kvm_queue_exception(&svm->vcpu, UD_VECTOR);
+	return 1;
+}
+
+static int task_switch_interception(struct vcpu_svm *svm,
+				    struct kvm_run *kvm_run)
+{
+	pr_unimpl(&svm->vcpu, "%s: task switch is unsupported\n", __FUNCTION__);
+	kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
+	return 0;
+}
+
+static int cpuid_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	svm->next_rip = svm->vmcb->save.rip + 2;
+	kvm_emulate_cpuid(&svm->vcpu);
+	return 1;
+}
+
+static int emulate_on_interception(struct vcpu_svm *svm,
+				   struct kvm_run *kvm_run)
+{
+	if (emulate_instruction(&svm->vcpu, NULL, 0, 0, 0) != EMULATE_DONE)
+		pr_unimpl(&svm->vcpu, "%s: failed\n", __FUNCTION__);
+	return 1;
+}
+
+static int cr8_write_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	emulate_instruction(&svm->vcpu, NULL, 0, 0, 0);
+	if (irqchip_in_kernel(svm->vcpu.kvm))
+		return 1;
+	kvm_run->exit_reason = KVM_EXIT_SET_TPR;
+	return 0;
+}
+
+static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	switch (ecx) {
+	case MSR_IA32_TIME_STAMP_COUNTER: {
+		u64 tsc;
+
+		rdtscll(tsc);
+		*data = svm->vmcb->control.tsc_offset + tsc;
+		break;
+	}
+	case MSR_K6_STAR:
+		*data = svm->vmcb->save.star;
+		break;
+#ifdef CONFIG_X86_64
+	case MSR_LSTAR:
+		*data = svm->vmcb->save.lstar;
+		break;
+	case MSR_CSTAR:
+		*data = svm->vmcb->save.cstar;
+		break;
+	case MSR_KERNEL_GS_BASE:
+		*data = svm->vmcb->save.kernel_gs_base;
+		break;
+	case MSR_SYSCALL_MASK:
+		*data = svm->vmcb->save.sfmask;
+		break;
+#endif
+	case MSR_IA32_SYSENTER_CS:
+		*data = svm->vmcb->save.sysenter_cs;
+		break;
+	case MSR_IA32_SYSENTER_EIP:
+		*data = svm->vmcb->save.sysenter_eip;
+		break;
+	case MSR_IA32_SYSENTER_ESP:
+		*data = svm->vmcb->save.sysenter_esp;
+		break;
+	default:
+		return kvm_get_msr_common(vcpu, ecx, data);
+	}
+	return 0;
+}
+
+static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
+	u64 data;
+
+	if (svm_get_msr(&svm->vcpu, ecx, &data))
+		kvm_inject_gp(&svm->vcpu, 0);
+	else {
+		svm->vmcb->save.rax = data & 0xffffffff;
+		svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32;
+		svm->next_rip = svm->vmcb->save.rip + 2;
+		skip_emulated_instruction(&svm->vcpu);
+	}
+	return 1;
+}
+
+static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	switch (ecx) {
+	case MSR_IA32_TIME_STAMP_COUNTER: {
+		u64 tsc;
+
+		rdtscll(tsc);
+		svm->vmcb->control.tsc_offset = data - tsc;
+		break;
+	}
+	case MSR_K6_STAR:
+		svm->vmcb->save.star = data;
+		break;
+#ifdef CONFIG_X86_64
+	case MSR_LSTAR:
+		svm->vmcb->save.lstar = data;
+		break;
+	case MSR_CSTAR:
+		svm->vmcb->save.cstar = data;
+		break;
+	case MSR_KERNEL_GS_BASE:
+		svm->vmcb->save.kernel_gs_base = data;
+		break;
+	case MSR_SYSCALL_MASK:
+		svm->vmcb->save.sfmask = data;
+		break;
+#endif
+	case MSR_IA32_SYSENTER_CS:
+		svm->vmcb->save.sysenter_cs = data;
+		break;
+	case MSR_IA32_SYSENTER_EIP:
+		svm->vmcb->save.sysenter_eip = data;
+		break;
+	case MSR_IA32_SYSENTER_ESP:
+		svm->vmcb->save.sysenter_esp = data;
+		break;
+	case MSR_K7_EVNTSEL0:
+	case MSR_K7_EVNTSEL1:
+	case MSR_K7_EVNTSEL2:
+	case MSR_K7_EVNTSEL3:
+		/*
+		 * only support writing 0 to the performance counters for now
+		 * to make Windows happy. Should be replaced by a real
+		 * performance counter emulation later.
+		 */
+		if (data != 0)
+			goto unhandled;
+		break;
+	default:
+	unhandled:
+		return kvm_set_msr_common(vcpu, ecx, data);
+	}
+	return 0;
+}
+
+static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
+	u64 data = (svm->vmcb->save.rax & -1u)
+		| ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32);
+	svm->next_rip = svm->vmcb->save.rip + 2;
+	if (svm_set_msr(&svm->vcpu, ecx, data))
+		kvm_inject_gp(&svm->vcpu, 0);
+	else
+		skip_emulated_instruction(&svm->vcpu);
+	return 1;
+}
+
+static int msr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (svm->vmcb->control.exit_info_1)
+		return wrmsr_interception(svm, kvm_run);
+	else
+		return rdmsr_interception(svm, kvm_run);
+}
+
+static int interrupt_window_interception(struct vcpu_svm *svm,
+				   struct kvm_run *kvm_run)
+{
+	svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR);
+	svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
+	/*
+	 * If the user space waits to inject interrupts, exit as soon as
+	 * possible
+	 */
+	if (kvm_run->request_interrupt_window &&
+	    !svm->vcpu.arch.irq_summary) {
+		++svm->vcpu.stat.irq_window_exits;
+		kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
+		return 0;
+	}
+
+	return 1;
+}
+
+static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
+				      struct kvm_run *kvm_run) = {
+	[SVM_EXIT_READ_CR0]           		= emulate_on_interception,
+	[SVM_EXIT_READ_CR3]           		= emulate_on_interception,
+	[SVM_EXIT_READ_CR4]           		= emulate_on_interception,
+	[SVM_EXIT_READ_CR8]           		= emulate_on_interception,
+	/* for now: */
+	[SVM_EXIT_WRITE_CR0]          		= emulate_on_interception,
+	[SVM_EXIT_WRITE_CR3]          		= emulate_on_interception,
+	[SVM_EXIT_WRITE_CR4]          		= emulate_on_interception,
+	[SVM_EXIT_WRITE_CR8]          		= cr8_write_interception,
+	[SVM_EXIT_READ_DR0] 			= emulate_on_interception,
+	[SVM_EXIT_READ_DR1]			= emulate_on_interception,
+	[SVM_EXIT_READ_DR2]			= emulate_on_interception,
+	[SVM_EXIT_READ_DR3]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR0]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR1]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR2]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR3]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR5]			= emulate_on_interception,
+	[SVM_EXIT_WRITE_DR7]			= emulate_on_interception,
+	[SVM_EXIT_EXCP_BASE + UD_VECTOR]	= ud_interception,
+	[SVM_EXIT_EXCP_BASE + PF_VECTOR] 	= pf_interception,
+	[SVM_EXIT_EXCP_BASE + NM_VECTOR] 	= nm_interception,
+	[SVM_EXIT_INTR] 			= nop_on_interception,
+	[SVM_EXIT_NMI]				= nop_on_interception,
+	[SVM_EXIT_SMI]				= nop_on_interception,
+	[SVM_EXIT_INIT]				= nop_on_interception,
+	[SVM_EXIT_VINTR]			= interrupt_window_interception,
+	/* [SVM_EXIT_CR0_SEL_WRITE]		= emulate_on_interception, */
+	[SVM_EXIT_CPUID]			= cpuid_interception,
+	[SVM_EXIT_INVD]                         = emulate_on_interception,
+	[SVM_EXIT_HLT]				= halt_interception,
+	[SVM_EXIT_INVLPG]			= emulate_on_interception,
+	[SVM_EXIT_INVLPGA]			= invalid_op_interception,
+	[SVM_EXIT_IOIO] 		  	= io_interception,
+	[SVM_EXIT_MSR]				= msr_interception,
+	[SVM_EXIT_TASK_SWITCH]			= task_switch_interception,
+	[SVM_EXIT_SHUTDOWN]			= shutdown_interception,
+	[SVM_EXIT_VMRUN]			= invalid_op_interception,
+	[SVM_EXIT_VMMCALL]			= vmmcall_interception,
+	[SVM_EXIT_VMLOAD]			= invalid_op_interception,
+	[SVM_EXIT_VMSAVE]			= invalid_op_interception,
+	[SVM_EXIT_STGI]				= invalid_op_interception,
+	[SVM_EXIT_CLGI]				= invalid_op_interception,
+	[SVM_EXIT_SKINIT]			= invalid_op_interception,
+	[SVM_EXIT_WBINVD]                       = emulate_on_interception,
+	[SVM_EXIT_MONITOR]			= invalid_op_interception,
+	[SVM_EXIT_MWAIT]			= invalid_op_interception,
+};
+
+
+static int handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	u32 exit_code = svm->vmcb->control.exit_code;
+
+	kvm_reput_irq(svm);
+
+	if (svm->vmcb->control.exit_code == SVM_EXIT_ERR) {
+		kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+		kvm_run->fail_entry.hardware_entry_failure_reason
+			= svm->vmcb->control.exit_code;
+		return 0;
+	}
+
+	if (is_external_interrupt(svm->vmcb->control.exit_int_info) &&
+	    exit_code != SVM_EXIT_EXCP_BASE + PF_VECTOR)
+		printk(KERN_ERR "%s: unexpected exit_ini_info 0x%x "
+		       "exit_code 0x%x\n",
+		       __FUNCTION__, svm->vmcb->control.exit_int_info,
+		       exit_code);
+
+	if (exit_code >= ARRAY_SIZE(svm_exit_handlers)
+	    || !svm_exit_handlers[exit_code]) {
+		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
+		kvm_run->hw.hardware_exit_reason = exit_code;
+		return 0;
+	}
+
+	return svm_exit_handlers[exit_code](svm, kvm_run);
+}
+
+static void reload_tss(struct kvm_vcpu *vcpu)
+{
+	int cpu = raw_smp_processor_id();
+
+	struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu);
+	svm_data->tss_desc->type = 9; /* available 32/64-bit TSS */
+	load_TR_desc();
+}
+
+static void pre_svm_run(struct vcpu_svm *svm)
+{
+	int cpu = raw_smp_processor_id();
+
+	struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu);
+
+	svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
+	if (svm->vcpu.cpu != cpu ||
+	    svm->asid_generation != svm_data->asid_generation)
+		new_asid(svm, svm_data);
+}
+
+
+static inline void svm_inject_irq(struct vcpu_svm *svm, int irq)
+{
+	struct vmcb_control_area *control;
+
+	control = &svm->vmcb->control;
+	control->int_vector = irq;
+	control->int_ctl &= ~V_INTR_PRIO_MASK;
+	control->int_ctl |= V_IRQ_MASK |
+		((/*control->int_vector >> 4*/ 0xf) << V_INTR_PRIO_SHIFT);
+}
+
+static void svm_set_irq(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm_inject_irq(svm, irq);
+}
+
+static void svm_intr_assist(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb *vmcb = svm->vmcb;
+	int intr_vector = -1;
+
+	if ((vmcb->control.exit_int_info & SVM_EVTINJ_VALID) &&
+	    ((vmcb->control.exit_int_info & SVM_EVTINJ_TYPE_MASK) == 0)) {
+		intr_vector = vmcb->control.exit_int_info &
+			      SVM_EVTINJ_VEC_MASK;
+		vmcb->control.exit_int_info = 0;
+		svm_inject_irq(svm, intr_vector);
+		return;
+	}
+
+	if (vmcb->control.int_ctl & V_IRQ_MASK)
+		return;
+
+	if (!kvm_cpu_has_interrupt(vcpu))
+		return;
+
+	if (!(vmcb->save.rflags & X86_EFLAGS_IF) ||
+	    (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) ||
+	    (vmcb->control.event_inj & SVM_EVTINJ_VALID)) {
+		/* unable to deliver irq, set pending irq */
+		vmcb->control.intercept |= (1ULL << INTERCEPT_VINTR);
+		svm_inject_irq(svm, 0x0);
+		return;
+	}
+	/* Okay, we can deliver the interrupt: grab it and update PIC state. */
+	intr_vector = kvm_cpu_get_interrupt(vcpu);
+	svm_inject_irq(svm, intr_vector);
+	kvm_timer_intr_post(vcpu, intr_vector);
+}
+
+static void kvm_reput_irq(struct vcpu_svm *svm)
+{
+	struct vmcb_control_area *control = &svm->vmcb->control;
+
+	if ((control->int_ctl & V_IRQ_MASK)
+	    && !irqchip_in_kernel(svm->vcpu.kvm)) {
+		control->int_ctl &= ~V_IRQ_MASK;
+		push_irq(&svm->vcpu, control->int_vector);
+	}
+
+	svm->vcpu.arch.interrupt_window_open =
+		!(control->int_state & SVM_INTERRUPT_SHADOW_MASK);
+}
+
+static void svm_do_inject_vector(struct vcpu_svm *svm)
+{
+	struct kvm_vcpu *vcpu = &svm->vcpu;
+	int word_index = __ffs(vcpu->arch.irq_summary);
+	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
+	int irq = word_index * BITS_PER_LONG + bit_index;
+
+	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+	if (!vcpu->arch.irq_pending[word_index])
+		clear_bit(word_index, &vcpu->arch.irq_summary);
+	svm_inject_irq(svm, irq);
+}
+
+static void do_interrupt_requests(struct kvm_vcpu *vcpu,
+				       struct kvm_run *kvm_run)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb_control_area *control = &svm->vmcb->control;
+
+	svm->vcpu.arch.interrupt_window_open =
+		(!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) &&
+		 (svm->vmcb->save.rflags & X86_EFLAGS_IF));
+
+	if (svm->vcpu.arch.interrupt_window_open && svm->vcpu.arch.irq_summary)
+		/*
+		 * If interrupts enabled, and not blocked by sti or mov ss. Good.
+		 */
+		svm_do_inject_vector(svm);
+
+	/*
+	 * Interrupts blocked.  Wait for unblock.
+	 */
+	if (!svm->vcpu.arch.interrupt_window_open &&
+	    (svm->vcpu.arch.irq_summary || kvm_run->request_interrupt_window))
+		control->intercept |= 1ULL << INTERCEPT_VINTR;
+	 else
+		control->intercept &= ~(1ULL << INTERCEPT_VINTR);
+}
+
+static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
+{
+	return 0;
+}
+
+static void save_db_regs(unsigned long *db_regs)
+{
+	asm volatile ("mov %%dr0, %0" : "=r"(db_regs[0]));
+	asm volatile ("mov %%dr1, %0" : "=r"(db_regs[1]));
+	asm volatile ("mov %%dr2, %0" : "=r"(db_regs[2]));
+	asm volatile ("mov %%dr3, %0" : "=r"(db_regs[3]));
+}
+
+static void load_db_regs(unsigned long *db_regs)
+{
+	asm volatile ("mov %0, %%dr0" : : "r"(db_regs[0]));
+	asm volatile ("mov %0, %%dr1" : : "r"(db_regs[1]));
+	asm volatile ("mov %0, %%dr2" : : "r"(db_regs[2]));
+	asm volatile ("mov %0, %%dr3" : : "r"(db_regs[3]));
+}
+
+static void svm_flush_tlb(struct kvm_vcpu *vcpu)
+{
+	force_new_asid(vcpu);
+}
+
+static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu)
+{
+}
+
+static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+	u16 fs_selector;
+	u16 gs_selector;
+	u16 ldt_selector;
+
+	pre_svm_run(svm);
+
+	save_host_msrs(vcpu);
+	fs_selector = read_fs();
+	gs_selector = read_gs();
+	ldt_selector = read_ldt();
+	svm->host_cr2 = kvm_read_cr2();
+	svm->host_dr6 = read_dr6();
+	svm->host_dr7 = read_dr7();
+	svm->vmcb->save.cr2 = vcpu->arch.cr2;
+
+	if (svm->vmcb->save.dr7 & 0xff) {
+		write_dr7(0);
+		save_db_regs(svm->host_db_regs);
+		load_db_regs(svm->db_regs);
+	}
+
+	clgi();
+
+	local_irq_enable();
+
+	asm volatile (
+#ifdef CONFIG_X86_64
+		"push %%rbp; \n\t"
+#else
+		"push %%ebp; \n\t"
+#endif
+
+#ifdef CONFIG_X86_64
+		"mov %c[rbx](%[svm]), %%rbx \n\t"
+		"mov %c[rcx](%[svm]), %%rcx \n\t"
+		"mov %c[rdx](%[svm]), %%rdx \n\t"
+		"mov %c[rsi](%[svm]), %%rsi \n\t"
+		"mov %c[rdi](%[svm]), %%rdi \n\t"
+		"mov %c[rbp](%[svm]), %%rbp \n\t"
+		"mov %c[r8](%[svm]),  %%r8  \n\t"
+		"mov %c[r9](%[svm]),  %%r9  \n\t"
+		"mov %c[r10](%[svm]), %%r10 \n\t"
+		"mov %c[r11](%[svm]), %%r11 \n\t"
+		"mov %c[r12](%[svm]), %%r12 \n\t"
+		"mov %c[r13](%[svm]), %%r13 \n\t"
+		"mov %c[r14](%[svm]), %%r14 \n\t"
+		"mov %c[r15](%[svm]), %%r15 \n\t"
+#else
+		"mov %c[rbx](%[svm]), %%ebx \n\t"
+		"mov %c[rcx](%[svm]), %%ecx \n\t"
+		"mov %c[rdx](%[svm]), %%edx \n\t"
+		"mov %c[rsi](%[svm]), %%esi \n\t"
+		"mov %c[rdi](%[svm]), %%edi \n\t"
+		"mov %c[rbp](%[svm]), %%ebp \n\t"
+#endif
+
+#ifdef CONFIG_X86_64
+		/* Enter guest mode */
+		"push %%rax \n\t"
+		"mov %c[vmcb](%[svm]), %%rax \n\t"
+		SVM_VMLOAD "\n\t"
+		SVM_VMRUN "\n\t"
+		SVM_VMSAVE "\n\t"
+		"pop %%rax \n\t"
+#else
+		/* Enter guest mode */
+		"push %%eax \n\t"
+		"mov %c[vmcb](%[svm]), %%eax \n\t"
+		SVM_VMLOAD "\n\t"
+		SVM_VMRUN "\n\t"
+		SVM_VMSAVE "\n\t"
+		"pop %%eax \n\t"
+#endif
+
+		/* Save guest registers, load host registers */
+#ifdef CONFIG_X86_64
+		"mov %%rbx, %c[rbx](%[svm]) \n\t"
+		"mov %%rcx, %c[rcx](%[svm]) \n\t"
+		"mov %%rdx, %c[rdx](%[svm]) \n\t"
+		"mov %%rsi, %c[rsi](%[svm]) \n\t"
+		"mov %%rdi, %c[rdi](%[svm]) \n\t"
+		"mov %%rbp, %c[rbp](%[svm]) \n\t"
+		"mov %%r8,  %c[r8](%[svm]) \n\t"
+		"mov %%r9,  %c[r9](%[svm]) \n\t"
+		"mov %%r10, %c[r10](%[svm]) \n\t"
+		"mov %%r11, %c[r11](%[svm]) \n\t"
+		"mov %%r12, %c[r12](%[svm]) \n\t"
+		"mov %%r13, %c[r13](%[svm]) \n\t"
+		"mov %%r14, %c[r14](%[svm]) \n\t"
+		"mov %%r15, %c[r15](%[svm]) \n\t"
+
+		"pop  %%rbp; \n\t"
+#else
+		"mov %%ebx, %c[rbx](%[svm]) \n\t"
+		"mov %%ecx, %c[rcx](%[svm]) \n\t"
+		"mov %%edx, %c[rdx](%[svm]) \n\t"
+		"mov %%esi, %c[rsi](%[svm]) \n\t"
+		"mov %%edi, %c[rdi](%[svm]) \n\t"
+		"mov %%ebp, %c[rbp](%[svm]) \n\t"
+
+		"pop  %%ebp; \n\t"
+#endif
+		:
+		: [svm]"a"(svm),
+		  [vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)),
+		  [rbx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBX])),
+		  [rcx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RCX])),
+		  [rdx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDX])),
+		  [rsi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RSI])),
+		  [rdi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDI])),
+		  [rbp]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBP]))
+#ifdef CONFIG_X86_64
+		  , [r8]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R8])),
+		  [r9]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R9])),
+		  [r10]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R10])),
+		  [r11]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R11])),
+		  [r12]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R12])),
+		  [r13]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R13])),
+		  [r14]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R14])),
+		  [r15]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R15]))
+#endif
+		: "cc", "memory"
+#ifdef CONFIG_X86_64
+		, "rbx", "rcx", "rdx", "rsi", "rdi"
+		, "r8", "r9", "r10", "r11" , "r12", "r13", "r14", "r15"
+#else
+		, "ebx", "ecx", "edx" , "esi", "edi"
+#endif
+		);
+
+	if ((svm->vmcb->save.dr7 & 0xff))
+		load_db_regs(svm->host_db_regs);
+
+	vcpu->arch.cr2 = svm->vmcb->save.cr2;
+
+	write_dr6(svm->host_dr6);
+	write_dr7(svm->host_dr7);
+	kvm_write_cr2(svm->host_cr2);
+
+	load_fs(fs_selector);
+	load_gs(gs_selector);
+	load_ldt(ldt_selector);
+	load_host_msrs(vcpu);
+
+	reload_tss(vcpu);
+
+	local_irq_disable();
+
+	stgi();
+
+	svm->next_rip = 0;
+}
+
+static void svm_set_cr3(struct kvm_vcpu *vcpu, unsigned long root)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->save.cr3 = root;
+	force_new_asid(vcpu);
+
+	if (vcpu->fpu_active) {
+		svm->vmcb->control.intercept_exceptions |= (1 << NM_VECTOR);
+		svm->vmcb->save.cr0 |= X86_CR0_TS;
+		vcpu->fpu_active = 0;
+	}
+}
+
+static int is_disabled(void)
+{
+	u64 vm_cr;
+
+	rdmsrl(MSR_VM_CR, vm_cr);
+	if (vm_cr & (1 << SVM_VM_CR_SVM_DISABLE))
+		return 1;
+
+	return 0;
+}
+
+static void
+svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
+{
+	/*
+	 * Patch in the VMMCALL instruction:
+	 */
+	hypercall[0] = 0x0f;
+	hypercall[1] = 0x01;
+	hypercall[2] = 0xd9;
+}
+
+static void svm_check_processor_compat(void *rtn)
+{
+	*(int *)rtn = 0;
+}
+
+static struct kvm_x86_ops svm_x86_ops = {
+	.cpu_has_kvm_support = has_svm,
+	.disabled_by_bios = is_disabled,
+	.hardware_setup = svm_hardware_setup,
+	.hardware_unsetup = svm_hardware_unsetup,
+	.check_processor_compatibility = svm_check_processor_compat,
+	.hardware_enable = svm_hardware_enable,
+	.hardware_disable = svm_hardware_disable,
+
+	.vcpu_create = svm_create_vcpu,
+	.vcpu_free = svm_free_vcpu,
+	.vcpu_reset = svm_vcpu_reset,
+
+	.prepare_guest_switch = svm_prepare_guest_switch,
+	.vcpu_load = svm_vcpu_load,
+	.vcpu_put = svm_vcpu_put,
+	.vcpu_decache = svm_vcpu_decache,
+
+	.set_guest_debug = svm_guest_debug,
+	.get_msr = svm_get_msr,
+	.set_msr = svm_set_msr,
+	.get_segment_base = svm_get_segment_base,
+	.get_segment = svm_get_segment,
+	.set_segment = svm_set_segment,
+	.get_cs_db_l_bits = kvm_get_cs_db_l_bits,
+	.decache_cr4_guest_bits = svm_decache_cr4_guest_bits,
+	.set_cr0 = svm_set_cr0,
+	.set_cr3 = svm_set_cr3,
+	.set_cr4 = svm_set_cr4,
+	.set_efer = svm_set_efer,
+	.get_idt = svm_get_idt,
+	.set_idt = svm_set_idt,
+	.get_gdt = svm_get_gdt,
+	.set_gdt = svm_set_gdt,
+	.get_dr = svm_get_dr,
+	.set_dr = svm_set_dr,
+	.cache_regs = svm_cache_regs,
+	.decache_regs = svm_decache_regs,
+	.get_rflags = svm_get_rflags,
+	.set_rflags = svm_set_rflags,
+
+	.tlb_flush = svm_flush_tlb,
+
+	.run = svm_vcpu_run,
+	.handle_exit = handle_exit,
+	.skip_emulated_instruction = skip_emulated_instruction,
+	.patch_hypercall = svm_patch_hypercall,
+	.get_irq = svm_get_irq,
+	.set_irq = svm_set_irq,
+	.queue_exception = svm_queue_exception,
+	.exception_injected = svm_exception_injected,
+	.inject_pending_irq = svm_intr_assist,
+	.inject_pending_vectors = do_interrupt_requests,
+
+	.set_tss_addr = svm_set_tss_addr,
+};
+
+static int __init svm_init(void)
+{
+	return kvm_init(&svm_x86_ops, sizeof(struct vcpu_svm),
+			      THIS_MODULE);
+}
+
+static void __exit svm_exit(void)
+{
+	kvm_exit();
+}
+
+module_init(svm_init)
+module_exit(svm_exit)
diff --git a/arch/x86/kvm/svm.h b/arch/x86/kvm/svm.h
new file mode 100644
index 000000000000..5fd50491b555
--- /dev/null
+++ b/arch/x86/kvm/svm.h
@@ -0,0 +1,325 @@
+#ifndef __SVM_H
+#define __SVM_H
+
+enum {
+	INTERCEPT_INTR,
+	INTERCEPT_NMI,
+	INTERCEPT_SMI,
+	INTERCEPT_INIT,
+	INTERCEPT_VINTR,
+	INTERCEPT_SELECTIVE_CR0,
+	INTERCEPT_STORE_IDTR,
+	INTERCEPT_STORE_GDTR,
+	INTERCEPT_STORE_LDTR,
+	INTERCEPT_STORE_TR,
+	INTERCEPT_LOAD_IDTR,
+	INTERCEPT_LOAD_GDTR,
+	INTERCEPT_LOAD_LDTR,
+	INTERCEPT_LOAD_TR,
+	INTERCEPT_RDTSC,
+	INTERCEPT_RDPMC,
+	INTERCEPT_PUSHF,
+	INTERCEPT_POPF,
+	INTERCEPT_CPUID,
+	INTERCEPT_RSM,
+	INTERCEPT_IRET,
+	INTERCEPT_INTn,
+	INTERCEPT_INVD,
+	INTERCEPT_PAUSE,
+	INTERCEPT_HLT,
+	INTERCEPT_INVLPG,
+	INTERCEPT_INVLPGA,
+	INTERCEPT_IOIO_PROT,
+	INTERCEPT_MSR_PROT,
+	INTERCEPT_TASK_SWITCH,
+	INTERCEPT_FERR_FREEZE,
+	INTERCEPT_SHUTDOWN,
+	INTERCEPT_VMRUN,
+	INTERCEPT_VMMCALL,
+	INTERCEPT_VMLOAD,
+	INTERCEPT_VMSAVE,
+	INTERCEPT_STGI,
+	INTERCEPT_CLGI,
+	INTERCEPT_SKINIT,
+	INTERCEPT_RDTSCP,
+	INTERCEPT_ICEBP,
+	INTERCEPT_WBINVD,
+	INTERCEPT_MONITOR,
+	INTERCEPT_MWAIT,
+	INTERCEPT_MWAIT_COND,
+};
+
+
+struct __attribute__ ((__packed__)) vmcb_control_area {
+	u16 intercept_cr_read;
+	u16 intercept_cr_write;
+	u16 intercept_dr_read;
+	u16 intercept_dr_write;
+	u32 intercept_exceptions;
+	u64 intercept;
+	u8 reserved_1[44];
+	u64 iopm_base_pa;
+	u64 msrpm_base_pa;
+	u64 tsc_offset;
+	u32 asid;
+	u8 tlb_ctl;
+	u8 reserved_2[3];
+	u32 int_ctl;
+	u32 int_vector;
+	u32 int_state;
+	u8 reserved_3[4];
+	u32 exit_code;
+	u32 exit_code_hi;
+	u64 exit_info_1;
+	u64 exit_info_2;
+	u32 exit_int_info;
+	u32 exit_int_info_err;
+	u64 nested_ctl;
+	u8 reserved_4[16];
+	u32 event_inj;
+	u32 event_inj_err;
+	u64 nested_cr3;
+	u64 lbr_ctl;
+	u8 reserved_5[832];
+};
+
+
+#define TLB_CONTROL_DO_NOTHING 0
+#define TLB_CONTROL_FLUSH_ALL_ASID 1
+
+#define V_TPR_MASK 0x0f
+
+#define V_IRQ_SHIFT 8
+#define V_IRQ_MASK (1 << V_IRQ_SHIFT)
+
+#define V_INTR_PRIO_SHIFT 16
+#define V_INTR_PRIO_MASK (0x0f << V_INTR_PRIO_SHIFT)
+
+#define V_IGN_TPR_SHIFT 20
+#define V_IGN_TPR_MASK (1 << V_IGN_TPR_SHIFT)
+
+#define V_INTR_MASKING_SHIFT 24
+#define V_INTR_MASKING_MASK (1 << V_INTR_MASKING_SHIFT)
+
+#define SVM_INTERRUPT_SHADOW_MASK 1
+
+#define SVM_IOIO_STR_SHIFT 2
+#define SVM_IOIO_REP_SHIFT 3
+#define SVM_IOIO_SIZE_SHIFT 4
+#define SVM_IOIO_ASIZE_SHIFT 7
+
+#define SVM_IOIO_TYPE_MASK 1
+#define SVM_IOIO_STR_MASK (1 << SVM_IOIO_STR_SHIFT)
+#define SVM_IOIO_REP_MASK (1 << SVM_IOIO_REP_SHIFT)
+#define SVM_IOIO_SIZE_MASK (7 << SVM_IOIO_SIZE_SHIFT)
+#define SVM_IOIO_ASIZE_MASK (7 << SVM_IOIO_ASIZE_SHIFT)
+
+struct __attribute__ ((__packed__)) vmcb_seg {
+	u16 selector;
+	u16 attrib;
+	u32 limit;
+	u64 base;
+};
+
+struct __attribute__ ((__packed__)) vmcb_save_area {
+	struct vmcb_seg es;
+	struct vmcb_seg cs;
+	struct vmcb_seg ss;
+	struct vmcb_seg ds;
+	struct vmcb_seg fs;
+	struct vmcb_seg gs;
+	struct vmcb_seg gdtr;
+	struct vmcb_seg ldtr;
+	struct vmcb_seg idtr;
+	struct vmcb_seg tr;
+	u8 reserved_1[43];
+	u8 cpl;
+	u8 reserved_2[4];
+	u64 efer;
+	u8 reserved_3[112];
+	u64 cr4;
+	u64 cr3;
+	u64 cr0;
+	u64 dr7;
+	u64 dr6;
+	u64 rflags;
+	u64 rip;
+	u8 reserved_4[88];
+	u64 rsp;
+	u8 reserved_5[24];
+	u64 rax;
+	u64 star;
+	u64 lstar;
+	u64 cstar;
+	u64 sfmask;
+	u64 kernel_gs_base;
+	u64 sysenter_cs;
+	u64 sysenter_esp;
+	u64 sysenter_eip;
+	u64 cr2;
+	u8 reserved_6[32];
+	u64 g_pat;
+	u64 dbgctl;
+	u64 br_from;
+	u64 br_to;
+	u64 last_excp_from;
+	u64 last_excp_to;
+};
+
+struct __attribute__ ((__packed__)) vmcb {
+	struct vmcb_control_area control;
+	struct vmcb_save_area save;
+};
+
+#define SVM_CPUID_FEATURE_SHIFT 2
+#define SVM_CPUID_FUNC 0x8000000a
+
+#define MSR_EFER_SVME_MASK (1ULL << 12)
+#define MSR_VM_CR       0xc0010114
+#define MSR_VM_HSAVE_PA 0xc0010117ULL
+
+#define SVM_VM_CR_SVM_DISABLE 4
+
+#define SVM_SELECTOR_S_SHIFT 4
+#define SVM_SELECTOR_DPL_SHIFT 5
+#define SVM_SELECTOR_P_SHIFT 7
+#define SVM_SELECTOR_AVL_SHIFT 8
+#define SVM_SELECTOR_L_SHIFT 9
+#define SVM_SELECTOR_DB_SHIFT 10
+#define SVM_SELECTOR_G_SHIFT 11
+
+#define SVM_SELECTOR_TYPE_MASK (0xf)
+#define SVM_SELECTOR_S_MASK (1 << SVM_SELECTOR_S_SHIFT)
+#define SVM_SELECTOR_DPL_MASK (3 << SVM_SELECTOR_DPL_SHIFT)
+#define SVM_SELECTOR_P_MASK (1 << SVM_SELECTOR_P_SHIFT)
+#define SVM_SELECTOR_AVL_MASK (1 << SVM_SELECTOR_AVL_SHIFT)
+#define SVM_SELECTOR_L_MASK (1 << SVM_SELECTOR_L_SHIFT)
+#define SVM_SELECTOR_DB_MASK (1 << SVM_SELECTOR_DB_SHIFT)
+#define SVM_SELECTOR_G_MASK (1 << SVM_SELECTOR_G_SHIFT)
+
+#define SVM_SELECTOR_WRITE_MASK (1 << 1)
+#define SVM_SELECTOR_READ_MASK SVM_SELECTOR_WRITE_MASK
+#define SVM_SELECTOR_CODE_MASK (1 << 3)
+
+#define INTERCEPT_CR0_MASK 1
+#define INTERCEPT_CR3_MASK (1 << 3)
+#define INTERCEPT_CR4_MASK (1 << 4)
+#define INTERCEPT_CR8_MASK (1 << 8)
+
+#define INTERCEPT_DR0_MASK 1
+#define INTERCEPT_DR1_MASK (1 << 1)
+#define INTERCEPT_DR2_MASK (1 << 2)
+#define INTERCEPT_DR3_MASK (1 << 3)
+#define INTERCEPT_DR4_MASK (1 << 4)
+#define INTERCEPT_DR5_MASK (1 << 5)
+#define INTERCEPT_DR6_MASK (1 << 6)
+#define INTERCEPT_DR7_MASK (1 << 7)
+
+#define SVM_EVTINJ_VEC_MASK 0xff
+
+#define SVM_EVTINJ_TYPE_SHIFT 8
+#define SVM_EVTINJ_TYPE_MASK (7 << SVM_EVTINJ_TYPE_SHIFT)
+
+#define SVM_EVTINJ_TYPE_INTR (0 << SVM_EVTINJ_TYPE_SHIFT)
+#define SVM_EVTINJ_TYPE_NMI (2 << SVM_EVTINJ_TYPE_SHIFT)
+#define SVM_EVTINJ_TYPE_EXEPT (3 << SVM_EVTINJ_TYPE_SHIFT)
+#define SVM_EVTINJ_TYPE_SOFT (4 << SVM_EVTINJ_TYPE_SHIFT)
+
+#define SVM_EVTINJ_VALID (1 << 31)
+#define SVM_EVTINJ_VALID_ERR (1 << 11)
+
+#define SVM_EXITINTINFO_VEC_MASK SVM_EVTINJ_VEC_MASK
+
+#define	SVM_EXITINTINFO_TYPE_INTR SVM_EVTINJ_TYPE_INTR
+#define	SVM_EXITINTINFO_TYPE_NMI SVM_EVTINJ_TYPE_NMI
+#define	SVM_EXITINTINFO_TYPE_EXEPT SVM_EVTINJ_TYPE_EXEPT
+#define	SVM_EXITINTINFO_TYPE_SOFT SVM_EVTINJ_TYPE_SOFT
+
+#define SVM_EXITINTINFO_VALID SVM_EVTINJ_VALID
+#define SVM_EXITINTINFO_VALID_ERR SVM_EVTINJ_VALID_ERR
+
+#define	SVM_EXIT_READ_CR0 	0x000
+#define	SVM_EXIT_READ_CR3 	0x003
+#define	SVM_EXIT_READ_CR4 	0x004
+#define	SVM_EXIT_READ_CR8 	0x008
+#define	SVM_EXIT_WRITE_CR0 	0x010
+#define	SVM_EXIT_WRITE_CR3 	0x013
+#define	SVM_EXIT_WRITE_CR4 	0x014
+#define	SVM_EXIT_WRITE_CR8 	0x018
+#define	SVM_EXIT_READ_DR0 	0x020
+#define	SVM_EXIT_READ_DR1 	0x021
+#define	SVM_EXIT_READ_DR2 	0x022
+#define	SVM_EXIT_READ_DR3 	0x023
+#define	SVM_EXIT_READ_DR4 	0x024
+#define	SVM_EXIT_READ_DR5 	0x025
+#define	SVM_EXIT_READ_DR6 	0x026
+#define	SVM_EXIT_READ_DR7 	0x027
+#define	SVM_EXIT_WRITE_DR0 	0x030
+#define	SVM_EXIT_WRITE_DR1 	0x031
+#define	SVM_EXIT_WRITE_DR2 	0x032
+#define	SVM_EXIT_WRITE_DR3 	0x033
+#define	SVM_EXIT_WRITE_DR4 	0x034
+#define	SVM_EXIT_WRITE_DR5 	0x035
+#define	SVM_EXIT_WRITE_DR6 	0x036
+#define	SVM_EXIT_WRITE_DR7 	0x037
+#define SVM_EXIT_EXCP_BASE      0x040
+#define SVM_EXIT_INTR		0x060
+#define SVM_EXIT_NMI		0x061
+#define SVM_EXIT_SMI		0x062
+#define SVM_EXIT_INIT		0x063
+#define SVM_EXIT_VINTR		0x064
+#define SVM_EXIT_CR0_SEL_WRITE	0x065
+#define SVM_EXIT_IDTR_READ	0x066
+#define SVM_EXIT_GDTR_READ	0x067
+#define SVM_EXIT_LDTR_READ	0x068
+#define SVM_EXIT_TR_READ	0x069
+#define SVM_EXIT_IDTR_WRITE	0x06a
+#define SVM_EXIT_GDTR_WRITE	0x06b
+#define SVM_EXIT_LDTR_WRITE	0x06c
+#define SVM_EXIT_TR_WRITE	0x06d
+#define SVM_EXIT_RDTSC		0x06e
+#define SVM_EXIT_RDPMC		0x06f
+#define SVM_EXIT_PUSHF		0x070
+#define SVM_EXIT_POPF		0x071
+#define SVM_EXIT_CPUID		0x072
+#define SVM_EXIT_RSM		0x073
+#define SVM_EXIT_IRET		0x074
+#define SVM_EXIT_SWINT		0x075
+#define SVM_EXIT_INVD		0x076
+#define SVM_EXIT_PAUSE		0x077
+#define SVM_EXIT_HLT		0x078
+#define SVM_EXIT_INVLPG		0x079
+#define SVM_EXIT_INVLPGA	0x07a
+#define SVM_EXIT_IOIO		0x07b
+#define SVM_EXIT_MSR		0x07c
+#define SVM_EXIT_TASK_SWITCH	0x07d
+#define SVM_EXIT_FERR_FREEZE	0x07e
+#define SVM_EXIT_SHUTDOWN	0x07f
+#define SVM_EXIT_VMRUN		0x080
+#define SVM_EXIT_VMMCALL	0x081
+#define SVM_EXIT_VMLOAD		0x082
+#define SVM_EXIT_VMSAVE		0x083
+#define SVM_EXIT_STGI		0x084
+#define SVM_EXIT_CLGI		0x085
+#define SVM_EXIT_SKINIT		0x086
+#define SVM_EXIT_RDTSCP		0x087
+#define SVM_EXIT_ICEBP		0x088
+#define SVM_EXIT_WBINVD		0x089
+#define SVM_EXIT_MONITOR	0x08a
+#define SVM_EXIT_MWAIT		0x08b
+#define SVM_EXIT_MWAIT_COND	0x08c
+#define SVM_EXIT_NPF  		0x400
+
+#define SVM_EXIT_ERR		-1
+
+#define SVM_CR0_SELECTIVE_MASK (1 << 3 | 1) /* TS and MP */
+
+#define SVM_VMLOAD ".byte 0x0f, 0x01, 0xda"
+#define SVM_VMRUN  ".byte 0x0f, 0x01, 0xd8"
+#define SVM_VMSAVE ".byte 0x0f, 0x01, 0xdb"
+#define SVM_CLGI   ".byte 0x0f, 0x01, 0xdd"
+#define SVM_STGI   ".byte 0x0f, 0x01, 0xdc"
+#define SVM_INVLPGA ".byte 0x0f, 0x01, 0xdf"
+
+#endif
+
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
new file mode 100644
index 000000000000..fc494aff5d8b
--- /dev/null
+++ b/arch/x86/kvm/vmx.c
@@ -0,0 +1,2671 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * This module enables machines with Intel VT-x extensions to run virtual
+ * machines without emulation or binary translation.
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ *
+ * Authors:
+ *   Avi Kivity   <avi@qumranet.com>
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "irq.h"
+#include "vmx.h"
+#include "segment_descriptor.h"
+#include "mmu.h"
+
+#include <linux/kvm_host.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/sched.h>
+#include <linux/moduleparam.h>
+
+#include <asm/io.h>
+#include <asm/desc.h>
+
+MODULE_AUTHOR("Qumranet");
+MODULE_LICENSE("GPL");
+
+static int bypass_guest_pf = 1;
+module_param(bypass_guest_pf, bool, 0);
+
+struct vmcs {
+	u32 revision_id;
+	u32 abort;
+	char data[0];
+};
+
+struct vcpu_vmx {
+	struct kvm_vcpu       vcpu;
+	int                   launched;
+	u8                    fail;
+	u32                   idt_vectoring_info;
+	struct kvm_msr_entry *guest_msrs;
+	struct kvm_msr_entry *host_msrs;
+	int                   nmsrs;
+	int                   save_nmsrs;
+	int                   msr_offset_efer;
+#ifdef CONFIG_X86_64
+	int                   msr_offset_kernel_gs_base;
+#endif
+	struct vmcs          *vmcs;
+	struct {
+		int           loaded;
+		u16           fs_sel, gs_sel, ldt_sel;
+		int           gs_ldt_reload_needed;
+		int           fs_reload_needed;
+		int           guest_efer_loaded;
+	} host_state;
+	struct {
+		struct {
+			bool pending;
+			u8 vector;
+			unsigned rip;
+		} irq;
+	} rmode;
+};
+
+static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
+{
+	return container_of(vcpu, struct vcpu_vmx, vcpu);
+}
+
+static int init_rmode_tss(struct kvm *kvm);
+
+static DEFINE_PER_CPU(struct vmcs *, vmxarea);
+static DEFINE_PER_CPU(struct vmcs *, current_vmcs);
+
+static struct page *vmx_io_bitmap_a;
+static struct page *vmx_io_bitmap_b;
+
+static struct vmcs_config {
+	int size;
+	int order;
+	u32 revision_id;
+	u32 pin_based_exec_ctrl;
+	u32 cpu_based_exec_ctrl;
+	u32 cpu_based_2nd_exec_ctrl;
+	u32 vmexit_ctrl;
+	u32 vmentry_ctrl;
+} vmcs_config;
+
+#define VMX_SEGMENT_FIELD(seg)					\
+	[VCPU_SREG_##seg] = {                                   \
+		.selector = GUEST_##seg##_SELECTOR,		\
+		.base = GUEST_##seg##_BASE,		   	\
+		.limit = GUEST_##seg##_LIMIT,		   	\
+		.ar_bytes = GUEST_##seg##_AR_BYTES,	   	\
+	}
+
+static struct kvm_vmx_segment_field {
+	unsigned selector;
+	unsigned base;
+	unsigned limit;
+	unsigned ar_bytes;
+} kvm_vmx_segment_fields[] = {
+	VMX_SEGMENT_FIELD(CS),
+	VMX_SEGMENT_FIELD(DS),
+	VMX_SEGMENT_FIELD(ES),
+	VMX_SEGMENT_FIELD(FS),
+	VMX_SEGMENT_FIELD(GS),
+	VMX_SEGMENT_FIELD(SS),
+	VMX_SEGMENT_FIELD(TR),
+	VMX_SEGMENT_FIELD(LDTR),
+};
+
+/*
+ * Keep MSR_K6_STAR at the end, as setup_msrs() will try to optimize it
+ * away by decrementing the array size.
+ */
+static const u32 vmx_msr_index[] = {
+#ifdef CONFIG_X86_64
+	MSR_SYSCALL_MASK, MSR_LSTAR, MSR_CSTAR, MSR_KERNEL_GS_BASE,
+#endif
+	MSR_EFER, MSR_K6_STAR,
+};
+#define NR_VMX_MSR ARRAY_SIZE(vmx_msr_index)
+
+static void load_msrs(struct kvm_msr_entry *e, int n)
+{
+	int i;
+
+	for (i = 0; i < n; ++i)
+		wrmsrl(e[i].index, e[i].data);
+}
+
+static void save_msrs(struct kvm_msr_entry *e, int n)
+{
+	int i;
+
+	for (i = 0; i < n; ++i)
+		rdmsrl(e[i].index, e[i].data);
+}
+
+static inline int is_page_fault(u32 intr_info)
+{
+	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
+			     INTR_INFO_VALID_MASK)) ==
+		(INTR_TYPE_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK);
+}
+
+static inline int is_no_device(u32 intr_info)
+{
+	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
+			     INTR_INFO_VALID_MASK)) ==
+		(INTR_TYPE_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK);
+}
+
+static inline int is_invalid_opcode(u32 intr_info)
+{
+	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
+			     INTR_INFO_VALID_MASK)) ==
+		(INTR_TYPE_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK);
+}
+
+static inline int is_external_interrupt(u32 intr_info)
+{
+	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VALID_MASK))
+		== (INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK);
+}
+
+static inline int cpu_has_vmx_tpr_shadow(void)
+{
+	return (vmcs_config.cpu_based_exec_ctrl & CPU_BASED_TPR_SHADOW);
+}
+
+static inline int vm_need_tpr_shadow(struct kvm *kvm)
+{
+	return ((cpu_has_vmx_tpr_shadow()) && (irqchip_in_kernel(kvm)));
+}
+
+static inline int cpu_has_secondary_exec_ctrls(void)
+{
+	return (vmcs_config.cpu_based_exec_ctrl &
+		CPU_BASED_ACTIVATE_SECONDARY_CONTROLS);
+}
+
+static inline int cpu_has_vmx_virtualize_apic_accesses(void)
+{
+	return (vmcs_config.cpu_based_2nd_exec_ctrl &
+		SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
+}
+
+static inline int vm_need_virtualize_apic_accesses(struct kvm *kvm)
+{
+	return ((cpu_has_vmx_virtualize_apic_accesses()) &&
+		(irqchip_in_kernel(kvm)));
+}
+
+static int __find_msr_index(struct vcpu_vmx *vmx, u32 msr)
+{
+	int i;
+
+	for (i = 0; i < vmx->nmsrs; ++i)
+		if (vmx->guest_msrs[i].index == msr)
+			return i;
+	return -1;
+}
+
+static struct kvm_msr_entry *find_msr_entry(struct vcpu_vmx *vmx, u32 msr)
+{
+	int i;
+
+	i = __find_msr_index(vmx, msr);
+	if (i >= 0)
+		return &vmx->guest_msrs[i];
+	return NULL;
+}
+
+static void vmcs_clear(struct vmcs *vmcs)
+{
+	u64 phys_addr = __pa(vmcs);
+	u8 error;
+
+	asm volatile (ASM_VMX_VMCLEAR_RAX "; setna %0"
+		      : "=g"(error) : "a"(&phys_addr), "m"(phys_addr)
+		      : "cc", "memory");
+	if (error)
+		printk(KERN_ERR "kvm: vmclear fail: %p/%llx\n",
+		       vmcs, phys_addr);
+}
+
+static void __vcpu_clear(void *arg)
+{
+	struct vcpu_vmx *vmx = arg;
+	int cpu = raw_smp_processor_id();
+
+	if (vmx->vcpu.cpu == cpu)
+		vmcs_clear(vmx->vmcs);
+	if (per_cpu(current_vmcs, cpu) == vmx->vmcs)
+		per_cpu(current_vmcs, cpu) = NULL;
+	rdtscll(vmx->vcpu.arch.host_tsc);
+}
+
+static void vcpu_clear(struct vcpu_vmx *vmx)
+{
+	if (vmx->vcpu.cpu == -1)
+		return;
+	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 0, 1);
+	vmx->launched = 0;
+}
+
+static unsigned long vmcs_readl(unsigned long field)
+{
+	unsigned long value;
+
+	asm volatile (ASM_VMX_VMREAD_RDX_RAX
+		      : "=a"(value) : "d"(field) : "cc");
+	return value;
+}
+
+static u16 vmcs_read16(unsigned long field)
+{
+	return vmcs_readl(field);
+}
+
+static u32 vmcs_read32(unsigned long field)
+{
+	return vmcs_readl(field);
+}
+
+static u64 vmcs_read64(unsigned long field)
+{
+#ifdef CONFIG_X86_64
+	return vmcs_readl(field);
+#else
+	return vmcs_readl(field) | ((u64)vmcs_readl(field+1) << 32);
+#endif
+}
+
+static noinline void vmwrite_error(unsigned long field, unsigned long value)
+{
+	printk(KERN_ERR "vmwrite error: reg %lx value %lx (err %d)\n",
+	       field, value, vmcs_read32(VM_INSTRUCTION_ERROR));
+	dump_stack();
+}
+
+static void vmcs_writel(unsigned long field, unsigned long value)
+{
+	u8 error;
+
+	asm volatile (ASM_VMX_VMWRITE_RAX_RDX "; setna %0"
+		       : "=q"(error) : "a"(value), "d"(field) : "cc");
+	if (unlikely(error))
+		vmwrite_error(field, value);
+}
+
+static void vmcs_write16(unsigned long field, u16 value)
+{
+	vmcs_writel(field, value);
+}
+
+static void vmcs_write32(unsigned long field, u32 value)
+{
+	vmcs_writel(field, value);
+}
+
+static void vmcs_write64(unsigned long field, u64 value)
+{
+#ifdef CONFIG_X86_64
+	vmcs_writel(field, value);
+#else
+	vmcs_writel(field, value);
+	asm volatile ("");
+	vmcs_writel(field+1, value >> 32);
+#endif
+}
+
+static void vmcs_clear_bits(unsigned long field, u32 mask)
+{
+	vmcs_writel(field, vmcs_readl(field) & ~mask);
+}
+
+static void vmcs_set_bits(unsigned long field, u32 mask)
+{
+	vmcs_writel(field, vmcs_readl(field) | mask);
+}
+
+static void update_exception_bitmap(struct kvm_vcpu *vcpu)
+{
+	u32 eb;
+
+	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
+	if (!vcpu->fpu_active)
+		eb |= 1u << NM_VECTOR;
+	if (vcpu->guest_debug.enabled)
+		eb |= 1u << 1;
+	if (vcpu->arch.rmode.active)
+		eb = ~0;
+	vmcs_write32(EXCEPTION_BITMAP, eb);
+}
+
+static void reload_tss(void)
+{
+#ifndef CONFIG_X86_64
+
+	/*
+	 * VT restores TR but not its size.  Useless.
+	 */
+	struct descriptor_table gdt;
+	struct segment_descriptor *descs;
+
+	get_gdt(&gdt);
+	descs = (void *)gdt.base;
+	descs[GDT_ENTRY_TSS].type = 9; /* available TSS */
+	load_TR_desc();
+#endif
+}
+
+static void load_transition_efer(struct vcpu_vmx *vmx)
+{
+	int efer_offset = vmx->msr_offset_efer;
+	u64 host_efer = vmx->host_msrs[efer_offset].data;
+	u64 guest_efer = vmx->guest_msrs[efer_offset].data;
+	u64 ignore_bits;
+
+	if (efer_offset < 0)
+		return;
+	/*
+	 * NX is emulated; LMA and LME handled by hardware; SCE meaninless
+	 * outside long mode
+	 */
+	ignore_bits = EFER_NX | EFER_SCE;
+#ifdef CONFIG_X86_64
+	ignore_bits |= EFER_LMA | EFER_LME;
+	/* SCE is meaningful only in long mode on Intel */
+	if (guest_efer & EFER_LMA)
+		ignore_bits &= ~(u64)EFER_SCE;
+#endif
+	if ((guest_efer & ~ignore_bits) == (host_efer & ~ignore_bits))
+		return;
+
+	vmx->host_state.guest_efer_loaded = 1;
+	guest_efer &= ~ignore_bits;
+	guest_efer |= host_efer & ignore_bits;
+	wrmsrl(MSR_EFER, guest_efer);
+	vmx->vcpu.stat.efer_reload++;
+}
+
+static void reload_host_efer(struct vcpu_vmx *vmx)
+{
+	if (vmx->host_state.guest_efer_loaded) {
+		vmx->host_state.guest_efer_loaded = 0;
+		load_msrs(vmx->host_msrs + vmx->msr_offset_efer, 1);
+	}
+}
+
+static void vmx_save_host_state(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (vmx->host_state.loaded)
+		return;
+
+	vmx->host_state.loaded = 1;
+	/*
+	 * Set host fs and gs selectors.  Unfortunately, 22.2.3 does not
+	 * allow segment selectors with cpl > 0 or ti == 1.
+	 */
+	vmx->host_state.ldt_sel = read_ldt();
+	vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel;
+	vmx->host_state.fs_sel = read_fs();
+	if (!(vmx->host_state.fs_sel & 7)) {
+		vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel);
+		vmx->host_state.fs_reload_needed = 0;
+	} else {
+		vmcs_write16(HOST_FS_SELECTOR, 0);
+		vmx->host_state.fs_reload_needed = 1;
+	}
+	vmx->host_state.gs_sel = read_gs();
+	if (!(vmx->host_state.gs_sel & 7))
+		vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel);
+	else {
+		vmcs_write16(HOST_GS_SELECTOR, 0);
+		vmx->host_state.gs_ldt_reload_needed = 1;
+	}
+
+#ifdef CONFIG_X86_64
+	vmcs_writel(HOST_FS_BASE, read_msr(MSR_FS_BASE));
+	vmcs_writel(HOST_GS_BASE, read_msr(MSR_GS_BASE));
+#else
+	vmcs_writel(HOST_FS_BASE, segment_base(vmx->host_state.fs_sel));
+	vmcs_writel(HOST_GS_BASE, segment_base(vmx->host_state.gs_sel));
+#endif
+
+#ifdef CONFIG_X86_64
+	if (is_long_mode(&vmx->vcpu))
+		save_msrs(vmx->host_msrs +
+			  vmx->msr_offset_kernel_gs_base, 1);
+
+#endif
+	load_msrs(vmx->guest_msrs, vmx->save_nmsrs);
+	load_transition_efer(vmx);
+}
+
+static void vmx_load_host_state(struct vcpu_vmx *vmx)
+{
+	unsigned long flags;
+
+	if (!vmx->host_state.loaded)
+		return;
+
+	++vmx->vcpu.stat.host_state_reload;
+	vmx->host_state.loaded = 0;
+	if (vmx->host_state.fs_reload_needed)
+		load_fs(vmx->host_state.fs_sel);
+	if (vmx->host_state.gs_ldt_reload_needed) {
+		load_ldt(vmx->host_state.ldt_sel);
+		/*
+		 * If we have to reload gs, we must take care to
+		 * preserve our gs base.
+		 */
+		local_irq_save(flags);
+		load_gs(vmx->host_state.gs_sel);
+#ifdef CONFIG_X86_64
+		wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE));
+#endif
+		local_irq_restore(flags);
+	}
+	reload_tss();
+	save_msrs(vmx->guest_msrs, vmx->save_nmsrs);
+	load_msrs(vmx->host_msrs, vmx->save_nmsrs);
+	reload_host_efer(vmx);
+}
+
+/*
+ * Switches to specified vcpu, until a matching vcpu_put(), but assumes
+ * vcpu mutex is already taken.
+ */
+static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u64 phys_addr = __pa(vmx->vmcs);
+	u64 tsc_this, delta;
+
+	if (vcpu->cpu != cpu) {
+		vcpu_clear(vmx);
+		kvm_migrate_apic_timer(vcpu);
+	}
+
+	if (per_cpu(current_vmcs, cpu) != vmx->vmcs) {
+		u8 error;
+
+		per_cpu(current_vmcs, cpu) = vmx->vmcs;
+		asm volatile (ASM_VMX_VMPTRLD_RAX "; setna %0"
+			      : "=g"(error) : "a"(&phys_addr), "m"(phys_addr)
+			      : "cc");
+		if (error)
+			printk(KERN_ERR "kvm: vmptrld %p/%llx fail\n",
+			       vmx->vmcs, phys_addr);
+	}
+
+	if (vcpu->cpu != cpu) {
+		struct descriptor_table dt;
+		unsigned long sysenter_esp;
+
+		vcpu->cpu = cpu;
+		/*
+		 * Linux uses per-cpu TSS and GDT, so set these when switching
+		 * processors.
+		 */
+		vmcs_writel(HOST_TR_BASE, read_tr_base()); /* 22.2.4 */
+		get_gdt(&dt);
+		vmcs_writel(HOST_GDTR_BASE, dt.base);   /* 22.2.4 */
+
+		rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp);
+		vmcs_writel(HOST_IA32_SYSENTER_ESP, sysenter_esp); /* 22.2.3 */
+
+		/*
+		 * Make sure the time stamp counter is monotonous.
+		 */
+		rdtscll(tsc_this);
+		delta = vcpu->arch.host_tsc - tsc_this;
+		vmcs_write64(TSC_OFFSET, vmcs_read64(TSC_OFFSET) + delta);
+	}
+}
+
+static void vmx_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	vmx_load_host_state(to_vmx(vcpu));
+}
+
+static void vmx_fpu_activate(struct kvm_vcpu *vcpu)
+{
+	if (vcpu->fpu_active)
+		return;
+	vcpu->fpu_active = 1;
+	vmcs_clear_bits(GUEST_CR0, X86_CR0_TS);
+	if (vcpu->arch.cr0 & X86_CR0_TS)
+		vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
+	update_exception_bitmap(vcpu);
+}
+
+static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu->fpu_active)
+		return;
+	vcpu->fpu_active = 0;
+	vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
+	update_exception_bitmap(vcpu);
+}
+
+static void vmx_vcpu_decache(struct kvm_vcpu *vcpu)
+{
+	vcpu_clear(to_vmx(vcpu));
+}
+
+static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu)
+{
+	return vmcs_readl(GUEST_RFLAGS);
+}
+
+static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
+{
+	if (vcpu->arch.rmode.active)
+		rflags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
+	vmcs_writel(GUEST_RFLAGS, rflags);
+}
+
+static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
+{
+	unsigned long rip;
+	u32 interruptibility;
+
+	rip = vmcs_readl(GUEST_RIP);
+	rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
+	vmcs_writel(GUEST_RIP, rip);
+
+	/*
+	 * We emulated an instruction, so temporary interrupt blocking
+	 * should be removed, if set.
+	 */
+	interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
+	if (interruptibility & 3)
+		vmcs_write32(GUEST_INTERRUPTIBILITY_INFO,
+			     interruptibility & ~3);
+	vcpu->arch.interrupt_window_open = 1;
+}
+
+static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
+				bool has_error_code, u32 error_code)
+{
+	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
+		     nr | INTR_TYPE_EXCEPTION
+		     | (has_error_code ? INTR_INFO_DELIEVER_CODE_MASK : 0)
+		     | INTR_INFO_VALID_MASK);
+	if (has_error_code)
+		vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code);
+}
+
+static bool vmx_exception_injected(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	return !(vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK);
+}
+
+/*
+ * Swap MSR entry in host/guest MSR entry array.
+ */
+#ifdef CONFIG_X86_64
+static void move_msr_up(struct vcpu_vmx *vmx, int from, int to)
+{
+	struct kvm_msr_entry tmp;
+
+	tmp = vmx->guest_msrs[to];
+	vmx->guest_msrs[to] = vmx->guest_msrs[from];
+	vmx->guest_msrs[from] = tmp;
+	tmp = vmx->host_msrs[to];
+	vmx->host_msrs[to] = vmx->host_msrs[from];
+	vmx->host_msrs[from] = tmp;
+}
+#endif
+
+/*
+ * Set up the vmcs to automatically save and restore system
+ * msrs.  Don't touch the 64-bit msrs if the guest is in legacy
+ * mode, as fiddling with msrs is very expensive.
+ */
+static void setup_msrs(struct vcpu_vmx *vmx)
+{
+	int save_nmsrs;
+
+	save_nmsrs = 0;
+#ifdef CONFIG_X86_64
+	if (is_long_mode(&vmx->vcpu)) {
+		int index;
+
+		index = __find_msr_index(vmx, MSR_SYSCALL_MASK);
+		if (index >= 0)
+			move_msr_up(vmx, index, save_nmsrs++);
+		index = __find_msr_index(vmx, MSR_LSTAR);
+		if (index >= 0)
+			move_msr_up(vmx, index, save_nmsrs++);
+		index = __find_msr_index(vmx, MSR_CSTAR);
+		if (index >= 0)
+			move_msr_up(vmx, index, save_nmsrs++);
+		index = __find_msr_index(vmx, MSR_KERNEL_GS_BASE);
+		if (index >= 0)
+			move_msr_up(vmx, index, save_nmsrs++);
+		/*
+		 * MSR_K6_STAR is only needed on long mode guests, and only
+		 * if efer.sce is enabled.
+		 */
+		index = __find_msr_index(vmx, MSR_K6_STAR);
+		if ((index >= 0) && (vmx->vcpu.arch.shadow_efer & EFER_SCE))
+			move_msr_up(vmx, index, save_nmsrs++);
+	}
+#endif
+	vmx->save_nmsrs = save_nmsrs;
+
+#ifdef CONFIG_X86_64
+	vmx->msr_offset_kernel_gs_base =
+		__find_msr_index(vmx, MSR_KERNEL_GS_BASE);
+#endif
+	vmx->msr_offset_efer = __find_msr_index(vmx, MSR_EFER);
+}
+
+/*
+ * reads and returns guest's timestamp counter "register"
+ * guest_tsc = host_tsc + tsc_offset    -- 21.3
+ */
+static u64 guest_read_tsc(void)
+{
+	u64 host_tsc, tsc_offset;
+
+	rdtscll(host_tsc);
+	tsc_offset = vmcs_read64(TSC_OFFSET);
+	return host_tsc + tsc_offset;
+}
+
+/*
+ * writes 'guest_tsc' into guest's timestamp counter "register"
+ * guest_tsc = host_tsc + tsc_offset ==> tsc_offset = guest_tsc - host_tsc
+ */
+static void guest_write_tsc(u64 guest_tsc)
+{
+	u64 host_tsc;
+
+	rdtscll(host_tsc);
+	vmcs_write64(TSC_OFFSET, guest_tsc - host_tsc);
+}
+
+/*
+ * Reads an msr value (of 'msr_index') into 'pdata'.
+ * Returns 0 on success, non-0 otherwise.
+ * Assumes vcpu_load() was already called.
+ */
+static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata)
+{
+	u64 data;
+	struct kvm_msr_entry *msr;
+
+	if (!pdata) {
+		printk(KERN_ERR "BUG: get_msr called with NULL pdata\n");
+		return -EINVAL;
+	}
+
+	switch (msr_index) {
+#ifdef CONFIG_X86_64
+	case MSR_FS_BASE:
+		data = vmcs_readl(GUEST_FS_BASE);
+		break;
+	case MSR_GS_BASE:
+		data = vmcs_readl(GUEST_GS_BASE);
+		break;
+	case MSR_EFER:
+		return kvm_get_msr_common(vcpu, msr_index, pdata);
+#endif
+	case MSR_IA32_TIME_STAMP_COUNTER:
+		data = guest_read_tsc();
+		break;
+	case MSR_IA32_SYSENTER_CS:
+		data = vmcs_read32(GUEST_SYSENTER_CS);
+		break;
+	case MSR_IA32_SYSENTER_EIP:
+		data = vmcs_readl(GUEST_SYSENTER_EIP);
+		break;
+	case MSR_IA32_SYSENTER_ESP:
+		data = vmcs_readl(GUEST_SYSENTER_ESP);
+		break;
+	default:
+		msr = find_msr_entry(to_vmx(vcpu), msr_index);
+		if (msr) {
+			data = msr->data;
+			break;
+		}
+		return kvm_get_msr_common(vcpu, msr_index, pdata);
+	}
+
+	*pdata = data;
+	return 0;
+}
+
+/*
+ * Writes msr value into into the appropriate "register".
+ * Returns 0 on success, non-0 otherwise.
+ * Assumes vcpu_load() was already called.
+ */
+static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	struct kvm_msr_entry *msr;
+	int ret = 0;
+
+	switch (msr_index) {
+#ifdef CONFIG_X86_64
+	case MSR_EFER:
+		ret = kvm_set_msr_common(vcpu, msr_index, data);
+		if (vmx->host_state.loaded) {
+			reload_host_efer(vmx);
+			load_transition_efer(vmx);
+		}
+		break;
+	case MSR_FS_BASE:
+		vmcs_writel(GUEST_FS_BASE, data);
+		break;
+	case MSR_GS_BASE:
+		vmcs_writel(GUEST_GS_BASE, data);
+		break;
+#endif
+	case MSR_IA32_SYSENTER_CS:
+		vmcs_write32(GUEST_SYSENTER_CS, data);
+		break;
+	case MSR_IA32_SYSENTER_EIP:
+		vmcs_writel(GUEST_SYSENTER_EIP, data);
+		break;
+	case MSR_IA32_SYSENTER_ESP:
+		vmcs_writel(GUEST_SYSENTER_ESP, data);
+		break;
+	case MSR_IA32_TIME_STAMP_COUNTER:
+		guest_write_tsc(data);
+		break;
+	default:
+		msr = find_msr_entry(vmx, msr_index);
+		if (msr) {
+			msr->data = data;
+			if (vmx->host_state.loaded)
+				load_msrs(vmx->guest_msrs, vmx->save_nmsrs);
+			break;
+		}
+		ret = kvm_set_msr_common(vcpu, msr_index, data);
+	}
+
+	return ret;
+}
+
+/*
+ * Sync the rsp and rip registers into the vcpu structure.  This allows
+ * registers to be accessed by indexing vcpu->arch.regs.
+ */
+static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP);
+	vcpu->arch.rip = vmcs_readl(GUEST_RIP);
+}
+
+/*
+ * Syncs rsp and rip back into the vmcs.  Should be called after possible
+ * modification.
+ */
+static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu)
+{
+	vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]);
+	vmcs_writel(GUEST_RIP, vcpu->arch.rip);
+}
+
+static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+{
+	unsigned long dr7 = 0x400;
+	int old_singlestep;
+
+	old_singlestep = vcpu->guest_debug.singlestep;
+
+	vcpu->guest_debug.enabled = dbg->enabled;
+	if (vcpu->guest_debug.enabled) {
+		int i;
+
+		dr7 |= 0x200;  /* exact */
+		for (i = 0; i < 4; ++i) {
+			if (!dbg->breakpoints[i].enabled)
+				continue;
+			vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address;
+			dr7 |= 2 << (i*2);    /* global enable */
+			dr7 |= 0 << (i*4+16); /* execution breakpoint */
+		}
+
+		vcpu->guest_debug.singlestep = dbg->singlestep;
+	} else
+		vcpu->guest_debug.singlestep = 0;
+
+	if (old_singlestep && !vcpu->guest_debug.singlestep) {
+		unsigned long flags;
+
+		flags = vmcs_readl(GUEST_RFLAGS);
+		flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
+		vmcs_writel(GUEST_RFLAGS, flags);
+	}
+
+	update_exception_bitmap(vcpu);
+	vmcs_writel(GUEST_DR7, dr7);
+
+	return 0;
+}
+
+static int vmx_get_irq(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 idtv_info_field;
+
+	idtv_info_field = vmx->idt_vectoring_info;
+	if (idtv_info_field & INTR_INFO_VALID_MASK) {
+		if (is_external_interrupt(idtv_info_field))
+			return idtv_info_field & VECTORING_INFO_VECTOR_MASK;
+		else
+			printk(KERN_DEBUG "pending exception: not handled yet\n");
+	}
+	return -1;
+}
+
+static __init int cpu_has_kvm_support(void)
+{
+	unsigned long ecx = cpuid_ecx(1);
+	return test_bit(5, &ecx); /* CPUID.1:ECX.VMX[bit 5] -> VT */
+}
+
+static __init int vmx_disabled_by_bios(void)
+{
+	u64 msr;
+
+	rdmsrl(MSR_IA32_FEATURE_CONTROL, msr);
+	return (msr & (MSR_IA32_FEATURE_CONTROL_LOCKED |
+		       MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
+	    == MSR_IA32_FEATURE_CONTROL_LOCKED;
+	/* locked but not enabled */
+}
+
+static void hardware_enable(void *garbage)
+{
+	int cpu = raw_smp_processor_id();
+	u64 phys_addr = __pa(per_cpu(vmxarea, cpu));
+	u64 old;
+
+	rdmsrl(MSR_IA32_FEATURE_CONTROL, old);
+	if ((old & (MSR_IA32_FEATURE_CONTROL_LOCKED |
+		    MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
+	    != (MSR_IA32_FEATURE_CONTROL_LOCKED |
+		MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
+		/* enable and lock */
+		wrmsrl(MSR_IA32_FEATURE_CONTROL, old |
+		       MSR_IA32_FEATURE_CONTROL_LOCKED |
+		       MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED);
+	write_cr4(read_cr4() | X86_CR4_VMXE); /* FIXME: not cpu hotplug safe */
+	asm volatile (ASM_VMX_VMXON_RAX : : "a"(&phys_addr), "m"(phys_addr)
+		      : "memory", "cc");
+}
+
+static void hardware_disable(void *garbage)
+{
+	asm volatile (ASM_VMX_VMXOFF : : : "cc");
+}
+
+static __init int adjust_vmx_controls(u32 ctl_min, u32 ctl_opt,
+				      u32 msr, u32 *result)
+{
+	u32 vmx_msr_low, vmx_msr_high;
+	u32 ctl = ctl_min | ctl_opt;
+
+	rdmsr(msr, vmx_msr_low, vmx_msr_high);
+
+	ctl &= vmx_msr_high; /* bit == 0 in high word ==> must be zero */
+	ctl |= vmx_msr_low;  /* bit == 1 in low word  ==> must be one  */
+
+	/* Ensure minimum (required) set of control bits are supported. */
+	if (ctl_min & ~ctl)
+		return -EIO;
+
+	*result = ctl;
+	return 0;
+}
+
+static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf)
+{
+	u32 vmx_msr_low, vmx_msr_high;
+	u32 min, opt;
+	u32 _pin_based_exec_control = 0;
+	u32 _cpu_based_exec_control = 0;
+	u32 _cpu_based_2nd_exec_control = 0;
+	u32 _vmexit_control = 0;
+	u32 _vmentry_control = 0;
+
+	min = PIN_BASED_EXT_INTR_MASK | PIN_BASED_NMI_EXITING;
+	opt = 0;
+	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PINBASED_CTLS,
+				&_pin_based_exec_control) < 0)
+		return -EIO;
+
+	min = CPU_BASED_HLT_EXITING |
+#ifdef CONFIG_X86_64
+	      CPU_BASED_CR8_LOAD_EXITING |
+	      CPU_BASED_CR8_STORE_EXITING |
+#endif
+	      CPU_BASED_USE_IO_BITMAPS |
+	      CPU_BASED_MOV_DR_EXITING |
+	      CPU_BASED_USE_TSC_OFFSETING;
+	opt = CPU_BASED_TPR_SHADOW |
+	      CPU_BASED_ACTIVATE_SECONDARY_CONTROLS;
+	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS,
+				&_cpu_based_exec_control) < 0)
+		return -EIO;
+#ifdef CONFIG_X86_64
+	if ((_cpu_based_exec_control & CPU_BASED_TPR_SHADOW))
+		_cpu_based_exec_control &= ~CPU_BASED_CR8_LOAD_EXITING &
+					   ~CPU_BASED_CR8_STORE_EXITING;
+#endif
+	if (_cpu_based_exec_control & CPU_BASED_ACTIVATE_SECONDARY_CONTROLS) {
+		min = 0;
+		opt = SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
+			SECONDARY_EXEC_WBINVD_EXITING;
+		if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS2,
+					&_cpu_based_2nd_exec_control) < 0)
+			return -EIO;
+	}
+#ifndef CONFIG_X86_64
+	if (!(_cpu_based_2nd_exec_control &
+				SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES))
+		_cpu_based_exec_control &= ~CPU_BASED_TPR_SHADOW;
+#endif
+
+	min = 0;
+#ifdef CONFIG_X86_64
+	min |= VM_EXIT_HOST_ADDR_SPACE_SIZE;
+#endif
+	opt = 0;
+	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_EXIT_CTLS,
+				&_vmexit_control) < 0)
+		return -EIO;
+
+	min = opt = 0;
+	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_ENTRY_CTLS,
+				&_vmentry_control) < 0)
+		return -EIO;
+
+	rdmsr(MSR_IA32_VMX_BASIC, vmx_msr_low, vmx_msr_high);
+
+	/* IA-32 SDM Vol 3B: VMCS size is never greater than 4kB. */
+	if ((vmx_msr_high & 0x1fff) > PAGE_SIZE)
+		return -EIO;
+
+#ifdef CONFIG_X86_64
+	/* IA-32 SDM Vol 3B: 64-bit CPUs always have VMX_BASIC_MSR[48]==0. */
+	if (vmx_msr_high & (1u<<16))
+		return -EIO;
+#endif
+
+	/* Require Write-Back (WB) memory type for VMCS accesses. */
+	if (((vmx_msr_high >> 18) & 15) != 6)
+		return -EIO;
+
+	vmcs_conf->size = vmx_msr_high & 0x1fff;
+	vmcs_conf->order = get_order(vmcs_config.size);
+	vmcs_conf->revision_id = vmx_msr_low;
+
+	vmcs_conf->pin_based_exec_ctrl = _pin_based_exec_control;
+	vmcs_conf->cpu_based_exec_ctrl = _cpu_based_exec_control;
+	vmcs_conf->cpu_based_2nd_exec_ctrl = _cpu_based_2nd_exec_control;
+	vmcs_conf->vmexit_ctrl         = _vmexit_control;
+	vmcs_conf->vmentry_ctrl        = _vmentry_control;
+
+	return 0;
+}
+
+static struct vmcs *alloc_vmcs_cpu(int cpu)
+{
+	int node = cpu_to_node(cpu);
+	struct page *pages;
+	struct vmcs *vmcs;
+
+	pages = alloc_pages_node(node, GFP_KERNEL, vmcs_config.order);
+	if (!pages)
+		return NULL;
+	vmcs = page_address(pages);
+	memset(vmcs, 0, vmcs_config.size);
+	vmcs->revision_id = vmcs_config.revision_id; /* vmcs revision id */
+	return vmcs;
+}
+
+static struct vmcs *alloc_vmcs(void)
+{
+	return alloc_vmcs_cpu(raw_smp_processor_id());
+}
+
+static void free_vmcs(struct vmcs *vmcs)
+{
+	free_pages((unsigned long)vmcs, vmcs_config.order);
+}
+
+static void free_kvm_area(void)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu)
+		free_vmcs(per_cpu(vmxarea, cpu));
+}
+
+static __init int alloc_kvm_area(void)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		struct vmcs *vmcs;
+
+		vmcs = alloc_vmcs_cpu(cpu);
+		if (!vmcs) {
+			free_kvm_area();
+			return -ENOMEM;
+		}
+
+		per_cpu(vmxarea, cpu) = vmcs;
+	}
+	return 0;
+}
+
+static __init int hardware_setup(void)
+{
+	if (setup_vmcs_config(&vmcs_config) < 0)
+		return -EIO;
+	return alloc_kvm_area();
+}
+
+static __exit void hardware_unsetup(void)
+{
+	free_kvm_area();
+}
+
+static void fix_pmode_dataseg(int seg, struct kvm_save_segment *save)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+
+	if (vmcs_readl(sf->base) == save->base && (save->base & AR_S_MASK)) {
+		vmcs_write16(sf->selector, save->selector);
+		vmcs_writel(sf->base, save->base);
+		vmcs_write32(sf->limit, save->limit);
+		vmcs_write32(sf->ar_bytes, save->ar);
+	} else {
+		u32 dpl = (vmcs_read16(sf->selector) & SELECTOR_RPL_MASK)
+			<< AR_DPL_SHIFT;
+		vmcs_write32(sf->ar_bytes, 0x93 | dpl);
+	}
+}
+
+static void enter_pmode(struct kvm_vcpu *vcpu)
+{
+	unsigned long flags;
+
+	vcpu->arch.rmode.active = 0;
+
+	vmcs_writel(GUEST_TR_BASE, vcpu->arch.rmode.tr.base);
+	vmcs_write32(GUEST_TR_LIMIT, vcpu->arch.rmode.tr.limit);
+	vmcs_write32(GUEST_TR_AR_BYTES, vcpu->arch.rmode.tr.ar);
+
+	flags = vmcs_readl(GUEST_RFLAGS);
+	flags &= ~(X86_EFLAGS_IOPL | X86_EFLAGS_VM);
+	flags |= (vcpu->arch.rmode.save_iopl << IOPL_SHIFT);
+	vmcs_writel(GUEST_RFLAGS, flags);
+
+	vmcs_writel(GUEST_CR4, (vmcs_readl(GUEST_CR4) & ~X86_CR4_VME) |
+			(vmcs_readl(CR4_READ_SHADOW) & X86_CR4_VME));
+
+	update_exception_bitmap(vcpu);
+
+	fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
+	fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
+	fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
+	fix_pmode_dataseg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
+
+	vmcs_write16(GUEST_SS_SELECTOR, 0);
+	vmcs_write32(GUEST_SS_AR_BYTES, 0x93);
+
+	vmcs_write16(GUEST_CS_SELECTOR,
+		     vmcs_read16(GUEST_CS_SELECTOR) & ~SELECTOR_RPL_MASK);
+	vmcs_write32(GUEST_CS_AR_BYTES, 0x9b);
+}
+
+static gva_t rmode_tss_base(struct kvm *kvm)
+{
+	if (!kvm->arch.tss_addr) {
+		gfn_t base_gfn = kvm->memslots[0].base_gfn +
+				 kvm->memslots[0].npages - 3;
+		return base_gfn << PAGE_SHIFT;
+	}
+	return kvm->arch.tss_addr;
+}
+
+static void fix_rmode_seg(int seg, struct kvm_save_segment *save)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+
+	save->selector = vmcs_read16(sf->selector);
+	save->base = vmcs_readl(sf->base);
+	save->limit = vmcs_read32(sf->limit);
+	save->ar = vmcs_read32(sf->ar_bytes);
+	vmcs_write16(sf->selector, save->base >> 4);
+	vmcs_write32(sf->base, save->base & 0xfffff);
+	vmcs_write32(sf->limit, 0xffff);
+	vmcs_write32(sf->ar_bytes, 0xf3);
+}
+
+static void enter_rmode(struct kvm_vcpu *vcpu)
+{
+	unsigned long flags;
+
+	vcpu->arch.rmode.active = 1;
+
+	vcpu->arch.rmode.tr.base = vmcs_readl(GUEST_TR_BASE);
+	vmcs_writel(GUEST_TR_BASE, rmode_tss_base(vcpu->kvm));
+
+	vcpu->arch.rmode.tr.limit = vmcs_read32(GUEST_TR_LIMIT);
+	vmcs_write32(GUEST_TR_LIMIT, RMODE_TSS_SIZE - 1);
+
+	vcpu->arch.rmode.tr.ar = vmcs_read32(GUEST_TR_AR_BYTES);
+	vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
+
+	flags = vmcs_readl(GUEST_RFLAGS);
+	vcpu->arch.rmode.save_iopl
+		= (flags & X86_EFLAGS_IOPL) >> IOPL_SHIFT;
+
+	flags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
+
+	vmcs_writel(GUEST_RFLAGS, flags);
+	vmcs_writel(GUEST_CR4, vmcs_readl(GUEST_CR4) | X86_CR4_VME);
+	update_exception_bitmap(vcpu);
+
+	vmcs_write16(GUEST_SS_SELECTOR, vmcs_readl(GUEST_SS_BASE) >> 4);
+	vmcs_write32(GUEST_SS_LIMIT, 0xffff);
+	vmcs_write32(GUEST_SS_AR_BYTES, 0xf3);
+
+	vmcs_write32(GUEST_CS_AR_BYTES, 0xf3);
+	vmcs_write32(GUEST_CS_LIMIT, 0xffff);
+	if (vmcs_readl(GUEST_CS_BASE) == 0xffff0000)
+		vmcs_writel(GUEST_CS_BASE, 0xf0000);
+	vmcs_write16(GUEST_CS_SELECTOR, vmcs_readl(GUEST_CS_BASE) >> 4);
+
+	fix_rmode_seg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
+	fix_rmode_seg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
+	fix_rmode_seg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
+	fix_rmode_seg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
+
+	kvm_mmu_reset_context(vcpu);
+	init_rmode_tss(vcpu->kvm);
+}
+
+#ifdef CONFIG_X86_64
+
+static void enter_lmode(struct kvm_vcpu *vcpu)
+{
+	u32 guest_tr_ar;
+
+	guest_tr_ar = vmcs_read32(GUEST_TR_AR_BYTES);
+	if ((guest_tr_ar & AR_TYPE_MASK) != AR_TYPE_BUSY_64_TSS) {
+		printk(KERN_DEBUG "%s: tss fixup for long mode. \n",
+		       __FUNCTION__);
+		vmcs_write32(GUEST_TR_AR_BYTES,
+			     (guest_tr_ar & ~AR_TYPE_MASK)
+			     | AR_TYPE_BUSY_64_TSS);
+	}
+
+	vcpu->arch.shadow_efer |= EFER_LMA;
+
+	find_msr_entry(to_vmx(vcpu), MSR_EFER)->data |= EFER_LMA | EFER_LME;
+	vmcs_write32(VM_ENTRY_CONTROLS,
+		     vmcs_read32(VM_ENTRY_CONTROLS)
+		     | VM_ENTRY_IA32E_MODE);
+}
+
+static void exit_lmode(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.shadow_efer &= ~EFER_LMA;
+
+	vmcs_write32(VM_ENTRY_CONTROLS,
+		     vmcs_read32(VM_ENTRY_CONTROLS)
+		     & ~VM_ENTRY_IA32E_MODE);
+}
+
+#endif
+
+static void vmx_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.cr4 &= KVM_GUEST_CR4_MASK;
+	vcpu->arch.cr4 |= vmcs_readl(GUEST_CR4) & ~KVM_GUEST_CR4_MASK;
+}
+
+static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
+{
+	vmx_fpu_deactivate(vcpu);
+
+	if (vcpu->arch.rmode.active && (cr0 & X86_CR0_PE))
+		enter_pmode(vcpu);
+
+	if (!vcpu->arch.rmode.active && !(cr0 & X86_CR0_PE))
+		enter_rmode(vcpu);
+
+#ifdef CONFIG_X86_64
+	if (vcpu->arch.shadow_efer & EFER_LME) {
+		if (!is_paging(vcpu) && (cr0 & X86_CR0_PG))
+			enter_lmode(vcpu);
+		if (is_paging(vcpu) && !(cr0 & X86_CR0_PG))
+			exit_lmode(vcpu);
+	}
+#endif
+
+	vmcs_writel(CR0_READ_SHADOW, cr0);
+	vmcs_writel(GUEST_CR0,
+		    (cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON);
+	vcpu->arch.cr0 = cr0;
+
+	if (!(cr0 & X86_CR0_TS) || !(cr0 & X86_CR0_PE))
+		vmx_fpu_activate(vcpu);
+}
+
+static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
+{
+	vmcs_writel(GUEST_CR3, cr3);
+	if (vcpu->arch.cr0 & X86_CR0_PE)
+		vmx_fpu_deactivate(vcpu);
+}
+
+static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
+{
+	vmcs_writel(CR4_READ_SHADOW, cr4);
+	vmcs_writel(GUEST_CR4, cr4 | (vcpu->arch.rmode.active ?
+		    KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON));
+	vcpu->arch.cr4 = cr4;
+}
+
+#ifdef CONFIG_X86_64
+
+static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER);
+
+	vcpu->arch.shadow_efer = efer;
+	if (efer & EFER_LMA) {
+		vmcs_write32(VM_ENTRY_CONTROLS,
+				     vmcs_read32(VM_ENTRY_CONTROLS) |
+				     VM_ENTRY_IA32E_MODE);
+		msr->data = efer;
+
+	} else {
+		vmcs_write32(VM_ENTRY_CONTROLS,
+				     vmcs_read32(VM_ENTRY_CONTROLS) &
+				     ~VM_ENTRY_IA32E_MODE);
+
+		msr->data = efer & ~EFER_LME;
+	}
+	setup_msrs(vmx);
+}
+
+#endif
+
+static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+
+	return vmcs_readl(sf->base);
+}
+
+static void vmx_get_segment(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+	u32 ar;
+
+	var->base = vmcs_readl(sf->base);
+	var->limit = vmcs_read32(sf->limit);
+	var->selector = vmcs_read16(sf->selector);
+	ar = vmcs_read32(sf->ar_bytes);
+	if (ar & AR_UNUSABLE_MASK)
+		ar = 0;
+	var->type = ar & 15;
+	var->s = (ar >> 4) & 1;
+	var->dpl = (ar >> 5) & 3;
+	var->present = (ar >> 7) & 1;
+	var->avl = (ar >> 12) & 1;
+	var->l = (ar >> 13) & 1;
+	var->db = (ar >> 14) & 1;
+	var->g = (ar >> 15) & 1;
+	var->unusable = (ar >> 16) & 1;
+}
+
+static u32 vmx_segment_access_rights(struct kvm_segment *var)
+{
+	u32 ar;
+
+	if (var->unusable)
+		ar = 1 << 16;
+	else {
+		ar = var->type & 15;
+		ar |= (var->s & 1) << 4;
+		ar |= (var->dpl & 3) << 5;
+		ar |= (var->present & 1) << 7;
+		ar |= (var->avl & 1) << 12;
+		ar |= (var->l & 1) << 13;
+		ar |= (var->db & 1) << 14;
+		ar |= (var->g & 1) << 15;
+	}
+	if (ar == 0) /* a 0 value means unusable */
+		ar = AR_UNUSABLE_MASK;
+
+	return ar;
+}
+
+static void vmx_set_segment(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+	u32 ar;
+
+	if (vcpu->arch.rmode.active && seg == VCPU_SREG_TR) {
+		vcpu->arch.rmode.tr.selector = var->selector;
+		vcpu->arch.rmode.tr.base = var->base;
+		vcpu->arch.rmode.tr.limit = var->limit;
+		vcpu->arch.rmode.tr.ar = vmx_segment_access_rights(var);
+		return;
+	}
+	vmcs_writel(sf->base, var->base);
+	vmcs_write32(sf->limit, var->limit);
+	vmcs_write16(sf->selector, var->selector);
+	if (vcpu->arch.rmode.active && var->s) {
+		/*
+		 * Hack real-mode segments into vm86 compatibility.
+		 */
+		if (var->base == 0xffff0000 && var->selector == 0xf000)
+			vmcs_writel(sf->base, 0xf0000);
+		ar = 0xf3;
+	} else
+		ar = vmx_segment_access_rights(var);
+	vmcs_write32(sf->ar_bytes, ar);
+}
+
+static void vmx_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l)
+{
+	u32 ar = vmcs_read32(GUEST_CS_AR_BYTES);
+
+	*db = (ar >> 14) & 1;
+	*l = (ar >> 13) & 1;
+}
+
+static void vmx_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	dt->limit = vmcs_read32(GUEST_IDTR_LIMIT);
+	dt->base = vmcs_readl(GUEST_IDTR_BASE);
+}
+
+static void vmx_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	vmcs_write32(GUEST_IDTR_LIMIT, dt->limit);
+	vmcs_writel(GUEST_IDTR_BASE, dt->base);
+}
+
+static void vmx_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	dt->limit = vmcs_read32(GUEST_GDTR_LIMIT);
+	dt->base = vmcs_readl(GUEST_GDTR_BASE);
+}
+
+static void vmx_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
+{
+	vmcs_write32(GUEST_GDTR_LIMIT, dt->limit);
+	vmcs_writel(GUEST_GDTR_BASE, dt->base);
+}
+
+static int init_rmode_tss(struct kvm *kvm)
+{
+	gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT;
+	u16 data = 0;
+	int r;
+
+	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
+	if (r < 0)
+		return 0;
+	data = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE;
+	r = kvm_write_guest_page(kvm, fn++, &data, 0x66, sizeof(u16));
+	if (r < 0)
+		return 0;
+	r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE);
+	if (r < 0)
+		return 0;
+	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
+	if (r < 0)
+		return 0;
+	data = ~0;
+	r = kvm_write_guest_page(kvm, fn, &data, RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
+			sizeof(u8));
+	if (r < 0)
+		return 0;
+	return 1;
+}
+
+static void seg_setup(int seg)
+{
+	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
+
+	vmcs_write16(sf->selector, 0);
+	vmcs_writel(sf->base, 0);
+	vmcs_write32(sf->limit, 0xffff);
+	vmcs_write32(sf->ar_bytes, 0x93);
+}
+
+static int alloc_apic_access_page(struct kvm *kvm)
+{
+	struct kvm_userspace_memory_region kvm_userspace_mem;
+	int r = 0;
+
+	mutex_lock(&kvm->lock);
+	if (kvm->arch.apic_access_page)
+		goto out;
+	kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT;
+	kvm_userspace_mem.flags = 0;
+	kvm_userspace_mem.guest_phys_addr = 0xfee00000ULL;
+	kvm_userspace_mem.memory_size = PAGE_SIZE;
+	r = __kvm_set_memory_region(kvm, &kvm_userspace_mem, 0);
+	if (r)
+		goto out;
+	kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
+out:
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+
+/*
+ * Sets up the vmcs for emulated real mode.
+ */
+static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
+{
+	u32 host_sysenter_cs;
+	u32 junk;
+	unsigned long a;
+	struct descriptor_table dt;
+	int i;
+	unsigned long kvm_vmx_return;
+	u32 exec_control;
+
+	/* I/O */
+	vmcs_write64(IO_BITMAP_A, page_to_phys(vmx_io_bitmap_a));
+	vmcs_write64(IO_BITMAP_B, page_to_phys(vmx_io_bitmap_b));
+
+	vmcs_write64(VMCS_LINK_POINTER, -1ull); /* 22.3.1.5 */
+
+	/* Control */
+	vmcs_write32(PIN_BASED_VM_EXEC_CONTROL,
+		vmcs_config.pin_based_exec_ctrl);
+
+	exec_control = vmcs_config.cpu_based_exec_ctrl;
+	if (!vm_need_tpr_shadow(vmx->vcpu.kvm)) {
+		exec_control &= ~CPU_BASED_TPR_SHADOW;
+#ifdef CONFIG_X86_64
+		exec_control |= CPU_BASED_CR8_STORE_EXITING |
+				CPU_BASED_CR8_LOAD_EXITING;
+#endif
+	}
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, exec_control);
+
+	if (cpu_has_secondary_exec_ctrls()) {
+		exec_control = vmcs_config.cpu_based_2nd_exec_ctrl;
+		if (!vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
+			exec_control &=
+				~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES;
+		vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control);
+	}
+
+	vmcs_write32(PAGE_FAULT_ERROR_CODE_MASK, !!bypass_guest_pf);
+	vmcs_write32(PAGE_FAULT_ERROR_CODE_MATCH, !!bypass_guest_pf);
+	vmcs_write32(CR3_TARGET_COUNT, 0);           /* 22.2.1 */
+
+	vmcs_writel(HOST_CR0, read_cr0());  /* 22.2.3 */
+	vmcs_writel(HOST_CR4, read_cr4());  /* 22.2.3, 22.2.5 */
+	vmcs_writel(HOST_CR3, read_cr3());  /* 22.2.3  FIXME: shadow tables */
+
+	vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS);  /* 22.2.4 */
+	vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
+	vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
+	vmcs_write16(HOST_FS_SELECTOR, read_fs());    /* 22.2.4 */
+	vmcs_write16(HOST_GS_SELECTOR, read_gs());    /* 22.2.4 */
+	vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
+#ifdef CONFIG_X86_64
+	rdmsrl(MSR_FS_BASE, a);
+	vmcs_writel(HOST_FS_BASE, a); /* 22.2.4 */
+	rdmsrl(MSR_GS_BASE, a);
+	vmcs_writel(HOST_GS_BASE, a); /* 22.2.4 */
+#else
+	vmcs_writel(HOST_FS_BASE, 0); /* 22.2.4 */
+	vmcs_writel(HOST_GS_BASE, 0); /* 22.2.4 */
+#endif
+
+	vmcs_write16(HOST_TR_SELECTOR, GDT_ENTRY_TSS*8);  /* 22.2.4 */
+
+	get_idt(&dt);
+	vmcs_writel(HOST_IDTR_BASE, dt.base);   /* 22.2.4 */
+
+	asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return));
+	vmcs_writel(HOST_RIP, kvm_vmx_return); /* 22.2.5 */
+	vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0);
+	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0);
+	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, 0);
+
+	rdmsr(MSR_IA32_SYSENTER_CS, host_sysenter_cs, junk);
+	vmcs_write32(HOST_IA32_SYSENTER_CS, host_sysenter_cs);
+	rdmsrl(MSR_IA32_SYSENTER_ESP, a);
+	vmcs_writel(HOST_IA32_SYSENTER_ESP, a);   /* 22.2.3 */
+	rdmsrl(MSR_IA32_SYSENTER_EIP, a);
+	vmcs_writel(HOST_IA32_SYSENTER_EIP, a);   /* 22.2.3 */
+
+	for (i = 0; i < NR_VMX_MSR; ++i) {
+		u32 index = vmx_msr_index[i];
+		u32 data_low, data_high;
+		u64 data;
+		int j = vmx->nmsrs;
+
+		if (rdmsr_safe(index, &data_low, &data_high) < 0)
+			continue;
+		if (wrmsr_safe(index, data_low, data_high) < 0)
+			continue;
+		data = data_low | ((u64)data_high << 32);
+		vmx->host_msrs[j].index = index;
+		vmx->host_msrs[j].reserved = 0;
+		vmx->host_msrs[j].data = data;
+		vmx->guest_msrs[j] = vmx->host_msrs[j];
+		++vmx->nmsrs;
+	}
+
+	vmcs_write32(VM_EXIT_CONTROLS, vmcs_config.vmexit_ctrl);
+
+	/* 22.2.1, 20.8.1 */
+	vmcs_write32(VM_ENTRY_CONTROLS, vmcs_config.vmentry_ctrl);
+
+	vmcs_writel(CR0_GUEST_HOST_MASK, ~0UL);
+	vmcs_writel(CR4_GUEST_HOST_MASK, KVM_GUEST_CR4_MASK);
+
+	if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
+		if (alloc_apic_access_page(vmx->vcpu.kvm) != 0)
+			return -ENOMEM;
+
+	return 0;
+}
+
+static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u64 msr;
+	int ret;
+
+	if (!init_rmode_tss(vmx->vcpu.kvm)) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	vmx->vcpu.arch.rmode.active = 0;
+
+	vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
+	set_cr8(&vmx->vcpu, 0);
+	msr = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
+	if (vmx->vcpu.vcpu_id == 0)
+		msr |= MSR_IA32_APICBASE_BSP;
+	kvm_set_apic_base(&vmx->vcpu, msr);
+
+	fx_init(&vmx->vcpu);
+
+	/*
+	 * GUEST_CS_BASE should really be 0xffff0000, but VT vm86 mode
+	 * insists on having GUEST_CS_BASE == GUEST_CS_SELECTOR << 4.  Sigh.
+	 */
+	if (vmx->vcpu.vcpu_id == 0) {
+		vmcs_write16(GUEST_CS_SELECTOR, 0xf000);
+		vmcs_writel(GUEST_CS_BASE, 0x000f0000);
+	} else {
+		vmcs_write16(GUEST_CS_SELECTOR, vmx->vcpu.arch.sipi_vector << 8);
+		vmcs_writel(GUEST_CS_BASE, vmx->vcpu.arch.sipi_vector << 12);
+	}
+	vmcs_write32(GUEST_CS_LIMIT, 0xffff);
+	vmcs_write32(GUEST_CS_AR_BYTES, 0x9b);
+
+	seg_setup(VCPU_SREG_DS);
+	seg_setup(VCPU_SREG_ES);
+	seg_setup(VCPU_SREG_FS);
+	seg_setup(VCPU_SREG_GS);
+	seg_setup(VCPU_SREG_SS);
+
+	vmcs_write16(GUEST_TR_SELECTOR, 0);
+	vmcs_writel(GUEST_TR_BASE, 0);
+	vmcs_write32(GUEST_TR_LIMIT, 0xffff);
+	vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
+
+	vmcs_write16(GUEST_LDTR_SELECTOR, 0);
+	vmcs_writel(GUEST_LDTR_BASE, 0);
+	vmcs_write32(GUEST_LDTR_LIMIT, 0xffff);
+	vmcs_write32(GUEST_LDTR_AR_BYTES, 0x00082);
+
+	vmcs_write32(GUEST_SYSENTER_CS, 0);
+	vmcs_writel(GUEST_SYSENTER_ESP, 0);
+	vmcs_writel(GUEST_SYSENTER_EIP, 0);
+
+	vmcs_writel(GUEST_RFLAGS, 0x02);
+	if (vmx->vcpu.vcpu_id == 0)
+		vmcs_writel(GUEST_RIP, 0xfff0);
+	else
+		vmcs_writel(GUEST_RIP, 0);
+	vmcs_writel(GUEST_RSP, 0);
+
+	/* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */
+	vmcs_writel(GUEST_DR7, 0x400);
+
+	vmcs_writel(GUEST_GDTR_BASE, 0);
+	vmcs_write32(GUEST_GDTR_LIMIT, 0xffff);
+
+	vmcs_writel(GUEST_IDTR_BASE, 0);
+	vmcs_write32(GUEST_IDTR_LIMIT, 0xffff);
+
+	vmcs_write32(GUEST_ACTIVITY_STATE, 0);
+	vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0);
+	vmcs_write32(GUEST_PENDING_DBG_EXCEPTIONS, 0);
+
+	guest_write_tsc(0);
+
+	/* Special registers */
+	vmcs_write64(GUEST_IA32_DEBUGCTL, 0);
+
+	setup_msrs(vmx);
+
+	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, 0);  /* 22.2.1 */
+
+	if (cpu_has_vmx_tpr_shadow()) {
+		vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, 0);
+		if (vm_need_tpr_shadow(vmx->vcpu.kvm))
+			vmcs_write64(VIRTUAL_APIC_PAGE_ADDR,
+				page_to_phys(vmx->vcpu.arch.apic->regs_page));
+		vmcs_write32(TPR_THRESHOLD, 0);
+	}
+
+	if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
+		vmcs_write64(APIC_ACCESS_ADDR,
+			     page_to_phys(vmx->vcpu.kvm->arch.apic_access_page));
+
+	vmx->vcpu.arch.cr0 = 0x60000010;
+	vmx_set_cr0(&vmx->vcpu, vmx->vcpu.arch.cr0); /* enter rmode */
+	vmx_set_cr4(&vmx->vcpu, 0);
+#ifdef CONFIG_X86_64
+	vmx_set_efer(&vmx->vcpu, 0);
+#endif
+	vmx_fpu_activate(&vmx->vcpu);
+	update_exception_bitmap(&vmx->vcpu);
+
+	return 0;
+
+out:
+	return ret;
+}
+
+static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (vcpu->arch.rmode.active) {
+		vmx->rmode.irq.pending = true;
+		vmx->rmode.irq.vector = irq;
+		vmx->rmode.irq.rip = vmcs_readl(GUEST_RIP);
+		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
+			     irq | INTR_TYPE_SOFT_INTR | INTR_INFO_VALID_MASK);
+		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1);
+		vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip - 1);
+		return;
+	}
+	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
+			irq | INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK);
+}
+
+static void kvm_do_inject_irq(struct kvm_vcpu *vcpu)
+{
+	int word_index = __ffs(vcpu->arch.irq_summary);
+	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
+	int irq = word_index * BITS_PER_LONG + bit_index;
+
+	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+	if (!vcpu->arch.irq_pending[word_index])
+		clear_bit(word_index, &vcpu->arch.irq_summary);
+	vmx_inject_irq(vcpu, irq);
+}
+
+
+static void do_interrupt_requests(struct kvm_vcpu *vcpu,
+				       struct kvm_run *kvm_run)
+{
+	u32 cpu_based_vm_exec_control;
+
+	vcpu->arch.interrupt_window_open =
+		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
+		 (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
+
+	if (vcpu->arch.interrupt_window_open &&
+	    vcpu->arch.irq_summary &&
+	    !(vmcs_read32(VM_ENTRY_INTR_INFO_FIELD) & INTR_INFO_VALID_MASK))
+		/*
+		 * If interrupts enabled, and not blocked by sti or mov ss. Good.
+		 */
+		kvm_do_inject_irq(vcpu);
+
+	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+	if (!vcpu->arch.interrupt_window_open &&
+	    (vcpu->arch.irq_summary || kvm_run->request_interrupt_window))
+		/*
+		 * Interrupts blocked.  Wait for unblock.
+		 */
+		cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
+	else
+		cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING;
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+}
+
+static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
+{
+	int ret;
+	struct kvm_userspace_memory_region tss_mem = {
+		.slot = 8,
+		.guest_phys_addr = addr,
+		.memory_size = PAGE_SIZE * 3,
+		.flags = 0,
+	};
+
+	ret = kvm_set_memory_region(kvm, &tss_mem, 0);
+	if (ret)
+		return ret;
+	kvm->arch.tss_addr = addr;
+	return 0;
+}
+
+static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
+{
+	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
+
+	set_debugreg(dbg->bp[0], 0);
+	set_debugreg(dbg->bp[1], 1);
+	set_debugreg(dbg->bp[2], 2);
+	set_debugreg(dbg->bp[3], 3);
+
+	if (dbg->singlestep) {
+		unsigned long flags;
+
+		flags = vmcs_readl(GUEST_RFLAGS);
+		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
+		vmcs_writel(GUEST_RFLAGS, flags);
+	}
+}
+
+static int handle_rmode_exception(struct kvm_vcpu *vcpu,
+				  int vec, u32 err_code)
+{
+	if (!vcpu->arch.rmode.active)
+		return 0;
+
+	/*
+	 * Instruction with address size override prefix opcode 0x67
+	 * Cause the #SS fault with 0 error code in VM86 mode.
+	 */
+	if (((vec == GP_VECTOR) || (vec == SS_VECTOR)) && err_code == 0)
+		if (emulate_instruction(vcpu, NULL, 0, 0, 0) == EMULATE_DONE)
+			return 1;
+	return 0;
+}
+
+static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 intr_info, error_code;
+	unsigned long cr2, rip;
+	u32 vect_info;
+	enum emulation_result er;
+
+	vect_info = vmx->idt_vectoring_info;
+	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
+
+	if ((vect_info & VECTORING_INFO_VALID_MASK) &&
+						!is_page_fault(intr_info))
+		printk(KERN_ERR "%s: unexpected, vectoring info 0x%x "
+		       "intr info 0x%x\n", __FUNCTION__, vect_info, intr_info);
+
+	if (!irqchip_in_kernel(vcpu->kvm) && is_external_interrupt(vect_info)) {
+		int irq = vect_info & VECTORING_INFO_VECTOR_MASK;
+		set_bit(irq, vcpu->arch.irq_pending);
+		set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
+	}
+
+	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
+		return 1;  /* already handled by vmx_vcpu_run() */
+
+	if (is_no_device(intr_info)) {
+		vmx_fpu_activate(vcpu);
+		return 1;
+	}
+
+	if (is_invalid_opcode(intr_info)) {
+		er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
+		if (er != EMULATE_DONE)
+			kvm_queue_exception(vcpu, UD_VECTOR);
+		return 1;
+	}
+
+	error_code = 0;
+	rip = vmcs_readl(GUEST_RIP);
+	if (intr_info & INTR_INFO_DELIEVER_CODE_MASK)
+		error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE);
+	if (is_page_fault(intr_info)) {
+		cr2 = vmcs_readl(EXIT_QUALIFICATION);
+		return kvm_mmu_page_fault(vcpu, cr2, error_code);
+	}
+
+	if (vcpu->arch.rmode.active &&
+	    handle_rmode_exception(vcpu, intr_info & INTR_INFO_VECTOR_MASK,
+								error_code)) {
+		if (vcpu->arch.halt_request) {
+			vcpu->arch.halt_request = 0;
+			return kvm_emulate_halt(vcpu);
+		}
+		return 1;
+	}
+
+	if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) ==
+	    (INTR_TYPE_EXCEPTION | 1)) {
+		kvm_run->exit_reason = KVM_EXIT_DEBUG;
+		return 0;
+	}
+	kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
+	kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK;
+	kvm_run->ex.error_code = error_code;
+	return 0;
+}
+
+static int handle_external_interrupt(struct kvm_vcpu *vcpu,
+				     struct kvm_run *kvm_run)
+{
+	++vcpu->stat.irq_exits;
+	return 1;
+}
+
+static int handle_triple_fault(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
+	return 0;
+}
+
+static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	unsigned long exit_qualification;
+	int size, down, in, string, rep;
+	unsigned port;
+
+	++vcpu->stat.io_exits;
+	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
+	string = (exit_qualification & 16) != 0;
+
+	if (string) {
+		if (emulate_instruction(vcpu,
+					kvm_run, 0, 0, 0) == EMULATE_DO_MMIO)
+			return 0;
+		return 1;
+	}
+
+	size = (exit_qualification & 7) + 1;
+	in = (exit_qualification & 8) != 0;
+	down = (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_DF) != 0;
+	rep = (exit_qualification & 32) != 0;
+	port = exit_qualification >> 16;
+
+	return kvm_emulate_pio(vcpu, kvm_run, in, size, port);
+}
+
+static void
+vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
+{
+	/*
+	 * Patch in the VMCALL instruction:
+	 */
+	hypercall[0] = 0x0f;
+	hypercall[1] = 0x01;
+	hypercall[2] = 0xc1;
+}
+
+static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	unsigned long exit_qualification;
+	int cr;
+	int reg;
+
+	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
+	cr = exit_qualification & 15;
+	reg = (exit_qualification >> 8) & 15;
+	switch ((exit_qualification >> 4) & 3) {
+	case 0: /* mov to cr */
+		switch (cr) {
+		case 0:
+			vcpu_load_rsp_rip(vcpu);
+			set_cr0(vcpu, vcpu->arch.regs[reg]);
+			skip_emulated_instruction(vcpu);
+			return 1;
+		case 3:
+			vcpu_load_rsp_rip(vcpu);
+			set_cr3(vcpu, vcpu->arch.regs[reg]);
+			skip_emulated_instruction(vcpu);
+			return 1;
+		case 4:
+			vcpu_load_rsp_rip(vcpu);
+			set_cr4(vcpu, vcpu->arch.regs[reg]);
+			skip_emulated_instruction(vcpu);
+			return 1;
+		case 8:
+			vcpu_load_rsp_rip(vcpu);
+			set_cr8(vcpu, vcpu->arch.regs[reg]);
+			skip_emulated_instruction(vcpu);
+			if (irqchip_in_kernel(vcpu->kvm))
+				return 1;
+			kvm_run->exit_reason = KVM_EXIT_SET_TPR;
+			return 0;
+		};
+		break;
+	case 2: /* clts */
+		vcpu_load_rsp_rip(vcpu);
+		vmx_fpu_deactivate(vcpu);
+		vcpu->arch.cr0 &= ~X86_CR0_TS;
+		vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0);
+		vmx_fpu_activate(vcpu);
+		skip_emulated_instruction(vcpu);
+		return 1;
+	case 1: /*mov from cr*/
+		switch (cr) {
+		case 3:
+			vcpu_load_rsp_rip(vcpu);
+			vcpu->arch.regs[reg] = vcpu->arch.cr3;
+			vcpu_put_rsp_rip(vcpu);
+			skip_emulated_instruction(vcpu);
+			return 1;
+		case 8:
+			vcpu_load_rsp_rip(vcpu);
+			vcpu->arch.regs[reg] = get_cr8(vcpu);
+			vcpu_put_rsp_rip(vcpu);
+			skip_emulated_instruction(vcpu);
+			return 1;
+		}
+		break;
+	case 3: /* lmsw */
+		lmsw(vcpu, (exit_qualification >> LMSW_SOURCE_DATA_SHIFT) & 0x0f);
+
+		skip_emulated_instruction(vcpu);
+		return 1;
+	default:
+		break;
+	}
+	kvm_run->exit_reason = 0;
+	pr_unimpl(vcpu, "unhandled control register: op %d cr %d\n",
+	       (int)(exit_qualification >> 4) & 3, cr);
+	return 0;
+}
+
+static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	unsigned long exit_qualification;
+	unsigned long val;
+	int dr, reg;
+
+	/*
+	 * FIXME: this code assumes the host is debugging the guest.
+	 *        need to deal with guest debugging itself too.
+	 */
+	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
+	dr = exit_qualification & 7;
+	reg = (exit_qualification >> 8) & 15;
+	vcpu_load_rsp_rip(vcpu);
+	if (exit_qualification & 16) {
+		/* mov from dr */
+		switch (dr) {
+		case 6:
+			val = 0xffff0ff0;
+			break;
+		case 7:
+			val = 0x400;
+			break;
+		default:
+			val = 0;
+		}
+		vcpu->arch.regs[reg] = val;
+	} else {
+		/* mov to dr */
+	}
+	vcpu_put_rsp_rip(vcpu);
+	skip_emulated_instruction(vcpu);
+	return 1;
+}
+
+static int handle_cpuid(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	kvm_emulate_cpuid(vcpu);
+	return 1;
+}
+
+static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
+	u64 data;
+
+	if (vmx_get_msr(vcpu, ecx, &data)) {
+		kvm_inject_gp(vcpu, 0);
+		return 1;
+	}
+
+	/* FIXME: handling of bits 32:63 of rax, rdx */
+	vcpu->arch.regs[VCPU_REGS_RAX] = data & -1u;
+	vcpu->arch.regs[VCPU_REGS_RDX] = (data >> 32) & -1u;
+	skip_emulated_instruction(vcpu);
+	return 1;
+}
+
+static int handle_wrmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
+	u64 data = (vcpu->arch.regs[VCPU_REGS_RAX] & -1u)
+		| ((u64)(vcpu->arch.regs[VCPU_REGS_RDX] & -1u) << 32);
+
+	if (vmx_set_msr(vcpu, ecx, data) != 0) {
+		kvm_inject_gp(vcpu, 0);
+		return 1;
+	}
+
+	skip_emulated_instruction(vcpu);
+	return 1;
+}
+
+static int handle_tpr_below_threshold(struct kvm_vcpu *vcpu,
+				      struct kvm_run *kvm_run)
+{
+	return 1;
+}
+
+static int handle_interrupt_window(struct kvm_vcpu *vcpu,
+				   struct kvm_run *kvm_run)
+{
+	u32 cpu_based_vm_exec_control;
+
+	/* clear pending irq */
+	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+	cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING;
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+	/*
+	 * If the user space waits to inject interrupts, exit as soon as
+	 * possible
+	 */
+	if (kvm_run->request_interrupt_window &&
+	    !vcpu->arch.irq_summary) {
+		kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
+		++vcpu->stat.irq_window_exits;
+		return 0;
+	}
+	return 1;
+}
+
+static int handle_halt(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	skip_emulated_instruction(vcpu);
+	return kvm_emulate_halt(vcpu);
+}
+
+static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	skip_emulated_instruction(vcpu);
+	kvm_emulate_hypercall(vcpu);
+	return 1;
+}
+
+static int handle_wbinvd(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	skip_emulated_instruction(vcpu);
+	/* TODO: Add support for VT-d/pass-through device */
+	return 1;
+}
+
+static int handle_apic_access(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	u64 exit_qualification;
+	enum emulation_result er;
+	unsigned long offset;
+
+	exit_qualification = vmcs_read64(EXIT_QUALIFICATION);
+	offset = exit_qualification & 0xffful;
+
+	er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
+
+	if (er !=  EMULATE_DONE) {
+		printk(KERN_ERR
+		       "Fail to handle apic access vmexit! Offset is 0x%lx\n",
+		       offset);
+		return -ENOTSUPP;
+	}
+	return 1;
+}
+
+/*
+ * The exit handlers return 1 if the exit was handled fully and guest execution
+ * may resume.  Otherwise they set the kvm_run parameter to indicate what needs
+ * to be done to userspace and return 0.
+ */
+static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu,
+				      struct kvm_run *kvm_run) = {
+	[EXIT_REASON_EXCEPTION_NMI]           = handle_exception,
+	[EXIT_REASON_EXTERNAL_INTERRUPT]      = handle_external_interrupt,
+	[EXIT_REASON_TRIPLE_FAULT]            = handle_triple_fault,
+	[EXIT_REASON_IO_INSTRUCTION]          = handle_io,
+	[EXIT_REASON_CR_ACCESS]               = handle_cr,
+	[EXIT_REASON_DR_ACCESS]               = handle_dr,
+	[EXIT_REASON_CPUID]                   = handle_cpuid,
+	[EXIT_REASON_MSR_READ]                = handle_rdmsr,
+	[EXIT_REASON_MSR_WRITE]               = handle_wrmsr,
+	[EXIT_REASON_PENDING_INTERRUPT]       = handle_interrupt_window,
+	[EXIT_REASON_HLT]                     = handle_halt,
+	[EXIT_REASON_VMCALL]                  = handle_vmcall,
+	[EXIT_REASON_TPR_BELOW_THRESHOLD]     = handle_tpr_below_threshold,
+	[EXIT_REASON_APIC_ACCESS]             = handle_apic_access,
+	[EXIT_REASON_WBINVD]                  = handle_wbinvd,
+};
+
+static const int kvm_vmx_max_exit_handlers =
+	ARRAY_SIZE(kvm_vmx_exit_handlers);
+
+/*
+ * The guest has exited.  See if we can fix it or if we need userspace
+ * assistance.
+ */
+static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
+{
+	u32 exit_reason = vmcs_read32(VM_EXIT_REASON);
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 vectoring_info = vmx->idt_vectoring_info;
+
+	if (unlikely(vmx->fail)) {
+		kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+		kvm_run->fail_entry.hardware_entry_failure_reason
+			= vmcs_read32(VM_INSTRUCTION_ERROR);
+		return 0;
+	}
+
+	if ((vectoring_info & VECTORING_INFO_VALID_MASK) &&
+				exit_reason != EXIT_REASON_EXCEPTION_NMI)
+		printk(KERN_WARNING "%s: unexpected, valid vectoring info and "
+		       "exit reason is 0x%x\n", __FUNCTION__, exit_reason);
+	if (exit_reason < kvm_vmx_max_exit_handlers
+	    && kvm_vmx_exit_handlers[exit_reason])
+		return kvm_vmx_exit_handlers[exit_reason](vcpu, kvm_run);
+	else {
+		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
+		kvm_run->hw.hardware_exit_reason = exit_reason;
+	}
+	return 0;
+}
+
+static void vmx_flush_tlb(struct kvm_vcpu *vcpu)
+{
+}
+
+static void update_tpr_threshold(struct kvm_vcpu *vcpu)
+{
+	int max_irr, tpr;
+
+	if (!vm_need_tpr_shadow(vcpu->kvm))
+		return;
+
+	if (!kvm_lapic_enabled(vcpu) ||
+	    ((max_irr = kvm_lapic_find_highest_irr(vcpu)) == -1)) {
+		vmcs_write32(TPR_THRESHOLD, 0);
+		return;
+	}
+
+	tpr = (kvm_lapic_get_cr8(vcpu) & 0x0f) << 4;
+	vmcs_write32(TPR_THRESHOLD, (max_irr > tpr) ? tpr >> 4 : max_irr >> 4);
+}
+
+static void enable_irq_window(struct kvm_vcpu *vcpu)
+{
+	u32 cpu_based_vm_exec_control;
+
+	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+}
+
+static void vmx_intr_assist(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 idtv_info_field, intr_info_field;
+	int has_ext_irq, interrupt_window_open;
+	int vector;
+
+	update_tpr_threshold(vcpu);
+
+	has_ext_irq = kvm_cpu_has_interrupt(vcpu);
+	intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD);
+	idtv_info_field = vmx->idt_vectoring_info;
+	if (intr_info_field & INTR_INFO_VALID_MASK) {
+		if (idtv_info_field & INTR_INFO_VALID_MASK) {
+			/* TODO: fault when IDT_Vectoring */
+			if (printk_ratelimit())
+				printk(KERN_ERR "Fault when IDT_Vectoring\n");
+		}
+		if (has_ext_irq)
+			enable_irq_window(vcpu);
+		return;
+	}
+	if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) {
+		if ((idtv_info_field & VECTORING_INFO_TYPE_MASK)
+		    == INTR_TYPE_EXT_INTR
+		    && vcpu->arch.rmode.active) {
+			u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK;
+
+			vmx_inject_irq(vcpu, vect);
+			if (unlikely(has_ext_irq))
+				enable_irq_window(vcpu);
+			return;
+		}
+
+		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field);
+		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN,
+				vmcs_read32(VM_EXIT_INSTRUCTION_LEN));
+
+		if (unlikely(idtv_info_field & INTR_INFO_DELIEVER_CODE_MASK))
+			vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE,
+				vmcs_read32(IDT_VECTORING_ERROR_CODE));
+		if (unlikely(has_ext_irq))
+			enable_irq_window(vcpu);
+		return;
+	}
+	if (!has_ext_irq)
+		return;
+	interrupt_window_open =
+		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
+		 (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
+	if (interrupt_window_open) {
+		vector = kvm_cpu_get_interrupt(vcpu);
+		vmx_inject_irq(vcpu, vector);
+		kvm_timer_intr_post(vcpu, vector);
+	} else
+		enable_irq_window(vcpu);
+}
+
+/*
+ * Failure to inject an interrupt should give us the information
+ * in IDT_VECTORING_INFO_FIELD.  However, if the failure occurs
+ * when fetching the interrupt redirection bitmap in the real-mode
+ * tss, this doesn't happen.  So we do it ourselves.
+ */
+static void fixup_rmode_irq(struct vcpu_vmx *vmx)
+{
+	vmx->rmode.irq.pending = 0;
+	if (vmcs_readl(GUEST_RIP) + 1 != vmx->rmode.irq.rip)
+		return;
+	vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip);
+	if (vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK) {
+		vmx->idt_vectoring_info &= ~VECTORING_INFO_TYPE_MASK;
+		vmx->idt_vectoring_info |= INTR_TYPE_EXT_INTR;
+		return;
+	}
+	vmx->idt_vectoring_info =
+		VECTORING_INFO_VALID_MASK
+		| INTR_TYPE_EXT_INTR
+		| vmx->rmode.irq.vector;
+}
+
+static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 intr_info;
+
+	/*
+	 * Loading guest fpu may have cleared host cr0.ts
+	 */
+	vmcs_writel(HOST_CR0, read_cr0());
+
+	asm(
+		/* Store host registers */
+#ifdef CONFIG_X86_64
+		"push %%rdx; push %%rbp;"
+		"push %%rcx \n\t"
+#else
+		"push %%edx; push %%ebp;"
+		"push %%ecx \n\t"
+#endif
+		ASM_VMX_VMWRITE_RSP_RDX "\n\t"
+		/* Check if vmlaunch of vmresume is needed */
+		"cmpl $0, %c[launched](%0) \n\t"
+		/* Load guest registers.  Don't clobber flags. */
+#ifdef CONFIG_X86_64
+		"mov %c[cr2](%0), %%rax \n\t"
+		"mov %%rax, %%cr2 \n\t"
+		"mov %c[rax](%0), %%rax \n\t"
+		"mov %c[rbx](%0), %%rbx \n\t"
+		"mov %c[rdx](%0), %%rdx \n\t"
+		"mov %c[rsi](%0), %%rsi \n\t"
+		"mov %c[rdi](%0), %%rdi \n\t"
+		"mov %c[rbp](%0), %%rbp \n\t"
+		"mov %c[r8](%0),  %%r8  \n\t"
+		"mov %c[r9](%0),  %%r9  \n\t"
+		"mov %c[r10](%0), %%r10 \n\t"
+		"mov %c[r11](%0), %%r11 \n\t"
+		"mov %c[r12](%0), %%r12 \n\t"
+		"mov %c[r13](%0), %%r13 \n\t"
+		"mov %c[r14](%0), %%r14 \n\t"
+		"mov %c[r15](%0), %%r15 \n\t"
+		"mov %c[rcx](%0), %%rcx \n\t" /* kills %0 (rcx) */
+#else
+		"mov %c[cr2](%0), %%eax \n\t"
+		"mov %%eax,   %%cr2 \n\t"
+		"mov %c[rax](%0), %%eax \n\t"
+		"mov %c[rbx](%0), %%ebx \n\t"
+		"mov %c[rdx](%0), %%edx \n\t"
+		"mov %c[rsi](%0), %%esi \n\t"
+		"mov %c[rdi](%0), %%edi \n\t"
+		"mov %c[rbp](%0), %%ebp \n\t"
+		"mov %c[rcx](%0), %%ecx \n\t" /* kills %0 (ecx) */
+#endif
+		/* Enter guest mode */
+		"jne .Llaunched \n\t"
+		ASM_VMX_VMLAUNCH "\n\t"
+		"jmp .Lkvm_vmx_return \n\t"
+		".Llaunched: " ASM_VMX_VMRESUME "\n\t"
+		".Lkvm_vmx_return: "
+		/* Save guest registers, load host registers, keep flags */
+#ifdef CONFIG_X86_64
+		"xchg %0,     (%%rsp) \n\t"
+		"mov %%rax, %c[rax](%0) \n\t"
+		"mov %%rbx, %c[rbx](%0) \n\t"
+		"pushq (%%rsp); popq %c[rcx](%0) \n\t"
+		"mov %%rdx, %c[rdx](%0) \n\t"
+		"mov %%rsi, %c[rsi](%0) \n\t"
+		"mov %%rdi, %c[rdi](%0) \n\t"
+		"mov %%rbp, %c[rbp](%0) \n\t"
+		"mov %%r8,  %c[r8](%0) \n\t"
+		"mov %%r9,  %c[r9](%0) \n\t"
+		"mov %%r10, %c[r10](%0) \n\t"
+		"mov %%r11, %c[r11](%0) \n\t"
+		"mov %%r12, %c[r12](%0) \n\t"
+		"mov %%r13, %c[r13](%0) \n\t"
+		"mov %%r14, %c[r14](%0) \n\t"
+		"mov %%r15, %c[r15](%0) \n\t"
+		"mov %%cr2, %%rax   \n\t"
+		"mov %%rax, %c[cr2](%0) \n\t"
+
+		"pop  %%rbp; pop  %%rbp; pop  %%rdx \n\t"
+#else
+		"xchg %0, (%%esp) \n\t"
+		"mov %%eax, %c[rax](%0) \n\t"
+		"mov %%ebx, %c[rbx](%0) \n\t"
+		"pushl (%%esp); popl %c[rcx](%0) \n\t"
+		"mov %%edx, %c[rdx](%0) \n\t"
+		"mov %%esi, %c[rsi](%0) \n\t"
+		"mov %%edi, %c[rdi](%0) \n\t"
+		"mov %%ebp, %c[rbp](%0) \n\t"
+		"mov %%cr2, %%eax  \n\t"
+		"mov %%eax, %c[cr2](%0) \n\t"
+
+		"pop %%ebp; pop %%ebp; pop %%edx \n\t"
+#endif
+		"setbe %c[fail](%0) \n\t"
+	      : : "c"(vmx), "d"((unsigned long)HOST_RSP),
+		[launched]"i"(offsetof(struct vcpu_vmx, launched)),
+		[fail]"i"(offsetof(struct vcpu_vmx, fail)),
+		[rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])),
+		[rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])),
+		[rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])),
+		[rdx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDX])),
+		[rsi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RSI])),
+		[rdi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDI])),
+		[rbp]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBP])),
+#ifdef CONFIG_X86_64
+		[r8]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R8])),
+		[r9]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R9])),
+		[r10]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R10])),
+		[r11]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R11])),
+		[r12]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R12])),
+		[r13]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R13])),
+		[r14]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R14])),
+		[r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])),
+#endif
+		[cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2))
+	      : "cc", "memory"
+#ifdef CONFIG_X86_64
+		, "rbx", "rdi", "rsi"
+		, "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15"
+#else
+		, "ebx", "edi", "rsi"
+#endif
+	      );
+
+	vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);
+	if (vmx->rmode.irq.pending)
+		fixup_rmode_irq(vmx);
+
+	vcpu->arch.interrupt_window_open =
+		(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0;
+
+	asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS));
+	vmx->launched = 1;
+
+	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
+
+	/* We need to handle NMIs before interrupts are enabled */
+	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
+		asm("int $2");
+}
+
+static void vmx_free_vmcs(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (vmx->vmcs) {
+		on_each_cpu(__vcpu_clear, vmx, 0, 1);
+		free_vmcs(vmx->vmcs);
+		vmx->vmcs = NULL;
+	}
+}
+
+static void vmx_free_vcpu(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	vmx_free_vmcs(vcpu);
+	kfree(vmx->host_msrs);
+	kfree(vmx->guest_msrs);
+	kvm_vcpu_uninit(vcpu);
+	kmem_cache_free(kvm_vcpu_cache, vmx);
+}
+
+static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id)
+{
+	int err;
+	struct vcpu_vmx *vmx = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
+	int cpu;
+
+	if (!vmx)
+		return ERR_PTR(-ENOMEM);
+
+	err = kvm_vcpu_init(&vmx->vcpu, kvm, id);
+	if (err)
+		goto free_vcpu;
+
+	vmx->guest_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!vmx->guest_msrs) {
+		err = -ENOMEM;
+		goto uninit_vcpu;
+	}
+
+	vmx->host_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!vmx->host_msrs)
+		goto free_guest_msrs;
+
+	vmx->vmcs = alloc_vmcs();
+	if (!vmx->vmcs)
+		goto free_msrs;
+
+	vmcs_clear(vmx->vmcs);
+
+	cpu = get_cpu();
+	vmx_vcpu_load(&vmx->vcpu, cpu);
+	err = vmx_vcpu_setup(vmx);
+	vmx_vcpu_put(&vmx->vcpu);
+	put_cpu();
+	if (err)
+		goto free_vmcs;
+
+	return &vmx->vcpu;
+
+free_vmcs:
+	free_vmcs(vmx->vmcs);
+free_msrs:
+	kfree(vmx->host_msrs);
+free_guest_msrs:
+	kfree(vmx->guest_msrs);
+uninit_vcpu:
+	kvm_vcpu_uninit(&vmx->vcpu);
+free_vcpu:
+	kmem_cache_free(kvm_vcpu_cache, vmx);
+	return ERR_PTR(err);
+}
+
+static void __init vmx_check_processor_compat(void *rtn)
+{
+	struct vmcs_config vmcs_conf;
+
+	*(int *)rtn = 0;
+	if (setup_vmcs_config(&vmcs_conf) < 0)
+		*(int *)rtn = -EIO;
+	if (memcmp(&vmcs_config, &vmcs_conf, sizeof(struct vmcs_config)) != 0) {
+		printk(KERN_ERR "kvm: CPU %d feature inconsistency!\n",
+				smp_processor_id());
+		*(int *)rtn = -EIO;
+	}
+}
+
+static struct kvm_x86_ops vmx_x86_ops = {
+	.cpu_has_kvm_support = cpu_has_kvm_support,
+	.disabled_by_bios = vmx_disabled_by_bios,
+	.hardware_setup = hardware_setup,
+	.hardware_unsetup = hardware_unsetup,
+	.check_processor_compatibility = vmx_check_processor_compat,
+	.hardware_enable = hardware_enable,
+	.hardware_disable = hardware_disable,
+
+	.vcpu_create = vmx_create_vcpu,
+	.vcpu_free = vmx_free_vcpu,
+	.vcpu_reset = vmx_vcpu_reset,
+
+	.prepare_guest_switch = vmx_save_host_state,
+	.vcpu_load = vmx_vcpu_load,
+	.vcpu_put = vmx_vcpu_put,
+	.vcpu_decache = vmx_vcpu_decache,
+
+	.set_guest_debug = set_guest_debug,
+	.guest_debug_pre = kvm_guest_debug_pre,
+	.get_msr = vmx_get_msr,
+	.set_msr = vmx_set_msr,
+	.get_segment_base = vmx_get_segment_base,
+	.get_segment = vmx_get_segment,
+	.set_segment = vmx_set_segment,
+	.get_cs_db_l_bits = vmx_get_cs_db_l_bits,
+	.decache_cr4_guest_bits = vmx_decache_cr4_guest_bits,
+	.set_cr0 = vmx_set_cr0,
+	.set_cr3 = vmx_set_cr3,
+	.set_cr4 = vmx_set_cr4,
+#ifdef CONFIG_X86_64
+	.set_efer = vmx_set_efer,
+#endif
+	.get_idt = vmx_get_idt,
+	.set_idt = vmx_set_idt,
+	.get_gdt = vmx_get_gdt,
+	.set_gdt = vmx_set_gdt,
+	.cache_regs = vcpu_load_rsp_rip,
+	.decache_regs = vcpu_put_rsp_rip,
+	.get_rflags = vmx_get_rflags,
+	.set_rflags = vmx_set_rflags,
+
+	.tlb_flush = vmx_flush_tlb,
+
+	.run = vmx_vcpu_run,
+	.handle_exit = kvm_handle_exit,
+	.skip_emulated_instruction = skip_emulated_instruction,
+	.patch_hypercall = vmx_patch_hypercall,
+	.get_irq = vmx_get_irq,
+	.set_irq = vmx_inject_irq,
+	.queue_exception = vmx_queue_exception,
+	.exception_injected = vmx_exception_injected,
+	.inject_pending_irq = vmx_intr_assist,
+	.inject_pending_vectors = do_interrupt_requests,
+
+	.set_tss_addr = vmx_set_tss_addr,
+};
+
+static int __init vmx_init(void)
+{
+	void *iova;
+	int r;
+
+	vmx_io_bitmap_a = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
+	if (!vmx_io_bitmap_a)
+		return -ENOMEM;
+
+	vmx_io_bitmap_b = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
+	if (!vmx_io_bitmap_b) {
+		r = -ENOMEM;
+		goto out;
+	}
+
+	/*
+	 * Allow direct access to the PC debug port (it is often used for I/O
+	 * delays, but the vmexits simply slow things down).
+	 */
+	iova = kmap(vmx_io_bitmap_a);
+	memset(iova, 0xff, PAGE_SIZE);
+	clear_bit(0x80, iova);
+	kunmap(vmx_io_bitmap_a);
+
+	iova = kmap(vmx_io_bitmap_b);
+	memset(iova, 0xff, PAGE_SIZE);
+	kunmap(vmx_io_bitmap_b);
+
+	r = kvm_init(&vmx_x86_ops, sizeof(struct vcpu_vmx), THIS_MODULE);
+	if (r)
+		goto out1;
+
+	if (bypass_guest_pf)
+		kvm_mmu_set_nonpresent_ptes(~0xffeull, 0ull);
+
+	return 0;
+
+out1:
+	__free_page(vmx_io_bitmap_b);
+out:
+	__free_page(vmx_io_bitmap_a);
+	return r;
+}
+
+static void __exit vmx_exit(void)
+{
+	__free_page(vmx_io_bitmap_b);
+	__free_page(vmx_io_bitmap_a);
+
+	kvm_exit();
+}
+
+module_init(vmx_init)
+module_exit(vmx_exit)
diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h
new file mode 100644
index 000000000000..d52ae8d7303d
--- /dev/null
+++ b/arch/x86/kvm/vmx.h
@@ -0,0 +1,324 @@
+#ifndef VMX_H
+#define VMX_H
+
+/*
+ * vmx.h: VMX Architecture related definitions
+ * Copyright (c) 2004, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ *
+ * A few random additions are:
+ * Copyright (C) 2006 Qumranet
+ *    Avi Kivity <avi@qumranet.com>
+ *    Yaniv Kamay <yaniv@qumranet.com>
+ *
+ */
+
+/*
+ * Definitions of Primary Processor-Based VM-Execution Controls.
+ */
+#define CPU_BASED_VIRTUAL_INTR_PENDING          0x00000004
+#define CPU_BASED_USE_TSC_OFFSETING             0x00000008
+#define CPU_BASED_HLT_EXITING                   0x00000080
+#define CPU_BASED_INVLPG_EXITING                0x00000200
+#define CPU_BASED_MWAIT_EXITING                 0x00000400
+#define CPU_BASED_RDPMC_EXITING                 0x00000800
+#define CPU_BASED_RDTSC_EXITING                 0x00001000
+#define CPU_BASED_CR8_LOAD_EXITING              0x00080000
+#define CPU_BASED_CR8_STORE_EXITING             0x00100000
+#define CPU_BASED_TPR_SHADOW                    0x00200000
+#define CPU_BASED_MOV_DR_EXITING                0x00800000
+#define CPU_BASED_UNCOND_IO_EXITING             0x01000000
+#define CPU_BASED_USE_IO_BITMAPS                0x02000000
+#define CPU_BASED_USE_MSR_BITMAPS               0x10000000
+#define CPU_BASED_MONITOR_EXITING               0x20000000
+#define CPU_BASED_PAUSE_EXITING                 0x40000000
+#define CPU_BASED_ACTIVATE_SECONDARY_CONTROLS   0x80000000
+/*
+ * Definitions of Secondary Processor-Based VM-Execution Controls.
+ */
+#define SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES 0x00000001
+#define SECONDARY_EXEC_WBINVD_EXITING		0x00000040
+
+
+#define PIN_BASED_EXT_INTR_MASK                 0x00000001
+#define PIN_BASED_NMI_EXITING                   0x00000008
+#define PIN_BASED_VIRTUAL_NMIS                  0x00000020
+
+#define VM_EXIT_HOST_ADDR_SPACE_SIZE            0x00000200
+#define VM_EXIT_ACK_INTR_ON_EXIT                0x00008000
+
+#define VM_ENTRY_IA32E_MODE                     0x00000200
+#define VM_ENTRY_SMM                            0x00000400
+#define VM_ENTRY_DEACT_DUAL_MONITOR             0x00000800
+
+/* VMCS Encodings */
+enum vmcs_field {
+	GUEST_ES_SELECTOR               = 0x00000800,
+	GUEST_CS_SELECTOR               = 0x00000802,
+	GUEST_SS_SELECTOR               = 0x00000804,
+	GUEST_DS_SELECTOR               = 0x00000806,
+	GUEST_FS_SELECTOR               = 0x00000808,
+	GUEST_GS_SELECTOR               = 0x0000080a,
+	GUEST_LDTR_SELECTOR             = 0x0000080c,
+	GUEST_TR_SELECTOR               = 0x0000080e,
+	HOST_ES_SELECTOR                = 0x00000c00,
+	HOST_CS_SELECTOR                = 0x00000c02,
+	HOST_SS_SELECTOR                = 0x00000c04,
+	HOST_DS_SELECTOR                = 0x00000c06,
+	HOST_FS_SELECTOR                = 0x00000c08,
+	HOST_GS_SELECTOR                = 0x00000c0a,
+	HOST_TR_SELECTOR                = 0x00000c0c,
+	IO_BITMAP_A                     = 0x00002000,
+	IO_BITMAP_A_HIGH                = 0x00002001,
+	IO_BITMAP_B                     = 0x00002002,
+	IO_BITMAP_B_HIGH                = 0x00002003,
+	MSR_BITMAP                      = 0x00002004,
+	MSR_BITMAP_HIGH                 = 0x00002005,
+	VM_EXIT_MSR_STORE_ADDR          = 0x00002006,
+	VM_EXIT_MSR_STORE_ADDR_HIGH     = 0x00002007,
+	VM_EXIT_MSR_LOAD_ADDR           = 0x00002008,
+	VM_EXIT_MSR_LOAD_ADDR_HIGH      = 0x00002009,
+	VM_ENTRY_MSR_LOAD_ADDR          = 0x0000200a,
+	VM_ENTRY_MSR_LOAD_ADDR_HIGH     = 0x0000200b,
+	TSC_OFFSET                      = 0x00002010,
+	TSC_OFFSET_HIGH                 = 0x00002011,
+	VIRTUAL_APIC_PAGE_ADDR          = 0x00002012,
+	VIRTUAL_APIC_PAGE_ADDR_HIGH     = 0x00002013,
+	APIC_ACCESS_ADDR		= 0x00002014,
+	APIC_ACCESS_ADDR_HIGH		= 0x00002015,
+	VMCS_LINK_POINTER               = 0x00002800,
+	VMCS_LINK_POINTER_HIGH          = 0x00002801,
+	GUEST_IA32_DEBUGCTL             = 0x00002802,
+	GUEST_IA32_DEBUGCTL_HIGH        = 0x00002803,
+	PIN_BASED_VM_EXEC_CONTROL       = 0x00004000,
+	CPU_BASED_VM_EXEC_CONTROL       = 0x00004002,
+	EXCEPTION_BITMAP                = 0x00004004,
+	PAGE_FAULT_ERROR_CODE_MASK      = 0x00004006,
+	PAGE_FAULT_ERROR_CODE_MATCH     = 0x00004008,
+	CR3_TARGET_COUNT                = 0x0000400a,
+	VM_EXIT_CONTROLS                = 0x0000400c,
+	VM_EXIT_MSR_STORE_COUNT         = 0x0000400e,
+	VM_EXIT_MSR_LOAD_COUNT          = 0x00004010,
+	VM_ENTRY_CONTROLS               = 0x00004012,
+	VM_ENTRY_MSR_LOAD_COUNT         = 0x00004014,
+	VM_ENTRY_INTR_INFO_FIELD        = 0x00004016,
+	VM_ENTRY_EXCEPTION_ERROR_CODE   = 0x00004018,
+	VM_ENTRY_INSTRUCTION_LEN        = 0x0000401a,
+	TPR_THRESHOLD                   = 0x0000401c,
+	SECONDARY_VM_EXEC_CONTROL       = 0x0000401e,
+	VM_INSTRUCTION_ERROR            = 0x00004400,
+	VM_EXIT_REASON                  = 0x00004402,
+	VM_EXIT_INTR_INFO               = 0x00004404,
+	VM_EXIT_INTR_ERROR_CODE         = 0x00004406,
+	IDT_VECTORING_INFO_FIELD        = 0x00004408,
+	IDT_VECTORING_ERROR_CODE        = 0x0000440a,
+	VM_EXIT_INSTRUCTION_LEN         = 0x0000440c,
+	VMX_INSTRUCTION_INFO            = 0x0000440e,
+	GUEST_ES_LIMIT                  = 0x00004800,
+	GUEST_CS_LIMIT                  = 0x00004802,
+	GUEST_SS_LIMIT                  = 0x00004804,
+	GUEST_DS_LIMIT                  = 0x00004806,
+	GUEST_FS_LIMIT                  = 0x00004808,
+	GUEST_GS_LIMIT                  = 0x0000480a,
+	GUEST_LDTR_LIMIT                = 0x0000480c,
+	GUEST_TR_LIMIT                  = 0x0000480e,
+	GUEST_GDTR_LIMIT                = 0x00004810,
+	GUEST_IDTR_LIMIT                = 0x00004812,
+	GUEST_ES_AR_BYTES               = 0x00004814,
+	GUEST_CS_AR_BYTES               = 0x00004816,
+	GUEST_SS_AR_BYTES               = 0x00004818,
+	GUEST_DS_AR_BYTES               = 0x0000481a,
+	GUEST_FS_AR_BYTES               = 0x0000481c,
+	GUEST_GS_AR_BYTES               = 0x0000481e,
+	GUEST_LDTR_AR_BYTES             = 0x00004820,
+	GUEST_TR_AR_BYTES               = 0x00004822,
+	GUEST_INTERRUPTIBILITY_INFO     = 0x00004824,
+	GUEST_ACTIVITY_STATE            = 0X00004826,
+	GUEST_SYSENTER_CS               = 0x0000482A,
+	HOST_IA32_SYSENTER_CS           = 0x00004c00,
+	CR0_GUEST_HOST_MASK             = 0x00006000,
+	CR4_GUEST_HOST_MASK             = 0x00006002,
+	CR0_READ_SHADOW                 = 0x00006004,
+	CR4_READ_SHADOW                 = 0x00006006,
+	CR3_TARGET_VALUE0               = 0x00006008,
+	CR3_TARGET_VALUE1               = 0x0000600a,
+	CR3_TARGET_VALUE2               = 0x0000600c,
+	CR3_TARGET_VALUE3               = 0x0000600e,
+	EXIT_QUALIFICATION              = 0x00006400,
+	GUEST_LINEAR_ADDRESS            = 0x0000640a,
+	GUEST_CR0                       = 0x00006800,
+	GUEST_CR3                       = 0x00006802,
+	GUEST_CR4                       = 0x00006804,
+	GUEST_ES_BASE                   = 0x00006806,
+	GUEST_CS_BASE                   = 0x00006808,
+	GUEST_SS_BASE                   = 0x0000680a,
+	GUEST_DS_BASE                   = 0x0000680c,
+	GUEST_FS_BASE                   = 0x0000680e,
+	GUEST_GS_BASE                   = 0x00006810,
+	GUEST_LDTR_BASE                 = 0x00006812,
+	GUEST_TR_BASE                   = 0x00006814,
+	GUEST_GDTR_BASE                 = 0x00006816,
+	GUEST_IDTR_BASE                 = 0x00006818,
+	GUEST_DR7                       = 0x0000681a,
+	GUEST_RSP                       = 0x0000681c,
+	GUEST_RIP                       = 0x0000681e,
+	GUEST_RFLAGS                    = 0x00006820,
+	GUEST_PENDING_DBG_EXCEPTIONS    = 0x00006822,
+	GUEST_SYSENTER_ESP              = 0x00006824,
+	GUEST_SYSENTER_EIP              = 0x00006826,
+	HOST_CR0                        = 0x00006c00,
+	HOST_CR3                        = 0x00006c02,
+	HOST_CR4                        = 0x00006c04,
+	HOST_FS_BASE                    = 0x00006c06,
+	HOST_GS_BASE                    = 0x00006c08,
+	HOST_TR_BASE                    = 0x00006c0a,
+	HOST_GDTR_BASE                  = 0x00006c0c,
+	HOST_IDTR_BASE                  = 0x00006c0e,
+	HOST_IA32_SYSENTER_ESP          = 0x00006c10,
+	HOST_IA32_SYSENTER_EIP          = 0x00006c12,
+	HOST_RSP                        = 0x00006c14,
+	HOST_RIP                        = 0x00006c16,
+};
+
+#define VMX_EXIT_REASONS_FAILED_VMENTRY         0x80000000
+
+#define EXIT_REASON_EXCEPTION_NMI       0
+#define EXIT_REASON_EXTERNAL_INTERRUPT  1
+#define EXIT_REASON_TRIPLE_FAULT        2
+
+#define EXIT_REASON_PENDING_INTERRUPT   7
+
+#define EXIT_REASON_TASK_SWITCH         9
+#define EXIT_REASON_CPUID               10
+#define EXIT_REASON_HLT                 12
+#define EXIT_REASON_INVLPG              14
+#define EXIT_REASON_RDPMC               15
+#define EXIT_REASON_RDTSC               16
+#define EXIT_REASON_VMCALL              18
+#define EXIT_REASON_VMCLEAR             19
+#define EXIT_REASON_VMLAUNCH            20
+#define EXIT_REASON_VMPTRLD             21
+#define EXIT_REASON_VMPTRST             22
+#define EXIT_REASON_VMREAD              23
+#define EXIT_REASON_VMRESUME            24
+#define EXIT_REASON_VMWRITE             25
+#define EXIT_REASON_VMOFF               26
+#define EXIT_REASON_VMON                27
+#define EXIT_REASON_CR_ACCESS           28
+#define EXIT_REASON_DR_ACCESS           29
+#define EXIT_REASON_IO_INSTRUCTION      30
+#define EXIT_REASON_MSR_READ            31
+#define EXIT_REASON_MSR_WRITE           32
+#define EXIT_REASON_MWAIT_INSTRUCTION   36
+#define EXIT_REASON_TPR_BELOW_THRESHOLD 43
+#define EXIT_REASON_APIC_ACCESS         44
+#define EXIT_REASON_WBINVD		54
+
+/*
+ * Interruption-information format
+ */
+#define INTR_INFO_VECTOR_MASK           0xff            /* 7:0 */
+#define INTR_INFO_INTR_TYPE_MASK        0x700           /* 10:8 */
+#define INTR_INFO_DELIEVER_CODE_MASK    0x800           /* 11 */
+#define INTR_INFO_VALID_MASK            0x80000000      /* 31 */
+
+#define VECTORING_INFO_VECTOR_MASK           	INTR_INFO_VECTOR_MASK
+#define VECTORING_INFO_TYPE_MASK        	INTR_INFO_INTR_TYPE_MASK
+#define VECTORING_INFO_DELIEVER_CODE_MASK    	INTR_INFO_DELIEVER_CODE_MASK
+#define VECTORING_INFO_VALID_MASK       	INTR_INFO_VALID_MASK
+
+#define INTR_TYPE_EXT_INTR              (0 << 8) /* external interrupt */
+#define INTR_TYPE_EXCEPTION             (3 << 8) /* processor exception */
+#define INTR_TYPE_SOFT_INTR             (4 << 8) /* software interrupt */
+
+/*
+ * Exit Qualifications for MOV for Control Register Access
+ */
+#define CONTROL_REG_ACCESS_NUM          0x7     /* 2:0, number of control reg.*/
+#define CONTROL_REG_ACCESS_TYPE         0x30    /* 5:4, access type */
+#define CONTROL_REG_ACCESS_REG          0xf00   /* 10:8, general purpose reg. */
+#define LMSW_SOURCE_DATA_SHIFT 16
+#define LMSW_SOURCE_DATA  (0xFFFF << LMSW_SOURCE_DATA_SHIFT) /* 16:31 lmsw source */
+#define REG_EAX                         (0 << 8)
+#define REG_ECX                         (1 << 8)
+#define REG_EDX                         (2 << 8)
+#define REG_EBX                         (3 << 8)
+#define REG_ESP                         (4 << 8)
+#define REG_EBP                         (5 << 8)
+#define REG_ESI                         (6 << 8)
+#define REG_EDI                         (7 << 8)
+#define REG_R8                         (8 << 8)
+#define REG_R9                         (9 << 8)
+#define REG_R10                        (10 << 8)
+#define REG_R11                        (11 << 8)
+#define REG_R12                        (12 << 8)
+#define REG_R13                        (13 << 8)
+#define REG_R14                        (14 << 8)
+#define REG_R15                        (15 << 8)
+
+/*
+ * Exit Qualifications for MOV for Debug Register Access
+ */
+#define DEBUG_REG_ACCESS_NUM            0x7     /* 2:0, number of debug reg. */
+#define DEBUG_REG_ACCESS_TYPE           0x10    /* 4, direction of access */
+#define TYPE_MOV_TO_DR                  (0 << 4)
+#define TYPE_MOV_FROM_DR                (1 << 4)
+#define DEBUG_REG_ACCESS_REG            0xf00   /* 11:8, general purpose reg. */
+
+
+/* segment AR */
+#define SEGMENT_AR_L_MASK (1 << 13)
+
+#define AR_TYPE_ACCESSES_MASK 1
+#define AR_TYPE_READABLE_MASK (1 << 1)
+#define AR_TYPE_WRITEABLE_MASK (1 << 2)
+#define AR_TYPE_CODE_MASK (1 << 3)
+#define AR_TYPE_MASK 0x0f
+#define AR_TYPE_BUSY_64_TSS 11
+#define AR_TYPE_BUSY_32_TSS 11
+#define AR_TYPE_BUSY_16_TSS 3
+#define AR_TYPE_LDT 2
+
+#define AR_UNUSABLE_MASK (1 << 16)
+#define AR_S_MASK (1 << 4)
+#define AR_P_MASK (1 << 7)
+#define AR_L_MASK (1 << 13)
+#define AR_DB_MASK (1 << 14)
+#define AR_G_MASK (1 << 15)
+#define AR_DPL_SHIFT 5
+#define AR_DPL(ar) (((ar) >> AR_DPL_SHIFT) & 3)
+
+#define AR_RESERVD_MASK 0xfffe0f00
+
+#define MSR_IA32_VMX_BASIC                      0x480
+#define MSR_IA32_VMX_PINBASED_CTLS              0x481
+#define MSR_IA32_VMX_PROCBASED_CTLS             0x482
+#define MSR_IA32_VMX_EXIT_CTLS                  0x483
+#define MSR_IA32_VMX_ENTRY_CTLS                 0x484
+#define MSR_IA32_VMX_MISC                       0x485
+#define MSR_IA32_VMX_CR0_FIXED0                 0x486
+#define MSR_IA32_VMX_CR0_FIXED1                 0x487
+#define MSR_IA32_VMX_CR4_FIXED0                 0x488
+#define MSR_IA32_VMX_CR4_FIXED1                 0x489
+#define MSR_IA32_VMX_VMCS_ENUM                  0x48a
+#define MSR_IA32_VMX_PROCBASED_CTLS2            0x48b
+
+#define MSR_IA32_FEATURE_CONTROL                0x3a
+#define MSR_IA32_FEATURE_CONTROL_LOCKED         0x1
+#define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED  0x4
+
+#define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT	9
+
+#endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
new file mode 100644
index 000000000000..5902c5cbc1bb
--- /dev/null
+++ b/arch/x86/kvm/x86.c
@@ -0,0 +1,3146 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * derived from drivers/kvm/kvm_main.c
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ *
+ * Authors:
+ *   Avi Kivity   <avi@qumranet.com>
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include <linux/kvm_host.h>
+#include "segment_descriptor.h"
+#include "irq.h"
+#include "mmu.h"
+
+#include <linux/kvm.h>
+#include <linux/fs.h>
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+#include <linux/mman.h>
+#include <linux/highmem.h>
+
+#include <asm/uaccess.h>
+#include <asm/msr.h>
+
+#define MAX_IO_MSRS 256
+#define CR0_RESERVED_BITS						\
+	(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
+			  | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \
+			  | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG))
+#define CR4_RESERVED_BITS						\
+	(~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\
+			  | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE	\
+			  | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR	\
+			  | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE))
+
+#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR)
+#define EFER_RESERVED_BITS 0xfffffffffffff2fe
+
+#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
+#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
+
+struct kvm_x86_ops *kvm_x86_ops;
+
+struct kvm_stats_debugfs_item debugfs_entries[] = {
+	{ "pf_fixed", VCPU_STAT(pf_fixed) },
+	{ "pf_guest", VCPU_STAT(pf_guest) },
+	{ "tlb_flush", VCPU_STAT(tlb_flush) },
+	{ "invlpg", VCPU_STAT(invlpg) },
+	{ "exits", VCPU_STAT(exits) },
+	{ "io_exits", VCPU_STAT(io_exits) },
+	{ "mmio_exits", VCPU_STAT(mmio_exits) },
+	{ "signal_exits", VCPU_STAT(signal_exits) },
+	{ "irq_window", VCPU_STAT(irq_window_exits) },
+	{ "halt_exits", VCPU_STAT(halt_exits) },
+	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
+	{ "request_irq", VCPU_STAT(request_irq_exits) },
+	{ "irq_exits", VCPU_STAT(irq_exits) },
+	{ "host_state_reload", VCPU_STAT(host_state_reload) },
+	{ "efer_reload", VCPU_STAT(efer_reload) },
+	{ "fpu_reload", VCPU_STAT(fpu_reload) },
+	{ "insn_emulation", VCPU_STAT(insn_emulation) },
+	{ "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) },
+	{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) },
+	{ "mmu_pte_write", VM_STAT(mmu_pte_write) },
+	{ "mmu_pte_updated", VM_STAT(mmu_pte_updated) },
+	{ "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) },
+	{ "mmu_flooded", VM_STAT(mmu_flooded) },
+	{ "mmu_recycled", VM_STAT(mmu_recycled) },
+	{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) },
+	{ NULL }
+};
+
+
+unsigned long segment_base(u16 selector)
+{
+	struct descriptor_table gdt;
+	struct segment_descriptor *d;
+	unsigned long table_base;
+	unsigned long v;
+
+	if (selector == 0)
+		return 0;
+
+	asm("sgdt %0" : "=m"(gdt));
+	table_base = gdt.base;
+
+	if (selector & 4) {           /* from ldt */
+		u16 ldt_selector;
+
+		asm("sldt %0" : "=g"(ldt_selector));
+		table_base = segment_base(ldt_selector);
+	}
+	d = (struct segment_descriptor *)(table_base + (selector & ~7));
+	v = d->base_low | ((unsigned long)d->base_mid << 16) |
+		((unsigned long)d->base_high << 24);
+#ifdef CONFIG_X86_64
+	if (d->system == 0 && (d->type == 2 || d->type == 9 || d->type == 11))
+		v |= ((unsigned long) \
+		      ((struct segment_descriptor_64 *)d)->base_higher) << 32;
+#endif
+	return v;
+}
+EXPORT_SYMBOL_GPL(segment_base);
+
+u64 kvm_get_apic_base(struct kvm_vcpu *vcpu)
+{
+	if (irqchip_in_kernel(vcpu->kvm))
+		return vcpu->arch.apic_base;
+	else
+		return vcpu->arch.apic_base;
+}
+EXPORT_SYMBOL_GPL(kvm_get_apic_base);
+
+void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data)
+{
+	/* TODO: reserve bits check */
+	if (irqchip_in_kernel(vcpu->kvm))
+		kvm_lapic_set_base(vcpu, data);
+	else
+		vcpu->arch.apic_base = data;
+}
+EXPORT_SYMBOL_GPL(kvm_set_apic_base);
+
+void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr)
+{
+	WARN_ON(vcpu->arch.exception.pending);
+	vcpu->arch.exception.pending = true;
+	vcpu->arch.exception.has_error_code = false;
+	vcpu->arch.exception.nr = nr;
+}
+EXPORT_SYMBOL_GPL(kvm_queue_exception);
+
+void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr,
+			   u32 error_code)
+{
+	++vcpu->stat.pf_guest;
+	if (vcpu->arch.exception.pending && vcpu->arch.exception.nr == PF_VECTOR) {
+		printk(KERN_DEBUG "kvm: inject_page_fault:"
+		       " double fault 0x%lx\n", addr);
+		vcpu->arch.exception.nr = DF_VECTOR;
+		vcpu->arch.exception.error_code = 0;
+		return;
+	}
+	vcpu->arch.cr2 = addr;
+	kvm_queue_exception_e(vcpu, PF_VECTOR, error_code);
+}
+
+void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code)
+{
+	WARN_ON(vcpu->arch.exception.pending);
+	vcpu->arch.exception.pending = true;
+	vcpu->arch.exception.has_error_code = true;
+	vcpu->arch.exception.nr = nr;
+	vcpu->arch.exception.error_code = error_code;
+}
+EXPORT_SYMBOL_GPL(kvm_queue_exception_e);
+
+static void __queue_exception(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->queue_exception(vcpu, vcpu->arch.exception.nr,
+				     vcpu->arch.exception.has_error_code,
+				     vcpu->arch.exception.error_code);
+}
+
+/*
+ * Load the pae pdptrs.  Return true is they are all valid.
+ */
+int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
+{
+	gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT;
+	unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2;
+	int i;
+	int ret;
+	u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
+
+	mutex_lock(&vcpu->kvm->lock);
+	ret = kvm_read_guest_page(vcpu->kvm, pdpt_gfn, pdpte,
+				  offset * sizeof(u64), sizeof(pdpte));
+	if (ret < 0) {
+		ret = 0;
+		goto out;
+	}
+	for (i = 0; i < ARRAY_SIZE(pdpte); ++i) {
+		if ((pdpte[i] & 1) && (pdpte[i] & 0xfffffff0000001e6ull)) {
+			ret = 0;
+			goto out;
+		}
+	}
+	ret = 1;
+
+	memcpy(vcpu->arch.pdptrs, pdpte, sizeof(vcpu->arch.pdptrs));
+out:
+	mutex_unlock(&vcpu->kvm->lock);
+
+	return ret;
+}
+
+static bool pdptrs_changed(struct kvm_vcpu *vcpu)
+{
+	u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
+	bool changed = true;
+	int r;
+
+	if (is_long_mode(vcpu) || !is_pae(vcpu))
+		return false;
+
+	mutex_lock(&vcpu->kvm->lock);
+	r = kvm_read_guest(vcpu->kvm, vcpu->arch.cr3 & ~31u, pdpte, sizeof(pdpte));
+	if (r < 0)
+		goto out;
+	changed = memcmp(pdpte, vcpu->arch.pdptrs, sizeof(pdpte)) != 0;
+out:
+	mutex_unlock(&vcpu->kvm->lock);
+
+	return changed;
+}
+
+void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
+{
+	if (cr0 & CR0_RESERVED_BITS) {
+		printk(KERN_DEBUG "set_cr0: 0x%lx #GP, reserved bits 0x%lx\n",
+		       cr0, vcpu->arch.cr0);
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) {
+		printk(KERN_DEBUG "set_cr0: #GP, CD == 0 && NW == 1\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) {
+		printk(KERN_DEBUG "set_cr0: #GP, set PG flag "
+		       "and a clear PE flag\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
+#ifdef CONFIG_X86_64
+		if ((vcpu->arch.shadow_efer & EFER_LME)) {
+			int cs_db, cs_l;
+
+			if (!is_pae(vcpu)) {
+				printk(KERN_DEBUG "set_cr0: #GP, start paging "
+				       "in long mode while PAE is disabled\n");
+				kvm_inject_gp(vcpu, 0);
+				return;
+			}
+			kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
+			if (cs_l) {
+				printk(KERN_DEBUG "set_cr0: #GP, start paging "
+				       "in long mode while CS.L == 1\n");
+				kvm_inject_gp(vcpu, 0);
+				return;
+
+			}
+		} else
+#endif
+		if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
+			printk(KERN_DEBUG "set_cr0: #GP, pdptrs "
+			       "reserved bits\n");
+			kvm_inject_gp(vcpu, 0);
+			return;
+		}
+
+	}
+
+	kvm_x86_ops->set_cr0(vcpu, cr0);
+	vcpu->arch.cr0 = cr0;
+
+	mutex_lock(&vcpu->kvm->lock);
+	kvm_mmu_reset_context(vcpu);
+	mutex_unlock(&vcpu->kvm->lock);
+	return;
+}
+EXPORT_SYMBOL_GPL(set_cr0);
+
+void lmsw(struct kvm_vcpu *vcpu, unsigned long msw)
+{
+	set_cr0(vcpu, (vcpu->arch.cr0 & ~0x0ful) | (msw & 0x0f));
+}
+EXPORT_SYMBOL_GPL(lmsw);
+
+void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
+{
+	if (cr4 & CR4_RESERVED_BITS) {
+		printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if (is_long_mode(vcpu)) {
+		if (!(cr4 & X86_CR4_PAE)) {
+			printk(KERN_DEBUG "set_cr4: #GP, clearing PAE while "
+			       "in long mode\n");
+			kvm_inject_gp(vcpu, 0);
+			return;
+		}
+	} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE)
+		   && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
+		printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if (cr4 & X86_CR4_VMXE) {
+		printk(KERN_DEBUG "set_cr4: #GP, setting VMXE\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+	kvm_x86_ops->set_cr4(vcpu, cr4);
+	vcpu->arch.cr4 = cr4;
+	mutex_lock(&vcpu->kvm->lock);
+	kvm_mmu_reset_context(vcpu);
+	mutex_unlock(&vcpu->kvm->lock);
+}
+EXPORT_SYMBOL_GPL(set_cr4);
+
+void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
+{
+	if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) {
+		kvm_mmu_flush_tlb(vcpu);
+		return;
+	}
+
+	if (is_long_mode(vcpu)) {
+		if (cr3 & CR3_L_MODE_RESERVED_BITS) {
+			printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");
+			kvm_inject_gp(vcpu, 0);
+			return;
+		}
+	} else {
+		if (is_pae(vcpu)) {
+			if (cr3 & CR3_PAE_RESERVED_BITS) {
+				printk(KERN_DEBUG
+				       "set_cr3: #GP, reserved bits\n");
+				kvm_inject_gp(vcpu, 0);
+				return;
+			}
+			if (is_paging(vcpu) && !load_pdptrs(vcpu, cr3)) {
+				printk(KERN_DEBUG "set_cr3: #GP, pdptrs "
+				       "reserved bits\n");
+				kvm_inject_gp(vcpu, 0);
+				return;
+			}
+		}
+		/*
+		 * We don't check reserved bits in nonpae mode, because
+		 * this isn't enforced, and VMware depends on this.
+		 */
+	}
+
+	mutex_lock(&vcpu->kvm->lock);
+	/*
+	 * Does the new cr3 value map to physical memory? (Note, we
+	 * catch an invalid cr3 even in real-mode, because it would
+	 * cause trouble later on when we turn on paging anyway.)
+	 *
+	 * A real CPU would silently accept an invalid cr3 and would
+	 * attempt to use it - with largely undefined (and often hard
+	 * to debug) behavior on the guest side.
+	 */
+	if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT)))
+		kvm_inject_gp(vcpu, 0);
+	else {
+		vcpu->arch.cr3 = cr3;
+		vcpu->arch.mmu.new_cr3(vcpu);
+	}
+	mutex_unlock(&vcpu->kvm->lock);
+}
+EXPORT_SYMBOL_GPL(set_cr3);
+
+void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
+{
+	if (cr8 & CR8_RESERVED_BITS) {
+		printk(KERN_DEBUG "set_cr8: #GP, reserved bits 0x%lx\n", cr8);
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+	if (irqchip_in_kernel(vcpu->kvm))
+		kvm_lapic_set_tpr(vcpu, cr8);
+	else
+		vcpu->arch.cr8 = cr8;
+}
+EXPORT_SYMBOL_GPL(set_cr8);
+
+unsigned long get_cr8(struct kvm_vcpu *vcpu)
+{
+	if (irqchip_in_kernel(vcpu->kvm))
+		return kvm_lapic_get_cr8(vcpu);
+	else
+		return vcpu->arch.cr8;
+}
+EXPORT_SYMBOL_GPL(get_cr8);
+
+/*
+ * List of msr numbers which we expose to userspace through KVM_GET_MSRS
+ * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST.
+ *
+ * This list is modified at module load time to reflect the
+ * capabilities of the host cpu.
+ */
+static u32 msrs_to_save[] = {
+	MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
+	MSR_K6_STAR,
+#ifdef CONFIG_X86_64
+	MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
+#endif
+	MSR_IA32_TIME_STAMP_COUNTER,
+};
+
+static unsigned num_msrs_to_save;
+
+static u32 emulated_msrs[] = {
+	MSR_IA32_MISC_ENABLE,
+};
+
+#ifdef CONFIG_X86_64
+
+static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
+{
+	if (efer & EFER_RESERVED_BITS) {
+		printk(KERN_DEBUG "set_efer: 0x%llx #GP, reserved bits\n",
+		       efer);
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	if (is_paging(vcpu)
+	    && (vcpu->arch.shadow_efer & EFER_LME) != (efer & EFER_LME)) {
+		printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n");
+		kvm_inject_gp(vcpu, 0);
+		return;
+	}
+
+	kvm_x86_ops->set_efer(vcpu, efer);
+
+	efer &= ~EFER_LMA;
+	efer |= vcpu->arch.shadow_efer & EFER_LMA;
+
+	vcpu->arch.shadow_efer = efer;
+}
+
+#endif
+
+/*
+ * Writes msr value into into the appropriate "register".
+ * Returns 0 on success, non-0 otherwise.
+ * Assumes vcpu_load() was already called.
+ */
+int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
+{
+	return kvm_x86_ops->set_msr(vcpu, msr_index, data);
+}
+
+/*
+ * Adapt set_msr() to msr_io()'s calling convention
+ */
+static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data)
+{
+	return kvm_set_msr(vcpu, index, *data);
+}
+
+
+int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
+{
+	switch (msr) {
+#ifdef CONFIG_X86_64
+	case MSR_EFER:
+		set_efer(vcpu, data);
+		break;
+#endif
+	case MSR_IA32_MC0_STATUS:
+		pr_unimpl(vcpu, "%s: MSR_IA32_MC0_STATUS 0x%llx, nop\n",
+		       __FUNCTION__, data);
+		break;
+	case MSR_IA32_MCG_STATUS:
+		pr_unimpl(vcpu, "%s: MSR_IA32_MCG_STATUS 0x%llx, nop\n",
+			__FUNCTION__, data);
+		break;
+	case MSR_IA32_UCODE_REV:
+	case MSR_IA32_UCODE_WRITE:
+	case 0x200 ... 0x2ff: /* MTRRs */
+		break;
+	case MSR_IA32_APICBASE:
+		kvm_set_apic_base(vcpu, data);
+		break;
+	case MSR_IA32_MISC_ENABLE:
+		vcpu->arch.ia32_misc_enable_msr = data;
+		break;
+	default:
+		pr_unimpl(vcpu, "unhandled wrmsr: 0x%x\n", msr);
+		return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kvm_set_msr_common);
+
+
+/*
+ * Reads an msr value (of 'msr_index') into 'pdata'.
+ * Returns 0 on success, non-0 otherwise.
+ * Assumes vcpu_load() was already called.
+ */
+int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata)
+{
+	return kvm_x86_ops->get_msr(vcpu, msr_index, pdata);
+}
+
+int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
+{
+	u64 data;
+
+	switch (msr) {
+	case 0xc0010010: /* SYSCFG */
+	case 0xc0010015: /* HWCR */
+	case MSR_IA32_PLATFORM_ID:
+	case MSR_IA32_P5_MC_ADDR:
+	case MSR_IA32_P5_MC_TYPE:
+	case MSR_IA32_MC0_CTL:
+	case MSR_IA32_MCG_STATUS:
+	case MSR_IA32_MCG_CAP:
+	case MSR_IA32_MC0_MISC:
+	case MSR_IA32_MC0_MISC+4:
+	case MSR_IA32_MC0_MISC+8:
+	case MSR_IA32_MC0_MISC+12:
+	case MSR_IA32_MC0_MISC+16:
+	case MSR_IA32_UCODE_REV:
+	case MSR_IA32_PERF_STATUS:
+	case MSR_IA32_EBL_CR_POWERON:
+		/* MTRR registers */
+	case 0xfe:
+	case 0x200 ... 0x2ff:
+		data = 0;
+		break;
+	case 0xcd: /* fsb frequency */
+		data = 3;
+		break;
+	case MSR_IA32_APICBASE:
+		data = kvm_get_apic_base(vcpu);
+		break;
+	case MSR_IA32_MISC_ENABLE:
+		data = vcpu->arch.ia32_misc_enable_msr;
+		break;
+#ifdef CONFIG_X86_64
+	case MSR_EFER:
+		data = vcpu->arch.shadow_efer;
+		break;
+#endif
+	default:
+		pr_unimpl(vcpu, "unhandled rdmsr: 0x%x\n", msr);
+		return 1;
+	}
+	*pdata = data;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kvm_get_msr_common);
+
+/*
+ * Read or write a bunch of msrs. All parameters are kernel addresses.
+ *
+ * @return number of msrs set successfully.
+ */
+static int __msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs *msrs,
+		    struct kvm_msr_entry *entries,
+		    int (*do_msr)(struct kvm_vcpu *vcpu,
+				  unsigned index, u64 *data))
+{
+	int i;
+
+	vcpu_load(vcpu);
+
+	for (i = 0; i < msrs->nmsrs; ++i)
+		if (do_msr(vcpu, entries[i].index, &entries[i].data))
+			break;
+
+	vcpu_put(vcpu);
+
+	return i;
+}
+
+/*
+ * Read or write a bunch of msrs. Parameters are user addresses.
+ *
+ * @return number of msrs set successfully.
+ */
+static int msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs __user *user_msrs,
+		  int (*do_msr)(struct kvm_vcpu *vcpu,
+				unsigned index, u64 *data),
+		  int writeback)
+{
+	struct kvm_msrs msrs;
+	struct kvm_msr_entry *entries;
+	int r, n;
+	unsigned size;
+
+	r = -EFAULT;
+	if (copy_from_user(&msrs, user_msrs, sizeof msrs))
+		goto out;
+
+	r = -E2BIG;
+	if (msrs.nmsrs >= MAX_IO_MSRS)
+		goto out;
+
+	r = -ENOMEM;
+	size = sizeof(struct kvm_msr_entry) * msrs.nmsrs;
+	entries = vmalloc(size);
+	if (!entries)
+		goto out;
+
+	r = -EFAULT;
+	if (copy_from_user(entries, user_msrs->entries, size))
+		goto out_free;
+
+	r = n = __msr_io(vcpu, &msrs, entries, do_msr);
+	if (r < 0)
+		goto out_free;
+
+	r = -EFAULT;
+	if (writeback && copy_to_user(user_msrs->entries, entries, size))
+		goto out_free;
+
+	r = n;
+
+out_free:
+	vfree(entries);
+out:
+	return r;
+}
+
+/*
+ * Make sure that a cpu that is being hot-unplugged does not have any vcpus
+ * cached on it.
+ */
+void decache_vcpus_on_cpu(int cpu)
+{
+	struct kvm *vm;
+	struct kvm_vcpu *vcpu;
+	int i;
+
+	spin_lock(&kvm_lock);
+	list_for_each_entry(vm, &vm_list, vm_list)
+		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+			vcpu = vm->vcpus[i];
+			if (!vcpu)
+				continue;
+			/*
+			 * If the vcpu is locked, then it is running on some
+			 * other cpu and therefore it is not cached on the
+			 * cpu in question.
+			 *
+			 * If it's not locked, check the last cpu it executed
+			 * on.
+			 */
+			if (mutex_trylock(&vcpu->mutex)) {
+				if (vcpu->cpu == cpu) {
+					kvm_x86_ops->vcpu_decache(vcpu);
+					vcpu->cpu = -1;
+				}
+				mutex_unlock(&vcpu->mutex);
+			}
+		}
+	spin_unlock(&kvm_lock);
+}
+
+int kvm_dev_ioctl_check_extension(long ext)
+{
+	int r;
+
+	switch (ext) {
+	case KVM_CAP_IRQCHIP:
+	case KVM_CAP_HLT:
+	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
+	case KVM_CAP_USER_MEMORY:
+	case KVM_CAP_SET_TSS_ADDR:
+	case KVM_CAP_EXT_CPUID:
+		r = 1;
+		break;
+	default:
+		r = 0;
+		break;
+	}
+	return r;
+
+}
+
+long kvm_arch_dev_ioctl(struct file *filp,
+			unsigned int ioctl, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	long r;
+
+	switch (ioctl) {
+	case KVM_GET_MSR_INDEX_LIST: {
+		struct kvm_msr_list __user *user_msr_list = argp;
+		struct kvm_msr_list msr_list;
+		unsigned n;
+
+		r = -EFAULT;
+		if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list))
+			goto out;
+		n = msr_list.nmsrs;
+		msr_list.nmsrs = num_msrs_to_save + ARRAY_SIZE(emulated_msrs);
+		if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list))
+			goto out;
+		r = -E2BIG;
+		if (n < num_msrs_to_save)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(user_msr_list->indices, &msrs_to_save,
+				 num_msrs_to_save * sizeof(u32)))
+			goto out;
+		if (copy_to_user(user_msr_list->indices
+				 + num_msrs_to_save * sizeof(u32),
+				 &emulated_msrs,
+				 ARRAY_SIZE(emulated_msrs) * sizeof(u32)))
+			goto out;
+		r = 0;
+		break;
+	}
+	default:
+		r = -EINVAL;
+	}
+out:
+	return r;
+}
+
+void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	kvm_x86_ops->vcpu_load(vcpu, cpu);
+}
+
+void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->vcpu_put(vcpu);
+	kvm_put_guest_fpu(vcpu);
+}
+
+static int is_efer_nx(void)
+{
+	u64 efer;
+
+	rdmsrl(MSR_EFER, efer);
+	return efer & EFER_NX;
+}
+
+static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_cpuid_entry2 *e, *entry;
+
+	entry = NULL;
+	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
+		e = &vcpu->arch.cpuid_entries[i];
+		if (e->function == 0x80000001) {
+			entry = e;
+			break;
+		}
+	}
+	if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) {
+		entry->edx &= ~(1 << 20);
+		printk(KERN_INFO "kvm: guest NX capability removed\n");
+	}
+}
+
+/* when an old userspace process fills a new kernel module */
+static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
+				    struct kvm_cpuid *cpuid,
+				    struct kvm_cpuid_entry __user *entries)
+{
+	int r, i;
+	struct kvm_cpuid_entry *cpuid_entries;
+
+	r = -E2BIG;
+	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
+		goto out;
+	r = -ENOMEM;
+	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent);
+	if (!cpuid_entries)
+		goto out;
+	r = -EFAULT;
+	if (copy_from_user(cpuid_entries, entries,
+			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
+		goto out_free;
+	for (i = 0; i < cpuid->nent; i++) {
+		vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function;
+		vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax;
+		vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx;
+		vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx;
+		vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx;
+		vcpu->arch.cpuid_entries[i].index = 0;
+		vcpu->arch.cpuid_entries[i].flags = 0;
+		vcpu->arch.cpuid_entries[i].padding[0] = 0;
+		vcpu->arch.cpuid_entries[i].padding[1] = 0;
+		vcpu->arch.cpuid_entries[i].padding[2] = 0;
+	}
+	vcpu->arch.cpuid_nent = cpuid->nent;
+	cpuid_fix_nx_cap(vcpu);
+	r = 0;
+
+out_free:
+	vfree(cpuid_entries);
+out:
+	return r;
+}
+
+static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
+{
+	int r;
+
+	r = -E2BIG;
+	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
+		goto out;
+	r = -EFAULT;
+	if (copy_from_user(&vcpu->arch.cpuid_entries, entries,
+			   cpuid->nent * sizeof(struct kvm_cpuid_entry2)))
+		goto out;
+	vcpu->arch.cpuid_nent = cpuid->nent;
+	return 0;
+
+out:
+	return r;
+}
+
+static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
+{
+	int r;
+
+	r = -E2BIG;
+	if (cpuid->nent < vcpu->arch.cpuid_nent)
+		goto out;
+	r = -EFAULT;
+	if (copy_to_user(entries, &vcpu->arch.cpuid_entries,
+			   vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
+		goto out;
+	return 0;
+
+out:
+	cpuid->nent = vcpu->arch.cpuid_nent;
+	return r;
+}
+
+static inline u32 bit(int bitno)
+{
+	return 1 << (bitno & 31);
+}
+
+static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function,
+			  u32 index)
+{
+	entry->function = function;
+	entry->index = index;
+	cpuid_count(entry->function, entry->index,
+		&entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
+	entry->flags = 0;
+}
+
+static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
+			 u32 index, int *nent, int maxnent)
+{
+	const u32 kvm_supported_word0_x86_features = bit(X86_FEATURE_FPU) |
+		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
+		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
+		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
+		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
+		bit(X86_FEATURE_SEP) | bit(X86_FEATURE_PGE) |
+		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
+		bit(X86_FEATURE_CLFLSH) | bit(X86_FEATURE_MMX) |
+		bit(X86_FEATURE_FXSR) | bit(X86_FEATURE_XMM) |
+		bit(X86_FEATURE_XMM2) | bit(X86_FEATURE_SELFSNOOP);
+	const u32 kvm_supported_word1_x86_features = bit(X86_FEATURE_FPU) |
+		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
+		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
+		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
+		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
+		bit(X86_FEATURE_PGE) |
+		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
+		bit(X86_FEATURE_MMX) | bit(X86_FEATURE_FXSR) |
+		bit(X86_FEATURE_SYSCALL) |
+		(bit(X86_FEATURE_NX) && is_efer_nx()) |
+#ifdef CONFIG_X86_64
+		bit(X86_FEATURE_LM) |
+#endif
+		bit(X86_FEATURE_MMXEXT) |
+		bit(X86_FEATURE_3DNOWEXT) |
+		bit(X86_FEATURE_3DNOW);
+	const u32 kvm_supported_word3_x86_features =
+		bit(X86_FEATURE_XMM3) | bit(X86_FEATURE_CX16);
+	const u32 kvm_supported_word6_x86_features =
+		bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY);
+
+	/* all func 2 cpuid_count() should be called on the same cpu */
+	get_cpu();
+	do_cpuid_1_ent(entry, function, index);
+	++*nent;
+
+	switch (function) {
+	case 0:
+		entry->eax = min(entry->eax, (u32)0xb);
+		break;
+	case 1:
+		entry->edx &= kvm_supported_word0_x86_features;
+		entry->ecx &= kvm_supported_word3_x86_features;
+		break;
+	/* function 2 entries are STATEFUL. That is, repeated cpuid commands
+	 * may return different values. This forces us to get_cpu() before
+	 * issuing the first command, and also to emulate this annoying behavior
+	 * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */
+	case 2: {
+		int t, times = entry->eax & 0xff;
+
+		entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
+		for (t = 1; t < times && *nent < maxnent; ++t) {
+			do_cpuid_1_ent(&entry[t], function, 0);
+			entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
+			++*nent;
+		}
+		break;
+	}
+	/* function 4 and 0xb have additional index. */
+	case 4: {
+		int index, cache_type;
+
+		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+		/* read more entries until cache_type is zero */
+		for (index = 1; *nent < maxnent; ++index) {
+			cache_type = entry[index - 1].eax & 0x1f;
+			if (!cache_type)
+				break;
+			do_cpuid_1_ent(&entry[index], function, index);
+			entry[index].flags |=
+			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+			++*nent;
+		}
+		break;
+	}
+	case 0xb: {
+		int index, level_type;
+
+		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+		/* read more entries until level_type is zero */
+		for (index = 1; *nent < maxnent; ++index) {
+			level_type = entry[index - 1].ecx & 0xff;
+			if (!level_type)
+				break;
+			do_cpuid_1_ent(&entry[index], function, index);
+			entry[index].flags |=
+			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
+			++*nent;
+		}
+		break;
+	}
+	case 0x80000000:
+		entry->eax = min(entry->eax, 0x8000001a);
+		break;
+	case 0x80000001:
+		entry->edx &= kvm_supported_word1_x86_features;
+		entry->ecx &= kvm_supported_word6_x86_features;
+		break;
+	}
+	put_cpu();
+}
+
+static int kvm_vm_ioctl_get_supported_cpuid(struct kvm *kvm,
+				    struct kvm_cpuid2 *cpuid,
+				    struct kvm_cpuid_entry2 __user *entries)
+{
+	struct kvm_cpuid_entry2 *cpuid_entries;
+	int limit, nent = 0, r = -E2BIG;
+	u32 func;
+
+	if (cpuid->nent < 1)
+		goto out;
+	r = -ENOMEM;
+	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
+	if (!cpuid_entries)
+		goto out;
+
+	do_cpuid_ent(&cpuid_entries[0], 0, 0, &nent, cpuid->nent);
+	limit = cpuid_entries[0].eax;
+	for (func = 1; func <= limit && nent < cpuid->nent; ++func)
+		do_cpuid_ent(&cpuid_entries[nent], func, 0,
+				&nent, cpuid->nent);
+	r = -E2BIG;
+	if (nent >= cpuid->nent)
+		goto out_free;
+
+	do_cpuid_ent(&cpuid_entries[nent], 0x80000000, 0, &nent, cpuid->nent);
+	limit = cpuid_entries[nent - 1].eax;
+	for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func)
+		do_cpuid_ent(&cpuid_entries[nent], func, 0,
+			       &nent, cpuid->nent);
+	r = -EFAULT;
+	if (copy_to_user(entries, cpuid_entries,
+			nent * sizeof(struct kvm_cpuid_entry2)))
+		goto out_free;
+	cpuid->nent = nent;
+	r = 0;
+
+out_free:
+	vfree(cpuid_entries);
+out:
+	return r;
+}
+
+static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu,
+				    struct kvm_lapic_state *s)
+{
+	vcpu_load(vcpu);
+	memcpy(s->regs, vcpu->arch.apic->regs, sizeof *s);
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+static int kvm_vcpu_ioctl_set_lapic(struct kvm_vcpu *vcpu,
+				    struct kvm_lapic_state *s)
+{
+	vcpu_load(vcpu);
+	memcpy(vcpu->arch.apic->regs, s->regs, sizeof *s);
+	kvm_apic_post_state_restore(vcpu);
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
+				    struct kvm_interrupt *irq)
+{
+	if (irq->irq < 0 || irq->irq >= 256)
+		return -EINVAL;
+	if (irqchip_in_kernel(vcpu->kvm))
+		return -ENXIO;
+	vcpu_load(vcpu);
+
+	set_bit(irq->irq, vcpu->arch.irq_pending);
+	set_bit(irq->irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+long kvm_arch_vcpu_ioctl(struct file *filp,
+			 unsigned int ioctl, unsigned long arg)
+{
+	struct kvm_vcpu *vcpu = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	int r;
+
+	switch (ioctl) {
+	case KVM_GET_LAPIC: {
+		struct kvm_lapic_state lapic;
+
+		memset(&lapic, 0, sizeof lapic);
+		r = kvm_vcpu_ioctl_get_lapic(vcpu, &lapic);
+		if (r)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(argp, &lapic, sizeof lapic))
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_SET_LAPIC: {
+		struct kvm_lapic_state lapic;
+
+		r = -EFAULT;
+		if (copy_from_user(&lapic, argp, sizeof lapic))
+			goto out;
+		r = kvm_vcpu_ioctl_set_lapic(vcpu, &lapic);;
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_INTERRUPT: {
+		struct kvm_interrupt irq;
+
+		r = -EFAULT;
+		if (copy_from_user(&irq, argp, sizeof irq))
+			goto out;
+		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_SET_CPUID: {
+		struct kvm_cpuid __user *cpuid_arg = argp;
+		struct kvm_cpuid cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vcpu_ioctl_set_cpuid(vcpu, &cpuid, cpuid_arg->entries);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_SET_CPUID2: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid,
+				cpuid_arg->entries);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_GET_CPUID2: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid,
+				cpuid_arg->entries);
+		if (r)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_GET_MSRS:
+		r = msr_io(vcpu, argp, kvm_get_msr, 1);
+		break;
+	case KVM_SET_MSRS:
+		r = msr_io(vcpu, argp, do_set_msr, 0);
+		break;
+	default:
+		r = -EINVAL;
+	}
+out:
+	return r;
+}
+
+static int kvm_vm_ioctl_set_tss_addr(struct kvm *kvm, unsigned long addr)
+{
+	int ret;
+
+	if (addr > (unsigned int)(-3 * PAGE_SIZE))
+		return -1;
+	ret = kvm_x86_ops->set_tss_addr(kvm, addr);
+	return ret;
+}
+
+static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
+					  u32 kvm_nr_mmu_pages)
+{
+	if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+
+	kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages);
+	kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages;
+
+	mutex_unlock(&kvm->lock);
+	return 0;
+}
+
+static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm)
+{
+	return kvm->arch.n_alloc_mmu_pages;
+}
+
+gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
+{
+	int i;
+	struct kvm_mem_alias *alias;
+
+	for (i = 0; i < kvm->arch.naliases; ++i) {
+		alias = &kvm->arch.aliases[i];
+		if (gfn >= alias->base_gfn
+		    && gfn < alias->base_gfn + alias->npages)
+			return alias->target_gfn + gfn - alias->base_gfn;
+	}
+	return gfn;
+}
+
+/*
+ * Set a new alias region.  Aliases map a portion of physical memory into
+ * another portion.  This is useful for memory windows, for example the PC
+ * VGA region.
+ */
+static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
+					 struct kvm_memory_alias *alias)
+{
+	int r, n;
+	struct kvm_mem_alias *p;
+
+	r = -EINVAL;
+	/* General sanity checks */
+	if (alias->memory_size & (PAGE_SIZE - 1))
+		goto out;
+	if (alias->guest_phys_addr & (PAGE_SIZE - 1))
+		goto out;
+	if (alias->slot >= KVM_ALIAS_SLOTS)
+		goto out;
+	if (alias->guest_phys_addr + alias->memory_size
+	    < alias->guest_phys_addr)
+		goto out;
+	if (alias->target_phys_addr + alias->memory_size
+	    < alias->target_phys_addr)
+		goto out;
+
+	mutex_lock(&kvm->lock);
+
+	p = &kvm->arch.aliases[alias->slot];
+	p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
+	p->npages = alias->memory_size >> PAGE_SHIFT;
+	p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT;
+
+	for (n = KVM_ALIAS_SLOTS; n > 0; --n)
+		if (kvm->arch.aliases[n - 1].npages)
+			break;
+	kvm->arch.naliases = n;
+
+	kvm_mmu_zap_all(kvm);
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+
+out:
+	return r;
+}
+
+static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
+{
+	int r;
+
+	r = 0;
+	switch (chip->chip_id) {
+	case KVM_IRQCHIP_PIC_MASTER:
+		memcpy(&chip->chip.pic,
+			&pic_irqchip(kvm)->pics[0],
+			sizeof(struct kvm_pic_state));
+		break;
+	case KVM_IRQCHIP_PIC_SLAVE:
+		memcpy(&chip->chip.pic,
+			&pic_irqchip(kvm)->pics[1],
+			sizeof(struct kvm_pic_state));
+		break;
+	case KVM_IRQCHIP_IOAPIC:
+		memcpy(&chip->chip.ioapic,
+			ioapic_irqchip(kvm),
+			sizeof(struct kvm_ioapic_state));
+		break;
+	default:
+		r = -EINVAL;
+		break;
+	}
+	return r;
+}
+
+static int kvm_vm_ioctl_set_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
+{
+	int r;
+
+	r = 0;
+	switch (chip->chip_id) {
+	case KVM_IRQCHIP_PIC_MASTER:
+		memcpy(&pic_irqchip(kvm)->pics[0],
+			&chip->chip.pic,
+			sizeof(struct kvm_pic_state));
+		break;
+	case KVM_IRQCHIP_PIC_SLAVE:
+		memcpy(&pic_irqchip(kvm)->pics[1],
+			&chip->chip.pic,
+			sizeof(struct kvm_pic_state));
+		break;
+	case KVM_IRQCHIP_IOAPIC:
+		memcpy(ioapic_irqchip(kvm),
+			&chip->chip.ioapic,
+			sizeof(struct kvm_ioapic_state));
+		break;
+	default:
+		r = -EINVAL;
+		break;
+	}
+	kvm_pic_update_irq(pic_irqchip(kvm));
+	return r;
+}
+
+/*
+ * Get (and clear) the dirty memory log for a memory slot.
+ */
+int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
+				      struct kvm_dirty_log *log)
+{
+	int r;
+	int n;
+	struct kvm_memory_slot *memslot;
+	int is_dirty = 0;
+
+	mutex_lock(&kvm->lock);
+
+	r = kvm_get_dirty_log(kvm, log, &is_dirty);
+	if (r)
+		goto out;
+
+	/* If nothing is dirty, don't bother messing with page tables. */
+	if (is_dirty) {
+		kvm_mmu_slot_remove_write_access(kvm, log->slot);
+		kvm_flush_remote_tlbs(kvm);
+		memslot = &kvm->memslots[log->slot];
+		n = ALIGN(memslot->npages, BITS_PER_LONG) / 8;
+		memset(memslot->dirty_bitmap, 0, n);
+	}
+	r = 0;
+out:
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+
+long kvm_arch_vm_ioctl(struct file *filp,
+		       unsigned int ioctl, unsigned long arg)
+{
+	struct kvm *kvm = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	int r = -EINVAL;
+
+	switch (ioctl) {
+	case KVM_SET_TSS_ADDR:
+		r = kvm_vm_ioctl_set_tss_addr(kvm, arg);
+		if (r < 0)
+			goto out;
+		break;
+	case KVM_SET_MEMORY_REGION: {
+		struct kvm_memory_region kvm_mem;
+		struct kvm_userspace_memory_region kvm_userspace_mem;
+
+		r = -EFAULT;
+		if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
+			goto out;
+		kvm_userspace_mem.slot = kvm_mem.slot;
+		kvm_userspace_mem.flags = kvm_mem.flags;
+		kvm_userspace_mem.guest_phys_addr = kvm_mem.guest_phys_addr;
+		kvm_userspace_mem.memory_size = kvm_mem.memory_size;
+		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 0);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_SET_NR_MMU_PAGES:
+		r = kvm_vm_ioctl_set_nr_mmu_pages(kvm, arg);
+		if (r)
+			goto out;
+		break;
+	case KVM_GET_NR_MMU_PAGES:
+		r = kvm_vm_ioctl_get_nr_mmu_pages(kvm);
+		break;
+	case KVM_SET_MEMORY_ALIAS: {
+		struct kvm_memory_alias alias;
+
+		r = -EFAULT;
+		if (copy_from_user(&alias, argp, sizeof alias))
+			goto out;
+		r = kvm_vm_ioctl_set_memory_alias(kvm, &alias);
+		if (r)
+			goto out;
+		break;
+	}
+	case KVM_CREATE_IRQCHIP:
+		r = -ENOMEM;
+		kvm->arch.vpic = kvm_create_pic(kvm);
+		if (kvm->arch.vpic) {
+			r = kvm_ioapic_init(kvm);
+			if (r) {
+				kfree(kvm->arch.vpic);
+				kvm->arch.vpic = NULL;
+				goto out;
+			}
+		} else
+			goto out;
+		break;
+	case KVM_IRQ_LINE: {
+		struct kvm_irq_level irq_event;
+
+		r = -EFAULT;
+		if (copy_from_user(&irq_event, argp, sizeof irq_event))
+			goto out;
+		if (irqchip_in_kernel(kvm)) {
+			mutex_lock(&kvm->lock);
+			if (irq_event.irq < 16)
+				kvm_pic_set_irq(pic_irqchip(kvm),
+					irq_event.irq,
+					irq_event.level);
+			kvm_ioapic_set_irq(kvm->arch.vioapic,
+					irq_event.irq,
+					irq_event.level);
+			mutex_unlock(&kvm->lock);
+			r = 0;
+		}
+		break;
+	}
+	case KVM_GET_IRQCHIP: {
+		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
+		struct kvm_irqchip chip;
+
+		r = -EFAULT;
+		if (copy_from_user(&chip, argp, sizeof chip))
+			goto out;
+		r = -ENXIO;
+		if (!irqchip_in_kernel(kvm))
+			goto out;
+		r = kvm_vm_ioctl_get_irqchip(kvm, &chip);
+		if (r)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(argp, &chip, sizeof chip))
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_SET_IRQCHIP: {
+		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
+		struct kvm_irqchip chip;
+
+		r = -EFAULT;
+		if (copy_from_user(&chip, argp, sizeof chip))
+			goto out;
+		r = -ENXIO;
+		if (!irqchip_in_kernel(kvm))
+			goto out;
+		r = kvm_vm_ioctl_set_irqchip(kvm, &chip);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_GET_SUPPORTED_CPUID: {
+		struct kvm_cpuid2 __user *cpuid_arg = argp;
+		struct kvm_cpuid2 cpuid;
+
+		r = -EFAULT;
+		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
+			goto out;
+		r = kvm_vm_ioctl_get_supported_cpuid(kvm, &cpuid,
+			cpuid_arg->entries);
+		if (r)
+			goto out;
+
+		r = -EFAULT;
+		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
+			goto out;
+		r = 0;
+		break;
+	}
+	default:
+		;
+	}
+out:
+	return r;
+}
+
+static void kvm_init_msr_list(void)
+{
+	u32 dummy[2];
+	unsigned i, j;
+
+	for (i = j = 0; i < ARRAY_SIZE(msrs_to_save); i++) {
+		if (rdmsr_safe(msrs_to_save[i], &dummy[0], &dummy[1]) < 0)
+			continue;
+		if (j < i)
+			msrs_to_save[j] = msrs_to_save[i];
+		j++;
+	}
+	num_msrs_to_save = j;
+}
+
+/*
+ * Only apic need an MMIO device hook, so shortcut now..
+ */
+static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu,
+						gpa_t addr)
+{
+	struct kvm_io_device *dev;
+
+	if (vcpu->arch.apic) {
+		dev = &vcpu->arch.apic->dev;
+		if (dev->in_range(dev, addr))
+			return dev;
+	}
+	return NULL;
+}
+
+
+static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu,
+						gpa_t addr)
+{
+	struct kvm_io_device *dev;
+
+	dev = vcpu_find_pervcpu_dev(vcpu, addr);
+	if (dev == NULL)
+		dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr);
+	return dev;
+}
+
+int emulator_read_std(unsigned long addr,
+			     void *val,
+			     unsigned int bytes,
+			     struct kvm_vcpu *vcpu)
+{
+	void *data = val;
+
+	while (bytes) {
+		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+		unsigned offset = addr & (PAGE_SIZE-1);
+		unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
+		int ret;
+
+		if (gpa == UNMAPPED_GVA)
+			return X86EMUL_PROPAGATE_FAULT;
+		ret = kvm_read_guest(vcpu->kvm, gpa, data, tocopy);
+		if (ret < 0)
+			return X86EMUL_UNHANDLEABLE;
+
+		bytes -= tocopy;
+		data += tocopy;
+		addr += tocopy;
+	}
+
+	return X86EMUL_CONTINUE;
+}
+EXPORT_SYMBOL_GPL(emulator_read_std);
+
+static int emulator_read_emulated(unsigned long addr,
+				  void *val,
+				  unsigned int bytes,
+				  struct kvm_vcpu *vcpu)
+{
+	struct kvm_io_device *mmio_dev;
+	gpa_t                 gpa;
+
+	if (vcpu->mmio_read_completed) {
+		memcpy(val, vcpu->mmio_data, bytes);
+		vcpu->mmio_read_completed = 0;
+		return X86EMUL_CONTINUE;
+	}
+
+	gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+
+	/* For APIC access vmexit */
+	if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
+		goto mmio;
+
+	if (emulator_read_std(addr, val, bytes, vcpu)
+			== X86EMUL_CONTINUE)
+		return X86EMUL_CONTINUE;
+	if (gpa == UNMAPPED_GVA)
+		return X86EMUL_PROPAGATE_FAULT;
+
+mmio:
+	/*
+	 * Is this MMIO handled locally?
+	 */
+	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
+	if (mmio_dev) {
+		kvm_iodevice_read(mmio_dev, gpa, bytes, val);
+		return X86EMUL_CONTINUE;
+	}
+
+	vcpu->mmio_needed = 1;
+	vcpu->mmio_phys_addr = gpa;
+	vcpu->mmio_size = bytes;
+	vcpu->mmio_is_write = 0;
+
+	return X86EMUL_UNHANDLEABLE;
+}
+
+static int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
+			       const void *val, int bytes)
+{
+	int ret;
+
+	ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes);
+	if (ret < 0)
+		return 0;
+	kvm_mmu_pte_write(vcpu, gpa, val, bytes);
+	return 1;
+}
+
+static int emulator_write_emulated_onepage(unsigned long addr,
+					   const void *val,
+					   unsigned int bytes,
+					   struct kvm_vcpu *vcpu)
+{
+	struct kvm_io_device *mmio_dev;
+	gpa_t                 gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+
+	if (gpa == UNMAPPED_GVA) {
+		kvm_inject_page_fault(vcpu, addr, 2);
+		return X86EMUL_PROPAGATE_FAULT;
+	}
+
+	/* For APIC access vmexit */
+	if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
+		goto mmio;
+
+	if (emulator_write_phys(vcpu, gpa, val, bytes))
+		return X86EMUL_CONTINUE;
+
+mmio:
+	/*
+	 * Is this MMIO handled locally?
+	 */
+	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
+	if (mmio_dev) {
+		kvm_iodevice_write(mmio_dev, gpa, bytes, val);
+		return X86EMUL_CONTINUE;
+	}
+
+	vcpu->mmio_needed = 1;
+	vcpu->mmio_phys_addr = gpa;
+	vcpu->mmio_size = bytes;
+	vcpu->mmio_is_write = 1;
+	memcpy(vcpu->mmio_data, val, bytes);
+
+	return X86EMUL_CONTINUE;
+}
+
+int emulator_write_emulated(unsigned long addr,
+				   const void *val,
+				   unsigned int bytes,
+				   struct kvm_vcpu *vcpu)
+{
+	/* Crossing a page boundary? */
+	if (((addr + bytes - 1) ^ addr) & PAGE_MASK) {
+		int rc, now;
+
+		now = -addr & ~PAGE_MASK;
+		rc = emulator_write_emulated_onepage(addr, val, now, vcpu);
+		if (rc != X86EMUL_CONTINUE)
+			return rc;
+		addr += now;
+		val += now;
+		bytes -= now;
+	}
+	return emulator_write_emulated_onepage(addr, val, bytes, vcpu);
+}
+EXPORT_SYMBOL_GPL(emulator_write_emulated);
+
+static int emulator_cmpxchg_emulated(unsigned long addr,
+				     const void *old,
+				     const void *new,
+				     unsigned int bytes,
+				     struct kvm_vcpu *vcpu)
+{
+	static int reported;
+
+	if (!reported) {
+		reported = 1;
+		printk(KERN_WARNING "kvm: emulating exchange as write\n");
+	}
+#ifndef CONFIG_X86_64
+	/* guests cmpxchg8b have to be emulated atomically */
+	if (bytes == 8) {
+		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+		struct page *page;
+		char *addr;
+		u64 val;
+
+		if (gpa == UNMAPPED_GVA ||
+		   (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
+			goto emul_write;
+
+		if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK))
+			goto emul_write;
+
+		val = *(u64 *)new;
+		page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+		addr = kmap_atomic(page, KM_USER0);
+		set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
+		kunmap_atomic(addr, KM_USER0);
+		kvm_release_page_dirty(page);
+	}
+emul_write:
+#endif
+
+	return emulator_write_emulated(addr, new, bytes, vcpu);
+}
+
+static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg)
+{
+	return kvm_x86_ops->get_segment_base(vcpu, seg);
+}
+
+int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address)
+{
+	return X86EMUL_CONTINUE;
+}
+
+int emulate_clts(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->set_cr0(vcpu, vcpu->arch.cr0 & ~X86_CR0_TS);
+	return X86EMUL_CONTINUE;
+}
+
+int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long *dest)
+{
+	struct kvm_vcpu *vcpu = ctxt->vcpu;
+
+	switch (dr) {
+	case 0 ... 3:
+		*dest = kvm_x86_ops->get_dr(vcpu, dr);
+		return X86EMUL_CONTINUE;
+	default:
+		pr_unimpl(vcpu, "%s: unexpected dr %u\n", __FUNCTION__, dr);
+		return X86EMUL_UNHANDLEABLE;
+	}
+}
+
+int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value)
+{
+	unsigned long mask = (ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U;
+	int exception;
+
+	kvm_x86_ops->set_dr(ctxt->vcpu, dr, value & mask, &exception);
+	if (exception) {
+		/* FIXME: better handling */
+		return X86EMUL_UNHANDLEABLE;
+	}
+	return X86EMUL_CONTINUE;
+}
+
+void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
+{
+	static int reported;
+	u8 opcodes[4];
+	unsigned long rip = vcpu->arch.rip;
+	unsigned long rip_linear;
+
+	rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS);
+
+	if (reported)
+		return;
+
+	emulator_read_std(rip_linear, (void *)opcodes, 4, vcpu);
+
+	printk(KERN_ERR "emulation failed (%s) rip %lx %02x %02x %02x %02x\n",
+	       context, rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]);
+	reported = 1;
+}
+EXPORT_SYMBOL_GPL(kvm_report_emulation_failure);
+
+struct x86_emulate_ops emulate_ops = {
+	.read_std            = emulator_read_std,
+	.read_emulated       = emulator_read_emulated,
+	.write_emulated      = emulator_write_emulated,
+	.cmpxchg_emulated    = emulator_cmpxchg_emulated,
+};
+
+int emulate_instruction(struct kvm_vcpu *vcpu,
+			struct kvm_run *run,
+			unsigned long cr2,
+			u16 error_code,
+			int no_decode)
+{
+	int r;
+
+	vcpu->arch.mmio_fault_cr2 = cr2;
+	kvm_x86_ops->cache_regs(vcpu);
+
+	vcpu->mmio_is_write = 0;
+	vcpu->arch.pio.string = 0;
+
+	if (!no_decode) {
+		int cs_db, cs_l;
+		kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
+
+		vcpu->arch.emulate_ctxt.vcpu = vcpu;
+		vcpu->arch.emulate_ctxt.eflags = kvm_x86_ops->get_rflags(vcpu);
+		vcpu->arch.emulate_ctxt.mode =
+			(vcpu->arch.emulate_ctxt.eflags & X86_EFLAGS_VM)
+			? X86EMUL_MODE_REAL : cs_l
+			? X86EMUL_MODE_PROT64 :	cs_db
+			? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;
+
+		if (vcpu->arch.emulate_ctxt.mode == X86EMUL_MODE_PROT64) {
+			vcpu->arch.emulate_ctxt.cs_base = 0;
+			vcpu->arch.emulate_ctxt.ds_base = 0;
+			vcpu->arch.emulate_ctxt.es_base = 0;
+			vcpu->arch.emulate_ctxt.ss_base = 0;
+		} else {
+			vcpu->arch.emulate_ctxt.cs_base =
+					get_segment_base(vcpu, VCPU_SREG_CS);
+			vcpu->arch.emulate_ctxt.ds_base =
+					get_segment_base(vcpu, VCPU_SREG_DS);
+			vcpu->arch.emulate_ctxt.es_base =
+					get_segment_base(vcpu, VCPU_SREG_ES);
+			vcpu->arch.emulate_ctxt.ss_base =
+					get_segment_base(vcpu, VCPU_SREG_SS);
+		}
+
+		vcpu->arch.emulate_ctxt.gs_base =
+					get_segment_base(vcpu, VCPU_SREG_GS);
+		vcpu->arch.emulate_ctxt.fs_base =
+					get_segment_base(vcpu, VCPU_SREG_FS);
+
+		r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
+		++vcpu->stat.insn_emulation;
+		if (r)  {
+			++vcpu->stat.insn_emulation_fail;
+			if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
+				return EMULATE_DONE;
+			return EMULATE_FAIL;
+		}
+	}
+
+	r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
+
+	if (vcpu->arch.pio.string)
+		return EMULATE_DO_MMIO;
+
+	if ((r || vcpu->mmio_is_write) && run) {
+		run->exit_reason = KVM_EXIT_MMIO;
+		run->mmio.phys_addr = vcpu->mmio_phys_addr;
+		memcpy(run->mmio.data, vcpu->mmio_data, 8);
+		run->mmio.len = vcpu->mmio_size;
+		run->mmio.is_write = vcpu->mmio_is_write;
+	}
+
+	if (r) {
+		if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
+			return EMULATE_DONE;
+		if (!vcpu->mmio_needed) {
+			kvm_report_emulation_failure(vcpu, "mmio");
+			return EMULATE_FAIL;
+		}
+		return EMULATE_DO_MMIO;
+	}
+
+	kvm_x86_ops->decache_regs(vcpu);
+	kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags);
+
+	if (vcpu->mmio_is_write) {
+		vcpu->mmio_needed = 0;
+		return EMULATE_DO_MMIO;
+	}
+
+	return EMULATE_DONE;
+}
+EXPORT_SYMBOL_GPL(emulate_instruction);
+
+static void free_pio_guest_pages(struct kvm_vcpu *vcpu)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(vcpu->arch.pio.guest_pages); ++i)
+		if (vcpu->arch.pio.guest_pages[i]) {
+			kvm_release_page_dirty(vcpu->arch.pio.guest_pages[i]);
+			vcpu->arch.pio.guest_pages[i] = NULL;
+		}
+}
+
+static int pio_copy_data(struct kvm_vcpu *vcpu)
+{
+	void *p = vcpu->arch.pio_data;
+	void *q;
+	unsigned bytes;
+	int nr_pages = vcpu->arch.pio.guest_pages[1] ? 2 : 1;
+
+	q = vmap(vcpu->arch.pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
+		 PAGE_KERNEL);
+	if (!q) {
+		free_pio_guest_pages(vcpu);
+		return -ENOMEM;
+	}
+	q += vcpu->arch.pio.guest_page_offset;
+	bytes = vcpu->arch.pio.size * vcpu->arch.pio.cur_count;
+	if (vcpu->arch.pio.in)
+		memcpy(q, p, bytes);
+	else
+		memcpy(p, q, bytes);
+	q -= vcpu->arch.pio.guest_page_offset;
+	vunmap(q);
+	free_pio_guest_pages(vcpu);
+	return 0;
+}
+
+int complete_pio(struct kvm_vcpu *vcpu)
+{
+	struct kvm_pio_request *io = &vcpu->arch.pio;
+	long delta;
+	int r;
+
+	kvm_x86_ops->cache_regs(vcpu);
+
+	if (!io->string) {
+		if (io->in)
+			memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data,
+			       io->size);
+	} else {
+		if (io->in) {
+			r = pio_copy_data(vcpu);
+			if (r) {
+				kvm_x86_ops->cache_regs(vcpu);
+				return r;
+			}
+		}
+
+		delta = 1;
+		if (io->rep) {
+			delta *= io->cur_count;
+			/*
+			 * The size of the register should really depend on
+			 * current address size.
+			 */
+			vcpu->arch.regs[VCPU_REGS_RCX] -= delta;
+		}
+		if (io->down)
+			delta = -delta;
+		delta *= io->size;
+		if (io->in)
+			vcpu->arch.regs[VCPU_REGS_RDI] += delta;
+		else
+			vcpu->arch.regs[VCPU_REGS_RSI] += delta;
+	}
+
+	kvm_x86_ops->decache_regs(vcpu);
+
+	io->count -= io->cur_count;
+	io->cur_count = 0;
+
+	return 0;
+}
+
+static void kernel_pio(struct kvm_io_device *pio_dev,
+		       struct kvm_vcpu *vcpu,
+		       void *pd)
+{
+	/* TODO: String I/O for in kernel device */
+
+	mutex_lock(&vcpu->kvm->lock);
+	if (vcpu->arch.pio.in)
+		kvm_iodevice_read(pio_dev, vcpu->arch.pio.port,
+				  vcpu->arch.pio.size,
+				  pd);
+	else
+		kvm_iodevice_write(pio_dev, vcpu->arch.pio.port,
+				   vcpu->arch.pio.size,
+				   pd);
+	mutex_unlock(&vcpu->kvm->lock);
+}
+
+static void pio_string_write(struct kvm_io_device *pio_dev,
+			     struct kvm_vcpu *vcpu)
+{
+	struct kvm_pio_request *io = &vcpu->arch.pio;
+	void *pd = vcpu->arch.pio_data;
+	int i;
+
+	mutex_lock(&vcpu->kvm->lock);
+	for (i = 0; i < io->cur_count; i++) {
+		kvm_iodevice_write(pio_dev, io->port,
+				   io->size,
+				   pd);
+		pd += io->size;
+	}
+	mutex_unlock(&vcpu->kvm->lock);
+}
+
+static struct kvm_io_device *vcpu_find_pio_dev(struct kvm_vcpu *vcpu,
+					       gpa_t addr)
+{
+	return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr);
+}
+
+int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
+		  int size, unsigned port)
+{
+	struct kvm_io_device *pio_dev;
+
+	vcpu->run->exit_reason = KVM_EXIT_IO;
+	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
+	vcpu->run->io.size = vcpu->arch.pio.size = size;
+	vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
+	vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = 1;
+	vcpu->run->io.port = vcpu->arch.pio.port = port;
+	vcpu->arch.pio.in = in;
+	vcpu->arch.pio.string = 0;
+	vcpu->arch.pio.down = 0;
+	vcpu->arch.pio.guest_page_offset = 0;
+	vcpu->arch.pio.rep = 0;
+
+	kvm_x86_ops->cache_regs(vcpu);
+	memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4);
+	kvm_x86_ops->decache_regs(vcpu);
+
+	kvm_x86_ops->skip_emulated_instruction(vcpu);
+
+	pio_dev = vcpu_find_pio_dev(vcpu, port);
+	if (pio_dev) {
+		kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data);
+		complete_pio(vcpu);
+		return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_pio);
+
+int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
+		  int size, unsigned long count, int down,
+		  gva_t address, int rep, unsigned port)
+{
+	unsigned now, in_page;
+	int i, ret = 0;
+	int nr_pages = 1;
+	struct page *page;
+	struct kvm_io_device *pio_dev;
+
+	vcpu->run->exit_reason = KVM_EXIT_IO;
+	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
+	vcpu->run->io.size = vcpu->arch.pio.size = size;
+	vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
+	vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = count;
+	vcpu->run->io.port = vcpu->arch.pio.port = port;
+	vcpu->arch.pio.in = in;
+	vcpu->arch.pio.string = 1;
+	vcpu->arch.pio.down = down;
+	vcpu->arch.pio.guest_page_offset = offset_in_page(address);
+	vcpu->arch.pio.rep = rep;
+
+	if (!count) {
+		kvm_x86_ops->skip_emulated_instruction(vcpu);
+		return 1;
+	}
+
+	if (!down)
+		in_page = PAGE_SIZE - offset_in_page(address);
+	else
+		in_page = offset_in_page(address) + size;
+	now = min(count, (unsigned long)in_page / size);
+	if (!now) {
+		/*
+		 * String I/O straddles page boundary.  Pin two guest pages
+		 * so that we satisfy atomicity constraints.  Do just one
+		 * transaction to avoid complexity.
+		 */
+		nr_pages = 2;
+		now = 1;
+	}
+	if (down) {
+		/*
+		 * String I/O in reverse.  Yuck.  Kill the guest, fix later.
+		 */
+		pr_unimpl(vcpu, "guest string pio down\n");
+		kvm_inject_gp(vcpu, 0);
+		return 1;
+	}
+	vcpu->run->io.count = now;
+	vcpu->arch.pio.cur_count = now;
+
+	if (vcpu->arch.pio.cur_count == vcpu->arch.pio.count)
+		kvm_x86_ops->skip_emulated_instruction(vcpu);
+
+	for (i = 0; i < nr_pages; ++i) {
+		mutex_lock(&vcpu->kvm->lock);
+		page = gva_to_page(vcpu, address + i * PAGE_SIZE);
+		vcpu->arch.pio.guest_pages[i] = page;
+		mutex_unlock(&vcpu->kvm->lock);
+		if (!page) {
+			kvm_inject_gp(vcpu, 0);
+			free_pio_guest_pages(vcpu);
+			return 1;
+		}
+	}
+
+	pio_dev = vcpu_find_pio_dev(vcpu, port);
+	if (!vcpu->arch.pio.in) {
+		/* string PIO write */
+		ret = pio_copy_data(vcpu);
+		if (ret >= 0 && pio_dev) {
+			pio_string_write(pio_dev, vcpu);
+			complete_pio(vcpu);
+			if (vcpu->arch.pio.count == 0)
+				ret = 1;
+		}
+	} else if (pio_dev)
+		pr_unimpl(vcpu, "no string pio read support yet, "
+		       "port %x size %d count %ld\n",
+			port, size, count);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_pio_string);
+
+int kvm_arch_init(void *opaque)
+{
+	int r;
+	struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque;
+
+	r = kvm_mmu_module_init();
+	if (r)
+		goto out_fail;
+
+	kvm_init_msr_list();
+
+	if (kvm_x86_ops) {
+		printk(KERN_ERR "kvm: already loaded the other module\n");
+		r = -EEXIST;
+		goto out;
+	}
+
+	if (!ops->cpu_has_kvm_support()) {
+		printk(KERN_ERR "kvm: no hardware support\n");
+		r = -EOPNOTSUPP;
+		goto out;
+	}
+	if (ops->disabled_by_bios()) {
+		printk(KERN_ERR "kvm: disabled by bios\n");
+		r = -EOPNOTSUPP;
+		goto out;
+	}
+
+	kvm_x86_ops = ops;
+	kvm_mmu_set_nonpresent_ptes(0ull, 0ull);
+	return 0;
+
+out:
+	kvm_mmu_module_exit();
+out_fail:
+	return r;
+}
+
+void kvm_arch_exit(void)
+{
+	kvm_x86_ops = NULL;
+	kvm_mmu_module_exit();
+}
+
+int kvm_emulate_halt(struct kvm_vcpu *vcpu)
+{
+	++vcpu->stat.halt_exits;
+	if (irqchip_in_kernel(vcpu->kvm)) {
+		vcpu->arch.mp_state = VCPU_MP_STATE_HALTED;
+		kvm_vcpu_block(vcpu);
+		if (vcpu->arch.mp_state != VCPU_MP_STATE_RUNNABLE)
+			return -EINTR;
+		return 1;
+	} else {
+		vcpu->run->exit_reason = KVM_EXIT_HLT;
+		return 0;
+	}
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_halt);
+
+int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
+{
+	unsigned long nr, a0, a1, a2, a3, ret;
+
+	kvm_x86_ops->cache_regs(vcpu);
+
+	nr = vcpu->arch.regs[VCPU_REGS_RAX];
+	a0 = vcpu->arch.regs[VCPU_REGS_RBX];
+	a1 = vcpu->arch.regs[VCPU_REGS_RCX];
+	a2 = vcpu->arch.regs[VCPU_REGS_RDX];
+	a3 = vcpu->arch.regs[VCPU_REGS_RSI];
+
+	if (!is_long_mode(vcpu)) {
+		nr &= 0xFFFFFFFF;
+		a0 &= 0xFFFFFFFF;
+		a1 &= 0xFFFFFFFF;
+		a2 &= 0xFFFFFFFF;
+		a3 &= 0xFFFFFFFF;
+	}
+
+	switch (nr) {
+	default:
+		ret = -KVM_ENOSYS;
+		break;
+	}
+	vcpu->arch.regs[VCPU_REGS_RAX] = ret;
+	kvm_x86_ops->decache_regs(vcpu);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_hypercall);
+
+int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
+{
+	char instruction[3];
+	int ret = 0;
+
+	mutex_lock(&vcpu->kvm->lock);
+
+	/*
+	 * Blow out the MMU to ensure that no other VCPU has an active mapping
+	 * to ensure that the updated hypercall appears atomically across all
+	 * VCPUs.
+	 */
+	kvm_mmu_zap_all(vcpu->kvm);
+
+	kvm_x86_ops->cache_regs(vcpu);
+	kvm_x86_ops->patch_hypercall(vcpu, instruction);
+	if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu)
+	    != X86EMUL_CONTINUE)
+		ret = -EFAULT;
+
+	mutex_unlock(&vcpu->kvm->lock);
+
+	return ret;
+}
+
+static u64 mk_cr_64(u64 curr_cr, u32 new_val)
+{
+	return (curr_cr & ~((1ULL << 32) - 1)) | new_val;
+}
+
+void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
+{
+	struct descriptor_table dt = { limit, base };
+
+	kvm_x86_ops->set_gdt(vcpu, &dt);
+}
+
+void realmode_lidt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
+{
+	struct descriptor_table dt = { limit, base };
+
+	kvm_x86_ops->set_idt(vcpu, &dt);
+}
+
+void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw,
+		   unsigned long *rflags)
+{
+	lmsw(vcpu, msw);
+	*rflags = kvm_x86_ops->get_rflags(vcpu);
+}
+
+unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr)
+{
+	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
+	switch (cr) {
+	case 0:
+		return vcpu->arch.cr0;
+	case 2:
+		return vcpu->arch.cr2;
+	case 3:
+		return vcpu->arch.cr3;
+	case 4:
+		return vcpu->arch.cr4;
+	case 8:
+		return get_cr8(vcpu);
+	default:
+		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
+		return 0;
+	}
+}
+
+void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
+		     unsigned long *rflags)
+{
+	switch (cr) {
+	case 0:
+		set_cr0(vcpu, mk_cr_64(vcpu->arch.cr0, val));
+		*rflags = kvm_x86_ops->get_rflags(vcpu);
+		break;
+	case 2:
+		vcpu->arch.cr2 = val;
+		break;
+	case 3:
+		set_cr3(vcpu, val);
+		break;
+	case 4:
+		set_cr4(vcpu, mk_cr_64(vcpu->arch.cr4, val));
+		break;
+	case 8:
+		set_cr8(vcpu, val & 0xfUL);
+		break;
+	default:
+		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
+	}
+}
+
+static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i)
+{
+	struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i];
+	int j, nent = vcpu->arch.cpuid_nent;
+
+	e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
+	/* when no next entry is found, the current entry[i] is reselected */
+	for (j = i + 1; j == i; j = (j + 1) % nent) {
+		struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j];
+		if (ej->function == e->function) {
+			ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
+			return j;
+		}
+	}
+	return 0; /* silence gcc, even though control never reaches here */
+}
+
+/* find an entry with matching function, matching index (if needed), and that
+ * should be read next (if it's stateful) */
+static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e,
+	u32 function, u32 index)
+{
+	if (e->function != function)
+		return 0;
+	if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index)
+		return 0;
+	if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) &&
+		!(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
+		return 0;
+	return 1;
+}
+
+void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
+{
+	int i;
+	u32 function, index;
+	struct kvm_cpuid_entry2 *e, *best;
+
+	kvm_x86_ops->cache_regs(vcpu);
+	function = vcpu->arch.regs[VCPU_REGS_RAX];
+	index = vcpu->arch.regs[VCPU_REGS_RCX];
+	vcpu->arch.regs[VCPU_REGS_RAX] = 0;
+	vcpu->arch.regs[VCPU_REGS_RBX] = 0;
+	vcpu->arch.regs[VCPU_REGS_RCX] = 0;
+	vcpu->arch.regs[VCPU_REGS_RDX] = 0;
+	best = NULL;
+	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
+		e = &vcpu->arch.cpuid_entries[i];
+		if (is_matching_cpuid_entry(e, function, index)) {
+			if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
+				move_to_next_stateful_cpuid_entry(vcpu, i);
+			best = e;
+			break;
+		}
+		/*
+		 * Both basic or both extended?
+		 */
+		if (((e->function ^ function) & 0x80000000) == 0)
+			if (!best || e->function > best->function)
+				best = e;
+	}
+	if (best) {
+		vcpu->arch.regs[VCPU_REGS_RAX] = best->eax;
+		vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx;
+		vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx;
+		vcpu->arch.regs[VCPU_REGS_RDX] = best->edx;
+	}
+	kvm_x86_ops->decache_regs(vcpu);
+	kvm_x86_ops->skip_emulated_instruction(vcpu);
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);
+
+/*
+ * Check if userspace requested an interrupt window, and that the
+ * interrupt window is open.
+ *
+ * No need to exit to userspace if we already have an interrupt queued.
+ */
+static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu,
+					  struct kvm_run *kvm_run)
+{
+	return (!vcpu->arch.irq_summary &&
+		kvm_run->request_interrupt_window &&
+		vcpu->arch.interrupt_window_open &&
+		(kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF));
+}
+
+static void post_kvm_run_save(struct kvm_vcpu *vcpu,
+			      struct kvm_run *kvm_run)
+{
+	kvm_run->if_flag = (kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
+	kvm_run->cr8 = get_cr8(vcpu);
+	kvm_run->apic_base = kvm_get_apic_base(vcpu);
+	if (irqchip_in_kernel(vcpu->kvm))
+		kvm_run->ready_for_interrupt_injection = 1;
+	else
+		kvm_run->ready_for_interrupt_injection =
+					(vcpu->arch.interrupt_window_open &&
+					 vcpu->arch.irq_summary == 0);
+}
+
+static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	int r;
+
+	if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED)) {
+		pr_debug("vcpu %d received sipi with vector # %x\n",
+		       vcpu->vcpu_id, vcpu->arch.sipi_vector);
+		kvm_lapic_reset(vcpu);
+		r = kvm_x86_ops->vcpu_reset(vcpu);
+		if (r)
+			return r;
+		vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
+	}
+
+preempted:
+	if (vcpu->guest_debug.enabled)
+		kvm_x86_ops->guest_debug_pre(vcpu);
+
+again:
+	r = kvm_mmu_reload(vcpu);
+	if (unlikely(r))
+		goto out;
+
+	kvm_inject_pending_timer_irqs(vcpu);
+
+	preempt_disable();
+
+	kvm_x86_ops->prepare_guest_switch(vcpu);
+	kvm_load_guest_fpu(vcpu);
+
+	local_irq_disable();
+
+	if (signal_pending(current)) {
+		local_irq_enable();
+		preempt_enable();
+		r = -EINTR;
+		kvm_run->exit_reason = KVM_EXIT_INTR;
+		++vcpu->stat.signal_exits;
+		goto out;
+	}
+
+	if (vcpu->arch.exception.pending)
+		__queue_exception(vcpu);
+	else if (irqchip_in_kernel(vcpu->kvm))
+		kvm_x86_ops->inject_pending_irq(vcpu);
+	else
+		kvm_x86_ops->inject_pending_vectors(vcpu, kvm_run);
+
+	vcpu->guest_mode = 1;
+	kvm_guest_enter();
+
+	if (vcpu->requests)
+		if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
+			kvm_x86_ops->tlb_flush(vcpu);
+
+	kvm_x86_ops->run(vcpu, kvm_run);
+
+	vcpu->guest_mode = 0;
+	local_irq_enable();
+
+	++vcpu->stat.exits;
+
+	/*
+	 * We must have an instruction between local_irq_enable() and
+	 * kvm_guest_exit(), so the timer interrupt isn't delayed by
+	 * the interrupt shadow.  The stat.exits increment will do nicely.
+	 * But we need to prevent reordering, hence this barrier():
+	 */
+	barrier();
+
+	kvm_guest_exit();
+
+	preempt_enable();
+
+	/*
+	 * Profile KVM exit RIPs:
+	 */
+	if (unlikely(prof_on == KVM_PROFILING)) {
+		kvm_x86_ops->cache_regs(vcpu);
+		profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip);
+	}
+
+	if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
+		vcpu->arch.exception.pending = false;
+
+	r = kvm_x86_ops->handle_exit(kvm_run, vcpu);
+
+	if (r > 0) {
+		if (dm_request_for_irq_injection(vcpu, kvm_run)) {
+			r = -EINTR;
+			kvm_run->exit_reason = KVM_EXIT_INTR;
+			++vcpu->stat.request_irq_exits;
+			goto out;
+		}
+		if (!need_resched())
+			goto again;
+	}
+
+out:
+	if (r > 0) {
+		kvm_resched(vcpu);
+		goto preempted;
+	}
+
+	post_kvm_run_save(vcpu, kvm_run);
+
+	return r;
+}
+
+int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	int r;
+	sigset_t sigsaved;
+
+	vcpu_load(vcpu);
+
+	if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_UNINITIALIZED)) {
+		kvm_vcpu_block(vcpu);
+		vcpu_put(vcpu);
+		return -EAGAIN;
+	}
+
+	if (vcpu->sigset_active)
+		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
+
+	/* re-sync apic's tpr */
+	if (!irqchip_in_kernel(vcpu->kvm))
+		set_cr8(vcpu, kvm_run->cr8);
+
+	if (vcpu->arch.pio.cur_count) {
+		r = complete_pio(vcpu);
+		if (r)
+			goto out;
+	}
+#if CONFIG_HAS_IOMEM
+	if (vcpu->mmio_needed) {
+		memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8);
+		vcpu->mmio_read_completed = 1;
+		vcpu->mmio_needed = 0;
+		r = emulate_instruction(vcpu, kvm_run,
+					vcpu->arch.mmio_fault_cr2, 0, 1);
+		if (r == EMULATE_DO_MMIO) {
+			/*
+			 * Read-modify-write.  Back to userspace.
+			 */
+			r = 0;
+			goto out;
+		}
+	}
+#endif
+	if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) {
+		kvm_x86_ops->cache_regs(vcpu);
+		vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
+		kvm_x86_ops->decache_regs(vcpu);
+	}
+
+	r = __vcpu_run(vcpu, kvm_run);
+
+out:
+	if (vcpu->sigset_active)
+		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+
+	vcpu_put(vcpu);
+	return r;
+}
+
+int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	vcpu_load(vcpu);
+
+	kvm_x86_ops->cache_regs(vcpu);
+
+	regs->rax = vcpu->arch.regs[VCPU_REGS_RAX];
+	regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX];
+	regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX];
+	regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX];
+	regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI];
+	regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI];
+	regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP];
+	regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP];
+#ifdef CONFIG_X86_64
+	regs->r8 = vcpu->arch.regs[VCPU_REGS_R8];
+	regs->r9 = vcpu->arch.regs[VCPU_REGS_R9];
+	regs->r10 = vcpu->arch.regs[VCPU_REGS_R10];
+	regs->r11 = vcpu->arch.regs[VCPU_REGS_R11];
+	regs->r12 = vcpu->arch.regs[VCPU_REGS_R12];
+	regs->r13 = vcpu->arch.regs[VCPU_REGS_R13];
+	regs->r14 = vcpu->arch.regs[VCPU_REGS_R14];
+	regs->r15 = vcpu->arch.regs[VCPU_REGS_R15];
+#endif
+
+	regs->rip = vcpu->arch.rip;
+	regs->rflags = kvm_x86_ops->get_rflags(vcpu);
+
+	/*
+	 * Don't leak debug flags in case they were set for guest debugging
+	 */
+	if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep)
+		regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	vcpu_load(vcpu);
+
+	vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax;
+	vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx;
+	vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx;
+	vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx;
+	vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi;
+	vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi;
+	vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp;
+	vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp;
+#ifdef CONFIG_X86_64
+	vcpu->arch.regs[VCPU_REGS_R8] = regs->r8;
+	vcpu->arch.regs[VCPU_REGS_R9] = regs->r9;
+	vcpu->arch.regs[VCPU_REGS_R10] = regs->r10;
+	vcpu->arch.regs[VCPU_REGS_R11] = regs->r11;
+	vcpu->arch.regs[VCPU_REGS_R12] = regs->r12;
+	vcpu->arch.regs[VCPU_REGS_R13] = regs->r13;
+	vcpu->arch.regs[VCPU_REGS_R14] = regs->r14;
+	vcpu->arch.regs[VCPU_REGS_R15] = regs->r15;
+#endif
+
+	vcpu->arch.rip = regs->rip;
+	kvm_x86_ops->set_rflags(vcpu, regs->rflags);
+
+	kvm_x86_ops->decache_regs(vcpu);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+static void get_segment(struct kvm_vcpu *vcpu,
+			struct kvm_segment *var, int seg)
+{
+	return kvm_x86_ops->get_segment(vcpu, var, seg);
+}
+
+void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l)
+{
+	struct kvm_segment cs;
+
+	get_segment(vcpu, &cs, VCPU_SREG_CS);
+	*db = cs.db;
+	*l = cs.l;
+}
+EXPORT_SYMBOL_GPL(kvm_get_cs_db_l_bits);
+
+int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	struct descriptor_table dt;
+	int pending_vec;
+
+	vcpu_load(vcpu);
+
+	get_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
+	get_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
+	get_segment(vcpu, &sregs->es, VCPU_SREG_ES);
+	get_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
+	get_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
+	get_segment(vcpu, &sregs->ss, VCPU_SREG_SS);
+
+	get_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
+	get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);
+
+	kvm_x86_ops->get_idt(vcpu, &dt);
+	sregs->idt.limit = dt.limit;
+	sregs->idt.base = dt.base;
+	kvm_x86_ops->get_gdt(vcpu, &dt);
+	sregs->gdt.limit = dt.limit;
+	sregs->gdt.base = dt.base;
+
+	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
+	sregs->cr0 = vcpu->arch.cr0;
+	sregs->cr2 = vcpu->arch.cr2;
+	sregs->cr3 = vcpu->arch.cr3;
+	sregs->cr4 = vcpu->arch.cr4;
+	sregs->cr8 = get_cr8(vcpu);
+	sregs->efer = vcpu->arch.shadow_efer;
+	sregs->apic_base = kvm_get_apic_base(vcpu);
+
+	if (irqchip_in_kernel(vcpu->kvm)) {
+		memset(sregs->interrupt_bitmap, 0,
+		       sizeof sregs->interrupt_bitmap);
+		pending_vec = kvm_x86_ops->get_irq(vcpu);
+		if (pending_vec >= 0)
+			set_bit(pending_vec,
+				(unsigned long *)sregs->interrupt_bitmap);
+	} else
+		memcpy(sregs->interrupt_bitmap, vcpu->arch.irq_pending,
+		       sizeof sregs->interrupt_bitmap);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+static void set_segment(struct kvm_vcpu *vcpu,
+			struct kvm_segment *var, int seg)
+{
+	return kvm_x86_ops->set_segment(vcpu, var, seg);
+}
+
+int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	int mmu_reset_needed = 0;
+	int i, pending_vec, max_bits;
+	struct descriptor_table dt;
+
+	vcpu_load(vcpu);
+
+	dt.limit = sregs->idt.limit;
+	dt.base = sregs->idt.base;
+	kvm_x86_ops->set_idt(vcpu, &dt);
+	dt.limit = sregs->gdt.limit;
+	dt.base = sregs->gdt.base;
+	kvm_x86_ops->set_gdt(vcpu, &dt);
+
+	vcpu->arch.cr2 = sregs->cr2;
+	mmu_reset_needed |= vcpu->arch.cr3 != sregs->cr3;
+	vcpu->arch.cr3 = sregs->cr3;
+
+	set_cr8(vcpu, sregs->cr8);
+
+	mmu_reset_needed |= vcpu->arch.shadow_efer != sregs->efer;
+#ifdef CONFIG_X86_64
+	kvm_x86_ops->set_efer(vcpu, sregs->efer);
+#endif
+	kvm_set_apic_base(vcpu, sregs->apic_base);
+
+	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
+
+	mmu_reset_needed |= vcpu->arch.cr0 != sregs->cr0;
+	vcpu->arch.cr0 = sregs->cr0;
+	kvm_x86_ops->set_cr0(vcpu, sregs->cr0);
+
+	mmu_reset_needed |= vcpu->arch.cr4 != sregs->cr4;
+	kvm_x86_ops->set_cr4(vcpu, sregs->cr4);
+	if (!is_long_mode(vcpu) && is_pae(vcpu))
+		load_pdptrs(vcpu, vcpu->arch.cr3);
+
+	if (mmu_reset_needed)
+		kvm_mmu_reset_context(vcpu);
+
+	if (!irqchip_in_kernel(vcpu->kvm)) {
+		memcpy(vcpu->arch.irq_pending, sregs->interrupt_bitmap,
+		       sizeof vcpu->arch.irq_pending);
+		vcpu->arch.irq_summary = 0;
+		for (i = 0; i < ARRAY_SIZE(vcpu->arch.irq_pending); ++i)
+			if (vcpu->arch.irq_pending[i])
+				__set_bit(i, &vcpu->arch.irq_summary);
+	} else {
+		max_bits = (sizeof sregs->interrupt_bitmap) << 3;
+		pending_vec = find_first_bit(
+			(const unsigned long *)sregs->interrupt_bitmap,
+			max_bits);
+		/* Only pending external irq is handled here */
+		if (pending_vec < max_bits) {
+			kvm_x86_ops->set_irq(vcpu, pending_vec);
+			pr_debug("Set back pending irq %d\n",
+				 pending_vec);
+		}
+	}
+
+	set_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
+	set_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
+	set_segment(vcpu, &sregs->es, VCPU_SREG_ES);
+	set_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
+	set_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
+	set_segment(vcpu, &sregs->ss, VCPU_SREG_SS);
+
+	set_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
+	set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
+				    struct kvm_debug_guest *dbg)
+{
+	int r;
+
+	vcpu_load(vcpu);
+
+	r = kvm_x86_ops->set_guest_debug(vcpu, dbg);
+
+	vcpu_put(vcpu);
+
+	return r;
+}
+
+/*
+ * fxsave fpu state.  Taken from x86_64/processor.h.  To be killed when
+ * we have asm/x86/processor.h
+ */
+struct fxsave {
+	u16	cwd;
+	u16	swd;
+	u16	twd;
+	u16	fop;
+	u64	rip;
+	u64	rdp;
+	u32	mxcsr;
+	u32	mxcsr_mask;
+	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
+#ifdef CONFIG_X86_64
+	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
+#else
+	u32	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
+#endif
+};
+
+/*
+ * Translate a guest virtual address to a guest physical address.
+ */
+int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
+				    struct kvm_translation *tr)
+{
+	unsigned long vaddr = tr->linear_address;
+	gpa_t gpa;
+
+	vcpu_load(vcpu);
+	mutex_lock(&vcpu->kvm->lock);
+	gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, vaddr);
+	tr->physical_address = gpa;
+	tr->valid = gpa != UNMAPPED_GVA;
+	tr->writeable = 1;
+	tr->usermode = 0;
+	mutex_unlock(&vcpu->kvm->lock);
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
+
+	vcpu_load(vcpu);
+
+	memcpy(fpu->fpr, fxsave->st_space, 128);
+	fpu->fcw = fxsave->cwd;
+	fpu->fsw = fxsave->swd;
+	fpu->ftwx = fxsave->twd;
+	fpu->last_opcode = fxsave->fop;
+	fpu->last_ip = fxsave->rip;
+	fpu->last_dp = fxsave->rdp;
+	memcpy(fpu->xmm, fxsave->xmm_space, sizeof fxsave->xmm_space);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
+
+	vcpu_load(vcpu);
+
+	memcpy(fxsave->st_space, fpu->fpr, 128);
+	fxsave->cwd = fpu->fcw;
+	fxsave->swd = fpu->fsw;
+	fxsave->twd = fpu->ftwx;
+	fxsave->fop = fpu->last_opcode;
+	fxsave->rip = fpu->last_ip;
+	fxsave->rdp = fpu->last_dp;
+	memcpy(fxsave->xmm_space, fpu->xmm, sizeof fxsave->xmm_space);
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
+void fx_init(struct kvm_vcpu *vcpu)
+{
+	unsigned after_mxcsr_mask;
+
+	/* Initialize guest FPU by resetting ours and saving into guest's */
+	preempt_disable();
+	fx_save(&vcpu->arch.host_fx_image);
+	fpu_init();
+	fx_save(&vcpu->arch.guest_fx_image);
+	fx_restore(&vcpu->arch.host_fx_image);
+	preempt_enable();
+
+	vcpu->arch.cr0 |= X86_CR0_ET;
+	after_mxcsr_mask = offsetof(struct i387_fxsave_struct, st_space);
+	vcpu->arch.guest_fx_image.mxcsr = 0x1f80;
+	memset((void *)&vcpu->arch.guest_fx_image + after_mxcsr_mask,
+	       0, sizeof(struct i387_fxsave_struct) - after_mxcsr_mask);
+}
+EXPORT_SYMBOL_GPL(fx_init);
+
+void kvm_load_guest_fpu(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu->fpu_active || vcpu->guest_fpu_loaded)
+		return;
+
+	vcpu->guest_fpu_loaded = 1;
+	fx_save(&vcpu->arch.host_fx_image);
+	fx_restore(&vcpu->arch.guest_fx_image);
+}
+EXPORT_SYMBOL_GPL(kvm_load_guest_fpu);
+
+void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu->guest_fpu_loaded)
+		return;
+
+	vcpu->guest_fpu_loaded = 0;
+	fx_save(&vcpu->arch.guest_fx_image);
+	fx_restore(&vcpu->arch.host_fx_image);
+	++vcpu->stat.fpu_reload;
+}
+EXPORT_SYMBOL_GPL(kvm_put_guest_fpu);
+
+void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	kvm_x86_ops->vcpu_free(vcpu);
+}
+
+struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
+						unsigned int id)
+{
+	return kvm_x86_ops->vcpu_create(kvm, id);
+}
+
+int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
+{
+	int r;
+
+	/* We do fxsave: this must be aligned. */
+	BUG_ON((unsigned long)&vcpu->arch.host_fx_image & 0xF);
+
+	vcpu_load(vcpu);
+	r = kvm_arch_vcpu_reset(vcpu);
+	if (r == 0)
+		r = kvm_mmu_setup(vcpu);
+	vcpu_put(vcpu);
+	if (r < 0)
+		goto free_vcpu;
+
+	return 0;
+free_vcpu:
+	kvm_x86_ops->vcpu_free(vcpu);
+	return r;
+}
+
+void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	vcpu_load(vcpu);
+	kvm_mmu_unload(vcpu);
+	vcpu_put(vcpu);
+
+	kvm_x86_ops->vcpu_free(vcpu);
+}
+
+int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	return kvm_x86_ops->vcpu_reset(vcpu);
+}
+
+void kvm_arch_hardware_enable(void *garbage)
+{
+	kvm_x86_ops->hardware_enable(garbage);
+}
+
+void kvm_arch_hardware_disable(void *garbage)
+{
+	kvm_x86_ops->hardware_disable(garbage);
+}
+
+int kvm_arch_hardware_setup(void)
+{
+	return kvm_x86_ops->hardware_setup();
+}
+
+void kvm_arch_hardware_unsetup(void)
+{
+	kvm_x86_ops->hardware_unsetup();
+}
+
+void kvm_arch_check_processor_compat(void *rtn)
+{
+	kvm_x86_ops->check_processor_compatibility(rtn);
+}
+
+int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
+{
+	struct page *page;
+	struct kvm *kvm;
+	int r;
+
+	BUG_ON(vcpu->kvm == NULL);
+	kvm = vcpu->kvm;
+
+	vcpu->arch.mmu.root_hpa = INVALID_PAGE;
+	if (!irqchip_in_kernel(kvm) || vcpu->vcpu_id == 0)
+		vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
+	else
+		vcpu->arch.mp_state = VCPU_MP_STATE_UNINITIALIZED;
+
+	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+	if (!page) {
+		r = -ENOMEM;
+		goto fail;
+	}
+	vcpu->arch.pio_data = page_address(page);
+
+	r = kvm_mmu_create(vcpu);
+	if (r < 0)
+		goto fail_free_pio_data;
+
+	if (irqchip_in_kernel(kvm)) {
+		r = kvm_create_lapic(vcpu);
+		if (r < 0)
+			goto fail_mmu_destroy;
+	}
+
+	return 0;
+
+fail_mmu_destroy:
+	kvm_mmu_destroy(vcpu);
+fail_free_pio_data:
+	free_page((unsigned long)vcpu->arch.pio_data);
+fail:
+	return r;
+}
+
+void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
+{
+	kvm_free_lapic(vcpu);
+	kvm_mmu_destroy(vcpu);
+	free_page((unsigned long)vcpu->arch.pio_data);
+}
+
+struct  kvm *kvm_arch_create_vm(void)
+{
+	struct kvm *kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL);
+
+	if (!kvm)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
+
+	return kvm;
+}
+
+static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu)
+{
+	vcpu_load(vcpu);
+	kvm_mmu_unload(vcpu);
+	vcpu_put(vcpu);
+}
+
+static void kvm_free_vcpus(struct kvm *kvm)
+{
+	unsigned int i;
+
+	/*
+	 * Unpin any mmu pages first.
+	 */
+	for (i = 0; i < KVM_MAX_VCPUS; ++i)
+		if (kvm->vcpus[i])
+			kvm_unload_vcpu_mmu(kvm->vcpus[i]);
+	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+		if (kvm->vcpus[i]) {
+			kvm_arch_vcpu_free(kvm->vcpus[i]);
+			kvm->vcpus[i] = NULL;
+		}
+	}
+
+}
+
+void kvm_arch_destroy_vm(struct kvm *kvm)
+{
+	kfree(kvm->arch.vpic);
+	kfree(kvm->arch.vioapic);
+	kvm_free_vcpus(kvm);
+	kvm_free_physmem(kvm);
+	kfree(kvm);
+}
+
+int kvm_arch_set_memory_region(struct kvm *kvm,
+				struct kvm_userspace_memory_region *mem,
+				struct kvm_memory_slot old,
+				int user_alloc)
+{
+	int npages = mem->memory_size >> PAGE_SHIFT;
+	struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot];
+
+	/*To keep backward compatibility with older userspace,
+	 *x86 needs to hanlde !user_alloc case.
+	 */
+	if (!user_alloc) {
+		if (npages && !old.rmap) {
+			down_write(&current->mm->mmap_sem);
+			memslot->userspace_addr = do_mmap(NULL, 0,
+						     npages * PAGE_SIZE,
+						     PROT_READ | PROT_WRITE,
+						     MAP_SHARED | MAP_ANONYMOUS,
+						     0);
+			up_write(&current->mm->mmap_sem);
+
+			if (IS_ERR((void *)memslot->userspace_addr))
+				return PTR_ERR((void *)memslot->userspace_addr);
+		} else {
+			if (!old.user_alloc && old.rmap) {
+				int ret;
+
+				down_write(&current->mm->mmap_sem);
+				ret = do_munmap(current->mm, old.userspace_addr,
+						old.npages * PAGE_SIZE);
+				up_write(&current->mm->mmap_sem);
+				if (ret < 0)
+					printk(KERN_WARNING
+				       "kvm_vm_ioctl_set_memory_region: "
+				       "failed to munmap memory\n");
+			}
+		}
+	}
+
+	if (!kvm->arch.n_requested_mmu_pages) {
+		unsigned int nr_mmu_pages = kvm_mmu_calculate_mmu_pages(kvm);
+		kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
+	}
+
+	kvm_mmu_slot_remove_write_access(kvm, mem->slot);
+	kvm_flush_remote_tlbs(kvm);
+
+	return 0;
+}
+
+int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
+	       || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
+}
diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c
new file mode 100644
index 000000000000..79586003397a
--- /dev/null
+++ b/arch/x86/kvm/x86_emulate.c
@@ -0,0 +1,1912 @@
+/******************************************************************************
+ * x86_emulate.c
+ *
+ * Generic x86 (32-bit and 64-bit) instruction decoder and emulator.
+ *
+ * Copyright (c) 2005 Keir Fraser
+ *
+ * Linux coding style, mod r/m decoder, segment base fixes, real-mode
+ * privileged instructions:
+ *
+ * Copyright (C) 2006 Qumranet
+ *
+ *   Avi Kivity <avi@qumranet.com>
+ *   Yaniv Kamay <yaniv@qumranet.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4
+ */
+
+#ifndef __KERNEL__
+#include <stdio.h>
+#include <stdint.h>
+#include <public/xen.h>
+#define DPRINTF(_f, _a ...) printf(_f , ## _a)
+#else
+#include <linux/kvm_host.h>
+#define DPRINTF(x...) do {} while (0)
+#endif
+#include <linux/module.h>
+#include <asm/kvm_x86_emulate.h>
+
+/*
+ * Opcode effective-address decode tables.
+ * Note that we only emulate instructions that have at least one memory
+ * operand (excluding implicit stack references). We assume that stack
+ * references and instruction fetches will never occur in special memory
+ * areas that require emulation. So, for example, 'mov <imm>,<reg>' need
+ * not be handled.
+ */
+
+/* Operand sizes: 8-bit operands or specified/overridden size. */
+#define ByteOp      (1<<0)	/* 8-bit operands. */
+/* Destination operand type. */
+#define ImplicitOps (1<<1)	/* Implicit in opcode. No generic decode. */
+#define DstReg      (2<<1)	/* Register operand. */
+#define DstMem      (3<<1)	/* Memory operand. */
+#define DstMask     (3<<1)
+/* Source operand type. */
+#define SrcNone     (0<<3)	/* No source operand. */
+#define SrcImplicit (0<<3)	/* Source operand is implicit in the opcode. */
+#define SrcReg      (1<<3)	/* Register operand. */
+#define SrcMem      (2<<3)	/* Memory operand. */
+#define SrcMem16    (3<<3)	/* Memory operand (16-bit). */
+#define SrcMem32    (4<<3)	/* Memory operand (32-bit). */
+#define SrcImm      (5<<3)	/* Immediate operand. */
+#define SrcImmByte  (6<<3)	/* 8-bit sign-extended immediate operand. */
+#define SrcMask     (7<<3)
+/* Generic ModRM decode. */
+#define ModRM       (1<<6)
+/* Destination is only written; never read. */
+#define Mov         (1<<7)
+#define BitOp       (1<<8)
+#define MemAbs      (1<<9)      /* Memory operand is absolute displacement */
+#define String      (1<<10)     /* String instruction (rep capable) */
+#define Stack       (1<<11)     /* Stack instruction (push/pop) */
+
+static u16 opcode_table[256] = {
+	/* 0x00 - 0x07 */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x08 - 0x0F */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x10 - 0x17 */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x18 - 0x1F */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x20 - 0x27 */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	SrcImmByte, SrcImm, 0, 0,
+	/* 0x28 - 0x2F */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x30 - 0x37 */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x38 - 0x3F */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
+	0, 0, 0, 0,
+	/* 0x40 - 0x47 */
+	DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg,
+	/* 0x48 - 0x4F */
+	DstReg, DstReg, DstReg, DstReg,	DstReg, DstReg, DstReg, DstReg,
+	/* 0x50 - 0x57 */
+	SrcReg | Stack, SrcReg | Stack, SrcReg | Stack, SrcReg | Stack,
+	SrcReg | Stack, SrcReg | Stack, SrcReg | Stack, SrcReg | Stack,
+	/* 0x58 - 0x5F */
+	DstReg | Stack, DstReg | Stack, DstReg | Stack, DstReg | Stack,
+	DstReg | Stack, DstReg | Stack, DstReg | Stack, DstReg | Stack,
+	/* 0x60 - 0x67 */
+	0, 0, 0, DstReg | SrcMem32 | ModRM | Mov /* movsxd (x86/64) */ ,
+	0, 0, 0, 0,
+	/* 0x68 - 0x6F */
+	0, 0, ImplicitOps | Mov | Stack, 0,
+	SrcNone  | ByteOp  | ImplicitOps, SrcNone  | ImplicitOps, /* insb, insw/insd */
+	SrcNone  | ByteOp  | ImplicitOps, SrcNone  | ImplicitOps, /* outsb, outsw/outsd */
+	/* 0x70 - 0x77 */
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	/* 0x78 - 0x7F */
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	/* 0x80 - 0x87 */
+	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImm | ModRM,
+	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM,
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
+	/* 0x88 - 0x8F */
+	ByteOp | DstMem | SrcReg | ModRM | Mov, DstMem | SrcReg | ModRM | Mov,
+	ByteOp | DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	0, ModRM | DstReg, 0, DstMem | SrcNone | ModRM | Mov | Stack,
+	/* 0x90 - 0x9F */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, ImplicitOps | Stack, ImplicitOps | Stack, 0, 0,
+	/* 0xA0 - 0xA7 */
+	ByteOp | DstReg | SrcMem | Mov | MemAbs, DstReg | SrcMem | Mov | MemAbs,
+	ByteOp | DstMem | SrcReg | Mov | MemAbs, DstMem | SrcReg | Mov | MemAbs,
+	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | String, ImplicitOps | String,
+	/* 0xA8 - 0xAF */
+	0, 0, ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | String, ImplicitOps | String,
+	/* 0xB0 - 0xBF */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xC0 - 0xC7 */
+	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM,
+	0, ImplicitOps | Stack, 0, 0,
+	ByteOp | DstMem | SrcImm | ModRM | Mov, DstMem | SrcImm | ModRM | Mov,
+	/* 0xC8 - 0xCF */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xD0 - 0xD7 */
+	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
+	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
+	0, 0, 0, 0,
+	/* 0xD8 - 0xDF */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xE0 - 0xE7 */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xE8 - 0xEF */
+	ImplicitOps | Stack, SrcImm|ImplicitOps, 0, SrcImmByte|ImplicitOps,
+	0, 0, 0, 0,
+	/* 0xF0 - 0xF7 */
+	0, 0, 0, 0,
+	ImplicitOps, ImplicitOps,
+	ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM,
+	/* 0xF8 - 0xFF */
+	ImplicitOps, 0, ImplicitOps, ImplicitOps,
+	0, 0, ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM
+};
+
+static u16 twobyte_table[256] = {
+	/* 0x00 - 0x0F */
+	0, SrcMem | ModRM | DstReg, 0, 0, 0, 0, ImplicitOps, 0,
+	ImplicitOps, ImplicitOps, 0, 0, 0, ImplicitOps | ModRM, 0, 0,
+	/* 0x10 - 0x1F */
+	0, 0, 0, 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x20 - 0x2F */
+	ModRM | ImplicitOps, ModRM, ModRM | ImplicitOps, ModRM, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x30 - 0x3F */
+	ImplicitOps, 0, ImplicitOps, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x40 - 0x47 */
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	/* 0x48 - 0x4F */
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
+	/* 0x50 - 0x5F */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x60 - 0x6F */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x70 - 0x7F */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0x80 - 0x8F */
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
+	/* 0x90 - 0x9F */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xA0 - 0xA7 */
+	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0,
+	/* 0xA8 - 0xAF */
+	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0,
+	/* 0xB0 - 0xB7 */
+	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, 0,
+	    DstMem | SrcReg | ModRM | BitOp,
+	0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov,
+	    DstReg | SrcMem16 | ModRM | Mov,
+	/* 0xB8 - 0xBF */
+	0, 0, DstMem | SrcImmByte | ModRM, DstMem | SrcReg | ModRM | BitOp,
+	0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov,
+	    DstReg | SrcMem16 | ModRM | Mov,
+	/* 0xC0 - 0xCF */
+	0, 0, 0, DstMem | SrcReg | ModRM | Mov, 0, 0, 0, ImplicitOps | ModRM,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xD0 - 0xDF */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xE0 - 0xEF */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	/* 0xF0 - 0xFF */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/* EFLAGS bit definitions. */
+#define EFLG_OF (1<<11)
+#define EFLG_DF (1<<10)
+#define EFLG_SF (1<<7)
+#define EFLG_ZF (1<<6)
+#define EFLG_AF (1<<4)
+#define EFLG_PF (1<<2)
+#define EFLG_CF (1<<0)
+
+/*
+ * Instruction emulation:
+ * Most instructions are emulated directly via a fragment of inline assembly
+ * code. This allows us to save/restore EFLAGS and thus very easily pick up
+ * any modified flags.
+ */
+
+#if defined(CONFIG_X86_64)
+#define _LO32 "k"		/* force 32-bit operand */
+#define _STK  "%%rsp"		/* stack pointer */
+#elif defined(__i386__)
+#define _LO32 ""		/* force 32-bit operand */
+#define _STK  "%%esp"		/* stack pointer */
+#endif
+
+/*
+ * These EFLAGS bits are restored from saved value during emulation, and
+ * any changes are written back to the saved value after emulation.
+ */
+#define EFLAGS_MASK (EFLG_OF|EFLG_SF|EFLG_ZF|EFLG_AF|EFLG_PF|EFLG_CF)
+
+/* Before executing instruction: restore necessary bits in EFLAGS. */
+#define _PRE_EFLAGS(_sav, _msk, _tmp)					\
+	/* EFLAGS = (_sav & _msk) | (EFLAGS & ~_msk); _sav &= ~_msk; */ \
+	"movl %"_sav",%"_LO32 _tmp"; "                                  \
+	"push %"_tmp"; "                                                \
+	"push %"_tmp"; "                                                \
+	"movl %"_msk",%"_LO32 _tmp"; "                                  \
+	"andl %"_LO32 _tmp",("_STK"); "                                 \
+	"pushf; "                                                       \
+	"notl %"_LO32 _tmp"; "                                          \
+	"andl %"_LO32 _tmp",("_STK"); "                                 \
+	"andl %"_LO32 _tmp","__stringify(BITS_PER_LONG/4)"("_STK"); "	\
+	"pop  %"_tmp"; "                                                \
+	"orl  %"_LO32 _tmp",("_STK"); "                                 \
+	"popf; "                                                        \
+	"pop  %"_sav"; "
+
+/* After executing instruction: write-back necessary bits in EFLAGS. */
+#define _POST_EFLAGS(_sav, _msk, _tmp) \
+	/* _sav |= EFLAGS & _msk; */		\
+	"pushf; "				\
+	"pop  %"_tmp"; "			\
+	"andl %"_msk",%"_LO32 _tmp"; "		\
+	"orl  %"_LO32 _tmp",%"_sav"; "
+
+/* Raw emulation: instruction has two explicit operands. */
+#define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \
+	do { 								    \
+		unsigned long _tmp;					    \
+									    \
+		switch ((_dst).bytes) {					    \
+		case 2:							    \
+			__asm__ __volatile__ (				    \
+				_PRE_EFLAGS("0", "4", "2")		    \
+				_op"w %"_wx"3,%1; "			    \
+				_POST_EFLAGS("0", "4", "2")		    \
+				: "=m" (_eflags), "=m" ((_dst).val),        \
+				  "=&r" (_tmp)				    \
+				: _wy ((_src).val), "i" (EFLAGS_MASK));     \
+			break;						    \
+		case 4:							    \
+			__asm__ __volatile__ (				    \
+				_PRE_EFLAGS("0", "4", "2")		    \
+				_op"l %"_lx"3,%1; "			    \
+				_POST_EFLAGS("0", "4", "2")		    \
+				: "=m" (_eflags), "=m" ((_dst).val),	    \
+				  "=&r" (_tmp)				    \
+				: _ly ((_src).val), "i" (EFLAGS_MASK));     \
+			break;						    \
+		case 8:							    \
+			__emulate_2op_8byte(_op, _src, _dst,		    \
+					    _eflags, _qx, _qy);		    \
+			break;						    \
+		}							    \
+	} while (0)
+
+#define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \
+	do {								     \
+		unsigned long _tmp;					     \
+		switch ((_dst).bytes) {				             \
+		case 1:							     \
+			__asm__ __volatile__ (				     \
+				_PRE_EFLAGS("0", "4", "2")		     \
+				_op"b %"_bx"3,%1; "			     \
+				_POST_EFLAGS("0", "4", "2")		     \
+				: "=m" (_eflags), "=m" ((_dst).val),	     \
+				  "=&r" (_tmp)				     \
+				: _by ((_src).val), "i" (EFLAGS_MASK));      \
+			break;						     \
+		default:						     \
+			__emulate_2op_nobyte(_op, _src, _dst, _eflags,	     \
+					     _wx, _wy, _lx, _ly, _qx, _qy);  \
+			break;						     \
+		}							     \
+	} while (0)
+
+/* Source operand is byte-sized and may be restricted to just %cl. */
+#define emulate_2op_SrcB(_op, _src, _dst, _eflags)                      \
+	__emulate_2op(_op, _src, _dst, _eflags,				\
+		      "b", "c", "b", "c", "b", "c", "b", "c")
+
+/* Source operand is byte, word, long or quad sized. */
+#define emulate_2op_SrcV(_op, _src, _dst, _eflags)                      \
+	__emulate_2op(_op, _src, _dst, _eflags,				\
+		      "b", "q", "w", "r", _LO32, "r", "", "r")
+
+/* Source operand is word, long or quad sized. */
+#define emulate_2op_SrcV_nobyte(_op, _src, _dst, _eflags)               \
+	__emulate_2op_nobyte(_op, _src, _dst, _eflags,			\
+			     "w", "r", _LO32, "r", "", "r")
+
+/* Instruction has only one explicit operand (no source operand). */
+#define emulate_1op(_op, _dst, _eflags)                                    \
+	do {								\
+		unsigned long _tmp;					\
+									\
+		switch ((_dst).bytes) {				        \
+		case 1:							\
+			__asm__ __volatile__ (				\
+				_PRE_EFLAGS("0", "3", "2")		\
+				_op"b %1; "				\
+				_POST_EFLAGS("0", "3", "2")		\
+				: "=m" (_eflags), "=m" ((_dst).val),	\
+				  "=&r" (_tmp)				\
+				: "i" (EFLAGS_MASK));			\
+			break;						\
+		case 2:							\
+			__asm__ __volatile__ (				\
+				_PRE_EFLAGS("0", "3", "2")		\
+				_op"w %1; "				\
+				_POST_EFLAGS("0", "3", "2")		\
+				: "=m" (_eflags), "=m" ((_dst).val),	\
+				  "=&r" (_tmp)				\
+				: "i" (EFLAGS_MASK));			\
+			break;						\
+		case 4:							\
+			__asm__ __volatile__ (				\
+				_PRE_EFLAGS("0", "3", "2")		\
+				_op"l %1; "				\
+				_POST_EFLAGS("0", "3", "2")		\
+				: "=m" (_eflags), "=m" ((_dst).val),	\
+				  "=&r" (_tmp)				\
+				: "i" (EFLAGS_MASK));			\
+			break;						\
+		case 8:							\
+			__emulate_1op_8byte(_op, _dst, _eflags);	\
+			break;						\
+		}							\
+	} while (0)
+
+/* Emulate an instruction with quadword operands (x86/64 only). */
+#if defined(CONFIG_X86_64)
+#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)           \
+	do {								  \
+		__asm__ __volatile__ (					  \
+			_PRE_EFLAGS("0", "4", "2")			  \
+			_op"q %"_qx"3,%1; "				  \
+			_POST_EFLAGS("0", "4", "2")			  \
+			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
+			: _qy ((_src).val), "i" (EFLAGS_MASK));		\
+	} while (0)
+
+#define __emulate_1op_8byte(_op, _dst, _eflags)                           \
+	do {								  \
+		__asm__ __volatile__ (					  \
+			_PRE_EFLAGS("0", "3", "2")			  \
+			_op"q %1; "					  \
+			_POST_EFLAGS("0", "3", "2")			  \
+			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
+			: "i" (EFLAGS_MASK));				  \
+	} while (0)
+
+#elif defined(__i386__)
+#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)
+#define __emulate_1op_8byte(_op, _dst, _eflags)
+#endif				/* __i386__ */
+
+/* Fetch next part of the instruction being emulated. */
+#define insn_fetch(_type, _size, _eip)                                  \
+({	unsigned long _x;						\
+	rc = do_insn_fetch(ctxt, ops, (_eip), &_x, (_size));		\
+	if (rc != 0)							\
+		goto done;						\
+	(_eip) += (_size);						\
+	(_type)_x;							\
+})
+
+/* Access/update address held in a register, based on addressing mode. */
+#define address_mask(reg)						\
+	((c->ad_bytes == sizeof(unsigned long)) ? 			\
+		(reg) :	((reg) & ((1UL << (c->ad_bytes << 3)) - 1)))
+#define register_address(base, reg)                                     \
+	((base) + address_mask(reg))
+#define register_address_increment(reg, inc)                            \
+	do {								\
+		/* signed type ensures sign extension to long */        \
+		int _inc = (inc);					\
+		if (c->ad_bytes == sizeof(unsigned long))		\
+			(reg) += _inc;					\
+		else							\
+			(reg) = ((reg) & 				\
+				 ~((1UL << (c->ad_bytes << 3)) - 1)) |	\
+				(((reg) + _inc) &			\
+				 ((1UL << (c->ad_bytes << 3)) - 1));	\
+	} while (0)
+
+#define JMP_REL(rel) 							\
+	do {								\
+		register_address_increment(c->eip, rel);		\
+	} while (0)
+
+static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt,
+			      struct x86_emulate_ops *ops,
+			      unsigned long linear, u8 *dest)
+{
+	struct fetch_cache *fc = &ctxt->decode.fetch;
+	int rc;
+	int size;
+
+	if (linear < fc->start || linear >= fc->end) {
+		size = min(15UL, PAGE_SIZE - offset_in_page(linear));
+		rc = ops->read_std(linear, fc->data, size, ctxt->vcpu);
+		if (rc)
+			return rc;
+		fc->start = linear;
+		fc->end = linear + size;
+	}
+	*dest = fc->data[linear - fc->start];
+	return 0;
+}
+
+static int do_insn_fetch(struct x86_emulate_ctxt *ctxt,
+			 struct x86_emulate_ops *ops,
+			 unsigned long eip, void *dest, unsigned size)
+{
+	int rc = 0;
+
+	eip += ctxt->cs_base;
+	while (size--) {
+		rc = do_fetch_insn_byte(ctxt, ops, eip++, dest++);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+/*
+ * Given the 'reg' portion of a ModRM byte, and a register block, return a
+ * pointer into the block that addresses the relevant register.
+ * @highbyte_regs specifies whether to decode AH,CH,DH,BH.
+ */
+static void *decode_register(u8 modrm_reg, unsigned long *regs,
+			     int highbyte_regs)
+{
+	void *p;
+
+	p = &regs[modrm_reg];
+	if (highbyte_regs && modrm_reg >= 4 && modrm_reg < 8)
+		p = (unsigned char *)&regs[modrm_reg & 3] + 1;
+	return p;
+}
+
+static int read_descriptor(struct x86_emulate_ctxt *ctxt,
+			   struct x86_emulate_ops *ops,
+			   void *ptr,
+			   u16 *size, unsigned long *address, int op_bytes)
+{
+	int rc;
+
+	if (op_bytes == 2)
+		op_bytes = 3;
+	*address = 0;
+	rc = ops->read_std((unsigned long)ptr, (unsigned long *)size, 2,
+			   ctxt->vcpu);
+	if (rc)
+		return rc;
+	rc = ops->read_std((unsigned long)ptr + 2, address, op_bytes,
+			   ctxt->vcpu);
+	return rc;
+}
+
+static int test_cc(unsigned int condition, unsigned int flags)
+{
+	int rc = 0;
+
+	switch ((condition & 15) >> 1) {
+	case 0: /* o */
+		rc |= (flags & EFLG_OF);
+		break;
+	case 1: /* b/c/nae */
+		rc |= (flags & EFLG_CF);
+		break;
+	case 2: /* z/e */
+		rc |= (flags & EFLG_ZF);
+		break;
+	case 3: /* be/na */
+		rc |= (flags & (EFLG_CF|EFLG_ZF));
+		break;
+	case 4: /* s */
+		rc |= (flags & EFLG_SF);
+		break;
+	case 5: /* p/pe */
+		rc |= (flags & EFLG_PF);
+		break;
+	case 7: /* le/ng */
+		rc |= (flags & EFLG_ZF);
+		/* fall through */
+	case 6: /* l/nge */
+		rc |= (!(flags & EFLG_SF) != !(flags & EFLG_OF));
+		break;
+	}
+
+	/* Odd condition identifiers (lsb == 1) have inverted sense. */
+	return (!!rc ^ (condition & 1));
+}
+
+static void decode_register_operand(struct operand *op,
+				    struct decode_cache *c,
+				    int inhibit_bytereg)
+{
+	unsigned reg = c->modrm_reg;
+	int highbyte_regs = c->rex_prefix == 0;
+
+	if (!(c->d & ModRM))
+		reg = (c->b & 7) | ((c->rex_prefix & 1) << 3);
+	op->type = OP_REG;
+	if ((c->d & ByteOp) && !inhibit_bytereg) {
+		op->ptr = decode_register(reg, c->regs, highbyte_regs);
+		op->val = *(u8 *)op->ptr;
+		op->bytes = 1;
+	} else {
+		op->ptr = decode_register(reg, c->regs, 0);
+		op->bytes = c->op_bytes;
+		switch (op->bytes) {
+		case 2:
+			op->val = *(u16 *)op->ptr;
+			break;
+		case 4:
+			op->val = *(u32 *)op->ptr;
+			break;
+		case 8:
+			op->val = *(u64 *) op->ptr;
+			break;
+		}
+	}
+	op->orig_val = op->val;
+}
+
+static int decode_modrm(struct x86_emulate_ctxt *ctxt,
+			struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	u8 sib;
+	int index_reg = 0, base_reg = 0, scale, rip_relative = 0;
+	int rc = 0;
+
+	if (c->rex_prefix) {
+		c->modrm_reg = (c->rex_prefix & 4) << 1;	/* REX.R */
+		index_reg = (c->rex_prefix & 2) << 2; /* REX.X */
+		c->modrm_rm = base_reg = (c->rex_prefix & 1) << 3; /* REG.B */
+	}
+
+	c->modrm = insn_fetch(u8, 1, c->eip);
+	c->modrm_mod |= (c->modrm & 0xc0) >> 6;
+	c->modrm_reg |= (c->modrm & 0x38) >> 3;
+	c->modrm_rm |= (c->modrm & 0x07);
+	c->modrm_ea = 0;
+	c->use_modrm_ea = 1;
+
+	if (c->modrm_mod == 3) {
+		c->modrm_val = *(unsigned long *)
+			decode_register(c->modrm_rm, c->regs, c->d & ByteOp);
+		return rc;
+	}
+
+	if (c->ad_bytes == 2) {
+		unsigned bx = c->regs[VCPU_REGS_RBX];
+		unsigned bp = c->regs[VCPU_REGS_RBP];
+		unsigned si = c->regs[VCPU_REGS_RSI];
+		unsigned di = c->regs[VCPU_REGS_RDI];
+
+		/* 16-bit ModR/M decode. */
+		switch (c->modrm_mod) {
+		case 0:
+			if (c->modrm_rm == 6)
+				c->modrm_ea += insn_fetch(u16, 2, c->eip);
+			break;
+		case 1:
+			c->modrm_ea += insn_fetch(s8, 1, c->eip);
+			break;
+		case 2:
+			c->modrm_ea += insn_fetch(u16, 2, c->eip);
+			break;
+		}
+		switch (c->modrm_rm) {
+		case 0:
+			c->modrm_ea += bx + si;
+			break;
+		case 1:
+			c->modrm_ea += bx + di;
+			break;
+		case 2:
+			c->modrm_ea += bp + si;
+			break;
+		case 3:
+			c->modrm_ea += bp + di;
+			break;
+		case 4:
+			c->modrm_ea += si;
+			break;
+		case 5:
+			c->modrm_ea += di;
+			break;
+		case 6:
+			if (c->modrm_mod != 0)
+				c->modrm_ea += bp;
+			break;
+		case 7:
+			c->modrm_ea += bx;
+			break;
+		}
+		if (c->modrm_rm == 2 || c->modrm_rm == 3 ||
+		    (c->modrm_rm == 6 && c->modrm_mod != 0))
+			if (!c->override_base)
+				c->override_base = &ctxt->ss_base;
+		c->modrm_ea = (u16)c->modrm_ea;
+	} else {
+		/* 32/64-bit ModR/M decode. */
+		switch (c->modrm_rm) {
+		case 4:
+		case 12:
+			sib = insn_fetch(u8, 1, c->eip);
+			index_reg |= (sib >> 3) & 7;
+			base_reg |= sib & 7;
+			scale = sib >> 6;
+
+			switch (base_reg) {
+			case 5:
+				if (c->modrm_mod != 0)
+					c->modrm_ea += c->regs[base_reg];
+				else
+					c->modrm_ea +=
+						insn_fetch(s32, 4, c->eip);
+				break;
+			default:
+				c->modrm_ea += c->regs[base_reg];
+			}
+			switch (index_reg) {
+			case 4:
+				break;
+			default:
+				c->modrm_ea += c->regs[index_reg] << scale;
+			}
+			break;
+		case 5:
+			if (c->modrm_mod != 0)
+				c->modrm_ea += c->regs[c->modrm_rm];
+			else if (ctxt->mode == X86EMUL_MODE_PROT64)
+				rip_relative = 1;
+			break;
+		default:
+			c->modrm_ea += c->regs[c->modrm_rm];
+			break;
+		}
+		switch (c->modrm_mod) {
+		case 0:
+			if (c->modrm_rm == 5)
+				c->modrm_ea += insn_fetch(s32, 4, c->eip);
+			break;
+		case 1:
+			c->modrm_ea += insn_fetch(s8, 1, c->eip);
+			break;
+		case 2:
+			c->modrm_ea += insn_fetch(s32, 4, c->eip);
+			break;
+		}
+	}
+	if (rip_relative) {
+		c->modrm_ea += c->eip;
+		switch (c->d & SrcMask) {
+		case SrcImmByte:
+			c->modrm_ea += 1;
+			break;
+		case SrcImm:
+			if (c->d & ByteOp)
+				c->modrm_ea += 1;
+			else
+				if (c->op_bytes == 8)
+					c->modrm_ea += 4;
+				else
+					c->modrm_ea += c->op_bytes;
+		}
+	}
+done:
+	return rc;
+}
+
+static int decode_abs(struct x86_emulate_ctxt *ctxt,
+		      struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc = 0;
+
+	switch (c->ad_bytes) {
+	case 2:
+		c->modrm_ea = insn_fetch(u16, 2, c->eip);
+		break;
+	case 4:
+		c->modrm_ea = insn_fetch(u32, 4, c->eip);
+		break;
+	case 8:
+		c->modrm_ea = insn_fetch(u64, 8, c->eip);
+		break;
+	}
+done:
+	return rc;
+}
+
+int
+x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc = 0;
+	int mode = ctxt->mode;
+	int def_op_bytes, def_ad_bytes;
+
+	/* Shadow copy of register state. Committed on successful emulation. */
+
+	memset(c, 0, sizeof(struct decode_cache));
+	c->eip = ctxt->vcpu->arch.rip;
+	memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
+
+	switch (mode) {
+	case X86EMUL_MODE_REAL:
+	case X86EMUL_MODE_PROT16:
+		def_op_bytes = def_ad_bytes = 2;
+		break;
+	case X86EMUL_MODE_PROT32:
+		def_op_bytes = def_ad_bytes = 4;
+		break;
+#ifdef CONFIG_X86_64
+	case X86EMUL_MODE_PROT64:
+		def_op_bytes = 4;
+		def_ad_bytes = 8;
+		break;
+#endif
+	default:
+		return -1;
+	}
+
+	c->op_bytes = def_op_bytes;
+	c->ad_bytes = def_ad_bytes;
+
+	/* Legacy prefixes. */
+	for (;;) {
+		switch (c->b = insn_fetch(u8, 1, c->eip)) {
+		case 0x66:	/* operand-size override */
+			/* switch between 2/4 bytes */
+			c->op_bytes = def_op_bytes ^ 6;
+			break;
+		case 0x67:	/* address-size override */
+			if (mode == X86EMUL_MODE_PROT64)
+				/* switch between 4/8 bytes */
+				c->ad_bytes = def_ad_bytes ^ 12;
+			else
+				/* switch between 2/4 bytes */
+				c->ad_bytes = def_ad_bytes ^ 6;
+			break;
+		case 0x2e:	/* CS override */
+			c->override_base = &ctxt->cs_base;
+			break;
+		case 0x3e:	/* DS override */
+			c->override_base = &ctxt->ds_base;
+			break;
+		case 0x26:	/* ES override */
+			c->override_base = &ctxt->es_base;
+			break;
+		case 0x64:	/* FS override */
+			c->override_base = &ctxt->fs_base;
+			break;
+		case 0x65:	/* GS override */
+			c->override_base = &ctxt->gs_base;
+			break;
+		case 0x36:	/* SS override */
+			c->override_base = &ctxt->ss_base;
+			break;
+		case 0x40 ... 0x4f: /* REX */
+			if (mode != X86EMUL_MODE_PROT64)
+				goto done_prefixes;
+			c->rex_prefix = c->b;
+			continue;
+		case 0xf0:	/* LOCK */
+			c->lock_prefix = 1;
+			break;
+		case 0xf2:	/* REPNE/REPNZ */
+			c->rep_prefix = REPNE_PREFIX;
+			break;
+		case 0xf3:	/* REP/REPE/REPZ */
+			c->rep_prefix = REPE_PREFIX;
+			break;
+		default:
+			goto done_prefixes;
+		}
+
+		/* Any legacy prefix after a REX prefix nullifies its effect. */
+
+		c->rex_prefix = 0;
+	}
+
+done_prefixes:
+
+	/* REX prefix. */
+	if (c->rex_prefix)
+		if (c->rex_prefix & 8)
+			c->op_bytes = 8;	/* REX.W */
+
+	/* Opcode byte(s). */
+	c->d = opcode_table[c->b];
+	if (c->d == 0) {
+		/* Two-byte opcode? */
+		if (c->b == 0x0f) {
+			c->twobyte = 1;
+			c->b = insn_fetch(u8, 1, c->eip);
+			c->d = twobyte_table[c->b];
+		}
+
+		/* Unrecognised? */
+		if (c->d == 0) {
+			DPRINTF("Cannot emulate %02x\n", c->b);
+			return -1;
+		}
+	}
+
+	if (mode == X86EMUL_MODE_PROT64 && (c->d & Stack))
+		c->op_bytes = 8;
+
+	/* ModRM and SIB bytes. */
+	if (c->d & ModRM)
+		rc = decode_modrm(ctxt, ops);
+	else if (c->d & MemAbs)
+		rc = decode_abs(ctxt, ops);
+	if (rc)
+		goto done;
+
+	if (!c->override_base)
+		c->override_base = &ctxt->ds_base;
+	if (mode == X86EMUL_MODE_PROT64 &&
+	    c->override_base != &ctxt->fs_base &&
+	    c->override_base != &ctxt->gs_base)
+		c->override_base = NULL;
+
+	if (c->override_base)
+		c->modrm_ea += *c->override_base;
+
+	if (c->ad_bytes != 8)
+		c->modrm_ea = (u32)c->modrm_ea;
+	/*
+	 * Decode and fetch the source operand: register, memory
+	 * or immediate.
+	 */
+	switch (c->d & SrcMask) {
+	case SrcNone:
+		break;
+	case SrcReg:
+		decode_register_operand(&c->src, c, 0);
+		break;
+	case SrcMem16:
+		c->src.bytes = 2;
+		goto srcmem_common;
+	case SrcMem32:
+		c->src.bytes = 4;
+		goto srcmem_common;
+	case SrcMem:
+		c->src.bytes = (c->d & ByteOp) ? 1 :
+							   c->op_bytes;
+		/* Don't fetch the address for invlpg: it could be unmapped. */
+		if (c->twobyte && c->b == 0x01 && c->modrm_reg == 7)
+			break;
+	srcmem_common:
+		/*
+		 * For instructions with a ModR/M byte, switch to register
+		 * access if Mod = 3.
+		 */
+		if ((c->d & ModRM) && c->modrm_mod == 3) {
+			c->src.type = OP_REG;
+			break;
+		}
+		c->src.type = OP_MEM;
+		break;
+	case SrcImm:
+		c->src.type = OP_IMM;
+		c->src.ptr = (unsigned long *)c->eip;
+		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		if (c->src.bytes == 8)
+			c->src.bytes = 4;
+		/* NB. Immediates are sign-extended as necessary. */
+		switch (c->src.bytes) {
+		case 1:
+			c->src.val = insn_fetch(s8, 1, c->eip);
+			break;
+		case 2:
+			c->src.val = insn_fetch(s16, 2, c->eip);
+			break;
+		case 4:
+			c->src.val = insn_fetch(s32, 4, c->eip);
+			break;
+		}
+		break;
+	case SrcImmByte:
+		c->src.type = OP_IMM;
+		c->src.ptr = (unsigned long *)c->eip;
+		c->src.bytes = 1;
+		c->src.val = insn_fetch(s8, 1, c->eip);
+		break;
+	}
+
+	/* Decode and fetch the destination operand: register or memory. */
+	switch (c->d & DstMask) {
+	case ImplicitOps:
+		/* Special instructions do their own operand decoding. */
+		return 0;
+	case DstReg:
+		decode_register_operand(&c->dst, c,
+			 c->twobyte && (c->b == 0xb6 || c->b == 0xb7));
+		break;
+	case DstMem:
+		if ((c->d & ModRM) && c->modrm_mod == 3) {
+			c->dst.type = OP_REG;
+			break;
+		}
+		c->dst.type = OP_MEM;
+		break;
+	}
+
+done:
+	return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0;
+}
+
+static inline void emulate_push(struct x86_emulate_ctxt *ctxt)
+{
+	struct decode_cache *c = &ctxt->decode;
+
+	c->dst.type  = OP_MEM;
+	c->dst.bytes = c->op_bytes;
+	c->dst.val = c->src.val;
+	register_address_increment(c->regs[VCPU_REGS_RSP], -c->op_bytes);
+	c->dst.ptr = (void *) register_address(ctxt->ss_base,
+					       c->regs[VCPU_REGS_RSP]);
+}
+
+static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt,
+				struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc;
+
+	rc = ops->read_std(register_address(ctxt->ss_base,
+					    c->regs[VCPU_REGS_RSP]),
+			   &c->dst.val, c->dst.bytes, ctxt->vcpu);
+	if (rc != 0)
+		return rc;
+
+	register_address_increment(c->regs[VCPU_REGS_RSP], c->dst.bytes);
+
+	return 0;
+}
+
+static inline void emulate_grp2(struct x86_emulate_ctxt *ctxt)
+{
+	struct decode_cache *c = &ctxt->decode;
+	switch (c->modrm_reg) {
+	case 0:	/* rol */
+		emulate_2op_SrcB("rol", c->src, c->dst, ctxt->eflags);
+		break;
+	case 1:	/* ror */
+		emulate_2op_SrcB("ror", c->src, c->dst, ctxt->eflags);
+		break;
+	case 2:	/* rcl */
+		emulate_2op_SrcB("rcl", c->src, c->dst, ctxt->eflags);
+		break;
+	case 3:	/* rcr */
+		emulate_2op_SrcB("rcr", c->src, c->dst, ctxt->eflags);
+		break;
+	case 4:	/* sal/shl */
+	case 6:	/* sal/shl */
+		emulate_2op_SrcB("sal", c->src, c->dst, ctxt->eflags);
+		break;
+	case 5:	/* shr */
+		emulate_2op_SrcB("shr", c->src, c->dst, ctxt->eflags);
+		break;
+	case 7:	/* sar */
+		emulate_2op_SrcB("sar", c->src, c->dst, ctxt->eflags);
+		break;
+	}
+}
+
+static inline int emulate_grp3(struct x86_emulate_ctxt *ctxt,
+			       struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc = 0;
+
+	switch (c->modrm_reg) {
+	case 0 ... 1:	/* test */
+		/*
+		 * Special case in Grp3: test has an immediate
+		 * source operand.
+		 */
+		c->src.type = OP_IMM;
+		c->src.ptr = (unsigned long *)c->eip;
+		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		if (c->src.bytes == 8)
+			c->src.bytes = 4;
+		switch (c->src.bytes) {
+		case 1:
+			c->src.val = insn_fetch(s8, 1, c->eip);
+			break;
+		case 2:
+			c->src.val = insn_fetch(s16, 2, c->eip);
+			break;
+		case 4:
+			c->src.val = insn_fetch(s32, 4, c->eip);
+			break;
+		}
+		emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
+		break;
+	case 2:	/* not */
+		c->dst.val = ~c->dst.val;
+		break;
+	case 3:	/* neg */
+		emulate_1op("neg", c->dst, ctxt->eflags);
+		break;
+	default:
+		DPRINTF("Cannot emulate %02x\n", c->b);
+		rc = X86EMUL_UNHANDLEABLE;
+		break;
+	}
+done:
+	return rc;
+}
+
+static inline int emulate_grp45(struct x86_emulate_ctxt *ctxt,
+			       struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc;
+
+	switch (c->modrm_reg) {
+	case 0:	/* inc */
+		emulate_1op("inc", c->dst, ctxt->eflags);
+		break;
+	case 1:	/* dec */
+		emulate_1op("dec", c->dst, ctxt->eflags);
+		break;
+	case 4: /* jmp abs */
+		if (c->b == 0xff)
+			c->eip = c->dst.val;
+		else {
+			DPRINTF("Cannot emulate %02x\n", c->b);
+			return X86EMUL_UNHANDLEABLE;
+		}
+		break;
+	case 6:	/* push */
+
+		/* 64-bit mode: PUSH always pushes a 64-bit operand. */
+
+		if (ctxt->mode == X86EMUL_MODE_PROT64) {
+			c->dst.bytes = 8;
+			rc = ops->read_std((unsigned long)c->dst.ptr,
+					   &c->dst.val, 8, ctxt->vcpu);
+			if (rc != 0)
+				return rc;
+		}
+		register_address_increment(c->regs[VCPU_REGS_RSP],
+					   -c->dst.bytes);
+		rc = ops->write_emulated(register_address(ctxt->ss_base,
+				    c->regs[VCPU_REGS_RSP]), &c->dst.val,
+				    c->dst.bytes, ctxt->vcpu);
+		if (rc != 0)
+			return rc;
+		c->dst.type = OP_NONE;
+		break;
+	default:
+		DPRINTF("Cannot emulate %02x\n", c->b);
+		return X86EMUL_UNHANDLEABLE;
+	}
+	return 0;
+}
+
+static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt,
+			       struct x86_emulate_ops *ops,
+			       unsigned long memop)
+{
+	struct decode_cache *c = &ctxt->decode;
+	u64 old, new;
+	int rc;
+
+	rc = ops->read_emulated(memop, &old, 8, ctxt->vcpu);
+	if (rc != 0)
+		return rc;
+
+	if (((u32) (old >> 0) != (u32) c->regs[VCPU_REGS_RAX]) ||
+	    ((u32) (old >> 32) != (u32) c->regs[VCPU_REGS_RDX])) {
+
+		c->regs[VCPU_REGS_RAX] = (u32) (old >> 0);
+		c->regs[VCPU_REGS_RDX] = (u32) (old >> 32);
+		ctxt->eflags &= ~EFLG_ZF;
+
+	} else {
+		new = ((u64)c->regs[VCPU_REGS_RCX] << 32) |
+		       (u32) c->regs[VCPU_REGS_RBX];
+
+		rc = ops->cmpxchg_emulated(memop, &old, &new, 8, ctxt->vcpu);
+		if (rc != 0)
+			return rc;
+		ctxt->eflags |= EFLG_ZF;
+	}
+	return 0;
+}
+
+static inline int writeback(struct x86_emulate_ctxt *ctxt,
+			    struct x86_emulate_ops *ops)
+{
+	int rc;
+	struct decode_cache *c = &ctxt->decode;
+
+	switch (c->dst.type) {
+	case OP_REG:
+		/* The 4-byte case *is* correct:
+		 * in 64-bit mode we zero-extend.
+		 */
+		switch (c->dst.bytes) {
+		case 1:
+			*(u8 *)c->dst.ptr = (u8)c->dst.val;
+			break;
+		case 2:
+			*(u16 *)c->dst.ptr = (u16)c->dst.val;
+			break;
+		case 4:
+			*c->dst.ptr = (u32)c->dst.val;
+			break;	/* 64b: zero-ext */
+		case 8:
+			*c->dst.ptr = c->dst.val;
+			break;
+		}
+		break;
+	case OP_MEM:
+		if (c->lock_prefix)
+			rc = ops->cmpxchg_emulated(
+					(unsigned long)c->dst.ptr,
+					&c->dst.orig_val,
+					&c->dst.val,
+					c->dst.bytes,
+					ctxt->vcpu);
+		else
+			rc = ops->write_emulated(
+					(unsigned long)c->dst.ptr,
+					&c->dst.val,
+					c->dst.bytes,
+					ctxt->vcpu);
+		if (rc != 0)
+			return rc;
+		break;
+	case OP_NONE:
+		/* no writeback */
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+int
+x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
+{
+	unsigned long memop = 0;
+	u64 msr_data;
+	unsigned long saved_eip = 0;
+	struct decode_cache *c = &ctxt->decode;
+	int rc = 0;
+
+	/* Shadow copy of register state. Committed on successful emulation.
+	 * NOTE: we can copy them from vcpu as x86_decode_insn() doesn't
+	 * modify them.
+	 */
+
+	memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
+	saved_eip = c->eip;
+
+	if (((c->d & ModRM) && (c->modrm_mod != 3)) || (c->d & MemAbs))
+		memop = c->modrm_ea;
+
+	if (c->rep_prefix && (c->d & String)) {
+		/* All REP prefixes have the same first termination condition */
+		if (c->regs[VCPU_REGS_RCX] == 0) {
+			ctxt->vcpu->arch.rip = c->eip;
+			goto done;
+		}
+		/* The second termination condition only applies for REPE
+		 * and REPNE. Test if the repeat string operation prefix is
+		 * REPE/REPZ or REPNE/REPNZ and if it's the case it tests the
+		 * corresponding termination condition according to:
+		 * 	- if REPE/REPZ and ZF = 0 then done
+		 * 	- if REPNE/REPNZ and ZF = 1 then done
+		 */
+		if ((c->b == 0xa6) || (c->b == 0xa7) ||
+				(c->b == 0xae) || (c->b == 0xaf)) {
+			if ((c->rep_prefix == REPE_PREFIX) &&
+				((ctxt->eflags & EFLG_ZF) == 0)) {
+					ctxt->vcpu->arch.rip = c->eip;
+					goto done;
+			}
+			if ((c->rep_prefix == REPNE_PREFIX) &&
+				((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) {
+				ctxt->vcpu->arch.rip = c->eip;
+				goto done;
+			}
+		}
+		c->regs[VCPU_REGS_RCX]--;
+		c->eip = ctxt->vcpu->arch.rip;
+	}
+
+	if (c->src.type == OP_MEM) {
+		c->src.ptr = (unsigned long *)memop;
+		c->src.val = 0;
+		rc = ops->read_emulated((unsigned long)c->src.ptr,
+					&c->src.val,
+					c->src.bytes,
+					ctxt->vcpu);
+		if (rc != 0)
+			goto done;
+		c->src.orig_val = c->src.val;
+	}
+
+	if ((c->d & DstMask) == ImplicitOps)
+		goto special_insn;
+
+
+	if (c->dst.type == OP_MEM) {
+		c->dst.ptr = (unsigned long *)memop;
+		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->dst.val = 0;
+		if (c->d & BitOp) {
+			unsigned long mask = ~(c->dst.bytes * 8 - 1);
+
+			c->dst.ptr = (void *)c->dst.ptr +
+						   (c->src.val & mask) / 8;
+		}
+		if (!(c->d & Mov) &&
+				   /* optimisation - avoid slow emulated read */
+		    ((rc = ops->read_emulated((unsigned long)c->dst.ptr,
+					   &c->dst.val,
+					  c->dst.bytes, ctxt->vcpu)) != 0))
+			goto done;
+	}
+	c->dst.orig_val = c->dst.val;
+
+special_insn:
+
+	if (c->twobyte)
+		goto twobyte_insn;
+
+	switch (c->b) {
+	case 0x00 ... 0x05:
+	      add:		/* add */
+		emulate_2op_SrcV("add", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x08 ... 0x0d:
+	      or:		/* or */
+		emulate_2op_SrcV("or", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x10 ... 0x15:
+	      adc:		/* adc */
+		emulate_2op_SrcV("adc", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x18 ... 0x1d:
+	      sbb:		/* sbb */
+		emulate_2op_SrcV("sbb", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x20 ... 0x23:
+	      and:		/* and */
+		emulate_2op_SrcV("and", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x24:              /* and al imm8 */
+		c->dst.type = OP_REG;
+		c->dst.ptr = &c->regs[VCPU_REGS_RAX];
+		c->dst.val = *(u8 *)c->dst.ptr;
+		c->dst.bytes = 1;
+		c->dst.orig_val = c->dst.val;
+		goto and;
+	case 0x25:              /* and ax imm16, or eax imm32 */
+		c->dst.type = OP_REG;
+		c->dst.bytes = c->op_bytes;
+		c->dst.ptr = &c->regs[VCPU_REGS_RAX];
+		if (c->op_bytes == 2)
+			c->dst.val = *(u16 *)c->dst.ptr;
+		else
+			c->dst.val = *(u32 *)c->dst.ptr;
+		c->dst.orig_val = c->dst.val;
+		goto and;
+	case 0x28 ... 0x2d:
+	      sub:		/* sub */
+		emulate_2op_SrcV("sub", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x30 ... 0x35:
+	      xor:		/* xor */
+		emulate_2op_SrcV("xor", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x38 ... 0x3d:
+	      cmp:		/* cmp */
+		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x40 ... 0x47: /* inc r16/r32 */
+		emulate_1op("inc", c->dst, ctxt->eflags);
+		break;
+	case 0x48 ... 0x4f: /* dec r16/r32 */
+		emulate_1op("dec", c->dst, ctxt->eflags);
+		break;
+	case 0x50 ... 0x57:  /* push reg */
+		c->dst.type  = OP_MEM;
+		c->dst.bytes = c->op_bytes;
+		c->dst.val = c->src.val;
+		register_address_increment(c->regs[VCPU_REGS_RSP],
+					   -c->op_bytes);
+		c->dst.ptr = (void *) register_address(
+			ctxt->ss_base, c->regs[VCPU_REGS_RSP]);
+		break;
+	case 0x58 ... 0x5f: /* pop reg */
+	pop_instruction:
+		if ((rc = ops->read_std(register_address(ctxt->ss_base,
+			c->regs[VCPU_REGS_RSP]), c->dst.ptr,
+			c->op_bytes, ctxt->vcpu)) != 0)
+			goto done;
+
+		register_address_increment(c->regs[VCPU_REGS_RSP],
+					   c->op_bytes);
+		c->dst.type = OP_NONE;	/* Disable writeback. */
+		break;
+	case 0x63:		/* movsxd */
+		if (ctxt->mode != X86EMUL_MODE_PROT64)
+			goto cannot_emulate;
+		c->dst.val = (s32) c->src.val;
+		break;
+	case 0x6a: /* push imm8 */
+		c->src.val = 0L;
+		c->src.val = insn_fetch(s8, 1, c->eip);
+		emulate_push(ctxt);
+		break;
+	case 0x6c:		/* insb */
+	case 0x6d:		/* insw/insd */
+		 if (kvm_emulate_pio_string(ctxt->vcpu, NULL,
+				1,
+				(c->d & ByteOp) ? 1 : c->op_bytes,
+				c->rep_prefix ?
+				address_mask(c->regs[VCPU_REGS_RCX]) : 1,
+				(ctxt->eflags & EFLG_DF),
+				register_address(ctxt->es_base,
+						 c->regs[VCPU_REGS_RDI]),
+				c->rep_prefix,
+				c->regs[VCPU_REGS_RDX]) == 0) {
+			c->eip = saved_eip;
+			return -1;
+		}
+		return 0;
+	case 0x6e:		/* outsb */
+	case 0x6f:		/* outsw/outsd */
+		if (kvm_emulate_pio_string(ctxt->vcpu, NULL,
+				0,
+				(c->d & ByteOp) ? 1 : c->op_bytes,
+				c->rep_prefix ?
+				address_mask(c->regs[VCPU_REGS_RCX]) : 1,
+				(ctxt->eflags & EFLG_DF),
+				register_address(c->override_base ?
+							*c->override_base :
+							ctxt->ds_base,
+						 c->regs[VCPU_REGS_RSI]),
+				c->rep_prefix,
+				c->regs[VCPU_REGS_RDX]) == 0) {
+			c->eip = saved_eip;
+			return -1;
+		}
+		return 0;
+	case 0x70 ... 0x7f: /* jcc (short) */ {
+		int rel = insn_fetch(s8, 1, c->eip);
+
+		if (test_cc(c->b, ctxt->eflags))
+			JMP_REL(rel);
+		break;
+	}
+	case 0x80 ... 0x83:	/* Grp1 */
+		switch (c->modrm_reg) {
+		case 0:
+			goto add;
+		case 1:
+			goto or;
+		case 2:
+			goto adc;
+		case 3:
+			goto sbb;
+		case 4:
+			goto and;
+		case 5:
+			goto sub;
+		case 6:
+			goto xor;
+		case 7:
+			goto cmp;
+		}
+		break;
+	case 0x84 ... 0x85:
+		emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0x86 ... 0x87:	/* xchg */
+		/* Write back the register source. */
+		switch (c->dst.bytes) {
+		case 1:
+			*(u8 *) c->src.ptr = (u8) c->dst.val;
+			break;
+		case 2:
+			*(u16 *) c->src.ptr = (u16) c->dst.val;
+			break;
+		case 4:
+			*c->src.ptr = (u32) c->dst.val;
+			break;	/* 64b reg: zero-extend */
+		case 8:
+			*c->src.ptr = c->dst.val;
+			break;
+		}
+		/*
+		 * Write back the memory destination with implicit LOCK
+		 * prefix.
+		 */
+		c->dst.val = c->src.val;
+		c->lock_prefix = 1;
+		break;
+	case 0x88 ... 0x8b:	/* mov */
+		goto mov;
+	case 0x8d: /* lea r16/r32, m */
+		c->dst.val = c->modrm_val;
+		break;
+	case 0x8f:		/* pop (sole member of Grp1a) */
+		rc = emulate_grp1a(ctxt, ops);
+		if (rc != 0)
+			goto done;
+		break;
+	case 0x9c: /* pushf */
+		c->src.val =  (unsigned long) ctxt->eflags;
+		emulate_push(ctxt);
+		break;
+	case 0x9d: /* popf */
+		c->dst.ptr = (unsigned long *) &ctxt->eflags;
+		goto pop_instruction;
+	case 0xa0 ... 0xa1:	/* mov */
+		c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
+		c->dst.val = c->src.val;
+		break;
+	case 0xa2 ... 0xa3:	/* mov */
+		c->dst.val = (unsigned long)c->regs[VCPU_REGS_RAX];
+		break;
+	case 0xa4 ... 0xa5:	/* movs */
+		c->dst.type = OP_MEM;
+		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->dst.ptr = (unsigned long *)register_address(
+						   ctxt->es_base,
+						   c->regs[VCPU_REGS_RDI]);
+		if ((rc = ops->read_emulated(register_address(
+		      c->override_base ? *c->override_base :
+					ctxt->ds_base,
+					c->regs[VCPU_REGS_RSI]),
+					&c->dst.val,
+					c->dst.bytes, ctxt->vcpu)) != 0)
+			goto done;
+		register_address_increment(c->regs[VCPU_REGS_RSI],
+				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
+							   : c->dst.bytes);
+		register_address_increment(c->regs[VCPU_REGS_RDI],
+				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
+							   : c->dst.bytes);
+		break;
+	case 0xa6 ... 0xa7:	/* cmps */
+		c->src.type = OP_NONE; /* Disable writeback. */
+		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->src.ptr = (unsigned long *)register_address(
+				c->override_base ? *c->override_base :
+						   ctxt->ds_base,
+						   c->regs[VCPU_REGS_RSI]);
+		if ((rc = ops->read_emulated((unsigned long)c->src.ptr,
+						&c->src.val,
+						c->src.bytes,
+						ctxt->vcpu)) != 0)
+			goto done;
+
+		c->dst.type = OP_NONE; /* Disable writeback. */
+		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->dst.ptr = (unsigned long *)register_address(
+						   ctxt->es_base,
+						   c->regs[VCPU_REGS_RDI]);
+		if ((rc = ops->read_emulated((unsigned long)c->dst.ptr,
+						&c->dst.val,
+						c->dst.bytes,
+						ctxt->vcpu)) != 0)
+			goto done;
+
+		DPRINTF("cmps: mem1=0x%p mem2=0x%p\n", c->src.ptr, c->dst.ptr);
+
+		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
+
+		register_address_increment(c->regs[VCPU_REGS_RSI],
+				       (ctxt->eflags & EFLG_DF) ? -c->src.bytes
+								  : c->src.bytes);
+		register_address_increment(c->regs[VCPU_REGS_RDI],
+				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
+								  : c->dst.bytes);
+
+		break;
+	case 0xaa ... 0xab:	/* stos */
+		c->dst.type = OP_MEM;
+		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->dst.ptr = (unsigned long *)register_address(
+						   ctxt->es_base,
+						   c->regs[VCPU_REGS_RDI]);
+		c->dst.val = c->regs[VCPU_REGS_RAX];
+		register_address_increment(c->regs[VCPU_REGS_RDI],
+				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
+							   : c->dst.bytes);
+		break;
+	case 0xac ... 0xad:	/* lods */
+		c->dst.type = OP_REG;
+		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
+		c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
+		if ((rc = ops->read_emulated(register_address(
+				c->override_base ? *c->override_base :
+						   ctxt->ds_base,
+						 c->regs[VCPU_REGS_RSI]),
+						 &c->dst.val,
+						 c->dst.bytes,
+						 ctxt->vcpu)) != 0)
+			goto done;
+		register_address_increment(c->regs[VCPU_REGS_RSI],
+				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
+							   : c->dst.bytes);
+		break;
+	case 0xae ... 0xaf:	/* scas */
+		DPRINTF("Urk! I don't handle SCAS.\n");
+		goto cannot_emulate;
+	case 0xc0 ... 0xc1:
+		emulate_grp2(ctxt);
+		break;
+	case 0xc3: /* ret */
+		c->dst.ptr = &c->eip;
+		goto pop_instruction;
+	case 0xc6 ... 0xc7:	/* mov (sole member of Grp11) */
+	mov:
+		c->dst.val = c->src.val;
+		break;
+	case 0xd0 ... 0xd1:	/* Grp2 */
+		c->src.val = 1;
+		emulate_grp2(ctxt);
+		break;
+	case 0xd2 ... 0xd3:	/* Grp2 */
+		c->src.val = c->regs[VCPU_REGS_RCX];
+		emulate_grp2(ctxt);
+		break;
+	case 0xe8: /* call (near) */ {
+		long int rel;
+		switch (c->op_bytes) {
+		case 2:
+			rel = insn_fetch(s16, 2, c->eip);
+			break;
+		case 4:
+			rel = insn_fetch(s32, 4, c->eip);
+			break;
+		default:
+			DPRINTF("Call: Invalid op_bytes\n");
+			goto cannot_emulate;
+		}
+		c->src.val = (unsigned long) c->eip;
+		JMP_REL(rel);
+		c->op_bytes = c->ad_bytes;
+		emulate_push(ctxt);
+		break;
+	}
+	case 0xe9: /* jmp rel */
+	case 0xeb: /* jmp rel short */
+		JMP_REL(c->src.val);
+		c->dst.type = OP_NONE; /* Disable writeback. */
+		break;
+	case 0xf4:              /* hlt */
+		ctxt->vcpu->arch.halt_request = 1;
+		goto done;
+	case 0xf5:	/* cmc */
+		/* complement carry flag from eflags reg */
+		ctxt->eflags ^= EFLG_CF;
+		c->dst.type = OP_NONE;	/* Disable writeback. */
+		break;
+	case 0xf6 ... 0xf7:	/* Grp3 */
+		rc = emulate_grp3(ctxt, ops);
+		if (rc != 0)
+			goto done;
+		break;
+	case 0xf8: /* clc */
+		ctxt->eflags &= ~EFLG_CF;
+		c->dst.type = OP_NONE;	/* Disable writeback. */
+		break;
+	case 0xfa: /* cli */
+		ctxt->eflags &= ~X86_EFLAGS_IF;
+		c->dst.type = OP_NONE;	/* Disable writeback. */
+		break;
+	case 0xfb: /* sti */
+		ctxt->eflags |= X86_EFLAGS_IF;
+		c->dst.type = OP_NONE;	/* Disable writeback. */
+		break;
+	case 0xfe ... 0xff:	/* Grp4/Grp5 */
+		rc = emulate_grp45(ctxt, ops);
+		if (rc != 0)
+			goto done;
+		break;
+	}
+
+writeback:
+	rc = writeback(ctxt, ops);
+	if (rc != 0)
+		goto done;
+
+	/* Commit shadow register state. */
+	memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs);
+	ctxt->vcpu->arch.rip = c->eip;
+
+done:
+	if (rc == X86EMUL_UNHANDLEABLE) {
+		c->eip = saved_eip;
+		return -1;
+	}
+	return 0;
+
+twobyte_insn:
+	switch (c->b) {
+	case 0x01: /* lgdt, lidt, lmsw */
+		switch (c->modrm_reg) {
+			u16 size;
+			unsigned long address;
+
+		case 0: /* vmcall */
+			if (c->modrm_mod != 3 || c->modrm_rm != 1)
+				goto cannot_emulate;
+
+			rc = kvm_fix_hypercall(ctxt->vcpu);
+			if (rc)
+				goto done;
+
+			kvm_emulate_hypercall(ctxt->vcpu);
+			break;
+		case 2: /* lgdt */
+			rc = read_descriptor(ctxt, ops, c->src.ptr,
+					     &size, &address, c->op_bytes);
+			if (rc)
+				goto done;
+			realmode_lgdt(ctxt->vcpu, size, address);
+			break;
+		case 3: /* lidt/vmmcall */
+			if (c->modrm_mod == 3 && c->modrm_rm == 1) {
+				rc = kvm_fix_hypercall(ctxt->vcpu);
+				if (rc)
+					goto done;
+				kvm_emulate_hypercall(ctxt->vcpu);
+			} else {
+				rc = read_descriptor(ctxt, ops, c->src.ptr,
+						     &size, &address,
+						     c->op_bytes);
+				if (rc)
+					goto done;
+				realmode_lidt(ctxt->vcpu, size, address);
+			}
+			break;
+		case 4: /* smsw */
+			if (c->modrm_mod != 3)
+				goto cannot_emulate;
+			*(u16 *)&c->regs[c->modrm_rm]
+				= realmode_get_cr(ctxt->vcpu, 0);
+			break;
+		case 6: /* lmsw */
+			if (c->modrm_mod != 3)
+				goto cannot_emulate;
+			realmode_lmsw(ctxt->vcpu, (u16)c->modrm_val,
+						  &ctxt->eflags);
+			break;
+		case 7: /* invlpg*/
+			emulate_invlpg(ctxt->vcpu, memop);
+			break;
+		default:
+			goto cannot_emulate;
+		}
+		/* Disable writeback. */
+		c->dst.type = OP_NONE;
+		break;
+	case 0x06:
+		emulate_clts(ctxt->vcpu);
+		c->dst.type = OP_NONE;
+		break;
+	case 0x08:		/* invd */
+	case 0x09:		/* wbinvd */
+	case 0x0d:		/* GrpP (prefetch) */
+	case 0x18:		/* Grp16 (prefetch/nop) */
+		c->dst.type = OP_NONE;
+		break;
+	case 0x20: /* mov cr, reg */
+		if (c->modrm_mod != 3)
+			goto cannot_emulate;
+		c->regs[c->modrm_rm] =
+				realmode_get_cr(ctxt->vcpu, c->modrm_reg);
+		c->dst.type = OP_NONE;	/* no writeback */
+		break;
+	case 0x21: /* mov from dr to reg */
+		if (c->modrm_mod != 3)
+			goto cannot_emulate;
+		rc = emulator_get_dr(ctxt, c->modrm_reg, &c->regs[c->modrm_rm]);
+		if (rc)
+			goto cannot_emulate;
+		c->dst.type = OP_NONE;	/* no writeback */
+		break;
+	case 0x22: /* mov reg, cr */
+		if (c->modrm_mod != 3)
+			goto cannot_emulate;
+		realmode_set_cr(ctxt->vcpu,
+				c->modrm_reg, c->modrm_val, &ctxt->eflags);
+		c->dst.type = OP_NONE;
+		break;
+	case 0x23: /* mov from reg to dr */
+		if (c->modrm_mod != 3)
+			goto cannot_emulate;
+		rc = emulator_set_dr(ctxt, c->modrm_reg,
+				     c->regs[c->modrm_rm]);
+		if (rc)
+			goto cannot_emulate;
+		c->dst.type = OP_NONE;	/* no writeback */
+		break;
+	case 0x30:
+		/* wrmsr */
+		msr_data = (u32)c->regs[VCPU_REGS_RAX]
+			| ((u64)c->regs[VCPU_REGS_RDX] << 32);
+		rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data);
+		if (rc) {
+			kvm_inject_gp(ctxt->vcpu, 0);
+			c->eip = ctxt->vcpu->arch.rip;
+		}
+		rc = X86EMUL_CONTINUE;
+		c->dst.type = OP_NONE;
+		break;
+	case 0x32:
+		/* rdmsr */
+		rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data);
+		if (rc) {
+			kvm_inject_gp(ctxt->vcpu, 0);
+			c->eip = ctxt->vcpu->arch.rip;
+		} else {
+			c->regs[VCPU_REGS_RAX] = (u32)msr_data;
+			c->regs[VCPU_REGS_RDX] = msr_data >> 32;
+		}
+		rc = X86EMUL_CONTINUE;
+		c->dst.type = OP_NONE;
+		break;
+	case 0x40 ... 0x4f:	/* cmov */
+		c->dst.val = c->dst.orig_val = c->src.val;
+		if (!test_cc(c->b, ctxt->eflags))
+			c->dst.type = OP_NONE; /* no writeback */
+		break;
+	case 0x80 ... 0x8f: /* jnz rel, etc*/ {
+		long int rel;
+
+		switch (c->op_bytes) {
+		case 2:
+			rel = insn_fetch(s16, 2, c->eip);
+			break;
+		case 4:
+			rel = insn_fetch(s32, 4, c->eip);
+			break;
+		case 8:
+			rel = insn_fetch(s64, 8, c->eip);
+			break;
+		default:
+			DPRINTF("jnz: Invalid op_bytes\n");
+			goto cannot_emulate;
+		}
+		if (test_cc(c->b, ctxt->eflags))
+			JMP_REL(rel);
+		c->dst.type = OP_NONE;
+		break;
+	}
+	case 0xa3:
+	      bt:		/* bt */
+		c->dst.type = OP_NONE;
+		/* only subword offset */
+		c->src.val &= (c->dst.bytes << 3) - 1;
+		emulate_2op_SrcV_nobyte("bt", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0xab:
+	      bts:		/* bts */
+		/* only subword offset */
+		c->src.val &= (c->dst.bytes << 3) - 1;
+		emulate_2op_SrcV_nobyte("bts", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0xb0 ... 0xb1:	/* cmpxchg */
+		/*
+		 * Save real source value, then compare EAX against
+		 * destination.
+		 */
+		c->src.orig_val = c->src.val;
+		c->src.val = c->regs[VCPU_REGS_RAX];
+		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
+		if (ctxt->eflags & EFLG_ZF) {
+			/* Success: write back to memory. */
+			c->dst.val = c->src.orig_val;
+		} else {
+			/* Failure: write the value we saw to EAX. */
+			c->dst.type = OP_REG;
+			c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
+		}
+		break;
+	case 0xb3:
+	      btr:		/* btr */
+		/* only subword offset */
+		c->src.val &= (c->dst.bytes << 3) - 1;
+		emulate_2op_SrcV_nobyte("btr", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0xb6 ... 0xb7:	/* movzx */
+		c->dst.bytes = c->op_bytes;
+		c->dst.val = (c->d & ByteOp) ? (u8) c->src.val
+						       : (u16) c->src.val;
+		break;
+	case 0xba:		/* Grp8 */
+		switch (c->modrm_reg & 3) {
+		case 0:
+			goto bt;
+		case 1:
+			goto bts;
+		case 2:
+			goto btr;
+		case 3:
+			goto btc;
+		}
+		break;
+	case 0xbb:
+	      btc:		/* btc */
+		/* only subword offset */
+		c->src.val &= (c->dst.bytes << 3) - 1;
+		emulate_2op_SrcV_nobyte("btc", c->src, c->dst, ctxt->eflags);
+		break;
+	case 0xbe ... 0xbf:	/* movsx */
+		c->dst.bytes = c->op_bytes;
+		c->dst.val = (c->d & ByteOp) ? (s8) c->src.val :
+							(s16) c->src.val;
+		break;
+	case 0xc3:		/* movnti */
+		c->dst.bytes = c->op_bytes;
+		c->dst.val = (c->op_bytes == 4) ? (u32) c->src.val :
+							(u64) c->src.val;
+		break;
+	case 0xc7:		/* Grp9 (cmpxchg8b) */
+		rc = emulate_grp9(ctxt, ops, memop);
+		if (rc != 0)
+			goto done;
+		c->dst.type = OP_NONE;
+		break;
+	}
+	goto writeback;
+
+cannot_emulate:
+	DPRINTF("Cannot emulate %02x\n", c->b);
+	c->eip = saved_eip;
+	return -1;
+}
diff --git a/drivers/Kconfig b/drivers/Kconfig
index f4076d9e9902..08d4ae201597 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -90,8 +90,6 @@ source "drivers/dca/Kconfig"
 
 source "drivers/auxdisplay/Kconfig"
 
-source "drivers/kvm/Kconfig"
-
 source "drivers/uio/Kconfig"
 
 source "drivers/virtio/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index d92d4d82d001..9e1f808e43cf 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -47,7 +47,6 @@ obj-$(CONFIG_SPI)		+= spi/
 obj-$(CONFIG_PCCARD)		+= pcmcia/
 obj-$(CONFIG_DIO)		+= dio/
 obj-$(CONFIG_SBUS)		+= sbus/
-obj-$(CONFIG_KVM)		+= kvm/
 obj-$(CONFIG_ZORRO)		+= zorro/
 obj-$(CONFIG_MAC)		+= macintosh/
 obj-$(CONFIG_ATA_OVER_ETH)	+= block/aoe/
diff --git a/drivers/kvm/Kconfig b/drivers/kvm/Kconfig
deleted file mode 100644
index c83e1c9b5129..000000000000
--- a/drivers/kvm/Kconfig
+++ /dev/null
@@ -1,57 +0,0 @@
-#
-# KVM configuration
-#
-config HAVE_KVM
-       bool
-
-menuconfig VIRTUALIZATION
-	bool "Virtualization"
-	depends on HAVE_KVM || X86
-	default y
-	---help---
-	  Say Y here to get to see options for using your Linux host to run other
-	  operating systems inside virtual machines (guests).
-	  This option alone does not add any kernel code.
-
-	  If you say N, all options in this submenu will be skipped and disabled.
-
-if VIRTUALIZATION
-
-config KVM
-	tristate "Kernel-based Virtual Machine (KVM) support"
-	depends on HAVE_KVM && EXPERIMENTAL
-	select PREEMPT_NOTIFIERS
-	select ANON_INODES
-	---help---
-	  Support hosting fully virtualized guest machines using hardware
-	  virtualization extensions.  You will need a fairly recent
-	  processor equipped with virtualization extensions. You will also
-	  need to select one or more of the processor modules below.
-
-	  This module provides access to the hardware capabilities through
-	  a character device node named /dev/kvm.
-
-	  To compile this as a module, choose M here: the module
-	  will be called kvm.
-
-	  If unsure, say N.
-
-config KVM_INTEL
-	tristate "KVM for Intel processors support"
-	depends on KVM
-	---help---
-	  Provides support for KVM on Intel processors equipped with the VT
-	  extensions.
-
-config KVM_AMD
-	tristate "KVM for AMD processors support"
-	depends on KVM
-	---help---
-	  Provides support for KVM on AMD processors equipped with the AMD-V
-	  (SVM) extensions.
-
-# OK, it's a little counter-intuitive to do this, but it puts it neatly under
-# the virtualization menu.
-source drivers/lguest/Kconfig
-
-endif # VIRTUALIZATION
diff --git a/drivers/kvm/Makefile b/drivers/kvm/Makefile
deleted file mode 100644
index cf18ad46e987..000000000000
--- a/drivers/kvm/Makefile
+++ /dev/null
@@ -1,10 +0,0 @@
-#
-# Makefile for Kernel-based Virtual Machine module
-#
-
-kvm-objs := kvm_main.o x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o ioapic.o
-obj-$(CONFIG_KVM) += kvm.o
-kvm-intel-objs = vmx.o
-obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
-kvm-amd-objs = svm.o
-obj-$(CONFIG_KVM_AMD) += kvm-amd.o
diff --git a/drivers/kvm/i8259.c b/drivers/kvm/i8259.c
deleted file mode 100644
index b3cad632f3d5..000000000000
--- a/drivers/kvm/i8259.c
+++ /dev/null
@@ -1,449 +0,0 @@
-/*
- * 8259 interrupt controller emulation
- *
- * Copyright (c) 2003-2004 Fabrice Bellard
- * Copyright (c) 2007 Intel Corporation
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this software and associated documentation files (the "Software"), to deal
- * in the Software without restriction, including without limitation the rights
- * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
- * copies of the Software, and to permit persons to whom the Software is
- * furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
- * THE SOFTWARE.
- * Authors:
- *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
- *   Port from Qemu.
- */
-#include <linux/mm.h>
-#include "irq.h"
-#include "kvm.h"
-
-/*
- * set irq level. If an edge is detected, then the IRR is set to 1
- */
-static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level)
-{
-	int mask;
-	mask = 1 << irq;
-	if (s->elcr & mask)	/* level triggered */
-		if (level) {
-			s->irr |= mask;
-			s->last_irr |= mask;
-		} else {
-			s->irr &= ~mask;
-			s->last_irr &= ~mask;
-		}
-	else	/* edge triggered */
-		if (level) {
-			if ((s->last_irr & mask) == 0)
-				s->irr |= mask;
-			s->last_irr |= mask;
-		} else
-			s->last_irr &= ~mask;
-}
-
-/*
- * return the highest priority found in mask (highest = smallest
- * number). Return 8 if no irq
- */
-static inline int get_priority(struct kvm_kpic_state *s, int mask)
-{
-	int priority;
-	if (mask == 0)
-		return 8;
-	priority = 0;
-	while ((mask & (1 << ((priority + s->priority_add) & 7))) == 0)
-		priority++;
-	return priority;
-}
-
-/*
- * return the pic wanted interrupt. return -1 if none
- */
-static int pic_get_irq(struct kvm_kpic_state *s)
-{
-	int mask, cur_priority, priority;
-
-	mask = s->irr & ~s->imr;
-	priority = get_priority(s, mask);
-	if (priority == 8)
-		return -1;
-	/*
-	 * compute current priority. If special fully nested mode on the
-	 * master, the IRQ coming from the slave is not taken into account
-	 * for the priority computation.
-	 */
-	mask = s->isr;
-	if (s->special_fully_nested_mode && s == &s->pics_state->pics[0])
-		mask &= ~(1 << 2);
-	cur_priority = get_priority(s, mask);
-	if (priority < cur_priority)
-		/*
-		 * higher priority found: an irq should be generated
-		 */
-		return (priority + s->priority_add) & 7;
-	else
-		return -1;
-}
-
-/*
- * raise irq to CPU if necessary. must be called every time the active
- * irq may change
- */
-static void pic_update_irq(struct kvm_pic *s)
-{
-	int irq2, irq;
-
-	irq2 = pic_get_irq(&s->pics[1]);
-	if (irq2 >= 0) {
-		/*
-		 * if irq request by slave pic, signal master PIC
-		 */
-		pic_set_irq1(&s->pics[0], 2, 1);
-		pic_set_irq1(&s->pics[0], 2, 0);
-	}
-	irq = pic_get_irq(&s->pics[0]);
-	if (irq >= 0)
-		s->irq_request(s->irq_request_opaque, 1);
-	else
-		s->irq_request(s->irq_request_opaque, 0);
-}
-
-void kvm_pic_update_irq(struct kvm_pic *s)
-{
-	pic_update_irq(s);
-}
-
-void kvm_pic_set_irq(void *opaque, int irq, int level)
-{
-	struct kvm_pic *s = opaque;
-
-	pic_set_irq1(&s->pics[irq >> 3], irq & 7, level);
-	pic_update_irq(s);
-}
-
-/*
- * acknowledge interrupt 'irq'
- */
-static inline void pic_intack(struct kvm_kpic_state *s, int irq)
-{
-	if (s->auto_eoi) {
-		if (s->rotate_on_auto_eoi)
-			s->priority_add = (irq + 1) & 7;
-	} else
-		s->isr |= (1 << irq);
-	/*
-	 * We don't clear a level sensitive interrupt here
-	 */
-	if (!(s->elcr & (1 << irq)))
-		s->irr &= ~(1 << irq);
-}
-
-int kvm_pic_read_irq(struct kvm_pic *s)
-{
-	int irq, irq2, intno;
-
-	irq = pic_get_irq(&s->pics[0]);
-	if (irq >= 0) {
-		pic_intack(&s->pics[0], irq);
-		if (irq == 2) {
-			irq2 = pic_get_irq(&s->pics[1]);
-			if (irq2 >= 0)
-				pic_intack(&s->pics[1], irq2);
-			else
-				/*
-				 * spurious IRQ on slave controller
-				 */
-				irq2 = 7;
-			intno = s->pics[1].irq_base + irq2;
-			irq = irq2 + 8;
-		} else
-			intno = s->pics[0].irq_base + irq;
-	} else {
-		/*
-		 * spurious IRQ on host controller
-		 */
-		irq = 7;
-		intno = s->pics[0].irq_base + irq;
-	}
-	pic_update_irq(s);
-
-	return intno;
-}
-
-void kvm_pic_reset(struct kvm_kpic_state *s)
-{
-	s->last_irr = 0;
-	s->irr = 0;
-	s->imr = 0;
-	s->isr = 0;
-	s->priority_add = 0;
-	s->irq_base = 0;
-	s->read_reg_select = 0;
-	s->poll = 0;
-	s->special_mask = 0;
-	s->init_state = 0;
-	s->auto_eoi = 0;
-	s->rotate_on_auto_eoi = 0;
-	s->special_fully_nested_mode = 0;
-	s->init4 = 0;
-}
-
-static void pic_ioport_write(void *opaque, u32 addr, u32 val)
-{
-	struct kvm_kpic_state *s = opaque;
-	int priority, cmd, irq;
-
-	addr &= 1;
-	if (addr == 0) {
-		if (val & 0x10) {
-			kvm_pic_reset(s);	/* init */
-			/*
-			 * deassert a pending interrupt
-			 */
-			s->pics_state->irq_request(s->pics_state->
-						   irq_request_opaque, 0);
-			s->init_state = 1;
-			s->init4 = val & 1;
-			if (val & 0x02)
-				printk(KERN_ERR "single mode not supported");
-			if (val & 0x08)
-				printk(KERN_ERR
-				       "level sensitive irq not supported");
-		} else if (val & 0x08) {
-			if (val & 0x04)
-				s->poll = 1;
-			if (val & 0x02)
-				s->read_reg_select = val & 1;
-			if (val & 0x40)
-				s->special_mask = (val >> 5) & 1;
-		} else {
-			cmd = val >> 5;
-			switch (cmd) {
-			case 0:
-			case 4:
-				s->rotate_on_auto_eoi = cmd >> 2;
-				break;
-			case 1:	/* end of interrupt */
-			case 5:
-				priority = get_priority(s, s->isr);
-				if (priority != 8) {
-					irq = (priority + s->priority_add) & 7;
-					s->isr &= ~(1 << irq);
-					if (cmd == 5)
-						s->priority_add = (irq + 1) & 7;
-					pic_update_irq(s->pics_state);
-				}
-				break;
-			case 3:
-				irq = val & 7;
-				s->isr &= ~(1 << irq);
-				pic_update_irq(s->pics_state);
-				break;
-			case 6:
-				s->priority_add = (val + 1) & 7;
-				pic_update_irq(s->pics_state);
-				break;
-			case 7:
-				irq = val & 7;
-				s->isr &= ~(1 << irq);
-				s->priority_add = (irq + 1) & 7;
-				pic_update_irq(s->pics_state);
-				break;
-			default:
-				break;	/* no operation */
-			}
-		}
-	} else
-		switch (s->init_state) {
-		case 0:		/* normal mode */
-			s->imr = val;
-			pic_update_irq(s->pics_state);
-			break;
-		case 1:
-			s->irq_base = val & 0xf8;
-			s->init_state = 2;
-			break;
-		case 2:
-			if (s->init4)
-				s->init_state = 3;
-			else
-				s->init_state = 0;
-			break;
-		case 3:
-			s->special_fully_nested_mode = (val >> 4) & 1;
-			s->auto_eoi = (val >> 1) & 1;
-			s->init_state = 0;
-			break;
-		}
-}
-
-static u32 pic_poll_read(struct kvm_kpic_state *s, u32 addr1)
-{
-	int ret;
-
-	ret = pic_get_irq(s);
-	if (ret >= 0) {
-		if (addr1 >> 7) {
-			s->pics_state->pics[0].isr &= ~(1 << 2);
-			s->pics_state->pics[0].irr &= ~(1 << 2);
-		}
-		s->irr &= ~(1 << ret);
-		s->isr &= ~(1 << ret);
-		if (addr1 >> 7 || ret != 2)
-			pic_update_irq(s->pics_state);
-	} else {
-		ret = 0x07;
-		pic_update_irq(s->pics_state);
-	}
-
-	return ret;
-}
-
-static u32 pic_ioport_read(void *opaque, u32 addr1)
-{
-	struct kvm_kpic_state *s = opaque;
-	unsigned int addr;
-	int ret;
-
-	addr = addr1;
-	addr &= 1;
-	if (s->poll) {
-		ret = pic_poll_read(s, addr1);
-		s->poll = 0;
-	} else
-		if (addr == 0)
-			if (s->read_reg_select)
-				ret = s->isr;
-			else
-				ret = s->irr;
-		else
-			ret = s->imr;
-	return ret;
-}
-
-static void elcr_ioport_write(void *opaque, u32 addr, u32 val)
-{
-	struct kvm_kpic_state *s = opaque;
-	s->elcr = val & s->elcr_mask;
-}
-
-static u32 elcr_ioport_read(void *opaque, u32 addr1)
-{
-	struct kvm_kpic_state *s = opaque;
-	return s->elcr;
-}
-
-static int picdev_in_range(struct kvm_io_device *this, gpa_t addr)
-{
-	switch (addr) {
-	case 0x20:
-	case 0x21:
-	case 0xa0:
-	case 0xa1:
-	case 0x4d0:
-	case 0x4d1:
-		return 1;
-	default:
-		return 0;
-	}
-}
-
-static void picdev_write(struct kvm_io_device *this,
-			 gpa_t addr, int len, const void *val)
-{
-	struct kvm_pic *s = this->private;
-	unsigned char data = *(unsigned char *)val;
-
-	if (len != 1) {
-		if (printk_ratelimit())
-			printk(KERN_ERR "PIC: non byte write\n");
-		return;
-	}
-	switch (addr) {
-	case 0x20:
-	case 0x21:
-	case 0xa0:
-	case 0xa1:
-		pic_ioport_write(&s->pics[addr >> 7], addr, data);
-		break;
-	case 0x4d0:
-	case 0x4d1:
-		elcr_ioport_write(&s->pics[addr & 1], addr, data);
-		break;
-	}
-}
-
-static void picdev_read(struct kvm_io_device *this,
-			gpa_t addr, int len, void *val)
-{
-	struct kvm_pic *s = this->private;
-	unsigned char data = 0;
-
-	if (len != 1) {
-		if (printk_ratelimit())
-			printk(KERN_ERR "PIC: non byte read\n");
-		return;
-	}
-	switch (addr) {
-	case 0x20:
-	case 0x21:
-	case 0xa0:
-	case 0xa1:
-		data = pic_ioport_read(&s->pics[addr >> 7], addr);
-		break;
-	case 0x4d0:
-	case 0x4d1:
-		data = elcr_ioport_read(&s->pics[addr & 1], addr);
-		break;
-	}
-	*(unsigned char *)val = data;
-}
-
-/*
- * callback when PIC0 irq status changed
- */
-static void pic_irq_request(void *opaque, int level)
-{
-	struct kvm *kvm = opaque;
-	struct kvm_vcpu *vcpu = kvm->vcpus[0];
-
-	pic_irqchip(kvm)->output = level;
-	if (vcpu)
-		kvm_vcpu_kick(vcpu);
-}
-
-struct kvm_pic *kvm_create_pic(struct kvm *kvm)
-{
-	struct kvm_pic *s;
-	s = kzalloc(sizeof(struct kvm_pic), GFP_KERNEL);
-	if (!s)
-		return NULL;
-	s->pics[0].elcr_mask = 0xf8;
-	s->pics[1].elcr_mask = 0xde;
-	s->irq_request = pic_irq_request;
-	s->irq_request_opaque = kvm;
-	s->pics[0].pics_state = s;
-	s->pics[1].pics_state = s;
-
-	/*
-	 * Initialize PIO device
-	 */
-	s->dev.read = picdev_read;
-	s->dev.write = picdev_write;
-	s->dev.in_range = picdev_in_range;
-	s->dev.private = s;
-	kvm_io_bus_register_dev(&kvm->pio_bus, &s->dev);
-	return s;
-}
diff --git a/drivers/kvm/ioapic.c b/drivers/kvm/ioapic.c
deleted file mode 100644
index f8236774c1b4..000000000000
--- a/drivers/kvm/ioapic.c
+++ /dev/null
@@ -1,402 +0,0 @@
-/*
- *  Copyright (C) 2001  MandrakeSoft S.A.
- *
- *    MandrakeSoft S.A.
- *    43, rue d'Aboukir
- *    75002 Paris - France
- *    http://www.linux-mandrake.com/
- *    http://www.mandrakesoft.com/
- *
- *  This library is free software; you can redistribute it and/or
- *  modify it under the terms of the GNU Lesser General Public
- *  License as published by the Free Software Foundation; either
- *  version 2 of the License, or (at your option) any later version.
- *
- *  This library is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- *  Lesser General Public License for more details.
- *
- *  You should have received a copy of the GNU Lesser General Public
- *  License along with this library; if not, write to the Free Software
- *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- *
- *  Yunhong Jiang <yunhong.jiang@intel.com>
- *  Yaozu (Eddie) Dong <eddie.dong@intel.com>
- *  Based on Xen 3.1 code.
- */
-
-#include "kvm.h"
-#include "x86.h"
-
-#include <linux/kvm.h>
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/smp.h>
-#include <linux/hrtimer.h>
-#include <linux/io.h>
-#include <asm/processor.h>
-#include <asm/page.h>
-#include <asm/current.h>
-#include "irq.h"
-#if 0
-#define ioapic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg)
-#else
-#define ioapic_debug(fmt, arg...)
-#endif
-static void ioapic_deliver(struct kvm_ioapic *vioapic, int irq);
-
-static unsigned long ioapic_read_indirect(struct kvm_ioapic *ioapic,
-					  unsigned long addr,
-					  unsigned long length)
-{
-	unsigned long result = 0;
-
-	switch (ioapic->ioregsel) {
-	case IOAPIC_REG_VERSION:
-		result = ((((IOAPIC_NUM_PINS - 1) & 0xff) << 16)
-			  | (IOAPIC_VERSION_ID & 0xff));
-		break;
-
-	case IOAPIC_REG_APIC_ID:
-	case IOAPIC_REG_ARB_ID:
-		result = ((ioapic->id & 0xf) << 24);
-		break;
-
-	default:
-		{
-			u32 redir_index = (ioapic->ioregsel - 0x10) >> 1;
-			u64 redir_content;
-
-			ASSERT(redir_index < IOAPIC_NUM_PINS);
-
-			redir_content = ioapic->redirtbl[redir_index].bits;
-			result = (ioapic->ioregsel & 0x1) ?
-			    (redir_content >> 32) & 0xffffffff :
-			    redir_content & 0xffffffff;
-			break;
-		}
-	}
-
-	return result;
-}
-
-static void ioapic_service(struct kvm_ioapic *ioapic, unsigned int idx)
-{
-	union ioapic_redir_entry *pent;
-
-	pent = &ioapic->redirtbl[idx];
-
-	if (!pent->fields.mask) {
-		ioapic_deliver(ioapic, idx);
-		if (pent->fields.trig_mode == IOAPIC_LEVEL_TRIG)
-			pent->fields.remote_irr = 1;
-	}
-	if (!pent->fields.trig_mode)
-		ioapic->irr &= ~(1 << idx);
-}
-
-static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val)
-{
-	unsigned index;
-
-	switch (ioapic->ioregsel) {
-	case IOAPIC_REG_VERSION:
-		/* Writes are ignored. */
-		break;
-
-	case IOAPIC_REG_APIC_ID:
-		ioapic->id = (val >> 24) & 0xf;
-		break;
-
-	case IOAPIC_REG_ARB_ID:
-		break;
-
-	default:
-		index = (ioapic->ioregsel - 0x10) >> 1;
-
-		ioapic_debug("change redir index %x val %x\n", index, val);
-		if (index >= IOAPIC_NUM_PINS)
-			return;
-		if (ioapic->ioregsel & 1) {
-			ioapic->redirtbl[index].bits &= 0xffffffff;
-			ioapic->redirtbl[index].bits |= (u64) val << 32;
-		} else {
-			ioapic->redirtbl[index].bits &= ~0xffffffffULL;
-			ioapic->redirtbl[index].bits |= (u32) val;
-			ioapic->redirtbl[index].fields.remote_irr = 0;
-		}
-		if (ioapic->irr & (1 << index))
-			ioapic_service(ioapic, index);
-		break;
-	}
-}
-
-static void ioapic_inj_irq(struct kvm_ioapic *ioapic,
-			   struct kvm_vcpu *vcpu,
-			   u8 vector, u8 trig_mode, u8 delivery_mode)
-{
-	ioapic_debug("irq %d trig %d deliv %d\n", vector, trig_mode,
-		     delivery_mode);
-
-	ASSERT((delivery_mode == IOAPIC_FIXED) ||
-	       (delivery_mode == IOAPIC_LOWEST_PRIORITY));
-
-	kvm_apic_set_irq(vcpu, vector, trig_mode);
-}
-
-static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
-				       u8 dest_mode)
-{
-	u32 mask = 0;
-	int i;
-	struct kvm *kvm = ioapic->kvm;
-	struct kvm_vcpu *vcpu;
-
-	ioapic_debug("dest %d dest_mode %d\n", dest, dest_mode);
-
-	if (dest_mode == 0) {	/* Physical mode. */
-		if (dest == 0xFF) {	/* Broadcast. */
-			for (i = 0; i < KVM_MAX_VCPUS; ++i)
-				if (kvm->vcpus[i] && kvm->vcpus[i]->arch.apic)
-					mask |= 1 << i;
-			return mask;
-		}
-		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-			vcpu = kvm->vcpus[i];
-			if (!vcpu)
-				continue;
-			if (kvm_apic_match_physical_addr(vcpu->arch.apic, dest)) {
-				if (vcpu->arch.apic)
-					mask = 1 << i;
-				break;
-			}
-		}
-	} else if (dest != 0)	/* Logical mode, MDA non-zero. */
-		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-			vcpu = kvm->vcpus[i];
-			if (!vcpu)
-				continue;
-			if (vcpu->arch.apic &&
-			    kvm_apic_match_logical_addr(vcpu->arch.apic, dest))
-				mask |= 1 << vcpu->vcpu_id;
-		}
-	ioapic_debug("mask %x\n", mask);
-	return mask;
-}
-
-static void ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
-{
-	u8 dest = ioapic->redirtbl[irq].fields.dest_id;
-	u8 dest_mode = ioapic->redirtbl[irq].fields.dest_mode;
-	u8 delivery_mode = ioapic->redirtbl[irq].fields.delivery_mode;
-	u8 vector = ioapic->redirtbl[irq].fields.vector;
-	u8 trig_mode = ioapic->redirtbl[irq].fields.trig_mode;
-	u32 deliver_bitmask;
-	struct kvm_vcpu *vcpu;
-	int vcpu_id;
-
-	ioapic_debug("dest=%x dest_mode=%x delivery_mode=%x "
-		     "vector=%x trig_mode=%x\n",
-		     dest, dest_mode, delivery_mode, vector, trig_mode);
-
-	deliver_bitmask = ioapic_get_delivery_bitmask(ioapic, dest, dest_mode);
-	if (!deliver_bitmask) {
-		ioapic_debug("no target on destination\n");
-		return;
-	}
-
-	switch (delivery_mode) {
-	case IOAPIC_LOWEST_PRIORITY:
-		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm, vector,
-				deliver_bitmask);
-		if (vcpu != NULL)
-			ioapic_inj_irq(ioapic, vcpu, vector,
-				       trig_mode, delivery_mode);
-		else
-			ioapic_debug("null lowest prio vcpu: "
-				     "mask=%x vector=%x delivery_mode=%x\n",
-				     deliver_bitmask, vector, IOAPIC_LOWEST_PRIORITY);
-		break;
-	case IOAPIC_FIXED:
-		for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) {
-			if (!(deliver_bitmask & (1 << vcpu_id)))
-				continue;
-			deliver_bitmask &= ~(1 << vcpu_id);
-			vcpu = ioapic->kvm->vcpus[vcpu_id];
-			if (vcpu) {
-				ioapic_inj_irq(ioapic, vcpu, vector,
-					       trig_mode, delivery_mode);
-			}
-		}
-		break;
-
-		/* TODO: NMI */
-	default:
-		printk(KERN_WARNING "Unsupported delivery mode %d\n",
-		       delivery_mode);
-		break;
-	}
-}
-
-void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
-{
-	u32 old_irr = ioapic->irr;
-	u32 mask = 1 << irq;
-	union ioapic_redir_entry entry;
-
-	if (irq >= 0 && irq < IOAPIC_NUM_PINS) {
-		entry = ioapic->redirtbl[irq];
-		level ^= entry.fields.polarity;
-		if (!level)
-			ioapic->irr &= ~mask;
-		else {
-			ioapic->irr |= mask;
-			if ((!entry.fields.trig_mode && old_irr != ioapic->irr)
-			    || !entry.fields.remote_irr)
-				ioapic_service(ioapic, irq);
-		}
-	}
-}
-
-static int get_eoi_gsi(struct kvm_ioapic *ioapic, int vector)
-{
-	int i;
-
-	for (i = 0; i < IOAPIC_NUM_PINS; i++)
-		if (ioapic->redirtbl[i].fields.vector == vector)
-			return i;
-	return -1;
-}
-
-void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
-{
-	struct kvm_ioapic *ioapic = kvm->arch.vioapic;
-	union ioapic_redir_entry *ent;
-	int gsi;
-
-	gsi = get_eoi_gsi(ioapic, vector);
-	if (gsi == -1) {
-		printk(KERN_WARNING "Can't find redir item for %d EOI\n",
-		       vector);
-		return;
-	}
-
-	ent = &ioapic->redirtbl[gsi];
-	ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
-
-	ent->fields.remote_irr = 0;
-	if (!ent->fields.mask && (ioapic->irr & (1 << gsi)))
-		ioapic_deliver(ioapic, gsi);
-}
-
-static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr)
-{
-	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
-
-	return ((addr >= ioapic->base_address &&
-		 (addr < ioapic->base_address + IOAPIC_MEM_LENGTH)));
-}
-
-static void ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len,
-			     void *val)
-{
-	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
-	u32 result;
-
-	ioapic_debug("addr %lx\n", (unsigned long)addr);
-	ASSERT(!(addr & 0xf));	/* check alignment */
-
-	addr &= 0xff;
-	switch (addr) {
-	case IOAPIC_REG_SELECT:
-		result = ioapic->ioregsel;
-		break;
-
-	case IOAPIC_REG_WINDOW:
-		result = ioapic_read_indirect(ioapic, addr, len);
-		break;
-
-	default:
-		result = 0;
-		break;
-	}
-	switch (len) {
-	case 8:
-		*(u64 *) val = result;
-		break;
-	case 1:
-	case 2:
-	case 4:
-		memcpy(val, (char *)&result, len);
-		break;
-	default:
-		printk(KERN_WARNING "ioapic: wrong length %d\n", len);
-	}
-}
-
-static void ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len,
-			      const void *val)
-{
-	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
-	u32 data;
-
-	ioapic_debug("ioapic_mmio_write addr=%p len=%d val=%p\n",
-		     (void*)addr, len, val);
-	ASSERT(!(addr & 0xf));	/* check alignment */
-	if (len == 4 || len == 8)
-		data = *(u32 *) val;
-	else {
-		printk(KERN_WARNING "ioapic: Unsupported size %d\n", len);
-		return;
-	}
-
-	addr &= 0xff;
-	switch (addr) {
-	case IOAPIC_REG_SELECT:
-		ioapic->ioregsel = data;
-		break;
-
-	case IOAPIC_REG_WINDOW:
-		ioapic_write_indirect(ioapic, data);
-		break;
-#ifdef	CONFIG_IA64
-	case IOAPIC_REG_EOI:
-		kvm_ioapic_update_eoi(ioapic, data);
-		break;
-#endif
-
-	default:
-		break;
-	}
-}
-
-void kvm_ioapic_reset(struct kvm_ioapic *ioapic)
-{
-	int i;
-
-	for (i = 0; i < IOAPIC_NUM_PINS; i++)
-		ioapic->redirtbl[i].fields.mask = 1;
-	ioapic->base_address = IOAPIC_DEFAULT_BASE_ADDRESS;
-	ioapic->ioregsel = 0;
-	ioapic->irr = 0;
-	ioapic->id = 0;
-}
-
-int kvm_ioapic_init(struct kvm *kvm)
-{
-	struct kvm_ioapic *ioapic;
-
-	ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL);
-	if (!ioapic)
-		return -ENOMEM;
-	kvm->arch.vioapic = ioapic;
-	kvm_ioapic_reset(ioapic);
-	ioapic->dev.read = ioapic_mmio_read;
-	ioapic->dev.write = ioapic_mmio_write;
-	ioapic->dev.in_range = ioapic_in_range;
-	ioapic->dev.private = ioapic;
-	ioapic->kvm = kvm;
-	kvm_io_bus_register_dev(&kvm->mmio_bus, &ioapic->dev);
-	return 0;
-}
diff --git a/drivers/kvm/iodev.h b/drivers/kvm/iodev.h
index eb9e8a71843a..c14e642027b2 100644
--- a/drivers/kvm/iodev.h
+++ b/drivers/kvm/iodev.h
@@ -16,7 +16,7 @@
 #ifndef __KVM_IODEV_H__
 #define __KVM_IODEV_H__
 
-#include "types.h"
+#include <linux/kvm_types.h>
 
 struct kvm_io_device {
 	void (*read)(struct kvm_io_device *this,
diff --git a/drivers/kvm/irq.c b/drivers/kvm/irq.c
deleted file mode 100644
index 59b47c55fc76..000000000000
--- a/drivers/kvm/irq.c
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
- * irq.c: API for in kernel interrupt controller
- * Copyright (c) 2007, Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
- * Place - Suite 330, Boston, MA 02111-1307 USA.
- * Authors:
- *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
- *
- */
-
-#include <linux/module.h>
-
-#include "kvm.h"
-#include "x86.h"
-#include "irq.h"
-
-/*
- * check if there is pending interrupt without
- * intack.
- */
-int kvm_cpu_has_interrupt(struct kvm_vcpu *v)
-{
-	struct kvm_pic *s;
-
-	if (kvm_apic_has_interrupt(v) == -1) {	/* LAPIC */
-		if (kvm_apic_accept_pic_intr(v)) {
-			s = pic_irqchip(v->kvm);	/* PIC */
-			return s->output;
-		} else
-			return 0;
-	}
-	return 1;
-}
-EXPORT_SYMBOL_GPL(kvm_cpu_has_interrupt);
-
-/*
- * Read pending interrupt vector and intack.
- */
-int kvm_cpu_get_interrupt(struct kvm_vcpu *v)
-{
-	struct kvm_pic *s;
-	int vector;
-
-	vector = kvm_get_apic_interrupt(v);	/* APIC */
-	if (vector == -1) {
-		if (kvm_apic_accept_pic_intr(v)) {
-			s = pic_irqchip(v->kvm);
-			s->output = 0;		/* PIC */
-			vector = kvm_pic_read_irq(s);
-		}
-	}
-	return vector;
-}
-EXPORT_SYMBOL_GPL(kvm_cpu_get_interrupt);
-
-static void vcpu_kick_intr(void *info)
-{
-#ifdef DEBUG
-	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
-	printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
-#endif
-}
-
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
-{
-	int ipi_pcpu = vcpu->cpu;
-
-	if (waitqueue_active(&vcpu->wq)) {
-		wake_up_interruptible(&vcpu->wq);
-		++vcpu->stat.halt_wakeup;
-	}
-	if (vcpu->guest_mode)
-		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
-}
-
-void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu)
-{
-	kvm_inject_apic_timer_irqs(vcpu);
-	/* TODO: PIT, RTC etc. */
-}
-EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs);
-
-void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
-{
-	kvm_apic_timer_intr_post(vcpu, vec);
-	/* TODO: PIT, RTC etc. */
-}
-EXPORT_SYMBOL_GPL(kvm_timer_intr_post);
diff --git a/drivers/kvm/irq.h b/drivers/kvm/irq.h
deleted file mode 100644
index 6e023dc3f848..000000000000
--- a/drivers/kvm/irq.h
+++ /dev/null
@@ -1,196 +0,0 @@
-/*
- * irq.h: in kernel interrupt controller related definitions
- * Copyright (c) 2007, Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
- * Place - Suite 330, Boston, MA 02111-1307 USA.
- * Authors:
- *   Yaozu (Eddie) Dong <Eddie.dong@intel.com>
- *
- */
-
-#ifndef __IRQ_H
-#define __IRQ_H
-
-#include <linux/mm_types.h>
-#include <linux/hrtimer.h>
-#include <asm/kvm.h>
-#include "iodev.h"
-#include "kvm.h"
-
-struct kvm;
-struct kvm_vcpu;
-
-typedef void irq_request_func(void *opaque, int level);
-
-struct kvm_kpic_state {
-	u8 last_irr;	/* edge detection */
-	u8 irr;		/* interrupt request register */
-	u8 imr;		/* interrupt mask register */
-	u8 isr;		/* interrupt service register */
-	u8 priority_add;	/* highest irq priority */
-	u8 irq_base;
-	u8 read_reg_select;
-	u8 poll;
-	u8 special_mask;
-	u8 init_state;
-	u8 auto_eoi;
-	u8 rotate_on_auto_eoi;
-	u8 special_fully_nested_mode;
-	u8 init4;		/* true if 4 byte init */
-	u8 elcr;		/* PIIX edge/trigger selection */
-	u8 elcr_mask;
-	struct kvm_pic *pics_state;
-};
-
-struct kvm_pic {
-	struct kvm_kpic_state pics[2]; /* 0 is master pic, 1 is slave pic */
-	irq_request_func *irq_request;
-	void *irq_request_opaque;
-	int output;		/* intr from master PIC */
-	struct kvm_io_device dev;
-};
-
-struct kvm_pic *kvm_create_pic(struct kvm *kvm);
-void kvm_pic_set_irq(void *opaque, int irq, int level);
-int kvm_pic_read_irq(struct kvm_pic *s);
-void kvm_pic_update_irq(struct kvm_pic *s);
-
-#define IOAPIC_NUM_PINS  KVM_IOAPIC_NUM_PINS
-#define IOAPIC_VERSION_ID 0x11	/* IOAPIC version */
-#define IOAPIC_EDGE_TRIG  0
-#define IOAPIC_LEVEL_TRIG 1
-
-#define IOAPIC_DEFAULT_BASE_ADDRESS  0xfec00000
-#define IOAPIC_MEM_LENGTH            0x100
-
-/* Direct registers. */
-#define IOAPIC_REG_SELECT  0x00
-#define IOAPIC_REG_WINDOW  0x10
-#define IOAPIC_REG_EOI     0x40	/* IA64 IOSAPIC only */
-
-/* Indirect registers. */
-#define IOAPIC_REG_APIC_ID 0x00	/* x86 IOAPIC only */
-#define IOAPIC_REG_VERSION 0x01
-#define IOAPIC_REG_ARB_ID  0x02	/* x86 IOAPIC only */
-
-/*ioapic delivery mode*/
-#define	IOAPIC_FIXED			0x0
-#define	IOAPIC_LOWEST_PRIORITY		0x1
-#define	IOAPIC_PMI			0x2
-#define	IOAPIC_NMI			0x4
-#define	IOAPIC_INIT			0x5
-#define	IOAPIC_EXTINT			0x7
-
-struct kvm_ioapic {
-	u64 base_address;
-	u32 ioregsel;
-	u32 id;
-	u32 irr;
-	u32 pad;
-	union ioapic_redir_entry {
-		u64 bits;
-		struct {
-			u8 vector;
-			u8 delivery_mode:3;
-			u8 dest_mode:1;
-			u8 delivery_status:1;
-			u8 polarity:1;
-			u8 remote_irr:1;
-			u8 trig_mode:1;
-			u8 mask:1;
-			u8 reserve:7;
-			u8 reserved[4];
-			u8 dest_id;
-		} fields;
-	} redirtbl[IOAPIC_NUM_PINS];
-	struct kvm_io_device dev;
-	struct kvm *kvm;
-};
-
-struct kvm_lapic {
-	unsigned long base_address;
-	struct kvm_io_device dev;
-	struct {
-		atomic_t pending;
-		s64 period;	/* unit: ns */
-		u32 divide_count;
-		ktime_t last_update;
-		struct hrtimer dev;
-	} timer;
-	struct kvm_vcpu *vcpu;
-	struct page *regs_page;
-	void *regs;
-};
-
-#ifdef DEBUG
-#define ASSERT(x)  							\
-do {									\
-	if (!(x)) {							\
-		printk(KERN_EMERG "assertion failed %s: %d: %s\n",	\
-		       __FILE__, __LINE__, #x);				\
-		BUG();							\
-	}								\
-} while (0)
-#else
-#define ASSERT(x) do { } while (0)
-#endif
-
-static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
-{
-	return kvm->arch.vpic;
-}
-
-static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
-{
-	return kvm->arch.vioapic;
-}
-
-static inline int irqchip_in_kernel(struct kvm *kvm)
-{
-	return pic_irqchip(kvm) != NULL;
-}
-
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
-int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu);
-int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu);
-int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu);
-int kvm_create_lapic(struct kvm_vcpu *vcpu);
-void kvm_lapic_reset(struct kvm_vcpu *vcpu);
-void kvm_pic_reset(struct kvm_kpic_state *s);
-void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
-void kvm_free_lapic(struct kvm_vcpu *vcpu);
-u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu);
-void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8);
-void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value);
-
-struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
-				       unsigned long bitmap);
-u64 kvm_get_apic_base(struct kvm_vcpu *vcpu);
-void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data);
-int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest);
-void kvm_ioapic_update_eoi(struct kvm *kvm, int vector);
-int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda);
-int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig);
-void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu);
-int kvm_ioapic_init(struct kvm *kvm);
-void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
-int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
-int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
-void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
-void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
-void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
-void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
-void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
-
-#endif
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
deleted file mode 100644
index bf6a3b330a3d..000000000000
--- a/drivers/kvm/kvm.h
+++ /dev/null
@@ -1,289 +0,0 @@
-#ifndef __KVM_H
-#define __KVM_H
-
-/*
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- */
-
-#include <linux/types.h>
-#include <linux/hardirq.h>
-#include <linux/list.h>
-#include <linux/mutex.h>
-#include <linux/spinlock.h>
-#include <linux/signal.h>
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/preempt.h>
-#include <asm/signal.h>
-
-#include <linux/kvm.h>
-#include <linux/kvm_para.h>
-
-#include "types.h"
-
-#include "x86.h"
-
-#define KVM_MAX_VCPUS 4
-#define KVM_MEMORY_SLOTS 8
-/* memory slots that does not exposed to userspace */
-#define KVM_PRIVATE_MEM_SLOTS 4
-
-#define KVM_PIO_PAGE_OFFSET 1
-
-/*
- * vcpu->requests bit members
- */
-#define KVM_REQ_TLB_FLUSH          0
-
-
-struct kvm_vcpu;
-extern struct kmem_cache *kvm_vcpu_cache;
-
-struct kvm_guest_debug {
-	int enabled;
-	unsigned long bp[4];
-	int singlestep;
-};
-
-/*
- * It would be nice to use something smarter than a linear search, TBD...
- * Thankfully we dont expect many devices to register (famous last words :),
- * so until then it will suffice.  At least its abstracted so we can change
- * in one place.
- */
-struct kvm_io_bus {
-	int                   dev_count;
-#define NR_IOBUS_DEVS 6
-	struct kvm_io_device *devs[NR_IOBUS_DEVS];
-};
-
-void kvm_io_bus_init(struct kvm_io_bus *bus);
-void kvm_io_bus_destroy(struct kvm_io_bus *bus);
-struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr);
-void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
-			     struct kvm_io_device *dev);
-
-struct kvm_vcpu {
-	struct kvm *kvm;
-	struct preempt_notifier preempt_notifier;
-	int vcpu_id;
-	struct mutex mutex;
-	int   cpu;
-	struct kvm_run *run;
-	int guest_mode;
-	unsigned long requests;
-	struct kvm_guest_debug guest_debug;
-	int fpu_active;
-	int guest_fpu_loaded;
-	wait_queue_head_t wq;
-	int sigset_active;
-	sigset_t sigset;
-	struct kvm_vcpu_stat stat;
-
-#ifdef CONFIG_HAS_IOMEM
-	int mmio_needed;
-	int mmio_read_completed;
-	int mmio_is_write;
-	int mmio_size;
-	unsigned char mmio_data[8];
-	gpa_t mmio_phys_addr;
-#endif
-
-	struct kvm_vcpu_arch arch;
-};
-
-struct kvm_memory_slot {
-	gfn_t base_gfn;
-	unsigned long npages;
-	unsigned long flags;
-	unsigned long *rmap;
-	unsigned long *dirty_bitmap;
-	unsigned long userspace_addr;
-	int user_alloc;
-};
-
-struct kvm {
-	struct mutex lock; /* protects everything except vcpus */
-	struct mm_struct *mm; /* userspace tied to this vm */
-	int nmemslots;
-	struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
-					KVM_PRIVATE_MEM_SLOTS];
-	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
-	struct list_head vm_list;
-	struct file *filp;
-	struct kvm_io_bus mmio_bus;
-	struct kvm_io_bus pio_bus;
-	struct kvm_vm_stat stat;
-	struct kvm_arch arch;
-};
-
-/* The guest did something we don't support. */
-#define pr_unimpl(vcpu, fmt, ...)					\
- do {									\
-	if (printk_ratelimit())						\
-		printk(KERN_ERR "kvm: %i: cpu%i " fmt,			\
-		       current->tgid, (vcpu)->vcpu_id , ## __VA_ARGS__); \
- } while (0)
-
-#define kvm_printf(kvm, fmt ...) printk(KERN_DEBUG fmt)
-#define vcpu_printf(vcpu, fmt...) kvm_printf(vcpu->kvm, fmt)
-
-int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id);
-void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);
-
-void vcpu_load(struct kvm_vcpu *vcpu);
-void vcpu_put(struct kvm_vcpu *vcpu);
-
-void decache_vcpus_on_cpu(int cpu);
-
-
-int kvm_init(void *opaque, unsigned int vcpu_size,
-		  struct module *module);
-void kvm_exit(void);
-
-#define HPA_MSB ((sizeof(hpa_t) * 8) - 1)
-#define HPA_ERR_MASK ((hpa_t)1 << HPA_MSB)
-static inline int is_error_hpa(hpa_t hpa) { return hpa >> HPA_MSB; }
-struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva);
-
-extern struct page *bad_page;
-
-int is_error_page(struct page *page);
-int kvm_is_error_hva(unsigned long addr);
-int kvm_set_memory_region(struct kvm *kvm,
-			  struct kvm_userspace_memory_region *mem,
-			  int user_alloc);
-int __kvm_set_memory_region(struct kvm *kvm,
-			    struct kvm_userspace_memory_region *mem,
-			    int user_alloc);
-int kvm_arch_set_memory_region(struct kvm *kvm,
-				struct kvm_userspace_memory_region *mem,
-				struct kvm_memory_slot old,
-				int user_alloc);
-gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn);
-struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
-void kvm_release_page_clean(struct page *page);
-void kvm_release_page_dirty(struct page *page);
-int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
-			int len);
-int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
-int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
-			 int offset, int len);
-int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
-		    unsigned long len);
-int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len);
-int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len);
-struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn);
-int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn);
-void mark_page_dirty(struct kvm *kvm, gfn_t gfn);
-
-void kvm_vcpu_block(struct kvm_vcpu *vcpu);
-void kvm_resched(struct kvm_vcpu *vcpu);
-void kvm_load_guest_fpu(struct kvm_vcpu *vcpu);
-void kvm_put_guest_fpu(struct kvm_vcpu *vcpu);
-void kvm_flush_remote_tlbs(struct kvm *kvm);
-
-long kvm_arch_dev_ioctl(struct file *filp,
-			unsigned int ioctl, unsigned long arg);
-long kvm_arch_vcpu_ioctl(struct file *filp,
-			 unsigned int ioctl, unsigned long arg);
-void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
-void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
-
-int kvm_dev_ioctl_check_extension(long ext);
-
-int kvm_get_dirty_log(struct kvm *kvm,
-			struct kvm_dirty_log *log, int *is_dirty);
-int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
-				struct kvm_dirty_log *log);
-
-int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
-				   struct
-				   kvm_userspace_memory_region *mem,
-				   int user_alloc);
-long kvm_arch_vm_ioctl(struct file *filp,
-		       unsigned int ioctl, unsigned long arg);
-void kvm_arch_destroy_vm(struct kvm *kvm);
-
-int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
-int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
-
-int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
-				    struct kvm_translation *tr);
-
-int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
-int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
-int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs);
-int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs);
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg);
-int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
-
-int kvm_arch_init(void *opaque);
-void kvm_arch_exit(void);
-
-int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu);
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu);
-
-void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
-void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
-void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
-struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id);
-int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
-void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
-
-int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
-void kvm_arch_hardware_enable(void *garbage);
-void kvm_arch_hardware_disable(void *garbage);
-int kvm_arch_hardware_setup(void);
-void kvm_arch_hardware_unsetup(void);
-void kvm_arch_check_processor_compat(void *rtn);
-int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
-
-void kvm_free_physmem(struct kvm *kvm);
-
-struct  kvm *kvm_arch_create_vm(void);
-void kvm_arch_destroy_vm(struct kvm *kvm);
-
-int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
-int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
-
-static inline void kvm_guest_enter(void)
-{
-	account_system_vtime(current);
-	current->flags |= PF_VCPU;
-}
-
-static inline void kvm_guest_exit(void)
-{
-	account_system_vtime(current);
-	current->flags &= ~PF_VCPU;
-}
-
-static inline int memslot_id(struct kvm *kvm, struct kvm_memory_slot *slot)
-{
-	return slot - kvm->memslots;
-}
-
-static inline gpa_t gfn_to_gpa(gfn_t gfn)
-{
-	return (gpa_t)gfn << PAGE_SHIFT;
-}
-
-enum kvm_stat_kind {
-	KVM_STAT_VM,
-	KVM_STAT_VCPU,
-};
-
-struct kvm_stats_debugfs_item {
-	const char *name;
-	int offset;
-	enum kvm_stat_kind kind;
-	struct dentry *dentry;
-};
-extern struct kvm_stats_debugfs_item debugfs_entries[];
-
-#endif
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index ae2a1bf640bc..4026d7d64296 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -15,9 +15,9 @@
  *
  */
 
-#include "kvm.h"
 #include "iodev.h"
 
+#include <linux/kvm_host.h>
 #include <linux/kvm.h>
 #include <linux/module.h>
 #include <linux/errno.h>
diff --git a/drivers/kvm/kvm_svm.h b/drivers/kvm/kvm_svm.h
deleted file mode 100644
index a0e415daef5b..000000000000
--- a/drivers/kvm/kvm_svm.h
+++ /dev/null
@@ -1,45 +0,0 @@
-#ifndef __KVM_SVM_H
-#define __KVM_SVM_H
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/list.h>
-#include <asm/msr.h>
-
-#include "svm.h"
-#include "kvm.h"
-
-static const u32 host_save_user_msrs[] = {
-#ifdef CONFIG_X86_64
-	MSR_STAR, MSR_LSTAR, MSR_CSTAR, MSR_SYSCALL_MASK, MSR_KERNEL_GS_BASE,
-	MSR_FS_BASE,
-#endif
-	MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
-};
-
-#define NR_HOST_SAVE_USER_MSRS ARRAY_SIZE(host_save_user_msrs)
-#define NUM_DB_REGS 4
-
-struct kvm_vcpu;
-
-struct vcpu_svm {
-	struct kvm_vcpu vcpu;
-	struct vmcb *vmcb;
-	unsigned long vmcb_pa;
-	struct svm_cpu_data *svm_data;
-	uint64_t asid_generation;
-
-	unsigned long db_regs[NUM_DB_REGS];
-
-	u64 next_rip;
-
-	u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS];
-	u64 host_gs_base;
-	unsigned long host_cr2;
-	unsigned long host_db_regs[NUM_DB_REGS];
-	unsigned long host_dr6;
-	unsigned long host_dr7;
-};
-
-#endif
-
diff --git a/drivers/kvm/lapic.c b/drivers/kvm/lapic.c
deleted file mode 100644
index 8c74bf184a07..000000000000
--- a/drivers/kvm/lapic.c
+++ /dev/null
@@ -1,1087 +0,0 @@
-
-/*
- * Local APIC virtualization
- *
- * Copyright (C) 2006 Qumranet, Inc.
- * Copyright (C) 2007 Novell
- * Copyright (C) 2007 Intel
- *
- * Authors:
- *   Dor Laor <dor.laor@qumranet.com>
- *   Gregory Haskins <ghaskins@novell.com>
- *   Yaozu (Eddie) Dong <eddie.dong@intel.com>
- *
- * Based on Xen 3.1 code, Copyright (c) 2004, Intel Corporation.
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- */
-
-#include "kvm.h"
-#include "x86.h"
-
-#include <linux/kvm.h>
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/smp.h>
-#include <linux/hrtimer.h>
-#include <linux/io.h>
-#include <linux/module.h>
-#include <asm/processor.h>
-#include <asm/msr.h>
-#include <asm/page.h>
-#include <asm/current.h>
-#include <asm/apicdef.h>
-#include <asm/atomic.h>
-#include <asm/div64.h>
-#include "irq.h"
-
-#define PRId64 "d"
-#define PRIx64 "llx"
-#define PRIu64 "u"
-#define PRIo64 "o"
-
-#define APIC_BUS_CYCLE_NS 1
-
-/* #define apic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg) */
-#define apic_debug(fmt, arg...)
-
-#define APIC_LVT_NUM			6
-/* 14 is the version for Xeon and Pentium 8.4.8*/
-#define APIC_VERSION			(0x14UL | ((APIC_LVT_NUM - 1) << 16))
-#define LAPIC_MMIO_LENGTH		(1 << 12)
-/* followed define is not in apicdef.h */
-#define APIC_SHORT_MASK			0xc0000
-#define APIC_DEST_NOSHORT		0x0
-#define APIC_DEST_MASK			0x800
-#define MAX_APIC_VECTOR			256
-
-#define VEC_POS(v) ((v) & (32 - 1))
-#define REG_POS(v) (((v) >> 5) << 4)
-
-static inline u32 apic_get_reg(struct kvm_lapic *apic, int reg_off)
-{
-	return *((u32 *) (apic->regs + reg_off));
-}
-
-static inline void apic_set_reg(struct kvm_lapic *apic, int reg_off, u32 val)
-{
-	*((u32 *) (apic->regs + reg_off)) = val;
-}
-
-static inline int apic_test_and_set_vector(int vec, void *bitmap)
-{
-	return test_and_set_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
-}
-
-static inline int apic_test_and_clear_vector(int vec, void *bitmap)
-{
-	return test_and_clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
-}
-
-static inline void apic_set_vector(int vec, void *bitmap)
-{
-	set_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
-}
-
-static inline void apic_clear_vector(int vec, void *bitmap)
-{
-	clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
-}
-
-static inline int apic_hw_enabled(struct kvm_lapic *apic)
-{
-	return (apic)->vcpu->arch.apic_base & MSR_IA32_APICBASE_ENABLE;
-}
-
-static inline int  apic_sw_enabled(struct kvm_lapic *apic)
-{
-	return apic_get_reg(apic, APIC_SPIV) & APIC_SPIV_APIC_ENABLED;
-}
-
-static inline int apic_enabled(struct kvm_lapic *apic)
-{
-	return apic_sw_enabled(apic) &&	apic_hw_enabled(apic);
-}
-
-#define LVT_MASK	\
-	(APIC_LVT_MASKED | APIC_SEND_PENDING | APIC_VECTOR_MASK)
-
-#define LINT_MASK	\
-	(LVT_MASK | APIC_MODE_MASK | APIC_INPUT_POLARITY | \
-	 APIC_LVT_REMOTE_IRR | APIC_LVT_LEVEL_TRIGGER)
-
-static inline int kvm_apic_id(struct kvm_lapic *apic)
-{
-	return (apic_get_reg(apic, APIC_ID) >> 24) & 0xff;
-}
-
-static inline int apic_lvt_enabled(struct kvm_lapic *apic, int lvt_type)
-{
-	return !(apic_get_reg(apic, lvt_type) & APIC_LVT_MASKED);
-}
-
-static inline int apic_lvt_vector(struct kvm_lapic *apic, int lvt_type)
-{
-	return apic_get_reg(apic, lvt_type) & APIC_VECTOR_MASK;
-}
-
-static inline int apic_lvtt_period(struct kvm_lapic *apic)
-{
-	return apic_get_reg(apic, APIC_LVTT) & APIC_LVT_TIMER_PERIODIC;
-}
-
-static unsigned int apic_lvt_mask[APIC_LVT_NUM] = {
-	LVT_MASK | APIC_LVT_TIMER_PERIODIC,	/* LVTT */
-	LVT_MASK | APIC_MODE_MASK,	/* LVTTHMR */
-	LVT_MASK | APIC_MODE_MASK,	/* LVTPC */
-	LINT_MASK, LINT_MASK,	/* LVT0-1 */
-	LVT_MASK		/* LVTERR */
-};
-
-static int find_highest_vector(void *bitmap)
-{
-	u32 *word = bitmap;
-	int word_offset = MAX_APIC_VECTOR >> 5;
-
-	while ((word_offset != 0) && (word[(--word_offset) << 2] == 0))
-		continue;
-
-	if (likely(!word_offset && !word[0]))
-		return -1;
-	else
-		return fls(word[word_offset << 2]) - 1 + (word_offset << 5);
-}
-
-static inline int apic_test_and_set_irr(int vec, struct kvm_lapic *apic)
-{
-	return apic_test_and_set_vector(vec, apic->regs + APIC_IRR);
-}
-
-static inline void apic_clear_irr(int vec, struct kvm_lapic *apic)
-{
-	apic_clear_vector(vec, apic->regs + APIC_IRR);
-}
-
-static inline int apic_find_highest_irr(struct kvm_lapic *apic)
-{
-	int result;
-
-	result = find_highest_vector(apic->regs + APIC_IRR);
-	ASSERT(result == -1 || result >= 16);
-
-	return result;
-}
-
-int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-	int highest_irr;
-
-	if (!apic)
-		return 0;
-	highest_irr = apic_find_highest_irr(apic);
-
-	return highest_irr;
-}
-EXPORT_SYMBOL_GPL(kvm_lapic_find_highest_irr);
-
-int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (!apic_test_and_set_irr(vec, apic)) {
-		/* a new pending irq is set in IRR */
-		if (trig)
-			apic_set_vector(vec, apic->regs + APIC_TMR);
-		else
-			apic_clear_vector(vec, apic->regs + APIC_TMR);
-		kvm_vcpu_kick(apic->vcpu);
-		return 1;
-	}
-	return 0;
-}
-
-static inline int apic_find_highest_isr(struct kvm_lapic *apic)
-{
-	int result;
-
-	result = find_highest_vector(apic->regs + APIC_ISR);
-	ASSERT(result == -1 || result >= 16);
-
-	return result;
-}
-
-static void apic_update_ppr(struct kvm_lapic *apic)
-{
-	u32 tpr, isrv, ppr;
-	int isr;
-
-	tpr = apic_get_reg(apic, APIC_TASKPRI);
-	isr = apic_find_highest_isr(apic);
-	isrv = (isr != -1) ? isr : 0;
-
-	if ((tpr & 0xf0) >= (isrv & 0xf0))
-		ppr = tpr & 0xff;
-	else
-		ppr = isrv & 0xf0;
-
-	apic_debug("vlapic %p, ppr 0x%x, isr 0x%x, isrv 0x%x",
-		   apic, ppr, isr, isrv);
-
-	apic_set_reg(apic, APIC_PROCPRI, ppr);
-}
-
-static void apic_set_tpr(struct kvm_lapic *apic, u32 tpr)
-{
-	apic_set_reg(apic, APIC_TASKPRI, tpr);
-	apic_update_ppr(apic);
-}
-
-int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest)
-{
-	return kvm_apic_id(apic) == dest;
-}
-
-int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda)
-{
-	int result = 0;
-	u8 logical_id;
-
-	logical_id = GET_APIC_LOGICAL_ID(apic_get_reg(apic, APIC_LDR));
-
-	switch (apic_get_reg(apic, APIC_DFR)) {
-	case APIC_DFR_FLAT:
-		if (logical_id & mda)
-			result = 1;
-		break;
-	case APIC_DFR_CLUSTER:
-		if (((logical_id >> 4) == (mda >> 0x4))
-		    && (logical_id & mda & 0xf))
-			result = 1;
-		break;
-	default:
-		printk(KERN_WARNING "Bad DFR vcpu %d: %08x\n",
-		       apic->vcpu->vcpu_id, apic_get_reg(apic, APIC_DFR));
-		break;
-	}
-
-	return result;
-}
-
-static int apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source,
-			   int short_hand, int dest, int dest_mode)
-{
-	int result = 0;
-	struct kvm_lapic *target = vcpu->arch.apic;
-
-	apic_debug("target %p, source %p, dest 0x%x, "
-		   "dest_mode 0x%x, short_hand 0x%x",
-		   target, source, dest, dest_mode, short_hand);
-
-	ASSERT(!target);
-	switch (short_hand) {
-	case APIC_DEST_NOSHORT:
-		if (dest_mode == 0) {
-			/* Physical mode. */
-			if ((dest == 0xFF) || (dest == kvm_apic_id(target)))
-				result = 1;
-		} else
-			/* Logical mode. */
-			result = kvm_apic_match_logical_addr(target, dest);
-		break;
-	case APIC_DEST_SELF:
-		if (target == source)
-			result = 1;
-		break;
-	case APIC_DEST_ALLINC:
-		result = 1;
-		break;
-	case APIC_DEST_ALLBUT:
-		if (target != source)
-			result = 1;
-		break;
-	default:
-		printk(KERN_WARNING "Bad dest shorthand value %x\n",
-		       short_hand);
-		break;
-	}
-
-	return result;
-}
-
-/*
- * Add a pending IRQ into lapic.
- * Return 1 if successfully added and 0 if discarded.
- */
-static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
-			     int vector, int level, int trig_mode)
-{
-	int orig_irr, result = 0;
-	struct kvm_vcpu *vcpu = apic->vcpu;
-
-	switch (delivery_mode) {
-	case APIC_DM_FIXED:
-	case APIC_DM_LOWEST:
-		/* FIXME add logic for vcpu on reset */
-		if (unlikely(!apic_enabled(apic)))
-			break;
-
-		orig_irr = apic_test_and_set_irr(vector, apic);
-		if (orig_irr && trig_mode) {
-			apic_debug("level trig mode repeatedly for vector %d",
-				   vector);
-			break;
-		}
-
-		if (trig_mode) {
-			apic_debug("level trig mode for vector %d", vector);
-			apic_set_vector(vector, apic->regs + APIC_TMR);
-		} else
-			apic_clear_vector(vector, apic->regs + APIC_TMR);
-
-		if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
-			kvm_vcpu_kick(vcpu);
-		else if (vcpu->arch.mp_state == VCPU_MP_STATE_HALTED) {
-			vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
-			if (waitqueue_active(&vcpu->wq))
-				wake_up_interruptible(&vcpu->wq);
-		}
-
-		result = (orig_irr == 0);
-		break;
-
-	case APIC_DM_REMRD:
-		printk(KERN_DEBUG "Ignoring delivery mode 3\n");
-		break;
-
-	case APIC_DM_SMI:
-		printk(KERN_DEBUG "Ignoring guest SMI\n");
-		break;
-	case APIC_DM_NMI:
-		printk(KERN_DEBUG "Ignoring guest NMI\n");
-		break;
-
-	case APIC_DM_INIT:
-		if (level) {
-			if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
-				printk(KERN_DEBUG
-				       "INIT on a runnable vcpu %d\n",
-				       vcpu->vcpu_id);
-			vcpu->arch.mp_state = VCPU_MP_STATE_INIT_RECEIVED;
-			kvm_vcpu_kick(vcpu);
-		} else {
-			printk(KERN_DEBUG
-			       "Ignoring de-assert INIT to vcpu %d\n",
-			       vcpu->vcpu_id);
-		}
-
-		break;
-
-	case APIC_DM_STARTUP:
-		printk(KERN_DEBUG "SIPI to vcpu %d vector 0x%02x\n",
-		       vcpu->vcpu_id, vector);
-		if (vcpu->arch.mp_state == VCPU_MP_STATE_INIT_RECEIVED) {
-			vcpu->arch.sipi_vector = vector;
-			vcpu->arch.mp_state = VCPU_MP_STATE_SIPI_RECEIVED;
-			if (waitqueue_active(&vcpu->wq))
-				wake_up_interruptible(&vcpu->wq);
-		}
-		break;
-
-	default:
-		printk(KERN_ERR "TODO: unsupported delivery mode %x\n",
-		       delivery_mode);
-		break;
-	}
-	return result;
-}
-
-static struct kvm_lapic *kvm_apic_round_robin(struct kvm *kvm, u8 vector,
-				       unsigned long bitmap)
-{
-	int last;
-	int next;
-	struct kvm_lapic *apic = NULL;
-
-	last = kvm->arch.round_robin_prev_vcpu;
-	next = last;
-
-	do {
-		if (++next == KVM_MAX_VCPUS)
-			next = 0;
-		if (kvm->vcpus[next] == NULL || !test_bit(next, &bitmap))
-			continue;
-		apic = kvm->vcpus[next]->arch.apic;
-		if (apic && apic_enabled(apic))
-			break;
-		apic = NULL;
-	} while (next != last);
-	kvm->arch.round_robin_prev_vcpu = next;
-
-	if (!apic)
-		printk(KERN_DEBUG "vcpu not ready for apic_round_robin\n");
-
-	return apic;
-}
-
-struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
-		unsigned long bitmap)
-{
-	struct kvm_lapic *apic;
-
-	apic = kvm_apic_round_robin(kvm, vector, bitmap);
-	if (apic)
-		return apic->vcpu;
-	return NULL;
-}
-
-static void apic_set_eoi(struct kvm_lapic *apic)
-{
-	int vector = apic_find_highest_isr(apic);
-
-	/*
-	 * Not every write EOI will has corresponding ISR,
-	 * one example is when Kernel check timer on setup_IO_APIC
-	 */
-	if (vector == -1)
-		return;
-
-	apic_clear_vector(vector, apic->regs + APIC_ISR);
-	apic_update_ppr(apic);
-
-	if (apic_test_and_clear_vector(vector, apic->regs + APIC_TMR))
-		kvm_ioapic_update_eoi(apic->vcpu->kvm, vector);
-}
-
-static void apic_send_ipi(struct kvm_lapic *apic)
-{
-	u32 icr_low = apic_get_reg(apic, APIC_ICR);
-	u32 icr_high = apic_get_reg(apic, APIC_ICR2);
-
-	unsigned int dest = GET_APIC_DEST_FIELD(icr_high);
-	unsigned int short_hand = icr_low & APIC_SHORT_MASK;
-	unsigned int trig_mode = icr_low & APIC_INT_LEVELTRIG;
-	unsigned int level = icr_low & APIC_INT_ASSERT;
-	unsigned int dest_mode = icr_low & APIC_DEST_MASK;
-	unsigned int delivery_mode = icr_low & APIC_MODE_MASK;
-	unsigned int vector = icr_low & APIC_VECTOR_MASK;
-
-	struct kvm_vcpu *target;
-	struct kvm_vcpu *vcpu;
-	unsigned long lpr_map = 0;
-	int i;
-
-	apic_debug("icr_high 0x%x, icr_low 0x%x, "
-		   "short_hand 0x%x, dest 0x%x, trig_mode 0x%x, level 0x%x, "
-		   "dest_mode 0x%x, delivery_mode 0x%x, vector 0x%x\n",
-		   icr_high, icr_low, short_hand, dest,
-		   trig_mode, level, dest_mode, delivery_mode, vector);
-
-	for (i = 0; i < KVM_MAX_VCPUS; i++) {
-		vcpu = apic->vcpu->kvm->vcpus[i];
-		if (!vcpu)
-			continue;
-
-		if (vcpu->arch.apic &&
-		    apic_match_dest(vcpu, apic, short_hand, dest, dest_mode)) {
-			if (delivery_mode == APIC_DM_LOWEST)
-				set_bit(vcpu->vcpu_id, &lpr_map);
-			else
-				__apic_accept_irq(vcpu->arch.apic, delivery_mode,
-						  vector, level, trig_mode);
-		}
-	}
-
-	if (delivery_mode == APIC_DM_LOWEST) {
-		target = kvm_get_lowest_prio_vcpu(vcpu->kvm, vector, lpr_map);
-		if (target != NULL)
-			__apic_accept_irq(target->arch.apic, delivery_mode,
-					  vector, level, trig_mode);
-	}
-}
-
-static u32 apic_get_tmcct(struct kvm_lapic *apic)
-{
-	u64 counter_passed;
-	ktime_t passed, now;
-	u32 tmcct;
-
-	ASSERT(apic != NULL);
-
-	now = apic->timer.dev.base->get_time();
-	tmcct = apic_get_reg(apic, APIC_TMICT);
-
-	/* if initial count is 0, current count should also be 0 */
-	if (tmcct == 0)
-		return 0;
-
-	if (unlikely(ktime_to_ns(now) <=
-		ktime_to_ns(apic->timer.last_update))) {
-		/* Wrap around */
-		passed = ktime_add(( {
-				    (ktime_t) {
-				    .tv64 = KTIME_MAX -
-				    (apic->timer.last_update).tv64}; }
-				   ), now);
-		apic_debug("time elapsed\n");
-	} else
-		passed = ktime_sub(now, apic->timer.last_update);
-
-	counter_passed = div64_64(ktime_to_ns(passed),
-				  (APIC_BUS_CYCLE_NS * apic->timer.divide_count));
-
-	if (counter_passed > tmcct) {
-		if (unlikely(!apic_lvtt_period(apic))) {
-			/* one-shot timers stick at 0 until reset */
-			tmcct = 0;
-		} else {
-			/*
-			 * periodic timers reset to APIC_TMICT when they
-			 * hit 0. The while loop simulates this happening N
-			 * times. (counter_passed %= tmcct) would also work,
-			 * but might be slower or not work on 32-bit??
-			 */
-			while (counter_passed > tmcct)
-				counter_passed -= tmcct;
-			tmcct -= counter_passed;
-		}
-	} else {
-		tmcct -= counter_passed;
-	}
-
-	return tmcct;
-}
-
-static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
-{
-	u32 val = 0;
-
-	if (offset >= LAPIC_MMIO_LENGTH)
-		return 0;
-
-	switch (offset) {
-	case APIC_ARBPRI:
-		printk(KERN_WARNING "Access APIC ARBPRI register "
-		       "which is for P6\n");
-		break;
-
-	case APIC_TMCCT:	/* Timer CCR */
-		val = apic_get_tmcct(apic);
-		break;
-
-	default:
-		apic_update_ppr(apic);
-		val = apic_get_reg(apic, offset);
-		break;
-	}
-
-	return val;
-}
-
-static void apic_mmio_read(struct kvm_io_device *this,
-			   gpa_t address, int len, void *data)
-{
-	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
-	unsigned int offset = address - apic->base_address;
-	unsigned char alignment = offset & 0xf;
-	u32 result;
-
-	if ((alignment + len) > 4) {
-		printk(KERN_ERR "KVM_APIC_READ: alignment error %lx %d",
-		       (unsigned long)address, len);
-		return;
-	}
-	result = __apic_read(apic, offset & ~0xf);
-
-	switch (len) {
-	case 1:
-	case 2:
-	case 4:
-		memcpy(data, (char *)&result + alignment, len);
-		break;
-	default:
-		printk(KERN_ERR "Local APIC read with len = %x, "
-		       "should be 1,2, or 4 instead\n", len);
-		break;
-	}
-}
-
-static void update_divide_count(struct kvm_lapic *apic)
-{
-	u32 tmp1, tmp2, tdcr;
-
-	tdcr = apic_get_reg(apic, APIC_TDCR);
-	tmp1 = tdcr & 0xf;
-	tmp2 = ((tmp1 & 0x3) | ((tmp1 & 0x8) >> 1)) + 1;
-	apic->timer.divide_count = 0x1 << (tmp2 & 0x7);
-
-	apic_debug("timer divide count is 0x%x\n",
-				   apic->timer.divide_count);
-}
-
-static void start_apic_timer(struct kvm_lapic *apic)
-{
-	ktime_t now = apic->timer.dev.base->get_time();
-
-	apic->timer.last_update = now;
-
-	apic->timer.period = apic_get_reg(apic, APIC_TMICT) *
-		    APIC_BUS_CYCLE_NS * apic->timer.divide_count;
-	atomic_set(&apic->timer.pending, 0);
-	hrtimer_start(&apic->timer.dev,
-		      ktime_add_ns(now, apic->timer.period),
-		      HRTIMER_MODE_ABS);
-
-	apic_debug("%s: bus cycle is %" PRId64 "ns, now 0x%016"
-			   PRIx64 ", "
-			   "timer initial count 0x%x, period %lldns, "
-			   "expire @ 0x%016" PRIx64 ".\n", __FUNCTION__,
-			   APIC_BUS_CYCLE_NS, ktime_to_ns(now),
-			   apic_get_reg(apic, APIC_TMICT),
-			   apic->timer.period,
-			   ktime_to_ns(ktime_add_ns(now,
-					apic->timer.period)));
-}
-
-static void apic_mmio_write(struct kvm_io_device *this,
-			    gpa_t address, int len, const void *data)
-{
-	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
-	unsigned int offset = address - apic->base_address;
-	unsigned char alignment = offset & 0xf;
-	u32 val;
-
-	/*
-	 * APIC register must be aligned on 128-bits boundary.
-	 * 32/64/128 bits registers must be accessed thru 32 bits.
-	 * Refer SDM 8.4.1
-	 */
-	if (len != 4 || alignment) {
-		if (printk_ratelimit())
-			printk(KERN_ERR "apic write: bad size=%d %lx\n",
-			       len, (long)address);
-		return;
-	}
-
-	val = *(u32 *) data;
-
-	/* too common printing */
-	if (offset != APIC_EOI)
-		apic_debug("%s: offset 0x%x with length 0x%x, and value is "
-			   "0x%x\n", __FUNCTION__, offset, len, val);
-
-	offset &= 0xff0;
-
-	switch (offset) {
-	case APIC_ID:		/* Local APIC ID */
-		apic_set_reg(apic, APIC_ID, val);
-		break;
-
-	case APIC_TASKPRI:
-		apic_set_tpr(apic, val & 0xff);
-		break;
-
-	case APIC_EOI:
-		apic_set_eoi(apic);
-		break;
-
-	case APIC_LDR:
-		apic_set_reg(apic, APIC_LDR, val & APIC_LDR_MASK);
-		break;
-
-	case APIC_DFR:
-		apic_set_reg(apic, APIC_DFR, val | 0x0FFFFFFF);
-		break;
-
-	case APIC_SPIV:
-		apic_set_reg(apic, APIC_SPIV, val & 0x3ff);
-		if (!(val & APIC_SPIV_APIC_ENABLED)) {
-			int i;
-			u32 lvt_val;
-
-			for (i = 0; i < APIC_LVT_NUM; i++) {
-				lvt_val = apic_get_reg(apic,
-						       APIC_LVTT + 0x10 * i);
-				apic_set_reg(apic, APIC_LVTT + 0x10 * i,
-					     lvt_val | APIC_LVT_MASKED);
-			}
-			atomic_set(&apic->timer.pending, 0);
-
-		}
-		break;
-
-	case APIC_ICR:
-		/* No delay here, so we always clear the pending bit */
-		apic_set_reg(apic, APIC_ICR, val & ~(1 << 12));
-		apic_send_ipi(apic);
-		break;
-
-	case APIC_ICR2:
-		apic_set_reg(apic, APIC_ICR2, val & 0xff000000);
-		break;
-
-	case APIC_LVTT:
-	case APIC_LVTTHMR:
-	case APIC_LVTPC:
-	case APIC_LVT0:
-	case APIC_LVT1:
-	case APIC_LVTERR:
-		/* TODO: Check vector */
-		if (!apic_sw_enabled(apic))
-			val |= APIC_LVT_MASKED;
-
-		val &= apic_lvt_mask[(offset - APIC_LVTT) >> 4];
-		apic_set_reg(apic, offset, val);
-
-		break;
-
-	case APIC_TMICT:
-		hrtimer_cancel(&apic->timer.dev);
-		apic_set_reg(apic, APIC_TMICT, val);
-		start_apic_timer(apic);
-		return;
-
-	case APIC_TDCR:
-		if (val & 4)
-			printk(KERN_ERR "KVM_WRITE:TDCR %x\n", val);
-		apic_set_reg(apic, APIC_TDCR, val);
-		update_divide_count(apic);
-		break;
-
-	default:
-		apic_debug("Local APIC Write to read-only register %x\n",
-			   offset);
-		break;
-	}
-
-}
-
-static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr)
-{
-	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
-	int ret = 0;
-
-
-	if (apic_hw_enabled(apic) &&
-	    (addr >= apic->base_address) &&
-	    (addr < (apic->base_address + LAPIC_MMIO_LENGTH)))
-		ret = 1;
-
-	return ret;
-}
-
-void kvm_free_lapic(struct kvm_vcpu *vcpu)
-{
-	if (!vcpu->arch.apic)
-		return;
-
-	hrtimer_cancel(&vcpu->arch.apic->timer.dev);
-
-	if (vcpu->arch.apic->regs_page)
-		__free_page(vcpu->arch.apic->regs_page);
-
-	kfree(vcpu->arch.apic);
-}
-
-/*
- *----------------------------------------------------------------------
- * LAPIC interface
- *----------------------------------------------------------------------
- */
-
-void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (!apic)
-		return;
-	apic_set_tpr(apic, ((cr8 & 0x0f) << 4));
-}
-
-u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-	u64 tpr;
-
-	if (!apic)
-		return 0;
-	tpr = (u64) apic_get_reg(apic, APIC_TASKPRI);
-
-	return (tpr & 0xf0) >> 4;
-}
-EXPORT_SYMBOL_GPL(kvm_lapic_get_cr8);
-
-void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (!apic) {
-		value |= MSR_IA32_APICBASE_BSP;
-		vcpu->arch.apic_base = value;
-		return;
-	}
-	if (apic->vcpu->vcpu_id)
-		value &= ~MSR_IA32_APICBASE_BSP;
-
-	vcpu->arch.apic_base = value;
-	apic->base_address = apic->vcpu->arch.apic_base &
-			     MSR_IA32_APICBASE_BASE;
-
-	/* with FSB delivery interrupt, we can restart APIC functionality */
-	apic_debug("apic base msr is 0x%016" PRIx64 ", and base address is "
-		   "0x%lx.\n", apic->vcpu->arch.apic_base, apic->base_address);
-
-}
-
-u64 kvm_lapic_get_base(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.apic_base;
-}
-EXPORT_SYMBOL_GPL(kvm_lapic_get_base);
-
-void kvm_lapic_reset(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic;
-	int i;
-
-	apic_debug("%s\n", __FUNCTION__);
-
-	ASSERT(vcpu);
-	apic = vcpu->arch.apic;
-	ASSERT(apic != NULL);
-
-	/* Stop the timer in case it's a reset to an active apic */
-	hrtimer_cancel(&apic->timer.dev);
-
-	apic_set_reg(apic, APIC_ID, vcpu->vcpu_id << 24);
-	apic_set_reg(apic, APIC_LVR, APIC_VERSION);
-
-	for (i = 0; i < APIC_LVT_NUM; i++)
-		apic_set_reg(apic, APIC_LVTT + 0x10 * i, APIC_LVT_MASKED);
-	apic_set_reg(apic, APIC_LVT0,
-		     SET_APIC_DELIVERY_MODE(0, APIC_MODE_EXTINT));
-
-	apic_set_reg(apic, APIC_DFR, 0xffffffffU);
-	apic_set_reg(apic, APIC_SPIV, 0xff);
-	apic_set_reg(apic, APIC_TASKPRI, 0);
-	apic_set_reg(apic, APIC_LDR, 0);
-	apic_set_reg(apic, APIC_ESR, 0);
-	apic_set_reg(apic, APIC_ICR, 0);
-	apic_set_reg(apic, APIC_ICR2, 0);
-	apic_set_reg(apic, APIC_TDCR, 0);
-	apic_set_reg(apic, APIC_TMICT, 0);
-	for (i = 0; i < 8; i++) {
-		apic_set_reg(apic, APIC_IRR + 0x10 * i, 0);
-		apic_set_reg(apic, APIC_ISR + 0x10 * i, 0);
-		apic_set_reg(apic, APIC_TMR + 0x10 * i, 0);
-	}
-	update_divide_count(apic);
-	atomic_set(&apic->timer.pending, 0);
-	if (vcpu->vcpu_id == 0)
-		vcpu->arch.apic_base |= MSR_IA32_APICBASE_BSP;
-	apic_update_ppr(apic);
-
-	apic_debug(KERN_INFO "%s: vcpu=%p, id=%d, base_msr="
-		   "0x%016" PRIx64 ", base_address=0x%0lx.\n", __FUNCTION__,
-		   vcpu, kvm_apic_id(apic),
-		   vcpu->arch.apic_base, apic->base_address);
-}
-EXPORT_SYMBOL_GPL(kvm_lapic_reset);
-
-int kvm_lapic_enabled(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-	int ret = 0;
-
-	if (!apic)
-		return 0;
-	ret = apic_enabled(apic);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(kvm_lapic_enabled);
-
-/*
- *----------------------------------------------------------------------
- * timer interface
- *----------------------------------------------------------------------
- */
-
-/* TODO: make sure __apic_timer_fn runs in current pCPU */
-static int __apic_timer_fn(struct kvm_lapic *apic)
-{
-	int result = 0;
-	wait_queue_head_t *q = &apic->vcpu->wq;
-
-	atomic_inc(&apic->timer.pending);
-	if (waitqueue_active(q)) {
-		apic->vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
-		wake_up_interruptible(q);
-	}
-	if (apic_lvtt_period(apic)) {
-		result = 1;
-		apic->timer.dev.expires = ktime_add_ns(
-					apic->timer.dev.expires,
-					apic->timer.period);
-	}
-	return result;
-}
-
-static int __inject_apic_timer_irq(struct kvm_lapic *apic)
-{
-	int vector;
-
-	vector = apic_lvt_vector(apic, APIC_LVTT);
-	return __apic_accept_irq(apic, APIC_DM_FIXED, vector, 1, 0);
-}
-
-static enum hrtimer_restart apic_timer_fn(struct hrtimer *data)
-{
-	struct kvm_lapic *apic;
-	int restart_timer = 0;
-
-	apic = container_of(data, struct kvm_lapic, timer.dev);
-
-	restart_timer = __apic_timer_fn(apic);
-
-	if (restart_timer)
-		return HRTIMER_RESTART;
-	else
-		return HRTIMER_NORESTART;
-}
-
-int kvm_create_lapic(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic;
-
-	ASSERT(vcpu != NULL);
-	apic_debug("apic_init %d\n", vcpu->vcpu_id);
-
-	apic = kzalloc(sizeof(*apic), GFP_KERNEL);
-	if (!apic)
-		goto nomem;
-
-	vcpu->arch.apic = apic;
-
-	apic->regs_page = alloc_page(GFP_KERNEL);
-	if (apic->regs_page == NULL) {
-		printk(KERN_ERR "malloc apic regs error for vcpu %x\n",
-		       vcpu->vcpu_id);
-		goto nomem_free_apic;
-	}
-	apic->regs = page_address(apic->regs_page);
-	memset(apic->regs, 0, PAGE_SIZE);
-	apic->vcpu = vcpu;
-
-	hrtimer_init(&apic->timer.dev, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
-	apic->timer.dev.function = apic_timer_fn;
-	apic->base_address = APIC_DEFAULT_PHYS_BASE;
-	vcpu->arch.apic_base = APIC_DEFAULT_PHYS_BASE;
-
-	kvm_lapic_reset(vcpu);
-	apic->dev.read = apic_mmio_read;
-	apic->dev.write = apic_mmio_write;
-	apic->dev.in_range = apic_mmio_range;
-	apic->dev.private = apic;
-
-	return 0;
-nomem_free_apic:
-	kfree(apic);
-nomem:
-	return -ENOMEM;
-}
-EXPORT_SYMBOL_GPL(kvm_create_lapic);
-
-int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-	int highest_irr;
-
-	if (!apic || !apic_enabled(apic))
-		return -1;
-
-	apic_update_ppr(apic);
-	highest_irr = apic_find_highest_irr(apic);
-	if ((highest_irr == -1) ||
-	    ((highest_irr & 0xF0) <= apic_get_reg(apic, APIC_PROCPRI)))
-		return -1;
-	return highest_irr;
-}
-
-int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu)
-{
-	u32 lvt0 = apic_get_reg(vcpu->arch.apic, APIC_LVT0);
-	int r = 0;
-
-	if (vcpu->vcpu_id == 0) {
-		if (!apic_hw_enabled(vcpu->arch.apic))
-			r = 1;
-		if ((lvt0 & APIC_LVT_MASKED) == 0 &&
-		    GET_APIC_DELIVERY_MODE(lvt0) == APIC_MODE_EXTINT)
-			r = 1;
-	}
-	return r;
-}
-
-void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (apic && apic_lvt_enabled(apic, APIC_LVTT) &&
-		atomic_read(&apic->timer.pending) > 0) {
-		if (__inject_apic_timer_irq(apic))
-			atomic_dec(&apic->timer.pending);
-	}
-}
-
-void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (apic && apic_lvt_vector(apic, APIC_LVTT) == vec)
-		apic->timer.last_update = ktime_add_ns(
-				apic->timer.last_update,
-				apic->timer.period);
-}
-
-int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu)
-{
-	int vector = kvm_apic_has_interrupt(vcpu);
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	if (vector == -1)
-		return -1;
-
-	apic_set_vector(vector, apic->regs + APIC_ISR);
-	apic_update_ppr(apic);
-	apic_clear_irr(vector, apic);
-	return vector;
-}
-
-void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-
-	apic->base_address = vcpu->arch.apic_base &
-			     MSR_IA32_APICBASE_BASE;
-	apic_set_reg(apic, APIC_LVR, APIC_VERSION);
-	apic_update_ppr(apic);
-	hrtimer_cancel(&apic->timer.dev);
-	update_divide_count(apic);
-	start_apic_timer(apic);
-}
-
-void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
-{
-	struct kvm_lapic *apic = vcpu->arch.apic;
-	struct hrtimer *timer;
-
-	if (!apic)
-		return;
-
-	timer = &apic->timer.dev;
-	if (hrtimer_cancel(timer))
-		hrtimer_start(timer, timer->expires, HRTIMER_MODE_ABS);
-}
-EXPORT_SYMBOL_GPL(kvm_migrate_apic_timer);
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
deleted file mode 100644
index c26d83f86a3a..000000000000
--- a/drivers/kvm/mmu.c
+++ /dev/null
@@ -1,1806 +0,0 @@
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * This module enables machines with Intel VT-x extensions to run virtual
- * machines without emulation or binary translation.
- *
- * MMU support
- *
- * Copyright (C) 2006 Qumranet, Inc.
- *
- * Authors:
- *   Yaniv Kamay  <yaniv@qumranet.com>
- *   Avi Kivity   <avi@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-
-#include "vmx.h"
-#include "kvm.h"
-#include "x86.h"
-#include "mmu.h"
-
-#include <linux/types.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/module.h>
-#include <linux/swap.h>
-
-#include <asm/page.h>
-#include <asm/cmpxchg.h>
-#include <asm/io.h>
-
-#undef MMU_DEBUG
-
-#undef AUDIT
-
-#ifdef AUDIT
-static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg);
-#else
-static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg) {}
-#endif
-
-#ifdef MMU_DEBUG
-
-#define pgprintk(x...) do { if (dbg) printk(x); } while (0)
-#define rmap_printk(x...) do { if (dbg) printk(x); } while (0)
-
-#else
-
-#define pgprintk(x...) do { } while (0)
-#define rmap_printk(x...) do { } while (0)
-
-#endif
-
-#if defined(MMU_DEBUG) || defined(AUDIT)
-static int dbg = 1;
-#endif
-
-#ifndef MMU_DEBUG
-#define ASSERT(x) do { } while (0)
-#else
-#define ASSERT(x)							\
-	if (!(x)) {							\
-		printk(KERN_WARNING "assertion failed %s:%d: %s\n",	\
-		       __FILE__, __LINE__, #x);				\
-	}
-#endif
-
-#define PT64_PT_BITS 9
-#define PT64_ENT_PER_PAGE (1 << PT64_PT_BITS)
-#define PT32_PT_BITS 10
-#define PT32_ENT_PER_PAGE (1 << PT32_PT_BITS)
-
-#define PT_WRITABLE_SHIFT 1
-
-#define PT_PRESENT_MASK (1ULL << 0)
-#define PT_WRITABLE_MASK (1ULL << PT_WRITABLE_SHIFT)
-#define PT_USER_MASK (1ULL << 2)
-#define PT_PWT_MASK (1ULL << 3)
-#define PT_PCD_MASK (1ULL << 4)
-#define PT_ACCESSED_MASK (1ULL << 5)
-#define PT_DIRTY_MASK (1ULL << 6)
-#define PT_PAGE_SIZE_MASK (1ULL << 7)
-#define PT_PAT_MASK (1ULL << 7)
-#define PT_GLOBAL_MASK (1ULL << 8)
-#define PT64_NX_SHIFT 63
-#define PT64_NX_MASK (1ULL << PT64_NX_SHIFT)
-
-#define PT_PAT_SHIFT 7
-#define PT_DIR_PAT_SHIFT 12
-#define PT_DIR_PAT_MASK (1ULL << PT_DIR_PAT_SHIFT)
-
-#define PT32_DIR_PSE36_SIZE 4
-#define PT32_DIR_PSE36_SHIFT 13
-#define PT32_DIR_PSE36_MASK \
-	(((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT)
-
-
-#define PT_FIRST_AVAIL_BITS_SHIFT 9
-#define PT64_SECOND_AVAIL_BITS_SHIFT 52
-
-#define PT_SHADOW_IO_MARK (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
-
-#define VALID_PAGE(x) ((x) != INVALID_PAGE)
-
-#define PT64_LEVEL_BITS 9
-
-#define PT64_LEVEL_SHIFT(level) \
-		(PAGE_SHIFT + (level - 1) * PT64_LEVEL_BITS)
-
-#define PT64_LEVEL_MASK(level) \
-		(((1ULL << PT64_LEVEL_BITS) - 1) << PT64_LEVEL_SHIFT(level))
-
-#define PT64_INDEX(address, level)\
-	(((address) >> PT64_LEVEL_SHIFT(level)) & ((1 << PT64_LEVEL_BITS) - 1))
-
-
-#define PT32_LEVEL_BITS 10
-
-#define PT32_LEVEL_SHIFT(level) \
-		(PAGE_SHIFT + (level - 1) * PT32_LEVEL_BITS)
-
-#define PT32_LEVEL_MASK(level) \
-		(((1ULL << PT32_LEVEL_BITS) - 1) << PT32_LEVEL_SHIFT(level))
-
-#define PT32_INDEX(address, level)\
-	(((address) >> PT32_LEVEL_SHIFT(level)) & ((1 << PT32_LEVEL_BITS) - 1))
-
-
-#define PT64_BASE_ADDR_MASK (((1ULL << 52) - 1) & ~(u64)(PAGE_SIZE-1))
-#define PT64_DIR_BASE_ADDR_MASK \
-	(PT64_BASE_ADDR_MASK & ~((1ULL << (PAGE_SHIFT + PT64_LEVEL_BITS)) - 1))
-
-#define PT32_BASE_ADDR_MASK PAGE_MASK
-#define PT32_DIR_BASE_ADDR_MASK \
-	(PAGE_MASK & ~((1ULL << (PAGE_SHIFT + PT32_LEVEL_BITS)) - 1))
-
-#define PT64_PERM_MASK (PT_PRESENT_MASK | PT_WRITABLE_MASK | PT_USER_MASK \
-			| PT64_NX_MASK)
-
-#define PFERR_PRESENT_MASK (1U << 0)
-#define PFERR_WRITE_MASK (1U << 1)
-#define PFERR_USER_MASK (1U << 2)
-#define PFERR_FETCH_MASK (1U << 4)
-
-#define PT64_ROOT_LEVEL 4
-#define PT32_ROOT_LEVEL 2
-#define PT32E_ROOT_LEVEL 3
-
-#define PT_DIRECTORY_LEVEL 2
-#define PT_PAGE_TABLE_LEVEL 1
-
-#define RMAP_EXT 4
-
-#define ACC_EXEC_MASK    1
-#define ACC_WRITE_MASK   PT_WRITABLE_MASK
-#define ACC_USER_MASK    PT_USER_MASK
-#define ACC_ALL          (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK)
-
-struct kvm_rmap_desc {
-	u64 *shadow_ptes[RMAP_EXT];
-	struct kvm_rmap_desc *more;
-};
-
-static struct kmem_cache *pte_chain_cache;
-static struct kmem_cache *rmap_desc_cache;
-static struct kmem_cache *mmu_page_header_cache;
-
-static u64 __read_mostly shadow_trap_nonpresent_pte;
-static u64 __read_mostly shadow_notrap_nonpresent_pte;
-
-void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte)
-{
-	shadow_trap_nonpresent_pte = trap_pte;
-	shadow_notrap_nonpresent_pte = notrap_pte;
-}
-EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes);
-
-static int is_write_protection(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.cr0 & X86_CR0_WP;
-}
-
-static int is_cpuid_PSE36(void)
-{
-	return 1;
-}
-
-static int is_nx(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.shadow_efer & EFER_NX;
-}
-
-static int is_present_pte(unsigned long pte)
-{
-	return pte & PT_PRESENT_MASK;
-}
-
-static int is_shadow_present_pte(u64 pte)
-{
-	pte &= ~PT_SHADOW_IO_MARK;
-	return pte != shadow_trap_nonpresent_pte
-		&& pte != shadow_notrap_nonpresent_pte;
-}
-
-static int is_writeble_pte(unsigned long pte)
-{
-	return pte & PT_WRITABLE_MASK;
-}
-
-static int is_dirty_pte(unsigned long pte)
-{
-	return pte & PT_DIRTY_MASK;
-}
-
-static int is_io_pte(unsigned long pte)
-{
-	return pte & PT_SHADOW_IO_MARK;
-}
-
-static int is_rmap_pte(u64 pte)
-{
-	return pte != shadow_trap_nonpresent_pte
-		&& pte != shadow_notrap_nonpresent_pte;
-}
-
-static gfn_t pse36_gfn_delta(u32 gpte)
-{
-	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
-
-	return (gpte & PT32_DIR_PSE36_MASK) << shift;
-}
-
-static void set_shadow_pte(u64 *sptep, u64 spte)
-{
-#ifdef CONFIG_X86_64
-	set_64bit((unsigned long *)sptep, spte);
-#else
-	set_64bit((unsigned long long *)sptep, spte);
-#endif
-}
-
-static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache,
-				  struct kmem_cache *base_cache, int min)
-{
-	void *obj;
-
-	if (cache->nobjs >= min)
-		return 0;
-	while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
-		obj = kmem_cache_zalloc(base_cache, GFP_KERNEL);
-		if (!obj)
-			return -ENOMEM;
-		cache->objects[cache->nobjs++] = obj;
-	}
-	return 0;
-}
-
-static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc)
-{
-	while (mc->nobjs)
-		kfree(mc->objects[--mc->nobjs]);
-}
-
-static int mmu_topup_memory_cache_page(struct kvm_mmu_memory_cache *cache,
-				       int min)
-{
-	struct page *page;
-
-	if (cache->nobjs >= min)
-		return 0;
-	while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
-		page = alloc_page(GFP_KERNEL);
-		if (!page)
-			return -ENOMEM;
-		set_page_private(page, 0);
-		cache->objects[cache->nobjs++] = page_address(page);
-	}
-	return 0;
-}
-
-static void mmu_free_memory_cache_page(struct kvm_mmu_memory_cache *mc)
-{
-	while (mc->nobjs)
-		free_page((unsigned long)mc->objects[--mc->nobjs]);
-}
-
-static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
-{
-	int r;
-
-	kvm_mmu_free_some_pages(vcpu);
-	r = mmu_topup_memory_cache(&vcpu->arch.mmu_pte_chain_cache,
-				   pte_chain_cache, 4);
-	if (r)
-		goto out;
-	r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache,
-				   rmap_desc_cache, 1);
-	if (r)
-		goto out;
-	r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8);
-	if (r)
-		goto out;
-	r = mmu_topup_memory_cache(&vcpu->arch.mmu_page_header_cache,
-				   mmu_page_header_cache, 4);
-out:
-	return r;
-}
-
-static void mmu_free_memory_caches(struct kvm_vcpu *vcpu)
-{
-	mmu_free_memory_cache(&vcpu->arch.mmu_pte_chain_cache);
-	mmu_free_memory_cache(&vcpu->arch.mmu_rmap_desc_cache);
-	mmu_free_memory_cache_page(&vcpu->arch.mmu_page_cache);
-	mmu_free_memory_cache(&vcpu->arch.mmu_page_header_cache);
-}
-
-static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
-				    size_t size)
-{
-	void *p;
-
-	BUG_ON(!mc->nobjs);
-	p = mc->objects[--mc->nobjs];
-	memset(p, 0, size);
-	return p;
-}
-
-static struct kvm_pte_chain *mmu_alloc_pte_chain(struct kvm_vcpu *vcpu)
-{
-	return mmu_memory_cache_alloc(&vcpu->arch.mmu_pte_chain_cache,
-				      sizeof(struct kvm_pte_chain));
-}
-
-static void mmu_free_pte_chain(struct kvm_pte_chain *pc)
-{
-	kfree(pc);
-}
-
-static struct kvm_rmap_desc *mmu_alloc_rmap_desc(struct kvm_vcpu *vcpu)
-{
-	return mmu_memory_cache_alloc(&vcpu->arch.mmu_rmap_desc_cache,
-				      sizeof(struct kvm_rmap_desc));
-}
-
-static void mmu_free_rmap_desc(struct kvm_rmap_desc *rd)
-{
-	kfree(rd);
-}
-
-/*
- * Take gfn and return the reverse mapping to it.
- * Note: gfn must be unaliased before this function get called
- */
-
-static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn)
-{
-	struct kvm_memory_slot *slot;
-
-	slot = gfn_to_memslot(kvm, gfn);
-	return &slot->rmap[gfn - slot->base_gfn];
-}
-
-/*
- * Reverse mapping data structures:
- *
- * If rmapp bit zero is zero, then rmapp point to the shadw page table entry
- * that points to page_address(page).
- *
- * If rmapp bit zero is one, (then rmap & ~1) points to a struct kvm_rmap_desc
- * containing more mappings.
- */
-static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
-{
-	struct kvm_mmu_page *sp;
-	struct kvm_rmap_desc *desc;
-	unsigned long *rmapp;
-	int i;
-
-	if (!is_rmap_pte(*spte))
-		return;
-	gfn = unalias_gfn(vcpu->kvm, gfn);
-	sp = page_header(__pa(spte));
-	sp->gfns[spte - sp->spt] = gfn;
-	rmapp = gfn_to_rmap(vcpu->kvm, gfn);
-	if (!*rmapp) {
-		rmap_printk("rmap_add: %p %llx 0->1\n", spte, *spte);
-		*rmapp = (unsigned long)spte;
-	} else if (!(*rmapp & 1)) {
-		rmap_printk("rmap_add: %p %llx 1->many\n", spte, *spte);
-		desc = mmu_alloc_rmap_desc(vcpu);
-		desc->shadow_ptes[0] = (u64 *)*rmapp;
-		desc->shadow_ptes[1] = spte;
-		*rmapp = (unsigned long)desc | 1;
-	} else {
-		rmap_printk("rmap_add: %p %llx many->many\n", spte, *spte);
-		desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
-		while (desc->shadow_ptes[RMAP_EXT-1] && desc->more)
-			desc = desc->more;
-		if (desc->shadow_ptes[RMAP_EXT-1]) {
-			desc->more = mmu_alloc_rmap_desc(vcpu);
-			desc = desc->more;
-		}
-		for (i = 0; desc->shadow_ptes[i]; ++i)
-			;
-		desc->shadow_ptes[i] = spte;
-	}
-}
-
-static void rmap_desc_remove_entry(unsigned long *rmapp,
-				   struct kvm_rmap_desc *desc,
-				   int i,
-				   struct kvm_rmap_desc *prev_desc)
-{
-	int j;
-
-	for (j = RMAP_EXT - 1; !desc->shadow_ptes[j] && j > i; --j)
-		;
-	desc->shadow_ptes[i] = desc->shadow_ptes[j];
-	desc->shadow_ptes[j] = NULL;
-	if (j != 0)
-		return;
-	if (!prev_desc && !desc->more)
-		*rmapp = (unsigned long)desc->shadow_ptes[0];
-	else
-		if (prev_desc)
-			prev_desc->more = desc->more;
-		else
-			*rmapp = (unsigned long)desc->more | 1;
-	mmu_free_rmap_desc(desc);
-}
-
-static void rmap_remove(struct kvm *kvm, u64 *spte)
-{
-	struct kvm_rmap_desc *desc;
-	struct kvm_rmap_desc *prev_desc;
-	struct kvm_mmu_page *sp;
-	struct page *page;
-	unsigned long *rmapp;
-	int i;
-
-	if (!is_rmap_pte(*spte))
-		return;
-	sp = page_header(__pa(spte));
-	page = pfn_to_page((*spte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT);
-	mark_page_accessed(page);
-	if (is_writeble_pte(*spte))
-		kvm_release_page_dirty(page);
-	else
-		kvm_release_page_clean(page);
-	rmapp = gfn_to_rmap(kvm, sp->gfns[spte - sp->spt]);
-	if (!*rmapp) {
-		printk(KERN_ERR "rmap_remove: %p %llx 0->BUG\n", spte, *spte);
-		BUG();
-	} else if (!(*rmapp & 1)) {
-		rmap_printk("rmap_remove:  %p %llx 1->0\n", spte, *spte);
-		if ((u64 *)*rmapp != spte) {
-			printk(KERN_ERR "rmap_remove:  %p %llx 1->BUG\n",
-			       spte, *spte);
-			BUG();
-		}
-		*rmapp = 0;
-	} else {
-		rmap_printk("rmap_remove:  %p %llx many->many\n", spte, *spte);
-		desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
-		prev_desc = NULL;
-		while (desc) {
-			for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i)
-				if (desc->shadow_ptes[i] == spte) {
-					rmap_desc_remove_entry(rmapp,
-							       desc, i,
-							       prev_desc);
-					return;
-				}
-			prev_desc = desc;
-			desc = desc->more;
-		}
-		BUG();
-	}
-}
-
-static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte)
-{
-	struct kvm_rmap_desc *desc;
-	struct kvm_rmap_desc *prev_desc;
-	u64 *prev_spte;
-	int i;
-
-	if (!*rmapp)
-		return NULL;
-	else if (!(*rmapp & 1)) {
-		if (!spte)
-			return (u64 *)*rmapp;
-		return NULL;
-	}
-	desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
-	prev_desc = NULL;
-	prev_spte = NULL;
-	while (desc) {
-		for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i) {
-			if (prev_spte == spte)
-				return desc->shadow_ptes[i];
-			prev_spte = desc->shadow_ptes[i];
-		}
-		desc = desc->more;
-	}
-	return NULL;
-}
-
-static void rmap_write_protect(struct kvm *kvm, u64 gfn)
-{
-	unsigned long *rmapp;
-	u64 *spte;
-
-	gfn = unalias_gfn(kvm, gfn);
-	rmapp = gfn_to_rmap(kvm, gfn);
-
-	spte = rmap_next(kvm, rmapp, NULL);
-	while (spte) {
-		BUG_ON(!spte);
-		BUG_ON(!(*spte & PT_PRESENT_MASK));
-		rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte);
-		if (is_writeble_pte(*spte))
-			set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK);
-		kvm_flush_remote_tlbs(kvm);
-		spte = rmap_next(kvm, rmapp, spte);
-	}
-}
-
-#ifdef MMU_DEBUG
-static int is_empty_shadow_page(u64 *spt)
-{
-	u64 *pos;
-	u64 *end;
-
-	for (pos = spt, end = pos + PAGE_SIZE / sizeof(u64); pos != end; pos++)
-		if ((*pos & ~PT_SHADOW_IO_MARK) != shadow_trap_nonpresent_pte) {
-			printk(KERN_ERR "%s: %p %llx\n", __FUNCTION__,
-			       pos, *pos);
-			return 0;
-		}
-	return 1;
-}
-#endif
-
-static void kvm_mmu_free_page(struct kvm *kvm, struct kvm_mmu_page *sp)
-{
-	ASSERT(is_empty_shadow_page(sp->spt));
-	list_del(&sp->link);
-	__free_page(virt_to_page(sp->spt));
-	__free_page(virt_to_page(sp->gfns));
-	kfree(sp);
-	++kvm->arch.n_free_mmu_pages;
-}
-
-static unsigned kvm_page_table_hashfn(gfn_t gfn)
-{
-	return gfn;
-}
-
-static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
-					       u64 *parent_pte)
-{
-	struct kvm_mmu_page *sp;
-
-	if (!vcpu->kvm->arch.n_free_mmu_pages)
-		return NULL;
-
-	sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
-	sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
-	sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
-	set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
-	list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
-	ASSERT(is_empty_shadow_page(sp->spt));
-	sp->slot_bitmap = 0;
-	sp->multimapped = 0;
-	sp->parent_pte = parent_pte;
-	--vcpu->kvm->arch.n_free_mmu_pages;
-	return sp;
-}
-
-static void mmu_page_add_parent_pte(struct kvm_vcpu *vcpu,
-				    struct kvm_mmu_page *sp, u64 *parent_pte)
-{
-	struct kvm_pte_chain *pte_chain;
-	struct hlist_node *node;
-	int i;
-
-	if (!parent_pte)
-		return;
-	if (!sp->multimapped) {
-		u64 *old = sp->parent_pte;
-
-		if (!old) {
-			sp->parent_pte = parent_pte;
-			return;
-		}
-		sp->multimapped = 1;
-		pte_chain = mmu_alloc_pte_chain(vcpu);
-		INIT_HLIST_HEAD(&sp->parent_ptes);
-		hlist_add_head(&pte_chain->link, &sp->parent_ptes);
-		pte_chain->parent_ptes[0] = old;
-	}
-	hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) {
-		if (pte_chain->parent_ptes[NR_PTE_CHAIN_ENTRIES-1])
-			continue;
-		for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i)
-			if (!pte_chain->parent_ptes[i]) {
-				pte_chain->parent_ptes[i] = parent_pte;
-				return;
-			}
-	}
-	pte_chain = mmu_alloc_pte_chain(vcpu);
-	BUG_ON(!pte_chain);
-	hlist_add_head(&pte_chain->link, &sp->parent_ptes);
-	pte_chain->parent_ptes[0] = parent_pte;
-}
-
-static void mmu_page_remove_parent_pte(struct kvm_mmu_page *sp,
-				       u64 *parent_pte)
-{
-	struct kvm_pte_chain *pte_chain;
-	struct hlist_node *node;
-	int i;
-
-	if (!sp->multimapped) {
-		BUG_ON(sp->parent_pte != parent_pte);
-		sp->parent_pte = NULL;
-		return;
-	}
-	hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link)
-		for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) {
-			if (!pte_chain->parent_ptes[i])
-				break;
-			if (pte_chain->parent_ptes[i] != parent_pte)
-				continue;
-			while (i + 1 < NR_PTE_CHAIN_ENTRIES
-				&& pte_chain->parent_ptes[i + 1]) {
-				pte_chain->parent_ptes[i]
-					= pte_chain->parent_ptes[i + 1];
-				++i;
-			}
-			pte_chain->parent_ptes[i] = NULL;
-			if (i == 0) {
-				hlist_del(&pte_chain->link);
-				mmu_free_pte_chain(pte_chain);
-				if (hlist_empty(&sp->parent_ptes)) {
-					sp->multimapped = 0;
-					sp->parent_pte = NULL;
-				}
-			}
-			return;
-		}
-	BUG();
-}
-
-static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
-{
-	unsigned index;
-	struct hlist_head *bucket;
-	struct kvm_mmu_page *sp;
-	struct hlist_node *node;
-
-	pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
-	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
-	bucket = &kvm->arch.mmu_page_hash[index];
-	hlist_for_each_entry(sp, node, bucket, hash_link)
-		if (sp->gfn == gfn && !sp->role.metaphysical) {
-			pgprintk("%s: found role %x\n",
-				 __FUNCTION__, sp->role.word);
-			return sp;
-		}
-	return NULL;
-}
-
-static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
-					     gfn_t gfn,
-					     gva_t gaddr,
-					     unsigned level,
-					     int metaphysical,
-					     unsigned access,
-					     u64 *parent_pte,
-					     bool *new_page)
-{
-	union kvm_mmu_page_role role;
-	unsigned index;
-	unsigned quadrant;
-	struct hlist_head *bucket;
-	struct kvm_mmu_page *sp;
-	struct hlist_node *node;
-
-	role.word = 0;
-	role.glevels = vcpu->arch.mmu.root_level;
-	role.level = level;
-	role.metaphysical = metaphysical;
-	role.access = access;
-	if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) {
-		quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level));
-		quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1;
-		role.quadrant = quadrant;
-	}
-	pgprintk("%s: looking gfn %lx role %x\n", __FUNCTION__,
-		 gfn, role.word);
-	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
-	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
-	hlist_for_each_entry(sp, node, bucket, hash_link)
-		if (sp->gfn == gfn && sp->role.word == role.word) {
-			mmu_page_add_parent_pte(vcpu, sp, parent_pte);
-			pgprintk("%s: found\n", __FUNCTION__);
-			return sp;
-		}
-	sp = kvm_mmu_alloc_page(vcpu, parent_pte);
-	if (!sp)
-		return sp;
-	pgprintk("%s: adding gfn %lx role %x\n", __FUNCTION__, gfn, role.word);
-	sp->gfn = gfn;
-	sp->role = role;
-	hlist_add_head(&sp->hash_link, bucket);
-	vcpu->arch.mmu.prefetch_page(vcpu, sp);
-	if (!metaphysical)
-		rmap_write_protect(vcpu->kvm, gfn);
-	if (new_page)
-		*new_page = 1;
-	return sp;
-}
-
-static void kvm_mmu_page_unlink_children(struct kvm *kvm,
-					 struct kvm_mmu_page *sp)
-{
-	unsigned i;
-	u64 *pt;
-	u64 ent;
-
-	pt = sp->spt;
-
-	if (sp->role.level == PT_PAGE_TABLE_LEVEL) {
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
-			if (is_shadow_present_pte(pt[i]))
-				rmap_remove(kvm, &pt[i]);
-			pt[i] = shadow_trap_nonpresent_pte;
-		}
-		kvm_flush_remote_tlbs(kvm);
-		return;
-	}
-
-	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
-		ent = pt[i];
-
-		pt[i] = shadow_trap_nonpresent_pte;
-		if (!is_shadow_present_pte(ent))
-			continue;
-		ent &= PT64_BASE_ADDR_MASK;
-		mmu_page_remove_parent_pte(page_header(ent), &pt[i]);
-	}
-	kvm_flush_remote_tlbs(kvm);
-}
-
-static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte)
-{
-	mmu_page_remove_parent_pte(sp, parent_pte);
-}
-
-static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm)
-{
-	int i;
-
-	for (i = 0; i < KVM_MAX_VCPUS; ++i)
-		if (kvm->vcpus[i])
-			kvm->vcpus[i]->arch.last_pte_updated = NULL;
-}
-
-static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
-{
-	u64 *parent_pte;
-
-	++kvm->stat.mmu_shadow_zapped;
-	while (sp->multimapped || sp->parent_pte) {
-		if (!sp->multimapped)
-			parent_pte = sp->parent_pte;
-		else {
-			struct kvm_pte_chain *chain;
-
-			chain = container_of(sp->parent_ptes.first,
-					     struct kvm_pte_chain, link);
-			parent_pte = chain->parent_ptes[0];
-		}
-		BUG_ON(!parent_pte);
-		kvm_mmu_put_page(sp, parent_pte);
-		set_shadow_pte(parent_pte, shadow_trap_nonpresent_pte);
-	}
-	kvm_mmu_page_unlink_children(kvm, sp);
-	if (!sp->root_count) {
-		hlist_del(&sp->hash_link);
-		kvm_mmu_free_page(kvm, sp);
-	} else
-		list_move(&sp->link, &kvm->arch.active_mmu_pages);
-	kvm_mmu_reset_last_pte_updated(kvm);
-}
-
-/*
- * Changing the number of mmu pages allocated to the vm
- * Note: if kvm_nr_mmu_pages is too small, you will get dead lock
- */
-void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages)
-{
-	/*
-	 * If we set the number of mmu pages to be smaller be than the
-	 * number of actived pages , we must to free some mmu pages before we
-	 * change the value
-	 */
-
-	if ((kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages) >
-	    kvm_nr_mmu_pages) {
-		int n_used_mmu_pages = kvm->arch.n_alloc_mmu_pages
-				       - kvm->arch.n_free_mmu_pages;
-
-		while (n_used_mmu_pages > kvm_nr_mmu_pages) {
-			struct kvm_mmu_page *page;
-
-			page = container_of(kvm->arch.active_mmu_pages.prev,
-					    struct kvm_mmu_page, link);
-			kvm_mmu_zap_page(kvm, page);
-			n_used_mmu_pages--;
-		}
-		kvm->arch.n_free_mmu_pages = 0;
-	}
-	else
-		kvm->arch.n_free_mmu_pages += kvm_nr_mmu_pages
-					 - kvm->arch.n_alloc_mmu_pages;
-
-	kvm->arch.n_alloc_mmu_pages = kvm_nr_mmu_pages;
-}
-
-static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
-{
-	unsigned index;
-	struct hlist_head *bucket;
-	struct kvm_mmu_page *sp;
-	struct hlist_node *node, *n;
-	int r;
-
-	pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
-	r = 0;
-	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
-	bucket = &kvm->arch.mmu_page_hash[index];
-	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link)
-		if (sp->gfn == gfn && !sp->role.metaphysical) {
-			pgprintk("%s: gfn %lx role %x\n", __FUNCTION__, gfn,
-				 sp->role.word);
-			kvm_mmu_zap_page(kvm, sp);
-			r = 1;
-		}
-	return r;
-}
-
-static void mmu_unshadow(struct kvm *kvm, gfn_t gfn)
-{
-	struct kvm_mmu_page *sp;
-
-	while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) {
-		pgprintk("%s: zap %lx %x\n", __FUNCTION__, gfn, sp->role.word);
-		kvm_mmu_zap_page(kvm, sp);
-	}
-}
-
-static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn)
-{
-	int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn));
-	struct kvm_mmu_page *sp = page_header(__pa(pte));
-
-	__set_bit(slot, &sp->slot_bitmap);
-}
-
-struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
-{
-	gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
-
-	if (gpa == UNMAPPED_GVA)
-		return NULL;
-	return gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
-}
-
-static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
-			 unsigned pt_access, unsigned pte_access,
-			 int user_fault, int write_fault, int dirty,
-			 int *ptwrite, gfn_t gfn)
-{
-	u64 spte;
-	int was_rmapped = is_rmap_pte(*shadow_pte);
-	struct page *page;
-
-	pgprintk("%s: spte %llx access %x write_fault %d"
-		 " user_fault %d gfn %lx\n",
-		 __FUNCTION__, *shadow_pte, pt_access,
-		 write_fault, user_fault, gfn);
-
-	/*
-	 * We don't set the accessed bit, since we sometimes want to see
-	 * whether the guest actually used the pte (in order to detect
-	 * demand paging).
-	 */
-	spte = PT_PRESENT_MASK | PT_DIRTY_MASK;
-	if (!dirty)
-		pte_access &= ~ACC_WRITE_MASK;
-	if (!(pte_access & ACC_EXEC_MASK))
-		spte |= PT64_NX_MASK;
-
-	page = gfn_to_page(vcpu->kvm, gfn);
-
-	spte |= PT_PRESENT_MASK;
-	if (pte_access & ACC_USER_MASK)
-		spte |= PT_USER_MASK;
-
-	if (is_error_page(page)) {
-		set_shadow_pte(shadow_pte,
-			       shadow_trap_nonpresent_pte | PT_SHADOW_IO_MARK);
-		kvm_release_page_clean(page);
-		return;
-	}
-
-	spte |= page_to_phys(page);
-
-	if ((pte_access & ACC_WRITE_MASK)
-	    || (write_fault && !is_write_protection(vcpu) && !user_fault)) {
-		struct kvm_mmu_page *shadow;
-
-		spte |= PT_WRITABLE_MASK;
-		if (user_fault) {
-			mmu_unshadow(vcpu->kvm, gfn);
-			goto unshadowed;
-		}
-
-		shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn);
-		if (shadow) {
-			pgprintk("%s: found shadow page for %lx, marking ro\n",
-				 __FUNCTION__, gfn);
-			pte_access &= ~ACC_WRITE_MASK;
-			if (is_writeble_pte(spte)) {
-				spte &= ~PT_WRITABLE_MASK;
-				kvm_x86_ops->tlb_flush(vcpu);
-			}
-			if (write_fault)
-				*ptwrite = 1;
-		}
-	}
-
-unshadowed:
-
-	if (pte_access & ACC_WRITE_MASK)
-		mark_page_dirty(vcpu->kvm, gfn);
-
-	pgprintk("%s: setting spte %llx\n", __FUNCTION__, spte);
-	set_shadow_pte(shadow_pte, spte);
-	page_header_update_slot(vcpu->kvm, shadow_pte, gfn);
-	if (!was_rmapped) {
-		rmap_add(vcpu, shadow_pte, gfn);
-		if (!is_rmap_pte(*shadow_pte))
-			kvm_release_page_clean(page);
-	}
-	else
-		kvm_release_page_clean(page);
-	if (!ptwrite || !*ptwrite)
-		vcpu->arch.last_pte_updated = shadow_pte;
-}
-
-static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
-{
-}
-
-static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
-{
-	int level = PT32E_ROOT_LEVEL;
-	hpa_t table_addr = vcpu->arch.mmu.root_hpa;
-	int pt_write = 0;
-
-	for (; ; level--) {
-		u32 index = PT64_INDEX(v, level);
-		u64 *table;
-
-		ASSERT(VALID_PAGE(table_addr));
-		table = __va(table_addr);
-
-		if (level == 1) {
-			mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL,
-				     0, write, 1, &pt_write, gfn);
-			return pt_write || is_io_pte(table[index]);
-		}
-
-		if (table[index] == shadow_trap_nonpresent_pte) {
-			struct kvm_mmu_page *new_table;
-			gfn_t pseudo_gfn;
-
-			pseudo_gfn = (v & PT64_DIR_BASE_ADDR_MASK)
-				>> PAGE_SHIFT;
-			new_table = kvm_mmu_get_page(vcpu, pseudo_gfn,
-						     v, level - 1,
-						     1, ACC_ALL, &table[index],
-						     NULL);
-			if (!new_table) {
-				pgprintk("nonpaging_map: ENOMEM\n");
-				return -ENOMEM;
-			}
-
-			table[index] = __pa(new_table->spt) | PT_PRESENT_MASK
-				| PT_WRITABLE_MASK | PT_USER_MASK;
-		}
-		table_addr = table[index] & PT64_BASE_ADDR_MASK;
-	}
-}
-
-static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu,
-				    struct kvm_mmu_page *sp)
-{
-	int i;
-
-	for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
-		sp->spt[i] = shadow_trap_nonpresent_pte;
-}
-
-static void mmu_free_roots(struct kvm_vcpu *vcpu)
-{
-	int i;
-	struct kvm_mmu_page *sp;
-
-	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
-		return;
-#ifdef CONFIG_X86_64
-	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
-		hpa_t root = vcpu->arch.mmu.root_hpa;
-
-		sp = page_header(root);
-		--sp->root_count;
-		vcpu->arch.mmu.root_hpa = INVALID_PAGE;
-		return;
-	}
-#endif
-	for (i = 0; i < 4; ++i) {
-		hpa_t root = vcpu->arch.mmu.pae_root[i];
-
-		if (root) {
-			root &= PT64_BASE_ADDR_MASK;
-			sp = page_header(root);
-			--sp->root_count;
-		}
-		vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
-	}
-	vcpu->arch.mmu.root_hpa = INVALID_PAGE;
-}
-
-static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
-{
-	int i;
-	gfn_t root_gfn;
-	struct kvm_mmu_page *sp;
-
-	root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT;
-
-#ifdef CONFIG_X86_64
-	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
-		hpa_t root = vcpu->arch.mmu.root_hpa;
-
-		ASSERT(!VALID_PAGE(root));
-		sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
-				      PT64_ROOT_LEVEL, 0, ACC_ALL, NULL, NULL);
-		root = __pa(sp->spt);
-		++sp->root_count;
-		vcpu->arch.mmu.root_hpa = root;
-		return;
-	}
-#endif
-	for (i = 0; i < 4; ++i) {
-		hpa_t root = vcpu->arch.mmu.pae_root[i];
-
-		ASSERT(!VALID_PAGE(root));
-		if (vcpu->arch.mmu.root_level == PT32E_ROOT_LEVEL) {
-			if (!is_present_pte(vcpu->arch.pdptrs[i])) {
-				vcpu->arch.mmu.pae_root[i] = 0;
-				continue;
-			}
-			root_gfn = vcpu->arch.pdptrs[i] >> PAGE_SHIFT;
-		} else if (vcpu->arch.mmu.root_level == 0)
-			root_gfn = 0;
-		sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
-				      PT32_ROOT_LEVEL, !is_paging(vcpu),
-				      ACC_ALL, NULL, NULL);
-		root = __pa(sp->spt);
-		++sp->root_count;
-		vcpu->arch.mmu.pae_root[i] = root | PT_PRESENT_MASK;
-	}
-	vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root);
-}
-
-static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr)
-{
-	return vaddr;
-}
-
-static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
-				u32 error_code)
-{
-	gfn_t gfn;
-	int r;
-
-	pgprintk("%s: gva %lx error %x\n", __FUNCTION__, gva, error_code);
-	r = mmu_topup_memory_caches(vcpu);
-	if (r)
-		return r;
-
-	ASSERT(vcpu);
-	ASSERT(VALID_PAGE(vcpu->arch.mmu.root_hpa));
-
-	gfn = gva >> PAGE_SHIFT;
-
-	return nonpaging_map(vcpu, gva & PAGE_MASK,
-			     error_code & PFERR_WRITE_MASK, gfn);
-}
-
-static void nonpaging_free(struct kvm_vcpu *vcpu)
-{
-	mmu_free_roots(vcpu);
-}
-
-static int nonpaging_init_context(struct kvm_vcpu *vcpu)
-{
-	struct kvm_mmu *context = &vcpu->arch.mmu;
-
-	context->new_cr3 = nonpaging_new_cr3;
-	context->page_fault = nonpaging_page_fault;
-	context->gva_to_gpa = nonpaging_gva_to_gpa;
-	context->free = nonpaging_free;
-	context->prefetch_page = nonpaging_prefetch_page;
-	context->root_level = 0;
-	context->shadow_root_level = PT32E_ROOT_LEVEL;
-	context->root_hpa = INVALID_PAGE;
-	return 0;
-}
-
-void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
-{
-	++vcpu->stat.tlb_flush;
-	kvm_x86_ops->tlb_flush(vcpu);
-}
-
-static void paging_new_cr3(struct kvm_vcpu *vcpu)
-{
-	pgprintk("%s: cr3 %lx\n", __FUNCTION__, vcpu->cr3);
-	mmu_free_roots(vcpu);
-}
-
-static void inject_page_fault(struct kvm_vcpu *vcpu,
-			      u64 addr,
-			      u32 err_code)
-{
-	kvm_inject_page_fault(vcpu, addr, err_code);
-}
-
-static void paging_free(struct kvm_vcpu *vcpu)
-{
-	nonpaging_free(vcpu);
-}
-
-#define PTTYPE 64
-#include "paging_tmpl.h"
-#undef PTTYPE
-
-#define PTTYPE 32
-#include "paging_tmpl.h"
-#undef PTTYPE
-
-static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level)
-{
-	struct kvm_mmu *context = &vcpu->arch.mmu;
-
-	ASSERT(is_pae(vcpu));
-	context->new_cr3 = paging_new_cr3;
-	context->page_fault = paging64_page_fault;
-	context->gva_to_gpa = paging64_gva_to_gpa;
-	context->prefetch_page = paging64_prefetch_page;
-	context->free = paging_free;
-	context->root_level = level;
-	context->shadow_root_level = level;
-	context->root_hpa = INVALID_PAGE;
-	return 0;
-}
-
-static int paging64_init_context(struct kvm_vcpu *vcpu)
-{
-	return paging64_init_context_common(vcpu, PT64_ROOT_LEVEL);
-}
-
-static int paging32_init_context(struct kvm_vcpu *vcpu)
-{
-	struct kvm_mmu *context = &vcpu->arch.mmu;
-
-	context->new_cr3 = paging_new_cr3;
-	context->page_fault = paging32_page_fault;
-	context->gva_to_gpa = paging32_gva_to_gpa;
-	context->free = paging_free;
-	context->prefetch_page = paging32_prefetch_page;
-	context->root_level = PT32_ROOT_LEVEL;
-	context->shadow_root_level = PT32E_ROOT_LEVEL;
-	context->root_hpa = INVALID_PAGE;
-	return 0;
-}
-
-static int paging32E_init_context(struct kvm_vcpu *vcpu)
-{
-	return paging64_init_context_common(vcpu, PT32E_ROOT_LEVEL);
-}
-
-static int init_kvm_mmu(struct kvm_vcpu *vcpu)
-{
-	ASSERT(vcpu);
-	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
-
-	if (!is_paging(vcpu))
-		return nonpaging_init_context(vcpu);
-	else if (is_long_mode(vcpu))
-		return paging64_init_context(vcpu);
-	else if (is_pae(vcpu))
-		return paging32E_init_context(vcpu);
-	else
-		return paging32_init_context(vcpu);
-}
-
-static void destroy_kvm_mmu(struct kvm_vcpu *vcpu)
-{
-	ASSERT(vcpu);
-	if (VALID_PAGE(vcpu->arch.mmu.root_hpa)) {
-		vcpu->arch.mmu.free(vcpu);
-		vcpu->arch.mmu.root_hpa = INVALID_PAGE;
-	}
-}
-
-int kvm_mmu_reset_context(struct kvm_vcpu *vcpu)
-{
-	destroy_kvm_mmu(vcpu);
-	return init_kvm_mmu(vcpu);
-}
-EXPORT_SYMBOL_GPL(kvm_mmu_reset_context);
-
-int kvm_mmu_load(struct kvm_vcpu *vcpu)
-{
-	int r;
-
-	mutex_lock(&vcpu->kvm->lock);
-	r = mmu_topup_memory_caches(vcpu);
-	if (r)
-		goto out;
-	mmu_alloc_roots(vcpu);
-	kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
-	kvm_mmu_flush_tlb(vcpu);
-out:
-	mutex_unlock(&vcpu->kvm->lock);
-	return r;
-}
-EXPORT_SYMBOL_GPL(kvm_mmu_load);
-
-void kvm_mmu_unload(struct kvm_vcpu *vcpu)
-{
-	mmu_free_roots(vcpu);
-}
-
-static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu,
-				  struct kvm_mmu_page *sp,
-				  u64 *spte)
-{
-	u64 pte;
-	struct kvm_mmu_page *child;
-
-	pte = *spte;
-	if (is_shadow_present_pte(pte)) {
-		if (sp->role.level == PT_PAGE_TABLE_LEVEL)
-			rmap_remove(vcpu->kvm, spte);
-		else {
-			child = page_header(pte & PT64_BASE_ADDR_MASK);
-			mmu_page_remove_parent_pte(child, spte);
-		}
-	}
-	set_shadow_pte(spte, shadow_trap_nonpresent_pte);
-}
-
-static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu,
-				  struct kvm_mmu_page *sp,
-				  u64 *spte,
-				  const void *new, int bytes,
-				  int offset_in_pte)
-{
-	if (sp->role.level != PT_PAGE_TABLE_LEVEL) {
-		++vcpu->kvm->stat.mmu_pde_zapped;
-		return;
-	}
-
-	++vcpu->kvm->stat.mmu_pte_updated;
-	if (sp->role.glevels == PT32_ROOT_LEVEL)
-		paging32_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
-	else
-		paging64_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
-}
-
-static bool need_remote_flush(u64 old, u64 new)
-{
-	if (!is_shadow_present_pte(old))
-		return false;
-	if (!is_shadow_present_pte(new))
-		return true;
-	if ((old ^ new) & PT64_BASE_ADDR_MASK)
-		return true;
-	old ^= PT64_NX_MASK;
-	new ^= PT64_NX_MASK;
-	return (old & ~new & PT64_PERM_MASK) != 0;
-}
-
-static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, u64 old, u64 new)
-{
-	if (need_remote_flush(old, new))
-		kvm_flush_remote_tlbs(vcpu->kvm);
-	else
-		kvm_mmu_flush_tlb(vcpu);
-}
-
-static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
-{
-	u64 *spte = vcpu->arch.last_pte_updated;
-
-	return !!(spte && (*spte & PT_ACCESSED_MASK));
-}
-
-void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
-		       const u8 *new, int bytes)
-{
-	gfn_t gfn = gpa >> PAGE_SHIFT;
-	struct kvm_mmu_page *sp;
-	struct hlist_node *node, *n;
-	struct hlist_head *bucket;
-	unsigned index;
-	u64 entry;
-	u64 *spte;
-	unsigned offset = offset_in_page(gpa);
-	unsigned pte_size;
-	unsigned page_offset;
-	unsigned misaligned;
-	unsigned quadrant;
-	int level;
-	int flooded = 0;
-	int npte;
-
-	pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
-	++vcpu->kvm->stat.mmu_pte_write;
-	kvm_mmu_audit(vcpu, "pre pte write");
-	if (gfn == vcpu->arch.last_pt_write_gfn
-	    && !last_updated_pte_accessed(vcpu)) {
-		++vcpu->arch.last_pt_write_count;
-		if (vcpu->arch.last_pt_write_count >= 3)
-			flooded = 1;
-	} else {
-		vcpu->arch.last_pt_write_gfn = gfn;
-		vcpu->arch.last_pt_write_count = 1;
-		vcpu->arch.last_pte_updated = NULL;
-	}
-	index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
-	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
-	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) {
-		if (sp->gfn != gfn || sp->role.metaphysical)
-			continue;
-		pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
-		misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1);
-		misaligned |= bytes < 4;
-		if (misaligned || flooded) {
-			/*
-			 * Misaligned accesses are too much trouble to fix
-			 * up; also, they usually indicate a page is not used
-			 * as a page table.
-			 *
-			 * If we're seeing too many writes to a page,
-			 * it may no longer be a page table, or we may be
-			 * forking, in which case it is better to unmap the
-			 * page.
-			 */
-			pgprintk("misaligned: gpa %llx bytes %d role %x\n",
-				 gpa, bytes, sp->role.word);
-			kvm_mmu_zap_page(vcpu->kvm, sp);
-			++vcpu->kvm->stat.mmu_flooded;
-			continue;
-		}
-		page_offset = offset;
-		level = sp->role.level;
-		npte = 1;
-		if (sp->role.glevels == PT32_ROOT_LEVEL) {
-			page_offset <<= 1;	/* 32->64 */
-			/*
-			 * A 32-bit pde maps 4MB while the shadow pdes map
-			 * only 2MB.  So we need to double the offset again
-			 * and zap two pdes instead of one.
-			 */
-			if (level == PT32_ROOT_LEVEL) {
-				page_offset &= ~7; /* kill rounding error */
-				page_offset <<= 1;
-				npte = 2;
-			}
-			quadrant = page_offset >> PAGE_SHIFT;
-			page_offset &= ~PAGE_MASK;
-			if (quadrant != sp->role.quadrant)
-				continue;
-		}
-		spte = &sp->spt[page_offset / sizeof(*spte)];
-		while (npte--) {
-			entry = *spte;
-			mmu_pte_write_zap_pte(vcpu, sp, spte);
-			mmu_pte_write_new_pte(vcpu, sp, spte, new, bytes,
-					      page_offset & (pte_size - 1));
-			mmu_pte_write_flush_tlb(vcpu, entry, *spte);
-			++spte;
-		}
-	}
-	kvm_mmu_audit(vcpu, "post pte write");
-}
-
-int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
-{
-	gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
-
-	return kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
-}
-
-void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
-{
-	while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
-		struct kvm_mmu_page *sp;
-
-		sp = container_of(vcpu->kvm->arch.active_mmu_pages.prev,
-				  struct kvm_mmu_page, link);
-		kvm_mmu_zap_page(vcpu->kvm, sp);
-		++vcpu->kvm->stat.mmu_recycled;
-	}
-}
-
-int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
-{
-	int r;
-	enum emulation_result er;
-
-	mutex_lock(&vcpu->kvm->lock);
-	r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code);
-	if (r < 0)
-		goto out;
-
-	if (!r) {
-		r = 1;
-		goto out;
-	}
-
-	r = mmu_topup_memory_caches(vcpu);
-	if (r)
-		goto out;
-
-	er = emulate_instruction(vcpu, vcpu->run, cr2, error_code, 0);
-	mutex_unlock(&vcpu->kvm->lock);
-
-	switch (er) {
-	case EMULATE_DONE:
-		return 1;
-	case EMULATE_DO_MMIO:
-		++vcpu->stat.mmio_exits;
-		return 0;
-	case EMULATE_FAIL:
-		kvm_report_emulation_failure(vcpu, "pagetable");
-		return 1;
-	default:
-		BUG();
-	}
-out:
-	mutex_unlock(&vcpu->kvm->lock);
-	return r;
-}
-EXPORT_SYMBOL_GPL(kvm_mmu_page_fault);
-
-static void free_mmu_pages(struct kvm_vcpu *vcpu)
-{
-	struct kvm_mmu_page *sp;
-
-	while (!list_empty(&vcpu->kvm->arch.active_mmu_pages)) {
-		sp = container_of(vcpu->kvm->arch.active_mmu_pages.next,
-				  struct kvm_mmu_page, link);
-		kvm_mmu_zap_page(vcpu->kvm, sp);
-	}
-	free_page((unsigned long)vcpu->arch.mmu.pae_root);
-}
-
-static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
-{
-	struct page *page;
-	int i;
-
-	ASSERT(vcpu);
-
-	if (vcpu->kvm->arch.n_requested_mmu_pages)
-		vcpu->kvm->arch.n_free_mmu_pages =
-					vcpu->kvm->arch.n_requested_mmu_pages;
-	else
-		vcpu->kvm->arch.n_free_mmu_pages =
-					vcpu->kvm->arch.n_alloc_mmu_pages;
-	/*
-	 * When emulating 32-bit mode, cr3 is only 32 bits even on x86_64.
-	 * Therefore we need to allocate shadow page tables in the first
-	 * 4GB of memory, which happens to fit the DMA32 zone.
-	 */
-	page = alloc_page(GFP_KERNEL | __GFP_DMA32);
-	if (!page)
-		goto error_1;
-	vcpu->arch.mmu.pae_root = page_address(page);
-	for (i = 0; i < 4; ++i)
-		vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
-
-	return 0;
-
-error_1:
-	free_mmu_pages(vcpu);
-	return -ENOMEM;
-}
-
-int kvm_mmu_create(struct kvm_vcpu *vcpu)
-{
-	ASSERT(vcpu);
-	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
-
-	return alloc_mmu_pages(vcpu);
-}
-
-int kvm_mmu_setup(struct kvm_vcpu *vcpu)
-{
-	ASSERT(vcpu);
-	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
-
-	return init_kvm_mmu(vcpu);
-}
-
-void kvm_mmu_destroy(struct kvm_vcpu *vcpu)
-{
-	ASSERT(vcpu);
-
-	destroy_kvm_mmu(vcpu);
-	free_mmu_pages(vcpu);
-	mmu_free_memory_caches(vcpu);
-}
-
-void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
-{
-	struct kvm_mmu_page *sp;
-
-	list_for_each_entry(sp, &kvm->arch.active_mmu_pages, link) {
-		int i;
-		u64 *pt;
-
-		if (!test_bit(slot, &sp->slot_bitmap))
-			continue;
-
-		pt = sp->spt;
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
-			/* avoid RMW */
-			if (pt[i] & PT_WRITABLE_MASK)
-				pt[i] &= ~PT_WRITABLE_MASK;
-	}
-}
-
-void kvm_mmu_zap_all(struct kvm *kvm)
-{
-	struct kvm_mmu_page *sp, *node;
-
-	list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
-		kvm_mmu_zap_page(kvm, sp);
-
-	kvm_flush_remote_tlbs(kvm);
-}
-
-void kvm_mmu_module_exit(void)
-{
-	if (pte_chain_cache)
-		kmem_cache_destroy(pte_chain_cache);
-	if (rmap_desc_cache)
-		kmem_cache_destroy(rmap_desc_cache);
-	if (mmu_page_header_cache)
-		kmem_cache_destroy(mmu_page_header_cache);
-}
-
-int kvm_mmu_module_init(void)
-{
-	pte_chain_cache = kmem_cache_create("kvm_pte_chain",
-					    sizeof(struct kvm_pte_chain),
-					    0, 0, NULL);
-	if (!pte_chain_cache)
-		goto nomem;
-	rmap_desc_cache = kmem_cache_create("kvm_rmap_desc",
-					    sizeof(struct kvm_rmap_desc),
-					    0, 0, NULL);
-	if (!rmap_desc_cache)
-		goto nomem;
-
-	mmu_page_header_cache = kmem_cache_create("kvm_mmu_page_header",
-						  sizeof(struct kvm_mmu_page),
-						  0, 0, NULL);
-	if (!mmu_page_header_cache)
-		goto nomem;
-
-	return 0;
-
-nomem:
-	kvm_mmu_module_exit();
-	return -ENOMEM;
-}
-
-/*
- * Caculate mmu pages needed for kvm.
- */
-unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm)
-{
-	int i;
-	unsigned int nr_mmu_pages;
-	unsigned int  nr_pages = 0;
-
-	for (i = 0; i < kvm->nmemslots; i++)
-		nr_pages += kvm->memslots[i].npages;
-
-	nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000;
-	nr_mmu_pages = max(nr_mmu_pages,
-			(unsigned int) KVM_MIN_ALLOC_MMU_PAGES);
-
-	return nr_mmu_pages;
-}
-
-#ifdef AUDIT
-
-static const char *audit_msg;
-
-static gva_t canonicalize(gva_t gva)
-{
-#ifdef CONFIG_X86_64
-	gva = (long long)(gva << 16) >> 16;
-#endif
-	return gva;
-}
-
-static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
-				gva_t va, int level)
-{
-	u64 *pt = __va(page_pte & PT64_BASE_ADDR_MASK);
-	int i;
-	gva_t va_delta = 1ul << (PAGE_SHIFT + 9 * (level - 1));
-
-	for (i = 0; i < PT64_ENT_PER_PAGE; ++i, va += va_delta) {
-		u64 ent = pt[i];
-
-		if (ent == shadow_trap_nonpresent_pte)
-			continue;
-
-		va = canonicalize(va);
-		if (level > 1) {
-			if (ent == shadow_notrap_nonpresent_pte)
-				printk(KERN_ERR "audit: (%s) nontrapping pte"
-				       " in nonleaf level: levels %d gva %lx"
-				       " level %d pte %llx\n", audit_msg,
-				       vcpu->arch.mmu.root_level, va, level, ent);
-
-			audit_mappings_page(vcpu, ent, va, level - 1);
-		} else {
-			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);
-			struct page *page = gpa_to_page(vcpu, gpa);
-			hpa_t hpa = page_to_phys(page);
-
-			if (is_shadow_present_pte(ent)
-			    && (ent & PT64_BASE_ADDR_MASK) != hpa)
-				printk(KERN_ERR "xx audit error: (%s) levels %d"
-				       " gva %lx gpa %llx hpa %llx ent %llx %d\n",
-				       audit_msg, vcpu->arch.mmu.root_level,
-				       va, gpa, hpa, ent,
-				       is_shadow_present_pte(ent));
-			else if (ent == shadow_notrap_nonpresent_pte
-				 && !is_error_hpa(hpa))
-				printk(KERN_ERR "audit: (%s) notrap shadow,"
-				       " valid guest gva %lx\n", audit_msg, va);
-			kvm_release_page_clean(page);
-
-		}
-	}
-}
-
-static void audit_mappings(struct kvm_vcpu *vcpu)
-{
-	unsigned i;
-
-	if (vcpu->arch.mmu.root_level == 4)
-		audit_mappings_page(vcpu, vcpu->arch.mmu.root_hpa, 0, 4);
-	else
-		for (i = 0; i < 4; ++i)
-			if (vcpu->arch.mmu.pae_root[i] & PT_PRESENT_MASK)
-				audit_mappings_page(vcpu,
-						    vcpu->arch.mmu.pae_root[i],
-						    i << 30,
-						    2);
-}
-
-static int count_rmaps(struct kvm_vcpu *vcpu)
-{
-	int nmaps = 0;
-	int i, j, k;
-
-	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
-		struct kvm_memory_slot *m = &vcpu->kvm->memslots[i];
-		struct kvm_rmap_desc *d;
-
-		for (j = 0; j < m->npages; ++j) {
-			unsigned long *rmapp = &m->rmap[j];
-
-			if (!*rmapp)
-				continue;
-			if (!(*rmapp & 1)) {
-				++nmaps;
-				continue;
-			}
-			d = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
-			while (d) {
-				for (k = 0; k < RMAP_EXT; ++k)
-					if (d->shadow_ptes[k])
-						++nmaps;
-					else
-						break;
-				d = d->more;
-			}
-		}
-	}
-	return nmaps;
-}
-
-static int count_writable_mappings(struct kvm_vcpu *vcpu)
-{
-	int nmaps = 0;
-	struct kvm_mmu_page *sp;
-	int i;
-
-	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
-		u64 *pt = sp->spt;
-
-		if (sp->role.level != PT_PAGE_TABLE_LEVEL)
-			continue;
-
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
-			u64 ent = pt[i];
-
-			if (!(ent & PT_PRESENT_MASK))
-				continue;
-			if (!(ent & PT_WRITABLE_MASK))
-				continue;
-			++nmaps;
-		}
-	}
-	return nmaps;
-}
-
-static void audit_rmap(struct kvm_vcpu *vcpu)
-{
-	int n_rmap = count_rmaps(vcpu);
-	int n_actual = count_writable_mappings(vcpu);
-
-	if (n_rmap != n_actual)
-		printk(KERN_ERR "%s: (%s) rmap %d actual %d\n",
-		       __FUNCTION__, audit_msg, n_rmap, n_actual);
-}
-
-static void audit_write_protection(struct kvm_vcpu *vcpu)
-{
-	struct kvm_mmu_page *sp;
-	struct kvm_memory_slot *slot;
-	unsigned long *rmapp;
-	gfn_t gfn;
-
-	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
-		if (sp->role.metaphysical)
-			continue;
-
-		slot = gfn_to_memslot(vcpu->kvm, sp->gfn);
-		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
-		rmapp = &slot->rmap[gfn - slot->base_gfn];
-		if (*rmapp)
-			printk(KERN_ERR "%s: (%s) shadow page has writable"
-			       " mappings: gfn %lx role %x\n",
-			       __FUNCTION__, audit_msg, sp->gfn,
-			       sp->role.word);
-	}
-}
-
-static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg)
-{
-	int olddbg = dbg;
-
-	dbg = 0;
-	audit_msg = msg;
-	audit_rmap(vcpu);
-	audit_write_protection(vcpu);
-	audit_mappings(vcpu);
-	dbg = olddbg;
-}
-
-#endif
diff --git a/drivers/kvm/mmu.h b/drivers/kvm/mmu.h
deleted file mode 100644
index cbfc272262df..000000000000
--- a/drivers/kvm/mmu.h
+++ /dev/null
@@ -1,44 +0,0 @@
-#ifndef __KVM_X86_MMU_H
-#define __KVM_X86_MMU_H
-
-#include "kvm.h"
-
-static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
-{
-	if (unlikely(vcpu->kvm->arch.n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
-		__kvm_mmu_free_some_pages(vcpu);
-}
-
-static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
-{
-	if (likely(vcpu->arch.mmu.root_hpa != INVALID_PAGE))
-		return 0;
-
-	return kvm_mmu_load(vcpu);
-}
-
-static inline int is_long_mode(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_X86_64
-	return vcpu->arch.shadow_efer & EFER_LME;
-#else
-	return 0;
-#endif
-}
-
-static inline int is_pae(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.cr4 & X86_CR4_PAE;
-}
-
-static inline int is_pse(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.cr4 & X86_CR4_PSE;
-}
-
-static inline int is_paging(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.cr0 & X86_CR0_PG;
-}
-
-#endif
diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
deleted file mode 100644
index 56b88f7e83ef..000000000000
--- a/drivers/kvm/paging_tmpl.h
+++ /dev/null
@@ -1,461 +0,0 @@
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * This module enables machines with Intel VT-x extensions to run virtual
- * machines without emulation or binary translation.
- *
- * MMU support
- *
- * Copyright (C) 2006 Qumranet, Inc.
- *
- * Authors:
- *   Yaniv Kamay  <yaniv@qumranet.com>
- *   Avi Kivity   <avi@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-
-/*
- * We need the mmu code to access both 32-bit and 64-bit guest ptes,
- * so the code in this file is compiled twice, once per pte size.
- */
-
-#if PTTYPE == 64
-	#define pt_element_t u64
-	#define guest_walker guest_walker64
-	#define FNAME(name) paging##64_##name
-	#define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK
-	#define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK
-	#define PT_INDEX(addr, level) PT64_INDEX(addr, level)
-	#define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
-	#define PT_LEVEL_MASK(level) PT64_LEVEL_MASK(level)
-	#define PT_LEVEL_BITS PT64_LEVEL_BITS
-	#ifdef CONFIG_X86_64
-	#define PT_MAX_FULL_LEVELS 4
-	#define CMPXCHG cmpxchg
-	#else
-	#define CMPXCHG cmpxchg64
-	#define PT_MAX_FULL_LEVELS 2
-	#endif
-#elif PTTYPE == 32
-	#define pt_element_t u32
-	#define guest_walker guest_walker32
-	#define FNAME(name) paging##32_##name
-	#define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK
-	#define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK
-	#define PT_INDEX(addr, level) PT32_INDEX(addr, level)
-	#define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
-	#define PT_LEVEL_MASK(level) PT32_LEVEL_MASK(level)
-	#define PT_LEVEL_BITS PT32_LEVEL_BITS
-	#define PT_MAX_FULL_LEVELS 2
-	#define CMPXCHG cmpxchg
-#else
-	#error Invalid PTTYPE value
-#endif
-
-#define gpte_to_gfn FNAME(gpte_to_gfn)
-#define gpte_to_gfn_pde FNAME(gpte_to_gfn_pde)
-
-/*
- * The guest_walker structure emulates the behavior of the hardware page
- * table walker.
- */
-struct guest_walker {
-	int level;
-	gfn_t table_gfn[PT_MAX_FULL_LEVELS];
-	pt_element_t ptes[PT_MAX_FULL_LEVELS];
-	gpa_t pte_gpa[PT_MAX_FULL_LEVELS];
-	unsigned pt_access;
-	unsigned pte_access;
-	gfn_t gfn;
-	u32 error_code;
-};
-
-static gfn_t gpte_to_gfn(pt_element_t gpte)
-{
-	return (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT;
-}
-
-static gfn_t gpte_to_gfn_pde(pt_element_t gpte)
-{
-	return (gpte & PT_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT;
-}
-
-static bool FNAME(cmpxchg_gpte)(struct kvm *kvm,
-			 gfn_t table_gfn, unsigned index,
-			 pt_element_t orig_pte, pt_element_t new_pte)
-{
-	pt_element_t ret;
-	pt_element_t *table;
-	struct page *page;
-
-	page = gfn_to_page(kvm, table_gfn);
-	table = kmap_atomic(page, KM_USER0);
-
-	ret = CMPXCHG(&table[index], orig_pte, new_pte);
-
-	kunmap_atomic(table, KM_USER0);
-
-	kvm_release_page_dirty(page);
-
-	return (ret != orig_pte);
-}
-
-static unsigned FNAME(gpte_access)(struct kvm_vcpu *vcpu, pt_element_t gpte)
-{
-	unsigned access;
-
-	access = (gpte & (PT_WRITABLE_MASK | PT_USER_MASK)) | ACC_EXEC_MASK;
-#if PTTYPE == 64
-	if (is_nx(vcpu))
-		access &= ~(gpte >> PT64_NX_SHIFT);
-#endif
-	return access;
-}
-
-/*
- * Fetch a guest pte for a guest virtual address
- */
-static int FNAME(walk_addr)(struct guest_walker *walker,
-			    struct kvm_vcpu *vcpu, gva_t addr,
-			    int write_fault, int user_fault, int fetch_fault)
-{
-	pt_element_t pte;
-	gfn_t table_gfn;
-	unsigned index, pt_access, pte_access;
-	gpa_t pte_gpa;
-
-	pgprintk("%s: addr %lx\n", __FUNCTION__, addr);
-walk:
-	walker->level = vcpu->arch.mmu.root_level;
-	pte = vcpu->arch.cr3;
-#if PTTYPE == 64
-	if (!is_long_mode(vcpu)) {
-		pte = vcpu->arch.pdptrs[(addr >> 30) & 3];
-		if (!is_present_pte(pte))
-			goto not_present;
-		--walker->level;
-	}
-#endif
-	ASSERT((!is_long_mode(vcpu) && is_pae(vcpu)) ||
-	       (vcpu->cr3 & CR3_NONPAE_RESERVED_BITS) == 0);
-
-	pt_access = ACC_ALL;
-
-	for (;;) {
-		index = PT_INDEX(addr, walker->level);
-
-		table_gfn = gpte_to_gfn(pte);
-		pte_gpa = gfn_to_gpa(table_gfn);
-		pte_gpa += index * sizeof(pt_element_t);
-		walker->table_gfn[walker->level - 1] = table_gfn;
-		walker->pte_gpa[walker->level - 1] = pte_gpa;
-		pgprintk("%s: table_gfn[%d] %lx\n", __FUNCTION__,
-			 walker->level - 1, table_gfn);
-
-		kvm_read_guest(vcpu->kvm, pte_gpa, &pte, sizeof(pte));
-
-		if (!is_present_pte(pte))
-			goto not_present;
-
-		if (write_fault && !is_writeble_pte(pte))
-			if (user_fault || is_write_protection(vcpu))
-				goto access_error;
-
-		if (user_fault && !(pte & PT_USER_MASK))
-			goto access_error;
-
-#if PTTYPE == 64
-		if (fetch_fault && is_nx(vcpu) && (pte & PT64_NX_MASK))
-			goto access_error;
-#endif
-
-		if (!(pte & PT_ACCESSED_MASK)) {
-			mark_page_dirty(vcpu->kvm, table_gfn);
-			if (FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn,
-			    index, pte, pte|PT_ACCESSED_MASK))
-				goto walk;
-			pte |= PT_ACCESSED_MASK;
-		}
-
-		pte_access = pt_access & FNAME(gpte_access)(vcpu, pte);
-
-		walker->ptes[walker->level - 1] = pte;
-
-		if (walker->level == PT_PAGE_TABLE_LEVEL) {
-			walker->gfn = gpte_to_gfn(pte);
-			break;
-		}
-
-		if (walker->level == PT_DIRECTORY_LEVEL
-		    && (pte & PT_PAGE_SIZE_MASK)
-		    && (PTTYPE == 64 || is_pse(vcpu))) {
-			walker->gfn = gpte_to_gfn_pde(pte);
-			walker->gfn += PT_INDEX(addr, PT_PAGE_TABLE_LEVEL);
-			if (PTTYPE == 32 && is_cpuid_PSE36())
-				walker->gfn += pse36_gfn_delta(pte);
-			break;
-		}
-
-		pt_access = pte_access;
-		--walker->level;
-	}
-
-	if (write_fault && !is_dirty_pte(pte)) {
-		bool ret;
-
-		mark_page_dirty(vcpu->kvm, table_gfn);
-		ret = FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn, index, pte,
-			    pte|PT_DIRTY_MASK);
-		if (ret)
-			goto walk;
-		pte |= PT_DIRTY_MASK;
-		kvm_mmu_pte_write(vcpu, pte_gpa, (u8 *)&pte, sizeof(pte));
-		walker->ptes[walker->level - 1] = pte;
-	}
-
-	walker->pt_access = pt_access;
-	walker->pte_access = pte_access;
-	pgprintk("%s: pte %llx pte_access %x pt_access %x\n",
-		 __FUNCTION__, (u64)pte, pt_access, pte_access);
-	return 1;
-
-not_present:
-	walker->error_code = 0;
-	goto err;
-
-access_error:
-	walker->error_code = PFERR_PRESENT_MASK;
-
-err:
-	if (write_fault)
-		walker->error_code |= PFERR_WRITE_MASK;
-	if (user_fault)
-		walker->error_code |= PFERR_USER_MASK;
-	if (fetch_fault)
-		walker->error_code |= PFERR_FETCH_MASK;
-	return 0;
-}
-
-static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
-			      u64 *spte, const void *pte, int bytes,
-			      int offset_in_pte)
-{
-	pt_element_t gpte;
-	unsigned pte_access;
-
-	gpte = *(const pt_element_t *)pte;
-	if (~gpte & (PT_PRESENT_MASK | PT_ACCESSED_MASK)) {
-		if (!offset_in_pte && !is_present_pte(gpte))
-			set_shadow_pte(spte, shadow_notrap_nonpresent_pte);
-		return;
-	}
-	if (bytes < sizeof(pt_element_t))
-		return;
-	pgprintk("%s: gpte %llx spte %p\n", __FUNCTION__, (u64)gpte, spte);
-	pte_access = page->role.access & FNAME(gpte_access)(vcpu, gpte);
-	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
-		     gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte));
-}
-
-/*
- * Fetch a shadow pte for a specific level in the paging hierarchy.
- */
-static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
-			 struct guest_walker *walker,
-			 int user_fault, int write_fault, int *ptwrite)
-{
-	hpa_t shadow_addr;
-	int level;
-	u64 *shadow_ent;
-	unsigned access = walker->pt_access;
-
-	if (!is_present_pte(walker->ptes[walker->level - 1]))
-		return NULL;
-
-	shadow_addr = vcpu->arch.mmu.root_hpa;
-	level = vcpu->arch.mmu.shadow_root_level;
-	if (level == PT32E_ROOT_LEVEL) {
-		shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
-		shadow_addr &= PT64_BASE_ADDR_MASK;
-		--level;
-	}
-
-	for (; ; level--) {
-		u32 index = SHADOW_PT_INDEX(addr, level);
-		struct kvm_mmu_page *shadow_page;
-		u64 shadow_pte;
-		int metaphysical;
-		gfn_t table_gfn;
-		bool new_page = 0;
-
-		shadow_ent = ((u64 *)__va(shadow_addr)) + index;
-		if (is_shadow_present_pte(*shadow_ent)) {
-			if (level == PT_PAGE_TABLE_LEVEL)
-				break;
-			shadow_addr = *shadow_ent & PT64_BASE_ADDR_MASK;
-			continue;
-		}
-
-		if (level == PT_PAGE_TABLE_LEVEL)
-			break;
-
-		if (level - 1 == PT_PAGE_TABLE_LEVEL
-		    && walker->level == PT_DIRECTORY_LEVEL) {
-			metaphysical = 1;
-			if (!is_dirty_pte(walker->ptes[level - 1]))
-				access &= ~ACC_WRITE_MASK;
-			table_gfn = gpte_to_gfn(walker->ptes[level - 1]);
-		} else {
-			metaphysical = 0;
-			table_gfn = walker->table_gfn[level - 2];
-		}
-		shadow_page = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1,
-					       metaphysical, access,
-					       shadow_ent, &new_page);
-		if (new_page && !metaphysical) {
-			pt_element_t curr_pte;
-			kvm_read_guest(vcpu->kvm, walker->pte_gpa[level - 2],
-				       &curr_pte, sizeof(curr_pte));
-			if (curr_pte != walker->ptes[level - 2])
-				return NULL;
-		}
-		shadow_addr = __pa(shadow_page->spt);
-		shadow_pte = shadow_addr | PT_PRESENT_MASK | PT_ACCESSED_MASK
-			| PT_WRITABLE_MASK | PT_USER_MASK;
-		*shadow_ent = shadow_pte;
-	}
-
-	mmu_set_spte(vcpu, shadow_ent, access, walker->pte_access & access,
-		     user_fault, write_fault,
-		     walker->ptes[walker->level-1] & PT_DIRTY_MASK,
-		     ptwrite, walker->gfn);
-
-	return shadow_ent;
-}
-
-/*
- * Page fault handler.  There are several causes for a page fault:
- *   - there is no shadow pte for the guest pte
- *   - write access through a shadow pte marked read only so that we can set
- *     the dirty bit
- *   - write access to a shadow pte marked read only so we can update the page
- *     dirty bitmap, when userspace requests it
- *   - mmio access; in this case we will never install a present shadow pte
- *   - normal guest page fault due to the guest pte marked not present, not
- *     writable, or not executable
- *
- *  Returns: 1 if we need to emulate the instruction, 0 otherwise, or
- *           a negative value on error.
- */
-static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
-			       u32 error_code)
-{
-	int write_fault = error_code & PFERR_WRITE_MASK;
-	int user_fault = error_code & PFERR_USER_MASK;
-	int fetch_fault = error_code & PFERR_FETCH_MASK;
-	struct guest_walker walker;
-	u64 *shadow_pte;
-	int write_pt = 0;
-	int r;
-
-	pgprintk("%s: addr %lx err %x\n", __FUNCTION__, addr, error_code);
-	kvm_mmu_audit(vcpu, "pre page fault");
-
-	r = mmu_topup_memory_caches(vcpu);
-	if (r)
-		return r;
-
-	/*
-	 * Look up the shadow pte for the faulting address.
-	 */
-	r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault,
-			     fetch_fault);
-
-	/*
-	 * The page is not mapped by the guest.  Let the guest handle it.
-	 */
-	if (!r) {
-		pgprintk("%s: guest page fault\n", __FUNCTION__);
-		inject_page_fault(vcpu, addr, walker.error_code);
-		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
-		return 0;
-	}
-
-	shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
-				  &write_pt);
-	pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
-		 shadow_pte, *shadow_pte, write_pt);
-
-	if (!write_pt)
-		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
-
-	/*
-	 * mmio: emulate if accessible, otherwise its a guest fault.
-	 */
-	if (shadow_pte && is_io_pte(*shadow_pte))
-		return 1;
-
-	++vcpu->stat.pf_fixed;
-	kvm_mmu_audit(vcpu, "post page fault (fixed)");
-
-	return write_pt;
-}
-
-static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr)
-{
-	struct guest_walker walker;
-	gpa_t gpa = UNMAPPED_GVA;
-	int r;
-
-	r = FNAME(walk_addr)(&walker, vcpu, vaddr, 0, 0, 0);
-
-	if (r) {
-		gpa = gfn_to_gpa(walker.gfn);
-		gpa |= vaddr & ~PAGE_MASK;
-	}
-
-	return gpa;
-}
-
-static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
-				 struct kvm_mmu_page *sp)
-{
-	int i, offset = 0;
-	pt_element_t *gpt;
-	struct page *page;
-
-	if (sp->role.metaphysical
-	    || (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) {
-		nonpaging_prefetch_page(vcpu, sp);
-		return;
-	}
-
-	if (PTTYPE == 32)
-		offset = sp->role.quadrant << PT64_LEVEL_BITS;
-	page = gfn_to_page(vcpu->kvm, sp->gfn);
-	gpt = kmap_atomic(page, KM_USER0);
-	for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
-		if (is_present_pte(gpt[offset + i]))
-			sp->spt[i] = shadow_trap_nonpresent_pte;
-		else
-			sp->spt[i] = shadow_notrap_nonpresent_pte;
-	kunmap_atomic(gpt, KM_USER0);
-	kvm_release_page_clean(page);
-}
-
-#undef pt_element_t
-#undef guest_walker
-#undef FNAME
-#undef PT_BASE_ADDR_MASK
-#undef PT_INDEX
-#undef SHADOW_PT_INDEX
-#undef PT_LEVEL_MASK
-#undef PT_DIR_BASE_ADDR_MASK
-#undef PT_LEVEL_BITS
-#undef PT_MAX_FULL_LEVELS
-#undef gpte_to_gfn
-#undef gpte_to_gfn_pde
-#undef CMPXCHG
diff --git a/drivers/kvm/segment_descriptor.h b/drivers/kvm/segment_descriptor.h
deleted file mode 100644
index 56fc4c873389..000000000000
--- a/drivers/kvm/segment_descriptor.h
+++ /dev/null
@@ -1,29 +0,0 @@
-#ifndef __SEGMENT_DESCRIPTOR_H
-#define __SEGMENT_DESCRIPTOR_H
-
-struct segment_descriptor {
-	u16 limit_low;
-	u16 base_low;
-	u8  base_mid;
-	u8  type : 4;
-	u8  system : 1;
-	u8  dpl : 2;
-	u8  present : 1;
-	u8  limit_high : 4;
-	u8  avl : 1;
-	u8  long_mode : 1;
-	u8  default_op : 1;
-	u8  granularity : 1;
-	u8  base_high;
-} __attribute__((packed));
-
-#ifdef CONFIG_X86_64
-/* LDT or TSS descriptor in the GDT. 16 bytes. */
-struct segment_descriptor_64 {
-	struct segment_descriptor s;
-	u32 base_higher;
-	u32 pad_zero;
-};
-
-#endif
-#endif
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
deleted file mode 100644
index e606f6d18669..000000000000
--- a/drivers/kvm/svm.c
+++ /dev/null
@@ -1,1725 +0,0 @@
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * AMD SVM support
- *
- * Copyright (C) 2006 Qumranet, Inc.
- *
- * Authors:
- *   Yaniv Kamay  <yaniv@qumranet.com>
- *   Avi Kivity   <avi@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-#include "x86.h"
-#include "kvm_svm.h"
-#include "x86_emulate.h"
-#include "irq.h"
-#include "mmu.h"
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/vmalloc.h>
-#include <linux/highmem.h>
-#include <linux/sched.h>
-
-#include <asm/desc.h>
-
-MODULE_AUTHOR("Qumranet");
-MODULE_LICENSE("GPL");
-
-#define IOPM_ALLOC_ORDER 2
-#define MSRPM_ALLOC_ORDER 1
-
-#define DB_VECTOR 1
-#define UD_VECTOR 6
-#define GP_VECTOR 13
-
-#define DR7_GD_MASK (1 << 13)
-#define DR6_BD_MASK (1 << 13)
-
-#define SEG_TYPE_LDT 2
-#define SEG_TYPE_BUSY_TSS16 3
-
-#define SVM_FEATURE_NPT  (1 << 0)
-#define SVM_FEATURE_LBRV (1 << 1)
-#define SVM_DEATURE_SVML (1 << 2)
-
-static void kvm_reput_irq(struct vcpu_svm *svm);
-
-static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu)
-{
-	return container_of(vcpu, struct vcpu_svm, vcpu);
-}
-
-unsigned long iopm_base;
-unsigned long msrpm_base;
-
-struct kvm_ldttss_desc {
-	u16 limit0;
-	u16 base0;
-	unsigned base1 : 8, type : 5, dpl : 2, p : 1;
-	unsigned limit1 : 4, zero0 : 3, g : 1, base2 : 8;
-	u32 base3;
-	u32 zero1;
-} __attribute__((packed));
-
-struct svm_cpu_data {
-	int cpu;
-
-	u64 asid_generation;
-	u32 max_asid;
-	u32 next_asid;
-	struct kvm_ldttss_desc *tss_desc;
-
-	struct page *save_area;
-};
-
-static DEFINE_PER_CPU(struct svm_cpu_data *, svm_data);
-static uint32_t svm_features;
-
-struct svm_init_data {
-	int cpu;
-	int r;
-};
-
-static u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000};
-
-#define NUM_MSR_MAPS ARRAY_SIZE(msrpm_ranges)
-#define MSRS_RANGE_SIZE 2048
-#define MSRS_IN_RANGE (MSRS_RANGE_SIZE * 8 / 2)
-
-#define MAX_INST_SIZE 15
-
-static inline u32 svm_has(u32 feat)
-{
-	return svm_features & feat;
-}
-
-static inline u8 pop_irq(struct kvm_vcpu *vcpu)
-{
-	int word_index = __ffs(vcpu->arch.irq_summary);
-	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
-	int irq = word_index * BITS_PER_LONG + bit_index;
-
-	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
-	if (!vcpu->arch.irq_pending[word_index])
-		clear_bit(word_index, &vcpu->arch.irq_summary);
-	return irq;
-}
-
-static inline void push_irq(struct kvm_vcpu *vcpu, u8 irq)
-{
-	set_bit(irq, vcpu->arch.irq_pending);
-	set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
-}
-
-static inline void clgi(void)
-{
-	asm volatile (SVM_CLGI);
-}
-
-static inline void stgi(void)
-{
-	asm volatile (SVM_STGI);
-}
-
-static inline void invlpga(unsigned long addr, u32 asid)
-{
-	asm volatile (SVM_INVLPGA :: "a"(addr), "c"(asid));
-}
-
-static inline unsigned long kvm_read_cr2(void)
-{
-	unsigned long cr2;
-
-	asm volatile ("mov %%cr2, %0" : "=r" (cr2));
-	return cr2;
-}
-
-static inline void kvm_write_cr2(unsigned long val)
-{
-	asm volatile ("mov %0, %%cr2" :: "r" (val));
-}
-
-static inline unsigned long read_dr6(void)
-{
-	unsigned long dr6;
-
-	asm volatile ("mov %%dr6, %0" : "=r" (dr6));
-	return dr6;
-}
-
-static inline void write_dr6(unsigned long val)
-{
-	asm volatile ("mov %0, %%dr6" :: "r" (val));
-}
-
-static inline unsigned long read_dr7(void)
-{
-	unsigned long dr7;
-
-	asm volatile ("mov %%dr7, %0" : "=r" (dr7));
-	return dr7;
-}
-
-static inline void write_dr7(unsigned long val)
-{
-	asm volatile ("mov %0, %%dr7" :: "r" (val));
-}
-
-static inline void force_new_asid(struct kvm_vcpu *vcpu)
-{
-	to_svm(vcpu)->asid_generation--;
-}
-
-static inline void flush_guest_tlb(struct kvm_vcpu *vcpu)
-{
-	force_new_asid(vcpu);
-}
-
-static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer)
-{
-	if (!(efer & EFER_LMA))
-		efer &= ~EFER_LME;
-
-	to_svm(vcpu)->vmcb->save.efer = efer | MSR_EFER_SVME_MASK;
-	vcpu->arch.shadow_efer = efer;
-}
-
-static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
-				bool has_error_code, u32 error_code)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	svm->vmcb->control.event_inj = nr
-		| SVM_EVTINJ_VALID
-		| (has_error_code ? SVM_EVTINJ_VALID_ERR : 0)
-		| SVM_EVTINJ_TYPE_EXEPT;
-	svm->vmcb->control.event_inj_err = error_code;
-}
-
-static bool svm_exception_injected(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	return !(svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_VALID);
-}
-
-static int is_external_interrupt(u32 info)
-{
-	info &= SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID;
-	return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR);
-}
-
-static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	if (!svm->next_rip) {
-		printk(KERN_DEBUG "%s: NOP\n", __FUNCTION__);
-		return;
-	}
-	if (svm->next_rip - svm->vmcb->save.rip > MAX_INST_SIZE)
-		printk(KERN_ERR "%s: ip 0x%llx next 0x%llx\n",
-		       __FUNCTION__,
-		       svm->vmcb->save.rip,
-		       svm->next_rip);
-
-	vcpu->arch.rip = svm->vmcb->save.rip = svm->next_rip;
-	svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
-
-	vcpu->arch.interrupt_window_open = 1;
-}
-
-static int has_svm(void)
-{
-	uint32_t eax, ebx, ecx, edx;
-
-	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) {
-		printk(KERN_INFO "has_svm: not amd\n");
-		return 0;
-	}
-
-	cpuid(0x80000000, &eax, &ebx, &ecx, &edx);
-	if (eax < SVM_CPUID_FUNC) {
-		printk(KERN_INFO "has_svm: can't execute cpuid_8000000a\n");
-		return 0;
-	}
-
-	cpuid(0x80000001, &eax, &ebx, &ecx, &edx);
-	if (!(ecx & (1 << SVM_CPUID_FEATURE_SHIFT))) {
-		printk(KERN_DEBUG "has_svm: svm not available\n");
-		return 0;
-	}
-	return 1;
-}
-
-static void svm_hardware_disable(void *garbage)
-{
-	struct svm_cpu_data *svm_data
-		= per_cpu(svm_data, raw_smp_processor_id());
-
-	if (svm_data) {
-		uint64_t efer;
-
-		wrmsrl(MSR_VM_HSAVE_PA, 0);
-		rdmsrl(MSR_EFER, efer);
-		wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK);
-		per_cpu(svm_data, raw_smp_processor_id()) = NULL;
-		__free_page(svm_data->save_area);
-		kfree(svm_data);
-	}
-}
-
-static void svm_hardware_enable(void *garbage)
-{
-
-	struct svm_cpu_data *svm_data;
-	uint64_t efer;
-#ifdef CONFIG_X86_64
-	struct desc_ptr gdt_descr;
-#else
-	struct desc_ptr gdt_descr;
-#endif
-	struct desc_struct *gdt;
-	int me = raw_smp_processor_id();
-
-	if (!has_svm()) {
-		printk(KERN_ERR "svm_cpu_init: err EOPNOTSUPP on %d\n", me);
-		return;
-	}
-	svm_data = per_cpu(svm_data, me);
-
-	if (!svm_data) {
-		printk(KERN_ERR "svm_cpu_init: svm_data is NULL on %d\n",
-		       me);
-		return;
-	}
-
-	svm_data->asid_generation = 1;
-	svm_data->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1;
-	svm_data->next_asid = svm_data->max_asid + 1;
-	svm_features = cpuid_edx(SVM_CPUID_FUNC);
-
-	asm volatile ("sgdt %0" : "=m"(gdt_descr));
-	gdt = (struct desc_struct *)gdt_descr.address;
-	svm_data->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS);
-
-	rdmsrl(MSR_EFER, efer);
-	wrmsrl(MSR_EFER, efer | MSR_EFER_SVME_MASK);
-
-	wrmsrl(MSR_VM_HSAVE_PA,
-	       page_to_pfn(svm_data->save_area) << PAGE_SHIFT);
-}
-
-static int svm_cpu_init(int cpu)
-{
-	struct svm_cpu_data *svm_data;
-	int r;
-
-	svm_data = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL);
-	if (!svm_data)
-		return -ENOMEM;
-	svm_data->cpu = cpu;
-	svm_data->save_area = alloc_page(GFP_KERNEL);
-	r = -ENOMEM;
-	if (!svm_data->save_area)
-		goto err_1;
-
-	per_cpu(svm_data, cpu) = svm_data;
-
-	return 0;
-
-err_1:
-	kfree(svm_data);
-	return r;
-
-}
-
-static void set_msr_interception(u32 *msrpm, unsigned msr,
-				 int read, int write)
-{
-	int i;
-
-	for (i = 0; i < NUM_MSR_MAPS; i++) {
-		if (msr >= msrpm_ranges[i] &&
-		    msr < msrpm_ranges[i] + MSRS_IN_RANGE) {
-			u32 msr_offset = (i * MSRS_IN_RANGE + msr -
-					  msrpm_ranges[i]) * 2;
-
-			u32 *base = msrpm + (msr_offset / 32);
-			u32 msr_shift = msr_offset % 32;
-			u32 mask = ((write) ? 0 : 2) | ((read) ? 0 : 1);
-			*base = (*base & ~(0x3 << msr_shift)) |
-				(mask << msr_shift);
-			return;
-		}
-	}
-	BUG();
-}
-
-static __init int svm_hardware_setup(void)
-{
-	int cpu;
-	struct page *iopm_pages;
-	struct page *msrpm_pages;
-	void *iopm_va, *msrpm_va;
-	int r;
-
-	iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER);
-
-	if (!iopm_pages)
-		return -ENOMEM;
-
-	iopm_va = page_address(iopm_pages);
-	memset(iopm_va, 0xff, PAGE_SIZE * (1 << IOPM_ALLOC_ORDER));
-	clear_bit(0x80, iopm_va); /* allow direct access to PC debug port */
-	iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT;
-
-
-	msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
-
-	r = -ENOMEM;
-	if (!msrpm_pages)
-		goto err_1;
-
-	msrpm_va = page_address(msrpm_pages);
-	memset(msrpm_va, 0xff, PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER));
-	msrpm_base = page_to_pfn(msrpm_pages) << PAGE_SHIFT;
-
-#ifdef CONFIG_X86_64
-	set_msr_interception(msrpm_va, MSR_GS_BASE, 1, 1);
-	set_msr_interception(msrpm_va, MSR_FS_BASE, 1, 1);
-	set_msr_interception(msrpm_va, MSR_KERNEL_GS_BASE, 1, 1);
-	set_msr_interception(msrpm_va, MSR_LSTAR, 1, 1);
-	set_msr_interception(msrpm_va, MSR_CSTAR, 1, 1);
-	set_msr_interception(msrpm_va, MSR_SYSCALL_MASK, 1, 1);
-#endif
-	set_msr_interception(msrpm_va, MSR_K6_STAR, 1, 1);
-	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_CS, 1, 1);
-	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_ESP, 1, 1);
-	set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_EIP, 1, 1);
-
-	for_each_online_cpu(cpu) {
-		r = svm_cpu_init(cpu);
-		if (r)
-			goto err_2;
-	}
-	return 0;
-
-err_2:
-	__free_pages(msrpm_pages, MSRPM_ALLOC_ORDER);
-	msrpm_base = 0;
-err_1:
-	__free_pages(iopm_pages, IOPM_ALLOC_ORDER);
-	iopm_base = 0;
-	return r;
-}
-
-static __exit void svm_hardware_unsetup(void)
-{
-	__free_pages(pfn_to_page(msrpm_base >> PAGE_SHIFT), MSRPM_ALLOC_ORDER);
-	__free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), IOPM_ALLOC_ORDER);
-	iopm_base = msrpm_base = 0;
-}
-
-static void init_seg(struct vmcb_seg *seg)
-{
-	seg->selector = 0;
-	seg->attrib = SVM_SELECTOR_P_MASK | SVM_SELECTOR_S_MASK |
-		SVM_SELECTOR_WRITE_MASK; /* Read/Write Data Segment */
-	seg->limit = 0xffff;
-	seg->base = 0;
-}
-
-static void init_sys_seg(struct vmcb_seg *seg, uint32_t type)
-{
-	seg->selector = 0;
-	seg->attrib = SVM_SELECTOR_P_MASK | type;
-	seg->limit = 0xffff;
-	seg->base = 0;
-}
-
-static void init_vmcb(struct vmcb *vmcb)
-{
-	struct vmcb_control_area *control = &vmcb->control;
-	struct vmcb_save_area *save = &vmcb->save;
-
-	control->intercept_cr_read = 	INTERCEPT_CR0_MASK |
-					INTERCEPT_CR3_MASK |
-					INTERCEPT_CR4_MASK |
-					INTERCEPT_CR8_MASK;
-
-	control->intercept_cr_write = 	INTERCEPT_CR0_MASK |
-					INTERCEPT_CR3_MASK |
-					INTERCEPT_CR4_MASK |
-					INTERCEPT_CR8_MASK;
-
-	control->intercept_dr_read = 	INTERCEPT_DR0_MASK |
-					INTERCEPT_DR1_MASK |
-					INTERCEPT_DR2_MASK |
-					INTERCEPT_DR3_MASK;
-
-	control->intercept_dr_write = 	INTERCEPT_DR0_MASK |
-					INTERCEPT_DR1_MASK |
-					INTERCEPT_DR2_MASK |
-					INTERCEPT_DR3_MASK |
-					INTERCEPT_DR5_MASK |
-					INTERCEPT_DR7_MASK;
-
-	control->intercept_exceptions = (1 << PF_VECTOR) |
-					(1 << UD_VECTOR);
-
-
-	control->intercept = 	(1ULL << INTERCEPT_INTR) |
-				(1ULL << INTERCEPT_NMI) |
-				(1ULL << INTERCEPT_SMI) |
-		/*
-		 * selective cr0 intercept bug?
-		 *    	0:   0f 22 d8                mov    %eax,%cr3
-		 *	3:   0f 20 c0                mov    %cr0,%eax
-		 *	6:   0d 00 00 00 80          or     $0x80000000,%eax
-		 *	b:   0f 22 c0                mov    %eax,%cr0
-		 * set cr3 ->interception
-		 * get cr0 ->interception
-		 * set cr0 -> no interception
-		 */
-		/*              (1ULL << INTERCEPT_SELECTIVE_CR0) | */
-				(1ULL << INTERCEPT_CPUID) |
-				(1ULL << INTERCEPT_INVD) |
-				(1ULL << INTERCEPT_HLT) |
-				(1ULL << INTERCEPT_INVLPGA) |
-				(1ULL << INTERCEPT_IOIO_PROT) |
-				(1ULL << INTERCEPT_MSR_PROT) |
-				(1ULL << INTERCEPT_TASK_SWITCH) |
-				(1ULL << INTERCEPT_SHUTDOWN) |
-				(1ULL << INTERCEPT_VMRUN) |
-				(1ULL << INTERCEPT_VMMCALL) |
-				(1ULL << INTERCEPT_VMLOAD) |
-				(1ULL << INTERCEPT_VMSAVE) |
-				(1ULL << INTERCEPT_STGI) |
-				(1ULL << INTERCEPT_CLGI) |
-				(1ULL << INTERCEPT_SKINIT) |
-				(1ULL << INTERCEPT_WBINVD) |
-				(1ULL << INTERCEPT_MONITOR) |
-				(1ULL << INTERCEPT_MWAIT);
-
-	control->iopm_base_pa = iopm_base;
-	control->msrpm_base_pa = msrpm_base;
-	control->tsc_offset = 0;
-	control->int_ctl = V_INTR_MASKING_MASK;
-
-	init_seg(&save->es);
-	init_seg(&save->ss);
-	init_seg(&save->ds);
-	init_seg(&save->fs);
-	init_seg(&save->gs);
-
-	save->cs.selector = 0xf000;
-	/* Executable/Readable Code Segment */
-	save->cs.attrib = SVM_SELECTOR_READ_MASK | SVM_SELECTOR_P_MASK |
-		SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK;
-	save->cs.limit = 0xffff;
-	/*
-	 * cs.base should really be 0xffff0000, but vmx can't handle that, so
-	 * be consistent with it.
-	 *
-	 * Replace when we have real mode working for vmx.
-	 */
-	save->cs.base = 0xf0000;
-
-	save->gdtr.limit = 0xffff;
-	save->idtr.limit = 0xffff;
-
-	init_sys_seg(&save->ldtr, SEG_TYPE_LDT);
-	init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16);
-
-	save->efer = MSR_EFER_SVME_MASK;
-	save->dr6 = 0xffff0ff0;
-	save->dr7 = 0x400;
-	save->rflags = 2;
-	save->rip = 0x0000fff0;
-
-	/*
-	 * cr0 val on cpu init should be 0x60000010, we enable cpu
-	 * cache by default. the orderly way is to enable cache in bios.
-	 */
-	save->cr0 = 0x00000010 | X86_CR0_PG | X86_CR0_WP;
-	save->cr4 = X86_CR4_PAE;
-	/* rdx = ?? */
-}
-
-static int svm_vcpu_reset(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	init_vmcb(svm->vmcb);
-
-	if (vcpu->vcpu_id != 0) {
-		svm->vmcb->save.rip = 0;
-		svm->vmcb->save.cs.base = svm->vcpu.arch.sipi_vector << 12;
-		svm->vmcb->save.cs.selector = svm->vcpu.arch.sipi_vector << 8;
-	}
-
-	return 0;
-}
-
-static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
-{
-	struct vcpu_svm *svm;
-	struct page *page;
-	int err;
-
-	svm = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
-	if (!svm) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	err = kvm_vcpu_init(&svm->vcpu, kvm, id);
-	if (err)
-		goto free_svm;
-
-	page = alloc_page(GFP_KERNEL);
-	if (!page) {
-		err = -ENOMEM;
-		goto uninit;
-	}
-
-	svm->vmcb = page_address(page);
-	clear_page(svm->vmcb);
-	svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT;
-	svm->asid_generation = 0;
-	memset(svm->db_regs, 0, sizeof(svm->db_regs));
-	init_vmcb(svm->vmcb);
-
-	fx_init(&svm->vcpu);
-	svm->vcpu.fpu_active = 1;
-	svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
-	if (svm->vcpu.vcpu_id == 0)
-		svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP;
-
-	return &svm->vcpu;
-
-uninit:
-	kvm_vcpu_uninit(&svm->vcpu);
-free_svm:
-	kmem_cache_free(kvm_vcpu_cache, svm);
-out:
-	return ERR_PTR(err);
-}
-
-static void svm_free_vcpu(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	__free_page(pfn_to_page(svm->vmcb_pa >> PAGE_SHIFT));
-	kvm_vcpu_uninit(vcpu);
-	kmem_cache_free(kvm_vcpu_cache, svm);
-}
-
-static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	int i;
-
-	if (unlikely(cpu != vcpu->cpu)) {
-		u64 tsc_this, delta;
-
-		/*
-		 * Make sure that the guest sees a monotonically
-		 * increasing TSC.
-		 */
-		rdtscll(tsc_this);
-		delta = vcpu->arch.host_tsc - tsc_this;
-		svm->vmcb->control.tsc_offset += delta;
-		vcpu->cpu = cpu;
-		kvm_migrate_apic_timer(vcpu);
-	}
-
-	for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
-		rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
-}
-
-static void svm_vcpu_put(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	int i;
-
-	++vcpu->stat.host_state_reload;
-	for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
-		wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
-
-	rdtscll(vcpu->arch.host_tsc);
-}
-
-static void svm_vcpu_decache(struct kvm_vcpu *vcpu)
-{
-}
-
-static void svm_cache_regs(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
-	vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
-	vcpu->arch.rip = svm->vmcb->save.rip;
-}
-
-static void svm_decache_regs(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX];
-	svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP];
-	svm->vmcb->save.rip = vcpu->arch.rip;
-}
-
-static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu)
-{
-	return to_svm(vcpu)->vmcb->save.rflags;
-}
-
-static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
-{
-	to_svm(vcpu)->vmcb->save.rflags = rflags;
-}
-
-static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg)
-{
-	struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save;
-
-	switch (seg) {
-	case VCPU_SREG_CS: return &save->cs;
-	case VCPU_SREG_DS: return &save->ds;
-	case VCPU_SREG_ES: return &save->es;
-	case VCPU_SREG_FS: return &save->fs;
-	case VCPU_SREG_GS: return &save->gs;
-	case VCPU_SREG_SS: return &save->ss;
-	case VCPU_SREG_TR: return &save->tr;
-	case VCPU_SREG_LDTR: return &save->ldtr;
-	}
-	BUG();
-	return NULL;
-}
-
-static u64 svm_get_segment_base(struct kvm_vcpu *vcpu, int seg)
-{
-	struct vmcb_seg *s = svm_seg(vcpu, seg);
-
-	return s->base;
-}
-
-static void svm_get_segment(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg)
-{
-	struct vmcb_seg *s = svm_seg(vcpu, seg);
-
-	var->base = s->base;
-	var->limit = s->limit;
-	var->selector = s->selector;
-	var->type = s->attrib & SVM_SELECTOR_TYPE_MASK;
-	var->s = (s->attrib >> SVM_SELECTOR_S_SHIFT) & 1;
-	var->dpl = (s->attrib >> SVM_SELECTOR_DPL_SHIFT) & 3;
-	var->present = (s->attrib >> SVM_SELECTOR_P_SHIFT) & 1;
-	var->avl = (s->attrib >> SVM_SELECTOR_AVL_SHIFT) & 1;
-	var->l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1;
-	var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1;
-	var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1;
-	var->unusable = !var->present;
-}
-
-static void svm_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	dt->limit = svm->vmcb->save.idtr.limit;
-	dt->base = svm->vmcb->save.idtr.base;
-}
-
-static void svm_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	svm->vmcb->save.idtr.limit = dt->limit;
-	svm->vmcb->save.idtr.base = dt->base ;
-}
-
-static void svm_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	dt->limit = svm->vmcb->save.gdtr.limit;
-	dt->base = svm->vmcb->save.gdtr.base;
-}
-
-static void svm_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	svm->vmcb->save.gdtr.limit = dt->limit;
-	svm->vmcb->save.gdtr.base = dt->base ;
-}
-
-static void svm_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
-{
-}
-
-static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-#ifdef CONFIG_X86_64
-	if (vcpu->arch.shadow_efer & EFER_LME) {
-		if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
-			vcpu->arch.shadow_efer |= EFER_LMA;
-			svm->vmcb->save.efer |= EFER_LMA | EFER_LME;
-		}
-
-		if (is_paging(vcpu) && !(cr0 & X86_CR0_PG)) {
-			vcpu->arch.shadow_efer &= ~EFER_LMA;
-			svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME);
-		}
-	}
-#endif
-	if ((vcpu->arch.cr0 & X86_CR0_TS) && !(cr0 & X86_CR0_TS)) {
-		svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
-		vcpu->fpu_active = 1;
-	}
-
-	vcpu->arch.cr0 = cr0;
-	cr0 |= X86_CR0_PG | X86_CR0_WP;
-	cr0 &= ~(X86_CR0_CD | X86_CR0_NW);
-	svm->vmcb->save.cr0 = cr0;
-}
-
-static void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
-{
-       vcpu->arch.cr4 = cr4;
-       to_svm(vcpu)->vmcb->save.cr4 = cr4 | X86_CR4_PAE;
-}
-
-static void svm_set_segment(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	struct vmcb_seg *s = svm_seg(vcpu, seg);
-
-	s->base = var->base;
-	s->limit = var->limit;
-	s->selector = var->selector;
-	if (var->unusable)
-		s->attrib = 0;
-	else {
-		s->attrib = (var->type & SVM_SELECTOR_TYPE_MASK);
-		s->attrib |= (var->s & 1) << SVM_SELECTOR_S_SHIFT;
-		s->attrib |= (var->dpl & 3) << SVM_SELECTOR_DPL_SHIFT;
-		s->attrib |= (var->present & 1) << SVM_SELECTOR_P_SHIFT;
-		s->attrib |= (var->avl & 1) << SVM_SELECTOR_AVL_SHIFT;
-		s->attrib |= (var->l & 1) << SVM_SELECTOR_L_SHIFT;
-		s->attrib |= (var->db & 1) << SVM_SELECTOR_DB_SHIFT;
-		s->attrib |= (var->g & 1) << SVM_SELECTOR_G_SHIFT;
-	}
-	if (seg == VCPU_SREG_CS)
-		svm->vmcb->save.cpl
-			= (svm->vmcb->save.cs.attrib
-			   >> SVM_SELECTOR_DPL_SHIFT) & 3;
-
-}
-
-/* FIXME:
-
-	svm(vcpu)->vmcb->control.int_ctl &= ~V_TPR_MASK;
-	svm(vcpu)->vmcb->control.int_ctl |= (sregs->cr8 & V_TPR_MASK);
-
-*/
-
-static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
-{
-	return -EOPNOTSUPP;
-}
-
-static int svm_get_irq(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	u32 exit_int_info = svm->vmcb->control.exit_int_info;
-
-	if (is_external_interrupt(exit_int_info))
-		return exit_int_info & SVM_EVTINJ_VEC_MASK;
-	return -1;
-}
-
-static void load_host_msrs(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_X86_64
-	wrmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base);
-#endif
-}
-
-static void save_host_msrs(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_X86_64
-	rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base);
-#endif
-}
-
-static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *svm_data)
-{
-	if (svm_data->next_asid > svm_data->max_asid) {
-		++svm_data->asid_generation;
-		svm_data->next_asid = 1;
-		svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ALL_ASID;
-	}
-
-	svm->vcpu.cpu = svm_data->cpu;
-	svm->asid_generation = svm_data->asid_generation;
-	svm->vmcb->control.asid = svm_data->next_asid++;
-}
-
-static unsigned long svm_get_dr(struct kvm_vcpu *vcpu, int dr)
-{
-	return to_svm(vcpu)->db_regs[dr];
-}
-
-static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value,
-		       int *exception)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	*exception = 0;
-
-	if (svm->vmcb->save.dr7 & DR7_GD_MASK) {
-		svm->vmcb->save.dr7 &= ~DR7_GD_MASK;
-		svm->vmcb->save.dr6 |= DR6_BD_MASK;
-		*exception = DB_VECTOR;
-		return;
-	}
-
-	switch (dr) {
-	case 0 ... 3:
-		svm->db_regs[dr] = value;
-		return;
-	case 4 ... 5:
-		if (vcpu->arch.cr4 & X86_CR4_DE) {
-			*exception = UD_VECTOR;
-			return;
-		}
-	case 7: {
-		if (value & ~((1ULL << 32) - 1)) {
-			*exception = GP_VECTOR;
-			return;
-		}
-		svm->vmcb->save.dr7 = value;
-		return;
-	}
-	default:
-		printk(KERN_DEBUG "%s: unexpected dr %u\n",
-		       __FUNCTION__, dr);
-		*exception = UD_VECTOR;
-		return;
-	}
-}
-
-static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	u32 exit_int_info = svm->vmcb->control.exit_int_info;
-	struct kvm *kvm = svm->vcpu.kvm;
-	u64 fault_address;
-	u32 error_code;
-
-	if (!irqchip_in_kernel(kvm) &&
-		is_external_interrupt(exit_int_info))
-		push_irq(&svm->vcpu, exit_int_info & SVM_EVTINJ_VEC_MASK);
-
-	fault_address  = svm->vmcb->control.exit_info_2;
-	error_code = svm->vmcb->control.exit_info_1;
-	return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code);
-}
-
-static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	int er;
-
-	er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0);
-	if (er != EMULATE_DONE)
-		kvm_queue_exception(&svm->vcpu, UD_VECTOR);
-	return 1;
-}
-
-static int nm_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
-	if (!(svm->vcpu.arch.cr0 & X86_CR0_TS))
-		svm->vmcb->save.cr0 &= ~X86_CR0_TS;
-	svm->vcpu.fpu_active = 1;
-
-	return 1;
-}
-
-static int shutdown_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	/*
-	 * VMCB is undefined after a SHUTDOWN intercept
-	 * so reinitialize it.
-	 */
-	clear_page(svm->vmcb);
-	init_vmcb(svm->vmcb);
-
-	kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
-	return 0;
-}
-
-static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */
-	int size, down, in, string, rep;
-	unsigned port;
-
-	++svm->vcpu.stat.io_exits;
-
-	svm->next_rip = svm->vmcb->control.exit_info_2;
-
-	string = (io_info & SVM_IOIO_STR_MASK) != 0;
-
-	if (string) {
-		if (emulate_instruction(&svm->vcpu,
-					kvm_run, 0, 0, 0) == EMULATE_DO_MMIO)
-			return 0;
-		return 1;
-	}
-
-	in = (io_info & SVM_IOIO_TYPE_MASK) != 0;
-	port = io_info >> 16;
-	size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT;
-	rep = (io_info & SVM_IOIO_REP_MASK) != 0;
-	down = (svm->vmcb->save.rflags & X86_EFLAGS_DF) != 0;
-
-	return kvm_emulate_pio(&svm->vcpu, kvm_run, in, size, port);
-}
-
-static int nop_on_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	return 1;
-}
-
-static int halt_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	svm->next_rip = svm->vmcb->save.rip + 1;
-	skip_emulated_instruction(&svm->vcpu);
-	return kvm_emulate_halt(&svm->vcpu);
-}
-
-static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	svm->next_rip = svm->vmcb->save.rip + 3;
-	skip_emulated_instruction(&svm->vcpu);
-	kvm_emulate_hypercall(&svm->vcpu);
-	return 1;
-}
-
-static int invalid_op_interception(struct vcpu_svm *svm,
-				   struct kvm_run *kvm_run)
-{
-	kvm_queue_exception(&svm->vcpu, UD_VECTOR);
-	return 1;
-}
-
-static int task_switch_interception(struct vcpu_svm *svm,
-				    struct kvm_run *kvm_run)
-{
-	pr_unimpl(&svm->vcpu, "%s: task switch is unsupported\n", __FUNCTION__);
-	kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
-	return 0;
-}
-
-static int cpuid_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	svm->next_rip = svm->vmcb->save.rip + 2;
-	kvm_emulate_cpuid(&svm->vcpu);
-	return 1;
-}
-
-static int emulate_on_interception(struct vcpu_svm *svm,
-				   struct kvm_run *kvm_run)
-{
-	if (emulate_instruction(&svm->vcpu, NULL, 0, 0, 0) != EMULATE_DONE)
-		pr_unimpl(&svm->vcpu, "%s: failed\n", __FUNCTION__);
-	return 1;
-}
-
-static int cr8_write_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	emulate_instruction(&svm->vcpu, NULL, 0, 0, 0);
-	if (irqchip_in_kernel(svm->vcpu.kvm))
-		return 1;
-	kvm_run->exit_reason = KVM_EXIT_SET_TPR;
-	return 0;
-}
-
-static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	switch (ecx) {
-	case MSR_IA32_TIME_STAMP_COUNTER: {
-		u64 tsc;
-
-		rdtscll(tsc);
-		*data = svm->vmcb->control.tsc_offset + tsc;
-		break;
-	}
-	case MSR_K6_STAR:
-		*data = svm->vmcb->save.star;
-		break;
-#ifdef CONFIG_X86_64
-	case MSR_LSTAR:
-		*data = svm->vmcb->save.lstar;
-		break;
-	case MSR_CSTAR:
-		*data = svm->vmcb->save.cstar;
-		break;
-	case MSR_KERNEL_GS_BASE:
-		*data = svm->vmcb->save.kernel_gs_base;
-		break;
-	case MSR_SYSCALL_MASK:
-		*data = svm->vmcb->save.sfmask;
-		break;
-#endif
-	case MSR_IA32_SYSENTER_CS:
-		*data = svm->vmcb->save.sysenter_cs;
-		break;
-	case MSR_IA32_SYSENTER_EIP:
-		*data = svm->vmcb->save.sysenter_eip;
-		break;
-	case MSR_IA32_SYSENTER_ESP:
-		*data = svm->vmcb->save.sysenter_esp;
-		break;
-	default:
-		return kvm_get_msr_common(vcpu, ecx, data);
-	}
-	return 0;
-}
-
-static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
-	u64 data;
-
-	if (svm_get_msr(&svm->vcpu, ecx, &data))
-		kvm_inject_gp(&svm->vcpu, 0);
-	else {
-		svm->vmcb->save.rax = data & 0xffffffff;
-		svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32;
-		svm->next_rip = svm->vmcb->save.rip + 2;
-		skip_emulated_instruction(&svm->vcpu);
-	}
-	return 1;
-}
-
-static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	switch (ecx) {
-	case MSR_IA32_TIME_STAMP_COUNTER: {
-		u64 tsc;
-
-		rdtscll(tsc);
-		svm->vmcb->control.tsc_offset = data - tsc;
-		break;
-	}
-	case MSR_K6_STAR:
-		svm->vmcb->save.star = data;
-		break;
-#ifdef CONFIG_X86_64
-	case MSR_LSTAR:
-		svm->vmcb->save.lstar = data;
-		break;
-	case MSR_CSTAR:
-		svm->vmcb->save.cstar = data;
-		break;
-	case MSR_KERNEL_GS_BASE:
-		svm->vmcb->save.kernel_gs_base = data;
-		break;
-	case MSR_SYSCALL_MASK:
-		svm->vmcb->save.sfmask = data;
-		break;
-#endif
-	case MSR_IA32_SYSENTER_CS:
-		svm->vmcb->save.sysenter_cs = data;
-		break;
-	case MSR_IA32_SYSENTER_EIP:
-		svm->vmcb->save.sysenter_eip = data;
-		break;
-	case MSR_IA32_SYSENTER_ESP:
-		svm->vmcb->save.sysenter_esp = data;
-		break;
-	case MSR_K7_EVNTSEL0:
-	case MSR_K7_EVNTSEL1:
-	case MSR_K7_EVNTSEL2:
-	case MSR_K7_EVNTSEL3:
-		/*
-		 * only support writing 0 to the performance counters for now
-		 * to make Windows happy. Should be replaced by a real
-		 * performance counter emulation later.
-		 */
-		if (data != 0)
-			goto unhandled;
-		break;
-	default:
-	unhandled:
-		return kvm_set_msr_common(vcpu, ecx, data);
-	}
-	return 0;
-}
-
-static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
-	u64 data = (svm->vmcb->save.rax & -1u)
-		| ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32);
-	svm->next_rip = svm->vmcb->save.rip + 2;
-	if (svm_set_msr(&svm->vcpu, ecx, data))
-		kvm_inject_gp(&svm->vcpu, 0);
-	else
-		skip_emulated_instruction(&svm->vcpu);
-	return 1;
-}
-
-static int msr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
-{
-	if (svm->vmcb->control.exit_info_1)
-		return wrmsr_interception(svm, kvm_run);
-	else
-		return rdmsr_interception(svm, kvm_run);
-}
-
-static int interrupt_window_interception(struct vcpu_svm *svm,
-				   struct kvm_run *kvm_run)
-{
-	svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR);
-	svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
-	/*
-	 * If the user space waits to inject interrupts, exit as soon as
-	 * possible
-	 */
-	if (kvm_run->request_interrupt_window &&
-	    !svm->vcpu.arch.irq_summary) {
-		++svm->vcpu.stat.irq_window_exits;
-		kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
-		return 0;
-	}
-
-	return 1;
-}
-
-static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
-				      struct kvm_run *kvm_run) = {
-	[SVM_EXIT_READ_CR0]           		= emulate_on_interception,
-	[SVM_EXIT_READ_CR3]           		= emulate_on_interception,
-	[SVM_EXIT_READ_CR4]           		= emulate_on_interception,
-	[SVM_EXIT_READ_CR8]           		= emulate_on_interception,
-	/* for now: */
-	[SVM_EXIT_WRITE_CR0]          		= emulate_on_interception,
-	[SVM_EXIT_WRITE_CR3]          		= emulate_on_interception,
-	[SVM_EXIT_WRITE_CR4]          		= emulate_on_interception,
-	[SVM_EXIT_WRITE_CR8]          		= cr8_write_interception,
-	[SVM_EXIT_READ_DR0] 			= emulate_on_interception,
-	[SVM_EXIT_READ_DR1]			= emulate_on_interception,
-	[SVM_EXIT_READ_DR2]			= emulate_on_interception,
-	[SVM_EXIT_READ_DR3]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR0]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR1]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR2]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR3]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR5]			= emulate_on_interception,
-	[SVM_EXIT_WRITE_DR7]			= emulate_on_interception,
-	[SVM_EXIT_EXCP_BASE + UD_VECTOR]	= ud_interception,
-	[SVM_EXIT_EXCP_BASE + PF_VECTOR] 	= pf_interception,
-	[SVM_EXIT_EXCP_BASE + NM_VECTOR] 	= nm_interception,
-	[SVM_EXIT_INTR] 			= nop_on_interception,
-	[SVM_EXIT_NMI]				= nop_on_interception,
-	[SVM_EXIT_SMI]				= nop_on_interception,
-	[SVM_EXIT_INIT]				= nop_on_interception,
-	[SVM_EXIT_VINTR]			= interrupt_window_interception,
-	/* [SVM_EXIT_CR0_SEL_WRITE]		= emulate_on_interception, */
-	[SVM_EXIT_CPUID]			= cpuid_interception,
-	[SVM_EXIT_INVD]                         = emulate_on_interception,
-	[SVM_EXIT_HLT]				= halt_interception,
-	[SVM_EXIT_INVLPG]			= emulate_on_interception,
-	[SVM_EXIT_INVLPGA]			= invalid_op_interception,
-	[SVM_EXIT_IOIO] 		  	= io_interception,
-	[SVM_EXIT_MSR]				= msr_interception,
-	[SVM_EXIT_TASK_SWITCH]			= task_switch_interception,
-	[SVM_EXIT_SHUTDOWN]			= shutdown_interception,
-	[SVM_EXIT_VMRUN]			= invalid_op_interception,
-	[SVM_EXIT_VMMCALL]			= vmmcall_interception,
-	[SVM_EXIT_VMLOAD]			= invalid_op_interception,
-	[SVM_EXIT_VMSAVE]			= invalid_op_interception,
-	[SVM_EXIT_STGI]				= invalid_op_interception,
-	[SVM_EXIT_CLGI]				= invalid_op_interception,
-	[SVM_EXIT_SKINIT]			= invalid_op_interception,
-	[SVM_EXIT_WBINVD]                       = emulate_on_interception,
-	[SVM_EXIT_MONITOR]			= invalid_op_interception,
-	[SVM_EXIT_MWAIT]			= invalid_op_interception,
-};
-
-
-static int handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	u32 exit_code = svm->vmcb->control.exit_code;
-
-	kvm_reput_irq(svm);
-
-	if (svm->vmcb->control.exit_code == SVM_EXIT_ERR) {
-		kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
-		kvm_run->fail_entry.hardware_entry_failure_reason
-			= svm->vmcb->control.exit_code;
-		return 0;
-	}
-
-	if (is_external_interrupt(svm->vmcb->control.exit_int_info) &&
-	    exit_code != SVM_EXIT_EXCP_BASE + PF_VECTOR)
-		printk(KERN_ERR "%s: unexpected exit_ini_info 0x%x "
-		       "exit_code 0x%x\n",
-		       __FUNCTION__, svm->vmcb->control.exit_int_info,
-		       exit_code);
-
-	if (exit_code >= ARRAY_SIZE(svm_exit_handlers)
-	    || !svm_exit_handlers[exit_code]) {
-		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
-		kvm_run->hw.hardware_exit_reason = exit_code;
-		return 0;
-	}
-
-	return svm_exit_handlers[exit_code](svm, kvm_run);
-}
-
-static void reload_tss(struct kvm_vcpu *vcpu)
-{
-	int cpu = raw_smp_processor_id();
-
-	struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu);
-	svm_data->tss_desc->type = 9; /* available 32/64-bit TSS */
-	load_TR_desc();
-}
-
-static void pre_svm_run(struct vcpu_svm *svm)
-{
-	int cpu = raw_smp_processor_id();
-
-	struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu);
-
-	svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
-	if (svm->vcpu.cpu != cpu ||
-	    svm->asid_generation != svm_data->asid_generation)
-		new_asid(svm, svm_data);
-}
-
-
-static inline void svm_inject_irq(struct vcpu_svm *svm, int irq)
-{
-	struct vmcb_control_area *control;
-
-	control = &svm->vmcb->control;
-	control->int_vector = irq;
-	control->int_ctl &= ~V_INTR_PRIO_MASK;
-	control->int_ctl |= V_IRQ_MASK |
-		((/*control->int_vector >> 4*/ 0xf) << V_INTR_PRIO_SHIFT);
-}
-
-static void svm_set_irq(struct kvm_vcpu *vcpu, int irq)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	svm_inject_irq(svm, irq);
-}
-
-static void svm_intr_assist(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	struct vmcb *vmcb = svm->vmcb;
-	int intr_vector = -1;
-
-	if ((vmcb->control.exit_int_info & SVM_EVTINJ_VALID) &&
-	    ((vmcb->control.exit_int_info & SVM_EVTINJ_TYPE_MASK) == 0)) {
-		intr_vector = vmcb->control.exit_int_info &
-			      SVM_EVTINJ_VEC_MASK;
-		vmcb->control.exit_int_info = 0;
-		svm_inject_irq(svm, intr_vector);
-		return;
-	}
-
-	if (vmcb->control.int_ctl & V_IRQ_MASK)
-		return;
-
-	if (!kvm_cpu_has_interrupt(vcpu))
-		return;
-
-	if (!(vmcb->save.rflags & X86_EFLAGS_IF) ||
-	    (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) ||
-	    (vmcb->control.event_inj & SVM_EVTINJ_VALID)) {
-		/* unable to deliver irq, set pending irq */
-		vmcb->control.intercept |= (1ULL << INTERCEPT_VINTR);
-		svm_inject_irq(svm, 0x0);
-		return;
-	}
-	/* Okay, we can deliver the interrupt: grab it and update PIC state. */
-	intr_vector = kvm_cpu_get_interrupt(vcpu);
-	svm_inject_irq(svm, intr_vector);
-	kvm_timer_intr_post(vcpu, intr_vector);
-}
-
-static void kvm_reput_irq(struct vcpu_svm *svm)
-{
-	struct vmcb_control_area *control = &svm->vmcb->control;
-
-	if ((control->int_ctl & V_IRQ_MASK)
-	    && !irqchip_in_kernel(svm->vcpu.kvm)) {
-		control->int_ctl &= ~V_IRQ_MASK;
-		push_irq(&svm->vcpu, control->int_vector);
-	}
-
-	svm->vcpu.arch.interrupt_window_open =
-		!(control->int_state & SVM_INTERRUPT_SHADOW_MASK);
-}
-
-static void svm_do_inject_vector(struct vcpu_svm *svm)
-{
-	struct kvm_vcpu *vcpu = &svm->vcpu;
-	int word_index = __ffs(vcpu->arch.irq_summary);
-	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
-	int irq = word_index * BITS_PER_LONG + bit_index;
-
-	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
-	if (!vcpu->arch.irq_pending[word_index])
-		clear_bit(word_index, &vcpu->arch.irq_summary);
-	svm_inject_irq(svm, irq);
-}
-
-static void do_interrupt_requests(struct kvm_vcpu *vcpu,
-				       struct kvm_run *kvm_run)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	struct vmcb_control_area *control = &svm->vmcb->control;
-
-	svm->vcpu.arch.interrupt_window_open =
-		(!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) &&
-		 (svm->vmcb->save.rflags & X86_EFLAGS_IF));
-
-	if (svm->vcpu.arch.interrupt_window_open && svm->vcpu.arch.irq_summary)
-		/*
-		 * If interrupts enabled, and not blocked by sti or mov ss. Good.
-		 */
-		svm_do_inject_vector(svm);
-
-	/*
-	 * Interrupts blocked.  Wait for unblock.
-	 */
-	if (!svm->vcpu.arch.interrupt_window_open &&
-	    (svm->vcpu.arch.irq_summary || kvm_run->request_interrupt_window))
-		control->intercept |= 1ULL << INTERCEPT_VINTR;
-	 else
-		control->intercept &= ~(1ULL << INTERCEPT_VINTR);
-}
-
-static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
-{
-	return 0;
-}
-
-static void save_db_regs(unsigned long *db_regs)
-{
-	asm volatile ("mov %%dr0, %0" : "=r"(db_regs[0]));
-	asm volatile ("mov %%dr1, %0" : "=r"(db_regs[1]));
-	asm volatile ("mov %%dr2, %0" : "=r"(db_regs[2]));
-	asm volatile ("mov %%dr3, %0" : "=r"(db_regs[3]));
-}
-
-static void load_db_regs(unsigned long *db_regs)
-{
-	asm volatile ("mov %0, %%dr0" : : "r"(db_regs[0]));
-	asm volatile ("mov %0, %%dr1" : : "r"(db_regs[1]));
-	asm volatile ("mov %0, %%dr2" : : "r"(db_regs[2]));
-	asm volatile ("mov %0, %%dr3" : : "r"(db_regs[3]));
-}
-
-static void svm_flush_tlb(struct kvm_vcpu *vcpu)
-{
-	force_new_asid(vcpu);
-}
-
-static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu)
-{
-}
-
-static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-	u16 fs_selector;
-	u16 gs_selector;
-	u16 ldt_selector;
-
-	pre_svm_run(svm);
-
-	save_host_msrs(vcpu);
-	fs_selector = read_fs();
-	gs_selector = read_gs();
-	ldt_selector = read_ldt();
-	svm->host_cr2 = kvm_read_cr2();
-	svm->host_dr6 = read_dr6();
-	svm->host_dr7 = read_dr7();
-	svm->vmcb->save.cr2 = vcpu->arch.cr2;
-
-	if (svm->vmcb->save.dr7 & 0xff) {
-		write_dr7(0);
-		save_db_regs(svm->host_db_regs);
-		load_db_regs(svm->db_regs);
-	}
-
-	clgi();
-
-	local_irq_enable();
-
-	asm volatile (
-#ifdef CONFIG_X86_64
-		"push %%rbp; \n\t"
-#else
-		"push %%ebp; \n\t"
-#endif
-
-#ifdef CONFIG_X86_64
-		"mov %c[rbx](%[svm]), %%rbx \n\t"
-		"mov %c[rcx](%[svm]), %%rcx \n\t"
-		"mov %c[rdx](%[svm]), %%rdx \n\t"
-		"mov %c[rsi](%[svm]), %%rsi \n\t"
-		"mov %c[rdi](%[svm]), %%rdi \n\t"
-		"mov %c[rbp](%[svm]), %%rbp \n\t"
-		"mov %c[r8](%[svm]),  %%r8  \n\t"
-		"mov %c[r9](%[svm]),  %%r9  \n\t"
-		"mov %c[r10](%[svm]), %%r10 \n\t"
-		"mov %c[r11](%[svm]), %%r11 \n\t"
-		"mov %c[r12](%[svm]), %%r12 \n\t"
-		"mov %c[r13](%[svm]), %%r13 \n\t"
-		"mov %c[r14](%[svm]), %%r14 \n\t"
-		"mov %c[r15](%[svm]), %%r15 \n\t"
-#else
-		"mov %c[rbx](%[svm]), %%ebx \n\t"
-		"mov %c[rcx](%[svm]), %%ecx \n\t"
-		"mov %c[rdx](%[svm]), %%edx \n\t"
-		"mov %c[rsi](%[svm]), %%esi \n\t"
-		"mov %c[rdi](%[svm]), %%edi \n\t"
-		"mov %c[rbp](%[svm]), %%ebp \n\t"
-#endif
-
-#ifdef CONFIG_X86_64
-		/* Enter guest mode */
-		"push %%rax \n\t"
-		"mov %c[vmcb](%[svm]), %%rax \n\t"
-		SVM_VMLOAD "\n\t"
-		SVM_VMRUN "\n\t"
-		SVM_VMSAVE "\n\t"
-		"pop %%rax \n\t"
-#else
-		/* Enter guest mode */
-		"push %%eax \n\t"
-		"mov %c[vmcb](%[svm]), %%eax \n\t"
-		SVM_VMLOAD "\n\t"
-		SVM_VMRUN "\n\t"
-		SVM_VMSAVE "\n\t"
-		"pop %%eax \n\t"
-#endif
-
-		/* Save guest registers, load host registers */
-#ifdef CONFIG_X86_64
-		"mov %%rbx, %c[rbx](%[svm]) \n\t"
-		"mov %%rcx, %c[rcx](%[svm]) \n\t"
-		"mov %%rdx, %c[rdx](%[svm]) \n\t"
-		"mov %%rsi, %c[rsi](%[svm]) \n\t"
-		"mov %%rdi, %c[rdi](%[svm]) \n\t"
-		"mov %%rbp, %c[rbp](%[svm]) \n\t"
-		"mov %%r8,  %c[r8](%[svm]) \n\t"
-		"mov %%r9,  %c[r9](%[svm]) \n\t"
-		"mov %%r10, %c[r10](%[svm]) \n\t"
-		"mov %%r11, %c[r11](%[svm]) \n\t"
-		"mov %%r12, %c[r12](%[svm]) \n\t"
-		"mov %%r13, %c[r13](%[svm]) \n\t"
-		"mov %%r14, %c[r14](%[svm]) \n\t"
-		"mov %%r15, %c[r15](%[svm]) \n\t"
-
-		"pop  %%rbp; \n\t"
-#else
-		"mov %%ebx, %c[rbx](%[svm]) \n\t"
-		"mov %%ecx, %c[rcx](%[svm]) \n\t"
-		"mov %%edx, %c[rdx](%[svm]) \n\t"
-		"mov %%esi, %c[rsi](%[svm]) \n\t"
-		"mov %%edi, %c[rdi](%[svm]) \n\t"
-		"mov %%ebp, %c[rbp](%[svm]) \n\t"
-
-		"pop  %%ebp; \n\t"
-#endif
-		:
-		: [svm]"a"(svm),
-		  [vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)),
-		  [rbx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBX])),
-		  [rcx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RCX])),
-		  [rdx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDX])),
-		  [rsi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RSI])),
-		  [rdi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDI])),
-		  [rbp]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBP]))
-#ifdef CONFIG_X86_64
-		  , [r8]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R8])),
-		  [r9]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R9])),
-		  [r10]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R10])),
-		  [r11]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R11])),
-		  [r12]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R12])),
-		  [r13]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R13])),
-		  [r14]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R14])),
-		  [r15]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R15]))
-#endif
-		: "cc", "memory"
-#ifdef CONFIG_X86_64
-		, "rbx", "rcx", "rdx", "rsi", "rdi"
-		, "r8", "r9", "r10", "r11" , "r12", "r13", "r14", "r15"
-#else
-		, "ebx", "ecx", "edx" , "esi", "edi"
-#endif
-		);
-
-	if ((svm->vmcb->save.dr7 & 0xff))
-		load_db_regs(svm->host_db_regs);
-
-	vcpu->arch.cr2 = svm->vmcb->save.cr2;
-
-	write_dr6(svm->host_dr6);
-	write_dr7(svm->host_dr7);
-	kvm_write_cr2(svm->host_cr2);
-
-	load_fs(fs_selector);
-	load_gs(gs_selector);
-	load_ldt(ldt_selector);
-	load_host_msrs(vcpu);
-
-	reload_tss(vcpu);
-
-	local_irq_disable();
-
-	stgi();
-
-	svm->next_rip = 0;
-}
-
-static void svm_set_cr3(struct kvm_vcpu *vcpu, unsigned long root)
-{
-	struct vcpu_svm *svm = to_svm(vcpu);
-
-	svm->vmcb->save.cr3 = root;
-	force_new_asid(vcpu);
-
-	if (vcpu->fpu_active) {
-		svm->vmcb->control.intercept_exceptions |= (1 << NM_VECTOR);
-		svm->vmcb->save.cr0 |= X86_CR0_TS;
-		vcpu->fpu_active = 0;
-	}
-}
-
-static int is_disabled(void)
-{
-	u64 vm_cr;
-
-	rdmsrl(MSR_VM_CR, vm_cr);
-	if (vm_cr & (1 << SVM_VM_CR_SVM_DISABLE))
-		return 1;
-
-	return 0;
-}
-
-static void
-svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
-{
-	/*
-	 * Patch in the VMMCALL instruction:
-	 */
-	hypercall[0] = 0x0f;
-	hypercall[1] = 0x01;
-	hypercall[2] = 0xd9;
-}
-
-static void svm_check_processor_compat(void *rtn)
-{
-	*(int *)rtn = 0;
-}
-
-static struct kvm_x86_ops svm_x86_ops = {
-	.cpu_has_kvm_support = has_svm,
-	.disabled_by_bios = is_disabled,
-	.hardware_setup = svm_hardware_setup,
-	.hardware_unsetup = svm_hardware_unsetup,
-	.check_processor_compatibility = svm_check_processor_compat,
-	.hardware_enable = svm_hardware_enable,
-	.hardware_disable = svm_hardware_disable,
-
-	.vcpu_create = svm_create_vcpu,
-	.vcpu_free = svm_free_vcpu,
-	.vcpu_reset = svm_vcpu_reset,
-
-	.prepare_guest_switch = svm_prepare_guest_switch,
-	.vcpu_load = svm_vcpu_load,
-	.vcpu_put = svm_vcpu_put,
-	.vcpu_decache = svm_vcpu_decache,
-
-	.set_guest_debug = svm_guest_debug,
-	.get_msr = svm_get_msr,
-	.set_msr = svm_set_msr,
-	.get_segment_base = svm_get_segment_base,
-	.get_segment = svm_get_segment,
-	.set_segment = svm_set_segment,
-	.get_cs_db_l_bits = kvm_get_cs_db_l_bits,
-	.decache_cr4_guest_bits = svm_decache_cr4_guest_bits,
-	.set_cr0 = svm_set_cr0,
-	.set_cr3 = svm_set_cr3,
-	.set_cr4 = svm_set_cr4,
-	.set_efer = svm_set_efer,
-	.get_idt = svm_get_idt,
-	.set_idt = svm_set_idt,
-	.get_gdt = svm_get_gdt,
-	.set_gdt = svm_set_gdt,
-	.get_dr = svm_get_dr,
-	.set_dr = svm_set_dr,
-	.cache_regs = svm_cache_regs,
-	.decache_regs = svm_decache_regs,
-	.get_rflags = svm_get_rflags,
-	.set_rflags = svm_set_rflags,
-
-	.tlb_flush = svm_flush_tlb,
-
-	.run = svm_vcpu_run,
-	.handle_exit = handle_exit,
-	.skip_emulated_instruction = skip_emulated_instruction,
-	.patch_hypercall = svm_patch_hypercall,
-	.get_irq = svm_get_irq,
-	.set_irq = svm_set_irq,
-	.queue_exception = svm_queue_exception,
-	.exception_injected = svm_exception_injected,
-	.inject_pending_irq = svm_intr_assist,
-	.inject_pending_vectors = do_interrupt_requests,
-
-	.set_tss_addr = svm_set_tss_addr,
-};
-
-static int __init svm_init(void)
-{
-	return kvm_init(&svm_x86_ops, sizeof(struct vcpu_svm),
-			      THIS_MODULE);
-}
-
-static void __exit svm_exit(void)
-{
-	kvm_exit();
-}
-
-module_init(svm_init)
-module_exit(svm_exit)
diff --git a/drivers/kvm/svm.h b/drivers/kvm/svm.h
deleted file mode 100644
index 5fd50491b555..000000000000
--- a/drivers/kvm/svm.h
+++ /dev/null
@@ -1,325 +0,0 @@
-#ifndef __SVM_H
-#define __SVM_H
-
-enum {
-	INTERCEPT_INTR,
-	INTERCEPT_NMI,
-	INTERCEPT_SMI,
-	INTERCEPT_INIT,
-	INTERCEPT_VINTR,
-	INTERCEPT_SELECTIVE_CR0,
-	INTERCEPT_STORE_IDTR,
-	INTERCEPT_STORE_GDTR,
-	INTERCEPT_STORE_LDTR,
-	INTERCEPT_STORE_TR,
-	INTERCEPT_LOAD_IDTR,
-	INTERCEPT_LOAD_GDTR,
-	INTERCEPT_LOAD_LDTR,
-	INTERCEPT_LOAD_TR,
-	INTERCEPT_RDTSC,
-	INTERCEPT_RDPMC,
-	INTERCEPT_PUSHF,
-	INTERCEPT_POPF,
-	INTERCEPT_CPUID,
-	INTERCEPT_RSM,
-	INTERCEPT_IRET,
-	INTERCEPT_INTn,
-	INTERCEPT_INVD,
-	INTERCEPT_PAUSE,
-	INTERCEPT_HLT,
-	INTERCEPT_INVLPG,
-	INTERCEPT_INVLPGA,
-	INTERCEPT_IOIO_PROT,
-	INTERCEPT_MSR_PROT,
-	INTERCEPT_TASK_SWITCH,
-	INTERCEPT_FERR_FREEZE,
-	INTERCEPT_SHUTDOWN,
-	INTERCEPT_VMRUN,
-	INTERCEPT_VMMCALL,
-	INTERCEPT_VMLOAD,
-	INTERCEPT_VMSAVE,
-	INTERCEPT_STGI,
-	INTERCEPT_CLGI,
-	INTERCEPT_SKINIT,
-	INTERCEPT_RDTSCP,
-	INTERCEPT_ICEBP,
-	INTERCEPT_WBINVD,
-	INTERCEPT_MONITOR,
-	INTERCEPT_MWAIT,
-	INTERCEPT_MWAIT_COND,
-};
-
-
-struct __attribute__ ((__packed__)) vmcb_control_area {
-	u16 intercept_cr_read;
-	u16 intercept_cr_write;
-	u16 intercept_dr_read;
-	u16 intercept_dr_write;
-	u32 intercept_exceptions;
-	u64 intercept;
-	u8 reserved_1[44];
-	u64 iopm_base_pa;
-	u64 msrpm_base_pa;
-	u64 tsc_offset;
-	u32 asid;
-	u8 tlb_ctl;
-	u8 reserved_2[3];
-	u32 int_ctl;
-	u32 int_vector;
-	u32 int_state;
-	u8 reserved_3[4];
-	u32 exit_code;
-	u32 exit_code_hi;
-	u64 exit_info_1;
-	u64 exit_info_2;
-	u32 exit_int_info;
-	u32 exit_int_info_err;
-	u64 nested_ctl;
-	u8 reserved_4[16];
-	u32 event_inj;
-	u32 event_inj_err;
-	u64 nested_cr3;
-	u64 lbr_ctl;
-	u8 reserved_5[832];
-};
-
-
-#define TLB_CONTROL_DO_NOTHING 0
-#define TLB_CONTROL_FLUSH_ALL_ASID 1
-
-#define V_TPR_MASK 0x0f
-
-#define V_IRQ_SHIFT 8
-#define V_IRQ_MASK (1 << V_IRQ_SHIFT)
-
-#define V_INTR_PRIO_SHIFT 16
-#define V_INTR_PRIO_MASK (0x0f << V_INTR_PRIO_SHIFT)
-
-#define V_IGN_TPR_SHIFT 20
-#define V_IGN_TPR_MASK (1 << V_IGN_TPR_SHIFT)
-
-#define V_INTR_MASKING_SHIFT 24
-#define V_INTR_MASKING_MASK (1 << V_INTR_MASKING_SHIFT)
-
-#define SVM_INTERRUPT_SHADOW_MASK 1
-
-#define SVM_IOIO_STR_SHIFT 2
-#define SVM_IOIO_REP_SHIFT 3
-#define SVM_IOIO_SIZE_SHIFT 4
-#define SVM_IOIO_ASIZE_SHIFT 7
-
-#define SVM_IOIO_TYPE_MASK 1
-#define SVM_IOIO_STR_MASK (1 << SVM_IOIO_STR_SHIFT)
-#define SVM_IOIO_REP_MASK (1 << SVM_IOIO_REP_SHIFT)
-#define SVM_IOIO_SIZE_MASK (7 << SVM_IOIO_SIZE_SHIFT)
-#define SVM_IOIO_ASIZE_MASK (7 << SVM_IOIO_ASIZE_SHIFT)
-
-struct __attribute__ ((__packed__)) vmcb_seg {
-	u16 selector;
-	u16 attrib;
-	u32 limit;
-	u64 base;
-};
-
-struct __attribute__ ((__packed__)) vmcb_save_area {
-	struct vmcb_seg es;
-	struct vmcb_seg cs;
-	struct vmcb_seg ss;
-	struct vmcb_seg ds;
-	struct vmcb_seg fs;
-	struct vmcb_seg gs;
-	struct vmcb_seg gdtr;
-	struct vmcb_seg ldtr;
-	struct vmcb_seg idtr;
-	struct vmcb_seg tr;
-	u8 reserved_1[43];
-	u8 cpl;
-	u8 reserved_2[4];
-	u64 efer;
-	u8 reserved_3[112];
-	u64 cr4;
-	u64 cr3;
-	u64 cr0;
-	u64 dr7;
-	u64 dr6;
-	u64 rflags;
-	u64 rip;
-	u8 reserved_4[88];
-	u64 rsp;
-	u8 reserved_5[24];
-	u64 rax;
-	u64 star;
-	u64 lstar;
-	u64 cstar;
-	u64 sfmask;
-	u64 kernel_gs_base;
-	u64 sysenter_cs;
-	u64 sysenter_esp;
-	u64 sysenter_eip;
-	u64 cr2;
-	u8 reserved_6[32];
-	u64 g_pat;
-	u64 dbgctl;
-	u64 br_from;
-	u64 br_to;
-	u64 last_excp_from;
-	u64 last_excp_to;
-};
-
-struct __attribute__ ((__packed__)) vmcb {
-	struct vmcb_control_area control;
-	struct vmcb_save_area save;
-};
-
-#define SVM_CPUID_FEATURE_SHIFT 2
-#define SVM_CPUID_FUNC 0x8000000a
-
-#define MSR_EFER_SVME_MASK (1ULL << 12)
-#define MSR_VM_CR       0xc0010114
-#define MSR_VM_HSAVE_PA 0xc0010117ULL
-
-#define SVM_VM_CR_SVM_DISABLE 4
-
-#define SVM_SELECTOR_S_SHIFT 4
-#define SVM_SELECTOR_DPL_SHIFT 5
-#define SVM_SELECTOR_P_SHIFT 7
-#define SVM_SELECTOR_AVL_SHIFT 8
-#define SVM_SELECTOR_L_SHIFT 9
-#define SVM_SELECTOR_DB_SHIFT 10
-#define SVM_SELECTOR_G_SHIFT 11
-
-#define SVM_SELECTOR_TYPE_MASK (0xf)
-#define SVM_SELECTOR_S_MASK (1 << SVM_SELECTOR_S_SHIFT)
-#define SVM_SELECTOR_DPL_MASK (3 << SVM_SELECTOR_DPL_SHIFT)
-#define SVM_SELECTOR_P_MASK (1 << SVM_SELECTOR_P_SHIFT)
-#define SVM_SELECTOR_AVL_MASK (1 << SVM_SELECTOR_AVL_SHIFT)
-#define SVM_SELECTOR_L_MASK (1 << SVM_SELECTOR_L_SHIFT)
-#define SVM_SELECTOR_DB_MASK (1 << SVM_SELECTOR_DB_SHIFT)
-#define SVM_SELECTOR_G_MASK (1 << SVM_SELECTOR_G_SHIFT)
-
-#define SVM_SELECTOR_WRITE_MASK (1 << 1)
-#define SVM_SELECTOR_READ_MASK SVM_SELECTOR_WRITE_MASK
-#define SVM_SELECTOR_CODE_MASK (1 << 3)
-
-#define INTERCEPT_CR0_MASK 1
-#define INTERCEPT_CR3_MASK (1 << 3)
-#define INTERCEPT_CR4_MASK (1 << 4)
-#define INTERCEPT_CR8_MASK (1 << 8)
-
-#define INTERCEPT_DR0_MASK 1
-#define INTERCEPT_DR1_MASK (1 << 1)
-#define INTERCEPT_DR2_MASK (1 << 2)
-#define INTERCEPT_DR3_MASK (1 << 3)
-#define INTERCEPT_DR4_MASK (1 << 4)
-#define INTERCEPT_DR5_MASK (1 << 5)
-#define INTERCEPT_DR6_MASK (1 << 6)
-#define INTERCEPT_DR7_MASK (1 << 7)
-
-#define SVM_EVTINJ_VEC_MASK 0xff
-
-#define SVM_EVTINJ_TYPE_SHIFT 8
-#define SVM_EVTINJ_TYPE_MASK (7 << SVM_EVTINJ_TYPE_SHIFT)
-
-#define SVM_EVTINJ_TYPE_INTR (0 << SVM_EVTINJ_TYPE_SHIFT)
-#define SVM_EVTINJ_TYPE_NMI (2 << SVM_EVTINJ_TYPE_SHIFT)
-#define SVM_EVTINJ_TYPE_EXEPT (3 << SVM_EVTINJ_TYPE_SHIFT)
-#define SVM_EVTINJ_TYPE_SOFT (4 << SVM_EVTINJ_TYPE_SHIFT)
-
-#define SVM_EVTINJ_VALID (1 << 31)
-#define SVM_EVTINJ_VALID_ERR (1 << 11)
-
-#define SVM_EXITINTINFO_VEC_MASK SVM_EVTINJ_VEC_MASK
-
-#define	SVM_EXITINTINFO_TYPE_INTR SVM_EVTINJ_TYPE_INTR
-#define	SVM_EXITINTINFO_TYPE_NMI SVM_EVTINJ_TYPE_NMI
-#define	SVM_EXITINTINFO_TYPE_EXEPT SVM_EVTINJ_TYPE_EXEPT
-#define	SVM_EXITINTINFO_TYPE_SOFT SVM_EVTINJ_TYPE_SOFT
-
-#define SVM_EXITINTINFO_VALID SVM_EVTINJ_VALID
-#define SVM_EXITINTINFO_VALID_ERR SVM_EVTINJ_VALID_ERR
-
-#define	SVM_EXIT_READ_CR0 	0x000
-#define	SVM_EXIT_READ_CR3 	0x003
-#define	SVM_EXIT_READ_CR4 	0x004
-#define	SVM_EXIT_READ_CR8 	0x008
-#define	SVM_EXIT_WRITE_CR0 	0x010
-#define	SVM_EXIT_WRITE_CR3 	0x013
-#define	SVM_EXIT_WRITE_CR4 	0x014
-#define	SVM_EXIT_WRITE_CR8 	0x018
-#define	SVM_EXIT_READ_DR0 	0x020
-#define	SVM_EXIT_READ_DR1 	0x021
-#define	SVM_EXIT_READ_DR2 	0x022
-#define	SVM_EXIT_READ_DR3 	0x023
-#define	SVM_EXIT_READ_DR4 	0x024
-#define	SVM_EXIT_READ_DR5 	0x025
-#define	SVM_EXIT_READ_DR6 	0x026
-#define	SVM_EXIT_READ_DR7 	0x027
-#define	SVM_EXIT_WRITE_DR0 	0x030
-#define	SVM_EXIT_WRITE_DR1 	0x031
-#define	SVM_EXIT_WRITE_DR2 	0x032
-#define	SVM_EXIT_WRITE_DR3 	0x033
-#define	SVM_EXIT_WRITE_DR4 	0x034
-#define	SVM_EXIT_WRITE_DR5 	0x035
-#define	SVM_EXIT_WRITE_DR6 	0x036
-#define	SVM_EXIT_WRITE_DR7 	0x037
-#define SVM_EXIT_EXCP_BASE      0x040
-#define SVM_EXIT_INTR		0x060
-#define SVM_EXIT_NMI		0x061
-#define SVM_EXIT_SMI		0x062
-#define SVM_EXIT_INIT		0x063
-#define SVM_EXIT_VINTR		0x064
-#define SVM_EXIT_CR0_SEL_WRITE	0x065
-#define SVM_EXIT_IDTR_READ	0x066
-#define SVM_EXIT_GDTR_READ	0x067
-#define SVM_EXIT_LDTR_READ	0x068
-#define SVM_EXIT_TR_READ	0x069
-#define SVM_EXIT_IDTR_WRITE	0x06a
-#define SVM_EXIT_GDTR_WRITE	0x06b
-#define SVM_EXIT_LDTR_WRITE	0x06c
-#define SVM_EXIT_TR_WRITE	0x06d
-#define SVM_EXIT_RDTSC		0x06e
-#define SVM_EXIT_RDPMC		0x06f
-#define SVM_EXIT_PUSHF		0x070
-#define SVM_EXIT_POPF		0x071
-#define SVM_EXIT_CPUID		0x072
-#define SVM_EXIT_RSM		0x073
-#define SVM_EXIT_IRET		0x074
-#define SVM_EXIT_SWINT		0x075
-#define SVM_EXIT_INVD		0x076
-#define SVM_EXIT_PAUSE		0x077
-#define SVM_EXIT_HLT		0x078
-#define SVM_EXIT_INVLPG		0x079
-#define SVM_EXIT_INVLPGA	0x07a
-#define SVM_EXIT_IOIO		0x07b
-#define SVM_EXIT_MSR		0x07c
-#define SVM_EXIT_TASK_SWITCH	0x07d
-#define SVM_EXIT_FERR_FREEZE	0x07e
-#define SVM_EXIT_SHUTDOWN	0x07f
-#define SVM_EXIT_VMRUN		0x080
-#define SVM_EXIT_VMMCALL	0x081
-#define SVM_EXIT_VMLOAD		0x082
-#define SVM_EXIT_VMSAVE		0x083
-#define SVM_EXIT_STGI		0x084
-#define SVM_EXIT_CLGI		0x085
-#define SVM_EXIT_SKINIT		0x086
-#define SVM_EXIT_RDTSCP		0x087
-#define SVM_EXIT_ICEBP		0x088
-#define SVM_EXIT_WBINVD		0x089
-#define SVM_EXIT_MONITOR	0x08a
-#define SVM_EXIT_MWAIT		0x08b
-#define SVM_EXIT_MWAIT_COND	0x08c
-#define SVM_EXIT_NPF  		0x400
-
-#define SVM_EXIT_ERR		-1
-
-#define SVM_CR0_SELECTIVE_MASK (1 << 3 | 1) /* TS and MP */
-
-#define SVM_VMLOAD ".byte 0x0f, 0x01, 0xda"
-#define SVM_VMRUN  ".byte 0x0f, 0x01, 0xd8"
-#define SVM_VMSAVE ".byte 0x0f, 0x01, 0xdb"
-#define SVM_CLGI   ".byte 0x0f, 0x01, 0xdd"
-#define SVM_STGI   ".byte 0x0f, 0x01, 0xdc"
-#define SVM_INVLPGA ".byte 0x0f, 0x01, 0xdf"
-
-#endif
-
diff --git a/drivers/kvm/types.h b/drivers/kvm/types.h
deleted file mode 100644
index 1c4e46decb22..000000000000
--- a/drivers/kvm/types.h
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- *
- */
-
-#ifndef __KVM_TYPES_H__
-#define __KVM_TYPES_H__
-
-#include <asm/types.h>
-
-/*
- * Address types:
- *
- *  gva - guest virtual address
- *  gpa - guest physical address
- *  gfn - guest frame number
- *  hva - host virtual address
- *  hpa - host physical address
- *  hfn - host frame number
- */
-
-typedef unsigned long  gva_t;
-typedef u64            gpa_t;
-typedef unsigned long  gfn_t;
-
-typedef unsigned long  hva_t;
-typedef u64            hpa_t;
-typedef unsigned long  hfn_t;
-
-struct kvm_pio_request {
-	unsigned long count;
-	int cur_count;
-	struct page *guest_pages[2];
-	unsigned guest_page_offset;
-	int in;
-	int port;
-	int size;
-	int string;
-	int down;
-	int rep;
-};
-
-#endif /* __KVM_TYPES_H__ */
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
deleted file mode 100644
index 11ca2340d38f..000000000000
--- a/drivers/kvm/vmx.c
+++ /dev/null
@@ -1,2673 +0,0 @@
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * This module enables machines with Intel VT-x extensions to run virtual
- * machines without emulation or binary translation.
- *
- * Copyright (C) 2006 Qumranet, Inc.
- *
- * Authors:
- *   Avi Kivity   <avi@qumranet.com>
- *   Yaniv Kamay  <yaniv@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-
-#include "kvm.h"
-#include "x86.h"
-#include "x86_emulate.h"
-#include "irq.h"
-#include "vmx.h"
-#include "segment_descriptor.h"
-#include "mmu.h"
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/sched.h>
-#include <linux/moduleparam.h>
-
-#include <asm/io.h>
-#include <asm/desc.h>
-
-MODULE_AUTHOR("Qumranet");
-MODULE_LICENSE("GPL");
-
-static int bypass_guest_pf = 1;
-module_param(bypass_guest_pf, bool, 0);
-
-struct vmcs {
-	u32 revision_id;
-	u32 abort;
-	char data[0];
-};
-
-struct vcpu_vmx {
-	struct kvm_vcpu       vcpu;
-	int                   launched;
-	u8                    fail;
-	u32                   idt_vectoring_info;
-	struct kvm_msr_entry *guest_msrs;
-	struct kvm_msr_entry *host_msrs;
-	int                   nmsrs;
-	int                   save_nmsrs;
-	int                   msr_offset_efer;
-#ifdef CONFIG_X86_64
-	int                   msr_offset_kernel_gs_base;
-#endif
-	struct vmcs          *vmcs;
-	struct {
-		int           loaded;
-		u16           fs_sel, gs_sel, ldt_sel;
-		int           gs_ldt_reload_needed;
-		int           fs_reload_needed;
-		int           guest_efer_loaded;
-	} host_state;
-	struct {
-		struct {
-			bool pending;
-			u8 vector;
-			unsigned rip;
-		} irq;
-	} rmode;
-};
-
-static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
-{
-	return container_of(vcpu, struct vcpu_vmx, vcpu);
-}
-
-static int init_rmode_tss(struct kvm *kvm);
-
-static DEFINE_PER_CPU(struct vmcs *, vmxarea);
-static DEFINE_PER_CPU(struct vmcs *, current_vmcs);
-
-static struct page *vmx_io_bitmap_a;
-static struct page *vmx_io_bitmap_b;
-
-static struct vmcs_config {
-	int size;
-	int order;
-	u32 revision_id;
-	u32 pin_based_exec_ctrl;
-	u32 cpu_based_exec_ctrl;
-	u32 cpu_based_2nd_exec_ctrl;
-	u32 vmexit_ctrl;
-	u32 vmentry_ctrl;
-} vmcs_config;
-
-#define VMX_SEGMENT_FIELD(seg)					\
-	[VCPU_SREG_##seg] = {                                   \
-		.selector = GUEST_##seg##_SELECTOR,		\
-		.base = GUEST_##seg##_BASE,		   	\
-		.limit = GUEST_##seg##_LIMIT,		   	\
-		.ar_bytes = GUEST_##seg##_AR_BYTES,	   	\
-	}
-
-static struct kvm_vmx_segment_field {
-	unsigned selector;
-	unsigned base;
-	unsigned limit;
-	unsigned ar_bytes;
-} kvm_vmx_segment_fields[] = {
-	VMX_SEGMENT_FIELD(CS),
-	VMX_SEGMENT_FIELD(DS),
-	VMX_SEGMENT_FIELD(ES),
-	VMX_SEGMENT_FIELD(FS),
-	VMX_SEGMENT_FIELD(GS),
-	VMX_SEGMENT_FIELD(SS),
-	VMX_SEGMENT_FIELD(TR),
-	VMX_SEGMENT_FIELD(LDTR),
-};
-
-/*
- * Keep MSR_K6_STAR at the end, as setup_msrs() will try to optimize it
- * away by decrementing the array size.
- */
-static const u32 vmx_msr_index[] = {
-#ifdef CONFIG_X86_64
-	MSR_SYSCALL_MASK, MSR_LSTAR, MSR_CSTAR, MSR_KERNEL_GS_BASE,
-#endif
-	MSR_EFER, MSR_K6_STAR,
-};
-#define NR_VMX_MSR ARRAY_SIZE(vmx_msr_index)
-
-static void load_msrs(struct kvm_msr_entry *e, int n)
-{
-	int i;
-
-	for (i = 0; i < n; ++i)
-		wrmsrl(e[i].index, e[i].data);
-}
-
-static void save_msrs(struct kvm_msr_entry *e, int n)
-{
-	int i;
-
-	for (i = 0; i < n; ++i)
-		rdmsrl(e[i].index, e[i].data);
-}
-
-static inline int is_page_fault(u32 intr_info)
-{
-	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
-			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK);
-}
-
-static inline int is_no_device(u32 intr_info)
-{
-	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
-			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK);
-}
-
-static inline int is_invalid_opcode(u32 intr_info)
-{
-	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
-			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK);
-}
-
-static inline int is_external_interrupt(u32 intr_info)
-{
-	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VALID_MASK))
-		== (INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK);
-}
-
-static inline int cpu_has_vmx_tpr_shadow(void)
-{
-	return (vmcs_config.cpu_based_exec_ctrl & CPU_BASED_TPR_SHADOW);
-}
-
-static inline int vm_need_tpr_shadow(struct kvm *kvm)
-{
-	return ((cpu_has_vmx_tpr_shadow()) && (irqchip_in_kernel(kvm)));
-}
-
-static inline int cpu_has_secondary_exec_ctrls(void)
-{
-	return (vmcs_config.cpu_based_exec_ctrl &
-		CPU_BASED_ACTIVATE_SECONDARY_CONTROLS);
-}
-
-static inline int cpu_has_vmx_virtualize_apic_accesses(void)
-{
-	return (vmcs_config.cpu_based_2nd_exec_ctrl &
-		SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
-}
-
-static inline int vm_need_virtualize_apic_accesses(struct kvm *kvm)
-{
-	return ((cpu_has_vmx_virtualize_apic_accesses()) &&
-		(irqchip_in_kernel(kvm)));
-}
-
-static int __find_msr_index(struct vcpu_vmx *vmx, u32 msr)
-{
-	int i;
-
-	for (i = 0; i < vmx->nmsrs; ++i)
-		if (vmx->guest_msrs[i].index == msr)
-			return i;
-	return -1;
-}
-
-static struct kvm_msr_entry *find_msr_entry(struct vcpu_vmx *vmx, u32 msr)
-{
-	int i;
-
-	i = __find_msr_index(vmx, msr);
-	if (i >= 0)
-		return &vmx->guest_msrs[i];
-	return NULL;
-}
-
-static void vmcs_clear(struct vmcs *vmcs)
-{
-	u64 phys_addr = __pa(vmcs);
-	u8 error;
-
-	asm volatile (ASM_VMX_VMCLEAR_RAX "; setna %0"
-		      : "=g"(error) : "a"(&phys_addr), "m"(phys_addr)
-		      : "cc", "memory");
-	if (error)
-		printk(KERN_ERR "kvm: vmclear fail: %p/%llx\n",
-		       vmcs, phys_addr);
-}
-
-static void __vcpu_clear(void *arg)
-{
-	struct vcpu_vmx *vmx = arg;
-	int cpu = raw_smp_processor_id();
-
-	if (vmx->vcpu.cpu == cpu)
-		vmcs_clear(vmx->vmcs);
-	if (per_cpu(current_vmcs, cpu) == vmx->vmcs)
-		per_cpu(current_vmcs, cpu) = NULL;
-	rdtscll(vmx->vcpu.arch.host_tsc);
-}
-
-static void vcpu_clear(struct vcpu_vmx *vmx)
-{
-	if (vmx->vcpu.cpu == -1)
-		return;
-	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 0, 1);
-	vmx->launched = 0;
-}
-
-static unsigned long vmcs_readl(unsigned long field)
-{
-	unsigned long value;
-
-	asm volatile (ASM_VMX_VMREAD_RDX_RAX
-		      : "=a"(value) : "d"(field) : "cc");
-	return value;
-}
-
-static u16 vmcs_read16(unsigned long field)
-{
-	return vmcs_readl(field);
-}
-
-static u32 vmcs_read32(unsigned long field)
-{
-	return vmcs_readl(field);
-}
-
-static u64 vmcs_read64(unsigned long field)
-{
-#ifdef CONFIG_X86_64
-	return vmcs_readl(field);
-#else
-	return vmcs_readl(field) | ((u64)vmcs_readl(field+1) << 32);
-#endif
-}
-
-static noinline void vmwrite_error(unsigned long field, unsigned long value)
-{
-	printk(KERN_ERR "vmwrite error: reg %lx value %lx (err %d)\n",
-	       field, value, vmcs_read32(VM_INSTRUCTION_ERROR));
-	dump_stack();
-}
-
-static void vmcs_writel(unsigned long field, unsigned long value)
-{
-	u8 error;
-
-	asm volatile (ASM_VMX_VMWRITE_RAX_RDX "; setna %0"
-		       : "=q"(error) : "a"(value), "d"(field) : "cc");
-	if (unlikely(error))
-		vmwrite_error(field, value);
-}
-
-static void vmcs_write16(unsigned long field, u16 value)
-{
-	vmcs_writel(field, value);
-}
-
-static void vmcs_write32(unsigned long field, u32 value)
-{
-	vmcs_writel(field, value);
-}
-
-static void vmcs_write64(unsigned long field, u64 value)
-{
-#ifdef CONFIG_X86_64
-	vmcs_writel(field, value);
-#else
-	vmcs_writel(field, value);
-	asm volatile ("");
-	vmcs_writel(field+1, value >> 32);
-#endif
-}
-
-static void vmcs_clear_bits(unsigned long field, u32 mask)
-{
-	vmcs_writel(field, vmcs_readl(field) & ~mask);
-}
-
-static void vmcs_set_bits(unsigned long field, u32 mask)
-{
-	vmcs_writel(field, vmcs_readl(field) | mask);
-}
-
-static void update_exception_bitmap(struct kvm_vcpu *vcpu)
-{
-	u32 eb;
-
-	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
-	if (!vcpu->fpu_active)
-		eb |= 1u << NM_VECTOR;
-	if (vcpu->guest_debug.enabled)
-		eb |= 1u << 1;
-	if (vcpu->arch.rmode.active)
-		eb = ~0;
-	vmcs_write32(EXCEPTION_BITMAP, eb);
-}
-
-static void reload_tss(void)
-{
-#ifndef CONFIG_X86_64
-
-	/*
-	 * VT restores TR but not its size.  Useless.
-	 */
-	struct descriptor_table gdt;
-	struct segment_descriptor *descs;
-
-	get_gdt(&gdt);
-	descs = (void *)gdt.base;
-	descs[GDT_ENTRY_TSS].type = 9; /* available TSS */
-	load_TR_desc();
-#endif
-}
-
-static void load_transition_efer(struct vcpu_vmx *vmx)
-{
-	int efer_offset = vmx->msr_offset_efer;
-	u64 host_efer = vmx->host_msrs[efer_offset].data;
-	u64 guest_efer = vmx->guest_msrs[efer_offset].data;
-	u64 ignore_bits;
-
-	if (efer_offset < 0)
-		return;
-	/*
-	 * NX is emulated; LMA and LME handled by hardware; SCE meaninless
-	 * outside long mode
-	 */
-	ignore_bits = EFER_NX | EFER_SCE;
-#ifdef CONFIG_X86_64
-	ignore_bits |= EFER_LMA | EFER_LME;
-	/* SCE is meaningful only in long mode on Intel */
-	if (guest_efer & EFER_LMA)
-		ignore_bits &= ~(u64)EFER_SCE;
-#endif
-	if ((guest_efer & ~ignore_bits) == (host_efer & ~ignore_bits))
-		return;
-
-	vmx->host_state.guest_efer_loaded = 1;
-	guest_efer &= ~ignore_bits;
-	guest_efer |= host_efer & ignore_bits;
-	wrmsrl(MSR_EFER, guest_efer);
-	vmx->vcpu.stat.efer_reload++;
-}
-
-static void reload_host_efer(struct vcpu_vmx *vmx)
-{
-	if (vmx->host_state.guest_efer_loaded) {
-		vmx->host_state.guest_efer_loaded = 0;
-		load_msrs(vmx->host_msrs + vmx->msr_offset_efer, 1);
-	}
-}
-
-static void vmx_save_host_state(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-
-	if (vmx->host_state.loaded)
-		return;
-
-	vmx->host_state.loaded = 1;
-	/*
-	 * Set host fs and gs selectors.  Unfortunately, 22.2.3 does not
-	 * allow segment selectors with cpl > 0 or ti == 1.
-	 */
-	vmx->host_state.ldt_sel = read_ldt();
-	vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel;
-	vmx->host_state.fs_sel = read_fs();
-	if (!(vmx->host_state.fs_sel & 7)) {
-		vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel);
-		vmx->host_state.fs_reload_needed = 0;
-	} else {
-		vmcs_write16(HOST_FS_SELECTOR, 0);
-		vmx->host_state.fs_reload_needed = 1;
-	}
-	vmx->host_state.gs_sel = read_gs();
-	if (!(vmx->host_state.gs_sel & 7))
-		vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel);
-	else {
-		vmcs_write16(HOST_GS_SELECTOR, 0);
-		vmx->host_state.gs_ldt_reload_needed = 1;
-	}
-
-#ifdef CONFIG_X86_64
-	vmcs_writel(HOST_FS_BASE, read_msr(MSR_FS_BASE));
-	vmcs_writel(HOST_GS_BASE, read_msr(MSR_GS_BASE));
-#else
-	vmcs_writel(HOST_FS_BASE, segment_base(vmx->host_state.fs_sel));
-	vmcs_writel(HOST_GS_BASE, segment_base(vmx->host_state.gs_sel));
-#endif
-
-#ifdef CONFIG_X86_64
-	if (is_long_mode(&vmx->vcpu))
-		save_msrs(vmx->host_msrs +
-			  vmx->msr_offset_kernel_gs_base, 1);
-
-#endif
-	load_msrs(vmx->guest_msrs, vmx->save_nmsrs);
-	load_transition_efer(vmx);
-}
-
-static void vmx_load_host_state(struct vcpu_vmx *vmx)
-{
-	unsigned long flags;
-
-	if (!vmx->host_state.loaded)
-		return;
-
-	++vmx->vcpu.stat.host_state_reload;
-	vmx->host_state.loaded = 0;
-	if (vmx->host_state.fs_reload_needed)
-		load_fs(vmx->host_state.fs_sel);
-	if (vmx->host_state.gs_ldt_reload_needed) {
-		load_ldt(vmx->host_state.ldt_sel);
-		/*
-		 * If we have to reload gs, we must take care to
-		 * preserve our gs base.
-		 */
-		local_irq_save(flags);
-		load_gs(vmx->host_state.gs_sel);
-#ifdef CONFIG_X86_64
-		wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE));
-#endif
-		local_irq_restore(flags);
-	}
-	reload_tss();
-	save_msrs(vmx->guest_msrs, vmx->save_nmsrs);
-	load_msrs(vmx->host_msrs, vmx->save_nmsrs);
-	reload_host_efer(vmx);
-}
-
-/*
- * Switches to specified vcpu, until a matching vcpu_put(), but assumes
- * vcpu mutex is already taken.
- */
-static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u64 phys_addr = __pa(vmx->vmcs);
-	u64 tsc_this, delta;
-
-	if (vcpu->cpu != cpu) {
-		vcpu_clear(vmx);
-		kvm_migrate_apic_timer(vcpu);
-	}
-
-	if (per_cpu(current_vmcs, cpu) != vmx->vmcs) {
-		u8 error;
-
-		per_cpu(current_vmcs, cpu) = vmx->vmcs;
-		asm volatile (ASM_VMX_VMPTRLD_RAX "; setna %0"
-			      : "=g"(error) : "a"(&phys_addr), "m"(phys_addr)
-			      : "cc");
-		if (error)
-			printk(KERN_ERR "kvm: vmptrld %p/%llx fail\n",
-			       vmx->vmcs, phys_addr);
-	}
-
-	if (vcpu->cpu != cpu) {
-		struct descriptor_table dt;
-		unsigned long sysenter_esp;
-
-		vcpu->cpu = cpu;
-		/*
-		 * Linux uses per-cpu TSS and GDT, so set these when switching
-		 * processors.
-		 */
-		vmcs_writel(HOST_TR_BASE, read_tr_base()); /* 22.2.4 */
-		get_gdt(&dt);
-		vmcs_writel(HOST_GDTR_BASE, dt.base);   /* 22.2.4 */
-
-		rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp);
-		vmcs_writel(HOST_IA32_SYSENTER_ESP, sysenter_esp); /* 22.2.3 */
-
-		/*
-		 * Make sure the time stamp counter is monotonous.
-		 */
-		rdtscll(tsc_this);
-		delta = vcpu->arch.host_tsc - tsc_this;
-		vmcs_write64(TSC_OFFSET, vmcs_read64(TSC_OFFSET) + delta);
-	}
-}
-
-static void vmx_vcpu_put(struct kvm_vcpu *vcpu)
-{
-	vmx_load_host_state(to_vmx(vcpu));
-}
-
-static void vmx_fpu_activate(struct kvm_vcpu *vcpu)
-{
-	if (vcpu->fpu_active)
-		return;
-	vcpu->fpu_active = 1;
-	vmcs_clear_bits(GUEST_CR0, X86_CR0_TS);
-	if (vcpu->arch.cr0 & X86_CR0_TS)
-		vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
-	update_exception_bitmap(vcpu);
-}
-
-static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu)
-{
-	if (!vcpu->fpu_active)
-		return;
-	vcpu->fpu_active = 0;
-	vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
-	update_exception_bitmap(vcpu);
-}
-
-static void vmx_vcpu_decache(struct kvm_vcpu *vcpu)
-{
-	vcpu_clear(to_vmx(vcpu));
-}
-
-static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu)
-{
-	return vmcs_readl(GUEST_RFLAGS);
-}
-
-static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
-{
-	if (vcpu->arch.rmode.active)
-		rflags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
-	vmcs_writel(GUEST_RFLAGS, rflags);
-}
-
-static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
-{
-	unsigned long rip;
-	u32 interruptibility;
-
-	rip = vmcs_readl(GUEST_RIP);
-	rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
-	vmcs_writel(GUEST_RIP, rip);
-
-	/*
-	 * We emulated an instruction, so temporary interrupt blocking
-	 * should be removed, if set.
-	 */
-	interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
-	if (interruptibility & 3)
-		vmcs_write32(GUEST_INTERRUPTIBILITY_INFO,
-			     interruptibility & ~3);
-	vcpu->arch.interrupt_window_open = 1;
-}
-
-static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
-				bool has_error_code, u32 error_code)
-{
-	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
-		     nr | INTR_TYPE_EXCEPTION
-		     | (has_error_code ? INTR_INFO_DELIEVER_CODE_MASK : 0)
-		     | INTR_INFO_VALID_MASK);
-	if (has_error_code)
-		vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code);
-}
-
-static bool vmx_exception_injected(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-
-	return !(vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK);
-}
-
-/*
- * Swap MSR entry in host/guest MSR entry array.
- */
-#ifdef CONFIG_X86_64
-static void move_msr_up(struct vcpu_vmx *vmx, int from, int to)
-{
-	struct kvm_msr_entry tmp;
-
-	tmp = vmx->guest_msrs[to];
-	vmx->guest_msrs[to] = vmx->guest_msrs[from];
-	vmx->guest_msrs[from] = tmp;
-	tmp = vmx->host_msrs[to];
-	vmx->host_msrs[to] = vmx->host_msrs[from];
-	vmx->host_msrs[from] = tmp;
-}
-#endif
-
-/*
- * Set up the vmcs to automatically save and restore system
- * msrs.  Don't touch the 64-bit msrs if the guest is in legacy
- * mode, as fiddling with msrs is very expensive.
- */
-static void setup_msrs(struct vcpu_vmx *vmx)
-{
-	int save_nmsrs;
-
-	save_nmsrs = 0;
-#ifdef CONFIG_X86_64
-	if (is_long_mode(&vmx->vcpu)) {
-		int index;
-
-		index = __find_msr_index(vmx, MSR_SYSCALL_MASK);
-		if (index >= 0)
-			move_msr_up(vmx, index, save_nmsrs++);
-		index = __find_msr_index(vmx, MSR_LSTAR);
-		if (index >= 0)
-			move_msr_up(vmx, index, save_nmsrs++);
-		index = __find_msr_index(vmx, MSR_CSTAR);
-		if (index >= 0)
-			move_msr_up(vmx, index, save_nmsrs++);
-		index = __find_msr_index(vmx, MSR_KERNEL_GS_BASE);
-		if (index >= 0)
-			move_msr_up(vmx, index, save_nmsrs++);
-		/*
-		 * MSR_K6_STAR is only needed on long mode guests, and only
-		 * if efer.sce is enabled.
-		 */
-		index = __find_msr_index(vmx, MSR_K6_STAR);
-		if ((index >= 0) && (vmx->vcpu.arch.shadow_efer & EFER_SCE))
-			move_msr_up(vmx, index, save_nmsrs++);
-	}
-#endif
-	vmx->save_nmsrs = save_nmsrs;
-
-#ifdef CONFIG_X86_64
-	vmx->msr_offset_kernel_gs_base =
-		__find_msr_index(vmx, MSR_KERNEL_GS_BASE);
-#endif
-	vmx->msr_offset_efer = __find_msr_index(vmx, MSR_EFER);
-}
-
-/*
- * reads and returns guest's timestamp counter "register"
- * guest_tsc = host_tsc + tsc_offset    -- 21.3
- */
-static u64 guest_read_tsc(void)
-{
-	u64 host_tsc, tsc_offset;
-
-	rdtscll(host_tsc);
-	tsc_offset = vmcs_read64(TSC_OFFSET);
-	return host_tsc + tsc_offset;
-}
-
-/*
- * writes 'guest_tsc' into guest's timestamp counter "register"
- * guest_tsc = host_tsc + tsc_offset ==> tsc_offset = guest_tsc - host_tsc
- */
-static void guest_write_tsc(u64 guest_tsc)
-{
-	u64 host_tsc;
-
-	rdtscll(host_tsc);
-	vmcs_write64(TSC_OFFSET, guest_tsc - host_tsc);
-}
-
-/*
- * Reads an msr value (of 'msr_index') into 'pdata'.
- * Returns 0 on success, non-0 otherwise.
- * Assumes vcpu_load() was already called.
- */
-static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata)
-{
-	u64 data;
-	struct kvm_msr_entry *msr;
-
-	if (!pdata) {
-		printk(KERN_ERR "BUG: get_msr called with NULL pdata\n");
-		return -EINVAL;
-	}
-
-	switch (msr_index) {
-#ifdef CONFIG_X86_64
-	case MSR_FS_BASE:
-		data = vmcs_readl(GUEST_FS_BASE);
-		break;
-	case MSR_GS_BASE:
-		data = vmcs_readl(GUEST_GS_BASE);
-		break;
-	case MSR_EFER:
-		return kvm_get_msr_common(vcpu, msr_index, pdata);
-#endif
-	case MSR_IA32_TIME_STAMP_COUNTER:
-		data = guest_read_tsc();
-		break;
-	case MSR_IA32_SYSENTER_CS:
-		data = vmcs_read32(GUEST_SYSENTER_CS);
-		break;
-	case MSR_IA32_SYSENTER_EIP:
-		data = vmcs_readl(GUEST_SYSENTER_EIP);
-		break;
-	case MSR_IA32_SYSENTER_ESP:
-		data = vmcs_readl(GUEST_SYSENTER_ESP);
-		break;
-	default:
-		msr = find_msr_entry(to_vmx(vcpu), msr_index);
-		if (msr) {
-			data = msr->data;
-			break;
-		}
-		return kvm_get_msr_common(vcpu, msr_index, pdata);
-	}
-
-	*pdata = data;
-	return 0;
-}
-
-/*
- * Writes msr value into into the appropriate "register".
- * Returns 0 on success, non-0 otherwise.
- * Assumes vcpu_load() was already called.
- */
-static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	struct kvm_msr_entry *msr;
-	int ret = 0;
-
-	switch (msr_index) {
-#ifdef CONFIG_X86_64
-	case MSR_EFER:
-		ret = kvm_set_msr_common(vcpu, msr_index, data);
-		if (vmx->host_state.loaded) {
-			reload_host_efer(vmx);
-			load_transition_efer(vmx);
-		}
-		break;
-	case MSR_FS_BASE:
-		vmcs_writel(GUEST_FS_BASE, data);
-		break;
-	case MSR_GS_BASE:
-		vmcs_writel(GUEST_GS_BASE, data);
-		break;
-#endif
-	case MSR_IA32_SYSENTER_CS:
-		vmcs_write32(GUEST_SYSENTER_CS, data);
-		break;
-	case MSR_IA32_SYSENTER_EIP:
-		vmcs_writel(GUEST_SYSENTER_EIP, data);
-		break;
-	case MSR_IA32_SYSENTER_ESP:
-		vmcs_writel(GUEST_SYSENTER_ESP, data);
-		break;
-	case MSR_IA32_TIME_STAMP_COUNTER:
-		guest_write_tsc(data);
-		break;
-	default:
-		msr = find_msr_entry(vmx, msr_index);
-		if (msr) {
-			msr->data = data;
-			if (vmx->host_state.loaded)
-				load_msrs(vmx->guest_msrs, vmx->save_nmsrs);
-			break;
-		}
-		ret = kvm_set_msr_common(vcpu, msr_index, data);
-	}
-
-	return ret;
-}
-
-/*
- * Sync the rsp and rip registers into the vcpu structure.  This allows
- * registers to be accessed by indexing vcpu->arch.regs.
- */
-static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP);
-	vcpu->arch.rip = vmcs_readl(GUEST_RIP);
-}
-
-/*
- * Syncs rsp and rip back into the vmcs.  Should be called after possible
- * modification.
- */
-static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu)
-{
-	vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]);
-	vmcs_writel(GUEST_RIP, vcpu->arch.rip);
-}
-
-static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
-{
-	unsigned long dr7 = 0x400;
-	int old_singlestep;
-
-	old_singlestep = vcpu->guest_debug.singlestep;
-
-	vcpu->guest_debug.enabled = dbg->enabled;
-	if (vcpu->guest_debug.enabled) {
-		int i;
-
-		dr7 |= 0x200;  /* exact */
-		for (i = 0; i < 4; ++i) {
-			if (!dbg->breakpoints[i].enabled)
-				continue;
-			vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address;
-			dr7 |= 2 << (i*2);    /* global enable */
-			dr7 |= 0 << (i*4+16); /* execution breakpoint */
-		}
-
-		vcpu->guest_debug.singlestep = dbg->singlestep;
-	} else
-		vcpu->guest_debug.singlestep = 0;
-
-	if (old_singlestep && !vcpu->guest_debug.singlestep) {
-		unsigned long flags;
-
-		flags = vmcs_readl(GUEST_RFLAGS);
-		flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
-
-	update_exception_bitmap(vcpu);
-	vmcs_writel(GUEST_DR7, dr7);
-
-	return 0;
-}
-
-static int vmx_get_irq(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 idtv_info_field;
-
-	idtv_info_field = vmx->idt_vectoring_info;
-	if (idtv_info_field & INTR_INFO_VALID_MASK) {
-		if (is_external_interrupt(idtv_info_field))
-			return idtv_info_field & VECTORING_INFO_VECTOR_MASK;
-		else
-			printk(KERN_DEBUG "pending exception: not handled yet\n");
-	}
-	return -1;
-}
-
-static __init int cpu_has_kvm_support(void)
-{
-	unsigned long ecx = cpuid_ecx(1);
-	return test_bit(5, &ecx); /* CPUID.1:ECX.VMX[bit 5] -> VT */
-}
-
-static __init int vmx_disabled_by_bios(void)
-{
-	u64 msr;
-
-	rdmsrl(MSR_IA32_FEATURE_CONTROL, msr);
-	return (msr & (MSR_IA32_FEATURE_CONTROL_LOCKED |
-		       MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
-	    == MSR_IA32_FEATURE_CONTROL_LOCKED;
-	/* locked but not enabled */
-}
-
-static void hardware_enable(void *garbage)
-{
-	int cpu = raw_smp_processor_id();
-	u64 phys_addr = __pa(per_cpu(vmxarea, cpu));
-	u64 old;
-
-	rdmsrl(MSR_IA32_FEATURE_CONTROL, old);
-	if ((old & (MSR_IA32_FEATURE_CONTROL_LOCKED |
-		    MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
-	    != (MSR_IA32_FEATURE_CONTROL_LOCKED |
-		MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED))
-		/* enable and lock */
-		wrmsrl(MSR_IA32_FEATURE_CONTROL, old |
-		       MSR_IA32_FEATURE_CONTROL_LOCKED |
-		       MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED);
-	write_cr4(read_cr4() | X86_CR4_VMXE); /* FIXME: not cpu hotplug safe */
-	asm volatile (ASM_VMX_VMXON_RAX : : "a"(&phys_addr), "m"(phys_addr)
-		      : "memory", "cc");
-}
-
-static void hardware_disable(void *garbage)
-{
-	asm volatile (ASM_VMX_VMXOFF : : : "cc");
-}
-
-static __init int adjust_vmx_controls(u32 ctl_min, u32 ctl_opt,
-				      u32 msr, u32 *result)
-{
-	u32 vmx_msr_low, vmx_msr_high;
-	u32 ctl = ctl_min | ctl_opt;
-
-	rdmsr(msr, vmx_msr_low, vmx_msr_high);
-
-	ctl &= vmx_msr_high; /* bit == 0 in high word ==> must be zero */
-	ctl |= vmx_msr_low;  /* bit == 1 in low word  ==> must be one  */
-
-	/* Ensure minimum (required) set of control bits are supported. */
-	if (ctl_min & ~ctl)
-		return -EIO;
-
-	*result = ctl;
-	return 0;
-}
-
-static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf)
-{
-	u32 vmx_msr_low, vmx_msr_high;
-	u32 min, opt;
-	u32 _pin_based_exec_control = 0;
-	u32 _cpu_based_exec_control = 0;
-	u32 _cpu_based_2nd_exec_control = 0;
-	u32 _vmexit_control = 0;
-	u32 _vmentry_control = 0;
-
-	min = PIN_BASED_EXT_INTR_MASK | PIN_BASED_NMI_EXITING;
-	opt = 0;
-	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PINBASED_CTLS,
-				&_pin_based_exec_control) < 0)
-		return -EIO;
-
-	min = CPU_BASED_HLT_EXITING |
-#ifdef CONFIG_X86_64
-	      CPU_BASED_CR8_LOAD_EXITING |
-	      CPU_BASED_CR8_STORE_EXITING |
-#endif
-	      CPU_BASED_USE_IO_BITMAPS |
-	      CPU_BASED_MOV_DR_EXITING |
-	      CPU_BASED_USE_TSC_OFFSETING;
-	opt = CPU_BASED_TPR_SHADOW |
-	      CPU_BASED_ACTIVATE_SECONDARY_CONTROLS;
-	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS,
-				&_cpu_based_exec_control) < 0)
-		return -EIO;
-#ifdef CONFIG_X86_64
-	if ((_cpu_based_exec_control & CPU_BASED_TPR_SHADOW))
-		_cpu_based_exec_control &= ~CPU_BASED_CR8_LOAD_EXITING &
-					   ~CPU_BASED_CR8_STORE_EXITING;
-#endif
-	if (_cpu_based_exec_control & CPU_BASED_ACTIVATE_SECONDARY_CONTROLS) {
-		min = 0;
-		opt = SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
-			SECONDARY_EXEC_WBINVD_EXITING;
-		if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS2,
-					&_cpu_based_2nd_exec_control) < 0)
-			return -EIO;
-	}
-#ifndef CONFIG_X86_64
-	if (!(_cpu_based_2nd_exec_control &
-				SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES))
-		_cpu_based_exec_control &= ~CPU_BASED_TPR_SHADOW;
-#endif
-
-	min = 0;
-#ifdef CONFIG_X86_64
-	min |= VM_EXIT_HOST_ADDR_SPACE_SIZE;
-#endif
-	opt = 0;
-	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_EXIT_CTLS,
-				&_vmexit_control) < 0)
-		return -EIO;
-
-	min = opt = 0;
-	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_ENTRY_CTLS,
-				&_vmentry_control) < 0)
-		return -EIO;
-
-	rdmsr(MSR_IA32_VMX_BASIC, vmx_msr_low, vmx_msr_high);
-
-	/* IA-32 SDM Vol 3B: VMCS size is never greater than 4kB. */
-	if ((vmx_msr_high & 0x1fff) > PAGE_SIZE)
-		return -EIO;
-
-#ifdef CONFIG_X86_64
-	/* IA-32 SDM Vol 3B: 64-bit CPUs always have VMX_BASIC_MSR[48]==0. */
-	if (vmx_msr_high & (1u<<16))
-		return -EIO;
-#endif
-
-	/* Require Write-Back (WB) memory type for VMCS accesses. */
-	if (((vmx_msr_high >> 18) & 15) != 6)
-		return -EIO;
-
-	vmcs_conf->size = vmx_msr_high & 0x1fff;
-	vmcs_conf->order = get_order(vmcs_config.size);
-	vmcs_conf->revision_id = vmx_msr_low;
-
-	vmcs_conf->pin_based_exec_ctrl = _pin_based_exec_control;
-	vmcs_conf->cpu_based_exec_ctrl = _cpu_based_exec_control;
-	vmcs_conf->cpu_based_2nd_exec_ctrl = _cpu_based_2nd_exec_control;
-	vmcs_conf->vmexit_ctrl         = _vmexit_control;
-	vmcs_conf->vmentry_ctrl        = _vmentry_control;
-
-	return 0;
-}
-
-static struct vmcs *alloc_vmcs_cpu(int cpu)
-{
-	int node = cpu_to_node(cpu);
-	struct page *pages;
-	struct vmcs *vmcs;
-
-	pages = alloc_pages_node(node, GFP_KERNEL, vmcs_config.order);
-	if (!pages)
-		return NULL;
-	vmcs = page_address(pages);
-	memset(vmcs, 0, vmcs_config.size);
-	vmcs->revision_id = vmcs_config.revision_id; /* vmcs revision id */
-	return vmcs;
-}
-
-static struct vmcs *alloc_vmcs(void)
-{
-	return alloc_vmcs_cpu(raw_smp_processor_id());
-}
-
-static void free_vmcs(struct vmcs *vmcs)
-{
-	free_pages((unsigned long)vmcs, vmcs_config.order);
-}
-
-static void free_kvm_area(void)
-{
-	int cpu;
-
-	for_each_online_cpu(cpu)
-		free_vmcs(per_cpu(vmxarea, cpu));
-}
-
-static __init int alloc_kvm_area(void)
-{
-	int cpu;
-
-	for_each_online_cpu(cpu) {
-		struct vmcs *vmcs;
-
-		vmcs = alloc_vmcs_cpu(cpu);
-		if (!vmcs) {
-			free_kvm_area();
-			return -ENOMEM;
-		}
-
-		per_cpu(vmxarea, cpu) = vmcs;
-	}
-	return 0;
-}
-
-static __init int hardware_setup(void)
-{
-	if (setup_vmcs_config(&vmcs_config) < 0)
-		return -EIO;
-	return alloc_kvm_area();
-}
-
-static __exit void hardware_unsetup(void)
-{
-	free_kvm_area();
-}
-
-static void fix_pmode_dataseg(int seg, struct kvm_save_segment *save)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-
-	if (vmcs_readl(sf->base) == save->base && (save->base & AR_S_MASK)) {
-		vmcs_write16(sf->selector, save->selector);
-		vmcs_writel(sf->base, save->base);
-		vmcs_write32(sf->limit, save->limit);
-		vmcs_write32(sf->ar_bytes, save->ar);
-	} else {
-		u32 dpl = (vmcs_read16(sf->selector) & SELECTOR_RPL_MASK)
-			<< AR_DPL_SHIFT;
-		vmcs_write32(sf->ar_bytes, 0x93 | dpl);
-	}
-}
-
-static void enter_pmode(struct kvm_vcpu *vcpu)
-{
-	unsigned long flags;
-
-	vcpu->arch.rmode.active = 0;
-
-	vmcs_writel(GUEST_TR_BASE, vcpu->arch.rmode.tr.base);
-	vmcs_write32(GUEST_TR_LIMIT, vcpu->arch.rmode.tr.limit);
-	vmcs_write32(GUEST_TR_AR_BYTES, vcpu->arch.rmode.tr.ar);
-
-	flags = vmcs_readl(GUEST_RFLAGS);
-	flags &= ~(X86_EFLAGS_IOPL | X86_EFLAGS_VM);
-	flags |= (vcpu->arch.rmode.save_iopl << IOPL_SHIFT);
-	vmcs_writel(GUEST_RFLAGS, flags);
-
-	vmcs_writel(GUEST_CR4, (vmcs_readl(GUEST_CR4) & ~X86_CR4_VME) |
-			(vmcs_readl(CR4_READ_SHADOW) & X86_CR4_VME));
-
-	update_exception_bitmap(vcpu);
-
-	fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
-	fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
-	fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
-	fix_pmode_dataseg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
-
-	vmcs_write16(GUEST_SS_SELECTOR, 0);
-	vmcs_write32(GUEST_SS_AR_BYTES, 0x93);
-
-	vmcs_write16(GUEST_CS_SELECTOR,
-		     vmcs_read16(GUEST_CS_SELECTOR) & ~SELECTOR_RPL_MASK);
-	vmcs_write32(GUEST_CS_AR_BYTES, 0x9b);
-}
-
-static gva_t rmode_tss_base(struct kvm *kvm)
-{
-	if (!kvm->arch.tss_addr) {
-		gfn_t base_gfn = kvm->memslots[0].base_gfn +
-				 kvm->memslots[0].npages - 3;
-		return base_gfn << PAGE_SHIFT;
-	}
-	return kvm->arch.tss_addr;
-}
-
-static void fix_rmode_seg(int seg, struct kvm_save_segment *save)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-
-	save->selector = vmcs_read16(sf->selector);
-	save->base = vmcs_readl(sf->base);
-	save->limit = vmcs_read32(sf->limit);
-	save->ar = vmcs_read32(sf->ar_bytes);
-	vmcs_write16(sf->selector, save->base >> 4);
-	vmcs_write32(sf->base, save->base & 0xfffff);
-	vmcs_write32(sf->limit, 0xffff);
-	vmcs_write32(sf->ar_bytes, 0xf3);
-}
-
-static void enter_rmode(struct kvm_vcpu *vcpu)
-{
-	unsigned long flags;
-
-	vcpu->arch.rmode.active = 1;
-
-	vcpu->arch.rmode.tr.base = vmcs_readl(GUEST_TR_BASE);
-	vmcs_writel(GUEST_TR_BASE, rmode_tss_base(vcpu->kvm));
-
-	vcpu->arch.rmode.tr.limit = vmcs_read32(GUEST_TR_LIMIT);
-	vmcs_write32(GUEST_TR_LIMIT, RMODE_TSS_SIZE - 1);
-
-	vcpu->arch.rmode.tr.ar = vmcs_read32(GUEST_TR_AR_BYTES);
-	vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
-
-	flags = vmcs_readl(GUEST_RFLAGS);
-	vcpu->arch.rmode.save_iopl
-		= (flags & X86_EFLAGS_IOPL) >> IOPL_SHIFT;
-
-	flags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM;
-
-	vmcs_writel(GUEST_RFLAGS, flags);
-	vmcs_writel(GUEST_CR4, vmcs_readl(GUEST_CR4) | X86_CR4_VME);
-	update_exception_bitmap(vcpu);
-
-	vmcs_write16(GUEST_SS_SELECTOR, vmcs_readl(GUEST_SS_BASE) >> 4);
-	vmcs_write32(GUEST_SS_LIMIT, 0xffff);
-	vmcs_write32(GUEST_SS_AR_BYTES, 0xf3);
-
-	vmcs_write32(GUEST_CS_AR_BYTES, 0xf3);
-	vmcs_write32(GUEST_CS_LIMIT, 0xffff);
-	if (vmcs_readl(GUEST_CS_BASE) == 0xffff0000)
-		vmcs_writel(GUEST_CS_BASE, 0xf0000);
-	vmcs_write16(GUEST_CS_SELECTOR, vmcs_readl(GUEST_CS_BASE) >> 4);
-
-	fix_rmode_seg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
-	fix_rmode_seg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
-	fix_rmode_seg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
-	fix_rmode_seg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
-
-	kvm_mmu_reset_context(vcpu);
-	init_rmode_tss(vcpu->kvm);
-}
-
-#ifdef CONFIG_X86_64
-
-static void enter_lmode(struct kvm_vcpu *vcpu)
-{
-	u32 guest_tr_ar;
-
-	guest_tr_ar = vmcs_read32(GUEST_TR_AR_BYTES);
-	if ((guest_tr_ar & AR_TYPE_MASK) != AR_TYPE_BUSY_64_TSS) {
-		printk(KERN_DEBUG "%s: tss fixup for long mode. \n",
-		       __FUNCTION__);
-		vmcs_write32(GUEST_TR_AR_BYTES,
-			     (guest_tr_ar & ~AR_TYPE_MASK)
-			     | AR_TYPE_BUSY_64_TSS);
-	}
-
-	vcpu->arch.shadow_efer |= EFER_LMA;
-
-	find_msr_entry(to_vmx(vcpu), MSR_EFER)->data |= EFER_LMA | EFER_LME;
-	vmcs_write32(VM_ENTRY_CONTROLS,
-		     vmcs_read32(VM_ENTRY_CONTROLS)
-		     | VM_ENTRY_IA32E_MODE);
-}
-
-static void exit_lmode(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.shadow_efer &= ~EFER_LMA;
-
-	vmcs_write32(VM_ENTRY_CONTROLS,
-		     vmcs_read32(VM_ENTRY_CONTROLS)
-		     & ~VM_ENTRY_IA32E_MODE);
-}
-
-#endif
-
-static void vmx_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.cr4 &= KVM_GUEST_CR4_MASK;
-	vcpu->arch.cr4 |= vmcs_readl(GUEST_CR4) & ~KVM_GUEST_CR4_MASK;
-}
-
-static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
-{
-	vmx_fpu_deactivate(vcpu);
-
-	if (vcpu->arch.rmode.active && (cr0 & X86_CR0_PE))
-		enter_pmode(vcpu);
-
-	if (!vcpu->arch.rmode.active && !(cr0 & X86_CR0_PE))
-		enter_rmode(vcpu);
-
-#ifdef CONFIG_X86_64
-	if (vcpu->arch.shadow_efer & EFER_LME) {
-		if (!is_paging(vcpu) && (cr0 & X86_CR0_PG))
-			enter_lmode(vcpu);
-		if (is_paging(vcpu) && !(cr0 & X86_CR0_PG))
-			exit_lmode(vcpu);
-	}
-#endif
-
-	vmcs_writel(CR0_READ_SHADOW, cr0);
-	vmcs_writel(GUEST_CR0,
-		    (cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON);
-	vcpu->arch.cr0 = cr0;
-
-	if (!(cr0 & X86_CR0_TS) || !(cr0 & X86_CR0_PE))
-		vmx_fpu_activate(vcpu);
-}
-
-static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
-{
-	vmcs_writel(GUEST_CR3, cr3);
-	if (vcpu->arch.cr0 & X86_CR0_PE)
-		vmx_fpu_deactivate(vcpu);
-}
-
-static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
-{
-	vmcs_writel(CR4_READ_SHADOW, cr4);
-	vmcs_writel(GUEST_CR4, cr4 | (vcpu->arch.rmode.active ?
-		    KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON));
-	vcpu->arch.cr4 = cr4;
-}
-
-#ifdef CONFIG_X86_64
-
-static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER);
-
-	vcpu->arch.shadow_efer = efer;
-	if (efer & EFER_LMA) {
-		vmcs_write32(VM_ENTRY_CONTROLS,
-				     vmcs_read32(VM_ENTRY_CONTROLS) |
-				     VM_ENTRY_IA32E_MODE);
-		msr->data = efer;
-
-	} else {
-		vmcs_write32(VM_ENTRY_CONTROLS,
-				     vmcs_read32(VM_ENTRY_CONTROLS) &
-				     ~VM_ENTRY_IA32E_MODE);
-
-		msr->data = efer & ~EFER_LME;
-	}
-	setup_msrs(vmx);
-}
-
-#endif
-
-static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-
-	return vmcs_readl(sf->base);
-}
-
-static void vmx_get_segment(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-	u32 ar;
-
-	var->base = vmcs_readl(sf->base);
-	var->limit = vmcs_read32(sf->limit);
-	var->selector = vmcs_read16(sf->selector);
-	ar = vmcs_read32(sf->ar_bytes);
-	if (ar & AR_UNUSABLE_MASK)
-		ar = 0;
-	var->type = ar & 15;
-	var->s = (ar >> 4) & 1;
-	var->dpl = (ar >> 5) & 3;
-	var->present = (ar >> 7) & 1;
-	var->avl = (ar >> 12) & 1;
-	var->l = (ar >> 13) & 1;
-	var->db = (ar >> 14) & 1;
-	var->g = (ar >> 15) & 1;
-	var->unusable = (ar >> 16) & 1;
-}
-
-static u32 vmx_segment_access_rights(struct kvm_segment *var)
-{
-	u32 ar;
-
-	if (var->unusable)
-		ar = 1 << 16;
-	else {
-		ar = var->type & 15;
-		ar |= (var->s & 1) << 4;
-		ar |= (var->dpl & 3) << 5;
-		ar |= (var->present & 1) << 7;
-		ar |= (var->avl & 1) << 12;
-		ar |= (var->l & 1) << 13;
-		ar |= (var->db & 1) << 14;
-		ar |= (var->g & 1) << 15;
-	}
-	if (ar == 0) /* a 0 value means unusable */
-		ar = AR_UNUSABLE_MASK;
-
-	return ar;
-}
-
-static void vmx_set_segment(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-	u32 ar;
-
-	if (vcpu->arch.rmode.active && seg == VCPU_SREG_TR) {
-		vcpu->arch.rmode.tr.selector = var->selector;
-		vcpu->arch.rmode.tr.base = var->base;
-		vcpu->arch.rmode.tr.limit = var->limit;
-		vcpu->arch.rmode.tr.ar = vmx_segment_access_rights(var);
-		return;
-	}
-	vmcs_writel(sf->base, var->base);
-	vmcs_write32(sf->limit, var->limit);
-	vmcs_write16(sf->selector, var->selector);
-	if (vcpu->arch.rmode.active && var->s) {
-		/*
-		 * Hack real-mode segments into vm86 compatibility.
-		 */
-		if (var->base == 0xffff0000 && var->selector == 0xf000)
-			vmcs_writel(sf->base, 0xf0000);
-		ar = 0xf3;
-	} else
-		ar = vmx_segment_access_rights(var);
-	vmcs_write32(sf->ar_bytes, ar);
-}
-
-static void vmx_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l)
-{
-	u32 ar = vmcs_read32(GUEST_CS_AR_BYTES);
-
-	*db = (ar >> 14) & 1;
-	*l = (ar >> 13) & 1;
-}
-
-static void vmx_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	dt->limit = vmcs_read32(GUEST_IDTR_LIMIT);
-	dt->base = vmcs_readl(GUEST_IDTR_BASE);
-}
-
-static void vmx_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	vmcs_write32(GUEST_IDTR_LIMIT, dt->limit);
-	vmcs_writel(GUEST_IDTR_BASE, dt->base);
-}
-
-static void vmx_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	dt->limit = vmcs_read32(GUEST_GDTR_LIMIT);
-	dt->base = vmcs_readl(GUEST_GDTR_BASE);
-}
-
-static void vmx_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt)
-{
-	vmcs_write32(GUEST_GDTR_LIMIT, dt->limit);
-	vmcs_writel(GUEST_GDTR_BASE, dt->base);
-}
-
-static int init_rmode_tss(struct kvm *kvm)
-{
-	gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT;
-	u16 data = 0;
-	int r;
-
-	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
-	if (r < 0)
-		return 0;
-	data = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE;
-	r = kvm_write_guest_page(kvm, fn++, &data, 0x66, sizeof(u16));
-	if (r < 0)
-		return 0;
-	r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE);
-	if (r < 0)
-		return 0;
-	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
-	if (r < 0)
-		return 0;
-	data = ~0;
-	r = kvm_write_guest_page(kvm, fn, &data, RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
-			sizeof(u8));
-	if (r < 0)
-		return 0;
-	return 1;
-}
-
-static void seg_setup(int seg)
-{
-	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
-
-	vmcs_write16(sf->selector, 0);
-	vmcs_writel(sf->base, 0);
-	vmcs_write32(sf->limit, 0xffff);
-	vmcs_write32(sf->ar_bytes, 0x93);
-}
-
-static int alloc_apic_access_page(struct kvm *kvm)
-{
-	struct kvm_userspace_memory_region kvm_userspace_mem;
-	int r = 0;
-
-	mutex_lock(&kvm->lock);
-	if (kvm->arch.apic_access_page)
-		goto out;
-	kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT;
-	kvm_userspace_mem.flags = 0;
-	kvm_userspace_mem.guest_phys_addr = 0xfee00000ULL;
-	kvm_userspace_mem.memory_size = PAGE_SIZE;
-	r = __kvm_set_memory_region(kvm, &kvm_userspace_mem, 0);
-	if (r)
-		goto out;
-	kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
-out:
-	mutex_unlock(&kvm->lock);
-	return r;
-}
-
-/*
- * Sets up the vmcs for emulated real mode.
- */
-static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
-{
-	u32 host_sysenter_cs;
-	u32 junk;
-	unsigned long a;
-	struct descriptor_table dt;
-	int i;
-	unsigned long kvm_vmx_return;
-	u32 exec_control;
-
-	/* I/O */
-	vmcs_write64(IO_BITMAP_A, page_to_phys(vmx_io_bitmap_a));
-	vmcs_write64(IO_BITMAP_B, page_to_phys(vmx_io_bitmap_b));
-
-	vmcs_write64(VMCS_LINK_POINTER, -1ull); /* 22.3.1.5 */
-
-	/* Control */
-	vmcs_write32(PIN_BASED_VM_EXEC_CONTROL,
-		vmcs_config.pin_based_exec_ctrl);
-
-	exec_control = vmcs_config.cpu_based_exec_ctrl;
-	if (!vm_need_tpr_shadow(vmx->vcpu.kvm)) {
-		exec_control &= ~CPU_BASED_TPR_SHADOW;
-#ifdef CONFIG_X86_64
-		exec_control |= CPU_BASED_CR8_STORE_EXITING |
-				CPU_BASED_CR8_LOAD_EXITING;
-#endif
-	}
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, exec_control);
-
-	if (cpu_has_secondary_exec_ctrls()) {
-		exec_control = vmcs_config.cpu_based_2nd_exec_ctrl;
-		if (!vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
-			exec_control &=
-				~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES;
-		vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control);
-	}
-
-	vmcs_write32(PAGE_FAULT_ERROR_CODE_MASK, !!bypass_guest_pf);
-	vmcs_write32(PAGE_FAULT_ERROR_CODE_MATCH, !!bypass_guest_pf);
-	vmcs_write32(CR3_TARGET_COUNT, 0);           /* 22.2.1 */
-
-	vmcs_writel(HOST_CR0, read_cr0());  /* 22.2.3 */
-	vmcs_writel(HOST_CR4, read_cr4());  /* 22.2.3, 22.2.5 */
-	vmcs_writel(HOST_CR3, read_cr3());  /* 22.2.3  FIXME: shadow tables */
-
-	vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS);  /* 22.2.4 */
-	vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
-	vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
-	vmcs_write16(HOST_FS_SELECTOR, read_fs());    /* 22.2.4 */
-	vmcs_write16(HOST_GS_SELECTOR, read_gs());    /* 22.2.4 */
-	vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
-#ifdef CONFIG_X86_64
-	rdmsrl(MSR_FS_BASE, a);
-	vmcs_writel(HOST_FS_BASE, a); /* 22.2.4 */
-	rdmsrl(MSR_GS_BASE, a);
-	vmcs_writel(HOST_GS_BASE, a); /* 22.2.4 */
-#else
-	vmcs_writel(HOST_FS_BASE, 0); /* 22.2.4 */
-	vmcs_writel(HOST_GS_BASE, 0); /* 22.2.4 */
-#endif
-
-	vmcs_write16(HOST_TR_SELECTOR, GDT_ENTRY_TSS*8);  /* 22.2.4 */
-
-	get_idt(&dt);
-	vmcs_writel(HOST_IDTR_BASE, dt.base);   /* 22.2.4 */
-
-	asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return));
-	vmcs_writel(HOST_RIP, kvm_vmx_return); /* 22.2.5 */
-	vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0);
-	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0);
-	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, 0);
-
-	rdmsr(MSR_IA32_SYSENTER_CS, host_sysenter_cs, junk);
-	vmcs_write32(HOST_IA32_SYSENTER_CS, host_sysenter_cs);
-	rdmsrl(MSR_IA32_SYSENTER_ESP, a);
-	vmcs_writel(HOST_IA32_SYSENTER_ESP, a);   /* 22.2.3 */
-	rdmsrl(MSR_IA32_SYSENTER_EIP, a);
-	vmcs_writel(HOST_IA32_SYSENTER_EIP, a);   /* 22.2.3 */
-
-	for (i = 0; i < NR_VMX_MSR; ++i) {
-		u32 index = vmx_msr_index[i];
-		u32 data_low, data_high;
-		u64 data;
-		int j = vmx->nmsrs;
-
-		if (rdmsr_safe(index, &data_low, &data_high) < 0)
-			continue;
-		if (wrmsr_safe(index, data_low, data_high) < 0)
-			continue;
-		data = data_low | ((u64)data_high << 32);
-		vmx->host_msrs[j].index = index;
-		vmx->host_msrs[j].reserved = 0;
-		vmx->host_msrs[j].data = data;
-		vmx->guest_msrs[j] = vmx->host_msrs[j];
-		++vmx->nmsrs;
-	}
-
-	vmcs_write32(VM_EXIT_CONTROLS, vmcs_config.vmexit_ctrl);
-
-	/* 22.2.1, 20.8.1 */
-	vmcs_write32(VM_ENTRY_CONTROLS, vmcs_config.vmentry_ctrl);
-
-	vmcs_writel(CR0_GUEST_HOST_MASK, ~0UL);
-	vmcs_writel(CR4_GUEST_HOST_MASK, KVM_GUEST_CR4_MASK);
-
-	if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
-		if (alloc_apic_access_page(vmx->vcpu.kvm) != 0)
-			return -ENOMEM;
-
-	return 0;
-}
-
-static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u64 msr;
-	int ret;
-
-	if (!init_rmode_tss(vmx->vcpu.kvm)) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	vmx->vcpu.arch.rmode.active = 0;
-
-	vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
-	set_cr8(&vmx->vcpu, 0);
-	msr = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
-	if (vmx->vcpu.vcpu_id == 0)
-		msr |= MSR_IA32_APICBASE_BSP;
-	kvm_set_apic_base(&vmx->vcpu, msr);
-
-	fx_init(&vmx->vcpu);
-
-	/*
-	 * GUEST_CS_BASE should really be 0xffff0000, but VT vm86 mode
-	 * insists on having GUEST_CS_BASE == GUEST_CS_SELECTOR << 4.  Sigh.
-	 */
-	if (vmx->vcpu.vcpu_id == 0) {
-		vmcs_write16(GUEST_CS_SELECTOR, 0xf000);
-		vmcs_writel(GUEST_CS_BASE, 0x000f0000);
-	} else {
-		vmcs_write16(GUEST_CS_SELECTOR, vmx->vcpu.arch.sipi_vector << 8);
-		vmcs_writel(GUEST_CS_BASE, vmx->vcpu.arch.sipi_vector << 12);
-	}
-	vmcs_write32(GUEST_CS_LIMIT, 0xffff);
-	vmcs_write32(GUEST_CS_AR_BYTES, 0x9b);
-
-	seg_setup(VCPU_SREG_DS);
-	seg_setup(VCPU_SREG_ES);
-	seg_setup(VCPU_SREG_FS);
-	seg_setup(VCPU_SREG_GS);
-	seg_setup(VCPU_SREG_SS);
-
-	vmcs_write16(GUEST_TR_SELECTOR, 0);
-	vmcs_writel(GUEST_TR_BASE, 0);
-	vmcs_write32(GUEST_TR_LIMIT, 0xffff);
-	vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
-
-	vmcs_write16(GUEST_LDTR_SELECTOR, 0);
-	vmcs_writel(GUEST_LDTR_BASE, 0);
-	vmcs_write32(GUEST_LDTR_LIMIT, 0xffff);
-	vmcs_write32(GUEST_LDTR_AR_BYTES, 0x00082);
-
-	vmcs_write32(GUEST_SYSENTER_CS, 0);
-	vmcs_writel(GUEST_SYSENTER_ESP, 0);
-	vmcs_writel(GUEST_SYSENTER_EIP, 0);
-
-	vmcs_writel(GUEST_RFLAGS, 0x02);
-	if (vmx->vcpu.vcpu_id == 0)
-		vmcs_writel(GUEST_RIP, 0xfff0);
-	else
-		vmcs_writel(GUEST_RIP, 0);
-	vmcs_writel(GUEST_RSP, 0);
-
-	/* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */
-	vmcs_writel(GUEST_DR7, 0x400);
-
-	vmcs_writel(GUEST_GDTR_BASE, 0);
-	vmcs_write32(GUEST_GDTR_LIMIT, 0xffff);
-
-	vmcs_writel(GUEST_IDTR_BASE, 0);
-	vmcs_write32(GUEST_IDTR_LIMIT, 0xffff);
-
-	vmcs_write32(GUEST_ACTIVITY_STATE, 0);
-	vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0);
-	vmcs_write32(GUEST_PENDING_DBG_EXCEPTIONS, 0);
-
-	guest_write_tsc(0);
-
-	/* Special registers */
-	vmcs_write64(GUEST_IA32_DEBUGCTL, 0);
-
-	setup_msrs(vmx);
-
-	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, 0);  /* 22.2.1 */
-
-	if (cpu_has_vmx_tpr_shadow()) {
-		vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, 0);
-		if (vm_need_tpr_shadow(vmx->vcpu.kvm))
-			vmcs_write64(VIRTUAL_APIC_PAGE_ADDR,
-				page_to_phys(vmx->vcpu.arch.apic->regs_page));
-		vmcs_write32(TPR_THRESHOLD, 0);
-	}
-
-	if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
-		vmcs_write64(APIC_ACCESS_ADDR,
-			     page_to_phys(vmx->vcpu.kvm->arch.apic_access_page));
-
-	vmx->vcpu.arch.cr0 = 0x60000010;
-	vmx_set_cr0(&vmx->vcpu, vmx->vcpu.arch.cr0); /* enter rmode */
-	vmx_set_cr4(&vmx->vcpu, 0);
-#ifdef CONFIG_X86_64
-	vmx_set_efer(&vmx->vcpu, 0);
-#endif
-	vmx_fpu_activate(&vmx->vcpu);
-	update_exception_bitmap(&vmx->vcpu);
-
-	return 0;
-
-out:
-	return ret;
-}
-
-static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-
-	if (vcpu->arch.rmode.active) {
-		vmx->rmode.irq.pending = true;
-		vmx->rmode.irq.vector = irq;
-		vmx->rmode.irq.rip = vmcs_readl(GUEST_RIP);
-		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
-			     irq | INTR_TYPE_SOFT_INTR | INTR_INFO_VALID_MASK);
-		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1);
-		vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip - 1);
-		return;
-	}
-	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
-			irq | INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK);
-}
-
-static void kvm_do_inject_irq(struct kvm_vcpu *vcpu)
-{
-	int word_index = __ffs(vcpu->arch.irq_summary);
-	int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
-	int irq = word_index * BITS_PER_LONG + bit_index;
-
-	clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
-	if (!vcpu->arch.irq_pending[word_index])
-		clear_bit(word_index, &vcpu->arch.irq_summary);
-	vmx_inject_irq(vcpu, irq);
-}
-
-
-static void do_interrupt_requests(struct kvm_vcpu *vcpu,
-				       struct kvm_run *kvm_run)
-{
-	u32 cpu_based_vm_exec_control;
-
-	vcpu->arch.interrupt_window_open =
-		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
-		 (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
-
-	if (vcpu->arch.interrupt_window_open &&
-	    vcpu->arch.irq_summary &&
-	    !(vmcs_read32(VM_ENTRY_INTR_INFO_FIELD) & INTR_INFO_VALID_MASK))
-		/*
-		 * If interrupts enabled, and not blocked by sti or mov ss. Good.
-		 */
-		kvm_do_inject_irq(vcpu);
-
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
-	if (!vcpu->arch.interrupt_window_open &&
-	    (vcpu->arch.irq_summary || kvm_run->request_interrupt_window))
-		/*
-		 * Interrupts blocked.  Wait for unblock.
-		 */
-		cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
-	else
-		cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
-}
-
-static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
-{
-	int ret;
-	struct kvm_userspace_memory_region tss_mem = {
-		.slot = 8,
-		.guest_phys_addr = addr,
-		.memory_size = PAGE_SIZE * 3,
-		.flags = 0,
-	};
-
-	ret = kvm_set_memory_region(kvm, &tss_mem, 0);
-	if (ret)
-		return ret;
-	kvm->arch.tss_addr = addr;
-	return 0;
-}
-
-static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
-{
-	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-
-	set_debugreg(dbg->bp[0], 0);
-	set_debugreg(dbg->bp[1], 1);
-	set_debugreg(dbg->bp[2], 2);
-	set_debugreg(dbg->bp[3], 3);
-
-	if (dbg->singlestep) {
-		unsigned long flags;
-
-		flags = vmcs_readl(GUEST_RFLAGS);
-		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
-}
-
-static int handle_rmode_exception(struct kvm_vcpu *vcpu,
-				  int vec, u32 err_code)
-{
-	if (!vcpu->arch.rmode.active)
-		return 0;
-
-	/*
-	 * Instruction with address size override prefix opcode 0x67
-	 * Cause the #SS fault with 0 error code in VM86 mode.
-	 */
-	if (((vec == GP_VECTOR) || (vec == SS_VECTOR)) && err_code == 0)
-		if (emulate_instruction(vcpu, NULL, 0, 0, 0) == EMULATE_DONE)
-			return 1;
-	return 0;
-}
-
-static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 intr_info, error_code;
-	unsigned long cr2, rip;
-	u32 vect_info;
-	enum emulation_result er;
-
-	vect_info = vmx->idt_vectoring_info;
-	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
-
-	if ((vect_info & VECTORING_INFO_VALID_MASK) &&
-						!is_page_fault(intr_info))
-		printk(KERN_ERR "%s: unexpected, vectoring info 0x%x "
-		       "intr info 0x%x\n", __FUNCTION__, vect_info, intr_info);
-
-	if (!irqchip_in_kernel(vcpu->kvm) && is_external_interrupt(vect_info)) {
-		int irq = vect_info & VECTORING_INFO_VECTOR_MASK;
-		set_bit(irq, vcpu->arch.irq_pending);
-		set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
-	}
-
-	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
-		return 1;  /* already handled by vmx_vcpu_run() */
-
-	if (is_no_device(intr_info)) {
-		vmx_fpu_activate(vcpu);
-		return 1;
-	}
-
-	if (is_invalid_opcode(intr_info)) {
-		er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
-		if (er != EMULATE_DONE)
-			kvm_queue_exception(vcpu, UD_VECTOR);
-		return 1;
-	}
-
-	error_code = 0;
-	rip = vmcs_readl(GUEST_RIP);
-	if (intr_info & INTR_INFO_DELIEVER_CODE_MASK)
-		error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE);
-	if (is_page_fault(intr_info)) {
-		cr2 = vmcs_readl(EXIT_QUALIFICATION);
-		return kvm_mmu_page_fault(vcpu, cr2, error_code);
-	}
-
-	if (vcpu->arch.rmode.active &&
-	    handle_rmode_exception(vcpu, intr_info & INTR_INFO_VECTOR_MASK,
-								error_code)) {
-		if (vcpu->arch.halt_request) {
-			vcpu->arch.halt_request = 0;
-			return kvm_emulate_halt(vcpu);
-		}
-		return 1;
-	}
-
-	if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) ==
-	    (INTR_TYPE_EXCEPTION | 1)) {
-		kvm_run->exit_reason = KVM_EXIT_DEBUG;
-		return 0;
-	}
-	kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
-	kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK;
-	kvm_run->ex.error_code = error_code;
-	return 0;
-}
-
-static int handle_external_interrupt(struct kvm_vcpu *vcpu,
-				     struct kvm_run *kvm_run)
-{
-	++vcpu->stat.irq_exits;
-	return 1;
-}
-
-static int handle_triple_fault(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
-	return 0;
-}
-
-static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	unsigned long exit_qualification;
-	int size, down, in, string, rep;
-	unsigned port;
-
-	++vcpu->stat.io_exits;
-	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
-	string = (exit_qualification & 16) != 0;
-
-	if (string) {
-		if (emulate_instruction(vcpu,
-					kvm_run, 0, 0, 0) == EMULATE_DO_MMIO)
-			return 0;
-		return 1;
-	}
-
-	size = (exit_qualification & 7) + 1;
-	in = (exit_qualification & 8) != 0;
-	down = (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_DF) != 0;
-	rep = (exit_qualification & 32) != 0;
-	port = exit_qualification >> 16;
-
-	return kvm_emulate_pio(vcpu, kvm_run, in, size, port);
-}
-
-static void
-vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
-{
-	/*
-	 * Patch in the VMCALL instruction:
-	 */
-	hypercall[0] = 0x0f;
-	hypercall[1] = 0x01;
-	hypercall[2] = 0xc1;
-}
-
-static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	unsigned long exit_qualification;
-	int cr;
-	int reg;
-
-	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
-	cr = exit_qualification & 15;
-	reg = (exit_qualification >> 8) & 15;
-	switch ((exit_qualification >> 4) & 3) {
-	case 0: /* mov to cr */
-		switch (cr) {
-		case 0:
-			vcpu_load_rsp_rip(vcpu);
-			set_cr0(vcpu, vcpu->arch.regs[reg]);
-			skip_emulated_instruction(vcpu);
-			return 1;
-		case 3:
-			vcpu_load_rsp_rip(vcpu);
-			set_cr3(vcpu, vcpu->arch.regs[reg]);
-			skip_emulated_instruction(vcpu);
-			return 1;
-		case 4:
-			vcpu_load_rsp_rip(vcpu);
-			set_cr4(vcpu, vcpu->arch.regs[reg]);
-			skip_emulated_instruction(vcpu);
-			return 1;
-		case 8:
-			vcpu_load_rsp_rip(vcpu);
-			set_cr8(vcpu, vcpu->arch.regs[reg]);
-			skip_emulated_instruction(vcpu);
-			if (irqchip_in_kernel(vcpu->kvm))
-				return 1;
-			kvm_run->exit_reason = KVM_EXIT_SET_TPR;
-			return 0;
-		};
-		break;
-	case 2: /* clts */
-		vcpu_load_rsp_rip(vcpu);
-		vmx_fpu_deactivate(vcpu);
-		vcpu->arch.cr0 &= ~X86_CR0_TS;
-		vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0);
-		vmx_fpu_activate(vcpu);
-		skip_emulated_instruction(vcpu);
-		return 1;
-	case 1: /*mov from cr*/
-		switch (cr) {
-		case 3:
-			vcpu_load_rsp_rip(vcpu);
-			vcpu->arch.regs[reg] = vcpu->arch.cr3;
-			vcpu_put_rsp_rip(vcpu);
-			skip_emulated_instruction(vcpu);
-			return 1;
-		case 8:
-			vcpu_load_rsp_rip(vcpu);
-			vcpu->arch.regs[reg] = get_cr8(vcpu);
-			vcpu_put_rsp_rip(vcpu);
-			skip_emulated_instruction(vcpu);
-			return 1;
-		}
-		break;
-	case 3: /* lmsw */
-		lmsw(vcpu, (exit_qualification >> LMSW_SOURCE_DATA_SHIFT) & 0x0f);
-
-		skip_emulated_instruction(vcpu);
-		return 1;
-	default:
-		break;
-	}
-	kvm_run->exit_reason = 0;
-	pr_unimpl(vcpu, "unhandled control register: op %d cr %d\n",
-	       (int)(exit_qualification >> 4) & 3, cr);
-	return 0;
-}
-
-static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	unsigned long exit_qualification;
-	unsigned long val;
-	int dr, reg;
-
-	/*
-	 * FIXME: this code assumes the host is debugging the guest.
-	 *        need to deal with guest debugging itself too.
-	 */
-	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
-	dr = exit_qualification & 7;
-	reg = (exit_qualification >> 8) & 15;
-	vcpu_load_rsp_rip(vcpu);
-	if (exit_qualification & 16) {
-		/* mov from dr */
-		switch (dr) {
-		case 6:
-			val = 0xffff0ff0;
-			break;
-		case 7:
-			val = 0x400;
-			break;
-		default:
-			val = 0;
-		}
-		vcpu->arch.regs[reg] = val;
-	} else {
-		/* mov to dr */
-	}
-	vcpu_put_rsp_rip(vcpu);
-	skip_emulated_instruction(vcpu);
-	return 1;
-}
-
-static int handle_cpuid(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	kvm_emulate_cpuid(vcpu);
-	return 1;
-}
-
-static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
-	u64 data;
-
-	if (vmx_get_msr(vcpu, ecx, &data)) {
-		kvm_inject_gp(vcpu, 0);
-		return 1;
-	}
-
-	/* FIXME: handling of bits 32:63 of rax, rdx */
-	vcpu->arch.regs[VCPU_REGS_RAX] = data & -1u;
-	vcpu->arch.regs[VCPU_REGS_RDX] = (data >> 32) & -1u;
-	skip_emulated_instruction(vcpu);
-	return 1;
-}
-
-static int handle_wrmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
-	u64 data = (vcpu->arch.regs[VCPU_REGS_RAX] & -1u)
-		| ((u64)(vcpu->arch.regs[VCPU_REGS_RDX] & -1u) << 32);
-
-	if (vmx_set_msr(vcpu, ecx, data) != 0) {
-		kvm_inject_gp(vcpu, 0);
-		return 1;
-	}
-
-	skip_emulated_instruction(vcpu);
-	return 1;
-}
-
-static int handle_tpr_below_threshold(struct kvm_vcpu *vcpu,
-				      struct kvm_run *kvm_run)
-{
-	return 1;
-}
-
-static int handle_interrupt_window(struct kvm_vcpu *vcpu,
-				   struct kvm_run *kvm_run)
-{
-	u32 cpu_based_vm_exec_control;
-
-	/* clear pending irq */
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
-	cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
-	/*
-	 * If the user space waits to inject interrupts, exit as soon as
-	 * possible
-	 */
-	if (kvm_run->request_interrupt_window &&
-	    !vcpu->arch.irq_summary) {
-		kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
-		++vcpu->stat.irq_window_exits;
-		return 0;
-	}
-	return 1;
-}
-
-static int handle_halt(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	skip_emulated_instruction(vcpu);
-	return kvm_emulate_halt(vcpu);
-}
-
-static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	skip_emulated_instruction(vcpu);
-	kvm_emulate_hypercall(vcpu);
-	return 1;
-}
-
-static int handle_wbinvd(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	skip_emulated_instruction(vcpu);
-	/* TODO: Add support for VT-d/pass-through device */
-	return 1;
-}
-
-static int handle_apic_access(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	u64 exit_qualification;
-	enum emulation_result er;
-	unsigned long offset;
-
-	exit_qualification = vmcs_read64(EXIT_QUALIFICATION);
-	offset = exit_qualification & 0xffful;
-
-	er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
-
-	if (er !=  EMULATE_DONE) {
-		printk(KERN_ERR
-		       "Fail to handle apic access vmexit! Offset is 0x%lx\n",
-		       offset);
-		return -ENOTSUPP;
-	}
-	return 1;
-}
-
-/*
- * The exit handlers return 1 if the exit was handled fully and guest execution
- * may resume.  Otherwise they set the kvm_run parameter to indicate what needs
- * to be done to userspace and return 0.
- */
-static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu,
-				      struct kvm_run *kvm_run) = {
-	[EXIT_REASON_EXCEPTION_NMI]           = handle_exception,
-	[EXIT_REASON_EXTERNAL_INTERRUPT]      = handle_external_interrupt,
-	[EXIT_REASON_TRIPLE_FAULT]            = handle_triple_fault,
-	[EXIT_REASON_IO_INSTRUCTION]          = handle_io,
-	[EXIT_REASON_CR_ACCESS]               = handle_cr,
-	[EXIT_REASON_DR_ACCESS]               = handle_dr,
-	[EXIT_REASON_CPUID]                   = handle_cpuid,
-	[EXIT_REASON_MSR_READ]                = handle_rdmsr,
-	[EXIT_REASON_MSR_WRITE]               = handle_wrmsr,
-	[EXIT_REASON_PENDING_INTERRUPT]       = handle_interrupt_window,
-	[EXIT_REASON_HLT]                     = handle_halt,
-	[EXIT_REASON_VMCALL]                  = handle_vmcall,
-	[EXIT_REASON_TPR_BELOW_THRESHOLD]     = handle_tpr_below_threshold,
-	[EXIT_REASON_APIC_ACCESS]             = handle_apic_access,
-	[EXIT_REASON_WBINVD]                  = handle_wbinvd,
-};
-
-static const int kvm_vmx_max_exit_handlers =
-	ARRAY_SIZE(kvm_vmx_exit_handlers);
-
-/*
- * The guest has exited.  See if we can fix it or if we need userspace
- * assistance.
- */
-static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
-{
-	u32 exit_reason = vmcs_read32(VM_EXIT_REASON);
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 vectoring_info = vmx->idt_vectoring_info;
-
-	if (unlikely(vmx->fail)) {
-		kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
-		kvm_run->fail_entry.hardware_entry_failure_reason
-			= vmcs_read32(VM_INSTRUCTION_ERROR);
-		return 0;
-	}
-
-	if ((vectoring_info & VECTORING_INFO_VALID_MASK) &&
-				exit_reason != EXIT_REASON_EXCEPTION_NMI)
-		printk(KERN_WARNING "%s: unexpected, valid vectoring info and "
-		       "exit reason is 0x%x\n", __FUNCTION__, exit_reason);
-	if (exit_reason < kvm_vmx_max_exit_handlers
-	    && kvm_vmx_exit_handlers[exit_reason])
-		return kvm_vmx_exit_handlers[exit_reason](vcpu, kvm_run);
-	else {
-		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
-		kvm_run->hw.hardware_exit_reason = exit_reason;
-	}
-	return 0;
-}
-
-static void vmx_flush_tlb(struct kvm_vcpu *vcpu)
-{
-}
-
-static void update_tpr_threshold(struct kvm_vcpu *vcpu)
-{
-	int max_irr, tpr;
-
-	if (!vm_need_tpr_shadow(vcpu->kvm))
-		return;
-
-	if (!kvm_lapic_enabled(vcpu) ||
-	    ((max_irr = kvm_lapic_find_highest_irr(vcpu)) == -1)) {
-		vmcs_write32(TPR_THRESHOLD, 0);
-		return;
-	}
-
-	tpr = (kvm_lapic_get_cr8(vcpu) & 0x0f) << 4;
-	vmcs_write32(TPR_THRESHOLD, (max_irr > tpr) ? tpr >> 4 : max_irr >> 4);
-}
-
-static void enable_irq_window(struct kvm_vcpu *vcpu)
-{
-	u32 cpu_based_vm_exec_control;
-
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
-	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
-}
-
-static void vmx_intr_assist(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 idtv_info_field, intr_info_field;
-	int has_ext_irq, interrupt_window_open;
-	int vector;
-
-	update_tpr_threshold(vcpu);
-
-	has_ext_irq = kvm_cpu_has_interrupt(vcpu);
-	intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD);
-	idtv_info_field = vmx->idt_vectoring_info;
-	if (intr_info_field & INTR_INFO_VALID_MASK) {
-		if (idtv_info_field & INTR_INFO_VALID_MASK) {
-			/* TODO: fault when IDT_Vectoring */
-			if (printk_ratelimit())
-				printk(KERN_ERR "Fault when IDT_Vectoring\n");
-		}
-		if (has_ext_irq)
-			enable_irq_window(vcpu);
-		return;
-	}
-	if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) {
-		if ((idtv_info_field & VECTORING_INFO_TYPE_MASK)
-		    == INTR_TYPE_EXT_INTR
-		    && vcpu->arch.rmode.active) {
-			u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK;
-
-			vmx_inject_irq(vcpu, vect);
-			if (unlikely(has_ext_irq))
-				enable_irq_window(vcpu);
-			return;
-		}
-
-		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field);
-		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN,
-				vmcs_read32(VM_EXIT_INSTRUCTION_LEN));
-
-		if (unlikely(idtv_info_field & INTR_INFO_DELIEVER_CODE_MASK))
-			vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE,
-				vmcs_read32(IDT_VECTORING_ERROR_CODE));
-		if (unlikely(has_ext_irq))
-			enable_irq_window(vcpu);
-		return;
-	}
-	if (!has_ext_irq)
-		return;
-	interrupt_window_open =
-		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
-		 (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
-	if (interrupt_window_open) {
-		vector = kvm_cpu_get_interrupt(vcpu);
-		vmx_inject_irq(vcpu, vector);
-		kvm_timer_intr_post(vcpu, vector);
-	} else
-		enable_irq_window(vcpu);
-}
-
-/*
- * Failure to inject an interrupt should give us the information
- * in IDT_VECTORING_INFO_FIELD.  However, if the failure occurs
- * when fetching the interrupt redirection bitmap in the real-mode
- * tss, this doesn't happen.  So we do it ourselves.
- */
-static void fixup_rmode_irq(struct vcpu_vmx *vmx)
-{
-	vmx->rmode.irq.pending = 0;
-	if (vmcs_readl(GUEST_RIP) + 1 != vmx->rmode.irq.rip)
-		return;
-	vmcs_writel(GUEST_RIP, vmx->rmode.irq.rip);
-	if (vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK) {
-		vmx->idt_vectoring_info &= ~VECTORING_INFO_TYPE_MASK;
-		vmx->idt_vectoring_info |= INTR_TYPE_EXT_INTR;
-		return;
-	}
-	vmx->idt_vectoring_info =
-		VECTORING_INFO_VALID_MASK
-		| INTR_TYPE_EXT_INTR
-		| vmx->rmode.irq.vector;
-}
-
-static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 intr_info;
-
-	/*
-	 * Loading guest fpu may have cleared host cr0.ts
-	 */
-	vmcs_writel(HOST_CR0, read_cr0());
-
-	asm(
-		/* Store host registers */
-#ifdef CONFIG_X86_64
-		"push %%rdx; push %%rbp;"
-		"push %%rcx \n\t"
-#else
-		"push %%edx; push %%ebp;"
-		"push %%ecx \n\t"
-#endif
-		ASM_VMX_VMWRITE_RSP_RDX "\n\t"
-		/* Check if vmlaunch of vmresume is needed */
-		"cmpl $0, %c[launched](%0) \n\t"
-		/* Load guest registers.  Don't clobber flags. */
-#ifdef CONFIG_X86_64
-		"mov %c[cr2](%0), %%rax \n\t"
-		"mov %%rax, %%cr2 \n\t"
-		"mov %c[rax](%0), %%rax \n\t"
-		"mov %c[rbx](%0), %%rbx \n\t"
-		"mov %c[rdx](%0), %%rdx \n\t"
-		"mov %c[rsi](%0), %%rsi \n\t"
-		"mov %c[rdi](%0), %%rdi \n\t"
-		"mov %c[rbp](%0), %%rbp \n\t"
-		"mov %c[r8](%0),  %%r8  \n\t"
-		"mov %c[r9](%0),  %%r9  \n\t"
-		"mov %c[r10](%0), %%r10 \n\t"
-		"mov %c[r11](%0), %%r11 \n\t"
-		"mov %c[r12](%0), %%r12 \n\t"
-		"mov %c[r13](%0), %%r13 \n\t"
-		"mov %c[r14](%0), %%r14 \n\t"
-		"mov %c[r15](%0), %%r15 \n\t"
-		"mov %c[rcx](%0), %%rcx \n\t" /* kills %0 (rcx) */
-#else
-		"mov %c[cr2](%0), %%eax \n\t"
-		"mov %%eax,   %%cr2 \n\t"
-		"mov %c[rax](%0), %%eax \n\t"
-		"mov %c[rbx](%0), %%ebx \n\t"
-		"mov %c[rdx](%0), %%edx \n\t"
-		"mov %c[rsi](%0), %%esi \n\t"
-		"mov %c[rdi](%0), %%edi \n\t"
-		"mov %c[rbp](%0), %%ebp \n\t"
-		"mov %c[rcx](%0), %%ecx \n\t" /* kills %0 (ecx) */
-#endif
-		/* Enter guest mode */
-		"jne .Llaunched \n\t"
-		ASM_VMX_VMLAUNCH "\n\t"
-		"jmp .Lkvm_vmx_return \n\t"
-		".Llaunched: " ASM_VMX_VMRESUME "\n\t"
-		".Lkvm_vmx_return: "
-		/* Save guest registers, load host registers, keep flags */
-#ifdef CONFIG_X86_64
-		"xchg %0,     (%%rsp) \n\t"
-		"mov %%rax, %c[rax](%0) \n\t"
-		"mov %%rbx, %c[rbx](%0) \n\t"
-		"pushq (%%rsp); popq %c[rcx](%0) \n\t"
-		"mov %%rdx, %c[rdx](%0) \n\t"
-		"mov %%rsi, %c[rsi](%0) \n\t"
-		"mov %%rdi, %c[rdi](%0) \n\t"
-		"mov %%rbp, %c[rbp](%0) \n\t"
-		"mov %%r8,  %c[r8](%0) \n\t"
-		"mov %%r9,  %c[r9](%0) \n\t"
-		"mov %%r10, %c[r10](%0) \n\t"
-		"mov %%r11, %c[r11](%0) \n\t"
-		"mov %%r12, %c[r12](%0) \n\t"
-		"mov %%r13, %c[r13](%0) \n\t"
-		"mov %%r14, %c[r14](%0) \n\t"
-		"mov %%r15, %c[r15](%0) \n\t"
-		"mov %%cr2, %%rax   \n\t"
-		"mov %%rax, %c[cr2](%0) \n\t"
-
-		"pop  %%rbp; pop  %%rbp; pop  %%rdx \n\t"
-#else
-		"xchg %0, (%%esp) \n\t"
-		"mov %%eax, %c[rax](%0) \n\t"
-		"mov %%ebx, %c[rbx](%0) \n\t"
-		"pushl (%%esp); popl %c[rcx](%0) \n\t"
-		"mov %%edx, %c[rdx](%0) \n\t"
-		"mov %%esi, %c[rsi](%0) \n\t"
-		"mov %%edi, %c[rdi](%0) \n\t"
-		"mov %%ebp, %c[rbp](%0) \n\t"
-		"mov %%cr2, %%eax  \n\t"
-		"mov %%eax, %c[cr2](%0) \n\t"
-
-		"pop %%ebp; pop %%ebp; pop %%edx \n\t"
-#endif
-		"setbe %c[fail](%0) \n\t"
-	      : : "c"(vmx), "d"((unsigned long)HOST_RSP),
-		[launched]"i"(offsetof(struct vcpu_vmx, launched)),
-		[fail]"i"(offsetof(struct vcpu_vmx, fail)),
-		[rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])),
-		[rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])),
-		[rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])),
-		[rdx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDX])),
-		[rsi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RSI])),
-		[rdi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDI])),
-		[rbp]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBP])),
-#ifdef CONFIG_X86_64
-		[r8]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R8])),
-		[r9]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R9])),
-		[r10]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R10])),
-		[r11]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R11])),
-		[r12]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R12])),
-		[r13]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R13])),
-		[r14]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R14])),
-		[r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])),
-#endif
-		[cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2))
-	      : "cc", "memory"
-#ifdef CONFIG_X86_64
-		, "rbx", "rdi", "rsi"
-		, "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15"
-#else
-		, "ebx", "edi", "rsi"
-#endif
-	      );
-
-	vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);
-	if (vmx->rmode.irq.pending)
-		fixup_rmode_irq(vmx);
-
-	vcpu->arch.interrupt_window_open =
-		(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0;
-
-	asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS));
-	vmx->launched = 1;
-
-	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
-
-	/* We need to handle NMIs before interrupts are enabled */
-	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
-		asm("int $2");
-}
-
-static void vmx_free_vmcs(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-
-	if (vmx->vmcs) {
-		on_each_cpu(__vcpu_clear, vmx, 0, 1);
-		free_vmcs(vmx->vmcs);
-		vmx->vmcs = NULL;
-	}
-}
-
-static void vmx_free_vcpu(struct kvm_vcpu *vcpu)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-
-	vmx_free_vmcs(vcpu);
-	kfree(vmx->host_msrs);
-	kfree(vmx->guest_msrs);
-	kvm_vcpu_uninit(vcpu);
-	kmem_cache_free(kvm_vcpu_cache, vmx);
-}
-
-static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id)
-{
-	int err;
-	struct vcpu_vmx *vmx = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
-	int cpu;
-
-	if (!vmx)
-		return ERR_PTR(-ENOMEM);
-
-	err = kvm_vcpu_init(&vmx->vcpu, kvm, id);
-	if (err)
-		goto free_vcpu;
-
-	vmx->guest_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL);
-	if (!vmx->guest_msrs) {
-		err = -ENOMEM;
-		goto uninit_vcpu;
-	}
-
-	vmx->host_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL);
-	if (!vmx->host_msrs)
-		goto free_guest_msrs;
-
-	vmx->vmcs = alloc_vmcs();
-	if (!vmx->vmcs)
-		goto free_msrs;
-
-	vmcs_clear(vmx->vmcs);
-
-	cpu = get_cpu();
-	vmx_vcpu_load(&vmx->vcpu, cpu);
-	err = vmx_vcpu_setup(vmx);
-	vmx_vcpu_put(&vmx->vcpu);
-	put_cpu();
-	if (err)
-		goto free_vmcs;
-
-	return &vmx->vcpu;
-
-free_vmcs:
-	free_vmcs(vmx->vmcs);
-free_msrs:
-	kfree(vmx->host_msrs);
-free_guest_msrs:
-	kfree(vmx->guest_msrs);
-uninit_vcpu:
-	kvm_vcpu_uninit(&vmx->vcpu);
-free_vcpu:
-	kmem_cache_free(kvm_vcpu_cache, vmx);
-	return ERR_PTR(err);
-}
-
-static void __init vmx_check_processor_compat(void *rtn)
-{
-	struct vmcs_config vmcs_conf;
-
-	*(int *)rtn = 0;
-	if (setup_vmcs_config(&vmcs_conf) < 0)
-		*(int *)rtn = -EIO;
-	if (memcmp(&vmcs_config, &vmcs_conf, sizeof(struct vmcs_config)) != 0) {
-		printk(KERN_ERR "kvm: CPU %d feature inconsistency!\n",
-				smp_processor_id());
-		*(int *)rtn = -EIO;
-	}
-}
-
-static struct kvm_x86_ops vmx_x86_ops = {
-	.cpu_has_kvm_support = cpu_has_kvm_support,
-	.disabled_by_bios = vmx_disabled_by_bios,
-	.hardware_setup = hardware_setup,
-	.hardware_unsetup = hardware_unsetup,
-	.check_processor_compatibility = vmx_check_processor_compat,
-	.hardware_enable = hardware_enable,
-	.hardware_disable = hardware_disable,
-
-	.vcpu_create = vmx_create_vcpu,
-	.vcpu_free = vmx_free_vcpu,
-	.vcpu_reset = vmx_vcpu_reset,
-
-	.prepare_guest_switch = vmx_save_host_state,
-	.vcpu_load = vmx_vcpu_load,
-	.vcpu_put = vmx_vcpu_put,
-	.vcpu_decache = vmx_vcpu_decache,
-
-	.set_guest_debug = set_guest_debug,
-	.guest_debug_pre = kvm_guest_debug_pre,
-	.get_msr = vmx_get_msr,
-	.set_msr = vmx_set_msr,
-	.get_segment_base = vmx_get_segment_base,
-	.get_segment = vmx_get_segment,
-	.set_segment = vmx_set_segment,
-	.get_cs_db_l_bits = vmx_get_cs_db_l_bits,
-	.decache_cr4_guest_bits = vmx_decache_cr4_guest_bits,
-	.set_cr0 = vmx_set_cr0,
-	.set_cr3 = vmx_set_cr3,
-	.set_cr4 = vmx_set_cr4,
-#ifdef CONFIG_X86_64
-	.set_efer = vmx_set_efer,
-#endif
-	.get_idt = vmx_get_idt,
-	.set_idt = vmx_set_idt,
-	.get_gdt = vmx_get_gdt,
-	.set_gdt = vmx_set_gdt,
-	.cache_regs = vcpu_load_rsp_rip,
-	.decache_regs = vcpu_put_rsp_rip,
-	.get_rflags = vmx_get_rflags,
-	.set_rflags = vmx_set_rflags,
-
-	.tlb_flush = vmx_flush_tlb,
-
-	.run = vmx_vcpu_run,
-	.handle_exit = kvm_handle_exit,
-	.skip_emulated_instruction = skip_emulated_instruction,
-	.patch_hypercall = vmx_patch_hypercall,
-	.get_irq = vmx_get_irq,
-	.set_irq = vmx_inject_irq,
-	.queue_exception = vmx_queue_exception,
-	.exception_injected = vmx_exception_injected,
-	.inject_pending_irq = vmx_intr_assist,
-	.inject_pending_vectors = do_interrupt_requests,
-
-	.set_tss_addr = vmx_set_tss_addr,
-};
-
-static int __init vmx_init(void)
-{
-	void *iova;
-	int r;
-
-	vmx_io_bitmap_a = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
-	if (!vmx_io_bitmap_a)
-		return -ENOMEM;
-
-	vmx_io_bitmap_b = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
-	if (!vmx_io_bitmap_b) {
-		r = -ENOMEM;
-		goto out;
-	}
-
-	/*
-	 * Allow direct access to the PC debug port (it is often used for I/O
-	 * delays, but the vmexits simply slow things down).
-	 */
-	iova = kmap(vmx_io_bitmap_a);
-	memset(iova, 0xff, PAGE_SIZE);
-	clear_bit(0x80, iova);
-	kunmap(vmx_io_bitmap_a);
-
-	iova = kmap(vmx_io_bitmap_b);
-	memset(iova, 0xff, PAGE_SIZE);
-	kunmap(vmx_io_bitmap_b);
-
-	r = kvm_init(&vmx_x86_ops, sizeof(struct vcpu_vmx), THIS_MODULE);
-	if (r)
-		goto out1;
-
-	if (bypass_guest_pf)
-		kvm_mmu_set_nonpresent_ptes(~0xffeull, 0ull);
-
-	return 0;
-
-out1:
-	__free_page(vmx_io_bitmap_b);
-out:
-	__free_page(vmx_io_bitmap_a);
-	return r;
-}
-
-static void __exit vmx_exit(void)
-{
-	__free_page(vmx_io_bitmap_b);
-	__free_page(vmx_io_bitmap_a);
-
-	kvm_exit();
-}
-
-module_init(vmx_init)
-module_exit(vmx_exit)
diff --git a/drivers/kvm/vmx.h b/drivers/kvm/vmx.h
deleted file mode 100644
index d52ae8d7303d..000000000000
--- a/drivers/kvm/vmx.h
+++ /dev/null
@@ -1,324 +0,0 @@
-#ifndef VMX_H
-#define VMX_H
-
-/*
- * vmx.h: VMX Architecture related definitions
- * Copyright (c) 2004, Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
- * Place - Suite 330, Boston, MA 02111-1307 USA.
- *
- * A few random additions are:
- * Copyright (C) 2006 Qumranet
- *    Avi Kivity <avi@qumranet.com>
- *    Yaniv Kamay <yaniv@qumranet.com>
- *
- */
-
-/*
- * Definitions of Primary Processor-Based VM-Execution Controls.
- */
-#define CPU_BASED_VIRTUAL_INTR_PENDING          0x00000004
-#define CPU_BASED_USE_TSC_OFFSETING             0x00000008
-#define CPU_BASED_HLT_EXITING                   0x00000080
-#define CPU_BASED_INVLPG_EXITING                0x00000200
-#define CPU_BASED_MWAIT_EXITING                 0x00000400
-#define CPU_BASED_RDPMC_EXITING                 0x00000800
-#define CPU_BASED_RDTSC_EXITING                 0x00001000
-#define CPU_BASED_CR8_LOAD_EXITING              0x00080000
-#define CPU_BASED_CR8_STORE_EXITING             0x00100000
-#define CPU_BASED_TPR_SHADOW                    0x00200000
-#define CPU_BASED_MOV_DR_EXITING                0x00800000
-#define CPU_BASED_UNCOND_IO_EXITING             0x01000000
-#define CPU_BASED_USE_IO_BITMAPS                0x02000000
-#define CPU_BASED_USE_MSR_BITMAPS               0x10000000
-#define CPU_BASED_MONITOR_EXITING               0x20000000
-#define CPU_BASED_PAUSE_EXITING                 0x40000000
-#define CPU_BASED_ACTIVATE_SECONDARY_CONTROLS   0x80000000
-/*
- * Definitions of Secondary Processor-Based VM-Execution Controls.
- */
-#define SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES 0x00000001
-#define SECONDARY_EXEC_WBINVD_EXITING		0x00000040
-
-
-#define PIN_BASED_EXT_INTR_MASK                 0x00000001
-#define PIN_BASED_NMI_EXITING                   0x00000008
-#define PIN_BASED_VIRTUAL_NMIS                  0x00000020
-
-#define VM_EXIT_HOST_ADDR_SPACE_SIZE            0x00000200
-#define VM_EXIT_ACK_INTR_ON_EXIT                0x00008000
-
-#define VM_ENTRY_IA32E_MODE                     0x00000200
-#define VM_ENTRY_SMM                            0x00000400
-#define VM_ENTRY_DEACT_DUAL_MONITOR             0x00000800
-
-/* VMCS Encodings */
-enum vmcs_field {
-	GUEST_ES_SELECTOR               = 0x00000800,
-	GUEST_CS_SELECTOR               = 0x00000802,
-	GUEST_SS_SELECTOR               = 0x00000804,
-	GUEST_DS_SELECTOR               = 0x00000806,
-	GUEST_FS_SELECTOR               = 0x00000808,
-	GUEST_GS_SELECTOR               = 0x0000080a,
-	GUEST_LDTR_SELECTOR             = 0x0000080c,
-	GUEST_TR_SELECTOR               = 0x0000080e,
-	HOST_ES_SELECTOR                = 0x00000c00,
-	HOST_CS_SELECTOR                = 0x00000c02,
-	HOST_SS_SELECTOR                = 0x00000c04,
-	HOST_DS_SELECTOR                = 0x00000c06,
-	HOST_FS_SELECTOR                = 0x00000c08,
-	HOST_GS_SELECTOR                = 0x00000c0a,
-	HOST_TR_SELECTOR                = 0x00000c0c,
-	IO_BITMAP_A                     = 0x00002000,
-	IO_BITMAP_A_HIGH                = 0x00002001,
-	IO_BITMAP_B                     = 0x00002002,
-	IO_BITMAP_B_HIGH                = 0x00002003,
-	MSR_BITMAP                      = 0x00002004,
-	MSR_BITMAP_HIGH                 = 0x00002005,
-	VM_EXIT_MSR_STORE_ADDR          = 0x00002006,
-	VM_EXIT_MSR_STORE_ADDR_HIGH     = 0x00002007,
-	VM_EXIT_MSR_LOAD_ADDR           = 0x00002008,
-	VM_EXIT_MSR_LOAD_ADDR_HIGH      = 0x00002009,
-	VM_ENTRY_MSR_LOAD_ADDR          = 0x0000200a,
-	VM_ENTRY_MSR_LOAD_ADDR_HIGH     = 0x0000200b,
-	TSC_OFFSET                      = 0x00002010,
-	TSC_OFFSET_HIGH                 = 0x00002011,
-	VIRTUAL_APIC_PAGE_ADDR          = 0x00002012,
-	VIRTUAL_APIC_PAGE_ADDR_HIGH     = 0x00002013,
-	APIC_ACCESS_ADDR		= 0x00002014,
-	APIC_ACCESS_ADDR_HIGH		= 0x00002015,
-	VMCS_LINK_POINTER               = 0x00002800,
-	VMCS_LINK_POINTER_HIGH          = 0x00002801,
-	GUEST_IA32_DEBUGCTL             = 0x00002802,
-	GUEST_IA32_DEBUGCTL_HIGH        = 0x00002803,
-	PIN_BASED_VM_EXEC_CONTROL       = 0x00004000,
-	CPU_BASED_VM_EXEC_CONTROL       = 0x00004002,
-	EXCEPTION_BITMAP                = 0x00004004,
-	PAGE_FAULT_ERROR_CODE_MASK      = 0x00004006,
-	PAGE_FAULT_ERROR_CODE_MATCH     = 0x00004008,
-	CR3_TARGET_COUNT                = 0x0000400a,
-	VM_EXIT_CONTROLS                = 0x0000400c,
-	VM_EXIT_MSR_STORE_COUNT         = 0x0000400e,
-	VM_EXIT_MSR_LOAD_COUNT          = 0x00004010,
-	VM_ENTRY_CONTROLS               = 0x00004012,
-	VM_ENTRY_MSR_LOAD_COUNT         = 0x00004014,
-	VM_ENTRY_INTR_INFO_FIELD        = 0x00004016,
-	VM_ENTRY_EXCEPTION_ERROR_CODE   = 0x00004018,
-	VM_ENTRY_INSTRUCTION_LEN        = 0x0000401a,
-	TPR_THRESHOLD                   = 0x0000401c,
-	SECONDARY_VM_EXEC_CONTROL       = 0x0000401e,
-	VM_INSTRUCTION_ERROR            = 0x00004400,
-	VM_EXIT_REASON                  = 0x00004402,
-	VM_EXIT_INTR_INFO               = 0x00004404,
-	VM_EXIT_INTR_ERROR_CODE         = 0x00004406,
-	IDT_VECTORING_INFO_FIELD        = 0x00004408,
-	IDT_VECTORING_ERROR_CODE        = 0x0000440a,
-	VM_EXIT_INSTRUCTION_LEN         = 0x0000440c,
-	VMX_INSTRUCTION_INFO            = 0x0000440e,
-	GUEST_ES_LIMIT                  = 0x00004800,
-	GUEST_CS_LIMIT                  = 0x00004802,
-	GUEST_SS_LIMIT                  = 0x00004804,
-	GUEST_DS_LIMIT                  = 0x00004806,
-	GUEST_FS_LIMIT                  = 0x00004808,
-	GUEST_GS_LIMIT                  = 0x0000480a,
-	GUEST_LDTR_LIMIT                = 0x0000480c,
-	GUEST_TR_LIMIT                  = 0x0000480e,
-	GUEST_GDTR_LIMIT                = 0x00004810,
-	GUEST_IDTR_LIMIT                = 0x00004812,
-	GUEST_ES_AR_BYTES               = 0x00004814,
-	GUEST_CS_AR_BYTES               = 0x00004816,
-	GUEST_SS_AR_BYTES               = 0x00004818,
-	GUEST_DS_AR_BYTES               = 0x0000481a,
-	GUEST_FS_AR_BYTES               = 0x0000481c,
-	GUEST_GS_AR_BYTES               = 0x0000481e,
-	GUEST_LDTR_AR_BYTES             = 0x00004820,
-	GUEST_TR_AR_BYTES               = 0x00004822,
-	GUEST_INTERRUPTIBILITY_INFO     = 0x00004824,
-	GUEST_ACTIVITY_STATE            = 0X00004826,
-	GUEST_SYSENTER_CS               = 0x0000482A,
-	HOST_IA32_SYSENTER_CS           = 0x00004c00,
-	CR0_GUEST_HOST_MASK             = 0x00006000,
-	CR4_GUEST_HOST_MASK             = 0x00006002,
-	CR0_READ_SHADOW                 = 0x00006004,
-	CR4_READ_SHADOW                 = 0x00006006,
-	CR3_TARGET_VALUE0               = 0x00006008,
-	CR3_TARGET_VALUE1               = 0x0000600a,
-	CR3_TARGET_VALUE2               = 0x0000600c,
-	CR3_TARGET_VALUE3               = 0x0000600e,
-	EXIT_QUALIFICATION              = 0x00006400,
-	GUEST_LINEAR_ADDRESS            = 0x0000640a,
-	GUEST_CR0                       = 0x00006800,
-	GUEST_CR3                       = 0x00006802,
-	GUEST_CR4                       = 0x00006804,
-	GUEST_ES_BASE                   = 0x00006806,
-	GUEST_CS_BASE                   = 0x00006808,
-	GUEST_SS_BASE                   = 0x0000680a,
-	GUEST_DS_BASE                   = 0x0000680c,
-	GUEST_FS_BASE                   = 0x0000680e,
-	GUEST_GS_BASE                   = 0x00006810,
-	GUEST_LDTR_BASE                 = 0x00006812,
-	GUEST_TR_BASE                   = 0x00006814,
-	GUEST_GDTR_BASE                 = 0x00006816,
-	GUEST_IDTR_BASE                 = 0x00006818,
-	GUEST_DR7                       = 0x0000681a,
-	GUEST_RSP                       = 0x0000681c,
-	GUEST_RIP                       = 0x0000681e,
-	GUEST_RFLAGS                    = 0x00006820,
-	GUEST_PENDING_DBG_EXCEPTIONS    = 0x00006822,
-	GUEST_SYSENTER_ESP              = 0x00006824,
-	GUEST_SYSENTER_EIP              = 0x00006826,
-	HOST_CR0                        = 0x00006c00,
-	HOST_CR3                        = 0x00006c02,
-	HOST_CR4                        = 0x00006c04,
-	HOST_FS_BASE                    = 0x00006c06,
-	HOST_GS_BASE                    = 0x00006c08,
-	HOST_TR_BASE                    = 0x00006c0a,
-	HOST_GDTR_BASE                  = 0x00006c0c,
-	HOST_IDTR_BASE                  = 0x00006c0e,
-	HOST_IA32_SYSENTER_ESP          = 0x00006c10,
-	HOST_IA32_SYSENTER_EIP          = 0x00006c12,
-	HOST_RSP                        = 0x00006c14,
-	HOST_RIP                        = 0x00006c16,
-};
-
-#define VMX_EXIT_REASONS_FAILED_VMENTRY         0x80000000
-
-#define EXIT_REASON_EXCEPTION_NMI       0
-#define EXIT_REASON_EXTERNAL_INTERRUPT  1
-#define EXIT_REASON_TRIPLE_FAULT        2
-
-#define EXIT_REASON_PENDING_INTERRUPT   7
-
-#define EXIT_REASON_TASK_SWITCH         9
-#define EXIT_REASON_CPUID               10
-#define EXIT_REASON_HLT                 12
-#define EXIT_REASON_INVLPG              14
-#define EXIT_REASON_RDPMC               15
-#define EXIT_REASON_RDTSC               16
-#define EXIT_REASON_VMCALL              18
-#define EXIT_REASON_VMCLEAR             19
-#define EXIT_REASON_VMLAUNCH            20
-#define EXIT_REASON_VMPTRLD             21
-#define EXIT_REASON_VMPTRST             22
-#define EXIT_REASON_VMREAD              23
-#define EXIT_REASON_VMRESUME            24
-#define EXIT_REASON_VMWRITE             25
-#define EXIT_REASON_VMOFF               26
-#define EXIT_REASON_VMON                27
-#define EXIT_REASON_CR_ACCESS           28
-#define EXIT_REASON_DR_ACCESS           29
-#define EXIT_REASON_IO_INSTRUCTION      30
-#define EXIT_REASON_MSR_READ            31
-#define EXIT_REASON_MSR_WRITE           32
-#define EXIT_REASON_MWAIT_INSTRUCTION   36
-#define EXIT_REASON_TPR_BELOW_THRESHOLD 43
-#define EXIT_REASON_APIC_ACCESS         44
-#define EXIT_REASON_WBINVD		54
-
-/*
- * Interruption-information format
- */
-#define INTR_INFO_VECTOR_MASK           0xff            /* 7:0 */
-#define INTR_INFO_INTR_TYPE_MASK        0x700           /* 10:8 */
-#define INTR_INFO_DELIEVER_CODE_MASK    0x800           /* 11 */
-#define INTR_INFO_VALID_MASK            0x80000000      /* 31 */
-
-#define VECTORING_INFO_VECTOR_MASK           	INTR_INFO_VECTOR_MASK
-#define VECTORING_INFO_TYPE_MASK        	INTR_INFO_INTR_TYPE_MASK
-#define VECTORING_INFO_DELIEVER_CODE_MASK    	INTR_INFO_DELIEVER_CODE_MASK
-#define VECTORING_INFO_VALID_MASK       	INTR_INFO_VALID_MASK
-
-#define INTR_TYPE_EXT_INTR              (0 << 8) /* external interrupt */
-#define INTR_TYPE_EXCEPTION             (3 << 8) /* processor exception */
-#define INTR_TYPE_SOFT_INTR             (4 << 8) /* software interrupt */
-
-/*
- * Exit Qualifications for MOV for Control Register Access
- */
-#define CONTROL_REG_ACCESS_NUM          0x7     /* 2:0, number of control reg.*/
-#define CONTROL_REG_ACCESS_TYPE         0x30    /* 5:4, access type */
-#define CONTROL_REG_ACCESS_REG          0xf00   /* 10:8, general purpose reg. */
-#define LMSW_SOURCE_DATA_SHIFT 16
-#define LMSW_SOURCE_DATA  (0xFFFF << LMSW_SOURCE_DATA_SHIFT) /* 16:31 lmsw source */
-#define REG_EAX                         (0 << 8)
-#define REG_ECX                         (1 << 8)
-#define REG_EDX                         (2 << 8)
-#define REG_EBX                         (3 << 8)
-#define REG_ESP                         (4 << 8)
-#define REG_EBP                         (5 << 8)
-#define REG_ESI                         (6 << 8)
-#define REG_EDI                         (7 << 8)
-#define REG_R8                         (8 << 8)
-#define REG_R9                         (9 << 8)
-#define REG_R10                        (10 << 8)
-#define REG_R11                        (11 << 8)
-#define REG_R12                        (12 << 8)
-#define REG_R13                        (13 << 8)
-#define REG_R14                        (14 << 8)
-#define REG_R15                        (15 << 8)
-
-/*
- * Exit Qualifications for MOV for Debug Register Access
- */
-#define DEBUG_REG_ACCESS_NUM            0x7     /* 2:0, number of debug reg. */
-#define DEBUG_REG_ACCESS_TYPE           0x10    /* 4, direction of access */
-#define TYPE_MOV_TO_DR                  (0 << 4)
-#define TYPE_MOV_FROM_DR                (1 << 4)
-#define DEBUG_REG_ACCESS_REG            0xf00   /* 11:8, general purpose reg. */
-
-
-/* segment AR */
-#define SEGMENT_AR_L_MASK (1 << 13)
-
-#define AR_TYPE_ACCESSES_MASK 1
-#define AR_TYPE_READABLE_MASK (1 << 1)
-#define AR_TYPE_WRITEABLE_MASK (1 << 2)
-#define AR_TYPE_CODE_MASK (1 << 3)
-#define AR_TYPE_MASK 0x0f
-#define AR_TYPE_BUSY_64_TSS 11
-#define AR_TYPE_BUSY_32_TSS 11
-#define AR_TYPE_BUSY_16_TSS 3
-#define AR_TYPE_LDT 2
-
-#define AR_UNUSABLE_MASK (1 << 16)
-#define AR_S_MASK (1 << 4)
-#define AR_P_MASK (1 << 7)
-#define AR_L_MASK (1 << 13)
-#define AR_DB_MASK (1 << 14)
-#define AR_G_MASK (1 << 15)
-#define AR_DPL_SHIFT 5
-#define AR_DPL(ar) (((ar) >> AR_DPL_SHIFT) & 3)
-
-#define AR_RESERVD_MASK 0xfffe0f00
-
-#define MSR_IA32_VMX_BASIC                      0x480
-#define MSR_IA32_VMX_PINBASED_CTLS              0x481
-#define MSR_IA32_VMX_PROCBASED_CTLS             0x482
-#define MSR_IA32_VMX_EXIT_CTLS                  0x483
-#define MSR_IA32_VMX_ENTRY_CTLS                 0x484
-#define MSR_IA32_VMX_MISC                       0x485
-#define MSR_IA32_VMX_CR0_FIXED0                 0x486
-#define MSR_IA32_VMX_CR0_FIXED1                 0x487
-#define MSR_IA32_VMX_CR4_FIXED0                 0x488
-#define MSR_IA32_VMX_CR4_FIXED1                 0x489
-#define MSR_IA32_VMX_VMCS_ENUM                  0x48a
-#define MSR_IA32_VMX_PROCBASED_CTLS2            0x48b
-
-#define MSR_IA32_FEATURE_CONTROL                0x3a
-#define MSR_IA32_FEATURE_CONTROL_LOCKED         0x1
-#define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED  0x4
-
-#define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT	9
-
-#endif
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
deleted file mode 100644
index b37c0093d728..000000000000
--- a/drivers/kvm/x86.c
+++ /dev/null
@@ -1,3148 +0,0 @@
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * derived from drivers/kvm/kvm_main.c
- *
- * Copyright (C) 2006 Qumranet, Inc.
- *
- * Authors:
- *   Avi Kivity   <avi@qumranet.com>
- *   Yaniv Kamay  <yaniv@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-
-#include "kvm.h"
-#include "x86.h"
-#include "x86_emulate.h"
-#include "segment_descriptor.h"
-#include "irq.h"
-#include "mmu.h"
-
-#include <linux/kvm.h>
-#include <linux/fs.h>
-#include <linux/vmalloc.h>
-#include <linux/module.h>
-#include <linux/mman.h>
-#include <linux/highmem.h>
-
-#include <asm/uaccess.h>
-#include <asm/msr.h>
-
-#define MAX_IO_MSRS 256
-#define CR0_RESERVED_BITS						\
-	(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
-			  | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \
-			  | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG))
-#define CR4_RESERVED_BITS						\
-	(~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\
-			  | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE	\
-			  | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR	\
-			  | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE))
-
-#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR)
-#define EFER_RESERVED_BITS 0xfffffffffffff2fe
-
-#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
-#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
-
-struct kvm_x86_ops *kvm_x86_ops;
-
-struct kvm_stats_debugfs_item debugfs_entries[] = {
-	{ "pf_fixed", VCPU_STAT(pf_fixed) },
-	{ "pf_guest", VCPU_STAT(pf_guest) },
-	{ "tlb_flush", VCPU_STAT(tlb_flush) },
-	{ "invlpg", VCPU_STAT(invlpg) },
-	{ "exits", VCPU_STAT(exits) },
-	{ "io_exits", VCPU_STAT(io_exits) },
-	{ "mmio_exits", VCPU_STAT(mmio_exits) },
-	{ "signal_exits", VCPU_STAT(signal_exits) },
-	{ "irq_window", VCPU_STAT(irq_window_exits) },
-	{ "halt_exits", VCPU_STAT(halt_exits) },
-	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
-	{ "request_irq", VCPU_STAT(request_irq_exits) },
-	{ "irq_exits", VCPU_STAT(irq_exits) },
-	{ "host_state_reload", VCPU_STAT(host_state_reload) },
-	{ "efer_reload", VCPU_STAT(efer_reload) },
-	{ "fpu_reload", VCPU_STAT(fpu_reload) },
-	{ "insn_emulation", VCPU_STAT(insn_emulation) },
-	{ "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) },
-	{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) },
-	{ "mmu_pte_write", VM_STAT(mmu_pte_write) },
-	{ "mmu_pte_updated", VM_STAT(mmu_pte_updated) },
-	{ "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) },
-	{ "mmu_flooded", VM_STAT(mmu_flooded) },
-	{ "mmu_recycled", VM_STAT(mmu_recycled) },
-	{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) },
-	{ NULL }
-};
-
-
-unsigned long segment_base(u16 selector)
-{
-	struct descriptor_table gdt;
-	struct segment_descriptor *d;
-	unsigned long table_base;
-	unsigned long v;
-
-	if (selector == 0)
-		return 0;
-
-	asm("sgdt %0" : "=m"(gdt));
-	table_base = gdt.base;
-
-	if (selector & 4) {           /* from ldt */
-		u16 ldt_selector;
-
-		asm("sldt %0" : "=g"(ldt_selector));
-		table_base = segment_base(ldt_selector);
-	}
-	d = (struct segment_descriptor *)(table_base + (selector & ~7));
-	v = d->base_low | ((unsigned long)d->base_mid << 16) |
-		((unsigned long)d->base_high << 24);
-#ifdef CONFIG_X86_64
-	if (d->system == 0 && (d->type == 2 || d->type == 9 || d->type == 11))
-		v |= ((unsigned long) \
-		      ((struct segment_descriptor_64 *)d)->base_higher) << 32;
-#endif
-	return v;
-}
-EXPORT_SYMBOL_GPL(segment_base);
-
-u64 kvm_get_apic_base(struct kvm_vcpu *vcpu)
-{
-	if (irqchip_in_kernel(vcpu->kvm))
-		return vcpu->arch.apic_base;
-	else
-		return vcpu->arch.apic_base;
-}
-EXPORT_SYMBOL_GPL(kvm_get_apic_base);
-
-void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data)
-{
-	/* TODO: reserve bits check */
-	if (irqchip_in_kernel(vcpu->kvm))
-		kvm_lapic_set_base(vcpu, data);
-	else
-		vcpu->arch.apic_base = data;
-}
-EXPORT_SYMBOL_GPL(kvm_set_apic_base);
-
-void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr)
-{
-	WARN_ON(vcpu->arch.exception.pending);
-	vcpu->arch.exception.pending = true;
-	vcpu->arch.exception.has_error_code = false;
-	vcpu->arch.exception.nr = nr;
-}
-EXPORT_SYMBOL_GPL(kvm_queue_exception);
-
-void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr,
-			   u32 error_code)
-{
-	++vcpu->stat.pf_guest;
-	if (vcpu->arch.exception.pending && vcpu->arch.exception.nr == PF_VECTOR) {
-		printk(KERN_DEBUG "kvm: inject_page_fault:"
-		       " double fault 0x%lx\n", addr);
-		vcpu->arch.exception.nr = DF_VECTOR;
-		vcpu->arch.exception.error_code = 0;
-		return;
-	}
-	vcpu->arch.cr2 = addr;
-	kvm_queue_exception_e(vcpu, PF_VECTOR, error_code);
-}
-
-void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code)
-{
-	WARN_ON(vcpu->arch.exception.pending);
-	vcpu->arch.exception.pending = true;
-	vcpu->arch.exception.has_error_code = true;
-	vcpu->arch.exception.nr = nr;
-	vcpu->arch.exception.error_code = error_code;
-}
-EXPORT_SYMBOL_GPL(kvm_queue_exception_e);
-
-static void __queue_exception(struct kvm_vcpu *vcpu)
-{
-	kvm_x86_ops->queue_exception(vcpu, vcpu->arch.exception.nr,
-				     vcpu->arch.exception.has_error_code,
-				     vcpu->arch.exception.error_code);
-}
-
-/*
- * Load the pae pdptrs.  Return true is they are all valid.
- */
-int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
-{
-	gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT;
-	unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2;
-	int i;
-	int ret;
-	u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
-
-	mutex_lock(&vcpu->kvm->lock);
-	ret = kvm_read_guest_page(vcpu->kvm, pdpt_gfn, pdpte,
-				  offset * sizeof(u64), sizeof(pdpte));
-	if (ret < 0) {
-		ret = 0;
-		goto out;
-	}
-	for (i = 0; i < ARRAY_SIZE(pdpte); ++i) {
-		if ((pdpte[i] & 1) && (pdpte[i] & 0xfffffff0000001e6ull)) {
-			ret = 0;
-			goto out;
-		}
-	}
-	ret = 1;
-
-	memcpy(vcpu->arch.pdptrs, pdpte, sizeof(vcpu->arch.pdptrs));
-out:
-	mutex_unlock(&vcpu->kvm->lock);
-
-	return ret;
-}
-
-static bool pdptrs_changed(struct kvm_vcpu *vcpu)
-{
-	u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
-	bool changed = true;
-	int r;
-
-	if (is_long_mode(vcpu) || !is_pae(vcpu))
-		return false;
-
-	mutex_lock(&vcpu->kvm->lock);
-	r = kvm_read_guest(vcpu->kvm, vcpu->arch.cr3 & ~31u, pdpte, sizeof(pdpte));
-	if (r < 0)
-		goto out;
-	changed = memcmp(pdpte, vcpu->arch.pdptrs, sizeof(pdpte)) != 0;
-out:
-	mutex_unlock(&vcpu->kvm->lock);
-
-	return changed;
-}
-
-void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
-{
-	if (cr0 & CR0_RESERVED_BITS) {
-		printk(KERN_DEBUG "set_cr0: 0x%lx #GP, reserved bits 0x%lx\n",
-		       cr0, vcpu->arch.cr0);
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) {
-		printk(KERN_DEBUG "set_cr0: #GP, CD == 0 && NW == 1\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) {
-		printk(KERN_DEBUG "set_cr0: #GP, set PG flag "
-		       "and a clear PE flag\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
-#ifdef CONFIG_X86_64
-		if ((vcpu->arch.shadow_efer & EFER_LME)) {
-			int cs_db, cs_l;
-
-			if (!is_pae(vcpu)) {
-				printk(KERN_DEBUG "set_cr0: #GP, start paging "
-				       "in long mode while PAE is disabled\n");
-				kvm_inject_gp(vcpu, 0);
-				return;
-			}
-			kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
-			if (cs_l) {
-				printk(KERN_DEBUG "set_cr0: #GP, start paging "
-				       "in long mode while CS.L == 1\n");
-				kvm_inject_gp(vcpu, 0);
-				return;
-
-			}
-		} else
-#endif
-		if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
-			printk(KERN_DEBUG "set_cr0: #GP, pdptrs "
-			       "reserved bits\n");
-			kvm_inject_gp(vcpu, 0);
-			return;
-		}
-
-	}
-
-	kvm_x86_ops->set_cr0(vcpu, cr0);
-	vcpu->arch.cr0 = cr0;
-
-	mutex_lock(&vcpu->kvm->lock);
-	kvm_mmu_reset_context(vcpu);
-	mutex_unlock(&vcpu->kvm->lock);
-	return;
-}
-EXPORT_SYMBOL_GPL(set_cr0);
-
-void lmsw(struct kvm_vcpu *vcpu, unsigned long msw)
-{
-	set_cr0(vcpu, (vcpu->arch.cr0 & ~0x0ful) | (msw & 0x0f));
-}
-EXPORT_SYMBOL_GPL(lmsw);
-
-void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
-{
-	if (cr4 & CR4_RESERVED_BITS) {
-		printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if (is_long_mode(vcpu)) {
-		if (!(cr4 & X86_CR4_PAE)) {
-			printk(KERN_DEBUG "set_cr4: #GP, clearing PAE while "
-			       "in long mode\n");
-			kvm_inject_gp(vcpu, 0);
-			return;
-		}
-	} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE)
-		   && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
-		printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if (cr4 & X86_CR4_VMXE) {
-		printk(KERN_DEBUG "set_cr4: #GP, setting VMXE\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-	kvm_x86_ops->set_cr4(vcpu, cr4);
-	vcpu->arch.cr4 = cr4;
-	mutex_lock(&vcpu->kvm->lock);
-	kvm_mmu_reset_context(vcpu);
-	mutex_unlock(&vcpu->kvm->lock);
-}
-EXPORT_SYMBOL_GPL(set_cr4);
-
-void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
-{
-	if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) {
-		kvm_mmu_flush_tlb(vcpu);
-		return;
-	}
-
-	if (is_long_mode(vcpu)) {
-		if (cr3 & CR3_L_MODE_RESERVED_BITS) {
-			printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");
-			kvm_inject_gp(vcpu, 0);
-			return;
-		}
-	} else {
-		if (is_pae(vcpu)) {
-			if (cr3 & CR3_PAE_RESERVED_BITS) {
-				printk(KERN_DEBUG
-				       "set_cr3: #GP, reserved bits\n");
-				kvm_inject_gp(vcpu, 0);
-				return;
-			}
-			if (is_paging(vcpu) && !load_pdptrs(vcpu, cr3)) {
-				printk(KERN_DEBUG "set_cr3: #GP, pdptrs "
-				       "reserved bits\n");
-				kvm_inject_gp(vcpu, 0);
-				return;
-			}
-		}
-		/*
-		 * We don't check reserved bits in nonpae mode, because
-		 * this isn't enforced, and VMware depends on this.
-		 */
-	}
-
-	mutex_lock(&vcpu->kvm->lock);
-	/*
-	 * Does the new cr3 value map to physical memory? (Note, we
-	 * catch an invalid cr3 even in real-mode, because it would
-	 * cause trouble later on when we turn on paging anyway.)
-	 *
-	 * A real CPU would silently accept an invalid cr3 and would
-	 * attempt to use it - with largely undefined (and often hard
-	 * to debug) behavior on the guest side.
-	 */
-	if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT)))
-		kvm_inject_gp(vcpu, 0);
-	else {
-		vcpu->arch.cr3 = cr3;
-		vcpu->arch.mmu.new_cr3(vcpu);
-	}
-	mutex_unlock(&vcpu->kvm->lock);
-}
-EXPORT_SYMBOL_GPL(set_cr3);
-
-void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
-{
-	if (cr8 & CR8_RESERVED_BITS) {
-		printk(KERN_DEBUG "set_cr8: #GP, reserved bits 0x%lx\n", cr8);
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-	if (irqchip_in_kernel(vcpu->kvm))
-		kvm_lapic_set_tpr(vcpu, cr8);
-	else
-		vcpu->arch.cr8 = cr8;
-}
-EXPORT_SYMBOL_GPL(set_cr8);
-
-unsigned long get_cr8(struct kvm_vcpu *vcpu)
-{
-	if (irqchip_in_kernel(vcpu->kvm))
-		return kvm_lapic_get_cr8(vcpu);
-	else
-		return vcpu->arch.cr8;
-}
-EXPORT_SYMBOL_GPL(get_cr8);
-
-/*
- * List of msr numbers which we expose to userspace through KVM_GET_MSRS
- * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST.
- *
- * This list is modified at module load time to reflect the
- * capabilities of the host cpu.
- */
-static u32 msrs_to_save[] = {
-	MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
-	MSR_K6_STAR,
-#ifdef CONFIG_X86_64
-	MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
-#endif
-	MSR_IA32_TIME_STAMP_COUNTER,
-};
-
-static unsigned num_msrs_to_save;
-
-static u32 emulated_msrs[] = {
-	MSR_IA32_MISC_ENABLE,
-};
-
-#ifdef CONFIG_X86_64
-
-static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
-{
-	if (efer & EFER_RESERVED_BITS) {
-		printk(KERN_DEBUG "set_efer: 0x%llx #GP, reserved bits\n",
-		       efer);
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	if (is_paging(vcpu)
-	    && (vcpu->arch.shadow_efer & EFER_LME) != (efer & EFER_LME)) {
-		printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n");
-		kvm_inject_gp(vcpu, 0);
-		return;
-	}
-
-	kvm_x86_ops->set_efer(vcpu, efer);
-
-	efer &= ~EFER_LMA;
-	efer |= vcpu->arch.shadow_efer & EFER_LMA;
-
-	vcpu->arch.shadow_efer = efer;
-}
-
-#endif
-
-/*
- * Writes msr value into into the appropriate "register".
- * Returns 0 on success, non-0 otherwise.
- * Assumes vcpu_load() was already called.
- */
-int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
-{
-	return kvm_x86_ops->set_msr(vcpu, msr_index, data);
-}
-
-/*
- * Adapt set_msr() to msr_io()'s calling convention
- */
-static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data)
-{
-	return kvm_set_msr(vcpu, index, *data);
-}
-
-
-int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
-{
-	switch (msr) {
-#ifdef CONFIG_X86_64
-	case MSR_EFER:
-		set_efer(vcpu, data);
-		break;
-#endif
-	case MSR_IA32_MC0_STATUS:
-		pr_unimpl(vcpu, "%s: MSR_IA32_MC0_STATUS 0x%llx, nop\n",
-		       __FUNCTION__, data);
-		break;
-	case MSR_IA32_MCG_STATUS:
-		pr_unimpl(vcpu, "%s: MSR_IA32_MCG_STATUS 0x%llx, nop\n",
-			__FUNCTION__, data);
-		break;
-	case MSR_IA32_UCODE_REV:
-	case MSR_IA32_UCODE_WRITE:
-	case 0x200 ... 0x2ff: /* MTRRs */
-		break;
-	case MSR_IA32_APICBASE:
-		kvm_set_apic_base(vcpu, data);
-		break;
-	case MSR_IA32_MISC_ENABLE:
-		vcpu->arch.ia32_misc_enable_msr = data;
-		break;
-	default:
-		pr_unimpl(vcpu, "unhandled wrmsr: 0x%x\n", msr);
-		return 1;
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(kvm_set_msr_common);
-
-
-/*
- * Reads an msr value (of 'msr_index') into 'pdata'.
- * Returns 0 on success, non-0 otherwise.
- * Assumes vcpu_load() was already called.
- */
-int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata)
-{
-	return kvm_x86_ops->get_msr(vcpu, msr_index, pdata);
-}
-
-int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
-{
-	u64 data;
-
-	switch (msr) {
-	case 0xc0010010: /* SYSCFG */
-	case 0xc0010015: /* HWCR */
-	case MSR_IA32_PLATFORM_ID:
-	case MSR_IA32_P5_MC_ADDR:
-	case MSR_IA32_P5_MC_TYPE:
-	case MSR_IA32_MC0_CTL:
-	case MSR_IA32_MCG_STATUS:
-	case MSR_IA32_MCG_CAP:
-	case MSR_IA32_MC0_MISC:
-	case MSR_IA32_MC0_MISC+4:
-	case MSR_IA32_MC0_MISC+8:
-	case MSR_IA32_MC0_MISC+12:
-	case MSR_IA32_MC0_MISC+16:
-	case MSR_IA32_UCODE_REV:
-	case MSR_IA32_PERF_STATUS:
-	case MSR_IA32_EBL_CR_POWERON:
-		/* MTRR registers */
-	case 0xfe:
-	case 0x200 ... 0x2ff:
-		data = 0;
-		break;
-	case 0xcd: /* fsb frequency */
-		data = 3;
-		break;
-	case MSR_IA32_APICBASE:
-		data = kvm_get_apic_base(vcpu);
-		break;
-	case MSR_IA32_MISC_ENABLE:
-		data = vcpu->arch.ia32_misc_enable_msr;
-		break;
-#ifdef CONFIG_X86_64
-	case MSR_EFER:
-		data = vcpu->arch.shadow_efer;
-		break;
-#endif
-	default:
-		pr_unimpl(vcpu, "unhandled rdmsr: 0x%x\n", msr);
-		return 1;
-	}
-	*pdata = data;
-	return 0;
-}
-EXPORT_SYMBOL_GPL(kvm_get_msr_common);
-
-/*
- * Read or write a bunch of msrs. All parameters are kernel addresses.
- *
- * @return number of msrs set successfully.
- */
-static int __msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs *msrs,
-		    struct kvm_msr_entry *entries,
-		    int (*do_msr)(struct kvm_vcpu *vcpu,
-				  unsigned index, u64 *data))
-{
-	int i;
-
-	vcpu_load(vcpu);
-
-	for (i = 0; i < msrs->nmsrs; ++i)
-		if (do_msr(vcpu, entries[i].index, &entries[i].data))
-			break;
-
-	vcpu_put(vcpu);
-
-	return i;
-}
-
-/*
- * Read or write a bunch of msrs. Parameters are user addresses.
- *
- * @return number of msrs set successfully.
- */
-static int msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs __user *user_msrs,
-		  int (*do_msr)(struct kvm_vcpu *vcpu,
-				unsigned index, u64 *data),
-		  int writeback)
-{
-	struct kvm_msrs msrs;
-	struct kvm_msr_entry *entries;
-	int r, n;
-	unsigned size;
-
-	r = -EFAULT;
-	if (copy_from_user(&msrs, user_msrs, sizeof msrs))
-		goto out;
-
-	r = -E2BIG;
-	if (msrs.nmsrs >= MAX_IO_MSRS)
-		goto out;
-
-	r = -ENOMEM;
-	size = sizeof(struct kvm_msr_entry) * msrs.nmsrs;
-	entries = vmalloc(size);
-	if (!entries)
-		goto out;
-
-	r = -EFAULT;
-	if (copy_from_user(entries, user_msrs->entries, size))
-		goto out_free;
-
-	r = n = __msr_io(vcpu, &msrs, entries, do_msr);
-	if (r < 0)
-		goto out_free;
-
-	r = -EFAULT;
-	if (writeback && copy_to_user(user_msrs->entries, entries, size))
-		goto out_free;
-
-	r = n;
-
-out_free:
-	vfree(entries);
-out:
-	return r;
-}
-
-/*
- * Make sure that a cpu that is being hot-unplugged does not have any vcpus
- * cached on it.
- */
-void decache_vcpus_on_cpu(int cpu)
-{
-	struct kvm *vm;
-	struct kvm_vcpu *vcpu;
-	int i;
-
-	spin_lock(&kvm_lock);
-	list_for_each_entry(vm, &vm_list, vm_list)
-		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-			vcpu = vm->vcpus[i];
-			if (!vcpu)
-				continue;
-			/*
-			 * If the vcpu is locked, then it is running on some
-			 * other cpu and therefore it is not cached on the
-			 * cpu in question.
-			 *
-			 * If it's not locked, check the last cpu it executed
-			 * on.
-			 */
-			if (mutex_trylock(&vcpu->mutex)) {
-				if (vcpu->cpu == cpu) {
-					kvm_x86_ops->vcpu_decache(vcpu);
-					vcpu->cpu = -1;
-				}
-				mutex_unlock(&vcpu->mutex);
-			}
-		}
-	spin_unlock(&kvm_lock);
-}
-
-int kvm_dev_ioctl_check_extension(long ext)
-{
-	int r;
-
-	switch (ext) {
-	case KVM_CAP_IRQCHIP:
-	case KVM_CAP_HLT:
-	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
-	case KVM_CAP_USER_MEMORY:
-	case KVM_CAP_SET_TSS_ADDR:
-	case KVM_CAP_EXT_CPUID:
-		r = 1;
-		break;
-	default:
-		r = 0;
-		break;
-	}
-	return r;
-
-}
-
-long kvm_arch_dev_ioctl(struct file *filp,
-			unsigned int ioctl, unsigned long arg)
-{
-	void __user *argp = (void __user *)arg;
-	long r;
-
-	switch (ioctl) {
-	case KVM_GET_MSR_INDEX_LIST: {
-		struct kvm_msr_list __user *user_msr_list = argp;
-		struct kvm_msr_list msr_list;
-		unsigned n;
-
-		r = -EFAULT;
-		if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list))
-			goto out;
-		n = msr_list.nmsrs;
-		msr_list.nmsrs = num_msrs_to_save + ARRAY_SIZE(emulated_msrs);
-		if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list))
-			goto out;
-		r = -E2BIG;
-		if (n < num_msrs_to_save)
-			goto out;
-		r = -EFAULT;
-		if (copy_to_user(user_msr_list->indices, &msrs_to_save,
-				 num_msrs_to_save * sizeof(u32)))
-			goto out;
-		if (copy_to_user(user_msr_list->indices
-				 + num_msrs_to_save * sizeof(u32),
-				 &emulated_msrs,
-				 ARRAY_SIZE(emulated_msrs) * sizeof(u32)))
-			goto out;
-		r = 0;
-		break;
-	}
-	default:
-		r = -EINVAL;
-	}
-out:
-	return r;
-}
-
-void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
-{
-	kvm_x86_ops->vcpu_load(vcpu, cpu);
-}
-
-void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
-{
-	kvm_x86_ops->vcpu_put(vcpu);
-	kvm_put_guest_fpu(vcpu);
-}
-
-static int is_efer_nx(void)
-{
-	u64 efer;
-
-	rdmsrl(MSR_EFER, efer);
-	return efer & EFER_NX;
-}
-
-static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
-{
-	int i;
-	struct kvm_cpuid_entry2 *e, *entry;
-
-	entry = NULL;
-	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
-		e = &vcpu->arch.cpuid_entries[i];
-		if (e->function == 0x80000001) {
-			entry = e;
-			break;
-		}
-	}
-	if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) {
-		entry->edx &= ~(1 << 20);
-		printk(KERN_INFO "kvm: guest NX capability removed\n");
-	}
-}
-
-/* when an old userspace process fills a new kernel module */
-static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
-				    struct kvm_cpuid *cpuid,
-				    struct kvm_cpuid_entry __user *entries)
-{
-	int r, i;
-	struct kvm_cpuid_entry *cpuid_entries;
-
-	r = -E2BIG;
-	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
-		goto out;
-	r = -ENOMEM;
-	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent);
-	if (!cpuid_entries)
-		goto out;
-	r = -EFAULT;
-	if (copy_from_user(cpuid_entries, entries,
-			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
-		goto out_free;
-	for (i = 0; i < cpuid->nent; i++) {
-		vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function;
-		vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax;
-		vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx;
-		vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx;
-		vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx;
-		vcpu->arch.cpuid_entries[i].index = 0;
-		vcpu->arch.cpuid_entries[i].flags = 0;
-		vcpu->arch.cpuid_entries[i].padding[0] = 0;
-		vcpu->arch.cpuid_entries[i].padding[1] = 0;
-		vcpu->arch.cpuid_entries[i].padding[2] = 0;
-	}
-	vcpu->arch.cpuid_nent = cpuid->nent;
-	cpuid_fix_nx_cap(vcpu);
-	r = 0;
-
-out_free:
-	vfree(cpuid_entries);
-out:
-	return r;
-}
-
-static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
-				    struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
-{
-	int r;
-
-	r = -E2BIG;
-	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
-		goto out;
-	r = -EFAULT;
-	if (copy_from_user(&vcpu->arch.cpuid_entries, entries,
-			   cpuid->nent * sizeof(struct kvm_cpuid_entry2)))
-		goto out;
-	vcpu->arch.cpuid_nent = cpuid->nent;
-	return 0;
-
-out:
-	return r;
-}
-
-static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
-				    struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
-{
-	int r;
-
-	r = -E2BIG;
-	if (cpuid->nent < vcpu->arch.cpuid_nent)
-		goto out;
-	r = -EFAULT;
-	if (copy_to_user(entries, &vcpu->arch.cpuid_entries,
-			   vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
-		goto out;
-	return 0;
-
-out:
-	cpuid->nent = vcpu->arch.cpuid_nent;
-	return r;
-}
-
-static inline u32 bit(int bitno)
-{
-	return 1 << (bitno & 31);
-}
-
-static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function,
-			  u32 index)
-{
-	entry->function = function;
-	entry->index = index;
-	cpuid_count(entry->function, entry->index,
-		&entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
-	entry->flags = 0;
-}
-
-static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
-			 u32 index, int *nent, int maxnent)
-{
-	const u32 kvm_supported_word0_x86_features = bit(X86_FEATURE_FPU) |
-		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
-		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
-		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
-		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
-		bit(X86_FEATURE_SEP) | bit(X86_FEATURE_PGE) |
-		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
-		bit(X86_FEATURE_CLFLSH) | bit(X86_FEATURE_MMX) |
-		bit(X86_FEATURE_FXSR) | bit(X86_FEATURE_XMM) |
-		bit(X86_FEATURE_XMM2) | bit(X86_FEATURE_SELFSNOOP);
-	const u32 kvm_supported_word1_x86_features = bit(X86_FEATURE_FPU) |
-		bit(X86_FEATURE_VME) | bit(X86_FEATURE_DE) |
-		bit(X86_FEATURE_PSE) | bit(X86_FEATURE_TSC) |
-		bit(X86_FEATURE_MSR) | bit(X86_FEATURE_PAE) |
-		bit(X86_FEATURE_CX8) | bit(X86_FEATURE_APIC) |
-		bit(X86_FEATURE_PGE) |
-		bit(X86_FEATURE_CMOV) | bit(X86_FEATURE_PSE36) |
-		bit(X86_FEATURE_MMX) | bit(X86_FEATURE_FXSR) |
-		bit(X86_FEATURE_SYSCALL) |
-		(bit(X86_FEATURE_NX) && is_efer_nx()) |
-#ifdef CONFIG_X86_64
-		bit(X86_FEATURE_LM) |
-#endif
-		bit(X86_FEATURE_MMXEXT) |
-		bit(X86_FEATURE_3DNOWEXT) |
-		bit(X86_FEATURE_3DNOW);
-	const u32 kvm_supported_word3_x86_features =
-		bit(X86_FEATURE_XMM3) | bit(X86_FEATURE_CX16);
-	const u32 kvm_supported_word6_x86_features =
-		bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY);
-
-	/* all func 2 cpuid_count() should be called on the same cpu */
-	get_cpu();
-	do_cpuid_1_ent(entry, function, index);
-	++*nent;
-
-	switch (function) {
-	case 0:
-		entry->eax = min(entry->eax, (u32)0xb);
-		break;
-	case 1:
-		entry->edx &= kvm_supported_word0_x86_features;
-		entry->ecx &= kvm_supported_word3_x86_features;
-		break;
-	/* function 2 entries are STATEFUL. That is, repeated cpuid commands
-	 * may return different values. This forces us to get_cpu() before
-	 * issuing the first command, and also to emulate this annoying behavior
-	 * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */
-	case 2: {
-		int t, times = entry->eax & 0xff;
-
-		entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
-		for (t = 1; t < times && *nent < maxnent; ++t) {
-			do_cpuid_1_ent(&entry[t], function, 0);
-			entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
-			++*nent;
-		}
-		break;
-	}
-	/* function 4 and 0xb have additional index. */
-	case 4: {
-		int index, cache_type;
-
-		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
-		/* read more entries until cache_type is zero */
-		for (index = 1; *nent < maxnent; ++index) {
-			cache_type = entry[index - 1].eax & 0x1f;
-			if (!cache_type)
-				break;
-			do_cpuid_1_ent(&entry[index], function, index);
-			entry[index].flags |=
-			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
-			++*nent;
-		}
-		break;
-	}
-	case 0xb: {
-		int index, level_type;
-
-		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
-		/* read more entries until level_type is zero */
-		for (index = 1; *nent < maxnent; ++index) {
-			level_type = entry[index - 1].ecx & 0xff;
-			if (!level_type)
-				break;
-			do_cpuid_1_ent(&entry[index], function, index);
-			entry[index].flags |=
-			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
-			++*nent;
-		}
-		break;
-	}
-	case 0x80000000:
-		entry->eax = min(entry->eax, 0x8000001a);
-		break;
-	case 0x80000001:
-		entry->edx &= kvm_supported_word1_x86_features;
-		entry->ecx &= kvm_supported_word6_x86_features;
-		break;
-	}
-	put_cpu();
-}
-
-static int kvm_vm_ioctl_get_supported_cpuid(struct kvm *kvm,
-				    struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
-{
-	struct kvm_cpuid_entry2 *cpuid_entries;
-	int limit, nent = 0, r = -E2BIG;
-	u32 func;
-
-	if (cpuid->nent < 1)
-		goto out;
-	r = -ENOMEM;
-	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
-	if (!cpuid_entries)
-		goto out;
-
-	do_cpuid_ent(&cpuid_entries[0], 0, 0, &nent, cpuid->nent);
-	limit = cpuid_entries[0].eax;
-	for (func = 1; func <= limit && nent < cpuid->nent; ++func)
-		do_cpuid_ent(&cpuid_entries[nent], func, 0,
-				&nent, cpuid->nent);
-	r = -E2BIG;
-	if (nent >= cpuid->nent)
-		goto out_free;
-
-	do_cpuid_ent(&cpuid_entries[nent], 0x80000000, 0, &nent, cpuid->nent);
-	limit = cpuid_entries[nent - 1].eax;
-	for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func)
-		do_cpuid_ent(&cpuid_entries[nent], func, 0,
-			       &nent, cpuid->nent);
-	r = -EFAULT;
-	if (copy_to_user(entries, cpuid_entries,
-			nent * sizeof(struct kvm_cpuid_entry2)))
-		goto out_free;
-	cpuid->nent = nent;
-	r = 0;
-
-out_free:
-	vfree(cpuid_entries);
-out:
-	return r;
-}
-
-static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu,
-				    struct kvm_lapic_state *s)
-{
-	vcpu_load(vcpu);
-	memcpy(s->regs, vcpu->arch.apic->regs, sizeof *s);
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-static int kvm_vcpu_ioctl_set_lapic(struct kvm_vcpu *vcpu,
-				    struct kvm_lapic_state *s)
-{
-	vcpu_load(vcpu);
-	memcpy(vcpu->arch.apic->regs, s->regs, sizeof *s);
-	kvm_apic_post_state_restore(vcpu);
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
-				    struct kvm_interrupt *irq)
-{
-	if (irq->irq < 0 || irq->irq >= 256)
-		return -EINVAL;
-	if (irqchip_in_kernel(vcpu->kvm))
-		return -ENXIO;
-	vcpu_load(vcpu);
-
-	set_bit(irq->irq, vcpu->arch.irq_pending);
-	set_bit(irq->irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-long kvm_arch_vcpu_ioctl(struct file *filp,
-			 unsigned int ioctl, unsigned long arg)
-{
-	struct kvm_vcpu *vcpu = filp->private_data;
-	void __user *argp = (void __user *)arg;
-	int r;
-
-	switch (ioctl) {
-	case KVM_GET_LAPIC: {
-		struct kvm_lapic_state lapic;
-
-		memset(&lapic, 0, sizeof lapic);
-		r = kvm_vcpu_ioctl_get_lapic(vcpu, &lapic);
-		if (r)
-			goto out;
-		r = -EFAULT;
-		if (copy_to_user(argp, &lapic, sizeof lapic))
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_SET_LAPIC: {
-		struct kvm_lapic_state lapic;
-
-		r = -EFAULT;
-		if (copy_from_user(&lapic, argp, sizeof lapic))
-			goto out;
-		r = kvm_vcpu_ioctl_set_lapic(vcpu, &lapic);;
-		if (r)
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_INTERRUPT: {
-		struct kvm_interrupt irq;
-
-		r = -EFAULT;
-		if (copy_from_user(&irq, argp, sizeof irq))
-			goto out;
-		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
-		if (r)
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_SET_CPUID: {
-		struct kvm_cpuid __user *cpuid_arg = argp;
-		struct kvm_cpuid cpuid;
-
-		r = -EFAULT;
-		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
-			goto out;
-		r = kvm_vcpu_ioctl_set_cpuid(vcpu, &cpuid, cpuid_arg->entries);
-		if (r)
-			goto out;
-		break;
-	}
-	case KVM_SET_CPUID2: {
-		struct kvm_cpuid2 __user *cpuid_arg = argp;
-		struct kvm_cpuid2 cpuid;
-
-		r = -EFAULT;
-		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
-			goto out;
-		r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid,
-				cpuid_arg->entries);
-		if (r)
-			goto out;
-		break;
-	}
-	case KVM_GET_CPUID2: {
-		struct kvm_cpuid2 __user *cpuid_arg = argp;
-		struct kvm_cpuid2 cpuid;
-
-		r = -EFAULT;
-		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
-			goto out;
-		r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid,
-				cpuid_arg->entries);
-		if (r)
-			goto out;
-		r = -EFAULT;
-		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_GET_MSRS:
-		r = msr_io(vcpu, argp, kvm_get_msr, 1);
-		break;
-	case KVM_SET_MSRS:
-		r = msr_io(vcpu, argp, do_set_msr, 0);
-		break;
-	default:
-		r = -EINVAL;
-	}
-out:
-	return r;
-}
-
-static int kvm_vm_ioctl_set_tss_addr(struct kvm *kvm, unsigned long addr)
-{
-	int ret;
-
-	if (addr > (unsigned int)(-3 * PAGE_SIZE))
-		return -1;
-	ret = kvm_x86_ops->set_tss_addr(kvm, addr);
-	return ret;
-}
-
-static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
-					  u32 kvm_nr_mmu_pages)
-{
-	if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES)
-		return -EINVAL;
-
-	mutex_lock(&kvm->lock);
-
-	kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages);
-	kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages;
-
-	mutex_unlock(&kvm->lock);
-	return 0;
-}
-
-static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm)
-{
-	return kvm->arch.n_alloc_mmu_pages;
-}
-
-gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
-{
-	int i;
-	struct kvm_mem_alias *alias;
-
-	for (i = 0; i < kvm->arch.naliases; ++i) {
-		alias = &kvm->arch.aliases[i];
-		if (gfn >= alias->base_gfn
-		    && gfn < alias->base_gfn + alias->npages)
-			return alias->target_gfn + gfn - alias->base_gfn;
-	}
-	return gfn;
-}
-
-/*
- * Set a new alias region.  Aliases map a portion of physical memory into
- * another portion.  This is useful for memory windows, for example the PC
- * VGA region.
- */
-static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
-					 struct kvm_memory_alias *alias)
-{
-	int r, n;
-	struct kvm_mem_alias *p;
-
-	r = -EINVAL;
-	/* General sanity checks */
-	if (alias->memory_size & (PAGE_SIZE - 1))
-		goto out;
-	if (alias->guest_phys_addr & (PAGE_SIZE - 1))
-		goto out;
-	if (alias->slot >= KVM_ALIAS_SLOTS)
-		goto out;
-	if (alias->guest_phys_addr + alias->memory_size
-	    < alias->guest_phys_addr)
-		goto out;
-	if (alias->target_phys_addr + alias->memory_size
-	    < alias->target_phys_addr)
-		goto out;
-
-	mutex_lock(&kvm->lock);
-
-	p = &kvm->arch.aliases[alias->slot];
-	p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
-	p->npages = alias->memory_size >> PAGE_SHIFT;
-	p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT;
-
-	for (n = KVM_ALIAS_SLOTS; n > 0; --n)
-		if (kvm->arch.aliases[n - 1].npages)
-			break;
-	kvm->arch.naliases = n;
-
-	kvm_mmu_zap_all(kvm);
-
-	mutex_unlock(&kvm->lock);
-
-	return 0;
-
-out:
-	return r;
-}
-
-static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
-{
-	int r;
-
-	r = 0;
-	switch (chip->chip_id) {
-	case KVM_IRQCHIP_PIC_MASTER:
-		memcpy(&chip->chip.pic,
-			&pic_irqchip(kvm)->pics[0],
-			sizeof(struct kvm_pic_state));
-		break;
-	case KVM_IRQCHIP_PIC_SLAVE:
-		memcpy(&chip->chip.pic,
-			&pic_irqchip(kvm)->pics[1],
-			sizeof(struct kvm_pic_state));
-		break;
-	case KVM_IRQCHIP_IOAPIC:
-		memcpy(&chip->chip.ioapic,
-			ioapic_irqchip(kvm),
-			sizeof(struct kvm_ioapic_state));
-		break;
-	default:
-		r = -EINVAL;
-		break;
-	}
-	return r;
-}
-
-static int kvm_vm_ioctl_set_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
-{
-	int r;
-
-	r = 0;
-	switch (chip->chip_id) {
-	case KVM_IRQCHIP_PIC_MASTER:
-		memcpy(&pic_irqchip(kvm)->pics[0],
-			&chip->chip.pic,
-			sizeof(struct kvm_pic_state));
-		break;
-	case KVM_IRQCHIP_PIC_SLAVE:
-		memcpy(&pic_irqchip(kvm)->pics[1],
-			&chip->chip.pic,
-			sizeof(struct kvm_pic_state));
-		break;
-	case KVM_IRQCHIP_IOAPIC:
-		memcpy(ioapic_irqchip(kvm),
-			&chip->chip.ioapic,
-			sizeof(struct kvm_ioapic_state));
-		break;
-	default:
-		r = -EINVAL;
-		break;
-	}
-	kvm_pic_update_irq(pic_irqchip(kvm));
-	return r;
-}
-
-/*
- * Get (and clear) the dirty memory log for a memory slot.
- */
-int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
-				      struct kvm_dirty_log *log)
-{
-	int r;
-	int n;
-	struct kvm_memory_slot *memslot;
-	int is_dirty = 0;
-
-	mutex_lock(&kvm->lock);
-
-	r = kvm_get_dirty_log(kvm, log, &is_dirty);
-	if (r)
-		goto out;
-
-	/* If nothing is dirty, don't bother messing with page tables. */
-	if (is_dirty) {
-		kvm_mmu_slot_remove_write_access(kvm, log->slot);
-		kvm_flush_remote_tlbs(kvm);
-		memslot = &kvm->memslots[log->slot];
-		n = ALIGN(memslot->npages, BITS_PER_LONG) / 8;
-		memset(memslot->dirty_bitmap, 0, n);
-	}
-	r = 0;
-out:
-	mutex_unlock(&kvm->lock);
-	return r;
-}
-
-long kvm_arch_vm_ioctl(struct file *filp,
-		       unsigned int ioctl, unsigned long arg)
-{
-	struct kvm *kvm = filp->private_data;
-	void __user *argp = (void __user *)arg;
-	int r = -EINVAL;
-
-	switch (ioctl) {
-	case KVM_SET_TSS_ADDR:
-		r = kvm_vm_ioctl_set_tss_addr(kvm, arg);
-		if (r < 0)
-			goto out;
-		break;
-	case KVM_SET_MEMORY_REGION: {
-		struct kvm_memory_region kvm_mem;
-		struct kvm_userspace_memory_region kvm_userspace_mem;
-
-		r = -EFAULT;
-		if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
-			goto out;
-		kvm_userspace_mem.slot = kvm_mem.slot;
-		kvm_userspace_mem.flags = kvm_mem.flags;
-		kvm_userspace_mem.guest_phys_addr = kvm_mem.guest_phys_addr;
-		kvm_userspace_mem.memory_size = kvm_mem.memory_size;
-		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 0);
-		if (r)
-			goto out;
-		break;
-	}
-	case KVM_SET_NR_MMU_PAGES:
-		r = kvm_vm_ioctl_set_nr_mmu_pages(kvm, arg);
-		if (r)
-			goto out;
-		break;
-	case KVM_GET_NR_MMU_PAGES:
-		r = kvm_vm_ioctl_get_nr_mmu_pages(kvm);
-		break;
-	case KVM_SET_MEMORY_ALIAS: {
-		struct kvm_memory_alias alias;
-
-		r = -EFAULT;
-		if (copy_from_user(&alias, argp, sizeof alias))
-			goto out;
-		r = kvm_vm_ioctl_set_memory_alias(kvm, &alias);
-		if (r)
-			goto out;
-		break;
-	}
-	case KVM_CREATE_IRQCHIP:
-		r = -ENOMEM;
-		kvm->arch.vpic = kvm_create_pic(kvm);
-		if (kvm->arch.vpic) {
-			r = kvm_ioapic_init(kvm);
-			if (r) {
-				kfree(kvm->arch.vpic);
-				kvm->arch.vpic = NULL;
-				goto out;
-			}
-		} else
-			goto out;
-		break;
-	case KVM_IRQ_LINE: {
-		struct kvm_irq_level irq_event;
-
-		r = -EFAULT;
-		if (copy_from_user(&irq_event, argp, sizeof irq_event))
-			goto out;
-		if (irqchip_in_kernel(kvm)) {
-			mutex_lock(&kvm->lock);
-			if (irq_event.irq < 16)
-				kvm_pic_set_irq(pic_irqchip(kvm),
-					irq_event.irq,
-					irq_event.level);
-			kvm_ioapic_set_irq(kvm->arch.vioapic,
-					irq_event.irq,
-					irq_event.level);
-			mutex_unlock(&kvm->lock);
-			r = 0;
-		}
-		break;
-	}
-	case KVM_GET_IRQCHIP: {
-		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
-		struct kvm_irqchip chip;
-
-		r = -EFAULT;
-		if (copy_from_user(&chip, argp, sizeof chip))
-			goto out;
-		r = -ENXIO;
-		if (!irqchip_in_kernel(kvm))
-			goto out;
-		r = kvm_vm_ioctl_get_irqchip(kvm, &chip);
-		if (r)
-			goto out;
-		r = -EFAULT;
-		if (copy_to_user(argp, &chip, sizeof chip))
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_SET_IRQCHIP: {
-		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
-		struct kvm_irqchip chip;
-
-		r = -EFAULT;
-		if (copy_from_user(&chip, argp, sizeof chip))
-			goto out;
-		r = -ENXIO;
-		if (!irqchip_in_kernel(kvm))
-			goto out;
-		r = kvm_vm_ioctl_set_irqchip(kvm, &chip);
-		if (r)
-			goto out;
-		r = 0;
-		break;
-	}
-	case KVM_GET_SUPPORTED_CPUID: {
-		struct kvm_cpuid2 __user *cpuid_arg = argp;
-		struct kvm_cpuid2 cpuid;
-
-		r = -EFAULT;
-		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
-			goto out;
-		r = kvm_vm_ioctl_get_supported_cpuid(kvm, &cpuid,
-			cpuid_arg->entries);
-		if (r)
-			goto out;
-
-		r = -EFAULT;
-		if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid))
-			goto out;
-		r = 0;
-		break;
-	}
-	default:
-		;
-	}
-out:
-	return r;
-}
-
-static void kvm_init_msr_list(void)
-{
-	u32 dummy[2];
-	unsigned i, j;
-
-	for (i = j = 0; i < ARRAY_SIZE(msrs_to_save); i++) {
-		if (rdmsr_safe(msrs_to_save[i], &dummy[0], &dummy[1]) < 0)
-			continue;
-		if (j < i)
-			msrs_to_save[j] = msrs_to_save[i];
-		j++;
-	}
-	num_msrs_to_save = j;
-}
-
-/*
- * Only apic need an MMIO device hook, so shortcut now..
- */
-static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu,
-						gpa_t addr)
-{
-	struct kvm_io_device *dev;
-
-	if (vcpu->arch.apic) {
-		dev = &vcpu->arch.apic->dev;
-		if (dev->in_range(dev, addr))
-			return dev;
-	}
-	return NULL;
-}
-
-
-static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu,
-						gpa_t addr)
-{
-	struct kvm_io_device *dev;
-
-	dev = vcpu_find_pervcpu_dev(vcpu, addr);
-	if (dev == NULL)
-		dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr);
-	return dev;
-}
-
-int emulator_read_std(unsigned long addr,
-			     void *val,
-			     unsigned int bytes,
-			     struct kvm_vcpu *vcpu)
-{
-	void *data = val;
-
-	while (bytes) {
-		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
-		unsigned offset = addr & (PAGE_SIZE-1);
-		unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
-		int ret;
-
-		if (gpa == UNMAPPED_GVA)
-			return X86EMUL_PROPAGATE_FAULT;
-		ret = kvm_read_guest(vcpu->kvm, gpa, data, tocopy);
-		if (ret < 0)
-			return X86EMUL_UNHANDLEABLE;
-
-		bytes -= tocopy;
-		data += tocopy;
-		addr += tocopy;
-	}
-
-	return X86EMUL_CONTINUE;
-}
-EXPORT_SYMBOL_GPL(emulator_read_std);
-
-static int emulator_read_emulated(unsigned long addr,
-				  void *val,
-				  unsigned int bytes,
-				  struct kvm_vcpu *vcpu)
-{
-	struct kvm_io_device *mmio_dev;
-	gpa_t                 gpa;
-
-	if (vcpu->mmio_read_completed) {
-		memcpy(val, vcpu->mmio_data, bytes);
-		vcpu->mmio_read_completed = 0;
-		return X86EMUL_CONTINUE;
-	}
-
-	gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
-
-	/* For APIC access vmexit */
-	if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
-		goto mmio;
-
-	if (emulator_read_std(addr, val, bytes, vcpu)
-			== X86EMUL_CONTINUE)
-		return X86EMUL_CONTINUE;
-	if (gpa == UNMAPPED_GVA)
-		return X86EMUL_PROPAGATE_FAULT;
-
-mmio:
-	/*
-	 * Is this MMIO handled locally?
-	 */
-	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
-	if (mmio_dev) {
-		kvm_iodevice_read(mmio_dev, gpa, bytes, val);
-		return X86EMUL_CONTINUE;
-	}
-
-	vcpu->mmio_needed = 1;
-	vcpu->mmio_phys_addr = gpa;
-	vcpu->mmio_size = bytes;
-	vcpu->mmio_is_write = 0;
-
-	return X86EMUL_UNHANDLEABLE;
-}
-
-static int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
-			       const void *val, int bytes)
-{
-	int ret;
-
-	ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes);
-	if (ret < 0)
-		return 0;
-	kvm_mmu_pte_write(vcpu, gpa, val, bytes);
-	return 1;
-}
-
-static int emulator_write_emulated_onepage(unsigned long addr,
-					   const void *val,
-					   unsigned int bytes,
-					   struct kvm_vcpu *vcpu)
-{
-	struct kvm_io_device *mmio_dev;
-	gpa_t                 gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
-
-	if (gpa == UNMAPPED_GVA) {
-		kvm_inject_page_fault(vcpu, addr, 2);
-		return X86EMUL_PROPAGATE_FAULT;
-	}
-
-	/* For APIC access vmexit */
-	if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
-		goto mmio;
-
-	if (emulator_write_phys(vcpu, gpa, val, bytes))
-		return X86EMUL_CONTINUE;
-
-mmio:
-	/*
-	 * Is this MMIO handled locally?
-	 */
-	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
-	if (mmio_dev) {
-		kvm_iodevice_write(mmio_dev, gpa, bytes, val);
-		return X86EMUL_CONTINUE;
-	}
-
-	vcpu->mmio_needed = 1;
-	vcpu->mmio_phys_addr = gpa;
-	vcpu->mmio_size = bytes;
-	vcpu->mmio_is_write = 1;
-	memcpy(vcpu->mmio_data, val, bytes);
-
-	return X86EMUL_CONTINUE;
-}
-
-int emulator_write_emulated(unsigned long addr,
-				   const void *val,
-				   unsigned int bytes,
-				   struct kvm_vcpu *vcpu)
-{
-	/* Crossing a page boundary? */
-	if (((addr + bytes - 1) ^ addr) & PAGE_MASK) {
-		int rc, now;
-
-		now = -addr & ~PAGE_MASK;
-		rc = emulator_write_emulated_onepage(addr, val, now, vcpu);
-		if (rc != X86EMUL_CONTINUE)
-			return rc;
-		addr += now;
-		val += now;
-		bytes -= now;
-	}
-	return emulator_write_emulated_onepage(addr, val, bytes, vcpu);
-}
-EXPORT_SYMBOL_GPL(emulator_write_emulated);
-
-static int emulator_cmpxchg_emulated(unsigned long addr,
-				     const void *old,
-				     const void *new,
-				     unsigned int bytes,
-				     struct kvm_vcpu *vcpu)
-{
-	static int reported;
-
-	if (!reported) {
-		reported = 1;
-		printk(KERN_WARNING "kvm: emulating exchange as write\n");
-	}
-#ifndef CONFIG_X86_64
-	/* guests cmpxchg8b have to be emulated atomically */
-	if (bytes == 8) {
-		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
-		struct page *page;
-		char *addr;
-		u64 val;
-
-		if (gpa == UNMAPPED_GVA ||
-		   (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
-			goto emul_write;
-
-		if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK))
-			goto emul_write;
-
-		val = *(u64 *)new;
-		page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
-		addr = kmap_atomic(page, KM_USER0);
-		set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
-		kunmap_atomic(addr, KM_USER0);
-		kvm_release_page_dirty(page);
-	}
-emul_write:
-#endif
-
-	return emulator_write_emulated(addr, new, bytes, vcpu);
-}
-
-static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg)
-{
-	return kvm_x86_ops->get_segment_base(vcpu, seg);
-}
-
-int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address)
-{
-	return X86EMUL_CONTINUE;
-}
-
-int emulate_clts(struct kvm_vcpu *vcpu)
-{
-	kvm_x86_ops->set_cr0(vcpu, vcpu->arch.cr0 & ~X86_CR0_TS);
-	return X86EMUL_CONTINUE;
-}
-
-int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long *dest)
-{
-	struct kvm_vcpu *vcpu = ctxt->vcpu;
-
-	switch (dr) {
-	case 0 ... 3:
-		*dest = kvm_x86_ops->get_dr(vcpu, dr);
-		return X86EMUL_CONTINUE;
-	default:
-		pr_unimpl(vcpu, "%s: unexpected dr %u\n", __FUNCTION__, dr);
-		return X86EMUL_UNHANDLEABLE;
-	}
-}
-
-int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value)
-{
-	unsigned long mask = (ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U;
-	int exception;
-
-	kvm_x86_ops->set_dr(ctxt->vcpu, dr, value & mask, &exception);
-	if (exception) {
-		/* FIXME: better handling */
-		return X86EMUL_UNHANDLEABLE;
-	}
-	return X86EMUL_CONTINUE;
-}
-
-void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
-{
-	static int reported;
-	u8 opcodes[4];
-	unsigned long rip = vcpu->arch.rip;
-	unsigned long rip_linear;
-
-	rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS);
-
-	if (reported)
-		return;
-
-	emulator_read_std(rip_linear, (void *)opcodes, 4, vcpu);
-
-	printk(KERN_ERR "emulation failed (%s) rip %lx %02x %02x %02x %02x\n",
-	       context, rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]);
-	reported = 1;
-}
-EXPORT_SYMBOL_GPL(kvm_report_emulation_failure);
-
-struct x86_emulate_ops emulate_ops = {
-	.read_std            = emulator_read_std,
-	.read_emulated       = emulator_read_emulated,
-	.write_emulated      = emulator_write_emulated,
-	.cmpxchg_emulated    = emulator_cmpxchg_emulated,
-};
-
-int emulate_instruction(struct kvm_vcpu *vcpu,
-			struct kvm_run *run,
-			unsigned long cr2,
-			u16 error_code,
-			int no_decode)
-{
-	int r;
-
-	vcpu->arch.mmio_fault_cr2 = cr2;
-	kvm_x86_ops->cache_regs(vcpu);
-
-	vcpu->mmio_is_write = 0;
-	vcpu->arch.pio.string = 0;
-
-	if (!no_decode) {
-		int cs_db, cs_l;
-		kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
-
-		vcpu->arch.emulate_ctxt.vcpu = vcpu;
-		vcpu->arch.emulate_ctxt.eflags = kvm_x86_ops->get_rflags(vcpu);
-		vcpu->arch.emulate_ctxt.mode =
-			(vcpu->arch.emulate_ctxt.eflags & X86_EFLAGS_VM)
-			? X86EMUL_MODE_REAL : cs_l
-			? X86EMUL_MODE_PROT64 :	cs_db
-			? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;
-
-		if (vcpu->arch.emulate_ctxt.mode == X86EMUL_MODE_PROT64) {
-			vcpu->arch.emulate_ctxt.cs_base = 0;
-			vcpu->arch.emulate_ctxt.ds_base = 0;
-			vcpu->arch.emulate_ctxt.es_base = 0;
-			vcpu->arch.emulate_ctxt.ss_base = 0;
-		} else {
-			vcpu->arch.emulate_ctxt.cs_base =
-					get_segment_base(vcpu, VCPU_SREG_CS);
-			vcpu->arch.emulate_ctxt.ds_base =
-					get_segment_base(vcpu, VCPU_SREG_DS);
-			vcpu->arch.emulate_ctxt.es_base =
-					get_segment_base(vcpu, VCPU_SREG_ES);
-			vcpu->arch.emulate_ctxt.ss_base =
-					get_segment_base(vcpu, VCPU_SREG_SS);
-		}
-
-		vcpu->arch.emulate_ctxt.gs_base =
-					get_segment_base(vcpu, VCPU_SREG_GS);
-		vcpu->arch.emulate_ctxt.fs_base =
-					get_segment_base(vcpu, VCPU_SREG_FS);
-
-		r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
-		++vcpu->stat.insn_emulation;
-		if (r)  {
-			++vcpu->stat.insn_emulation_fail;
-			if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
-				return EMULATE_DONE;
-			return EMULATE_FAIL;
-		}
-	}
-
-	r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
-
-	if (vcpu->arch.pio.string)
-		return EMULATE_DO_MMIO;
-
-	if ((r || vcpu->mmio_is_write) && run) {
-		run->exit_reason = KVM_EXIT_MMIO;
-		run->mmio.phys_addr = vcpu->mmio_phys_addr;
-		memcpy(run->mmio.data, vcpu->mmio_data, 8);
-		run->mmio.len = vcpu->mmio_size;
-		run->mmio.is_write = vcpu->mmio_is_write;
-	}
-
-	if (r) {
-		if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
-			return EMULATE_DONE;
-		if (!vcpu->mmio_needed) {
-			kvm_report_emulation_failure(vcpu, "mmio");
-			return EMULATE_FAIL;
-		}
-		return EMULATE_DO_MMIO;
-	}
-
-	kvm_x86_ops->decache_regs(vcpu);
-	kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags);
-
-	if (vcpu->mmio_is_write) {
-		vcpu->mmio_needed = 0;
-		return EMULATE_DO_MMIO;
-	}
-
-	return EMULATE_DONE;
-}
-EXPORT_SYMBOL_GPL(emulate_instruction);
-
-static void free_pio_guest_pages(struct kvm_vcpu *vcpu)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(vcpu->arch.pio.guest_pages); ++i)
-		if (vcpu->arch.pio.guest_pages[i]) {
-			kvm_release_page_dirty(vcpu->arch.pio.guest_pages[i]);
-			vcpu->arch.pio.guest_pages[i] = NULL;
-		}
-}
-
-static int pio_copy_data(struct kvm_vcpu *vcpu)
-{
-	void *p = vcpu->arch.pio_data;
-	void *q;
-	unsigned bytes;
-	int nr_pages = vcpu->arch.pio.guest_pages[1] ? 2 : 1;
-
-	q = vmap(vcpu->arch.pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
-		 PAGE_KERNEL);
-	if (!q) {
-		free_pio_guest_pages(vcpu);
-		return -ENOMEM;
-	}
-	q += vcpu->arch.pio.guest_page_offset;
-	bytes = vcpu->arch.pio.size * vcpu->arch.pio.cur_count;
-	if (vcpu->arch.pio.in)
-		memcpy(q, p, bytes);
-	else
-		memcpy(p, q, bytes);
-	q -= vcpu->arch.pio.guest_page_offset;
-	vunmap(q);
-	free_pio_guest_pages(vcpu);
-	return 0;
-}
-
-int complete_pio(struct kvm_vcpu *vcpu)
-{
-	struct kvm_pio_request *io = &vcpu->arch.pio;
-	long delta;
-	int r;
-
-	kvm_x86_ops->cache_regs(vcpu);
-
-	if (!io->string) {
-		if (io->in)
-			memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data,
-			       io->size);
-	} else {
-		if (io->in) {
-			r = pio_copy_data(vcpu);
-			if (r) {
-				kvm_x86_ops->cache_regs(vcpu);
-				return r;
-			}
-		}
-
-		delta = 1;
-		if (io->rep) {
-			delta *= io->cur_count;
-			/*
-			 * The size of the register should really depend on
-			 * current address size.
-			 */
-			vcpu->arch.regs[VCPU_REGS_RCX] -= delta;
-		}
-		if (io->down)
-			delta = -delta;
-		delta *= io->size;
-		if (io->in)
-			vcpu->arch.regs[VCPU_REGS_RDI] += delta;
-		else
-			vcpu->arch.regs[VCPU_REGS_RSI] += delta;
-	}
-
-	kvm_x86_ops->decache_regs(vcpu);
-
-	io->count -= io->cur_count;
-	io->cur_count = 0;
-
-	return 0;
-}
-
-static void kernel_pio(struct kvm_io_device *pio_dev,
-		       struct kvm_vcpu *vcpu,
-		       void *pd)
-{
-	/* TODO: String I/O for in kernel device */
-
-	mutex_lock(&vcpu->kvm->lock);
-	if (vcpu->arch.pio.in)
-		kvm_iodevice_read(pio_dev, vcpu->arch.pio.port,
-				  vcpu->arch.pio.size,
-				  pd);
-	else
-		kvm_iodevice_write(pio_dev, vcpu->arch.pio.port,
-				   vcpu->arch.pio.size,
-				   pd);
-	mutex_unlock(&vcpu->kvm->lock);
-}
-
-static void pio_string_write(struct kvm_io_device *pio_dev,
-			     struct kvm_vcpu *vcpu)
-{
-	struct kvm_pio_request *io = &vcpu->arch.pio;
-	void *pd = vcpu->arch.pio_data;
-	int i;
-
-	mutex_lock(&vcpu->kvm->lock);
-	for (i = 0; i < io->cur_count; i++) {
-		kvm_iodevice_write(pio_dev, io->port,
-				   io->size,
-				   pd);
-		pd += io->size;
-	}
-	mutex_unlock(&vcpu->kvm->lock);
-}
-
-static struct kvm_io_device *vcpu_find_pio_dev(struct kvm_vcpu *vcpu,
-					       gpa_t addr)
-{
-	return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr);
-}
-
-int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
-		  int size, unsigned port)
-{
-	struct kvm_io_device *pio_dev;
-
-	vcpu->run->exit_reason = KVM_EXIT_IO;
-	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
-	vcpu->run->io.size = vcpu->arch.pio.size = size;
-	vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
-	vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = 1;
-	vcpu->run->io.port = vcpu->arch.pio.port = port;
-	vcpu->arch.pio.in = in;
-	vcpu->arch.pio.string = 0;
-	vcpu->arch.pio.down = 0;
-	vcpu->arch.pio.guest_page_offset = 0;
-	vcpu->arch.pio.rep = 0;
-
-	kvm_x86_ops->cache_regs(vcpu);
-	memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4);
-	kvm_x86_ops->decache_regs(vcpu);
-
-	kvm_x86_ops->skip_emulated_instruction(vcpu);
-
-	pio_dev = vcpu_find_pio_dev(vcpu, port);
-	if (pio_dev) {
-		kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data);
-		complete_pio(vcpu);
-		return 1;
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(kvm_emulate_pio);
-
-int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
-		  int size, unsigned long count, int down,
-		  gva_t address, int rep, unsigned port)
-{
-	unsigned now, in_page;
-	int i, ret = 0;
-	int nr_pages = 1;
-	struct page *page;
-	struct kvm_io_device *pio_dev;
-
-	vcpu->run->exit_reason = KVM_EXIT_IO;
-	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
-	vcpu->run->io.size = vcpu->arch.pio.size = size;
-	vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
-	vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = count;
-	vcpu->run->io.port = vcpu->arch.pio.port = port;
-	vcpu->arch.pio.in = in;
-	vcpu->arch.pio.string = 1;
-	vcpu->arch.pio.down = down;
-	vcpu->arch.pio.guest_page_offset = offset_in_page(address);
-	vcpu->arch.pio.rep = rep;
-
-	if (!count) {
-		kvm_x86_ops->skip_emulated_instruction(vcpu);
-		return 1;
-	}
-
-	if (!down)
-		in_page = PAGE_SIZE - offset_in_page(address);
-	else
-		in_page = offset_in_page(address) + size;
-	now = min(count, (unsigned long)in_page / size);
-	if (!now) {
-		/*
-		 * String I/O straddles page boundary.  Pin two guest pages
-		 * so that we satisfy atomicity constraints.  Do just one
-		 * transaction to avoid complexity.
-		 */
-		nr_pages = 2;
-		now = 1;
-	}
-	if (down) {
-		/*
-		 * String I/O in reverse.  Yuck.  Kill the guest, fix later.
-		 */
-		pr_unimpl(vcpu, "guest string pio down\n");
-		kvm_inject_gp(vcpu, 0);
-		return 1;
-	}
-	vcpu->run->io.count = now;
-	vcpu->arch.pio.cur_count = now;
-
-	if (vcpu->arch.pio.cur_count == vcpu->arch.pio.count)
-		kvm_x86_ops->skip_emulated_instruction(vcpu);
-
-	for (i = 0; i < nr_pages; ++i) {
-		mutex_lock(&vcpu->kvm->lock);
-		page = gva_to_page(vcpu, address + i * PAGE_SIZE);
-		vcpu->arch.pio.guest_pages[i] = page;
-		mutex_unlock(&vcpu->kvm->lock);
-		if (!page) {
-			kvm_inject_gp(vcpu, 0);
-			free_pio_guest_pages(vcpu);
-			return 1;
-		}
-	}
-
-	pio_dev = vcpu_find_pio_dev(vcpu, port);
-	if (!vcpu->arch.pio.in) {
-		/* string PIO write */
-		ret = pio_copy_data(vcpu);
-		if (ret >= 0 && pio_dev) {
-			pio_string_write(pio_dev, vcpu);
-			complete_pio(vcpu);
-			if (vcpu->arch.pio.count == 0)
-				ret = 1;
-		}
-	} else if (pio_dev)
-		pr_unimpl(vcpu, "no string pio read support yet, "
-		       "port %x size %d count %ld\n",
-			port, size, count);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(kvm_emulate_pio_string);
-
-int kvm_arch_init(void *opaque)
-{
-	int r;
-	struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque;
-
-	r = kvm_mmu_module_init();
-	if (r)
-		goto out_fail;
-
-	kvm_init_msr_list();
-
-	if (kvm_x86_ops) {
-		printk(KERN_ERR "kvm: already loaded the other module\n");
-		r = -EEXIST;
-		goto out;
-	}
-
-	if (!ops->cpu_has_kvm_support()) {
-		printk(KERN_ERR "kvm: no hardware support\n");
-		r = -EOPNOTSUPP;
-		goto out;
-	}
-	if (ops->disabled_by_bios()) {
-		printk(KERN_ERR "kvm: disabled by bios\n");
-		r = -EOPNOTSUPP;
-		goto out;
-	}
-
-	kvm_x86_ops = ops;
-	kvm_mmu_set_nonpresent_ptes(0ull, 0ull);
-	return 0;
-
-out:
-	kvm_mmu_module_exit();
-out_fail:
-	return r;
-}
-
-void kvm_arch_exit(void)
-{
-	kvm_x86_ops = NULL;
-	kvm_mmu_module_exit();
-}
-
-int kvm_emulate_halt(struct kvm_vcpu *vcpu)
-{
-	++vcpu->stat.halt_exits;
-	if (irqchip_in_kernel(vcpu->kvm)) {
-		vcpu->arch.mp_state = VCPU_MP_STATE_HALTED;
-		kvm_vcpu_block(vcpu);
-		if (vcpu->arch.mp_state != VCPU_MP_STATE_RUNNABLE)
-			return -EINTR;
-		return 1;
-	} else {
-		vcpu->run->exit_reason = KVM_EXIT_HLT;
-		return 0;
-	}
-}
-EXPORT_SYMBOL_GPL(kvm_emulate_halt);
-
-int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
-{
-	unsigned long nr, a0, a1, a2, a3, ret;
-
-	kvm_x86_ops->cache_regs(vcpu);
-
-	nr = vcpu->arch.regs[VCPU_REGS_RAX];
-	a0 = vcpu->arch.regs[VCPU_REGS_RBX];
-	a1 = vcpu->arch.regs[VCPU_REGS_RCX];
-	a2 = vcpu->arch.regs[VCPU_REGS_RDX];
-	a3 = vcpu->arch.regs[VCPU_REGS_RSI];
-
-	if (!is_long_mode(vcpu)) {
-		nr &= 0xFFFFFFFF;
-		a0 &= 0xFFFFFFFF;
-		a1 &= 0xFFFFFFFF;
-		a2 &= 0xFFFFFFFF;
-		a3 &= 0xFFFFFFFF;
-	}
-
-	switch (nr) {
-	default:
-		ret = -KVM_ENOSYS;
-		break;
-	}
-	vcpu->arch.regs[VCPU_REGS_RAX] = ret;
-	kvm_x86_ops->decache_regs(vcpu);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(kvm_emulate_hypercall);
-
-int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
-{
-	char instruction[3];
-	int ret = 0;
-
-	mutex_lock(&vcpu->kvm->lock);
-
-	/*
-	 * Blow out the MMU to ensure that no other VCPU has an active mapping
-	 * to ensure that the updated hypercall appears atomically across all
-	 * VCPUs.
-	 */
-	kvm_mmu_zap_all(vcpu->kvm);
-
-	kvm_x86_ops->cache_regs(vcpu);
-	kvm_x86_ops->patch_hypercall(vcpu, instruction);
-	if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu)
-	    != X86EMUL_CONTINUE)
-		ret = -EFAULT;
-
-	mutex_unlock(&vcpu->kvm->lock);
-
-	return ret;
-}
-
-static u64 mk_cr_64(u64 curr_cr, u32 new_val)
-{
-	return (curr_cr & ~((1ULL << 32) - 1)) | new_val;
-}
-
-void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
-{
-	struct descriptor_table dt = { limit, base };
-
-	kvm_x86_ops->set_gdt(vcpu, &dt);
-}
-
-void realmode_lidt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
-{
-	struct descriptor_table dt = { limit, base };
-
-	kvm_x86_ops->set_idt(vcpu, &dt);
-}
-
-void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw,
-		   unsigned long *rflags)
-{
-	lmsw(vcpu, msw);
-	*rflags = kvm_x86_ops->get_rflags(vcpu);
-}
-
-unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr)
-{
-	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
-	switch (cr) {
-	case 0:
-		return vcpu->arch.cr0;
-	case 2:
-		return vcpu->arch.cr2;
-	case 3:
-		return vcpu->arch.cr3;
-	case 4:
-		return vcpu->arch.cr4;
-	case 8:
-		return get_cr8(vcpu);
-	default:
-		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
-		return 0;
-	}
-}
-
-void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
-		     unsigned long *rflags)
-{
-	switch (cr) {
-	case 0:
-		set_cr0(vcpu, mk_cr_64(vcpu->arch.cr0, val));
-		*rflags = kvm_x86_ops->get_rflags(vcpu);
-		break;
-	case 2:
-		vcpu->arch.cr2 = val;
-		break;
-	case 3:
-		set_cr3(vcpu, val);
-		break;
-	case 4:
-		set_cr4(vcpu, mk_cr_64(vcpu->arch.cr4, val));
-		break;
-	case 8:
-		set_cr8(vcpu, val & 0xfUL);
-		break;
-	default:
-		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
-	}
-}
-
-static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i)
-{
-	struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i];
-	int j, nent = vcpu->arch.cpuid_nent;
-
-	e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
-	/* when no next entry is found, the current entry[i] is reselected */
-	for (j = i + 1; j == i; j = (j + 1) % nent) {
-		struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j];
-		if (ej->function == e->function) {
-			ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
-			return j;
-		}
-	}
-	return 0; /* silence gcc, even though control never reaches here */
-}
-
-/* find an entry with matching function, matching index (if needed), and that
- * should be read next (if it's stateful) */
-static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e,
-	u32 function, u32 index)
-{
-	if (e->function != function)
-		return 0;
-	if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index)
-		return 0;
-	if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) &&
-		!(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
-		return 0;
-	return 1;
-}
-
-void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
-{
-	int i;
-	u32 function, index;
-	struct kvm_cpuid_entry2 *e, *best;
-
-	kvm_x86_ops->cache_regs(vcpu);
-	function = vcpu->arch.regs[VCPU_REGS_RAX];
-	index = vcpu->arch.regs[VCPU_REGS_RCX];
-	vcpu->arch.regs[VCPU_REGS_RAX] = 0;
-	vcpu->arch.regs[VCPU_REGS_RBX] = 0;
-	vcpu->arch.regs[VCPU_REGS_RCX] = 0;
-	vcpu->arch.regs[VCPU_REGS_RDX] = 0;
-	best = NULL;
-	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
-		e = &vcpu->arch.cpuid_entries[i];
-		if (is_matching_cpuid_entry(e, function, index)) {
-			if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
-				move_to_next_stateful_cpuid_entry(vcpu, i);
-			best = e;
-			break;
-		}
-		/*
-		 * Both basic or both extended?
-		 */
-		if (((e->function ^ function) & 0x80000000) == 0)
-			if (!best || e->function > best->function)
-				best = e;
-	}
-	if (best) {
-		vcpu->arch.regs[VCPU_REGS_RAX] = best->eax;
-		vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx;
-		vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx;
-		vcpu->arch.regs[VCPU_REGS_RDX] = best->edx;
-	}
-	kvm_x86_ops->decache_regs(vcpu);
-	kvm_x86_ops->skip_emulated_instruction(vcpu);
-}
-EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);
-
-/*
- * Check if userspace requested an interrupt window, and that the
- * interrupt window is open.
- *
- * No need to exit to userspace if we already have an interrupt queued.
- */
-static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu,
-					  struct kvm_run *kvm_run)
-{
-	return (!vcpu->arch.irq_summary &&
-		kvm_run->request_interrupt_window &&
-		vcpu->arch.interrupt_window_open &&
-		(kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF));
-}
-
-static void post_kvm_run_save(struct kvm_vcpu *vcpu,
-			      struct kvm_run *kvm_run)
-{
-	kvm_run->if_flag = (kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
-	kvm_run->cr8 = get_cr8(vcpu);
-	kvm_run->apic_base = kvm_get_apic_base(vcpu);
-	if (irqchip_in_kernel(vcpu->kvm))
-		kvm_run->ready_for_interrupt_injection = 1;
-	else
-		kvm_run->ready_for_interrupt_injection =
-					(vcpu->arch.interrupt_window_open &&
-					 vcpu->arch.irq_summary == 0);
-}
-
-static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	int r;
-
-	if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED)) {
-		pr_debug("vcpu %d received sipi with vector # %x\n",
-		       vcpu->vcpu_id, vcpu->arch.sipi_vector);
-		kvm_lapic_reset(vcpu);
-		r = kvm_x86_ops->vcpu_reset(vcpu);
-		if (r)
-			return r;
-		vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
-	}
-
-preempted:
-	if (vcpu->guest_debug.enabled)
-		kvm_x86_ops->guest_debug_pre(vcpu);
-
-again:
-	r = kvm_mmu_reload(vcpu);
-	if (unlikely(r))
-		goto out;
-
-	kvm_inject_pending_timer_irqs(vcpu);
-
-	preempt_disable();
-
-	kvm_x86_ops->prepare_guest_switch(vcpu);
-	kvm_load_guest_fpu(vcpu);
-
-	local_irq_disable();
-
-	if (signal_pending(current)) {
-		local_irq_enable();
-		preempt_enable();
-		r = -EINTR;
-		kvm_run->exit_reason = KVM_EXIT_INTR;
-		++vcpu->stat.signal_exits;
-		goto out;
-	}
-
-	if (vcpu->arch.exception.pending)
-		__queue_exception(vcpu);
-	else if (irqchip_in_kernel(vcpu->kvm))
-		kvm_x86_ops->inject_pending_irq(vcpu);
-	else
-		kvm_x86_ops->inject_pending_vectors(vcpu, kvm_run);
-
-	vcpu->guest_mode = 1;
-	kvm_guest_enter();
-
-	if (vcpu->requests)
-		if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
-			kvm_x86_ops->tlb_flush(vcpu);
-
-	kvm_x86_ops->run(vcpu, kvm_run);
-
-	vcpu->guest_mode = 0;
-	local_irq_enable();
-
-	++vcpu->stat.exits;
-
-	/*
-	 * We must have an instruction between local_irq_enable() and
-	 * kvm_guest_exit(), so the timer interrupt isn't delayed by
-	 * the interrupt shadow.  The stat.exits increment will do nicely.
-	 * But we need to prevent reordering, hence this barrier():
-	 */
-	barrier();
-
-	kvm_guest_exit();
-
-	preempt_enable();
-
-	/*
-	 * Profile KVM exit RIPs:
-	 */
-	if (unlikely(prof_on == KVM_PROFILING)) {
-		kvm_x86_ops->cache_regs(vcpu);
-		profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip);
-	}
-
-	if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
-		vcpu->arch.exception.pending = false;
-
-	r = kvm_x86_ops->handle_exit(kvm_run, vcpu);
-
-	if (r > 0) {
-		if (dm_request_for_irq_injection(vcpu, kvm_run)) {
-			r = -EINTR;
-			kvm_run->exit_reason = KVM_EXIT_INTR;
-			++vcpu->stat.request_irq_exits;
-			goto out;
-		}
-		if (!need_resched())
-			goto again;
-	}
-
-out:
-	if (r > 0) {
-		kvm_resched(vcpu);
-		goto preempted;
-	}
-
-	post_kvm_run_save(vcpu, kvm_run);
-
-	return r;
-}
-
-int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
-{
-	int r;
-	sigset_t sigsaved;
-
-	vcpu_load(vcpu);
-
-	if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_UNINITIALIZED)) {
-		kvm_vcpu_block(vcpu);
-		vcpu_put(vcpu);
-		return -EAGAIN;
-	}
-
-	if (vcpu->sigset_active)
-		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
-
-	/* re-sync apic's tpr */
-	if (!irqchip_in_kernel(vcpu->kvm))
-		set_cr8(vcpu, kvm_run->cr8);
-
-	if (vcpu->arch.pio.cur_count) {
-		r = complete_pio(vcpu);
-		if (r)
-			goto out;
-	}
-#if CONFIG_HAS_IOMEM
-	if (vcpu->mmio_needed) {
-		memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8);
-		vcpu->mmio_read_completed = 1;
-		vcpu->mmio_needed = 0;
-		r = emulate_instruction(vcpu, kvm_run,
-					vcpu->arch.mmio_fault_cr2, 0, 1);
-		if (r == EMULATE_DO_MMIO) {
-			/*
-			 * Read-modify-write.  Back to userspace.
-			 */
-			r = 0;
-			goto out;
-		}
-	}
-#endif
-	if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) {
-		kvm_x86_ops->cache_regs(vcpu);
-		vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
-		kvm_x86_ops->decache_regs(vcpu);
-	}
-
-	r = __vcpu_run(vcpu, kvm_run);
-
-out:
-	if (vcpu->sigset_active)
-		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
-
-	vcpu_put(vcpu);
-	return r;
-}
-
-int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
-{
-	vcpu_load(vcpu);
-
-	kvm_x86_ops->cache_regs(vcpu);
-
-	regs->rax = vcpu->arch.regs[VCPU_REGS_RAX];
-	regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX];
-	regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX];
-	regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX];
-	regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI];
-	regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI];
-	regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP];
-	regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP];
-#ifdef CONFIG_X86_64
-	regs->r8 = vcpu->arch.regs[VCPU_REGS_R8];
-	regs->r9 = vcpu->arch.regs[VCPU_REGS_R9];
-	regs->r10 = vcpu->arch.regs[VCPU_REGS_R10];
-	regs->r11 = vcpu->arch.regs[VCPU_REGS_R11];
-	regs->r12 = vcpu->arch.regs[VCPU_REGS_R12];
-	regs->r13 = vcpu->arch.regs[VCPU_REGS_R13];
-	regs->r14 = vcpu->arch.regs[VCPU_REGS_R14];
-	regs->r15 = vcpu->arch.regs[VCPU_REGS_R15];
-#endif
-
-	regs->rip = vcpu->arch.rip;
-	regs->rflags = kvm_x86_ops->get_rflags(vcpu);
-
-	/*
-	 * Don't leak debug flags in case they were set for guest debugging
-	 */
-	if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep)
-		regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
-{
-	vcpu_load(vcpu);
-
-	vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax;
-	vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx;
-	vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx;
-	vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx;
-	vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi;
-	vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi;
-	vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp;
-	vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp;
-#ifdef CONFIG_X86_64
-	vcpu->arch.regs[VCPU_REGS_R8] = regs->r8;
-	vcpu->arch.regs[VCPU_REGS_R9] = regs->r9;
-	vcpu->arch.regs[VCPU_REGS_R10] = regs->r10;
-	vcpu->arch.regs[VCPU_REGS_R11] = regs->r11;
-	vcpu->arch.regs[VCPU_REGS_R12] = regs->r12;
-	vcpu->arch.regs[VCPU_REGS_R13] = regs->r13;
-	vcpu->arch.regs[VCPU_REGS_R14] = regs->r14;
-	vcpu->arch.regs[VCPU_REGS_R15] = regs->r15;
-#endif
-
-	vcpu->arch.rip = regs->rip;
-	kvm_x86_ops->set_rflags(vcpu, regs->rflags);
-
-	kvm_x86_ops->decache_regs(vcpu);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-static void get_segment(struct kvm_vcpu *vcpu,
-			struct kvm_segment *var, int seg)
-{
-	return kvm_x86_ops->get_segment(vcpu, var, seg);
-}
-
-void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l)
-{
-	struct kvm_segment cs;
-
-	get_segment(vcpu, &cs, VCPU_SREG_CS);
-	*db = cs.db;
-	*l = cs.l;
-}
-EXPORT_SYMBOL_GPL(kvm_get_cs_db_l_bits);
-
-int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs)
-{
-	struct descriptor_table dt;
-	int pending_vec;
-
-	vcpu_load(vcpu);
-
-	get_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
-	get_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
-	get_segment(vcpu, &sregs->es, VCPU_SREG_ES);
-	get_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
-	get_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
-	get_segment(vcpu, &sregs->ss, VCPU_SREG_SS);
-
-	get_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
-	get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);
-
-	kvm_x86_ops->get_idt(vcpu, &dt);
-	sregs->idt.limit = dt.limit;
-	sregs->idt.base = dt.base;
-	kvm_x86_ops->get_gdt(vcpu, &dt);
-	sregs->gdt.limit = dt.limit;
-	sregs->gdt.base = dt.base;
-
-	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
-	sregs->cr0 = vcpu->arch.cr0;
-	sregs->cr2 = vcpu->arch.cr2;
-	sregs->cr3 = vcpu->arch.cr3;
-	sregs->cr4 = vcpu->arch.cr4;
-	sregs->cr8 = get_cr8(vcpu);
-	sregs->efer = vcpu->arch.shadow_efer;
-	sregs->apic_base = kvm_get_apic_base(vcpu);
-
-	if (irqchip_in_kernel(vcpu->kvm)) {
-		memset(sregs->interrupt_bitmap, 0,
-		       sizeof sregs->interrupt_bitmap);
-		pending_vec = kvm_x86_ops->get_irq(vcpu);
-		if (pending_vec >= 0)
-			set_bit(pending_vec,
-				(unsigned long *)sregs->interrupt_bitmap);
-	} else
-		memcpy(sregs->interrupt_bitmap, vcpu->arch.irq_pending,
-		       sizeof sregs->interrupt_bitmap);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-static void set_segment(struct kvm_vcpu *vcpu,
-			struct kvm_segment *var, int seg)
-{
-	return kvm_x86_ops->set_segment(vcpu, var, seg);
-}
-
-int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs)
-{
-	int mmu_reset_needed = 0;
-	int i, pending_vec, max_bits;
-	struct descriptor_table dt;
-
-	vcpu_load(vcpu);
-
-	dt.limit = sregs->idt.limit;
-	dt.base = sregs->idt.base;
-	kvm_x86_ops->set_idt(vcpu, &dt);
-	dt.limit = sregs->gdt.limit;
-	dt.base = sregs->gdt.base;
-	kvm_x86_ops->set_gdt(vcpu, &dt);
-
-	vcpu->arch.cr2 = sregs->cr2;
-	mmu_reset_needed |= vcpu->arch.cr3 != sregs->cr3;
-	vcpu->arch.cr3 = sregs->cr3;
-
-	set_cr8(vcpu, sregs->cr8);
-
-	mmu_reset_needed |= vcpu->arch.shadow_efer != sregs->efer;
-#ifdef CONFIG_X86_64
-	kvm_x86_ops->set_efer(vcpu, sregs->efer);
-#endif
-	kvm_set_apic_base(vcpu, sregs->apic_base);
-
-	kvm_x86_ops->decache_cr4_guest_bits(vcpu);
-
-	mmu_reset_needed |= vcpu->arch.cr0 != sregs->cr0;
-	vcpu->arch.cr0 = sregs->cr0;
-	kvm_x86_ops->set_cr0(vcpu, sregs->cr0);
-
-	mmu_reset_needed |= vcpu->arch.cr4 != sregs->cr4;
-	kvm_x86_ops->set_cr4(vcpu, sregs->cr4);
-	if (!is_long_mode(vcpu) && is_pae(vcpu))
-		load_pdptrs(vcpu, vcpu->arch.cr3);
-
-	if (mmu_reset_needed)
-		kvm_mmu_reset_context(vcpu);
-
-	if (!irqchip_in_kernel(vcpu->kvm)) {
-		memcpy(vcpu->arch.irq_pending, sregs->interrupt_bitmap,
-		       sizeof vcpu->arch.irq_pending);
-		vcpu->arch.irq_summary = 0;
-		for (i = 0; i < ARRAY_SIZE(vcpu->arch.irq_pending); ++i)
-			if (vcpu->arch.irq_pending[i])
-				__set_bit(i, &vcpu->arch.irq_summary);
-	} else {
-		max_bits = (sizeof sregs->interrupt_bitmap) << 3;
-		pending_vec = find_first_bit(
-			(const unsigned long *)sregs->interrupt_bitmap,
-			max_bits);
-		/* Only pending external irq is handled here */
-		if (pending_vec < max_bits) {
-			kvm_x86_ops->set_irq(vcpu, pending_vec);
-			pr_debug("Set back pending irq %d\n",
-				 pending_vec);
-		}
-	}
-
-	set_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
-	set_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
-	set_segment(vcpu, &sregs->es, VCPU_SREG_ES);
-	set_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
-	set_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
-	set_segment(vcpu, &sregs->ss, VCPU_SREG_SS);
-
-	set_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
-	set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg)
-{
-	int r;
-
-	vcpu_load(vcpu);
-
-	r = kvm_x86_ops->set_guest_debug(vcpu, dbg);
-
-	vcpu_put(vcpu);
-
-	return r;
-}
-
-/*
- * fxsave fpu state.  Taken from x86_64/processor.h.  To be killed when
- * we have asm/x86/processor.h
- */
-struct fxsave {
-	u16	cwd;
-	u16	swd;
-	u16	twd;
-	u16	fop;
-	u64	rip;
-	u64	rdp;
-	u32	mxcsr;
-	u32	mxcsr_mask;
-	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
-#ifdef CONFIG_X86_64
-	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
-#else
-	u32	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
-#endif
-};
-
-/*
- * Translate a guest virtual address to a guest physical address.
- */
-int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
-				    struct kvm_translation *tr)
-{
-	unsigned long vaddr = tr->linear_address;
-	gpa_t gpa;
-
-	vcpu_load(vcpu);
-	mutex_lock(&vcpu->kvm->lock);
-	gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, vaddr);
-	tr->physical_address = gpa;
-	tr->valid = gpa != UNMAPPED_GVA;
-	tr->writeable = 1;
-	tr->usermode = 0;
-	mutex_unlock(&vcpu->kvm->lock);
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
-{
-	struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
-
-	vcpu_load(vcpu);
-
-	memcpy(fpu->fpr, fxsave->st_space, 128);
-	fpu->fcw = fxsave->cwd;
-	fpu->fsw = fxsave->swd;
-	fpu->ftwx = fxsave->twd;
-	fpu->last_opcode = fxsave->fop;
-	fpu->last_ip = fxsave->rip;
-	fpu->last_dp = fxsave->rdp;
-	memcpy(fpu->xmm, fxsave->xmm_space, sizeof fxsave->xmm_space);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
-{
-	struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
-
-	vcpu_load(vcpu);
-
-	memcpy(fxsave->st_space, fpu->fpr, 128);
-	fxsave->cwd = fpu->fcw;
-	fxsave->swd = fpu->fsw;
-	fxsave->twd = fpu->ftwx;
-	fxsave->fop = fpu->last_opcode;
-	fxsave->rip = fpu->last_ip;
-	fxsave->rdp = fpu->last_dp;
-	memcpy(fxsave->xmm_space, fpu->xmm, sizeof fxsave->xmm_space);
-
-	vcpu_put(vcpu);
-
-	return 0;
-}
-
-void fx_init(struct kvm_vcpu *vcpu)
-{
-	unsigned after_mxcsr_mask;
-
-	/* Initialize guest FPU by resetting ours and saving into guest's */
-	preempt_disable();
-	fx_save(&vcpu->arch.host_fx_image);
-	fpu_init();
-	fx_save(&vcpu->arch.guest_fx_image);
-	fx_restore(&vcpu->arch.host_fx_image);
-	preempt_enable();
-
-	vcpu->arch.cr0 |= X86_CR0_ET;
-	after_mxcsr_mask = offsetof(struct i387_fxsave_struct, st_space);
-	vcpu->arch.guest_fx_image.mxcsr = 0x1f80;
-	memset((void *)&vcpu->arch.guest_fx_image + after_mxcsr_mask,
-	       0, sizeof(struct i387_fxsave_struct) - after_mxcsr_mask);
-}
-EXPORT_SYMBOL_GPL(fx_init);
-
-void kvm_load_guest_fpu(struct kvm_vcpu *vcpu)
-{
-	if (!vcpu->fpu_active || vcpu->guest_fpu_loaded)
-		return;
-
-	vcpu->guest_fpu_loaded = 1;
-	fx_save(&vcpu->arch.host_fx_image);
-	fx_restore(&vcpu->arch.guest_fx_image);
-}
-EXPORT_SYMBOL_GPL(kvm_load_guest_fpu);
-
-void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
-{
-	if (!vcpu->guest_fpu_loaded)
-		return;
-
-	vcpu->guest_fpu_loaded = 0;
-	fx_save(&vcpu->arch.guest_fx_image);
-	fx_restore(&vcpu->arch.host_fx_image);
-	++vcpu->stat.fpu_reload;
-}
-EXPORT_SYMBOL_GPL(kvm_put_guest_fpu);
-
-void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
-{
-	kvm_x86_ops->vcpu_free(vcpu);
-}
-
-struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
-						unsigned int id)
-{
-	return kvm_x86_ops->vcpu_create(kvm, id);
-}
-
-int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
-{
-	int r;
-
-	/* We do fxsave: this must be aligned. */
-	BUG_ON((unsigned long)&vcpu->arch.host_fx_image & 0xF);
-
-	vcpu_load(vcpu);
-	r = kvm_arch_vcpu_reset(vcpu);
-	if (r == 0)
-		r = kvm_mmu_setup(vcpu);
-	vcpu_put(vcpu);
-	if (r < 0)
-		goto free_vcpu;
-
-	return 0;
-free_vcpu:
-	kvm_x86_ops->vcpu_free(vcpu);
-	return r;
-}
-
-void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
-{
-	vcpu_load(vcpu);
-	kvm_mmu_unload(vcpu);
-	vcpu_put(vcpu);
-
-	kvm_x86_ops->vcpu_free(vcpu);
-}
-
-int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu)
-{
-	return kvm_x86_ops->vcpu_reset(vcpu);
-}
-
-void kvm_arch_hardware_enable(void *garbage)
-{
-	kvm_x86_ops->hardware_enable(garbage);
-}
-
-void kvm_arch_hardware_disable(void *garbage)
-{
-	kvm_x86_ops->hardware_disable(garbage);
-}
-
-int kvm_arch_hardware_setup(void)
-{
-	return kvm_x86_ops->hardware_setup();
-}
-
-void kvm_arch_hardware_unsetup(void)
-{
-	kvm_x86_ops->hardware_unsetup();
-}
-
-void kvm_arch_check_processor_compat(void *rtn)
-{
-	kvm_x86_ops->check_processor_compatibility(rtn);
-}
-
-int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
-{
-	struct page *page;
-	struct kvm *kvm;
-	int r;
-
-	BUG_ON(vcpu->kvm == NULL);
-	kvm = vcpu->kvm;
-
-	vcpu->arch.mmu.root_hpa = INVALID_PAGE;
-	if (!irqchip_in_kernel(kvm) || vcpu->vcpu_id == 0)
-		vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
-	else
-		vcpu->arch.mp_state = VCPU_MP_STATE_UNINITIALIZED;
-
-	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
-	if (!page) {
-		r = -ENOMEM;
-		goto fail;
-	}
-	vcpu->arch.pio_data = page_address(page);
-
-	r = kvm_mmu_create(vcpu);
-	if (r < 0)
-		goto fail_free_pio_data;
-
-	if (irqchip_in_kernel(kvm)) {
-		r = kvm_create_lapic(vcpu);
-		if (r < 0)
-			goto fail_mmu_destroy;
-	}
-
-	return 0;
-
-fail_mmu_destroy:
-	kvm_mmu_destroy(vcpu);
-fail_free_pio_data:
-	free_page((unsigned long)vcpu->arch.pio_data);
-fail:
-	return r;
-}
-
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-	kvm_free_lapic(vcpu);
-	kvm_mmu_destroy(vcpu);
-	free_page((unsigned long)vcpu->arch.pio_data);
-}
-
-struct  kvm *kvm_arch_create_vm(void)
-{
-	struct kvm *kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL);
-
-	if (!kvm)
-		return ERR_PTR(-ENOMEM);
-
-	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
-
-	return kvm;
-}
-
-static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu)
-{
-	vcpu_load(vcpu);
-	kvm_mmu_unload(vcpu);
-	vcpu_put(vcpu);
-}
-
-static void kvm_free_vcpus(struct kvm *kvm)
-{
-	unsigned int i;
-
-	/*
-	 * Unpin any mmu pages first.
-	 */
-	for (i = 0; i < KVM_MAX_VCPUS; ++i)
-		if (kvm->vcpus[i])
-			kvm_unload_vcpu_mmu(kvm->vcpus[i]);
-	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-		if (kvm->vcpus[i]) {
-			kvm_arch_vcpu_free(kvm->vcpus[i]);
-			kvm->vcpus[i] = NULL;
-		}
-	}
-
-}
-
-void kvm_arch_destroy_vm(struct kvm *kvm)
-{
-	kfree(kvm->arch.vpic);
-	kfree(kvm->arch.vioapic);
-	kvm_free_vcpus(kvm);
-	kvm_free_physmem(kvm);
-	kfree(kvm);
-}
-
-int kvm_arch_set_memory_region(struct kvm *kvm,
-				struct kvm_userspace_memory_region *mem,
-				struct kvm_memory_slot old,
-				int user_alloc)
-{
-	int npages = mem->memory_size >> PAGE_SHIFT;
-	struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot];
-
-	/*To keep backward compatibility with older userspace,
-	 *x86 needs to hanlde !user_alloc case.
-	 */
-	if (!user_alloc) {
-		if (npages && !old.rmap) {
-			down_write(&current->mm->mmap_sem);
-			memslot->userspace_addr = do_mmap(NULL, 0,
-						     npages * PAGE_SIZE,
-						     PROT_READ | PROT_WRITE,
-						     MAP_SHARED | MAP_ANONYMOUS,
-						     0);
-			up_write(&current->mm->mmap_sem);
-
-			if (IS_ERR((void *)memslot->userspace_addr))
-				return PTR_ERR((void *)memslot->userspace_addr);
-		} else {
-			if (!old.user_alloc && old.rmap) {
-				int ret;
-
-				down_write(&current->mm->mmap_sem);
-				ret = do_munmap(current->mm, old.userspace_addr,
-						old.npages * PAGE_SIZE);
-				up_write(&current->mm->mmap_sem);
-				if (ret < 0)
-					printk(KERN_WARNING
-				       "kvm_vm_ioctl_set_memory_region: "
-				       "failed to munmap memory\n");
-			}
-		}
-	}
-
-	if (!kvm->arch.n_requested_mmu_pages) {
-		unsigned int nr_mmu_pages = kvm_mmu_calculate_mmu_pages(kvm);
-		kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
-	}
-
-	kvm_mmu_slot_remove_write_access(kvm, mem->slot);
-	kvm_flush_remote_tlbs(kvm);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
-	       || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
-}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
deleted file mode 100644
index dfb8091971a9..000000000000
--- a/drivers/kvm/x86.h
+++ /dev/null
@@ -1,602 +0,0 @@
-#/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * This header defines architecture specific interfaces, x86 version
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- */
-
-#ifndef KVM_X86_H
-#define KVM_X86_H
-
-#include <linux/types.h>
-#include <linux/mm.h>
-
-#include <linux/kvm.h>
-#include <linux/kvm_para.h>
-
-#include <asm/desc.h>
-
-#include "types.h"
-
-#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
-#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
-#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS|0xFFFFFF0000000000ULL)
-
-#define KVM_GUEST_CR0_MASK \
-	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE \
-	 | X86_CR0_NW | X86_CR0_CD)
-#define KVM_VM_CR0_ALWAYS_ON \
-	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE | X86_CR0_TS \
-	 | X86_CR0_MP)
-#define KVM_GUEST_CR4_MASK \
-	(X86_CR4_VME | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE | X86_CR4_VMXE)
-#define KVM_PMODE_VM_CR4_ALWAYS_ON (X86_CR4_PAE | X86_CR4_VMXE)
-#define KVM_RMODE_VM_CR4_ALWAYS_ON (X86_CR4_VME | X86_CR4_PAE | X86_CR4_VMXE)
-
-#define INVALID_PAGE (~(hpa_t)0)
-#define UNMAPPED_GVA (~(gpa_t)0)
-
-#define DE_VECTOR 0
-#define UD_VECTOR 6
-#define NM_VECTOR 7
-#define DF_VECTOR 8
-#define TS_VECTOR 10
-#define NP_VECTOR 11
-#define SS_VECTOR 12
-#define GP_VECTOR 13
-#define PF_VECTOR 14
-
-#define SELECTOR_TI_MASK (1 << 2)
-#define SELECTOR_RPL_MASK 0x03
-
-#define IOPL_SHIFT 12
-
-#define KVM_ALIAS_SLOTS 4
-
-#define KVM_PERMILLE_MMU_PAGES 20
-#define KVM_MIN_ALLOC_MMU_PAGES 64
-#define KVM_NUM_MMU_PAGES 1024
-#define KVM_MIN_FREE_MMU_PAGES 5
-#define KVM_REFILL_PAGES 25
-#define KVM_MAX_CPUID_ENTRIES 40
-
-extern spinlock_t kvm_lock;
-extern struct list_head vm_list;
-
-struct kvm_vcpu;
-struct kvm;
-
-enum {
-	VCPU_REGS_RAX = 0,
-	VCPU_REGS_RCX = 1,
-	VCPU_REGS_RDX = 2,
-	VCPU_REGS_RBX = 3,
-	VCPU_REGS_RSP = 4,
-	VCPU_REGS_RBP = 5,
-	VCPU_REGS_RSI = 6,
-	VCPU_REGS_RDI = 7,
-#ifdef CONFIG_X86_64
-	VCPU_REGS_R8 = 8,
-	VCPU_REGS_R9 = 9,
-	VCPU_REGS_R10 = 10,
-	VCPU_REGS_R11 = 11,
-	VCPU_REGS_R12 = 12,
-	VCPU_REGS_R13 = 13,
-	VCPU_REGS_R14 = 14,
-	VCPU_REGS_R15 = 15,
-#endif
-	NR_VCPU_REGS
-};
-
-enum {
-	VCPU_SREG_CS,
-	VCPU_SREG_DS,
-	VCPU_SREG_ES,
-	VCPU_SREG_FS,
-	VCPU_SREG_GS,
-	VCPU_SREG_SS,
-	VCPU_SREG_TR,
-	VCPU_SREG_LDTR,
-};
-
-#include "x86_emulate.h"
-
-#define KVM_NR_MEM_OBJS 40
-
-/*
- * We don't want allocation failures within the mmu code, so we preallocate
- * enough memory for a single page fault in a cache.
- */
-struct kvm_mmu_memory_cache {
-	int nobjs;
-	void *objects[KVM_NR_MEM_OBJS];
-};
-
-#define NR_PTE_CHAIN_ENTRIES 5
-
-struct kvm_pte_chain {
-	u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
-	struct hlist_node link;
-};
-
-/*
- * kvm_mmu_page_role, below, is defined as:
- *
- *   bits 0:3 - total guest paging levels (2-4, or zero for real mode)
- *   bits 4:7 - page table level for this shadow (1-4)
- *   bits 8:9 - page table quadrant for 2-level guests
- *   bit   16 - "metaphysical" - gfn is not a real page (huge page/real mode)
- *   bits 17:19 - common access permissions for all ptes in this shadow page
- */
-union kvm_mmu_page_role {
-	unsigned word;
-	struct {
-		unsigned glevels : 4;
-		unsigned level : 4;
-		unsigned quadrant : 2;
-		unsigned pad_for_nice_hex_output : 6;
-		unsigned metaphysical : 1;
-		unsigned access : 3;
-	};
-};
-
-struct kvm_mmu_page {
-	struct list_head link;
-	struct hlist_node hash_link;
-
-	/*
-	 * The following two entries are used to key the shadow page in the
-	 * hash table.
-	 */
-	gfn_t gfn;
-	union kvm_mmu_page_role role;
-
-	u64 *spt;
-	/* hold the gfn of each spte inside spt */
-	gfn_t *gfns;
-	unsigned long slot_bitmap; /* One bit set per slot which has memory
-				    * in this shadow page.
-				    */
-	int multimapped;         /* More than one parent_pte? */
-	int root_count;          /* Currently serving as active root */
-	union {
-		u64 *parent_pte;               /* !multimapped */
-		struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
-	};
-};
-
-/*
- * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
- * 32-bit).  The kvm_mmu structure abstracts the details of the current mmu
- * mode.
- */
-struct kvm_mmu {
-	void (*new_cr3)(struct kvm_vcpu *vcpu);
-	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err);
-	void (*free)(struct kvm_vcpu *vcpu);
-	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva);
-	void (*prefetch_page)(struct kvm_vcpu *vcpu,
-			      struct kvm_mmu_page *page);
-	hpa_t root_hpa;
-	int root_level;
-	int shadow_root_level;
-
-	u64 *pae_root;
-};
-
-struct kvm_vcpu_arch {
-	u64 host_tsc;
-	int interrupt_window_open;
-	unsigned long irq_summary; /* bit vector: 1 per word in irq_pending */
-	DECLARE_BITMAP(irq_pending, KVM_NR_INTERRUPTS);
-	unsigned long regs[NR_VCPU_REGS]; /* for rsp: vcpu_load_rsp_rip() */
-	unsigned long rip;      /* needs vcpu_load_rsp_rip() */
-
-	unsigned long cr0;
-	unsigned long cr2;
-	unsigned long cr3;
-	unsigned long cr4;
-	unsigned long cr8;
-	u64 pdptrs[4]; /* pae */
-	u64 shadow_efer;
-	u64 apic_base;
-	struct kvm_lapic *apic;    /* kernel irqchip context */
-#define VCPU_MP_STATE_RUNNABLE          0
-#define VCPU_MP_STATE_UNINITIALIZED     1
-#define VCPU_MP_STATE_INIT_RECEIVED     2
-#define VCPU_MP_STATE_SIPI_RECEIVED     3
-#define VCPU_MP_STATE_HALTED            4
-	int mp_state;
-	int sipi_vector;
-	u64 ia32_misc_enable_msr;
-
-	struct kvm_mmu mmu;
-
-	struct kvm_mmu_memory_cache mmu_pte_chain_cache;
-	struct kvm_mmu_memory_cache mmu_rmap_desc_cache;
-	struct kvm_mmu_memory_cache mmu_page_cache;
-	struct kvm_mmu_memory_cache mmu_page_header_cache;
-
-	gfn_t last_pt_write_gfn;
-	int   last_pt_write_count;
-	u64  *last_pte_updated;
-
-	struct i387_fxsave_struct host_fx_image;
-	struct i387_fxsave_struct guest_fx_image;
-
-	gva_t mmio_fault_cr2;
-	struct kvm_pio_request pio;
-	void *pio_data;
-
-	struct kvm_queued_exception {
-		bool pending;
-		bool has_error_code;
-		u8 nr;
-		u32 error_code;
-	} exception;
-
-	struct {
-		int active;
-		u8 save_iopl;
-		struct kvm_save_segment {
-			u16 selector;
-			unsigned long base;
-			u32 limit;
-			u32 ar;
-		} tr, es, ds, fs, gs;
-	} rmode;
-	int halt_request; /* real mode on Intel only */
-
-	int cpuid_nent;
-	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
-	/* emulate context */
-
-	struct x86_emulate_ctxt emulate_ctxt;
-};
-
-struct kvm_mem_alias {
-	gfn_t base_gfn;
-	unsigned long npages;
-	gfn_t target_gfn;
-};
-
-struct kvm_arch{
-	int naliases;
-	struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
-
-	unsigned int n_free_mmu_pages;
-	unsigned int n_requested_mmu_pages;
-	unsigned int n_alloc_mmu_pages;
-	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
-	/*
-	 * Hash table of struct kvm_mmu_page.
-	 */
-	struct list_head active_mmu_pages;
-	struct kvm_pic *vpic;
-	struct kvm_ioapic *vioapic;
-
-	int round_robin_prev_vcpu;
-	unsigned int tss_addr;
-	struct page *apic_access_page;
-};
-
-struct kvm_vm_stat {
-	u32 mmu_shadow_zapped;
-	u32 mmu_pte_write;
-	u32 mmu_pte_updated;
-	u32 mmu_pde_zapped;
-	u32 mmu_flooded;
-	u32 mmu_recycled;
-	u32 remote_tlb_flush;
-};
-
-struct kvm_vcpu_stat {
-	u32 pf_fixed;
-	u32 pf_guest;
-	u32 tlb_flush;
-	u32 invlpg;
-
-	u32 exits;
-	u32 io_exits;
-	u32 mmio_exits;
-	u32 signal_exits;
-	u32 irq_window_exits;
-	u32 halt_exits;
-	u32 halt_wakeup;
-	u32 request_irq_exits;
-	u32 irq_exits;
-	u32 host_state_reload;
-	u32 efer_reload;
-	u32 fpu_reload;
-	u32 insn_emulation;
-	u32 insn_emulation_fail;
-};
-
-struct descriptor_table {
-	u16 limit;
-	unsigned long base;
-} __attribute__((packed));
-
-struct kvm_x86_ops {
-	int (*cpu_has_kvm_support)(void);          /* __init */
-	int (*disabled_by_bios)(void);             /* __init */
-	void (*hardware_enable)(void *dummy);      /* __init */
-	void (*hardware_disable)(void *dummy);
-	void (*check_processor_compatibility)(void *rtn);
-	int (*hardware_setup)(void);               /* __init */
-	void (*hardware_unsetup)(void);            /* __exit */
-
-	/* Create, but do not attach this VCPU */
-	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
-	void (*vcpu_free)(struct kvm_vcpu *vcpu);
-	int (*vcpu_reset)(struct kvm_vcpu *vcpu);
-
-	void (*prepare_guest_switch)(struct kvm_vcpu *vcpu);
-	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
-	void (*vcpu_put)(struct kvm_vcpu *vcpu);
-	void (*vcpu_decache)(struct kvm_vcpu *vcpu);
-
-	int (*set_guest_debug)(struct kvm_vcpu *vcpu,
-			       struct kvm_debug_guest *dbg);
-	void (*guest_debug_pre)(struct kvm_vcpu *vcpu);
-	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
-	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
-	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
-	void (*get_segment)(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg);
-	void (*set_segment)(struct kvm_vcpu *vcpu,
-			    struct kvm_segment *var, int seg);
-	void (*get_cs_db_l_bits)(struct kvm_vcpu *vcpu, int *db, int *l);
-	void (*decache_cr4_guest_bits)(struct kvm_vcpu *vcpu);
-	void (*set_cr0)(struct kvm_vcpu *vcpu, unsigned long cr0);
-	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);
-	void (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
-	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
-	void (*get_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
-	void (*set_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
-	void (*get_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
-	void (*set_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
-	unsigned long (*get_dr)(struct kvm_vcpu *vcpu, int dr);
-	void (*set_dr)(struct kvm_vcpu *vcpu, int dr, unsigned long value,
-		       int *exception);
-	void (*cache_regs)(struct kvm_vcpu *vcpu);
-	void (*decache_regs)(struct kvm_vcpu *vcpu);
-	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
-	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
-
-	void (*tlb_flush)(struct kvm_vcpu *vcpu);
-
-	void (*run)(struct kvm_vcpu *vcpu, struct kvm_run *run);
-	int (*handle_exit)(struct kvm_run *run, struct kvm_vcpu *vcpu);
-	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
-	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
-				unsigned char *hypercall_addr);
-	int (*get_irq)(struct kvm_vcpu *vcpu);
-	void (*set_irq)(struct kvm_vcpu *vcpu, int vec);
-	void (*queue_exception)(struct kvm_vcpu *vcpu, unsigned nr,
-				bool has_error_code, u32 error_code);
-	bool (*exception_injected)(struct kvm_vcpu *vcpu);
-	void (*inject_pending_irq)(struct kvm_vcpu *vcpu);
-	void (*inject_pending_vectors)(struct kvm_vcpu *vcpu,
-				       struct kvm_run *run);
-
-	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
-};
-
-extern struct kvm_x86_ops *kvm_x86_ops;
-
-int kvm_mmu_module_init(void);
-void kvm_mmu_module_exit(void);
-
-void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
-int kvm_mmu_create(struct kvm_vcpu *vcpu);
-int kvm_mmu_setup(struct kvm_vcpu *vcpu);
-void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
-
-int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
-void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
-void kvm_mmu_zap_all(struct kvm *kvm);
-unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
-void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);
-
-enum emulation_result {
-	EMULATE_DONE,       /* no further processing */
-	EMULATE_DO_MMIO,      /* kvm_run filled with mmio request */
-	EMULATE_FAIL,         /* can't emulate this instruction */
-};
-
-int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run,
-			unsigned long cr2, u16 error_code, int no_decode);
-void kvm_report_emulation_failure(struct kvm_vcpu *cvpu, const char *context);
-void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
-void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
-void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw,
-		   unsigned long *rflags);
-
-unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr);
-void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long value,
-		     unsigned long *rflags);
-int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *data);
-int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
-
-struct x86_emulate_ctxt;
-
-int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
-		     int size, unsigned port);
-int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
-			   int size, unsigned long count, int down,
-			    gva_t address, int rep, unsigned port);
-void kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
-int kvm_emulate_halt(struct kvm_vcpu *vcpu);
-int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address);
-int emulate_clts(struct kvm_vcpu *vcpu);
-int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr,
-		    unsigned long *dest);
-int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr,
-		    unsigned long value);
-
-void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
-void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr0);
-void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr0);
-void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr0);
-unsigned long get_cr8(struct kvm_vcpu *vcpu);
-void lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
-void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
-
-int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata);
-int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data);
-
-void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
-void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
-void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2,
-			   u32 error_code);
-
-void fx_init(struct kvm_vcpu *vcpu);
-
-int emulator_read_std(unsigned long addr,
-		      void *val,
-		      unsigned int bytes,
-		      struct kvm_vcpu *vcpu);
-int emulator_write_emulated(unsigned long addr,
-			    const void *val,
-			    unsigned int bytes,
-			    struct kvm_vcpu *vcpu);
-
-unsigned long segment_base(u16 selector);
-
-void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
-void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
-		       const u8 *new, int bytes);
-int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
-void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
-int kvm_mmu_load(struct kvm_vcpu *vcpu);
-void kvm_mmu_unload(struct kvm_vcpu *vcpu);
-
-int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);
-
-int kvm_fix_hypercall(struct kvm_vcpu *vcpu);
-
-int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code);
-
-int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3);
-int complete_pio(struct kvm_vcpu *vcpu);
-
-static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
-{
-	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);
-
-	return (struct kvm_mmu_page *)page_private(page);
-}
-
-static inline u16 read_fs(void)
-{
-	u16 seg;
-	asm("mov %%fs, %0" : "=g"(seg));
-	return seg;
-}
-
-static inline u16 read_gs(void)
-{
-	u16 seg;
-	asm("mov %%gs, %0" : "=g"(seg));
-	return seg;
-}
-
-static inline u16 read_ldt(void)
-{
-	u16 ldt;
-	asm("sldt %0" : "=g"(ldt));
-	return ldt;
-}
-
-static inline void load_fs(u16 sel)
-{
-	asm("mov %0, %%fs" : : "rm"(sel));
-}
-
-static inline void load_gs(u16 sel)
-{
-	asm("mov %0, %%gs" : : "rm"(sel));
-}
-
-#ifndef load_ldt
-static inline void load_ldt(u16 sel)
-{
-	asm("lldt %0" : : "rm"(sel));
-}
-#endif
-
-static inline void get_idt(struct descriptor_table *table)
-{
-	asm("sidt %0" : "=m"(*table));
-}
-
-static inline void get_gdt(struct descriptor_table *table)
-{
-	asm("sgdt %0" : "=m"(*table));
-}
-
-static inline unsigned long read_tr_base(void)
-{
-	u16 tr;
-	asm("str %0" : "=g"(tr));
-	return segment_base(tr);
-}
-
-#ifdef CONFIG_X86_64
-static inline unsigned long read_msr(unsigned long msr)
-{
-	u64 value;
-
-	rdmsrl(msr, value);
-	return value;
-}
-#endif
-
-static inline void fx_save(struct i387_fxsave_struct *image)
-{
-	asm("fxsave (%0)":: "r" (image));
-}
-
-static inline void fx_restore(struct i387_fxsave_struct *image)
-{
-	asm("fxrstor (%0)":: "r" (image));
-}
-
-static inline void fpu_init(void)
-{
-	asm("finit");
-}
-
-static inline u32 get_rdx_init_val(void)
-{
-	return 0x600; /* P6 family */
-}
-
-static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
-{
-	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
-}
-
-#define ASM_VMX_VMCLEAR_RAX       ".byte 0x66, 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMLAUNCH          ".byte 0x0f, 0x01, 0xc2"
-#define ASM_VMX_VMRESUME          ".byte 0x0f, 0x01, 0xc3"
-#define ASM_VMX_VMPTRLD_RAX       ".byte 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMREAD_RDX_RAX    ".byte 0x0f, 0x78, 0xd0"
-#define ASM_VMX_VMWRITE_RAX_RDX   ".byte 0x0f, 0x79, 0xd0"
-#define ASM_VMX_VMWRITE_RSP_RDX   ".byte 0x0f, 0x79, 0xd4"
-#define ASM_VMX_VMXOFF            ".byte 0x0f, 0x01, 0xc4"
-#define ASM_VMX_VMXON_RAX         ".byte 0xf3, 0x0f, 0xc7, 0x30"
-
-#define MSR_IA32_TIME_STAMP_COUNTER		0x010
-
-#define TSS_IOPB_BASE_OFFSET 0x66
-#define TSS_BASE_SIZE 0x68
-#define TSS_IOPB_SIZE (65536 / 8)
-#define TSS_REDIRECTION_SIZE (256 / 8)
-#define RMODE_TSS_SIZE (TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
-
-#endif
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
deleted file mode 100644
index 50b133f68743..000000000000
--- a/drivers/kvm/x86_emulate.c
+++ /dev/null
@@ -1,1913 +0,0 @@
-/******************************************************************************
- * x86_emulate.c
- *
- * Generic x86 (32-bit and 64-bit) instruction decoder and emulator.
- *
- * Copyright (c) 2005 Keir Fraser
- *
- * Linux coding style, mod r/m decoder, segment base fixes, real-mode
- * privileged instructions:
- *
- * Copyright (C) 2006 Qumranet
- *
- *   Avi Kivity <avi@qumranet.com>
- *   Yaniv Kamay <yaniv@qumranet.com>
- *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4
- */
-
-#ifndef __KERNEL__
-#include <stdio.h>
-#include <stdint.h>
-#include <public/xen.h>
-#define DPRINTF(_f, _a ...) printf(_f , ## _a)
-#else
-#include "kvm.h"
-#include "x86.h"
-#define DPRINTF(x...) do {} while (0)
-#endif
-#include "x86_emulate.h"
-#include <linux/module.h>
-
-/*
- * Opcode effective-address decode tables.
- * Note that we only emulate instructions that have at least one memory
- * operand (excluding implicit stack references). We assume that stack
- * references and instruction fetches will never occur in special memory
- * areas that require emulation. So, for example, 'mov <imm>,<reg>' need
- * not be handled.
- */
-
-/* Operand sizes: 8-bit operands or specified/overridden size. */
-#define ByteOp      (1<<0)	/* 8-bit operands. */
-/* Destination operand type. */
-#define ImplicitOps (1<<1)	/* Implicit in opcode. No generic decode. */
-#define DstReg      (2<<1)	/* Register operand. */
-#define DstMem      (3<<1)	/* Memory operand. */
-#define DstMask     (3<<1)
-/* Source operand type. */
-#define SrcNone     (0<<3)	/* No source operand. */
-#define SrcImplicit (0<<3)	/* Source operand is implicit in the opcode. */
-#define SrcReg      (1<<3)	/* Register operand. */
-#define SrcMem      (2<<3)	/* Memory operand. */
-#define SrcMem16    (3<<3)	/* Memory operand (16-bit). */
-#define SrcMem32    (4<<3)	/* Memory operand (32-bit). */
-#define SrcImm      (5<<3)	/* Immediate operand. */
-#define SrcImmByte  (6<<3)	/* 8-bit sign-extended immediate operand. */
-#define SrcMask     (7<<3)
-/* Generic ModRM decode. */
-#define ModRM       (1<<6)
-/* Destination is only written; never read. */
-#define Mov         (1<<7)
-#define BitOp       (1<<8)
-#define MemAbs      (1<<9)      /* Memory operand is absolute displacement */
-#define String      (1<<10)     /* String instruction (rep capable) */
-#define Stack       (1<<11)     /* Stack instruction (push/pop) */
-
-static u16 opcode_table[256] = {
-	/* 0x00 - 0x07 */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x08 - 0x0F */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x10 - 0x17 */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x18 - 0x1F */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x20 - 0x27 */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	SrcImmByte, SrcImm, 0, 0,
-	/* 0x28 - 0x2F */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x30 - 0x37 */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x38 - 0x3F */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
-	/* 0x40 - 0x47 */
-	DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg,
-	/* 0x48 - 0x4F */
-	DstReg, DstReg, DstReg, DstReg,	DstReg, DstReg, DstReg, DstReg,
-	/* 0x50 - 0x57 */
-	SrcReg | Stack, SrcReg | Stack, SrcReg | Stack, SrcReg | Stack,
-	SrcReg | Stack, SrcReg | Stack, SrcReg | Stack, SrcReg | Stack,
-	/* 0x58 - 0x5F */
-	DstReg | Stack, DstReg | Stack, DstReg | Stack, DstReg | Stack,
-	DstReg | Stack, DstReg | Stack, DstReg | Stack, DstReg | Stack,
-	/* 0x60 - 0x67 */
-	0, 0, 0, DstReg | SrcMem32 | ModRM | Mov /* movsxd (x86/64) */ ,
-	0, 0, 0, 0,
-	/* 0x68 - 0x6F */
-	0, 0, ImplicitOps | Mov | Stack, 0,
-	SrcNone  | ByteOp  | ImplicitOps, SrcNone  | ImplicitOps, /* insb, insw/insd */
-	SrcNone  | ByteOp  | ImplicitOps, SrcNone  | ImplicitOps, /* outsb, outsw/outsd */
-	/* 0x70 - 0x77 */
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	/* 0x78 - 0x7F */
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	/* 0x80 - 0x87 */
-	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImm | ModRM,
-	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM,
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
-	/* 0x88 - 0x8F */
-	ByteOp | DstMem | SrcReg | ModRM | Mov, DstMem | SrcReg | ModRM | Mov,
-	ByteOp | DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	0, ModRM | DstReg, 0, DstMem | SrcNone | ModRM | Mov | Stack,
-	/* 0x90 - 0x9F */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	0, 0, 0, 0, ImplicitOps | Stack, ImplicitOps | Stack, 0, 0,
-	/* 0xA0 - 0xA7 */
-	ByteOp | DstReg | SrcMem | Mov | MemAbs, DstReg | SrcMem | Mov | MemAbs,
-	ByteOp | DstMem | SrcReg | Mov | MemAbs, DstMem | SrcReg | Mov | MemAbs,
-	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
-	ByteOp | ImplicitOps | String, ImplicitOps | String,
-	/* 0xA8 - 0xAF */
-	0, 0, ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
-	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
-	ByteOp | ImplicitOps | String, ImplicitOps | String,
-	/* 0xB0 - 0xBF */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xC0 - 0xC7 */
-	ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM,
-	0, ImplicitOps | Stack, 0, 0,
-	ByteOp | DstMem | SrcImm | ModRM | Mov, DstMem | SrcImm | ModRM | Mov,
-	/* 0xC8 - 0xCF */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xD0 - 0xD7 */
-	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
-	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
-	0, 0, 0, 0,
-	/* 0xD8 - 0xDF */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xE0 - 0xE7 */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xE8 - 0xEF */
-	ImplicitOps | Stack, SrcImm|ImplicitOps, 0, SrcImmByte|ImplicitOps,
-	0, 0, 0, 0,
-	/* 0xF0 - 0xF7 */
-	0, 0, 0, 0,
-	ImplicitOps, ImplicitOps,
-	ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM,
-	/* 0xF8 - 0xFF */
-	ImplicitOps, 0, ImplicitOps, ImplicitOps,
-	0, 0, ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM
-};
-
-static u16 twobyte_table[256] = {
-	/* 0x00 - 0x0F */
-	0, SrcMem | ModRM | DstReg, 0, 0, 0, 0, ImplicitOps, 0,
-	ImplicitOps, ImplicitOps, 0, 0, 0, ImplicitOps | ModRM, 0, 0,
-	/* 0x10 - 0x1F */
-	0, 0, 0, 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x20 - 0x2F */
-	ModRM | ImplicitOps, ModRM, ModRM | ImplicitOps, ModRM, 0, 0, 0, 0,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x30 - 0x3F */
-	ImplicitOps, 0, ImplicitOps, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x40 - 0x47 */
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	/* 0x48 - 0x4F */
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov,
-	/* 0x50 - 0x5F */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x60 - 0x6F */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x70 - 0x7F */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0x80 - 0x8F */
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	ImplicitOps, ImplicitOps, ImplicitOps, ImplicitOps,
-	/* 0x90 - 0x9F */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xA0 - 0xA7 */
-	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0,
-	/* 0xA8 - 0xAF */
-	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0,
-	/* 0xB0 - 0xB7 */
-	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, 0,
-	    DstMem | SrcReg | ModRM | BitOp,
-	0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov,
-	    DstReg | SrcMem16 | ModRM | Mov,
-	/* 0xB8 - 0xBF */
-	0, 0, DstMem | SrcImmByte | ModRM, DstMem | SrcReg | ModRM | BitOp,
-	0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov,
-	    DstReg | SrcMem16 | ModRM | Mov,
-	/* 0xC0 - 0xCF */
-	0, 0, 0, DstMem | SrcReg | ModRM | Mov, 0, 0, 0, ImplicitOps | ModRM,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xD0 - 0xDF */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xE0 - 0xEF */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-	/* 0xF0 - 0xFF */
-	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
-};
-
-/* EFLAGS bit definitions. */
-#define EFLG_OF (1<<11)
-#define EFLG_DF (1<<10)
-#define EFLG_SF (1<<7)
-#define EFLG_ZF (1<<6)
-#define EFLG_AF (1<<4)
-#define EFLG_PF (1<<2)
-#define EFLG_CF (1<<0)
-
-/*
- * Instruction emulation:
- * Most instructions are emulated directly via a fragment of inline assembly
- * code. This allows us to save/restore EFLAGS and thus very easily pick up
- * any modified flags.
- */
-
-#if defined(CONFIG_X86_64)
-#define _LO32 "k"		/* force 32-bit operand */
-#define _STK  "%%rsp"		/* stack pointer */
-#elif defined(__i386__)
-#define _LO32 ""		/* force 32-bit operand */
-#define _STK  "%%esp"		/* stack pointer */
-#endif
-
-/*
- * These EFLAGS bits are restored from saved value during emulation, and
- * any changes are written back to the saved value after emulation.
- */
-#define EFLAGS_MASK (EFLG_OF|EFLG_SF|EFLG_ZF|EFLG_AF|EFLG_PF|EFLG_CF)
-
-/* Before executing instruction: restore necessary bits in EFLAGS. */
-#define _PRE_EFLAGS(_sav, _msk, _tmp)					\
-	/* EFLAGS = (_sav & _msk) | (EFLAGS & ~_msk); _sav &= ~_msk; */ \
-	"movl %"_sav",%"_LO32 _tmp"; "                                  \
-	"push %"_tmp"; "                                                \
-	"push %"_tmp"; "                                                \
-	"movl %"_msk",%"_LO32 _tmp"; "                                  \
-	"andl %"_LO32 _tmp",("_STK"); "                                 \
-	"pushf; "                                                       \
-	"notl %"_LO32 _tmp"; "                                          \
-	"andl %"_LO32 _tmp",("_STK"); "                                 \
-	"andl %"_LO32 _tmp","__stringify(BITS_PER_LONG/4)"("_STK"); "	\
-	"pop  %"_tmp"; "                                                \
-	"orl  %"_LO32 _tmp",("_STK"); "                                 \
-	"popf; "                                                        \
-	"pop  %"_sav"; "
-
-/* After executing instruction: write-back necessary bits in EFLAGS. */
-#define _POST_EFLAGS(_sav, _msk, _tmp) \
-	/* _sav |= EFLAGS & _msk; */		\
-	"pushf; "				\
-	"pop  %"_tmp"; "			\
-	"andl %"_msk",%"_LO32 _tmp"; "		\
-	"orl  %"_LO32 _tmp",%"_sav"; "
-
-/* Raw emulation: instruction has two explicit operands. */
-#define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \
-	do { 								    \
-		unsigned long _tmp;					    \
-									    \
-		switch ((_dst).bytes) {					    \
-		case 2:							    \
-			__asm__ __volatile__ (				    \
-				_PRE_EFLAGS("0", "4", "2")		    \
-				_op"w %"_wx"3,%1; "			    \
-				_POST_EFLAGS("0", "4", "2")		    \
-				: "=m" (_eflags), "=m" ((_dst).val),        \
-				  "=&r" (_tmp)				    \
-				: _wy ((_src).val), "i" (EFLAGS_MASK));     \
-			break;						    \
-		case 4:							    \
-			__asm__ __volatile__ (				    \
-				_PRE_EFLAGS("0", "4", "2")		    \
-				_op"l %"_lx"3,%1; "			    \
-				_POST_EFLAGS("0", "4", "2")		    \
-				: "=m" (_eflags), "=m" ((_dst).val),	    \
-				  "=&r" (_tmp)				    \
-				: _ly ((_src).val), "i" (EFLAGS_MASK));     \
-			break;						    \
-		case 8:							    \
-			__emulate_2op_8byte(_op, _src, _dst,		    \
-					    _eflags, _qx, _qy);		    \
-			break;						    \
-		}							    \
-	} while (0)
-
-#define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \
-	do {								     \
-		unsigned long _tmp;					     \
-		switch ((_dst).bytes) {				             \
-		case 1:							     \
-			__asm__ __volatile__ (				     \
-				_PRE_EFLAGS("0", "4", "2")		     \
-				_op"b %"_bx"3,%1; "			     \
-				_POST_EFLAGS("0", "4", "2")		     \
-				: "=m" (_eflags), "=m" ((_dst).val),	     \
-				  "=&r" (_tmp)				     \
-				: _by ((_src).val), "i" (EFLAGS_MASK));      \
-			break;						     \
-		default:						     \
-			__emulate_2op_nobyte(_op, _src, _dst, _eflags,	     \
-					     _wx, _wy, _lx, _ly, _qx, _qy);  \
-			break;						     \
-		}							     \
-	} while (0)
-
-/* Source operand is byte-sized and may be restricted to just %cl. */
-#define emulate_2op_SrcB(_op, _src, _dst, _eflags)                      \
-	__emulate_2op(_op, _src, _dst, _eflags,				\
-		      "b", "c", "b", "c", "b", "c", "b", "c")
-
-/* Source operand is byte, word, long or quad sized. */
-#define emulate_2op_SrcV(_op, _src, _dst, _eflags)                      \
-	__emulate_2op(_op, _src, _dst, _eflags,				\
-		      "b", "q", "w", "r", _LO32, "r", "", "r")
-
-/* Source operand is word, long or quad sized. */
-#define emulate_2op_SrcV_nobyte(_op, _src, _dst, _eflags)               \
-	__emulate_2op_nobyte(_op, _src, _dst, _eflags,			\
-			     "w", "r", _LO32, "r", "", "r")
-
-/* Instruction has only one explicit operand (no source operand). */
-#define emulate_1op(_op, _dst, _eflags)                                    \
-	do {								\
-		unsigned long _tmp;					\
-									\
-		switch ((_dst).bytes) {				        \
-		case 1:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"b %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 2:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"w %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 4:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"l %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 8:							\
-			__emulate_1op_8byte(_op, _dst, _eflags);	\
-			break;						\
-		}							\
-	} while (0)
-
-/* Emulate an instruction with quadword operands (x86/64 only). */
-#if defined(CONFIG_X86_64)
-#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)           \
-	do {								  \
-		__asm__ __volatile__ (					  \
-			_PRE_EFLAGS("0", "4", "2")			  \
-			_op"q %"_qx"3,%1; "				  \
-			_POST_EFLAGS("0", "4", "2")			  \
-			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
-			: _qy ((_src).val), "i" (EFLAGS_MASK));		\
-	} while (0)
-
-#define __emulate_1op_8byte(_op, _dst, _eflags)                           \
-	do {								  \
-		__asm__ __volatile__ (					  \
-			_PRE_EFLAGS("0", "3", "2")			  \
-			_op"q %1; "					  \
-			_POST_EFLAGS("0", "3", "2")			  \
-			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
-			: "i" (EFLAGS_MASK));				  \
-	} while (0)
-
-#elif defined(__i386__)
-#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)
-#define __emulate_1op_8byte(_op, _dst, _eflags)
-#endif				/* __i386__ */
-
-/* Fetch next part of the instruction being emulated. */
-#define insn_fetch(_type, _size, _eip)                                  \
-({	unsigned long _x;						\
-	rc = do_insn_fetch(ctxt, ops, (_eip), &_x, (_size));		\
-	if (rc != 0)							\
-		goto done;						\
-	(_eip) += (_size);						\
-	(_type)_x;							\
-})
-
-/* Access/update address held in a register, based on addressing mode. */
-#define address_mask(reg)						\
-	((c->ad_bytes == sizeof(unsigned long)) ? 			\
-		(reg) :	((reg) & ((1UL << (c->ad_bytes << 3)) - 1)))
-#define register_address(base, reg)                                     \
-	((base) + address_mask(reg))
-#define register_address_increment(reg, inc)                            \
-	do {								\
-		/* signed type ensures sign extension to long */        \
-		int _inc = (inc);					\
-		if (c->ad_bytes == sizeof(unsigned long))		\
-			(reg) += _inc;					\
-		else							\
-			(reg) = ((reg) & 				\
-				 ~((1UL << (c->ad_bytes << 3)) - 1)) |	\
-				(((reg) + _inc) &			\
-				 ((1UL << (c->ad_bytes << 3)) - 1));	\
-	} while (0)
-
-#define JMP_REL(rel) 							\
-	do {								\
-		register_address_increment(c->eip, rel);		\
-	} while (0)
-
-static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt,
-			      struct x86_emulate_ops *ops,
-			      unsigned long linear, u8 *dest)
-{
-	struct fetch_cache *fc = &ctxt->decode.fetch;
-	int rc;
-	int size;
-
-	if (linear < fc->start || linear >= fc->end) {
-		size = min(15UL, PAGE_SIZE - offset_in_page(linear));
-		rc = ops->read_std(linear, fc->data, size, ctxt->vcpu);
-		if (rc)
-			return rc;
-		fc->start = linear;
-		fc->end = linear + size;
-	}
-	*dest = fc->data[linear - fc->start];
-	return 0;
-}
-
-static int do_insn_fetch(struct x86_emulate_ctxt *ctxt,
-			 struct x86_emulate_ops *ops,
-			 unsigned long eip, void *dest, unsigned size)
-{
-	int rc = 0;
-
-	eip += ctxt->cs_base;
-	while (size--) {
-		rc = do_fetch_insn_byte(ctxt, ops, eip++, dest++);
-		if (rc)
-			return rc;
-	}
-	return 0;
-}
-
-/*
- * Given the 'reg' portion of a ModRM byte, and a register block, return a
- * pointer into the block that addresses the relevant register.
- * @highbyte_regs specifies whether to decode AH,CH,DH,BH.
- */
-static void *decode_register(u8 modrm_reg, unsigned long *regs,
-			     int highbyte_regs)
-{
-	void *p;
-
-	p = &regs[modrm_reg];
-	if (highbyte_regs && modrm_reg >= 4 && modrm_reg < 8)
-		p = (unsigned char *)&regs[modrm_reg & 3] + 1;
-	return p;
-}
-
-static int read_descriptor(struct x86_emulate_ctxt *ctxt,
-			   struct x86_emulate_ops *ops,
-			   void *ptr,
-			   u16 *size, unsigned long *address, int op_bytes)
-{
-	int rc;
-
-	if (op_bytes == 2)
-		op_bytes = 3;
-	*address = 0;
-	rc = ops->read_std((unsigned long)ptr, (unsigned long *)size, 2,
-			   ctxt->vcpu);
-	if (rc)
-		return rc;
-	rc = ops->read_std((unsigned long)ptr + 2, address, op_bytes,
-			   ctxt->vcpu);
-	return rc;
-}
-
-static int test_cc(unsigned int condition, unsigned int flags)
-{
-	int rc = 0;
-
-	switch ((condition & 15) >> 1) {
-	case 0: /* o */
-		rc |= (flags & EFLG_OF);
-		break;
-	case 1: /* b/c/nae */
-		rc |= (flags & EFLG_CF);
-		break;
-	case 2: /* z/e */
-		rc |= (flags & EFLG_ZF);
-		break;
-	case 3: /* be/na */
-		rc |= (flags & (EFLG_CF|EFLG_ZF));
-		break;
-	case 4: /* s */
-		rc |= (flags & EFLG_SF);
-		break;
-	case 5: /* p/pe */
-		rc |= (flags & EFLG_PF);
-		break;
-	case 7: /* le/ng */
-		rc |= (flags & EFLG_ZF);
-		/* fall through */
-	case 6: /* l/nge */
-		rc |= (!(flags & EFLG_SF) != !(flags & EFLG_OF));
-		break;
-	}
-
-	/* Odd condition identifiers (lsb == 1) have inverted sense. */
-	return (!!rc ^ (condition & 1));
-}
-
-static void decode_register_operand(struct operand *op,
-				    struct decode_cache *c,
-				    int inhibit_bytereg)
-{
-	unsigned reg = c->modrm_reg;
-	int highbyte_regs = c->rex_prefix == 0;
-
-	if (!(c->d & ModRM))
-		reg = (c->b & 7) | ((c->rex_prefix & 1) << 3);
-	op->type = OP_REG;
-	if ((c->d & ByteOp) && !inhibit_bytereg) {
-		op->ptr = decode_register(reg, c->regs, highbyte_regs);
-		op->val = *(u8 *)op->ptr;
-		op->bytes = 1;
-	} else {
-		op->ptr = decode_register(reg, c->regs, 0);
-		op->bytes = c->op_bytes;
-		switch (op->bytes) {
-		case 2:
-			op->val = *(u16 *)op->ptr;
-			break;
-		case 4:
-			op->val = *(u32 *)op->ptr;
-			break;
-		case 8:
-			op->val = *(u64 *) op->ptr;
-			break;
-		}
-	}
-	op->orig_val = op->val;
-}
-
-static int decode_modrm(struct x86_emulate_ctxt *ctxt,
-			struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	u8 sib;
-	int index_reg = 0, base_reg = 0, scale, rip_relative = 0;
-	int rc = 0;
-
-	if (c->rex_prefix) {
-		c->modrm_reg = (c->rex_prefix & 4) << 1;	/* REX.R */
-		index_reg = (c->rex_prefix & 2) << 2; /* REX.X */
-		c->modrm_rm = base_reg = (c->rex_prefix & 1) << 3; /* REG.B */
-	}
-
-	c->modrm = insn_fetch(u8, 1, c->eip);
-	c->modrm_mod |= (c->modrm & 0xc0) >> 6;
-	c->modrm_reg |= (c->modrm & 0x38) >> 3;
-	c->modrm_rm |= (c->modrm & 0x07);
-	c->modrm_ea = 0;
-	c->use_modrm_ea = 1;
-
-	if (c->modrm_mod == 3) {
-		c->modrm_val = *(unsigned long *)
-			decode_register(c->modrm_rm, c->regs, c->d & ByteOp);
-		return rc;
-	}
-
-	if (c->ad_bytes == 2) {
-		unsigned bx = c->regs[VCPU_REGS_RBX];
-		unsigned bp = c->regs[VCPU_REGS_RBP];
-		unsigned si = c->regs[VCPU_REGS_RSI];
-		unsigned di = c->regs[VCPU_REGS_RDI];
-
-		/* 16-bit ModR/M decode. */
-		switch (c->modrm_mod) {
-		case 0:
-			if (c->modrm_rm == 6)
-				c->modrm_ea += insn_fetch(u16, 2, c->eip);
-			break;
-		case 1:
-			c->modrm_ea += insn_fetch(s8, 1, c->eip);
-			break;
-		case 2:
-			c->modrm_ea += insn_fetch(u16, 2, c->eip);
-			break;
-		}
-		switch (c->modrm_rm) {
-		case 0:
-			c->modrm_ea += bx + si;
-			break;
-		case 1:
-			c->modrm_ea += bx + di;
-			break;
-		case 2:
-			c->modrm_ea += bp + si;
-			break;
-		case 3:
-			c->modrm_ea += bp + di;
-			break;
-		case 4:
-			c->modrm_ea += si;
-			break;
-		case 5:
-			c->modrm_ea += di;
-			break;
-		case 6:
-			if (c->modrm_mod != 0)
-				c->modrm_ea += bp;
-			break;
-		case 7:
-			c->modrm_ea += bx;
-			break;
-		}
-		if (c->modrm_rm == 2 || c->modrm_rm == 3 ||
-		    (c->modrm_rm == 6 && c->modrm_mod != 0))
-			if (!c->override_base)
-				c->override_base = &ctxt->ss_base;
-		c->modrm_ea = (u16)c->modrm_ea;
-	} else {
-		/* 32/64-bit ModR/M decode. */
-		switch (c->modrm_rm) {
-		case 4:
-		case 12:
-			sib = insn_fetch(u8, 1, c->eip);
-			index_reg |= (sib >> 3) & 7;
-			base_reg |= sib & 7;
-			scale = sib >> 6;
-
-			switch (base_reg) {
-			case 5:
-				if (c->modrm_mod != 0)
-					c->modrm_ea += c->regs[base_reg];
-				else
-					c->modrm_ea +=
-						insn_fetch(s32, 4, c->eip);
-				break;
-			default:
-				c->modrm_ea += c->regs[base_reg];
-			}
-			switch (index_reg) {
-			case 4:
-				break;
-			default:
-				c->modrm_ea += c->regs[index_reg] << scale;
-			}
-			break;
-		case 5:
-			if (c->modrm_mod != 0)
-				c->modrm_ea += c->regs[c->modrm_rm];
-			else if (ctxt->mode == X86EMUL_MODE_PROT64)
-				rip_relative = 1;
-			break;
-		default:
-			c->modrm_ea += c->regs[c->modrm_rm];
-			break;
-		}
-		switch (c->modrm_mod) {
-		case 0:
-			if (c->modrm_rm == 5)
-				c->modrm_ea += insn_fetch(s32, 4, c->eip);
-			break;
-		case 1:
-			c->modrm_ea += insn_fetch(s8, 1, c->eip);
-			break;
-		case 2:
-			c->modrm_ea += insn_fetch(s32, 4, c->eip);
-			break;
-		}
-	}
-	if (rip_relative) {
-		c->modrm_ea += c->eip;
-		switch (c->d & SrcMask) {
-		case SrcImmByte:
-			c->modrm_ea += 1;
-			break;
-		case SrcImm:
-			if (c->d & ByteOp)
-				c->modrm_ea += 1;
-			else
-				if (c->op_bytes == 8)
-					c->modrm_ea += 4;
-				else
-					c->modrm_ea += c->op_bytes;
-		}
-	}
-done:
-	return rc;
-}
-
-static int decode_abs(struct x86_emulate_ctxt *ctxt,
-		      struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	int rc = 0;
-
-	switch (c->ad_bytes) {
-	case 2:
-		c->modrm_ea = insn_fetch(u16, 2, c->eip);
-		break;
-	case 4:
-		c->modrm_ea = insn_fetch(u32, 4, c->eip);
-		break;
-	case 8:
-		c->modrm_ea = insn_fetch(u64, 8, c->eip);
-		break;
-	}
-done:
-	return rc;
-}
-
-int
-x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	int rc = 0;
-	int mode = ctxt->mode;
-	int def_op_bytes, def_ad_bytes;
-
-	/* Shadow copy of register state. Committed on successful emulation. */
-
-	memset(c, 0, sizeof(struct decode_cache));
-	c->eip = ctxt->vcpu->arch.rip;
-	memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
-
-	switch (mode) {
-	case X86EMUL_MODE_REAL:
-	case X86EMUL_MODE_PROT16:
-		def_op_bytes = def_ad_bytes = 2;
-		break;
-	case X86EMUL_MODE_PROT32:
-		def_op_bytes = def_ad_bytes = 4;
-		break;
-#ifdef CONFIG_X86_64
-	case X86EMUL_MODE_PROT64:
-		def_op_bytes = 4;
-		def_ad_bytes = 8;
-		break;
-#endif
-	default:
-		return -1;
-	}
-
-	c->op_bytes = def_op_bytes;
-	c->ad_bytes = def_ad_bytes;
-
-	/* Legacy prefixes. */
-	for (;;) {
-		switch (c->b = insn_fetch(u8, 1, c->eip)) {
-		case 0x66:	/* operand-size override */
-			/* switch between 2/4 bytes */
-			c->op_bytes = def_op_bytes ^ 6;
-			break;
-		case 0x67:	/* address-size override */
-			if (mode == X86EMUL_MODE_PROT64)
-				/* switch between 4/8 bytes */
-				c->ad_bytes = def_ad_bytes ^ 12;
-			else
-				/* switch between 2/4 bytes */
-				c->ad_bytes = def_ad_bytes ^ 6;
-			break;
-		case 0x2e:	/* CS override */
-			c->override_base = &ctxt->cs_base;
-			break;
-		case 0x3e:	/* DS override */
-			c->override_base = &ctxt->ds_base;
-			break;
-		case 0x26:	/* ES override */
-			c->override_base = &ctxt->es_base;
-			break;
-		case 0x64:	/* FS override */
-			c->override_base = &ctxt->fs_base;
-			break;
-		case 0x65:	/* GS override */
-			c->override_base = &ctxt->gs_base;
-			break;
-		case 0x36:	/* SS override */
-			c->override_base = &ctxt->ss_base;
-			break;
-		case 0x40 ... 0x4f: /* REX */
-			if (mode != X86EMUL_MODE_PROT64)
-				goto done_prefixes;
-			c->rex_prefix = c->b;
-			continue;
-		case 0xf0:	/* LOCK */
-			c->lock_prefix = 1;
-			break;
-		case 0xf2:	/* REPNE/REPNZ */
-			c->rep_prefix = REPNE_PREFIX;
-			break;
-		case 0xf3:	/* REP/REPE/REPZ */
-			c->rep_prefix = REPE_PREFIX;
-			break;
-		default:
-			goto done_prefixes;
-		}
-
-		/* Any legacy prefix after a REX prefix nullifies its effect. */
-
-		c->rex_prefix = 0;
-	}
-
-done_prefixes:
-
-	/* REX prefix. */
-	if (c->rex_prefix)
-		if (c->rex_prefix & 8)
-			c->op_bytes = 8;	/* REX.W */
-
-	/* Opcode byte(s). */
-	c->d = opcode_table[c->b];
-	if (c->d == 0) {
-		/* Two-byte opcode? */
-		if (c->b == 0x0f) {
-			c->twobyte = 1;
-			c->b = insn_fetch(u8, 1, c->eip);
-			c->d = twobyte_table[c->b];
-		}
-
-		/* Unrecognised? */
-		if (c->d == 0) {
-			DPRINTF("Cannot emulate %02x\n", c->b);
-			return -1;
-		}
-	}
-
-	if (mode == X86EMUL_MODE_PROT64 && (c->d & Stack))
-		c->op_bytes = 8;
-
-	/* ModRM and SIB bytes. */
-	if (c->d & ModRM)
-		rc = decode_modrm(ctxt, ops);
-	else if (c->d & MemAbs)
-		rc = decode_abs(ctxt, ops);
-	if (rc)
-		goto done;
-
-	if (!c->override_base)
-		c->override_base = &ctxt->ds_base;
-	if (mode == X86EMUL_MODE_PROT64 &&
-	    c->override_base != &ctxt->fs_base &&
-	    c->override_base != &ctxt->gs_base)
-		c->override_base = NULL;
-
-	if (c->override_base)
-		c->modrm_ea += *c->override_base;
-
-	if (c->ad_bytes != 8)
-		c->modrm_ea = (u32)c->modrm_ea;
-	/*
-	 * Decode and fetch the source operand: register, memory
-	 * or immediate.
-	 */
-	switch (c->d & SrcMask) {
-	case SrcNone:
-		break;
-	case SrcReg:
-		decode_register_operand(&c->src, c, 0);
-		break;
-	case SrcMem16:
-		c->src.bytes = 2;
-		goto srcmem_common;
-	case SrcMem32:
-		c->src.bytes = 4;
-		goto srcmem_common;
-	case SrcMem:
-		c->src.bytes = (c->d & ByteOp) ? 1 :
-							   c->op_bytes;
-		/* Don't fetch the address for invlpg: it could be unmapped. */
-		if (c->twobyte && c->b == 0x01 && c->modrm_reg == 7)
-			break;
-	srcmem_common:
-		/*
-		 * For instructions with a ModR/M byte, switch to register
-		 * access if Mod = 3.
-		 */
-		if ((c->d & ModRM) && c->modrm_mod == 3) {
-			c->src.type = OP_REG;
-			break;
-		}
-		c->src.type = OP_MEM;
-		break;
-	case SrcImm:
-		c->src.type = OP_IMM;
-		c->src.ptr = (unsigned long *)c->eip;
-		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		if (c->src.bytes == 8)
-			c->src.bytes = 4;
-		/* NB. Immediates are sign-extended as necessary. */
-		switch (c->src.bytes) {
-		case 1:
-			c->src.val = insn_fetch(s8, 1, c->eip);
-			break;
-		case 2:
-			c->src.val = insn_fetch(s16, 2, c->eip);
-			break;
-		case 4:
-			c->src.val = insn_fetch(s32, 4, c->eip);
-			break;
-		}
-		break;
-	case SrcImmByte:
-		c->src.type = OP_IMM;
-		c->src.ptr = (unsigned long *)c->eip;
-		c->src.bytes = 1;
-		c->src.val = insn_fetch(s8, 1, c->eip);
-		break;
-	}
-
-	/* Decode and fetch the destination operand: register or memory. */
-	switch (c->d & DstMask) {
-	case ImplicitOps:
-		/* Special instructions do their own operand decoding. */
-		return 0;
-	case DstReg:
-		decode_register_operand(&c->dst, c,
-			 c->twobyte && (c->b == 0xb6 || c->b == 0xb7));
-		break;
-	case DstMem:
-		if ((c->d & ModRM) && c->modrm_mod == 3) {
-			c->dst.type = OP_REG;
-			break;
-		}
-		c->dst.type = OP_MEM;
-		break;
-	}
-
-done:
-	return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0;
-}
-
-static inline void emulate_push(struct x86_emulate_ctxt *ctxt)
-{
-	struct decode_cache *c = &ctxt->decode;
-
-	c->dst.type  = OP_MEM;
-	c->dst.bytes = c->op_bytes;
-	c->dst.val = c->src.val;
-	register_address_increment(c->regs[VCPU_REGS_RSP], -c->op_bytes);
-	c->dst.ptr = (void *) register_address(ctxt->ss_base,
-					       c->regs[VCPU_REGS_RSP]);
-}
-
-static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt,
-				struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	int rc;
-
-	rc = ops->read_std(register_address(ctxt->ss_base,
-					    c->regs[VCPU_REGS_RSP]),
-			   &c->dst.val, c->dst.bytes, ctxt->vcpu);
-	if (rc != 0)
-		return rc;
-
-	register_address_increment(c->regs[VCPU_REGS_RSP], c->dst.bytes);
-
-	return 0;
-}
-
-static inline void emulate_grp2(struct x86_emulate_ctxt *ctxt)
-{
-	struct decode_cache *c = &ctxt->decode;
-	switch (c->modrm_reg) {
-	case 0:	/* rol */
-		emulate_2op_SrcB("rol", c->src, c->dst, ctxt->eflags);
-		break;
-	case 1:	/* ror */
-		emulate_2op_SrcB("ror", c->src, c->dst, ctxt->eflags);
-		break;
-	case 2:	/* rcl */
-		emulate_2op_SrcB("rcl", c->src, c->dst, ctxt->eflags);
-		break;
-	case 3:	/* rcr */
-		emulate_2op_SrcB("rcr", c->src, c->dst, ctxt->eflags);
-		break;
-	case 4:	/* sal/shl */
-	case 6:	/* sal/shl */
-		emulate_2op_SrcB("sal", c->src, c->dst, ctxt->eflags);
-		break;
-	case 5:	/* shr */
-		emulate_2op_SrcB("shr", c->src, c->dst, ctxt->eflags);
-		break;
-	case 7:	/* sar */
-		emulate_2op_SrcB("sar", c->src, c->dst, ctxt->eflags);
-		break;
-	}
-}
-
-static inline int emulate_grp3(struct x86_emulate_ctxt *ctxt,
-			       struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	int rc = 0;
-
-	switch (c->modrm_reg) {
-	case 0 ... 1:	/* test */
-		/*
-		 * Special case in Grp3: test has an immediate
-		 * source operand.
-		 */
-		c->src.type = OP_IMM;
-		c->src.ptr = (unsigned long *)c->eip;
-		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		if (c->src.bytes == 8)
-			c->src.bytes = 4;
-		switch (c->src.bytes) {
-		case 1:
-			c->src.val = insn_fetch(s8, 1, c->eip);
-			break;
-		case 2:
-			c->src.val = insn_fetch(s16, 2, c->eip);
-			break;
-		case 4:
-			c->src.val = insn_fetch(s32, 4, c->eip);
-			break;
-		}
-		emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
-		break;
-	case 2:	/* not */
-		c->dst.val = ~c->dst.val;
-		break;
-	case 3:	/* neg */
-		emulate_1op("neg", c->dst, ctxt->eflags);
-		break;
-	default:
-		DPRINTF("Cannot emulate %02x\n", c->b);
-		rc = X86EMUL_UNHANDLEABLE;
-		break;
-	}
-done:
-	return rc;
-}
-
-static inline int emulate_grp45(struct x86_emulate_ctxt *ctxt,
-			       struct x86_emulate_ops *ops)
-{
-	struct decode_cache *c = &ctxt->decode;
-	int rc;
-
-	switch (c->modrm_reg) {
-	case 0:	/* inc */
-		emulate_1op("inc", c->dst, ctxt->eflags);
-		break;
-	case 1:	/* dec */
-		emulate_1op("dec", c->dst, ctxt->eflags);
-		break;
-	case 4: /* jmp abs */
-		if (c->b == 0xff)
-			c->eip = c->dst.val;
-		else {
-			DPRINTF("Cannot emulate %02x\n", c->b);
-			return X86EMUL_UNHANDLEABLE;
-		}
-		break;
-	case 6:	/* push */
-
-		/* 64-bit mode: PUSH always pushes a 64-bit operand. */
-
-		if (ctxt->mode == X86EMUL_MODE_PROT64) {
-			c->dst.bytes = 8;
-			rc = ops->read_std((unsigned long)c->dst.ptr,
-					   &c->dst.val, 8, ctxt->vcpu);
-			if (rc != 0)
-				return rc;
-		}
-		register_address_increment(c->regs[VCPU_REGS_RSP],
-					   -c->dst.bytes);
-		rc = ops->write_emulated(register_address(ctxt->ss_base,
-				    c->regs[VCPU_REGS_RSP]), &c->dst.val,
-				    c->dst.bytes, ctxt->vcpu);
-		if (rc != 0)
-			return rc;
-		c->dst.type = OP_NONE;
-		break;
-	default:
-		DPRINTF("Cannot emulate %02x\n", c->b);
-		return X86EMUL_UNHANDLEABLE;
-	}
-	return 0;
-}
-
-static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt,
-			       struct x86_emulate_ops *ops,
-			       unsigned long memop)
-{
-	struct decode_cache *c = &ctxt->decode;
-	u64 old, new;
-	int rc;
-
-	rc = ops->read_emulated(memop, &old, 8, ctxt->vcpu);
-	if (rc != 0)
-		return rc;
-
-	if (((u32) (old >> 0) != (u32) c->regs[VCPU_REGS_RAX]) ||
-	    ((u32) (old >> 32) != (u32) c->regs[VCPU_REGS_RDX])) {
-
-		c->regs[VCPU_REGS_RAX] = (u32) (old >> 0);
-		c->regs[VCPU_REGS_RDX] = (u32) (old >> 32);
-		ctxt->eflags &= ~EFLG_ZF;
-
-	} else {
-		new = ((u64)c->regs[VCPU_REGS_RCX] << 32) |
-		       (u32) c->regs[VCPU_REGS_RBX];
-
-		rc = ops->cmpxchg_emulated(memop, &old, &new, 8, ctxt->vcpu);
-		if (rc != 0)
-			return rc;
-		ctxt->eflags |= EFLG_ZF;
-	}
-	return 0;
-}
-
-static inline int writeback(struct x86_emulate_ctxt *ctxt,
-			    struct x86_emulate_ops *ops)
-{
-	int rc;
-	struct decode_cache *c = &ctxt->decode;
-
-	switch (c->dst.type) {
-	case OP_REG:
-		/* The 4-byte case *is* correct:
-		 * in 64-bit mode we zero-extend.
-		 */
-		switch (c->dst.bytes) {
-		case 1:
-			*(u8 *)c->dst.ptr = (u8)c->dst.val;
-			break;
-		case 2:
-			*(u16 *)c->dst.ptr = (u16)c->dst.val;
-			break;
-		case 4:
-			*c->dst.ptr = (u32)c->dst.val;
-			break;	/* 64b: zero-ext */
-		case 8:
-			*c->dst.ptr = c->dst.val;
-			break;
-		}
-		break;
-	case OP_MEM:
-		if (c->lock_prefix)
-			rc = ops->cmpxchg_emulated(
-					(unsigned long)c->dst.ptr,
-					&c->dst.orig_val,
-					&c->dst.val,
-					c->dst.bytes,
-					ctxt->vcpu);
-		else
-			rc = ops->write_emulated(
-					(unsigned long)c->dst.ptr,
-					&c->dst.val,
-					c->dst.bytes,
-					ctxt->vcpu);
-		if (rc != 0)
-			return rc;
-		break;
-	case OP_NONE:
-		/* no writeback */
-		break;
-	default:
-		break;
-	}
-	return 0;
-}
-
-int
-x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
-{
-	unsigned long memop = 0;
-	u64 msr_data;
-	unsigned long saved_eip = 0;
-	struct decode_cache *c = &ctxt->decode;
-	int rc = 0;
-
-	/* Shadow copy of register state. Committed on successful emulation.
-	 * NOTE: we can copy them from vcpu as x86_decode_insn() doesn't
-	 * modify them.
-	 */
-
-	memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
-	saved_eip = c->eip;
-
-	if (((c->d & ModRM) && (c->modrm_mod != 3)) || (c->d & MemAbs))
-		memop = c->modrm_ea;
-
-	if (c->rep_prefix && (c->d & String)) {
-		/* All REP prefixes have the same first termination condition */
-		if (c->regs[VCPU_REGS_RCX] == 0) {
-			ctxt->vcpu->arch.rip = c->eip;
-			goto done;
-		}
-		/* The second termination condition only applies for REPE
-		 * and REPNE. Test if the repeat string operation prefix is
-		 * REPE/REPZ or REPNE/REPNZ and if it's the case it tests the
-		 * corresponding termination condition according to:
-		 * 	- if REPE/REPZ and ZF = 0 then done
-		 * 	- if REPNE/REPNZ and ZF = 1 then done
-		 */
-		if ((c->b == 0xa6) || (c->b == 0xa7) ||
-				(c->b == 0xae) || (c->b == 0xaf)) {
-			if ((c->rep_prefix == REPE_PREFIX) &&
-				((ctxt->eflags & EFLG_ZF) == 0)) {
-					ctxt->vcpu->arch.rip = c->eip;
-					goto done;
-			}
-			if ((c->rep_prefix == REPNE_PREFIX) &&
-				((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) {
-				ctxt->vcpu->arch.rip = c->eip;
-				goto done;
-			}
-		}
-		c->regs[VCPU_REGS_RCX]--;
-		c->eip = ctxt->vcpu->arch.rip;
-	}
-
-	if (c->src.type == OP_MEM) {
-		c->src.ptr = (unsigned long *)memop;
-		c->src.val = 0;
-		rc = ops->read_emulated((unsigned long)c->src.ptr,
-					&c->src.val,
-					c->src.bytes,
-					ctxt->vcpu);
-		if (rc != 0)
-			goto done;
-		c->src.orig_val = c->src.val;
-	}
-
-	if ((c->d & DstMask) == ImplicitOps)
-		goto special_insn;
-
-
-	if (c->dst.type == OP_MEM) {
-		c->dst.ptr = (unsigned long *)memop;
-		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->dst.val = 0;
-		if (c->d & BitOp) {
-			unsigned long mask = ~(c->dst.bytes * 8 - 1);
-
-			c->dst.ptr = (void *)c->dst.ptr +
-						   (c->src.val & mask) / 8;
-		}
-		if (!(c->d & Mov) &&
-				   /* optimisation - avoid slow emulated read */
-		    ((rc = ops->read_emulated((unsigned long)c->dst.ptr,
-					   &c->dst.val,
-					  c->dst.bytes, ctxt->vcpu)) != 0))
-			goto done;
-	}
-	c->dst.orig_val = c->dst.val;
-
-special_insn:
-
-	if (c->twobyte)
-		goto twobyte_insn;
-
-	switch (c->b) {
-	case 0x00 ... 0x05:
-	      add:		/* add */
-		emulate_2op_SrcV("add", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x08 ... 0x0d:
-	      or:		/* or */
-		emulate_2op_SrcV("or", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x10 ... 0x15:
-	      adc:		/* adc */
-		emulate_2op_SrcV("adc", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x18 ... 0x1d:
-	      sbb:		/* sbb */
-		emulate_2op_SrcV("sbb", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x20 ... 0x23:
-	      and:		/* and */
-		emulate_2op_SrcV("and", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x24:              /* and al imm8 */
-		c->dst.type = OP_REG;
-		c->dst.ptr = &c->regs[VCPU_REGS_RAX];
-		c->dst.val = *(u8 *)c->dst.ptr;
-		c->dst.bytes = 1;
-		c->dst.orig_val = c->dst.val;
-		goto and;
-	case 0x25:              /* and ax imm16, or eax imm32 */
-		c->dst.type = OP_REG;
-		c->dst.bytes = c->op_bytes;
-		c->dst.ptr = &c->regs[VCPU_REGS_RAX];
-		if (c->op_bytes == 2)
-			c->dst.val = *(u16 *)c->dst.ptr;
-		else
-			c->dst.val = *(u32 *)c->dst.ptr;
-		c->dst.orig_val = c->dst.val;
-		goto and;
-	case 0x28 ... 0x2d:
-	      sub:		/* sub */
-		emulate_2op_SrcV("sub", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x30 ... 0x35:
-	      xor:		/* xor */
-		emulate_2op_SrcV("xor", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x38 ... 0x3d:
-	      cmp:		/* cmp */
-		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x40 ... 0x47: /* inc r16/r32 */
-		emulate_1op("inc", c->dst, ctxt->eflags);
-		break;
-	case 0x48 ... 0x4f: /* dec r16/r32 */
-		emulate_1op("dec", c->dst, ctxt->eflags);
-		break;
-	case 0x50 ... 0x57:  /* push reg */
-		c->dst.type  = OP_MEM;
-		c->dst.bytes = c->op_bytes;
-		c->dst.val = c->src.val;
-		register_address_increment(c->regs[VCPU_REGS_RSP],
-					   -c->op_bytes);
-		c->dst.ptr = (void *) register_address(
-			ctxt->ss_base, c->regs[VCPU_REGS_RSP]);
-		break;
-	case 0x58 ... 0x5f: /* pop reg */
-	pop_instruction:
-		if ((rc = ops->read_std(register_address(ctxt->ss_base,
-			c->regs[VCPU_REGS_RSP]), c->dst.ptr,
-			c->op_bytes, ctxt->vcpu)) != 0)
-			goto done;
-
-		register_address_increment(c->regs[VCPU_REGS_RSP],
-					   c->op_bytes);
-		c->dst.type = OP_NONE;	/* Disable writeback. */
-		break;
-	case 0x63:		/* movsxd */
-		if (ctxt->mode != X86EMUL_MODE_PROT64)
-			goto cannot_emulate;
-		c->dst.val = (s32) c->src.val;
-		break;
-	case 0x6a: /* push imm8 */
-		c->src.val = 0L;
-		c->src.val = insn_fetch(s8, 1, c->eip);
-		emulate_push(ctxt);
-		break;
-	case 0x6c:		/* insb */
-	case 0x6d:		/* insw/insd */
-		 if (kvm_emulate_pio_string(ctxt->vcpu, NULL,
-				1,
-				(c->d & ByteOp) ? 1 : c->op_bytes,
-				c->rep_prefix ?
-				address_mask(c->regs[VCPU_REGS_RCX]) : 1,
-				(ctxt->eflags & EFLG_DF),
-				register_address(ctxt->es_base,
-						 c->regs[VCPU_REGS_RDI]),
-				c->rep_prefix,
-				c->regs[VCPU_REGS_RDX]) == 0) {
-			c->eip = saved_eip;
-			return -1;
-		}
-		return 0;
-	case 0x6e:		/* outsb */
-	case 0x6f:		/* outsw/outsd */
-		if (kvm_emulate_pio_string(ctxt->vcpu, NULL,
-				0,
-				(c->d & ByteOp) ? 1 : c->op_bytes,
-				c->rep_prefix ?
-				address_mask(c->regs[VCPU_REGS_RCX]) : 1,
-				(ctxt->eflags & EFLG_DF),
-				register_address(c->override_base ?
-							*c->override_base :
-							ctxt->ds_base,
-						 c->regs[VCPU_REGS_RSI]),
-				c->rep_prefix,
-				c->regs[VCPU_REGS_RDX]) == 0) {
-			c->eip = saved_eip;
-			return -1;
-		}
-		return 0;
-	case 0x70 ... 0x7f: /* jcc (short) */ {
-		int rel = insn_fetch(s8, 1, c->eip);
-
-		if (test_cc(c->b, ctxt->eflags))
-			JMP_REL(rel);
-		break;
-	}
-	case 0x80 ... 0x83:	/* Grp1 */
-		switch (c->modrm_reg) {
-		case 0:
-			goto add;
-		case 1:
-			goto or;
-		case 2:
-			goto adc;
-		case 3:
-			goto sbb;
-		case 4:
-			goto and;
-		case 5:
-			goto sub;
-		case 6:
-			goto xor;
-		case 7:
-			goto cmp;
-		}
-		break;
-	case 0x84 ... 0x85:
-		emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0x86 ... 0x87:	/* xchg */
-		/* Write back the register source. */
-		switch (c->dst.bytes) {
-		case 1:
-			*(u8 *) c->src.ptr = (u8) c->dst.val;
-			break;
-		case 2:
-			*(u16 *) c->src.ptr = (u16) c->dst.val;
-			break;
-		case 4:
-			*c->src.ptr = (u32) c->dst.val;
-			break;	/* 64b reg: zero-extend */
-		case 8:
-			*c->src.ptr = c->dst.val;
-			break;
-		}
-		/*
-		 * Write back the memory destination with implicit LOCK
-		 * prefix.
-		 */
-		c->dst.val = c->src.val;
-		c->lock_prefix = 1;
-		break;
-	case 0x88 ... 0x8b:	/* mov */
-		goto mov;
-	case 0x8d: /* lea r16/r32, m */
-		c->dst.val = c->modrm_val;
-		break;
-	case 0x8f:		/* pop (sole member of Grp1a) */
-		rc = emulate_grp1a(ctxt, ops);
-		if (rc != 0)
-			goto done;
-		break;
-	case 0x9c: /* pushf */
-		c->src.val =  (unsigned long) ctxt->eflags;
-		emulate_push(ctxt);
-		break;
-	case 0x9d: /* popf */
-		c->dst.ptr = (unsigned long *) &ctxt->eflags;
-		goto pop_instruction;
-	case 0xa0 ... 0xa1:	/* mov */
-		c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
-		c->dst.val = c->src.val;
-		break;
-	case 0xa2 ... 0xa3:	/* mov */
-		c->dst.val = (unsigned long)c->regs[VCPU_REGS_RAX];
-		break;
-	case 0xa4 ... 0xa5:	/* movs */
-		c->dst.type = OP_MEM;
-		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->dst.ptr = (unsigned long *)register_address(
-						   ctxt->es_base,
-						   c->regs[VCPU_REGS_RDI]);
-		if ((rc = ops->read_emulated(register_address(
-		      c->override_base ? *c->override_base :
-					ctxt->ds_base,
-					c->regs[VCPU_REGS_RSI]),
-					&c->dst.val,
-					c->dst.bytes, ctxt->vcpu)) != 0)
-			goto done;
-		register_address_increment(c->regs[VCPU_REGS_RSI],
-				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
-							   : c->dst.bytes);
-		register_address_increment(c->regs[VCPU_REGS_RDI],
-				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
-							   : c->dst.bytes);
-		break;
-	case 0xa6 ... 0xa7:	/* cmps */
-		c->src.type = OP_NONE; /* Disable writeback. */
-		c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->src.ptr = (unsigned long *)register_address(
-				c->override_base ? *c->override_base :
-						   ctxt->ds_base,
-						   c->regs[VCPU_REGS_RSI]);
-		if ((rc = ops->read_emulated((unsigned long)c->src.ptr,
-						&c->src.val,
-						c->src.bytes,
-						ctxt->vcpu)) != 0)
-			goto done;
-
-		c->dst.type = OP_NONE; /* Disable writeback. */
-		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->dst.ptr = (unsigned long *)register_address(
-						   ctxt->es_base,
-						   c->regs[VCPU_REGS_RDI]);
-		if ((rc = ops->read_emulated((unsigned long)c->dst.ptr,
-						&c->dst.val,
-						c->dst.bytes,
-						ctxt->vcpu)) != 0)
-			goto done;
-
-		DPRINTF("cmps: mem1=0x%p mem2=0x%p\n", c->src.ptr, c->dst.ptr);
-
-		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
-
-		register_address_increment(c->regs[VCPU_REGS_RSI],
-				       (ctxt->eflags & EFLG_DF) ? -c->src.bytes
-								  : c->src.bytes);
-		register_address_increment(c->regs[VCPU_REGS_RDI],
-				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
-								  : c->dst.bytes);
-
-		break;
-	case 0xaa ... 0xab:	/* stos */
-		c->dst.type = OP_MEM;
-		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->dst.ptr = (unsigned long *)register_address(
-						   ctxt->es_base,
-						   c->regs[VCPU_REGS_RDI]);
-		c->dst.val = c->regs[VCPU_REGS_RAX];
-		register_address_increment(c->regs[VCPU_REGS_RDI],
-				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
-							   : c->dst.bytes);
-		break;
-	case 0xac ... 0xad:	/* lods */
-		c->dst.type = OP_REG;
-		c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
-		c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
-		if ((rc = ops->read_emulated(register_address(
-				c->override_base ? *c->override_base :
-						   ctxt->ds_base,
-						 c->regs[VCPU_REGS_RSI]),
-						 &c->dst.val,
-						 c->dst.bytes,
-						 ctxt->vcpu)) != 0)
-			goto done;
-		register_address_increment(c->regs[VCPU_REGS_RSI],
-				       (ctxt->eflags & EFLG_DF) ? -c->dst.bytes
-							   : c->dst.bytes);
-		break;
-	case 0xae ... 0xaf:	/* scas */
-		DPRINTF("Urk! I don't handle SCAS.\n");
-		goto cannot_emulate;
-	case 0xc0 ... 0xc1:
-		emulate_grp2(ctxt);
-		break;
-	case 0xc3: /* ret */
-		c->dst.ptr = &c->eip;
-		goto pop_instruction;
-	case 0xc6 ... 0xc7:	/* mov (sole member of Grp11) */
-	mov:
-		c->dst.val = c->src.val;
-		break;
-	case 0xd0 ... 0xd1:	/* Grp2 */
-		c->src.val = 1;
-		emulate_grp2(ctxt);
-		break;
-	case 0xd2 ... 0xd3:	/* Grp2 */
-		c->src.val = c->regs[VCPU_REGS_RCX];
-		emulate_grp2(ctxt);
-		break;
-	case 0xe8: /* call (near) */ {
-		long int rel;
-		switch (c->op_bytes) {
-		case 2:
-			rel = insn_fetch(s16, 2, c->eip);
-			break;
-		case 4:
-			rel = insn_fetch(s32, 4, c->eip);
-			break;
-		default:
-			DPRINTF("Call: Invalid op_bytes\n");
-			goto cannot_emulate;
-		}
-		c->src.val = (unsigned long) c->eip;
-		JMP_REL(rel);
-		c->op_bytes = c->ad_bytes;
-		emulate_push(ctxt);
-		break;
-	}
-	case 0xe9: /* jmp rel */
-	case 0xeb: /* jmp rel short */
-		JMP_REL(c->src.val);
-		c->dst.type = OP_NONE; /* Disable writeback. */
-		break;
-	case 0xf4:              /* hlt */
-		ctxt->vcpu->arch.halt_request = 1;
-		goto done;
-	case 0xf5:	/* cmc */
-		/* complement carry flag from eflags reg */
-		ctxt->eflags ^= EFLG_CF;
-		c->dst.type = OP_NONE;	/* Disable writeback. */
-		break;
-	case 0xf6 ... 0xf7:	/* Grp3 */
-		rc = emulate_grp3(ctxt, ops);
-		if (rc != 0)
-			goto done;
-		break;
-	case 0xf8: /* clc */
-		ctxt->eflags &= ~EFLG_CF;
-		c->dst.type = OP_NONE;	/* Disable writeback. */
-		break;
-	case 0xfa: /* cli */
-		ctxt->eflags &= ~X86_EFLAGS_IF;
-		c->dst.type = OP_NONE;	/* Disable writeback. */
-		break;
-	case 0xfb: /* sti */
-		ctxt->eflags |= X86_EFLAGS_IF;
-		c->dst.type = OP_NONE;	/* Disable writeback. */
-		break;
-	case 0xfe ... 0xff:	/* Grp4/Grp5 */
-		rc = emulate_grp45(ctxt, ops);
-		if (rc != 0)
-			goto done;
-		break;
-	}
-
-writeback:
-	rc = writeback(ctxt, ops);
-	if (rc != 0)
-		goto done;
-
-	/* Commit shadow register state. */
-	memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs);
-	ctxt->vcpu->arch.rip = c->eip;
-
-done:
-	if (rc == X86EMUL_UNHANDLEABLE) {
-		c->eip = saved_eip;
-		return -1;
-	}
-	return 0;
-
-twobyte_insn:
-	switch (c->b) {
-	case 0x01: /* lgdt, lidt, lmsw */
-		switch (c->modrm_reg) {
-			u16 size;
-			unsigned long address;
-
-		case 0: /* vmcall */
-			if (c->modrm_mod != 3 || c->modrm_rm != 1)
-				goto cannot_emulate;
-
-			rc = kvm_fix_hypercall(ctxt->vcpu);
-			if (rc)
-				goto done;
-
-			kvm_emulate_hypercall(ctxt->vcpu);
-			break;
-		case 2: /* lgdt */
-			rc = read_descriptor(ctxt, ops, c->src.ptr,
-					     &size, &address, c->op_bytes);
-			if (rc)
-				goto done;
-			realmode_lgdt(ctxt->vcpu, size, address);
-			break;
-		case 3: /* lidt/vmmcall */
-			if (c->modrm_mod == 3 && c->modrm_rm == 1) {
-				rc = kvm_fix_hypercall(ctxt->vcpu);
-				if (rc)
-					goto done;
-				kvm_emulate_hypercall(ctxt->vcpu);
-			} else {
-				rc = read_descriptor(ctxt, ops, c->src.ptr,
-						     &size, &address,
-						     c->op_bytes);
-				if (rc)
-					goto done;
-				realmode_lidt(ctxt->vcpu, size, address);
-			}
-			break;
-		case 4: /* smsw */
-			if (c->modrm_mod != 3)
-				goto cannot_emulate;
-			*(u16 *)&c->regs[c->modrm_rm]
-				= realmode_get_cr(ctxt->vcpu, 0);
-			break;
-		case 6: /* lmsw */
-			if (c->modrm_mod != 3)
-				goto cannot_emulate;
-			realmode_lmsw(ctxt->vcpu, (u16)c->modrm_val,
-						  &ctxt->eflags);
-			break;
-		case 7: /* invlpg*/
-			emulate_invlpg(ctxt->vcpu, memop);
-			break;
-		default:
-			goto cannot_emulate;
-		}
-		/* Disable writeback. */
-		c->dst.type = OP_NONE;
-		break;
-	case 0x06:
-		emulate_clts(ctxt->vcpu);
-		c->dst.type = OP_NONE;
-		break;
-	case 0x08:		/* invd */
-	case 0x09:		/* wbinvd */
-	case 0x0d:		/* GrpP (prefetch) */
-	case 0x18:		/* Grp16 (prefetch/nop) */
-		c->dst.type = OP_NONE;
-		break;
-	case 0x20: /* mov cr, reg */
-		if (c->modrm_mod != 3)
-			goto cannot_emulate;
-		c->regs[c->modrm_rm] =
-				realmode_get_cr(ctxt->vcpu, c->modrm_reg);
-		c->dst.type = OP_NONE;	/* no writeback */
-		break;
-	case 0x21: /* mov from dr to reg */
-		if (c->modrm_mod != 3)
-			goto cannot_emulate;
-		rc = emulator_get_dr(ctxt, c->modrm_reg, &c->regs[c->modrm_rm]);
-		if (rc)
-			goto cannot_emulate;
-		c->dst.type = OP_NONE;	/* no writeback */
-		break;
-	case 0x22: /* mov reg, cr */
-		if (c->modrm_mod != 3)
-			goto cannot_emulate;
-		realmode_set_cr(ctxt->vcpu,
-				c->modrm_reg, c->modrm_val, &ctxt->eflags);
-		c->dst.type = OP_NONE;
-		break;
-	case 0x23: /* mov from reg to dr */
-		if (c->modrm_mod != 3)
-			goto cannot_emulate;
-		rc = emulator_set_dr(ctxt, c->modrm_reg,
-				     c->regs[c->modrm_rm]);
-		if (rc)
-			goto cannot_emulate;
-		c->dst.type = OP_NONE;	/* no writeback */
-		break;
-	case 0x30:
-		/* wrmsr */
-		msr_data = (u32)c->regs[VCPU_REGS_RAX]
-			| ((u64)c->regs[VCPU_REGS_RDX] << 32);
-		rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data);
-		if (rc) {
-			kvm_inject_gp(ctxt->vcpu, 0);
-			c->eip = ctxt->vcpu->arch.rip;
-		}
-		rc = X86EMUL_CONTINUE;
-		c->dst.type = OP_NONE;
-		break;
-	case 0x32:
-		/* rdmsr */
-		rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data);
-		if (rc) {
-			kvm_inject_gp(ctxt->vcpu, 0);
-			c->eip = ctxt->vcpu->arch.rip;
-		} else {
-			c->regs[VCPU_REGS_RAX] = (u32)msr_data;
-			c->regs[VCPU_REGS_RDX] = msr_data >> 32;
-		}
-		rc = X86EMUL_CONTINUE;
-		c->dst.type = OP_NONE;
-		break;
-	case 0x40 ... 0x4f:	/* cmov */
-		c->dst.val = c->dst.orig_val = c->src.val;
-		if (!test_cc(c->b, ctxt->eflags))
-			c->dst.type = OP_NONE; /* no writeback */
-		break;
-	case 0x80 ... 0x8f: /* jnz rel, etc*/ {
-		long int rel;
-
-		switch (c->op_bytes) {
-		case 2:
-			rel = insn_fetch(s16, 2, c->eip);
-			break;
-		case 4:
-			rel = insn_fetch(s32, 4, c->eip);
-			break;
-		case 8:
-			rel = insn_fetch(s64, 8, c->eip);
-			break;
-		default:
-			DPRINTF("jnz: Invalid op_bytes\n");
-			goto cannot_emulate;
-		}
-		if (test_cc(c->b, ctxt->eflags))
-			JMP_REL(rel);
-		c->dst.type = OP_NONE;
-		break;
-	}
-	case 0xa3:
-	      bt:		/* bt */
-		c->dst.type = OP_NONE;
-		/* only subword offset */
-		c->src.val &= (c->dst.bytes << 3) - 1;
-		emulate_2op_SrcV_nobyte("bt", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0xab:
-	      bts:		/* bts */
-		/* only subword offset */
-		c->src.val &= (c->dst.bytes << 3) - 1;
-		emulate_2op_SrcV_nobyte("bts", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0xb0 ... 0xb1:	/* cmpxchg */
-		/*
-		 * Save real source value, then compare EAX against
-		 * destination.
-		 */
-		c->src.orig_val = c->src.val;
-		c->src.val = c->regs[VCPU_REGS_RAX];
-		emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags);
-		if (ctxt->eflags & EFLG_ZF) {
-			/* Success: write back to memory. */
-			c->dst.val = c->src.orig_val;
-		} else {
-			/* Failure: write the value we saw to EAX. */
-			c->dst.type = OP_REG;
-			c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
-		}
-		break;
-	case 0xb3:
-	      btr:		/* btr */
-		/* only subword offset */
-		c->src.val &= (c->dst.bytes << 3) - 1;
-		emulate_2op_SrcV_nobyte("btr", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0xb6 ... 0xb7:	/* movzx */
-		c->dst.bytes = c->op_bytes;
-		c->dst.val = (c->d & ByteOp) ? (u8) c->src.val
-						       : (u16) c->src.val;
-		break;
-	case 0xba:		/* Grp8 */
-		switch (c->modrm_reg & 3) {
-		case 0:
-			goto bt;
-		case 1:
-			goto bts;
-		case 2:
-			goto btr;
-		case 3:
-			goto btc;
-		}
-		break;
-	case 0xbb:
-	      btc:		/* btc */
-		/* only subword offset */
-		c->src.val &= (c->dst.bytes << 3) - 1;
-		emulate_2op_SrcV_nobyte("btc", c->src, c->dst, ctxt->eflags);
-		break;
-	case 0xbe ... 0xbf:	/* movsx */
-		c->dst.bytes = c->op_bytes;
-		c->dst.val = (c->d & ByteOp) ? (s8) c->src.val :
-							(s16) c->src.val;
-		break;
-	case 0xc3:		/* movnti */
-		c->dst.bytes = c->op_bytes;
-		c->dst.val = (c->op_bytes == 4) ? (u32) c->src.val :
-							(u64) c->src.val;
-		break;
-	case 0xc7:		/* Grp9 (cmpxchg8b) */
-		rc = emulate_grp9(ctxt, ops, memop);
-		if (rc != 0)
-			goto done;
-		c->dst.type = OP_NONE;
-		break;
-	}
-	goto writeback;
-
-cannot_emulate:
-	DPRINTF("Cannot emulate %02x\n", c->b);
-	c->eip = saved_eip;
-	return -1;
-}
diff --git a/drivers/kvm/x86_emulate.h b/drivers/kvm/x86_emulate.h
deleted file mode 100644
index 7db91b9bdcd4..000000000000
--- a/drivers/kvm/x86_emulate.h
+++ /dev/null
@@ -1,186 +0,0 @@
-/******************************************************************************
- * x86_emulate.h
- *
- * Generic x86 (32-bit and 64-bit) instruction decoder and emulator.
- *
- * Copyright (c) 2005 Keir Fraser
- *
- * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4
- */
-
-#ifndef __X86_EMULATE_H__
-#define __X86_EMULATE_H__
-
-struct x86_emulate_ctxt;
-
-/*
- * x86_emulate_ops:
- *
- * These operations represent the instruction emulator's interface to memory.
- * There are two categories of operation: those that act on ordinary memory
- * regions (*_std), and those that act on memory regions known to require
- * special treatment or emulation (*_emulated).
- *
- * The emulator assumes that an instruction accesses only one 'emulated memory'
- * location, that this location is the given linear faulting address (cr2), and
- * that this is one of the instruction's data operands. Instruction fetches and
- * stack operations are assumed never to access emulated memory. The emulator
- * automatically deduces which operand of a string-move operation is accessing
- * emulated memory, and assumes that the other operand accesses normal memory.
- *
- * NOTES:
- *  1. The emulator isn't very smart about emulated vs. standard memory.
- *     'Emulated memory' access addresses should be checked for sanity.
- *     'Normal memory' accesses may fault, and the caller must arrange to
- *     detect and handle reentrancy into the emulator via recursive faults.
- *     Accesses may be unaligned and may cross page boundaries.
- *  2. If the access fails (cannot emulate, or a standard access faults) then
- *     it is up to the memop to propagate the fault to the guest VM via
- *     some out-of-band mechanism, unknown to the emulator. The memop signals
- *     failure by returning X86EMUL_PROPAGATE_FAULT to the emulator, which will
- *     then immediately bail.
- *  3. Valid access sizes are 1, 2, 4 and 8 bytes. On x86/32 systems only
- *     cmpxchg8b_emulated need support 8-byte accesses.
- *  4. The emulator cannot handle 64-bit mode emulation on an x86/32 system.
- */
-/* Access completed successfully: continue emulation as normal. */
-#define X86EMUL_CONTINUE        0
-/* Access is unhandleable: bail from emulation and return error to caller. */
-#define X86EMUL_UNHANDLEABLE    1
-/* Terminate emulation but return success to the caller. */
-#define X86EMUL_PROPAGATE_FAULT 2 /* propagate a generated fault to guest */
-#define X86EMUL_RETRY_INSTR     2 /* retry the instruction for some reason */
-#define X86EMUL_CMPXCHG_FAILED  2 /* cmpxchg did not see expected value */
-struct x86_emulate_ops {
-	/*
-	 * read_std: Read bytes of standard (non-emulated/special) memory.
-	 *           Used for instruction fetch, stack operations, and others.
-	 *  @addr:  [IN ] Linear address from which to read.
-	 *  @val:   [OUT] Value read from memory, zero-extended to 'u_long'.
-	 *  @bytes: [IN ] Number of bytes to read from memory.
-	 */
-	int (*read_std)(unsigned long addr, void *val,
-			unsigned int bytes, struct kvm_vcpu *vcpu);
-
-	/*
-	 * read_emulated: Read bytes from emulated/special memory area.
-	 *  @addr:  [IN ] Linear address from which to read.
-	 *  @val:   [OUT] Value read from memory, zero-extended to 'u_long'.
-	 *  @bytes: [IN ] Number of bytes to read from memory.
-	 */
-	int (*read_emulated) (unsigned long addr,
-			      void *val,
-			      unsigned int bytes,
-			      struct kvm_vcpu *vcpu);
-
-	/*
-	 * write_emulated: Read bytes from emulated/special memory area.
-	 *  @addr:  [IN ] Linear address to which to write.
-	 *  @val:   [IN ] Value to write to memory (low-order bytes used as
-	 *                required).
-	 *  @bytes: [IN ] Number of bytes to write to memory.
-	 */
-	int (*write_emulated) (unsigned long addr,
-			       const void *val,
-			       unsigned int bytes,
-			       struct kvm_vcpu *vcpu);
-
-	/*
-	 * cmpxchg_emulated: Emulate an atomic (LOCKed) CMPXCHG operation on an
-	 *                   emulated/special memory area.
-	 *  @addr:  [IN ] Linear address to access.
-	 *  @old:   [IN ] Value expected to be current at @addr.
-	 *  @new:   [IN ] Value to write to @addr.
-	 *  @bytes: [IN ] Number of bytes to access using CMPXCHG.
-	 */
-	int (*cmpxchg_emulated) (unsigned long addr,
-				 const void *old,
-				 const void *new,
-				 unsigned int bytes,
-				 struct kvm_vcpu *vcpu);
-
-};
-
-/* Type, address-of, and value of an instruction's operand. */
-struct operand {
-	enum { OP_REG, OP_MEM, OP_IMM, OP_NONE } type;
-	unsigned int bytes;
-	unsigned long val, orig_val, *ptr;
-};
-
-struct fetch_cache {
-	u8 data[15];
-	unsigned long start;
-	unsigned long end;
-};
-
-struct decode_cache {
-	u8 twobyte;
-	u8 b;
-	u8 lock_prefix;
-	u8 rep_prefix;
-	u8 op_bytes;
-	u8 ad_bytes;
-	u8 rex_prefix;
-	struct operand src;
-	struct operand dst;
-	unsigned long *override_base;
-	unsigned int d;
-	unsigned long regs[NR_VCPU_REGS];
-	unsigned long eip;
-	/* modrm */
-	u8 modrm;
-	u8 modrm_mod;
-	u8 modrm_reg;
-	u8 modrm_rm;
-	u8 use_modrm_ea;
-	unsigned long modrm_ea;
-	unsigned long modrm_val;
-	struct fetch_cache fetch;
-};
-
-struct x86_emulate_ctxt {
-	/* Register state before/after emulation. */
-	struct kvm_vcpu *vcpu;
-
-	/* Linear faulting address (if emulating a page-faulting instruction). */
-	unsigned long eflags;
-
-	/* Emulated execution mode, represented by an X86EMUL_MODE value. */
-	int mode;
-
-	unsigned long cs_base;
-	unsigned long ds_base;
-	unsigned long es_base;
-	unsigned long ss_base;
-	unsigned long gs_base;
-	unsigned long fs_base;
-
-	/* decode cache */
-
-	struct decode_cache decode;
-};
-
-/* Repeat String Operation Prefix */
-#define REPE_PREFIX  1
-#define REPNE_PREFIX    2
-
-/* Execution mode, passed to the emulator. */
-#define X86EMUL_MODE_REAL     0	/* Real mode.             */
-#define X86EMUL_MODE_PROT16   2	/* 16-bit protected mode. */
-#define X86EMUL_MODE_PROT32   4	/* 32-bit protected mode. */
-#define X86EMUL_MODE_PROT64   8	/* 64-bit (long) mode.    */
-
-/* Host execution mode. */
-#if defined(__i386__)
-#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT32
-#elif defined(CONFIG_X86_64)
-#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT64
-#endif
-
-int x86_decode_insn(struct x86_emulate_ctxt *ctxt,
-		    struct x86_emulate_ops *ops);
-int x86_emulate_insn(struct x86_emulate_ctxt *ctxt,
-		     struct x86_emulate_ops *ops);
-
-#endif				/* __X86_EMULATE_H__ */
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
new file mode 100644
index 000000000000..28940e1a9713
--- /dev/null
+++ b/include/asm-x86/kvm_host.h
@@ -0,0 +1,601 @@
+#/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * This header defines architecture specific interfaces, x86 version
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#ifndef ASM_KVM_HOST_H
+#define ASM_KVM_HOST_H
+
+#include <linux/types.h>
+#include <linux/mm.h>
+
+#include <linux/kvm.h>
+#include <linux/kvm_para.h>
+#include <linux/kvm_types.h>
+
+#include <asm/desc.h>
+
+#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
+#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
+#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS|0xFFFFFF0000000000ULL)
+
+#define KVM_GUEST_CR0_MASK \
+	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE \
+	 | X86_CR0_NW | X86_CR0_CD)
+#define KVM_VM_CR0_ALWAYS_ON \
+	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE | X86_CR0_TS \
+	 | X86_CR0_MP)
+#define KVM_GUEST_CR4_MASK \
+	(X86_CR4_VME | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE | X86_CR4_VMXE)
+#define KVM_PMODE_VM_CR4_ALWAYS_ON (X86_CR4_PAE | X86_CR4_VMXE)
+#define KVM_RMODE_VM_CR4_ALWAYS_ON (X86_CR4_VME | X86_CR4_PAE | X86_CR4_VMXE)
+
+#define INVALID_PAGE (~(hpa_t)0)
+#define UNMAPPED_GVA (~(gpa_t)0)
+
+#define DE_VECTOR 0
+#define UD_VECTOR 6
+#define NM_VECTOR 7
+#define DF_VECTOR 8
+#define TS_VECTOR 10
+#define NP_VECTOR 11
+#define SS_VECTOR 12
+#define GP_VECTOR 13
+#define PF_VECTOR 14
+
+#define SELECTOR_TI_MASK (1 << 2)
+#define SELECTOR_RPL_MASK 0x03
+
+#define IOPL_SHIFT 12
+
+#define KVM_ALIAS_SLOTS 4
+
+#define KVM_PERMILLE_MMU_PAGES 20
+#define KVM_MIN_ALLOC_MMU_PAGES 64
+#define KVM_NUM_MMU_PAGES 1024
+#define KVM_MIN_FREE_MMU_PAGES 5
+#define KVM_REFILL_PAGES 25
+#define KVM_MAX_CPUID_ENTRIES 40
+
+extern spinlock_t kvm_lock;
+extern struct list_head vm_list;
+
+struct kvm_vcpu;
+struct kvm;
+
+enum {
+	VCPU_REGS_RAX = 0,
+	VCPU_REGS_RCX = 1,
+	VCPU_REGS_RDX = 2,
+	VCPU_REGS_RBX = 3,
+	VCPU_REGS_RSP = 4,
+	VCPU_REGS_RBP = 5,
+	VCPU_REGS_RSI = 6,
+	VCPU_REGS_RDI = 7,
+#ifdef CONFIG_X86_64
+	VCPU_REGS_R8 = 8,
+	VCPU_REGS_R9 = 9,
+	VCPU_REGS_R10 = 10,
+	VCPU_REGS_R11 = 11,
+	VCPU_REGS_R12 = 12,
+	VCPU_REGS_R13 = 13,
+	VCPU_REGS_R14 = 14,
+	VCPU_REGS_R15 = 15,
+#endif
+	NR_VCPU_REGS
+};
+
+enum {
+	VCPU_SREG_CS,
+	VCPU_SREG_DS,
+	VCPU_SREG_ES,
+	VCPU_SREG_FS,
+	VCPU_SREG_GS,
+	VCPU_SREG_SS,
+	VCPU_SREG_TR,
+	VCPU_SREG_LDTR,
+};
+
+#include <asm/kvm_x86_emulate.h>
+
+#define KVM_NR_MEM_OBJS 40
+
+/*
+ * We don't want allocation failures within the mmu code, so we preallocate
+ * enough memory for a single page fault in a cache.
+ */
+struct kvm_mmu_memory_cache {
+	int nobjs;
+	void *objects[KVM_NR_MEM_OBJS];
+};
+
+#define NR_PTE_CHAIN_ENTRIES 5
+
+struct kvm_pte_chain {
+	u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
+	struct hlist_node link;
+};
+
+/*
+ * kvm_mmu_page_role, below, is defined as:
+ *
+ *   bits 0:3 - total guest paging levels (2-4, or zero for real mode)
+ *   bits 4:7 - page table level for this shadow (1-4)
+ *   bits 8:9 - page table quadrant for 2-level guests
+ *   bit   16 - "metaphysical" - gfn is not a real page (huge page/real mode)
+ *   bits 17:19 - common access permissions for all ptes in this shadow page
+ */
+union kvm_mmu_page_role {
+	unsigned word;
+	struct {
+		unsigned glevels : 4;
+		unsigned level : 4;
+		unsigned quadrant : 2;
+		unsigned pad_for_nice_hex_output : 6;
+		unsigned metaphysical : 1;
+		unsigned access : 3;
+	};
+};
+
+struct kvm_mmu_page {
+	struct list_head link;
+	struct hlist_node hash_link;
+
+	/*
+	 * The following two entries are used to key the shadow page in the
+	 * hash table.
+	 */
+	gfn_t gfn;
+	union kvm_mmu_page_role role;
+
+	u64 *spt;
+	/* hold the gfn of each spte inside spt */
+	gfn_t *gfns;
+	unsigned long slot_bitmap; /* One bit set per slot which has memory
+				    * in this shadow page.
+				    */
+	int multimapped;         /* More than one parent_pte? */
+	int root_count;          /* Currently serving as active root */
+	union {
+		u64 *parent_pte;               /* !multimapped */
+		struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
+	};
+};
+
+/*
+ * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
+ * 32-bit).  The kvm_mmu structure abstracts the details of the current mmu
+ * mode.
+ */
+struct kvm_mmu {
+	void (*new_cr3)(struct kvm_vcpu *vcpu);
+	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err);
+	void (*free)(struct kvm_vcpu *vcpu);
+	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva);
+	void (*prefetch_page)(struct kvm_vcpu *vcpu,
+			      struct kvm_mmu_page *page);
+	hpa_t root_hpa;
+	int root_level;
+	int shadow_root_level;
+
+	u64 *pae_root;
+};
+
+struct kvm_vcpu_arch {
+	u64 host_tsc;
+	int interrupt_window_open;
+	unsigned long irq_summary; /* bit vector: 1 per word in irq_pending */
+	DECLARE_BITMAP(irq_pending, KVM_NR_INTERRUPTS);
+	unsigned long regs[NR_VCPU_REGS]; /* for rsp: vcpu_load_rsp_rip() */
+	unsigned long rip;      /* needs vcpu_load_rsp_rip() */
+
+	unsigned long cr0;
+	unsigned long cr2;
+	unsigned long cr3;
+	unsigned long cr4;
+	unsigned long cr8;
+	u64 pdptrs[4]; /* pae */
+	u64 shadow_efer;
+	u64 apic_base;
+	struct kvm_lapic *apic;    /* kernel irqchip context */
+#define VCPU_MP_STATE_RUNNABLE          0
+#define VCPU_MP_STATE_UNINITIALIZED     1
+#define VCPU_MP_STATE_INIT_RECEIVED     2
+#define VCPU_MP_STATE_SIPI_RECEIVED     3
+#define VCPU_MP_STATE_HALTED            4
+	int mp_state;
+	int sipi_vector;
+	u64 ia32_misc_enable_msr;
+
+	struct kvm_mmu mmu;
+
+	struct kvm_mmu_memory_cache mmu_pte_chain_cache;
+	struct kvm_mmu_memory_cache mmu_rmap_desc_cache;
+	struct kvm_mmu_memory_cache mmu_page_cache;
+	struct kvm_mmu_memory_cache mmu_page_header_cache;
+
+	gfn_t last_pt_write_gfn;
+	int   last_pt_write_count;
+	u64  *last_pte_updated;
+
+	struct i387_fxsave_struct host_fx_image;
+	struct i387_fxsave_struct guest_fx_image;
+
+	gva_t mmio_fault_cr2;
+	struct kvm_pio_request pio;
+	void *pio_data;
+
+	struct kvm_queued_exception {
+		bool pending;
+		bool has_error_code;
+		u8 nr;
+		u32 error_code;
+	} exception;
+
+	struct {
+		int active;
+		u8 save_iopl;
+		struct kvm_save_segment {
+			u16 selector;
+			unsigned long base;
+			u32 limit;
+			u32 ar;
+		} tr, es, ds, fs, gs;
+	} rmode;
+	int halt_request; /* real mode on Intel only */
+
+	int cpuid_nent;
+	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
+	/* emulate context */
+
+	struct x86_emulate_ctxt emulate_ctxt;
+};
+
+struct kvm_mem_alias {
+	gfn_t base_gfn;
+	unsigned long npages;
+	gfn_t target_gfn;
+};
+
+struct kvm_arch{
+	int naliases;
+	struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
+
+	unsigned int n_free_mmu_pages;
+	unsigned int n_requested_mmu_pages;
+	unsigned int n_alloc_mmu_pages;
+	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
+	/*
+	 * Hash table of struct kvm_mmu_page.
+	 */
+	struct list_head active_mmu_pages;
+	struct kvm_pic *vpic;
+	struct kvm_ioapic *vioapic;
+
+	int round_robin_prev_vcpu;
+	unsigned int tss_addr;
+	struct page *apic_access_page;
+};
+
+struct kvm_vm_stat {
+	u32 mmu_shadow_zapped;
+	u32 mmu_pte_write;
+	u32 mmu_pte_updated;
+	u32 mmu_pde_zapped;
+	u32 mmu_flooded;
+	u32 mmu_recycled;
+	u32 remote_tlb_flush;
+};
+
+struct kvm_vcpu_stat {
+	u32 pf_fixed;
+	u32 pf_guest;
+	u32 tlb_flush;
+	u32 invlpg;
+
+	u32 exits;
+	u32 io_exits;
+	u32 mmio_exits;
+	u32 signal_exits;
+	u32 irq_window_exits;
+	u32 halt_exits;
+	u32 halt_wakeup;
+	u32 request_irq_exits;
+	u32 irq_exits;
+	u32 host_state_reload;
+	u32 efer_reload;
+	u32 fpu_reload;
+	u32 insn_emulation;
+	u32 insn_emulation_fail;
+};
+
+struct descriptor_table {
+	u16 limit;
+	unsigned long base;
+} __attribute__((packed));
+
+struct kvm_x86_ops {
+	int (*cpu_has_kvm_support)(void);          /* __init */
+	int (*disabled_by_bios)(void);             /* __init */
+	void (*hardware_enable)(void *dummy);      /* __init */
+	void (*hardware_disable)(void *dummy);
+	void (*check_processor_compatibility)(void *rtn);
+	int (*hardware_setup)(void);               /* __init */
+	void (*hardware_unsetup)(void);            /* __exit */
+
+	/* Create, but do not attach this VCPU */
+	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
+	void (*vcpu_free)(struct kvm_vcpu *vcpu);
+	int (*vcpu_reset)(struct kvm_vcpu *vcpu);
+
+	void (*prepare_guest_switch)(struct kvm_vcpu *vcpu);
+	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
+	void (*vcpu_put)(struct kvm_vcpu *vcpu);
+	void (*vcpu_decache)(struct kvm_vcpu *vcpu);
+
+	int (*set_guest_debug)(struct kvm_vcpu *vcpu,
+			       struct kvm_debug_guest *dbg);
+	void (*guest_debug_pre)(struct kvm_vcpu *vcpu);
+	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
+	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
+	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
+	void (*get_segment)(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg);
+	void (*set_segment)(struct kvm_vcpu *vcpu,
+			    struct kvm_segment *var, int seg);
+	void (*get_cs_db_l_bits)(struct kvm_vcpu *vcpu, int *db, int *l);
+	void (*decache_cr4_guest_bits)(struct kvm_vcpu *vcpu);
+	void (*set_cr0)(struct kvm_vcpu *vcpu, unsigned long cr0);
+	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);
+	void (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
+	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
+	void (*get_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
+	void (*set_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
+	void (*get_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
+	void (*set_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt);
+	unsigned long (*get_dr)(struct kvm_vcpu *vcpu, int dr);
+	void (*set_dr)(struct kvm_vcpu *vcpu, int dr, unsigned long value,
+		       int *exception);
+	void (*cache_regs)(struct kvm_vcpu *vcpu);
+	void (*decache_regs)(struct kvm_vcpu *vcpu);
+	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
+	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
+
+	void (*tlb_flush)(struct kvm_vcpu *vcpu);
+
+	void (*run)(struct kvm_vcpu *vcpu, struct kvm_run *run);
+	int (*handle_exit)(struct kvm_run *run, struct kvm_vcpu *vcpu);
+	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
+	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
+				unsigned char *hypercall_addr);
+	int (*get_irq)(struct kvm_vcpu *vcpu);
+	void (*set_irq)(struct kvm_vcpu *vcpu, int vec);
+	void (*queue_exception)(struct kvm_vcpu *vcpu, unsigned nr,
+				bool has_error_code, u32 error_code);
+	bool (*exception_injected)(struct kvm_vcpu *vcpu);
+	void (*inject_pending_irq)(struct kvm_vcpu *vcpu);
+	void (*inject_pending_vectors)(struct kvm_vcpu *vcpu,
+				       struct kvm_run *run);
+
+	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
+};
+
+extern struct kvm_x86_ops *kvm_x86_ops;
+
+int kvm_mmu_module_init(void);
+void kvm_mmu_module_exit(void);
+
+void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
+int kvm_mmu_create(struct kvm_vcpu *vcpu);
+int kvm_mmu_setup(struct kvm_vcpu *vcpu);
+void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
+
+int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
+void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
+void kvm_mmu_zap_all(struct kvm *kvm);
+unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
+void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);
+
+enum emulation_result {
+	EMULATE_DONE,       /* no further processing */
+	EMULATE_DO_MMIO,      /* kvm_run filled with mmio request */
+	EMULATE_FAIL,         /* can't emulate this instruction */
+};
+
+int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			unsigned long cr2, u16 error_code, int no_decode);
+void kvm_report_emulation_failure(struct kvm_vcpu *cvpu, const char *context);
+void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
+void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
+void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw,
+		   unsigned long *rflags);
+
+unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr);
+void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long value,
+		     unsigned long *rflags);
+int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *data);
+int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
+
+struct x86_emulate_ctxt;
+
+int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
+		     int size, unsigned port);
+int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
+			   int size, unsigned long count, int down,
+			    gva_t address, int rep, unsigned port);
+void kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
+int kvm_emulate_halt(struct kvm_vcpu *vcpu);
+int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address);
+int emulate_clts(struct kvm_vcpu *vcpu);
+int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr,
+		    unsigned long *dest);
+int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr,
+		    unsigned long value);
+
+void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
+void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr0);
+void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr0);
+void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr0);
+unsigned long get_cr8(struct kvm_vcpu *vcpu);
+void lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
+void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
+
+int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata);
+int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data);
+
+void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
+void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
+void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2,
+			   u32 error_code);
+
+void fx_init(struct kvm_vcpu *vcpu);
+
+int emulator_read_std(unsigned long addr,
+		      void *val,
+		      unsigned int bytes,
+		      struct kvm_vcpu *vcpu);
+int emulator_write_emulated(unsigned long addr,
+			    const void *val,
+			    unsigned int bytes,
+			    struct kvm_vcpu *vcpu);
+
+unsigned long segment_base(u16 selector);
+
+void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
+void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
+		       const u8 *new, int bytes);
+int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
+void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
+int kvm_mmu_load(struct kvm_vcpu *vcpu);
+void kvm_mmu_unload(struct kvm_vcpu *vcpu);
+
+int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);
+
+int kvm_fix_hypercall(struct kvm_vcpu *vcpu);
+
+int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code);
+
+int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3);
+int complete_pio(struct kvm_vcpu *vcpu);
+
+static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
+{
+	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);
+
+	return (struct kvm_mmu_page *)page_private(page);
+}
+
+static inline u16 read_fs(void)
+{
+	u16 seg;
+	asm("mov %%fs, %0" : "=g"(seg));
+	return seg;
+}
+
+static inline u16 read_gs(void)
+{
+	u16 seg;
+	asm("mov %%gs, %0" : "=g"(seg));
+	return seg;
+}
+
+static inline u16 read_ldt(void)
+{
+	u16 ldt;
+	asm("sldt %0" : "=g"(ldt));
+	return ldt;
+}
+
+static inline void load_fs(u16 sel)
+{
+	asm("mov %0, %%fs" : : "rm"(sel));
+}
+
+static inline void load_gs(u16 sel)
+{
+	asm("mov %0, %%gs" : : "rm"(sel));
+}
+
+#ifndef load_ldt
+static inline void load_ldt(u16 sel)
+{
+	asm("lldt %0" : : "rm"(sel));
+}
+#endif
+
+static inline void get_idt(struct descriptor_table *table)
+{
+	asm("sidt %0" : "=m"(*table));
+}
+
+static inline void get_gdt(struct descriptor_table *table)
+{
+	asm("sgdt %0" : "=m"(*table));
+}
+
+static inline unsigned long read_tr_base(void)
+{
+	u16 tr;
+	asm("str %0" : "=g"(tr));
+	return segment_base(tr);
+}
+
+#ifdef CONFIG_X86_64
+static inline unsigned long read_msr(unsigned long msr)
+{
+	u64 value;
+
+	rdmsrl(msr, value);
+	return value;
+}
+#endif
+
+static inline void fx_save(struct i387_fxsave_struct *image)
+{
+	asm("fxsave (%0)":: "r" (image));
+}
+
+static inline void fx_restore(struct i387_fxsave_struct *image)
+{
+	asm("fxrstor (%0)":: "r" (image));
+}
+
+static inline void fpu_init(void)
+{
+	asm("finit");
+}
+
+static inline u32 get_rdx_init_val(void)
+{
+	return 0x600; /* P6 family */
+}
+
+static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
+{
+	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
+}
+
+#define ASM_VMX_VMCLEAR_RAX       ".byte 0x66, 0x0f, 0xc7, 0x30"
+#define ASM_VMX_VMLAUNCH          ".byte 0x0f, 0x01, 0xc2"
+#define ASM_VMX_VMRESUME          ".byte 0x0f, 0x01, 0xc3"
+#define ASM_VMX_VMPTRLD_RAX       ".byte 0x0f, 0xc7, 0x30"
+#define ASM_VMX_VMREAD_RDX_RAX    ".byte 0x0f, 0x78, 0xd0"
+#define ASM_VMX_VMWRITE_RAX_RDX   ".byte 0x0f, 0x79, 0xd0"
+#define ASM_VMX_VMWRITE_RSP_RDX   ".byte 0x0f, 0x79, 0xd4"
+#define ASM_VMX_VMXOFF            ".byte 0x0f, 0x01, 0xc4"
+#define ASM_VMX_VMXON_RAX         ".byte 0xf3, 0x0f, 0xc7, 0x30"
+
+#define MSR_IA32_TIME_STAMP_COUNTER		0x010
+
+#define TSS_IOPB_BASE_OFFSET 0x66
+#define TSS_BASE_SIZE 0x68
+#define TSS_IOPB_SIZE (65536 / 8)
+#define TSS_REDIRECTION_SIZE (256 / 8)
+#define RMODE_TSS_SIZE (TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
+
+#endif
diff --git a/include/asm-x86/kvm_x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h
new file mode 100644
index 000000000000..7db91b9bdcd4
--- /dev/null
+++ b/include/asm-x86/kvm_x86_emulate.h
@@ -0,0 +1,186 @@
+/******************************************************************************
+ * x86_emulate.h
+ *
+ * Generic x86 (32-bit and 64-bit) instruction decoder and emulator.
+ *
+ * Copyright (c) 2005 Keir Fraser
+ *
+ * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4
+ */
+
+#ifndef __X86_EMULATE_H__
+#define __X86_EMULATE_H__
+
+struct x86_emulate_ctxt;
+
+/*
+ * x86_emulate_ops:
+ *
+ * These operations represent the instruction emulator's interface to memory.
+ * There are two categories of operation: those that act on ordinary memory
+ * regions (*_std), and those that act on memory regions known to require
+ * special treatment or emulation (*_emulated).
+ *
+ * The emulator assumes that an instruction accesses only one 'emulated memory'
+ * location, that this location is the given linear faulting address (cr2), and
+ * that this is one of the instruction's data operands. Instruction fetches and
+ * stack operations are assumed never to access emulated memory. The emulator
+ * automatically deduces which operand of a string-move operation is accessing
+ * emulated memory, and assumes that the other operand accesses normal memory.
+ *
+ * NOTES:
+ *  1. The emulator isn't very smart about emulated vs. standard memory.
+ *     'Emulated memory' access addresses should be checked for sanity.
+ *     'Normal memory' accesses may fault, and the caller must arrange to
+ *     detect and handle reentrancy into the emulator via recursive faults.
+ *     Accesses may be unaligned and may cross page boundaries.
+ *  2. If the access fails (cannot emulate, or a standard access faults) then
+ *     it is up to the memop to propagate the fault to the guest VM via
+ *     some out-of-band mechanism, unknown to the emulator. The memop signals
+ *     failure by returning X86EMUL_PROPAGATE_FAULT to the emulator, which will
+ *     then immediately bail.
+ *  3. Valid access sizes are 1, 2, 4 and 8 bytes. On x86/32 systems only
+ *     cmpxchg8b_emulated need support 8-byte accesses.
+ *  4. The emulator cannot handle 64-bit mode emulation on an x86/32 system.
+ */
+/* Access completed successfully: continue emulation as normal. */
+#define X86EMUL_CONTINUE        0
+/* Access is unhandleable: bail from emulation and return error to caller. */
+#define X86EMUL_UNHANDLEABLE    1
+/* Terminate emulation but return success to the caller. */
+#define X86EMUL_PROPAGATE_FAULT 2 /* propagate a generated fault to guest */
+#define X86EMUL_RETRY_INSTR     2 /* retry the instruction for some reason */
+#define X86EMUL_CMPXCHG_FAILED  2 /* cmpxchg did not see expected value */
+struct x86_emulate_ops {
+	/*
+	 * read_std: Read bytes of standard (non-emulated/special) memory.
+	 *           Used for instruction fetch, stack operations, and others.
+	 *  @addr:  [IN ] Linear address from which to read.
+	 *  @val:   [OUT] Value read from memory, zero-extended to 'u_long'.
+	 *  @bytes: [IN ] Number of bytes to read from memory.
+	 */
+	int (*read_std)(unsigned long addr, void *val,
+			unsigned int bytes, struct kvm_vcpu *vcpu);
+
+	/*
+	 * read_emulated: Read bytes from emulated/special memory area.
+	 *  @addr:  [IN ] Linear address from which to read.
+	 *  @val:   [OUT] Value read from memory, zero-extended to 'u_long'.
+	 *  @bytes: [IN ] Number of bytes to read from memory.
+	 */
+	int (*read_emulated) (unsigned long addr,
+			      void *val,
+			      unsigned int bytes,
+			      struct kvm_vcpu *vcpu);
+
+	/*
+	 * write_emulated: Read bytes from emulated/special memory area.
+	 *  @addr:  [IN ] Linear address to which to write.
+	 *  @val:   [IN ] Value to write to memory (low-order bytes used as
+	 *                required).
+	 *  @bytes: [IN ] Number of bytes to write to memory.
+	 */
+	int (*write_emulated) (unsigned long addr,
+			       const void *val,
+			       unsigned int bytes,
+			       struct kvm_vcpu *vcpu);
+
+	/*
+	 * cmpxchg_emulated: Emulate an atomic (LOCKed) CMPXCHG operation on an
+	 *                   emulated/special memory area.
+	 *  @addr:  [IN ] Linear address to access.
+	 *  @old:   [IN ] Value expected to be current at @addr.
+	 *  @new:   [IN ] Value to write to @addr.
+	 *  @bytes: [IN ] Number of bytes to access using CMPXCHG.
+	 */
+	int (*cmpxchg_emulated) (unsigned long addr,
+				 const void *old,
+				 const void *new,
+				 unsigned int bytes,
+				 struct kvm_vcpu *vcpu);
+
+};
+
+/* Type, address-of, and value of an instruction's operand. */
+struct operand {
+	enum { OP_REG, OP_MEM, OP_IMM, OP_NONE } type;
+	unsigned int bytes;
+	unsigned long val, orig_val, *ptr;
+};
+
+struct fetch_cache {
+	u8 data[15];
+	unsigned long start;
+	unsigned long end;
+};
+
+struct decode_cache {
+	u8 twobyte;
+	u8 b;
+	u8 lock_prefix;
+	u8 rep_prefix;
+	u8 op_bytes;
+	u8 ad_bytes;
+	u8 rex_prefix;
+	struct operand src;
+	struct operand dst;
+	unsigned long *override_base;
+	unsigned int d;
+	unsigned long regs[NR_VCPU_REGS];
+	unsigned long eip;
+	/* modrm */
+	u8 modrm;
+	u8 modrm_mod;
+	u8 modrm_reg;
+	u8 modrm_rm;
+	u8 use_modrm_ea;
+	unsigned long modrm_ea;
+	unsigned long modrm_val;
+	struct fetch_cache fetch;
+};
+
+struct x86_emulate_ctxt {
+	/* Register state before/after emulation. */
+	struct kvm_vcpu *vcpu;
+
+	/* Linear faulting address (if emulating a page-faulting instruction). */
+	unsigned long eflags;
+
+	/* Emulated execution mode, represented by an X86EMUL_MODE value. */
+	int mode;
+
+	unsigned long cs_base;
+	unsigned long ds_base;
+	unsigned long es_base;
+	unsigned long ss_base;
+	unsigned long gs_base;
+	unsigned long fs_base;
+
+	/* decode cache */
+
+	struct decode_cache decode;
+};
+
+/* Repeat String Operation Prefix */
+#define REPE_PREFIX  1
+#define REPNE_PREFIX    2
+
+/* Execution mode, passed to the emulator. */
+#define X86EMUL_MODE_REAL     0	/* Real mode.             */
+#define X86EMUL_MODE_PROT16   2	/* 16-bit protected mode. */
+#define X86EMUL_MODE_PROT32   4	/* 32-bit protected mode. */
+#define X86EMUL_MODE_PROT64   8	/* 64-bit (long) mode.    */
+
+/* Host execution mode. */
+#if defined(__i386__)
+#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT32
+#elif defined(CONFIG_X86_64)
+#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT64
+#endif
+
+int x86_decode_insn(struct x86_emulate_ctxt *ctxt,
+		    struct x86_emulate_ops *ops);
+int x86_emulate_insn(struct x86_emulate_ctxt *ctxt,
+		     struct x86_emulate_ops *ops);
+
+#endif				/* __X86_EMULATE_H__ */
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
new file mode 100644
index 000000000000..a85d5b6943de
--- /dev/null
+++ b/include/linux/kvm_host.h
@@ -0,0 +1,289 @@
+#ifndef __KVM_HOST_H
+#define __KVM_HOST_H
+
+/*
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/types.h>
+#include <linux/hardirq.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/preempt.h>
+#include <asm/signal.h>
+
+#include <linux/kvm.h>
+#include <linux/kvm_para.h>
+
+#include <linux/kvm_types.h>
+
+#include <asm/kvm_host.h>
+
+#define KVM_MAX_VCPUS 4
+#define KVM_MEMORY_SLOTS 8
+/* memory slots that does not exposed to userspace */
+#define KVM_PRIVATE_MEM_SLOTS 4
+
+#define KVM_PIO_PAGE_OFFSET 1
+
+/*
+ * vcpu->requests bit members
+ */
+#define KVM_REQ_TLB_FLUSH          0
+
+
+struct kvm_vcpu;
+extern struct kmem_cache *kvm_vcpu_cache;
+
+struct kvm_guest_debug {
+	int enabled;
+	unsigned long bp[4];
+	int singlestep;
+};
+
+/*
+ * It would be nice to use something smarter than a linear search, TBD...
+ * Thankfully we dont expect many devices to register (famous last words :),
+ * so until then it will suffice.  At least its abstracted so we can change
+ * in one place.
+ */
+struct kvm_io_bus {
+	int                   dev_count;
+#define NR_IOBUS_DEVS 6
+	struct kvm_io_device *devs[NR_IOBUS_DEVS];
+};
+
+void kvm_io_bus_init(struct kvm_io_bus *bus);
+void kvm_io_bus_destroy(struct kvm_io_bus *bus);
+struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr);
+void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
+			     struct kvm_io_device *dev);
+
+struct kvm_vcpu {
+	struct kvm *kvm;
+	struct preempt_notifier preempt_notifier;
+	int vcpu_id;
+	struct mutex mutex;
+	int   cpu;
+	struct kvm_run *run;
+	int guest_mode;
+	unsigned long requests;
+	struct kvm_guest_debug guest_debug;
+	int fpu_active;
+	int guest_fpu_loaded;
+	wait_queue_head_t wq;
+	int sigset_active;
+	sigset_t sigset;
+	struct kvm_vcpu_stat stat;
+
+#ifdef CONFIG_HAS_IOMEM
+	int mmio_needed;
+	int mmio_read_completed;
+	int mmio_is_write;
+	int mmio_size;
+	unsigned char mmio_data[8];
+	gpa_t mmio_phys_addr;
+#endif
+
+	struct kvm_vcpu_arch arch;
+};
+
+struct kvm_memory_slot {
+	gfn_t base_gfn;
+	unsigned long npages;
+	unsigned long flags;
+	unsigned long *rmap;
+	unsigned long *dirty_bitmap;
+	unsigned long userspace_addr;
+	int user_alloc;
+};
+
+struct kvm {
+	struct mutex lock; /* protects everything except vcpus */
+	struct mm_struct *mm; /* userspace tied to this vm */
+	int nmemslots;
+	struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
+					KVM_PRIVATE_MEM_SLOTS];
+	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
+	struct list_head vm_list;
+	struct file *filp;
+	struct kvm_io_bus mmio_bus;
+	struct kvm_io_bus pio_bus;
+	struct kvm_vm_stat stat;
+	struct kvm_arch arch;
+};
+
+/* The guest did something we don't support. */
+#define pr_unimpl(vcpu, fmt, ...)					\
+ do {									\
+	if (printk_ratelimit())						\
+		printk(KERN_ERR "kvm: %i: cpu%i " fmt,			\
+		       current->tgid, (vcpu)->vcpu_id , ## __VA_ARGS__); \
+ } while (0)
+
+#define kvm_printf(kvm, fmt ...) printk(KERN_DEBUG fmt)
+#define vcpu_printf(vcpu, fmt...) kvm_printf(vcpu->kvm, fmt)
+
+int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id);
+void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);
+
+void vcpu_load(struct kvm_vcpu *vcpu);
+void vcpu_put(struct kvm_vcpu *vcpu);
+
+void decache_vcpus_on_cpu(int cpu);
+
+
+int kvm_init(void *opaque, unsigned int vcpu_size,
+		  struct module *module);
+void kvm_exit(void);
+
+#define HPA_MSB ((sizeof(hpa_t) * 8) - 1)
+#define HPA_ERR_MASK ((hpa_t)1 << HPA_MSB)
+static inline int is_error_hpa(hpa_t hpa) { return hpa >> HPA_MSB; }
+struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva);
+
+extern struct page *bad_page;
+
+int is_error_page(struct page *page);
+int kvm_is_error_hva(unsigned long addr);
+int kvm_set_memory_region(struct kvm *kvm,
+			  struct kvm_userspace_memory_region *mem,
+			  int user_alloc);
+int __kvm_set_memory_region(struct kvm *kvm,
+			    struct kvm_userspace_memory_region *mem,
+			    int user_alloc);
+int kvm_arch_set_memory_region(struct kvm *kvm,
+				struct kvm_userspace_memory_region *mem,
+				struct kvm_memory_slot old,
+				int user_alloc);
+gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn);
+struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
+void kvm_release_page_clean(struct page *page);
+void kvm_release_page_dirty(struct page *page);
+int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
+			int len);
+int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
+int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
+			 int offset, int len);
+int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
+		    unsigned long len);
+int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len);
+int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len);
+struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn);
+int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn);
+void mark_page_dirty(struct kvm *kvm, gfn_t gfn);
+
+void kvm_vcpu_block(struct kvm_vcpu *vcpu);
+void kvm_resched(struct kvm_vcpu *vcpu);
+void kvm_load_guest_fpu(struct kvm_vcpu *vcpu);
+void kvm_put_guest_fpu(struct kvm_vcpu *vcpu);
+void kvm_flush_remote_tlbs(struct kvm *kvm);
+
+long kvm_arch_dev_ioctl(struct file *filp,
+			unsigned int ioctl, unsigned long arg);
+long kvm_arch_vcpu_ioctl(struct file *filp,
+			 unsigned int ioctl, unsigned long arg);
+void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
+void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
+
+int kvm_dev_ioctl_check_extension(long ext);
+
+int kvm_get_dirty_log(struct kvm *kvm,
+			struct kvm_dirty_log *log, int *is_dirty);
+int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
+				struct kvm_dirty_log *log);
+
+int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
+				   struct
+				   kvm_userspace_memory_region *mem,
+				   int user_alloc);
+long kvm_arch_vm_ioctl(struct file *filp,
+		       unsigned int ioctl, unsigned long arg);
+void kvm_arch_destroy_vm(struct kvm *kvm);
+
+int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
+int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
+
+int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
+				    struct kvm_translation *tr);
+
+int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
+int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
+int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs);
+int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs);
+int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
+				    struct kvm_debug_guest *dbg);
+int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
+
+int kvm_arch_init(void *opaque);
+void kvm_arch_exit(void);
+
+int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu);
+void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu);
+
+void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
+void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
+void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
+struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id);
+int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
+void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
+
+int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
+void kvm_arch_hardware_enable(void *garbage);
+void kvm_arch_hardware_disable(void *garbage);
+int kvm_arch_hardware_setup(void);
+void kvm_arch_hardware_unsetup(void);
+void kvm_arch_check_processor_compat(void *rtn);
+int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
+
+void kvm_free_physmem(struct kvm *kvm);
+
+struct  kvm *kvm_arch_create_vm(void);
+void kvm_arch_destroy_vm(struct kvm *kvm);
+
+int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
+int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
+
+static inline void kvm_guest_enter(void)
+{
+	account_system_vtime(current);
+	current->flags |= PF_VCPU;
+}
+
+static inline void kvm_guest_exit(void)
+{
+	account_system_vtime(current);
+	current->flags &= ~PF_VCPU;
+}
+
+static inline int memslot_id(struct kvm *kvm, struct kvm_memory_slot *slot)
+{
+	return slot - kvm->memslots;
+}
+
+static inline gpa_t gfn_to_gpa(gfn_t gfn)
+{
+	return (gpa_t)gfn << PAGE_SHIFT;
+}
+
+enum kvm_stat_kind {
+	KVM_STAT_VM,
+	KVM_STAT_VCPU,
+};
+
+struct kvm_stats_debugfs_item {
+	const char *name;
+	int offset;
+	enum kvm_stat_kind kind;
+	struct dentry *dentry;
+};
+extern struct kvm_stats_debugfs_item debugfs_entries[];
+
+#endif
diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h
new file mode 100644
index 000000000000..1c4e46decb22
--- /dev/null
+++ b/include/linux/kvm_types.h
@@ -0,0 +1,54 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ */
+
+#ifndef __KVM_TYPES_H__
+#define __KVM_TYPES_H__
+
+#include <asm/types.h>
+
+/*
+ * Address types:
+ *
+ *  gva - guest virtual address
+ *  gpa - guest physical address
+ *  gfn - guest frame number
+ *  hva - host virtual address
+ *  hpa - host physical address
+ *  hfn - host frame number
+ */
+
+typedef unsigned long  gva_t;
+typedef u64            gpa_t;
+typedef unsigned long  gfn_t;
+
+typedef unsigned long  hva_t;
+typedef u64            hpa_t;
+typedef unsigned long  hfn_t;
+
+struct kvm_pio_request {
+	unsigned long count;
+	int cur_count;
+	struct page *guest_pages[2];
+	unsigned guest_page_offset;
+	int in;
+	int port;
+	int size;
+	int string;
+	int down;
+	int rep;
+};
+
+#endif /* __KVM_TYPES_H__ */
-- 
cgit v1.2.3


From 5736199afba8a8bb60a1ea282ab72857d6b16400 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Mon, 17 Dec 2007 14:21:40 +0800
Subject: KVM: Move kvm_vcpu_kick() to x86.c

Moving kvm_vcpu_kick() to x86.c. Since it should be
common for all archs, put its declarations in <linux/kvm_host.h>

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/irq.c       | 20 --------------------
 arch/x86/kvm/irq.h       |  2 --
 arch/x86/kvm/x86.c       | 20 ++++++++++++++++++++
 include/linux/kvm_host.h |  1 +
 4 files changed, 21 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index 07a09aad4fd6..e5714759e97f 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -63,26 +63,6 @@ int kvm_cpu_get_interrupt(struct kvm_vcpu *v)
 }
 EXPORT_SYMBOL_GPL(kvm_cpu_get_interrupt);
 
-static void vcpu_kick_intr(void *info)
-{
-#ifdef DEBUG
-	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
-	printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
-#endif
-}
-
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
-{
-	int ipi_pcpu = vcpu->cpu;
-
-	if (waitqueue_active(&vcpu->wq)) {
-		wake_up_interruptible(&vcpu->wq);
-		++vcpu->stat.halt_wakeup;
-	}
-	if (vcpu->guest_mode)
-		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
-}
-
 void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu)
 {
 	kvm_inject_apic_timer_irqs(vcpu);
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index c376b5b41bba..53c7f48254be 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -85,6 +85,4 @@ void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
 void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
 void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
 
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
-
 #endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 5902c5cbc1bb..733bff65d9a9 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3144,3 +3144,23 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 	return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
 	       || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
 }
+
+static void vcpu_kick_intr(void *info)
+{
+#ifdef DEBUG
+	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
+	printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
+#endif
+}
+
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
+{
+	int ipi_pcpu = vcpu->cpu;
+
+	if (waitqueue_active(&vcpu->wq)) {
+		wake_up_interruptible(&vcpu->wq);
+		++vcpu->stat.halt_wakeup;
+	}
+	if (vcpu->guest_mode)
+		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
+}
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index a85d5b6943de..953b50aa0e61 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -250,6 +250,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm);
 
 int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
 
 static inline void kvm_guest_enter(void)
 {
-- 
cgit v1.2.3


From ec10f4750db5e810cafef114a592428f8242b640 Mon Sep 17 00:00:00 2001
From: Zhang Xiantao <xiantao.zhang@intel.com>
Date: Mon, 17 Dec 2007 20:27:27 +0800
Subject: KVM: Expose ioapic to ia64 save/restore APIs

IA64 also needs to see ioapic structure in irqchip.

Signed-off-by: xiantao.zhang@intel.com <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/linux/kvm.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index f0bebd61ff77..de9f28d96ced 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -53,6 +53,8 @@ struct kvm_irqchip {
 		char dummy[512];  /* reserving space */
 #ifdef CONFIG_X86
 		struct kvm_pic_state pic;
+#endif
+#if defined(CONFIG_X86) || defined(CONFIG_IA64)
 		struct kvm_ioapic_state ioapic;
 #endif
 	} chip;
-- 
cgit v1.2.3


From dfc5aa00cbe888d3a9ea97775bbac74cb1c1a1d8 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Tue, 18 Dec 2007 19:47:18 +0200
Subject: KVM: MMU: Add cache miss statistic

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/mmu.c         | 1 +
 arch/x86/kvm/x86.c         | 1 +
 include/asm-x86/kvm_host.h | 1 +
 3 files changed, 3 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 9a57e1a01449..8f12ec52ad86 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -715,6 +715,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 			pgprintk("%s: found\n", __FUNCTION__);
 			return sp;
 		}
+	++vcpu->kvm->stat.mmu_cache_miss;
 	sp = kvm_mmu_alloc_page(vcpu, parent_pte);
 	if (!sp)
 		return sp;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 733bff65d9a9..fa9e42c03741 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -73,6 +73,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) },
 	{ "mmu_flooded", VM_STAT(mmu_flooded) },
 	{ "mmu_recycled", VM_STAT(mmu_recycled) },
+	{ "mmu_cache_miss", VM_STAT(mmu_cache_miss) },
 	{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) },
 	{ NULL }
 };
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 28940e1a9713..ced1bebabbc8 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -289,6 +289,7 @@ struct kvm_vm_stat {
 	u32 mmu_pde_zapped;
 	u32 mmu_flooded;
 	u32 mmu_recycled;
+	u32 mmu_cache_miss;
 	u32 remote_tlb_flush;
 };
 
-- 
cgit v1.2.3


From b209749f528488c4c0d20a42c0fbcbf49e6933b3 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Mon, 22 Oct 2007 16:50:39 +0200
Subject: KVM: local APIC TPR access reporting facility

Add a facility to report on accesses to the local apic tpr even if the
local apic is emulated in the kernel.  This is basically a hack that
allows userspace to patch Windows which tends to bang on the tpr a lot.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/lapic.c       | 21 +++++++++++++++++++++
 arch/x86/kvm/x86.c         | 25 +++++++++++++++++++++++++
 include/asm-x86/kvm_host.h |  1 +
 include/linux/kvm.h        | 17 +++++++++++++++++
 include/linux/kvm_host.h   |  2 +-
 5 files changed, 65 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 4076331b01ee..50c3f3a8dd3d 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -551,6 +551,23 @@ static u32 apic_get_tmcct(struct kvm_lapic *apic)
 	return tmcct;
 }
 
+static void __report_tpr_access(struct kvm_lapic *apic, bool write)
+{
+	struct kvm_vcpu *vcpu = apic->vcpu;
+	struct kvm_run *run = vcpu->run;
+
+	set_bit(KVM_REQ_REPORT_TPR_ACCESS, &vcpu->requests);
+	kvm_x86_ops->cache_regs(vcpu);
+	run->tpr_access.rip = vcpu->arch.rip;
+	run->tpr_access.is_write = write;
+}
+
+static inline void report_tpr_access(struct kvm_lapic *apic, bool write)
+{
+	if (apic->vcpu->arch.tpr_access_reporting)
+		__report_tpr_access(apic, write);
+}
+
 static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
 {
 	u32 val = 0;
@@ -568,6 +585,9 @@ static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
 		val = apic_get_tmcct(apic);
 		break;
 
+	case APIC_TASKPRI:
+		report_tpr_access(apic, false);
+		/* fall thru */
 	default:
 		apic_update_ppr(apic);
 		val = apic_get_reg(apic, offset);
@@ -677,6 +697,7 @@ static void apic_mmio_write(struct kvm_io_device *this,
 		break;
 
 	case APIC_TASKPRI:
+		report_tpr_access(apic, true);
 		apic_set_tpr(apic, val & 0xff);
 		break;
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 513258c797ca..c2b80884447e 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -684,6 +684,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_SET_TSS_ADDR:
 	case KVM_CAP_EXT_CPUID:
+	case KVM_CAP_VAPIC:
 		r = 1;
 		break;
 	default:
@@ -1055,6 +1056,15 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
+static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu,
+					   struct kvm_tpr_access_ctl *tac)
+{
+	if (tac->flags)
+		return -EINVAL;
+	vcpu->arch.tpr_access_reporting = !!tac->enabled;
+	return 0;
+}
+
 long kvm_arch_vcpu_ioctl(struct file *filp,
 			 unsigned int ioctl, unsigned long arg)
 {
@@ -1148,6 +1158,21 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 	case KVM_SET_MSRS:
 		r = msr_io(vcpu, argp, do_set_msr, 0);
 		break;
+	case KVM_TPR_ACCESS_REPORTING: {
+		struct kvm_tpr_access_ctl tac;
+
+		r = -EFAULT;
+		if (copy_from_user(&tac, argp, sizeof tac))
+			goto out;
+		r = vcpu_ioctl_tpr_access_reporting(vcpu, &tac);
+		if (r)
+			goto out;
+		r = -EFAULT;
+		if (copy_to_user(argp, &tac, sizeof tac))
+			goto out;
+		r = 0;
+		break;
+	};
 	default:
 		r = -EINVAL;
 	}
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index ced1bebabbc8..6e649af08d19 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -211,6 +211,7 @@ struct kvm_vcpu_arch {
 	int mp_state;
 	int sipi_vector;
 	u64 ia32_misc_enable_msr;
+	bool tpr_access_reporting;
 
 	struct kvm_mmu mmu;
 
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index de9f28d96ced..850f5ef76636 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -72,6 +72,7 @@ struct kvm_irqchip {
 #define KVM_EXIT_FAIL_ENTRY       9
 #define KVM_EXIT_INTR             10
 #define KVM_EXIT_SET_TPR          11
+#define KVM_EXIT_TPR_ACCESS       12
 
 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
 struct kvm_run {
@@ -130,6 +131,12 @@ struct kvm_run {
 			__u32 longmode;
 			__u32 pad;
 		} hypercall;
+		/* KVM_EXIT_TPR_ACCESS */
+		struct {
+			__u64 rip;
+			__u32 is_write;
+			__u32 pad;
+		} tpr_access;
 		/* Fix the size of the union. */
 		char padding[256];
 	};
@@ -202,6 +209,13 @@ struct kvm_signal_mask {
 	__u8  sigset[0];
 };
 
+/* for KVM_TPR_ACCESS_REPORTING */
+struct kvm_tpr_access_ctl {
+	__u32 enabled;
+	__u32 flags;
+	__u32 reserved[8];
+};
+
 #define KVMIO 0xAE
 
 /*
@@ -229,6 +243,7 @@ struct kvm_signal_mask {
 #define KVM_CAP_USER_MEMORY 3
 #define KVM_CAP_SET_TSS_ADDR 4
 #define KVM_CAP_EXT_CPUID 5
+#define KVM_CAP_VAPIC 6
 
 /*
  * ioctls for VM fds
@@ -274,5 +289,7 @@ struct kvm_signal_mask {
 #define KVM_SET_LAPIC             _IOW(KVMIO,  0x8f, struct kvm_lapic_state)
 #define KVM_SET_CPUID2            _IOW(KVMIO,  0x90, struct kvm_cpuid2)
 #define KVM_GET_CPUID2            _IOWR(KVMIO, 0x91, struct kvm_cpuid2)
+/* Available with KVM_CAP_VAPIC */
+#define KVM_TPR_ACCESS_REPORTING  _IOWR(KVMIO,  0x92, struct kvm_tpr_access_ctl)
 
 #endif
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 953b50aa0e61..9ff5904c5072 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -35,7 +35,7 @@
  * vcpu->requests bit members
  */
 #define KVM_REQ_TLB_FLUSH          0
-
+#define KVM_REQ_REPORT_TPR_ACCESS  2
 
 struct kvm_vcpu;
 extern struct kmem_cache *kvm_vcpu_cache;
-- 
cgit v1.2.3


From b93463aa59d67b21b4921e30bd5c5dcc7c35ffbd Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Thu, 25 Oct 2007 16:52:32 +0200
Subject: KVM: Accelerated apic support

This adds a mechanism for exposing the virtual apic tpr to the guest, and a
protocol for letting the guest update the tpr without causing a vmexit if
conditions allow (e.g. there is no interrupt pending with a higher priority
than the new tpr).

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/lapic.c     | 51 +++++++++++++++++++++++++++++++++++++++++++-
 arch/x86/kvm/lapic.h     |  6 ++++++
 arch/x86/kvm/x86.c       | 55 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/kvm.h      |  7 ++++++
 include/linux/kvm_para.h |  2 ++
 5 files changed, 120 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 50c3f3a8dd3d..e7513bb98af1 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -815,7 +815,8 @@ void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
 
 	if (!apic)
 		return;
-	apic_set_tpr(apic, ((cr8 & 0x0f) << 4));
+	apic_set_tpr(apic, ((cr8 & 0x0f) << 4)
+		     | (apic_get_reg(apic, APIC_TASKPRI) & 4));
 }
 
 u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu)
@@ -1104,3 +1105,51 @@ void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
 		hrtimer_start(timer, timer->expires, HRTIMER_MODE_ABS);
 }
 EXPORT_SYMBOL_GPL(kvm_migrate_apic_timer);
+
+void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu)
+{
+	u32 data;
+	void *vapic;
+
+	if (!irqchip_in_kernel(vcpu->kvm) || !vcpu->arch.apic->vapic_addr)
+		return;
+
+	vapic = kmap_atomic(vcpu->arch.apic->vapic_page, KM_USER0);
+	data = *(u32 *)(vapic + offset_in_page(vcpu->arch.apic->vapic_addr));
+	kunmap_atomic(vapic, KM_USER0);
+
+	apic_set_tpr(vcpu->arch.apic, data & 0xff);
+}
+
+void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu)
+{
+	u32 data, tpr;
+	int max_irr, max_isr;
+	struct kvm_lapic *apic;
+	void *vapic;
+
+	if (!irqchip_in_kernel(vcpu->kvm) || !vcpu->arch.apic->vapic_addr)
+		return;
+
+	apic = vcpu->arch.apic;
+	tpr = apic_get_reg(apic, APIC_TASKPRI) & 0xff;
+	max_irr = apic_find_highest_irr(apic);
+	if (max_irr < 0)
+		max_irr = 0;
+	max_isr = apic_find_highest_isr(apic);
+	if (max_isr < 0)
+		max_isr = 0;
+	data = (tpr & 0xff) | ((max_isr & 0xf0) << 8) | (max_irr << 24);
+
+	vapic = kmap_atomic(vcpu->arch.apic->vapic_page, KM_USER0);
+	*(u32 *)(vapic + offset_in_page(vcpu->arch.apic->vapic_addr)) = data;
+	kunmap_atomic(vapic, KM_USER0);
+}
+
+void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr)
+{
+	if (!irqchip_in_kernel(vcpu->kvm))
+		return;
+
+	vcpu->arch.apic->vapic_addr = vapic_addr;
+}
diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
index 447b654aefbb..676c396c9cee 100644
--- a/arch/x86/kvm/lapic.h
+++ b/arch/x86/kvm/lapic.h
@@ -18,6 +18,8 @@ struct kvm_lapic {
 	struct kvm_vcpu *vcpu;
 	struct page *regs_page;
 	void *regs;
+	gpa_t vapic_addr;
+	struct page *vapic_page;
 };
 int kvm_create_lapic(struct kvm_vcpu *vcpu);
 void kvm_free_lapic(struct kvm_vcpu *vcpu);
@@ -41,4 +43,8 @@ int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
 int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
 void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
 
+void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr);
+void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu);
+void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu);
+
 #endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c2b80884447e..e7eac27adb7f 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1173,6 +1173,19 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		r = 0;
 		break;
 	};
+	case KVM_SET_VAPIC_ADDR: {
+		struct kvm_vapic_addr va;
+
+		r = -EINVAL;
+		if (!irqchip_in_kernel(vcpu->kvm))
+			goto out;
+		r = -EFAULT;
+		if (copy_from_user(&va, argp, sizeof va))
+			goto out;
+		r = 0;
+		kvm_lapic_set_vapic_addr(vcpu, va.vapic_addr);
+		break;
+	}
 	default:
 		r = -EINVAL;
 	}
@@ -2214,6 +2227,9 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
 	}
 
 	switch (nr) {
+	case KVM_HC_VAPIC_POLL_IRQ:
+		ret = 0;
+		break;
 	default:
 		ret = -KVM_ENOSYS;
 		break;
@@ -2421,6 +2437,29 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu,
 					 vcpu->arch.irq_summary == 0);
 }
 
+static void vapic_enter(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+	struct page *page;
+
+	if (!apic || !apic->vapic_addr)
+		return;
+
+	page = gfn_to_page(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT);
+	vcpu->arch.apic->vapic_page = page;
+}
+
+static void vapic_exit(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (!apic || !apic->vapic_addr)
+		return;
+
+	kvm_release_page_dirty(apic->vapic_page);
+	mark_page_dirty(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT);
+}
+
 static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	int r;
@@ -2435,6 +2474,8 @@ static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
 	}
 
+	vapic_enter(vcpu);
+
 preempted:
 	if (vcpu->guest_debug.enabled)
 		kvm_x86_ops->guest_debug_pre(vcpu);
@@ -2444,6 +2485,14 @@ again:
 	if (unlikely(r))
 		goto out;
 
+	if (vcpu->requests)
+		if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
+				       &vcpu->requests)) {
+			kvm_run->exit_reason = KVM_EXIT_TPR_ACCESS;
+			r = 0;
+			goto out;
+		}
+
 	kvm_inject_pending_timer_irqs(vcpu);
 
 	preempt_disable();
@@ -2469,6 +2518,8 @@ again:
 	else
 		kvm_x86_ops->inject_pending_vectors(vcpu, kvm_run);
 
+	kvm_lapic_sync_to_vapic(vcpu);
+
 	vcpu->guest_mode = 1;
 	kvm_guest_enter();
 
@@ -2506,6 +2557,8 @@ again:
 	if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
 		vcpu->arch.exception.pending = false;
 
+	kvm_lapic_sync_from_vapic(vcpu);
+
 	r = kvm_x86_ops->handle_exit(kvm_run, vcpu);
 
 	if (r > 0) {
@@ -2527,6 +2580,8 @@ out:
 
 	post_kvm_run_save(vcpu, kvm_run);
 
+	vapic_exit(vcpu);
+
 	return r;
 }
 
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 850f5ef76636..b27a381db5e8 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -216,6 +216,11 @@ struct kvm_tpr_access_ctl {
 	__u32 reserved[8];
 };
 
+/* for KVM_SET_VAPIC_ADDR */
+struct kvm_vapic_addr {
+	__u64 vapic_addr;
+};
+
 #define KVMIO 0xAE
 
 /*
@@ -291,5 +296,7 @@ struct kvm_tpr_access_ctl {
 #define KVM_GET_CPUID2            _IOWR(KVMIO, 0x91, struct kvm_cpuid2)
 /* Available with KVM_CAP_VAPIC */
 #define KVM_TPR_ACCESS_REPORTING  _IOWR(KVMIO,  0x92, struct kvm_tpr_access_ctl)
+/* Available with KVM_CAP_VAPIC */
+#define KVM_SET_VAPIC_ADDR        _IOW(KVMIO,  0x93, struct kvm_vapic_addr)
 
 #endif
diff --git a/include/linux/kvm_para.h b/include/linux/kvm_para.h
index e4db25ffdb52..6af91a506fbf 100644
--- a/include/linux/kvm_para.h
+++ b/include/linux/kvm_para.h
@@ -12,6 +12,8 @@
 /* Return values for hypercalls */
 #define KVM_ENOSYS		1000
 
+#define KVM_HC_VAPIC_POLL_IRQ            1
+
 #ifdef __KERNEL__
 /*
  * hypercalls use architecture specific
-- 
cgit v1.2.3


From 774ead3ad9bcbc05ef6aaebb9bdf8b4c3126923b Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Wed, 26 Dec 2007 13:57:04 +0200
Subject: KVM: Disable vapic support on Intel machines with FlexPriority

FlexPriority accelerates the tpr without any patching.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/svm.c         | 6 ++++++
 arch/x86/kvm/vmx.c         | 3 ++-
 arch/x86/kvm/x86.c         | 4 +++-
 include/asm-x86/kvm_host.h | 1 +
 4 files changed, 12 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 3d4b71a94440..33f77b044c52 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1653,6 +1653,11 @@ static void svm_check_processor_compat(void *rtn)
 	*(int *)rtn = 0;
 }
 
+static bool svm_cpu_has_accelerated_tpr(void)
+{
+	return false;
+}
+
 static struct kvm_x86_ops svm_x86_ops = {
 	.cpu_has_kvm_support = has_svm,
 	.disabled_by_bios = is_disabled,
@@ -1661,6 +1666,7 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.check_processor_compatibility = svm_check_processor_compat,
 	.hardware_enable = svm_hardware_enable,
 	.hardware_disable = svm_hardware_disable,
+	.cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr,
 
 	.vcpu_create = svm_create_vcpu,
 	.vcpu_free = svm_free_vcpu,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index fc494aff5d8b..d1167fc303d6 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -192,7 +192,7 @@ static inline int cpu_has_secondary_exec_ctrls(void)
 		CPU_BASED_ACTIVATE_SECONDARY_CONTROLS);
 }
 
-static inline int cpu_has_vmx_virtualize_apic_accesses(void)
+static inline bool cpu_has_vmx_virtualize_apic_accesses(void)
 {
 	return (vmcs_config.cpu_based_2nd_exec_ctrl &
 		SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
@@ -2565,6 +2565,7 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.check_processor_compatibility = vmx_check_processor_compat,
 	.hardware_enable = hardware_enable,
 	.hardware_disable = hardware_disable,
+	.cpu_has_accelerated_tpr = cpu_has_vmx_virtualize_apic_accesses,
 
 	.vcpu_create = vmx_create_vcpu,
 	.vcpu_free = vmx_free_vcpu,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e7eac27adb7f..1f48ec871035 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -684,9 +684,11 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_SET_TSS_ADDR:
 	case KVM_CAP_EXT_CPUID:
-	case KVM_CAP_VAPIC:
 		r = 1;
 		break;
+	case KVM_CAP_VAPIC:
+		r = !kvm_x86_ops->cpu_has_accelerated_tpr();
+		break;
 	default:
 		r = 0;
 		break;
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 6e649af08d19..44b89259f6c4 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -329,6 +329,7 @@ struct kvm_x86_ops {
 	void (*check_processor_compatibility)(void *rtn);
 	int (*hardware_setup)(void);               /* __init */
 	void (*hardware_unsetup)(void);            /* __exit */
+	bool (*cpu_has_accelerated_tpr)(void);
 
 	/* Create, but do not attach this VCPU */
 	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
-- 
cgit v1.2.3


From 7ec54588210df29ea637e6054489bc942c0ef371 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Thu, 20 Dec 2007 19:18:23 -0500
Subject: KVM: Add kvm_read_guest_atomic()

In preparation for a mmu spinlock, add kvm_read_guest_atomic()
and use it in fetch() and prefetch_page().

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/paging_tmpl.h | 28 ++++++++++++++++------------
 include/linux/kvm_host.h   |  2 ++
 virt/kvm/kvm_main.c        | 20 ++++++++++++++++++++
 3 files changed, 38 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 7f83f5557d5e..136a65d72b0a 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -316,10 +316,12 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 					       metaphysical, access,
 					       shadow_ent, &new_page);
 		if (new_page && !metaphysical) {
+			int r;
 			pt_element_t curr_pte;
-			kvm_read_guest(vcpu->kvm, walker->pte_gpa[level - 2],
-				       &curr_pte, sizeof(curr_pte));
-			if (curr_pte != walker->ptes[level - 2])
+			r = kvm_read_guest_atomic(vcpu->kvm,
+						  walker->pte_gpa[level - 2],
+						  &curr_pte, sizeof(curr_pte));
+			if (r || curr_pte != walker->ptes[level - 2])
 				return NULL;
 		}
 		shadow_addr = __pa(shadow_page->spt);
@@ -429,9 +431,8 @@ static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr)
 static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 				 struct kvm_mmu_page *sp)
 {
-	int i, offset = 0;
-	pt_element_t *gpt;
-	struct page *page;
+	int i, offset = 0, r = 0;
+	pt_element_t pt;
 
 	if (sp->role.metaphysical
 	    || (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) {
@@ -441,15 +442,18 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 
 	if (PTTYPE == 32)
 		offset = sp->role.quadrant << PT64_LEVEL_BITS;
-	page = gfn_to_page(vcpu->kvm, sp->gfn);
-	gpt = kmap_atomic(page, KM_USER0);
-	for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
-		if (is_present_pte(gpt[offset + i]))
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+		gpa_t pte_gpa = gfn_to_gpa(sp->gfn);
+		pte_gpa += (i+offset) * sizeof(pt_element_t);
+
+		r = kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &pt,
+					  sizeof(pt_element_t));
+		if (r || is_present_pte(pt))
 			sp->spt[i] = shadow_trap_nonpresent_pte;
 		else
 			sp->spt[i] = shadow_notrap_nonpresent_pte;
-	kunmap_atomic(gpt, KM_USER0);
-	kvm_release_page_clean(page);
+	}
 }
 
 #undef pt_element_t
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 9ff5904c5072..a020fb280540 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -167,6 +167,8 @@ void kvm_release_page_clean(struct page *page);
 void kvm_release_page_dirty(struct page *page);
 int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
 			int len);
+int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
+			  unsigned long len);
 int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
 int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
 			 int offset, int len);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 678e80561b74..8d0b7c16c2f7 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -541,6 +541,26 @@ int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len)
 }
 EXPORT_SYMBOL_GPL(kvm_read_guest);
 
+int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
+			  unsigned long len)
+{
+	int r;
+	unsigned long addr;
+	gfn_t gfn = gpa >> PAGE_SHIFT;
+	int offset = offset_in_page(gpa);
+
+	addr = gfn_to_hva(kvm, gfn);
+	if (kvm_is_error_hva(addr))
+		return -EFAULT;
+	pagefault_disable();
+	r = __copy_from_user_inatomic(data, (void __user *)addr + offset, len);
+	pagefault_enable();
+	if (r)
+		return -EFAULT;
+	return 0;
+}
+EXPORT_SYMBOL(kvm_read_guest_atomic);
+
 int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
 			 int offset, int len)
 {
-- 
cgit v1.2.3


From d7824fff896a1698a07a8046dc362f4500c302f7 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Sun, 30 Dec 2007 12:29:05 +0200
Subject: KVM: MMU: Avoid calling gfn_to_page() in mmu_set_spte()

Since gfn_to_page() is a sleeping function, and we want to make the core mmu
spinlocked, we need to pass the page from the walker context (which can sleep)
to the shadow context (which cannot).

[marcelo: avoid recursive locking of mmap_sem]

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/mmu.c         | 55 +++++++++++++++++++++++++++++++++++++++++-----
 arch/x86/kvm/paging_tmpl.h | 23 ++++++++++++++-----
 include/asm-x86/kvm_host.h |  5 +++++
 3 files changed, 73 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 3b91227969a5..c0b757be7b99 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -890,11 +890,10 @@ struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
 static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 			 unsigned pt_access, unsigned pte_access,
 			 int user_fault, int write_fault, int dirty,
-			 int *ptwrite, gfn_t gfn)
+			 int *ptwrite, gfn_t gfn, struct page *page)
 {
 	u64 spte;
 	int was_rmapped = is_rmap_pte(*shadow_pte);
-	struct page *page;
 
 	pgprintk("%s: spte %llx access %x write_fault %d"
 		 " user_fault %d gfn %lx\n",
@@ -912,8 +911,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 	if (!(pte_access & ACC_EXEC_MASK))
 		spte |= PT64_NX_MASK;
 
-	page = gfn_to_page(vcpu->kvm, gfn);
-
 	spte |= PT_PRESENT_MASK;
 	if (pte_access & ACC_USER_MASK)
 		spte |= PT_USER_MASK;
@@ -979,6 +976,11 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
 	int level = PT32E_ROOT_LEVEL;
 	hpa_t table_addr = vcpu->arch.mmu.root_hpa;
 	int pt_write = 0;
+	struct page *page;
+
+	down_read(&current->mm->mmap_sem);
+	page = gfn_to_page(vcpu->kvm, gfn);
+	up_read(&current->mm->mmap_sem);
 
 	for (; ; level--) {
 		u32 index = PT64_INDEX(v, level);
@@ -989,7 +991,7 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
 
 		if (level == 1) {
 			mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL,
-				     0, write, 1, &pt_write, gfn);
+				     0, write, 1, &pt_write, gfn, page);
 			return pt_write || is_io_pte(table[index]);
 		}
 
@@ -1005,6 +1007,7 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
 						     NULL);
 			if (!new_table) {
 				pgprintk("nonpaging_map: ENOMEM\n");
+				kvm_release_page_clean(page);
 				return -ENOMEM;
 			}
 
@@ -1347,6 +1350,43 @@ static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
 	return !!(spte && (*spte & PT_ACCESSED_MASK));
 }
 
+static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
+					  const u8 *new, int bytes)
+{
+	gfn_t gfn;
+	int r;
+	u64 gpte = 0;
+
+	if (bytes != 4 && bytes != 8)
+		return;
+
+	/*
+	 * Assume that the pte write on a page table of the same type
+	 * as the current vcpu paging mode.  This is nearly always true
+	 * (might be false while changing modes).  Note it is verified later
+	 * by update_pte().
+	 */
+	if (is_pae(vcpu)) {
+		/* Handle a 32-bit guest writing two halves of a 64-bit gpte */
+		if ((bytes == 4) && (gpa % 4 == 0)) {
+			r = kvm_read_guest(vcpu->kvm, gpa & ~(u64)7, &gpte, 8);
+			if (r)
+				return;
+			memcpy((void *)&gpte + (gpa % 8), new, 4);
+		} else if ((bytes == 8) && (gpa % 8 == 0)) {
+			memcpy((void *)&gpte, new, 8);
+		}
+	} else {
+		if ((bytes == 4) && (gpa % 4 == 0))
+			memcpy((void *)&gpte, new, 4);
+	}
+	if (!is_present_pte(gpte))
+		return;
+	gfn = (gpte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
+	vcpu->arch.update_pte.gfn = gfn;
+	vcpu->arch.update_pte.page = gfn_to_page(vcpu->kvm, gfn);
+}
+
 void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 		       const u8 *new, int bytes)
 {
@@ -1367,6 +1407,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 	int npte;
 
 	pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
+	mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes);
 	mutex_lock(&vcpu->kvm->lock);
 	++vcpu->kvm->stat.mmu_pte_write;
 	kvm_mmu_audit(vcpu, "pre pte write");
@@ -1437,6 +1478,10 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 	}
 	kvm_mmu_audit(vcpu, "post pte write");
 	mutex_unlock(&vcpu->kvm->lock);
+	if (vcpu->arch.update_pte.page) {
+		kvm_release_page_clean(vcpu->arch.update_pte.page);
+		vcpu->arch.update_pte.page = NULL;
+	}
 }
 
 int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 136a65d72b0a..3d7846ba26e1 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -245,6 +245,7 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 {
 	pt_element_t gpte;
 	unsigned pte_access;
+	struct page *npage;
 
 	gpte = *(const pt_element_t *)pte;
 	if (~gpte & (PT_PRESENT_MASK | PT_ACCESSED_MASK)) {
@@ -256,8 +257,14 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 		return;
 	pgprintk("%s: gpte %llx spte %p\n", __FUNCTION__, (u64)gpte, spte);
 	pte_access = page->role.access & FNAME(gpte_access)(vcpu, gpte);
+	if (gpte_to_gfn(gpte) != vcpu->arch.update_pte.gfn)
+		return;
+	npage = vcpu->arch.update_pte.page;
+	if (!npage)
+		return;
+	get_page(npage);
 	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
-		     gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte));
+		     gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte), npage);
 }
 
 /*
@@ -265,7 +272,8 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
  */
 static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 			 struct guest_walker *walker,
-			 int user_fault, int write_fault, int *ptwrite)
+			 int user_fault, int write_fault, int *ptwrite,
+			 struct page *page)
 {
 	hpa_t shadow_addr;
 	int level;
@@ -321,8 +329,10 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 			r = kvm_read_guest_atomic(vcpu->kvm,
 						  walker->pte_gpa[level - 2],
 						  &curr_pte, sizeof(curr_pte));
-			if (r || curr_pte != walker->ptes[level - 2])
+			if (r || curr_pte != walker->ptes[level - 2]) {
+				kvm_release_page_clean(page);
 				return NULL;
+			}
 		}
 		shadow_addr = __pa(shadow_page->spt);
 		shadow_pte = shadow_addr | PT_PRESENT_MASK | PT_ACCESSED_MASK
@@ -333,7 +343,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 	mmu_set_spte(vcpu, shadow_ent, access, walker->pte_access & access,
 		     user_fault, write_fault,
 		     walker->ptes[walker->level-1] & PT_DIRTY_MASK,
-		     ptwrite, walker->gfn);
+		     ptwrite, walker->gfn, page);
 
 	return shadow_ent;
 }
@@ -362,6 +372,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 	u64 *shadow_pte;
 	int write_pt = 0;
 	int r;
+	struct page *page;
 
 	pgprintk("%s: addr %lx err %x\n", __FUNCTION__, addr, error_code);
 	kvm_mmu_audit(vcpu, "pre page fault");
@@ -388,9 +399,11 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 		return 0;
 	}
 
+	page = gfn_to_page(vcpu->kvm, walker.gfn);
+
 	mutex_lock(&vcpu->kvm->lock);
 	shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
-				  &write_pt);
+				  &write_pt, page);
 	pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
 		 shadow_pte, *shadow_pte, write_pt);
 
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 44b89259f6c4..20597bc16744 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -224,6 +224,11 @@ struct kvm_vcpu_arch {
 	int   last_pt_write_count;
 	u64  *last_pte_updated;
 
+	struct {
+		gfn_t gfn;          /* presumed gfn during guest pte update */
+		struct page *page;  /* page corresponding to that gfn */
+	} update_pte;
+
 	struct i387_fxsave_struct host_fx_image;
 	struct i387_fxsave_struct guest_fx_image;
 
-- 
cgit v1.2.3


From aaee2c94f7a1f7726e360a6cfb40173bd552bcff Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Thu, 20 Dec 2007 19:18:26 -0500
Subject: KVM: MMU: Switch to mmu spinlock

Convert the synchronization of the shadow handling to a separate mmu_lock
spinlock.

Also guard fetch() by mmap_sem in read-mode to protect against alias
and memslot changes.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/mmu.c         | 48 +++++++++++++++++++++++++---------------------
 arch/x86/kvm/paging_tmpl.h | 10 ++++++----
 arch/x86/kvm/vmx.c         |  2 --
 include/linux/kvm_host.h   |  3 ++-
 virt/kvm/kvm_main.c        |  3 +--
 5 files changed, 35 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index c0b757be7b99..834698d24595 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -971,16 +971,12 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
 {
 }
 
-static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
+static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write,
+			   gfn_t gfn, struct page *page)
 {
 	int level = PT32E_ROOT_LEVEL;
 	hpa_t table_addr = vcpu->arch.mmu.root_hpa;
 	int pt_write = 0;
-	struct page *page;
-
-	down_read(&current->mm->mmap_sem);
-	page = gfn_to_page(vcpu->kvm, gfn);
-	up_read(&current->mm->mmap_sem);
 
 	for (; ; level--) {
 		u32 index = PT64_INDEX(v, level);
@@ -1022,9 +1018,17 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
 {
 	int r;
 
-	mutex_lock(&vcpu->kvm->lock);
-	r = __nonpaging_map(vcpu, v, write, gfn);
-	mutex_unlock(&vcpu->kvm->lock);
+	struct page *page;
+
+	down_read(&current->mm->mmap_sem);
+	page = gfn_to_page(vcpu->kvm, gfn);
+
+	spin_lock(&vcpu->kvm->mmu_lock);
+	r = __nonpaging_map(vcpu, v, write, gfn, page);
+	spin_unlock(&vcpu->kvm->mmu_lock);
+
+	up_read(&current->mm->mmap_sem);
+
 	return r;
 }
 
@@ -1045,7 +1049,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
 
 	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
 		return;
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 #ifdef CONFIG_X86_64
 	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
 		hpa_t root = vcpu->arch.mmu.root_hpa;
@@ -1053,7 +1057,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
 		sp = page_header(root);
 		--sp->root_count;
 		vcpu->arch.mmu.root_hpa = INVALID_PAGE;
-		mutex_unlock(&vcpu->kvm->lock);
+		spin_unlock(&vcpu->kvm->mmu_lock);
 		return;
 	}
 #endif
@@ -1067,7 +1071,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
 		}
 		vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
 	}
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
 	vcpu->arch.mmu.root_hpa = INVALID_PAGE;
 }
 
@@ -1270,9 +1274,9 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu)
 	r = mmu_topup_memory_caches(vcpu);
 	if (r)
 		goto out;
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 	mmu_alloc_roots(vcpu);
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
 	kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
 	kvm_mmu_flush_tlb(vcpu);
 out:
@@ -1408,7 +1412,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 
 	pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
 	mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes);
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 	++vcpu->kvm->stat.mmu_pte_write;
 	kvm_mmu_audit(vcpu, "pre pte write");
 	if (gfn == vcpu->arch.last_pt_write_gfn
@@ -1477,7 +1481,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 		}
 	}
 	kvm_mmu_audit(vcpu, "post pte write");
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
 	if (vcpu->arch.update_pte.page) {
 		kvm_release_page_clean(vcpu->arch.update_pte.page);
 		vcpu->arch.update_pte.page = NULL;
@@ -1493,15 +1497,15 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
 	gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
 	up_read(&current->mm->mmap_sem);
 
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 	r = kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
 	return r;
 }
 
 void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
 {
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 	while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
 		struct kvm_mmu_page *sp;
 
@@ -1510,7 +1514,7 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
 		kvm_mmu_zap_page(vcpu->kvm, sp);
 		++vcpu->kvm->stat.mmu_recycled;
 	}
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
 }
 
 int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
@@ -1642,10 +1646,10 @@ void kvm_mmu_zap_all(struct kvm *kvm)
 {
 	struct kvm_mmu_page *sp, *node;
 
-	mutex_lock(&kvm->lock);
+	spin_lock(&kvm->mmu_lock);
 	list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
 		kvm_mmu_zap_page(kvm, sp);
-	mutex_unlock(&kvm->lock);
+	spin_unlock(&kvm->mmu_lock);
 
 	kvm_flush_remote_tlbs(kvm);
 }
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 3d7846ba26e1..a35b83a4fef2 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -387,7 +387,6 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 	 */
 	r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault,
 			     fetch_fault);
-	up_read(&current->mm->mmap_sem);
 
 	/*
 	 * The page is not mapped by the guest.  Let the guest handle it.
@@ -396,12 +395,13 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 		pgprintk("%s: guest page fault\n", __FUNCTION__);
 		inject_page_fault(vcpu, addr, walker.error_code);
 		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
+		up_read(&current->mm->mmap_sem);
 		return 0;
 	}
 
 	page = gfn_to_page(vcpu->kvm, walker.gfn);
 
-	mutex_lock(&vcpu->kvm->lock);
+	spin_lock(&vcpu->kvm->mmu_lock);
 	shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
 				  &write_pt, page);
 	pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
@@ -414,13 +414,15 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 	 * mmio: emulate if accessible, otherwise its a guest fault.
 	 */
 	if (shadow_pte && is_io_pte(*shadow_pte)) {
-		mutex_unlock(&vcpu->kvm->lock);
+		spin_unlock(&vcpu->kvm->mmu_lock);
+		up_read(&current->mm->mmap_sem);
 		return 1;
 	}
 
 	++vcpu->stat.pf_fixed;
 	kvm_mmu_audit(vcpu, "post page fault (fixed)");
-	mutex_unlock(&vcpu->kvm->lock);
+	spin_unlock(&vcpu->kvm->mmu_lock);
+	up_read(&current->mm->mmap_sem);
 
 	return write_pt;
 }
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index c39493feba46..3d251f894a8d 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1477,7 +1477,6 @@ static int alloc_apic_access_page(struct kvm *kvm)
 	struct kvm_userspace_memory_region kvm_userspace_mem;
 	int r = 0;
 
-	mutex_lock(&kvm->lock);
 	down_write(&current->mm->mmap_sem);
 	if (kvm->arch.apic_access_page)
 		goto out;
@@ -1491,7 +1490,6 @@ static int alloc_apic_access_page(struct kvm *kvm)
 	kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
 out:
 	up_write(&current->mm->mmap_sem);
-	mutex_unlock(&kvm->lock);
 	return r;
 }
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index a020fb280540..2714068ee8bc 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -104,7 +104,8 @@ struct kvm_memory_slot {
 };
 
 struct kvm {
-	struct mutex lock; /* protects everything except vcpus */
+	struct mutex lock; /* protects the vcpus array and APIC accesses */
+	spinlock_t mmu_lock;
 	struct mm_struct *mm; /* userspace tied to this vm */
 	int nmemslots;
 	struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 8d0b7c16c2f7..3c4fe26096fc 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -165,6 +165,7 @@ static struct kvm *kvm_create_vm(void)
 
 	kvm->mm = current->mm;
 	atomic_inc(&kvm->mm->mm_count);
+	spin_lock_init(&kvm->mmu_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
 	mutex_init(&kvm->lock);
 	kvm_io_bus_init(&kvm->mmio_bus);
@@ -552,9 +553,7 @@ int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
 	addr = gfn_to_hva(kvm, gfn);
 	if (kvm_is_error_hva(addr))
 		return -EFAULT;
-	pagefault_disable();
 	r = __copy_from_user_inatomic(data, (void __user *)addr + offset, len);
-	pagefault_enable();
 	if (r)
 		return -EFAULT;
 	return 0;
-- 
cgit v1.2.3


From 571008daccc17c03ccec810922c2bcaed86b15c1 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng.yang@intel.com>
Date: Wed, 2 Jan 2008 14:49:22 +0800
Subject: KVM: x86 emulator: Only allow VMCALL/VMMCALL trapped by #UD

When executing a test program called "crashme", we found the KVM guest cannot
survive more than ten seconds, then encounterd kernel panic. The basic concept
of "crashme" is generating random assembly code and trying to execute it.

After some fixes on emulator insn validity judgment, we found it's hard to
get the current emulator handle the invalid instructions correctly, for the
#UD trap for hypercall patching caused troubles. The problem is, if the opcode
itself was OK, but combination of opcode and modrm_reg was invalid, and one
operand of the opcode was memory (SrcMem or DstMem), the emulator will fetch
the memory operand first rather than checking the validity, and may encounter
an error there. For example, ".byte 0xfe, 0x34, 0xcd" has this problem.

In the patch, we simply check that if the invalid opcode wasn't vmcall/vmmcall,
then return from emulate_instruction() and inject a #UD to guest. With the
patch, the guest had been running for more than 12 hours.

Signed-off-by: Feng (Eric) Liu <eric.e.liu@intel.com>
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/svm.c         |  2 +-
 arch/x86/kvm/vmx.c         |  2 +-
 arch/x86/kvm/x86.c         | 18 +++++++++++++++---
 include/asm-x86/kvm_host.h |  4 +++-
 4 files changed, 20 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 33f77b044c52..de755cb1431d 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -942,7 +942,7 @@ static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	int er;
 
-	er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0);
+	er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, EMULTYPE_TRAP_UD);
 	if (er != EMULATE_DONE)
 		kvm_queue_exception(&svm->vcpu, UD_VECTOR);
 	return 1;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 3d251f894a8d..ad36447e696e 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1863,7 +1863,7 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	}
 
 	if (is_invalid_opcode(intr_info)) {
-		er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
+		er = emulate_instruction(vcpu, kvm_run, 0, 0, EMULTYPE_TRAP_UD);
 		if (er != EMULATE_DONE)
 			kvm_queue_exception(vcpu, UD_VECTOR);
 		return 1;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e3b3141db13c..8a90403272e2 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1840,9 +1840,10 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 			struct kvm_run *run,
 			unsigned long cr2,
 			u16 error_code,
-			int no_decode)
+			int emulation_type)
 {
 	int r;
+	struct decode_cache *c;
 
 	vcpu->arch.mmio_fault_cr2 = cr2;
 	kvm_x86_ops->cache_regs(vcpu);
@@ -1850,7 +1851,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 	vcpu->mmio_is_write = 0;
 	vcpu->arch.pio.string = 0;
 
-	if (!no_decode) {
+	if (!(emulation_type & EMULTYPE_NO_DECODE)) {
 		int cs_db, cs_l;
 		kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
 
@@ -1884,6 +1885,16 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 					get_segment_base(vcpu, VCPU_SREG_FS);
 
 		r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
+
+		/* Reject the instructions other than VMCALL/VMMCALL when
+		 * try to emulate invalid opcode */
+		c = &vcpu->arch.emulate_ctxt.decode;
+		if ((emulation_type & EMULTYPE_TRAP_UD) &&
+		    (!(c->twobyte && c->b == 0x01 &&
+		      (c->modrm_reg == 0 || c->modrm_reg == 3) &&
+		       c->modrm_mod == 3 && c->modrm_rm == 1)))
+			return EMULATE_FAIL;
+
 		++vcpu->stat.insn_emulation;
 		if (r)  {
 			++vcpu->stat.insn_emulation_fail;
@@ -2640,7 +2651,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		vcpu->mmio_read_completed = 1;
 		vcpu->mmio_needed = 0;
 		r = emulate_instruction(vcpu, kvm_run,
-					vcpu->arch.mmio_fault_cr2, 0, 1);
+					vcpu->arch.mmio_fault_cr2, 0,
+					EMULTYPE_NO_DECODE);
 		if (r == EMULATE_DO_MMIO) {
 			/*
 			 * Read-modify-write.  Back to userspace.
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 20597bc16744..4702b04b979a 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -415,8 +415,10 @@ enum emulation_result {
 	EMULATE_FAIL,         /* can't emulate this instruction */
 };
 
+#define EMULTYPE_NO_DECODE	    (1 << 0)
+#define EMULTYPE_TRAP_UD	    (1 << 1)
 int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run,
-			unsigned long cr2, u16 error_code, int no_decode);
+			unsigned long cr2, u16 error_code, int emulation_type);
 void kvm_report_emulation_failure(struct kvm_vcpu *cvpu, const char *context);
 void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
 void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
-- 
cgit v1.2.3


From 6f723c7911e7827091586ae63f4040874eeb75e5 Mon Sep 17 00:00:00 2001
From: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Date: Tue, 8 Jan 2008 08:04:50 +0100
Subject: KVM: Portability: Move kvm_fpu to asm-x86/kvm.h

This patch moves kvm_fpu asm-x86/kvm.h to allow every architecture to
define an own representation used for KVM_GET_FPU/KVM_SET_FPU.

Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Acked-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/asm-x86/kvm.h | 15 +++++++++++++++
 include/linux/kvm.h   | 17 -----------------
 2 files changed, 15 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h
index 17afa81d6d66..7a71120426a3 100644
--- a/include/asm-x86/kvm.h
+++ b/include/asm-x86/kvm.h
@@ -114,6 +114,21 @@ struct kvm_sregs {
 	__u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
 };
 
+/* for KVM_GET_FPU and KVM_SET_FPU */
+struct kvm_fpu {
+	__u8  fpr[8][16];
+	__u16 fcw;
+	__u16 fsw;
+	__u8  ftwx;  /* in fxsave format */
+	__u8  pad1;
+	__u16 last_opcode;
+	__u64 last_ip;
+	__u64 last_dp;
+	__u8  xmm[16][16];
+	__u32 mxcsr;
+	__u32 pad2;
+};
+
 struct kvm_msr_entry {
 	__u32 index;
 	__u32 reserved;
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index b27a381db5e8..4de4fd2d8607 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -142,23 +142,6 @@ struct kvm_run {
 	};
 };
 
-
-/* for KVM_GET_FPU and KVM_SET_FPU */
-struct kvm_fpu {
-	__u8  fpr[8][16];
-	__u16 fcw;
-	__u16 fsw;
-	__u8  ftwx;  /* in fxsave format */
-	__u8  pad1;
-	__u16 last_opcode;
-	__u64 last_ip;
-	__u64 last_dp;
-	__u8  xmm[16][16];
-	__u32 mxcsr;
-	__u32 pad2;
-};
-
-
 /* for KVM_TRANSLATE */
 struct kvm_translation {
 	/* in */
-- 
cgit v1.2.3


From a03d7f4b544f699bbdd3cf14692bd8f476cb9d24 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Tue, 15 Jan 2008 13:10:15 -0200
Subject: KVM: Put kvm_para.h include outside __KERNEL__

kvm_para.h potentially contains definitions that are to be used by userspace,
so it should not be included inside the __KERNEL__ block. To protect its own
data structures, kvm_para.h already includes its own __KERNEL__ block.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Acked-by: Amit Shah <amit.shah@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/linux/kvm_para.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/kvm_para.h b/include/linux/kvm_para.h
index 6af91a506fbf..5497aac0d2f8 100644
--- a/include/linux/kvm_para.h
+++ b/include/linux/kvm_para.h
@@ -14,12 +14,12 @@
 
 #define KVM_HC_VAPIC_POLL_IRQ            1
 
-#ifdef __KERNEL__
 /*
  * hypercalls use architecture specific
  */
 #include <asm/kvm_para.h>
 
+#ifdef __KERNEL__
 static inline int kvm_para_has_feature(unsigned int feature)
 {
 	if (kvm_arch_para_features() & (1UL << feature))
-- 
cgit v1.2.3


From 2f52d58c92d971bf421f461ad06eb93fb4f34981 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Wed, 16 Jan 2008 12:49:30 +0200
Subject: KVM: Move apic timer migration away from critical section

Migrating the apic timer in the critical section is not very nice, and is
absolutely horrible with the real-time port.  Move migration to the regular
vcpu execution path, triggered by a new bitflag.

Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/x86/kvm/irq.h       | 2 +-
 arch/x86/kvm/lapic.c     | 3 +--
 arch/x86/kvm/x86.c       | 5 ++++-
 include/linux/kvm_host.h | 6 ++++++
 4 files changed, 12 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index 53c7f48254be..fa5ed5d59b5d 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -83,6 +83,6 @@ void kvm_pic_reset(struct kvm_kpic_state *s);
 void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
 void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
 void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
-void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
+void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
 
 #endif
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index e7513bb98af1..2cbee9479ce4 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -1092,7 +1092,7 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
 	start_apic_timer(apic);
 }
 
-void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
+void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
 {
 	struct kvm_lapic *apic = vcpu->arch.apic;
 	struct hrtimer *timer;
@@ -1104,7 +1104,6 @@ void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
 	if (hrtimer_cancel(timer))
 		hrtimer_start(timer, timer->expires, HRTIMER_MODE_ABS);
 }
-EXPORT_SYMBOL_GPL(kvm_migrate_apic_timer);
 
 void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu)
 {
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index f0493e7dcf0c..8f94a0b89dff 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2518,13 +2518,16 @@ again:
 	if (unlikely(r))
 		goto out;
 
-	if (vcpu->requests)
+	if (vcpu->requests) {
+		if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests))
+			__kvm_migrate_apic_timer(vcpu);
 		if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
 				       &vcpu->requests)) {
 			kvm_run->exit_reason = KVM_EXIT_TPR_ACCESS;
 			r = 0;
 			goto out;
 		}
+	}
 
 	kvm_inject_pending_timer_irqs(vcpu);
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 2714068ee8bc..ea4764b0a2f4 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -35,6 +35,7 @@
  * vcpu->requests bit members
  */
 #define KVM_REQ_TLB_FLUSH          0
+#define KVM_REQ_MIGRATE_TIMER      1
 #define KVM_REQ_REPORT_TPR_ACCESS  2
 
 struct kvm_vcpu;
@@ -277,6 +278,11 @@ static inline gpa_t gfn_to_gpa(gfn_t gfn)
 	return (gpa_t)gfn << PAGE_SHIFT;
 }
 
+static inline void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
+{
+	set_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests);
+}
+
 enum kvm_stat_kind {
 	KVM_STAT_VM,
 	KVM_STAT_VCPU,
-- 
cgit v1.2.3


From bb52d82f45df3a2661d88befba7c79a7db8be496 Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Thu, 13 Dec 2007 16:14:27 -0800
Subject: [SCSI] tgt: use scsi_init_io instead of scsi_alloc_sgtable

If we export scsi_init_io()/scsi_release_buffers() instead of
scsi_{alloc,free}_sgtable() from scsi_lib than tgt code is much more
insulated from scsi_lib changes. As a bonus it will also gain bidi
capability when it comes.

[jejb: rebase on to sg_table and fix up rejections]

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_lib.c     | 20 +++++++++-----------
 drivers/scsi/scsi_tgt_lib.c | 28 +++++-----------------------
 include/scsi/scsi_cmnd.h    |  4 ++--
 3 files changed, 16 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 7c4c889c5221..69fd62183bab 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -746,7 +746,7 @@ static struct scatterlist *scsi_sg_alloc(unsigned int nents, gfp_t gfp_mask)
 	return mempool_alloc(sgp->pool, gfp_mask);
 }
 
-int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+static int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 {
 	int ret;
 
@@ -762,15 +762,11 @@ int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 	return ret;
 }
 
-EXPORT_SYMBOL(scsi_alloc_sgtable);
-
-void scsi_free_sgtable(struct scsi_cmnd *cmd)
+static void scsi_free_sgtable(struct scsi_cmnd *cmd)
 {
 	__sg_free_table(&cmd->sg_table, SCSI_MAX_SG_SEGMENTS, scsi_sg_free);
 }
 
-EXPORT_SYMBOL(scsi_free_sgtable);
-
 /*
  * Function:    scsi_release_buffers()
  *
@@ -788,7 +784,7 @@ EXPORT_SYMBOL(scsi_free_sgtable);
  *		the scatter-gather table, and potentially any bounce
  *		buffers.
  */
-static void scsi_release_buffers(struct scsi_cmnd *cmd)
+void scsi_release_buffers(struct scsi_cmnd *cmd)
 {
 	if (cmd->use_sg)
 		scsi_free_sgtable(cmd);
@@ -800,6 +796,7 @@ static void scsi_release_buffers(struct scsi_cmnd *cmd)
 	cmd->request_buffer = NULL;
 	cmd->request_bufflen = 0;
 }
+EXPORT_SYMBOL(scsi_release_buffers);
 
 /*
  * Function:    scsi_io_completion()
@@ -1001,7 +998,7 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
  * Returns:     0 on success
  *		BLKPREP_DEFER if the failure is retryable
  */
-static int scsi_init_io(struct scsi_cmnd *cmd)
+int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 {
 	struct request     *req = cmd->request;
 	int		   count;
@@ -1016,7 +1013,7 @@ static int scsi_init_io(struct scsi_cmnd *cmd)
 	/*
 	 * If sg table allocation fails, requeue request later.
 	 */
-	if (unlikely(scsi_alloc_sgtable(cmd, GFP_ATOMIC))) {
+	if (unlikely(scsi_alloc_sgtable(cmd, gfp_mask))) {
 		scsi_unprep_request(req);
 		return BLKPREP_DEFER;
 	}
@@ -1036,6 +1033,7 @@ static int scsi_init_io(struct scsi_cmnd *cmd)
 	cmd->use_sg = count;
 	return BLKPREP_OK;
 }
+EXPORT_SYMBOL(scsi_init_io);
 
 static struct scsi_cmnd *scsi_get_cmd_from_req(struct scsi_device *sdev,
 		struct request *req)
@@ -1081,7 +1079,7 @@ int scsi_setup_blk_pc_cmnd(struct scsi_device *sdev, struct request *req)
 
 		BUG_ON(!req->nr_phys_segments);
 
-		ret = scsi_init_io(cmd);
+		ret = scsi_init_io(cmd, GFP_ATOMIC);
 		if (unlikely(ret))
 			return ret;
 	} else {
@@ -1132,7 +1130,7 @@ int scsi_setup_fs_cmnd(struct scsi_device *sdev, struct request *req)
 	if (unlikely(!cmd))
 		return BLKPREP_DEFER;
 
-	return scsi_init_io(cmd);
+	return scsi_init_io(cmd, GFP_ATOMIC);
 }
 EXPORT_SYMBOL(scsi_setup_fs_cmnd);
 
diff --git a/drivers/scsi/scsi_tgt_lib.c b/drivers/scsi/scsi_tgt_lib.c
index 01e03f3f6ffa..91630baea532 100644
--- a/drivers/scsi/scsi_tgt_lib.c
+++ b/drivers/scsi/scsi_tgt_lib.c
@@ -331,8 +331,7 @@ static void scsi_tgt_cmd_done(struct scsi_cmnd *cmd)
 
 	scsi_tgt_uspace_send_status(cmd, tcmd->itn_id, tcmd->tag);
 
-	if (scsi_sglist(cmd))
-		scsi_free_sgtable(cmd);
+	scsi_release_buffers(cmd);
 
 	queue_work(scsi_tgtd, &tcmd->work);
 }
@@ -353,25 +352,6 @@ static int scsi_tgt_transfer_response(struct scsi_cmnd *cmd)
 	return 0;
 }
 
-static int scsi_tgt_init_cmd(struct scsi_cmnd *cmd, gfp_t gfp_mask)
-{
-	struct request *rq = cmd->request;
-	int count;
-
-	cmd->use_sg = rq->nr_phys_segments;
-	if (scsi_alloc_sgtable(cmd, gfp_mask))
-		return -ENOMEM;
-
-	cmd->request_bufflen = rq->data_len;
-
-	dprintk("cmd %p cnt %d %lu\n", cmd, scsi_sg_count(cmd),
-		rq_data_dir(rq));
-	count = blk_rq_map_sg(rq->q, rq, scsi_sglist(cmd));
-	BUG_ON(count > cmd->use_sg);
-	cmd->use_sg = count;
-	return 0;
-}
-
 /* TODO: test this crap and replace bio_map_user with new interface maybe */
 static int scsi_map_user_pages(struct scsi_tgt_cmd *tcmd, struct scsi_cmnd *cmd,
 			       unsigned long uaddr, unsigned int len, int rw)
@@ -397,9 +377,11 @@ static int scsi_map_user_pages(struct scsi_tgt_cmd *tcmd, struct scsi_cmnd *cmd,
 	}
 
 	tcmd->bio = rq->bio;
-	err = scsi_tgt_init_cmd(cmd, GFP_KERNEL);
-	if (err)
+	err = scsi_init_io(cmd, GFP_KERNEL);
+	if (err) {
+		scsi_release_buffers(cmd);
 		goto unmap_rq;
+	}
 
 	return 0;
 
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index a457fca66f61..09d44f91dbdb 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -127,8 +127,8 @@ extern void *scsi_kmap_atomic_sg(struct scatterlist *sg, int sg_count,
 				 size_t *offset, size_t *len);
 extern void scsi_kunmap_atomic_sg(void *virt);
 
-extern int scsi_alloc_sgtable(struct scsi_cmnd *, gfp_t);
-extern void scsi_free_sgtable(struct scsi_cmnd *);
+extern int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask);
+extern void scsi_release_buffers(struct scsi_cmnd *cmd);
 
 extern int scsi_dma_map(struct scsi_cmnd *cmd);
 extern void scsi_dma_unmap(struct scsi_cmnd *cmd);
-- 
cgit v1.2.3


From 30b0c37b27485a9cb897bfe3824f6f517b8c80d6 Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Thu, 13 Dec 2007 13:47:40 +0200
Subject: [SCSI] implement scsi_data_buffer

In preparation for bidi we abstract all IO members of scsi_cmnd,
that will need to duplicate, into a substructure.

- Group all IO members of scsi_cmnd into a scsi_data_buffer
  structure.
- Adjust accessors to new members.
- scsi_{alloc,free}_sgtable receive a scsi_data_buffer instead of
  scsi_cmnd. And work on it.
- Adjust scsi_init_io() and  scsi_release_buffers() for above
  change.
- Fix other parts of scsi_lib/scsi.c to members migration. Use
  accessors where appropriate.

- fix Documentation about scsi_cmnd in scsi_host.h

- scsi_error.c
  * Changed needed members of struct scsi_eh_save.
  * Careful considerations in scsi_eh_prep/restore_cmnd.

- sd.c and sr.c
  * sd and sr would adjust IO size to align on device's block
    size so code needs to change once we move to scsi_data_buff
    implementation.
  * Convert code to use scsi_for_each_sg
  * Use data accessors where appropriate.

- tgt: convert libsrp to use scsi_data_buffer

- isd200: This driver still bangs on scsi_cmnd IO members,
  so need changing

[jejb: rebased on top of sg_table patches fixed up conflicts
and used the synergy to eliminate use_sg and sg_count]

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libsrp.c        |  4 +--
 drivers/scsi/scsi.c          |  2 +-
 drivers/scsi/scsi_error.c    | 30 ++++++++-------------
 drivers/scsi/scsi_lib.c      | 63 ++++++++++++++++++--------------------------
 drivers/scsi/sd.c            |  4 +--
 drivers/scsi/sr.c            | 25 +++++++++---------
 drivers/usb/storage/isd200.c |  8 +++---
 include/scsi/scsi_cmnd.h     | 36 +++++++++++++++----------
 include/scsi/scsi_eh.h       |  8 +++---
 include/scsi/scsi_host.h     |  4 +--
 10 files changed, 85 insertions(+), 99 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libsrp.c b/drivers/scsi/libsrp.c
index 5cff0204227d..6d6a76e65a6c 100644
--- a/drivers/scsi/libsrp.c
+++ b/drivers/scsi/libsrp.c
@@ -426,8 +426,8 @@ int srp_cmd_queue(struct Scsi_Host *shost, struct srp_cmd *cmd, void *info,
 
 	sc->SCp.ptr = info;
 	memcpy(sc->cmnd, cmd->cdb, MAX_COMMAND_SIZE);
-	sc->request_bufflen = len;
-	sc->request_buffer = (void *) (unsigned long) addr;
+	sc->sdb.length = len;
+	sc->sdb.table.sgl = (void *) (unsigned long) addr;
 	sc->tag = tag;
 	err = scsi_tgt_queue_command(sc, itn_id, (struct scsi_lun *)&cmd->lun,
 				     cmd->tag);
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 1a9fba6a9f92..b35d19472caa 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -757,7 +757,7 @@ void scsi_finish_command(struct scsi_cmnd *cmd)
 				"Notifying upper driver of completion "
 				"(result %x)\n", cmd->result));
 
-	good_bytes = cmd->request_bufflen;
+	good_bytes = scsi_bufflen(cmd);
         if (cmd->request->cmd_type != REQ_TYPE_BLOCK_PC) {
 		drv = scsi_cmd_to_driver(cmd);
 		if (drv->done)
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 547e85aa414f..4f4edc388c07 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -617,29 +617,25 @@ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses,
 	ses->cmd_len = scmd->cmd_len;
 	memcpy(ses->cmnd, scmd->cmnd, sizeof(scmd->cmnd));
 	ses->data_direction = scmd->sc_data_direction;
-	ses->bufflen = scmd->request_bufflen;
-	ses->buffer = scmd->request_buffer;
-	ses->use_sg = scmd->use_sg;
-	ses->resid = scmd->resid;
+	ses->sdb = scmd->sdb;
 	ses->result = scmd->result;
 
+	memset(&scmd->sdb, 0, sizeof(scmd->sdb));
+
 	if (sense_bytes) {
-		scmd->request_bufflen = min_t(unsigned,
-		                       SCSI_SENSE_BUFFERSIZE, sense_bytes);
+		scmd->sdb.length = min_t(unsigned, SCSI_SENSE_BUFFERSIZE,
+					 sense_bytes);
 		sg_init_one(&ses->sense_sgl, scmd->sense_buffer,
-		                                       scmd->request_bufflen);
-		scmd->request_buffer = &ses->sense_sgl;
+			    scmd->sdb.length);
+		scmd->sdb.table.sgl = &ses->sense_sgl;
 		scmd->sc_data_direction = DMA_FROM_DEVICE;
-		scmd->use_sg = 1;
+		scmd->sdb.table.nents = 1;
 		memset(scmd->cmnd, 0, sizeof(scmd->cmnd));
 		scmd->cmnd[0] = REQUEST_SENSE;
-		scmd->cmnd[4] = scmd->request_bufflen;
+		scmd->cmnd[4] = scmd->sdb.length;
 		scmd->cmd_len = COMMAND_SIZE(scmd->cmnd[0]);
 	} else {
-		scmd->request_buffer = NULL;
-		scmd->request_bufflen = 0;
 		scmd->sc_data_direction = DMA_NONE;
-		scmd->use_sg = 0;
 		if (cmnd) {
 			memset(scmd->cmnd, 0, sizeof(scmd->cmnd));
 			memcpy(scmd->cmnd, cmnd, cmnd_size);
@@ -676,10 +672,7 @@ void scsi_eh_restore_cmnd(struct scsi_cmnd* scmd, struct scsi_eh_save *ses)
 	scmd->cmd_len = ses->cmd_len;
 	memcpy(scmd->cmnd, ses->cmnd, sizeof(scmd->cmnd));
 	scmd->sc_data_direction = ses->data_direction;
-	scmd->request_bufflen = ses->bufflen;
-	scmd->request_buffer = ses->buffer;
-	scmd->use_sg = ses->use_sg;
-	scmd->resid = ses->resid;
+	scmd->sdb = ses->sdb;
 	scmd->result = ses->result;
 }
 EXPORT_SYMBOL(scsi_eh_restore_cmnd);
@@ -1700,8 +1693,7 @@ scsi_reset_provider(struct scsi_device *dev, int flag)
 	memset(&scmd->cmnd, '\0', sizeof(scmd->cmnd));
     
 	scmd->scsi_done		= scsi_reset_provider_done_command;
-	scmd->request_buffer		= NULL;
-	scmd->request_bufflen		= 0;
+	memset(&scmd->sdb, 0, sizeof(scmd->sdb));
 
 	scmd->cmd_len			= 0;
 
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 69fd62183bab..d5e77e9b3a9c 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -440,7 +440,7 @@ EXPORT_SYMBOL_GPL(scsi_execute_async);
 static void scsi_init_cmd_errh(struct scsi_cmnd *cmd)
 {
 	cmd->serial_number = 0;
-	cmd->resid = 0;
+	scsi_set_resid(cmd, 0);
 	memset(cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE);
 	if (cmd->cmd_len == 0)
 		cmd->cmd_len = COMMAND_SIZE(cmd->cmnd[0]);
@@ -746,25 +746,25 @@ static struct scatterlist *scsi_sg_alloc(unsigned int nents, gfp_t gfp_mask)
 	return mempool_alloc(sgp->pool, gfp_mask);
 }
 
-static int scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+static int scsi_alloc_sgtable(struct scsi_data_buffer *sdb, int nents,
+			      gfp_t gfp_mask)
 {
 	int ret;
 
-	BUG_ON(!cmd->use_sg);
+	BUG_ON(!nents);
 
-	ret = __sg_alloc_table(&cmd->sg_table, cmd->use_sg,
-			       SCSI_MAX_SG_SEGMENTS, gfp_mask, scsi_sg_alloc);
+	ret = __sg_alloc_table(&sdb->table, nents, SCSI_MAX_SG_SEGMENTS,
+			       gfp_mask, scsi_sg_alloc);
 	if (unlikely(ret))
-		__sg_free_table(&cmd->sg_table, SCSI_MAX_SG_SEGMENTS,
+		__sg_free_table(&sdb->table, SCSI_MAX_SG_SEGMENTS,
 				scsi_sg_free);
 
-	cmd->request_buffer = cmd->sg_table.sgl;
 	return ret;
 }
 
-static void scsi_free_sgtable(struct scsi_cmnd *cmd)
+static void scsi_free_sgtable(struct scsi_data_buffer *sdb)
 {
-	__sg_free_table(&cmd->sg_table, SCSI_MAX_SG_SEGMENTS, scsi_sg_free);
+	__sg_free_table(&sdb->table, SCSI_MAX_SG_SEGMENTS, scsi_sg_free);
 }
 
 /*
@@ -786,15 +786,10 @@ static void scsi_free_sgtable(struct scsi_cmnd *cmd)
  */
 void scsi_release_buffers(struct scsi_cmnd *cmd)
 {
-	if (cmd->use_sg)
-		scsi_free_sgtable(cmd);
+	if (cmd->sdb.table.nents)
+		scsi_free_sgtable(&cmd->sdb);
 
-	/*
-	 * Zero these out.  They now point to freed memory, and it is
-	 * dangerous to hang onto the pointers.
-	 */
-	cmd->request_buffer = NULL;
-	cmd->request_bufflen = 0;
+	memset(&cmd->sdb, 0, sizeof(cmd->sdb));
 }
 EXPORT_SYMBOL(scsi_release_buffers);
 
@@ -829,7 +824,7 @@ EXPORT_SYMBOL(scsi_release_buffers);
 void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 {
 	int result = cmd->result;
-	int this_count = cmd->request_bufflen;
+	int this_count = scsi_bufflen(cmd);
 	struct request_queue *q = cmd->device->request_queue;
 	struct request *req = cmd->request;
 	int clear_errors = 1;
@@ -837,8 +832,6 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 	int sense_valid = 0;
 	int sense_deferred = 0;
 
-	scsi_release_buffers(cmd);
-
 	if (result) {
 		sense_valid = scsi_command_normalize_sense(cmd, &sshdr);
 		if (sense_valid)
@@ -861,9 +854,11 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 				req->sense_len = len;
 			}
 		}
-		req->data_len = cmd->resid;
+		req->data_len = scsi_get_resid(cmd);
 	}
 
+	scsi_release_buffers(cmd);
+
 	/*
 	 * Next deal with any sectors which we were able to correctly
 	 * handle.
@@ -871,7 +866,6 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 	SCSI_LOG_HLCOMPLETE(1, printk("%ld sectors total, "
 				      "%d bytes done.\n",
 				      req->nr_sectors, good_bytes));
-	SCSI_LOG_HLCOMPLETE(1, printk("use_sg is %d\n", cmd->use_sg));
 
 	if (clear_errors)
 		req->errors = 0;
@@ -1002,35 +996,30 @@ int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 {
 	struct request     *req = cmd->request;
 	int		   count;
-
-	/*
-	 * We used to not use scatter-gather for single segment request,
-	 * but now we do (it makes highmem I/O easier to support without
-	 * kmapping pages)
-	 */
-	cmd->use_sg = req->nr_phys_segments;
+	struct scsi_data_buffer *sdb = &cmd->sdb;
 
 	/*
 	 * If sg table allocation fails, requeue request later.
 	 */
-	if (unlikely(scsi_alloc_sgtable(cmd, gfp_mask))) {
+	if (unlikely(scsi_alloc_sgtable(sdb, req->nr_phys_segments,
+					gfp_mask))) {
 		scsi_unprep_request(req);
 		return BLKPREP_DEFER;
 	}
 
 	req->buffer = NULL;
 	if (blk_pc_request(req))
-		cmd->request_bufflen = req->data_len;
+		sdb->length = req->data_len;
 	else
-		cmd->request_bufflen = req->nr_sectors << 9;
+		sdb->length = req->nr_sectors << 9;
 
 	/* 
 	 * Next, walk the list, and fill in the addresses and sizes of
 	 * each segment.
 	 */
-	count = blk_rq_map_sg(req->q, req, cmd->request_buffer);
-	BUG_ON(count > cmd->use_sg);
-	cmd->use_sg = count;
+	count = blk_rq_map_sg(req->q, req, sdb->table.sgl);
+	BUG_ON(count > sdb->table.nents);
+	sdb->table.nents = count;
 	return BLKPREP_OK;
 }
 EXPORT_SYMBOL(scsi_init_io);
@@ -1086,9 +1075,7 @@ int scsi_setup_blk_pc_cmnd(struct scsi_device *sdev, struct request *req)
 		BUG_ON(req->data_len);
 		BUG_ON(req->data);
 
-		cmd->request_bufflen = 0;
-		cmd->request_buffer = NULL;
-		cmd->use_sg = 0;
+		memset(&cmd->sdb, 0, sizeof(cmd->sdb));
 		req->buffer = NULL;
 	}
 
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 24eba3118b5a..51a5557f42dd 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -519,7 +519,7 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq)
 		SCpnt->cmnd[4] = (unsigned char) this_count;
 		SCpnt->cmnd[5] = 0;
 	}
-	SCpnt->request_bufflen = this_count * sdp->sector_size;
+	SCpnt->sdb.length = this_count * sdp->sector_size;
 
 	/*
 	 * We shouldn't disconnect in the middle of a sector, so with a dumb
@@ -926,7 +926,7 @@ static struct block_device_operations sd_fops = {
 static int sd_done(struct scsi_cmnd *SCpnt)
 {
 	int result = SCpnt->result;
- 	unsigned int xfer_size = SCpnt->request_bufflen;
+	unsigned int xfer_size = scsi_bufflen(SCpnt);
  	unsigned int good_bytes = result ? 0 : xfer_size;
  	u64 start_lba = SCpnt->request->sector;
  	u64 bad_lba;
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index 1fcee16fa36d..50ba49250203 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -231,7 +231,7 @@ out:
 static int sr_done(struct scsi_cmnd *SCpnt)
 {
 	int result = SCpnt->result;
-	int this_count = SCpnt->request_bufflen;
+	int this_count = scsi_bufflen(SCpnt);
 	int good_bytes = (result == 0 ? this_count : 0);
 	int block_sectors = 0;
 	long error_sector;
@@ -379,17 +379,18 @@ static int sr_prep_fn(struct request_queue *q, struct request *rq)
 	}
 
 	{
-		struct scatterlist *sg = SCpnt->request_buffer;
-		int i, size = 0;
-		for (i = 0; i < SCpnt->use_sg; i++)
-			size += sg[i].length;
+		struct scatterlist *sg;
+		int i, size = 0, sg_count = scsi_sg_count(SCpnt);
 
-		if (size != SCpnt->request_bufflen && SCpnt->use_sg) {
+		scsi_for_each_sg(SCpnt, sg, sg_count, i)
+			size += sg->length;
+
+		if (size != scsi_bufflen(SCpnt)) {
 			scmd_printk(KERN_ERR, SCpnt,
 				"mismatch count %d, bytes %d\n",
-				size, SCpnt->request_bufflen);
-			if (SCpnt->request_bufflen > size)
-				SCpnt->request_bufflen = size;
+				size, scsi_bufflen(SCpnt));
+			if (scsi_bufflen(SCpnt) > size)
+				SCpnt->sdb.length = size;
 		}
 	}
 
@@ -397,12 +398,12 @@ static int sr_prep_fn(struct request_queue *q, struct request *rq)
 	 * request doesn't start on hw block boundary, add scatter pads
 	 */
 	if (((unsigned int)rq->sector % (s_size >> 9)) ||
-	    (SCpnt->request_bufflen % s_size)) {
+	    (scsi_bufflen(SCpnt) % s_size)) {
 		scmd_printk(KERN_NOTICE, SCpnt, "unaligned transfer\n");
 		goto out;
 	}
 
-	this_count = (SCpnt->request_bufflen >> 9) / (s_size >> 9);
+	this_count = (scsi_bufflen(SCpnt) >> 9) / (s_size >> 9);
 
 
 	SCSI_LOG_HLQUEUE(2, printk("%s : %s %d/%ld 512 byte blocks.\n",
@@ -416,7 +417,7 @@ static int sr_prep_fn(struct request_queue *q, struct request *rq)
 
 	if (this_count > 0xffff) {
 		this_count = 0xffff;
-		SCpnt->request_bufflen = this_count * s_size;
+		SCpnt->sdb.length = this_count * s_size;
 	}
 
 	SCpnt->cmnd[2] = (unsigned char) (block >> 24) & 0xff;
diff --git a/drivers/usb/storage/isd200.c b/drivers/usb/storage/isd200.c
index 178e8c2a8a2f..0db488624ab1 100644
--- a/drivers/usb/storage/isd200.c
+++ b/drivers/usb/storage/isd200.c
@@ -415,14 +415,14 @@ static void isd200_set_srb(struct isd200_info *info,
 		sg_init_one(&info->sg, buff, bufflen);
 
 	srb->sc_data_direction = dir;
-	srb->request_buffer = buff ? &info->sg : NULL;
-	srb->request_bufflen = bufflen;
-	srb->use_sg = buff ? 1 : 0;
+	srb->sdb.table.sgl = buff ? &info->sg : NULL;
+	srb->sdb.length = bufflen;
+	srb->sdb.table.nents = buff ? 1 : 0;
 }
 
 static void isd200_srb_set_bufflen(struct scsi_cmnd *srb, unsigned bufflen)
 {
-	srb->request_bufflen = bufflen;
+	srb->sdb.length = bufflen;
 }
 
 
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index 09d44f91dbdb..c6478bb6f963 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -11,6 +11,11 @@ struct request;
 struct Scsi_Host;
 struct scsi_device;
 
+struct scsi_data_buffer {
+	struct sg_table table;
+	unsigned length;
+	int resid;
+};
 
 /* embedded in scsi_cmnd */
 struct scsi_pointer {
@@ -61,15 +66,11 @@ struct scsi_cmnd {
 	/* These elements define the operation we are about to perform */
 #define MAX_COMMAND_SIZE	16
 	unsigned char cmnd[MAX_COMMAND_SIZE];
-	unsigned request_bufflen;	/* Actual request size */
 
 	struct timer_list eh_timeout;	/* Used to time out the command. */
-	void *request_buffer;		/* Actual requested buffer */
 
 	/* These elements define the operation we ultimately want to perform */
-	struct sg_table sg_table;
-	unsigned short use_sg;	/* Number of pieces of scatter-gather */
-
+	struct scsi_data_buffer sdb;
 	unsigned underflow;	/* Return error if less than
 				   this amount is transferred */
 
@@ -79,10 +80,6 @@ struct scsi_cmnd {
 				   reconnects.   Probably == sector
 				   size */
 
-	int resid;		/* Number of bytes requested to be
-				   transferred less actual number
-				   transferred (0 if not supported) */
-
 	struct request *request;	/* The command we are
 				   	   working on */
 
@@ -133,18 +130,29 @@ extern void scsi_release_buffers(struct scsi_cmnd *cmd);
 extern int scsi_dma_map(struct scsi_cmnd *cmd);
 extern void scsi_dma_unmap(struct scsi_cmnd *cmd);
 
-#define scsi_sg_count(cmd) ((cmd)->use_sg)
-#define scsi_sglist(cmd) ((cmd)->sg_table.sgl)
-#define scsi_bufflen(cmd) ((cmd)->request_bufflen)
+static inline unsigned scsi_sg_count(struct scsi_cmnd *cmd)
+{
+	return cmd->sdb.table.nents;
+}
+
+static inline struct scatterlist *scsi_sglist(struct scsi_cmnd *cmd)
+{
+	return cmd->sdb.table.sgl;
+}
+
+static inline unsigned scsi_bufflen(struct scsi_cmnd *cmd)
+{
+	return cmd->sdb.length;
+}
 
 static inline void scsi_set_resid(struct scsi_cmnd *cmd, int resid)
 {
-	cmd->resid = resid;
+	cmd->sdb.resid = resid;
 }
 
 static inline int scsi_get_resid(struct scsi_cmnd *cmd)
 {
-	return cmd->resid;
+	return cmd->sdb.resid;
 }
 
 #define scsi_for_each_sg(cmd, sg, nseg, __i)			\
diff --git a/include/scsi/scsi_eh.h b/include/scsi/scsi_eh.h
index d21b8913ceb3..1e08be1466ee 100644
--- a/include/scsi/scsi_eh.h
+++ b/include/scsi/scsi_eh.h
@@ -68,16 +68,14 @@ extern int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len,
 extern int scsi_reset_provider(struct scsi_device *, int);
 
 struct scsi_eh_save {
+	/* saved state */
 	int result;
 	enum dma_data_direction data_direction;
 	unsigned char cmd_len;
 	unsigned char cmnd[MAX_COMMAND_SIZE];
+	struct scsi_data_buffer sdb;
 
-	void *buffer;
-	unsigned bufflen;
-	unsigned short use_sg;
-	int resid;
-
+	/* new command support */
 	struct scatterlist sense_sgl;
 };
 
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 0fd4746ee39d..cb2bcab41dfb 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -136,9 +136,9 @@ struct scsi_host_template {
 	 * the done callback is invoked.
 	 *
 	 * This is called to inform the LLD to transfer
-	 * cmd->request_bufflen bytes. The cmd->use_sg speciefies the
+	 * scsi_bufflen(cmd) bytes. scsi_sg_count(cmd) speciefies the
 	 * number of scatterlist entried in the command and
-	 * cmd->request_buffer contains the scatterlist.
+	 * scsi_sglist(cmd) returns the scatterlist.
 	 *
 	 * return values: see queuecommand
 	 *
-- 
cgit v1.2.3


From 6f9a35e2dafa0f855ab051c11bdbf739745ff6f5 Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Thu, 13 Dec 2007 13:50:53 +0200
Subject: [SCSI] bidirectional command support

At the block level bidi request uses req->next_rq pointer for a second
bidi_read request.
At Scsi-midlayer a second scsi_data_buffer structure is used for the
bidi_read part. This bidi scsi_data_buffer is put on
request->next_rq->special. Struct scsi_cmnd is not changed.

- Define scsi_bidi_cmnd() to return true if it is a bidi request and a
  second sgtable was allocated.

- Define scsi_in()/scsi_out() to return the in or out scsi_data_buffer
  from this command This API is to isolate users from the mechanics of
  bidi.

- Define scsi_end_bidi_request() to do what scsi_end_request() does but
  for a bidi request. This is necessary because bidi commands are a bit
  tricky here. (See comments in body)

- scsi_release_buffers() will also release the bidi_read scsi_data_buffer

- scsi_io_completion() on bidi commands will now call
  scsi_end_bidi_request() and return.

- The previous work done in scsi_init_io() is now done in a new
  scsi_init_sgtable() (which is 99% identical to old scsi_init_io())
  The new scsi_init_io() will call the above twice if needed also for
  the bidi_read command. Only at this point is a command bidi.

- In scsi_error.c at scsi_eh_prep/restore_cmnd() make sure bidi-lld is not
  confused by a get-sense command that looks like bidi. This is done
  by puting NULL at request->next_rq, and restoring.

[jejb: update to sg_table and resolve conflicts
also update to blk-end-request and resolve conflicts]

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_error.c |   3 ++
 drivers/scsi/scsi_lib.c   | 106 +++++++++++++++++++++++++++++++++++++++-------
 include/scsi/scsi_cmnd.h  |  19 ++++++++-
 include/scsi/scsi_eh.h    |   1 +
 4 files changed, 113 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 4f4edc388c07..045a0868fc7b 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -618,9 +618,11 @@ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses,
 	memcpy(ses->cmnd, scmd->cmnd, sizeof(scmd->cmnd));
 	ses->data_direction = scmd->sc_data_direction;
 	ses->sdb = scmd->sdb;
+	ses->next_rq = scmd->request->next_rq;
 	ses->result = scmd->result;
 
 	memset(&scmd->sdb, 0, sizeof(scmd->sdb));
+	scmd->request->next_rq = NULL;
 
 	if (sense_bytes) {
 		scmd->sdb.length = min_t(unsigned, SCSI_SENSE_BUFFERSIZE,
@@ -673,6 +675,7 @@ void scsi_eh_restore_cmnd(struct scsi_cmnd* scmd, struct scsi_eh_save *ses)
 	memcpy(scmd->cmnd, ses->cmnd, sizeof(scmd->cmnd));
 	scmd->sc_data_direction = ses->data_direction;
 	scmd->sdb = ses->sdb;
+	scmd->request->next_rq = ses->next_rq;
 	scmd->result = ses->result;
 }
 EXPORT_SYMBOL(scsi_eh_restore_cmnd);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index d5e77e9b3a9c..681ed1b830f5 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -64,6 +64,8 @@ static struct scsi_host_sg_pool scsi_sg_pools[] = {
 };
 #undef SP
 
+static struct kmem_cache *scsi_bidi_sdb_cache;
+
 static void scsi_run_queue(struct request_queue *q);
 
 /*
@@ -790,9 +792,37 @@ void scsi_release_buffers(struct scsi_cmnd *cmd)
 		scsi_free_sgtable(&cmd->sdb);
 
 	memset(&cmd->sdb, 0, sizeof(cmd->sdb));
+
+	if (scsi_bidi_cmnd(cmd)) {
+		struct scsi_data_buffer *bidi_sdb =
+			cmd->request->next_rq->special;
+		scsi_free_sgtable(bidi_sdb);
+		kmem_cache_free(scsi_bidi_sdb_cache, bidi_sdb);
+		cmd->request->next_rq->special = NULL;
+	}
 }
 EXPORT_SYMBOL(scsi_release_buffers);
 
+/*
+ * Bidi commands Must be complete as a whole, both sides at once.
+ * If part of the bytes were written and lld returned
+ * scsi_in()->resid and/or scsi_out()->resid this information will be left
+ * in req->data_len and req->next_rq->data_len. The upper-layer driver can
+ * decide what to do with this information.
+ */
+void scsi_end_bidi_request(struct scsi_cmnd *cmd)
+{
+	blk_end_bidi_request(cmd->request, 0, scsi_out(cmd)->resid,
+							scsi_in(cmd)->resid);
+	scsi_release_buffers(cmd);
+
+	/*
+	 * This will goose the queue request function at the end, so we don't
+	 * need to worry about launching another command.
+	 */
+	scsi_next_command(cmd);
+}
+
 /*
  * Function:    scsi_io_completion()
  *
@@ -854,9 +884,15 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 				req->sense_len = len;
 			}
 		}
+		if (scsi_bidi_cmnd(cmd)) {
+			/* will also release_buffers */
+			scsi_end_bidi_request(cmd);
+			return;
+		}
 		req->data_len = scsi_get_resid(cmd);
 	}
 
+	BUG_ON(blk_bidi_rq(req)); /* bidi not support for !blk_pc_request yet */
 	scsi_release_buffers(cmd);
 
 	/*
@@ -982,28 +1018,16 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 	scsi_end_request(cmd, -EIO, this_count, !result);
 }
 
-/*
- * Function:    scsi_init_io()
- *
- * Purpose:     SCSI I/O initialize function.
- *
- * Arguments:   cmd   - Command descriptor we wish to initialize
- *
- * Returns:     0 on success
- *		BLKPREP_DEFER if the failure is retryable
- */
-int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb,
+			     gfp_t gfp_mask)
 {
-	struct request     *req = cmd->request;
-	int		   count;
-	struct scsi_data_buffer *sdb = &cmd->sdb;
+	int count;
 
 	/*
 	 * If sg table allocation fails, requeue request later.
 	 */
 	if (unlikely(scsi_alloc_sgtable(sdb, req->nr_phys_segments,
 					gfp_mask))) {
-		scsi_unprep_request(req);
 		return BLKPREP_DEFER;
 	}
 
@@ -1022,6 +1046,50 @@ int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 	sdb->table.nents = count;
 	return BLKPREP_OK;
 }
+
+/*
+ * Function:    scsi_init_io()
+ *
+ * Purpose:     SCSI I/O initialize function.
+ *
+ * Arguments:   cmd   - Command descriptor we wish to initialize
+ *
+ * Returns:     0 on success
+ *		BLKPREP_DEFER if the failure is retryable
+ *		BLKPREP_KILL if the failure is fatal
+ */
+int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+{
+	int error = scsi_init_sgtable(cmd->request, &cmd->sdb, gfp_mask);
+	if (error)
+		goto err_exit;
+
+	if (blk_bidi_rq(cmd->request)) {
+		struct scsi_data_buffer *bidi_sdb = kmem_cache_zalloc(
+			scsi_bidi_sdb_cache, GFP_ATOMIC);
+		if (!bidi_sdb) {
+			error = BLKPREP_DEFER;
+			goto err_exit;
+		}
+
+		cmd->request->next_rq->special = bidi_sdb;
+		error = scsi_init_sgtable(cmd->request->next_rq, bidi_sdb,
+								    GFP_ATOMIC);
+		if (error)
+			goto err_exit;
+	}
+
+	return BLKPREP_OK ;
+
+err_exit:
+	scsi_release_buffers(cmd);
+	if (error == BLKPREP_KILL)
+		scsi_put_command(cmd);
+	else /* BLKPREP_DEFER */
+		scsi_unprep_request(cmd->request);
+
+	return error;
+}
 EXPORT_SYMBOL(scsi_init_io);
 
 static struct scsi_cmnd *scsi_get_cmd_from_req(struct scsi_device *sdev,
@@ -1639,6 +1707,14 @@ int __init scsi_init_queue(void)
 		return -ENOMEM;
 	}
 
+	scsi_bidi_sdb_cache = kmem_cache_create("scsi_bidi_sdb",
+					sizeof(struct scsi_data_buffer),
+					0, 0, NULL);
+	if (!scsi_bidi_sdb_cache) {
+		printk(KERN_ERR "SCSI: can't init scsi bidi sdb cache\n");
+		return -ENOMEM;
+	}
+
 	for (i = 0; i < SG_MEMPOOL_NR; i++) {
 		struct scsi_host_sg_pool *sgp = scsi_sg_pools + i;
 		int size = sgp->size * sizeof(struct scatterlist);
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index c6478bb6f963..de28aab820b0 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -2,12 +2,12 @@
 #define _SCSI_SCSI_CMND_H
 
 #include <linux/dma-mapping.h>
+#include <linux/blkdev.h>
 #include <linux/list.h>
 #include <linux/types.h>
 #include <linux/timer.h>
 #include <linux/scatterlist.h>
 
-struct request;
 struct Scsi_Host;
 struct scsi_device;
 
@@ -158,4 +158,21 @@ static inline int scsi_get_resid(struct scsi_cmnd *cmd)
 #define scsi_for_each_sg(cmd, sg, nseg, __i)			\
 	for_each_sg(scsi_sglist(cmd), sg, nseg, __i)
 
+static inline int scsi_bidi_cmnd(struct scsi_cmnd *cmd)
+{
+	return blk_bidi_rq(cmd->request) &&
+		(cmd->request->next_rq->special != NULL);
+}
+
+static inline struct scsi_data_buffer *scsi_in(struct scsi_cmnd *cmd)
+{
+	return scsi_bidi_cmnd(cmd) ?
+		cmd->request->next_rq->special : &cmd->sdb;
+}
+
+static inline struct scsi_data_buffer *scsi_out(struct scsi_cmnd *cmd)
+{
+	return &cmd->sdb;
+}
+
 #endif /* _SCSI_SCSI_CMND_H */
diff --git a/include/scsi/scsi_eh.h b/include/scsi/scsi_eh.h
index 1e08be1466ee..25071d5d9bf8 100644
--- a/include/scsi/scsi_eh.h
+++ b/include/scsi/scsi_eh.h
@@ -74,6 +74,7 @@ struct scsi_eh_save {
 	unsigned char cmd_len;
 	unsigned char cmnd[MAX_COMMAND_SIZE];
 	struct scsi_data_buffer sdb;
+	struct request *next_rq;
 
 	/* new command support */
 	struct scatterlist sense_sgl;
-- 
cgit v1.2.3


From d3f46f39b7092594b498abc12f0c73b0b9913bde Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Tue, 15 Jan 2008 11:11:46 -0600
Subject: [SCSI] remove use_sg_chaining

With the sg table code, every SCSI driver is now either chain capable
or broken (or has sg_tablesize set so chaining is never activated), so
there's no need to have a check in the host template.

Also tidy up the code by moving the scatterlist size defines into the
SCSI includes and permit the last entry of the scatterlist pools not
to be a power of two.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 arch/ia64/hp/sim/simscsi.c            |  1 -
 drivers/infiniband/ulp/srp/ib_srp.c   |  1 -
 drivers/scsi/3w-9xxx.c                |  1 -
 drivers/scsi/3w-xxxx.c                |  1 -
 drivers/scsi/BusLogic.c               |  1 -
 drivers/scsi/NCR53c406a.c             |  1 -
 drivers/scsi/a100u2w.c                |  1 -
 drivers/scsi/aacraid/linit.c          |  1 -
 drivers/scsi/aha1740.c                |  1 -
 drivers/scsi/aic7xxx/aic79xx_osm.c    |  1 -
 drivers/scsi/aic7xxx/aic7xxx_osm.c    |  1 -
 drivers/scsi/aic7xxx_old.c            |  1 -
 drivers/scsi/arcmsr/arcmsr_hba.c      |  1 -
 drivers/scsi/dc395x.c                 |  1 -
 drivers/scsi/dpt_i2o.c                |  1 -
 drivers/scsi/eata.c                   |  1 -
 drivers/scsi/hosts.c                  |  1 -
 drivers/scsi/hptiop.c                 |  1 -
 drivers/scsi/ibmmca.c                 |  1 -
 drivers/scsi/ibmvscsi/ibmvscsi.c      |  1 -
 drivers/scsi/initio.c                 |  1 -
 drivers/scsi/iscsi_tcp.c              |  1 -
 drivers/scsi/lpfc/lpfc_scsi.c         |  2 -
 drivers/scsi/mac53c94.c               |  1 -
 drivers/scsi/megaraid.c               |  1 -
 drivers/scsi/megaraid/megaraid_mbox.c |  1 -
 drivers/scsi/megaraid/megaraid_sas.c  |  1 -
 drivers/scsi/mesh.c                   |  1 -
 drivers/scsi/nsp32.c                  |  1 -
 drivers/scsi/pcmcia/sym53c500_cs.c    |  1 -
 drivers/scsi/qla1280.c                |  1 -
 drivers/scsi/qla2xxx/qla_os.c         |  2 -
 drivers/scsi/qla4xxx/ql4_os.c         |  1 -
 drivers/scsi/qlogicfas.c              |  1 -
 drivers/scsi/scsi_lib.c               | 74 +++++++++--------------------------
 drivers/scsi/stex.c                   |  1 -
 drivers/scsi/sym53c416.c              |  1 -
 drivers/scsi/sym53c8xx_2/sym_glue.c   |  1 -
 drivers/scsi/u14-34f.c                |  1 -
 drivers/scsi/ultrastor.c              |  1 -
 drivers/scsi/wd7000.c                 |  1 -
 include/scsi/scsi.h                   | 19 +++++++++
 include/scsi/scsi_host.h              | 13 ------
 43 files changed, 37 insertions(+), 111 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/hp/sim/simscsi.c b/arch/ia64/hp/sim/simscsi.c
index 6ef9b5219930..7661bb065fa5 100644
--- a/arch/ia64/hp/sim/simscsi.c
+++ b/arch/ia64/hp/sim/simscsi.c
@@ -360,7 +360,6 @@ static struct scsi_host_template driver_template = {
 	.max_sectors		= 1024,
 	.cmd_per_lun		= SIMSCSI_REQ_QUEUE_LEN,
 	.use_clustering		= DISABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 static int __init
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index f2d2c7e2c76b..195ce7c12319 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -1571,7 +1571,6 @@ static struct scsi_host_template srp_template = {
 	.this_id			= -1,
 	.cmd_per_lun			= SRP_SQ_SIZE,
 	.use_clustering			= ENABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 	.shost_attrs			= srp_host_attrs
 };
 
diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c
index 1c244832c6c8..b4912d1cee2a 100644
--- a/drivers/scsi/3w-9xxx.c
+++ b/drivers/scsi/3w-9xxx.c
@@ -1990,7 +1990,6 @@ static struct scsi_host_template driver_template = {
 	.max_sectors		= TW_MAX_SECTORS,
 	.cmd_per_lun		= TW_MAX_CMDS_PER_LUN,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.shost_attrs		= twa_host_attrs,
 	.emulated		= 1
 };
diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c
index 59716ebeb10c..d09532162217 100644
--- a/drivers/scsi/3w-xxxx.c
+++ b/drivers/scsi/3w-xxxx.c
@@ -2261,7 +2261,6 @@ static struct scsi_host_template driver_template = {
 	.max_sectors		= TW_MAX_SECTORS,
 	.cmd_per_lun		= TW_MAX_CMDS_PER_LUN,	
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.shost_attrs		= tw_host_attrs,
 	.emulated		= 1
 };
diff --git a/drivers/scsi/BusLogic.c b/drivers/scsi/BusLogic.c
index ead47c143ce0..4d3ebb1af490 100644
--- a/drivers/scsi/BusLogic.c
+++ b/drivers/scsi/BusLogic.c
@@ -3575,7 +3575,6 @@ static struct scsi_host_template Bus_Logic_template = {
 	.unchecked_isa_dma = 1,
 	.max_sectors = 128,
 	.use_clustering = ENABLE_CLUSTERING,
-	.use_sg_chaining = ENABLE_SG_CHAINING,
 };
 
 /*
diff --git a/drivers/scsi/NCR53c406a.c b/drivers/scsi/NCR53c406a.c
index 137d065db3da..6961f78742ae 100644
--- a/drivers/scsi/NCR53c406a.c
+++ b/drivers/scsi/NCR53c406a.c
@@ -1065,7 +1065,6 @@ static struct scsi_host_template driver_template =
      .cmd_per_lun       	= 1			/* commands per lun */, 
      .unchecked_isa_dma 	= 1			/* unchecked_isa_dma */,
      .use_clustering    	= ENABLE_CLUSTERING,
-     .use_sg_chaining           = ENABLE_SG_CHAINING,
 };
 
 #include "scsi_module.c"
diff --git a/drivers/scsi/a100u2w.c b/drivers/scsi/a100u2w.c
index d3a6d15fb77a..f608d4a1d6da 100644
--- a/drivers/scsi/a100u2w.c
+++ b/drivers/scsi/a100u2w.c
@@ -1071,7 +1071,6 @@ static struct scsi_host_template inia100_template = {
 	.sg_tablesize		= SG_ALL,
 	.cmd_per_lun 		= 1,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 static int __devinit inia100_probe_one(struct pci_dev *pdev,
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 61be22774e99..0e8267c1e915 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -1032,7 +1032,6 @@ static struct scsi_host_template aac_driver_template = {
 	.cmd_per_lun    		= AAC_NUM_IO_FIB,
 #endif
 	.use_clustering			= ENABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 	.emulated                       = 1,
 };
 
diff --git a/drivers/scsi/aha1740.c b/drivers/scsi/aha1740.c
index be58a0b097c7..7c45d88a205b 100644
--- a/drivers/scsi/aha1740.c
+++ b/drivers/scsi/aha1740.c
@@ -563,7 +563,6 @@ static struct scsi_host_template aha1740_template = {
 	.sg_tablesize     = AHA1740_SCATTER,
 	.cmd_per_lun      = AHA1740_CMDLUN,
 	.use_clustering   = ENABLE_CLUSTERING,
-	.use_sg_chaining  = ENABLE_SG_CHAINING,
 	.eh_abort_handler = aha1740_eh_abort_handler,
 };
 
diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c
index 3c4efa4ae543..014654792901 100644
--- a/drivers/scsi/aic7xxx/aic79xx_osm.c
+++ b/drivers/scsi/aic7xxx/aic79xx_osm.c
@@ -766,7 +766,6 @@ struct scsi_host_template aic79xx_driver_template = {
 	.max_sectors		= 8192,
 	.cmd_per_lun		= 2,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.slave_alloc		= ahd_linux_slave_alloc,
 	.slave_configure	= ahd_linux_slave_configure,
 	.target_alloc		= ahd_linux_target_alloc,
diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c
index bff7f8c59856..99a3b33a3233 100644
--- a/drivers/scsi/aic7xxx/aic7xxx_osm.c
+++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c
@@ -747,7 +747,6 @@ struct scsi_host_template aic7xxx_driver_template = {
 	.max_sectors		= 8192,
 	.cmd_per_lun		= 2,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.slave_alloc		= ahc_linux_slave_alloc,
 	.slave_configure	= ahc_linux_slave_configure,
 	.target_alloc		= ahc_linux_target_alloc,
diff --git a/drivers/scsi/aic7xxx_old.c b/drivers/scsi/aic7xxx_old.c
index bcb0b870320c..3bfd9296bbfa 100644
--- a/drivers/scsi/aic7xxx_old.c
+++ b/drivers/scsi/aic7xxx_old.c
@@ -11141,7 +11141,6 @@ static struct scsi_host_template driver_template = {
 	.max_sectors		= 2048,
 	.cmd_per_lun		= 3,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 #include "scsi_module.c"
diff --git a/drivers/scsi/arcmsr/arcmsr_hba.c b/drivers/scsi/arcmsr/arcmsr_hba.c
index d80dba913a75..f4a202e8df26 100644
--- a/drivers/scsi/arcmsr/arcmsr_hba.c
+++ b/drivers/scsi/arcmsr/arcmsr_hba.c
@@ -122,7 +122,6 @@ static struct scsi_host_template arcmsr_scsi_host_template = {
 	.max_sectors    	= ARCMSR_MAX_XFER_SECTORS,
 	.cmd_per_lun		= ARCMSR_MAX_CMD_PERLUN,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.shost_attrs		= arcmsr_host_attrs,
 };
 #ifdef CONFIG_SCSI_ARCMSR_AER
diff --git a/drivers/scsi/dc395x.c b/drivers/scsi/dc395x.c
index f93c73c0ba53..22ef3716e786 100644
--- a/drivers/scsi/dc395x.c
+++ b/drivers/scsi/dc395x.c
@@ -4763,7 +4763,6 @@ static struct scsi_host_template dc395x_driver_template = {
 	.eh_bus_reset_handler   = dc395x_eh_bus_reset,
 	.unchecked_isa_dma      = 0,
 	.use_clustering         = DISABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 
diff --git a/drivers/scsi/dpt_i2o.c b/drivers/scsi/dpt_i2o.c
index 19cce125124c..c9dd8392aab2 100644
--- a/drivers/scsi/dpt_i2o.c
+++ b/drivers/scsi/dpt_i2o.c
@@ -3340,7 +3340,6 @@ static struct scsi_host_template driver_template = {
 	.this_id		= 7,
 	.cmd_per_lun		= 1,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 #include "scsi_module.c"
 MODULE_LICENSE("GPL");
diff --git a/drivers/scsi/eata.c b/drivers/scsi/eata.c
index 05163cefec12..8be3d76656fa 100644
--- a/drivers/scsi/eata.c
+++ b/drivers/scsi/eata.c
@@ -524,7 +524,6 @@ static struct scsi_host_template driver_template = {
 	.this_id = 7,
 	.unchecked_isa_dma = 1,
 	.use_clustering = ENABLE_CLUSTERING,
-	.use_sg_chaining = ENABLE_SG_CHAINING,
 };
 
 #if !defined(__BIG_ENDIAN_BITFIELD) && !defined(__LITTLE_ENDIAN_BITFIELD)
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 5ea1f986220c..880c78bff0e1 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -342,7 +342,6 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 	shost->use_clustering = sht->use_clustering;
 	shost->ordered_tag = sht->ordered_tag;
 	shost->active_mode = sht->supported_mode;
-	shost->use_sg_chaining = sht->use_sg_chaining;
 
 	if (sht->supported_mode == MODE_UNKNOWN)
 		/* means we didn't set it ... default to INITIATOR */
diff --git a/drivers/scsi/hptiop.c b/drivers/scsi/hptiop.c
index 890f44fa0e1a..ff149ad6bc4e 100644
--- a/drivers/scsi/hptiop.c
+++ b/drivers/scsi/hptiop.c
@@ -906,7 +906,6 @@ static struct scsi_host_template driver_template = {
 	.unchecked_isa_dma          = 0,
 	.emulated                   = 0,
 	.use_clustering             = ENABLE_CLUSTERING,
-	.use_sg_chaining            = ENABLE_SG_CHAINING,
 	.proc_name                  = driver_name,
 	.shost_attrs                = hptiop_attrs,
 	.this_id                    = -1,
diff --git a/drivers/scsi/ibmmca.c b/drivers/scsi/ibmmca.c
index db004a450732..4d15a62914e9 100644
--- a/drivers/scsi/ibmmca.c
+++ b/drivers/scsi/ibmmca.c
@@ -1501,7 +1501,6 @@ static struct scsi_host_template ibmmca_driver_template = {
           .sg_tablesize   = 16,
           .cmd_per_lun    = 1,
           .use_clustering = ENABLE_CLUSTERING,
-          .use_sg_chaining = ENABLE_SG_CHAINING,
 };
 
 static int ibmmca_probe(struct device *dev)
diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c
index 30819012898f..78d46a900bb5 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.c
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.c
@@ -1600,7 +1600,6 @@ static struct scsi_host_template driver_template = {
 	.this_id = -1,
 	.sg_tablesize = SG_ALL,
 	.use_clustering = ENABLE_CLUSTERING,
-	.use_sg_chaining = ENABLE_SG_CHAINING,
 	.shost_attrs = ibmvscsi_attrs,
 };
 
diff --git a/drivers/scsi/initio.c b/drivers/scsi/initio.c
index a10a5c74b48d..0cc8868ea35d 100644
--- a/drivers/scsi/initio.c
+++ b/drivers/scsi/initio.c
@@ -2833,7 +2833,6 @@ static struct scsi_host_template initio_template = {
 	.sg_tablesize		= SG_ALL,
 	.cmd_per_lun		= 1,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 static int initio_probe_one(struct pci_dev *pdev,
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index e5be5fd4ef58..b6f99dfbb038 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -1933,7 +1933,6 @@ static struct scsi_host_template iscsi_sht = {
 	.eh_device_reset_handler= iscsi_eh_device_reset,
 	.eh_host_reset_handler	= iscsi_eh_host_reset,
 	.use_clustering         = DISABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.slave_configure        = iscsi_tcp_slave_configure,
 	.proc_name		= "iscsi_tcp",
 	.this_id		= -1,
diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index 6483c62730b3..fc5c3a42b05a 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -1459,7 +1459,6 @@ struct scsi_host_template lpfc_template = {
 	.scan_finished		= lpfc_scan_finished,
 	.this_id		= -1,
 	.sg_tablesize		= LPFC_DEFAULT_SG_SEG_CNT,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.cmd_per_lun		= LPFC_CMD_PER_LUN,
 	.use_clustering		= ENABLE_CLUSTERING,
 	.shost_attrs		= lpfc_hba_attrs,
@@ -1482,7 +1481,6 @@ struct scsi_host_template lpfc_vport_template = {
 	.sg_tablesize		= LPFC_DEFAULT_SG_SEG_CNT,
 	.cmd_per_lun		= LPFC_CMD_PER_LUN,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.shost_attrs		= lpfc_vport_attrs,
 	.max_sectors		= 0xFFFF,
 };
diff --git a/drivers/scsi/mac53c94.c b/drivers/scsi/mac53c94.c
index a035001f4438..b12ad7c7c673 100644
--- a/drivers/scsi/mac53c94.c
+++ b/drivers/scsi/mac53c94.c
@@ -402,7 +402,6 @@ static struct scsi_host_template mac53c94_template = {
 	.sg_tablesize	= SG_ALL,
 	.cmd_per_lun	= 1,
 	.use_clustering	= DISABLE_CLUSTERING,
-	.use_sg_chaining = ENABLE_SG_CHAINING,
 };
 
 static int mac53c94_probe(struct macio_dev *mdev, const struct of_device_id *match)
diff --git a/drivers/scsi/megaraid.c b/drivers/scsi/megaraid.c
index 765c24d2bc38..4d59ae8491a4 100644
--- a/drivers/scsi/megaraid.c
+++ b/drivers/scsi/megaraid.c
@@ -4490,7 +4490,6 @@ static struct scsi_host_template megaraid_template = {
 	.sg_tablesize			= MAX_SGLIST,
 	.cmd_per_lun			= DEF_CMD_PER_LUN,
 	.use_clustering			= ENABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 	.eh_abort_handler		= megaraid_abort,
 	.eh_device_reset_handler	= megaraid_reset,
 	.eh_bus_reset_handler		= megaraid_reset,
diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c
index 24e32e446e76..6db77c00e3ee 100644
--- a/drivers/scsi/megaraid/megaraid_mbox.c
+++ b/drivers/scsi/megaraid/megaraid_mbox.c
@@ -361,7 +361,6 @@ static struct scsi_host_template megaraid_template_g = {
 	.eh_host_reset_handler		= megaraid_reset_handler,
 	.change_queue_depth		= megaraid_change_queue_depth,
 	.use_clustering			= ENABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 	.sdev_attrs			= megaraid_sdev_attrs,
 	.shost_attrs			= megaraid_shost_attrs,
 };
diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c
index d7ec921865c4..672c759ac24d 100644
--- a/drivers/scsi/megaraid/megaraid_sas.c
+++ b/drivers/scsi/megaraid/megaraid_sas.c
@@ -1192,7 +1192,6 @@ static struct scsi_host_template megasas_template = {
 	.eh_timed_out = megasas_reset_timer,
 	.bios_param = megasas_bios_param,
 	.use_clustering = ENABLE_CLUSTERING,
-	.use_sg_chaining = ENABLE_SG_CHAINING,
 };
 
 /**
diff --git a/drivers/scsi/mesh.c b/drivers/scsi/mesh.c
index 7470ff39ab22..651d09b08f2a 100644
--- a/drivers/scsi/mesh.c
+++ b/drivers/scsi/mesh.c
@@ -1843,7 +1843,6 @@ static struct scsi_host_template mesh_template = {
 	.sg_tablesize			= SG_ALL,
 	.cmd_per_lun			= 2,
 	.use_clustering			= DISABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 };
 
 static int mesh_probe(struct macio_dev *mdev, const struct of_device_id *match)
diff --git a/drivers/scsi/nsp32.c b/drivers/scsi/nsp32.c
index 28161dc95e0d..7fed35372150 100644
--- a/drivers/scsi/nsp32.c
+++ b/drivers/scsi/nsp32.c
@@ -281,7 +281,6 @@ static struct scsi_host_template nsp32_template = {
 	.cmd_per_lun			= 1,
 	.this_id			= NSP32_HOST_SCSIID,
 	.use_clustering			= DISABLE_CLUSTERING,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 	.eh_abort_handler       	= nsp32_eh_abort,
 	.eh_bus_reset_handler		= nsp32_eh_bus_reset,
 	.eh_host_reset_handler		= nsp32_eh_host_reset,
diff --git a/drivers/scsi/pcmcia/sym53c500_cs.c b/drivers/scsi/pcmcia/sym53c500_cs.c
index 969b9387a0c3..3454a5714749 100644
--- a/drivers/scsi/pcmcia/sym53c500_cs.c
+++ b/drivers/scsi/pcmcia/sym53c500_cs.c
@@ -692,7 +692,6 @@ static struct scsi_host_template sym53c500_driver_template = {
      .sg_tablesize		= 32,
      .cmd_per_lun		= 1,
      .use_clustering		= ENABLE_CLUSTERING,
-     .use_sg_chaining		= ENABLE_SG_CHAINING,
      .shost_attrs		= SYM53C500_shost_attrs
 };
 
diff --git a/drivers/scsi/qla1280.c b/drivers/scsi/qla1280.c
index c94906abfee3..68c0d09ffe78 100644
--- a/drivers/scsi/qla1280.c
+++ b/drivers/scsi/qla1280.c
@@ -4204,7 +4204,6 @@ static struct scsi_host_template qla1280_driver_template = {
 	.sg_tablesize		= SG_ALL,
 	.cmd_per_lun		= 1,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index aba1e6d48066..3954ed2d7b51 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -131,7 +131,6 @@ static struct scsi_host_template qla2x00_driver_template = {
 	.this_id		= -1,
 	.cmd_per_lun		= 3,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.sg_tablesize		= SG_ALL,
 
 	/*
@@ -163,7 +162,6 @@ struct scsi_host_template qla24xx_driver_template = {
 	.this_id		= -1,
 	.cmd_per_lun		= 3,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.sg_tablesize		= SG_ALL,
 
 	.max_sectors		= 0xFFFF,
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index d3f86646cb08..2e2b9fedffcc 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -94,7 +94,6 @@ static struct scsi_host_template qla4xxx_driver_template = {
 	.this_id		= -1,
 	.cmd_per_lun		= 3,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.sg_tablesize		= SG_ALL,
 
 	.max_sectors		= 0xFFFF,
diff --git a/drivers/scsi/qlogicfas.c b/drivers/scsi/qlogicfas.c
index 1769f965eedf..1e874f1fb5c6 100644
--- a/drivers/scsi/qlogicfas.c
+++ b/drivers/scsi/qlogicfas.c
@@ -197,7 +197,6 @@ static struct scsi_host_template qlogicfas_driver_template = {
 	.sg_tablesize		= SG_ALL,
 	.cmd_per_lun		= 1,
 	.use_clustering		= DISABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 static __init int qlogicfas_init(void)
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 4560713ecb14..e1c7eebfe907 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/bio.h>
+#include <linux/bitops.h>
 #include <linux/blkdev.h>
 #include <linux/completion.h>
 #include <linux/kernel.h>
@@ -34,13 +35,6 @@
 #define SG_MEMPOOL_NR		ARRAY_SIZE(scsi_sg_pools)
 #define SG_MEMPOOL_SIZE		2
 
-/*
- * The maximum number of SG segments that we will put inside a scatterlist
- * (unless chaining is used). Should ideally fit inside a single page, to
- * avoid a higher order allocation.
- */
-#define SCSI_MAX_SG_SEGMENTS	128
-
 struct scsi_host_sg_pool {
 	size_t		size;
 	char		*name;
@@ -48,19 +42,26 @@ struct scsi_host_sg_pool {
 	mempool_t	*pool;
 };
 
-#define SP(x) { x, "sgpool-" #x }
+#define SP(x) { x, "sgpool-" __stringify(x) }
+#if (SCSI_MAX_SG_SEGMENTS < 32)
+#error SCSI_MAX_SG_SEGMENTS is too small (must be 32 or greater)
+#endif
 static struct scsi_host_sg_pool scsi_sg_pools[] = {
 	SP(8),
 	SP(16),
-#if (SCSI_MAX_SG_SEGMENTS > 16)
-	SP(32),
 #if (SCSI_MAX_SG_SEGMENTS > 32)
-	SP(64),
+	SP(32),
 #if (SCSI_MAX_SG_SEGMENTS > 64)
+	SP(64),
+#if (SCSI_MAX_SG_SEGMENTS > 128)
 	SP(128),
+#if (SCSI_MAX_SG_SEGMENTS > 256)
+#error SCSI_MAX_SG_SEGMENTS is too large (256 MAX)
 #endif
 #endif
 #endif
+#endif
+	SP(SCSI_MAX_SG_SEGMENTS)
 };
 #undef SP
 
@@ -692,42 +693,16 @@ static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int error,
 	return NULL;
 }
 
-/*
- * Like SCSI_MAX_SG_SEGMENTS, but for archs that have sg chaining. This limit
- * is totally arbitrary, a setting of 2048 will get you at least 8mb ios.
- */
-#define SCSI_MAX_SG_CHAIN_SEGMENTS	2048
-
 static inline unsigned int scsi_sgtable_index(unsigned short nents)
 {
 	unsigned int index;
 
-	switch (nents) {
-	case 1 ... 8:
+	BUG_ON(nents > SCSI_MAX_SG_SEGMENTS);
+
+	if (nents <= 8)
 		index = 0;
-		break;
-	case 9 ... 16:
-		index = 1;
-		break;
-#if (SCSI_MAX_SG_SEGMENTS > 16)
-	case 17 ... 32:
-		index = 2;
-		break;
-#if (SCSI_MAX_SG_SEGMENTS > 32)
-	case 33 ... 64:
-		index = 3;
-		break;
-#if (SCSI_MAX_SG_SEGMENTS > 64)
-	case 65 ... 128:
-		index = 4;
-		break;
-#endif
-#endif
-#endif
-	default:
-		printk(KERN_ERR "scsi: bad segment count=%d\n", nents);
-		BUG();
-	}
+	else
+		index = get_count_order(nents) - 3;
 
 	return index;
 }
@@ -1603,20 +1578,7 @@ struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost,
 	 * this limit is imposed by hardware restrictions
 	 */
 	blk_queue_max_hw_segments(q, shost->sg_tablesize);
-
-	/*
-	 * In the future, sg chaining support will be mandatory and this
-	 * ifdef can then go away. Right now we don't have all archs
-	 * converted, so better keep it safe.
-	 */
-#ifdef ARCH_HAS_SG_CHAIN
-	if (shost->use_sg_chaining)
-		blk_queue_max_phys_segments(q, SCSI_MAX_SG_CHAIN_SEGMENTS);
-	else
-		blk_queue_max_phys_segments(q, SCSI_MAX_SG_SEGMENTS);
-#else
-	blk_queue_max_phys_segments(q, SCSI_MAX_SG_SEGMENTS);
-#endif
+	blk_queue_max_phys_segments(q, SCSI_MAX_SG_CHAIN_SEGMENTS);
 
 	blk_queue_max_sectors(q, shost->max_sectors);
 	blk_queue_bounce_limit(q, scsi_calculate_bounce_limit(shost));
diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c
index e3fab3a6aed7..72f6d8015358 100644
--- a/drivers/scsi/stex.c
+++ b/drivers/scsi/stex.c
@@ -1123,7 +1123,6 @@ static struct scsi_host_template driver_template = {
 	.this_id			= -1,
 	.sg_tablesize			= ST_MAX_SG,
 	.cmd_per_lun			= ST_CMD_PER_LUN,
-	.use_sg_chaining		= ENABLE_SG_CHAINING,
 };
 
 static int stex_set_dma_mask(struct pci_dev * pdev)
diff --git a/drivers/scsi/sym53c416.c b/drivers/scsi/sym53c416.c
index 1f6fd1680335..6325901e5093 100644
--- a/drivers/scsi/sym53c416.c
+++ b/drivers/scsi/sym53c416.c
@@ -840,6 +840,5 @@ static struct scsi_host_template driver_template = {
 	.cmd_per_lun =		1,
 	.unchecked_isa_dma =	1,
 	.use_clustering =	ENABLE_CLUSTERING,
-	.use_sg_chaining =	ENABLE_SG_CHAINING,
 };
 #include "scsi_module.c"
diff --git a/drivers/scsi/sym53c8xx_2/sym_glue.c b/drivers/scsi/sym53c8xx_2/sym_glue.c
index e939f3813233..d39107b7669b 100644
--- a/drivers/scsi/sym53c8xx_2/sym_glue.c
+++ b/drivers/scsi/sym53c8xx_2/sym_glue.c
@@ -1681,7 +1681,6 @@ static struct scsi_host_template sym2_template = {
 	.eh_host_reset_handler	= sym53c8xx_eh_host_reset_handler,
 	.this_id		= 7,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 	.max_sectors		= 0xFFFF,
 #ifdef SYM_LINUX_PROC_INFO_SUPPORT
 	.proc_info		= sym53c8xx_proc_info,
diff --git a/drivers/scsi/u14-34f.c b/drivers/scsi/u14-34f.c
index 4bc5407f9695..662c00451be4 100644
--- a/drivers/scsi/u14-34f.c
+++ b/drivers/scsi/u14-34f.c
@@ -451,7 +451,6 @@ static struct scsi_host_template driver_template = {
                 .this_id                 = 7,
                 .unchecked_isa_dma       = 1,
                 .use_clustering          = ENABLE_CLUSTERING,
-                .use_sg_chaining         = ENABLE_SG_CHAINING,
                 };
 
 #if !defined(__BIG_ENDIAN_BITFIELD) && !defined(__LITTLE_ENDIAN_BITFIELD)
diff --git a/drivers/scsi/ultrastor.c b/drivers/scsi/ultrastor.c
index 75eca6b22db5..f385dce8dfbe 100644
--- a/drivers/scsi/ultrastor.c
+++ b/drivers/scsi/ultrastor.c
@@ -1204,6 +1204,5 @@ static struct scsi_host_template driver_template = {
 	.cmd_per_lun       = ULTRASTOR_MAX_CMDS_PER_LUN,
 	.unchecked_isa_dma = 1,
 	.use_clustering    = ENABLE_CLUSTERING,
-	.use_sg_chaining   = ENABLE_SG_CHAINING,
 };
 #include "scsi_module.c"
diff --git a/drivers/scsi/wd7000.c b/drivers/scsi/wd7000.c
index b4304ae78527..c975c01b3a02 100644
--- a/drivers/scsi/wd7000.c
+++ b/drivers/scsi/wd7000.c
@@ -1671,7 +1671,6 @@ static struct scsi_host_template driver_template = {
 	.cmd_per_lun		= 1,
 	.unchecked_isa_dma	= 1,
 	.use_clustering		= ENABLE_CLUSTERING,
-	.use_sg_chaining	= ENABLE_SG_CHAINING,
 };
 
 #include "scsi_module.c"
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index 702fcfeb37f1..0f55f5b24896 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -10,6 +10,25 @@
 
 #include <linux/types.h>
 
+/*
+ * The maximum number of SG segments that we will put inside a
+ * scatterlist (unless chaining is used). Should ideally fit inside a
+ * single page, to avoid a higher order allocation.  We could define this
+ * to SG_MAX_SINGLE_ALLOC to pack correctly at the highest order.  The
+ * minimum value is 32
+ */
+#define SCSI_MAX_SG_SEGMENTS	128
+
+/*
+ * Like SCSI_MAX_SG_SEGMENTS, but for archs that have sg chaining. This limit
+ * is totally arbitrary, a setting of 2048 will get you at least 8mb ios.
+ */
+#ifdef ARCH_HAS_SG_CHAIN
+#define SCSI_MAX_SG_CHAIN_SEGMENTS	2048
+#else
+#define SCSI_MAX_SG_CHAIN_SEGMENTS	SCSI_MAX_SG_SEGMENTS
+#endif
+
 /*
  *	SCSI command lengths
  */
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index cb2bcab41dfb..5c58d594126a 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -39,9 +39,6 @@ struct blk_queue_tags;
 #define DISABLE_CLUSTERING 0
 #define ENABLE_CLUSTERING 1
 
-#define DISABLE_SG_CHAINING 0
-#define ENABLE_SG_CHAINING 1
-
 enum scsi_eh_timer_return {
 	EH_NOT_HANDLED,
 	EH_HANDLED,
@@ -445,15 +442,6 @@ struct scsi_host_template {
 	 */
 	unsigned ordered_tag:1;
 
-	/*
-	 * true if the low-level driver can support sg chaining. this
-	 * will be removed eventually when all the drivers are
-	 * converted to support sg chaining.
-	 *
-	 * Status: OBSOLETE
-	 */
-	unsigned use_sg_chaining:1;
-
 	/*
 	 * Countdown for host blocking with no commands outstanding
 	 */
@@ -598,7 +586,6 @@ struct Scsi_Host {
 	unsigned unchecked_isa_dma:1;
 	unsigned use_clustering:1;
 	unsigned use_blk_tcq:1;
-	unsigned use_sg_chaining:1;
 
 	/*
 	 * Host has requested that no further requests come through for the
-- 
cgit v1.2.3


From c639d14e2f70d086842d37fb8e4f5ab9bd56af2e Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <tomof@acm.org>
Date: Wed, 23 Jan 2008 01:32:01 +0900
Subject: [SCSI] scsi_debug: add XDWRITEREAD_10 support

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Douglas Gilbert <dougg@torque.net>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_debug.c | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 include/scsi/scsi.h       |  1 +
 2 files changed, 71 insertions(+)

(limited to 'include')

diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index d810aa7aee40..1541c174937a 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -280,6 +280,8 @@ static int resp_write(struct scsi_cmnd * SCpnt, unsigned long long lba,
 		      unsigned int num, struct sdebug_dev_info * devip);
 static int resp_report_luns(struct scsi_cmnd * SCpnt,
 			    struct sdebug_dev_info * devip);
+static int resp_xdwriteread(struct scsi_cmnd *scp, unsigned long long lba,
+			    unsigned int num, struct sdebug_dev_info *devip);
 static int fill_from_dev_buffer(struct scsi_cmnd * scp, unsigned char * arr,
                                 int arr_len);
 static int fetch_to_dev_buffer(struct scsi_cmnd * scp, unsigned char * arr,
@@ -334,6 +336,7 @@ static void get_data_transfer_info(unsigned char *cmd,
 		break;
 	case WRITE_10:
 	case READ_10:
+	case XDWRITEREAD_10:
 		*lba = cmd[5] + (cmd[4] << 8) +	(cmd[3] << 16) + (cmd[2] << 24);
 		*num = cmd[8] + (cmd[7] << 8);
 		break;
@@ -542,6 +545,28 @@ int scsi_debug_queuecommand(struct scsi_cmnd * SCpnt, done_funct_t done)
 	case WRITE_BUFFER:
 		errsts = check_readiness(SCpnt, 1, devip);
 		break;
+	case XDWRITEREAD_10:
+		if (!scsi_bidi_cmnd(SCpnt)) {
+			mk_sense_buffer(devip, ILLEGAL_REQUEST,
+					INVALID_FIELD_IN_CDB, 0);
+			errsts = check_condition_result;
+			break;
+		}
+
+		errsts = check_readiness(SCpnt, 0, devip);
+		if (errsts)
+			break;
+		if (scsi_debug_fake_rw)
+			break;
+		get_data_transfer_info(cmd, &lba, &num);
+		errsts = resp_read(SCpnt, lba, num, devip);
+		if (errsts)
+			break;
+		errsts = resp_write(SCpnt, lba, num, devip);
+		if (errsts)
+			break;
+		errsts = resp_xdwriteread(SCpnt, lba, num, devip);
+		break;
 	default:
 		if (SCSI_DEBUG_OPT_NOISE & scsi_debug_opts)
 			printk(KERN_INFO "scsi_debug: Opcode: 0x%x not "
@@ -1948,6 +1973,50 @@ static int resp_report_luns(struct scsi_cmnd * scp,
 				    min((int)alloc_len, SDEBUG_RLUN_ARR_SZ));
 }
 
+static int resp_xdwriteread(struct scsi_cmnd *scp, unsigned long long lba,
+			    unsigned int num, struct sdebug_dev_info *devip)
+{
+	int i, j, ret = -1;
+	unsigned char *kaddr, *buf;
+	unsigned int offset;
+	struct scatterlist *sg;
+	struct scsi_data_buffer *sdb = scsi_in(scp);
+
+	/* better not to use temporary buffer. */
+	buf = kmalloc(scsi_bufflen(scp), GFP_ATOMIC);
+	if (!buf)
+		return ret;
+
+	offset = 0;
+	scsi_for_each_sg(scp, sg, scsi_sg_count(scp), i) {
+		kaddr = (unsigned char *)kmap_atomic(sg_page(sg), KM_USER0);
+		if (!kaddr)
+			goto out;
+
+		memcpy(buf + offset, kaddr + sg->offset, sg->length);
+		offset += sg->length;
+		kunmap_atomic(kaddr, KM_USER0);
+	}
+
+	offset = 0;
+	for_each_sg(sdb->table.sgl, sg, sdb->table.nents, i) {
+		kaddr = (unsigned char *)kmap_atomic(sg_page(sg), KM_USER0);
+		if (!kaddr)
+			goto out;
+
+		for (j = 0; j < sg->length; j++)
+			*(kaddr + sg->offset + j) ^= *(buf + offset + j);
+
+		offset += sg->length;
+		kunmap_atomic(kaddr, KM_USER0);
+	}
+	ret = 0;
+out:
+	kfree(buf);
+
+	return ret;
+}
+
 /* When timer goes off this function is called. */
 static void timer_intr_handler(unsigned long indx)
 {
@@ -1981,6 +2050,7 @@ static int scsi_debug_slave_alloc(struct scsi_device * sdp)
 	if (SCSI_DEBUG_OPT_NOISE & scsi_debug_opts)
 		printk(KERN_INFO "scsi_debug: slave_alloc <%u %u %u %u>\n",
 		       sdp->host->host_no, sdp->channel, sdp->id, sdp->lun);
+	set_bit(QUEUE_FLAG_BIDI, &sdp->request_queue->queue_flags);
 	return 0;
 }
 
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index 0f55f5b24896..82251575a9b4 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -102,6 +102,7 @@ extern const unsigned char scsi_command_size[8];
 #define READ_TOC              0x43
 #define LOG_SELECT            0x4c
 #define LOG_SENSE             0x4d
+#define XDWRITEREAD_10        0x53
 #define MODE_SELECT_10        0x55
 #define RESERVE_10            0x56
 #define RELEASE_10            0x57
-- 
cgit v1.2.3